University of of of of of

CHARACTERIZATION OF NATURAL ANTIOXIDANTS OF MEALS OF BORAGE AND EVENING PRIMROSE

BY

CMAHINDA WETTASING HE, B.Sc., M.St.

A thesis submined to the School of Graduate Studiesin panial fulfilment of the requirements for thedegree of the

Doctor of Philosophy

Depanment of Biochemistry MemorialUniversity of Newfoundland

March, 1999

St. John's Newfoundland Canada

THIS WORK IS DEDICATEDTO MY LOVING PARENTS,WIFE&SON

ABSTRACT

Antioxidant efficacy of borageandevening primrose meals in a cooked comminuted pork model systemwasinvestigated. Borage meal. at 2%(w/w).reduced the formation of !hiobarbituric acid·reactive substances (TBARS). hexanal and tOlal

\'olatiles in treated samples by 21. 30 and)1)01..respectively. Formation of TBARS.

hexanal and IOlal volatiles in samples containing 20/.(w/w)evening primrose mealwas

r~ucedon day-' oftheSlorageby44. 73and63%. respectively. Since meals demonstrated antioxidantpro~rties..their crude extracts were prepared under optimum extraction conditions v.ilich were determined by employing response surface methodology (RSM). Effects ofthreevariables. namely the solvent content in the aqueous extraction medium (x,. %.v/v).extraction temperature(x~.0c)and time(Xl'min) on antioxidant efficacy(Y)oftheextracts were investigated. RSMpredictedthat the maximum antioxidant activity of borage extractwasreachedwhenXl>x! and x, were 52% ethanol (v/v). 72"Cand62 min. respectively.Thecorresponding optimum extraction conditions for evening primrose were pmlicted tobe56% acetone (v/v). 71"Cand47 min.

re~tively.Finedpolynomial models for borageandevening primroseweresignificant (p s; 0.05) and reproduc:ible (CV<5%). Verification experiments carried out to determine the adequacy ofthemodels showedthat the predicted response values were well in agreementwiththeobserved values (r>0.95). Crude extractspreparedunder

optimum extraction conditions. were subjected to Sephadex LH-20 column chromatography. For both types of extracts. six fractions (I-VI) were obtained and their contents of total. hydrophobicand hydrophillic phenolics determined. Borage fractions I-VI consisted of 283. 129. 140. 366. 280and 341 mg of phenolics as sinapic acid equivalents. respectively. while evening primrose fractions t.vl cOnlainN 158. ] 13. 369.

402. 279 and 445 mg of phenolics as catechin equivalents. respectively. Borage fractions contained more of hydropbillicthan hydrophobic phenolics whereas evening primrose fractions contained high amounts of bothtypesof phenolics. A qualitative vanillin test wasemployed to determinethe presence or absence of condensed lalUtins in borageand evening primrose fractions. Borage fractions did not contain condensed tannins as evidenced by a negative vanillin lest, but fractions III-VI of evening prirnrost did.

Ultraviolet(UV)spectra of borage fractions indicated possible presence of phenolic acids while those for evening primrose ftactions suggested possible presence of procyanidins.

Antioxidant efficacies of borage and evening primrosecrudeextracts and their fractions (additives) were investigated in jl-carotene-linoleate. cooked comminuted pork.

bulk stripped com oilandstripped com oil-in-watet emulsion systems. In general. all additives exhibited varying antioxidant activities in all fourtypesof modelsySlmlS investigated. After a 2 h assayperiod, borageandevening primrose additiveswereable to retain 27-79% and 37-84% ofinitiaJcontent of !Harotene, respectively, as compared toI()O/,retentioninthe control. Incookedcomminutedporkmodel systems, borageand

iii

e\'ening primrose additives inhibiu:dthefonnation ofTBARS.hexanaland total \'olatiles (onday~3of storage) to varyingdC'grces(19-97%) as compared tothe: control. On day-3 of storage. 200 ppm (as sinspic acid equivalents) of borage addith'es inhibited the formation of conjugated dienes. hexanal and total volatiles in bulk stripped-com oil and itsoil-in~wateremulsions by21~95%as compared to the control. Inhibition of fonnation of oxidation products in samples treated with 200 ppm (as catechin equivalents) of evening primrose additives. on day-). ranged from 17 to 94% as compared tothe: control.

In general. borage additives'A'ttC'bener antioxidants in bulk stripped com oil systems than their emulsion coWlterpam. Evening primrose additives perfonned well in almost all systems examined. These effects were amibuted tothedifferent affinities of active compounds to various phases and interfaces of the model systems. In an attempt to investigate the antioxidant mechanisms of additives. iron (II) chelating. reactive-oxygen species (ROS) scavengingand2.2-diphenyl-l-picrylhydrazyl (DPPH) radical scavenging activities of borageandevening primrose additives were also determined. Iron (II) binding capacities of borageandevening primrose additives ranged from )) 1010001••

depending uponthetypeandconcentration or additives investigated. In general, borage and evening primrose additives exerted strong ROS scavenging properties of 100% in most cases. Borageandevening primrose additives also exhibited strong DPPH radical scavenging aclivities. Anattemptwasmade 10 correlate iron (II) chelating, ROS

;v

and DPPH sca\'enging activities ofmeadditives to the antioxidant activities brought about by respective additives in model systems: correlations ranged from weak (r<0.6) to very slrong (r>0.9). depending mainly upon the nalure of the model system involved. as delennined by linear regression analysis.

For the first time. major phenolic antioxidants present in the borage and evening primrose extracts were identified. Presence of rosmarinic, syringic and sinapic acids in borage extractwasconfirmed by chromatographic as well as UV. mass and nuclear magnetic resonance (NMR) spectroscopic data. Instrumental analysis of isolated evening primrose phenolics allowed identification of(+)catechin. (·)epicatechin and gaJlic acid as the major active compounds responsible for antioxidant activity of its extracts.

ACKNOWLEDGEMENTS

My sincc:rc: gratitude is conveyed toDr.F. Shahidi for providing me with financial suppon andtheinvaluable guidance throughout my programme. I alsowish to thank my supervisory comminee members.Dr.A.M.Martin andDr. J. Banoub. for guidance throughout my programme. Many thanks to Dr. R. Amarov.;cz for helping mewith chromatographic techniques. [would also like to extend my appreciation10Drs. C.

Jablonski. B. Gregory and L.K. Thompson for allowing me to use the spectrometers.

Thanks to my colleagues in Dr. Shahidi's research team for creating a pleasant environment10work.

vi

TABLE OF em'IITENTS

ABSTRACT . ACKNOWLEDGEMENTS TABLE OF CONTENTS . LIST OF FIGURES LIST OF TABLES LIST OF A88REVIATlONS .

.. ii vi . ... vii xiii

xx-xii CHAPTER I

CHAPTER1

INTRODUCTION.

LITERATURE REVIEW .

.

...

^\

. . . 6 2.1 Mechanism of lipid oxidation: initiationandpropagation 7

2.\.\ Factorsthaiaffect lipid oxidation 2.1.1.1 Role of oxygenandoxygen species . 2.1.1.2 Role of transition metals 2.1.1.3 Role of ionizing radiation

2.1.1.4 Role of ultraviolet(UV)radiation and visible light 2.1.1.5 Role ofmz:ymes .

II ..\\

... 16 17 . . . 18 .. ... 20 2.2 Mechanism of lipid oxidation: hydroperoxide decomposition

andtermination of chain reactions . 21

2.3 Effects of lipid oxidation on food quality . . . .24 2.4 link betv.·eeT1 oxidation products in human disease

conditions . . . 3 I

2.5 Role of reactive-oxygen species (ROS) in development of

human disease conditions. . 35

2.5.1 Role of antioxidant enzymesinprevention of ROS injury 39 2.5.2 Role of dietary antioxidantsinprevention of ROS injury 39

vii

2.6 Control of lipid oxidation in foods . 40 2.6.1 Synthetic amioxidants .

2.6.2 Natural antioxidants

.... 41 . ... 44

2.6.2.1 Sources of natural antioxidants . 45

2.6.2.1.1 Spices

2.6.2.1.2 Oilseeds . . . • . . . . . 2.6.2.1.3 Cer~als..

2.6.2.1.4 Beverages andh~rbs

. 45 . . . 50 .55 .58

2.6.2.2 Extraction techniques . .. 60

2.6.6.2.1 Polynomial models and response surface methodology (RSM)

2.6.3 Evaluation of antioxidants

.. 66 .... 67

. 67

71 76 2.6.3.1 Meat model systems.

2.6.3.2 Bulk oil and emulsion systems.

2.6.3.3 Free radical- and hydrogen peroxide-scavenging assays . 2.6.4 Identification of active compounds

CHAPTER3 MATERIALS AND METHODS

3.1 Materials.

3.2 Methods

... 78 ... 79 . .. 79 ... 80 ... 80 3.2.1 Preparation of borage and evening primrose meals

3.2.2 Assessment of antioxidant activity of borage and evening

primrose meals. . . .80

3.2.3 Preparation ofthe crude exttacts . 81

3.2.4 Determination ofthe content of total phenolics 81

3.2.5 Response surface methodology (RSM) . . 82

3.2.6 Detennination of the content of hydrophillic and

hydrophobic phenolics. . 85

3.2.7 Qualitative detection of vanillin positive compounds in borageandevening primrose crude extracts . . . . 86

viii

3.2.8 Evaluation of antioxidant activity ofthecrude eXlraCt5 ... 86 3.2.8.1 p-<:arotene.linoleate model system 86

3.2.8.2 Cooked comminuted pork model system 87

3.2.8.2.1 Determination or 1- thiobarbituric acid reactive

substances(TBARS) . . . .. 88

3.2.8.2.2 Static headspace gas

chromatographic analysis . . 88

3.2.8.2.3 Detcnnination or moisture

content orpork . . . .89 3.2.8.2.4 Detennination of crude

protein content of pork . . . • . . . .90 3.2.8.2.5 Determination of total lipid

content of pork . . . • . . . . ...91 3.2.8.2.6 Determination of ash content

of pork. . ...91

3.2.8.2.7 Analysis of fany acid

composition of lipids . . . • • • • . . . 92 3.2.8.3 Bulkstrippedcomoil model system

3.2.8.3.1 Detcnnination of conjugated dienes (CD) . 3.2.8.3.2 Determination of fatty acid

composition or stripped com oil

3.2.8.4 Stripped com oil-in-water emulsion systems.

3.2.9 Evaluation of iron (II) chclating activities of borageandevening primrosecrudeextracts 3.2.10 Evaluation of reactive-oxygenspctics (ROS) and DPPHfreeradical.scavenging efficacies ofthc crude extract.

3.2.10.1 Hydrogen peroxide-scavenging assay.

3.2.10.2 Hydroxyl radical-scavenging assay ix

. 93

.94

. . . 94 .... 94

... 9S

.96 . ... 96 . 97

3.2.10.3 Superoxide radical-scavenging assay 3.2.10.4 DPPH frtt radical-scavenging assay

.... 98 . . . 98 3.2.11 Column chromatographic fractionation of

crude extracts . .99

3.2.11.1 Evaluation of antioxidant activity of column chromatographic fractions

3.2.11.2 Evaluation of iron (II) chelating and reactive-oxygen species as well as DPPH free radical-scavenging activities of fractions

100

100 3.2.12 Thin-layer chromatography<TLC) . . . • . 100 3.2.13 High performance liquid chromatographic

(HPlC) analysis of active compounds. 102

3.2.14 Ultraviolet(UV)spet:tro.scopy of purified

compounds . . . • • • . 102

3.2.15 Mass~troscopyofputifiedcompounds. 103

3.2.16 Proton

eHl

and carbon (uC) nuclear

magnetic resonance(NMR)sp«troscopy 103

3.2.17 Tukey's studentized range test 103

CHAPTER 4 RESULTS AND DISCUSSION 104

4.1 Antioxidant activity of meals of borage and evening

primrose. 104

4.2 locating an appropriate experimental region forRSM 113

4.3 Experimentaldesignfor RSM 118

4.4 location ofcritica1(optimal)extraction conditions 121 4.5 Sephadex LH-20 column chromatography of crude extracts

of borageand evening primrose meals . 134

4.6 Oxidative stability of modelsystems as affected by borage and evening primrose crude extraelSandtheir fractions 143 4.6.1 Antioxidant efficacy of borageand evening primrose crude

extractsandtheir fractionsina Jkarotene-linoliate model

system 144

4.6.2 Antioxidant efficacy of borageandevening primrose crude extractsandtheir fractionsina cooked comminuted pork

modelsystem . 148

4.6.3 Antioxidant efficacy of borage and evening primrose crude extractsand their fractions in a bulk stripped com oil model

system 155

4.6.4 Antioxidanl efficacy of borage and evening primrose crude extracts and their fractions ina stripped com oil·in-water

emulsion system . 163

4.1 Iron(II)chelating capacity of borage and evenin" ycimrose

crude extractsandtheir fractions .. 111

4.7.1 Relationship between iron (II) chelating capacityand antioxidant activity of borage and evening primrose

additives in model systems. 173

4.8 Hydrogen peroxide(Hl02}-scavenging capacity of borage ande"'ening primrose crude extractsandtheir fractions 182 4.8.1 Relationship between hydrogen peroxide-scavenging

capacityand antioxidant activity of borage and evening

primrose additives in model systems . 187

4.9 Free radical-scavenging capacity of borageandevening primrose crude extractsandtheir fractions . 197 4.9.1 Relationships between free radical-

sc.avenging capacityandantioxidant activity of borageandevening primroseadditi~in

model systems . . . . .. 205

4.10 Elucidation of chemical structures of active compounds of

borageand evening primrose extracts .231

4.10.1 Thin-layer chromatography(TLC)of borage fractions and high performance liquid

,hromalOpaphyofTLCspots . . . .231

4.10.2 Thin--Iayer chromatography (TLC) of evening primrose fractions and high performanceliquid chromatography of TLC

SJlOts. . ... 234

4.11 Instrumental analysis of activecomponents of borageand

evening primrose extracts . . . . .. 237

4.11.1 4.11.2 4.11.3

St:ruetutaI analysis of compound A Structuralanalysis ofcompound8 Structuralanalysis of compound C

xi

. . . 231 . . . . 242 ....•••••••••.... 245

4.11.4 4.11.5 4.11.6 4.12

4.ll

Structural analysis of compound 0 Structural analysis of compoundE Structural analysis of compoundF

Thin-layer chromatographic (TLC) quantification ofisolallo.-d compounds

Structure-antioxidant activity relationships of idenlifk-d compounds

. . . . 247 . . . . 253 258 . . . . 261

.... , 264 4.13.1 Rosmarinic acid (compoundAof borage

extract) . ,.264

4.13.2 Syringic acid (compound Bof borage

extract) . . ... 265

4.13,1 Sinapic acid (compound C of borage

extract) . . , 266

4.13.4 (+)CatechinandHepicatechin (compounds

o

andE of evening primrose extract) . . . . 267 4.13.5 Gallic acid (compound F of evening

primrose extract) , , .. , . . . • • • . . .. 268 SUMMARY AND CONCLUTIONS .

REFERENCES APPENDIX1 . APPENDIX 2

xii

269 . 272 .. 303 . . . 371

LIST OF FIGURES

Figure2.1 Graphical representation of thestepsin"'olved in lipid

oxidation . . .. 8

Figure2.2 Initiation and propagation steps of the classical lipid autoxidation mechanism (A) and iron induCl.ld

decomposition of hydroperoxides (B). . . 9

Figure2.3 Formation of hydroperoxidesviaene reaction; involvement of Iriplelandsinglet oxygen in this process . . . ..13 Figure 2.4 Four electron reduction of oxygen into water (A).

supcroxide dismutalion reaction(B)and supcroltide driven

Fenton reaction. . ... 14

Figure 2.5 Regiospec:ificity of Iipoxygena.se (A) and involvement of lipoxygenase in the generation of free radicals(B). . 22 Figure2.6 Decomposition of hydroperoxides and subsequent

termination of chain reactions. . . .23 Figure2.7 Reaction of maJonalde:hyde. a secondary oxidation product

with amino compounds to produce an aminoiminopropcne

schiffbase . . . .2S

Figure2.8 Mechanism ofhexanaJfonnation ..28

Figure2.9 Mechanismof propanal formation. . 29

Figure2.10 Cycle of events associated with free radicaJ·medialed

inflammation and subsequent schemic injury 38

Figure2.11 Chemical structures of synthetic antioxidants . . . • • . . . . 42 Figure2.12 Mechanism ofactionof phenolic antioxidants . . . • • . . . 43 Figure2.13 Chemical structures ofrosemaryantioxidants . . .46 Figure2.14 Chemicalstructuresoforegano (Al and thyme (B)

antioxidants . . .. 48

xiii

Figure 2.15 Anlioxidants ofCapsicumfrulcs~ns(A). blackpepper(8)

andtunneric (C) . . _ . . . 49

Figure 2.16 Chemical SUUClUres of sesame antioxidants (A)andIignans

(B) . . .... 51

Figu~2.17 Chemical SUUCtures ofthemost active phenolic antioxidant

in canolameal(A)andtocopherols (8) .. . 53

Figure2.18 Chemical structures of soybean (A) and peanut (8)

antioxidants 54

F;gu~2.19 Chemical structures of antioxidants of rapseed meal . . 56 Figure 2.20 Chemical struclures of rice (A) and oat (8) antioxidants .... 57 Figure 2.21 Chemical struclures of black lea antioxidants 59 Figure 2.22 Chemical structures of green tea antioxidants 61 Figure 2.23 Chemical structures ofSchisandraceae antioxidants . . 62 Figure 2.24 Sleps involved intheformation of TBA·MA adduct 70 Figure 3.1 Graphical rqwesentation oftheface-centred cubedesign 84 Figure 4.1 Day-)gas chromatograms of cooked comminutedpork

treatedwithborage meal. (A) Control. (8) 1%.w/wmeal.

(C)2%.w/w meal 105

Figure 4.2 Day-] gas chromatograms of cooked comminutedpork treatedwithevmingprimrose:meal. (A) Control. (8) 1%,

w/wmeat.(C)2%,w!wmeal . . 106

Figure 4.3 Effect of borage meal on formation ofTBARS (A), hexanal (B) and total volatiles(C)in a cooked comminuted pork

modelsystem . 11 0

xiv

Figure 4.4 Effect of evening primrose meal on formation of TBARS (A).hexanal(B)andtotal volatiles(C) in acook~d

comminuted porkmodel system 111

Figure 4.5 Effect of varying extraction conditions on the antioxidant activity of crude extracts of bor.lge meal in a l}<arotene-

linoleate model system 114

Figure4.6 EffC(;t of varying extraction conditions on the antioxidant activity of crude extracts of evening primrose meal in a13-

carotene-Iinoleate model system 115

Figure 4.7 Response swface and its contour plot depicting w dependence of antioxidant activity of crude extracts of boragemealina l3<arotene-linoleate model system on combined effett of extraction mediumandtemperature 122 Figure4.8 Response surface and its contour plot depicting the

dependence of antioxidant activity of crude extracts of borage meal in a p-carotene-Iinoleate model system on combined effect of extraction mediumandtime 123 Figure 4.9 Response surface and its contour plot depicting the

dependence of antioxidant activity of crude extrac15 of boragemealina lkarotene-Iinoleate model system on combined effect of extractiontemperature andtime . 124 Figure4.10 Response surface and its contour plot depicting the

dependence of antioxidant activity of cl"Ude extracts of evening primrosemealinalkarotene-Iinoleate model system on combined effect of eXllaCtion medium and

temperature: . 125

Figure 4.11 Response surfaceand its contour plot depicting the dependence of antioxidant activity ofcrudeextracts of evening primrose mealina j3-carotene-linoleale model system on combined effect of extraction mediwnand

time. 126

Figuu 4.12 Response surface and its contour plot depicting the dependence of antioxidant activity of crude extracts of eveningprimrosemeal in a~lene-linolcate model system on combined effect of extraction temperatureand

time. 127

Figuu4.13 Relationship between predictedandobservedrespon~for

borage 132

Figure4.14 Relationship between predicted and obser....ed responses for

evening primrose 133

Figure4.15 Column chromatographic fraclion profile of borage(A)and

eveningprimrose (B) crude extraclS 135

Figure4.16 Ultraviolet(UV)spettra of borage crude extractandits

fractions 137

Figure4.17 Ultraviolet(UV)spettra of evening primrose crude extract

and its fraclions 138

Figure4.18 Effect of borage andevening primrose additives on oxidativestabilityof~tene(after2h) in a (k:arolene.

!inolcale model system 146

Figure4.19 Effect of borageandevening primroseadditives on formation ofTBARSina cooked comminuted pork model

system (on day-) . 150

Figure4.20 Effecl of borageandevening primrose: additives on formation ofhexanalina cooked comminuted pork model

system (onday~) . 152

Figure 4.21 Effect of borageandevening primrose additives on fonnation oftotalvolatiles in a cooked comminuted pork

model system (on day-) . 154

Figure 4.22 Effect of borage andevening primrose additives on fonnalion of conjugated dienes in a bulk stripped com oil

model system (on day-3) . IS7

xvi

Figure 4.2) Effect ofborage and evening primrose additives on formation of hexanal in a bulk stripped com oil model

system (on day-3) . 159

Figure 4.24 Effect of borage andevening primrose additives on formation of total volatiles in a bulkstripped com oil model

system (on day-) . 161

Figure 4.25 Effect of borage and evening primrose additives on formation of conjugated dienes in a stripped com oil-in-

\\<l[er emulsion system (on day-) . 164

Figure 4.26 Effect of borage andevening primrose additives on formation of hexanal in a stripped com oil-in-water

emulsion system (on day-) . 167

Figure 4.27 Effect of borageandevening primrose additives on formation of total volatiles in astrippedcom oil-in-water

emulsion system (on day-) . 169

Figure 4.28 Relationships between iron (II) chelating capacities of borage (A)/evening primrose (8) additives and stability of

~carotene in a ~-earotene-Iinoleate model system

containingtherespective additives 176

Figure 4.29 Relationshipsbetwm1iron (II) chelating capacity of borage additivesandformation of TBARS (A),hexanal(8)and total volatiles (C) in a cooked comminuted pork model

system containing the respective additives 177

Figure 4.30 Relationships between iron (II) chelating capacity of evening primrose additivesandformation of T8ARS (A).

hexanal (8)andtotalvolatiles (qina cooked comminuted pork model system containingtherespective additives . 178 Figure 4.31 Relationships between iron (II) chelating capacity of borage

additives and fonnation of conjugated dienes (A). hexanal (8) andtotalvolatiles(qina bulkstripped. com oil model

system containingtherespective additives 180

lCYii

Figure~.3:! Relationships between iron (II) chelating capacity of evening primrose additivesandformation of conjugatc:d dienes (A), hexanal (8)andtotal volatiles (C) in a bulk stripped com oil model system containing the respective

additives . 181

Figure 4.33 Relationships between iron (II) chelaling capacity of borage additivesandformation of conjugated dienes (A). hexanal (8) and total volatiles(C)in a stripped com oil-in-water emulsion containing the respective additives 183 Figure 4.34 Relationships between iron (II) chelatin& capacity of

evening primrose additives and formation of conjugated dienes (A). hexanal (8) and total volatiles(C)in a stripped com oil-in-water emulsion containing the respective

additives. 184

Figure 4.35 Hydrogen peroxide-scavenging capacities of borageand

evening primrose additives. 18S

Figure 4.36 Relationships between hydrogen peroxide.scavenging capacity of borage (Ayevening primrose (8) additivesand stability of Jkarotene in a fkarotene-linoleate model system containing the respective additives 188 Figure 4.37 Relationships bet'Attn hydrogen peroxide-scavenging

capacity of borage additivesandformation ofTBARS (A), hexanal(8)andtotal volatiles(C)ina cooked comminuted pork modelsySlCmcontainingtherespective additives. 190 Figure 4.38 Relationships between hydrogen peroxide-scavenging

capacity of eveningprimroseadditivesandformation of TBARS (A),hexanal(8)andtotal voliUiles(C)in a cooked comminuted pork model system containingtherespective

additives. 191

Figure 4.39 Relationships between hydrogen peroxide-scavenging capacity of borage additivesandformation of conjugated dienes (A), hexanal (8) and total volatiles(C)in a bulk stripped corn oil model system containingtherespective

additives 193

xviii

Figure4.40 Relationships between hydrogen peroxide-scavenging capacity of evening primrose additivesand formation of conjugated dimes (A),hexanal (8)andtotal volatiles(C)in a bulk stripped com oil model system containingthe

respective additives . . 194

Figure -1041 Relationships between hydrogen peroxide-scavenging capacity of borage additivesandformation of conjugated dienes (A), hexanal(8)andtotal volatiles(C)in astripped com oil-in·water emulsion containing the respective

additives. 195

Figure 4.42 Relationships between hydrogen peroxide-scavenging capacity of evening primrose additives and fonnation of conjugated dienes (A), hexanal(8)and total volatiles(C)in a stripped com oil-in-water emulsion containing the

respective additives. 196

Figure4,4j Superoxide radical-scavenging capacity of borage and

evening primrose additives. . 198

Figure4.44 Mechanism by whichthe superoxide radicalsaregenerated andsubsequent reaction of superoxidewithnitroblue

tetrazotium indicator. . 199

Figure 4.45 Hydroxyl radical-scavenging capacity of borageandevening primroseadditives . . . .. 202 Figure 4.46 Organicfree radical (DPPH}-scavcngina capacity of borage

and evenini primrose additives. . . .. 204

Figure 4.41 Relationships between superoxide radical-scavenging capacity of borage {A)/eveningprimrose(8) additivesand stability of !l-carotene in a Ikarotene-linoleatt model system containingthe respC'Ctive additives . .. 206 Figure 4.48 Relationships between hydroxyl radical·scavenging capacity

of borage (A)/evening primrose (8) additivesandstability of~-(arolenein a~.-carotene-Iinoleate model system

containingtherespective additives . . .. 201

xix

Figure-JA9 Relationships between organic fr~e radical (DPPH)- scavenging capacity of borage (A)/evening primrose (B) additives and stability of~-carO!enein a~--carotene- linoleate model system containingth~ respe<:tiv~additives .. 209 Figure 4.50 Relationships between superoxide radical-scavenging

capacity of borage additives and formation of TBARS (A).

hexanaJ (B) and total volatiles(e)in a cooked comminuted pork model system containing the respe<:tive additives. . 210 Figure 4.51 Relationships between superoxide radical-scavenging

capacity of evening primrose additives and formation of TBARS (Al, hexanal (B) and total volatiles(C)in a cooked comminuted pork model system containing the respe<:tive

additives. . 211

Figure 4.52 Relationships between hydroxyl radical-scavenging capacity of borage additives and formation of TSARS (A), hexanal (B) and total volatiles(C)in a cooked comminuted pork model system containing the respective additives 212 Figure 4.53 Relationships between hydroxyl radica!-scavengingcapacity

of evening primrose additivesandformation of TBARS (A), nexana! (B) and total volatiles(C)in a cooked comminuted pork model system containing the respective

additives. . . 214

Figure 4.54 Relationships between organic free radical (DPPH)- scavenging capacity of borage additives and formation of TBARS (A), nexana! (B) and total volatiles(C)in a cooked comminuted pork model system containing the respective

additives. .. 215

Figure 4.55 Relationships between organic free radical (DPPH)- scavenging capacity of evening primrose additives and formation of TSARS (A),hexanai (B)andtotal volatiles (C)in a cooked comminuted pork model systems containing

the respective additives . .. 216

Figure4.56 Relationships between superoxide radical-scavenging capacity of bomge additives and formation of conjugated dienes (A), hexanal (B) and total volatiles (C) in a bulk stripped com oil model systems containing the respective

additives. . ... 118

Figure4.57 Relationships between supet'Oxide radical-scavenging capacity of evening primrose additivesandformation of conjugated dienes (A). htxanai (B)andtotal volatiles (C) in a bulk stripped com oil model systems containing the respective additives . . . .. 119 Figure4.58 Relationships between hydroxyl radical-scavenging capacity

of borage addilives and fonnation of conjugated dienes (A), hexanal (B) and total volaliles (C) in a bulk stripped com oil model systems conwning the respettive additives. . . .. 220 Figure4.59 Relationships between hydroxyl radical-scavenging capacity

of evening primrose additivesandformation of conjugated dimes (A),hcxanaJ(B) and total volatiles (C) in a bulk stripped com oil model systems containingtherespective

additives . . _ 221

Figure 4.60 Relationships between organic free radical (DPPH)- scavenging capacity of borage additives and formation of conjugated dienes (A), hc:xanal (B) and tOlal volatiles (C) in a bulk stripped com oilmodelsystem containing the respective additives . . . .. 222 Figure4.6t Relationships between organic free radical (DPPH~

scavenging capac:ity of eveningprimroseadditivesand formation of conjugated dimes (A),hexanaI(8)andtotal volatiles (C)inabulkSlrippedcom oil modelsystem containingtherespective additives

Figure4.62 Relationships between superoxide radical-scavenging capacity of borage additives and formation of conjugated dienes (A), hexanal (B) andtotalvolatiles(C)in a stripped com oil-in-waltt emulsions containing the respective additives

.. 223

... 224

Figure 4.63 Relationships between superoxide radical-scavenging capacity of evening priITU'OSC additives and foonation of conjugated dienes (A), hexanal (8) and total volatiles (C) in a stripped com oil-in-water emulsions containing the

respective additives. .. 225

Figure 4.64 Relationships between hydroxyl radical·scavengingcapacity of borage additivesandformation of conjugated dienes (A).

hexanal (8)andtocal volatiles(C)in a stripped com oil·in- ....ilter emulsions containingthe respective additives . . . .. 226 Figure 4.65 Relationships between hydroxyl radical-scavenging capacity

of evening primrose additivesandformation of conjugated dienes (A), hexanal (8) and total volatiles (C) in a stripped com oil-in-water emulsions containing the respective

additives . . . . .. 228

Figure 4.66 Relationships betw~n organic free radical (DPPH)- scavenging capacity of borage addilivesandformation of conjugated dienes (A), hennal (8)andtotal volatiles(C)in a stripped com oil·in-....'8ler emulsions containingthe

respective addilives . . n9

Figure 4.67 Relationships betw=n organic free radical (DPPH)- scavenging capacity of evening primrose additivesand formation of conjugated dienes (A), hennal (8) and total volatiles(C) in a stripped com oil-in·water emulsions containingtherespective additives . . . .. 230 Figure 4.68 UVspectra of compounds A. 8andC . ....•••..•.•... 238 Figure 4.69 UVspectra of compounds 0, EandF . . . . .. 249 Figure A.I Dependence ofthe absorbance of sinapic acid·metal

complex at 725 nm onthecontent of sinapic acid inthe

medium 303

Figure A.2 Dependence of meabsorbaDccof catechin·metal complex at 725 nm onthecontent ofcatechininthemedium . . .. 304

xxii

Figur~A.3 Depend~nc~of the absorbance of~arou:neat 470run on

th~concenlration of j}-carotene inth~assay mediwn . 305 Figur~A.4 Dependence of theabsorbanc~of malonaldehyde(MA)-

TOA complex at 532nmon the concentration ofMA 306 Figur~A.5 Dependence ofthe:absorbanc~ratio(469nml530run)on

the concentration of free Fe!- in~M. 307

Figure.'\.6 Dependence of the absorbance of hydrogen peroxide at 230 nmon the concentration of hydrogen peroxide: inthe assay

m~dium .. 308

FigureA.7 Electron paramagneticresonanc~(EPR) spectra ofDMPO·

OH adductas affected by borage additives . . . .. 309 FigureA.8 Electron paramagnetic resonance (EPR) spectra ofDMPO-

OH adductas affected by~veningprimrose additives. 310 Figur~.'\.9 Electron paramagneticresonance(EPR) spectra of DPPH

free radicalsasaffectedby borage additives . . . .. 311 Figure A.IO Electron paramagnetic monance(EPR)spectra of DPPH

free radicalsasaffectedby eveningprimrose additives .

...

312 Figur~A.ll Massspectrumof compound A 313 Figure/1\.12 'H-NMRspectrumof compound A in acetone d-6 . ...31' FigureA.13 'H.'H-COSY spectrum of compoundAin acetone: d-6 .

...

31S Figur~A.14 Mass spectrum of compound 0 . 316 Figur~A.15 'H_NMR spectrum of compound 0 in methanol d-4 .

...

317

FigureA.l6 Mass spectrumof compoundC .

..

318

Figure A.l7 'H·NMRspettrum.of compoundCin methanol d-4. . 319 Figur~A.18 'H, [H·COSY spectrum of compound Cinmethanol d-4. . 320

xxiii

FigureA,19 Mass spectrum of compound0

Figure A.10 'H_NMR spectrum of compound0 in acetone d-6 Figure A.2l 'H,'H-COSY spectrum of compound0 in acetone d-6

.321 312 . . . 323 Figure A.:!2 IJC·NMR spectrum of compound0 in acetom: d·6 324

Figure A.23 Mass spectrum of compound E 325

Figure A.24 'H_NMRspectrum of compound E in methanol d-4 326 FigureA.l5 (H,IH-COSY spectrum of compound E in methanol d·4 327 FigureA.l6 llC_NMR spectrum of compound E in m:thanol d-4 328

Figure A.27 Mass spectrum of compound F

...

329

FigureA.l8 'H_NMRspectrum of compound F in methanol d-4

....

330

xxiv

LIST OF TABLES

Table 3.1 Face-centred cube design. . ... 83

Table 4.1 Fany acid composition of lardandstripped com oil 101 Table 4.2 Face-ttntred cube design and experimental response values

for borage. 119

T<lble 4.3 Face-centred cube design and experimental response values

for evening primrose. 120

T<lble 4.4 Estimated regression coefficients of the quadratic polynomial models for borageandevening primrose .. 129 Table 4.5 Critical factor levelsandresponse at maximwn point. 13\

Table 4.6 Column chromatographic data for borageandevening

primrose.. 136

Table 4.7 Ultraviolet (UV)spectraldata for borage andevening primrose crudeextract:. andtheir fractions 139 Table 4.8 Contents of toW. hydrophillicandhydrophobic phenolics of

borage crude exttae:tandits fractions. 141

Table 4.9 Contents of total. hydrophillic and hydrophobic phenolics of evening primrose crude extractandits fractions 142 Table 4.10 Concentration(~M)and proponion(YD)of chelated iron (II)

by borage etude extractandits fractions 172

Table 4.11 Concentration(p.M)andproponion of chelated iron {II} by eveningprimrosecrude extractandits fractions 174 Table 4.12 Rrvaluesandantioxidant activities of various borage

phenolicsandauthentic standards resolved on TLC plates. . .. 232 Table 4.13 Rrvaluesandantioxidant activities of various evening

primrose phenolicsandauthenticstandardsresolved on TLC

plates. . _ ..23S

Tabl.: 4.14 Mass spectral fragmentation pattern of compound A Table 4.15 Prolon assignments for compoundA ..

Table 4.16 Mass spectral fragmentation pattern of compoundB Table~.17 Proton assignments for compound B . Table 4.18 Mass spectral fragmentation pattern of compound C Table 4.19 Proton assignments for compound C .

.... 239 . 241 . ... 243 .... 244 .. 246 . . . 248 Table 4.20 Mass spectral fragmentation pattern of compound 0 251

Table4.21 Prolon assignments for compound0 . .. 252

Table 4.21 Carbonassi~tsfor compound 0 . 154

Table4.23 Massspectral fragmentation pattern of compoundE . 255

Table 4.14 ProlOn assignments for compound E 157

Table 4.25 Carbon assignments for compound E . . 259

Table4.26 Mass spectral fragmentation pattern of compoundF . .. 260

Table 4.27 Proton assignments for compound F 262

Table 4.28 Contents of isolated crude compounds in fractions. crude

extractsandmeaJs . . 263

Table A.l Effects of borage additives at a concentration of 100ppm as sinapic acid equivaJents on stability of a Jk.arotene- linoleate model system maintained atso-c . . 3) I TableA.2 Effects of borage additives al a concentration of200ppm

as sinapic acid equivalents on stability of a p-carotene-

!inoleatemodel systemmaintained at 50"C. 332 Table A.) Effects of evening primrose additives at a concentration of

100 ppm as catechin equivalents on stability of ap..

carotene-Hncleale model system maintained at 500c . . . 333

>eM

Table A.4 Effects of evening primrose additives at a concenuation of 200 ppm as catedun equivalents on stability of ap...

carotene-linoleate model system mainlained at SO"C . . . .. 334 TableA.5 Effects of borage additives at a concentration of 100 ppm

as sinapic acid equivalents on formation of TBARS in a cooked comminuted porL: model system stored at 40(: 335 Table A.6 Effects of borage additi\·es at a concentration of 200 ppm

as sinapic acid equivalents on formation of TBARS in a cooked comminuted pork model system stored at 4^GC 336 Table A.7 Effects of evening primrose additives at a concentration of

100 ppm as catechin equivalents on formation ofTBARS in a cooked comminuted pork model system stored at 4°C m Table A.8 Effects of evening primrose additives al a concentration of

200 ppm as catechin equivalents on formation ofTBARS in a cooked comminutedpork model system stored at 4^CC

...

338 Table A.9 Effects of borage additives at a concentration of 100 ppm

as sinapic acid equivalents on fonnation of hexanaI in a cooked comminuted pork m<Kk1 system stored at 40(:

..

339 Table A.IO Effects of borage additives at a concentration of 200 ppm

as sinapic acid equivalents on formation of hexanaIina cooked comminuted pork model system stored at 4"C 340 Table A.II Effects of evening primrose additives at a concenuation of

100 ppm as catechin equivalents on formation of hexanaI in a cooked comminutedpork model system stored at 40C 341 Table A.12 Effects of evening primrose additives at a concentration of

200 ppm as catechin equivalents on formation of hexanal in a cooked comminuted pork model system stored at 4cC . . .. 342 Table A.i3 Effects of borage additives at a concentration of 100 ppm

as sinapic acid equivalents on formation oftota! volatiles in a cooked comminuted pork model system stored at 4"C 343

xxvii

TableA.14 Effects of borage additives at a concentration of 200 ppm as sinapic acid equivalenls on formation of IotaIvolatiles in a cooked comminuted pork model system stored at 4°C 344 TableA.IS Effects of evening primrose additives at a concentration of

100 ppmas catechin equivalents on formation of lotal volatiles in a cooked comminuted pork model system ston.'d

M~ . . . .3~

TableA.16 Effects of evening primrose additives al a concentration of lOappmas calechin equivalents on formation of total volatiles in a cooked comminuted pork model system stored

at 4°C 346

TableA.17 Effects of borage additives al a concentration of100ppm as sinapic acid equivalents on foonalion of conjugalcd dienes in a bulk stripped com oil model system stored at

60°C. 347

Table Al8 Effects of borage additives at a concentration of 200 ppm as sinapic acid equivalents on formation of conjugatcd dienes in a bulk stripped com oil model system stored at

600C. . 348

TableA.19 Effects of evening primrose additives at a concentration of 100ppm as catechin equivalents on formation of conjugated dienes in a bulk stripped com oil model system stored at

6O"C. 349

Table A20 Effects of evening primrose additives at a concentration of 200 ppm as catechin equivalents on formation ofconjugated dienes in a bulk stripped com oil model system stored at

6O'C . . 350

Table A2l Effects of borage additives at a concentration of 100 ppm as sinapic acid equivalents on formation of hexanai in a bulk stripped corn oil model system stored at6O"C . ..351 Table A12 Effects of borage additives at a concentration of 200 ppm

as sinapic acid equivalents on formation of hexanai in a bulk stripped corn oil model system stored at 6O"C . . ... 352

xxviii

TableA.23 Effects of evening primrose additives at a concenlration of 100 ppmascatechin equivalents on formation of hexanal in a bulk stripped com oil model system stored at6O"C . . ... 353 Table A.2-i Effects of evening primrose additives at a concelllration of

200 ppmascatechin equivalents on fonnation of he.'(anal in a bulk stripped com oil model system stored at6<rC . . .. 354 TableA.15 Effects of borage additives at a concentration of100ppm

as sinapic acid equivalents on formation of total volatiles in a bulk stripped com oil model system stored at6O"C . 355 TableA.26 Effects of borage additives at a concentration of 200 ppm

as sinapic acid equivalents on formation of total volatiles in a bulk stripped com oil mlXkl system stom! at6O"C . .356 TableA.27 Effects of evening primrose additives at a concentration of

100 ppmascatechin equivalents on formation of tmal volatiles in a bulk stripped com oil model system stored at

6O"C. 357

TableA.:!8 Effects of evening primrose additives ataconcentration of 200 ppmascatechin equivalents on formation of total volatiles in a bulk stripped com oil model system stored at

6O"C .. 358

TableA.29 Effects of borage additives ataconcentration of 100 ppm assinapic acid equivalentsonformation of conjugated dienes in a stripped com oH-in-water emulsion system

stored at60°C 359

TableA.30 Effects of borage additives ataconcentration of 200 ppm assinapic acid equivalents on formation of conjugatt:d dienesina strippedcomoil-in-water emulsion system

stored at6lrC 360

Table A.31 Effects of evening primrose additives at a concentration of 100 ppm as catechin equivalents on fonnation ofconjugated dienes in a stripped com oil-in-water emulsion system stored at 6QOC

xxix

... 361

Table A.32 Effects of eveningpri~additives al a concentralion of 200 ppm as catechin equivalents on formation of conjugated dienes in astri~com oil-in-"''ater emulsion system

stored al6O"C . . . .362

Table A.33 Effects of borage additives at a concentralion of 100ppm as sinapic acid equivalents on formation of hexanal in a stripped com oil-in-water emulsion system stored at 6O"C . . .. 363 Table A.34 Effects of borage additives at a concentration of 200 ppm

as sinapic acid equivalents on fonnation of hexanal in a stripped com oil-in-water emulsion system stored at 6O"C . 364 Table A.35 Effects of evening primrose additives at a concentration of

[00 ppm as catechin equivalents on formation of hexanal in a slripped com oH-in-water emulsion system slOred at

60"C. 365

Table A.36 Effects of evening primrose additives at a concentration of 200 ppm as catechin equivalents on formation of hexanal in a stripped com oil-in-water emulsion system stored at

6lI'C. . 366

Table A.37 Effects of bofage additives at a concentration of 100 ppm as sinapic acid equivalents on formation of total volatiles in a stripped com oil-in-water emulsion system stored at

6O"C. . 367

Table A.38 Effects of borage additivesata concentration of 200 ppm as sinapic acid equivalents on fannation of total volatiles in a stripped com oil-in-waler emulsion system stored at

6lI'C. . ....368

Table A.39 Effects of evening primrose additives at a concentration of 100 ppm as catechin equivalents on formation of total volatiles in astripped com oil-in-water emulsion system

stored at6O"C .369

Table A.40 Effects of evening primrose additives at a concentration of 200 ppm as catechin equivalents on formation of total volatilesinastripped com oil-in-water emulsion system

stored at6O"C 370

ANOVA AOAC AOCS ATP BHA BHT CD COSY

CRn

CV DMPO DNA DHA DPPH EDTA EPA EPR FAME FlO FDA

LISTorABBREVIATIONS

-Analysis of variance

-Association of Official Analytical Chemists"

-American Oil Chemists' Society -Adenosintriphosphate -Butylated hydroxyanisole -Butylated hydroxytoluene -Conjugated dienes -Correlation spectroscopy -Completely randomized block design -coefficient of variance -S,S-Dimethyl-I-pyrrole-N-oxide -Deoxyribonucleic acid -Docosahexaenoic acid

·2.2·Diph<nyl·\·picrylhyd<azyl -Ethylenediaminetetraaeetic acid -Eicosapentaenoic acid -Electron paramagnetic resonance -Fatty acid methyl esters -Flame ionization detector -Food andDrugAdministration

xxxi

OC GLM HPLC HS [R

LD LDL LM MA MS m/z NADPH ND

NMR OS[

PO ppm PUFA PV QM

-Gas chromatography -General linear model

-High performance liquid chromatography -Headspace

-Infrared -lethal dose -Low-density lipoprotein -Linear model -Malonaldehyde -Mass spectrometry -mass to charge ratio

-Nicotimamide adenine dinucleotide phosphate -Not detected

·nanometre

-Nuclear magnetic resonance -Oil StabilityInstrument .Propyl gallate

·Partspermillion -Polyunsaturated fatty acids

·Peroxide value -Quadratic model

xxxii

R' RNA ROS RSM RSREG SAS SOD TBA TBARS TBHQ TCA TLC TMS UV USDA UHP

WOF w/v wlw

-Correlation coefficient -Coefficient of detennination -Ribonucleic acid -Reactive-oxygen spedes -Response surface methodology -Response surface regression -Statistical Analysis System -Superoxide dismutase -2·thiobarbituric acid

-Thiobarbituric acid-reactive substances -tertiary-Butyl hydroquinone -Trichloroacetic acid .Thin-layer chromalOgraphy -Tetramethylsilane -Ultraviolet

-United States Department of Agriculture -Ultra h.igh purity

-Volume by volume -Warmed-over flavour -We:ight by volume -Weight by weight

xxxiii

CHAPTER I INTRODUCTION

Lipidoxidationisofgreatconcern to

me

food industry andconsumers becauseitleads to the development ofundesirable off·8avoursandpotentially lOlOe reaction products (Maillard etat,\996; ShahidiandWanasundara, 1992). Syntheticanrimc:idants such asbutylated hydroxy,,"sol. (BHA), bulylaled hydroxytoluene (Bill), "niary-butylhyckoqWnone (TBHQ) and propyl gallate(PO)maybeadded tofood produCtS to recardlipidoxidation (Wmata and Lorenz,1996). Howeva'. use of syntheticantioxidantsin foodproductsisunder strict regulation due tothepotentialhealth hazardscausedbysuchcompounds(Heniarac:hchyetaI., 1996).Therefore, searchfornaI\IIaI antioxidants as alternatives to synthetic ones is of great interestamongresearchers,SeveralsourcesofRINrIIIlltioxidams areknownandsome of themare currentlyused in avarietyoffoodproducts(Melba etaI., 1994;inIlanietaI.,1982).

Extracts of herbssuchas rosemary (BraccoctaI.•1981),thyme(1natanidaI.,1982) andsage (Pizzocaroetai" 1994),oiIsccds such IS sesame(FukudaetaI., 1985;OsawactaI., 1985).

cano~(WanasundaraetaI., 1994),IIox(Amorow>:zetaI., 1990; OomohetaI.. 1995), soybeon (ChenetaI.,1995)....t _ (Duh....t Yen, 1995),cen:aIs_ a srice(AsamIrIietaI., 1996)....tbarley (MaiI1ardetaI., 1996),spiees_ a s """(Kramer,1985),...-d (S!loIIMti etaI., 1994.),twmeri<(CbipaultetaI., 1955),!PDF'- (LeeetaI., 1986)....ttmJarceI<

(HettiarachchyetaI., 1996), ....t beverap _ a sblacktea(IWchetaI., 1989)....tgreen tea (AmarowiczandShahidi, 1996;R.uchetaI., 1989)~beenreportedtobeIIIIioxidItivein

various model systems Theartioxidantactivityoftheseextractshasalways been anributed to their phenolic constituents. For instance,the antioxidantsinrosemaryextractshavebeen identified as phenolicssuch asrosmarinicacid,rosemarydiphenol and rosmanol(Houlihanet al., 1984; Nakataniand lnatani, 1984). Several. authors have reponedthepresence of flavonoids such as eatechinsinextracts of green and black teas. These phenoliccompounds can retard lipid olcidationbydonating ahydrogenatom or an efectron to chaininitiatingfree radicalssuchasthehydroxyl and superolrideradicals(CaoetaI.. 1991; Shahidiand Wanasundara, 1992). They canalsoneutralizethesubstrale-derived freeradicals suchasthe fany acid freeradicalsand&lkoX)'radicals(Cao ecaI.,1991; Packer and<Hazer,1990).koch etaI.(1989)reponed that teaextraelSwerecapaNe ofscavenging reactiveoxygenspecies (ROS),namelyhydrogen peroxide,supero>eideand hydroxyl _ Th,propertyofplal1l extractshasanimponantIdeinreurdinglipidoxidationinfoodproducts andIMngtissues.

lncorporation ofsuchextrICtSin tunInfoodsnotaNypreserves theirwholesomeness.but sIsoreducesI!leriskof~ortheroscIerosUandcancer (Ames,\983; Nill1lOO. 1990;

RamatathnameIaI., 1995).

PIanl_

wlIenodded to~foods,cansIso reducethelossofa-tocopherol (Jtice.Ewns etaI.,1996).PIarI:phenolicsCII'lregenerate a·

tocopherolfromtocophe:rylheradicalbydonatingan dectron or • hydrogen atom. Some phenotic compounds _ Utplant ...are"'Il'lft<dtorewdlipid _ tI1roullI>

chelation oftransitionmetal ionssuch asthoseof iron,copper andmIngIDe5e(Rioe-EVIIIS d aI.,I996).

Asalready _ naturalontioxicWws ..themealsof - .MeeuiIyand abunda.ttlyavailable. Borage(Borogoofficina/ist.)andcvming prinrose(o;,lOIhera hielmis).oil.scc:d cropsgrownmainlyinNorthAmerica.EuropeandAustralia.haveearnedan important placeinthephannIoeutical industry due10thehi@hcontent[l9-lrloand9 - 10'1t (w/w).intxnge and evri\gprimrose:5teds..respec:tivdy) ofy-linolenicacid(GLA)intheir seed oils (GibsonetaI.,1992;.Ra!wnatu11aeIal.•1994;Rtd6enetaI.•I99S). Borageand evening primrose oils havebeenused for treating severalskin disorders (Chapkin and Charmicheal, 1990;EnglereIIll., 1991, 1992). Themeals afteroilremoval mayretainI S!Jbstantialamount ofphenoIjcantioxidants(Luand Foo, 1995)which maybeextrICtedby employing a properectraetionteemique.Itmaybenecessary10evaludctheextractinseveral modelsystemsusingdifferatIlIIIIytic.aJt~inorderto

cnw.

valid condusion ontheir antioxidantdIiacy. _ 0Ulh0n _

""""ed

thIlthe _ aaMtyofpIool phenolicsdiifas~aoc:ord8ltllO the physicalond _ . . - . . .ofthemoddsyslem .. whM:hthey.,...-eII.F _ ..aI. (1994)~thIl~- . such as Troktx: (an «-tocopherol witbouIIskxwcbai:nbydroc:a'bon).-em:n:c&ccr.oein bulk oil systems whereas hydrophobicanboxidIds,u:Ilua~1ftmoreeffecUveinoil-in- water enJJ.Isions. Thisphenomenon is dueprimarilytothedifI"erencialaffiMiesofthe antioxidantcompounds for oil. . IndoiJ.waleriDcerfaces in bulk oil IlIdoiI·in-water

emulsions,respecIiveIy. This_~~_bythe_llIIUreofll1e

antioxidative compoundsirM>Md (Frookd. 1996).

One oflhe importantaspects of the extraction of antioxidarivecompoundsfrom plant materials is theselection of appropriate extractionconditions. Itisnot~toapplythe cooditionsusedfor one kindofplantmaterial to anotherbecause thediverse nature ofnatural antioxidantsmakes thegeneralized extraction conditionsinefficient. However,asetof optimum extraction conditions for a partiaJlar material canbeobtainedbyemployingresponse surface methodology(RSM),a toolusedbymanyresearchers topredict optimumexperimental conditions to maximize various responses(Gaoand Mazza,1996; WanasundaraandShahidi, 1996)

Based upon the literature evidences for antioxidantandradical intercepting propenies of plant extracts,anhypothesis wasmadethat borageandeveningprimrose extracts might also possess similar activities due to the presence of phenolic antioxidants.

Also based upon literatureevidencesfor the involvement ofvarious phenolic compounds in antioxidantrnethanisms, this hypothesis was further extended that borageand evening primrose phenolics might panicipateinsimilarmechanismsas other phenolics. Since antioxidant activities of phenolic compounds largelydependupon their affinittes to different phases and interfaces of model systemsbeingemployed to evaluate them, itwas thought that crude plant antioxidants mightalsoletinasimilarfashion. Therefore, in order to investigatethesehypotheses,severalobjectives were considered. These objectives were: (I) to evaluatetheantioxidantefIicacies of borageandeveningprimrose mealsina meat modelsystem,(2) to optimize extraction conditionsfor both borageand evening primrose toobtain extracts with IUgh antioxidantactivity, (3) to evaluatethe

antioxidant efficacies of crude eXlracts as well as their fractions in different model systems., (4) to examine the metal chelatingandreactive oxygen species (ROS) as well as organic free radical-scavenging activities of borageandevening primrose crude extractsandtheir fractions in order to understand the antioxidant mechanisms involved, (5) to elucidate the chemical structures of active compounds present in the extracts of borage and evening primrose meals and (6) to quantify the identified compounds in the extracts as well as in the source materials.

CHAPTERZ LITERATUR£REVIEW

Lipids are heterogeneous compounds which primarily serve as a source of fuel for both plants and animals. Apartfrom providing I condensed source of energy for living organisms(9kCalfg), theyplayseveral imponant functions in foods. Lipids contribute 10 food quality by providing organoleptic character notesthaimake them appealing for consumption. These propenies include flavour, colour. textureandmouthfeel.. In human nutrition, food lipidJ not only provide lhe essentiaJ rany acids such as linoleicandlinolenic acids, but also serveIS •sourceandIcarrierfor flt-solubtevitaminssuch asA,O. Eand K (St.Angelo,1996). Upids.however, undergo oxidation reactionswhichshorten the sbeIf-lif.ofliiXd... foods.Duringproduction. _ and Rangeprececting actualconsumption, foodlipidsundergovarious deterioration processes that involve mierobial, physicalandlllOSI~IIllIy,_ modes. TheIan"~ehuac:teriud by enzymaticandnonenz:ymacil:oxiduioaofIipiciswhichCIUIeundcsirIbIechangesin flavour, colour,andnutritionaJvalue. Deoxygenation.ain.:,!btpKkagingandocher techniques havesolvedsome of theseproblems, buttheroleofantioxidants CIRnOt be overlooked (NamiJci, 1990).Inthis chapter, the mechanism of lipid oxidlrion, toxtcityand diseasesassociatedwith the consumption of oxidizedlipids, methods of

assessma

lipid oxidation and role of antioxidantsinIipickontainiDgfoods will bediscussed.

2,1 Mtch.llism or lipid oaidat": illitiatioll.ad propaptiotl

When the olcidAtion of a lipid is monitored experimentally (by measuring the oxygen uptakc or peroxide value) itis found that the course of the oxidationshowsthree distinct phases. During thefirstphase(induction period), the oxidation takes place slowly, bul at a unifomt ralc. After oxidation reaches a certain point the reaction enters a second phase (propagalion) with asharpaa:der"ating ratc,andthe cventual ratc is many limes greater than that observcd in lhe initial phase. Thethird phase (termination) is characterized by a decrease in the rate of oxidation (Figure 2.1).

Figurc 2.2(A) outlines the initiationandpropagation steps of the classical lipid oxidation mechanism. In a peroxide-free system, lipid pcroltidation is initiated when a hydrogen atom is abstracted from • rnethy\ene group (-CH2group) adjacent to a double bondof an unsaturated fattyacid(HalliwellandGuttcridae.1985). free radicals canbe defined as species with an unpaired dearon,andthe reactivity of oxygen-derivedfree nt.dicals varies fromrelativelylow.asinthecase oftheoxygenmcMadeitself. to very high,as in the case of the short·livedandhighly reactive hydroxylradic.al("OH) (Packer andGlazer. 1990; Pocker. 1994). Hydrogen~_ ...edfromfattyocHj.bylUshlY reactive oxygen species (ROS) such IS ·OM,hence any reaction or process which forms ROS would definitely stimulate lipid oxidation. Hydrogen abstraction is easier in unsaturated fatty acids thanintheir saturated countetparts, thwmakingthem more susceptible to ROS attack.

Upon abstraction of. singlehydrogen 110mfrom •ftttyacid, the fatty acid carbon

Figure 2.\ Graphical representation of the steps involved in lipid oxidation.

InitietiOn

I

Ftropagation

Time

Figure 2.2 Initiation andpropagation stepsof the classical lipid autoxidation m«hanism (A)andiron induced decomposition of hydropc:roxides (8).

(A)

(8)

RH

Initiator

ROOH + R

I

R.QOH + Fe2+-e:ompfex _ F.3+-comptex + OW + R-O A1koxyl radical

R·OOH + Fe3+-compMIx _ F.2+--complex + H+ + R-OO Peroxyl radical

10

is left with an unpaired electron, i.e. a fatty acid free radical is fonned This radical is stabilized by a molecular rearrangement into a conjugated diene (Esterbauert!tal., 199\) which can undergo reactions such as cross--Iinking with fatty acid molecules. Under aerobic conditions, however, the most likely reaction is with oxygen and the product is a peroxyl radical (Halliwell and Gut1eridge, 1985). Theperoxyl radicals in tum Conn cyclic peroxides and are sufficiently reactive to abstract a hydrogen atom fromratty acid chains of other lipid molecules, fonning a lipid hydroperoxideand anewranyacid free radical Thus, the process of lipid olcidation is propagated by a free radical chain reaction, one initial hydrogen abstraction potentially leading to the formation of many lipid hydroperoxides and cyclic peroxides, collectively known as lipid peroxides. Pure lipid peroxides are reported tobe stable at physiological temperatures, but rapidly decompose in the presence of transitionmr..alcomplexes, espt(:iaUyironsalts (packerandGlazer, 1990).Ferrous ions reduce lipid peroxides to aJkoxyl radicals, while ferric ionscanfonn both alkoxy and peroxy radicals with lipid peroxides [Figure 2.2.(8»). Final decomposition products of thereaction between lipid peroxides and iron or copper complexes include hydrocarbongasessuchas ethane, ethyleneandpentane, carbonyl compounds suchasaldehydesandketonesISwellualcohols (EslerbauertlaI.,1991;

Wettasinghe and Shahidi, 1996). Mechanisms involvedinthedecomposition of hydroperoxides wiDbedetailedina later section.

\I

2.1.1 Futon.bl.fred lipido.idalioa

The major sources of primary catalysts that initiate oxidationillvivoandill vUro have been identified as oxygen and oxygen species. transitionmetalions and their complexes (haem proteins), electromagnetic radiation and enzymes. Thecontribution of these factors to lipid oxidation willbereviewed in the following sections

2.t.l.tRole ofollYI"aadoK)'ltilsptcits

The electronic structure of oxygenhastwo unpaired electrons at energy levels of p antibondins,intriplet state(~I.. Korycka-Dhal and Richardson, 1978). The reaction of oxygenwithground state molecules ofsingletmultiplicity (i.e., polyunsaturated rany acids, PUFA) is spin forbidden. However. this barrier doesnotapply 10 reactions which involve single electrons, hydrogen atoms,andmolccules containjng unpaired electrons such as transition metal complexesandfree radi<:aJs. Therefore,thetriplet state oxygen can reactwithothermolecules to yield ROS sucb as hydrogen peroxide(H~).

superoxide «h-),andhydroxyl radicalrOH)(Kanneret

m.,

1987). Tripletoxygen can also undergoenergytransition to generate singlet oxygen(It.and IAJ(SimicandTaylor, 1988). Pholosensitizers such as natural pigments (chiorophyUs, Bavinsand haem pigments) can generate singlet oxygen from triplet oxygen. Pbotosensitizers absorb visible or near-UV light tobecome electronica1ly excited. They can transferenergy(photons) onto triplet stale oxygeD molecules upontheU'returning 10groundstalewhich generales singlet oxygen (Krinsky, 1977). Singlet: oxygen.in1WD initiales lipid oxidation by an "ene"

12

reaction (Figure 2.3) where olefenic groups (-C=C-)

or

fillyacids are converted 10 their corresponding allyl hydroperoxidcs (Adam, 1975).

The singlet oxygen is generated mainly by photoexcitation of triplet oxygen and its reaction with PUFA only forms peroxides which do not trigger a radical chain reaction.It has been reponed that singlet oxygen-mediated lipid oxidation is more imponanl in food than in biological systems (Wassennan, 1979). Singlet oxygen, while a reactive oxygen species, lacks an unpaired electron and is therefore not a free radicalby definition. Free radical derivatives of oxygen include supe10xide radical(~-),hydroxyl radical ("OH), hydroperOlcy radical<Ho.:⁰) ,and nitric oxide ("NO) (HalliwellandGutteridge. 1985;

Packer and Glazer, 1990; Borg, 1993; Beckmaneta/.,1994).

Superoxide radical (<h'" ) is generatedbyfour electron reduction of molecular oxygen into water (Figure 2.4{A)]. This radical is also formed in aerobic cells due to electron leakage from the electron transport chain. Supcroxide radical(0:2Jis also formed by activated phagocytes (monocyte5. macrophage5, eosinophilsandneutrophils) and the production of 01'"is an important factorinthe killing of bacteria by phagocytes (Halliwell and Gutteridge, 1985; PackerandGlazer, 1990).Inliving organisms,0:2'"is removed by theenzymescalled superoxide dismutases (SOD). Thedismutation reaction of 01- is shown in Figure 2.4(8).

The H201,formed. due to 4 electron reduction of Ch into H:zOanddismutatKln of O2- ,is not a free radical, but an oxidizing agent. It,inthepresenceofCh-andtransition

13

Figure 1.3 Formation of hydroperoxidesviacne reaction: involvement of tripletand singlet oxygen in this process.

FV\ ~:::,..h'._, F\I\

^{+ H}

R R' sensitizer R • R'

OOH

R~I"""""'"

R'

OOH

;=v-\

R R'

Conjugated hydroperoxide

OOH

r-J-\

R R'

Non-conjugated hydropef'Oxide

"

Figure :!.4 Four electron reduction of oxygen into "iller (A). superoxide dismutation reaction(B)and superoxide driven Fenton reaction(C).

(A)

(6)

202 + 2H+ Superoxide diSmutase.. H 2

0 2

+ 02

(e)

Oxidized metal ion + 02 - Reduced metalion + 02

",o,U.~.~.

Overall readion

"

metal ions. can generate ·OH viathesuperoxide-driven Fenton reaction (HaUiweiland Gutteridge. 1985). This reaaion canbesummuiz.ed asshownin Figure 2.4(C). The 'OH, formed by this reaction&lid4 e- reduction of(h.is l\ighly reactiveandcauses damage to deoxyribonudcic .dd (DNA)and initiates lipid oxidation (packerandGlazer.

1990).

The hydroperoX)' radical(H(h0)is fonnedby photooKidation of lhe0.:.... Packer and Glazer (1990) reponed that HOt" is more reactivethan0:-and readily difusable across biological membranes where it attacks unsaturated (auy acids. Nitric oxide \NO), a vasodialating factor, isreleased bytheendothelium in response to stimulation with • variety of substances. Itis also producedbyplatelets inwhichit inhibits platelet aggregation.Thebio1og)ca1 aaivity of NO' is limitedbythe eonc::urrentpresenceof<h....

which reacts rapidly to formthe peroxynitrittanion (OONOl (Iuliano tlaJ.,1997).

However, protonated OONO- can decompose into ·OHwhichinturninitiates lipid oxidation(Beckmanelal., 1994).

Theoccurrence of kOS infoodsis incvitabic due to their biological nature.

Kannerel al.(1986) reportedthatmuscle lipidoxidation is initiacedbyROSand haem proteins. ROS also initiate lipid peroxidation in vegetableand animalfatsandoils (Bradley and Min, 1992; Rawls and van Santen, 1970). Free radical species of 0;: can directly abstract hydrogen atoms from methylene IfOUps Idjucent to oldenie groups of fany acids resulting in lhe formation offattyacid free:radicals. Inmeats. HzOl.u described elsewhere, can generate 'OHinthepresence:ofFe²•viaFenton reaction(Kanner

16

t!I 0/.•19&7;Kannerand Doll. 1991;Kanner.1994).

2.1.1.2 Roleoftn.,itiNI_dah

Transition mecal ions such asthoseofiron,. copper, magnesium. manganeseand zinc are abundantly present in both living organismsandfoods of both plantandanimal origin (Schaich, 1980;T~hivanganaMorrissey,1985). Theypanicipate in direct and indirect initiation of lipid oxidation (Schaich, 1980).Higher valence state metals such as iron, copper, manganese, nickelandcobalt areknownto panicipatc in direct initiation of lipidoxidation viaelectron transferandlipid alkyl radical formation.

Lower valence state metals can directly initiatelipidoxidation via formation of ROSwhichcanabstracta hydrogen atomfrommethylenegroupsadjacent10 double bonds of unsaNraiedfattyacids IcMIingto •he radical chainreaction(K.annu

e'

^oJ.,

1987, 1986). Fe?' canbeoxidized toFe)'whilereducing<>zto

az-;

~-innun can generate·OHvia supcroxide-drivalFentonrelICtion.

Indirectinitiationoflipidoxidationbymel&ltonsoccurswhenpreformed hydroperoxides(LOOH)are oxidized orreducedto formrtdicalsSl.tCh u LO·and

Loo·

(MinottiandAust, 1992). LO·andLoo·can increase theTatcof initiationbyabstracting hydrogen atoms from methylene groups adjacent(0doublebondsof unsaturatedratty acids. The redox potential ofothermeulssuchISmanganeseandcobaltue too low to cause hydropero,ode decompositionin aqueous systems. but they may Cltalyze hydropero,ode decomposition, espec:iaDyin non-polarmedia.byformation ofmetal-

17 hydroperoxide complexes (Kanner et01.. 1986, 1987).

One of the major problems encounteredby the meat indust!)' is the rapid deterioration of cooked meat quality due to lipid oxidation (Decker and Xu. 1998; Kanner el aI.,1986). In muscle tissue, iron exists in a protein-bound Conn in myoglobin, heamoglobin, ferritin and transferrin. All of these haem proteins havebeen shown 10 catalyzelipid oxidation (Decker and Welch, 1990; Love and Pearson, 1971; Love, 1987, 1988; Wenasinghe and ShaJljdi, 1996). Several authors reported that the nonhaem iron, as opposed to haem iron, is the principle prooxidantinmuscle (Kannerel aI.,1981),but rtUs idea is not widely accepted. Several other authors have claimed thata1ka!i and alkali- eanh metals also contribute to muscle lipid peroxidationby displacing catalytic iron ions ITom haem compounds (Rhee et01.,1983; Wettasinghe and Stwtidi, 1996).

1.1.1.3 Role orioailiac radiltioa

Theionizing radiation of principle concern in chemicalandbiological systems are charged panicles such as efectrons (p..particlesand8-rays), protonsand cr.-panicles, and electromagnetic waves or photons such u X-raysandy-rays(Bielski.1976). These panicles and electromagnetic WIves canionize atomsandmoleculesbyejectingelectrons from them, thus forming apositivelycharged speciesin theparentnwerial(Sctwch, 1980). Electrons ejectedintheionization process may themselvesbe sufficiently energetic to producefunherionizationandexcitation. Iftheirenergyis less than 100 eV, their range is shan,andresulting secondary ionizationswill beclose to the primary ionization