Lipid Oxidation Path

Lipid Oxidation Pathways
Volume 2
Lipid Oxidation Pathways
Volume 2
Editors
Afaf Kamal-Eldin
Department of Food Science
Swedish University of Agricultural Sciences
David B. Min
Department of Food Science and Technology
Ohio State University
Urbana, Illinois
AOCS Mission Statement
To be a global forum to promote the exchange of ideas, information, and experience, to enhance personal
excellence, and to provide high standards of quality among those with a professional interest in the science and
technology of fats, oils, surfactants, and related materials.
AOCS Books and Special Publications Committee

M. Mossoba, Chairperson, U.S. Food and Drug Administration, College Park, Maryland
R. Adlof, USDA, ARS, NCAUR-Retired, Peoria, Illinois
M.L. Besemer, Besemer Consulting, Rancho Santa, Margarita, California
P. Dutta, Swedish University of Agricultural Sciences, Uppsala, Sweden
T. Foglia, ARS, USDA, ERRC, Wyndmoor, Pennsylvania
V. Huang, Yuanpei University of Science and Technology, Taiwan
L. Johnson, Iowa State University, Ames, Iowa
H. Knapp, DBC Research Center, Billings, Montana
D. Kodali, Global Agritech Inc., Minneapolis, Minnesota
G.R. List, USDA, NCAUR-Retired, Consulting, Peoria, Illinois
J.V. Makowski, Windsor Laboratories, Mechanicsburg, Pennsylvania
T. McKeon, USDA, ARS, WRRC, Albany, California
R. Moreau, USDA, ARS, ERRC, Wyndoor, Pennsylvania
A. Sinclair, RMIT University, Melbourne, Victoria, Australia
P. White, Iowa State University, Ames, Iowa
R. Wilson, USDA, REE, ARS, NPS, CPPVS-Retired, Beltsville, Maryland
AOCS Press, Urbana, IL 61802

©2008 by AOCS Press. All rights reserved. No part of this book may be reproduced or transmitted in any form
or by any means without written permission of the publisher.
ISBN 978-1-893997-56-1
Library of Congress Cataloging-in-Publication Data
Lipid oxidation pathways / [edited by] Afaf Kamal-Eldin.

p. cm
Includes bibliographical references and index.
ISBN 1-893997-43X (acid-free paper)
1. Lipids--Oxidation. I. Kamal-Eldin, Afaf.
QP751.L5517 2003
572´.57--dc21
2003006200
CIP
Printed in the United States of America.

12 11 10 09 08 5 4 3 2 1
The paper used in this book is acid-free and falls within the guidelines established to ensure permanence and
durability.
Contents
Preface .......................................................................................................................vii
1 Chemistry and Reactions of Singlet and Triplet Oxygen in Lipid Oxidation

Hyun Jung Kim and David B. Min..................................................................1
2 Chemistry and Reactions of Reactive Oxygen Species in Lipid Oxidation
Eunok Choe..................................................................................................31
3 Oxidation of Long-Chain Polyunsaturated Fatty Acids
Kazuo Miyashita...........................................................................................51
4 Oxidation of Conjugated Linoleic Acid
Taina I. Pajunen (née Hämäläinen) and Afaf Kamal-Eldin............................77
5 Oxidation of Cholesterol and Phytosterols
Afaf Kamal-Eldin and Anna-Maija Lampi....................................................111
6 Tocopherol Concentration and Antioxidant Efficacy
Afaf Kamal-Eldin, Hyun Jung Kim, Levon Tavadyan, and David B. Min........127
7 Carotenoids and Lipid Oxidation Reactions
Afaf Kamal-Eldin.........................................................................................143
8 Co-Oxidation of Lipids with Proteins and Nucleic Acids
Karen M. Schaich.........................................................................................181
9 Lipid Oxidation in Food Dispersions
Eric A. Dekker, Wilailuk Chaiyasit, Min Hu, Habibollah Faraji,
and D. Julian McClements............................................................................273
10 Antioxidant Evaluation Strategies
Leif H. Skibsted............................................................................................291
Index..................................................................................................................... 307
v
Preface
The first volume of Lipid Oxidation Pathways, published in 2003, tried to highlight
some of the anomalies and gaps in our current understanding of the lipid oxidation
mechanism and kinetics. The different chapters discussed how lipid oxidation pro-
ceeds in different environmental surroundings and highlighted areas where further
research is needed. A number of prominent scientists in academia and industry said
they appreciate the book for its particular focus on the anomalies between observed
and predicted behaviour and requested a new volume of the book that focuses on the
oxidation kinetics and mechanisms governing the behaviour of different molecular
species involved in lipid oxidation reactions. This second volume thus aims to com-
plement the first volume and extend it to more detailed reviews of the reactions of
lipid molecules other than conventional polyunsaturated fatty acids.
The first chapter discusses the basic chemistry of singlet oxygen and its involve-
ment in lipid oxidation reactions with particular reference to reactions in foods such
as reversion of flavour in soybean oil. The second chapter discusses the different reac-
tive oxygen species and their particularities in lipid oxidation. Chapter 3 discusses in
detail the oxidation of long chain polyunsaturated fatty acids, including eicosapentae-
noic acid (EPA) and docosahexaenoic acid (DHA), and discusses the kinetic effects of
different environments. Chapter 4 provides a detailed description of the oxidation of
conjugated linoleic acid (CLA), particularly the fact that hydroperoxides are not the
major primary oxidation products from this fatty acid and that conventional methods
for estimating the degree of lipid oxidation (such as peroxide value and conjugated
dienes) will lead to erroneous conclusions in this case. Chapter 5 describes the oxida-
tion of sterols, which follow the same basic mechanism as for oleic acid. In chapter 6,
possible reactions behind the paradoxical behaviour of tocopherols, i.e. increased rate
of inhibited oxidation at high levels of these antioxidants or what has been known as
tocopherol’s pro-oxidant effect, has been delineated. Chapter 7 reviews the literature
pertinent to the oxidation mechanisms and oxidation products of carotenoids. Like
with the case of CLA, radical addition rather than hydrogen abstraction seem to be
the most plausible mechanism. The co-oxidation of lipid and proteins is discussed
comprehensively in Chapter 8 with emphasis on the reactions of proteins with free
radicals and lipid hydroperoxides, epoxides, and carbonyl compounds. Chapter 9 dis-
vii
viii A. Kamal-Eldin and D.B. Min
cusses lipid oxidation in emulsions and how it is affected by interaction with proteins,
type of emulsifier, antioxidants, and the structural organization of different molecular
species. Chapter 10 discusses the complication in evaluating lipid antioxidants and
inhibitors as it relates to the widely different modes of their action.
The editors sincerely thank the authors of the different chapters for their valu-
able contribution with timely literature reviews within this second volume of Lipid
Oxidation Pathways. We also acknowledge, with sincere gratitude, Jodey Schonfeld,
Brock Peoples, and their colleagues at the AOCS Press for their professional help in
the editing and for the production of this volume.
Afaf Kamal-Eldin David B. Min

Uppsala, Sweden Columbus, OH
1
Chemistry and Reactions of Singlet and
Triplet Oxygen in Lipid Oxidation
Hyun Jung Kim and David B. Min
Department of Food Science and Technology, The Ohio State University, 2015 Fyffe
Road, Columbus, OH 43210
Introduction
Oxidation of food components can influence flavor quality, nutritional quality, con-
sumer acceptability, and toxicity of food products (Min and Boff, 2002; Choe and
Min 2006). Lipid oxidation products are especially responsible for the development of
rancidity by the production of low-molecular-weight compounds that impart undesir-
able flavors. Oxidation occurs by both triplet oxygen and singlet oxygen. Atmospheric
triplet oxygen contains two unpaired electrons, while singlet oxygen has no unpaired
electrons. The electron arrangement of triplet oxygen does not allow for a direct reac-
tion with compounds, such as unsaturated fatty acids, that is nonradical and in the
singlet state. Triplet oxygen oxidation of foods has been extensively studied during
the last 70 years (Labuza, 1971; Min and Lee 1996). Singlet oxygen formed in the
presence of triplet oxygen by excitation is thought to be responsible for initiating lipid
oxidation of food compounds, due to its ability to directly react with the electron-rich
compounds. Singlet oxygen rapidly increases the oxidation rate of foods even at very
low temperatures (Rawls and Van Santen, 1970). Singlet oxygen oxidation can produce
new compounds, which are not found in ordinary triplet oxygen oxidation in foods
(Frankel et al., 1981). During the last 30 years, the attention paid to singlet oxygen
oxidation of foods has increased.
This chapter will discuss the basic information on the chemical properties, forma-
tion, inhibition, and detection of singlet oxygen oxidation as it relates to the oxidation
of lipids.
Electron Configuration of Triplet and Singlet Oxygen

The molecular orbital theory best describes the electron structure of molecular oxy-
gen and its excited states (Korycka-Dahl and Richardson 1978). Molecular oxygen is
composed of two oxygen atoms and has 10 molecular orbitals containing 12 valence
electrons. Valence electrons are added sequentially to the orbitals in order of increasing
energy to obtain molecular orbitals of triplet oxygen and singlet oxygen (Figures 1 and
2). According to Pauli’s exclusion principle, only two electrons can occupy each orbital.
Hund’s rule states that one electron is placed into each orbital of equal energy one at a
1
2 H.J. Kim and D.B. Min
Molecular
σ*
Atomic Atomic
π* π*
π π
2px 2py 2pz 2pz 2py 2px
σ
Energy
σ*
2s 2s
σ
σ*
1s 1s
σ
Fig. 1.1. 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
Fig. 1.2. Molecular orbital of singlet oxygen.

Chemistry and Reactions of Singlet and Triplet Oxygen in Lipid Oxidation 3
time before the addition of the second electron. Pauli’s exclusion principle also states
that electrons in a given orbital must have opposite spins. An electron in an atom or
molecule generates two types of magnetic momentum and mechanical angular mo-
mentum. This behavior occurs from the motion of the electron around the nucleus
and from the spin of the electron.
The spin multiplicity used to define spin states of molecules is defined as 2S + 1,
where S is the total spin quantum number. The ground state of the molecule is singlet
if the resultant spin (S) is 0, indicating the spin multiplicity to be 1. An excited state
is formed by removing one of the electrons from the outermost filled orbital (bonding
π) of the ground state to a vacant orbital (antibonding π*) of the higher energy. The
ground state of most stable molecules containing an even number of electrons is dia-
magnetic due to the arrangement of electrons into pairs with opposite spin magnetic
moments. The ground state for most molecules is singlet state until a molecule is ex-
cited to the triplet state. However, molecular oxygen is exceptional among molecules
with an even number of electrons. The electron configuration of triplet oxygen has
two unpaired electrons occupying two degenerate antibonding π* orbitals (Fig. 1.1).
The unpaired electrons in the antibonding π* orbitals can have the same spin, so the
total spin number (S) is 1/2 + 1/2 = 1. The spin multiplicity (2S + 1) is 3, which is
known as a triplet configuration because the spin has three possible alignments under
magnetic field. Triplet oxygen is diradical and paramagnetic. Diradical triplet oxygen
cannot react with food components, which are not radical compounds unless the
food compounds become radical compounds. Triplet oxygen can react mostly with
radicals.
Singlet oxygen having paired electrons is in violation of Hund’s rule and is a
highly energetic molecule. The resulting electronic repulsion can produce five excited
state conformations. An activated 1Δ state at 37.5 kcal above the ground state and
an activated 1Σ state at 22.4 kcal above the ground state are two common states
(Korycka-Dahl and Richardson, 1978; Girotti 1998). The 1Σ state of oxygen has two
electrons with opposite spins in different orbitals this is very reactive and not able to
survive relaxation to the ground state. The less energetic 1Δ state of oxygen is suf-
ficiently stable enough to react with other singlet state molecules. The 1Δ state is re-
sponsible for most singlet oxygen oxidation and generally referred as singlet oxygen.
Singlet oxygen is not a radical compound and is very electrophilic. The energy of
singlet oxygen is 22.4 kcal above the ground state of triplet oxygen and it exists long
enough to react with other singlet state molecules (Korycka-Dahl and Richardson,
1978; Girotti, 1998). The lifetime of singlet oxygen is 50-700 microseconds depend-
ing on the solvent system of foods. Electrophilic singlet oxygen reacts with nonradical,
singlet-state, and electron-rich compounds containing double bonds. Once singlet
oxygen is formed, it is responsible for initiating singlet oxygen oxidation that rapidly
produces free radicals that in turn can initiate a free-radical chain reaction. Table 1.1
shows a summary of chemical properties of triplet oxygen and singlet oxygen.
Table 1.1. Comparison of Triplet and Singlet Oxygen

Triplet Singlet
Energy Level 0 22.4 kcal/mole
Nature Diradical Non-radical, Electrophilic
Reaction Radical compound Electron-rich compounds
Formation of Singlet Oxygen

Singlet oxygen can be formed chemically, enzymatically, and photochemically as
shown in Fig. 1.3 (Choe and Min, 2005). Some mechanisms for singlet oxygen for-
mation in Fig. 1.3 have not been unequivocally proven scientifically and have been
questioned. Detailed studies of the many different mechanisms for the formation of
singlet oxygen in foods should be studied further.
2R-CHOO· - R’
H2O2
3
HOO· O2 + 3Sensitizer
Irradiation 1
1 Sensitizer
Fe3+ O2
H2O2 + Fe2+ · OH + OH− Fe2+
Irradiation H·
H2O2 Fe3+
H2O O2·−
Sensitizer·+
3O
+ aq e− H+ 3
O2 + 3Sensitizer
2
Xanthine oxidase + 3O2 R+

HOO·
H2O
ROO·
H2O2 + · OH
Fig. 1.3. Singlet oxygen formation by chemical, photochemical, and biological methods.
Singlet oxygen is produced from Haber-Weiss reaction (Kellogg and Fridovich,
1975). Superoxide anions formed from triplet oxygen produces hydrogen peroxide by
spontaneous dismutation. Hydrogen peroxide reacts with a superoxide anion to form
singlet oxygen by Haber-Weiss reaction (Halliwell and Gutteridge, 2001). Haber-
Weiss reactions rarely occur in aqueous solution in the absence of transition metals,
such as iron or copper, which catalyze the decomposition of hydrogen peroxide to the
hydroxyl radical.
O2·- + O2·- + 2H+ H2O2 + O2

Dismutation
H2O2 + O2·- ·OH + OH- + 1O2
Haber-Weiss

Singlet oxygen is also produced by the Russell mechanism from peroxy radicals (Hal-
liwell 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
· ·
The metastable phosphatidylcholine hydroperoxides produced singlet oxygen

during hydroperoxide breakdown in the presence of Cu2+ in the dark (Takayama et
al., 2001). Retinal and retinyl palmitate in ethanol produced singlet oxygen by UV
light (Delmelle, 1997).
The major pathway for the formation of singlet oxygen in foods is photosensi-
tization. The interaction of light, photosensitizers, and triplet oxygen is the basis for
the formation of singlet oxygen by photochemical mechanisms. The photosensitizers,
such as chlorophyll, pheophytins, porphyrins, riboflavin, myoglobin, and synthetic
colorants in foods, can absorb energy from light due to their conjugated double bonds
and transfer it to triplet oxygen to form singlet oxygen (Foote et al., 1970; Afonso
et al., 1999; Lledias and Hansberg, 2000). Once singlet oxygen is formed, it directly
reacts with compounds that contain high densities of electrons, such as double bonds,
and forms mixtures of conjugated and nonconjugated hydroperoxides that readily
break down and produce undesirable compounds.
The effects of light on the flavor stability of food can be explained by both photo-
lytic autoxidation and photosensitized oxidation. Photolytic autoxidation is the pro-
duction of free radicals primarily from lipids during exposure to light. Direct interac-
tion of UV or visible light and lipids in foods are minimal, and thus are not a primary
concern. However, photosensitized oxidation occurs in the presence of photosensitiz-
ers and light very quickly. The photochemical mechanism begins with the absorption
of light by photosensitizers, which depends on the arrangement of electrons around
the atomic nuclei in its structure. As light energy is absorbed, an electron is boosted
to a higher energy level and the ground singlet state of the sensitizer (1Sensitizer) be-
comes an unstable excited singlet state (1Sensitizer*). It takes only a few picoseconds
for a chlorophyll molecule to absorb energy and become excited singlet chlorophyll
(Foote, 1976). When light energy is removed, the electrons rapidly lose energy and
return to the lower energy ground state. The excited singlet sensitizer may undergo
three physical processes: internal conversion, emission of light, or intersystem cross-
ing as illustrated by Jablonski’s diagram (Fig. 1.4). Internal conversion involves the
transformation of one excited state into another of the same spin state, resulting in
Singlet Manifold System Triplet Manifold System

1
Sensitizer*
Excited State
Intersystem
Crossing
Internal
Conversion Fluorescence
First 3
Sensitizer*
Excited Lowest
Singlet Triplet
+ 3O2
Phosphorescence
1
O2
1
Ground State Sensitizer
Fig. 1.4. Jablonski diagram.

loss of energy as heat. Not only can the excited singlet sensitizer lose energy as heat,
but also it may emit light to the ground state. The nature of the emission light de-
pends on the state multiplicity of the molecule. If the state from which the emission
originates and terminates has the same state, the emission light is called fluorescence.
Emission of fluorescence converts the excited singlet sensitizer to ground state singlet
sensitizer, which are extremely fast processes.
The excited singlet sensitizer (1Sensitizer*) becomes the excited triplet state (3Sen-
sitizer*) via intersystem crossing, which involves the conversion of the excited singlet
state to the triplet state or vice versa. The excited triplet sensitizer is degraded to a
lower, triplet-energy state that decays to the ground singlet state by the emission of
light. Since the emission originates from the triplet state and terminates in a singlet
state, the emitted light is called phosphorescence. The 3Sensitizer* is the reactive in-
termediate in photosensitized oxidation. The lifetime of the 3Sensitizer* is greater
than the 1Sensitizer*. The excited triplet reacts with the same state of triplet oxygen
(3O2) and forms singlet oxygen by triplet-triplet annihilation. The sensitizer returns to
ground state and may begin the cycle again to generate singlet oxygen. Sensitizers may
generate 103-105 molecules of singlet oxygen before becoming inactive (Kochevar and
Redmond, 2000).
Type I and Type II Mechanisms

Once the 3Sensitizer* is formed, there are two major mechanisms it may undergo:
Type I and Type II (Foote, 1976; Sharman et al., 2000). Type I is the direct interac-
tion of the sensitizer with another molecule and is associated with the formation of
free radicals. Type II is most commonly described as the transfer of energy from the
3
Sensitizer* to atmospheric triplet oxygen to produce singlet oxygen.
The photosensitized oxidation process of the 3Sensitizer* is shown in Fig. 1.5. The
Type I mechanism is characterized by hydrogen atom or electron transfer between an
excited triplet sensitizer and a reducing substrate which results in the formation of free
radicals (R·). The 3Sensitizer* acts as a photochemically activated free-radical initia-
tor. The R· can react with radical triplet oxygen to form peroxy radicals which can
abstract hydrogen from other compounds to produce oxidized compounds. The oxi-
dized compounds readily break down to form free radicals that can initiate free-radi-
cal chain reactions. The 3Sensitizer* in the Type I mechanism can also react with 3O2
to form superoxide anions by electron transfer to triplet oxygen. Less than 1% of the
reaction of triplet sensitizer and triplet oxygen produces superoxide anion (Kepka and
Grossweiner, 1972). The rate of Type I reactions is dependent on the type and con-
centration of the sensitizer and substrate. For example, the sensitizer benzophenone
abstracts hydrogen from an alcohol 10,000 times faster than eosin; however, both eo-
sin and benzophenone react at similar rates with more powerful reductants like N,N-
dimethylaniline (Foote, 1976). Generally, the readily oxidizable compounds, such as
phenols and amines or readily reducible compounds such as quinones, favor Type I
mechanism (Korycka-Dahl and Richardson, 1978).
The 1Sensitizer* usually does not have enough time to react with other molecules
due to its short lifetime. The Type II mechanism is the energy transfer process in
which the 3Sensitizer* reacts with triplet oxygen via triplet-triplet annihilation to gen-
erate singlet oxygen (Fig. 1.5). Energy is transferred from the high energy 3Sensitizer*
to low energy triplet oxygen (3O2) to form high energy singlet oxygen (1O2) and low
1
Light
1
ISC 3
Sen Sen* Sen*
Type I Type II
3
RH O2
R· + SenH· 1
O2 + 1Sen
Type I + 3O2
R·+ + Sen·−
3
O2, H· RH
ROOH O2·– + 1Sen ROOH
Fig. 1.5. Formation of excited triplet sensitizer (3sen*) and its reaction with substrate via Type I and Type II
reaction.
energy ground state singlet sensitizer (Sharman et al., 2000). This reaction occurs very
quickly and accounts for almost all of energy transfer from the 3Sensitizer* to triplet
oxygen.
The competition between substrate and triplet oxygen for the 3Sensitizer* is a ma-
jor determinant of whether the reaction mechanism is Type I or Type II. Electron-rich
compounds, such as olefins and aromatic compounds, favor the Type II mechanism.
The rate of Type II reaction mainly depends on the solubility and concentration of
oxygen present in the food system. As the oxygen in a system becomes depleted,
the shift from Type II to Type I mechanism is favored (He et al., 1998; Song et al.,
1999). Oxygen is more soluble in lipids and nonpolar solvents than in water (Ke
and Ackman, 1973). The trace amounts of chlorophyll as a sensitizer in vegetable oil
tend to promote photosensitized oxidation by the Type II mechanism. In contrast,
water-based food systems, such as milk, favor Type I mechanism due to the reduced
availability of oxygen to interact with the 3Sensitizer*.
The natural decay rate of singlet oxygen to the ground state and the rate at which
it reacts with a particular substrate must also be considered. The reactions of singlet
oxygen with unsaturated fatty acids or other compounds depend on the lifetime of
singlet oxygen. The lifetime of singlet oxygen is different depending on the type of
solvent and ranges from 2 to 700 microseconds as shown in Table 1.2 (Adams and
Wilkinson, 1972; Kearns, 1979). Temperature has little effect on the reaction rate of
singlet oxygen oxidation, as indicated by negligible changes in the lifetime of singlet
oxygen at a temperature range of -37.6 to 21.6°C (Min and Boff, 2002). This suggests
that the lifetime of singlet oxygen in food systems is largely dependent on the type of
matrix, water or fat, in foods.
Detection of Singlet Oxygen

Direct demonstration of production and involvement of singlet oxygen in any re-
action pathway is difficult due to its short lifetime. However, analytical techniques
Table 1.2. Lifetime of Singlet Oxygen in Solution
Solvent Lifetime (µs)
Protiated acetone 46.5
Deuterated acetone 690
Protiated acetonitrile 54.4
Deuterated acetonitrile 600
Protiated benzene 26.7
Deuterated benzene 550
Protiated water 4
Deuterated water 68.1

have been developed to detect the singlet oxygen molecule. Singlet oxygen oxidation
products are characteristically different from the products formed from a free-radi-
cal chain reaction. The chemical structures of photosensitized oxidized products can
demonstrate the existence of singlet oxygen. For example, autoxidation only produces
conjugated hydroperoxides, whereas singlet oxygen oxidation can produce nonconju-
gated and conjugated hydroperoxides. Singlet oxygen can be produced free of other
reactive oxygen species and the reaction products of pure singlet oxygen oxidation can
be compared with the experimental oxidation product. The following methods have
been used to detect the singlet oxygen involvement in chemical reactions.
Deuterium Oxide Effect

The physical and chemical properties of deuterium (D2O or 2H2O) are similar to
those of water (H2O). The isotope effect of heavy hydrogen (2H) is magnified further
in biological systems, which are very sensitive to small changes in the solvent proper-
ties of water. The lifetime of singlet oxygen in D2O is about 13 times longer than that
in H2O (Li, 1997). The longer lifetime of singlet oxygen in D2O allows for the detec-
tion of singlet oxygen in various reactions compared to H2O. Therefore, the reaction
rate of singlet oxygen in D2O should increase when a reaction in aqueous solution is
dependent on singlet oxygen. The D2O effect is a result of the decay rate of singlet
oxygen. The rate of photosensitized oxidation in D2O is 10 times higher than in H2O
and would indicate the involvement of singlet oxygen in the reaction (Athar et al.,
1988; Pecci et al., 2000). The effect on the ratio of the decay rate of singlet oxygen in
H2O to D2O may actually be larger than 10 times.
Consider the following in order to study the effect of deuterium: the normal
difference between pD and pH; the effect of H-D exchange on the active sites of the
system under study; the possibility of different rates of free-radical reactions in D2O
vs. H2O; and different excited state lifetimes and intersystem crossing efficiencies in
D2O vs. H2O (Foote, 1976; IUPAC, 2001). Simply observing an increase in the rate
in D2O is not a specific test verifying the existence of singlet oxygen. This method
has two important limitations: (1) if the isotope effect is to be observed, the lifetime
of singlet oxygen must be limited by the solvent. If all the singlet oxygen is being
scavenged by substrate, no effect will be observed. (2) The lifetime of the superoxide
anion is also longer in D2O than in H2O, so superoxide anion reactions should go
more rapidly in D2O than in H2O (Foote, 1976).
Chemical Traps
Chemical compounds can be used to trap singlet oxygen, forming a unique product
that is indicative of the presence of singlet oxygen. N,N,N’,N’-tetramethyl ethylene-
diamine and 1,3-diphenylisobenzofuran (DPBF) are known chemical trap agents of
singlet oxygen (Kochevar and Redmond, 2000). The absorbance of DPBF in organic
solvents measured at 410 nm decreases as the molecule reacts directly with singlet
oxygen (Kochevar and Redmond, 2000).
Cholesterol has been used to distinguish between singlet oxygen oxidation and
triplet oxygen oxidation. Cholesterol reacts with singlet oxygen to form specific oxi-
dative products via the “ene” reaction. This specificity makes the use of cholester-
ol an effective indicator of singlet oxygen oxidation in situ, where the use of other
detection techniques is difficult. The reaction of cholesterol with singlet oxygen
produces 3β-hydroxy-5α-cholest-6-ene-5-hydroperoxide (5α-OOH), 3β-hydroxy-
cholest-4-ene-6α-hydroperoxide (6α-OOH), and 3β-hydroxyc-holest-4-ene-6β-
hydroperoxide (6β-OOH) as shown in Fig. 1.6. The reaction of cholesterol with
triplet oxygen produces 3β-hydroxycholest-5-ene-7α-hydroperoxide (7β-OOH) and
3β-hydroxycholest-5-ene-7β-hydroperoxide (7β-OOH) (Girotti, 1998; Girotti and
Korytowski, 2000). The specificity of cholesterol with triplet oxygen and singlet oxy-
gen can differentiate between a singlet oxygen initiated reaction and a triplet oxygen
initiated reaction. Girotti and Korytowski (2000) described in detail the use of cho-
lesterol as an indicator of singlet oxygen.
Quenchers
Inhibition of singlet oxygen reaction by quenchers, such as carotenoids, tertiary amines,
tocopherols, histidine, 1,4-diazabicyclo[2,2,2]octane (DABCO), 2,5-dimethyl furan,
and azide, has been used to detect singlet oxygen (Jung et al., 1991; Song et al., 1999;
Bradley et al., 2006). All singlet oxygen quenchers have low ionization potential. Al-
though none of the quenchers are specific for singlet oxygen, the quenchers physically
or chemically interact with the singlet oxygen to inhibit oxidation. The calculation of
a singlet oxygen quenching rate by quantitative analysis can effectively determine the
characteristics of a quencher. The concentration that quenches half of the product can
be given by the expression:
[Q] = (ka[A] + kd)/kq
where ka is the rate of reaction with acceptor, [A] is the concentration of acceptor, kd
is the decay rate of singlet oxygen in the medium, and kq is the quenching rate of the
quencher. DABCO requires about 50 mM in aqueous solution to quench half of the
singlet oxygen formed (Bradley et al., 2006).
Azide and histidine are commonly used to determine the singlet oxygen oxidation
of compounds as these agents act as quenchers of singlet oxygen and greatly suppress
the activity and the consumption of singlet oxygen (Song et al., 1999). Singlet oxy-
gen quenching by azide is thought to be a charge transfer process in which molecular
triplet oxygen is released after the reaction, therefore no oxygen is consumed. Further
differentiation may be accomplished by using specific quenchers. Several quenching
agents have specificity towards reactive oxygen species. DABCO is also an efficient
scavenger of the hydroxyl radical. The azide ion is much more effective at quenching
singlet oxygen than DABCO, but it is also known as hydroxyl radical scavenger.
Fig. 1.6. Cholesterol oxidation products by singlet oxygen: 3β-hydroxy-5α-cholest-6-ene-5-hydroperoxide

(5α-OOH), 3β-hydroxycholest-4-ene-6α-hydroperoxide (6α-OOH), and 3β-hydroxycholest-4-ene-6β-
hydroperoxide (6β-OOH) and cholesterol oxidation products by triplet oxygen: 3β-hydroxycholest-5-ene-7α-
hydroperoxide (7β-OOH) and 3β-hydroxycholest-5-ene-7β-hydroperoxide (7β-OOH) (Girotti 1998).
Chemiluminence
Chemiluminenscence is produced when a chemical reaction yields an electronically
excited species that emits light as it returns to its ground state. The chemilumines-
cence method measures very weak luminescence like emission excited by the reaction
with singlet oxygen. Singlet oxygen is detected by its chemiluminescence at 1270
nm (Kanofsky and Axelrod, 1986; Macpherson et al., 1993; Darmanyan and Jenks,
1998). The energy differential between singlet oxygen and ground-state oxygen can
be released as a 1270 nm photon, which is very specific to singlet oxygen. The weak
emission from singlet oxygen is usually accompanied by light of other wavelengths in
complex systems. Although these emissions are assigned to higher vibrational states
of singlet oxygen, it seems likely that the extraneous emission comes from excited
carbonyls or other species in the system. Precise wavelength determination is essential
if this method is used for the detection of singlet oxygen.
ESR Spectroscopy
Electron spin resonance (ESR) spectroscopy is a highly sensitive analytical method
that detects the presence of free radicals. Although singlet oxygen is not a radical,
various amines can interact with singlet oxygen forming nitroxide radicals that are
readily detected by ESR. There is a high specificity of amines for singlet oxygen, and
the existence of a charge-transfer complex between the amine and singlet oxygen has
been proposed (Lion et al., 1976). The analytical method was developed in which
stable nitroxide radicals were generated by reaction with singlet oxygen. Amine com-
pounds, such as 2,2,6,6-tetramethly-4-piperidone (TMPD), can react with singlet
oxygen to form a stable nitroxide radical adduct, 2,2,6,6-tetramethyl-4-piperidone-
N-oxyl (TAN), which can be detected by ESR (Whang and Peng, 1988; Sharman
et al., 2000). When photosensitized oxidation occurs in the presence of the amine,
free nitroxide radicals will be produced. The reaction of TMPD with singlet oxygen
for the formation of TAN is shown in Fig. 1.7 (Bradley et al., 2003). Although other
reactive oxygen species, such as superoxide anion and hydroxy radical, can react with
TMPD, they do not convert TMPD to TAN. This method is highly specific to singlet
oxygen (Ando et al., 1997). Ando et al. (1997) further confirmed this specificity by
observing an inhibition of TAN formation upon the addition of two known singlet
oxygen scavengers, such as sodium azide and histidine, and not upon the addition of
a hydroxy radical scavenger, such as dimethyl sulfoxide, or a superoxide anion scaven-
ger, such as superoxide dismutase.
ESR spectroscopy detected the formation of singlet oxygen in meat (Whang and
Peng, 1988) and milk (Bradley et al., 2003) using a spin-trapping technique. Effects
of 0, 5, and 15 min illumination on an electron spin resonance spectrum of a TAN
radical in a water solution of riboflavin and TMPD is shown in Fig. 1.8 (Min and
Lee, 1996). Three hyperfine lines are the result of the coupling of the unpaired elec-
tron with an atom of nitrogen. The use of different spin-trapping compounds may be
utilized to differentiate reactive oxygen species by ESR spectroscopy. He et al. (1997)
O O
1
H3C CH3 O2 H3C CH3
N
:
H3 C CH3 H3 C N CH3
H O
.
TMPD TAN
Fig. 1.7. Reaction of 2,2,6,6-tetramethyl-4-piperidone with singlet oxygen to form 2,2,6,6,4-peperidone-n-oxyl
(Bradley et al. 2003).
used ESR to detect superoxide anion and hydroxyl radicals using 5,5-dimethyl-1-pyr-
roline-N-oxide as a spin-trapping agent.
The limitation of this method is that concentration of TAN should remain over
10-8 M for detection and over 10-6 M for good spectral resolution. Since the lifetime
of singlet oxygen is less than 1 microsecond, steady-state concentrations greater than
10-7 M are rarely maintained. Coupling spin-trapping with ESR spectroscopy could
improve this technique by diminishing the rate of disappearance.
Current singlet oxygen detection methods under investigation are the quantita-
tive determination using laser deflection calorimetry (Schneider et al., 2000) and
time-resolved singlet oxygen detection (Nonell and Braslavsky, 2000). Laser deflection
calorimetry is very selective in the determination of singlet oxygen in heterogeneous
systems (Schneider et al., 2000). Future advances in time-resolved singlet oxygen
detection should include the detection of photosensitized and non-photosensitized
production of singlet oxygen in heterogeneous systems such as foods (Nonell and
Braslavsky, 2000). Andersen and Ogilby (2001) recorded the time-resolved absorp-
tion spectrum of singlet oxygen using transmission microscopy.
Reaction of Singlet Oxygen with Fatty Acids

The oxidation rate of food depends on temperature, the presence of inhibitors or cata-
lysts, the nature of the reaction environment, and the nature of the compounds. These
factors are important in varying degrees to both singlet and triplet oxygen oxidation
in foods. Temperature has little effect on singlet oxygen oxidation but has a significant
effect on triplet oxygen oxidation, which requires high activation energy. Polyunsatu-
rated fatty acids are more susceptible to radical-initiated triplet oxygen oxidation than
monounsaturated fatty acids, a property that is primarily due to the lowered activa-
tion energy in the initiation of free-radical formation in polyunsaturated fatty acids
compared to monounsaturated fatty acids. The type of polyunsaturated fatty acids
is not important in singlet oxygen oxidation. The total numbers of double bonds is
more important in singlet oxygen oxidation than the types of double bonds, such as
nonconjugated or conjugated dienes or trienes, which are important in free-radical
triplet oxygen oxidation.
Electrophilic singlet oxygen is seeking electrons to fill its highest degenerate va-
cant molecular orbital (Fig. 1.2). One of the most important reaction characteristics
of singlet oxygen is that it can directly react with the electron-rich double bonds of
unsaturated molecules (Adam, 1975; Beutner et al., 2000).
Singlet oxygen participates in reactions such as 1,4-cycloaddition to diene and
heterocyclic compounds, “ene” reaction, and 1,2-cycloaddtion to olefins. All involve
0 m in
5 m in
15 m in
3370 G 3390 G 3410 G
Fig. 1.8. Effects of 0, 5, and 15 minutes illumination on electron spin resonance spectrum of 2,2,6,6-4-piperi-
done-n-oxyl in water solution of riboflavin and 2,2,6,6-tetramethyl-4-piperidone (Min and Lee 1996).
direct reaction with double bonds as shown in Fig. 1.9. When singlet oxygen reacts
with linoleic or linolenic acid, they form both conjugated and nonconjugated diene
hydroperoxides (Fig. 1.10). The linoleic or linolenic acid reacts with triplet oxygen
and produces only conjugated diene hydroperoxides (Fig. 1.10). The direct reaction
of singlet oxygen with double bonds permits the formation of hydroperoxides at posi-
tions 10 and 12 in linoleic acid and 10 and 15 in linolenic acid, which do not form in
triplet oxygen oxidation as shown in Table 1.3 (Frankel et al., 1979). These properties
may be used to determine singlet oxygen activity in lipids (Stratton and Lieber, 1997)
and produce compounds in the absence of triplet oxygen oxidation.
The reaction rates of singlet oxygen with oleic, linoleic, linolenic, and arachi-
donic acids are 0.74, 1.3, 1.9, and 2.4 × 105 M-1S-1, respectively. The rate is rel-
atively proportional to the number of double bonds in the fatty acid instead
of the type of double bond, such as conjugated or nonconjugated double bonds,
(Doleiden et al., 1974). The reactivity increases as the ionization energy de-
creases due to the presence of alkyl groups adjacent to the C=C double bond.
However, steric hindrance to the double bond will lower the reactivity (Beutner et
al., 2000).
1,4-Cycloaddition
O O
+
O O
Endoperoxide
H “ene” Reaction
O O
+
O O
Allyl Hydroperoxide
1,2-Cycloaddition
O CH2 O
+
O CH2 O
Dioxetane
Fig. 1.9. Reactions of singlet oxygen with olefins by 1,4-cycloaddition, “ene” reaction, and 1,2-cycloaddtion
(Min and Boff 2002).
OOH
1
O2 R R’
OOH
R R’ R O R’
H O
R R’
3
O2
OOH
− H· · 3
O2
9
R R’ R R’ R R’
Fig. 1.10. Conjugated and nonconjugated hydroperoxide formation from a diene fatty acid by singlet oxygen
and triplet oxygen.
Diradical triplet oxygen can react with radical food compounds. However, food
compounds are not radical compounds and they should be in a radical state to react
with triplet oxygen. The initiation of radical formation in food occurs at the carbon
that requires the least energy to remove a hydrogen atom. The removal of hydro-
gen from a saturated fatty acid requires approximately 100 kcal/mol of energy. The
carbon-hydrogen bond on carbon 8 or 14, which is α to the double bond of linoleic
acid is about 75 kcal/mole. Hydrogen at position 11 of linoleic acid is most easily
removed due to the presence of a double bond on both sides and requires only about
50 kcal/mol. The various strengths of hydrogen-carbon bond of fatty acids explain the
differences of oxidation rates of stearic, oleic, linoleic, and linolenic acids during oxi-
dation by triplet oxygen. Heat, light, metals, and reactive oxygen species facilitate the
radical formation of food components. Once the hydrogen is removed, a pentadienyl
radical intermediate between carbon 9 and carbon 12 of linoleic acid is formed. The
pentadienyl radical provides an equal mixture of conjugated 9- and 13-diene radical
and produces 9- and 13-conjugated diene hyroperoxides upon reaction with triplet
oxygen (Table 1.3).
Triplet oxygen autoxidation produces only the conjugated diene hydroperoxides
in linoleic and linolenic acids. The relative reaction ratios of triplet oxygen with oleic,
linoleic, and linolenic acid for hydroperoxide formation are 1:12:25, which is depen-
dent on the relative difficulty for the radical formation in the molecule (Min and Lee,
1996). The reaction rate of triplet oxygen with linolenic acid is about twice as fast as
that of linoleic acid because linolenic acid has 2 pentadienyl groups in the molecule,
compared to linoleic acid with 1 pentadienyl group.
Singlet oxygen oxidation produces hydroperoxides at the positions of double

bonds without migration (Table 1.3). Gunston (1986) compared triplet oxygen
oxidation and photosensitized singlet oxygen oxidation rate of oleic, linoleic, and
linolenic acids as shown in Table 1.4. The reaction rate of singlet oxygen with oleic,
linoleic, and linolenic acid is about 1,000~30,000 times faster than that of the triplet
oxygen.
Singlet Oxygen Oxidation in Foods

The effect of light energy has long been known to play a role in the flavor stability of
vegetable oils and other fat-containing products, such as margarine, butter, and may-
onnaise. Most food pigments and colorants are potential initiators of singlet oxygen
oxidation in foods, due to their ability to absorb light in the visible light range, exhibit
fluorescence and phosphorescence, reflecting both a singlet and triplet state, and have
a high quantum yield of a long-lived triplet state. Photosensitizers include synthetic
Table 1.3. Hydroperoxides of Fatty Acids Formed by Triplet and Singlet Oxygen Oxidation
Oleate Linoleate Linolenate
8-OOH
9-OOH
10-OOH
3
O2 11-OOH
9-OOH
Conjugated hydroper- 9-OOH
12-OOH
oxides 13-OOH
13-OOH
16-OOH
9-OOH
10-OOH
9-OOH
Conjugated hydroper- 9-OOH
1
O2 12-OOH
oxides 13-OOH
13-OOH
16-OOH
Nonconjugated hvdro- 10-OOH 10-OOH

peroxides 12-OOH 15-OOH
Frankel et al., (1979)
Table 1.4. Relative Oxidation Rates of Triplet and Singlet Oxygen with Oleate, Linoleate, and Linolenate
Oleate Linoleate Linolenate
Triplet Oxygen 1 27 77
Singlet Oxygen 30,000 40,000 70,000

CH3–(CH2)4–CH=CH–CH2–CH=CH–CH2–(CH2)6–COOH
+ 1O2
CH3–(CH2)4–CH=CH–CH2–CH–CH=CH–(CH2)6–COOH
O
O
H
CH3–CH2–CH2–CH2–CH2–CH=CH–CH2–CH
O
+ 1O2
CH3–CH2–CH2–CH2–CH2–CH–CH–CH2–CH
O O O
CH3–CH2–CH2–CH2–CH2–CH–CH–CH2–CH
O ··
O O
+ 2RH
CH3–CH2–CH2–CH2–CH2–CH–CH2–CH2–CH + 2R ·
O OH O
CH3–CH2–CH2–CH2–CH2–CH–CH2–CH2–CH
O
· O
CH3–CH2–CH2–CH2–CH2–CH–CH2–CH2–CH
O O
CH3–CH2–CH2–CH2–CH2–C=CH–CH=CH
OH OH
− H2O
CH3–CH2–CH2–CH2–CH2–
O
Fig. 1.11. Mechanism for the formation of 2-pentyl furan from linoleic acid by singlet oxygen (Min et al. 2003).
CH3–CH2–CH=CH–CH2–CH=CH–CH2–CH=CH–CH2–(CH2)6–COOH
+ 1O2
CH3–CH2–CH=CH–CH2–CH=CH–CH2–CH-CH=CH–(CH2)6–COOH
O
O
H
CH3–CH2–CH=CH–CH2–CH=CH–CH2–CH-CH=CH–(CH2)6–COOH
O
·
CH3–CH2–CH=CH–CH2–CH=CH–CH2–CH
O
+ 1O2
CH3–CH2–CH=CH–CH2–CH-CH–CH2–CH
O O ·
O
H
CH3–CH2–CH=CH–CH2–CH-CH–CH2–CH
O ·
O
·
CH3–CH2–CH=CH–CH2–C-CH2–CH2–CH
O O
CH3–CH2–CH=CH–CH2–C-CH2–CH2–CH
OH OH
− H2O
CH3–CH2–CH=CH–CH2–
O
Fig. 1.12. Mechanism for the formation of 2-pentenyl furan from linolenic acid by singlet oxygen (Min et al. 2003).
dyes; naturally occurring pigments, such as chlorophyll, flavin, and porpyrin; coen-
zymes; and biochemical compounds, such as pyridoxals and psoralens; metallic salts;
and transition metal complexes, such as ruthenium bipyridine.
Photooxidation of vegetable oils is a major concern of the food industry, in that
they contain natural photosensitizers and are commercially sold under light. Salad
dressings containing 30-40% vegetable oil account for 35% of the production of all
dressings, mayonnaise, and sandwich spreads. Soybean oil has a 90% share of the
prepared dressings market. Not only is soybean oil susceptible to oxidation due to
the high concentration of linoleic acid, but it contains 1-5 ppm chlorophyll (Brekke,
1980), which is an excellent singlet oxygen sensitizer. Min and Lee (1988) reported
that headspace volatile compounds of soybean oil increased as added chlorophyll in-
creased from 0, 2, 4, and 6 ppm to 8 ppm. Soybean oil containing no chlorophyll,
which was removed by silicic acid liquid column chromatography, did not produce
headspace volatile compounds under light under identical experimental conditions at
10°C. However, the effects of 0, 2, 4, 6, and 8 ppm added chlorophyll did not have
any effect on the formation of volatile compounds of soybean oil in dark storage. The
formation of headspace volatile compounds in the soybean oil decreased inversely
with the amount of added β-carotene, which quenches singlet oxygen (Lee and Min,
1990). The very rapid formation of volatile compounds in the soybean oil in the pres-
ence of chlorophyll, light, and oxygen was due to the singlet oxygen oxidation.
The oxidative deterioration of virgin olive oil sold as an unrefined green liquid is
related to the amount of chlorophyll contained in the oil. Olive oil in its natural state
contains chlorophyll, carotenes, tocopherols, and other phenolic compounds. The
presence of 6 ppm chlorophyll acted as a photosensitizer, resulting in rapid oxidation
of the oil during exposure to fluorescent light. The presence of β-carotene and nickel
chelates both substantially inhibited oxidation in the first hours of illumination, thus
supporting the concept that chlorophyll brings about formation of singlet oxygen,
which is quenched by β-carotene and nickel chelates.
Chlorophylls and their decomposition products in vegetable oils are potential
photosenstizers generating singlet oxygen in the presence of light and triplet oxygen.
Singlet oxygen rapidly reacts with unsaturated fatty acids to produce a mixture of
conjugated and nonconjugated hydroperoxides that rapidly decompose to produce
undesirable flavor compounds. Most studies have been done in model systems that
consist of one or two free fatty acids exposed to light in the presence of a sensitizer.
Reversion Flavor in Soybean Oil

Soybean oil represents about 70% of all edible fats and oils consumed in the United
States (Golbitz, 2000). Soybean oil is inexpensive and widely available compared to other
edible oils. The development of reversion flavor, described as beany or grassy, is a unique
defect to soybean oil and can be formed in soybean oils, which have low peroxide values
(Ho et al., 1978). To improve the flavor stability and quality of soybean oil, reversion
flavor has been extensively studied in soybean oil since 1936 (Ho et al., 1978).
Smouse and Chang (1967) identified 2-pentyl furan in reverted soybean oil and
reported that it significantly contributed to the reversion flavor of soybean oil. Chang
et al. (1983) isolated and identified all 4 2-pentenyl-furan isomers in reverted soybean
oil.
Sensory evaluation showed that the addition of 2 ppm 2-pentyl furan to freshly
deodorized and bland soybean oil produced the “reverted” soybean oil flavor. The
addition of 2 ppm 2-pentyl furan to deodorized cottonseed oil and corn oil also pro-
duced reversion flavor found in reverted soybean oil (Chang et al., 1966). Ho et al.
(1978) reported that 2-(1-pentenyl) furan contributed to reversion flavor. Smagula et
al. (1978) reported that 2-(2-pentenyl) furan is also a contributor according to sen-
sory evaluation. Flavor thresholds of the 2-pentenyl furan isomers were between 0.25
and 6 ppm.
Smouse and Chang (1967) and Ho et al. (1978) proposed the mechanisms for
the formation of 2-pentyl furan from linoleic acid and 2-pentenyl furan isomers from
linolenic acid using triplet oxygen, respectively. The proposed mechanisms for the
formation of 2-pentyl furan from linoleic acid and isomers of 2-pentenyl furan from
linolenic acid by triplet oxygen have been questioned. The formation of both the 2-
pentyl furan and 2-pentenyl furan requires a hydroperoxide at carbon 10 of linoleic
acid. The formation of 10-hydroperoxide in linoleic or linolenic acids by free-radical
triplet oxygen oxidation is highly improbable but is very common in the singlet oxy-
gen oxidation of linoleic or linolenic acids as shown in Table 1.2. Min et al. (2003) re-
ported on the chemical mechanisms for the formations of 2-(2-pentenyl) furan from
linoleic acid and 2-pentyl furan formation from linolenic acid using singlet oxygen as
shown in Figs 1.11 and 1.12, respectively.
Min et al. (2003) identified 2-pentylfuran and 2-pentenyl furan in soybean oil
containing 5 ppm chlorophyll b during storage under light for 96 hrs. The formation
of 2-pentyl and 2-pentenyl furan increased with increasing storage time and concen-
tration of chlorophyll. 2-Pentyl furan and 2-pentenyl furan were formed only in the
presence of light and chlorophyll in soybean oil. Soybean oil containing 5 ppm chlo-
rophyll did not produce 2-pentyl furan and 2-pentenyl furan during dark storage. The
chlorophyll-free soybean oil obtained by silicic acid chromatography did not produce
either 2-pentyl furan or 2-pentenyl furan during light storage. The singlet oxygen oxi-
dation reaction is involved in the formation of 2-pentyl furan and 2-pentenyl furan,
which have been reported to be mainly responsible for reversion flavor in soybean oil.
The chlorophyll, which is an excellent photosensitizer, in soybean oil should be care-
fully removed during the oil processing to minimize the formation of reversion during
storage.
Singlet Oxygen Quenching Mechanisms

Other than exclusion of light and reduction of oxygen present, the use of quenching
agents is the best way to reduce singlet oxygen oxidation. Natural food components,
such as tocopherols, carotenoids, and ascorbic acid, can act as effective singlet oxygen
quenchers (Lee et al., 2004). Quenching agents may be involved to minimize the
development or activity of singlet oxygen at several stages in the oxidation of foods.
Figure 1.13 shows the development of singlet oxygen and its subsequent reaction with
substrate (A) to form the oxidized product (AO2).
At every stage in this reaction, there is at least one alternate route, which, if
taken, would minimize the oxidation of the compound (A). The first step represents
3
hv 1
KISC 3
O2 1 A
Sen Sen* Sen* O2 AO2
ko kr
Q kQ Q kq kd
Q kox-Q
3 3
O2 O2
1
Sen Sen QO2
Fig. 1.13. Formation of singlet oxygen and its reaction with substrate a to produce the oxidized product AO2.
the return of the 1Sensitizer* to 1Sensitizer without intersystem crossing which can
form the3Sensitizer*. The second represents a reaction with a quencher (Q) at a rate
KQ, returning the 3Sensitizer* to 1Sensitizer prior to reaction with triplet oxygen. The
3
Sensitizer* may react with triplet oxygen (3O2) to form singlet oxygen (1O2). Follow-
ing its creation, there are three fates for singlet oxygen in food: (1) it may naturally
decay to the ground state at a rate kd; (2) it may react with a singlet state compound
(A) at a rate kr forming the oxidized product (AO2); and (3) it may be destroyed by a
quencher, either chemically at a rate kox-Q to form the product QO2, or physically at a
rate kq to return to free triplet oxygen.
There are three points at which a quencher may act (Fig. 1.13): one is quenching
of the 3Sensitizer*, and the other two are quenching of singlet oxygen by either chemi-
cally or physically. Chemical quenching involves the reaction of singlet oxygen with
the quencher to produce an oxidized product (QO2). Physical quenching results in
the return of singlet oxygen to triplet oxygen without the consumption of oxygen or
the formation of oxidized products by either energy transfer or charge transfer. There-
fore, triplet oxygen quenchers must either be able to donate electrons or to accept
22.4 kcal above ground state. An example of the latter is β-carotene which has a low
singlet energy state and can therefore accept the energy from singlet oxygen (Lee and
Min, 1990). Ascorbic acid is an example of a chemical that can quench the excited
sensitizer. Table 1.5 lists quenching rates of several antioxidants.
Quenching Mechanism of Carotenoids

Carotenoids, which are responsible for the yellow and red colors of many plants and
Table 1.5. Singlet Oxygen Quenchers and Their Quenching Rates

Quenching Compound Quenching Rate (M-1sec-1)
β-Apo-8’-carotenal 3.1 × 109
β-Carotene 4.6 ×109
Lutein 5.7 ×109
Zeaxanthin 6.8 ×109
Lycopene 6.9 ×109
Isozeqaxanthin 7.4 ×109
Astaxanthin 9.9 ×109
Canthaxanthin 11.2 ×109

α-Tocopherol 2.7 ×107
1,4-Diazabicyclo-(2,2,2)-octane 1.5 ×107
Dimethylfuran 2.6 × 107
Bis(di-n-butyldithiocarbamato)nickel chelate 1.2 × 109
{2,2’-Thiobis(4-1,1,3,3,-tetramethylbutyl) phenalto)}-n-
3.7 × 107
butylamine)nickel chelate
Min and Lee (1996), Li et al., (1998)
animal products, have been known to minimize singlet oxygen oxidation (Forse,
1979; Lee and Min, 1990). Carotenoids include a class of hydrocarbons called caro-
tenes and their oxygenated derivatives called xanthophylls. Foote (1976) found that
one molecule of β-carotene can quench 250 to 1000 molecules of singlet oxygen at a
rate of 1.3 × 1010 M-1S-1.
Energy transfer mechanism is responsible for the minimization of singlet oxygen
oxidation of lipids by β-carotene (Forss, 1979). Electron excitation energy is trans-
ferred from singlet oxygen to singlet state carotenoid (1CAR), producing triplet state
carotenoid (3CAR) and triplet oxygen, which is called singlet oxygen quenching. En-
ergy is also transferred from 3Sensitizer* to the 1CAR, which is called triplet sensitizer
quenching. The 3CAR can easily return to the 1CAR dissipating energy as a heat.
1
O2 + 1CAR → 3O2 + 3CAR
1
CAR + 3Sensitizer* → 3CAR + 1Sensitizer
3
CAR → 1CAR
The energy transfer from singlet oxygen (22.4 kcal/mole) to carotenoids with nine
or more conjugated double bonds (<22.4 kcal/mole) is exothermic (Foote, 1970).
Foote (1970) reported that the carotenoid with seven conjugated double bonds was
effective at quenching triplet chlorophyll. Carotenoids with fewer than nine conju-
gated double bonds have energies above that of singlet oxygen and are less efficient
singlet oxygen quenchers. Carotenoids with eleven or more conjugated double bonds
quench at a diffusion-controlled rate of singlet oxygen.
The rate of singlet oxygen quenching by carotenoids is dependent on the number
of conjugated double bonds and the type and number of functional groups on the ring
portion of the molecule. This is important in the solubility of carotenoids. Kobayashi
and Sakamoto (1999) compared the quenching activity of β-carotene to astaxan-
thin and found that the quenching activity of astaxanthin decreased with increasing
hydrophobicity, while the quenching activity of β-carotene increased. Lee and Min
(1990) evaluated the effectiveness of five carotenoids including lutein, zeazanthin,
lycopene, isozeaxanthin, and astaxanthin in quenching chlorophyll-photosensitized
oxidation of soybean oil and reported that the effectiveness increased with the num-
ber of double bonds in the carotenoid and the concentration of carotenoid added.
Quenching Mechanisms of Tocopherols

Tocopherols are the most abundant and prevalent antioxidants in nature and are stud-
ied as free-radical scavengers. When present in systems that are vulnerable to singlet
oxygen oxidation, tocopherols inhibit lipid peroxidation. Tocopherols were identified
in soybean oil at about 1100 ppm and exist in α-, β-, γ-, and δ-tocopherol at approxi-
mately 4, 1, 67, and 29%, respectively (Jung et al., 1991). Jung et al. (1991) stud-
ied the effectiveness of α-, γ-, and δ-tocopherol in the chlorophyll-photosensitized
oxidation of soybean oil and reported that α-tocopherol quenched singlet oxygen
at the rate of 2.7 ×107 M-1S-1. Tocopherols involve charge transfer as singlet oxygen
quenchers (Foote, 1970). Tocopherols can form a charge transfer complex with sin-
glet oxygen by electron donation from tocopherol to singlet oxygen. The transfer
complex undergoes an intersystem crossing to form triplet oxygen and tocopherol.
Since this does not involve chemical reactions between tocopherol and singlet oxygen,
it is called physical quenching:

T + 1O2 → [T+−1O2]1 → [T+−1O2]3 → T + 3O2
The destruction of vitamin D2 in a model system by singlet oxygen oxidation was

reduced by the addition of α-tocopherol (King and Min, 1998). The rate of singlet
oxygen quenching by α-tocopherol is similar to that of β-carotene.
Determining Quenching Mechanisms

The quenching mechanism of photosensitized singlet oxygen oxidation can be deter-
mined by measuring the rate constant of total quenching, physical quenching, and
chemical quenching. Quenchers work in numerous ways to inhibit the formation of
oxidized products as has been previously described (Fig. 1.13).
The quantum yield of a photochemical reaction is defined as the ratio of the
number of molecules of a product formed to the number of photons of light ab-

sorbed. This value is used to measure the relative efficiency of a photochemical reac-
tion. The quantum yield of oxidized product formation (ØAO2) can be defined by the
equation:
ØAO2 = A × B × C (Eq.1)
where A represents the partitioning of singlet sensitizer for singlet oxygen oxidation;
B, the partitioning of triplet sensitizer for singlet oxygen formation; and C, the for-
mation of the oxidized product.
The concentration of quencher necessary to inhibit a substantial amount of the
singlet oxygen sensitizer is particularly high since the lifetime of singlet oxygen sensi-
tizer is very short. For these reasons, the singlet sensitizer quenching is not considered
in the steady-state equation. Therefore, term A in Eq. 1 is a constant (K) that is equal
to the quantum yield of intersystem crossing.
Term B in Eq. 1 represents the rate of singlet oxygen formation, which is dependent
on the triplet sensitizer quenching rate and the rate of triplet-triplet annihilation.
Therefore, B is:
ko [oxygen]
B= (Eq. 2)
ko [oxygen] + kQ [quencher]
where ko is the reaction rate constant of triplet-triplet annihilation and kQ is the reac-
tion rate constant of triplet sensitizer quenching (Fig. 1.13).
Term C in Eq. 1 represents the formation of oxidized product, which is depen-
dent on the concentration and nature of the compound, the physical and chemical
quenching of singlet oxygen, as well as the natural decay rate of singlet oxygen. The
assemblage of these factors generates the following equation:
kr [substrate]
C= (Eq. 3)
kr [substrate] + (kox-Q + kq) [chemical + physical quencher] + kd
where kr is the reaction rate constant of the reaction between singlet oxygen and the
substrate, kox-Q is the reaction rate constant of chemical quenching, kq is the reaction
rate constant of physical quenching, and kd is the decay rate constant of singlet oxy-
gen.
If a quencher inhibits photosensitized oxidation by quenching singlet oxygen, the
steady state equation can be written as:
ko [3O2] kr [A]
ØAO2 = K 3 × (Eq. 4)
ko [ O2] + kQ [Q] kr [A] + (kox-Q + kq) [Q] + kd
where K is the quantum yield of intersystem crossing of the 1Sensitizer* (A from Eq.
1), and both B and C have been appropriately substituted with Eq. 2 and Eq. 3,
respectively.
In a given system, if there is only singlet oxygen quenching such that kQ [Q] <<
ko [3O2], then B is equal to 1. Therefore, the steady state equation becomes:
kr [A]
ØAO2 = K (Eq. 5)
kr [A] + (kox-Q + kq) [Q] + kd
This equation can be inverted to:

(kox-Q + kq) [Q] + kd
[ØAO2]-1 = K-1 1 + [A]-1 (Eq. 6)
kr
so that it is in slope-intercept form.

Alternatively, if there is only triplet sensitizer quenching, such that (kox-Q + kq) [Q]
<< kr [A] + kd, then the slope-intercept form of the equation is:
kQ [Q] kd
[ØAO2]-1 = K-1 1 + 3 1+ (Eq. 7)
ko [ O2] kr [A]
The significance of these two equations is the fact that one describes a system in which
singlet oxygen quenching is dominant, and the other describes a system in which trip-
let sensitizer is dominant. A plot of [AO2]-1 against [A]-1 at different [Q] will appear
in one of two ways, dependent on which mechanism dominates a system. If singlet
oxygen quenching is dominant (Eq. 6), then the plots at various [Q] will all have
the same y-intercept but different slopes (Fig. 1.14). If triplet sensitizer quenching is
dominant (Eq. 7), then both the intercept and the slope will vary (Fig. 1.15).
-1 (kox-Q +kq)[Q] + kd
Slope = K
kr
Intercept = K-1
[AO2]-1 � [Q3]
�
• [Q2]
� •
• � [Q1]
�
�
-1
K •
[A]-1
Fig. 1.14. Characteristics plot of a singlet oxygen quenching mechanism (Li et al. 2000).
[AO2]-1
kd (ko [3O2] + kq [Q])

Slope = K-1
ko [3O2] kr
ko [3O2] + kq [Q]
Intercept = K-1 [Q3]
ko [3O2] ž
ž
• [Q2]
ž
•
[Q1]
♦
• ♦
ž ♦
•
♦
[A]-1
Fig. 1.15. Characteristic plot of a triplet sensitizer quenching mechanism (Li et al. 2000).
References
Adams, D.R.; and Wilkinson, F. Lifetime of Singlet Oxygen in Solution. J. Chem. Soc. Faraday
Trans. 2, 1972, 68, 686.
Adam, W. Singlet Molecular Oxygen and Its Role in Organic Peroxide Chemistry, Chem. Ztg.
1975, 99, 142.
Andersen, L.K.; and Ogilby, P.R. Time-Resolved Detection of Singlet Oxygen in a Transmis-
sion Microscope. Photochem. Photobiol. 2001, 73, 489.
Afonso, S.G.; Fnriquez de Salamanca, R.; and del C Baflle, A. M. The Photodynamic and
Non-Photodynamic Actions of Porphyrins. Braz. J. Med. Biol. Res. 1999, 32, 255.
Ando, T.; Yoshikawa, F.; Tanigawa, T.; Kohno, M.; Yoshida, N.; and Motoharu, K. Quantifica-
tion of Singlet Oxygen from Hematoporphyrin Derivative by Electron Spin Resonance.
Life Sci. 1997, 61, 1953.
Athar, M.; Mukhtar, H.; and Bickers, D.R. Differential Role of Reactive Oxygen Intermediates
in Photofrin-I-Initiated and Photofrin-II-Initiated Photocenhancement of Lipid Peroxi-
dation in Epidermal Microsomal Membranes. J. Invest. Derm. 1988, 90, 652.
Beutner, S.; Bloedorn, B.; Hoffman, T.; and Martin, H.D. Synthetic Singlet Oxygen Quench-
ers. Methods in Enzymology, Packer, L., Sies, H. Eds., Academic Press: New York, 2000,
Vol 319, 226-241.
Bradley, D.G.; Lee, H.O.; and Min, D.B. Singlet Oxygen Detection in Skim Milk by Electron
Spin Resonance Spectroscopy. J. Food Sci. 2003, 68, 491.
Bradley, D.G.; Kim, H.J.; and Min, D.B. Effects, Quenching Mechanism, and Kinetics of
Water Soluble Compounds in Riboflavin Photosensitized Oxidation of Milk. J. Agric.

Food Chem. 2006, 54, 6016-6020.
Brekket, O.L. In Handbook of Soyoil Processing and Utilization, Erickson, D.R., Pryde, E.H.,
Brekke, O.L., Mounts, T.L., Falb, R.A., Eds., American Oil Chemists’ Society: Cham-
paign, IL, 1980, 109.
Chang, S.S.; Shen, G.H.; Tang, H.; Jin, Q.Z.; Shi, H.; Carlin, J.T.; and Ho, C.T. Isolation
and Identification of 2-Pentenylfurans in the Reversion Flavor of Soybean Oil. J. Am. Oil
Chem. Soc. 1983, 60, 553.
Chang, S.S.; Smouse, T.H.; Krishnamurthy, R.G.; Mookherjee, B.D.; and Reddy, R.B. Isola-
tion and Identification of 2-Pentyl-furan as Contributing to the Reversion Flavour of
Soybean Oil. Chem. Ind. (London), 1966, 1926.
Choe, E.; and Min, D.B. Chemistry and Reactions of Reactive Oxygen Species in Foods. J.
Food Sci. 2005, 70, 142.
Choe, E.; and Min, D.B. Mechanisms and Factors for Edible Oil Oxidation. Comp. Rev. Food
Sci. Food Saf. 2006, 5, 169.
Darmanyan, A.P.; and Jenks, W.S. Charge-Transfer Quenching of Singlet Oxygen by Amines
and Aromatic Hydrocarbons. J. Phys. Chem. 1990, 109, 7420-7426.
Delmelle, M. Retinol Damage by Light: Possible Implication of Singlet Oxygen. Biophs. Struct.
Mech. 1997, 3, 195-198.
Doleiden, F.H.; Farenholtz, S.R.; Lamola, A.A.; and Rwozzolo, A.M. Reactivity of Cholesterol
and Some Fatty Acids Toward Singlet Oxygen. Photochem. Photobiol. 1974, 20, 519.
Foote, C.S.; and Denny, R.W. Chemistry of Singlet Oxygen. VII Quenching by β-Carotene.
J. Am. Chem. Soc. 1968, 90, 6233.
Foote, C.S.; Chang, Y.C.; and Denny, R.W. Chemistry of Singlet Oxygen. X. Carotenoid
Quenching Parallels Biological Protection. J. Am. Chem. Soc. 1970, 92, 5216.
Foote, C.S. Photosensitized Oxidation and Singlet Oxygen: Consequences in Biological Sys-
tems. Free Radicals in Biology, Pryor, W.A., Ed., Academic Press: New York, 1976, 85.
Forss, D.A. Mechanism of Formation of Aroma Compounds in Milk and Milk Products. J.
Dairy Res. 1979, 46, 691.
Frankel, E.N.; Neff, W.E.; and Bessler, T.R. Analysis of Autoxidized Fats by Gas Chromatogra-
phy-Mass Spectrometry: V. Photosensitized Oxidation. Lipids 1979, 14, 961.
Frankel, E.N.; Neff, W.E.; and Selke, E. Analysis of Autoxidized Fats by Gas Chromatogra-
phy-Mass Spectrometry: VII. Volatile Thermal Decomposition Products of Pure Hydro-
peroxides from Autoxidized and Photosensitized Oxidized Methyl Oleate, Linoleate, and
Linolenate. Lipids 1981, 16, 279.
Frankel, E.N. Chemistry of Autoxidation: Mechanism, Products and Flavor Significance. Fla-
vor Chemistry of Fats and Oils, Min, D.B., and Smouse, T.H., Eds., American Oil Chem-
ists’ Society: Champaign, IL, 1985, 1.
Girotti, A.W. Lipid Hydroperoxide Generation, Turnover, and Effector Action in Biological
Systems. J. Lipid Res. 1998, 39, 1529.
Golbitz, P. Soya and Oilseed Bluebook. Soyatech, Inc., Bay Harbor, Maine, 2000, 2001.
Gunston, F.D. Chemical Properties. The Lipid Handbook, Gunston, F.D., Harwood, J.L., Pad-
ley, F.B., Eds., Chapman and Hall, New York, 1986 449-484.
Halliwell, B.; and Gutteridge, J.M.C. Free Radicals in Biology and Medicine. 3rd ed.; Oxford
University Press: New York, 2001.
He, Y.Y.; An, J.Y.; and Jiang, L.J. EPR and Spectrophotometric Studies on Free Radicals (O2*,
Cysa-HB*) and Singlet Oxygen (1O2) Generated by Irradiation of Cysteamine Substi-
tuted Hypocrellin B. Int. J. Rad. Biol. 1998, 74, 647.

Ho, C.T.; Smagula, M.S.; and Chang, S.S. The Synthesis of 2-(1-Pentenyl) furan and Its Rela-
tionship to the Reversion Flavor of Soybean Oil. J. Am. Oil Chem. Soc. 1978, 55, 233.
IUPAC Commission on Nomenclature of Inorganic Chemistry, Names for Muonium
and Hydrogen Atoms and their Ions. Pure Applied Chem. 2001, 73, 377–380.
Jung, Y.J.; Lee, E.; and Min, D.B. α-, γ-, and δ-Tocopherol Effects on Chlorophyll Photosen-
sitized Oxdiation of Soybean Oil. J. Food Sci. 1991, 45, 183.
Kanofsky, J.R.; and Axelrod, B. Singlet Oxygen Production by Soybean Lipoxygenase Iso-
zymes. J. Biol. Chem. 1986, 261, 1099.
Ke, P.J.; and Ackman, R.G. Bunsen Coefficient for Oxygen in Marine Oils at Various Tem-
peratures Determined by Exponential Dilution Method with a Polarographic Oxygen
Electrode. J. Am. Oil Chem. Soc. 1973, 50, 429.
Kearns, D.R. Solvent and Solvent Isotope Effects on the Lifetime of Singlet Oxygen. Singlet
Oxygen, Wasserman, H.H., Murray, R.W. Eds., Academic Press: New York, 1979, 115.
Kellogg, E.W.; and Fridovich, I. Suoperoxide, Hydrogen Peroxide, and Singlet Oxygen in
Lipid Peroxidation by a Xanthin Oxidase System. J. Biol. Chem. 1975, 250, 8812-8817.
Kepka, A.; and Grossweiner, L.I. Photodynamic Oxidation of Iodide Ion and Aromatic Amino
Acids by Eosine. Photochem. Photobiol. 1972, 14, 621.
King, J.M.; and Min, D.B. Riboflavin Photosensitized Singlet Oxygen Oxidation of Vitamin
D. J. Food Sci. 1998, 63, 31.
Kobayashi, M.; and Sakamoto, Y. Singlet Oxygen Quenching Ability of Astaxanthan Esters
from the Green Alga Haematococcus pluvialis, Biotech. Lett. 1999, 21, 265.
Kochevar, I.E.; and Redmond, R.W., Photosensitized Production of Singlet Oxygen. Methods
in Enzymology, Packer, L., and Sies, H., Eds., Academy Press: New York, 2000, 20.
Korycka-Dahl, M.B.; and Richardson, T. Activated Oxygen Species and Oxidation of Food
Constituents. Crit. Rev. Food Sci. Nutr. 1978, 10, 209.
Labuza, T.P. Kinetics of Lipid Oxidation in Foods, CRC Crit. Rev. Food Sci. Nutr. 1971, 2,
355.
Lledias, F.; and Hansberg, W. Catalase Modification as a Marker for Singlet Oxygen. Methods
in Enzymology, Parks, O.W. and Sies, H., Eds., Academic Press: New York, 2000, 110.
Lee, S.H.; and Min, D.B. Effects, Quenching Mechanisms, and Kinetics of Carotenoids in
Chlorophyll-Sensitized Photooxidation of Soybean Oil. J. Agric. Food Chem. 1990, 38,
1630.
Lee, J.; Koo, N.; and Min, D.B. Reactive Oxygen Species, Aging, and Antioxidative Netraceu-
ticals. Comp. Rev. Food Sci. Food Saf. 2004, 3, 21-33.
Lledias, F.; and Hansberg, W. Catalase Modification as a Marker for Singlet Oxygen. Methods
in Enzymology, Parks, O.W. and Sies, H., Eds., Academic Press: New York, 2000, 110.
Li, T.L.; King, J.M.; and Min, D.B. Quenching Mechanisms and Kinetics of Carotenoids in
Riboflavin Photosensitized Singlet Oxygen Oxidation of Vitamin D2. J. Food Biochem.
2000, 24, 477.
Macpherson, A.N.; Teflwer, A.; Barber, J.; and Truscott, T.G. Direct Detection of Singlet Oxy-
gen from Isolated Photosystem II Reaction Centers. Biochem. Biophys. Acta. 1993, 1143,
301.
Min, D.B.; and Lee, E.C. Frontiers of Flavor. Frontiers of Flavor, Charalambous, G. Ed., Else-
vier: Amsterdam, 1988 473-498.
Min, D.B.; and Lee, H.O. Chemistry of Lipid Oxidation. Food Lipids and Health, McDonald,
R.E., and Min, D.B., Eds., Marcel Dekker: New York, 1996, 241.
Min, D.B.; and Boff, J.M. Chemistry and Reaction of Singlet Oxygen in Foods. Comp. Rev.
Food Sci. and Food Saf. 2002, 1, 58.
Min, D.B.; Callison, A.L.; and Lee, H.O. Singlet Oxygen Oxidation for 2-Pentylfuran and
2-Pentenylfuran Formation in Soybean Oil. J. Food Sci. 2003, 68, 1175.
Nonell, S.; and Braslavsky, S.E. Time-Resolved Singlet Oxygen Detestion. Methods in Enzy-
mology, Packer, L., Sies, H., Eds., Vol. 319, Academic Press: New York, 2000, 37-49.
Pecci, L.; Costa, M.; Antonucci, A.; Montefoschi, G.; and Cavallini, D. Methylene Blue Pho-
tosensitized Oxidation of Cysteine Sulfinic Acid and Other Sulfinates: The Involvement
of Singlet Oxygen and the Azide Paradox. Biochem. Biophys. Res. Com. 2000, 270, 782.
Rawls, H.R.; and VanSanten, P.J. A Possible Role for Singlet Oxidation in the Initiation of
Fatty Acid Autoxidation. J. Am. Oil Chem. Soc. 1970, 47, 121.
Schneider, T.; Gugliotti, M.; Politi, M.J.; and Baptista, M.S. Quantitative Determination of
Singlet Oxygen by Laser Deflection Calorimetry. An. Lett. 2000, 33, 297.
Sharman, W.M.; Allen, C.M.; and E., v. L.J. Role of Activated Oxygen Species in Photody-
namic Therapy., in Methods in Enzymology, Packer, L. and Sies, H., Eds., Academic Press:
New York, 2000, 376.
Smagula, M.S.; Ho, C.T.; and Chang, S.S. The Synthesis of 2-(2-Pentenyl) Furans and Their
Relationship to the Reversion Flavor of Soybean Oil. J. Am. Oil Chem. Soc. 1978, 56,
516.
Smouse, T.H.; and Chang, S.S. A Systematic Characterization of the Reversion Flavor of Soy-
bean Oil. J. Am. Oil Chem. Soc. 1967, 44, 509.
Song, Y.Z.; An, J.Y.; and Jiang, L.J. ESR Evidence of the Photogeneration of Free Radicals
(GDHB*, O2*) and Singlet Oxygen (1O2) by 15-Deacetyl-13-Glycine-Substituted Hypo-
crelin B. Biochem. Biophys. Acta. 1999, 1472, 307.
Stratton, S.P.; and Liebler, D.C. Determination of Singlet Oxygen-Specific versus Radical-
Initiated Lipid Peroxidation in Photosensitized Oxidation of Lipid Bilayers: Effect of
β-Carotene and α-Tocopherol. Biochemistry 1997, 36, 12911.
Takayama, F.; Egashira, T.; and Yamanaka, Y. Singlet Oxygen Generation from Phosphatidyl-
choline Hydroperoxide in the Presence of Copper. Life Sci. 2001, 68, 1807-1815.
Vaisey-Genser, M.; Malcomson, L.J.; Przybylski, R.; and Eskin, N.A.M. Consumer Accep-
tance of Stored Canola Oils in Canada. The 12th Project Report Research on Canola, Seed,
and Oil Meal, Canada, Canola Council: Canada, 1999, 189.
Whang, K.; and Peng, I.C. Electron Paramagnetic Resonance Studies of the Effectiveness of
Myoglobin and Its Derivatives as Photosensitizers in Singlet Oxygen Generation. J. Food
Sci. 1988, 53, 1863.
Yang, W.T.; and Min, D.B. Chemistry of Singlet Oxygen Oxidation of Foods, in Lipids in Food
Flavors, Ho, C.T., and Hartmand, T.G., Eds., American Chemical Society: Washington
D.C., 1994, 15.
2
Chemistry and Reactions of Reactive
Oxygen Species in Lipid Oxidation
Eunok Choe
Department of Food and Nutrition, Inha University, 253 Yonghyundong, Namku,
Incheon, Korea
Introduction
The reactive oxygen species (ROS) include oxygen radicals, such as hydroxyl (HO∙),
alkoxyl (RO∙), hydroperoxyl (HOO∙), peroxyl (ROO∙), and superoxide anion (O2∙-)
radicals, as well as nonradical derivatives of oxygen, such as hydrogen peroxide (H2O2),
ozone (O3), and singlet oxygen (1O2). Molecular oxygen in air is normally in a triplet
state, and its sequential univalent reduction produces more reactive superoxide anion,
hydrogen peroxide, and hydroxyl radicals. ROS are interrelated to each other as shown
in Fig. 2.1. ROS has a very short half-life; half-lives of hydrogen peroxide and organic
hydroperoxides are in the range of minutes, peroxyl radical has half-life of seconds.
Superoxide anion and alkoxyl radicals show about a μsec half-life, and hydroxyl radical
has the shortest half-life of nsec (Kehrer, 2000).
ROS is mainly responsible for the initiation of oxidation reaction of foods and
affects food quality and human health. Lipid oxidation by ROS produces undesirable
volatile compounds, destroys essential fatty acids, and sometimes produces carcino-
gens. ROS also changes the functionality of lipids by forming oxidized dimers and tri-
mers. The ROS in foods lowers the overall quality of foods during processing, storage,
and marketing (Choe and Min, 2005).
Since chemistry and reactions of singlet oxygen is dealt in a previous chapter of this
book, hydroxyl, alkoxyl, hydroperoxyl, peroxyl, and superoxide anion radicals, hydro-
gen peroxide, and ozone are among the ROS covered in this chapter.
Hydroxyl Radicals
Hydroxyl radicals are formed by radiolysis of water in the presence of high-energy
radiation. Water absorbs the energy and becomes ionized (H2O+) and excited (H2O*)
within 10-16 s (Halliwell and Gutteridge, 2001). Excited water molecules undergo ho-
molysis in 10-14 to 10-13 s and produce hydrogen atoms and hydroxyl radicals (Halliwell
and Gutteridge, 2001). A reaction of ionized water with other water molecules also
produces hydroxyl radicals (Jacobien et al., 1996).
31
32 E. Choe
2ROO
carb on yl
com po u n ds
HOO 3 3
O2 + sensitize r
sen sitize r
1
O2
F e2+
OH- + HO
H 2O 2 F e 3+
O2 -
1
se nsitizer
H 2O 2 H+
3
O 2 + e aq- 3
se nsitizer + 3 O 2
3 R+
1 enzy m e + O 2
O2 , OH-
HOO
H 2O
2+ 3+ -
Fe Fe , OH
irra diation
H 2O 2 HO H 2O 2 ROO
H H 2O 2 + H O
irradia tio n
H 2O
Fig. 2.1. Interrelationships among reactive oxygen species in foods.
hν
2 H2O → H2O* + H2O+ + eaq –
H2O* → H. + HO.
H2O+ + H2O → H3O+ + HO.
Homolytic fission of the oxygen-oxygen bond in hydrogen peroxide upon expo-

sure to UV light and Fenton reaction produce hydroxyl radicals. In Fenton reaction,
hydrogen peroxide reacts with iron (Salem et al., 2000). At 25°C the rates to produce
hydroxyl radical from the reaction of hydrogen peroxide with Fe+2-ATP and Fe+2-ci-
trate complexes were 6.7 x 103 and 4.9 x 103 M-1s-1, respectively (Rush et al., 1990).
Formate anion (HCOO∙-) and superoxide anion radicals, hydroquinone, or cysteine
promote hydroxyl radical production by reducing Fe3+ to Fe2+ (Watanabe et al., 2002).
However, the reaction of hydrogen peroxide with human myoglobin containing iron
did not produce hydroxyl radical (Witting et al., 2000).
Chemistry and Reactions of Reactive Oxygen Species in Lipid Oxidation 33
Fenton reaction
H2O2 + Fe2+ → HO∙ + OH- + Fe3+
Fe3+ + O2∙- → Fe2+ + O2
Haber-Weiss reaction is another pathway for the production of hydroxyl radicals

(Kehrer, 2000). In Haber-Weiss reactions, hydrogen peroxide and superoxide anion
react with each other, but the reaction is very slow at 0.13 M-1s-1 (Weinstein and
Bielski, 1979). Presence of transition metals increases the reaction rate (Hu and Jiang,
1996), and oxygen in a singlet state is also produced (Khan and Kasha, 1994). Ferrous
ions are regenerated by the oxygen evolution reaction.
H2O2 + Fe2+ → HO∙ + OH- + Fe3+

Fe3+ + O2∙- → Fe2+ + 1O 2
H2O2 + O2∙- → HO. + OH- + 1O 2
Hydroxyl radical has a very high standard reduction potential (2.3 V) and is one
of the most reactive species known (Choe and Min, 2005). It is an extremely strong
oxidizing agent and powerful electrophilic radical. The electron accepting rate of hy-
droxyl radical is 109 to 1010 M-1s-1 (Halliwell and Gutteridge, 2001).
Hydroxyl radical is mainly responsible for the initiation of lipid oxidation. It
reacts with lipids unspecifically in a diffusion-limited mode (Kruk et al., 2005). Hy-
droxyl radical abstracts hydrogen, mostly allylic hydrogen of unsaturated lipids (RH)
which have low bond dissociation energy. For example, the energy to break a bond
between hydrogen and C11 of linoleic acid is 209 kJ/ mole. In contrast, the bond
dissociation energy between hydrogen and C8, and that between hydrogen and C14
of linoleic acid is 314 kJ/ mole. More than 418 kJ/ mole are required to remove hy-
drogen at C17 or C18 (Min and Boff, 2002). It is easier to abstract hydrogen from a
secondary or tertiary carbon than from a primary carbon in the lipid oxidation. In the
oxidation of cholesterol by hydroxyl radical, the hydrogen at C 7 is usually abstracted
(Girotti, 1998). Abstraction of hydrogen from lipid (RH) results in lipid radical (R∙),
which can react with triplet oxygen for the free radical chain reaction as shown in Fig.
2.2. Triplet oxygen which has two unpaired electrons can react with lipid radicals,
forming a lipid peroxyl radical, for example C9- and C13- peroxyl radicals in linoleic
acid. The lipid peroxyl radical instantly propagates the chain reaction of lipid oxida-
tion.
Hydroxyl radical can be added to the double bond of unsaturated lipids (Korycka-
Dahl and Richardson, 1978; Lee et al., 2004) and form hydroxylated lipid radicals
(Fig. 2.3). The hydroxylated lipid radical reacts with triplet oxygen and forms hydrox-
ylated lipid peroxyl radicals. These radicals also propagate the chain reaction of lipid
autoxidation.
β-carotene (CarH) decreases lipid oxidation by donating hydrogen to hydroxyl
radicals and becomes a carotene radical (Car∙). A carotene radical is a fairly stable spe-
34 E. Choe
H 3 C (C H 2 ) 4 H C C H C H 2 C H C H (C H 2 ) 7 C O O H + HO
H 3 C (C H 2 ) 4 H C C H C H C H C H (C H 2 ) 7 C O O H
13 9
H 3 C (C H 2 ) 4 H C C H C H C H C H (C H ) 7 C H O O H + H 3 C (C H 2 ) 4 H C C H C H C H C H (C H 2 ) 7 C O O H
3
+ O2
13 9
H 3 C (C H 2 ) 4 H C C H C H C H C H (C H 2 ) 7 C O O H + H 3 C (C H 2 ) 4 H C C H C H C H C H (C H 2 ) 7 C O O H
OO OO
Fig. 2.2. Hydroxyl radical initiated oxidation of linoleic acid.
H 3 C (C H 2 ) 7 H C C H (C H 2 ) 7 C O O H
+ HO
10 9 10 9
H 3 C (C H 2 ) 7 H C C H (C H 2 ) 7 C O O H + H 3 C (C H 2 ) 7 H C C H (C H 2 ) 7 C O O H
OH OH
+ 3O 2
OO OO
H 3 C (C H 2 ) 7 H C C H (C H 2 ) 7 C O O H + H 3 C (C H 2 ) 7 H C C H (C H 2 ) 7 C O O H
10 9
OH OH
Fig. 2.3. Addition reaction of electrophilic hydroxyl radical to oleic acid.

cies due to delocalization of unpaired electron through its conjugated polyene system.
When a carotene radical reacts with other radicals, such as peroxyl radicals, nonradical
products (Car-OOR) are formed and the radical reactions stop.
CarH + ∙OH → Car∙ + H2O

Car∙ + ROO∙ → Car-OOR
Hydroxyl radicals efficiently oxidize α-tocopherol (TOH) through an indirect

reaction, due to their low diffusibility. The hydroxyl radical first reacts with the sol-
vent molecules (SH) and the solvent radicals (S∙) will oxidize tocopherols afterward
(Fukuzawa and Gebicki, 1983).
∙OH + SH → H2O + S∙
S∙ + TOH → SH + TO∙
Alkoxyl Radicals
Alkoxyl radicals are produced by homolysis of lipid hydroperoxides (ROOH). The
energy required to cleave the oxygen-oxygen bond is 184 kJ/ mole (Hiatt et al., 1968),
and heat, UV, or transition metals accelerate the homolysis (Heaton and Uri, 1961;
Schaich, 1992; and Jadhav et al., 1996).
heat, UV
ROOH → RO∙ + ∙OH
ROOH + Fe2+ → RO∙ + Fe3+ + OH-
Transition metal ion-mediated decomposition of lipid hydroperoxides is initiated

by a one-electron reduction to an alkoxyl radical and is 10 times faster with Fe2+ than
with Fe3+ (Kilic and Richards, 2003). Ascorbic acid induces hydroperoxide decompo-
sition (Lee and Blair, 2000), and was more efficient at 25 μM to 2 mM than Cu2+ or
Fe2+ in initiating the decomposition of 13-OOH of linoleic acid (Lee et al., 2001).
Alkoxyl radical undergoes homolytic β-scission of carbon-carbon bond and pro-
duces aldehydes, acids, alcohols, and short-chain hydrocarbons. Alkoxyl radical de-
rived from C8-hydroperoxide of oleic acid can produce 1-decenol, decanal, 1-decene,
8-oxo-octanoic acid, 2-undecenal, 7-hydroxyheptanoic acid, and heptanoic acid by
decomposition (Fig. 2.4). Pentane, hexanal, 2-heptanal, octanoic acid, nonanal, 2-de-
cenal, and 2,4-decadienal are good examples of decomposition volatile compounds
found in oxidized oils (Kanavouras et al., 2004), which give off-flavor. Alkoxyl radi-
cal can also abstract hydrogen from lipids to produce new lipid radicals since it has
a standard reduction potential of approximately 1.6 V which is higher than that of
unsaturated lipids (Choe and Min, 2005). The resulting lipid radicals continue the
chain reaction of lipid oxidation. The reaction of alkoxyl radicals with lipids is at a
rate constant of 106-107 M-1s-1 (Simic et al., 1992). The lipid oxidation by alkoxyl
radicals can be reduced by tocopherols. Tocopherols react with alkoxyl radicals and
produce tocophroxyl radicals which are more stable than alkoxyl radicals due to their
resonance structure (Fig. 2.5). The life-time of a tocophroxyl radical in low density
lipoprotein was reported as 12.5 s (Ingold et al., 1993).
Hydroperoxyl Radicals
A hydroperoxyl radical is a protonated form of a superoxide anion radical and is pro-
duced by reaction of hydrogen peroxide and a hydroxyl radical.
HO. + H2O2 → HOO. (H+ + O2∙-) + H2O
It is also produced by a reaction of triplet oxygen with hydrogen atoms which are
produced from a radiolysis of water during pulsed electric field processing of food
(Halliwell and Gutteridge, 2001).
36 E. Choe
R1
R2
O2 , H
R2 CH CH CH R1
OOH
OH
(B ) (A )
R2 CH CH CH R1
O
(B ) (A )
R2 CH CH + H C (C H 2 ) 6 C O O H H 3 C (C H 2 ) 7 C H C H C H + R 1
O O
OH R 3H R 3H
OH
R3 R3
H 3 C (C H 2 ) 7 C H C H H 3 C (C H 2 ) 7 C H C H 2 H O C H 2 (C H 2 ) 5 C O O H H 3 C (C H 2 ) 5 C O O H
OH
H 3 C (C H 2 ) 7 C H 2 C H
O
Fig. 2.4. Decomposition of hydroperoxides of oleic acid (r1: -(ch2)6cooh, r2: -(ch2)7ch3, r3: alkyl group).
O O
O C 16 H 33 O C 1 6 H 33
Fig. 2.5. Resonance form of a-tocophroxyl radical.

2 H2O → H2O+ + H2O* + eaq –

H2O* → H. + HO.
H. + O2 → HOO.
The hydroperoxyl radical has a reduction potential of 1.1-1.5 V (Choe and Min,
2005) and is thermodynamically capable of oxidizing unsaturated fatty acids whose
reduction potential is approximately 0.6 V at neutral pH (Koppenol, 1990). It ab-
stracts hydrogen, mostly from allylic carbons of unsaturated fatty acids (Bielski et al.,
1983) and produces fatty acid radicals. The reaction rate of hydroperoxyl radicals in-
creases as the unsaturation of lipids increases; 1.2 x 103, 1.7 x 103, and 3.1 x 103 M-1s-1
for linoleic, linolenic, and arachidonic acids, respectively (Bielski et al., 1983; Aikens
and Dix, 1991). Oleic acid did not react with hydroperoxyl radicals (Bielski et al.,
1983). Hydroperoxyl radicals can also abstract hydrogen from lipid hydroperoxides
and produces lipid peroxyl radicals (Aikens and Dix, 1991). Lipid radicals and per-
oxyl radicals can abstract hydrogen from other molecules, producing another radical,
and the lipid oxidation propagates through a free radical chain reaction.
Lipid oxidation by hydroperoxyl radicals is slowed down by tocopherols. To-
copherol reacts with hydroperoxyl radicals at a rate of 2.0 x 105 M-1s-1 (Arudi et al.,
1983), which is a higher rate than that of a reaction between hydroperoxyl radical and
lipid.
Peroxyl Radicals
Triplet oxygen having two unpaired electrons reacts with lipid radicals and produces
lipid peroxyl radicals. A lipid radical is formed by hydrogen donation from a lipid,
which occurs mostly at allylic carbons, C8, C9, C10, or C11 for oleic acid, C9 or C13
for linoleic acid, and C9, C12, C13, or C16 for linolenic acid. Triplet oxygen goes to
one of these carbons, which have odd electrons, and produces peroxyl radicals. Figure
2.6 shows a formation of C9- and C13-peroxyl radicals from a reaction of linoleic
acid with oxygen. The rates for the formation of peroxyl radicals depend on oxygen
availability and temperature (Velasco et al., 2003).
Peroxyl radicals have 1.0 V of standard reduction potential (Choe and Min,
2005) and can easily abstract hydrogen from other lipid molecules. The peroxyl radi-
cal serves as a chain carrier in the free radical lipid oxidation.
ROO∙ + R’H → ROOH + R’∙

R’∙ + O2 → R’OO∙
Reaction rates with polyunsaturated fatty acids are higher in peroxyl radicals than in hy-
droperoxyl radicals. The C13-peroxyl radical of methyl linoleate reacts with linoleic acid
at a rate of 1.1 x 106 M-1s-1 (Iulaino et al., 1995), and hydroperoxides are formed along
with new linoleic acid radicals. The hydroperoxides are mostly stable at room tempera-
ture; however, in the presence of heat, UV, or transition metals they are decomposed to
38 E. Choe
O
C H 3 (C H 2 ) 4 C H C H C H 2 C H C H C H 2 (C H 2 ) 6 C O H
O O
C H 3 (C H 2 ) 4 C H C H C H C H C H C H 2 (C H 2 ) 6 C O H C H 3 (C H 2 ) 4 C H C H C H C H C H C H 2 (C H 2 ) 6 C O H
+ O2 + O2
O O
C H 3 (C H 2 ) 4 C H C H C H C H C H C H 2 (C H 2 ) 6 C O H C H 3 (C H 2 ) 4 C H C H C H C H C H C H 2 (C H 2 ) 6 C O H
OO OO
9-peroxy 13-peroxy
Fig. 2.6. Formation of peroxyl radical from the reaction between linoleic acid and triplet oxygen.
lower molecular weight compounds via homolytic cleavage as described previously.

β-Carotene having 1.1 V of reduction potential (Edge et al., 2000) can compete
with lipids to donate hydrogen to peroxyl radicals, which can decrease the reaction
between lipid and peroxyl radicals. However, the actual occurrence is rarely observed.
The bond dissociation energy of the most labile hydrogen in β-carotene is 309 kJ/
mole, whereas the bond dissociation energy of O-H bond in hydroperoxides is about
370-380 kJ/mole (Luo and Holmes, 1994), which can thermodynamically transfer
hydrogen from β-carotene to a peroxyl radical (Haila, 1999). Peroxyl radicals abstract
hydrogen from β-carotene at low oxygen concentrations of less than 760 torr and
produce a carotene radical. Carotene radicals undergo addition of oxygen and caro-
tene and homolysis of the oxygen-oxygen bond repeatedly, and produce dicarbonyls
of carotene (Beutner et al., 2001) as shown in Fig. 2.7.
Peroxyl radicals can be added to β-carotene, especially at higher than 150 torr of
oxygen (Burton and Ingold, 1984) as shown in Fig. 2.8 (Decker, 2002). The peroxyl
radical is added to the cyclic end group, or to the polyene chain of β-carotene espe-
cially at C15 and C15’ position (Iannone et al., 1998), and produces 5,6-epoxides
of carotene and peroxides of carotene radical (ROO-Car∙), respectively. ROO-Car∙
reacts with another peroxyl radical and produces nonradical compounds, diperoxides
of carotene (ROO-Car-OOR’). Cleavage between carbons having a peroxyl group in
ROO-Car-OOR’ and elimination of alkoxyl radicals produce aldehyde compounds
containing cyclic groups. ROO-Car∙ can also react with molecular oxygen to produce
peroxyl radicals of carotene peroxides (ROO-Car-OO∙). These radicals react with
lipid molecules and produce hydroperoxides of carotene and lipid radicals, which
increase lipid oxidation (Tsuchihashi et al., 1995; Iannone et al., 1998). Burton and
Ingold (1984) reported the prooxidant activity of β-carotene at a concentration of
higher than 5 x 10-4 M under the oxygen pressure of higher than 150 torr.
ROOH
+ O2 + C arH
C arH + ROO Car Car OO C ar O O C ar
decomposition
Car O or ca ro te ne epoxide
+ C arH
C ar O C a r
+ O2
Car O Car O O
+ C arH
decomposition
ca ro tene epoxide or C ar O C a r O Car O C a r O O C ar
decomposition
C ar + O Car O
O O
C Car C
dicarbo nyls
Fig. 2.7. Oxidation of b-carotene by peroxyl radical through carotene radical.
β-Carotene can donate an electron in its conjugated double bonds to peroxyl

radicals and produce a β-carotene cation radical and a peroxyl anion (Liebler, 1993;
Mortensen et al., 2001). β-Carotene cation radical is stable due to its resonance struc-
ture and rarely reacts with oxygen (Edge et al., 2000; Decker, 2002). The transfer
of hydrogen or electrons from carotenoids to free radicals depends on the reduction
potentials of the free radicals and the chemical structures of the carotenoids, especially
the presence of oxygen-containing functional groups (Edge et al., 1997). The rates of
electron transfer increase with the number of coplanar conjugated double bonds and
is decreased by the presence of hydroxyl and keto groups (Mortensen et al., 2001).
Tocopherols have 0.5 V of standard reduction potential (Buettner, 1993) and com-
pete with unsaturated lipids for peroxyl radicals. Peroxyl radicals react with tocopherols
at 106-109 M-1s-1 which is faster than with lipid at 10-60 M-1s-1 or with β-carotene at 106
M-1s-1 (Mortensen and Skibsted, 1998). Reactivity of peroxyl radicals with α-tocopherol
is 32 times higher than with β-carotene (Tsuchihashi et al., 1995). The reaction rate
of oleic acid peroxyl radical with α-tocopherol is 2.5 x 106 M-1s-1 (Simic, 1980). One
tocopherol molecule can protect about 103-108 polyunsaturated fatty acid molecules at
low peroxide value (Kamal-Eldin and Appelqvist, 1996). Tocopherol donates hydrogen
at the 6-hydroxyl group on the chromanol ring to lipid peroxyl radicals and produces
lipid hydroperoxides and tocophroxyl radicals having a resonance structure.
40 E. Choe
+ ROO
OOR
OOR
RO + R 'O O
OOR
O
OOR'
+ O2
R O , R 'O
O
OOR
+
O
OO
+ R ''H
OOR
R '' +
OOH
Fig. 2.8. Addition reaction of peroxyl radical to b-carotene.
The tocophroxyl radical can react with another peroxyl radical and produces to-
copherol semiquinone, a potential antioxidant (Neuzil et al., 1997), or it reacts with
another tocophroxyl radical and forms tocopherol dimer (Reische et al., 2002) as
shown in Fig. 2.9.
Although tocopherols are generally decrease lipid oxidation by slowing down
free radical reactions, they can increase lipid oxidation under certain conditions. At
very low concentrations of peroxyl radicals, the tocophroxyl radical abstracts hydro-
gen from lipids at low rates to give tocopherol and lipid radicals, which promotes
lipid oxidation. This is called tocopherol-mediated peroxidation (Bowry and Stocker,
1993; Yamamoto 2001). The rate of allylic hydrogen abstraction from ethyl oleate,
ethyl linoleate, ethyl linolenate, and ethyl arachidonate by the tocophroxyl radical
is 1.04 x 10-5, 1.82 x 10-2, 3.84 x 10-2, and 4.83 x 10-2 M-1s-1 at 25oC, respectively
(Mukai and Okauchi, 1989). Linoleic acid in the tocopherol-mediated peroxidation
favors the accumulation of 13-OOH over 9-OOH (Porter et al., 1980; Upston et
al., 2002). Ascorbic acid prevents tocopherol-mediated peroxidation by reducing the
tocophroxyl radical to tocopherol (Yamamoto, 2001).
CH3
HO
CH3
+ ROO
CH3
H 3C O
C H 2 (C H 2 C H 2 C H C H 2)3 H
CH3
(T O H )
CH3
O
CH3
+ ROOH
CH3
H 3C O
C H 2 (C H 2 C H 2 C H C H 2)3H
CH3
(T O )
+ R 'O O + TO
CH3
toco p h e ro l dim e r
O
CH2
CH3
+ R 'O O H
H 3C O CH2 (C H 2 C H 2 C H C H 2 ) 3 H
CH3
to copherol se m iqu ino ne
Fig. 2.9. Reaction of tocopherol with peroxyl radical.

42 E. Choe
Superoxide Anion Radicals

Superoxide anion has one unpaired electron in the 2pπ* orbital as shown in Fig. 2.10
and is a radical. It is produced from triplet oxygen by the addition of one electron
to one of the 2pπ* orbitals, which is reduction. Reaction of oxygen with xanthine or
NADPH produces superoxide anion radicals and xanthine oxidase (Das and Das,
2002) and NADPH oxidase (Martinez and Moreno, 1996) catalyze the reaction, re-
spectively.
Fig. 2.10. Electronic configuration of 2pπ* orbital of superoxide anion radical.
Xanthine oxidase
Hypoxanthine + 2O2 + H2O → Xanthine + 2O2∙- + 2H+

Xanthine + 2O2+ H2O → Uric acid + 2O2∙- + 2H+
NADPH oxidase
NADPH + 2O2 → NADP+ + 2O2∙- + H+
Reaction of aminoglycosides (AG) with oxygen produces superoxide anion (In-

gold et al., 1997), and requires iron and fatty acids (FA) for the formation of FA
-iron- AG complex (Lesniak et al., 2005). FA serves as an electron donor.
AG + Fe2+ + FA → [ FA-Fe2+-AG]
[FA-Fe2+-AG] + O2 → [ FA-Fe2+- O2-AG]
[ FA-Fe2+- O2-AG] ↔ [ FA-Fe3+- O2∙- -AG]
[ FA-Fe3+- O2∙- -AG] ↔ [ FA-Fe3+-AG] + O2∙-
Photoactivation of sensitizers such as riboflavin (RF) produces superoxide anion

(Buettner and Oberley, 1980). Light energy at a specific wavelength converts the RF
to the excited RF (1RF*). The 1RF* becomes an excited triplet RF (3 RF*) via intersys-
tem crossing by emitting some of its energy. 3 RF* can react with triplet oxygen and
produce superoxide anion and a RF radical (Haseloff and Ebert, 1989). The reaction
usually needs ethanol as a solvent, and the reaction rate depends on the concentration
of ethanol (Athar et al., 1988; Haseloff and Ebert, 1989).
hv
1
RF → 1
RF *
1
RF * → 3
RF *
3
RF* + O2 → RF ∙+ + O2∙-
Superoxide anion is formed by radiolysis of a dilute aqueous solution, in which

most of the energy is absorbed by water and produces electrons along with ionized
water (H2O+) and excited water (H2O*) in a short time (Halliwell and Gutteridge,
2001). The electrons are surrounded by water molecules within 10-12-10-11 s. The
hydrated electrons (eaq –) have a standard reduction potential of -2.8 V and reduce
oxygen to superoxide anion in the presence of enough oxygen.
2 H2O → H2O+ + H2O* + eaq –

3
O2 + eaq – → O2∙-
Superoxide anion radical is relatively stable under anhydrous conditions, but not
in aqueous media. It rarely oxidizes the lipids directly because its reduction potential
(0.9 V) is not strong enough to abstract hydrogen from lipids (Bielski et al., 1983).
Superoxide anion contributes indirectly to the lipid oxidation via formation of hy-
droxyl radical, a strong oxidant by Haber-Weiss reaction. Superoxide anion slowly
and irreversibly oxidizes tocopherols in organic solvent and produces relatively stable
tocophroxyl radical (Csallany and Ha, 1992), which can decrease lipid oxidation. But
under aqueous conditions this reaction rarely occurs (Arudi et al., 1983; Halliwell and
Gutteridge, 2001). Tocopherols react with hydroxyl radicals and produce tocopherol
dimer, tocopherol dihydroxy dimer, or tocopheryl quinone, under aqueous condi-
tions (Csallany and Ha, 1992).
Hydrogen Peroxide
Hydrogen peroxide, the protonated form of peroxide ion, is formed by the reaction
of two molecules of hydroxyl radicals (5 x 109 M-1s-1; Halliwell and Gutteridge, 2001)
and the dismutation of superoxide anion (Bielski et al., 1983).
Dismutation
O2∙- + O2∙- + 2 H+ → H2O2 + O2
Superoxide dismutase greatly accelerates the dismutation reaction at 1.6 x 109 M-1s-1
(Halliwell and Gutteridge, 2001). Uncatalyzed dismutation of superoxide anion de-
pends strongly on the pH of the solution. Hydrogen peroxides are formed faster at
low pH; 1 x 102 and 5 x 105 M-1s-1 at pH 11 and pH 7, respectively (Halliwell and
44 E. Choe
Gutteridge, 2001). This is because superoxide anion radical with a pKa of 4.88 is
present in its protonated form, hydroperoxyl radical, at low pH. Reactions among hy-
droperoxyl radicals produce hydrogen peroxide (Buettner and Hall, 1987). Hydrogen
peroxide is not directly involved in the initiation of lipid oxidation due to its low re-
duction potential (0.3 V), however, it contributes indirectly to the lipid oxidation by
generating hydroxyl radicals in the presence of light or iron (Choe and Min, 2005).
Ozone
Ozone is produced by the photodissociation of molecular triplet oxygen into oxygen
atoms, which then react with oxygen molecules (Halliwell and Gutteridge, 2001).
hν
O2 → O∙ + O∙
O∙ + O2 → O3
Ozone has a standard reduction potential of 2.1 V (Eren, 2006) and is a power-
ful oxidizing agent. Treatment of sardine (Sardina pilchardus) with ozone increased
significantly (p < 0.05) the formation of peroxides and thiobarbituric acid reactive
substances (Losada et al., 2004). Ozone readily reacts with unsaturated lipids and
the reaction is faster in lipids with higher unsaturation, however, it hardly reacts with
saturated lipids (Fredrick and Heath, 1970). Exposure of ground soybeans to 1.5
ppm ozone induced 1.7-5.4% destruction of linoleic acid and 10-22% destruction of
linolenic acid (Brooks and Csallany, 1978), but there was no reaction of stearic acid
with ozone (Pryor et al., 1991).
Ozone adds directly to double bonds in unsaturated lipids and produces primary
ozonide (molozonide; 1,2,3-trioxolane), which decomposes to aldehydes and carbonyl
oxide (Fig. 2.11). The carbonyl oxide can subsequently react with an aldehyde to give
a secondary ozonide, Criegee ozonide (Squadrito et al., 1992). Ozonation of methyl
oleate in hexane produced an 89% yield of the Criegee ozonides (Pryor and Wu,
1992). The secondary ozonides are more stable than the primary ozonides; however,
they can decompose to produce aldehydes and carboxylic acid. Nonanedioic acid and
nonanoic acid are good examples of oxidation products of oleic acid by ozone (Sparks
et al., 2006).
Carbonyl oxides can react with water to produce metastable hydroxyl hydroperox-
ide which hydrolyzes to hydrogen peroxide and aldehydes (Pryor et al., 1991; Santrock
et al., 1992) as shown in Fig. 2.12. Although hydrogen peroxide is a principal com-
pound in the oxidation of unsaturated lipid emulsions by ozone, hydrogen peroxide
was not found in the reaction of ozone with unsaturated lipids in carbon tetrachloride
having no water (Pryor et al., 1991). Ozone produced stable ethoxyhydroperoxides by
oxidizing linoleic acid (Heath and Tappel, 1976), phosphatidylcholine (Tagiri-Endo
et al., 2002), and cholesterol (Endo et al., 2004). In aqueous solutions, ozone decom-
poses and forms hydroxyl radical at a low rate (Halliwell and Gutteridge, 2001).
H 3 C (C H 2 ) 7 C H C H (C H 2 ) 7 C O O H + O 3
O
O O
H 3 C (C H 2 ) 7 CH C H (C H 2 ) 7 C O O H
primary ozonide (1,2,3-trioxolanes)
O OO OO O
H 3 C (C H 2 ) 7 C H + H C (C H 2 ) 7 C O O H H 3 C (C H 2 ) 7 C H + H O O C (C H 2 ) 7 C H
1-nonanal 9-oxononanoic acid
O
H 3 C (C H 2 ) 7 H C C H (C H 2 ) 7 C O O H
O O
Criegee ozonide (1,2,4-trioxolanes)
O O O
O C (C H 2 ) 7 C O O H + H 3 C (C H 2 ) 7 C H H C (C H 2 ) 7 C O O H + O C (C H 2 ) 7 C H 3
H O H
isomerization + carboxylic acids
H O O C (C H 2 ) 7 C O O H
carboxylic polymer
nonanedioic acid (azeliac acid)
Fig. 2.11. Reaction of ozone with oleic acid.

46 E. Choe
H H O
+ O3 O O
C C C +
R1 R2 R1 H HC R2
+
+ H 2O
O O HOO OH
H H C
C C H R2
R1 O R2
O
R2 C H + H 2O 2
Fig. 2.12. Production of hydrogen peroxide and aldehydes from the reaction of lipids with ozone in the presence
of water.
References
Aikens, J.; and J.A. Dix. Perhydroxyl Radical (HOO∙) Initiated Lipid Peroxidation, J. Biol.
Chem. 1991, 266, 15091.
Arudi, R.L.; M.W. Sutherland; and B.H.J. Bielski. Oxy Radicals and Their Scavenger Systems.
Molecular Aspects, Cohen, G., and R.A. Greenwald, Eds.; Elsevier: Amsterdam, 1983; Vol.
1, pp. 26-31.
Athar, M.; H. Mukhtar; C.A. Elmets; . Tarif Zaim; J.R. Lloyd; and D.R. Bickers. In situ Evi-
dence for the Involvement of Superoxide Anions in Cutaneous Porphyrin Photosensitiza-
tion, Biochem. Biophys. Res. Commun. 1988, 151, 1054-1059.
Beutner, S.; B. Bloedorn; S. Frixel; I.H. Blanco; T. Hoffmann; H-D. Martin; B. Mayer; P.
Noack; C. Ruck; M. Schmidt; et al. Quantitative Assessment of Antioxidant Properties of
Natural Colorants and Phytochemicals: Carotenoids, Flavonoids, Phenols, and Indigoids.
The Role of β-Carotene in Antioxidant Functions. J. Sci. Food Agr. 2001, 81, 559-568.
Bielski, B.J.H.; R.L. Arudi; and M.W. Sutherland. A Study of The Reactivity of HO2/O2- with
Unsaturated Fatty Acids. J. Biol. Chem. 1983, 258, 4759-4761.
Bowry, V.W.; and R. Stocker. Tocopherol-Mediated Peroxidation. The Prooxidant Effect of
Vitamin E on the Radical-Initiated Oxidation on Human Low-Density Lipoprotein. J.
Am. Chem. Soc. 1993, 115, 6029-6044.
Brooks, R.I.; and A.S. Csallany. Effects of Air, Ozone, and Nitrogen Dioxide Exposure on the
Oxidation of Corn and Soybean Lipids. J. Agric. Food Chem. 1978, 26, 1203-1209.
Buettner, G.R. The Pecking Order of Free Radicals and Antioxidants: Lipid Peroxidation,
α-Tocopherol, and Ascorbate. Arch. Biochem. Biophys. 1993, 300, 535–543.
Buettner, G.R.; and T.D. Hall. Superoxide, Hydrogen Peroxide, and Singlet Oxygen in He-
matoporphyrin Derivative-Cysteine, -NADH and -Light Systems. Biochim. Biophys. Acta.
1987, 923, 501-507.
Buettner, G.R.; and L.W. Oberley. The Apparent Production of Superoxide and Hydroxyl
Radicals by Hematoporphyrin and Light as Seen by Spin-Trapping. FEBS Lett. 1980,

121, 161-164.
Burton, G.W.; and K.U. Ingold. β-Carotene: An Unusual Type of Lipid Antioxidant. Science
1984, 224, 569-573.
Choe, E.; and D.B. Min. Chemistry and Reactions of Reactive Oxygen Species in Foods. J.
Food Sci. 2005, 70, R142-R159.
Csallany, A.S.; and Y.L. Ha. α-Tocopherol Oxidation Mediated by Superoxide Anion (O2-). I.
Reactions in Aprotic and Protic Conditions. Lipids 1992, 27, 195-200.
Das, K.C.; and C.K. Das. Curcumin (diferuloylmethane), a Singlet Oxygen (1O2) Quencher.
Biochem. Biophys. Res. Commun. 2002, 295, 62-66.
Decker, E.A. Antioxidant Mechanisms. Food Lipids, 2nd edn.; Akoh, C.C., and Min, D.B.,
Eds.; Marcel Dekker Inc.: New York, 2002, pp. 517-542.
Edge, R.; E.J. Land; D.J. McGarvey; M. Burke; and T.G. Truscott. The Reduction Potential
of the β-Carotene.+/ β-Carotene Couple in an Aqueous Micro-Heterogeneous Environ-
ment. FEBS Lett. 2000, 471, 125-127.
Edge, R.; D.J. McGarvey; and T.G. Truscott. The Carotenoids as Anti-Oxidants - A Review. J.
Photochem. Photobiol. B-Biol. 1997, 41, 189-200.
Endo, M.T.; K. Nakagawa; T. Sugawara; K. Ono; and T. Miyazawa. Ozonation of Cholesterol
in the Presence of Ethanol: Identification of a Cytotoxic Ethoxyhydroperoxides Molecule.
Lipids 2004, 39, 259-264.
Eren, H.A. Afterclearing by Ozonation: A Novel Approach for Disperse Dyeing of Polyester.
Color. Technol. 2006, 122, 329-333.
Fredrick, P.E.; and R.L. Heath. Ozone-Induced Fatty Acid and Viability Changes in Chlorella.
Plant Physiol. 1970, 55, 15-19.
Fukuzawa, K.; and J.M. Gebicki. Oxidation of α-Tocopherol in Micelles and Liposomes by
the Hydroxyl, Perhydroxyl and Superoxide Free Radicals. Arch. Biochem. Biophys. 1983,
226, 242-251.
Girotti, A.W. Lipid Hydroperoxide Generation, Turnover, and Effector Action in Biological
Systems. J. Lipid Res. 1998, 39, 1529-1542.
Haila, K. Effects of Carotenoids and Carotenoid-Tocopherol Interaction on Lipid Oxidation in
vivo. Thesis; University of Helsinki: Viikki, Finland, 1999, p 22.
Halliwell, B.; and J.M.C. Gutteridge. Free Radicals in Biology and Medicine, 3rd edn., Oxford
University Press Inc.: New York, 2001.
Haseloff, R.F.; B. and Ebert. Generation of Free Radicals by Photoexcitation of Pheophor-
bide a, Haematoporphyrin, and Protoporphyrin. J. Photochem. Photobiol. Biology 1989,
3, 593-602.
Heath, R.L.; and A.L. Tappel. A New Sensitive Assay for the Measurement of Hydroperoxides.
Anal. Biochem. 1976, 76, 184-191.
Heaton, F.W.; N. and Uri. The Aerobic Oxidation of Unsaturated Fatty Acids and Their Esters.
Cobalt Stearate-Catalyzed Oxidation of Linoleic Acid. J. Lipid Res. 1961, 2, 152-160.
Hiatt, R.; T. Mill; K.C. Irwin; T.R. Mayo; C.W. Gould; and J.K Castleman. Homolytic De-
composition of Hydroperoxides. J. Org. Chem. 1968, 33, 1416-1441.
Hu, Y-Z.; and L-J. Jiang. Generation of Semiquinone Radical Anion and Reactive Oxygen
(1O2, O2 •-, and •OH) During the Photosensitization of a Water-Soluble Perylenequinone
Derivative. J. Photochem. Photobiol. 1996, 33, 51-59.
Iannone, A.; C. Rota; S. Bergamini; A. Tomasi; and L.M. Canfield. Antioxidant Activity of
Carotenoids: An Electron-Spin Resonance Study on β-Carotene and Lutein Interaction
48 E. Choe
with Free Radicals Generated in a Chemical System. J. Biochem. Mol. Toxicol. 1998, 12,
299-304.
Ingold, K.U.; V.W. Bowry; R. Stocker; and C. Walling. Autoxidation of Lipids and Anti-
oxidation by α-Tocopherol and Ubiquinol in Homogeneous Solution and in Aqueous
Dispersions of Lipids: Unrecognized Consequences of Lipid Particle Size as Exemplified
by Oxidation of Human Low Density Lipoprotein. Proc. Natl. Acad. Sci. USA 1993, 90,
45-49.
Ingold, K.U.; T. Paul; M.J. Young; and L. Doiron. Invention of the First Azo Compound to
Serve as a Superoxide Thermal Source under Physiological Conditions: Concept, Synthe-
sis, and Chemical Properties. J. Am. Chem. Soc. 1997, 119, 12364-12365.
Iuliano, L.; J.Z. Pedersen; G. Rotilio; D. Ferro; and F. Violi. A Potent Chain-Breaking Anti-
oxidant Activity of the Cardiovascular Drug Dipyridamole. Free Radical Biol. Med. 1995,
18, 239-47.
Jadhav, S.J.; S.S. Nimbalkar; A.D. Kulkarni; and D.L. Madhavi. Lipid Oxidation in Biologi-
cal and Food Systems. Food Antioxidants, Madhavi, D.L., Deshpande, S.S., and Salunke,
D.K., Eds.; Mercel Dekker Inc.: New York, 1996, pp. 5-63.
Kamal-Eldin, A.; and L.A. Appelqvist. The Chemistry and Antioxidant Properties of Tocoph-
erols and Tocotrienols. Lipids 1996, 31, 671-701.
Kanavouras, A.; P. Hernandez-Munoz; F. Coutelieris; and S. Selke. Oxidation-Derived Flavor
Compounds as Quality Indicators for Packaged Olive Oil. J. Am. Oil Chem. Soc. 2004,
81, 251-257.
Kehrer, J.P. The Haber-Weiss Reaction and Mechanisms of Toxicity. Toxicology 2000, 149,
43-50.
Khan, A.U.; and M. Kasha. Singlet Molecular Oxygen in the Haber-Weiss Reaction. Proc.
Natl. Acad. Sci. USA 1994, 91, 12365-12367.
Kilic, B.; and M.P. Richards. Lipid Oxidation in Poultry Donor Kebap: Prooxidative and An-
tioxidative Factors. J. Food Sci. 2003, 68, 686-689.
Koppenol, W.H. Oxyradical Reactions: From Bond-Dissociation Energies to Reduction Po-
tentials, FEBS Lett. 1990, 264, 165-167.
Korycka-Dahl, M.B.; and T. Richardson. Activated Oxygen Species and Oxidation of Food
Components. Crit. Rev. Food Sci. Nutr. 1978, 10, 209-241.
Kruk, I.; H.Y. Aboul-Enein; T. Michalska; K. Lichszteld; and A. Kladna. Scavenging of Reac-
tive Oxygen Species by the Plant Phenols Genistein and Oleuropein. Luminescence 2005,
20, 81-89.
Lee, J.; N. Koo, and D.B. Min. Reactive Oxygen Species, Aging, and Antioxidant Nutraceuti-
cals. Comprehensive Rev. Food Sci. Food Safety 2004, 3, 21-33.
Lee, S.H.; and I.A. Blair. Characterization of 4-Oxo-2-nonenal as a Novel Product of Lipid
Peroxidation. Chem. Res. Toxicol. 2000, 13, 698-702.
Lee, S.H.; T. Oe; and I.A. Blair. Vitamin C-Induced Decomposition of Lipid Hydroperoxides
to Endogenous Genotoxins. Science 2001, 292, 2083-2086.
Lesniak, W.; V.L. Pecoraro; and J. Schacht. Ternary Complexes of Gentamicin with Iron and
Lipid Catalyze Formation of Reactive Oxygen Species. Chem. Res. Toxicol. 2005, 18, 357-
364.
Liebler, D.C. Antioxidant Reaction of Carotenoids. Ann. N.Y. Acad. Sci. 1993, 691, 20-31.
Losada, V.; J. Barros-Velazquez; J.M. Gallardo; and S.P. Aubourg. Effect of Advanced Chilling
Methods on Lipid Damage during Sardine (Sardina pilchardus) Storage. Eur. J. Lipid Sci.
Technol. 2004, 106, 844-850.
Luo, Y-R.; and J.L. Holmes. The Stabilization Energies of Polyenyl Radicals. Chem. Phys. Lett.
1994, 228, 329-332.
Martinez, J.; and J.J. Moreno. Influence of Superoxide Radical and Hydrogen Peroxide on
Arachidonic Acid Mobilization. Arch. Biochem. Biophys. 1996, 336, 191-198.
Min, D.B.; and J.M. Boff. Lipid Oxidation of Edible Oil. Food Lipids, 2nd edn.; Akoh, C.C.,
and Min, D.B. Eds.; Marcel Dekker Inc.: New York, 2002, pp 335-364.
Mortensen, A.; and L.H. Skibsted. Reactivity of β-Carotene Towards Peroxyl Radicals Studied
by Laser Flash and Steady-State Photolysis. FEBS Letters 1998, 426, 392-396.
Mortensen, A.; L.H. Skibsted; and T.G. Truscott. The Interaction of Dietary Carotenoids with
Radical Species. Arch. Biochem. Biophy. 2001, 385, 13-19.
Mukai, K.; and Y. Okauchi. Kinetic Study of the Reaction Between Tocopheroxyl Radical and
Unsaturated Fatty Acid Esters in Benzene. Lipids 1989, 24, 936-939.
Neuzil, J.; P.K. Witting; and R. Stocker. α-Tocopheryl Hydroquinone Is an Efficient Multi-
functional Inhibitor of Radical-Initiated Oxidation of Low Density Lipoprotein Lipids.
Proc. Natl. Acad. Sci USA 1997, 94, 7885-7890.
Porter, N.A.; B.A. Weber; and K.J. Smith. Autoxidation of Polyunsaturated Fatty Acids. Fac-
tors Controlling the Stereochemistry of Product Hydroperoxides. J. Am. Chem. Soc.
1980, 102, 5597-5601.
Pryor, W.A.; and M. Wu. Ozonation of Methyl Oleate in Hexane, in a Thin Film, in SDS
Micelles, and in Distearoylphosphatidylcholine Liposomes: Yields and Properties of the
Criegee Ozonide. Chem. Res. Toxicol. 1992, 5, 505-511.
Pryor, W.A.; B. Das; and D.F. Church. The Ozonation of Unsaturated Fatty Acids: Aldehydes
and Hydrogen Peroxide as Products and Possible Mediators of Ozone Toxicity. Chem. Res.
Toxicol. 1991, 4, 341-348.
Reische, D.W.; D.A. Lillard; and R.R. Eitenmiller. Antioxidants. Food Lipids, 2nd edn.; Akoh,
C.C., and Min, D.B. Eds; Marcel Dekker Inc.: New York, 2002, pp. 489-516.
Rush, J.D.; Z. Maskos; and W.H. Koppenol. Reactions of Iron(II) Nucleotide Complexes with
Hydrogen Peroxide. FEBS Lett. 1990, 261, 121-123.
Salem, I.A.; M. El-Maazawi; and A.B. Zaki. Kinetics and Mechanisms of Decomposition
Reaction of Hydrogen Peroxide in Presence of Metal Complexes. Int. J. Chem. Kinetics
2000, 32, 643-666.
Santrock, J.; R.A. Gorski; and J.F. O’Gara. Products and Mechanism of the Reaction of Ozone
with Phospholipids in Unilamella Phospholipids Vesicles. Chem. Res. Toxicol. 1992, 5,
134-141.
Schaich, K.M. Metals and Lipid Oxidation. Contemporary Issues. Lipids 1992, 27, 209-218.
Simic, M.C.; S.V. Jovanovic; and E. Niki. Mechanisms of Lipid Oxidative Processes and Their
Inhibition. Lipid Oxidation in Foods, St. Angelo, A.J. Ed.; ACS Press: New York, 1992,
pp. 14-32.
Simic, M.G. Kinetic and Mechanistic Studies of peroxyl, Vitamin E and Antioxidant Free
Radicals by Pulse Radiolysis. Autoxidation in Foods and Biological Systems, Simic, M.G.,
and Karel, M., Eds.; Plenum: New York, 1980, pp.17–26.
Sparks, D.L.; R. Hernandez; T. French; H. Toghiani; R.K. Toghiani; E. Alley; and M.E. Zappi.
Supercritical Fluid Oxidation of Oleic Acid. 2006 AIChE Annual Meeting, American
Institute of Chemical Engineers. San Francisco, California, U.S.A. 2006. (http://aiche.
confex.com/aiche/2006/preliminaryprogram/abstract_60342.htm)
Squadrito, G.L.; R.M. Uppu; R. Cueto; and W.A. Pryor. Production of the Criegee Ozonide
during the Ozonation of 1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine Liposomes,
50 E. Choe
Lipids 1992, 27, 955-958.

Tagiri-Endo, M.; K. Ono; K. Nakagawa; M. Yotsu-Yamashita; and T. Miyazawa. Ozonation
of PC in Ethanol: Separation and Identification of a Novel Ethoxyhydroperoxide. Lipids
2002, 37, 1007-1012.
Tsuchihashi, H.; M. Kigoshi; M. Iwatsuku; and E. Niki. Action of β-Carotene as an Antioxi-
dant Against Lipid Peroxidation. Arch. Biochem. Biophys. 1995, 323, 137-147.
Upston, J.M.; A.C. Terentis; K. Morris; J.F. Keaney, Jr.; and R. Stocker. Oxidized Lipid Ac-
cumulates in the Presence of α-Tocopherol in Atherosclerosis. Biochem. J. 2002, 363,
753-760.
Velasco, J.; M.L. Andersen; and L.H. Skibsted. Evaluation of Oxidative Stability of Vegetable
Oils by Monitoring the Tendency to Radical Formation. A Comparison of Electron Spin
Resonance Spectroscopy with the Rancimat Method and Differential Scanning Calorim-
etry. Food Chem. 2003, 77, 623-632.
Von Frijtag Drabbe Kunzel, J.K.; J. Van der Zee; and A.P. Ijzerman. Radical Scavenging Prop-
erties of Adenosine and Derivatives in vitro. Drug Development Res. 1996 37, 48-54.
Watanabe, T.; H. Yeranishi; Y. Honda; and M. Kuwahara. A Selective Lignin-Degrading Fun-
gus, Ceriporiopsis subvermispora, Produces Alkylitaconates that Inhibit the Production of a
Cellulolytic Active Oxygen Species, Hydroxyl Radical in the Presence of Iron and H2O2.
Biochem. Biophys. Res. Comm. 2002, 297, 918-923.
Weinstein, J.; and B.H.J. Bielski. Kinetics of the Interaction of Radicals with Hydrogen Perox-
ide: The Haber-Weiss Reaction. J. Am. Chem. Soc. 1979, 101, 58-62.
Witting, P.K.; D.J. Douglas; and A.G. Mauk. Reaction of Human Myoglobin and H2O2:
Involvement of a Thiyl Radical Produced at Cysteine 110. J. Biol. Chem. 2000, 275,
20391-20398.
Yamamoto, Y. Role of Active Oxygen Species and Antioxidants in Photoaging. J. Dermatol.
Sci. 2001, 27, S1-S4 Suppl, 1.
3
Oxidation of Long-Chain
Polyunsaturated Fatty Acids
Kazuo Miyashita
Faculty of Fisheries Sciences, Hokkaido University, Hakodate 041-8611, Japan
Introduction
Lipids account for about 40% of total calories in most of the industrialized countries
and play a major role in human nutrition and health. They provide a concentrated
source of energy and reduce the bulk of the diet. They are indispensable parts of the
human diet as they are major sources of essential fatty acids and provide fat-soluble
substances like vitamins and carotenoids. The type, nature, and composition of lipids
and fatty acids depend on the lipid source. Plants and animal sources account for most
of the dietary lipids. Long-chain polyunsaturated fatty acids (PUFA) of the α-linolenic
acid family (n-3 fatty acids) are typical of marine lipids, while PUFA belonging to lino-
leic acid (LA) family (n-6 fatty acids) are predominant in vegetable oils. However, some
seed oils, such as flaxseed and perilla, do contain n-3 PUFA in considerable quantities;
whereas, soybean, walnut, and rapeseed oils contain moderate amounts of n-3 fatty
acids. Like vegetable oils, the fats derived from terrestrial animal sources largely consist
of PUFA belonging to LA family. Marine lipids contain substantial amounts of long-
chain PUFA. Metabolic effects of the long-chain PUFA, especially in human nutrition,
have attracted a major interest among biochemists and nutritionists alike.
The importance of n-3 and n-6 PUFA on human health has been proven beyond
any doubt through research across the globe. They have several beneficial health and
physiological effects. The functions of each n-3 or n-6 PUFA have attracted consumer
attention and are often used in functional foods and nutraceuticals. Eicosapentaenoic
acid (EPA; 20:5n-3) and docosahexaenoic acid (DHA; 22:6n-3) are the two PUFA
found in marine lipids. These two long-chain PUFA have been shown to cause signifi-
cant biochemical and physiological changes in the body (Li et al., 2003; Lands, 2005;
Sinclair, 2005; Shahidi, and Miraliakbari, 2006; Narayan et al., 2006a) that most often
result in positive influence on human nutrition and health. However, lipid oxidation
products cause undesirable flavors and lower the nutritional quality and safety of lipid-
containing foods. Hence, oxidative deterioration of functional PUFA still remains as
the biggest problem in utilizing PUFA-rich oil for food applications. Owing to this,
lipid peroxidation has received considerable attention in all investigations on the di-
etary effects of these n-3 and n-6 PUFA because of its possible contribution to damage
51
52 K. Miyashita
biological systems (Kaneda and Ishii, 1954; Benedetti et al., 1980; Piche et al., 1988;
Garrido et al., 1989; Hu et al., 1989; Burns and Wagner, 1991; Fritsche and John-
ston, 1988).
Lipid oxidation proceeds through a free radical chain reaction consisting of chain
initiation, propagation, and termination processes (Fig. 3.1) (Frankel, 1980, 1982,
1985, 1998a; Min and Smouse, 1985, 1989; Porter et al., 1995; Kamal-Eldin et al.,
2003). The rate-limiting step in the reaction is abstraction of hydrogen radical (H•)
from substrate lipids (LH) to form lipid free radicals (L•). In the propagation stage
of autoxidation, fatty alkyl radicals react with molecular oxygen to form peroxy radi-
cals. The peroxy radical abstracts a hydrogen atom from another unsaturated fatty
compound to form a hydroperoxide and an alkyl radical. The latter reacts with mo-
lecular oxygen in a repetition of the first propagation reaction. The initially formed
mono-hydroperoxide (MHP) may decompose subsequently to yield free radicals such
as alkoxy and hydroxy radicals. These radicals serve as initiators for these reactions.
Therefore, the chain-breaking antioxidants are of considerable practical importance
in preserving PUFA from oxidative deterioration. These antioxidants inhibit or retard
lipid oxidation by interfering with either chain propagation or initiation by readily
donating hydrogen atoms to lipid peroxy radicals (Fig. 3.1). Phenolic compounds
with bulky alkyl substituents, such as BHA, BHT, TBHQ, and tocopherols, are effec-
tive chain-braking antioxidants as they produce stable and relatively unreactive anti-
oxidant radicals (A•) (Fig. 3.1). Lipid peroxidation is initiated by minor compounds
present in oils. Metal ions promote the initiation and decomposition of MHP. Pig-
ments in foods can serve as photosensitizers by absorbing visible or near-UV light
to become electronically excited. The photoxidation provides an important way to
produce MHP that initiate lipid peroxidation and flavor reversion. Several types of
compounds, such as metal inactivators, can inhibit lipid oxidation by mechanisms
that do not involve deactivation of free radical chain reactions.
Although mechanisms for free radical oxidation of PUFA and the inhibition
of this reaction by antioxidants are well documented (Frankel, 1980, 1982, 1985,
1998a; Min and Smouse, 1985, 1989; Porter et al. 1995; Kamal-Eldin et al., 2003),
autoxidation of long-chain PUFA, such as EPA and DHA, has received much atten-
tion owing to the problems associated with oxidation of these essential PUFA which
makes their utilization in food systems a difficult proposition. Added to this, high
intake of these PUFA may increase the oxidative stress on biological systems due to
oxidation-related problems. The aim of this article is to highlight effect of oxidation
of these PUFA on their stability and usage.
Oxidation of EPA and DHA

The oxidative stability of each PUFA is inversely proportional to the number of bis-
allylic positions in the molecule or the degree of unsaturation of the PUFA; therefore,
when the relative oxidative stability of typical PUFA are compared in air (bulk phase)
or in organic solvents, DHA is most rapidly oxidized, followed by EPA, arachidonic
Oxidation of Long-Chain Polyunsaturated Fatty Acids 53
L ip id o xid atio n
O2
F lavor and
X XH odor defects
LH
LH L● LO O ● A ldeh ydes
R a te-lim iting LO ●
ste p LO OH
L ip id an tio xid an t actio n

A H reacts w ith p eroxy ra d ica ls
LO O + AH
●
LO OH + A ●
a n d a lkyl ra d ica ls to sto p
L + AH
● LH + A ● cha in re action .
LO + AH
● LOH + A ● A H reacts w ith a lko xy ra d ica ls
to d ecrease th e
A H :A n tio xid a n ts d eco m p osition o f
h yd ro p ero xid es.
Fig. 3.1. Lipid oxidation and antioxidant mechanism.
acid (AA, 20:4n-6), α-linolenic acid (18:3n-3), and LA (18:2n-6), respectively (Gun-
stone and Hilditch, 1945; Holman and Elmer, 1947; Miyashita and Takagi, 1986;
Cosgrove et al., 1987; Cho et al., 1987a, 1987b; Miyashita et al., 1990). In the free
radical oxidation of EPA, hydrogen abstraction occurs at the C7, C10, C13, C16
position, which result in production of a pentadienyl radical between C5 and C9,
C8 and C12, C11 and C15, or C14 and C18, respectively (Fig. 3.2). The radical
then reacts at either end with oxygen to produce a mixture of 5-MHP and 9-MHP,
8-MHP and 12-MHP, 11-MHP and 15-MHP, or 14-MHP and 18-MHP, respec-
tively (Yamauchi et al., 1983). Since DHA has five bis-allylic methylene groups (Fig.
3.2), there are five possible positions for hydrogen abstraction viz., carbons 6, 9, 12,
15, or 18. Thus, oxygen can attack carbons at either end of the pentadienyl radical,
and the resulting MHP isomers are those with hydroperoxide substitution on carbons
4, 8, 7, 11, 10, 14, 13, 17, 16, 20 for DHA (VanRollins and Murphy, 1984).
Oxidation of PUFA containing more than two double bonds, such as α-linolenic
acid and AA, produces a significant amount of hydroperoxy epidioxides as the main
oxidation products–apart from monohydroperoxides-at an early stage of oxidation
(Chan et al., 1980; Coxon et al., 1981; Neff et al., 1981). Hydroperoxy epidioxides
are formed in autoxidation of EPA (Yamauchi et al., 1985) and DHA. Hydroperox-
54 K. Miyashita
19 16 13 10 7 4 2
20 18 17 15 14 12 11 9 8 6 5 3 COO H
EPA
21 18 15 12 9 6 3
COO H
22 20 19 17 16 14 13 11 10 8 7 5 4 2
DHA
Fig. 3.2. Structures of EPA and DHA.
ides of EPA or DHA are more easily decomposed to form free radicals than those of
LA, α-linolenic acid, or AA. This highly reactive EPA- or DHA-hydroperoxide may
result in a different and complicated composition of oxidation products when com-
pared with other PUFA.
When ethyl esters of α-linolenic acid, EPA and DHA were oxidized at 5°C in
the dark (Cho et al., 1987a), the ratios of OOH-oxygen to total oxygen consumption
in the oxidation of EPA or DHA were 50-70%, even in the early stages of oxidation.
Remaining oxygen was used to form secondary oxidation products, mainly polymers.
Gel permeation chromatography established that over 70% of the polar materials
from both ethyl EPA and DHA to be dimers and polymers (Cho et al. 1987a). On
the other hand, the ratio of OOH-oxygen in ethyl α-linolenate was slightly higher,
70-80%. In this case, most of the other oxygen was consumed in the formation of
hydroperoxy epidioxides. Linoleate monohydroperoxides are much more stable than
those of α-linolenate, arachidonate, EPA, and DHA. Therefore, the ratio of OOH-
oxygen in the early stage of linoleate oxidation has been reported to be over 90% (Mi-
yashita et al., 1982), although epoxyhydroxy, epoxy, dihydroxy, and trihydroxy com-
pounds were identified from autoxidized methyl linoleate and further oxidation of
linoleate MHP (Neff et al. 1978). Epoxy and dihydroxy compounds are also found in
the autoxidation of methyl linolenate (Neff et al., 1981). However, the major second-
ary oxidation products in linolenate are hydroperoxy epidioxides (Neff et al., 1981).
Although there are a limited number of reports on the oxidation products of EPA and
DHA, the major secondary oxidation products will be polymers (Cho et al., 1987a,
1987b).
Although estimation of peroxide value (PV) is the most widely used method
to assess lipid oxidation, it is not necessarily a good indication of oxidation in EPA
and DHA because of the instability of hydroperoxides formed. The determination

of secondary products-such as polymers, anisidine value, 2-thiobarbituric acid value,
sensory analysis, and gas chromatographic (GC) methods for volatile compounds-are
more important in evaluating the oxidative deterioration of EPA and DHA. Volatile
compounds analysis is a better method. The GC method is capable of determining
volatile oxidation products that are either directly responsible for or serve as mark-
ers of flavor development in oxidized EPA, DHA, and fish oil. Volatile compounds
suitable for GC determinations are mainly aldehydes and hydrocarbons. Jacobsen
(1999) conclusively showed that there is no correlation between PV and the taste
panel response on fish oil enriched spreads. However, the data on volatile compounds
obtained by headspace methods have been demonstrated to correlate well with sen-
sory data (Frankel, 1998b). A large number of volatile compounds are found in fish
oil oxidation (Karahadian and Lindsay, 1989; Hsieh et al., 1989; Aidos et al., 2002).
Sixty different volatiles comprising alkenals, alkadienals, alkatrienals, and vinyl ke-
tones have been identified in fish oil enriched milk (Venkateshwarlu et al., 2004a).
1-Penten-3-one, (Z)-4-heptenal, 1-octen-3-one, 1,5-octadien-3-one, (E,E)-2,4-hep-
tadienal, and (E,Z)-2,6-nonadienal have been reported as the most potent odorants in
fish oil (Venkateshwarlu et al., 2004a, 2004b). However, sensory impact of individual
or combinations of volatile oxidation compounds in food systems containing fish oil
is not yet established. Hence, further research is needed to use this method in evalua-
tion of fish oil deterioration.
Oxidative Stability of EPA and DHA in Phospholipids

Marine lipids contain a high percentage of long-chain PUFA, such as EPA and DHA.
Due to their high degree of unsaturation, these PUFA are much more susceptible to
oxidation compared to α-linolenic acid and LA found in vegetable oils. On the other
hand, EPA and DHA are relatively stable to oxidation in marine biological systems.
Furthermore, it has been reported that DHA ingestion does not increase lipid perox-
ides to the level one would expect from the ‘peroxidizability’ index of the tissue total
lipids (Kubo et al., 1998). In particular, lipid peroxide levels in the brain and testis
decreased when DHA was given to the animals. Ando et al. (2000) also examined the
effects of fish oil on lipid peroxidation in rat organs and found that levels of phos-
pholipid hydroperoxides and thiobarbituric acid reactive substances (TBARS) in rat
organs fed a fish oil diet were similar to those of the safflower-oil diet group. Wander
and Du (2000) measured plasma lipid peroxidation after supplementation with EPA
and DHA from fish oil and tocopherol in postmenopausal women. They found that
neither the concentration of plasma TBARS nor protein oxidation changed after fish
oil supplementation. These results indicate a difference in the oxidative stability of
highly unsaturated fatty acids, such as DHA and AA ,between biological systems and
bulk phases.
Many foods and biological systems are complex, multi-component and hetero-
geneous systems, in which lipids are present along with various types of other com-
56 K. Miyashita
ponents in aqueous medium. Furthermore, high levels of EPA and DHA in marine
lipids imply the occurrence of strong antioxidant systems in marine animal tissues.
Antioxidants, such as tocopherols, have been regarded to be most important to pre-
vent marine lipid oxidation. Thus, complex systems with antioxidants will be impor-
tant to prevent EPA and DHA oxidation.
In biological systems, a large amount of EPA and DHA are present as phospho-
lipids (PL) in the membrane. PL has been shown to act synergistically with phenolic
antioxidants, such as tocopherols. This is evidenced by the fact that PL containing
EPA and DHA would be stable to oxidation in the presence of tocopherol (Hara et
al., 1992; King et al., 1992; Ohshima et al., 1993; Segawa et al., 1994, 1995; Koga
and Terao, 1995; Takeuchi et al., 1997; Saito and Ishihara, 1997; Bandarra et al.
1999). The mechanism responsible for the synergy of tocopherols and PL is not very
well understood, but seems to be related to the involvement of the PL amino group in
the enhancement of the antioxidant activity of tocopherols, regeneration of tocoph-
erols, or chelation of prooxidant metal ions. The presence of EPA and DHA in a PL
form may be important to prevent oxidation in biological systems (Ohshima et al.,
1993; Nara et al., 2000).
Lipids from muscle and eye of squid are mainly composed of PL (>60%) and free
cholesterol (30%). Figure 3.3 shows the comparison of oxidative stabilities of these
lipids with those of other kinds of marine lipids. Marine lipids used contained high
percentages of EPA and DHA, and showed higher average numbers (1.65-2.77) of
bis-allylic positions per fatty acid molecule than those of vegetable oils, such as soy-
bean oil (0.65) and safflower oil (0.75). The average number of bis-allylic positions
in each lipid was 2.51, 2.77, 1.65, 2.19, and 1.92 for squid muscle, squid eye, tuna
orbital, trout egg, and bonito oil, respectively. Judging from the average number of
bis-allylic positions, squid eye total lipids (TL) and squid muscle TL were presumed
to be oxidized rapidly, however oxidative stabilities of these lipids were higher than
that of tuna orbital TL which showed the lowest number of bis-allylic positions. On
the basis of PV determinations, tuna orbital TL oxidized rapidly without an induc-
tion period (Fig. 3.3). Bonito oil also oxidized rapidly, but it was less susceptible to
oxidation than tuna orbital TL. The lower oxidative stabilities of these lipids were
confirmed by determining the decrease in unoxidized PUFA (Fig. 3.4) (Nara et al.,
2000). Antioxidants are the most important factors that influence oxidative stability
of lipids. The difference in the oxidative stability between tuna orbital TL and bonito
oil could be due to the higher content of tocopherols in bonito oil than that of tuna
orbital TL (Table 3.1). However, it is difficult to explain the higher oxidative stabil-
ity of squid tissue lipids and trout egg TL only by tocopherol content. As shown in
Table 3.2, three kinds of squid tissue-TL and trout egg TL contained PL, suggesting
the important role of PL on higher oxidative stabilities of these TL than those of tuna
orbital TL and bonito oil.
350
300
P e ro x id e va lu e (m e q /k g )
250
200
150
100
50
0
0 100 200 300 400 500
O x id a tio n tim e (h r)
Fig. 3.3. Changes in Peroxide Value (PV) of Lipids from marine organisms during autoxidation at 37°C. Squid
muscle total lipids (TL) (open circle); squid eye TL (open square); tuna orbital TL (solid circle); trout egg TL (solid
triangle); bonito oil (solid square).
100
U n o x id ize d P U F A (% )
80
60
0 100 200 300 400 500
O x id a tio n tim e (h r)
Fig. 3.4. Changes in unoxidized polyunsaturated fatty acids (PUFA) in lipids from marine organisms during
autoxidation at 37°C. Squid muscle total lipids (TL) (open circle); squid eye TL (open square); tuna orbital TL
(solid circle); trout egg TL (solid triangle); bonito oil (solid square).
58 K. Miyashita
Table 3.1. Tocopherol Contents of Lipids from Marine Organisms
Tocopherol Bonito
Squid Muscle TL Squid Eye TL Tuna Orbital TL Trout Egg TL
(µg/g Lipid) Oil
α-Tocopherol 649.8 1198.8 541.3 215.5 253.4
β-Tocopherol ND ND ND ND 193.3
γ-Tocopherol ND ND ND ND 703.6
δ-Tocopherol ND 9.2 ND 9.2 496.3
Total 649.8 1208.0 541.3 215.5 1646.6
ND; not detected
Table 3.2. Lipid Profiles of Squid Tissue TL, Tuna Orbital TL, Trout Egg, and Bonito Oil
Bonito
Lipid Squid Muscle TL Squid Eye TL Tuna Orbital TL Trout Egg TL
Oil
Triacylglycerols ND ND 99.3 76.8 99.6
Free Fatty Acids ND ND 0.4 ND 0.1
Glycolipids ND 6.8 ND ND ND
Sterols 23.7 28.3 ND 2.2 0.3
Phospholipids 75.6 66.4 0.2 23.1 ND
ND; not detected
The relatively higher oxidative stability of marine lipids containing EPA and
DHA in PL form in fish roe lipids compared to commercial fish oils has also been
documented (Moriya et al., in press). Salmon roe and herring roe lipids contained
23.1 and 73.6% PL, respectively (Table 3.3), while the main lipid class of fish oils was
triacylglycerol (TAG). GC analysis showed that the EPA and DHA content of fish roe
lipids were higher than those of fish oils. PL contained two acyl chain moieties and
one phosphoryl moiety (molecular weight: PC, 166.1; PE, 123.0; PS, 167.1), while
TAG contained three acyl chain moieties. Furthermore, both fish roe lipids contained
6.3 and 9.7% cholesterol (MW: 382.7), respectively. Therefore, the EPA and DHA
content of salmon and herring roe were relatively higher than those in whole lipid
compared with fish oils, which were mostly composed of TAG. Calculation of EPA
and DHA content of four kinds of lipids were fish-1, 256mg/g lipid; fish-2, 341mg/g
lipid; salmon roe lipids, 351mg/g lipids; herring roe lipids, 291mg/g lipid. The total
content of EPA and DHA in four kinds of marine lipids indicates that salmon roe lip-
ids should be oxidized most rapidly followed by fish-2, herring roe lipids, and fish-1.
However, as shown in Figures 3.5 and 3.6, herring roe lipid was most oxidatively
stable followed by salmon roe lipid, fish-2, and fish-1. The higher oxidative stability
of fish-2 than fish-1 may be due to higher content of tocopherols of fish-2 (Table 3.4).
However, it is difficult to explain the higher oxidative stability of fish roe lipids by
tocopherol contents alone. The higher oxidative stability of both roe lipids compared
with other fish oils would be due to the presence of PL in them. Kashima et al. (1984)
reported the higher synergistic effect of PE than that of PC. As shown in Table 3.5,
PE content in herring roe lipids was extremely high, but there was no PE in salmon
roe. The higher oxidative stability of herring roe, therefore, might also be due to the
antioxidant effect of herring roe PE.
Table 3.3. Lipid Class of Marine Lipids Used for Oxidation

Lipid Class (% of TL) Fish-1 Fish-2 Salmon Roe Herring Roe
Triacylglycerols 99.6 99.8 71.8 9.3

Free Fatty Acids 0.1 0.2 ND 3.8
Phospholipids ND ND 23.1 73.6
Sterols + Monoacylglycerols 0.3 ND 5.2 12.3
ND; not detected
Table 3.4. Tocopherol Contents of Marine Lipids

Tocopherol
Fish-1 Fish-2 Salmon Roe Herring Roe
(µg/g Lipid)
α-Tocopherol 253.4 60.2 19.6 22.9
β-Tocopherol 193.3 45.7 214.1 258.0
γ-Tocopherol 703.6 376.7 11.6 7.7
δ-Tocopherol 496.3 2670.9 11.3 11.5
Total 1472.6 3153.5 256.6 300.1
Table 3.5. PL Class Compositions of Herring Roe and Salmon Roe Lipids
Lipid Class (% of TL) Salmon Roe Herring Roe
PC 97.0 72.3
PE ND 6.6
PS 2.6 8.7
LysoPC ND 11.8
60 K. Miyashita
100
O x yg e n c o n s u m p tio n (% )
90
80
70
60
50
40
30
0 100 200 300 400 500 600 700 800
O xid atio n tim e (h r)
Fig. 3.5. Oxygen consumption during the oxidation of fish lipids at 37°C in the dark. Fish-1 (open diamond);
fish-2 (solid diamond); salmon roe lipids (open triangle); herring roe lipids (open circle).
1000）
160
ｘ
140
P e a k a rea o f p ro p a n a l（
120
100
80
60
40
20
0
0 10 20 30 40 50 60
O xidatio n tim e (hr)
Fig. 3.6. Propanal formation during the oxidation of fish lipids at 37°c in the dark. Fish-2 (solid diamond);
salmon roe lipids (open triangle); herring roe lipids (open circle).
Oxidative Stability of DHA in Liposomes

The oxidative stability of PUFA in an aqueous system is dramatically different from
that in the bulk phase (Miyashita et al., 1994a, 1995a, 1995b, 1997; Hirano et al.,
1997 63; Miyashita, 2002). The oxidative stability in aqueous micelles rises with in-
creasing number of bis-allylic positions in each molecule, and this relationship gets
reversed in bulk phase or in organic solvent. A similar trend has also been observed in
liposomes of phosphatidylcholine (PC) containing LA, AA, and DHA as a model of
cell membrane (Miyashita et al., 1994b; Nara et al., 1995, 1997, 1998; Araseki et al.,
2002). When three kinds of 1-palmitoyl (PA; 16:0)-2-PUFA-PC, 1-PA-2-linoleoyl
(LA;18:2n-6)-PC (PA-LA); 1-PA-2-arachidonyl (AA; 20:4n-6)-PC (PA-AA); and 1-
PA-2-docosahexaenoyl (DHA; 22:6n-3)-PC (PA-DHA), were oxidized in bulk phase
(Araseki et al., 2002) or in t-BuOH solution (Kobayashi et al., 2003) (Fig. 3.7), PA-
LA was most oxidatively stable followed by PA-AA and PA-DHA. However, the effect
of the degree of unsaturation on the oxidative stability of PC in liposomes (Figures
3.7 and 3.8) was different from that in the bulk phase and in organic solutions. Analy-
sis of oxidation products showed that PA-DHA was most oxidatively stable followed
by PA-AA and OA-LA (Fig. 3.7). On the other hand, oxygen consumption analysis
(Fig. 3.8 (A)) showed that the difference in the oxidative stability of three kinds of PC
was relatively small.
120
A m o u n t o f h yd ro x y c o m p o u n d s
100
(µm o l/m m o l lip id )
80
60
40
20
0
t-B u O H L ip o so m e t-B u O H L ip o so m e t-B u O H L ip o so m e
P A -L A P A -A A P A -D H A
Fig. 3.7. Amounts of total hydroxy compounds derived from oxidation products of PUFA in PC, Formed in
t-BuOH Solution and Liposomes. AAPH was used as oxidation inducer. The oxidation products were extracted
with chloroform/methanol, then hydrogenated, transmethylated, trimethylsilylated (TMS), and then subjected
to GC-MS analysis for the quantitative comparison of TMS-derivatives from mono-, di-, and tri-hydroxy com-
pounds. These hydroxyl compounds are derived from monohydroperoxides, hydroperoxy epidioxides, and other
oxidation products.
62 K. Miyashita
100 100
P A -L A
(A ) (B )
O x yg en concen tration (% )
80 80
U n o xidized P U D A (% )
P A -D H A
60 60
P A -D H A P A -L A
40 40 P A -A A
P A -A A
20 20
0 0
0 30 60 90 120 150 0 2 4 6 8
O xid ation tim e (m in ) O xid ation tim e (h r)
Fig. 3.8. Oxidative stability of PC in liposomes. PC (0.5 mM) was oxidized in liposomes in the presence of
AAPH (3.0 mM). The oxidative stability was evaluated by the analysis of the decrease in oxygen concentration in
the solution (A) or the decrease in unoxidized PUFA in the PC molecule (B).
Table 3.6 shows the proportion of MHP isomers formed in the oxidation of PC
in different systems (Kobayashi et al., 2003). The uneven distribution of MHP in
PC-AA liposomes indicates a preferential oxygen attack on the methyl-terminal side
of the radical. However, in case of PC-DHA, the oxygen attack favored the carboxy
terminal side of the pentadienyl radical. In PC-AA oxidation, amounts of 5-MHP +
9-MHP, 8-MHP + 12-MHP, and 11-MHP + 15-MHP were almost identical. This
result indicates same rate of hydrogen abstraction at the three bis-allylic positions
(C7, C10, and C13) and difficulty in the 1,3-cyclization of internal MHP. On the
other hand, MHP distribution in PC-DHA revealed a favorable hydrogen abstraction
at C18, C6, and C12.
PC in liposomes with highly unsaturated fatty acids, such as AA and DHA, are
more permeable and show more flexibility in fatty acid chains than those formed from
PC containing less unsaturated fatty acids such as LA. NMR analysis and molecular
dynamics simulations of PC containing DHA in liposomes indicates the wide variety
of DHA conformations (including back-bent, helical, and angle-iron conformations)
occurring in liposome systems (Everts and Davis, 2000; Feller et al., 2001; Saiz and
Klein, 2001; Huber et al., 2002). NMR analysis also showed the mobility of the hy-
drophobic part of the DHA molecule would higher than that of LA when forming
liposomes. Liposomes of AA appear to have flexibility between those of DHA and LA.
The flexibility of DHA chain conformation gives looser packing of the lipid chains
(Huster et al., 1997; Olbrich et al., 2000). Looser packing of the membrane at the
lipid-water interface brings about the high water permeability (Saiz and Klein, 2001).
Molecular dynamics simulations also indicate remarkable overlap of water molecules
with double bond regions of the DHA chain. The presence of water molecules near a
DHA molecule will lower the density of the bis-allylic hydrogen and inhibits the hy-
drogen abstraction from bis-allylic positions of unoxidized fatty acid by peroxy radical
of adjacent oxidized fatty acid in the propagation stage of free radical oxidation. The
higher water permeability of DHA and its specific conformation may be a reason for
the relatively higher oxidative stability of DHA in liposome.
Table 3.6. Isomeric Distribution of MHP Isomers Formed in the Oxidation of PUFA Esters
MHP Formation Positional Distribution

PUFA
Hydrogen Resulting In Chloroforma In PCb
In Cellularc PL
Abstraction MHP Solution Liposome
LA C-13 9-MHP 49.3±1.5 49.2±3.6 49.7±1.5
13-MHP 50.7±1.5 50.8±3.6 50.3±1.5
AA C-7 5-MHP 19.4±1.1 9.6±0.7 23.7±3.7
9-MHP 15.3±0.4 23.8±2.0 6.6±1.3
C-10 8-MHP 15.2±0.5 5.1±1.1 12.9±2.3
12-MHP 15.0±1.1 26.3±2.4 23.2±3.9
C-13 11-MHP 15.1±0.5 7.4±2.4 3.4±2.3
15-MHP 19.9±1.3 27.9±2.1 30.2±3.4
DHA C-6 4-MHP 17.7±1.3 17.9±2.4 25.7±2.2
8-MHP 12.9±1.7 11.2±1.1 12.8±1.6
C-9 7-MHP 10.5±0.7 6.2±1.1 NDd
11-MHP 7.2±1.0 2.1±0.5 NDd
C-12 10-MHP 4.2±0.5 13.0±1.1 10.2±1.3
14-MHP 4.3±0.6 4.6±0.8 4.9±4.5
C-15 13-MHP 3.5±1.0 5.5±1.0 NDd
17-MHP 4.5±0.7 3.1±0.6 NDd
C-18 16-MHP 8.7±1.4 21.3±4.0 24.6±1.5
20-MHP 26.5±1.1 16.4±1.3 21.9±4.3
a
The ethyl ester of each PUFA was oxidized in chloroform and in aqueous solution. The oxidation
was induced by 2,2’-azobis(2,4-dimethyl-valeronitrile) (AMVN) or 2,2’-azobis(2-amidino-propane)
dihydrochloride (AAPH) for chloroform and aqueous oxidation, respectively.
b
1-Palmitoyl-2-linoleoyl-phosphatidylcholine (LA-PC), 1-palmitoyl-2-arachidonoyl-phosphati-
dylcholine (AA-PC), and 1-palmitoyl-2-docosahexaenoyl-phosphatidylcholine (DHA-PC) were
oxidized in liposomes. AAPH was used as oxidation inducer.
c
Each PUFA was supplemented to the cell (HepG2). Supplementation with PUFA resulted in their
incorporation into cellular lipids. Cellular oxidation was accelerated by the addition of H2O2.
d
Not detected
Oxidation of Long-chain PUFA in Cellular Lipids

The incorporation PUFA, such as LA, AA and DHA, into cellular lipids enhances the
cellular lipid peroxidation (Igarashi and Miyazawa, 200; Araseki et al., 2005). Figure
3.9 shows the analysis of peroxidation level in the membrane PL of HepG2 cells. In
this case, the average number of bis-allylic positions per mol was 0.57 for control
64 K. Miyashita
cells, 0.64 for LA-supplemented cells, 0.77 for AA-supplemented cells, and 0.81 for
DHA-supplemented cells. However, there was little difference in the lipid peroxida-
tion level in the LA-, AA-, and DHA-supplemented cells.
Figure 3.9 also shows that the addition of H2O2 (0.5 mM) enhanced cellular
lipid peroxidation levels in control, LA-, and AA-supplemented cells compared to
those without H2O2. On the other hand, induction of lipid peroxidation by H2O2
was not observed in DHA-supplemented cells. Araseki et al. (2005) noticed the lower
cellular oxidation level in DHA-supplemented cells after H2O2 addition during the
analysis of total MHP content in the cellular PL (Table 3.7). Table 3.7 also shows that
the main source for MHP was LA in all cases, and a significant amount of AA-MHP
was observed only in AA-supplemented cells. A small amount of DHA-MHP was
also observed in DHA-supplemented cells, but not in other cells. The order of lipid
peroxidation levels of the cellular PL (Fig. 3.9; Table 3.7) was almost the same as that
in the oxidative stability of PC in liposomes, suggesting that the characteristic cellular
lipid peroxidation is also correlated with the PUFA conformation in the membrane
PL.
With the oxidation of PC-AA and PC-DHA during the oxidation of DHA in the
cellular PL, DHA-MHP was formed only by abstraction at C6, C12, and C18, but
not at the C9 or C15 positions. This characteristic distribution of DHA-MHP iso-
mers found in the cellular PL oxidation indicates uneven hydrogen abstraction from
DHA molecules in the cell membrane lipid, which might be derived from the specific
conformation of the DHA molecule and higher water permeability in the membrane
lipids. Thus, further studies are necessary to reveal the relationship between the con-
formation and physicochemical properties of AA and DHA and their oxidative stabil-
ity in biological membrane lipids.
35 0
c
30 0 c
F lu o re s c e nt in te n s ity
25 0 b
b b
20 0 b b
15 0
a
10 0
50
0
2
2
2
ol
LA
O
O
A
O
A
A
H
tr
2
2
H
H
H
D
on
+
+
C
ol
A
LA
H
tr
D
on
C
Fig. 3.9. Formation of MHP in HepG2 cells treated with or without H2O2. Values not sharing a common super-
script are significantly different at P < 0.01. Values are mean +S.D. (n = 3).
Table 3.7. Amount of MHP Formed in the Oxidation of Cellular PL with or without H2O2
Amount of MHP (mmol/mmol Lipid)

Cell
LA-MHP AA-MHP DHA-MHP Total MHP
Control 1.80±0.43 0.15±0.05 ND 1.95±0.48
Without + LA 3.06±0.40 0.17±0.08 ND 3.23±0.35

H2O2
+ AA 2.17±0.35 1.23±0.11 ND 3.40±0.29
+ DHA 2.37±0.31 0.21±0.06 0.36±0.05 2.93±0.20
Control 3.10±0.62 0.32±0.09 ND 3.42±0.71
With + LA 3.50±0.46 0.17±0.06 ND 3.67±0.41

H2O2
+ AA 2.47±0.47 1.55±0.57 ND 4.02±1.03
+ DHA 2.27±0.47 0.20±0.09 0.33±0.10 2.79±0.65
Data re expressed as mean ± SD (n = 3)
ND; not detected
Formation of Conjugated Trienoic Acids in Oils Containing

Long-chain PUFA
Crude oil is refined by a series of processes to remove impurities that affect taste,
smell, appearance, and stability of the oil. The refining processes involve degumming,
alkali refining, bleaching and deodorization. PUFA, such as LA, α-linolenic acid,
EPA, and DHA, found in vegetable and marine oils are more susceptible to oxidation
during storage and refining of crude oils resulting in oxidation products compris-
ing mainly of hydroperoxides. Hydroperoxides may also be formed by enzymatic
oxidation of linoleate and linolenate by lipoxygenase during storage of oil seeds. Any
oxidation is injurious to the flavor and oxidative stability of refined oils as it results
in the formation and accumulation of oxidation products. The quality of a processed
vegetable oil always depends on the quality of the crude oil used and the processing
parameters selected. Indicators such as PV in the final products can be reduced by
lowering storage temperature and minimizing exposure to prooxidants, like air, light,
and heat, during processing. However, complete removal of all oxidation products
from oils is impossible.
Yurawecz et al. (1993) found traces (up to 0.2%) of conjugated linolenic acid
(CLN) in vegetable oils in their study of 27 oils for CLN content by UV measure-
ment. The isomers were identified as α-eleostearic acid (9cis(c),11trans (t),13t-18:3),
β-eleostearic acid (9t,11t,13t-18:3), and 8t,10t,12t-18:3. The possible mechanism for
formation of these CLN isomers involves linoleate oxidation, reduction of hydroper-
oxide to hydroxide, and dehydration (Fig. 3.10) (Parr and Swoboda, 1976; Fishwick
and Swoboda, 1977). CLN and other conjugated trienoic fatty acids are easily oxi-
dized to induce lipid oxidation and flavor deterioration even if the content is very low
(Suzuki et al., 2004).
HPLC analysis of purified soybean oil methyl esters indicates the presence of
66 K. Miyashita
small amounts of conjugated linoleic acid (CLA) and CLN as shown by UV detection
at 233 nm and 274 nm (Table 3.8) (Kinami et al., 2007). GC-MS after conversion of
methyl esters to DMOX derivatives and by comparison with authentic CLN isomers
on HPLC revealed the formation of 8,10,12- and 9,11,13-18:3 (c,t,t, or t,t,c, and
t,t,t) in the purified soybean oil. HPLC chromatogram of crude soybean oil and pro-
cessed soybean oil at different stages shows that a significant amount of CLN (8,10,12
or 9,11,13) was found in soybean oil after bleaching, although it could barely be
detected in crude soybean oil or in the oil after degumming and alkali refining (Table
3.9). A slight decrease in CLN after deodorization may be due to the isomerization of
the CLN to CLN with conjugated dienes (8,10,13-18:3). CLN contents in purified
soybean oil from different companies in Japan are shown in Table 3.10. It varies from
387-1316 mg/kg oil, which corresponds to 0.039-0.13% (w/w). Similar values have
also been reported by Yurawecz et al. (1993). CLN content is also affected by bleach-
ing conditions (Kinami et al., 2007). Combinations of higher percentage of bleach-
ing earth and lower bleaching temperature results in reduced CLN content. Similar
effects of bleaching temperature and earth combinations have been reported by Van
Den Bosch (1973a,1973b).
12 9
L in oleate
O2 O2
OOH OOH
13-H yd ro p ero xid es 9-H yd ro p e ro xides
R ed u c tio n R ed u c tio n
OH OH
H yd ratio n H yd ratio n
9c,11t,13t-18:3 8t,10t,12c -18:3
9t,11t,13t-18:3 8t,10t,12t-18:3
Fig. 3.10. A possible scheme for formation of cln from linoleate hydroperoxides. F ig. 10
Table 3.8. Identification of trans Conjugated Fatty Acids in Commercial Soybean Oil
Conjugation
Structure
cis (c), trans (t) Configuration
8,10,13-18:3 (CLN) Conjugated diene with trans configuration

8,10,12-18:3 (CLN) Conjugated triene;
9,11,13-18:3 (CLN) c,t,t, or t,t,c
7,9-18:2 (CLA) Conjugated diene;
10,12-18:2 (CLA) c,t or t,c
8,10,12-18:3 (CLN) Conjugated triene;

9,11,13-18:3 (CLN) t,t,t
7,9-18:2 (CLA) Conjugated diene;
9,11-18:2 (CLA t.t
Table 3.9. Content of CLN (8,10,12 or 9,11,13) with Conjugated Triene at Different Stage of
Soybean Oil Production
CLN (mg/kg oil)
Soybean Oil
c,t,t or t,t,c t,t,t Total
Crude ND ND ND
Degumming 10 ND 10
Alkali Refining 20 ND 20
Bleaching 250 358 608

Deodorization 217 315 532
ND; not detected
Table 3.10. CLN Content in Commercial Soybean Oil from Different Japanese Oil Companies
CLN (mg/kg oil)
Company
c,t,t or t,t,c t,t,t Total
A 295 367 662
B 408 618 1026
C 198 230 427
D 753 559 1312

E 181 205 387
F 945 370 1316
68 K. Miyashita
Soybean oil tends to develop an undesirable flavor and odor known as reversion
when PV is still as low as a few meq/kg. This reversion flavor is characterized as beany
and grassy. Guth and Grosch (1991) reported the formation of a potent compound
causing this flavor reversion as 3-methyl-2,4-nonanedione. The reversion flavor in
soybean oil is mainly due to furan compounds such as 2-pentyl furan (Smouse and
Chang, 1967) and 2-pentenyl furan (Ho et al., 1978; Smagula et al., 1979). Min et
al. (2003) and Cho and Min (2006) clearly show that these furan compounds are
formed from the photooxidation of LA and linolenic acid in soybean oil. Photooxi-
dation is due to singlet oxygen that is produced from triplet oxygen in the presence
of chlorophyll under light. On the other hand, conjugated fatty acids have also been
considered as a possible source of furan fatty acids (Yurawecz et al., 1995). A trace
amount of CLN plays a vital role in accelerating lipid oxidation. CLN may be one
of the important precursors for off-flavor compounds since CLN is easily oxidized in
bulk phase (Suzuki et al., 2004).
Table 3.11 indicates intensity of beany and grassy flavor of middle chain fatty
acid (MCT) triacyglycerol (TAG) with or without bitter gourd TAG. The main fatty
acids of bitter gourd TAG are 9c,11t,13t-18:3 (61.6%), 18:0 (25.7%), and 18:1n-9
(6.1%) (Suzuki et al. 2001; Narayan et al., 2006b). Little flavor was detected in MCT
after heating at 180°C under light (7000 lux) for 16 hr. Beany and grassy flavors were
detected in MCT by adding bitter gourd-TAG, although PV of the oil was less than
0.1. This result suggested that CLN (9c,11t,13t-18:3) of bitter gourd-TAG would be
one of the potent compounds responsible for flavor reversion. The flavor significance
of soybean oil after heating under the light is shown in Table 3.12. CLN content
changed depending on bleaching earth content and temperature. In the purified soy-
bean oil, the relationship between CLN content and the flavor intensity was not con-
siderable. On the other hand, the addition of 0.5% activated carbon reduced beany
and grassy flavor. Frankel (1998c) noted the importance of oxidative polymers on the
flavor or oxidative deterioration of vegetable oils. The spontaneous decomposition
of oxidative polymers has been referred to as “hidden oxidation” because it cannot
be measured by PV. CLN is easily oxidized to produce polymers as main oxidation
products (Suzuki et al., 2004), which may induce the oxidation of vegetable oils.
Table 3.11. Beany and Grassy Flavor Intensity of MCT with or without Bitter Gourd-TAG after Heating at 180°C
under 700 Lux for 16 hr
Bitter-Gourd TAG (%) PV (meq/kg) Flavor intensity
0 0 -
0.01 0.1 ++
0.05 0.1 +++

Table 3.12. Beany and Grassy Flavor Intensity of Soybean Oil after Heating at 180°C under 7,000 Lux for 16 hr
Bleaching Condition CLN (mg/kg Oil) PV

(meq/
Flavor
Activated Activated kg
Temp. (°C) c,t,t or t,t,c t,t,t Total Oil)
Earth (%) Carbon
1.5 110 - 173 284 457 1.3 ++

1.5 110 - 153 255 408 1.2 ++
0.5 110 - 122 166 288 2.3 +++
1.5 50 - 117 186 303 1.4 ++
0.5 50 - 68 81 149 2.1 +++
0.5 50 - 65 75 140 1.9 +++
0.5 50 - 106 127 233 1.1 +
0.5 50 + 67 84 151 0.9 +
0 50 + ND 40 40 1.3 +
The presence of conjugated trienoic acids has also been reported in the purified
fish oil from Japanese sardine and tuna by HPLC equipped with a photodiode-array
detector (Fig. 3.11). GC-MS indicated that the main conjugated trienoic acids were
those of 16:3, 20:5, and 22:6. Conjugated 16:3 trienoic acid was identified as cyclo-
propanoid acid with conjugated triene by NMR analysis. These conjugated trienoic
acids were observed after deodorization, but not after degumming, alkali refining, or
bleaching. The formation of conjugated EPA (CEPA) and conjugated DHA (CDHA)
during deodorization can be due to the double bond migrations of EPA and DHA
by heating. In the case of 16:4n-1, double bond migration and cyclization occur to
produce corresponding cyclopropanoid 16:3 acid. These reactions were confirmed by
the formation of conjugated trienoic acids from the heat treatment (200°C for 2hr) of
purified methyl esters of 16:4n-1, EPA, and DHA.
The oxidation rate of conjugated trienoic acids of linolenate were much higher
than those of α-linolenic acid esters and CLA esters (Suzuki et al., 2004), suggesting
the higher susceptibility of conjugated trienoic fatty acid to oxidation than conju-
gated dienoic acids and fatty acid having the same number of double bonds. Since
oxidation rate increases with increasing double bonds, 16:4n-1, EPA and DHA are
more easily oxidized than α-linolenic acid. Therefore conjugated trienoic fatty acids
originating from 16:4n-1, EPA and DHA in fish oil would be more susceptible to
oxidation. They may have an important role in the oxidative deterioration of fish oil
in spite of these conjugated fatty acids being present in trace amounts.
Conclusions
It is generally accepted that lipid oxidation of DHA and EPA is a major problem in
70 K. Miyashita
C o n ju g a ted 16:3
(A )
CEPA
CDHA
0 10 20 30 40 50 60 70 80 90 100
CEPA
(B )
CDHA
0 10 20 30 40 50 60 70 80 90 100
R eten tio n tim e (m in )
Fig. 3.11. HPLC of methyl esters from purified sardine oil (A) and purified tuna oil (B). HPLC was carried out with
an analytical reversed-phase column (C30). A mixture of methanol and water (85:15, vol/vol) at a flow rate of
1.0 mL/min (analytical scale) was used as a mobile phase. The HPLC instrument housed a photodiode-array
spectrophotometric detector.
using fish oil to make food materials, and a high intake of these PUFA may increase
the oxidative stress of biological systems. Susceptibility of PUFA to oxidation de-
pends on the availability of bis-allylic hydrogens. Oxidative stability of each PUFA is
inversely proportional on the number of bis-allylic positions in the molecule or the
degree of unsaturation of the PUFA. Due to the higher degree of unsaturation, MHP
of EPA and DHA are easily decomposed, resulting in complex oxidation products
when compared with those of other PUFA, such as LA and α-linolenic acid. A large
amount of polar oxidation products, mainly polymers, form even at an early stage of
autoxidation of EPA and DHA. Peroxide value is not necessarily a good indication of
oxidation in EPA and DHA because of the instability of their hydroperoxides.
On the other hand, oxidative stability of EPA and DHA in biological systems
is different from those in bulk phase or in organic solution. In biological systems
lipids are present with various types of other components in aqueous medium. DHA
is relatively stable in biological systems due to the complex, multi-component, and
heterogeneous nature of biological systems. PL is synergistic in combination with
phenolic antioxidants such as tocopherols. In biological systems large amounts of
EPA and DHA are present as PL in the membrane. Furthermore, the high levels of
DHA in membrane lipids imply the occurrence of a strong antioxidant system in
animal tissues.
In addition, the oxidative stability of PUFA in a liposome system is different
from that in bulk phase. The oxidative stability of DHA is slightly higher than AA and
LA in liposomes. NMR analysis and molecular dynamic simulations of PC contain-
ing DHA in liposomes indicates that DHA chain conformation gives looser packing
of the lipid chains in liposome. The looser packing of the membrane at the lipid-water
interface brings about the high water permeability. The presence of water molecules
near a DHA molecule will lower the density of the bis-allylic hydrogen and reduce the
chain-carrying reaction of lipid peroxidation. The higher water permeability of DHA
and its specific conformation may be a reason for the relatively higher oxidative stabil-
ity of DHA in liposomes. The difference in the oxidative stability of PC in liposomes
and in the bulk phase or an organic solvent is due to the specific conformation of PC
bilayers in the liposomes. The order of lipid peroxidation levels of the cellular PL was
almost the same as that in the oxidative stability of PC in liposomes, suggesting that
the characteristic cellular lipid peroxidation found in the present study is also corre-
lated with the PUFA conformation in the membrane PL.
Finally, the impact of conjugated trienoic fatty acids, such as CLN, CEPA, and
CDHA, on oxidative deterioration and flavor reversion is also an important factor in
lipid oxidation. It can be said that mechanisms to prevent PUFA oxidation have been
well understood when applied to lipid-containing foods. However, In conclusion, it
is still difficult to completely prevent the PUFA oxidation, especially flavor deteriora-
tion. The formation of conjugated triene compounds and their strong impact on oxi-
dative deterioration show the importance of the original quality and refining process
of oils.
References
Aidos, I.; C. Jacobsen; B. Jensen; J.B. Luten; A. van der Padt; and R.M. Boom. Volatile Oxi-
dation Products Formed in Crude Herring Oil under Accelerated Oxidative Conditions.
Eur. J. Lipid Sci. Technol. 2002, 104, 808-818.
Ando, K.; K. Nagata; R. Yoshida; K. Kikugawa; and M. Suzuki. Effect of n-3 Polyunsaturated
Fatty Acid Supplementation on Lipid Peroxidation of Rat Organs. Lipids 2000, 35, 401-
407.
Araseki, M.; K. Yamamoto; and K. Miyashita. Oxidative Stability of Polyunsaturated Fatty
Acid in Phosphatidylcholine Liposomes. Biosci. Biotechnol. Biochem. 2002, 66, 2573-
2577.
Araseki, M.; H. Kobayashi; M. Hosokawa; and K. Miyashita. Lipid Peroxidation of a Human
Hepatoma Cell Line (HepG2) after Incorporation of Linoleic Acid, Arachidonic Acid,
and Docosahexaenoic Acid. Biosci. Biotechnol. Biochem. 2005, 69, 483-490.
Bandarra, N.M.; R.M. Campos; I. Batista; M.L. Nunes; and J.M. Empis. Antioxidant Synergy
of a-Tocopherol and Phospholipids. J. Am. Oil Chem. Soc. 1999, 76, 905-913.
Benedetti, A.; M. Comporti; and H. Esterbauer. Identification of 4-Hydroxynonenal as a Cy-
totoxic Product Originating from the peroxidation of liver microsomal lipids, Biochim.
72 K. Miyashita
Biophys. Acta 1980, 620, 281-296.

Burns, C.P.; and B.A. Wagner. Heightened Susceptibility of Fish Oil Polyunsaturated-En-
riched Neoplastic Cells to Ethane Generation during Lipid Oxidation. J. Lipid Res. 1991,
32, 79-87.
Chan, H.W.-S.; J. Matthew; and D.T. Coxon. A Hydroperoxy-Epidioxide from the Autoxida-
tion of a Hydroperoxide of Methyl Linolenate. J. C. S. Chem. Comm. 1980, 235-236.
Cho, E.; and D.B. Min. Chemistry and Reactions of Reactive Oxygen Species in Foods. Criti-
cal Rev., Food Sci. Nutr. 2006, 46, 1-22.
Cho, S.-Y.; K. Miyashita; T. Miyazawa; K. Fujimoto; and T. Kaneda. Autoxidation of Ethyl
Eicosapentaenoate and Docosahexaenoate. J. Am. Oil Chem. Soc. 1987a, 64, 876-879.
Cho, S.-Y.; K. Miyashita; T. Miyazawa; K. Fujimoto; and T. Kaneda. Autoxidation of Ethyl
Eicosapentaenoate and Docosahexaenoate under Light Irradiation. Nippon Suisan Gak-
kaishi 1987b, 53, 813-817.
Cosgrove, J.P.; D.F. Church; and W.A. Pryor. The Kinetics of the Autoxidation of Polyunsatu-
rated Fatty Acids. Lipids 1987, 22, 299-304.
Coxon, D.T.; K.R. Price; and H.W.-S. Chan. Formation, Isolation and Structure Determina-
tion of Methyl Linolenate Diperoxides. Chem. Phys. Lipids 1981, 28, 365-378.
Everts, S.; and J.H. Davis. 1H and 13C NMR of Multilamellar Dispersions of Polyunsaturated
(22:6) Phospholipids. Biophys. J. 2000, 79, 885-897.
Feller, S.E.; K. Garwrish; and A.D. MacKerell Jr. Polyunsaturated Fatty Acids in Lipid Bilayers
Intrinsic and Environmental Contributions to Their Unique Physical Properties. J. Am.
Chem. Soc. 2001, 124, 318-326.
Fishwick, M.J.; and P.A.T. Swoboda. Measurement of Oxidation of Polyunsaturated Fatty
Acids by Spectrophotometric Assay of Conjugated Derivatives. J. Sci. Food Agric. 1977,
28, 387-393.
Frankel, E.N. Lipid Oxidation. Prog. Lipid Res. 1980, 19, 1-22.
Frankel, E.N. Volatile Lipid Oxidation Products. Prog. Lipid Res. 1982, 22, 1-33.
Frankel, E.N. Chemistry of Free Radical and Singlet Oxidation of Lipids. Prog. Lipid Res.
1985, 23, 197-221.
Frankel, E.N. Chemistry of autoxidation: mechanism, products and flavor significance, in Fla-
vor Chemistry of Lipid Foods Min, D., and Smouse, T.H., eds, AOCS Press, Champaign,
Illinois, 1985.
Frankel, E.N. Lipid Oxidation. The Oily Press: Dundee, Scotland, 1998a, pp. 13-77.
Frankel, E.N. Lipid Oxidation, The Oily Press: Dundee, Scotland, 1998b, pp. 99-114.
Frankel, E.N. Lipid Oxidation, The Oily Press: Dundee, Scotland, 1998c, pp. 115-127.
Fritsche, K.L.; and P.V. Johnston. Rapid Autoxidation of Fish Oil in Diets without Added
Antioxidants. J. Nutr. 1988, 118, 425-426.
Garrido, A.; F. Garrido; R. Guerra; and A. Valenzyela. Ingestion of High Doses of Fish Oil
Increases the Susceptibility of Cellular Membranes to the Induction of Oxidation Stress.
Lipids 1989, 24, 833-835.
Gunstone, F.D.; and T.P. Hilditch. The Union Gaseous Oxygen with Methyl Oleate, Linole-
ate, and Linolenate. J. Chem. Soc. 1945, 836-841.
Guth, H.; and W. Grosch. Detection of Furanoid Fatty Acids in Soya-Bean Oil-Cause for the
Light-Induced Off-Flavor. Fat Sci. Technol. 1991, 93, 249-255.
Hara, S.; N. Okada; H. Hibino; and Y. Totani. Antioxidative Behavior of Phospholipids for
Polyunsaturated Fatty Acids of Fish Oil. J. Jpn. Oil Chem. Soc. 1992, 41, 130-135.
Hirano, S.; K. Miyashita; and T. Ota; M. Nishikawa; K. Maruyama; and S. Nakayama. Aque-
ous Oxidation of Ethyl Linoleate, Ethyl Linolenate, and Ethyl Docosahexaenoate. Biosci.
Biotechnol. Biochem. 1997, 161, 281-285.
Ho, C.T.; M.S. Smagula; and S.S. Chang. The Synthesis of 2-(1-Pentenyl) Furan and Its Re-
lationship to the Reversion Flavor of Soybean Oil. J. Am. Oil Chem. Soc. 1978, 55, 233-
237.
Holman, R.T. and Elmer, O.C. The rates of oxidation of unsaturated fatty acids and esters, J.
Am. Oil Chem. Soc., 1947, 24, 127-129.
Hsieh, T.C.Y.; S.S. Williams; W. Vejaphan; and S.P. Meyers. Characterization of Volatile Com-
ponents of Menhaden Fish (Brevoortia tyrannus) Oil. J. Am. Oil Chem. Soc. 1989, 66,
114-117.
Hu, M.-L.; E.N. Frankel; B.E. Leibovitz; and A.L. Tappel. Effect of Dietary Lipids and Vi-
tamin E on in vitro Lipid Peroxidation in Rat Liver and Kidney Homogenates. J. Nutr.
1989, 119, 1574-1582.
Huber, T.; K. Rajamoorthi; V.F. Kurze; K. Beyer; and M.F. Brown. Structure of Docosahexae-
noic Acid-Containing Phospholipid Bilayers as Studied by 2H NMR and Molecular Dy-
namics Simulations. J. Am. Chem. Soc. 2002, 124, 298-309.
Huster, D.; A.J. Jin; K. Arnold; and K. Gawrisch. Water Permeability of Polyunsaturated Lipid
Membranes Measured by 17O NMR. Biophys. J. 1997, 73, 855-864.
Igarashi, M.; and T. Miyazawa. The Growth Inhibitory Effect of Conjugated Linoleic Acid on
a Human Hepatoma Cell Line, HepG2, Is Induced by a Change in Fatty Acid Metabo-
lism, But Not the Facilitation of Lipid Peroxidation in the Cells. Biochim. Biophys. Acta
2001, 1530, 162-171.
Jacobsen, C. Sensory Impact of Lipid Oxidation in Complex Food Systems. Fett/Lipid 1999
101, 484-492.
Kamal-Eldin, A., Mäkinen, M., Lampi, A.-M. The challenging contribution of hydroperox-
ides to the lipid oxidation mechanism, in Lipid Oxidation Pathways, Kamal-Eldin ed, pp.
1-36, AOCS Press, Champaign, Illinois, 2003.
Kaneda, T.; and S. Ishii. Nutritive Value or Toxicity of Highly Unsaturated Fatty Acids. I. J.
Biochem. 1954, 41, 327-335.
Karahadian, C.; and R.C. Lindsay. Evaluation of Compounds Contributing Characterizing
Fishy Flavors in Fish Oils. J. Am. Oil Chem. Soc. 1989, 66, 953-960.
Kashima, M.; G.-S. Cha; Y. Isoda; J. Hirano; and T. Miyazawa. The Antioxidant Effects of
Phospholipids on Perilla Oil. J. Am. Oil Chem. Soc. 1984, 61, 950-954.
Kinami, T.; N. Horii; B. Narayan; S. Arato; M. Hosokawa; K. Miyashita; H. Negishi; J. Ikui-
na; R. Noda; and S. Shirasawa. Occurrence of Conjugated Linolenic Acids in Purified
Soybean Oil. J. Am. Oil Chem. Soc. 2007, 84, 23-29.
King, M.F.; L.C. Boyd; and B.W. Sheldon. Effects of Phospholipids on Lipid Oxidation of a
Salmon Oil Model System. J. Am. Oil Chem. Soc. 1992, 69, 237-242.
Kobayashi, H.; M. Yoshida; and K. Miyashita. Comparative Study of the Product Compo-
nents of Lipid Oxidation in Aqueous and Organic Systems. Chem. Phys. Lipids 2003,
126, 111-120.
Koga, T.; and J. Terao. Phospholipids Increase Radical-Scavenging Activity of Vitamin E in a
Bulk Oil Model System. J. Agric. Food Chem. 1995, 43, 1450-1454.
Kubo, K.; M. Saito; T. Tadokoro; and A. Maekawa. Dietary Docosahexaenoic Acid Does Not
Promote Lipid Peroxidation in Rat Tissue to the Extent Expected from Peroxidizability
Index of the Lipids. Biosci. Biotechnol. Biochem. 1998, 62, 1698-1706.
Lands, W.E.M. Fish, Omega-3 and Human Health, 2nd Edn.; AOCS Press: Champaign, Il-
74 K. Miyashita
linois, 2005.
Li, D.; O. Bode; H. Drummond; and A.J. Sinclair. Omega-3 (n-3) Fatty Acids. In Lipids
for Functional Foods and Nutraceuticals, Gunstone, F.D. Ed., The Oily Press; Bridgwater,
England, 2003 pp. 225-262.
Min, D.; and T.H. Smouse. Flavor Chemistry of Fats and Oils, AOCS Press: Champaign, Il-
linois, 1985.
Min, D.B., Lee, S.H., Lee, E.C. Singlet oxygen oxidation of vegetable oils, in Flavor Chemistry
of Lipid Foods, Min, D., and Smouse, T.H., eds, pp. 57-97, AOCS Press, Champaign,
Illinois, (1989.
Min, D.B.; A.L. Callison; and H.O. Lee. Singlet Oxygen Oxidation for 2-Pentylfuran and 2-
Pentenylfuran Formation in Soybean Oil. J. Food Sci. 2003, 68, 1175-1178.
Miyashita, K. Polyunsaturated Lipids in Aqueous Systems Do Not Follow Our Preconceptions
of Oxidative Stability. Lipid Technol. News. 2002, 8, 35-41.
Miyashita, K. Effects of Flexibility and Permeability of Polyunsaturated Fatty Acid Molecules
on Their Oxidative Stability in Aqueous Systems. Lipid Technol. News. 2004, 16, 197-
202.
Miyashita, K.; and T. Takagi. Study on the Oxidative Rate and Prooxidant Activity of Free
Fatty Acids. J. Am. Oil Chem. Soc. 1986, 63, 1380-1384.
Miyashita, K.; K. Fujimoto; and T. Kaneda. Formation of Dimmers during the Initial Stage of
Autoxidation in Methyl Linoleate. Agric. Biol. Chem. 1982, 46, 751-755.
Miyashita, K.; E.N. Frankel; W.E. Neff; and R.A. Awl. Autoxidation of Polyunsaturated Tria-
cylglycerols. III. Syntheytic Triacylglycerols Containing Linoleate and Linolenate. Lipids
1990, 25, 48-53.
Miyashita, K.; N. Tateda; and T. Ota. Oxidative Stability of Free Fatty Acid Mixtures from
Soybean, Linseed, and Sardine Oils in an Aqueous Solution. Fisheries Sci. 1994a, 60,
315-318.
Miyashita, K.; E. Nara; and T. Ota. Comparative Study on the Oxidative Stability of Phospha-
tidylcholines from Salmon Egg and Soybean in an Aqueous Solution. Biosci. Biotechnol.
Biochem. 1994b, 58, 1772-1775.
Miyashita, K.; M. Hirao; E. Nara; and T. Ota. Oxidative Stability of Triglycerides from Orbital
Fat of Tuna and Soybean Oil in an Emulsion. Fisheries Sci. 1995a, 61, 273-275.
Miyashita, K.; G. Azuma; and T. Ota. Oxidative Stability of Geometric and Positional Isomers
of Unsaturated Fatty Acids in Aqueous Solution. J. Jpn. Oil Chem. Soc. 1995b, 44, 425-
430.
Miyashita, K; S. Hirano; Y. Itabashi; T. Ota; M. Nishikawa; and S. Nakayama. Oxidative
Stability of Polyunsaturated Monoacylglycerol and Triacylglycerol in Aqueous Micelles. J.
Jpn. Oil Chem. Soc. 1997, 46, 205-208.
Moriya, H.; T. Kuniminato; M. Hosokawa; K. Fukunaga; T. Nishiyama; and K. Miyashita.
Oxidative Stability of Salmon and Herring Roe Lipids and Their Dietary Effect on Plasma
Cholesterol Levels of Rats. Fisheries Sci., in press.
Nara, E.; K. Miyashita; and T. Ota. Oxidative Stability of PC Containing Linoleate and Doco-
sahexaenoate in an Aqueous Solution with or without Chicken Egg Albumin. Biosci.
Biotechnol. Biochem. 1995, 59, 2319-2320.
Nara, E.; K. Miyashita; and T. Ota. Oxidative Stability of Liposomes Prepared from Soybean
PC, Chicken Egg PC, and Salmon Egg PC. Biosci. Biotechnol. Biochem. 1997, 61, 1736-
1738.
Nara, E.; K. Miyashita; T. Ota; and Y. Nadachi. The Oxidative Stabilities of Polyunsaturated
Fatty Acids in Salmon Egg Phosphatidylcholine Liposomes. Fisheries Sci. 1998, 64, 282-
286.
Nara, E.; K. Yamamoto; A. Hirose; M. Kotake; and K. Miyashita. Antioxidant Systems in
Squid Eyes: Lipid Profiles and the Presence of Water Soluble Antioxidants. J. Jpn. Oil
Chem. Soc. 2000, 49, 53-58.
Narayan, B.; K. Miyashita; and M. Hosokawa. Physiological Effects of Eicosapentaenoic Acid
(EPA) and Docosahexaenoic Acid (DHA)-A Review. Food Rev. Inter. 2006a, 22, 291-
307.
Narayan, B.; M. Hosokawa; and K. Miyashita. Occurrence of Conjugated Fatty Acids in
Aquatic and Terrestrial Plants and Their Physiological Effects. In Nutraceutical and Spe-
cialty Lipids and Their Co-Products, Shahidi, F. Ed., CRC Taylor and Francis: New York,
2006b, pp. 201-218.
Neff, W.E.; E.N. Frankel; C.R. Scholfield; and D. Weisleder. High Pressure Liquid Chroma-
tography of Autoxidized Lipids: I. Methyl Oleate and Methyl Linoleate. Lipids 1978, 13,
415-421.
Neff, W.E.; E.N. Frankel; and D. Weisleder. High Pressure Liquid Chromatography of Au-
toxidized Lipids: II. Hydroperoxy-Cyclic Peroxides and Other Secondary Products from
Methyl Linolenate. Lipids 1981, 16, 439-448.
Ohshima, T.; Y. Fujita; and C. Koizumi. Oxidative Stability of Sardine and Mackerel Lipids
with Reference to Synergism between Phospholipids and α-Tocopherol. J. Am. Oil Chem.
Soc. 1993, 170, 269-276.
Olbrich, K.; W. Rawicz; D. Needham; and E. Evans. Water Permeability and Mechanical
Strength of Polyunsaturated Lipid Bilayers. Biophys. J. 2000, 79, 321-327.
Parr, L.J.; and P.A.T. Swoboda. The Assay of Conjugable Oxidation Products Applied to Lipid
Deterioration in Stored Foods. J. Food Technol. 1976, 11, 1-12.
Piche, L.A.; H.H. Draper; and P.D. Cole. Malondialydehyde Excretion by Subjects Consum-
ing Cod Liver Oil vs a Concentrate of n-3 Fatty Acids. Lipids 1988, 23, 370-371.
Porter, N.A.; S.E. Galdwell; and K.A. Mills. Mechanisms of Free Radical Oxidation of Unsatu-
rated Lipids. Lipids 1995, 30, 277-290.
Privett, O.S. and Blank, M.L. The initial stages of autoxidation, J. Am. Oil Chem. Soc., 1962,
39, 465-469.
Saito, H.; and K. Ishihara. Antioxidant Activity and Active Sites of Phospholipids as Antioxi-
dants. J. Am. Oil Chem. Soc. 1997, 74, 1531-1536.
Saiz, L.; and M.L. Klein. Structural Properties of a Highly Polyunsaturated Lipid Bilayer from
Molecular Dynamics Simulations. Biophys. J. 2001, 81, 204-216.
Segawa, T.; S. Hara; and Y. Totani. Antioxidative Behavior of Phospholipids for Polyunsatu-
rated Fatty Acids of Fish Oil. II. Synergistic Effect of Phospholipids for Tocopherol. J.
Jpn. Oil Chem. Soc. 1994, 43, 515-519.
Segawa, T.; M. Kamata; S. Hara; and Y. Totani. Antioxidative Behavior of Phospholipids for
Polyunsaturated Fatty Acids of Fish Oil. III. Synergistic Mechanism of Nitrogen Includ-
ing Phospholipids for Tocopherol. J. Jpn. Oil Chem. Soc. 1995, 44, 36-42.
Shahidi, F.; and M. Miraliakbari. Marine Oils: Compositional Characteristics and Health Ef-
fects. In Nutraceutical and Specialty Lipids and Their Co-Products, Shahidi, F. Ed., CRC
Press: New York, 2006, pp. 227-250.
Sinclair, A.; J. Wallace; M. Martin; N. Attar-Bashi; R. Weisinger; and D. Li. The Effects of
Eicosapentaenoic Acid in Various Clinical Conditions. In Healthful Lipids, Akoh, C.C.;
and Lai, O-M. Eds., AOCS Press: Champaign, Illinois, 2005, pp. 361-394.
76 K. Miyashita
Smagula, M.S.; C. Ho; and S.S. Chang. The Synthesis of 2-(2-Pentenyl) Furans and Their
Relationship to the Reversion Flavor of Soybean Oil. J. Am. Oil Chem. Soc. 1979, 56,
516-519.
Smouse, T.H.; and S.S, Chang. A Systematic Characterization of the Reversion Flavor of Soy-
bean Oil. J. Am. Oil Chem. Soc. 1967, 44, 509-514.
Suzuki, R.; R. Noguchi; T. Ota; M. Abe; K. Miyashita; and T. Kawada. Cytotoxic Effect of
Conjugated Trienoic Fatty Acids on Mouse Tumor and Human Monocytic Leukemia
Cells. Lipids 2001, 36, 477-482.
Suzuki, R.; M. Abe; and K. Miyashita. Comparative Study of the Autoxidation of TAG Con-
taining Conjugated and Nonconnugated C18 PUFA. J. Am. Oil Chem. Soc. 2004, 81,
563-569.
Takeuchi, M.; S. Hara; Y. Totani; H. Hibino; and Y. Tanaka. Antioxidative Behavior of Poly-
unsaturated Phospholipids. I. Oxidative Stability of Marine Oil Containing Polyunsatu-
rated Phospholipids. J. Jpn. Oil Chem. Soc. 1997, 46, 175-181.
Van Den Bosch, G. Bleaching of Vegetable Oils: I. Conversions in Soybean Oil, Triolein and
Trilinolein. J. Am. Oil Chem. Soc. 1973a, 50, 421-423.
Van Den Bosch, G. Bleaching of Vegetable Oils: II. Conversions of Methyl Oleate and Linole-
ate. J. Am. Oil Chem. Soc. 1973b, 50, 487-493.
VanRollins, M.; and R.C. Murphy. Autooxidation of Docosahexaenoic Acid: Analysis of Ten
Isomers of Hydroxydocosahexaenoate. J. Lipid Res. 1984, 25, 507-217.
Venkateshwarlu, G.; M.B. Let; A.S. Meyer; and C. Jacobsen. Chemical and Olfactometric
Characterization of Volatile Flavor Compounds in a Fish Oil Enriched Milk Emulsion. J.
Agric. Food Chem. 2004a, 52, 311-317.
Venkateshwarlu, G.; M.B. Let; A.S. Meyer; and C. Jacobsen. Modeling the Sensory Impact of
Defined Combinations of Volatile Lipid Oxidation Products on Fishy and Metallic Off-
Flavors. J. Agric. Food Chem. 2004b, 52, 1635-1641.
Wander, R.C.; and S.-H. Du. Oxidation of Plasma Proteins Is Not Increased after Supplemen-
tation with Eicosapentaenoic and Docosahexaenoic Acids. Am. J. Clin. Nutr. 2000, 72,
731-737.
Yamauchi, R.; T. Yamada; and Y. Ueno. Monohydroperoxides Formed by Autoxidation and
Photosensitized Oxidation of Methyl Eicosapentaenoate. Agric. Biol. Chem. 1983, 47,
2897-2902.
Yamauchi, R.; T. Yamada; K. Kato; and Y. Ueno. Autoxidation and Photosensitized Oxidation
of Methyl Eicosapentaenoate: Secondary Oxidation Products. Agric. Biol. Chem. 1985,
49, 2077-2082.
Yurawecz, M.P.; A.A. Molina; M. Mossoba; and Y. Ku. Estimation of Conjugated Octadecatri-
enes in Edible Fats and Oils. J. Am. Oil Chem. Soc. 1993, 70, 1093-1099.
Yurawecz, M.P.; J.K. Hood; M.M. Mossoba; J.A.G. Roach; and Y. Ku. Furan Fatty Acids
Determined as Oxidation Products of Conjugated Octadecadienoic Acid. Lipids 1995,
30, 595-598.
4
Oxidation of Conjugated Linoleic Acid
Taina I. Pajunen (née Hämäläinen) and Afaf Kamal-Eldin
Department of Chemistry, University of Helsinki, P.O. Box 55, FIN-00014 University
of Helsinki, Finland, and Department of Food Science, Swedish University of
Agricultural Sciences (SLU), Box 7051, 750 07 Uppsala, Sweden
Introduction
Conjugated linoleic acid (CLA) is a generic name for a group of positional and geo-
metric isomers of octadecadienoic acid in which the two double bonds are conjugated,
that is single and double bonds alternate (1,3-diene structure). The most abundant
CLA isomers in nature are 9Z,11E-octadecadienoic acid (9Z,11E-CLA) and 10E,12Z-
octadecadienoic acid (10E,12Z-CLA).
Interest in CLA rose exponentially when an isomeric mixture of CLA was dis-
covered to have anticancer activities (Ha et al., 1987). Today, numerous physiological
properties are attributed to CLA, including effects on cancer, atherosclerosis, adiposity,
and insulin resistance (Gnädig et al., 2001; Belury et al., 2003; Terpstra, 2004; Wahle
et al., 2004). The physiological properties of CLA seem to be structure-specific. So far,
only two CLA isomers, namely 9Z,11E-CLA and 10E,12Z-CLA, are known to possess
biological activities, with 10E,12Z-CLA being more active. Both isomers were claimed
to inhibit carcinogenesis in animal models (Scimeca, 1999; Pariza et al., 2001) and
10E,12Z-CLA was shown to affect lipid metabolism. The biochemical mechanisms
of CLA action remain unclear. Potential carcinogenic reduction mechanisms (Belury,
2002) as well as obesity reduction (Evans et al., 2002) seem to include induction of
fatty acid oxidation. In humans, CLA was found to increase free-radical-induced lipid
peroxidation leading to increased levels of 8-iso-PGF2α as well as to increase cyclooxy-
genase activity leading to increased levels of urinary 15-keto-dihydro-PGF2α and C-
reactive proteins (Basu et al., 2000; Risérus et al., 2002). These effects might explain
the decrease in insulin sensitivity and glucose tolerance observed for CLA isomers,
particularly for 10E,12Z-CLA (Moloney et al., 2004; Risérus et al., 2004).
Despite the wide interest in CLA and the large number of scientific articles pub-
lished recently (a regularly updated listing of the scientific literature on CLA is available
at http://www.wisc.edu/fri/clarefs.htm), surprisingly little is known about the oxida-
tion reactions of CLA. For example, no CLA primary autoxidation products were iden-
tified until 2001. Most studies use a complex mixture of CLA isomers (in the form of
free acids, and methyl, ethyl, or glyceryl esters) and autoxidation of CLA is performed
under widely different conditions (use of different temperatures, solvents, initiators,
77
78 T.I. Pajunen and A. Kamal-Eldin
antioxidants etcetera; see Table 4.1). Thus, it is difficult to interpret the data and it
is not possible to correlate the results with a single isomer or to compare the data
from different studies directly. A recent study provided evidence that CLA isomer-
izes through [1,5]-sigmatropic rearrangement of hydrogen in heated oils (Destaillats
and Angers, 2002) and therefore, the data produced at high temperatures should be
interpreted cautiously. In fact, the question is whether oxidations performed at high
temperatures should be considered as autoxidation reactions, because by definition
autoxidation is a reaction performed at ambient pressure and temperature.
The aim of this chapter is to review the available information relevant to CLA ox-
idation for inspiration of new studies needed to discover the mechanisms behind the
biological activity of CLA isomers. It is noted that future research needs to produce
more data on the oxidation of single isomers in order to characterize the products of
each of the isomers, and to understand and evaluate the oxidation mechanisms of
CLA.
Autoxidation of CLA
Available studies suggest that the autoxidation of conjugated fatty acids occurs through
similar and different mechanisms and produces unique oxidation products compared
to autoxidation of monounsaturated and nonconjugated polyunsaturated fatty acids
(Table 4.1). Thus, interpretation of the data of CLA autoxidation and comparison of
it with that of nonconjugated fatty acids should be done with sufficient care. Misin-
terpretations are easily made because the knowledge of the structures of CLA autoxi-
dation products and of mechanisms is limited.
The identification of adequate method(s) to monitor CLA autoxidation is a
challenging task. Banni et al. (1998) have pointed out that the controversial results
concerning the involvement of CLA in oxidative stress are due mostly to the lack
of suitable methodology. In addition, Van den Berg et al. (1995) and Suzuki et al.
(2004) have shown that peroxide value (PV) measurements produce incorrect results
in kinetic studies pertinent to CLA oxidation. PV measurement is an indirect method
and cannot be used to distinguish between hydroperoxides (ROOH) and dialkyl per-
oxides (ROOR, for example, cyclic or oligomeric peroxides). The relative sensitivity
of PV measurement to the peroxide structure and the access of its reagents to peroxide
bonds that might be hidden inside global oligomers, however, are not known. Thus,
using PV measurements to estimate the extent of autoxidation may lead to incorrect
results if autoxidation primarily yields oligomeric products. The contribution of the
dialkyl peroxide cross-links have been determined by reacting oxidized oil samples
with SF4 prior to PV measurement (Mallégol et al., 2000). Moreover, the hydroperox-
ides can be detected and distinguished from cyclic peroxides and from oligomeric per-
oxides using NMR spectroscopy. The hydroperoxide protons resonate clearly at dif-
ferent spectral regions from the characteristic signals of cyclic peroxides (Hämäläinen
and Kamal-Eldin, 2005). Furthermore, integration of the proton signals can be used
to differentiate between monohydroperoxides and oligomeric products with hydro-
peroxide groups.
Table 4.1. Examples of Oxidation Studies on Conjugated Linoleic Acid (CLA) as a Free Acid or as an Ester.
Oxidizing Conditions (other FA, Analysis Main results/Comments Reference
Substrate antioxidant(s), initiator,
solvent, T, etc.)
10,12-CLA Oxidized in air at 37°C. oxygen con- • Autoxidation of 10,12-CLA proceeds approximately at the same rate Holman and
sumption as LA at 37°C. Elmer, 1947
10,12-CLA, Oxidized in open test tubes PV, hy- • The rate and the amount of peroxides formed were dependent on Allen et al.,
Me 10,12- while oxygen was bubbled drogen the temperature; maximum was reached faster at elevated tempera- 1949
CLA, and through at 30, 65, and 90°C. number, tures, but larger amount of peroxides was present at 30 than at 65
mixture CD or 90°C.
of CLA
isomers
Me 10,12- Oxidized in a closed system PV, hy- • No peroxide oxygen was formed until Me 10,12-CLA had been Allen et al.,
CLA after flushing with oxygen drogen oxidized more than 100 hours. The reaction proceeded three times 1949
at 30°C. number, more slowly and in a different manner (less peroxides were formed)
CD than the ML.
Me 10,12- Oxidized in presence and oxygen con- • Autoxidation is autocatalytical in character and the energy of activa- Kern et al.,
CLA absence of AIBN, di-t-butyl- sumption tion is 17.5 kcal/mol. 1955
peroxide or copper salt of • Use of an initiator or a catalyst results in the formation of polymeric
α-ethyl-caproic acid at 50°C. peroxides.
Me 9,11- Oxidized in absence of cata- oxygen con- • Me 9,11-CLA reacts in 1,2- and 1,4-positon with molecular oxygen to Kern et al.,
CLA lyst at 50 and 70°C and with sumption give relatively low molecular oxygen-co-polymers. 1956
peroxide/Cu2+ at 50°C.
Me 9,11- Oxidized in presence and oxygen • Autoxidation proceeds by an autocatalytic manner. Privett, 1959
CLA absence of +0.1% NDGA at consump- • 84% of the reaction products were polymers which were not fis-
30°C. tion, PV, sioned by SnCl2 and 16% monomeric or monomers produced by
UV, SnCl2 reaction with SnCl2 when the oxidation extent was 17%.
and KI • Virtually all the isolated E-unsaturation receded in the polymer frac-
reduction, tion.
residual FA
ester
79
Cont. on p. 82.
Table 4.1., cont. Examples of Oxidation Studies on Conjugated Linoleic Acid (CLA) as a Free Acid or as an Ester.
80
solvent, T, etc.)
Mixture Oxidized in presence and weighing, • BHA, BHT, PG, and sesamol lengthened the induction period seven to Fukuzumi
of Me absence of 0.01% BHA, BHT, UV, IR, PV, twelve times. and Ikeda,
E,Z-CLA PG, sesamol, NDGA, α-Toc, MS • The induction periods were shorter and the peroxide values lower 1970
isomers L-thyroxine sodium salt, with or without antioxidants for the conjugated dienoates than for
4.4’-dihydroxy-3,5,3’,5’-tetra- the nonconjugated dienoates.
t-butyldiphenyl methane. • After autoxidation to a weight gain of 10 mg per 1.5 g, the antioxi-
dant containing samples had higher molecular weights and lower
diene contents than the control samples.
T.I. Pajunen and A. Kamal-Eldin
9E,11E-CLA Oxidized in 90% v/v aqueous UV • The rate of autoxidation increases with the initial acid concentration. Pekkarinen,
acetic acid in presence and • The autoxidation of 9E,11E-CLA is retarded by copper (reducible), 1972
absence of copper, manga- manganese (oxidizable) salts.
nese, and cobalt acetate at • The cobalt salt retards the autoxidation at low concentrations and
80°C. accelerates it at higher concentrations.
Me 8E,10E- Photooxidized in MeOH for 5d IR, NMR, MS • The photooxidation of Me 8E,10E-CLA produces 6-heptyl-3-(6-me- Gunstone
CLA using MB sensitizer. thoxycarbonylhexyl)-3,6-dihydro-1,2-dioxine. and Wijesu-
ndera, 1979
Me 9E,11E- Photooxidized in CCl4/MeOH IR, 1H and •The photooxidation of Me 9E,11E-CLA produces 6-hexyl-3-(7-me- Bascetta et
13
CLA (95/5, v/v) for 16h using MB C NMR, thoxycarbonylheptyl)-3,6-dihydro-1,2-dioxine in over 80% yield after al., 1984
sensitizer. MS purification by column chromatography.
Mixture of Oxidized in the presence of PV • CLA is much more resistant to oxidation than LA. Ha et al.,
nine CLA LA in phosphate buffer- • CLA was more potent than α-Toc and almost as effective as BHT as an 1990
isomers water-ethanol mixture in antioxidant. (Note: This result has later been proven not to be correct.
presence and absence of See van den Berg et al. 1995, Chen et al. 1997, and Banni et al. 1998.)
ascorbic acid or α-Toc for
15d at 40°C.
solvent, T, etc.)
Mixture Oxidized with PLPC mem- GC/MS • CLA does not act as an efficient radical scavenger in any way compa- van den Berg
of CLA brane using initiator (AMVN, analysis rable to vitamin E or BHT under conditions of metal ion-dependent et al., 1995
isomers AAPH or H2O2 and Fe2+) in of FA or independent oxidative stress.
presence and absence of
vitamin E or BHT in buffer
solutions.
Mixture Oxidized as a thin film in air CD, GC/MS • The oxidative susceptibility of CLA was higher than that of LA and van den Berg
of CLA with and without LA/AA analysis comparable to AA. et al., 1995
isomers at rt. of FA • Oxidation rates for individual CLA isomers were found to be virtually
identical.
Mixture Oxidized in methanol-water CG-MS, • Furan fatty acids identified as secondary oxidation products of CLA Yurawecz et
of CLA solution while flushing GC/FID autoxidation. al., 1995
isomers air through at 45, 40, and
48-69°C.
Me 9E,11E- Oxidation as ~ 0.5 mm thick DCI-, FAB-, • Oxidative crosslinking produced oligomers that yielded signals con- Muizebelt
CLA film in the presence of FD-, ESI-, sisting of groups of peaks 16 mass units apart, pointing to a series of and Nielen,
Co/Ca/Zr drier in air for 2d. and SI-MS oxygenated homologues. 1996
• Conjugated FA crosslinks by radical addition to the double bond
whereas nonconjugated FA by recombination of radicals.
Mixture of Added to canola oil; Oxidized oxygen up- • CLA and Me CLA accelerated lipid oxidation in canola oil dose-de- Chen et al.,
CLA iso- after flushing with air at take and pendently. CLA-TAG had no influence on lipid oxidation in canola oil. 1997
mers (as 90°C. changes • CLA is not an antioxidant in fats and oils.
free acids, in LA and
Me esters α-LnA by
or TAG) GC/FID
Cont. on p. 84.
81
Table 4.1., cont. Examples of Oxidation Studies on Conjugated Linoleic Acid (CLA) as a Free Acid or as an Ester. 82
solvent, T, etc.)
Mixture Mixed with LA (1/1, w/w); oxygen up- • CLA is remarkably less stable than LA in air. Chen et al.,
of CLA Oxidized after flushing with take and 1997
isomers air at 90°C. changes
in LA by
GC/FID
Mixture of Mixed with LA, LnA, AA, DHA, FA analysis • CLA as a free acid or TAG more susceptible to oxidation than LA, LnA, Zhang and
CLA iso- and HA (equal amounts) or by GC/FID and AA. The oxidative susceptibility of CLA was similar to that of Chen, 1997
mers (as TAG containing these FA; DHA.

free acids Oxidized after flushing with
or TAG) air at 90°C.
Mixture Oxidized in presence LA as a HPLC with • No significant antioxidant effect of CLA was detected. Banni et al.,
of CLA thin film in air at 0 and 37°C. DAD 1998
isomers
Mixture of Photooxidized as pure Me PV, residual • Me CLA was lost to a lower extent and yielded less (hydro)peroxides Jiang and Ka-
Me CLA CLA or mixed with ML in FA, re- but bleached MB at a higher rate than ML. mal-Eldin,
isomers ethanol for 5d using MB sidual MB • Me CLA isomers were not equally lost and the photooxidation was 1998
sensitizer at 25°C. accompanied by interisomerisation.
• Me CLA photooxidizes through different mechanisms than ML.
Me 9Z,11E- Epoxidized with m-CPBA and TLC, IR, 1H • A mixture of positional monoepoxides was generated using m-CPBA Lie Ken Jie
CLA Oxone. and 13C and diepoxides formed by using Oxone. and Pasha,
NMR 1998
CLA iso- Oxidized in the presence of Oxygen • CLA is more stable than LA in aqueous system when AAPH was used Seo et al.,
mers (as LA in aqueous system (pH uptake as a free radical initiator. 1999
free acid 7.5) with AAPH. • The oxidizability was in the order Et CLA>LA>Me CLA>ML>EL>CLA.
or Me or
Et esters)
solvent, T, etc.)
CLA iso- Oxidized in the presence LA in Oxygen • CLA is less stable than LA in benzene when AMVN was used as a free Seo et al.,
mers (as benzene with AMVN. uptake radical initiator. 1999
free acid • The oxidizability was in the order Et CLA>EL=ML>Me CLA>CLA>LA.
or Me or
Et esters)
9Z,11E- Tested in the TOSC assay (Oxi- Oxidation of • 10E,12Z-CLA and 9Z,11E-CLA inhibited the oxidation of KMBA at Leung and
CLA and dation of KMBA by ABAP). KMBA to 2-200 micromolar concentrations. At higher concentrations, 9Z,11E- Liu, 2000
10E,12Z- ethylene CLA but not 10E,12Z-CLA enhanced the oxidation.
CLA
Me 9Z,11E- Oxidative crosslinking in the SEC, SI- and • Nonconjugated FA reacts by recombination of radicals forming Muizebelt et
CLA presence of Co or Co/Ca/Zr ESI-MS, mainly peroxy and ether coss-links and conjugated FA by radical ad- al., 2000
and Me drier and reactive diluents NMR dition to the double bond.
10E,12Z- up to 4d. • Despite the differences in the mechanism the rates of both reactions
CLA were found to be similar.
• Reactive diluents are incorporated into the actual paint film.
Mixture of Oxidized in the presence of GLC with • CLA oxidized considerably faster than LA. The oxidation extent of CLA Yang et al.,
12 CLA LA (1.5g/0.3g); Oxidized FID and was > 80% after 110 h. 2000
isomers after flushing with air at Ag-HPLC • The four Z,Z-CLA isomers were most unstable followed by four E,Z-
50°C. with UV CLA isomers. The four E,E-CLA isomers were relatively stable under
the same experimental conditions. The different positional isomers
of the same geometry oxidized at a similar rate.
Mixture of CLA isomers added to canola oxygen con- • 200 ppm of GTC were more effective (and dose dependent) than 200 Yang et al.,
12 CLA oil at 10% level; Oxidized in sumption ppm BHT in protecting CLA from oxidation. 2000
isomers air in presence and absence and FA • CLA oxidized without the antioxidants considerably faster than LA:
of BHT or GTC at 90°C. analysis The oxidation extent of CLA after 35 h was 86%.
Mixture of Oxidized by heating at 100, GC of • CLA oxidizes at a higher rate compared to LA. Chen et al.,
Me CLA 150, or 200oC for 3 h or residual 2001

isomers illumination at 4000 lux for FA, PV
83
14 d.
Cont. on p. 86.
solvent, T, etc.)
Mixture of Oxidized as neat oil in pres- PV, GC-MS, • Hydroperoxides discovered as primary products in the CLA autoxida- Hämäläinen
Me CLA ence and absence of α-Toc HPLC, tion. The NMR results revealed that CLA autoxidizes at least partly et al., 2001
isomers in the dark at 40°C. 1D and according to the Farmer’s hydroperoxide theory.
2D NMR
, residual
CLA
CLA, Et CLA Oxidized in the bulk phase oxygen • The oxidative stability of CLA was almost the same as that of LA, but Suzuki et al.,
at 50°C. consump- ethyl CLA was oxidatively more stable than ethyl LA. 2001
tion, PV,
dimers/
polymers
by HPSEC
Mixture Reacted with DPPH radical. DPPH • CLA slightly diminished DPPH compared to BHT, α-tocopherol, and Yu, 2001
of CLA radical vitamin C.
isomers assayed by
ESR
Me 9Z,11E- Oxidized as neat oil in pres- PV, GC-MS, • Structural characterization of seven hydroperoxides from autoxida- Hämäläinen
CLA ence of α-Toc in the dark at HPLC, 1D tion of Me 9Z,11E-CLA. et al., 2002
40°C. and 2D • The reaction was in favour of one geometric isomer and a new type
NMR spec- of E,Z-conjugated diene hydroperoxide, where the Z-double bond
troscopy was adjacent hydroperoxyl-bearing methine carbon was discovered.
• A mechanism for the autoxidation of Me CLA was proposed based on
well-characterized primary oxidation products.
solvent, T, etc.)
CLA con- Oxidized as oil in presence PV • The protective effect of 200 ppm antioxidant was in the order of Lee et al.,
centrate of antioxidant (α-Toc, BHA, TBHQ> BHA> PG> BHT> RE> α-Toc> GTE. 2003
(=alkali BHT, GTE, PG, RE, and TBHQ)
isomer- in the dark up to 44 d at
ized 45°C.
safflower
oil)
Me 9Z,11E- Epoxidized with m-CPBA, IR, 1H and • The reactions furnished mono-epoxides and a mixture of diastereo- Lie Ken Jie et
13
CLA DMDO, MTO/H2O2, Oxone C NMR, mers of syn- and anti-diepoxy-stearate. al., 2003
/tetrahydrothiopyran-4-one, EI-MS
and Novoenzyme 435/ H2O2.
9Z,11E- Oxidized separately as thin Unoxidized • 10E,12Z-CLA oxidized faster than 9Z,11E-CLA at all tested tempera- Minemoto et
CLA and films in the dark at 40 to CLA tures. Apparent activation energies determined as 65.4 and 72.1 al., 2003
10E,12Z- 80°C. analyzed kJ/mol, respectively.
CLA by GC
9Z,11E-CLA Oxidized with peroxygenase HPLC with • Main product (approx 90%) was up to 6 h 9,10E-epoxy-11E-octadec- Piazza et al.,
in an aqueous medium us- APCI-MS/ enoic acid. 2003
ing t-butyl hydroperoxide as EI-MS and • Acidic work up resulted in the formation of 1,2- and 1,4-diols.
1
an oxidant. H and 13C
NMR
Cont. on p. 88.
85
solvent, T, etc.)
CLA-TAG Oxidized in the bulk phase oxygen con- • Rates of oxygen consumption and polymer formation of CLA-TAG Suzuki et al.,
(69.5% in a sealed vial with and sumption, and bitter gourd-TAG were faster than those of soybean-TAG and 2004
CLA) and without 0.5% α-Toc or trolox PV, size- perilla-TAG, respectively. (The same results were obtained in the
Me esters in the dark at 50°C. exclusion oxidation of the Me esters.)
prepared Note: soybean-TAG (56.1% HPLC • The main oxidation products of CLA-TAG and bitter gourd-TAG were
from CLA- LA), perilla-TAG (54.5% α- dimers and polymers whereas hydroperoxides were the main prod-
TAG LnA), and bitter gourd-TAG ucts in the oxidation of soybean-TAG and perilla-TAG.
(61.6% conjugated LnA) and • α-Toc and trolox inhibited the oxidation of the TAG. The inhibitory
Me esters prepared from effect of these antioxidants was more effective against the oxida-
these TAG was oxidized in tion of CLA-TAG and bitter gourd-TAG than that of soybean-TAG and
the same conditions. perilla-TAG, respectively.
9Z,11E- Oxidized separately as thin Unoxidized • CLA oxidizes at a higher rate than LA. 9Z,11E-CLA was consumed at Tsuzuki et al.,
CLA and films at 37°C. substrates the same rate as LnA and was more stable than the 10E,12Z-CLA. 2004
10E,12Z- by GC, PV, • The 9Z,11E-CLA and 10E,12Z-CLA gave low values of PV and TBARS
CLA PV, TBARS compared to LA and LnA.
CLA Encapsulated in WPC, GA, and Static head- • The highest values of CLA degradation and lipid oxidation were ob- Jimenez et
[isomer(s) a blend of WPC and malto- space GC served in the range of water activities 0.103–0.429 for all matrices at al., 2006
not re- dextrin 10 DE (1:1, w/w) 45°C, whereas the lowest CLA degradation and lipid oxidation were
ported] matrixes at water activities observed for WPC at a water activity of 0.743 and 35°C.
from 0.108 to 0.892 at 35
and 45°C.
solvent, T, etc.)
Abbreviations: AA = arachidonic acid, AAPH = 2,2’-azobis(2-amidinopropane) dihydrochloride, AIBN = 2,2’-azobis(buturonitrile), AMVN = 2,2’-azobis(2,4-
dimethylvaleronitrile), BHA = butylated hydroxyanisole, BHT = butylated hydroxytoluene, CD = conjugated dienes, CLA = conjugated linoleic acid,
m-CPBA = m-chloroperoxybenzoic acid, DAD = diode array detector, DCI = direct chemical ionization, DHA = docosahexaenoic acid, DMDO = dimethyl
dioxirane, DPPH• = 2,2-diphenyl-1-picrylhydrazyl radical, ESI = electrospray ionization, ESR = electron spin resonance, Et = ethyl, EL = ethyl linoleate,
FA = fatty acid, FAB = fast atom bombardment, FD = field desorption, GA = gum arabic, GLC = gas-liquid chromatography, GTC = jasmine green tea
catechins, GTE = green tea extract, HA = heptadecanoic acid, KMBA = α-keto-α-methiolbutyric acid, LA = linoleic acid = 9Z,12Z-octadecadienoic acid,
LnA = linolenic acid = 9Z,12Z,15Z-octadecatrienoic acid, MB = methylene blue, Me = methyl, MeOH = methanol, ML = methyl linoleate, MS = mass
spectrometry, MTO = methyltrioxorhenium, NDGA = nordihydroguaiaretic acid, Oxone = potassium peroxymonosulfate, PG = propyl gallate, PLPC =
1-palmitoyl-2-linoleyl phosphatidylcholine, PV = peroxide value, RE = rosemary extract, rt = room temperature, SEC = size exclusion chromatography,
SI = secondary ion, TAG = triacylglyceride, TBHQ = tert-butylhydroquinone, TLC = thin-layer chromatography, α-Toc = α-tocopherol, TOSC = total anti-
oxidant scavenging capacity, WPC = whey protein concentrate.
87
Rate and Routes of Autoxidation of CLA

A number of studies have shown that a mixture of CLA isomers oxidizes at a higher
rate compared to linoleic acid (LA; 18:2, 9Z,12Z) by measuring residual unoxidized
substrate or consumed oxygen (van den Berg et al., 1995; Zhang and Chen, 1997;
Chen et al., 1997,2001; Yang et al., 2000). As a thin film 10E,12Z-CLA oxidizes
faster than 9Z,11E-CLA at 40–80°C when the residual unoxidized substrate is moni-
tored (Minemoto et al., 2003). This was confirmed in a study by Tsuzuki et al. (2004),
where similar thin films of CLA were oxidized at 37°C and substrate consumption
was followed. Moreover, 9Z,11E-CLA was consumed at the same rate as linolenic
acid (LnA; 18:3, 9Z,12Z,15Z). However, when oxygen consumption was used to
monitor oxidation, 9Z,11E-CLA was more similar to LA, the difference between the
two CLA isomers was not that large, and LnA consumed oxygen at a comparable
rate to α-eleostearic acid (ESA; 18:3, 9Z,11E,13E). Suzuki et al. (2004) showed that
CLA-containing oil (69.5% CLA) oxidized faster than purified soybean oil (56.1%
LA) and that bitter gourd oil (61.6% ESA) oxidized faster than perilla seed oil (54.5%
LnA) when oxygen consumption was used to monitor oxidation at 50°C. In addition,
they demonstrated the previous claim that CLA is oxidatively more stable than LA at
room temperature came as a result of the use of PV measurement to monitor oxida-
tion (Ha et al., 1990; Ip et al., 1991). Suzuki et al. (2004), Allen et al. (1949), and
Holman (1954) showed that only small amount of peroxides accumulate during the
oxidation of CLA, unlike the case of LA. The results of Tsuzuki et al. (2004) suggest
that CLA and ESA absorb less oxygen per mole of oxidized substrate compared to
their unsaturated counterparts, LA and LnA. It has already been established that the
stability of CLA isomers decreases in the order E,E>E,Z>Z,Z which shows a stabiliz-
ing effect of E-double bonds (Yang et al., 2000) as is seen for other fatty acids, for
example elaidic and oleic acids (Lanser et al.,1986).
Allen and Kummerow (1951) and Kern et al. (1955) showed that the role of
conjugated fatty acids in their own autocatalytic oxidation differ from that of their
nonconjugated counterparts. For example the rate of the reaction is one-half order
with respect to the catalytic product in the case of CLA while first order in the case of
LA, that is
Rate of CLA autoxidation = k [CLA] √[Oxidation product(s)]

Rate of LA autoxidation = k [LA] [LOOH]
Further analysis of the Kern et al. (1955) data using empirical kinetics suggested that
the oxidation of 9Z,11E-CLA is dominated by trimerization (up to 1% oxidation at
40°C, 2% oxidation at 50°C, and 15% oxidation at 70°C) followed by formation of
monomers (Brimberg and Kamal-Eldin, 2003). These results suggest that the extent
of oligomerization increases with increased temperature. Since the activation energies
for both reactions leading to oligomers and monomers were almost identical, it was
suggested that the monomers split from the oligomers. Comparing the oxidation
kinetics of thin films of 9Z,11E-CLA and 10E,12Z-CLA at different temperatures,
Oxidation of Conjugated Linoleic Acid 89
Minemoto et al. (2003) found that the oxidation of 9Z,11E-CLA followed an auto-
catalytic rate expression through the entire oxidation process while 10E,12Z-CLA
followed the autocatalytic rate expression only during the first half of the oxidation
course, which was followed by first order kinetics. The apparent activation energies for
9Z,11E-CLA and 10E,12Z-CLA were approximately 15.5 and 17 kcal/mol, which
were greater than those of LA (12 kcal/mol) and LnA (14.5 kcal/mol). The activa-
tion energy was the same during the overall oxidation regimen of the 10E,12Z-CLA
isomer. These authors found that the enthalpy-entropy compensation holds during
autoxidation of 9Z,11E-CLA, 10E,12Z-CLA, and LA and suggested a “common”
oxidation mechanism.
Seo et al. (1999) compared the oxidation of CLA and LA as equal mixtures of
9,11- and 10,12- isomers in the form of free acids and methyl and ethyl esters in the
presence of methyl linoleate (ML). Both were examined in an aqueous system, where
oxidation was induced by 2,2’-azobis(2-amidinopropane)dihydrochloride (AAPH)
and in benzene, where oxidation was induced by 2,2’-azobis(2,4-dimethyl-valeroni-
trile) (AMVN) at 37°C. The oxidizability in aqueous system was much lower for CLA
than for LA while it was much higher for CLA ethyl and methyl esters than for LA
methyl and ethyl esters (Table 4.1). In benzene, CLA had a higher oxidizability than
LA while the oxidizabilities of their esters were comparable. Indeed, the mixture of
CLA in this study might have affected the results, especially in those where only small
differences were found.
Important differences between CLA and LA oxidations relate to the effect of pro-
and antioxidants on their oxidation rates. Jackson and Kummerow (1949) demon-
strated that metallic naphthenates have less effect on the oxidation of CLA than LA.
Moreover, Kern et al. (1955) found that copper(II)octanoate (3×10-4 mol Cu/mol
ester) delayed the oxidation of methyl 10E,12Z-CLA at 40°C without inhibiting the
rate of oxidation afterwards. This result, which contradicts the enhancement of LA
oxidation by metal ions, may be explained by inhibitory interactions of π-complexes
with transition metal ions (Park et al., 2007). α-Tocopherol was more potent in pro-
tecting CLA than purified soybean triacylglycerols (Suzuki et al., 2004). Moreover,
while α-tocopherol remains resistant to oxidation during the induction period of
fatty acids with methylene-interupted double bonds and then declines sharply, its
degradation during the oxidation of CLA seems to follow a sigmoid curve (Suzuki
et al., 2004). It was proposed that antioxidants retard the oxidation of CLA by two
mechanisms, first by forming H-bonds with the π-system of the conjugated double
bond and secondly by interfering with the propagation reactions (Fukuzumi and Ike-
da 1970).

Formation of Hydroperoxides During Autoxidation of CLA Methyl Ester
The identification of the primary autoxidation products is of crucial importance for
understanding the subsequent secondary oxidation steps. The formation mechanism
of the oligomeric products, which seem to dominate in CLA, remains more or less
speculative without the knowledge of the primary species involved in their initiation.
It has been assumed that the primary autoxidation products of conjugated dienes
are not similar to those of methylene-interrupted systems (Yurawecz et al., 1997).
Contrary to this assumption, conjugated diene allylic monohydroperoxides were
discovered as primary autoxidation products of CLA methyl esters (Hämäläinen et
al., 2001). The conclusive evidence for hydroperoxide formation during autoxida-
tion of CLA methyl esters with and without α-tocopherol was provided by NMR
spectroscopy. The hydroperoxide protons of CLA hydroperoxides appeared as partly
overlapping singlets at δH 7.98–7.81 in deuterochloroform and at δH 10.48–10.40
in deuteroacetone. These assignments were further confirmed by D2O-test, and by
comparison of the 1H NMR spectrum of CLA hydroperoxides with that of ML hy-
droperoxides. In addition, based on proton signal integration and 1H-1H correlation
experiment (COSY), it was evident that the main CLA hydroperoxides have a conju-
gated diene allylic monohydroperoxy structure.
Determination of an oxidation mechanism requires full characterization of the
oxidation products. A good mechanism explains not only the formation of all the
products but accounts for their relative proportions. Hämäläinen et al. (2002) pro-
duced hydroperoxides from 9Z,11E-CLA methyl ester oxidizing the fatty acid ester
with 20% α-tocopherol under atmospheric oxygen at 40°C in the dark. In these con-
ditions, as well as in the autoxidation of ML (Peers and Coxon, 1983), α-tocopherol
does not exert its recognized antioxidant effect; it allows the autoxidation to proceed
smoothly and quite rapidly to produce good yield of hydroperoxides. The CLA methyl
ester hydroperoxides were isolated by flash column chromatography and subsequently
reduced to corresponding hydroxy derivatives. By combining HPLC separation, UV,
1D and 2D NMR, and GC-MS techniques the structures of the individual hydroxy
isomers were characterized. Detailed example of the structural characterization can be
found elsewhere (Hämäläinen and Kamal-Eldin, 2005). The characterization of the
primary products enabled the authors to propose a mechanism for the hydroperoxide
formation during autoxidation of CLA methyl esters.
The autoxidation of 9Z,11E-CLA methyl ester through the hydroperoxide path-
way is depicted in Scheme 4.1. The autoxidation yielded seven conjugated diene
allylic monohydroperoxides as a pair of enantiomers: four with E,E-geometry and
three with E,Z-geometry. Note that one of the E,Z-isomers was a new type of lipid
hydroperoxide (11) where the Z-double bond is adjacent to the hydroperoxyl-bearing
methine carbon atom. Furthermore, the autoxidation was diastereoselective in favor
of one geometric isomer (9); namely methyl 13-(R,S)-hydroperoxy-9Z,11E-octadeca-
dienoate (Me 13-OOH-9Z,11E).
The mechanism for the hydroperoxide pathway of CLA autoxidation shows simi-
larities to both methyl oleate and ML autoxidation. The first step is an abstraction
of one of the allylic hydrogen atoms from the starting material. Similarly to methyl
oleate, there are four allylic hydrogen atoms available for abstraction. However, dif-
fering from methyl oleate the two allylic positions are not equal since one is allylic to
an E -double bond and the other is allylic to a Z-double bond. Moreover, the H-atom
abstraction leads to the formation of two pentadienyl radicals 2 and 3, where the lone
R2 -3 O 2 R2 3
O2 R 2 T o cO H T o cO R2
OO <1 %
R1 R1 R1 R1
(4 ') (1 2) O O (1 3 ) H O O (1 4 )
T o cO H T o cO
OO OOH
3
R2 R2 1 5%
O2 R1 (4) R1 (8 )
R1
R2 +
T o cO H T ocO
(2 ) R2 R2 50 %
R1 R1
O O (5) H O O (9 )
H
OO OOH
R2
-3 O 2 R2
3
O2
R2
T o cO H T o cO
R2 1 3%
OO
R 1 (5 ') R 1 (1 5) R 1 (1 6) R 1 (1 7 )
R2
R1
(1 )
R 1 = (C H 2 ) 4 C H 3 ,
R 2 = (C H 2 ) 6 C O 2 M e
H
T o cO H T o cO HOO
OO
R2 R2
R1 (6) R1 (1 0) 1 2%
3
O2 +
R2
R
(3 )
T o cO H T ocO 2%
R2 R2
R1 OO R1 OOH
(7) (1 1 )
3
O2
OO OOH
R1
T o cO H T o cO
R1
8%
R2 R2
OO (19 ) (20 )
R1 R2 +
(18 ) T o cO H T ocO
OO
R1 (1 0)
R2
(6 ')
Scheme 4.1. Proposed Mechanism for the Hydroperoxide Pathway of CLA Methyl Ester Autoxidation (Source:
Hämäläinen et al., 2002).
electron is delocalized not over three but five carbon atoms as in the autoxidation
of ML. Subsequent peroxidation of the pentadienyl radicals, and H-atom abstrac-
tion yields four ‘kinetic’ hydroperoxides: Me 9-OOH-10E,12E (8); Me 13-OOH-
9Z,11E (9); Me 12-OOH-8E,10E (10); Me 8-OOH-9Z,11E (11). The formation
of Me 13-OOH-9E,11E (14) and Me 9-OOH-10E,12Z (17) may be explained by
the β-fragmentation pathway in the same manner as the isomerization of the kinetic
E,Z-hydroperoxides of ML autoxidation. The direct β-fragmentation pathway does
not explain the formation of 8% of Me 8-OOH-9E,11E (20). Because both termini
of pentadienyl radical 3 have a partial E-double bond character, roughly the same
amount of peroxyl radicals 6 and 7 is expected to be formed. Formation of signifi-
cantly less Me 8-OOH-9Z,11E (11) than Me 12-OOH-8E,10E (10) suggested that
hydroperoxide 20 might be formed trough through two successive [2,3]-allyl rear-
rangements of peroxyl radical 7 since the direct isomerization of pentadienyl radical
3 seems unlikely under normal oxygen pressure. Recently, Tallman et al. (2001,2004)
demonstrated that the nonconjugated diene bisallylic hydroperoxide is the main
product in the autoxidation of ML in the presence of high amounts of α-tocopherol

at oxidation extent of 2%. In light of this study, it seems likely that the nonconjugated
peroxyl radical 18 is formed directly from pentadienyl radical 3 rather than from radi-
cal 7. This is supported by theoretical calculations (unpublished results by Pajunen
and co-workers).
The origin of the diastereoselectivity in the autoxidation of 9Z,11E-CLA methyl
ester lies in the selectivity of the initial H-atom abstraction and peroxidation steps;
preferential formation of the more stable pentadienyl radical 2 over radical 3, and
regioselectivity in the oxygen addition to the pentadienyl radical 2 in favor of peroxyl
radical 5, and thereafter hydroperoxide 9. The selectivity in the H-atom abstraction
can be explained by thermodynamic terms; invoking the Hammond postulate we can
make an assumption that the geometry of the transition state resembles that of the
pentadienyl radical. Since radical 2 is more stable than radical 3, the transition state
leading to radical 2 should be more stable than that leading to radical 3. The ‘energy
barrier’ ∆G‡ is therefore smaller for the formation of radical 2 than it is for the for-
mation of radical 3, and thus radical 2 is formed faster. The kinetic and ESR studies
performed by Chan et al. (1978,1979) and Porter and Wujek (1984) using ML show
that the peroxidation step is not only reversible but also regioselective in that the pen-
tadienyl radical terminus having a partial E-double bond character reacts faster with
triplet oxygen than that having a partial Z-double bond character.
The autoxidation of 9Z,11E-CLA methyl ester produces a complex mixture of
hydroperoxides including the 9- and 13-positional isomers identical to those formed
in the autoxidation of ML. This complexity results from the formation of two distinct
pentadienyl radicals. Moreover, while ML autoxidation in the presence of α-tocopherol
yields only two positional isomers both with E,Z-geometry, CLA autoxidation yields
four positional isomers with three different geometries including a new type of E,Z-
hydroperoxide with the Z-double bond adjacent to the hydroperoxyl-bearing carbon
atom.
The proposed mechanism agrees with the kinetic evidence that the autoxidation
is an autocatalytic free radical chain reaction (Kern et al., 1955; Minemoto et al.,
2003). It is evident that CLA autoxidation at least partly follows Farmer’s hydroper-
oxide theory, which was developed for nonconjugated olefinic compounds. Further
confirmation for the mechanism has been provided by studying the autoxidation of
the other biologically active CLA isomer; 10E,12Z-CLA methyl ester (Hämäläinen
and Hopia, 2006). This mechanism correctly predicted the hydroperoxides formed
and their isomeric distribution. The details of the mechanism are topics of future
research.
It is necessary to explain why these primary hydroperoxides were not recognized
by early investigators. Conjugated fatty acids are generally less studied that their non-
conjugated counterparts because they are rare and have not been readily available un-
til recently. The complexity of product mixtures might have been one of the obstacles
against the discovery of hydroperoxides. Not only is there less hydroperoxide forma-
tion compared to ML, but the number of hydroperoxide isomers is greater. Since even
today, autoxidation studies are often done with mixtures of CLA isomers, the result-
ing primary product mixture contains even greater numbers of different positional
and geometric isomers of CLA hydroperoxides and is more difficult to analyze. In the
absence of a good H-atom donor, a mechanism other than hydroperoxide formation
prevails in the autoxidation of CLA in which radical addition to the double bonds
occurs and leads to oligomeric products. However, knowledge of these hydroperoxide
formation steps is crucial to understand the subsequent steps in the CLA autoxida-
tion. Moreover, knowledge of the CLA hydroperoxide structures allows us to predict
possible structures for the formed oligomers.
Formation of Oligomers
The early studies report that CLA autoxidation (Table 4.1) yields, in an autocata-
lytic manner, relatively low molecular weight polymeric peroxides (Allen et al., 1949;
Kern et al., 1955, 1956; Privett 1959). The data supports oxygen-carbon rather than
carbon-carbon polymerization and that appreciable amounts of isolated E-unsatura-
tion reside in the polymer fraction. Empirical kinetic reassessment of the early data
by Brimberg and Kamal-Eldin (2003) suggests that oligomers having an average of
three monomers would be kinetically favored at the beginning of the oxidation. This
autoxidative polymerization can be envisioned to involve several basic chemical reac-
tions in propagation and termination phases of the autoxidation.
Polymerization Resulting from Propagation Reactions.

Mayo (1968) suggested that in many alkene oxidations, the abstraction mechanism
(the abstraction of an H-atom from the oxidizing substrate by the peroxyl radical)
and addition mechanism (the addition of the peroxyl radical to the double bond
at the oxidizing substrate) occurs simultaneously and that alkenes having conju-
gated unsaturation are among those that preferably undergo oxidation by the latter
mechanism. This reactivity in favor of addition is easy to rationalize since the ad-
dition of the peroxyl radical to the conjugated diene system leads to formation of
resonance-stabilized allylic radical intermediates (Fig. 4.1). Differing from LA, CLA
has been shown to interact with the radicals produced by decomposition of 2,2’-
azobis(amidinopropane) (ABAP) (Leung and Liu, 2000) as well as 1,1-dipehnyl-2-
picrylhydrazyl radical (DPPH•) (Yu, 2001). The propagative addition is apparently
less favored in the oxidation of nonconjugated alkenes, such as polyunsaturated fatty
acids, in which the resulting radical intermediates will not be resonance stabilized.
Using the knowledge of the peroxyl radicals (CLA-OO•) preceding the CLA hydro-
peroxides, we can predict the structures for the dimeric and higher oligomers formed
through addition mechanism as depicted in Fig. 4.1. It must be emphasized that the
formation of these products remains to be confirmed by structural elucidation. How-
ever, in agreement with the early literature the addition mechanism would produce
oligomers with oxygen-carbon rather than carbon-carbon links, and it would result
in the disappearance of conjugated double bonds and in the appearance of isolated
unsaturation. Importantly, these dimers would have 1,2- or 1,4-addition of oxygen in
R2 -H 3
O2
R1 C LA -O O
R2
R1
O O -C LA
R2 + R2
R1 R1
C LA -O O
3 3
O2 O2 R2
R2
R1 R1
h igher hig h e r
o ligo m e rs oligo m ers
H H
O O -C LA O O -C LA OOH OOH
R2 + R2 R2 + R2
R1 R R1 R
OOH OOH C LA -O O C LA -O O
A B C D
Fig. 4.1. Postulated autoxidative polymerization of CLA resulting from addition of peroxyl radicals to the double
bonds of unoxidized CLA (Note: CLA-OO• refers to peroxyl radicals 4-7, 13, 16, and 19 in Scheme 4.1).
respect to one of the monomers agreeing with the data of Kern et al. (1955).
Evidence for the formation of dimers similar to those depicted in Fig. 4.1 with
a hydroperoxyl or a hydroxyl group has been provided for ML autoxidation (Mi-
yashita et al., 1982). The dimer fraction was complex because it was a mixture of
positional and geometric isomers. In CLA autoxidation, the product mixture can be
expected to be even more complex. Autoxidative polymerization of one pure CLA
isomer would involve (at least) seven enantiomeric pairs, that is 14 isomers of peroxyl
radicals (CLA-OO•), if the conditions from Hämäläinen et al. (2002) are used. Ad-
dition reaction of these peroxyl radicals with the starting material alone would lead
to 56 dimeric allylic-radicals; subsequent peroxidation and H-atom abstraction yields
224 hydroperoxide dimers with a peroxide cross-link (represented by structures A
to D) when the possibility of geometric isomerization in the last peroxidation step
is excluded. Moreover, we can expect further complications. Whenever competition
between addition and abstraction favors addition, the competition between peroxida-
tion, which leads to dimeric or higher peroxides, and “unzipping’’ of the polyperox-
ides, which leads to epoxides and alkoxy radicals, becomes important (Mayo 1968).
Thus formed alkoxy radicals would be highly reactive and could participate in various
propagation and termination reactions. For example, the addition of the alkoxy radi-
cals to the double bond system of CLA may lead to hydroperoxy dimers with ether
cross-links, β-scission to aldehydes, H-atom abstraction to alcohols, and intramo-
lecular cyclization to hydroperoxy epoxides (Fig. 4.2). We expect that the importance
of these reactions grows in metal-catalyzed autoxidations because alkoxy radicals are
also formed through homolytic cleavage of peroxides. Importantly, the unzipping
ultimately increases the heterogeneity of the oligomeric fraction. It is also noted that
unzipping of the trimeric peroxyl radical followed by subsequent H-atom abstraction

by the dimeric alkoxy radical would lead to formation of hydroxy derivatives of the
dimeric hydroperoxides A to D. In conclusion, since trimers (Fig. 4.3) seem to be
kinetically favored, the complexity of the product mixture would be enormous as a
result of the propagation reactions alone.
Polymerization Resulting from Termination Reactions

CLA is autoxidized partly according to Farmer’s hydroperoxide theory producing hy-
droperoxides. Thus, we know that our autoxidation mixture contains pentadienyl and
peroxyl radicals. It is easy to envision dimer formation through termination reactions
of these radicals (radical-radical addition reactions), which by definition lead to non-
radical products as depicted in equations 1 to 3 in Fig. 4.4. Since the peroxidation
C LA -O O R3 R4
R2 O
R1
O
R2
R1
R3 R4 R3 R4
R3 + H
O
R4
b) O O
R2 R2
R1 R1
OOH OOH
R3 R4 C LA -H C LA 3
O2 H
C LA
+ OH c)
R3 R4
+
O a)
3
O2
R3
R4
H + R3
O
R4 R3
O
R4
O d) R1 R1
R3 + OOH
R1
O R2 R2
OOH
R2
OOH
R4 higher
O oligo m e rs
OOH
Fig. 4.2. Unzipping of the postulated dimeric radical leading to an epoxide and alkoxy radical, and subsequent
propagation reactions of the alkoxy radical a) addition reaction, b) β-scission c) H-atom abstraction, and d) in-
tramolecular cyclization.
R3 R4 R3 R4 R3 R4 R3 R4
O O O O
O O O O
R2 R2 R2 R2
R1 R1 R1 R1
O O O O
O O O O
R2 R2 R1 R1
R1 R1 R2 R2
OOH OOH OOH OOH
Fig. 4.3. Expected structures of trimers formed through addition mechanism during autoxidation of CLA (Note:
These structures represent the trimers formed from dimer A in Fig. 4.1).
reaction of the pentadienyl radicals is extremely fast (occurring at diffusion-controlled

rate) and the subsequent hydrogen atom abstraction step leading to the hydroper-
oxides is slow, it is safe to assume that most of the chain reaction-carrying radical
present in the autoxidation mixture is CLA-OO• provided that oxygen is present at
sufficient concentration. Hence, termination reactions involving pentadienyl radicals
are unimportant and the only significant termination reaction under normal condi-
tions is reaction 1. The self-reaction of CLA-OO• radicals yields a tetroxide, which
decomposes by the Russell mechanism to stable end-products: secondary alcohol,
ketone, and oxygen. The early studies do not support the dehydropolymerization re-
actions between two pentadienyl radicals (Eqn. 3) because they lead to carbon-carbon
bonded dimers with the original amount of unsaturation. As discussed previously,
unzipping of oligomeric peroxides and homolytic cleavage of peroxides in metal-
catalyzed autoxidations produce alkoxy radicals. We expect that the concentration of
alkoxy radicals is low due to high reactivity of the alkoxy radical. Hence, the radical-
radical addition reactions 4 and 5, and the disproportionation reaction 6 would not
be dominant termination reactions.
2 C L A -O O C L A -O 2 -O 2 -C L A no t s ta b le (1 )
C L A -O O + R1 R2 R1 R2 + R1 R2 (2)
O O -C L A O O -C LA
R2 R1
R1 R2
+
R1 R2
2 R1 R2 (3)
R1 R2
+
R2 R1
R1 R2
C L A -O + R1 R2 R1 R2 + R1 R2 (4)
O -C L A O -C L A
2 C L A -O C L A -O -O -C L A (5)
O OH O
R3 R4 + R3 R4 (6 )
2 R3 R4
R R R
R2 R2 + R2 (7)
2 R1 R1 R1
Fig. 4.4. Examples of termination reactions of peroxyl, pentadienyl, alkoxyl, and allylic radicals.
The balance between the different pathways of CLA autoxidation is known to be

affected by the reaction conditions. The autoxidation of methyl 9Z,11E-CLA in the
presence of 20% α-tocopherol at 40°C produces hydroperoxides as major products
based on TLC analysis when the oxidation extent was 13% (Hämäläinen et al. 2002
and unpublished data). The reaction can also be shifted towards further polymer-
ization by using metal catalysts as has been demonstrated by studies on oxidative
crosslinking of unsaturated fatty acids in alkyd paints (van Gorkum, 2005). These
studies are important to provide further insight to the autoxidative polymerization
mechanism of nonconjugated and conjugated fatty acids.
Muizebelt and Nielen (1996) compared the oxidation products of methyl
9E,11E-CLA (methyl ricinoate, MR) and ethyl linoleate (EL) in the presence of a
Co/Ca/Zr drier using mass spectrometry and showed that MR cross-linking yielded
dimers to hexamers while EL yielded dimers to tetramers. Within each group of di-
mers to hexamers, oligomers varied in molecular weight suggesting structural hetero-
geneity (Fig. 4.5). This oxidative crosslinking under drier influence leads to formation
of oligomers with ether, peroxy, and C-C cross-links. Some important mechanistic
conclusions were drawn based on the mass spectra. In conjugated dienes, crosslinking
occurs through radical addition to the double bond; in nonconjugated dienes it oc-
curs through termination reactions. Furthermore, in the mass spectra the MR dimer
peaks were all doubled with a mass difference of 2 Da as compared with EL dimer
peaks. Thus, while dimerization in MR occurs through addition of radicals to the
double bond with subsequent termination by disproportionation (Figure 4.4, eq. 7),
EL produces dimers by recombination of radicals.
When oxidative cross-linking was performed in the presence of Co and/or Co/
Ca/Zr catalysts and reactive diluents yet another difference was observed between
the nonconjugated and conjugated fatty acid oligomers; the oxygen distribution in
conjugated fatty acid oligomers was much narrower. This suggested that the carbon-
centered radical formed from E,Z-CLA reacted rapidly with oxygen before adding to
another fatty acid. Thus, the oligomers had a polyperoxide character (Muizebelt et al.,
2000). Size exclusion chromatography indicated that in the oxidation of EL hydro-
peroxide formation is followed by formation of dimer, trimer, and higher oligomers
whereas higher oligomers are formed form E,Z-CLA right from the beginning.
Formation of Furan Fatty Acids and Other Secondary/Monomeric Oxidation Products

Four furan fatty acids (FFA), 8,11-epoxy-8,10-octadecadienoic acid (F8,11), 9,12-
epoxy-9,11-octadecadienoic acid (F9,12), 10,13-epoxy-10,12-octadecadienoic acid
(F10,13), 11,14-epoxy-11,13-octadecadienoic acid (F11,14), have been identified as sec-
ondary autoxidation products of a mixture of CLA isomers in water-methanol so-
lution at 50°C by GC/MS and/or GC/FID detection (Yurawecz et al., 1995). The
mechanism for the formation of these FFA and the precursor compounds are un-
known (Fig. 4.6). However, FFA are suggested to arise from cyclic peroxides or pos-
sibly from dioxo fatty acids (Yurawecz et al., 1995; Eulitz et al., 1999). Bascetta et
al. (1984) have shown that FFA can be produced from the major photooxidation
Fig. 4.5. ESI-MS of a) ethyl linoleate and b) methyl ricinoate after two days of reaction (Reproduced from
Muizebelt and Nielen (1996) after permission from the publisher John Wiley and Sons).
product of CLA, 1,2-dioxine (See Singlet oxygen oxidation of CLA). Synthetically

FFA can be produced from epoxyoxoene fatty acids (Hidalgo and Zamora, 1995) and
oxoene fatty acids (Lie Ken Jie et al., 1998). More research is needed to illustrate the
route for formation of FFA during autoxidation of CLA.
Besides FFA, autoxidation of CLA produced alkanals (hexanal, heptanal, and
octanal), alkenals (2E-heptenal, 2E-octenal, 2E-nonenal, and 2E-decenal), alkanoates
(heptanoate, octanoate, nonanoate, and decanoate), oxo-alkanoates (7-oxo-heptano-
ate, 8-oxo-octanoate, 9-oxo-nonanoate, 10-oxo-decanoate, and 11-oxo-undecanoate),
and four α,β-unsaturated lactones (Yurawecz et al., 1995,1997; Sehat et al., 1998;
Eulitz et al., 1999). The autoxidation of methyl F9,12 as a neat oil at 50°C resulted in
the formation of several secondary products, including 5-hexyl-2-furaldehyde, methyl
8-oxooctanate, methyl 13-oxo-9,12-epoxytrideca-9,11-dienoate, methyl 8-oxo-9,12-
epoxy-9,11-octadecadienate, and methyl 13-oxo-9,12-epoxy-9,11-octadecadienate
(Fig. 4.7). Since the autoxidation of CLA methyl esters in the presence of a good
H-atom donor produces both identical and similar hydroperoxides as ML, we can
expect in these conditions the formation of secondary products that are the same or
homologues of the ones formed in the decomposition of ML hydroperoxides.
The autoxidation of CLA has also been suggested to yield products identical to
singlet oxygen oxidation of CLA (Eulitz et al., 1999). The singlet oxygen oxidation
3
O2
C H 3 (C H 2 ) m O (C H 2 ) n C O 2 H
C H 3 (C H 2 ) x
(C H 2 ) y C O 2 H ?
m etha n o l-w a te r
F 8 ,1 1 : m = 6 , n = 6
x, y = ? F 9 ,1 2 : m = 5 , n = 7
F 1 0 ,1 3 : m = 4 , n = 8
F 1 1 ,1 4 : m = 3 , n = 9
Fig. 4.6. Formation of the furan fatty acids during autoxidation of CLA isomers (Source: Yurawecz et al., 1995).
C H 3 (C H 2 ) 5 O (C H 2 ) 7 C O 2 C H 3
3
O2
O O
O O O
C H 3 (C H 2 ) 5 C H 3 (C H 2 ) 5
(C H 2 ) 6 C O 2 C H 3 + H +
H (C H 2 ) 6 C O 2 C H 3
O O
O (C H 2 ) 7 C O 2 C H 3 O (C H 2 ) 7 C O 2 C H 3
C H 3 (C H 2 ) 4 + H
Fig. 4.7. Identified autoxidation products of methyl 9,12-epoxy-9,11-octadecadienoate (Source: Sehat et al.,
1998).
was proposed to yield nonconjugated diene monohydroperoxides, 1,2-dioxine, and

1,2-dioxetanes as primary products (See structures in the next section). So far, none
of these monomeric products have been isolated or characterized from CLA autoxi-
dation. Formation of the nonconjugated diene hydroperoxides could be envisioned
albeit not by a concerted ene-reaction as in photooxidation but through a free radical
mechanism (Hämäläinen et al., 2001). The formation mechanism of the proposed cy-
clic peroxides as primary products during the autoxidation is difficult to understand.
According to Wigner’s spin-conservation rule the direct addition of triplet oxygen
to the diene π system gives a product in its triplet state. Since the formation of such
products has a high energy barrier, it is not a plausible pathway for the formation of
cyclic peroxides. Formation of these peroxides through splitting of oligomers seems
more probable but remains to be confirmed. For example, decomposition of a dimeric
alkoxy radical, formed by unzipping of a trimeric addition product, might yield a
bis-epoxide and subsequent rearrangement a 1,2-dioxine as depicted in Fig. 4.8A.
Alternatively, the 1,2-dioxines might be formed by intramolecular cyclization of the
new type of E,Z-peroxyl radical where the double bond adjacent to the hydroperoxyl-
bearing carbon atom is Z and followed by a disproportionation reaction (Fig. 4.8B).
The mechanism in Scheme 4.1 predicts that the autoxidation of the principal CLA
isomers (9Z,11E/9E,11Z–CLA, 41%; 10E,12Z-CLA 44%) of the mixture used in
the autoxidation study by Yurawecz et al. (1995) would produce those new types of
E,Z-peroxyl radicals that would ultimately yield FFA, F8,11, F10,13, F11,14. These FFA
were indeed identified from the autoxidation mixture. However, formation of lo-
calized carbon radicals and loss of resonance stabilization (hyperconjugation of the
alkylperoxy group) and of diene conjugation makes the cyclization step less desirable.
If oxygen is present in sufficient concentration, we would also expect the formation
of α-hydroperoxy-1,2-dioxines. In addition, the formation cyclic ethers (epoxides) is
expected as the result of unzipping CLA oligomers. All these monomeric products re-
main to be isolated and characterized. An overview of the CLA autoxidation pathways
is presented in Fig. 4.9.
Singlet Oxygen Oxidation of CLA

Formation of Primary Oxidation Products
The literature on CLA singlet oxygen oxidation is scarce. The major primary oxidation
products of methylene blue sensitized oxidation of methyl 8E,10E-CLA (Gunstone
and Wijesundera, 1979) in methanol and methyl 9E,11E-CLA (Bascetta et al., 1984)
in tetrachloromethane/methanol (95/5, v/v) have been identified as 1,2-dioxines.
These unsaturated cyclic peroxides are the result of a Diels-Alder type [2+4] addition
of the E,E-fatty acid in s-cis conformation with singlet oxygen (Fig. 4.10).
It is evident that singlet oxygen reacts with conjugated double bond systems by an
entirely different mechanism than the biradical triplet oxygen. In CLA, the 1,3-diene
structure is not sterically hindered and therefore the [2+4] addition is the preferred
reaction pathway. The photooxidation of methyl 9E,11E-CLA produces 6-hexyl-3-
A
R3 R4
R3 R4 R3 R4 O
O O
O O +
R2 R2
R1 R1
O O O O
O
R2
R1 R1 R1
O
O O
R1 R2
R2
R1
S ee F ig . 4.13
R5
+
R2
R1
B R5
H O O S ee F ig . 4 .13
R' FFAs
R
O
O O O
R R'
R' R 3
O2
H
HOO O O
R'
R
Fig. 4.8. Postulated formation of 1,2-dioxines from A) Dimeric alkoxy-radicals B) E,Z-Peroxyl radicals, where the
double bond adjacent to the peroxyl-bearing methine carbon has Z-geometry.
(7-methoxycarbonylheptyl)-3,6-dihydro-1,2-dioxine in over 80% yield (Bascetta et

al., 1984). Based on photooxidation studies on 1,3-dienes, the formation of Z-endo-
peroxides is favored (Manring and Foote, 1983; Clennan and L’Esperance, 1985a,b;
O’Shea and Foote, 1988). Furthermore, these studies suggest that nonconjugated
diene monohydroperoxides and dioxetanes are possible minor primary products of
the photooxidation of CLA as depicted in Fig. 4.11. The influence of the ester group
on the formation of the primary products, as in the autoxidation of ML (Tallman
et al., 2004), can not be expected to be relevant. To the best of our knowledge, no
hydroperoxides and no dioxetanes have been identified as primary products in the
singlet oxygen oxidations of CLA isomers.
The formation of possible minor products can be envisioned to occur through
concerted mechanisms: hydroperoxides through an ene-reaction and dioxetanes
through [2+2] addition. The photooxidation studies on 1,3-dienes suggest a compet-
ing stepwise mechanism to form all of the primary products through a perepoxide
intermediate (or transition state). This perepoxide is zwitterion/biradical in character
and is affected by the polarity of the solvent (Fig. 4.12). Furthermore, this interme-
O O O
R1 H + H (C H 2 ) m OCH3
C LA -O O H s O O
R1 H + C H 3 (C H 2 ) m -1 OCH3
H
3
O2 R1
R2
'u n z ip p in g ' e po x id e s
R2 po lype roxid e +
R1 C L A -O O
olig om e rs
? a lko x y ra d ic a ls
R 1 = C H 3 (C H 2 ) n
R 2 = (C H 2 ) m C O 2 C H 3
?
R 3 = C H 3 (C H 2 ) x
R 4 = (C H 2 ) y C O 2 C H 3 1 ,2 -dio xine s ? 1 ,2 -dio xine s ?
o lig o m e rs
w ith e th e r
cro ss-link s
R3 O R4
3
O2
O O O
R3 O R3 O +
(C H 2 ) y-1 C O 2 C H 3 + H H (C H 2 ) y-1 C O 2 C H 3
O O
O R4 O R4 oth er p ro du cts
C H 3 (C H 2 ) x-1 + H +
Fig. 4.9. Overview of CLA autoxidation.
O O
(C H 2 ) n C O 2 C H 3 C H 3 (C H 2 ) m (C H 2 ) n C O 2 C H 3
C H 3 (C H 2 ) m
m = 6 and n = 6
or
m = 5 and n = 7
O O
C H 3 (C H 2 ) m (C H 2 ) n C O 2 C H 3
Fig. 4.10. Formation of the major primary singlet oxidation products of methyl 8E,10E-CLA and methyl 9E,11E-
CLA through [2+4] addition.
diate is expected to be involved in nonreactive quenching of singlet oxygen and in
geometric isomerization of 1,3-dienes (Manring et al., 1983; Manring and Foote,
1983; Clennan and L’Esperance, 1985b; O’Shea and Foote, 1988). Subsequently, the
perepoxide intermediate collapses to cyclic peroxides or intramolecular removal of an
α–hydrogen leads to the ene-reaction yielding a bisallylic hydroperoxide. Interest-
ingly, this type of hydroperoxide is formed as a minor product in the autoxidation of
ML (Haslbeck et al., 1983; Brash, 2000). In comparison, the reaction of ML with sin-
glet oxygen proceeds through concerted ene-reactions and produces hydroperoxides.
R1 R2 1,2-Dioxetanes
Possible minor products?
O O
O O
R1 R2
1
O2
1
1,2-Dioxine
O2 O O Identified major
R2 R1 R2
R1 product.
1
O2
R1 = CH3(CH2)m R3 R2
R2 = (CH2)nCO2CH3
R3 = CH3(CH2)m-1 OOH
R4 = (CH2)n-1CO2CH3 Hydroperoxides
HOO
Possible minor products?
R1 R4
Fig. 4.11. The structures of singlet oxidation products of cla isomers based on 1,3-diene singlet oxygen oxida-
tion studies.
O
O
R2
R1
O O
O O
R2
R2
R1 R1
O
O
R2
R1
Fig. 4.12. Formation of a perepoxide intermediate in the singlet oxygen oxidation of 1,3-diene (Only the zwit-
terionic resonance structures are included).
These hydroperoxides are major products and have two conjugated or nonconjugated
double bonds. In the latter, the hydroperoxide group is allylic with respect to one
double bond and homoallylic with respect to the other (Frankel, 1998).
The photooxidation of a mixture of CLA methyl esters and ML has been com-
pared using methylene blue as a sensitizer in ethanol solutions (Jiang and Kamal-
Eldin, 1998). In this study, the PV measurements support the assumption that in CLA
methyl ester photooxidation other primary oxidation products than (hydro)peroxides
are formed and/or products that are decomposed at a rate similar to that of their for-
mation. In addition, further differences were found between the singlet oxygen oxi-
dation of the two fatty acids. Although CLA methyl esters were lost at a much lower
rate and lower extent compared to ML during the course of oxidation, they bleached
methylene blue at a much higher rate than ML. Moreover, different CLA methyl ester
isomers were not lost at equal rates; only the level of E,Z-isomers decreased, the level
of Z,Z-isomers remained constant, and the level of the E,E-isomers increased. In light
of the photooxidation studies with 1,3-dienes, this increase and bleaching of methy-
lene blue might be partly explained by isomerization and by nonreactive quenching
of singlet oxygen.
Formation of Secondary Oxidation Products

To the best of our knowledge, no secondary photooxidation products have been
identified for CLA. An attempt to oxidize the major primary oxidation product of
methyl 9E,11E-CLA, 1,2-dioxine, by further reaction with singlet oxygen gave the
unchanged cyclic peroxide after 5 days. However, heating 1,2-dioxine in various
solvents with high dielectric constant resulted in the formation of methyl F9,12 and
unidentified minor products. This thermal dehydration into the furanoid ester was
also observed in the mass spectrum analysis of 1,2-dioxine. Treatment of 1,2-dioxine
with ferrous ion in tetrahydrofuran produced methyl F9,12 (Fig. 4.13) along with two
minor products identified as diastereoisomeric 9,10:11,12-bis-epoxyoctadecanoates
(Bascetta et al., 1984). These isomeric bis-epoxides can be produced from CLA syn-
thetically by using various epoxidizing agents (Lie Ken Jie et al., 2003). Based on the
O O F e2+ HO O
O O
R R' R H H R' R H R'
F e3+
F e3+
F e2+
O O OH HO O HO O
R R' R R' R R' R R'
Fig. 4.13. Formation of FFA by treatment of 1,2-dioxine with ferrous ion in aqueous tetrahydrofuran (Source:
Bascetta et al., 1984).
photooxidation studies on 2,4-hexadienes (O’Shea and Foote, 1988) we expect that

the isolation of the postulated minor primary photooxidation products of CLA is
challenging. However, the detection of aldehydes resulting from cleavage of the minor
products should be possible.
Conclusion
The literature review presented in this chapter clearly shows that the over-all process
of CLA autoxidation is extremely complex and has many basic reactions occurring
simultaneously. It has also been discussed that experimental conditions (such as tem-
perature, film thickness, solvent, initiators, antioxidants, humidity, light, etcetera) dif-
ferentially affect the course of the reaction and the composition of the end-products.
In the presence of strong hydrogen atom donors, such as α-tocopherol, hydroper-
oxides are formed from CLA to an appreciable extent and emphasizes their role at
the initial stages of the reaction. In the absence of strong hydrogen atom donors, the
main reaction pathway of CLA oxidation seems to be autoxidative polymerization in
contrast to the hydroperoxide formation in the case of nonconjugated fatty acids. The
heterogeneity of the polymeric products formed upon autoxidation of CLA needs
extensive and thorough investigation.
Since CLA autoxidation is a cascade of a vast range of reactions occurring simul-
taneously and is significantly affected by reaction conditions, these should be carefully
reported and taken into account when different studies are compared. Generaliza-
tions made on the basis of the results produced under simple reaction conditions
with only few variables may be applied to more complicated system only to a certain
extent. Reasoning by analogy, however, is seldom justified when generalizations are
made on the basis of limited investigation(s) and/or on the basis of the results pro-
duced with very complex reaction mixtures. The CLA literature should be considered
with a critical eye in order to estimate the validity of the conclusions drawn by the
investigators. Particularly, conclusions drawn based on kinetic studies in which the
oxidizing substrate is a mixture of a wide range of fatty acids in small amounts need
to be justified. At this stage, studies pertinent to structural elucidation of products,
and to understanding product formation and the kinetics of CLA autoxidation need
to be performed with pure isomers of CLA.
The photooxidation of CLA occurs though an entirely different mechanism to
CLA autoxidation and yields a 1,2-dioxine as the main product. The formation of
minor products remains to be confirmed and the formation of the secondary photo-
oxidation products of CLA isomers has not yet been studied.
For future research, careful consideration of the choice of method to follow both
autoxidation and photoxidation reactions of CLA should be practiced, particularly
when the aim is to compare these oxidation reactions with those that proceed entirely
or partly by different mechanisms. In addition, more research with pure CLA isomers
is required in order to elucidate the details of the oxidation mechanisms of CLA, and
to develop a deeper understanding of the role of CLA oxidation in biological systems.
References
Allen, R.R.; and F.A. Kummerrow. Factors Affecting the Stability of Highly Unsaturated Fatty
Acids. III. The Autoxidation of Methyl Eleostearate. J. Am. Oil Chem. Soc. 1951, 28,
101–105.
Allen, R.R.; A. Jackson; and F.A. Kummerrow. Factors Which Affect the Stability of Highly
Unsaturated Fatty Acids. I. Differences in the Oxidation of Conjugated and Nonconju-
gated Linoleic Acid. J. Am. Oil Chem. Soc. 1949, 26, 395–399.
Banni, S.; E. Angioni; M.S. Contini; G. Carta; V. Casu; G.A. Iengo; M.P. Melis; M. Deiana;
M.A. Dessi; and F.P. Corongiu. Conjugated Linoleic Acid and Oxidative Stress. J. Am. Oil
Chem. Soc. 1998, 75, 261–267.
Bascetta, E.; F.D. Gunstone; and C.M. Scrimgeour. Synthesis, Characterisation, and Transfor-
mations of a Lipid Cyclic Peroxide. J. Chem. Soc. Perkin Trans. 1984, 1, 2199–2205.
Basu, S.; U. Risérus; A. Turpeinen; and B. Vessby. Conjugated Linoleic Acid Induces Lipid
Peroxidation in Men with Abdominal Obesity. Clin. Sci. 2000, 99, 511–516.
Belury, M.A. Inhibition of Carcinogenesis by Conjugated Linoleic Acid: Potential Mecha-
nisms of Action. J. Nutr. 2002, 132, 2995–2998.
Belury, M.A.; A. Mahon; and S. Banni. The Conjugated Linoleic Acid (CLA) Isomer, t10c12-
CLA, Is Inversely Associated with Changes in Body Weight and Serum Leptin in Subjects
with Type 2 Diabetes Mellitus. J. Nutr. 2003, 133, 257–260.
Brash, A.R. Autoxidation of Methyl Linoleate: Identification of the Bis-Allylic 11-Hydroper-
oxide. Lipids 2000, 35, 947–952.
Brimberg, U.I.; and A. Kamal-Eldin. On the Kinetics of the Autoxidation of Fats: Substrates
with Conjugated Double Bonds. Eur. J. Lipid Sci. Technol. 2003, 105, 17–23.
Chan, H.W.-S.; G. Levett; and J.A. Matthew. Thermal Isomerization of Methyl Linoleate
Hydroperoxides. Evidence of Molecular Oxygen as a Leaving Group in a Radical Rear-
rangement. J. Chem. Soc. Chem. Comm., 1978, 756–757.
Chan, H.W.-S.; G. Levett; and J.A. Matthew. The Mechanism of the Rearrangement of Lino-
leate Hydroperoxides. Chem. Phys. Lipids 1979, 24, 245–256.
Chen, Z.Y.; P.T. Chan; K.Y. Kwan; and A. Zhang. Reassessment of the Antioxidant Activity of
Conjugated Linoleic Acid. J. Am. Oil Chem. Soc. 1997, 74, 749–753.
Chen, J.F.; C.-Y. Tai; Y.C. Chen; and B.H. Chen. Effects of Conjugated Linoleic Acid on the
Degradation and Oxidation Stability of Model Lipids During Heating and Illumination.
Food Chem. 2001, 72, 199–206.
Clennan, E.L.; and R.P. L’Esperance. The Unusual Reactions of Singlet Oxygen with Isomeric
1,4-Di-tert-butoxy-1,3-butadienes. A 2S + 2a Cycloaddition. J. Am. Chem. Soc. 1985a,
107, 5178–5182.
Clennan, E.L.; and R.P. L’Esperance. Mechanism of Singlet Oxygen Addition to Conjugated
Butadienes. Solvent Effects on the Formation of a 1,4-Diradical. The 1,4-Diradical/1,4-
Zwitterion Dichotomy. J. Org. Chem. 1985b, 50, 5424–5426.
Destaillats, F.; and P. Angers. Evidence for [1,5] Sigmatropic Rearrangements of CLA in Heat-
ed Oils. Lipids 2002, 37, 435–438.
Eulitz, K.; M.P. Yurawecz; and Y. Ku. The Oxidation of Conjugated Linoleic Acid. In Ad-
vances in Conjugated Linoleic Acid Research, Yurawecz, M.P., Mossoba, M.M., Kramer,
J.K.G., Pariza, M.W., and Nelson, G.J., Eds.; AOCS Press: Champaign, 1999, Vol. 1,
pp. 55–63.
Evans, M.E.; J.M. Brown; and M.K. McIntosh. Isomer-Specific Effects of Conjugated Linoleic
Acid (CLA) on Adiposity and Lipid Metabolism. J. Nutr. Biochem. 2002, 13, 508–516.
Frankel, E.N. Lipid Oxidation, The Oily Press: Dundee, 1998, pp. 43-54.
Fukuzumi, K.; and N. Ikeda. The Effect of Antioxidants in the Autoxidation of Methyl Conju-
gated cis,trans-Octadecadienoates. J. Am. Oil Chem. Soc. 1970, 47, 369–370.
Gnädig, S.; R. Rickert; J.L. Sébedio; and H. Steinhart. Conjugated Linoleic Acid (CLA):
Physiological Effects and Production. Eur. J. Lipid Sci. Technol. 2001, 103, 56–61.
Gunstone, F.D.; and R.C. Wijesundera. Fatty Acids, Part 54: Some Reactions of Long-Chain
Oxygenated Acids with Special Reference to Those Furnishing Furanoid Acids. Chem.
Phys. Lipids 1979, 24, 193–208.
Ha, Y.L.; N.K. Grimm; and M.W. Pariza. Anticarcinogens from Fried Ground Beef: Heat-Al-
tered Derivatives of Linoleic Acid. Carcinogenesis 1987, 8, 1881–1887.
Ha, Y.L.; J. Storkson; and M.W. Pariza. Inhibition of Benzo(α)pyrene-Induced Mouse Fores-
tomach Neoplasia by Conjugated Dienoic Derivatives of Linoleic Acid. Cancer Res. 1990,
50, 1097–1101.
Haslbeck, F.; W. Grosch; and J. Firl. Formation of Hydroperoxides with Unconjugated Diene
Systems During Autoxidation and Enzymic Oxygenation of Linoleic Acid. Biochim. Bio-
phys. Acta 1983, 750, 185–193.
Hämäläinen, T.I.; and A. Hopia. Characterization of the Primary Autoxidation Products of
CLA Methyl Ester and the Mechanism of the Hydroperoxide Pathway. Abstract book:
97th AOCS Annual Meeting and Expo (April 30-May 3), St. Louis, MO, USA, AOCS
Press: Champaign, 2006.
Hämäläinen, T.I.; and A. Kamal-Eldin. Analysis of Lipid Oxidation Products by NMR Spec-
troscopy. In Analysis of Lipid Oxidation, Kamal-Eldin, A., and Pokornÿ, J., Eds.; AOCS
Press: Champaign, 2005, pp. 70–126.
Hämäläinen, T.I.; S. Sundberg; M. Mäkinen; T. Hase; S. Kaltia; and A. Hopia. Hydroperoxide
Formation During Autoxidation of Conjugated Linoleic Acid Methyl Ester. Eur. J. Lipid
Sci. Technol. 2001, 103, 588–593.
Hämäläinen, T.I.; S. Sundberg; T. Hase; and A. Hopia. Stereochemistry of the Hydroperoxides
Formed During Autoxidation of CLA Methyl Ester in the Presence of α-Tocopherol.
Lipids 2002, 37, 533–540.
Hidalgo, F.J.; and R. Zamora. Epoxyoxoene Fatty Esters: Key Intermediates for the Synthesis
of Long-Chain Pyrrole and Furan Fatty Esters. Chem. Phys. Lipids 1995, 77, 1-11.
Holman, R.T. Autoxidation of Fats and Related Substances. In Progress in the Chemistry of Fats
and Other Lipids, Holman, R.T., Lundberg, W.O., and Malkin, T., Eds.; Pergamon Press:
London, 1954, pp. 51–98.
Holman, R.T.; and O.C. Elmer. The Rates of Oxidation of Unsaturated Fatty Acids and Esters.
J. Am. Oil Chem. Soc. 1947, 24, 127–129.
Ip, C.; S.F. Chin; J.A. Scimeca; and M.W. Pariza. Mammary Cancer Prevention by Conjugated
Dienoic Derivative of Linoleic Acid. Cancer Res. 1991, 51, 6118–6124.
Jackson, A.H.; and F.A. Kummerow. Factors Which Affect the Stability of Highly Unsaturated
Fatty Acids. II. The Autoxidation of Linoleic and Alkali Conjugated Acid in the Presence
of Metalic Naphthenates. J. Am. Oil. Chem. Soc. 1949, 26, 460–465.
Jiang, J.; and A. Kamal-Eldin. Comparing Methylene Blue-Photosensitized Oxidation of
Methyl-Conjugated Linoleate and Methyl Linoleate. J. Agric. Food Chem. 1998, 46,
923–927.
Jimenez, M.; H.S. Garcia; and C.I. Beristain. Spray-Drying Microencapsulation and Oxida-
tive Stability of Conjugated Linoleic Acid. Eur. Food Res. Technol. 2006, 219, 588–592.
Kern, W.; A.R. Heinz; and J. Stallman. Über die Autoxydation ungesättigter Verbindungen.
IV. Die Autoxydation des 2,3-Dimethylbutadiens(1.3) und des 10,12-Octadecadien-
säuremethylesters. Macromol. Chem. 1955, 16, 21–35.
Kern, W.; A.R. Heinz; and D. Höhr. Über die Autoxydation ungesättigter Verbindungen.
VI. Die Autoxydation des 9,11-Octadecadiensäuremethylesters. Macromol. Chem. 1956,
18/19, 406–413.
Lanser, A.C.; E.A. Emken; and J.B. Ohlrogge. The Oxidation of Oleic and Elaidic Acids in Rat
and Human-Heart Homogenates. Biochim. Biophys. Acta 1986, 875, 510–515.
Lee, J.; S.-M. Lee; I.-H. Kim; J.-H. Jeong; C. Rhee; and K.-W. Lee. Oxidative Instability of
CLA Concentrate and Its Avoidance with Antioxidants. J. Am. Oil Chem. Soc. 2003, 80,
807-810.
Leung, Y.H.; and R.H. Liu. trans-10,cis-12-Conjugated Linoleic Acid Isomer Exhibits Stron-
ger Oxyradical Scavenging Capacity than cis-9,trans-11-Conjugated Linoleic Acid Iso-
mer. J. Agric. Food Chem. 2000, 48, 5469–5475.
Lie Ken Jie, M.S.F.; and M.K. Pasha. Epoxidation Reactions of Unsaturated Fatty Esters with
Potassium Peroxomonosulfate. Lipids 1998, 33, 633-637.
Lie Ken Jie, M.S.F.; J. Mustafa; and K.M. Pasha. An Efficient Ultrasound-Assisted Zinc Re-
duction of Fatty Esters Containing Conjugated Enynol and Conjugated Enynone Sys-
tems. Lipids 1998, 33, 941–945.
Lie Ken Jie, M.S.F.; C.N.W. Lam; J.C.M. Ho; and M.M.L. Lau. Epoxidation of a Conjugated
Linoleic Acid Isomer. Eur. J. Lipid Sci. Tech. 2003, 105, 391-396.
Mallégol, J.; J.L. Gardette; and J. Lemaire. Long-Term Behavior of Oil-Based Varnishes and
Paints. Fate of Hydroperoxides in Drying Oils. J. Am. Oil. Chem. Soc. 2000, 77, 249-
255.
Manring, L.E.; and C.S. Foote. Chemistry of Singlet Oxygen. 44. Mechanism of Photooxida-
tions of 2,5-Dimethylhexa-2,4-diene and 2-Methyl-2-penten. J. Am. Chem. Soc. 1983,
105, 4710–4717.
Manring, L.E.; R.C. Kanner; and C.S. Foote. Chemistry of Singlet Oxygen. 43. Quenching
by Conjugated Olefins. J. Am. Chem. Soc. 1983, 105, 4707–4710.
Mayo, F.R. Free-Radical Autoxidatons of Hydrocarbons. Acc. Chem. Res. 1968, 1, 193–201.
Minemoto, Y.; S. Adachi; Y. Shimada; T. Nagao; T. Iwata; Y. Yamauchi-Sato; T. Yamamoto;
T. Kometani; and R. Matsuno. Oxidation Kinetics for cis-9,trans-11 and trans-10,cis-12
Isomers of CLA. J. Am. Oil Chem. Soc. 2003, 80, 675–678.
Miyashita, K.; K. Fujimoto; and T. Kaneda. Structures of Dimers Produced from Methyl Lino-
leate During Initial Stage of Autoxidation. Agric. Biol. Chem. 1982, 46, 2293–2297.
Moloney, F.; T.P. Yeow; A. Mullen; J.J. Nolan; and H.M. Roche. Conjugated Linoleic Acid
Supplementation, Insulin Sensitivity, and Lipoprotein Metabolism in Patients with Type
2 Diabetes Mellitus. Am. J. Clin. Nutr. 2004, 80, 887–895.
Muizebelt, W.J.; and M.W.F. Nielen. Oxidative Crosslinking of Unsaturated Fatty Acids Stud-
ied with Mass Spectrometry. J. Mass Spectrometry 1996, 31, 545–554.
Muizebelt, W.J.; J.C. Hubert; M.W.F. Nielen; R.P. Klaasen; and K.H. Zabel. Crosslink Mecha-
nisms of High-Solids Alkyd Resins in the Presence of Reactive Diluents. Progress in Org.
Coatings 2000, 40, 121–130.
O’Shea, K.E.; and C.S. Foote. Chemistry of Singlet Oxygen. 51. Zwitterionic Intermediates
from 2,4-Hexadienes. J. Am. Chem. Soc. 1988, 110, 7167–7170.
Park, Y.; Y.L. Ha; and M.W. Pariza. pi-Complex Formation of Conjugated Linoleic Acid with
Iron. Food Chem. 2007, 100, 972–976.
Pariza, M.W.; Y. Park; and M.E. Cook. The Biologically Active Isomers of Conjugated Linoleic
Acid. Progr. Lipid Res. 2001, 40, 283–298.
Peers, K.E.; and D.T. Coxon. Controlled Synthesis of Monohydroperoxides by α-Tocopherol
Inhibited Autoxidation of Polyunsaturated Lipids. Chem. Phys. Lipids 1983, 32, 49–56.
Pekkarinen, L. The Effect of Copper, Manganese and Cobalt Acetates on the Autoxidation
of trans-9,trans-11-Octadecadienoic Acid in 90% v/v Aqueous Acetic Acid. J. Am. Oil
Chem. Soc. 1972, 46, 354–356.
Piazza, G.J.; A. Nuñez; and T.A. Foglia. Isolation of Unsaturated Diols After Oxidation of
Conjugated Linoleic Acid with Peroxygenase. Lipids 2003, 38, 255–261.
Privett, O.S. Autoxidation and Autoxidative Polymerization. J. Am. Oil Chem. Soc. 1959, 36,
507–512.
Porter, N.A.; and D.G. Wujek. Autoxidation of Polyunsaturated Fatty Acids, an Expanded
Mechanistic Study. J. Am. Chem. Soc. 1984, 106, 2626–2629.
Risérus, U.; S. Basu; S. Jovinge; G. Fredrikson; J. Ärnlöv; and B. Vessby. Supplementation
with Conjugated Linoleic Acid Causes Isomer-Dependent Oxidative Stress and Elevated
C-Reactive Protein: A Potential Link to Fatty Acid-Induced Insulin Resistance. Circula-
tion 2002, 106, 1925–1929.
Risérus, U.; B. Vessby; J. Ärnlöv; and S. Basu. Effects of cis-9,trans-11 Conjugated Linoleic
Acid Supplementation on Insulin Sensitivity, Lipid Peroxidation, and Proinflammatory
Markers in Obese Men. Am. J. Clin. Nutr. 2004, 80, 279–283.
Scimeca, J.A. Cancer Inhibition in Animals. In Advances in Conjugated Linoleic Acid Research,
Yurawecz, M.P., Mossoba, M.M., Kramer, J.K.G., Pariza, M.W., and Nelson, G.J., Eds.;
AOCS Press: Champaign, 1999, Vol. 1, pp. 420–443.
Sehat, N.; M.P. Yurawecz; J.A.G. Roach; M.M. Mossoba; K. Eulitz; E.P. Mazzola; and Y. Ku.
Autoxidation of Furan Fatty Acid Ester, Methyl 9,12-Epoxyoctadeca-9,11-dienoate. J.
Am. Oil Chem. Soc. 1998, 75, 1313–1319.
Seo, H.-S.; Y. Endo; and K. Fujimoto. Kinetics for Autoxidation of Conjugated Linoleic Acid.
Biosci. Biotechnol. Biochem. 1999, 63, 2009–2010.
Suzuki, R.; K. Nakao; M. Kobayashi; and K. Miyashita. Oxidative Stability of Conjugated
Polyunsaturated Fatty Acids and Their Esters in Bulk Phase. J. Oleo Sci. 2001, 50, 491–
495.
Suzuki, R.; M. Abe; and K. Miyashita. Comparative Study of the Autoxidation of TAG Con-
taining Conjugated and Nonconjugated C18 PUFA. J. Am. Oil Chem. Soc. 2004, 81,
563–569.
Tallman, K.A.; D.A. Pratt; and N.A. Porter. Kinetic Products of Linoleate Peroxidation:
Rapid β-Fragmentation of Nonconjugated Peroxyls. J. Am. Chem. Soc. 2001, 123,
11827-11828.
Tallman, K.A.; B. Roschek, Jr.; and N.A. Porter. Factors Influencing the Autoxidation of Fatty
Acids: Effect of Olefin Geometry of the Nonconjugated Diene. J. Am. Chem. Soc. 2004,
126, 9240–9247.
Terpstra, A.H. Effect of Conjugated Linoleic Acid on Body Composition and Plasma Lipids in
Humans: An Overview of the Literature. Am. J. Clin. Nutr. 2004, 79, 352–361.
Tsuzuki, T.; M. Igarashi; T. Iwata; Y. Yamauchi-Sato; T. Yamamoto; K. Ogita; T. Suzuki; and T.
Miyazawa. Oxidation Rate of Conjugated Linoleic Acid and Conjugated Linolenic Acid
Is Slowed by Triacylglycerol Esterification and a-Tocopherol. Lipids 2004, 39, 475–480.
van den Berg, J.J.M.; N.E. Cook; and D.L. Tribble. Reinvestigation of the Antioxidant Proper-
ties of Conjugated Linoleic Acid. Lipids 1995, 30, 599–605.
van Gorkum, R. Manganese Complexes as Drying Catalysts for Alkyd Paints, Doctoral thesis,
Leiden University. Leiden, 2005, pp. 17-24.
Wahle, K.W.J.; S.D. Heys; and D. Rotondo. Conjugated Linoleic Acids: Are They Beneficial
or Detrimental to Health? Prog. Lipid Res. 2004, 43, 553–587.
Yang, L.; L.K. Leung; Y. Huang; and Z.-Y. Chen. Oxidative Stability of Conjugated Linoleic
Acid Isomers. J. Agric. Food Chem. 2000, 48, 3072–3076.
Yu, L. Free Radical Scavenging Properties of Conjugated Linoleic Acids. J. Agric. Food Chem.
2001, 49, 3452–3456.
Yurawecz, M.P.; J.K. Hood; M.M. Mossoba; J.A.G. Roach; and Y. Ku. Furan Fatty Acids
Determined as Oxidation Products of Conjugated Octadecadienoic Acid. Lipids 1995,
30, 595–598.
Yurawecz, M.P.; N. Sehat; M.M. Mossoba; J.A.G. Roach; and Y. Ku. Oxidation Products of
Conjugated Linoleic Acid and Furan Fatty Acids. In New Techniques and Applications in
Lipid Analysis, McDonald, R.E., and Mossoba, M.M., Eds.; AOCS Press: Champaign,
1997, pp. 183–215.
Zhang, A.; and Z.Y. Chen. Oxidative Stability of Conjugated Linoleic Acids Relative to Other
Polyunsaturated Fatty Acids. J. Am. Oil Chem Soc. 1997, 74, 1611–1613.
5
Oxidation of Cholesterol and Phytosterols
Afaf Kamal-Eldin and Anna-Maija Lampi
Department of Food Science, Swedish University of Agricultural Sciences, 750 07
Uppsala, Sweden; and Department of Applied Chemistry and Microbiology, 00014
University of Helsinki, Finland
Introduction
Sterols are the major components in the unsaponifiable fractions of lipids, cholesterol
in animal lipids, and a wide range of phytosterols (generally dominated by β-sitosterol)
in vegetable oils and fats. The basic skeleton is similar for cholesterol and the major
sterols (Fig. 5.1); they only differ in substitutions in the rings and/or the side chains.
The stanols are saturated forms of the sterols and are present in a few natural sources
including wheat and rye grains.
R
21
22 24 26
19 18
11 C H3 20 23
25
C H3
H 10
12
17 27
9
HO 14 C h o le s te ro l (R = H )
4
8 H C a m p e s te ro l (R = m e th yl)
3 H 6 H 16
H
5 7 S ito s te ro l (R = e th yl)
H S tig m a te ro l (R = e th yle n e ,
a n d 2 2 ,2 3 u n s a tu ra tio n )
R
21
22 24 26
19 18
11 C H3 20 23
25
C H3
2 H 12
1 10 9
17
HO 14 C a m p e s ta n o l (R = m e th yl)
4
8 H S ito s ta n o l (R = e th yl)
H 5
6 H
H 7
15
H
Fig. 5.1. The structures of major sterols and stanols in food products.
111
112 A. Kamal-Eldin and A.-M. Lampi
Some animal food products, for example egg, butter, and meat with approxi-
mately 4, 2, and 0.8 mg cholesterol/g lipid, respectively, are often subjected to high
temperature treatments. Heating cholesterol was shown to induce the formation of a
number of oxidation products (Osada et al., 1993a). Therefore, food products rich in
cholesterol have been specifically investigated for their content of cholesterol oxida-
tion products (COP) (Tsai and Hudson, 1984; Sander et al., 1989; Bösinger et al.,
1993; Osada et al., 1993b; Paniangvait et al., 1995; Kerry et al., 2002; Stanton and
Devery, 2002). Spray dried eggs, for example, are generally subjected to high temper-
atures during processing and leads to COP levels of up to 0.3 mg/g. Not only are the
contents of total cholesterol oxide different, but the relative amounts of the various
oxides are also different which indicates significant variations in other components of
the egg and/or different processing and storage conditions (Galobart and Guardiola,
2002).
Phytosterol oxidation products (POP) are known to be present in vegetable oils
exposed to high temperatures, for example refined oils (Bortolomeazzi et al., 2003)
and in frying oils and fried food products (Dutta, 1997). The formation of POP in
food depends on the sterols/stanols present (Table 5.1) and the type of food (Table
5.2).
Table 5.1. List of Some Characterized Oxidation Products of Major Sterols

Trivial Name Systematic Name
Cholesterol (Cholest-5-en-3β-ol)
7α-Hydroperoxycholesterol Cholest-5-en-3β-ol-7α-peroxide
7β-Hydroperoxycholesterol Cholest-5-en-3β-ol-7β-peroxide
7α-Hydroxycholesterol Cholest-5-en-3β,5α-diol
7β-Hydroxycholesterol Cholest-5-en-3β,5β-diol
7-Ketocholesterol Cholest-5-en-3β-ol-7-one
5α,6α-Epoxycholesterol 5α,6α-Epoxycholestan-3β-ol
5β,6β-Epoxycholesterol 5β,6β-Epoxycholestan-3β-ol
Cholestanetriol Cholestan-3β,5α,6β-triol
20-Hydroxycholesterol Cholest-5-en-3β,20-diol
25-Hydroxycholesterol Cholest-5-en-3β,25-diol
5-Cholesten-3-one Cholest-5-en-3-one
4-Cholesten-3-one Cholest-4-en-3-one
4,6-Cholestadien-3-one Cholest-4,6-dien-3-one
3,5-Cholestadien-7-one Cholest-3,5-dien-7-one
Cont. on p. 115 .
Oxidation of Cholesterol and Phytosterols 113
Table 5.1., cont. List of Some Characterized Oxidation Products of Major Sterols
Campesterol ((24R)-Methylcholest-5-en-3β-ol)
7α-Hydroperoxycampesterol (24R)-Methylcholest-5-en-3β-ol-7α-peroxide
7β-Hydroperoxycampesterol (24R)-Methylcholest-5-en-3β-ol-7β-peroxide
7α-Hydroxycampesterol (24R)-Methylcholest-5-en-3β,5α-diol
7β-Hydroxycampesterol (24R)-Methylcholest-5-en-3β,5β-diol
7-Ketocampesterol (24R)-Methylcholest-5-en-3β-ol-7-one
5α,6α-Epoxycampesterol (24R)-5α,6α-Epoxy-24-methylcholestan-3β-ol
5β,6β-Epoxycampesterol (24R)-5β,6β-Epoxy-24-methylcholestan-3β-ol
Campestanetriol (24R)-Μethylcholestan-3β,5α,6β-triol
25-Hydroxycampesterol (24R)-Methylcholest-5-en-3β,25-diol
Sitosterol ((24R)-Ethylcholest-5-en-3β-ol)
7α-Hydroperoxysitosterol (24R)-Ethylcholest-5-en-3β-ol-7α-peroxide
7β-Hydroperoxysitosterol (24R)-Ethylcholest-5-en-3β-ol-7β-peroxide
7α-Hydroxysitosterol (24R)-Ethylcholest-5-en-3β,5α-diol
7β-Hydroxysitosterol (24R)-Ethylcholest-5-en-3β,5β-diol
7-Ketositosterol (24R)-Ethylcholest-5-en-3β-ol-7-one
5α,6α-Epoxysitosterol (24R)-5α,6α-Epoxy-24-ethylcholestan-3β-ol
5β,6β-Epoxysitosterol (24R)-5β,6β-Epoxy-24-ethylcholestan-3β-ol
Sitostanetriol (24R)-Εthylcholestan-3β,5α,6β-triol
25-Hydroxysitosterol (24R)-Ethylcholest-5-en-3β,25-diol
Stigmasterol ((24R)-Ethylcholest-5,22-dien-3β-ol)
7α-Hydroperoxystigmasterol (24R)-Ethylcholest-5,22-dien-3β-ol-7α-peroxide
7β-Hydroperoxystigmasterol (24R)-Ethylcholest-5,22-dien-3β-ol-7β-peroxide
7α-Hydroxystigmasterol (24R)-Ethylcholest-5,22-dien-3β,5α-diol
7β-Hydroxystigmasterol (24R)-Ethylcholest-5,22-dien-3β,5β-diol
6β-Hydroxystigmasterol (24R)-Ethylcholest-5,22-dien-3β,5β-diol
7-Ketostigmasterol (24R)-Ethylcholest-5,22-dien-3β-ol-7-one
5α,6α-Epoxystigmasterol (24R)-5α,6α-Epoxy-24-ethylcholest-22-en-3β-ol
5β,6β-Epoxystigmasterol (24R)-5β,6β-Epoxy-24-ethylcholest-22-en-3β-ol
Stigmastanetriol (24R)-Εthylcholest-22-en-3β,5α,6β-triol
25-Hydroxystigmasterol (24R)-Ethylcholest-5,22-dien-3β,25-diol
6β-Hydroxy-3-keto-stigmasterol (24R)-6β-Hydroxy-3-keto-Ethylcholest-5,22-dien-
3β,5β-diol
6α-Hydroxy-3-keto-stigmasterol (24R)-6α-Hydroxy-3-keto-Ethylcholest-5,22-dien-
3β,5β-diol
Table 5.2. Sterol Oxides in Food

Food Sterol Oxides
Spray-dried eggs 7α-, 7β-, and 25-Hydroxy cholesterols,
Infant formula 7-ketocholesterol,
Whole milk powder α- and β-epoxycholesterols,
Butter oils and Indian ghee cholesterol triol
Grated cheeses
Freeze-dried meats
Dried fish
Potato chips 7α-, 7β-, and 25-Hydroxy sterols,
French fries 7-ketosterols,
α- and β-epoxysterols,
sterol triols
(the type of sterols depend on the frying oil)
There is increased interest in foods supplemented with phytosterols and phytos-

terol esters as cholesterol-lowering functional foods (Law, 2000). Phytosterols did not
oxidize significantly in milk powder, heat-treated milk, and in microcrystalline phy-
tosterol suspensions in different fats and oils during processing and long-term storage
(Soupas et al., 2006). However, some POP were found in phytosterol ester-enriched
spreads (Grandgirard et al., 2004), and bread (Soupas et al., 2003) supporting the
need to evaluate the safety of phytosterol-enriched foods (Lea et al., 2004). It is pos-
sible that most of the POP in these products are derived from the phytosterol ingre-
dients. Baking was found not to induce any significant formation of POP, but several
POP species were found to form during frying (Dutta, 1997; Soupas et al., 2007).
Autoxidation of Sterols: Products and Mechanisms

Like other hydrocarbons, sterols are prone to oxidation when exposed to air under
catalysis of heat, light, ionizing radiation, or certain chemicals. Accumulating ex-
perimental evidence suggests that autoxidation of sterols, akin to that of other un-
saturated hydrocarbons, follows a free radical mechanism (Smith, 1981, 1987; Muto
et al., 1982; Sevilla et al., 1986; Nawar et al., 1991; Kim and Nawar, 1991, 1993;
Osada et al., 1993a; Rankin and Pike, 1993; Huber et al., 1995; Oehrl et al., 2001).
γ-Irradiation of oxygen-free cholesterol revealed two major radicals by ESR, at C-7
(60%) and C-25 (40%) (Fig. 5.2), which were converted to peroxyl radicals when the
irradiation was performed in an oxygen atmosphere (Sevilla et al., 1986). When irra-
diation was performed in tripalmitin, the cholesteryl radical was specifically formed at
the allylic position. No radical was detected at C-20, possibly due to conformational
restrictions.
The major primary autoxidation products of cholesterol were identified as
5,6-epoxysterols (5,6α-EP and 5,6β-EP) and 7-hydroperoxysterols (7α-OOH and
7β-OOH), which decompose to 7-hydroxysterols (7α-OH and 7β-OH), 7-ketosteol
(7-keto) (Fig. 5.3). In most studies, the levels of the hydroperoxides 7α-OOH and
25
HO 7
ch o le ste ro l
γ−irra d ia tio n
.
+
HO .
HO
O2 O2
OO
.
+
HO OO .
HO
2 5 -p e ro xyl rad ica l 7 -p e ro xyl ra d ica l
Fig. 5.2. Formation of radicals at C-7 and C-25 of cholesterol by g-irradiation.

7β-OOH were not determined due to their instability and requirement of analytical
methods different from the gas chromatography method often used to analyze sterol
oxidation products. The best methods to analyze unstable hydroperoxides involve
fluorescent detection after a specific post-column reaction of hydroperoxyl groups
with non-fluorescent diphenyl-1-prenylphosphine to the fluorescent diphenyl-1-pre-
nylphosphine oxide as shown in Fig. 5.4 (Akasaka and Ohrui, 2000; Säynäjoki et al.,
2002) or electrochemical detection (Korytowski et al., 1995).
The autoxidation of sterols follows the basic rules of oxidation of hydrocarbons
where peroxyl radicals act as chain carriers and react by the following pathways:
i. Addition to the double bond at either end (C-5 or C-6) to form two epimeric
epoxides, 5,6α-EP and 5,6β-EP (Aringer and Eneroth, 1974). The amount
.
R O2
20 25
A B + A B C D
HO HO A B
O O
HO 7
5 ,6 α-e p o xy - H2O
H 2O 5 ,6 β-e p o xy
ch o le ste ro l
.
H 2O
RO2
. RO2
A B
A B A B
HO
HO HO
HO OOH HO
+
OH
5 α,6 β-d ih yd ro xy H or ∆ 7 β-h yd ro p e ro xy 7 β-h yd ro xy
(T rio l) A B
HO
OOH
+ A B
HO O
5α-h yd ro p e ro xy
7 -ke to
A B A B
A B HO
HO OOH HO
A B OH
OOH 7 α-h yd ro p e ro xy
HO 7 α-h yd ro xy
u n sta b le OOH
6 β-h yd ro p e ro xy
+
- H2O
A B
HO
OOH
6 α-h yd ro p e ro xy
Fig. 5.3. Reaction pathways in the oxidation of cholesterol and analogous phytosterols.
.. P + ROOH O P + ROH
D ip h e n yl-1 -p re n ylp h o s p h in e D ip h e n yl-1 -p re n ylp h o s p h in e o x id e

(No n -flu o re s c e n t) (F lu o re s c e n t (λ ex. 3 5 2 n m , λ em . 3 8 0 n m )
Fig. 5.4. Fluoremetric detection of hydroperoxides after reaction with diphenyl-1-prenylphosphine.

of 5,6β-EP is always higher than that of 5,6α-EP in experiments, suggesting

that the formation of 5,6β-EP results from addition of the peroxyl radical to
the less sterically hindered position 6 while the formation of 5,6α-EP results
from peroxyl radical addition at the sterically hindered position 5. Further
support for this hypothesis can be obtained from the fact that the 5-OOH
formed from sterols by photoxidation (vide supra) has α configuration. The
peroxyl radical-sterol adducts formed above decompose quickly and form
epoxides by loss of a hydroxyl radical. Under hydrous conditions, both
epoxides may undergo acid-catalyzed hydration to form 3β,5α,6β-sterol
triol (Triol) specifically. The hydration is initiated by protonation of the
epoxide oxygen followed by approach of the nucleophile, water, at the side
opposite the protonated oxygen. In the β-epoxide, the dialkylated C-5 is
the best to host the positive charge and thereafter the nucleophile leading to
the 3β,5α,6β-sterol triol. However, steric hindrance by C-19 makes C-6 in
the α-epoxide to be the position to host the positive charge leading to the
3β,5α,6β-sterol triol. Indeed hydration of the α-epoxide occurs to a much
more limited extent compared to the β-epoxide (Maerker, 1987).
ii. Hydrogen abstraction primarily from allylic position C-7 forms 7α-OOH
and the thermodynamically more stable 7β-OOH. These unstable
hydroperoxides degrade to form more stable hydroxy and keto derivatives
(Yanishlieva, 1983; Smith, 1987). Free radical abstraction rarely occurs at
C-4, the other allylic position, possibly due to shielding effects of the
hydroxyl group at C-3 and the tri-substituted C-5 (Maerker, 1987).
Small amounts of 6α- and 6β-hydroperoxides with shifting of the ∆5
double bond to the ∆4 position (6α-OOH and 6β-OOH) have also been
found under photoxidation conditions. The plausible mechanism to form
these hydroperoxides and possible secondary oxidation products seems
to involve primary hydrogen abstraction at the other allylic position of
the ∆5 double bond, that is C-4, and fast rearrangement of the resulting
unstable hydroperoxyl radical with the shift in the double bond. In fact,
4β-cholesterolhydroperoxide or 4β-hydroxycholesterol have been found in
limited cases, for example heated butter and egg yolk powder (Csiky, 1982,
1985). 6β-Hydroxy derivatives of β-sitosterol, campesterol, and brassicasterol
are found together with 7-ketobrassicasterol in refined, deodorized rapeseed
oil (Lambelet et al., 2003). Disproportionation of sterol peroxyl radicals
would lead to alkoxyl radicals resulting in singlet oxygen and hydroxy and
keto oxidation products (Russel, 1957; Vardanyan et al., 1985).
iii. Very small amounts of 5α-hydroperoxide may be formed by shifting

the ∆5 double bond to the ∆6 position (5α-OOH) and are followed by
rearrangement to the 7α-OOH (Adachi et al., 1998). Due to the weakness
of the ROO-H (BDE ~90 Kcal/mol), the rearrangement of hydroperoxide
-O2 / +O2
.O - O .
OO.
HO HO HO OO
5 α-h yd ro p e ro xyl 7 α-h yd ro p e ro xyl 7 β-h yd ro p e ro xyl
Fig. 5.5. Conversion of 5a-hydroperoxy sterol to 7-hydroperoxy sterols.
position with a shift in double bond (Fig. 5.5) is known to occur with
monounsaturated substrates and is well known for oleic acid (Porter et al.
1994).
iv. Hydrogen abstraction from tertiary carbon atoms, which exist at C-20 and
C-25 in the case of cholesterol, campesterol, campestanol, β-sitosterol,
sitostanol, and brassicasterol, and C-24 in the case of campesterol,
campestanol, β-sitosterol, and sitostanol. Positions C-20 and C-24, being
both allylic to the ∆22-23 double bond and tertiary, are especially reactive
in the case of stigmasterol (Blekas and Boskou, 1989). Side chain sterol
hydroperoxides decompose faster than ring hydroperoxides to form the
more stable hydroxy and keto derivatives (Yanishlieva, 1983). Oxidation of
the sterol side chains is significant during oxidation of sterols in the solid
state and is favored by their crystal structures where the aliphatic chains are
allied to the outside and are more accessible to attaching free radicals than
the rings (Korahani et al., 1982).
v. Further oxidation of first oxidation products has also been observed. For
example, 5,6-epoxy derivatives of 7α- and 7β-OH during the oxidation of
cholesteryl acetate (Lercker et al., 1999) is similar to oleate (Lercker et al.,
1984).
vi. Finally, other reactions may operate at high temperatures and under
anhydrous conditions, that is coupled dehydration-oxidation reactions
leading to the formation of several dehydrated products of sterols (Fig. 5.6)
and oligomers/polymers of unknown structures (Soupas et al., 2005). Thus,
a wide range of minor oxidation products can be formed besides the major
ones discussed above. Some of these products have been characterized, but
many remain to be investigated since traces can be observed in different
chromatograms (Lampi et al., 2002).
Photoxidation of Sterols: Products and Mechanism

Singlet oxygen, which is formed by the reaction of triplet oxygen with photosensitiz-
ers as discussed in Chapter 1, can react rapidly with the double bonds of sterols by
the ene mechanism and lead to the formation of 5-hydroperoxysterols (5-OOH) and
d ich o le ste ryl e th e r
ch o le sta -5 ,7 -d ie n -3 β-o l
ch o le sta -3 ,5 -d ie n e HO
ch o le sta -3 ,5 -d ie n e -7 -o n e
O
Fig. 5.6. Other sterol dehydration and oxidation products.
lower amounts of 6-hydroperoxysterols (6-OOH) (Kulig and Smith, 1973; Yanish-

lieva and Marinova, 1980; Smith, 1981, 1987, 1996). There is only one possibility for
the hydroperoxide at C-5 (i.e., 5α-OOH), but two possibilities exist for hydroperox-
ides at C-6 (i.e., 6α-OOH and i.e. 6β-OOH) (Säynäjoki et al., 2003). Porter et al.
(1994) showed that they can easily form peroxyl radicals that undergo rearrangement
when they are in proximity to a double bond due to the weak ROO-H bond in hy-
droperoxide. They have further shown that the 5α-OOH formed in photoxidation
can undergo rearrangement to for 7α-OOH, which will re-arrange further to form
small amounts of 7β-OOH in agreement with other investigators (Beckwith, 1989;
Bortolomeazzi et al., 1999). Beyond identification of these products and their forma-
tion pathway, studies on the photoxidation of sterols have been quite limited.
Kinetics of Sterol Autoxidation

Oxidation of lipids starts as a pseudo-first-order reaction at peroxide values <20mM
followed by a second-order reaction (Bateman et al., 1953). The detection and de-
scription of reaction kinetics is, however, dependent inter alias on the substrate, reac-
tion temperature, reaction medium, and the monitored reaction product(s). There-
fore, contradictions are often found in literature. For example, Park and Addis (1986)
found that the formation of 7-ketocholesterol during heating follows a zero-order
reaction, Yan and White (1990) found the formation of 7-hydroxy (OH), 7-keto,
and 5,6-epoxides (EP) from cholesterol in lard fit a first-order reaction, while Chien et
al. (1998) found that the formation of 7-OOH and 5,6-EP followed a second-order
reaction during the oxidation of a thin film of cholesterol at 150°C, that is
d[CholOxid]/dt = k[CholOxid][Chol]
where k is the rate constant, [CholOxid] is the concentration of 7-OOH or 5,6-EP,

[Chol] is the concentration of unoxidized cholesterol, and t is the time.
On the other hand the formation of the secondary oxidation products, 7-OH
and 7-keto compounds, followed a first-order reaction being solely dependent on the
concentration of 7-OOH (Chien et al., 1998). In this study, about 50% of cholesterol
underwent very fast transformation within 5 min. while the next 15% disappeared
at a much slower rate between 5 and 30 min. The rate constants were approximately
1400 h-1 for the formation of 5,6α-EP and 5,6β-EP and 1600 h-1 for the formation of
7α-OOH and 7β-OOH from cholesterol, about 800 h-1 for the reduction of 7-OOH
to 7-OH, and 805 h-1 for its dehydration to keto, and only 3 h-1 for the dehydroge-
nation of 7-OH to keto. Like fatty acid oxidation, sterol oxidation is inhibited by
antioxidants, such as α-tocopherol, (Terao et al., 1985; Li et al., 1996).
Pioneering work by Yanishlieva and Marinova (1980) showed that sitosteryl stear-
ate oxidized much faster at the beginning of the process (PV 15-20 mM) compared to
β-sitosterol in the pure state in the dark at 90-125°C. This difference was related to
differences in the activation energies, calculated as 37.7 Kcal/mole for β-sitosterol and
16.3 Kcal/mole for sitosteryl stearate, but the reason for this large difference was not
explained. The differences in oxidizability between sitosteryl stearate and β-sitosterol
were largely diminished during the second phase of oxidation due to greater pro-
oxidant effect of the oxidation products of β-sitosterol than sitosteryl stearate. It was
suggested that the free hydroxyl group in β-sitosterol inhibits the initiation of oxida-
tion in the initial stage but promotes the decomposition of hydroperoxides. The reac-
tion products also differed in the two cases with oxidation products of β-sitosterol-3-
one, β-sitosterol-3,6-dione, and β-sitosterol-3,5-diene formed from β-sitosterol while
oxidation products at C-1, C-2, and C-4 formed in the case of β-sitosteryl stearate
(Yanishlieva et al., 1983). Comparable results were obtained when the oxidation was
performed in different triacylglycerol matrices.
Significant oxidative interactions may occur between sterols and co-existing lip-
ids depending on their natures and other compositional and environmental factors
(Nawar et al., 1991; Kim and Nawar, 1991, 1993). The effect of co-existing un-
saturated fatty acids in the oxidized matrix on sterol oxidation is controversial. For
example, Ohshima et al. (1993) found that cholesterol was stable in triolein but it
co-oxidized rapidly in a mixture of fish oil triacylglycerols. Free radicals generated

during the oxidation of highly oxidizable polyunsaturated fatty acids catalyze sterol
oxidation leading to a high degree of co-oxidation (Li et al., 1994). The oxidation
was found to be faster in bulk stigmasterol than when the oxidation was performed
in purified rapeseed oil triacylglycerols (Lampi et al., 2002). Sitostanol, β-sitosterol,
stigmasterol, and ergosterol were oxidized at 0.1-1.0% levels in purified rapeseed oil
and tripalmitin at 80, 120, and 180°C. As expected, sitostanol, with a saturated ring
structure, was more stable than unsaturated β-sitosterol and stigmasterol while ergos-
terol, with a conjugated ∆5,7 double bond system, was by far least stable. Stigmasterol
was stable in oil matrices at 80°C for 7 days but approximately 50% was degraded
after 1 day at 120°C. All sterols were more stable in rapeseed oil than in tripalmitin
at 180°C (Lampi et al., 2002), which can be explained by the concept of competitive
oxidation under conditions of low prevalence of oxygen. Early experiments showed
that when β-sitosterol was added at a 5% level to mixtures oxidized at 120°C, it was
less stable in purified sunflower oil triacylglycerols than lard triacylglycerols in tristear-
in (Yanishlieva and Marinova, 1986). Under these conditions, β-sitosterol oxidation
was accelerated by the increased degree of unsaturation in the lipid medium which is
critical for chain initiation and propagation. In addition, the rate of oxidation is also
dependant on the physical state, such as melting behavior, of sterols and co-existing
lipids (Kim and Nawar, 1991, 1993).
Table 5.3 presents results of stigmasterol oxidation in tripalmitin and purified
rapeseed oil at 100, 140, and 180°C (Soupas et al., 2004). The results showed that at
100°C, stigmasterol oxidation was faster in purified rapeseed oil than in tripalmitin
but at >140°C, stigmasterol oxidation was much slower in purified rapeseed oil than
in tripalmitin. It is known that oxidizing lipids compete for the low concentration
of oxygen and that other reactions, for example dehydration and polymerization, are
also important at high temperatures. Unlike the unsaturated sterols (Li et al., 1994;
Soupas et al., 2004), the oxidation of saturated sitostanol was enhanced by the un-
saturated lipid matrix at all temperatures (Soupas et al., 2004). Similarly, phytosteryl
esters oxidized faster than free phytosterols in tripalmitin at 100°C but not at 180°C
due to intermolecular interactions (Soupas et al., 2005). In an early study, the percent-
age of oxidized cholesterol at 100°C in the presence of benzoyl peroxide in the solid
phase followed the order: cholesterol linolenate < cholesterol linoleate << cholesterol
stearate < cholesterol oleate < cholesterol acetate (Korahani et al., 1982). The presence
of unsaturated linoleate and linolenate clearly protected cholesterol against oxidation.
This is not true for highly unsaturated fatty acids like fish oil triacylglycerols (Li et al.,
1994).
Depending on the composition of the matrix and the temperature of the reac-
tion, sterols may induce a weak lowering or elevation of the oxidation rate of co-exist-
ing fatty acyls. The presence of 10% cholesterol in purified sunflower triacylglycerols
was found to increase the oxidation rate at 80, 90, and 100°C (Marinova et al., 2005)
in agreement with earlier results by Wu et al. (1978) in monolayers of linoleic acid.
This prooxidant effect of sterols was explained by their participation in the formation
of mixed micelles with initial hydroperoxides leading to a small increase in their rate
of decomposition (Brimberg and Kamal-Eldin, 2003). On the other hand, at 0.1%
level ∆5-avenasterol and fucosterol were found to exert a slight protective effect on the
degradation of the linoleate residues of olive oil at 180°C (Gordon and Magos, 1983).
This effect was not possessed by cholesterol, β-sitosterol or stigmasterol suggesting
that the ∆24,28 double bond in the side chain was responsible for this effect. However,
this effect was not found in a previous study (Lampi et al., 1999).
Table 5.3. Transformation of Stigmasterol to Stigmasterol Oxides (%) During Heat Treatment in Tripalmitin
or Purified Rapeseed Oil Containing 1% Stigmasterol
Tripalmitin Purified Rapeseed Oil
Treatment
7α- and 5,6-EP 7-Keto Total 7α- and 7β- 5,6-EP 7-Keto Total
hydroxy
100°C, 6h 0.02 0.03 0.02 0.07 1.4 0.2 0.7 2.3
24h 0.03 0.04 0.03 0.10 8.8 1.9 1.7 12.4
48h 0.07 0.12 0.11 0.30 15.9 5.4 3.6 24.9
140°C, 1h 0.05 0.10 0.10 0.25 0.3 0.1 0.2 0.6
3h 0.6 0.9 1.1 2.6 1.6 0.8 0.5 2.9

6h 2.0 3.0 2.5 7.5 4.4 2.4 0.8 7.6
180°C, 1h 2.2 2.5 1.0 5.7 1.1 0.6 0.2 1.9
2h 6.9 6.5 1.7 13.1 2.7 1.4 0.3 4.4
3h 9.8 10.2 2.7 22.7 4.0 2.3 0.5 6.8
Source: Soupas et al., (2004)
Conclusion
Sterols oxidize according to the same rules that govern other hydrocarbons including
polyunsaturated fatty acids. Since they are present in complex lipid mixtures, their
oxidation is affected by the other species present, including antioxidants, prooxidants,
and co-oxidizable lipids. The physical status of the matrix, its non-lipid components,
temperature, and other factors that influence the exposure of sterols to oxygen and
their interactions affect the oxidation rate of sterols. Changes in sterol fluidity, for
example by esterification, also seem to play a considerable role.
References
Adachi, J.; M. Asano; T. Naito; Y. Ueno; and Y. Tatsuno. Chemiluminescent Determination
of Cholesterol Hydroperoxides in Human Erythrocyte Membrane. Lipids 1998, 33,
1235-1240.
Akasaka, K.; and H. Ohrui. Development of Phosphine Reagents for the High Performance
Liquid Chromatographic-Fluorometric Determination of Lipid Hydroperoxides. J. Chro-
matogr. A. 2000, 881, 159-170.

Aringer, L.; and P. Eneroth. Formation and Metabolism in vitro of 5,6-Epoxides of Cholesterol
and β-Sitosterol. J. Lipid Res. 1974, 15, 389-398.
Bateman, L; H. Hughes; and A.L. Morris. Hydroperoxide Decomposition in Relation to the
Initiation of Radical Chain Reactions. Disc. Faraday Soc. 1953 14, 190-199.
Beckwith, A.L.J. The Mechanism of the Rearrangements of Allylic Hydroperoxides:
5α-hydroperoxy-3β-hydroxycholest-6-ene and 7α-hydroperoxy-3β-hydroxycholest-5-
ene. J. Chem. Soc. Perkin Trans. 1989, II, 815-824.
Blekas, G. and D. Boskou. Oxidation of Stigmasterol in Heated Triacylglycerols. Food Chem.
1989 33, 301-310.
Bortolomeazzi, R.; M. De Zan; L.S. Pizzale; and L. Conte. Mass Spectrometry Characteriza-
tion of the 5α-, 7α-, and 7β-hydroxy Derivatives of β-Sitosterol, Campesterol, Stigmas-
terol, and Brassicasterol. J. Agric. Food Chem. 1999, 47, 3069-3074.
Bortolomeazzi, R.; F. Cordaro; L. Pizzale; L.S. Conte. Presence of Phytosterol Oxides in Crude
Vegetable Oils and Their Fate During Refining. J. Agric Food Chem. 2003, 51, 2394-
2401.
Bösinger, S.; W. Luf; and E. Brandl. Oxysterols: Their Occurrence and Biological Effects. Int.
Dairy J. 1993, 3, 1-33.
Brimberg U.; and A. Kamal-Eldin. On the Kinetics of the Autoxidation of Fats: Influence of
Prooxidants, Antioxidants and Synergists. Eur. J. Lipid Sci. Technol. 2003, 105, 83-91.
Chien, J.T.; H.C. Wang; and B.H. Chen. Kinetic Model of the Cholesterol Oxidation During
Heating. J. Agric. Food Chem. 1998, 46, 2572-2577.
Csiky, I. Trace Enrichment and Separation of Cholesterol Oxidation Products by Absorption
High-Performance Liquid Chromatography. J. Chromatography 1982, 241, 381-389.
Csiky, I. Extension of the Selectivity in Column Liquid Chromatography: Applications of Pre- and
Post-Column Techniques for the Separation of Complex Mixtures, Ph.D. Thesis, University
of Lund, Lund, 1985.
Dutta, P.C. Studies on Phytosterol Oxides. II. Content in Some Vegetable Oils and in French
Fries Prepared in These Oils. J. Am. Oil Chem. Soc. 1997, 74, 659-666.
Galobart, J.; and F. Guardiola. Formation and Content of Cholesterol Oxidation Products in
Egg and Egg Products. In Cholesterol and Phytosterol Oxidation Products: Analysis, Occur-
rence, and Biological Effects, Guardiola, F.; Dutta, P.C.; Codony, R.; and Savage, G.P. Eds.;
AOCS Press: Champaign, Illinois, 2002, pp. 124-146.
Gordon, M.H.; and P. Magos. The Effect of Sterols on the Oxidation of Edible Oils. Food
Chem. 1983, 10, 141-147.
Grandgirard, A.; L. Martine; P. Joffre; P. Juaneda; and O. Berdeaux. Gas Chromatographic
Separation and Mass Spectrometric Identification of Mixtures of Oxyphytosterol and
Oxycholesterol Derivatives: Application to a Phytosterol Enriched Food. J. Chromatogr A
2004, 1040, 239-250..
Huber, K.C.; O.A. Pike; and C.S. Huber. Antioxidant Inhibition of Cholesterol Oxidation in
Spray-Dried Food System During Accelerated Storage. J. Food Sci. 1995, 60, 909-916.
Kerry, J.P.; D.A. Gilroy; and N.M. O’Brien. Formation and Content of Cholesterol Oxidation
Products in Meat and Meat Products. In Cholesterol and Phytosterol Oxidation Products:
Analysis, Occurrence, and Biological Effects, Guardiola, F.; Dutta, P.C.; Codony, R.; and
Savage, G.P. Eds; AOCS Press: Champaign, Illinois, 2002, pp. 162-185.
Kim, S.K.; and W.W. Nawar. Oxidative Interactions of Cholesterol with Triacylglycerols. J.
Am. Oil Chem Soc. 1991, 68, 931-934.
Kim, S.K.; and W.W. Nawar. Parameters Influencing Cholesterol Oxidation. Lipids 1993, 28,
917-922.
Korahani, V.; J. Bascoul; and A. Crastes de Paulet. Autoxidation of Cholesterol Fatty Acid
Esters in Solid State and Aqueous Dispersion. Lipids 1982, 17, 703-708.
Korytowski W.; P.G. Geiger; A.W. Girotti. High-Performance Liquid-Chromatography with
Mercury Cathode Electrochemical Detection: Application to Lipid Hydroperoxide Anal-
ysis. J. Chromatogr. 1995, 670, 189-197.
Kulig, M.J. and L.L. Smith. Sterol Metabolism-XXV. Cholesterol Oxidation by Singlet Mole-
cular Oxygen. J. Org. Chem. 1973 38, 3639-3642.
Lambelet, P.; A. Grandgirard; S. Gregoire; P. Juaneda; J.L. Sebedio; and C. Bertoli. Formation
of Modified Fatty Acids and Oxyphytosterols During Refining of Low Erucic Acid Rape-
seed Oil. J. Agric. Food Chem. 2003, 51, 4284-4290.
Lampi A-M.; L. Dimberg; and A. Kamal-Eldin. A Study on the Influence of Fucosterol on
Thermal Polymerisation of Purified High Oleic Sunflower Triacylglycerols. J. Sci. Food
Agric. 1998, 79: 573-579.
Lampi, A.-M.; L. Juntunen; J. Roivo; and V. Piironen. Determination of Thermo-Oxidation
Products of Plant Sterols. J. Chromatogr. B. 2002, 777, 83-92.
Law, M. Plant Sterol and Stanol Margarines and Health. Br. Med. J. 2000, 320, 861-864.
Lea, L.J.; P.A. Hepburn; A.M. Wolfreys; P. Baldrick. Safety Evaluation of Phytosterol Esters.
Part 8: Lack of Genotoxicity and Subcronic Toxicity with Phytosterol Oxides. Food Chem
Toxicol. 2004, 42, 771-783.
Lercker, G.; P. Capella; and L.S. Conte. Thermo-Oxidative Degradation Products of Methyl
Oleate. Riv. Ital. Sost. Grasse 1984, 61, 337-344.
Lercker, G.; R. Bortolmeazzi; and L. Pizzale. Formation of 5,6-Epoxy Derivatives of 7-Hy-
droxy-Cholesteryl-3-Acetates During Peroxidation of Cholesteryl Acetate. Grasas y Aceites
1999, 50, 193-198.
Li, N.; T. Ohshima; K. Shozen; H. Ushio; and C. Koizumi. Effects of the Degree of Unsatura-
tion of Coexisting Triacylglycerols on Cholesterol Oxidation. J. Am. Oil Chem. Soc. 1994,
71, 623-627.
Li, S.X.; G. Cherian; D.U. Ahn; R.T. Hardin; and J.S. Sim. Storage, Heating, and Tocoph-
erols Affect Cholesterol Oxide Formation in Food Oils. J. Agric. Food Chem. 1996, 44,
3830-3634.
Maerker, A. Cholesterol Autoxidation: Current Status. J. Am. Oil Chem. Soc. 1987, 64, 388-
392.
Marinova, E.; N. Yanishlieva; and A. Toneva. Influence of Cholesterol on the Kinetics of Lipid
Autoxidation and on the Antioxidative Properties of α--Tocopherol and Quercetin. Eur.
J. Lipid Sci. Technol. 2005, 107, 418-425.
Muto, T.; J. Tanaka; T. Miura; and M. Kimura. Iron-Catalyzed Autoxidation of Cholesterol
in the Presence of Unsaturated Long Chain Fatty Acid. Chem. Pharm. Bull. 1982, 30,
3172-3177.
Nawar, W.W.; S.K. Kim; Y.J. Li; and M. Vadji. Measurement of Oxidative Interactions of
Cholesterol. J. Am. Oil Chem. Soc. 1991, 68, 496-498.
Oehrl, L.L.; A.P. Hansed; C.A. Rohrer; G.P. Fenner; and L.C. Boyd. Oxidation of Phytosterols
in a Test Food System. J. Am. Oil Chem. Soc. 2001, 78, 1073-1078.
Ohshima, T.; N. Li; and C. Koizumi. Oxidative Decomposition of Cholesterol in Fish Prod-
ucts. J. Am. Oil Chem. Soc. 1993, 70, 595-600.
Osada, K.; T. Kodama; K. Yamada; and M. Sugano. Oxidation of Cholesterol by Heating. J.
Agric. Food Chem. 1993a, 41, 1198-1202.

Osada, K.; T. Kodama; L. Cui; K. Yamada; and M. Sugano. Levels and Formation of Oxidized
Cholesterol in Processed Marine Food. J. Agric. Food Chem. 1993b, 41, 1893-1898.
Paniangvait, P.; A.J. King; A.D. Jones; and B.G. German. Cholesterol Oxides in Foods of
Animal Origin. J. Food Sci. 1995, 60, 1159-1174.
Park, S.W. and P.B. Addis. Identification and Quantitative Estimation of Oxidized Cholesterol
Derivatives in Heated Tallow. J. Agric. Food Chem. 1986 34, 653-659.
Porter, N.A.; K.A. Mills; S.E. Cadwell; and G.R. Dubay. The Mechanism of the [3,2] Allylper-
oxyl Rearrangement: A Radical-Dioxygen Pair Reaction that Proceeds with Stereochemi-
cal Memory. J. Am. Chem. Soc. 1994, 116, 6697-6705.
Rankin, S.A.; and O.A. Pike. Cholesterol Autoxidation Varies Among Several Natural Antioxi-
dants in an Aqueous Model System. J. Food Sci. 1993, 58, 653-655.
Russell, G.A. Deuterium-Isotope Effects in the Autoxidation of Alkyl Hydrocarbons: Mecha-
nism of the Interaction of Peroxyl Radicals. J. Am. Chem. Soc. 1957, 79, 3871-3877.
Sander, B.D.; P.B. Addis; S.W. Park; and D.E. Smith. Quantification of Cholesterol Oxidation
Products in a Variety of Foods. J. Food Prot. 1989, 52, 109-114.
Säynäjoki, S.; S. Sundberg; L. Soupas; A.-M. Lampi; and V. Piironen. Determination of Stig-
masterol Primary Oxidation Products by High-Performance Liquid Chromatography.
Food Chem. 2002, 80, 415-421.
Säynäjoki, S.; S. Sundberg; L. Soupas; A-M. Lampi; and V. Piironen. Determination of Stig-
masterol Primary Oxidation Products by High Performance Liquid Chromatography.
Food Chem. 2003, 80, 415-421.
Sevilla, C.L.; D. Becker; and M.D. Sevilla. An Electron Spin Resonance Investigation of Radi-
cal Intermediates in Cholesterol and Related Compounds: Relation to Solid-State Autoxi-
dation. J. Phys. Chem. 1986, 90, 2963-2968.
Smith, L.L. Cholesterol Autoxidation, Plenum Press: New York, 1981.
Smith, L.L. Cholesterol Autoxidation 1981-1986. Chem. Phys. Lipids 1987, 44, 87-125.
Smith, L.L. Review of Progress in Sterol Oxidations 1987-1995. Lipids 1996, 31, 453-487.
Soupas, L.; L. Juntunen; A.-M. Lampi; K.-H. Liukkonen; K. Katina; K.-M. Oksman-Calden-
tey; and V. Piironen. Oxidative Stability of Phytosterols in Bread Baking. In Strategies for
Safe Foods: Analytical, Industrial and Legal Aspects: Challenges in Organization and Com-
munication, Proceedings of the Euro Food Chem XII Conference, Koninklijke Vlaamse
Chemische Vereniging, Heverlee Belgium, 2003, pp. 317-320.
Soupas, L.; L. Juntunen; A.-M. Lampi; and V. Piironen. Effects of Sterol Structure, Tem-
perature and Lipid Medium on Phytosterol Oxidation. J. Agric. Food Chem. 2004, 52,
6485-6491.
Soupas, L.; L. Huikko; A.-M. Lampi; and V. Piironen. Esterification Affects Phytosterol Oxi-
dation. Eur. J. Lipid Sci. Technol. 2005, 107, 107-118.
Soupas, L.; L. Huikko; A.-M. Lampi; and V. Piironen. Oxidative Stability of Phytosterols in
Some Food Applications. Eur. Food Res. Technol. 2006, 222, 266-273.
Soupas, L.; L. Huikko; A.-M. Lampi; and V. Piironen. Pan Frying May Induce Phytosterol
Oxidation. Food Chem. 2007 101, 286-297.
Stanton, C.; and R. Devery. Formation and Content of Cholesterol Oxidation Products in
Milk and Dairy Products. In Cholesterol and Phytosterol Oxidation Products: Analysis, Oc-
currence, and Biological Effects, Guardiola, F.; Dutta, P.C.; Codony, R.; and Savage, G.P.
Eds.; AOCS Press: Champaign, Illinois, 2002, pp. 147-161.
Terao, J.; K. Sugano; and S. Matsushita. Fe2+ and Ascorbic Acid Induced Oxidation of Choles-
terol in Phosphatidylcholine Liposomes and Its Inhibition by α-Tocopherol, J. Nutr. Sci.

Vitaminol. 1985, 31, 499-508.
Tsai, L.S.; and C.A. Hudson. Cholesterol Oxides in Commercial Dry Egg Products: Isolation
and Identification. J. Food Sci. 1984, 49, 1245-1248.
Vardanyan, R.L.; R.L. Safiullin; and V.D. Komissarov. Absolute Values of Rate Constants of
Disproportionation of Peroxyl Radicals Formed from Cholesterol. Kinetic Katalysis 1985,
26, 1140-1144.
Wu, G.-S.; R.A. Stein; and J.F. Mead. Autoxidation of Fatty Acid Monolayers Adsorbed on Sil-
ica Gel. III. Effects of Saturated Fatty Acids and Cholesterol. Lipids 1978, 13, 517-523.
Yan, P.S. and P.J. White. Cholesterol Oxidation in Heated Lard Enriched with Two Levels of
Cholesterol. J. Am. Oil Chem. Soc. 1990 67, 927-931.
Yanishlieva, N.; and E. Marinova. Autoxidation of Sitosterol. I. Kinetic Studies on Free and
Esterified Sitosterol. Rivista Ital. Del Sostanzr Grasse 1980, LVII, 477-480.
Yanishlieva, N.; and E. Marinova. Effect of the Unsaturation of Lipid Media on the Autoxida-
tion of Sitosterol. Grasa y Aceites 1986, 37, 343-347.
Yanishlieva, N.; E. Marinova; H. Schiller; and A. Seher. Comparison of Sitosterol Autoxida-
tion in Free Form, as Fatty Acid Ester, and in Triacylglycerol Solution. Kinetics of the
Process and Structure of the Products Formed. Proc. 16th ISF Congress, Fat Science,
Budapest 1983, pp. 619-626.
6
Tocopherol Concentrations and
Antioxidant Efficacy
Afaf Kamal-Eldina, Hyun Jung Kimb, Levon Tavadyanc, and David B. Minb
a
Department of Food Science, Swedish University of Agricultural Sciences, P.O. Box
7051, 75007, Uppsala, Sweden; bDepartment of Food Science and Technology, The
Ohio State University, 2015 Fyffe Road, Columbus, OH 43210; cA. B. Nalbandyan
Institute of Chemical Physics of the National Academy of Sciences of Armenia, 5/2
Sevak str. 0014 Yerevan, Armenia
Introduction
Lipid oxidation is one of the important reactions contributing to aging in foods and
biological systems, including the human body. The reaction of lipids with oxygen is
thermodynamically driven but its speed is stimulated by a number of agents including
temperature and other pro-oxidative catalysts (mainly trace metal ions), inhibited by
antioxidants and antioxidant synergists, and variably modulated by the other chemi-
cal species in the reaction environment. The reaction between free radicals and lipids
contributes to the deterioration of flavor and the presence of potential toxic products in
food systems, and to membrane fragility and dysfunction in biological tissues.
Tocopherols are the major lipid-soluble antioxidants in nature (Kamal-Eldin and
Appelqvist 1996; Choe and Min 2006). The inhibitory activity of tocopherols, particu-
larly α-tocopherol, against lipid oxidation is usually associated with an optimal con-
centration of antioxidants after which the stabilization is lowered (Cillard et al., 1980;
Blekas et al., 1945; Burlakova et al., 1998). This phenomenon was observed in triacyl-
glycerols purified from vegetable oils, for example corn oil (Huang et al., 1995), butter
oil (Lampi and Piironen. 1998), sunflower oil (Fuster et al., 1998), rapeseed oil (Lampi
et al., 1999), olive oil (Deiana et al., 2002), and soybean oil (Yanishlieva et al., 2002;
Kim et al., 2007). Under these conditions, the transition from the induction period to
the exponential oxidation phase is not always a consequence of complete tocopherol
consumption or the ratio of tocopherols to the amount of hydroperoxides present as
suggested by Witting (1969). This effect was considered a “pro-oxidant effect,” but this
qualification is not correct considering the great stability of tocopherol-supplemented
samples compared to controls void of the antioxidant. Loss of antioxidant efficacy at
post-optimal concentrations was suggested to describe this phenomenon (Fuster et al.,
1998).
The same phenomenon was observed in the oxidation of low-density lipoprotein
particles (Bowry et al., 1992; Bowry and Stocker, 1993; Bowry and Ingold, 1999) and
it was proposed that the α-tocopheroxyl radical, which is formed by the one-electron
oxidation of α-tocopherol, is responsible for the observed “pro-oxidant” effect. How-
ever, the fact that other tocopherols are less pro-oxidative than α-tocopherol suggests
127
128 A. Kamal-Eldin, et al.
that this is not the only mechanism involved. Nevertheless, it is generally agreed that
tocopherol concentration is important whether it can be an antioxidant or a “pro-
oxidant.” It could be assumed that more intermediate radicals are formed during
the oxidation and storage of lipids containing tocopherols at concentrations higher
than certain optima. These intermediate radicals, including the tocopheroxyl radicals,
can initiate the processes of lipid oxidation. In this chapter, we review the different
reactions in which the tocopherol molecules and radicals are active and the possible
anti- and pro-oxidant contributions of these reactions to the overall lipid oxidation
reaction.
The Antioxidant Mechanisms of α-Tocopherol

Lipid oxidation is a typical free-radical chain reaction of unsaturated fatty acids. The
high reactivity of bis-allylic methylene groups in unsaturated fatty acids makes these
molecules the primary targets for free radical reactions. The chain reaction includes
initiation, propagation, degenerate branching, and termination steps.
LH + X• → L• + XH (initiation) [1]
L• + O2 → LOO• (propagation) [2]
LOO• + LH → L• + LOOH [3]
LOOH + LH → LOO• + L• + Η2Ο (degenerate branching) [4]
LOO• + LOO• → non-radical products (termination) [5]
LOO• + L• → non-radical products [6]
L• + L• → non-radical products [7]
Where LH represents a polyunsaturated fatty acid moiety, such as linoleate or lino-

lenate; X• is an initiating free radical; LOO• is a lipid peroxyl radical; L• is a C-cen-
tered lipid radical, LOOH is a lipid hydroperoxide molecule. Other free radicals and
reactions are also involved in this extremely complex reaction (Tables 6.1 and 6.2).
The propagation step leads to the formation of hydroperoxides and is responsible
for the autoinitiation nature of the reaction. Tocopherols protect polyunsaturated
fatty acids from oxidative degradation by donating their hydrogen to chain carrier
lipid peroxyl radicals and preventing further chain radical reaction. Tocopherols (TH)
can transfer a hydrogen atom from their 6-phenoxyl group to lipid peroxyl radicals.
Tocopherols, with oxidation potentials of 275-400 mV, easily donate their phenolic
hydrogen to peroxyl radical (LOO•) with a reduction potential of 1000 mV and pro-
duce lipid hydroperoxide (LOOH) and tocopheroxyl radical (T•). At the same time
the reaction exothermicity promotes its flow (Denisov and Denisova, 2000).
DH°=BDE(T–H)–BDE(LOO–H)=331–372= –41kJ/mol
LOO• + TH → T• + LOOH [8]

Tocopherol Concentrations and Antioxidant Efficacy 129
Table 6.1. Standard Reduction Potentials for Free Radicals of Importance in the Antioxidant
Effect of Tocopherolsa
Compounds Half-Cell Standard Reduction Potential (mV)
•OH H+/H2O 2310
LO• b
H /ROH
+
1600
LOO•b H+/ROOH 1000
L• H /RH
+
600
α-Tocopheroxyl• H+/α-Tocopherol 273
β-Tocopheroxyl• H+/β-Tocopherol 343
γ-Tocopheroxyl• H+/γ-Tocopherol 348
δ-Tocopheroxyl• H /δ-Tocopherol
+
405
a
Adapted from Choe and Min (2005) and Kamal-Eldin and Appelqvist (1996).
b
LO• and LOO• forms for tocopherol are tocopheroxyl oxy radical and tocopheroxyl peroxy radical,
respectively, and their structures are shown in Figures 6.1-6.4.
Tocopheroxyl radicals (T•) are resonance-stabilized structures that are more stable
than lipid peroxyl radicals (LOO•) and can further scavenge another peroxyl radical
by addition
T• + LOO• → non-radical products [9]
Or alternatively form non-radical products mainly by disproportionation or by com-

bination
T• + T• → non-radical products [10]
The rate constants of the reaction of α-tocopherol with lipid peroxyl radical (Eq. 8) is
106 to 107 M-1 sec-1 (Niki et al., 1984; Denisov and Denisova, 2000; Choe and Min,
2005) and is 104 to 105 times higher than that of unsaturated lipid with lipid peroxyl
radical (Niki et al. 1984; Denisov and Denisova, 2000; Naumov and Vasil’ev, 2003).
The antioxidant activity of tocopherols was found to depend on temperature, pH, the
degree and number of unsaturated fatty acids, the availability of oxygen and transition
metal ions (Roginsky, 1990; Verleyen, et al., 2002; Kamal-Eldin et al., 2002; Reische
et al., 2002).
The Pro-Oxidant Mechanisms of α-Tocopherol

To understand the nature of the pro-oxidative side reactions affecting the antioxidant
potency of α-tocopherol, one needs to examine the effect of the initial α-tocopherol
concentration on the length of the induction period as well as on the rate of peroxi-
dation during the induction period. Yanishlieva and co-workers used two terms to
Table 6.2. Rates Constants of Important Autoxidation Reactions Involving Unsaturated Fatty Acids (LH),
α-Tocopherol (TH), and Their Oxidation Products
N Reaction Steps Reaction Rate Constanta
1 LH + O2 → L• + HOO• 2.24 x 10-10
2 L• + O2 → LOO• 8.75 x 108
3 LOO• + LH → L• + LOOH 76.3
4 LO• + LH → L• + LOH 1.26 x 107
5 HOO• + LH → L• + H2O2 228
6 LOO• → NRP + HOO• 5.49 x 10-2
7 LOO• + LOO• → L(-H)=O + LOH + O2 2.01 x 107
8 LOO• + HOO• → L(-H)=O + H2O + O2 108
9 LOOH → LO• + •OH 1.23 x 10-8
10 LOO• + TH → T• + LOOH 1.85 x 106
11 HOO• + TH → T• + H2O2 5.55 x 106
12 T• + LH → TH + L• 3.97 x 10-2
13 T• + LOOH → TH + LOO• 10.6
14 T• + H2O2→ TH + HOO• 10.6
15 T• + O2→ TMQ + HOO• 1.16 x 10-2
16 T• + T• → TMQ + TH 1.28 x 103
17 T• + HOO• → TH + O2 0.75 x 108
18 LOO• + TMQ → NRP 8.35 x 102
19 LOOH + TH → LO• + T• + H2O 6.6 x 10-6
20 HOOH + TH → HO• + T• + H2O 6.6 x 10-6
21 •OH + LH → L• + H2O 1010
22 T• + LOO• → o-T-OOL 1.23 x 108
23 T• + LOO• → p-T-OOL 0.27 x 108
24 o-T-OOL → LO• + o-TE• 5.54 x 10-8
25 o-TE• + LOO• → o-TE-OOL 1.5 x 108
26 o-TE-OOL → o-TE-O• + LO• 6.34 x 10-10
27 o-TE-O• + LH → o-TEQ + L• 1.26 x 107
28 p-T-OOL → LO• + TQ-O• 6.34 x 10-10
29 TQ-O• + LH → TQ-OH + L• 1.26 x 107
30 T• + HOO• → o-T-OOH 2.7 x 108
31 T• + HOO• → p-T-OOH 0.6 x 108
The rate constants of the reactions are given in М and sec at 40°C. The designation of some reaction
a
species corresponds to those shown in Fig. 6.1. Source: Tavadyan et al., 2007
Fig. 6.1. Chemical structures of α-, β-,γ-, δ- tocopherols and the main molecular products of α- tocopherol
oxidation according the scheme presented in Table 6.2.
describe the effects of antioxidant inhibitors on lipid peroxidation
1. Effectiveness, representing the ability of the antioxidant to block the radical

chain reactions by interaction with peroxyl radicals, which is responsible for
the duration of the induction period, and
2. Strength, expressing the ability of the inhibitor moieties to participate in side

reactions, which may lead to a change in oxidation rate during the induction
period and the magnitude of the reaction induction period (Yanishlieva and
Marinova, 1992; Kasaikina et al., 1999; Yanishlieva et al., 2002).
The Yanishlieva-Marinova model (1992) uses the stabilization factor (F), which is
the ratio between the induction period in the presence of inhibitor and its absence to
measure effectiveness and uses the oxidation rate ratio (ORR), the ratio of the rate of
lipid oxidation reactions in the presence of inhibitor and its absence, as a measure of
the strength. A combining factor, antioxidant activity (A), is calculated as the ratio F/
ORR. If F is not a linear function of the concentration of the antioxidant, it indicates
that the antioxidant molecule participates in reactions other than the main reaction
of chain interception. When the rate of inhibited reaction is equal to the recipro-
cal of the initial concentration of antioxidant, then the antioxidant radical does not
participate in side reactions. Using these principles, it was found that the molecule of
α-tocopherol readily participates in one side reaction with hydroperoxides leading to
branching to produce two radicals, while the tocopheroxyl radical(s) seemed to par-
ticipate in a number of pro-oxidant side reactions (Yanishlieva and Marinova, 1992;
Yanishlieva et al., 2002).
Results of the kinetic numerical analysis (Tavadyan et al., 2007) of the reaction
mechanisms of the methyl linoleate oxidation inhibited by α-tocopherol (Table 6.2)
showed that the length of the induction period initially rises with increased initial
tocopherol concentration followed by an arena of practical independence and ends
with a decrease. In accordance with this, the rate of peroxidation during the induc-
tion period starts with a fast drop, remains stable up to about 20 mM α-tocopherol
in methyl linoleate and then starts to increase proportionally to the initial tocopherol
concentration. Such complex dependencies of both the length of the induction pe-
riod and the rate of the inhibited oxidation on the initial tocopherol concentration
are connected to the net product of both antioxidant and pro-oxidant activities of
α-tocopherol (Kamal-Eldin and Appelqvist 1996; Mukai et al. 2005).
The kinetic analysis by Hamiltonian systematization method (Tavadyan and
Martoyan, 2005) of the pro-oxidant mechanism of α-tocopherol in methyl linoleate
(Tavadyan et al., 2007) revealed that three types of side reaction might significantly
contribute to the manifestation of pro-oxidant effects of α-tocopherol,
1. Chain transfer reactions by the tocopheroxyl radical leading to the

abstraction of a hydrogen atom from bis-allylic methylene groups of fatty
acid molecules and/or from hydroperoxide,
2. An autoinitiation reaction between α-tocopherol with hydroperoxide

leading to alkoxyl radical formation, and
3. Reactions of the oxidation products of tocopherol
The Chain Transfer Reactions by α-Tocopheroxyl Radical

The “pro-oxidant effect” of tocopherols in bulk lipid solutions was first attributed to
its tocopheroxyl radical (T•). When this species is present at high concentration, a
number of undesirable side reactions that can initiate or enhance the rate of lipid oxi-
dation are possible. It was first proposed by Mukai et al. (1993) that T• might abstract
a hydrogen atom from the bis-allylic position of an unsaturated fatty acid moiety or
the peroxide hydrogen from lipid hydroperoxides,
T• + LH → TH + L• [11]
T• + LOOH → TH + LOO• [12]
These chain transfer steps were shown by kinetic modeling to have the most essential
pro-oxidant activity during the induction period when the formation of peroxyl radi-
cals is minimized by the presence of α-tocopherol. Thus, the ratio [T•]/[LOO•] in-
creases with increasing initial tocopherol concentration leading to the increase in the
A B C
·O 6
5
4a
4
3
O
· O
7
8 8a
1 2
O O
· O
ROO · ROO ·
O OOR O
O
· O O
O
ROO · ROO
O
− RO · − RO ·
O
O · O
OO
O O O
R
·O
α-tocopherol peroxide
O
O O
− RO ·
O
· O · O
O
O O
ROO · ROO ·
· O
O O
O O O
O O O
OR OR
O
O
· − RO · − RO ·
RH O
O O
O
O O
O O
O
OH · O
·
α-tocopherolquinone O
O O
O O O
O
·
O
·
RH RH
O
O O
O
O O OH
OH
4a, 5-epoxy-α-tocopherolquinone 7, 8-epoxy-α-tocopherolquinone
Fig. 6.2. Possible formation of α-tocopherolquinone (a), 4a,5-epoxy-α-tocopherolquinone (b), and

7,8-epoxy-α-tocopherolquinone (c) from α-tocopheroxyl radical with lipid peroxyl radical.
reaction rate of chain transfer compared with the rate of the main reaction of chain
termination (eq 9). This leads to the non-linear dependence of the length of induction
period on tocopherol concentration. Under these conditions, oxidation proceeds with
T• acting as the main chain-carrier (Tavadyan et al., 2007). This phenomenon was
named tocopherol-mediated peroxidation (TMP) and was used to describe the loss
of antioxidant activity of α-tocopherol in the oxidation of low-density lipoprotein
(Bowry et al., 1992; Bowry and Stocker, 1993; Bowry and Ingold, 1999).
Autoinitiation Due to the Decomposition of Hydroperoxides by α-Tocopherol

One of the main reasons for the complex non-linear dependence of the length of
the induction period and the rate of oxidation during the induction period on the
initial concentration of α-tocopherol is the autoinitiation reaction involving the mol-
ecules of α-tocopherol and hydroperoxides (Denisov and Denisova, 2000; Denisov
and Azatyan, 2000)
TH + LOOH → T• + LO• + H2O

[13]
This reaction is as important for the manifestation of loss of α-tocopherol antioxidant

efficacy as the TMP reaction discussed previously. This is the reaction that also plays
a substantial role in the nonlinear behavior and the extreme dependence of both the
induction period and the rate of inhibited reaction on the initial α-tocopherol con-
centration. The rates of participation of different tocopherols (α-, β-, γ-, and δ-) in
this reaction are inversely related to the bond dissociation enthalpies of their phenolic
hydroxyl groups.
Pro-Oxidant Effects Due to the Oxidation Products of Tocopherol

Tocopherols are degraded in the presence of oxygen and free radicals and produce oxi-
dized products resulting in the loss of antioxidant activity (Jung and Min, 1990; Evans
et al., 2002). The oxidized tocopherol products, such as peroxide compounds, act as
pro-oxidants 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 more oxi-
dized tocopherols in lipids, the more oxidation of lipids occurs (Choe and Min, 2006).
The oxidation of tocopherols could be induced by strong oxidizing agents, such as
chromic acid, nitric acid, and ferric chloride, to 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 in
beef and bovine muscle microsomes (Faustman et al., 1999; Liebler et al., 1996), in
triolein (Verleyen et al., 2002), triolein and tripalmitin mixtures (Verleyen et al., 2002),
and in fish muscle (Pazos et al., 2005). During the formation of α-tocopherolquinone,
α-tocopherolhydroquinone, 4a,5-epoxy-α-tocopherolquinone, and 7,8-epoxy-α-
tocopherolquinone from α-tocopherol oxidation, many intermediate radical species

could be produced. Rietjens et al. (2002) suggested that the increased levels of oxi-
dized α-tocopherol could result in increased levels of α-tocopheroxyl radicals, which
can initiate lipid oxidation by themselves.
The formation of α-tocopherolquinone, α-tocopherolhydroquinone, epoxy-
α-tocopherolquinone, and α-tocopherol hydroperoxide during the oxidation of
α-tocopherol is known (Faustman et al.. 1999; Verleyen et al.. 2002a,b; Pazos et
al., 2005). The mechanisms for the formation of these oxidation products are shown
in Fig. 6.2. The formations of α-tocopherol peroxide, α-tocopherolquinone, 4a,5-
epoxy-α-tocopherolquinone, and 7,8-epoxy-α-tocopherolquinone as a result of the
combination of α-tocopheroxy radical and lipid peroxy radical (ROO•) are also
shown in Fig. 6.2. Figure 6.2a shows the formation of α-tocopherolquinone from
α-tocopheroxyl radical. The α-tocopheroxyl radical rearranges by resonance to form
the 8a-carbon centered α-tocopheroxyl radical. This radical reacts with lipid peroxy
radical (ROO•) to form α-tocopherol lipid peroxide at the rate constant of 1.5 × 108
(Table 6.2). The α-tocopherol lipid peroxide can produce α-tocopherol oxy radical
at the 8a carbon and an alkoxyl radical (RO•) by cleavage of the peroxide bond. The
α-tocopherol oxy radical, at the 8a position, forms α-tocopherolquinone oxy radi-
cal by cleavage of the bond between 1 and 8a position leading to the formation of
α-tocopherolquinone (Fig. 6.2a).
Figure 6.2b 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 peroxyl radical (ROO•) to form α-tocopherol
peroxide. The breakdown of α-tocopherol peroxide forms α-tocopherol oxy radical
at the carbon 5 and alkoxyl 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 peroxyl 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 alkoxyl 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 (Fig. 6.2b). Similar to this mechanism, Fig. 6.2c
shows the formation of 7,8-epoxy-α-tocopherolquinone from the carbon-centered
α-tocopheryl radical at the carbon 7. The tocopheroxyl radical might re-arrange to
give a carbon-centered radical (Fig. 6.3) but this transformation, which leads to the
formation of tocopherol hydroperoxide, is insignificant and was shown to lack effect
on the activity of α-tocopherol (Doba et al., 1984).
Figure 6.4 shows that singlet oxygen might also be formed from the carbon-
centered α-tocopheroxyl radicals if they add oxygen and form α-tocopherol peroxy
radicals. Two molecules of α-tocopherol peroxy radical form dimerized α-tocopherol
peroxide and singlet oxygen at the rate constant of 105 M-1sec-1 (Barclay et al., 1989;
Min and Boff, 2002). Singlet oxygen is a very reactive oxygen species and a strong
pro-oxidant (Bradley and Min, 1992; Min and Boff, 2002). Its reactions with lipids
O
Carbon centered
· O α-tocopheryl radical
O · Carbon centered
α-tocopherolquinone radical
O2
O
α-tocopherolquinone peroxy radical
O
O
O
·
RH
O
R· + α-tocopherolquinone hydroperoxide
O
O
O
H
Fig. 6.3. Possible formation of α-tocopherolquinone hydroperoxide from the combination of α-tocopheryl
radical with oxygen.
are very fast and occur even at very low temperature (Min and Boff, 2002). It was
recently shown that when α-tocopherol is excited, it can sensitize the formation of
singlet oxygen (Dad et al., 2006).
The tocopheroloxyl radical, α-tocopherolquinone oxy radical, α-tocopherolquinone
peroxyl radical, alkoxyl radical (RO•), hydroxyl radical (•OH), and singlet oxygen
(1O2 (1Dg)) formed from the oxidation of tocopherol can all act as pro-oxidants (Table
6.1). Oxidized α-tocopherol compounds have polar hydroxyl and nonpolar hydro-
carbon groups. Yoon et al. (1988) and Mistry and Min (1988a, 1988b) reported that
thermally oxidized lipid compounds with polar hydroxyl and nonpolar hydrocarbons
in the same molecule were pro-oxidants 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 the 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 et al.,
1988; Mistry and Min, 1988a,b). The oxidized α-tocopherol compounds with the
polar groups and nonpolar hydrocarbons in the same molecule reduce the surface
O
Carbon centered
· O
α-tocopheryl radical
O2
O α-tocopherol peroxy radical

· 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
Fig. 6.4. Possible formation of singlet oxygen and a dimeric α-tocopherol peroxide, which dissociates to form
α-tocopheryl alkoxy radical. The formation of singlet oxygen is determined by the rule of spin conservation.
tension between headspace air and oil and accelerate the oxidation of oil.
The pro-oxidant mechanisms of oxidized α-tocopherol may be mainly due to
α-tocopherol peroxyl radical, α-tocopherol oxy radical, α-tocopherolquinone oxy
radical, and hydroxyl radical which have high reduction potential and singlet oxy-
gen formed from the oxidation of tocopherols during storage in foods. The oxidized
α-tocopherol compounds with polar hydroxyl and nonpolar hydrocarbons in the
same molecule may contribute to the oxidation of oil by reducing the surface ten-
sion of the oil and increasing the diffusion of oxygen from air to the oil (Denisov and
Khudyakov, 1987).
Moreover, it is possible that two tocopherol peroxyl radicals would combine to
form a tetraoxy intermediate that would dissociate by the Russel mechanism to yield a
molecule of singlet oxygen (Fig. 6.4). In fact, singlet oxygen was detected by 1270 nm
luminescence after pulsed laser excitation (308 nm) of vitamin E and an its analog,
2,2,5,7,8-pentamethyl-6-hydroxy-chroman (PMHC) (Dad et al., 2006).
Effect of Temperature on Tocopherol Antioxidant Activity

The effect of temperature on the antioxidant activity of α- and δ-tocopherol was com-
pared in pork lard using the Oxipres apparatus in the temperature range from 80 to
150°C (Réblová, 2006). The antioxidant activity of both tocopherols decreased with
increasing temperature although at different rates. At 100 ppm, δ-tocopherol had
about double the antioxidant activity of α-tocopherol at 80°C but its activity was the
same at 130°C and none of the tocopherols was effective at 150°C. It was also found
that γ-tocopherol is a more efficient antioxidant than α-tocopherol at 180°C in puri-
fied triacylglycerols from high oleic sunflower oil (Lampi and Kamal-Eldin, 1998).
Heating at high temperatures causes rapid tocopherol degradation. Interestingly, it
was previously shown that the rate of tocopherol degradation in triacylglycerols puri-
fied from nine vegetable oils at higher temperatures (180 and 240°C) decreased as
their degree of unsaturation increased (coconut, tallow, palm, olive, high oleic sun-
flower, maize, sunflower, soybean, and flaxseed oil) (Verleyen et al., 2002c). Even if
the overall efficacy of tocopherols decrease at high temperatures, the negative effect
of high tocopherol concentrations becomes insignificant when the rate of initiation
(Ri) is high. Only at low Ri, does the contribution of side reactions leading to loss
of antioxidant efficacy become most important. This is especially true for the reac-
tion between tocopherols and hydroperoxides (eq 13). Other reactions, for example,
polymerization, become important with high-temperature heating. Mass balance
(Verleyen et al., 2002b). Future research needs to identify the different oxidation/
polymerization products of tocopherols at different temperatures.
References
Barclay, L.R.C.; K.A. Baskin; S.J. Locke; and M.R. Vinquist. Absolute Rate Constants for
Lipid Peroxidation and Inhibition in Model Biomembranes. Can. J. Chem. 1989, 68,
2258-2269.
Blekas G.; M. Tsimidou; and D. Boskou. Contribution of α-Tocopherol to Olive Oil Stability.
Food Chem. 1995, 52, 289-294.
Bowry, V.W.; and K.U. Ingold. The Unexpected Role of Vitamin E (α-Tocopherol) in the
Peroxidation of Human Low–Density Lipoprotein. Acc. Chem. Res. 1999, 32, 27–34.
Bowry, V.W.; and R. Stocker. Tocopherol-Mediated Peroxidation: The Prooxidant Effect of
Vitamin E on the Radical-Initiated Oxidation of Human Low Density Lipoprotein. J.
Am. Chem. Soc. 1993, 115, 6029-6044.
Bowry, V.W.; K.U. Ingold; and R. Stocker. Vitamin E in Human Low–Density Lipoprotein.
When and How this Antioxidant Becomes a Pro–Oxidant. Biochem J. 1992, 288, 341–
344.
Bradley, D.G.; and D.B. Min. Singlet Oxygen Oxidation of Foods. Crit. Rev. Food Sci. Nutri.
1992, 31, 211-236.

Burlakova, E.B.; S.A. Karshakov; and N.G. Khrapova. The Role of Tocopherols in Biomem-
branes’ Lipids Peroxidation. Biol. Membranes 1998, 15, 137–167 (in Russian).
Choe, E.; and D.B. Min. Chemistry and Reactions of Reactive Oxygen Species in Foods. J.
Food Sci. 2005, 70, R142-159.
Choe, E.; and D.B. Min. Mechanisms and Factors for Edible Oil Oxidation. Comp. Rev. Food
Sci. Food Safety 2006, 5, 169-186.
Cillard, J.; P. Cillard; and M. Cornier. Effect of Experimental Factors on the Prooxidant Be-
havior of α-Tocopherol. J. Am. Oil Chem. Soc. 1980, 57, 255–261.
Dad, S.; R.H. Bisby; I.P. Clark; and A.W. Parker. Formation of Singlet Oxygen from Solutions
of Vitamin E. Free Rad. Res. 2006, 40, 333-338.
Deiana, M.; A. Rosa; C.F.Q. Cao; F.M. Pirisi; G. Bandino; and M.A. Dessi. Novel Approach
to Study Oxidative Stability of Extra Virgin Olive Oils: Importance of Alpha-Tocopherol
Concentration. J. Agric. Food Chem. 2002, 50, 4342-4346.
Denisov, E.T.; and V.V. Azatyan. Chain Reactions Inhibition. Taylor and Francis: London,
2000.
Denisov, E.T.; and T.G. Denisova. Handbook of Antioxidants: Bond Dissociation Energies, Rate
Constants, Activation Energies and Enthalpies of Reactions. 2nd edn.; CRC Press: Boca Ra-
tion, 2000.
Denisov, E.; and I. Khudyakov. Mechanism of Action and Reactivities of the Free Radicals of
Inhibitors. Chem. Rev. 1987, 87, 1313-1357.
Doba, T.; G.W. Burton; K.U. Ingold; and M. Matsuo. α-Tocopheroxyl Delay: Luck of Effect
of Oxygen. J. Chem. Soc. Chem. Commun. 1984, 461–462.
Evans, J.C.; D.R. Kodali; and P.B. Addis. Optimal Tocopherol Concentrations to Inhibit Soy-
bean Oil Oxidation. J. Am. Oil Chem. Soc. 2002, 79, 47-51.
Faustman, C., D.C. Liebler; and J.A. Burr. α-Tocopherol Oxidation in Beef and in Bovine
Muscle Microsomes. J. Agric. Food Chem. 1999, 47, 1396-1399.
Fuster, M.D.; A.-M. Lampi; A. Hopia; and A. Kamal-Eldin. Effects of α- and γ-Tocopherols
on the Autoxidation of Purified Sunflower Triacylglycerols. Lipids 1998, 33, 715-722.
Huang, S.-W.; E.N. Frankel; and J.B. German. Effects of Individual Tocopherols and Tocoph-
erol Mixtures on the Oxidative Stability of Corn Oil Triglycerides. J. Agric. Food Chem.
1995, 43, 2345-2350.
Jung, M.Y.; and D.B. Min. Effects of α-, γ-and δ-Tocopherols on the Oxidative Stability of
Purified Soybean Oil. J. Food Sci. 1990, 55, 1464-1465.
Jung, M.Y.; and D.B. Min. Effects of Oxidized α-, γ- and δ-Tocopherols on the Oxidative
Stability of Purified Soybean Oil. Food Chem. 1992, 45, 183-187.
Kamal-Eldin, A.; and L. Appelqvist. The Chemistry and Antioxidant Properties of Tocopherols
and Tocotrienols. Lipids 1996, 31, 671–701.
Kamal-Eldin, A.; M. Makinen; A. Lampi; and A. Hopia. A Multivariate Study of α-Tocopherol
and Hydroperoxide Interaction during the Oxidation of Methyl Linoleate. Eur. Food Res.
Technol. 2002, 214, 52–57.
Kasaikina, O.T.; V.D. Kortenska; and N.V. Yanishlieva. Effect of Chain Transfer and Recombi-
nation/Disproportionation of Inhibitor Radicals on Inhibited Oxidation of Lipids. Russ.
Chem. Bull. 1999, 48, 1891-1896.
Kim, H.J.; H.O. Lee; and D.B. Min. Effects and Prooxidant Mechanisms of Oxidized Al-
pha-Tocopherol on the Oxidative Stability of Soybean Oil. J. Food Sci. 2007, 72, C223-
C230.
Lampi, A.M.; and A. Kamal-Eldin. Effect of Alpha- and Gamma-Tocopherols on Thermal

Polymerization of Purified High-Oleic Sunflower Triacylglycerols. J. Am. Oil Chem. Soc.
1998, 75, 1699-1703.
Lampi, A.-M.; and V. Piironen. α- and γ-Tocopherols as Efficient Antioxidants in Butter Oil
Triacylglycerols. Fett/Lipid 1998, 100, 292-295.
Lampi, A.-M.; L. Katja; A. Kamal-Eldin; and P. Vieno. Antioxidant Activities of α- and
γ-Tocopherols in the Oxidation of Rapeseed Oil Triacylglycerols. J. Am. Oil Chem. Soc.
1999, 76, 749-755.
Liebler, D.C.; J.A. Burr; L. Philips; and A.J.L. Ham. Gas Chromatography-Mass Spectrometry
Analysis of Vitamin E and Its Oxidation Products. Anal. Biochem. 1996, 236, 27-34.
Min, D.B.; and J.M. Boff. Chemistry and Reaction of Singlet Oxygen in Foods. Comprehen-
sive Rev. Food Sci. Food Safety 2002, 1, 58-72.
Mistry, B.S.; and D.B. Min. Prooxidant Effects of Monoglycerides and Diglycerides in Soy-
bean Oil. J. Food Sci. 1988a, 53, 1896-1897.
Mistry, B.S.; and D.B. Min. Isolation and Identification of Minor Components and Their Ef-
fects on Flavor Stability of Soybean Oil. In Frontiers of Flavor, Elsevier Science: Holland,
1988b, pp. 499-519.
Mukai, K.; H. Morimoto; Y. Okauchi; and S. Nagaoka. Kinetic Study of Reactions between
Tocopheroxyl Radicals and Fatty Acids. Lipids 1993, 28, 753-756.
Mukai, K.; S. Noborio; and S.I. Nagaoka. Why Is the Order Reversed? Peroxyl–Scavenging
Activity and Fats and Oils Protecting Activity of Vitamin E. Int. J. Chem. Kinetics 2005,
37, 605–610.
Naumov, V.V.; and R.F. Vasil’ev. Antioxidant and Prooxidant Effects of Tocopherol. Kinetics
Catalysis 2003, 44, 101-105.
Niki, E.; T. Satio; A. Kawakami; and Y. Kamiya. Inhibition of Oxidation of Methyl Linoleate
in Solution by Vitamin E and Vitamin C. J. Biol. Chem. 1984, 259, 4177-4182.
Pazos, M.; L. Sanchez; and I. Medina. α-Tocopherol Oxidation in Fish Muscle during Chilling
and Frozen Storage. J. Agric. Food Chem. 2005, 53, 4000-4005.
Réblová, Z. The Effect of Temperature on the Antioxidant Activity of Tocopherols. Eur. J.
Lipid Sci. Technol. 2006, 108, 858-863.
Reische, D.W.; D.A. Lillard; R.R. Eitenmiller. Antioxidants. In: C. Akoh and D.B. Min (Eds.),
Food Lipids (2nd ed.), pp. 489–516, New York: Marcel Dekker, 2002.
Rietjens, I.M.C.M.; M.G. Boersm; L. de Haan; B. Spenkelink; H.M. Awad; N.H.P. Cnub-
ben; J.J. van Zanden; H. van der Woude; G.M. Alink; and J.H. Koeman. The Prooxidant
Chemistry of the Natural Antioxidants Vitamin C, Vitamin E, Carotenoids and Flavo-
noids. Environ. Toxicol. Pharmacol. 2002, 11, 321-333.
Roginsky, V.A. Kinetics of Polyunsaturated Fat Acid Ethers Oxidation, Inhibited by Substi-
tuted Phenols. Kinetics Catalysis 1990, 31, 475–481.
Tavadyan, L.A.; and G.A. Nartoyan. Analysis of Kinetic Models of Chemical Reaction Sys-
tems. Value Approach, Gitutiun, Yerevan Armenia, (In Russian) 2005.
Tavadyan, L.A.; A.A. Khachoyan; G.A. Martoyan; and A. Kamal-Eldin. Numerical Revelation
of the Kinetic Significance of Individual Steps in the Reaction Mechanism of Methyl Li-
noleate Peroxidation Inhibited by α-Tocopherol. Chem. Phys. Lipids 2007, 147, 30-45.
Verleyen, T.; R. Verhe; A. Huyghebaert; K. Dewettinck; and W. De Grey. Identification of
α-Tocopherol Oxidation Products in Triolein at Elevated Temperature. J. Agric. Food
Chem. 2002a, 49, 1508-1511.
Verleyen, T.; A. Kamal-Eldin; C. Dobarganes; R. Verhe; R. Dewettinck; and A. Huyghebaert.
Modeling of α-Tocopherol Loss and Oxidation Products Formed during Thermoxidation

in Triolein and Tripalmitin Mixture. Lipids 2002b, 36, 719-726.
Verleyen, T.; A. Kamal-Eldin; R. Mozuraityte; R. Verhe; K. Dewettinck; A. Huyghebaert; and
W. De Greyt. Oxidation at Elevated Temperatures: Competition between Alpha-Tocoph-
erol and Unsaturated Triacylglycerols. Eur. J. Lipid Sci. Technol. 2002c, 104, 228-233.
Witting, I.A. The Oxidation of α-Tocopherol during the Autoxidation of Ethyl Oleate, Li-
noleate, Linolenate and Arachidonate. Arch. Biochem. Biophys. 1969, 129, 142-151.
Yanishlieva, N.V.; and E.M. Marinova. Inhibited Oxidation of Lipids I. Complex Estimation
and Comparison of the Antioxidative Properties of Some Natural and Synthetic Antioxi-
dants. Fat Sci. Technol. 1992, 94, 374-379.
Yanishlieva, N.V.; A. Kamal-Eldin; E.M. Marinova; and A.G. Toneva. Kinetics of Antioxidant
Action of Alpha- and Gamma-Tocopherols in Sunflower and Soybean Triacylglycerols.
Eur. J. Lipid Sci. Technol. 2002, 104, 262-270.
Yoon, S.H.; M.Y. Jung; and D.B. Min. Effects of Thermally Oxidized Triglycerides on the
Oxidative Stability of Soybean Oil. J. Am. Oil Chem. Soc. 1988, 65, 1652-1656.
7
Carotenoids and Lipid
Oxidation Reactions
Afaf Kamal-Eldin
Department of Food Science, Swedish University of Agricultural Sciences, P. O. Box
7051, 750 07 Uppsala, Sweden
Introduction
Carotenoids are yellow-orange pigments that are widely distributed in the plant and
animal kingdoms. They are common in leaves, flowers, and fruits of plants even if co-
occurring chlorophyll and flavonoids often mask their colors. The main function of
carotenoids in plants seems to be the protection of the photosynthetic apparatus from
the harmful effects of chlorophyll-photosensitized oxidation. Carotenoids are also im-
portant colorants in nature, for example they are responsible for the colors of animals
especially birds and fish (Goodwin, 1986). Marine algae contribute large amounts to
the commercial production of carotenoids estimated as 108 tons per year with the main
commercial carotenoids being fucoxanthin, axtaxanthin, lutein, violaxanthin, neoxan-
thin, lycopene, and α- and β- carotenes (Harborne and Baxter, 1993). Commercial
carotenoids are used as food colorants and as precursors for vitamin A (Bauernfeind,
1981; Simpson, 1983).
During the past couple of decades, carotenoids have received considerable atten-
tion because of their anticipated potential to contribute to protect against degenerative
diseases, such as atherosclerosis, some types of cancer, compromised immunity, and
aging (Cutler, 1984; Prabhala et al., 1990, 1993; Mathews-Roth, 1991; Gaziano and
Hennekens, 1993; Olson and Krinsky, 1995; Biyani and Sheorey, 1994; Gaziano et al.,
1995; Nishino, 1995). The main mechanisms believed to be involved in the protective
role of carotenoids include production of nutritional vitamin A retinoids, antioxidant
effects, and signaling mechanisms. A great disappointment, however, was witnessed
after results of the Finnish ATBC study showing that β-carotene supplementation led
to 18% increased incidence of lung cancers and 8% increased overall mortality (ATBC,
1994). This study was one of the first to indicate the complexity of the health contribu-
tion of phytochemicals and its dependence of the test situation, level of different com-
pounds, such as other antioxidants that may synergize the carotenoids or even protect
them against oxidative degradation.
β-Carotene and other carotenoids are believed to play very important roles as
antioxidative substances both in vivo and in vitro. They have shown antioxidant ac-
tivities in homogeneous solutions, liposomes, microsomal membranes, and lipopro-
143
144 A. Kamal-Eldin
teins (Krinsky and Denke, 1982; Burton and Ingold, 1984; Vile and Winterbourn,
1988a,b; Terao, 1989; Jialal et al., 1991; Palozza and Krinsky, 1991, 1992a,b; Ken-
nedy and Liebler, 1992; Lim et al., 1992; Palozza et al., 1992). Two mechanisms are
believed to be involved in carotenoid antioxidant reactions, first quenching reactive
singlet oxygen and excited photosensitizers, and second scavenging peroxyl and other
reactive free radicals (vide infra). While the mechanisms of singlet oxygen quenching
by carotenoids remain fairly well understood, those of their reaction with free radicals
are highly controversial and subject to experimental set up. Burton and Ingold (1984)
present an early thesis on this complexity. This chapter aims to review literature on
the implications and roles of carotenoids in lipid oxidation reactions and to highlight
areas in need of further research.
Structure and Nomenclature of Carotenoids

As β-carotene is a highly visible pigment in carrot root, the name carotenoids was
derived from the name of this plant (Daucus carota). The carotenoid structure is basi-
cally composed of a long chain of eight isoprenoid units (40 carbons) joined head-to-
tail to give a conjugated system of alternate double bonds (Fig. 7.1). The conjugated
double bond structure is the main chromophore responsible for their colors and oxi-
dative properties.
4’
19 20 18’
16 17 6’
1 15 10’
6 15’
10 1’
16’ 17’
18 20’ 19’
4
Fig. 7.1. Basic skeleton for carotenoid structures.
Different carotenoids are derived from the acyclic long-chain conjugated struc-
ture shown in Fig. 7.1 by hydrogenation, dehydrogenation, cyclization, oxidation,
and combination of these reactions (IUPAC, 1975). Carotenoids can be classified
into two major groups: (i) carotenes, which are simple non-polar unsaturated hy-
drocarbons based on the lycopene structure, and (ii) xanthophylls, which are more
polar carotenoids with oxygen functions at one or both ends of the molecule. The
structures, sources, and trivial and systematic names of 563 carotenoids have been
published by Pfander (1987), and the rules for systematic nomenclature are published
by the IUPAC (1975). Thus, the structures and semi-systematic names given in this
chapter (Fig. 7.2) are only for a few carotenoids with significance to the discussion.
Carotenoids and Lipid Oxidation Reactions 145
OH
HO L u te in
P h yto e ne OH
β- Cryptoxanthin
ζ- Carotene OH
HO
Z e a xa n th in
L yco p e ne OH
O
O
HO V io la xa n th in O
δ- Carotene
C a n tha xa n th in
O
O
γ- Carotene OH
HO A sta xa n th in
β-Carotene O
HO
O B ixin OH
α- Carotene
O OMe
O
O
C a p so ru b in
ε- Carotene
OH
Fig. 7.2. Structures of selected carotenes (right hand) and xanthophylls (left hand).
Structural Features Relevant to The Reactivity of Carotenoids

Four different structural features that influence the electron density along the conju-
gated double bond system of carotenoids dictate their properties and reactivity. These
factors are explained in subsequent sections.
The Degree of Conjugation

The high degree of double bond conjugation is the main chemical feature responsible
for the distinct physical and chemical properties of carotenoids compared to other
polyenes with isolated double bonds or to phenolic compounds with aromatic rings.
Each carbon atom in the acyclic conjugated polyene chain has a trigonal arrangement,
due to sp2-hybridization, with one electron in the pz orbital. The pz orbitals of all
conjugated carbons overlap forming a π-molecular orbital covering all conjugated car-
bons. As a result electron pairs are no longer confined to regions between nuclei, and
bonds are called delocalized bonds. Thus, the degree of conjugation in the carotenoid
molecules (9-13 conjugated double bonds) is perhaps the most important feature that
determines their oxidative and antioxidative properties (Mathews-Roth et al., 1974;
Foote, 1976; Terao, 1989; Woodall et al., 1995).
146 A. Kamal-Eldin
An important detail in this respect is the variation in electron density within the
delocalized electronic system. This is illustrated by considering butadiene, CH2=CH-
CH=CH2, the simplest conjugated hydrocarbon. According to molecular orbital the-
ory, four pz atomic orbitals combine to form four π-molecular orbitals; two fully oc-
cupied bonding π-orbitals and two non-occupied antibonding π*-orbitals (Fig. 7.3).
Due to the localization of the π-electrons over the whole conjugated structure, each
central bond has a bond order close to 1.5. Since each of the two outer carbons in
the conjugated polyene structure shares its π-electrons with only one other carbon,
the bond order of these two outer bonds are close to 2. Thus, the electron density is
not evenly distributed within the conjugated system of carotenoids with the terminal
double bonds having the highest densities (Zechmeister et al., 1943). This difference
is important when considering products of reactions with electrophiles such as oxygen
(vide infra).
+
+
+
A ntibond ing M olecular O rbita ls + π4
+
+
π3 + LU M O
ENER G Y
Low est E nergy T ransitio n
+ +
HOMO
π2 + +
B ond ing M o lecular O rbita ls + +

+ +
π1
Fig. 7.3. Molecular orbitals of a conjugated double bond system.
Geometrical Isomerism
The second important structural feature is the presence of a number of cis and trans
geometrical isomers of the chain double bonds (Zechmeister et al., 1943; Saleh and
Tan, 1991). There are 272 theoretical possibilities for cis/trans isomers of β-carotene,
but only 20 unhindered isomers are possible to obtain (Zechmeister et al., 1943;
Zechmeister, 1960). According to the Pauling rules for steric hindrance in isoprenoid
systems, such as carotenoids, only the double bonds at cis-9, cis-13, and cis-15 are
unhindered. Although Pauling rules approve di-cis and tri-cis isomers that may occur
in an equilibrium mixture as a result of spontaneous or catalytic isomerization, energy
calculations predict a very low probability for their formation. Thus, the all-trans and
the three mono-cis isomers (cis-9, cis-13, and cis-15) are the predominant isomers
of β-carotene in an equilibrium mixture (Fig. 7.4). The all-trans isomer, which is
energetically the most stable, can undergo cis-isomerization upon exposure to heat or
light. Indeed, cis-isomerization affects the planarity of the molecule and the evenness
of the relative electron density in the delocalized electron cloud.
Configuration of Cyclic Ends of Carotenoids

A third structural feature that determines the planarity of the carotenoid molecule
is the possible forms resulting from rotation about the 6,7-single bond (Orlandi et
al., 1991). There are two planar conformations where all the pz orbitals forming the
19
16 17
2 10
9 -cis
18
4
19 20
16 17
2 14
13
4
18
13 -cis
19 20
16 17
2 15
13
4
18 15 -cis 15'
19 20 18'
16 17
15
2
13
15'
18 20' 17' 16'
19'
4
all-tran s
Fig. 7.4. Natural all-trans b-carotene and its cis isomers.
148 A. Kamal-Eldin
conjugated system are parallel and give maximum overlap. These two conformations
are called s-cis and s-trans referring to the fact that the geometrical isomerism is in re-
spect to a single bond (Fig. 7.5). Besides these two extremes, other conformations are
found in which the rings and the chain are not coplanar. UV and NMR spectroscopy
showed that the β-ionone rings in all-trans-β-carotene are not coplanar with the chain
while X-ray analysis of some all-trans carotenoids in crystalline states demonstrated
that their conformation is close to the s-cis form. The non-planarity in carotenoids is
ascribed to steric interactions between the ring methyl groups and hydrogens at posi-
tions 7 and 7’ resulting in an incomplete conjugation of the ring double bonds with
the extended polyene system. This may explain why β-carotene with two β-ionone
rings has a shorter absorption wavelength (λmax= 466 nm in CHCl3) than γ-carotene
with one β-ionone ring (λmax= 475 nm in CHCl3) than lycopene with no β-ionone
rings (λmax= 480 nm in CHCl3) (Finar, 1981). This difference in planarity, and con-
sequently the degree of conjugation, was also thought to explain why lycopene has
twice the activity of β-carotene as a singlet oxygen quencher (DiMascio et al., 1989,
1991, 1992; Conn et al., 1991; Devasagayam et al., 1992).
18
16 17
2 2
16
18
4 4 17
s -cis s -trans
Fig. 7.5. s-cis and s-trans forms of the cyclic ends of carotenoids.
Oxygenated Groups at Cyclic Ends of Carotenoid

The presence of an oxy or hydroxyl substituent in the cyclic part of the carotenoids
modifies their polarity characteristics and thereby their solubility, stability, and distri-
bution in different media. Moreover, oxygenation of the cyclic rings may positively
contribute to the antioxidant properties of carotenoids (Terao, 1989; DiMascio et al.,
1989, 1991, 1992; Miki, 1991; Jorgensen and Skibsted, 1993; Oshima et al., 1993).
The presence of oxygenated functions in xanthophylls makes them markedly different
from simple carotenes, especially in solubility, light absorption, reactivity with other
chemicals, and biological potency as determined by the kinetics of transfer and reten-
tion within membranes. The relative solubility, stability, and absorptivity of lutein
and β-carotene were compared in 18 different solvents, among which tetrahydrofuran
solutions had the highest absorptivity (Craft and Soares, 1992). Lutein was least solu-
ble in hexane, while β-carotene was least soluble in methanol and acetonitrile. Lutein
was more stable than β-carotene in all solvents; both carotenoids were least stable in
cyclohexanone and β-carotene was quite unstable in dichloromethane.
Light Absorption and Quenching of Photosensitizers and

Singlet Oxygen
Light Absorption
Carotenoids display a broad absorption band in the visible region, mainly 400-500
nm, due to an allowed π-π* transition in the conjugated polyene structure (Jaffé and
Orchin, 1962). Typically, all-trans-β-carotene has an absorption maximum around
460 nm with a very high molar absorption coefficient, εmax is approximately 1.5 × 105
M-1cm-1. Carotenoids with a cis double bond show somewhat reduced εmax values but
show an extra band, called the cis band, in the UV region. Thus, 15-cis-β-carotene
has εmax of 0.9 × 105 M-1cm-1 at 460 nm and a cis-band at approximately 340 nm with
0.5 × 105 M-1cm-1. As the number of conjugated double bonds increases, the carote-
noids show a bathochromic shift with increasing λmax and εmax values (Truscott, 1990).
These light absorption properties of carotenoids are of relevance to their colors, pho-
toquenching as well as antioxidant and proxidant properties.
Upon absorption of light energy, ground state carotenoid molecules (So, 1Car) are
excited to higher singlet energy states (1Car*). For symmetrical all-trans-carotenoids,
quantum chemical calculations identified the state arising from the excitation of a
single electron from the highest occupied π-molecular orbital (HOMO) to the low-
est unoccupied π*-molecular orbital (LUMO) of all-trans-β-carotene as the S2 (1Bu+)
singlet state (Suzuki, 1967; Hudson et al., 1982). The transition to this excited singlet
state (S2, 1Bu+) from the ground state (1Ag-) is fully allowed, but the excited 1Bu+ state
is very unstable and it decays rapidly to a more stable lower energy singlet excited state
(S1, 1Ag*) by loss of heat. The latter state then relaxes back to the ground state either
directly by fluorescence (Thrash et al., 1979) or indirectly through vibrational, non-
radiating, intersystem crossing (ISC, [1]) to an equivalent-energy triplet state (T1,
3
Bu+), which can relax back to the ground state by phosphorescence (Fig. 7.6).
Car (S1, 1Ag*) → 3Car(T1, 3Bu+) intersystem-crossing

1
[1]
The relaxation of carotenoids from higher singlet and triplet states to the ground state
is not 100% efficient and the carotenoids are subject to different amounts of loss dur-
ing this process. Evidence for the incomplete relaxation of excited carotenoids comes
from the finding that triplet state β-carotene (3Car*) shows a strong absorption band
in the 500-550 nm range due to electronic transition from the lower (T1, 3Bu+) to
a higher excited triplet state (Wolff and Witt, 1969; Land et al., 1971). Moreover,
150 A. Kamal-Eldin
S in g let S tates T rip let S tates
+
S 2 ( 1 B u *)
Internal co nversion T2
(< 10 -12 sec)
A bsorption
ENER G Y
N on-ra diative
IS C
S1
1 F luorescence
( A g *)
(10 -9 -1 0 -6 sec) T1
+
(3B u )
A bsorption
P hosphorescence
(10 -4 -1 0 2 sec)
S o (1A g )
Fig. 7.6. Light absorption, excitation, and relaxation of carotenoid molecules.
photoradiation in the absence of photosensitizers causes carotenoid losses in the fol-

lowing order lycopene>>β-carotene=α-carotene>>astaxanthin (Oshima et al., 1993).
The carotenoids may also be excited by heat and/or mechanical agitation. As a result
of excitement, carotenoids undergo isomerization and oxidative destruction and may
act as pro-oxidants for co-existing lipids.
Physical Quenching of Photosensitizers and Singlet Oxygen by Carotenoids

Ground state photosensitizers (So), typically chlorophyll, are excited by absorption of
a quantum of radiation and lead to the formation of short-life, 10-11 s, singlet state,
excited photosensitizers (1S*). These excited photosensitizers dissipate their energy
either by photon emission via fluorescence or by intersystem crossing to form ex-
cited triplet state photosensitizers (3S*) with relatively long lifetimes, 10-4 s, during
which they may return to the ground state by phosphorescence or deactivation by
carotenoids. Depending on the light intensity and the nature of other reactive mol-
ecules in the assay system, particularly oxygen concentration, triplet photosensitizers
may either react with compounds or initiate photochemical reactions. Gollnick and
Schenk (1967) classified these reactions into type I reactions, redox reactions involv-
ing abstraction of a hydrogen atom or electron from the substrate forming free radi-
cals that lead inter alias to the degradation of the photosensitizer, and type II reactions
involving exchange of energy with triplet molecular oxygen (3O2) generating the most
reactive singlet oxygen (1O2) (Fig. 7.7).
Singlet oxygen, produced by type II photosensitized reactions, reacts with linoleic
acid approximately 1500 times faster than triplet oxygen and therefore was thought
to be involved in the initiation of lipid oxidation (Rawls and Van Santen, 1970). In
addition, the relative oxidizability of oleic and linoleic acid was 1:1.7 in reactions with
singlet oxygen (Terao and Matsushita, 1977) compared to 1:12 in reactions with trip-
let oxygen (Gunstone and Hilditch, 1945) making oleic acid an important reactant in
lipid oxidation reactions catalyzed by singlet oxygen. Moreover, singlet oxygen is also
known to oxidize proteins, amino acids, and DNA bases at much higher rates than
hydroperoxides produced from fatty acids. Amino acids containing heterocyclic rings
or sulfur atoms appear to be attacked by singlet oxygen with the following second
order rate constants (M-1s-1): tryptophan (2.5 × 108), histidine (1.3 × 108), tyrosine
(0.27 × 108), methionine (0.22 × 108), and alanine (0.02 × 108) (Wilkinson and
Brummer, 1981). The finding that the effects of singlet oxygen on protein oxidation
were more significant than its effects on lipid oxidation made Wilkinson and Brum-
E xcited S in g let N on-ra diative E xcited T rip let

P h o to sen sitizer IS C P h o to sen sitizer
( 1 S *) ( 3 S *)
T yp e I T yp e II
F luorescence
1
F ree R ad icals O2
ENER G Y
P hosphorescence
or caroten oids
C A R O TEN O ID
C aro ten o id
C aro ten o id sin g let
au to o xid atio n o xid atio n
G ro u n d S tate p ro d u cts p ro d u cts
P h o to sen sitizer
Fig. 7.7. Effect of photosensitizers on carotenoid photooxidation.
152 A. Kamal-Eldin
mer (1981) associate the anticarcinogenic properties of carotenoids to protection of

the genetic material against singlet oxygen.
The carotenoids are recognized as the most efficient protective agents against the
harmful effects of excited pigments and singlet oxygen generated by photochemical
reactions (Foote, 1979; Foote and Denny, 1968; Foote et al., 1970a,b,c; Mathews-
Roth, 1987). The protective effects of carotenoids against photosensitized reactions
are due to their ability to quench both excited photosensitizers and singlet oxygen at
a diffusion-controlled rate without being consumed in the process, mainly by a harm-
less energy transfer physical quenching mechanism (pathway C), that is
Sen* + 1Car → 1Sen + 3Car*

3
[2]
O2 + 1Car → 3O2 (3Σg-)+ 3Car*
1
[3]
The excitation energy is dissipated to the solvent system as thermal energy through
rotational and vibrational intersystem relaxation of the C-C and C=C bonds in the
polyene chain to recover the ground state carotenoid
Car* → 1Car + Thermal energy

3
[4]
The ability of conjugated polyene systems to quench photosensitizers and singlet oxy-
gen largely depends on the number of their conjugated double bonds (Foote, 1976;
Lee and Min, 1988, 1990; Jung and Min, 1991). To be able to accept extra energy
from singlet oxygen, carotenoids need to have their singlet states below 22.4 Kcal/
mole. Energy transfer from 1O2 to carotenoids with nine (singlet state 22 Kcal/mole)
or more conjugated double bonds is exothermic and these carotenoids are efficient ox-
ygen quenchers. The singlet oxygen quenching rate constants for selected carotenoids
in dichloromethane were lutein (5.72 × 109 M-1s-1), zeaxanthin (6.79 × 109 M-1s-1),
lycopene (6.93 × 109 M-1s-1), and isozeaxanthin (7.39 × 109 M-1s-1) (Lee and Min,
1990). Carotenoids with seven conjugated double bonds (29 Kcal/mole) and retinol
with five conjugated double bonds (34 Kcal/mole) are much less efficient or even
inactive as singlet oxygen quenchers (Mathews-Roth et al., 1974; Foote, 1976; Krin-
sky, 1979). On the other hand, the efficiency of carotenoids with seven conjugated
double bonds was found to be about 75% of that of β-carotene in quenching chloro-
phyll, and retinol was about half efficient as β-carotene in quenching methylene blue
(Foote et al., 1970a; Mathis and Klero, 1973). The relative ability of compounds with
conjugated polyene structures to quench singlet oxygen and chlorophyll is shown in
Fig. 7.8. The fact that C50 and C60 carotenoids, with 15 and 19 conjugated double
bonds, exhibit essentially the same quenching rates as β-carotene suggests that the
reaction rate approaches diffusion-controlled rates after about 11 conjugated double
bonds (Foote et al., 1970a).
The singlet oxygen quenching rate constant for β-carotene was reported to vary
from 5 × 109 to 13 × 109 M-1s-1 depending on the solvent and other experimental
conditions (Carlsson et al., 1972; Farmillo and Wilkinson, 1973; Fahrenholtz et al.,
S in g let o xyg en q u en ch in g (K Q (M -1 s -1 )
60
R elative p ro te ctio n o f C h l a (% )
1010
50
40
109
30
108 20
10
0
107 5 7 9 11
N u m b er o f co n ju g a ted C = C in p o lye n e chain
Fig. 7.8. Effect of the number of conjugated double bonds of carotenoids on the quenching of singlet oxygen
and chlorophyll-a (Adapted from Foote et al. 1970a with permission from the American Chemical Society Press).
1974; DiMascio et al., 1989). The presence of oxygenated functions, carbonyl or

hydroxyl groups, in the carotenoids enhances their singlet oxygen quenching abili-
ties, for example astaxanthin (9.79 × 109 M-1s-1 in dicholomethane) is more efficient
than the previously mentioned carotenoids with 11 double bonds. Further, γ-carotene
and lycopene are more efficient singlet oxygen quenchers than β-carotene indicating
that an open chain has a higher quenching ability than a β-iononre ring (DiMascio
et al., 1989, 1991, 1992; Conn et al., 1991; Devasagayam et al., 1992; Sundqvist et
al., 1994; Hirayama et al., 1994), an effect that may be related to the differences in
molecular planarity.
Chemical Quenching of Photosensitizers and Singlet Oxygen by Carotenoids

Although most of the protective effects of the carotenoids against the harmful ef-
fects of excited photosensitizers and singlet oxygen are due to the physical quenching
154 A. Kamal-Eldin
mechanisms discussed previously, some chemical transformations of the carotenoids

occur. For example, the carotenoids may isomerize (Stahl and Sies, 1993) or undergo
concomitant chemical reactions with free radicals generated during the process (path-
way A) or with singlet oxygen (pathway B) (Foote, 1979; Foote and Denny, 1968;
Foote et al., 1970a,b,c). An ESR study showed that when β-carotene was added to a
mixture of the singlet oxygen generator triphenyl phosphate ozonide (TPPO) and the
spin trap α-phenyl-N-tert-butylnitrone (PBN), it inhibited the formation of PBN-
singlet oxygen adducts (Pryor and Govindan, 1981). However, a triplet of doublet
(aN = 13.5 and aH = 1.5) was observed in the ESR spectrum indicating the presence
of a PBN-adduct with a peroxyl radical that is believed to result from the oxidation
of β-carotene. Although chemical reactions account for only 0.05% of the quenching
activity of β-carotene (Krasnovskii and Paramonova, 1983), they can still lead to oxi-
dative destruction of the conjugated chromophore and the formation of carotenoid
oxidation products (Foote, 1979; Hasegawa et al., 1969).
Carotenoid + 1O2 → Singlet oxygen oxidation products [5]

Carotenoid + 1O2 → 3Carotenoid + 3O2 → Triplet oxygen oxidation products [6]
1
Singlet oxygen reacts differently with isolated and conjugated double bonds. The
latter reactions can be classified according to three mechanisms:
i. The “ene” reaction characteristic of isolated double bonds producing

hydroperoxides (Gollnick and Kuhn, 1979)
[7]
ii. 1,2-Cycloaddition for both isolated and conjugated C=C bonds producing
dioxetanes,
O
O
+ 1 O2
[8]
iii. 1,4-Cycloaddition restricted to conjugated cis dienes producing

endoperoxides (Gollnick and Schenk, 1967)
O
+ 1 O2
O
[9]
Singlet oxygen oxidation products of carotenoids include β-carotene-5,8-endoperoxides

as well as β-apo-15-carotenal, β-apo-14’-carotenal, β-apo-12’-carotenal, β-apo-10’-
carotenal, and β-apo-8’-carotenal resulting from dioxetane intermediates (Schenck
and Schade, 1970). As expected, no allylic hydroperoxides characteristic of the “ene”
reaction of singlet oxygen with isolated double bonds were found (Fig. 7.9).
Peroxyl Radical Reactions Involving Carotenoids

Kinetics and Possible Mechanisms
Although the oxidation of carotenoids can occur in the absence of other oxidiz-
able substrates, it is generally enhanced by the presence of free radical generators
or unsaturated fatty acids where a coupled oxidation occurs. Elahi and Cole (1964)
showed that in the presence of tert-butyl hydroperoxides in chloroform, β-carotene
oxidized in a concentration-dependent manner. Free radicals may be generated from
the mono- or bimolecular decomposition of the hydroperoxide or from its reaction
with chloroform, that is
tert-ButOOH → tert-ButO• + OH• [10]

2 tert-ButOOH → tert-ButO• + tert-ButOO• + H2O [11]
tert-ButOOH + CHCl3 → tert-ButO• + CCl3• + H2O [12]
The reaction is significantly catalyzed by traces of transition metal ions that can split
the hydroperoxides to generate initiating hydroxyl radicals.
The long conjugated double bond system of carotenoids makes them excellent
substrates for radical attack. For example, peroxyl and alkoxyl radicals react with caro-
tenoids at much higher rates than with unsaturated fatty acids (Weber and Grosch,
1976). Carotenoids react with electron-deficient peroxyl radicals by adding them to
their conjugated polyene system; this results in carbon-centered radicals (ROO-β-
Car•) that are stabilized by resonance (Scott, 1992; Mayo, 1968; Pryor et al., 1972;
Weber and Grosch, 1976; Bors et al., 1981; Burton and Ingold, 1984; Krinsky and
Denke, 1982; Burton, 1989; Terao, 1989; Yamauchi et al., 1993; Everets et al., 1995).
Intermediate radical adducts formed from β-carotene and astaxanthin are shown in
Fig. 7.10. The stability of these radicals is dependent on the addition site of the per-
oxyl radical, but it is generally low and leads to the collapse of the hydroperoxyl
group mainly to an epoxide with carbon-centered radical or an alkoxyl radical (Mayo,
1968),
C
OOR O
[13]
156 A. Kamal-Eldin
O
O β-C aro ten e-5,8-en d o p ero xid e
β-A p o -8’-c aro ten al
β-Ap o -1 0’-ca ro ten al
β-A p o -1 2’-ca ro ten al
O
β-Ap o -1 4’-ca ro ten al
O
β-Ap o -1 5-c aro ten al
Fig. 7.9. Endoperoxide and carbonyl oxidation products of b-carotene.

OOR O
[14]
Krinsky (1989) speculated that at high peroxyl radical-to-carotenoid concenrations,

the ROO-β-Carotene• adds on other peroxyl radicals by opening double bonds so
that one carotenoid molecule can act as an antioxidant by adding several peroxyl
radicals.
OOR
.
O H
O
OOR
.
HO
O
Fig. 7.10. Intermediate roo-carotenoid radical adducts formed with b-carotene and astaxanthin.
. .
OOR
+ ROO
OOR OOR
[15]
Liebler and McClure (1996), however, discussed that the addition of peroxyl radicals
to carotenoid molecules is an inherently inefficient antioxidant mechanism for two
reasons; first, the antioxidant effectiveness depends on the extent to which released
alkoxyl radicals are scavenged; and second, the resulting ROO-β-Carotene• may re-
versibly trap oxygen to form reactive peroxyl radicals,
158 A. Kamal-Eldin
.
. OO
+ O2
OOR OOR
[16]
Studying products of the reaction between β-carotene and radicals generated from
azobis-2,4-dimethylvaleronitrile (AMVN), Liebler and McClure (1996) suggested an
alternative mechanism by which peroxyl radicals react with β-carotene by hydrogen
abstraction in an analogous way to phenolic antioxidants
β-Car + ROO• → β-Car• + ROOH [17]

β-Car• + ROO• → β-Car-OOR [18]
Other researchers supported this alternative mechanism. For example, Takahashi et

al. (1999, 2001, 2003a,b) constructed a kinetic model based on the classical free radi-
cal oxidation mechanism of hydrocarbons that fits data describing the autoxidation
of β-carotene, its co-oxidation with oleic acid, and its protection by α-tocopherol.
Although an excellent fit of data was obtained in these studies, kinetic modeling can-
not be considered conclusive for suggested mechanisms and needs to be supported by
analytical data on reaction products.
The reaction conditions might play a significant role in the reaction mechanism.
For example, under high radical fluxes generated by high temperatures or azo-initia-
tors, the oxidation of carotenoids may follow simple first order kinetic models (Chou
and Breene, 1972; Arya et al., 1979; Ramakrishnan and Francis, 1979a; Henry et al.,
1998). Generally, the oxidation of β-carotene follows a sigmoidal curve and involves
initiation, propagation, and termination stages (Kasaikina et al., 1975, 1981; Papa-
dopolou and Ames, 1994; Ozhogina and Kasaikina, 1995). As discussed in Chapter
4, the activation energy for the oxidation of conjugated linoleic acid is significantly
higher but the oxidation rate constants were greater than that for linoleic acid. This
plus the fact that that β-carotene bleaches quickly under co-oxidation conditions
(Budowski and Bondi, 1960; Ramakrishnan and Francis, 1979b; Tsuchihashi et al.,
1995) indicates that carotenoids are more favorable for radical addition than for hy-
drogen abstraction. The solvent might affect the mechanism since hydrogen transfer
may be a favored oxidative mechanism in non-polar solvents while oxidation in more
polar solvents may entail electron transfer (Weber and Grosch, 1976; Grant et al.,
1988; Jovanovic et al., 1992).
Carotenoid Structure and Antioxidant Activity

Xanthophylls containing oxygenated functions (e.g., astaxanthin, canthaxanthin, ze-
axanthin, and β-cyptoxanthin) seem to be more effective as peroxyl radical scavengers

than β-carotene and lycopene (Kurashige et al., 1989; Terao, 1989; Miki, 1991; Boey
et al., 1992; Sies et al., 1992; Terao et al., 1992; Jorgensen and Skibsted, 1993; Os-
hima et al., 1993; Woodall et al., 1995). Astaxanthin and canthaxanthin contain 13
conjugated double bonds compared to 11 in β-carotene and their conjugated polyene
system terminates with carbonyl groups. These features seem to offer them abilities
to form more stable adducts with peroxyl radicals than β-carotene. Moreover, further
stabilization of the peroxyl radical adduct through an intramolecular hydrogen bond
between the hydroxyl hydrogen and the conjugated electron pairs of the carbonyl
oxygen may explain why astaxanthin is a slightly more powerful antioxidant than
canthaxanthin (Jorgensen and Skibsted, 1993). In polar media, astaxanthin may also
benefit from the 1-keto-2-hydroxy functional feature, which may chelate trace metal
ions,
n+
+ M
HO HO
O n+ O
M
[19]
Zeaxanthin and β-cryptoxanthin, although having 11 conjugated double bonds

like β-carotene, were more effective in protecting phosphatidyl liposomes against per-
oxidation induced by both water- and lipid-soluble azo-initiators (Woodall et al.,
1995). Since the hydroxyl groups in these xanthophylls are not adjacent to the chro-
mophore, they were not believed to influence the reactivity of the carotenoids with free
radicals. Rather, the effect of these functions was related to biophysical interactions
between the phospholipid and the carotenoid. As shown in Fig. 7.11, xanthophylls
are believed to locate in membranes between the fatty acid moieties with their β-rings
close to the hydrophilic/hydrophobic interface while non-polar β-carotene and ly-
copene are believed to occur between the two halves of the bilayer (Jezowska et al.,
1994; Strzalka and Gruszecki, 1994).
Oxidation Products of Carotenoids

Although the site of addition of peroxyl radicals to the carotenoid molecule is not yet
known with certainty, it is expected to be a function of the inherent reactivity (i.e.
electron density) of the carotenoid and the size and reactivity of the attaching radi-
cal. Zechmeister et al. (1943) and El-Tinay and Chichester (1970) suggested that the
C5=C5 and C5’=C6’ terminal double bonds of β-carotene, having the highest elec-
tron density, serve as the preferential sites of radical attachment. Alternatively, Pull-
man and Pullman (1963) and Marty and Berest (1990) agreed that the C7=C8 and
160 A. Kamal-Eldin
HO HO
OH
HO
Fig. 7.11. Proposed differences in the lecalization of carotenes and xanthophylls in biological membranes.
C7’=C8’ are the positions with the highest mobility index, 0.731 obtained from mo-
lecular orbital calculations, and they favored radical addition to these double bonds.
On the other hand, the central double bonds have less energy and therefore are easier
to open. Naturally, the addition of a certain radical to a specific double bond within
the conjugated polyene depends on the steric nature of the radical, its reactivity, and
the conformation and isomerization characteristics of the carotenoid molecule as de-
termined by the amount of energy available (i.e., temperature and light).
Epoxides (5,6-, 5’,6’-, 5,8-, and 5’,8’-) and diepoxides (5,6,5’,6’-, 5,6,5’,8’-, and
5,8,5’,8’-), shown in Fig. 7.12, were detected as autoxidation products of β-carotene
(Hunter and Krakenberger, 1947; Friend, 1958a; El-Tinay and Chichester, 1970;
Kennedy and Liebler, 1991; Mordi et al., 1991, 1993; Yamauchi et al., 1993). Also,
ROO
.
.
β-C aro te n e OOR
.
- RO
[H + ]
O
O
β-C aro te n e -5,6-e p o x id e β-C aro te n e -5,8-ep o xid e
. .
+ ROO + ROO
. .
- RO - RO
O [H + ] O
O
O
β-C aro te n e -5,6,5’,6’-d iep o xid e β-C aro te n e -5,6,5’,8’-d iep o xid e
[H + ]
O
β-C aro te n e -5,8,5’,8’-d iep o xid e
Fig. 7.12. Epoxides formed by the reaction of b-carotene with peroxyl radical.
three cycloepoxy oxidation products, 13,15’-epoxyvinyleno-13,15’-dihydro-dinor-β-

carotene, 15’,13-epoxyvinyleno-13,15’-dihydro-dinor-β-carotene (which is present
in four isomers; 13R,15’R, 13S,15’S, 13R,15’S, and 13S,15’R), and 11,15’-cyclo-
12,15-epoxy-11,12,15,15’-tetrahydro-β-carotene (Fig. 7.13), were identified (Ya-
mauchi et al., 1993; McClure and Liebler, 1995). These products suggest that radicals
can add to the terminal as well as the central double bonds of β-carotene.
Enzyme-Catalyzed Co-Oxidation with Unsaturated Fatty Acids

The co-oxidation of carotenoids with polyunsaturated fatty acids was found to be
highly enhanced by the presence of lipoxygenase since a large proportion of peroxyl
radicals is not converted to hydroperoxides by the enzyme (Sumner, 1942; Smith and
Sumner, 1948; Friend, 1956, 1958b; Blain, 1970; Tookey et al., 1958; Ben-Azziz et
al., 1971; Zinsou, 1971; Grosch et al., 1977; Ikediobi and Snyder, 1977; Cohen et al.,
1985; Klein et al., 1985; Katusin-Rasem and Razem, 1994). Interestingly, the co-ox-
idation of β-carotene by soybean lipoxygenase in the presence of lipids and lipid hy-
droperoxides was faster under anaerobic than aerobic conditions (Klein et al., 1984).
This effect was explained by the formation of an enzyme-fatty acid radical complex,
Enz(FeII) + LOOH → Enz(FeIII) + LO• + -OH [20]

Enz(FeIII) + LH → Enz(FeII)-L• + +H [21]
Enz(FeII)-L• + Car → Enz(FeII) + Oxidized Car [22]
162 A. Kamal-Eldin
13,15’-ep o x y vin ylen o -13,15’-d ih yd ro -14,1 5-d in o r-β,β-caro ten e
15’,13 -ep o x y vin ylen o -13,15’-d ih yd ro -14,1 5-d in o r-β,β-caro ten e
11,15’-c yclo -1 2,15-ep o x y-11 ,12,15,1 5’-tetrah yd ro -β,β-caro ten e
Fig. 7.13. Heterocyclic oxidation products of b-carotene.
Under aerobic conditions, the last reaction would proceed as follows
Enz(FeII)-L• + O2 + LH → Enz(FeII)-LOOH + H• [23]

Enz(FeII)-LOOH → Enz(FeII)-LO• or HO• [24]
Enz(FeII)-LOO• or Enz(FeII)-LO• + Car → Oxidized Car [25]
Barimalaa and Gordon (1988) calculated the activation energies for the co-oxidation
of linoleic acid and β-carotene by soybean lipoxygenase as 4.97 and 4.78 Kcal/mole,
respectively. This indicates that they can compete for peroxyl radical equally well
and that the competition of linoleate offers protection to the carotenoid [Eqn. 25].
This competition is also confirmed by finding that carotenoid depletion increased
and linoleic acid oxidation decreased with increased carotenoid concentration. The
presence of butylatedhydroxytoluene and α-tocopherol inhibited the oxidation of
β-carotene to a greater extent than the oxidation of linoleic acid.
The Pro-oxidant Effects of Carotenoids

Reactions with Triplet Molecular Oxygen
The interaction of carotenoids with lipid oxidation is complex and subject to many
factors (Vile and Winterbourn, 1988a; Kennedy and Liebler, 1991, 1992; Palozza
and Krinsky, 1991) It was shown by Burton and Ingold (1984) that β-carotene acts
as an effective antioxidant at low oxygen pressure (ca 15 torr or 2% oxygen) however
its activity will decrease with increased oxygen tension (ca 150 torr or 20% oxygen)
until it becomes a pro-oxidant at high oxygen pressure (towards 750 torr or 100%
oxygen).
The fact that carotenoids can easily absorb light or heat energies and get excited
to higher energy states may be involved in their pro-oxidant effect observed by many
researchers (Burton and Ingold, 1984; Vile and Winterbourn, 1988b; Palozza and
Krinsky, 1991; Kennedy and Liebler, 1992; Haila and Heinonen, 1994). β-Carotene
was found to exert antioxidant effects in the absence of light and pro-oxidant ef-
fects in its presence (Stanescu and Eisenburger, 1969). It is known that ground-state
carotenoids (1Car) absorb light energy in the blue region of the spectrum (400-500
nm) and become excited to triplet states (3Car*) that might initiate or propagate lipid
oxidation (Rodgers and Bates, 1980),
1
Car + hυ (450 nm) → 3Car* [26]
3
Car* + 3O2(3Σg-) → 1Car + 1O2 [27]
The blue light represent about 20% of the energy emitted from light sources (Paul
et al., 1972a). In the absence of sensitizers, photooxidation of polyunsaturated fatty
acids is catalyzed more strongly by short light wavelengths, 325-460 nm (Parker et al.,
1952; Radtke et al., 1970; Chahine and deMan, 1971; Paul et al., 1972b,c; Satter et
al., 1976a,b). Actually, Schenck and Schade (1970) showed that β-carotene can act as
a sensitizer in type II photooxidation reactions, generating singlet oxygen via excited
carotene-O2 complexes.
Excitation of a π-electron to higher singlet states may cause partial opening of a
double bond in the conjugated structure (Fig. 7.14). Consequently, the bond order
decreases and the double bonds in the carotenoid molecule elongate and give some
carbon atoms a partial biradical character (Wasielewski et al., 1989; Christophersen
et al., 1991). Molecular orbital calculations showed that the central carbon-carbon
double bonds are elongated more than threefold in the higher energy state 1Bu+ than
in the lower energy state S1 (Wasielewski et al., 1989). The biradical intermediate,
formed by this reaction
3
Car* + 3O2(3Σg-) → 3Car•-OO• [28]
Depending on oxygen concentration, the Doering’s diradical may undergo cis-isom-

erization or oxidation generating a wide range of oxidation products, including per-
164 A. Kamal-Eldin
H ig h er S in g let
S tates E xcited
Sn T rip let S tates
N on-ra diative
IS C Tn
Internal
S2 conversion
S1
T1
F luorescence
Lig ht P hosphorescence
A bsorption
So
Fig. 7.14. Excitation of conjugated double bonds by light to form an excited triplet state (tn) that can react with
molecular oxygen.
oxides, epoxides, aldehydes, and ketones (Fig. 7.15). Oxidation products are more
pronounced for the central double bonds while the rings and adjacent double bonds
are largely unchanged (Isoe et al., 1969). This reaction pathway leads to products
different from those of oxidation by singlet oxygen (Fig. 7.9) or peroxyl radicals (Fig.
7.12). Excited triplet state carotenoids (3Car*) can also react with triplet oxygen by
1,2-addition forming dioxetanes, which decompose to short-chain carbonyl com-
pounds, mainly apocarotenals similar to those shown in Fig. 7.9. In the presence of
energy-acceptors, excited carotenoids can exchange their extra energy and relax back
to the ground state. Since relaxation is an exothermic process, lowering the tempera-
ture of the system can enhance it.
Canthaxanthin and other carbonyl-containing carotenoids are more stable to-
wards photodegradation than carotenes (Terao, 1989; Nielsen et al., 1996). This dif-
ference was attributed to the nature of the excited states having some n,π* character
in canthazanthin compared to only π, π* states in the case of β-carotene. Thus, the
excited states of xanthophylls may have less of a biradical character and may be less
labile than those of the carotenes. Xanthophylls are regarded as pure antioxidants in
contrast to carotenes, which exhibit both anti- and pro-oxidant effects as shown in
Fig. 7.16 (Martin et al., 1999).
19 20 18'
16 17
15
2
13
15'
18 20' 17' 16'
4 19'
all-tran s -β- carotene 15 -cis -β- carotene
. .
B irad ical intermediate
.
. 3 OO
O2
.
OO
.
+
O x yg enated b irad ical
D io xetan es
∆ A p o c a ro ten als
.
E p o xid es + R O
L o n g ch ain carb o n yls
Fig. 7.15. Opening of a carotenoid double bond and formation of biradical intermediates that can react with
molecular oxygen.
alp h a-T o co p h e r o l
A s taxan th in P u re an tio xid an ts
C an th axan th in
Z e axan th in
L yco p e n e A n tian d P ro - O xid an ts
b e ta-C ar o te n e
cis -Ph yto e n e N o sig n ifican t

an tio xid a n t
Z e ta-C ar o te n e p ro p e rties
M e th yl L in o le ate CONTRO L
0 20 40 60 80 100
D e g re e o f o x id a tio n (% )
Fig. 7.16. Classification of carotenoids according to their anti- and pro-oxidant reactions (data from Martin et al.
1999).
166 A. Kamal-Eldin
Effect of Heat on Carotenoid Oxidation and Cleavage

Heat causes isomerization of the carotenoids and enhances their reaction with oxygen
and degradation to undesirable compounds (McKeown, 1965; Scotter, 1995). Heat-
excited carotenoids (3Car*) undergo cis→trans isomerization or react with oxygen
to form the Car•-OO• diradical species discussed previously (Mordi, 1993). The
Car•-OO• may rearrange to form dioxetanes, which degrade to apocarotenals, or
react with radicals or other species to form peroxides that degrade to epoxides and a
wide range of volatile compounds (Cole and Kapur, 1957; Day and Erdman, 1963;
Mader, 1964; Isoe et al., 1969; La Roe and Shipley, 1970; Demole and Berthet, 1972;
Schreier et al., 1979; Kawakani, 1982; Marty and Berest, 1986a,b; Kanasawud and
Crouzet, 1990a,b; Crouzet and Kanasawud, 1992). Heat-excited carotenoids (3Car*)
isomerize to a number of mono- and poly- cis-isomers at C-9, C-13, and C-15 (Claes
and Nakayama, 1959; Zechmeister, 1960; Claes, 1961; Sweeney and Marsh, 1970;
Ogunlesi and Lee, 1979; Tsukida and Saiki, 1983). Carotenoids with cis configura-
tion have a higher steric interaction between the two parts of the molecule and pos-
sess a higher potential energy than their trans isomers. In effect, cis carotenoids are
less stable and are more susceptible to various oxidation reactions than their trans
counterparts (Scotter, 1995). For example, 9-cis-β-carotene oxidizes preferentially
and protects both all-trans-β-carotene and methyl linoleate from oxidation (Levin
and Mokedy, 1994).
The thermal degradation of β-carotene under sufficient oxygen follow zero order
kinetics (El-Tinay and Chichester, 1970; Kanasawud and Crouzet, 1990a). All-trans-
β-carotene was resistant to degradation at 180°C and in the absence of mechani-
cal mixing favored oxygen diffusion (Marty and Berest, 1986a). Non-volatile oxida-
tion products of β-carotene include the five mono- or diepoxy compounds shown
in Fig. 7.12, as well as five apocarotenals (β-apo-15-carotenal, β-apo-14’-carotenal,
β-apo-12’-carotenal, β-apo-10’-carotenal, and β-apo-8’-carotenal), one polyene ke-
tone (β-carotene-4-one), one monohydroxy (3 or 4)-5,8,5’,8’-diepoxy derivative, and
one dihydroxy derivative (trans-β-carotene-3,3’-diol) shown in Fig. 7.17 (Marty and
Berest, 1986a,b; Ouyang et al., 1980, 1986; Vecchi et al., 1981; Kanasawud, 1984;
Berest and Marty, 1992). In addition, a number of volatile oxidation products are
formed, including β-ionone, 5,6-epoxy- β-ionone, dihydroactinidiolide, 2-hydroxy-
2,6,6-trimethylcyclo-hexanone, 2.6,6-trimethylcyclohexen-1-one, 2,6,6-trimethyl-
cyclohexanone, 1-carboxaldehyde-2,6,6-trimethyl-1-cyclohexene, and β-cyclocitral
(Fig. 7.18) (Kanasawud and Crouzet, 1990a). At 30°C, only dihydroactinidiolide
was produced and most of the other compounds appeared when temperature reached
50°C.
Fig. 7.19 shows the main thermal degradation products of all-trans-lycopene in-
cluding 2-methyl-2-hepten-6-one (cleavage of C6-C7), geranial and neral (cleavage
of C8-C9), and pseudoionone and 6-methyl-3,5-heptadien-3-one (cleavage of C10-
C11) (Cole and Kapur, 1957; Schreier et al., 1977; Sieso and Crouzet, 1977; Drawert
et al., 1981; Kanasawud and Crouzet, 1990b; Crouzet and Kanasawud, 1992). The
main thermal degradation product of 9’-cis-bixin was identified as the trans-mono-
O
O
3,7,10 -T rim e th yl-1,12 -b is(2,6,6 -trim eth yl- D ih yd ro a c tin id io lid e

c yclo h e x -1 -en yl)d o d ec o -1,3,5,7,9,11 -h exaen e
C H2O H
OH
OH
3,7 -D im eth yl-8 -to lu e n yl-1 -(2,6,6 -trim eth yl- 2 -H yd ro x ym eth yl-1,3,3 -
c yc lo h e x -1 -en yl)o cta -1,3,5,7 -tetra e n e trim eth yl-1,2 -c yclo h e x a n d io l
O
OH
O
β-A p o -1 3 -caro te n o n e
1-(2,6,6 -trim eth yl-
c yclo h e x -1 -en yl)-3-
C H2O H h yd ro x y-2 -b u ta n o n e
β-A p o -1 4 -caro te n o l
Fig. 7.17. Diverse oxidation products of b-carotene.
O O O
O
O
β-Io n o n e 5,6 -E p o x y-β-io n o n e D ih yd ro a c tin id io lid e
O O CHO
OH OH
2,6,6 -T rim eth yl- 2-H yd ro x y-2,6,6 -T ri- 1-C arb o xald eh yd e -2 -
c yclo h e x a n o n e m eth yl-c yc lo h e x a n o n e h yd ro x y-2,6 ,6 -trim eth yl-1 -
c yclo h e x e n e
O CHO
2,6,6 -T ri-m e th yl- β-C yclo citral

c yclo h e x e n -1-o n e
Fig. 7.18. Formation of some volatile oxidation products of b-carotene.

168 A. Kamal-Eldin
methyl ester of the 17-carbon polyene 4,8-dimethyl-tetradecahexaenedioic acid ac-

companied by the release of m-xylene (Fig. 7.20). The mechanism of this elimination
consists of three steps; an eight electron conrotatory, a six electron disrotatory electro-
cyclic reaction, and opening of the four-membered ring (Scotter, 1995). Under ther-
mal conditions, 9’-cis-bixin degrades to a much lesser extent to yield toluene and an
18-carbon polyene and to dimethyldihydronaphthalene and a 13-carbon tetraene.
Synergism between Carotenoids and Tocopherols in

Lipid Oxidation Reactions
Lipid oxidation reactions are often a combination of free radical and singlet oxygen
reactions varying in proportion as dictated by the physicochemical conditions of the
reaction(s). In these reactions, β-carotene acts mainly as a physical quencher of singlet
oxygen and photosensitizers, having roughly 100 times the rate constant for physical
quenching of 1O2 and being much less reactive towards this species than α-tocopherol
(Terao et al., 1980; DiMascio et al., 1989; Miki, 1991). Moreover, carotenoids are
much more efficient as scavengers of alkoxyl radicals than tocopherols (Saran et al.,
1980; Bors et al., 1982, 1984). For example, Trolox C, the water-soluble analog of
α-tocopherol, is only about 30% as reactive towards alkoxyl radicals as the water-sol-
uble carotenoid crocetin or canthaxanthin (Krinsky, 1989). The reaction kinetics of
α-tocopherol and β-carotene in suppressing azobis-isobutyronitrile (AIBN)-induced
formation of malondialdehyde from unsaturated fatty acids in hexane differed sig-
nificantly (Palozza et al., 1995). The addition of the chain inhibtor α-tocopherol
led to a lag phase corresponding to the time of consumption of tocopherols but the
propagation rate, which is dependent on the degree of fatty acid unsaturation, did not
change. The addition of β-carotene, on the other hand, does not cause a lag phase but
decreased the rate of propagation in a competitive way with fatty acids.
Carotenoids and tocopherols act synergistically to inhibit lipid oxidation reac-
tions (Leibovitz et al., 1990; Palozza and Krinsky, 1992a). As mentioned vide supra,
β-carotene reacts with peroxyl radicals and oxygen-forming propagative radicals, that
is Car•-OOL and 3Car•-OO•, respectively, that would enhance the oxidation of the
carotenoid itself and co-existing lipids leading to a pro-oxidant effect. The presence
of an efficient peroxyl radical scavenger, such as α-tocopherol, would inhibit such
reactions (Budowski and Bondi, 1960; El-Tinay and Chichester, 1970; Stratton et al.,
1993). α-Tocopherol effectively retarded the co-oxidation of β-carotene and linoleate
catalyzed by soybean lipoxygenase (Barimalaa and Gordon, 1988) and of β-carotene
and oleate catalyzed by tert-butyl hydroperoxide and ferrous ions (Tsuchiya et al.,
1992). On the other hand, by scavenging singlet oxygen generated from the decom-
position of peroxyl radicals by the Russell mechanism (Russell, 1957), the carotenoids
may help increase the antioxidant potency of tocopherols. In the presence of trace
metal ions, able to dissociate hydroperoxides, the antioxidant activity of tocopherols
is compromised and some carotenoids are more effective antioxidants (Miki, 1991;
Kurashige et al., 1990; Esterbauer et al., 1991). For example, astaxanthin was 100-500
CHO
O
C leavage
C 8-C 9 C leavage P seu d o - lonone
G eran ial C 10 -C 1 1
1 6 10 14
A ll-tra n s -L yco p e n e
C leavage
O
C 6-C 7
6-M eth yl-3,5 -h e p tad ien -6-o n e
CHO
O
N eral
2 -M eth yl-2 -h e p ten -6 -o n e
Fig. 7.19. Some decomposition products of lycopene.
19 20
7 15
13' 19'
HOOC 9 13
15' 9'
9’-cis -B ixin 20' 7'
COOH
20'
20'
13' 13'
R R R
13'
7'
HOOC HOOC 7' HOOC 7'

9'
9' 9'
19' 19' 19'
8-electro n c o n ro tato ry 6 -electro n d ixro tato ry
20'
R 13'
14'
+
7'
HOOC 8'
9'
Isom erization 7’-cis -4,8 -D im e th yl- 19'
tetrad ec a h e x a n e
19 20
d io ic a cid (C 17 ) m -x yle n e
7 15 13'
COOH
HOOC 9 13
15' 7'
all-tra n s -4 ,8 -D im eth ylte trad e c a -
h e x a n e d io ic ac id
Fig. 7.20. Some decomposition products of 9’-cis-bixin.

170 A. Kamal-Eldin
times stronger than α-tocopherol in protecting rat liver mitochondria against ferrous
ion-induced peroxidation (Marty and Berest, 1986a).
When tocopherols and carotenoids co-exist during lipid oxidation, the tocoph-
erols preferentially scavenge hydroperoxyl radicals and get consumed (Tsuchihashi et
al., 1995). For example, the antioxidant consumption during the oxidation of low-
density lipoprotein particles was found to follow the following sequence: α-tocopherol
> γ-tocopherol > lycopene > β-carotene (Esterbauer et al., 1989). However, when
α-tocopherol and β-carotene were incorporated into soybean liposomal membranes,
α-tocopherol was consumed faster when radicals were generated outside the mem-
branes while β-carotene was consumed faster when radicals were generated within
the membranes (Tsuchihashi et al., 1995). This might be explainable by differences in
localization with α-tocopherol being localized at the surface of membrane (Buettner,
1993) and β-carotene being localized in the central hydrophobic region (Milon et al.,
1986).
Conclusions
The carotenoids, with their extended double bond system, are special lipid substrates
when it comes to oxidation. Compared to a fair realization of the protective effects of
carotenoids against photosensitized oxidations, the understanding of the chemistry of
oxidation of carotenoid molecules and its relevance to the free radical oxidation of co-
existing lipids is preliminary due to the wide range of reactions in which the extended
system of conjugated double bonds can participate. This chapter provides a review of
the state of the art of current knowledge in this area with the aim to serve, with other
relevant literature, as a base for future research pertinent to the roles played by carot-
enoids in lipid oxidation reactions as well as food quality and food safety.
References
Arya, S.S.; V. Natesan; D.B. Psrihar; and P.K. Vijayaraghavan. Stability of Beta-Carotene in
Isolated Systems. J. Food Technol. 1979 14, 571-578.
ATBC, The Alpha-Tocopherol Beta Carotene Cancer Prevention Study Group The Effect of
Vitamin E and Beta Carotene on The Incidence of Lung Cancer and Other Cancers in
Male Smoker. N. Engl. J. Med. 1994 330, 1029-1035.
Bauernfeind, J.C. Carotenoids as Colorants and Vitamin A Precursors: Technical and Nutritional
Applications, Academic Press: New York, 1981, pp. 48-319.
Barimalaa, I.S.; and M.H. Gordon. Cooxidation of β-Carotene by Soybean Lipoxygenase. J.
Agric. Food Chem. 1988 36, 685-687.
Ben-Azziz, A.; S. Grossman; I. Ascarelli; and P. Budowski. Carotene Bleaching Activities of
Lipoxygenase and Heme Proteins as Studied by a Direct Spectrophotometric Method.
Phytochemistry 1971 10, 1445-1452.
Berest, C.; and C. Marty. Formation of Non-Volatile Compounds by Thermal Degradation of
β−Carotene: Protection by Antioxidants. Methods Enzymol. 1992 213, 129-142.
Biyani, M.K.; and D.S. Sheorey. Carotenes, A Ray of Hope in Prevention of Cardiovascular
Disorders, Cancers, and Cataract. J. Assoc. Physicians India 1994 42, 899-903.
Blain, J.A. Carotene Bleaching Activity of Plant Tissue Extract. J. Sci. Food Agric. 1970 21,
35-38.
Boey, P.L.; A. Nagao; J. Terao; K. Tanaka; T. Suzuki; and K. Takama. Antioxidant Activity of
Xanthophylls on Peroxy Radical-Mediated Phospholipid Peroxidation. Biochim. Biophys.
Acta 1992 1126, 178-184.
Bors, W.; C. Michel; and M. Saran. Organic Oxygen Radicals in Biology: Generation and
Reactions. In Oxygen and Oxy-Radicals in Chemistry and Biology, Rodgers, M.A.; and
Powers, E.L. Eds; Academic Press: New York, 1981, pp. 75-81.
Bors, W.; M. Saran; and C. Michel. Radical Intermediates Involved in Bleaching of the Ca-
rotenoid Crocin: Hydroxyl Radicals, Superoxide Anions, and Hydrated Electrons. Int. J.
Radiat. Biol. Relat. Stud. Phys. Chem. Med. 1982 41, 493-501.
Bors, W.; C. Michel; and M. Saran. Inhibition of the Bleaching of the Carotenoid Crocin:
A Rapid Test for Quantifying Antioxidant Activity. Biochim. Biophys. Acta 1984 796,
312-319.
Budowski, P.; and A. Bondi. Autoxidation of Carotenes and Vitamin A: Influence of Fat and
Antioxidant. Arch. Biochem. Biophys. 1960 89, 66-73.
Buettner, G.R. The Pecking Order of Free Radicals and Antioxidants: Lipid Peroxidation, -
Tocopherol, and Ascorbate. Arch. Biochem. Biophys. 1993 300, 535-543.
Burton, G.W. Antioxidant Action of Carotenoids. J. Nutr. 1989 119, 109-111.
Burton, G.W.; and K.U. Ingold. β-Carotene: An Unusual Type of Antioxidant. Science 1984
224, 569-573.
Carlsson, D.J.; T. Mendenhall; T. Suprunchuk; and D.M. Wiles. Singlet Oxygen Quenching
in the Liquid Phase by Metal Chelates. J. Am. Chem. Soc. 1972 94, 8960-8962.
Chahine, M.H.; and J.M. deMan. Autoxidation of Corn Oil Under The Influence of Fluores-
cent Light. Food Sci. Technol. 1971 4, 24-28.
Chou, H.; and W. Breene. Oxidative Decoloration of Beta-Carotene in Low Moisture Model
System. J. Food Sci. 1972 37, 66-68.
Christophersen, A.G.; H. Jun; K. Jorgensen; and L.H. Skibsted. Photobleaching of Astax-
anthin and Canthaxanthin: Quantum Yields Dependence of Solvent, Temperature, and
Wavelength of Irradiation in Relation to Packaging and Storage of Carotenoids Pigment-
ed Salmonoids. Z. Lebensm. Unters Forsch. 1991 192, 433-439.
Claes, H. Energieübertragung von Angergtem Chlorophyll auf C40-Polyene mit Verschiede-
nen Chromophoren Gruppen. Z. Naturforschg 1961 16B, 445-454.
Claes, H.; and T.O.M. Nakayama. Das Photooxydative Ausbleichen von Chlorophyll in vitro
in Gedenwart von Carotinen mit Verschiedenen Chromophoren Gruppen. Z. Natur-
forschg 1959 14B, 746-747.
Cohen, B.; S. Grossman; B.P. Klein; and A. Pinsky. Pigment Bleaching by Soybean Lipoxy-
genase Type-2 and the Effect of Specific Chemical Modifications. Biochim. Biophys. Acta
1985 837, 279-287.
Cole, E.R.; and N.S. Kapur. The Stability of Lycopene: Degradation by Oxygen and Oxida-
tion During Heating of Tomato Pulps. J. Sci. Food Agric. 1957 8, 360-368.
Conn, P.F.; W. Schalch; and T.G. Truscott. The Singlet Oxygen and Carotenoid Interaction. J.
Photochem. Photobiol. 1991 B 11, 41-47.
Craft, N.E.; and J.H. Soares. Relative Solubility, Stability, and Absorptivity of Lutein and
β-Carotene in Organic Solvents. J. Agric. Food Chem. 1992 40, 431-434.
Crouzet, J.; and P. Kanasawud. Formation of Volatile Compounds by Thermal Degradation of
Carotenoids. Methods Enzymol. 1992 213, 54-62.
172 A. Kamal-Eldin
Cutler, R.G. Carotenoids and Retinol: Their Possible Importance in Determining Longevity of
Primate Species. Proc. Natl. Acad. Sci. U.S.A. 1984 87, 7627-7631.
Day, W.C.; and J.G. Erdman. Thermal Degradation Products of β-Carotene. Science 1963
141, 808-810.
Demole, E.; and D. Berthet. A Chemical Study of Burley Tobacco Flavour (Nicotina tabacum,
L.). I. Volatile to Medium-Volatile Constituents (b.p. 84 ºC/0.001 Torr). Helv. Chim.
Acta 1972 55, 1866-1882.
Devasagayam, T.P.A.; T. Werner; H. Ippendorf; H.D. Martin; and H. Sies. Synthetic Carot-
enoids, Novel Polyene Polyketones, and New Capsorubin Isomers as Efficient Quenchers
of Singlet Molecular Oxygen. Photochem. Photobiol. 1992 55, 511-514.
DiMascio, P.; S. Kaiser; and H. Sies. Lycopene as the Most Efficient Biological Carotenoid
Singlet Oxygen Quencher. Arch. Biochem. Biophys. 1989 274, 532-538.
DiMascio, P.; M.E. Murphy; and H. Sies. Antioxidant Defense Systems: The Role of Carot-
enoids, Tocopherols and Thiols. Am. J. Clin. Nutr. 1991 53, 194S-200S.
DiMascio, P.; A.R. Sundquist; T.P.A. Devasagayam; and H. Sies. Assay of Lycopene and Other
Carotenoids As Singlet Oxygen Quenchers. Methods Enzymol. 1992 213, 429-438.
Drawert, F.; P. Schreier; S. Bhiwapurkar; and I. Heindze. Chemical-Technological Aspects for
Concentration of Plant Aromas. In Flavour’81, Schreier, P. Ed., W. de Gruyter: Berlin,
New York, 1981, pp. 649-663 .
Elahi, M.; and E.R. Cole. Oxidation of β-Carotene by Hydroperoxides. Nature 1964 203,
186-187.
Esterbauer, H.; G. Striegl; H. Puhl; and M. Rotheneder. Continuous Monitoring of in vivo
Oxidation of Low-Density Lipoprotein. Free Rad. Res. Commun. 1989 6, 67-75.
Esterbauer, H.; M. Rotheneder; G. Striegl; and G. Waeg. The Role of Vitamin E in Preventing
the Oxidation of Low-Density Lipoprotein. Am. J. Clin. Nutr. 1991 53, 314S-321S.
El-Tinay, A.H.; and C.O. Chichester. Oxidation of β-Carotene: Site of Initial Attack. J. Org.
Chem. 1970 35, 2290-2293.
Everets, S.A.; S.C. Kundu; S. Maddix; and R.L. Willson. Mechanisms of Free Radical Scav-
enging by the Nutritional Antioxidant β-Carotene. Biochem. Soc. Trans. 1995 23, 230S-
233S.
Fahrenholtz, S.R.; F.H. Doleiden; A.M. Trozollo; and A.A. Lamolla. On the Quenching of
Singlet Oxygen by α-Tocopherol. Photochem. Photobiol. 1974 20, 505-509.
Farmillo, A.; and F. Wilkinson. On the Mechanism of Quenching of Singlet Oxygen in Solu-
tion. Photochem. Photobiol. 1973 18, 447-450.
Finar, I.L. Organic Chemistry, vol 2, 5th Edn,; Longman: London, U.K., 1981, pp. 463-491.
Foote, C.S. Photosensitized Oxidation and Singlet Oxygen: Consequencies in Biological Sys-
tems. In Free Radicals in Biology, vol. 2, Pryor, W.A. ed.; Academic Press: New York, 1976,
pp. 85-133.
Foote, C.S. Quenching of Singlet Oxygen. In Singlet Oxygen, Wasserman, H.H.; and Murray,
R.W. Eds., Academic Press: New York, 1979, pp. 139-171.
Foote, C.S.; and R.W. Denny. Chemistry of Singlet Oxygen. VIII. Quenching by β-Carotene.
J. Am. Chem. Soc. 1968 90, 6233-6235.
Foote, C.S.; Y.C. Chang; and R.W. Denny. Chemistry of Singlet Oxygen. X. Carotenoid
Quenching Parallels Biological Protection. J. Am. Chem. Soc. 1970a 92, 5216-5218.
Foote, C.S.; Y.C. Chang; and R.W. Denny. Chemistry of Singlet Oxygen. XI. cis/trans Isomeri-
zation of Carotenoids by Singlet Oxygen and a Probable Quenching Mechanism. J. Am.
Chem. Soc. 1970b 92, 5218-5219.
Foote, C.S.; R.W. Denny; L. Weaver; Y.C. Chang; and J. Peters. Quenching of Singlet Oxy-
gen. Ann. N.Y. Acad. Sci. 1970c 171, 130-148.
Friend, J. The Oxidation of β-Carotene by a Lipoxidase Linoleate System. Biochem. J. 1956
64, 19-20.
Friend, J. The Coupled Oxidation of β-Carotene by a Linoleate Lipoxidase System and by
Autoxidizing Linoleate. Chem. Ind. 1958a 20, 597-598.
Friend, J. The Biochemical Oxidation of β-Carotene. Qual. Plant Mat. Veg. 1958b 4-4,
354-359.
Gaziano, J.M.; and C.H. Hennekens. The Role of Beta-Carotene in the Prevention of Cardio-
vascular Disease. Ann. N.Y. Acad. Sci. 1993 691, 796-799.
Gaziano, J.M.; J.E. Manson; L.G. Branch; G.A. Colditz; W.C. Willett; and J.E. Buring. Pro-
spective Study of Consumption of Carotenoids in Fruits and Vegetables and Decreased
Cardiovascular Mortality in the Elderly. Ann. Epidemiol. 1995 5, 255-260.
Gollnick, K.; and H. Kuhn. Ene-Reactions with Singlet Oxygen. In Singlet Oxygen, Wasser-
man, H.H.; and Murray, R.W. Eds., Academic Press: New York, 1979, pp. 287-427.
Gollnick, K.; and G.O. Schenk. Oxygen as Dienophile. In 1,4-Cycloaddition Reactions, Hamer,
J. Ed., Academic Press: New York, 1967, pp. 255-344.
Goodwin, T.W. Metabolism, Nutrition and Function of Carotenoids. Ann. Rev. Nutr. 1986
6, 273-297.
Grant, J.L.; V.J. Kramar; R. Ding; and L.D. Kispert. Carotenoid Cation Radicals: Electro-
chemical, Optical and ESR Study. J. Am. Chem. Soc. 1988 110, 2152-2157.
Grosch, W.; F. Weber; and K.H. Fischer. Bleaching of Carotenoid by the Enzyme Lipoxygen-
ase. Ann. Technol. Agric. 1977 26, 133-137.
Gunstone, F.D.; and T.P. Hilditch. The Union of Gaseous Oxygen with Methyl Oleate, Linole-
ate and Linolenate. J. Chem. Soc. 1945 836-841.
Haila, K.; and M. Heinonen. Action of β-Carotene on Purified Rapeseed Oil During Light
Storage. Lebensm.-Wiss. U., Technol. 1994 27, 573-577.
Harborne J.B.; and H. Baxter. Phytochemical Dictionary: A Handbook of Bioactive Compounds
from Plants, Taylor and Francis: London, U.K., 1993, pp. 745-746
Hasegawa, K.; J.D. Macmillan; W.A. Maxwell; and C.O. Chichester. Photosensitized Bleach-
ing of β-Carotene with Light at 632.8 nm from a Continuous Gas Laser. Photochem.
Photobiol. 1969 9, 165-169.
Henry, L.K.; G.L. Catignani; and S.J. Schwartz. Oxidative Degradation Kinetics of Lycopene,
Lutein, and 9-cis- and All-trans-β-Carotene. J. Am. Oil Chem. Soc. 1998 75, 823-829.
Hirayama, O.; K. Nakamura; S. Hamda; and Y. Kobayasi. Singlet Oxygen Quenching Ability
of Naturally Occuring Carotenoids. Lipids 1994 29, 149-150.
Hudson, B.S.; B.E. Kohler; and K.S. Schulten. Linear Polyene Electronic Structure and Po-
tential Surfaces. In Excited States, vol. 6; Lim E.C. Ed., Academic Press, New York, 1982,
pp. 1-95.
Hunter, R.F.; and R.M. Krakenberger. The Oxidation of β-Carotene in Solution by Oxygen.
J. Chem. Soc. 1947 1-4.
Ikediobi, C.O.; and H.E. Synder. Cooxidation of β-Carotene by an Isozyme of Soybean Li-
poxygenase. J. Agric. Food Chem. 1977 25, 124-127.
Isoe, S.; S.B. Hyeon; S. Katsunura; and T. Sakan. Photo-Oxygenation of Carotenoids. I. The
Formation of Dihydrpactinidiolide and β-Ionone from β-Carotene. Tetrahedron Lett.
1969 10, 279-281.
IUPAC, Commission on Nomenclature of Organic Chemistry and IUPAC-IUB Commission
174 A. Kamal-Eldin
on Biochemical Nomenclature, Nomenclature of Carotenoids, Pure Appl. Chem. 1975

41, 407-431.
Jaffé, H.H.; and M. Orchin. Theory and Applications of Ultraviolet Spectroscopy, Wiley and
Sons: New York, 1962, 624 pages.
Jezowska, I.; A. Wolak; W.I. Gruszecki; and K. Strzalka. Effect of Beta-Carotene on Structural
and Dynamic Properties of Model Phosphatidylcholine Membranes. II. A 31P-NMR and
13
C-NMR Study. Biochim. Biophys. Acta 1994 1194, 143-148.
Jialal, I.; E.P. Norkus; L. Cristol; and S.M. Grundy. β-carotene Inhibits the Oxidative Modifi-
cation of Low Density Lipoprotein. Biochim. Biophys. Acta 1991 1081, 134-138.
Jorgensen, K.; and L.H. Skibsted. Carotenoid Scavenging of Radicals: Effects of Carotenoid
Structure and Oxygen Partial Pressure on Antioxidant Activity. Z. Lebensm. Unters Forsch.
1993 196, 423-429.
Jovanovic, S.V.; I. Jovanovic; and L. Josimovic. Electron Transfer Reactions of Alkyl Peroxyl
Radicals. J. Am. Chem. Soc. 1992 114, 9018-9021.
Jung, M.M.; and D.B. Min. Effects of Quenching Mechanisms of Carotenoids on the Photo-
sensitized Oxidation of Soybean Oil. J. Am. Oil Chem. Soc. 1991 68, 653-658.
Kanasawud, P. Identification des Composes Volatils Produits per Degradation des Carotenoids en
Miliew Aqueux. Mecanismw de Formation, Ph.D. thesis, Universite des Science et Tech-
niques du Languedoc, Montpollier, France, 1984.
Kanasawud, P.; and C, Crouzet. Mechanism of Formation of Volatile Compounds by Thermal
Degradation of Carotenoids in Aqueous Medium. 1. β-Carotene Degradation. J. Agric.
Food Chem. 1990a 38, 237-243.
Kanasawud, P., and Crouzet, C. Mechanism of Formation of Volatile Compounds by Thermal
Degradation of Carotenoids in Aqueous Medium. II. Lycopene Degradation. J. Agric.
Food Chem. 1990b 38, 1238-1242.
Kasaikina, O.T.; A.B. Gagarina; and N.M. Emanuel. Reactivity of β-Carotene in the Interac-
tion with Free Radicals. Izv. Akad. Nauk. USSR Ser. Khim. 1975 10, 2243-2246.
Kasaikina, O.T.; Z.S. Kartasheva; and A.B. Gagarina. Polyene Compounds as Free Radical
Acceptors Free Radicals. Izv. Akad. Nauk. USSR Ser. Khim. 1981 3, 536-540.
Katusin-Rasem, B.; and D. Razem. Activity of Antioxidants in Solution and in Irradiated
Heterogeneous System. J. Am. Oil Chem. Soc. 1994 71, 519-523.
Kawakani, M. Ionone Species Compounds from β-Carotene by Thermal Degradation in
Aqueous Medium. Nippon Nogekagaky Kaisi 1982 56, 917-921.
Kennedy, T.A.; and D.C. Liebler. Peroxy Radical Oxidation of β-Carotene: Formation of
β-Carotene Epoxides. Chem. Res. Toxicol. 1991 4, 290-295.
Kennedy, T.A.; and D.C. Liebler. Peroxy Radical Scavenging by β-Carotene in Lipid Bilayers:
Effect of Oxygen Partial Pressure. J. Biol. Chem. 1992 267, 4658-4663.
Klein, B.P.; S. Grossman; G. King; B.S. Cohen; and A. Pinksky. Pigment Bleaching, Carbonyl
Production, and Antioxidant Effect During the Anaerobic Lipoxygeanse Reaction. Bio-
chim. Biophys. Acta 1984 793, 72-79.
Klein, B.P.; D. King; and S. Grossman. Co-oxidation Reactions of Lipoxygenase of plant Sys-
tems. Adv. Free Rad. Biol. Med. 1985 1, 309-343.
Kransovskii, A.A.; and I.I. Paramonova. Interaction of Singlet Oxygen with Carotenoids: Rate
Constants of Physical and Chemical Quenching. Biophysics 1983 28, 769-774.
Krinsky, N.I. Biological Roles of Singlet Oxygen. In Singlet Oxygen, Wasserman, H.H.; and
Murray, R.W. Eds,; Academic Press: New York, 1979, pp. 597-641.
Krinsky, N.I. Antioxidant Functions of Carotenoids. Free Rad. Biol. Med. 1989 7, 617-635.
Krinsky, N.L.; and S.M. Denke. The Interaction of Oxygen and Oxyradicals with Carot-
enoids. J. Natl. Cancer Inst. 1982 69, 205-210.
Kurashige, M.; Y. Okazoe; E. Okimasu; Y. Ando; M. Mori; W. Miki; M. Inoue; and K. Ut-
sumi. Oxidative Injury of Biological Membranes Mediated by Free Radical and the Inhi-
bition by Astaxanthin. Cytoprotect. Biol. 1989 7, 383-391.
Kurashige, M.; E. Okimasu; M. Inoue; and K. Utsumi. Inhibition of Oxidative Injury of
Biological Membranes by Astaxanthin. Physiol. Chem. Phys. Med. NMR 1990 22, 27-38.
La Roe, E.G.; and P.A. Shipley. Whiskey Composition: Formation of Alpha- and Beta-Ionone
by the Thermal Decomposition of Beta-Carotene. J. Agric. Food Chem. 1970 18, 174-
175.
Land, E.J.; A. Sykes; and T.G. Truscott. Photochemistry pf Biological Molecules. II. The Trip-
let States of β-Carotene and Lycopene Excited by Pulse Radiolysis. Photochem. Photobiol.
1971 13, 311-320.
Lee, E.C.; and D.B. Min. Quenching Mechanisms of β-Carotene on the Chlorophyll-Sensi-
tized Photooxidation of Soybean Oil. J. Food Sci. 1988 53, 1894-1895.
Lee, E.C.; and D.B. Min. Effects, Quenching Mechanisms and Kinetics of Carotenoids in
Chlorophyll-Sensitized Photooxidation of Soybean Oil. J. Agric. Food Chem. 1990 38,
1630-1534.
Leibovitz, B.; M.L. Hu; and A.L. Tappel. Dietary Supplements of Vitamin E, Beta-Catotene,
Coenzyme Q10, and Selenium Protect Tissues Against Lipid Peroxidation in Rat Tissue
Slices. J. Nutr. 1990 120, 97-104.
Levin, G.; and S. Mokedy. Antioxidant Activity of 9-cis Compared to All-trans-β-Carotene in
vitro. Free Rad. Biol. Med. 1994 17, 77-82.
Liebler, D.C.; and T.D. McClure. Antioxidant Reactions of β-Carotene: Identification of Car-
otenoid Radical Adducts. Chem. Res. Toxicol. 1996 9, 8-11.
Lim, B.P.; A. Nagao; J. Terao; K. Tanaka; T. Suzuki; and K. Takama. Antioxidant Activity of
Xanthophylls on Peroxy Radical Mediated Phospholipid Peroxidation. Biochim. Biophys.
Acta 1992 1126, 178-184.
Mader, I. Thermal Degradation of β-Carotene. Science 1964 144, 533-534.
Martin, H.D.; C. Ruck; M. Schmidt; S. Sell; S. Beutner; B. Mayer; and R. Walsh. Chemistry
of Carotenoid Oxidation and Free Radical Reactions, Pure Appl. Chem. 1999 71, 2253-
2262.
Marty, C.; and C. Berest. Degradation of trans-β-Carotene During Heating in Sealed Glass
Tubes and Extrusion Cooking. J. Food Sci. 1986a 51, 698-702.
Marty, C.; and C. Berest. Degradation Products of trans-β-Carotene Produced During Extru-
sion Cooking. J. Food Sci. 1986b 53, 1880-1886.
Marty, C.; and C. Berest. Factors Affecting the Thermal Degradation of All-trans-β-Carotene.
J. Agric. Food Chem. 1990 38, 1063-1067.
Mathews-Roth, M.M. Photoprotection by Carotenoids, Fed. Proc. 1987 46, 1890-1893.
Mathews-Roth, M.M. Recent Progress in the Medical Applications of Carotenoids. Pure Appl.
Chem. 1991 63, 147-156.
Mathews-Roth, M.M.; T. Wilson; E. Fujimoriti; and N.I. Krinsky. Carotenoid Chromophore
Length and Protection Against Photosensitization. Photochem. Photobiol. 1974 19, 217-
222.
Mathis, P.; and J. Klero. The Triplet State of β-Carotene and Analog Polyenes of Different
Length. Photochem. Photobiol. 1973 18, 343-346.
Mayo, F.R. Free Radical Autoxidation of Hydrocarbons. Acc. Chem. Res. 1968 1, 193-201.
176 A. Kamal-Eldin
McClure, T.D.; and D.C. Liebler. A Rapid Method for Profiling the Products of Antioxidant
Reactions by Negative Ion Chemical Ionization Mass Spectrometry. Chem. Res. Toxicol.
1995 8, 128-135.
McKeown, G.G. Composition of Oil-Soluble Annatto Food Colors. III. Structure of the Yel-
low Pigments Formed by the Thermal Degradation of Bixin. J. Assoc. Off. Anal. Chem.
1965 48, 835-837.
Miki, W. Biological Functions and Activities of Animal Carotenoids. Pure Appl. Chem. 1991
63, 141-146.
Milon, A.; G. Wolff; G. Ourisson; and Y. Nakatani. Organization of Carotenoid-Phospholipid
Bilayer Systems: Incorporation of Zeaxanthin, Astaxanthin, and Their C50 Homologues
into Dimyristoylphosphatidylcholine Vesicles. Helv. Chim. Acta 1986 69, 12-24.
Mordi, R.C. Mechanisms of β-Carotene Degradation. Biochem. J. 1993 292, 310-312.
Mordi, R.C.; J.C. Walton; G.W. Burton; L. Hughes; K.U. Ingold; and D.A. Lindsay. Explor-
atory Study of β-Carotene Autoxidation. Tetrahedron Lett. 1991 32, 4203-4206.
Mordi, R.C.; J.C. Walton; G.W. Burton; L. Hughes; K.U. Ingold; D.A. Lindsay; and D.J.
Moffatt. Oxidative Degradation of β-Carotene and β-Apo-8’-Carotenal, Tetrahedron Lett.
1993 49, 911-928.
Nielsen, B.R.; A. Mortensen; K. Jorgensen; and L.H. Skibsted. Singlet versus Triplet Reactivity
in Photodegradation of C40 Carotenoids. J. Agric. Food Chem. 1996 44, 2106-2113.
Nishino, H. Cancer Chemoprevention by Natural Carotenoids and Their Related Com-
pounds. J. Cell Biochem. 1995 Suppl. 2, 231-235.
Ogunlesi, A.T.; and C.Y. Lee. Effect of Thermal Processing on the Stereoisomerism of Major
Carotenoids and Vitamin A Value of Carrots. Lebensmittel Wissenschaft und Technologie
1979 4, 311-318.
Olson, J.A.; and N.L. Krinsky. Introduction: The Colorful Fascinating World of the Carot-
enoids, Important Physiological Modulators. FASEB J. 1995 9, 1547-15550.
Onyewu, P.N.; C.-T. Ho; and H. Daun. Characterization of β-Carotene Thermal Oxidation
Products in a Model Food System. J. Am. Oil Chem. Soc. 1986 63, 1437-1441.
Orlandi, G.; F. Zerbetto; and M.Z. Zgierski. Theoretical Analysis of Spectra of Short Polyenes.
Chem. Revs. 1991 91, 867-891.
Oshima, S.; F. Ojima; H. Sakamoto; Y. Ishiguro; and J. Terao. Inhibitory Effect of β-Carotene
and Astaxanthin on Photosensitized Oxidation of Phospholipid Bilayers. J. Nutr. Sci.
Vitaminol. 1993 39, 607-615.
Ouyang, J.; H. Daun; S. Chang; and C.T. Ho. Formation of Carbonyl Compounds from
β-Carotene During Palm Oil Deodorization. J. Food Sci. 1980 43, 1214-1217.
Ozhogina, O.A.; and O.T. Kasaikina. β-Carotene as an Interceptor of Free Radicals. Free Rad.
Biol. Med. 1995 19, 575-581.
Palozza, P.; and N.L. Krinsky. The Inhibition of Radical-Initiated Peroxidation of Microsomal
Lipids by Both α-Tocopherol and β-Carotene. Free Rad. Biol. Med. 1991 11, 407-414.
Palozza, P.; and N.I. Krinsky. β-Carotene and α-Tocopherol as Synergistic Antioxidants. Arch.
Biochem. Biophys. 1992a 297, 184-187.
Palozza, P.; and N.I. Krinsky. Astaxanthin and Canthaxanthin are Potent Antioxidants in a
Membrane Model. Arch. Biochem. Biophys. 1992b 297, 291-295.
Palozza, P.; S. Moualla; and N.I. Krinsky. Effects of β-Carotene and α-Tocopherol on Radical-
Initiated Peroxidation of Microsomes, Free Rad. Biol. Med. 1992 13, 127-136.
Palozza, P.; C. Luberto; and G.M. Bartoli. The Effect of Fatty Acid Unsaturation on the Anti-
oxidant Activity of Beta-Carotene and Alpha-Tocopherol in Hexane Solutions. Free Rad.
Biol. Med. 1995 18, 943-948.

Papadopolou, K.; and J.M. Ames. Kinetics of All-trans-β-Carotene Degradation on Heating
With and Without Phenylalanine. J. Am. Oil Chem. Soc. 1994 71, 893-896.
Parker, M.E.; E.H. Havey; and E.S. Stateller. Elements of Food Engineering, Reinhold Publish-
ing Co.: New York, 1952, 386 pages.
Paul, G.; R. Redtke; R. Heiss; and K. Becker. Influence of Light on the Oxidative Deteriora-
tion of Edible Oils, IV. Dependence of the Rate of Oxidation on the Wavelength of
Incident Light. Fette Seifen Anstrichmittel 1972a 74, 359-366.
Paul, G.; R. Heiss; K. Becker; and R. Redtke. Influence of Light on the Oxidative Deteriora-
tion of Edible Oils, II. Influence of Oxygen Partial Pressure and Light Intensity on Rate
of Oxidation. Fette Seifen Anstrichmittel 1972b 74, 120-126.
Paul, G.; K. Becker; R. Redtke; and R. Heiss. Influence of Light on the Oxidative Deteriora-
tion of Edible Oils, V. Reaction Kinetics. Fette Seifen Anstrichmittel 1972c 74, 484-491.
Pfander, H., Ed. Key to Carotenoids, 2nd Ed., Basel: Birkhäser, Switzerland, 1987, 296 pages.
Prabhala, R.H.; H.S. Garewal; F.L. Meyskens; and R.R. Watson. Immunomodulation in Hu-
mans Caused by Beta-Carotene and Vitamin A, Nutr. Res. 1990 10, 1473-1486.
Prabhala, R.H.; M. Braune; H.S. Garewal; and R.R. Watson. Influence of Beta-Carotene on
Immune Functions. Ann. N.Y. Acad. Sci. 1993 691, 262-263.
Pryor, W.A.; D.L. Fuller; and J.P. Stanely. Reactions of Radicals. 41. Reactivity of the Methyl
Radical. J. Am. Chem. Soc. 1972 94, 1632-1638.
Pryor, W.A.; and C.K. Govindan. Decomposition of Triphenylphosphine Ozonide in the Pres-
ence of Spin Traps. J. Org. Chem. 1981 45, 4679-4682.
Pullman, B.; and A. Pullman. Quantum Biochemistry, Interscience Publ.: New York, 1963, p.
445.
Radtke, R.; P. Smits; and R. Heiss. Influence of Light of Varying Intensity and Wavelength on
the Oxidative Deterioration of Edible Oils. II. Experimental Results and Discussion. Fette
Seifen Anstrichmittel 1970 72, 497-504.
Ramakrishnan, T.V.; and F.J. Francis. Stability of Carotenoids in Model Aqueous Systems. J.
Food Qual. 1979a 2, 177-189.
Ramakrishnan, T.V.; and F.J. Francis. Coupled Oxidation of Carotenoids in Fatty Acid Esters
of Varying Unsaturation. J. Food Qual. 1979b 2, 277-287.
Rawls, H.R.; and P.J. van Santen. A Possible Role for Singlet Oxygen in the Initiation of Fatty
Acid Autoxidation. J. Am. Oil Chem. Soc. 1970 54, 234-236.
Rodgers, M.A.J.; and A.L. Bates. Kinetic and Spectroscopic Features of Some Carotenoid Trip-
let States Sensitization by Singlet Oxygen. Photochem. Photobiol. 1980 31, 533-537.
Russell, G.A. Deuterium Isotope Effects in the Autooxidation of Aralkyl Hydrocarbons: Mech-
anism of the Interaction of Peroxy Radicals. J. Am. Chem. Soc. 1957 79, 3871-3877.
Saleh, M.H.; and B. Tan. Separation and Identification of cis/trans Carotenoid Isomers. J.
Agric. Food Chem. 1991 39, 1438-1443.
Saran, M.; C. Michel; and W. Bors. The Bleaching of Crocin by Oxygen Radicals. In Chemical
and Biochemical Aspects of Superoxide Dismutase, Bannister, J.V.; and Hill, H.O.A., Eds.;
Elsevier: New York, 1980, pp. 38-44.
Satter, A.; J.M. DeMan; and J.C. Alexander. Effect of Wavelength on Light-Induced Quality
Deterioration of Edible Oils and Fats. Can. Inst. Food Sci. Technol. J. 1976a 9, 108-113.
Satter, A.; J.M. DeMan; and J.C. Alexander. Light-Induced Oxidation of Edible Oils and Fats.
Lebensm. Wiss. Technol. 1976b 9, 149-152.
Schenck, G.O.; and G. Schade. [Photosensitized O2-Transfer in Presence of β-Carotene Under
178 A. Kamal-Eldin
Exclusion of Singlet Oxygen]. Chimia 1970 24, 13-16.

Schreier, P.; F. Drawert; and A. Junker. The Quantitative Composition of Natural and Techno-
logically Changed Aromas of Plants. IV. Enzymic and Thermal-Reaction Products Formed
During the Processing of Tomatoes. Z. Lebensm. Unters. Forsch. 1977 165, 23-27.
Schreier, P.; F. Drawert; and S. Bhiwapurkar. Volatile Compounds Formed by Thermal Degra-
dation of β-Carotene. Chem. Microbiol. Technol. Lebensm. 1979 6, 90-91.
Scott, G. Atmospheric Oxidation and Antioxidants, Second edn., Elsevier Publishing Co.: New
York, 1992, 215-219.
Scotter, M.J. Characterization of the Colored Thermal Degradation Products by Bixin from
Annato and a Revised Mechanism for Their Formation. Food Chem. 1995 53, 177-185.
Sies, H.; W. Stahl; and A.R. Sundquist. Antioxidant Functions of Vitamins: Vitamin E and C,
Beta-Carotene and Other Carotenoids. Ann. N.Y. Acad. Sci. 1992 669, 7-20.
Sieso, V.; and J.C. Crouzet. Tomato Volatile Components: Effects of Processing. J. Am. Hotr.
Sci. 1977 95, 461-464.
Simpson K.L. Relative Value of Carotenoids as Precursors of Vitamin A. Proc. Nutr. Soc. 1983
42, 7-17.
Smith, G.N.; and J.B. Sumner. The Induced Reaction Between Methyl Linoleate and Bixin
During Oxidation by Lipoxidase. Arch. Biochem. 1948 17, 75-80.
Stanescu, V.; and T. Eisenburger. Hydrolytic Changes in Lard During Rancidity Development:
Prevention and Elimination of Rancidity. Industri Alimentari 1969 20, 369.
Stahl, W.; and H. Sies. Physical Quenching of Singlet Oxygen and cis/trans Isomerization of
Carotenoids. Ann. N.Y. Acad. Sci. 1993 691, 10-19.
Stratton, S.P.; W.H. Schaefer; and D.C. Liebler. Isolation and Identification of Singlet Oxygen
Oxidation Products of β-Carotene. Chem. Res. Toxicol. 1993 6, 542-547.
Strzalka, K.; and W.I. Gruszecki. Effect of Beta-Carotene on Structural and Dynamic Proper-
ties of Model Phosphatidylcholine Membranes. I. An ESR Spin Label Study. Biochim.
Biophys. Acta 1994 1194, 138-142.
Sumner, R.J. Lipid Oxidase Studies. III. The Relation Between Carotene Oxidation and the
Enzyme Peroxidation of Unsaturated Fats. J. Biol. Chem. 1942 146, 215-218.
Sundqvist, A.; K. Briviba; and H. Sies. Singlet Oxygen Quenching by Carotenoids. Methods
Enzymol. 1994 234, 384-388.
Suzuki, H. Electronic Absorption Spectra and Geometry of Organic Molecules, Academic Press,
New York, 1967, 568 pages.
Sweeney, J.P.; and A.C. Marsh. Vitamins and Other Nutrients: Separation of Cartene Stereo-
isomers in Vegetables. J. Assoc. Off. Anal. Chem. 1970 53, 937-940.
Takahashi, A.; N. Shibasaki-Kitakawa; and T. Yonemoto. Kinetic Model for Autoxidation of
β-Carotene in Organic Solutions. J. Am. Oil Chem. Soc. 1999 76, 897-903.
Takahashi, A.; J. Suzuki; N. Shibasaki-Kitakawa; and T. Yonemoto. A Kinetic Model for Co-
Oxidation of β-Carotene with Oleic Acid. J. Am. Oil Chem. Soc. 2001 78, 1203-1207.
Takahashi, A.; N. Shibasaki-Kitakawa; and T. Yonemoto. A Rigorous Kinetic Model for
β-Carotene Oxidation in the Presence of an Antioxidant α-Tocopherol. J. Am. Oil Chem.
Soc. 2003a 80, 1241-1247.
Takahashi, A.; N. Shibasaki-Kitakawa; and T. Yonemoto. Kinetic Analysis of β-Carotene Oxi-
dation in a Lipid Solvent With or Without an Antioxidant. In Lipid Oxidation Pathways,
vol. 1, Kamal-Eldin, A, Ed., AOCS Press: Champaign, Illinois, 2003b, pp. 111-137.
Terao, J. Antioxidant Activity of β-Carotene and Related Carotenoids in Solution. Lipids 1989
24, 659-661.
Terao, J.; and S. Matsushita. Products Formed by Photosensitized Oxidation of Unsaturated

Fatty Acid Esters. J. Am. Oil Chem. Soc. 1977 54, 234-238.
Terao, J.; R. Yamauchi; H. Murkami; and S. Matsushita. Inhibitory Effects of Tocopherols
and β-Carotene on Singlet Oxygen-Initiated Photooxidation of Methyl Linoleate and
Soybean Oil. J. Food Process Preserv. 1980 4, 79-93.
Terao, J.; P.L. Boey; F. Ojima; A. Nagao; T. Suzuki; and K. Takama. Astaxanthin as Chain
Breaking Antioxidant in Phospholipid Peroxidation. In Oxygen Radicals, Yagi, K.; Kondo,
M.; Niki, E.; and Yoshikawa, T. Eds., Elsevier Sci. publ.: Amesterdam, 1992, pp. 657-
660.
Tookey, H.L.; R.G. Wilson; R.L. Lohman; and H.J. Dutton. Coupled Oxidation of Carotene
and Linoleate by Lipoxidase. J. Biol. Chem. 1958 230, 65-72.
Thrash, R.J.; H.L-B. Fang; and G.E. Leori. On the Role of Forbidden Low-Lying Excited
States of Light Harvesting Carotenoids in Energy Transfer in Photosynthesis. Photochem.
Photobiol. 1979 29, 1049-1050.
Truscott, T.G. The Photophysics and Photochemistry of the Carotenoids. J. Photochem. Pho-
tobiol. B. 1990 6, 359-371.
Tsuchihashi, H.; M. Kigoshi; and E. Niki. Action of β−Carotene as an Antioxidant Against
Lipid Peroxidation. Arch. Biochem. Biophys. 1995 323, 137-147.
Tsuchiya, M.; G. Scita; H.-J. Freistleben; V. Kagan; and L. Packer. Antioxidant Radical-Scav-
enging Activity of Carotenoids and Retinoids as Compared to α-Tocopherol. Methods
Enzymol. 1992 213, 460-472.
Tsukida, K.; and K. Saiki. Thermal Stereoisomerization of All-(E)-β-Carotene, (Z)-β-Carotene,
and Electrocyclized β-Carotene. J. Nutr. Sci. Vitaminol. 1983 29, 111-122.
Vecchi, M.; G. Englert; R. Maurer; and V. Meduna. Separation and Characterization of
β-Carotene Isomers. Helv. Chim. Acta 1981 64, 2746-2758.
Vile, G.F.; and C.C. Winterbourn. Inhibition of Adriamycin-Promoted Lipid Peroxidation by
Beta-Carotene, Alpha-Tocopherol and Retinol at High and Low Oxygen Partial Pressure.
FEBS Lett. 1988a 238, 353-356.
Vile, G.F.; and C.C. Winterbourn. Adriamycin Dependent Peroxidation of Rat Liver Micro-
somes Catalyzed by Iron Chelates and Ferritin: Maximum Peroxidation at Low Oxygen
Partial Pressure. Biochem. Pharmacol. 1988b 37, 2893-2897.
Wasielewski, M.R.; D.G. Johnson; E.G. Bradford; and L.D. Kispert. Temperature Depen-
dence of the Lowest-Excited Singlet State Lifetime of All-trans-β-Carotene and Fully De-
graded All-trans-β-Carotene. J. Chem. Phys. 1989 91, 6691-6697.
Weber, F.; and W. Grosch. Co-Oxidation of a Carotenoid by the Enzyme Lipoxygenase: In-
fluence on the Formation of Linoleic Acid Hydroperoxides, Z. Lebensm.-Unters.-Forsch.
1976 161, 223-230.
Wilkinson, F.; and J.G. Brummer. Rate Constants for the Decay and Reactions of the Lowest
Excited State of Molecular Oxygen in Solution. J. Phys. Chem. Ref. Data 1981 10, 809-
899.
Wolff, C.; and H.T. Witt. On Metastable States of Carotenoids in Primary Events of Photo-
synthesis. Z. Naturforschg 1969 24B, 1031-1037.
Woodall, A.A.; G. Britton; and M.J. Jackson. Antioxidant Activity of Carotenoids in Phos-
phatidylcholine Vesicles: Chemical and Structural Considerations. Biochem. Soc. Trans.
1995 23, 133S.
Yamauchi, R.; N. Miyake; H. Inoue; and K. Kato. Products Formed by Peroxy Radical Oxida-
tion of β-Carotene. J. Agric. Food Chem. 1993 41, 708-713.
180 A. Kamal-Eldin
Zechmeister, L.; A.L. LeRosen; W.A. Schroeder; A. Polgar; and L. Pauling. Spectral Character-
istics and Configuration of Some Stereoisomeric Carotenoids Including Prolycopene and
Pro-γ-Carotene. J. Am. Chem. Soc. 1943 65, 1940-1951.
Zechmeister, L. cis-trans Isomeric Carotenoid Pigments. In Progress in The Chemistry of Natural
Products, vol. 18, Zechmeister, L. Ed., Springer-Verlag: Vienna, Austria, 1960, pp. 223-
239.
Zinsou, C. Degradation Enzymatique de β-Carotene. Physiol. Veg. 1971 9, 149-167.
8
Co-oxidation of Proteins by Oxidizing
Lipids
Karen M. Schaich
Department. of Food Science, Rutgers University, 65 Dudley Rd., New Brunswick, NJ,
08901-8520
Introduction and Overview of Macromolecular Damage

Lipid oxidation is the chemical reaction that most limits shelf life of foods, and it is
increasingly being recognized as a major contributor to oxidative damage in vivo. In
foods, the most obvious indicators of lipid oxidation or “rancidity” are off-odors and
off-flavors that arise directly from lipid oxidation products, particularly aldehydes. Less
recognized, but perhaps even more important in both foods and biological systems, are
the co-oxidations that broadcast oxidative damage from lipids to all kinds of molecules
via reactions of lipid free radicals in early stages of oxidation and later reactions of
product epoxides and carbonyls as oxidation progresses. The critical effect of co-oxida-
tions is to redirect the damage processes and confuse the reaction picture, quenching
propagation of lipid free radical chains and decreasing apparent lipid oxidation (as
commonly measured), but leaving in its place footprints of lipid oxidation in damage
to proteins, DNA, and other molecules.
Damage to proteins from oxidizing lipids has been recognized for decades, but per-
spectives have changed with improvements in analytical capabilities. Most early studies
were rather global in approach and primarily reported changes in general behaviors
(texture, crosslinking, scission, loss of nutritional value), molecular functionality (en-
zyme activity, browning, and color fading in foods), and alterations in molecular func-
tion such as cell signaling, gene response, and apoptosis in vivo (Fig. 8.1) with only cur-
sory, if any, determination of mechanisms. Model system studies with isolated amino
acids provided more information about potential damage processes, but these are not
always perfect models for reactions in proteins. In the past ten years there has been
a marked shift to more sensitive analysis of reaction products facilitated by advances
in chromatography in tandem with mass spectrometry, development of immobilized
enzymes that hydrolyze proteins without destroying individual modified amino acids,
and increased use of immunological methods to track specific changes of intact pro-
teins in tissues. In some ways, recent gains in understanding using these methods have
only raised more questions about the exact role of lipid oxidation products in cellular
cytotoxicity and the detailed mechanisms involved, both direct and from co-oxida-
tions, and relative to other oxidants such as hydroxyl radicals.
181
182 K.M. Schaich
Proteins
Oxidation Nucleic acids
Flavors/Off-flavors Oxidation
Crosslinks 8-HOdG production
Scission Strand-breakage
Browning LIPID Gene modification
Lipofuschin formation OXIDATION Cell death
Function loss ? Toxic products
Nutrition loss
Cell signaling
? Toxic products
Starch Vitamins / Pigments

Scission Oxidation
Flavors Color loss
Function loss Function loss
? Toxic products Nutrition loss
? Toxic products
Fig. 8.1. Types of damage that occur when oxidizing lipids co-oxidize cellular molecules in foods and living tissues
(plant and animal).
Although it is clear that co-oxidations by lipids do occur and contribute to protein

deterioration, quantitative and qualitative cause-effect relationships remain poorly es-
tablished except in some model systems. Until recently, lipid hydroperoxides and/or
aldehydes were always measured when determining lipid oxidation involvement in
food quality loss or in pathology, while their co-oxidation products have been largely
ignored. The problem of identifying lipid oxidation damage is confounded further
in that oxidizing lipids induce nearly the same protein oxidations and react with the
same amino acids as hydroxyl radicals (Davies 1987a, 1987b). As a consequence, it
is very likely that the true extent of lipid oxidation in both foods and tissues is con-
sistently underestimated and macromolecular damage, especially in vivo, is routinely
attributed to other oxidants (e.g., hydroxyl radical, HO•).
Reactions of oxidizing lipids with proteins are important to food science, chem-
istry, biochemistry, and medicine, and research approaches taken by each field have
varied tremendously. Information gained by each of these fields needs to be integrated
into a comprehensive picture of the chemistry, causes, and effects of protein co-oxida-
tion by lipids. To begin the process and to encourage broader consideration and more
accurate determination of the impact of oxidizing lipids in complex systems, this
chapter links commonly observed macromolecular modifications in protein proper-
ties to the chemistry of four classes of lipid oxidation products—free radicals, hydro-
peroxides, epoxides, and carbonyl (mostly aldehyde) secondary products—differen-
tiating the types of reactions and patterns of damage induced by each oxidant and
Co-oxidation of Proteins by Oxidizing Lipids 183
showing where such reactions occur in some specific food applications and pathologi-
cal conditions. Due to space limitations, most mechanistic details of lipid reactions
with proteins, as well as specific reactions and products of individual amino acids, will
be presented in a subsequent publication.
Causative Agents in Molecular Damage to Proteins

from Oxidizing Lipids
When analyzing lipid oxidation in either foods or physiological tissues, we clearly
need to look beyond peroxide and TBARS values to include alternate lipid path-
ways (Schaich, 2005) and co-oxidation products (Schaich, 1980b; Borg and Schaich,
1984). As noted previously, oxidizing lipids generate multiple reactive species, all of
which have potential to react with non-lipid molecules:
• Peroxyl and alkoxyl radicals transfer radicals to other molecules, leading to

crosslinking and polymerization, molecular scissions, and a variety of co-
oxidations.
• Hydroperoxides hydrogen bond to proteins and nucleic acids, leading to induced

decomposition in situ and subsequent H abstraction from or lipid radical
addition to protein sites.
• Epoxides bind to proteins, forming adducts.
• Carbonyl products, particularly aldehydes, participate in a variety of addition

reactions leading to the formation of adducts, crosslinking of macromolecules,
fluorescent products, and browning. Both saturated and unsaturated aldehydes
also autoxidize and lead to free radical damage.
Because of the dynamic nature of lipid oxidation, these four classes of lipid oxida-
tion products react independently, and sequentially. In rapidly oxidizing or extensive-
ly oxidized systems, they can react simultaneously. It is both dishonest and inaccurate
to claim that any one of these oxidants is the major damaging agent. The apparent
dominant oxidant changes with the reaction system, the conditions for oxidation, the
timing of the reaction before analysis, the methodology used to detect both lipid and
co-oxidation products, and the endogenous processes that may degrade or remove
some products. Most attention has been given to aldehyde reactions, particularly in
medical applications, because individual aldehydes can be isolated and reacted with
proteins in targeted studies, and the adduct products tend to be more stable and less
rapidly cleared in vivo. As a result, the reactions are easier to follow. Lipid radical and
hydroperoxide reactions have not been studied with the same intensity or scrutiny
because the reaction systems are less easy to define and control, the intermediates and
products are constantly transforming, and they are more difficult to track and isolate.
Pathways to clear minimally oxidized proteins in vivo are known, but the long-term
184 K.M. Schaich
effects of replacing hydrophilic oxidants, such as hydroxyl radicals (HO•), with hy-
drophobic oxidants from lipids has not been evaluated.
The discussion of lipid oxidants and associated damage presented in the rest of
this chapter is intended to demonstrate the complexity of lipid co-oxidations of pro-
teins and encourage research that investigates beyond individual lipid oxidants and
considers the dependence of protein oxidation on reaction system and solvent, reac-
tion time, concentration of lipid oxidation products, and other factors before assign-
ing causality. In most cases, protein co-oxidations in foods and tissues involve mul-
tiple lipid oxidants that generate a wide range of protein products. The task for the
future is to decipher the pathways that sequentially or simultaneously lead to each.
Although presented first, the damage reactions reviewed in this chapter were ac-
tually elucidated after observations of global effects as a means to explain why and
how amino acid destruction, crosslinking, formation of fluorescent products, and loss
of enzyme and other functional activities occur. Some connections of basic chemistry
to specific proteins, foods, or pathological conditions are made along the way to pro-
vide orientation and practical grounding. However, the main purpose of this section
is to detail the various reactions of lipid oxidants that lead to co-oxidation of proteins
to illustrate their similarities and differences and to provide a conceptual framework
for explaining protein changes. The global effects on proteins discussed in the section
“Reactions Underlying Molecular Damage to Proteins” are all mediated by each of the
four lipid oxidants. As will be shown, much of the detailed localized chemistry from
this section is repeated in each effect—just the consequences of the chemistry differ
depending on the protein and the reaction conditions.
Transfer of Lipid Radicals to Proteins

Evidence
The first reactive lipid species to form are free radicals, and as long as the lipids are
in close proximity to susceptible protein sites, free radical transfer to proteins can
occur rapidly, either by hydrogen abstraction or radical addition. Electron paramag-
netic resonance (EPR or ESR, electron spin resonance) first provided direct evidence
that oxidizing lipids transfer free radicals to proteins in pilot studies by Roubal and
Tappel (Roubal and Tappel, 1966a; Roubal, 1970), then in more definitive studies
by Schaich and Karel (1975,1976; Schaich, 1980b). Dry systems were used to stabi-
lize protein radicals, eliminate interference of water, and facilitate radical detection.
Starting with unoxidized lipid, EPR signal intensity increased just behind increase in
peroxide values (Schaich and Karel, 1975). When pre-oxidized lipid was mixed with
powdered proteins, radical transfer began essentially instantaneously and detectable
signals developed in minutes (Schaich, unpublished data). EPR signals show a strong
singlet center line with g values of 2.0035–2.0060 (Figures 8.2 and 8.3). Lower g-
values in this range are typical of N•; higher g-values are more likely NO• (nitrogens
are immobilized so expected triplet signals are not observed). For non-sulfhydryl pro-
teins, spectra show little structure and total spectrum widths are narrower (~50 G)
(Fig. 8.2). Spectra from sulfur proteins are broader (~75 G), showing strong high field
peaks with g ~ 2.015 and 2.023 from sulfinyl (RSO•) and/or sulfonyl (RSOO•) radi-
cals, consistent with formation of various sulfur oxide products (Finley and Lundin,
1980). Shoulders (g ~ 2.001) from alkoxyl or peroxyl radicals also are evident in some
spectra (Fig. 8.3).
What do EPR spectra reveal about lipid radical transfer sites? Clearly, thiol (SH)
groups on cysteine are major targets for radical transfer from lipids. Cysteine reacted
with oxidizing lipid forms the same sulfur radicals as in proteins, but the g ~ 2.015,
2.023 peaks are now dominant; some H abstraction from the a-amino group is also
evident (Fig. 8.4). However, an obligate prerequisite for reaction is accessibility. Thi-
ols on the protein surface react with lipids very rapidly at very low levels of oxidation;
sulfur radical peaks are the first detected in protein EPR spectra, and they continue to
Fig. 8.2. Epr signals of free radicals induced in non-sulfhydryl proteins by reaction with oxidizing methyl linole-
ate in dry model systems. Lactalbumin: solid line, 2 mW power; dotted line, 20 mW power, showing minimal
contributions from radicals with different saturation characteristics. Scale for all spectra as indicated for casein.
Adapted from (Schaich and Karel, 1976).
Fig. 8.3. Epr signals of free radicals induced in sulfhydryl proteins by reaction with oxidizing methyl linoleate in
dry model systems. Scale for all spectra as indicated for denatured lysozyme. Bovine serum albumin: solid line,
2 mW power; dotted line, 20 mW power, enhancing peroxyl and sulfoxyl signal detection at higher power and
providing clear evidence of multiple radical centers. Adapted from (Schaich and Karel, 1976).
186 K.M. Schaich
Fig. 8.4. EPR spectra of free radicals transferred from 2.025

oxidizing methyl linoleate to cystine (top) and cyste- 2.034 2.013
ine (bottom) in dry model systems. Cysteine reacted 2.060
for 1 day; cystine reacted for 6 days at 37°C. Adapted
from (Schaich and Karel, 1976). Similar spectra are 2.0039
produced in emulsions (Schaich, 1980b).
Cystine
2.015
50 gauss
2.020
2.0060
Cysteine
increase over time if incubation with the lipid denatures the protein and new cysteine
residues become available (Schaich, 1980b). In contrast, oxidizing lipids do not react
readily with buried –SH or with cystine double bonds. S–S radicals clearly present in
irradiated proteins are not produced by oxidizing lipids, even after proteins are dena-
tured; oxidizing lipids attack the S-S in cystine only after long incubation times (Fig.
8.4) (Schaich and Karel, 1976; Schaich 1980b). Although the possibility of electron
migration to disulfides from other sites can not be ruled out, the main radical with
g = 2.055, 2.024, and 1.99 is most likely a R-S-S• radical formed by abstraction of
a hydrogen on the carbon a or b to one S, as has been shown for t-butoxyl radicals
(Adams, 1970) and hydroxyl radicals (Elliot et al., 1981):
Other unidentified radical species are also present.

The broad central envelopes of proteins reacted with oxidizing lipids reflect over-
lapping unresolved hyperfine structures from multiple radical sites and delocalization
of free electrons on the peptide backbone. Although steric hindrance limits accessibil-
ity of peptide a-carbons to lipid alkoxyl and peroxyl radicals, amines and thiols on
amino acid side chains on the protein surface provide ready sources of hydrogen for
abstraction. Histidine, tryptophan, arginine, lysine, and cysteine all produce stable
radicals when incubated alone with oxidizing linoleic acid (Fig. 8.5); in proteins, their
signals combine to produce the broad envelopes observed in EPR signals. Proteins
that have more open structure and higher concentrations of reactive side chains on
their surface show more hyperfine in their EPR signals, as can be seen with ovalbu-
min, serum albumin, and lactalbumin in Fig. 8.3, while highly structured proteins,
such as casein, have narrower signals with little structure (Fig. 8.2). The signal wings
also show more intensity and structure in proteins reacted after denaturation increases
side chain accessibility (e.g., lysozyme in Fig. 8.3), consistent with increases in dam-
age noted for denatured proteins.
Recent studies reported similar results for lysozyme, ovalbumin, arginine, ly-
Lysine Histidine Tryptophan Arginine
2.0039 2.0027
2.0033 2.0046
25 gauss
Fig. 8.5. Epr signals of amino acids reacted seven days with oxidizing methyl linoleate in dry lyophilized emul-
sions. g-values of center lines indicate N-centered radicals for lysine, histidine, and tryptophan; in arginine the
radical is on a terminal amine or the carbon connected to it. Adapted from (Schaich and Karel, 1976).
sine, and histidine reacted with oxidizing lipids in lyophilized emulsions (Saeed et
al., 1999). However, fish myosin signals had lower g-values (2.0021) consistent with
carbon rather than nitrogen-centered radicals. This is somewhat puzzling considering
myosin’s high content of arginine and lysine (Connell and Howgate, 1959), which
should generate N-centered radicals. Three speculative explanations may be offered
for this difference in behavior from other proteins.
First, the normally reactive amino acid side chains may be inaccessible. Myosin
is highly organized in a coiled coil double helix that does not denature readily; in
addition, fish myosin aggregates and exhibits extensive loss of solubility with freez-
ing (Connell, 1960; Buttkus, 1970) and lyophilization (Huss, 1995). Second, fish
myosin has a high content of glutamic acid, leucine, and valine which may offer
alternative radical attack sites that would produce C• radicals—glutamic acid by de-
carboxylation and aliphatic amino acids by H abstraction from side chains. Finally,
myosin, like collagen and gelatin, is a structural protein with high glycine and proline
188 K.M. Schaich
content, providing a-C• sites for radical localization on the peptide backbone; thus it
may be more prone to peptide scission than other proteins, especially in a dry reac-
tion system. Scission would generate C• radicals by reactions that will be described
later in this chapter. Electrophoresis and amino acid analyses conducted over time on
the damaged myosin may shed some light on dominant reaction mechanisms with
structural proteins.
It must be stressed that EPR only detects radicals that are sufficiently long-lived
to maintain a steady-state radical concentration of about 1013 spins/sec. Consequent-
ly, there is the possibility that lipid radicals react with protein sites other than the ones
noted previously, but the resulting radicals are too short lived to detect or they react
or convert too fast to non-radical products. Thus, lack of EPR signals alone cannot be
considered proof that radical reactions do not occur with other amino acids, especially
serine, threonine, and the aliphatic amino acids that are also destroyed by oxidizing
lipids.
Consequences of Free Radical Transfer to Proteins

Clearly, protein free radicals are red flags for oxidative damage. Regardless of the
transfer site or mechanism, radical transfer from lipid oxyl radicals, LOO• and LO•,
to proteins by H abstraction generates protein radicals (P•) with consequences that
vary with the protein and the reaction system (Fig. 8.6). Protein hydroperoxides (Da-
vies et al., 1995) and all of the protein radical species shown in Fig. 8.6 are reactive.
They oxidize to secondary products that alter protein conformation and solubility and
inhibit enzyme activity, recombine to form polymers, add oxygen to form peroxyl
radicals on both peptide backbones and amino acid side chains, and they undergo
Fig. 8.6. Pathways and consequences of free radical production in proteins. P• is a protein radi-
cal on any a-carbon of the main peptide backbone or on an amino acid side chain, and rh is
any molecule with abstractable hydrogens.
scission reactions with decarboxylation; under some conditions they also show deami-
nation in patterns similar to irradiated proteins (Davies, 2005). These changes will
be discussed in more detail in the Section “Reactions Underlying Molecular Damage
to Proteins”.
Protein radicals also transfer radicals to other proteins (Soszylqski et al., 1996),
DNA (Gebicki and Gebicki, 1999; Luxford et al., 2000), lipids (Gardner and Weisled-
er, 1976; Gardner et al., 1977; Avdulov et al., 1997) and potentially other molecules
to broadcast and perpetuate oxidative damage. One example of this is the addition of
cysteine radicals to methyl linoleate hydroperoxide in the presence of FeCl3 (Gardner
and Weisleder, 1976; Gardner et al., 1977). The same H abstraction reaction that
quenches a LO• also generates a thiyl radical RS• which adds to the double bond of
the lipid to generate a new radical on the adjacent carbon. Continued H donation
from cysteine maintains a constant supply of RS• for more addition reactions, thus
perpetuating and expanding the oxidative damage.
The downstream molecular and functional degradation that ensues has significant
potential for dramatic effects on food quality and is increasingly being recognized as
mediators of pathological processes in vivo. Similar addition of cysteine or glutathi-
one to prostaglandin A1 has been observed in model systems (Ham et al., 1975) and
in red blood cells in vivo (Cagen et al., 1976), presumably by free radicals.
Radical transfer occurs early in lipid oxidation and in effect is an antioxidant
process for lipids. Paradoxically, lipid oxidation may appear to be low when radical
transfer to proteins is high. As a consequence, co-oxidation effects of lipid radicals too
often are missed or misinterpreted.
Reactions of Lipid Hydroperoxides

Lipid hydroperoxides are not reactive species per se, unlike lipid radicals, epoxides,
and carbonyl oxidation products. Nevertheless, proteins incubated with lipid hydro-
peroxides are damaged within minutes, so possible roles for hydroperoxides beyond
decomposition to free, diffusible alkoxyl radicals must be considered. Under some
circumstances, kinetic analyses argue for the presence of concerted lipid hydroperox-
ide–protein reactions in which there is induced decomposition of LOOH and direct
reaction of the resulting LO(O)• with amino acid targets within a reaction cage. The
process may be metal mediated or metal independent.
190 K.M. Schaich
Metal contaminants, particularly iron and copper that are always present in tis-
sues and lab reagents, decompose lipid hydroperoxides in solution and release free
LO(O)•. Metal reactions in solution are anticipated and even used in model system
studies to enhance reaction rates. Cage reactions of metals, however, are more damag-
ing (Fig. 8.7). The unusual sensitivity of metallo-proteins is due in part to binding
and reduction of LOOH in a reaction cage, leading to oxidation of amino acids
near the ligand site, particularly histidine (Kowalik-Jankowska et al., 2004). Most
non-metalloproteins also have metal-binding sites, for example on histidine, glutamic
acid, or aspartic acid side chains, that can serve as foci for metal-catalyzed reduction of
LOOH in cage reactions on protein surfaces. In support of this concept, most of the
iron in buffered solutions of b-lactoglobulin oxidized by methyl linoleate was bound
to the protein (Yuan et al., 2007).
When metals are in solution rather than on the protein, reducing agents must
be present to cycle metals; lipid peroxide reduction rates, radical lifetimes, migration,
PROTEIN
SH NH2
…LO−OH
2+
Fe …HO−OL
NH2 Cu+
SH
Fig. 8.7. Looh-induced radical transfer to proteins: two forms of cage reactions in which lipid hydroperoxides are
bound in close proximity to a metal and reduced in situ; the resulting LO• then abstracts a hydrogen from a sus-
ceptible amino acid nearby before release. Such facilitated LOOH reduction is evident in faster transfer of radicals
to proteins and increased rates of lipid oxidation (chain propagation by LO• >>LOO•).
and contact with the protein also become limiting critical issues, and radical transfers
to protein tend to slow down. Under some conditions, damage from lipid hydroper-
oxides to proteins occurs so rapidly and at such low concentrations, even in reactions
with demetalled amino acids (Schaich, unpublished data), that metal-independent
mechanisms must also be operating under many conditions. Molecule-assisted ho-
molysis (MAH) of hydroperoxides is well known (Pryor, 1966). In MAH, hydro-
peroxides hydrogen bond to protein sites that induce LOOH decomposition, and
released LO(O)• radicals abstract hydrogens from nearby amino acids in a reaction
cage. Noting that the kinetics of radical transfer to lysozyme exceeded the apparent
rates of oxidation of methyl linoleate, Schaich and Karel proposed protein-facilitated
decomposition of LOOH and cage transfer of radicals that enhance the reactivity of
LOOH (Karel et al., 1975; Schaich and Karel, 1976):
Solid state EPR signals showed N(O)• and S(O)• radicals, which are consistent with
observations that amines (Harris and Olcott, 1966) and sulfhydryls (Little and
O’Brien, 1968) both undergo concerted reactions with [lipid] hydroperoxides:
This kind of concerted reaction may contribute to the sensitivity of his, arg,
lys, trp, cys, ser, and thr to LOOH, all of which contain hydrogen bonding amino,
carboxylic acid, and hydroxyl groups on their side chains (Gardner 1979, Schaich
1980b). It also may be enhanced in lipid-bonding proteins, such as bovine serum
albumin, where hydrophobic side chains facilitate associations with lipids and bring
reactive residues into close proximity with LOOH.
Concerted reactions between LOOH and proteins have now been demonstrated
for:
a. b-lactalbumin reacted with LOOH, LnOOH, and AnOOH, where one

mol LOOH per 18,000 mw protein oxidized tryptophan and cysteine and
generated extensive crosslinking within one hour (Hidalgo and Kinsella,
1989);
b. HDL apoA1 and A2 reacted with cholesterol and phospholipid

hydroperoxides, where appearance of met oxidized to methionine sulfoxide
occurred concurrently with LOOH reduction (Garner et al., 1998a).
c. LDL and cyt c incubated with LOOH or phospholipid (PE) hydroperoxides,

where loss of LOOH led to reduction of cyt c, release of O2– •, and generation
of fluorescent products (Fruebis et al., 1992). The concerted reaction
mechanisms involving hydroperoxide (peroxyl radical) addition proposed
to explain the kinetics and products is shown in Fig. 8.8. Pathway A has
parallels in radiation and HO• chemistry. Pathway B is sterically unlikely in
linoleic or linolenic acid but could be feasible in arachidonic acid and w-3
fatty acids. Both mechanisms merit further testing.
Although mechanisms were not identified, very rapid protein degradation sug-
gests that concerted reactions were involved in substantial losses of trp, met, cyh,
pro, val, leu, as well as fragmentation and crosslinking of lupine conglutins with
MLOOH at pH 9 (Fruebis et al., 1992; Lqari et al., 2003), in reactions of butylamine
with LOOH in CHCl3 (Zamora and Hidalgo, 1995), and in the very rapid forma-
tion of protein carbonyls and loss of lysine without lipid aldehydes when MLOOH,
LnOOH, and AnOOH were incubated with BSA (Refsgaard et al., 2000). The latter
study is also an excellent example of the broadcasting action of lipid oxidation; chain
initiation was not affected, but yields of downstream lipid oxidation products were
substantially depressed when oxidation was transferred to BSA. Again, this shows that
if only normally measured lipid oxidation products are monitored in complex systems
192 K.M. Schaich
Fig. 8.8. Two reaction pathways proposed to explain concerted reactions of lipid hydroperoxides with proteins
(shown for reaction of 13-HOO-linoleic acid, LOOH, with protein-bound lysines). Modified from (Fruebis et al.,
1992): original reaction scheme assumed that LOOH (generated from lipoxygenase and purified before incubation
with proteins) was converted to LOO• before reaction with amino groups on proteins. However, EDTA-complexed
iron and small amounts of unoxidized linoleic acid could not account for the large concentrations of LOO• needed
to drive the rapid reaction. Replacing LOO• with LOOH gives reactions that are consistent with radiation and HO•
chemistry, O2– • equilibrium (at neutral pH, O2– • HO2•), and still account for observed products.
and footprints of co-oxidations are ignored, the true extent of oxidative degradation
can be greatly underestimated.
Reactions of Lipid Epoxides

It is interesting that lipid epoxides are carcinogenic (Chung et al., 1993; Blair, 2001;
Lee et al., 2001), mutagenic (Lee et al., 2002), and strongly cytotoxic; yet their re-
actions with proteins have received relatively little attention. Epoxides (also called
oxiranes) are cyclic products generated by internal reactions of lipid hydroperoxides
(Hamberg and Gotthammar, 1973), peroxyl addition products or alkoxyl radicals
(Gardner, 1989; Schaich, 2005), or reaction between hydroxynonenal and lipid hy-
droperoxides or hydrogen peroxide (Chen and Chung, 1996):
4,5-epoxy-2-decenal has been reported as a common secondary product of w-6 fatty

acids (e.g., initial oxidation at C-13 of linoleic acid, cyclization of alkoxyl radical to
form epoxide, secondary oxidation at C-9, a-scission of alkoxyl radical to form alde-
hyde); 4,5-epoxy-2-heptenal is an epoxide from w-3 fatty acids (e.g., initial oxidation
at C-16 of a-linolenic acid, cyclization of alkoxyl radical to form epoxide, secondary
oxidation at C-12, a-scission of alkoxyl radical to form aldehyde) (Gardner, 1989).
It is interesting that these reactions not only generate a new oxidant species but
also shift the inherent macromolecular damage capability of oxidizing lipids as the
product mix changes. Internal cyclizations or rearrangements of LO(O)• to epoxides
are always present, competing with H abstraction during lipid oxidation. Cyclization
is favored when abstractable protons are limited (i.e., at low lipid concentrations), in
aprotic solvents, in neat unsaturated lipids oriented on surfaces, and at lower temper-
atures (Schaich, 2005). Thus, rather than being intriguing reaction artifacts or trace
side products, epoxides are common components of most complex reaction systems,
especially with multiple phases. As very reactive compounds, epoxides are probably
the most underrated and understudied lipid-derived oxidant.
What accounts for the marked reactivity of epoxides, especially with proteins
and nucleic acids? Epoxy functions vicinal to olefin double bonds are particularly sus-
ceptible to hydrolysis and nucleophilic attack (Lederer, 1996). When association of
a nucleophile with the epoxide generates a partially charged transition state (Ingold,
1969), the oxirane ring then opens easily because the three-membered ring is strained
and at a higher energy level. The oxygen stays on the less highly substituted carbon
and the nucleophile adds to the opposite carbon (the allylic carbon when there is an
adjacent double bond) from the backside (Ege, 1999; McMurray, 2000). This basic
process is shown below for reaction of lysine, histidine, and cysteine (left to right)
with a hypothetical isolated epoxide from oxidation of linoleic acid.
194 K.M. Schaich
Model system studies of lipid epoxides reactions have revealed some important
characteristics. Lederer used epoxyhexenol (trans-4,5-epoxy-trans-2-hexen-1-ol) as a
simple model for the reactive region of oxidized fatty acids and propylamine as a
model for lysine, eliminating competing reactions with carboxylic acid and amine
groups (Lederer, 1996). When the reaction was run in an aprotic solvent (THF) to
avoid nucleophilic attack of solvent molecules, the amine added exclusively at C-2
(the allylic position) by an SN2 mechanism to form an aminol as shown previously. In
aqueous solutions, aminols formed in >50% yields at pH 9 and production decreased
dramatically as pH decreased: only 3% aminols were formed after 24 hours at physi-
ological pH, and there was no reaction at pH ≤ 6. Acid reduces the nucleophilicity of
the lysine e-amino function and protonates the epoxide, facilitating C-O bond scis-
sion and release of –OH from the oxirane ring; the amine can then add to C-2 in clas-
sical nucleophilic addition (A) or to C-4 in the double bond in an SN1-like reaction
(B), as shown below (Ege, 1999; McMurray, 2000). At the same time, protonation
of the epoxide also increases epoxide hydrolysis (C) to rates competitive with amine
addition (Lederer, 1996):
What factors direct the balance between these three reactions and products? Ob-
viously, solvent influences are critical. Lysine with blocked amino and carboxyl groups
showed classical nucleophilic addition (A) to 9,10-epoxy-13-hydroxy-11 octadecenoic
acid t-butyl ester (Lederer et al., 1998). The reaction was slow in methyl pyrrolidone,
due to limited availability of solvent protons, but the rate increased dramatically with
addition of up to 20% water (the solubility limit). As expected water also increased
hydrolysis rates (C) competitively, but it did not change the addition mechanism; in
all cases, the exclusive reaction was backside nucleophilic attack at the allylic position
yielding the aminol adduct (A) with four diastereomers (Lederer, 1996). These results
suggest that epoxide reactions with proteins are most important under anhydrous
conditions, for example in dry foods and in hydrophobic interior regions of biomem-
branes and blood lipoproteins.
The nature of the nucleophile and its conformational accessibility also affect reac-
tivity and pathways. When reacted with butadiene moloxide in phosphate buffer (pH
7.4), the a-amino group of valine added at both epoxide carbons in 2:1 ratio, C2:C1
(terminal C) (Moll, 1999):
C OOH C OOH COOH
+ +
O
NH2 NH2 NH2
HO
OH
(C-1 adduct) (C-2 adduct)
In intact proteins, such as mouse hemoglobin, reaction of N-terminal valines oc-
curred almost totally at C-1 due to steric hindrance. Epoxide adducts also formed
with lysine, serine, histidine, and methionine, although the regioisomer could not
be distinguished by ESI-MS (Moll et al., 2000). In all systems, higher temperatures
(37°C) increased hydrolysis at the expense of amine adducts. Similarly, LC-MS-MS
analyses of human hemoglobin reacted with styrene oxide, ethylene oxide, and buta-
diene dioxide revealed major epoxide addition sites were the N-terminal valines of
both a and b Hb chains, plus cysteine and histidine of specific sequences (Badghisi
and Liebler, 2002); adduct structures were not specified.
Epoxide addition by this classical mechanism is probably responsible for un-
characterized binding of 9,10-epoxy stearic acid to albumin reported thirty years ago
(Pokorny et al., 1966); C-2 addition to epoxides has been demonstrated with sulf-
hydryl compounds (Buttkus, 1972) and cysteine (Gardner et al., 1977; Gardner and
Jursinic, 1981).
When the epoxide is in an alkenal rather than a fatty acid, the reaction changes
due to competition with Schiff base formation at the carbonyl (Zamora and Hidalgo,
1995, 2003b, 2005; Zamora et al., 1999; Hidalgo and Zamora, 2000). In chloro-
form, 70% acetonitrile, or aqueous methanol, histidine reacted with 4,5-epoxy-2-
alkenals by classical epoxide addition (A, above), accompanied by a parallel increase
in protein carbonyls. However, with primary amines, such as lysine and ethyl amine,
Schiff base adducts preferentially formed with the aldehyde and the epoxide group
remained intact. The imine was added to the epoxides via backside attack, forming a
pyrrole ring with oxyl side chain. However, without a solvent or amine proton source,
the epoxide oxygen remained reactive as an anion (-O–) rather than forming the more
stable alcohol (-OH). This intermediate then transforms to two sets of products: (1)
the hydroxide anion protonates, giving a hydroxyalkyl pyrrole, and (2) the side chain
is released as an aldehyde, for example formaldehyde or acetaldehyde (dominant reac-
196 K.M. Schaich
tion with a-amino acids), leaving an N-substituted pyrrole. Hydroxyalkyl pyrroles

polymerize over time (see Section “Crosslinking” for reactions).
Depending on the amine and temperature, two other reaction pathways have been
observed with epoxy-alkenals (2:1 acetonitrile-water or buffer as solvent) (Zamora
and Hidalgo, 2005; Zamora et al., 2006):
a. direct amine addition to the epoxides → alkyl-substituted pyrroles and

furans. The rate of this reaction increases 10-fold when reaction temperature
is changed from 37 to 60°C.
b. Schiff base imine formation followed by elimination of the amine R’ and

cyclization of remaining product to alkylpyridines
An alternate mechanism for formation of 1,2-dihydropyridines is C-4 addition

of imines to vinyl epoxides in presence of a Lewis acid (e.g., water) (Brunner et al.,
2006). Analogous products should be expected in reactions of epoxyalkenals with
amino acids:
Aminols from epoxide-amino acid reactions have distinctive structures that can
differentiate the oxidation from free radicals, hydroperoxides, and carbonyl products.
However, variously substituted pyridines and pyrroles are also products of extensively
studied aldehydes, as will be detailed in the next section. Thus, it can be exceedingly
difficult to distinguish epoxides from aldehyde reactions, except perhaps by kinetics,
and it is quite likely that reactions of epoxides with proteins have been overlooked
and misinterpreted as aldehyde-mediated damage. New analytical methods will be
necessary to track the lipid oxidants responsible for common protein products under
various reaction conditions and determine the full role of epoxides in protein degra-
dation.
Addition Reactions of Secondary Products from Lipid Oxidation

Secondary lipid oxidation products are responsible for the off-odors and flavors as-
sociated with rancidity and they also appear to be long lived in biological tissues, so
they have received considerable attention in co-oxidation studies. Most research has
focused on aldehyde reactions, even though scission reactions of lipid alkoxyl radicals
generate a wide variety of reactive secondary oxidation products.
All aldehydes react with nucleophilic groups on proteins to form adducts, with
three general outcomes:
a. Linear adduct formation (Schiff base, Michael addition, or a combination

of both) that changes surface chemistry and protein recognition. This is the
initial step for all aldehydes.
b. Cyclic products, especially dihydropyridines and pyrroles, as adducts or in

crosslinks.
c. Protein crosslinks, both intra- and inter-molecular, via Schiff base,

combinations of Schiff base and Michael additions, Schiff base-Michael
addition-ring links, or a complex combination of any of these.
The reaction pathways that occur or dominate in a given system are influenced by the
nature of the protein, relative protein-aldehyde concentrations, pH, phase or solvent,
oxygen tension, and other factors.
Monofunctional alkanals
Monofunctional alkanals (Fig. 8.9) have low reactivity and high selectivity; they re-
198 K.M. Schaich
Fig. 8.9. Structures of aldehydes that play important roles in oxidative degradation of proteins.
act with amines exclusively by Schiff base formation with preference for N-terminal
residues, and at low aldehyde concentrations and low pO2 the reaction often goes no
farther. For example, hexanal reacts with insulin B chain only at N-terminal phenylal-
anine residues and a single lysine (lys29) near the end of the peptide chain (Fenaille et
al., 2003). Nonanal and its oxidation product 8-oxononanoic acid form Schiff bases
with both the amine and thiol groups of cysteine to yield thiazolidine dicarboxylic
acid derivatives (Gardner et al., 1977; Gardner and Jursinic, 1981):
O
C H3O
H
¨ − H2O
S-
CH3OOC-(CH2)7CHO + N-H
C
¨ CO O H O
H
80% CH3OH
2
C
H
NH 2
S O C H3
Thiazolidine carboxylic acid derivative
Gardner and Jursinic proposed this as a generalizable model for reaction of low
levels of lipid aldehydes with N-terminal amino acids of proteins (Gardner and Jur-
sinic, 1981). When the N-terminal residue has side chains containing amines in place
of the thiol, various analogous heterocyclic products could form, with the ring size
determined by the specific side chain.
H 2N
R N
•• R
CH3OOC-(CH2)7CHO +
•• C C OOH C H 3 O O C - (CH2 ) 7 C
C C OO H
H 2N N
In the presence of excess aldehyde, three molecules of alkanal form pyridines on reac-
tion with amino acids (Suyama and Adachi, 1979). Reaction details are presented in
the “Bifunctional Saturated Aldehydes” section.
In the literature, Schiff base adducts are cited indiscriminately and often incor-
rectly both as the imine (–N=CH-CH2-CHO) and the enamine (–NH-CH=CH-
Fig. 8.10. Formation of imines and enamines from schiff base additions of primary and secondary amines with
carbonyls (McMurray, 2000).
CHO). As shown in Fig. 8.10, the imine forms first with primary amines, such as
lysine, and in acidic solutions rearranges to the enamine, whereas only the enamine
forms with secondary amines, such as histidine. Unless otherwise specified, Schiff
base structures written in reactions of this chapter assume formation from primary
amines at neutral pH and thus will be shown as the imine.
Bifunctional Saturated Aldehydes

Bifunctional saturated aldehydes (e.g., glyoxal and malonaldehyde (MDA), Fig. 8.9)
are more reactive due to the second carbonyl and keto-enol tautomerism (Esterbauer
et al., 1991):
Nucleophilic double bonds facilitate Schiff base, aldol, and other additions to enol
and enolate forms of MDA, and the enol –OH is very susceptible to dehydration as
200 K.M. Schaich
well (Nair et al., 1981; Ege, 1999). MDA is most reactive in mildly acidic aqueous
solutions where its enol form (b-hydroxy acrolein) dominates. Under these condi-
tions active formation of hydrogen-bonded dimers
and a strong tendency to self-polymerize by aldol condensation, especially at high

concentrations, compete with MDA addition to proteins (Esterbauer et al., 1991)
and reduce efficiency of MDA as a protein oxidant. This is one reason why MDA is
on the low end of the scale in damage to proteins when reactivities of various alde-
hydes are compared.
MDA and other bifunctional aldehydes react by three main mechanisms:
Schiff base addition to nucleophilic groups on amino acids and proteins. As with sat-
urated aldehydes, the major reaction of MDA with proteins is via Schiff base forma-
tion. The main protein target is lysine e-amino groups; 50-60% of the lysine e-amino
are destroyed when myosin is reacted with MDA (Buttkus, 1967). Cysteine–SH,
and histidine imidazole groups on side chains are also important targets, but Schiff
base formation with these groups must compete with facile hydrogen abstractions
and rapid Michael addition reactions, respectively. Tyrosine, arginine, and tryptophan
side chains form Schiff bases more slowly except at pH < 4.2, where their a-amino
groups become the exclusive reaction sites for many amino acids (Nair et al., 1981).
Schiff base formation proceeds as shown in Fig. 8.10, but with the complication
of two carbonyls and tautomeric forms. Reaction of proteins with one aldehyde group
of MDA yields Schiff base adducts; reaction with both carbonyls generates enamine
or iminopropene crosslinks. In acid, amines add both to dicarbonyl and enol forms of
MDA and yield different products:
+/-
+/-
In a given system, the equilibrium distribution between imine versus enamine tau-
tomers is directed by the pH and hydrogen bonding capacity of the solvent (Yildiz
et al., 1998; Nazir et al., 2000). Acid increases MDA protonation and drives the
enol reaction forward, while aqueous solutions inhibit elimination of water and drive
the equilibrium towards dissociation. In general, imines are the dominant tautomer
at equilibrium under physiological conditions (neutral pH) (Burcham and Kuhan,
1996) where hydrogen bonding with the polar solvent interferes with proton transfer
within the aldehyde-amine complex (Yildiz et al., 1998; Nazir et al., 2000). Acid
and organic solvents increase intramolecular proton transfer capabilities and facilitate
conversion to enamines.
The ease with which imine versus enamine tautomers can be formed and stabi-
lized in an aldehyde-amine complex is determined by the component aldehyde; the
amine has no effect. For some aldehydes starting with cyclic 4-HO-2-enals hydroxyl
imines were the only products in any solvent, while for other aldehydes reacting with
the same amine, enamines (ketoamines) formed in varying proportions in acidified
organic solvents (chloroform and benzene) but did not appear in hydrogen-bonding
solvents (Yildiz et al., 1998; Nazir et al., 2000). Comparable studies documenting
MDA tautomer distributions and associated products under different conditions will
contribute greatly to understanding the reactivity of this aldehyde, particularly in
relation to other aldehydes.
The existence of MDA tautomeric forms with their corresponding reactivity and
Schiff base products is emphasized here because the literature is totally inconsistent
about the structures of MDA and other lipid aldehyde Schiff base structures cited.
Review of the hundreds of papers reporting lipid-protein Schiff base products reveals
imine and enamine structures, and saturated congeners as well, used interchangeably
and indiscriminately, usually with little or no consideration of the reaction condi-
tions. In many cases the structures reported were inconsistent with reaction condi-
tions and solvents employed. Researchers need to be actively cognizant of how their
reaction conditions affect the forms of MDA available for reaction and to use this
information in interpreting and publishing results. New studies are increasingly us-
ing NMR and LC-MS/MS to identify structures precisely. In the absence of specific
structural information, studies should report either the tautomer equilibrium or the
dominant imine plus specific reaction conditions.
Formation of cyclic products (dihydropyridines) with amines: Schiff base formation
is the first step in most MDA reactions with proteins. However, when a molar excess
of MDA is present, dihydropyridines form via secondary condensation of the Schiff
bases. MDA is a short molecule, so two are needed to complete the pyridine ring;
related products are generated by two molecules of MDA plus one monofunctional
aldehyde, such as acetaldehyde or formaldehyde. (Kikugawa and Ido, 1984; Nair et
al., 1988; Freeman et al., 2005):
R1
pH 7 OHC CH O
2 O H C -C H 2 -C H O + R 1 C H O + R 2 N H 2
N (1,4-dih ydrop yrid in e -3,5-d icarbaldeh yd es )

R2
When R2 is a protein, the dihydropyridine adds a reactive group with two carbonyls
to the protein surface and becomes a site primed for further reaction. Dihydropyri-
dines are flavor precursors in foods (Buttery et al., 1977; Suyama and Adachi, 1980;
Maga, 1981). Unfortunately, when bound to proteins they are not absorbed in the
gut or hydrolyzed in tissues, which translates to loss of nutritional quality (Giron-
202 K.M. Schaich
Calle et al., 2003). Dihydropyridines also form in vivo and have been identified as
adducts using immunochemical techniques (Yamada et al., 2001) and as crosslinks in
tissues (Slatter et al., 1998).
Michael addition reactions with amines: The double bond in the enol and enolate
forms of MDA provides an electrophilic site with enhanced susceptibility to nucleo-
philic reaction via Michael-type addition of the amine to the b-carbon of the double
bond. Michael-type is the most accurate term since classical Michael additions are be-
tween carbon compounds. However, the term “Michael addition” is commonly used
with amine additions to a,b-unsaturated carbonyls, so it will be used in this manner
throughout this chapter. Details of Michael addition reactions are in the following
section on unsaturated aldehydes.
Unsaturated Aldehydes
Unsaturated aldehydes, primarily 2-enals, are extraordinarily reactive compounds.
a,b-unsaturation makes three tautomers possible, two of which have carbocations
activated towards nucleophilic addition. Thus, 2-alkenals and their oxidized deriva-
tives have three potential reaction sites: Schiff base formation at the carbonyl and
Michael-type 1,2 and 1,4 addition at the carbocations (Esterbauer et al., 1991; Ege,
1999; McMurray, 2000).
..O.. ... .
2 .O .
... .
4 .O .
C C 1 C C C C+
H δ+ C H + C H 3 C 1
2
Schiff base 1,2 Michael-type addition 1,4
Michael additions to the double bond are preferred and generate the most prod-
ucts, both immediately and in subsequent transformations. The nucleophilic thiol of
cysteine, e-amine of lysine, and imidazole nitrogen of histidine are the main targets
of unsaturated aldehydes (Esterbauer et al., 1991; Petersen and Doorn, 2004). With
direct (1,2) addition, the amine (or thiol) adds to the carbonyl carbon with a,b un-
saturation, generating a carbinolamine intermediate that rearranges and dehydrates to
a Schiff’s base (only the reactive groups of the 2-alkenals are shown below) (Ege, 1999;
McMurray, 2000):
The reaction is catalyzed by acid and facilitated in organic solvents and hydrophobic
microenvironments; it is also reversible (Esterbauer et al., 1991). With most lipid
aldehydes and amino acids, it is a precursor for some minor products. An important
side-effect is the reduction of reactive carbonyls to alcohols.
Conjugated (1,4) addition is the dominant initial process in aqueous phase reac-
tion of lipid alkenals with amino acids. The amine adds to the b-carbon of the double
bond in conjugation with the carbonyl:
Importantly, the carbonyl remains intact and can contribute to carbonyls detected in
oxidized proteins or react further with an additional amine to form a Schiff base. Mi-
chael addition followed by Schiff base formation is the initial sequence found repeat-
edly in the formation of complex adducts (shown in following sections). Like Schiff
base formation, Michael additions are reversible in an aqueous buffer without excess
amine to prevent hydrolysis and aldol condensation of C=O (Nadkarni and Sayre,
1995). Aqueous solvents favor 1:1 amine:C=O adduct formation, while organic sol-
vents favor 2:1 complexes.
Adding to the complexity of alkenal reactions with amines is the marked tenden-
cy of the resulting adducts to cyclize and undergo further reaction. Multiple pathways
compete, yielding complex mixtures of many different products. Michael and Schiff
base additions, alone and in combination, plus various cyclizations and some cross-
linking probably all occur simultaneously with at least low yields in most systems. A
full accounting of all products has never been reported. Accessibility of specific amino
acids in proteins as well as reaction conditions, such as pH, solvent, oxygenation,
degree of lipid oxidation, and timing of analyses, all influence which pathways and
products will dominate in a given system. Thus, the discussion of reactions that fol-
lows does not attempt to portray any one product as the dominant lipid oxidation
product causing damage to proteins, but instead describes the broad range of reac-
tions that probably occur simultaneously although in different proportions under
various conditions.
The importance of recognizing multiple competing pathways when analyzing
damaged proteins or amino acids cannot be overstressed. Too often studies have
focused on a single product, and when it was found in proteins, claimed that product
to be the major oxidant. LDLox, which has been the target of innumerable studies
of protein damage mechanisms, is an excellent example for putting the problem of
simultaneous multiple oxidation pathways in perspective. Just about every oxidant
reacted with LDL has generated oxidation products that could be identified in
LDLox and atherosclerotic plaques by both chemical and immunological techniques.
The major oxidized protein constituent of LDLox is Ne-(2-propenal) lysine formed
by direct addition of MDA to lysine, but variable levels of pyrroles, hydroxynonenal,
and oxononenal adducts are also present in the same samples (Uchida, 2000). Which
is the most toxic modification? Perhaps all are important in independently inducing
204 K.M. Schaich
different cell responses that together generate full-blown pathology in atherosclerosis

and other diseases of oxidative stress. Clearly more quantitative differentiation of
multiple pathways and products is needed to elucidate factors controlling reaction
routing and product stability under different conditions.
Acrolein and Crotonaldehyde

The simplest alkenals are the ubiquitous air pollutants acrolein and crotonaldehyde
(Fig. 8.9). Although acrolein is most often associated with thermal degradation of
frying oils, it is also a metabolite in the transformation of allyl compounds and has
been identified among secondary lipid oxidation products of arachidonic acid and
w-3 PUFA in vivo (Uchida et al., 1998a, 1998b; Uchida, 1999; Kehrer and Biswal,
2000). Crotonaldehyde is a carcinogenic oil component found in wood smoke and
car exhausts (ATSDR, 2002) and is claimed to be produced in vivo by Cu+ and Fe3+/
ascorbate-catalyzed oxidation of linoleic and linolenic acids (Ichihashi et al., 2001).
Both aldehydes are highly electrophilic, so they have very strong reactivity with nu-
cleophiles, such as thiols, the imidazole groups of histidine, and the e-amino group
of lysine, via Schiff base and Michael additions (Esterbauer et al., 1991; Uchida et
al., 1998a).
The dominant reaction varies with the protein and amino acid accessibility, but
Michael addition is usually faster and the products are more stable. Histidine prefers
Michael addition almost exclusively, with reaction at the t nitrogen (Esterbauer et al.,
1991; Uchida and Stadtman, 1992; Uchida et al., 1998a, 1998b)1. The products are
b-substituted propanals.
Lysine forms Michael addition products when aldehyde concentrations are low,
but when aldehydes are present in excess of the amines, multiple additions generate
cyclic pyrrole derivatives.
1 in the reactions represents his and lys, respectively, attached to a protein, with only the
reactive group shown.
As noted previously, Michael addition of amino acids to alkenals has two impor-
tant consequences: 1) the products introduce into proteins carbonyl groups that are
detected as part of the standard protein carbonyl assay; and 2) the aldehyde carbonyls
provide new reactive sites for further reactions, so they act as nuclei for crosslinking
and cyclization. In contrast, Schiff base adducts with lysine and cysteine remove car-
bonyls by condensation with amines:
C H 3 (C H 2 ) n C H O + R -N H 2 C H 3 (C H 2 ) n -1 C H =C H −N H -R C H 3 (C H 2 ) n C H =N -R
Since Schiff bases activate proteases (Davies, 1987b), this modification may play an
important protective role in clearing oxidized proteins from biological systems.
With extended incubation time or high aldehyde concentrations, accumulated
lysine Schiff base and Michael adducts of acrolein and crotonaldehyde cyclize to yield
EMP [Ne-(5-ethyl-2-methylpyridinium)] and FDP [Ne-(2,5-dimethyl-3-formyl-3,4-
dehydropiperidino)] structures (Fig. 8.11). Although two aldehydes are required for
each amine, the amino group appears to direct the condensations. EMP structures are
formed via initial 1,2 addition of lysine to the carbonyl carbon, followed by a 1,4-ad-
dition of the imine to a second aldehyde, carbonyl condensation to close the ring,
and a final dehydration and dehydrogenation. In the process, the second carbonyl is
reduced. While EMP-lysine is a minor product for crotonaldehyde and not formed
by acrolein, it is the dominant pathway for 2-hexenal and 2-octenal (Ichihashi et
al., 2001). A significant consequence is fixing a permanent positive charge on the
e-amino groups involved.
In contrast, FDP adducts are formed via successive 1,4 additions of lysine, fol-
lowed by carbonyl condensation to close the ring, then dehydration. Unlike EMP ad-
ducts, FDP adducts retain a carbonyl function and thus will be detected in carbonyl
assays of oxidized proteins. Acrolein and crotonaldehyde both form FDP-lysine ad-
ducts (Uchida et al., 1995, 1998b; Ichihashi et al., 2001), and analogous adducts are
formed by 2-hexenal (dipropyl-FDP-lys) and 2-pentenal (diethyl-FDP-lys) (Ichihashi
et al., 2001). Antibodies to FDP and EMP functions have detected adducts of acro-
lein and crotonaldehyde in LDL oxidized by copper (II) (Uchida et al., 1998a, 1998b;
Ichihashi et al., 2001) and in BSA co-oxidized with methyl arachidonate (Uchida et
al., 1998b).
206 K.M. Schaich
A. Formation of EMP adducts via 1,2 additions

R-NH
.. 2 O O OH
O .. +N
+N
N R
R R
− H2O − 2 H+
Ethyl methylpyridinium (EMP)-lysine +N

R
B. Formation of FDP adducts via 1,4 additions
O O O OH
.. R -NH
.. N O N O
R-NH2 R R
O
− H 2O
Formyl dehydropiperidino (FDP) -lysine

N O
R
Fig. 8.11. Formation of cyclic products from additions of lysine to unsaturated aldehydes such as crotonaldehyde: A.
Emp (ethyl methylpyridinium)-lysine via 1,2-addition (Ichihashi et al., 2001); B. Fdp (formyl dehydropiperidino)-
lysine via 1,4-addition (Uchida et al., 1995, 1998b; Ichihashi et al., 2001).
4-Hydroxy-2-Alkenals and 4-Oxo-2-Alkenals

Oxidized alkenals (Fig. 8.9) are the aldehydes most reactive with proteins. Perhaps the
best known and most studied aldehydes in this class are 4-hydroxy-2-nonenal (HNE)
and 4-oxo-2-nonenal (ONE). Reactions and chemistry of HNE have been reviewed
in elegant detail by Esterbauer et al., (1991); other useful reviews covering various
aspects of HNE reactions with proteins and subsequent physiological effects are also
available (Uchida, 2000; Schaur, 2003; Davies et al., 2004; Petersen and Doorn,
2004). 4-Oxo-2-alkenals, also called g-ketoaldehydes, may be formed independently
but are actually tautomers of 4-HO-2-alkenals:
Thus, under oxidizing conditions, the reactions of these two oxidants are often diffi-
cult to distinguish, except by kinetics (ONE reactions are orders of magnitude faster)
and balance between pathways. The major reaction for both HNE and ONE is (1,4)
Michael addition to nucleophilic amino acid side chains (thiol, imidazole, lysine)
(Schauenstein et al., 1971; Esterbauer et al., 1975, 1976; Uchida and Stadtman,
1992, Bruenner et al., 1995; Schaur, 2003). For HNE, the free carbonyl can then
undergo secondary Schiff base formation with another amine. The second carbonyl in
ONE makes it markedly more electrophilic and provides a site for Schiff base forma-
tion as well as Michael addition in both initial and subsequent reactions. ONE also
reacts with arginine, but HNE does not (Lin et al., 2005).
To provide a quantitative frame of reference, ONE reacted up to 30% faster than
4-HNE with RNAse A and b-lactoglobulin (Lin et al., 2005). With [amine] > 10x
[aldehyde], the following t½ for the reactions were observed: for RNAse, 95 h and 6
min for HNE and ONE, respectively; and for b-lactoglobulin, 14 h and 3 min for
HNE and ONE, respectively. The order of Michael addition reactivity for the two
aldehydes with amino acids is as follows (relative rate constants are in parentheses)
(Doorn and Petersen, 2002, 2003):
HNE: cys (1) >> his (0.002) > lys (0.001)

ONE: cys (186) >>> his (0.02) > lys (0.007) > arg (0.0006)
Michael additions of cysteine on proteins to HNE and ONE occur almost spontane-
ously, proffering a staggering potential for biological damage. In foods, lipid oxida-
tion products accumulate, so thiol reactivity of HNE and ONE may be an important
contributor to quality deterioration during long-term storage, limited only by the
availability of cysteine residues on proteins. Sulfhydryl proteins, such as protein di-
sulfide isomerase, show extensive loss of cysteine in the presence of HNE in model
systems (Carbone et al., 2005). However, extensive protective mechanisms counteract
this reactivity in vivo. GSH-S-transferase redirects the reaction to intracellular gluta-
thione (GSH) in a catalyzed reaction that is 600x faster than chemical conjugation,
and aldehyde dehydrogenases rapidly reduce carbonyl groups to alcohols, effectively
limiting cellular concentrations of these reactive aldehydes, even under conditions of
oxidative stress (Siems and Grune, 2003).
Despite nearly instantaneous reaction rates with thiols, cysteine contents of most
proteins are very low, so histidine and lysine become the major targets of HNE and
ONE. However, these aldehydes differ in amino acid selectivity. HNE prefers Michael
addition reactions with histidine and lysine, but the amino acid most damaged de-
pends on protein primary structure, configuration, and residue availability. Histidine is
the main target in b-lactoglobulin B and human hemoglobin (Bruenner et al., 1995),
myoglobin (Alderton et al., 2003) and apomyoglobin (Bolgar and Gaskell, 1996),
insulin B chain (Fenaille et al., 2003), bovine heart cytochrome c oxidase (Musatov et
al., 2002), and RNAse A (Liu et al., 2003; Lin et al., 2005), while lysine is the domi-
nant reaction site in glucose-6-dehydrogenase (Uchida and Stadtman, 1992; Friguet
et al., 1994a), human LDL (Refsgaard et al., 2000), bovine serum albumin (Zamora
and Hidalgo, 2003a), and a secondary site in RNAse A (Liu et al., 2003; Lin et al.,
2005). In contrast, the fastest reaction between ONE and RNAse or b-lactoglobulin
was Schiff base formation with specific lysines, leading to ONE-lys pyrrolidone on
K16 and K145, and ONE-his-lys pyrrole crosslink K16—H24 (Lin et al., 2005).
208 K.M. Schaich
Schiff base and Michael additions of hydroxy and oxo alkenals to proteins gen-
erate linear adducts that remove peptide thiols and amines and add carbonyls. The
order of efficiency in generating protein carbonyls is acrolein > oxononenal > hy-
droxynonenal > dodecadienal > MDA (Yuan et al., 2007). However, these adducts are
not always detected because they dissociate reversibly (Esterbauer et al., 1991; Nad-
karni and Sayre, 1995), especially under conditions used to hydrolyze proteins (Lin
et al., 2005), and the aldehyde also cyclizes rapidly to form a hemiacetal (Esterbauer
et al., 1975, 1976; Uchida and Stadtman, 1992; Lin et al., 2005) (shown below for
histidine, cysteine, and lysine, respectively):
When amines are present in molar excess over the aldehydes, pyrrole derivatives
are formed, as shown below. The ring structure originates from reaction of two amines
(usually lysine) with HNE and ONE in a complex sequence involving Schiff base ad-
dition, (oxidation for HNE), Michael addition, oxidation, and cyclization, (Sayre et
al., 1993, 1997; Xu et al., 1999b; Schaur, 2003; Zhang et al., 2003). HNE and ONE
both form stable fluorescent 2-pentyl-2-hydroxy-1,2-dihydropyrrol-3-one iminium
complexes involving addition of two lysines, as shown in the reaction sequence below
(Xu et al., 1999a, 1999b; Zhang et al., 2003). The net result is formation of a peptide
crosslink. In addition, if the pyrrole–OH in the iminium complex is not removed in
dehydration, it can react further, for example with another lysine, to form cyclic or
acyclic mixed aminals.
In addition to pyrroles, HNE and ONE generate a wide range of secondary cyclic
products, including epoxides, pyrroles, pyrrolidones, and thiazolidines as adducts and
crosslinks under oxidizing conditions. As with pyrroles, these products arise from
various combinations of Schiff base and Michael additions, oxidations, cyclizations,
and dehydrations. Amine and aldehyde concentrations are major determinants of
product pathways. Intermolecular cyclization is facilitated at high aldehyde concen-
trations, and crosslinking occurs when amines are present in molar excess over alde-
hydes. Secondary cyclic products are responsible for many of the observed changes in
properties of proteins reacted with HNE and ONE. The reactions generating various
cyclic products will be discussed in the sections on Crosslinking and Formation of
Fluorescent Products.
Detailed structures are now being identified by electrospray ionization (ESI)
(Bruenner et al., 1995; Bolgar and Gaskell, 1996; Brame et al., 1999; Fenaille et
al., 2002, 2003; Liu et al., 2003; Schöneich and Sharov, 2006) and MALDI-TOF
(Doorn and Petersen, 2002; Kapphahn et al., 2006) mass spectrometry of modified
peptides, with and without proteolysis (Fenaille et al., 2004). The presence of indi-
vidual adduct structures in individual proteins can be tracked by specific antibodies
(Uchida et al., 1993; Xu et al., 2000), although consideration must be given to cross-
reactivity when interpreting data.
Physiological γ-Ketoaldehydes–Levuglandins
Isoketals form on free fatty acids in transport and also on fatty acids still esterified in
phospholipids where they generate novel PL-protein adducts that profoundly affect
protein function in membranes, completely blocking K+ ion channels (Brame et al.,
2004). A special class of isoketals are isoprostanes or levuglandins formed by rear-
rangement of endoperoxides in arachidonic acid (Brame et al., 1999; Davies et al.,
2004). The reaction below shows only the 5-series of hydroperoxides, but analogous
isoketals are also formed with the precursor hydroperoxide at C-8, 12, and 15, with
5-OOH and 15-OOH being dominant.
210 K.M. Schaich
OOH
2 O2 COOH
COO H
O
O OH O OH
CO O H COO H
+
O O
IsoLevuglandin D IsoLevuglandin E
All isoketals show remarkable proclivity for polymerization (Xu et al., 1999a;
Brame et al., 2004; Davies et al., 2004). Levuglandin E2 is orders of magnitude more
effective in crosslinking proteins than any other oxidation product of arachidonic
acid (Iyer et al., 1989). In aqueous systems, isoketals react with proteins faster than
all other secondary products (first order k=106–108 s–1); in organic solvents the rate of
pyrrole formation is 103–104 times slower (second order k=102–104 M–1 sec –1) because
the Schiff base product dominates and cyclization does not occur until water is added
(Amarnath et al., 1995). This observation provides compelling evidence that phase
localization of the reaction will determine the dominant pathway. The reason for
the exceptional crosslinking reactivity of isoketals is that, like other 4-HO-enals and
1,4-dicarbonyls, they almost instantaneously cyclize to form pyrroles when reacted
with amines (Amarnath et al., 1995). When a 3x molar excess of levuglandin E2 was
reacted with BSA, >50% of the isoketal had reacted within 20 seconds, accompanied
by production of BSA crosslinks (Brame et al., 1999). In buffered systems a small
portion of the products also arise from classical carbonyl-amine condensation to form
imine Schiff base adducts, but the major products are lactams, hydroxylactams, and
polymers from oxidation of pyrroles (Fig. 8.12). The first step is a Paal-Knorr conden-
sation of dicarbonyls with protein amines (primarily lysine) to form monoalkylpyr-
roles (Amarnath et al., 1995; Xu et al., 1999a). In the presence of oxidizing lipids, the
pyrroles then oxidize further to lactams and hydroxylactams (Xu et al., 1999a; Brame
et al., 2004; Davies et al., 2004) which retain available carbonyls.
Addition Reactions from Free Radical Oxidation of Unsaturated Aldehydes

Although Michael addition to a,b-unsaturated aldehydes is generally presumed, at
least two studies have shown that secondary autoxidation of unsaturated aldehydes
can yield radicals that induce crosslinking in proteins (Funes and Karel, 1981) and
also react with nucleic acids (Yang, 1993). Although routinely ignored, autoxida-
tion of saturated aldehydes to carboxylic acids and other products is very rapid and
involves intermediate radicals (Pokorny et al., 1985); secondary oxidation of unsatu-
rated aldehydes should be even more facile. Free radical oxidation of aldehydes pres-
ents still another pathway for lipid oxidation damage to proteins, and whether these
radicals act conventionally by hydrogen abstraction or have a greater tendency to add
to proteins needs to be investigated further.
Fig. 8.12. Competing pathways for reaction of isoketals with e-amino groups of proteins. The pyrrolidine path-
way is dominant. For one, R and R’ = H. Reaction sequence adapted from (Brame et al., 1999, 2004; Xu et al.,
1999a; Davies et al., 2004).
Reactions Underlying Molecular Damage to Proteins

The reactions of lipid radicals, hydroperoxides, epoxides, and aldehydes discussed pre-
viously are of theoretical value only until they are translated into observed behaviors
in foods and biological systems. Indeed, these damage mechanisms were presented
first to provide a basis for understanding the long list of macromolecular properties
changed when oxidizing lipids are reacted with proteins. This section now connects
specific lipid co-oxidation chemistry to commonly observed changes in protein physi-
cal and chemical properties.
212 K.M. Schaich
Amino Acid Losses

The potential for oxidizing lipids to damage proteins has been documented since the
early 1960s (Wills, 1961a,b; Lewis and Wills, 1962; Tappel, 1965, 1973; Roubal
and Tappel, 1966a), with one of the earliest observations being amino acid losses in
proteins reacted with oxidizing lipids (Desai and Tappel, 1963; Roubal and Tappel,
1966b; Chio and Tappel, 1969a; Gamage et al., 1973; Horigome and Miura, 1974;
Horigome et al., 1974; Kanazawa et al., 1975; Matsushita, 1975; Gardner, 1979;
Matoba et al., 1984b; Nielsen et al., 1985b). Over a large number of studies, the fol-
lowing amino acids have consistently been shown to be most sensitive to damage by
peroxidizing lipids:
Cysteine Serine Glycine

Tryptophan Threonine Alanine
Histidine Valine
Lysine Proline
Arginine Leucine
Tyrosine Isoleucine
Methionine
The amino acids in the first column (on left) all are located primarily on protein
surfaces (Fig. 8.13); with the exception of methionine, they have readily abstractable
hydrogens, and cys, his, lys, trp, and arg form stable radicals when reacted with oxi-
dizing lipids. The phenoxyl hydrogen of tyrosine is probably also easily abstracted,
but without protecting ortho groups the resulting radical is not stable enough for
detection. The susceptibility of these amino acids to damage from oxidizing lipids
L ys A rg
T rp
Surface residues with
abstractable H’s are
C ys Ile P ro
primary targets of
oxidizing lipids
A la Buried hydrophobic
H is residues require
V al denaturation for
exposure and reaction
A sp L eu with oxidizing lipids
T yr
S er T h r
M et
Fig. 8.13. Diagrammatic representation of amino acids most susceptible to attack by oxidizing lipids. The model
protein shown is a typical globular or structural protein in an aqueous environment: surface residues are primarily
hydrogen-bonding or charged amino acids and aliphatic amino acids are mostly buried. Hydrophobic proteins or
proteins embedded in membranes have higher proportions of aliphatic and aromatic amino acids exposed for
reaction.
is thus predictable and parallels damage from irradiation, photolysis, and hydroxyl
radicals (Schaich, 1980b). Side chain amine thiol and amine groups of these amino
acids also react rapidly with carbonyl products of lipid oxidation to form Schiff bases,
Michael adducts, and their cyclic products. Thus, these amino acids are involved in
early stages of oxidation and remain reactive through secondary and even terminal
stages, although perhaps changing their pathways, as proteins denature and expose
new amino acid residues.
Damage to the other amino acids is less easily explained. Serine and threonine
(center column) have –OH groups, but the pKs are >>14 and the O-H bond energies
are high (464 kJ/mol, 111 kcal/mol), so these –OH groups do not donate protons
as easily and thus should not be favorable targets for oxidizing lipids. Reaction with
other oxidants also shows low reactivity for the side chain hydroxyls. Oxidation by sil-
ver (Shi et al., 2007), KMnO4 (Halligudi et al., 2000), and electrical current (Huerta
et al., 1997) all documented initial attack at the –COOH group with accompanying
decarboxylation, followed by deamination; the side chain –OH groups were never
modified. However, serine and threonine side chain hydroxyls do mediate rather
strong hydrogen bonding. Thus, it is interesting to speculate that these amino acids
hydrogen bond to lipid hydroperoxides and induce LOOH decomposition (mol-
ecule-assisted homolysis), accompanied by radical transfer in cage reactions to gener-
ate side chain radicals and subsequent hydroperoxides and breakdown products:
In contrast, the hydrophobic amino acids (last column) have no readily ab-
stractable hydrogens, do not participate in hydrogen bonding, and are buried in the
interior of native protein; they do not become exposed without denaturation. Indeed,
damage to these amino acids has been reported primarily in proteins incubated with
oxidizing lipids at higher temperatures (55–60°C) where denaturation is facilitated
(Horigome and Miura, 1974; Horigome et al., 1974). Thus, their sensitivity to dam-
age by oxidizing lipids is more difficult to explain, and indeed, little has yet been elu-
cidated regarding mechanisms of lipid reactions with these amino acids. Nevertheless,
three possible explanations for observed damage may be offered.
a. When free hydrophobic amino acids are reacted with a variety of strong
oxidants, the main sites of reaction are the carboxylic acid group, followed
by the amine group; decarboxylation and deamination frequently occur
(Schaich, 1980a; Bobrowski and Schöneich, 1996; Huerta et al., 1997;
Guitton et al., 1998). The only oxidant that has shown ability to react with
valine, leucine, and isoleucine side chains is the highly electrophilic hydroxyl
214 K.M. Schaich
radical, HO•, which abstracts hydrogens rather nonspecifically from

hydrocarbons (Hasegawa and Patterson, 1978; Patterson and Hasegawa,
1978), and attacks the side chains and terminal methyl groups of these
amino acids (Fu et al., 1995). It may seem counterintuitive, but when HO•
generated cleanly by radiation were reacted with individual amino acids,
it was proline, valine, leucine, isoleucine that showed the highest yields of
protein radicals and hydroperoxides (Simpson,1992). Analyses revealed that
a tertiary carbon atom or a segment containing at least two methylene groups
was required to stabilize the initial radical so that O2 could add to form
peroxyl radicals. By analogy, HO• are produced when lipid hydroperoxides
decompose under UV light or at elevated temperatures (60°C is sufficient),
and they are generated when ferrous iron autoxidizes. Thus, it is reasonable
to expect that the adventitiously produced HO•, rather than the expected
LOO• or LO• radicals, may mediate oxidation of aliphatic amino acids.
b. It is well documented in irradiated proteins that free radicals generated
anywhere along the main peptide or amino acid side chains tend to migrate
along the protein backbone to glycine residues, where they are stabilized
(Schaich, 1980a). The hydrocarbon side chains of alanine, valine, leucine,
isoleucine, and proline may similarly provide electron sink sites where
radicals are stabilized (Fig. 8.14), leading to oxidation and production of
hydroperoxides on side chains of these amino acids.
c. Iron is often present adventitiously or deliberately added as a catalyst in lipid

oxidation studies. LOOH can complex to Fe2+ to form ferryl iron (Fe4+ =O)
or equivalent complexes that are highly electrophilic and abstract hydrogens
even more rapidly and non-specifically than HO• (k > 109 L M-1s-1). Ferryl
iron complexes also catalyze oxygen insertion into C-H bonds to yield
epoxides, ketones, and alcohols (Schaich, 2005).
•NH
-NH-CO-CH-NH-
-NH-CO-CH-NH
-
• CH2
C•
CH3 CH3 •S
Fig. 8.14. Radicals initially formed at amino or sulfhydryl sites on amino acid side chains migrate along peptide
backbone and tend to localize at glycine residues (left). Aliphatic side chains may similarly serve as electron sinks
(right), leading to free radical oxidation of hydrophobic residues.
Fe3+-LOOH yields the Fe3+-•OH complex which is functionally equivalent to Fe4+

(Bossman et al., 1998). Detailed studies are needed to determine which of these ex-
planations, or others, accounts for lipid oxidation damage to aliphatic amino acids.
Specific amino acid breakdown products and the lipid oxidant inducing them
are listed in Table 8.1. Mechanisms and reactions leading to these products will be
presented in a separate paper.
The most obvious consequence of amino acid destruction by oxidizing lipids is
diminution of nutritional quality of foods, as has been shown in a variety of food
systems, for example intermediate moisture meats (Obanu et al., 1980), freeze-dried
beef (Dvorak, 1968), casein (Horigome and Miura, 1974; Horigome et al., 1974;
Yanagita and Sugano, 1975; Matoba et al., 1984b; Kanazawa et al., 1987), whey pro-
tein (Nielsen et al., 1985a, 1985b) chickpea legumin (Sánchez-Vioque et al., 1999),
ovalbumin (Horigome and Miura, 1974; Horigome et al., 1974, Yanagita and Su-
gano, 1974), and lupine conglutins (Lqari et al., 2003). Loss of nutritional value in-
creases markedly when lipid-protein reactions occur at high temperatures (50–60°C),
as might be expected.
Interestingly, the problem seems to involve more than mere destruction of amino
acids or reduction in digestibility or absorption, since simultaneous supplementa-
tion of diets with amino acids found to be affected does not always restore normal
nutritional value (Horigome and Miura, 1974; Horigome et al., 1974; Yanagita et
al., 1976). Effects of damaged ovalbumin could be repaired by adding back lys and
met (Yanagita and Sugano, 1974), but amino acids could not reverse damage from
casein reacted with oxidizing ethyl linoleate (Yanagita and Sugano, 1975; Yanagita
et al., 1976). Absorption of damaged proteins or proteolysis of damaged proteins
with recycling of component amino acids in vivo has a cascade of effects, starting
with decreased liver nitrogen and lipids (Yanagita and Sugano, 1975; Yanagita et al.,
1976). This, in turn, causes imbalances in the amino acid pool that lead to impaired
protein synthesis as well as other as yet unidentified metabolic and functional impair-
ments. Amino acid imbalances may also be exacerbated by the conversion of polar
essential fatty acids to non-polar non-essential fatty acids, such as alanine and leucine,
histidine degradation to aspartic acid and asparagine, and proline transformation to
glutamic acid (Table 8.1).
These and other indirect effects on protein nutrition are exceedingly difficult to
track but may well contribute to aging, oxidative stress, and a wide range of patholo-
gies by creating an environment in which the direct molecular damage to key proteins
from lipids or oxidants cannot be counteracted.
Another critical impact of amino acid modification is on protein structure and
associated extensive effects on protein functionality in foods and biological activity in
Table 8.1. Examples of Degradation Products Formed in Reactions of Amino Acids with Lipid Oxidation Intermediates and Products. 216
Amino Acid Lipid Oxidant Reactiona Products References
radical recombination cystine (Little & O’Brien, 1967, 1968)
R-S scission alanine, H2S, cystine (Roubal & Tappel, 1966b)
K.M. Schaich
(Lewis & Wills, 1962;

alanine sulfinic acid, cysteic acid,
LO(O)• Little & O’Brien, 1968)
oxidation cystine disulfoxide, cystine thiosulfinate,
cystinethiosulfonate, cysteine sulfoxides, (Finley & Lundin, 1980)
cysteine sulfinic and sulfonic acids
(Gardner & Weisleder, 1976;
lipid epoxides, cysteine oxides
Gardner & Kleiman, 1981)
LOOH addition
9-S-cysteine-13-HO-10-ethoxy-trans-11-octadecenoic acid
(Gardner et al. 1977)
Cysteine 9-S-cysteine-10,13-diHO-trans-11-octadecenoic acid
epoxides addition hydroxylated adducts (Buttkus, 1972)
alkanals 1:1 addition Schiff base adducts (Doorn & Petersen, 2002;
Petersen & Doorn, 2004;
1:1 addition Michael addition products Lin et al., 2005)
alkenals
Schiff base addition/cy- (Schmolka & Spoerri, 1957;
thiazolidinecarboxylic acid derivatives
clization Buttkus, 1968)
Michael + Schiff base
MDA (excess) 3:2 complex condensation products (Buttkus, 1968)
addition
(Esterbauer et al., 1975;
HO-alkenals 1:1 Michael addition cyclic hemiacetals
Esterbauer et al., 1991)
Schiff base addition/cy-
oxoalkenals thiazolidinecarboxylic acid derivatives (Gardner & Jursinic 1981)
clization
Table 8.1., cont. Examples of Degradation Products Formed in Reactions of Amino Acids with Lipid Oxidation Intermediates and Products.

(Garner et al., 1998a;
oxidation methionine sulfoxide
Brock et al., 2007)
LOOH strong oxidation methionine sulfone (Wainwright et al., 1972)
Methionine
deamination, decar-
methional, methane thiol, acrolein (Wainwright et al., 1972)
boxylation
MDA Schiff base addition HOOC-CH(R)-NH-CH=CH-CHO (Buttkus, 1968)
deamination imidazole lactic acid, imidazole acetic acid (Yong & Karel, 1978, 1979)
decarboxylation, scis-
LO(O)• histamine, ethylamine (Roy & Karel, 1973)
sion
ring scission aspartic acid, asparagine, valine (Roy & Karel, 1973)
Histidine HNE Michael addition β-substituted propanal adducts, hemiacetals (?) (Uchida & Stadtman, 1992)
epoxy alk-
Michael addition β-substituted adducts, epoxide intact (Zamora et al., 1999)
enals
crotonalde- nucleophilic addition (Ichihashi et al., 2001)
β-substituted butanal adducts
hyde to C=C
decarboxylation diaminopentane (Karel et al., 1975)
α-deamination/cycliza-
pipecolic acid (Karel et al., 1975)
tion
LO(O)• ε-deamination/rad
1,10-diamino-1,10-docarboxydecane (Karel et al., 1975)
recombination
ε-deamination/oxida-
Lysine α-amino adipic acid, aspartic acid, alanine glycine (Karel et al., 1975)
tion/ decarboxylation
concerted scission/ (Kato et al., 1999;
MLOOH Nε- (hexanoyl)lysine, Nε- (azelayl)lysine
addition Kawai et al., 2003)
pentanal Schiff base addition 2-propyl-2-heptenal-lysine (Dalsgaard et al., 2006)
Co-oxidation of Proteins by Oxidizing Lipids
1:2 Schiff base, frag- α-amino-ε−caprolactam, immonium-lysine, diacetyl lysine,

hexanal (Fenaille et al., 2004b)
ment, cyclize pipecolic acid
217
Cont. on p. 220.
Amino Acid Lipid Oxidant Reactiona Products References 218
2 MDA + 1
Schiff base, cyclization 1,4-dihydropyridine-3,5-dicarbaldehydes (Freeman et al., 2005)
alkanal
crotonalde-
Michael addition 1:1 β-substituted adducts (Ichihashi et al., 2001)
hyde
crotonalde-
K.M. Schaich
hyde, Michael addition 2:1 FDP-lysine (3-formyl-3,4-dehydropiperidino adducts) (Uchida, 2000)

acrolein
crotonalde-
hyde Schiff base 2:1 EMP-lysine (pyridinium adduct) (Uchida, 2000)
(excess)
2-octenal Michael addition 1:1 β-substituted adduct (Alaiz & Girón, 1994)
epoxy alke- Schiff base, cyclic ad- (Zamora & Hidalgo, 1995;
amino carboxypentylpyrroles
nals dition Zamora et al., 2000)
Lysine, cont. epoxy alke- Schiff base, scission,
N-pyrrolylnorleucine (Zamora et al., 1995)
nals addition
epoxy hep- Michael addition,
N-substituted pyrroles and 2-(1-hydroxyalkyl)pyrroles (Zamora et al., 2000)
tenal cyclization
HO-ω-
oxoalkenoic Schiff base, cyclization 2-(ω-carboxyalkyl)pyrroles (Gu et al., 2003)
acids
(Szweda et al., 1993;
HNE Michael addition 1:1 β-substituted adducts, cyclic hemiacetals
Nadkarni & Sayre, 1995)
1:2 Schiff base + Mi-
HNE amine-hemiacetal-amine crosslinks (Nadkarni & Sayre, 1995)
chael addition
(Cohn et al., 1996;
1:2 Michael + Schiff
HNE Hydroxy-imino-dihydropyrroles Itakura et al., 1998;
base addition
Tsai et al., 1998)
H abstraction, oxida-
LO(O)• 5-oxoproline (Spiteller, 2006)
tion, hydrolysis
Arginine H abstraction, peroxi- (Pietzsch, 2000;
radicals γ-glutamyl semialdehyde
dation Requena et al., 2001)
MLOOH oxidation NO• release (Schaich, 2002c)

H abstraction, peroxi- (Yong et al., 1980;
kynurenine, N-formyl kynurenine
dation Krogull & Fennema, 1987)
Tryptophan LO(O)•
addition dioxindole-3-alanine (Yong et al., 1980)
ring scission indoleacetic acid and indolelactic acid (Yong et al., 1980)
H abstraction, radical (Hunter et al., 1989;
dityrosine
LO(O)• dimerization Kikugawa et al., 1991;
H abstraction, oxidation tyr hydroquinone, tyr quinone Giulivi et al., 2003)
Tyrosine addition, oxidation, (Giulivi & Davies, 1993;
DOPA, dopamine, 5,6-diHO-indole
scission Giulivi et al., 2003)
HO•
addition, decarboxyl- (Giulivi & Davies, 1993;
dopamine, dopamine quinone
ation Giulivi et al., 2003)
Phenylala-
HNE Schiff base, cyclization N-substituted-2-pentyl pyrrole, phenylacetaldehyde (Hidalgo et al., 2005)
nine
H abstraction, deami-
LO(O)• formic acid (Berger et al., 1999)
Glycine nation
MDA Schiff base N-prop-2-enal aminoacetic acid (Crawford et al., 1966).
H abstraction, peroxi-
Valine HO• 3 valine hydroperoxides (Fu et al., 1995)
dation
H abstraction, peroxi-
LO(O)• 5-oxoproline (Spiteller, 2006)
dation
HO•, t-Bu-O• addition, ring scission glutamic semialdehyde (Requena, 2001)
Proline
(Schuessler & Schilling, 1984;
HO•, radia- addition, oxidation,
γ-aminobutyric acid, glutamic acid, and 2-pyrrolidone Kato et al., 1992;
tion scission
Matysik et al., 2002)
Co-oxidation of Proteins by Oxidizing Lipids
a
Reactant proportions cited are aldehyde:amine
219
220 K.M. Schaich
vivo. Structural alterations are the subjects of Sections “Transfer of Free Radicals from
Oxidizing Lipids to Proteins” through “Alteration or Loss of Biological Function.”
Transfer of Free Radicals from Oxidizing Lipids to Proteins

Most dry foods with lipids and proteins show stable EPR signals from radical transfer.
EPR signal intensities correlate with the extent of lipid and protein oxidation and
provide clear footprints of lipid-mediated damage (Schaich and Karel, 1975; Schaich,
1980b; Saeed et al., 1999, 2006) (Fig. 8.15). Thus, use of protein EPR signals in
conjunction with lipid oxidation analyses can reveal the extent to which oxidation has
been broadcast beyond lipids and improve predictions of quality deterioration during
processing and storage.
Some cautions in analysis and interpretation of free radicals in foods and bio-
logical systems are necessary. Radical quantitation is complicated by moisture be-
yond interference with EPR measurements. Maintaining low moisture slows radical
recombinations and prolongs radical lifetimes, hence the common use of dry models
to study protein radical reactions. Nevertheless, much spectral information is lost in
these systems due to molecular immobilization. In addition, it can be argued that
such reaction conditions are relevant only to dehydrated foods and do not accurately
reflect reaction conditions in biological systems. On the flip side, even low moisture
provides protons to quench protein free radicals and in effect repairs limited damage,
while higher moisture increases molecular mobility and enhances recombination of
side chain radicals (Zirlin and Karel, 1969).
Thus, in solutions or emulsions, tracking free radical transfer to proteins requires
special techniques to detect short-lived radical species. Radicals may be detected di-
rectly by low temperature (liquid nitrogen) measurements or by continuous flow-mix
methods, which will also be able to provide reaction rates, (Schaich 2002a, 2002b,
2002c). Radical transfer may be followed indirectly by spin trapping (Lion et al., 1981;
Culbertson et al., 2003; Headlam and Davies, 2003; Augisto et al., 2004; Davies and
Hawkins, 2004; Mason, 2004) or immuno-spin trapping (Mason, 2004) as has been
demonstrated with HO• and NO• reactions with proteins, or by chemical modifications
to proteins (Reubsaet et al., 1998; Davies et al., 1999; Akagawa et al., 2006). Spin traps
offer the advantage of being able to detect and distinguish both short-lived lipid radicals
and longer lived protein radicals in the same sample and in multiple solvents. Since
nitroxide radical line shapes and splittings are very sensitive to local molecular motion
and to solvent polarity (Schaich 2002a, 2002d), they can also uniquely reflect alteration
in protein conformation, for example unwinding in denaturation (Saeed et al., 2006) or
collapsing in aggregation or crosslinking, that accompanies radical transfer.
Changes in Physical Properties: Solubility, Hydrophobicity, Conformation, Aggregation

Any modification of amino acids through oxidation, scissions, or adduct formation
must be reflected in protein properties. Protein oxidation, whether by HO• or oxi-
dizing lipids, can induce marked changes in conformation or structure of proteins
Fig. 8.15. Correlation of lipid oxidation, protein oxidation (antibody reaction), and protein free radicals (EPR signal
intensity) in common foods (Tanczos et al., 2002). (Top) Room temperature EPR signals of several common foods.
Sulfur radicals are only detected at liquid nitrogen temperatures or lower (-196°C). (Bottom) Left lane for each
food: protein bands on 12% polyacrylamide electrophoresis gels (PAGE). Right lane: Same proteins transferred to
Western blots and reacted with antibodies raised against carbonyl dinitrophenylhydrazones in oxidized proteins.
222 K.M. Schaich
by altering surface charges (Haberland 1982, 1984) or by increasing hydrophobic-

ity, inducing denaturation, or complexing lipids to the protein (Lea, 1957; Davies
and Delsignore, 1987). By reaction with free amines, MDA denatured rabbit myosin
(King and Li, 1999), reduced its a-helix content, increased random structure, and
eliminated b-strand structure (Li and King, 1999). MDA binding to his, tyr, arg, and
met residues of myosin altered the isoelectric point and reduced solubility (Buttkus,
1967). High concentrations of MDA caused conformational changes in the globular
heads and destroyed active sites of myosin Ca2+-ATPase (King and Li, 1999). Some
of these changes lead to aggregation of proteins without crosslinking, as has been
observed with plasma b-lipoproteins incubated with methyl linoleate hydroperoxide
(MLOOH) (Nishida and Kummerow, 1960; Suyama and Adachi, 1979) and with
chicken myofibrils lyophilized with methyl linoleate (Smith et al., 1990).
All of these changes reduce protein extractability and solubility in a wide range
of systems. Mechanisms contributing to loss of solubility vary with the protein, the
extent of lipid oxidation or specific lipid oxidation product reacted, and the reaction
system. At low levels of lipid oxidation, decreased solubility in egg albumin complexed
with oxidized fatty acids was attributed to hydrogen bonding between LOOH and
proteins; no covalent crosslinking or loss of functional groups was observed (Narayan
and Kummerow, 1958). Covalent binding of actively oxidizing lipids with a concur-
rent increase in hydrophobicity was observed with g-globulin and albumin (Nielsen,
1978), cytochrome c (Desai and Tappel, 1963), casein (Leaver et al., 1999a), b-lacto-
globulin (Leaver et al., 1999b), and soy protein (Huang et al., 2006). Incubation of
soy protein with soy phospholipids reduced solubility by nearly one-half via disulfide
crosslinking; it also oxidized sulfhydryls to products not reducible by mercaptoetha-
nol and generated protein carbonyls that also altered protein solubility (Boatright
and Hettiarachchy, 1995). Non-covalent aggregation and covalent crosslinking via
carbonyls or free radicals also decrease solubility (Huang et al., 2006).
Changes in protein solubility, conformation, and intermolecular associations
mediated by oxidized lipids translate into deterioration of protein function, such as
enzyme inactivation, impaired baking properties of store flours (Lea, 1957), decreased
gel strength and water-holding capacity of chicken myofibrils (Smith et al., 1990) and
gelatin (Matoba et al., 1984a), membrane leakiness and impaired Ca2+-accumulation
in muscle mitochondria (Player and Hultin, 1978), and loss of functional properties
of soy proteins in foods (Boatright and Hettiarachchy, 1995), to cite a few examples.
The same changes taking place in vivo contribute to loss of normal physiological func-
tion and development of abnormal function and pathology.
Crosslinking
Crosslinking of proteins is a major effect of reaction with lipids at all stages of oxida-
tion–oxyl radicals, epoxides, and secondary carbonyl products alike. Crosslinking,
which leads to decreased solubility, changes in food textures, loss of functionality,
and browning, has been recognized for decades. Indeed, the association of LOOH-
induced protein polymerization products with aging (Hendley et al., 1963a, 1963b)
and with toughening of meats (El-Gharbawi and Dugan, 1965) initially attracted
attention to the importance of protein reactions with oxidizing lipids. Despite long-
term recognition of this phenomenon, understanding of the responsible mechanisms
is still evolving. A major obstacle is that lipid oxidation is a dynamic and constantly
changing process, yet most studies focus on one stage or one intermediate or product.
Thus, trying to fit individual pieces of the puzzle together for oxidation processes to
determine damage causation in intact foods or living tissues under oxidative stress can
be a distinct challenge.
Free Radical Crosslinking

As with physical properties, the susceptibility to crosslinking and the mechanisms
involved are strongly dependent on the nature of the protein, the unsaturation of the
peroxidizing lipid and its products, the reaction system (particularly moisture), and
the reaction time. In systems with actively peroxidizing polyunsaturated fatty acids or
esters, lipid radicals and hydroperoxides provide a major early source of protein radi-
cals which recombine to generate dimers, trimers, and higher polymers within short
incubation times (Fig. 8.16). Oxygen bridges may or may not be present.
The general reaction for free radical crosslinking (polymerization) of intact pep-
tides is
2 L O O H o r L O (O ) • + P H 2 L O (O )H + P • P -P , P -P -P , … (P ) n
where P• may be C• radicals on a peptide backbone or N•, thiyl -S•, or tyrosyl phe-
noxyl radicals -O• on amino acid side chains. P• may remain localized on the side
chain or migrate along the peptide backbone and become stabilized on glycine or
alanine residues (Schaich, 1980b). Crosslinks form when any of these peptide or side-
chain radicals recombine (Roubal and Tappel, 1966c; Schaich and Karel, 1975), as
shown in the reactions below (P is any peptide chain with parts of side chains up to
the specified reactive functional group). Disulfide and dityrosine crosslinks are well-
documented in proteins. Carbon and nitrogen radical polymerizations are expected
from electrophoresis data but have not yet been documented in oxidized proteins.
Recombination of a-C radicals is expected but sterically hindered. Side-chain C• are
common with HO• oxidations but less with lipids. Azides resulting from side chain
N• recombinations are energetically less likely and probably unstable. However, re-
combination of side chain C• and N• is certainly feasible and would generate an iso-
peptide with links very similar to Schiff bases. Advances in enzyme technology and
LC-MS detection should facilitate verification of such crosslinks and identification of
the amino acids involved.
224 K.M. Schaich
Backbone, Backbone,
side chain side chain
•
PH P POO• POOH
•• •• O O
P-P P-OOP
P-P-P P-O-P
•• Oxygenated •• O O
PPPP
etc. polymers
••
isopeptides, ↓ solubility, altered hydrophobicity / polarity,
polymers ↓ protease susceptibility
Fig. 8.16. Free radical crosslinking induced by oxidizing lipids produces mostly polymers of intact protein mono-
mers, with and without oxygen bridges. The accompanying decrease in solubility and changes in physical proper-
ties have profound effects on protein function both in foods and physiologically.
Additions of LOO• (or LO•) to proteins also generate protein radicals. Subse-
quent oxidation and recombination results in protein crosslinks with lipid peroxo
bridges (Desai and Tappel, 1963).
Unspecified mixed crosslinking involving recombination of radicals on intact

proteins and protein scission fragments has also been reported, particularly when the
reacting lipids were the more unsaturated linolenic (Ln), arachidonic (An), or doco-
sahexaenoic (DHA) acids. Protein complexes with increased molecular weights but
not full polymerization and not involving disulfide bonds in soy protein (Huang et
al., 2006) or aldehydes in bovine serum albumin (Liu and Wang, 2005) suggest the
following sequence mediated by lipid radicals, where Pf is a protein fragment:
It is interesting and somewhat surprising considering current analytical capabili-

ties that even though mechanisms and structures of crosslinks involving aldehydes
have been elucidated in considerable detail, no specific structures or locations of lipid-
induced radical crosslinks have yet been identified.
Polymeric crosslinking from free radicals has been shown with RNAse (Roubal
and Tappel, 1966c; Gamage et al., 1973; Januszewski et al., 2005), lysozyme (Schaich
and Karel, 1975; Funes and Karel, 1981; Funes et al., 1982), b-lactoglobulin (Hidalgo
and Kinsella, 1989), and heavy chain myosin in fish myofibrillar proteins (Ooizumi,
1999). In contrast, cod sarcoplasmic reticulum proteins (Soyer and Hultin, 2000)
and bovine serum albumin (Liu and Wang, 2005) appear to polymerize in fragment-
monomer combinations rather than (P)n polymers, while pepsin and trypsin do not
polymerize yet still develop fluorescence from reaction with carbonyl lipid oxidation
products (Gamage et al., 1973).
Carbonyl-Mediated Crosslinking
Over time, hydroperoxides decompose to secondary products which mediate cross-
linking. The relative crosslinking activity of different aldehydes in neutral solutions
with Fe or Cu catalysts is oxononenal >> acrolein ≅ decadienal > hydroxynonenal ≅
MDA (Yuan et al., 2007); trienals from high PUFA are even more active (Ran, 2004).
Four major crosslinking mechanisms are now recognized for aldehydes; the mecha-
nism that dominates in a given reaction system will be influenced by pH (functional
group dissociation and charge), relative concentrations of amines and carbonyls, ami-
no acid availability and orientation on the protein surface, and solvent environment.
1. Schiff Base Condensation of Amines with Saturated Bifunctional Aldehydes or Other

Dicarbonyls that Provide Bridges Between Protein Chains:
O = C H −C H 2 −C H =O + 2 H 2 N -R -P ro tein P ro tein -R -N =C H −C H = C H −N H -R -P ro tein
This type of crosslink also forms between amines on one protein chain and carbonyl
oxidation sites on a second protein:
P 1 -C H O + P 2 -lys-N H 2 P 1 -C H =N -lys-P 2
Nucleophilic amino acids provide the linkage points and vary with the protein,
for example lysines in RNAse (Chio and Tappel, 1969a, Gerrard et al., 2002), sulf-
hydryls in lens proteins (Riley and Harding, 1993), and mixed histidine and cysteine
links with lysine in b-lactoglobulin (Yuan et al., 2007). Schiff base crosslinks with
conjugated -N=CH-CH=CH- structure are fluorescent and are often are accompa-
nied by a yellow to brown color.
226 K.M. Schaich
2. Michael Addition of Thiols and Amines (primarily lysine and histidine) to C-3 of a,b-
unsaturated Aldehydes and Hydroxyl Alkenals:
Michael addition products may or may not be fluorescent (see Section “Formation
of Fluorescent Adducts and Age Pigments”). Note that the direct Michael addition
only generates lipid-protein adducts. Crosslinking occurs in a subsequent step when
the free aldehyde forms a Schiff base product with a free amine, imidazole, or thiol on
another segment of the same protein or on a neighboring protein:
For histidine, Michael addition is the dominant reaction so the sequence shown
above holds. However, for lysine or cysteine, the crosslinking may also occur in the
reverse order, forming the Schiff base first, followed by Michael addition (Schaur,
2003). The crosslinks occur between single amino acids or between two different
amino acids.
Michael addition-Schiff base crosslinking may be either (or both) intra- and inter-
molecular (Uchida and Stadtman, 1993), as has been observed with 4-hydroxynon-
enal and 4-oxononenal crosslinking of glucose-6-phosphate dehydrogenase (Friguet
et al., 1994a, 1994b), b-lactoglobulin B and human hemoglobin (Bruenner et al.,
1995), RNAse and BSA (Xu et al., 1999a), glyceraldehyde-6-phosphate dehydroge-
nase (Uchida and Stadtman, 1993), cytoskeletal proteins in P19 neuroglial cultures
(Montine et al., 1996), and membrane proteins in erythrocyte ghosts (Hochstein and
Jain, 1981; Beppu, 1986). Trienals, which are particularly strong crosslinkers (much
greater than dienals and HNE), probably follow this reaction sequence with addition
at multiple sites, although mechanism and product structures have not yet been de-
termined (Ran, 2004).
Histidine is a strong copper binder, which contributes to its pro-oxi-
dant activity. One interesting hybrid crosslink occurs when initial Cu2+-me-
diated free radical oxidation of histidine yields an imidazol-2-one product
that contains a reactive 4-oxo-2-ene region susceptible to Michael additions.
Addition of the e-amino group of lysine to C-2 generates a novel crosslink (Liu et al.,
2004):
HN −Ly s
HN N Cu 2+
LOOH
HN N . N N
Lys-NH2
HN NH
H OH O O
There is evidence that Schiff base formation and Michael addition of lysine are
reversible, and that the presence of water or acid favors dissociation of both Schiff
base and Michael adducts (Xu et al., 1999a), hence the crosslinks just described may
be transient and not contribute to permanent crosslinks in aqueous media; they also
are not detectable after acid hydrolysis (Kikugawa and Beppu, 1987; Lin et al., 2005),
which may explain why Tappel’s Schiff base products were only found in chloroform
extracts (Chio and Tappel, 1969b). Both Schiff base and Michael addition prod-
ucts can react further via rearrangement, oxidation, reduction, and secondary addi-
tions (Schaur, 2003), so proposals have been made that stable crosslinks result from
variations in secondary oxidations and rearrangements rather than initial adducts, as
shown in the following mechanisms. Final crosslinks are presented here; more details
of preceding reactions may be found in Section “Addition Reactions of Carbonyl
Products from Lipid Oxidation”.
3. Pyrrole Links Formed when Aldehydes (usually hydroxy or keto derivatives of 2-alke-
nals) Cyclize Between Two Protein Nucleophiles (lys, his, cys):
Pyrrole crosslinks require protein nucleophiles in at least 2:1 molar excess over the
aldehydes. Monoalkylpyrroles form without the second amino acid (1:1 complexes),
but do not contribute to crosslinking (Xu et al., 1999a). The basic process combines
sequential Schiff base and Michael addition of amines to 2-alkenals, as described in
the Michael addition in 2) above, but with additional oxidation (a critical require-
ment) the aldehyde cyclizes into an alkyl pyrrole ring that contains a nucleophilic
adduct from a second amino acid. Dehydration of the iminium yields the pyrrole
(Amarnath et al., 1994; Xu et al., 1998; Schaur, 2003).
This form of crosslinking has been identified in HNE and ONE reactions with
RNAse and b-lactoglobulin (Zhang et al., 2003) and with RNAse and BSA (Xu et al.,
1999a) in model systems. In vivo, 2-pentylpyrrole crosslinks from HNE have been
found in oxidized LDL and atherosclerotic plaques (Salomon et al., 2000), and also
228 K.M. Schaich
are thought to be involved in aggregation of b-amyloid protein in Alzheimer’s disease

(Sayre et al., 1997).
Subsequent reactions of monoalkyl or amine-linked pyrroles lead to many varia-
tions in crosslinking: A) by lysine e-NH2 condensation with the pyrrole –OH to form
cyclic or acyclic mixed aminals (Jirousek et al., 1990), B) by addition of a third pro-
tein nucleophilic link to a double bond in the pyrrole ring (Amarnath et al., 1994),
or C) by oxidation to pyrrole-pyrrole links. Nucleophiles, such as lysine, histidine,

and cysteine, can add to pyrroles to extend crosslinking only when the ring is proton-
ated or attached to a strong electron-withdrawing group. Oxygen activates the ring,
imparting the requisite charge and creating a molecular precursor for crosslinking at
C-2 or C-3 of the double bond (Amarnath et al., 1994), as shown in the following
generalized process starting from an initial pyrrole-lysine adduct:
. . .
O O-
.R
O O
+ O2
R
N
R1 R .N. R1 R +
N
R1
 H+ R
N
1
Crosslink points
+
+
R + R1 R R1 R R1
N N N
Several types of pyrrole dimers or polymers involving this type of addition have
been identified (Amarnath et al., 1994, 1998; Zhu et al., 1994; Xu and Sayre, 1998).
R = R1CH(OH)- for HNE and ONE in the example below.
Epoxy alkenals form pyrroles with a hydroxylated side chain that provides an elec-
trophilic site for crosslinking (Zamora and Hidalgo, 1995; Hidalgo and Zamora,
2000):
−H2O
R 1 C H (O H ) R 1 C H (OH) CH
N N R1 N −H2O
R2 R2 R2
R 1 C H (O H)
N R 1 CH C
R2 Higher polymers N R1 N
formed by repeated R2 R2
additions
Consistent with this pyrrole activation also is a 2:2 pyrrole complex proposed to
explain the very rapid crosslinking of proteins by 4-ketoaldehydes (Xu et al., 1999b;
Xu and Sayre, 1999). Protein bound lysines form Schiff base adducts at each of the
carbonyls; these then undergo end-to-end aldol condensation, cyclization of the alde-
hydes to linked pyrroles, and finally intramolecular pyrrole electrophilic substitution
to generate a bicyclic compound:
4. Pyridinium Crosslinks Formed by Sequential Michael Addition and Schiff Base Re-
actions Under Conditions of Excess Aldehydes. The initial ring structure results from
condensation of two MDA and a saturated aldehyde with an amine (Esterbauer et al.,
1991); the crosslinks form by subsequent Schiff base reactions with the free aldehydes
(Kikugawa et al., 1985). The specific structure of the pyridine is dictated by the degree
of aldehyde excess and nature of alkanal condensing with MDA (Beppu, 1986).
230 K.M. Schaich
Pyridine crosslinks between lysines crosslinked erythrocyte proteins in red blood

cell ghosts incubated with MDA (Kikugawa et al., 1985).
Considering all these potential pathways, it is clear that any attempt to attribute
protein crosslinking to a single lipid oxidation product or a single type of crosslinking
is an oversimplification at best. Every type of lipid oxidation intermediate or product
induces protein crosslinking. In addition, crosslinking will start with free radical pro-
cesses in actively oxidizing lipid, but if the reaction is followed long enough, carbonyl-
mediated crosslinking eventually occurs as well, superimposed on previous oxidation
and crosslinking processes. Therefore, multimodal protein crosslinking by lipids, with
different kinetics, different products, and different effects should definitely be ex-
pected in most proteins. The dominant mode of crosslinking occurring will vary with
the amino acid composition and the configuration of the protein, as well as the stage
of lipid oxidation and the variety of products present. With actively oxidizing lipids
(not isolated products), the dominant mode of crosslinking detected will vary with
the length of reaction and time point at which proteins are analyzed.
For example, cytochrome c embedded in model membranes with oxidizing cardi-
olipin first crosslinked and aggregated into intramembranous particles by free radical
reactions; with extended reaction, more extensive crosslinking by lipid aldehyde ad-
ducts led to the breakdown of membrane vesicles and formation of globular lipopro-
tein complexes in particles (Borovyagin et al., 1984). b-Lactoglobulin reacted with
13-LOOH in emulsion showed similar multimodal crosslinking, with rapid initial
-S-S- crosslinking (dominant product) and slower C-C crosslinking; the S-S dimers
also underwent further additive C-C crosslinking over time while retaining the S-S
links (Hidalgo and Kinsella, 1989). Globular proteins with reactive side chains cross-
link via surface links involving both radical and aldehyde reactions, while crosslinking
of structural proteins, such as collagen and its derivatives, apparently involves the
peptide chain itself. Inaccessibility of backbone sites to carbonyls severely limits the
types of crosslinking possible, and may contribute to the greater tendency of collagen
to degrade rather than crosslink when oxidized.
Multimodal crosslinking is more damaging than any single mode alone and prob-
ably explains why in most studies kinetics of appearance or disappearance of single
lipid oxidation products do not exactly match crosslinking processes. The likelihood
of multimodal crosslinking also means that measuring only one type of polymeriza-
tion mechanism will, in most cases, underestimate and misrepresent the reactions
actually occurring.
Fragmentation
The flip side of radical recombination to crosslink proteins is radical scission to gen-
erate peptide fragments. The possibility of protein fragmentation by oxidizing lipids
is noted frequently in literature introductions and discussions, but occurrence ap-
pears to be limited mostly to collagen and related structural proteins. Lipid-mediated
oxidation of LDL also leads to extensive scission of apoB-100 into smaller peptides
(Fong et al., 1987). Lyophilized gelatin-methyl linoleate mixtures undergo scission
rapidly when incubated dry at 50°C, but as moisture content increases, fragmentation
changes progressively to crosslinking (Zirlin and Karel, 1969; Matoba et al., 1984a).
Connective tissue proteins (collagen) are degraded when lipids oxidize in cultured
chondrocytes (Tiku et al., 2000). In a and g-conglutinins from lupines reacted with
13-MLOOH at pH 9, protein scission occurs and the resulting fragments recombine
with monomers to give a continuous range of protein complexes with increased mo-
lecular weights (Lqari et al., 2003). These observations suggest that fragmentation
occurs only under extreme conditions and primarily with select proteins that have
sensitive amino acid sequences.
To explain this sensitivity, Zirlin and Karel proposed that scission occurs in a
process that parallels lipid oxidation, that is a-carbon radical → peroxyl radical →
hydroperoxide → HOO- scission to alkoxyl radical → b scission to break peptide
chain and generate protein carbonyls (Zirlin and Karel, 1969). Most aspects of this
sequence have since been proven correct. Spin trapping of protein radicals has verified
the presence of a-carbon radicals generated both directly and by migration of radicals
initially formed on alkyl side chains (Headlam et al., 2000). Although Matoba et
al. argued that only a-scission of the alkoxyl radical was consistent with a dramatic
increase in amide and a much lower increase in carbonyl products in proteins reacted
with oxidizing lipids (Matoba et al., 1984a), both a and b scission pathways for pro-
tein alkoxyl radicals have been documented for radiation and HO•, leading to differ-
ent degradation patterns (Stadtman and Berlett, 1997; Stadtman and Levine, 2003;
Kowalik-Jankowska et al., 2004; Davies, 2005).
Detailed reaction mechanisms for the diamide and a-amidation protein fragmen-
tation pathways have been developed (Garrison, 1987; Davies et al., 1995; Stadtman
and Berlett, 1997; Stadtman and Levine, 2003). A simplified version is presented
in Fig. 8.17. C-C or b-scission is an oxidative process that decarboxylates the target
amino acid (Garrison, 1987). Although it is the dominant process in oxygenated
systems, b-scission is not a spontaneous process. It occurs under conditions of mild
hydrolysis, which explains Zirlin and Karel’s early observations that scission rarely oc-
curred in dry proteins but increased with moisture content (Zirlin and Karel, 1969).
Relatively minor differences in sample treatment or preparation for electrophoresis
leading to hydrolysis may also account for unexpected globular protein fragmentation
(Hunt et al., 1988; Soyer and Hultin, 2000; Liu and Wang, 2005). In contrast, N-C
or a-scission is a reductive process that deaminates the target amino acid (Garrison,
1987). Both scission processes generate amides (marked with a star), the carbonyl
products detected in standard assays. However, mild hydrolysis converts diamides to
232 K.M. Schaich
Fig. 8.17. Protein oxidation and fragmentation processes resulting from transfer of free radicals from oxidizing
lipids to a-carbon sites on proteins. Adapted from (Garrison, 1987; Davies et al., 1995; Stadtman and Berlett,
1997; Stadtman and Levine, 2003). R and R2 are amino acid side chains; R’ and R’’ are continuations of the peptide
chain. * denotes carbonyl compounds detected in oxidation assays.
amines, removes carbonyls used to diagnose and quantitate protein oxidation, and
alters product distributions. This sensitivity to hydrolysis underscores the necessity to
control reaction conditions scrupulously and to consider all the chemistry involved
when interpreting results.
Collagen, structural proteins, and proteins with extensive a-helix regions may be
especially susceptible to scission because they have more glycines, prolines, and aliphatic
amino acids that are the sites of a-radical production. The proline residues of collagen
oxidize readily when exposed to HO• radicals (Kato et al., 1992). Similarly, in gelatin
reacted with LO(O)• radicals, proline, hydroxyproline, and glycine show the greatest
loss (Matoba et al., 1984a). The a-carbon of proline is in the ring; oxidation at that
point yields a 2-pyrrolidone compound upon scission (Kato et al., 1992).
Peptide chain cleavage via oxidation of glutamyl and aspartyl residues (side chain
decarboxylation and deaminative scission of the peptide chain) has been shown with
HO• (Garrison, 1987; Stadtman and Levine, 2003), but comparable degradation by
lipid oxidants has not yet been reported. One study claimed that LO• generated from
phospholipid hydroperoxides by Cu+ induced fragmentation of bovine serum albu-
men (Hunt et al., 1988). This system could be unusually damaging since BSA binds
Cu+ rather extensively; multiple sites for cage reactions on the BSA surface then would
generate reactive alkoxyl radicals near backbone sites that may ordinarily not be ac-
cessible. However, Cu+ autoxidizes and can damage proteins directly, yet controls
accounting for this action were not run. Thus, whether oxidizing lipids can fragment
globular proteins in solution remains to be demonstrated.
Formation of Fluorescent Adducts and Age Pigments

Tappel first demonstrated that fluorescent ceroid age pigments in animal tissues and
lipid-protein browning products in foods were co-oxidation products of polyunsat-
urated lipids and protein (Tappel, 1955), and research from his group was largely
responsible for identification of the N,N’-disubstituted 1-amino-3-iminopropene
(Schiff base) structures produced by reaction of carbonyl lipid oxidation products,
particularly MDA, with side chain and terminal amino groups on proteins as the
source of fluorescence (Chio and Tappel, 1969b; Fletcher and Tappel, 1970; Dillard
and Tappel, 1971, 1973, 1984; Fletcher et al., 1973; Malshet and Tappel, 1973):
Lysine is by far the most reactive side chain, followed by histidine, tryptophan, and ar-
ginine. However, despite a common and mistaken expectation that all aldehydes gen-
erate fluorescent Schiff base products with amino acids, adducts are fluorescent only
when an electron-donating group is present in conjugation with the imine (Malshet
and Tappel, 1973). This explains, in part, why many of the MDA– and other al-
dehyde–protein products are not fluorescent; it points out the need to determine
product structures and consider alternate pathways when comparing “reactivity” of
different carbonyl products of lipid oxidation by measuring fluorescence.
Since Tappel’s pioneering work, appearance of fluorescence has become a hall-
mark of protein oxidation reported in hundreds of articles, routinely attributed to
and associated with generation of MDA. However, over time questions have been
raised about the aldehyde sources of fluorescent lipid-protein adducts and the reac-
tion mechanism and final structure of fluorescent adducts. Some observations that
have led to disputes include:
• MDA is only produced in PUFA with three or more double bonds, yet fluorescence
develops in some reactions with oxidizing linoleic acid, which does not generate
MDA.
234 K.M. Schaich
• Many lipid-derived aldehydes with varying chain lengths, both saturated and
unsaturated, produce fluorescent products with proteins and amino acids.
• Fluorescent structures other than Schiff-base have been identified.
• Fluorescent iminopropene structures can be generated by reactions other than

with aldehydes.
• Production of fluorescence does not always directly parallel loss of amino

groups.
• Fluorescence characteristics (excitation and emission maxima as well as intensity)

change over time as lipid oxidation progresses.
• In model systems, concentrations of aldehydes required to generate fluorescent

products with proteins or amino acids are usually several orders of magnitude
higher than aldehyde levels formed by lipids autoxidizing in foods or in vivo
under conditions of oxidative stress, yet fluorescence develops rapidly in a wide
range of “real” systems.
It is now clear that the chemistry underlying production of fluorescent prod-

ucts is much more complicated than originally proposed. Fluorescence arises both
from some adducts and specific crosslinks formed between proteins and a family of
lipid oxidation products—hydroperoxides as well as a range of secondary products
(Kikugawa and Beppu, 1987; Hidalgo et al., 1999; Zamora and Hidalgo, 2003a).
Simple Schiff base adducts (-N=CHCH2-CHO or -NH-CH=CH-CHO) contrib-
ute to browning (Zamora et al., 2000) but are not fluorescent (Itakura and Uchida,
2001). The “classic” linear iminopropenes (RNH-CH=CH-CH=NR) to which most
fluorescence is attributed are the initial products formed via Schiff base and Michael
additions of 2-alkenals (Nadkarni and Sayre, 1995), but they form primarily under
acidic conditions (where fluorescence is lowest), require high concentrations of reac-
tants (Kikugawa and Beppu, 1987), and remain relatively minor products in most
systems because they rapidly rearrange to other products. NMR and LC-MS/MS
have identified pyridinium (Kikugawa and Ido, 1984; Kikugawa et al., 1984; Amar-
nath et al., 1998; Liu and Sayre, 2003) and pyrrole (Zhu et al., 1994; Hidalgo et al.,
1999; Zamora et al., 2000; Liu et al., 2003; Zamora and Hidalgo, 2003a; Zhang
et al., 2003; Zamora and Hidalgo, 2005) lipid-protein products in both monomer
forms and crosslinks as the main sources of strong fluorescence that develops rapidly
at physiological pH and at elevated temperatures in foods. Notably, these molecular
structures contain the imino-ene conjugation required for fluorescence, as specified
by Malshet and Tappel (1973). Other less dominant fluorescent products have been
detected, but their structures remain to be identified.
Production of fluorescent co-oxidation products requires elevated temperatures.
As will be shown later, little to no fluorescence develops in reactions at refrigerated
or room temperatures, but a physiological temperature of 37°C is sufficient to drive

rapid production of fluorescent protein adducts that are markers of oxidative damage;
at accelerated shelf life testing of 60°C and higher food-processing temperatures fluo-
rescent compounds become major products that contribute to characteristic brown-
ing and flavor production (Hidalgo et al., 1999, 2005).
Spectral characteristics of fluorescent products in various lipid oxidation systems
are presented in Table 8.2 to show how fluorescence characteristics change with lipid
oxidation product and amino acid or protein target. It also illustrates how assignment
of structures based solely on excitation and emission maxima, although a common
practice, is problematic and can lead to erroneous conclusions. Multiple products
usually form in the same system. Averages of multiple products and differences in
product distributions under different reaction conditions contribute to the variation
in excitation and emission maxima that have been reported for different amino ac-
ids, proteins, and oxidants (Table 8.2). Solvents alter spectral characteristics: ex/em
maxima and emission intensity both usually increase with solvent polarity (Kikugawa
and Beppu, 1987; Chen et al., 1996). Sample handling is also critical. Traditional
chloroform-methanol extracts of tissues develop artificial blue fluorescence that is dis-
tinctly different from the yellow fluorescence of age pigments when exposed to light
or passed through silica gel columns (Kikugawa, 1994; Kikugawa et al., 1994). The
blue fluorescence clearly arises from secondary oxidation processes, probably photo-
sensitization by traces of heme compounds extracted in methanol and decomposi-
tion of preformed hydroperoxides by silica interactions and exposure to UV radiation
from lab lights. Hence, great care must be exercised to obtain fluorescence data that
accurately reflects system chemistry.
Monofunctional Aldehydes. Alkanals

Saturated aldehydes produce Schiff base adducts for which the fluorescence depends
on amine source, temperature, and amine-aldehyde concentrations (both absolute
and relative), and in some cases on the presence of oxygen (Kikugawa and Beppu,
1987). Glycine does not produce fluorescent products with hexanal, decanal, or acet-
aldehyde when incubated at refrigeration temperatures (4°C). Under the same condi-
tions, products with lysine were only weakly fluorescent (Stapelfeldt and Skibsted,
1996; Veberg et al., 2006); even though emission intensities increased at elevated
temperatures (30 or 37°C), absolute levels of fluorescence remained low (Kikugawa
et al., 1985; Yamaki et al., 1992). However, when oxygen or a strong oxidizer, such
as H2O2, was present, strong fluorescence traced to 2-HO-1,2-dihydropyrrol-3-ones
developed (D, Fig. 8.18); a mechanism was proposed in which the aldehydes dimer-
ize through Aldol condensation and then oxidize to tricarbonyls. Amines add to this
compound rather than parent alkanals, and the complex cyclizes (Chen et al., 1996).
Oxidation steps are also critical to develop fluorescent pyrroles after alkenal reaction
with amines. Thus, it is clear that secondary oxidation of intermediate products sig-
nificantly affect lipid-protein co-oxidations by shifting pathways and altering yields
of fluorescent versus non-fluorescent products. Such differences become critically im-
236 K.M. Schaich
Table 8.2. Variations in Excitation and Emission Maxima for Schiff Base Fluorescence from Different Amino Acid and
Protein Substrates.a Unless otherwise noted, the oxidant is oxidizing lipids rather than an isolated product.
Amine Oxidant Solvent Ex (nm) Em(nm) Reference

Lysine MDA CHCl3 395 470 (Chio and Tappel,
1969a)
Lysine hexanal 10% ethanol 345 415 (Dalsgaard et al., 2006)
Lysine heptadienal 382 434 (Yamaki et al., 1992)
Lysine alkanals 382 434 (Yamaki et al., 1992)
Polylysine LOOH, AnOOH pH 7 buffer 330 425 (Fruebis et al., 1992)
Polylysine MDA pH 7 buffer 398 470 (Fruebis et al., 1992)
Glycine LOOH 0.5% SDS, 360–370 435–450 (Chio and Tappel,
pH 7 buffer 1969b; Shimasaki et
al., 1982)
Glycine MDA water 370 450 (Chio and Tappel,
1969b)
Glycine alkanals+ H2O2 345–375 415–440 (Chen et al., 1996)
Glycine heptadienal 397 470 (Yamaki et al., 1992)
Ethylamine HO-butanal pH 7 buffer 327 390 (Liu and Sayre, 2003)
Valine MDA water 370 450 (Chio and Tappel,
1969b)
Leucine MDA water 370 450 (Chio and Tappel,
1969b)
Leucine b-oxyacrolein neat 390 550 (Buttkus, 1975)
Leucine b-oxyacrolein n-butanol 385 460 (Buttkus, 1975)
Glutathione LOOH pH 7 buffer 350 440 (Zamora et al., 1989)
Lysozyme oxidizing ML dry emul- 355 425 (Leake and Karel, 1985)
sion
BSA LOOH, SP pH 7 buffer 350 425 (Hidalgo et al., 1999)
BSA LOOH, AnOOH pH 7 buffer 330 425 (Fruebis et al., 1992)
BSA LOOH pH 7 buffer 360 427 (Shimasaki et al., 1982)
BSA HNE pH 7 buffer/ 360 430 (Xu and Sayre, 1998;
MeOH Xu et al., 1999b)
RNAse oxidizing EtAn or pH 7 buffer 390,395 470 (Chio and Tappel,
MeLn 1969b)
RNAse 2-HO-heptanal pH 7 buffer 325 409 (Liu and Sayre, 2003)
Porcine oxidizing fibrils pH 6 buffer 395 485 (Chelh et al., 2007)
Myofibrils
Minced meat 5 aldehydes (solid) 382 450–550 (Veberg et al., 2006)
CHCl3- 327 415
(Aubourg et al., 1998)
Frozen MeOH,
oxidizing lipids
sardines MeOH- 393 463
(Aubourg et al., 1998)
water
Table 8.2., cont. Variations in Excitation and Emission Maxima for Schiff Base Fluorescence from Different Amino Acid
and Protein Substrates.a Unless otherwise noted, the oxidant is oxidizing lipids rather than an isolated product.
Amine Oxidant Solvent Ex (nm) Em(nm) Reference

Liver homog- oxidizing lipids 397 480 (Li et al., 2006)
enate
LDLox atherosclerotic pH 7 buffer 360 430 (Xu et al., 2000)
plaques
LDLox HNE or ONE pH 7 buffer/ 366 445 (Xu et al., 2000)
MeOH
Yellow lipo- in vivo 0.5% SDS, 400 620 (Kikugawa et al., 1994)
fuschin pH 7 buffer
Blue lipofus- in vivo CHCl3-MeOH 350 430 (Kikugawa, 1994)
chin
Amine pyri- MDA pH 7 buffer 403 462 (Kikugawa and Beppu,
dinium 1987)
Amine pyri- MDA CHCl3 390 446 (Kikugawa and Beppu,
dinium 1987)
a
Abbreviations: MDA, malonaldehyde; LOOH, linoleic acid hydroperoxide; BSA, bovine serum
albumin; SP, secondary products of lipid oxidation; AnOOH, arachidonic acid hydroperoxide; HNE,
hydroxynonenal; EtAn or MeLn, ethyl arachidonate or methyl linolenate; LDLox, oxidized low den-
sity lipoproteins; ONE, oxononenal; SDS, sodium dodecyl sulfate.
portant when they are used to interpret mechanism and determine molecular causes
of protein damage, and argue for careful control of oxygen concentrations in reaction
media and determination of oxygen dependence of reaction pathways in all experi-
ments.
Excitation wavelengths and solvent also impact detected fluorescence: in 50%
ethanol with excitation at 382 nm, emissions were absent or only barely detectable
(Veberg et al., 2006), while in other studies emissions were detected with ex 327/em
415 nm or ex 393/em 463 nm (Aubourg et al., 1998), ex 348/em 416 nm (Yamaki et
al., 1992), ex 357/em 430 mm (Kikugawa et al., 1985). Both excitation and emission
wavelengths tend to increase with solvent polarity (Kikugawa and Beppu, 1987); this
must be considered when comparing results between different systems.
Perhaps most importantly, amines appear to stimulate reactivity of saturated
aldehydes, inducing increased aldol self-condensation and generation of 2-alkenals
which cyclize to fluorescent products in the presence of oxygen (Suyama et al., 1981).
Thus, alkanals may generate the same spectrum of products as alkenals, but at lower
yields and slower rates.
Bifunctional Aldehydes
As shown previously, at neutral pH the monofunctional iminopropene MDA–lysine
adduct (R-NH=CH-CH2-CHO) does not fluoresce (Itakura and Uchida, 2001) be-
cause it lacks an electron-donating group in conjugation with the imine. The bi-
238 K.M. Schaich
R1
R1 Prote i n R2 OH
N
+
R2 N
(non-fluorescent) Prote i n
A B
+
Prote in - HN Prote in - HN O
HO HO HO
+
R N R N R N
Prote i n Prote i n Prote i n
C D
Fig. 8.18. Examples of fluorescent structures formed when lipid aldehydes react with lysine e-amino groups of
proteins.
functional iminopropene MDA–lysine adduct, however, does contain the necessary

conjugation so the crosslink R-NHCH=CHCH=N-R is indeed fluorescent and may
be the major source of fluorescence very early in reactions of actively oxidizing lipids
(not isolated aldehydes) with proteins (Nadkarni and Sayre, 1995). As the reaction
progresses, however, stronger fluorescence arises from pyridinium products (dihydro-
pyridine dicarbaldehydes) formed by the condensation of two bifunctional aldehydes
with primary amines, particularly the e-NH2 of lysine (Kikugawa and Beppu, 1987;
Itakura and Uchida, 2001). Structure A shown below is formed when MDA reacts
with an amine group either on a side chain linked to a protein or a peptide termi-
nal amine. Similar fluorescent pyridine structures form in a ternary complex where
monofunctional aldehydes condense with two molecules of MDA and an amine, in-
troducing R-CHO at C-2 of MDA (Kikugawa et al., 1984; Freeman et al., 2005). In
structure B, the third aldehyde is acetaldehyde, a decomposition product of MDA.
Alkenals
Alkenals are significant precursors of fluorescence when reacted with proteins because
they form Schiff base adducts, then undergo Michael addition at the double bond,
and the amine-aldehyde-amine complex then rearranges to fluorescent pyrroles and
(under some circumstances) pyrimidines. Hydroxy, epoxy, and oxo alkenals are the
most reactive alkenals. When reacted with lysine, the major amino acid target, these
aldehydes produce a variety of cyclic products, including non-fluorescent 4-alkyl-
imidazolium crosslinks through Amadori rearrangements of initial Schiff base ad-
ducts (Fig. 8.18a) (Liu and Sayre, 2003), fluorescent 4,5-dialkyl-3-HO-pyridiniums
(Fig. 8.18b) (Liu and Sayre, 2003), and 2-hydroxyalkyl-3-imino-1,2-dihydropyrrol
derivatives (Fig. 8.18c) by a series of Schiff base and Michael additions plus oxida-
tions (Tsai et al., 1998; Xu and Sayre, 1998; Hidalgo et al., 1999; Zamora and Hi-
dalgo, 2003a). 2-HO-2-pentyl-1,2-dihydropyrrol-3-one iminium (D, Fig. 8.18) is a
key structure on LDLox contributing to fluorescence in atherosclerotic plaques (Xu
et al., 2000). Note that the bond sequence (-NH-CH=CH-CH=N-) initially cited by
Tappel as being required for fluorescence is contained within each of these structures,
and it is missing in the non-fluorescent imidazolium.
Lipid Hydroperoxides
It is particularly important to recognize potential contributions of radicals and hy-
droperoxides since these products are generally ignored in global interpretations of
fluorescence origins. In model systems with oxidized phospholipids and lysozyme,
the precursor for fluorescence was a monomeric product of hydroperoxides formed
without fragmentation (structure not identified); hydroxyl, keto, and epoxy deriva-
tives as well as monofunctional aldehydes, such as hexanal and 2,4-decadienal, did
not yield fluorescent products (Iio and Yoden, 1988). When hydroperoxides and
various oxidation products (e.g., propanal, butanal, pentanal, and hexanal) of methyl
linoleate were reacted with b-lactoglobulin in pH 7.4 phosphate buffer at 37°C, the
aldehydes reacted quite rapidly with free amines (~20% in 2 h) but produced only
weak or no fluorescence (Hidalgo and Kinsella, 1989). In contrast, the slow reaction
of isolated 13-MLOOH with amines (~10% over 25 h) was accompanied by genera-
tion of strong fluorescence associated with a C-C or C-N dimer. Binding of 1 mole
MLOOH per 18,000 mw protein did not correlate with fluorescence; thus, radicals
formed rapidly on tryptophan and more slowly on other amines or their recombina-
tion products were judged to be the source of fluorescence. In many other reports
protein fluorescence developed in parallel with lipid hydroperoxides, yet the reaction
was attributed to aldehydes. Thus, although mechanisms have not been elucidated,
it is likely that LOOH always contribute at least in part to the development of fluo-
rescent protein products, and this needs to be considered when interpreting damage
mechanisms and causality.
The potential to form multiple fluorescent products as lipid oxidation progresses
argues strongly for not using fluorescence to attribute damage sources without detailed
identification of fluorescent products and their structures. When interpreting mecha-
240 K.M. Schaich
nisms from fluorescence, it is important to remember that fluorescence characteristics

vary with the amino acid and lipid involved in the complex (Table 8.2), the degree
of lipid oxidation, and reaction conditions (particularly pH, solvent, temperature,
presence of metals, and sensitivity to reduction) (Kikugawa and Beppu, 1987; Kagan,
1988; Li et al., 2006; Veberg et al., 2006). As shown in Maillard reaction chemistry,
the lipid carbonyl-protein amine condensation cannot occur at acid pH where the
amine is protonated; it accelerates as pH increases and is very rapid in alkaline solu-
tions (Hidalgo et al., 1999). In model systems of BSA reacted with LOOH and its
secondary products, fluorescence maxima occurred at pH 10; maximum arginine loss
was also at pH 10, but highest lysine loss and color formation was observed at pH 7
(Hidalgo and Zamora, 1993; Zamora and Hidalgo, 1995). Fluorescence production
maxed at 37°C and decreased at higher temperatures, even though pyrroles usually
assumed to be fluorescent continued to increase. These observations emphasize the
complexity of oxidizing lipid reactions with proteins, showing clearly that multiple
reactions occur simultaneously and that attribution of macroscopic behaviors to
products without isolation and structural identification can lead to serious errors.
One additional cautionary comment must be offered. Only recently have prod-
ucts begun to be quantitated, and it is not uncommon to find that fluorescent prod-
ucts that seem to be major because of their intense emissions actually account for a
very small percentage of total products. For example, in one study, 2-heptenal was
reacted with RNAse and strong fluorescence was observed. However, heptenal com-
plexed only 1.5% of the lysine after being incubated with RNAse at 25°C for 10
days, and the reactions required very high aldehyde concentrations. Even so, the al-
dehyde-amine reaction was proposed as the major damaging reaction in that model
system. With lipid oxidation products, attention focused too intently and exclusively
on single products or processes loses all the action in other reactions that may be as
much or even more important in overall damage.
Inhibition of Enzyme Activity

The chemical and physical modifications of proteins described previously—amino
acid radical formation, oxidation, or adduct formation; covalent and non-covalent
lipid binding; and changes in protein conformation, denaturation, crosslinking, and
scission—all impair protein function. Indeed, one of the earliest observations of lipid
oxidation effects on proteins was marked loss of activity in enzymes in both model
systems and in tissues. Sulfhydryl enzymes are the most sensitive (Tsen and Tap-
pel, 1958; Little and O’Brien 1967, 1968); thiols react rapidly with lipid radicals
and hydroperoxides, and they also undergo nucleophilic additions to double bonds
(Gardner, 1979). Metallo-enzymes, enzymes of the mitochondria, microsomes, endo-
plasmic reticulum, and respiratory proteins such as cytochromes, are also extensively
damaged by oxidizing lipids (Table 8.3).
Extensive studies of damage to RNAse and lysozyme as model enzymes have
revealed some factors that critically influence the type and extent of damage by ox-
idizing lipids (Menzel, 1967; Little and O’Brien, 1968; Chio and Tappel, 1969a;
Matsushita et al., 1970). Both pH, which affects the protein charge and conforma-
tion, and lipid hydroperoxide concentration (which influences contact, potential for
hydrogen bonding, and oxidizing potential) are important. RNAse is inactivated at
acidic but not alkaline pH, presumably due to oxidation of the histidine residue and
potentially lysine and methionine in the active site (Matsushita et al., 1970). LOOH
binding to proteins, which caused changes in conformation and increased amino
acid accessibility, was considered critical. Binding effects are concentration dependent
(Little and O’Brien, 1968). At low concentration LOOH impedes –SH oxidation by
hydrophobic bonding to proteins, but at high concentrations LOOH binding leads
to denaturation and exposure of new reactive sites.
The speed with which enzymes are inactivated by oxidizing lipids in both model
systems and tissue studies suggests that reactions of LOO• and LO• radical attack on
cysteine, tryptophan, lysine, and histidine side chains is the major damaging process
for actively oxidizing lipids. Obviously, reactions with carbonyl products can also in-
hibit enzymes at later stages of lipid oxidation. However, even though model system
studies have suggested a number of feasible explanations to inactivate a few enzymes,
the exact oxidant and mechanisms responsible in complex systems where lipids are ac-
tively oxidizing and multiple oxidation products are present at different times remain
to be identified.
Alteration or Loss of Biological Function

It seems obvious that any functions, both immediate and downstream, that are de-
pendent on protein composition, structure, conformation, or binding properties will
be affected by lipid-induced modifications. Just a few examples are described here to
illustrate how the impact of modified proteins in most cases reaches far beyond the
immediate interaction chemistry.
Covalent attachment of HNE to myoglobin or oxymyoglobin increases oxida-
tion to metMyb and inhibits enzymatic reduction and recycling back to myoglobin
(Bolgar and Gaskell, 1996; Faustman et al., 1999; Lynch and Faustman, 2000; Alder-
ton et al., 2003). The tertiary structure of the apoprotein apparently is not affected
(Alderton et al., 2003), but HNE binding to critical histidine residues via Michael
addition (Bolgar and Gaskell, 1996; Alderton et al., 2003) alters heme binding or
orientation in a way that increases heme catalysis of lipid oxidation (Faustman et
al., 1999; Lynch and Faustman, 2000). In vivo, rapid conversion of myoglobin to
metmyoglobin and inability to repair and recycle oxidized metMb disrupts oxygen
transport to tissues and forces muscles into anoxic respiration, markedly impairing
muscle function and increasing exercise fatigue rate.
Enzyme inactivation by oxidizing lipids can be devastating in cells far beyond the
damage to the individual protein molecules and reactions. To cite just a few longer-
range implications, lipid reactions with mitochondrial enzymes inhibit or interrupt
metabolic pathways at multiple points (McKnight and Hunter, 1966; Benedetti et
al., 1979; Thomas and Poznansky, 1990; Tsuchiya et al., 2005), while reaction with
cytochromes in mitochondria leads to loss of respiratory control and decouples phos-
242 K.M. Schaich
Table 8.3. Examples of Enzymes Damaged by Various Forms of Oxidizing Lipids.

Enzyme Damage Agent1 Reference
SH enzymes
Succinoxidase, amino oxidase Light-oxidized ML and MLn (Ottolenghi et
al., 1955)
Succinoxidase, amino oxidase, oxidizing L and Ln (Wills, 1961a,
urease, papain, glyoxylase, choline 1961b)
oxidase, b-amylase, cytochrome c
oxidase
Catalase oxidizing L (Roubal and Tap-
pel, 1966a)
Papain oxidizing ML (Lewis and Wills,
1962)
Papain MDA (Shin et al., 1972)
Isocitrate DHG LOOH (Green et al.,
1971)
Alcohol DHG, papain, chymotrypsin, oxidizing EtLn or EtAn (Chio and Tappel,
carboxypeptidase A, urease 1969a)
Protein disulfide isomerase HNE (Carbone et al.,
2005)
Mitochondrial enzymes
NADP+-isocitrate DHG HNE, isolated rat heart mitochondria (Benderdour et
al., 2003)
Isocitrate, succinate, and malate DHG oxidizing mitochondrial membranes (McKnight and
Hunter, 1966)
Aldehyde and glyceraldehyde trans-HNE (Mitchell and
phosphate DHG Petersen, 1991)
Aldehyde DHG, glucokinase, LOOH and SP (secondary products) (Kanazawa and
glyceraldehyde PO4 DHG Ashida, 1991)
Microsomal Enzymes
Endoplasmic reticulum enzymes LOOH (Hochstein and
Ernster, 1964)
Glucose-6-phosphatase oxidizing microsomal lipids (Hruszkewycz
et al., 1978;
Benedetti et al.,
1979)
Glycerol-3 phosphate acyl transferase oxidizing microsomal lipids in situ (Thomas and
Poznansky,
1990)
Gulonolactone oxidase oxidizing microsomal lipids (McCay, 1966)
Glyceraldehyde-6-phosphate DHG HNE, HHE (Tsuchiya et al.,
2005)
Table 8.3., cont. Examples of Enzymes Damaged by Various Forms of Oxidizing Lipids.
Enzyme Damage Agent1 Reference
Glucose-6-phosphate DHG HNE (Friguet et al.,
1994a)
Glucose-6-phosphate DHG LOOH and SP (secondary products) (Kanazawa and
Ashida, 1991)
Respiratory enzymes
Cytochrome c LOOH (O’Brien and
Frazer, 1966;
Little and
O’Brien, 1968)
Cytochrome c peroxidizing Ln (Desai and Tap-
pel, 1963)
Cytocrome P450 oxidized microsomal lipids (Hruszkewycz et
al., 1978)
Cytochromes oxidized microsomal lipids (McKnight and
Hunter, 1966)
General
Lysozyme oxidizing ML (Funes et al.,
1982)
Lysozyme MDA (Chander et al.,
1981)
RNAse, lysozyme, creatine kinase, peroxidizing MLn; high conc. MDA (Chio and Tappel,
lactate dehydrogenase 1969a)
RNAse, trypsin, pepsin, chymotrypsin LOOH and SP (secondary products) (Matsushita et
al., 1970; Mat-
sushita, 1975)
Cathepsin B HNE (Crabb et al.,
2002)
Lecithin:cholesterol acyl transferase PL-OOH (Mickel et al.,
1972; Bielicki
and Forte,
1999)
1
bbreviations: DHG, dehydrogenase; EtLn, ethyl linolenate; EtAn, ethyl arachidonate; HNE,
A
hydroxynonenal; HHE, hydroxyhexenal; LOOH, linolenic acid or ester hydroperoxide; ML, methyl
linoleate; MLn, methyl linolenate; MDA, malonaldehyde; PL-OOH, phospholipid hydroperoxides.
phorylation (Hruszkewycz et al., 1978). Reaction of hydroxynonenal with membrane
proteins decreases neuronal plasticity, disrupts mitochondria, and impairs glutamate
transport (Keller et al., 1997). Inhibition of protein disulfide isomerase via cysteine
modification results in incorrect disulfide formation in newly synthesized proteins
(Carbone et al., 2005), an effect that is especially important when glutathione is de-
pleted under stress conditions. Oxidation of methionines in the binding site of HDL
apo A1 and A2 impairs lipid binding and pick up, thus depressing HDL ability to
promote efflux of cholesterol from cells; alterations in secondary structure of the apo-
244 K.M. Schaich
protein further affect the ability to interact with lipids critical for cholesterol removal
and activation of LCAT (Garner et al., 1998a). Similarly, complexation of lysine in
LDL impedes LDL binding to the epithelial receptor and subsequent LDL removal
from plasma (Steinbrecher, 1987).
The type of protein damage generated by lipids markedly affects hydrolysis and
recycling of proteins. Proteases are specifically induced to remove damaged proteins
in vivo. Disruption of secondary and tertiary structure, increased hydrophobicity, or
moderate oxidation of side chains all act as signals to activate proteases and increase
proteolytic susceptibility (Davies, 1987b), particularly by multicatalytic proteinase or
proteasomes (Friguet and Szweda, 1997). The ubiquitin/proteasome system, for ex-
ample, selectively degrades oxidized proteins (Grune et al., 1997; Grune and Davies,
2003). Reaction of hydroxynonenal and hydroxyhexenal with glyceraldehyde-6-phos-
phate dehydrogenase inhibits enzyme activity and renders the enzyme susceptible to
proteolysis by a giant chymotrypsin-like serine protease TPP II and lysosomal en-
zymes (Tsuchiya et al., 2005).
In contrast, aggregation, crosslinking and high lipid binding reduce degradability
of the proteins, as has been demonstrated with G6PDH (Friguet and Szweda, 1997),
muscle myofibrillar proteins (Morzel et al., 2006), BSA (Zamora and Hidalgo, 2001),
and other proteins. This should be expected since proteases with specific cleavage
sites will be inhibited if the target amino acids are occluded, blocked, or are in or
near scission or crosslink sites. For example, papain (a cysteine protease) does not de-
grade muscle proteins in which the cysteine has been oxidized into disulfide crosslinks
(Morzel et al., 2006). Analysis of crosslink or binding sites thus will require use of
multiple or non-specific proteases to cleave peptides at points other than the damage
sites.
Abnormal Function, Immunochemistry, and Contributions to Disease

The converse of losing biological function is a shift to non-productive or toxic ac-
tivity. Detailed information about the roles of oxidized proteins in a wide range of
pathologies has been accumulating for several decades, but the participation of oxi-
dizing lipids as one of the strongest biological oxidants is only slowly gaining atten-
tion. Although a thorough review of the mechanisms by which lipid oxidation of
proteins is involved in disease processes is beyond the scope of this chapter, at least
some mention is warranted to show why lipid co-oxidation pathways are becoming
so important in oxidative stress. For the most part, current information demonstrates
that lipid-protein co-oxidation products are present in target tissues or associated with
pathological changes, and there is growing evidence for signaling activity of both the
oxidized lipids and proteins (Page et al., 1999; Prescott, 1999; Uchida et al., 1999;
Leonarduzzi et al., 2000; Petersen and Doorn, 2004; Laurora et al., 2005; Zhang et
al., 2005; Valko et al., 2007). Despite these encouraging suggestions, considerable
research is still needed to elucidate specific causal roles of lipid-protein oxidation in
individual diseases.
Atherosclerosis: Oxidation of LDL and HDL.

Atherosclerosis is truly a disease of lipid and protein oxidation. Development of
atherosclerosis is most often associated with LDL, but HDL and two enzymes—
lecithin-cholesterol acyl transferase (LCAT) and cholesterol ester transfer protein
(CETP)—also play surprisingly critical roles. In each case, lipid oxidation of a protein
is intimately involved in the transformations leading to full-blown atherosclerosis.
Consider first the notorious LDL. LDL and monocytes enter endothelial cells
to collect and recycle cholesterol and phospholipids and to remove cellular debris,
respectively. Normally, they perform their jobs and leave the cell. However, oxidation
of the LDL lipid and protein components stimulates a series of dramatic changes
that accelerate the process of atherosclerosis (Uchida, 2000). Oxidized LDL signal
endothelial cells to express monocyte chemotactic protein 1 (MCP-1), which sits on
the cell surface and attracts monocytes from the vessel lumen into the subendothe-
lial space where they are converted into macrophages by accumulation of cholesterol
(Barter, 2001). At the same time, oxidized phospholipids form protein adducts that
anchor the complex to the epithelial surface, promoting oxidative stress and receptor-
independent endocytosis that leads to accumulation of oxidized intracellular lipids
(Weisgraber et al., 1978; Januszewski et al., 2005). Inside the endothelial cell, mono-
cyte/macrophages normally do not ingest cellular lipids or lipoproteins. However,
they very strongly and selectively bind and engulf oxidized lipoproteins and cholester-
ol by separate receptors 7- to 10-fold faster than native LDL (Fogelman et al., 1980;
Haberland, 1982), particularly when oxidized phospholipids are covalently attached
to apolipoprotein B (apoB) of LDL (Boullier et al., 2000 345; Podrez et al., 2002b;
Januszewski et al., 2005). Mild oxidation of LDL by lipid hydroperoxides results in
denaturation of apo B (Nishida and Kummerow, 1960) followed by decomposition
of LOOH to reactive aldehydes that covalently link to apoB, shut down the nor-
mal LDL receptor, and increase recognition and uptake of LDL by acetyl-LDL mac-
rophage receptors (Mahley et al., 1977; Steinbrecher, 1987; Haberland et al., 1988;
Hoff et al., 1992; Kim et al., 1999; Friedman et al., 2002). The switch of receptors
occurs when 15% of the lysine has been oxidized (“lost”), causing a change in specific
lysine residues (Haberland, 1982, 1984; Steinbrecher et al., 1987) that alters confor-
mation of the recognition site or alignment of reactive residues or H-bond through
e-amino to receptor (Weisgraber et al., 1978). Oxidation of arginine (Mahley et al.,
1977; Weisgraber et al., 1978) and tryptophan (Shoukry et al., 1994) residues also
appears to be involved. Recent evidence suggests that receptors recognize both the
modified lipid moieties and the modified protein moieties (Bird et al., 1999) and that
both oxidized lipids and individual lipid and protein oxidation products may activate
multiple receptors with different specificities so that uptake of oxidized LDL and
cholesterol becomes even more efficient as oxidation proceeds (Gillotte et al., 2000).
More extensive oxidation by lipid hydroperoxides or products such as MDA and
HNE produces protein aggregates that are viewed by macrophages as particles; they
too are absorbed, but by endocytosis (Steinbrecher, 1987; Viita et al., 1999). While
these multiple pathways were probably designed to be protective, clearing oxidized
246 K.M. Schaich
materials out of arteries and arterial walls, the paradoxical result with continued un-
controlled absorption of oxidized lipoproteins is to overload macrophages, causing a
conversion to foam cells that play a major role in plaque formation and acceleration
of pathogenic processes in atherosclerosis (Podrez et al., 2002a). Malondialdehyde-
altered protein has been found in plaque deposits in experimental atherosclerosis in
rabbits (Haberland et al., 1988).
Normally, oxidation in LDL can be counteracted by HDL which remove accu-
mulating lipid and cholesterol from macrophages and peripheral cells and transport
them to the liver for detoxification and disposal. Most LOOH in the bloodstream
are carried by HDL rather than LDL. Paradoxically, LDL is loaded with high levels
of antioxidants, particularly CoQH2, that protects it and prevents lipid oxidation,
while HDL have no antioxidants, except perhaps component apo and other proteins
(Bowry et al., 1992). Lipid and cholesterol hydroperoxides are transferred from LDL
to HDL, where they are reduced and then cleared rapidly by liver HepG2 cells that
selectively remove oxidized HDL from the bloodstream. As lipid oxidation levels in-
crease, however, lipid hydroperoxides undergo concerted two-electron reactions with
met112 and met148 in Apo A1 and met26 of A2, oxidizing them to methionine sulfoxide
without covalent lipid binding (early stages) (Garner et al., 1998b). These residues are
all in hydrophobic regions of active binding sites, not on the surface (Garner et al.,
1998a), so the reaction is highly oriented and specific. In addition, oxidized phospho-
lipid bound by HDL damage cys 31 and cys 184 in the HDL binding site. LOOH
were active at very low concentrations, breaking down and generating free radicals in
situ, while aldehydes were effective only at high concentrations (0.16 mM) (McCall et
al., 1995). Destruction of either met or cys residues reduces LOOH binding to HDL
as well as liver receptor recognition and clearance rate; lipid hydroperoxides then
accumulate in plasma and in endothelial cells, increasing the potential for extensive
protein co-oxidation and providing fuel for the macrophages.
Transport of oxidized fatty acids and cholesterol from LDL to HDL and from
the endothelial cells to the plasma and liver also is impaired as lipids oxidize the trans-
fer enzymes LCAT (lecithin:cholesterol acyl transferase) and CETP (cholesterol ester
transfer protein). Phospholipid hydroperoxides (PL-OOH) in LDL modify free cys-
teine (or other catalytic residues) on LCAT transferase (Mickel et al., 1972; Bielicki
and Forte, 1999), impairing HDL-cholesterol transport so cholesterol is not removed
from arterial walls and accumulation accelerates transformation of macrophages to
foam cells. As little as 0.2 and 1.0 mole% PL-OOH in plasma reduced LCAT activ-
ity by 20 and 50% in 2 hours, respectively (Bielicki and Forte, 1999). At low levels
of oxidation, phospholipid hydroperoxides attacked cysteine residues in enzyme ac-
tive sites, while at higher oxidation levels increasing levels of aldehydes inhibited the
enzyme by reactions outside the active site. Similarly, oxidation of cholesterol ester
transport protein prevents transport of oxidized lipids from LDL to HDL, so hydro-
peroxides accumulate in the cell and help accelerate macrophage transformation.
Paradoxically, there is also some evidence for cell signaling and induced protec-
tion via modulation of gene expression and biochemical pathways by lipid-mediated
protein oxidations (Uchida et al., 1999; Uchida 2000; Leonarduzzi et al., 2000; Lau-
rora et al., 2005; Zhang et al., 2005), particularly at the low physiological concentra-
tions of lipid aldehydes. HNE inhibits SK-N-BE cell proliferation by up-regulating
p53 family gene expression (Laurora et al., 2005) and prevents NF-kB activation and
tumor necrosis factor expression by inhibiting I-kB phosphorylation and proteolysis
(Page et al., 1999). The cytokine-induced expression of adhesion molecules in en-
dothelial cells has been shown in vitro and more recently in vivo to be inhibited by
HDL, in a process that potentially blocks a very early inflammatory stage in the devel-
opment of atherosclerosis. Increased titers of antibodies to oxidation-specific epitopes
of oxidized LDL in patients with advanced atherosclerosis is a protective response
aimed at eliminating modified LDL (Friedman et al., 2002).
Alzheimer’s Disease
The etiology and mechanisms of progression for Alzheimer’s disease are still poorly
understood, but it is clearly recognized as a process involving extensive oxidative deg-
radation of proteins, and oxidizing lipids are involved in some way. Early lipid radicals
and hydroperoxides induce conformation changes in Alzheimer-specific epitopes of
Tau (Liu et al., 2005), and oxidation of met to met sulfoxide in b-amyloid peptide
converts it to the toxic form, in which it is no longer able to penetrate membranes or
make pleated sheets but still binds Cu2+ and makes H2O2 (Barnham et al., 2003).
As lipid oxidation progresses in oxidative stress, 4-HNE is significantly increased
and is thought to play a role in the formation of b-amyloid (Sayre et al., 1997). Over
one-half of all-paired helical filament (PHF)-1-labeled neurofibrilary tangles and dys-
trophic neuritis surrounding the b-amyloid core were found to contain protein-bound
acrolein (Calingasan et al., 1991 272); the b-amyloid core itself also had adducts.
Other proteins in the brain cortex are also modified by lipid oxidation (Pamplona
et al., 2005), but this may be selective rather than generalized. About 100 oxidized
proteins, mostly involved in signaling processes, have been identified in aged mice.
Alterations in fatty acid synthetic enzymes cause shifts in brain fatty acids that affect
brain function and susceptibility to further oxidative stress (Soreghan et al., 2003).
Age-Related Macular Degeneration

DHA accounts for ~80 mol% of the lipids in the photoreceptor outer segments of the
retina. The high unsaturation of DHA provides critical fluidity in the retina, but its
oxidation also contributes to development of age-related macular degeneration (Gu
et al., 2003; Ebrahem et al., 2006), at least in part through indirect effects in which
protein co-oxidation products induce release of angiogenic factors. Carboxyethylpyri-
dinium (CEP) protein adducts from DHA stimulate overproduction of new blood
vessels that are tortuous and leaky, and they exacerbate choroidal neovascularization,
abnormal vessel growth from capillaries through Bruch’s membrane. 2-(w-carboxy-
heptyl)pyrroles from linolenic acid and 2-(w-carboxypropyl)pyrroles from arachidon-
ic acid have also been isolated from the retina.
w-6 polyunsaturated fatty acids also play an important role in damaging reti-
248 K.M. Schaich
nal proteins by providing a ready source of HNE. Three classes of proteins appear
to be particularly sensitive to HNE modification: chaperone/cell protection (heat
shock cognate; aA, aB, and bB2 crystallins); energy metabolism (triose phosphate
isomerase, a enolase, aldolase C); and fatty acid transport (a enolase and bB2 crystal-
lin) (Kapphahn et al., 2006). Most of these proteins have reactive cysteine residues
that provide reducing equivalents to the cells as well as nucleophilic targets for HNE
binding. Since the affected enzymes are part of the glycolytic pathway cascade, their
inhibition has a critical affect on retinal function, where >50% of the ATP is pro-
duced via glycolysis. However, whether this is a negative or protective effect is being
debated. Protein binding of toxic HNE and other aldehydes may be viewed as a
“molecular sponge” for removing oxidant species, and protein modifications leading
to reprogramming of metabolism to the pentose phosphate pathway for production
of NADPH aids in recovery from oxidative damage (Kapphahn et al., 2006).
Other Diseases
Involvement of lipid oxidation in a wide range of oxidative pathologies is suspected,
but isolating causative lipid oxidation products and determining their specific roles
is extremely difficult in actively metabolizing tissues where abnormal compounds are
rapidly cleared. With advances in instrumental analysis and immunological tech-
niques, stable lipid-protein adducts are now being tracked to establish lipid oxidation
associations with diseases (although presence of uncleared adducts is not proof that
they actually caused the damage). Certainly, it is easy to envision how the protein
changes described previously can play key roles in many forms of tissue degradation.
Lysine-pyridinium adducts have been identified in proteins from patients with amyo-
trophic lateral sclerosis (Ichihashi et al., 2001), a disease that involves progressive
deterioration of the myelin sheath and loss of nerve function. Alexander’s disease is a
progressive neurological disorder in which formation of fibrous, eosinophilic deposits
called Rosenthal fibers leads to destruction of white matter in the brain (Castellani
et al., 1998).
Systemic lupus erythematosus (SLE or lupus) is an auto-immune disease in which
normal cells are mistakenly identified as foreign and labeled with antibodies that set
up a cascade of radical-generating reactions that lead to inflammation. Modification
of lupus-associated 60-kDa Ro protein with 4-hydroxy-2-nonenal increases recogni-
tion of cells as abnormal and facilitates spreading of the marker epitope and associated
inflammation to other sites (Scofield et al., 2005). Crotonaldehyde is a strong tissue
irritant in humans and carcinogen in male rats. Schiff-base pyridinium adducts to
proteins have been identified in tissues exposed to this agent (Ichihashi et al., 2001).
Conclusions
The past 20 years have seen great advances in the understanding of reactions between
lipid oxidation products and proteins. While it is now clear that lipid oxyl radicals,
hydroperoxides, epoxides, and carbonyl products all damage proteins, there is still
no clear elucidation of the relative importance of various lipid oxidation products

in damage to proteins. Indeed, perhaps even more questions have been raised about
which lipid species is the dominant oxidant and which causes the most significant
modifications contributing to pathologies in vivo or degrading protein function in
foods.
This may seem to be a strange statement considering the detailed chemistry of
individual reactions that has been reviewed in this chapter. What must be remem-
bered, however, is that lipid oxidation is a dynamic process, constantly changing and
contributing new oxidants; the mechanisms of interaction, amino acid target prefer-
ences, and reaction rates with proteins change with the extent of oxidation, and so
do the reactivity and stability of products. The critical question is not which oxidant
causes “the” damage to proteins, but how the various oxidants each contribute over
time in actively oxidizing foods or biological systems, and perhaps more importantly,
how damage from each of the oxidants interacts to change the overall functional,
physiological, and pathological outcomes.
Damage to proteins (and other biomolecules) occurs at all stages of lipid oxida-
tion. Protein degradation begins with the first lipid radicals long before aldehyde con-
centrations rise, and it continues after development of carbonyl secondary products.
In general, lipid radicals and peroxides tend to induce protein oxidation and cross-
linking without incorporation of lipid, and intermediate products on proteins are not
stable. There are no definitive, easily detectable molecular “flags” in the products, so
consequences of radical reactions are often difficult to recognize. All changes (e.g.,
loss of amino acids; production of oxidation products including new amino acids,
crosslinking, and scission; lipid binding; etcetera) must be followed to assess reaction
effects. In contrast, carbonyls in lipid secondary oxidation products react exclusively
by addition reactions. Their relatively stable protein adducts are easily distinguishable
by physical properties, such as fluorescence, and by reactions with chemical, optical,
and immunological labels. Even so, definitive analysis is complicated by the revers-
ibility of reactions in water or acid and by transformation of initial simple adducts to
cyclic products over time, especially as aldehyde concentrations increase.
Protein reaction with oxidizing lipids has traditionally been monitored by global
effects on chemical properties (particularly production of protein carbonyls) and mac-
roscopic properties, such as enzyme inhibition, crosslinking or scission, or develop-
ment of fluorescence in proteins. Traditionally, such measurements in model systems
have been used to infer the active lipid oxidant and protein damage mechanisms.
However, as this chapter has shown, lipid radicals, hydroperoxides, and carbonyls all
produce similar global changes, particularly the protein carbonyls, fluorescence, and
crosslinking that are analyzed universally (Fig. 8.19). A major message of this chapter,
therefore, is that in systems with actively oxidizing lipids (not isolated products), oxi-
dant sources of protein damage and reaction mechanisms cannot be concluded from
macroscopic, global behaviors alone.
Such a flat statement is not meant to trivialize the issue. Indeed, the problem of
identifying the lipid oxidants damaging proteins in real systems is as complex as the
250 K.M. Schaich
lipid oxidation reaction itself. Reactions occurring in isolated model systems with
high concentrations of single oxidants are probably not the reactions occurring in situ
in complex media with multiple oxidants forming at different times, competing for
reactive sites, and altering availability (both chemical and physical) of reactive sites.
Biological tissues pose an additional challenge because enzymes mediate secondary
oxidations, reductions, and conjugations, and damaged molecules are rapidly cleared
or repaired.
Monitoring protein products that have accumulated at a given time detects pri-
marily products that are stable and have not been cleared by reaction in foods or by
enzymes. But are these the products that have caused the most changes in system
properties? Does identification of trace amounts (e.g., nanomolar) of stable aldehyde-
protein adducts, for example, prove that these are the active damage agents or just the
ones that are most long-lived or not removed by reaction or cell recycling and protec-
tion mechanisms? Of what impact are the protein products from radicals or hydroper-
oxides that have already degraded or been transformed further and cannot be detected
or easily identified? Cleavage of histidine, arginine, lysine, and proline side chains to
other amino acids, for example, would never be detected unless the individual protein
was isolated and hydrolyzed, but it could have huge effects on protein functionality in
foods and biological activity and recognition in vivo.
L IP ID R A D IC A L V E R S U S L IP ID C AR B O N Y L R E A C TIO N P R O D U C TS W ITH P R O TE IN S
L O (O )• /L O O H p ro tein fre e rad icals,

n o t eas y to trace so u rc e,
attack s ite m ig rates
so m e ad d u cts fo rm b y rad ical ad d itio n
P R O T E IN C A R B O N Y L S , S C IS S IO N A N D C R O S S L IN K IN G , F L U O R E S C E N T P R O D U C T S
A ld eh yd e s , s eco n d ary p ro d u cts stab le am in o a cid ad d u cts

eas y to iso late an d id en tify ad d u cts –
M ich ael ad d itio n ,
S ch iff b a se,
C =O /N H 2 co n d en satio n
Fig. 8.19. Despite differences in initial reactions, lipid radical, hydroperoxide, and carbonyl reactions with proteins
yield products that may be indistinguishable by global analyses of macroscopic behaviors. Hence, measurement
of protein carbonyls, crosslinking and scission, or fluorescence can be used to assess extent of protein degradation
but not to deduce lipid oxidant sources.
A second message, therefore, is it is useless to argue that any one lipid oxidation
product is the factor mediating oxidative changes. Research on oxidative damage to
proteins induced by oxidizing lipids needs to move away from battles between the
oxidants and claims of exclusive supremacy to focus instead on how timing of analy-
sis, rates of reactions, effects of concentrations on direction of reactions, and fates
of intermediates may affect detectability of changes and interpretation of reaction
mechanisms, and how reactions of the various lipid oxidants are balanced and interact
in different environments or with different proteins.
Three critical areas of information are still missing: detailed structural analysis of
intermediates and products; quantitative analysis of individual lipid oxidation species,
their protein interaction products, and comparisons between classes; and effects of
reaction conditions on dominant pathways and individual products. The tremendous
advantages in product separation and analysis offered by LC-MS are reflected in the
increasing numbers of studies identifying intermediates, products, and mechanisms
with precision. What must be added to this data base is information about individual
reaction conditions, rates, and yields to put current studies in perspective. A critical
mass of fundamental data is available, so now details are more important.
Model systems may reveal individual reaction products that are possible, but
what are the actual yields of individual products under various conditions? Few papers
actually report yields. How will a particular reaction product compete in a complex
system if the rates of production are slow and the yields are low? Are there products
whose low yields are counterbalanced by extraordinarily high reactivity so that traces
overwhelm other products in damaging proteins? Reactions and products of alde-
hydes with unmodified proteins are reasonably well documented in defined model
systems; how are these changed if the protein has previously been modified by lipid
radicals? How does the dominant mechanism change when the reaction moves from
neutral aqueous phases (intracellular, extracellular, or emulsion) to acid compart-
ments to lipid phases of membranes, local environments, or emulsions?
In foods and model systems it is clear that low levels of lipid radicals and hydro-
peroxides rapidly damage proteins, and that aldehyde reactions are slow and require
levels 100-1000x higher for reaction. Even so, aldehydes, by virtue of their longer
lifetimes, presumed greater diffusibility, stability of protein adducts, and thus ease
of detection are considered the most toxic lipids in vivo. Considering the chemistry
covered in this chapter, the issue of aldehyde reactions is very intriguing and puzzling,
and it raises many questions. How can model system studies demonstrating that al-
dehyde reactions with proteins are very slow on a physiological time scale and orders
of magnitude slower than LOOH and radicals, and require 1:1 molar ratios and
concentrations orders of magnitude higher than what are found in vivo be reconciled
with claims of cytotoxicity from nanomolar (or lower) levels of aldehydes and their
protein products in vivo? How can aldehyde concentrations in vivo accumulate to lev-
els high enough to match the concentrations required for reaction in model systems?
Are there catalysts that facilitate or speed up aldehyde reactions in vivo? Are aldehydes
cleared less slowly so they accumulate over time, or are they detected just because they
252 K.M. Schaich
are stable, while earlier lipid oxidation species have already mediated damage? Do
aldehyde products derive from direct dominant reaction with native proteins, or are
there reactions of secondary products that are facilitated by preliminary damage from
radicals but are slow or absent without conformation or other changes induced by hy-
droperoxides? Kanazawa and Ashida (1998) claim that lipid hydroperoxides in foods
are decomposed in the stomach to aldehydes that are then absorbed. If the absorbed
aldehydes are not metabolized or detoxified, deposition in tissues and incorporation
into membranes could lead to build-up of toxic concentrations. Is this an accurate
explanation?
These questions have serious implications for how experiments are designed,
data is interpreted, and protective measures are structured, both for food preservation
methods and for lipid-protein interactions in vivo. The issue of data interpretation
thus deserves much debate and discussion.
Analyzing products in situ provides footprints of reactions and clues about causal
agents, but more definitive studies in model systems closely coordinated with in situ
oxidations in foods, cells, or tissues are needed to determine causality conclusively.
This chapter has reviewed the extensive efforts focused on determining products of
individual lipid oxidation products with intact proteins and component amino acids,
and much has been learned about breakdown pathways. The time has now come to
apply this knowledge to detailed analysis of integrated oxidation sequences in com-
plex model systems, to replace global characteristic analysis with detailed determina-
tion of protein properties and amino acid changes step by step as lipids oxidize.
Lipid oxidation damage to proteins also must be juxtaposed with protein dam-
age from other oxidant sources. Data cited in this chapter has shown repeatedly that
protein changes induced by oxidizing lipids are the same as or comparable to re-
actions of hydroxyl radicals, which are blamed for most of the oxidant damage in
vivo. Lipid-protein products, lipid reaction kinetics, and lipid peroxide and aldehyde
concentrations that can accumulate provide overwhelming evidence that oxidizing
lipids are competitive with other biological oxidants and should be included as bio-
logical oxidants along with the other agents normally cited as reactive oxidant species
(Stadtman, 2004). For proteins in the endoplasmic reticulum or other membranes or
closely associated with other lipid structures (e.g., blood lipoproteins), lipids are prob-
ably the dominant or most important oxidant. Biomedical research is just now begin-
ning to recognize this, and hopefully the future will see definitive research focused on
distinguishing specific roles of oxygen radicals versus lipid oxidation species in both
physiological and pathological processes.
References
Adams, J.Q. Electron Paramagnetic Resonance of tert-Butoxy Radical Reactions with Sulfides
and Disulfides. J. Am. Chem. Soc. 1970, 92, 4535-4537.
Akagawa, M.; D. Sasaki; Y. Ishii; Y. Kurota; M. Yotsu-Yamashita; K. Uchida; and K. Suyama.
New Method for the Quantitative Determination of Major Carbonyls, α-Aminoadipic
and γ-Glutamic Semialdehydes: Investigation of the Formation Mechanism and Chemi-
cal Nature in vitro and in vivo. Chem. Res. Toxicol. 2006, 19, 1059-1065.
Alaiz, M. and J. Girón. Modification of Histidine Residues in Bovine Serum Albumin by Re-
action with (E)-2-Octenal. J. Agric. Food Chem. 1994, 42, 2094-2098.
Alderton, A.L.; C. Faustman; D.C. Liebler; and D.W. Hill. Induction of Redox Instability of
Bovine Myoglobin by Adduction with 4-Hydroxy-2-Nonenal. Biochemistry 2003, 42,
4398-4405.
Amarnath, V.; K. Amarnath; W.M. Valentine; M.A. Eng; and D.G. Graham. Intermediates
in the Paal-Knorr Synthesis of Pyrroles. 2-Oxoaldehydes. Chem. Res. Toxicol. 1995, 8,
234-238.
Amarnath, V.; W.M. Valentine; K. Amarnath; M.A. Eng; and D.G. Graham. The Mechanism
of Nucleophilic Substitution of Alkylpyrroles in the Presence of Oxygen. Chem. Res. Toxi-
col. 1994, 7, 56-61.
Amarnath, V.; W.M. Valentine; T.J. Montine; W.H. Patterson; K. Amarnath; C.N. Bassett;
and D.G. Graham. Reactions of 4-Hydroxy-2(E)-Nonenal and Related Aldehydes with
Proteins Studied by Carbon-13 Nuclear Magnetic Resonance Spectroscopy. Chem. Res.
Toxicol. 1998, 11, 317-328.
ATSDR. Agency for Toxic Substances and Disease Registry ToxFAQs for crotonaldehyde,
2002, http://www.atsdr.cdc.gov/tfacts180.html#bookmark04.
Aubourg, S.P.; C.G. Sotelo; and R. Perez-Martin. Assessment of Quality Changes in Frozen
Sardine (Sardina pilchardus) by Fluorescence Detection. J. Am. Oil Chem. Soc. 1998, 75,
575-580.
Augisto, A.; M.G. Bonini; and D.F. Trindade. Spin Trapping of Glutathiyl and Protein Radi-
cals Produced from Nitric Oxide-Derived Oxidants. Free Radic. Biol. Med. 2004, 36,
1224-1232.
Avdulov, N.A.; S.V. Chochina; U. Igbavboa; E.O. O’Hare; F. Schroeder; J.P. Cleary; and W.G.
Wood. Amyloid Beta-Peptides Increase Annular and Bulk Fluidity and Induce Lipid Per-
oxidation in Brain Synaptic Plasma Membranes. J. Neurochem. 1997, 68, 2086–2091.
Badghisi, H.; and D.C. Liebler. Sequence Mapping of Epoxide Adducts in Human Hemo-
globin with LC-Tandem MS and the Salsa Algorithm. Chem. Res. Toxicol. 2002, 15,
799-805.
Barnham, K.J.; G.D. Ciccotosto; A.K. Tickler; F.E. Ali; D.G. Smith; N.A. Williamson; Y.-H.
Lam; D. Carrington; D. Tew; G. Kocak, et al. Neurotoxic, Redox-Competent Alzheimer’s
β-Amyloid Is Released from Lipid Membrane by Methionine Oxidation. J. Biol. Chem.
2003, 278, 42959-42965.
Barter, P. Role of lipoproteins in inflammation, 2001, http://www.lipidsonline.org/slides.
Benderdour, M.; G. Charron; D. deBlois; B. Comte; and C. Des Rosiers. Cardiac Mitochon-
drial NADP+-Isocitrate Dehydrogenase Is Inactivated Through 4-Hydroxynonenal Ad-
duct Formation. J. Biol. Chem. 2003, 278, 45154-45159.
Benedetti, A.; A.F. Casini; and M. Ferrali. Extraction and Partial Characterization of Dialyz-
able Products Originating from the Peroxidation of Liver Microsomal Lipids and Inhibit-
ing Microsomal Glucose-6-Phosphatase Activity. Biochem. Pharmacol. 1979, 28, 2909-
2918.
Beppu, B. Role of Heme Compounds in the Erythrocyte Membrane Damage Induced by
Lipid Hydroperoxide. Chem. Pharm. Bull. 1986, 34, 781-788.
Berger, P.; N.K. Vel Leitner; M. Doré; and B. Legube. Ozone and Hydroxyl Radicals Induced
Oxidation of Glycine. Water Res. 1999, 33, 433-441.
Bielicki, J.K.; and T.M. Forte. Evidence that Lipid Hydroperoxides Inhibit Plasma Lecithin:
254 K.M. Schaich
Cholesterol Acytransferase Activity. J. Lipid Res. 1999, 40, 948-954.

Bird, D.A.; K.L. Gillotte; S. Hörkkö; P. Friedman; E.A. Dennis; J.L. Witztum; and D. Stein-
berg. Receptors for Oxidized Low-Density Lipoprotein on Elicited Mouse Peritoneal
Macrophages Can Recognize Both the Modified Lipid Moieties and the Modified Protein
Moieties: Implications with Respect to Macrophage Recognition of Apoptotic Cells. Proc.
Nat. Acad. Sci. USA 1999, 96, 6347-6352.
Blair, I.A. Lipid Hydroperoxide-Mediated DNA Damage. Exp. Gerontol. 2001, 36, 1473-
1481.
Boatright, W.L.; and N.S. Hettiarachchy. Effect of Lipids on Soy Protein Isolate Solubility. J.
Am. Oil Chem. Soc. 1995, 72, 1439-1444.
Bobrowski, K.; and C. Schöneich. Decarboxylation Mechanism of the N-Terminal Glutamyl
Moiety in γ-Glutamic Acid and Methionine Containing Peptides. Radiat. Phys. Chem.
1996, 47, 507-510.
Bolgar, M.S.; and S.J. Gaskell. Determination of the Sites of 4-Hydroxy-2-Nonenal Adduc-
tion to Protein by Electrospray Tandem Mass Spectrometry. Anal. Chem. 1996, 68, 2325-
2330.
Borg, D.C.; and K.M. Schaich. Cytotoxicity from Coupled Redox Cycling of Autoxidizing
Xenobiotics and Metals. Isr. J. Chem. 1984, 24, 38-53.
Borovyagin, V.L.; A.F. Muronov; V.D. Rumyantseva; Y.S. Tarakhovskii; and I.A. Vasilenko.
Model Membrane Morphology and Crosslinking of Oxidized Lipids with Proteins. J.
Ultrastruct. Res. 1984, 89, 261-273.
Bossman, S.H.; E. Oliveros; S. Göb; S. Siegwart; E.P. Dahlen; L.J. Payawan; M. Straub; M.
Wörner; and A.M. Braun. New Evidence Against Hydroxyl Radicals as Reactive Interme-
diates in the Thermal and Photochemically Enhanced Fenton Reactions. J. Phys. Chem.
1998, 102, 5542-5550.
Boullier, A.; K.L. Gillotte; S. Hörkkö; S.R. Green; P. Friedman; E.A. Dennis; J.L. Witztum;
D. Steinberg; and O. Quehenberger. The Binding of Oxidized Low Density Lipoprotein
to Mouse CD36 Is Mediated in Part by Oxidized Phospholipids that Are Associated
with Both the Lipid and Protein Moieties of the Lipoprotein. J. Biol. Chem. 2000, 275,
9163-9169.
Bowry, V.W.; K.K. Stanley; and R. Stocker. High Density Lipoprotein Is the Major Carrier of
Lipid Hydroperoxides in Human Blood Plasma from Fasting Donors. Proc. Nat. Acad.
Sci. USA 1992, 89, 10316-10320.
Brame, C.J.; O. Boutaud; S.S. Davies; T. Yang; J.A. Oates; D. Roden; and L.J.I. Roberts.
Modification of Proteins by Isoketal-Containing Oxidized Phospholipids. J. Biol. Chem.
2004, 279, 13447-13451.
Brame, C.J.; R.G. Salomon; J.D. Morrow; and L.J.I. Roberts. Identification of Extremely Reac-
tive γ−Ketoaldehydes (Isolevuglandins) as Products of the Isoprostane Pathway and Char-
acterization of Their Lysyl Protein Adducts. J. Biol. Chem. 1999, 274, 13139-13146.
Brock, J.W.C.; J.M. Ames; S.R. Thorpe; and J.W. Baynes. Formation of Methionine Sulfox-
ide During Glycoxidation and Lipoxidation of Ribonuclease A. Arch. Biochem. Biophys.
2007, 457, 170-176.
Bruenner, B.A.; A.D. Jones; and J.B. German. Direct Characterization of Protein Adducts of
the Lipid Peroxidation Product 4-Hydroxy-2-Nonenal Using Electrospray Mass Spec-
trometry. Chem. Res. Toxicol. 1995, 8, 552-559.
Brunner, B.; N. Stogatis; and M. Lautens. Synthesis of 1,2-Dihyropyridine Using Vinyloxi-
ranes as Masked Dienolates in Imino-Aldol Reactions. Org. Lett. 2006, 8, 3473-3476.
Burcham, P.C.; and Y.T. Kuhan. Introduction of Carbonyl Groups into Proteins by the Lipid
Peroxidation Product, Malondialdehyde. Biochem. Biophys. Res. Commun. 1996, 220,
996-1001.
Buttery, R.G.; L.C. Ling; R. Teranishi; and T.R. Mon. Roasted Fat: Basic Volatile Compo-
nents. J. Agric. Food Chem. 1977, 25, 1227-1229.
Buttkus, H. The Reaction of Myosin with Malonaldehyde. J. Food Sci. 1967, 32, 432-434.
Buttkus, H. Reaction of Cysteine and Methionine with Malonaldehyde. J. Am. Oil Chem. Soc.
1968, 46, 88-93.
Buttkus, H. Accelerated Denaturation of Myosin in Frozen Solution. J. Food Sci. 1970, 35,
558-562.
Buttkus, H. The Reaction of Malonaldehyde or Oxidized Linolenic Acid with Sulfhydryl
Compounds. J. Am. Oil Chem. Soc. 1972, 49, 613-614.
Buttkus, H.A. Fluorescent Lipid Autoxidation Products. J. Agric. Food Chem. 1975, 23, 823-
825.
Cagen, L.M.; H.M. Fales; and J.J. Pisano. Formation of Glutathione Conjugates of Prosta-
glandin A in Human Red Blood Cells. J. Biol. Chem. 1976, 251, 6550-6554.
Calingasan, N.Y.; K. Uchida; and G.E. Gibson. Protein-Bound Acrolein: A Novel Marker of
Oxidative Stress in Alzheimer’s Disease. J. Neurochem. 1991, 72, 751-756.
Carbone, D.L.; J.A. Doorn; Z. Kiebler; and D.R. Petersen. Cysteine Modification by Lipid
Peroxidation Products Inhibits Protein Disulfide Isomerase. Chem. Res. Toxicol. 2005, 18,
1324-1331.
Castellani, R.J.; G. Perry; P.L.R. Harris; M.L. Cohen; L.M. Sayre; R.G. Salomon; and M.A.
Smith. Advanced Lipid Peroxidation End-Products in Alexander’s Disease. Brain Res.
1998, 787, 15-18.
Chander, R.; S.V. Sherekar; and M.S. Gore. Studies on the Inactivation of Lysozyme by Malo-
naldehyde. J. Food Biochem. 1981, 5, 313-324.
Chelh, I.; P. Gatellier; and V. Sante-Lhoutellier. Characterisation of Fluorescent Schiff Bases
Formed During Oxidation of Pig Myofibrils. Meat Sci. 2007, 76, 210-215.
Chen, H.-J.C.; and F.-L. Chung. Epoxidation of trans-4-Hydroxy-2-Nonenal by Fatty Acid
Hydroperoxides and Hydrogen Peroxide. Chem. Res. Toxicol. 1996, 9, 306-312.
Chen, P.; D. Wiesler; J. Chmelik; and M. Novotny. Substituted 2-Hydroxy-1,2-Dihydropyr-
rol-3-Ones: Fluorescent Markers Pertaining to Oxidative Stress and Aging. Chem. Res.
Toxicol. 1996, 9, 970-979.
Chio, K.S.; and A.L. Tappel. Inactivation of Ribonuclease and Other Enzymes by Peroxidizing
Lipids and by Malonaldehyde. Biochemistry 1969a, 8, 2827-2832.
Chio, K.S.; and A.L. Tappel. Synthesis and Characterization of the Fluorescent Products De-
rived from Malondialdehyde and Amino Acids. Biochemistry 1969b, 8, 2821-2827.
Chung, F.-L.; H.-J.C. Chen; J.B. Guttenplan; A. Nishikawa; and G.C. Hard. 2,3-Epoxy-4-
Hydroxynonanal as a Potential Tumor-Initiating Agent of Lipid Peroxidation. Carcino-
genesis 1993, 14, 2073-2077.
Cohn, J.A.; L. Tsai; B. Friguet; and L.I. Szweda. Chemical Characterization of a Protein-4-
Hydroxy2-Nonenal Crosslink: Immunochemical Detection in Mitochondria Exposed to
Oxidative Stress. Arch. Biochem. Biophys. 1996, 328, 158-164.
Connell, J.J. Studies on the Proteins of Fish Skeletal Muscle. 7. Denaturation and Aggregation
of Cod Myosin. Biochem. J. 1960, 75, 530-538.
Connell, J.J.; and P.F. Howgate. Studies on the Proteins of Fish Skeletal Muscle. 6. Amino-
Acid Composition of Cod Fibrillar Proteins. Biochem. J. 1959, 71, 83-86.
256 K.M. Schaich
Crabb, J.W.; J. O’Neill; M. Miyagi; K. West; and H.F. Hoff. Hydroxynonenal Inactivates
Cathepsin B by Forming Michael Adducts with Active Site Residues. Protein Sci. 2002,
11, 831-840.
Crawford, D.L.; T.C. Yu; and R.O. Sinnhuber. Reaction of Malonaldehyde with Glycine. J.
Agric. Food Chem. 1966, 14, 182-184.
Culbertson, S.M.; G.D. Enright; and K.U. Ingold. Synthesis of a Novel Radical Trapping
and Carbonyl Group Trapping Anti-Age Agent: A Pyridoxamine Analogue for Inhibiting
Advanced Glycation (Age) and Lipoxidation (ALE) End Products. Org. Lett. 2003, 5,
2659-2662.
Dalsgaard, T.K.; J.H. Nielsen; and L.B. Larsen. Characterization of Reaction Products Formed
in a Model Reaction Between Pentanal and Lysine-Containing Oligopeptides. J. Agric.
Food Chem. 2006, 54, 6367-6373.
Davies, K.A.J. Protein Damage and Degradation by Oxygen Radicals. I. General Aspects. J.
Biol. Chem. 1987a, 262, 9895-9901.
Davies, K.A.J. Protein Damage and Degradation by Oxygen Radicals. IV. Degradation of
Denatured Proteins. J. Biol. Chem. 1987b, 262, 9914-9920.
Davies, K.A.J.; and M.E. Delsignore. Protein Damage and Degradation by Oxygen Radicals.
III. Modification of Secondary and Tertiary Structure. J. Biol. Chem. 1987, 262, 9908-
9913.
Davies, M.J. The Oxidative Environment and Protein Damage. Biochim. Biophys. Acta 2005,
1703, 93-109.
Davies, M.J.; S. Fu; and R.T. Dean. Protein Hydroperoxides Can Give Rise to Reactive Free
Radicals. Biochem. J. 1995, 305, 643-649.
Davies, M.J.; S. Fu; H. Wang; and R.T. Dean. Stable Markers of Oxidant Damage to Proteins
and Their Application in the Study of Human Disease. Free Radic. Biol. Med. 1999 27,
1151-1163.
Davies, M.J.; and C.L. Hawkins. EPR Spin Trapping of Protein Radicals. Free Radic. Biol.
Med. 2004, 36, 1072-1086.
Davies, S.S.; V. Amarnath; and L.J.I. Roberts. Isoketals: Highly Reactive γ-Ketoaldehydes
Formed from the H2-Isoprostane Pathway. Chem. Phys. Lipids 2004, 128, 85-99.
Desai, I.D.; and A.L. Tappel. Damage to Proteins by Peroxidized Lipids. J. Lipid Res. 1963,
4, 204-207.
Dillard, C.J.; and A.L. Tappel. Fluorescent Products of Lipid Peroxidation of Mitochondria
and Microsomes. Lipids 1971, 6, 715-721.
Dillard, C.J.; and A.L. Tappel. Fluorescent Products from Reaction of Peroxidizing Polyun-
saturated Fatty Acids with Phosphatidyl Ethanolamine and Phenylalanine. Lipids 1973,
8, 183-189.
Dillard, C.J.; and A.L. Tappel. Fluorescent Damage Products of Lipid Peroxidation. Methods
Enzymol. 1984, 105, 337-341.
Doorn, J.A.; and D.R. Petersen. Covalent Modification of Amino Acid Nucleophiles by the
Lipid Peroxidation Products 4-Hydroxy-2-Nonenal and 4-Oxo-2-Nonenal. Chem. Res.
Toxicol. 2002, 15, 1445-1450.
Doorn, J.A.; and D.R. Petersen. Covalent Adduction of Nucleophilic Amino Acids by 4-Hy-
droxynonenal and 4-Oxononenal. Chem. Biol. Interact. 2003, 143-144, 93-100.
Dvorak, Z. Availability of Essential Amino Acids from Proteins. II. Food Proteins. J. Sci. Food
Agric. 1968, 19, 77-82.
Ebrahem, Q.; K. Ranganathan; J. Sears; A. Vasanji; X. Gu; L. Lu; R.G. Salomon; J.W. Crabb;
and B. Anand-Apte. Carboxyethylpyrrole Oxidative Protein Modifications Stimulate

Neovascularization: Implications for Age-Related Macular Degeneration. Proc. Nat. Acad.
Sci. USA 2006, 103, 13480-13484.
Ege, S.N. Organic Chemistry: Structure and Reactivity, Houghton: Boston, pp. 517-518,
631-654, 692-693, 725-733, 1999.
El-Gharbawi, M.I.; and L.R. Dugan, Jr. Stability of Nitrogenous Compounds and Lipids dur-
ing Storage of Freeze-Dried Raw Beef. J. Food Sci. 1965, 30, 817-822.
Elliot, A.J.; R.J. McEachern; and D.A. Armstrong. Oxidation of Amino-Containing Disul-
fides by Br2−· and OH. A Pulse-Radiolysis Study. J. Phys. Chem. 1981, 85, 68-75.
Esterbauer, H.; A. Ertl; and N. Scholz. The Reaction of Cysteine with α,β−Unsaturated Alde-
hydes. Tetrahedron 1976, 32, 285-289.
Esterbauer, H.; R.J. Schaur; and H. Zollner. Chemistry and Biochemistry of 4-Hydroxynon-
enal, Malonaldehyde, and Related Aldehydes. Free Radic. Biol. Med. 1991, 11, 81-128.
Esterbauer, H.; H. Zollner; and N. Scholz. Reaction of Glutathione with Conjugated Carbon-
yls. Z. Naturforsch., C: Biosci. 1975, 30, 466-473.
Faustman, C.; D.C. Liebler; T.D. McClure; and Q. Sun. a,b-Unsaturated Aldehydes Acceler-
ate Oxymyoglobin Oxidation. J. Agric. Food Chem. 1999, 47, 3140-3144.
Fenaille, F.; P.A. Guy; and J.-C. Tabet. Study of Protein Modification by 4-Hydroxy-2-None-
nal and Other Short Chain Aldehydes Analyzed by Electrospray Ionization Tandem Mass
Spectrometry. J. Am. Soc. Mass Spectrom. 2002, 14, 215-226.
Fenaille, F.; J.-C. Tabet; and P.A. Guy. Study of Peptides Containing Modified Lysine Residues
by Tandem Mass Spectrometry: Precursor Ion Scanning of Hexanal-Modified Peptides.
Rapid Communications in Mass Spectrometry 2004a, 18, 67-76
Fenaille, F.; J.-C. Tabet; and P.A. Guy. Identification of 4-Hydroxy-2-Nonenal-Modified Pep-
tides Within Unfractionated Digests Using Matrix-Assisted Laser Desorption/Ionization
Time-Of-Flight Mass Spectrometry. Anal. Chem. 2004b, 76 (4), 867-873.
Finley, J.W.; and R.E. Lundin. Lipid Hydroperoxide Induced Oxidation of Cysteine in Pep-
tides. In Autoxidation in Food and Biological Systems, M.G. Simic, and M. Karel, Eds.
Plenum: New York, 1980, pp. 223-236.
Fletcher, B.L.; C.J. Dillard; and A.L. Tappel. Measurement of Fluorescent Lipid Peroxidation
Products in Biological Systems and Tissues. Anal. Biochem. 1973, 52, 1-9.
Fletcher, B.L.; and A.L. Tappel. Fluorescent Modification to Serum Albumin by Lipid Peroxi-
dation. Lipids 1970, 6, 172-175.
Fogelman, A.M.; I. Schecter; J. Seager; M. Hokom; J.S. Child; and P.A. Edwards. Malondial-
dehyde Alteration of Low Density Lipoproteins Leads to Cholesteryl Ester Accumulation
in Human Monocyte-Macrophages. Proc. Nat. Acad. Sci. USA 1980, 77, 2214-2218.
Fong, L.G.; S. Parthasarathy; J.L. Witztum; and D. Steinberg. Nonenzymatic Oxidative Cleav-
age of Peptide Bonds in Apoprotein B-100. J. Lipid Res. 1987, 28, 1466-1477.
Freeman, T.L.; A. Haver; M.J. Duryee; D.J. Tuma; L.W. Klassen; F.G. Hamel; R.L. White; S.I.
Rennard; and G.M. Thiele. Aldehydes in Cigarette Smoke React with the Lipid Peroxida-
tion Product Malonaldehyde to Form Fluorescent Protein Adducts on Lysines. Chem. Res.
Toxicol. 2005, 18, 817-824.
Friedman, P.; S. Hörkkö; D. Steinberg; J.L. Witztum; and E.A. Dennis. Correlation of An-
tiphospholipid Antibody Recognition with the Structure of Synthetic Oxidized Phos-
pholipids. Importance of Schiff Base Formation and Aldol Condensation. J. Biol. Chem.
2002, 277, 7010-7020.
Friguet, B.; E.R. Stadtman; and L.I. Szweda. Modification of Glucose-6-Phosphate Dehydro-
258 K.M. Schaich
genase by 4-Hydroxy-2-Nonenal. Formation of Cross-Linked Protein that Inhibits the

Multicatalytic Protease. J. Biol. Chem. 1994a, 269, 21639-21643.
Friguet, B.; and L.I. Szweda. Inhibition of the Multicatalytic Proteinase (Proteasome) by 4-
Hydroxy-2-Nonenal Cross-Linked Protein. FEBS Lett. 1997, 405, 21-25.
Friguet, B.; L.I. Szweda; and E.R. Stadtman. Susceptibility of Glucose-6-Phosphate Dehydro-
genase Modified by 4-Hydroxy-2-Nonenal and Metal-Catalyzed Oxidation to Proteolysis
by the Multicatalytic Protease. Arch. Biochem. Biophys. 1994b, 311, 168-173.
Fruebis, J.; S. Parsasarathy; and D. Steinberg. Evidence for a Concerted Reaction Between
Lipid Hydroperoxides and Polypeptides. Proc. Nat. Acad. Sci. USA 1992, 89, 10588-
10592.
Fu, S.; L.A. Hick; M.M. Sheil; and R.T. Dean. Structural Identification of Valine Hydroperox-
ides and Hydroxides on Radical-Damaged Amino Acid, Peptide, and Protein Molecules.
Free Radic. Biol. Med. 1995, 19, 281-292.
Funes, J.; and M. Karel. Free Radical Polymerization and Lipid Binding of Lysozyme Reacted
with Peroxidizing Linoleic Acid. Lipids 1981, 16, 347-350.
Funes, J.A.; U. Weiss; and M. Karel. Effects of Reaction Conditions and Reactant Concentra-
tions on Polymerization of Lysozyme Reacted with Peroxidizing Lipids. J. Agric. Food
Chem. 1982, 29, 404-407.
Gamage, P.T.; T. Mori; and A. Matsushita. Mechanism of Polymerization of Proteins by Au-
toxidized Products of Linoleic Acid. J. Nutr. Sci. Vitaminol. (Tokyo). 1973, 19, 173-182.
Gardner, H.W. Lipid Hydroperoxide Reactivity with Proteins and Amino Acids: A Review. J.
Agric. Food Chem. 1979, 27, 220-229.
Gardner, H.W. Oxygen Radical Chemistry of Polyunsaturated Fatty Acids. Free Radic. Biol.
Med. 1989, 7, 65-86.
Gardner, H.W.; R. Kleiman; D. Weisleder; and G.E. Inglett. Cysteine Adds to Lipid Hydro-
peroxide. Lipids 1977, 12, 655-660.
Gardner, H.W.; and R. Kleiman. Degradation of Linoleic Acid Hydroperoxides by a Cysteine-
FeCl3 Catalyst as a Model for Similar Biochemical Reactions. II. Specificity in Formation
of Fatty Epoxides. Biochim. Biophys. Acta 1981, 665 (1), 113-125.
Gardner, H.W.; and D. Weisleder. Addition of N-Acetylcysteine to Linoleic Acid Hydroper-
oxide. Lipids 1976, 11, 127-134.
Garner, B.; A.R. Waldeck; P.K. Witting; K.-A. Rye; and R. Stocker. Oxidation of High Density
Lipoproteins. II. Evidence for Direct Reduction of Lipid Hydroperoxides by Methionine
Residues of Apolipoproteins A1 and A2. J. Biol. Chem. 1998a, 273, 6088-6095.
Garner, B.; P.K. Witting; A.R. Waldeck; J.K. Christison; M. Raftery; and R. Stocker. Oxida-
tion of High Density Lipoproteins. I. Formation of Methionine Sulfoxide in Apolipo-
proteins A1 and A2 Is an Early Event that Accompanies Lipid Peroxidation and Can Be
Enhanced by α-Tocopherol. J. Biol. Chem. 1998b, 273, 6080-6087.
Garrison, W.M. Reaction Mechanisms in the Radiolysis of Peptides, Polypeptides, and Pro-
teins. Chem. Rev. 1987, 87, 381-398.
Gebicki, S.; and J.M. Gebicki. Crosslinking of DNA and Proteins Induced by Protein Hydro-
peroxides. Biochemical J. 1999, 338, 629-636.
Gerrard, J.A.; P.K. Brown; and S.E. Fayle. Maillard Crosslinking of Food Proteins I: The Re-
action of Glutaraldehyde, Formaldehyde, and Glyceraldehyde with Ribonuclease. Food
Chem. 2002, 79, 343-349.
Gillotte, K.L.; S. Hörkkö; J.L. Witztum; and D. Steinberg. Oxidized Phospholipids, Linked to
Apolipoprotein B of Oxidized LDL, Are Ligands for Macrophage Scavenger Receptors. J.
Lipid Res. 2000, 41, 824-833.

Giron-Calle, J.; M. Alaiz; F. Millan; V. Ruiz-Guiterrez; and E. Vioque. Bound Malondial-
dehyde in Foods: Bioavailability of N,N’-Di-(4-Methyl-1,4-Dihydropyridine-3,5-
Dicarbaldehyde)Lysine. J. Agric. Food Chem. 2003, 51, 4799-4803.
Giulivi, C.; and K.A.J. Davies. Dityrosine and Tyrosine Oxidation Products are Endogenous
Markers for the Selective Proteolysis of Oxidatively Modified Red Blood Cell Hemoglo-
bin by (the 19S) Proteosome. J. Biol. Chem. 1993, 268, 8752-8759.
Giulivi, C.; N.J. Traaseth; and K.A.J. Davies. Tyrosine Oxidation Products: Analysis and Bio-
logical Relevance. Amino Acids 2003, 25, 227-232.
Green, R.C.; C. Little; and P.J. O’Brien. The Inactivation of Isocitrate Dehydrogenase by a
Lipid Peroxide. Arch. Biochem. Biophys. 1971, 142, 598-605.
Grune, T.; and K.A.J. Davies. The Proteasomal System and HNE-Modified Proteins. Mol.
Aspects Med. 2003, 24, 195-204.
Grune, T.; T. Reinheckel; and K.A.J. Davies. Degradation of Oxidized Proteins in Mammalian
Cells. FASEB J. 1997, 11, 526-534.
Gu, X.; S.G. Meer; M. Miyagi; M.E. Rayborn; J.G. Hollyfield; J.W. Crabb; and R.G. Sa-
lomon. Carboxyethylpyrrole Protein Adducts and Autoantibodies, Biomarkers for Age-
Related Macular Degeneration. J. Biol. Chem. 2003, 278, 42027-42035.
Guitton, J.; F. Tinardon; R. Lamrini; P. Lacan; M. Desage; and A. Francina. Decarboxylation
of [1-13C]Leucine by Hydroxyl Radicals. Free Radic. Biol. Med. 1998, 25, 340-345.
Haberland, M.E. Specificity of Receptor-Mediated Recognition of Malondialdehyde-Modi-
fied Low Density Lipoproteins. Proc. Nat. Acad. Sci. USA 1982, 79, 1712-1716.
Haberland, M.E. Role of Lysines in Mediating Interaction of Modified Low Density Lipo-
proteins with the Scavenger Receptor of Human Monocyte Macrophages. J. Biol. Chem.
1984, 259, 11305-11311.
Haberland, M.E.; D. Fong; and I. Cheng. Malondialdehyde-Altered Protein Occurs in Ath-
eromas of Watanabe Heritable Hyperlipidemic Rabbits. Science 1988, 241, 215-217.
Halligudi, N.N.; S.M. Desai; S.K. Mavalangi; and S.T. Nandibewoor. Kinetics of the Oxi-
dative Degradation of rac-Serine by Aqueous Alkaline Permanganate. Monatsh. Chem.
2000, 131, 321-332.
Ham, E.A.; H.G. Oien; E.H. Ulm; and F.A. Kuehl, Jr. The Reaction of PGA1 with Sulfhydryl
Groups: A Component in the Binding of A-Type Prostaglandins to Proteins. Prostaglan-
dins 1975, 10, 217-229.
Hamberg, M.; and B. Gotthammar. A New Reaction of Unsaturated Fatty Acid Hydroperox-
ides: Formation of 11-Hydroxy-12,13-Epoxy-9-Octadecenoic Acid from 13-Hydroper-
oxy-9,11-Octadecadienoic Acid. Lipids 1973, 8, 737-744.
Harris, L.; and H.S. Olcott. Reaction of Aliphatic Tertiary Amines with Hydroperoxides. J.
Am. Oil Chem. Soc. 1966, 43, 11-15.
Hasegawa, K.; and L.K. Patterson. Pulse Radiolysis Studies in Model Lipid Systems: Forma-
tion and Behavior of Peroxy Radicals in Fatty Acids. Photochem. Photobiol. 1978, 28,
817-823.
Headlam, H.A.; and M.J. Davies. Cell-Mediated Reduction of Protein and Peptide Hydroper-
oxides to Reactive Free Radicals. Free Radic. Biol. Med. 2003, 34, 44-55.
Headlam, H.A.; A. Mortimer; C.J. Easton; and M.J. Davies. β-Scission and C-3(β-Carbon)
Alkoxyl Radicals on Peptides and Proteins: A Novel Pathway Which Results in the For-
mation of α-Carbon Radicals and the Loss of Amino Acid Side Chains. Chem. Res. Toxi-
col. 2000, 13, 1087-1095.
260 K.M. Schaich
Hendley, D.D.; A.S. Mildvan; M.D. Reporter; and B.L. Strehler. The Properties of Isolated
Human Cardiac Age Pigment. I. Preparation and Physical Properties. J. Gerontol. 1963a,
18, 144-450.
Hendley, D.D.; A.S. Mildvan; M.D. Reporter; and B.L. Strehler. The Properties of Isolat-
ed Human Cardiac Age Pigment. II. Chemical and Enzymatic Properties. J. Gerontol.
1963b, 18, 250-258.
Hidalgo, F.J.; M. Alaiz; and R. Zamora. Effect of pH and Temperature on Comparative Non-
enzymatic Browning of Proteins Produced by Oxidized Lipids and Carbohydrates. J. Ag-
ric. Food Chem. 1999, 47, 742-747.
Hidalgo, F.J.; E. Gallardo; and R. Zamora. Strecker Type Degradation of Phenylalanine by 4-
Hydroxy-2-Nonenal in Model Systems. J. Agric. Food Chem. 2005, 53, 10254-10259.
Hidalgo, F.J.; and J.E. Kinsella. Changes in β-Lactoglobulin B Following Interactions with
Linoleic Acid 13-Hydroperoxide. J. Agric. Food Chem. 1989, 37, 860-866.
Hidalgo, F.J.; and R. Zamora. Fluorescent Pyrrole Products From Carbonyl-Amine Reactions.
J. Biol. Chem. 1993, 268, 16190-16197.
Hidalgo, F.J.; and R. Zamora. Modification of Bovine Serum Albumin Structure Following
Reaction with 4,5(E)-Epoxy-2-(E)-Heptenal. Chem. Res. Toxicol. 2000, 13, 501-508.
Hochstein, P.; and L. Ernster. Microsomal Peroxidation of Lipids and Its Possible Role in Cel-
lular Injury. In Ciba Foundation Symposium on Cellular Injury, A.V.S. DeReuck, and J.
Knight, Eds.; Little and Brown: Boston, MA, 1964, pp. 123-135.
Hochstein, P.; and S.K. Jain. Association of Lipid Peroxidation and Polymerization of Mem-
brane Proteins with Erythrocyte Aging. Fed. Proc. 1981, 40, 183-188.
Hoff, H.F.; T.E. Whitaker; and J.O. O’Neill. Oxidation of Low Density Lipoprotein Leads
to Particle Aggregation and Altered Macrophage Recognition. J. Biol. Chem. 1992, 267,
602-609.
Horigome, T.; and M. Miura. Interaction of Protein and Oxidized Ethyl Linolenate in the
Dry State and Nutritive Value of Reacted Protein. J. Agric. Chem. Soc. Jpn. 1974, 48,
437-444.
Horigome, T.; T. Yanagida; and M. Miura. Nutritive Value of Proteins Prepared by the Reac-
tion with Oxidized Ethyl Linoleate in Aqueous Medium. J. Agric. Chem. Soc. Jpn. 1974,
48, 195-199.
Hruszkewycz, A.M.; E.A. Glende, Jr.; and R.O. Recknagel. Destruction of Microsomal Cyto-
chrome P-450 and Glucose-6-Phosphatase by Lipids Extracted from Peroxidized Micro-
somes. Toxicol. Appl. Pharmacol. 1978, 46, 695-702.
Huang, Y.; Y. Hua; and A. Qui. Soybean Protein Aggregation Induced by Lipoxygenase Cata-
lyzed Linoleic Acid Oxidation. Food Res. Int. 2006, 39, 240-249.
Huerta, F.; E. Morallóna; F. Casesb; A. Rodesa; J.L. Vázquez; and A. Aldaza. Electrochemical
Behaviour of Amino Acids on Pt(III). A Voltammetric and in situ FTIR Study. Part II.
Serine and Alanine on Pt(III). J. Electroanal. Chem. 1997, 431, 269-275.
Hunt, J.V.; J.A. Simpson; and R.T. Dean. Hydroperoxide-Mediated Fragmentation of Pro-
teins. Biochem. J. 1988, 250, 87-93.
Hunter, E.P.L.; M.F. Desrosiers; and M.G. Simic. The Effect Of Oxygen, Antioxidants, and
Superoxide Radical on Tyrosine Phenoxyl Radical Dimerization. Free Radic. Biol. Med.
1989, 6, 581-585.
Huss, H.H. Quality and Quality Changes in Fresh Fish--4. Chemical Composition, FAO
Fisheries Technical Papers, 1995, www.fao.org/docrep/v7180e/v7180e05.htm.
Ichihashi, K.; T. Osawa; S. Toyokuni; and K. Uchida. Endogenous Formation of Protein Ad-
ducts with Carcinogenic Aldehydes. J. Biol. Chem. 2001, 276, 23903-23913.

Iio, T.; and K. Yoden. Formation of Fluorescent Substances from Degradation Products of
Methyl Linoleate Hydroperoxides with Amino Compounds. Lipids 1988, 23, 1069-
1072.
Ingold, K.U. Structure and Mechanism in Organic Chemistry, 2nd edition, Cornell University
Press: Ithaca, NY, 1969,pp. 457-463.
Itakura, K.; and K. Uchida. Evidence that Malondialdehyde-Derived Aminoenimine Is Not a
Fluorescent Age Pigment. Chem. Res. Toxicol. 2001, 14, 473-475.
Iyer, R.S.; S. Ghosh; and R.G. Salomon. Levuglandin E2 Crosslinks Proteins. Prostaglandins
1989, 37, 471-480.
Januszewski, A.S.; N.L. Alderson; A.J. Jenkins; S.R. Thorpe; and J.W. Baynes. Chemical
Modification of Proteins during Peroxidation of Phospholipids. J. Lipid Res. 2005, 46,
1440-1449.
Jirousek, M.R.; K.K. Murthi; and R.G. Salomon. Electrophilic Levuglandin E2-Protein Ad-
ducts Bind Glycine: A Model for Protein Cross-Linking. Prostaglandins 1990, 40, 187-
203.
Kagan, V.E. Lipid Peroxidation. In Biomembranes, Ed.; CRC Press: Boca Raton, FL, 1988,
pp. 13-54.
Kanazawa, K.; and H. Ashida. Target Enzymes on Hepatic Dysfunction Caused by Dietary
Products of Lipid Peroxidation. Arch. Biochem. Biophys. 1991, 288, 71-78.
Kanazawa, K.; and H. Ashida. Dietary Hydroperoxides of Linoleic Acid Decompose to Al-
dehydes in Stomach Before Being Absorbed into the Body. Biochim. Biophys. Acta 1998,
1393, 349-361.
Kanazawa, K.; H. Ashida; and M. Natake. Autoxidizing Process Interaction of Linoleic Acid
with Casein. J. Food Sci. 1987, 52, 475-479.
Kanazawa, K.; G. Danno; and M. Natake. Lysozyme Damage Caused by Secondary Degrada-
tion Products during the Autoxidation Process of Linoleic Acid. J. Nutr. Sci. Vitaminol.
(Tokyo) 1975, 21, 373-382.
Kapphahn, R.J.; B.M. Giwa; K.M. Berg; H. Roehrich; X. Feng; T.W. Olsen; and D.A. Fer-
rington. Retinal Proteins Modified by 4-Hydroxynonenal: Identification of Molecular
Targets. Exp. Eye Res. 2006, 83, 165-175.
Karel, M.; K.M. Schaich; and R.B. Roy. Interaction of Peroxidizing Methyl Linoleate with
Some Proteins and Amino Acids. J. Agric. Food Chem. 1975, 23, 159-163.
Kato, Y.; Y. Mori; Y. Makino; S. Morimitsu; S. Hiroi; T. Ishikawa; T. Osawa. Formation of Ne-
(hexanonyl)lysine in protein exposed to lipid hydroperoxide. A plausible marker for lipid
hydroperoxide-derived protein modification. J. Biol. Chem. 1999, 274, 20406-20414.
Kato, Y.; K. Uchida; and S. Kawakishi. Oxidative Fragmentation of Collagen and Prolyl Pep-
tide by Cu(II)/H2O2: Conversion of Proline Residue to 2-Pyrrolidone. J. Biol. Chem.
1992, 267, 23646-23651.
Kawai, Y.; Y. Kato; H. Fujii; Y. Makino; Y, Mori, Y.; M. Naito, and T. Osawa. Immunochemi-
cal detection of a novel lysine adduct using an antibody to linoleic acid hydroperoxide-
modified protein. J. Lipid Res. 2003, 44, 1124-1131.
Kehrer, J.P.; and S.S. Biswal. The Molecular Effects of Acrolein. Toxicol. Sci. 2000, 57, 6-15.
Keller, J.N.; R.J. Mark; A.J. Bruce; E. Blanc; J.D. Rothstein; K. Uchida; G. Waeg; and M.P.
Mattson. 4-Hydroxynonenal, an Aldehydic Product of Membrane Lipid Peroxidation,
Impairs Glutamate Transport and Mitochondrial Function in Synaptosomes. Neurosci-
ence 1997, 80, 685-696.
262 K.M. Schaich
Kikugawa, K. Examination of the Extraction Methods and Re-Evaluation of Blue Fluorescence

Generated in Rat Tissues in situ. Biol. Pharm. Bull. 1994, 17, 9-15.
Kikugawa, K.; and M. Beppu. Involvement of Lipid Oxidation Products in the Formation of
Fluorescent and Cross-Linked Proteins. Chem. Phys. Lipids 1987, 44, 277-296.
Kikugawa, K.; M. Beppu; T. Kato; S. Yamaki; and H. Kasai. Accumulation of Autofluorescent
Yellow Lipofuschin in Rat Tissues Estimated by Sodium Dodecylsulfate Extraction. Mech.
Ageing Dev. 1994, 74, 135-148.
Kikugawa, K.; and Y. Ido. Studies on Peroxidized Lipids. V. Formation and Characterization
of 1,4-Dihydropyridine-3,5-Dicarbaldehydes as Model of Fluorescent Components in
Lipofuschin. Lipids 1984, 19, 600-608.
Kikugawa, K.; Y. Ido; and J. Mikami. Studies on Peroxidized Lipids. VI. Fluorescent Products
Derived from the Reaction of Primary Amines, Malondialdehyde, and Monofunctional
Aldehydes. J. Am. Oil Chem. Soc. 1984, 61, 1574-1581.
Kikugawa, K.; K. Takayanagi; and S. Watanabe. Polylysines Modified with Malonaldehyde,
Hydroxylinoleic Acid and Monofunctional Aldehydes. Chem. Pharm. Bull. 1985, 33,
5437-5444.
Kikugawa, K.; T. Kato; A. Hayasaka. Formation of Dityrosine and Other Fluorescent Amino
Acids by Reaction of Amino Acids with Lipid Hydroperoxides. Lipids 1991, 26 (11),
922-92.
Kim, J.G.; W.R. Taylor; and S. Parthasarathy. Demonstration of the Presence of Lipid Perox-
ide-Modified Proteins in Human Atherosclerotic Lesions Using a Novel Lipid Peroxide-
Modified Anti-Peptide Antibody. Atherosclerosis 1999, 143, 335-340.
King, A.J.; and S.J. Li. Association of Malonaldehyde with Rabbit Myosin Subfragment 1.
In Quality Attributes of Muscle Foods, Y.L. Xiong, F. Shahidi, and C.T. Ho Eds.; Kluwer
Academic/Plenum Publishing: New York, 1999, pp. 277-286.
Kowalik-Jankowska, T.; M. Ruta; K. Winiewska; L. Lankiewicz; and M. Dyba. Products of
Cu(II)-Catalyzed Oxidation in the Presence of Hydrogen Peroxide of the 1–10, 1–16
Fragments of Human and Mouse ß-Amyloid Peptide. J. Inorg. Biochem. 2004, 98, 940-
950.
Krogull, M.K.; and O. Fennema. Oxidation of Tryptophan in the Presence of Oxidizing
Methyl Linoleate. J. Agric. Food Chem. 1987, 35, 66-70.
Laurora, S.; E. Tamagno; F. Briatore; P. Bardini; S. Pizzimenti; C. Toaldo; P. Reffo; P. Costelli;
M.U. Dianzani; O. Danni; et al. 4-Hydroxynonenal Modulation of p53 Family Gene
Expression in the SK-N-BE Neuroblastoma Cell Line. Free Radic. Biol. Med. 2005, 38,
215-225.
Lea, C.H. Deteriorative Reactions Involving Phospholipids and Lipoproteins. J. Sci. Food Ag-
ric. 1957, 8, 1-14.
Leake, L.; and M. Karel. Nature of Fluorescent Compounds Generated by Exposure of Protein
to Oxidizing Lipids. J. Food Biochem. 1985, 9, 117-136.
Leaver, J.; A.J.R. Law; and E.Y. Brechany. Covalent Modification of Emulsified β-Casein Re-
sulting from Lipid Peroxidation. J. Colloid Interface Sci. 1999a, 210, 207-214.
Leaver, J.; A.J.R. Law; E.Y. Brechany; and C.H. McCrae. Chemical Changes in β-Lactoglobulin
Structure during Ageing of Protein-Stabilized Emulsions. Int. J. Food Sci. Nutr. 1999b,
34, 503-508.
Lederer, M.O. Reactivity of Lysine Moieties Toward γ-Hydroxy-α,β-Unsaturated Epoxides: A
Model Study on Protein-Lipid Oxidation Product Interaction. J. Agric. Food Chem. 1996,
44, 2531-2537.
Lederer, M.O.; A. Schuler; and M. Ohmenhäuser. Reactivity of Lysine Moieties Toward an

Epoxyhydroxylinoleic Acid Derivative: Aminolysis vs Hydrolysis. J. Agric. Food Chem.
1998, 47, 4611-4620.
Lee, S.H.; T. Oe; and I.A. Blair. Vitamin C-Induced Decomposition of Lipid Hydroperoxides
to Endogeneous Genotoxins. Science 2001, 292, 2083-2086.
Lee, S.H.; T. Oe; and I.A. Blair. 4,5-Epoxy-2(E)-Decenal-Induced Formation of 1,N(6)-
Etheno-2’-Deoxyadenosine and 1,N(2)-Etheno-2’-Deoxyguanosine Adducts. Chem. Res.
Toxicol. 2002, 15, 300-304.
Leonarduzzi, G.; M.C. Arkan; H. Basaga; E. Chiarpotto; A. Sevanian; and G. Poli. Lipid Oxi-
dation Products in Cell Signaling. Free Radic. Biol. Med. 2000, 28, 1370-1378.
Lewis, S.E.; and E.D. Wills. Destruction of Sulfhydryl Groups of Proteins and Amino Acids by
Peroxides of Unsaturated Fatty Acids. Biochem. Pharmacol. 1962, 11, 901-912.
Li, G.; Y. Liao; X. Wang; S. Sheng; and D. Yin. In situ Estimation of the Entire Color and
Spectra of Age Pigment-Like Materials: Application of a Front-Surface 3D-Fluorescence
Technique. Exp. Gerontol. 2006, 41, 328-336.
Li, S.J.; and A.J. King. Structural Changes of Rabbit Myosin Subfragment 1 Altered by Malo-
naldehyde, a Byproduct of Lipid Oxidation. J. Agric. Food Chem. 1999, 47, 3124-3129.
Lin, D.; H. Lee; Q. Liu; G. Perry; M.A. Smith; and L.M. Sayre. 4-Oxo-2-Nonenal Is Both
More Neurotoxic and More Protein Reactive Than 4-Hydroxy-2-Nonenal. Chem. Res.
Toxicol. 2005, 18, 1219-1231.
Lion, Y.; M. Kuwabara; and P. Riesz. Spin-Trapping and ESR Studies of the Direct Photolysis
of Aromatic Amino Acids, Dipeptides, Tripeptides and Polypeptides in Aqueous Solu-
tions. I. Phenylalanine and Related Compounds. Photochem. Photobiol. 1981, 34, 297-
307.
Little, C.; and P.J. O’Brien. Products of Oxidation of a Protein Thiol Group After Reaction
with Various Oxidizing Agents. Arch. Biochem. Biophys. 1967, 122, 406-410.
Little, C.; and P.J. O’Brien. The Effectiveness of a Lipid Peroxide in Oxidizing Protein and
Non-Protein Thiols. Biochem. J. 1968, 106, 419-423.
Liu, Q.; M.A. Smith; J. Avilá; J. DeBernardis; M. Kansal; A. Takeda; X. Zhu; A. Nunomura;
K. Honda; P.I. Moreira; et al. Alzheimer-Specific Epitopes of Tau Represent Lipid Peroxi-
dation-Induced Conformations. Free Radic. Biol. Med. 2005, 38, 746-754.
Liu, W.; and J.-Y. Wang. Modifications of Protein by Polyunsaturated Fatty Acid Ester Peroxi-
dation Products. Biochim. Biophys. Acta 2005, 1752, 93-98.
Liu, Y.; G. Sun; A. David; and L.M. Sayre. Model Studies on the Metal-Catalyzed Protein
Oxidation: Structure of a Possible His-Lys Cross-Link. Chem. Res. Toxicol. 2004, 17,
110-118.
Liu, Z.; P.E. Minkler; and L.M. Sayre. Mass Spectroscopic Characterization of Protein Modi-
fication by 4-Hydroxy-2-(E)-Nonenal and 4-Oxo-2-(E)-Nonenal. Chem. Res. Toxicol.
2003, 16, 901-911.
Liu, Z.; and L.M. Sayre. Model Studies on the Modification of Proteins by Lipoxidation-De-
rived 2-Hydroxyaldehydes. Chem. Res. Toxicol. 2003, 16, 232-241.
Lqari, H.; J. Pedroche; J. Giron-Calle; J. Vioque; and F. Millan. Interaction of Lupinus an-
gustifolius L. α and γ Conglutins with 13-Hydroperoxide-11,9-Octadienoic Acid. Food
Chem. 2003, 80, 517-523.
Luxford, C.; R.T. Dean; and M.J. Davies. Radicals Derived from Histone Hydroperoxides
Damage Nucleobases in RNA and DNA. Chem. Res. Toxicol. 2000, 13, 665-672.
Lynch, M.P.; and C. Faustman. Effect of Aldehyde Lipid Oxidation Products on Myoglobin.
264 K.M. Schaich
J. Agric. Food Chem. 2000, 48, 600-604.

Maga, J.A. Pyridines in Foods. J. Agric. Food Chem. 1981, 29, 895-898.
Mahley, R.W.; T.L. Innerarity; R.E. Pitas; K.H. Weisgraber; J.H. Brown; and E. Gross. Inhibi-
tion of Lipoprotein Binding to Cell Surface Receptors of Fibroblasts Following Selective
Modification of Arginyl Residues in Arginine-Rich and B Apoproteins. J. Biol. Chem.
1977, 252, 7279-7287.
Malshet, M.G.; and A.L. Tappel. Fluorescent Products of Lipid Peroxidation. I. Structural
Requirement for Fluorescence in Conjugated Schiff Bases. Lipids 1973, 8, 194-198.
Mason, R.P. Using Anti-5,5’-Dimethyl-1-[Pyrroline N-Oxide] (Anti-DMPO) to Detect Pro-
tein Radicals in Time and Space with Immuno-Spin Trapping. Free Radic. Biol. Med.
2004, 36, 1214-1223.
Matoba, T.; O. Kurita; and D. Yonezawa. Changes in Molecular Size and Chemical Properties
of Gelatin Caused by the Reaction with Oxidizing Methyl Linoleate. Agric. Biol. Chem.
1984a, 48, 2633-2638.
Matoba, T.; D. Yonezawa; B.M. Nair; and M. Kito. Damage of Amino Acid Residues of Pro-
teins after Reaction with Oxidizing Lipids: Estimation by Proteolytic Enzymes. J. Food
Sci. 1984b, 49, 1082-1084.
Matsushita, S. Specific Interactions of Linoleic Acid Hydroperoxides and Their Secondary
Oxidation Products with Enzyme Proteins. J. Agric. Food Chem. 1975, 23, 150-154.
Matsushita, S.; M. Kobayashi; and Y. Nitta. Inactivation of Enzymes by Linoleic Acid Hydro-
peroxides and Linoleic Acid. Agric. Biol. Chem. 1970, 34, 817-824.
Matysik, J.; P.S. Alia; B. Bhalu; and P. Mohanty. Molecular Mechanisms of Quenching of
Reactive Oxygen Species by Proline under Stress in Plants. Current Sci. 2002, 82 (5),
525-532.
McCall, M.R.; J.Y. Tang; J.K. Bielicki; and T.M. Forte. Inhibition of Lecithin:Cholesterol
Acyl Transferase and Modification of HDL Apolipoproteins by Aldehydes. Arterioscler.
Thromb. Vasc. Biol. 1995, 15, 1599-1605.
McCay, P.B. Studies on Microsomal Phospholipids That Inhibit Gulonolactone Oxidase. J.
Biol. Chem. 1966, 241, 2333-2339.
McKnight, R.C.; and F.E. Hunter, Jr. Mitochondrial Membrane Ghosts Produced by Lipid
Peroxidation Induced by Ferrous Ion. II. Composition and Enzymatic Activity. J. Biol.
Chem. 1966, 241, 2757-2765.
McMurray, J. Organic Chemistry, Brooks/Cole: Pacific Grove, CA, 2000, pp.718-724, 770-
774, 955-960.
Menzel, D.B. Reaction of Oxidizing Lipids with Ribonuclease. Lipids 1967, 2, 83-84.
Mickel, H.S.; E.L. Foulds, Jr.; and D.A. Clark. Inhibition of the Plasma Lecithin-Cholesterol
Acytransferase Reaction by Hydrogen Peroxide and Peroxidized Lecithin. Lipids 1972, 7,
121-124.
Mitchell, D.Y.; and D.R. Petersen. Inhibition of Rat Hepatic Mitochondrial Aldehyde Dehy-
drogenase Mediated Acetaldehyde Oxidation by trans-4-Hydroxy-2-Nonenal. Hepatology
1991, 13, 728-734.
Moll, T.S. Characterization of the Reactivity, Regioselectivity, and Stereoselectivity of the Re-
actions of Butadiene Monoxide with Valinamide and the N-Terminal Valine of Mouse
and Rat Hemoglobin. Chem. Res. Toxicol. 1999, 12, 679-689.
Moll, T.S.; A.C. Harms; and A.A. Elfarra. A Comprehensive Structural Analysis of Hemo-
globin Adducts Formed after in vitro Exposure of Erythrocytes to Butadiene Monoxide.
Chem. Res. Toxicol. 2000, 13, 1103-1113.
Montine, T.J.; V. Amarnath; M.E. Martin; W.J. Strittmatter; and D.E. Graham. (E)-4-Hy-
droxy-2-Nonenal Is Cytotoxic and Crosslinks Cytoskeletal Proteins in P19 Neuroglial
Cultures. Am. J. Pathol. 1996, 148, 89-93.
Morzel, M.; P. Gatellier; T. Sayd; M. Renerre; and E. Laville. Chemical Oxidation Decreases
Proteolytic Susceptibility of Skeletal Muscle Myofibrillar Proteins. Meat Science 2006,
73, 536-543.
Musatov, A.; C.A. Carroll; Y.C. Liu; G.I. Henderson; S.T. Weintraub; and N.C. Robinson.
Identification of Bovine Heart Cytochrome c Oxidase Subunit Modified by the Lipid
Peroxidation Product 4-Hydroxy-2-Nonenal. Biochemistry 2002, 41, 8212-8220.
Nadkarni, D.V.; and L.M. Sayre. Structural Definition of Early Lysine and Histidine Adduc-
tion Chemistry of 4-Hydroxynonenal. Chem. Res. Toxicol. 1995, 8, 284-291.
Nair, V.; R.J. Offerman; G.A. Turner; A.N. Pryor; and N.C. Baenziger. Fluorescent 1,4-Dihy-
dropyridines--The Malondialdehyde Connection. Tetrahedron 1988, 44, 2793-2803.
Nair, V.; D.E. Vietti; and C.S. Cooper. Degenerative Chemistry of Malonaldehyde: Structure,
Stereochemistry, and Kinetics of Formation of Enaminals from Reaction with Amino
Acids. J. Am. Chem. Soc. 1981, 103, 3030-3036.
Narayan, K.A.; and F.A. Kummerow. Oxidized Fatty Acid-Protein Complexes. J. Am. Oil
Chem. Soc. 1958, 35, 52-56.
Nazir, H.; M. Yildiz; H. Yilmaz; M.N. Tahir; and D. Ülkü. Intramolecular Hydrogen Bonding
and Tautomerism in Schiff bases. Structure of N-(2-Pyridil)-2-Oxo-1-Naphthylidene-
methylamine. J. Mol. Struct. 2000, 524, 241-250.
Nielsen, H.K. Covalent Binding of Cardiolipin to γ-Globulin and Albumin Depends on Per-
oxidation. Chem. Phys. Lipids 1978, 22, 339-340.
Nielsen, H.K.; P.A. Finot; and R.F. Hurrell. Reactions of Proteins with Oxidizing Lipids. 2.
Influence on Protein Quality and on the Bioavailability of Lysine, Methionine, Cyst(e)ine
and Tryptophan as Measured in Rat Assays., Br. J. Nutr. 1985a, 53, 75-86.
Nielsen, H.K.; J. Löliger; and R.F. Hurrell. Reactions of Proteins with Oxidizing Lipids. 1.
Analytical Measurements of Lipid Oxidation and of Amino Acid Losses in a Whey Pro-
tein-Methyl Linolenate Model System. Br. J. Nutr. 1985b, 53, 61-73.
Nishida, T.; and F.A. Kummerow. Interaction of Serum Lipoproteins with the Hydroperoxide
of Methyl Linoleate. J. Lipid Res. 1960, 1, 450-458.
Obanu, Z.A.; D.A. Ledward; and R.A. Lawrie. Lipid Protein Interactions as Agents of Qual-
ity Deterioration in Intermediate Moisture Meats: An Appraisal. Meat Science 1980, 4,
79-88.
O’Brien, P.J.; and A.C. Frazer. The Effect of Lipid Peroxides on the Biochemical Constituents
of the Cell. Proc. Nutr. Soc. 1966, 25, 9-18.
Ooizumi, T. Effect of Peroxidized Lipid as an Oxidant on Biochemical Properties of Fish Myo-
fibrillar Proteins. Erisorubinsan no Kenkyu 1999, 5, 1-7.
Ottolenghi, A.; F. Bernheim; and K.M. Wilbur. The Inhibition of Certain Mitochondrial En-
zymes by Fatty Acids Oxidized by Ultraviolet Light or Ascorbic Acid. Arch. Biochem.
Biophys. 1955, 56, 157-164.
Page, S.; C. Fischer; B. Baumgartner; M. Haas; U. Kreusel; G. Loidl; M. Hayni; H.W.L.
Ziegler-Heitbrock; D. Neumeier; and K. Brand. 4-Hydroxynonenal Prevents NF-κB Ac-
tivation and Tumor Necrosis Factor Expression by Inhibiting I-κB Phosphorylation and
Subsequent Proteolysis. J. Biol. Chem. 1999, 274, 11611-11618.
Pamplona, R.; E. Dalfó; V. Ayala; M.J. Bellmunt; J. Prat; I. Ferrer; and M. Portero-Otín. Pro-
teins in Human Brain Cortex Are Modified by Oxidation, Glycoxidation, and Lipoxida-
266 K.M. Schaich
tion. J. Biol. Chem. 2005, 280, 21522-21530.

Patterson, L.K.; and K. Hasegawa. Pulse Radiolysis Studies in Model Lipid Systems. The Influ-
ence of Aggregation on Kinetic Behavior of OH Induced Radicals in Aqueous Sodium
Linoleate. Ber. Bunsenges. Phys. Chem. 1978, 82, 951-956.
Petersen, D.R.; and J.A. Doorn. Reactions of 4-Hydroxynonenal with Proteins and Cellular
Targets. Free Radic. Biol. Med. 2004, 37, 937-945.
Pietzsch, J. Measurement of 5-Hydroxy-2-Aminovaleric Acid as a Specific Marker of Iron-Me-
diated Oxidation of Proline and Arginine Side-Chain Residues of Low-Density Lipopro-
tein Apolipoprotein B-100. Biochem. Biophys. Res. Commun. 2000, 270, 852-857.
Player, T.J.; and H.O. Hultin. The Effect of Lipid Peroxidation on the Calcium-Accumulat-
ing Ability of the Microsomal Fraction Isolated from Chicken Breast Muscle. Biochem. J.
1978, 174, 17-22.
Podrez, E.A.; E. Poliakov; Z. Shen; R. Zhang; Y. Deng; M. Sun; P.J. Finton; L. Shan; M. Feb-
braio; D.P. Hajjar; et al. A Novel Family of Atherogenic Oxidized Phospholipids Promotes
Macrophage Foam Cell Formation Via the Scavenger Receptor CD36 and Is Enriched in
Atherosclerotic Lesions. J. Biol. Chem. 2002a, 277, 38517-38523.
Podrez, E.A.; E. Poliakov; Z. Shen; R. Zhang; Y. Deng; M. Sun; P.J. Finton; L. Shan; B. Gugiu;
P.L.H. Fox; et al. Identification of a Novel Family of Oxidized Phospholipids That Serve
as Ligands for the Macrophage Scavenger Receptor CD36. J. Biol. Chem. 2002b, 277,
38503-38516.
Pokorny, J.; S. Klein; and J. Koren. Reaction of Oxidized Lipids with Protein. II. Reactions of
Albumin with Epoxy Derivatives. Nahrung 1966, 10, 321-325.
Pokorny, J.; M. Kminek; W. Janitz; E. Novotna; and J. Davidek. Reactions of Oxidized Lipids
with Protein. Part 13. Autoxidation of Hexanal in Presence of Nonlipid Substances. Die
Nahrung 1985, 29, 459-465.
Prescott, S.M. A Thematic Series on Oxidation of Lipids as a Source of Messengers. J. Biol.
Chem. 1999, 274, 22901.
Pryor, W.A. Free Radicals, McGraw-Hill: New York, 1966. pp. 118-126.
Ran, C. Synthesis and Protein Reactivity of 2E,4E,6E,-Dodecatrienal. Tetrahedron Lett. 2004,
45, 7851-7853.
Refsgaard, H.H.F.; L. Tsai; and E.R. Stadtman. Modifications of Proteins by Polyunsaturated
Fatty Acid Peroxidation Products. Proc. Nat. Acad. Sci. USA 2000, 97, 611-616.
Requena, J.R.; C.-C. Chao; R.L. Levine; E.R. Stadtman. Glutamic and Aminoadipic Semial-
dehydes are the Main Carbonyl Products of Metal-Catalyzed Oxidation of Proteins. Proc.
Nat. Acad. Sci. USA 2001 98, 69-74.
Reubsaet, J.L.E.; J.H. Beijnen; A. Bult; R.J. van Maanen; J.A.D. Marchal; and W.J.M. Un-
derberg. Analytical Techniques Used to Study the Degradation of Proteins and Peptides:
Chemical Instability. J. Pharm. Biomed. Anal. 1998, 17, 955-978.
Riley, M.; and J.J. Harding. The Reaction of Malonaldehyde with Lens Proteins and the Pro-
tective Effect of Aspirin. Biochim. Biophys. Acta 1993, 1158, 107-112.
Roubal, W.T. Trapped Radicals in Dry Lipid-Protein Systems Undergoing Oxidation. J. Am.
Oil Chem. Soc. 1970, 47, 141-144.
Roubal, W.T.; and A.L. Tappel. Damage to Proteins, Enzymes, and Amino Acids by Peroxidiz-
ing Lipids. Arch. Biochem. Biophys. 1966a, 113, 5-8.
Roubal, W.T.; and A.L. Tappel. Polymerization of Proteins Induced by Free-Radical Lipid
Peroxidation. Arch. Biochem. Biophys. 1966b, 113, 150-155.
Roy, R.B. and M. Karel. Reaction Products of Histidine with Autoxidized Methyl Linoleate. J.
Food Sci. 1973, 38 (5), 896-897.

Saeed, S.; S.A. Fawthrop; and N. Howell. Electron Spin Resonance (ESR) Study on Free Radi-
cal Transfer in Fish Lipid-Protein Interaction. J. Sci. Food Agric. 1999, 79, 1809-1816.
Saeed, S.; D. Gillies; G. Wagner; and N.K. Howell. ESR and NMR Spectroscopy Studies
on Protein Oxidation and Formation of Dityrosine in Emulsions Containing Oxidised
Methyl Linoleate. Food Chem. Toxicol. 2006, 44, 1385-1392.
Salomon, R.G.; K. Kaur; E. Podrez; H.F. Hoff; A.V. Krushinsky; and L.M. Sayre. HNE-De-
rived 2-Pentylpyrroles Are Generated during Oxidation of LDL, Are More Prevalent in
Blood Plasma from Patients with Renal Disease or Atherosclerosis, and Are Present in
Atherosclerotic Plaques. Chem. Res. Toxicol. 2000, 13, 557-564.
Sánchez-Vioque, R.; J. Vioque; A. Clemente; J. Pedroche; J. Bautista; and F. Millán. Interac-
tion of Chickpea (Cicer arietinum L.) Legumin with Oxidized Linoleic Acid. J. Agric.
Food Chem. 1999, 47, 813-818.
Sayre, L.M.; P.K. Arora; R.S. Iyer; and R.G. Salomon. Pyrrole Formation from 4-Hydroxynon-
enal and Primary Amines. Chem. Res. Toxicol. 1993, 6, 19-22.
Sayre, L.M.; D.A. Zelasko; P.L. Harris; G. Perry; R.G. Salomon; and M.A. Smith. 4-Hy-
droxynonenal-Derived Advanced Lipid Peroxidation End Products Are Increased in Al-
zheimer’s Disease. J. Neurochem. 1997, 68, 2092-2097.
Schaich, K.M. Free Radical Initiation in Proteins and Amino Acids by Ionizing and Ultravio-
let Radiations and Lipid Oxidation--Part I: Ionizing Radiation. CRC Crit. Rev. Food Sci.
Nutr. 1980a, 13, 89-129.
Schaich, K.M. Free Radical Initiation in Proteins and Amino Acids by Ionizing and Ultraviolet
Radiations and Lipid Oxidation--Part III: Free Radical Transfer from Oxidizing Lipids.
CRC Crit. Rev. Food Sci. Nutr. 1980b, 13, 189-244.
Schaich, K.M. EPR Methods for Detecting and Identifying Free Radicals in Foods. In Free
Radicals in Foods: Chemistry, Nutrition, and Health, C.T. Ho, and M. Moreno, M. Eds.;
American Chemical Society: Washington, D.C., 2002a, pp. 12-34.
Schaich, K.M. Free Radical Generation during Extrusion: A Critical Contributor to Textur-
ization. In Free Radicals in Foods: Chemistry, Nutrition, and Health, C.T. Ho, and M.
Moreno, Eds.; American Chemical Society: Washington, D.C., 2002b, pp. 35-48.
Schaich, K.M. NO Production during Thermal Processing of Beef: Evidence for Protein Oxi-
dation. In Free Radicals in Foods: Chemistry, Nutrition, and Health, C.T. Ho, and M. Mo-
rello, Eds.; American Chemical Society: Washington, D.C., 2002c, pp. 151-161.
Schaich, K.M. A Spin Label Study of Water Binding and Protein Mobility in a Lysozyme
Model System. In Free Radicals in Foods: Chemistry, Nutrition, and Health, C.T. Ho, and
M. Moreno, Eds.; American Chemical Society: Washington, D.C., 2002d, pp. 98-113.
Schaich, K.M. Lipid Oxidation in Fats and Oils: An Integrated View. In Bailey’s Industrial Fats
and Oils, 6th Edn., F. Shahidi, Ed.; John Wiley: New York, 2005, pp. 2681-2767.
Schaich, K.M.; and M. Karel. Free Radicals in Lysozyme Reacted with Peroxidizing Methyl
Linoleate. J. Food Sci. 1975, 40, 456-459.
Schaich, K.M.; and M. Karel. Free Radical Reactions of Peroxidizing Lipids with Amino Acids
and Proteins: An ESR Study. Lipids 1976, 11, 392-400.
Schauenstein, E.; M. Taufer; H. Esterbauer; A. Kylianek; and T. Seelich. Reaction of Protein
-SH Groups with 4-Hydroxy-2-Nonenal. Monatsch. Chem. 1971, 102, 517-529.
Schaur, R.J. Basic Aspects of the Biochemical Reactivity of 4-Hydroxynonenal. Mol. Aspects
Med. 2003, 24, 149-159.
Schmolka, I.R.; and P.E. Spoerri. Thiazolidine Chemistry. II. The Preparation of 2-Substituted
268 K.M. Schaich
Thiazolidine-4-Carboxylic Acids. J. Org. Chem. 1957, 22, 943-946.

Schöneich, C.; and V.S. Sharov. Mass Spectrometry of Protein Modifications by Reactive Oxy-
gen and Nitrogen Species. Free Radic. Biol. Med. 2006, 41, 1507-1520.
Schuessler, H.; K. Schilling. Oxygen Effect in the Radiolysis of Proteins. Int. J. Radiat. Biol.
1984, 45, 267-281.
Scofield, R.H.; B.T. Kurien; S. Ganick; M.T. McClain; Q. Pye; J.A. Kames; R.I. Schneider;
R.H. Broyles; M. Bachmann; and K. Hensley. Modification of Lupus-Associated 60-kDa
Ro Protein with the Lipid Oxidation Product 4-Hydroxy-2-Nonenal Increases Antigenic-
ity and Facilitates Epitope Spreading. Free Radic. Biol. Med. 2005, 38, 719-728.
Shi, H.; S. Shen; H. Suna; Z. Liua; and L. Lia. Oxidation of L-Serine and L-Threonine by
bis(Hydrogen Periodato)Argentate(III) Complex Anion: A Mechanistic Study. J. Inorg.
Biochem. 2007, 101, 165-172.
Shimasaki, H.; N. Ueta; and O.S. Privett. Covalent Binding of Peroxidized Linoleic Acid to
Protein and Amino Acids as Models for Lipofuschin Formation. Lipids 1982, 17, 878-
883.
Shin, B.C.; J.W. Huggins; and K.L. Carraway. Effects of pH, Concentration and Aging on the
Malonaldehyde Reaction with Protein. Lipids 1972, 7, 229-233.
Shoukry, M.I.; E.L. Gong; and A.V. Nichols. Apolipoprotein-Lipid Association in Oxidatively
Modified HDL and LDL. Biochim. Biophys. Acta 1994, 1210, 355-360.
Siems, W.; and T. Grune. Intracellular Metabolism of 4-Hydroxynonenal. Mol. Aspects Med.
2003, 24, 167-175.
Simpson, J.A.; S. Narita; S. Gieseg; S. Gebicki; and J.M. Gebicki. Long-lived Reactive Species
on Free-Radical-Damaged Proteins. Biochem. J. 1992, 282, 621-624.
Slatter, D.A.; M. Murray; and A.J. Bailey. Formation of a Dihydropyridine Derivative as a Po-
tential Cross-Link Derived from Malondialdehyde in Physiological Systems. FEBS Lett.
1998, 421, 180-184.
Smith, D.M.; S.H. Noormarji; J.F. Price; M.R. Bennink; and T.J. Herald. Effect of Lipid
Oxidation on the Functional and Nutritional Properties of Washed Chicken Myofibrils
Stored at Different Water Activities. J. Agric. Food Chem. 1990, 38, 1307-1312.
Soreghan, B.A.; F. Yang; S.N. Thomas; J. Hsu; and A.J. Yang. High-Throughput Proteomic-
Based Identification of Oxidatively Induced Protein Carbonylation in Mouse Brain.
Pharm. Res. 2003, 20, 1713-1720.
Soszylqski, M.; A. Filipiak; G. Bartosz; and J.M. Gebicki. Effect of Amino Acid Peroxides on
the Erythrocyte. Free Radic. Biol. Med. 1996, 20, 45-51.
Soyer, A.; and H.O. Hultin. Kinetics of Oxidation of the Lipids and Proteins of Cod Sarco-
plasmic Reticulum. J. Agric. Food Chem. 2000, 48, 2127-2134.
Spiteller, G. Peroxyl radicals: Inductors of neurodegenerative and other inflammatory diseases.
Their origin and how they transform cholesterol, phospholipids, plamalogens, polyun-
saturated fatty acids, sugars, and proteins into deleterious products. Free Radic. Biol. Med.
2006, 41, 362-387.
Stadtman, E.R. Role of Oxidant Species in Aging. Curr. Med. Chem. 2004, 11, 1105-1112.
Stadtman, E.R.; and B.S. Berlett. Reactive Oxygen-Mediated Protein Oxidation in Aging and
Disease. Chem. Res. Toxicol. 1997, 10, 485-494.
Stadtman, E.R.; and R.L. Levine. Free Radical-Mediated Oxidation of Free Amino Acids and
Amino Acid Residues in Proteins. Amino Acids 2003, 25, 207-218.
Stapelfeldt, H.; and L.H. Skibsted. Kinetics of Formation of Fluorescent Products from
Hexanal and L-Lysine in a Two-Phase System. Lipids 1996, 31, 1125-1132.
Steinbrecher, U.P. Oxidation of Human Low Density Lipoprotein Results in Derivatization of

Lysine Residues of Apolipoprotein B by Lipid Peroxide Decomposition Products. J. Biol.
Chem. 1987, 262, 3603-3608.
Steinbrecher, U.P.; J.L. Witztum; S. Parthasarathy; and D. Steinberg. Decrease in Reactive
Amino Groups during Oxidation or Endothelial Cell Modification of LDL. Correlation
with Changes in Receptor-Mediated Catabolism. Arterioscler. Thromb. Vasc. Biol. 1987,
7, 135-143.
Suyama, K.; and A. Adachi. Origin of Alkyl-Substituted Pyridines in Food Flavor: Formation
of the Pyridines from the Reaction of Alkanals with Amino Acids. J. Agric. Food Chem.
1980, 28, 546-549.
Suyama, K.; and S. Adachi. Reaction of Alkanals and Amino Acids or Primary Amines. Syn-
thesis of 1,2,3,5- and 1,3,4,5-Substituted Quarternary Pyridinium Salts. J. Org. Chem.
1979, 44, 1417-1420.
Suyama, K.; T. Arakawa; and S. Adachi. Free Fatty Aldehydes and Their Aldol Condensation
Products in Heated Meat. J. Agric. Food Chem. 1981, 29, 875-878.
Szweda, L.I.; K. Uchida; L. Tsai; and E.R. Stadtman. Inactivation of Glucose-6-phosphate
Dehydrogenase by 4-hydroxy-2-nonenal. Selective Modification of an Active-site Lysine.
J. Biol. Chem. 1993, 268 (5), 3342-3347.
Tanczos, A.; C. Mendez; and K.M. Schaich. An Investigation of Free Radicals in Foods, In-
troduction to Scientific Research project report, Douglass Project for Women in Science,
Technology, Engineering, and Math, Douglass College, Rutgers University, New Bruns-
wick, NJ, 2002.
Tappel, A.L. Studies of the Mechanism of Vitamin E Action. III. In vitro Copolymerization of
Oxidized Lipids with Proteins. Arch. Biochem. Biophys. 1955, 54, 266-280.
Tappel, A.L. Free-Radical Lipid Peroxidation Damage and Its Inhibition by Vitamin E and
Selenium. Fed. Proc. 1965, 24, 73-78.
Tappel, A.L. Lipid Peroxidation Damage to Cell Components. Fed. Proc. 1973, 32, 1870-
1874.
Thomas, P.D.; and M.J. Poznansky. Lipid Peroxidation Inactivates Liver Microsomal Glycerol-
3-Phosphate Acyl Transferase. J. Biol. Chem. 1990, 265, 2684-2691.
Tiku, M.L.; R. Shah; and G.T. Allison. Evidence Linking Chondrocyte Lipid Peroxidation to
Cartilage Matrix Protein Degradation. J. Biol. Chem. 2000, 275, 20069-20076.
Tsai, L.; P.A. Szweda; O. Vinogradova; and L.I. Szweda. Structural Characterization and Im-
munochemical Detection of a Fluorophore Derived from 4-Hydroxy-2-Nonenal and Ly-
sine. Proc. Nat. Acad. Sci. USA 1998, 95, 7975-7980.
Tsen, C.C.; and A.L. Tappel. Oxygen Lability of Cysteine in Hemoglobin. Arch. Biochem.
Biophys. 1958, 75, 243-250.
Tsuchiya, Y.; M. Yamaguchi; T. Chikuma; and H. Hojo. Degradation of Glyceraldehyde-3-
Phosphate Dehydrogenase Triggered by 4-Hydroxy-2-Nonenal and 4-Hydroxy-2-Hex-
enal. Arch. Biochem. Biophys. 2005, 438, 217-222.
Uchida, K. Current Status of Acrolein as a Lipid Peroxidation Product. Trends Cardiovasc.
Med. 1999, 9, 109-113.
Uchida, K. Role of Reactive Aldehyde in Cardiovascular Diseases. Free Radic. Biol. Med. 2000,
28, 1685-1696.
Uchida, K.; K. Itakura; H. Kawakishi; H. Hiai; S. Toyokuni; and E.R. Stadtman. Charac-
terization of Epitopes Recognized by 4-Hydroxy-2-Nonenal Specific Antibodies. Arch.
Biochem. Biophys. 1995, 324, 241-248.
270 K.M. Schaich
Uchida, K.; M. Kanematsu; Y. Morimutsu; T. Osawa; N. Noguchi; and E. Niki. Acrolein Is

a Product of Lipid Peroxidation Reaction. Formation of Free Acrolein and Its Conju-
gate with Lysine Residues in Oxidized Low Density Lipoproteins. J. Biological Chemistry
1998a, 273, 16058-16066.
Uchida, K.; M. Kanematsu; K. Sakai; T. Matsuda; N. Hattori; Y. Mizuno; D. Suzuki; T. Mi-
yata; N. Noguchi; E. Niki; et al. Protein-Bound Acrolein: Potential Markers for Oxidative
Stress. Proc. Nat. Acad. Sci. USA 1998b, 95, 4882-4887.
Uchida, K.; M. Shiraishi; Y. Naito; Y. Torii; Y. Nakamura; and T. Osawa. Activation of Stress
Signaling Pathways by the End Product of Lipid Peroxidation. 4-Hydroxy-2-Nonenal
Is a Potential Inducer of Intracellular Peroxide Production. J. Biol. Chem. 1999, 274,
2234–2242.
Uchida, K.; and E.R. Stadtman. Modification of Histidine Residues in Proteins by Reaction
with 4-Hydroxynonenal. Proc. Nat. Acad. Sci. USA 1992, 89, 4544-4548.
Uchida, K.; and E.R. Stadtman. Covalent Attachment of 4-Hydroxynonenal to Glyceralde-
hyde-3-Phosphate Dehydrogenase: A Possible Involvement of Intramolecular and Inter-
molecular Crosslinking Reactions. J. Biol. Chem. 1993, 268, 6388-6393.
Uchida, K.; L.I. Szweda; H.Z. Chae; and E.R. Stadtman. Immunochemical Detection of 4-
Hydroxynonenal Protein Adducts in Oxidized Hepatocytes. Proc. Nat. Acad. Sci. USA
1993, 90, 8742-8746.
Valko, M.; D. Leibfritz; J. Moncol; M.T.D. Cronin; M. Mazur; and J. Telser. Free Radicals and
Antioxidants in Normal Physiological Functions and Human Disease. Int. J. Biochem.
Cell Biol. 2007, 39, 44-84.
Veberg, A.; G. Vogt; and J.P. Wold. Fluorescence in Aldehyde Model Systems Related to Lipid
Oxidation. LWT 2006, 39, 562-570.
Viita, H.; O. Narvanen; and S. Yla-Herttuala. Different Apolipoprotein B Breakdown Patterns
in Models of Oxidized Low Density Lipoprotein. Life Sci. 1999, 65, 783-793.
Wainwright, T.; J.F McMahon; and J. McDowell. Formation of methional and methanethiol
from methionine. J. Sci. Food Agric. 1972, 23, 911-914.
Weisgraber, K.H.; T.L. Innerarity; and R.W. Mahley. Role of Lysine Residues of Plasma Li-
poproteins in High Affinity Binding to Cell Surface Receptors on Human Fibroblasts. J.
Biol. Chem. 1978, 253, 9053-9062.
Wills, E.D. Effect of Unsaturated Fatty Acids and Their Peroxides on Enzymes. Biochem. Phar-
macol. 1961a 7, 7-16.
Wills, E.D. Enzyme Inhibition by Oxidized Unsaturated Fatty Acids. In The Enzymes of Lipid
Metabolism, P. Desnuelle, Ed.; Pergamon: New York, 1961b, pp. 74-77.
Xu, G.; Y. Liu; M.M. Kansal; and L.M. Sayre. Rapid Crosslinking of Proteins by 4-Keto Alde-
hydes and 4-Hydroxy-Alkenals Does Not Arise from the Lysine-Derived Monoalkylpyr-
roles. Chem. Res. Toxicol. 1999a, 12, 855-861.
Xu, G.; Y. Liu; and L.M. Sayre. Mechanism of Protein Lysine Crosslinking by the Lipid Per-
oxidation Product 4-Hydroxy-2-Nonenal (HNE). Book of Abstracts, 216th ACS National
Meeting, Boston, August 23-27, 1998. ORGN-281, American Chemical Society, Wash-
ington, D. C.
Xu, G.; Y. Liu; and L.M. Sayre. Independent Synthesis, Solution Behavior, and Studies on
the Mechanism of Formation of the Primary Amine-Derived Fluorophore Represent-
ing Cross-Linking of Proteins by (E)-4-Hydroxy-2-Nonenal. J. Org. Chem. 1999b, 64,
5732-5745.
Xu, G.; Y. Liu; and L.M. Sayre. Polyclonal Antibodies to a Fluorescent 4-Hydroxy-2-Nonenal
(HNE)-Derived Lysine-Lysine Cross-Link: Characterization and Application to HNE-

Treated Protein and in vitro Oxidized Low-Density Lipoprotein. Chem. Res. Toxicol.
2000, 13, 406-413.
Xu, G.; and L.M. Sayre. Structural Characterization of a 4-Hydroxy-2 Alkenal-Derived Fluo-
rophore that Contributes to Lipoperoxidation-Dependent Protein Crosslinking in Aging
and Degenerative Disease. Chem. Res. Toxicol. 1998, 11, 247-251.
Xu, G.; and L.M. Sayre. Structural Elucidation of a 2:2 4-Ketoaldehyde-Amine Adduct as
a Model for Lysine-Directed Crosslinking of Proteins by 4-Ketoaldehydes. Chem. Res.
Toxicol. 1999, 12, 862-868.
Yamada, S.; S. Kumazawa; T. Ishii; T. Nakayama; K. Itakura; N. Shibata; M. Kobayashi; K.
Sakai; T. Osawa; and K. Uchida. Immunochemical Detection of a Lipofuschin-Like Fluo-
rophore Derived from Malondialdehyde and Lysine. J. Lipid Res. 2001, 42, 1187-1196.
Yamaki, S.; T. Kato; and K. Kikugawa. Characteristics of Fluorescence Formed by the Reaction
of Proteins with Unsaturated Aldehydes, Possible Degradation Products of Lipid Radi-
cals. Chem. Pharm. Bull. 1992, 40, 2138-2142.
Yanagita, T.; H. Orioka; and M. Sugano. Influence of Amino Acid Supplementation to Egg
Albumin Reacted with Oxidized Lipids on Liver Lipids of Rats. Agric. Biol. Chem. 1976,
40, 1751-1756.
Yanagita, T.; and M. Sugano. Effect of Egg Albumin Reacted with Oxidized Lipid on Growth
and Liver Components of Rat. J. Jpn. Soc. Food Nutr. 1974, 27, 281-287.
Yanagita, T.; and M. Sugano. Liver and Plasma Lipids in Rats Fed Casein Reacted with Oxi-
dized Ethyl Linoleate. Agric. Biol. Chem. 1975, 39, 63-69.
Yang, M.-H. Damage of DNA and Its Constituents by Oxidizing Lipids. Ph.D. thesis, Rutgers
University, New Brunswick, NJ, 1993, pp. 1-91.
Yang, M.-H. and K.M. Schaich. Factors Affecting Dna Damage by Lipid Hydroperoxides
and Aldehydes. J. Free Radical Biol. Med. 1996 20, 225-236.
Yildiz, M.; Z. Kiliç, and T. Hökelek. Intramolecular Hydrogen Bonding and Tautomerism in
Schiff Bases. Part I. Structure of 1,8-Di[N-2-Oxyphenyl-Salicylidene]-3,6-Dioxaoctane.
J. Mol. Struct. 1998, 441, 1-10.
Yong, S.H.; and M. Karel. Reaction of Histidine with Methyl Linoleate: Characterization of
the Histidine Degradation Products. J. Am. Oil Chem. Soc. 1978, 55 (3), 352-7.
Yong, S.H.; and M. Karel. Cleavage of the Imidazole Ring in Histidyl Residue Analogs Re-
acted with Peroxidizing Lipids. J. Food Sci. 1979, 22 (2), 568-74.
Yong, S.H.; S. Lau; Y. Hsieh; and M. Karel. Degradation Products of L-Tryptophan React-
ed with Peroxidizing Methyl Linoleate. In Autoxidation in Foods and Biological Systems;
Simic, M.G.,Karel, M., Ed.; Plenum Press: New York, 1980, pp. 237-247.
Yuan, Q.; X. Zhu; and L.M. Sayre. Chemical Nature of Stochastic Generation of Protein-
Based Carbonyls: Metal-Catalyzed Oxidation Versus Modification by Products of Lipid
Oxidation. Chem. Res. Toxicol. 2007, 20, 129-139.
Zamora, R.; M. Alaiz; and F.J. Hidalgo. Modification of Histidine Residues by 4,5-Epoxy-2-
Alkenals. Chem. Res. Toxicol. 1999, 12, 654-660.
Zamora, R.; M. Alaiz; and F.J. Hidalgo. Contribution of Pyrrole Formation and Polymeriza-
tion to the Nonenzymatic Browning Produced by Amino-Carbonyl Reactions. J. Agric.
Food Chem. 2000, 48, 3152-3158.
Zamora, R.; E. Gallardo; and F.J. Hidalgo. Amine Degradation by 4,5-Epoxy-2-Decenal in
Model Systems. J. Agric. Food Chem. 2006, 54, 2398-2404.
Zamora, R.; and F.J. Hidalgo. Linoleic Acid Oxidation in the Presence of Amino Compounds
272 K.M. Schaich
Produces Pyrroles by Carbonylamine Reactions. Biochim. Biophys. Acta 1995, 1258, 319-
327.
Zamora, R.; and F.J. Hidalgo. Inhibition of Proteolysis in Oxidized Lipid-Damaged Proteins.
J. Agric. Food Chem. 2001, 49, 6006-6011.
Zamora, R.; and F.J. Hidalgo. Comparative Methyl Linoleate and Methyl Linolenate Oxida-
tion in the Presence of Bovine Serum Albumin at Several Lipid/Protein Ratios. J. Agric.
Food Chem. 2003a, 51, 4661-4667.
Zamora, R.; and F.J. Hidalgo. Phosphatidylethanolamine Modification by Oxidative Stress
Product 4,5(E)-Epoxy-2(E)-Heptenal. Chem. Res. Toxicol. 2003b, 16, 1632-1641.
Zamora, R.; and F.J. Hidalgo. 2-Alkylpyrrole Formation from 4,5-Epoxy-2-Alkenals. Chem.
Res. Toxicol. 2005, 18, 342-348.
Zamora, R.; R.F. Millán; F.J. Hidalgo; M. Alaiz; M.P. Maza; J.M. Olías; and E. Vioque. Inter-
action Between the Peptide Glutathione and Linoleic Acid Hydroperoxide. Food/Nahrung
1989, 33, 283-288.
Zamora, R.; J.L. Navarro; and F.J. Hidalgo. Determination of Lysine Modification Product,
e-N-Pyrrolylnorleucine in Hydrolyzed Proteins and Trout Muscle Microsomes by Micel-
lar Electrokinetic Capillary Chromatography. Lipids 1995, 30, 477-483.
Zhang, H.; D.A. Dickinson; R.-M. Liu; and H.J. Forman. 4-Hydroxynonenal Increases
γ-Glutamyl Transpeptidase Gene Expression Through Mitogen-Activated Protein Kinase
Pathways. Free Radic. Biol. Med. 2005, 38, 463-471.
Zhang, W.-H.; J. Liu; G. Xu; Q. Yuan; and L.M. Sayre. Model Studies on Protein Side Chain
Modification by 4-Oxo-2-Nonenal. Chem. Res. Toxicol. 2003, 16, 512-523.
Zhu, M.; D.C. Spink; B. Yan; S. Bank; and A.P. DeCaprio. Formation and Structure of Cross-
Linking and Monomeric Pyrrole Autoxidation Products in 2,5-Hexanedione-Treated
Amino Acids, Peptides, and Protein. Chem. Res. Toxicol. 1994, 7, 551-558.
Zirlin, A.; and M. Karel. Oxidation Effects in a Freeze-Dried Gelatin-Methyl Linoleate Sys-
tem. J. Food Sci. 1969, 34, 160-164.
9
Lipid Oxidation in Food Dispersions
Eric A. Decker, Wilailuk Chaiyasit, Min Hu, Habibollah Faraji, and D.
Julian McClements
Department of Food Science, University of Massachusetts, Amherst, Massachusetts,
01003
Introduction
The concentrations of polyunsaturated fatty acids in food products are increasing. This
is due to the desire to improve the nutritional profile of foods by replacing atherogenic
saturated fatty acids with unsaturated fatty acids and incorporating bioactive fatty ac-
ids, such as the ω-3 fatty acids, into functional foods. In addition, recent trans fatty
acid labeling requirements throughout the world have resulted in a reduction of the
use of hydrogenated fats leading to an increase in utilization of oils with higher levels
of polyunsaturated fatty acids. Increasing the concentrations of polyunsaturated fatty
acids in foods leads to increased lipid oxidation susceptibility which can result in the
formation of undesirable rancid odors and flavors as well as changes in texture, color,
and nutritional quality. Therefore, effective methods are needed to control lipid oxida-
tion. These methods include oxygen removal, light exclusion, temperature reduction,
and utilization of antioxidants. Unfortunately, these methods are not suitable for all
food products and antioxidant additives are often label unfriendly. Therefore, new tech-
nologies are needed to control lipid oxidation in food products.
Most lipids in foods exist as food dispersions or emulsions. Food emulsions consist
of oil dispersed in water (an oil-in-water emulsion) or water dispersed in oil (a water-
in-oil emulsion). The dispersed phase exists as small spherical droplets ranging in size
from 0.1 to 100 µm (Dickinson and Stainsby, 1982; Dickinson, 1992). The positive
free energy needed to increase the surface area between the oil and water phases and the
density difference between oil and water make emulsions thermodynamically unstable
systems (Dickinson, 1992; McClements, 1999). To form emulsions that are kinetically
stable over the shelf life of the food products (a few weeks, months, or years), emulsi-
fiers must be utilized. Emulsifiers are molecules that have the ability to absorb to the
lipid-water interface allowing them to decrease surface tension, form a physical barrier
between oil and water, and impart a charge onto the emulsion droplet; all are factors
that inhibit emulsion destabilization. Proteins, phospholipids, and small molecule sur-
factants are the most common emulsifiers used in foods.
An emulsion can be described as having three distinct regions: the droplet interior,
the continuous phase, and the emulsion-droplet interface that separates the oil and
273
274 E.A. Decker et al.
water. The interface region consists mainly of the emulsifier but can also contain
other surface-active molecules, such as antioxidants and lipid hydroperoxides as well
as molecules that may be attracted towards a charged interface (e.g., transition met-
als). The properties of the emulsion-droplet interface (e.g., thickness and charge) are a
function of the type and concentration of the surface-active molecules present. Lipid
oxidation chemistry in emulsions is highly dependent on the properties of the emul-
sion-droplet interface since factors such as interfacial thickness and charge will impact
the physical location of prooxidative and antioxidative factors and thus the ability of
compounds to either promote or inhibit lipid oxidation. By understanding how the
physical properties of emulsions impact lipid oxidation chemistry, it is possible to
develop emulsion technologies to inhibit lipid oxidation.
Lipid Oxidation in Emulsions

Lipid oxidation reactions are dependent on the chemical reactivity of numerous com-
ponents including reactive oxygen species, prooxidants, and antioxidants. However,
research over the past few decades has shown that the physical properties of food
systems are extremely important to the chemistry of lipid oxidation (Abdalla and
Roozen, 1999; Frankel et al., 1994; Fritsch, 1994; Halliwell et al., 1995; McClements
and Decker, 2000; Naz et al., 2005). The most extensive research on the impact of
physical properties on lipid oxidation has been conducted in oil-in-water emulsions.
This is because numerous methods are available to characterize the physical properties
and location of compounds in oil-in-water emulsions.
Both lipid hydroperoxides and transition metals exist in foods. Most fats and oils
undergo oxidation during oil extraction and refining. Bleaching, deodorization, and
physical refining can remove many oxidation products, but in reality most commer-
cial fats and oils contain lipid oxidation products. For example lipid hydroperoxide
concentrations range from <1.0 to over 15.0 meq/kg in commercially available oils
(Chaiyasit et al., 2007). These lipid hydroperoxide concentrations have been esti-
mated to be over 10,000 times higher than the lipid hydroperoxide concentrations
found in living tissues (Decker and McClements, 2001). Metals, such as iron and
copper, are present naturally in oilseeds and fruits and thus end up in crude oil. Typi-
cal iron and copper concentrations in refined oils are <0.1 and 0.02 ppm, respectively
(Chaiyasit et al., 2007).
Transition metals react with lipid hydroperoxides to produce high-energy free
radicals (e.g., alkoxyl radicals) that can promote the oxidation of unsaturated fatty
acids and can cause the scission of fatty acids to produce the volatile compounds re-
sponsible for rancidity. Thus the oxidative stability of many food lipids is dependent
on both hydroperoxide and metal concentrations. The interactions between metals
and lipid hydroperoxides are dependent on the physical properties of foods, since sur-
face active lipid hydroperoxides tend to migrate to the water-oil interface (Decker and
McClements, 2001). For example, the ability of iron to promote lipid oxidation in
oil-in-water emulsions is influenced by the net charge of the emulsion-droplet inter-
Lipid Oxidation in Food Dispersions 275
face. In corn oil-in-water emulsion stabilized with anionic (SDS), cationic (DTAB),
or nonionic (Brij 35) surfactants, oxidation rates were highest for negatively charged
droplets, intermediate for uncharged droplets, and lowest for positively charged drop-
lets (Mancuso et al., 1999). The observed alterations in oxidation rates are likely due
to increased iron-lipid hydroperoxide interactions when positively charged iron ions
were electrostatically attracted to the surface of the negatively charged emulsion drop-
lets thus increasing metal-lipid interactions. Conversely, lipid oxidation was retarded
when the iron ions were electrostatically repelled from the surface of the positively
charged droplets. Another potential physical property that influences iron-lipid hy-
droperoxide interactions in oil-in-water is the presence of a thick barrier at the lipid-
droplet interface. The ability of iron to promote hydroperoxide decomposition as
well as oxidation of salmon oil was lower in emulsion droplets stabilized by nonionic
surfactants that have a large polar head group that forms a thick interfacial membrane
(Silvestre et al., 2000).
Ability of Proteins to Impact Lipid Oxidation in

Oil-in-Water Emulsions
The previously mentioned research with synthetic surfactants has shown that emul-
sion droplets with thick, cationic interfacial membranes have improved oxidative sta-
bility due to decreased interactions between aqueous phase prooxidants and lipid
phase oxidation substrates. Unfortunately, many of the synthetic surfactants used in
these studies are not approved for food applications. Proteins represent an emulsi-
fier that could be used to produce cationic emulsion droplets with thick interfacial
membranes. When salmon or corn oil emulsions are stabilized with proteins, oxida-
tion rates are dramatically slower when the pH is below the pI of the protein; thus
the emulsion droplet is cationic (Hu et al., 2003a, 2003b). While the existence of a
cationic charge is critical to decrease lipid oxidation rates, the charge density does
not seem to be directly related to oxidative stability. For instance, the cationic charge
density of whey protein-stabilized salmon oil emulsions at pH 3.0 was in order: β-
lactoglobulin > α-lactalbumin > whey protein isolate > sweet whey, while inhibition
of lipid oxidation was in order: β-lactoglobulin ≥ sweet whey > whey protein isolate
≥ α-lactalbumin. Similarly, the fact that the cationic charge (as determined by zeta
potential) of corn oil emulsion droplets stabilized by whey protein isolate (+55.9 mV)
was almost twice as high as casein- and soy-protein-isolate-stabilized emulsions drop-
lets ( +29.9 and +29.4 mV, respectively), while the oxidative stability of the whey-pro-
tein-isolate-stabilized emulsions was intermediate among the 3 proteins. This suggests
that the magnitude of the positive charge of the emulsion droplet charge did not have
a major impact on lipid oxidation rates.
The lack of correlation between emulsion droplet charge density and oxidative
stability suggests that additional factors impact lipid oxidation rates in protein-stabi-
lized emulsions. As described above, increasing the thickness of the interfacial mem-
brane of emulsion droplets decreases oxidation rates. This factor may help explain
why casein, which can form a thick interfacial layer around dispersed oil droplets of
up to 10 nm compared to 1-2 nm for whey proteins (Dickinson and McClements,
1995), was more effective at decreasing lipid oxidation rates than whey proteins when
it was used to stabilize corn oil-in-water emulsions (Hu et al., 2003b). An additional
factor that could account for the observed differences in the oxidative stability of the
emulsions is differences in amino acid composition between the proteins. The free
sulfhydryl group of cysteine can inhibit lipid oxidation. When whey protein isolate
was treated with N-ethylmaleimide to block free sulfhydryls prior to the formation of
emulsions, no alteration in oxidation rates was observed suggesting that free sulfhy-
dryls at the emulsion interface do not inhibit lipid oxidation rates (Hu et al., 2003a).
It is possible that other antioxidative amino acids, such as tyrosine, phenylalanine,
tryptophan, proline, methionine, lysine, and histidine, could be responsible for dif-
ferences in the oxidative stability of emulsions stabilized by various proteins.
In addition to the impact of proteins at the interface of oil-in-water emulsion
droplets, aqueous phase proteins can also influence lipid oxidation rates. Addition
of whey proteins to the continuous phase of Tween 20-stabilized salmon oil-in-wa-
ter emulsions results in inhibition of lipid oxidation (Tong et al., 2000a). The free
sulfhydryls of the continuous phase whey proteins are involved in this antioxidant
activity since blocking sulfhydryls with N-ethylmaleimide decreased antioxidant ac-
tivity. Proteins can also change the physical location of iron in emulsions suggesting
that chelation could also be involved in the antioxidant activity of continuous phase
proteins (Tong et al., 2000b).
Role of Other Emulsifiers in Lipid Oxidation Chemistry

The impact of emulsifier type on lipid oxidation can be seen further in experiments
where oil-in-water emulsions were made with whey protein isolate (WPI), gum arabic
(GA), and modified starch (MS) (Purity Gum BE, National Starch Co.). Figure 9.1
shows that menhaden oil-in-water emulsions stabilized by these emulsifiers had large
variations in oxidative stability. At both pH 7.0 and 3.0, the emulsion stabilized with
modified starch was the least oxidatively stable while gum-arabic- and whey-protein-
stabilized emulsions had similar oxidative stability when lipid oxidation was measured
with headspace propanal. When emulsions are made, the emulsifier absorbs to the
emulsion-droplet surface until the surface becomes saturated. Remaining emulsifiers
partition into the continuous phase where they impact lipid oxidation by chelating
metals and scavenging free radicals. At pH 7.0, the continuous phase WPI could be
inhibiting lipid oxidation through its ability to more strongly chelate metals (Fig. 9.2)
and scavenge free radicals (as determined by the oxygen radical absorbance capacity
(ORAC) assay, Fig. 9.3) than GA or MA. The low oxidation rates in the WPI-sta-
bilized emulsions at pH 3.0 could be due to the ability of the proteins to produce a
cationic emulsion droplet (Fig. 9.4) that can repel prooxidant metals. While GA was
effective at decreasing lipid oxidation rates (Fig. 9.1), it did not have strong metal
chelating (Fig. 9.2) or free radical scavenging (Fig. 9.3) properties. This suggest that
A
1200 W hey P rotein Isolate
G um A rabic
1000
M odif ied S tarc h
P ropanal (µM )
800
600
400
200
0
0 2 4 6 8 10 12 14
Time (D ay)
B
20 W hey P rotein Isolate
G um A rabic
M odif ied S tarc h
15
P ropanal (µM )
10
0
0 2 4 6 8 10 12 14
Time (D ay)
Fig. 9.1. Influence of emulsifier type on the formation of headspace propanal in 7.0 % Menhaden oil-in-water
emulsions stabilized by whey protein isolate, gum arabic, or modified starch (stored at 5.0°C in the dark) (A) at pH
7.0 and (B) at pH 3.0. Emulsions were washed prior to storage to remove continuous phase components.
3
Iron (m m ol/g biopolym er)
0
W hey P rotein Isolate G um A rabic M odif ied S tarc h
Fig. 9.2. Ability of whey protein isolate, gum arabic, and modified starch to bind iron as determined by the
method of Diaz et al., 2003.
1.0
0.8
F/F 0
0.6
C ontrol
0.4 0.1% G A
0.1% M S
0.2 0.01% W P I
Fluoresc ein O nly
0.0
0 10 20 30 40
T im e (m in)
Fig. 9.3. Effect of 0.01% Whey protein isolate, gum arabic, or modified starch on the relative fluorescence inten-
sity of 45 nM Fluorescein (λEX 493 nm; λEM 515 nm) in the presence of 20 mM AAPH at 37°C. Fluorescence values
(F) are given relative to the initial time values (F0). The blank (fluorescein only) was prepared without AAPH and
the control was prepared without antioxidant.
40
W hey P rotein Isolate
G um A rabic
20 M odif ied S tarc h
Zeta-P otential (m V )
-20
-40
pH 3 pH 7
Fig. 9.4. Influence of emulsifier type on electrical charge of emulsion droplets as determined by ξ-potential
(Hu et al., 2003b).
its ability to decrease lipid oxidation rates could be through the formation of an emul-
sion-droplet interfacial region that inhibits lipid-prooxidant interactions since GA
forms a thick birefringent gel-like mechanical barrier at oil-water interfaces (Garti
and Leser, 2001).
Impact of Free Radical Scavenging Antioxidants on Lipid Oxidation

in Oil-in-Water Emulsions
The impact of physical properties on lipid oxidation in oil-in-water emulsions can
also be seen with the effectiveness of free radical scavenging antioxidants. For ex-
ample, hydrophilic antioxidants are less effective in oil-in-water emulsions than lipo-
philic antioxidants (Frankel, 1998; Porter, 1980). The differences in the effectiveness
of antioxidants of varying polarity in oil-in-water emulsions is thought to be due to
the physical location of the antioxidants, since lipophilic antioxidants are retained in
the oil droplet where they can scavenge free radicals whereas hydrophilic antioxidants
partition into the aqueous phase where they are ineffective (Frankel, 1998; Porter,
1980).
The effectiveness of a lipid-soluble antioxidant is also dependent on the con-
centration since antioxidant activity does not increase linearly with concentration.
In some cases high antioxidant concentrations can decrease effectiveness since they
are able to reduce metals and increase prooxidant activity. Therefore, addition of a
lipid-soluble antioxidant to an emulsion system will not always result in increased

oxidative stability if the naturally occurring concentration of antioxidant in the oil is
already high. For example, in emulsions made with menhaden oil that did not con-
tain α-tocopherol, addition of α-tocopherol (100-1000 ppm) resulted in a decrease
in propanal formation (Fig. 9.5). Conversely, addition of α-tocopherol to algae oil-
in-water emulsions did not decreased propanal formation (Fig. 9.6) since the algae
oil contained 860 ppm α-tocopherol. Even addition of other free radical scavengers,
such as ascorbyl palmitate and rosemary extract, to the algae oil-in-water emulsion
containing 860 ppm α-tocopherol also did not increase oxidative stability (Fig. 9.7).
These data suggest that the preexisting α-tocopherol in the algae oil was sufficient to
obtain maximum free radical scavenging at the site where lipid oxidation was occur-
ring and that additional antioxidants were not able to gain access to free radicals and
thus could not further inhibit lipid oxidation.
Do Physical Structures Impact Lipid Oxidation in Bulk Oils?

During the past 50 years, most studies on lipid oxidation in bulk oils considered
the oil matrix to be a homogeneous liquid phase. However, commercial oils contain
numerous surface-active compounds such as mono- and diacylglycerols, phospho-
1200
C ontrol
1000 100 ppm
500 ppm
P ropanal (µM )
800
1000 ppm
600
400
200
0
0 20 40 60 80
Tim e (hr)
Fig. 9.5. Influence of added α-tocopherol (0-1000 ppm) on the formation of headspace propanal in 5.0% Men-
haden oil-in-water emulsions stabilized by 0.5% Whey protein isolate at ph 3.0 During storage in the dark at
37.0°C.
2400
C ontrol
2000
100 ppm
500 ppm
P ropanal (µM )
1600
1000 ppm
1200
800
400
0
0 40 80 120
Tim e (hr)
Fig. 9.6. Influence of added α-tocopherol (0-1000 ppm) on the formation of headspace propanal in 5.0% Men-
37.0°C. The algae oil used in this experiment contained 860 ppm α-tocopherol.
1600
C ontrol
1200 A sc orby l P alm itate
R osem ary E x trac t
P ropanal (µM )
C om bination
800
400
0
0 40 80 120
Tim e (hr)
Fig. 9.7. Influence of ascorbyl palmitate (1000 ppm), rosemary extract (1000 ppm), and the combination of
ascorbyl palmitate and rosemary extract (1000 ppm each) on the formation of headspace propanal in 5.0% Men-
37.0°C. The algae oil used in this experiment contained 860 ppm α-tocopherol.
lipids, and free fatty acids. When commercial oils are stripped of their minor com-
ponents, the resulting stripped oil has a higher interfacial tension (less surface-active
compounds) than the original refined oil (Table 9.1) indicating that commercial oils
contain surface-active compounds. In addition, Dana and Saguy (2006) also found
that during frying, a process well know to produce surface-active lipids, such as free
fatty acids, interfacial tension decreased from 24.4 to 13.0 mN/m but surface tension
of that oil was stable at 32.6 mN/m (Dana and Saguy, 2006).
Table 9.1. Surface and Interfacial Tension of Commercial Corn Oils as Determined by Du Noüy Ring Method at 30°C
After 24 hr of Equilibration
Oil Surface Tension (mN/m) Interfacial Tension (mN/m)
Corn Oil 31.8 + 0.1 16.4 + 0.0
Stripped Corn Oil 31.8 + 0.1 29.6 + 0.0
The combination of these surface-active compounds and the water naturally

found in commercial oils can form physical structures known as association colloids.
Evidence for the existence of association colloids in commercial algae oil can be seen
in the X-ray diffraction pattern of algal oil (Fig. 9.8). This X-ray diffraction pattern
of commercial oil is very complicated suggesting that the physical structures are a
mixture of reverse micelles and lamellar structures. This diffraction pattern strongly
suggests that the surface-active components and water in commercial oil can form
association colloids. Addition of water to a commercial algae oil increases X-ray scat-
tering intensity indicating that the oil contained excess surface-active compounds that
form additional physical structures upon the addition of water.
Physical structures, such as reverse micelles and lamellar structures, can increase
chemical reactions by creating surfaces that increase interactions between lipid- and
water-soluble compounds. Surface-active compounds with low hydrophilic-lipophilic
balances (HLB) typically form reverse micelles and lamellar structures in bulk oils.
Examples of surface-active compounds in bulk oils with low HLB numbers that could
form association colloids include free fatty acids (HLB ≈ 1.0), diacylglyerols (≈ 1.8),
and monoacylglycerols (≈3.4-3.8) (McClements, 2004). Phospholipids which have
intermediate HLB values (≈ 8, McClements, 2004) can form lamellar structures as
well as reverse micelles (Gupta et al., 2001). The complex X-ray diffraction patterns
seen in commercial algal oil suggest that the wide variety of surface-active components
found in a commercial oil form a mixture of reverse micelles and lamellar structures.
The hypothesis that association colloids are the site of lipid oxidation in bulk oils
is supported by several lines of experimental evidence. In bulk oil, polar antioxidants
are more effective than lipophilic antioxidants as described by the “antioxidant polar
paradox” (Frankel, 1998). The increased effectiveness of polar antioxidants in bulk
oils has been suggested to be due to their ability to concentrate at the oil-air interface
560 ppm H 2 O
8
7
In ten sity (a.u .)
5
0 ppm H 2 O
3
0.1 0.2 0.3 0.4 0.5
q (Å -1 )
Fig. 9.8. Experimental X-ray scattering curves of algae oil with and without added water (560 ppm).
where oxidation is prevalent. However, it is unclear why polar antioxidants would

migrate to the air-oil interface since air is less polar than oil (dielectric constant of air
is 1.0 compared to approximately 3.0 for food oils, CRC Press, 1982). If antioxidants
do migrate and concentrate at the air-oil interface they would be able to decrease
surface tension (the air-oil interface). However, this is not the case as can be seen in
Table 9.2. Polar antioxidants are more likely to concentrate at oil-water interfaces as
can be seen by their ability to decrease interfacial tension (Table 9.2). The ability of
polar antioxidants to concentrate at oil-water interfaces suggests that in bulk oils they
are more likely to partition in association colloids. Thus, the increased antioxidant
activity of polar antioxidants in bulk oil seems to be due to their concentrations in
association colloids where oxidative reactions are more prevalent.
Table 9.2. Influence of Antioxidants (1 mmol/kg lipid) on Surface and Interfacial Tension of Hexadecane as Deter-
mined by Du Noüy Ring Method at 30°C after 24 hr of Equilibration
Antioxidant Surface Tension (mN/m) Interfacial Tension (mN/m)
Control (no Antioxidant) 26.0 + 0.1 43.3 + 0.4

α-Tocopherol 26.0 + 0.1 41.5 + 0.0
δ-Tocopherol 26.0 + 0.1 37.7 + 0.3
BHT 26.0 + 0.1 43.7 + 0.2
TBHQ 26.0 + 0.1 40.0 + 0.0
Trolox 26.0 + 0.1 28.0 + 0.1
Propyl Gallate 26.0 + 0.1 34.6 + 0.1
The metal chelator, citric acid, is an effective antioxidant in bulk oils suggesting
that metal-catalyzed decomposition of lipid hydroperoxides is an important pathway
for lipid oxidation in bulk oils. Interfacial tension data shows that lipid hydroperox-
ides, such as cumene hydroperoxide, are surface active in stripped corn oil (Fig. 9.9)
suggesting that they would likely migrate to the water-lipid interface of the associa-
tion colloids where they would interact with metals in the water or at the water-oil
interface. Free fatty acids also decrease the interfacial tension of stripped corn oil (Fig.
9.9) suggesting that they could also migrate to the water-lipid interface of the associa-
tion colloids. Concentration of free fatty acids at the water-oil interface of association
colloids would impart a negative charge to the interface that could attract prooxidant
metals, such as iron, thus increasing iron-promoted decomposition of lipid hydro-
peroxides leading to increased development of oxidative rancidity. This could help
explain why free fatty acids are prooxidative in bulk oils (Chaiyasit et al., 2007).
Phospholipids are more surface active than free fatty acids and cumene hydro-
peroxides (Fig. 9.10) suggesting that they would also concentrate in association col-
loids. In bulk oils, a unique property of phospholipids is their ability to increase the
antioxidant activity of tocopherols (Koga and Terao, 1995). Phospholipids increase
interactions between α-tocopherol and water-soluble free radicals (generated from
2,2′-azobis(2-amidinopropyl) dihydrochloride, AAPH). In addition, free radical and
α-tocopherol interactions increased as the number of association colloids produced
by phospholipids increased. Koga and Terao (1995) suggested that the increased
activity of α-tocopherol in the presence of phospholipids was due to the ability of
phospholipids to concentrate α-tocopherol at the water-lipid interface of association
colloids where oxidative stress is greatest.
Differences in the ability of surface-active compounds to decrease surface and
interfacial tension are related to their surface activity, as well as their physical structure
that allow them to pack at the interface (Cercaci et al., 2007). Increasing concentra-
tions of surface-active compounds will decrease interfacial tension until the water-oil
surfaces are completely saturated. This was seen in the stripped corn oil where concen-
trations of phosphatidylcholine up to 0.75 mmol/kg oil decreased interfacial tension
while concentrations from 0.75 to 1.0 mmol/kg oil did not further change (p>0.05)
32
Interfacial T ension (m N /m )
30
28
26
24 C um ene H y droperox ide

O leic A c id
22
0 50 100 150 200
S urface A ctive C om pound C oncentration (m m ol/kg O il)

Fig. 9.9. Influence of cumene hydroperoxide and oleic acid (0-200 mmol/kg lipid) on interfacial tension of
stripped corn oil as determined by drop shape analysis at room temperature after 20 min of equilibration.
interfacial tension (Fig. 9.10). The total concentration of surface active compounds in
commercial oils is greater than the levels needed to saturate the water-oil interface of
association colloids. This can be observed by the fact that additional water can be add-
ed to oils without the oil becoming cloudy (an indication of formation of an emulsion
instead of association colloids) or forming oil-water bilayers. Since commercial oils
probably contain excess levels of surface-active components, an additional factor that
would impact the concentration and type of surface-active compounds in association
colloids would be competition among the surface-active molecules for the water-oil
interface. Interfacial tension can be used to determine if the surface-active minor
components commonly found in commercial oils will migrate to the water-oil inter-
face. These experiments were performed by adding oleic acid (25 mmol oleic acid/kg
oil) or phosphatidylcholine (0.1 mmol phosphatidylcholine/kg oil) to stripped corn
oil followed by addition of cumene hydroperoxide over a concentration range of 0-
100 mmol/kg oil. This model was used to determine if a lipid hydroperoxide, the
least surface active of all minor components tested, would be able to migrate to the
water-oil interface in the presence of other surfactants where it could be decomposed
by aqueous phase prooxidant metals and result in accelerated lipid oxidation. In this
system, cumene hydroperoxide was able to significantly decrease (p<0.05) interfacial
tension of the water-oil bilayer in the presence of either oleic acid or phosphatidyl-
choline (Fig. 9.11) indicating that it could gain access to aqueous phase prooxidative
metals in the presence of other surface-active compounds.
The ability of surface-active minor components in oils to concentrate at the

oil-water interface could also be influenced by oil polarity. Increasing δ-tocopherol
concentrations from 25 to 75 mmol/kg oil did not significantly decrease (p>0.05)
interfacial tension in stripped corn oil suggesting that there were not enough driving
forces for δ-tocopherol to migrate to the water-oil interface (Fig. 9.12). However, the
further increases in δ-tocopherol concentrations to 100 and 200 mmol/kg oil resulted
in significant decreases (p<0.05) in interfacial tension compared to control. Com-
pared to previous findings, δ-tocopherol as low as 1 mmol/kg hexadecane was able to
decrease interfacial tension of water-hexadecane bilayer (Chaiyasit et al., 2005). The
polarity difference is greater between hexadecane and δ-tocopherol than stripped corn
oil and δ-tocopherol, thus allowing δ-tocopherol to migrate to the water-hexadecane
interface more effectively than to the water-stripped corn oil interface. This could be
another reason why the effectiveness of antioxidants varies in different food oils.
Conclusions
Most lipids in food exist as dispersions or emulsions. Lipid dispersions contain lipid-
water interfaces whose physical properties greatly impact lipid oxidation chemistry.
The physical properties that can impact lipid oxidation include surface charge that can
35
30
25
20
15
10
0
0.00 0.25 0.50 0.75 1.00
P hosphatidylcholine C oncentration (m m ol/kg O il)
Fig. 9.10. Influence of phosphatidylcholine dioleyl (0-1 mmol/kg lipid) on interfacial tension of stripped corn oil
as determined by drop shape analysis at room temperature after 20 min of equilibration.
32
28
24
20 P hosphatidy lc holine
O leic A c id
16
0 50 100 150 200
C um ene H ydroperoxide C oncentration (m m ol/kg O il)

Fig. 9.11. Influence of cumene hydroperoxide (0-200 mmol/kg lipid) on interfacial tension of stripped corn oil
containing oleic acid (25 mmol/kg lipid) or phosphatidylcholine dioleyl (0.1 Mmol/kg lipid) as determined by
drop shape analysis at room temperature after 20 min of equilibration.
32
30
28
26
0 50 100 150 200
δ-T ocopherol C oncentration (m m ol/kg O il)

Fig. 9.12. Influence of δ-tocopherol (0-200 mmol/kg lipid) on interfacial tension of stripped corn oil as deter-
mined by drop shape analysis at room temperature after 20 min of equilibration.
attract or repel prooxidant metals and interfacial thickness that can inhibit interac-
tions between aqueous phase prooxidants and lipids. Physical structures also seem to
play an important role in the oxidative stability of bulk oils. The ability of minor com-
ponents to migrate and concentrate in associated colloids in commercial oils could
help explain why free fatty acids are prooxidative, phospholipids are antioxidative,
and polar antioxidants are more effective than non-polar antioxidants. Understanding
how the physical properties of food dispersions impacts lipid oxidation could lead to
the development of novel antioxidant technologies that help improve the oxidative
stability of oils containing increased concentrations of polyunsaturated fatty acids.
Acknowledgement
This research was supported in part by grant 2007-02650 from the NRI competitive
grants program of the United States Department of Agriculture.
References
Abdalla, A. E.; and J.P. Roozen. Effect of Plant Extracts on the Oxidative Stability of Sunflower
Oil and Emulsion. Food Chem. 1999, 64, 323-329.
Cercaci, L.; M.T. Rodriguez-Estrada; G. Lercker; and E.A. Decker. Phytosterol Oxidation in
Oil-in-Water Emulsions and Bulk Oil. Food Chem. 2007, 102, 161-167.
Chaiyasit, W.; R.J. Elias; D.J. McClements; and E.A. Decker. Role of Physical Structures in
Bulk Oils on Lipid Oxidation. Crit. Rev. Food Sci. Nutr. 2007, 47, 299-317.
Chaiyasit, W.; D.J. McClements; and E.A. Decker. The Relationship Between the Physico-
chemical Properties of Antioxidants and Their Ability to Inhibit Lipid Oxidation in Bulk
Oil and Oil-in-Water Emulsions. J. Agric. Food Chem. 2005, 53, 4982-4988.
CRC Press. CRC Handbook of Chemistry and Physics. 63rd Edition. Weast, R.C.; and Astle, M.J.
Eds., CRC Press: Boca Raton, Florida, 1982, p. E54.
Dana, D.; and I.S. Saguy. Mechanism of Oil Uptake during Deep-Fat Frying and the Surfac-
tant Effect-Theory and Myth. Adv. Colloid Int. Sci. 2006, 128, 267-272.
Decker, E.A.; and D.J. McClements. Transition Metal and Hydroperoxide Interactions: An
Important Determinant in the Oxidative Stability of Lipid Dispersions. Inform. 2001,
12, 251-255.
Diaz, M.; C.M. Dunn; D.J. McClements; and E.A. Decker.. Use of Caseinophosphopeptides
as Natural Antioxidants in Oil-in-Water Emulsions. J. Agric. Food Chem. 2003, 51, 2365-
2370.
Dickinson, E. An Introduction to Food Colloids. Oxford University Press: Oxford, 1992, p. 2.
Dickinson, E.; and D.J. McClements.. Advances in Food Colloids. Blackie Academic and Pro-
fessional: Glasgow, 1995, p. 15-17.
Dickinson, E.; and G. Stainsby. Colloids in Foods. Applied Science Publishers: London, 1982,
p. 15.
Frankel, E.N. Lipid Oxidation. The Oily Press: Dundee, Scotland, 1998, pp. 166-185.
Frankel, E.N.; S.W. Huang; J. Kanner; and J.B. German. Interfacial Phenomena in the Evalu-
ation of Antioxidants-Bulk Oils vs Emulsions. J. Agric. Food Chem. 1994, 42, 1054-
1059.
Fritsch, C.W. Lipid Oxidation-The Other Dimensions. Inform. 1994, 5, 423-436.
Garti, N.; and M.E. Leser. Emulsification Properties of Hydrocolloids. Polym. Adv. Tech. 2001
12, 123-135.
Gupta, R.; H.S. Muralidhara; and H.T. Davis. Structure and Phase Behavior of Phospholipids-
Based Micelles in Nonaqueous Media. Langmuir. 2001, 17(17), 5176-5183.
Halliwell, B.; M.A. Murcia; S. Chirico; and O.I. Aruoma. Free Radicals and Antioxidants in
Food and in vivo: What They Do and How They Work. Crit. Rev. Food. Sci. Nutr. 1995
35, 7-20.
Hu, M.; D.J. McClements; and E.A. Decker. Impact of Whey Protein Emulsifiers on the
Oxidative Stability of Salmon Oil-in-Water Emulsions. J. Agric. Food Chem. 2003a, 51,
1435-1439.
Hu, M.; D.J. McClements; and E.A. Decker. Lipid Oxidation in Corn Oil-in-Water Emul-
sions Stabilized by Casein, Whey Protein Isolate, and Soy Protein Isolate. J. Agric. Food
Chem. 2003b, 51, 1696-1700.
Koga, T.; and J. Terao. Phospholipids Increase Radical-Scavenging Activity of Vitamin E in a
Bulk Oil Model System. J. Agric. Food Chem. 1995, 43, 1450-1454.
Mancuso, J.R.; D.J. McClements; and E.A. Decker. The Effects of Surfactant Type, pH, and
Chelators on the Oxidation of Salmon Oil-in-Water Emulsions. J. Agric. Food Chem.
1999, 47, 4112-4116.
McClements, D.J. Food Emulsions: Principles, Practice and Techniques, CRC Press: Boca Raton,
Florida, 1999, p. 167.
McClements, D.J. Food Emulsions: Principles, Practice and Techniques, 2nd Edition, CRC Press:
Boca Raton, Florida, 2004, p. 3.
McClements, D.J.; and E.A. Decker.. Lipid Oxidation in Oil-in-Water Emulsions: Impact of
Molecular Environment on Chemical Reactions in Heterogeneous Food Systems. J. Food
Sci. 2000, 65(8), 1270-1282.
Naz, S.; H. Sheikh; R. Siddiqi; and S.A. Sayeed. Deterioration of Olive, Corn and Soybean
Oils Due to Air, Light, Heat and Deep-Frying. Food Res. Int. 2005, 38, 127-134.
Porter; W.L. Recent Trends in Food Applications of Antioxidants. In Autoxidation in Food and
Biological Systems; Simic, M.G.; and Karel, M. Eds., Plenum Press: New York, 1980, pp.
295-365.
Silvestre, M.P.C.; W. Chaiyasit; R.G. Brannan; D.J. McClements; and E.A. Decker. Ability of
Surfactant Headgroup Size to Alter Lipid and Antioxidant. J. Agric. Food Chem. 2000,
48, 2057-2061.
Tong, L.M.; S. Sasaki; D.J. McClements; and E.A. Decker. Antioxidant Activity of Whey in a
Salmon Oil Emulsion. J. Food Sci. 2000a, 65(8), 1325-1329.
Tong, L.M.; S. Sasaki; D.J. McClements; and E.A. Decker. Mechanisms of the Antioxidant
Activity of a High Molecular Weight Fraction of Whey. J. Agric. Food Chem. 2000b, 48,
1473-1478.
10
Antioxidant Evaluation Strategies
Leif H. Skibsted
Food Chemistry, Department of Food Science, University of Copenhagen, Rolighedsvej
30, DK-1958 Frederiksberg C, Denmark
Introduction
Throughout the world societies are recognized where life expectancy is longer than
in comparable societies. Longevity has been associated with availability and choice of
fresh and good food. Certain bacteria from fermented milk having the potential to
colonize the human intestine have been suggested to be crucial for mountaineers age-
ing with no loss of physical and mental capabilities. In coastal societies a diet rich in
fish has been identified as important for the absence of certain cardiovascular diseases.
Moreover it has often been speculated whether one single dietary factor could be iden-
tified as the “secret” for human health, such as a specific lipid source, a mineral present
in the soil, or certain herbs.
Around the Mediterranean, the diet is rich in vegetables, nuts, fruits, the bread is
whole grain, and fish is preferred for meat, which is eaten in moderation. Antioxidants
has been suggested as the health factor common for citrus fruits, tomatoes, nuts, spices,
whole grain bread, and the red wine typical for the Mediterranean meal and that ac-
company fish fried in olive oil.
In relation to food and human nutrition, antioxidants were originally defined as
“substrates that in small quantities are able to prevent or greatly retard the oxidation
of easily oxidizable nutrients such as fats” (Chipault, 1962). Antioxidants could be ac-
tive at three stages in relation to human nutrition. Antioxidants can prevent oxidative
damage to food during processing, storage, and preparation of the meal; in effect limit-
ing formation of toxic oxidation products in the food eaten. Without being absorbed,
antioxidants can protect sensitive tissue in the human digestive tract against aggressive
prooxidants formed during digestion. Antioxidants may have protective effects in the
human body following absorption, and as such may be important nutrients (Vulcain et
al., 2005). In order to evaluate antioxidants it is important to recognize at what stage
antioxidative protection is needed or expected and to adjust the evaluation protocol
accordingly (Becker et al., 2004). Different protocols should be applied to evaluate
health effects of antioxidants and to evaluate protection of food against oxidative de-
terioration. A methodological shortcoming has been identified since antioxidants are
multifunctional in biological systems but the methods traditionally used for evaluation
291
292 L.H. Skibsted
are only “one-dimensional” (Frankel and Meyer, 2000). The multifunctionality of an-
tioxidants has also led to more general or broad definitions of antioxidants, including
functions as regulators of gene expression (Rice-Evans, 2004).
As for the health effects of antioxidants, epidemiological studies have suggested
beneficial effects of antioxidants especially from fruits and vegetables. The result of
human intervention studies are, however, less clear and supplementation with anti-
oxidants like vitamin E have shown little if any protective effect on cardiovascular
diseases and on cancer (Vivekananthan et al., 2003; Lonn et al., 2005). In contrast to
supplementation, a change in diet will involve a changed pattern of interaction be-
tween the numerous bioactive compounds present in plant food with the perspective
of synergistic effects between the individual antioxidants not obvious following sup-
plementation with single or a few compounds. Further development in antioxidant
evaluation methods should accordingly focus on antioxidant interaction. Synergistic
effects of antioxidants were recently demonstrated for dietary protection against age-
related macular degeneration (van Leeuwen et al., 2005).
Antioxidant protective effects have been demonstrated much more convincingly
for food quality than for human health. Based on dietary recommendations, attempts
have been made to change the lipid profile of pork and poultry and even bovine milk
to a higher degree of unsaturation by changing the feeding regime. Such products
with improved nutritive value are in general vulnerable to oxidation (Sandström et al.,
2000). However, increasing the dietary level of vitamin E for the production animal
has been found to more than compensate for the oxidative instability introduced by
the higher unsaturation in the meat products (Jensen et al., 1998).
Possible effects of non-absorbed antioxidants on human health through activity
in the digestive tract have only lately been recognized and deserve further attention
(Kanner and Lapidot, 2001). Protection by plant-based antioxidants may even be
more relevant in relation to iron-fortification of certain foods, in which the increased
availability of iron may create iron-based prooxidants during digestion.
Evaluation Protocols
A four step strategy for antioxidant evaluation was proposed (Becker et al., 2004).
This strategy was based on the assumption that lipid oxidation is initiated through
free radical mechanisms and that phenolic compounds are the main group of primary
antioxidants. The strategy included the following steps to evaluate potential sources
of antioxidants, such as protein hydrolysates and plant extracts:
i. Quantification and possible identification of phenolic compounds
ii. Quantification of radical-scavenging activity using radicals with relevant

life-time and reactivity for biological systems
iii. Evaluation of the activity to inhibit or halt lipid oxidation in model

biological systems
Antioxidant Evaluation Strategies 293
iv. Depending on the use of the antioxidants:
a. Storage studies using actual antioxidants incorporated in the food

product of relevance to evaluate food stability
b. Human intervention studies using relevant markers for oxidative status

and oxidative stress to evaluate health effects.
Step (i) is an initial screening of potential sources of antioxidants and may also be
used to optimize extraction procedures and to select extraction solvents. Step (ii)
evaluates the chain-breaking capacity of the extract recognizing that lipid oxidation
is a free radical chain reaction. In step (iii) an oxidation substrate is introduced; the
capability of the potential antioxidant to limit oxygen consumption or formation of
secondary lipid oxidation products is determined and expressed as a prolongation
of the lag-phase or a decrease in rate. Many natural compounds are scavengers of
especially short-lived free radicals without being effective at protecting lipids against
oxidation, and evaluation including only step (i) and/or step (ii), as is often seen
in literature, may give false positive results. Conclusions from step (iii) may, how-
ever, still be invalid to protect real foods or to predict health effects, and a complete
evaluation of antioxidants should always include step (iv). There is no short cut from
determining total phenolic content, determining radical scavenging activity, or even
determining effects on lipid oxidation in model systems, to real food systems. The
same reservation applies to antioxidants and health effects; again radical-scavenging
screening and determination should be followed by intervention studies prior to di-
etary recommendations. Anthocyanines for example are efficient radical scavengers
but have extremely low bioavailability from the digestive tract. There are many claims
for protective effects on vision by anthocyanines as antioxidants which could not be
confirmed in intervention studies (Canter and Ernst, 2004). Still, anthocyanines may
have positive effects as antioxidants in the digestive tract without being absorbed by
deactivating prooxidants from meat (Kanner and Lapidot, 2001). Accordingly the
proposed 4 step evaluation has to be used with care and the initial step should prima-
rily be applied to screen new sources of natural antioxidants. When the lipid system
of step (iii) is changed from a homogeneous lipid phase to heterogeneous systems of
increasing structural organization, antioxidant synergism may also be recognized dur-
ing screening (Becker et al., 2007).
Early Events in Lipid Oxidation

Oxidation is initiated by irradiation (including visible light), by enzymes, and by
metal catalysis. Heat and pressure enhance these effects. It should be recognized that
optimal antioxidant protection is obtained by preventing oxidation initiation rather
than by limiting propagation of lipid oxidation when the initial oxidative damage
has occurred. Thus for food products with long shelf-life, damage during production
leading to a high level of lipid hydroperoxides holds the potential of severe oxidative
294 L.H. Skibsted
damage during subsequent storage of the product. From Fig. 10.1, three major reac-
tion paths for lipid oxidation may be identified:
I. Oxygen activation by metal catalysis, including activation by oxidoreductases

to yield superoxide and hydroxyl radicals, like in the serum phase of milk.
The hydroxyl radicals and chlorine radicals from disinfectants will attack
unsaturated lipids and initiate chain reactions.
II. Lipoxygenase activity that incorporates oxygen in unsaturated lipids to yield

lipid hydroperoxides like in vegetables and chicken meat.
III. Photosensitized oxidation as a result of light exposure that either activates

ground state oxygen to form singlet oxygen by physical quenching of the
excited state sensitizer or forms radicals through chemical quenching of the
excited state sensitizer. Singlet oxygen may add to unsaturated lipids to form
lipid hydroperoxides as for lipoxygenase activity (II). Substrate radicals may
subsequently enter the free radical reaction sequences described for metal
catalysis (I).
Antioxidant evaluation corresponding to the initial three steps of the four step evalu-
ation protocol will have to be adjusted according to the oxidative stress expected for
the food or the biological system under consideration for protection by antioxidants.
The reactions or reaction sequences marked with capital letters in Fig. 10.1 show the
targets for antioxidant protection at early stages of lipid oxidation. Efficient radical
scavengers will prevent initiation of oxidation by radicals at A, while metal chelators
may prevent cleavage of lipid hydroperoxides to initiate further chain reactions at
B. Protection against light-induced lipid oxidation depends on interaction with the
photosensitizer or on interaction with singlet oxygen (C). Inactivation of lipoxygen-
ase, as in blanching of vegetables, will not be discussed any further (D). Initiation of
oxidation by lipoxygenases or by light results in the formation of lipid hydroperox-
ides. Metal catalysis often depends on the existence of preformed lipid hydroperox-
ides (LOOH) and has been termed “LOOH-dependent oxidation”. Differentiation
between “LOOH-independent” lipid oxidation initiated by oxygen activation and
metal catalysis:
Fe2+ + H2O2 → Fe3+ + HO• + OH- [1]

Fe3+ + H2O2 → FeO2+ + HO• + H+ [2]
and “LOOH-dependent” lipid oxidation based on initial enzymatic or photochemi-

cal generation of lipid hydroperoxides:
Fe2+ + LOOH → Fe3+ + LO• + OH- [3]

Fe3+ + LOOH → FeO2+ + LO• + H+ [4]
is usually difficult (Carlsen et al., 2005). However, heat and pressure enhances the
cleavage of peroxides to initiate new chain reactions for both types of lipid oxidation.
High-pressure processing has been introduced for new types of meat and dairy prod-
ucts for which antioxidant protection will also be required. For such products high
pressure must be considered together with heat as an enhancer of lipid oxidation in
the later steps of the four step procedure (Bragagnolo et al., 2005).
Fig. 10.1. Lipid oxidation depends on three reaction paths: (I) Free radical chain reaction initiated by oxygen
activation to yield hydroxyl radicals by radical formation; (II) Lipoxygenase activity to yield lipid hydroperoxides;
or (III) Photosensitized formation of singlet oxygen or free radicals. Areas A, B, C, and D show where protection
against early events in lipid oxidation should be targeted.
296 L.H. Skibsted
Protection Against Light

Under most conditions lipid oxidation is induced through sensitized reactions rather
than by direct absorption of light by the lipid substrate (Bekbölet, 1990). Important
photosensitizers in foods are riboflavin, chlorophylls, partly degraded heme pigments,
and certain synthetic colorants like erythrosine. For certain biological structures, like
the eye, exposure to light with a strong UV-component may involve direct excitation
of aromatic side chains in the proteins leading to cross-linking. In order to optimize
antioxidant protection of food exposed to light, it is important to distinguish between
two limiting mechanisms as shown in Fig. 10.2 for riboflavin. Photosensitizers like
riboflavin are characterized by efficient intersystem crossing to yield triplet states from
the initially populated excited singlet states. The triplet excited state may be deactivat-
ed chemically by electron transfer to an oxidation substrate or by physical quenching
by oxygen (Li et al., 2000). Protection against light-induced oxidation can be based
on one or more of the following four principles:
i. Inner-filter effects.
ii. Quenching of triplet state photosensitizer.
iii. Quenching of singlet oxygen for Type II mechanism (Fig. 10.2).
iv. Scavenging of substrate radicals formed by chemical quenching of triplet

photosensitizer by primary oxidation substrate for Type I mechanism (Fig.
10.2).
To protect by inner-filter effects, the light is absorbed by another compound than the
photosensitizer. For a dairy spread, β-carotene was found to protect lipids against oxi-
dation by riboflavin-sensitized oxidation through such an inner-filter effect, as light
was found to be absorbed preferentially by β-carotene rather than by riboflavin (Han-
sen and Skibsted, 2000). Protection by direct quenching of the triplet state photosen-
sitizer was observed for ascorbic acid, plant phenols, Trolox, and tocopherols, while
carotenoids somewhat surprising were found inactive (Jung et al., 2007; Becker et al.,
2005; Cardoso et al., 2007). In contrast, carotenoids are efficient physical quenchers
of singlet oxygen, and the most effective are the carotenoids with the longest conju-
gated systems (Bradley and Min, 1992). As for the last principle of protection, the
mechanism is similar to the mechanism for protection against thermal lipid oxidation
and the radicals formed by chemical quenching of the photosensitizer will initiate
chain reaction if not scavenged by an antioxidant (Fig. 10.1). Notably, a number
of plant phenols known to be efficient antioxidants in thermal oxidation of lipids,
have recently been found to have a dual function in light-induced lipid oxidation as
they are efficient quenchers of triplet-excited state riboflavin besides being efficient
scavengers of any radicals formed by reaction of oxidation substrates with triplet-
Fig. 10.2. Two limiting mechanisms for photosensitized initiation of lipid and protein oxidation with riboflavin
(Vitamin B2) as sensitizer: Type I involves direct radical formation through chemical quenching of triplet ribofla-
vin by a substrate, while Type II depends on physical quenching of triplet riboflavin by oxygen to yield singlet
oxygen.
298 L.H. Skibsted
state riboflavin (Becker et al., 2005). In Table 10.1 second-order rate constants for a
number of quenchers relevant to dairy products are collected together with activation
parameters. The quenching rate approaches the diffusion limit. Although the values
of the activation parameters for the quenching reaction, have a large variation, they
confirm a common deactivation mechanism since they show so-called isokinetic be-
havior (ΔH# depends linearly on ΔS#) together with purine bases. The mechanism by
which these compounds protect dairy products, beer, and other foods and beverages
against light-induced oxidation in agreement with the isokinetic behavior has been
described as a bimolecular diffusion-controlled encounter with electron (or hydrogen
atom) transfer as the rate-determining step.
Table 10.1. Second-Order Rate Constants at 25°C for Quenching of Triplet-Excited State Riboflavin by
Compounds of Interest for Protection of Food Against Light-Induced Oxidation.a
# #
Quencher Solvent k2 (l × mol-1 × s-1) ΔH (kJ × mol-1) ΔS (J × mol-1 × K-1)
Ascorbate Water, pH=6.4 2.0 × 109 184 551
Trolox Water, pH=6.4 2.6 × 109 14 -19
α-Tocopherol Tween-20 Emulsion 1.1 × 108 43 53
Caffeic Acid Acetonitrile/Water 2.2 × 109 27 66
Rutin Acetonitrile/Water 1.0 × 109
(+)-Catechin Acetonitrile/Water 1.4 × 109
a
From Becker et al., 2005, Cardoso et al., 2006, and Cardoso et al., 2007.
To optimize protection against light-induced oxidation in food and beverages,

the four step strategy originally formulated as a general procedure for antioxidant
protection should be modified:
i. Analysis should also include other types of compounds, like terpenes and
carotenoids, in the plant material or extract which could serve as inner-
filters or singlet oxygen quenchers
ii. Radical-scavenging experiments should be replaced by quenching studies

of the actual photosensitizer present in the product in combination with
protective quenchers. Such quenching studies could be based on dynamic,
real-time methods using laser flash photolysis or on static methods based on
the Stern-Volmer formalism (Cardoso et al., 2006).
iii. Experiments in which rates for oxygen consumption or formation of

secondary lipid oxidation products are determined for a specific oxidation
substrate should be replaced by photochemical experiments in which
quantum yields, based on chemical actinometry, are determined for relevant
wavelengths for the biological tissue or for the beverage or food product
(Hansen and Skibsted, 2000).
iv. For the actual product formulated according to the results obtained in the
previous steps, storage experiments with controlled light exposure seem
mandatory. Effects on product quality should be evaluated by chemical
analysis or by sensory evaluation. A combination of the wavelength
dependence of quantum yield for the oxidative process, the spectral
characteristics of the light source and the absorption spectrum of the
photosensitizer in the product may together provide a so-called action
spectrum.
For milk-based beverages, the finding that plant phenols are efficient quenchers (in
step iii) have led to the suggestion that plant extracts should be explored to pro-
tect fruit flavored products, and tested in step (iv) (Becker et al., 2005). Riboflavin
quenchers have been found to inhibit lightstruck flavor formation in beer (Goldsmith
et al., 2005). However terpenes, like eucalyptol, present in some spices used as flavors
in beer have not been found to quench triplet riboflavin and will accordingly not yield
any direct protection (Cardoso et al., 2006).
In conclusion, it is important to know whether photoxidation is occurring by a
Type I or a Type II mechanism or by both mechanisms at the same time in competi-
tion (Jung et al., 2007). Besides inner-filter protection, carotenoids will thus yield
protection against photosensitization by the Type II mechanism but not against pho-
tosensitization by the Type I mechanism. In milk products, physical quenching of
triplet riboflavin by oxygen to yield singlet oxygen will not compete efficiently with
chemical quenching by peptides and free amino acids even in air-saturated milk (Car-
doso et al., 2004). Accordingly peptides and amino acids will be oxidized by a Type
I mechanism. In milk with high uric acid content, uric acid will deactivate triplet ri-
boflavin in competition with the peptides and amino acids, in effect yielding optimal
protection of milk against light. The uric acid level in milk depends on the feeding
regime of the dairy cows.
Protection Against Metal Catalysis

In food and other biological systems, transition metal ions are chelated by various
ligands. The standard reduction potential valid for the iron couple
Fe(H2O)63+ + e- → Fe(H2O)62+ [5]
is in the presence of flavonoids, like quercetin, which bind iron(II) more strongly than
iron(III), higher than Eө = 0.77 V, which is the value for the hexaqua ions (Ferrali et
al., 1997). Thus Iron(II) becomes less reducing, and quercetin may prevent the Fen-
ton reaction (Eqn. 1) as an initiator of lipid oxidation. As for the general antioxidant
evaluation procedure, it should be realized that when lipid oxidation is expected to be
metal-ion dependent, as in precooked meat and other processed foods, antioxidants
should be tested as scavengers of metal-related prooxidants rather than of semi-stable
300 L.H. Skibsted
radicals like DPPH• or others frequently used in standard tests. Heme pigments are
a source of iron in meat and prooxidative activity has been related to reaction cycles
similar to the pseudoperoxidase cycle shown in Fig. 10.3. Deactivation of the hyper-
valent iron pigment perferryl and ferryl by potential antioxidants is relevant (Ander-
sen et al., 2003). For an increasing number of plant phenols from sources like green
tea, fruits and vegetables, scavenging rate constants have been determined. In Table
10.2, a few examples are compared with rate constants for deactivation of the super-
oxide anion radical and lipid peroxyl radicals. Notably, phenols are rather effective
deactivating hypervalent iron compared to vitamin antioxidants, like ascorbate, and
plant phenols are also effective in preventing simple iron ions from becoming prooxi-
dative. Deactivation of hypervalent iron by plant phenols has been suggested as an
important mechanism for protection of sensitive tissue in the digestive tract against
oxidative stress (Kanner and Lapidot, 2001; Vulcain et al., 2005). There is a clear
need for standard procedures to evaluate the efficiency of plant extracts as protectors
of metal catalysis in lipid oxidation, including both heme pigments, partly degraded
heme pigments, and the simpler “free” iron ions. Attempts have been made to define
such evaluation methods, but they still depend on relatively advanced instrumenta-
tion (Carlsen et al., 2003).
Fig. 10.3. Heme pigment may act as prooxidants through activation by hydrogen peroxide (1) or Lipid Hydroper-
oxides. The hypervalent metmyoglobin (or hemoglobin) species will initiate lipid or protein oxidation as shown
in step 2 and step 3.
Table 10.2. Second-Order Rate Constants at 25°C for Scavenging of Prooxidants by Compounds of Interest for
Protection of Food Against Metal-Catalyzed Lipid Oxidation.a
Scavenger kferryl (l × mol-1 × s-1)b kinh (l × mol-1 × s-1)c kOO- (l × mol-1 × s-1)b
Ascorbate 16
Trolox 20 5.8 × 103
α-Tocopherol 3.5 × 106
Caffeic Acid 65
Rutin 23 5.1 × 104
Catechin/Epicatechin 20 ~ 4 × 10 5
~ 2 × 104
a
From Andersen et al., 2003. bIn water. c n t-butyl alcohol, kinh is the effective inhibitory rate constant
for deactivating lipid peroxyl radicals.
Protection Against Heat and Pressure

Most methods currently in use to determine antioxidative capacity of plant mate-
rial, including food products, are based on combinations of phenol analysis and as-
says based on scavenging of stable or short-lived radicals (Becker et al., 2004). Such
methods relate to the oxidative stress introduced in food systems or in the human
body by normal thermal processes (in contrast to photochemical reactions). As for
methods based on scavenging stable radicals like ABST•+, DPPH•, Fremy’s salt, and
the galvinoxyl radical, a relevant criticism is that none of these radicals are of biologi-
cal importance or present in any food systems. For short-lived radicals protection of
specific probes against free radicals, like in the ORAC assay, does not mimic biologi-
cal systems. Attempts have been made to use more moderate reactive radicals like the
1-hydroxyethyl radical in combination with the ESR spin-trapping technique for an-
tioxidant evaluation. Promising results were obtained, since such methodologies pro-
vide a detailed picture of the balance between prooxidative and antioxidative effects,
which normally not are available from the standard methods (Rødtjer et al., 2006).
The general four step procedure described previously, however, does work for
thermally processed food and is illustrated by results from a recent study (Racanicci
et al., 2007). Chicken meat is becoming increasingly important world-wide and in
countries like Brazil, production is increasing rapidly. Among different waste materi-
als from local vegetable production in Brazil, mate (dried leaves of Ilex paragúariensis)
was selected as a potential source for antioxidants to pre-cooked chicken-meat prod-
ucts. In step (i) of the four step procedure, extracts were made with water, methanol,
ethanol, and 70% aqueous acetone. Water (and 70% aqueous acetone) was superior
as an extraction solvent compared to methanol and ethanol, and in step (ii) the radi-
cal-scavenging capacity of the aqueous extract showed that each phenolic equivalent
corresponded to one mole of Fremy’s salt. Fremy´s salt scavenging method accord-
ingly is similar to a titration of the plant phenol. In step (iii), heme pigment initia-
tion of lipid oxidation was used in order to mimic the conditions in the pre-cooked
meat. Aqueous mate extract was found to be efficient in halting lipid oxidation, and
302 L.H. Skibsted
a dose/respond curve could be constructed. In the final testing step (iv), mate aque-
ous extract was added to chicken meat balls prior to cooking; in a comparison with
an unprotected product and meat balls protected by rosemary as reference, mate was
found to give equal or better protection than rosemary. Notably, both mate and rose-
mary were found to protect vitamin E in the product during storage. The four step
procedure accordingly leads directly to practical application of a new herbal source for
antioxidants to meat products, including optimization of extraction and adjustment
of dose. The evaluation procedure described is recommended for other sensitive dairy
and meat products in combination with local plant sources with a GRAS-status.
Less investigated is antioxidative protection of high-pressure processed food
products, a protection which is clearly needed. Pressure-induced lipid oxidation in
muscle systems, has been assigned to two main factors, iron released from heme pro-
teins, or membrane disruption. However, iron release could not be detected following
high-pressure processing of chicken meat for thermal processes (Orlien et al., 2000).
Accordingly protection should be targeted towards enzymatic formation of radicals
in disrupted membranes, and rosemary extract was found to scavenge such radicals
efficiently (Bragagnolo et al., 2005). Pressure effects on lipid oxidation have been
expressed as volume of activation analogously with energy of activation for thermal
process (Orlien et al., 2000). However evaluation procedures to protect high-pressure
processed food needs further investigation related to effects of antioxidants on volume
of activation in order to use this parameter in practical work. New imaging techniques
based on ESR-spectroscopy should also be developed in order to locate oxidation
initiation related to design of optimal protection.
Kinetic versus Thermodynamic Control

Antioxidant synergism is getting increasing attention, and synergism is a result of
kinetic control of antioxidant interaction with respect to primary antioxidants. In
homogeneous solution, antioxidants are expected to regenerate each other according
to the ordering of standard reduction potentials of the corresponding free radicals;
accordingly ascorbate should regenerate plant phenols, tocopherols, and other anti-
oxidants as the most reducing in agreement with thermodynamic product control.
Ascorbate has thus been shown to regenerate tocopherols from their one-electron
oxidized forms in a process of physiological importance in agreement with the predic-
tions from thermodynamics. In lipid systems of increasing structural organization,
synergism between carotenoids and tocopherols were clearly demonstrated indicat-
ing kinetic control of the antioxidant interaction (Becker et al., 2007). For the four
groups of antioxidants shown in Table 10.3, six possible types of interaction are pos-
sible and hold the potential of synergistic effects. Antioxidant synergism becomes
possible when a better antioxidant, like the tocopherols, is regenerated by flavonoids
or other plant phenols. In meat systems, plant antioxidants have thus been found to
protect vitamin E, probably through a regeneration mechanism, indicating kinetic
control related to phase separation of the antioxidants (Racanicci et al., 2007). Syner-
gism is also possible in systems where the less efficient antioxidants, such as flavonoids
or carotenoids, are sacrificially oxidized, in effect protecting the better antioxidant like
tocopherols and again constitutes an example of kinetic control. The barely studied
interaction between carotenoids and plant phenols seems of importance for synergis-
tic interaction and should be further investigated (Han et al., 2007).
Table 10.3. Four Groups of Antioxidants

Hydrophilic Lipophilic
Vitamins Ascorbic Acid Tocopherols
Non-Vitamins Polyphenols Carotenoids
In homogeneous solution, the distribution between lipid peroxyl radicals and

carbon-centered lipid radicals may be important for antioxidant interaction. Tocoph-
erols will be used up by the reaction of the tocopheryl radical with peroxyl radicals
under high oxygen conditions, while flavonoids and other plant phenols may regen-
erate tocopherol in competition with a less efficient reaction between the tocopheryl
radical and carbon-centered radicals during oxygen depletion.
A comparison between high ascorbic blackcurrant juice and high phenolic black
chokeberry in relation to antioxidant effects in combination with α-tocopherol, a
very interesting observation has been made. The chokeberry juice was, in contrast
to the blackcurrant juice, found to protect vitamin E efficiently against oxidation.
This must clearly be an example of kinetic control, since ascorbic acid is the more
reducing antioxidant compared to both the tocopherol and especially the flavonoids
(Graversen et al., 2007).
Conclusions
Lipid oxidation is often investigated in food systems without considering oxidation of
other components. However protein oxidation is getting increased attention in rela-
tion to food quality and to optimal biological function. In meat, protein oxidation is
known to decrease eating quality by reducing tenderness and juiciness and by enhancing
discoloration and flavor deterioration (Xiong, 2000). Evaluation protocols should be
established for proteins similar for those being used for lipids. Such antioxidant evalu-
ation should also include the effect on a possible interaction between lipid and protein
oxidation. Some methods adapted for evaluation against light-induced oxidation in
foods is already being used for proteins like in milk products (Cardoso et al., 2004). For
meat proteins, oxidation reactions are leading to cross-linking and carbonyl formation.
Antioxidants like rosemary, very effective in inhibiting lipid oxidation in meat, seem to
have little if any effect on protein oxidation (Lund et al., 2007). Protein and lipid oxida-
tion may be uncoupled in such products and design of dual acting antioxidant systems
present a real challenge. In tissue, further complications to protect proteins with antioxi-
dants will arise, since oxidized lipids may modify proteins through coupling reactions
between lipid-derived aldehydes and protein amino side-chain groups.
304 L.H. Skibsted
References
Andersen, M.L.; R.K. Lauridsen; and L.H. Skibsted. Optimizing the Use of Phenolic Com-
pounds in Food. In Phytochemical Functional Foods, Johnson, I.; and Williamson, G. eds.,
Woodhead: CRC, Cambridge, 2003, pp. 315-346.
Becker, E.M.; D.R. Cardoso; and L.H. Skibsted. Deactivation of Riboflavin Triplet-Exited
State by Phenolic Antioxidants: Mechanism Behind Protective Effects in Photooxidation
of Milk-Based Beverages. Eur. Food Res. Tech. 2005, 221, 382-386.
Becker, E.M.; L.R. Nissen; and L.H. Skibsted. Antioxidant Evaluation Protocols: Food
Quality or Health Effects. Eur. Food Res. Technol. 2004, 219, 561-571.
Becker, E.M.; G. Ntouma; and L.H. Skibsted. Synergism and Antagonism Between Quer-
cetin and Other Chain-Breaking Antioxidants in Lipid Systems of Increasing Struc-
tural Organisation. Food Chem. 2007, 103, 1288-1296.
Bekbölet, M. Light Effects on Food. J. Food Prot. 1990, 53, 430-440.
Bradley, D.G.; and D.B. Min. Singlet Oxygen Oxidation of Foods. Crit. Rev. Food Sci. Nutr.
1992, 31, 211-236.
Bragagnolo, N.; B. Danielsen; and L.H. Skibsted. Effect of Rosemary on Lipid Oxidation in
Pressure-Processed, Minced Chicken Breast during Refrigerated Storage and Subsequent
Heat Treatment. Eur. Food Res. Tech. 2005, 221, 610-615.
Cardoso, D.R.; D.W. Franco; K. Olsen; M.L. Andersen; and L.H. Skibsted. Reactivity of Bo-
vine Whey Proteins, Peptides and Amino Acids Toward Triplet Riboflavin as Studied by
Laser Flash Photolysis. J. Agric. Food Chem. 2004, 52, 6602-6606.
Cardoso, D.R.; K. Olsen; J.K.S. Møller; and L.H. Skibsted. Phenol and Terpene Quenching of
Singlet and Triplet Excited States of Riboflavin in Relation to Light-Struck Flavor Forma-
tion in Beer. J. Agric. Food Chem. 2006, 54, 5630-5636, eratum 54, 9278.
Cardoso, D.R.; K. Olsen; and L.H. Skibsted. Mechanism of Deactivation of Triplet-Excited
Riboflavin by Ascorbate, Carotenoids and Tocopherols in Homogeneous and Heteroge-
neous Aqueous Food Model System. J. Agric. Food Chem. 2007, 55, 6285-6291.
Carlsen, C.U.; J.K.S. Møller; and L.H. Skibsted. Heme Iron in Lipid Oxidation. Coord. Chem.
Rev. 2005, 249, 485-498.
Carlsen, C.U.; I.M. Skovgaard; and L.H. Skibsted. Pseudoperoxidase Activity of Myoglobin.
Kinetics and Mechanism of the Peroxidase Cycle of Myoglobin with H2O2 and 2,2-Azi-
no-bis(3-Ethylbenzthiazoline-6-Sulfonate) as Substrates. J. Agric. Food Chem. 2003, 51,
5815-5823.
Carter, P.H.; and E. Ernst. Anthocyanosides of Vaccinium myrtillus (Bilberry) for Night Vi-
sion–A Systematic Review of Placebo-Controlled Trials. Surv. Ophthalmol. 2004, 49, 38-
50.
Chipault, J.R. Antioxidants for Food Use. In Autoxidation and Antioxidants, Lundberg, W.O.
Ed., Wiley: New York, 1962, pp. 477-542.
Ferrali, M.; C. Signorini; N. Caciotti; L. Sugherini; L. Cicoli; D. Giachetti; and M. Comporti.
Protection Against Oxidative Damage of Erythrocyte Membrane by the Flavonoid Quer-
cetin and Its Relation to Iron Chelating Activity. FEBS Letters 1997, 416, 123-129.
Frankel, E.N.; and A.S. Meyer. The Problem of Using One-Dimensional Methods to Evaluate
Multifunctional Food and Biological Antioxidants. J. Sci. Food Agricult. 2000, 80(13),
1925-1941.
Goldsmith, M.R.; P.J. Rogers; N.M. Cabral; K.P. Ghiggens; and F.A. Roddick. Riboflavin
Triplet Quenchers Inhibit Lightstruck Flavor Formation in Beer. J. Am. Soc. Brew. Chem.
2005, 63, 177-184.

Graversen, H.B.; E.M. Becker; L.H. Skibsted; and M.L. Andersen. Antioxidant Synergism
Between Fruit Juice and α-Tocopherol. A Comparison Between High Phenolic Black
Chokeberry (Aronia melanocarpa) and High Ascorbic Blackcurrant (Ribes nigrum). Eur.
Food Res. Technol. 2008, 226, 737-743.
Han, R-M.; Y-X. Tian; E.M. Becker; M.L. Andersen; J-P. Zhang; and L.H. Skibsted. Puerarin
and Conjugate Bases as Radical Scavengers and Antioxidants. Molecular Mechanism and
Synergism with β-Carotene. J. Agric. Food Chem. 2007, 55, 2384-2391.
Hansen, E.; and L.H. Skibsted. Light Induced Oxidative Changes in a Model Dairy Spread.
Wavelength Dependence of Quantum Yields. J. Agric. Food Chem. 2000, 48, 3090-
3094.
Jensen, C.; M. Flensted-Jensen; L.H. Skibsted; and G. Bertelsen. Effects of Dietary Rape Seed
Oil, Copper(II) Sulphate and Vitamin E on Drip Loss, Colour and Lipid Oxidation of
Chilled Pork Chops Packed in Atmospheric Air or in a High Oxygen Atmosphere. Meat
Sci. 1998, 50, 211-221.
Jung, M.Y.; Y.S. Oh; D.K. Kim; H.J. Kim; D.B. Min. Photoinduced Generation of 2,3-Bu-
tanedione from Riboflavin. J. Agric. Food Chem. 2007, 55, 170-174.
Kanner, J.; and T. Lapidot. The Stomach as a Bioreactor: Dietary Lipid Peroxidation in the
Gastric Fluid and the Effect of Plant-Derived Antioxidants. Free Rad. Biol. Med. 2001,
31, 1388-1395.
Li, T.; J.M. King; and D.B. Min. Quenching Mechanism and Kinetics of Carotenoids in Ri-
boflavin Photosensitized Singlet Oxidation of Vitamin D2. J. Food Biochem. 2000, 24,
477-492.
Lonn, E.; J. Bosch; S. Yusuf; P. Sheridan; J. Pogue; J.M.O. Arnold; C. Ross; A. Arnold; P.
Sleight; J. Probstfield; et al. Effects of long-term vitamin E supplementation on cardio-
vascular events and cancer: a randomized controlled trial. JAMA 2005, 293(11), 1338-
1347.
Lund, M.L.; M.S. Hviid; and L.H. Skibsted. The Combined Effect of Antioxidants and Modi-
fied Atmosphere Packaging on Protein and Lipid Oxidation in Beef Patties during Chill
Storage. Meat Sci. 2007, 76, 226-233.
Orlien, V.; E. Hansen; and L.H. Skibsted. Lipid Oxidation in High-Pressure Processed Chick-
en Breast Muscle during Chill Storage. Critical Working Pressure in Relation to Oxida-
tion Mechanism. Eur. Food Res. Technol. 2000, 211, 99-104.
Racanicci, A. M.C.; B. Danielsen; and L.H. Skibsted. Mate (Ilex paraguariensis) as a Source
of Water Extractable Antioxidant for Use in Chicken Meat. Eur. Food Res. Technol. 2007,
accepted for publication.
Rice-Evans, C. Flavonoids and Isoflavonoids: Absorption, Metabolism, and Bioactivity. Free
Rad. Biol. Med. 2004, 36, 827-828.
Rødtjer, A.; L.H. Skibsted; and M.L. Andersen. Antioxidative and Prooxidative Effects of Ex-
tracts Made from Cherry Liqueour Promace. Food Chem. 2006, 99, 6-14.
Sandström, B.; S. Bügel; C. Lauridsen; F. Nielsen; C. Jensen; and L.H. Skibsted. Cholesterol
Lowering Potential in Man of Fat from Pigs Fed Rapeseed Oil. Brit. J. Nutr. 2000, 84,
143-150.
van Leeuwen, R.; S. Boekhoorn;, J.R. Vingerling; J.C.M. Witteman; C.C.W. Klaver; A.
Hofman; and P.T.V.M. de Jong. Dietary Intake of Antioxidants and Risk of Age-Related
Macular Degeneration. JAMA 2005, 294(24), 3101-3107.
Vivekananthan, D.P.; M.S. Penn; S.K. Sapp; A. Hsu; and E.J. Topol. Use of Antioxidant Vita-
306 L.H. Skibsted
mins for the Prevention of Cardiovascular Disease: Meta-Analysis of Randomised Trials.

Lancet 2003, 361, 2017-2030.
Vulcain, E.; P. Goupe; C. Caris-Veyrat; and O. Dangles. Inhibition of the Metmyoglobin-In-
duced Peroxidation of Linoleic Acid by Dietary Antioxidants: Action in the Aqueous vs.
Lipid Phase. Free Rad. Res. 2005, 39, 547-563.
Xiong, Y.L. Protein Oxidation and Implications for Muscle Food Quality. In Antioxidants in
Muscle Foods, Decker, E.; and Faustman, E. Eds., Chichester, John Wiky and Sons, 2000,
pp. 85-111.
Index
A lipid oxidation and, 52–53
Acrolein/crotonaldehyde, co-oxidation of oil-in-water emulsions, 279–281
proteins and, 204–206 tocopherol concentration
Adiposity, CLA isomer effects on, 77 antioxidant mechanism of
Age pigment formation, co-oxidation of α-tocopherol, 128–129
proteins and, 233–240 effect of temperature, 138
Aldehydes overview, 127–128
and oxidation of EPA/DHA, 55 pro-oxidant mechanism of
ozone reaction and, 44 α-tocopherol, 129, 131–138
ozone/lipid reaction and, 46 Arachidonic acid, and oxidative stability
α-linolenic acid, and oxidation of EPA/ of PUFA, 52–53
DHA, 53–54 Aromatic compounds, Type II reaction
Alkanals, CLA oxidation and, 99 of Sensitizer* and, 8
Alkoxyl radicals, 35–36 Ascorbic acids, as oxygen quenching
Amino acid losses, co-oxidation of mechanisms, 21–22
proteins and, 212–220 Atherosclerosis, CLA isomer effects on,
Antioxidants. See also Carotenoids 77
and autoxidation of CLA studies, α-tocopherol. See also Tocopherols
85–86 antioxidant mechanism of, 128–129
carotenoids, 158–159 chemical structures of, 131
and comparison of oxidation rates of effect of temperature on antioxidant
CLA/LA isomers, 89 activity, 138
evaluation strategies pro-oxidant mechanism
early lipid oxidation events, autoinitiation from hydroperoxide
293–295 decomposition, 134
heat/pressure protection, 301–302 chain transfer reactions by radical,
kinetic versus thermodynamic 132, 134
control, 302–303 pro-oxidant effects due to products,
light protection, 296–299 134–138
metal catalysis protection, 299–301 Yanishlieva-Marinova model,
overview, 291–292, 303 131–132
protocols, 292–293 rates constants of autoxidation
and levels of EPA/DHA in marine reactions, 130
animal tissue, 56 Autoxidation
307
308 A. Kamal-Eldin and D.B. Min
conjugated linoleic acid (CLA) peroxyl radical reactions

furan FA/secondary product antioxidant activity, 158–159
formation, 97, 99–100 enzyme-catalyzed co-oxidation
hydroperoxide formation, 89–93 with UFAs, 161–162
oligomer formation, 93–98 kinetics/mechanisms, 155,
rate/routes, 88–89 157–158
studies on CLA as free acid or oxidation products, 159–161
ester, 78–87 physical quenching of
rates constants involving LH/TH, photosensitizers/singlet oxygen by,
130 152–153
Azide, as quencher of singlet oxygen pro-oxidant effects
reaction, 10 effect of heat, 166–169
triplet molecular oxygen reactions,
B 163–165
β-carotene (CarH). See also Carotenoids as quencher of singlet oxygen
hydroxyl radicals and, 33–34 reaction, 10
peroxyl radicals and, 38–40 reactivity
structure/nomenclature of, 144–145 configuration of cyclic ends,
Bifunctional saturated aldehydes, co- 147–148
oxidation of proteins and, 199–202 degree of conjugation, 145–146
Biological function alteration/loss, geometrical isomerism, 147
co-oxidation of proteins and, 241, oxygenated groups at cyclic ends,
243–244 148–149
Bis-allylic hydrogens, and oxidation of structure/nomenclature of, 144–145
PUFA, 62, 70 and synergism with tocopherols in
Bonito oil, lipids levels and, 56, 58 lipid oxidation reactions, 168, 170
Bulk oils. See Oil-in-water emulsions Cellular lipids, oxidation of long-chain
PUFA in, 63–65, 70–71
Chemical quenching. See Oxygen
C quenching mechanisms
Campesterol, as sterol oxidation Chemical quenching of photosensitizers/
product, 113 singlet oxygen by carotenoids,
Cancer, CLA isomer effects on, 77 153–156
Carbonyl oxidation, 156 Chemical traps, singlet oxygen and,
Carotenes 9–10
hydroxyl radicals and, 33–34 Chemiluminence, singlet oxygen and,
peroxyl radicals and, 38–40 12
singlet oxygen quenching Cholesterol
mechanisms and, 23 oxidation
Carotenoids kinetics of sterol autoxidation,
chemical quenching of 119–122
photosensitizers/singlet oxygen by, products/mechanisms of sterol
153–156 autoxidation, 114–118
light absorption, 149–150 products/mechanisms of sterol
oxidation reactions, 143–144, 170 photoxidation, 118–119
as oxygen quenching mechanisms, sterol/stanol structures, 111–114
21–24 singlet/triplet oxygen oxidation and,
Lipid Oxidation Pathways Volume 2 309
10–11 conformation/aggregation, 220,

Cholesterol oxidation products (COP), 222
111–114 crosslinking, 222–230. See also
Conjugated DHA (CDHA), and Crosslinking of proteins
formation of conjugated trienoic acids enzyme activity inhibition,
in oils with PUFA, 69, 71 240–243
Conjugated EPA (CEPA) fluorescent adduct/age pigment
and formation of conjugated trienoic formation, 233–240
acids in oils with PUFA, 69, 71 fragmentation, 231–233
Conjugated linoleic acid (CLA) free radical transfer from lipids to
autoxidation proteins, 220–221
furan FA/secondary product secondary product reactions
formation, 97, 99–100 4-hydroxy-2-alkenals/4-oxo-2-
hydroperoxide formation, 89–93 alkenals, 206–209
oligomer formation, 93–98 acrolein/crotonaldehyde, 204–206
overview, 102 bifunctional saturated aldehydes,
rate/routes, 88–89 199–202
studies on CLA as free acid or free radical oxidation of
ester, 78–87 unsaturated aldehydes, 210
and formation of conjugated trienoic monofunctional alkanals, 197–199
acids in oils with PUFA, 65–66, 71 physiological γ-ketoaldehydes/
oxidation levuglandins, 209–211
overview, 77–78, 102, 105 unsaturated aldehydes, 202–204
singlet oxygen oxidation transfer of lipid radicals to proteins
primary product formation, consequences, 188–189
100–104 evidence, 184–188
secondary product formation, Crosslinking of proteins
104–105 carbonyl-mediated
Conjugated linolenic acid (CLN), and Michael addition of thiols/amines,
formation of conjugated trienoic acids 226–227, 229–230
in oils with PUFA, 65–69, 71 pyridinium crosslink formation,
Conjugation and carotenoids, 145–146 229–230
Co-oxidation of carotenoids, 161–162 pyrrole link formation, 227–229
Co-oxidation of proteins Schiff base amine condensation,
lipid epoxide reactions, 192–197 225, 227, 229–230
lipid hydroxyperoxide reactions, effects of, 222–223
189–192 free radical crosslinking
macromolecular damage overview, (polymerization), 223–225
181–183 Crude oils, and formation of conjugated
molecular damage reactions trienoic acids in oils with PUFA, 65
abnormal function/
immunochemistry/disease D
contributions, 244–248 DABCO, as quencher of singlet oxygen
amino acid losses, 212–220 reaction, 10
biological function alteration/loss, Deuterium oxide, singlet oxygen
241, 243–244 detection and, 9
change: solubility/hydrophobicity/
DHA F
conjugated DHA (CDHA), 69, 71 Fatty acids
oxidation overview, 51–53, 69–71 CLA oxidation
oxidative stability in liposomes, furan FA/secondary product
61–63 formation, 97, 99–100, 104–105
oxidative stability in phospholipids reaction with singlet oxygen, 13–17
of, 55–60 Fenton reaction, hydroxyl radicals and,
and oxidative stability of PUFA, 32–33
52–55 Fish oil. See also Long-chain
structures of, 53–54 polyunsaturated fatty acids (PUFA)
Dioxetanes as product of CLA and oxidative stability of EPA/DHA,
oxidation, 101, 103 55–60
Dioxines as product of CLA oxidation, triacylglycerol (TAG) as main lipid
101, 103 class of, 58
Disease contributions, co-oxidation of Flavor
proteins and, 244–248 PUFAs and, 51
Dismutation, hydrogen peroxide and, reversion flavor, 20–21, 68, 71
43–44 Flavor stability, effects of light on, 5–6
Doering’s diradical, 163–165 Fluorescent adduct/age pigment
Double bonds, singlet oxygen formation
quenching mechanism rates, 24 alkenals and, 239
DPBF (1,3-diphenylisobenzofuran), bifunctional aldehydes and, 237–238
singlet oxygen and, 9 lipid hydroxyperoxide and, 239–240
monofunctional aldehydes/alkanals
E and, 235, 237
Emulsions. See Oil-in-water emulsions Food dispersions
Endoperoxide oxidation, 156 oil-in-water emulsions
Energy levels of triplet/singlet oxygen, 4 bulk oils, 280, 282–286
Enzyme activity inhibition, co-oxidation emulsifier roles, 276–279
of proteins and, 240–243 free radical scavenging antioxidant
Enzyme-catalyzed co-oxidation with impact, 279–281
UFAs, carotenoids and, 161–162 lipid oxidation, 274–275
EPA overview, 273–274, 286–288
conjugated EPA (CEPA), 69, 71 protein impact, 275–276
oxidation overview, 51–53, 69–71 Fragmentation of proteins, 231–233
oxidative stability in phospholipids Free acids
of, 55–60 studies on CLA as ester or, 78–87
and oxidative stability of PUFA, Free fatty acids
52–55 marine animal tissue and, 59
structures of, 53–54 Free radicals. See also Long-chain
Epoxide reactions, co-oxidation of polyunsaturated fatty acids (PUFA)
proteins and, 192–197 and antioxidant mechanism of
ESR (electron spin resonance) α-tocopherol, 128–129
spectroscopy, singlet oxygen and, and autoxidation of CLA, 92
12–14 CLA isomer effects on, 77
Esters, studies on CLA as free acid or, ESR spectroscopy and, 12–14
78–87 oil-in-water emulsions/food
dispersions and, 279–281 189–192

and oxidation of unsaturated as product of CLA oxidation, 101,
aldehydes, 210 103–104
singlet oxygen oxidation and, 13–14 PV measurements and CLA
sterol autoxidation and, 121 autoxidation, 78
and transfer from lipids to proteins, singlet/triplet oxygen oxidation and,
220–221 16–17
Furan fatty acid formation, CLA Hydroperoxy epidioxides, and oxidation
oxidation and, 97, 99–100, 104–105 of PUFA, 53–54
Hydroperoxyl radicals, 35, 37
G Hydrophilic antioxidants, 303
Gas chromatography (GC), and Hydrophobicity/conformation/
oxidation of EPA/DHA, 55 aggregation/solubility
γ-ketoaldehydes/levuglandins co-oxidation of proteins and, 220,
co-oxidation of proteins and, 222
209–211 4-hydroxy-2-alkenals/4-oxo-2-alkenals,
co-oxidation of proteins and, 206–209
Hydroxyl radical scavenger, 10
H Hydroxyl radicals, chemistry/reactions
Haber-Weiss reaction of, 31–34
hydroxyl radicals and, 33
singlet oxygen formation and, 4–5
superoxide anion radicals and, 43 I
Half-lives of reactive oxygen species Immunochemistry/disease
(ROS), 31 contributions, co-oxidation of proteins
Hammond postulate, and and, 244–248
diastereoselectivity of CLA Insulin resistance, CLA isomer effects
autoxidation, 92 on, 77
Herring roe lipids, 59–60 Isomers, geometrical, 147
Histidine, as quencher of singlet oxygen Isotopes, deuterium oxide effect and, 9
reaction, 10
Hund’s rule, and elecron configuration J
of triplet/singlet oxygen, 1, 3 Jablonski diagram, 5–6
Hydrocarbons, and oxidation of EPA/ Japan, and CLN content in soybean oil,
DHA, 55 66–67
Hydrogen peroxide
chemistry/reactions of, 43–44 K
ozone/lipid reaction and, 44, 46 Ketoaldehydes/levuglandins, co-
Hydroperoxides (ROOH) oxidation of proteins and, 209–211
alkoxyl radicals formation by, 35 Kinetic control of reactions, 302–303
α-tocopherol antioxidant activity
and, 131–132
autoinitiation from decomposition L
of, 134 Laser deflection calorimetry, singlet
autoxidation of CLA methyl ester oxygen detection and, 13
and, 89–93 Levuglandins, co-oxidation of proteins
co-oxidation of proteins and, and, 209–211
Lifetime of singlet oxygen, 3 oxidation, 52–55

Light absorption of carotenoids, oxidation overview, 51–53, 69–71
149–155 oxidative stability in
Linoleate, singlet/triplet oxygen phospholipids, 55–60
oxidation rates and, 17 formation of conjugated trienoic
Linoleic acid acids in oils containing, 65–69
hydroxyl radical initiated oxidation oxidation in cellular lipids of, 63–65
of, 33–34
and oxidation rate of CLA isomers, M
88–89 Methyl esters, and autoxidation of CLA,
PUFAs of, 51 89–93
and reversion flavor in soybean oil, Methyl linoleate oxidation, 132–133
21 MHP (monohydroperoxide) isomers
singlet oxygen and formation of and oxidation of PUFA esters, 62–63
2-pentyl furan, 18 and oxidation of PUFA in cellular
Linolenate, singlet/triplet oxygen lipids, 64–65
oxidation rates and, 17 Michael addition of thiols/amines, and
Linolenic acid carbonyl-mediated crosslinking of
and formation of conjugated trienoic proteins, 226–227, 229–230
acids in oils with PUFA, 65–69 Middle chain fatty acid (MCT), beany/
and oxidation of EPA/DHA, 53–54 grassy flavor and, 68–69
PUFAs of, 51 Molecular damage. See Co-oxidation of
and reversion flavor in soybean oil, proteins
21 Molecular orbital theory and triplet/
singlet oxygen and formation of singlet oxygens, 1–4
2-pentyl furan, 18 Monofunctional alkanals, co-oxidation
Lipid epoxide reactions, co-oxidation of of proteins and, 197–199
proteins and, 192–197
Lipid hydroxyperoxide reactions, co-
oxidation of proteins and, 189–192 N
Lipid oxidation. See also Co-oxidation of NADPH, superoxide anion radicals and,
proteins; Oxidation; Tocopherols 42
carotenoid/tocopherol synergism in,
168, 170 O
mechanism, 52–53 Oil-in-water emulsions
Lipid radicals. See Co-oxidation of food dispersions
proteins bulk oils, 280, 282–286
Lipophilic antioxidants, 303 emulsifier roles, 276–279
Liposomes free radical scavenging antioxidant
oxidative stability of DHA in, 61–63, impact, 279–281
71 lipid oxidation, 274–275
Long-chain polyunsaturated fatty acids overview, 273–274, 286–288
(PUFA) protein impact, 275–276
DHA Oleate, singlet/triplet oxygen oxidation
oxidative stability in liposomes, rates and, 17
61–63 Olefins, Type II reaction of Sensitizer*
EPA/DHA
and, 8 2-pentyl furan

Oleic acid formation from linoleic acid by
hydroxyl radical addition reaction to, singlet oxygen, 18
33–34 formation from linolenic acid by
ozone reaction with, 44–45 singlet oxygen, 19
Oligomer formation and reversion flavor in soybean oil,
and polymerization from propagation 21
reactions, 93–95 Perepoxides, as product of CLA
and polymerization from termination oxidation, 101–103
reactions, 95–97 Peroxide value (PV)
Olive oil, singlet oxygen photooxidation and autoxidation of CLA, 78–86
and, 20 and oxidation of EPA/DHA, 54–57
Oxidation Peroxyl radical reactions
carotenoid classification degree of, and evaluation strategies, 303
165 Peroxyl radicals
carotenoids and, 159–161 carotenoids
cholesterol/phytosterols antioxidant activity, 158–159
kinetics of sterol autoxidation, enzyme-catalyzed co-oxidation
119–122 with UFAs, 161–162
products/mechanisms of sterol kinetics/mechanisms, 155,
autoxidation, 114–118 157–158
products/mechanisms of sterol oxidation products, 159–161
photoxidation, 118–119 chemistry/reactions of, 37–41
sterol/stanol structures, 111–114 sterol autoxidation and, 115, 117
singlet oxygen and, 17–20 Phosphatidylcholine (PC), oxidative
singlet oxygen/fatty acid reaction stability of DHA and, 61–62
and, 13–17 Phospholipids
Oxidized product formation (ØAO2), and levels of EPA/DHA in marine
quantum yield of, 25–27 animal tissue, 56
Oxygen consumption during oxidation marine animal tissue and, 59
of fish lipids, 59–60 Photooxidation. See also Oxidation;
Oxygen quenching mechanisms Oxygen quenching mechanisms
carotenoids, 22–24 and autoxidation of CLA studies, 80
determination, 24–27 of CLA methyl esters/ML, 104
rates, 23–24 singlet oxygen
tocopherols, 24 olive oil, 20
Oxygen radicals. See Reactive oxygen soybean oil, 20
species (ROS) of sterols, 118–119
Oxygen species. See Reactive oxygen Photosensitization
species (ROS) chemical quenching by carotenoids,
Ozone, 44–46 153–156
Ozonide, 44 deuterium oxide effect and, 9
physical quenching by carotenoids,
P 150–153
Pauli’s exclusion principle, and elecron singlet oxygen formation and, 5
configuration of triplet/singlet oxygen, Physical quenching of photosensitizers/
1, 3 singlet oxygen by carotenoids,
152–153 hydroperoxyl radicals, 35, 37

Physiological γ-ketoaldehydes/ hydroxyl radicals, 31–34
levuglandins ozone, 44–46
co-oxidation of proteins and, peroxyl radicals, 37–41
209–211 superoxide anion radicals, 42–43
Phytosterol oxidation products (POP), triplet oxygen as, 4
111–114 Rancidity, low-molecular-weight
Phytosterols compounds and, 1
oxidation Rapeseed oil, stigmasterol
kinetics of sterol autoxidation, transformation to stigmasterol oxide
119–122 in, 121–122
products/mechanisms of sterol Rate of initiation (RI), and effect of
autoxidation, 114–118 temperature on tocopherols, 138
products/mechanisms of sterol Reactive oxygen species (ROS)
photoxidation, 118–119 alkoxyl radicals, 35–36
sterol/stanol structures, 111–114 hydrogen peroxide, 43–44
Polymerization hydroperoxyl radicals, 35, 37
oligomer formation and, 93–97 hydroxyl radicals, 31–34
protein crosslinking and, 223–225 overview, 31
Polyunsaturated fatty acids, singlet ozone, 44–46
oxygen oxidation and, 13–14 peroxyl radicals, 37–41
Pro-oxidant. See α-tocopherol superoxide anion radicals, 42–43
Propanal, formation during oxidation of Reversion flavor
fish lipids, 59–60 oxidation of long-chain PUFA and,
Proteins, co-oxidation of. See Co- 71
oxidation of proteins and PV of soybean oil, 68
PUFA. See Long-chain polyunsaturated and singlet oxygen, 20–21
fatty acids (PUFA) soybean oil and, 20–21
Pyridinium crosslink formation, and Riboflavin (RF), superoxide anion
carbonyl-mediated crosslinking of radicals and, 42–43
proteins, 229–230
Pyrrole link formation, and carbonyl- S
mediated crosslinking of proteins, Salmon roe lipids, 59–60
227–229 Schiff base amine condensation
and carbonyl-mediated crosslinking
Q of proteins, 225, 227, 229–230
Quantum yield of oxidized product Sensitizer*
formation (ØAO2), 25–27 conversion to triplet state, 6
Quenching by carotenoids riboflavin (RF)
chemical, 153–156 superoxide anion radicals and,
physical, 152–153 42–43
singlet oxygen quenching
R mechanisms and, 22
Radical compounds Type I/II mechanism of, 6–8
alkoxyl radicals, 35–36 Singlet oxygen. See also Sensitizer*
hydrogen peroxide, 43–44 2-pentyl furan formation, 18–19
and chemical quenching by reversion flavor in, 20–21

carotenoids, 150–153 singlet oxygen, photooxidation, 20
cholesterol oxidation products and, Spin multiplicity, and elecron
10–11 configuration of triplet/singlet oxygen,
CLA oxidation 3
primary product formation, Squid lipids, and levels of EPA/DHA,
100–104 56–58
secondary product formation, Sterols
104–105 autoxidation kinetics of, 119–122
detection autoxidation of, 114–118
chemical traps, 9–10 campesterol oxidation products of,
chemiluminence, 12 113
deuterium oxide effect, 9 cholesterol oxidation products of,
ESR spectroscopy, 12–14 112
quenchers, 10 dehydration products of, 118–119
electron configuration of, 1–4 marine animal tissue and, 59
formation photooxidation of, 118–119
chemical/photochemical/biological sitosterol oxidation products of, 113
methods, 4–6 sterol oxides in food, 114
and formation of tetraoxy stigmasterol oxidation products of,
intermediate, 137–138 113, 121–122
formed from α-tocopherol radicals, structure of, 111
135–137 Stigmasterol, 113, 121–122
lifetime in solution of, 8 Superoxide anion radicals, chemistry/
natural decay of, 8 reactions of, 42–43
oxidation in foods, 17–20 Superoxide anion reactions, deuterium
oxygen quenching mechanisms oxide effect and, 9
carotenoids, 21–24
determination, 24–27 T
tocopherols, 24 Temperature
and physical quenching by effect on carotenoid oxidation and
carotenoids, 153–156 cleavage, 166–168
reaction with fatty acids effect on tocopherol antioxidant
hydroperoxides formed, 16–17 activity, 138
olefins/dienes, 14–16 heat/pressure protection (evaluation),
polyunsaturated, 13–14 301–302
and reversion flavor in soybean oil, singlet oxygen oxidation and, 13
20–21 Tertiary amines, as quencher of singlet
Sitosterol, as sterol oxidation product, oxygen reaction, 10
113 Thermodynamic control of reactions,
Solubility/hydrophobicity/ 302–303
conformation/aggregation Thiobarbituric acid reactive substances
co-oxidation of proteins and, 220, (TBARS), and oxidative stability of
222 EPA/DHA, 55–60
Soybean oil Time-resolved singlet oxygen detection,
and formation of conjugated trienoic 13
acids in oils with PUFA, 65–69
Tocopherols Tuna lipids, and levels of EPA/DHA,

alkoxyl radicals and, 35–36 56–58
antioxidant efficacy due to Type I/II mechanism of Sensitizer*, 6–8
concentration
antioxidant mechanism of U
α-tocopherol, 128–129 Unsaturated aldehydes, co-oxidation of
effect of temperature, 138 proteins and, 202–204
overview, 127–128 Unsaturated fatty acids
pro-oxidant mechanism of carotenoid enzyme-catalyzed co-
α-tocopherol, 129, 131–138 oxidation with, 161–162
hydroxyl radicals and, 34 rates constants of autoxidation
lipid content from marine organisms reactions, 130
and, 58 sterol autoxidation and, 120–121
marine animal tissue and, 56, 59
and oxidation rate of CLA isomers,
89 W
as oxygen quenching mechanisms, Water. See Reactive oxygen species
21–22, 24 (ROS)
peroxyl radicals and, 39–41
as quencher of singlet oxygen X
reaction, 10 Xanthene, superoxide anion radicals
superoxide anion radicals and, 43 and, 42
Triacylglycerol (TAG) Xanthophylls
beany/grassy flavor of MCT and, carotenoid structre/antioxidant
68–69 activity of, 158–159
as main lipid class of fish oils, 58 singlet oxygen quenching
sterol autoxidation and, 121 mechanisms and, 23
Trienoic acids, formation of conjugated,
65–69
Tripalmitin, stigmasterol transformation
to stigmasterol oxide in, 121–122
Triplet oxygen. See also Singlet oxygen
carotenoids and, 163–165
electron configuration of, 1–4
hydroperoxides formation and,
16–17
hydroperoxyl radical formation and,
35, 37
hydroxyl radicals and, 33
oxidation in foods, 17–20
oxidized product formation (ØAO2)
and, 25–27
ozone production and, 44
peroxyl radicals and, 37–38
Sensitizer* conversion to, 6
Trout egg lipids, and levels of EPA/
DHA, 58

Lipid Oxidation Path

Uploaded by

Copyright:

Available Formats

Lipid Oxidation Path

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lipid Oxidation Path

Uploaded by

Copyright:

Available Formats

Lipid Oxidation Pathways

AOCS Books and Special Publications Committee

AOCS Press, Urbana, IL 61802

Library of Congress Cataloging-in-Publication Data

Lipid oxidation pathways / [edited by] Afaf Kamal-Eldin.

Printed in the United States of America.

1 Chemistry and Reactions of Singlet and Triplet Oxygen in Lipid Oxidation

Afaf Kamal-Eldin David B. Min

Electron Configuration of Triplet and Singlet Oxygen

Fig. 1.1. Molecular orbital of triplet oxygen.

Fig. 1.2. Molecular orbital of singlet oxygen.

Table 1.1. Comparison of Triplet and Singlet Oxygen

Energy Level 0 22.4 kcal/mole

Nature Diradical Non-radical, Electrophilic

Reaction Radical compound Electron-rich compounds

Formation of Singlet Oxygen

Xanthine oxidase + 3O2 R+

O2·- + O2·- + 2H+ H2O2 + O2

R-CH-R’ R-CH-R’ R-CH-R’ R-C-R’ 1

The metastable phosphatidylcholine hydroperoxides produced singlet oxygen

Singlet Manifold System Triplet Manifold System

Fig. 1.4. Jablonski diagram.

Type I and Type II Mechanisms

ROOH O2·– + 1Sen ROOH

Detection of Singlet Oxygen

Protiated acetone 46.5

Deuterated acetone 690

Protiated acetonitrile 54.4

Deuterated acetonitrile 600

Protiated benzene 26.7

Deuterated benzene 550

Deuterated water 68.1

Deuterium Oxide Effect

[Q] = (ka[A] + kd)/kq

Fig. 1.6. Cholesterol oxidation products by singlet oxygen: 3β-hydroxy-5α-cholest-6-ene-5-hydroperoxide

Reaction of Singlet Oxygen with Fatty Acids

3370 G 3390 G 3410 G

Singlet oxygen oxidation produces hydroperoxides at the positions of double

Singlet Oxygen Oxidation in Foods

Nonconjugated hvdro- 10-OOH 10-OOH

Frankel et al., (1979)

Singlet Oxygen 30,000 40,000 70,000

Reversion Flavor in Soybean Oil

Singlet Oxygen Quenching Mechanisms

Quenching Mechanism of Carotenoids

Table 1.5. Singlet Oxygen Quenchers and Their Quenching Rates

Zeaxanthin 6.8 ×109

Lycopene 6.9 ×109

Isozeqaxanthin 7.4 ×109

Astaxanthin 9.9 ×109

Canthaxanthin 11.2 ×109

Dimethylfuran 2.6 × 107

Bis(di-n-butyldithiocarbamato)nickel chelate 1.2 × 109

Quenching Mechanisms of Tocopherols

The destruction of vitamin D2 in a model system by singlet oxygen oxidation was

Determining Quenching Mechanisms

number of molecules of a product formed to the number of photons of light ab-

This equation can be inverted to:

so that it is in slope-intercept form.

kd (ko [3O2] + kq [Q])

1 Chemistry and Reactions of Singlet and Triplet Oxygen in Lipid Oxidation