Oxidation Mechanism of Riboflavin Destruction And

Home , Tocopherol, Tocotrienol

OXIDATION MECHANISM OF RIBOFLAVIN DESTRUCTION AND

ANTIOXIDANT MECHANISM OF TOCOTRIENOLS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

The Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

Hyun Jung Kim, M.S.

* * * * *

The Ohio State University

2007

Dissertation Committee: Approved by

Professor David B. Min, Adviser Professor V.M. Balasubramaniam Professor Lynn Knipe Adviser Professor Luis E. Rodriguez-Saona Graduate Program in Professor Stephanie A. Smith Food Science and Nutrition

ABSTRACT

Singlet oxygen formed in the presence of triplet oxygen and photosensitizer under

light accelerates the oxidation and causes the nutritional loss of foods. Riboflavin, an

essential vitamin B2 is a good photosensitizer to form singlet oxygen and it is very unstable under light. The compound formed from riboflavin under light was studied by solid-phase microextraction (SPME)-gas chromatography (GC)-mass spectrometry (MS) analysis. Only one major compound in the riboflavin solution of a phosphate buffer (0.1

M, pH 6.5) under light was formed and increased as the light exposure time increased.

The major compound from riboflavin solution was positively identified as 2,3- butanedione by a combination of gas chromatographic retention time, mass spectrum and odor evaluation of authentic 2,3-butanedione. The addition of sodium azide, a singlet oxygen quencher, to riboflavin solution minimized the formation of 2,3-butanedione.

Singlet oxygen was involved in the formation of 2,3-butanedione. The 2,3-butanedione was produced from the reaction between riboflavin and singlet oxygen. Singlet oxygen was formed from triplet oxygen by riboflavin photosensitization mechanism.

ii Tocotrienols are vitamin E family with tocopherols which are major lipid soluble antioxidants. The effects, quenching mechanisms, and kinetics of tocotrienols on the autoxidation and the chlorophyll-photosensitized oxidation of lard were studied to improve the oxidative stability of lard. The effects of 0, 100, 200, 300, 500, and 1000 ppm of α-, β-, γ-, and δ-tocotrienol on the oxidative stability of lard during the storage of

7 days in the dark at 55°C were studied. The oxidation of lard was determined by

measuring headspace oxygen in the sample bottle using gas chromatography and

peroxide value daily. As the storage time increased, the headspace oxygen contents

decreased and peroxide value increased. The lard containing α-, β-, γ-, and δ-tocotrienol

had higher headspace oxygen contents and lower peroxide values than the lard without

tocotrienol. Tukey’s test showed that the 100 ppm α-or β-tocotrienol significantly

increased the headspace oxygen contents and decreased peroxide values of lard during

storage (p<0.05). As the concentration of α- or β-tocotrienol increased from 100 to 200,

300, 500, and 1000 ppm, the antioxidative activities decreased. Tukey’s test showed that

the γ- or δ-tocotrienol significantly increased the headspace oxygen contents and

decreased peroxide values of lard during storage (p<0.05) but there were no significant

differences between the concentrations of 100, 200, 300, 500, and 1000 ppm (p>0.05).

The optimum concentration of α-, β-, γ-, and δ-tocotrienol to increase the oxidative

stability of lard could be 100 ppm. The antioxidative activities increased in δ- > γ- > β- >

iii α-tocotrienol. The selection of type and optimum concentration of tocotrienol not only

minimizes the oxidation of lipids but also will be economically important. Samples of

0.1, 0.25, and 0.4 M lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b and

0, 0.3, 0.6, or 0.9 mM α-, β-, γ-, and δ-tocotrienol were prepared and stored at 3,000 lux for 4 hours to study the effects, quenching mechanisms, and kinetics of α-, β-, γ-, and δ- tocotrienol on the chlorophyll photosensitized oxidation of lard. The oxidation of sample was determined by measuring the headspace oxygen content in air-tight sample bottles using gas chromatography and the peroxide value. α-, β-, γ-, and δ-Tocotrienol had significant antioxidative effects on the chlorophyll photosensitized oxidation of lard at p<0.05. Chlorophyll in methylene chloride at 3,000 lux produced singlet oxygen at 1.089

µmole oxygen/mL headspace/hr. The steady state kinetic study showed that α-, β-, γ-,

and δ-tocotrienol acted as antioxidants in chlorophyll photosensitized oxidation of lard by

quenching singlet oxygen. The reaction rate constant of singlet oxygen with lard was 6.5

× 104 M-1sec-1. The singlet oxygen quenching rates of α-, β-, γ-, and δ-tocotrienol were

2.16 × 107, 1.99 × 107, 2.05 × 107, and 0.80 × 107 M-1sec-1, respectively.

α-Tocopherol in foods is oxidized during the processing and storage. The effects of 0, 250, 500, 1000 and 1500 ppm of oxidized α-tocopherol on the oxidative stability of

purified soybean oil in the dark at 55°C for 6 days were determined by measuring the

peroxide values and headspace oxygen contents. As the concentrations of oxidized α-

iv tocopherol increased, the peroxide values increased and the headspace oxygen contents decreased during the 6 days of storage. Tukey’s test showed that oxidized α-tocopherol had a significant effect on the peroxide value and headspace oxygen disappearance of oil at p<0.05. The results showed that oxidized α-tocopherol compounds acted as prooxidant in purified soybean oil. The prooxidant mechanisms of oxidized α-tocopherol may be due to intermediate compounds such as α-tocopherol peroxy radical, α- tocopherol oxy radical, hydroxy radical, and singlet oxygen which are formed during the oxidation of α-tocopherol. The oxidized α-tocopherol contained polar and nonpolar groups in the same molecule. The polar group of oxidized α-tocopherol may reduce the surface tension of oil to increase the transfer of headspace oxygen to oil and accelerate the oil oxidation. The prevention of tocopherol oxidation and removal of oxidized tocopherol could improve the oxidative stability of foods.

Dedicated to my mother, my father and my brother

ACKNOWLEDGMENTS

I must thank to my adviser, Dr. David B. Min for his valuable guidance, support, patience and encouragement throughout my Ph. D. study. He has been a great teacher and a considerable mentor who is so dedicated to show me the joy of learning. It has been my greatest pleasure to be his student under his enthusiasm and expertise.

I would like to thank my committee members, Dr. V.M. Balasubramaniam, Dr.

Lynn Knipe, Dr. Luis E. Rodriguez-Saona and Dr. Stephanie Smith for their guidance, suggestion and advice during my study. Also I deeply appreciate to my former undergraduate and M.S. advisors, Dr. Byong Ki Kim and Dr. Khee Choon Rhee for their constant encouragement, moral support and sincere care to lead my step forward.

I greatly thank to my colleagues and friends, Robert King, Yettella Ramesh

Reddy, Yoon-Hee Lee, Naeemah Hall, Timothy Chapman, and SeungRan Yoo for their help and friendship. Special thanks to Hyun Ju Lee who has encouraged me and prayed for me throughout my study.

Most importantly, I sincerely thank to my mother, father and brother for their unconditional love and strong support that make me do anything.

vii

VITA

November 26, 1977………..……… Born in Seoul, Republic of Korea

1996-2000…………………………. B.S. in Food Engineering Dankook University, Cheonan, Korea

2001-2003…………………………. M.S. in Food Science and Technology Texas A&M University, College Station, TX

2003-2005…………………………. Ph.D. study in Food Science and Human Nutrition University of Illinois, Urbana-Champaign, IL

2005-2007…………………………. Ph.D. in Food Science and Technology The Ohio State University, Columbus, OH

PUBLICATIONS

1. Kim HJ, Lee HO, Min DB. 2007. Effects and prooxidant mechanisms of oxidized α- tocopherol on the oxidative stability of soybean oil. J Food Sci 72(4): 777-82.

2. Bartee SD, Kim HJ, Min DB. 2007. Effects of antioxidants on the oxidative stability of oils containing arachidonic, docosapentaenoic and docosahexaenoic acids. J Am Oil Chem Soc 84(4): 362-8.

3. Kim HJ, Hahm TS, Min DB. 2007. Hydroperoxide as prooxidant in the oxidative stability of soybean oil. J Am Oil Chem Soc 84(4): 349-55.

viii 4. Jia M, Kim HJ, Min DB. 2007. Effect of soybean oil and oxidized soybean oil on the stability of β-carotene. Food Chem 103:695-700.

5. Jung MY, Oh YS, Kim DK, Kim HJ, Min DB. 2007. Photoinduced generation of 2,3- butanedione from riboflavin. J Agric Food Chem 55(1):170-4.

6. Kim HJ, Lee MY, Min DB. 2006. Singlet oxygen oxidation rates of α-, γ-, and δ- tocopherols. J Food Sci 71(8):C860-3.

7. Player ME, Kim HJ, Lee HO, Min DB. 2006. Stability of α-, γ-, or δ-tocopherols during soybean oil oxidation. J Food Sci 71(8):C854-9.

8. Daniel RL, Kim HJ, Min DB. 2006. Hydrogenation and interesterification effects on the oxidative stability and melting point of soybean oil. J Agric Food Chem 54:6011- 15.

9. Bradley DG, Kim HJ, Min DB. 2006. Effects, quenching mechanism, and kinetics of water soluble compounds in riboflavin photosensitized oxidation of milk. J Agric Food Chem 54:6016-20.

10. Huang R, Kim HJ, Min DB. 2006. Photosensitizing effect of riboflavin, lumiflavin and lumichrome on the generation of volatiles in soymilk. J Agric Food Chem 54:2359-64.

PUBLISHED ABSTRACTS

1. Kim HJ, Min DB. Effects, quenching mechanisms and kinetics of α-, β-, γ-, and δ- tocotrienol on chlorophyll photosensitized oxidation of lard. IFT Annual Meeting. Chicago, IL. July, 2007.

2. Kim HJ, Min DB. Mechanism of riboflavin destruction under light. OARDC Annual Conference. Columbus, OH. April, 2007.

3. Kim HJ, Bartee S, Min DB. Oxidative stability of oils containing omega-3 and omega-6 fatty acids. IFT Annual Meeting. Orlando, FL. June, 2006.

4. Kim HJ, Daniels RL, Min DB. Hydrogenation and interesterification effects on the oxidative stability and melting point of soybean oil. IFT Annual Meeting. Orlando, FL. June, 2006.

ix 5. Kim HJ, Bradley DG, Min DB. Effects, quenching mechanisms, and kinetics of water soluble compounds in riboflavin photosensitized oxidation of milk. IFT Annual Meeting. Orlando, FL. June, 2006.

FIELD OF STUDY

Major Field: Food Science and Nutrition

TABLE OF CONTENTS

Page

Abstract...... ii

Dedication...... vi

Acknowledgments...... vii

Vita...... viii

List of Tables ...... xiii

List of Figures...... xiv

Chapters

1. Introduction...... 1

2. Literature Review...... 5

2.1. The oxidation of foods...... 5 2.1.1. Chemistry of triplet and singlet oxygen...... 5 2.1.2. Formation of singlet oxygen ...... 8 2.1.3. Photosensitizer ...... 12 2.1.4. Triplet oxygen oxidation...... 18 2.1.5. Singlet oxygen oxidation ...... 20 2.2. Chemistry and antioxidative activity of tocopherols and tocotrienols...... 22 2.2.1. Chemical and structural characteristics ...... 22 2.2.2. Vitamin E activity...... 24 2.2.3. Stability during processing and storage ...... 25 2.2.4. Antioxidative activity of tocopherols...... 28 2.2.5. Antioxidative activity of tocotrienols ...... 33 2.2.6. Prooxidative activity of tocopherols ...... 35

3. Photoinduced Generation of 2,3-Butanedione from Riboflavin...... 38 3.1 Abstract...... 38 3.2 Introduction...... 39 3.3 Materials and Methods...... 41 3.4 Results and Discussion ...... 45 3.5 Conclusion ...... 51 3.6 References...... 53

4. Effects of α-, β-, γ-, and δ-Tocotrienol on the Oxidative Stability of Lard...... 61 4.1 Abstract...... 61 4.2 Introduction...... 62 4.3 Materials and Methods...... 64 4.4 Results and Discussion ...... 66 4.5 Conclusion ...... 74 4.6 References...... 76

5. Effects, Quenching Mechanisms, and Kinetics of α-, β-, γ-, and δ-Tocotrienol on Chlorophyll Photosensitized Oxidation of Lard ...... 89 5.1 Abstract...... 89 5.2 Introduction...... 90 5.3 Materials and Methods...... 93 5.4 Results and Discussion ...... 95 5.5 Conclusion ...... 102 5.6 References...... 103

6. Effects and Prooxidant Mechanisms of Oxidized α-Tocopherol on the Oxidative Stability of Soybean Oil...... 120 6.1 Abstract...... 120 6.2 Introduction...... 121 6.3 Materials and Methods...... 123 6.4 Results and Discussion ...... 126 6.5 Conclusion ...... 135 6.6 References...... 137

Conclusion ...... 149

Bibliography ...... 152

xii

LIST OF TABLES

Table Page

2.1. Tocopherol and tocotrienol contents in food products ...... 27

2.2. Bond dissociation energy (BDE) of water, phenol, and tocopherols...... 31

2.3. Standard one-electron reduction potentials for common free radicals ...... 32

4.1. Fatty acid composition of lard ...... 80

5.1. Effects of α-, β-, γ-, and δ-tocotrienol at 1.2 mM on headspace oxygen and peroxide value of 0.4 M lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs ...... 108

5.2. The linear regressions from Figure 5.7, 5.8, 5.9, and 5.10 and quenching rates of α-, β-, γ-, and δ-tocotrienol...... 109

6.1. Reaction rates of the lipid (RH) and tocopherol (T)...... 142

xiii

LIST OF FIGURES

Figure Page

2.1. Molecular orbital of triplet oxygen...... 7

2.2. Molecular orbital of singlet oxygen...... 8

2.3. Singlet oxygen formation by chemical, photochemical, and biological mechanisms...... 10

2.4. Chemical mechanism for the formation of singlet oxygen in the presence of sensitizer, light and triplet oxygen ...... 11

2.5. Structure of riboflavin...... 13

2.6. Photosensitization of riboflavin and Type I and Type II mechanisms ...... 15

2.7. Structure of chlorophyll ...... 17

2.8. Mechanisms of triplet oxygen oxidation with linoleic acid...... 19

2.9. Structure of tocopherols and tocotrienols ...... 23

3.1. Effects of 0, 3, 6, and 12 hrs on the volatile compounds from riboflavin solution under light ...... 56

3.2. Mass spectra of gas chromatographic peak with the retention time of 6.5 minutes (A) and authentic 2,3-butanedione (B) ...... 57

3.3. Proposed mechanism for the formation of 2,3-butanedione from riboflavin and singlet oxygen ...... 58

3.4. Effects of 0, 0.5, 1.0 and 5.0 mM sodium azide on the volatile compounds from riboflavin solution under light for 12 hrs...... 59

xiv 3.5. Formation of 2,3-butanedione from riboflavin in 0.1 M sodium phosphate buffer solution with pH 4.5, 6.5, and 8.5 under light for 12 hrs...... 60

4.1. Effect of 0, 100, 200, 300, 500, and 1000 ppm α-tocotrienol on the headspace oxygen of lard during dark storage for 7 days at 55°C...... 81

4.2. Effect of 0, 100, 200, 300, 500, and 1000 ppm β-tocotrienol on the headspace oxygen of lard during dark storage for 7 days at 55°C...... 82

4.3. Effect of 0, 100, 200, 300, 500, and 1000 ppm γ-tocotrienol on the headspace oxygen of lard during dark storage for 7 days at 55°C...... 83

4.4. Effect of 0, 100, 200, 300, 500, and 1000 ppm δ-tocotrienol on the headspace oxygen of lard during dark storage for 7 days at 55°C...... 84

4.5. Effect of 0, 100, 200, 300, 500, and 1000 ppm α-tocotrienol on the peroxide value of lard during dark storage for 7 days at 55°C...... 85

4.6. Effect of 0, 100, 200, 300, 500, and 1000 ppm β-tocotrienol on the peroxide value of lard during dark storage for 7 days at 55°C...... 86

4.7. Effect of 0, 100, 200, 300, 500, and 1000 ppm γ-tocotrienol on the peroxide value of lard during dark storage for 7 days at 55°C...... 87

4.8. Effect of 0, 100, 200, 300, 500, and 1000 ppm δ-tocotrienol on the peroxide value of lard during dark storage for 7 days at 55°C...... 88

5.1. Headspace oxygen depletion of 0, 0.1, 0.25, and 0.4 M lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs ...... 110

5.2. Schematic diagram for the formation of oxidized product via singlet oxygen oxidation under light: Sen = chlorophyll; A = lard; AO2 = oxidized lard; Q = tocotrienol ...... 111

5.3. Effect of 0, 0.3, 0.6, or 1.2 mM α-tocotrienol on the headspace oxygen depletion of lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs...... 112

5.4. Effect of 0, 0.3, 0.6, or 0.9 mM β-tocotrienol on the headspace oxygen depletion of lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs...... 113

xv 5.5. Effect of 0, 0.3, 0.6, or 0.9 mM γ-tocotrienol on the headspace oxygen depletion of lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs...... 114

5.6. Effect of 0, 0.3, 0.6, or 0.9 mM δ-tocotrienol on the headspace oxygen depletion of lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs...... 115

5.7. Relation of slope/intercept in Figure 5.3 to the concentration of α-tocotrienol . 116

5.8. Relation of slope/intercept in Figure 5.4 to the concentration of β-tocotrienol.. 117

5.9. Relation of slope/intercept in Figure 5.5 to the concentration of γ-tocotrienol .. 118

5.10. Relation of slope/intercept in Figure 5.6 to the concentration of δ-tocotrienol.. 119

6.1. Effects of 0, 250, 500, 1000 and 1500 ppm of oxidized α-tocopherol on the peroxide value of purified soybean oil during dark storage of 6 days at 55°C... 143

6.2. Effects of 0, 250, 500, 1000 and 1500 ppm of oxidized α-tocopherol on the headspace oxygen content (%) of purified soybean oil during dark storage of 6 days at 55°C ...... 144

6.3. Suggested possible formation of α-tocopherolquinone (A), 4a, 5-epoxy-α- tocopherolquinone (B), and 7, 8-epoxy-α-tocopherolquinone (C) from α- tocopheryl radical with triplet oxygen ...... 145

6.4. Suggested possible formation of α-tocopherolquinone (A), 4a, 5-epoxy-α- tocopherolquinone (B), and 7, 8-epoxy-α-tocopherolquinone (C) from α- tocopheryl radical with acyl peroxy radical...... 146

6.5. Suggested possible formation of α-tocopherolquinone hydroperoxide from α- tocopheryl radical with acyl peroxy radical...... 147

6.6. Suggested possible formation of dimerized α-tocopherol peroxide, α-tocopheryl oxy radical, and 7, 8-epoxy-α-tocopheryl oxy radical from α-tocopheryl radical ...... 148

xvi

CHAPTER 1

INTRODUCTION

The oxidation of foods causes significant nutritional losses, produces undesirable off-flavor, toxic, and color compounds, and lowers the sensory acceptability of foods to consumers (Nawar 1996; Min and Boff 2002a; Min and Boff 2002b). Oxidation reactions can occur due to the combination of diradical triplet oxygen and non-radical singlet oxygen. Triplet oxygen oxidation in foods is a typical free-radical chain reaction which needs high activation energy. The free-radical chain reaction includes initiation, propagation, and termination step. The propagation step is a slow step and responsible for the autocatalytic nature of triplet oxygen oxidation, which is called autoxidation

(Buettner 1993; Lee and others 2004). The reaction between free radicals and lipids, proteins, or vitamins in foods contributes to the loss of nutrients, the deterioration of flavor and the formation of potential toxic products (Frankel 1984; Nawar 1996; Min and

Boff 2002b).

1 Singlet oxygen can be formed in foods from triplet oxygen in the presence of a sensitizer such as chlorophyll or riboflavin under light (Bradley and Min 1992). A sensitizer becomes an excited singlet state after absorbing energy from light. The excited state sensitizer is formed from the excited singlet state through the intersystem crossing mechanism. The excited triplet sensitizer reacts with triplet molecular oxygen to form singlet oxygen by energy transfer (Min and Boff 2002a). Singlet oxygen can directly react with electron-rich double bonds without the formation of free radical intermediates.

The oxidation rate of singlet oxygen is 1,000-100,000 times faster than the ordinary triplet oxygen with food components (Doleiden and others 1974). Singlet oxygen rapidly accelerates the oxidation of foods even at -196°C (Andersen and others 2005). Singlet oxygen has been known to be involved in the reversion flavor in soybean oil (Min and others 2003), the sunlight flavor in milk (Jung and others 1998), oxidation of vitamin D in milk under light (Li and Min 1998), and decomposition of essential amino acids in foods stored under light (Bisby and others 1999).

The role of ordinary triplet oxygen in the autoxidation of foods has been extensively studied for last 70 year. However, the significant adverse effects of singlet oxygen on the quality of food have been recognized during last 30 years, especially in the area of lipids and vitamins which are typically the most sensitive to oxidative damage.

Riboflavin as vitamin B2 is a good photosensitizer for the formation of singlet oxygen under light (Bradley and others 2003). The singlet oxygen formed from riboflavin could destroy electron-rich riboflavin in foods by singlet oxygen oxidation. The compounds formed from riboflavin under light were identified as lumichrome and lumiflavin which

2 were produced from the excited triplet riboflavin (Huang and others 2006). However, the

reaction products between riboflavin and singlet oxygen have not been reported and the

chemical mechanisms of the compounds formed from riboflavin by singlet oxygen have

not been studied.

Antioxidants delay the onset of oxidation or slow the rate at which it proceeds in foods (Reische and others 2002). Tocopherols are the major lipid soluble antioxidants in nature (Burton and Ingold 1981; Kitts 1997; Belitz and Grosch 1999; Lee and others

2004). Tocopherols protect lipids from the oxidation by donating their hydrogen to lipid peroxy radicals (Nawar 1996). Tocotrienols are derivatives of 6-chromanol ring having phytyl group as the same as tocopherols but having three isolated double bonds in their phytyl group. Tocopherols have been reported to prevent food oxidation by scavenging free radicals (Jung and Min 1990; Kamal-Eldin and Appelqvist 1996; Belitz and Grosch

1999; Evans and others 2002; Reische and others 2002) and quenching singlet oxygen by charge transfer mechanism (Foote 1979; Choe and Min 2005). Although tocotrienols have similar structure with tocopherols, the antioxidant effect and chemical mechanism of tocotrienols has seldom been studied in foods.

Tocopherols can be antioxidants, neutral or prooxidants depending on temperature, pH, concentration, the presence of other compounds near to tocopherols, and their chemical characteristics (Kamal-Eldin and Appelqvist 1996; Rietjens and others 2002).

Tocopherols themselves are degraded by oxidation reaction (Verleyen and others 2001b).

The oxidized tocopherol products have been identified and reported as prooxidants in foods (Jung and Min 1992; Liebler and others 1996; Verleyen and others 2001a;

3 Verleyen and others 2001b). However, the detailed chemical mechanisms for the

formation and prooxidant effects of oxidized α-tocopherol in foods have not been reported.

The objectives of this dissertation were (1) to separate and identify the compounds formed from riboflavin by singlet oxygen under light and to study the chemical mechanisms for the formation of these compounds, (2) to study the quantitative and qualitative effects of α-, β-, γ-, and δ-tocotrienol on the oxidative stability of lard, (3) to determine the effects, quenching mechanisms and kinetics of α-, β-, γ-, and δ-

tocotrienol on the chlorophyll photosensitized oxidation of lard, and (4) to study the

effect of oxidized α-tocopherol on the oxidative stability of soybean oil and to postulate the formation and prooxidant mechanisms of oxidized α-tocopherol compounds during

soybean oil oxidation.

CHAPTER 2

LITERATURE REVIEW

2.1. The oxidation of foods

2.1.1. Chemistry of triplet and singlet oxygen

The oxidation of foods is due to the combination of triplet oxygen and singlet oxygen (Bradley and Min 1992; Choe and Min 2006). The most abundant oxygen in the air is triplet oxygen which was discovered in 1775 (Bradley and Min 1992; Min and Boff

2002a). The oxygen is a diatomic molecule and has paramagnetic characteristics which are due to the parallel spins of two outer electrons in the 2p π antibonding orbitals of the oxygen molecule (Bradley and Min 1992; Min and Boff 2002a). The molecular orbital of triplet oxygen is shown in Figure 2.1. Triplet oxygen exhibits three distinctive energy states under the electromagnetic field. The ground state of oxygen is called triplet state and diradical compound. Radical compounds can readily react with radical compounds and nonradical compounds react with nonradical compounds according to spin

5 conservation. Triplet oxygen, a diradical compound can react with only radical

compounds.

Singlet oxygen, another type of oxygen was discovered in 1934 and it has been studied for the last 30 years in food chemistry, biochemistry, medicine, and organic chemistry (Bradley and Min 1992). The molecular orbital of singlet oxygen is shown in

Figure 2.2. Molecular orbital of singlet oxygen is different with that of triplet oxygen.

One orbital of 2p π antibonding orbitals has paired electrons and the other is completely empty. Singlet oxygen is nonradical and electrophilic state. Singlet oxygen gives one energy level under the electromagnetic field.

Singlet oxygen is responsible for singlet oxygen oxidation in foods (Min and Boff

2002a; Choe and Min 2005). The energy of singlet oxygen is 22.4 kcal/mole above the ground state of triplet oxygen (Korycka-Dahl and Richardson 1978; Girotti 1998). The nonradical electrophilic singlet oxygen readily reacts with nonradical singlet state electron-rich compounds containing double bonds. The life time of singlet oxygen is from 2 to 700 microseconds depending on the solvent types (Long and Kearns 1975;

Choe and Min 2006). The activation energy of singlet oxygen is 0 to 6 kcal/mole which is low enough for the oxidation reactions (Min and Boff 2002a). Singlet oxygen has been studied in lipids and vitamins, which are most sensitive to the oxidative damage by singlet oxygen (Jung and Min 1991; Jung and others 1998; King and Min 1998; Huang and others 2004a).

Molecular σ* Atomic Atomic π* π*

π π 2px 2py 2pz 2pz 2py 2px σ

Energy σ*

2s 2s σ

σ* 1s 1s σ

Figure 2.1. Molecular orbital of triplet oxygen

Molecular σ* Atomic Atomic π* π*

π π 2px 2py 2pz 2pz 2py 2px σ

Energy σ*

2s 2s σ

σ* 1s 1s σ

Figure 2.2. Molecular orbital of singlet oxygen

2.1.2. Formation of singlet oxygen

Singlet oxygen can be formed chemically, enzymatically, and photochemically as shown in Figure 2.3. The formation of singlet oxygen in the presence of sensitizer and

8 triplet oxygen under light is most prevalent and shown in Figure 2.4. Photosensitizers

such as chlorophyll, myoglobin, riboflavin, porphyrins, pheopytins and synthetic colorant

can absorb energy from light and transfer it to react with triplet oxygen and form singlet

oxygen (Foote and Denny 1968; Jung and others 1991; King and Min 1998; Baron and

Andersen 2002; Huang and others 2004a).

The chemical mechanisms for the formation of singlet oxygen by photosensitizers under light are shown in Figure 2.4. When a ground singlet state photosensitizer (1Sen) is exposed to light, it becomes an excited singlet state (1Sen*). Fluorescence or heat is generated from 1Sen* by light emission or internal conversion, respectively. The 1Sen* produces an excited triplet sensitizer (3Sen*) by the intersystem crossing mechanism

(ISC). The emission of phosphorescence converts 3Sen* to ground state 1Sen. The 3Sen* reacts with nonradical triplet oxygen and loses higher energy to form singlet oxygen by triplet-triplet annihilation mechanism. The lifetime of 3Sen* is longer than 1Sen*. The returned 1Sen can begin another cycle of singlet oxygen formation. Sensitizers could generate 103 to 105 molecules of singlet oxygen before they are inactive (Kochevar and

Redmond 2000).

2R-CHOO· - R’

H2O2 3 3 2HOO· O2 + Sensitizer Irradiation 1 1O Sensitizer Fe3+ 2 2+ − 2+ H2O2 + Fe ·OH + OH Fe

Irradiation H· 3+ H2O2 Fe − H2O O · 2 Sensitizer·+

− + 3 H 3 3 O 2 + aqe O2 + Sensitizer + 3 R Xanthine oxidase + O2 HOO· H2O ROO· H2O2 + ·OH

Figure 2.3. Singlet oxygen formation by chemical, photochemical, and biological mechanisms

Excited 1 State Sen* Intersystem Crossing k = 1- 20×108 sec-1 Fluorescence hv 8 -1 3 k = 2×10 sec Sen*

3 Phosphorescence O2 9 -1 k = 10-104 sec-1 k = 1- 3×10 sec

Ground 1Sen 1 State O2

Figure 2.4. Chemical mechanism for the formation of singlet oxygen in the presence of sensitizer, light and triplet oxygen

The excited triplet sensitizer (3Sen*) reacts with triplet oxygen and can produce

superoxide anion by electron transfer (Choe and Min 2005). Superoxide anion produces

hydrogen peroxide by spontaneous dismutation. Hydrogen peroxide reacts with

superoxide anion to form singlet oxygen by Haber-Weiss reaction (Halliwell and

Gutteridge 2001). Haber-Weiss reaction occurs in the presence of transition metals such

as iron or copper.

11 - - + O2· + O2· + 2H H2O2 + O2 Dismutation - - 1 H2O2 + O2· ·OH + OH + O2 Haber-Weiss

Singlet oxygen is also produced by the Russell mechanism from peroxy radicals

(Halliwell and Gutteridge 2001).

R-CH-R’ R-CH-R’ R-CH-R’ R-C-R’= 1 O + O O + + O2 O O Russell H O · ·

2.1.3. Photosensitizer

2.1.3.1. Riboflavin

Riboflavin is a 7, 8-dimethyl-10-(1’-D-ribityl) isoalloxazine having many conjugated double bonds (Figure 2.5). Riboflavin is the prosthetic group of flavin enzymes, flavin mononucleotide (FMN) and flavin-adenine dinucleotide (FAD). FMN and FAD function as coenzymes that catalyze various oxidation-reduction processes which are important in metabolism of protein (Choe and others 2005). Riboflavin as vitamin B2 is required for red blood cell formation and respiration, antibody production, and regulating human growth and reproduction (Belitz and Grosch 1999; Siassi and

Ghadirian 2005). Deficiency of riboflavin causes the disorders in the skin and mucous membranes, particularly cracks and sores in the corners of the mouth, lesions of the lips,

12 and a red and sore tongue (Choe and others 2005; Siassi and Ghadirian 2005). The

recommended dietary reference intake of riboflavin is 1.1 to 1.6 mg/day depending on

age, sex, pregnancy, and lactation (Institute of Medicine 1998).

CH2OH CHOH CHOH CHOH

CH2

CH3 N N O NH CH3 N O

Figure 2.5. Structure of riboflavin

Riboflavin is stable during food processing and storage in the dark. However, riboflavin is very sensitive to UV and visible light. The 98% of riboflavin was destroyed during 96 hours storage under fluorescence light at room temperature, while only 4% was

13 lost in the dark (Huang and others 2004b). The loss of riboflavin depends on light

intensity, exposure time, and wavelength, packaging materials and food processing.

Riboflavin (RF) has been known as a photosensitizer which can be excited by

light and reacts with food components (Type I) or with triplet oxygen (Type II) as shown

in Figure 2.6. The 3RF* acts as a photochemically activated free-radical initiator for

radical compound (R·) formation. The radical compound (R·) initiates the free-radical

3 3 chain reaction. The RF* in Type I pathway can also react with O2 to form superoxide

– 3 3 – anion (O2· ) by electron transfer from RF* to O2. But the reaction to form O2· occurs less than 1% of the reaction between 3RF* and 3O2 (Kepka and Grossweiner 1972). The

3 3 1 RF* reacts with O2 to form singlet oxygen ( O2) by Type II mechanism by triplet-triplet

3 3 annihilation (Figure 2.6). More than 99% of the reaction between RF* and O2 produces

1 O2 (Kepka and Grossweiner 1972). The reaction rate of Type I or Type II mechanism depends on the solubility and concentration of oxygen in food system. As the oxygen in a system decreases, Type II mechanism is shifted to Type I. Oxygen is more soluble in lipids than in water (Ke and Ackman 1973). Type I reaction of riboflavin is favored due to the low concentration of oxygen in water matrix and easy oxidation-reduction property of riboflavin (McGinnis and others 1999).

Bradley and others (2003) proved the formation of singlet oxygen by riboflavin under light using electron spin resonance spectroscopy. The reaction rate between riboflavin and singlet oxygen was 1010 M-1sec-1 (Huang and others 2004b). The effect of riboflavin on the oxidation of amino acids, lipids, and vitamins under light has been reported (Jung and others 1998; King and Min 1998; Li and Min 1998; Huang and others

14 2004a). Sodium azide and ascorbic acid as singlet oxygen quenchers reduced the riboflavin-photosensitized oxidation (Huang and others 2004b).

Light ISC RF 1RF 3RF*

Type I Type II

3 RH O2

R· + RFH· 3 1 1 Type I + O2 O2 + RF − R·+ + RF·

3 O2, H· RH

– 1 ROOH O2· + RF ROOH

Figure 2.6. Photosensitization of riboflavin and Type I and Type II mechanisms

15 2.1.3.2. Chlorophyll

Chlorophyll is a green-colored pigment having a magnesium ion at the center of tetrapyrrole rings and a long phytol chain (Figure 2.7). Chlorophyll a (blue-green) and chlorophyll b (yellow-green) are found together in green plants in a ratio of 3:1 (Belitz and Grosch 1999). They are lipophilic due to the presence of the phytol group. The important characteristic of chlorophyll is the absorption of visible light between 400-500 nm and 600-700 nm. Chlorophyll is light sensitive and easily degraded into colorless compounds. The photodegradation of chlorophyll results in the opening of tetrapyrrole ring and destroy of porphyrin rings.

Chlorophyll acts as sensitizer to produce singlet oxygen in the presence of light and atmospheric triplet oxygen and accelerates the oxidation of foods. As the concentration of chlorophyll in purified soybean oil stored under light increased, the oxidative stability of oil decreased, but not in the dark (Jung and others 1991). Rahmani and Saari Csallany (1998) reported that the oxidation of virgin olive oil containing pheophytin which is the degradation product of chlorophyll was accelerated by fluorescent light. Chlorophyll acts as a photosensitizer through mostly Type II mechanism which forms singlet oxygen (Rawls and VanSanten 1970).

H CH2=CH R a, R: CH3 b, R: CHO H3C CH2CH3 N N H Mg H N N CH3 H3C

CH2 H O CO2CH3 H2C

CO2

CH3 CH3 CH3 CH3

Figure 2.7. Structure of chlorophyll

17 2.1.4. Triplet oxygen oxidation

Autoxidation by triplet oxygen and photosensitized oxidation by singlet oxygen are responsible for food oxidation. Triplet oxygen is a diradical compound and can react with radical compounds. Nonradical singlet state food components do not react with triplet oxygen due to spin conservation. Food components should be in a radical state to react with radical triplet oxygen for oxidation reaction. The hydrogen atom with the weakest bond on the carbon of food component will be removed first to become radicals.

The mechanism of triplet oxygen oxidation with linoleic acid is shown in Figure 2.8. The energy required to break the carbon-hydrogen bond on the C11 of linoleic acid is about

50 kcal/mole (Min and Boff 2002a). The double bonds at C9 and C12 decrease the carbon-hydrogen bond at C11 by withdrawing electrons. The carbon-hydrogen bond on the C8 or C14 of linoleic acid is about 75 kcal/mole and the carbon-hydrogen bond on the saturated carbon without any double bond next to it is approximately 100 kcal/mole (Min and Boff 2002a). The various strengths of carbon-hydrogen bond of fatty acids explain the differences of oxidation rates of stearic, oleic, linoleic and linolenic acids during oxidation. The relative oxidation rates for stearic, oleic, linoleic and linolenic acids are 1,

100, 1200, and 2500, respectively (Min and Boff 2002b).

The autoxidation has three steps: initiation, propagation, and termination (Figure

2.8). Initiation is a step for the formation of free alkyl radicals. The weakest carbon- hydrogen bond of linoleic acid is the one at C11 and the hydrogen at C11 will be

18 100 kcal/mole 50 kcal/mole 75 kcal/mole

CH3–(CH2)3–CH2–CH=CH–CH2–CH=CH–CH2–(CH2)6–COOH 14 13 12 11 10 9 Initiation – H (Metal, Energy) · 13 12 11 10 9 CH –(CH ) –CH–CH=CH–CH=CH–(CH ) –COOH 3 2 4 · 2 7 + O2

13 12 11 10 9 CH3–(CH2)4–CH–CH=CH–CH=CH–(CH2)7–COOH O Propagation + H O· · 12 11 10 CH3–(CH2)4–CH–CH=CH–CH=CH–(CH2)7–COOH O

Hydroperoxide O Decomposition H – · OH

CH3–(CH2)4–CH–CH=CH–CH=CH–(CH2)7–COOH O· O CH3–(CH2)3– CH2 · + C–CH=CH–CH=CH–(CH2)7–COOH H + H Termination ·

CH3–(CH2)3–CH3 Pentane

Figure 2.8. Mechanisms of triplet oxygen oxidation with linoleic acid

19 removed first to form radical at C11. The radical at C11 will be rearranged to form a

conjugated pentadienyl radical at C9 or C13 with trans double bond (Figure 2.8). Heat,

light, metals, and reactive oxygen species facilitate the radical formation of food

components. Triplet oxygen can react with conjugated double bond radicals of linoleic

acid and produce peroxyl radical at C9 or C13. The peroxyl radical with the standard

one-electron reduction potential of 1000 mV easily abstracts hydrogen from other fatty

acids and produces hydroperoxide and another alkyl radical (Choe and Min 2005). This

chain reaction is called free radical chain reaction and propagation step. The chain

reactions of free alkyl radicals and peroxyl radicals accelerate the oxidation. The alkoxyl

radicals react with other alkoxyl radicals or are decomposed to nonradical products. The

formation of nonradical volatile and nonvolatile compounds at the end of oxidation is

called the termination step (Figure 2.8).

2.1.5. Singlet oxygen oxidation

Singlet oxygen is non-radical, singlet state, and electrophilic compound due to its highest degenerate vacant molecular orbital (Figure 2.2). Singlet oxygen can react with non-radical, singlet state and electron-rich compounds containing double bonds without the formation of a radical. Singlet oxygen directly reacts with double bonds of the compound through the mechanisms of cycloaddition and “ene” reaction (Bradley and

Min 1992; Min and Boff 2002b). The reactions between singlet oxygen and

20 hydrocarbons with one or more double bonds produce endoperoxides, ally

hydroperoxides, or dioxetanes (Bradley and Min 1992; Choe and Min 2005).

Singlet oxygen oxidation depends on the number of double bonds instead of types of double bonds, such as conjugated or nonconjugated double bonds. The reaction rates between singlet oxygen and oleic, linoleic, linolenic, and arachidonic acids are 0.7, 1.3,

1.9, and 2.4 × 105 M-1sec-1, respectively (Vever-Bizet and others 1989), which are

relatively proportional to the number of double bonds. The reaction rate of singlet

oxygen with food components is much faster than that of triplet oxygen. The reaction

rates of triplet oxygen and singlet oxygen with linoleic acid are 8.9 × 101 and 1.3 × 105

M-1sec-1, respectively (Min and Boff 2002a).

Singlet oxygen can readily react with electron-rich compounds in foods such as

fatty acids, amino acids, and vitamins. Singlet oxygen has been known to be involved in

the reversion flavor in soybean oil (Min and others 2003), the sunlight flavor in milk

(Jung and others 1998; Bradley and others 2003), destruction of riboflavin (Huang and

others 2004b; Huang and others 2006), the oxidation of vitamin D in milk (King and Min

1998; Li and Min 1998), the decomposition of essential amino acids in foods (Bisby and

others 1999) and the oxidation of pork and turkey meat exposed to light (Whang and

Peng 1988).

21 2.2. Chemistry and antioxidative activity of tocopherols and tocotrienols

2.2.1. Chemical and structural characteristics

α-, β-, γ -, and δ-Tocopherols and α-, β-, γ -, and δ-tocotrienols are all derivatives of 6-chromanol having similar characteristics (Azzi and Stocker 2000). α-, β-, γ -, and δ-

Tocopherols consist of one hydroxyl group and one or more methyl groups at the 5, 7, or

8 position of the chromanol ring with a 16-carobon saturated phytyl group (Figure 2.9).

The phytyl group has three chiral centers at carbons 2, 4’, and 8’. All naturally occurring tocopherol isomers have the R-configuration at all three positions in their phytyl group,

RRR-α-tocopherol (2D, 4’D, 8’D) (Gregory 1996).

α-, β-,γ -, and δ-Tocotrienols have the same pattern on the chromanol ring but three isolated double bonds in their phytyl groups at the positions 3’, 7’, and 11’ (Figure

2.9). The presence of the double bonds at 3’ and 7’ of the phytyl group have four cis/trans geometrical isomers per tocotrienol. There are a total of eight isomers for each tocotrienol.

The positions of methyl groups on the chromanol ring determine the forms of α-,

β-, γ-, and δ-tocopherols and tocotrienols as shown in Figure 2.9. α-Tocopherol or tocotrienol has three methyl groups on the chromanol ring at the 5, 7, and 8 positions, while β-and γ-tocopherols or tocotrienols have two at the 5 and 8 positions and 7 and 8 positions, respectively. δ-Tocopherol or tocotrienol has only one methyl group at the 8 position. The chromanol ring only is responsible for the antioxidant potential of the tocopherols and tocotrienols. The phytyl group which is very lipophilic has no effect on

22 the chemical reactivity of antioxidants but is important for proper positioning in the biomembranes (Suzuki and Packer 1993).

R1 4a HO 5 4 CH3 CH3 CH3 6 3 CH3 7 8 1 2 8a 2′ 4′ 6′ 8′ 10′ 12′ R2 O 1′ 3′ 5′ 7′ 9′ 11′ CH3 CH3 Tocopherol

HO CH3 CH3 CH3 CH3

R2 O CH3 CH3 Tocotrienol

Trivial Name Chemical Name R1 R2

α-Tocopherol/Tocotrienol 5,7,8-Trimethyltocopherol/tocotrienol CH3 CH3

β-Tocopherol/Tocotrienol 5,8-Dimethyltocopherol/tocotrienol CH3 H

γ-Tocopherol/Tocotrienol 7,8-Dimethyltocopherol/tocotrienol H CH3 δ-Tocopherol/Tocotrienol 8-Methyltocopherol/tocotrienol H H

Figure 2.9. Structure of tocopherols and tocotrienols

23 2.2.2. Vitamin E activity

α-, β-, γ- and δ-Tocopherol and α-, β-, γ- and δ-tocotrienol have vitamin E activity. Vitamin E activity is defined in terms of International Units (IU) equivalent to the activity of α-tocopheryl acetate (Bramley and others 2000). Since the IU does not reflect the bioavailability, biological activity is now defined in terms of α-tocopherol equivalent (α-TE mg) (Burton and others 1998). The biological activities of RRR-α, β-,

γ-, and δ-tocopherol are 1.0, 0.5, 0.1, and 0.03 α-TE mg, respectively (Bramley and others 2000). The activities of α-, β-, and γ-tocotrienol are 0.3, 0.05, and 0.01 α-TE mg, respectively (Drotleff and Ternes 1999). The natural α-tocopherol is biologically most potent. α-Tocotrienol only has 30% of the vitamin E activity and β-tocotrienol has 10% of the biological activity of β-tocopherol. The biological activity of vitamin E depends on structure specific including the presence or absence of methyl groups on the chromanol ring, the stereochemistry of chiral carbon centers, and branching or desaturation of the side chain (Azzi and Stocker 2000).

The recommended daily intake of vitamin E is 15 mg (Institute of Medicine 2000).

A tolerable upper intake level is 1000 mg per day (Bendich and Machlin 1988). The recommended daily intake increases as the content of unsaturated fatty acids in a diet increases (Belitz and Grosch 1999). The requirements of α-tocopherol based on the contents of monoene, diene and triene and hexaene fatty acids are 0.09, 0.6, 0.9, and 1.8 mg/g fatty acid, respectively (Belitz and Grosch 1999). Tocopherol has been associated with the reduction of heart disease, delay of Alzheimer’s disease, and prevention of

24 cancer (Schneider 2005). Tocotrienols are more mobile within the biological membrane

(Theriault and others 1999). Tocotrienols have been shown to prevent cardiovascular

disease and cancer and to lower cholesterols (Watkins and others 1993; Theriault and

others 1999).

2.2.3. Stability during processing and storage

The stability of tocopherols and tocotrienols are affected by processing and storage of foods. Tocopherols and tocotrienols are retained 60~70% through out the edible oil extraction and refining process although there is some loss during the deodorization step (Bramley and others 2000). Tocopherol contents in rapeseed oil are

794 and 749 ppm for hexane extracted oil and pressed oil, respectively. γ- and δ-

Tocotrienols of safflower oil were from 3.8 to 7.0 and from 7.5 to 7.8 ppm at different

roasting temperature (Lee and others 2004). Tocopherols in crude and deodorized

soybean oil were 1670 and 1138 ppm, respectively (Jung and others 1989). The

deodorization processing removed 31.8% tocopherols of crude soybean oil. Although

total tocopherol content decreased during processing, the relative compositions of α-, β-,

γ-, and δ-tocopherols in soybean oils were constant (Jung and others 1989).

Milling and breadmaking in industries cause the losses of tocopherols and

tocotrienols. Tocopherol and tocotrienol contents become concentrated in certain

fractions such as bran and germ (Bramley and others 2000). The total tocopherol and

25 tocotrienol contents of wheat is 49 ppm but those of wheat germ and wheat bran are 1920

and 911 ppm, respectively (Table 2.1). Wennemark and Jaegerstad (1992) reported that

dough making resulted in 20 to 40% losses of vitamin E in French bread and wheat/rye

bread.

The losses of vitamin E during the storage of vegetable oil were observed. Corn oil lost 25% of α- and γ-tocopherols for the 15 months of storage at 4°C (Bauernfeind

1980). Safflower oil lost 70% of tocopherols during 3 months at 37°C (Bauernfeind

1980). The storage of 6 months at room temperature resulted in an increase of

hydroperoxides concentration in vegetable oil which negatively correlated with the loss

of total tocopherol content (Nourooz-Zadeh 1998).

Tocopherols Tocotrienols Total Products (ppm) (ppm) (ppm) α β γ δ α β Oils Wheatgerm 1330 710 260 271 26 181 2778 Soybean 116 34 737 275 2 1 1165 Palm 256 - 316 70 146 32 820 Canola 272 0.1 423 - 0.4 - 685 Sunflower 613 17 19 - - - 649 Corn 134 18 412 39 - - 603 Roasted sesame 4 - 584 9 - - 597 Rapeseed 252 - 314 - - - 566 Safflower 386~520 8.6~12.4 2.4~7.7 - - - 379~540 Olive 168~226 - - - - - 168~226

Cereals Corn 6 - 45 - 3 - 54 Wheat 10 7 - - 4 28 49 Rye 16 4 - - 15 8 43 Oats 5 1 - - 11 2 19 Barley 2 0.4 0.3 0.1 11 3 17 Wheat germ 1153 660 - - 26 81 1920 Wheat bran 163 101 - - 110 537 911

Nuts Almonds 452 3 19 1 2 - 477 Hazelnuts 215 - 1 0.1 - - 216 Peanuts 114 - 84 - - - 198

Seeds Sunflower 495 27 - - - - 522 Sesame - - 227 - - - 227

Table 2.1. Tocopherol and tocotrienol contents in food products

27 2.2.4. Antioxidative activity of tocopherols

Tocopherols are the best known antioxidants in nature to prevent lipid oxidation

(Burton and Ingold 1981; Min and Boff 2002a; Lee and others 2004). Tocopherols compete with unsaturated fatty acids for lipid peroxy radicals. Tocopherols donate a hydrogen atom at the 6-hydroxy group on its chroman ring to lipid peroxy radical.

Tocopherol (TH) with a reduction potential of 300-400 mV easily donates hydrogen to lipid peroxy radical (ROO·) with a reduction potential of 1000 mV and produces lipid hydroperoxide (ROOH) and tocopheroxy radical (T·).

TH + ROO· T· + ROOH

Tocopheroxy radicals (T·) are resonance structures which are more stable than lipid peroxy radicals (ROO·). The reaction rate of α-tocopherol with lipid peroxy radical is 107 M-1sec-1 (Niki and others 1984; Choe and Min 2005) and is 105 to 106 times faster than that of unsaturated lipid with lipid peroxy radical (Niki and others 1984; Naumov and Vasil’ev 2003). Tocopherols take away the radicals from the oxidizing fatty acids to prevent further radical chain reactions. One tocopherol molecule can protect about 103 to

108 polyunsaturated fatty acid molecules at low peroxide value (Kamal-Eldin and

Appelqvist 1996). Tocopheroxy radical (T·) can interact with other compounds or each other depending on the lipid oxidation rates. Tocopheroxy radicals may react with lipid

28 peroxy radicals or another tocopheroxy radical and form non radical products (Kamal-

Eldin and Appelqvist 1996).

T· + ROO· T-OOR

T· + T· T-T

The effectiveness of tocopherols as antioxidants depends on the chemical and physical characteristics. The chemical structures of α-, β-, γ-, and δ-tocopherol support a

hydrogen-donating power in the order of α- > β- or γ- > δ-tocopherol (Burton and Ingold

1981; Wright and others 2001; Evans and others 2002). Isnardy and others (2003)

reported that α-tocopherol degraded faster than γ- or δ-tocopherol in purified rapeseed oil.

α-Tocopherol was completely destroyed during the oxidation of soybean oil while most

of γ- and δ-tocopherol remained (Player and others 2006). Tocopherols are the chain-

breaking antioxidants by donating their phenolic hydrogen to lipid free radicals (Nawar

1996). As the antioxidant activity or hydrogen-donating ability of tocopherol is high, the

stability of tocopherol in vegetable oil is low (Jung and Min 1990). The reaction rates of

α-, γ-, and δ-tocopherol with lipid peroxy radical at 30°C were 2.4 ×106 M-1sec-1, 1.6 ×

106 M-1sec-1, and 0.7 ×106 M-1sec-1, respectively (Belitz and Grosch 1999). α-

Tocopherol had the highest antioxidant activity in vegetable oil with the least stability

during storage (Huang and others 1994; Kamal-Eldin and Appelqvist 1996).

The antioxidant activities of α-, β-, γ-, and δ-tocopherol can be explained by bond

dissociation energy (BDE). The BDE measures the bond strength in a chemical bond.

29 The BDE in antioxidant reactivity indicates the driving force for the hydrogen transfer

from the phenolic antioxidant to lipid radical. The BDE depends on the strength of O-H

bond in the phenolic antioxidant. The lower BDE means the easier donation of hydrogen

to lipid peroxy radical by cleaving the O-H bond.

The BDE of water, phenol, and α-, β-, γ- and δ-tocopherol are shown in Table 2.2.

The O-H bond energy for phenol (87 kcal/mol) is considerably lower than water (119.3 kcal/mol). The BDE of hydroxyl group on the chromanol ring of α-, β-, γ-, and δ- tocopherol are 75.8, 77.7, 78.2, and 79.8 kcal/mol, respectively (Wright and others 2001).

The O-H bond of hydroxyl group in α-tocopherol can be more easily cleaved to donate a hydrogen atom to lipid peroxy radical than that of β-, γ-, or δ-tocopherol. α-Tocopherol

is a fully methylated chromanol (Figure 2.9) which is more sterically hindered than β-, γ-,

or δ-tocopherol. Sterically hindered phenols are the most active antioxidants for

scavenging lipid peroxy radicals by donating hydrogen-atoms (Wright and others 2001).

The fully methylated α-tocopherol having lower BDE could be more potent as a

hydrogen donor than β-, γ- or δ-tocopherol (Kamal-Eldin and Appelqvist 1996; Wright

and others 2001). The BDE also indicates the stabilization of the resulting radical. The

most stable radicals can be derived from their parent compounds with the lowest BDE,

which will be the most efficient hydrogen atom donors. α-Tocopheroxy radical will be

the most stable radicals among tocopherol homologues.

BDE Compound (kcal/mol) Water 119.3 Phenol 87.0 α-tocopherol 75.8 β-tocopherol 77.7 γ-tocopherol 78.2 δ-tocopherol 79.8

(Berkowitz and others 1994; Bordwell and Liu 1996; Wright and others 2001)

Table 2.2. Bond dissociation energy (BDE) of water, phenol, and tocopherols

The standard one-electron reduction potentials of common free radicals and α-, β-,

γ-, and δ-tocopherols are listed in Table 2.3. The reduction potentials of lipid alkyl, peroxy, and alkoxyl radicals are 600, 1000, and 1600 mV, respectively. To work as an antioxidant and prevent lipid oxidation, the reduction potential of a free radical scavenger should be lower than 600 mV which is a reduction potential of lipid alkyl radical (Lee and others 2004). Tocopherols have lower standard one-electron reduction potential than lipid alkyl, peroxy, and alkoxyl radicals (Table 2.3). Therefore, tocopherols can be easily

31 oxidized by donating a hydrogen atom to lipid peroxy radical which will be a reduced

form (Buettner 1993). The reduction potentials of α-, β-, γ-, and δ-tocopherol are 270,

345, 350, and 405 mV, respectively. Therefore, α-tocopherol is the best reducing agent

to donate a hydrogen atom.

Standard reduction potential Compounds Half-cell (mV) + ·OH H /H2O 2310 RO·a H+/ROH 1600 ROO·a H+/ROOH 1000 R· a H+/RH 600 α-Tocopheryl· H+/α-Tocopherol 270 β-Tocopheryl· H+/β-Tocopherol 345 γ-Tocopheryl· H+/γ-Tocopherol 350 δ-Tocopheryl· H+/δ-Tocopherol 405 a RO·, ROO·, and R· are lipid alkoxy, peroxy, and alkyl radical, respectively. b RO· and ROO· forms in tocopherol (Chapter 6) are tocopheryl oxy radical and tocopheryl peroxy radical, respectively, and the structures are shown in Figure 6.3 to 6.6.

(Kamal-Eldin and Appelqvist 1996; Choe and Min 2005)

Table 2.3. Standard one-electron reduction potentials for common free radicals

32 Tocopherol can quench singlet oxygen at the rate of 107 M-1sec-1 (Jung and others

1991). The singlet oxygen quenching effect of tocopherols decreases in the order of α-,

β-, γ-, and δ-tocopherol (Jung and others 1991; Mukai and others 1993). Foote (1979) found that tocopherol forms a charge transfer complex with singlet oxygen by an electron donation from tocopherol to singlet oxygen. The singlet state of the tocopherol-singlet oxygen complex undergoes intersystem crossing (ISC) into the triplet state. They produce tocopherol and triplet oxygen which is less reactive.

ISC 1 + 1 + 1 3 T + O2 → [T − O2]1 → [T − O2]3 → T + O2

When there is no chemical reaction between tocopherol and singlet oxygen, physical quenching occurs (Choe and Min 2005). Tocopherols produce oxidized tocopherols when they react irreversibly with singlet oxygen in chemical quenching

(Choe and Min 2005).

2.2.5. Antioxidative activity of tocotrienols

Tocotrienols are similar in structure with tocopherols and donate hydrogen from the hydroxyl group of a chromanol ring to reduce lipid radicals. Tocotrienols are thought to be more potent in antioxidant properties than α-tocopherol (Serbinova and others

1991). The unsaturated phytyl group in tocotrienols makes them penetrate into tissues

33 which have saturated fatty layers such as brain and liver more efficiently (Suzuki and

Packer 1993). Tocotrienols have shown to lower plasma cholesterol levels

(Hasselwander and others 2002) and to reduce lipid and non-lipid related risk factors for

cardiovascular disease (Newaz and Nawal 1999). Tocotrienols had better anti-tumor

activity than α-tocopherol (Newaz and Nawal 1999; Conte and others 2004).

Tocotrienols have been reported to be better antioxidants than tocopherols in

foods due to their higher recycling efficiency from chromanoxyl radicals which correlates

with inhibition of lipid oxidation (Lehmann and Slover 1976; Ping and May 2000). Little

information exists on tocotrienol antioxidant properties in foods because of lower

biological vitamin E activity than tocopherols. Lehmann and Slover (1976) first reported

that γ- and δ-tocotrienol had better antioxidant effects than α-tocotrienol at the

concentration of 500 ppm which was similar to α-tocopherol antioxidant potential. γ-

Tocotrienol was a more effective antioxidant than α-tocotrienol and α-tocopherol under

heating conditions in stripped oils (Feng 1996). Wagner and others (2001) reported that

δ- and γ-tocotrienols increased the shelf-life of coconut fat at ambient temperature and the antioxidative potential at frying conditions increased in the order of α-tocopherol=α- tocotrienol<β-tocotrienol<γ-tocopherol<γ-tocotrienol<δ-tocopherol<δ-tocotrienol.

Schroeder and others (2006) showed that α-tocotrienol in palm olein acted as a more effective antioxidant in a simulated frying experiment than α-tocopherol.

34 2.2.6. Prooxidative activity of tocopherols

Tocopherols can be antioxidants, neutral or prooxidants depending on temperature, pH, concentration, the presence of other compounds near to tocopherols, and their chemical characteristics (Gregory 1996; Kamal-Eldin and Appelqvist 1996; Verleyen and others 2001b; Reische and others 2002).

The prooxidant effect of α-tocopherol has been proposed to be induced by hydrogen abstraction between the tocopheroxyl radical (T·) and lipid molecules or lipid hydroperoxides (Terao and Matsushita 1986).

RH + T· TH + R·

ROOH + T· TH + ROO·

The reaction rate constants of α-tocopheroxyl radicals with polyunsaturated fatty acids or with hydroperoxides of polyunsaturated fatty acids were reported in the range of

10-5 to 10-1 M-1sec-1 (Kamal-Eldin and Appelqvist 1996). The reactions are very slow compared to the antioxidative reactions of α-tocopherol and the termination reaction of lipid autoxidation. The resulting antioxidant radical must not propagate the chain reaction and will not undergo hydrogen-abstraction reactions or react with oxygen to form another peroxy radical. Thus, these reactions cannot totally explain the prooxidant effect of α-tocopherol.

35 The alternative prooxidant mechanism of tocopherols has been suggested to be more significant in the presence of high levels of hydroperoxides (Hicks and Gebicki

1981). This reaction mechanism involves hydrogen bonding between tocopherol (TO-H) and lipid hydroperoxide (RO-O-H). The peroxide abstracts a hydrogen from the tocopherol and cleaves the O-O bond in the peroxide. As a result of this reaction, the alkoxyl radical (RO·) is formed and then propagates lipid oxidation due to its high reduction potential.

H TOH + ROOH ROO RO· + H2O + TO· H

Tocopherols are degraded in the presence of molecular oxygen and produce oxidized products resulting in the loss of antioxidant activity. The oxidized tocopherol products act as prooxidants in lipids. The addition of oxidized α-, γ-, and δ-tocopherols to soybean oil lowered the oxidative stability of soybean oil (Jung and Min 1992). The oxidation of tocopherols by strong oxidizing agents such as chromic acid, nitric acid, and ferric chloride generally produce lactones, quinines, and many degradation products

(Kamal-Eldin and Appelqvist 1996). α-Tocopherolquinone, α-tocopherolhydroquinone,

4a, 5-epoxy-α-tocopherolquinone, and 7, 8-epoxy-α-tocopherolquinone have been reported as the oxidation products of α-tocopherol (Liebler and others 1996; Faustman and others 1999; Verleyen and others 2001a; Verleyen and others 2001b; Pazos and

36 others 2005). During the oxidation of α-tocopherol, many intermediate compounds could be produced. Rietjens and others (2002) suggested that increased levels of oxidized α- tocopherol could result in increased levels of α-tocopherol radicals, which can initiate lipid peroxidation by themselves.

The concentration of tocopherol is important whether it will be an antioxidant or prooxidant. Generally, the antioxidant activity of tocopherol is greatest at lower concentration and decreases at higher concentration. The optimum concentrations of α-,

γ-, and δ-tocopherol to increase the oxidative stability of oil were 100, 250-500, and 500-

1000 ppm, respectively (Jung and Min 1990; Evans and others 2002). α-Tocopherol at high concentrations acts as a prooxidant during the oxidation of lipids resulting in the increase of hydroperoxide levels and conjugated dienes (Jung and Min 1990; Bowry and

Stocker 1993; Evans and others 2002; Yoshida and others 2003). It is important to prevent the oxidation of tocopherol and remove the oxidized tocopherols.

CHAPTER 3

PHOTOINDUCED GENERATION OF 2,3-BUTANEDIONE FROM RIBOFLAVIN

3.1 Abstract

The volatile compound formation from riboflavin solution of 0.1 M phosphate buffer at pH 6.5 under light for 15 hours was studied by SPME-GC/MS analysis. Only one major compound in the riboflavin solution was formed and increased as the light exposure time increased. The light exposed riboflavin solution had a buttery odor. The compound formed from riboflavin solution under light was analyzed by GC and olfactometry. The major volatile compound eluted from GC had a buttery odor. The buttery odor compound was positively identified as 2,3-butanedione by a combination of gas chromatographic retention time, mass spectrum and odor evaluation of authentic 2,3- butanedione. The addition of sodium azide, a singlet oxygen quencher, to riboflavin solution minimized the formation of 2,3-butanedione. Singlet oxygen was involved in the formation of 2,3-butanedione. The 2,3-butanedione was produced from the reaction

38 between riboflavin and singlet oxygen. Singlet oxygen was formed from triplet oxygen by riboflavin photosensitization mechanism. This is the first report on the oxidation between riboflavin and singlet oxygen in food and biological systems.

3.2 Introduction

Riboflavin, vitamin B2, is an active part of the coenzymes of flavin mononucleotide and flavin adenine dinucleotide, which catalyze many oxidation- reduction reactions in biological systems. These coenzymes play essential roles in several dehydrogenases and oxidases. Riboflavin exists in milk, eggs, meats, vegetables and many other food products (Szczesniak and others 1971; Gliszczynska and Koziolowa

1999; USDA 2004). Milk is the most important source of riboflavin in the diets in the

United States and many other nations. Riboflavin contents in whole and skim milk is

1.5-2.0 µg/mL (Jung and others 1998). Light destroys riboflavin in milk rapidly (Allen and Parks 1979; Jung and others 1998; Lee and others 1998). Riboflavin in foods is extremely unstable under light, but very stable in the dark (Ahmad and others 2004;

Huang and others 2004a).

Riboflavin has complex photochemical properties (Kim and others 1993; Edwards and others 1999; Criado and others 2003) and has been extensively studied as a photosensitizer in foods (Jung and others 1995; Jung and others 1998; King and Min

2002; Rosenthal and others 2003; Huang and others 2004b) and biological systems

39 (Lucius and others 1998; Grzelak and others 2001). Riboflavin produced singlet oxygen

from ordinary triplet oxygen under light by the excited triplet riboflavin and triplet

oxygen annihilation mechanism (Choe and others 2005). The direct detection of singlet

oxygen was extremely difficult because it has only about two microseconds of life time

depending on solvent (Bradley and others 2003). The singlet oxygen formed by

riboflavin photosensitization was trapped by 2, 2, 6,

6-tetramethyl-4-piperidone and produced stable 2, 2, 6, 6-tetramethyl-4-

piperidone-1-oxyl radical. The 2, 2, 6, 6-tetramethyl-4-piperidone-1-oxyl radical was

detected by electron spin resonance spectroscopy (Bradley and others 2003). The

electron spin resonance spectrum of 2, 2, 6, 6-tetramethyl-4-piperidone-1-oxyl radical

formed from riboflavin under light had three hyperfine lines and the hyperfine coupling

constant and G-factor were 16.1G and 2.0048, respectively (Bradley and others ). The

reaction rate between riboflavin and singlet oxygen was 1.01 × 1010 M-1sec-1 (Huang and others 2004a). This explains the extremely fast degradation of riboflavin in foods under light. Sodium azide reduces the degradation of riboflavin under light by quenching singlet oxygen at the rate of 1.55 × 107 M-1sec-1 (Huang and others 2004a). The nonvolatile compounds formed from riboflavin under light were lumichrome and lumiflavin (Huang and others 2006). The lumichrome and lumiflavin were produced from the excited triplet riboflavin by cleaving 3, 4, 5-trihydroxy-2-pentanone and 2, 3, 4- trihydroxybutanal, respectively. Singlet oxygen was not involved in the production of lumichrome and lumiflavin (Huang and others 2006). The reaction products between riboflavin and triplet or singlet oxygen, the identification and chemical mechanisms of

40 the volatile compounds formed from riboflavin, and the flavor qualities of the volatile

compounds formed from riboflavin under light have not been reported in the literature.

The objectives of this study were to identify the volatile compounds formed from riboflavin under light and to study the chemical mechanisms for the formation of volatile compounds under light and the flavor properties of the volatile compounds from riboflavin.

3.3 Materials and Methods

Materials

Riboflavin and sodium azide were purchased from Sigma Chemical Co. (St. Louis,

MO). Millipore purification system with a filter of Progard 2 and Milli-Q Plus ultra-pure water system with Purification Pak were purchased from Millipore Co (Bedford, MA).

Serum bottles (10 mL), aluminum caps, and Teflon™-coated septa were purchased from

Supelco, Inc. (Bellefonte, PA). The 15 mm × 1.5 mm magnetic bars were purchased from Bel-Art Products (Pequannock, NJ).

Riboflavin solution preparation

Phosphate buffer (0.1 M, pH 6.5) was prepared with the freshly purified water which was first purified by a Millipore purification system with a filter of Progard and then further purified by Milli-Q Plus ultra-pure water system with Purification Pak.

41 Riboflavin solution (0.5 mM) in the phosphate buffer (0.1 M, pH 6.5) was prepared and

20 mL of the riboflavin solution was, in triplicate, transferred into a 100 mL-Erlenmeyer

flask. The flask was sealed with a parafilm and the riboflavin solution was stored in a

light storage box of 3000 lux for 0, 3, 6, 9, 12 and 15 hrs at 25oC. The riboflavin

solutions stored under light for 0, 3, 6, 9, 12 and 15 hrs were analyzed for volatile

compounds.

Analysis of headspace volatile compounds of riboflavin solution by SPME-GC

The 3 mL of the riboflavin sample solution stored under light or in dark was

pipetted into a 15 mL serum bottle having a 15 mm × 1.5 mm magnetic bar. The serum

bottle was air-tightly sealed with a Teflon-coated rubber septum and aluminum cap.

Headspace volatile compounds of riboflavin serum bottle were analyzed by SPME

method with a 75 µm Carboxen/polydimethylsiloxane fiber (Supelco Inc., Bellefonte,

PA). The sample serum bottle was kept in a 40 °C water bath for 20 min to equilibrate the sample temperature and to have a good reproducibility for SPME analysis. The

SPME fiber was inserted into the headspace of temperature equilibrated riboflavin solution bottle and exposed for 20 minutes in a 40 °C water bath. The riboflavin solution in the serum bottle was continuously stirred with a 15 mm × 1.5 mm magnetic bar to increase the extraction of headspace compounds and to improve the reproducibility of

SPME analysis. The volatile compounds adsorbed in the SPME fiber was desorbed into the injection port of a gas chromatograph (Shimadzu GC-14B, Shimadzu, Japan) for 5 minutes at 250 °C. The injection port was fitted with a 0.75 mm internal diameter

42 splitless glass liner. Supelco wax 10-fused silica capillary column (60 m × 0.32 mm,

0.25 µm film thickness; Supelco Inc., Bellefonte, PA), a flame ionization detector, and helium gas were used for gas chromatography. The GC oven temperature was held at 40

°C for 2 minutes, increased to 170 °C at 7 °C/minute and held at 170 °C for 2 minutes.

Gas chromatography-olfactometry

The odor characteristics of the effluents from the gas chromatograph were

analyzed by Gas chromatography-Olfactometry. The effluent from capillary gas

chromatographic column was split to a sniffing port by a glass seal Y connector (Supelco

Inc., Bellefonte, PA). The riboflavin solution under light for 15 hrs was used for the gas

chromatography-olfactometry analysis. The SPME method was used for the extraction of

volatile compounds from the sample headspace as described above. The gas

chromatographic conditions and column used for the gas chromatography-olfactometry

analysis were the same as SPME-GC analysis.

Gas chromatography-mass spectrometry (GC/MS)

The identification of the volatile compounds was carried out with a Gas

Chromatograph-Mass Spectrometer (Perkin-Elmer). The electron ionization of mass

spectra was 70 eV. The gas chromatographic conditions for GC/MS were identical to

those used for the SPME-GC analysis.

43 Effect of sodium azide on volatile compound formation of riboflavin solution

To study the possible involvement of singlet oxygen in the volatile compound formation of the riboflavin solution, the 0.5 mM riboflavin solution containing 0, 0.5, 1.0 and 5.0 mM sodium azide were prepared. The riboflavin solutions (20 mL) were transferred into 100 mL-Erlenmeyer flasks, and the flasks were sealed with parafilm and stored in the light storage box at 3,000 lux for 12 hrs.

Effects of pH on the formation of 2,3-butanedione from riboflavin after storage under light

The samples contained riboflavin in 0.1 M phosphate buffer with pH 4.5, 6.5 and

8.5 were prepared as previously described. The samples were stored under light for 12 hours. The quantities of the 2,3-butanedione formed in the samples were analyzed with the standard curves obtained from the added known amount of authentic 2,3-butanedione in the each tested buffer solution.

Statistical analysis

All the experiments for the analysis of volatile compounds in the riboflavin solutions were done in triplicate. Data were analyzed using Microsoft Office Excel

Program. Comparisons for mean value differences were done by T-test. The p-value

≤0.05 was considered to be significantly different.

44 3.4 Results and Discussion

Riboflavin solution preparation

It was difficult to have a reproducible analysis for the volatile compounds formed from riboflavin solution under light in a preliminary work. The quality of water was extremely important for the determination of volatile compounds from riboflavin solution under light. The stored deionized or distilled water did not give a good reproducible result. The stored water might have absorbed various volatile compounds from environment. The absorbed volatile compounds in the stored water might have interfered with the volatile compounds formed from riboflavin solution under light during gas chromatograph analysis.

Distilled water was first purified by a Millipore purification system with a filter of

Progard and then further purified by Milli-Q Plus ultra-pure water system with

Purification Pak for the preparation of riboflavin solution. The riboflavin solution prepared with the freshly purified water provided good reproducible gas chromatograms for the volatile compounds of riboflavin solution. It was extremely important to use the freshly purified water for the preparation of riboflavin solution. It is also very important to seal the sample flask with a parafilm for the storage under fluorescence light (3000 lux). These might explain the difficulties of separation and identification of the volatile compounds formed from riboflavin under light which have not been reported in literatures.

45 Analysis of headspace volatile compounds of riboflavin solution by SPME-GC

Aqueous solution of riboflavin (0.5 mM) was prepared and stored under light at

3,000 lux for 0, 3, 6, 9, 12 and 15 hrs. Some of the sample serum bottles were completely wrapped with aluminum foils to protect the riboflavin solution from light.

The wrapped serum bottles were designated as the samples stored in dark. Therefore, there were riboflavin solution serum bottles stored under light and in dark.

The gas chromatograms of volatile compounds from the riboflavin solutions under light after 0, 3, 6, and 12 hrs are shown in Figure 3.1. The fresh riboflavin solution without any storage under light or in dark showed several small gas chromatographic peaks as shown in Figure 3.1 at 0 hr. The buffer solution without riboflavin also showed the same gas chromatogram of 0 hr riboflavin solution. The several small peaks shown in

Figure 3.1 at 0 hr were not from riboflavin. The several peaks at extremely low concentration might be due to the column bleeding or noise signal of gas chromatograph.

The light produced and increased only one major peak with the retention time of

6.5 minutes (Figure 3.1). The major peak with the retention time of 6.5 minutes increased as the storage time under light increased from 0, 3, 6, or 9 to 12 hrs. The area of the gas chromatographic peak with the retention time of 6.5 minutes in the riboflavin solution under light for 0, 3, 6, 9, 12 or 15 hrs were 0, 2.34, 6.78, 10.46, 12.18, or 11.71

(mV × sec), respectively. The gas chromatographic peak area at the retention time of 6.5 minutes from the riboflavin solution under light for 12 hrs was significantly higher than those for 3, 6, and 9 hrs at p<0.05. The peak area did not increase as the storage time under light increased from 12 to 15 hrs and no significant difference at p>0.05. The

46 riboflavin solution did not produce any volatile compounds during the storage from 0 to

15 hrs in dark. The buffer solution without riboflavin under light for 15 hrs also did not

produce any volatile compounds. The results suggested that both riboflavin and light

were required to produce the gas chromatographic peak with the retention time of 6.5

minutes.

Gas chromatography and olfactometry

The aqueous riboflavin solution under light had a buttery odor. The odor of the effluent from a sniffing port of gas chromatograph with the retention time of 6.5 minutes was a buttery flavor. The buttery odor from the gas chromatographic peak at the retention time of 6.5 minutes was the same odor of the riboflavin solution stored under light. The result suggested that the gas chromatographic peak with the retention time of

6.5 minutes was responsible for the odor formed from riboflavin solution under light.

Identification of volatile compound

The mass spectrum of the gas chromatographic peak with the retention time of 6.5 minutes is shown in Figure 3.2 (A). The mass spectrum was tentatively identified as 2,3-

butanedione. The gas chromatographic retention time of authentic 2,3-butanedione was

6.5 minutes. The mass spectrum of the authentic 2,3-butatnedione shown in Figure 3.2

(B) was the same as the mass spectrum of Figure 3.2 (A). The unknown compound of

buttery odor in riboflavin solution with the retention time of 6.5 minutes was 2,3-

butanedione. The authentic 2,3-butanedione was added to the buffer solution at the level

47 of 0.1 µg/mL. The odor characteristics of the buffer solution with the authentic 2,3-

butanedione was exactly the same as that of riboflavin solution under light. The buttery

odor compound with the retention time of 6.5 minutes was positively identified as 2,3-

butanedione by the combination of gas chromatographic retention time, mass spectrum,

and odor characteristic of authentic 2,3-butadione.

Mechanism for 2,3-butanedione formation from riboflavin under light

The proposed mechanism for the formation of 2,3-butanedione from the

riboflavin under light is shown in Figure 3.3. Riboflavin is a photosensitizer for the

formation of singlet oxygen (Jung and others 1995; Jung and others 1998; King and

others 2002; Rosenthal and others 2003; Huang and others 2004a, Huang and others

2004b). The riboflavin in singlet state absorbs light and becomes the excited singlet state

riboflavin. The excited singlet riboflavin becomes the excited triplet riboflavin by

intersystem crossing mechanism (Choe and others 2005). The excited triplet riboflavin

reacts with triplet oxygen to form singlet state riboflavin and singlet state oxygen by

triplet-triplet annihilation (Choe and others 2005). Singlet oxygen is an electrophilic

molecule and reacts with electron rich compounds such as riboflavin, linolenic acid or

aromatic amino acids (Min and Boff 2002). Singlet oxygen reacts with riboflavin which

has several double bonds and then forms riboflavin endoperoxide through 6, 9-addition as

shown in Figure 3.3 (Jung and others 1995; Min and Boff 2002). The reaction rate

between riboflavin and singlet oxygen was very fast at the rate of 1.01 × 1010 M-1sec-1

(Huang and others 2004a). Singlet oxygen is directly added to the electron rich double

48 bond of riboflavin and also produces dioxetane through 7, 8-cycloaddition (Bradley and

Min 1992; Min and Boff 2002). The 7, 8-dioxetane of riboflavin endoperoxide at 6 and 9

produces 2,3-butanedione by scissions as shown in Figure 3.3. This mechanism

suggested that riboflavin under light was destroyed by singlet oxygen oxidation and

produced a volatile compound, 2,3-butanedione.

Effect of sodium azide on volatile compound formation of riboflavin solution under light

To study the possibility of singlet oxygen involvement in the formation of the volatile compound from the self-sensitized photooxidation of riboflavin, sodium azide, a well known singlet oxygen quencher (Foote 1979; Haag 1987), was added to the riboflavin solution. The prepared solution was stored in the light box at 3,000 lux for 12 hrs. The gas chromatograms of riboflavin solution containing sodium azide are shown in

Figure 3.4. As the addition of sodium azide increased from 0, 0.5, and 1.0 to 5.0 mM, the peak sizes of the volatile compound with the retention time of 6.5 minutes decreased.

The gas chromatographic peak area with retention time of 6.5 minutes in riboflavin solutions containing 0, 0.25, 0.5, 1.0, or 5.0 mM sodium azide under light for 12 hrs were

12.37, 6.30, 4.35, 3.19, or 1.69 (mV × sec), respectively. The addition of 5.0 mM sodium azide reduced 86% of the formation of the volatile compound with the retention time of

6.5 minutes. This result suggested that singlet oxygen was involved in the formation of the buttery odor compound with the gas chromatographic retention time of 6.5 minutes in the riboflavin solution under light for 12 hrs. This also indicated that singlet oxygen

49 reacted with riboflavin to produce 2,3-butanedione and riboflavin was destroyed by

singlet oxygen.

Effect of pH on 2,3-butanedione formation from riboflavin under light

The quantities of 2,3-butanedione formed from riboflavin in the 0.1 M phosphate buffer with different pHs were analyzed to study the effects of pH on the 2,3-butanedione formation. The quantifications of the 2,3-butanedione in the samples were based on the standard curves obtained from the added known amount of authentic 2,3-butanedione in the each tested buffer solution. Figure 3.5 shows the quantity of 2,3-butanedione formed from riboflavin in 0.1 M phosphate buffer solution with pH 4.5, 6.5 and 8.5 during 12 hr light illumination. The pH of the buffer solution greatly affected the formation of 2,3- butanedione from riboflavin. The highest contents of 2,3-butanedione was formed at pH

6.5, and followed by at pH 4.5 and 8.5, in a decreasing order. The contents of 2,3- butanedione formed in the buffer solutions with pH 4.5, 6.5, and 8.5 were 0.92, 0.77, and

0.17 µg/mL, respectively. It has been previously reported that riboflavin was most stable at pH 6.5, followed by pH 4.5 and 8.5 (Huang and others 2006). It is interesting to note that the 2,3-butanedione formation was closely related with the stability of riboflavin at the different pH. The higher the stability of the riboflavin under light, the higher the 2,3- butanedione formation from the phosphate buffer. To check the sodium phosphate effect on the 2,3-butanedione formation, the 2,3-butanedione formation from riboflavin in purified water was analyzed and compared with that in 0.1 M phosphate buffer at pH 6.5 after 12 hr light illumination. In the purified water, 2,3-butanedione was also the only

50 major volatile compound formed from riboflavin after light exposure as in the phosphate buffer. But the 2,3-butanedione content (0.82 µg/mL) formed in the purified water was somewhat higher than that (0.57 µg/mL) in the phosphate buffer of pH 6.5. The result showed that sodium phosphate did not affect the mechanism for the formation of 2,3- butanedione, but slightly accelerated its formation. We also tested the effects of riboflavin contents in the sodium phosphate buffer (0.1 M, pH 6.5) on the 2,3- butanedione formation. Lower riboflavin concentration in buffered solution induced the significantly lower content of 2,3-butanedione in the solution after 12 hr light exposure.

The concentrations of 2,3-butandione formed from 0.25 mM riboflavin and 0.5 mM riboflavin in sodium phosphate buffer solution (pH 6.5) were 0.53 and 0.82 µg/mL, respectively.

3.5 Conclusion

Riboflavin in a buffer solution under light produced a buttery odor. The buttery odor compound was positively identified as 2,3-butanedione by a combination of gas chromatographic retention time, mass spectrum and odor evaluation with authentic 2,3- butanedione. Sodium azide reduced the formation of 2,3-butanedione from riboflavin solution under light. The 2,3-butadione was formed by the reaction between singlet oxygen and riboflavin. The detailed mechanism for the formation of 2,3-butanedione from riboflavin and singlet oxygen under light was presented. The buttery odor

51 compound from riboflavin under light could affect the flavor quality of foods containing riboflavin under light. This paper reports the formation of buttery odor 2,3-butanedione formed from riboflavin under light for the first time. The previously identified compounds formed from riboflavin under light are nonvolatile lumichrome and lumiflavin. The formation of nonvolatile lumichrome and lumiflavin and volatile 2, 3- butanedione from riboflavin solution under light may explain the rapid destruction of riboflavin in foods under light.

52 3.6 References

Ahmad I, Fasihullah Q, Vaid FHM. 2004. A study of simultaneous photolysis and photoaddition reactions of riboflavin in aqueous solution. J Photochem Photobiol B: Biology 75:13-20.

Allen C, Parks OW. 1979. Phtoodegradation of riboflavin in milks exposed to fluorescent light. J Dairy Sci 62:1377-79.

Bradley DG, Min DB. 1992. Singlet oxygen oxidation of foods. Crit Rev Food Sci Nutr 31:211-36.

Bradley DG, Lee HO, Min DB. Singlet oxygen detection in skim milk by electron spin resonance spectroscopy. J Food Sci 68:491-4.

Choe E, Huang R, Min DB. 2005. Chemical reactions and stability of riboflavin in foods. J Food Sci 70:R28-36.

Criado S, Castillo C, Yppolito R, Bertolotti S, Garcia NA. 2003. The role of 4- and 5- aminosalicylic acids in a riboflavin-photosensitized process. J Photochem Photobiol A: Chemistry 155:115-8.

Edwards AM, Bueno C, Saldano A, Silva E, Kassab K, Polo L, Jori G. 1999. Photochemical and pharmacokinetic properties of selected flavins. J Photochem Photobiol B: Biology 48:36-41.

Foote CS. 1979. Quenching of singlet oxygen. In: Wasserman H, Murray RW, editors. Singlet oxygen. New York: Academy Press. p. 139-71.

53 Gliszczynska A, Koziolowa A. 1999. Flavins and Flavoproteins. In: Proceedings of the 13th International Symposium, Konstanz, Germany, Aug. 29-Sept. 4, p.875-8.

Grzelak A, Rychlik B, Bartosz G. 2001. Light-dependent generation of reactive oxygen species in cell culture media. Free Radical Biol Med 30:1418-25.

1 Haag WR, Mill T. 1987. Rate constants for interaction of singlet oxygen ( Dg) with azide ion in water. Photochem Photobiol 45:317-21.

Huang R, Choe E, Min DB. 2004a. Kinetics for singlet oxygen formation by riboflavin photosensitization and the reaction between riboflavin and singlet oxygen. J Food Sci 69:C726-32.

Huang R, Choe E, Min DB. 2004b. Effects of riboflavin photosensitized oxidation on the volatile compounds of soymilk. J Food Sci 69:C733-8.

Huang R, Kim HJ, Min DB. 2006. Photosensitizing effect of riboflavin, lumiflavin, and lumichrome on the generation of volatiles in soy milk. J Agric Food Chem 54:2359-64.

Jung MY, Kim SK, Kim SY. 1995. Riboflavin-sensitized photooxidation of ascorbic acid: kinetics and amino acid effect. Food Chem 53:397-403.

Jung MY, Yoon SH, Lee HO, Min DB. 1998. Singlet oxygen and ascorbic acid effects on dimethyl disulfide and off-flavor in skim milk exposed to light. J Food Sci 63:408-12.

Kim H, Kirschenbaum LJ, Rosenthal I, Riesz P. 1993. Photosensitized formation of ascorbate radicals by riboflavin: an ESR study. Photochem Photobiol 57:777-84.

54 King JM, Min DB. 2002. Riboflavin-photosensitized singlet oxygen oxidation product of

vitamin D2. J Am Oil Chem Soc 79:983-7.

Lee KH, Jung MY, Kim SY. Effects of ascorbic acid on the light-induced riboflavin degradation and color changes in milks. J Agric Food Chem 46:407-10.

Lucius R, Mentlein R, Sievers J. 1998. Riboflavin-mediated axonal degeneration of postnatal retinal ganglion cells in vitro is related to the formation of free radicals. Free Radical Biol Med 24:798-808.

Min DB, Boff JM. 2002. Chemistry and reaction of singlet oxygen in foods. Comp Rev Food Sci Food Saf 1:58-72.

Rosenthal A, Deliza R, Cabral LMC, Cabral LC, Farias CAA, Domingues AM. 2003. Effect of enzymatic treatment and filtration on sensory characteristics and physical stability of soymilk. Food Control 14:187-92.

Szczesniak T, Karabin L, Szczepankowska M, Wituch K. 1971. Biosynthesis of riboflavin by Ashbya gossypii. I. Influence of fats of animal origin on riboflavin production. Acta Microbiol Pol Ser B: Microbiol Appl 3:29-34.

USDA National Nutrient Database for Standard Reference, Release 16 Nutrient Lists. 2004.

0 hr

3 hr

6.5 min

6 hr

12 hr

0 10 20 Retention time (min)

Figure 3.1. Effects of 0, 3, 6, and 12 hrs on the volatile compounds from riboflavin solution under light

(A)

(B)

Figure 3.2. Mass spectra of gas chromatographic peak with the retention time of 6.5 minutes (A) and authentic 2,3-butanedione (B)

R H H R H C N N O C 3 C 1O N N O 2 H3C O NH O NH H3C C N H C N H 3 C O H O Riboflavin Endoperoxide

H R H R H3C C N O H C C 1 N 3 N N O O2 O O O O O O NH O O NH N N H3C C H3C C H O H O Dioxetane

O R CH3 HC N N O O C Scission + O C NH HC N CH3 O O 2, 3-Butanedione R = Ribose

Figure 3.3. Proposed mechanism for the formation of 2,3-butanedione from riboflavin and singlet oxygen

6.5 min 0 mM

0.5 mM

1.0 mM

5.0 mM

0 10 20

Retention time (min)

Figure 3.4. Effects of 0, 0.5, 1.0 and 5.0 mM sodium azide on the volatile compounds from riboflavin solution under light for 12 hrs

1.0

pH 4.5 phosphate buffer pH 6.5 phosphate buffer 0.8 g/mL) pH 8.5 phosphate buffer µ

0.6

0.4

0.2 2,3-butanedione content (

0.0 036912

Illumination Time (hr)

Figure 3.5. Formation of 2,3-butanedione from riboflavin in 0.1 M sodium phosphate buffer solution with pH 4.5, 6.5 and 8.5 under light for 12 hrs

CHAPTER 4

EFFECTS OF α-, β-, γ-, AND δ-TOCOTRIENOL ON THE OXIDATIVE STABILITY

OF LARD

4.1 Abstract

The effects of 0, 100, 200, 300, 500, and 1000 ppm of α-, β-, γ-, and δ-tocotrienol on the oxidative stability of lard during the storage of 7 days in the dark at 55°C were studied. The oxidation of lard was determined by measuring headspace oxygen in the sample bottle using gas chromatography and peroxide value daily. As the storage time increased, the headspace oxygen contents of lard samples decreased and peroxide value increased. The lard containing α-, β-, γ-, and δ-tocotrienol had higher headspace oxygen contents and lower peroxide values than the lard without tocotrienol. Tukey’s test showed that the 100 ppm α-or β-tocotrienol significantly increased the headspace oxygen contents and decreased peroxide values of lard during storage (p<0.05). As the concentration of α- or β-tocotrienol increased from 100 to 200, 300, 500, and 1000 ppm,

61 the antioxidative activities decreased. Tukey’s test showed that the γ- or δ-tocotrienol significantly increased the headspace oxygen contents and decreased peroxide values of lard during storage (p<0.05) but there were no significant differences between the concentrations of 100, 200, 300, 500, and 1000 ppm (p>0.05). The optimum concentration of α-, β-, γ-, and δ-tocotrienol to increase the oxidative stability of lard could be 100 ppm. The antioxidative activities were in the order of δ- > γ- > β- > α- tocotrienol. The selection of type and optimum concentration of tocotrienol not only minimizes the oxidation of lipids but also will be economically important.

4.2 Introduction

Lipid oxidation causes a significant nutritional loss, produces off-flavor and potential toxic compounds and lowers the sensory perception and safety of foods. The oxidation not only makes the food less acceptable or unacceptable to consumers but also causes great economic losses to the food industry.

Tocopherols are powerful natural antioxidants used in various lipid foods.

Tocopherols prevent lipid oxidation by donating hydrogen from the hydroxyl group of its chromanol ring (Jung and Min 1990; Kamal-Eldin and Appelqvist 1996; Belitz and

Grosch 1999; Evans and others 2002; Reische and others 2002). Tocotrienols are a family of vitamin E with tocopherols and have a similar structure to tocopherols.

Tocotrienols consist of a hydroxyl group and one or more methyl groups at the 5, 7, or 8

62 position of the chromanol ring with a 16-carbon phytyl group (Azzi and Stocker 2000).

Only phytyl group of tocotrienol is different with tocopherol, which has three isolated

double bonds at the 3’, 7’, and 11’ position (Kamal-Eldin and Appelqvist 1996).

Tocotrienol can donate hydrogen from the hydroxyl group of a chroman ring to lipid

radicals as tocopherol does.

Palm oil contains up to 800 mg/kg tocotrienols, mainly consisting of γ-tocotrienol and α-tocotrienol (Sundram and Gapor 1992). Tocotrienols are also found in cereal

grains (Sundram and Gapor 1992).

There is some information available on the biological effects of tocotrienols.

Tocotrienols acted as antioxidants in rats and in human lipoproteins (Suarna and others

1993) and lowered plasma concentration of atherogenic LDL cholesterol in pigs (Qureshi

and others 1991). α-Tocotrienol exhibited greater peroxy radical scavenging than α-

tocopherol in liposome membranes (Suzuki and others 1993). Serbinova and others

(1991) reported the higher antioxidant activity of α-tocotrienol than α-tocopherol in

liposome was due to its more uniform distribution in the membrane bilayer. Although

tocotrienols have a similar structure to tocopherols, the antioxidant effect of tocotrienols

has seldom been studied in foods. The objective of this study was to determine the

quantitative and qualitative effects of α-, β-, γ-, and δ-tocotrienol on the oxidative

stability of lard.

63 4.3 Materials and Methods

Materials

Pork fat was obtained from the Department of Animal Science at The Ohio State

University (Columbus, OH). α-, β-, γ-, and δ-Tocotrienol were obtained from Davos

Life Science (Singapore). Teflon-coated rubber septa, aluminum caps, and serum bottles were purchased from Supelco, Inc. (Bellefonte, PA). Acetic acid and chloroform were purchased from Fisher Scientific Co. (Fair Lawn, NJ) and Mallinckrodt Baker

(Phillipsburg, NJ), respectively.

Sample preparation

Lard was collected from pork fat by placing in an oven at 70°C for 2 hrs. The peroxide value of collected lards was 0.4 meq/kg lard. Lards containing 0, 100, 200, 300,

500, and 1000 ppm α-, β-, γ-, or δ-tocotrienol were prepared. The 1.5 g of sample was

transferred to a 10-mL serum bottle and sealed airtight with a Teflon coated septum and

an aluminum cap. Sample bottles were placed in a Blue M oven (Blueland, IL) at 55°C

for 7 days. The oxidative stability of lard was determined by measuring headspace

oxygen content in the sample bottle and peroxide value each day during 7 days of storage.

Fatty acid analysis

Fatty acid composition of lard was determined by the gas chromatography using

AOCS method Ch 2a-94 (AOCS 2004).

64 Determination of headspace oxygen

Headspace oxygen in the sample bottle was measured by injecting a 100 µL

headspace gas into a Hewlett-Packard 5890 gas chromatograph (GC) equipped with a

thermal conductivity detector (Huang and others 2004). A stainless steel column (1.8 m

× 0.32 cm) packed with 80/100 Molecular Sieve 13X (Alltech Asso Inc., IL) was used.

The high purity (99.995%) helium gas was used as a carrier gas at 20 mL/min. The

temperature of the injection port, oven, and detector were 120, 40, and 150°C,

respectively. The gas chromatographic peak was measured by electronic count using a

HP 3396A integrator. The electronic count was converted and reported as % oxygen

content.

Determination of peroxide value

The peroxide value of sample was determined by AOCS method Cd 8-53 (AOCS

2004). The 1.5g of lard was transferred into a dry 125-mL Erlenmeyer flask and 15 mL

of acetic acid/chloroform (3:2) solution were added. The flask was swirled and then 0.25

mL saturated KI solution was added. After allowing the solution to set for 1 min, 15 mL

of distilled water was added. This was titrated against 0.1 N Na2S2O3 after the addition of 1 mL of starch indicator until the blue color disappeared. The peroxide value was expressed as milliequivalent of peroxides per kilogram lard (meq/kg lard).

Statistical analysis

65 All experiments were done in duplicate. One-way analysis of variance and

Fisher’s least significant differences (LSD) test were used to analyze the effects of α-, β-,

γ-, and δ-tocotrienols on the headspace oxygen contents and peroxide values using SAS version 9.1 (SAS® Institute, Cary, NC).

4.4 Results and Discussion

Fatty acid composition of lard

The fatty acid composition of lard was determined by gas chromatography and showed in Table 4.1. The fatty acids of lard were 26% palmitic acid, 14% stearic acid,

44% oleic acid, and 10% linoleic acid. Lard has less unsaturated fatty acids than vegetable oils. The higher the unsaturated fatty acids in oils, the faster oxidation occurs.

Soybean, safflower or sunflower oil that contains high amounts of unsaturated fatty acids showed much shorter induction period than coconut or palm kernel oil which has low levels of unsaturated fatty acids (Tan and others 2002). The relative autoxidation rates of stearic acid, oleic acid, linoleic acid, and linolenic acid are 1:100:1200:2500 (Min and

Boff 2002). Therefore, lard should have better oxidative stability than vegetable oils.

Lard contains only less than 18 ppm of total tocopherols compared to soybean oil having 1,200 ppm tocopherols (Choe and Min 2006) and does not have tocotrienols.

Therefore, lard was chosen to study the antioxidative activities of α-, β-, γ-, and δ- tocotrienol.

66 Reproducibility and correlation of headspace oxygen and peroxide value analyses

The coefficients of variation for measuring headspace oxygen and peroxide value for 5 replicates were 1.0 and 1.8%, respectively (data not shown). The low coefficients of variations for the analyses by headspace oxygen and peroxide value were considered as good reproducible and acceptable methods to study the oxidation of lard.

The relationship between the headspace oxygen contents and peroxide values of lard with tocotrienols were highly correlated. The correlation coefficients (R2) for lard with α-, β-, γ-, and δ-tocotrienol were 0.96, 0.93, 0.90, and 0.92, respectively. The good relationship suggested that lipid oxidation of lard could be evaluated by the combination of headspace oxygen and peroxide value analyses.

Effects of tocotrienols on the oxidative stability of lard

The oxidative stability of lard was determined by the combination of headspace oxygen content and peroxide value. The effects of α-, β-, γ-, and δ-tocotrienol on the headspace oxygen in lard sample bottles during the storage of 7 days at 55°C are shown in Figure 4.1, 4.2, 4.3, and 4.4, respectively. As the storage time increased from to 0 day to 7 day, the headspace oxygen of lard decreased. The headspace oxygen of the lard with

0 ppm tocotrienol stored for 0 day and 7 day were 21.06% and 10.69%, respectively. The disappeared headspace oxygen in a gas-tight sample bottle could be due to the reaction between oxygen and lard. The continuous disappearance of headspace oxygen in the sample bottle indicates the progress of lipid oxidation during storage (Kim and others

2007).

67 The mean of headspace oxygen contents for lards containing 0, 100, 200, 300,

500, or 1000 ppm α-tocotrienol stored for 0, 1, 2, 3, 4, 5, 6, and 7 day at 55°C was determined. The means of headspace oxygen for lards with 0, 100, 200, 300, 500 and

1000 ppm α-tocotrienol were 16.86, 19.36, 18.76, 18.53, 18.09, and 17.47%, respectively

(Figure 4.1.). LSD test showed that the headspace oxygen of lard with 100 ppm α-

tocotrienol was significantly different compared to the control (p<0.05). As the

concentration increased, the antioxidative activities of α-tocotrienol decreased (Figure

4.1) but there were no significant differences among the concentration of 200, 300, 500,

and 1000 ppm (p>0.05). The 100 ppm α-tocotrienol was the most effective to decrease

the loss of headspace oxygen.

The means of headspace oxygen contents for lard with 0, 100, 200, 300, 500, and

1000 ppm β-tocotrienol stored for up to 7 days at 55°C were 16.86, 19.80, 19.24, 18.78,

18.45, and 17.42%, respectively (Figure 4.2.). As the concentration of β-tocotrienol

increased, the headspace oxygen contents decreased. The headspace oxygen of lard with

100 and 200 ppm β-tocotrienol was significantly different compared to the control

(p<0.05). The optimum concentration of β-tocotrienol to decrease the loss of headspace

oxygen was 100-200 ppm.

The headspace oxygen contents of lard with 0, 100, 200, 300, 500, and 1000 ppm

γ-tocotrienol stored for 7 days in dark at 55°C are shown in Figure 4.3. The means of

headspace oxygen of lard with 0, 100, 200, 300, 500 and 1000 ppm γ-tocotrienol stored

for 0, 1, 2, 3, 4, 5, 6, and 7 day were 16.86, 19.60, 19.62, 19.60, 19.75, and 19.49%,

68 respectively. LSD test showed that γ-tocotrienol at 100, 200, 300, 500, and 1000 ppm

significantly decreased the loss of headspace oxygen in lard sample bottle (p<0.05), but

there were no significant differences among the concentrations of 100, 200, 300, 500, and

1000 ppm (p>0.05). The optimum concentration of γ-tocotrienol to decrease the

headspace oxygen depletion could be 100 ppm since the antioxidative effects of five

different concentrations were not significantly different.

Figure 4.4 shows the headspace oxygen of lard containing 0, 100, 200, 300, 500, and 1000 ppm δ-tocotrienol during the storage of 7 days in dark at 55°C. The means of headspace oxygen contents for lard with 0, 100, 200, 300, 500 and 1000 ppm δ- tocotrienol stored for up to 7 days at 55°C were 16.86, 19.63, 19.79, 19.80, 19.82, and

19.83%, respectively. LSD test showed that δ-tocotrienol at 100, 200, 300, 500, and

1000 ppm had significant antioxidative effect on the headspace oxygen of lard (p<0.05) but there were no significant differences among the concentrations of 100, 200, 300, 500, and 1000 ppm (p>0.05). The optimum concentration of δ-tocotrienol to decrease the headspace oxygen depletion could be 100 ppm.

The means of headspace oxygen contents of lard samples having 0, 100, 200, 300,

500 and 1000 ppm α-, β-, γ-, or δ-tocotrienol stored for 0, 1, 2, 3, 4, 5, 6, and 7 day at

55°C were determined to compare the antioxidative activities of each tocotrienol. The means of headspace oxygen of lard with α-, β-, γ-, and δ-tocotrienol were 18.18, 18.42,

19.15, and 19.29%, respectively. The antioxidative activities were in δ- > γ- > β- > α- tocotrienol. The antioxidative activity of α-tocotrienol was significantly different

69 compared to δ-tocotrienol (p<0.05). The β- and γ-tocotrienol activities were not significantly different with δ-tocotrienol (p>0.05).

The effects of α-, β-, γ-, and δ-tocotrienol on the peroxide value of lard during the storage of 7 days at 55°C are shown in Figure 4.5, 4.6, 4.7, and 4.8, respectively. As the storage time increased from 0 to 7 day at 55°C, the peroxide value of lard increased. The peroxide values of lard with 0 ppm tocotrienol stored for 0 day and 7 day were 0.4 and

33.4 meq/kg lard, respectively. Peroxide value measures hydroperoxide which is a transient product of lipid oxidation. As the lipid oxidation progresses during the storage, hydroperoxide contents in lipid increase and peroxide value increases.

The mean of peroxide values for lards with 0, 100, 200, 300, 500 or 1000 ppm α- tocotrienol stored for 0, 1, 2, 3, 4, 5, 6, and 7 day at 55°C was determined. The means of peroxide values for lards with 0, 100, 200, 300, 500 and 1000 ppm α-tocotrienol were

13.17, 4.85, 7.17, 8.90, 11.80, and 16.10 meq/kg lard, respectively (Figure 4.5). As the concentration of α-tocotrienol increased, the peroxide value of lard increased. The peroxide value of lard with 100 ppm α-tocotrienol was significantly different compared

to the control (p<0.05). There were no significant differences between 100 ppm and 200,

300, or 500 ppm α-tocotrienol (p>0.05). The 1000 ppm α-tocotrienol acted as

prooxidant on the peroxide value during the storage of 6 days but the antioxidative

activity increased at 7 day. The optimum concentration of α-tocotrienol to decrease the

formation of hydroperoxides in lard was 100 ppm.

The mean of peroxide values for lards with 0, 100, 200, 300, 500 or 1000 ppm β-

tocotrienol stored for up to 7 day at 55°C was calculated. The means of peroxide values

70 for lard with 0, 100, 200, 300, 500 and 1000 ppm β-tocotrienol were 13.17, 3.82, 4.08,

4.96, 5.45, and 7.80 meq/kg lard, respectively (Figure 4.6.). As the concentration of β- tocotrienol increased, the peroxide values of lard increased during storage. LSD test showed that 100 ppm β-tocotrienol had significant lower peroxide value than the control

(p<0.05). Concentration of 200, 300, and 500 ppm were not significantly different compared to 100 ppm β-tocotrienol (p>0.05). The optimum concentration of β-

tocotrienol to decrease the peroxide value of lard could be 100 ppm.

The peroxide values of lard containing 0, 100, 200, 300, 500, and 1000 ppm γ-

tocotrienol stored for 7 days in dark at 55°C are shown in Figure 4.7. The means of

peroxide values for lard with 0, 100, 200, 300, 500 and 1000 ppm γ-tocotrienol stored for

up to 7 day were 13.17, 3.33, 3.35, 3.33, 3.54, and 3.79 meq/kg lard, respectively. γ-

Tocotrienol effectively decreased the peroxide values as the storage time increased.

There were no significant differences among the concentration of 100, 200, 300, 500, and

1000 ppm γ-tocotrienol (p>0.05).

The peroxide values of lard containing 0, 100, 200, 300, 500, and 1000 ppm δ-

tocotrienol stored for 7 days in dark at 55°C are shown in Figure 4.8. The means of

peroxide values for lard with 0, 100, 200, 300, 500 and 1000 ppm δ-tocotrienol stored for

up to 7 day were 13.17, 3.42, 3.12, 2.93, 3.00, and 2.89 meq/kg lard, respectively. The

peroxide values of lard with 100, 200, 300, 500, and 1000 ppm δ-tocotrienol were

significantly lower than the control (p<0.05). There were no significant differences

among the concentration of 100, 200, 300, 500, and 1000 ppm (p>0.05).

71 The mean of peroxide values of lard with 0, 100, 200, 300, 500 and 1000 ppm α-,

β-, γ-, or δ-tocotrienol stored for 0, 1, 2, 3, 4, 5, 6, and 7 day at 55°C was calculated to compare the antioxidative activities of each tocotrienol. The means of peroxide values of lard with α-, β-, γ-, and δ-tocotrienol were 10.33, 6.55, 5.08, and 4.76 meq/kg lard, respectively. The antioxidative activities were of δ- > γ- > β- > α-tocotrienol. The antioxidative effect of α-tocotrienol to reduce the formation of hydroperoxides was significantly lower than that of β-, γ-, or δ-tocotrienol (p<0.05). There were no significant differences between β-, or γ-, and δ-tocotrienol (p>0.05). δ-Tocotrienol most effectively minimized the formation of hydroperoxides during the storage.

The results in this study showed that the type and concentration of tocotrienol were important on the oxidative stability of lard during storage in the dark at 55°C.

Tocotrienols have the same structure as tocopherols except a phytyl group on the chroman ring. The antioxidant activity of the tocopherols and tocotrienols is mainly due to their ability to donate their hydrogen on the hydroxyl group of chromanol ring to lipid free-radicals (Reische and others 2002). Wright and others (2001) determined the bond dissociation energy of tocopherols, which measures the bond strength of O-H bond in tocopherols. The bond dissociation energies of α-, β-, γ-, and δ-tocopherol are 75.8, 77.7,

78.2, and 79.8 kcal/mole, respectively (Wright and others 2001). The O-H bond of hydroxyl group in α-tocopherol can be more easily cleaved to donate a hydrogen atom to lipid radical than β-, γ-, or δ-tocopherol. The antioxidant activity of tocopherols are in the order of α- > β- > γ- > δ-tocopherol based on their chemical structure. However, the

72 reversed order, δ- > β- or γ- > α-tocopherol were obtained when relative antioxidant activities were compared in fats and oils (Wagner and Elmadfa 2000; Wagner and others

2001; Isanardy and others 2003). The results of tocotrienols in this study also showed the order of δ- > γ- > β- > α-tocotrienol. The relative antioxidative activities of tocopherols and tocotrienols are not only dependent on their absolute chemical reactivity toward lipid free radicals, but also on many other possible side reactions by tocopherol concentrations, temperature, light, type of substrate and solvent, and other chemical species acting as prooxidants or synergists in the system (Kamal-Eldin and Appelqvist 1996).

Many studies showed that tocopherol concentration is important whether it can be an antioxidant or prooxidant in foods (Jung and Min 1990; Huang and others 1994;

Huang and others 1995; Evans and others 2002). The antioxidative activity of tocopherol is greatest at lower concentration and decreases at higher concentration. The optimum concentrations of α-, γ-, and δ-tocopherol to minimize the oxidation of soybean oils were

100 ppm, 250-500 ppm, and 500-1000 ppm, respectively (Jung and Min 1990; Huang and others 1994; Evans and others 2002). The optimum concentrations of α-, γ-, and δ- tocopherol were related to the oxidative stability of each tocopherol (Gottstein and

Grosch 1990). The oxidative stability of tocopherol in soybean oil was in δ- > γ- > α- tocopherol (Player and others 2006). δ-Tocopherol is more stable than α-tocopherol during oxidation. Therefore, α-tocotrienol, having the same structure as α-tocopherol, could be more easily oxidized. The antioxidative activity of δ-tocotrienol, which is the most stable, was the greatest and the effect of α-tocotrienol was the lowest among the four types of tocotrienols.

73 Tocotrienols are thought to be more potent in antioxidant properties than α- tocopherol (Serbinova and others 1991; Serbinova and Packer 1994). The unsaturated side chain (a phytyl group) in tocotrienols made them penetrate more efficiently into tissues which have saturated fatty layers (Suzuki and others 1993). Tocotrienols have shown excellent effects in biological system to lower plasma cholesterol levels

(Hasselwander and others 2002) and to reduce lipid and non-lipid related risk factors for cardiovascular disease (Newaz and Nawal 1999). Tocotrienols have been reported to be better antioxidants than tocopherols in some foods. Wagner and others (2001) reported that δ- and γ-tocotrienol increased the shelf-life of coconut fat at ambient temperature and frying temperature. They found that tocotrienols extended the induction period of coconut fat longer than their corresponding tocopherol. Feng (1996) reported that γ- tocotrienol had a higher antioxidative ability than α-tocotrienol in vegetable oils at heating temperature. The higher antioxidative activities of tocotrienol than tocopherol might be due to its more uniform distribution in the lipids.

4.5 Conclusion

α-, β-, γ-, and δ-Tocotrienol at 100, 200, 300, 500, and 1000 ppm minimized the oxidation of lard during the storage of 7 days in the dark at 55°C. As the concentration of

α- or β-tocotrienol increased from 100 to 200, 300, 500, and 1000 ppm, the antioxidative activities of α- or β-tocotrienol decreased. There were no significant differences on the

74 antioxidative activities of γ- or δ-tocotrienol at 100, 200, 300, 500, and 1000 ppm at

p>0.05. The 100 ppm of α-, β-, γ- or δ-tocotrienol could be the optimum concentration

to improve the oxidative stability of lard. The antioxidative activities were δ->γ->β->α-

tocotrienol. The antioxidative activity of δ-tocotrienol was the most effective among four types of tocotrienol.

75 4.6 References

[AOCS] American Oil Chemists’ Society. 2004. Official methods and recommended practices of the AOCS. 5th ed. Champaign, Ill.:AOCS.

Azzi A, Stocker A. 2000. Vitamin E: non-antioxidant roles. Prog in Lipid Res 39: 231-55. Belitz H-D, Grosch W. 1999. Food Chemistry, 2nd ed. Berlin: Springer. pp 207-8, 381-2, 606.

Belitz H-D, Grosch W. 1999. Food Chemistry, 2nd edition. Berlin: Springer. pp 207-8, 381-2, 606.

Choe E, Min DB. 2006. Mechanisms and factors for edible oil oxidation. Compr Rev Food Sci Food Saf 5: 169-86.

Evans JC, Kodali DR, Addis PB. 2002. Optimal tocopherol concentrations to inhibit soybean oil oxidation. J Am Oil Chem Soc 79: 47-51.

Feng HP. 1996. Preparative techniques for isolation of vitamin E homologs and evaluation of their antioxidant activities. Dissertation-Abstract-International, -B 56: 207.

Gottstein T, Grosch W. 1990. Model study of different antioxidant properties of α- and γ- tocopherol in fats. Fat Sci and Tech 92: 139-44.

Hasselwander O, Krämer K, Hoppe PP, Oberfrank U, Baldenius K, Schröder H, Wolfgan K, Rainer B, Nowakowsky B. 2002. Effects of feeding various tocotrienol sources on plasma lipids and aortic atherosclerotic lesions in cholesterol-fed rabbits. Food Res Int 35:245-51.

76 Huang S-W, Frankel EN, German JB. 1994. Antioxidant activity of α- and γ-tocopherols in bulk oils and in oil-in-water emulsions. J Agric and Food Chem 42:2108-14.

Huang S-W, Frankel EN, German JB. 1995. Effects of individual tocopherols and tocopherol mixtures on the oxidative stability of corn oil triglycerides. J Agric and Food Chem 43:2345-50.

Huang R, Choe E, Min DB. 2004. Effects of riboflavin photosensitized oxidation on the volatile compounds of soymilk. J Food Sci 69: C733-8.

Isnardy B, Wagner K-H, Elmadfa I. 2003. Effects of α-, γ-, and δ-tocopherols on the autoxidation of purified rapeseed oil triacylglycerols in a system containing low oxygen. J Agric Food Chem 51: 7775-80.

Jung MY, Min DB. 1990. Effects of α-, γ-, and δ-tocopherols on the oxidative stability of soybean oil. J Food Sci 55: 1464-5.

Kamal-Eldin A, Appelqvist L-Å. 1996. The chemistry and antioxidant properties of tocopherols and tocotrienols. Lipids 31:671-701.

Kim HJ, Hahm TS, Min DB. 2007. Hydroperoxide as a prooxidant in the oxidative stability of soybean oil. J Am Oil Chem Soc 84:349-355.

Min DB, Boff JM. 2002. Chemistry and reaction of singlet oxygen in foods. Compr Rev Food Sci Food Saf 1: 58-72.

Newaz MA, Nawal NN. 1999. Effect of γ-tocotrienol on blood pressure, lipid peroxidation and total antioxidant status in spontaneously hypertensive rats. Clin Exp Hypertension 21:1297-313.

77 Player ME, Kim HJ, Lee HO, Min DB. 2006. Stability of α-, γ-, or δ-tocopherol during soybean oil oxidation. J Food Sci 71:C456-C60.

Qureshi AA, Qureshi N, Hasler-Rapacz JO, Weber FE, Chaudhary V, Crenshaw TC, Gapor A, Ong AS, Chong YH, Peterson D. 1991. Dietary tocotrienols reduce concentrations of plasma cholesterol, apolipoprotein B, thrombox and B2, and platelet factor 4 in pigs with inherited hyperlipodemias. Am J Clin Nutr 53:1042S-6S.

Reische DW, Lillard DA, Eitenmiller RR. 2002. Antioxidants. In: Akoh CC and Min DB. Food Lipids. 2nd edition. New York: Marcel-Dekker. p 489-516.

Serbinova E, Kagan V, Han D, Packer L. 1991. Free radical recycling and intra- membrane mobility in the antioxidative properties of α-tocopherol and α-tocotrienol. Free Rad Biol Med 10: 263-76.

Serbinova EA, Packer L. 1994. Antioxidant properties of α-tocopherol and α-tocotrienol. Meth Enz 234:354-66.

Suarna C, Hood LR, Dean RT, Stocker R. 1993. Comparative antioxidant activity of tocotrienols and other natural lipid soluble antioxidants in a homogeneous system, and in rat and human lipoproteins. Biochem Biophys Acta 1166: 163-70.

Sundram K, Gapor A. 1992. Vitamin E from palm oil: its extraction and nutritional properties. Lipid Technol Nov/Dec: 137-41.

Suzuki YJ, Tsuchiya M, Wassall SR, Choo YM, Govil G, Kagan VE, Packer L. 1993. Structural and dynamic membrane properties of α-tocopherol and α-tocotrienol: implication to the molecular mechanism of their antioxidant potency. Biochem 32: 10692-9.

78 Tan CP, Che-Man YB, Jinap S, Yusoff MSA. 2002. Effects of microwave heating on the quality characteristics and thermal properties of RBD palm olein. Innovative Food Sci Emerge Tech 3:157-63.

Wagner K-H, Elmadfa I. 2000. Effects of tocopherols and their mixtures on the oxidative stability of olive oil and linseed oil under heating. Euro J Lipid Sci Technol 102: 624-9.

Wagner K-H, Yoruba F, Elmadfa I. 2001. Antioxidative potential of tocotrienols and tocopherols in coconut fat at different oxidation temperature. Euro J Lipid Sci Technol 103: 746-51.

Wright JS, Johnson ER, Dilation GA. 2001. Predicting the activity of phenolic antioxidants: theoretical method, analysis of substituent effects, and application to major families of antioxidants. J Am Chem Soc 123:1173-83.

Fatty acid % Myristic acid 2 Palmitic acid 26 Stearic acid 14 Oleic acid 44 Linoleic acid 10

Table 4.1. Fatty acid composition of lard

14 0 ppm (a) 100 ppm (b)

Headspace Oxygen (%) 12 200 ppm (ab) 300 ppm (ab) 10 500 ppm (ab) 1000 ppm (ab)

8 01234567 Storage (Day)

Figure 4.1. Effect of 0, 100, 200, 300, 500, and 1000 ppm α-tocotrienol on the headspace oxygen of lard during dark storage for 7 days at 55°C (The concentrations followed by different letters are significantly different at p<0.05)

0 ppm (a) 100 ppm (b) Headspace Oxygen (%) 12 200 ppm (bc) 300 ppm (abc) 10 500 ppm (abc) 1000 ppm (ac)

8 01234567 Storage (Day)

Figure 4.2. Effect of 0, 100, 200, 300, 500, and 1000 ppm β-tocotrienol on the headspace oxygen of lard during dark storage for 7 days at 55°C (The concentrations followed by different letters are significantly different at p<0.05)

14 0 ppm (a) 100 ppm (b)

Headspace Oxygen (%) 12 200 ppm (b) 300 ppm (b) 10 500 ppm (b) 1000 ppm (b)

8 01234567 Storage (Day)

Figure 4.3. Effect of 0, 100, 200, 300, 500, and 1000 ppm γ-tocotrienol on the headspace oxygen of lard during dark storage for 7 days at 55°C (The concentrations followed by different letters are significantly different at p<0.05)

14 0 ppm (a) 100 ppm (b)

Headspace Oxygen (%) 12 200 ppm (b) 300 ppm (b) 10 500 ppm (b) 1000 ppm (b)

8 01234567 Storage (Day)

Figure 4.4. Effect of 0, 100, 200, 300, 500, and 1000 ppm δ-tocotrienol on the headspace oxygen of lard during dark storage for 7 days at 55°C (The concentrations followed by different letters are significantly different at p<0.05)

0 ppm (ac) 35 100 ppm (b) 200 ppm (bc) 30 300 ppm (abc) 500 ppm (abc) 25 1000 ppm (a)

15 Peroxide Value (meq/kg) Peroxide Value 10

0 01234567 Storage (Day)

Figure 4.5. Effect of 0, 100, 200, 300, 500, and 1000 ppm α-tocotrienol on the peroxide value of lard during dark storage for 7 days at 55°C (The concentrations followed by different letters are significantly different at p<0.05)

0 ppm (a) 35 100 ppm (b) 200 ppm (b) 30 300 ppm (b) 500 ppm (b) 25 1000 ppm (ab)

15 Peroxide Value (meq/kg) Value Peroxide 10

0 01234567 Storage (Day)

Figure 4.6. Effect of 0, 100, 200, 300, 500, and 1000 ppm β-tocotrienol on the peroxide value of lard during dark storage for 7 days at 55°C (The concentrations followed by different letters are significantly different at p<0.05)

35 0 ppm (a) 100 ppm (b) 200 ppm (b) 30 300 ppm (b) 500 ppm (b) 25 1000 ppm (b)

15 Peroxide Value (meq/kg) 10

0 01234567 Storage (Day)

Figure 4.7. Effect of 0, 100, 200, 300, 500, and 1000 ppm γ-tocotrienol on the peroxide value of lard during dark storage for 7 days at 55°C (The concentrations followed by different letters are significantly different at p<0.05)

35 0 ppm (a) 100 ppm (b) 30 200 ppm (b) 300 ppm (b) 500 ppm (b) 25 1000 ppm (b)

15 Peroxide Value (meq/kg) 10

0 01234567 Storage (Day)

Figure 4.8. Effect of 0, 100, 200, 300, 500, and 1000 ppm δ-tocotrienol on the peroxide value of lard during dark storage for 7 days at 55°C (The concentrations followed by different letters are significantly different at p<0.05)

CHAPTER 5

EFFECTS, QUENCHING MECHANISMS, AND KINETICS OF α-, β-, γ-, AND δ-

TOCOTRIENOL ON CHLOROPHYLL PHOTOSENSITIZED OXIDATION OF LARD

5.1 Abstract

The effects, quenching mechanisms and kinetics of α-, β-, γ-, and δ-tocotrienol on

the chlorophyll photosensitized oxidation of lard were studied. Samples of 0.1, 0.25, and

0.4 M lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b and 0, 0.3, 0.6, or

0.9 mM α-, β-, γ-, and δ-tocotrienol were prepared and stored at 3,000 lux for 4 hours.

The oxidation of sample was determined by measuring the headspace oxygen content in

air tight sample bottles using gas chromatography and the peroxide value. α-, β-, γ-, and

δ-Tocotrienols had significant antioxidative effects on the chlorophyll photosensitized

oxidation of lard at p<0.05. Chlorophyll in methylene chloride at 3,000 lux produced

singlet oxygen at 1.089 µmole oxygen/mL headspace/hr. The steady state kinetic study

showed that α-, β-, γ-, and δ-tocotrienol acted as antioxidants in chlorophyll

89 photosensitized oxidation of lard by quenching singlet oxygen. The reaction rate

constant of singlet oxygen with lard was 6.5 × 104 M-1sec-1. The singlet oxygen quenching rates of α-, β-, γ-, and δ-tocotrienol were 2.16 × 107, 1.99 × 107, 2.05 × 107, and 0.80 × 107 M-1sec-1, respectively.

5.2 Introduction

Lipid oxidation in foods is due to the combination of triplet oxygen and singlet

oxygen oxidation (Bradley and Min 1992; Min and Boff 2002; Choe and Min 2006).

Singlet oxygen could be formed by ordinary triplet oxygen in the presence of a sensitizer

such as chlorophyll under light (Bradley and Min 1992). Singlet oxygen can directly

react with the double bonds of compounds without the formation of free radicals. The

reaction of singlet oxygen is very rapid due to the low activation energy for chemical

reaction (Bradley and Min 1992; Min and Boff 2002). The rate of singlet oxygen with

linoleic acid is about 1500 times greater than that of ordinary triplet oxygen (Min and

Boff 2002; Choe and Min 2005). Singlet oxygen has been known to be involved in

reversion flavor in soybean oil (Min and others 2003), sunlight flavor in milk (Bradley

1991; Jung and others 1998), destruction of riboflavin (Huang and others 2004a; Huang

and others 2006), oxidation of vitamin D in milk (King and Min 1998), and oxidation of

pork and turkey meat exposed to light (Whang and Peng 1988).

90 Vitamin E which is in the family of α-, β-, γ-, and δ-tocopherols and α-, β-, γ-, and δ-tocotrienols is the major lipid soluble antioxidant compound (Kamal-Eldin and

Appelqvist 1996; Reische and others 2002; Yoshida and others 2003). Tocopherols and tocotrienols are derivatives of 6-chromanol. α-, β-, γ-, and δ-Tocopherols consist of one

hydroxyl group and one or more methyl groups at the 5, 7, or 8 position of the chromanol

ring with a 16-carbon saturated phytyl chain (Azzi and Stocker 2000). α-, β-, γ-, and δ-

Tocotrienols have the same pattern on the chromanol ring but three isolated double bonds

in their phytyl tails at the 3’, 7’, and 11’ position (Kamal-Eldin and Appelqvist 1996).

Tocopherols have been reported to minimize food oxidation by scavenging free radicals

(Jung and Min 1990; Kamal-Eldin and Appelqvist 1996; Belitz and Grosch 1999; Evans

and others 2002; Reische and others 2002) and quenching singlet oxygen by charge

transfer mechanism (Foote 1979; Jung and others 1991; Choe and Min 2005). Although

tocotrienols have a similar structure to tocopherols, the antioxidant effect of tocotrienols

has seldom been studied in foods. Wagner and others (2001) studied the free radical

scavenging properties of tocotrienols at 60°C and 160°C in coconut fat. Schroeder and

others (2006) reported the antioxidant synergism of tocotrienols and carotenoids in palm

oil during deep-fat frying. Some information is available on their biological effects.

Tocotrienols acted as antioxidants in rats and in human lipoproteins (Suarna and others

1993) and lowered plasma concentration of atherogenic LDL cholesterol in pigs (Qureshi

and others 1991). α-Tocotrienol exhibited greater peroxy radical scavenging than α-

tocopherol in liposomal membranes (Suzuki and others 1993). Serbinova and others

(1991) reported the higher antioxidant activity of α-tocotrienol than α-tocopherol in

91 liposomes was due to its more uniform distribution in the membrane bilayer and its

stronger disordering effect on membrane lipids.

Palm oil contains a high amount of tocotrienols (up to 800 mg/kg), mainly consisting of γ-tocotrienol and α-tocotrienol (Sundram and Gapor 1992). Tocotrienols are also found in cereal grains. Rice bran, barley, wheat germ, and oat have 585, 910,

190, and 210 mg/kg of tocotrienols, respectively (Sundram and Gapor 1992).

The study of photosensitized oxidation in foods has been carried out occasionally due to the difficulty of complete removal of photosensitizers and antioxidants in a food system. The combination of adsorbents including activated silicic acid, Celite, powdered sugar, and activated charcoal was used to remove photosensitizers and antioxidants from oils (Lee and Min 1990; Jung and Min 1991; Kulas and Ackman 2001). However, this method was very slow and did not assure the complete removal of photosensitizers and antioxidants from the oil. Lard is animal lipid and has been used in deep-fat frying and as shortening in the food industry (Belitz and Grosch 1999). Lard contains only 18 ppm of total tocopherols compared to soybean oil having 1,200 ppm (Choe and Min 2006) and it does not contain tocotrienols. Therefore, lard can be a suitable substrate for the study of photosensitized oxidation. The antioxidant mechanisms and kinetics of α-, β-, γ-, and δ- tocotrienols on photosensitized singlet oxygen oxidation of foods have never been studied. The objectives of this research were to determine the effects, quenching mechanisms, and kinetics of α-, β-, γ-, and δ-tocotrienols on the chlorophyll photosensitized oxidation of lard.

92 5.3 Materials and Methods

Materials

Pork fat was obtained from Department of Animal Science at The Ohio State

University (Columbus, OH). α-, β-, γ-, and δ-Tocotrienol were obtained from Davos

Life Science (Singapore). Chlorophyll b and methylene chloride were purchased from

Sigma Chemical Co. (St Louis, MO) and Fisher Scientific Co. (Pittsburgh, PA), respectively. Teflon-coated rubber septa, aluminum caps, and serum bottles were purchased from Supelco, Inc. (Bellefonte, PA).

Fatty acid analysis

Fatty acid composition of lard was determined by the gas chromatography using

AOCS method Ch 2a-94 (AOCS 2004).

Sample preparation for effects of tocotrienols on chlorophyll photosensitized oxidation of lard

Lard was carefully collected from pork fat after placing in an oven at 70°C for 2 hrs. The peroxide value of the collected lard was 0.4 meq/kg lard. α-, β-, γ-, and δ-

Tocotrienols at 0 and 1.2 mM were added to 0.4 M lard in methylene chloride containing

4.4 × 10-6 M chlorophyll b. Three mL of sample was transferred to a 10-mL serum bottle and sealed airtight with a Teflon coated septum and an aluminum cap. Sample bottles were placed in a light storage box (70 × 50 × 60 cm) described by Yang (1994) and King

93 and Min (1998). The inside wall of the light box were covered with mirrors to provide

uniform light intensity to samples. The sample bottles held in a plate were rotated under

a 100 watt tungsten light bulb at 3,000 Lux. Samples in duplicate were stored for 4 hours

and analyzed for headspace oxygen and peroxide value.

Sample preparation for determination of quenching mechanisms and kinetics of

tocotrienols

The quenching mechanisms and kinetics of α-, β-, γ-, and δ-tocotrienol were

studied using the steady state kinetics (Foote and Denny 1968; Jung and Min 1991;

Huang and others 2004b). Samples of 0.1, 0.25, and 0.4 M lard in methylene chloride

containing 4.4 × 10-6 M chlorophyll b and 0, 0.3, 0.6, and 0.9 mM α-, β-, γ-, and δ- tocotrienol were prepared. A 3 mL aliquot of the prepared sample was transferred to a

10-mL serum bottle and sealed airtight with a Teflon septum and an aluminum cap in duplicate. Samples in duplicate were placed in the light box for 4 hours and analyzed for headspace oxygen.

Oxidative stability analysis

The oxidation of lard sample was evaluated by measuring headspace oxygen and peroxide value. Headspace oxygen in the sample bottles was measured by injecting a

100 µL headspace gas into a Hewlett-Packard 5890 gas chromatograph (GC) equipped with a thermal conductivity detector (Huang and others 2004). A stainless steel column

(1.8 m × 0.32 cm) packed with 80/100 Molecular Sieve 13X (Alltech Asso Inc., IL) was

94 used. High purity (99.995%) helium gas was used as a carrier gas at 20 mL/min. The temperature of injection port, oven, and detector were 120, 40, and 150°C, respectively.

The gas chromatographic peak was measured by electronic count using a HP 3396A integrator. The depleted headspace oxygen was expressed as µmole of oxygen per mL of headspace gas (King and Min 1998; Huang and others 2004b). One mL of air contains

20.946% of oxygen which is equal to 9.35 µmole of oxygen (Parker 1982). The peroxide value of samples was determined by AOCS method Cd 8-53 (AOCS 2004).

Statistical analysis

All experiments were done in duplicate. One-way analysis of variance and

Tukey’s test were used to analyze the effects of α-, β-, γ-, and δ-tocotrienols on the headspace oxygen content and peroxide value at α = 0.05 using XLSTAT (Microsoft

2007).

5.4 Results and Discussion

Fatty acid composition of lard

The fatty acid composition of lard determined by the gas chromatography was

26% palmitic acid, 14% stearic acid, 44% oleic acid, and 10% linoleic acid. The average molecular weight of fatty acid in lard was 263 g/mole. The average molecular weight of lard triglycerides was 826 g/mole, which was derived from 3 fatty acid + 1 glycerol – 3

95 water = 3(263 g/mole) + 1(92 g/mole) – 3(18 g/mole). The molar concentration (M) of

lard in methylene chloride was calculated with the average molecular weight of lard

triglycerides (Jung and Min 1991).

Reproducibility of headspace oxygen and peroxide value analyses

The coefficients of variation for measuring headspace oxygen and peroxide value for 5 replicates were 1.9 and 2.4%, respectively (data not shown). The low coefficients of variations showed that the headspace oxygen and peroxide value analyses in sample bottles were considered as reproducible and acceptable methods to study the oxidation of lard.

Effects of α-, β-, γ-, and δ-tocotrienol on chlorophyll photosensitized oxidation of lard

The effects of α-, β-, γ-, and δ-tocotrienol at 1.2 mM on the headspace oxygen

and peroxide values of 0.4 M lard in methylene chloride containing chlorophyll b are

shown in Table 5.1. The correlation coefficient (r2) between headspace oxygen content and peroxide value was -0.94. Preliminary study showed that the headspace oxygen and peroxide value of lard without chlorophyll under light for 4 hours did not change. The headspace oxygen and peroxide value of lard with chlorophyll in the dark for 4 hours did not change. The changes of headspace oxygen and peroxide value occurred only in the presence of chlorophyll and light. The effects of α-, β-, γ-, and δ-tocotrienols on the headspace oxygen and peroxide value of samples without chlorophyll were not present.

α-, β-, γ-, and δ-Tocotrienols had antioxidant effects in the presence of chlorophyll under

96 light during 4 hours of storage. The antioxidative effects were in the order of α- tocotrienol > β-tocotrienol = γ-tocotrienol > δ-tocotrienol. Tukey’s test showed the lard in methylene chloride containing 1.2 mM α-, β-, γ-, and δ-tocotrienols had significantly

higher headspace oxygen and lower peroxide values compared to the control lard sample

containing 0 mM tocotrienol at p<0.05.

Determination of singlet oxygen formation rate

The headspace oxygen in the sample bottle with lard in methylene chloride

containing 4.4 × 10-6 M chlorophyll b decreased as the concentration of lard increased from 0 to 0.1, 0.25, and 0.4 M. The headspace oxygen contents in the sample having 0,

0.1, 0.25, and 0.4 M lard in methylene chloride after 4 hrs of light storage were 20.95,

17.15, 15.22, and 13.97%, respectively. Figure 5.1 is the reciprocal plot of the depleted headspace oxygen versus the lard concentration, which can be used to determine the singlet oxygen formation rate (King and Min 1998). The intercept of the plot is the inverse of the singlet oxygen formation rate, which was 0.229 mL headspace/µmole O2

(Figure 5.1). The inverse of the intercept is 4.36 µmole O2/ mL headspace/ 4hrs. The rate of singlet oxygen formation by 4.4 × 10-6 M chlorophyll b in methylene chloride with lard was 1.09 µmole O2/ mL headspace/ hr. King and Min (1998) reported that the rate of singlet oxygen formation by 15 ppm riboflavin in 12% water and 88% acetone system was 1.57 µmole O2/ mL headspace/ hr.

Determination of quenching mechanisms and kinetics of α-, β-, γ-, and δ-tocotrienol

97 A schematic diagram for the formation of singlet oxygen by sensitizer and oxidized product (AO2) by singlet oxygen oxidation under light is shown in Figure 5.2

(Jung and Min 1991). When chlorophyll (Sen) absorbs light energy, it becomes an

excited singlet chlorophyll (1Sen*) and then becomes an excited triplet chlorophyll

3 ( Sen*) by an intersystem crossing (KISC) mechanism. The energy of excited triplet

3 1 chlorophyll is transferred to ordinary triplet oxygen ( O2) to produce singlet oxygen ( O2) by triplet-triplet annihilation. Singlet oxygen reacts with substrates (A) to form oxidized products (AO2), or singlet oxygen is quenched physically or chemically by quenchers or is decayed without radiation.

The steady-state kinetic equation for singlet oxygen oxidation in the presence of chlorophyll as photosensitizer is as follows (Foote 1979):

+ + d[AO ] − − k [Q] (k k − )[Q] k − 2 1 = 1 + Q + q ox Q d × 1 { } K (1 3 )(1 [A] ) dt ko [ O2 ] kr where [AO2] is the concentration of oxidized products; K, the rate of singlet oxygen formation (the quantum yield of intersystem crossing); kr, reaction rate constant of substrate with singlet oxygen; A, substrate; kq, reaction rate constant of physical singlet

oxygen quenching; kox-Q, reaction rate constant of chemical singlet oxygen quenching; Q, quencher; and kd, decay rate of singlet oxygen. In this study, the substrate (A) is lard and the quencher (Q) is tocotrienol.

If a quencher reduced the photosensitized oxidation of food by only singlet oxygen quenching, the steady-state kinetic equation for singlet oxygen formation is

98 simplified due to kq[Q] = 0 as following (Foote 1979; Jung and others 1991; Huang and others 2004b):

+ + d[O ] − d[AO ] − − (k k − )[Q] k − {− 2 } 1 = { 2 } 1 = K 1 (1+ q ox Q d ×[A] 1 ) dt dt k r

-1 -1 This equation is in slope-intercept form. When plotting [AO2] versus [A] at various concentrations of quencher, the intercept is equal to K-1 and the slope is equal to

(k + k − )[Q] + k K −1 × q ox Q d . The ratio of the slope to intercept (S/I) will be kr

+ + (kq kox−Q )[Q] kd , which is independent of O2 concentration. The intercepts of the kr plots at various concentration of a quencher will be independent of [Q] and the slope will be dependent on [Q] (Foote 1979). Then, the plot of S/I against [Q] will be obtained as a

k (k + k − ) line graph with an intercept of d and a slope of q ox Q . kr kr

When there is no quencher added in an experimental system, the equation can be

simplified to the following (Foote 1979; Jung and others 1991; Huang and others 2004b):

d[AO ] − − k − { 2 } 1 = K 1 (1+ d ×[A] 1 ) dt k r

-1 -1 The intercept and slope of the plot of [AO2] versus [A] at various

-1 -1  kd  concentrations of quencher is K and K   , respectively. The ratio of the slope to  kr 

kd intercept will be . If the decay rate of singlet oxygen (kd) is known, the reaction rate kr

99 constant (kr) between lard and singlet oxygen can be determined (Jung and Min 1991;

Huang and others 2004b; Kim and others 2006).

The reciprocal plot of depleted headspace oxygen against the lard concentration of 0.1, 0.25, and 0.4 M in methylene chloride at 0, 0.3, 0.6, 0.9 or 1.2 mM α-, β-, γ-, or δ- tocotrienol stored under light for 4 hrs are shown in Figure 5.3, 5.4, 5.5, and 5.6. The intercepts were independent of the concentrations of α-, β-, γ-, or δ-tocotrienol. The

slopes increased as the concentration of α-, β-, γ-, or δ-tocotrienol increased. Since the

intercepts of different concentration of α-, β-, γ-, or δ-tocotrienol were not significantly

different and the slopes increased according to the increase of α-, β-, γ-, or δ-tocotrienol

concentration, tocotrienols quenched singlet oxygen to reduce chlorophyll

photosensitized oxidation of lard.

-1 -1 The linear regression for the plot of [AO2] versus [A] containing 0 mM α-, β-,

γ-, and δ-tocotrienol was y = 0.39x + 0.229 (r2 = 0.99), where y is the reciprocals of

depleted headspace oxygen and x is the reciprocals of lard concentration (Figure 5.1).

The slope over the intercept of the regression line was 0.17. The slope/intercept is (kd/kr)

4 -1 = 0.17 and the decay rate of singlet oxygen (kd) in methylene chloride was 1.1 × 10 sec

(Salokhiddinov and others 1981). The reaction rate (kr) between lard and singlet oxygen is 6.5 × 104 M-1sec-1. The reaction rate between ordinary triplet oxygen and soybean oil is about 10 M-1sec-1 (Min and Bradley 2002). This suggested the oxidation with singlet oxygen is much faster than triplet oxygen oxidation. Jung and Min (1991) reported the reaction rate between soybean oil and singlet oxygen was 1.0 ×105 M-1sec-1. The reaction

100 rate with singlet oxygen and lard having less unsaturated fatty acid was lower than that of

soybean oil having more unsaturated fatty acids.

The ratios of slope/intercept of the plots containing 0, 0.3, 0.6, and 1.2 mM α-

tocotrienol were 0.316, 0.389, 0.513, and 0.408, respectively. The slope/intercept at 1.2

mM did not increase as the concentration of α-tocotrienol increased from 0.9 to 1.2 mM.

This suggested that α-tocotrienol at the higher concentration of 1.2 mM quenched not

only singlet oxygen but also excited triplet chlorophyll (Figure 5.2). To determine the

total singlet oxygen quenching rate (kq + kox-Q) of α-tocotrienol, the slope/intercept against the concentration of α-tocotrienol at 0, 0.3, and 0.6 mM was plotted (Figure5.7).

The linear regression equation of the plot of slope/intercept against concentration of α- tocotrienol was y = 333.74x + 0.308, where y is slope/intercept and x is concentration of

α-tocotrienol (M). The slope of this regression line was (kq + kox-Q)/kr = 333.74 and the

4 -1 -1 kr was 6.5 × 10 M sec . Therefore, the singlet oxygen quenching rate (kq + kox-Q) of α- tocotrienol was calculated to 2.16 × 107 M-1sec-1.

The singlet oxygen quenching rates of β-, γ-, and δ-tocotrienol at 0, 0.3, 0.6, and

0.9 mM (Figure 5.8., 5.9, and 5.10) were calculated as same as α-tocotrienol and are shown in Table 5.2. The singlet oxygen quenching rates of α-, β-, γ-, and δ-tocotrienol were 2.16 ×107, 1.99 ×107, 2.05 ×107, and 0.80 ×107 M-1sec-1, respectively. Jung and others (1991) reported that singlet oxygen quenching rates of α-tocopherol in purified soybean oil were 2.6 × 107 M-1sec-1. α-, β-, γ-, and δ-Tocotrienols were excellent

101 antioxidants in photosensitized singlet oxygen oxidation and they can be used as natural

singlet oxygen quenchers to improve the oxidative stabilities of foods stored under light.

5.5 Conclusion

Lard was successfully used as a model system to study the effects, quenching mechanisms, and kinetics of α-, β-, γ-, and δ-tocotrienol on photosensitized oxidation of

lipids. The 4.4 × 10-6 M chlorophyll b in methylene chloride at 3,000 lux produced singlet oxygen at 1.089 µmole oxygen/mL headspace/hr. The reaction rate constant of singlet oxygen with lard were 6.5 × 104 M-1sec-1. α-, β-, γ-, and δ-Tocotrienols acted as antioxidants in chlorophyll photosensitized oxidation of lard by quenching singlet oxygen.

The singlet oxygen quenching rates of α-, β-, γ-, and δ-tocotrienol were 2.16 × 107, 1.99

× 107, 2.05 × 107, and 0.80 × 107 M-1sec-1, respectively. α-Tocotrienol at the high concentration of 1.2 mM quenched both singlet oxygen and excited triplet chlorophyll.

α-, β-, γ-, and δ-Tocotrienol were excellent antioxidants in photosensitized singlet oxygen oxidation.

102 5.6 References

[AOCS] American Oil Chemists’ Society. 2004. Official methods and recommended practices of the AOCS. 5th ed. Champaign, Ill.:AOCS.

Azzi A, Stocker A. 2000. Vitamin E: non-antioxidant roles. Prog in Lipid Res 39: 231-55. Belitz H-D, Grosch W. 1999. Food Chemistry, 2nd ed. Berlin: Springer. pp 207-8, 381-2, 606.

Bradley DG. 1991. Riboflavin photosensitized singlet oxygen oxidation in milk products. Ph. D. Dissertation. The Ohio State University, Columbus.

Bradley DG, Min DB. 1992. Singlet oxygen oxidation in foods. Crit Rev Food Sci Nutri 31: 211-36.

Choe E, Min DB. 2005. Chemistry and reactions of reactive oxygen species in foods. J Food Sci 70: R142-R59.

Choe E, Min DB. 2006. Mechanisms and factors for edible oil oxidation. Compr Rev Food Sci Food Saf 5: 169-86.

Evans JC, Kodali DR, Addis PB. 2002. Optimal tocopherol concentrations to inhibit soybean oil oxidation. J Am Oil Chem Soc 79: 47-51.

Foote CS. 1979. Quenching of singlet oxygen. In: Wasserman HH and Murray RW. Singlet Oxygen. New York: Academic Press. p 139-73.

Foote CS, Denny RW. 1968. Chemistry of singlet oxygen. VII Quenching by β-carotene. J Am Oil Chem Soc 90: 6233-5.

103

Huang R, Choe E, Min DB. 2004a. Effects of riboflavin photosensitized oxidation on the volatile compounds of soymilk. J Food Sci 69: C733-C8.

Huang R, Choe E, Min DB. 2004b. Kinetics for singlet oxygen formation by riboflavin photosensitization and the reaction between riboflavin and singlet oxygen. J Food Sci 69: C726-C32.

Huang R, Kim HJ, Min DB. 2006. Photosensitizing effect of riboflavin, lumiflavin, and lumichrome on the generation of volatiles in soy milk. J Agric Food Chem 54: 2359-64.

Jung MY, Choe E, Min DB. 1991. Effects of α-, γ-, and δ-tocopherols on the chlorophyll photosensitized oxidation of soybean oil. J Food Sci 56: 807-15.

Jung MY, Lee HO, Min DB. 1998. Singlet oxygen and ascorbic acid effects on dimethyl disulfide and off-flavor in skim milk exposed to light. J Food Sci 63: 408-12.

Jung MY, Min DB. 1990. Effects of α-, γ-, and δ-tocopherols on the oxidative stability of soybean oil. J Food Sci 55: 1464-5.

Jung MY, Min DB. 1991. Effects of quenching mechanisms of carotenoids on the photosensitized oxidation of soybean oil. J Am Oil Chem Soc 68: 653-7.

Kamal-Eldin A, Appelqvist L-Å. 1996. The chemistry and antioxidant properties of tocopherols and tocotrienols. Lipids 31:671-701.

Kim HJ, Lee MY, Min DB. 2006. Singlet oxygen oxidation rates of α-, γ-, and δ- tocopherols. J Food Sci 71: C465-C8.

104 King JM, Min DB. 1998. Riboflavin photosensitized singlet oxygen oxidation of Vitamin D. J Food Sci 63: 31-4.

Kulas E, Ackman RG. 2001. Properties of α-, γ-, and δ-tocopherol in purified fish oil triacylglycerols. J Am Oil Chem Soc 78: 361-7.

Lee SW, Min DB. 1990. Effects, quenching mechanisms, and kinetics of carotenoids in chlorophyll-sensitized photooxidation of soybean oil. J Agric Food Chem 38: 1630-4.

Min DB, Boff JM. 2002. Chemistry and reaction of singlet oxygen in foods. Compr Rev Food Sci Food Saf 1: 58-72.

Min DB, Callison AL, Lee HO. 2003. Singlet oxygen oxidation for 2-pentylfuran and 2- pentenylfuran formation in soybean oil. J Food Sci 68: 1175-8.

Parker SP. 1982. McGraw-Hill Encyclopedia of Science and Technology, New York: McGraw-Hill.

Reische DW, Lillard DA, Eitenmiller RR. 2002. Antioxidants. In: Akoh CC and Min DB. Food Lipids. 2nd edition. New York: Marcel-Dekker. p 489-516.

Salokhiddinov KI, Byteva IM, Gurinovich GP. 1981. Lifetime of singlet oxygen in different solvents. Zh Prikl Spektrosk 34: 892.

105 Schroeder MT, Becker EM, Skibsted LH. 2006. Molecular mechanism of antioxidant synergism of tocotrienols and carotenoids in palm oil. J Agric Food Chem 54: 3445-53.

Serbinova E, Kagan V, Han D, Packer L. 1991. Free radical recycling and intra- membrane mobility in the antioxidative properties of α-tocopherol and α-tocotrienol. Free Rad Biol Med 10: 263-76.

Sundram K, Gapor A. 1992. Vitamin E from palm oil: its extraction and nutritional properties. Lipid Technol Nov/Dec: 137-41.

Wagner K-H, Wotruba F, Elmadfa I. 2001. Antioxidative potential of tocotrienols and tocopherols in coconut fat at different oxidation temperature. Eur J Lipid Sci Technol 103: 746-51.

Whang K, Peng IC. 1988. Electron paramagnetic resonance studies of the effectiveness of myoglobin and its derivatives as photosensitizers in singlet oxygen generation. J Food Sci 53: 1863-5.

Yang WT. 1994. Effects of synthetic food colorants on singlet oxygen oxidation of foods. Ph. D. Dissertation. The Ohio State University, Columbus.

106 Yoshida Y, Niki E, Noguchi N. 2003. Comparative study on the action of tocopherols and tocotrienols as antioxidant: chemical and physical effects. Chem Phys Lipids 123: 63-75.

107

Headspace oxygen Peroxide value Sample (%) (meq/kg) Without tocotrienol 13.97a 13.65a α-Tocotrienol 17.08b 7.72b β-Tocotrienol 16.59b 8.76c γ-Tocotrienol 16.83b 8.44c δ-Tocotrienol 15.85b 9.87d a Different letters in the same column are significantly different at α = 0.05.

Table 5.1. Effects of α-, β-, γ-, and δ-tocotrienol at 1.2 mM on headspace oxygen and peroxide value of 0.4 M lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs

108

Tocotrienol Linear Regressiona Quenching Rate (M-1sec-1)

α y = 333.47x + 0.3081 2.16 × 107 β y = 307.65x + 0.1490 1.99 × 107 γ y = 315.67x + 0.2013 2.05 × 107 δ y = 123.34x + 0.1133 0.80 × 107 a y: Slope/Intercept in M, x: Concentration of tocotrienol in M

Table 5.2. The linear regressions from Figure 5.7, 5.8, 5.9, and 5.10 and quenching rates of α-, β-, γ-, and δ-tocotrienol

109

0.8

0.7

0.6 ) 2

0.5 mole O mole µ µ µ µ

0.4

0.3 y = 0.0389x + 0.2294 2 (mL headspace/ R = 0.99 0.2 1/depleted headspace oxygen

0.1

0 0 5 10 15 1/Lard (1/M)

Figure 5.1. Headspace oxygen depletion of 0.1, 0.25, and 0.4 M lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs

110

hv 3 1 KISC 3 O2 1 A Sen Sen* Sen* O2 AO2 ko kr

Q k Q KQ kq d

Q kox-Q

3 3 O2 O2 1 Sen Sen QO2

Figure 5.2. Schematic diagram for the formation of oxidized product via singlet oxygen oxidation under light: Sen = chlorophyll; A = lard; AO2 = oxidized lard; Q = tocotrienol

111

1.6

1.4

1.2 ) 2

1.0 mole O mole µ µ µ µ

0.8

0.6

(mL headspace/ 0 mM 0.4 1/Depleted headspace oxygen 0.3 mM 0.6 mM 0.2 1.2 mM

0.0 051015 1/Lard (1/M)

Figure 5.3. Effect of 0, 0.3, 0.6, or 1.2 mM α-tocotrienol on the headspace oxygen depletion of lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs

112

1.4

0 mM 1.2 0.3 mM 0.6 mM

) 0.9 mM 2 1 mole O mole µ µ µ µ 0.8

0.6

(mL headspace/ 0.4 1/Depleted headspace oxygen

0.2

0 051015 1/Lard (1/M)

Figure 5.4. Effect of 0, 0.3, 0.6, or 0.9 mM β-tocotrienol on the headspace oxygen depletion of lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs

113

1.8

1.6 0 mM 0.3 mM 1.4 0.6 mM 0.9 mM 1.2 mole O2) mole µ µ µ µ 1.0

0.8

0.6 (mL headspace/

1/Depleted headspace oxygen 0.4

0.2

0.0 051015 1/Lard (M)

Figure 5.5. Effect of 0, 0.3, 0.6, or 0.9 mM γ-tocotrienol on the headspace oxygen depletion of lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs

114

1.0

0.9 0 mM 0.3 mM 0.8 0.6 mM )

2 0.9 mM 0.7

mole O mole 0.6 µ µ µ µ

0.5

0.4

0.3 (mL headspace/ 1/Depleted headspace oxygen 0.2

0.1

0.0 0 5 10 15 1/Lard (1/M)

Figure 5.6. Effect of 0, 0.3, 0.6, or 0.9 mM δ-tocotrienol on the headspace oxygen

depletion of lard in methylene chloride containing 4.4 × 10-6 M chlorophyll b under light for 4 hrs

115

0.6

0.5

0.4

y = 333.47x + 0.3081 0.3 2 R = 0.98 Slope/Intercept 0.2

0.1

0.0 0.0000 0.0002 0.0004 0.0006 0.0008

α-Tocotrienol (M)

Figure 5.7. Relation of slope/intercept in Figure 5.3 to the concentration of α-tocotrienol

116

0.5

0.4

0.3

y = 307.65x + 0.149 0.2 2 Slope/Intercept R = 0.96

0.1

0 0 0.0002 0.0004 0.0006 0.0008 0.001

β-Tocotrienol (M)

Figure 5.8. Relation of slope/intercept in Figure 5.4 to the concentration of β-tocotrienol

117

0.6

0.5

0.4

0.3

y = 370.69x + 0.1851 2 Slope/Intercept 0.2 R = 0.95

0.1

0.0 0 0.0002 0.0004 0.0006 0.0008 0.001

γ-Tocotrienol (M)

Figure 5.9. Relation of slope/intercept in Figure 5.5 to the concentration of γ-tocotrienol

118

0.25

0.2

0.15

0.1 y = 123.34x + 0.1133 R2 = 0.98 Slope/Intercept

0.05

0 0 0.0002 0.0004 0.0006 0.0008 0.001

δ-Tocotrienol (M)

Figure 5.10. Relation of slope/intercept in Figure 5.6 to the concentration of δ-tocotrienol

119

CHAPTER 6

EFFECTS AND PROOXIDANT MECHANISMS OF OXIDIZED α-TOCOPHEROL

ON THE OXIDATIVE STABILITY OF SOYBEAN OIL

6.1 Abstract

α-Tocopherol was oxidized in methanol containing methylene blue for 30 hrs under light. The effects of 0, 250, 500, 1000 and 1500 ppm of oxidized α-tocopherol on the oxidative stability of purified soybean oil in the dark at 55°C were studied by measuring the peroxide values and headspace oxygen contents in sample bottles. As the concentrations of oxidized α-tocopherol increased, the peroxide values increased and the headspace oxygen contents decreased during the 6 days of storage. The oxidized α- tocopherol compounds acted as prooxidant on the peroxide values and headspace oxygen contents of purified soybean oil. Tukey’s test showed that oxidized α-tocopherol had a significant effect on the peroxide value and headspace oxygen disappearance of oil at p<0.05. The prooxidant mechanisms of oxidized α-tocopherol may be due to α-

120 tocopherol peroxy radical, α-tocopherol oxy radical, hydroxy radical, and singlet oxygen

formed from the α-tocopherol. The oxidized α-tocopherol containing polar and nonpolar

groups in the same molecule may reduce the surface tension of oil to increase the transfer

of headspace oxygen to oil and accelerate the oil oxidation.

6.2 Introduction

Tocopherols are the most important natural antioxidants present in vegetable oils

(Huang and others 1995; Wagner and Elmadfa 2000; Warner 2005). Tocopherols protect

lipids from oxidation by donating hydrogen from their phenolic group on the chromanol

ring to peroxy radical in propagation step (Nawar 1996; Kamal-Eldin and Appelqvist

1996). The reaction rate of α-tocopherol with lipid peroxy radical is 107 M-1sec-1 (Niki and others 1984; Choe and Min 2005) and is 105 to 106 times faster than that of unsaturated lipid with lipid peroxy radical (Niki and others 1984; Naumov and Vasil’ev

2003). Soybean oil contains about 1000-1500 ppm of tocopherols and the compositions of α-, β-, γ-, and δ-tocopherol are 4, 1, 66, and 29%, respectively (Jung and others 1989).

The optimum concentrations of α-, γ-, and δ-tocopherol to increase the oxidative stability of oil were 100, 250 ~ 500, and 500 ~ 1000 ppm, respectively (Jung and Min 1990;

Yoshida and others 1994; Evans and others 2002). However, the antioxidant activity of tocopherols decreased and became prooxidant above the optimum concentrations (Jung

121 and Min 1990; Bowry and Stocker 1993; Huang and others 1995; Evans and others 2002;

Rietjens and others 2002).

Tocopherols are degraded by oxidation reaction (Verleyen and others 2001a).

Tocopherols were rapidly destroyed in soybean and rapeseed oil at frying conditions

(Miyagawa and others 1991; Marquez-Ruiz and others 1999). Tocopherols donate hydrogen to peroxy radicals as antioxidant and form tocopherol radicals. Tocopherol radicals combine with another lipid peroxy radicals in a termination reaction yielding nonradical tocopherol oxidation products (Mukai and others 1993; Verleyen and others

2001a). Tocopherol radicals also dimerize with another tocopherol radical to form dimers and/or trimers (Kamal-Eldin and Appelqvist 1996). α-Tocopherolquinone, α-

tocopherolhydroquinone, 4a, 5-epoxy-α-tocopherolquinone, and 7, 8-epoxy-α-

tocopherolquinone from α-tocopherol oxidation have been identified in beef and bovine

muscle microsomes (Fasutman and others 1999; Liebler and others 1995), in triolein

(Verleyen and others 2001a) and triolein and tripalmitin mixture (Verleyen and others

2001b) and in fish muscle (Pazos and others 2005). Pazos and others (2005) reported

that α-tocopherol in chilled and frozen fish muscle significantly decreased and α-

tocopherolquinone, α-tocopherolhydroquinone, and epoxy-α-tocopherolquinone

increased during storage. However, the detailed chemical mechanisms for the formation

and prooxidant properties of oxidized α-tocopherol have not been reported. The

objectives of this research were to study the effect of oxidized α-tocopherol on the

oxidative stability of soybean oil and to postulate the formation and prooxidant

mechanisms of oxidized α-tocopherol compounds during soybean oil oxidation.

122 6.3 Materials and Methods

Materials

The refined, bleached and deodorized (RBD) soybean oil from Archer Daniels

Midlands (Decatur, IL) and α-tocopherol from Sigma Chemical Co. (St. Louis, MO)

were purchased.

Purification of soybean oil

The RBD soybean oil was purified by the method of Jung and Min (1992). Fifty grams of soybean oil were passed through a chromatographic column (55 × 4 cm) packed with a series of 100 g of activated silicic acid (100 mesh, Mallinkrodt Co., Paris, KA), 30 g of a 2:1 mixture of activated charcoal (J.T. Baker Chemical Co., Phillipsburg, NJ) and

Celite, 120 g of a 2:1 mixture of powdered sugar and Celite, and 100 g of activated silicic acid. The flow rate of oil through the column was 2 mL/hr.

Preparation of oxidized α-tocopherol

Oxidized α-tocopherol was prepared by the method of Jung and Min (1992). One

gram of α-tocopherol and 120 µmol of methylene blue were dissolved in 200 mL of

methanol. Twenty five milliliters of the α-tocopherol sample were transferred to a 100

mL serum bottle and sealed airtight with a Teflon coated stopper. The α-tocopherol

sample was stored in a light box of 4000 Lux for 30 hrs at 28 °C. One hundred milliliters

of hexane were added to extract the oxidized tocopherols and methylene blue was

123 removed by washing several times with deionized water. Hexane was removed using a

rotary vacuum evaporator at room temperature. Oxidized α-tocopherol was then obtained.

Characterization of oxidized α-tocopherol compounds by infrared spectrometry

A small drop of oxidized α-tocopherol sample was smeared onto a sodium

chloride disc. The infrared spectrum was determined with the Beckman Acculab 2

Infrared Spectrometer (Beckman Instruments Inc., Fullerton, CA).

Sample preparation for effect of oxidized α-tocopherol on the oxidative stability of oil

To study the effect of oxidized α-tocopherol on the oxidative stability of oil,

purified soybean oils containing 0, 250, 500, 1000 and 1500 ppm of oxidized α-

tocopherol were prepared. Five grams of sample were transferred into a 30 mL serum

bottle and sealed airtight with Teflon coated septum and aluminum cap (Supelco,

Bellefonte, PA). The sample bottles were stored in duplicate for 6 days in a Blue M oven

(Blueland, IL) at 55 °C. The oxidative stability of soybean oil was determined by

measuring peroxide value and headspace oxygen of sample bottles.

Determination of peroxide value

Peroxide values of the samples were determined using AOCS method Cd 8-53

(AOCS 2004). Five grams of oil were transferred into a dry 250 mL Erlenmeyer flask

and 30 mL of acetic acid/chloroform (3:2) were added. The flask was swirled and 0.5

124 mL saturated KI solution was added. After allowing the solution to set for 1 min, 30 mL

of distilled water and 0.5 mL of starch indicator were added. This was titrated against 0.1

N Na2S2O3 until the blue color disappeared. The peroxide value was expressed as milliequivalent of peroxides per kg oil.

Determination of headspace oxygen

The headspace oxygen was determined by injecting 50 µL headspace gas of a sample bottle into a HP 5890 gas chromatograph equipped with a thermal conductivity detector. A stainless steel column (1.8 m × 0.32 cm) with 80/100 Molecular Sieve 13×

(Alltech Assoc. Inc., Deerfield, IL) was used. The flow rate of helium gas was 20 mL/min. The temperatures of oven, injector, and thermal conductivity detector were 40,

120 and 150 °C, respectively. Electronic counts of 50 µL headspace gas of a sample bottle were quantified by a Hewlett Packard 3390 electronic integrator (Huang and others

2004).

Statistical Analysis

The peroxide values and headspace oxygen contents were the mean values of duplicate samples. The data were analyzed by Analysis of variance and Tukey’s test at α

= 0.05 using the Statistical Analysis System (SAS Inst., Cary, NC, 2004).

125 6.4 Results and Discussion

Characterization of purified soybean oil and oxidized α-tocopherol

The soybean oil that passed through the column packed with silicic acid, charcoal, celite and sugar was designated as purified soybean oil. The purified soybean oil was colorless, tasteless, and odorless and peroxide value was 0. The analyses showed that free fatty acids, phospholipids, tocopherols, chlorophylls, carotenoids and oxidized compounds were not present in the purified soybean oil. The column packed with silicic acid, charcoal, celite and sugar effectively removed all minor compounds from RBD soybean oil.

The oxidized α-tocopherol compounds prepared for 30 hrs under light was characterized by infrared spectrometry and peroxide value determination. The infrared spectrum showed the broad absorption peak at 3330 cm-1 which may be due to the intermolecular hydrogen bond of -O-H group of tocopherol or TOO-H of tocopherol hydroperoxide and the peak at 1730 cm-1 which is due to the C=O stretching vibration of

aromatic compounds. The peak at 1730 cm-1 is possibly due to the quinone group of

tocopherol (Conley 1972; Silverstein and others 1981). The peroxide value determined

by AOCS (2004) showed that the oxidized α-tocopherol contains hydroperoxide. The

combined results of infrared spectrum and peroxide value determination showed the

oxidized α-tocopherol compounds contain hydroxyl group, carbonyl group and

hydroperoxide. Faustman and others (1999), Verleyen and others (2001a, 2001b) and

Pazos and others (2005) identified α-tocopherolquinone, α-tocopherolhydroquinone, and

126 epoxy-α-tocopherolquinone from α-tocopherol oxidation. α-Tocopherolquinone, α-

tocopherolhydroquinone, epoxy-α-tocopherolquinone, and tocopherol hydroperoxide

contain hydroxyl, carbonyl and hydroperoxyl groups. Our infrared spectrum and

peroxide value analysis suggested the presence of α-tocopherolquinone, α-

tocopherolhydroquinone, and epoxy-α-tocopherolquinone in our oxidized α-tocopherol.

Effects of oxidized α-tocopherol on the oxidative stability of soybean oil

The effects of oxidized α-tocopherol on the oxidative stability of soybean oil were determined by measuring both peroxide values and headspace oxygen contents of samples. The effects of 0, 250, 500, 1000 and 1500 ppm of oxidized α-tocopherol on the

peroxide values of purified soybean oil during 6 days of storage at 55 °C are shown in

Figure 6.1. The coefficient of variation for peroxide value analyses of oil was 2%.

Peroxide values of soybean oil containing 0, 250, 500, 1000 and 1500 ppm of oxidized α-

tocopherol gradually increased during storage. As the concentration of oxidized α-

tocopherol in soybean oil increased from 0 to 250, 500, 1000 and 1500 ppm, the mean of

peroxide values for 6 days of storage increased from 37.5 to 43.5, 49.1, 54.9 and 61.4

meq/kg oil, respectively. Tukey’s test showed that the peroxide values of soybean oil

containing oxidized α-tocopherol were significantly different with the peroxide value of

the control oil containing no oxidized α-tocopherol at p<0.05.

The effects of oxidized α-tocopherol on the headspace oxygen contents of

purified soybean oils during storage at 55 °C are shown in Figure 6.2. The coefficient of

127 variation for headspace oxygen analyses was 3%. The headspace oxygen content of a

fresh sample was 20.7%. The headspace oxygen of soybean oil containing oxidized α-

tocopherol decreased as the storage time increased. As the oxidized α-tocopherol

increased from 0 to 250, 500, 1000 and 1500 ppm, the mean of headspace oxygen content

of soybean oil for 6 days of storage decreased from 15.5 to 14.2, 13.1, 12.8 and 11.2%,

respectively. Tukey’s test showed that the headspace oxygen contents of soybean oil

containing 250, 500, 1000 and 1500 ppm oxidized α-tocopherol were lower than that of

the control oil containing no oxidized α-tocopherol at p<0.05. The linear coefficient (r)

between the peroxide value and headspace oxygen content of soybean oil during storage

was -0.99. That is, as the headspace oxygen decreased, the peroxide value increased.

The combined results of peroxide values and headspace oxygen contents indicate that

oxidized α-tocopherol had prooxidant effect on the oxidation of purified soybean oil

during storage. The prevention of tocopherol oxidation and the removal of oxidized

tocopherols during processing and storage could improve the oxidative stability of

soybean oil.

Formations of α-tocopherol peroxy radical, α-tocopherol oxy radical, and hydroxy radical

α-Tocopherolquinone, α-tocopherolhydroquinone, epoxy-α-tocopherolquinone, and α-tocopherol hydroperoxide formed during the oxidation of α-tocopherol were identified by Faustman and others (1999), Verleyen and others (2001a, 2001b) and Pazos and others (2005). The mechanisms for the formation of α-tocopherolquinone, 4a, 5-

128 epoxy-α-tocopherolquinone, 7, 8-epoxy-α-tocopherolquinone and α-tocopherolquinone

hydroperoxide are postulated in Figure 6.3, 6.4, 6.5, and 6.6. The standard reduction

potentials for common free radicals are shown in Table 2.3. The reaction rates of lipid

oxidation in the presence of tocopherol are in Table 6.1. When the published data on the

reaction rates and reduction potentials of α-tocopherol derivatives were not available, the

estimated values based on the data shown in Table 2.3 and 6.1 were used to discuss the

reaction mechanisms and kinetics of α-tocopherol derivatives in Figure 6.3, 6.4, 6.5, and

6.6.

The formations of α-tocopherol peroxide, α-tocopherolquinone, 4a, 5-epoxy-α-

tocopherolquinone and 7, 8-epoxy-α-tocopherolquinone from α-tocopheroxy radical and

oxygen are shown in Figure 6.3. Figure 6.3-A shows the formation of α-

tocopherolquinone from α-tocopheroxy radical. The α-tocopheroxy radical formed from

α-tocopherol by donating hydrogen to unsaturated lipid peroxy radical forms 8a-carbon

centered α-tocopheryl radical by resonance. The 8a-carbon centered α-tocopheryl radical

reacts with oxygen to form α-tocopherol peroxy radical at the rate of 107 M-1sec-1 (Table

6.1). The α-tocopherol peroxy radical has the reduction potential of approximately 1000 mV (Table 2.3) and can abstract a hydrogen from unsaturated fatty acids to form α- tocopherol hydroperoxide at the rate of 101~102 M-1sec-1 (Table 6.1). The peroxide bond strength of α-tocopherol hydroperoxide is about 44 kcal/mole (Hiatt and others 1968).

The α-tocopherol hydroperoxide is cleaved to form α-tocopherol oxy radical and

hydroxy radical (·OH) (Figure 6.3-A). The α-tocopherol oxy radical forms α-

129 tocopherolquinone oxy radical at the 2 position by cleaving the bond between 1 and 8a.

The α-tocopherolquinone oxy radical has the reduction potential of about 1600 mV

(Table 2.3) and forms α-tocopherolquinone by abstracting a hydrogen from unsaturated lipids at the rate of 107 M-1sec-1 (Table 6.1).

Figure 6.3-B shows the formation of 4a, 5-epoxy-α-tocopherolquinone from the carbon centered α-tocopheryl radical at the 5 position. The 5-carbon centered α- tocopheryl radical reacts with oxygen to form α-tocopherol peroxy radical and then forms

α-tocopherol hydroperoxide by abstracting a hydrogen from unsaturated lipids. The α- tocopherol hydroperoxide is cleaved to form α-tocopherol oxy radical and hydroxy radical (·OH) (Figure 6.3-B). The α-tocopherol oxy radical at the 5 position is cyclized to form 4a, 5-epoxy-α-tocopheryl radical at the 8a position. The 4a, 5-epoxy-α- tocopheryl radical reacts with oxygen to form 4a, 5-epoxy-α-tocopherol peroxy radical and then form 4a, 5-epoxy-α-tocopherol hydroperoxide. The cleavage of 4a, 5-epoxy-α- tocopherol hydroperoxide forms 4a, 5-epoxy-α-tocopherolquinone oxy radical at the carbon 2 position and hydroxy radical (·OH). The 4a, 5-epoxy-α-tocopherolquinone oxy radical forms 4a, 5-epoxy-α-tocopherolquinone as shown in Figure 6.3-B.

Figure 6.3-C shows the formation of 7, 8-epoxy-α-tocopherolquinone from the carbon centered α-tocopheryl radical at the 7 position. The 7-carbon centered α- tocopheryl radical reacts with oxygen to form α-tocopherol peroxy radical and then forms

α-tocopherol hydroperoxide by abstracting a hydrogen. The α-tocopherol hydroperoxide produces α-tocopherol oxy radical and hydroxy radical (·OH) (Figure 6.3-C). The α-

130 tocopherol oxy radical at the 7 position forms 7, 8-epoxy-α-tocopheryl radical at the 8a position and then produces 7, 8-epoxy-α-tocopherol peroxy radical by reacting oxygen.

The 7, 8-epoxy-α-tocopherol peroxy radical forms 7, 8-epoxy-α-tocopherol

hydroperoxide. The 7, 8-epoxy-α-tocopherol hydroperoxide forms 7, 8-epoxy-α-

tocopherolquinone oxy radical at the carbon 2 position and hydroxy radical (·OH). The 7,

8-epoxy-α-tocopherolquinone oxy radical forms 7, 8-epoxy-α-tocopherolquinone as

shown in Figure 6.3-C. α-Tocopherol peroxy radical, α-tocopherol oxy radical, α-

tocopherolquinone oxy radical and hydroxy radical formed from the oxidation of α-

tocopherol are prooxidants (Table 2.3).

The formations of α-tocopherol peroxide, α-tocopherolquinone, 4a, 5-epoxy-α-

tocopherolquinone, and 7, 8-epoxy-α-tocopherolquinone from α-tocopheroxy radical and

lipid peroxy radical (ROO·) are shown in Figure 6.4. Figure 6.4-A shows the formation

of α-tocopherolquinone from α-tocopheroxy radical. The α-tocopheroxy radical forms

8a-carbon centered α-tocopheryl radical by resonance. The 8a-carbon centered α-

tocopheryl radical reacts with lipid peroxy radical (ROO·) to form α-tocopherol peroxide

at the rate of 106 M-1sec-1 (Table 6.1). The α-tocopherol peroxide can produce α-

tocopherol oxy radical at the 8a carbon and alkoxy radical (RO·) by cleaving peroxide

bond. The α-tocopherol oxy radical at the 8a position forms α-tocopherolquinone oxy

radical by cleaving the bond between 1 and 8a position. The α-tocopherolquinone oxy

radical forms α-tocopherolquinone (Figure 6.4-A).

131 Figure 6.4-B shows the formation of 4a, 5-epoxy-α-tocopherolquinone from the carbon centered α-tocopheryl radical at the 5 position. The 5-carbon centered α- tocopheryl radical reacts with lipid peroxy radical (ROO·) to form α-tocopherol peroxide.

The breakdown of α-tocopherol peroxide forms α-tocopherol oxy radical at the carbon 5

and alkoxy radical (RO·). The α-tocopherol oxy radical is cyclized to form 4a, 5-epoxy-

α-tocopheryl radical at the 8a position. The 4a, 5-epoxy-α-tocopheryl radical can react

with lipid peroxy radical (ROO·) to form 4a, 5-epoxy-α-tocopherol peroxide. The 4a, 5-

epoxy-α-tocopherol peroxide produces 4a, 5-epoxy-α-tocopherol oxy radical at the

carbon 8a and alkoxy radical (RO·). The 4a, 5-epoxy-α-tocopherol oxy radical at the

carbon 8a forms 4a, 5-epoxy-α-tocopherolquinone oxy radical at the carbon 2 and then

produces 4a, 5-epoxy-α-tocopherolquinone by abstracting a hydrogen from unsaturated

lipids (Figure 6.4-B).

Figure 6.4-C shows the chemical mechanism for the formation of 7, 8-epoxy-α-

tocopherolquinone from the carbon centered α-tocopheryl radical at the carbon 7. The 7-

carbon centered α-tocopheryl radical reacts with lipid peroxy radical (ROO·) to form α-

tocopherol peroxide which is cleaved to α-tocopherol oxy radical and alkoxy radical

(RO·). The α-tocopherol oxy radical is cyclized to form 7, 8-epoxy-α-tocopheryl radical

at the 8a position and then forms 7, 8-epoxy-α-tocopherol peroxide by reacting with lipid

peroxy radical (ROO·). The 7, 8-epoxy-α-tocopherol peroxide forms 7, 8-epoxy-α-

tocopherol oxy radical and alkoxy radical (RO·) by cleaving the peroxide bond. The 7, 8-

epoxy-α-tocopherol oxy radical forms 7, 8-epoxy-α-tocopherolquinone oxy radical at the

132 2 position which generates 7, 8-epoxy-α-tocopherolquinone by abstracting a hydrogen

from unsaturated lipids (Figure 6.4-C). α-Tocopherol peroxy radical, α-tocopherol oxy

radical, and α-tocopherolquinone oxy radical are prooxidants (Table 2.3).

Formation of α-tocopherolquinone peroxy radical and singlet oxygen

The formation of α-tocopherolquinone peroxy radical from carbon centered α- tocopheryl radical is shown in Figure 6.5. The carbon centered α-tocopherolquinone radical at the carbon 2 reacts with triplet oxygen to form α-tocopherolquinone peroxy radical at the rate of about 107 M-1sec-1 (Table 6.1). The α-tocopherolquinone peroxy

radical with the reduction potential of about 1000 mV (Table 2.3) can abstract a hydrogen

from unsaturated lipids to form α-tocopherolquinone hydroperoxide at the rate of

101~102 M-1sec-1 (Table 6.1) and is prooxidant.

Figure 6.6 shows the singlet oxygen formation from the carbon centered α- tocopheryl radical. The carbon centered α-tocopheryl radical reacts with oxygen and forms α-tocopherol peroxy radical. Two moles of α-tocopherol peroxy radical form

dimerized α-tocopherol peroxide and singlet oxygen at the rate of 105 M-1sec-1 (Barclay and others 1989; Min and Boff 2002). The singlet oxygen thus formed is a strong prooxidant (Bradley and Min 1992; Min and Boff 2002).

133 α-Tocopherol peroxy radical, α-tocopherol oxy radical, hydroxy radical, singlet oxygen, and oxidized α-tocopherol compounds as prooxidants

The tocopherol peroxy radical (TOO·), tocopherol oxy radical (TO·), α- tocopherolquinone oxy radical, α-tocopherolquinone peroxy radical, alkoxy radical (RO·),

1 hydroxy radical (·OH) and singlet oxygen ( O2) formed from the oxidation of tocopherol are prooxidants (Table 2.3). Oxidized α-tocopherol compounds have polar hydroxyl and

nonpolar hydrocarbon groups. Yoon and others (1988) and Mistry and others (1988a,

1988b) reported that thermally oxidized lipid compounds with polar hydroxyl and

nonpolar hydrocarbons in the same molecule were prooxidants in soybean oil during

storage. They reported that the oxidized lipids with hydroxyl and/or carbonyl groups

were less soluble in the soybean oil and moved to the surface of oil. The oxidized oils

having polar and nonpolar groups decreased the surface tension between air and oil and

increased the transportation of oxygen from air to oil to accelerate the oxidation of oil

(Yoon and others 1988; Mistry and others 1988a, 1988b). The oxidized α-tocopherol

compounds with the polar groups and nonpolar hydrocarbons in the same molecule

reduce the surface tension between headspace air and oil and accelerate the oxidation of

oil.

The prooxidant mechanisms of oxidized α-tocopherol may be mainly due to α-

tocopherol peroxy radical, α-tocopherol oxy radical, α-tocopherolquinone oxy radical, hydroxy radical and singlet oxygen formed from the oxidation of tocopherols during storage in foods. The oxidized α-tocopherol compounds with polar hydroxyl and

134 nonpolar hydrocarbons in the same molecule may contribute to the oxidation of oil by

reducing the surface tension of oil and increasing the diffusion of oxygen from air to oil.

Possible prooxidant mechanism of tocopherol above the optimum concentration

Tocopherol is antioxidant up to the optimum concentration and is prooxidant above the concentration (Olcott and Van der Veen 1968; Jung and Min 1990; Bowry and

Stocker 1993; Huang and others 1994; Huang and others 1995; Yoshida and others 1994;

Evans and others 2002; Rietjens and others 2002). It could be assumed that the higher the concentration of tocopherol in foods, the higher the amounts of the peroxy radical, oxy radical, hydroxy radical of tocopherol, and singlet oxygen formed from tocopherol during the storage of foods. The antioxidant effect of tocopherol will be greater than the prooxidant effect up to the optimum concentration and then, the prooxidant effect is greater than the antioxidant effect above the optimum concentration.

6.5 Conclusion

The oxidized α-tocopherol compounds acted as prooxidant on the peroxide values and headspace oxygen contents of purified soybean oil. The prooxidant mechanisms of oxidized α-tocopherol may be due to α-tocopherol peroxy radical, α-tocopherol oxy

radical, hydroxy radical, and singlet oxygen formed from the α-tocopherol. The oxidized

α-tocopherol containing polar and nonpolar groups in the same molecule may reduce the

135 surface tension of oil to increase the transfer of oxygen to oil and accelerate the oil oxidation. The prooxidant effect of α-tocopherol at high concentration could be due to high amounts of peroxy radical and oxy radical of α-tocopherol, hydroxyl radical and singlet oxygen formed from tocopherol during storage. The prevention of tocopherol oxidation and removal of oxidized tocopherols could improve the oxidative stability of soybean oil.

136 6.6 References

AOCS. 2004. Official Methods and Recommended Practices of the AOCS. 5th ed. Champaign, IL: AOCS.

Barclay LRC, Baskin KA, Locke SJ, Vinquist MR. 1989. Absolute rate constants for lipid peroxidation and inhibition in model biomembranes. Can J Chem 68:2258-69.

Bradley DG, Min DB. 1992. Singlet oxygen oxidation of foods. Crt Rev Food Sci and Nutri. 31:211-36.

Bowry VW, Stocker R. 1993. Tocopherol-mediated peroxidation. The prooxidant effect of vitamin E on the radical-initiated oxidation of human low-density lipoprotein. J Am Chem Soc 115:6029-44.

Choe E, Min DB. 2005. Chemistry and reactions of reactive oxygen species in foods. J Food Sci 70:R142-59.

Conley RT. 1972. Infrared Spectroscopy. Boston, MA: Allyn and Bacon, Inc.

Evans JC, Kodali DR, Addis PB. 2002. Optimal tocopherol concentrations to inhibit soybean oil oxidation. J Am Oil Chem Soc 79:47-51.

Faustman C, Liebler DC, Burr JA. 1999. α-Tocopherol oxidation in beef and in bovine muscle microsomes. J Agric Food Chem. 47:1396-99.

Hasegawa K, Patterson LK. 1978. Pulse radiolysis in model lipid systems: formation and behavior of peroxyl radicals in fatty acids. Photochem Photobiol 28:817-23.

137 Hiatt R, Mill T, Irwin KC, Mayo TR, Gould CW, Castleman JK. 1968. Homolytic decomposition of hydroperoxides. J Org Chem 33:1416-41.

Huang S-W, Frankel EN, German JB. 1995. Effects of individual tocopherols and tocopherol mixtures on the oxidative stability of corn oil triglycerides. J Agric Food Chem 43:2345-50.

Huang R, Choe E, Min DB. 2004. Effects of riboflavin photosensitized oxidation on the volatile compounds of soymilk. J Food Sci 69:C733-8.

Jung MY, Yoon SH, Min DB. 1989. Effects of processing steps on the contents of minor compounds and oxidation of soybean oil. J Am Oil Chem Soc 66:118-20.

Jung MY, Min DB. 1990. Effects of α-, γ-and δ-tocopherols on the oxidative stability of purified soybean oil. J Food Sci 55:1464-5.

Jung MY, Min DB. 1992. Effects of oxidized α-, γ- and δ-tocopherols on the oxidative stability of purified soybean oil. Food Chem 45:183-7.

Kamal-Eldin A, Appelqvist L-Å. 1996. The chemistry and antioxidant properties of tocopherols and tocotrienols. Lipids 31:671-701.

Kasaikina OT, Kortenska VD, Yanishlieva NV. 1999. Effect of chain transfer and recombination/disproportionation of inhibitor radicals on inhibited oxidation of lipids. Russ Chem Bull 48:1891-6.

Kaouadji MN, Jore D, Ferradini C, Patterson LK. 1987. Radiolytic scanning of vitamin E/vitamin C oxidation reduction mechanisms. Bioelectrochem Bioenerg 18:59-70.

138 Liebler DC, Burr JA, Philips L, Ham AJL. 1996. Gas chromatography-mass spectrometry analysis of vitamin E and its oxidation products. Anal Biochem 236:27-34.

Marquez-Ruiz G, Polvillo M, Jorge N, Ruiz-Mendez M, Dobarganes M. 1999. Influence of used frying oil quality and natural tocopherol content on oxidative stability of fried potatoes. J Am Oil Chem Soc 76:421-5.

Melchert H-U, Pollok D, Pabel E, Rubach K, Stan H-J. 2002. Determination of tocopherols, tocopherolquinones and tocopherolhydroquinones by gas chromatography- mass spectrometry and preseparation with lipophilic gel chromatography. J Chrom A 976:215-20.

Min DB, Boff JM. 2002. Chemistry and reaction of singlet oxygen in foods. Comprehensive Rev Food Sci Food Safety 1:58-72.

Mistry BS, Min DB. 1988a. Prooxidant effects of monoglycerides and diglycerides in soybean oil. J Food Sci 53:1896-7.

Mistry BS, Min DB. 1988b. Isolation and identification of minor components and their effects on flavor stability of soybean oil. In: Frontiers of Flavor. Holland: Elsevier Science p 499-519.

Miyagawa K, Hirai K, Takezoe R. 1991. Tocopherol and fluorescence levels in deep-fat frying oil and their measurement for oil assessment. J Am Oil Chem Soc 68:163-6.

Mukai K, Morimoto H, Okauchi Y, Nagaoka S. 1993. Kinetic study of reactions between tocopheroxyl radicals and fatty acids. Lipids 28:753-6.

139 Naumov VV, Vasil’ev RF. 2003. Antioxidant and prooxidant effects of tocopherol. Kinetics and Catalysis 44:101-5.

Nawar W. 1996. Lipids. In: Fennema OR, editor. Food Chemistry. 3rd ed. New York: Marcel Dekker p 225-319.

Niki E, Satio T, Kawakami A, Kamiya Y. 1984. Inhibition of oxidation of methyl linoleate in solution by vitamin E and vitamin C. J Biol Chem 259:4177-82.

Olcott HS, Van der Veen J. 1968. Comparison of antioxidant activities of tocol and its methyl derivatives. Lipids 3:331-4.

Pazos M, Sanchez L, Medina I. 2005. α-Tocopherol oxidation in fish muscle during chilling and frozen storage. J Agric Food Chem 53:4000-5.

Remorova AA, Roginsky VA.1991. Rate constants for the reaction of α-tocopherol phenoxy radicals with unsaturated fatty acid esters and the contribution of this reaction to the kinetics of inhibition of lipid peroxidation. Kinet Catal 32:726-31.

Rietjens IMCM, Boersma MG, de Haan L, Spenkelink B, Awad HM, Cnubben NHP, van Zanden JJ, van der Woude H, Alink GM, Koeman JH. 2002. The prooxidant chemistry of the natural antioxidants vitamin C, vitamin E, carotenoids and flavonoids. Environmental Toxicology and Pharmacology 11:321-33.

SAS. 2004. Statistical Analysis System Procedure Guide: Cary, NC.

Silverstein RM, Bassler GC, Morril TC. 1981. Spectrometric Identification of Organic Compounds. New York: John Wiley &Sons.

140 Small RD Jr., Scaiano JC, Patterson LK. 1979. Radical processes in lipids. A laser photolysis study of t-butoxy radical reactivity toward fatty acids. Photochem Photobiol 29:49-51.

Verleyen T, Verhe R, Huyghebaert A, Dewettinck K, De Grey W. 2001a. Identification of α-tocopherol oxidation products in triolein at elevated temperature. J Agric Food Chem 49:1508-11.

Verleyen T, Kamal-Eldin A, Dobarganes C, Verhe R, Dewettinck R, Huyghebaert A. 2001b. Modeling of α-tocopherol loss and oxidation products formed during thermoxidation in triolein and tripalmitin mixture. Lipids 36:719-26.

Wagner K-H, Elmadfa I. 2000. Effects of tocopherols and their mixtures on the oxidative stability of olive oil and linseed oil under heating. Eur J Lipid Sci Technol 102: 624-9.

Warner K. 2005. Effects on the flavor and oxidative stability of stripped soybean and sunflower oils with added pure tocopherols. J Agric Food Chem 53: 9906-10.

Yoon SH, Jung MY, Min DB. 1988. Effects of thermally oxidized triglycerides on the oxidative stability of soybean oil. J Am Oil Chem Soc 65:1652-6.

Yoshida H, Kajimoto G, Emura S. 1994. Antioxidant effects of δ-tocopherols at different concentrations in oils during microwave heating. J Am Oil Chem Soc 70:989-95.

141

Reaction k (M-1sec-1) Reference

3 × 108 Hasegawa and Patterson (1978) R· + 3O → ROO· 2 5 × 106 Kasaikina and others (1999) RO· + RH → R· + ROH 1 × 107 Small and others (1979) Niki and others (1984) ROO· + RH → R· + ROOH 10 ~ 100 Naumov and Vasil’ev (2003) R· + ROOH → ROO· + RH 1 × 105 Naumov and Vasil’ev (2003) 0.07 Remorova and Roginsky (1991) T· + RH → T + R· 0.5 Naumov and Vasil’ev (2003) T· + ROOH → T + ROO· 0.1~0.5 Mukai and others (1993) T + ROO· → T· + ROOH 1 × 106 Niki and others (1984) 1 2TOO· → TOOT + O2 - - 1 5 (2ROO· → ROOR + O2) 1 ×10 Barclay and others (1989) TOOT → 2TO· - - TO· + RH → TOH + R· - - ROO· + T· → ROOT 2.5 × 106 Kaouadji and others (1987) R· + TOO· → ROOT - - ROOT → RO· + TO· - - TOO· + T → TOOH + T· - Bowry and Stocker (1993) TOOH → TO· + ·OH - -

Table 6.1. Reaction rates of the lipid (RH) and tocopherol (T)

142

120 0 ppm 250 ppm 100 500 ppm 1000 ppm 1500 ppm

40 Peroxide Value (meq/kg)

0 0123456 Storage (Day)

Figure 6.1. Effects of 0, 250, 500, 1000 and 1500 ppm of oxidized α-tocopherol on the

peroxide value of purified soybean oil during dark storage of 6 days at 55 °C

143

25 0 ppm 250 ppm 500 ppm 20 1000 ppm 1500 ppm

10 Headspace Oxygen (%)

0 0123456 Storage (Day)

Figure 6.2. Effects of 0, 250, 500, 1000 and 1500 ppm of oxidized α-tocopherol on the

headspace oxygen content (%) of purified soybean oil during dark storage of 6 days at 55

°C

144 (A) 4a (B) (C) · O 5 4 O O 6 3 · 7 1 2 8 · 8a O O O

O2 O2 OO O O O · O · O O · OO 3 O2 RH RH OOH O O O

O O O O O · HOO − · OH − · OH RH O · O O O O O O O · O O H α-tocopherol hydroperoxide O O O

− · O · OH · O O O O 2 O2 O O O O O· O O O O O O O · O · RH RH O O O O O RH· O O O OO O O O H H − OH − OH O · · OH O O O α -tocopherolquinone O O O O · O · RH RH O O O O O OH OH O

4a, 5-epoxy-α-tocopherolquinone 7, 8-epoxy-α-tocopherolquinone

Figure 6.3. Suggested possible formation of α-tocopherolquinone (A), 4a, 5-epoxy-α- tocopherolquinone (B), and 7, 8-epoxy-α-tocopherolquinone (C) from α-tocopheryl radical with triplet oxygen

145 (A) 4a (B) · O 5 4 O (C) O 6 3 · 7 1 2 8 · 8a O O O ROO· ROO· O OOR O O · O O O ROO · ROO − RO − RO · O · O O · O O O O O R O · O α-tocopherol peroxide O O O − RO · · O · O O O ROO O O · ROO· · O O O O O O O O O O O O R R O · − RO · − RO · RH O O O O O O O O O O OH · ·

α-tocopherolquinone O O O O O O O · O · RH RH O O O O O OH OH O

4a, 5-epoxy-α-tocopherolquinone 7, 8-epoxy-α-tocopherolquinone

Figure 6.4. Suggested possible formation of α-tocopherolquinone (A), 4a, 5-epoxy-α- tocopherolquinone (B), and 7, 8-epoxy-α-tocopherolquinone (C) from α-tocopheryl radical with acyl peroxy radical

146

O Carbon centered · O α-tocopheryl radical

Carbon centered O · α-tocopherolquinone radical

O α-tocopherolquinone peroxy radical O O

O·

O α-tocopherolquinone hydroperoxide O O O H

Figure 6.5. Suggested possible formation of α-tocopherolquinone hydroperoxide from α- tocopheryl radical with acyl peroxy radical

147

O Carbon centered · α O -tocopheryl radical

O α-tocopherol peroxy radical O · OO

Dimerization

O O 1 + O2 O O O O

Dimerized α-tocopherol peroxide

O O or · O O O O · α-tocopheryl oxy radical 7, 8-epoxy-α-tocopheryl oxy radical

Figure 6.6. Suggested possible formation of dimerized α-tocopherol peroxide, α- tocopheryl oxy radical, and 7, 8-epoxy-α-tocopheryl oxy radical from α-tocopheryl radical

148

CONCLUSION

1. 2,3-Butanedione formed from the riboflavin solution under light was positively

identified by the combination of gas chromatographic retention time, mass spectrum,

and odor evaluation.

2. Riboflavin produced singlet oxygen in the presence of triplet oxygen under light.

3. The 2,3-butanedione was produced from the reaction between riboflavin and singlet

oxygen.

4. Sodium azide quenched singlet oxygen, prevented the formation of 2,3-butanedione,

and minimized the destruction of riboflavin under light.

5. α-, β-, γ-, and δ-Tocotrienol at 100, 200, 300, 500, and 1000 ppm minimized the

oxidative stability of lard during the storage of 7 days in dark at 55°C.

6. As the concentration of α- or β-tocotrienol increased from 100 to 200, 300, 500, and

1000 ppm, the antioxidative activities of α- or β-tocotrienol decreased.

149 7. There were no significant differences on the antioxidative effects of γ- or δ-

tocotrienol between the concentrations of 100, 200, 300, 500, and 1000 ppm (p>0.05).

8. The optimum concentration of α-, β-, γ-, or δ-tocotrienol to minimize the oxidation of

lard could be 100 ppm.

9. The antioxidative activities of tocotrienol in lard were δ- > γ- > β- > α-tocotrienol.

10. The selection of optimum concentration and type of tocotrienol not only can

minimize the oxidation of lipids but also be economically important.

11. α-, β-, γ-, and δ-Tocotrienols had significant antioxidative effects on the chlorophyll

photosensitized oxidation of lard by quenching singlet oxygen at p<0.05.

12. Chlorophyll b in methylene chloride at 3,000 Lux produced singlet oxygen at 1.089

µmole oxygen/mL headspace/hr.

13. The reaction rate constant of singlet oxygen with lard was 6.5 × 104 M-1sec-1.

14. The singlet oxygen quenching rates of α-, β-, γ-, and δ-tocotrienol were 2.16 × 107,

1.99 × 107, 2.05 × 107, and 0.80 × 107 M-1sec-1, respectively.

150 15. α-, β-, γ-, and δ-Tocotrienol could be used as good singlet oxygen quenchers in foods.

16. The oxidized α-tocopherol compounds formed during the oxidation of α-tocopherol

acted as prooxidants in purified soybean oil (p<0.05).

17. As the concentrations of oxidized α-tocopherol increased from 0 ppm to 250, 500,

1000, and 1500 ppm, the oxidative stability of purified soybean oil decreased during

the 6 days of storage.

18. The prooxidant mechanisms of oxidized α-tocopherol may be due to the

intermediated compounds such as α-tocopherol peroxy radical, α-tocopherol oxy

radical, hydroxy radical, and singlet oxygen which are formed during the oxidation of

α-tocopherol.

19. The oxidized α-tocopherol contains polar and nonpolar groups in the same molecule

which may reduce the surface tension of oil to increase the transfer of headspace

oxygen to oil and accelerate the oil oxidation.

20. It is important to prevent the oxidation of tocopherol and remove the oxidized

tocopherol in order to improve the oxidative stability of foods.

151

BIBLIOGRAPHY

Andersen ML, Velasco J, Skipsted LH. 2005. Analysis of lipid oxidation by ESR spectroscopy. In: Kamel-Eldin A, Pokorny J, editors. Analysis of lipid oxidation. Champaign, IL: AOCS Press.

Azzi A, Stocker A. 2000. Vitamin E: non-antioxidant roles. Prog Lipid Res 39:231-55.

Baron CP, Andersen H. 2002. Myoglobin-induced lipid oxidation. A review. J Agric Food Chem 50:3887-97.

Bauernfeind J. 1980. Tocopherols in foods. In: Machlin LJ, editors. Vitamin E: A Comprehensive Treatise. New York: Marcel-Dekker. p 99-167.

Belitz H-D, Grosch W. 1999. Food Chemistry, 2nd edition. Berlin: Springer. pp 207-8, 381-2, 606.

Bendich A, Machlin LJ. 1988. Safety of oral intake of vitamin E. Am J Clin Nutri 48:612-9.

Berkowitz J, Ellison GB, Gutman D. 1994. Three methods to measure Rh bond-energies. J Phys Chem 98:2744-65.

Bisby RH, Morgan CG, Hamblett I, Gorman AA. 1999. Quenching of singlet oxygen by Trolox C, ascorbate and amino acids: effects of pH and temperature. J Phys Chem 103:7454-9.

Bordwell FG, Liu WZ. 1996. Solvent effects on hemolytic bond dissociation energies of hydroxylic acids. J Am Chem Soc 118:10819-23.

Bowry VW, Stocker R. 1993. Tocopherol-mediated peroxidation: The prooxidant effect of vitamin E on the radical-initiated oxidation of human low-density lipoprotein. J Am Chem Soc 115:6029-44.

Bradley DG, Lee HO, Min DB. 2003. Singlet oxygen detection in skim milk by electron spin resonance spectroscopy. J Food Sci 68:491-4.

152

Bradley DG, Min DB. 1992. Singlet oxygen oxidation in foods. Crit Rev Food Sci Nutri 31:211-36.

Bramley PM, Elmadfa I, Kafatos A, Kelly FG, Manios Y, Roxborough HE, Schuch W, Sheehy PJA, Wagner K-H. 2000. Review: Vitamin E. J Sci Food Agric 80:913-38.

Buettner G. 1993. The pecking order of free radicals and antioxidants: lipid peroxidation, α-tocopherol, and ascorbate. Arch Biochem Biophys 300:535-43.

Burton GH, Ingold KU. 1981. Autoxidation of biological molecules. 1. The antioxidant activity of vitamin E and related chain breaking phenolic antioxidants in vitro. J Am Chem Soc 103:6472-7.

Burton GW, Traber MG, Acuff RV, Walters DN. 1998. Human plasma and tissue α- tocopherol concentrations in response to supplementation with deuterated natural and synthetic vitamin E. Am J Clin Nutri 67:669-84.

Choe E, Huang R, Min DB. 2005. Chemical reactions and stability of riboflavin in foods. J Food Sci 70:R28-36.

Choe E, Min DB. 2005. Chemistry and reactions of reactive oxygen species in foods. J Food Sci 70: R142-59.

Choe E, Min DB. 2006. Mechanisms and factors for edible oil oxidation. Comp Rev Food Sci Food Saf 5:169-86.

Conte C, Floridi A, Aisa C, Piroddi M, A. F, Galli F. 2004. Gamma-tocotrienol metabolism and antiproliferative effect in prostate cancer cells. Annals New York Academy Sci 1031:391-4.

Doleiden FH, Farenholtz SR, Lamola AA, Trozzolo AM. 1974. Reactivity of cholesterol and some fatty acids toward singlet oxygen. Photochem Photobiol 20:519-21.

Drotleff AM, Ternes W. 1999. Cis/trans isomers of tocotrienols - occurrence and bioavailability. Euro Food Res Tech 210:1-8.

Evans JC, Kodali DR, Addis PB. 2002. Optimal tocopherol concentrations to inhibit soybean oil oxidation. J Am Oil Chem Soc 79:47-51.

Faustman C, Liebler DC, Burr JA. 1999. α-Tocopherol oxidation in beef and in bovine muscle microsomes. J Agric Food Chem 47:1396-9.

153 Feng HP. 1996. Preparative techniques for isolation of vitamin E homologs and evaluation of their antioxidant activities. Dissertation-Abstracts-International, -B 56:207.

Foote CS. 1979. Quenching of singlet oxygen. In: Wasserman HH, Murray RW editors. Singlet Oxygen. New York: Academic Press. p 139-73.

Foote CS, Denny RW. 1968. Chemistry of singlet oxygen. VII Quenching by β-carotene. J Am Chem Soc 90:6233-5.

Frankel EN. 1984. Chemistry of free radical and singlet oxidation of lipids. Prog Lipid Res 23:197-321.

Girotti AW. 1998. Lipid hydroperoxide generation, turnover, and effector action in biological systems. J Lipid Res 39:1529-42.

Gregory JF. 1996. Vitamins. In: Fennema OR, editor. Food Chemistry. 3rd edition. New York: Marcel-Dekker. p 553-7.

Halliwell B, Gutteridge JMC. 2001. Free radicals in biology and medicine, 3rd edition. New York: Oxford University Press.

Hicks M, Gebicki JM. 1981. Inhibition of peroxidation in linoleic acid membranes by nitroxide radicals, butylated hydroxy toluene and α-tocopherol. Arch Biochem Biophys 210:56-63.

Huang R, Choe E, Min DB. 2004a. Effects of riboflavin photosensitized oxidation on the volatile compounds of soymilk. J Food Sci 69:C733-8.

Huang R, Choe E, Min DB. 2004b. Kinetics for singlet oxygen formation by riboflavin photosensitization and the reaction between riboflavin and singlet oxygen. J Food Sci 69:C726-32.

Huang R, Kim HJ, Min DB. 2006. Photosensitizing effect of riboflavin, lumiflavin, and lumichrome on the generation of volatiles in soy milk. J Agric Food Chem 54:2359-64.

Huang S-W, Frankel EN, German JB. 1994. Antioxidant activity of α- and γ-tocopherols in bulk oils and in oil-in-water emulsions. J Agric and Food Chem 42:2108-14.

154 Institute of Medicine. 1998. Dietary reference intakes for thiamin, riboflavin, niacin, vitamin B6, folate, vitamin B12, pantothenic acid, biotin, and choline- a report of the standing committee on the scientific evaluation of dietary reference intakes and its panel on folate, other B vitamins, and choline and subcommittee on upper reference levels of nutrients. Washington D.C.: National Academy Press.

Institute of Medicine. 2000. Food and nutrition board. Dietary reference intakes: vitamin C, vitamin E, selenium, and carotenoids. Washington D.C.: National Academy Press

Jung MY, Choe E, Min DB. 1991. Effects of α-, γ-, and δ-tocopherols on the chlorophyll photosensitized oxidation of soybean oil. J Food Sci 56:807-15.

Jung MY, Lee HO, Min DB. 1998. Singlet oxygen and ascorbic acid effects on dimethyl disulfide and off-flavor in skim milk exposed to light. J Food Sci 63:408-12.

Jung MY, Min DB. 1990. Effects of α-, γ-, and δ-tocopherols on the oxidative stability of soybean oil. J Food Sci 55:1464-5.

Jung MY, Min DB. 1991. Effects of quenching mechanisms of carotenoids on the photosensitized oxidation of soybean oil. J Am Oil Chem Soc 68:653-7.

Jung MY, Min DB. 1992. Effects of oxidized α-, γ-, and δ-tocopherols on the oxidative stability of purified soybean oil. Food Chem 43:183-7.

Jung MY, Yoon SH, Min DB. 1989. Effects of processing steps on the contents of minor compounds and oxidation of soybean oil. J Am Oil Chem Soc 66:118-20.

Kamal-Eldin A, Appelqvist L-A. 1996. The chemistry and antioxidant properties of tocopherols and tocotrienols. Lipids 31:671-701.

Ke PJ, Ackman RG. 1973. Bunsen coefficient for oxygen in marine oils at various temperatures determined by exponential dilution method with a polarographic oxygen electrode. J Am Oil Chem Soc 50:429-35.

Kepka A, Grossweiner LI. 1972. Photodynamic oxidation of iodide ion and aromatic amino acids by eosine. Photochem Photobiol 14:621-39.

King JM, Min DB. 1998. Riboflavin photosensitized singlet oxygen oxidation of Vitamin D. J Food Sci 63: 31-4.

155 Kitts DD. 1997. An evaluation of the multiple effects of the antioxidant vitamins. Trends Food Sci Tech 8:198-203.

Kochevar IE, Redmond RW. 2000. Photosensitized production of singlet oxygen. Met Enzymol 319:20-8.

Korycka-Dahl MB, Richardson T. 1978. Activated oxygen species and oxidation of food constituents. Crit Rev Food Sci Nutri 10:209-40.

Lee YC, Oh SW, Chang J, Kim IH. 2004. Chemical composition and oxidative stability of safflower oil prepared from safflower seed roasted with different temperature. Food Chem 84:1-6.

Lehmann J, Slover HT. 1976. Relative autoxidative and photolytic stabilities of tocols and tocotrienols. Lipids 11:853-7.

Li TL, Min DB. 1998. Stability and photosensitized singlet oxygen oxidation of vitamin D. J Food Sci 63:413-7.

Liebler DC, Burr JA, Philips L, Han AJL. 1996. Gas chromatography-mass spectrometry analysis of vitamin E and its oxidation products. Anal Biochem 236:27-34.

Long CA, Kearns DR. 1975. Radiationless decay of singlet molecular oxygen in solution. II. Temperature dependence and solvent effects. J Am Chem Soc 97:2018-20.

McGinnis BD, Adams VD, Middlebrooks EJ. 1999. Evaluation of methylene blue and riboflavin for the photosensitized degradation of ethylene glycol. Environ Int 25:953-9.

Min DB, Boff JM. 2002a. Chemistry and reaction of singlet oxygen in foods. Comp Rev Food Sci Food Saf 1:58-72.

Min DB, Boff JM. 2002b. Lipid oxidation of edible oil. In: Akoh CC, Min DB editors. Food Lipids. 2nd edition. New York: Marcel-Dekker. p 335-63.

Min DB, Callison AL, Lee HO. 2003. Singlet oxygen oxidation for 2-pentylfuran and 2- pentenylfuran formation in soybean oil. J Food Sci 68:1175-8.

Mukai K, Morimoto H, Okauchi Y, Nagaoka S. 1993. Kinetic study of reactions between tocopheroxyl radicals and fatty acids. Lipids 28:753-6.

Naumov VV, Vasil’ev RF. 2003. Antioxidant and prooxidant effects of tocopherol. Kinet Cat 44:101-5.

156 Nawar W. 1996. Lipids. In: Fennema OR, editor. Food Chemistry. 3rd edition. New York: Marcel-Dekker. p 225-319.

Newaz MA, Nawal NN. 1999. Effect of γ-tocotrienol on blood pressure, lipid peroxidation and total antioxidant status in spontaneously hypertensive rats. Clin Exp Hypertension 21:1297-313.

Niki E, Satio T, Kawakami A, Y. K. 1984. Inhibition of oxidation of methyl linoleate in solution by vitamin E and vitamin C. J Biol Chem 259:4177-82.

Nourooz-Zadeh J. 1998. Measurement of hydroperoxides in edible oils and fats using the ferrous-oxidation in xylenol orange (FOX) assay and application of the method to estimation of dietary hydroperoxide intakes in the United Kingdom. Rec Res Develop Oil Chem 2:53-61.

Pazos M, Sanchez L, Medina I. 2005. α-Tocopherol oxidation in fish muscle during chilling and frozen storage. J Agric Food Chem 53:4000-5.

Ping BTY, May CY. 2000. Valuable phytonutrients in commercial red palm olein. Palm Oil Developments 32:20-5.

Player ME, Kim HJ, Lee HO, Min DB. 2006. Stability of α-, γ-, or δ-tocopherol during soybean oil oxidation. J Food Sci 71:C456-C60.

Rahmani M, Saari Csallany A. 1998. Role of minor constituents in the photooxidation of virgin olive oil. J Am Oil Chem Soc 75:837-43.

Rawls HR, VanSanten PJ. 1970. A possible role for singlet oxidation in the initiation of fatty acid autoxidation. J Am Oil Chem Soc 47:121-5.

Reische DW, Lillard DA, Eitenmiller RR. 2002. Antioxidants. In: Akoh CC, Min DB editors. Food Lipids. 2nd edition. New York: Marcel-Dekker. p 489-516.

Schneider C. 2005. Chemistry and biology of vitamin E. Mol Nutri Food Res 49:7-30.

Schroeder MT, Becker EM, Skibsted LH. 2006. Molecular mechanism of antioxidant synergism of tocotrienols and carotenoids in palm oil. J Agric Food Chem 54:3445-53.

157 Serbinova E, Kagan V, Han D, Packer L. 1991. Free radical recycling and intra- membrane mobility in the antioxidative properties of α-tocopherol and α-tocotrienol. Free Radical Biol Med 10:263-75.

Siassi F, Ghadirian P. 2005. Riboflavin deficiency and esophageal cancer: A case control-household study in the Caspian Literal of Iran. Cancer Detect Prev 29:464-9.

Suzuki YJ, Packer L. 1993. Inhibition of NF-kappa B activation by vitamin E derivatives. Biochem Biophys Res Comm 193:277-83.

Terao J, Matsushita S. 1986. The peroxidizing effect of α-tocopherol on autoxidation of methyl linoleate in bulk phase. Lipids 21:255-60.

Theriault A, Wang Q, Gapor A, Adeli K. 1999. Effects of γ-tocotrienol on ApoB synthesis, degradation, and secretion in HepG2 cells. Arteriosclerosis, Thrombosis, and Vascular Biology 18:704-12.

Verleyen T, Kamal-Eldin A, Dobarganes C, Verhe R, Dewettinck R, Huyghebaert A. 2001a. Modeling of α-tocopherol loss and oxidation products formed during thermoxidation in triolein and tripalmitin mixture. Lipids 36:719-26.

Verleyen T, Verhe R, Huyghebaert A, Dewettinck K, De Grey W. 2001b. Identification of α-tocopherol oxidation products in triolein at elevated temperature. J Agric Food Chem 49:1508-11.

Vever-Bizet C, Dellinger M, Brault D, Rougee M, Bensasson RV. 1989. Singlet molecular oxygen quenching by saturated and unsaturated fatty-acids and by cholesterol. Photochem Photobiol 50:321-5.

Wagner K-H, Wotruba F, Elmadfa I. 2001. Antioxidative potential of tocotrienols and tocopherols in coconut fat at different oxidation temperatures. Euro J Lipid Sci Tech 103:746-51.

Watkins T, Lenz P, Gapor A, Struck M, Tomeo A, Bierenbaum M. 1993. Gamma- tocotrienol as a hypocholesterolemic and antioxidant agent in rats fed atherogenic diets. Lipids 28:1113-8.

Wennemark B, Jaegerstad M. 1992. Breadmaking and storage of various wheat fractions affect vitamin E. J Food Sci 57: 1205-9.

Whang K, Peng IC. 1988. Electron paramagnetic resonance studies of the effectiveness of myoglobin and its derivatives as photosensitizers in singlet oxygen generation J Food Sci 53:1863-5.

158 Wright JS, Johnson ER, DiLabio GA. 2001. Predicting the activity of phenolic antioxidants: theoretical method, analysis of substituent effects, and application to major families of antioxidants. J Am Chem Soc 123:1173-83.

Yoshida Y, Niki E, Noguchi N. 2003. Comparative study on the action of tocopherols and tocotrienols as antioxidant: chemical and physical effects. Chem Phys Lipids 123:63- 75.

159