Identification of antioxidants in dark soy sauce

1.4 The balance of free radicals/reactive species and antioxidants 6 2.3 Isolation of low molecular mass components from ethyl acetate extract 9 2.9 Detection and determination of maltol

IDENTIFICATION OF ANTIOXIDANTS IN DARK SOY SAUCE

Acknowledgements

This journey is not possible without the help of so many people First and foremost, I would especially like to express my most sincere and profound appreciation to my supervisor, Professor Barry Halliwell, for his constant guidance, invaluable

suggestion and critical comments throughout this work

I am also grateful to all my colleagues and friends inside/outside Professor

Halliwell’s group, who help me in this way or that way, especially to Dr Tang Soon Yew for his useful discussions and many technical support in cell culture, Dr Andrew Jenner and Dr Lee Chung-Yung for their useful suggestions, Dr Shui Guanghou, Professor Markus Wenk’s group, for his help in using LC-TOF-MS, Dr Koh Hwee Ling, Department of Pharmacy, for her help in using FT-IR, Dr Mark Richards, Agilent Technologies Singapore, for his help in LC-APCI-Ion trap -MS/MS

measurement, and Professor Yong Eu Leong, Department of Obstetrics &

Gynecology, for his help in using triple quadrupole LC-MS/MS

My heartfelt thanks also go to the Department of Biochemistry, the Centre of Life Sciences, the National University of Singapore for holding so many exciting talks and creating a wonderful academic atmosphere

Last but not least, I thank my wife, Shen Ping, for her patience and consistent support

I would like to dedicate this thesis to my son, Bo Qian

1.4 The balance of free radicals/reactive species and antioxidants 6

2.3 Isolation of low molecular mass components from ethyl acetate extract 9

2.9 Detection and determination of maltol metabolites in human urine samples 13

2.9.1 Standard preparation 13

2.9.3 HPLC-DAD detection of maltol metabolites and determination

2.9.4 HPLC-MS/MS detection of maltol metabolites 15

4.1 Separation and characterization of low molecular mass components 19 4.2 Content of maltol and its contribution to the total antioxidant activity

Summary

Soy sauce is a traditional fermented seasoning in Asian countries, that has high

antioxidant activity in vitro and some antioxidant activity in vivo We attempted to

identify the major antioxidants present, using the

2,2’-azinobis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) assay as a guide 4H-pyran-4-one (maltol) was one of several active compounds found in an ethyl acetate extract of dark soy sauce (DSS) and was present at millimolar concentrations

3-Hydroxy-2-methyl-in DSS However, most of the antioxidant activity was present 3-Hydroxy-2-methyl-in colored fractions, two of which (CP1 and CP2) were obtained by gel filtration chromatography Their structural characteristics based on nuclear magnetic resonance (NMR) and

electrospray-ionization time-of-flight mass spectrometry (ESI-TOF-MS) analysis suggest that carbohydrate-containing pigments such as melanoidins are the major

contributors to the high antioxidant capacity of DSS In vitro, maltol, CP1 and CP2

can protect against hypochlorous acid (HOCl)-mediated DNA damage

dose-dependently Furthermore, dark soy sauce potentially inhibits the growth of colon cancer HT 29 cells at high concentrations, while it decreases the up-regulation of cyclooxygenase-2 (COX-2) expression in LPS (lipopolysaccharide) -induced HT 29 cells at low concentrations

List of Tables

Table II Within- and between-assay precision and recoveries of the assay

Table III The observed ions in TOF-MS spectrum of CP1 Three series were

observed Within each series, m/zs of the doubly-charged or the singly-charged

Table IV Inhibition of HOCl-induced DNA damage by maltol, the colored

List of Figures and Chart

Figure 1 Trolox equivalent antioxidant capacity (TEAC) values

per g/ml of dark soy sauce (DSS) and three fractions: methanol

extract residue (MeOH-R), ethyl acetate extract (EtOAc-extract)

and ethyl acetate extract residue (EtoAc-R) Values are mean ±

Figure 2 (A) Typical HPLC chromatogram of ethyl acetate extract

The absorbance was monitored at 270 nm (B) Trolox equivalent

antioxidant capacity (TEAC) values of 10 fractions of ethyl acetate

Figure 3 EI-MS spectrum of compound 1 and the proposed mechanism

Figure 4 The HPLC chromatograms of (a) ethyl acetate extract of dark

soy sauce, (b) compound1 and (c) authentic maltol (d) The overlaid spectra

of maltol and compound 1; the match factor is 999.967 (It is generally

considered to be matched well, if the match factor is no less than 990.) 24

Figure 5 Structure of 3-hydroxy-2-methyl-4H-pyran-4-one (maltol) 25 Figure 6 Standard curve of maltol at five concentrations, 0.25, 0.5 1.0,

Figure 7 (a) limit of detection (LOD): a typical chromatogram of maltol

of 8 μM;.(b) limit of quantification (LOQ): a typical chromatogram of

Figure 8 Typical chromatograms of (a) dark soy sauce extract and

(b) maltol standard (1.0mM) under the following HPLC conditions,

Mobile phase: A: 0.1% formic acid; B: methanol 10% of B for 15min,

10%-90% of B in 6min, 90% of B for another 5 min; Flow rate:

1.0 ml/min; Column: Agilent ZORBAX SB-C18 (4.6 mm i.d × 250mm);

Figure 9 The scavenging effects of dark soy sauce (a), maltol and trolox (b),

Figure 10 Typical HPLC chromatograms of (a) urine sample collected at

1 hour after the subject orally taken 70 mg maltol, and (b) the same urine

sample as in (a) digested with β-glucuronidase And the UV spectrum of

(c) peak 1, whose retention time (RT) at 5.12 min, agrees well with that of

Figure 11 (A) The total ion chromatogram of MS/MS scanning of maltol

glucuronide from urine sample (B) The mass spectrum of the peak with

retention time at 3.12 min in total ion chromatogram A: The ion with

m/z 303 is the protonated ion of maltol glucuronide; whereas the product ion

Figure 12 (A).Total ion chromatogram of MS/MS scanning of maltol sulfate

in urine sample (B) MS spectrum of the peak with retention time at 5.2 min

in total ion chromatogram A (C) MS spectrum of the peak with retention time

at 1.7 min in total ion chromatogram A: The ion with m/z 207 is the protonated

ion of maltol sulfate; whereas the product ion with m/z 127 is protonated ion

Figure 13 (A) Total ion chromatogram of multiple reaction monitoring (MRM)

scan of maltol glucuronide (303/127) and maltol sulfate (207/127) in

urine sample.(B) Extract ion chromatogram of MRM scan of maltol

glucuronide (303/127) (C) Extract ion chromatogram of MRM scan of

Figure 14 The typical HPLC chromatograms of (a) urine sample with subject

taking 30 ml dark soy sauce and (b) that urine sample digested with enzyme:

inset is the UV spectrum of peak with RT at 14.8 min which agrees with that

Figure 15 Time course of digestion of urine samples with 5000U

Figure 16 (A) The maltol amount excreted in urine after one subject took

6 mg maltol or 30 ml dark soy sauce mixed with plain boiled rice The

accumulated maltol amounts excreted in urine for such two cases are also

Figure 17 The average total maltol (standardized with creatinine) measured

in the different time point urine samples of 24 young healthy subjects who

orally took 30 ml of dark soy sauce mixed with 200 gram of plain boiled rice

Figure 18 The absorbance and ABTS•+ scavenging activity of MeOH-R

fraction under acidic and basic conditions The MeOH-R samples were

incubated in 6 M HCl or 4.2 M NaOH in vacuo at 110°C for 18 hours

The colored components became insoluble in acidic condition, while in

alkaline condition they were stable Removal of insoluble colored

components dramatically decreased the antioxidant activity of the acidic

hydrolysate, indicating that the colored components could greatly contribute

to the total antioxidant activity (TAA) of dark soy sauce Values are

mean ± SD, n=3 ***Comparision between the ABTS•+ scavenging activity

of sample + 6 M HCl and that of sample + H2O (*** p < 0.001) 43

Figure 19 (A) Overlain Sephadex G-75 gel filtration chromatograms of

EtOAc-R and MeOH-R of dark soy sauce Fraction 26 to 34 of EtOAc-R

were combined as Colored Product 1 (CP1), and fraction 3 to 9 of MeOH-R

were combined as Colored Product 2 (CP2) The fragmentation range of

Sephadex G75 gel filtration chromatography is 1000 ~ 50 000 Dalton

(B) The correlation of ABTS•+ scavenging activity and absorbance of

CP1 and CP2 at 470 nm Values are mean ± SD, n=3 44 Figure 20 (a) 1H-NMR spectrum of CP1, (b) 1H-NMR spectrum of

CP2, and (c) 13C-NMR spectrum of CP1 45

Figure 21 TOF-MS spectrum of CP1 The inset shows a typical doubly

charged ion The peaks observed at around 1001.3 show the isotopic

pattern at 0.5 Dalton distance 47

Figure 22 Decrease of cell viability of HT 29 cells 24, 48 and 72 hours after incubation with various concentrations of dark soy sauce (DSS) (A), and DSS Nondialysable fraction (B) Results are mean±SD, n=3 52

Figure 23 Cell viability tested on dark soy sauce only or plus catalase (1000 units/ml) 53

Figure 24 Time course of COX-2 expression and its inhibition by dark

soy sauce in lipopolysaccharide (LPS)-induced HT-29 cells Cells were

pretreated with dark soy sauce (5 μl/ml) for half of an hour and then

induced with LPS (100ng/ml) for various time indicated Protein levels

were estimated by Western Blot analysis as described in “Materials and Methods” Lane 1 is untreated HT-29 cells; lane 2, HT-29 cells treated

with LPS (100 ng/ml) only); and lane 3, HT-29 cells simultaneously

treated with dark soy sauce (5 μl/ml) and LPS (100 ng/ml) 54

Figure 25 Inhibitory effect of dark soy on COX-2 expression in LPS-

induced HT-29 cells Lane 1, untreated HT-29 cells; Lane 2, HT-29

cells treated with 1 μl/ml dark soy sauce(DSS); Lane 3, with 5 μl/ml

DSS; Lane 4, with LPS (100 ng/ml); Lane 5, with DSS (1 μl/ml) and LPS (100 ng/ml); Lane 6, with DSS (5 μl/ml) and LPS (100 ng/ml) The western blot is from a single experiment, but is representative of 3

independent experiments 55

Figure 26 The effect of maltol on COX-2 expression in LPS-induced

HT-29 cells.Lane 1 is untrated HT-29 cells; lane 2, HT-29 cells treated

with maltol (100 μM); lane 3, maltol (500 μM); lane 4; treated with LPS

(100 ng/ml); lane 5, maltol (100 μM) and LPS (100 ng/ml); lane 6, maltol

(500 μM) and LPS (100 ng/ml) 55

Chart 1 The flow chart of fractionation and isolation of maltol, CP1 and CP2 from dark soy sauce 10

List of Symbols

ABTS 2,2’-azinobis(3-ethylbenzothiazoline-6-sulfonic acid)

APCI-ITMS atmospheric pressure chemical ionization-ion trap mass spectrometry

CP1/2 the colored product 1/2

DAD diode array detection

EI-MS electron impact – mass spectrometry

ESI-MS electrospray-ionization mass spectrometry

EtOAc-R ethyl acetate extract residue

FTIR Fourier transform infrared

HEMF 4-hydroxy-2(or 5)-ethyl-5(or 2)-methyl-3(2H)-furanone

HPLC high performance liquid chromatography

LC-MS/MS liquid chromatography-mass spectrometry/mass spectrometry

LOD limit of detection

LOQ limit of quantification

MeOH-R methanol extract – residue

NMR nuclear magnetic resonance

SPE solid phase extraction

TAA total antioxidant activity

TEAC trolox equivalent antioxidant capacity

(From Tina Chang’s poem, Ode to Soy Sauce [1])

Soy sauce is a traditional fermented seasoning of East Asian countries and is currently used in cooking worldwide Not only do the flavor components of soy sauce improve the taste of many types of foods, but its coloring ingredients can enhance the

appearance of the dipped food or the mixed soup Moreover, recent studies indicate that some ingredients in soy sauce are potentially beneficial to human health, showing effects such as anticarcinogenesis, antihypertension and antihyperlipidemia [2-4]

1.1 A brief history of soy sauce

The history of soy sauce goes back over three thousand years The origin of soy sauce

is generally considered to be in China, where soy sauce is called jiang-you, the extract

of jiang Jiang was first recorded in the books of Zhou dynasty (1121-256 B.C.), while Jiang-you was first mentioned in a book of ‘Qimenyaoshu’, Jia Sixie, Bei-Wei

dynasty (220-265) [5]

It is speculated that, to preserve foodstuffs against periods of scarcity, the ancient

Chinese accidently discovered Jiang when they mixed salt with meat, fish and

vegetables, and innocently incorporated some harmless aeroborne fungi, which

fermented the raw material after a long period of culture With the introduction of Buddism to China, meat was excluded from the diet of the Buddist monks Plant-

Jian Zhen, in Tang Dynasty (618-907) [7] But some consider that Japanese soy

sauce originated from that brought back by a Japanese Buddist monk, Kakushin, from

China in Song Dynasty (960-1279) [8-10] It is in Japan that the making of soy sauce

was modernized and exported to Europe and North America Between 17th and 18thcentury, a large quantity of soy sauce was exported from Japan by a Dutch company

to Europe [8] As early as in 1867 Japanese soy sauce was taken along by immigrants

to Hawaii [11] In 1972, the Kikkoman Company opened a modern soy sauce plant at Walworth, Wisconsin [8]

Soy sauce was brought to East Asian countries by the Chinese immigrants In

Singapore, it is said that the small-scale manufacturing of soy sauce started by a small

number of Xin-hui Cantonese, a sub-dialect group from Guangdong Province of

South China, just a couple of years after the first arrival of Chinese immigrants in the early nineteenth century [12] After being successful, many of them shifted their business to other more profitable fields Only a few survived For example, Chuen Cheong Food Industries, which produces Tiger brand sauce, was established in 1930, earlier than other local famous soy sauce manufacturers, such as Yeo Hiap Seng (1938) and Tai Hua Food Industries (1949) The business and technologies of soy sauce fermentation have been passed from father to son Now the fourth generation is

in the charge of the business [13, 14]

3 

1.2 The methodology of preparation of soy sauce

The ratios of starting raw materials, soy bean and wheat, are different for Chinese style and Japanese style: around 50:50 for Japanese style, while approximately 60:40 for Chinese style [15] For both styles of traditionally fermented soy sauce, the basic procedure of production is very similar The method of soy sauce production

practiced locally is based on the traditional Chinese fermentation process, the main steps of which are as follows [16]:

Raw material preparation:

Soy beans or defatted soy flakes are soaked and then cooked, while wheat is roasted

The Koji fermentation process

The raw materials are mixed and inoculated with mold (seed Koji), and then cultured

for 2-3 days with controlled temperature and moisture The mold grows enough to provide the enzymes necessary to hydrolyze the raw materials, thus, then, also called

sauce Koji

The mash fermentation

The sauce Koji is poured into fermentation tanks and mixed with saline water, aging

for 3 months to 2 years, upon the culture temperature During the fermentation period,

the enzymes from Koji mold hydrolyze most of the protein to amino acids and

low-molecular-weight peptides, and some polysaccharides into simple sugars

Refinement

The aged mash is filtered The filtrate is called raw soy sauce The raw soy sauce is further pasteurized and finally bottled To make dark soy sauce, the aging process is

After World War II, when the supplies of soy beans were scarce, many manufacturers applied chemical soy sauce, acid hydrolyzed vegetable proteins But the flavor of such products is poor To keep some flavor as of fermented soy sauce, semi-chemical and blended soy sauce were produced Semi-chemical soy sauce is produced by further fermentation of acid hydrolyzed products; while blended soy sauce is produced by mixing fermented soy sauce with chemical soy sauce and/or enzymatically

hydrolyzed vegetable protein As Asian economies have grown, fermented soy sauce becomes the main product in Asian markets The United States is left to be the largest chemical soy sauce producer in the world Japanese producers once proposed the world-trade regulations that the manufacturing methods should be included in the label But the US suggested that individual countries should make their own decisions

on labeling [18] In Singapore, soy sauce can be made from soy beans with or

without other foodstuffs, by “either enzymic reaction or acid hydrolysis or by both methods” [19], although local products are traditionally fermented But there is no labeling requirement on the manufacturing process

5 

1.3 The Functional components of soy sauce

In China, soy sauce has been traditionally used for treatment of anorexia, ulcers post thermal burn etc, recorded in Traditional Chinese medicinal books (Su Jing,

Xinxiubencao, Tang Dynasty, 659; Sun Simiao, Qianjinfang, Tang Dynasty, 625)

[7] Modern scientific research has provided some supporting evidence for such usage: e.g soy sauce can promote gastric juice secretion in humans; soy sauce has

antimicrobial activity, due to the synergistic effects of NaCl, ethanol, pH, and

preservatives [20]

Soy sauce has a variety of biologically active effects, such as hypotensive,

anticarcinogenic, anticataract, and antiplatelet Nicotianamine was found to be the major bioactive components inhibiting angiotensin I-converting enzyme 4-hydroxy-

3(2H)-furanone derivatives, namely, 4-hydroxy-2(or 5)-ethyl-5(or furanone, 4-hydroxy-5-methyl-3(2H)-furanone and 4-hydroxy-2,5-dimethyl-3(2H)-

2)-methyl-3(2H)-furanone, are antioxidants, also having anticarcinogenic and anticataract activities Shoyu-flavones, derivatives of daidzein, genistein, and 8-hydroxygenistein, are

antioxidants and histidine decarboxylase inhibitors carboline and 1-methyl-β-carboline are found to be the active antiplatelet components [20]

1-methyl-1,2,3,4,-tetrahydro-β-During the fermentation process, the soybean and wheat proteins are degraded into peptides and amino acids Thus the IgE-mediated hypersensitive response to wheat is markedly reduced in soy sauce Furthermore, the soy sauce polysaccharides that

cannot be hydrolyzed during the fermentation process can inhibit hyaluronidase

activity and histamine release In vitro, these polysaccharides can increase the

production of IgA from Peyer’s patch cells In vivo, soy sauce polysaccharides can

6 

increase the concentration of IgA in the intestines of mice, suppress passive cutaneous anaphylaxis reaction in the ears of allergy model mice, and regulate the balance of Th1/Th2 cell response in mice, potentially enhancing host defenses [21-23] Clinically, oral administration of soy sauce polysaccharides can improve allergic symptoms of patients with perennial allergic rhinitis or seasonal allergic rhinitis [23] Soy sauce polysaccharides can also inhibit pancreatic lipase activity and reduce the absorption of lipid in mice and humans [4]

1.4 The balance of free radicals/reactive species and antioxidants

A free radical is any species capable of independent existence containing one or more unpaired electrons [24] Free radicals can be formed by homolytic or heterolytic

fission of a covalent bond For example, UV-induced homolytic fission of the O-O bond in H2O2 can produce hydroxyl radical, OH• Reactive oxygen species (ROS) includes not only the oxygen radicals (O2• − and OH•) but also non-radical derivatives

of O2 (H2O2, HOCl and O3) Univalent reduction of molecular oxygen can form a variety of ROS Mitochondrial electron transport chain is the main cellular source of ROS Various oxidases in cell membranes, cytosol, and other organelles can transfer single electrons onto dioxygen,e.g Cytochrome P450, a monooxygenase in the

endoplasmic reticulum, is able to reduce dioxygen to superoxide radicals Transition metals, e.g iron, having the capability of donating and/or receiving electrons, play an important role in oxygen radical formation Inflammation can also result in excessive production of free radicals Free radicals and ROS are not only formed endogenously, but also introduced outside [25]

Our body has evolved antioxidant defense systems Small molecules and antioxidant enzymes are two major categories Antioxidant enzymes, such as superoxide

7 

dismutase, catalase, and glutathione reductase, are important for antioxidant defense [24]

Imbalance between formation of free radicals/reactive oxygen species and levels of

antioxidants in vivo has been suggested to play a role in the development of various

diseases, such as atherosclerosis, diabetes, rheumatoid arthritis, cancer and

neurodegenerative diseases [24, 26] Some (but not all) studies show that nutritional antioxidants can decrease oxidative damage in the human body and may have

beneficial effects on disease prevention [26 – 29] This has led to a growing interest in antioxidants from natural products [26 – 31] Several papers have alluded to the presence of antioxidants in soy sauce [32 – 36]

Tiger brand soy sauce is a local brand sauce with an 80-years history [13, 14] Our group’s studies showed that dark soy sauce, especially Tiger brand products, had

extremely high total antioxidant activity (TAA) in vitro [36] as judged by the ability

to scavenge the nitrogen-centred ABTS•+ radical, an assay that is frequently used to

assess the antioxidant activity of beverages, food extracts and body fluids [37] In vivo

Dark soy sauce of this brand also decreased lipid peroxidation in human volunteers [38] In this study, using Tiger brand products, we attempted to identify the major components that contribute to the high antioxidant activity of dark soy sauce

Chapter 2 Materials and Methods

Dark soy sauce (Tiger brand, Chuen Cheong Food Industries, Singapore) was

purchased from a local supermarket

2.2 ABTS assay

This was carried out as described in Ref [36, 37]

2,2ʹ-Azino-bis[3-ethylbenzothiazoline-6-sulfonate] (ABTS) in water (7 mM final concentration) was oxidized using potassium persulfate (2.45 mM final concentration) for at least 12 h in the dark The ABTS•+ solution was diluted to an absorbance of 0.70 ± 0.02 at 734 nm (Beckman UV – VIS spectrophotometer, Model DU640B, UK) with phosphate

buffered saline (PBS 10 mM, pH 7.4) Extracts of dark soy sauce (10 μl) or trolox standard (10 μl) were added to 1 ml of ABTS•+ solution Absorbance was measured

1 min after initial mixing Antioxidant properties of fractions of dark soy sauce

extracts were expressed as Trolox equivalent antioxidant capacity (TEAC), calculated from at least three different concentrations of extract tested in the assay and giving a linear response

2.3 Isolation of low molecular mass components from ethyl acetate extract

Dark soy sauce (6.4 L) was extracted three times with a four-fold volume of methanol The methanol extracts were combined, filtered by filter paper and evaporated to

dryness under vacuum at 40ºC (MeOH-extract, yield 2.6 kg) No significant retention

of antioxidant components from soy sauce by the filter paper was detected

The residue after methanol extraction (MeOH-R) yielded 1.4 kg The MeOH-extract was suspended in water and partitioned with ethyl acetate three times

The ethyl acetate fractions were combined and evaporated to dryness under vacuum at 30ºC (EtOAc-extract, yield 11.6 g) The remaining aqueous fraction (EtOAc-R) yielded 2.6 kg (Chart 1.)

As the EtOAc-extract exhibited strong ABTS radical scavenging activity, it was

subjected to flash chromatographic separation with a silica gel RP18 (particle size 40 – 63 µm, Merck KGaA, Darmsadt, Germany) packed column (6 × 42 cm) eluting with methanol and water Fractions eluted with 10% methanol had the highest ABTS•+radical scavenging activity and were further purified by a prep-HPLC (Agilent 1100 Series, equipped with a fraction collector) using a ZORBAX SB-C18 PreP HT

column (21.2 × 250 mm, 7 µm) (Agilent, USA) at 20 ml/min with methanol—0.1% formic acid in MilliQ water (10:90, v/v) as mobile phase Ten fractions showing antioxidant activity in the ABTS assay, Fr.1 to 10, were obtained Fraction 9 was found to have the most ABTS scavenging activity and yielded Compound 1 (41mg)

2.4 Fractionation of colored components

Approximately, 1 gm of ethyl acetate extract residue (EtOAc-R) was resuspended in

25 ml distilled water and dialyzed against distilled water for seven days, using a

(1) extracted with 4 fold methanol (MeOH)

3 times

(1) filtered by filter paper, evaporated to dryness under vacuum at 40 ºC (2) and suspended in water;

(3) partitioned with ethyl acetate (EtOAc) three times

Dark soy sauce

(1) Fractionated with RP18 column chromatography (2) Further fractionated with prep HPLC Eluted with methanol – 0.1% formic acid

Colored Product 1 (CP1)

(1) dialyzed against distilled water (2) fractionated with gel filtration chromatography

Chart 1 The flow chart of fractionation and isolation of maltol, CP1 and CP2 from dark soy sauce  

cellulose dialysis tubing (Pierce, Rockford, USA; molecular weight cutoff 3500) Initially, we investigated the influence of dialysis time and temperature (room

temperature, ~25ºC, and cold-room temperature, ~4ºC), on the antioxidant capacity of the colored products as measured by the ABTS assay and found no significant effect For convenience, the dialysis experiments were carried out at room temperature The non-dialyzable fraction was freeze-dried

Approximately, 85 mg of the freeze-dried product was dissolved in 5 ml of water and loaded onto a fine Sephadex G-75 gel filtration chromatography column (2 × 100 cm) The colored fractions (absorbance at 470 nm), 4 ml per tube, were collected The fractions from No 26 to 34 possessed significantly higher TEAC values and

consequently were combined as Colored Product 1 (CP1) (yield 61 mg)

The MeOH-R fraction (approximately 2 g) was dialyzed against distilled water The non-dialyzable fraction (85mg) was fractionated with gel chromatography in the same way as EtOAc-R The fractions from No 3 to 9 were combined as Colored Product 2 (CP2) (yield 47mg) (Chart 1.)

2.5 HPLC determination of maltol in dark soy sauce

Dark soy sauce (10 ml) was extracted with 40 ml methanol on an orbital shaker

(SLOS-20, Seoulin Bioscience, Seoul, Korea) at a speed of 150 rpm for 24 h, and then centrifuged at 3000g for 30 min This procedure was repeated three times The

supernatants were pooled and dried using a rotary evaporator under vacuum at 40ºC The residue was dissolved in 20 ml water and extracted three times with 20 ml ethyl acetate The ethyl acetate extracts were combined and evaporated to dryness at 30ºC Prior to HPLC analysis, the dried samples were dissolved in 10 ml methanol-0.1%

formic acid (1:9) and then filtered through 0.45 µm disposable nylon filters (Agilent Technology, USA)

Analysis was performed using an Agilent 1100 HPLC with an Agilent ZORBAX C18 column (4.6 × 250 mm, 5 µm), maintained at 35ºC The mobile phase was formic acid in MilliQ water (0.1%, v/v) (A) and methanol (B) with a gradient program as follows: 10% of B for 15 min, 10 – 90% of B in 6 min, 90% of B for another 5 min, with flow rate at 1 ml/min The injection volume for all samples was 10 μl and

SB-absorbance was monitored at 270 nm Spectra were recorded from 190 to 400 nm

2.6 Mass spectrometry

An Agilent XCT Plus ion trap mass spectrometer (ITMS) (Agilent Technology, US) was used to analyze the fractions, Fr 1 – 10, obtained from ethyl acetate extracts Atmospheric pressure chemical ionization (APCI) MS was performed in the positive mode The dry gas and vaporizer temperatures were 350 and 400ºC, respectively Compound 1 was also analyzed by an electron impact (EI) MS spectrometer (Agilent Technologies), sample dissolved in methanol and introduced by a gas

chromatography (GC) interface (Agilent Technologies)

For the analysis of CP1, Electrospray-Ionization Mass Spectrometry (ESI-MS) was performed using a Waters Micromass Q-Tof micro mass spectrometer (Waters, USA) For acquiring mass spectra, sample was directly infused at a speed of 10 µl/min The capillary and sample cone voltages were maintained at 3.0 kV and 50 V, respectively The source and desolvation temperatures were 80 and 250ºC, respectively The mass

spectra were acquired from m/z 100 to 5000 in the positive ion mode

2.7 Fourier transfer infrared spectrometry (FTIR)

IR spectra (KBr disc) were recorded on a JASCO FT/IR-430 spectrometer (Japan)

2.8 Nuclear magnetic resonance spectrometry (NMR)

The NMR spectra of CP1 and CP2 and compound 1 were recorded on a Bruker

Advance AMX500 NMR spectrometer (Rheinstetten, Germany) at 500.13 MHz (1H) and 125.75 MHz (13C), respectively Compound 1 was dissolved in methanol-d4; CP1

and CP2 were dissolved in D2O

2.9 Detection and determination of maltol metabolites in human urine samples

2.9.1 Standard preparation

Maltol glucuronide was isolated and puried from the urine samples of a healthy volunteer using solid phase extraction (SPE), reverse-phase (RP) C-18 flash column chromatography and preparative HPLC The purity of the isolated compound was checked on an analytical HPLC system Its identity was confirmed based on NMR data

Maltol sulfate was synthesized based on a published method [39] To the solution of

200 mg maltol in pyridine, 300 mg pyridine-sulfur trioxide complex was added and the mixture was stirred at 4 °C for 60 hours The precipitating solid in the reaction mixture was filtered off and washed with CHCl3, then dried in vacuo to give 390 mg

solid The solid was then dissolved in water and further purified using a preparative HPLC (Agilent Technologies)

A series of different concentrations (0.05 mM - 1 mM) of maltol standard solutions were prepared in 10% methanol/0.1% formic acid- MilliQ water

2.9.2 Sample preparation

To test how fast maltol is metabolized, the urine samples at different time points (0.5,

1, 2, 3, 4, 5 and 6h) were collected from a human volunteer who had consumed 6 mg maltol or 30 ml dark soy sauce mixed with 200 gram of plain boiled rice The

washtime between these two experiments were more than one week The urine

samples were also collected from 24 young health volunteers in an observer-blinded, randomized, placebo controlled, crossover clinical study [38]

The urine samples were deproteinized with three volumes of methanol, then

centrifuged at 1500 × g for 10 min For detection of maltol metabolites, the

supernatants were directly injected onto HPLC column For determination of total maltol content in urine, these supernatants were dried under N2 flow, digested with

200 μl 11.1 mg/ml β-D-galactonidase in 0.5 mM acetate buffer at 37°C overnight, and then centrifuged at 20,000 × g for 15 min at 4°C The supernatants were collected and analyzed immediately using HPLC

2.9.3 HPLC-DAD detection of maltol metabolites and determination of total maltol content in urine

HPLC analysis was performed on an agilent 1100 HPLC system, equipped with a diode array detector The samples were introduced by an autosampler A

Phenomenex column (4.6 × 250 mm, 5.0 μm) was used, with temperature kept at

35 °C in a thermostat column compartment UV spectra (190-400 nm) were recorded

The mobile phase was composed of 10% methanol in 0.1% formic acid, with a

running rate of 1ml/min

2.9.4 HPLC-MS/MS detection of maltol metabolites

LC-MS/MS analysis of maltol metabolites in urine was performed on an Agilent 1200

LC system (Palo Alto, CA, USA) coupled to an API3200 Triple-Quadrupole mass

spectrometer (Applied Biosystems/MDS SCIEX, Foster City, USA) A Turbo V

source with ESI prober was used for the analysis Data acquisition was performed

with Analyst 1.4.2 software (AB MDS Sciex)

The chromatographic separation was performed on a 150 × 2.0 (i.d.) mm Phenomenex

Synergi 4µ Polar-RP 80 Å column (Phenomenex, CA, USA) with a column oven

temperature of 35°C The mobile phase was premixed 10% methanol in 0.1% formic

acid Flow rate was 0.4 ml/min Standards or samples were introduced into the LC

using an Agilent 1200 G1367B autosampler and injection volume was 10.0 μl

The ion source was operated in the positive mode For survey scan (Q1 MS scan

mode), ion-spray voltage and temperature were set at 5000 V and 600oC Ion source

gas 1(GS1) and ion source gas 2 (GS2) were set at 40 and 45 psi Interface heater (ihe)

was set as on The curtain gas (CUR), declustering potential (CE) and entrance energy

(EP) settings were at 10 psi, 50 V and 10 V, respectively

In product ion scan (MS2) mode, for maltol glucuronide, parent ion 303 was used,

and scan range was set at 50 – 400; whereas for maltol sulphate, parent ion 207, scan

range was 50 -300 For both chemicals, collision energy (CE) was set at 5 V In

multiple reaction monitoring (MRM) mode, the ion pairs monitored for maltol

glucuronide and maltol sulphate were 303/127 and 207/127 respectively

2.10 GC-MS analysis of DNA base modification

(E260 = 50 μg/ml) Aliquots of 100 μg DNA were hydrolyzed by adding 0.5 ml of 60% (v/v) formic acid and heating at 140 °C for 45 min in an evacuated, sealed hydrolysis tube The internal standards were added to the cooled hydrolyzed samples and freeze-dried Derivatization was carried out for 2 hour at 25 °C using BSTFA (+1%

TMCS)/acetonitrile/ethanethiol (16:3:1(v/v)) mixture

2.10.2 GC-MS analysis

The GC-MS analysis was performed as previously reported [41, 42] Derivatized

samples were analyzed by GC-MS (Agilent gas chromatography 6890 interfaced with Agilent 5973 mass selective detector) The injection port and the GC-MS interface were kept at 250 and 290 °C, respectively Separations were carried out on a fused silica gel capillary column (12 m × 0.2 mm i.d.) coated with cross-linked 5%

phenylmethylsiloxane (film thickness 0.33 μm) (Agilent) Helium was the carrier gas

with a flow of 1.0 ml/min Derivatized samples (1.0 μl) were injected into the GC port

using a split ratio of 8:1 Column temperature was increased from 125 to 175 °C at

8 °C/min after 2 min at 125 °C, then increased from 175 to 220 °C at 30 °C/min, held

at 220 °C for 1 min, and finally increased from 220 to 290 °C at 40 °C/min and held

at 290 °C for 2 min Selected-ion monitoring was performed using the

electron-ionization mode with the ion source maintained at 189 °C

2.11 Cell culture

Human colorectal adenocarcinoma cells (HT29), obtained from the American Type Culture Collection, were cultured in McCoy’s 5A media with 1% (v/v) penicillin, 10% (v/v) fetal bovine serum (FBS), 5% CO2 at 37 °C The cells were maintained in the logarithmic growth phase by routine passage every 2-3 days

2.12 Assessment of cell viability

Viability was assessed by the MTT (3-(4,5-dimethyl-2-yl)-2,5-diphenyltetrazolium bromide) method as previously reported [43, 44] Cells in McCoy’s 5A full media were seeded overnight at a density around of 1.5 × 104 cells per well in 96-well plates After treatment with various concentration of dark soy sauce in full medium, 200 μl

of MTT (0.5mg/ml final concentration) dissolved in McCoy’s 5A only (with FBS) Cells were incubated at 37 °C in darkness for 1 hour and then MTT was removed 200

μl of dimethyl sulfoxide (DMSO) was added to solubilize the formazan formed After shaking in the dark for 15 min, absorbance at 550 nm was measured using a

microplate reader (SpectraMax190, Molecular Device)

2.13 Western blot analysis

Experiments were performed as describe in [43, 45] Control and treated cells were washed twice with cold PBS buffer, incubated in lysis buffer on ice for 30 min, and then scraped into Eppendorf tubes Cells were centrifuged at 13800 rpm at 4 °C for 15 min using a desktop centrifuge (Eppendorf) to remove unbroken cells, nuclei and other organelles The protein concentration of the resulting supernatant was measured with the DC protein assay Kit (Bio-Rad), and 20 μg of protein was boiled for 10 min, and then loaded in a 12.5% (v/v) SDS – PAGE gel The separated proteins were then transferred to nitrocellulose membranes and probed with antibodies against COX-2, β-actin, followed by the appropriate horseradish peroxidase-conjugated secondary antibodies

2.14 Statistical analysis

The mean values were calculated from data taken from at least three separate

experiments Where significance testing was performed, a Student’s t-test was used; P-values of, 0.05 were considered to be statistically significant

Chapter 4 Results and Discussion

4.1 Separation and characterization of low molecular mass components

Dark soy sauce was extracted with methanol, and the methanol extract was further partitioned with ethyl acetate as described in Experimental section (Chart 1) The three fractions of dark soy sauce, methanol extract residue (MeOH-R), ethyl acetate extract (EtOAc-extract), and ethyl acetate extract residue (EtOAc-R), were subjected

to the ABTS assay The MeOH-R fraction contributed about 60% of the total

antioxidant activity (TAA) of dark soy sauce However, the EtOAc-extract showed the strongest antioxidant activity, approximately 12-fold higher than that of the dark soy sauce (Figure 1)

After pretreatment with a reverse phase (RP)-C18 column (mainly to remove benzoic acid, which has no antioxidant activity in the ABTS assay), the EtOAc-extract was separated with preparative high performance liquid chromatography (HPLC) resulting

in 10 fractions (Figure 2A) Among them, the ABTS assay (Figure 2B) showed that Fr.9 has by far the strongest activity Fr.9 was further purified to yield compound 1

The positive atmospheric pressure chemical ionization - ion trap mass spectrometry (APCI-ITMS) spectrum of Compound 1 showed a protonated molecular ion peak at

m/z 127 [M + H]+ The electron impact – mass spectrometry (EI-MS) spectrum also

shows its molecular ion at m/z 126 (Figure 3) The major fragment ions (m/z 97, 71, 55) were observed The molecular ion m/z 126 can be decomposed to a furan ion (m/z 97) by losing one H and CO The furan fragment ion (m/z 97) can be further

decomposed to ion m/z 55 by losing CH3CO Via another pathway, the molecular ion

(m/z 126) can give the fragment ion m/z 71 by losing CH2CH2 and CO The

mechanism for these fragment ion formation is proposed in Figure 3 The ion

fragment pattern agrees with that previously reported [46] The proton and carbon 13 nuclear magnetic resonance (1H- and 13C-NMR) spectral data of compound 1 are listed in Table I The 1H-NMR spectrum of compound 1 (500MHz, methanol-d4)

showed two cis-olefinic proton signals at δH 7.94 (1H, d, J = 5.5 Hz) and 6.39 (1H, d,

J = 5.5 Hz) In addition, the signal of one methyl group directly attached to an olefinic

carbon at δH 2.35 (3H, s) was observed The 13C-NMR spectrum showed one

carbonyl carbon signal at δC175.3, and four olefinic carbon signals at δc 156.3, 152.2, 144.6 and 114.4, indicating the presence of a pyranone ring The signal at δc 14.2 corresponded to the carbon of the methyl group The Fourier transform infrared (FTIR)

spectrum of compound 1 shows the major absorbance bands (cm-1): 3258, 3069, 1655,

1617, 1560, 1459, 1257, 1199 Compound 1 was also analyzed by high performance

liquid chromatography – diode array detection (HPLC-DAD) Its retention time and ultraviolet (UV) spectra (absorption maxima at 210 nm and 275 nm) agreed very closely with those obtained from authentic maltol (Figure 4) So the MS spectrum, NMR data, FTIR spectrum, the retention time and UV spectrum support the identity

of compound 1 as 3-hydroxy-2-methyl-4H-pyran-4-one (maltol) (Figure 5)

Fractions 1-8 and 10 were also analyzed by APCI-ITMS, suggesting these fractions contain molecules with molecular weight ranging from 124-214 (Appendix: Table A-1)

200 400 600 800 1000 1200 1400 1600 1800

1

2 3

4 5

Figure 3 EI-MS spectrum of compound 1 and the proposed mechanism for the

formation of fragment ions

Figure 4 The HPLC chromatograms of (a) ethyl acetate extract of dark soy sauce, (b)

compound1 and (c) authentic maltol (d) The overlaid spectra of maltol and

compound 1; the match factor is 999.967 (It is generally considered to be matched

well, if the match factor is no less than 990.)

160 *Fr.2a9 *maltol

Match factor:

999.967

(d)

Figure 5 Structure of 3-hydroxy-2-methyl-4H-pyran-4-one (maltol)

Table I 1H- and 13C-NMR data of Compound 1

4.2 Content of maltol and its contribution to the TAA of dark soy sauce

A considerable part of the TAA of the dark soy sauce was contained in MeOH-R and EtOAc-R fractions and in fractions 1-8 and 10 (Figure 1, 2) Although maltol was identified as the active fraction of Fr 9, we needed to determine its overall

contribution to TAA Thus an HPLC method was developed for determination of

maltol in dark soy sauce

Maltol standard was dissolved in methanol-0.1% formic acid (10:90, v/v) yielding concentrations of 0.25, 0.5, 1.0, 1.5, 2.0 mM Three different sets of standard

solutions were prepared and analyzed each day and in three continuous days

Calibration curves for the quantification of maltol were obtained by plotting amount (mM) against peak area The calibration curve (Figure 6) was linear within the

investigated concentration range (0.25 – 2.0 mM) with the following regression

in a peak with a height ten times that of the baseline noise The LOD and LOQ were 8

and 25 μM respectively Figure 7 shows a typical chromatogram at such concentration The accuracy was evaluated through recovery studies by adding known amounts of maltol at different levels (0.25, 0.5, 1.0, 1.5, 2.0 mM) to the sample The unspiked samples and each of the spiked samples were analyzed in triplicate The recoveries at the different concentration levels were in the range 89.8 to 94.2% (Table II) Using this validated method, the average concentration of maltol in dark soy sauce was determined to 1.15 ± 0.04 mM (n=5) One typical HPLC chromatogram of ethyl acetate extract of dark soy sauce is shown in Figure 8

Based on the ABTS assay, the total antioxidant activity of dark soy sauce and matol was evaluated The antioxidant activity of maltol (TEAC value) was 2.67 ± 0.05 mM (n=5) (Figure 9) This concentration of maltol was calculated to contribute around 2.42% of the TAA of dark soy sauce

Table II Within- and between-assay precision and recoveries of the assay used to measure maltol

Concentration

level/mM

Within-assay precision (%

RSD, n=3)

Between-assay precision (% RSD, n=3)

0 2000 4000 6000 8000 10000