Deodorizing Activity of Hop Bitter Acids and Their Oxidation Products Against Allyl Methyl Sulfide, a Major Contributor to Unpleasant Garlic-Associated Breath and Body Odor

Abstract

Garlic is a spice widely used worldwide, but ingestion of garlic can cause unpleasant breath odor that can be offensive in interpersonal interactions. Among several sulfur-containing components of garlic, allyl methyl sulfide is considered the primary causative agent of unpleasant garlic breath and body odor. We discovered that hop cone powder exhibits potent deodorizing activity against allyl methyl sulfide. Oxidation products of the hop bitter acids humulinone and hulupone were detected in a partially purified sample of hop cone powder. Oxidation products of the α-acids cohumulinone and n-humulinone showed approximately 10- and 15-fold stronger deodorizing activity than the parent α-acids, respectively. The deodorizing activity of oxidation products of β-acids was comparable to that of n-humulinone. It is presumed that the oxidation products of hop powder play an important role in the strong deodorizing activity of hop cone powder against allyl methyl sulfide.

1. Introduction

The number of odor compound has been estimated at over 400,000 and malodorous gases have a variety of chemical structures, which include nitrogen compounds, sulfur compounds, aldehydes, esters and short-chain fatty acids. On the other hand, we have several deodorization methods, which can be divided into four categories according to the mechanism of deodorization: physical methods such as activated carbon to adsorb malodorous gases, biological methods to control odor-producing bacteria using antimicrobial agents, chemical methods to chemically deactivate malodorous gases and sensory methods to mask unpleasant odors with fragrances. As an example of a chemical deodorization mechanism, Yasuda and Arakawa showed that polyphenol compounds exhibited strong deodorizing activity against thiol compounds and proposed that the mechanism is an addition reaction between thiol compounds and polyphenol ones [1]. We reportedly exhibited that 1,8-cineole reduced odor gas concentrations by adsorption through physical interaction with odor gas molecules [2].

Regarding to unpleasant food odors, they can result from microbial growth, lipid oxidation, and enzyme-catalyzed degradation of food components. Perhaps the most common example of enzymatic degradation producing a foul odor is that of garlic. Garlic (Allium sativum L.) has been used as a seasoning worldwide since ancient times. However, ingestion of garlic can cause unpleasant halitosis and body odor, resulting in personal discomfort and social embarrassment. Previous studies have identified several causative sulfur-containing compounds responsible for the unpleasant halitosis and body odor associated with garlic, including diallyl disulfide, allyl methyl disulfide, diallyl disulfide, diallyl trisulfide, and allyl methyl sulfide (AMS) [3,4,5,6]. The first four volatile compounds are detected at high concentrations in human exhaled breath immediately after garlic ingestion and decrease to baseline levels within 2–3 h. By contrast, AMS appears rapidly in human exhaled breath and can persist for up to 30 h [7]. This organosulfur compound is believed to be the primary cause of the unpleasant odor caused by garlic consumption.

Hops, derived from the hop plant (Humulus lupulus L), are an essential component of beer brewing, along with malt, yeast, and water. Hops play important roles in brewing by providing a favorable bitterness and attractive aroma and maintaining beer foam. Hops also inhibit bacterial growth and function as a natural preservative [8]. These functions are attributed to the lupulin gland inside the hop cone.

In addition to antibacterial activity, hop extract reportedly exhibits a wide range of other physiologic activities, including anti-oxidant, anti-inflammatory, anti-cancer, anti-obesity, and other clinically important activities [9].

Bitter acids are a class of bitter compounds that consist primarily of two related series: α-acids (humulones) and β-acids (lupulones). The major α-acids include n-humulone (or humulone), cohumulone, and adhumulone, whereas the corresponding β-acids include n-lupulone (or lupulone), colupulone, and adlupulone (Figure 1).

Our screening program to isolate natural products exhibiting deodorizing activity previously revealed that 1,8-cineole exhibits potent deodorizing activity against various odors [2]. As AMS causes unwanted halitosis and body odor, we sought to identify compounds with deodorizing activity against AMS and developed a new deodorant against the unpleasant odor of garlic. Here, we report newly discovered deodorant substances from the hop plant and discuss the possible mechanism of their deodorizing activity against AMS.

2. Materials and Methods

2.1. Chemicals and Materials

Hop cone powder was obtained from Nippon Funmatsu Yakuhin Co., Ltd. (Osaka, Japan). Silica gel (Wakosil C-300) was purchased from FUJIFILM Wako Pure Chemical Industries (Tokyo, Japan). AMS was obtained from Alfa Aesar (Ward Hill, MA, USA). Hop acid standard (α-acids and β-acids mixture, ICE 4, composed of a mixture of 42.58% α-acids and 26.54% β-acids) and humulinones standard (ICS-Hum 1) were obtained from Labor Veritas AG (Zurich, Switzerland). Xanthohumol C was obtained from MedChem Express (Princeton, NJ, USA). Allyl methyl sulfoxide (AMSO) was purchased from aromaLAB GmbH (Martinsried, Germany).

2.2. Analysis of Deodorizing Activity

For gas chromatography (GC) analysis of deodorizing activity, a predetermined amount of either samples suspended in 1 mL of deionized water or freeze dried powders alone, was placed in a 30-mL glass vial with a butyl rubber cap (Nichiden Rika Glass Co., Kobe, Japan), to which 2 μL of 0.01% AMS was added using a microliter syringe (Hamilton Company, Reno, NV, USA), and the mixture was allowed to stand for a predetermined time (usually 2 h, reaching almost the maximum level of the activity) at 25 °C, after which 2 mL of headspace gas was analyzed on a Shimadzu (Kyoto, Japan) GC-2014AF GC instrument. Deodorizing activity is expressed as the percentage decrease in odor gas concentration in the headspace of the glass vial and calculated according to the following equation: deodorizing activity (%) = (C − S)/C × 100, where C represents the peak area of AMS in the headspace of the reaction mixture without test sample (control), and S represents the peak area of the reaction mixture containing the test sample.

2.3. Determination of Deodorant Activity from Hop Corn Powder

A total of 200 g of finely ground hop cone powder was macerated overnight in 50% aqueous ethanol at room temperature and then filtered. The filtered extract was concentrated under vacuum to obtain 20 g of crude hop extract. To the crude extracted material, 200 mL of methanol was added and stirred with a magnetic mixer for 10 min. The sample was then centrifuged (1500× g) for 10 min, and the supernatant was collected and concentrated under vacuum. A total of 17.3 g of methanol-soluble material showing AMS deodorizing activity was obtained, of which the extraction yield was 86.5%. The methanol soluble fraction (8.1 g) was subjected to silica gel column chromatography (Wakogel^® C-300, silica gel; 170 mL, column size; 4.2 cm id × 25 cm) and eluted with hexane/ethyl acetate (v/v) mixture as the eluent to provide Fraction I (100:15, 250 mL), Fraction II (100:30, 500 mL), Fraction III (100:50, 200 mL), Fraction IV (100:60, 300 mL), and Fraction V (100:70, 500 mL). Each fraction was concentrated under vacuum and assayed for deodorizing activity.

The second round of separation was performed using the same silica gel chromatography system, as follows (column size: 2 cm id × 30 cm). The silica gel column was loaded with material from Fraction II (1.1 g) and successively fractionated with hexane/ethyl acetate in the ratios 100:20 (v/v; 400 mL), 100:30 (v/v; 100 mL), 100:50 (v/v; 100 mL), and a total of 100 fractions (4 g/fraction) were obtained. Strong deodorizing activity was observed in fractions 19–23, 25–29, 31–33, 61–64, and 70–74. The main fractions were combined to prepare fractions A (19–23), B (25–29), C (31–33), D (61–64), and E (70–74), respectively. The extraction and fractionation procedures are shown in Figure 2.

2.4. Preparation of α-Acids Fraction

Samples of α-acids were prepared according to the protocol reported by Taniguchi et al. [10]. A total of 20 g of α-acids and β-acids mixture standard (hop acid standard) was dissolved in 200 mL of hexane and partitioned using 0.24 M disodium carbonate (200 mL). The α-acids were then selectively extracted into aqueous solution. The aqueous solution was acidified using 6 N HCl, and free α-acids were extracted using hexane. After washing with saturated NaCl, the hexane layer was dried over anhydrous sodium sulfate and concentrated to dryness to give the α-acids fraction (9.71 g).

2.5. Synthesis of Humulinones

Humulinones were prepared from the α-acids fraction using the method described by Cook et al. [11]. The α-acids fraction (2.05 g) and cumene hydroperoxide (1 mL) were dissolved in diethyl ether (20 mL), and 35 mL of saturated aq. sodium bicarbonate was added to the solution. The resulting bi-layer that formed was kept at room temperature for 4 days in a sealed flask. Under these conditions, sodium salts of humulinones deposited at the border of the bi-layer. These sodium salts were filtered and washed using cold diethyl ether and water. The washed sodium salts were then dissolved in 60 mL of methanol containing 1% phosphoric acid, and then 600 mL of 0.5 N HCl was added to the solution, which was then partitioned twice with 600 mL of hexane. The final hexane layer was concentrated to dryness to give humulinones (300 mg).

2.6. Instrumentation and Measurement Conditions

An LC-20 AT Prominence HPLC system (Shimadzu, Kyoto, Japan) was used to analyze the hop components in fraction E, in which the deodorizing activity was greater than that of other fractions. Elution was performed using a Shim-pack Velox C18 column (50 mm × 3.0 mm id, particle size 1.8 µm; Shimadzu). The mobile phases consisted of ultrapure water (solvent A) and methanol (solvent B), with an elution program of A:B (95:5) for 10 min, followed by 5:95 for 10 min. The column oven was maintained at 45 °C, the flow rate at 0.2 mL/min, and the injection volume of MeOH solution containing 1% fraction E was 2 µL.

Mass spectrometry (MS) analysis of the components eluted from HPLC separation was performed using a Q Exactive instrument (Thermo Fisher Scientific, San Jose, CA, USA) equipped with an electrospray ionization (ESI) source running protonated or deprotonated molecular ions. The heated capillary temperature and probe heater temperature were set at 330 °C. Full-scan mass acquisition was performed by scanning the range m/z 100–1500. Compounds were identified by comparison of both UV and ESI/MS spectra with literature values or by comparison with authentic standards.

For purification of bitter acids and their oxidation products, gradient elution was performed using an LC-20 AT Prominence HPLC system (Shimadzu) equipped with an SPD-M20A Prominence diode array detector. A Shim-pack Scepter C18 column (10 mm id × 250 mm, particle size 5 µm) was used as the semi-preparative HPLC column, and a Shim-pack Scepter C18 column (4.6 mm id × 250 mm, particle size 5 µm) was used to evaluate the purity of each compound.

GC/MS analyses were performed using a GCMS-TQ8040 gas chromatograph (Shimadzu) as follows: column, Intercap Pure-WAX (0.25 mm id × 30 m, 0.25 µm [GL Sciences, Tokyo, Japan]); temperature, 50 °C; first 5.0 min, ramp to 250 °C (20 °C/min) with a 10-min hold; carrier gas, helium, 1.43 mL/min; injection volume of prepared sample, 1 µL; GCMS interface temperature, 250 °C; ion source temperature, 200 °C; scan mode, Q3 scan, run at 1 mL/min for 10 min.

2.7. Separation of Individual α-Acids

A total of 2 mL of methanol solution of the α-acids fraction at a concentration of 100 mg/mL was prepared, and after filtration, individual α-acids (cohumulone and n-humulone) were isolated by HPLC using semi-preparative columns. Chromatography was performed using a gradient of a 1000/0.2 (v/v) mixture of water and phosphoric acid (85%) (containing 0.02% [w/v] EDTA) (solvent A), acetonitrile (solvent B), and water (solvent C). The gradient was programmed as follows: 0–26.67 min, 10% B in A; 26.67–30 min, 52% B in A; 30–32.67 min, 75% B in A; 32.67–37.67 min, 85% B in A; 37.67–41.31 min, 10% B in C; 41.31–51 min, 10% B in A; 51 min, elution stopped. The flow rate of the mobile phase was 5 mL/min, and the column oven temperature was maintained at 40 °C. Forty microliters of sample was repeatedly injected. Elution was monitored using a photodiode array detector in scan mode from 200 nm to 700 nm. Cohumulone and n-humulone eluted at 37.0 and 38.0 min, respectively. To remove phosphate and EDTA, each eluate was diluted (5×) with distilled water (HPLC grade), applied to an Oasis HLB column (Waters, Milford, MA, USA), developed with water, and eluted with methanol (HPLC grade).

2.8. Separation of Individual α-Acid Oxidation Products

Each α-acid oxidation product (cohumulinone and n-humulinone) was isolated from humulinones synthesized from the α-acids fraction by repeated semi-preparative HPLC in the manner used for the isolation of α-acid congeners.

2.9. Separation of Individual β-Acids and Their Oxidation Products

Purified colupulone and n-/adlupulone (overlap) were obtained from α-acids and β-acids mixture standard using the following method: 100 mg of mixture standard was dissolved in 1 mL of methanol and subjected to repeated semi-preparative HPLC. Chromatography was performed as follows: sample injection volume, 50 µL; mobile phase flow rate, 4 mL/min, with the following gradient solvent elution program: 0–5 min, 10% solvent B in solvent A; 5–7.67 min, 52% B in A; 7.67–12.67 min, 75% B in A; 12.67–33 min, 85% B in A; 33 min, elution stopped. Colupulone eluted at 40.2 min and n-/adlupulone at 41.4 min. Phosphoric acid and EDTA in each eluate were removed using an Oasis HLB column, as described in the Separation of individual α-acids section.

β-Acid oxidation products (cohulupone and n-/adhulupone) were prepared as follows. First, 100 g of mixture standard was stored at 60 °C for 120 h and then extracted with ethanol (EtOH; 500 mL). The resulting extract (24.5 g) was subjected to Diaion HP-20 (Mitsubishi Chemical, Tokyo, Japan) column chromatography (column size; 3.0 cm id × 25 cm) and then eluted stepwise with 10% EtOH (1 L, 40 mg), 30% EtOH (1 L, 60 mg), 50% EtOH (4 L, 170 mg), 60% EtOH (4 L, 12.71 g), and 80% EtOH (2 L, 5.19 g). Hulupones (170 mg) eluted in the 50% EtOH fraction. UV spectral data for the 50% EtOH fraction showed λ_max values of 256 and 328 nm, identical to those reported in the literature for hulupones [12]. Purified cohulupone and n-/adhulupone (overlap) were isolated using the following method: 100 mg of the 50% EtOH elution sample was dissolved in 1 mL of methanol and subjected to repeated semi-preparative HPLC. Chromatography was performed as follows: sample injection volume, 50 µL; mobile phase flow rate, 4 mL/min, with the following gradient solvent elution program: 0–5 min, 10% solvent B in solvent A; 5–7.67 min, 52% B in A; 7.67–12.67 min, 75% B in A; 12.67–33 min, 85% B in A; 33 min, elution stopped. Cohulupone eluted at 14.2 min and n-/adhulupone at 14.9 min. Phosphoric acid and EDTA in each eluate were removed using an Oasis HLB column, as described in the Separation of individual α-acids section.

2.10. GC-MS/MS Analysis of the Deodorizing Effect of n-Humulinone Against AMS

The products of the reaction of AMS with n-humulinone were identified based on comparison of the mass spectra of the target compounds in the reaction with those of reference compounds. A 300-µL volume of methanolic solution of n-humulinone (10 mg/mL) was placed in a 30-mL GC vial. After air drying, 100 µL of methanol was added and mixed with 1 µL of AMS (750 µg/mL), and the mixture was vortexed for 5 s and then incubated for 2 h at 25 °C.

The deodorizing effect of n-humulinone against AMS was analyzed by GC–tandem mass spectrometry (MS/MS) using a Shimadzu GCMS-TQ8040 instrument. An InterCap Pure-WAX column (0.25 mm id × 30 m, film thickness 0.25 µm) was used as the analytical column. The GC oven program utilized an initial oven temperature of 50 °C, which was held for 5 min and then raised at 20 °C/min to 250 °C and finally held for 10 min. Helium as the carrier gas was set at a flow of 1.43 mL/min. The injection temperature was set at 250 °C, and the injection volume was 1 μL. Detector parameters used for GC-MS/MS analyses were as follows: interface temperature, 250 °C; ion source temperature, 200 °C. MS was performed using Q3 scan with a scanning range of m/z 2–500.

3. Results and Discussion

3.1. Potent Deodorizing Compounds Against AMS Extracted from Hop Con Powder

We conducted a screening program of natural materials for potential use as deodorants. For this purpose, we screened hydroethanolic extracts of more than 200 natural materials. The screening results showed that a hydroethanolic extract of dried hop powder exhibited potent deodorizing activity against AMS. The activity of this extract was 61% at a concentration of 1 mg/30-mL glass vial in the assay system. To obtain a more purified active principle, the methanol soluble fraction obtained from the hydroethanolic extract, which was 96% (1 mg/vial) active, was subjected to silica gel chromatography and fractionated using a hexane/ethyl acetate mixture. Among fractions I–V, the deodorizing activity against AMS peaked in Fraction II (eluent, hexane/ethyl acetate [100:30]), with 97.2% activity at 0.5 mg/vial and 46.5% at 0.1 mg/vial. Fraction II was further purified using the same silica gel chromatography system to obtain five active fractions, A, B, C, D, and E.

Table 1. Deodorizing activity and mass of each silica gel chromatography fraction, in grams.

Fraction	Deodorizing Activity *			Mass of Fraction (mg)
	Concentration (mg/Vial)
	0.05	0.1	0.5
A	28.5	66.0	96.7	63.9
B	37.3	50.4	97.8	63.5
C	49.5	57.9	98.0	35.4
D	60.5	85.6	98.0	9.6
E	60.4	87.9	>99.0	19.0

* Deodorizing activity was defined as percent reduction in headspace odorant concentration following deodorant addition in vitro.

shows the deodorizing activity and mass of each fraction.

To determine the qualitative chemical profile of the active principle, fraction E (which exhibited higher deodorant activity than the other fractions) was analyzed by LC-MS/MS. The 220-nm UV chromatogram of fraction E showed two dominating peaks at 2.1 min (Peak 1) and 15.8 min (Peak 2) (Figure 3).

Table 2. LC/MS/MS data for the constituents found in peaks 1 and 2 of silica gel column chromatography fraction E.

LC/UV Chromatogram Peak	Compound	MW	Molecular Formula	Molecularion [M−H]− or [M+H]+ (m/z) *	MS/MS Fragmentions [M−H]− or [M+H]+ (m/z) *	Proposed Structure
Peak 1 (RT, 2.8)	1	318	C9H26O4	317.1754	248.1054 233.0817 205.0510 180.0432 111.0455	Cohulupone
Peak 2 (RT, 15.8)	2	278	C18H30O2	279.2317	173.1245 109.0880 95.0771	Unknown
	3	352	C21H20O5	353.1383	233.0808 191.0294 109.0517 81.0655	Xanthohumol C
	4	448	C25H36O7	449.2534	381.1907 347.1854 265.1070 215.0703	Unknown
	5	364	C20H28O6	363.1816	249.1129 209.1180 179.0712 139.0763 111.0450	Cohumulinone
	6	434	C25H38O6	433.2601	276.1369 233.0615	Unknown

*: MS and MS/MS analyses of compounds 1, 5, and 6 were performed in the negative ion mode, and those of compounds 2, 3, and 4 were performed in positive ion mode. RT: retention time.

shows the MS and MS/MS data for the compounds in Peaks 1 and 2, in which one and five precursor ions were detected, respectively.

Of the six compounds detected in fraction E, compounds 2, 4, and 6 were not assigned, whereas compounds 1, 3, and 5 were presumed to be hulupone, xanthohumol C, and humulinone, respectively. The m/z value of 317.1754 for the deprotonated molecular ion of compound 1 was in good agreement with the m/z value of 317.2 for the molecular ion of cohulupone reported by Garcia-Villalba et al. [13]. On the other hand, the MS/MS spectrum showed major peaks at m/z 248.1054, 233.0817, 205.0510, 180.0432, and 111.0455. These fragment ion spectra were consistent with the literature data for cohulupone [14]. The fragment ions at m/z 248.1054, 233,0817, and 205.0510 were deduced to be [M-H-C₅H₉]⁻, [M-H-C₅H₉-CH₃]⁻, and [M-H-C₅H₉-C₃H₇]⁻, respectively.

For compound 3, the m/z value for the protonated molecular ion of m/z 353.1383 corresponded well with that of xanthohumol C (molecular formula, C₂₁H₂₀O₅), whereas the MS/MS spectrum revealed the presence of a peak at m/z 233.0808 (molecular formula, C₁₃H₁₂O₄), for which fragment ions were speculated to be due to cleavage of the intermolecular C-C linkage present in the chalcone moiety of xanthohumol C (Figure 1e). Xanthohumol is the most abundant prenylated flavonoid in the hop plant, and several other prenylated flavonoids derived from hops are known, including xanthohumol C [15].

Compound 5 exhibited a deprotonated molecular ion at m/z 363.1816, indicating the molecular formula C₂₀H₂₇O₆, which corresponds with that of cohumulinone [13]. By contrast, MS analysis indicated that the molecular ion of pure cohumulinone separated from the international calibration standard, DCHA-Humulinones ICS-Hum1, was m/z 363.1811. The MS/MS spectra were composed of the following fragment ions: m/z 249.1129, 179.0712, 167.1075, 139.0763, 111.0450, and 85.0657. Among these MS/MS ions, four fragment ions, m/z 249.1129, 167.1075, 139.0763, and 111.0450, corresponded to those of cohumulinone (m/z 249, 167, 139, and 111), as reported by Taniguchi et al. [16].

We noted that fraction E, which exhibits strong deodorizing activity, contained xanthohumol C and bitter acid-related compounds such as humulinones and hulupones. Therefore, we first examined the deodorizing activity of the bitter acids, which consisted of a mixture of α-acids and β-acids and xanthohumol C. Xanthohumol C showed low levels of deodorizing activity, 0% at 0.3 mg/vial, 6.1% at 1 mg/vial, and 31.3% at 3 mg/vial. By contrast, a mixture of α-acids and β-acids showed 25.8% deodorizing activity at 1 mg/vial and 50.5% at 3 mg/vial, which were relatively low activity levels.

It is well known that oxidation products of bitter acids are formed in hops stored for a long period. Taniguchi et al. developed a simple autoxidation model of hops and showed that humulinone and hulupone are the major products of oxidation. Furthermore, they showed that α-acids and β-acids decrease to <5% of their initial amounts after storage at 60 °C for 96 or 72 h [10].

The effects of storage temperature and storage duration on the deodorizing activity of the mixture standard against AMS were also investigated. As can be seen from Figure 4a, after 48 h of storage at a concentration of 10 mg/vial, the deodorizing activity increased from 48% at 20 °C to 85% at 60 °C.

Figure 4b shows a time-course analysis of the change in deodorizing activity of the mixture standard at 60 °C. Deodorizing activity increased with time, reaching a maximum level after 72 to 96 h. At that point, >95% of the bitter acids were potentially transformed into oxidation products, as described earlier in this section [10]. These data indicate that the oxidation products obtained from the mixture standard exhibit stronger deodorizing activity than the parent compounds.

Deodorizing activity analyses were conducted as described in the Section 2. Values are expressed as the mean of two independent assays.

Because the deodorizing activity of mixtures of α-acids and β-acids is enhanced by their oxidative transformation, the deodorizing activity of the individual compounds was of particular interest. The initial focus was on the separation of each congener of α-acids and β-acids from the mixture standard. The HPLC chromatogram of the mixture standard showed five major peaks corresponding to the main components of α-acids and β-acids (Figure 5): cohumulone (P1), n-humulone (P2), adhumulone (P3), colupulone (P4), and n-lupulone/adlupulone (P5). Each congener of α-acids and β-acids was identified by comparing the order of HPLC elution with that of compounds described in the literature and by comparing m/z values with those of standards ([17] and

Table 3. MS parameters of α-acids, β-acids, and their oxidation products separated. Using semi-preparative HPLC.

Compound	Molecular Formula	MS (m/z) *
		Found	Reported **
cohumulone	C20H28O5	346.60	347.3
n-humulone	C20H30O5	360.66	361.2
cohumulinone	C20H28O6	363.18	363.2
n-humulinone	C21H30O6	377.19	377.3
colupulone	C26H38H4	399.25	399.3
n-/adlupulone	C26H38O4	412.66	413.3
cohulupone	C19H26O4	316.60	317.2
n-/adhulupone	C20H28O4	330.62	331.2

* Negative ion mode; ** Garcia-Villalba et al. [13].

).

In order to compare the deodorizing activities of individual pure α-acid components with those of the corresponding oxidation products, pure samples of cohumulone and n-humulone were isolated by repetitive semi-preparative HPLC from the α-acids fraction prepared from the mixture standard. Under the chromatographic conditions described in the Section 2, 30 mg of pure cohumulone (purity; 98.4%) and 109 mg of pure n-humulone (purity; 99.5%) were obtained.

Cohumulinone and n-humulinone were also isolated from humulinones synthesized from the α-acids fraction. The yields of pure cohumulinone and n-humulinone were 36 mg (purity; 98.2%) and 40 mg (purity; 98.4%), respectively. In the same procedure, pure samples of colupulone and n-/adlupulone were isolated from the mixture standard, yielding 22 mg of colupulone (purity; 98.1%) and 24 mg of n-/adlupulone (purity; 98.5%).

Purified cohulupone and n-/adhulupone were isolated from oxidation products of the mixture standard using Diaion HP-20 column chromatography, followed by repetitive semi-preparative HPLC. Hulupones (170 mg) eluted in the 50% EtOH fraction of the chromatogram, yielding 25 mg of cohulupone (purity; 93.1%) and 26 mg of n-/adhulupone (purity; 93.5%). The purity of the oxidation products of the lupulones was inferior to that of the corresponding parent compounds. This was attributed to the presence of minor peaks, possibly decomposition products, that partially overlapped with hulupone peaks observed in the chromatogram of the 50% EtOH fraction.

The m/z values for the deprotonated molecular ions of cohumulone, n-humulone, colupulone, and n-/adlupulone and their oxidation products are shown in Table 3. These m/z values were in good agreement with those described in the literature [13].

The deodorizing activity of the major α-acid components and their oxidation products was investigated using isolated pure samples. As shown in Figure 6, the oxidative conversion of cohumulone and n-humulone significantly enhanced the deodorizing activity against AMS. This enhancement was dose dependent across the concentration range tested. The DA30 value, the deodorant concentration providing a 30% decrease in the odor gas concentration, was approximately 10-fold lower for cohumulinone than cohumulone, and the DA50 value was approximately 15-fold lower for n-humulinone than n-humulone (Figure 6a,b). These data indicate that the deodorizing activity against AMS of the different α-acid congeners and their oxidation products differs. These data also clearly indicate that the oxidation products of α-acids, especially n-humulinone, play an important role in the deodorization of AMS.

The deodorizing activity of individual lupulones and their oxidation products, hulupones, was also examined. Test samples were prepared at ambient temperature (25 °C) and analyzed for deodorizing activity according to the method described in the Section 2. The deodorizing activity of colupulone and n-/adlupulone at a concentration of 3 mg/vial was 11% and 15%, respectively, and that of cohulupone and n-/adhulupone was 27.5% and 36.0%, respectively. As in the case of the α-acids, oxidative transformation of the β-acids enhanced the deodorizing activity against AMS. However, the level of enhancement was rather low compared with that of α-acids.

Other studies have shown that lupulone is unstable and readily oxidizes to hulupone in the presence of air [18,19]. High sensitivity of hulupone to oxidation has also been reported [20]. Furthermore, these compounds are temperature sensitive [19]. In view of the ready oxidation of lupulones, the behavior of lupulones during the analysis of deodorizing activity was examined. In order to measure the deodorizing activity under conditions that minimize the degradation of lupulones, colupulone (98.1% purity) and n-/adlupulone (98.5% purity) were each dissolved in cold methanol (concentration 3 mg/mL), and 3-fold serial dilutions were then prepared. Each cold methanol solution was kept in the dark in a refrigerator for 48 h to remove residual methanol (usually methanol was removed at ambient temperature), and then the respective deodorizing activity was determined. As shown in Figure 7, both lupulone-derived compounds exhibited strong deodorizing activity against AMS, comparable to that of n-humulinone.

The effect of temperature on lupulone-derived compounds in the above analysis was quantitatively investigated using HPLC. The HPLC peak for lupulone-derived compounds was unexpectedly split into three peaks: cohulupone (RT, 32.2 min; [M−H]⁻, 317.1754; peak area, 51%), unknown #1 (RT, 33.2 min; peak area, 32%), and colupulone (RT, 40.5 min; [M−H]⁻, 399.2536; peak area, 13%). The n-/adlupulone peak was also split into three peaks: n-/adhulupone (RT, 33.1 min; [M−H]⁻, 331.1912; peak area, 26%), unknown #2 (RT, 8.0 min; peak area, 17%), and the parent compound, n-/adlupulone (RT, 41.8 min; [M−H]⁻, 413.2695; peak area, 57%) (Figure 8).

It is possible that oxidation of lupulones occurred during the deodorizing activity measurement and resulted in the formation of hulupones and unknown compounds. The parent compounds, colupulone and n-/adlupulone, were confirmed by their m/z values in the negative ion mode, at 399.2536 and 317.1754, respectively. These results suggested that the oxidation products of lupulone play a dominant role in the higher deodorizing activity against AMS.

The deprotonated molecular ions of unknown #1 and unknown #2 were m/z 416.2563 and m/z 430.2719, corresponding to those of hydroxytricyclocolupone and hydroxytricyclolupone, respectively. In addition, it was further inferred that the ions of hydroxytricyclocolupone at m/z 346.1783, 303.1236, and 287.1287 represented the [M-H-C₅H₉]⁻, [M-H-C₃H₇-C₅H₉]⁻, and [M-H-C₃H₇O-C₅H₉]⁻ fragment ions, respectively. In the same way, the ions of hydroxytricyclolupone at m/z 360.194, 317.1391, and 301.1443 represented the [M-H-C₅H₉]⁻, [M-H-C₃H₇-C₅H₉]⁻, and [M-H-C₃H₇O-C₅H₉]⁻ fragment ions, respectively.

In the presence of oxygen, β-acids are transformed into a large number of lupulone derivatives [21,22]. Haseleu, Intelmann, and Hofmann, proposed a reaction route leading from colupulone to the tricyclic β-acid transformation products, including the above-mentioned oxidation products [21]. Based on combined analysis of the precursor ion data, product ion data, and literature information, unknowns #1 and #2 were tentatively identified as hydroxytricyclocolupone and hydroxytricyclolupone, respectively. Although the deodorizing activity of these two oxidation products of lupulones was of interest, the low yields did not allow for successful analysis of the activity.

Hop corn contains significant amounts of polyphenolic compounds, including catechins, proanthocyanidins, and tannins [23], which exhibit strong deodorizing activities against methyl mercaptan. However, in the present assay system, these compounds showed no deodorizing activity against AMS at a concentration of 3 mg/vial.

3.2. Deodorizing Mechanism of Humulinone Against AMS

As mentioned earlier, the deodorization mechanism of polyphenolic compounds against thiol gas has been proposed to involve the addition reaction of thiol compounds to polyphenolic compounds [1]. Against such a background, the mechanism by which the concentration of AMS is reduced by oxidation products of bitter acids is of interest. In order to investigate the deodorization mechanism of α-acid oxidation products against AMS, products from the reaction of n-humulinone with AMS were analyzed using GC-MS/MS. The chromatogram of the reaction solution containing n-humulinone and AMS was compared with that of a solution containing AMS alone (control) to determine if new peaks were apparent in the former solution. A new peak detectable only in the mixture of n-humulinone and AMS was observed (Figure 9). The chromatogram showed an additional peak at m/z 104, for which the molecular ion was consistent with that of AMSO. Furthermore, the fragmentation spectrum of the newly appearing peak was compared with that of authentic AMSO, and a good match was obtained. This suggests that AMS was converted to AMSO.

Identification of AMSO was performed via GC-MS/MS. Comparison of the mass spectra of AMSO with the reference standard was performed with the aid of the NIST 14 Mass Spectral Library.

Scheffler et al. investigated the odor of breast milk and urine after consumption of raw garlic by nursing mothers and identified AMS as the major odor metabolite. They also identified two metabolites, AMSO and allyl methyl sulfone (i.e., AMSO₂), which were odorless, as determined by GC-olfactometry [24,25]. They assumed that these two metabolites were oxidation products of AMS.

Shimizu et al. reported that water extracts of Rubus suavissimus (RSW) exhibit potent deodorizing activity against AMS [26]. They demonstrated that the partially purified fraction of RSW with polyphenolic compounds exhibits strong deodorizing activity against AMS. The activity disappeared under anaerobic conditions, suggesting that the deodorant mechanism involves the oxidative conversion of AMS to AMSO, mediated by polyphenolic compounds. Based on literature reports and GC-MS/MS spectra data for a reaction mixture of n-humulinone with AMS, we hypothesized that AMS is converted to the odorless compound AMSO.

Taniguchi et al. identified 4-hydroxyallohumulinones as oxidation products of humulinones and proposed a plausible reaction pathway to 4-hydroxyallohumulinones and/or 4-perhydroxyallohumulinones [10]. We considered that the latter peroxides could oxidize AMS to give the corresponding sulfoxide (AMSO) (Figure 10).

We confirmed the presence of 4′-hydroxyallohumulinone in the reaction solution of AMS with n-humulinone by HPLC analysis. The LC/UV (240 nm) chromatogram showed small peaks at RT 14.27 min (compound A) and RT 15.46 min (compound B) (chromatogram not shown). Compound A was examined by LC/MS/MS, and the deprotonated molecular ion showed m/z 393.1918, indicating the molecular formula as C₂₁H₃₀O₇. Further MS/MS analysis gave four fragmental peaks (m/z 393.1920, 263.1286, 223.1337, and 125.0606). The pattern corresponded to the data for 4′-hydroxyallohumulinone (m/z 393.19, 263.13, 223.13, and 125.06) reported by Ferreira and Collin [27]. Based on these findings, compound A was identified as 4′-hydroxyallohumulinone.

Compound B was also examined by LC/MS/MS, and the deprotonated molecular ion showed m/z 409.1864, indicating the molecular formula as C₂₁H₃₀O₈ [compound A+O]. The MS/MS chromatogram gave four peaks, which were also observed in the MS/MS analysis of compound A, and an additional three peaks (m/z 409.1865, 377.1603, and 295.1182). Among these additional peaks, the peak of m/z 377.1603 could be assigned by O₂ dissociation from the parental ion. Compound B had a structure closely related to that of compound A, but no data for a direct comparison have been reported.

In contrast, Intelman and Hofmann identified hydroxyl-alloisohumulones (compound C) and hydroperoxy-alloisohumulones (compound D) as the oxidative products of iso-α-acids, and proposed the reaction pathway [28]. According to their MS/MS data of compound D, the presence of a fragmental peak, formed by O2 cleavage from the parental ion, strongly suggested that compound D was the hydroperoxy analog of compound C. Therefore, we considered that compound B was also the hydroperoxy analog of compound A.

Figure 10 depicts the plausible deodorizing mechanism of n-humulinone against AMS. Unfortunately, however, we could not isolate sufficient amounts of compound A and B to analyze their deodorizing activity. Further research will be required to establish the validity of our hypothetical deodorization mechanism of bitter acid–derived compounds against AMS.

4. Conclusions

As mentioned in the Introduction, bitter acids reportedly exhibit multiple bio-active effects that could be beneficial in the medicine and pharmaceutical fields. Here, we present a new biological property of bitter acids and their derivatives, that is, potent deodorizing activity against AMS. The oxidative transformation products of these bitter acids in particular can be considered the major constituents responsible for this activity, which could be exploited to control unpleasant breath and body odor resulting from the consumption of garlic.

5. Patents

Japan Patent; Hop-derived deodorant (P7578312).

Deodorant with hop cone as the main ingredient. Patent No. P7578312.