Flavoids in Health and Disease (Catherine Rice Evans-Lester Packer) PDF

flavonoids in
Health and
kond Edition
Revised and Expanded
e d i t e d by
Catherine fl. Rice-Euans

King's College London
London, England
Lester Packer
University of Southern California School of Pharmacy
Los Angeles, California
n
MARCEL
MARCEL DEKKER, INC.
NEW YORK BASEL
Although great care has been taken to provide accurate and current information, neither the author(s) nor the publisher, nor anyone else associated
with this publication, shall be liable for any loss, damage, or liability directly
or indirectly caused or alleged to be caused by this book. The material contained herein is not intended to provide specic advice or recommendations
for any specic situation.
Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used only for identication and explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
A catalog record for this book is available from the Library of Congress.
ISBN: 0-8247-4234-6
This book is printed on acid-free paper.
Headquarters
Marcel Dekker, Inc.,
270 Madison Avenue, New York, NY 10016
tel: 212-696-9000; fax: 212-685-4540
Eastern Hemisphere Distribution
Marcel Dekker AG
Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland
tel: 41-61-260-6300; fax: 41-61-260-6333
World Wide Web
http://www.dekker.com
The publisher oers discounts on this book when ordered in bulk quantities.
For more information, write to Special Sales/Professional Marketing at the
headquarters address above.
Copyright n 2003 by Marcel Dekker, Inc.
All Rights Reserved.
Neither this book nor any part may be reproduced or transmitted in any form
or by any means, electronic or mechanical, including photocopying, microlming, and recording, or by any information storage and retrieval system,
without permission in writing from the publisher.
Current printing (last digit):
10 9 8 7 6 5 4 3 2 1
PRINTED IN THE UNITED STATES OF AMERICA
OXIDATIVE STRESS AND DISEASE

Series Editors
LESTER PACKER, PH.D.

ENRIQUE CADENAS, M.D., PH.D.
Los Angeles, California
1
2
3
4.
5
6
7
8
9
Oxidative Stress in Cancer, AIDS, and Neurodegenerative Diseases,

edited by Luc Montagnier, Rene Olivier, and Catherine Pasquier
Understanding the Process of Aging. The Roles of Mitochondria, Free
Radicals, and Antioxidants, edited by Enrique Cadenas and Lester
Packer
Redox Regulation of Cell Signaling and Its Clinical Application, edited
by Lester Packer and Junji Yodoi
Antioxidants in Diabetes Management, edited by Lester Packer, Peter
Rosen, Hans J Tritschler, George L King, and Angelo Azzi
Free Radicals in Brain Pathophysiology, edited by Giuseppe Poll,
Enrique Cadenas, and Lester Packer
Nutraceuticals in Health and Disease Prevention, edited by Klaus
Kramer, Peter-Paul Hoppe, and Lester Packer
Environmental Stressors in Health and Disease, edited by Jurgen
Fuchs and Lester Packer
Handbook of Antioxidants Second Edition, Revised and Expanded,
edited by Enrique Cadenas and Lester Packer
Flavonoids in Health and Disease. Second Edition, Revised and
Expanded, edited by Catherine A Rice-Evans and Lester Packer
Related Volumes
Vitamin E in Health and Disease" Biochemistry and Clinical Applications, edited by Lester Packer and Jurgen Fuchs
Vitamin A in Health and Disease, edited by Rune Blomhoff
Free Radicals and Oxidation Phenomena in Biological Systems, edited
by Marcel Roberfroid and Pedro Buc Calderon
Biothiols in Health and Disease, edited by Lester Packer and Enrique

Cadenas
Handbook of Antioxidants, edited by Ennque Cadenas and Lester
Packer
Handbook of Synthetic Antioxidants, edited by Lester Packer and
Enrique Cadenas
Vitamin C in Health and Disease, edited by Lester Packer and Jurgen
Fuchs
Lipoic Acid in Health and Disease, edited by Jurgen Fuchs, Lester
Packer, and Guido Zimmer
Additional Volumes in Preparation
Series Introduction
Oxygen is a dangerous friend. Overwhelming evidence indicates that oxidative stress can lead to cell and tissue injury. However, the same free radicals
that are generated during oxidative stress are produced during normal
metabolism and thus are involved in both human health and disease.
Free radicals are molecules with an odd number of electrons. The odd,
or unpaired, electron is highly reactive as it seeks to pair with
another free electron.
Free radicals are generated during oxidative metabolism and energy
production in the body.
Free radicals are involved in:
Enzyme-catalyzed reactions
Electron transport in mitochondria
Signal transduction and gene expression
Activation of nuclear transcription factors
Oxidative damage to molecules, cells, and tissues
Antimicrobial action of neutrophils and macrophages
Aging and disease
Normal metabolism is dependent on oxygen, a free radical. Through
evolution, oxygen was chosen as the terminal electron acceptor for respiration. The two unpaired electrons of oxygen spin in the same direction; thus,
oxygen is a biradical, but is not a very dangerous free radical. Other oxygenderived free radical species, such as superoxide or hydroxyl radicals, formed
during metabolism or by ionizing radiation are stronger oxidants and are
therefore more dangerous.
In addition to research on the biological eects of these reactive oxygen
species, research on reactive nitrogen species has been gathering momentum.
NO, or nitrogen monoxide (nitric oxide), is a free radical generated by NO
synthase (NOS). This enzyme modulates physiological responses such as
vasodilation or signaling in the brain. However, during inammation, syniii
iv
Series Introduction
thesis of NOS (iNOS) is induced. This iNOS can result in the overproduction
of NO, causing damage. More worrisome, however, is the fact that excess NO
can react with superoxide to produce the very toxic product peroxynitrite.
Oxidation of lipids, proteins, and DNA can result, thereby increasing the
likelihood of tissue injury.
Both reactive oxygen and nitrogen species are involved in normal cell
regulation in which oxidants and redox status are important in signal
transduction. Oxidative stress is increasingly seen as a major upstream
component in the signaling cascade involved in inammatory responses,
stimulating adhesion molecule and chemoattractant production. Hydrogen
peroxide, which breaks down to produce hydroxyl radicals, can also activate
NF-nB, a transcription factor involved in stimulating inammatory
responses. Excess production of these reactive species is toxic, exerting
cytostatic eects, causing membrane damage, and activating pathways of
cell death (apoptosis and/or necrosis).
Virtually all diseases thus far examined involve free radicals. In most
cases, free radicals are secondary to the disease process, but in some instances
free radicals are causal. Thus, there is a delicate balance between oxidants
and antioxidants in health and disease. Their proper balance is essential for
ensuring healthy aging.
The term oxidative stress indicates that the antioxidant status of cells
and tissues is altered by exposure to oxidants. The redox status is thus
dependent on the degree to which a cells components are in the oxidized
state. In general, the reducing environment inside cells helps to prevent
oxidative damage. In this reducing environment, disulde bonds (SS) do
not spontaneously form, because sulfhydryl (SH) groups kept in the reduced
state prevent protein misfolding or aggregation. This reducing environment
is maintained by oxidative metabolism and by the action of antioxidant
enzymes and substances such as glutathione, thioredoxin, vitamins E and C,
and enzymes such as superoxide dismutase (SOD), catalase, and the
selenium-dependent glutathione and thioredoxin hydroperoxidases, which
serve to remove reactive oxygen species.
Changes in the redox status and depletion of antioxidants occur during
oxidative stress. The thiol redox status is a useful index of oxidative stress
mainly because metabolism and NADPH-dependent enzymes maintain cell
glutathione (GSH) almost completely in its reduced state. Oxidized glutathione (glutathione disulde, GSSG) accumulates under conditions of oxidant exposure, and this changes the ratio of oxidized to reduced glutathione;
an increased ratio indicates oxidative stress. Many tissues contain large
amounts of glutathione, 24 mM in erythrocytes or neural tissues and up
to 8 mM in hepatic tissues. Reactive oxygen and nitrogen species can directly
Series Introduction
react with glutathione to lower the levels of this substance, the cells primary
preventative antioxidant.
Current hypotheses favor the idea that lowering oxidative stress can
have a clinical benet. Free radicals can be overproduced or the natural
antioxidant system defenses weakened, rst resulting in oxidative stress, and
then leading to oxidative injury and disease. Examples of this process
include heart disease and cancer. Oxidation of human low-density lipoproteins is considered the rst step in the progression and eventual development of atherosclerosis, leading to cardiovascular disease. Oxidative DNA
damage initiates carcinogenesis.
Compelling support for the involvement of free radicals in disease
development comes from epidemiological studies showing that an enhanced
antioxidant status is associated with reduced risk of several diseases. Vitamin
E and prevention of cardiovascular disease is a notable example. Elevated
antioxidant status is also associated with decreased incidence of cataracts
and cancer, and some recent reports have suggested an inverse correlation
between antioxidant status and occurrence of rheumatoid arthritis and
diabetes mellitus. Indeed, the number of indications in which antioxidants
may be useful in the prevention and/or the treatment of disease is increasing.
Oxidative stress, rather than being the primary cause of disease, is
more often a secondary complication in many disorders. Oxidative stress
diseases include inammatory bowel diseases, retinal ischemia, cardiovascular disease and restenosis, AIDS, ARDS, and neurodegenerative diseases
such as stroke, Parkinsons disease, and Alzheimers disease. Such indications may prove amenable to antioxidant treatment because there is a clear
involvement of oxidative injury in these disorders.
In this new series of books, the importance of oxidative stress in
diseases associated with organ systems of the body will be highlighted by
exploring the scientic evidence and the medical applications of this knowledge. The series will also highlight the major natural antioxidant enzymes
and antioxidant substances such as vitamins E, A, and C, avonoids,
polyphenols, carotenoids, lipoic acid, and other nutrients present in food
and beverages.
Oxidative stress is an underlying factor in health and disease. More and
more evidence indicates that a proper balance between oxidants and antioxidants is involved in maintaining health and longevity and that altering
this balance in favor of oxidants may result in pathological responses causing
functional disorders and disease. This series is intended for researchers in the
basic biomedical sciences and clinicians. The potential for healthy aging and
disease prevention necessitates gaining further knowledge about how oxidants and antioxidants aect biological systems.
vi
Series Introduction
Flavonoids and avonoid-rich botanical extracts increasingly have

been a subject of interest for scientic research, conferences in the scientic
community, the herbal products and dietary supplement industry, and the
consumer. This interest has led to research exploring the molecular basis of
their action in health. Flavonoids in Health and Disease, Second Edition,
Revised and Expanded highlights the recent advances in this rapidly developing eld of study.
Lester Packer
Enrique Cadenas
Preface
The evidence for avonoid-rich food components as cardioprotective, neuroprotective, and chemopreventive agents is steadily accumulating. However,
their mechanisms of action still remain to be elucidated. Although there are
hundreds of avonoid molecules in dietary plants, the major components of
current interest for their benecial health eects are members of the avonol,
avanol, avanone, anthocyanin, and hydroxycinnamate families, as well as
specic stilbene and chalcone structures. Key molecules of common interest
are the avanol components of grapes, wine, and teas (epicatechin/catechin
family with their procyanindin oligomers and gallate esters) in relation to
cardioprotection; the berry fruit avonoids, especially in the context of
neuroprotection; and citrus avanones and the chemopreventive properties
of a variety of molecules including the green tea constituents resveratrol and
quercetin. In addition, a number of more unusual sources of avonoids are
presented and discussed as potential lead molecules or pro-nutraceuticals for
the future.
Flavonoids in Health and Disease, Second Edition, Revised and Expanded,
reects the major advances made in this exciting area of research over the last
5 years and the involvement of avonoids and phenolic compounds in
biochemical, physiological, and pharmacological processes that are implicated in disease prevention. In particular, the eld has gained much momentum through the knowledge that avonoids are modied on absorption by the
small intestine, through conjugation and metabolism, and by the large
intestine, through the actions of the colonic microora, and by subsequent
hepatic metabolism of the components that are absorbed. Thus, the avonoid
forms reaching the cells and tissues after leaving the gastrointestinal tract are
chemically, biologically, and, in many instances, functionally distinct from
the dietary forms, such features underlying their bioactivities. This new
edition thoroughly scrutinizes and updates these aspects. In addition, the
current thinking on and evidence for the modes of action of in vivo avonoid
metabolite forms, in particular under conditions of oxidative stress, are
vii
viii
Preface
addressed. It is clear from the studies reported in the literature since the rst
edition that the vision of avonoids functioning as hydrogen-donating
antioxidants in vivo is receding and that their antioxidant ability to protect
cells and tissues from oxidative stress is more likely to be through their inuences on signal transduction pathways and gene expression, the key
putative modes of action of these exciting physiological molecules. All these
aspects are explored in this volume.
In addition, the book provides a thorough update on the avonoid and
phenolic proles of dietary plants and herbs and the advances made in the
analysis of these molecules in the environment of the plant, in chemical
systems, on interaction with cells, and in vivo in biological uids.
Catherine A. Rice-Evans
Lester Packer
Contents
Series Introduction (Lester Packer and Enrique Cadenas)

Preface
Contributors
Part I.
Occurrence and Analysis
1. Phenolic Acids in Fruits and Vegetables

Annie Fleuriet and Jean-Jacques Macheix
2. Flavonoids in Herbs
Piergiorgio Pietta, Claudio Gardana, and Annamaria Pietta
3. The Relationship Between the Phenolic Composition
and the Antioxidant Activity of Fruits and Vegetables
Anna R. Proteggente, Sheila Wiseman, Frans H. M. M.
van de Put, and Catherine A. Rice-Evans
Part II.
iii
vii
xi
43
71
Characterization, Chemical, and Biological Properties
4. Applications of Flavonoid Analysis and Identication Techniques: Isoavones (Phytoestrogens) and 3-Deoxyanthocyanins
Ewald E. Swinny and Kenneth R. Markham
97
5. Synthesis, Identication, Quantication and Chemical

Reactivity of Methylated Flavan-3-ols
Cecile Cren-Olive and Christian Rolando
123
6. Investigation of Flavonoids and Their In Vivo Metabolite

Forms Using Tandem Mass Spectrometry
Gunter G. C. Kuhnle
145
ix
Contents
7. Eects of Flavonoids on the Oxidation of Low-Density

Lipoprotein and Atherosclerosis
Michael Aviram and Bianca Fuhrman
Part III.
Cell Interactions In Vivo and In Vitro
8. Phytochemicals and Brain Aging: A Multiplicity of Eects

Kuresh A. Youdim and James A. Joseph
9. Flavonoids: Neuroprotective Agents? Modulation of Oxidative StressInduced MAP Kinase Signal Transduction
Hagen Schroeter and Jeremy P. E. Spencer
10.
Mitochondrial Actions of Flavonoids and Isoavonoids

Clinton S. Boyd and Enrique Cadenas
11. Gene Modulation of HaCaT Cells Induced by Pine Bark

Extract
Bertrand Henri Rihn and Claude Saliou
12. Cytoprotective and Cytotoxic Eects of Flavonoids
Jeremy P. E. Spencer, Hagen Schroeter, and Catherine A.
Rice-Evans
Part IV.
165
205
233
273
303
309
Absorption, Metabolism, and Bioavailability
13. Understanding the Bioavailability of Flavonoids Through

Studies in Caco-2 Cells
Thomas Walle, Richard A. Walgren, U. Kristina Walle,
Alema Galijatovic, and Jaya b. Vaidyanathan
349
14. Metabolism in the Small Intestine and Gastrointestinal Tract

Jeremy P. E. Spencer, Surjit Kaila Singh Srai,
and Catherine A. Rice-Evans
363
15. Absorption of Quercetin Glycosides

Andrea J. Day and Gary Williamson
391
16. Bioavailability of Flavanol Monomers

Jennifer L. Donovan and Andrew L. Waterhouse
413
17. Absorption and Metabolism of Hydroxycinnamates

Andreas R. Rechner
441
Index
459
Contributors
Michael Aviram, D.Sc. Lipid Research Laboratory, Technion Faculty of

Medicine, Rambam Medical Center, Haifa, Israel
Clinton S. Boyd Department of Cell and Molecular Pharmacology and
Toxicology, University of Southern California School of Pharmacy, Los
Angeles, California, U.S.A.
Enrique Cadenas, M.D., Ph.D. Department of Cell and Molecular Pharmacology and Toxicology, University of Southern California School of Pharmacy, Los Angeles, California, U.S.A.
Cecile Cren-Olive
dAscq, France
Lille University of Science and Technology, Villeneuve
Andrea J. Day Procter Department of Food Science, University of Leeds,

Leeds, England
Jennifer L. Donovan Laboratory of Drug Disposition and Pharmacogenetics, Medical University of South Carolina, Charleston, South Carolina,
U.S.A.
Annie Fleuriet, Dr. Sci. Department of Plant Biochemistry and Molecular
Biology, University of Montpellier II, Montpellier, France
Bianca Fuhrman, D. Sc. Lipid Research Laboratory, Technion Faculty of
Medicine, Rambam Medical Center, Haifa, Israel
Alema Galijatovic, Ph.D. Department of Molecular Pharmacology and
Experimental Therapeutics, Medical University of South Carolina, Charleston, South Carolina, U.S.A.
Claudio Gardana Institute of Biomedical TechnologiesNational Council
of Research, Segrate, Italy
xi
xii
Contributors
James A. Joseph, Ph.D. Jean Mayer USDA Human Nutrition Center on

Aging at Tufts University, Boston, Massachusetts, U.S.A.
Gunter G. C. Kuhnle, Dr. rer. nat. Wolfson Center for Age-Related Diseases, Kings College London, London, England
Jean-Jacques Macheix, Dr. Sci. Department of Plant Biochemistry and
Molecular Biology, University of Montpellier II, Montpellier, France
Kenneth R. Markham, Ph.D., F.N.Z.I.C., F.R.S.N.Z. New Zealand Institute for Industrial Research and Development, Lower Hutt, New Zealand
Annamaria Pietta, Pharm. D.
Farmacia Dr. Carlo Bravi, Brescia, Italy
Piergiorgio Pietta, Ph.D. Institute of Biomedical TechnologiesNational

Council of Research, Segrate, Italy
Anna R. Proteggente, Pharm. D. Wolfson Center for Age-Related Diseases, Kings College London, London, England
Andreas R. Rechner Wolfson Center for Age-Related Diseases, Kings
College London, London, England
Catherine A. Rice-Evans, Ph.D., D.Sc., F.R.C.Path. Wolfson Center for
Age-Related Diseases, Kings College London, London, England
Bertrand Henri Rihn
Christian Rolando
dAscq, France
Teaching Hospital of Brabois, Vandoeuvre, France
Lille University of Science and Technology, Villeneuve
Claude Saliou Johnson & Johnson Consumer Products Worldwide, Skillman, New Jersey, U.S.A.
Hagen Schroeter, Ph.D. Department of Cell and Molecular Pharmacology
and Toxicology, University of Southern California School of Pharmacy,
Los Angeles, California, U.S.A., and Wolfson Center for Age-Related
Diseases, Kings College London, London, England
Jeremy P. E. Spencer, Ph.D. Wolfson Center for Age-Related Diseases,
Kings College London, London, England
Surjit Kaila Singh Srai, Ph.D. Department of Biochemistry and Molecular
Biology, Royal Free University College Medical School, London, England
Ewald E. Swinny, Ph.D. Department of Horticulture, Viticulture, and
Enology, University of Adelaide, Adelaide, South Australia
Contributors
xiii
Jaya b. Vaidyanathan Department of Cell and Molecular Pharmacology

and Experimental Therapeutics, Medical University of South Carolina,
Charleston, South Carolina, U.S.A.
Frans H. M. M. van de Put, Ph.D.
The Netherlands
Unilever Health Institute, Vlaardingen,
Richard A. Walgren, M.D., Ph.D. Department of Cell and Molecular

Pharmacology and Experimental Therapeutics, Medical University of South
Carolina, Charleston, South Carolina, U.S.A.
Thomas Walle, Ph.D. Department of Cell and Molecular Pharmacology
and Experimental Therapeutics, Medical University of South Carolina,
U. Kristina Walle, R. Ph. Department of Cell and Molecular Pharmacology and Experimental Therapeutics, Medical University of South Carolina,
Andrew L. Waterhouse, Ph.D. Department of Viticulture and Enology,
University of California, Davis, Sacramento, California
Gary Williamson, Ph.D.
Institute of Food Research, Norwich, England
Sheila Wiseman, Ph.D.

erlands
Unilever Health Institute, Vlaardingen, The Neth-
Kuresh A. Youdim, Ph.D. Wolfson Center for Age-Related Diseases,

Kings College London, London, England
1
Phenolic Acids in Fruits
and Vegetables
Annie Fleuriet and Jean-Jacques Macheix
University of Montpellier II
Montpellier, France
I.
INTRODUCTION
The several thousand polyphenols that have been described in plants can be
grouped into distinct classes, most of which are found in fruits and vegetables
[1,2]. Distinctions among these classes are drawn rst on the basis of the number
of constitutive carbon atoms and then in light of the structure of the basic
skeleton. Phenolic acids belong to two dierent classes, hydroxybenzoic acids
(HBA) and hydroxycinnamic acids (HCA), which derive from two nonphenolic
molecules, benzoic and cinnamic acid, respectively. In contrast to other phenolic
compounds, HBA and HCA present an acidic character because of the presence
of one carboxylic group in the molecule. They are widely represented in plants,
although their distribution may strongly vary with species, cultivar, and
physiological stage. They clearly play a role both in the interactions between
the plant and its biotic or abiotic environment and in the organoleptic and
nutritional qualities of fruits, vegetables, and derived products, e.g., fruit juices,
wines, and ciders. Furthermore, their antioxidant properties are essential in the
stability of food products and in antioxidant defense mechanisms of biological
systems. These last aspects are largely developed elsewhere in this volume.
Plant organs consumed by humans as vegetables have various botanical
origins, e.g., leaves, stems, shoots, owers, roots, rhizomes, tubers, bulbs, seeds,
pods, and even some eshy fruits. In some cases, it is not easy to distinguish
between fruits and vegetables, as there is no concordance between the botanical
denitions and the common use of plant organs by the consumer. For instance,
1
Fleuriet and Macheix
bean pods, tomatoes, eggplant fruits, and sweet peppers are fruits in a botanical
sense, whereas they generally are commercially marketed as vegetables.
Qualitative and quantitative determinations of phenolic acids, especially the
combined forms, have been signicantly improved during the last two decades,
allowing one to draw a general picture of their distribution in fruits and
vegetables and their importance as food constituents. In the comprehensive
reviews on these topics that have already been published [15] most of the
oldest references may be found. In the present review, our attention is focused on
the presence and content of phenolic acids in fruits (mainly eshy fruits with their
seeds) and vegetables, and on the main parameters that can modify them.
II.
ANALYSIS
Soluble HBA or HCA derivatives are frequently extracted from fruits and
vegetables with ethanol or methanol-water solutions (80/20, v/v), using low
temperatures and adding an antioxidant to prevent oxidation during the extraction
procedure. Chemical or enzymatic hydrolysis of the plant material is necessary
when phenolic acids are linked to cell wall constituents to give insoluble forms
[6]. Apolar solvents or supercritical carbon dioxide may be useful to extract
phenolic lipids [7,8]. In the case of acylated avonoids, solvents must be adapted
to the characteristics of the avonoid itself, e.g., acidic methanol for fruit
anthocyanins, although some artefacts may appear under these conditions.
Purication of the raw extract is essential. This may be performed in a rst
stage by removing chlorophylls and carotenoids and in a second stage by extracting
phenolic acids with ethyl acetate from the depigmented aqueous extract, using a
method previously described for fruits [2]. A preliminary analysis on a polyamide
column has the advantage of separating the two groups of HCA derivatives:
glucose derivatives on the one hand and quinic, tartaric, malic, or galactaric
derivatives on the other [7]. Paper chromatography, classical or high-performance
thin-layer chromatography, and column chromatography have been used extensively since the 1960s to separate phenolic acids, both before and after hydrolysis
of esters and glycosides. Furthermore, separation of phenolic acid conjugates has
greatly progressed thanks to high-performance capillary electrophoresis [9,10] and
high-performance liquid chromatography (HPLC), which also allows quantitative
determinations. In particular, the development of reversed-phase columns has
greatly improved the separation performance of HCA and HBA derivatives [7].
In addition to analytical separations, the identication of phenolic acids has
greatly beneted from the development of modern techniques (infrared [IR] and
nuclear magnetic resonance [NMR] spectroscopy, mass spectrometry, etc.), that
have added to the accurate knowledge of the structure of natural phenolic
molecules [7]. New analytical approaches, including Raman spectroscopy, also
Phenolic Acids in Fruits and Vegetables
allow in situ detection of HCA covalenty linked to cell wall constituents [6].
Some early approximate identications have now been rectied, but there may be
others as yet unrecognized [4].
In some unusual cases, spectrophotometric estimation of a major phenolic
acid may be performed directly in plant extracts, such as chlorogenic acid in
apples, pears, or potatoes [2,11], but this gives approximative information. From
a quantitative point of view, HPLC techniques appear to be the most suitable, and
they have been widely developed for estimating individual plant phenolic acids in
their native forms [7]. Numerous examples concerning fruits and vegetables have
already been reported [1,2]. Nevertheless, given the diversity and complexity of
the combined forms naturally present, it has often been easier to determine
phenolic acids released after hydrolysis of the extract, although some molecules
might then be degraded.
A rapid uorometric determination of p-coumaric, protocatechuic, and
gallic acids has also been proposed in persimmon [12], but interference with
other phenolic compounds is likely. Moreover, the radical scavenging activities of
HBA and HCA may be used for their quantitative determination by chemiluminescence in the presence of hydrogen peroxide [13].
III.
OCCURRENCE IN FRUITS AND VEGETABLES
In most cases, phenolic acids are not found in a free state, except in trace levels,
but as combined forms, either soluble and then accumulated in the vacuole or
insoluble when linked to cell wall components. Nevertheless, some exceptional
situations can cause phenolic acids to accumulate in the free form [2]: brutal
extraction conditions, physiological disturbances, contamination by microorganisms, anaerobiosis, processing of fruit juices, and winemaking. As they also
accumulate when plant extracts are submitted to hydrolysis, the free HCA
prole may characterize the plant material, and it has been used to discriminate
between blood and blond oranges [14]. In rare cases, for example, in Capsicum
species, the balance between free and combined forms may serve as a chemotaxonomic criterion: free phenolic acids are present in fruits of C. annum,
whereas only the glycosylated forms appear in C. frutescens [15].
A.
Hydroxybenzoic Acids
HBAs have a general structure of the C6-C3 type derived directly from benzoic
acid (Fig. 1), and variations in structure lie in the hydroxylations and methoxylations of the aromatic cycle. They are mainly present in fruits and vegetables in
the form of O-glucosides, but glucose esters of p-hydroxybenzoic, vanillic, or
syringic acids have also been reported, e.g., in garden cress (Table 1). In most of
Figure 1 Chemical structure of hydroxybenzoic acids and some derivatives

identified in fruits and vegetables.
3
5
3
1
14
6
1
19.9
37
11-44
<0.5-6
<0.5-4
1036
2-4
1
3
418
8 - 67
68 -189
6 -16
1
1
4-7
1
410
30-62
<10-52
0-16
3-6
2-9
45
12
Extreme value for several cultivars (milligrams per kilogram fresh weight); t, traces,, not detectable.
Determined after hydrolysis in most cases.
Source: Refs. 13,19,123.
Derivatives
p-HB glucoside
p-HB glucose
V glucoside
V glucose
Sy glucoside
Sy glucose
P glucoside
G glucoside
G glucose
G quinic acid
Ellagic acid
2-11 212
<0.55
38
66
91
3
3
2
45
2
6
129.6
37.7
Black
Horseradish Onion Garden Potato
Tomato Cherry Plum Strawberry Blackberry currant Blueberry
leaf
peel cress peel
Free acidsb
p-hydroxybenzoic
00.07
( p-HB)
Protocatechuic (P)
t
Vanillic (V)
0.072.275 <0.51
Salicylic
0.04
<0.5-1
Gallic (G)
t0.46
Syringic (Sy)
t11
Grape
Contents of Hydroxybenzoic Acids in Vegetables and Ripe Fruitsa
Compounds
Table 1

5
the important species of fruits and vegetables, because HBA conjugates are only
found in trace concentrations, their identication is dicult. The presence of free
HBA likely corresponds to degradation products from conjugates forms, during
either extraction or subsequent hydrolysis. For example, salicylic, p-hydroxybenzoic, vanillic, gentisic, 3,4-dihydroxybenzoic, syringic, p-coumaric, and gallic
acids were identied in the fruit of Diospyros lotus, whereas no information was
reported about native forms [16].
Three HBAs ( p-hydroxybenzoic, vanillic, and protocatechuic) are apparently universal in the angiosperms, and others (e.g., syringic, gallic, salicylic)
are also frequently present in either complex structures, i.e., hydrolysable
tannins, or as simple derivatives in combination with sugars or organic acids.
Gallic acid, hexahydroxydiphenic acid, and pentagalloylglucose (Fig. 1)
are also constituents of hydrolyzable tannins. In addition, very low concentrations of gallic acid are found in fruits in the form of esters with quinic acid
(theogallin) or glucose (glucogallin) and in the form of glucosides. Glucogallin
has also been identied in persimmon and isolated only from astringent
immature fruit, whereas free gallic acid was found in immature fruit of both
astringent and nonastringent varieties [2]. Glucogallin was thus proposed as a
good index for distinguishing between astringent and nonastringent varieties.
Gallic acid is also found combined with naringenin in fruits of Acacia farnesiana
or with ()epicatechin to form epicatechin 3-O-gallate, a constituent of unripe
grapes [2] (Fig. 1). Ellagic acid, a dimer of gallic acid, is a component of
ellagitannins, but it has also been reported in the free form and as arabinoside,
acetylxyloside, or acetylarabinoside in raspberry and strawberry [1719].
p-Hydroxybenzoic and vanillic acids are also present in numerous fruits
and vegetables [1], and the native forms are frequently simple combinations
with glucose (Table 1). Other derivatives have been detected in certain fruits
[1,2]: the methyl ester of p-hydroxybenzoic acid in passion fruit, 3,4-dihydroxybenzoic aldehyde in banana, a phenylpropene benzoic acid derivative in fruits
of Jamaican Piper species, and benzoyl esters and other derivatives in the fruits
of Aniba riparia. Dierent new glycosides of HBA showing radical-scavenging
activity [e.g., a new guaiacylglycerol-vanillic acid ether (Fig. 1)] have been
identied in the fruits of Boreava orientalis [20].
Syringic acid or its glucoside has been reported in grape, plum, and some
vegetables (Table 1), but its distribution appears to be very limited. It is very
likely that p-hydroxybenzoic, vanillic, and syringic acids derive, at least
partially, from the degradation of certain lignied zones of the fruit when these
exist (stone, seed teguments, etc.).
Protocatechuic acid is found in a number of soft fruits and vegetables in
the form of glucosides (Table 1), generally much less abundantly than those of
p-hydroxybenzoic acid [1,2], except in onion peel, where it is prominent [21].
Salicylic and gentisic acids have been reported in very small quantities in the
fruits of certain Solanaceae (tomato, eggplant, pepper), Cucurbitaceae (melon,

cucumber) and other species (e.g., kiwi fruit, grapefruit, grape).
Very low concentrations of p-hydroxybenzoic, protocatechuic, and
t-cinnamic acids have been reported in dierent species of mushrooms
(Agaricus and Lentinus species), along with traces of caeic acid [22].
B.
Hydroxycinnamic Acids
Among fruit and vegetable phenolics, HCA derivatives play an important role that
is due to their abundance and diversity. They all derive from cinnamic acid and are
essentially present as combined forms of four basic molecules: coumaric, caeic,
ferulic, and sinapic acids (Fig. 2). Two main types of soluble derivatives have been
identied (Fig. 2): (1) those involving an ester bond between the carboxylic
function of phenolic acid and one of the alcoholic groups of an organic compound
(e.g., quinic acid, glucose), for example, chlorogenic acid, which has been
identied in numerous fruits and vegetables; and (2) those that involve a bond
with one of the phenolic groups of the molecule, e.g., p-coumaric acid O-glucoside
in tomato fruit. The diversity of HCA conjugates thus results from the nature of the
bonds and that of the molecule(s) involved. In addition, for each of these
compounds, the presence of a double bond in the lateral chain leads to the possible
existence of two isomeric forms: cis (Z) and trans (E). Although native compounds
are mainly of the trans form, isomerization occurs during extraction, purication,
and processing under the eect of light or other chemical and physical factors.
1. Hydroxycinnamic Acids in Fruits
Quinic esters of HCA have been reported for a long time in fruits. The rst were
chlorogenic acid (5-O-caeoylquinic acid*) and p-coumaroylquinic acid in apple
[23]. Chlorogenic acid was subsequently found in many other fruits (Table 2),
often accompanied by other caeoylquinic isomers such as neochlorogenic acid
(3-O-caeoylquinic acid) and cryptochlorogenic acid (4-O-caeoylquinic acid)/
isochlorogenic acid (a mixture of several di-O-caeoylquinic acids) in coee
beans or coee pulp, apple, avocado, pineapple, cherry, peach, eggplant [1,2], and
in loquat fruit [24]. Red berries are particularly rich in caeoylquinic esters, which
confer on them, along with anthocyanins, high antioxidant activity [25,26]. The
presence of tri-or tetra-O-caeoylquinic acids in some fruits or leaves seems to be
rather uncommon [4].
* The nomenclature of HCA quinic esters is in conformity with the International Union of Pure and
Applied Chemistry (IUPAC) recommendations. Chlorogenic acid is thus 5-caeoylquinic acid and
not 3-caeoyquinic acid as it was originally called.
Figure 2 Chemical structure of hydroxycinnamic acids and some common

derivatives identified in fruits and vegetables.
24
t9
2.6
2
t
815
6-19
19-68
3.5
1271
5-11
11.2
t
15-48
Tomato
t13
t
12
8-25
40-450
t
11140
1-21
73-620
Cherry
134
2-12
t2
t-4
4-40
3-34
15142
6-100
88-771
2-7
Plum
29
2-3
t
30282
29-142
Peach
522
t
t2
t-2
2-9
37123
t
26-132
Apricot
t2
1427
t
t2
Strawberry
24
2-6
25
4-11
2-5
t3
1
41-52
3-6
Blackberry
13
11-15
2-4
12
13-21
10-14
3-10
12
3-5
38-48
19-30
2
Black
currant
8
t
510
25
t
3-15
18512075
2-5
5-7
t
3
Blueberry
28145
329907
5-43
95207
Loquat
Extreme values for several cultivars (milligrams per kilogram fresh weight); CQ, p CQ; FQ, caeoyl, p-coumaroyl, feruloylquinic acids; CT, p-CT; FT,
caeoyl, p-coumaroyl, feruloyl tartaric acids; CG, p-CG, FG; SG, glucose esters; C Gluc, p-C Gluc; F Gluc, glucoside derivatives; t, traces;, not
detectable.
b
Skin of white or red grapes.
Source: Refs. 13,24,66,68.
Sinapic
SG
0.9865
Ferulic
FQ
FG
F Gluc
FT
260
t-8
1.8484
t12
346
2-4
t-19
p-Coumaric
5-p CQ
4-p CQ
3-p CQ
p-CG
p-C Gluc
p-CT
10516
Grapeb
6621
26510
t-12
t-6
Pear
Caeic
5-CQ
4-CQ
3-CQ
CG
C Gluc
CT
Apple
Contents of Hydroxycinnamic Derivatives in Ripe Fruitsa
Phenolic acid
derivatives
Table 2

9
10
Quinic derivatives of other HCAs have also been identied in numerous

fruits, e.g., several isomers of p-coumaroylquinic acid in apple and 5-Oferuloylquinic acid in tomato [2]. Although quinic derivatives are generally
abundant in fruits, some contain none at all, e.g., grape, cranberry, and strawberry
(Table 2). Mixed quinic di-esters of caeic and ferulic acids are also present in
robusta coee beans [4].
Tartaric esters are limited to certain fruits of Vitis species and to some
vegetables of the Asteraceae family (Tables 2 and 3). HPLC separations during
the 1980s fully conrmed previous data by showing that the only combined
form of caeic acid in grape was in fact caeoyltartaric acid (=caftaric acid)
(Fig. 2). In addition, p-coumaroyl and feruloyl-tartaric acids (respectively named
coutaric and fertaric acids) were found in varying proportions according to
species and physiological stages [27]. Caeoylshikimic esters (Fig. 2) are not
widespread in plants, but they are very abundant in date fruit, where they
participate in enzymic browning [2].
HCA derivatives with other hydroxyacids have rarely been identied in
fruits, although p-coumaroylmalic acid is present in pear skin [28] and 2V-O-pcoumaroyl-, 2V-O-feruloylgalactaric acids, 2V-O-p-coumaroyl-, 2V-O-feruloyl-,
and 2V,4V-O-diferuloylglucaric acids in the peel of citrus fruits [29].
Since the identication of 1-O-p-coumaroylglucose (Fig. 2) and caeic
acid 3-O-glucoside in potato berry, numerous derivatives of HCA with simple
sugars have been identied in various fruits [2] (Table 2), and cinnamoylglucose
itself has been reported in blood orange [30]. Glucose esters and glucosides may
be present simultaneously, for example, in tomato fruit, where p-coumaric and
ferulic acids are present both as glucosides and as glucose esters (Fig. 2),
whereas caeic acid is only represented by caeoylglucose. Glucose esters of
sinapic acid have also been reported in tomato and in Boreava orientalis, where
it is present along with a glucosinolate salt [31,32]. Dierent new phenylpropanoid derivatives with simple sugars have been shown in the fresh fruit of
Piscrama quassioides [33]. Verbascoside (Fig. 3) is an example of a rather more
complex chemical combination that was identied in olives and in the fruit of
dierent members of the Oleaceae family, along with several other caeoyl
glycosides [2].
Although HCA derivatives with sugars and hydroxyacids are simultaneously present in numerous fruits [e.g., apple, tomato, cherry (Table 2)], several
exceptions should be reported. Glucose derivatives of HCA are not present or
are present only as traces in pear and in grape, whereas HCAs are only present in
the form of conjugates with sugars in strawberry and cranberry [1,2].
The presence of hydroxycinnamoyl amides in fruits and vegetables has
rarely been reported. Feruloyputrescine (Fig. 2) occurs in grapefruit and orange
juice [29] but has not been found in tangerine or lemon. The p-coumaroyl or
caeoyl amides of hydro-or dihydroxyphenylalanine have been reported in
11
cocoa [34]. Two new phenolic amides were isolated from the fruit of white
pepper (Piper nigrum L.), N-trans-feruloyltyramine (Fig. 2) and N-transferuloylpiperidine, together with some other derivatives of piperidine and
phenolics [2].
Acylation of anthocyanins with certain phenolic acids has been known for
a long time [35]. Grape has been studied extensively, and it was shown that pcoumaric acid plays a major role in the acylation of malvidin (Fig. 3) and of all
the other anthocyanins present, whereas caeic acid combines only with
malvidin 3-glucoside, a condition common in fruits and vegetables [35]. In
eggplant, delphinidin is acylated with coumaric and caeic acids; delphinidin 3( p-coumaroylrutinoside)-5-glucoside is a major pigment in purple-skinned
varieties. In the fruit of Solanum guineese (garden huckleberry), petunidin 3( p-coumaroyl-rutinoside)-5-glucoside forms at least 70% of anthocyanins and is
accompanied by very small quantities of several other acylated derivatives [36].
An extreme case concerns the blue berries of Dianella species, which contain
delphinidin tetraglucosides bearing p-coumaroyl groups on two, three, or four of
the sugars [37] (Fig. 3).
Flavonoid glycosides other than anthocyanins can also be acylated with
HCA, but they have only rarely been reported in fruits, e.g., in the form of
kaempferol p-coumaroylglycosides in Tribulus terrestris, 7-O-p-coumaroylglycoside-naringenin in Mabea caudata, or rhamnetin-3-p-coumaroylrhamninoside in Rhamnus petiolaris [38]. p-Coumaric and ferulic acids are also present
in combination with betanidin monoglucoside (Fig. 3) in fruits of Basella
rubra [39].
HCA may also be covalenty attached to aliphatic components of cutin and
suberin. The amount of covalently bound phenolic compounds (m-, p-coumaric
acids and avonoids) in tomato fruit cutin increased during fruit development
and accounted for as much as 6% of cutin membranes. Protoplasts isolated from
immature tomato fruit secrete a wall that has been shown to contain suberin, in
which phenolic compounds formed 25% of total monomers [3].
2. Hydroxycinnamic Acids in Vegetables and Cereals
Most of HCA conjugates previously described in fruits are also present in
vegetables [1,4], but concentrations may be very dierent according to the
botanical origin and the nature of the plant organ. An extensive review of the
dierent HCA conjugates encountered in most vegetables consumed was
published in 1999 [4], and here we only summarize some peculiar points.
Caeoylquinic esters have been reported in most vegetables (Table 3):
cabbages, endive, artichoke, potatoes, carrot, etc. In addition to the classical
dicaeoylquinic acids, diferuloylquinic acids are present in carrot root [40].
Several points must be underlined in Brassica vegetables: (1) 3-O-caeoyl-
Contents of Hydroxycinnamic Derivatives in Vegetablesa
p-Coumaric
5-p CQ
4-p CQ
3-p CQ
p-CG
p-C Gluc
p-C meso T
p-CM
Caeic
5-CQ
4-CQ
3-CQ
di-CQ
CG
C Gluc
CT
di-CT
CM
t-2
217
2-104
t
t
3
58
t10
2-19
6-120
310
1153
21-31
163-334
10-13
10-17
38-73
2-18
36124
539
10
433
11
12
160260b
189-230
2329
1621
1621
2271c
4-20
3-9
3
t6
2-3
5541
t
t-9
1-58
t8
3-23
6-48
20-22
813
t61
t-9
Phenolic acid
Rhubarb Faba bean
derivatives
Cabbage Broccoli Radish Lettuce Endive Artichoke Spinach Potato Carrot
leaf
pod
Table 3
1020
3840
Corchurus
olitorius
leaves
12
4273
10
1416
42-64
t83
t-17
321
34
t-31
t2
3-2
olitorius
leaves
Extreme values for several cultivars (milligrams per kilogram fresh weight); t, traces;, not detectable; CQ, p CQ; FQ, caeoyl, p-coumaroyl,
feruloylquinic acids; CT, p-CT; FT, caeoyl, p-coumaroyl, feruloyl tartaric acids; CM, p-CM; FM, caeoyl, p-coumaroyl, feruloyl malic acids; CG, p-CG,
FG; SG, glucose esters; C Gluc, p-C Gluc; F Gluc, glucoside derivatives.
b
Estimate for cynarine (1,3 di-CQ).
c
Up to 229.1 in potato peel (Ref. 123).
Source: Refs. 1,2,4,40,43,44,89,123.
Sinapic
SG
3-71
733
217
Ferulic
FQ
di-FQ
FG
F Gluc
FT
FM
Phenolic acid
Rhubarb Fava bean
derivatives
Cabbage Broccoli Radish Lettuce Endive Artichoke Spinach Potato Carrot
leaf
pod

13
14
Figure 3 Chemical structure of some complex hydroxycinnamic derivatives

identified in fruits and vegetables.
15
quinic acid is always isomers and a similar condition is found for p-coumaroyl
and feruloyl quinic esters; (2) feruloyl and sinapoyl glucose esters are important
in kale and red cabbage; (3) mixed feruloyl-sinapoyl esters of gentibiose are
present in broccoli [41]; (4) malic esters are present in radish tuber and leaf,
whereas quinic and glucose esters are present only as traces [1].
Chlorogenic acid is also detected in fennel teas prepared by infusion or
decoction [42] and in the leaves of Corchurus olitorius used as a vegetable for
soup [43]. Artichoke capitula is characterized by signicant amounts of
chlorogenic acid and various dicaeoylquinic esters, especially 1,3-dicaeoylquinic acid, known as cynarin [44]. In addition to the previous compounds,
3,5-dicaeoyl-4-succinylquinic acid is present in garland [45] and several
caeoyl-methylquinic acids with a strong antioxidant activity were characterized in bamboo shoots [46].
Along with quinic esters, caeoyl and dicaeoyltartaric acids are prominent in the leaves of some of the Asteraceae [1], e.g., lettuce, endive, and
chicory. Although they are rarely present in fruits, malic esters of HCA are more
frequently found in vegetables, e.g., in the leaves and pods of faba bean and in
lettuce or spinach leaves (Table 3). Nevertheless, in the latter case, the
prominent HCA conjugate is p-coumaroyl-meso-tartaric acid [1,47]. Tartronic
acid occurs as p-coumaroyl, feruloyl, and caeoyl-tartronic esters in the leaves
of mung bean (Vigna radiata) [48]. Rosmarinic acid, a caeic ester of 3,4dihydroxyphenyllactic acid (Fig. 3), is found at a high level in extracts of
various culinary and medicinal herbs (up to 1 g/kg fresh weight in thyme), where
it shows remarkable antioxidant activity [4951].
As previously shown in the case of fruits, sugar esters of HCA are also
present in numerous vegetables, especially p-coumaroyl, caeoyl, and sinapoyl
glucose esters in Brassiceae, spinach leaves, and rhubarb stalk (Table 3). Root
and/or derived cell cultures of red beet are rich in dierent HCA esters, e.g.,
several feruloylglucose conjugates, a feruloylsucrose monoester, a ferulicaspartic acid amide, and a feruloylglycerol glucuronide [52,53]. Furthermore,
red beet also contains low concentrations of two conjugates of HCA with
betacyanins (the major coloring substances of red beet): lampranthin I
( p-coumaroylbetanin) and lampranthin II (feruloylbetanin) [53]. In addition to
the case of tomato fruit previously reported (Table 2), HCA glucosides have
been identied in faba beans (leaves and pods) and are present as traces in
carrot [1].
Although the presence of chlorogenic acid itself has rarely been reported
in barley grains [54], HCAs, and ferulic acid in particular, are generally found as
insoluble forms in various glucidic fractions of the cell wall. These compounds
have not been reported in eshy fruits but exist in Graminaeae and some other
plants from which they are easily liberated by chemical or enzymatic hydrolysis.
Several reviews on the subject were published in 1999 [6,55], and only a brief
16
summary is given here. Ferulic and p-coumaric acids are bound through an ester
linkage to the arabinoxylans or xyloglucans of Gramineae (wheat, maize, barley,
rice, etc.), leaves, straw, and grain (bran and aleurone layer). A part of ferulic
acid also exists as dehydrodimers (Fig. 2) (e.g., in grasses, cereals, Chinese
water chestnut, sugar beet, carrot), which cross-link and strengthen the wall [56
58], and a small amount of ferulic acid is also found in the cell walls of the thick
cuticle of eshy scales of onions [21]. In some dicotyledons (e.g., sugar beet,
spinach, beans) ferulic and p-coumaric acid are also bound to the galactose or
arabinose residues of pectins [4,59].
Apolar esters of sterols and stanols with ferulic or p-coumaric acids have
been reported in corn bran and other cereals [60]. Furthermore, oats contain
numerous caeic and ferulic esters of glycerol, long-chain alkanols, and xhydroxyacids, in addition to avenanthramides (esters of anthranilic acid with
either p-coumaric, caeic, or ferulic acids) [61].
Suberized potato includes long-chain fatty acids and phenolic derivatives
[62]. Furthermore, in addition to the HCA esters of p-coumaric acid, a
signicant number of ligninlike monolignol structures exist within suberin [63].
IV.
FACTORS AFFECTING THE PHENOLIC ACID CONTENT
Accumulation of phenolic acids in fruits and vegetables varies strongly in

relation to dierent factors: (1) the genetic background, (2) the stage of
development of the plant organ, and (3) the environmental and culture conditions. All these changes involve the regulation of phenolic metabolism
(biosynthesis and degradation) and its integration in the program of cell and
tissue dierentiation, the control of gene expression, and the regulation of
enzyme activities and of their compartmentation. The biosynthetic pathway of
phenolic compounds is now well known and is not discussed here as it is not
specic to fruits or vegetables. The deamination of phenylalanine, previously
formed via the shikimate pathway, yields the nonphenolic cinnamic acid that is
the direct precursor of the dierent HCAs and of their coenzyme A (CoA) esters
through the general phenylpropanoid metabolism [64]. CoA esters of HCAs are
common precursors of various other classes of phenolic compounds (e.g., HBA,
anthocyanins, tannins, lignins). The structure and regulation of genes encoding
the enzymes of the general phenylpropanoid metabolism from a number of plant
species have been studied. Gene expression and enzyme activity are subject to
large uctuations in relation to endogenous and external factors (e.g., temperature, light, various types of stress) [64]. Furthermore, the enzymatic oxidation
of phenolic compounds is of vital importance to the quality of fruits, vegetables,
and their products, because of the formation of undesirable color and avor and
the loss of nutrients during processing (see Secs. V and VI).
A.
17
Changes in the Phenolic Acid Patterns According

to Species and Cultivars
Numerous factors may inuence considerable qualitative and quantitative modications in the phenolic acid patterns of fruits and vegetables from dierent
species and cultivars. Although HBAs are present in most fruits and vegetables,
the HBA content of fruits is generally low, except in certain fruits of the Rosaceae
family and in particular blackberry, in which protocatechuic and gallic acid
content may be very high: respectively, 189 and 67 mg/kg fresh weight in the
richest cultivars (Table 1). Great interspecic dierences in HBA exist in fruits
and vegetables with regard to both quality and quantity, and such dierences are
also found among the varieties of the same species. In fact, qualitative and
quantitative investigation of the native molecules of HBA derivatives is still
inconclusive and it is dicult to draw general and nal conclusions.
Comparing HCA content in numerous fruits and vegetables reveals
enormous variations among species (Tables 2 and 3). Chlorogenic acid itself
is especially abundant in coee beans (610% on a dry matter basis), C. olitorius
leaves (3.8 g/kg fresh weight [FW]), blueberries (2 g/kg FW), corn salad
(approximately 1 g/kg FW), loquat fruit (329907 mg/kg FW), eggplant (575
632 mg/kg FW), purple carrot (541 mg/kg FW), and artichoke (433 mg/kg FW),
whereas it is present only as traces in Cucurbitaceae [1,2,4,24,40,43]. Similar
variations are also frequently reported between cultivars of the same species, for
example, 26 to 510 mg/kg chlorogenic acid in apples [2,6568] or 6 to 621 mg/
kg caftaric acid in grapes [2].
The relative proportions of each HCA are a characteristic of fruit in the
mature stage. Caeic acid is frequently the most abundant phenolic acid. It
commonly exceeds 75% of total HCA in numerous fruits and vegetables (e.g.,
apple, plum, tomato, grape, red cabbage, endive, artichoke, potatoes) and may
even form almost the entire HCA content in extreme cases, such as eggplant or
certain blueberries. In some cases (e.g., pineapple, white currants, savoy cabbage, faba bean pod, spinach), p-coumaric acid is predominant, and in rare
casesin a few varieties of raspberry, for exampleonly traces of caeic acid
are found, whereas the other HCAs are dominant. Whereas ferulic acid usually
forms only a small percentage of total HCA in fruits and vegetables, it can reach
and even exceed 50% in peppers, some citrus, and some white grape cultivars
[27]. Sinapic acid has been reported more rarely in fruits and is generally only
observed as traces [31,32], whereas it may be abundant in various Brassica
vegetables (Table 3).
The balance of the various HCA conjugates also characterizes fruit and
vegetable species and cultivars. Thus, the HCA quinic ester patterns of stone and
pome fruit dier considerably: the 3-isomers are major constituents in cherry
and plum, whereas the 5-isomers are principally found in apple and pear
18
[[1,2,65,69,70] (Table 2)]. An identical dierence is observed between Brassica

(rich in 3-isomers) and Asteraceae (rich in 5-isomers) vegetables (Table 3). In
most cases (e.g., apple, tomato, artichoke, carrot), glycosylated derivatives are
distinctly less abundant than quinic esters, whereas the opposite proportion is
more rarely observed (kale, raspberry). The relative proportions of glucose
esters and glucosides of both HBAs and HCAs are also variable with the
dierent species of fruits and vegetables (Tables 1, 2, and 3).
HCA ester content can be selected, among other parameters, to discriminate between grape species, but the most reliable criterion when comparing
cultivars of the same species appears to be the percentage of each HCA, as
shown in the case of V. vinifera, where the percentage of p-coumaroyl and
caeoyltartaric esters can be used to discriminate between varieties for
taxonomic purposes [27].
B.
Changes with Tissular Localization
The highest levels of HCA derivatives are frequently found in the external part of
eshy fruits, as shown for chlorogenic acid in pear and peach and 3-caeoylquinic acid in cherry [2,70,71]. On the contrary, chlorogenic acid is often more
abundant in the core than in the peel of apple, although this distribution depends
on the cultivar [67,72]. Tomato is one of the better-known examples of HCA
distribution: quantity of quinic esters in ripe fruit was found to be higher in the
pulp than in the pericarp, whereas the opposite was found for glucose derivatives
[2]. Tissue compartmentation of p-coumaroylglucose, caeoyl, and 3,4-dimethoxycinnamoyl glycosides makes it possible to discriminate clearly between the
placenta and the pericarp of Capsicum frutescens [15]. In grape, although the
level was always higher in skin than in pulp, the percentage of caeoyltartaric
acid was higher in the pulp, whereas the opposite was true for p-coumaroyltartaric acid [27]. p-Coumaroylgalactaric and feruloylgalactaric acids are also more
abundant in the outer part of avedo and albedo of citrus [2], and ferulic and
sinapic acid concentrations are higher in sour orange peel [73]. Distribution of
HCA derivatives is even more complex in certain cases, such as pineapple: in
addition to the gradients between the inside and outside, there are very important
longitudinal gradients, probably related to dierent stages of maturity of each of
individual fruits that make up the pineapple [2].
The outer part of many plant organs consumed as vegetables also shows
the highest concentrations of HBA and HCA conjugates. For example, puree
from nonpeeled carrot roots contained 104 mg/kg chlorogenic acid, whereas
only 28.3 mg/kg were found after removing approximatively 2 mm of periderm
tissue [74]. In potato tuber, about 50% of the phenolic compounds (mostly
chlorogenic acid and other mono- or di-caeoylquinic esters) were located in the
peel; the remainder decreased in concentration from the outside to the center of
19
tubers [11]. In cereals, bran always presents the highest antioxidant activity,
which is due to the localization of bound phenolic acids in the grain: the outer
layers (husk, pericarp, testa, and aleurone layer) contain the greatest concentrations of total phenolics, whereas concentrations are considerably lower in the
endosperm. About 80% of ferulic acid of both rye and wheat grain was found in
the bran [25,75].
C.
Changes with Physiological Stage
Concentrations of phenolic acids in a plant organ result from a balance between

biosynthesis and further metabolism, including turnover and catabolism. Considerable variations are generally observed in the amount of phenolic acids according
to the physiological stage when plant organs are picked up to be consumed or
transformed by humans. This may concern each type of organ (leaves, owers,
stalks, tubers, roots, etc.), but the most spectacular cases are those of fruits, as
considerable variations in phenolic compounds occur during maturation.
Concentrations of soluble forms of phenolic acid conjugates (expressed
per unit of fresh or dry weight) are generally highest in young fruits, with a
maximum during the early weeks after blossoming and a rapid decrease during
fruit development [2,27]. In dierent apple cultivars, for example, maximal
concentrations of chlorogenic acid, p-coumaroylquinic acid, and p-coumaroylglucose were found in very young fruits, followed by a constant decrease [2,72].
These changes make it possible to divide the life of a fruit into two main periods.
During the rst (approximately 2 months in apple), HCA derivatives accumulate
in the fruit with a positive balance among in situ biosynthesis, migration, and
possible reutilization. However, in the second period, this balance becomes
negative and the overall HCA content in the fruit falls.
Loquat fruit represents an exception as the concentration of chlorogenic
acid increased during maturation and became more prominent than neochlorogenic acid and all other phenolic compounds identied in this fruit [24]. The
late accumulation of chlorogenic acid in this fruit results from the activation of
its metabolism and especially from an increase in the enzyme activities of
phenylalanine ammonia-lyase, 4-coumarate:CoA ligase, and hydroxycinnamoyl
CoA:quinate hydroxycinnamoyltransferase. Consequently, the metabolism of
chlorogenic acid may be considered to be a biochemical marker for maturation
of loquat fruit.
In certain fruits the disappearance of phenolic acids may occur in relation
to the biosynthesis of other phenolic compounds, for example, the decrease in
hydroxycinnamic conjugates during growth and maturation of Vitis vinifera
berries and the rapid increase in anthocyanin levels that occurs at the same time
in the red cultivars [2]. In fruits of chili pepper (Capsicum frutescens) the onset
of capsaicinoid accumulation and ligninlike material parallels the disappear-
20
ance of the three cinnamoyl glycosides, which may be considered a source of

precursors in capsaicinoid biosynthesis [15].
Quantitative changes are sometimes accompanied by qualitative ones.
Thus, in tomato (cv. cerasiforme), most HCA conjugates appear during growth,
and ripe fruit contains 11 dierent conjugates, whereas very young green fruit
contains only chlorogenic acid [2]. Some compounds are characteristic of a
physiological stage: chlorogenic acid forms 76% of total HCA derivatives in the
unripe fruit, then falls to 15% in ripe fruit. By contrast, HCA glucosides form
23% and 84% during the same periods, a nding that may suggest certain
metabolic relations between these compounds. Thus, growth and matutation of
the tomato fruit are characterized by dierent expressions of the metabolism of
HCA derivatives. In growing pulp it is mainly oriented toward the accumulation
of quinic derivatives, whereas glucose derivatives (particularly glucosides)
accumulate in the pericarp during maturation. A good correlation between
variations in activities of enzymes of phenylpropanoid pathway and accumulation of phenolic compounds is observed in tomato. These data led to the notion
that phenolic acids and their metabolism may be suitable markers of maturation.
Variations in the phenolic acid levels during growth and development of
vegetables and cereals are not so homogeneous as those reported for eshy fruits.
It depends mainly on two parameters: the nature of the plant organ (leaves,
tubers, grains, etc.) and the subcellular localization of the phenolic conjugates
that accumulate either as soluble forms in the vacuole or linked to the cell wall.
During the development of soft or durum wheat grains, ferulic acid is
mainly present as bound forms and accumulates during the milk stage, in
correlation with high activities of phenylalanine and tyrosine-ammonia-lyases
[76,77]. A decrease in ferulic acid level is then observed during grain ripening,
but this may be due in part to the formation of alkali-resistant bonds in crosslinked polymers in cell walls, parallel to the progressive decrease in the grain
water content. Peroxidases should play a role in the formation of these covalent
linkages [56] as their activity occurs even in the last stages of grain ripening
[77]. Changes in the ratios of dierent dehydrodiferulic acids were also observed
in the cell walls during the growth period of sugar beet root along with a
decrease of more than 50% [78]. These changes may be related to the major
expansion of the storage root during the latter part of the growth period.
The integration of phenolic metabolism in the program of plant development raises the question of the possible role of these substances in physiological
regulations. They have sometimes been implicated in the control of growth,
maturation, and abscission [2,64], but these aspects are rather speculative and
are not discussed here. HCA derivatives and coumarins may act as in situ
inhibitors of seed germination in berries or other eshy fruits, either directly or
indirectly through the control of oxygen consumption [79]. It is also well known
that various phenolic compounds play a role in the interactions between plants
21
and microorganisms [64], and allelopathic eects of HBA and HCA have been
reported in wheat root exudates [80]. Furthermore, they clearly participate in the
resistance of plants to biological and environmental stresses [64] and play an
important role in the browning capacity of plant organs [2].
D.
Changes with Environmental Factors
Secondary metabolism, and in particular phenolic metabolism, largely depends

on external factors such as light, temperature, and various stresses [64]. Although
this has been mainly studied in relation to avonoid production, we report here
only information related to phenolic acids.
Caeoyl quinic ester concentrations of potato tubers steadily increased
after light exposure, whereas they decline during prolonged storage in the dark
[81,82]. Rates of accumulation were inuenced by cultivar, storage period, and
light source. Furthermore, ratios of 5-:4-:3-caeoylquinic esters were modied,
but this had no eect on blackening of tubers. In Vitis vinifera berries, some
phenolic components of aroma (e.g., methyl vanillate, methyl syringate) were
signicantly less abundant after bunch shading, whereas this treatment did not
modify some other compounds (e.g., methyl salicylate, 4-vinylphenol) [83].
g-Irradiation treatment degradates phenolic acids in a rst step, but it later
stimulates phenylalanine ammonia-lyase and the biosynthesis of p-coumaric
acid, avanones, and avones in clementines [84]. p-Coumaric content was then
particularly high in irradiated fruits after 49 days of storage at 3 jC and could be
related to better resistance to pathogens. Similar data have been obtained in
potato: irradiation of tubers to inhibit sprouting caused an initial reduction in
phenolics, but an increased formation of chlorogenic acid and its isomers was
later observed during storage [11].
Two major types of observations reveal relations between phenolic acids
and temperature. First, various data link the overall eects of climate and local
environmental conditions with the accumulation of phenols. Another group of
results is taken from the frequent use of low temperatures in postharvest storage
of fruits and vegetables. In this case, physiological disturbances may occur even
when temperatures are maintained above 0 jC. Such eects are generally
referred to as chilling injury, and they frequently take the form of discoloration,
for which phenolic compounds may be directly responsible. Several examples
show great variations in phenolic compounds or in the enzymes of their
metabolism during cold, but the changes vary greatly, depending on the species
and cultivar [2]. Cold storage of apples did not induce important variations in
their chlorogenic acid content for up to 9 months, and the health benets of
phenolics should be maintained during long-term storage, although the response
may slightly dier with cultivars [66,67]. On the contrary, numerous other
examples indicate an increase in phenolic acids during cold storage: chlorogenic
22
acid in Anjou pear [2], ellagitannins and p-coumaroylglucose in strawberries

[19], a sucrose ester of ferulic acid in the peel of beetroot [53], chlorogenic acid
in dierent cultivars of potato tubers stored at 0 jC [11,85], whereas a decrease
was also observed at 5 jC in other cultivars [82]. p-Coumaric and caeic
derivatives also accumulate in pineapple during storage at 8 jC, and the content
in phenolic acids was multiplied 10-fold 15 days after the fruits were returned at
20 jC [2]. In tomato, a fairly specic action of low-temperature storage was
observed on chlorogenic acid metabolism. Among the enzymes tested, levels of
phenylalanine ammonia-lyase and hydroxycinnamoyl-quinate transferase, two
enzymes that allow synthesis of chlorogenic acid, increased considerably, in
relation to chlorogenic acid accumulation [2].
Storage of bean seeds at high temperature (35 jC) and humidity causes
textural defects along with an increase in free phenolic acids (caeic, pcoumaric, ferulic and sinapic acids), a decrease in soluble esters, and a strong
increase in ferulic acid bound to soluble pectins [59]. These modications result
in poor soaking imbibition of seeds and in prolonged cooking time.
Phenolic acids are directly implied in the response of plant organs to
dierent kinds of stresses [64]: mechanical (wounding), chemical (various types
of treatment), or microbiological (pathogen infection). Phenolic acids are
involved in resistance in two ways: (1) by contributing to the healing of wounds
by lignication of cell walls around wounded zones [86] and (2) through the
antimicrobial properties demonstrated for many of them [87]. The compounds
involved can be classied in three groups: (1) some are already present in the
plant, and their level generally increases after stress; (2) others are formed only
after injury but are derived from existing substances by hydrolysis or oxidation;
(3) still others are biosynthesized de novo and can be classied as phytoalexins.
The eect of wounding has been particularly well studied in fruits [2] and
in minimally processed fruits and vegetables [88]. The most immediate response
to wounding is the oxidation of preexisting phenolic compounds and hence their
degradation. Thus, the chlorogenic acid content of tomato fruit pericarp falls for
6 h after wounding, and activation of phenolic metabolism occurs later, with an
increase in phenylalanine ammonia-lyase activity and consequent accumulation
of chlorogenic acid, feruloyl-, p-coumaroyl, and sinapoyl-glucose [2]. Such an
accumulation of caeoylquinic acids or caeoyl-and dicaeoyl tartaric acids is
also observed in shredded carrots and in minimally processed lettuce leaves [89
91], although its intensity may depend on storage conditions, either in air or in
controlled atmospheres. Along with polyphenoloxidase and peroxidase activities, this increase in phenolic substrates is responsible of the browning of
wounded tissues that shortens storage life of the product. Finally, the third aspect
of response to wounding is the formation of healing tissues (wound lignin or
suberin) that protect plant organs from water loss and also form a physiological
barrier that prevents possible penetration of pathogens [2,11].
23
The increase in phenolic content with pathogen infection has been well
documented in cell suspension cultures, especially at a molecular level [64]. It
was postulated that as a defense mechanism, ferulic and p-coumaric acids are
esteried to wall polysaccharides, possibly rendering the wall resistant to fungal
enzymes either by masking the substrate or by altering the solubility properties
of these wall polysaccharides [56,64]. Phenolic acids also possess antimicrobial
properties, and chlorogenic acid and related compounds may function to arrest
Molinia fructicola, in quiescent infections associated with immature and ripening peach fruits [92]. Chlorogenic and p-coumaroylquinic acids in apple are
inhibitors of both Botrytis sp. spore germination and mycelial growth [2]. When
their eects on growth of certain fungi are compared, p-coumaroylquinic is
more inhibitory than chlorogenic at the same concentrations for B. cinerea and
Alternaria sp., whereas P. expansum is less sensitive. Both quinic derivatives are
stimulatory at low concentrations for Botrytis and Penicillium sp.
The acquisition of antimicrobial properties by phenolic compounds may
derive from oxidation or hydrolysis. First, o-quinones are generally more active
than o-diphenols and browning intensity is often greatest in highly resistant
plants, suggesting that black and brown pigments contribute to resistance.
Hydrolysis may be carried out by fungal pectic enzymes, as suggested by the
appearance of free p-coumaric, caeic, and ferulic acids in apple infected with
P. expansum [2]. In this case, the damage does not cause much browning around
the infection site because of the inhibition of the phenolase system by acids
released after hydrolysis of chlorogenic and p-coumaroylquinic acids in the
fruit. Again in apple, antifungal compounds, such as 4-hydroxybenzoic acid
produced after infection with Sclerotinia fructigena, are thought to be derived
from the transformation of chlorogenic acid by the fungus. Free phenolic acids
are the best inhibitors of growth of the fungi appearing during the postharvest
storage, and their structure/activity relationships have been studied in vitro [87].
An additional methoxy group caused increased activity of HBA and HCA
derivatives. Thus, ferulic and 2,5-methoxybenzoic acids showed a strong
inhibition against all fungi tested.
Certain phenolic compounds can be biosynthesized de novo after infection. These compounds, which do not exist before infection and which have
antimicrobial properties, are called phytoalexins. They are produced by plants as
defense mechanisms in response to microbial infection [64,93], but the accumulated compounds are often avonoids or coumarins, although benzoic acid
itself has been shown in apple after infection.
Both constitutive phenolic acids and phytoalexins may be involved in the
resistance of potatoes to Erwinia sp. [94]. This resistance could result from the
increase of caeic, chlorogenic, and ferulic acids and the formation of suberized
barriers after wound injury or pathogen attacks [11,63,94]. Both polyesterforming HCAs and ligninlike monolignols of potato suberin constitute a dense
24
covalent network capable of repelling water and protecting the cell wall from
pathogenic attack [63]. The involvement of phenolic acids in potato resistance is
nevertheless selective as they protect against some but not all pathogens:
although levels of chlorogenic acid increased after infection of potato tubers
with the fungus Phytophtora infestans, no dierences in the levels were observed in resistant or nonresistant cultivars [95].
V.
ROLE OF PHENOLIC ACIDS IN THE ORGANOLEPTIC

AND NUTRITIONAL QUALITY OF FRUITS
AND VEGETABLES
Phenolic acids contribute to the sensory and nutritional qualities of fruits,

vegetables, and derived foods. Directly or indirectly, they play a role in color,
astringency, bitterness, and aroma, and they also are of great interest to
humans, because of their antioxidant capacity [2,4].
A.
Phenolic Acids and Food Flavor
As reported, acylation of anthocyanins with p-coumaric and caeic acids is

common in fruits, and it is responsible for better color stability in fruit products
[35]. For example, it has been shown that the dierence of stability to light and heat
of dierent Sambucus species results from the degree of anthocyanin acylation
[96,97]. Diacylated anthocyanins are stabilized by the sandwich-type stacking
caused by hydrophobic interaction between the anthocyanidin ring and the two
aromatic acyl groups [98]. Intramolecular copigmentation involving p-coumaroylglucose units at three or four positions of delphinidin is responsible for the
exceptionally deep blue color of Daniella sp. berries [37]. Furthermore, numerous
avonoids and HCA derivatives play a role in the intermolecular copigmentation
by stabilizing the pigment in its colored form and being the cause of a bathochromic shift and of an increase in the absorbance in the visible band [98].
The color of plant organs may also be strongly modied by the appearance
of brown compounds, which generally result from the enzymatic oxidation of
phenolic compounds including caeic esters [2,99]. These melanin-type pigments may appear naturally during maturation of certain fruits, but they generally
occur after wounding and crushing of plant organs. The resultant discoloration
aects both commercial quality and nutritional parameters. These aspects are
discussed late in relation to the processing of fruits and vegetables.
The astringency of fruits results from the interaction of salivary proteins
with tannins or other phenolics. Although chlorogenic itself has sometimes been
reported to be astringent, HBAs play a major role as they participate in the
formation of hydrolysable tannins [100]. For example, ellagitannins (ellagic acid
25
esters of glucose) are responsible for the strong astringency of various fruits,
e.g., pomegranate, persimmon, chestnuts, and fruits of Rosaceae. Furthermore,
in rather rare cases, HBAs are also present in condensed tannins in the form of
epicatechin-gallate [27].
The role of phenolic acids in the bitterness of fruit and fruit products is
still a matter of discussion, but it was concluded that HCAs do not play any role
in the taste of wines, even at high concentrations of caftaric acid and
glutathionylcaftaric acid [27]. Verbascoside may contribute to bitterness in
olives, but its concentration is always low in comparison to oleuropein concentration [2]. Phenylpropanoid sucrose esters with several acetyl groups are also
responsible for the bitter taste of stone fruits of Prunus sp. [101].
The importance of phenolic acids in fruit aroma is low, though many
simple aromatic phenols may be released by enzymatic or chemical reactions
from glycosylated precursors during maturation or processing, e.g., in vanilla,
passion fruit, mango, and apricot [102]. Such transformations are also at the
origin of some aroma constituents in wines, ciders, and fruit juices, through the
degradation of HCA conjugates. Vanillic acid participates, in addition to
vanillin, in vanilla aroma, and cinnamaldehyde is the principal component of
cinnamon avor [103]. Ferulic acid is a potential precursor of o-avors in
stored citrus juice, and pasteurization increases both the release of free ferulic
acid from bound forms and the formation of p-vinyl guaiacol [29].
B.
Phenolic Acids as Antioxidants
Antioxidants play an important role in antioxidant defense mechanisms in biological systems, protecting lipids both in cells and in food products and having
inhibitory eects on mutagenesis and carcinogenesis. Attention is now focused on
natural antioxidants, since the use of synthetic antioxidants has been falling o
because of their suspected action as cancer promotors [104]. Most natural
antioxidants present a polyphenolic structure, and it is signicant that most papers
published since the early 1990s about the characterization of phenolic compounds,
and especially phenolic acids, concern their antioxidant activity (see reviews
104109; see also Ref. 110) and the dierent chapters of the present volume.
Along with numerous other phenolic compounds, hydroxycinnamates and gallic
acid derivatives (methyl and lauryl esters, propylgallate) act as free radical
acceptors and show strong antioxidant properties [4,6,50,105,108].
Many fruits and vegetables (e.g., grape, citrus, olive, black pepper, spices,
soya, cereals) and the derived foods and beverages are a good source of phenolic
antioxidants and constitute an important part of our daily diet [4,5,25,26,107
109,111115]. A good correlation between phenolic content and antioxidant
activity is often observed, as reported for monomeric and dimeric hydroxycinnamates of rye bran [75] and various caeoyl quinic esters in peach puree [116],
26
tart cherries [117], and prunes and prune juice [69,118]. Regular consumption of
phenolic antioxidants may provide protection against diseases, including cancer
and cardio-and cerebrovascular diseases [110], and it increases the serum
antioxidant capacity in humans, as shown after consumption of strawberries,
spinach, or red wine [119].
In addition to avonoids, HCA derivatives protect food products from
oxidation: for example, the remarkable stability of virgin olive oil is directly
related to its phenolic antioxidants [104,120,121]. They are also widely used as
food antioxidant additives to protect lipid structures [4,6] and are good
candidates for successful employment as topical protective agents against
ultraviolet (UV) radiationinduced skin damage [122]. Agricultural and industrial residues and by-products of plant origin (e.g., potato peel waste, grape
seeds, olive, apple or cranberry pomace, citrus peels and seeds) are attractive
and cheap sources of natural antioxidants [123125]. An extensive review of
these last aspects, including the inuence of processing conditions, has been
published [108], and it clearly shows that antioxidant capacity is often associated with the presence of phenolic acids (Table 4).
The relationships between chemical structures of HCA conjugates and
their antioxidant and free radical scavenging activity have been reported
[5,50,106,108,110,126]. Although this may vary with temperature and the nature
of the test used, antioxidant activity was always higher for free caeic acid than
Table 4 Main Phenolic Acids in Crude Extracts from

Agroindustrial Wastes
Residue
Durum wheat bran
Corn bran hemicellulose
Oat hulls
Potato peel extract
Apple pomace
Lemon seeds
Sweet orange seeds and peel
Citrus peel molasses
Grape marc
Grape pomace
Grape seeds
Cranberry pomace
a
Identified compoundsa
p-HB, P, G, V, Sy, C, 5-CQ, p-C, F
p-C, F, diF
p-HB, V, p-C, F
P, G, C, 5-CQ
5-CQ
C, p-C, F, S
C, p-C, F, S
p-C, F, S, FP
P, G, V
G, G 3-glucoside, G 4-glucoside, p-CT, CT
G, p-C
p-HB, G, p-C, 5-CQ
Phenolic acids: P, protocatechuic; p-HB, p-hydroxybenzoic; G, gallic; V, vanillic; Sy,

syringic; p-C, p-coumaric; p-CT, coutaric; C, caeic; 5-CQ, 5-caeoylquinic
(chlorogenic); CT, caftaric; F, ferulic; diF, diferulic; S, sinapic; FP, feruloylputrescine.
Source: Refs. 108,124,125.
27
for its glucose or quinic esters (e.g., chlorogenic acid), whereas it was lower for
ferulic acid [50,75]. For the less active HCAs, p-coumaric and ferulic acids,
esterication to tartaric acid may enhance ability to inhibit low-density lipoprotein (LDL) oxidation [127]. The presence of hydroxyl groups in the ortho
position increases antioxidant activity, as shown for caeic and rosmarinic
acids. Nevertheless, a limitation of the utilization of HCAs and their natural
esters as lipid protectors is their low liposolublity, and propyl esters have been
proposed to prevent oxidative deterioration of edible oils [128].
Although phenolics are generally considered good antioxidants, gallic acid
and its derivatives have also been reported simultaneously to exert pro-oxidant
eects on various biological molecules, and their consumption should be
regarded with caution [108,129].
VI.
CHANGES DURING FOOD PROCESSING
As reported, phenolic acids contribute to the organoleptic and nutritional qualities

of fruits and vegetables, but they also play an important role in the acquisition of
the sensory properties of derived products, fermented or not (e.g., juices, wines,
beers, ciders, purees, jam, jelly, minimally processed fruits or vegetables, oils). In
all cases, the phenolic prole is dierent from that of the original organ, as
qualitative and quantitative modications occur during processing. They mainly
result from three set of reactions: (1) the oxidative degradation of phenolic acids,
including enzymatic browning; (2) the release of free acids from conjugate forms;
and (3) the formation of complex structures of phenolic acids and other
chemicals, in particular proteins, tannins, and anthocyanins.
A.
Phenolic Acids and Enzymatic Browning
Enzymatic browning is often observed during fruit and vegetable processing, and
it results from an initial cellular decompartmentation. The formation of yellow
and brown pigments is controlled by phenolic compound levels, the presence of
oxygen, and polyphenoloxidase (PPO) activities. PPO is a membrane-bound
enzyme that catalyzes the hydroxylation of monophenols to o-diphenols and their
oxidation to quinones. Chlorogenic acid and its isomers in numerous fruits, and
other caeic esters in many others (e.g., caeoylshikimic in date, caeoyltartaric in
grape), are easily degraded by PPO [2,27,65,99], and it seems that esterication of
the carboxyl group of caeic acid with quinic acid or tartaric acid leads to an
increase in PPO activity in practically all cases. In fruit juice processing, unit
operations such as crushing, prepressing, enzymatic treatments, and pressing
provide opportunities for PPO activities, and HCAs and catechins are mainly
aected [130]. The rst PPO reaction products are quinone-colored substances,
28
which then rapidly condense and do or do not combine with amino or sulfhydryl
groups of proteins [65]. The relatively insoluble brown polymers formed are
generally eliminated when they disturb the quality of the product, such as fruit juice.
The oxidation of grape caftaric acid during vinication or juice preparation has drawn the attention of many researchers. It is now established that
oxidation of caftaric acid by PPO leads to the formation of 2-S-glutathionylcaftaric acid [27]. This compound results from a four-component reaction
involving caftaric acid, PPO, oxygen, and glutathione. It seems likely that the
presence of glutathione or cysteine derivatives of sinapyl alcohol in pineapple
juice results from similar oxidative reactions during juice processing [131]. As
S-glutathionylcaftaric acid is not inherently colored and is not a substrate for
PPO, its formation is therefore believed to limit most browning by trapping
caftaric acid quinones in the form of a stable glutathione substituted product.
Nevertheless, it may also be oxidized by fungus laccase during vinication and
may induce coupled oxidations of other grape phenolics.
In comparison to that of caeoyl conjugates, the contribution of pcoumaric, ferulic, and sinapic derivatives to browning is rather low, and they
even have been reported to act as PPO inhibitors [2]. Many other phenolic
compounds present in large quantities in fruits and vegetables (e.g., anthocyanins, avonol glycosides) do not appear to be direct substrates of PPO, but
chlorogenic acid or other caeoyl conjugates may participate in their coupled
oxidation, which increases the browning intensity [27,65,99]. Thus, apple
procyanidins are not themselves substrates for PPO, but they are oxidized by
a chlorogenic acid/chlorogenoquinone redox shuttle while the amount of
chlorogenic acid remains almost constant. Enzymatic oxidation of chlorogenic
acid may also be combined with nonenzymatic degradation of anthocyanins as
shown in eggplant, sweet cherry, and dEnte plum.
In addition to the enzymatic oxidation of phenolic acids, their autooxidation can also result in brown-colored pigments that are detrimental to food
quality. It is dependent on the physicochemical environment and strongly
increases with pH. For example, auto-oxidation of phenolic acids was responsible
for color loss of carrot puree, and processing treatments that reduce residual
oxygen may result in better color retention after processing and storage [132].
Enzymatic oxidation or auto-oxidation may also aect potato tuber either
during cold storage or during processing and cooking. Reviews on the chemical,
biochemical, and dietary roles of potato polyphenols have been published
[11,133]. They focus on chlorogenic acid, as this compound constitutes up to
90% of the total phenolic content in potato tuber. No correlation was found
between bruising-induced browning of potato tuber and chlorogenic acid, as
tyrosine and PPO activity plays the major role ([11] for a review). On the
contrary, chlorogenic acid seemed to be responsible for the bluishgray discoloration of boiled or steamed potatoes after exposure to air. This after-cooking
29
darkening appears to be due to the oxidation of a colorless ferrous ionchlorogenic acid complex to a dark ferric one [133].
Oxidation of chlorogenic acid during processing results in a dramatic loss in
concentration: only traces were recovered in an oxidized apple juice, whereas 14,
23, or 32 mg/100 mL was present in conventionnal juice, a rapidly pasteurized
juice, and nonoxidized juice, respectively [2]. As the antioxidant capacity of
apple and apple products is due to phenolic constituents rather than to ascorbic
acid [111,134], the chain-breaking activity decreased in apple puree when
chlorogenic acid was oxidized and brown polymers accumulated [135]. Inhibition
of browning during processing may then protect both the natural coloration of the
product and its antioxidant properties. This may be achieved by physical,
chemical, genetic, or molecular methods, and numerous treatments have been
proposed to prevent browning [27,65,99,136138]. The most widely used in
industry are heat, ascorbic acid, and sulte adjunctions, but use of sulting agents
has been limited and now banned because of health risks. These aspects are not
discussed further here, as they do not specically concern phenolic acids.
In contrast to the previous examples, oxidative treatments may promote
positive consequences in some other cases, as they decrease concentrations in
phenolic acids (and other phenolics) and limit subsequent oxidations. Therefore,
hyperoxidation of musts during wine processing has sometimes been proposed,
and this process particularly aected HCA derivatives [27]. In addition, a
peroxidasic treatment allowed the dimerization of ferulic acid in wheat bran
tissues and consequently decreased arabinoxylan solubility and increased
mechanical strength of the tissue [58].
B.
Changes in the Phenolic Acid Profile
Although free phenolic acids are rarely present in plant organs, they have often
been detected in derived products, e.g., fruit juices, wines, ciders, and oils
[2,27,69,118,130,139,140]. It is assumed that these free acids are formed by
partial degradation of the combined forms during extraction or processing. For
example, removal of olive bitterness requires treatments with sodium hydroxyde,
and, during this process, caeic acid has appeared, directly derived from alkaline
hydrolysis of verbascoside [141]. Hydrolysis of esteried forms of phenolic acids
may also be catalyzed by esterases, as shown during the preparation of prunes
and prune juice [69]. A decrease in bound phenolic acids (mainly ferulic acid)
was also observed during malting of cereals, and it may due to the action of
esterases that are induced during germination and act on various phenolic acid
esters linked to wall polysaccharides [142,143]. In addition to hydrolytic enzymes
issued either from the plant material or from the microrganisms naturally present
on the plant cuticle, commercial pectolytic enzymes are often used in fruit juice
processing to improve clarication, and they also can cause hydrolysis of HCA
30
esters [130]. HCA conjugates may also be transformed by endogenous microorganisms to cause o-avor in food products. For example, yeasts isolated from
unpasteurized apple juice are responsible for the formation of 4-vinyl guaicol as
they contain both feruloyl esterase and ferulic acid decarboxylase activity [144].
Analysis of wine phenolic acids reveals two important dierences in
comparison with the original grapes: (1) lower content of HCA esters and (2)
appearance of free p-hydroxybenzoic, gallic, p-coumaric, and caeic acids
[2,27]. The nal phenolic composition of a wine mainly depends on three sets
of parameters: (1) the phenolics initially present in grapes; (2) the extraction and
winemaking techniques, including fermentations; and (3) the chemical reactions
that occur during aging either in wood barrels or in bottles. In particular, when
seed crushing and long extraction times were employed, high levels of HBA
derivatives and S-glutathionylcaftaric acid were present in wines [145]. Whereas
red wines contain high levels of HCA conjugates, anthocyanins, and condensed
tannins, the phenolic prole of white wines essentially results from the presence
of avonols and phenolic acid derivatives [27]. Twelve new benzoic and
cinnamic derivatives (caeol ethyl tartrates, glucose esters of cinnamic, pcoumaric, ferulic, or vanillic acids; 4-O-glucosides of ferulic acid; ethyl
protocatechuate, or gallate) were identied in Riesling wine, where they are
responsible for antioxidant activity [146]. Nevertheless, the highest antioxidant
capacity was obtained for red wines, because of their high content of catechins
and related structures [147,148].
In addition to microbial or chemical processes of clarication, physical
ones may also have a great inuence on the phenolic prole of fruit juices,
especially for HCA. A signicantly higher oxidation level was observed on
ultraltration membranes than on microltration membranes of apple juices
[149]. Furthermore, chlorogenic acid content was one of the most relevant
variables to dierentiate juices claried according to the two membrane ltration
technologies. Important changes in the composition of the phenolic fraction of
kiwi was observed during processing [150]. In contrast to juice subjected to a
high-temperature short-period treatment, in pressed juice low levels of HCA and
HBA were found (respectively, 1.11 and 0.17 mg/L chlorogenic acid). Reduction
in the level of phenolic acid derivatives also occurred after clarication (0.71 mg/
L in comparison to 1.11 before clarication).
Thermal treatments also cause changes in the phenolic acids of food
products, as they accelerate chemical degradation and either promote or block
enzymatic activities. For example, the roasting of coee beans and the baking of
potatoes or cereal our induce a signicant loss in chlorogenic content [4,11]. In
the same way, the red-brown color of semidried pickled plums is obtained by
thermal processing that induces the oxidation of the three isomers of chlorogenic
acid [151]. In the case of quince, jellies contained lower concentrations of
chlorogenic acid than jams, because of the more severe thermal treatment to
31
which the fruit is subjected during the preparation of jelly [152,153]. An

increase in ellagic acid, explained by a release from ellagitannins, was also
observed during the thermal treatment of red raspberry to prepare jams [18],
whereas freezing and long-term frozen storage of raspberries induced a signicant decrease in the ellagic content [17].
Qualitative and quantitative changes in phenolic acids also occurred
during long-term storage of fruit juices or juice concentrates and extensive
degradation was observed at 25 jC [130]. Refrigerated storage of orange juice
also induced modications in the phenolic pattern, as HCA derivatives, in
contrast to avanones, remained in soluble form and were not present in the
cloud that appears in juice [154].
Phenolic compounds have sometimes been used to detect adulterations of
fruit juices and jams. Most of them are avonols, but some HCA derivatives can
be used since they are typical of some fruit species such as tartaric derivatives in
grape. For example, grape juice can be detected by the presence of caeoyl-,
p-coumaroyl-, and feruloyl-tartaric acids, whereas the presence of quinic esters of
HCA would imply adulteration with other fruits. A method has also been
developed for detection and quantitation of pulpwash, a lower-quality juice
product, in orange juice [9]: feruloyl and sinapoyl glucose, in addition to dierent
other phenolic compounds, were present in much larger amounts in pulpwash.
VII.
CONCLUSIONS
Considerable progress in the analysis of phenolic acid derivatives has been made
since the 1980s, and the diversity observed in eshy fruits and vegetables with
regard to their distribution is found in terms of quality and quantity. The raison
detre of phenolic acids and, more generally, of the totality of phenolic
compounds, is now better understood, and their role in the interface between
the plant and its environment has now been well established. Furthermore,
phenolic acids have great importance, on the one hand, as precursors for many
other phenolic molecules often found in fruits and vegetables (e.g., anthocyanins,
tannins, lignin) and, on the other hand, to the organoleptic and nutritional quality
of plant products that play an important role in the daily diet. The most important
biological activity of phenolic acids, along with other phenolic compounds, is
probably that of their antioxidant property and, consequently, their inhibitory
eects on mutagenesis, carcinogenesis, and human heart diseases. As shown,
many papers published since the early 1990s about phenolic acids in fruits and
vegetables concern their antioxidant activity. Nevertheless, frequently only
global activity has been reported, and more data are still necessary to explain
individual eects and synergistic interactions, e.g., by determining relative
health-promoting properties of the dierent chlorogenic acid isomers.
32
Modication of the phenolic pattern of fruits and vegetables can be

envisaged by means of plant selection both by conventional hybridization and
by use of new genetic engineering techniques in the near future. In light of their
potential role as antioxidants, as well as their inhibitory eect on plant pathogens,
phenolic acids may be viewed as positive constituents of fruits and vegetables.
Nevertheless, high phenolic acid concentrations may also be undesirable, as some
phenolic acids are good substrates to polyphenoloxidases and participate in
browning during storage or processing.
Genetic engineering, which is being pursued in plants for the manipulation
of plant secondary metabolism [64], might lead to the production of fruits and
vegetables in which the phenolic metabolism is over-or underexpressed. However, it must rst be determined whether plants with higher or lower phenolic
content are desired. The answer may depend on which of the following is
preferred: resistance to pathogens, improvement of organoleptic qualities, or
accumulation of one or several phenolics showing antioxidant properties and
interactions aecting human health. Suppression of genes involved in the
biosynthesis of polyphenoloxidase through antisense ribonucleic/acid (RNA),
which have already been obtained for potato [136], might lead to fruits and
vegetables with low browning capacity and a high level of monomeric phenolic
compounds with high antioxidant properties.
REFERENCES
1. Hermann K. Occurence and content of hydroxycinnamic and hydroxybenzoic acid
compounds in foods. Crit Rev Food Sci Nutr 1989; 28:315347.
2. Macheix JJ, Fleuriet A, Billot J. Fruit phenolics. Boca Raton, FL: CRC Press, 1990.
3. Macheix JJ, Fleuriet A. Phenolic acids in fruits. In: Rice-Evans CA, Packer L, eds.
Flavonoids in Health and Disease. New York: Marcel Dekker, 1998:3559.
4. Cliord MN. Chlorogenic acids and other cinnamatesnature, occurrence and
dietary burden. J Sci Food Agric 1999; 79:362372.
5. Robards K, Prenzler PD, Tucker G, Swatsitang P, Glover W. Phenolic compounds
and their role in oxidative processes in fruits. Food Chem 1999; 66:401436.
6. Kroon PA, Williamson G. Hydroxycinnamates in plants and food: current and
future perspectives. J Sci Food Agric 1999; 79:355361.
7. Ibrahim R, Barron D. Phenylpropanoids. In: Harborne JB, ed. Methods in Plant
Biochemistry: Vol. 1. Plant Phenolics. London: Academic Press, 1989:73111.
8. Shobha SV, Ravindranath B. Supercritical carbon dioxide and solvent extraction of
the phenolic lipids of cashew nut (Anacardium occidentale) shells. J Agric Food
Chem 1991; 39:22142217.
9. Cancalon PF, Bryan CR. Use of capillary electrophoresis for monitoring citrus
juice composition. J Chromatogr A 1993; 652:555561.
10. Tomas-Barberan FA. Capillary electrophoresis: a new technique in the analysis of
plant secondary metabolites. Phytochem Anal 1995; 6:177192.

11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
33
Friedman M. Chemistry, biochemistry, and dietary role of potato polyphenols: a

review. J Agric Food Chem 1997; 45:15231540.
Gorinstein S, Zemser M, Weisz M, Halevy S, Deutsch J, Tilis K, Feintuch D,
Guerra N, Fishman M, Bartnikowska E. Fluorometric analysis of phenolics in
persimmons. Biosci Biotech Biochem 1994; 58:10871092.
Yoshiki Y, Okubo K, Onuma M, Igarashi K. Chemiluminescence of benzoic and
cinnamic acids, and avonoids in the presence of aldehyde and hydrogen peroxide
or hydroxyl radical by Fenton reaction. Phytochemistry 1995; 39:225229.
Rapisarda P, Carollo G, Fallico B, Tomaselli F, Maccarone E. Hydroxycinnamic
acids as markers of italian blood orange juices. J Agric Food Chem 1998; 46:
464470.
Sukrasno N, Yeoman MM. Phenylpropanoid metabolism during growth and
development of Capsicum frutescens fruits. Phytochemistry 1993; 32:839844.
Ayaz FA, Kadioglu A. Changes in phenolic acid contents of Diospyros lotus L.
during fruit development. J Agric Food Chem 1997; 45:25392541.
Ancos de B, Gonzalez EM, Cano MP. Ellagic acid, vitamin C, and total phenolic
contents and radical scavenging capacity aected by freezing and frozen storage in
raspberry fruit. J Agric Food Chem 2000; 48:45654570.
Zafrilla P, Ferreres F, Tomas-Barberan FA. Eect of processing and storage on the
antioxidant ellagic acid derivatives and avonoids of red raspberry (Rubus idaeus)
jams. J Agric Food Chem 2001; 49:36513655.
Gil MI, Holcroft DM, Kader AA. Changes in strawberry anthocyanins and other
polyphenols in response to carbon dioxide treatments. J Agric Food Chem 1997;
45:16621667.
Sakushima A, Coskun M, Maoka T. Hydroxybenzoic acids from Boreava
orientalis. Phytochemistry 1995; 40:257261.
Ng A, Parker ML, Parr AJ, Saunders PK, Smith AC, Waldron KW. Physicochemical characteristics of onion (Allium cepa L.) tissues. J Agric Food Chem
2000; 48:56125617.
Mattila P, Konko K, Eurola M, Pihlava JM, Astola J, Vahteristo L, Hietaniemi V,
Kumpulainen J, Valtonen M, Piironen V. Contents of vitamins, mineral elements,
and some phenolic compounds in cultivated mushrooms. J Agric Food Chem
2001; 49:23432348.
Bradeld AE, Flood AE, Hulme AC, Williams AH. Chlorogenic acid in fruit trees.
Nature 1952; 170:168.
Ding CK, Chachin K, Ueda Y, Imahori Y, Wang CY. Metabolism of phenolic
compounds during loquat fruit development. J Agric Food Chem 2001; 49:2883
2888.
Kahkonen MP, Hopia AI, Vuorela HJ, Rauha JP, Pihlaja K, Kujala TS, Heinonen
M. Antioxidant activity of plant extracts containing phenolic compounds. J Agric
Food Chem 1999; 47:39543962.
Kahkonen MP, Hopia AI, Heinonen M. Berry phenolics and their antioxidant
activity. J Agric Food Chem 2001; 49:40764082.
Macheix JJ, Sapis JC, Fleuriet A. Phenolic compounds and polyphenoloxidase in
relation to browning in grapes and wines. Crit Rev Food Sci Nutr 1991; 30:441486.
34
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.

Oleszek W, Amiot MJ, Aubert SY. Identication of some phenolics in pear fruit.
J Agric Food Chem 1994; 42:12611265.
Naim M, Zehavi U, Nagy S, Rouse RL. Hydroxycinnamic acids as o-avor
precursors in Citrus fruits and their products. In: Ho CT, Lee CY, Huang MT, eds.
Series 506: Phenolic Compounds in Food and Their Eects on Health I. Analysis,
Occurrence, and Chemistry. Washington, DC: ACS, 1992:180191.
Mouly PP, Gaydou EM, Faure R, Estienne JM. Blood orange juice authentication
using cinnamic acid derivatives: variety dierentiations associated with avanone
glycoside content. J Agric Food Chem 1997; 45:373377.
Sakushima A, Coskun M, Tanker M, Tanker N. A sinapic acid ester from Boreava
orientalis. Phytochemistry 1994; 35:14811484.
Sakushima A, Coskun M, Maoka T. Sinapinyl but-3-enylglucosinolate from
Boreava orientalis. Phytochemistry 1995; 40:483485.
Yoshikawa K, Sugawara S, Arihara S. Phenylpropanoids and other secondary
metabolites from fresh fruits of Picrasma quassioides. Phytochemistry 1995; 40:
253256.
Sanbongi C, Osakabe N, Natsume M, Takizawa T, Gomi S, Osawa T. Antioxidative polyphenols isolated from Theobroma cacao. J Agric Food Chem 1998;
46:454457.
Mazza G, Miniati E. Anthocyanins in fruits, vegetables and grains. Boca Raton,
FL: CRC Press, 1993.
Price CL, Wrolstad RE. Anthocyanin pigments of royal Okanogan huckleberry
juice. J Food Sci 1995; 60:369374.
Bloor SJ. Deep blue anthocyanins from blue Dianella berries. Phytochemistry
2001; 58:923927.
zipek M, Calis I, Ertan M, Ruedi P. Rhammetin 3-p-coumaroylrhamninoside
O
from Rhamnus petiolaris. Phytochemistry 1994; 37:249253.
Glassgen WE, Metzger JW, Heuer S, Strack D. Betacyanins from fruits of Basella
rubra. Phytochemistry 1993; 33:15251527.
Alasalvar C, Grigor JM, Zhang D, Quantick PC, Shahidi F. Comparison of
volatiles, phenolics, sugars, antioxidant vitamins, and sensory quality of dierent
colored carrot varieties. J Agric Food Chem 2001; 49:14101416.
Plumb GW, Proce KR, Rhodes MJC, Williamson G. Antioxidant properties of the
major polyphenolic compounds in broccoli. Free Radic Res 1997; 27:429435.
Bilia AR, Fumarola M, Gallori S, Mazzi G, Vincieri FF. Identication by HPLCDAD and HPLC-MS analyses and quantication of constituents of fennel teas and
decoctions. J Agric Food Chem 2000; 48:47344738.
Azuma K, Nakayama M, Koshioka M, Ippoushi K, Yamaguchi Y, Kohata K,
Yamauchi Y, Ito H, Higashio H. Phenolic antioxidants from the leaves of
Corchorus olitorius L. J Agric Food Chem 1999; 47:39633966.
Ben Hod G, Basnizki Y, Zohary D, Mayer AM. Cynarin and chlorogenic acid
content in germinating seeds of globe artichoke (Cyanara scolymus L.). J Gen
Breed 1992; 46:6368.
Chuda Y, Suzuki M, Nagata T, Tsushida T. Contents and cooking loss of three
quinic acid derivatives from Garland (Chrysanthemum coronarium L.). J Agric
Food Chem 1998; 46:14371439.

46.
35
Kweon MH, Hwang HJ, Sung HC. Identication and antioxidant activity of novel
chlorogenic acid derivatives from bamboo (Phyllostachys edulis). J Agric Food
Chem 2001; 49:46464655.
47. Bergman M, Varshavsky L, Gottlieb HE, Grossman S. The antioxidant activity of
aqueous spinach extract: chemical identication of active fractions. Phytochemistry 2001; 58:143152.
48. Strack D, Hartfeld F, Austenfeld FA, Grotjahn L, Wray V. Coumaroyl-, caeoyland feruloyltartronates and their accumulation in mung bean. Phytochemistry
1985; 24:147150.
49. Cuvelier ME, Richard H, Berset C. Antioxidative activity and phenolic
composition of pilot-plant and commercial extracts of sage and rosemary. J Am
Oil Chem Soc 1996; 73:645652.
50. Chen JH, Ho C-T. Antioxidant activities of caeic acid and its related
hydroxycinnamic acid compounds. J Agric Food Chem 1997; 45:23742378.
51. Zheng W, Wang SY. Antioxidant activity and phenolic compounds in selected
herbs. J Agric Food Chem 2001; 49:51655170.
52. Bokern M, Heuer S, Wray V, Witte L, Macek T, Vanek T, Strack D. Ferulic acid
conjugates and betacyanins from cell cultures of Beta vulgaris. Phytochemistry
1991; 30:32613265.
53. Kujala TS, Loponen JM, Klika KD, Pihlaja K. Phenolics and betacyanins in red
beetroot (Beta vulgaris) root: distribution and eect of cold storage on the content
of total phenolics and three individual compounds. J Agric Food Chem 2000;
48:53385342.
54. Yu J, Vasanthan T, Temelli F. Analysis of phenolic acids in barley by highperformance liquid chromatography. J Agric Food Chem 2001; 49:4352 4358.
55. Faulds CB, Williamson G. The role of hydroxycinnamates in plant cell wall. J Sci
Food Agric 1999; 79:393395.
56. Iiyama K, Lam TBT, Stone BA. Covalent cross-links in the cell wall. Plant Physiol
1994; 104:315320.
57. Parr AJ, Ng A, Waldron KW. Ester-linked phenolic components of carrot cell
walls. J Agric Food Chem 1997; 45:24682471.
58. Peyron S, Abecassis J, Autran JC, Rouau X. Enzymatic oxidative treatments of
wheat bran layers: eects on ferulic acid composition and mechanical properties.
J Agric Food Chem 2001; 49:46944699.
59. Garcia E, Filisetti TMCC, Udaeta JEM, Lajolo FM. Hard-to-cook beans
(Phaseolus vulgaris): involvement of phenolic compounds and pectates. J Agric
Food Chem 1998; 46:21102116.
60. Norton RA. Isolation and identication of steryl cinnamic acid derivatives from
corn bran. Cereal Chem 1994; 71:111117.
61. Dimberg LH, Theander O, Lingnert H. Avenanthramides: a group of phenolic
antioxidants in oats. Cereal Chem 1993; 70:637641.
62. Lapierre C, Pollet B, Negrel J. The phenolic domain of potato suberin: structural
comparison with lignins. Phytochemistry 1996; 42:949953.
63. Yan B, Stark RE. Biosynthesis, molecular structure, and domain architecture of
potato suberin: a 13C NMR study using isotopically labeled precursors. J Agric
Food Chem 2000; 48:32983304.
36
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.

Dixon RA, Paiva NL. Stress-induced phenylpropanoid metabolism. Plant Cell
1995; 7:10851097.
Nicolas J, Richard-Forget FC, Goupy PM, Amiot MJ, Aubert SY. Enzymatic
browning reactions in apple and apple products. Crit Rev Food Sci Nutr 1994;
34:109157.
Golding JB, McGlasson WB, Wyllie SG, Leach DN. Fate of apple peel phenolics
during cool storage. J Agric Food Chem 2001; 49:22832289.
van der Sluis A, Dekker M, de Jager A, Jongen WMF. Activity and concentration
of polyphenolic antioxidants in apple: eect of cultivar, harvest year, and storage
conditions. J Agric Food Chem 2001; 49:36063613.
Podsedek A, Wilska-Jeszka J, Anders B, Markowski J. Compositional characterisation of some apple varieties. Eur Food Res Technol 2000; 210:268272.
Donovan JL, Meyer AS, Waterhouse AL. Phenolic composition and antioxidant
activity of prunes and prune juice (Prunus domestica). J Agric Food Chem 1998;
46:12471252.
Tomas-Barberan FA, Gil MI, Cremin P, Waterhouse AL, Hess-Pierce B, Kader
AA. HPLC-DAD-ESIMS analysis of phenolic compounds in nectarines, peaches,
and plums. J Agric Food Chem 2001; 49:47484760.
Chang S, Tan C, Frankel EN, Barrett DM. Low-density lipoprotein antioxidant
activity of phenolic compounds and polyphenol oxidase activity in selected
Clingstone peach cultivars. J Agric Food Chem 2000; 48:147151.
Mayr U, Treutter D, Santos-Buelga C, Bauer H, Feucht W. Developmental
changes in the phenol concentrations of Golden delicious apple fruits and
leaves. Phytochemistry 1995; 38:11511155.
Bocco A, Cuvelier ME, Richard H, Berset C. Antioxidant activity and phenolic
composition of citrus peel and seed extracts. J Agric Food Chem 1998; 46:2123
2129.
Talcott ST, Howard LR, Brenes CH. Antioxidant changes and sensory properties
of carrot puree processed with and without periderm tissue. J Agric Food Chem
2000; 48:13151321.
Andreasen MF, Landbo AK, Chritensen LP, Hansen A, Meyer AS. Antioxidant
eects of phenolic rye (Secale cereale L.) extracts, monomeric hydroxycinnamates, and ferulic acid dehydrodimers on human low-density lipoproteins. J Agric
Food Chem 2001; 49:40904096.
McCallum JA, Walker JRL. Phenolic biosynthesis during grain development in
wheat: changes in phenylalanine ammonia-lyase activity and soluble phenolic
content. J Cereal Sci 1990; 11:3549.
Regnier T, Macheix JJ. Changes in wall-bound phenolic acids, phenylalanine and
tyrosine ammonia-lyases, and peroxidases in developing durum wheat grains
(Triticum turgidum L. var. durum). J Agric Food Chem 1996; 44:17271730.
Wende G, Waldron KW, Smith AC, Brett CT. Developmental changes in
cell-wall ferulate and dehydroferulates in sugar beet. Phytochemistry 1999; 52:
819827.
Bewley JD, Black M. Seeds, Physiology of Development and Germination. 2nd ed.
New York: Plenum Press, 1994.
Wu H, Haig T, Pratley J, Lemerle D, An M. Allelochemicals in wheat (Triticum
37
aestivum L.): cultivar dierence in the exudation of phenolic acids. J Agric Food
Chem 2001; 49:37423745.
81. Griths DW, Bain H. Photo-induced changes in the concentration of individual
chlorogenic isomers in potato (Solanum tuberosum) tubers and their complexation
with ferric ions. Potato Res 1997; 40:307315.
82. Percival GC, Baird L. Inuence of storage upon light-induced chlorogenic acid
accumulation in potato tubers (Solanum tuberosum L.). J Agric Food Chem 2000;
48:24762482.
83. Bureau SM, Baumes RL, Razungles AJ. Eects of vine or bunch shading on the
glycosylated avor precursors in grapes of Vitis vinifera L. cv. Syrah. J Agric Food
Chem 2000; 48:12901297.
84. Oufedjik H, Mahrouz M, Amiot MJ, Lacroix M. Eect of g-irradiation on phenolic
compounds and phenylalanine ammonia-lyase activity during storage in relation to
peel injury from peel of Citrus clementina Hort. Ex. Tanaka. J Agric Food Chem
2000; 48:559565.
85. Lewis CE, Walker JRL, Lancaster JE. Changes in anthocyanin, avonoid and
phenolic acid concentrations during development and storage of coloured potato
(Solanum tuberosum L.) tubers. J Sci Food Agric 1999; 79:311316.
86. Nicholson RL, Hammerschmidt R. Phenolic compounds and their role in disease
resistance. Annu Rev Phytopathol 1992; 30:369389.
87. Lattanzio V, De Cicco V, Di Venere D, Lima G, Salerno M. Antifungal activity of
phenolics against fungi commonly encountered during storage. Ital J Food Sci
1994; 6:2330.
88. Saltveit ME. Physical and physiological changes in minimally processed fruits and
vegetables. In: Tomas-Barberan FA, Robin RD, eds. Phytochemistry of Fruits and
Vegetables. Oxford: Oxford University Press, 1997:205220.
89. Babic I, Amiot MJ, Nguyen-The C, Aubert S. Changes in phenolic content
in fresh ready-to-use shredded carrots during storage. J Food Sci 1993; 58:
351356.
90. Ferreres F, Gil MI, Castaner M, Tomas-Barberan FA. Phenolic metabolites in red
pigmented lettuce (Lactuca sativa): changes with minimal processing and cold
storage. J Agric Food Chem 1997; 45:42494254.
91. Cantos E, Espin JC, Tomas-Barberan FA. Eect of wounding on phenolic enzymes
in six minimally processed lettuce cultivars upon storage. J Agric Food Chem
2001; 49:322330.
92. Bostock RM, Wilcox SM, Wang G, Adaskaveg JE. Suppression of Monilinia
fructicola cutinase production by peach fruit surface phenolic acids. Physiol Mol
Plant Pathol 1999; 54:3750.
93. Kuc J. Phytoalexins, stress metabolism, and disease resistance in plants. Annu Rev
Phytopathol 1995; 33:275297.
94. Lyon GD. The biochemical basis of resistance of potatoes to soft rot Erwinia
spp.a review. Plant Pathol 1989; 38:313339.
95. Phukan SN, Baruah CK. Chlorogenic acid content of potato plant tissues in
relation to infection by Phytophtora infestans (Mont.) De Bary. Adv Plant Sci
1991; 2:218222.
96. Johansen OP, Andersen OM, Nerdal W, Aksnes DW. Cyanidin 3-[(6-p-
38
coumaroyl)-2-(xylosyl)-glucosideand other anthocyanins from fruits of Sambucus canadensis. Phytochemistry 1991; 30:41374141.
97. Nakatani N, Kikuzaki H, Hikida J, Ohba M, Inami O, Tamura I. Acylated anthocyanins from fruits of Sambucus canadensis. Phytochemistry 1995; 38:755757.
98. Brouillard R, Figueiredo P, Elhabiri M, Dangles O. Molecular interactions of
phenolic compounds in relation to the colour of fruit and vegetables. In: TomasBarberan FA, Robins RJ, eds. Proceeding of the Phytochemical Society of Europe:
Phytochemistry of Fruit and Vegetables. Oxford: Clarendon Press, 1997:2949.
99. Amiot MJ, Fleuriet A, Cheynier V, Nicolas J. Phenolic compounds and oxidative
mechanisms in fruit and vegetables. In: Toma`s-Barbera`n FA, Robins RJ, eds.
Phytochemistry of Fruit and Vegetables. Oxford: Clarendon Press, 1997:5185.
100. Cliord MN. Astringency. In: Tomas-Barberan FA, Robins RJ, eds. Proceeding of
the Phytochemical Society of Europe: Phytochemistry of fruit and vegetables.
Oxford: Clarendon Press, 1997:87107.
101. Shimazaki N, Mimaki Y, Sashida Y. Prunasin and acetylated phenylpropanoic acid
sucrose esters, bitter principles from the fruits of Prunus jamasakura and P.
maximowiczii. Phytochemistry 1991; 30:14751480.
102. Crouzet J, Sakho M, Chassagne D. Fruit aroma precursors with special reference to
phenolics. In: Tomas-Barberan FA, Robins RJ, eds. Proceedings of the
Phytochemical Society of Europe: Phytochemistry of Fruit and Vegetables.
Oxford: Clarendon Press, 1997:109123.
103. Friedman M, Kozukue N, Harden LA. Cinnamaldehyde content in foods
determined by gas chromatography-mass spectrometry. J Agric Food Chem
2000; 48:57025709.
104. Shahidi F, Janitha PK, Wanasundara PD. Phenolic antioxidants. Crit Rev Food Sci
Nutr 1992; 32:67103.
105. Huang MT, Ferraro T. Phenolic compounds in food and cancer prevention. In:
Huang MT, Ho CT, Lee CY, eds. ACS Symposium Series 507: Phenolic
Compounds in Food and Their Eects on Health. II. Antioxidants and Cancer
Prevention. Washington, DC: ACS, 1992:834.
106. Rice-Evans CA, Miller NJ, Paganga G. Structure-antioxidant activity relationships
of avonoids and phenolic acids. Free Rad Biol Med 1996; 20:933956.
107. Vinson JA, Hao Y, Su X, Zubik L. Phenol antioxidant quantity and quality in
foods: vegetables. J Agric Food Chem 1998; 46:36303634.
108. Moure A, Cruz JM, Franco D, Dominguez JM, Sineiro J, Dominguez H, Nunez
MJ, Parajo JC. Natural antioxidants from residual sources. Food Chem 2001;
72:145171.
109. Vinson JA, Su X, Zubik L, Bose P. Phenol antioxidant quantity and quality in
foods: fruits. J Agric Food Chem 2001; 49:53155321.
110. Rice-Evans CA, Packer L. Flavonoids in Health and Disease. New York: Marcel
Dekker, 1998.
111. Wang H, Cao G, Prior RL. Total antioxidant capacity of fruits. J Agric Food Chem
1996; 44:701705.
112. Heinonen IM, Lehtonen PJ, Hopia AI. Antioxidant activity of berry and fruit wines
and liquors. J Agric Food Chem 1998; 46:2531.
113. Rapisarda P, Tomaino A, Lo Cascio R, Bonina F, De Pasquale A, Saija A.
39
Antioxidant eectiveness as inuenced by phenolic content of fresh orange juices.

J Agric Food Chem 1999; 47:47184723.
114. Kaur C, Kapoor HC. Antioxidants in fruits and vegetablesthe millenniums
health. Int J Food Sci Technol 2001; 36:703725.
115. Richelle M, Tavazzi I, Oord E. Comparison of the antioxidant activity of
commonly consumed polyphenolic beverages (coee, cocoa, and tea) prepared per
cup serving. J Agric Food Chem 2001; 49:34383442.
116. Talcott ST, Howard LR, Brenes CH. Contribution of periderm material and
blanching time to the quality of pasteurized peach puree. J Agric Food Chem 2000;
48:45904596.
117. Wang H, Nair MG, Strasburg GM, Booren AM, Gray JI. Antioxidant polyphenols
from tart cherries (Prunus cerasus). J Agric Food Chem 1999; 47:840844.
118. Nakatani N, Kayano SI, Kikuzaki H, Sumino K, Katagiri K, Mitani T.
Identication, quantitative determination, and antioxidative activities of chlorogenic acid isomers in prune (Prunus domestica L.). J Agric Food Chem 2000;
48:55125516.
119. Cao G, Russell RM, Lischner N, Prior RL. Serum antioxidant capacity is increased
by consumption of strawberries, spinach, red wine or vitamin C in elderly women.
J Nutr 1998;128: 23832390.
120. Gutierrez F, Jimenez B, Ruiz A, Albi MA. Eect of olive ripeness on the
oxidative stability of virgin olive oil extracted from the varieties Picual and
Hojiblanca and on the dierent components involved. J Agric Food Chem 1999;
47:121127.
121. McDonald S, Prenzler PD, Antolovich M, Robards K. Phenolic content and
antioxidant activity of olive extracts. Food Chem 2001; 73:7384.
122. Saija A, Tomaino A, Lo Cascio R, Trombetta D, Proteggente A, De Pasquale A,
Uccella N, Bonina F. Ferulic and caeic acids as potential protective agents
against photooxidative skin damage. J Sci Food Agric 1999; 79:476480.
123. Rodriguez de Sotillo D, Hadley M, Holm ET. Phenolics in aqueous potato peel
extract: extraction, identication and degradation. J Food Sci 1994; 59:649651.
124. Zheng Z, Shetty K. Solid-state bioconversion of phenolics from cranberry pomace
and role of Lentinus edodes. b-glucosidase. J Agric Food Chem 2000; 48:895
900.
125. Manthey JA, Grohmann K. Phenols in citrus peel byproducts: concentrations
of hydroxycinnamates and polymethoxylated avones in citrus peel molasses. J
Agric Food Chem 2001; 49:32683273.
126. Natella F, Nardini M, Di Felice M, Scaccini C. Benzoic and cinnamic acid derivatives
as antioxidants: stucture-activity relation. J Agric Food Chem 1999; 47:14531459.
127. Meyer AS, Donovan JL, Pearson DA, Waterhouse AL, Frankel EN. Fruit hydroxycinnamic acids inhibit human low-density lipoprotein oxidation in vitro.
J Agric Food Chem 1998; 46:17831787.
128. Silva FAM, Borges F, Ferreira MA. Eects of phenolic propyl esters on the oxidative
stability of rened sunower oil. J Agric Food Chem 2001; 49:39363941.
129. Aruoma OI, Murcia A, Butler J, Halliwell B. Evaluation of the antioxidant and
pro-oxidant actions of gallic acid and derivatives. J Agric Food Chem 1993; 41:
18801885.
40
130.
Spanos GA, Wrolstad RE. Phenolics of apple, pear, and white grape juices and
their changes with processing and storagea review. J Agric Food Chem 1992;
40:14781487.
131. Wen L, Wrolstad RE, Hsu VL. Characterization of sinapyl derivatives in pineapple
(Ananas comosus L. Merill) juice. J Agric Food Chem 1999; 47:850853.
132. Talcott ST, Howard LR. Phenolic autoxidation is responsible for color degradation
in processed carrot puree. J Agric Food Chem 1999; 47:21092115.
133. Putz B. The current state of knowledge on blue and black spot or injury.
Kartoelbau 1995; 46:284286.
134. Miller NJ. Flavonoids and phenylpropanoids as contributors to the antioxidant
activity of fruit juices. In: Rice-Evans CA, Packer L, eds. Flavonoids in Health and
Disease. New York: Marcel Dekker, 1998:387403.
135. Nicoli MC, Calligaris S, Manzocco L. Eect of enzymatic and chemical oxidation
on the antioxidant capacity of catechin model systems and apple derivatives.
J. Agric. Food Chem., 2000; 48:45764580.
136. Coetzer C, Corsini D, Love S, Pavek J, Tumer N. Control of enzymatic browning
in potato (Solanum tuberosum L) by sense and antisense RNA from tomato
polyphenol oxidase. J Agric Food Chem 2001; 49:652657.
137. Lee CY, Whitaker JR. Enzymatic Browning and Its Prevention. ACS Symposium
Series, 600. Washington, DC: ACS, 1995.
138. Son SM, Moon KD, Lee CY. Inhibitory eects of various antibrowning agents on
apple slices. Food Chem 2001; 73:2330.
139. Pirisi F, Angioni A, Cabras P, Garau VL, Sanjust di Teulada ML, Kaim dos Santos
M, Bandino G. Phenolic compounds in virgin olive oils. I. Low-wavelength
quantitative determination of complex phenols by high-performance liquid
chromatography under isocratic elution. J Chromatogr A 1997; 768:207213.
140. Brenes M, Garcia A, Garcia P, Rios JJ, Garrido A. Phenolic compounds in Spanish
olive oils. J Agric Food Chem 1999; 47:35353540.
141. Brenes-Balbuena M, Garcia-Garcia P, Garrido-Fernandez A. Phenolic compounds
related to the black color formed during the processing of ripe olives. J Agric Food
Chem 1992; 40:11921196.
142. Maillard MN, Berset C. Evolution of antioxidant activity during kilning: role of
insoluble bound phenolic acids of barley and malt. J Agric Food Chem 1995;
43:17891793.
143. Subba Rao MVSST, Muralikrishna G. Non-starch polysaccharides and bound
phenolic acids from native and malted nger millet (Ragi, Eleusine coracana,
Indaf-15). Food Chem 2001; 72:187192.
144. Donaghy JA, Kelly PF, McKay A. Conversion of ferulic acid to 4-vinyl guaiacol
by yeasts isolated from unpasteurised apple juice. J Sci Food Agric 1999;
79:453456.
145. Meyer AS, Yi OS, Pearson DA, Waterhouse AL, Frankel EN. Inhibition of human
low-density lipoprotein oxidation in relation to composition of phenolic
antioxidants in grapes (Vitis vinifera). J Agric Food Chem 1997; 45:16381643.
146. Baderschneider B, Winterhalter P. Isolation and characterization of novel
benzoates, cinnamates, avonoids, and lignans from Riesling wine and screening
for antioxidant activity. J Agric Food Chem 2001; 49:27882798.

147.
148.
149.
150.
151.
152.
153.
154.
41
Waterhouse AL, Walzem RL. Nutrition of grape phenolics. In: Rice-Evans CA,
Packer L, eds. Flavonoids in Health and Disease. New York: Marcel Dekker,
1998:359385.
Landrault N, Poucheret P, Ravel P, Gasc F, Cros G, Teissedre P-L. Antioxidant
capacities and phenolics levels of french wines from dierent varieties and
vintages. J Agric Food Chem 2001; 49:33413348.
Mangas JJ, Suarez B, Picinelli A, Moreno J, Blanco D. Dierentiation by phenolic
prole of apple juices prepared according to two membrane techniques. J Agric
Food Chem 1997; 45:47774784.
Dawes HM, Keene JB. Phenolic composition of kiwifruit juice. J Agric Food
Chem 1999; 47:23982403.
Fu HY, Huang TC, Ho CT. Changes in phenolic compounds during plum
processing. In: Huang MT, Ho CT, Lee CY, eds. ACS Symposium Series 506:
Phenolic Compounds in Food and Their Eects on Health. I. Analysis,
Occurrence, and Chemistry. Washington, DC: ACS, 1992:287295.
Silva BM, Andrade PB, Mendes GC, Valentao P, Seabra RM, Ferreira MA.
Analysis of phenolic compounds in the evaluation of commercial quince jam
authenticity. J Agric Food Chem 2000; 48:28532857.
Silva BM, Andrade PB, Valentao P, Mendes GC, Seabra RM, Ferreira MA.
Phenolic prole in the evaluation of commercial quince jellies authenticity. Food
Chem 2000; 71:281285.
Gil-Izquierdo A, Gil MI, Ferreres F, Tomas-Barberan FA. In vitro availability of
avonoids and other phenolics in orange juice. J Agric Food Chem 2001; 49:1035
1041.
2
Flavonoids in Herbs
Piergiorgio Pietta and Claudio Gardana
Institute of Biomedical TechnologiesNational Council of Research
Segrate, Italy
Annamaria Pietta
Farmacia Dr. Carlo Bravi
Brescia, Italy
I.
INTRODUCTION
Flavonoids are a large group of polyphenolic compounds that occur commonly in plants. This group contains more than 8000 known compounds, and
this number is constantly growing because of the great structural diversity
arising from the various hydroxylation, methoxylation, glycosylation, and
acylation patterns.
Flavonoids are the pigments responsible for the shades of yellow, orange,
and red in owering plants. They are also important factors for plant growth,
development, and defense. Many avonoids are endowed with biological
activities, such as anti-inammatory, antiallergic, antischemic, antiplatelet, immunomodulatory, and antitumoral activities [13]. Flavonoids have also been
shown to inhibit several enzymes, including lipoxygenases and cyclooxygenases, mono-oxygenases, xanthine oxidase, mitochondrial succinoxidase,
reduced nicotinamide-adenine dinucleotide (NADH) oxidase, phospholipase A2,
topoisomerases, and protein kinases [46]. The biological activities of avonoids
are thoughy to be due mainly to their antioxidant properties [78], which are
displayed by limiting the production of reactive oxygen species (ROS) and/or
scavenging them.
Flavonoids are components of the diet of numerous herbivores and
omnivores, including humans [9]. They are principally found in fruits, vegeta43
44
Pietta et al.
bles, and popular drinks, such as red wine, tea, beer, and their intake may reach
1 g/day [10]. In addition, avonoids are present in various herbs* [11].
Approximately 50 species, from Achillea millefolium to Viola tricolor, have
been used as herbal remedies for their avonoid content; some are listed in
Table 1. These preparations have been reported to be eective for the treatment of
disorders of peripheral circulation and for the improvement of aquaresis. In
addition, avonoid-based herbal medicines are available in dierent countries as
anti-inammatory, antispasmodic, antiallergic, and antiviral remedies [1214].
The pharmacological eects of these phytomedicines are ascribed either to their
functions as radical scavengers, reductants, and metal chelators or to alternative
nonantioxidant functions, including the interaction with dierent enzymes, the
inhibition of calcium ion inux into the cells, and the regulation of cell signaling
[15] and gene expression [16]. However, it should be remembered that the health
benet properties of most medicinal plants high in avonoids cannot be assigned
exclusively to these compounds, since other components present in the phytocomplex may either directly contribute to or display a permissive role that
enhances the eects of avonoids. When examining dierent examples including
aquaretic, anti-inammatory, sedative, and antispasmodic herbs, it is found that
the observed pharmacological eect is due to avonoidic and nonavonoidic
constituents [17].
As natural products, herbs can greatly dier in their composition as a
result of genetic factors, climate, soil quality, and other external factors. Therefore, controlled cultivation and selection represent the rst steps to ensuring the
most consistent concentration of specic ingredients or groups of compounds.
Second, the production of the herbal ingredients by extracting the herbs with
solvents must be carefully monitored to select the components that are important
to the action and the ecacy of the product. To achieve consistent pharmaceutical quality, the analytical quality control is essential. This is not an easy
task, as herbs and related extracts are complex mixtures of constituents with
dierent physicochemical (i.e., analytical) characteristics. With avonoidcontaining herbs, however, phytochemical data are largely available: i.e., the
chemical nature of avonoids present in these herbs is known. Almost all the
avonoid classes are present in herbs with proven therapeutic activity, including
avonols, avones, and their dihydroderivatives; isoavones; catechins; ava-
* In botanical nomenclature, the word herb refers to non woody seed-producing plants that die down
at the end of the growing season. Currently, in the culinary arts, herbs are dened as vegetable
products used to add avor or aroma to food and beverages. However, in botanical medicine the
word herb refers to plants or plant parts used in freshed, dried, or extracted form for the treatment
of disease states, often of a chronic nature, or for improvement or maintenance of health. So here,
for the purpose of this chapter, herbs are considered as healing herbs (medicinal plants), as
distinguished from vegetables with aromatic and savory properties, that is, culinary herbs.
Flavonoids in Herbs
45
nolignans; and anthocyanins. Some of the additional phytochemicals are closely

related to avonoids such as phenolic and hydroxycinnamic acids, whereas
others have dierent chemical natures, including various terpenes (mostly
present in volatile oils), coumarin derivatives, phytosterols, and other speciescharacteristic constituents.
The analysis of the avonoid fraction in the raw herbs and in standardized
(i.e., having known potency) extracts may be accomplished by using dierent
approaches [1819], including high-performance liquid chromatography (HPLC),
capillary electrophoresis (CE), and mass spectrometry. HPLC coupled with
online ultraviolet (UV) detection and/or mass spectrometry (MS) allows data
on the chromatographic, UV, and MS behavior of the analytes to be obtained
from a single run. This approach remains the method of choice (1) to obtain
typical ngerprints for the herbal ingredient; (2) to assay single avonoids;
and (3) to detect evidence the presence of adulterants. CE has been proved a
valuable alternative to HPLC, because of its high selective power, which allows
detection of some avonoids not separable by HPLC. Unfortunately, CE has not
become as popular as HPLC, which remains the technique of choice for routine
quality control of avonoid-containing vegetables [20].
Typical examples of avonoid herbs examined by HPLC or CE have
already been described in a previous contribution of this volume series [11]. This
chapter aims to describe three mass techniques, the electrospray ionization MS,
(ESI-MS), atmospheric pressure chemical ionization MS (APCI-MS), and ion
trap MS (ITMS) techniques, and their application to the analysis of avonoids in
some standardized herbal extracts with proven therapeutic ecacy. In addition,
the avonoid composition of some commonly consumed vegetables with
aromatic or savory properties (culinary herbs) is described.
A.
Mass Spectrometry of Flavonoids
Flavonoid herbs have been increasingly studied by mass spectrometry (MS) since
the introduction of the thermospray (TSP), ESI, and APCI interfaces, which allow
direct coupling of MS with HPLC. Being characterized by soft ionization,
these techniques permit the analysis of avonoids in their native form without
derivatization [21]. TSP-MS was used, at rst, to analyze avonoids in dierent
plant extracts, such as Arnica montana, Gentianaceae species, Ginkgo biloba,
Calendula ocinalis, and Hypericum perforatum [11]. Unfortunately, TSP-MS
fails in the case of thermolabile compounds, such as the avonol-glycosides.
These compounds undergo fragmentation and yield mainly the aglycone
fragment [A+H]+, with the molecular ion [M+H]+ present in very low quantity.
This was one reason to switch to the ESI and APCI interfaces, which involve a
low level of fragmentation. Both ESI and APCI produce mainly molecular ions,
and they are particularly suitable for detecting intact molecular species present
Rutin, quercitrin, isoquercitrin,

astragalin, hyperoside
(1.52.4%); triterpene
saponins (>8%)
Procyanidins (>3%);
avonol and
avone derivatives (>1%)
Glycyrrhizin-glycosides (424%);
avanones, chalcones,
and isoavones (c1%)
Goldenrod
(Solidago virgaurea)
Hawthorn leaf/ower/fruit
(Crataegus monogyna,
oxyacantha)
Licorice root
(Glycyrhiza glabra)
avonol glycosides (2227%);

terpene lactones (57%)
Ginkgo leaf extract

(Ginkgo biloba)
Elderberry extract
Elder ower
(Sambucus nigra)
Calendula owers
(Calendula ofcinalis)
Chamomile owers
(Matricaria chamomilla)
Dyspeptic problems
Anti-inammatory,
demulcent, expectorant
65,66
63,64
62
Irritation of urinary tract

in cases of inammation
and renal gravel
Cardiotonic
61
60
59
57,58
56
55
54
Reference
Cognitive decits
Capillary weakness,
venous insuciency,
diarrhea
Oral mucosa inammation,
wound healing
Gastrointestinal complaints,
skin and mucosa
irritations
Feverish conditions,
common cold
Dyspeptic problems
Uses
Cardioactive, hypotensive,
coronary vasodilator
Antioxidant,
anti-inammatory,
antiallergic
Antioxidant,
neuroprotective,
PAF inhibitory
Aquaretic, analgesic,
anti-inammatory
Antispasmodic,
anti-inammatory,
antibacterial
Anti-inammatory,
aquaretic
Anti-inammatory
Vasoprotective, antioxidant,
astringent
Choleretic, hepatoprotective,
spasmolytic
Luteolin-glycosides (<1%);
caeoyl-quinic acids (<2%);
sesquiterpene lactones (<4%)
Anthocyanins (mainly malvidin,
cyanidin, and delphinidin
glycosides) (<24%)
Isoquercitrin, rutin, narcissin;
triterpene saponins
Apigenin and luteolin glycosides
( > 8%); volatile oil
(a-bisabolol, chamazulene)
Rutin, astragalin, hyperoside,
isoquercitrin (<3%);
triterpene acids (about 1%)
Anthocyanins (>15%)
Artichoke leaf
(Cynara scolymus)
Bilberry fruit extract

(Vaccinium myrtillus)
Alleged activity
Purported active componentsa
Herb
Table 1 Selected Flavonoid Herbs
46
Pietta et al.
Volatile oil (>2%); avones,

avanones
Vitexin, isovitexin, and
their c-glycosides,
luteolin glycosides (>2.5%)
Hyperoside (0.52%),
rutin (0.31.6%),
isoquercitrin (0.3%);
phoreglucinols (>4%);
naphthodianthrones (>0.75%)
Salicin and its esters (>10%);
naringenin glucosides
(>1.5%)
Hamamelitannin,
gallotannins (>12);
catechins, quercetin, and
kaempferol glycosides
Procyanidin dimers (about 2%);

quercetin and kaempferol
glycosides (about 1%)
Quercetin glycosides, mostly
quercetin-4V-glucoside
(0.51%); hydrolyzable
tannins (1015%);
salicylates (about 0.15%)
Flavanolignanes,
collectively called
sylimarin (1.53%)
The content refers to the dried herb.

PAF, platelet activating factor.
Witch hazel leaf/bark

(Hamamelis virginiana)
Willow bark (Salix spp.)
Orange peel, bitter

(Citrus aurantium)
Passionower
aboveground parts
(Passiora incarnata)
St. Johns wort
(Hypericum perforatum)
Milk thistle fruit

(Sylibum marianum)
Meadowsweet ower
(Filipendula ulmaria)
Linden owers
(Tilia cordata)
Anti-inammatory,
antipyretic,
analgesic
Astringent,
anti-inammatory,
vasoconstrictive,
antioxidant
Antidepressant
Sedative, anxiolytic
Aromatic bitter, choleretic
Hepatocellular protection
Diaphoretic
Diaphoretic, sedative
73,74
75,76
Skin injuries,
varicose veins,
hemorrhoids
72
71
70
69
68
67
Mild neuralgic pains,

osteoarthritis
Mild to moderate
depression
Toxic liver damage,

supportive treatment
in chronic liver
diseases, dyspeptic
complaints
Dyspeptic ailments,
appetite stimulant
Nervous unrest,
mild sleep disorders
Cold relief
Cold relief, nervous

tension
Flavonoids in Herbs
47
48
Pietta et al.
in herbal extracts. For further structural information, these interfaces may be

coupled to an ITMS analyzer to promote mass fragmentation, and this
arrangement provides data helpful in identifying the avonoids of interest.
1.
Electrospray Ionization Mass Spectrometry
ESI-MS produces ions as a result of the application of a potential to a owing

liquid, which causes the liquid to charge and spray. Electrospray forms very small
droplets of solvent containing the analytes. Usually, the solvent is removed by
heat and multicharged ions are produced. As previously stated, ESI has the
advantage over TSP of producing low fragmentation of avonoid derivatives.
To elucidate the dierence between TSP and ESI, the following example
is illustrative. TSP-MS of kaempferol-3-O-rutinoside (molecular weight [MW]
594 da) mainly produces kaempferol fragment (m/z 286) resulting from the loss
of rutinose ([M-rutinose+H]+), and the kaempferol-rhamnoside fragment (m/z
433, [M-glucose + H]+). The molecular ion (m/z 595 = [M+H]+) is present in
very low quantity. Conversely, ESI-MS of the same glycoside predominantly
yields the sodium and potassium adducts of the molecular ions: ([M+Na]+), m/z
617; ([M+K]+), m/z 633. Fragmentation ions are almost absent.
For this reason, ESI-MS is particularly suitable for direct analysis of
samples without preliminary chromatographic separation. As a result, specic
ngerprints of complex natural mixtures are easily and rapidly obtained. This
information on the overall components is particularly valuable for herbal
medicines, because they are in toto regarded as the active principle rather than
single constituents.
Closely related to ESI is APCI in that the source operates at nearatmospheric pressure. APCI produces almost molecular ions with very little
fragmentation, and it provides ngerprints of herbs [22].
2.
Ion Trap Mass Spectrometry
MS spectra with fragmentation of molecules require collision-activated dissociation (CAD) and triple quadrupole analyzers. In these instruments, the analysis
is performed as follows: the rst quadrupole selects the interesting ion (parent
ion), the second produces the fragments from the isolated ion, and the third
quadrupole analyzes the fragmentation products (daughter ion spectrum). These
steps (ion isolation, fragmentation, and analysis) can be repeated by addition of
n quadrupole devices (multisector mass spectrometer) to allow multiple MS/MS
experiments (MSn) to be performed.
As an alternative, MSn analysis can be carried out in the same physical
space by means of ITMS. This approach involves using combinations of direct
and rf-eld pulse sequences on trapped ions in a helium reagent gas atmosphere.
Flavonoids in Herbs
49
Besides being simple, ITMS oers signicant advantages in terms of sensitivity

over a triple quadrupole and it may play an important role in avonoid analysis.
To exemplify, ITMS is capable of isolating the ions m/z 271, 301, 353, 447,
and 609 from the negative ESI-MS spectrum of naringenin, quercetin, chlorogenic acid, quercitrin, and rutin and fragmenting them to produce an ion map,
which shows both isolated ions (parent m/z axis) and their fragments (product m/z
axis). In practice, rutin and quercitrin are identied from their molecular ions
(m/z 609 and 447) and from the same fragment (the aglycone quercetin, m/z 300).
Similarly, the identity of chlorogenic acid is given by the molecular ion (m/z 354)
and the fragment m/z 191, which represents quinic acid.
The techniques described allow three dierent analytical approaches:
infusion, direct injection, and injection after a separation step. The infusion is
the simplest method of sample introduction (by means of a syringe) into the
mass spectrometer. High sample volumes (50150 AL) at ow rates (310 AL/
min) are required, and these conditions facilitate the structural investigation of
the analytes subjected to a continuous infusion into the spectrometer.
In the second approach, the sample solution is injected by means of a
HPLC injector directly into the mass spectrometer, without using any chromatographic column. Direct injection involves low sample volumes (110 AL) and
has the advantage over the infusion method that no cleaning is needed after each
analysis. In addition, the analysis times are very short (12 min), thereby
permitting rapid screening of many samples. Furthermore, the direct injection
approach allows minimization of the ion suppression eects due to the matrix by
adding dierent concentrations of avonoid standards to the herbal sample, and
it may be considered for semiquantitative and rapid screening of herbal extracts.
The third approach involves coupling of a separative system (usually
HPLC) with the mass spectrometer. This procedure simultaneously provides
chromatographic, ultraviolet, and mass spectrometric data, and this range of
information may be very helpful when assessing the identity of principles
present in herbs. Further, HPLC coupled to MS (LC-MS) permits discrimination
of compounds with the same molecular masses and exclusion of interference
from the herbal matrix. Therefore, this approach remains the method of choice
for quantitative analyses.
3.
Sample Preparation
The herb is usually extracted with methanol or aqueous methanol at room

temperature or at 4050 jC, depending on the stability of its components. The
resulting crude extract may be puried to remove undesired constituents, such as
lipids, chlorophyll, sugars, organic acids, and salts. In the case of commercial
extracts, which are normally enriched in specic compounds, this step may be
50
Pietta et al.
eliminated. Similarly, the purication may not be necessary in the case of LCMS, since the analytes of interest are separated from the interfering compounds
during the chromatographic elution. By contrast, purication of the sample is
recommended in the infusion and direct injection approaches. Indeed, the
presence in the herbal matrix of dierent molecular species at concentrations
ranging from 1 to 10 mM can cause ion suppression: i.e., the MS analyzer fails to
detect the ions. In some circumstances the matrix eect may be reduced by
diluting the sample and/or lowering the ow rate. These expedients appear to be
successfull when highly sensitive and salt-compatible MS instruments are used.
4.
Alkali Adducts
In positive ESI-MS, some molecular species can form adducts with alkali
cations (sodium and potassium). In particular, potassium adducts are typical
of raw herbal samples, because vegetable matrices are rich in potassium salts.
Alkali adduct formation may be diminished by desalting the samples through
solid phase extraction (SPE). Diluting the sample solutions is a simple way to
replace potassium ions with sodium ions. The latter are the most common in
commercial extracts of herbs.
Not all avonoids are capable of yielding alkali adducts. Thus, avonol-3O-glycosides generate sodium or potassium adducts. By contrast, these adducts
are not obtained from avonol-4V-O-glycosides and avone glycosides.
So the spectrum of rutin (quercetin-3-O-rutinoside, MW 610 da) is characterized by the presence of the sodium adduct (m/z 633); the molecular ion ([MH]+,
m/z 611) is almost absent. Also, the abundance of the aglycone residue (m/z 302) is
low, indicating that the removal of the glucose residue is hindered. The same
behavior can be observed for other avonol-3-O-glycosides, as described for
Ginkgo biloba and St. Johns wort extracts (see Secs. II.A and II.B).
Conversely, in the case of spiraeoside (quercetin-4V-O-glucoside) no
adduct is formed. Its spectrum mainly presents the molecular ion ([MH]+ =
465) and a relevant fragmentation occurs. Likewise, the avone-glycoside
(lacking the 3-OH group) rhoifolin (apigenin-7-O-neohesperidoside, MW 578)
produces the m/z values 579 ([MH]+) and 270, corresponding to molecular and
aglycone residue ions, respectively. This nding may suggest that the presence
of the hydroxyl group at position 3 may be important for adduct formation.
However, this hypothesis is not appropriate, since avonol aglycones, such as
quercetin, kaempferol, and isorhamnetin, do not produce sodium adducts. What
seems crucial is the 3-O-glycosylation, which may favor the formation of a
crown-embedded stable cation.
Isoavones form molecular ion adducts. Likely, the isoavone ring at
position 3 plays the same role as the sugar moiety in 3-O-glycosyl avonols. In
Flavonoids in Herbs
51
fact, the ESI-MS spectrum of the isoavone biochanin A (MW 284 da) shows as
main ion the sodium adduct ([M+Na]+, 307 m/z).
Alkali adducts are also formed with other avonoid classes (see Secs. II.D.
and II.G).
II.
MASS SPECTROMETRY FINGERPRINTS OF SELECTED

FLAVONOID HERBS
A.
Ginkgo biloba
The dried green leaves of the ginkgo tree (Ginkgo biloba L., Fam. Ginkgoaceae) are the crude herb from which ginkgo extracts are obtained. According
to the monograph published by ESCOP [23], the only acceptable ginkgo
extracts are those obtained with an average herb-to-extract ratio of 50:1. The
extraction procedure is strictly standardized to eliminate unwanted constituents, such as fats, tannins, biavonoids, ginkgol, and ginkgolic acids. The
resulting extract consists of 2227% avonol-glycosides, 57% terpene
lactones (2.83.4% ginkgolides A, B, C, and J and 2.63.2% bilobalide),
small amounts of phenolic acids, and fewer than 5 parts per million (ppm)
ginkgolic acids (allergenic).
More than 300 papers have been published on the pharmacological
properties of this extract [23]. Experimentally documented activity includes
increased tolerance to hypoxia in brain tissue, improved learning capacity and
memory, inhibition of platelet activator factor (PAF) [24], improved cerebral
and peripheral circulation, neuroprotective eects [25], and reduction of retinal
edema. Although all the constituents of the ginkgo extract are considered to
contribute to the therapeutic eects, the ginkgo avonoids are assumed to play
an important role that is due to their free radical scavenging capacity.
The ngerprint of Ginkgo biloba extracts obtained by direct infusion in
ESI-MS is shown in Figure 1. The ions (m/z 617779) refer to dierent avonolglycosides. In particular, the ions m/z 617, 633, and 647 are due to the sodium
adducts of kaempferol-3-O-rutinoside, quercetin-3-O-rutinoside, and isorhamnetin-3-O-rutinoside, respectively. The ions m/z 763, 779, and 793 correspond to
the sodium adducts of the 3-O-[rhamnosyl-(1!2)-rhamnosyl-(1!6)-glucoside]
derivatives of kaempferol, quercetin, and isorhamnetin, respectively. Ions m/z
763 and 779 account also for the 3-O-[6VVV-p-coumaroylglucosyl-(1!2)rhamno2)rhamnoside] derivatives of kaempferol and quercetin. Finally, the ions m/z
431, 447, and 463 correspond to sodium adducts of ginkgolide A, ginkgolide B,
ginkgolide J, and ginkgolide C, respectively. Interestingly this approach has
allowed simultaneous detection of avonoid and terpene compounds of Ginkgo
biloba, thereby permitting rapid screening of many samples.
52
Pietta et al.
Figure 1 Typical positive ESI-MS mass spectrum of Ginkgo biloba extract. ESIMS, electrospray ionization mass spectrometry; GA, ginkgolide A; GB, ginkgolide
B; GC, ginkgolide C; GJ, ginkgolide J.
B.
St. Johns Wort
St. Johns wort consists of the dried aboveground parts of Hypericum perforatum
L. (Fam. Hypericaceae) gathered during the owering season. St. Johns wort
contains numerous compounds with documented biological activity. Components
that have received most attention are the naphthodianthrones hypericin and
pseudohypericin (up to 0.15%) and the phloroglucinols hyperforin and adhyperforin (up to 4%). Besides these specic components, St. Johns wort contains
signicant amounts of some very common plant metabolites [2627], such as
avonol derivatives, mainly hyperoside (0.52%), rutin (0.31.6%), quercitrin,
and isoquercitrin (up to 0.3%); biavones (amentoavone); procyanidins; and a
volatile oil (about 1%). Clinical proof of therapeutic ecacy for mild to moderate
depression [2829], restlessness, and irritability has been documented for St.
Johns wort ased on alcoholic extracts, with an herb-to-extract ratio in the range of
4:1 to 7:1 (i.e., 0.20.3% hypericins and 26% hyperforins, respectively).
ESI-MS in negative mode of Hypericum perforatum extracts provides
typical ngerprints showing three dierent classes of compounds, i.e., avonoids,
hypericins, and hyperforins (Fig. 2). The abundant ions at m/z 535 and 549 are due
to [M-H] ions of hyperforin and adhyperforin, respectively. Deprotonated
molecules ([M-H]) of hypericin (m/z 503), pseudohypericin (m/z 519), and
Flavonoids in Herbs
53
Figure 2 Typical negative ESI-MS mass spectrum of Hypericum perforatum

extract. ESI-MS, electrospray ionization mass spectrometry; M1, rutin; M2,
hyperforin; M3, pseudohypericin; M4, hypericin; M5, adhyperforin; M6, isoquercitrin/hyperoside.
rutin (m/z 609) are present in lower abundance. [M-H] ions of other known
avonoids can be detected, such as those of isoquercitrin and hyperoside (m/z
463), quercitrin (m/z 447), and quercetin (m/z 301). The ions m/z 466, 397, 383,
and 313 are produced by fragmentation of [M-H] ions of hyperforin, presumably by collisions in the skimmer region. As expected, ESI-MS of St. Johns
wort extracts in positive mode evidences produces cation adducts of isoquercitrin ([M+Na]+ = 487; [M+K]+ = 503) and rutin ([M+Na]+ = 639; [M+K]+ = 649).
ESI-MS may be applied to conrm dierences due to environmental
conditions. For example, in the case of H. perforatum owers collected in
northern and southern Italy, the northern sample was higher in rutin (m/z 609)
and pseudohypericin (m/z 519) and had low levels of hyperforin (m/z 535).
C.
Bilberry
Bilberry consists of the dried ripe fruit of Vaccinium myrtillus, a dwarf shrub of the
Ericaceae family. Dried bilberries contain 15% catechins, approximately 30%
invert sugar, and small amounts of avonol glycosides (e.g., astragalin, quercitrin,
isoquercitrin, hyperoside), phenolic acids (e.g., caeic and chlorogenic acids), and
anthocyanins, particularly glycosides of malvidin, cyanidin, and delphinidin [30].
54
Pietta et al.
Bilberry has shown vasoprotective, antiedematous, antioxidant, anti-inammatory, and astringent actions. In particular, a standardized extract enriched with
anthocyanins and their aglycons (anthocyanidins) is endowed with constant
biological activity useful in ophthalmology and treatment of vascular disorders
including capillary weakness, venous insuciency, and hemorrhoids [12].
The positive mass spectrum of this standardized bilberry extract is shown in
Figure 3. The most abundant ions are due to the molecular ions of anthocyanidins. In particular, the ions with m/z 287, 303, 317, and 331 are related to
cyanidin, delphinidin, petunidin, and malvidin, respectively. In addition, the
anthocyanins (glycoside derivatives) are also present as molecular ions: m/z 419,
449, 463, 479, and 493 correspond to cyanidin-3-O-arabinoside, cyanidin-3-Oglucoside/petunidin-3-O-arabinoside, delphinidin-3-O-glucoside/malvidin-3-Oarabinoside, petunidin-3-O-glucoside, and malvidin-3-O-glucoside, respectively.
By this approach, ESI-MS in the positive mode permits simultaneous detection of
dierent anthocyanidins and anthocyanins in bilberry extract, and the results are
in good agreement with those previously reported [30]. Interestingly, the
ngerprint of Vaccinium myrtillus diers from those of Catharanthus roseus,
Vaccinium macrocarpon (cranberry fruit), and Sambucus nigra (elder fruit),
Figure 3 Typical positive ESI-MS mass spectrum of Vaccinium myrtyllus extract.

ESI-MS, electrospray ionization mass spectrometry.
Flavonoids in Herbs
55
indicating that extracts containing the same class of compounds may be easily
dierentiated by ESI-MS.
D.
Camellia sinensis
Tea (Camellia sinensis, Fam. Theaceae) can be considered one of the most
important herbal antioxidants [31,32]. Indeed, fresh leaves of tea contain avanols
(catechins), avonols, catechin tannins (procyanidins), and phenolic acids, which
are all endowed with antioxidant properties. This composition is almost preserved
in green tea, i.e., in the variety obtained by steaming and drying the leaves
immediately after harvesting to prevent enzymatic modications (fermentation).
In green tea the group of avanols usually accounts for 1530% of dried leaves,
and it includes dierent catechins, e.g., epigallocatechin, epicatechin, epigallocatechin-3-O-gallate, gallocatechin-3-O-gallate, methyl-epigallocatechin-3-Ogallate, and epicatechin-3-O-gallate. Green tea as black (fermented) and oolong
(partially fermented) teas also contains caeine (2.55.5%). Because of the
presence of this methylxanthine, the intake of large doses of tea is not advisable,
and this property reduces the healthy eects, that are thought to be associated with
high consumption of tea [33]. To overcome this limitation a caeine-free green tea
extract containing about 80% catechins has been developed.
Figure 4 shows the ESI-MS positive mass spectrum of this decaeinated
green tea extract. The most abundant ion, m/z 481, is due to the sodium adduct of
epigallocatechin gallate (EGCg). The ions m/z 465, 329, and 291 are present in
lower abundance and correspond to the sodium adducts ([M+Na]+) of epicatechin gallate (ECg), epigallocatechin (EGC), and epicatechin/catechin
(EC/C), respectively.
E.
Soybean and Red Clover
The primary isoavones in soybeans (Glycine max) are genistein and daidzein
and their respective 7-O-h-glucosides. There are also small amounts of a third
isoavone, glycitein, and its glucoside, glycitin. In soybeans and in nonfermented
soy foods, isoavone glucosides are esteried with malonic or acetic acid [34].
The isoavone pattern of red clover (Trifolium pratense) diers from that
of soybean. In addition to genistein, daidzein, and their glucosides, red clover
contains formononetin (a precursor of equol) and biochanin-A in free and
glycosylated forms [35].
Isoavones have a chemical structure similar to that of mammalian estrogens and are referred as phytoestrogens [36]. Isoavones are quite weak,
possessing 1/1000 to 1/10,000 the estrogenic activity of 17-O-h-estradiol. How-
56
Pietta et al.
Figure 4 Typical positive ESI-MS mass spectrum of green tea extract. ESI-MS,
electrospray ionization mass spectrometry.
ever, their circulating concentrations in subjects consuming a moderate amount of

soy foods may be 1000-fold higher than peak levels of endogenous estrogens.
Furthermore, isoavones have a stronger binding anity for estrogen receptor h
(Erh) than Era [37], suggesting that the estrogenic eects of isoavones may be
tissue-selective. Indeed, some tissues predominantly contain one form of the
receptor or the other, and this may partly explain the health benets of
isoavones on bone and breast tissues [38]. Specically, genistein and daidzein
are thought to decrease osteoporosis, diminish the risk of estrogen-enhanced
carcinogenesis, relieve menopausal symptoms, and decrease heart disease [39].
In addition, isoavones possess antioxidant capacity and have eects on protein
synthesis, enzyme activity, growth factor action, and angiogenesis. Unfortunately, most of these health benets are based on epidemiological and
laboratory evidence, and only a few small and contradictory trials are presently
available. What seems necessary are some larger clinical studies, comparing
placebo with low and high isoavone intake [40,41]. As an example of an
isoavone-containing herb, Figure 5 shows the positive ESI mass spectrum of a
Flavonoids in Herbs
57
Figure 5 Typical positive ESI-MS mass spectrum of soybean extract. ESI-MS,

electrospray ionization mass spectrometry.
standardized soybean extract. The main ion (m/z 255) is due to the aglycone
daidzein, whereas the glycoside daidzin (m/z 417) and genistein (m/z 271) are
present in lower abundance.
F.
Propolis
Propolis is a resinous substance collected by honeybees from leaf buds and cracks
in the bark of various plants, mainly from poplar (Populus) species and, to a
lesser extent, beech, horse chestnut, birch, and conifer. Bees mix the original
propolis with beeswax and h-glucosidase they secrete during the propolis
collection. The resulting material is used by bees to seal holes in the hives,
exclude drafts, protect against external invaders, and mummify their carcasses.
Propolis has been used extensively in folk medicine for many years, and there is
substantial evidence indicating that propolis has antiseptic, antifungal, antibacterial, antiviral, anti-inammatory, and antioxidant properties [42].
Propolis cannot be used as raw material, it has to be puried by either
extraction with alcoholic solvents or supercritical uid extraction. Depending on
the geographical source and extraction processes, propolis extracts may dier
Figure 6 Negative APCI-IT-MS spectrum of propolis. APCI-IT-MS, atmospheric

pressure ionization ion trap mass spectrometry.

Flavonoids in Herbs
59
from each other. Unfortunately and unlike for other herbal extracts used to
prepare phytomedicines, a generally accepted standardized extract of propolis
has been not yet assessed. For this reason, three dierent samples of propolis are
considered, and their APCI mass spectra recorded in negative mode are shown in
Figures 68. In contrast to HPLC, which permits identication only a limited
number of avonoids (those with a reference standard) and may suggest the
presence of other avonoid like compounds (from the UV spectra), APCI-IT-MS
of demonstrates a larger array of avonoids, as summarized in Tables 57 of
Figs. 68. The identity of the ions is conrmed through their MS-MS fragmentation, as exemplied for pinocembrin, pinobanksin, and its acetate (Fig. 9). A
comparison of MS ngerprints indicates that sample 2 has a broader pattern of
compounds than sample 3, which in turn is more rich than sample 1.
MS may also provide a rapid comparison of samples obtained by dierent
extraction procedures. So, propolis obtained by supercritical uid extraction
produces a negative ESI mass spectrum mainly showing chrysin (m/z 253).
Apigenin/galangin (m/z 269) and naringenin (m/z 271) are also present at
signicant levels, whereas pinocembrin (m/z 255) is less abundant. In contrast,
the MS spectrum of propolis sample obtained by ethanol extraction evidences
higher levels of this avanone.

60
Pietta et al.
Figure 9
G.
MS/MS2 of a propolis extract.
Vitis vinifera
Polyphenolic oligomers of the avanoid type result from the condensation of two
or more avan-3-ols and are known as catechin tannins or oligomeric cyanidins
(OPCs). The most abundant sources of OPC are extracts of grape seed (Vitis
vinifera), standardized to contain 8085% procyanidins, and extracts of the bark
of the maritime pine (Pinus pinaster), concentrated to contain about 85%
procyanidins [43]. Both these extracts are characterized by the presence of
procyanidins (oligomers of catechin epicatechin and their 3-O-gallates) [44].
Procyanidins are potent antioxidants [32] and active inhibitors of collagenase,
elastase, hyaluronidase, and h-glucuronidase [45]. All these enzymes are involved
in the degradation of the main structural components of the extravascular matrix,
and their inhibition may aid maintenance of normal capillary permeability.
The negative mass spectrum of Vitis vinifera permits identication of the
ions related to proanthocyanidins, i.e., the monomer, dimer, trimer, tetramer,
and pentamer (Table 2), as previously reported for positive ESI-MS equipped
with a triple quadrupole [45]. The ion map of the same sample allows the
detection of the dierent parent ions and their fragment products. For example,
the ion m/z 729 (dimer gallate) produces the fragments m/z 711 (loss of water),
Flavonoids in Herbs
61
Table 2 Procyanidin Ions Detected in a

Standardized Vitis vinifera Seed Extract by
Negative Electrospray IonizationIon Trap
Mass Spectrometry
[M-H] (m/z)
169
289
441
577
729
865
881
1017
1153
1169
1305
1441
1593
Compound
Gallic acid
Monomer (catechin-epicatechin)
Monomer-gallate
Dimer
Dimer-gallate
Trimer
Dimer-digallate
Trimer-gallate
Tetramer
Trimer-digallate
Tetramer-gallate
Pentamer
Pentamer-gallate
577 (loss of gallate), 441 (loss of a monomer), and 289 (loss of a monomer
gallate). All the other oligomers yield analogous results.
H.
Flavonoids in Herbs with Aromatic or Savory Properties
The presence of polyphenols (e.g., avonoids and phenolic acid) has been
reported in dierent species of the Labiatae family, including Ocimum
basilicum (basil), Origanum vulgare (oregano), Origanum majorana (marjoram), Mentha var. (mint), Rosmarinus ocinalis (rosemary), and Thymus
vulgaris (thyme), and Umbelliferae family, such as Anethum graveolens (dill),
Coriandrum sativum (coriander), Cuminum sativum (cumin), Foeniculum
vulgare (fennel), and Petroselinum crispum (parsley) [46]. Many of these
species are consumed as ingredients in foods and beverages for their avoring
and aromatic properties and are regarded as culinary herbs. In most of these
culinary herbs, the avor is provided by a variety of compounds (Table 3)
present in their volatile oils [46,47]. In contrast, the avonoid and nonvolatile
phenolic (e.g., carnosol, rosmarinic acid, carnosic acid, ferulic and caeic
acids) fractions are not relevant for sensory eects, but they may contribute
antioxidant eects. However, this contribution is quite disputed, because only
modest amounts (just for avoring) of these vegetables are considered safe.
This limitation is suggested by the presence in these species of some
phytochemicals with potential adverse eects [46], such as apiole and
Apigenin derivatives [46]
Apigenin and
luteolin derivatives [51]
Apigenin, 1899 mg
Luteolin, 1142 mg
c1% (carvacrol, a-pinene,

cineole, borneol)
c1% (thymol, 3070%;

carvacrol, 315%;
a-pinene, a-terpineol, linalool
Satureya hortensis
(summer savory) and
Satureya montana
(winter savory)
Thymus vulgaris (thyme)
Apigenin and luteolin

derivatives [51]
Apigenin, 1899 mg
Luteolin, 1142 mg
derivatives [46]
Luteolin derivatives [51]
Luteolin, c4 mg
Apigenin, nevadensin,
salvigenin [77]
Apigenin and luteolin derivatives

Apigenin, 1899 mg
Luteolin, 1142 mg [51]
c1% (terpinenes, p-cymene,

carvacrol, estragole)
c0.5% (a-pinene, geraniol,
cineol, borneol)
c0.4% (menthol, 3040%;

menthone, 2030%; pulegone,
menthylacetate, limonene,
cineole)
c0.08% (linalool, c.45%;
estragol, c50%; methylcinnamate,
geraniol, cineole, safrole)
c0.6% (thymol, carvacrol,
monoterpene hydrocarbons,
borneol, linalool)
Flavonoidsa
Origanum majorana
(marjoram)
Rosmarinus ofcinalis
(rosemary)
Origanum vulgare
(oregano)
Ocimum basilicum (basil)
Labiatae
Mentha var. (mints)
Volatile oil main components [46]
Table 3 Selected Flavonoid Culinary Herbs
Caeic acid,
labiatic acid
Labiatic acid,
rosmarinic acid
Rosmarinic acid,
labiatic acid,
carnosic acid,
carnosol
Labiatic acid
Rosmarinic acid
Rosmarinic acid
Rosmarinic acid
Other phenolics [46]
62
Pietta et al.
14% (anethole, 7590%;

estragole, anise ketone,
monoterpene hydrocarbons)
47% (carvone, 5060%;

limonene, c40%)
0.41% (linalool,
decyl aldehyde, cineole)
25% (cuminaldehyde, ca. 50%;
monoterpene and
sesquiterpene hydrocarbons)
26% (anethole, 5070%;
fenchone, 1020%;
estragole, trace;
limonene, a-pinene)
0.050.3% (myristicin, c80%;
apiole, terpenes)
2.54% (carvone, 3560%;

a-pinene limonene,
phellandrene, eugenol, cineole)
Not detected
Apigenin,
luteolin derivatives;
hesperidin [46]
Quercetin kaempferol
derivatives [46]

derivatives [51]
Apigenin, 510630 mg
Luteolin, 04 mg
Quercetin, luteolin,
apigenin derivatives [46]
Coumarins
Quercetin derivatives [46]
Coumarins
Coumarins
Coumarins
Not detected
Coumarins
Other phenolics [46]
Quercetin, kaempferol,
isorhamnetin derivatives [51]
Quercetin, 48110 mg
Kaempferol, 1624 mg
Isorhamnetin, 1572 mg
Quercetin derivatives [46]
avonoidsa
The content of total avonoids of fresh vegetables, measured by high-performance liquid chromatographydiode array detection (HPLC-DAD) after acid
hydrolysis, is expressed as milligrams per 100 g fresh weight [51]. The avonoid content of savory and coriander, cumin, fennel, and anise seeds is not
known.
Pimpinella anisum
(anise seeds)
Petroselinum crispum
(parsley)
Foeniculum vulgare
(fennel seeds)
Carum carvi
(caraway seeds)
Coriandrum sativum
(coriander seeds)
Cuminum cyminum
(cumin seeds)
Umbelliferae
Anethum graveolens
(dill seeds)
Volatile oil main components [46]
Table 3 Selected avonoid Culinary Herbs
Flavonoids in Herbs
63
64
Pietta et al.
Table 4 m/z Values of Ions Produced by Liquid Chromatography Mass

Spectrometry of Flavonoid Glycosides Present in Some Culinary Herbs
HERB
[M-H]
Chives (Allium
schoenoprasum)
463
477
447
609
477
Dill (Anethum
graveolens)
Mint (Mentha var.)
Oregano (Origanum
vulgare)
Parsley (Petroselium
crispum)
Rosemary (Rosmarinus
ocinalis)
Thyme (Thymus
vulgaris)
Product
ions of
[M-H]
461
491
301
315
285
301
301,255,
179,151
285
315,300
593
579
609
285
271
301
577
607
269
299
795
591
799
635
635
563
593
605
635
503
283
283
285
269
593,299
269
299
563,269
593,299
285
463
447
287
285
461
285
489
285
445
609
269
285
Product ions
of aglycone
Suggested
namea
ND
300,271,151,107
ND
ND
179,151,121,107
Q-glu
I-glu
K-glu
Q-rut
Q-gluc
300,271,164,
151,136,107
175,151,133,107
ND
286,242,164,
151,125
151,149,117,107
284,256,177,
151,121
268,240,(151)
268,240,151,107
199,175,151,133
ND
284,137
151,149,117,107
284,256,(107)
151,149,117,107
284,256,(107)
L-acetyl-glu
K-gluc
I-gluc
151,135,107
175,151,149,
133,107
175,151,149,
133,107
175,151,(149),
133,107
151,149,117,107
L-rut
N-rut
H-rut
A-rut
D-rut
Ac-acetyl-glu-rut
Ac-rut
L-gly
A-acetyl-diglu
D-acetyl-apiosylglu
A-apiosylglu
D-apiosylglu
A-acetyl-apiosylglu
D-acetyl-apiosylglu
E-gluc
L-glu
L-gluc
L-acetyl-glu
A-gluc
L-diglu
Q, quercetin; I, isorhamnetin; K, kaempferol; L, luteolin; N, naringenin; H, hesperidin; A, apigenin;

Ac, acacetin; D, diosmetin; E, eriodictyol; Glu, glucoside; rut, rutinoside; gluc, glucuronide; gly,
glucoside.
Source: Adapted from Ref. 53.
Flavonoids in Herbs
65
myristicin (parsley), coumarins (anise, coriander, fennel, parsley), estragol

(anise, basil, fennel), safrole (basil), thuyone (sage), and thymol and carvacrol
(thyme, oregano, marjoram). Namely, coumarins are phototoxic; apiole and
myristicin are uterine stimulants and may cause damage to kidney; estragole
and safrole are alkylbenzenes with weak hepatotoxic potential; thujone is a
bicyclic diterpene (related to camphor) with neurotoxic potential; thymol and
carvacrol may induce nausea [4850,17]. Hence, only low consumption of
these culinary herbs (that means also low amounts of all phytochemicals,
including the antioxidant phenolics) is advisable. In terms of avonoid
composition, quercetin and kaempferol are the most abundant, followed by
luteolin, apigenin, hesperetin, and methoxylated avones [51]. The majority of
these avonoids are present as glycosides and in most studies are measured as
aglycones after acid hydrolysis [52]. In 2000, structural information on
avonoid glycosides present in commonly eaten herbs was obtained by
negative APCI-MS [53]. This approach provides the deprotonated molecular
ions (Mr) of glycosylated avonoids. In addition, source fragmentation of Mr
ions produces the aglycones, whose identity is conrmed by comparison of the
product-ion spectra obtained through low-energy collision-induced dissociation
(CID) MS-MS with those of aglycone standards (Table 4).
III.
CONCLUSIONS
It may be concluded that mass spectrometry allows the analysis of avonoids in

complex herbal matrices. Characterized by high specicity and sensitivity (up to
10 nM), ESI-MS and APCI-MS can produce ngerprints of herbal extracts
without prepurication steps. In addition, MS-MS ion trap methodology
provides reliable assignment of the compounds present in the extracts.
ACKNOWLEDGMENTS
Research stimulating this review was supplied by grants from Indena S.p.A
Italy and Specchiasol S.r.IItaly.
REFERENCES
1.
2.
Prior RL, Cao G. Flavonoids: diet and health relationships. Nutr Clin Care 2000;
3:279288.
Ielpo MTL, Basile A, Mirando R, Moscatello V, Nappo C, Sorbo S, Laghi E,
Ricciardi MM, Ricciardi L, Vuotto ML. Immunopharmacological properties of
avonoids. Fitoterapia 2000; 71:S101S109.
66
3.
Pietta et al.
Craig WJ. Health-promoting properties of common herbs. Am J Clin Nutr 1999;

70:491S499S.
4. Valerio LG, Kepa JK, Pickwell GV, Quattrochi LC. Induction of human
NADPH:quinone oxidoreductase (NQO1) gene expression by the avonol
quercetin. Toxicol Lett 2001; 119:4957.
5. Samman S, Wall PML, Farmakalidis E. Flavonoids and other phytochemicals
in relation to coronary disease. In: Basu TK, Temle NJ, Garg ML, eds. Antioxidants in Human Health and Disease. Wallingford, England: CABI, 1999:175
183.
6. Dugas AJ, Castaneda J, Bonin GC, Price KL, Fisher NH, Winston GW. Evaluation
of the total peroxyl radical scavenging capacity of avonoids: structure-activity
relationships. J Nat Prod 2000; 63:327331.
7. Zandi P, Gordon MH. Antioxidant activity of extracts from old tea leaves. Food
Chem 1999; 64:285288.
8. Pietta PG. Flavonoids as antioxidants. J Nat Prod 2000; 63:10351043.
9. Karakaya S, Nehir SEL. Quercetin, luteolin, apigenin and kaempferol contents of
some foods. Food Chem 1999; 66:289292.
10. Petersen J, Dwyer J. Flavonoids: dietary occurrence and biochemical activity. Nutr
Res 1998;18:19952018.
11. Pietta PG. Flavonoids in medicinal plants. In: Rice-Evans C, Packer L, eds.
Flavonoids in Health and Disease. New York: Marcel Dekker, 1998:61109.
12. Mills S, Bone K. Principles and Practice of Phytotherapy-Modern Herbal Medicine,
Edinburgh: Churchill Livingston, 2000.
13. Hudson J, Towers GHN. Phytomedicines as antivirals. Drugs Future 1999; 24:295
320.
14. Pietta PG, Pietta A. Fitomedicine e Nutrienti. 2d ed. Verona: Ricchiuto GM, 1998.
15. Lin JK, Liang YC, Lin-Shiau SY. Cancer chemoprevention by tea polyphenols
through mitotic signal transduction blockade. Biochem Pharmacol 1999; 58:911
915.
16. Ferguson LR. Role of plant polyphenolics in genomic stability. Mutat Res 2001;
47:89111.
17. Pietta PG. Phytomedicines: creating safer choices. In: Watson RR, ed.
Vegetables, Fruits and Herbs in Health Promotion. Boca Raton, FL: CRC Press,
2001:7384.
18. Pietta PG, Mauri PL. Analysis of avonoids in medicinal plants. Methods Enzymol
2001; 335:2646.
19. Pietta PG, Gardana C. Capillary electrophoresis and herbal drug analysis. In press.
20. Merkem HM, Beecher GR. Measurement of food avonoids by high performance
liquid chromatography. J Agric Food Chem 2000; 48:577599.
21. Mauri PL, Pietta PG. Electrospray characterization of selected medicinal plant
extracts. J Pharm Biomed Anal 2000; 23:6168.
22. Pietta PG, Gardana C, Cristoni S. Analytical methods for quality control of propolis.
Phytother Res. In press.
23. DeFeudis FV. Ginkgo biloba extract (EGb 761): from chemistry to the clinic.
Wiesbaden: Ullstein Medical, 1998.
Flavonoids in Herbs
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
67
Dutta-Roy AK. Inhibitory eect of ginkgo biloba extract on human platelet

aggregation. Platelets 1999; 10:298305.
Luo Y. Ginkgo biloba neuroprotection: Therapeutic implications in Alzehimers
disease. J Alzheimer Dis 2001; 3:401407.
Mauri PL, Pietta PG. Analysis of Hypericum perforatum extracts by high
performance liquid chromatography\electrospray mass spectrometry. Rapid Commun Mass Spectrom 2000; 14:9599.
Pietta PG, Gardana C, Pietta AM. Comparative evaluation of St. Johns wort from
dierent Italian regions. II Farmaco 2001; 56:491496.
Schrader E. Equivalence of St Johns wort extract (ZE 117) and uoxetine: a
randomized, controlled study in mild to moderate depression. Int Clin Psychopharmacol 2000; 15:6168.
Stevenson C, Ernst E. Safety of Hypericum in patients with depression: a
comparison with conventional antidepressant. CNS drugs 1999; 11:125132.
Morazzoni P, Bombardelli E. Vaccinium myrtillus. Fitoterapia 1996; 67:329.
Liu F, Ng TB. Antioxidative and free radical scavenging activities of selected
medicinal herbs. Life Sci 2000; 66:725735.
Pietta PG, Simonetti PL, Mauri PL. Antioxidant activity of selected medicinal
plants. J Agric Food Chem 1998; 46(11):44874490.
Mukhtar H, Ahmad N. Tea polyphenols: prevention of cancer and optimizing
health. Am J Clin Nutr 2000; 71(suppl):1698S1702S.
Mellenthin O, Galena R. Analysis of polyphenols using capillary zone electrophoresis and HPLC: detection of soy, lupin and pea protein in meat products.
J Agric Food Chem 1999; 47:594602.
Newall CA, Anderson LA, Phillison JD. Herbal medicines: a guide for health-care
professionals. London: Pharmaceutical Press, 1996.
Setchell, KD, Cassidy, A. Dietary isoavones: biological eects and relevance to
human health. J Nutr 1999; 129:425242639.
Kuiper GG, Lemmen JC, Carlsson B, Corton JC, Safe SH, van der Saag PT, van der
Burg B, Gustafsson JA. Interaction of estrogenic chemicals and phytoestrogens with
estrogen receptor beta. Endocrinology 1998; 138:42524263.
Hendrich S, Murphy PA. Isoavones, source and metabolism. In: Wildman REC,
ed. Handbook of Nutraceuticals and Functional Foods. Boca Raton, FL: CRC Press,
2001:5575.
Bidlack WR, Wang W. Designing functional foods to enhance health. In:
Bidlack WR, Omaye ST, Meskin MS, Topham DKW, eds. Phytochemicals as
Bioactive Agents. Lancaster, PA: Technomic, 2000:241270.
Mitchell JH. Phytoestrogens: involvement in breast and prostate cancer. In:
Wildman REC, ed. Handbook of Nutraceuticals and Functional Foods. Boca Raton,
FL: CRC Press, 2001:99112.
Banes S. What is the safe level of phytoestrogen intake? Phytoestrogens in food and
bone health. Final Meeting of the European Concerted Action on Phytoestrogens in
Food and Bone Health, Versailles, Oct 46, 2001.
Pietta PG, Gardana C, Pietta A. Propolis: a review. Fitoterapia 2002; 73(S1):720.
Packer L, Rimbach G, Virgili F. Antioxidant activity and biological properties of a
68
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
Pietta et al.
procyanidin-rich extract from pine bark, pycnogenol. Free Radic Biol Med 1999;
27:704724.
Lazarus SA, Adamson GE, Hammerstone JF, Schmitz HH. High performance liquid
chromatography\mass spectrometry analysis of procyanidins in foods and
beverages. J Agric Food Chem 1999; 47:36933701.
Gabetta B, Fuzzati N, Grino A, Lolla E, Pace R, Rulli T, Peterlongo F.
Characterization of procyanidins from grape seeds (Leucoselect). Fitoterapia 2000;
71(2):162175.
Leung AJ, Foster S. Encyclopedia of Common Natural Ingredients, 2nd ed. New
York: John Wiley, 1996.
Craig WJ. Health-promoting properties of common herbs. Am J Clin Nutr 1999;
70:491S499S.
Towers GHN, Page JE, Hudson JB. Light-mediated biological activities of natural
products from plants and fungi. Curr Org Chem 1997; 1:395414.
Gun M, Hobbs C, Upton R, Goldberg A. Botanical safety handbook. Boca Raton,
FL: CRC Press, 1997.
Brinker F. Herb contraindications and drug interactions. Sandy, OR: Eclectic
Medical, 1998.
Justesen U, Knuthsen P. Composition of avonoids in fresh herbs and calculation of
avonoid intake by use of herbs in traditional Danish dishes. Food Chem 2001;
73:245250.
Justesen U, Knuthsen P, Leth T. Quantitative analysis of avonols, avones and
avanones in fruits, vegetables and beverages by HPLC with photo-diode array and
mass spectrometric detection. J Chromatogr A 1998; A799:101110.
Justesen U. Negative atmospheric pressure chemical ionisation low-energy collision
activation mass spectrometry for the characterisation of avonoids in extracts of
fresh herbs. J Chromatogr A 2000; 902:369379.
Weggener T, Fintelmann V. Pharmacological properties and therapeutic prole of
artichoke. Wien Med Wochenschr 1999; 149:241247.
Bilberry fruit. In: Blumenthal M, Goldberg A, Brinckmann J, eds. Herbal MedicineExpanded Commission E Monographs. Newton, MA: Integrative Medicine
Communications, 2000:1621.
Calendula ower. In: Blumenthal M, Goldberg A, Brinckmann J, eds. Herbal
Medicine-Expanded Commission E Monographs. Newton, MA: Integrative
Medicine Communications, 2000:4446.
Chamomile ower. In: Blumenthal M, Goldberg A, Brinckmann J, eds. Herbal
Licorice root. In: Blumenthal M, Goldberg A, Brinckmann J, eds. Herbal MedicineExpanded Commission E Monographs. Newton, MA: Integrative Medicine
Elder ower. In: Blumenthal M, Goldberg A, Brinckmann J, eds. Herbal MedicineExpanded Commission E Monographs. Newton, MA: Integrative Medicine
Youdim KA, Martin A, Joseph JA. Incorporation of the elderberry anthocyanins by
Flavonoids in Herbs
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
69
endothelial cells increases protection against oxidative stress. Free Radic Biol Med
2000; 29:5160.
Ginkgo biloba leaf. In: Blumenthal M, Goldberg A, Brinckmann J, eds. Herbal
Goldenrod. In: Blumenthal M, Goldberg A, Brinckmann J, eds. Herbal MedicineExpanded Commission E Monographs. Newton, MA: Integrative Medicine
Hawtorn leaf with ower: Analytical, quality control and therapeutic monograph.
American Herbal Pharmacopoeia February 1999:129.
Tauchert M, Gildor A, Lipinski J. High-dose crataegus extract WS 1442 in the
treatment of NYHA stage II heart failure. Herz 1999; 24:465474.
Shibata S. A drug over the millenia: pharmacognosy, chemistry, and pharmacology
of licorice. J Pharm Sci Jpn 2000; 120:849962.
Schultz V, Hansel R, Tyler VE. Rational Phytotherapy 200. Berlin: Springer Verlag,
2001.
Linden ower. In: Blumenthal M, Goldberg A, Brinckmann J, eds. Herbal
Meadowsweet. In: Blumenthal M, Goldberg A, Brinckmann J, eds. Herbal
Milk thistle fruit. In: Blumenthal M, Goldberg A, Brinckmann J, eds. Herbal
Orange peel. In: Blumenthal M, Goldberg A, Brinckmann J, eds. Herbal MedicineExpanded Commission E Monographs. Newton, MA: Integrative Medicine
Berdonces JL. Attention decit and infantile hyperactivity. Rev Enfermeria 2001;
24:1114.
Nathan PJ. Hypericum perforatum: a non-selective reuptake inhibitor? A review of
the recent advances in its pharmacology. J Psychopharmacol 2001; 15:4754.
Willow bark. Analytical, quality control and therapeutic monograph. American
Herbal Pharmacopoeia December 1999:115.
Schmid B, Ludtke R, Selbmann HK, Kotter I, Tschirdewahn B, Schaner W, Heide
L. Ecacy and tolerability of a standardized willow bark extract in patients with
osteoarthritis: randomized placebo-controlled, double blind clinical trial. Phytother
Res 2001; 15:344350.
MakKay D. Hemorroids and varicose veins: a review of treatments options. Altern
Med Rev 2001; 6:126140.
Grimme H, Augustin M. Phytotherapy in chronic dermatoses and wounds: what is
the evidence? Forsch Komplementarmed 1999; 6:58.
Grayer RJ, Veitch NC, Kite KG, Price AM, Kobubun T. Distribution of
8-oxygenated leaf-surface avones in the genus Ocimum. Phytochemistry 2001;
56:559567.
3
The Relationship Between
the Phenolic Composition
and the Antioxidant Activity
of Fruits and Vegetables
Anna R. Proteggente and Catherine A. Rice-Evans
Kings College London
London, England
Sheila Wiseman and Frans H. M. M. van de Put

Unilever Health Institute
Vlaardingen, The Netherlands
I.
INTRODUCTION
Over the years, several benecial properties (anti-inammatory, anticarcinogenic,

antimutagenic) have been attributed to phenolic compounds [12], and a number
of studies have suggested that consumption of fruits and vegetables can reduce the
risk of cardiovascular diseases and cancer, potentially through the biological
actions of the phenolic components [38] as well as antioxidant vitamins.
The antioxidant activity is probably the most extensively studied aspect of
the bioactivity of phenolic compounds. Pure phenolic compounds are powerful
free radical scavengers in vitro, and this property has been demonstrated both
with synthetic free radicals [e.g., 2,2V-azinobis-(3-ethylbenzothiazoline-6sulfonic acid) (ABTS) and (2,2-diphenyl-l-picrylhydrazyl) (DPPH)] and with
physiologically relevant peroxyl radicals, hydroxyl radicals, and superoxide [9
11]. However, although increasing evidence for the role of oxidative damage in
pathological disorders suggests that antioxidant ability might be an important
factor in the eects of avonoids in vivo [12], one cannot infer that the potential
71
72
Proteggente et al.
benecial eects of phenolics exclusively depend on their antioxidant activity. In

this chapter, the relationships among the phenolic content and composition of
regularly consumed fruits and vegetables, their vitamin C concentrations, and
antioxidant potentials in vitro are reviewed [1320]. In addition, a comparison of
their H-donating antioxidant activities is made through three assays: Trolox
equivalent antioxidant capacity (TEAC), oxygen radical absorbance capacity
(ORAC), and ferric reducing ability of plasma (FRAP). The information that is
provided represents a useful compendium of the composition and the content of
major phenolic compounds in some common fruit and vegetables. This information can enhance understanding of the potential for uptake of specic phenolic
compounds and determine whether they may be present in vivo at concentrations
able to exert protective eects. Thus, by identifying the major components in
foods and by characterizing their biological activities and bioavailable forms, it
will be possible to provide useful information to illuminate the potential role of
these dietary components in protection against chronic degenerative diseases.
II.
ANTIOXIDANT ACTIVITY IN VITRO OF FRUITS

AND VEGETABLES
Over the years, a variety of assays to measure the total antioxidant capacity of
pure substances, biological uids, food extracts, and beverages have been
developed. Among the most commonly used are the Trolox equivalent antioxidant capacity (TEAC), oxygen radical absorbance capacity (ORAC), and
ferric reducing ability of plasma (FRAP) assays.
The TEAC assay, as in the modied version by Re and associates [21], is
based on the scavenging of the ABTS, 2,2V-azinobis-(3-ethylbenzothiazoline-6sulfonic acid) diammonium salt by hydrogen-donating compounds. The ORAC
assay, according to the method of Cao and colleagues [22], is based on the
oxidation of h-phycoerythrin, a uorescent protein, by 2,2V-azobis(2-methylpropionamidine dihydrochloride) (ABAP), a peroxyl radical generator, and the
inhibition of the formation of the uorescent protein by antioxidants. The FRAP,
according to the method of Benzie and Strain [23], is based on the reduction of a
ferric 2,4,6-tripyridyl-s-triazine complex to the ferrous form by antioxidants.
Although the three assays dier in the radical used, the radical-generating
system, and the kinetic of reaction, all are based on reactions occurring in a
hydrophilic system and refer the radical-scavenging ability of test mixtures to a
Trolox standard.
A comparison of antioxidant activities reported in literature, for a number
of regularly consumed fruits and vegetables, and those obtained by using the
TEAC, ORAC, and FRAP assays is presented in Tables 1, 2, and 3, respectively.
A 2002 study utilized these assays to assess comparatively the antioxidant
Phenol Composition and Antioxidant Activity
73
Table 1 Antioxidant Capacity, Trolox Equivalent Antioxidant

Capacity of Fruits and Vegetablesa
Fruit / vegetable
Strawberry
Raspberry
Red plum
Grapefruit
Orange
Red cabbage
Broccoli
Onion
Green grape
Spinach
Green cabbage
Pea
Cauliower
Leek
Lettuce
Pear
Apple
Peach
Banana
Tomato
a
b
Proteggente et al. [13]

2591
1846
1825
861
849
1377
648
532
594
757
492
440
295
240
171
282
343
244
181
255
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
68
10
28
53
25
49
25
29
72
54
18
18
16
11
12
19
13
9
39
14
Other studies
1725 F 103 [25]
580 F 320 [17]
640 F 270 [17]

160 F 60 [17]
270 F 30 [24]b
Micromolar Trolox equivalents/100 g fresh weight (FW).

Extrapolated from gure in original Ref. 24.
potentials of polyphenol-rich extracts of fruits and vegetables [13]. The hierarchy

of the antioxidant activities (Amol Trolox equivalents/100 g [FW] uncooked
portion size) for some common fruits and vegetables, as determined by these
authors by using the TEAC assay, was as follows: strawberries J raspberries =
red plums J red cabbages { grapefruit = oranges > spinach > broccoli >
green grapes i onions > green cabbages > peas > apples > cauliowers i pears
> tomatoes i peaches = leeks > bananas i lettuce. The hierarchy of antioxidant
potentials determined by the ORAC assay was somewhat dierent from the one
obtained by using the TEAC assay (red plums i strawberries > red cabbage >
oranges i raspberries J spinach > grapefruit > broccoli > green cabbage >
onions > green grapes > peaches > peas J pears i apples > cauliower i
tomatoes i leeks > bananas i lettuce). However, as for the TEACs hierarchy,
the anthocyanin-rich fruits exerted the highest antioxidant activity, followed
by the citrus fruits and the majority of the avonol-rich fruit and vegetables,
whereas the hydroxycinnamate-containing fruits and vegetables consistently
74
Proteggente et al.
Table 2 Antioxidant Capacity, Oxygen Radical Absorbance

Capacity, of Fruits and Vegetablesa
Fruit / Vegetable
Other studies
Strawberry
2437 F 95
Raspberry
1849 F 232
Red plum
Grapefruit
Orange
2564 F 185
1447 F 67
1904 F 259
1536 F 238 [28]

2060 F 233 [26]
2680 [27]
1820 F 80 [20]
2140 F 224 [26]
949 F 67 [28]
483 F 18 [26]
750 F 101 [26]
1970 [27]
Red cabbage
Broccoli
Onion
Green grape
Spinach
Green cabbage
Peas
Cauliower
Leek
Lettuce
Pear
2124
1335
988
872
1655
1180
704
425
413
319
587
F
F
F
F
F
F
F
F
F
F
F
68
62
30
48
115
68
62
44
15
37
50
Apple
560 F 18
Peach
Banana
764 F 49
331 F 59
Tomato
420 F 39
1270
446
1940
490
[27]
F 106 [28]
[27]
[27]
790 [27]
400
134
460
218
490
[27]
F 6 [28]
[27]
F 35 [28]
[27]
221
460
189
450
F 19 [28]
[27]
F 12 [28]
[27]
Micromolar Trolox equivalents/100 g fresh weight (FW).
elicited lower antioxidant activity. Also the hierarchy of the antioxidant activities
determined by the FRAP assay mirrored those observed by using the TEAC and
ORAC assays: strawberries { raspberries > red plums > red cabbage J
oranges i spinach > broccoli = grapefruits i green cabbage > green grapes >
apples i onions itomatoes i peaches i pears > cauliower = peas > bananas =
leeks > lettuce.
Previously reported TEAC values for onion and tomato extracts [17,24]
and for raspberry extracts [25] were consistent with those obtained by
Proteggente and coworkers [13]; the TEAC value for apple was dierent
because of varietal dierences. As for reports on antioxidant potentials that
75
Table 3 Antioxidant Capacity, Ferric

Reducing Ability of Plasma, of Fruits
and Vegetablesa
Fruit / Vegetable
Strawberry
Raspberry
Red plum
Grapefruit
Orange
Red cabbage
Broccoli
Onion
Green grape
Spinach
Green cabbage
Pea
Cauliower
Leek
Lettuce
Pear
Apple
Peach
Banana
Tomato
a

3352
2325
2057
829
1181
1870
833
369
519
1009
694
251
259
160
124
315
394
336
164
344
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
38
53
25
6
6
18
16
13
48
35
14
9
5
1
7
24
8
4
32
7
Micromolar Fe2+ equivalents/100 g fresh weight

(FW).
have been determined by using the ORAC assays, the values for strawberry and
raspberry measured by Kalt and associates [26] are in agreement with those
indicated by Proteggente and colleagues [13]. Similar ORAC values for a
number of other fruits and vegetables, such as strawberry, orange, apple, pear,
banana, onion, spinach, green cabbage, cauliower, and lettuce, have been
indicated [27]. However, previously reported ORAC values of plum, orange,
grapefruit, tomato, white grape, apple, pear, and banana [28] are much lower.
Thus, the literature data for the antioxidant activities of some fruits and
vegetables, obtained by the same or similar in vitro assays, consistently indicate
a high potential of certain foods such as berries, green leafy vegetables, and
citrus fruits. However, the ndings of dierent authors are not always
comparable because of the enormous variability of the antioxidant content in
fruits and vegetables, not only between varieties, but also in the same variety,
depending on the cultivation site, the climate, the stage of maturity of the fruit
or vegetable, and the sample preparation and extraction procedures. However,
76
Proteggente et al.
the data reviewed here present a compendium of the available information on

the antioxidant potentials of some of the most commonly consumed fruits and
vegetables to enhance understanding of the nutritional importance of some
foods in terms of antioxidant defense.
III.
PHENOLIC CONTENT OF FRUITS AND VEGETABLES
When evaluating the antioxidant potential of fruits and vegetables, it is of interest

to consider their total phenolic content, as presented in Table 4. Strawberry
extracts have been reported to possess the highest total phenolic content (400
mg/100 g FW), with respect not only to the other anthocyanin-rich fruits but also
to all the other fruit and vegetable extracts (Table 4) [13]. However, total
Table 4 Total Phenolic Content of Fruits and Vegetablesa
Fruit / Vegetable
Others
Strawberry
330 F 4
Raspberry
228 F 6
161265 [29]
86 [26]
95152, 91278 [20]
265303 [29]
121 [26]
Red plum
Grapefruit
Orange
Red cabbage
Broccoli
Onion
Green grape
Spinach
Green cabbage
Pea
Cauliower
Leek
Lettuce
Pear
Apple
Peach
Banana
Tomato
320
150
126
158
128
88
80
72
58
32
30
22
14
60
48
38
38
30
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
12
4
6
4
4
1
4
1
1
1
1
1
1
3
1
1
4
1
104b [18]
70116 [18]
49 [18]
52 [18]
23 [18]
110600 [30]
4377 [31]
38 [18]
Milligrams gallic acid equivalents/100 g fresh weight (FW), unless otherwise

stated.
b
Catechin equivalents.
77
phenolics in strawberry and raspberry extracts have also been indicated to range
from 161 to 265 mg and from 265 to 303 mg gallic acid equivalents (GAE)/100
g FW, respectively, depending on the method of extraction [29]. Furthermore,
Kalt and coworkers [26] have determined the total phenolic content of fresh
strawberries and raspberries and found even lower levels of phenolics: 86 and
121 mg GAE/100 g FW, in strawberries and raspberries, respectively. These
large dierences, however, should not be surprising when considering that the
phenolic content can vary enormously, depending not only on the variety of the
fruit or vegetable but also on their stage of maturity. For example, Wang and Lin
[20] have reported that the total phenolic content ranged from 95 to 152 mg
GAE/100 g FW in dierent varieties of ripe strawberries and from 91 to 278 mg
GAE/100 g FW, depending on ripeness.
For orange and grapefruit, which are rich in avanones, Proteggente and
associates [13] have found that the total phenolic content (126 mg and 150 mg/
100 g FW, respectively) was similar and much lower than that measured for berry
fruits. Other data on the phenolic content of these fruits were not available in the
literature.
The total phenolic content of apples of dierent varieties has been
reported to be in the range 110600 mg epicatechin equivalents/100 g FW
[30], which is much higher than the level of 48 mg GAE/100 g FW [13].
However, this large dierence may depend on the use of dierent standards.
The total phenolic content of the fruits such as pear, peach, and tomato was
found to range between 30 and 60 mg GAE/100 g FW by Proteggente and
colleagues [13], considerably lower than the levels measured by the same
authors for the berry and citrus fruits and most of the green leafy vegetables.
The measurements of total phenolic content of peach and tomato extracts by
Proteggente and coworkers [13] are essentially in the same range of levels
previously reported by Chang and associates [31] for peach and by Vinson and
colleagues [18] for tomato; no other literature data were available for a
comparison of the phenolic content of pears (Table 4). Among the vegetables,
Proteggente and coworkers [13] observed that broccoli extracts had the highest
total phenolic content (128 mg GAE/100 g FW) and lettuce extracts had the
lowest phenolic content (14 mg GAE/100 g FW). Similarly, Vinson and
associates [18] have found a total phenolic content, as catechin equivalents,
of 104 and 23 mg/100 g for broccoli and lettuce, respectively, and their values
for onion, spinach, and green cabbage are consistent with those reported by
Proggente and coworkers [13] (Table 4).
Furthermore, Proteggente and colleagues [13] have found that cauliower,
peas, and banana had a low total phenolic content, 30, 32, and 38 mg GAE/100 g
FW, respectively, whereas red cabbage extracts showed an appreciable phenolic
content (158 mg GAE/100 g FW), higher than that of some of the avanone- and
avanol-rich fruits and vegetables (Table 4).
78
Proteggente et al.
Figure 1 Representative HPLC-DAD (monitored over the range 200750 nm)

chromatogram of an anthocyanin-rich extract, strawberry. Peaks were derived at
(A) 320 nm and (B) 520 nm. HPLC-DAD, high-performance liquid chromatographydiode array detection.
IV.
79
PHENOLIC COMPOSITION OF FRUITS

AND VEGETABLES
Some of the most common fruits and vegetables containing four specic
families of phenolic compounds, anthocyanins, avanones, avonols, and
hydroxycinnamates, are characterized for their individual major components
and compared with reports from the leading literature of Macheix and coworkers
[32] and other researchers.
Anthocyanins have been found to be the major phenolics in strawberry,
raspberry, and red plum extracts (hydroxycinnamic acid derivatives and quercetin
and kaempferol conjugates are also contained in these fruits) [13], consistently
with previous reports by Macheix and associates [32]. More specically,
pelargonidin-3-glucoside was indicated as the major phenolic compound in
strawberry extracts, together with appreciable amounts of cyanidin-3-glucoside
(Fig. 1B), and cyanidin-3-glucoside and cyanidin-3-sophoroside were the major
components in red plum and raspberry extracts, respectively; red plum extracts
also contained minor amounts of cyanidin-3-rutinoside [13]. A summary of the
total anthocyanin content reported in the literature for strawberries, raspberries,
and red plums is presented in Table 5, which generally indicates good agreement
in the ndings of various authors [13,14,26,32,33].
Comparative total amounts of avanones in orange and grapefruit reported
in the literature are given in Table 6. Proteggente and colleagues [13] have
reported hesperidin as the major phenolic component in orange extracts,
accompanied by appreciable amounts of narirutin and neohesperedin (Fig. 2),
and naringin as the most important avanone and narirutin the second major
identiable component in grapefruit extracts. The data by Proteggente and
Table 5 Anthocyanin Content of Red Fruitsa
Reference
Cliord [33]
Kahkonen et al. [14]b
Kalt et al. [26]c
Macheix et al. [32]
a
Red plum
Strawberry
Raspberry
19.4
225
1.953
11.9
1535
20.4
7.6
2870
28.9
1060
22.9
41.4
23.6
Milligrams/100 grams fresh weight.

Data converted from milligrams/100 g dry weight reported in the original
reference, using a 9.7% and 14.9% dry matter value for strawberry and
raspberry, respectively.
c
Data converted from micromoles malvidin-3-glucoside/g fresh weight
reported in the original reference into milligrams malvidin-3-glucoside/
100 g fresh weight.
b
80
Proteggente et al.
Table 6 Flavanone Content of Citrus Fruitsa

Reference
Orange
Grapefruit

Tomas-Barberan and Cliord [35]b
Justesen et al. [44]c
Bronner and Beecher [34]d
Macheix et al. [32]e
130.9
140319
67.3
144180
270600
226.3
96543
262268
170280
Milligrams/100 g fresh weights.

Data obtained from the sum of hesperidin and narirutin for fresh and
hand-squeezed orange juices and naringin and narirutin for grapefruit
juices reported in the original reference.
c
Data obtained from the sum of hesperetin and naringenin (the aglycone
forms of hesperidin and narirutin, respectively) for orange pulp
reported in the original reference.
d
Data obtained from the sum of hesperidin and narirutin for orange
juice concentrates and naringin and narirutin for grapefruit juice
concentrates reported in the original reference.
e
Hesperidin only for oranges and naringin only for grapefruits (whole
fruit) reported in the original reference.
b

chromatogram of a flavanone-rich extract, orange. Peaks were derived at 320 nm.
HPLC-DAD, high-performance liquid chromatographydiode array detection.
81
coworkers [13] for citrus fruits are reasonably consistent with other reports on the
avanone content of fresh juices and concentrates [34,35], when accounting for
the diversity of the material analyzed (whole fruit, pulp only, fresh juice or
concentrates, etc.) and the method of analysis.
The avonol levels found in the literature for some fruits and vegetables are
reported in Table 7. Proteggente and associates [13] indicated quercetin-3,4Vdiglucoside and quercetin-4V-glucoside, together with traces of quercetin-3glucoside, as the major phenolic compounds in onion (Fig. 3). The amount of
avonols in onion reported by these authors is signicantly higher than that
indicated by Hollman and Arts [36] but similar to that reported by Moon and
colleagues [37]. It should, however, be noted that Proteggente and coworkers [13]
and Moon and associates [37] quantied on the basis of quercetin glucoside,
whereas Hollman and Arts [36] quantied on the basis of the aglycone quercetin
after acid hydrolysis. Furthermore, Proteggente and colleagues [13] quantied all
peaks relative to quercetin conjugates as quercetin-3-glucoside, and this method
may have resulted in an overestimation of the relative amounts of the three major
glucosides, particularly of the diglucoside. Therefore, the avonol levels reported
by these authors for broccoli, spinach, and green cabbage, quantied on the basis
of quercetin-3-glucoside, are also likely to be overestimated. Of course there
would also be dierences due to the use of dierent varieties of onions.
Also, Proteggente and coworkers [13] have found quercetin conjugates, as
identied from the spectral characteristics, to be the main components in lettuce,
which also contained small amounts of chlorogenic acid and anthocyanin
Table 7 Flavonol Content of Fruits and Vegetablesa
Reference
Proteggente
et al. [13]
Hollman and
Arts [36]b
Moon
et al. [37]c
Hertog
et al. [64]d
a
Green
grape
Onion
Leek
Lettuce
Green
Broccoli Spinach cabbage
1.3
65.9
91.2
16.8
7.1
52.5
15.0
34
1.47.9
3.7
50.5
1.2
34.7 F 6.3 3 F 2.3 1.4 F 1.4
10.2
<3
<3
Milligrams/100 g fresh weight.

Data obtained for quercetin, the aglycone, after acid hydrolysis, in the original reference.
c
Data obtained from the sum of the individual amounts of quercetin-3,4V-diglucoside, quercetin-4Vglucoside, and quercetin-3-glucoside reported in the original reference.
d
Data obtained for quercetin and kaempferol aglycones after acid hydrolysis in the original
reference.
b
82
Proteggente et al.

chromatogram of a flavonol-rich extract, onion. Peaks were derived at 320 nm.
HPLC-DAD, high-performance liquid chromatographydiode array detection.
conjugates in the red varieties, and kaempferol conjugates in leek, consistently with
ndings by Fattorusso and associates [38]. However, the former authors, who were
using high-performance liquid chromatography (HPLC) with diode array detection, could not precisely identify the individual components of these vegetables,
because of the complexity of their avonol glycoside mixtures, in which glycosides
are often acylated with hydroxycinnamic acids. In fact, Fattorusso and colleagues
[38], by using ultraviolet (UV) and nuclear magnetic resonance (NMR) spectroscopy, have identied, but not quantied, ve avonoid glucosides based on
the kaempferol aglycone, some of which were acylated with a 3-methoxy-4hydroxycinnamoyl moiety, in leek extracts. Others have also reported the presence
of 3-O-h-D-sophoroside-7-O-h-D-glucosides of kaempferol and quercetin, with
and without further acylation with ferulic, sinapic, and caeic acids, in cabbage by
using HPLC with mass spectrometric detection but with no indication of the
amounts [39,40]. Similarly, acylated avonol glycosides and gentiobiosides,
derivatives of spinacetin and patuletin, have been isolated from spinach leaves [41].
A summary of the levels of hydroxycinnamate derivatives and other
components among the hydroxycinnamate-containing fruits and vegetables is
presented in Table 8. Chlorogenic acid was identied as one of the major
phenolic components in pear, apple, tomato, and peach/nectarine by Proteggente
0.12.2
1.4f
2f
4.5f
<0.3f
0.8f
3.6 F 1.9f
<0.4f
6.2134
17.840.5e
6.451.8
4.2060g

Average of three dierent cultivars.
c
In apple pomace, data converted from milligrams per kilogram dry weight reported in the original reference, based on a 14.3% dry matter value.
d
In whole fruit.
e
In peel.
f
Quercetin only, after acid hydrolysis.
g
In the esh and in the peel, respectively.
h
In the skin.
6.4h
6.238.5d
7.2c
13.5c
15.0c
4.5
5.4
3.1
3.5
2.6
3.1
4.5
0.8
3.5
8.733h
628d
1.8b
0.8b
Tomas-Barberan
and Clifford
Stewart Clifford Justesen
Hertog
Macheix
[35]
et al. [43]
[66]
et al. [44] et al. [64] et al. [32]
5.1
2.6
Proteggente Schieber
et al. [13] et al. [65]
Hydroxycinnamates and Miscellaneous Phenolic Content of Certain Fruitsa
Hydroxycinnamates
Flavonols
Others
Apple
Hydroxycinnamates
Flavonols
Others
Tomato
Hydroxycinnamates
Flavonols
Others
Peach/
Hydroxycinnamates
Nectarine Flavonols
Others
Pear
Table 8

83
84
Proteggente et al.
and coworkers [13]. These authors also found that pear extracts contained
quercetin-3-glucoside and that other major components in apple extracts were
rutin, avonol conjugates, and phloridzin (Fig. 4). Literature data indicate a great
variability in phenolic content of apples and pears (Table 8), which may be due to
the lack of appropriate standards and the consequent unspecic quantication of
the various chlorogenic acid isomers, such as 4V- and 3V-caeoylquinic acids, and
of p-coumaroylquinic acids, feruloylquinic acids, p-coumaroylmalic esters, and
hydroxycinnamic acid glucose derivatives that have been reported to be present
in these fruits [32,42].
As for tomatoes, Proteggente and associates [13] indicated that they also
contained rutin and chalconaringenin, but no naringenin or naringenin glucosides,
in agreement with other reports [35,43,44].
Thus, although the available information on the individual phenolic
composition of fruits and vegetables is very fragmentary and still insucient,
from the data summarized in this section it is possible to form an idea of which
are those components of fruit and vegetables that are most likely to have a dietary
importance and whose potential biological actions would therefore be of interest.

chromatogram of a hydroxycinnamate-rich extract, apple. Peaks were derived
at 320 nm. HPLC-DAD, high-performance liquid chromatographydiode
array detection.
V.
85
VITAMIN C CONTENT OF FRUITS AND VEGETABLES
When considering the antioxidant potential of fresh fruits and vegetables, it is

important also to evaluate the vitamin C content of these foods, because ascorbic
acid is a very eective scavenger of free radicals both in vitro and in vivo. A
comparison of the total vitamin C content of some common fruits and vegetables
is shown in Table 9. As can be seen, the levels from the dierent reports [13,45]
are generally consistent, except for most of the green vegetables, such as
broccoli, spinach, lettuce, and cauliower. This may be accounted for by losses
of vitamin C during the freeze-drying used in the preparation of the material,
although this eect would be expected to apply in all cases.
According to Table 9, strawberries contained the highest total vitamin C
content (61 mg/100 g FW) of the fruits and vegetables analyzed, followed by the
citrus fruits (52 and 46 mg/100 g FW for grapefruit and orange, respectively).
Broccoli had a vitamin C content comparable to that of the citrus fruits, 45 mg/
100 g FW, and red cabbage also showed an appreciable amount of vitamin C (37
Table 9 Total Vitamin C in Fruits and Vegetablesa

Fruit / Vegetable
Strawberry
Raspberry
Red plum
Grapefruit
Orange
Red cabbage
Broccoli
Onion
Green grape
Spinach
Green cabbage
Pea
Cauliower
Leek
Lettuce
Pear
Apple
Peach
Banana
Tomato
a
b
Holland et al. [45]
61
26
5
52
46
37
45
6
2
7
28
22
15
16
<2
3
6
6
10
18
4090
32
4
36
4479
4060b
87
5
3
26
4060a
24
43
17
1030
3
3
3137
10
1030

Includes both red and green cabbage.
86
Proteggente et al.
mg/100 g FW). For the other fruits and vegetables reported by Proteggente and
colleagues [13], the vitamin C content ranged from <2 mg/100 g FW for lettuce
to 28 mg/100 g FW for green cabbage, with pear, green grape, apple, peach, red
plum, onion, spinach, and banana at the lower end and raspberry, pea, tomato,
leek, and cauliower at the higher end of the range.
VI.
RELATIONSHIPS AMONG ANTIOXIDANT ACTIVITY,

PHENOLIC CONTENT AND COMPOSITION, AND
VITAMIN C CONTENT OF FRUITS AND VEGETABLES
A number of studies have shown that the antioxidant capacity of fruits and
vegetables is strongly correlated to both vitamin C concentration and total
phenolic content. For example, Gardner and coworkers [19] have assessed the
antioxidant potential of various fruit juices, through their ability to reduce a
synthetic free radical, potassium nitrosodisulfonate (by using electron spin
resonance (ESR)), and Fe(III) (by using FRAP) and found a strong correlation
with both phenolics and vitamin C (r = 0.970.99 and 0.900.93, respectively,
depending on the procedure used to determine the antioxidant capacity). Also,
Prior and colleagues [15] observed a good correlation between total phenolic
concentrations (r = 0.84), as well as the anthocyanin content (r = 0.77), and
antioxidant activity (ORAC) of dierent cultivars of Vaccinium species. Furthermore, Guo and colleagues [27] and Velogliu and associates [16] have
observed a strong correlation between total phenolic content and antioxidant
activity among a variety of fruits, vegetables, and grain products. Consistently,
Proteggente and colleagues [13] have shown a good correlation between total
phenolic and vitamin C content and all the three measurements of antioxidant
activity they have used, particularly with the TEAC assay (Fig. 5). These authors
not only indicated the existence of a correlation between the phenolic content
and the antioxidant activity but also strongly suggested that the antioxidant
capacity is closely related to the phenolic composition, i.e., to the prevalence of
a particular class of phenolic components in a specic fruit or vegetable. In fact,
the hierarchy of antioxidant activities of the fruits and vegetables studied
showed that fruits that are rich in anthocyanins have a higher antioxidant
Figure 5 Correlation between total phenolics (milligrams gallic acid equivalents/

100 g and measurements of antioxidant activity, TEAC, FRAP, and ORAC, of fruit
and vegetables extracts. The correlation coefficients (r) and the correspondent
significance values (P ) are indicated. TEAC, Trolox equivalent antioxidant
capacity; FRAP, ferric reducing ability of plasma; ORAC, oxygen radical
absorbance capacity.
87
88
Proteggente et al.
potential than those rich in avanones, avanols, and hydroxycinnamates. This

pattern appears to be consistent with the hierarchy of antioxidant potential
observed for the diverse classes of phenolic compounds in a number of in vitro
systems. In fact, the anthocyanins generally have shown higher antioxidant
activity than other phenolic compounds, such as avonols and avanones, when
tested in in vitro systems [9,21,46,47].
However, when considering the antioxidant potential of fruits and vegetables, other aspects than the phenolic content must be taken into account, such as
the contribution of vitamin C to the total antioxidant activity of hydrophilic
extracts and the combined actions by the phenolic and the vitamin C components.
For example, the data from Proteggente and colleagues [13] showed that the
antioxidant activities of orange and grapefruit were much lower than those
measured for berry fruits, as expected by the lower antioxidant eciency of
avanones, in which citrus fruits are rich, compared to that of anthocyanins and
avonols when evaluated in in vitro systems [9,11,21]. However, the antioxidant
potential of orange and grapefruit was appreciably higher than that observed for
avonol-rich fruits and vegetables, although pure avonols, which possess all the
structural features, such as the o-dihydroxy structure in the B ring, the 2,3 double
bond in conjugation with a 4-oxo function in the C ring, and the 3- 5-OH groups
with 4-oxo function in the A and C rings, that have been identied as
determinants for maximal radical scavenging ability [9], have demonstrated
higher antioxidant potential than avanones in in vitro systems. The nding that
the antioxidant activity observed for the citrus fruits is higher than expected on
the basis of the phenolic quality and quantity is likely to be due to the high
vitamin C content of citrus fruits with respect to avonol-rich fruit and
vegetables. The vitamin C content of avonol-containing fruits and vegetables
reported by Proteggente and coworkers [13] was in fact very low, and therefore it
might be deduced that the phenolic components almost exclusively accounted for
the antioxidant activities of these foods. Consistently, these authors reported that
extracts from hydroxycinnamate-containing fruits and vegetables had the lowest
phenolic and vitamin C content, reected in the lowest antioxidant activity, as
predicted by the low antioxidant potential of the hydroxycinnamic acids [9].
VII.
DIETARY BURDEN OF PHENOLIC-RICH FRUITS

AND VEGETABLES
In 1983 the UK total diet study estimated that the total average intake for
selected avonols, quercetin, kaempferol, apigenin, and luteolin, from six food
groups, green and other vegetables, canned vegetables, fresh fruits, fruit
products and beverages, was 30 mg/day, and quercetin contributed 64% of this
total. Beverages contributed 82% of the total intake of these avonoids; in
89
particular, tea provided up to 92% of avonoids from beverages. More recent

estimates of avonoid dietary intake range from 10 to 100 mg/day, depending on
the population studied and the technique used [48]. Estimated average intakes of
selected avonols and avones from the UK total diet study are similar to the 23
mg/day estimated average intakes of selected avonols, avones, and avanones
from a Dutch food survey [49].
More recently, Scalbert and Williamson [50] have reviewed the main
dietary sources of phenolic compounds and have calculated that the total daily
intake, for a given diet containing some common fruits (i.e., apple and cherry),
vegetables (i.e., potato, tomato, lettuce, and onion), and beverages (i.e., orange
juice, red wine, coee, and black tea) as well as other foods (i.e., wheat bran and
dark chocolate), was equivalent to f1 g/day. These authors also indicated that, as
a general trend, phenolic acids account for approximately one-third of the total
phenolics, and avonols, avanols, avanones, and anthocyanins for two-thirds.
However, the proportion in which the dierent classes of phenolics are ingested
varies greatly, according to the foodstus consumed. Therefore, comprehensive
surveys of the content of the important phenolic classes in common foods are
essential in estimating the dietary phenolic intake and in investigating the impact
that dietary phenolics might have in terms of disease prevention and /or treatment.
VIII.
IS THERE A POTENTIAL FOR ANTIOXIDANT ACTIVITY

OF COMPONENTS OF FRUITS AND VEGETABLES
IN VIVO?
The data presented here indicate that red fruits, strawberries in particular, have an
enormous antioxidant potential in vitro, when compared with other types of fruits
such as citrus or apples, for both high anthocyanin and high vitamin C content,
and therefore could represent an excellent source of dietary antioxidants.
Recent studies are starting to build evidence that these foods help to
improve the antioxidant potential in vivo. For example, Pedersen and associates
[51] have shown that consumption of cranberry juice was able to increase the
plasma antioxidant capacity of healthy female volunteers, by determining a 30%
increase in vitamin C and a small but signicant increase in total phenols plasma
concentration.
Furthermore, a study by Manuel y Keenoy and colleagues [52] has
shown that supplementation with diosmin (90%) and hesperidin (10%),
avonoids found in lemons and oranges, respectively, induced a signicant
decrease in hemoglobin A1c in type 1 diabetic patients that was not related to
the glycemic control. In these patients, the decrease in hemoglobin A1c was
accompanied by an increase in glutathione peroxidase activity and in the
lag time of the copper-induced in vitro oxidability of nonhigh-density
90
Proteggente et al.
lipoproteins (non-HDLs), suggesting that the avonoid-induced decrease in

glycation was associated with an improvement of the antioxidant status of the
patients, specically of the antioxidant component represented by the thiolcontaining proteins. However, a more recent study has shown that supplementation with 500 mg/day of rutin, quercetin-3-rutinoside, had no eect on
plasma antioxidant status or markers of oxidative stress in healthy female
volunteers, although the plasma levels of quercetin, kaempferol, and isorhamnetin were signicantly elevated [53].
Other studies have evaluated the antioxidant potential in vivo of phenolicrich beverages and foods. It has been found that the antioxidant capacity of
plasma was signicantly enhanced in humans, after the ingestion of apple juice,
whose main antioxidant components were identied as ascorbic acid and
chlorogenic acid [54]. Furthermore, it has been observed that plasma total
antioxidant capacity increased signicantly after ingestion of red wine, green
tea, and black tea in healthy volunteers [55]. Moreover, an increase in plasma
antioxidant capacity and a reduction in plasma concentration of thiobarturic acid
reactive substances (TBARS) have been observed in healthy subjects after the
consumption of a meal of procyanidin-rich chocolate [56]. Also, it has been
shown that the total antioxidant capacity of serum can be increased signicantly,
after consumption of red wine, strawberries, and spinach, in elderly women [57].
Thus, there is abundant evidence to indicate that phenolic-rich foods and
beverages can have a signicant impact on the antioxidant capacity of plasma,
ultimately suggesting that avonoids may act as antioxidants in vivo without, of
course, excluding other biological actions or the eects of other components.
IX.
CONCLUSIONS
In this chapter, the phenolic composition and content of regularly consumed

fruits and vegetables are extensively discussed through the analysis of the
current relevant literature, in order to provide a comprehensive summary of the
current compositional and quantitative data on some avonoid-rich foods.
Furthermore, the formal relation of the in vitro antioxidant potential of these
fruits and vegetables to the quality of the phenolic and, to a lesser extent,
vitamin C content is emphasized. The potential for antioxidant activity of
avonoid-rich fruits and vegetables in vivo is also discussed. The data described
here allow identication of the potentially most eective fruits and vegetables
in terms of phenolic content and antioxidant activity. However, much research
is still needed: the elucidation of the metabolism and bioavailability of
avonoids in vivo, as well as of the amounts and the forms in which they are
taken up into cells and tissues, is crucial in order to establish the mechanisms
and the forms in which dietary phenolics may act in vivo [58]. Finally, it
91
remains to be established to what extent phenolics act as antioxidants or

enzymes and gene modulators, and how their actions may be relevant to
disease prevention [5963].
REFERENCES
Middleton E Jr, Kandaswami C, Theoharides TC. The eects of plant avonoids on
mammalian cells: implications for inammation, heart disease, and cancer.
Pharmacol Rev 2000; 52:673751.
2. Galati G, Teng S, Moridani MY, Chan TS, OBrien PG. Cancer chemoprevention
and apoptosis mechanisms induced by dietary polyphenolics. Drug Metabol Drug
Interact 2000; 17:311349.
3. Joseph JA, Shukitt-Hale B, Desinova NA, Prior RL, Cao G, Martin A, Taglialatela
G, Bickford PC. Long-term dietary strawberry, spinach or vitamin E supplementation retards the onset of age-related neuronal signal-transduction and cognitive
behavioural decits. J Neurosci 1998; 18:80478055.
4. Cao Y and Cao R. Angiogenesis inhibited by drinking tea. Nature 1999; 398:381.
5. Fuhrman B, Levy A, Aviram M. Consumption of red wine with meals reduces the
susceptibility of human plasma and LDL to lipid peroxidation. Am J Clin Nutr
1996; 61:549554.
6. Gaziano JM, Hennikens CH. Update on dietary antioxidants and cancer. Pathol Biol
1996; 44:4245.
7. Hertog MG, Kromhout D, Aravinis C, Blackburn H, Buzina R, Fidanza F,
Giampaoli S, Jansen A, Menotti A, Nedeljkovic S, Pekkarinen M, Simic BS,
Toshima H, Feskens EJM, Hollman PCH, Katan MB. Flavonoid intake and longterm risk of coronary heart disease and cancer in the seven countries study. Arch
Intern Med 1995; 155:381386.
8. Gandini S, Merzenich H, Robertson C, Boyle P. Meta-analysis of studies on breast
cancer risk and diet: the role of fruit and vegetable consumption and the intake of
associated micronutrients. Eur J Cancer 2000; 36:636646.
9. Rice-Evans CA, Miller NJ, Paganga G. Structure-antioxidant activity
relationships of avonoids and phenolic acids. Free Radic Biol Med 1996;
20:933956.
10. Visioli F, Bellomo G, Galli C. Free radical-scavenging properties of olive oil
polyphenols. Biochem Biophys Res Commun 1998; 247:6064.
11. Jovanovic SV, Steenken S, Simic MC, Hara Y. Antioxidant properties of
avonoids: reduction potentials and electron transfer reactions of avonoids
radicals. In: Rice-Evans CA, Packer L, eds. Flavonoids in Health and Disease. New
York: Marcel Dekker 1998:137161.
12. Halliwell B, Gutteridge JMC, eds. Free Radicals in Biology and Medicine, 3rd ed.
New York: Oxford University Press, 1999.
13. Proteggente AR, Pannala AS, Paganga G, van Buren L, Wagner E, Wiseman S, van
de Put F, Dacombe C, Rice-Evans CA. The antioxidant activity of regularly
consumed fruit and vegetables reects their phenolic and vitamin C composition.
Free Radic Res 2002; 36:217233.
1.
92
Proteggente et al.
14.
Kahkonen MP, Hopia AI, Vuorela HJ, Rauha J-P, Pihlaja K, Kujala TS, Heinonen
M. Antioxidant activity of plants extracts containing phenolic compounds. J Agric
Food Chem 1999; 47:39543962.
Prior RL, Cap G, Martin A, Soc E, McEwen J, OBrien C, Lischner N, Ehlenfeldt
M, Kalt W, Krewer G, Mainland CM. Antioxidant capacity as inuenced by total
phenolic and anthocyanin content, maturity, and variety of Vaccinium species.
J Agric Food Chem 1998; 46:26862693.
Velogliu YS, Mazza G, Gao L, Oomah BD. Antioxidant activity and total phenolics
in selected fruits, vegetables and grain products. J Agric Food Chem 1998;
46:41134117.
Paganga G, Miller N, Rice-Evans C. The polyphenolic content of fruit and
vegetables and their antioxidant activities: what does a serving constitute? Free
Radic Res 1999; 30:153162.
Vinson JA, Hao Y, Su X, Zubik L. Phenol antioxidant quantity and quality in foods:
vegetables. J Agric Food Chem 1998; 46:36303634.
Gardner PT, White TAC, McPhail DB, Duthie GG. The relative contributions of
vitamin C, carotenoids and phenolics to the antioxidant potential of fruit juices.
Food Chem 2000; 68:471.
Wang SY, Lin H-S. Antioxidant activity in fruits and leaves of blackberry,
raspberry, and strawberry varies with cultivar and developmental stage. J Agric
Food Chem 2000; 48:140146.
Re R, Pellegrini N, Proteggente AR, Pannala AS, Rice-Evans CA. Antioxidant
activity applying an improved ABTS radical cation decolorization assay. Free Radic
Biol Med 1999; 26:12311237.
Cao G, Verdon CP, Wu AHB, Wang H, Prior RL. Automated-assay of oxygen
radical absorbance capacity with the COBAS FARA-II. Clin Chem 1995; 41:1738
1744.
Benzie IFF, Strain JJ. Ferric reducing/antioxidant power assay: direct measure of
total antioxidant activity of biological uids and modied version for simultaneous
measurement of total antioxidant power and ascorbic acid concentration. Methods
Enzymol 1999; 299:1527.
Djuric Z, Powell LC. Antioxidant capacity of lycopene-containing foods. Int J Food
Sci Nutr 2001; 52:143149.
Deighton N, Brennan R, Finn C, Davies HV. Antioxidant properties of domesticated
and wild Rubus species. J Sci Food Agric 2000; 80:13071313.
Kalt W, Forney CF, Martin A, Prior RL. Antioxidant capacity, vitamin C, phenolics
and anthocyanins after fresh storage of small fruits. J Agric Food Chem 1999;
47:46384644.
Guo C, Cao G, Soc E, Prior RL. High-performance liquid chromatography coupled
with coulometric array detection of electroactive components in fruits and
vegetables: relationship to Oxygen Radical Absorbance Capacity. J Agric Food
Chem 1997; 45:17871796.
Wang H, Cao G, Prior RL. Total antioxidant capacity of fruits. J Agric Food Chem
1996; 44:701705.
Heinonen IM, Meyer AS, Frankel EN. Antioxidant activity of berry phenolics on
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
93
human low-density lipoprotein and liposome oxidation. J Agric Food Chem 1998;
46:41074112.
Sanoner P, Guyot S, Marnet N, Molle D, Drilleau J-F. Polyphenol proles of French
cider apple varieties (Malus domestica sp.). J Agric Food Chem 1999; 47:4847
4853.
Chang S, Tan C, Frankel EN, Barrett DM. Low-density lipoprotein antioxidant
activity of phenolic compounds and polyphenol oxidase activity in selected
clingstone peach cultivars. J Agric Food Chem 2000; 48:147151.
Macheix J-J, Fleuriet A, Billot J, eds. Fruit Phenolics. Boca Raton, FL: CRC Press,
1990.
Cliord MN. Anthocyaninsnature, occurrence and dietary burden. J Sci Food
Agric 2000; 80:10631072.
Bronner WE, Beecher GR. Extraction and measurement of prominent avonoids in
orange and grapefruit concentrates. J Chromatogr A 1995; 705:247256.
Tomas-Barberan FA, Cliord MN. Flavanones, chalcones and dihydrochalconesnature, occurrence and dietary burden. J Sci Food Agric 2000; 80:1073
1080.
Hollman PCH, Arts ICW. Flavonols, avones and avanolsnature, occurrence
and dietary burden. J Sci Food Agric 2000; 80:10811093.
Moon J-H, Nakata R, Oshima S, Inakuma T, Terao J. Accumulation of quercetin
conjugates in blood plasma after short-term ingestion of onion by women. Am J
Physiol Regul Integr Comp Physiol 2000; 279:R461R467.
Fattorusso E, Lanzotti V, Taglialatela-Scafati O, Cicala C. The avonoids of leek,
Allium porrum. Phytochemistry 2001; 57:565569.
Nielsen JK, Olsen CE, Petersen MK. Acylated avonol glycosides from cabbage
leaves. Phytochemistry 1993; 34:539544.
Nielsen JK, Nrbk R, Olsen CE. Kaempferol tetraglucosides from cabbage leaves.
Phytochemistry 1998; 49:21712176.
Ferreres F, Castaner M, Tomas-Barberan FA. Acylated avonol glycosides from
spinach leaves (Spinacia Oleracea). Phytochemistry 1997; 45:17011705.
Oleszek W, Amiot MJ, Aubert SY. Identication of some phenolic in pear fruit.
J Agric Food Chem 1994; 42:12611265.
Stewart AJ, Bozonnet S, Mullen W, Jenkins GI, Lean MEJ, Crozier A. occurrence
of avonols in tomatoes and tomato-based products. J Agric Food Chem 2000;
48:26632669.
Justesen U, Knuthsen P, Leth T. Quantitative analysis of avonols, avones and
avanones in fruits, vegetables and beverages by high-performance liquid
chromatography with photo-diode array and mass spectrometric detection.
J Chromatogr A 1998; 799:101110.
Holland B, Welch AA, Unwin ID, Buss DH, Paul AA, Southgate DAT. McCance
and Widdowsons the composition of Foods, 5th ed. London: Royal Society of
Chemistry, 1991.
Cao G, Soc E, Prior RL. Antioxidant and prooxidant behavior of avonoids:
structure-activity relationships. Free Radic Biol Med 1997; 22:749760.
Ghiselli A, Nardini M, Baldi A, Scaccini C. Antioxidant activity of dierent
94
Proteggente et al.
phenolic fractions separated from an Italian red wine. J Agric Food Chem 1998;
46:361367.
48. Wearne S. Estimating dietary intakes of avonoids. In: Rice-Evans C, ed. Wake Up
to Flavonoids. London: Royal Society of Medicine Press, 2000:3543.
49. Hertog MG, Hollman PC, Katan MB, Kromhout D. Intake of potentially
anticarcinogenic avonoids and their determinants in adults in The Netherlands.
Nutr Cancer 1993; 20:2129.
50. Scalbert A, Williamson G. Dietary intake and bioavailability of polyphenols. J Nutr
2000; 130:2073S2085S.
51. Pedersen CB, Kyle J, Jenkinson AM, Gardner PT, McPhail DB, Duthie GG. Eects
of blueberry and cranberry juice consumption on the plasma antioxidant capacity of
healthy female volunteers. Eur J Clin Nutr 2000; 54:405408.
52. Manuel y Keenoy B, Vertommen J, De Leeuw I. The eect of avonoid treatment
status in Type 1 diabetic patients. Diabetes Nutr Metab 1999; 12:256263.
53. Boyle SP, Dobson VL, Duthie SJ, Hinselwood DC, Kyle JA, Collins AR.
Bioavailability and eciency of rutin as an antioxidant: a human supplementation
study. Eur J Clin Nutr 2000; 54:774782.
54. Bitsch R, Netzel M, Carle E, Strass G, Kesenheimer B, Herbst M, Bitsch I.
Bioavailability of antioxidative compounds from Brettacher apple juice in humans.
Inn Food Sci Emerg Tech 2001; 1:245249.
55. Serani M, Laranjinha JAN, Almeida LM, Maiani G. Inhibition of human LDL
lipid peroxidation by phenol-rich beverages and their impact on plasma total
antioxidant capacity in humans. J Nutr Biochem 2000; 11:585590.
56. Rein D, Lotito S, Holt RR, Keen CL, Schmitz HH, Fraga CG. Epicatechin in human
plasma: in vivo determination and eect of chocolate consumption on plasma
oxidation status. J Nutr 2000; 130:2109S2114S.
57. Cao G, Russell RM, Lischner N, Prior RL. Serum antioxidant capacity is increased
by consumption of strawberries, spinach, red wine or vitamin C in elderly women.
J Nutr 1998; 128:23832390.
58. Rice-Evans C, Spencer JPE, Schroeter H, Rechner AR. Bioavailability of avonoids
and potential bioactive forms in vivo. Drug Metabol Drug Interact 2000; 17:291310.
59. Gerritsen ME, Carley WW, Ranges GE, Shen CP, Phan SA, Ligon GF, Perry CA.
Flavonoids inhibit cytokine-induced endothelial cell adhesion protein gene
expression. Am J Pathol 1995; 147:278292.
60. Panes J, Gerritsen ME, Anderson DC, Miyasaka M, Granger DM. Apigenin inhibits
tumor necrosis factor-induced intercellular adhesion molecule-1 upregulation in
vivo. Microcirculation 1996; 3:279286.
61. Soriani M, Rice-Evans CA, Tyrrell RM. Modulation of the UVA activation of haem
oxygenase, collagenase and cyclooxygenase gene expression by epigallocatechin in
human skin cells. FEBS Lett 1998; 439:253257.
62. Schroeter H, Williams RJ, Matin R, Iversen L, Rice-Evans CA. Phenolic
antioxidants attenuate neuronal cell death following uptake of oxidized low density
lipoprotein. Free Radic Biol Med 2000; 29:12221233.
63. Spencer JPE, Schroeter H, Kuhnle G, Srai SKS, Tyrrell RM, Hahn U, Rice-Evans C.
Epicatechin and its in vivo metabolite, 3V-O-methyl epicatechin, protect human
64.
65.
66.
95
broblasts from oxidative-stress-induced cell death involving caspace-3 activation.

Biochem J 2001; 354:493500.
Hertog MGL, Hollman PCH, Katan MB. Content of potentially anticarcinogenic
avonoids of 28 vegetables and 9 fruits commonly consumed in The Netherlands. J
Agric Food Chem 1992; 40:23792383.
Schieber A, Keller P, Carle R. Determination of phenolic acids and avonoids of
apple and pear by high-performance liquid chromatography. J Chromatogr A 2001;
910:265273.
Cliord MN. Chlorogenic acids and other cinnamatesnature, occurrence and
4
Applications of Flavonoid Analysis
and Identification Techniques:
Isoflavones (Phytoestrogens)
and 3-Deoxyanthocyanins
Ewald E. Swinny
University of Adelaide
Adelaide, South Australia
Kenneth R. Markham
New Zealand Institute for Industrial Research and Development
Lower Hutt, New Zealand
I.
INTRODUCTION
The avonoids are among the most numerous and widespread natural products
found in plants and have many diverse applications and properties. There have
been several texts published on avonoids, some comprising collections of
specialist chapters [15], and a number of more practically oriented texts have
appeared independently [611]. A previous edition of this book contains a
detailed treatise on the analysis and identication of avonoids in practice [11].
Rather than simply describe further techniques of avonoid analysis, we have
chosen in this chapter to present two signicant examples of the application of
techniques previously described. These applications demonstrate how the
various available techniques are used in concert to produce pure avonoids
and to establish their structures. Selected for this purpose are two classes of
avonoids that we see as having contemporary emerging importance, the
isoavones and the 3-deoxyanthocyanins.
97
98
Swinny and Markham
The isoavones have recently enjoyed widespread attention as potential

therapeutic agents, particularly in the area of womens health. They have acquired
the phytoestrogen label because they have estrogenic activity and have been
associated with prevention of breast and prostate cancer in addition to cardiovascular disease. There are several reports in the literature that describe the
analysis of phytoestrogens that occur in a variety of dietary legumes such as red
clover, soybeans, and kudzu root [1225]. There are a large number of dietary
phytoestrogen supplements available on the phytopharmaceutical market, and
assessments of some of these for their estrogenic activity and phytoestrogen
content are also well described in the literature [12,16].
The 3-deoxyanthocyanins are a rare group of anthocyanins that are dierent
from the normal anthocyanins in that they lack oxygenation at carbon-3 of the
avonoid heterocyclic ring. Generally anthocyanins are found in many owers,
fruits, vegetables, and red wines, etc., and they are responsible for the array of
impressive colors associated with many of these. Anthocyanin-rich foods have
attracted much attention in the health sector with regard to their reputed benecial
pharmacological and biological eects. The 3-deoxyanthocyanins are a unique
subgroup of these colored pigments. Their chemical and biological properties
have not been widely explored and the limited research on them may be
attributed to their very restricted occurrence in nature.
The analysis and identication of avonoids from natural sources, as
exemplied by isoavones and 3-deoxyanthocyanins, usually follows a series
of sequential steps. Each step constitutes the subject of a section of this chapter:
Section II. Structure and distribution
Section III. Preparation of plant material for extraction
Section IV. Extraction procedures
Section V. Chromatographic analysis: high-performance liquid chromatography (HPLC)
Section VI. Flavonoid quantication: ultraviolet/visible- (UV/VIS) spectroscopy and HPLC
Section VII. Isolation and purication procedures, e.g., 3-deoxyanthocyanins
Section VIII. Structure analysis, e.g., 3-deoxyanthocyanins
II.
STRUCTURE
The distribution of the isoavones in the plant kingdom is largely restricted to the
family Leguminoseae. There are over 350 known isoavones, making them the
largest group of compounds in the class of isoavonoids. The four most common
isoavones associated with phytoestrogenic herbs or extracts are daidzein,
Isoflavones and 3-Deoxyanthocyanins
99
Figure 1 Examples of isoflavones.
genistein, formononetin, and biochanin A (see Fig. 1). These are commonly
found to occur as the glucosides, glucoside malonate esters, or free aglycones.
The pool of known 3-deoxyanthocyanins is very small. The two better
known 3-deoxyanthocyanidins (aglycones) are apigeninidin and luteolinidin.
Their 5-O-glycoside derivatives occur in the brightly colored petals of Sinningia
cardinalis [26,27]. An acetylated diglucoside was found in the red fronds of the
New Zealand fern Blechnum novae-zealandiae [28], and a series of 5-di-, 5,7-di-,
7-di-, and 7-O-glycosides were reported to occur in the fern Blechnum procerum
[29]. More highly oxygenated pigments include tricetinidin, found in black tea
Camellia sinensis [30], and the methoxylated carajurin found in the antiinammatory leaves of Arrabidea chica [31]. Another one of these highly
oxygenated pigments, columnin, was reported to occur in the gesneriad Columnea banksii [26]. However, the structure of this compound has never been
proved. The structures of some of the aglycones are shown in Fig. 2.
III.
PREPARATION OF PLANT MATERIAL

FOR EXTRACTION
Some avonoids are unstable and are degraded by enzyme action in undried
freshly harvested plant material. The safest method for drying fresh material,
100
Figure 2
Swinny and Markham
Examples of 3-deoxyanthocyanidins.
particularly ower petals and fern fronds containing 3-deoxyanthocyanins, is

freeze-drying. The powder so obtained can be stored for future use in sealed
containers in a freezer. If avonoid quantication is the objective of the
subsequent analysis, snap freezing in liquid nitrogen immediately after harvest
is advisable. Drying of well spread out plant material in an oven at 100 jC is
also an acceptable method for larger amounts, such as isoavone-containing
material in red clover. Subsequent storage of the dried material in a sealed
container under refrigeration prevents further signicant avonoid losses. Airdrying plant material at room temperature is not recommended, as this
process can give rise to enzymic degradation, e.g., the conversion of glycosides to aglycones. A satisfactory alternative, especially with anthocyanincontaining material, is to extract the freshly harvested plant material by
chopping up the sample in a blender with the appropriate solvent. Enzyme
action is not a problem here if alcohol is included in the extracting medium
to denature plant enzymes.
IV.
101
EXTRACTION PROCEDURES
In avonoid analysis the extraction procedure selected is usually determined by

the types of avonoids to be extracted and the purpose of the extraction, whether
for qualitative or quantitative use. The literature abounds with dierent methods
that have been used for the extraction of isoavones. These include the use of
aqueous methanolic, ethanolic, and acetonitrile mixtures [1325]. Some of these
also contain a portion of hydrochloric acid and involve a reuxing process that
results in subsequent hydrolysis and conversion of the glycosides to the aglycones. In general, the isoavone aglycones together with their glycosides and
malonate esters present in herbs such as red clover are conveniently extracted by
soaking the nely ground and dried plant material in MeOH:H2O or EtOH:H2O
(80:20 to 90:10) over a period of about 24 h at room temperature. The extraction
may be enhanced by sonication in an ultrasonic bath. Quantitative yields of the
constituent avonoids are best obtained when two to three sequential extractions
of the weighed and dried plant material are pooled. The less polar aglycones may
be obtained qualitatively from the leaf surface simply by rinsing the intact plant
material in an organic solvent such as diethyl ether or ethyl acetate.
3-Deoxyanthocyanins are readily extracted with acidied aqueous or
alcoholic solvents at room temperature from ground or mashed material such
as petals, leaves, or fronds. Formic acid, acetic acid, and triuoroacetic acid
(TFA) are popular acid components as these are least likely to cause hydrolysis or
deacylation on work-up. The 3-deoxyanthocyanins apigeninidin-5-O-glucoside
and luteolinidin-5-O-glucoside are conveniently extracted into 0.5% TFA [27]
from freeze-dried Sinningia cardinalis petals.
V.
CHROMATOGRAPHIC ANALYSIS
Quantitative or qualitative analysis of avonoid extracts generally involves

some form of chromatography. In the case of isoavones in red clover and other
legumes, high-performance liquid chromatography (HPLC) is the popular
method of choice. HPLC is also very useful for the analysis of 3-deoxyanthocyanins and for monitoring of the sequential progress involved in preparative
isolation of these pigments. Quantitative analysis of the 3-deoxyanthocyanins is
best carried out by using absorption spectroscopy (discussed later).
A.
High-Performance Liquid Chromatography
The identication and quantication of health-promoting avonoids in herbal

extracts and plant-derived food supplements have gained increasing importance
in the phytopharmaceutical and health-food industries. HPLC technology has
102
Swinny and Markham
developed rapidly over the years to the point where it is now undoubtedly the
analytical method of choice for these avonoids. It has the advantage that it is fast
and reproducible, requires little sample, and can be used for both qualitative and
quantitative analysis as well as for preparative work.
1. High-Performance Liquid Chromatography Instrumentation
The fundamental components of any modern-day HPLC system are a solvent
delivery system, a sample injector, a column, a detector, and a computer with the
appropriate data acquisition and processing software. There are numerous HPLC
methods described in the literature for isoavones [1325] and for the common
anthocyanins, each method invokes dierent combinations of solvent systems,
columns, and detectors. HPLC has been interfaced with a variety of detection
methods such as ultraviolet/visible (UV/vis) spectrocopy and liquid chromatographymass spectrometry (LC-MS) [21,22]. In this chapter, however, discussion
is restricted to the most commonly used pairing in avonoid analysis, that of a
reverse-phase (RP-18) column and a UV/visible detector.
Standard solvent delivery systems now oer a choice of high-pressure or
low-pressure pumping units capable of delivering at least a binary solvent
gradient. The predominant column used is the reversed-phase octadecyl (RP-18)
column. The trend toward shorter analysis times and more economical solvent
consumption has led to an array of alternative column dimensions such as
narrow-bore, or shorter and thicker columns, or even variable particle size and
type. Generally a particle size of 5 Am or smaller and column dimension of 125
or 750 mm suce for most avonoid analyses. Multichannel, fast scanning
photodiode array (PDA) detectors coupled to computer-based data acquisition
and processing software have now become the norm. PDA detection is
extremely useful for avonoid analysis because it is able to provide a UV/
visible absorption spectrum for each peak. For the detection of anthocyanins, it
is essential that the detection extend into the visible region (to at least 600 nm).
Spectral libraries are a useful software feature that allow automatic tentative
identication of components.
2. Solvent System
The solvent mixture eluting through the column can remain the same throughout
the chromatographic run (an isocratic system), or the mixture may change in a
predetermined way (a gradient system). Flavonoid analysis is best achieved by
using a gradient. Typical binary gradient solvent programs used with RP-18
columns start with a high proportion of a polar solvent and gradually increase
the proportion of a less polar solvent [32,33]. In this way the analytes are
initially concentrated at the top of the column, progressively desorbed, and
eventually refocused into narrower bands that are detected as sharp peaks on
elution. The polar solvent is normally water-based and the less polar solvent
103
methanol- or acetonitrile-based. The aqueous solvent is commonly acidied to

prevent ionization of phenolic compounds, which can yield broadened peaks for
some compounds. This sort of solvent system is appropriate for all avonoids,
including anthocyanins, which require a strongly acidic environment to ensure
complete conversion to the stable avylium ion [34]. In most cases poor peak
separation and resolution result when an ordinary linear increase in the
proportion of the nonpolar solvent is used. Better chromatography can usually
be achieved when a more complex gradient system involving linear, stepped, or
curved increases is used. Such complex gradients vary from application to
application and are best determined by trial and error.
Solvent degassing or sparging is an essential step in solvent preparation.
There are two main reasons for degassing the mobile phase. Dissolved gas levels
are lowered by degassing, thereby reducing the risk of air bubbles entering and
remaining in the detector cell. Air bubbles trapped in the detector cell produce
an undesirable sawtooth detector output. Another reason for degassing is that
oxygen absorbs strongly at low UV wavelengths, resulting in unsuitable detector
response during the analysis. Newer HPLC systems have built in degassing
units. Helium sparging is a popular and ecient method of degassing. Alternatively solvents may be degassed by prior ultrasonic agitation, although this
method is not as eective as helium sparging.
Various solvent systems have been described in the literature for the
analysis of phytoestrogens [1325]. In general, good chromatography of clover
isoavones may be obtained by using an acidic water and acetonitrile gradient.
In our experience a combination of aqueous formic or acetic acid with
acetonitrile produces very satisfactory results. A typical chromatogram obtained
in our laboratory showing the common red clover isoavone glycosides and
aglycones detected at 260 nm is depicted in Fig. 3.
The 3-deoxyanthocyanins are conveniently analyzed by using a low-pH
gradient that is also very suitable for most avonoid types, including normal 3oxygenated anthocyanins (see Fig. 4) [27]. A gentle linear gradient gives good
separation over the retention period in which most avonol glycosides are
encountered. The 3-deoxyanthocyanins are detected in the wavelength range
470490 nm. Detection in the 320 to 350-nm range is suitable for most
common avones and avonols, whereas normal anthocyanins are detected at
510530 nm.
3. Peak Identification
A good starting point when identifying the avonoid peaks in a chromatogram is
to examine the associated online UV/visible absorption spectrum captured by the
PDA detector in the range from ca. 210 to 600 nm. Most avonoids exhibit
absorption peaks in two regions, one in the low-wavelength region, 210290 nm
(band II), and one in the longer-wavelength region, 320380 nm, or 470540 nm
104
Swinny and Markham
Figure 3 High-performance liquid chromatography (HPLC) chromatogram of red

clover extract recorded at 260 nm. Column: Merck Supersphere Lichrocart 125-4
RP-19 endcapped (4 Am, 4 mm 119 mm). Elution (0.8 mL /min) performed by
using a solvent system comprising solvent A (5% HCOOH) and solvent B
(CH3CN), mixed by using a linear gradient starting with 90% A, decreasing to 60%
A at 30 min, 20% A at 33 min, and 0% A at 39 min. Peak identities: 1, ononin; 2,
daidzein; 3, genistein; 4, formononetin; 5, biochanin A.
for anthocyanins (band I). The exact position of Emax of band I can give a good
indication of the type of avonoid. Thus, for example, the band I Emax for
isoavones appears as a low-intensity shoulder in the range 310330 nm and band
II is usually prominent in the region 250265 nm. The band I Emax for the 3deoxyanthocyanins, however, appears in the range 470490 nm and band II in the
range 270280 nm. Acylation with hydroxycinnamic acids can be recognized by a
third absorption band at about 310330 nm. Variation within the ranges mentioned is due primarily to the eect of varying oxygenation. Additional oxygenation in either ring shifts the band I (or II) absorption to longer wavelength. The
absorption spectra of a wide selection of avonoids are now available in the
literature [6,810,35]. Online spectra obtained from HPLC runs, however, may
vary somewhat from those in the references, as they are measured in the HPLC
solvent rather than the usual methanol or ethanol. The other point of caution is that
sometimes a single peak on the chromatogram may actually represent more than
one compound, thereby resulting in a distorted spectrum. The purity of a peak can
be checked by measuring the spectrum at dierent points across the peak.
The retention times on RP-18 are dependent on the relative anity of the
compound for the stationary and mobile phases [32,36]. Those parts of the
molecule that are capable of forming hydrogen bonds such as the C-4 carbonyl
105
Figure 4 High-performance liquid chromatography (HPLC) chromatogram of a

Sinningia cardinalis flower petal extract recorded at 470 nm. Column as for Fig. 3.
Elution (0.8 mL /min) performed by using a solvent system comprising solvent A
[1.5% H3PO4] and solvent B [HOAc-CH3CN-H3PO4-H2O (20:24:1.5:54.5)], mixed
by using a linear gradient starting with 80% A, decreasing to 33% A at 30 min, 10%
A at 33 min, and 0% A at 39.3 min. Peak identities: 1, luteolinidin-5-O-glucoside; 2,
apigeninidin-5-O-glucoside.
group and free hydroxyl groups have anity for the more polar mobile phase.
Flavonoids that have more of these groups elute earlier than those with fewer.
Alternatively, an increase in the number of hydrophobic groups (e.g., methylation or acylation) increases anity for the stationary phase and so increases
retention time. However, although genistein and biochanin A both have one
more hydroxyl than daidzein and formononetin, respectively, this hydroxyl is
strongly hydrogen-bonded to the C-4 carbonyl, resulting in an overall decrease
in polarity and hence longer relative retention times. The glucoside of each of
these isoavones has a shorter retention time than the corresponding aglycone as
a result of the increase in hydrophilic hydroxyls. Acylation, as seen with the
malonates found in red clover, increases the retention time by eectively
deleting a hydrophilic hydroxyl group. In the 3-deoxyanthocyanin group,
106
Swinny and Markham
luteolinidin 5-O-glucoside has one more free hydroxyl than apigeninidin 5-Oglucoside, causing it to be more polar and hence elute earlier.
Although it is possible to arrive at some reasonable conclusion regarding
the identity of a peak on the basis of a comparison of the retention time and
online spectrum with published data, it is always best to conrm the identity
by cochromatography against an authentic standard. A simple way to do this
is to inject a known concentration and volume of the standard and, separately,
of the plant extract. An amount of each of these is then combined and
injected, and a comparison is made between the detector response (peak
height or peak area) obtained and the response of the plant extract and the
standard solution. A corresponding increase in the resultant response usually
conrms the identity.
4. Sample Preparation
The key priorities associated with the preparation of a sample for HPLC analysis
are cleanup, matrix removal or modication, and optimal analyte concentration.
Cleanup of the sample is essential since impurities in the extract or sample
matrix may interfere with analyte detection and measurement. It is also vital that
the sample to be injected is free of solid matter to prevent clogging of the
injector or the guard column and the eventual deterioration of the analytical
column. Filtration or centrifugation of the extract is usually an eective rst step
in the cleanup process.
Sometimes the analyte concentration is too dilute for direct injection
and measurement and it becomes necessary to increase the concentration per
unit volume. Solid phase extraction (SPE) is a very eective and elegant
sample preparation procedure for most phenolics that can be used simultaneously to achieve cleanup, matrix removal or modication, and optimal
analyte concentration.
Typically the process involves retention and concentration of the analytes
on a RP-18 solid phase cartridge, while the impurities are washed o. To speed
up this process the cartridge or column may be attached to a vacuum manifold
processor that is connected to a vacuum source. The vacuum manifold processor
also allows a large number of samples to be processed simultaneously. The
analytes are selectively eluted as a concentrated solution by using a minimal
volume of the appropriate solvent.
A preparation of a red clover sample for HPLC injection would typically
involve the following steps. The ethanolic extract (typically 1 mL) is diluted
with 34 mL water and introduced to a water preconditioned disposable RP-18
500-mg cartridge. Unbound sugars and polar impurities are washed out with
water (35 mL), followed by 35 mL of 10% MeOH. The retained isoavones
are eluted with about 35 mL 8090% MeOH. This extract is dried down and
107
reconstituted in the appropriate volume of mobile phase or other solvent and

injected directly.
This procedure works equally well for the 3-deoxyanthocyanins except that
it is desirable to maintain an acidic medium (the water component in the eluant is
replaced by 0.1% HCOOH or TFA) throughout the process. SPE is particularly
useful when extracts or selected analytes are subjected to chemical reactions such
as acid or base hydrolysis. By-products formed and excess reagents such as HCl
and NaOH used in hydrolysis of sugars are conveniently removed.
VI.
FLAVONOID QUANTIFICATION
UV/visible absorption spectroscopy and HPLC with UV/visible detection are

both very useful for the determination of the quantities of individual or total
avonoids in plant extracts and herbal preparations. Quantication by UV/
visible spectroscopy is especially suitable for 3-deoxyanthocyanins and other
anthocyanins because they have a visible maximum that is usually free of
interference from other phenolic compounds.
A.
Quantification Using Ultraviolet /Visible Spectroscopy
Quantication of avonoids using UV/vis spectroscopy involves application of

the Beer-Lambert law:
A ecl
In this relationship A = absorbance (read from the spectrophotometer), e =
extinction coecient, c = concentration in moles per liter, and l is the path length
(centimeters) of the cell used. The e value is taken from the literature or calculated
by using a standard solution. Thus the concentrations of standard solutions of
the 3-deoxyanthocyanidins apigeninidin and luteolinidin may be determined
by using the published extinction coecients of 18000 M1 cm1 and 13,800
M1 cm1, respectively [37]. A solution of an unknown or unknowns can also be
quantied by using these e values. In this case, however, the level of the
unknown(s) is expressed in terms of equivalents of the standard relating to the e
value used. In order to maintain the linear relationship as required for the BeerLambert law to hold, solutions should be prepared such that the absorption (A) is
between 0.05 and 1.00 absorption units. Anthocyanin concentrations must always
be measured in acidic solution, usually 0.1 N HCl (aqueous or alcoholic), to
ensure that the anthocyanin is totally in the avylium form [6,38].
The concentration of isoavonoid standard solutions used for phytoestrogen quantication can be accurately and conveniently determined by UV/vis
spectroscopy using known e values [13,14,23]. A known volume of such a
108
Swinny and Markham
solution is injected into an HPLC (see Sec. VI. B.), and the peak area relativities
obtained from the standard and the sample are used to calculate the concentration of analyte. e Values for avonoids are available in the literature, and a
good compilation is reported by Jurd [35].
B.
Quantification Using High-Performance

Liquid Chromatography
HPLC has become a widely used and reliable technique for the accurate
quantication of avonoids, for example, for the determination of phytoestrogens
in herbal supplements and plant extracts [1325]. Typically a good chromatogram
shows well separated peaks representing those components of the plant extract that
absorb at a selected wavelength. For example, analysis of an isoavone extract
would ideally involve detection at the band II maximum of about 260 nm.
However, not all the peaks appearing in the chromatogram would necessarily
represent isoavones. An online spectral facility is very useful at this stage. With
this facility a complete absorption spectrum for each peak may be obtained,
thereby allowing the grouping of peaks into their appropriate avonoid classes.
When using a UV/vis detector the peak height, or peak area produced by
an electronic integrator, is a response of the UV/vis detector signal, which in
turn is a function of the molar absorptivity of the avonoid. Thus both the height
and the area of a well-resolved peak in a chromatogram are proportional to the
concentration of the avonoid detected. For quantication, either peak height or
peak area can be used. Peak area quantication is very popular for most
avonoid analysis. However, it is not always the most accurate. When peaks
are well resolved and consistently symmetrical, peak height quantication can
be far more accurate. This is also the preferred option when trace amounts are to
be determined. A number of operating conditions aect the peak height and area
measurement dierently. For example, a small change in the ow rate aects the
peak area measurement more than it aects the peak height measurement. When
the column is not thermostatted, the preferred option is peak area measurement.
Whatever the choice, accurate quantitative analysis is reliant on the detectors
operating in the linear range. For a more detailed discussion on this topic and
other aspects of HPLC theory and practice the reader is referred to more
specialized HPLC texts [39,40]. Discussion here is restricted to the use of peak
area measurement in two common methods of avonoid quantication, the
external standard method and the internal standard method.
1. External Standard Method
The external standard method is the simplest quantication method and is usually
used for straightforward analysis, especially those not requiring extensive sample
109
preparation, and for analyses in which the chromatography is highly reproducible

and the instrumentation is very dependable. The compound chosen as external
standard (ES) should ideally be the same avonoid as that being quantied.
Otherwise it should be closely related to the avonoids of interest and show
similar solvent solubility and spectral and chromatographic properties. The ES
should also be readily available in pure form and economically priced.
For isoavone analysis the four common isoavones biochanin A,
formononetin, daidzein, and genistein are all very suitable as external standards.
They are found in many herbal extracts exhibiting estrogenic properties. The
levels of isoavones may thus be conveniently expressed as equivalents of one
of these standards. Other standards used in our laboratory for general avonoid
quantication include rutin, naringenin, catechin, apigenin-7-glucoside, quercetin, and kaempferol.
With the ES method a number of concentrations of the standard solution,
also referred to as a calibration solution (CS), are prepared. The exact concentration of each CS can be established by UV/vis spectroscopy by using the
extinction coecient of the compound. A known volume of each CS is injected
and the response (peak area) vs. concentration is plotted to produce a calibration
plot. This plot should be linear and have a zero intercept. The sample to be
analyzed is prepared in the same solvent as the CS and injected, chromatographed, and detected exactly as the CS. The concentration of the relevant
components in the sample may then be determined graphically from the peak
integrals by using the calibration plot, or numerically by using a response factor
RF (or calibration factor). It is essential that the concentrations of the CS cover
the concentration range expected for the unknown sample. If the unknown
sample concentration falls outside the range of the CS injected, the sample
should be diluted or concentrated, whichever is applicable.
The RF method is a convenient numerical alternative to reading a value
from the calibration plot. The response factor for the CS is determined by
RF
CSarea
CSconcentration
For greater accuracy it is advisable to determine an average RF of a CS over a

range of concentrations. If the calibration plot is linear and the intercept is zero,
the RF is equivalent to the slope of the plot. After the RF is obtained, it can then
be used to calculate the avonoid level in the unknown sample or plant extract:
Flavonoid concentration
Flavonoidarea
RF
Conversion of the result to a more appropriate unit of concentration such as

milligrams per gram (mg/g) can be made if required, by incorporating factors
110
Swinny and Markham
such as amount of dry plant material extracted, volume of extraction solvent,

sample dilution, and injection volume. A practical example is the extraction of
98 mg of dry red clover leaves with 5 mL of 80% MeOH. Injection of 10 AL of
the prepared sample produces a chromatogram shown in Figure 3. The biochanin
A area in this chromatogram is 6,886,805 units. Injection of 10 AL of biochanin
A (CS), which has a concentration of 0.04 mg/mL, produces a peak with area
4,282,266 units (peak 5). The response factor is
RF
4;282;266
0:04
The biochanin A concentration in the extract is therefore

biochanin Aconc
6;886;805

4;282;266
0:04
0:06 mg=mL
Conversion to milligrams of biochanin A per gram of plant material would be
0:06mg=mL 5mL
1000mg
98mg
3:06 mg=g
The total isoavone content may be calculated in a similar way by summing the
total area of all the isoavone peaks and expressing the value in this case as
biochanin A equivalents. Alternatively, several standards may be used to
determine the total isoavone level. In commercial supplements the total
phytoestrogen content is often expressed as milligrams per gram or micrograms
per gram (mg/g or Ag/g), and is a measure of the total free aglycone content
[13,14,16,18,23,25]. This total content is actually a sum of the glucosides and
aglycones, but the level is expressed as free aglycone units because isoavones
are absorbed by the gut as aglycones. The total aglycone content is arrived at
by correcting for the molecular weight of the glucosides, which is almost twice
that of the aglycones. Hence the total isoavone concentrations indicated on
labels of phytoestrogen supplements are never just an expression of the arithmetic sum of the individual conjugate forms.
2. Internal Standard Method
An internal standard (IS) is used primarily to monitor the reliability of extraction,
sample preparation, chromatographic, and instrumentation procedure. The use of
an IS is strongly recommended if the sample preparation method consists of
several steps. In such a case a known amount of an internal standard would be
added to the sample at an early stage on the assumption that the internal standard
111
and the avonoid analytes will behave similarly during sample work-up and
throughout the chromatographic process. Alternatively the IS could be added to
the prepared sample immediately before HPLC if the sole concern is instrument
reliability, for example, injection volume reproducibility.
The choice of IS is usually not easy. The compound chosen should have
properties similar to those of the avonoids being quantied, and consideration
should be given to factors such as solubility in the extraction solvent and the
chromatographic and detection characteristics. Ideally this compound should
elute in a gap in the pattern of peaks, somewhere in the middle of the
chromatogram or in the region of the peaks being quantied. Some chromatograms of plant extracts are crowded with peaks, making nding a gap for an IS
dicult. The chromatogram of red clover (Fig. 3) recorded at 260 nm has an
abundance of glucoside and aglycone peaks, some minor and others major. Such
chromatograms can be simplied by including an acid hydrolysis step in the
sample preparation. This produces a chromatogram showing only the isoavone
aglycones. Extreme care, however, should be exercised when employing an acid
hydrolysis step in a quantication procedure since hydrochloric acid can degrade
avonoids. Several examples of the use of acid hydrolysis for this purpose have
been reported in the literature [13,24,25]. Enzymic hydrolysis is milder than acid
hydrolysis, although many attempts have been described as not very successful.
Cellulase from Aspergillus niger is reported to yield good results [41]. Several
examples of internal standards used to quantify isoavones accurately have
appeared in the literature. These include compounds such as 4-hydroxybenzophenone [18], avone [13], equilenin [16], and 2,4,4V-trihydroxybenzoin [14].
Once a suitable compound has been chosen as an IS, the procedure would
be to add an equal amount of the IS to the sample and to the calibration standard
CS (essentially this is the external standard), which is made up to dierent
concentrations. Unlike in the external standard method, here peak area or peak
height ratios are compared. The ratio of the peak area of the CS to the IS for
each CS prepared is determined. This ratio is plotted vs. the concentration of the
CS to produce a calibration plot. The avonoid concentration in the analyte can
then be determined directly from the calibration plot. As with the ES method the
avonoid concentration can also be determined numerically by using the
response factor (providing that the plot is linear and has a zero intercept):

RF
CSarea
ISarea
CSconcentration
Therefore,
Flavonoid concentration
Flavonoidarea
ISarea
RF
112
VII.
Swinny and Markham
ISOLATION AND PURIFICATION PROCEDURES
Many plant extracts contain a complex mixture of avonoids, some of which

might contribute directly to particular biological properties of the extract, such
as antioxidant, free radical scavenging, or estrogenic activity. The isolation of
individual biologically active avonoids from a complex mixture requires a
systematic approach involving a carefully considered selection of solvents and
chromatographic stationary supports.
In this section we discuss the isolation and purication of avonoids, with a
particular emphasis on 3-deoxyanthocyanins. Very few of these pigments have
been isolated from natural sources, and presented here is a three-step procedure
that we have found to be most useful: (1) initial cleanup of extract, (2) large-scale
fractionation, (3) nal purication.
A.
Initial Cleanup
In most plant extractions a signicant amount of carbohydrate ends up in the crude

extract. A preliminary removal of these carbohydrates from the extract can be
achieved by selection from the product range of non-ionic polystyrene (Amberlite
XAD) [42] or polyacrylic resins (Diaion HP resins) [43]. The procedure with 3deoxyanthocyanins involves passing the crude aqueous acidied (0.1% TFA)
extract through a column containing one of these materials. The sugars are not
adsorbed and are eluted from the column with additional amounts of acidied
water. The retained less polar compounds, the 3-deoxyanthocyanins and other
avonoids, are then washed from the column with acidied aqueous methanol. If
desired, at this stage some separation can be achieved by using a stepped gradient
of water to methanol. The methanol is removed by rotary evaporation and the
residual aqueous fraction is freeze-dried. If the crude extract contains some
methanol, this should be removed before application to the resin. It is possible
to employ a similar solvent regimen, with other adsorbents such as derivatized
silica gel (e.g., RP-18, cyanopropyl) or polyamide [8] to achieve this initial
cleanup. With avonoids other than anthocyanins the water need not be acidied.
B.
Large-Scale Fractionation
Polyamide and cellulose are both very economical and eective media for largescale fractionation. The freeze-dried extract is dissolved in a minimal amount of
0.5% TFA and is applied to a polyamide column previously equilibrated in 0.5%
TFA; elution is continued with the same solvent. A methanol component can be
introduced to the eluant with a gradual incremental increase. The 3-deoxyanthocyanins are brightly colored in an acid medium, a feature that makes their
separation in a glass column quite spectacular. For example, in the fractionation
113
of a Sinningia cardinalis extract, the apigeninidin-5-O-glucoside and luteolinidin5-O-glucoside are obvious as a yellow and reddish purple band, respectively. The
most convenient way to collect the eluted fractions is to use an automatic fraction
collector. Similar fractions can later be combined on the basis of HPLC analysis.
At this stage some of the fractions may contain essentially pure compounds,
which then require only minor subsequent cleanup.
When using cellulose for large-scale fractionation, the freeze-dried extract
is dissolved in 5% acetic acid and then applied to a glass column containing
cellulose preconditioned with 5% acetic acid. Depending on the rate and quality of
separation, elution can be continued with 5% acetic acid or stepped up gradually to
15% acetic acid. Once the target fractions have been collected and suitably
combined, an appropriate polymeric resin (e.g., Amberlite XAD or Diaion HP)
can be used to concentrate the fractions in preparation for nal purication.
C.
Final Purification
The hydroxypropylated cross-linked dextran, Sephadex LH-20, is very eective

for the nal purication of 3-deoxyanthocyanins (or avonoids in general). LH20 separation is based on molecular size and H-bonding interactions. The fraction
to be puried is dissolved in a minimal amount of MeOH:H2O, 40:60, containing
0.1% TFA and applied carefully to a column containing LH-20 preconditioned
with the same solvent. Elution is continued with this solvent. Although the 3deoxyanthocyanins are strikingly visible to the naked eye, a glass column also
allows the progress of the separation of colorless avonoids to be monitored
conveniently with a portable UV/visible lamp in a darkroom.
Medium-pressure (up to 6 bar) preparative chromatography employing
ready-to-use glass columns (e.g., Lobar) with a variety of stationary phases is
another excellent medium available for avonoid purication. Using a peristaltic pump and a reverse-phase column, the separation emulates that of
an analytical RP-18 column. By examining a chromatogram obtained with an
analytical column it is possible to design a suitable solvent system and gradient
and thereby obtain a similar chromatographic separation for a fraction needing
purication. The other advantage of these columns is that they have a relatively
large capacity and can tolerate larger quantities (up to a gram) than preparative
HPLC, for instance. Solvent systems we have found to be most useful in our
laboratories include 520% MeOH or CH3CN acidied (0.1%) with TFA or
formic or acetic acid. The fraction to be puried is dissolved in a minimal
volume of solvent and is then introduced to the top of the preconditioned
column via an in-line peristaltic pump. To obtain good chromatography the
analytes need to be concentrated on the top of the column before changing of
the gradient and solvent composition. The pump draws eluting solvent from the
reservoir and passes it through the column; then appropriate fractions are
114
Swinny and Markham
collected. The glass casing of these columns makes monitoring the separation
process with the naked eye possible for brightly colored pigments, or with a
portable UV lamp for other avonoids.
Preparative HPLC is often very useful for a dicult purication and
situations in which only a small amount of a pure compound is required. A
larger (10-mm diameter) column is needed and a higher ow rate than that used
in analytical work is used. Since only small amounts can be collected from each
run, the pure compounds are accumulated from several injections by collection
of appropriate peaks as detected by UV/visible or other detection. This
procedure can be automated through the use of automatic injection and timeor peak-based automated fraction collection.
One-dimensional paper chromatography is an inexpensive and convenient
purication technique. Milligram quantities can be obtained by running several
one-dimensional paper chromatograms. The appropriate bands are excised and
eluted and the fractions combined [9]. Contaminant polysaccharide material
may be removed subsequently by using a small RP-18 column.
VIII.
STRUCTURE ANALYSIS
It is always a tremendously satisfying experience for the avonoid scientist to

toil through the intricacies of the fractionation of an extract and eventually
obtain a pure avonoid. Even more exhilarating are the eventual identication
and characterizing of the compound, especially if it is novel. The 3-deoxyanthocyanins provide this sort of stimulation since they are very rare and not many
structural variations are known. UV/vis spectroscopy, acid hydrolysis, nuclear
magnetic resonance spectroscopy, and mass spectroscopy are valuable degradation and instrumental techniques that make structural identication of these
pigments possible. The following section focuses on the latter three techniques.
A.
Acid Hydrolysis
Acid hydrolysis of a avonoid glycoside leads to the separation of the aglycone

and the sugar entities, thereby enabling structural investigations to be carried out
on each portion independently. Besides its use for pure avonoids, it allows
valuable information such as the aglycone ratio in a crude extract to be obtained.
3-Deoxyanthocyanins are cleaved under the conditions of acid hydrolysis,
producing the stable 3-deoxyanthocyanidin aglycone and the liberated sugar
in the process. The procedure involves reuxing the pure pigment or extract in
2N HCl : MeOH (1:1) on a boiling water bath for 3040 min. The reaction may
be monitored by HPLC, the aglycone has a longer retention time than the
glycoside. The aglycone and sugars are conveniently separated on a small
115
reverse-phase column such as that used for SPE (see Sec. V. A.4). Identication
of the sugar liberated is commonly carried out by one-dimensional paper
chromatography (1D-PC) by comparison with a mix of authentic sugars. The
ve most frequently encountered sugars associated with avonoids, glucose,
galactose, rhamnose, xylose, and arabinose, are well separated by 1D-PC by
using n-BuOH:Pyr:H2O (6:4:3), visualizing spots with an aniline phthalate spray
reagent (ca. 3% in MeOH), or by thin-layer chromatography (TLC), using a dried
silica plate impregnated with 0.3 M KH2PO4, run in a BuOH:acetone:H2O (4:5:1)
mixture [9].
B.
Nuclear Magnetic Resonance Spectroscopy
Nuclear magnetic resonance (NMR) spectroscopy is an invaluable instrumental

technique for the structural determination of all avonoids including 3-deoxyanthocyanins. As well as providing information on the chemical environment of
each proton or carbon nucleus in the molecule, the technique can be employed to
determine linkages among nearby nuclei, often enabling a complete structure to
be assembled. The reader is referred to Refs. 44 and 45 for details of the
principles of NMR and general interpretation of NMR spectra.
Figure 5 The structures of 1 and 2 showing the main HMBC correlations.
116
Swinny and Markham
Dimethyl sulfoxide (DMSO-d6) and methanol (CD3OD) are both suitable

solvents for 3-deoxyanthocyanins. Complete conversion to the avylium ion
requires a concentration of 1% TFA-d when methanol-d4 is used and sometimes
higher levels when DMSO-d6 is used. A broad water peak in the proton NMR
spectrum can obscure signals in the sugar region, so both the solvent and the
sample should be as dry as possible. At the end of the NMR analysis the sample
may be recovered by evaporation of the CD3OD in vacuo. With DMSO the
sample can be recovered by adding water and applying the mixture to a small
RP-18 column.
The 3-deoxyanthocyanins 1 and 2, shown in Fig. 5, are examples used
later to show how NMR is used to establish the structure.
The rst requirement in an NMR analysis is to obtain basic one-dimensional proton (1H) and carbon (13C) spectra, including a 13C-DEPT (disproportionless enhancement by polarization transfer) spectrum. Typical proton spectra
for the pigments 1 and 2 are shown in Figs. 6 and 7, respectively [27,28]. As a
result of the relatively low natural abundance of the 13C isotope, 13C experiments require considerably longer periods (hours) to acquire sucient data to
yield a presentable spectrum, whereas proton spectra can be obtained in a few
minutes. A reasonable proton spectrum can be obtained with as little as 0.3 mg,
whereas 13C spectra generally require larger samples, typically more than 1 mg.
Figure 6 1H nuclear magnetic resonance (NMR) spectrum of 1, apigeninidin-5O-glucoside.
117
Figure 7 1H nuclear magnetic resonance (NMR) spectrum of 2, luteolinidin-5-Oh-D-[3-O-h-D-glucopyranosyl-2-O-acetylglucopyranoside].
The aglycone structures of 1 and 2 are determined from the patterns of

signals in the aromatic region 69.5 ppm in the proton spectrum. The B-ring (H2V,6V,3V,5V) and (H-2V,5V,6V) signal patterns dene the apigenin and luteolin nature
of 1 and 2, respectively. A distinctive feature in these spectra is the presence of
two downeld doublets that represent the protons H-3 and H-4. It is this feature
that readily distinguishes 3-deoxyanthocyanins from normal C-3 oxygenated
anthocyanins. The number of anomeric protons, two (1W and 1VW) in the case of 2
(Fig. 7), indicates the number of sugars present. The number of carbon atoms
and the number of hydrogens bonded to each carbon can be obtained from the
13
C spectrum and the DEPT spectrum. These carbon spectra are particularly
useful in establishing the number of sugar carbons and eventually the type of
sugar. The DEPT experiment allows identication of the carbons bearing two
attached hydrogens. The methylene protons of the glucose unit in 1 and 2 are
readily identied in this way, since their signals become inverted.
Following on from the one-dimensional proton and carbon spectra, the next
step is to employ more sophisticated two-dimensional (2D) NMR techniques to
help determine linkages within the molecule. In most of these experiments the
instrument automatically combines the results of these experiments and the data
118
Swinny and Markham
are presented as a 2D contour plot. The more common of these techniques include
homonuclear proton correlation spectroscopy (H,H-COSY), heteronuclear carbon correlation spectroscopy (H,C-COSY), heteronuclear multiple bond correlation (HMBC), and total correlation spectroscopy (TOCSY) [46].
The H,H-COSY or double quantum ltered correlation spectroscopy
(DQF-COSY) is the simplest 2D experiment to run and is usually run rst.
This technique is useful for the determination of linkages between adjacent
hydrogens. In this experiment the 1D proton spectrum is displayed along each
axis with a contour display of the same spectrum along the diagonal axis. Odiagonal peaks are seen where the corresponding protons are coupled, usually as
a result of vicinal or geminal coupling. In both compounds 1 and 2 the H,HCOSY shows a characteristic correlation between H-3 and H-4 and the
respective correlations between the protons in the A- and B-rings. Correlation
between the relevant sugar protons is particularly helpful when allocating the
signals in the sugar region of the proton spectrum. For instance, the downeld
H-2W signal in compound 2 is recognized from its correlation with the
identiable H-1W signal.
A TOCSY experiment enables linkages to be made among all of the
protons within a chain of coupled protons. This is especially useful in a
diglycoside, because it is sometimes dicult to establish which protons belong
to which of the sugar units. For example, in compound 2 the respective group of
protons belonging to each of the sugars can be established by examining the
respective correlations from the identiable anomeric protons.
The C,H-COSY or heteronuclear single quantum coherence (HSQC) experiment shows linkages between carbon and hydrogen nuclei in the same general
format as the H,H-COSY except that the 13C spectrum is displayed on one axis and
the proton spectrum on the other. The spectrum obtained shows correlations
between protons and the specic carbons to which they are directly attached. For
example, the C,H-COSY of 1 (Fig. 8) shows a very clear correlation between the
protons H-3 and H-4 and the carbons C-3 and C-4, respectively. Similar
correlations between the other protons and their respective carbons allow one to
compile a fairly accurate structure of the compound. The point of linkage of the
dierent moieties in the molecule can be determined by a long-range C,H-COSY
experiment called an HMBC. This technique shows correlations between protons
and carbons that are two, three, or four bonds away (depending on the instrument
parameters used), as shown in the structures for 1 and 2. This is particularly useful
when determining the point of attachment of the sugar to the aglycone.
C.
Mass Spectrometry
Mass spectrometry (MS) is used mainly in avonoid analysis for the conrmation of molecular weight, and the technique is rarely used without prior recourse
119
Figure 8 C,H-COSY spectrum of 1.
to the other spectroscopic techniques described. There are various choices of

mass spectrometric ionization methods available, including chemical ionization
and electrospray mass spectrometry (ESMS). Compound 2 is a monoacetate of a
luteolinidin diglycoside and typically a high-resolution ESMS of 2 gives a
molecular ion [M]+ at m/z 637.1747, which is consistent with that calculated,
that is, 637.1763 for the formula C29H33O16+.
IX.
CONCLUSIONS
The analysis and identication of avonoids, in particular 3-deoxyanthocyanins

and isoavones, are conveniently executed by a series of sequential steps as
discussed in this chapter. However, each class of avonoids, and in fact each new
120
Swinny and Markham
compound discovered, usually requires a unique modication of the standard

protocol. Of all the various techniques available to the avonoid scientist, high
performance liquid chromatography (HPLC) and nuclear magnetic resonance
spectroscopy (NMR) have proven to be extremely useful and invaluable.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Geissman TA, ed. The Chemistry of Flavonoid Compounds. Oxford: Pergamon

Press, 1962.
Harborne JB, Mabry TJ, Mabry H, eds. The Flavonoids. London: Chapman and
Hall, 1975.
Harborne JB, Mabry TJ, eds. The FlavonoidsAdvances in Research. London:
Chapman and Hall, 1982.
Harborne JB, ed. The FlavonoidsAdvances in Research Since 1980. London:
Harborne JB, ed. The FlavonoidsAdvances in Research Since 1986. London:
Harborne JB. Comparative Biochemistry of the Flavonoids. London: Academic
Press, 1967.
Ribereau-Gayon P. Plant Phenolics. Edinburgh: Oliver and Boyd, 1972.
Mabry TJ, Markham KR, Thomas MB. The Systematic Identication of Flavonoids.
New York: Springer-Verlag, 1970.
Markham KR. Techniques of Flavonoid Identication. London: Academic Press,
1982.
Markham KR. Flavones, avonols and their glycosides. In: Dey PM, Harborne JB,
eds. Methods in Plant Biochemistry. Vol. 1. Plant Phenolics. London: Academic
Press, 1989:197235.
Markham KR, Bloor SJ. Analysis and identication of avonoids in practice. In:
Rice-Evans CA, Packer L, eds. Flavonoids in Health and Disease. New York:
Marcel Dekker, 1998:133.
Liu J, Burdette JE, Xu H, Gu C, van Breemen RB, Bhat KPL, Booth N,
Constantinou AI, Pezzuto JM, Fong HHS, Farnsworth NR, Bolton JL. Evaluation of
estrogenic activity of plant extracts for the potential treatment of menopausal
symptoms. J Agric Food Chem 2001; 49:24722479.
Franke AA, Hankin JN, Yu MC, Maskarinec G, Low S, Custer LJ. Isoavone levels
in soy foods consumed by multiethnic populations in Singapore and Hawaii. J Agric
Food Chem 1999; 47:977986.
Murphy PA, Song T, Buseman G, Barua K, Beecher GR, Trainer D, Holden J.
Isoavones in retail and institutional soy foods. J Agric Food Chem 1999; 47:2697
2704.
Mazur W, Adlercreutz H. Naturally occurring oestrogens in food. Pure Appl Chem
1998; 9:17591776.
Setchell KDR, Brown NM, Desai P, Zimmer-Nechemias L, Wolfe BE, Brashear
WT, Kirscher AS, Cassidy A, Heubi JE. Bioavailability of pure isoavones in
121
healthy humans and analysis of commercial soy isoavone supplements. J Nutr

2001; 131:1362S1375S.
17. Hutabarat LS, Greeneld H, Mulholland M. Quantitative determination of
isoavones and coumestrol in soybean by liquid chromatography. J Chromatogr
A 2000; 886:5563.
18. Thomas BF, Zeisel SH, Busby MG, Hill JM, Mitchell RA, Scheler NM, Brown SS,
Bloeden LT, Dix KJ, Jecoat AR. Quantitative analysis of the principle soy
isoavones genistein, diadzein and glycetin, and their primary conjugated metabolites
in human plasma and urine using reversed-phase high-performance liquid
chromatography with ultraviolet detection. J Chromatogr B 2001; 760:191205.
19. Kledjus B, Vitamvasova D, Kuban V. Reversed-phase high-performance liquid
chromatographic determination of isoavones in plant materials after isolation by
solid-phase extraction. J Chromatogr A 1999; 839:261263.
20. de Rijke E, Zafra-Gomez, Ariese F, Brinkman UA, Gooijer C. Determination of
isoavone glucoside malonates in Trifolium pratense L. (red clover) extracts:
quantication and stability studies. J Chromatogr A. 2001; 932:5564.
21. Lin L, He X, Lindenmaier M, Yang J, Cleary M, Qiu S, Cordell GA. LC-ESI-MS
Study of the avonoid glycoside malonates of red clover (Trifolium pratense).
J Agric Food Chem 2000; 48:354365.
22. He X, Lin L, Lian L. Analysis of avonoids from red clover by liquid
chromatography electrospray mass spectrometry. J Chromatogr A 1996; 755:127
132.
23. Murphy PA, Song T, Buseman G, Barua K. Isoavones in soy-based infant
formulas. J Agric Food Chem 1997; 45:46354638.
24. Petterson H, Kiessling KH. Liquid chromatographic determination of the plant
estrogens coumestrol and isoavones in animal feed. J Assoc O Anal Chem 1984;
3:503506.
25. Wang H, Murphy PA. Isoavone content in commercial soybean foods. J Agric
Food Chem 1994; 42:16661673.
26. Harborne JB. 3-Desoxyanthocyanins and their systematic distribution in ferns and
gesnerads. Phytochemistry 1966; 5:589600.
27. Swinny EE, Bloor SJ, Wong H. 1H and 13C NMR assignments for the 3deoxyanthocyanins luteolinidin-5-O-glucoside and apigeninidin-5-O-glucoside.
Magn Reson Chem 2000; 38:10311033.
28. Swinny EE. A novel acetylated 3-deoxyanthocyanidin laminaribioside from the fern
Blechnum novae-zealandiae Z Naturforsch 2001; 56c:177180.
29. Crowden RK, Jarman SJ. 3-Deoxyanthocyanins from Blechnum procerum.
30. Zorn B, Garcia-Pineres AJ, Castro V, Murillo R, Mora G, Merfot I. 3-Desoxyanthocyanidins from Arrabidaea chica. Phytochemistry 2001; 56:831835.
31. Ozawa T. Separation of the components in black tea infusion by chromatography on
ToyopearlR. Agric Biol Chem 1982; 46:10791081.
32. Vande Casteele K, Van Sumere C, Geiger H. Separation of avonoids by
reversed-phase high-performance liquid chromatography. J Chromatogr A 1982;
240:8194.
122
33.
Swinny and Markham
Van Sumere C, Fache P, Vande Casteele K, DeCooman L, Everaert E, De Loose R,

Hutsebaut W. Improved extraction and reversed phase high-performance liquid
chromatographic separation of avonoids and the identication of Rosa cultivars.
Phytochem Anal 1993; 4:279292.
34. Strack D, Wray V. Anthocyanins. In: Harborne JB, ed. Methods in Plant
Biochemistry. Vol. 1. Plant Phenolics. London: Academic Press, 1989:122.
35. Jurd L. Spectral properties of avonoid compounds. In: Geissman TA, ed. The
Chemistry of Flavonoid Compounds. Oxford: Pergamon Press, 1962:107155.
36. Pietrogrande MC, Kahie YD. Eect of the mobile and stationary phases on rp-hplc
retention and selectivity of avonoid compounds. J Liquid Chromatogr 1994;
17:36553670.
37. Staord HA. Regulatory mechanisms in anthocyanin biosynthesis in rst internodes
of sorghum vulgare: eect of presumed inhibitors of protein synthesis. Plant Physiol
1966; 41:953961.
38. Francis JF, Analysis of anthocyanins. In: Markakis P, ed. Anthocyanins as Food
Colors. New York: Academic Press, 1982:181207.
39. Snyder LR, Kirkland JJ, Giajch JL. Practical HPLC Method Development. 2d ed.
New York: John Wiley & Sons, 1997.
40. Kromidas S. Practical Problem Solving in HPLC. Germany: Wiley-VCH Verlag
GmbH, 2000.
41. Liggins J, Bluck LJC, Coward WA, Bingham SA. Extraction and quantication of
daidzein and genistein in food. Anal Biochem 1998; 264:17.
42. Tomas-Barberan FA, Blasquez MA, Garcia-Viquera C, Ferreres F, Tomas-Lorente
F. A comparative study of dierent Amberlite XAD resins in avonoid analysis.
Phytochem Anal 1992; 3:178181.
43. Lu TS, Saito N, Yokoi M, Shigihara A, Honda T. An acylated peonidin glycoside in
the violet-blue owers of Pharbilis nil. Phytochemistry 1991; 30:23872390.
44. Sanders JK, Hunter BK. Modern NMR Spectroscopya Guide for Chemists. 2d ed.
Oxford: Oxford University Press, 1993.
45. Silverstein RM, Bassler GC, Morrill TC. Spectrometric Identication of Organic
Compounds. 4th ed. New York: John Wiley & Sons, 1981.
46. Markham, KR, Geiger H. 1H nuclear magnetic resonance spectroscopy of
avonoids and their glycosideshexadeuterodimethylsulfoxide. In: Harborne JB,
ed. The FlavonoidsAdvances in Research Since 1986. London: Chapman and
Hall, 1994:441497.
5
Synthesis, Identification,
Quantification, and Chemical
Reactivity of Methylated
Flavan-3-ols
Cecile Cren-Olive and Christian Rolando
Lille University of Science and Technology
Villeneuve dAscq, France
I.
INTRODUCTION
Flavan-3-ols are a large class of avanods ubiquitous in plants [15] and widely
found in a number of foods [6,7]. They represent an integral part of the human
diet and are considered to be key compounds in the relationship between health
and diet. Indeed, they are known to combat aging pathologies in which oxidative
stress is involved such as cancers [810], and cardiovascular [1113] and
neurodegenerative [14] diseases.
Because of the increasing signicance of these potential benecial roles,
understanding the mechanism by which they behave as antioxidants is essential.
However, polyphenols mainly circulate in blood as metabolites. For example, the
most-studied avan-3-ol, catechin, is present almost exclusively as methylated
metabolites (3V-methylcatechin in the majority) as well as sulfate and glucuronide
conjugates in plasma [1517]. So it is not the native forms but the methylated
forms that require further study.
Unfortunately, these metabolites are not commercially available and are
dicult to extract from enzymatic synthesis in the quantities requested for
biological or chemical studies. So whatever the study envisaged on the methylated avan-3-ols (investigation of their chemical reactivity, development of an
123
124
Cren-Olive and Rolando
analytical tool to characterize each site involved in the metabolism), the rst step
is the synthesis of a whole family of methylated compounds.
II.
SYNTHESES OF METHYLATED ANALOGUES OF

FLAVAN-3-ols
Various strategies have been attempted for the chemical synthesis of methylated
avan-3-ols, which can be divided into two groups: total enantioselective
synthesis [1824] and hemisynthesis [2531]. For total enantioselective synthesis, two strategies are predominant. The key step of the rst, leading to either
the catechin or the epigallocatechin gallate skeleton, consists of a stereospecic
cyclization of the Sharpless asymmetrical dihydroxylation product [18,20,21]
once the C6-C3-C6 skeleton is obtained either by base-catalyzed condensation
of the appropriate oxygenated acetophenone and benzaldehyde [20,21] or by the
coupling of cinnamyl alcohol derivative with a 3,5-dimethoxyphenol [18]. The
second strategy requires four main steps for synthesizing the epigallocatechin
gallate skeleton: (1) coupling of appropriate oxygenated acetophenone and
benzaldehyde; (2) cyclization of the chalcone directly to 3-aven; (3) hydroboration-oxidation, followed; (4) a two-step sequence involving oxidation with
the Dess-Martin periodinane followed by selective reduction with lithium trisec-butyl-borohydride (L-Selectride) [19]. But neither the choice of the starting
synthons nor the yields of the epigallocatechin gallate are satisfactory for
accessing to catechin analogues and methylated metabolites. Since total enantioselective syntheses of methylated avan-3-ols appear to be dicult, long, and
expensive, this total synthesis strategy is applied more specically when the
initial natural compound is not available or is available only with diculty in
pure enantiomeric form, as in the case of epigallocatechin gallate.
When polyphenol precursors are available in pure enantiomeric form from
the vegetal pool, strategies based on hemisyntheses seem much more appropriate.
However, it is well known that partial methylation of catechin, for example, does
not constitute a suitable method to synthesize methylated avan-3-ols since it
produces a complex untractable mixture of products in low yield [3235]. So the
hemisynthesis of methylated derivatives of avan-3-ols is thus very quickly
directed to strategies based on selective protection-deprotection of the A- and Brings [2530]. However, the choice of the reagent is rather limited, as catechin is
known to undergo quite readily a base-catalyzed epimerization at C-2 to form
ent-epicatechin through reversible opening of the C-ring via a B-ring quinone
methide intermediate, which requires a free phenolic OH at the C4V position
(Scheme 1) [36,37].
Until recently, the use of benzyl carbonate [28] or cyclic borate [25,28] as
protecting group led to the synthesis of essentially two dimethylated catechin
Methylated Flavan-3-ols
125
Scheme 1 Catechin epimerization and rearrangement in basic medium.

(Adapted from Ref. 36.)
analogues, the 3V,4V-dimethyl- and the 5,7-dimethylcatechin. But, these two

protecting groups have important disadvantages in the case of avan-3-ols.
The rst reagent, benzylchloroformate, is not, in fact, selective since it is not
specic to the catechol moiety. So in the presence of polyphenol, its regioselectivity depends only on the dierence of microscopic pKa between the dierent
hydroxyl functions present in the molecule, which is, for catechin and epicatechin, too tiny (pK3V-OH = 9.02; pK4V-OH = 9.12; pK5-OH = 9.43; pK7-OH = 9.58 in
water) [38] to obtain regiospecicity [28]. The second proposed protecting group
proposed, cyclic borate, oers selective protection of the B-ring under mild basic
conditions but it is rather delicate to use since the protected compound is not
isolated [25,28].
Since the year 2000 a strategy for the selective protection of catechin based
on the dierentiation between catechol and other phenols has been developed to
synthesize the four monomethylated isomers of (+)catechin in position,
respectively, 3V, 4V, 5, and 7, two dimethylated derivatives: the 5,7-dimethylcatechin and the 3V, 4V-dimethylcatechin and two trimethylated isomers of
(+)catechin in position, respectively, 3V, 5, 7, and 4V, 5, 7 [29,30]. The key
step is the dierentiation of the catechol ring of catechin from the resorcinol-like
ring by using reagents (dichlorodiphenylmethane or di-tert-butyldichlorosilane)
leading to the formation of permanent or transient dioxolane cycle. These B-ringprotected compounds open access to two dierent pathways: the rst leads to
partial or total methylation of the A-ring, the second to the methylation of the
same B-ring by means of specic protection of the A-ring and deprotection of the
B-ring. So 5,7-dimethylcatechin and 5- and 7-methylcatechin were synthesized
by dichlorodiphenyl methane protection of the catechol moiety, partial or total
126
Scheme 2 Synthesis of catechin A-ring methylated analogues: 5,7-dimethylcatechin, 5- and 7-methylcatechin. (Adapted from Ref. 30.)
methylation of the A-ring under standard conditions, followed by removal of the

protection by hydrogenolysis (Scheme 2) [29,30].
Unlike A-ring methylated derivatives, the monomethylated and dimethylated B-ring compounds were synthesized in two dierent ways. Whereas the two
hydroxysilyl monoethers (Scheme 3) obtained after reaction of catechin with ditert-butyldichlorosilane led to the two B-ring monomethylated isomers, the 3V,4Vdiphenylmethylenedioxycatechin (Scheme 4) initiated the synthesis of 3V,4Vdimethylcatechin [30]. The crucial step of these syntheses consists of the selective
and successive protection of the B- and A-rings. Indeed, after selectively protecting
the B-ring catechol moiety, protection of the A-ring, whose deprotection conditions dier from those of the B-ring, is required to allow selective deprotection of
the B-ring; therefore, selective methylation of the B-ring is possible.
More precisely, the protection of catechin by di-tert-butyldichlorosilane is
not stable and leads, after a rapid hydrolysis, to two hydroxysilyl monoesters of
the parent catechol [30], providing material for the synthesis of the two B-ring
monomethylated isomers on the B-ring (Scheme 3). The protection of the free
phenolic functions of these two ethers is achieved by benzylation under standard
conditions. After deprotection of the B-ring, i.e., desilylation induced by a
uoride ion source, the methylation of the position 3V or 4V is followed by the
removal of the benzylic group by hydrogenolysis, giving the target compounds
3V- and 4V-methylcatechin [30].
The same synthetic approach ensures the synthesis of the two trimethylated
isomers of (+)-catechin in position, respectively, 3V, 5, 7 and 4V, 5, 7 (Scheme 5).
In the case of 3V,4V-dimethylcatechin (Scheme 4), the most appropriate
protection of the A-ring is acetylation, but the use of these protecting groups
127
Scheme 3 Synthesis of catechin B-ring monomethylated analogues: 3V- and 4Vmethylcatechin. (Adapted from Ref. 30.)
requires the development of a new deprotection method. Indeed, the usual

deprotection of phenol acetate in slightly basic medium led to the formation of
a complex mixture because of the high instability of catechin. The only way to
achieve deprotection of 3,5,7-triacetyl-3V,4V-dimethylcatechin consists of the
action of a reagent that is both nucleophilic and reductive such as sodium sulte.
Unfortunately in these conditions the secondary alcohol is not deprotected;
Scheme 4 Synthesis of catechin B-ring dimethylated analogues: 3V,4V-dimethylcatechin. (Adapted from Ref. 30.)
128
Scheme 5
30.)
Synthesis of two catechin trimethylated isomers. (Adapted from Ref.
therefore, a new cycle of protection/deprotection to obtain the 3V,4V-dimethylcatechin is required (Scheme 4) [30].
Synthetic routes that are now available provide access to all monomethylated avan-3-ols, including the metabolites that are not available through natural
sources, allowing the development of new analytical methodologies to characterize avan-3-ols metabolites. Furthermore, the same strategy virtually gives access
to any polymethylated avan-3-ol analogues, especially the trimethylated series
with only one free phenol, which are key model compounds for establishing reliable relation-structure activities in biological tests and for determining thermodynamic constants.
III.
FREE RADICAL CHEMISTRY AND PHYSICOCHEMISTRY

CHARACTERISTICS OF METHYLATED FLAVAN-3-ols
The study of the free radical chemistry and physicochemistry characteristics

of methylated avan-3-ols is crucial to understand the mechanism by which
avan-3-ols behave as antioxidants. Indeed, on the one hand, physicochemical
parameters such as redox potential, scavenging and decay constants, and pKa are
of capital importance to the understanding of avan-3-ols biological eects. On
the other hand, it is the metabolites and not the native forms of avan-3-ols that
deserve further investigation since avan-3-ols are present almost exclusively as
methylated metabolites (3V-methylcatechin in the majority), as well as sulfate
and glucuronide conjugates in plasma [1517].
However, until recently methylated avan-3-ols were not readily available; so only a few studies [29,39] have examined their physicochemical
129
parameters in contrast to the abundant literature dedicated to the chemical

characteristics of avan-3-ols radicals [4049]. But there is much discussion
and contradiction in this literature regarding the phenoxyl radicals structure
[4143,4648], the reduction potentials [41,42,45,46,49], and therefore the
structure-activity relationship to the antioxidant activity. Thus it is interesting
to see how the study of the chemical reactivity of methylated avan-3-ols
enables us to understand the chemical reactivity of the native forms [29,39].
A.
Free Radical Chemistry of Methylated Flavan-3-ols
The free radical reactivity of methylated avan-3-ols has been investigated using a
ash photolysis experiment for the photochemical generation of radicals and their
characterization through the monitoring of their UV-visible spectra [29,31,39].
Phenoxyl radicals have been generated by dierent techniques: (1) by direct
photoionization of the polyphenol derivatives in their basic form and (2) by Hatom abstraction from phenolic OH by tert-butoxyl radicals generated by the
photoionization of tert-butyl peroxide in aprotic media (Fig. 1).
The study of the dimethylated compounds allows characterization of the
intrinsic reactivity of each ring of avan-3-ol. Whereas 3V,4V-dimethylcatechin
presents an absorbency at 495 nm, 5,7-dimethylcatechin shows an absorption
band at 380 nm (Table 1) [29]. During photo-oxidation experiments, another
band at 550 nm is visible for the radical from 3V,4V-dimethylcatechin, an
absorption band that has been ascribed to a fast further deprotonation of the
neutral resorcinol-like radical to the corresponding radical anion in basic
medium [29,39]. So the phenoxyl radicals of both rings have been described,
which will enable us to understand the more complex behavior exhibited by
catechin and monomethylated compounds (Table 1).
Figure 1 Techniques used to generate phenoxyl radicals.
130
Table 1 Spectral Characteristics of Phenoxyl Radicals Generated by

Photo-Oxidation and H-Abstractiona
Photo-oxidation
on phenolate
Compounds
3V,4V-Dimethylcatechin
5,7-Dimethylcatechin
Catechin
3V-Methylcatechin
4V-Methylcatechin
Models
of ring
A
B
max
(nm)
495/550
380
380
380/495
495
H-abstraction on phenol
E
max
(nm)
e
(mol1L1cm1)
495
380
380/495
380/495
495
2700
6600
2300
Delay after pulse: 200 ns for photo-oxidation experiments, 5 As for H-abstraction.

Source: Refs. 29 and 39.
a
Indeed, these results obtained on dimethylated compounds led to demonstration that H-abstraction of catechin by tert-butoxyl radicals is selective on the
A-ring, since 70% of the reactivity occurs on the A-ring (absorption at 495 nm),
whereas the 308-nm laser-induced photo-oxidation of catechin phenolate
appears selective on the B-ring with an absorbency at 380 nm [39]. Moreover,
the selectivity of direct irradiation experiments has been correlated to the
deprotonation sequence since only phenolates absorb the laser light and are
much more easily oxidized than neutral phenols. So although the protonation
sequence between the two rings has not been clearly established and is still
under discussion, these results indicate that the B-ring is slightly more acidic
than the A-ring [29], as conrmed by a study of catechin deprotonation followed
by NMR and aording the microscopic pKa [38] (Table 2). Indeed, the
successive deprotonations of the dierent phenolic functions of avan-3-ols
Table 2 Microscopic pKa of Each Catechin

Phenolic Function
Phenolic function
3V
4V
5
7
Source: Ref. 38.
Microscopic pKa
(F0.05)
9.02
9.12
9.43
9.58
131
induce great changes in the chemical environment of various carbons of the

skeleton, as reected in the 13C-NMR chemical shifts [5054]. More precisely,
the deprotonation of a phenolic function induces deshielding for the ipso and
ortho carbon atom and shielding for the para carbon atom. Such selective
behavior allows, with the unambiguous assignment of 13C-NMR signals of
avan-3-ols, the determination of the precise deprotonation site of catechin and
epicatechin [38]. The quantication of each existing species allows determination of the intrinsic pKa. NMR studies are the only method yielding these
microscopic pKa in the case of polyphenolic compounds.
It is now particularly interesting and possible to study the inuence of the
B-ring monomethylation on the reactivity of avan-3-ol, comparing the results
obtained for catechin and for two monomethylated compounds: 3V- and 4Vmethylcatechin, rst of which is the major metabolite currently identied as
circulating in plasma [39]. Although the B-ring monomethylation only enhances
the selectivity of the H-abstraction on the A-ring, it implies drastic changes in
the behavior on photo-oxidation, indicating that the physicochemical parameter
that is greatly aected by methylation is the relative acidity of each phenolic
position. More precisely, the methylation of the 3V position, the most acidic
position in catechin (Table 2), reduces the pKa dierence between the two rings
since during photo-oxidation experiments, both radicals issuing from each ring
appear equally (Table 1). Conversely, the methylation of the 4V position
drastically changes the protonation sequence of avan-3-ols: only the radical
absorbing at 495 nm exists (Table 1), indicating that in this case, the A-ring is
more acidic.
Study of the reactivity of methylated avan-3-ols leads to a better
understanding of the free radical chemical processes of the whole avan-3-ol
family since it shows that the mechanisms of electron and H-atom transfer are
radically dierent and are specic to one moiety of the avan-3-ols: electron
transfer involves the B-ring, whereas H-atom transfer involves the A-ring.
B.
Redox Properties of Flavan-3-ols
The redox potential of interest to understand the biological eects of avan-3-ols

is the one related to phenoxyl radicalphenate couple, as this potential is roughly
1 V lower than the potential of the phenoxyl radicalphenol couple, which
furthermore may transiently involve the oxidation of the aromatic atoms. Standard
potential can be measured by electrochemistry [49] or pulse radiolysis [4044].
However, determining the redox potential of polyphenolic compounds is a real
challenge since for these methods the measurement must be faster than the
subsequent reactions induced by the oxidation of the phenol group in order to
obtain the thermodynamic value. By using ultramicroelectrodes (electrodes with a
micrometer diameter), it has been shown that a very high scan rate, up to 1 million
132
V/s, can be reached; this implies that intermediates that have a lifetime in the
microsecond range can be characterized by direct electrochemical methods [55]. So
the use of fast cyclic voltametry allows determination of the Ej value of coniferyl
alcohol [56]; however, the Ej of a simple catechol such as caeic acid [57] cannot
be determined even using ultramicroelectrodes.
Therefore, the only way to obtain precise values is to use fast cyclic
voltametry on model compounds, for example, the two trimethylated avan-3ols, which oer only one free function hydroxyl. The redox potential of each
phenolic function using fast cyclic voltametry has been investigated [39] (Fig. 2)
and is summarized in Table 3.
As expected, the catechol B-ring is more oxidizable than the resorcinol
A-ring, and these results are in good agreement with the acid-base properties
discussed previously: on each ring, the more basic site is the most oxidable
phenolic function.
C.
Conclusion
The study of the physicochemistry and free radical chemistry characteristics

of methylated avan-3-ols allows [29,39] identication of the two avan-3-ol
Figure 2 High scan rate cyclic voltametry of 3V,5,7-trimethylcatechin (Cj =

0.98 mM) in acetonitrile-0,1M NBu4BF4 on a 10-Am-diameter glassy carbon
ultramicroelectrode (scan rate v = 11500 Vs1). SCE, saturated calomel
electrode reference.
133
Table 3 Redox Potentials of Different

Catechin Analogues
Source: Ref. 39.
radical families, characterization of the intrinsic reactivity of each ring as well

as determination of two thermodynamic values: redox potential and dissociation constant.
All these parameters permit the proposal of a new insight into the
antioxidative properties of avan-3-ols. Whereas in the past these properties
were explained only by H-abstraction process, these results indicate that these
activities also involve an electron transfer since the B-ring of avan-3-ols has
been shown to be the active moiety of the molecule.
IV.
CHARACTERIZATION OF THE METHYLATION SITE OF

FLAVAN-3-OLS BY MASS SPECTROMETRY
The characterization of the metabolism site via the methylation site of avan3-ols is essential for the identication of the metabolites circulating in the
plasma and thus for the understanding of the bioavailability of avan-3-ols.
Indeed, in spite of extensive and detailed studies on their metabolism, little is
known about the precise structures of the metabolites because of the lack of
suitable methodology.
Indeed, until recently [35], the identication and quantication of metabolites in biological samples were often performed indirectly after initial
hydrolysis conjugates with enzyme regardless of the technique used for the
analysis (Table 4): high-performance liquid chromatography combined with UV
[58,59], chemiluminescence [60], uorescence [61], electrochemical [62], mass
spectrometry [63] detection, capillary liquid chromatography coupled with
electrospray mass spectrometry [64], or gas chromatography coupled with mass
spectrometry [16,34,65]. The development of a new liquid chromatography
electrospray ionization mass spectrometry (LC ESI-MS/MS) method that uses
positive ion mode allows unambiguous characterization and dierentiation of
134
Table 4 Characteristics of Various Analytical Methods for the Analysis of

Flavan-3-ols and Their Metabolites in Biological Mediuma
Limit of detection of 3V-methylcatechin. GC/MS, gas chromatography/mass Spectrometry; UV,

ultraviolet; EC, electrochemical; HPLC, high-performance liquid chromatography; CL, chemiluminescence; FL, fluorescence; Nd, not determined; EGCG, epigallocatechin gallate.
each site involved in catechin metabolism [35] and thus allows analysis of crude
biological extracts [66].
A.
Catechin Fragmentation Under Electrospray Ionization

Mass Spectrometry Conditions
The fragmentation of catechin under ESI-MS/MS conditions can be rationalized

by three pathways depicted in Scheme 6 [35]. The rst, I, leads to the unique
A-ring product ion and can be unambiguously attributed to the retro Diels
Alder fragmentation, well known and characteristic of the fragmentation of
avan-3-ols under EI-MS conditions [67] and fast atom bombardment mass
spectrometry (FAB MS) using glycerol [67] or meta-nitrobenzyl-alcohol [67,68]
matrix. But whereas under EI and FAB MS conditions, both A- and B-ring
fragments appear in the spectrum, the MS/MS spectrum obtained using ESIMS/MS shows only one species, the o-hydroxylbenzylcation 1,3A+ ion at m/z
139. The two other fragmentation pathways II and III, which rst yield two
dierent product ions, lead nally to the same fragment ions after loss of water,
CO, and C2H4 (Scheme 6). The major product ions obtained in these two
pathways involve only the B-ring, the 1,2B+ ion at m/z 123 and the 1,4B+ ion at
m/z 165, which can further fragment by loss of water, giving rise to (1,4B+H2O) ion at m/z 147.
135
Scheme 6 Major fragmentations of protonated catechin. The insert gives the ion
nomenclature for protonated flavan-3-ols. (From Ref. 35.)
136
B.
Characterization of the Methylation Cycle

of Flavan-3-ols
The most useful fragmentations in terms of characterization of the methylation or

metabolism sites of catechin are those resulting in structurally informative i,jA+
and i,jB+ ions. Indeed, the study of the dierent mass shifts of the three diagnostic
ions 1,3A+, 1,2B+, 1,4B+ observed between catechin and its methylated analogues
(Fig. 3) allows unambiguous identication of the methylated cycle [35]. For
example, the 28-mass-unit shift of 1,2B+, 1,4B+ in the collision-activated dissociation (CAD) spectrum C of Figure 3 clearly indicates that this is the MS/MS
spectrum of 3V,4V-dimethylcatechin.
More interesting is that the substitution, in this case methylation, changes
the gas phase basicity of the substituted ring and creates a privileged fragmentation pathway [35]. For example, in Figure 3, whereas in the CAD spectra of
catechin A, and of 5,7-dimethylcatechin B, the base peak is the A-ring product
ion at respectively, m/z = 139 and 167 in the case of 3V,4V-dimethylcatechin B,
there is an inversion and the B-ring product ion at m/z = 151 becomes the base
peak. More precisely, the calculations of the proton anity of the dierent
models of A- and B-ring obtained using MNDO methods (Table 5) allow
rationalization of these dierent behaviors. The resorcinol-like ring (A-ring)
generally presents higher basicity than its isomeric catechol-like B-ring, as
suggested by the results obtained for the proton anity of 5,7-dihydroxybenzopyran (849 kJ mol1) and 1,2-dihydroxybenzene (768 kJ mol1). So the
protonation of the more favorable ring leads to the formation of A-ring base
peak at m/z = 139 for catechin. On the other hand, methylation, as expected,
increases the gas phase basicity of the substituted ring: the proton anities
obtained for methylated models such as methoxybenzene, 1,2-dimethoxyben-
Table 5 Proton Affinities Obtained by MNDO Calculations

Conducted at PM3 Levela for Different Catechin Models
Compounds
5,7-Dihydroxybenzopyran
1,2-Dihydroxybenzene
1,2-Dihydroxy, 4-methylbenzene
methoxybenzene
1,2-Dimethoxybenzene
1,2-Dimethoxy, 4-methylbenzene
Position of
protonation
Proton Affinity
(kJ mol1)
C6
C4
C5
849
768
795
807
804
813
Calculated from proton afnity (PA) = DHjf (H+) + DHjf (R) DHjf (RH+),
where DHjf (H+) = 1528 kJ mol_1.
Source: Ref. 35.
a
137
zene (807 kJ mol1) [69], 804, and 813 kJ mol1, are higher than those of
phenol (786 kJ mol1) [69], 1,2-dihydroxybenzene (804 kJ mol1), and 1,2dihydoxy, 4-methylbenzene (813 kJ mol1). So in the case of catechin
methylated analogues, the protonation of the most basic ring leads for 3V,4Vdimethylcatechin to the formation of a B-ring base peak and for 5,7-dimethylcatechin to an A-ring base peak. So substitution induces modication in the
spectrum of catechin-substituted analogues, which can be rationalized by the
examination of the proton anity of the substituted ring.
C.
Characterization of the Methylation Site of Flavan-3-ols
The modications observed in the relative intensity of the major product ions are
specic to the substitution (nature and position). In the case of the isomeric Bring monomethylcatechin, there is an inversion of base peak between the two
isomers as already detected for the dimethylated ones: the 1,3A+ ion at m/z 139 is
the base peak for the 3V-methylcatechin (Fig. 3, spectrum D), whereas the 1,2B+
ion at m/z 137 is for 4V-methylcatechin (Fig. 3, spectrum E). For the other couple
of isomers (on the A-ring), dierences appear on the major product ions 1,3A+
and 1,4B+. So these dierences can be used to determine precisely the site of
methylation on each cycle (3V versus 4V and 5 versus 7) [35].
In order to quantify these dierences and propose criteria to localize
precisely the substituent, branching ratios have been calculated: S relative
intensity of 1,3A+/S relative intensity of 1,3A+, 1,2B+, and 1,4B+ for position 3V
and 4V; and S relative intensity of 1,4B+/S relative intensity of 1,3A+and 1,4B+for
position 5 and 7, which gives 59 (3V-methyl) versus 36 (4V-methylated) for the Bring pair of isomers and 34 (5-methylation) versus 28 (7-methylation) for the Aring pair of isomers.
So isomeric methyl catechin with the same [M+H+] ion can be characterized and dierentiated on the basis of their CAD spectra alone: structurally
informative fragmentation allows one to infer the substitution pattern in the Aor B-ring, whereas the study of the relative intensities of the major product ions
through the determination of the branching ratio indicates the precise site of
substitution on each ring since the higher branching ratio of each isomer pair is
correlated with the substituent position (3V in the case of B-ring isomers and 5 in
the case of A-ring ones) [35].
D.
Application to Crude Biological Samples
Moreover, the sensitivity of this methodology appears to be excellent down to 30

pg/mL, which allows investigations of real biological samples. The LC-ESI-MS/
MS methodology has indeed been applied with success to the analysis of crude
samples obtained by rat liver homogenate [66]. The reconstructed mass chroma-
138
Figure 3 Collision-activated dissociation (CAD) spectra obtained for (A)

catechin, (B) 5,7-dimethylcatechin, (C) 3V,4V-dimethylcatechin, (D) 3V-methylcatechin, (E) 4V-methylcatechin. Voltage cone, 30 V, collision energy, 10 eV; argon
pressure, 2.8.103 mB. (From Ref. 35.)
togram at m/z 305 obtained on a C18 reversed-phase column exhibit two major
peaks (Fig. 4A). The MS/MS spectrum of the major peak (Fig. 4B), which
exhibits a high 139/137 ratio, can be condently attributed to 3V-methylcatechin,
whereas the more noisy MS/MS mass spectrum of the minor peak (Fig. 4C) can
be attributed to 4V-methylcatechin.
Moreover, it has been shown that this method can be generalized to other
series of avan-3-ols such as epicatechin where no standard is available and it is
139
Figure 4 Mass chromatogram and CAD spectra obtained for the two major
peaks of a mixture obtained from rat liver homogenate of catechin and analyzed by
LC-ESI-MS/MS. Voltage cone, 30 V; collision energy, 10 eV; argon pressure,
2.8.103 mB.
140
expected that this kind of technology with a mild ionization technique (ESI)
could be applied to other types of metabolites that are more polar and labile,
such as sulfate or glucuronide metabolites.
V.
CONCLUSIONS
The discovery of the numerous benecial eects of avan-3-ol metabolites on

human health stimulates researchers to investigate more thoroughly the methylated avan-3-ols chemistry: their synthesis, identication, quantication, and
chemical reactivity.
Recently, a new strategy based on successive and selective protections of
the various phenol functions present on avan-3-ols has allowed synthesis of a
whole family of methylated catechin analogues. These materials appear particularly useful for the development of a new analytical tool allowing the
identication of all avan-3-ol metabolites, for the study of their chemical
reactivity, and thus for understanding the mechanism by which these methylated
avan-3-ols behave as antioxidants.
Indeed, with these selectively methylated catechin analogues, it has
been possible to determine thermodynamic constants (redox potentials, pKa)
and to study the intrinsic reactivity of each catechin moiety. Each moiety,
the A- or B-ring, presents its own reactivity: whereas the A-ring is more
specically involved in H-atom transfer, the B-ring is specically involved in
electron transfer. These results lead to a new interpretation of the antioxidant
properties of these molecules: since B-ring of avan-3-ols has been shown to
be the active moiety of the molecule [12], the antioxidant activity involves
mainly an electron transfer and not only an H-atom transfer as is so often
proposed.
Moreover, the access to a whole family of selectively methylated analogues opens new areas of research in the elucidation of the key role of avan-3ol metabolites in human health by oering the possibility of establishing reliable
relation-structure activities in biological tests.
REFERENCES
1.
2.
3.
4.
Porter LJ. Flavans and proanthocyanidins. In: Harborne JB, ed. The Flavonoids:
Advances in Research Since 1986. London, Chapman and Hall, 1994:2356.
Bravo L. Polyphenols: chemistry, dietary sources, metabolism and nutritional
signicance. Nutr Rev 1998; 56:317333.
Kuhnau J. The avonoids: a class of semi-essential food components: their role in
human nutrition. World Rev Nutr Diet 1976; 24:117191.
Ferreira D, Bekker R. Oligomeric proanthocyanidins: naturally occurring Oheterocycles. Nat Prod Rep 1996; 13:411433.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
141
Peterson J, Dwyer J. Flavonoids: dietary occurrence and biochemical activity. Nutr

Res 1998; 18:19952018.
Arts IC, van de Putte B, Hollman PCH. Catechin contents of foods commonly
consumed in the Netherlands. 1. Fruits, vegetables, staple foods and processed
foods. J Agric Food Chem 2000; 48:17461751.
Arts IC, van de Putte B, Hollman PCH. Catechin contents of foods commonly
consumed in the Netherlands. 2. Tea, wine, fruit juices, and chocolate milk. J Agric
Food Chem 2000; 48:17521757.
Cao Y, Cao R. Angiogenesis inhibited by drinking tea. Nature 1999; 398:381.
Jankun J, Selman SH, Swiercz R, Skrypczak-Jankun E. Why drinking green tea
could prevent cancer. Nature 1997; 387:561.
Vergote D, Cren-Olive C, Chopin V, Toillon RA, Rolando C, Hondermarck H,
Le Bourhis X. (-)-Epigallocatechin (EGC) of green tea induces apoptosis of human
breast cancer cells but not their normal counterparts. Breast Cancer Treat 2002;
76:195201.
Leake DS. Eects of avonoids on the oxidation of low-density lipoproteins. In:
Rice-Evans C, Packer L, eds. Flavonoids in health and disease. New York: Marcel
Dekker, 1997:253276.
Cren-Olive C, Tessier E, Duriez P, Rolando C. Structure-activity relationship for
the inhibition of LDL oxidation by catechin methylated metabolites and analogues.
Free Rad Biol Med. In press.
Mangiapane H, Thomson J, Salter A, Brown S, Bell GD, White DA. The inhibition
of the oxidation of low density lipoprotein by (+)-catechin, a naturally occurring
avonoid. Biochem Pharmacol 1992; 43:445450.
Smith MA, Perry G, Richey PL, Sayre LM, Anderson VE, Beal MF, Kowall N.
Oxidative damage in Alzheimers. Nature 1996; 382:120121.
Lee MJ, Wang ZY, Li H, Chen L, Sun Y, Gobbo S, Balentine DA, Yang CS.
Analysis of plasma and urinary tea polyphenols in human subjects. Cancer
Epidemiol, Biomarkers Prev 1995; 4:393399.
Donovan JL, Bell JR, Kasim-Karadas S, German JB, Walzem RL, Hansen RJ,
Waterhouse AL. Catechin is present as metabolites in human plasma after
consumption of red wine. J Nutr 1999; 129:16621668.
Bell JR, Donovan JL, Wong R, Waterhouse AL, German JB, Walzem RL, KasimKaradas S. (+)-Catechin in human plasma after ingestion of a single serving of
reconstituted red wine. Am J Clin Nutr 2000; 71:103108.
Li L, Chan TH. Enantioselective synthesis of epigallocatechin-3-gallate (EGCG),
the active polyphenol component from green tea. Org Lett 2001; 3:739741.
Zaveri N. Synthesis of a 3,4,5-trimethoxybenzoyl ester analogue of epigallocatechin-3-gallate (EGCG): a potential route to the natural product green tea catechin,
EGCG. Org Lett 2001; 3:843846.
Van Rensburg H, van Heerden PS, Bezuidenhoudt BCB, Ferreira D. Enantioselective synthesis of the four catechin diastereomer derivatives. Tetrahedron Lett
1997; 38:30893092.
Van Rensburg H, van Heerden PS, Ferreira D. Enantioselective synthesis of
avonoids. Part 3. trans and cis-Flavan-3-ols methyl ether acetates. J Chem Soc
Perkin Trans 1 1997; 34153421.
142
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.

Nay B, Monti JP, Nuhrich A, Deeux G, Merillon JM, Vercauteren J. Methods in
synthesis of avonoids. Part 2. High yield access to both enantiomers of catechin.
Tetrahedron Lett 2000; 41:3904939051.
Nay B, Arnaudinaud V, Peyrat JF, Nuhrich A, Deeux G, Merillon JM,
Vercauteren J. Total synthesis of isotopically labelled avonoids, 2: 13C-labelled
(F)-catechin from potassium [13C]cyanide. Eur J Org Chem 2000; 12791283.
Birch AJ, W. Clark-Lewis J, Robertson AV. Relative and absolute congurations of
catechins and epicatechins. J Chem Soc 1957; 35863588.
Hathway DE, Seakins JWT. Autoxidation of polyphenols. Part III. Autoxidation in
neutral aqueous solution of avans related to catechin. J Chem Soc 1957; 15621566.
Yoshida K, Ikeda Y, Tsukamoto G. Japanese Patent 1982; 120584.
Akimoto K, Sugimoto I. Degradation of (+) cyanidanol-3 by sodium sulphite in
aqueous solution. II. Reactivity of several (+) cyanidanol-3 dervatives with sodium
sulphite. Chem Pharm Bull 1984; 32:31483154.
Van Dyk MS, Steynberg JP, Steynberg PJ, Ferreira D. Selective O-methylation of
polyhydroxyavan-3-ols via benzylcarbonates. Tetrahedron Lett 1990; 31:2643
2646.
Cren-Olive C, Lebrun S, Hapiot P, Pinson J, Rolando C. Selective protection of
catechin gives access to the intrinsic reactivity of the two phenol rings during Habstraction and photo-oxidation. Tetrahedron Lett 2000; 41:58475851.
Cren-Olive C, Lebrun S, Rolando C. An ecient synthesis of all mono and tri-Omethylated analogues of (+)-catechin including the major metabolites through
sequential protection of catechol ring. J Chem Soc Perkin Trans 1 2002; 6:821830.
Cren-Olive C. Synthe`se, Physico-Chimie et Analyse de Flavan-3-ols. Ph.D.
dissertation, Universite des Sciences et Technologies de Lille, Lille, 2001.
Sweeny JG, Iacobucci GA. Regiospecicity of (+)-catechin methylation. J Org
Chem 1979; 44:22982299.
Steynberg JP, Burger JFW, Young DA, Desmond A, Brandt EV, Steenkamp JA,
Ferreira D. Oligomeric avanoids. Part 3. Structure and synthesis of phlobatannins
related to (-)-Fisetinidol-(4a,6)-and (4a,8)-(+)-catechin prosetinidins. J Chem Soc
Perkin Trans 1 1988; 12:33233329.
Donavan JL, Luthria DL, Stremple P, Waterhouse AL J. Analysis of (+)-catechin,
(-)-epicatechin and their 3V- and 4V-O-methylated analogs: a comparison of sensitive
methods. Chromatogr B 1999; 726:277283.
Cren-Olive C, Deprez S, Lebrun S, Coddeville B, Rolando C. Characterization of
methylation site of avan-3-ols by LC ESI MS/MS. Rapid Commun Mass Spectrom
2000; 14:23122319.
Kiatagrajai P, Wellons JD, Golob L, White JD J. Kinetics of epimerization of (+)
catechin and its rearrangement to catechinic acid. Org Chem 1982; 47:29102912.
Kennedy JA, Munro MHG, Powell HKJ, Porter LJ, Foo Y. The protonation
reactions of catechin, epicatechin and related compounds. Aust J Chem 1984;
38:885892.
Cren-Olive C, Wierulesky JM, Maes E, Rolando C. Catechin and epicatechin
deprotonation followed by 13C NMR. Tetrahedron Lett 2002: 43:45454549.
Cren-Olive C, Hapiot P, Pinson J, Rolando C. Free radical chemistry of avan-3-
143
ols: determination thermodynamics parameters and of kinetic reactivity from short

(ns) to long (ms) time scale. J Am Chem Soc 2002; 124:1402714038.
40. Steenken S, Neta P. One-redox potentials of phenols: hydroxy-and aminophenols
and related compounds of biological interest. J Phys Chem 1982; 86:36613667.
41. Jovanovic SV, Steenken S, Tosic M, Marjanovic B, Simic MG. Flavonoids as
antioxidants. J Am Chem Soc 1994; 116:48464851.
42. Jovanovic SV, Hara Y, Steenken S, Simic MG. Antioxidant potential of
gallocatechins: a pulse radiolysis and laser photolysis study. J Am Chem Soc
1995; 117:98819888.
43. Jovanovic SV, Steenken S, Hara Y, Simic MG. Reduction potentials of avonoid
and model phenoxyl radicals: which ring in avonoids is responsible for
antioxydant activity? J Chem Soc Perkin Trans 2 1996; 11:24972504.
44. Jovanovic SV, Steenken S, Simic MG, Hara Y. Antioxidant properties of
avonoids: reduction potentials and electron transfer reaction reactions of avonoid
radicals. In: Rice-Evans C, Packer L, eds. Flavonoids in Health and Disease. New
York: Marcel Dekker, 1997:137161.
45. Bors W, Heller W, Michel C. The chemistry of avonoids. In: Rice-Evans C, Packer
L, eds. Flavonoids in Health and Disease. New York: Marcel Dekker, 1997:111
136.
46. Bors W, Michel C. Antioxidant capacity of avanols and gallate esters: pulse
radiolysis studies. Free Radic Biol Med 1999; 27:14131426.
47. Bors W, Michel C, Schikora S. Interaction of avonoids with ascorbate and
determination of their univalent redox potentials: a pulse radiolysis study. Free
Radic Biol Med 1995; 19:4552.
48. Bors W, Michel C, Stettmaier K. Electron paramagnetic resonance studies of radical
species of proanthocyanidins and gallate esters. Arch Biochem Biophys 2000;
374:347355.
49. Hodnick WF, Milosavljevic EB, Nelson JH, Pardini RS. Electrochemistry of
avonoids: relationships between redox potentials, inhibition of mitochondrial
respiration, and production of oxygen radicals by avonoids. Biochem Pharmacol
1988; 37:26072611.
50. Agrawal PK, Schneider HJ. Deprotonation induced 13 C NMR shifts in phenol and
avonoids. Tetrahedron Lett 1983; 24:177180.
51. Jarosszewski JW, Matzen L, Frolund B, Krogsgaard P. Neuroactive polyamine
wasp toxins: nuclear magnetic resonance spectroscopic analysis of the protolytic
properties of philanthotoxin-343. J Med Chem 1996; 39:515521.
52. Berger S. The pH dependence of phenolphthalein: a 13C NMR study. Tetrahedron
1981; 37:16071611.
53. Haran R, Nepveu-Juras F, Laurent J.P. Les amines bioge`nes (phenol-et catecholamines): attribution des spectres de resonance 13C et etude de la deprotonation. Org
Magn Reson 1979; 12:153158.
54. Maciel GE, James RV. Solvent eects on the 13C chemical shift of the substituted
carbon atom of phenol. J Am Chem Soc 1964; 86:38933894.
55. Andrieux CP, Hapiot P, Saveant JM. Fast kinetics by means of direct and indirect
electrochemical techniques. Chem Rev 1990; 90:723738.
144
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.

Hapiot P, Pinson J, Neta P, Francesch C, Mhamdi F, Rolando C, Schneider S.
Mechanism of oxidative coupling of coniferyl alcohol. Phytochemistry 1994;
36:10131020.
Hapiot P, Neudeck A, Pinson J, Fulcrand H, Neta P, Rolando C. Oxidation of caeic
acid and related hydroxycinnamic acids. J Electroanal Chem 1996; 405:169176.
Goto T, Yoshida Y, Kiso M, Nagashima H. Simultaneous analysis of individual
catechins and caeine in green tea. J Chromatogr A 1996; 749:295299.
Maiani G, Serani M, Salucci M, Azzini E, Ferro-Luzzi A. Application of a new
high-performance liquid chromatographic method for measuring selected polyphenols in human plasma. J Chromatogr B 1997; 692:311317.
Nakagawa K, Miyazawa T. Chemiluminescence-high-performance liquid chromatographic determination of tea catechin, (-)-epigallocatechin-3-gallate, at picomole
levels in rat and human plasma. Anal Biochem 1997; 248:4149.
Ho Y, Lee YL, Hsu KY. Determination of (+)-catechin in plasma by highperformance liquid chromatography using uorescence detection. J Chromatogr B
1995; 665:383389.
Lee MJ, Prabhu S, Sheng S, Meng X, Li C, Yang CS. An improved method for the
determination of green and black tea polyphenols in biomatrices by highperformance liquid chromatography with coulometric array detection. Anal
Biochem 2000; 279:164169.
Yong YL, Kwokei JNg, Shenjiang Y. Characterization of avonoids by liquid
chromatography-tandem mass spectrometry. J Chromatogr A 1993; 629: 389
393.
Dalluge JJ, Nelson BC, Thomas JB, Welch MJ, Sander LC. Capillary liquid
chromatography/electrospray mass spectrometry for the separation and detection of
catechins in green tea and human plasma. Rapid Commun Mass Spectrom 1997;
11:17531756.
Luthria DL, Jones AD, Donavan JL, Waterhouse AL. GC/MS determination of
catechin and epicatechin levels in human plasma. J High Resol Chromatogr 1997;
20:621623.
Cren-Olive C, Lenoir M, Croq-Lemarrec F, Coddeville B, Salzet M, Rolando C.
Characterization of metabolism site of avan-3-ols by LC ESI MS/MS. 49th
American Society for Mass Spectrometry Conference, Chicago, May 2731, 2001.
Miketova P, Schram KH, Whitney JL, Kerns EH, Valvic S, Timmermann BN, Volk
KJ. Mass spectrometry of selected components of biological interest in green tea
extracts. J Nat Prod 1998; 61:461467.
Stobneicki M, Popenda M. Flavan-3-ols from seeds of Lupinus angustifolius.
P.J. Linstrom and W.G. Mallard, eds., NIST Chemistry WebBook, NIST Standard
Reference Database Number 69, July 2001, National Institute of Standards and
Technology, Gaithersburg MD, 20899 (http://webbook.nist.gov).
6
Investigation of Flavonoids
and Their In Vivo Metabolite Forms
Using Tandem Mass Spectrometry
Gunter G. C. Kuhnle
London, England
I.
INTRODUCTION
The purpose of this chapter is to review dierent methods to investigate

avonoids and their in vivo metabolites using mass spectrometry (MS). In
particular, the focus is the elucidation of the structure of in vivo avonoid
metabolite forms. Table 1 shows the structure and m/z values for the compounds
described in this chapter.
The detection and characterization of avonoid conjugates and metabolites
are crucial for the investigation of their bioactivity. During their passage through
the gastrointestinal tract into the blood circulation and potential target organs,
several modications occur through metabolism in the small intestine, the liver,
and degrading enzymes of the colonic microora. Further modications may
occur after cellular uptake. The major in vivo avonoid conjugates and metabolites are O-glucuronidated, O-sulfated, and, in the case of catechol structures, Omethylated derivatives [1]. In addition, the colonic microora generate breakdown products by ring cleavage, producing secondary metabolites such as
phenolic acids with varying saturated chain lengths and other derivations. Mass
spectrometric techniques provide a sensitive and powerful tool for both the
detection and the structural elucidation of such metabolites.
Most methods developed for the mass spectrometric investigation of
avonoids have focused on biological material from plants, as a result of the
importance of avonoids as natural products and potential medicinal prepara145
146
Kuhnle
Table 1 Structures and Names for Selected Flavones and Flavanolsa
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
a
Trivial name
[M+H+]+
R1
R2
R3
R4
Apigenin
Luteolin
Acacetin
Chrysoeriol
Kaempferol
Quercetin
Myricetin
Isorhamnetin
Chrysin
Galengin
Kaempferid
Eriodyctol
Naringenin
Isosakuramnetin
(Epi)catechin
271
287
285
301
287
303
319
317
253
269
299
287
271
285
291
OH
OH
OCH3
OH
OH
OH
OH
OH
H
H
OCH3
OH
H
H
OH
H
OH
H
OCH3
H
OH
OH
OCH3
H
H
H
OH
OH
OCH3
OH
H
H
H
H
H
H
OH
H
H
H
H
H
H
H
H
H
H
H
H
OH
OH
OH
OH
H
OH
OH
H
H
H
OH
All structural information refers to structure i, except for number 15, (epi)catechin, which refers to
structure ii.
tions. However, these compounds dier from the expected metabolites in in vivo
mammalian systems in several ways: rst, avonoids occur in plants normally as
glycosides with one or more sugar residues (with the exception of avan-3-ols),
and second, the glycosylation is not limited to O-linkages; indeed, C-glycosides
are common.
Mass spectrometric investigations of avonoidsespecially regarding
fragmentation reactions and structural elucidationhave been undertaken using
electron impact (EI) and chemical ionization (CI) techniques [26]. Even though
avonoids and their glycosides are polar and nonvolatile, it has been possible to
yield intense signals for the molecular ion of the aglycones. In contrast, it has
been dicult to obtain data for glycosides, even by using derivatized compounds
[7]. The introduction of soft ionization techniques, which allow the generation of
intact ions of larger molecules without fragmentation, such as electrospray
Mass Spectrometry
147
ionization (ESI) [8,9] and atmospheric pressure chemical ionization (APCI), in

recent years increased the application of mass spectrometry to the analysis of
these compounds. Using these techniques, it is possible to obtain intense signals
for the quasi-molecular ions, [M+H+]+ and [MH+], even for the glycosides.
Furthermore, both techniques allow the hyphenation of mass spectrometric
detectors with chromatographic separation devices such as high-performance
liquid chromatography (HPLC/MS).
Structural information can be obtained by tandem MS experiments. The
technique mainly used is low-energy collision-induced decomposition (CID MS/
MS). With instruments equipped with a quadrupole ion trap (QIT) mass analyzer,
several consecutive tandem MS experiments on sequential product ions can be
performed (MSn), permitting structural information from one single analysis.
Characteristic fragmentation patterns can also be used for the identication of
certain compounds in LC/MS.
II.
MASS SPECTROMETRY OF FLAVONOIDS
There are several possibilities to investigate avonoids and their in vivo

metabolites by using the mass spectrometric methods described. The main
two methods are direct infusion using a syringe and ow injection either with or
without chromatographic separation. Using direct infusion, the sample solution
is infused directly into the ion source of the mass spectrometer, normally by
using a syringe and a syringe pump. This method allows a long and thorough
investigation of the sample, including the acquisition of data for several
consecutive fragmentation steps (MSn experiments), and is therefore mainly
used for structural characterization. However, this method normally requires a
puried sample as a separation is not possible. Without purication, matrix
eects and contaminants can lead to ion suppression eects and thereby
decrease the sensitivity of the instrument. Furthermore, a large sample amount
is necessary for longer investigations as the sample is normally infused with a
ow rate of about 3 to 10 AL/min.
Using ow injection reduces the sample amount required and allows the
separation of the sample components before the mass spectrometric analysis.
However, ow injection allows only a short investigation time for each signal,
which may be too short for a thorough analysis using tandem MS experiments.
A.
Experimental
With respect to mass spectrometric analysis, the most interesting physicochemical properties of avonoids are the proton anity and pKa values of the hydroxyl
148
Kuhnle
groups, as these groups are the main sites of protonation or deprotonation and
therefore ion formation. Unfortunately, few data have been published so far.
However, Cecile Cren-Olive and Christian Rolando provide some data on avan3-ols in their chapter (see Chap. 5).
The large number of hydroxyl groups associated with avonoids would
suggest a basic environment and negative ion scan. However, many specic
avonoid structures are unstable under basic conditions and are thus likely to
decompose at pH above neutrality. The decomposition of these compounds
mainly involves the formation of quinonic structures from the catechol group at
the B-ring. For this reason, a neutral orbetteracidic environment is
normally necessary.
1.
Ion Polarity
Flavonoids can be detected in both positive and negative ion modes, even under
acidic conditions. Whereas the positive ion mode often generates higher yields,
the noise level is lower in the negative ion mode, thus improving the quality of
the signals. Furthermore, the fragmentation pathway can be inuenced by the
ion polarity [10], and it has been reported that phenolic compounds show less
fragmentation in negative mode than in positive ion mode [11]. Thus, using
positive ions can be advantageous for structure elucidation, whereas the
negative ion mode is advantageous for the detection of compounds. Investigations show that the optimal ionization polarity depends very much on the
compound used. For this reason, preliminary investigations regarding the
polarity used are important.
2.
Atmospheric Pressure Chemical Ionization

and Electrospray Ionization
The most commonly applied ionization methods for LC-MS are electrospray
ionization (ESI) and atmospheric pressure chemical ionization (APCI). In ESI, the
ions are considered to be preformed in solution [12] and subsequently extracted in
the spray; in APCI the molecules become ionized inside the source by using a
corona discharge. In ESI, the sample is sprayed into a mist composed of small
charged droplets by using a high voltage and an assisting nebulizing gas. In APCI
sources, the sample is evaporated by using high temperatures of up to 600jC and
ionized by using a corona discharge. Though avonoids and their metabolites are
unstable at high temperatures, the high temperatures in the APCI source do not
seem to cause any damage to these compounds. Main advantages of the APCI
source over the ESI source are the increased range for ow rates and the potential
to obtain ions from aqueous solutions even at ow rates well above 1 mL/min.
However, there are no studies comparing the best conditions for the
investigations of avonoids using APCI or ESI. Investigations using pesticides
Mass Spectrometry
149
[13] showed that the response of the compounds depends largely on their
chemical properties: i.e., APCI in the positive mode was more sensitive for basic
nonionic compounds and positive ESI was more sensitive for positively charged
ones. For acidic compounds, negative ESI proved to be the best ionization
method. In general, it proved that compounds working well with positive APCI
also worked well with positive ESI, but not necessarily vice versa. However,
these data cannot be used unconditionally for the investigation of avonoids and
their metabolites, and further investigation is required.
3.
Adduct Formation
Depending on the solvent used, the formation of adducts, for example,

[M+Na+]+ or [M+formiate+]+ is frequently observed in the positive ion mode.
For a larger abundance of the ion under investigation, but also for an accurate
quantication, it is often necessary to remove those adducts. This can be done
either by using in source fragmentation or by increasing either the transfer
capillary temperature or the intensity of the drying gas.
B.
Fragmentation
1.
Fragment Nomenclature
To obtain structural information, compounds can be fragmented by several

techniques. The technique mainly employed with soft ionization techniques is
low-energy collision-induced dissociation (CID). To describe the fragments
generated, the nomenclature proposed by Ma and colleagues [14] is generally
used. Compared with the system introduced by Marby and Markham [7], this
nomenclature is conceptually more similar to that commonly used for the
description of carbohydrate fragmentation in CID spectra of glycoconjugates
[15]. In this nomenclature, i,jA+ and i,jB+ labels are used to designate primary
product ions containing intact A- and B-rings. The superscripts i and j indicate
the bonds of the C-ring that have been cleaved (Fig. 1). Further fragmentations
of the A- or B-ring products, in particular the loss of small molecules like H2O
or CO2, are indicated by the combined use of the respective fragment and the
lost molecules (e.g., [0,2A+H2O]+).
2.
Fragmentation Mechanisms and Major Fragments
With respect to the elucidation of unknown metabolites, the most useful

fragments are those involving the cleavage of two bonds in the C-ring, in
particular 1/3, 0/2, and 0/4 [14], leading to fragments shown in Fig. 2.
Protonation on either the ether oxygen atom in the C-ring or the C-3 atom
leads to two dierent quasi-molecular ions, which can both undergo a retro-
150
Kuhnle
Figure 1 Nomenclature and selected product ionsderived from retro-DielsAlder reaction - of protonated compounds under low-energy CID as proposed in
Ref. 14. In ijX+, the superscript number (ij) denotes the C-ring bond cleaved, the
letter X describes the ring remaining in the fragment.
Figure 2 Mechanisms proposed for the fragmentation of flavonoid [M+H+]+ ions

to generate 1,3A+ and 1,3B+ ions [14] via a retro-Diels-Alder reaction. An alternative
is pathway III leading to a 2,4B+ fragment, also via a retro-Diels-Alder reaction.
Mass Spectrometry
151
Diels-Alder (RDA) reaction, leading to dierent fragmentation products (1,3A+

and 1,3B+ ions, pathway Ia and Ib in Fig. 2). An alternative to this is pathway III,
resulting in 0,4B+ ions, which can undergo a further fragmentation by the loss of
water, leading to [0,4B+H2O]+. This pathway seems to be specic for CID as it
has not been observed under EI conditions [7]. 0,2A+ and 0,2B+ ions can be
formed after a proposed pathway II. Protonation at the C-3 and C-2 position may
lead to a cleavage of bonds 0 and 2 in the C-ring and to the formation of these
ions (Fig. 3) [14]. Other fragments include the loss of small molecules such as
H2O, CO, or CO2. Table 2 and Table 3 show major fragments resulting from
retro-Diels-Alder reactions and their intensities for selected compounds in
positive and negative ion modes, which can be used for structure assignment
and identication of these compounds and their metabolites [14,16].
Useful fragments for interpreting structures are those including only one
possible site for metabolic modication of the A- or B-ring, showing the actual
site of modication. However, few fragments are described with these properties, for example, an [MH+CO2CO] fragment of luteolin (2) after loss of
the 5-hydroxyl group [16]. But most fragments described do not reveal the
actual site of modication within the respective ring.
3.
Fragmentation of In Vivo Metabolites
A main group of in vivo metabolites of avonoids are conjugates with

glucuronic acid, which demonstrate behavior in tandem MS experiments similar
to that of other O-glycosides. O-Glycosides normally show the aglycone as an
Figure 3 Mechanism proposed for the fragmentation of flavonoid [M+H+]+ ions

leading to 0,2A+ and 0,2B+ ions, as proposed in Ref. 14.
153
163
145
121
119
(100)
(7)
(12)
(10)
(24)
A+
1,3
See Table 1 for details.

Source: Ref. 14.
Base peak
(100%)
1,3 +
A
0,4 +
B
[0,4B+H2O]+
0,2 +
B
1,3 +
B
0,2 +
A
[0,2ACO]+
[1,3B2H]+
[1,4A+2H]+
[1,3AC2H2O]+
Compound
A
153
179
161
137
135
(100)
(10)
(10)
(9)
(21)
1,3 +
2
[M+H+
CH3]+
153 (9.7)
177 (0.6)
159 (2.4)
135 (1.1)
133 (4.3)
3
[M+H+
CH3]+
153 (3.0)
175 (0.7)
151 (0.9)
149 (1.0)
165
137
133
127
111
(73)
(48)
(21)
(11)
(16)
165
137
149
127
111
121 (8.1)
(90)
(20)
(24)
(14)
(22)
137 (4.8)
153 (100)
A+
6
[M+H+
CO]+
153 (39)
1,3
165
137
165
127
111
(8.1)
(12)
(81)
(11)
(20)
165
137
163
127
111
(6.8)
(1.4)
(2.2)
(1.4)
(1.4)
151 (5.3)
[M+H+
CH3]+
153 (4.4)
[M+H+H2O
CO]+
153 (63)
153 (39)
Table 2 Principal Product Ions Including an Intact A- or C-Ring with Their Relative Intensities (Base Peak Equals 100) for
Selected Flavonoids Generated from [M+H+]+ Ions by Low-Energy Collision-Induced Dissociationa
152
Kuhnle
151
149
107
117
(10)
(36)
(1)
(1)
133 (1)
151 (4)
151 (1)
[MH+
[MH+] [MH+]
CO2]
a
See Table 1 for details.
Source: Ref. 16.
1,3
1,2
A
A
1,4
[ B2H]
[1,3ACO2]
1,3
B
[1,2ACO]
1,2
B
[1,2ACOCO2]
1,4
A
Base Peak
Compound
151 (67)
121 (1)
107 (1)
179 (100)
A
1,2
10
[MH+] [MH+]
9
A
1,3
A
12
107 (4)
132 (1)
125 (1)
107 (1)
135 (3)
151 (100) 151 (100)
1,3
11
151 (12)
[MH+
C2H2O]
14
125 (1)
107 (3) 107 (1)
A
1,3
13
Table 3 Principal Product Ions Including Intact A- or C-Ring with Intensities (Base Peak Equals 100) for Selected Flavonoids
Generated from [MH+] Ions by Low-Energy Collision-Induced Dissociationa
Mass Spectrometry
153
154
Kuhnle
Figure 4 Product ion spectra of epicatechin-O-glucuronide (top) and quercetinO-glucuronide. Whereas the spectrum for quercetin-O-glucuronide only shows the
neutral loss of glucuronic acid (m/z 303), the spectrum for epicatechin-Oglucuronide showsapart from the loss of water and other small molecules
several characteristic fragments like [1,3A++glucuronic acid]+ (m/z 315).
Mass Spectrometry
155
intense fragment ion. However, O-glycosides with more than one sugar residue
can undergo a rearrangement with an internal sugar residue loss [17]. Monoglycosides often show a fragmentation pattern similar to that of their aglycone,
indicating that it is also possible to elucidate structures from the glycosides. The
5- and 7-O-glucuronidated epicatechin exhibits a fragmentation pattern very
similar to that of the aglycone, showing similar fragments including the
metabolic modication (e.g., [1,3A++glucuronic acid]+). However, other compounds produce a completely dierent product ion spectrum; for example, the
product ion spectrum of the quercetin-O-glucuronide exhibits the neutral loss of
only glucuronic acid, but no other fragments, even though they are present in the
product ion spectrum of the aglycone (Fig. 4).
O-Methylated compounds often show the neutral loss of the methyl group,
however, not all avonoids behave in this manner. Butin contrast to the Oglucuronidesmost O-methylated compounds generate several product ions,
which allow identication and structure elucidation.
However, the inuence of a modication of the aglycone, for example, by
O-methylation or O-glucuronidation, must not be underestimated, as this has an
inuence on the fragmentation mechanisms and thereby the abundance of
certain fragment ions.
III.
STRUCTURE ELUCIDATION
Structural elucidation of metabolites focuses mainly on the kind and the site of
modication. The main conjugations expected for in vivo metabolites are Omethylation, O-glucuronidation, and O-sulfation. Theyand possible combinationscan be distinguished easily by the mass they add to the original molecule.
In contrast to this, the detection of the site of modication is more dicult as
most product ions still have an intact A- or C-ring, which is the main site of
metabolism.
To elucidate the structure of avanoid metabolites and investigate metabolic
processes, tandem MS experiments (fragmentations) can be performed. O-Glucuronidated compounds normally exhibit a neutral loss of glucuronic acid,
showing an intense signal for the aglycone. In contrast to this, O-methylated
compounds often show only a very small signal for the original (unmethylated)
compound. Table 4 gives a compilation of product ions and their intensities for
some possible avonoid metabolites. Metabolites in general often exhibit a
fragmentation pattern similar to that of the original compounds, generating
[i,jA++glucuronic acid]+ or [i,jB++CH3]+ ions, for example, epicatechin-5O-hD-glucuronide shows an intense signal at m/z 315, corresponding to the glucuronidated RDA product of the A-ring ([1,3A++glucuronic acid]+). Comparing these
ions with the corresponding product ions of standard compounds can reveal a lot
156
Kuhnle
Table 4 Molecular Ion and Major Fragments (Intensity > 10%) and Their
Intensities (Percentage of the Intensity of the Most Abundant Signal) of
Flavonoid Metabolites, Acquired by Direct Infusion by Using Electrospray Mass
Spectrometry and Low-Energy Collision-Induced Dissociation
Compound
Kaempferol-O-h-D-glucuronide
Naringenin-O-h-D-glucuronide
Hesperetin-O-h-D-glucuronide
[M+H+]+
Major fragments
9
8
30
[MGlucuronic acid+H+]+ (100)

[MH2O+H+]+ (30)
[MCH3+H+]+ (<1) [MH2O+H+]+ (12)
[1,2B++CH3]+ (100) 1,3A+ (28)
[1,5B++CH3]+ (30)
[0,5A++Glucuronic acid]+ (35)
[MCH3+H+]+ (<1)
O-Methylcatechin
Epicatechin-O-h-D-glucuronide
3V-O-Methylquercetin [14]
Source: Data from the authors unless indicated otherwise.
of information concerning the structure of the metabolites under investigation.

However, it remains dicult to discriminate between a modication within the Aor the C-ring, respectively (for example, at the 5- or 7-hydroxy group), as these
rings appear to be very stable under the CID conditions applied; for example, the
binding moiety of glucuronic acid on quercetin could not be resolved by using a
tandem MS experiment [18]. Thus, the main dierences between spectra are the
intensities for product ions of the respective compounds, as the modication
inuences the stability of the fragment ions and thus their abundance. By having
standards available, it is often possible to compare the intensities of the respective
fragments to obtain further information concerning the actual site of modication.
For example, the comparison of the mass spectra for 3V-O-methylated and 4V-Omethylated epicatechin show dierences in the intensity for several product ions,
especially for the RDA product 1,3A (see Fig. 5). The probable reason for the
higher intensity of this product is that the methylation prevents the 4V-hydroxyl
group from being deprotonized; thus the molecule becomes ionized on a
dierent site, leading to the altered fragmentation pattern (see Chap. 5).
Complexes of avonoidsfor example, with transition metalsshow a
fragmentation pattern dierent from that of the fragmentation pattern of [M+H+]+
or [MH+] ions. For this reason using those complexes may be a solution for
this problem as they provide additional information regarding the structure of the
compound under investigation. Sodium complexes [M+Na+]+ generate a more
complex fragmentation pattern, but consistent pathways for the determination of
glycosylation sites have not yet been found. Ternary complexes with cobalt(II)
Mass Spectrometry
157
Figure 5 Product ion spectra of 3V-O-methylated (top) and 4V-O-methylated epicatechin, acquired in negative ion mode. The base peak (m/z 259) corresponds
with the loss of CO2. The main difference between both spectra is the intensity of the
1,2
A ion (m/z 137) which is much more abundant for the 4V-methylated compound.
and 2,2V-bipyridine of avonoids have, however, proved to be very useful for the
dierentiation between glycoside attachment positions (3 vs. 5 position), as they
generate a dierent fragmentation patterns [19]. This could also be useful to
dierentiate between dierent conjugates. Furthermore, complexation of avonoids with transition metals increases the signal intensity by about one order of
magnitude [20].
IV.
LIQUID CHROMATOGRAPHYMASS SPECTROMETRY
A.
Acquisition Modes
Using LC-MS as an analytical tool for the identication of metabolites can be

regarded as a process of structural elucidation of potential metabolites by
158
Kuhnle
interpretation of product ion mass spectra of selected precursor ions, guided by

concepts of biotransformation pathways [21]. Using an MS detector allows the
application of the information acquired by structure elucidation to the detection
of these compounds in biological samples. The capabilities of this technique
allow a selective and sensitive detection of the compounds (and their metabolites)
under investigation.
The basic experiment that can be performed is a full ion scan during the
complete chromatographic run to look for the original compound and possible
metabolites by their calculated mass. As there is a large number of possible
metabolites, searching manually for possible products can be very timeconsuming. Therefore, metabolic proling software such as Metabolite ID,
which searches for all possible metabolites, can be very helpful. In addition,
these programs often include sophisticated background subtraction algorithms,
which reduce the noise resulting from the biological matrix (Fig. 6).
In spite of the abilities of a full scan to detect all metabolites in a sample, it
does have major disadvantages: rst, the detected mass is often insucient for the
denite identication of a compound; even together with a UV spectrum and the
HPLC retention time, there is still some uncertainty regarding multiple modications or isomerization. Second, the signal-to-noise ratio is much higher in the
full scan modeeven with noise reduction algorithmsthus decreasing the
sensitivity. Depending on the mass analyzer used, dierent recording modes such
as selected ion monitoring (SIM) or selected reaction monitoring (SRM) can
increase the sensitivity and improve the results remarkably.
In the SIM mode, the mass spectrometer monitors only the signal for a
single mass-to-charge ratio (m/z). This recording mode is more selective and
therefore provides a higher signal-to-noise ratio. In instruments that apply a
mass lter as analyzersuch as quadrupole instrumentsthis also directly
increases the sensitivity. While these analyzers scan through the desired mass
range for each full scan, monitoring each m/z only for a short time, they can
spend much more time if they focus on a single ion, thus increasing the
sensitivity. However, this advantage applies only for the kind of instruments
in which only the selected ion(s) can reach the detector and trigger a signal.
Other types of analyzers, such as ion traps or time-ight-mass spectrometers
(TOF), can provide only a simulated SIM. In normal ion traps, there is almost
no advantage in using SIM, as the sensitivity of the instrument is still determined
by the overall amount of ions trapped, regardless their m/z. As in the rst step,
all ions are trapped; a contaminant ion with a high intensity can still suppress the
signal of the ion(s) under investigation, even though it does not appear in the
spectrum. There are, however, techniques that lead to an improved sensitivity
for SIM with quadrupole ion traps (QITs), for example, selected ion storage
(SIS), in which selected ions are stored inside the trap, or the application of an
RF voltage to assure that only selected ions are stored in the trap. With these
Mass Spectrometry
159
Figure 6 Acquisition modes in tandem MS with two mass analyzers. Multiple

arrows indicate a full scan, a single arrow a single ion scan (mass filter fixed on a
single m/z). In the neutral loss scan, a full scan with a constant mass difference is
performed with both analyzers, selecting only ions which show a neutral loss.
(Graphic adapted from Ref. 23.)
techniques, the advantage of SIM in quadrupole instruments can also be

achieved with QIT.
SRM is an extended SIM mode, which uses the tandem MS capabilities of
the instrument. As with SIM, ions of a certain m/z are selected and isolated, but
these ions are fragmented, normally by collision-induced dissociation (CID).
On the product ions formed, a selected ion monitoring is performed. Selecting
specic fragments, this technique provides a very high specicity and thus a
high signal-to-noise ratio, thereby increasing the sensitivity. In triple quadrupole
instruments, selecting the parent ion with the rst quadrupole and performing
160
Kuhnle
either a SIM or a full scan with the third quadrupole performs this. Performing a
full scan on the product ions (product ion scan) provides fragment ion spectra for
the compounds under investigation with the appropriate mass that allows
structure elucidation.
A special form of this technique is the scan for neutral losses, which is a
speciality of triple quadrupole instruments and only available with these
instruments. Neutral losses occur frequently during CID of metabolites. For
example, epicatechin-5-O-glucuronide loses glucuronic acid as a neutral
during CID, with epicatechin as detectable ion. Therefore, neutral loss scans
are an important and valuable tool for the investigation of metabolites. In this
scan mode, both quadrupoles perform a full scan with a xed mass-to-charge
ratio dierence: i.e., the second mass ltering quadrupole is always set to a
mass-to-charge ratio lower than the settings of the rst quadrupole, corresponding to the anticipated neutral loss. Thus, only ions that lose an appropriate
neutral, for example, glucuronic acid, are detected.
Another specialty of triple quadrupole instruments is the parent ion scan,
which is the reverse form of a product ion scan. Whereas in a product ion scan the
rst quadrupole is xed to a certain mass-to-charge ratio and the postfragmentation chamber quadrupole is either xed or in scanning mode, the opposite
arrangement is used for the parent ion scan: i.e., all ions undergo a collisioninduced fragmentation, but only ions generating a fragment of a certain massto-charge ratio generate a signal. Applying a parent ion scan for a characteristic
fragment, it is possible to detect yet unknown metabolites. For example, a
precursor ion scan for the 1,3A+-ion, the RDA product of epicatechin (m/z 139),
can reveal all epicatechin metabolites with an unmodied A-ring [22]. This kind
of acquisition can also be performed with a single quadrupole instrument by
using the in-source fragmentation and monitoring a selected fragment by using
SIM. But, in contrast to triple quadrupole instruments, it is not possible to
determine the mass of the precursor ion with a single quadrupole instrument.
In contrast to the full scan, these acquisition modes require more
information on the compounds under investigation. Especially for SRM mode
acquisitions, information on the product ion(s) of the respective compounds is
[16] necessary to perform an analysis. These techniques provide higher sensitivity, but information can be lost as unexpected metabolites remain undetected.
Therefore, these methods are suitable only for the detection and quantication of
known (or at least predicted) metabolites, even though a precursor ion scan or
neutral loss scan can detect a large range of possible metabolites. However, as
some metabolic reactions lead to an altered fragmentation pattern, there is still
the chance that some products are missed.
Whereas a full scan reveals only the mass of a compound in the sample
under investigation, a product ion scan is normally limited to a certain number of
precursor ions, as a large number of simultaneously conducted product ion scans
Mass Spectrometry
161
decreases the sensitivity. Sophisticated acquisition software allows dependent

tandem MS experiments by selecting certain ions from a full scan (either the
most intense ion from a list or from all ions in a full scan), thus providing
product ion spectra for the most abundant ions in a sample. This is an advantage
for the characterization of metabolites, as a large number of product ion spectra
can be acquired without any eect on the sensitivity. Nevertheless, there are also
disadvantages, especially with biological samples. Having a high background
either by a biological matrix or sample contamination may confuse the ion
selection algorithm and thereby prevent the fragmentation of the appropriate
ions. For this reason, it is necessary either to remove all contaminations and
background ions from the sample or to use algorithms to prevent unwanted ions
from triggering the acquisition algorithm.
V.
CONCLUSIONS
Mass spectrometry provides a very powerful and versatile tool for the investigation of avonoid in vivo metabolites. This technique allows both structural
characterization with tandem MS experiments and the detection of metabolites
by using LC/MS.
Characteristic fragments, mainly deriving from a retro-Diels-Alder reaction within the C-ring, allow the elucidation of the structure of the metabolites. Additional information, such as proton anity or pKa values, allows one
to distinguish between dierent isomers, such as 3V- and 4V-O-methylated
compounds. The capabilities of quadrupole or quatrupole ion trap (QIT)
instruments allow the detection of even small amounts of the metabolites in
biological samples.
ACKNOWLEDGMENT
I am grateful to Anna Przyborowska for her collaboration.
REFERENCES
1.
2.
Kuhnle G, Spencer JPE, Schroeter H, Shenoy B, Debnam E, Srai S, Rice-Evans C,

Hahn U. Epicatechin and catechin are O-methylated and glucuronidated in the small
intestine. Biochem Biophys Res Commun 2000; 277(5):507512.
Porter QN, Baldas J. Mass Spectrometry of Heterocyclic Compounds. New York:
Wiley Interscience, 1971.
162
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Kuhnle
Kingston DGI, Fales HM. Methane chemical ionization mass spectrometry of
avonoids. Tetrahedron 1973; 29:40834086.
Kingston DGI. Mass spectrometry of organic compounds. VI. Electron-impact
spectra of avonoid compounds. Tetrahedron 1971; 27:26912700.
Pelter A, Stainton P, Barber M. The mass spectra of oxygen heterocylces. II. The
mass spectra of some avonoids. J Heterocylc Chem 1965; 2:262271.
Grayer RJ. Flavonoids. Harborne JB, ed. In: Methods in Plant Biochemistry,
London: Academic Press, 1989:283323.
Marby TJ, Markham KR. Mass spectrometry of avonoids, In Harborne JB, Mabry
TJ, Mabry H, eds. The Flavonoids. New York: Academic Press, 1975:p 79126.
Yamashita M, Fenn JB. Electrospray ion source: another variation of the free-jet
theme. J Phys Chem 1984; 88:44514459.
Aleksandrow ML, Gall LN, Krasnov NV, Nikolaev VI, Shkurow VA. Mass spectrometric analysis of thermally unstable compounds of low volatility by extraction of
ions from solution at atmospheric pressure. J Anal Chem USSR 1985; 40:12271236.
Cuyckens F, Rozenberg R, de Homan E, Clacys M. Structure characterization of
avonoid O-diglycosides by positive and negative nano-electrospray ionization ion
trap mass spectrometry. J Mass Spectrom 2001; 36(11):12031210.
McDonald S, Prenzler PD, Antolovich M, Robards K. Phenolic content and
antioxidant activity in olive extracts. Food Chem 2001; 73:7384.
Kebarle P, Tang L. From ions in solution to ions in the gas phase. Anal Chem 1993;
65(22):972A986A.
Thurman EM, Ferrer I, Barcelo D. Choosing between atmospheric pressure
chemical ionization and electrospray interfaces for the HPLC/MS analysis of
pesticides. Anal Chem 2001; 73(22):54415449.
Ma Y-L, Li QM, Heuvel HVD, Claeys M. Characterization of avone and avonol
aglycones by collision-induced dissociation tandem mass spectrometry. Rapid
Commun Mass Spectrom 1997; 11:13571364.
Domon B, Costello C. Glycoconjugate. 1988; 5:397.
Fabre N, Rustan I, Homann ED, Quetin-Leclercq J. Determination of avone,
avonol, and avanone aglycones by negative ion liquid chromatography electrospray ion trap mass spectrometry. J Am Soc Mass Spectrom 2001; 12(6):707715.
Ma Y-L, Vedernikowa I, Heuvel HVD, Clayes M. Internal glucose residue loss in
protonated O-diglycosyl avonoids upon low-energy collision-induced dissociation.
J Am Soc Mass Spectrom 2000; 11(2):136144.
Wittig J, Hederich M, Graefe EU, Veit M. Identication of quercetin glucuronides
in human plasma by high-performance liquid chromatography-tandem mass
spectrometry. J Chromatogr B, 2001; 753:237243.
Sattereld M, Brodbelt JS. Structural characterization of avonoid glycosides by
collisionally activated dissociation of metal complexes. J Am Soc Mass Spectrom
2001; 12(5):537549.
Sattereld M, Brodbelt JS. Enhanced detection of avonoids by metal complexation
and electrospray ionization mass spectrometry. Anal Chem 2000; 72(24):5898
5906.
Mass Spectrometry
21.
22.
23.
163
Bu H-Z, Poglod M, Micetich RG, Khan JK. Novel sample preparation method
facilitating identication of urinary drug metabolites by liquid chromatographytandem mass spectrometry. J Chromatogr B 2000; 738:259265.
Zeeb DJ, Nelson BC, Albert K, Dalluge JJ. Separation and identication of twelve
catechins in tea using liquid chromatography/atmospheric pressure chemical
ionization-mass spectrometry. Anal Chem 2000; 72(20):50205026.
Griths WJ, Jonsson AP, Liu S, Rai DK, Wang Y. Electrospray and tandem mass
spectrometry in biochemistry. Biochem J 2001; 355:545561.
7
Effects of Flavonoids on the
Oxidation of Low-Density
Lipoprotein and Atherosclerosis
Michael Aviram and Bianca Fuhrman
Rambam Medical Center
Haifa, Israel
I.
INTRODUCTION
Atherosclerosis is the leading cause of morbidity and mortality among people

with a Western life-style. The early atherosclerotic lesion is characterized by the
accumulation of arterial foam cells derived mainly from cholesterol-loaded
macrophages [1,2]. Most of the accumulated cholesterol in foam cells originates
from plasma low-density lipoprotein (LDL), which is internalized into the cells
via the LDL receptor. Native LDL, however, does not induce cellular cholesterol accumulation, because the LDL receptor activity is downregulated by the
cellular cholesterol content [3,4]. LDL has to undergo oxidative modication in
order to be taken up by macrophages at an enhanced rate via the macrophage
scavenger receptor pathway, which, unlike the LDL receptor, are not subjected
to downregulation by cellular cholesterol [57]. The oxidative modication
hypothesis of atherosclerosis proposes that LDL oxidation play a pivotal role in
early atherogenesis [816]. This hypothesis is supported by evidence that LDL
oxidation occurs in vivo [13,17] and contributes to the clinical manifestation of
atherosclerosis. The uptake of oxidized LDL (Ox-LDL) via scavenger receptors
promotes cholesterol accumulation and foam cell formation [5,7,16,18]. In addition, Ox-LDL atherogenicity is related to recruitment of monocytes into the
intima [19], to stimulation of monocyte adhesion to the endothelium [20], and to
cytotoxicity to arterial cells [21,22].
165
166
Aviram and Fuhrman
II.
MECHANISMS OF LDL OXIDATION
A.
Oxidation of LDL in a Cell-Free System
Oxidation of LDL involves free radical attack on the lipoprotein components,

including cholesterol, phospholipids, fatty acids, and apolipoprotein B-100.
LDL oxidation results, rst, in the consumption of its antioxidants (mainly
vitamin E and carotenoids) then in substantial loss of polyunsaturated fatty
acids and of cholesterol, which is converted to oxysterols. A predominant
oxysterol formed at early stages of oxidation is 7-hydroperoxycholesterol, and
at later stages, 7-ketocholesterol is formed [23]. Both of these oxysterols are
formed as a result of an oxygenation at the 7-position. The polyunsaturated
groups of the esteried cholesteryl esters and of phospholipids are also major
targets for oxidation. The primary products formed are hydroperoxides, which
can undergo subsequent reduction to hydroxides and aldehydes. Nonenzymatic
peroxidation of arachidonic acid results in the formation of isoprostanes and
epoxyisoprostanes [24]. In the presence of transition metal ions, acyl hydroperoxides also undergo carbon-carbon bond cleavage to form reactive shortchain aldehydes.
During oxidation of LDL, apolipoprotein B also undergoes direct and
indirect modications. Direct attack of oxidants can oxidize amino acid side
chains and fragment the polypeptide backbone. Reactive lipid peroxidation
products, such as short-chain aldehydes, can form stable adducts with amino
acid residues in the apolipoprotein B-100 [25], which can then lead to
intermolecular cross-linking and to aggregation of lipoprotein particles [26].
Enrichment of LDL with lipid hydroperoxides appears to be an important rst
step in LDL oxidation. After depletion of LDL antioxidants, transition metal
ions catalyze propagation reactions, breakdown of lipid hydroperoxides, and
formation of reactive products, such as malondialdehyde and hydroxynonenal,
which are responsible for apolipoprotein B-100 modication. All these reactions
result in changes in the LDL structure, and the oxidatively modied LDL, which
can no longer bind to the LDL receptor, interacts with the macrophage
scavenger receptors, leading to the accumulation of cholesterol and oxidized
lipids and to foam cell formation.
B.
Macrophage-Mediated Oxidation of LDL
The process of LDL oxidation is unlikely to occur in plasma because plasma

contains high concentrations of antioxidants and of metal ion chelators. It is
more likely to occur within the artery wall, an environment, that is depleted
of antioxidants and therefore where the LDL is exposed to oxidative stress.
The identity of the cells responsible for the oxidation of LDL along atherogenesis in the arterial wall is uncertain. Monocyte-derived macrophages are
Oxidation of LDL and Atherosclerosis
167
likely candidates to induce the oxidation of LDL during early atherogenesis,

because they are prominent in arterial lesions and because they generate
reactive oxygen and nitrogen [27,28]. Macrophage-mediated oxidation of LDL
is considerably aected by the oxidative state in the cells, which depends on
the balance between cellular oxygenases and macrophage-associated antioxidants [29]. Macrophage binding of LDL to the LDL receptor initiates the
activation of cellular oxygenases [30,31]. LDL oxidation by arterial wall cells
was suggested to involve the activation of macrophage 15-lipoxygenase and
of nicotinamide-adenine dinucleotide phosphate (NADPH) oxidase [31,32].
When NADPH oxidase is activated, the cytosolic components of the NADPH
oxidase complex, P-47 and P-67, translocate to the plasma membrane, where
they form, together with the membrane-bound cytochrome b558, the active
NADPH oxidase complex. Both phospholipase A2, and phospholipase D can
induce macrophage NADPH oxidasedependent oxidation of LDL [33]. On
the other hand, macrophage antioxidants also contribute to the extent of cellmediated oxidation of LDL. Cellular reduced glutathione (GSH) is a most
potent antioxidant [34,35], and an inverse relationship was shown between the
extent of macrophage-mediated oxidation of LDL and the cellular reduced
glutathione content [35]. Macrophage-mediated oxidation of LDL can also
result from an initial cellular lipid peroxidation. When cultured macrophages
were exposed to ferrous ions, cellular lipid peroxidation took place [36,37].
These oxidized macrophages could easily oxidize the LDL lipids, even in
the absence of additional transition metal ions. LDL oxidation by oxidized
macrophages can also result from the transfer of peroxidized lipids from the
cell membranes to the LDL particle.
C.
Paraoxonase and LDL Oxidation
Human serum paraoxonase (PON 1) is an esterase that is physically

associated with high-density lipoprotein (HDL) and is also distributed in
tissues such as liver, kidney, and intestine [38,39]. Activities of PON 1,
which are routinely measured, include hydrolysis of organophosphates, such
as paraoxon (the active metabolite of the insecticide parathion); hydrolysis of
arylesters, such as phenyl acetate; and lactonase activities. Human serum
paraoxonase activity has been shown to be inversely related to the risk of
cardiovascular disease [40,41], as shown in atherosclerotic, hypercholesterolemic, and diabetic patients [4244]. In 1998 HDL-associated PON 1 was
shown to protect LDL, as well as the HDL particle itself, against oxidation
induced by either copper ions or free radical generators [45,46], and this
eect could be related to the hydrolysis of the specic lipoproteins oxidized
lipids such as cholesteryl linoleate hydroperoxides and oxidized phospholipids. Protection of HDL from oxidation by PON 1 was shown to preserve
168
Aviram and Fuhrman
the antiatherogenic eect of HDL in reverse cholesterol transport, as shown

by its benecial eect on HDL-mediated macrophage cholesterol eux [45].
These eects of PON 1 may be relevant to its benecial properties against
cardiovascular disease [40,41]. Antioxidants were shown to preserve PON 1
activity as they decrease the formation of lipid peroxides that can inactivate
PON 1 [47].
III.
FLAVONOIDS INHIBIT LDL OXIDATION
Clinical studies investigated the antioxidative eects of antioxidant supplementation of humans on ex vivo LDL oxidation [4852]. We have shown that
dietary supplementation of h-carotene of healthy subjects resulted in a moderate
inhibitory eect on the susceptibility of LDL to oxidative modication [5355]
in some, but not in all studied subjects. The combination of carotenoids with
vitamin E, in contrast, demonstrated a synergistic inhibitory eect on LDL
oxidation in all studied cases [56]. We showed that supplementation of vitamin E of atherosclerotic apolipoprotein Edecient mice (25 Ag/mouse/day for
3 months) inhibited LDL oxidation by 40% and the atherosclerotic lesion area
by 35% [57]. In humans, unlike in animal models, both vitamin E and carotenoids did not signicantly reduce atherosclerosis in primary prevention trials
[58]. This result may be related to insucient absorption, insucient potency,
and inappropriate tissue distribution.
Flavonoids are more potent antioxidants than carotenoids and vitamin E,
and mortality rate can be attributed to dierences in avonoid intake of
coronary heart disease across populations. Dietary consumption of avonoids
was shown to be inversely related to morbidity and mortality of coronary
heart disease [59]. Flavonoids constitute the largest and most studied group of
plant phenols. Over 4000 dierent avonoids have been identied to date.
They are usually found in plants as glycosides, and large compositional
dierences exist among dierent types of plants, even among dierent parts
of the same plant. Flavonoids are grouped into anthocyanins and anthoxantins
(Fig. 1). Anthocyanins are glycosides of anthocyanidin, and they are the most
important group of water-soluble plant pigments, responsible for the red, blue,
and purple colors of owers and fruits. Anthoxantins are colorless or colored
white to yellow; they include avonols, avanols, avones, avans, isoavones, and isoavans.
Flavonoids are powerful antioxidants, and their activity is related to their
chemical structures [6062]. Plant avonoids are multifunctional and can act as
reducing agents, as hydrogen atomdonating antioxidants, and as singlet oxygen quenchers. Some avonoids also act as antioxidants via their metal ion
chelation properties [63], thereby reducing the metals capacity to generate
169
Figure 1 General classification of flavonoids.
free radicals. Flavonoids can act as potent inhibitors of LDL oxidation, via
several mechanisms:
1.
2.
3.
4.
5.
6.
Scavenging of free radicals

Protection of the LDL-associated antioxidants a-tocopherol (vitamin
E) and carotenoids from oxidation
Regeneration of vitamin E from oxidized a-tocopherol
Chelation of transition metal ions
Protection of cells against oxidative damage, with result, inhibition of
cell-mediated oxidation of LDL, achieved via the potency of
avonoids to inhibit xanthine oxidase [64], NADPH oxidase [65], or
lipoxygenase [6668]
Preservation of serum paraoxonase (PON 1) activity, and as a result,
hydrolysis of LDL-associated lipid peroxides
The protection of LDL against copper ion or free radicalinduced oxidation

by avonoids depends on their structural properties in terms of their response
to copper ion, their partitioning between the aqueous and the lipophilic compartments within the LDL particle, and their hydrogen donating antioxidant
properties [63].
The avanol catechin prevented plasma lipid peroxidation that was induced
by azo compounds such as the water-soluble 2,2V-azobis, 2-amidinopropane
hydrochloride (AAPH) or the lipid-soluble 2,2V-azobis, 2,4-dimethylvaleronitrile
170
Aviram and Fuhrman
(AMVN). This antioxidant eect of catechin depends on its plasma concentration,

on the incubation time, and on the physical localization of the generated radicals.
As expected from its hydrophilic structure, however, catechin showed a higher
antioxidant capacity when the free radical reactions were initiated in the aqueous,
rather than the lipid phase [69]. Catechin was also shown to inhibit LDL oxidation
induced by copper ions, by cultured macrophages, or by vascular endothelial cells
[70]. Quercetin, rutin, luteolin, and kaempferol also inhibited copper ioninduced
LDL oxidation [63]; quercetin, rutin, and luteolin were more eective inhibitors of
copper ioninduced LDL oxidation than kaempferol, as they also chelate copper
ions. Morin, setin, quercetin, and gossypetin inhibited copper ioninduced LDL
oxidation and macrophage-mediated LDL oxidation, with an IC50 of 12 AM [63].
Other avonoids that were shown to inhibit LDL oxidation include the hydroxycinnamic acidderived phenolic acid caeic, ferulic, and p-coumaric acids [71]
and the isoavan glabridin [72,73]. Among the dierent groups of avonoids,
avonols, avanols, and the isoavans are the most potent protectors of LDL
against copper ioninduced oxidation. However, although possessing a similar OH
group arrangement, the avonol quercetin was a more potent antioxidant than the
avanol catechin, because of the 2-3 double bond and the 4-oxo structure present in
the quercetin ring C. Similarly, studies on structural aspects of the inhibitory eect
of glabridin on LDL oxidation revealed that the antioxidant eect of glabridin on
LDL oxidation resides mainly in the 2V-hydroxyl group of the isoavan B-ring
[73]. The hydrophobic moiety of the isoavan was also essential in order to obtain
the inhibitory eect of glabridin on LDL oxidation, and the position of the
hydroxyl groups at the B-ring signicantly aected the ability of glabridin to
inhibit LDL oxidation [73]. Flavonoids are also quite suitable for protecting cell
membranes from free radicalinduced oxidation, since they are both lipophilic and
hydrophilic, thus resulting in reduced cell-mediated oxidation of LDL. Being
partly inside and partly outside the cells plasma membrane, avonoids can
scavenge free radicals that are generated within the cells as well as free radicals
that attack the cell from the outside. Indeed, catechins from tea were shown to
protect erythrocyte membranes and rat liver microsomes from lipid peroxidation
[74]. Pretreatment of cells with avanols or avonols also protected the cells
against damage induced by reactive oxygen species [75]. We have also demonstrated that enrichment of macrophages with the isoavan glabridin protected the
cells from lipid peroxidation under oxidative stress [65].
IV.
PROTECTION OF LDL AGAINST OXIDATION

BY FLAVONOID-RICH NUTRIENTS
The average daily human intake of avonoids varies from as low as 25 mg to as

high as 1 g [7680]. After oral ingestion, some of the ingested avonoids are
171
absorbed from the gastrointestinal tract, and some of the absorbed avonoids are
metabolized by the gastrointestinal microora. Dierences in the bioavailability
of dierent avonoids exist and may be related to chemical structure dierences
[76,77]. The bioavailability and metabolic modications of avonoids determine the antioxidative capacity of these potent antioxidants in vivo. Major
dietary sources of avonoids and their chemical structures, along with their
eects on LDL oxidation and on atherosclerosis, are discussed in the sections
that follow (Fig. 2).
A.
Wine
Wine has been part of the human culture for over 6000 years, serving dietary and
socioreligious functions. Epidemiological studies of numerous populations reveal
a signicant lower cardiovascular mortality rate in individuals who have the habit
of daily moderate wine consumption [81,82]. The French paradox, i.e., the low
incidence of cardiovascular events in spite of diet high in saturated fat, was
attributed to the regular drinking of red wine in southern France [83]. The
benecial eect of red wine consumption against the development of atherosclerosis was attributed in part to its alcohol, but mostly to the antioxidant activity of
its polyphenols. Red wine contains a range of polyphenols derived from the skin
of the grape, with important biological activities [84,85]. Red wine contains the
avonols quercetin and myricetin (1020 mg/L), the avanols catechin and
epi(gallo)catechin (up to 270 mg/L), gallic acid (95 mg/L), condensed tannins
[catechin and epicatechin polymers (2500 mg/L)], and polymeric anthocyanidins.
Phenolic compounds in red wine are derived from the grapes skin, as well as from
the seeds, stems, and pulp, all of which are an important source of avanols that
are transferred to the wine during preparation together with the grape juice at the
rst stage of wine fermentation. On the contrary, white wines are usually made
from the free running juice, without the grape mash, and have no contact with the
grapes skin. This is thought to be the main reason for the relatively low
polyphenol content and the low antioxidant activity of white wine, in comparison
to those of red wine [8691]. In previous studies, red wine, which contains a much
higher concentration of polyphenols than white wine, was shown to be more
eective in inhibiting LDL oxidation [9295]. In 2001 we produced white wine
with red winelike antioxidant characteristics, by increasing the white wine
polyphenol content [96]. This was achieved by imposing grapeskin contact for
a short period in the presence of added alcohol, in order to augment the extraction
of grape skin polyphenols into the wine. We have analyzed the antioxidant
capacity of white wine samples obtained from whole squeezed grapes that were
stored for increasing periods before grape skin removal or from whole squeezed
grapes to which increasing concentrations of alcohol were added. White wine
obtained from the whole squeezed grapes, which were incubated for 18 hours with
172
Aviram and Fuhrman
173
18% alcohol, contained a 60% higher concentration of polyphenols than untreated

white wine and exhibited signicant antioxidant capacity against LDL oxidation,
almost similar to that of red wine. The antioxidant capacity of the white wine was
directly proportional to its polyphenol content (Fig. 3). Ingestion of red wine was
previously shown to be associated with increased serum antioxidant activity [97].
Administration of 400 mL of red wine to healthy human volunteers for a period of
2 weeks reduced the propensity of the volunteers LDL to lipid peroxidation in
response to copper ions. On the contrary, the resistance to oxidation of LDL
derived from subjects who consumed the same volume of white wine showed no
signicant change, in comparison to baseline LDL oxidation rates [85,92,98]. The
administration of red wine to healthy human volunteers for a period of 2 weeks
resulted in a substantial prolongation of the lag phase required for the initiation of
LDL oxidation by as much as 130 minutes, whereas consumption of a similar
volume of white wine had no signicant eect on LDL oxidation. In parallel, the
propensity of the volunteer LDL, which was obtained after red wine consumption,
to copper ioninduced lipid peroxidation was reduced in comparison to that of
LDL obtained at baseline, as measured by a 72% decrement in the content of the
lipoprotein-associated lipid peroxides, whereas after white wine consumption no
signicant eect was observed [92]. The antioxidant eect of dietary red wine
against LDL oxidation could be related to the elevation in polyphenol concentrations in the plasma and in the LDL particle. Thus, some phenolic substances
that exist in red wine are absorbed, bind to plasma LDL, and protect the
lipoprotein from oxidation [92]. The eect of red wine consumption on the
susceptibility of LDL to oxidation was also studied in the postprandial state [99].
Five volunteers consumed 300 mL of California red wine, containing 1500 mg/L
of total phenolic compounds. LDL isolated from plasma samples taken 1 and 2 h
after red wine ingestion was signicantly more resistant to copper ioninduced
oxidation than LDL obtained before wine consumption, as shown by 50% and
66% inhibition in aldehyde formation, respectively. These results were further
conrmed in a study [100] that demonstrated that red wine consumption increased
concentration plasma- and LDL-associated polyphenols and protected LDL
against copper ioninduced oxidation, as shown by increased lag time and
decreased LDL content of lipid peroxides and thiobarbituric acid reactive
substances (TBARSs) (by 34% and 22%, respectively).
Figure 2 Antiatherogenic effects of dietary polyphenolic flavonoids. Dietary

consumption of nutrients rich in flavonoids inhibits LDL oxidation, foam cell
formation, and development of aortic atherosclerotic lesions. Major flavonoidrich nutrients are shown, along with the chemical structure of their flavonoids.
CE, cholesteryl ester; Ox-LDL, oxidized low-density lipoprotein; UC, unesterified cholesterol.
174
Aviram and Fuhrman
Figure 3 Alcohol augments grape skin polyphenol extraction into white wine and
increases its antioxidant capacity. Whole squeezed Muscat grapes were incubated
for 18 h with increasing concentrations of alcohol up to 18%, after which the juice
was separated from the grape skin and allowed to ferment into wine. (A)
Polyphenol concentration in wine samples was determined. (B) Wine samples at a
final concentration of 2 AL/mL were added to LDL (100 mg of protein/L) and
incubated with 5 Amol/L CuSO4 for 2 h at 37 jC. LDL oxidation was measured by
LDL TBARS assay. LDL, low-density lipoprotein; TBARS, thiobarbituric acid
reactive substance. (C) Linear regression analysis of the total polyphenol concentration of wine and the wine-induced inhibition of LDL oxidation.
175
The eect of the nonalcoholic components of red wine was also studied
[101,102]. By using wine and alcohol-free red wine extract, it was shown that
although the alcohol component of the wine may be important for a favorable
lipid pattern, such potential health benets may be independent of the proposed
antioxidant eects of red wine [92,100,103,104]. In a 2001 study it was shown
that polyphenols in dealcoholized red wine can reduce in vivo lipid peroxidation, as measured by F2-isoprostanes, in smoking subjects, whereas no reduction
in lipid peroxidation was observed after red or white wine consumption [102]. In
2001 human intervention study [102], it was shown that alcohol-free red wine
extract can inhibit LDL oxidation ex vivo. A short-term ingestion of purple
grape juice reduced LDL susceptibility to oxidation in patients with coronary
artery disease [105,106].
Variations in the concentration and composition of avonoids among red
wines [107] may be responsible for the range of antioxidant potential exhibited
by dierent red wines. We compared the composition of two red wines, which
both increased the resistance of LDL ex vivo to oxidative modication in human
supplementation studies [92,100]. Both studies used red wine from Cabernet
Sauvignon cultivars, one grown in Israel [92] and the other in France [100].
After similar 400-mL daily consumption of red wine for 2 weeks, the inhibitory
eect on LDL oxidation was found to be much higher in the Israeli study [92].
Comparison of the polyphenol composition of both wines revealed that even
though the total polyphenol content of the wines was similar [1650 (Israeli) and
1800 (French) mg/L], the wines diered substantially in their avonol and
monomeric anthocyanin content. There is a wide variation in the avonol
content of dierent red wines throughout the world [108], and a major
determinant factor for this phenomenon is related to the amount of sunlight to
which the grapes are exposed during cultivation [109]. The avonol synthesis in
the skin of the grape is increased in response to sunlight so as to act as a yellow
lter against the harmful eect of ultraviolet (UV) light. Thus, the climatic
conditions under which grapes are grown may explain the increased content of
avonols in the Israeli red wines in comparison to that in the French red wine,
studied by Nigdikar and associates [100]. Another comparison of two studies
[103,110], both of which used alcohol-free red wine extracts, showed an
increase in the resistance of LDL to oxidation in only one study [103]. A
comparison of the wine composition showed that the concentrations of catechins
and anthocyanins were double in the wine that showed an ex vivo inhibitory
eect on LDL oxidation [103]. The direct eect of red wine consumption on the
development of atherosclerotic lesions was further studied in Eo mice that were
supplemented with 0.5 mL of red wine/day per mouse for a period of 6 weeks
[111,112]. LDL isolated after red wine consumption was less susceptible (by
3080%) to oxidation induced either by copper ions, by the free radical
generator AAPH, or by J-774 A.1 macrophages in culture, in comparison to
176
Aviram and Fuhrman
LDL isolated from the placebo (alcoholized water)-treated Eo mice [111,112].

The atherosclerotic lesion areas in Eo mice that consumed red wine were
signicantly reduced (by 40%), in comparison to the lesion areas in the
placebo-treated Eo mice (Fig. 4). In contrast to wine supplementation of young
Eo mice, in which development of atherosclerosis was substantially reduced,
administration of red wine to old, already atherosclerotic Eo mice, for up to 26
weeks, did not signicantly reduce the mature atherosclerosis development
[113]. In another study, administration of red wine, dealcoholized red wine, or
grape juice was shown to inhibit atherosclerosis development in hamsters.
However, grape juice was calculated to be much more eective than red wine
or dealcoholized red wine, at the same polyphenol dosage, in inhibiting
atherosclerosis and improving plasma lipid pattern and antioxidant activity
[114]. Phenolic substances in red wine were shown to inhibit LDL oxidation
in vitro [95]. In previous studies, red winederived phenolic acids [115, 116],
resveratrol [117], avonols (quercetin, myricetin) [68,118,119], catechins
[66,120], and the grape extract itself [121,122] have been shown to possess
antioxidant properties. The nding that ethanol and wine stripped of phenols did
not aect LDL oxidation further conrmed that the active antioxidant components in red wine are phenolic compounds [123]. Red wine fractionation
revealed major antioxidative potency to monomeric catechins, procyanidins,
monomeric anthocyanidins, and phenolic acids [123]. The avonol quercetin
and the avonol catechin were both tested for antioxidative and antiatherogenic
eects in the atherosclerotic Eo mice [111]. Eo mice at the age of 4 weeks were
supplemented for up to 6 weeks in their drinking water with placebo (1.1%
alcohol) or with catechin or quercetin (50 Ag/day/mouse). The atherosclerotic
lesion area was smaller by 39% or by 46%, respectively, in the treated mice than
in Eo mice that were treated with placebo (Fig. 4AE).
These results were associated with reduced susceptibility to oxidation (that
was induced by dierent modes such as copper ions, free radical generators, or
macrophages) of LDL isolated after quercetin, and to a lesser extent after
catechin consumption, in comparison to LDL isolated from the placebo group.
LDL isolated from Eo mice that consumed catechin or quercetin for 2 weeks was
also found to be less oxidized in its basal, not induced, state, in comparison to
LDL isolated from Eo mice that received placebo, as evidenced by 39% or 48%
reduced content of LDL-associated lipid peroxides, respectively (Fig. 4F) [111].
The inhibitory eect of moderate red wine ingestion against LDL oxidation may
also be explained in part by its eects on HDL-associated paraoxonase [124].
Daily moderate alcohol consumption was shown to increase serum paraoxonase
activity in middle-aged men [125]. Catechin, quercetin, and red wine consumption also increased serum paroxonase activity by 14%, 13%, and 75%,
respectively, in Eo mice (Fig. 4G) [111]. As paraoxonase protects against lipid
peroxidation (probably by hydrolyzing oxidized lipids) the benecial eect of
177
Figure 4 Consumption of red wine or its polyphenols catechin and quercetin by

Ej mice decreases the development of atherosclerotic lesions, inhibits oxidation of
LDL, and increases serum paraoxonase activity. (AE) The aortic arch from Ej
mice that consumed placebo, catechin, quercetin, or red wine was analyzed.
Results are expressed as the mean of the lesion area in square micrometers F SD.
*p < 0.05 (vs. placebo). (F) LDL was isolated from Ej mice that consumed placebo,
red wine (0.5 mL / day per mouse), catechin or quercetin (50 Ag/day per mouse) for
a period of 6 weeks. LDL (100 Ag of protein/mL) oxidative state (basal, not induced
oxidation) was determined as lipid peroxide levels. Results are expressed as mean
F S.D. (n = 3), *p < 0.01 (vs. placebo). (G) Paraoxanase activity was measured as
arylesterase activity in serum derived from Ej mice that consumed placebo,
catechin, quercetin, or red wine for 6 weeks. Results are expressed as mean F
S.D. of three separate determinations. *p < 0.01 vs. placebo. LDL, low-density
lipoprotein.
178
Aviram and Fuhrman
red wine polyphenols on paraoxonase activity can be considered an additional

antiatherogenic property of red wine.
B.
Licorice
Glycyrrhiza glabra, the licorice plant, has a history of consumption of more than
3000 years. The licorice root has long been used as a avoring and sweetening
agent. Licorice root has also been used medicinally for a wide range of therapeutic
functions, such as antibacterial, antiviral, anti-inammatory, antiallergic, and
antihepatotoxic functions. Minor components of licorice, mostly avonoids from
the isoavan and chalcon subclasses, were shown to possess antioxidative
properties. The antioxidative capability of licorice crude extract against LDL
oxidation was investigated in vitro and ex vivo [126]. LDL oxidation induced by
copper ions or by AAPH was inhibited by 90% with as little as 0.3 Ag of licorice
root extract/mL. Licorice ethanolic extract inhibited LDL oxidation by a mechanism that involves scavenging of free radicals. The protective eect of licorice
root extract on the resistance of LDL to ex vivo oxidation was studied in
normolipidemic humans, as well as in hypercholesterolemic patients and in
atherosclerotic apolipoprotein Edecient (Eo) mice. LDL, which was isolated
from the plasma of 10 healthy volunteers after consumption of 100 mg of licorice
root ethanolic extract per day for a period of 2 weeks, was more resistant to copper
ioninduced oxidation, as well as to AAPH-induced oxidation, by 44% and by
36%, respectively, in comparison to LDL isolated before licorice supplementation.
Supplementation of licorice root extract (0.1 g/day) to hypercholesterolemic
patients for a period of 1 month was followed by an additional 1 month of
placebo consumption [127]. Licorice consumption resulted in a moderate reduction in the patients plasma susceptibility to oxidation (by 19%) and in an
increased resistance of the patients plasma LDL to oxidation (by 55%). After
an additional 1 month of placebo consumption, a reversal of the parameters
studied to baseline levels was noted. Licorice extract supplementation resulted
also in a 10% reduction in the patients systolic blood pressure, which was
sustained for an additional 1 month during the placebo consumption. Thus, dietary
consumption of licorice root extract by hypercholesterolemic patients may provide
a moderate hypocholesterolemic nutrient and a potent antioxidant agent, which
confers a health benet against cardiovascular disease. These eects were further
supported by the antioxidative eects of licorice extract in the atherosclerotic
apolipoprotein Edecient mice. Dietary supplementation of licorice (200 Ag/day/
mouse) to Eo mice for a period of 6 weeks resulted in a 80% reduction in the
susceptibility of their LDL to copper ioninduced oxidation in comparison to LDL
isolated from placebo-treated mice [126].
Licorice root contains avonoids with biological activities, several of
which were isolated and puried. Licochalcone B and D, isolated from the roots
179
of Glycyrrhiza inata, were shown to inhibit superoxide anion production in the

xanthine/xanthine oxidase system [128] and to possess free radical scavenging
activity toward the 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical. These phenolic compounds were also shown to be eective in protecting biological
systems against various oxidative processes. They inhibited mitochondrial lipid
peroxidation induced by Fe (III)adenosine diphosphate/NADH (ADP/NADH),
scavenged superoxide anions in microsomes, and protected red blood cells
against oxidative hemolysis [128]. Other antioxidant constituents that were
isolated from licorice were identied as the isoavans glabridin, hispaglabridin
A, hispaglabridin B, and 4-O-methyl glabridin and two chalcones, isoprenylchalcone and isolipuritegenin [72]. Among these compounds, glabridin constituted the major avonoid in the licorice root extract (500 mg/kg of ethanolic
root extract). On LDL incubation with glabridin, the latter was shown to bind to
the LDL particles and subsequently to protect them from oxidation [73,129].
Glabridin inhibited AAPH-induced LDL oxidation in a dose-dependent manner,
as shown by the inhibition of cholesteryl linoleate hydroperoxide (CL-OOH)
formation, as well as by the inhibition of aldehyde and lipid peroxide formation.
Addition of glabridin (30 AM) to LDL that was incubated with AAPH or with
copper ions also inhibited the formation of oxysterols (7-hydroxycholesterol,
7-ketocholesterol, and 5, 5-epoxycholesterol) by 65%, 70%, and 45%, respectively. Glabridin inhibited the consumption of h-carotene and that of lycopene
by 41% and 50%, respectively, after 1 h of LDL oxidation in the presence of
AAPH but did not protect the major LDL-associated antioxidant, vitamin E,
from oxidation [73] (Fig. 5). Finally, glabridin preserved the arylesterase
activity of human serum paraoxonase (PON 1), including its ability to hydrolyze
Ox-LDL cholesteryl linoleate hydroperoxides [47]. Administration of glabridin
to Eo mice in their drinking water was followed by analysis of its antioxidative
eect against LDL oxidation ex vivo [126]. Analysis of LDL derived from Eo
mice after consumption of glabridin revealed that glabridin was absorbed and
bound to the LDL particle. Although no glabridin could be detected in LDL of
control mice, LDL of mice that consumed glabridin (20 Ag/day/mouse)
contained about 2 nmol of glabridin/mg LDL protein. LDL derived from Eo
mice after consumption of 20 Ag glabridin/day/mouse for a period of 6 weeks
was signicantly more resistant (by 22%) than LDL derived from placebotreated mice to copper ioninduced oxidation. Most importantly, inhibition
of oxidative modication of LDL in Eo mice after glabridin consumption
was associated with a substantial reduction in the development of atherosclerotic lesions.
Glabridin consumption was shown to exert its antioxidative eects also
at the cellular level. Enrichment of mouse peritoneal macrophages with
glabridin either in vitro or in vivo (after its consumption by Eo mice) resulted
in 80% inhibition in macrophage-mediated oxidation of LDL, in comparison
180
Aviram and Fuhrman
Figure 5 The antioxidative effect of glabridin on LDL endogenous constituents

during AAPH-induced LDL oxidation. LDL (100 mg of protein/L) was incubated
for 3 h at 37 jC with 5 mM AAPH in the presence of 30 Ag glabridin. The extent of LDL oxidation was measured as formation of thiobarbituric acid reactive substances (TBARSs), lipid peroxides, cholesteryl linoleate hydroperoxide
(CLOOH), 7-hydroxycholesterol, 7-ketocholesterol, 5,6-epoxycholesterol, or consumption (oxidation) of h-carotene, lycopene, and vitamin E. Results are
expressed as percentage of inhibition relative to LDL incubated with AAPH in
absence of glabridin. LDL, low-density lipoprotein; AAPH, 2,2V-azobis, 2-amidinopropane hydrochloride.
to that of control cells [65]. This eect was secondary to the inhibition of
the macrophage NADPH oxidase, as reected by the decrement in superoxide
anion release. This latter eect was related to an inhibition in the translocation of the cytosolic component of NADPH oxidase P-47 to the plasma
membrane. The eects of glabridin described were associated with the inhibition (by 70%) of macrophage protein kinase C activity, which is required
for P-47phosphorylation and activation. Thus, glabridin-induced inhibition
of P-47 phosphorylation may be the primary event in its inhibitory eect
on NADPH oxidaseinduced macrophage-mediated oxidation of LDL. All the
inhibitory eects of glabridin on the events related to cell-mediated oxidation
of LDL required the hydroxyl groups on the isoavan B-ring. Since glabridin
inhibited oxidative processes both in macrophages and in LDL, these
mechanisms may be responsible for the attenuation of atherosclerosis in Eo
mice that consumed glabridin.
C.
181
Pomegranate
The pomegranate tree, which is said to have ourished in the Garden of Eden,
has been extensively used as a folk medicine in many cultures [130]. Edible
parts of pomegranate fruits (about 50% of total fruit weight) comprise 80% juice
and 20% seeds. Fresh juice contains 85% moisture, 10% total sugars, and a total
of 1.5% for pectin, ascorbic acid, polyphenols, and avonoids. Pomegranate
seeds are a rich source of crude bers, pectin, and sugars, and the pomegranate
peel has been shown to contain phenols from the condensed and hydrolysable
tannin class [131133]. The dried pomegranate seeds contain the steroidal estrogen estrone [134,135], the isoavonic phytoestrogens genistein and daidzein,
and the phytoestrogen coumestrol [136]. Content of soluble polyphenols in
pomegranate juice varies within the limits of 0.2% to 1.0%, depending on variety, and includes mainly anthocyanins (such as cyanidin-3-glycoside, cyanidin-3,
3-diglycoside, and delphindin-3-glucosid) and anthoxantins (such as catechins,
ellagic tannins, and gallic and ellagic acids [131133]). Pomegranate fermented
juice and cold-pressed pomegranate seeds possess antioxidant activity and can
reduce prostaglandin and leukotriene formation by inhibition of cyclooxygenases and lipoxygenases, respectively [137]. Pomegranate juice was shown to
possess antioxidant activity that was three times higher than the antioxidant
activity of red wine or of green tea [133]. The antioxidant activity was higher in
commercial juices extracted from whole pomegranates than in juices obtained
from arils only, suggesting that industrial processing extracts some of the
hydrolyzable tannins present in the fruit rind. The eect of pomegranate juice
on LDL oxidation was studied in vitro and ex vivo in healthy male volunteers
and in the atherosclerotic apolipoprotein Edecient (Eo) mice [138]. The in
vitro studies demonstrated a signicant dose-dependent antioxidant capability
of pomegranate juice against LDL oxidation, as well as against oxidation of
HDL. The mechanisms for the antioxidative eects of pomegranate juice against
lipoprotein oxidation could be related to its capacity to scavenge free radicals.
The water-soluble fractions of pomegranates inner and outer peels, but not
that of the seeds, were even stronger antioxidants against LDL oxidation than
the juice [138].
LDL derived from human healthy volunteers after consumption of
pomegranate juice (Wonderful cultivation, 50 mL/day of pomegranate juice,
equivalent to 1.5 mmol total polyphenols/day), for a period of 2 weeks, was
found to be more resistant to copper ioninduced oxidation than LDL obtained
before pomegranate juice supplementation [138]. This eect was demonstrated
by a 43% prolongation of the LDL oxidation lag time after 2 weeks of juice
consumption, in comparison to that of LDL obtained before juice consumption,
and this latter eect was accompanied by a signicant 10% increment in plasma
total antioxidant status.
182
Aviram and Fuhrman
Pomegranate juice consumption by healthy humans also resulted in

increased activity of their serum paraoxonase. We have extended our studies
on the antioxidative eects of pomegranate juice consumption to hypertensive
patients [139,140]. The eect of pomegranate juice consumption for a period of
2 weeks by 10 hypertensive patients on their blood pressure was small but
signicant. In 7 of 10 hypertensive patients studied, serum angiotensin-converting enzyme (ACE) activity was signicantly decreased by 36%. Pomegranate
juice was shown to exhibit directly a dose-dependent inhibitory eect on serum
ACE activity. Because ACE inhibitors are metabolized by cytochrome P-450
enzymes, serum ACE activity can be signicantly aected by modulating P-450
enzyme activity [141]. Therefore, we next analyzed the eect of pomegranate
juice on cytochrome P-450 enzymes [139]. Pomegranate juice decreased the
activities of cytochrome P-450 3A4, 2D6, 2E1, and 2B6 by 40%, 30%, 20%, and
60%, respectively. In hypertensive patients treated with ACE inhibitors, the
inhibitory eect of pomegranate juice consumption on the P-450 enzymes can
possibly decrease P-450-mediated drug (the ACE inhibitor) breakdown, and
hence, serum ACE activity may be further decreased in these patients. We have
indeed observed in three hypertensive patients who were treated with the ACE
inhibitor fosinopril (20 mg/day for 1 month) that their serum ACE activity
decreased after 2 weeks of pomegranate juice consumption (50 mL of juice
containing 1.5 mmol total polyphenols/day) by 26%. Taken together, the results
on the inhibitory eect of pomegranate juice on serum ACE activity, on one
hand, and on cytochrome P-450 enzymes, on the other hand, suggest that
pomegranate juice may also aect ACE activity indirectly, secondary to its
inhibitory eect on the cytochrome P-450 enzymes.
Consumption of pomegranate juice by Eo mice has also been shown to have
considerable antioxidative and antiatherogenic eects [138]. Pomegranate juice
consumption substantially reduced the propensity of Eo micederived LDL to
copper ioninduced oxidation and to macrophage-mediated oxidation of LDL by
reducing the oxidative capacity of the cells. The mechanism responsible for this
eect was associated with inhibition of the NADPH oxidase cytosolic factor p-47
translocation to the macrophage plasma membrane and hence, inhibition of
NADPH oxidase activation, reduction (by 49%) in superoxide anion release from
the macrophages, and elevation in cellular glutathione content by 25% [138].
This eect could also be related to reduced levels of macrophage-associated lipid
peroxides after pomegranate juice consumption, in comparison to levels of
macrophages isolated from control mice that consumed placebo (Fig. 6). Most
importantly, pomegranate juice supplementation to Eo mice reduced the size of
their atherosclerotic lesion and the number of foam cells in the lesion [138], in
comparison to those of control placebo-treated Eo mice that were supplemented
with water. Furthermore, pomegranate juice supplementation to Eo mice with already advanced atherosclerosis was still able to reduce the mices atherosclerotic
183
Figure 6 Pomegranate juice consumption by E mice reduces macrophagemediated LDL oxidation mechanisms. Mouse peritoneal macrophages (MPMs)
were isolated from the peritoneal fluid of control Ej mice or from mice that
consumed 12.5 AL of pomegranate juice (PJ)/mouse/day, for a period of 2 months.
(A) Cell-mediated LDL oxidation. The MPMs were incubated for 6 h at 37 jC with
LDL (100 Ag protein/mL) under oxidative stress (in the presence of 2 AM of
CuSO4). LDL oxidation was measured directly in the medium by the TBARS
assay. Results are expressed as mean F S.D. (n = 3). *p < 0.01 (vs. placebo). (B)
MPM lipid peroxidation was determined as cellular lipid peroxides. (C) Superoxide
anion release: The amount of superoxide anion release from the MPM to the
medium in response to 50 ng/mL of PMA was determined. (D) Total glutathione
was determined in MPM sonicate supernatant with the 5,5-dithiobis-2-nitrobenzoic
acid-glutathione reductase (NADPH) recycling assay. (E) Results are expressed
as mean F S.D. (n = 3). *p < 0.01 (vs. control). LDL, low-density lipoprotein;
TBARS, thiobarbituric acid reactive substance; PMA,
184
Aviram and Fuhrman
lesion size by 17%, in comparison to the size of atherosclerotic lesion in agematched placebo-treated mice [142].
D.
Tea
Tea drinking has been associated in epidemiological studies with a decreased risk
for cardiovascular disease [59,143145]. The terms green tea and black tea refer
to products manufactured from the leaf of the tea plant, Camellia sinensis. Green
tea is manufactured from fresh leaf and is rich in avonoids, especially avonols
from the catechin group, of which epigallocatechin gallate, epicatechin gallate,
and epicatechin account for 3040% of the green tea solids [146]. Black tea
manufacture includes an enzymatic step, in which most catechins are converted to
complex condensation products, such as the aavins or the arubigens. Green and
black tea also contain small amount of avonols, such as quercetin.
Absorption studies of tea polyphenols and their eects on LDL oxidation, and atherosclerosis have shown conicting results [147,148]. The eect
of green or black tea consumption on the resistance of LDL to oxidation was
studied in 45 human volunteers who for a period of 4 weeks consumed 900 mL
(6 cups) of green tea or black tea per day in comparison to mineral water
[149]. Consumption of tea (green or black) had no eect on the ex vivo
resistance of LDL to oxidation. Similar negative results have also been
demonstrated in other studies [150153]. Conversely, in another study [154],
ingestion of tea (300 mL after an overnight fast) produced a signicant
increase of plasma antioxidant capacity, which peaked at 3050 min after
consumption. Similarly [155], ingestion of 400 mL of freshly prepared green
tea resulted in a rapid absorption of the tea polyphenols, and it was associated
with an increase in plasma total antioxidant state, peaking at 2040 min post
ingestion. Consumption of 750 mL of black tea/day for 4 weeks by 14 healthy
volunteers revealed [156] that the lag time for LDL oxidation was signicantly
prolonged (from 54 to 62 min). In other studies [157,158], ingestion of black
and green tea, in comparison to ingestion of alcohol-free red or white wine or
water, resulted in a signicant increase in plasma total antioxidant capacity at
30 min after consumption, and red wine and green tea were the most efcient
in protecting LDL from oxidation [157], whereas black tea had a mild acute
eect [158].
The eect of green or black tea on LDL oxidation and atherosclerotic
lesion formation was also studied in animal models, including hypercholesterolemic rabbits [159] and hamsters [160], and apolipoprotein Edecient mice
[161]. These studies indicated that green tea consumption reduced the atherosclerotic plaque formation in hypercholesterolemic rabbits [159], whereas black
tea showed no signicant eect, although both green and black tea induced a
13% and 15% prolongation in the lag phase of LDL oxidation, respectively
185
[159]. Similar results were demonstrated in the hamster model [160], showing
that green tea was signicantly more eective than black tea in improving risk
factors for heart disease, including hypolipidemic and antioxidant eects.
Furthermore, supplementation of green tea extract (0.8 g/L) to apolipoprotein
Edecient mice signicantly attenuated (by 23%) development of atherosclerotic lesions, without changing the plasma lipid level, probably through the
potent antioxidative activity of the tea [161].
The potent antioxidative eects of tea are attributed to its polyphenols.
Green tea extract and catechin-rich fractions from green tea were shown to
inhibit the oxidation of LDL by endothelial cells [162]. Ingestion of 300 mg of
green tea polyphenol extract twice daily for 1 week by 22 male volunteers
resulted in increased resistance of LDL to oxidation [163]. Catechins or
theaavins (25400 Amol/L) [164], as well as epicatechin, epigallocatechin,
epicatechin gallate, epigallocatechin gallate, and gallic acid [165,166], which
were added to LDL, dose-dependently inhibited its oxidation. Among the
catechins, epigallocatechin gallate exerted the most marked eect in prolonging
LDL oxidation lag time [163]. Furthermore, addition of 1.5 AM of epicatechin
and epigallocatechin to a mixture of LDL and copper ion in the initiation phase
inhibited LDL oxidation, whereas higher concentrations were needed (10 AM of
epicatechin and 2 AM of epigallocatechin) for the inhibition of the LDL
oxidation propagation phase [164]. The mechanisms responsible for the inhibition of LDL oxidation by tea include inhibition in the ability of macrophages
to modify LDL oxidatively by decreasing macrophage production of superoxide
and chelation of iron ions [167] as well as regeneration of vitamin E in human
LDL [168].
E.
Olive Oil
The Mediterranean diet, rich in fresh fruits and vegetables, was shown to be
inversely related to the incidence of cardiovascular disease, as shown in the
Seven Countries Studies [59,169171]. Olive oil, the dietary fat of choice in the
Mediterranean area, in comparison to other vegetable oils has a peculiar fatty acid
composition. The monounsaturated oleic acid (C-18:1, n = 9) is the most abundant fatty acid in olive oil (5684%), whereas the polyunsaturated linoleic acid
(C-18:2, n = 6) ranges only from 3% to 21%. In addition, olive oil contains a
variety of minor components, including polyphenols (up to 800 mg/kg), which
provide the typical taste and aroma of extra virgin olive oil and confer on this oil
its stability to oxidation [172].
The benecial eects of the Mediterranean diet may stem from the high
content of the monounsaturated oleic acid, as well as from the polyphenols,
which are benecial in reducing LDL oxidation. LDL isolated from Greek
subjects consuming a diet naturally rich in olive oil was signicantly less
186
Aviram and Fuhrman
susceptible (by 12%) to oxidation as measured by conjugated diene formation,

in comparison to LDL isolated from U.S. subjects consuming a typical
American diet [173]. Furthermore, the proinammatory potential of mildly
oxidized LDL derived from Greek subjects measured as LDL promotion of
monocyte chemotaxis and adhesion to endothelial cells, was decreased by 42%,
in comparison to LDL derived from U.S. subjects. There was an inverse
correlation between the LDL oleic acid content and the stimulation of monocyte
chemotaxis and adhesion [173]. Consumption of a liquid diet supplemented
with oleic acid for 8 weeks by U.S. subjects resulted in an oleic acidenriched
LDL, which subsequently promoted a 52% reduction in monocyte chemotaxis
and a 77% reduction in monocyte adhesion, in comparison to linoleate-enriched
LDL [173]. This study suggested that LDL enriched with oleic acid is less
easily converted into the proinammatory minimally modied LDL. These
results are consistent with our study [174], which demonstrated that dietary
supplementation of olive oil to healthy human subjects (50 g/day) for a period
of 2 weeks increased the resistance of their LDL to oxidation [174]. This was
shown by a signicant reduction in LDL peroxides, thiobarbituric acid reactive
substances (TBARSs), and conjugated diene content by 73%, 28%, and 32%,
respectively. Furthermore, LDL obtained after olive oil supplementation demonstrated a 61% decrease in cellular uptake by macrophages [174]. These
benecial eects of olive oil consumption may be related, in addition to the
oleic acid, to the olive oil polyphenols. To determine whether the minor polar
components of virgin olive oil could have favorable eects on LDL composition
and susceptibility to oxidation, 10 normolipidemic subjects received dierent
oils in a crossover study (two diet periods of 3 weeks each). Subjects received
either virgin olive oil or oleic acidrich sunower oil. LDL oxidation, which
was measured as conjugated diene production, decreased only after intake of the
virgin olive oil diet, suggesting a mild antioxidative eect against LDL
oxidation for some minor components of virgin olive oil [175]. This was further
evidenced by dietary administration of extra virgin olive oil versus rened olive
oil to New Zealand white rabbits (NZW) [176]. This experiment resulted in
increased resistance of the rabbits LDL to oxidation only in the group of
animals that consumed extra virgin olive oil [176], suggesting that the benecial
eect of extra virgin olive oil over that of the rened olive oil is due to the
presence of phenolics, which are lost during the rening process [176].
Furthermore, administration of extra virgin olive oil, which contains 800 mg/
kg of phenolic compounds, in comparison to rened olive oil, which contains
only 60 mg/kg of phenolic compounds, to patients with peripheral vascular
disease resulted in a superior antioxidative eect of the extra virgin olive oil
against LDL oxidation [177]. Consumption of olive oil together with a dietary
supplement of sh oil by patients with peripheral vascular disease also resulted
in protection of LDL from oxidation [178,179]. However, contradictory results
187
were also presented. Consumption of 18 mg/day of phenols from extra virgin

olive oil for 3 weeks did not aect LDL oxidation [180], suggesting that dietary
studies in humans may vary according to the studied populations and the
experimental conditions.
Studies performed in vitro also demonstrated the unique antioxidant
properties of virgin olive oil phenolics against LDL oxidation [181183]. These
eects could be related to the ability of olive oil phenolics, such as tyrosol, to
bind LDL in vitro and thus to protect LDL and other phenolic compounds
previously bound to LDL from oxidation [182,183].
Some major phenols were extracted and puried from olive oil [172, 184],
and their biological properties have been investigated. Oleuropein and hydroxytyrosol are two major representative phenols in olive oil. Direct antioxidant
activities of oleuropein and hydroxytyrosol against LDL oxidation were demonstrated [185,186]. LDL incubation with oleuropein or with hydroxytyrosol
resulted in the inhibition of copper ioninduced LDL oxidation. Oleuropein and
hydroxytyrosol protection of LDL against oxidation can be related to the
potency of these compounds in scavenging free radicals. Both hydroxytyrosol
and oleuropein were shown to express scavenging capacity for superoxide
radical in a cell-free system (based on the generation of urate and superoxide
by xanthine/xanthine oxidase) and in a cellular system (based on the production
of superoxide by phorbol myristate acetate (PMA)-challenged human neutrophils) [187]. Oleuropein was shown also to increase the resistance of LDL to
oxidation after dietary supplementation to rabbits [188], and tyrosol could
prevent oxidized LDL-induced cytotoxicity in Caco-2 cells when present during
incubation [189]. Two other phenolic compounds of olive oil, 3,4-dihydroxyphenylethanol-elenolic acid and protocatecuic acid, were shown to be able to
inhibit LDL lipid peroxidation and to reduce the extent of its cytotoxic activity
[190]. Biological active compounds with antioxidative properties against LDL
oxidation were also extracted from the olive mill waste waters [191].
Olive oil, as well as olive oil phenolic constituents, were shown to aect
biological processes involved in atherogenesis, as well as LDL oxidation.
Polyphenols in olive oil possess benecial eects on cellular processes, including inhibition of lipoxygenase [192], production of leukotriene B4 [193],
and stimulation of nitric oxide (NO) release from macrophages [194]. The
polyphenol-rich olive oil waste water was shown to depress cell production
of superoxide anion [195], thus resulting in reduced cellular oxidative stress,
which can subsequently reduce the capacity of the cells to modify LDL
oxidatively. In addition, olive oilenriched diet supplementation to mice
resulted in a lower level of macrophage scavenger receptor (SRA-I, SRA-II,
and CD36) gene expression, which can account for reduced macrophage uptake
of oxidized LDL, reduced foam cell formation, and reduced atherosclerotic
lesion development [196].
188
V.
Aviram and Fuhrman
CONCLUSIONS
The benecial health eects attributed to the consumption of fruits and

vegetables are related at least in part to their antioxidant activity. Of special
interest is the inverse relationship between intake of dietary nutrients rich in
avonoids and cardiovascular diseases. This eect is attributed to the avonoids capability to inhibit LDL oxidation, macrophage foam cell formation, and
atherosclerosis. Fig. 7 summarizes some of our studies of total polyphenol
concentration in several fruit juices or wines and the capacity of the juices, when
compared on a similar total polyphenol concentration basis, to inhibit LDL
oxidation. Pomegranate juice, red wine, and cranberry juice contained the
Figure 7 Analysis of the total polyphenol concentration of several common fruit

juices, white wine, and red wine (A), and juice-induced inhibition of LDL oxidation
(B) expressed as IC50, which is the concentration needed to inhibit LDL oxidation
by 50%. LDL, low-density lipoprotein.
189
highest polyphenol concentration (Fig. 7A), and they also showed the most
potent antioxidant activity against LDL oxidation (Fig. 7B). As shown in Fig.
7B, the various juices exhibit dierent antioxidant capacities, which may be
related to qualitative dierences in the types of avonoids present in the juices,
which possess dierent antioxidant capabilities. Thus both avonoid quantity
and quality determine the antioxidant potency of the juices (Fig. 7).
Our current view on the major pathways by which avonoids protect
LDL against oxidative modications, and thereby reduce macrophage foam
cell formation and development of advanced atherosclerosis, is summarized in
Fig. 8. Flavonoids can protect LDL against cell-mediated oxidation via two
pathways: direct interaction of the avonoids with the lipoprotein and avonoid
accumulation in arterial macrophages. Flavonoids were shown to accumulate in
macrophages and to inhibit the activation of cellular NADPH oxidase via the
Figure 8 Major pathways by which flavonoids inhibit macrophage-mediated

oxidation of LDL. Flavonoids (Fl) affect LDL directly by their interaction with the
lipoprotein and inhibition (Fl ) of LDL oxidation. Flavonoids can also protect LDL
indirectly, by their accumulation in arterial cells and protection of arterial
macrophages against oxidative stress. This latter effect is associated with
inhibition (Fl ) of the formation of oxidized macrophages and reduction in the
capacity of macrophages to oxidize LDL. In addition, flavonoids preserve/increase
(Fl +) paraoxonase activity, thereby increasing hydrolysis of lipid peroxides either
in the LDL or in oxidized macrophages, resulting in further protection of LDL from
oxidation. LDL, low-density lipoprotein.
190
Aviram and Fuhrman
inhibition of protein kinase C (PKC) activity. This eect results in reduced

capacity of the cells to modify LDL oxidatively. Enrichment of macrophages
with avonoids inhibits macrophage lipid peroxidation, formation of lipid
peroxiderich macrophages, and cell-mediated LDL oxidation.
Furthermore, avonoids increase serum paraoxonase activity, resulting in
hydrolysis of lipid peroxides either in oxidized LDL or in lipid peroxiderich
macrophages and hence further prevent the formation of oxidized LDL. All these
eects of avonoids were demonstrated in vitro, as well as in vivo in humans and
in the atherosclerotic apolipoprotein Edecient mice, after dietary supplementation of nutrients rich in diverse avonoids or of puried avonoids. Dietary
supplementation of pomegranate juice rich in avonoids to atherosclerotic mice
indeed resulted in a signicant inhibition in the development of atherosclerotic
lesions, along with the protection of LDL against oxidation [197200].
REFERENCES
1. Schaner T, Taylor K, Bartucci EJ, Fischer-Dzoga K, Beenson JH, Glagov S,
Wissler R. Arterial foam cells with distinctive immunomorphologic and
histochemical features of macrophages. Am J Pathol 1980; 100:5780.
2. Gerrity RG. The role of monocytes in atherogenesis. Am J Pathol 1981;
103:181190.
3. Goldstein JL, Brown MS. Regulation of the mevalonate pathway. Nature 1990;
343:425430.
4. Brown MS, Goldstein JL. A receptor-mediated pathway for cholesterol homeostasis. Science 1986; 232:3447.
5. Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond
cholesterol: modications of low-density lipoprotein that increase its atherogenicity. N Engl J Med 1989; 320:915924.
6. Aviram M. Modied forms of low density lipoprotein and atherosclerosis.
Atherosclerosis 1993; 98:19.
7. Aviram M. Beyond cholesterol: modications of lipoproteins and increased
atherogenicity. In: Neri Serneri GG, Gensini GF, Abbate R, Prisco D, eds.
Atherosclerosis Inammation and Thrombosis. Florence, Italy: Scientic Press,
1993:1536.
8. Jialal I, Devaraj S. The role of oxidized low density lipoprotein in atherogenesis.
J Nutr 1996; 126:1053S1057S.
9. Steinberg D. Low density lipoprotein oxidation and its pathobiological
signicance. J Biol Chem 1997; 272:2096320966.
10. Berliner JA, Heinecke JW. The role of oxidized lipoproteins in atherosclerosis.
Free Radic Biol Med 1996; 20:707727.
11. Aviram M. Oxidative modication of low density lipoprotein and atherosclerosis.
Isr J Med Sci 1995; 31:241249.

12.
191
Witztum JL, Steinberg D. Role of oxidized low density lipoprotein in

atherogenesis. J Clin Invest 1991; 88:17851792.
13. Aviram M. Interaction of oxidized low density lipoprotein with macrophages in
atherosclerosis and the antiatherogenicity of antioxidants. Eur J Clin Chem Clin
Biochem 1996; 34:599608.
14. Kaplan M, Aviram M. Oxidized low density during lipoprotein: atherogenic and
proinammatory characteristics during macrophage foam cell formation: an inhibitory role for nutritional antioxidants and serum paraoxonase. Clin Chem Lab Med
1999; 37:777787.
15. Parthasarathy S, Santanam N, Auge N. Oxidized low-density lipoprotein, a twofaced janus in coronary artery disease? Biochem Pharmacol 1998; 56:279284.
16. Parthasarathy S, Rankin SM. The role of oxidized LDL in atherogenesis. Prog
Lipid Res 1992; 31:127143.
17. Herttuala SY. Is oxidized low density lipoprotein present in vivo? Curr Opin
Lipidol 1998; 9:337344.
18. Aviram M. The contribution of the macrophage receptor for oxidized LDL to its
cellular uptake. Biochem Biophys Res Commun 1991; 179:359365.
19. Kim JA, Territo MC, Wayner E, Carlos TM, Parhami F, Smith CW, Haberland
ME, Fogelman AM, Berliner JA. Partial characterization of leukocyte binding
molecules on endothelial cells induced by minimally oxidized LDL. Arterioscler
Thromb Vasc Biol 1994; 14:427433.
20. Khan NBV, Parthasarathy S, Alexander RW. Modied LDL and its constituents
augment cytokine-activated vascular cell adhesion molecule-1 gene expression in
human vascular endothelial cells. J Clin Invest 1995; 95:12621270.
21. Rangaswamy S, Penn MS, Saidel GM, Chisolm GM. Exogenous oxidized low
density lipoprotein injures and alters the barrier function of endothelium in rats in
vivo. Circ Res 1997; 80:3744.
22. Penn MS, Chisolm GM. Oxidized lipoproteins, altered cell function and
atherosclerosis. Atherosclerosis 1994; 108:S21S29.
23. Brown AJ, Leong SL, Dean RT, Jessup W. 7-hydroxycholesterol and its products
in oxidized low density lipoprotein and human atherosclerotic plaque. J Lipid Res
1997; 38:17301745.
24. Lynch SM, Morrow JD, Roberts II LJ, Frei B. Formation of non-cyclooxygenasederived prostanoids (F2-isoprostanes) in plasma and low density lipoprotein
exposed to oxidative stress in vitro. J Clin Invest 1994; 93:9981004.
25. Slatter DA, Paul RG, Murray M, Bailey AJ. Reactions of lipid derived
malondialdehyde with collagen. J Biol Chem 1999; 274:1966119669.
26. Jessup W, Mander EL, Dean RT. The intracellular storage and turnover of
apolipoprotein B of oxidized LDL in macrophages. Biochim Biophys Acta 1992;
1126:167177.
27. Chisolm GM, Hazen ST, Fox PL, Catchard MK. The oxidation of lipoproteins by
monocyte-macrophages. J Biol Chem 1999; 274:2595925962.
28. Parthasarathy S, Printz DJ, Boyd D, Joy L, Steinberg D. Macrophage oxidation of
low density lipoprotein generates a modied form recognized by the scavenger
receptor. Arteriosclerosis 1986; 6:505510.
192
Aviram and Fuhrman
29.
Aviram M, Fuhrman B. LDL oxidation by arterial wall macrophages depends on

the antioxidative status in the lipoprotein and in the cells: role of prooxidants vs.
antioxidants. Mol Cell Biochem 1998; 188:149159.
Aviram M, Rosenblat M. Macrophage mediated oxidation of extracellular low
density lipoprotein requires an initial binding of the lipoprotein to its receptor.
J Lipid Res 1994; 35:385398.
Aviram M, Rosenblat M, Etzioni A, Levy R. Activation of NADPH oxidase is
required for macrophage-mediated oxidation of low density lipoprotein.
Metabolism 1996; 45:10691079.
Herttuala YS, Rosenfeld ME, Parthasarathy S, Glass CK, Sigal E, Witztum JL,
Steinberg D. Colocalization of 15-lipoxygenase mRNA and protein with epitopes
of oxidized low density lipoprotein in macrophage-rich areas of atherosclerotic
lesions. Proc Natl Acad Sci USA 1990; 87:69596963.
Aviram M, Kent UM, Hollenberg PF. Microsomal cytochrome P450 catalyze the
oxidation of low density lipoprotein. Atherosclerosis 1999; 143:253260.
Meister A, Anderson ME. Glutathione. Annu Rev Biochem 1983; 52:711760.
Rosenblat M, Aviram M. Macrophage glutathione content and glutathione
peroxidase activity are inversely related to cell-mediated oxidation of LDL. Free
Radic Biol Med 1997; 24:305313.
Fuhrman B, Oiknine J, Aviram M. Iron induces lipid peroxidation in cultured
macrophages, increases their ability to oxidatively modify LDL and aect their
secretory properties. Atherosclerosis 1994; 111:6578.
Fuhrman B, Oiknine J, Keidar S, Kaplan M, Aviram M. 1997 Increased uptake of
low density lipoprotein (LDL) by oxidized macrophages is the result of enhanced
LDL receptor activity and of progressive LDL oxidation. Free Radic Biol Med
1994; 23:3446.
Mackness MI, Mackness B, Durrington PN, Connelly PW, Hegele RA. Paraoxonases biochemistry, genetics and relationship to plasma lipoproteins. Curr Opin
Lipidol 1996; 7:6976.
La Du BN, Adkins S, Kuo CL, Lipsig D. Studies on human serum paraoxonase/
arylesterase. Chem Biol Interact 1993; 87:2534.
Aviram M. Does paraoxonase play a role in susceptibility to cardiovascular
disease? Mol Med 1999; 5:381386.
La Du BN, Aviram M, Billecke S, Navab M, Primo-Parmo S, Sorenson RC,
Standiford TJ. On the physiological role(s) of the paraoxonases. Chem Biol
Interact 1999; 119/120: 379388.
Mackness MI, Harty D, Bhatnagar D, Winocour PH, Arrol S, Ishola M, Durrington
PN. Serum paraoxonase activity in familial hypercholesterolaemia and insulindependent diabetes mellitus. Atherosclerosis 1991; 86:193197.
Abbott CA, Mackness MI, Kumar S, Boulton AJ, Durrington PN. Serum paraoxonase activity, concentration, and phenotype distribution in diabetes mellitus
and its relationship to serum lipids and lipoproteins. Arterioscler Thromb Vasc
Biol 1995; 15:18121818.
Garin MC, James RW, Dussoix P, Blanche H, Passa P, Froguel P, Ruiz J.
Paraoxonase polymorphism Met-Leu54 is associated with modied serum
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
193
concentrations of the enzyme: a possible link between the paraoxonase gene and
increased risk of cardiovascular disease in diabetes. J Clin Invest 1997; 99:6266.
Aviram M, Rosenblat M, Bisgaier CL, Newton RS, Primo-Parmo SL, La Du BN.
Paraoxonase inhibits high density lipoprotein (HDL) oxidation and preserves its
functions: a possible peroxidative role for paraoxonase. J Clin Invest 1998;
101:15811590.
Aviram M, Billecke S, Sorenson R, Bisgaier C, Newton R, Rosenblat M, Erogul J,
Hsu C, Dunlp C, La Du BN. Paraoxonase active site required for protection against
LDL oxidation involves its free sulfhydryl group and is dierent from that required
for its arylesterase/paraoxonase activities: selective action of human paraoxonase
allozymes Q and R. Arterioscler Thromb Vasc Biol 1998; 18:1617624.
Aviram M, Rosenblat M, Billecke S, Erogul J, Sorenson R, Bisgaier CL, Newton
RS, La Du B. Human serum paraoxonase (PON 1) is inactivated by oxidized low
density lipoprotein and preserved by antioxidants. Free Radic Biol Med 1999;
26:892904.
Keaney JF, Frei B. Antioxidant protection of low-density lipoprotein and its role
in the prevention of atherosclerotic vascular disease. In: Frei B, ed. Natural
Antioxidants in Human Health and Disease. San Diego: Academic Press, 1994:
303352.
Aviram M. Antioxidants in restenosis and atherosclerosis. Curr Interven Cardiol
Rep 1999; 1:6678.
Vaya J, Aviram M. Nutritional antioxidants: mechanisms of action, analyses of
activities and medical application. Curr Med Chem 2001; 1:99117.
Aviram M. Review of human studies on oxidative damage and antioxidant
protection related to cardiovascular diseases. Free Radic Res 2000; 33:S85S97.
Aviram M. Macrophage foam cell formation during early atherogenesis is
determined by the balance between pro-oxidants and antioxidants in arterial cells
and blood lipoproteins. Antiox Redox Signal 1999; 1:585594.
Levy Y, Ben-Amotz A, Aviram M. Eect of dietary supplementation of hCarotene to humans on its binding to plasma LDL and on the lipoprotein
susceptibility to undergo oxidative modication: comparison of the synthetic all
trans isomer with the natural algae h-carotene. J Nutr Environ Med 1995;
5:1322.
Levy Y, Kaplan M, Ben Amotz A, Aviram M. The eect of dietary supplementation of h-carotene on human monocyte-macrophage-mediated oxidation of low
density lipoprotein. Isr J Med Sci 1996; 32:473478.
Fuhrman B, Ben-Yaish L, Attias J, Hayek T, Aviram M. Tomatos lycopene and
h-carotene inhibit low density lipoprotein oxidation and this eect depends on the
lipoprotein vitamin E content. Nutr Metab Cardiovasc Dis 1997; 7:433443.
Fuhrman B, Volkova N, Rosenblat M, Aviram M. Lycopene synergistically
inhibits LDL oxidation in combination with vitamin E, glabridin, rosmarinic acid,
carnosic acid, or garlic. Antiox Redox Signal 2000; 2:491506.
Maor I, Hayek T, Coleman R, Aviram M. Plasma LDL oxidation leads to its
aggregation in atherosclerotic apolipoprotein E-decient mice. Arterioscler
194
Aviram and Fuhrman
58.
Fuhrman B, Aviram M. Anti-atherogenicity of nutritional antioxidants. Drugs

2001; 4:8292.
Hertog MG, Kromhout D, Aravanis C, Blackburn H, Buzina R, Fidanza F,
Giampaoli S, Jansen A, Menotti A, Nadeljkovic S. Flavonoid intake and long-term
risk of coronary heart disease and cancer in the seven countries study. Arch Intern
Med 1995; 155:381386.
Rice-Evans CA, Miller NJ, Bolwell PG, Bramley PM, Pridham JB. The relative
antioxidant activities of plant-derived polyphenolic avonoids. Free Radic Res
1995; 22:375383.
Rice-Evans CA, Miller NJ, Paganga G. Structure-antioxidant activity relationships of avonoids and phenolic acids. Free Radic Biol Med 1996; 20:
933956.
Van Acker SABE, Van-den Berg DJ, Tromp MNJL, Grioen DH, van Bennekom
WP, Van der Vijgh WJF, Bast A. Structural aspects of antioxidants activity of
avonoids. Free Radic Biol Med 1996; 20:331342.
Brown JA, Khodr H, Hider RC, Rice-Evans C. Structural dependence of avonoid
interactions with Cu2+ ions: implications for their antioxidant properties. Biochem
J 1998; 330:11731178.
Chang WS, Chang YH, Lu FJ, Chiang HC. Inhibitory eects of phenolics on
xanthine oxidase. Anticancer Res 1994; 14:501506.
Rosenblat M, Belinky P, Vaya J, Levy R, Hayek T, Coleman R, Merchav S,
Aviram M. Macrophage enrichment with the isoavan glabridin inhibits NADPH
oxidase-induced cell mediated oxidation of low density lipoprotein. J Biol Chem
1999; 274:1379013799.
Hsiech R, German B, Kinsella J. Relative inhibitory potencies of avonoids on 12lipoxygenase of sh gill. Lipids 1988; 23:322326.
Baumann J, Bruchhausen F, Wurm G. Flavonoids and related compounds as
inhibitors of arachidonic acid peroxidation. Prostaglandins 1980; 20:627639.
Luiz da Silva E, Tsushida T, Terao J. Inhibition of mammalian 15-lipoxygenasedependent lipid peroxidation in low density lipoprotein by quercetin and quercetin
monoglucosides. Arch Biochem Biophys 1998; 349:313320.
Lotito SB, Fraga CG. (+) -Catechin prevents human plasma oxidation. Free Radic
Biol Med 1998; 24:435441.
Mangiapane H, Thomson J, Salter A, Brown S, Bell GD, White DA. The inhibition
of the oxidation of low density lipoprotein by (+)-catechin, a naturally occurring
avonoid. Biochem Pharmacol 1992; 43:445450.
De Whalley CV, Rankin SM, Hoult RS, Jessup W, Leake DS. Flavonoids inhibit
the oxidative modication of low density lipoproteins by macrophages. Biochem
Pharmacol 1990; 39:17431750.
Vaya J, Belinky PA, Aviram M. Antioxidant constituents from licorice roots:
isolation, structure elucidation and antioxidative capacity toward LDL oxidation.
Belinky PA, Aviram M, Fuhrman B, Rosenblat M, Vaya J. The antioxidative
eects of the isoavan glabridin on endogenous constituents of LDL during is
oxidation. Atherosclerosis 1998; 137:4961.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.

74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
195
Namiki M, Osawa T. Antioxidants/mutagens in foods. Basic Life Sci 1986;

39;131142.
Shimoi K, Masuda S, Furugori M, Esaki S, Kinae N. Radioprotective eect of
antioxidative avonoids in gamma-ray irradiated mice. Carcinogenesis 1986;
15:26692672.
Bravo L. Polyphenols: chemistry, dietary sources, metabolism, and nutritional
signicance. Nutr Rev 1998; 56:317333.
Leibovitz BE, Mueller JA. Bioavonoids and polyphenols: medical application.
J Optim Nutr 1993; 2:1735.
de Vries JH, Hollman PC, Meyboom S, Buysman MN, Zock PL, van Staveren
WA, Katan MB. Plasma concentrations and urinary excretion of the antioxidant
avonols quercetin and kaempferol as biomarkers for dietary intake. Am J Clin
Nutr 1998; 68:6065.
Hollman PCH. Bioavailability of avonoids. Eur J Clin Nutr 1997; 51(suppl
1):S66S69.
Hertog MGL, Hollman POH, Katan MB, Kromhout D. Intake of potentially
Nutr Cancer 1993; 20:2129.
German JB, Walzem RL. The health benets of wine. Annu Rev Nutr 2000;
20:561593.
Goldberg IJ, Mosca L, Piano MR, Fisher EA. Wine and your heart. Circulation
2001; 103:472475.
Renaud S, de Lorgeril M. Wine alcohol, platelets and the French paradox for
coronary heart disease. Lancet 1992; 339:15231526.
Soleas GJ, Diamandis EP, Goldberg DM. Wine as a biological uid: history,
production, and role in disease prevention. J Clin Lab Anal 1997; 11:287313.
Hertog MGL, Hollman PCH, van de Putte B. Content of potentially anticarcinogenic avonoids of tea infusions, wines, and fruit juices. J Agric Food Chem
1993; 41:12421246.
Serani M, Maiani G, Ferro-Luzzi A. Alcohol-free red wine enhances plasma
antioxidant capacity in humans. J Nutr 1998; 128:10031007.
Fuhrman B, Aviram M. White wine reduces the susceptibility of low density
lipoprotein to oxidation. Am J Clin Nutr 1996; 63(3):403404.
Lamuela-Raventos RM, de la Torre-Boronat MC. Benecial eects of white
wines. Drugs Exp Clin Res 1999; 25:121124.
Paganga G, Miller N, Rice-Evans CA. The polyphenolic content of fruit and
vegetables and their antioxidant activities: what does a serving constitute? Free
Radic Res 1999; 30:153162.
Rici VA, Stephan EM, Schneider SH, Khachadurian AK. Red wine inhibits the
cell-mediated oxidation of LDL and HDL. J Am Coll Nutr 1999; 18:137143.
Vinson JA, Hontz BA. Phenol antioxidant index: comparative antioxidant eectiveness of red and white wines. J Agric Food Chem 1995; 43:401403.
Fuhrman B, Lavy A, Aviram M. Consumption of red wine with meals reduces the
susceptibility of human plasma and LDL to undergo lipid peroxidation. Am J Clin
Nutr 1995; 61:549554.
196
Aviram and Fuhrman
93.
Frankel EN, Waterhouse AL, Teissedre PL. Principal phenolic phytochemicals

in selected Californian wines and their antioxidant activity in inhibiting
oxidation of human low-density lipoproteins. J Agric Food Chem 1995; 43:
890893.
Lairon D, Amiot MJ. Flavonoids in food and natural antioxidants in wine. Curr
Opin Lipidol 1999; 10:2328.
Frankel EN, Kanner J, German JB, Parks E, Kinsella JE. Inhibition of oxidation
human low-density lipoprotein by phenolic substances in red wine. Lancet 1993;
341:454457.
Fuhrman B, Volkova N, Aviram M. White wine with red wine-like properties:
increased extraction of grape skins-polyphenols improves the antioxidant capacity
of the derived white wine. J Agric Food Chem 2001; 49:31643168.
Whitehead TP, Robinson D, Allaway S, Syms J, Hale A. Eect of red wine
ingestion on the antioxidant capacity of serum. Clin Chem 1995; 41:3235.
Aviram M, Fuhrman B. Wine avonoids protect against LDL oxidation and
atherosclerosis. Annals NY Acad Sci 2002; 957:146161.
Miyagi Y, Miwa K, Inoue H. Inhibition of human low density lipoprotein
oxidation by avonoids in red wine and grape juice. Am J Cardiol 1997;
80:16271631.
Nigdikar SV, Williams N, Grin BA, Howard AH. Consumption of red wine
polyphenols reduces the susceptibility of low density lipoproteins to oxidation in
vivo. Am J Clin Nutr 1998; 68:258265.
de Rijke YB, Demacker PN, Assen NA, Sloots LM, Katan MB, Stalenhoef AF.
Red wine consumption does not aect oxidizability of low-density lipoprotein
volunteers. Am J Clin Nutr 1996; 63:329334.
Abu-Amsha CR, Burke V, Mori TA, Beilin LJ, Puddey IB, Croft KD. Red wine
polyphenols, in the absence of alcohol, reduce lipid peroxidative stress in smoking
subjects. Free Radic Biol Med 2001; 30:636642.
Chopra M, Fitzsimons PEE, Strain JJ, Thurnham DI, Howard AN. Nonalcoholic
red wine extract and quercetin inhibit LDL oxidation without aecting plasma
antioxidant vitamin and carotenoid concentrations. Clin Chem 2000; 46(8):1162
1170.
Fremon L, Belguendouz L, Delpal S. Antioxidant activity of resveratrol and
alcohol-free wine polyphenols related to LDL oxidation and polyunsaturated fatty
acids. Life Sci 1999; 64:25112521.
Stein JH, Keevil JG, Wiebe DA, Aeschlimann S, Folts JD. Purple grape juice
improves endothelial function and reduces the susceptibility of LDL cholesterol
to oxidation in patients with coronary artery disease. Circulation 1999; 100:1050
1055.
van Golde PH, Sloots LM, Vermeulen WP, Wielders JP, Hart HC, Bouma BN, van
de Wiel A. The role of alcohol in the anti low density lipoprotein oxidation activity
of red wine. Atherosclerosis 1999; 147:365370.
Ritchey JG, Waterhouse AL. A standard red wine: monomeric phenolic analysis of commercial Cabernet sauvignon wines. Am J Enol Viticul 1999; 50(1):
91100.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.

108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
197
McDonald MS, Hughes M, Burns J, Lean MEJ, Matthews D, Crozier A. Survey of

free and conjugated myricetin and quercetin content of red wines of dierent
geographical origin. J Agric Food Chem 1998; 46:368375.
Price SF, Breen PJ, Valladao M, Watson BT. Clusters sun exposure and quercetin
in Pinot noir grapes and wines. Am J Enol Viticul 1995; 46:187194.
Carbonneau MA, Leger CL, Monnier L, Bonnet C, Michel F, Fouret G.
Supplementation with wine phenolic compounds increases the antioxidant capacity
of plasma and vitamin E of low-density lipoprotein without changing the
lipoprotein Cu2+ oxidizability: possible explanation by phenolic location. Eur J
Clin Nutr 1997; 51:682690.
Hayek T, Fuhrman B, Vaya J, Rosenblat M, Belinky P, Coleman R, Elis A,
Aviram M. Reduced progression of atherosclerosis in the apolipoprotein E
decient mice following consumption of red wine, or its polyphenols quercetin or
catechin, is associated with reduced susceptibility of LDL to oxidation and
aggregation. Arterioscler Thromb Vasc Biol 1997; 17:27442752.
Aviram M, Hayek T, Fuhrman B. Red wine consumption inhibits LDL
oxidation and aggregation in humans and in atherosclerotic mice. Biofactors
1997; 6:415419.
Bentzon JF, Skovenborg E, Hansen C, Moler J, Saint-Cricq de Gaulejac N, Proch
J, Falk E. Red wine does not reduce mature atherosclerosis in apolipoprotein Edecient mice. Circulation 2001; 103:1681.
Vinson JA, Teufel K, Wu N. Red wine, dealcoholized red wine, and especially
grape juice, inhibit atherosclerosis in a hamster model. Atherosclerosis 2001;
156:6772.
Nardini M, DAquino M, Tomassi G, Gentili V, Di Felice M, Scaccini C.
Inhibition of human low density lipoprotein oxidation by caeic acid and other
hydroxycinnamic acid derivatives. Free Radic Biol Med 1995; 19:541552.
Abu-Amsha R, Croft KD, Puddey IB, Proudfoot JM, Beilin LJ. Phenolic content of
various beverages determines the extent of inhibition of human serum and low
density lipoprotein oxidation in vitro: identication and mechanism of some
cinnamic derivatives from red wine. Clin Sci 1996; 91:449458.
Frankel EN, Waterhouse AL, Kinsella JE. Inhibition of human LDL oxidation by
resveratrol. Lancet 1993; 341:11031104.
Manach C, Morand C, Texier O, Favier ML, Agullo G, Demigne C, Regerat F,
Remesy C. Quercetin metabolites in plasma of rats fed diets containing rutin or
quercetin. J Nutr 1995; 125:19111922.
Vinson JA, Dabbagh YA, Serry MM, Janj J. Plant avonoids, especially tea
avonols, are powerful antioxidants using an in vitro model for heart disease.
J Agric Food Chem 1995; 45:28002802.
Salah N, Miller NJ, Paganga G, Tijburg L, Bolwell GP, Rice Evans C.
Polyphenolic avanols as scavengers of aqueous phase radicals and as chainbreaking antioxidants. Arch Biochem Biophys 1995; 322:339346.
Lanninghamfoster L, Chen C, Chance DS, Loo G. Grape extract inhibits lipid
peroxidation of human low density lipoprotein. Biol Pharm Bull 1995;
18:13471351.
198
122.
Aviram and Fuhrman
Rao AV, Shen H, Agarwal A, Yatcilla MT, Agarwal S. Bioabsorption and in vivo
antioxidant properties of grape extract Biovin: a human intervention study. J Med
Food 2000; 3:1522.
123. Kerry NL, Abbey M. Red wine and fractionated phenolic compounds prepared
from red wine inhibit low density lipoprotein oxidation in vitro. Atherosclerosis
1997; 135:93102.
124. Fuhrman B, Aviram M. Paraoxonase activity is preserved by wine avonoids:
possible role in protection of LDL from lipid peroxidation. Annals NY Acad Sci
2002; 957:321324.
125. van der Gaag MS, van Tol A, Scheek LM, James RW, Urgert R, Schaafsma G,
Hendriks HFJ. Daily moderate alcohol consumption increases serum paraoxonase
activity: a diet controlled, randomized intervention study in middle-aged men.
Atheroclerosis 1999; 147:405410.
126. Fuhrman B, Buch S, Vaya J, Belinky PA, Coleman R, Hayek T, Aviram M.
Licorice extract and its major polyphenol glabridin protect low-density lipoprotein
against lipid peroxidation: in vitro and ex vivo studies in humans and in atherosclerotic apolipoprotein E-decient mice. Am J Clin Nutr 1997; 66:267275.
127. Fuhrman B, Volkova N, Kaplan M, Presser D, Attias J, Hayek T, Aviram M.
Antiatherosclerotic eects of licorice extract supplementation to hypercholesterolemic patients: increased resistance of their LDL to atherogenic modications,
reduced plasma lipids levels, and decreased systolic blood pressure. Nutrition
2002; 18:268273.
128. Haraguchi H, Ishikawa H, Mizutani K, Tamura Y, Kinoshita T. Antioxidative and
superoxide scavenging activities of retrochalcones in Glycyrrhiza inata. Bioorg
Med Chem 1998; 6:339347.
129. Belinky PA, Aviram M, Mahmood S, Vaya J. Structural aspects of the inhibitory
eect of glabridin on LDL oxidation. Free Radic Biol Med 1998; 24:14191429.
130. Langley P. Why a pomegranate? Br Med J 2000; 321:11531154.
131. Ben Nasr C, Ayed N, Metche M. Quantitative determination of the polyphenolic
content of pomegranate peel. Z Lebensm Unters Forsch 1996; 203:374378.
132. El-Toumy SAA, Marzouk MSA. A new ellagic acid glycoside from Punica
granatum L. Polyphenols Communications. Germany: Freising-Weihenstephan,
2000:127128.
133. Gil MI, Tomas-Barberan FA, Hess-Pierce B, Holcroft DM, Kader AA. Antioxidant
activity of pomegranate juice and its relationship with phenolic composition and
processing. J Agric Food Chem 2000; 48:45814589.
134. Heftaman E, Bennett ST. Identication of estrone in pomegranate seeds.
135. Moneam NMA, El Sharasky AS, Badreldin MM. Oestrogen content of
pomegranate seeds. J Chromatogr 1988; 438:438442.
136. Sharaf A, Nigm SAR. The oestrogenic acitvity of pomegranate seed oil.
J Endocrinol 1964; 29:9192.
137. Shubert YS, Lansky EP, Neeman I. Antioxidant and eicosanoid enzyme inhibition
properties of pomegranate seed oil and fermented juice avonoids. J Ethnopharmacol 1999; 66:1117.

138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
150.
151.
152.
153.
199
Aviram M, Dorenfeld L, Rosenblat M, Volkova N, Kaplan M, Hayek T, Presser D,

Fuhrman B. Pomegranate juice consumption reduces oxidative stress, atherogenic
modications to LDL, and platelet aggregation: studies in humans and in the
atherosclerotic apolipoprotein E decient mice. Am J Clin Nutr 2000; 71:1062
1076.
Aviram M, Dornfeld L. Pomegranate juice consumption inhibits serum angiotensin converting enzyme activity and reduces systolic blood pressure. Atherosclerosis 2001; 158:195198.
Aviram M, Dornfeld L, Kaplan M, Coleman R, Gaitini D, Nitecki S, Homan A,
Rosenblat M, Volkova N, Presser D, Attias J, Hayek T, Fuhrman B. Pomegranate
juice avonoids inhibit LDL oxidation and cardiovascular diseases: studies in
atherosclerotic mice and in humans. Proc Int Conf Mech Action Nutraceuticals
(ICMAN). 2002; 28:4962.
He K, Lyer KR, Hayes RN. Inactivation of cytochrome P-450 3A4 by bergamottin,
a component of grapefruit juice. Chem Res Toxicol 1998; 11:252259.
Kaplan M, Hayek T, Raz A, Coleman R, Dornfeld L, Vaya J, Aviram M.
Pomegranate juice supplementation to atherosclerotic mice reduces macrophages
lipid peroxidation, cellular cholesterol accumulation and development of
atherosclerosis. J Nutr 2001; 131:20822089.
Thelle DS. Coee, tea and coronary heart disease. Curr Opin Lipidol 1995;
6:2527.
Yang CS. Tea and health. Nutrition 1999; 15:946949.
Stensvold I, Tverdal A, Solvoll K, Foss PO. Tea consumption, relationship to
cholesterol, blood pressure and coronary and total mortality. Prev Med 1992;
21:546553.
Graham HN. Green tea composition, consumption and polyphenol chemistry. Prev
Med 1992; 21:334334.
Weisburger JH. Tea and health: the underlying mechanisms. Proc Soc Exp Biol
Med 1999; 220:271275.
Miyazawa T. Absorption, metabolism and antioxidative eects of tea catechins in
humans. Biofactors 2000; 13:5559.
Van het Hof KH, de Boer HS, Wiseman SA, Lien N, Westrate JA, Tijburg LB.
Consumption of green or black tea does not increase resistance of low density
lipoprotein to oxidation in humans. Am J Clin Nutr 1997; 66:11251132.
van het Hof KH, Wiseman SA, Yang CS, Tijburg LB. Plasma and lipoprotein
levels of tea catechins following repeated tea consumption. Proc Soc Exp Biol
Med 1999; 220:203209.
McAnlis GT, McEneny J, Pearce J, Young IS. Black tea consumption does not
protect low-density lipoprotein from oxidative modication. Eur J Clin Nutr 1998;
52:202206.
Princen HM, van Duyvennvoorde W, Buytenhek R, Blonk C, Tijburg LB, Langius
JA, Meinders AE, Pijl H. No eect of consumption of green and black tea on
plasma lipid and antioxidant levels and on LDL oxidation in smokers. Arterioscler
OReilly JD, Mallet AI, McAnlis GT, Young IS, Halliwell B, Sanders TA,
200
154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
166.
167.
Aviram and Fuhrman

Wiseman H. Consumption of avonoids in onions and black tea: lack of eect on
F2-isoprostanes and autoantibodies to oxidized LDL in healthy humans. Am J Clin
Nutr 2001; 73:10401044.
Serani M, Ghiselli A, Ferro-Luzzi A. In vivo antioxidant eect of green and black
tea in man. Eur J Clin Nutr 1996; 50:2832.
Benzie IFF, Szeto YT, Tomlinson B, Strain JJ. Drinking green tea leads to a rapid
increase in plasma antioxidant potential. In: Kumpulainen JT, Salonen JT, eds.
Natural Antioxidants and Anticarcinogens in Nutrition, Health and Disease.
London: The Royal Society of Chemistry, 1999:280282.
Ishikawa T, Suzukawa M, Ito T, Yioshida H, Ayaori M, Nishiwaki M, Yonemura
A, Hara Y, Nakamura H. Eect of tea avonoid supplementation on the susceptibility of low density lipoprotein to oxidative modication. Am J Clin Nutr
1997; 66:261266.
Serani M, Laranjinha JA, Almeida LM, Maiani G. Inhibition of human LDL lipid
peroxidation by phenol-rich beverages and their impact on plasma total antioxidant
capacity in humans. J Nutr Biochem 2000; 11:585590.
Hodgson JM, Puddey IB, Croft KD, Burke V, Mori TA, Caccetta RA, Beilin LJ.
Acute eects of ingstion of black and green tea on lipoprotein oxidation. Am J
Clin Nutr 2000; 71:11031107.
Tijburg LB, Wiseman SA, Meijer GW, Weststrate JA. Eects of green tea, black
tea and dietary lipophilic antioxidants on LDL oxidizability and atherosclerosis in
hypercholesterolaemic rabbits. Atherosclerosis 1997; 135:3747.
Vinson JA, Dabbagh YA. Eect of green and black tea supplementation on lipids,
lipid oxidation and brinogen in the hamster: mechanisms for the epidemiological
benets of tea drinking. FEBS Lett 1998; 14:4446.
Miura Y, Chiba T, Tomita I, Koizumi H, Miura S, Umegaki K, Hara Y, Ikeda M,
Tomita T. Tea catechins prevent the development of atherosclerosis in apoprotein
E-decient mice. J Nutr 2001; 131:2732.
Yang TTC, Koo MWL. Inhibitory eect of Chinese green tea on endothelia cellinduced LDL oxidation. Atherosclerosis 2000; 148:6773.
Miura Y, Chiba T, Miura S, Tomita I, Umegaki K, Ikeda M, Tomita T. Green
tea polyphenols (avan 3-ols) prevent oxidative modication of low density
lipoproteins: an ex vivo study in humans. J Nutr Biochem 2000; 11: 216222.
Yamanaka N, Oda O, Nagao S. Green tea catechins such as ()-epicatechin and
()-epigallocatechin accelerate Cu2+-induced low density lipoprotein oxidation
in propagation phase. FEBS Lett 1997; 20:230234.
Liu ZQ, Ma LP, Zhou B, Yang L, Liu ZL. Antioxidative eects of green tea
polyphenols on free radical initiated and photosensitized peroxidation of human
low density lipoprotein. Chem Phys Lipids 2000; 106:5363.
Osada K, Takahashi M, Hoshina S, Nakamura M, Nakamura S, Sugano M. Tea
catechins inhibit cholesterol oxidation accompanying oxidation of low density
lipoprotein in vitro. Comp Biochem Physiol C Pharmacol Toxicol Endocrinol
2001; 128:153164.
Yoshida H, Ishikawa T, Hosoai H, Suzukawa M, Ayaori M, Hisada T, Sawada S,
Yonemura A, Higashi K, Ito T, Nakajima K, Yamashita T, Tomiyasu K,
168.
169.
170.
171.
172.
173.
174.
175.
176.
177.
178.
179.
201
Nishiwaki M, Ohsuzu F, Nakamura H. Inhibitory eect of tea avonoids on the

ability of cells to oxidize low density lipoprotein. Biochem Pharmacol 1999;
58:16951703.
Zhu QY, Huang Y, Tsang D, Chen ZY. Regeneration of alpha-tocopherol in
human low-density lipoprotein by green tea catechin. J Agric Food Chem 1999;
47:20202025.
Keys A. Mediterranean diet and public health: personal reections. Am. J Clin
Nutr 1995; 61:1321S1323S.
Fidanza F, Puddu V, Imbimbo AB, Menotti A, Keys A. Coronary heart disease in
seven countries. VII. Five-year experience in rural Italy. Circulation 1970; 41:I63
I75.
Lopez-Miranda J, Gomez P, Castro P, Marin C, Paz E, Bravo MD, Blanco J,
Jimenez-Pereperez J, Fuentes F, Perez-Jimenez F. Mediterranean diet improves
low density lipoprotein susceptibility to oxidative modications. Med Clin (Barc)
2000; 115:361365.
Visioli F, Galli C. Natural antioxidants and prevention of coronary heart disease: a
potential role of olive oil and its minor constituents. Nutr Metab Cardiovasc Dis
1995; 5:306314.
Tsimikas S, Philis-Tsimikas A, Alexopoulos S, Sigari F, Lee C, Reaven PD. 1999
LDL isolated from Greek subjects on a typical diet or from American subjects on
an oleate-supplemented diet induces less monocyte chemotaxis and adhesion when
exposed to oxidative stress. Arterioscler Thromb Vasc Biol 1995; 19:122130.
Aviram M, Eias K. Dietary olive oil reduces low-density lipoprotein uptake by
macrophages and decreases the susceptibility of the lipoprotein to undergo lipid
peroxidation. Ann Nutr Metab 1993; 37:7584.
Nicolaiew N, Lemort N, Adorni L, Berra B, Montorfano G, Rapelli S, Cortesi N,
Jacotot B. Comparison between extra virgin olive oil and oleic acid rich sunower
oil: eects on postprandial lipemia and LDL susceptibility to oxidation. Ann Nutr
Metab 19981; 42:251260.
Wiseman SA, Mathot JN, de Fouw NJ, Tijburg LB. Dietary non-tocopherol
antioxidants present in extra virgin olive oil increase the resistance of low density
lipoprotein to oxidation in rabbits. Atherosclerosis 1996; 120:1523.
Ramirez-Tortosa MC, Urbano G, Lopez-Jurado M, Nestares T, Gomez MC, Mir
A, Ros E, Mataix J, Gil A. Extra-virgin olive oil increases the resistance of LDL to
oxidation more than rened olive oil in free-living men with peripheral vascular
disease. J Nutr 1999; 129:21772183.
Ramirez-Tortosa MC, Suarez A, Gomez MC, Mir A, Ros E, Mataix J, Gil A.
Eect of extra-virgin olive oil and sh-oil supplementation on plasma lipids
and susceptibility of low-density lipoprotein to oxidative alteration in freeliving Spanish male patients with peripheral vascular disease. Clin Nutr 1999;
18:167174.
Ramirez-Tortosa C, Lopez-Pedrosa JM, Suarez A, Ros E, Mataix J, Gil A. Olive
oil and sh oil-enriched diets modify plasma lipids and susceptibility of LDL to
oxidative modication in free-living male patients with peripheral vascular
disease: the Spanish Nutrition Study. Br J Nutr 1999; 82:3139.
202
Aviram and Fuhrman
180.
Vissers MN, Zock PL, Wiseman SA, Meyboom S, Katan MB. Eect of phenolrich extra virgin olive oil on markers of oxidation in healthy volunteers. Eur J Clin
Nutr 2001; 55:334341.
Caruso D, Berra B, Giavarini F, Cortesi N, Fedeli E, Galli G. Eect of virgin olive
oil phenolic compounds on in vitro oxidation of human low density lipoproteins.
Nutr Metab Cardiovasc Dis 1999; 9:102107.
Covas MI, Fito M, Lamuela-Raventos RM, Sebastia N, de la Torre-Boronat C,
Marrugat J. Virgin olive oil phenolic compounds: binding to human low density
lipoprotein (LDL) and eect on LDL oxidation. Int J Clin Pharmacol Res 2000;
20:4954.
Fito M, Fito M, Lamuela-Raventos RM, Vila J, Torrents L, de la Torre C,
Marrugat J. Protective eect of olive oil and its phenolic compounds against low
density lipoprotein oxidation. Lipids 2000; 35:633638.
Visioli F, Galli C. The eect of minor constituents of olive oil on cardiovascular
disease: new ndings. Nutr Rev 1998; 56:142147.
Visioli F, Bellomo G, Montedoro G, Galli C. Low density lipoprotein oxidation is
inhibited in vitro by olive oil constituents. Atherosclerosis 1995; 117:2532.
Visioli F, Galli C. Oleuropein protects low density lipoprotein from oxidation. Life
Sci 1994; 55:19651971.
Visioli F, Bellomo G, Galli C. Free radical-scavenging properties of olive oil
polyphenols. Biochem Biophys Res Commun 1998; 247:6064.
Coni E, Di Benedetto R, Di Pasquale M, Masella R, Modesti D, Mattei R, Carlini
EA. Protective eect of oleuropein, an olive oil biophenol, on low density
lipoprotein oxidizability in rabbits. Lipids 2000; 35:4554.
Giovannini C, Straface E, Modesti D, Coni E, Cantafora A, De Vincenzi M,
Malorni W, Masella R. Tyrosol, the major olive oil biophenol, protects against
oxidized-LDL-induced injury in Caco-2 cells. J Nutr 1999; 129:12691277.
Masella R, Cantafora A, Modesti D, Cardilli A, Gennaro L, Bocca A, Coni E.
Antioxidant activity of 3,4-DHPEA-EA and protocatechuic acid: a comparative
assessment with other olive oil biophenols. Redox Rep 1999; 4:113121.
Visioli F, Romani A, Mulinacci N, Zarini S, Conte D, Vincieri FF, Galli C.
Antioxidant and other biological activities of olive mill waste waters. J Agric Food
Chem 1999; 47:33973401.
de la Puerta R, Ruiz Gutierrez V, Hoult JR. Inhibition of leukocyte 5-lipoxygenase
by phenolics from virgin olive oil. Biochem Pharmacol 1999; 57:445449.
Petroni A, Blasevich M, Papini N, Salami M, Sala A, Galli C. Inhibition of
leukocyte leukotriene B4 production by an olive oil-derived phenol identied by
mass-spectrometry. Thromb Res 1997; 87:315322.
Visioli F, Bellosta S, Galli C. Oleuropein, the bitter principle of olives, enhances
nitric oxide production by mouse macrophages. Life Sci 1998; 62:541546.
Leger CL, Kadiri-Hassani N, Descomps B. Decreased superoxide anion production
in cultured human promonocyte cells (THP-1) due to polyphenol mixtures from
olive oil processing wastewaters. J Agric Food Chem 2000; 48:50615067.
Miles EA, Wallace FA, Calder PC. An olive oil-rich diet reduces scavenger
receptor mRNA in murine macrophages. Br J Nutr 2001; 85:185191.
181.
182.
183.
184.
185.
186.
187.
188.
189.
190.
191.
192.
193.
194.
195.
196.

197.
203
Aviram M. Macrophage-mediated oxidation of LDL, antioxidants, and atherosclerosis. J Isr Heart Soc 1998; 8:68.
198. Aviram M, Fuhrman B. Polyphenolic avonoids inhibit macrophage-mediated
oxidation of LDL and attenuate atherogenesis. Atherosclerosis 1998; 137(suppl):
S45S50.
199. Fuhrman B, Aviram M. Flavonoids protect LDL from oxidation and attenuate
atherosclerosis. Curr Opin Lipidol 2001; 12:4148.
200. Fuhrman B, Aviram M. Polyphenols and avonoids protect LDL against
atherogenic modications. In: Handbook of Antioxidants: Biochemical, Nutritional and Clinical Aspects, 2d ed. New York: Marcel Dekker, 2001:303336.
8
Phytochemicals and Brain Aging:
A Multiplicity of Effects
Kuresh A. Youdim
London, England
James A. Joseph
Jean Mayer USDA Human Nutrition Center on Aging at Tufts University
Boston, Massachusetts, U.S.A.
If we could give every individual the right amount of nourishment and

exercise, not too little and not too much, we would have found the safest
way to health.
Hippocrates c. 460377 B.C.E.
I.
INTRODUCTION
Throughout the millennia, humankind has pursued the concept of achieving

eternal youth. However, if an expanded existence is to be of benet, it should
be coupled with a high quality of life. Unfortunately, progress toward the goal of
healthy longevity has often lagged well behind the aspirations of philosophers.
Nonetheless, scientists have long been interested in the aging process, which has
received renewed attention as a result of the ever-increasing number of aged
persons and the signicant burden of age-related disease on national expenditure.
Changes in the optimal performance of biological systems invariably
impact health. However, alterations in the ecient functioning of the central
nervous system (CNS) have perhaps the most devastating repercussions. Among
those decits having the greatest impact are those associated with dementia, a
general term describing symptoms exhibited by people with various kinds of
cognitive impairment. Common disorders in which dementia is observed include
205
206
Youdim and Joseph
Alzheimers disease (AD); cerebrovascular disease, such as successive strokes or

lesions; Parkinsons disease (PD); Creutzfeldt-Jakob disease; alcoholism; and
certain traumas. Of greater concern perhaps are the changes in the optimal
performance of the CNS that occur simply as a function of aging, possibly
exacerbating common correlative motor and cognitive behavioral changes
exhibited in these conditions. Age-related decits in memory occur primarily
in secondary memory systems and are reected in the retrieval of newly acquired
information. The impairments in retrieval are attributed to decits in such
encoding processes as motivation, attention, processing depth, and organizational
skills and have been characterized in animals [13] and humans [4,5]. In contrast,
motor performance decits are thought to result from alterations in the striatal
dopamine system [6] or in the cerebellum [7], whereas age-related memory
decrements can result from alterations in either the hippocampus (which mediates
allocentric spatial navigation or place learning) or the striatum (which mediates
egocentric spatial orientation and response-cue learning). The mechanisms
involved in the losses of sensitivity in the various receptor systems and the
related loss in cognitive and motor behavior are the subjects of continued
research. The critical issue is that additional research has shown that this
sensitivity may increase further during aging [8].
Several factors contribute to losses of neuronal function and behavior
observed in brain aging (Fig. 1). Examples include age-related changes in
sensitivity to oxidative [911] and inammatory stresses [12,13], endogenous
antioxidant system [14] receptor sensitivity [1518], membrane alterations [19],
and calcium homeostasis [20]. With the involvement of these various pathophysiological processes, employing agents that nullify their actions could potentially
provide eective neuroprotective therapies. Thus, there is a growing interest in
the establishment of therapeutic strategies to combat oxidative stressinduced
Figure 1 Common changes observed in the aging brain that are associated
with dementia.
Phytochemicals and Brain Aging
207
damage to the CNS, and attention is turning to the potential neuroprotective

eects of dietary antioxidants, especially avonoids. Only recently have studies
been performed that focus on the potential for avonoids per se to mediate
neuroprotection. It is not clear whether the neuroprotective eects of avonoids
involve their reducing properties or some other mechanism independent of their
antioxidant activities. Their precise mechanisms of action in vivo depend on the
extent to which they are conjugated and metabolized during absorption (see
Chap. 12) and the ability of bioavailable compounds to localize within the brain.
II.
DE NOVO LOCALIZATION
The inherent diculties in the characterization, distribution, and localization of

avonoids within the body pose a major hurdle when attempting to investigate
mechanisms of avonoid bioactivity in vivo. Compounds reported to localize
within brain structures include epigallocatechin gallate [21], a major polyphenolic in tea and to a lesser degree in wine, and the citrus avonoids hesperetin
Figure 2 Potential interactions of bioavailable flavonoid conjugates with the

blood brain barrier (BBB).
208
Youdim and Joseph
[22] and naringenin together with its glucuronide conjugate [23]. Schroder-van
der Elst and associates [24,25] have also identied the synthetic avonoid Emd
49209, which is able to localize both in the adult rat brain and in the developing
fetal rat brain. The paucity of studies is due in part to a limited knowledge of
polyphenolic bioavailability and characterization of the circulating forms that
potentially interact with the CNS. Furthermore, knowledge about the physiological interaction between bioavailable avonoids and/or their conjugates with
the blood-brain barrier (BBB) is also limited (Fig. 2). This interaction is of
fundamental importance, considering that the functional role of the BBB is to
control the composition of extracellular uid in the CNS, sealing o entry of all
but the smallest molecules. Hence this barrier ultimately determines the fate of
dietary components such as avonoids within the CNS.
III.
NUTRITIONAL COMPONENTS AND THEIR ROLE

IN PREVENTING BRAIN AGING
Although evidence suggesting avonoids are able to localize in the brain is scarce,
growing awareness gained from epidemiological and dietary intervention studies
of humans and animals suggests that avonoid consumption may be important to
neuronal health. The contributory role of avonoids to the modulation of
neurodegeneration, especially age-related cognitive and motoric decline, in
protection against oxidative stress, cerebral ischemia/reperfusion injuries, and
other brain abnormalities is being extensively investigated (see Tables 1 and 2).
IV.
NEUROPROTECTIVE EFFECTS OF FRUIT

AND VEGETABLES
We and others have shown in rodents that dietary supplementation with extracts
prepared from strawberry, spinach, and blueberry imparted signicant protection
against neurological parameters sensitive to oxidative stress. These included
receptor sensitivity [16], cerebellar Purkinje cell activity and calcium buering
capacity [26], guanosine triphosphatase (GTPase) coupling/uncoupling, and
cognitive functions [2,11]. In particular, blueberry extract supplementation to
middle-aged or aged animals was found to have a profound eect. One striking
observation after supplementation was the signicant increase in oxotremorine
enhancement of dopamine release from isolated striatal slices. This is especially
important since maintaining the functional integrity of the striatal dopaminergic
system has a major impact on certain behavioral parameters [2729]. What
remains to be examined is whether this modulation in dopamine release is
attributable to an increase in neuronal sensitivity and/or neurogenesis of striatal
209
neurons. It is also unclear whether these fruit and vegetable extracts might have
promoted similar mechanisms toward age-related changes in cerebellar h-adrenergic function [30,31]. It has been postulated that the cerebellar noradrenergic
system, which shows age-related changes in h-adrenergic function, may underlie
certain age-related decits in motor learning [7]. Norepinephrine potentiates
GABA-induced inhibition of cerebellar Purkinje neurons via the h-adrenergic
receptor. In aged rats, h-adrenergic potentiation occurs in only 30% of the
recorded cerebellar Purkinje cells as compared with 7080% of those in younger
animals. However, despite these ndings, current of studies investigating fruit and
vegetable extracts have not investigated whether avonoid components were able
to localize within brain structures. Hence one can only hypothesize that the
neuroprotective actions, such as modulation of receptor systems, are mediated
from within the CNS or may reside in some kind of peripheral eect.
Although the site(s) where these extract mediate their eects are presently
unknown, the ndings suggest that in vitro antioxidant activities of these extracts
were not predictive in assessing their potency against neurological decits. In our
early studies [32,33], extracts were supplemented in the diet at 1.36 mmol Trolox
equivalents/kg diet, such that animals consumed an equal concentration of
antioxidants per day. However, extracts diered in the ability to aord protection
in parameters sensitive to oxidative stress. This illustrates that a simple measure of
in vitro antioxidant activity alone may not be sucient to argue potential health
benets, and that functional assessments need to be combined. More recently, in
2000, it was shown that two dierent cultivars of blueberry, when supplemented in
the diet on an equal-weight basis (20 g/kg-diet), also exhibited dierent degrees of
protection against memory and learning declines in aging rats [34]. Collectively
these ndings suggest that dierences in the bioavailability of the diverse array of
avonoid components found in these extracts and their biological potency on
entering the circulation play an important role. Moreover, it is not clear from these
studies whether the neuroprotective eects of avonoids against oxidative stress
involve their reducing properties or some other mechanism independent of their
antioxidant activities. In this regard, when two indices of antioxidant activity
reactive oxygen species (ROS) production and glutathione levels in the striatum
and cerebellum were examined [32], supplementations resulted in only modest
eects that could not totally account for the ecacy of these extracts, especially
with regard to their eects on motor and cognitive function.
Recent studies (Joseph, unpublished) appear to draw attention to a possible
interaction with signaling molecules, although whether this is a direct or indirect
property has yet to be elucidated. For example, blueberry supplementation was
found to induce alterations in age- and calcium-sensitive signaling molecules
associated with memory, especially the conversion of short-to long-term memory.
These include calcium-dependent protein kinase C (PKC), for which studies have
shown that its activity is important in formation of memory, particularly spatial
Concentration
Duration
20 g/kg diet
5 mg/dL
Blueberrya
Grape
polyphenols
2 mo
8 wk
1.36 m mol Trolox

8 wk
equivalents/kg diet
Strawberry,
spinach, and
blueberrya
Fruits and vegetables

1.36 m mol Trolox
8 mo
Strawberry
equivalents/kg diet
and spinacha
Flavonoid source
In diet
In diet
In diet
In diet
Route
Rat
Rat
Rat
Rat
Species
5% Ethanol in
drinking water
Normal aging
Hyperoxia
Normal aging
Normal aging
Normal aging
Stress type
Observations
Improved striatal dopamine

receptor sensitivity, cerebellar
Purkinje cell activity, calcium
buering capacity, GTPase
coupling/uncoupling, and
cognitive functions
receptor sensitivity, cerebellar
Purkinje cell activity, calcium
buering capacity, GTPase
coupling/uncoupling, and
behavioral and cognitive functions
Improved motor learning
and cerebellar h-adrenergic
receptor function and cerebellar
glutathione concentrations
Prevention of declines in and
cerebellar h-adrenergic
receptor function
receptor sensitivity,
striatal and cortical vitamin C
concentrations, and behavioral
and cognitive functions
Prevention of decrease in
synaptosomal Na, K-ATPase
activity and dopamine uptake
Table 1 The Neuroprotective Actions of Flavonoid-Rich Extracts Found in Dietary Sources
[130]
[34]
[30]
[31]
[129]
[33]
References
210
Youdim and Joseph
Concentration
Green tea
Chronic
0.5 and 1 mg/kg
Body weight
0.5% And 2%
Green tea
Green tea
0.5%
Tea
Green tea
Grape seed
25100 mg/kg
proanthocyanidin
Body weight
extract
Flavonoid source
Drinking
water
Drinking
water
Gavage
Route
2 wk
Orally
2/day
Intraperitofor 1 day
neal
Single dose Intranigral

infusion
3 wk
3 wk
1 wk
Duration
Mouse
Rat
Gerbil
Rat
Mouse
Species
Reduced infarction volume,

free radical production,
lipid peroxidation, DNA
damage, and apoptotic
cell numbers in striatum
Reduced infarction volume,
free radical production,
lipid peroxidation, DNA
damage, and apoptotic
cell numbers in striatum
Intranigral infusion decreased
lipid peroxidation in
substantia nigra. Co-infusion
reduced elevation in lipid
peroxidation in substantia nigra
and associated decrease in
striatal dopamine content. oral
administration had no eect
Dose-dependent reduction of
reactive oxygen species
formation, lipid peroxidation,
and DNA fragmentation
Observations
[49]
[53]
[133]
[132]
[131]
References
(continued on next page )
MPTP (24 mg/kg Protection against dopaminergic

neuronal loss, maintenance of
body weight IP)
dopamine and tyrosine
for 4 daysc
hydroxylase concentrations
Intranigral
infusion of
ferrous citrate
(4.2nM)
Ischemia/
reperfusion
Ischemia/
reperfusion
Intraperitoneal
injection of
TPA* (0.1 Ag)
Stress type

211
chronic
400 mL
1 day
(3 times/day)
Kombuchac
Black tea
Drinking
water
Oral
3yr
Drinking
longitudinal
water
1 mo
Drinking
water
Route
Human
Mouse
Rat
Rat
Species
Observations
Reduced TBA reactive substance

formation in frontal cortex,
occipital cortex, hippocampus,
and cerebellum
Cortical injection
Increased mitochondrial
with 01 M ferric
superoxide dismutase activity
chloride
of striatum and midbrain,
decreased TBA reactive
substances in cortex and
cerebellum of aged rats and
8-OHdG formation in cortex
Normal aging
Inhibited weight gain, increased
environmental awareness and
responsiveness and life span
Caused transient improvements

in performance, prevented
steady decline throughout day in
alertness and cognitive capacity,
increased alertness and
information processing capacity
Normal aging
Stress type
[137]
[136]
[135]
[134]
References
Aqueous extracts.
12-O-tetradecanoylphorbol-13-acetate.
c
Green tea supplementation preceded each MPTP injection.
d
Contains green tea extract as a main component, with ascorbic acid, sunower seed extract, dunaliella carotene, and natural vitamin E.
e
German form of the Japanese name for a lightly fermented tea beverage.
GTPase, guanosine triphosphatase; MPTP, N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine; TPA, ATPase, adenosine triphosphatase; DNA, deoxyribonucleic acid; TBA, 8-OHdG.
1 mL/kg
body weight
h-Catechinb
21 mo
Chronic
Rooibos tea
(Aspalathus linearis)
Duration
Concentration
Flavonoid source
Table 1 (continued )
212
Youdim and Joseph
213
Table 2 Neuroprotection and Neuroprotective Actions Afforded by Dietary and

Phytochemical Mixtures to the Brain
Properties
Herbs
S-113a and DX-9386b
Ding lang
(Policies fruticosum L.)
Qizhu tangc
TJ-960 (baicaleind)
. Improves cognitive performance

. Reduce age-related decits in behavioral
References
[138145]
[146150]
performance
. Modulates endogenous SOD levels

. Prevents cerebral oxidative injury in rats
. Reduces FeCl3-induced epilepsy, mediates
[151]
[152157]
increases in peroxidation by-products
. Inhibits cerebral ischemiainduced hippocampal
Guilingji (baicaleind)
Shou Xing Bu Zhi
Oren-gedoku-to (TJ-15)
and Toki-shakuyaku-san
(TJ-23)
Aged garlic extracte

(Allium sativum)
.
.
.
.
.
.
.
.
.
Spices
Curcumin and tusmeric
neuronal death
Modulates neurotransmitter levels, NOS activity
Protects against FeCl3-induced epilepsy
Reduces age-related increases in brain lipofuscin
Ameliorates cerebral ischemiainduced and
age-related learning and memory decits
Prevents reduction of acetylcholine content
in brain cerebral cortex, hippocampus, and
striatum; modulates neurotransmitter levels
Improves cognitive impairments in SAM
Protects against morphological changes in SAM
Reduces age-related increases in peroxidation
by-products
Exhibits immunomodulatory properties
Saron and crocinf
. Protects against retinal induced peroxidation

. Enhances glutathione levels
. Protects against ethanol-induced impairments
Red bell pepper

(Capsicum annuum L.)
in learning, inhibition of hippocampal LTP,

and synaptic plasticity
. Improves memory and acquisition performance
in SAM
[158]
[159,160]
[161169]
[170176]
[177,178]
[179,180]
[181]
Consists of Biota orientalis, Panax ginseng, and Schizandra chinesis.

Consists of Panax ginseng, Polygala tenuifola, Acorus gamineus, and Paoria cocas.
c
Consists of Rhizoma atractylodis, Poria, Radix notoginseng, and Radax astragali.
d
Major component found in extract believed to promote properties.
e
Contains S-allycisteine, S-allymercaptocysteine, allicin, and diallosuldes.
f
Crocetin di-gentobiose ester.
SOD, superoxide dismutase; NOS, nitric oxide synthase; SAM, senescence accelerated mice.
Source: Adapted from Ref. 182.
b
214
Youdim and Joseph
memory [35] and that treatment with PKC inhibitors impairs memory formation
[36]. It appears that training induces the calcium-induced translocation [37] of
PKC from the cytosol to the membrane subcellular fraction [38]. However, in
aging, there appear to be alterations in this translocation, which are correlated
with decrements in spatial memory. In this regard, Colombo and coauthors [37]
showed that young rats with the best performance in spatial memory also had the
highest PKC-g in the membrane fraction of the hippocampus and PKC-h2 in the
soluble cytosolic fraction.
Also important are the mitogen-activated protein (MAP) kinases (MAPKs),
critical in long-term memory formation. More specically there is a great amount
of data indicating that MAPKs are involved not only in hippocampal memory
formation but also in memory modulation in other brain structures [39]. Moreover, recent studies have indicated that the activation of these molecules is
sensitive to oxidative stress (see Chap. 9) and that they may serve as biochemical
signal integrators and molecular detectors for modulating coordinated responses
to extracellular signals in neurons, playing a critical role in synaptic plasticity.
Particularly important in this regard are the extracellular signal regulated kinases
(ERKs) 1/2 and the Jun kinases (JNKs). Studies have demonstrated the role of
ERK signaling cascades in diverse types of learning and memory such as
conditioned taste aversion [40], novel taste learning [41], spatial learning [42],
and inhibitory avoidance [43]. Findings have shown that ERK activities were
reduced in cortical brain slices of senescent rats (24 months) without declines in
the corresponding proteins [44]. With respect to this, preliminary data from our
lab (Joseph et al., unpublished) indicated that mice transgenic for amyloid
precursor protein and presenilin-1 mutations, maintained on blueberry supplemented diet from the time of weaning up until 12 months of age, showed Y-maze
performance equivalent to that of nontransgenic controls; that result correlated
signicantly with decreased sphingomyelin turnover and increased ERK, PKC,
and GTPase activities in both the striatum and the hippocampus. These ndings
suggest that blueberry extract may benet other parameters of cell signaling and
synaptic plasticity that are involved in learning and memory. Although there is do
direct evidence to suggest that blueberry avonoids directly mediate these eects
from within the CNS, recent in vitro studies appear to support the notion that
MAPK signaling can be inuenced by avonoids, such as those found in tea and
wine (see Chap. 9).
V.
NEUROPROTECTIVE EFFECTS OF FLAVONOID-RICH

BEVERAGES
The neuroprotective actions of polyphenolic-rich beverages, such as teas and red

wine, have also been investigated. Studies that examined the neuroprotective
215
properties of tea are shown in Table 1. These experiments were performed using
either simple aging models or the cerebral ischemia/reperfusion model, which
reproduces a number of pathophysiological features observed in age- and
disease-related brain dysfunction [45]. From these ndings (Table 1) one could
argue that protection aorded by tea polyphenolics against the various types of
decits induced by ischemic damage, N-methyl-4-phenyl-1,2,3,6-tetra hydropyridine (MPTP) or iron chloride, for example, could be due in part to inhibition
of oxidative and inammatory processes [4649], both of which inuence
behavior [50].
Weak associations between tea consumption and neurological disorders
such as Parkinsons disease have also been suggested [51]. Although the possible
mechanisms involved are unclear, tea avonoids have been shown to protect
against iron-induced decits in striatal neurotransmitter concentrations/turnover
[52], as well as to decrease oxidative damage in the substantia nigra [53], both of
which are common etiological features associated with Parkinsons disease.
However, in these studies tea avonoids were administered intravenously or by
infusion directly into the brain, and hence it is not clear how the outcomes relate
to the normal ingestion of tea per se and the consequences of its interactions and
biotransformations in the gastrointestinal tract before uptake.
Although the active phenolic component(s) in tea extract aording the
protection described are not known, potential candidates include the metabolites
of the major components catechin and epigallocatechin gallate, which supplemented alone aord protection against ischemia/reperfusion-induced memory
impairment [54] and cell death of hippocampal CA1 neurons [55]. Moreover,
Levites and coworkers [49] have also shown that decits induced by the
neurotoxin MPTP, which specically induce neurodegeneration in dopaminergic
neurons similar to that observed in Parkinsons disease, can be ameliorated by
green tea extract (0.5 and 1 mg/kg body weight). They also reported that
supplementation of epigallocatechin gallate alone provided signicant protection, which may account for the neuroprotective properties of the extract itself.
However, epigallocatechin gallate was supplemented at higher concentrations
(2 and 10 mg/kg body weight) than that present in the extract itself, suggesting
possible synergistic interactions of the various components of the tea extract.
However, although epigallocatechin gallate had previously been shown to cross
the blood-brain barrier [21], the studies described did not determine the
localization of the avonoids in the brain; hence one could only hypothesize
that their protective actions were mediated directly within the CNS and not
from the periphery.
Correlations between moderate wine consumption (34 glasses/day) and
the incidence of dementia and AD compared with that of nondrinkers [56,57]
have also been reported, although some argue the contrary [58]. Reports have
also shown that heavy drinkers displayed the poorest results on memory or
216
Youdim and Joseph
on intelligence tasks as compared with moderate drinkers or nondrinkers [59].

Orgogozo and colleagues [57] propose several possible protective mechanisms of
action regarding moderate wine consumption, including antioxidant and/or antiinammatory properties, as well as elevation of plasma apolipoprotein E levels.
Although there have been few direct demonstrations that the protective eects
associated with wine consumption are due to the avonoid content, there is some
evidence that they are not due to the alcohol content [60].
Attempts to elucidate which possible component(s) contributes neuroprotection have been focused on the stilbene resveratrol. This is found in several
types of red wine and is considered one of the substances responsible for the
lower incidence of coronary heart diseases among moderate drinkers and has
been proposed to exhibit neuroprotective properties. Virgili and Contestabile
[61] reported that long-term administration of resveratrol to young adult rats
protected them from damage caused by systemic injection of the excitotoxin
kainic acid, in the olfactory cortex and the hippocampus. The same treatment,
however, is not able to give any signicant protection in an ex vivo model of
simulated ischemia on hippocampal slices. Although a direct demonstration of
the protective eects of wine or a delineation of its mechanisms has not been
shown, moderate wine drinking appears to promote some degree of protection
against senile dementia or AD.
VI.
HERBS AND SPICES
Alternative or complementary medicines are widely used in North America.

Consumers in the United States spend an estimated $1.5 billion per year on
herbal medicines with projected annual growth of 15%. Germany, the largest
market among American or Western European countries, had total sales in 1993
of $1.9 billion for plant-based allopathic medicines (half of these prescribed by
physicians) and 5 million prescriptions for ginkgo in 1988. The use of these
complementary medicines in dementia therapy varies according to the dierent
cultural traditions. The two most commonly used preparations, Gingko biloba
and ginseng, are discussed in the following sections. Further neuroprotective
herbs and spices, including preparations that contain Gingko biloba and ginseng,
are highlighted in Table 2.
A.
Ginkgo biloba (EGb 761)
By far the most extensively studied herbal mixture with respect to brain function
is Ginkgo biloba (EGb 761) [6264]. Extracts of the leaves have been used for
5000 years in traditional Chinese medicine for various purposes. Flavonoids such
as myricetin, quercetin, and kaempferol are general components in Ginkgo biloba
217
extract, which is often standardized to contain 24% ginkgo-avone glycosides

and 6% terpenoids including bilobalide and the ginkgolides A, B, C, M, and J
(20-carbon cage molecules with six ve-member rings).
A number of epidemiological studies have implicated Ginkgo biloba as a
potential neuroprotective mixture against age-related and AD-related dementia
[6567]. Although the cause and underlying pathophysiological features of AD
are unknown, prominent hypotheses as to the cause center around age-related
oxidative injury and inammatory processes [9] and propose that attenuation of
these processes by Ginkgo biloba underlies its neuroprotective actions. Further
actions promoted by Gingko biloba that may contribute to improvement in
dementia and other brain aging functions include the reduction in levels of ROS
[68,69]; reduction of age-related decits in oxygen and glucose delivery to the
brain by increasing cerebral blood ow [7075] and protection of membrane
fatty acids [7679], changes in which have been shown to correlate with
cognitive performance [80,81], interactions with the muscarinic cholinergic
system [82,83] involved in the performance of spatial tasks [84,85], antagonism
of platelet activating factor [8688]; protection of the striatal dopaminergic
system [89,90]; and inhibition of monoamine oxidase activity [9193], hence
maintaining levels of monoamines such as NE, 5-HT, and DA, which are known
to play essential roles in a variety of brain functions. However, these ndings
still dictate a signicant amount of validation before Ginkgo biloba is fully
accepted as a therapy against dementia. This progression is complicated further
in light of the fact that the relative content of ginkgolide and bilobalide
components and those of the various avonoids varies across preparations and
even seasons [94].
Although these observations highlight possible neuroprotective action by
Gingko biloba, Oken and colleagues [95] reported in 1998 that after closer
examination of the epidemiological data according to certain acceptable scientic criteria, only four studies suggest a potential protective eect of the extract.
As such, further validation is required before these herbal remedies are fully
accepted as therapies against dementia.
B.
Ginseng
Panax ginseng is one of the mostly widely used herbs in traditional Chinese
medicine. In addition to controlling functions related to stamina, fatigue, and
physical stress, ginseng has been reported to reduce neuronal death and protect
against ischemic damage [9698]. Possible mechanisms of neuroprotection
proposed include ability to increase the expression of nerve growth factor [99]
and intrinsic antioxidant activity [100,101]. Ginseng has also been investigated as
a potential therapy against amyotrophic lateral sclerosis (ALS), a motor neuron
disease [102]. Among the observations made in rats given ginseng in drinking
218
Youdim and Joseph
water were prolongation in onset of signs of motor impairment and survival. This
improvement in memory performance is in agreement with studies describing the
benecial eects of ginseng on memory and learning performance [103107]. In
addition, ginsengs ability to modulate the cholinergic and serotoninergic neurotransmitter systems, damage to which aects spatial working memory [108,109],
has also been proposed as a mechanism of neuroprotection [100]. In this regard,
behavioral paradigms aected by electroconvulsive shock through modulation of
the cholinergic neurotransmitter system are attenuated by ginseng supplementation [110], in particular within brain areas such as the hippocampus, known to be
involved in spatial memory tasks [111,112]. It has also been speculated that
ginseng acts by enhancing cholinergic systems such as choline acetyltransferase
important in the formation of memory [113].
Studies have also shown that behavioral impairments can be forestalled
by using individual ginsenosides [97,107,114122]. The observations made
after the application of these individual components appear to support the
notion that neuroprotection may result from interactions with neurotransmitter
systems. For example, ginsenosides have been shown to increase muscariniccholinergic receptor density and levels of acetylcholine in the brain [123].
Unfortunately, despite these observations few epidemiological studies have
been performed to complement these ndings. Indeed, a comprehensive
survey of the literature found only ve studies investigating the eects of
ginseng on human cognitive performance [124128], in three of which
signicant improvement in mental arithmetic and abstraction tests were
reported [125127].
VII.
CONCLUSIONS
At the present time, for some neurodegenerative disorders there is very little in
the form of treatment and what treatments are available are only eective for a
short period and are often associated with debilitating side eects. Current drug
therapy does not address the progressive nature of many of these diseases, and
ultimately the patient becomes severely disabled and requires nursing/hospital
care. In light of this it appears essential that novel strategies with potential to
delay the onset or even prevent the manifestation of certain processes believed
to contribute to neurological dysfunction be developed. As such renewed
attention is being paid to the application of avonoids commonly found in
fruits, vegetables, and beverages such as tea and wine. The ndings from these
studies highlight their ability to aord neuroprotection, yet evidence of direct
action within the brain is lacking. Only a few studies to date have reported that
avonoids localize within the brain. Although further studies are clearly
required to support these ndings, a more cautious approach must also be taken
219
when attempting to elucidate potential mechanisms of action. Although use of

the native avonoid in in vitro studies has some relevance to the in vivo
scenario, the application of the predominant in vivo physiological metabolites
that enter the circulation will undoubtedly prove more fruitful in examining
possible mechanisms of action.
REFERENCES
1.
Ingram DK, Jucker M, Spangler E. Behavioral manifestations of aging. In: Mohr

U, Cungworth DL, Capen CC, eds. Pathobiology of the Aging Rat. Washington,
DC: ILSI, 1994:149170.
2. Shukitt-Hale B, Mouzakis G, Joseph JA. Psychomotor and spatial memory
performance in aging male Fischer 344 rats. Exp Gerontol 1998; 33:615624.
3. Bartus RT. Drugs to treat age-related neurodegenerative problems: the nal
frontier of medical science? J Am Geriat Soc 1990; 38:680695.
4. Muir JL. Acetylcholine, aging, and Alzheimers disease. Pharmacol Biochem
Behav 1997; 56:687696.
5. West RL. An application of pre-frontal cortex function theory to cognitive aging.
Psychol Bull 1996; 120:272292.
6. Joseph JA, Hunt WA, Rabin BM, Dalton TK. Possible accelerated striatal aging
induced by 56Fe heavy-particle irradaition: implications for manned space ights.
Radiat Res 1992; 130:8893.
7. Bickford PC. Motor learning decits in aged rats are correlated with loss of
cerebellar noradrenergic function. Brain Res 1993; 620:133138.
8. Joseph JA, Villalobos-Molina R, Denisova N, Erat S, Cutler R, Strain JG. Age
dierences in sensitivity to H2O2 or NO-induced reductions in K+-evoked
dopamine release from superfused striatal slices: reversals by PBN or Trolox. Free
Radic Biol Med 1996; 20:821830.
9. Christen Y. Oxidative stress and Alzheimer disease. Am J Clin Nutr 2000;
71:621S629S.
10. Urano S, Sato Y, Otonari T, Makabe S, Susuki S, Ogata M, Endo T. Aging and
oxidative stress in neurodegeneration. Biofactors 1998; 7:103112.
11. Shukitt-Hale B. The eects of aging and oxidative stress on psychomotor and
cognitive behavior. Age 1999; 22:917.
12. Halliday G, Robinson SR, Shepherd C, Kril J. Alzheimers disease and
inammation: a review of cellular and therapeutic mechanisms. Clin Exp
Pharmacol Physiol 2000; 27:18.
13. Combs CK, Johnson DE, Karlo JC, Cannady SB, Landreth GE. Inammatory
mechanisms in Alzheimers disease: inhibition of beta-amyloid-stimulated
proinammatory responses and neurotoxicity by PPARgamma agonists. J Neurosci 2000; 20:558567.
14. Benzi G, Moretti A. Age- and peroxidative stress-related modications of the
cerebral enzymatic activities linked to mitochondria and the glutathione system.
220
15.
Youdim and Joseph
Egashira T, Takayama F, Yamanaka Y. Eects of bifemelane on muscarinic

receptors and choline acetyltransferase in the brains of aged rats following chronic
cerebral hypoperfusion induced by permanent occlusion of bilateral carotid
arteries. Jpn J Pharmacol 1996; 72:5765.
16. Joseph JA, Berger RE, Engel BT, Roth GS. Age-related changes in the nigrostriatum: a behavioral and biochemical analysis. J Gerontol 1978; 33: 643
649.
17. Gould TJ, Bickford PC. Age-related decits in the cerebellar beta-adrenergic
signal transduction cascade in Fischer 344 rats. J Pharmacol Exp Ther 1997;
281:965971.
18. Kornhuber J, Schoppmeyer K, Bendig C, Riederer P. Characterization of [3H]
pentazocine binding sites in post-mortem human frontal cortex. J Neural Transm
1996; 103:4553.
19. Youdim KA, Martin A, Joseph JA. Essential fatty acids and the brain: possible
health implications. Int J Dev Neurosci 2000; 18:382399.
20. Landeld PW, Eldridge JC. The glucocorticoid hypothesis of age-related
hippocampal neurodegeneration: role of dysregulated intraneuronal Ca2+. Ann
NY Acad Sci 1994; 746:308321.
21. Suganuma M, Okabe S, Oniyama M, Tada Y, Ito H, Fujiki H. Wide distribution of
[3H](-)-epigallocatechin gallate, a cancer preventive tea polyphenol, in mouse
tissue. Carcinogenesis 1998; 19:17711776.
22. Tsai T-H, Chen Y-F. Determination of unbound hesperetin in rat blood and brain
by microdialysis coupled to microbore liquid chromatography. J Food Drug Anal
2000; 8:331336.
23. Peng HW, Cheng FC, Huang YT, Chen CF, Tsai TH. Determination of naringenin
and its glucuronide conjugate in rat plasma and brain tissue by high-performance
liquid chromatography. J Chromatogr A 1998; 714:369374.
24. Schroder-van der Elst JP, van der Heide D, Rokos H, Morreale de Escobar G, Kohrle
J. Dierent tissue distribution, elimination and kinetics of thyroxine and its
conformational analog, the synthetic avonoid Emd 49209 in the rat. Endocrinology
1997; 138:7984.
25. Schroder-van der Elst JP, van der Heide D, Rokos H, Morreale de Escobar G,
Kohrle J. Synthetic avonoids cross the placenta in the rat and are found in the
fetal brain. Am J Physiol 1998; 274:E253E256.
26. Joseph JA, Strain JG, Jimenez ND, Fisher D. Oxidant injury in PC12 cellsa
possible model of calcium deregulation in aging. I. Selectivity of protection
against oxidative stress. J Neurochem 1997; 69:12521258.
27. Joseph JA, Shukitt-Hale B, McEwen J, Rabin B. Magnesium activation of GTP
hydrolysis or incubation in S-Adenosyl-1-methionine reverses 56Fe-induced
decrements in oxotremorine-enhancement of K+ evoked striatal dopamine release.
Radiat Res 1999; 152:637641.
28. Joseph JA, Kowatch MA, Maki T, Roth GS. Selective cross activation/inhibition
of second messenger systems and the reduction of age-related decits in the
muscarinic control of dopamine release from perfused rat striata. Brain Res 1990;
537:4048.

29.
221
Joseph JA, Bartus RT, Clody DE, Morgan D, Finch C, Beer B, Sesack S.
Psycomotor performance in the senescent rodent: reduction of decits via striatal
dopamine receptor up-regulation. Neurobiol Aging 1983; 4:313319.
30. Bickford PC, Shukitt-Hale B, Joseph JA. Eects of aging on cerebellar
noradrenergic function and motor learning: nutritional interventions. Mech Ageing
Dev 1999; 111:141154.
31. Bickford PC, Gould T, Briederick L, Chadman K, Pollock A, Young D, ShukittHale B, Joseph J. Antioxidant-rich diets improve cerebellar physiology and motor
learning in aged rats. Brain Res 2000; 866:211217.
32. Joseph JA, Shukitt-Hale B, Denisova NA, Bielinski D, Martin A, McEwen JJ,
Bickford PC. Reversals of age-related declines in neuronal signal transduction,
cognitive and motor behavioral decits with diets supplemented with blueberry,
spinach or strawberry dietary supplementation. J Neurosci 1999; 19:81148121.
33. Joseph JA, Shukitt-Hale B, Denisova NA, Prior RL, Cao G, Martin A, Taglialatela
G, Bickford PC. Long-term dietary strawberry, spinach, or vitamin E supplementation retards the onset of age-related neuronal signal-transduction and
cognitive behavioral decits. J Neurosci 1998; 18:80478055.
34. Youdim KA, Shukitt-Hale B, Martin A, Wang H, Denisova N, Bickford PC,
Joseph JA. Short-term dietary supplementation of blueberry polyphenolics:
benecial eects on aging brain performance and peripheral tissue function. Nutr
Neurosci 2000; 3:383397.
35. Micheau J, Riedel G. Protein kinases: which one is the memory molecule? Cell
Mol Life Sci 1999; 55:534548.
36. Serrano PA, Beniston DS, Oxonian MG, Rodriguez WA, Rosenzweig MR,
Bennett EL. Dierential eects of protein kinase inhibitors and activators on
memory formation in the 2 day old chick. Behav Neural Biol 1994; 61:6072.
37. Colombo PJ, Wetsel W, Gallagher MG. Spatial memory is related to hippocampal
subcellular concentrations of calcium-dependent protein kinase C isoforms in
young and aged rats. Proc Natl Acad Sci USA 1997; 94:14951499.
38. Van der Zee EA, Compaan JC, Bohus B, Luiten PG. Alterations in the
immunoreactivity for muscarinic acetylcholine receptors and colocalized PKC
gamma in mouse hippocampus induced by spatial discrimination learning.
Hippocampus 1995; 5:349362.
39. Sgambato V, Pages C, Rogard M, Besson MJ, Caboche J. Extracellular signal
regulated kinase (ERK) controls immediate early gene induction on corticostriatal
stimulation. J Neurosci 1998; 18:88148825.
40. Berman DE, Hazvi S, Rosenblum K, Seger R, Dudai Y. Specic and dierential
activation of mitogen-activated protein kinase cascades by unfamiliar taste in the
insular cortex of the behaving rat. J Neurosci 1998; 18:1003710044.
41. Swank MW, Sweatt JD. Increased histone acetyltransferase and lysine acetyltransferase activity and biphasic activation of the ERK / RSK cascade in insular
cortex during novel taste learning. J Neurosci 2001; 21:33833391.
42. Selcher JC, Atkins CM, Trzaskos JM, Paylor R, Sweatt JD. A necessity for
MAP kinase activation in mammalian spatial learning. Learn Mem 1999; 6:478
490.
222
Youdim and Joseph
43.
Schafe GE, Nadel NV, Sullivan GM, Harris A, LeDoux JE. Memory consolidation
for contextual and auditory fear conditioning is dependent on protein synthesis,
PKA, and MAP kinase. Learn Mem 1999; 6:97110.
Zhen X, Uryu K, Cai G, Johnson GP, Friedman E. Age-associated impairment in
brain MAPK signal pathways and the eect of restriction in Fischer 344 rats.
J Gerontol A Biol Sci Med Sci 1999; 54:B539B548.
Heiss WD. Therapy of cerebral ischemia. Zr Kardiol 1987; 76:8798.
Hofbauer R, Frass M, Gmeiner B, Handler S, Speiser W, Kapiotis S. The green tea
extract epigallocatechin gallate is able to reduce neutrophil transmigration through
monolayers of endothelial cells. Wiener Klin Wochenschr 1999; 111:278282.
Lin YL, Lin JK. (-)-Epigallocatechin-3-gallate blocks the induction of nitric oxide
synthase by down-regulating lipopolysaccharide-induced activity of transcription
factor nuclear factor-kappaB. Mol Pharmacol 1997; 52:465472.
Lin YL, Tsai SH, Lin-Shiau SY, Ho CT, Lin JK. Theaavin-3,3V-digallate from
black tea blocks the nitric oxide synthase by down-regulating the activation of NFkappaB in macrophages. Eur J Pharmacol 1999; 367:379388.
Levites Y, Weinreb O, Maor G, Youdim MBH, Mandel S. Green tea polyphenol (-)
epigallocatechin-3-gallate prevents N-methyl-4-phenyl-1,2,3,6-tetrahydropyridineinduced dopaminergic neurodegeneration. J Neurochem 2001; 78:10731082.
Hauss-Wegrzyniak B, Vannucchi MG, Wenk GL. Behavioral and ultrastructural
changes in by chronic neuroinammation in young rats. Brain Res 2000; 859:
157166.
Hellenbrand W, Boeing H, Robra BP, Siedler A, Vieregge P, Nischan P, Joerg J,
Oertel WH, Schneider E, Ulm G. Diet and Parkinsons disease II: a possible role
for the past intake of specic nutrients. Neurology 1996; 47:644650.
Kabuto H, Yokoi I, Mori A. Monoamine metabolites, iron induced seizures, and
the anticonvulsant eect of tannins. Neurochem Res 1992; 17:585590.
Lin AM, Chyi BY, Wu LY, Hwang LS, Ho LT. The antioxidative property of
green tea against iron-induced oxidative stress in rat brain. Chin J Physiol 1998;
41:189194.
Matsuoka Y, Hasegawa H, Okuda S, Muraki T, Uruno T, Kubota K. Ameliorative
eects of tea catechins on active oxygen-related nerve cell injuries. J Pharmacol
Exp Therapeut 1995; 274:602608.
Lee S, Suh S, Kim S. Protective eects of the green tea polyphenol (-) epigallocatechin gallate against hippocampal neuronal damage after transient global
ischemia in gerbils. Neurosci Lett 2000; 287:191194.
Launer LJ, Kalmijn S. Anti-oxidants and cognitive function: a review of clinical
and epidemiologic studies. J Neural Transm Suppl 1998; 53:18.
Orgogozo JM, Dartigues JF, Lafont S, Letenneur L, Commenges D, Salamon R,
Renaud S, Breteler MB. Wine consumption and dementia in the elderly: a
prospective community study in the Bordeaux area. Rev Neurol 1997;153: 185
192.
Leibovici D, Ritchie K, Ledesert B, Touchon J. The eects of wine and tobacco
consumption on cognitive performance in the elderly: a longitudinal study of
relative risk. Int J Epidemiol 1999; 28:7781.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.

59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
223
De Renzi E, Faglioni P, Nichelli P, Pignattari L. Intellectual and memory

impairment in moderate and heavy drinkers. Cortex 1984; 20:525533.
Puddey IB, K.D. C, Abdu-Amsha Caccetta R, Beilin LJ. Alcohol, free radicals and
antioxidants. Novartis Found Symp 1998; 216:5162.
Virgili M, Contestabile A. Partial neuroprotection of in vivo excitotoxic brain
damage by chronic administration of the red wine antioxidant agent, transresveratrol in rats. Neurosci Lett 2000; 281:123126.
Diamond BJ, Shieett SC, Feiwel N, Matheis RJ, Noskin O, Richards JA,
Schoenberger NE. Ginkgo biloba extract: mechanisms and clinical indications.
Arch Phys Med Rehab 2000; 81:668678.
Clostre F. Ginkgo biloba extract (EGb 761): state of knowledge in the dawn of the
year 2000. Ann Pharm Fr 1999; 57:1S888.
Kidd PM. A review of nutrients and botanicals in the integrative management of
cognitive dysfunction. Altern Med Rev 1999; 4:144161.
Simanyi M. Use of special Ginkgo biloba extract for cognitive disorders in the
elderly. Wiener Med Wochenschr 1999; 149:231234.
Le Bars PL, Kieser M, Itil KZ. A 26-week analysis of a double-blind, placebocontrolled trial of the ginkgo biloba extract EGb 761 in dementia. Dement Geriat
Cogn Disord 2000; 11:230237.
Le Bars PL, Katz MM, Berman N, Itil TM, Freedman AM, Schatzberg AF.
A placebo-controlled, double-blind, randomized trial of an extract of Ginkgo
biloba for dementia: North American EGb Study Group. JAMA 1997; 278:1327
1332.
Oyama Y, Chikahisa L, Ueha T, Kanemaru K, Noda K. Ginkgo biloba extract
protects brain neurons against oxidative stress induced by hydrogen peroxide.
Brain Res 1996; 712:349352.
Seif-El-Nasr M, El-Fattah AA. Lipid peroxide, phospholipids, glutathione levels
and superoxide dismutase activity in rat brain after ischaemia: eect of ginkgo
biloba extract. Pharmacol Res 1995; 32:273278.
Eckmann F. Cerebral insuciency treatment with Ginkgo-biloba extract: Time
of onset of eect in a double-blind study with 60 inpatients. Fortschr der Med
1990; 108:557560.
Gajewski A, Hensch SA. Ginkgo biloba and memory for a maze. Psychol Rep
1999; 84:481484.
Krieglstein J, Beck T, Seibert A. Inuence of an extract of Ginkgo biloba on
cerebral blood ow and metabolism. Life Sci 1986; 39:23272334.
Lamm K, Arnold W. The eect of blood ow promoting drugs on cochlear blood
ow, perilymphatic pO(2) and auditory function in the normal and noise-damaged
hypoxic and ischemic guinea pig inner ear. Hear Res 2000; 141:199219.
Oberpichler H, Beck T, Abdel-Rahman MM, Bielenberg GW, Krieglstein J.
Eects of Ginkgo biloba constituents related to protection against brain damage
caused by hypoxia. Pharmacol Res Comm 1988; 20:349368.
Rapin JR, Le Poncin Latte M. Cerebral glucose consumption: the eect of
Ginkgo biloba extract. Presse Med 1986; 15:14941497.
al-Zuhair H, Abd el-Fattah A, el-Sayed MI. The eect of meclofenoxate with
224
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
Youdim and Joseph

Ginkgo biloba extract or zinc on lipid peroxide, some free radical scavengers and
the cardiovascular system of aged rats. Pharmacol Res 1998; 38:6572.
Klein J, Chatterjee SS, Loelholz K. Phospholipid breakdown and choline release
under hypoxic conditions: inhibition by bilobalide, a constituent of Ginkgo biloba.
Brain Res 1997; 755:347350.
Rabin O, Drieu K, Grange E, Chang MC, Rapoport SI, Purdon AD. Eects of EGb
761 on fatty acid reincorporation during reperfusion following ischemia in the
brain of the awake gerbil. Mol Chem Neuropathol 1998; 34:79101.
Stoll S, Scheuer K, Pohl O, Muller WE. Ginkgo biloba extract (EGb 761)
independently improves changes in passive avoidance learning and brain
membrane uidity in the aging mouse. Pharmacopsychiatry 1996; 29:144149.
Suzuki H, Park SJ, Tamura M, Ando S. Eect of the long-term feeding of dietary
lipids on the learning ability, fatty acid composition of brain stem phospholipids
and synaptic membrane uidity in adult mice: a comparison of sardine oil diet with
palm oil diet. Mech Ageing Dev 1998; 16:119128.
Scheuer K, Rostock A, Bartsch R, Muller WE. Piracetam improves cognitive
performance by restoring neurochemical decits of the aged rat brain.
Pharmacopsychiatry 1999; 32:1016.
Chopin P, Briley M. Eects of four non-cholinergic cognitive enhancers in
comparison with tacrine and galanthamine on scopolamine-induced amnesia in
rats. Psychopharmacology 1992; 106:2630.
Deberdt W. Interaction between psychological and pharmacological treatment in
cognitive impairment. Life Sci 1994; 55:20572066.
Puumala T, Greijus S, Narinen K, Haapalinna A, Riekkinen P Sr, Sirvio J.
Stimulation of alpha-1 adrenergic receptors facilitates spatial learning in rats. Eur
Neuropsychopharmacol 1998; 8:1726.
Packard MG, Teather LA. Posttraining estradiol injections enhance memory in
ovariectomized rats: cholinergic blockade and synergism. Neurobiol Learn Mem
1997; 68:172188.
Akisu M, Kultursay N, Coker I, Huseyinov A. Platelet-activating factor is an
important mediator in hypoxic ischemic brain injury in the newborn rat: unarizine
and Ginkgo biloba extract reduce PAF concentration in the brain. Biol Neonate
1998; 74:439444.
Smith PF, Maclennan K, Darlington CL. The neuroprotective properties of the
Ginkgo biloba leaf: a review of the possible relationship to platelet-activating
factor (PAF). J Ethnopharmacol 1996; 50:131139.
Spinnewyn B, Blavet N, Clostre F, Bazan N, Braquet P. Involvement of plateletactivating factor (PAF) in cerebral post-ischemic phase in Mongolian gerbils.
Prostaglandins 1987; 34:337349.
Ramassamy C, Clostre F, Christen Y, Costentin J. Prevention by a Ginkgo biloba
extract (GBE 761) of the dopaminergic neurotoxicity of MPTP. J Pharm
Pharmacol 1990; 42:785789.
Ramassamy C, Christen Y, Clostre F, Costentin J. The Ginkgo biloba extract,
EGb761, increases synaptosomal uptake of 5-hydroxytryptamine: in-vitro and exvivo studies. J Pharm Pharmacol 1992; 44:943945.

91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
225
White HL, Scates PW, Cooper BR. Extracts of Ginkgo biloba leaves inhibit
monoamine oxidase. Life Sci 1996; 58:13151321.
Wu WR, Zhu XZ. Involvement of monoamine oxidase inhibition in neuroprotective
and neurorestorative eects of Ginkgo biloba extract against MPTP-induced
nigrostriatal dopaminergic toxicity in C57 mice. Life Sci 1999; 65:157164.
Pardon MC, Joubert C, Perez-Diaz F, Christen Y, Launay JM, Cohen-Salmon C. In
vivo regulation of cerebral monoamine oxidase activity in senescent controls and
chronically stressed mice by long-term treatment with Ginkgo biloba extract (EGb
761). Mech Ageing Dev 2000; 113:157168.
Sticher O. Quality of Ginkgo preparations. Planta Medica 1993; 59:211.
Oken BS, Storzbach DM, Kaye JA. The ecacy of Ginkgo biloba on cognitive
function in Alzheimer disease. Arch Neurol 1998; 55:14091415.
Choi SR, Saji H, Iida Y, Magata Y, Yokoyama A. Ginseng pretreatment protects
against transient global cerebral ischemia in the rat: measurement of local cerebral
glucose utilization by [14C]deoxyglucose autoradiography. Biol Pharma Bull
1996; 19:644646.
Wen TC, Yoshimura H, Matsuda S, Lim JH, Sakanaka M. Ginseng root prevents
learning disability and neuronal loss in gerbils with 5-minute forebrain ischemia.
Acta Neuropathol 1996; 91:1522.
Zhang YG, Liu TP. Ineuences of ginsenosides Rb1 and Rg1 on reversible focal
brain ischemia in rats. Chung Kuo Yao Li Hsueh Pao 1996; 17:4448.
Salim KN, McEwen BS, Chao HM. Ginsenoside Rb1 regulates ChAT, NGF and
trkA mRNA expression in the rat brain. Brain Res Mol Brain Res 1997; 47:177182.
Kitts DD, Wijewickreme AN, Hu C. Antioxidant properties of a North American
ginseng extract. Mol Cell Biochem 2000; 203:110.
Keum YS, Park KK, Lee JM, Chun KS, Park JH, Lee SK, Kwon H, Surh YJ.
Antioxidant and anti-tumor promoting activities of the methanol extract of heatprocessed ginseng. Cancer Lett 2000; 150:4148.
Jiang F, DeSilva S, Turnbull J. Benecial eect of ginseng root in SOD-1 (G93A)
transgenic mice. J Neurol Sci 2000; 180:5254.
Jaenicke B, Kim EJ, Ahn JW, Lee HS. Eect of Panax ginseng extract on passive
avoidance retention in old rats. Arch Pharmacol Res 1991; 14:2529.
Lasarova MB, Mosharrof AH, Petkov VD, Markovska VL, Petkov VV. Eect of
piracetam and of standardized ginseng extract on the electroconvulsive shockinduced memory disturbances in step-down passive avoidance. Acta Physiol
Pharmacol Bulg 1987; 13:1117.
Petkov VD, Mosharrof AH. Eects of standardized ginseng extract on learning,
memory and physical capabilities. Am J Chin Med 1987; 15:1929.
Petkov VD, Kehayov R, Belcheva S, Konstantinova E, Petkov VV, Getova D,
Markovska V. Memory eects of standardized extracts of Panax ginseng (G115),
Ginkgo biloba (GK 501) and their combination Gincosan (PHL-00701). Planta
Med 1993; 59:106114.
Petkov VD, Cao Y, Todorov I, Lazarova M, Getova D, Stancheva S, Alova L.
Behavioral eects of stem-leaves extract from panax ginseng C.A. Meyer. Acta
Physiol Pharmacol Bulg 1992; 18:4148.
226
Youdim and Joseph
108.
Lehmann O, Jeltsch H, Lehnardt O, Pain L, Lazarus C, Cassel JC. Combined

lesions of cholinergic and serotonergic neurons in the rat brain using 192 IgGsaporin and 5,7-dihydroxytryptamine: neurochemical and behavioural characterization. Eur J Neurosci 2000; 12:6779.
Balse E, Lazarus C, Kelche C, Jeltsch H, Jackisch R, Cassel JC. Intrahippocampal
grafts containing cholinergic and serotonergic fetal neurons ameliorate spatial
reference but not working memory in rats with mbria-fornix/cingular bundle
lesions. Brain Res Bull 1999; 49:263272.
Mingo NS, Cottrell GA, Mendonca A, Gombos Z, Eubanks JH, Burnham WM.
Amygdala-kindled and electroconvulsive seizures alter hippocampal expression
of the m1 and m3 muscarinic cholinergic receptor genes. Brain Res 1998; 810:9
15.
Vann SD, Brown MW, Erichsen JT, Aggleton JP. Fos imaging reveals dierential
patterns of hippocampal and parahippocampal subeld activation in rats in
response to dierent spatial memory tests. J Neurosci 2000; 20:27112718.
Herzog CD, Gandhi C, Bhattacharya P, Walsh TJ. Eects of intraseptal zolpidem
and chlordiazepoxide on spatial working memory and high-anity choline uptake
in the hippocampus. Neurobiol Learn Mem 2000; 73:168179.
Stancampiano R, Cocco S, Cugusi C, Sarais L, Fadda F. Serotonin and acetylcholine release response in the rat hippocampus during a spatial memory task.
Neuroscience 1999; 89:11351143.
Chepurnov SA, Chepurnova NE, Kholmanskikh SS, Berdiev RK, Park JK, Son
JO. Learning by mice of Y-maze using drinking reinforcement and aversive
olfactory stimulus (benecial eect of ginsenosides). Biull Eksp Biol Med 1996;
122:253257.
Gillis CN. Panax ginseng pharmacology: a nitric oxide link? Biochem Pharmacol
1997; 54:18.
Benishin CG, Lee R, Wang LC, Liu HJ. Eects of ginsenoside Rb1 on central
cholinergic metabolism. Pharmacology 1991; 42:223229.
Ma TC, Yu QH. Eect of 20(S)-ginsenoside-Rg2 and cyproheptadine on two-way
active avoidance learning and memory in rats. Arzneimittelforschung 1993;
43:10491052.
Ma TC, Yu QH, Chen MH. Eects of ginseng stem-leaves saponins on one-way
avoidance behavior in rats. Acta Pharmacol Sin 1991; 12:403406.
Liu M. Studies on the anti-aging and nootropic eects of ginsenoside Rg1 and its
mechanisms of actions. Prog Physiol Sci 1996; 27:139142.
Liu M, Zhang JT. Eects of ginsenoside Rg1 on c-fos gene expression and cAMP
levels in rat hippocampus. Acta Pharmacol Sin 1996; 17:171174.
Lim JH, Wen TC, Matsuda S, Tanaka J, Maeda N, Peng H, Aburaya J, Ishihara K,
Sakanaka M. Protection of ischemic hippocampal neurons by ginsenoside Rb1, a
main ingredient of ginseng root. Neurosci Res 1997; 28:191200.
Li Z, Guo YY, Wu CF, Li X, Wang JH. Protective eects of pseudoginsenosideF11 on scopolamine-induced memory impairment in mice and rats. J Pharm
Pharmacol 1999; 51:435440.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.

123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
134.
135.
136.
227
Zhang JT, Qu ZW, Liu Y, Deng HL. Preliminary study on antiamnestic

mechanism of ginsenoside Rg1 and Rb1. Chin Med J 1990; 103:932938.
Smith K, Engels H-J, Martin JA, Wirth JC. Ecacy of a standardised ginseng
extract to alter psychological function characteristics at rest and during exercise
stress. Med Sci Sport Exerc 1995; 27:S147.
DAngelo L, Grimaldi R, Caravagi M, Marcoli M, Perucca E, Lecchini S, Frigo
GM, Crema A. A double-blind placebo-controlled clinical study on the eect of
a standardized ginseng extract on psychomotor performance in healthy volunteers.
J Ethnopharmacol 1986; 16:1522.
Sorensen H, Sonne J. A double-masked study of the eect of ginseng on cognitive
functions. Curr Ther Res 1996; 57:959968.
Winther K, Ranlov C, Rein E, Mehlsen J. Russian root (Siberian Ginseng)
improves cognitive functions in middle-aged people whereas Ginkgo biloba seems
eective only in the elderly. J Neurol Sci 1997; 150:S90.
Thommessen B, Laake K. No identiable eect of ginseng (Gericomplex) as
an adjuvant in the treatment of geriatric patients. Aging (Milano) 1996; 8:417
420.
Joseph JA, Shukitt-Hale B, Denisova NA, Bielinski D, Martin A, McEwen JJ,
cognitive, and motor behavioral decits with blueberry, spinach, or strawberry
dietary supplementation. J Neurosci 1999; 19:81148121.
Sun GY, Xia J, Draczynska-Lusiak B, Simonyi A, Sun AY. Grape polyphenols
protect neurodegenerative changes induced by chronic ethanol administration.
Neuroreport 1999; 10:9396.
Bagchi D, Garg A, Krohn RL, Bagchi M, Bagchi DJ, Balmoori J, Stohs SJ.
Protective eects of grape seed proanthocyanidins and selcted antioxidants against
TPA-induced hepatic and brain lipid peroxidation and DNA fragmentation, and
peritoneal macrophage activation in mice. Gen Pharmacol 1998; 30:771776.
Hong JT, Ryu SR, Kim HJ, Lee JK, Lee SH, Kim DB, Yun YP, Ryu JH, Lee BM,
Kim PY. Neuroprotective eect of green tea extract in experimental ischemiareperfusion brain injury. Brain Res Bull 2000; 53:743749.
Hong JT, Ryu SR, Kim HJ, Lee JK, Lee SH, Yun YP, Lee BM, Kim PY.
Protective eect of green tea extract on ischemia/reperfusion-induced brain injury
in Mongolian gerbils. Brain Res 2001; 888:1118.
Inanami O, Asanuma T, Inukai N, Jin T, Shimokawa S, Kasai N, Nakano M, Sato
F, Kuwabara M. The suppression of age-related accumulation of lipid peroxides in
rat brain by administration of Rooibos tea (Aspalathus linearis). Neurosci Lett
1995; 196:8588.
Komatsu M, Hiramatsu M. The ecacy of an antioxidant cocktail on lipid
peroxide level and superoxide dismutase activity in aged rat brain and DNA
damage in iron-induced epileptogenic foci. Toxicology 2000; 148:143148.
Hartmann AM, Burleson LE, Holmes AK, Geist CR. Eects of chronic kombucha
ingestion on open-eld behaviors, longevity, appetitive behaviors, and organs in
c57-b1/6 mice: a pilot study. Nutrition 2000; 16:755761.
228
Youdim and Joseph
137.
Hindmarch I, Quinlan PT, Moore KL, Parkin C. The eects of black tea and other
beverages on aspects of cognition and psychomotor performance. Psychopharmacology 1998; 139:230238.
Nishiyama N, Zhou Y, Saito H. Benecial eects of DX-9386, a traditional
Chinese prescription, on memory disorder produced by lesioning the amygdala in
mice. Biol Pharm Bull 1994; 17:16791681.
Nishiyama N, Zhou Y, Saito H. Ameliorative eects of chronic treatment using
DX-9386, a traditional Chinese prescription, on learning performance and lipid
peroxide content in senescence accelerated mouse. Biol Pharm Bull 1994;
17:14811484.
Nishiyama N, Zhou Y, Takashina K, Saito H. Eects of DX-9386, a traditional
Chinese preparation, on passive and active avoidance performances in mice. Biol
Pharm Bull 1994; 17:14721476.
Nishiyama N, Wang YL, Saito H. Benecial eects of S-113m, a novel herbal
prescription, on learning impairment model in mice. Biol Pharm Bull 1995;
18:14981503.
Nishiyama N, Chu PJ, Saito H. An herbal prescription, S-113m, consisting of
biota, ginseng and schizandra, improves learning performance in senescence
accelerated mouse. Biol Pharm Bull 1996; 19:388393.
Smriga M, Saito H, Nishiyama N. Hoelen (Poria cocos Wolf) and ginseng (Panax
ginseng C. A. Meyer), the ingredients of a Chinese prescription DX-9386,
individually promote hippocampal long-term potentiation in vivo. Biol Pharm Bull
1995; 18:518522.
Zhang Y, Saito H, Nishiyama N. Improving eects of DX-9386, a traditional
Chinese medicinal prescription, on thymectomy-induced impairment of learning
behaviors in mice. Biol Pharm Bull 1994; 17:11991205.
Zhang Y, Saito H, Nishiyama N, Abe K. Eects of DX-9386, a traditional Chinese
medicinal prescription, on long-term potentiation in the dentate gyrus in rats. Biol
Pharm Bull 1994; 17:13371340.
Yagi H, Irino M, Matsushita T, Katoh S, Umezawa M, Tsuboyama T, Hosokawa
M, Akiguchi I, Tokunaga R, Takeda T. Spontaneous spongy degeneration of the
brain stem in SAM-P/8 mice, a newly developed memory-decient strain.
J Neuropathol Exp Neurol 1989; 48:577590.
Yen TT, Kalasz H, Matkovics B. The eects of Ding lang root extract on the
activities of peroxide metabolism enzymes in the rat. In: Matkovics B, Karmazsin
L, Kalasz H, eds. Radicals, Ions and Tissue Damage. Budapest: Akademiai Kiado,
1990:293298.
Yen TT. Stimulation of sexual performance in male rats with the root extract of
Ding lang (Policias fruticosum L.). Acta Physiol Hung 1990; 75:6167.
Yen TT, Dallo J, Knoll J. The eect of Ding lang and (-) deprenyl on the survival
rate of male rats. Acta Physiol Hung 1990; 75:301302.
Yen TT. Improvement of learning ability in mice and rats with the root extract of
Ding Lang (Policias fruticosum L.). Acta Physiol Hung 1990; 75:6976.
Xuejiang W, Magara T, Konishi T. Prevention and repair of cerebral ischemia-
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
229
reperfusion injury by Chinese herbal medicine, shengmai san, in rats. Free Radic
Res 1999; 31:449455.
Sugaya E, Yuyama N, Kajiwara K, Tsuda T, Ohguchi H, Shimizu-Nishikawa K,
Kimura M, Sugaya A. Regulation of gene expression by herbal medicinesa new
paradigm of gene therapy for multifocal abnormalities of genes. Res Commun Mol
Pathol Pharmacol 1999; 106:171180.
Inada K, Yokoi I, Kabuto H, Habu H, Mori A, Ogawa N. Age-related increase in
nitric oxide synthase activity in senescence accelerated mouse brain and the eect
of long-term administration of superoxide radical scavenger. Mech Ageing Dev
1996; 89:95102.
Hiramatsu M, Edamatsu R, Mori A. Free radicals, lipid peroxidation, SOD
activity, neurotransmitters and choline acetyltransferase activity in the aged rat
brain. EXS 1992; 62:213218.
Hamada H, Hiramatsu M, Edamatsu R, Mori A. Free radical scavenging action of
baicalein. Arch Biochem Biophys 1993; 306:261266.
Hiramatsu M, Haba K, Edamatsu R, Hamada H, Mori A. Increased choline
acetyltransferase activity by Chinese herbal medicine Sho-saiko-to-go-keishikashakuyaku-to in aged rat brain. Neurochem Res 1989; 14:249251.
Haba K, Ogawa N, Mori A. The eects of sho-saiko-to-go-keishi-ka-shakuyaku-to
(TJ-960) on ischemia-induced changes of brain acetylcholine and monoamine
levels in gerbils. Neurochem Res 1990; 15:487493.
Chen J. An experimental study on the anti-senility eects of shou xing bu zhi. Chin
J Mod Dev Trad Med 1989; 9:226227, 198.
Liu JK, Kabuto H, Hiramatsu M, Mori A. Eects of Guilingji on brain
monoamines and their metabolites in mice. Acta Med Okayama 1991; 45:217
222.
Liu J, Edamatsu R, Kabuto H, Mori A. Antioxidant action of guilingji in the brain
of rats with FeCl3-induced epilepsy. Free Radic Biol Med 1990; 9:451454.
Kabuto H, Asanuma M, Nishibayashi S, Iida M, Ogawa N. Chronic administration
of Oren-gedoku-to (TJ15) inhibits ischemia-induced changes in brain indoleamine
metabolism and muscarinic receptor binding in the Mongolian gerbil. Neurochem
Res 1997; 22:3336.
Fushitani S, Minakuchi K, Tsuchiya K, Takasugi M, Murakami K. Studies on
attenuation of post-ischemic brain injury by kampo medicines-inhibitory eects of
free radical production. II. J Pharm Soc Jpn 1995; 115:611617.
Xu J, Murakami Y, Matsumoto K, Tohda M, Watanabe H, Zhang S, Yu Q, Shen J.
Protective eect of Oren-gedoku-to (Huang-Lian-Jie-Du-Tang) against impairment of learning and memory induced by transient cerebral ischemia in mice.
Sasaki K, Sudo T, Kurusu T, Kiuchi T, Yoshizaki F. The mechanism of alteration
of monoamine metabolism in brain regions in marble burying behavior-isolated
housing mice and eect of oren-gedoku-to on this alteration. J Pharm Soc Jpn
2000; 120:559567.
Kondo Y, Kondo F, Asanuma M, Tanaka K, Ogawa N. Protective eect of oren-
230
166.
167.
168.
169.
170.
171.
172.
173.
174.
175.
176.
177.
178.
Youdim and Joseph

gedoku-to against induction of neuronal death by transient cerebral ischemia in the
C57BL/6 mouse. Neurochem Res 2000; 25:205209.
Komatsu M, Ueda Y, Hiramatsu M. Dierent changes in concentrations of
monoamines and their metabolites and amino acids in various brain regions by the
herbal medicine/Toki-Shakuyaku-San between female and male senescenceaccelerated mice (SAMP8). Neurochem Res 1999; 24:825831.
Toriizuka K, Hou P, Yabe T, Iijima K, Hanawa T, Cyong JC. Eects of Kampo
medicine, Toki-shakuyaku-san (Tang-Kuei-Shao-Yao-San), on choline acetyltransferase activity and norepinephrine contents in brain regions, and mitogenic
activity of splenic lymphocytes in ovariectomized mice. J Ethnopharmacol 2000;
71:133143.
Kishikawa M, Sakae M. Herbal medicine and the study of aging in senescenceaccelerated mice (SAMP1TA/Ngs). Exp Gerontol 1997; 32:229242.
Itoh T, Murai S, Saito H, Masuda Y. Eects of single and repeated administrations
of Toki-shakuyaku-san on the concentrations of brain neurotransmitters in mice.
Methods Find Exp Clin Pharmacol 1998; 20:1117.
Nishiyama N, Moriguchi T, Saito H. Benecial eects of aged garlic extract on
learning and memory impairment in the senescence-accelerated mouse. Exp
Gerontol 1997; 32:149160.
Moriguchi T, Saito H, Nishiyama N. Anti-ageing eect of aged garlic extract in
the inbred brain atrophy mouse model. Clin Exp Pharmacol Physiol 1997; 24:235
242.
Moriguchi T, Matsuura H, Itakura Y, Katsuki H, Saito H, Nishiyama N. Allixin, a
phytoalexin produced by garlic, and its analogues as novel exogenous substances
with neurotrophic activity. Life Sci 1997; 61:14131420.
Zhang Y, Moriguchi T, Saito H, Nishiyama N. Functional relationship between
age-related immunodeciency and learning deterioration. Eur J Neurosci 1998;
10:38693875.
Moriguchi T, Takashina K, Chu P-J, Saito H, Nishiyama N. Prolongation of life
span and improved learning in the senescence accelerated mouse produced by aged
garlic extract. Biol Pharm Bull 1994; 17:15891594.
Shimada A, Ohta A, Akiguchi I, Takeda T. Inbred SAM-P/10 as a mouse model of
spontaneous, inherited brain atrophy. J Neuropathol Exp Neurol 1992; 51:440
450.
Shimada A, Hosokawa M, Ohta A, Akiguchi I, Takeda T. Localization of
atrophy-prone areas in the aging mouse brain: comparison between the brain
atrophy model SAM-P/10 and the normal control SAM-R/1. Neuroscience 1994;
59:859869.
Kaul S, Krishnakantha TP. Inuence of retinol deciency and curcumin/turmeric
feeding on tissue microsomal membrane lipid peroxidation and fatty acids in rats.
Mol Cell Biochem 1997; 175:4348.
Kaul S, Krishnakanth TP. Eect of retinol deciency and curcumin or turmeric
feeding on brain Na(+)-K+ adenosine triphosphatase activity. Mol Cell Biochem
1994; 137:101107.

179.
180.
181.
182.
231
Sugiura M, Shoyama Y, Saito H, Abe K. Crocin (crocetin di-gentiobiose ester)

prevents the inhibitory eect of ethanol on long-term potentiation in the dentate
gyrus in vivo. J Pharmacol Exp Ther 1994; 271:703707.
Abe K, Saito H. Eects of saron extract and its constituent crocin on learning
behaviour and long-term potentiation. Phytother Res 2000; 14:149152.
Suganuma H, Hirano T, Inakuma T. Amelioratory eect of dietary ingestion with
red bell pepper on learning impairment in senescence-accelerated mice (SAMP8).
J Nutr Sci Vitaminol 1999; 45:143149.
Youdim KA, Spencer JPE, Schroeter H, Rice-Evans C. Flavonoids in neuroprotection. Biol Chem. In press.
9
Flavonoids: Neuroprotective
Agents? Modulation of Oxidative
StressInduced Map Kinase
Signal Transduction
Hagen Schroeter
Los Angeles, California, U.S.A.
London, England
Jeremy P. E. Spencer
London, England
I.
INTRODUCTION
The pathological mechanisms of neurodegeneration and the process of aging are

increasingly associated with oxidative stress mediated by reactive oxygen species/reactive nitrogen species (ROS/RNS) [1,2]. Several ndings substantiate the
susceptibility of the central nervous system (CNS) to oxidative stress:
The high density of mitochondria, a major cellular source of ROS, in brain
tissue [1,3]
The high rate of oxygen consumption [1,3]
A considerable pool of autoxidizable compounds such as dopamine, Ldopa, and noradrenalin [4]
The presence of locally high levels of potentially excitotoxic glutamate [5]
The high Ca2+ trac across membranes [6]
233
234
Schroeter and Spencer
The lipid composition of the neuronal membrane with its relatively

high content of polyunsaturated fatty acids compared to that of the
nonneuronal cell
Oxidative stress has been shown to contribute to the neuropathological processes
of a number of neurodegenerative disorders [1], including Alzheimers disease [7],
Parkinsons disease [8], and Huntingtons disease [9], as well as being implicated
in neuronal loss associated with age-related cognitive decline [2,10,11], cerebral
ischemia/reperfusion injuries [5,12], seizures, trauma, and neuroinammation
[13,14]. Several factors might enhance oxidative stressrelated neuronal decline
in age-related neurological disorders:
Decrements in energy availability due to reduced mitochondrial
function [15]
Loss of enzyme activities essential in the defense against ROS/RNS or
important in damage removal, especially superoxide dismutase (SOD),
catalase, glutathione peroxidase, and proteasomal functions [16,17]
Decline in the overall antioxidant status due to a low intake of vitamin C,
vitamin E, h-carotene, minerals (e.g., selenium), and other micronutrients based on age-related changes in food intake and digestion
[1820]
Accumulation of the amyloid h-peptide in Alzheimers disease and the
deposition of potentially pro-oxidant iron in certain brain regions,
especially the substantia nigra, in Parkinsons disease [21]
Indeed, increased iron levels in the substantia nigra [21], elevation of lipid
peroxidation [22], decline in glutathione concentrations [23], enhanced protein
oxidation [24], and increase in glycation end products [25] are biomarkers
associated with oxidative damage in neurodegeneration and aging.
II.
NEURONAL APOPTOSIS: INVOLVEMENT OF THE

MITOGEN-ACTIVATED PROTEIN KINASE CASCADE
A.
The Mitogen-Activated Protein Kinases
High levels of ROS and RNS can disrupt the normal redox state and shift cells
into the state of oxidative stress, hallmarked by intracellular increases in
products of lipid peroxidation, hydrogen peroxide, and elevated damage to
other biomolecules. It is often stated that the damage to important biomolecules
such as proteins and deoxyribonucleic acid (DNA) as a result of oxidative stress
leads to cell injury or death. Importantly, this oversimplication ignores a
number of stress response mechanisms that cells have developed to coordinate
reactions that ultimately determine the outcome of an oxidative insult. Among
Flavonoids: Neuroprotective Agents?
235
the main stress signaling pathways or central mediators in stress response to

oxidative insults are the mitogen-activated protein kinase (MAPK) cascades, the
phosphoinositol 3-kinase/Akt pathway (PI3-kinase/Akt), the nuclear factor nB
(NF-nB) signaling system, p53 activation, and the heat shock protein response.
In general, MAPK pathways transduce extracellular and intracellular stimuli into
cellular responses. These responses consist of direct phosphorylation of cytosolic
or nuclear target proteins and activation of transcription factors, which consequently modulate gene expression. Typically, MAPK cascades are composed
of three proteins, which are subsequently activated by phosphorylation: MAPK
kinase kinase (MAPKKK), MAPK kinase (MAPKK), and MAPK [26,27]. These
cascades may be initiated by a great variety of stimuli often via small guanosine
triphosphate(GTP)-binding proteins (Ras, Rho, Ral, Rap) [28,29].
Mammals express at least three distinctly regulated groups of MAPKs,
which may exist in dierent isoforms:
1.
2.
3.
extracellular signal-regulated kinases (ERK1/2)

c-Jun amino-terminal kinases (JNK1/2/3)
p38 kinases (p38a/h/g/y)
Each MAPKK can be activated by more than one MAPKKK, increasing the
complexity and diversity of MAPK signaling. All members of the MAPK family
require dual phosphorylation of a threonine and tyrosine residue within the
catalytic domain by their respective upstream kinase in order to be activated.
Hence ERK, JNK, and p38 contain the specic dual phosphorylation motif ThrGlu-Tyr, Thr-Pro-Tyr, and Thr-Gly-Try, respectively [27]. Besides the upstream
kinases, the activation of MAPKs critically depends on the activity of a special
family of dual specicity phosphatases, the MAP kinase phosphatases (MKPs),
which inactivate MAPK and therefore play an important role in the dynamics of
MAPK signaling.
Active MAPKs function as modulators for dierentiation, proliferation,
cell death, and survival. Commonly, the activation of ERK1/2 has been
linked to cell survival, whereas JNK and p38 [also called the stress-activated
protein kinases (SAPKs)] have been associated with apoptosis [30]. This
perspective is an oversimplication, and the actual roles are highly dependent
on the cell type, the state of cell development, the kind of stimulus, and the
context of stimulation.
B.
Extracellular Signal-Regulated Kinase
Extracellular signal-regulated kinase (ERK) is probably the best-characterized,

hence best-understood, member of the MAPK family. The activation of ERK1/2
in neurons triggers a wide variety of possible cellular responses ranging from
proliferation and dierentiation to cell survival and death.
236
1.
Activation of Extracellular Signal-Regulated Kinase
Classically, ERK is activated in the CNS via growth factors such as the neuronal
growth factor (NGF) [31]. The binding of neurotrophins to Trk receptors induces
the autophosphorylation of Trk, which subsequently allows the docking of the
Shc adaptor protein (Shc). This triggers the phosphorylation of Shc and
consequently the recruitment of the Grb2-SOS complex, which leads to the
activation of the small G-protein Ras. Ras-GTP initiates the subsequent
sequential phosphorylation of Raf1-kinase (a MAPKKK), MEK1/2 (a MAPKK),
and nally ERK1/2 (MAPK) [32] (Fig. 1). A novel pathway leading to the
activation of ERK1/2 in neuronal cells, which is wholly Ca2+-dependent,
Figure 1 The classic growth factormediated activation of ERK1/2. The figure

illustrates the binding of the neuronal growth factor (NGF) to its receptor (TrkA).
The subsequent autophosphorylation of TrkA allows the binding and of the adapter
protein Shc, which in turn triggers the recruitment of Grb2-SOS, subsequently
leading to a GDP/GTP exchange and the phosphorylation of Ras. After its
activation Ras-GTP initiates the subsequent sequential phosphorylation of Raf1kinase (MAPKKK), MEK (MAPKK), and finally ERK1/2 (MAPK). ERK, extracellular
signal-regulated kinase; GDP/GTP, guanosine diphosphate/guanosine triphosphate; MAPK, mitogen-activated protein kinase.
237
was recognized in the 1990s and gives insights to cell signaling processes in
which calcium homeostasis is a fundamental factor [31,33,34]. These ndings
implicate the increase in intracellular calcium levels mediated by glutamate
receptors with an activation of ERK1/2. This activation of ERK1/2 involves a
complex signal transduction involving a pertussis-sensitive G-protein [33] and
possibly phosphoinositol-3-kinase [34]. Evidence that Ca2+ might be an important regulator of ERK1/2 activation also emerges from the observation that the
activation of the neurospecic Ras-GRF1, an activator of the MAPK pathways, is
Ca2+/calmodulin-sensitive [31,35,36].
2.
Targets for Active Extracellular Signal-Regulated Kinase 1/2
Activation of ERK1/2 can lead to the phosphorylation of a wide array of

potential targets in cytosol or nucleus. For example, active ERK1/2 can activate
transcription factors [37] and eector kinasesthe MAPK-activated protein
kinases (MAPKAPKs) such as the mitogen- and stress-activated kinase 1
(MSK1) or the pp90 ribosomal S6 kinases (RSKs) [38]. RSK phosphorylates
the Bcl-2 family member BAD, thereby inhibiting its proapoptotic activity [32].
RSK and MSK1 are also potent activators of the cyclic adenosine monophosphate (cAMP) response element binding protein (CREB), a transcription
factor for Bcl-2 and therefore an important factor for cell survival [39]. This
would suggest that active ERK1/2 plays an important role in prosurvival signal
transduction processes. Indeed, dierent studies clearly indicate that under
certain conditions the activation of ERK1/2 is essential to neuronal survival
[30,32]. Furthermore, the activation of Ras, an initiator of the ERK1/2 signaling
cascade, has been shown also to activate the PI3-kinase/Akt pathway, another
important survival pathway in neurons, demonstrating possible interlinks of
dierent survival signals [40]. Interestingly, the protective eects initiated by
Ras activation involve the suppression of proapoptotic signals such as the
tumor suppressor protein p53 and the Bcl-2 family member Bax and are less
potent if either the ERK1/2 or the PI3-kinase/Akt pathway is selectively
inhibited [40]. Thus, this describes the outcome of a signal transduction process
as the sum of dierent actions toward the same target and underlines the
importance of interlinks among signaling cascades in promoting and enhancing
specic signals.
However, the actual role of ERK1/2 seems to be dependent on various
parameters since inhibition of ERK1/2 activation during focal ischemia [41],
oxidative stress [42], and a model for hippocampal seizures [43] attenuated
neuronal death and cellular injury, indicating a proapoptotic role for ERK1/2. In
addition, the inhibition of ERK1/2 activation has been demonstrated to protect a
mouse neuronal cell line and rat primary cortical neurons from oxidative stress
induced neurotoxicity [44].
238
C.
c-Jun Amino-Terminal Kinase
The stress-activated protein kinases (SAPKs) c-Jun amino-terminal kinase (JNK)

and p38 are involved in signal transduction processes that are complex with regard
to the large number of dierent stimuli capable of activating JNK or p38. At least
12 dierent MAPKKKs involved in the activation of SAPK have been overexpressed, of which the physiological functions, specicities, and regulations remain
elusive. JNK exists in mammals in three distinct isoforms, which are dierentially
expressed in dierent tissues. The expression pattern of the JNK isoforms may
even dier in cells of the same tissue. JNK3 expression is largely restricted to the
CNS, whereas JNK1 and JNK2 have a wider pattern of distribution [27,45,46].
1.
Activation of c-Jun Amino-Terminal Kinases
JNKs can be activated by a large range of dierent stimuli, including kainic acid
[47], microtubule-disrupting agents [48], double-stranded ribonucleic acid
(RNA) [49], viral infections [49], tumor necrosis factors, interleukins [50],
osmotic stress [51], and trophic factor withdrawal [30]. JNKs have also been
reported to be activated by oxidative stress, particularly that induced by ultraviolet radiation [47], hypoxia [52], ischemia [53], hydrogen peroxide [30,54],
peroxynitrite [55], malondialdehyde [56], and 4-hydroxynonenal [57].
In accord with the general scheme of the activation of MAPKs, the JNK
signaling cascade can be initiated through the activation of small G-proteins such
as Rac/cdc42, a protein belonging to the Rho family of small G-proteins. Rac/
cdc42 is activated by guanosine diphosphate/guanosine triphosphate (GDP/GTP)
exchange mediated by a variety of guanine nucleotide exchange factors (GEFs)
such as SOS, VAV, Tiam, and the neuron-specic Ras-GRF1 in response to a
stimulus [45]. Active Rac/cdc42 initiates the subsequent phosphorylation of
JNK-specic MAPKKKs such as MEKK2/3,which phosphorylates the
MAPKKs, and MKK4/7, which ultimately causes dual phosphorylation of JNKs
(MAPK) [27,45,46] (Fig. 2).
Several other MAPKKKs, have been reported to activate the JNK pathway.
These include members of the MEKK group (MEKK14), the mixed-lineage
protein kinases (MLK13) [58,59], and members of the apoptosis signaling kinase
(ASK) family [60] (Fig. 3). This variety of potential activators of JNKs reects the
broad array of possible stimuli that may lead to JNK phosphorylation. For
example, JNKs can be activated via the MAPKKK MEKK1, involving caspases
[61]; via the MAPKKK MEKK4, involving the p53-mediated expression of
growth arrest and DNA damageinduced protein (GADD45) [62]; or via the
tumor necrosis factor a(TNF-a)-induced activation of the ASK1 [60] (Fig. 3).
As a result of the diversity of stimuli and upstream events that lead to the
activation of JNKs, this signaling cascade is complex and not completely understood. Furthermore, the essential Ca2+/calmodulin binding to the neuron-specic
239
Figure 2 The activation of JNK. Rac/cdc42, a small G-protein, is activated by

GDP/GTP exchange mediated by guanine nucleotide exchange factors (GEFs).
Active Rac/cdc42 initiates the subsequent phosphorylation of the JNK-specific
MAPKKK MEKK2/3, which in turn phosphorylates the MAPKK MKK4/7, ultimately
resulting in the dual phosphorylation of JNK (MAPK). JNK, c-Jun amino-terminal
kinase; GDP/GTP, guanosine diphosphate/guanosine triphosphate; MAPK, mitogen-activated protein kinase.
Ras-GRF1 [35,36], which may also activate the JNK pathway [63], provides a
possible link between intracellular Ca2+ homeostasis and activation of JNKs.
The control of such diverse signaling processes is essential for cellular
functions, especially in view of the variety of dierent stimuli involved. It is
therefore important to ensure specicity in MAPK activation and function. In
the case of JNKs, mechanisms ensuring specicity involve scaolding proteins,
physical interactions between members of a given cascade, and ability of MAPK
to regulate indirectly the expression of ligands and inhibitors for receptors that
are involved in JNK signaling [45].
2.
Targets for Active c-Jun Amino-Terminal Kinase
Active JNKs have a wide range of potential phosphorylation targets in the nucleus
as well as in the cytoplasm (Fig. 4). Nuclear substrates for JNKs are transcription
240
Figure 3 Diverse stimuli are able to activate JNK. This picture exemplifies the
various pathways leading to JNK activation, indicating the complexity and diversity
of stimuli able to induce JNK-mediated signal transduction. JNK, c-Jun aminoterminal kinase.
factor proteins such as c-Jun [6466], ATF-2 [67], and ELK-1 [68]. As far as is
known, JNKs are the only kinases capable of phosphorylating c-Jun in vivo. c-Jun
is part of the activator protein 1 (AP-1) transcription factor, which exists as either
a Jun homodimer or as a Jun/Fos heterodimer [69]. The transcriptional activity of
the AP-1 complex is increased [69] following phosphorylation of Ser-63 and Ser73 of c-Jun by JNK [70]. In addition to c-Jun, JNKs also phosphorylate other AP1 proteins, including JunB and JunD [69,71]. The regulatory eect of JNKs on
AP-1 transcription may not be due only to phosphorylation of Jun but may also
involve JNK-regulated ubiquitin-mediated degradation of AP-1 proteins [72,73].
JNKs appear to be essential in cytokine- and stress-induced activation of AP-1 but
are not required for AP-1 activation in response to other stimuli.
Cytosolic substrates for JNKs include cytoskeletal proteins, the tumor
suppressor protein p53, the mitogen-activated kinase activating death domain
241
Figure 4 Potential cytosolic and nuclear targets for active JNK. JNK, c-Jun
amino-terminal kinase.
protein (MADD), glucocorticoid receptors, Bcl-2 and Bcl-xL, neurolaments,

tau, and STAT-3, a member of the signal transducer and activators of transcription (STAT) family (Fig. 4).
D.
c-Jun Amino-Terminal Kinase Signaling in Neuronal Death
Although JNK activation is often associated with cellular injury and degeneration,
it also plays an important part in embryonic morphogenesis, memory formation,
and immune defense. The proapoptotic action of JNKs in neurons was initially
investigated in neuronal cell death that followed the withdrawal of neurotrophic
factors [30]. It was found that JNK activation contributed to the apoptotic response
and that JNK-mediated apoptosis was suppressed by activation of survival
pathways. Studies in mice utilizing the disruption of the neuronal gene JNK3
conrmed the role of JNK in stress-induced neuronal apoptosis by demonstrating
that JNK3-knockout mice were developmentally normal but resistant to excitotoxin-induced neuronal apoptosis [74]. These ndings were supported by observations in mice with a mutation in the c-Jun gene that altered the JNK
phosphorylation sites of c-Jun and led to a resistance to kainic acidtriggered
death of hippocampal neurons [75]. Further support arose from data obtained by
using dominant-negative c-Jun mutants, which reduced sympathetic neuronal
242
death after NGF withdrawal [76]. A proapoptotic function has also been suggested
for JNK activation in NGF-withdrawal-induced neuronal death [77] in a hippocampal model of Huntingtons disease [78] and h-amyloid-induced neuronal
apoptosis [79]. However, the apoptotic process does not occur in JNK-knockout
mice [74] or mice expressing a mutant form of c-Jun lacking the JNK phosphorylation site [75]. In addition, the indirect inhibition of JNK by CEP-1347 protected
neuronal PC12 cells and sympathetic neurons in vitro from death after trophic
factor withdrawal, h-amyloid exposure, ultraviolet (UV) radiation and oxidative
stress [75,79,80]. Furthermore, CEP-1347 protected nigral neurons in vivo against
MPTP-induced neuronal death [81] implicating again the involvement of JNK in
neurodegeneration. Moreover, JNK activity and apoptosis in cerebellar granule
neurons were enhanced after inhibition of the prosurvival pathway PI3-kinase/Akt
[82]. Finally JNKs seem to be indirectly involved in other apoptotic pathways by
enhancing transcription of death receptors such as Fas-L [83] or by activating and
stabilizing the p53 protein [84,85]. Together these data strongly support JNK
involvement in apoptosis signaling in neurons.
A review in 2000 by Davis [45] of the role of JNK in apoptosis suggested a
mechanism of JNK-dependent apoptosis involving the mitochondria and caspase3. This hypothesis is based on observations in primary murine embryonic
broblasts (MEFs) lacking the genes for Jnk1 and Jnk2 (Jnk null; no JNK1/2
protein/activity). These Jnk null MEFs exhibit profound defects in stress-induced
(UV radiation, DNA alkylation, translational inhibition) apoptosis [86]. The
defect in the execution of apoptotic cell death was caused by the failure to initiate
the JNK-induced cytochrome c release from the mitochondria [86]. This malfunction is signicant since it is critical to the subsequent sequential activation of
Apaf-1 [87], the initiator caspase caspase-9 [88], and nally the eector caspase
caspase-3 [89], all of which are essential in the execution of apoptosis. Tournier
and colleagues [86] suggested that the apoptotic response is suppressed in Jnk null
MEF as a result of the absence of JNK, which is needed to initiate the apoptotic
cascade (Fig. 5). However, it is not clear yet by which molecular mechanism JNK
mediates the release of cytochrome c from the mitochondria. Although a c-Jun
activated transcription seems possible (Fig. 5) [75,76], it is not required for UVinduced apoptosis [86]. Several studies point to the JNK-mediated in vitro
phosphorylation/inactivation of the mitochondria-associated antiapoptotic proteins Bcl-2 and Bcl-XL [9092] as a possible mechanism, since Bcl-2/Bcl-XL is
known to regulate cytochrome c release (Fig. 5). Others have proposed an as yet
unknown adaptor protein as the mediator for JNK actions. Further studies are
needed to substantiate this hypothesis and establish the molecular mechanisms.
In summary it is becoming increasingly clear that JNK activation is
involved in apoptotic processes in vivo and in vitro either by direct eects of
JNK-mediated c-Jun phosphorylation, adverse eects on survival pathways, or
indirect eects on cell function via other mediator systems. But apoptosis is not
243
Figure 5 The possible involvement of JNK signaling in the execution of cell

death. Once activated, JNK can phosphorylate p53, which in term stabilizes and
activates this apoptotic mediator, resulting in the suppression of the antiapoptotic
protein Bcl-2 and the induction of the proapoptotic protein BAX. Active JNK may
also influence the expression of other proapoptotic molecules via the c-Jun/AP-1mediated regulation of their expression. The proposed function for active JNK as
mediator of the phosphorylation of the mitochondria-associated proteins Bcl-2 and
Bcl-XL alters their antiapoptotic functions and may subsequently result in the
release of cytochrome c or other apoptotic factors such as DIABLO/Smac from
the mitochondria. The release of cytochrome c in turn promotes the formation
of the apoptosome, subsequently leading to the activation of caspase-3, one of the
major executors of apoptosis. JNK, c-Jun amino-terminal kinase; AP-1, activator
protein 1.
the inevitable outcome of JNK signaling. In fact, apoptotic cell death might
rather be the sum of signal transduction processes involving both prosurvival
and proapoptotic pathways.
E.
Extracellular Signal-Regulated Kinase1/2 and c-Jun

Kinase Signaling and Oxidative Stress
Oxidative stress seems to be a major stimulus for MAPK signaling cascades.

However, this picture would appear to be too simplistic since a shift of the overall
244
reductive environment that most cells maintain under physiological conditions

toward reduction also leads to the activation of MAPK signal transduction with
cell survival or death as a possible consequence. The observation that multiple
pathways are sensitive to alteration in intracellular ROS/RNS concentrations
indicates that this might represent a common way that cells have evolved to
signal a large diversity of dierent stressful stimuli. Accordingly a large number of
redox stress-responsive transcription factors and genes have been identied [93].
Growth factor receptors have been shown to undergo enhanced phosphorylation in response to oxidative insults such as hydrogen peroxide or UV
radiation initiating ERK1/2 activation [33,94,95]. This is consistent with
demonstrations of the mitogenic eects of low concentrations of hydrogen
peroxide [96]. However, the inhibition of ERK1/2 activation has been shown to
protect a mouse neuronal cell line and rat primary cortical neurons from
oxidative stressinduced neurotoxicity [97], demonstrating a possible involvement in cell death. The JNK signaling cascade has been reported to be activated
by a wide range of dierent oxidants/reductants, including hydrogen peroxide
[30,54], lipid peroxidation products [56,57], dierent types of radiation [47],
modulators of intracellular glutathione status [98], peroxynitrite [55], glutamate
[99], dithiothreitol, and nitric oxide [100]. Although the links between redox
status and MAPK signaling have been known for some time, data demonstrating
the molecular basis of such links and identifying the sensors in this redox
response are few.
Accumulating evidence indicates that members of the MAPK family or
their upstream or downstream partners have such redox-sensitive motives. For
example, JNK itself exhibits a redox-sensitive cysteine residue that is not
present on ERK or p38. An additional mechanism in JNK redox regulation is
its binding to redox-sensitive proteins such as glutathione S-transferases
(GSHSTs) [101]. It has been shown that under nonstressed conditions JNK is
associated with GSHST, resulting in the inhibition of JNK activity, but that
GSHST dissociates from JNK after UV radiation or hydrogen peroxide treatment [101,102]. Forced expression of GSHST decreased JNK activity, increased
c-Jun ubiquitination, and reduced c-Jun-mediated transcription [101]. In addition, GSHST expression in NIH 3T3 cells led to the attenuation of hydrogen
peroxideinduced JNK activation as well to an increase in ERK1/2 activity
[102]. Similarly GSHST and the redox regulatory protein thioredoxin (Trx) bind
under nonstressed conditions to the apoptosis signal-regulating kinase ASK1
[103105], an upstream activator of JNK, and inhibit JNK activity. However,
oxidative insults cause the dissociation of the Trx-ASK1 and GSHST complex
and the subsequent activation of [103105].
Ca2+-Homeostasis is another important mediator/regulator of oxidative
stressinduced signaling, since both ERK and JNK are sensitive to changes in
intracellular Ca2+ concentrations [35,36,63,106]. This Ca2+ sensitivity of ERK/
245
JNK signaling might play a role in oxidative stressinduced signaling events,

since it has been demonstrated that oxidative insults often inuence normal Ca2+
homeostasis in cells. With regard to ERK1/2, Samanta and coworkers [107]
demonstrated for the rst time that hydrogen peroxideinduced activation of
ERK1/2 in primary neurons is strictly dependent on extracellular calcium. Ca2+dependent MAPK activation has also been suggested to play a role in glutamate
receptormediated neuronal death [34]. The inux of Ca2+ into the cytosol from
the extracellular space or from intracellular stores after a stressful stimulus can
activate the Ca2+/calmodulin kinases, which in turn can stimulate the activation
of all three MAPKs [108].
The eect of ROS/RNS on MAPK activation is complex and occurs at
multiple levels, further results are needed to elucidate the molecular basis of
these interactions.
F.
Nitric Oxide and Mitogen-Activated Protein Kinase

Signaling
NO. has been identied as a messenger, promoting Ca2+-dependent neurotransmitter release from synaptic storage vesicles. NO. modulates exocytosis
through cGMP-dependent protein phosphorylation cascades after the classic
activation of soluble guanylate cyclase. The high diusibility of NO. makes it
an ideal retrograde signal for the two forms of synaptic modulation required for
learning and memory, namely, long-term potentiation (LTP) in the hippocampus
and long-term depression in the cerebellum [109].
There is growing evidence for numerous cGMP-independent mechanisms
for NO.-mediated cell signaling. Besides having a role in dierentiation and
synaptic plasticity, NO. has been implicated in neuronal apoptosis and, consequently, in neurodegenerative diseases, specically when NO. production is
increased to toxic levels [1,2,110112]. In particular, NO. has been linked to the
phenomenon of excitotoxicity involving the overstimulation of the NMDA
receptor by glutamate, subsequently triggering a strong intracellular accumulation of Ca2+ [113]. This Ca2+ overload in neurons leads to a substantial
increase in the activity of Ca2+/calmodulin-dependent nitric oxide synthase
(neuronal NOS), resulting in a high intracellular concentration of NO., which
has been identied as a mediator of glutamate-induced neuronal death [110].
Furthermore, microglia activation during neuroinammation is associated with
induction of inducible nitric oxide synthase (iNOS) expression, leading to large
and sustained NO. production, which also appears to be causally linked to
neuronal apoptosis and neurodegeneration [114,115].
Conversely, there is growing evidence for cGMP-independent NO.-mediated cell signaling toward neuronal survival, and NO. has been implicated in
mechanisms protecting against stress-induced cell injury. The GTP-binding
246
protein Ras is an intermediate in the transduction of signals from membrane

receptor tyrosine kinases to MAP kinases [116] (Fig. 6). Ras appears to be a
common signaling target of reactive free radicals, including NO., and agents that
modulate the cellular redox status, including H 2O2 and GSH (111). NO.
activates Ras by S-nitrosation of a highly conserved cysteine residue, leading
to GDP/GTP exchange and downstream signaling [117], such as the activation
Figure 6 Potential antiapoptotic mechanisms mediated by nitric oxide. The

nitrosation of Ras mediated by nitric oxide potentially leads to the activation of
ERK1/2 or the PI3-kinase/AKT pathway, resulting in the suppression of
proapoptotic proteins such as p53 and BAX or in the upregulation of antioxidant
enzymes. Furthermore, Ras activation might mediate the phosphorylation/
activation of the cAMP response element binding protein (CREB), a transcription
factor for Bcl-2, and the phosphorylation/inactivation of the proapoptotic protein
BAD. In addition, the NO.-mediated nitrosation of JNK and caspases may also
contribute to the antiapoptotic properties of NO.. ERK, extracellular signal-regulated
kinase; PI3, phosphoinositol 3-kinase; JNK, c-Jun amino-terminal kinase.
247
of ERK1/2. ERK1/2 activation, as discussed earlier, may in turn result in the

phosphorylation/inactivation of the proapoptotic protein BAD, the expression of
antiapoptotic proteins such as Bcl-2, and the suppression of proapoptotic
proteins such as BAX (discussed previously) (Fig. 6). It has been proposed that
this cysteine residue may represent an important molecular redox trigger,
whereby cells can respond to the ambient redox status. In addition, NO. may
directly suppress JNK activation via S-nitrosation of a redox-sensitive residue
that is not present on ERK or p38 [100,118], thereby also supporting prosurvival
mechanisms (Fig. 6).
Ultimately, S-nitrosation may thus regulate several redox-senstive transcription factors, including NF-nB, AP-1, Sp-1, and p53, and modulate levels of
active c-fos and c-jun [119]. Indeed, the expression of some enzymes implicated
in antioxidant defenses such as the enzymes for glutathione synthesis [120] and
heme oxygenase exhibit ERK1/2 dependency in the regulation of their expression (Fig. 6). Furthermore, the transcription of Cu/ZnSOD is regulated by the
transcription factor Elk1 [121] and the promoter for MnSOD expression contain
binding sites for Sp1, AP-1, and CREB [122,123], all of which have been linked
to regulation by ERK1/2 [32,124] (Fig. 6).
In turn, MAPKs are also involved in the regulation of the gene expression
of all three NOS isoenzymes. The expression of iNOS (NOS-2) is regulated by
all three MAP kinase pathways in a variety of cell types [125129]. For
example, JNK and ERK1/2 pathways are necessary for lipopolysaccharide
(LPS-) and interferon-g-stimulated iNOS expression in mouse macrophage
cells, possibly via a-tumor necrosis factor secretion, whereas p38 inhibited
induction [130]. The induction of endothelial NOS (eNOS, NOS-3) by estrogen,
broblast growth factor, or epidermal growth factor in endothelial cells involves
the Ras-ERK pathway [131,132]. eNOS is phosphorylated, and thus activated,
by the serine/threonine protein kinase Akt, which is recruited to the cell
membrane by PI3-kinase as an antiapoptotic mechanism in the response of
endothelial cells to shear stress [133].
III.
IMPLICATIONS FOR FLAVONOIDS IN PROTECTION

AGAINST NEURODEGENERATION
As indicated in the previous discussion, there is increasing evidence that

oxidative stress contributes to the neuropathological mechanisms of disorders
that include Alzheimers [7], Parkinsons [8], and Huntingtons diseases [9];
neuronal loss associated with cognitive decline in aging [11]; and cerebral
ischemia and seizures [5]. In addition, the molecular mechanisms underlying
oxidative stressinduced neuronal damage are emerging and appear to involve
an apoptotic mode of death in which ERK1/2 [42,44,107] and JNK [10,45,46]
248
have been strongly implicated. Furthermore, it is becoming clear that products of

lipid peroxidation such as 4-hydroxy-2,3-nonenal (4-HNE), lipid hydroperoxides (LOOHs), and oxysterols are important mediators of oxidative stress
mediated apoptosis in the CNS [22,134136]. Consequently, there is a growing
interest in the establishment of therapeutic and dietary strategies to combat
oxidative stressinduced damage to the CNS, and attention is turning to the
potential neuroprotective eects of dietary antioxidants, especially avonoids.
IV.
FLAVONOIDS: BIOLOGICAL AND CELLULAR

PROPERTIES
The major emphasis in recent years has concerned the potent in vitro antioxidant
eects of avonoids described in numerous publications [137140]. However,
avonoids exhibit various eects on mammalian cells with interesting implications for inammation [141], cardiovascular disease [142,143], and cancer [144]
involving the modulation of redox functions, calcium homeostasis [145], and the
activity of various enzyme systems [146] (for more information please refer to
Chap. 8). The eects of avonoids are often pictured as benecial for cell
survival, preventive against oxidative insults, and anticarcinogenic. However,
the actions of avonoids are complex and often seemingly antagonistic or
paradoxical. For example, avonoids have been described as antioxidant agents
protecting against oxidative insults to cells and apoptosis [147149]; other
researchers have found avonoids to be prooxidants and proapoptotic [150
152]. Thus it becomes clear that the eects of avonoids depend on dierent
factors such as the specic compound used, the cell type, concentrations,
experimental design, and the general context in which avonoids are used.
Furthermore, data used to demonstrate biological eects of avonoids are often
observational and fail to explain the molecular/cellular basis of such observations or base the eects solely on the rather unspecic (though important)
antioxidant properties of avonoids.
A.
Flavonoids: Neuroprotective Agents In Vivo

and In Vitro?
Data on avonoids in the context of the CNS or CNS-derived cells are starting to
accumulate but are not as extensive as for other cells or tissues. Epidemiological
and dietary intervention studies in humans and animals in 2000 suggest that
avonoids may play a useful role in preventing neurodegeneration, especially
age-related cognitive, motoric, and mood decline, and protect against oxidative
stress as well as cerebral ischemia/reperfusion injuries. Studies in humans using
avonoid-containing plant extracts, such as Ginkgo biloba extracts, demonstrate
249
positive eects on cognitive function and memory performance in healthy

volunteers from all age groups [153155], as well as in patients with age- or
Alzheimers diseaseassociated dementia [156158]. Watanabe and associates
[159] reported that Ginkgo biloba supplementation in mice had neuromodulatory
eects as indicated by a more than three-fold increase of messenger RNA
(mRNA) expression for neuronal tyrosine/threonine phosphatases-1, microtubule-associated tau, prolactin, dierent calcium and chloride channels, as well as
transthyretin. Flavonoid-associated [e.g., citrus avonoids, (epi)catechin, anthocyanidins] antioxidant interventions have also been proposed to be benecial in
hypoxia/ischemia, seizures, Parkinsons disease, increased survival rate in brain
cancers, and general age-related neurodegeneration [160164]. However, since
the avonoids used in these studies were given as food extracts or preparations
containing other potential bioactive or antioxidant components rather than as pure
avonoid fractions, the results do not solely reect the eects of avonoids.
Furthermore, the mechanisms of action of avonoids remain speculative and may
involve the antioxidant properties, the modulation of receptor/protein function,
and calcium homeostasis or a combination thereof [165].
In vivo studies in animals demonstrate protective eects of the avonoids
epicatechin and quercetin against ischemia/reperfusion-induced neuronal injury
[166,167]. Furthermore, avonoids showed protective eects against an
increase in brain lipid peroxidation after vitamin E deprivation [168], attenuated
neuronal damage induced by ethanol administration [169,170], and reduced Oethyl-S,S-dipropyl phosphorodithioateinduced neurotoxicity in mice [171].
Interestingly, the oral administration of a catechin-containing antioxidant
preparation signicantly increased the life span of senescence-accelerated mice
[172] and increased SOD activity in the mitochondrial fraction of the striatum
and the midbrain, decreased products of lipid peroxidation in cortex and
cerebellum, and attenuated the iron(II)-chloride-induced formation of markers
of DNA damage in the cortex of aged rats [173]. The authors of this study relate
the observed eects mainly to the rather unspecic antioxidant activity of
catechin and other compounds contained in the preparation and give no further
insights about the mechanisms related to the increase in mitochondrial SOD
activity. Other investigators reported an attenuation of age-related decline in
cognitive function and behavioral decits after long-term dietary supplementation with anthocyanidin-rich foods or plant extracts [163,174,175]. The
investigators used several neuronal and behavioral parameters including dopamine release, GTP activity, calcium buering in striatal synaptosomes, rod
walking tasks, and water maze performance to elucidate possible mechanisms
involved [163,175].
In vitro studies on primary striatal neurons demonstrated the attenuation
of oxidized low-density lipoprotein(Ox-LDL)-induced neurotoxicity by phenolic compounds, indicating overall that avonoids, in particular epicatechin,
250
kaempferol, and cyanidin, are more eective in attenuating Ox-LDL-induced

neurotoxicity than hydroxycinnamates or vitamin C [176]. In subsequent studies
the same authors demonstrated that the protective eect of epicatechin against
Ox-LDL-induced neurotoxicity is seemingly independent of the classical hydrogen-donating antioxidant properties of epicatechin [177]. This conclusion was
based on the following observations:
1.
2.
3.
Epicatechin does not prevent the oxLDL-induced oxidative shift in the

intracellular redox state [177].
Common antioxidants, such as ascorbic acid, only weakly protect
against Ox-LDL-induced neurotoxicity when used in 10 times higher
concentrations than epicatechin [176].
3V-O-methyl-epicatechin, one of the major in vivo metabolites of
epicatechin and a compound with signicantly lower antioxidant
capacity than epicatechin, equals epicatechin in its ability to prevent
neuronal death [177].
The authors indicated that the neuroprotective eects of epicatechin and 3V-Omethyl-epicatechin might be based on their modulation of apoptotic cellular
signaling pathways [177] (discussed later). Interestingly, Spencer and colleagues
[178] reported that epicatechin and 3V-O-methyl-epicatechin are equally eective
in protecting primary cortical neurons against hydrogen peroxideinduced neuronal death, whereas the glucuronidated in vivo metabolite epicatechin-5-h-O-Dglucuronide exerts no signicant protection in the same model. This dierence in
bioactivity was based on the inability of epicatechin-5-h-O-D-glucuronide to
enter cells as determined by liquid chromatography mass spectrometry (LC-MS/
MS) analysis [178].
Observations in neuronal PC12 cells demonstrated the protective eects of
Ginkgo biloba extract against hydrogen peroxidemediated [179] or h-amyloidinduced neuronal death [180,181]. Other authors reported protective eect of
Ginkgo biloba extract against h-amyloid-mediated neurotoxicity in primary
hippocampal neurons [147,182] and implicated antioxidant eects and the
modulation of intracellular calcium levels in PC12 cells as possible mechanisms
of action [179]. Other lines of research focused on the binding of avonoids such
as apigenin, naringenin, kaempferol, and quercetin-3-O-glucoside to benzodiazepine binding sites [183] of dierent receptors, including the g-aminobutyric
acid A(GABA-A)-receptor [184,185] and adenosine receptors [186], and
investigated their anxiolytic potential [183] as a possible mechanism of action
in the CNS. In addition, the levels of neurotransmitters involved in the
pathophysiological mechanisms of mood disorders, such as 5-hydroxytryptamine, noradrenaline, and dopamine, have been found to be modied in cortex,
diencephalons, and brainstem of rats after the administration of a avonoidcontaining preparation [187,188]. Bastianetto and coworkers [148] report that
251
antioxidant eects are not the only mechanism of avonoid-mediated protection

against neuronal death and show attenuation of nitric oxideinduced activation of
protein kinase C (PKC) to be partially involved in the protection against
neurotoxicity. However, the precise mechanisms by which avonoids exert their
neuroprotective actions in vivo and in vitro are presently unknown, and it is
currently unclear whether or not these compounds function solely as hydrogendonating antioxidants or exert their neuroprotective actions independently of
such properties.
Furthermore, only a very few investigators have studied the inuence of
metabolism on the bioactivity of avonoids [177,178,189191]; thus most of
the eects reported, on the basis of in vitro experiments, cannot be extrapolated
to in vivo situations. For example, one of the most intensively investigated
avonoids, quercetin, is often only used as the aglycone in cell culture systems
and interesting biological activities such as modulation of the multidrugresistance protein [192], induction of apoptosis [193], inhibition of the expression of nitric oxide synthase and cyclooxygenase [194], protection against
oxidative stress [149], and modulation of the calcium homeostasis [179] have
been reported. However, in plants and therefore also in most plant-derived foods
and beverages, quercetin is predominantly present in a glycosylated form that
completely alters its bioactivity [195197]. If digested, the glycosides might be
taken up or be cleaved and the released quercetin metabolized so that there is
almost no free quercetin aglycone in circulation. Thus the bioactivities observed
when using the aglycone alone may be very dierent from the eects of the
derivatives of quercetin. It is therefore vital for cell culture experiments to
analyze, synthesize, and investigate the in vivo metabolites of the avonoid
under investigation.
V.
WHAT ARE THE POTENTIAL MECHANISMS AND

TARGETS FOR THE BIOACTIVITY OF FLAVONOIDS?
A.
Do Flavonoids Act as Antioxidants In Vivo?
There has been considerable interest in recent years in the cytoprotective and
neuroprotective eects of avonoids, especially in the context of their modes of
action as antioxidants. The electron-donating properties of avonoids are well
dened to explain their antioxidant properties in vitro [1014]. Structurally
important features dening the reduction potential of avonoids are the hydroxylation pattern, a 3V,4V-dihydroxy catechol structure in the B-ring, the planarity of
the molecule, and the presence of 2,3 unsaturation in conjugation with a 4-oxofunction in the C-ring. Many studies have described the antioxidant ecacy of
avonoids and demonstrated that these polyphenols can inhibit the oxidation of
lipids [12,15,16] and other biomolecules such as proteins and DNA [1719] in
252
vitro. In addition, the ability of avonoids to act as antioxidants in vitro may be

based on their metal-chelating properties [22,23]. However, although avonoids
react rapidly with ROS/RNS in chemical systems in vitro, their reactions in vivo
are dependent on the form that is bioavailable to cells and tissues. This becomes
especially important if one considers that avonoids undergo substantial modication when metabolized in mammals, resulting in mainly methylated and/or
glucuronidated or sulfated primary metabolites with often lower antioxidant
capacity (for further information, see Chap. 14). In addition, the maximal blood
plasma concentration of avonoids and their structurally related metabolites after
ingestion of avonoid-rich food or beverages rarely reaches 5 AM. This compares
to an average of 65100 AM ascorbic acid and relatively high amounts of other
antioxidants such as uric acid and vitamin E in the plasma. Furthermore, the
slight increases in the antioxidant capacity of blood plasma that follow avonoid
administration to humans or animals are of a very temporary nature and often
decline rapidly to reach baseline levels 24 h later. On the basis of such
considerations the role of avonoids as important antioxidants in vivo seems
unlikely (exception: the gastrointestinal tract).
B.
Flavonoids as Modulators of Protein Activities

and Intracellular Signaling?
Reports on the bioactivity of avonoids or avonoid-rich plant extracts demonstrated that these compounds are able to alter cellular function seemingly
independently of their antioxidant potential. These actions include alterations
of receptor function [e.g., GABA-A-receptor (183,185)] and enzyme activities
[e.g., mitochondrial calcium adenosine triphosphatases (ATPases) (198)], inuences on the calcium homeostasis [179], and alterations of intracellular cell
signaling processes, especially with regard to protein kinase C [148] and
mitogen-activated protein kinase (MAPK) signaling. Accumulating evidence
suggests that avonoids interact selectively within MAPK signaling cascades
[177,195197,199,200]. This could have important implications with regard
to their possible sites of action in neurons since members of the MAPK family
are involved in signaling processes with regard to neuronal survival, regeneration,
and death [10,45,46].
In 2001, we demonstrated that pretreatment of primary striatal neurons
with low-micromolar concentrations of the avan-3-ol epicatechin and the
avonol kaempferol potently protected against Ox-LDL-induced neuronal
apoptosis [177]. The neuronal death induced by Ox-LDL was characterized by
a time- and dose-dependent decrease in MTT reduction and membrane integrity
and increase in annexin-V binding, caspase-3-like protease activity, and DNA
fragmentation. In addition, it was demonstrated that oxLDL is a rapid activator
253
of ERK1/2 and JNK in striatal neurons, thereby highlighting the potential

importance of MAPK cascades as pivotal mediators of oxidative stress signaling
in neurons [177]. More importantly, the Ox-LDL-induced activation of ERK1/2
and JNK was strongly inhibited when striatal neurons were pretreated with
epicatechin or kaempferol at concentrations that blocked Ox-LDL-induced
neuronal death. However, preventing ERK1/2 activation by using the MEK
inhibitors PD98059 or U0126 had no eect on neuronal loss caused by Ox-LDL.
This nding strongly suggests that the neuroprotective eects of epicatechin and
kaempferol are not linked to their potent inhibitory action within the ERK1/2
cascade. The Ox-LDL-induced activation of JNK was also potently attenuated by
epicatechin and kaempferol pretreatments. Whereas other authors have reported a
direct link between JNK activation and neurotoxicity [7981], in the absence of a
commercially available inhibitor of JNK it remains to be demonstrated that OxLDL-induced JNK activation, as observed here, is directly linked to neuronal
apoptosis. However, the phosphorylation of the activator protein 1 (AP-1) protein
c-Jun on Ser-63 and Ser-73 by JNK causes increased transcriptional activity [70],
which has been linked to stress-induced apoptosis. Ox-LDL caused a pronounced
phosphorylation of c-Jun, which mirrored the activation of JNK. Signicantly,
the Ox-LDL-induced activation of c-Jun was abolished in neurons pre-exposed to
epicatechin. The complete inhibition of phosphorylation of c-Jun by avonoids
suggests that these polyphenolic compounds may have powerful antiapoptotic
actions in neurons. Interestingly, Ox-LDL also induced the cleavage of procaspase-3 and thereby the activation of caspase-3 in striatal neurons. The activation
of caspase-3 was completely blocked in avonoid pretreated neurons, providing
compelling evidence in support of a potent antiapoptotic action of avonoids in
these cells. This is further supported by the fact that blocking caspase-3 activity
with a selective inhibitor protected neurons against Ox-LDL-induced neurotoxicity [177]. These results are of special interest with regard to the proposed
links among JNK activation, impairment of mitochondria function, cytochrome
c release, and caspase activation (as discussed previously) [45,86].
Other reports of nonneuronal cells support a potential role for avonoids
as modulators of intracellular signaling, especially with regard to MAPKs.
Yoshizumi and associates [201] demonstrated that the avonol quercetin
attenuates the angiotensin IIinduced activation of JNK in cultured rat aortic
smooth muscle cells. Furthermore, in experiments with RAW 264.7 macrophages Wadsworth and colleagues [202] showed that quercetin inhibits the
lipopolysaccharide-(LPS)-induced TNF-a expression by inhibiting the activation of JNK, which subsequently results in a suppression of AP-1-DNA binding.
In addition, the avanol epigallocatechin has been suggested to suppress the
proliferation of vascular smooth muscle cells via the inhibition of JNK
activation and c-Jun expression [203].
254
Another interesting target for avonoids, which is potentially linked to

calcium homeostasis, cell signaling processes, and apoptosis is the mitochondrion. Mitochondria play a central role in oxidative stressinduced
apoptosis [204] since they contain cytochrome c, thought to be essential in
Fas-receptor-independent apoptosis [86,205]. The emerging view is that after
mitochondrial depolarization cytochrome c is released, committing the cell to
die by either an apoptotic mechanism, involving Apaf-1-mediated caspase-3
activation, or a necrotic process, due to the collapse of the mitochondrial
electron transport chain [204]. Thus, blocking the release of cytochrome c
might be the key to preventing neuronal death. However, other investigators
reported apoptotic processes involving the depolarization of the mitochondria
without translocation of cytochrome c to the cytosol [204,206], thus potentially indicating additional mechanisms leading to the recruitment of apoptosis-executing caspases.
Speculatively, the avonoid-mediated inhibition of apoptosis as demonstrated by Schroeter and coworkers [176,177] and others might occur by a
variety of means related to mitochondrial functions, rst, by blocking the
activation of JNK. This might take place upstream of JNK, inuencing one of
many MAPKKK activating proteins involved in transducing signals to JNK [or
at the level of maintaining the calcium homeostasis, which has been shown to be
important in JNK activation]. Second, avonoids might be able to upregulate the
expression of the antiapoptotic proteins Bcl-2 and Bcl-xL known to inhibit the
release of cytochrome c from the mitochondria and to be inactivated through
phosphorylation by JNK. Or, third, avonoids might modulate the mitochondrial
permeability transition (MPT) believed to be important in apoptotic cell death
by either opening a gateway for cytochrome c release from the mitochondria
[204] or activating other mitochondrial-related proapoptotic factors such as
DIABLO/smac [207,208]. Inhibitors of MPT pore opening such as cyclosporin
A bind to cyclophilin D, which is associated with the adenine nucleotide
transporter (ANT), part of the multiprotein complex of the MPT pore
[204,209], and inhibit mitochondrial depolarization [206], cytochrome c translocation [210], and cell death [206,210]. Conversely, MPT pore openers such as
the ANT-activator atractyloside trigger MPT pore opening, cytochrome c
release, and apoptosis [210]. Interestingly, the MPT pore possesses a benzodiazepine binding site [204], and the binding of ligands such as PK11195 [211,212]
and antagonists such as umazenil [213] have been shown to modulate the MPT
pore and thus the MPT. Since avonoids have been reported to bind to
benzodiazepine binding sites of GABA and adenosine receptors [183,185], it
might be speculated that these compounds exert an eect on the MPT pore and
therefore modulate cytochrome c release. Taken together avonoids might
inuence the release of proapoptotic cytochrome c from the mitochondrion,
thus providing an interesting hypothesis for future investigations.
1.
255
Flavonoids as Modulators of Adenosine Triphosphate Binding

Sites: Structure-Activity Relationship
As detailed, a variety of biological eects of avonoids may be related to their

binding to ATP-binding sites of various proteins, including transport ATPases
and protein kinases. In fact, quite a few commercially available protein kinase
inhibitors have been developed on the basis of the structure of quercetin. These
compounds include the widely used PI-3-K inhibitor LY294002 (IC50 c 15
AM; the IC50 of quercetin is approximately 48 AM), the MEK inhibitor
Figure 7 Structural comparison of epicatechin, quercetin, and widely used

protein kinase and ATPase inhibitors. Flavonoids might exert their effects by
modulating the activity of protein kinases and ATPases on the basis of binding to
ATP-binding sites. This picture demonstrates structural similarities of classical
kinase inhibitors, quercetin, and epicatechin. The proposed structural requirements for affinity to ATP-binding sites include (with varying degree of importance)
A- and C-ring moiety of flavonoids, the 5-hydroxyl group, and the 4-carbonyl
function, which are supposed to mimic the adenine moiety of ATP. Although the Bring moiety may not be directly involved in binding to the ATP-binding domain, it
seems to be a determinant of the protein specificity of the inhibitor. ATP,
adenosine triphosphate.
256
PD98059 (IC50 c 24 AM;), the JNK inhibitor SP600129 (IC50 c 90 nM), the
tyrosine kinase inhibitors aminogenistein (IC50 c 12 AM) and emodin (IC50 c
519 AM), and the CDK2 inhibitor L868276 (IC50 c 0.41.6 AM). The
structural characteristics responsible for interactions with ATP-binding sites
on a variety of proteins have been investigated [214216] (Fig. 7):
1.
2.
3.
4.
Rings A and B of the avonoid structure mimic the adenine ring

system of ATP.
The 5-hydroxyl group mimics the 6-amino group of ATP and may
interact with up to six protein residues of the target protein (the 7-OH
group seems not to be involved).
The 4-carbonyl function may enhance the inhibitor activity by
mimicking the N1 of adenine.
The B-ring moiety of the avonoid structure seems to be important for
orientating and directing rings A and B toward the ATP-binding site.
Derivatization of the hydroxyl groups on the B-ring has been implicated
in modulation of both protein specicity and inhibitor activity.
Thus these ndings substantiate the potential interaction of avonoids

with ATP-binding sites on proteins as a potential mechanism for their biological
activities. The specic biological eect of a particular avonoid depends on its
bioavailable form, the intracellular concentrations achieved in target tissues, and
its anity to the ATP-binding site of the protein of interest.
VI.
CONCLUSIONS
Epidemiological studies and investigations in vivo and in cell culture systems

increasingly indicate avonoids as potential neuroprotective agents. Thus,
exploring the mechanisms of the biological eects of avonoids might help to
develop dietary and therapeutic approaches against neurodegeneration.
The classical hydrogen-donating antioxidant properties of avonoids,
which are often used as a possible explanation for their biological activity,
may, after all, not be the most important determinant for their eects in vivo. This
conclusion is tenable when factors such as in vivo metabolism and the achieved
plasma and tissue concentrations of avonoids/metabolites are taken into
account. Therefore, the bioactivity of avonoids in vivo may rather be supported
by their ability to modulate protein functions, intracellular cell signaling, and
receptor activities on the basis of interactions, for example, with ATP-binding
sites and benzodiazepine-binding domains.
With regard to a potential role for avonoids against neurodegeneration, the
availability of avonoids and their metabolites to the brain remains to be
investigated further. In this context it needs to be considered that perhaps the
257
benecial eects of avonoids on memory and motoric functions and against

age-related cognitive decline are not exerted directly in the brain but occur in the
periphery and are reected subsequently in functions of the CNS. It might
therefore be that factors such as platelet aggregation, vascular tone, hormonal
changes, and immune responses modulated by avonoid intake induce changes in
the CNS, which in turn aect neurodegenerative processes.
To understand fully the molecular basis for the in vivo eects of avonoids
and to assess comprehensively their potential as neuroprotective agents, further
studies are needed, in particular with regard to metabolism, distribution across the
blood-brain barrier, and neuropharmacological properties.
REFERENCES
1.
Halliwell B. Reactive oxygen species and the central nervous system. J Neurochem
1992; 59(5):16091623.
2. Finkel T, Holbrook NJ. Oxidants, oxidative stress and the biology of ageing.
Nature 2000; 408(6809):239247.
3. Halliwell B, Gutteridge JM. The importance of free radicals and catalytic metal
ions in human diseases. Mol Aspects Med 1985; 8(2):89193.
4. Bindoli A, Rigobello MP, Deeble DJ. Biochemical and toxicological properties
of the oxidation products of catecholamines. Free Radic Biol Med 1992;13(4):
391405.
5. Coyle JT, Puttfarcken P. Oxidative stress, glutamate, and neurodegenerative
disorders. Science 1993; 262(5134):689695.
6. Orrenius S, Nicotera P. The calcium ion and cell death. J Neural Transm Suppl
1994; 43:111.
7. Behl C. Alzheimers disease and oxidative stress: implications for novel
therapeutic approaches. Prog Neurobiol 1999; 57(3):301323.
8. Zhang Y, Dawson VL, Dawson TM. Oxidative stress and genetics in the
pathogenesis of Parkinsons disease. Neurobiol Dis 2000; 7(4):240250.
9. Alexi T, Borlongan CV, Faull RL, Williams CE, Clark RG, Gluckman PD, Hughes
PE. Neuroprotective strategies for basal ganglia degeneration: Parkinsons and
Huntingtons diseases. Prog Neurobiol 2000; 60(5):409470.
10. Yuan J, Yankner BA. Apoptosis in the nervous system. Nature 2000; 407(6805):
802809.
11. Quinn J, Kaye J. The neurology of aging. Neurologist 2001; 7:98112.
12. Oliver CN, Starke-Reed PE, Stadtman ER, Liu GJ, Carney JM, Floyd RA.
Oxidative damage to brain proteins, loss of glutamine synthetase activity, and
production of free radicals during ischemia/reperfusion-induced injury to gerbil
brain. Proc Natl Acad Sci USA 1990; 87(13):51445147.
13. McGeer PL, McGeer EG. The inammatory response system of brain:
implications for therapy of Alzheimer and other neurodegenerative diseases.
Brain Res Rev 1995; 21(2):195218.
258
14.
McGeer EG, McGeer PL. The importance of inammatory mechanisms in

Alzheimer disease. Exp Gerontol 1998; 33(5):371378.
Shigenaga MK, Hagen TM, Ames BN. Oxidative damage and mitochondrial decay
in aging. Proc Natl Acad Sci USA 1994; 91(23):1077110778.
Tsay HJ, Wang P, Wang SL, Ku HH. Age-associated changes of superoxide dismutase and catalase activities in the rat brain. J Biomed Sci 2000; 7(6):466474.
Keller JN, Huang FF, Markesbery WR. Decreased levels of proteasome activity and
proteasome expression in aging spinal cord. Neuroscience 2000; 98(1):149156.
Wakimoto P, Block G. Dietary intake, dietary patterns, and changes with age: an
epidemiological perspective. J Gerontol A Biol Sci Med Sci 2001; 56 (spec no 2)
(2):6580.
Winkler S, Garg AK, Mekayarajjananonth T, Bakaeen LG, Khan E. Depressed taste and smell in geriatric patients. J Am Dent Assoc 1999; 130(12):
17591765.
Chau N, Tebi A, Creton C, Belbraouet S, Debry G. Relationship between plasma
retinol and infectious diseases in the elderly: a case-control study. Ann Nutr Metab
2000; 44(56):256262.
Mohanakumar KP, de Bartolomeis A, Wu RM, Yeh KJ, Sternberger LM, Peng SY,
Murphy DL, Chiueh CC. Ferrous-citrate complex and nigral degeneration: evidence for free-radical formation and lipid peroxidation. Ann NY Acad Sci 1994;
738:392399.
Keller JN, Mattson MP. Roles of lipid peroxidation in modulation of cellular
signaling pathways, cell dysfunction, and death in the nervous system. Rev
Neurosci 1998; 9(2):105116.
Ben-Yoseph O, Boxer PA, Ross BD. Assessment of the role of the glutathione and
pentose phosphate pathways in the protection of primary cerebrocortical cultures
from oxidative stress. J Neurochem 1996; 66(6):23292337.
Buttereld DA, Kanski J. Brain protein oxidation in age-related neurodegenerative
disorders that are associated with aggregated proteins. Mech Ageing Dev 2001;
122(9):945962.
Vitek MP, Bhattacharya K, Glendening JM, Stopa E, Vlassara H, Bucala R,
Manogue K, Cerami A. Advanced glycation end products contribute to amyloidosis
in Alzheimer disease. Proc Natl Acad Sci USA 1994; 91(11):47664770.
Garrington TP, Johnson GL. Organization and regulation of mitogen-activated
protein kinase signaling pathways. Curr Opin Cell Biol 1999; 11(2):211218.
Chang L, Karin M. Mammalian MAP kinase signalling cascades. Nature 2001;
410(6824):3740.
Huber LA, Ullrich O, Takai Y, Lutcke A, Dupree P, Olkkonen V, Virta H, de
Hoop MJ, Alexandrov K, Peter M. Mapping of Ras-related GTP-binding proteins
by GTP overlay following two-dimensional gel electrophoresis. Proc Natl Acad
Sci USA 1994; 91(17):78747878.
Arozarena I, Aaronson DS, Matallanas D, Sanz V, Ajenjo N, Tenbaum SP,
Teramoto H, Ighishi T, Zabala JC, Gutkind JS, Crespo P. The Rho family GTPase
Cdc42 regulates the activation of Ras/MAP kinase by the exchange factor RasGRF. J Biol Chem 2000; 275(34):2644126448.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.

30.
259
Xia Z, Dickens M, Raingeaud J, Davis RJ, Greenberg ME. Opposing eects of

ERK and JNK-p38 MAP kinases on apoptosis. Science 1995; 270:13261331.
31. Rosen LB, Ginty DD, Weber MJ, Greenberg ME. Membrane depolarization and
calcium inux stimulate MEK and MAP kinase via activation of Ras. Neuron
1994; 12(6):12071221.
32. Bonni A, Brunet A, West AE, Datta SR, Takasu MA, Greenberg ME. Cell survival
promoted by the Ras-MAPK signaling pathway by transcription-dependent and
-independent mechanisms. Science 1999; 286(5443):13581362.
33. Wang X, Martindale JL, Liu Y, Holbrook NJ. The cellular response to oxidative
stress: inuences of mitogen-activated protein kinase signalling pathways on cell
survival. Biochem J 1998; 333(pt 2):291300.
34. Perkinton MS, Sihra TS, Williams RJ. Ca(2+)-permeable AMPA receptors induce
phosphorylation of cAMP response element-binding protein through a phosphatidylinositol 3-kinase-dependent stimulation of the mitogen-activated protein
kinase signaling cascade in neurons. J Neurosci 1999; 19(14):58615874.
35. Zippel R, Balestrini M, Lomazzi M, Sturani E. Calcium and calmodulin are
essential for Ras-GRF1-mediated activation of the Ras pathway by lysophosphatidic acid. Exp Cell Res 2000; 258(2):403408.
36. Enslen H, Tokumitsu H, Stork PJ, Davis RJ, Soderling TR. Regulation of mitogenactivated protein kinases by a calcium/calmodulin-dependent protein kinase
cascade. Proc Natl Acad Sci USA 1996; 93(20):1080310808.
37. Whitmarsh AJ, Davis RJ. Transcription factor AP-1 regulation by mitogenactivated protein kinase signal transduction pathways. J Mol Med 1996; 74(10):
589607.
38. Sturgill TW, Ray LB, Erikson E, Maller JL. Insulin-stimulated MAP-2 kinase
phosphorylates and activates ribosomal protein S6 kinase II. Nature 1988; 334
(6184):715718.
39. Deak M, Clifton AD, Lucocq LM, Alessi DR. Mitogen- and stress-activated
protein kinase-1 (MSK1) is directly activated by MAPK and SAPK2/p38, and may
mediate activation of CREB. EMBO J 1998; 17(15):44264441.
40. Mazzoni IE, Said FA, Aloyz R, Miller FD, Kaplan D. Ras regulates sympathetic
neuron survival by suppressing the p53-mediated cell death pathway. J Neurosci
1999; 19(22):97169727.
41. Alessandrini A, Namura S, Moskowitz MA, Bonventre JV. MEK1 protein kinase
inhibition protects against damage resulting from focal cerebral ischemia. Proc
Natl Acad Sci USA 1999; 96(22):1286612869.
42. Stanciu M, Wang Y, Kentor R, Burke N, Watkins S, Kress G, Reynolds I, Klann E,
Angiolieri MR, Johnson JW, DeFranco DB. Persistent activation of ERK
contributes to glutamate-induced oxidative toxicity in a neuronal cell line and
primary cortical neuron cultures. J Biol Chem 2000; 275(16):1220012206.
43. Murray B, Alessandrini A, Cole AJ, Yee AG, Furshpan EJ. Inhibition of the p44/
42 MAP kinase pathway protects hippocampal neurons in a cell-culture model of
seizure activity. Proc Natl Acad Sci USA 1998; 95(20):1197511980.
44. Satoh T, Nakatsuka D, Watanabe Y, Nagata I, Kikuchi H, Namura S.
Neuroprotection by MAPK/ERK kinase inhibition with U0126 against oxidative
260
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.

stress in a mouse neuronal cell line and rat primary cultured cortical neurons.
Neurosci Lett 2000; 288(2):163166.
Davis RJ. Signal transduction by the JNK group of MAP kinases. Cell 2000;
109:239252.
Mielke K, Herdegen T. JNK and p38 stress kinasesdegenerative eectors of
signal-transduction-cascades in the nervous system. Prog Neurobiol 2000; 61
(1):4560.
Ferrer I, Planas AM, Pozas E. Radiation-induced apoptosis in developing rats and
kainic acid-induced excitotoxicity in adult rats are associated with distinctive
morphological and biochemical c-Jun/AP-1 (N) expression. Neurosci 1997; 80(2):
449458.
Gibson S, Widmann C, Johnson GL. Dierential involvement of MEK kinase 1
(MEKK1) in the induction of apoptosis in response to microtubule-targeted drugs
versus DNA damaging agents. J Biol Chem 1999; 274(16):1091610922.
Iordanov MS, Wong J, Newton DL, Rybak SM, Bright RK, Flavell RA, Davis
RJ, Magun BE. Dierential requirement for the stress-activated protein
kinase/c-Jun NH(2)-terminal kinase in RNA damage-induced apoptosis in
primary and in immortalized broblasts. Mol Cell Biol Res Commun 2000;
4(2):122128.
Meier R, Rouse J, Cuenda A, Nebreda AR, Cohen P. Cellular stresses and
cytokines activate multiple mitogen-activated-protein kinase kinase homologues in
PC12 and KB cells. Eur J Biochem 1996; 236(3):796805.
Meeker R, Fernandes A. Osmotic and glutamate receptor regulation of c-Jun
NH(2)-terminal protein kinase in neuroendocrine cells. Am J Physiol Endocrinol
Metab 2000; 279(3):E475E486.
Chihab R, Ferry C, Koziel V, Monin P, Daval JL. Sequential activation of activator
protein-1-related transcription factors and JNK protein kinases may contribute to
apoptotic death induced by transient hypoxia in developing brain neurons. Brain
Res Mol Brain Res 1998; 63(1):105120.
Herdegen T, Claret FX, Kallunki T, Martin-Villalba A, Winter C, Hunter T, Karin
M. Lasting N-terminal phosphorylation of c-Jun and activation of c-Jun N-terminal
kinases after neuronal injury. J Neurosci 1998; 18(14):51245135.
Mielke K, Brecht S, Dorst A, Herdegen T. Activity and expression of JNK1, p38
and ERK kinases, c-Jun N-terminal phosphorylation, and c-Jun promoter binding
in the adult rat brain following kainate-induced seizures. Neurosci 1999; 91(2):
471483.
Go YM, Patel RP, Maland MC, Park H, Beckman JS, Darley-Usmar VM, Jo H.
Evidence for peroxynitrite as a signaling molecule in ow-dependent activation of
c-Jun NH(2)-terminal kinase. Am J Physiol 1999; 277(4 pt 2):H1647H1653.
Anania FA, Womack L, Jiang M, Saxena NK. Aldehydes potentiate alpha(2)(I)
collagen gene activity by JNK in hepatic stellate cells. Free Radic Biol Med 2001;
30(8):846857.
Soh Y, Jeong KS, Lee IJ, Bae MA, Kim YC, Song BJ. Selective activation of the
c-Jun N-terminal protein kinase pathway during 4-hydroxynonenal-induced
apoptosis of PC12 cells. Mol Pharmacol 2000; 58(3):535541.

58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
261
Tibbles LA, Ing YL, Kiefer F, Chan J, Iscove N, Woodgett JR, Lassam NJ. MLK-3
activates the SAPK/JNK and p38/RK pathways via SEK1 and MKK3/6. EMBO J
1996; 15(24):70267035.
Hirai S, Izawa M, Osada S, Spyrou G, Ohno S. Activation of the JNK pathway
by distantly related protein kinases, MEKK and MUK. Oncogene 1996; 12(3):
641650.
Hoeich KP, Yeh WC, Yao Z, Mak TW, Woodgett JR. Mediation of TNF
receptor-associated factor eector functions by apoptosis signal-regulating kinase1 (ASK1). Oncogene 1999; 18(42):58145820.
Shiah SG, Chuang SE, Kuo ML. Involvement of Asp-Glu-Val-Asp-directed,
caspase-mediated mitogen-activated protein kinase kinase 1 Cleavage, c-Jun Nterminal kinase activation, and subsequent Bcl-2 phosphorylation for paclitaxelinduced apoptosis in HL-60 cells. Mol Pharmacol 2001; 59(2):254262.
Takekawa M, Saito H. A family of stress-inducible GADD45-like proteins mediate
activation of the stress-responsive MTK1/MEKK4 MAPKKK. Cell 1998; 95(4):
521530.
Kiyono M, Satoh T, Kaziro Y. G protein beta gamma subunit-dependent
Racguanine nucleotide exchange activity of Ras-GRF1/CDC25(Mm). Proc Natl
Acad Sci USA 1999; 96(9):48264831.
Adler V, Polotskaya A, Wagner F, Kraft AS. Ainity-puried c-Jun aminoterminal protein kinase requires serine/threonine phosphorylation for activity.
J Biol Chem 1992; 267(24):1700117005.
Kallunki T, Su B, Tsigelny I, Sluss HK, Derijard B, Moore G, Davis R, Karin M.
JNK2 contains a specicity-determining region responsible for ecient c-Jun
binding and phosphorylation. Genes Dev 1994; 8(24):29963007.
Gupta S, Barrett T, Whitmarsh AJ, Cavanagh J, Sluss HK, Derijard B, Davis RJ.
Selective interaction of JNK protein kinase isoforms with transcription factors.
EMBO J 1996; 15(11):27602770.
Kallunki T, Deng T, Hibi M, Karin M. c-Jun can recruit JNK to phosphorylate dimerization partners via specic docking interactions. Cell 1996; 87(5):
929939.
Cavigelli M, Dol F, Claret FX, Karin M. Induction of c-fos expression through
JNK-mediated TCF/Elk-1 phosphorylation. EMBO J 1995; 14(23):59575964.
Karin M, Liu Z, Zandi E. AP-1 function and regulation. Curr Opin Cell Biol 1997;
9(2):240246.
Pulverer BJ, Kyriakis JM, Avruch J, Nikolakaki E, Woodgett JR. Phosphorylation
of c-jun mediated by MAP kinases. Nature 1991; 353(6345):670674.
Buschmann T, Yin Z, Bhoumik A, Ronai Z. Amino-terminal-derived JNK
fragment alters expression and activity of c-Jun, ATF2, and p53 and increases
H2O2-induced cell death. J Biol Chem 2000; 275(22):1659016596.
Fuchs SY, Dolan L, Davis RJ, Ronai Z. Phosphorylation-dependent targeting of cJun ubiquitination by Jun N-kinase. Oncogene 1996; 13(7):15311535.
Fuchs SY, Xie B, Adler V, Fried VA, Davis RJ, Ronai Z. c-Jun NH2-terminal
kinases target the ubiquitination of their associated transcription factors. J Biol
Chem 1997; 272(51):3216332168.
262
74.
Yang DD, Kuan CY, Whitmarsh AJ, Rincon M, Zheng TS, Davis RJ, Rakic P,
Flavell RA. Absence of excitotoxicity-induced apoptosis in the hippocampus of
mice lacking the Jnk3 gene. Nature 1997; 389(6653):865870.
Behrens A, Sibilia M, Wagner EF. Amino-terminal phosphorylation of c-Jun
regulates stress-induced apoptosis and cellular proliferation. Nat Genet 1999;
21(3):326329.
Whiteld J, Neame SJ, Paquet L, Bernard O, Ham J. Dominant-negative c-Jun
promotes neuronal survival by reducing BIM expression and inhibiting
mitochondrial cytochrome c release. Neuron 2001; 29(3):629643.
Eilers A, Whiteld J, Shah B, Spadoni C, Desmond H, Ham J. Direct inhibition
of c-Jun N-terminal kinase in sympathetic neurons prevents c-jun promoter activation and NGF withdrawal-induced death. J Neurochem 2001; 76(5):14391454.
Liu YF. Expression of polyglutamine-expanded Huntingtin activates the SEK1JNK pathway and induces apoptosis in a hippocampal neuronal cell line. J Biol
Chem 1998; 273(44):2887328877.
Troy CM, Rabacchi SA, Xu Z, Maroney AC, Connors TJ, Shelanski ML, Greene
LA. Beta-Amyloid-induced neuronal apoptosis requires c-Jun N-terminal kinase
activation. J Neurochem 2001; 77(1):157164.
Maroney AC, Finn JP, Bozyczko-Coyne D, OKane TM, Ne NT, Tolkovsky AM,
Park DS, Yan CY, Troy CM, Greene LA. CEP-1347 (KT7515), an inhibitor of JNK
activation, rescues sympathetic neurons and neuronally dierentiated PC12 cells
from death evoked by three distinct insults. J Neurochem 1999; 73(5):19011912.
Saporito MS, Thomas BA, Scott RW. MPTP activates c-Jun NH(2)-terminal
kinase (JNK) and its upstream regulatory kinase MKK4 in nigrostriatal neurons in
vivo. J Neurochem 2000; 75(3):12001208.
Shimoke K, Yamagishi S, Yamada M, Ikeuchi T, Hatanaka H. Inhibition of
phosphatidylinositol 3-kinase activity elevates c-Jun N-terminal kinase activity in
apoptosis of cultured cerebellar granule neurons. Brain Res Dev Brain Res 1999;
112(2):245253.
Faris M, Latinis KM, Kempiak SJ, Koretzky GA, Nel A. Stress-induced Fas ligand
expression in T cells is mediated through a MEK kinase 1-regulated response
element in the Fas ligand promoter. Mol Cell Biol 1998; 18(9):54145424.
Buschmann T, Potapova O, Bar-Shira A, Ivanov VN, Fuchs SY, Henderson S,
Fried VA, Minamoto T, Alarcon-Vargas D, Pincus MR, Gaarde WA, Holbrook NJ,
Shiloh Y, Ronai Z. Jun NH2-terminal kinase phosphorylation of p53 on Thr-81 is
important for p53 stabilization and transcriptional activities in response to stress.
Mol Cell Biol 2001; 21(8):27432754.
Fuchs SY, Adler V, Pincus MR, Ronai Z. MEKK1/JNK signaling stabilizes and
activates p53. Proc Natl Acad Sci USA 1998; 95(18):1054110546.
Tournier C, Hess P, Yang DD, Xu J, Turner TK, Nimnual A, Bar-Sagi D, Jones
SN, Flavell RA, Davis RJ. Requirement of JNK for stress-induced activation of the
cytochrome c-mediated death pathway. Science 2000; 288(5467):870874.
Yoshida H, Kong YY, Yoshida R, Elia AJ, Hakem A, Hakem R, Penninger JM,
Mak TW. Apaf1 is required for mitochondrial pathways of apoptosis and brain
development. Cell 1998; 94(6):739750.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.

88.
263
Hakem R, Hakem A, Duncan GS, Henderson JT, Woo M, Soengas MS, Elia A, de
la Pompa JL, Kagi D, Khoo W, Potter J, Yoshida R, Kaufman SA, Lowe SW,
Penninger JM, Mak TW. Dierential requirement for caspase 9 in apoptotic
pathways in vivo. Cell 1998; 94(3):339352.
89. Woo M, Hakem R, Soengas MS, Duncan GS, Shahinian A, Kagi D, Hakem A,
McCurrach M, Khoo W, Kaufman SA, Senaldi G, Howard T, Lowe SW, Mak TW.
Essential contribution of caspase 3/CPP32 to apoptosis and its associated nuclear
changes. Genes Dev 1998; 12(6):806819.
90. Kang CD, Jang JH, Kim KW, Lee HJ, Jeong CS, Kim CM, Kim SH,
Chung BS. Activation of c-jun N-terminal kinase/stress-activated protein
kinase and the decreased ratio of Bcl-2 to Bax are associated with the autooxidized dopamine-induced apoptosis in PC12 cells. Neurosci Lett 1998;
256(1):3740.
91. Fan MY, Goodwin M, Vu T, Brantley-Finley C, Gaarde WA, Chambers TC.
Vinblastine-induced phosphorylation of Bcl-2 and Bcl-X-L is mediated by JNK
and occurs in parallel with inactivation of the Raf-1/MEK/ERK cascade. J Biol
Chem 2000; 275:2998029985.
92. Choi WS, Yoon SY, Chang, II, Choi EJ, Rhim H, Jin BK, Oh TH, Krajewski S,
Reed JC, Oh YJ. Correlation between structure of Bcl-2 and its inhibitory function
of JNK and caspase activity in dopaminergic neuronal apoptosis. J Neurochem
2000; 74(4):16211626.
93. Allen RG, Tresini M. Oxidative stress and gene regulation. Free Radic Biol Med
2000; 28(3):463499.
94. Lander HM, Ogiste JS, Pearce SF, Levi R, Novogrodsky A. Nitric oxidestimulated guanine nucleotide exchange on p21ras. J Biol Chem 1995; 270(13):
70177020.
95. Lander HM, Ogiste JS, Teng KK, Novogrodsky A. p21ras as a common signaling
target of reactive free radicals and cellular redox stress. J Biol Chem 1995;
270(36):2119521198.
96. Burdon RH. Superoxide and hydrogen peroxide in relation to mammalian cell
proliferation. Free Radic Biol Med 1995; 18(4):775794.
97. Satoh T, Nakatsuka D, Watanabe Y, Nagata I, Kikuchi H, Namura S.
Neuroprotection by MAPK/ERK kinase inhibition with U0126 against oxidative
stress in a mouse neuronal cell line and rat primary cultured cortical neurons.
Neurosci Lett 2000; 288(2):163166.
98. Oguro T, Hayashi M, Nakajo S, Numazawa S, Yoshida T. The expression of heme
oxygenase-1 gene responded to oxidative stress produced by phorone, a glutathione
depletor, in the rat liver: the relevance to activation of c-jun n-terminal kinase.
J Pharmacol Exp Ther 1998; 287(2):733778.
99. Schwarzschild MA, Cole RL, Hyman SE. Glutamate, but not dopamine, stimulates
stress-activated protein kinase and AP-1-mediated transcription in striatal neurons.
J Neurosci 1997; 17(10):34553466.
100. Park HS, Huh SH, Kim MS, Lee SH, Choi EJ. Nitric oxide negatively regulates cJun N-terminal kinase/stress-activated protein kinase by means of S-nitrosylation.
Proc Natl Acad Sci USA 2000; 97(26):1438214387.
264
101.
Adler V, Yin Z, Fuchs SY, Benezra M, Rosario L, Tew KD, Pincus MR, Sardana
M, Henderson CJ, Wolf CR, Davis RJ, Ronai Z. Regulation of JNK signaling by
GSTp. EMBO J 1999; 18(5):13211334.
Yin Z, Ivanov VN, Habelhah H, Tew K, Ronai Z. Glutathione S-transferase p
elicits protection against H2O2-induced cell death via coordinated regulation of
stress kinases. Cancer Res 2000; 60(15):40534057.
Saitoh M, Nishitoh H, Fujii M, Takeda K, Tobiume K, Sawada Y, Kawabata M,
Miyazono K, Ichijo H. Mammalian thioredoxin is a direct inhibitor of apoptosis
signal-regulating kinase (ASK) 1. EMBO J 1998; 17(9):25962606.
Yin ZM, Liu AH, Jiang Y. Glutathione S-transferase pi protects serum depletioninduced cell death by inhibiting ASK1-MKK7-JNK pathway in the 293 cells. Acta
Biochim Biophys Sin 2001; 33:185190.
Cho SG, Lee YH, Park HS, Ryoo K, Kang KW, Park J, Eom SJ, Kim MJ,
Chang TS, Choi SY, Shim J, Kim Y, Dong MS, Lee MJ, Kim SG, Ichijo H,
Choi EJ. Glutathione S-transferase mu modulates the stress-activated signals by
suppressing apoptosis signal-regulating kinase 1. J Biol Chem 2001; 276(16):
1274912755.
Nimnual AS, Yatsula BA, Bar-Sagi D. Coupling of Ras and Rac guanosine
triphosphatases through the Ras exchanger Sos. Science 1998; 279(5350):
560563.
Samanta S, Perkinton MS, Morgan M, Williams RJ. Hydrogen peroxide enhances
signal-responsive arachidonic acid release from neurons: role of mitogen-activated
protein kinase. J Neurochem 1998; 70(5):20822090.
Vanhoutte P, Barnier JV, Guibert B, Pages C, Besson MJ, Hipskind RA, Caboche
J. Glutamate induces phosphorylation of Elk-1 and CREB, along with c-fos
activation, via an extracellular signal-regulated kinase-dependent pathway in brain
slices. Mol Cell Biol 1999; 19(1):136146.
Garthwaite J, Boulton CL. Nitric oxide signaling in the central nervous system.
Annu Rev Physiol 1995; 57:683706.
Dawson VL, Dawson TM, London ED, Bredt DS, Snyder SH. Nitric oxide
mediates glutamate neurotoxicity in primary cortical cultures. Proc Natl Acad Sci
USA 1991; 88:63686371.
Lipton SA, Choi Y-B, Pan Z-H, Lei SZ, Chen H-SV, Sucher NJ, Loscalzo J, Singel
DJ, Stamler JS. A redox-based mechanism for the neuroprotective and neurodestructive eects of nitric oxide and related nitroso-compounds. Nature 1993;
364:626632.
Tatton WG, Olanow CW. Apoptosis in neurodegenerative diseases: the role of
mitochondria. Biochim Biophys Acta 1999; 1410:195213.
Nicholls DG, Ward MW. Mitochondrial membrane potential and neuronal
glutamate excitotoxicity: mortality and millivolts. Trends Neurosci 2000;
23:166174.
Liberatore GT, Jackson-Lewis V, Vukosavic S, Mandir AS, Vila M, McAulie
WG, Dawson VL, Dawson TM, Przedborski S. Inducible nitric oxide synthase
stimulates dopaminergic neurodegeneration in the MPTP model of Parkinson
disease. Nat Med 1999; 5(12):14031409.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.

115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
265
Kim WG, Mohney RP, Wilson B, Jeohn GH, Liu B, Hong JS. Regional dierence
in susceptibility to lipopolysaccharide-induced neurotoxicity in the rat brain: role
of microglia. J Neurosci 2000; 20(16):63096316.
Schaeer HF, Weber MJ. Mitogen-activated protein kinase: specic messages
from ubiquitous messengers. Mol Cell Biol 1999; 19:24352444.
Lander HM, Hajjar DP, Hempstead BL, Mirza UA, Chait BT, Campbell S,
Quilliam LA. A molecular redox switch on p21(ras). Structural basis for the nitric
oxide-p21(ras) interaction. J Biol Chem 1997; 272(7):43234326.
So HS, Park RK, Kim MS, Lee SR, Jung BH, Chung SY, Jun CD, Chung HT.
Nitric oxide inhibits c-Jun N-terminal kinase 2 (JNK2) via S-nitrosylation.
Biochem Biophys Res Commun 1998; 247(3):809813.
Marshall HE, Merchant K, Stamler JS. Nitrosation and oxidation in the regulation
of gene expression. FASEB J 2000; 14:18891900.
Zipper LM, Mulcahy RT. Inhibition of ERK and p38 MAP kinases inhibits binding
of Nrf2 and induction of GCS genes. Biochem Biophys Res Commun 2000;
278(2):484492.
Chang MS, Yoo HY, Rho HM. Positive and negative regulatory elements in the
upstream region of the rat Cu/Zn-superoxide dismutase gene. Biochem J 1999;
339(pt 2):335341.
Das KC, Lewis-Molock Y, White CW. Activation of NF-kappa B and elevation of
MnSOD gene expression by thiol reducing agents in lung adenocarcinoma (A549)
cells. Am J Physiol 1995; 269(5 pt 1):L588L602.
Seo SJ, Kim HT, Cho G, Rho HM, Jung G. Sp1 and C/EBP-related factor
regulate the transcription of human Cu/Zn SOD gene. Gene 1996; 178(12):
177185.
Sgambato V, Pages C, Rogard M, Besson MJ, Caboche J. Extracellular signalregulated kinase (ERK) controls immediate early gene induction on corticostriatal
stimulation. J Neurosci 1998; 18(21):88148825.
Kan H, Xie Z, Finkel MS. Norepinephrine-stimulated MAP kinase activity
enhances cytokine-induced NO production by rat cardiac myocytes. Am J Physiol
1999; 276(1 pt 2):H47H52.
Singh K, Balligand JL, Fischer TA, Smith TW, Kelly RA. Regulation of cytokineinducible nitric oxide synthase in cardiac myocytes and microvascular endothelial
cells. Role of extracellular signal-regulated kinases 1 and 2 (ERK1/ERK2) and
STAT1 alpha. J Biol Chem 1996; 271(2):11111117.
Chan ED, Winston BW, Uh ST, Wynes MW, Rose DM, Riches DW. Evaluation of
the role of mitogen-activated protein kinases in the expression of inducible nitric
oxide synthase by IFN-gamma and TNF-alpha in mouse macrophages. J Immunol
1999; 162:415422.
Da Silva J, Pierrat B, Mary JL, Lesslauer W. Blockade of p38 mitogen-activated
protein kinase pathways inhibits inducible nitric-oxide synthase expression in
mouse astrocytes. J Biol Chem 1997; 272:2837328380.
Bhat NR, Zhang P, Bhat AN. Cytokine induction of inducible nitric oxide synthase
in an oligodendrocyte cell line: role of p38 mitogen-activated protein kinase
activation. J Neurochem 1999; 72(2):472478.
266
130.
Chan ED, Riches DW. IFN-gamma + LPS induction of iNOS is modulated by

ERK, JNK/SAPK, and p38(mapk) in a mouse macrophage cell line. Am J Physiol
Cell Physiol 2001; 280(3):C441C450.
131. Chen Z, Yuhanna IS, Galcheva-Gargova Z, Karas RH, Mendelsohn ME, Shaul
PW. Estrogen receptor alpha mediates the nongenomic activation of endothelial
nitric oxide synthase by estrogen. J Clin Invest 1999; 103:401406.
132. Zheng J, Bird IM, Melsaether AN, Magness RR. Activation of the mitogenactivated protein kinase cascade is necessary but not sucient for basic broblast
growth factor- and epidermal growth factor-stimulated expression of endothelial
nitric oxide synthase in ovine fetoplacental artery endothelial cells. Endocrinology
1999; 140:13991407.
133. Dimmeler S, Fleming I, Fisslthaler B, Hermann C, Busse R, Zeiher AM.
Activation of nitric oxide synthase in endothelial cells by Akt-dependent
phosphorylation. Nature 1999; 399:601605.
134. Camandola S, Poli G, Mattson MP. The lipid peroxidation product 4-hydroxy-2,3nonenal increase AP-1-binding activity through caspase activation in neurons.
J Neurochem 2000; 74(1):159168.
135. Kumari MV, Yoneda T, Hiramatsu M. Eect of beta CATECHIN on the life
span of senescence accelerated mice (SAM-P8 strain). Biochem Mol Biol Int
1997; 41(5):10051011.
136. Mark RJ, Keller JN, Kruman I, Mattson MP. Basic FGF attenuates amyloid betapeptide-induced oxidative stress, mitochondrial dysfunction, and impairment of
Na+/K+-ATPase activity in hippocampal neurons. Brain Res 1997; 756(12):
205214.
137. Sichel G, Corsaro C, Scalia M, Di Bilio AJ, Bonomo RP. In vitro scavenger activity
of some avonoids and melanins against O2-(.). Free Radic Biol Med 1991;
11(1):18.
138. Bors W, Michel C, Schikora S. Interaction of avonoids with ascorbate and
determination of their univalent redox potentials: a pulse radiolysis study. Free
Radic Biol Med 1995; 19(1):4552.
139. Salah N, Miller NJ, Paganga G, Tijburg L, Bolwell GP, Rice-Evans C.
Polyphenolic avanols as scavengers of aqueous phase radicals and as chainbreaking antioxidants. Arch Biochem Biophys 1995; 322(2):339346.
of avonoids and phenolic acids. Free Radic Biol Med 1996; 20(7):933956.
141. Middleton E Jr, Kandaswami C. Eects of avonoids on immune and inammatory cell functions. Biochem Pharmacol 1992; 43(6):11671179.
142. Hertog MG, Feskens EJ, Hollman PC, Katan MB, Kromhout D. Dietary
antioxidant avonoids and risk of coronary heart disease: the Zutphen Elderly
Study. Lancet 1993; 342(8878):10071011.
143. Hirvonen T, Pietinen P, Virtanen M, Ovaskainen ML, Hakkinen S, Albanes D,
Virtamo J. Intake of avonols and avones and risk of coronary heart disease in
male smokers. Epidemiology 2001; 12(1):6267.
144. Edenharder R, von Petersdor I, Rauscher R. Antimutagenic eects of avonoids,
chalcones and structurally related compounds on the activity of 2-amino-3-
145.
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
157.
158.
159.
267
methylimidazo[4,5-f]quinoline (IQ) and other heterocyclic amine mutagens from

cooked food. Mutat Res 1993; 287(2):261274.
Yule DI, Kim ET, Williams JA. Tyrosine kinase inhibitors attenuate capacitative Ca2+ inux in rat pancreatic acinar cells. Biochem Biophys Res Commun
1994; 202(3):16971704.
Middleton E Jr, Kandaswami C, Theoharides TC. The eects of plant avonoids
on mammalian cells: implications for inammation, heart disease, and cancer.
Pharmacol Rev 2000; 52(4):673751.
Bastianetto S, Ramassamy C, Dore S, Christen Y, Poirier J, Quirion R. The Ginkgo
biloba extract (EGb 761) protects hippocampal neurons against cell death induced
by beta-amyloid. Eur J Neurosci 2000; 12(6):18821890.
Bastianetto S, Zheng WH, Quirion R. The Ginkgo biloba extract (EGb 761)
protects and rescues hippocampal cells against nitric oxide-induced toxicity:
involvement of its avonoid constituents and protein kinase C. J Neurochem 2000;
74(6):22682277.
Ishige K, Schubert D, Sagara Y. Flavonoids protect neuronal cells from
oxidative stress by three distinct mechanisms. Free Radic Biol Med 2001; 30(4):
433446.
Yamashita N, Kawanishi S. Distinct mechanisms of DNA damage in apoptosis
induced by quercetin and luteolin. Free Radic Res 2000; 33(5):623633.
Chan FL, Choi HL, Chen ZY, Chan PS, Huang Y. Induction of apoptosis in prostate
cancer cell lines by a avonoid, baicalin. Cancer Lett 2000; 160(2):219228.
Wenzel U, Kuntz S, Brendel MD, Daniel H. Dietary avone is a potent apoptosis
inducer in human colon carcinoma cells. Cancer Res 2000; 60(14):38233831.
Wesnes KA, Ward T, McGinty A, Petrini O. The memory enhancing eects of a
Ginkgo biloba/Panax ginseng combination in healthy middle-aged volunteers.
Psychopharmacologia 2000; 152(4):353361.
Kennedy DO, Scholey AB, Wesnes KA. The dose-dependent cognitive eects of
acute administration of Ginkgo biloba to healthy young volunteers. Psychopharmacologia 2000; 151(4):416423.
Cockle SM, Kimbe S, Hindmarch I. The eects of Gingko biloba extract (LI 1370)
supplementation on activities of daily living in free living older volunteers: a
questionnare survey. Hum Psychopharm Clin 2000; 15:227235.
Maurer K, Ihl R, Dierks T, Frolich L. Clinical ecacy of Gingko biloba special
extract Egb 761 in dementia of the Alzheimer type. Phytomedicine 1998; 5:
417424.
Oken BS, Storzbach DM, Kaye JA. The ecacy of Ginkgo biloba on cognitive
function in Alzheimer disease. Arch Neurol 1998; 55(11):14091415.
van Dongen MCJM, van Rossum E, Kessels AGH, Sielhorst HJG, Knipschild PG.
The ecacy of ginko for elderly people with dementia and age-associated memory
impairment: new results of a randomised clinical trial. J Am Geriatr Soc 2000;
48:11831194.
Watanabe CM, Wolram S, Ader P, Rimbach G, Packer L, Maguire JJ, Schultz
PG, Gohil K. The in vivo neuromodulatory eects of the herbal medicine Ginkgo
biloba. Proc Natl Acad Sci USA 2001; 98(12):65776580.
268
160.
Gramaglia A, Loi GF, Mongioj V, Baronzio GF. Increased survival in brain

metastatic patients treated with stereotactic radiotherapy, omega three fatty acids
and bioavonoids. Anticancer Res 1999; 19(6C):55835586.
Smith PF, Maclennan K, Darlington CL. The neuroprotective properties of the
Ginkgo biloba leaf: a review of the possible relationship to platelet-activating
factor (PAF). J Ethnopharmacol 1996; 50(3):131139.
de Rijk MC, Breteler MM, den Breeijen JH, Launer LJ, Grobbee DE, van der
Meche FG, Hofman A. Dietary antioxidants and Parkinson disease: The Rotterdam
Study. Arch Neurol 1997; 54(6):762765.
Joseph JA, Shukitt-Hale B, Denisova NA, Prior RL, Cao G, Martin A, Taglialatela
G, Bickford PC. Long-term dietary strawberry, spinach, or vitamin E
supplementation retards the onset of age-related neuronal signal-transduction
and cognitive behavioral decits. J Neurosci 1998; 18(19):80478055.
Logani S, Chen MC, Tran T, Le T, Raa RB. Actions of Ginkgo biloba related to
potential utility for the treatment of conditions involving cerebral hypoxia. Life Sci
2000; 67(12):13891396.
Joseph JA, Denisova N, Fisher D, Bickford P, Prior R, Cao G. Age-related
neurodegeneration and oxidative stress: putative nutritional intervention. Neurol
Clin 1998; 16(3):747755.
Inanami O, Watanabe Y, Syuto B, Nakano M, Tsuji M, Kuwabara M. Oral
administration of (-)catechin protects against ischemia-reperfusion-induced neuronal death in the gerbil. Free Radic Res 1998; 29:359365.
Shutenko Z, Henry Y, Pinard E, Seylaz J, Potier P, Berthet F, Girard P, Sercombe
R. Inuence of the antioxidant quercetin in vivo on the level of nitric oxide
determined by electron paramagnetic resonance in rat brain during global ischemia
and reperfusion. Biochem Pharmacol 1999; 57(2):199208.
Yamagishi M, Osakab N, Takizawa T, Osawa T. Cacao liquor polyphenols reduce
oxidative stress without maintaining alpha-tocopherol levels in rats fed a vitamin
E-decient diet. Lipids 2001; 36(1):6771.
Sun GY, Xia J, Draczynska-Lusiak B, Simonyi A, Sun AY. Grape polyphenols
protect neurodegenerative changes induced by chronic ethanol administration.
Neuroreport 1999; 10(1):9396.
La Grande L, Wang M, Watkins R, Ortiz D, Sanchez ME, Konst J, Lee C, Reyes
E. Protective eects of the avonoid mixture, sylimarin, on fetal rat brain and
liver. J Ethnopharmacol 1995; 65:5361.
Ray SD, Wong V, Rinkovsky A, Bagchi M, Raje RR, Bagchi D. Unique
organoprotective properties of a novel IH636 grape seed proanthocyanidin extract
on cadmium chloride-induced nephrotoxicity, dimethylnitrosamin (DMN)-induced
splenotoxicity and MOCAP-induced neurotoxicity in mice. Res Commun Mol
Pathol Pharmcol 2000; 107:105128.
Kruman I, Bruce-Keller AJ, Bredesen D, Waeg G, Mattson MP. Evidence that 4hydroxynonenal mediates oxidative stress-induced neuronal apoptosis. J Neurosci
1997; 17(13):50895100.
Komatsu M, Hiramatsu M. The eicacy of an antioxidant cocktail on lipid
peroxide level and superoxide dismutase activity in aged rat brain and
161.
162.
163.
164.
165.
166.
167.
168.
169.
170.
171.
172.
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
183.
184.
185.
186.
269
DNA damage in iron-induced epileptogenic foci. Toxicology 2000; 148(23):

143148.
Winter JC. The eects of an extract of Ginkgo biloba, EGb 761, on cognitive
behavior and longevity in the rat. Physiol Behav 1998; 63(3):425433.
Joseph JA, Shukitt-Hale B, Denisova NA, Bielinski D, Martin A, McEwen JJ,
cognitive, and motor behavioral decits with blueberry, spinach, or strawberry
dietary supplementation. J Neurosci 1999; 19(18):81148121.
Schroeter H, Williams RJ, Matin R, Iversen L, Rice-Evans CA. Phenolic
antioxidants attenuate neuronal cell death following uptake of oxidized lowdensity lipoprotein. Free Radic Biol Med 2000; 29(12):12221233.
Schroeter H, Spencer JPE, Rice-Evans C, Williams RJ. Flavonoids protect
neurones from oxidized low-density lipoprotein-induced apoptosis involving JNK,
c-jun and caspase-3. Biochem J 2001; 358:547557.
Spencer JP, Schroeter H, Crossthwaithe AJ, Kuhnle G, Williams RJ, Rice-Evans
C. Contrasting inuences of glucuronidation and O-methylation of epicatechin on
hydrogen peroxide-induced cell death in neurons and broblasts. Free Radic Biol
Med 2001; 31(9):11391146.
Wang H, Joseph JA. Structure-activity relationships of quercetin in antagonizing
hydrogen peroxide-induced calcium dysregulation in PC12 cells. Free Radic Biol
Med 1999; 27(56):683694.
Shin-Ya K, Kunigami T, Kim JS, Seto H. Protective eect of catechin against
beta-amyloid toxicity in hippocampal neurons and PC12 cells. J Neurochem 1997;
69:S42S52.
Yao Z, Drieu K, Papadopoulos V. The Ginkgo biloba extract EGb 761 rescues the
PC12 neuronal cells from beta-amyloid-induced cell death by inhibiting the
formation of beta-amyloid-derived diusible neurotoxic ligands. Brain Res 2001;
889(12):181190.
Dore S, Bastianetto S, Kar S, Quirion R. Protective and rescuing abilities of IGF-I
and some putative free radical scavengers against beta-amyloid-inducing toxicity
in neurons. Ann NY Acad Sci 1999; 890:356364.
Paladini AC, Marder M, Viola H, Wolfman C, Wasowski C, Medina JH.
Flavonoids and the central nervous system: from forgotten factors to potent
anxiolytic compounds. J Pharm Pharmacol 1999; 51(5):519526.
Dekermendjian K, Kahnberg P, Witt MR, Sterner O, Nielsen M, Liljefors T.
Structure-activity relationships and molecular modeling analysis of avonoids
binding to the benzodiazepine site of the rat brain GABA(A) receptor complex.
J Med Chem 1999; 42(21):43434350.
Medina JH, Viola H, Wolfman C, Marder M, Wasowski C, Calvo D, Paladini AC.
Overviewavonoids: a new family of benzodiazepine receptor ligands. Neurochem Res 1997; 22(4):419425.
Moro S, van Rhee AM, Sanders LH, Jacobson KA. Flavonoid derivatives as
adenosine receptor antagonists: a comparison of the hypothetical receptor binding
site based on a comparative molecular eld analysis model. J Med Chem 1998;
41(1):4652.
270
187.
Calapai G, Crupi A, Firenzuoli F, Costantino G, Inferrera G, Campo GM, Caputi

AP. Eects of Hypericum perforatum on levels of 5-hydroxytryptamine,
noradrenaline and dopamine in the cortex, diencephalon and brainstem of the
rat. J Pharm Pharmacol 1999; 51(6):723728.
Calapai G, Crupi A, Firenzuoli F, Inferrera G, Squadrito F, Parisi A, De Sarro G,
Caputi A. Serotonin, norepinephrine and dopamine involvement in the
antidepressant action of hypericum perforatum. Pharmacopsychiatry 2001;
34(2):4549.
Moon J, Tsushida T, Nakahara K, Terao J. Identication of quercetin 3-O-beta-Dglucuronide as an antioxidative metabolite in rat plasma after oral administration
of quercetin. Free Radic Biol Med 2001; 30(11):12741285.
Koga T, Meydani M. Eect of plasma metabolites of (+)-catechin and quercetin on
monocyte adhesion to human aortic endothelial cells. Am J Clin Nutr 2001;
73(5):941948.
Spencer JP, Schroeter H, Kuhnle G, Srai SK, Tyrrell RM, Hahn U, Rice-Evans C.
broblasts from oxidative-stress-induced cell death involving caspase-3 activation.
Biochem J 2001; 354(pt 3):493500.
Leslie EM, Mao Q, Oleschuk CJ, Deeley RG, Cole SP. Modulation of multidrug
resistance protein 1 (MRP1/ABCC1) transport and atpase activities by interaction
with dietary avonoids. Mol Pharmacol 2001; 59(5):11711180.
Rong Y, Yang EB, Zhang K, Mack P. Quercetin-induced apoptosis in the
monoblastoid cell line U937 in vitro and the regulation of heat shock proteins
expression. Anticancer Res 2000; 20(6B):43394345.
Raso GM, Meli R, Di Carlo G, Pacilio M, Di Carlo R. Inhibition of inducible nitric
oxide synthase and cyclooxygenase-2 expression by avonoids in macrophage
J774A.1. Life Sci 2001; 68(8):921931.
Saija A, Scalese M, Lanza M, Marzullo D, Bonina F, Castelli F. Flavonoids as
antioxidant agents: importance of their interaction with biomembranes. Free Radic
Biol Med 1995; 19(4):481486.
Morand C, Manach C, Crespy V, Remesy C. Quercetin 3-O-beta-glucoside is
better absorbed than other quercetin forms and is not present in rat plasma. Free
Radic Res 2000; 33(5):667676.
Cesquini M, Tenor AC, Torsoni MA, Stoppa GR, Pereira AL, Ogo SH. Quercetin
diminshes the binding of hemoglobin to the red blood cell membrane. J AntiAging
Med 2001; 4:5563.
Di Pietro A, Godinot C, Bouillant ML, Gautheron DC. Pig heart mitochondrial
ATPase: properties of puried and membrane-bound enzyme. Eects of
avonoids. Biochimie 1975; 57(8):959967.
Kobuchi H, Roy S, Sen CK, Nguyen HG, Packer L. Quercetin inhibits inducible
ICAM-1 expression in human endothelial cells through the JNK pathway. Am J
Physiol 1999; 277(3 pt 1):C403C411.
Kong AN, Yu R, Chen C, Mandlekar S, Primiano T. Signal transduction events
elicited by natural products: role of MAPK and caspase pathways in homeostatic
response and induction of apoptosis. Arch Pharm Res 2000; 23(1):116.
188.
189.
190.
191.
192.
193.
194.
195.
196.
197.
198.
199.
200.

201.
202.
203.
204.
205.
206.
207.
208.
209.
210.
211.
212.
213.
214.
271
Yoshizumi M, Tsuchiya K, Kirima K, Kyaw M, Suzaki Y, Tamaki T. Quercetin

inhibits Shc- and phosphatidylinositol 3-kinase-mediated c-Jun N-terminal kinase
activation by angiotensin II in cultured rat aortic smooth muscle cells. Mol
Pharmacol 2001; 60(4):656665.
Wadsworth TL, McDonald TL, Koop DR. Eects of Ginkgo biloba extract (EGb
761) and quercetin on lipopolysaccharide-induced signaling pathways involved
in the release of tumor necrosis factor-alpha. Biochem Pharmacol 2001; 62(7):
963974.
Lu LH, Lee SS, Huang HC. Epigallocatechin suppression of proliferation of
vascular smooth muscle cells: correlation with c-jun and JNK. Br J Pharmacol
1998; 124(6):12271237.
Green DR, Reed JC. Mitochondria and apoptosis. Science 1998; 281:13091312.
Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X.
Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex
initiates an apoptotic protease cascade. Cell 1997; 91(4):479489.
Kingham PJ, Pocock JM. Microglial apoptosis induced by chromogranin A is
mediated by mitochondrial depolarisation and the permeability transition but not
by cytochrome c release. J Neurochem 2000; 74(4):14521462.
Srinivasula SM, Hegde R, Saleh A, Datta P, Shiozaki E, Chai J, Lee RA, Robbins
PD, Fernandes-Alnemri T, Shi Y, Alnemri ES. A conserved XIAP-interaction
motif in caspase-9 and Smac/DIABLO regulates caspase activity and apoptosis.
Nature 2001; 410(6824):112116.
Goyal L. Cell death inhibition: keeping caspases in check. Cell 2001; 104(6):
805808.
Tafani M, Schneider TG, Pastorino JG, Farber JL. Cytochrome c-dependent
activation of caspase-3 by tumor necrosis factor requires induction of the
mitochondrial permeability transition. Am J Pathol 2000; 156(6):21112121.
Zamzami N, Marchetti P, Castedo M, Hirsch T, Susin SA, Masse B, Kroemer G.
Inhibitors of permeability transition interfere with the disruption of the
mitochondrial transmembrane potential during apoptosis. FEBS Lett 1996; 384(1):
5357.
Donahue RJ, Razmara M, Hoek JB, Knudsen TB. Direct inuence of the p53
tumor suppressor on mitochondrial biogenesis and function. FASEB J 2001;
15(3):635644.
Chelli B, Falleni A, Salvetti F, Gremigni V, Lucacchini A, Martini C.
Peripheral-type benzodiazepine receptor ligands: mitochondrial permeability
transition induction in rat cardiac tissue. Biochem Pharmacol 2001; 61(6):
695705.
Gale F, Sinnar S, Schwartz-Bloom RD. Diazepam promotes ATP recovery and
prevents cytochrome c release in hippocampal slices after in vitro ischemia.
J Neurochem 2000; 75(3):12421249.
Walker EH, Pacold ME, Perisic O, Stephens L, Hawkins PT, Wymann MP,
Williams RL. Structural determinants of phosphoinositide 3-kinase inhibition by
wortmannin, LY294002, quercetin, myricetin, and staurosporine. Mol Cell 2000;
6(4):909919.
272
215.
Conseil G, Baubichon-Cortay H, Dayan G, Jault JM, Barron D, Di Pietro A.

Flavonoids: a class of modulators with bifunctional interactions at vicinal ATPand steroid-binding sites on mouse P-glycoprotein. Proc Natl Acad Sci USA 1998;
95(17):98319836.
De Azevedo WF Jr, Mueller-Dieckmann HJ, Schulze-Gahmen U, Worland PJ,
Sausville E, Kim SH. Structural basis for specicity and potency of a avonoid
inhibitor of human CDK2, a cell cycle kinase. Proc Natl Acad Sci USA 1996;
93(7):27352740.
216.
10
Mitochondrial Actions of Flavonoids
and Isoflavonoids
Clinton S. Boyd and Enrique Cadenas
I.
INTRODUCTION
Flavonoids are a group of naturally occurring polyphenolic compounds,

ubiquitously distributed in plants (1). More than 4000 avonoids have been
identied and can be divided into the following subgroups of monomeric
structures: avones, avonols, avanols, avanones, and anthocyanidins.
Isoavonoids, including the isoavone genistein, are structural congeners
of avonoids. The high diversity of individual polyphenols within each of
these subgroups derives mainly from the number and conguration of
hydroxyl groups, and the nature and extent of alkylation. Numerous in vitro,
ex vivo, and in vivo studies have established that avonoids and isoavonoids exert a wide range of biological activities, including both cytoprotective and cytotoxic eects [18]. The biochemical targets that manifest the
eects of these polyphenolic compounds are distributed throughout the
subcellular compartments. This review focuses on one such compartment,
the mitochondrion.
Mitochondria play pivotal roles in both the life and the death of the
cell. First, mitochondrial integrity and the eciency of oxidative phosphorylation are essential for cellular bioenergetics and viability [9]. Second,
mitochondria appear to play a central role in apoptosis [10,11]. Specically,
mitochondria-dependent apoptotic pathways are responsive to intrinsic
proapoptotic stimuli such as oxidative stress. The latter is of particular
interest in that the oxyradical generation by the mitochondrial respiratory
273
274
Boyd and Cadenas
chain is the main contributor to the hydrogen peroxide (H2O2) steady-state

concentration in most cells [12]. Mitochondria can thus inuence the cellular
redox state, to which many redox-active biochemical species and pathways
are sensitive. Many of the biological eects of avonoids and isoavonoids
have been attributed to the antioxidant capacity of these compounds, as
determined in vitro, including their reducing properties (dened by their
redox potential) and their ability to scavenge radicals and chelate metals [1].
Thus, as a rst approach, it can be surmised that avonoids and isoavonoids
may modulate mitochondrial function through redox mechanisms. In fact,
early in vitro and ex vivo studies investigated structure-activity relationships
to correlate the redox potential of these compounds with their ability to
inhibit mitochondrial bioenergetics. These studies were extensively reviewed
in the rst volume of Flavonoids in Health and Disease [2] and are
summarized and extended in the current review.
However, evidence is accumulating from recent studies that avonoids
and isoavonoids may have more specic eects on biochemical pathways,
seemingly independent of their hydrogen-donating properties or of their ability
to alter the intracellular redox state. This new and exciting possibility is investigated in the current chapter with respect to mitochondrial function at the level
of (1) mitochondrial bioenergetics, (2) oxyradical generation and mitochondrial
antioxidant systems, and (3) mitochondria-dependent apoptosis. In particular,
the nding that mitogen-activated protein kinase (MAPK) signaling pathways
can be modulated by avonoids [7,13,14] has important implications for the
potential regulation of mitochondria-dependent apoptosis by such polyphenols.
The latter is considered with respect to the MAPK-dependent regulation of
antioxidant enzymes and the Bcl-2 family of proteins, which are known to
regulate cytochrome c release from mitochondria [11,15].
II.
THE EFFECT OF FLAVONOIDS AND ISOFLAVONES

ON MITOCHONDRIAL RESPIRATION AND
OXIDATIVE PHOSPHORYLATION
Mitochondria are described as the powerhouses of the cell, supplying the

adenosine triphosphate (ATP) demand for the multiple energy-dependent
cellular processes via oxidative phosphorylation. Thus the health of cells is
dependent on the functional state of these organelles. This is of particular
biological signicance considering the high energy demand of cancer cells and
may explain the cytotoxic or antineoplastic action of many avonoids [5,6]. For
example, potent inhibitors of mitochondrial respiration such as rotenone, an
isoavone derivative, and antimycin A result in cell death of tumors. For this
reason, mitochondrial respiration and oxidative phosphorylation represent a
Mitochondria, Flavonoids, and Isoflavonoids
275
likely starting point for an investigation into the eects of bioavonoids on

mitochondrial function.
A.
Inhibition of the Mitochondrial Electron Transport Chain
The mitochondrial electron transport chain (ETC) or respiratory chain comprises

a series of membrane-bound redox-active intermediates including avoproteins,
quinones, cytochromes, and iron-sulfur clusters [9] (Fig. 1). The latter facilitate
the thermodynamically controlled transfer of electrons from conjugate redox
pairs of low redox potential (E0V) [e.g., 320 mV for reduced nicotinamideadenine dinucleotide/oxidized nicotinamide-adenine dinucleotide (NADH/
NAD+)] to the final electron donor, O2 (with a high redox potential of +820
Figure 1 Mitochondrial bioenergetics, redox potentials, and the effect of flavonoids and isoflavonoids.
276
Boyd and Cadenas
mV for the 1/2O2/H2O pair). Iron-sulfur clusters generally have lower redox
potentials than cytochromes. Thus iron-sulfur clusters tend to be more concentrated at the electron donor (respiratory substrate) end of the ETC, whereas
cytochromes are more prominent at the electron acceptor (O2) end. Accordingly,
the cytochromes and iron-sulfur clusters are arranged into discrete complexes,
forming a sequential array of redox potentials (Fig. 1). These include NADHubiquinone oxidoreductase (complex I), succinate-ubiquinone oxidoreductase
(complex II), ubiquinol-cytochrome c oxidoreductase (complex III), and cytochrome c oxidase (complex IV). Approximate redox potentials for these complexes are 300, 220, +230, and +550 mV, respectively. Ubiquinone
(coenzyme Q; E0V = +0.04 mV) is an important electron carrier, common to
both complexes I and II and required to direct electron ow from the oxidation of
respiratory substrates to complexes III and IV. Another important shuttle is the
hemoprotein cytochrome c (E0V = +250 mV), located in the mitochondrial
intermembrane space. Cytochrome c receives electrons from complex III and is
then oxidized by cytochrome c oxidase, resulting in the ecient four-electron
reduction of O2 to H2O2, the terminal step in respiration [9].
As a whole, avonoids and isoavones cover almost the entire spectrum
of redox potentials inherent to the ETC [2,16,17]. Therefore, on electrochemical grounds, it is feasible that these polyphenolic compounds can interact
with mitochondrial redox centers and divert or impair electron ux through the
ETC (Fig. 1). There is also the potential for complex-specic interactions
depending on their redox potential. Extensive in vitro and ex vivo structureactivity studies were performed to determine the eect of a plethora of
avonoids and isoavones on mitochondrial electron transport and to correlate
potency with their redox potential [16,1821]. These and other studies were
extensively reviewed in the rst volume of this series [2]. The reader is referred
to the rst volume for more detailed information; only the key ndings are
summarized here. All the polyphenolic compounds tested were either inhibitory
or ineective toward complex I and complex II, dened as NADH-oxidase and
succino-oxidase activity, respectively [2]. None was found to exert a stimulatory
eect. From the onset, it should be noted that the multitude and complexity of the
avonoid and isoavone structures allow only the formulation of generalized
structure-activity rules.
1.
Complex I (Reduced Nicotinamide-Adenine

Dinucleotide-Oxidase)
NADH-oxidase activity was based on the O2 uptake in the presence of complex I

substrates (e.g., NADH) and is dependent on complexes I, III, and IV. Hydroxylation of the avonoid B-ring was an important determinant of inhibitory potency
toward NADH-oxidase [2]. The absence of B-ring hydroxyl substituents or their
277
methylation or glycosylation decreased inhibitory potency. The conjugation of

the C2,3 double bond (e.g., comparison of avonols and dihydroavonols) and the
presence of the 4-keto group were also very important structural features. A study
with a series of 3,5,7-trihydroxyavones revealed that inhibitory potency
correlated with the number and conguration of the B-ring hydroxyl groups
(i.e., their redox potential). In fact, the redox potentials of the latter 3,5,7trihydroxyavones falls within the range covered by NADH-ubiquinone
reductase (340 to +100 mV) [16,17]. Other studies involving a wider
group of avonoids conrmed the importance of B-ring hydroxyl group
conguration but found a weaker correlation with the number of hydroxyl
groups [2]. There were also strong indications that inhibitory potency is also
dependent on other structural features unrelated to redox potential. In summary, avonols were the most potent inhibitors of NADH-oxidase, and the
isoavone congener may also be more potent than the corresponding avonelike structure. The latter is not surprising considering that rotenone, an isoavone derivative, is a very potent and specic complex I inhibitor, routinely
used in mitochondrial research.
2.
Complex II (Succino-oxidase)
Succino-oxidase activity was based on the O2 uptake in the presence of complex

II substrates (e.g., succinate) and is dependent on complexes II, III, and IV.
Overall succino-oxidase activity was less sensitive to inhibition by avonoids
and isoavones than NADH-oxidase activity [2]. Only anthocyanidins showed a
greater potency toward succino-oxidase than NADH-oxidase. In the case of
succino-oxidase, there was a clearer relationship between both the number and
conguration of B-ring hydroxyl groups and inhibitory potency, as predicted by
studies with model phenolic groups [22,23]. For example, the order of potency
was as follows: chalcone > avone z avonol > dihydroavonol > anthocyanidin [2]. Catechins were inactive. Within the avones, a strong correlation
existed between the inhibitory potency and the number of hydroxyl groups, such
that 4 > 6 > 5 > 3 > 2 > 1 = 0. Indeed, the inhibitory potency of a series of 3,5,7trihydroxyavones toward succino-oxidase correlated with their E1/2 potentials,
which fall within the range of redox potentials inherent to succinate-ubiquinone
reductase (260 to +120 mV) [16,17]. Furthermore, the redox potentials of
avonoids determine their tendency to auto-oxidize, oxidize, with the subsequent generation of o-semiquinones and oxygen-centered radicals. In fact, autooxidizable avonoids have E1/2 values that are comparable to the one-electron
reduction potential range for superoxide anion (O2.) generation by quinone in
the NADH-ubiquinone reductase domain of the respiratory chain (70 to
+30 mV) [16]. Thus, some avonoids may damage complexes I and II through
a redox-cycling mechanism involving reactive quinoid and oxygen species.
278
Boyd and Cadenas
In summary, the free in vitro studies discussed indicate that avonoids do

show some specicity in inhibitory potency toward mitochondrial complex I or
complex II activity. However, complex I did appear to be more sensitive to a larger
number of avonoids, most likely because inhibitory potency is not only related to
redox potential. The chalcone butein was the most potent inhibitor of both
complexes I and II, with similar potency toward each complex. Polyphenolic
compounds such as quercetin, avone, and genistein were more potent against
complex I, whereas anthocyanidins were more potent against complex II. The
catechins were essentially inactive against both ETC complexes. Similarly, (+)epicatechin and (F)-catechin were found to have no eect on h-NADH oxidation
by rat liver submitochondrial particles, and taxifolin and (-)-epicatechin gallate
caused minor inhibition [24]. In contrast, pycnogenol, the French maritime pine
bark extract (PBE), competitively inhibited O2 consumption and h-NADH
oxidation by whole rat liver mitochondria and submitochondrial particles. The
decrease in electron transport activity was attributed to an inhibition of complexes I and II and a direct inhibition of complex III independent of an upstream
eect on complexes I and II. This indicates that natural plant extracts contain a
multitude of other polyphenolic phytochemicals and bioavonoids, besides the
catechins, which are likely to be potent inhibitors of all components of the
mitochondrial ETC.
B.
Inhibition of Mitochondrial Proton F0F1-Adenosine

Triphosphatase
The F0F1adenosine triphosphatase (ATPase) (or ATP synthase; complex V),

located in the mitochondrial inner membrane, catalyzes the synthesis of ATP
from adenosine diphosphate (ADP) (Fig. 1). Electron ux through the ETC
during respiration is associated with the vectorial transport of protons (H+) from
the matrix to the intermembrane space. Complexes I, III, and IV are the proton
pump, setting up the electrochemical gradient: a proton (pH) gradient and a
transmembrane potential (DCm). Proton reentry into the matrix through the F0
channel of complex V provides the energy to drive ATP synthesis by the F1
component. Mitochondrial ATP is exchanged for ADP in the cytoplasm by the
adenine nucleotide translocase (ANT), also located in the mitochondrial inner
membrane (Fig. 1).
Certain avonoids are capable of inhibiting a variety of ATPases, including
the oligomycin-sensitive mitochondrial F0F1-ATPase (complex V), Ca2+, Mg2+ATPase, and Na+, K+-ATPase (2,25). For example, the eect of quercetin on
ATPases and the implication for the regulation of glycolysis in tumor cells have
been reviewed [26]. Early ex vivo and in vitro structure-activity studies on
bovine F1-ATPase from whole mitochondria [18], submitochondrial particles
[27], or the puried enzyme [28] indicated that avonols (e.g., morin and
279
quercetin) were the most potent inhibitors of the avonoids tested. This
suggested the importance of the 3-hydroxyl group. In fact, very few avonoids
were found to have inhibitory aects on F1-ATPase. A more recent study
conrmed the inhibitory eect of avonoids on mitochondrial F1-ATPase from
rat brain and liver and extended the study to include other classes of polyphenolic
phytochemicals [25]. In particular, resveratrol and genistein were among the
most potent inhibitors, displaying noncompetitive kinetics. (+)-Catechin, (+)epicatechin, (-)-epicatechin, and (-)-epigallocatechin were ineective. Overall
mitochondrial F1-ATPase was relatively resistant to avonoids and isoavones in
comparison to complexes I and II. A reason for this may be that none of the
studies found a correlation between inhibitory potency and the number or
conguration of the B-ring hydroxyl groups.
The preceding ndings suggest that the redox potential is not a determining factor and that other mechanisms must be responsible for the observed
inhibition. One possibility is that avonoids having structural similarities with
the adenosine moiety may compete for the binding site of ADP/ATP of the F0F1ATPase, as has been reported for a variety of ATP-dependent enzymes and
steroid-binding sites [29]. This would be consistent with the ability of certain
avonoids to inhibit a number of dierent ATPase enzymes. Furthermore, the
ANT also possesses ADP-and ATP-binding domains [30]. Flavonoid hydroxyethylrutosides and procyanidolic oligomers were found to increase the respiratory control ratio (RCR) of isolated rat liver mitochondria by stimulating
ADP-dependent state 3 respiration [31]. The mechanism involved a promotion
of ADP uptake, which was sensitive to atractyloside, a selective inhibitor of
ANT activity. Hydroxyethylrutosides and procyanidolic oligomers are classied
as venotropic drugs, used in the treatment of chronic venous deciency or
arteriopathy. Thus, an increase in ANT activity could explain the ability of these
compounds to protect cultured human endothelial cells from the hypoxiainduced decrease in ATP content [31]. It is unclear how these compounds
would facilitate ADP uptake, but it may involve a switching of ANT to a
conformation that favors ADP transport over ATP transport.
C.
Respiratory Control, Coupling of Mitochondria, and the

Permeability Transition
As discussed, ATP synthesis is driven by the proton motive force of the

electrochemical gradient set up during the electron ow through the respiratory
chain. For this reason, the two processes of oxidation and phosphorylation are
described as being coupled (oxidative phosphorylation). In normally functioning,
tightly coupled mitochondria oxidation proceeds primarily when ATP is synthesized from ADP. The dependence of respiration on ADP levels is dened as
respiratory control, a key property of coupled mitochondria. The respiratory
280
Boyd and Cadenas
control ratio (RCR) is calculated as the ratio of the respiration rate in the presence
of ADP (i.e., state 3 respiration) and the rate of resting respiration in the absence
of ADP (i.e., state 4 respiration). Agents that uncouple oxidation from phosphorylation (known as uncouplers) remove respiratory control and thus allow state 4
respiration to proceed at maximal rates, but without the conservation of energy in
the form of ATP. Uncoupling is thus reected by a decrease in the RCR.
Dissipation of the electrochemical gradient (pH gradient and DCm) results
in rapid uncoupling. Classical uncouplers include protonophores, such as 2,4dinitrophenol and cyanide m-chlorophenylhydrazone, and the K+ ionophore
valinomycin. An in vitro model system investigated the ability of avonoids to
uncouple articial vesicles reconstituted with cytochrome oxidase [32]. Flavonoids uncoupled the vesicles by aecting both the transmembrane potential
dierence and the transmembrane pH dierence. Flavones were slightly more
eective than avanones, whereas the avonol quercetin exhibited hardly any
uncoupling activity. With respect to biological systems, it has been well
demonstrated that in plants avonoids act as uncouplers, possibly serving a role
as metabolic regulators [2]. Furthermore, the cytotoxicity of avonoids toward
cultured rat hepatocytes correlated strongly with their ability to dissipate the
mitochondrial transmembrane potential [6]. In isolated rat liver mitochondria,
quercetin, pinobanksin, pinocembrin, and derivatives were reported to stimulate
succinate-driven state 4 respiration, as would be consistent with an uncoupling
eect [33].
By denition, the uncoupling eect of certain avonoids should be
independent of their inhibitory eects on mitochondrial respiration or F0F1ATPase, suggesting an additional mode of action of avonoids against mitochondrial function. A collapse of the transmembrane potential is likely under
conditions in which the permeability barrier created by the mitochondrial inner
membrane is compromised (as occurs in the presence of ionophores). Calcium,
phosphate, oxidative stress, adenine nucleotide depletion, and membrane
depolarization can induce such a nonspecic increase in the permeability of
the inner membrane, in an event called the mitochondrial permeability transition
(MPT) [30,34]. The MPT can be selectively inhibited by cyclosporin A and is
believed to involve the assembly of a multiprotein complex to form a nonspecic pore that spans the inner and outer mitochondrial membranes. The latter
assembly is referred to as the permeability transition pore complex (PTPC) (Fig.
1). Its exact composition is unknown, but appears to comprise cyclophilin D,
ANT, the voltage-dependent anion channel (porin), and a benzodiazepinebinding site [10,30,34].
Quercetin and other avonoids were reported to prevent the MPT induced
by mefenamic acid or Ca2+ plus phosphate in isolated rat liver mitochondria
[33]. The avonoids were also found to prevent the oxidation of mitochondrial
protein sulfhydryl groups associated with the MPT. The authors proposed that
281
the protective eect of quercetin most likely involved the scavenging of free
radicals or an inhibition of respiration. As an alternative mechanism, it is interesting to note that avonoids can act as agonists or inverse agonists to the
benzodiazepine-binding site of the rat brain g-amino butyric acid A (GABA-A)
receptor complex [35]. Finally, the role of the MPT in mitochondria-dependent
apoptotic pathways and the impact of avonoid treatment are discussed in the
following section.
III.
OXYRADICAL PRODUCTION BY MITOCHONDRIA AND

THE ANTIOXIDANT EFFECTS OF FLAVONOIDS
Mitochondria are a main source of H2O2 production, contributing 4090% to the

cellular [H2O2]ss, depending on the tissue [12]. Superoxide anion (O2.), the
stoichiometric precursor of H2O2, is formed predominantly by ubisemiquinone
auto-oxidation during electron ux through the respiratory chain (Fig. 2).
Secondarily, O2. can be generated as a by-product of NADH-dehydrogenase
(complex I) activity. Once generated, O2. is vectorially released from the inner
membrane into the matrix, where it undergoes disproportionation to H2O2 by
Mn-superoxide dismutase (Mn-SOD). H2O2 that escapes the matrix glutathione
peroxidase activity can freely diuse from the mitochondria to the cytosol,
contributing to the cellular [H2O2]ss and redox state. Furthermore, monoamine
oxidase, located in the outer mitochondrial membrane, catalyzes the oxidative
deamination of monoamines, such as dopamine, with the concomitant production of H2O2 (Fig. 2). The latter route of H2O2 production could surpass that by
the inner mitochondrial membrane [12]. Thus O2. is continuously generated by
mitochondrial respiration and accounts for f2% of oxygen uptake by the
organelle under physiological conditions [12]. A further burst of O2. generation
can be observed under pathophysiological conditions, for example, on the
release of cytochrome c from the mitochondria in response to certain proapoptotic stimuli [36,37] (Fig. 2).
The antioxidant eects of avonoids in vitro have been extensively
described and are related to their ability to act as reducing agents (dened by
their redox potential), free radical scavengers, and metal chelators [1,3840].
Besides the intramitochondrial [H2O2]ss, mitochondria are rich in redox-active
species, including, inter alia, avoproteins, pyridine nucleotides, reduced
glutathione (GSH), and transition metals. It is anticipated that avonoids, by
virtue of their antioxidant properties, may inuence the intramitochondrial redox
state. Several in vitro studies have demonstrated that avonoids can protect
against lipid peroxidation induced in mitochondria by both nonenzymatic and
enzymatic sources [4143]. Nonenzymatic agents include the Fe2+-ascorbic acid
system and the alkyl radical generating system, whereas enzymatic mechanisms
282
Boyd and Cadenas
Figure 2 Oxyradical production by mitochondria and the role of antioxidant

enzymes: potential targets for flavonoids and isoflavonoids.
include complex I substrate (NADH-dependent) oxidation during mitochondrial

respiration. As an example of the former approach, an early in vitro investigation
with isolated rat brain mitochondria [41] conducted an extensive structureactivity study of the eects of avonoids on lipid peroxidation induced by the
Fe2+-ascorbic acid system. In general, although some avonoids actually promoted lipid peroxidation, the antiperoxidative action of the majority of the
compounds could be ranked according to those features known to confer the
most potent antioxidant properties to avonoids: (1) the number and conguration of hydroxylated substitutions on rings A and B, especially the 3V, 4Vdihydroxy catechol structure in ring B and a free 3-hydroxyl substitution; (2) the
planarity of molecule; (3) conjugation of the C2,3 double bond; and (4) the 4-keto
moiety in the C-ring [1, 17]. However, a more recent study found that
methylation of certain avonoids (e.g., quercetin, pinobanksin, and pinocembrin)
283
improved their ability to inhibit ADP/Fe2+-induced lipid peroxidation in isolated

rat liver mitochondria (33). This nding is consistent with the fact that increased
hydrophobicity would facilitate access of the avonoid into the membranes, the
site of lipid peroxidation [44]. Thus, the antioxidant potential of the avonoid
depends not only on its structure, but also on the type and source of the radical
and the localization of the oxidized molecular target (e.g., mitochondrial matrix
vs. membrane compartments).
The ability of certain avonoids to reduce lipid peroxidation in the
enzymatic, NADH-dependent model [42,43] most likely involved a dual eect:
the ability of the avonoids to scavenge radicals and a direct inhibition of
complex I activity. As discussed earlier, complex I activity is sensitive to most
avonoids in vitro; the inhibitory potency is only partially dependent on the
redox potential of the compounds. Furthermore, although methylation of a
avonoid such as quercetin promoted its antiperoxidative eect, it reduced the
inhibitory potency of the avonoids in relation to complex I and II respiration
[33]. Therefore, it is important to bear in mind that the protective vs. toxic eect
of a avonoid may be the net result of its ability to inhibit respiratory complexes,
scavenge radicals, and inuence other mitochondrial functions.
Flavonoids may also inuence antioxidant systems within the mitochondria,
which normally serve to protect mitochondria against the oxyradicals generated
during respiration. Such defenses could include glutathione peroxidase and two
SOD isoforms, which dier in their subcellular distribution. Cu, Zn-SOD is found
in the cytosol, whereas Mn-SO is abundant in the mitochondrial matrix. In 2001,
two studies altered the traditional view of SOD and mitochondrial function. First,
a Cu, Zn-SOD enzyme has been identied in the mitochondrial intermembrane
space [45]. The latter enzyme has properties similar to those of the cytosolic form
of the enzyme, and both appear to be encoded by the same gene. Second, evidence
has been presented that a percentage of mitochondrial O2. production is
vectorially released from the cytosolic side of the inner membrane space directly
into the intermembrane space [46]. Thus some O2. can escape the matrix MnSOD, potentially to be disproportionated to H2O2 by the intermembrane Cu, ZnSOD. A signicant point of this nding is that H2O2 produced in this manner
would escape glutathione peroxidase in the mitochondrial matrix.
The transcription of the Cu, Zn-SOD gene is regulated by the transcription
factor Elk1 (Ets domain protein-1) (47), whereas the promoter for Mn-SOD expression contains consensus sequences for Sp-1, Ap-1 (activator protein-1), and
CREB (c-AMP response element binding protein) [48,49]. The regulation of the
latter transcription factors has been linked to both the ERK and JNK groups of
MAPKs [50]. In turn, as discussed later, avonoids have been shown to inuence
MAPK (mitogen-activated protein kinase) signaling pathways, including ERK
and JNK (c-Jun amino-terminal kinase) [7,13,14], suggesting that avonoids may
modulate the expression of both cytosolic and mitochondrial SOD enzymes
284
Boyd and Cadenas
(Figs. 2 and 3). Some validity is given to the latter possibility by two in vivo
studies. First, the oral administration of a catechin-containing preparation to aged
rats increased SOD activity in the mitochondrial fraction of the striatum and
midbrain [51]. Second, the ability of (-)-catechin, orally administered to gerbils,
to protect dose-dependently against neuronal death induced by transient ischemia-reperfusion injury was associated with an increase in the superoxidescavenging ability of the brain tissue [52].
Figure 3 The role of MAPK signaling pathways in mitochondria-dependent

apoptosis: antioxidant enzymes, the Bcl-2 protein family, and the influence of
flavonoids and isoflavonoids. MAPK, mitogen-activated protein kinase.
IV.
285
APOPTOSIS AND FLAVONOIDS
Apoptosis, or programmed cell death, is a concerted, active process involving

the coordination of multiple signaling cascades [53]. Discrete signaling cascades
are activated in response to specic proapoptotic stimuli, although there may be
a degree of cross-talk between the apoptotic pathways. Key determinants of
apoptosis include the nature of proapoptotic stimulus, cell type, cellular redox
status, and preexisting activation state of the pro- versus antiapoptotic biochemical machinery.
The predominant eect of avonoid and isoavonoid supplementation in
ex vivo cell culture models appears to be one of promoting apoptosis [5457].
This is repeatedly observed in studies with transformed cancer cells, leading to
the descriptions cytoprotective and/or chemopreventive [6,58]. Two polyphenolic compounds that have been extensively studied in anticancer research
are quercetin and genistein, a avonoid and isoavone, respectively. However, ex
vivo studies with primary cultured cells in 2000 and 2001 showed that some
avonoids can prevent apoptosis promoted by agents that induce oxidative stress
[7,8,59]. The outcome of avonoid treatment is expected to show a complex
dependence on a number of factors, including the type of avonoid, its concentration, the type of cell (e.g., transformed versus nontransformed), the mechanisms of action of the avonoid, the nature of the proapoptotic stimulus, and the
specic apoptotic signaling pathway that is activated.
There is a growing recognition that mitochondria play a central role in
apoptosis [10,11]. Mitochondria-dependent apoptotic pathways are responsive to
intrinsic proapoptotic stimuli that cause perturbations in the intracellular
environment, particularly oxidative stress. In addition, other apoptotic signaling
pathways appear to converge onto the mitochondria. This raises the question of
whether avonoids may modulate events upstream or downstream of the
mitochondria-dependent apoptotic cascade, in addition to their potential eect
on mitochondrial respiration, oxidative phosphorylation, and oxyradical generation (as discussed earlier).
A.
Flavonoids and the Mitochondria-Dependent

Apoptotic Pathway
The hemoprotein cytochrome c, located in the intermembrane space, serves a

dual role in mitochondrial function [37]. First, it is essential for electron transfer
from complex III to cytochrome oxidase (complex IV) during mitochondrial
respiration. Second, cytochrome c is one of a number of apoptogenic or
apoptosis-inducing factors (AIFs) involved in mitochondria-dependent apoptosis
[10,11]. In response to certain proapoptotic stimuli, cytochrome c dissociates
286
Boyd and Cadenas
from the respiratory chain along the inner membrane and is released from the
intermembrane space into the cytosol (Fig. 4).
Cytochrome cdependent mitochondrial apoptosis involves a number of
concerted steps leading to the formation of the apoptosome and the downstream
activation of caspase-3 [10,11]. The predominant mechanism is believed to
involve the following events (Fig. 4): (1) a transient collapse of the mitochondrial transmembrane potential; (2) opening of the permeability transition pore
(PTP); (3) release of cytochrome c and procaspase-9; (4) recruitment of Apaf-1
(apoptosis protease activating factor-1) and dATP and assembly of the mitochondrial apoptosome; (5) activation of procaspase-9 by the apoptosome; and
(6) the caspase-9-catalyzed activation of caspase-3, the molecular executor of
apoptosis. Amplication through the caspase cascade leads to the downstream
Figure 4 Mitochondria-dependent apoptosis: Cytochrome-cdependent

caspase-3 activation and the role of the Bcl-2 protein family.
287
proteolysis of numerous death substrates [e.g., poly-(ADP-ribose) polymerase

(PARP)] and subsequent cellular damage. It should be noted, however, that
some dispute exists over the temporal relationship between the dissipation of the
mitochondrial transmembrane potential and the induction of the permeability
transition. Furthermore, it is unclear whether these two events are actually
necessary for the release of cytochrome c (and other apoptogenic factors), and
thus mitochondria-mediated apoptosis [37].
The Bcl-2 family of oncoproteins is known to play an important role
in apoptosis through their ability to regulate cytochrome c release from
mitochondria [11,15]. The antiapoptotic proteins Bcl-2 and Bcl-xL prevent
cytochrome c release, whereas the proapoptotic family members (e.g., Bad,
Bid, Bik, Bax) facilitate cytochrome c eux or block the protective eects of
Bcl-2 and Bcl-xL. The mechanism involved is unclear; the Bcl-2 family
proteins may interact directly with the MTP (mitochondrial permeability
transition) protein complex (the PTPC) or form independent ionic pores in
the outer mitochondrial membrane (Fig. 3). Nonetheless, cytochrome cdependent caspase-3 activation and changes in the expression or phosphorylation state
of Bcl-2 family proteins are taken as indicative of mitochondria-dependent
apoptotic pathways. It is important to remember that other apoptogenic proteins
are also present in the mitochondrial intermembrane space, including smac/
DIABLO and avoprotein (AIF) [10,11,60]. The release stimuli for the latter
factors, which are currently being elucidated, may also involve the permeability
transition or the Bcl-2 family proteins [37].
In 2001 a hypothesis was proposed that the redox state of cytochrome c,
after release into the cytoplasm, acts a fail-safe mechanism in the regulation of
apoptosis [61]. It is argued that cytochrome c induces apoptosis only if it is
present in the cytoplasm in the oxidized state. It is predicted that the redox state of
this hemoprotein will be sensitive to the cellular redox state such that cytosolic
GSH levels will maintain the protein in a reduced state. An increase in prooxidants (e.g., reactive oxygen species) or a decrease in GSH will switch
cytochrome c to the oxidized state, with the induction of apoptosis. Cytochrome
c released during apoptosis results in an impairment of the electron ux through
the respiratory chain with a subsequent burst of O2. generation by the
mitochondria [36,37] (Fig. 3). It could be proposed that this burst facilitates
the oxidation of cytochrome c. In this regard, the nding that avonoids such as
taxifolin, catechin, epicatechin, and, in particular, epicatechin gallate can directly
reduce cytochrome c in vitro is of interest [24]. Therefore, avonoids may protect
cells by promoting the reduction of cytochrome c directly or by preventing its
oxidation through modication of the cellular redox state. Specically, the latter
may involve scavenging the apoptotic burst of reactive oxygen species
produced by the mitochondria. In addition, it will be shown that cell culture
288
Boyd and Cadenas
studies are pointing to a role for avonoids and isoavones upstream of

cytochrome c release, i.e., prior to its oxidation in the cytosol.
The natural isoavone genistein is a potent inhibitor of tyrosine kinase and
induces cell cycle arrest and apoptosis in a number of tumor cell lines. An
early (1993) study found that the induction of G2/M cell cycle arrest by genistein
in MC-7 (human breast), Jurkat (human T-cell leukemia), and L-929 (mouse
transformed broblasts) tumor cells was consistently associated with an increase
in mitochondrial number and/or activity [54]. A more recent (2000) study with
the MRC-5 cell line (human lung broblasts) reported a similar nding [56].
Treatment with genistein or oxidative stress, induced by H2O2 or buthionine
sulfoximine (an inhibitor of intracellular GSH synthesis), resulted in cell cycle
arrest and an increase in mitochondrial mass and deoxyribonucleic acid (DNA)
content. Furthermore, studies have provided strong evidence that genisteininduced apoptosis is mediated by the mitochondria-dependent apoptotic pathway.
For example, in RPE-J cells, genistein treatment caused a reduction in the
mitochondrial transmembrane potential, an opening of the permeability transition
pore in a caspase-independent manner, and the release of cytochrome c [57].
Cytochrome c release was associated with the downstream activation of caspase-3
and subsequent nuclear condensation and DNA fragmentation. Interestingly, the
ANT inhibitor bongkrekic acid, a blocker of the MPT, prevented the apoptotic
eects of genistein described. The overexpression of Bcl-xL protein was found to
be responsible for the resistance of human prostate cancer PC-3 cells to
staurosporine-induced apoptosis [62]. Although the pro apoptotic Bax and BAD
proteins were successfully incorporated into the mitochondrial membranes, there
was no activation of mitochondria-dependent apoptosis. Treatment with genistein
restored the sensitivity to staurosporine by downregulating Bcl-xL protein
expression, thereby promoting the MPT [62]. Similarly, the avonoid quercetin
was found to restore sensitivity to CD95-(Apo1/Fas)-mediated apoptosis in HPBALL cells, which typically show a resistance to the latter proapoptotic stimulus
[55]. Quercetin treatment promoted the dissipation of the mitochondrial transmembrane potential, caspase-3 activation, and DNA fragmentation. Importantly,
quercetin induced apoptosis even though treatment was associated with a
reduction in the level of intracellular reactive oxygen species [55].
The activation of mitochondria-mediated apoptosis in transformed cells
appears to be a common feature of avonoids, too. For example, in a structureactivity study with human leukemia HL-60 cells [58], the ability of low micromolar concentrations of avonoids to induce mitochondria-mediated apoptosis
was as follows: apigenin > quercetin > myricetin > kaempferol. For all the
avonoids tested, the key biochemical events included elevated level of reactive
oxygen species, dissipation of the mitochondrial transmembrane potential, release
of cytochrome c, procaspase-9 processing, caspase-3 activation, and proteolytic
cleavage of PARP. Another study on the same leukemia cell line found that
289
theaavins, (-)-epigallocatechin-3-gallate, and penta-O-galloyl-h-d-glucose

[which is structurally related to (-)-epigallocatechin-3-gallate] could induce
apoptosis in an identical manner to that described [63]. Similarly, green tea
catechins, including (-)-epigallocatechin-3-gallate, (-)-epigallocatechin, and (-)epicatechin-3-gallate, were reported to induce mitochondria-dependent apoptosis
in human prostate cancer DU145 cells [64]. Observed events included an increase
in reactive oxygen species production and mitochondrial depolarization, but
treatment did not alter the expression of Bcl-2, Bcl-xL, or Bad. The order of
potency was as follows: (-)-epicatechin-3-gallate > (-)-epigallocatechin-3gallate > (-)-epigallocatechin > (-)-epicatechin, which was inactive (64). Finally,
avone was a potent inducer of apoptosis in HT-29 human colon carcinoma cells
without an eect on nontransformed murine colonocytes [65]. The latter eect
of avone was associated with a change in the messenger ribonucleic acid
(mRNA) levels of cell cycle-and apoptosis-related genes, including an upregulation of Bak mRNA without an eect on BAX, and a downregulation of NF-nB
and Bcl-xL mRNA levels. To reiterate, the proapoptotic eects of avonoids and
isoavones against cancer cells in ex vivo models may explain their cancer
chemopreventive activity [6,58]. One shortcoming of the studies discussed is
that, in nearly all cases, there was a lack of comparison of the eects of the
avonoids/isoavones in transformed cells versus the eects of their nontransformed counterparts.
Recent studies with primary cultures, particularly neurons, are nding a
protective eect for a variety of polyphenolic compounds, including avonoids,
against a number of proapoptotic agents. For example, the death of cultured rat
hippocampal cells, induced by nitric oxiderelated species, could be prevented by
pretreatment with the stilbene resveratrol or the avonoids quercetin and (+)catechin [4]. Similarly, the avonoid content of Ginkgo biloba extract (EGb 761)
was found to protect cultured rat hippocampal cells from the proapoptotic eects
of nitric oxiderelated species and h-amyloid-derived peptides [66,67]. Of
importance is the observation that the antiapoptotic eects of the avonoids,
but not resveratrol, appeared to be related to their ability to block the activation of
protein kinase C, rather than just their antioxidant potential [4,66]. Many
avonoids were found to protect against glutamate-induced excitotoxicity in
cultured rat primary neurons or the mouse hippocampal cell line HT-22, a model
system of oxidative stress [3]. Three distinct mechanisms of protection were observed: an increase in the intracellular GSH levels, a reduction of reactive oxygen
species, and the prevention of the NMDA (n-methyl-D-aspartate) receptormediated Ca2+ inux. Interestingly, the neuroprotective avonoids appeared to
operate specically by one of the possible mechanisms [3]. Of importance here is
the fact that mitochondria-mediated pathways are central to neuronal apoptosis,
especially with respect to the hippocampal glutamate excitotoxicity model or
oxidative stress.
290
Boyd and Cadenas
For example, the induction of apoptosis in cultured primary striatal

neurons by oxidized low-density lipoprotein (Ox-LDL) was prevented by
pretreatment with low micromolar concentrations of epicatechin and kaempferol
[7]. Although the avonoid pretreatments did not prevent an increase in
intracellular oxidative stress, there was a potent inhibition of MAPK signaling
and caspase-3 activation. Similarly, epicatechin protected against H2O2-induced
apoptosis in cultured human broblasts [8]. Epicatechin reduced caspase-3-like
protease activity and protected mitochondrial function from the damage induced
by the oxidative stress. Importantly, in both studies 3V-O-methyl-epicatechin, the
in vivo metabolite of epicatechin, displayed the same potency as epicatechin in
attenuating oxidative stressmediated cell death [7,8]. This suggests that the
antiapoptotic eects of these avonoids are not simply related to their hydrogendonating properties.
An ex vivo study with rat hepatocytes found a strong correlation between
the cytotoxicity of a avonoid and its ability to induce an early collapse of the
mitochondrial transmembrane potential [6]. In contrast, nontoxic avonoids had
no eect on the transmembrane potential. As discussed previously, the ability of
avonoids and isoavones to inhibit mitochondrial respiration and adenosine
triphosphate (ATP) synthesis or uncouple oxidative phosphorylation represents a
potential mechanism whereby they can trigger dissipation of the transmembrane
potential and thus cytochrome cdependent apoptosis. Thus, it is interesting to
note that epicatechin, which had little or no proapoptotic eect in cancer cells
[64] and exerted antiapoptotic eects in primary culture models [7,8,59], had
essentially no inhibitory eect against complex I, II, or III activity and F0F1ATPase activity [24,25]. However, the preceding discussion has also indicated
that the pro- or antiapoptotic eects of avonoids may be explained by more
specic mechanisms, potentially independently of their antioxidant capacity.
These include eects on protein kinase cascades and gene expression, including
MAPK and the Bcl-2 family of proteins.
B.
Mitogen-Activated Protein Kinases, Mitochondria,

and Flavonoids
Mitogen-activated protein kinases (MAPKs) are integral components of cellular

phosphorylation/dephosphorylation signalling cascades [50,68]. In response to
extracellular and intracellular signals, the activation of MAPKs is initiated by
upstream eector molecules and specic protein kinase cascades (e.g., tyrosine
kinases, MAPK kinase kinases, and MAPK kinases) and deactivated by MAPK
phosphatases. Mammalian MAPKs can be divided into at least 3 groups, with
identiable isoenzymes: (1) extracellular signal-related kinases (ERK1/2); (2) cJun N-terminal kinases (JNK1/2/3); (3) p38 kinases (p38a/h/g/y) [50,68] (Fig. 4).
Distinct signal transduction pathways activate the latter groups, although some
291
overlap may occur, depending on the stimuli. The activation of ERK1/2 by growth
factors involves a well-dened pathway, including the G-protein eector molecule c-Ras (p21ras), Raf-1, and MEK1/2. Both JNK and p38 respond to proinammatory cytokines and environmental stress, but their activation may occur via
common or parallel pathways. The activation of JNK and p38 by a variety of
environmental stressors, including ultraviolet (UV) irradiation, osmotic shock,
oxidative stress, proinammatory cytokines, and trophic factor withdrawal, has
led to their alternative designation as stress-activated protein kinases (SAPKs).
Likely eectors for SAPKs include the G-proteins Rac1 and Cdc42, and in some
cases, c-Ras. Downstream targets of MAPK include numerous cytosolic proteins,
nuclear proteins, and transcription factors. Thus, ultimately, the MAPKs can modulate gene and protein expression involved in cell dierentiation, proliferation,
survival, and death [69]. For example, it has already been discussed that both JNK
and ERK can regulate the gene expression of the antioxidant enzymes Cu, ZnSOD, and Mn-SOD [4749] (Fig. 4). Importantly, the JNKs are the only kinases
known to phosphorylate c-Jun in vivo, thereby activating this transcription factor,
which is a component of the activator protein 1 (AP-1) transcription factor complex. Typically ERK1/2 activation is associated with cell proliferation and survival,
whereas the SAPKs are predominantly associated with apoptotic events [70,71].
Activated MAPKs have been found to modulate mitochondria-mediated
apoptotic pathways, JNK in particular [71]. In fact, it appears that MAPK
signaling may represent a potential cross-link between the dierent apoptotic
pathways. JNK activation is not required for death-receptor-mediated apoptosis
but is required for caspase-9 activation by the mitochondrial pathway, induced by
a variety of proapoptotic stimuli [72,73]. JNK activation and c-Jun phosphorylation were found to be necessary to promote cytochrome c release from
mitochondria, with the sequential assembly of the apoptosome and caspase-3
activation. The molecular mechanism of this eect was unclear, but it appeared to
involve regulation of the expression and phosphorylation state of the Bcl-2
protein family and their recruitment to the mitochondrial outer membrane (Fig.
4). In a number of apoptotic models, JNK activation was associated with a
downregulation of the antiapoptotic Bcl-2 and Bcl-xL and upregulation of the
proapoptotic Bax and Bad [7477]. Two cell culture studies provided very strong
evidence that JNK activation resulted in the phosphorylation of Bcl-2 and Bcl-xL
ex vivo, with the induction of apoptosis [76,77]. In other words, Bcl-2 and Bcl-xL
appear to be substrates of JNK, and phosphorylation results in their inactivation,
thereby abolishing their ability to prevent cytochrome c release.
JNK can also stabilize the proapoptotic p53 through phosphorylation [78]
(Fig. 4). Once phosphorylated, p53 can impair mitochondria function and
induce the permeability transition through its mediator p53-regulated apoptosis-inducing protein 1 (p53AIP1), a mitochondrial protein [79]. Furthermore,
the expression of the proapoptotic Bax is regulated by p53 [80]. Alternatively,
292
Boyd and Cadenas
p53 might activate JNK via the MAPK kinase MEKK4, involving the p53mediated expression of the growth arrest and DNA damageinduced protein
(GADD45) [81]. Also, caspase-3 activation may, under certain conditions,
activate JNK downstream via MAPK kinase MEKK1 [82]. Finally the burst
of O2. generation by mitochondria associated with cytochrome c release
[36,37] could result in the activation of JNK, which is known to be sensitive
to H2O2 [70,71]. Activated JNK could, in turn, promote further cytochrome c
release (as discussed), thereby amplifying mitochondria-dependent apoptosis.
These observations strengthen the point that apoptotic pathways can be
interdependent, and JNK activation potentially play a central proapoptotic role.
In contrast, several lines of evidence suggest that the activation of the RasRaf-ERK1/2 pathway may oppose JNK-mediated eects with respect to mitochondria-dependent apoptosis (Fig. 4). ERK1/2 can phosphorylate and activate
mitogen-and stress-activated kinase 1 (MSK1) and pp90 ribosomal S6 kinase
(RSK) [83,84]. RSK can phosphorylate BAD, thereby inhibiting its proapoptotic eects [83]. Furthermore, both RSK and MSK1 are potent activators of
cyclic adenosine monophosphate (cAMP) element binding protein (CREB), a
transcription factor for Blc-2, and therefore important for cell survival [84]. The
upregulation or overexpression of Bcl-2 has been reported to be associated with
the inactivation of JNK [75,85]. Thus, ERK1/2 activation could ultimately lead to
JNK inactivation. Finally, the activation of c-Ras has been shown to result in a
suppression of Bax expression, via a mechanism that involved activation of both
the ERK1/2 signaling cascade and the phosphatidylinositol-3V-kinase (PI-3)/Akt
pathway [80].
Oxidative stress is a major stimulus for MAPK signaling cascades, as evidenced by the number of redox-responsive transcription factors and antioxidant
genes that are downstream targets of these enzymes [69]. The G-protein c-Ras is a
common signaling target for reactive oxygen species and nitric oxiderelated
species with important consequences for ERK1/2 signaling [8689]. JNK is also
activated by a number of agents that alter the redox state, including H2O2 [71,87],
peroxynitrite [90], and nitric oxiderelated species [8688]. Similarly to ERK1/2,
eector molecules upstream of JNK (e.g., c-Ras, Rac1 or Cdc42) appear to be
redox-sensitive. In addition, JNK can be directly aected by oxidative stress, an
eect that may be related to the redox-sensitive cysteine present in JNK, but not
ERK or p38 [91,92]. Furthermore, redox-sensitive proteins such as glutathione-Stransferase and thioredoxin can directly interact with JNK or apoptosis-signal
regulating kinase (ASK1), an upstream activator of JNK [9395]. Therefore,
certain avonoids may inuence MAPK pathways through an eect on the
cellular redox potential, depending on their intrinsic antioxidant capacity.
In fact, strong evidence is accumulating from dierent models that
avonoids can inuence MAPK signaling pathways, and this inuence may
represent a specic mechanism whereby avonoids can modulate apoptosis. At
293
low concentrations, epicatechin gallate and epigallocatechin gallate activated

ERK2, JNK1, and p38 in mammalian cell lines [14]. MAPK activation was
associated with an increase in the gene expression of c-Fos, c-Jun, and antioxidant
and detoxifying enzymes, such as glutathione-S-transferase. However, at higher
concentrations, these avonoids activated the caspase pathway and apoptosis,
that activation may reect a predominance of the JNK1 pathway over ERK. In
contrast, quercetin (in low micromolar concentrations) inhibited the activation
of the JNK pathway by proapoptotic stimuli, resulting in a downregulation of
AP-1 and ICAM-1 (intracellular adhesion molecule-1) expression in human
endothelial cells [13]. Similarly, a study with cultured primary striatal neurons
[7] found that pretreatment with low micromolar concentrations of epicatechin
or 3-O-methylepicatechin strongly inhibited the Ca2+-dependent activation of
ERK1/2 and JNK by oxidized low-density lipoprotein (oxLDL). Inhibition of
JNK activation by these compounds (and another avonoid, kaempferol)
abolished c-Jun phosphorylation. Inhibition of JNK activation, but not ERK1/
2 activation, was associated with cell survival and the neuroprotective eects of
the avonoids. Importantly, the avonoids also reduced procaspase-3 cleavage
(i.e., caspase-3 activation) and caspase-3-like protease activity induced by
oxLDL [7]. The latter is of particular relevance as the mitochondria-dependent
pathways typically mediate oxidative stressinduced apoptosis. Thus, the results
of this study strongly suggest a relationship between JNK and mitochondria that
can be modied by avonoid treatment.
In summary, the eect of a avonoid on MAPK signaling is likely to
depend on the concentration and type of avonoid, the preexisting degree of
activation of the main MAPK pathways, and the cellular environment, such as the
redox state to which MAPKs are sensitive. It is important to note that epicatechin
and its in vivo metabolite 3-O-methylepicatechin were equally potent in their
ability to prevent JNK activation and apoptosis in response to an oxidative stress
[7]. This nding suggests that the latter eect of these avonoids is not related to
their hydrogen-donating abilities. Flavonoids may inuence MAPK signaling
pathways by a direct inhibitory or stimulatory eect on the enzyme itself or by
modulation of signal transduction events upstream of the relevant MAPK (Fig.
4). Ultimately, an eect on MAPK signaling could represent a specic mechanism by which avonoids and isoavones inuence mitochondrial function.
V.
CONCLUSIONS
This review has demonstrated that avonoids can inuence mitochondrial

function, either at the level of respiration and oxidative phosphorylation or of
the mitochondrial apoptosome. These biochemical eects were elucidated in
vitro with isolated mitochondria or cell culture systems. However, the biological
294
Boyd and Cadenas
signicance of these eects to mitochondria in vivo needs to be claried and

depends on the bioavailability of avonoids [96].
After absorption, dietary avonoids are metabolized in the small intestine or
through xenobiotic metabolic systems in the liver. Main metabolic routes include
methylation, glucuronidation, and sulfation [97100]. In turn, tissue distribution
(including the ability to cross the blood-brain barrier), subcellular localization,
and access to mitochondria depend on the extent and type of metabolism. The in
vitro structure-activity studies discussed clearly showed that metabolism of the
hydroxyl groups (e.g., glycosylation, glucuronidation, or methylation) reduced the
antioxidant capacity of avonoids, their inhibitory potency toward complex II and
F0F1-ATPase activity, and their proapoptotic potential. Alternatively, methylation
may serve to enhance avonoid stability, membrane penetration, and access to
subcellular compartments (e.g., mitochondria). In other words, the in vivo
biochemical properties of avonoids cannot necessarily be predicted from their
in vitro properties. Future studies should (1) identify the biologically relevant
metabolic derivatives of dietary avonoids; (2) compare the biochemical eects
of these derivatives to those of their aglycones in vitro to predict likely
mitochondrial eects; and (3) determine the in vivo mitochondrial eects of
those dietary avonoids, which are known to be taken up by cells and thus are
likely to have access to mitochondria. Finally, an eort should be made to identify
specic and selective molecular targets of avonoids with respect to mitochondria
function, e.g., MAPK signaling, and to rationalize their importance as potential
therapeutic agents.
REFERENCES
1. Rice-Evans CA, Miller NJ, Paganga G. Structure-antioxidant activity relationship
of avonoids and phenolic acids. Free Radic Biol Med 1996; 20:933956.
2. Hodnick WF, Pardini RS. Inhibition of mitochondrial function by avonoids. In:
Rice-Evans CA, Packer L, eds. Flavonoids in Health and Disease. Vol. 1. New
3. Ishige K, Schubert D, Sagara Y. Flavonoids protect neuronal cells from oxidative
stress by three distinct mechanisms. Free Radic Biol Med 2001; 30:433446.
4. Bastianetto S, Zheng WH, Quirion R. Neuroprotective abilities of resveratrol and
other red wine constituents against nitric oxide-related toxicity in cultured hippocampal neurons. Br J Pharmacol 2000; 131:711720.
5. Colic M, Pavelic K. Molecular mechanisms of anticancer activity of natural
dietetic products. J Mol Med 2000; 78:333336.
6. Galati G, Teng S, Moridani MY, Chan TS, OBrien PJ. Cancer chemoprevention
and apoptosis mechanisms induced by dietary polyphenolics. Drug Metab Drug
Interact 2000; 17:311349.

7.
295
Schroeter H, Spencer JPE, Rice-Evans C, Williams RJ. Flavonoids protect

neurones from oxidized low-density lipoprotein-induced apoptosis involving JNK,
c-jun and caspase-3. Biochem J 2001; 358:547557.
8. Spencer JPE, Schroeter H, Kuhnle G, Srai SK, Tyrrell RM, Hahn U, Rice-Evans C.
Biochem J 2001; 354:493500.
9. Hate Y. The mitochondrial electron transport and oxidative phosphorylation
system. Annu Rev Biochem 1985; 54:10151069.
10. Green DR, Reed JC. Mitochondria and apoptosis. Science 1998; 281:13091312.
11. Adrain C, Martin SJ. The mitochondrial apoptosome: a killer unleashed by the
cytochrome seas. Trends Biochem Sci 2001; 26:390397.
12. Boveris A, Cadenas E. Mitochondrial production of hydrogen peroxide regulation
by nitric oxide and the role of ubisemiquinone. IUBMB Life 2000; 50:245250.
13. Kobuchi H, Roy S, Sen CK, Nguyen HG, Packer L. Quercetin inhibits inducible
Physiol Cell Physiol 1999; 277:C403C411.
14. Kong A-NT, Yu R, Chen C, Mandlekar S, Primiano T. Signal transduction events
response and induction of apoptosis. Arch Pharm Res 2000; 23:116.
15. Kluck RM, Bossy-Wetzel E, Green DR, Newmeyer DD. The release of
cytochrome c from mitochondria: a primary site for Bcl-2 regulation of apoptosis.
Science 1997; 275:11321136.
16. Hodnick WF, Milosavljevic EB, Nelson JH, Pardini RS. Electrochemistry of
1988; 37:26072611.
17. Van Acker SABE, van den Berg D-J, Tromp MNJL, Grioen DH, van Bennekom
WP, van der Vijgh WJF, Bast A. Structural aspects of antioxidant activity of
avonoids. Free Radic Biol Med 1996; 20:331342.
18. Bohmont C, Aaronson LM, Mann K, Pardini RS. Inhibition of mitochondrial
NADH oxidase, succinoxidase, and ATPase by naturally occurring avonoids.
J Nat Prod 1987; 50:427433.
19. Hodnick WF, Kung FS, Roettger WJ, Bohmont CW, Pardini RS. Inhibition of
mitochondrial respiration and production of toxic oxygen radicals by avonoids: a
structure-activity study. Biochem Pharmacol 1986; 35:23452357.
20. Hodnick WF, Bohmont CW, Capps C, Pardini RS. Inhibition of mitochondrial
NADH-oxidase (NADH-Coenzyme Q oxidoreductase) enzyme system by avonoids: a structure-activity study. Biochem Pharmacol 1987; 36:28732874.
21. Hodnick WF, Duval DL, Pardini RS. Inhibition of mitochondrial respiration and
cyanide-stimulated generation of reactive oxygen species by selected avonoids.
Biochem Pharmacol 1994; 47:573580.
22. Cheng SC, Pardini RS. Structure-inhibition relationships of various phenolic
compounds towards mitochondrial respiration. Pharmacol Res Commun 1978;
10:897910.
296
Boyd and Cadenas
23.
Cheng SC, Pardini RS. Inhibition of mitochondrial respiration by model phenolic

compounds. Biochem Pharmacol 1979; 28:16611667.
Moini H, Arroyo A, Vaya J, Packer L. Bioavonoid eects on the mitochondrial
respiratory electron transport chain and cytochrome c redox state. Redox Rep
1999; 4:3541.
Zheng J, Ramirez VD. Inhibition of mitochondrial proton F0F1-ATPase/ATP
synthase by polyphenolic phytochemicals. Br J Pharmacol 2000; 130:11151123.
Racker E. Eect of quercetin on ATP-driven pumps and glycolysis. Prog Clin Bio
Res 1986; 213:257271.
Lang DR, Racker E. Eects of quercetin and F1 inhibitor on mitochondrial
ATPase and energy-linked reactions in submitochondrial particles. Biochim
Biophys Acta 1974; 333:180186.
ATPase: properties of puried and membrane-bound enzyme: eects of
avonoids. Biochimie 1975; 57:959967.
Conseil G, Baubichon-Cortay H, Dayan G, Jault JM, Barron D, Di Pietro A.
Flavonoids: a class of modulators with bifunctional interactions at vicinal ATP-and
steroid-binding sites on mouse P-glycoprotein. Proc Natl Acad Sci USA 1998;
95:98319836.
Halestrap AP. The mitochondrial permeability transition: its molecular mechanism
and role in reperfusion injury. Biochem Soc Symp 1999; 66:181203.
Janssens D, Delaive E, Houbion A, Eliaers F, Remacle J, Michiels C. Eect of
venotropic drugs on the respiratory activity of isolated mitochondria and in
endothelial cells. Br J Pharmacol 2000; 130:15131524.
van Dijk C, Driessen AJ, Recourt K. The uncoupling eciency and anity of
avonoids for vesicles. Biochem Pharmacol 2000; 60:15931600.
Santos AC, Uyemura SA, Lopes JL, Bazon JN, Mingatto FE, Curti C. Eect of
naturally occurring avonoids on lipid peroxidation and membrane permeability
transition in mitochondria. Free Radic Biol Med 1998; 24:14551461.
Crompton M, Virji S, Doyle V, Johnson N, Ward JM. The mitochondrial permeability transition pore. Biochem Soc Symp 1999; 66:167179.
Dekermendjian K, Kahnberg P, Witt MR, Sterner O, Nielsen M, Liljefors T.
Structure-activity relationship and molecular modelling analysis of avonoids
binding to the benzodiazepine site of the rat brain GABA(A) receptor complex.
J Med Chem 1999; 42:43434350.
Cai J, Jones DP. Superoxide in apoptosis: mitochondrial generation triggered by
cytochrome c loss. J Biol Chem 1998; 273:1140111404.
Cai J, Yang J, Jones DP. Mitochondrial control of apoptosis: the role of cytochrome c. Biochim Biophys Acta 1998; 1366:139149.
Bors W, Heller W, Michel C. The chemistry of avonoids. In: Rice-Evans CA,
Packer L, eds. Flavonoids in Health and Disease. Vol. 1. New York: Marcel
Dekker, 1998:111136.
Jovanovic SV, Steenken S, Simic MG, Hara Y. Antioxidant properties of
avonoids: Reduction potentials and electron transfer reactions of avonoid
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
297
radicals. In: Rice-Evans, CA, Packer L, eds. Flavonoids in Health and Disease.
Vol. 1. New York: Marcel Dekker, 1998:137161.
40. Morel I, Cillard P, Cillard J. Flavonoid-metal interactions in biological systems.
In: Rice-Evans CA, Packer L, eds. Flavonoids in Health and Disease. Vol. 1. New
41. Ratty AK, Das NP. Eects of avonoids on nonenzymatic lipid peroxidation:
structure-activity relationship. Biochem Med Metab Biol 1988; 39:6979.
42. Gao Z, Huang K, Yang X, Xu H. Free radical scavenging and antioxidant activities
of avonoids extracted from the radix of Scutellaria baicalensis Georgi Biochim
Biophys Acta 1999; 1472:643650.
43. Haraguchi H, Yoshida N, Ishikawa H, Tamura Y, Mizutani K, Kinoshita T.
Protection of mitochondrial functions against oxidative stresses by isoavans from
Glycyrrhiza glabra. J Pharm Pharmacol 2000; 52:219223.
44. Arora A, Nair MG, Strasburg GM. Structure-activity relationships for antioxidant
activities of a series of avonoids in a liposomal system. Free Radic Biol Med
1998; 24:13551363.
45. Okado-Matsumoto A, Fridovich I. Subcellular distribution of superoxide
dismutases (SOD) in rat liver- Cu,Zn-SOD in mitochondria. J Biol Chem 2001;
276:3838838393.
46. Han D, Williams E, Cadenas E. Mitochondrial respiratory chain-dependent generation of superoxide anion and its release into the intermembrane space. Biochem J
2001; 353:411416.
47. Chang MS, Yoo HY, Rho HM. Positive and negative regulatory elements in the
upstream region of the rat Cu/Zn-superoxide dismutase gene. Biochem J 1999;
339:335341.
48. Das KC, Lewis-Molock Y, White CW. Activation of NF-kappa B and elevation of
MnSOD gene expression by thiol reducing agents in lung adenocarcinoma (A549)
cells. Am J Physiol 1995; 269:L588L602.
49. Seo SJ, Kim HT, Cho G, Rho HM, Jung G. Sp1 and C/EBP-related factor regulate
the transcription of human Cu/Zn SOD gene. Gene 1996; 178:177185.
50. Chang L, Karin M. Mammalian MAP kinase signalling cascades. Nature 2001;
410:3740.
51. Komatsu M, Hiramatsu M. The ecacy of an antioxidant cocktail on lipid
peroxide level and superoxide dismutase activity in aged rat brain and DNA
damage in iron-induced epileptogenic foci. Toxicology 2000; 148:143148.
52. Inanami O, Watanabe Y, Syuto B, Nakano M, Tsuji M, Kuwabara M. Oral
administration of (-)catechin protects against ischemia-reperfusion-induced neuronal death in the gerbil. Free Rad Res 1998; 29:359365.
53. Hengartner MO. The biochemistry of apoptosis. Nature 2000; 407:770775.
54. Pagliacci MC, Spinozzi F, Migliorati G, Fumi G, Smacchia M, Grignani F,
Riccardi C, Nicoletti I. Genistein inhibits tumour cell growth in vitro but enhances
mitochondrial reduction of tetrazolium salts: a further pitfall in the use of the
MTT assay for evaluating cell growth and survival. Eur J Cancer 1993; 29A:
15731577.
298
Boyd and Cadenas
55.
Russo M, Palumbo R, Tedesco I, Mazzarella G, Russo P, Iacomino G, Russo GL.

Quercetin and anti-CD95(Fas/Apo1) enhance apoptosis in HPB-ALL cell line.
FEBS Lett 1999; 462:322328.
Lee HC, Yin PH, Lu CY, Chi CW, Wei YH. Increase of mitochondria and
mitochondrial DNA in response to oxidative stress in human cells. Biochem J
2000; 348:425432.
Yoon HS, Moon SC, Kim ND, Park BS, Jeong MH, Yoo YH. Genistein induces
apoptosis of RPE-J cells by opening mitochondrial PTP. Biochem Biophys Res
Commun 2000; 276:151156.
Wang IK, Lin-Shiau SY, Lin JK. Induction of apoptosis by apigenin and related
avonoids through cytochrome c release and activation of caspase-9 and caspase-3
in leukaemia HL-60 cells. Eur J Cancer 1999; 35:15171525.
Schroeter H, Williams RJ, Matin R, Iversen L, Rice-Evans CA. Phenolic antioxidants attenuate neuronal cell death following uptake of oxidized low-density
lipoprotein. Free Radic Biol Med 2000; 29:12221233.
Susin SA, Lorenzo HK, Zamzami N, Marzo I, Snow BE, Brothers GM, Mangion J,
Jacotot E, Costantini P, Loeer M, Larochette N, Goodlett DR, Aebersold R,
Siderovski DP, Penninger JM, Kroemer G. Molecular characterization of
mitochondrial apoptosis-inducing factor. Nature 1999; 397:441446.
Hancock JT, Desikan R, Neill SJ. Does the redox status of cytochrome c act as a
fail-safe mechanism in the regulation of programmed cell death? Free Radic Biol
Med 2001; 31:697703.
Li X, Marani M, Mannucci R, Kinsey B, Andriani F, Nicoletti I, Denner L, Marcelli
M. Overexpression of BCL-XL underlies the molecular basis for resistance to
staurosporine-induced apoptosis in PC-3 cells. Cancer Res 2001; 61:16991706.
Pan MH, Lin JH, Lin-Shiau SY, Lin JK. Induction of apoptosis by penta-Ogalloyl-beta-D-glucose through activation of caspase-3 in human leukemia HL-60
cells. Eur J Pharmacol 1999; 381:171183.
Chung LY, Cheung TC, Kong SK, Fung KP, Choy YM, Chan ZY, Kwok TT.
Induction of apoptosis by green tea catechins in human prostate cancer DU145
cells. Life Sci 2001; 68:12071214.
Wenzel U, Kuntz S, Brendel MD, Daniel H. Dietary avone is a potent apoptosis
inducer in human colon carcinoma cells. Cancer Res 2000; 60:38233831.
Bastianetto S, Zheng W-H, Quirion R. The Ginkgo biloba extract (EGb 761)
protects and rescues hippocampal cells against nitric oxide-induced toxicity:
involvement of its avonoid constituents and protein kinase C. J Neurochem 2000;
74:22682277.
Bastianetto S, Ramassamy C, Dore S, Christen Y, Poirier J, Quirion R. The Ginkgo
biloba extract (EGb 761) protects hippocampal neurons against cell death induced
by beta-amyloid. Eur J Neurosci 2000; 12:18821890.
Schaeer HF, Weber MJ. Mitogen-activated protein kinase: specic messages
from ubiquitous messengers. Mol Cell Biol 1999; 19:24352444.
Allen RG, Tresini M. Oxidative stress and gene regulation. Free Radic Biol Med
2000; 28:463499.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.

70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
299
Xia Z, Dickens M, Raingeaud J, Davis RJ, Greenberg ME. Opposing eects of

ERK and JNK-p38 MAP kinases on apoptosis. Science 1995; 270:13261331.
Davis RJ. Signal transduction by the JNK group of MAP kinases. Cell 2000;
109:239252.
Tournier C, Hess P, Yang DD, Xu J, Turner TK, Nimnual A, Bar-Sagi D, Jones SN,
Flavell RA, Davis RJ. Requirement of JNK for stress-induced activation of the
cytochrome c-mediated death pathway. Science 2000; 288: 870874.
Whiteld J, Neame SJ, Paquet L, Bernard O, Ham J. Dominant-negative c-Jun
promotes neuronal survival by reducing BIM expression and inhibiting mitochondrial cytochrome c release. Neuron 2001; 29:629643.
Kang CD, Jang JH, Kim KW, Lee HJ, Jeong CS, Kim CM, Kim SH, Chung BS.
Activation of c-jun N-terminal kinase/stress-activated protein kinase and the
decreased ratio of Bcl-2 to Bax are associated with the auto-oxidized dopamineinduced apoptosis in PC12 cells. Neurosci Lett 1998; 256:3740.
Choi WS, Yoon SY, Chang, II, Choi EJ, Rhim H, Jin BK, Oh TH, Krajewski S,
Reed JC, Oh YJ. Correlation between structure of Bcl-2 and its inhibitory function
of JNK and caspase activity in dopaminergic neuronal apoptosis. J Neurochem
2000; 74:16211626.
Yamamoto K, Ichijo H, Korsmeyer SJ. BCL-2 is phosphorylated and inactivated
by an ASK1/Jun N-terminal protein kinase pathway normally activated at G(2)/M.
Mol Cell Biol 1999; 19:84698478.
Fan MY, Goodwin M, Vu T, Brantley-Finley C, Gaarde WA, Chambers TC.
Vinblastine-induced phosphorylation of Bcl-2 and Bcl-X-L is mediated by JNK
and occurs in parallel with inactivation of the Raf-1/MEK/ERK cascade. J Biol
Chem 2000; 275:2998029985.
Buschmann T, Potapova O, Bar-Shira A, Ivanov VN, Fuchs SY, Henderson S,
Fried VA, Minamoto T, Alarcon-Vargas D, Pincus MR, Gaarde WA, Holbrook NJ,
Shiloh Y, Ronai Z. Jun NH2-terminal kinase phosphorylation of p53 on Thr-81 is
important for p53 stabilization and transcriptional activities in response to stress.
Mol Cell Biol 2001; 21:27432754.
Oda K, Arakawa H, Tanaka T, Matsuda K, Tanikawa C, Mori T, Nishimori H,
Tamai K, Tokino T, Nakamura Y, Taya Y. p53AIP1, a potential mediator of p53dependent apoptosis, and its regulation by Ser-46-phosphorylated p53. Cell 2000;
102:849862.
Mazzoni IE, Said FA, Aloyz R, Miller FD, Kaplan D. Ras regulates sympathetic
neuron survival by suppressing the p53-mediated cell death pathway. J Neurosci
1999; 19:97169727.
Takekawa M, Saito H. A family of stress-inducible GADD45-like proteins
mediate activation of the stress-responsive MTK1/MEKK4 MAPKKK. Cell 1998;
95:521530.
Shiah SG, Chuang SE, Kuo ML. Involvement of Asp-Glu-Val-Asp-directed,
caspase-mediated mitogen-activated protein kinase kinase 1 cleavage, c-Jun Nterminal kinase activation, and subsequent Bcl-2 phosphorylation for paclitaxelinduced apoptosis in HL-60 cells. Mol Pharmacol 2001; 59:254262.
300
Boyd and Cadenas
83.
Bonni A, Brunet A, West AE, Datta SR, Takasu MA, Greenberg ME. Cell survival
promoted by the Ras-MAPK signaling pathway by transcription-dependent and
independent mechanisms. Science 1999; 286:13581362.
Deak M, Clifton AD, Lucocq LM, Alessi DR. Mitogen-and stress-activated protein
kinase-1 (MSK1) is directly activated by MAPK and SAPK2/p38, and may
mediate activation of CREB. EMBO J 1998; 17:44264441.
Park DS, Stefanis L, Yan CY, Farinelli SE, Greene LA. Ordering the cell
death pathway: dierential eects of BCL2, an interleukin-1-converting
enzyme family protease inhibitor, and other survival agents on JNK activation
in serum/nerve growth factor-deprived PC12 cells. J Biol Chem1996; 271:
2189821905.
Lander HM, Ogiste JS, Pearce SF, Levi R, Novogrodsky A. Nitric oxide-stimulated
guanine nucleotide exchange on p21ras. J Biol Chem 1995; 270:70177020.
Lander HM, Ogiste JS, Teng KK, Novogrodsky A. p21ras as a common signaling
target of reactive free radicals and cellular redox stress. J Biol Chem 1995;
270:2119521198.
Lander HM, Jacovina AT, Davis RJ, Tauras JM. Dierential activation of
mitogen-activated protein kinases by nitric oxide-related species. J Biol Chem
1996; 271:1970519709.
Lander HM, Hajjar DP, Hempstead BL, Mirza UA, Chait BT, Campbell S,
Quilliam LA. A molecular redox switch on p21ras. J Biol Chem 1997; 272:
43234326.
Go YM, Patel RP, Maland MC, Park H, Beckman JS, Darley-Usmar VM, Jo H.
Evidence for peroxynitrite as a signaling molecule in ow-dependent activation of
c-Jun NH2-terminal kinase. Am J Physiol 1999; 277:H1647H1653.
So HS, Park RK, Kim MS, Lee SR, Jung BH, Chung SY, Jun CD, Chung HT.
Nitric oxide inhibits c-Jun N-terminal kinase 2 (JNK2) via S-nitrosylation.
Biochem Biophys Res Commun 1998; 247:809813.
Park HS, Huh SH, Kim MS, Lee SH, Choi EJ. Nitric oxide negatively regulates cJun N-terminal kinase/stress-activated protein kinase by means of S-nitrosylation.
P Natl Acad Sci USA 2000; 97:1438214387.
Saitoh M, Nishitoh H, Fujii M, Takeda K, Tobiume K, Sawada Y, Kawabata M,
Miyazono K, Ichijo H. Mammalian thioredoxin is a direct inhibitor of apoptosis
signal-regulating kinase (ASK) 1. EMBO J 1998; 17:25962606.
Adler V, Yin Z, Fuchs SY, Benezra M, Rosario L, Tew KD, Pincus MR, Sardana
M, Henderson CJ, Wolf CR, Davis RJ, Ronai Z. Regulation of JNK signaling by
GSTp. EMBO J 1999; 18:13211334.
Cho SG, Lee YH, Park HS, Ryoo K, Kang KW, Park J, Eom SJ, Kim MJ, Chang
TS, Choi SY, Shim J, Kim Y, Dong MS, Lee MJ, Kim SG, Ichijo H, Choi EJ.
Glutathione S-transferase mu modulates the stress-activated signals by suppressing
apoptosis signal-regulating kinase 1. J Biol Chem 2001; 276:1274912755.
Rice-Evans C, Spencer JP, Schroeter H, Rechner AR. Bioavailability of avonoids and potential bioactive forms in vivo. Drug Metab Drug Interact 2000;
17:291310.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.

97.
98.
99.
100.
301
Spencer JP, Chowrimootoo G, Choudhury R, Debnam ES, Srai SK, Rice-Evans C.

The small intestine can both absorb and glucuronidate luminal avonoids. FEBS
Lett 1999; 458:224230.
Kuhnle G, Spencer JP, Chowrimootoo G, Schroeter H, Debnam ES, Srai SK, RiceEvans C, Hahn U. Resveratrol is absorbed in the small intestine as resveratrol
glucuronide. Biochem Biophys Res Commun 2000; 272:212217.
Kuhnle G, Spencer JP, Schroeter H, Shenoy B, Debnam ES, Srai SK, Rice-Evans
C, Hahn U. Epicatechin and catechin are O-methylated and glucuronidated in the
small intestine. Biochem Biophys Res Commun 2000; 277:507512.
Okushio K, Suzuki M, Matsumoto N, Nanjo F, Hara Y. Identication of (-)epicatechin metabolites and their metabolic fate in the rat. Drug Metab Dispos
1999; 27:309316.
11
Gene Modulation of HaCaT Cells
Induced by Pine Bark Extract
Bertrand Henri Rihn
Teaching Hospital of Brabois
Vandoeuvre, France
Claude Saliou
Johnson & Johnson Consumer Products Worldwide
Skillman, New Jersey, U.S.A.
Pine barks have been used for centuries as herbal remedies. During the 15th
century pine decoctions were used for wound healing, as noted by Minner
in 1497 in Thesaurus Medicaminum. Maritime Indians were the rst
reported to have used pine bark (decoction of white pine bark) to treat
Jacques Cartiers crew from scurvy during the winter of 1535. Maritime
Indians applied various barks to treat wounds and skin sores [1]. The
French maritime pine (Pinus maritima) bark extract (PBE) is a mixture
particularly rich in oligomeric procyanidins and other bioavonoids such as
taxifolin, catechin, and epicatechin. The French maritime pine grows on the
weather-beaten sand dunes of the Bay of Biscay in the southwestern corner
of France. Pine bark extracts have become popular again as dietary supplements. Beside their antioxidant functions, their other biological properties are
now being characterized. Genes are the Rosetta stone of human health and
disease. Thus gene expression analysis or genomics studies, using recently
developed complementary deoxyribonucleic acid (cDNA) arrays, help tremendously to identify markers of disease, therapeutic targets, and potential
pharmacological activities [2]. Using this approach, the eects of PBE on
the gene expression prole of the human keratinocyte (HaCaT) cell line
were investigated.
303
304
I.
Rihn and Saliou
INTRODUCTION
Numerous avonoids have been reported to aect cellular signaling processes,

from kinases to transcription factors. In fact, oligomeric procyanidins from pine
bark were found to prevent the activation of the proinammatory transcription
factor NF-n-B on ultraviolet (UV) exposure [3]. The same procyanidins are also
potent modulators of nitric oxide metabolism [4,5].
To understand the protective mechanism aorded by PBE-supplemented
cells, their basal gene expression prole was determined and compared to that of
nonsupplemented cells [6]. As expected, only a small proportion (83 genes) of the
588 genes analyzed were detected in either group. However, of these 83 genes, 39
genes showed an expression signicantly (more than two-fold) increased or
decreased (Table 1).
Interestingly, a group of overexpressed genes is involved in stress response.
Thioredoxin peroxidase 2 plays an important role in eliminating peroxides
generated during cellular metabolism and signaling cascades. This eect should
be related to the increase of gluthatione level in macrophage by PBE observed by
Rimbach and colleagues [7]. The ultraviolet (UV) excision repair protein
HHR23B is involved in nucleotide excision repair of deoxyribonucleic acid
(DNA) damage. Last, the heat shock protein HSP70 is generally expressed in
response to a stress. However, antioxidants such as curcumin have been suggested
to induce its expression, thereby increasing cell resistance to stress [8]. The
expression of the inhibitor for helix-loop-helix protein ID-3 is known to be
dysregulated in keratinocyte cell lines such as HaCaT [9]. The increased dynein
light chain 1 (dlc1) expression points to the regulation of intracellular tracking
through cytoskeleton interaction.
II.
PROTECTIVE EFFECT OF PINE BARK EXTRACT

AGAINST ULTRAVIOLET B
Using a similar approach, it is noteworthy that several UV-upregulated genes are

normalized after PBE pretreatment (Table 2). For instance, nucleoside diphosphate kinase B has a major role in the synthesis of nucleoside triphosphates [other
than adenosine triphosphate (ATP)]. In addition, nucleoside diphosphate kinase B
interacts with nuclease hypersensitive element (NHE) upstream of the c-myc gene
to transactivate c-myc [10]. The messenger level of thymosin h10 was increased
2.8-fold by UV and was normalized when cells were PBE-supplemented. This
protein plays an important role in the organization of the cytoskeleton by binding
to and sequestering actin monomers (g-actin) and, as a result, inhibits actin
polymerization. The heat shock 27-kd protein (HSP27) is known to be associated
with the cytoskeleton components a- and h-tubulin and the microtubules.

Table 1
305
Relative Expression of the Most and Least Expressed Genesa
GenBank
Genes overexpressed in PBE - supplemented cells
P/C
X69111
X67643
M11886
D21090
U32944
X67951
M36429
X00351
X53587
K00558
Inhibitor for helix - loop - helix protein ID - 3

Heat shock 70 - kd protein 1
HLA class 1 histocompatibility antigen C - 4 a chain
UV excision repair protein HHR 23B
Cytoplasmic dynein light chain 1
Thioredoxin peroxidase 2
Transducin h2
h - Actin
h4 - Integrin
a - Tubulin
10.3
5.7
4.9
4.7
4.6
3.9
3.7
2.0
1.7
1.6
Housekeeping genes
X56932
23 kd highly basic protein
1.1
Genes overexpressed in control cells

M22489
X01060
X59798
D13866
M74088
X06234
X06233
M28372
Bone morphogenetic protein 2A

CD71, transferin receptor protein
Cyclin D1 (G/S specic)
a - Catenin
Adenomatous polyposis coli protein
Calgranulin A
Calgranulin A
Cellular nucleic acid binding protein
0.08
0.07
0.07
0.07
0.05
0.04
0.04
0.04
HaCaT conuent cells were grown without and with 25 Ag/mL of PBE (Horphag Research
Ltd, Guernsey, UK). Twenty - four hours later the messengers were oligo - dT puried and [32P] dATP labeled. Approximately 1.106 CPM of each probe was used for lter hybridization. After
stringent washing, blots were exposed for phosphorimaging (BAS2000 BioImager). The cDNA dots
were analyzed, normalized to radioactivity per square millimeter, and the RNA abundance was
quantied as compared. P/C, ratio of relative abundancy of a given mRNA from PBE - supplemented
cells (P) or its equivalent from control cells (C); PBE, pine boric extract; dATP, IP - 3, CPM, HLA,
human leukocyte antigen; cDNA, complementary deoxyribonucleic acid; UV, ultraviolet, HaCaT,
human keratinocyte; mRNA, messenger ribonucleic acid.
a
Furthermore, HSP27, as do other chaperones, translocates to the nucleus during

heat shock. As for thymosine h10, HSP27 expression was normalized in PBEsupplemented cells. Consistent with the present results, other antioxidants have
been reported to prevent the upregulation of HSP27 [11]. Interestingly, anchorage dependence of butyrate-treated HT29 colon carcinoma cells was correlated
with a lower level of S19 messenger ribonucleic acid (mRNA) expression
306
Rihn and Saliou
Table 2
Cellsa
PBE Supplementation Reduces Ultraviolet Binduced Genes in HaCaT
Protein
60S Ribosomal
protein L6
40S Ribosomal
protein S19
Y Box binding
protein 1
G3PDH
Heat shock 27 kd
Nucleoside
diphosphate
kinase B
Thymosine h10
23 - kd Highly basic
protein
Genbank
CTRL
UV
PBE
PBE + UV
X69391
1.24
1.85
0.83
1.25
M81757
0.30
0.52
nd
0.34
M83234
0.28
0.54
0.28
0.19
X01677
X54079
L16785
0.95
0.94
0.11
1.03
6.46
0.26
0.60
1.16
0.16
0.60
1.07
0.07
M92381
X56932
0.60
1.00
1.65
1.00
nd
1.00
0.61
1.00
a
HaCaT cells were supplemented with PBE for 24 h before their exposure to sham (control cells)
or 150 mJ/cm2 UV (treated cells). mRNA were extracted 4 h after sham or UV exposure. The gene
expression of (1) control, (2) PBE - supplemented, (3) UV-treated, and (4) PBE-supplemented and
UV - treated was determined by high - density lter arrays (Atlas human cDNA expression arrays,
Clontech Laboratories, Inc.). One microgram of oligo - dT puried mRNA was [32P] - dATP - labeled.
Approximately 0.5.106 CPM of each probe was used for lter hybridization. After stringent washing,
blots were exposed for phosphorimaging (BAS2000 BioImager). The cDNA dots were analyzed,
normalized to radioactivity/per square millimeter by subtracting the nearest equivalent, and nonspecic dot considered as background, and the RNA abundance was quantied as compared to the
expression of 23 - kd highly basic protein (X56932). HaCaT, human keratinocyte; PBE, pine bark
extract; UV, ultraviolet; cDNA, complementary deoxyribonucleic acid; mRNA, messenger ribonucleic acid; [32P] - dATP, CPM, CTRL.
when compared with untreated cell [12]. Indeed, in our experiment, the level
of S19 mRNA expression was normalized by PBE treatment.
III.
CONCLUSIONS
Physiological and biochemical eects of PBE are, to date, not well studied: PBE
has been shown to inhibit platelet aggregation [13] and therefore proposed in
venous insuciency [14]. In 1999 some biochemical targets of PBE action were
proposed, e.g., inhibition of the respiratory electron transport chain [15]. PBE
also dose-dependently inhibited the activities of xanthine oxidase, xanthine
dehydrogenase, horseradish peroxidase, and lipoxygenase [16]. Nardili and
307
associates [17] demonstrated a clear eect of PBE on the protein kinases A and
C at 20 Ag/mL in vitro. Clinical trials with further evaluation on psoriatic
lesions are also required. There seems to be sucient evidence on use of PBE in
skin lesions, and PBE seems well tolerated, as indicated by trials in 2000 on
venous insuciency [14,18]. In addition, as dosage of ferrulic acid, a major
component of PBE, has become available [19], pharmacocinetic studies are
eventually possible.
ACKNOWLEDGMENTS
B.H.R. is indebted to M.C. Bottin and S. Mohr for excellent technical assistance.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
Chandler FR, Freeman L, Hooper SN. Herbal remedies of the Maritime Indians.
Gerhold D, Rushmore T, Caskey CT. DNA chips: promising toys have become
powerful tools. Trends Biochem Sci 1999; 24:168173.
Saliou C, Rimbach G, Moini H, McLaughlin L, Hosseini S, Lee J, Watson RR,
Packer L. Solar ultraviolet-induced erythema in human skin and nuclear factorkappa-B-dependent gene expression in keratinocytes are modulated by a French
maritime pine bark extract. Free Radic Biol Med 2001; 30:154160.
Packer L, Rimbach G, Virgili F. Antioxidant activity and biologic properties of a
procyanidin-rich extract from pine (Pinus maritima) bark, pycnogenol. Free Radic
Biol Med 1999; 27:704724.
Park YC, Rimbach G, Saliou C, Valacchi G, Packer L. Activity of monomeric,
dimeric, and trimeric avonoids on NO production, TNF-alpha secretion, and NFkappa B-dependent gene expression in RAW 264.7 macrophages. FEBS Lett 2000;
465:9397.
Rihn B, Saliou C, Bottin MC, Keith G, Packer L. From ancient remedies to
modern therapeutics: pine bark uses in skin disorders revisited. Phytother Res
2001; 15:7678.
Rimbach G, Virgili F, Park YC, Packer L. Eect of procyanidins from Pinus
maritima on glutathione levels in endothelial cells challenged by 3-morpholinosydnonimine or activated macrophages. Redox Rep 1999; 4:171177.
Sood A, Mathew R, Trachtman H. Cytoprotective eect of curcumin in human
proximal tubule epithelial cells exposed to shiga toxin. Biochem Biophys Res
Commun 2001; 283:3641
Langlands K, Down GA, Kealey T. ID proteins are dynamically expressed in
normal epidermis and dysrgulated in squamous cell carcinoma. Cancer Res 2000;
60:59295933.
308
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Rihn and Saliou

Postel EH, Berberich SJ, Flint SJ, Ferrone CA. Human c-myc transcription factor
PuF identied as nm23-H2 nucleoside diphosphate kinase, a candidate suppressor of
tumor metastasis. Science 1993; 261:428429.
Gorman AM, Heavey B, Creagh E, Cotter TG, Samali A. Antioxidant-mediated
inhibition of the heat shock response leads to apoptosis. FEBS Lett 1999;
445:98102.
Kondoh N, Schweinfest CW, Henderson KW, Papas TS. Dierential expression of
S19 ribosomal protein, laminin-binding protein, and human lymphocyte antigen
class I messenger RNAs associated with colon carcinoma progression and
dierentiation. Cancer Res 1992; 52:791796.
Araghi-Niknam M, Hosseini S, Larson D, Rohdewald P, Watson RR. Pine bark
extract reduces platelet aggregation. Integr Med 2000; 2:7377.
Arcangeli P. Pycnogenol in chronic venous insuciency. Fitoterapia 2000;71:236
244.
Moini H, Guo Q, Packer L. Enzyme inhibition and protein-binding action of the
procyanidin-rich French maritime pine bark extract, pycnogenol: eect on xanthine
oxidase. J Agric Food Chem 2000; 48:56309.
Moini H, Arroyo A, Vaya J, Packer L. Bioavonoid eects on the mitochondrial
respiratory electron transport chain and cytochrome c redox state. Redox Rep 1999;
4:3541.
Nardini M, Scaccini C, Packer L, Virgili F. In vitro inhibition of the activity of
phosphorylase kinase, protein kinase C and protein kinase A by caeic acid and a
procyanidin-rich pine bark (Pinus maritima) extract. Biochim Biophys Acta 2000;
1474:219225.
Petrassi C, Mastromarino A, Spartera C. Pycnogenol in chronic venous insuciency.
Phytomedicine 2000; 7:383383.
Virgili F, Pagana G, Bourne L, Rimbach G, Natella F, Rice-Evans C, Packer L.
Ferulic acid excretion as a marker of consumption of a French maritime pine (Pinus
maritima) bark extract. Free Radic Biol Med 2000; 28:12491256.
12
Cytoprotective and Cytotoxic Effects
of Flavonoids
Jeremy P. E. Spencer and Catherine A. Rice-Evans
London, England
Hagen Schroeter
London, England
I.
INTRODUCTION
Flavonoids have been ascribed a wide range of benecial properties related to

human health, including cancer [16]; cardiovascular diseases [7], including
coronary heart disease [2,4,8] and atherosclerosis [9]; inammation [4,10]; and
other diseases in which an increase in oxidative stress has been implicated [11
14]. Their ability to act as classical electron- (or hydrogen-) donating antioxidants in vitro has been intensively reported [1517] and used to explain their
protective eects against oxidative stress. Structurally important features that
dene this antioxidant activity are the hydroxylation pattern, in particular a 3V,4Vdihydroxy catechol structure in the B-ring and the presence of 2,3 unsaturation
in conjugation with a 4-oxo-function in the C-ring [17]. The antioxidant ecacy
of avonoids has been described for protection against oxidative damage to a
variety of cellular biomolecules. For example, avonoids inhibit the oxidation of
low-density lipoprotein [1822] and deoxyribonucleic acid (DNA) [2326] in
vitro. In addition, avonoids are eective scavengers of reactive nitrogen
species in the form of peroxynitrite [2730] and limit dopamine oxidation
mediated by peroxynitrite in a structure-dependent way involving oxidation or
309
310
Spencer et al.
nitration of the avonoid ring system [31]. Furthermore, their antioxidant

properties have also been attributed to their abilities to chelate transition metal
ions [3235] and their potential to quench singlet oxygen [36,37].
The ability of avonoids to act as eective antioxidants in vivo is dependent
on the extent of their conjugation and biotransformation on absorption. Flavonoids
are substrates for uridine diphosphate(UDP)-glucuronosyl transferases, catecholO-methyl transferases, and sulfotransferases in the small intestine and liver
[3843] and can also be degraded to secondary phenolic acid metabolites in the
colon [3840] (Chaps. 14 and 17). Most importantly it is now clear that those
avonoids with the most pronounced antioxidant activity, as demonstrated in
vitro, are those that are metabolized to the greatest extent in vivo [41]. This phase
I/II metabolism eectively lowers the antioxidant activity of the avonoid either
by aecting the B-ring catechol by O-methylation or by glucuronidating the Aring, thereby lowering hydrogen-donating ability and signicantly increasing the
rate of renal excretion. These observations have important implications for the
Figure 1 Flavonoids and their in vivo metabolite forms may act by direct
reaction with oxidizing species in the body, thereby reducing the accumulation of
end products of oxidative damage in the cell, or by modulation of intracellular
signaling events.
311
determination of their mechanism of protection against oxidative stress in vivo and

have provided evidence that their mode of protection in vivo may involve
processes that are independent of their classic antioxidant eects in scavenging
of reactive oxygen and reactive nitrogen species (Fig. 1).
This chapter examines the large body of evidence that demonstrates the
potential of avonoids to be both cytoprotective against oxidative stress (Sec. II)
and cytotoxic (Sec. III) in vitro. Many of these initial cytoprotection studies
characterized the avonoids as direct scavengers of free radical species, thereby
reducing the levels of oxidative damage to cell biomolecules (Sec. VI) and
attenuating the overall loss of cell function. More recent studies have also
highlighted that they may also act by mechanisms independent of classical Hdonating antioxidant reactions, such as specic interactions within intracellular
signalling pathways (Sec. IV). With all of these studies, however, caution must
be used in interpreting the results since in many instances, with the exception of
topical skin applications or eects in the oral cavity and gastrointestinal tract,
one must question whether the avonoids applied are present in the circulation
in that form and thus exposed to cells in vivo. With this in mind, the eects of
the more physiologically relevant in vivo avonoid metabolites on cells are also
reviewed (Sec. V).
II.
PROTECTION AGAINST OXIDATIVE STRESSINDUCED

CELL DAMAGE
The precise mechanisms by which avonoids may protect dierent cell

populations from oxidative insults are currently unknown. However, potential
mechanisms that involve their classical antioxidant properties, interactions with
mitochondria, modulation of intracellular signaling cascades, and stimulation of
adaptive responses have been proposed. Flavonoids, such as quercetin, have
been shown to be cytotoxic in many cell systems by mechanisms involving the
production of oxygen radicals through an auto-oxidation process dependent on
pH and the presence of oxygen. For example, quercetin has been observed to
initiate an adaptive response in low doses with the eect of protecting the cells
against higher doses of quercetin and other compounds, such as hydrogen
peroxide and mitomycin C [44]. However, most studies have not highlighted an
exact mechanism for the protective eects observed against oxidative cell
injury; often they have suggested that classical antioxidant processes are
involved. Other possible mechanisms might include an ability to increase
intracellular glutathione (GSH), the prevention of Ca2+ inux in the presence
of high levels of reactive oxygen species (ROS) [45], and direct interactions
within cell signaling cascades and with adenosine triphosphate (ATP) binding
sites (discussed in Sec. III).
312
A.
Spencer et al.
Protection Against Peroxide and Other Reactive

Oxygen Species
A number of studies have investigated the ability of avonoids to protect

against neuronal death induced by oxidative stress [4650] (Chap. 9). For
example, avonoids such as epicatechin and quercetin have been shown to
reduce the neurotoxicity induced by oxidized low-density lipoprotein [5053].
There has been an equivalent amount of interest in the ability of avonoids to
protect against oxidative injury, particularly that induced by hydrogen peroxide, in a variety of cell systems. For example, the avanol epicatechin has
been shown to evoke strong protection against cell damage, caspase-3
activation, and binding of annexin V-CY3.18 (an early marker of apoptosis)
induced by hydrogen peroxide in dermal broblasts [54,55] and primary
cortical neurons [55]. The protective eects of epicatechin were apparent at
low micromolar concentrations, similar to those reported to be detectable in
human plasma after oral ingestion of epicatechin [56] or green tea [57,58],
and were independent of direct interactions between the avonoid and the
peroxide. These ndings are consistent with those of other investigators who
have shown that the other main avanol, catechin, evokes signicant
protection against broblast (3T3 Swiss) damage induced by high levels of
reactive oxygen species (ROS) generated extracellularly by xanthine-xanthine
oxidase [59] and against hydrogen peroxideinduced damage to cultured rat
hepatocytes (BL-9) [60]. The protective eect of catechins has also been
studied in cultured human umbilical vein endothelial cells exposed to toxicity
induced by linoleic acid hydroperoxide (LOOH) [61]. Catechins interacted
with LOOH present in the medium or near the surface of membranes, but not
with LOOH incorporated into cellular membranes. In addition, the interaction
of catechin with a-tocopherol may provide synergistic protection against the
cytotoxicity of LOOH [61].
The avonol quercetin has been the subject of much interest in terms of its
benecial properties against oxidative stress [62] and has been shown to exert a
potent protective actions against hydrogen peroxideinduced apoptosis of rat
thymocytes [63] and cell death in rat hepatocytes (BL-9) [60], as well as reducing
oxidative stress and cell damage in liver tumor cells induced by AAPH (2,2azobis(2-aminopropane) dihydrochloride) [64]. Quercetin and another avonol,
kaempferol, as well as catechin and the avone taxifolin, have been observed to
suppress the cytotoxicity of O2. and H2O2 to Chinese hamster V79 cells, as
assessed by the ability of the avonoids to prevent the decrease in the number of
cell colonies induced by the oxidants [65]. Furthermore, quercetin has been
shown to protect cutaneous tissue-associated cells (human skin broblasts,
keratinocytes, and endothelial cells) from oxidative injury induced by buthionine
sulfoximine (BSO), an inhibitor of GSH synthesis [66].
313
Baicalein (5,6,7-trihydroxy-2-phenyl-4H-1-benzopyran-4-one), a naturally

occurring avone, has also been observed to prevent human dermal broblast
cell damage induced by reactive oxygen species such as hydrogen peroxide, tertbutyl hydroperoxide, and superoxide anions in a concentration-dependent
manner [67]. Interestingly, the protection evoked by baicalein was more
eective than that of the iron chelator deferoxamine, the hydroxyl radical
scavengers dimethyl sulfoxide and ethanol, the lipid peroxidation chain blocker
a-tocopherol, and the xanthine oxidase inhibitor allopurinol, indicating that the
protective eects may involve a more complex mechanism than simple direct
oxidant scavenging. Furthermore, Scutellaria baicalensis Georgi, which also
contains baicalein as well as the other avones, wogonin, wogonoside, and
baicalin, has been found to be eective in protecting human neuroblastoma SHSY5Y cells against hydrogen peroxideinduced cell damage and lipid peroxidation [68,69].
The avone glycoside diosmin and its aglycone form, diosmetin, are
reported to be eective inhibitors of cell membrane damage in primary cultures
of rat hepatocytes induced by erythromycin estolate and oxidative stress caused
by tert-butylhydroperoxide (tBH) [70]. Diosmetin, but not diosmin, protected
against tBH toxicity, and this protection was related to a decreased extent of
lipid peroxidation and reduced loss of glutathione. Metal complexes (Fe2+, Fe3+,
Cu2+, Zn2+) of the quercetin glycoside, rutin, and epicatechin have been found to
be more potent than the parent avonoids in protecting red blood cells against
asbestos-induced injury [71], suggesting that avonoid metal complexes may be
an additional eective therapy for the inammatory response associated with the
inhalation of asbestos bers.
These studies highlight the abilities of avonoids, in particular quercetin
and epicatechin, to reduce oxidative stress and cell death induced by ROS in
dierent cell systems. Although it is likely that avonoids can reduce
oxidative stressinduced damage to cells by direct interactions with the
oxidants in the extracellular environment, a few studies suggest that other
mechanisms are involved, as protection is also observed when the oxidant and
avonoid are not present together in the culture medium. In this case,
intracellular interactions between oxidant and avonoid may explain the
protection observed. However, as will be outlined later, other nonantioxidant
mechanisms have also been proposed.
B.
Protection Against Ultraviolet A /Ultraviolet BInduced

Cell Damage
Ultraviolet (UV) radiationinduced oxidative stress in skin cells [72,73], as

well as inltration of leukocytes and depletion of antigen-presenting cells [74],
314
Spencer et al.
plays an important role in the induction of immune suppression [74,75], DNA

damage [76] and photocarcinogenesis [76,77]. In particular, ultraviolet B (UVB) exposure may result in oxidative stress and skin injury that has been
associated with a variety of skin diseases including photoaging [78], inammation [74,75,7981], and cancer [75,76]. Many investigations have shown
that topical treatment with [8285] or oral consumption [86] of green tea
polyphenols inhibits UV-B-induced skin tumorigenesis in dierent animal
models. The most chemopreventive constituent in green tea, proposed to be
responsible for the biochemical or pharmacological eects observed in these
studies, is ()epigallocatechin-3-gallate, although treatment of the skin with
green tea polyphenols in general has been shown to modulate the biochemical
pathways involved in inammatory responses, cell proliferation, and responses
of chemical tumor promoters, as well as ultraviolet lightinduced inammatory markers of skin inammation [82,84,85].
The protective eects of green tea treatment on human skin either
topically or consumed orally against UV lightinduced inammatory or
carcinogenic responses are not well understood. Pretreatment of cultured sh
cells with quercetin inhibited both the photoinduction of cyclobutane pyrimidine dimer photolyase and enhancement of cellular attachment in a similar
manner to that of H7 (1-(5-isoquinoline solfunil)-2 methlypiperazine), a strong
inhibitor for protein kinase C [87], suggesting that interactions in cell signaling
processes might be involved. Epigallocatechin gallate (EGCG) has been shown
to exert preventive eects against photocarcinogenesis and phototoxicity in
mouse models. The topical application of EGCG to human skin before UV
irradiation (four times minimal erythema dose) caused decreases in UVinduced production of hydrogen peroxide (6890%) and nitric oxide (30
100%) in both the epidermis and the dermis in a time-dependent manner [85].
EGCG pretreatment also inhibited UV-induced inltration of inammatory
leukocytes, particularly CD11b(+) cells (a surface marker of monocytes/
macrophages and neutrophils), into the skin, reduced UV-induced epidermal
lipid peroxidation, was found to restore the UV-induced decrease in GSH
level, and protected against the decrease in glutathione peroxidase activity.
The topical application of EGCG to C3H/HeN mice before a single dose of
UV-B (90 mJ/cm2) exposure inhibited UV-B-induced inltration of leukocytes,
specically the CD11b+ cell type, and myeloperoxidase activity, a marker of
tissue inltration of leukocytes [82]. EGCG treatment was also found to
prevent UV-B-induced depletion in the number of antigen-presenting cells
when immunohistochemically detected as class II major histocompatibility
complex (MHC+) la+ cells. These data suggest that topical applications of
EGCG to skin may aord some protection against UV-B-induced immunosuppression, photoaging, inammatory dermatoses, and photocarcinogenesis.
III.
315
PRO-OXIDANT EFFECTS OF FLAVONOIDS

IN CELL SYSTEMS
The huge interest in avonoids and isoavones, fueled by their powerful

antioxidant and estrogenic eects, respectively, has led to their proposed use
as anticarcinogens and cardioprotective agents, prompting a dramatic increase in
their consumption as dietary supplements. However, there are a number of
studies that highlight the potentially toxic eects of excessive avonoid intake
[3,8891]. For example, at higher doses (high micromolar to low millimolar) in
cell studies, avonoids may act as cytotoxic agents [91]; as mutagens [92]; as
pro-oxidants that generate free radicals [90,93,94]; as inducers of apoptosis
[3,95]; and as inhibitors of key enzymes involved in hormone metabolism.
Flavonoids containing a free hydroxyl group at position 3 of the C ring, a free
hydroxyl group at position 7 of the A-ring, and a B ring with a catechol
(quercetin) or pyrogallol structure (epigallocatechin or epigallocatechin gallate),
or a structure that after metabolic activation is transformed into a catechol or a
pyrogallol, have been suggested to be the most genotoxic to eukaryotic cells.
However, these cytotoxic eects are concentration-dependent as avonols such
as quercetin and myricetin are damaging to cells at much lower concentrations
than avanols such as epicatechin or non-catechol-containing avanones such as
hesperetin. This nding is in agreement with an induction of revertants in
Salmonella typhimurium TA98 and the induction of chromosomal aberrations in
V79 cells by avonols such as quercetin [92]. Quercetin, in particular, although
possessing several biological activities that can be useful in protection against
oxidative stress [60,63], has been discussed as a compound that is potentially
cytotoxic and/or genotoxic to normal or cancer cells [96104].
A.
Generation of Reactive Oxygen Species by Flavonoids

in Cell Cultures Systems
Recently there has been interest in the ability of some avonoids to generate
cytotoxic amounts of hydrogen peroxide or express a pro-oxidant nature in some
cell culture models. In cell culture, chemical reactions between media and
avonoid are often ignored, in particular the potential artifactual generation
of ROS and hydrogen peroxide by reducing agents. It has been shown that
addition of high concentrations of strong reducing agents (high micromolar to
millimolar), such as avonoids, to cell culture media can lead to generation of
substantial amounts of H2O2 (10100 AM) [93]. This eect has led to concern
that some of the reported eects of avonoids such as quercetin and epigallocatechin gallate on cells in culture may be due to H2O2 generation, produced as a
result of the interaction of these compounds with cell culture media. This
316
Spencer et al.
compound-dependent generation may be viewed both negatively in that peroxide

is a known inducer of both apoptosis and necrosis in cells [50,55] and positively
in that small amounts of generated peroxide may result in an adaptive response in
cells that would act to protect them against a subsequent oxidative insult. It
should be noted, however, that artifactual generation of peroxide is negligible
when avonoids are exposed at physiologically relevant low concentrations in
media containing transferrin [50,55].
As mentioned, the ability of avonoids to generate peroxide seems to be
dependent on the presence of a pyrogallol or catechol structure; the former
avonoids generate more peroxide than the latter [in decreasing ability to
generate peroxide: myricetin > baicalein > quercetin > (-)-epicatechin > (+)catechin > setin = 7,8-dihydroxy avone] [94] (Fig. 2). The amount of peroxide
generated by myricetin (pyrogallol-type avonoid) was proportional to its
concentration and was dependent on the amount of dissolved oxygen in the
buer and was inhibited by the addition of superoxide dismutase [94]. These
results suggest that under specic culture conditions, where oxygen saturation is
high, some avonoids may generate small amounts of hydrogen peroxide by
donation of electrons from their pyrogallol or catechol structures to oxygen, to
form the superoxide anion radical and subsequently peroxide.
Other investigations have provided evidence that avonoids may also
induce DNA damage in cultured cells [105,106]. For example, morin and
naringenin [106] and kaempferol [107] induce a concentration-dependent
peroxidation of nuclear membrane lipids and DNA strand breaks in isolated
rat liver nuclei [106]. Quercetin and myricetin have been shown both to inhibit
growth of and cause dose-dependent increases in DNA strand breakage in Caco2, HepG2, HeLa cells, and normal human lymphocytes [24]. None of the
avonoids was observed to induce DNA base oxidation above normal cellular
levels, although all caused a depletion of reduced glutathione, which, in the case
of quercetin, occurred before cell death. These studies demonstrate the prooxidant activities of polyphenolic avonoids, which are generally considered as
antioxidants and anticarcinogens, and may suggest their possible dual role in
mutagenesis and carcinogenesis.
B.
Pro-Oxidant Effects and Anticarcinogenesis
The ability of avonoids to act as pro-oxidants in vitro and in cell systems has also
been suggested to be a potential mechanism by which they may act as anticarcinogenic compounds in vivo as a result of their abilities to promote death of
cancer cells. Quercetin has been thoroughly investigated for its abilities to express
antiproliferative eects [100,103,108,109] and induce death predominantly by an
apoptotic mechanism in cancer cell lines [95,9799,102105,110116]. For
example, the exposure of quercetin and the isoavone genistein to the colonic
317
Figure 2 Structures of some common flavonols: (A) myricetin, (B) baicalein, (C)
quercetin, (D) fisetin, and (E) kaempferol. Light shading highlights the presence of
a pyrogallol structure, whereas dark shading highlights the presence of a catechol
function. The ability to generate hydrogen peroxide decreases from (A) to (E).
318
Spencer et al.
cancer cell lines, Caco-2, HT-29, and rat nontransformed intestinal crypt cells,
IEC-6 induced strong chromatin condensation, a marker of apoptosis [109].
Furthermore, quercetin has been found to induce chromatin and nuclear fragmentation in human leukemia HL-60 cells [115], increase caspase-3 activation in the
malignant cell line HPB-ALL [113] and HL-60 cells [104], and cause activation of
caspase-9 and release of cytochrome c in HL-60 cells [104]. The inhibition of
cancer cell proliferation by quercetin may proceed via an inhibition of epidermal
growth factor (EGF) receptor phosphorylation, an integral part of the proliferation
mechanism in cultures of colonic tumor cells [114], or via inhibition of cell cycle
progression through transient M phase accumulation and subsequent G2 arrest, as
was observed in MCF-7 human breast cancer cells [111]. The many studies with
quercetin in this area would suggest that this avonol is a promising candidate
chemotherapy in vivo. However, for bioavailability data suggest that as a result of
the extensive metabolism of quercetin in the small intestine and liver the potential
for action in vivo may be limited [41].
Other avonoids have also been shown to be eective in cancer cell killing.
The avone apigenin is capable of inducing apoptotic cell death characterized by
DNA fragmentation and activation of caspases [95]. The involvement of hydrogen peroxide in the mechanism for cytotoxicity was ruled out in this study as
catalase failed to eliminate the cytotoxic eects. A synthetic avone acetic acid
(FAA) that has been reported to have antitumor activity against a variety of
transplanted tumors in mice inhibits the proliferation of endothelial cells by a
superoxide-dependent mechanism and induces apoptosis by a nitric oxide and
superoxide-independent mechanism [117]. In addition, pycnogenol, a preparation derived from pine bark, which contains high amounts of procyanidins [13],
selectively induces death in human mammary cancer cells (derived from human
brocystic mammary tissue) but not in normal human mammary MCF-10 cells
[118], indicating that avonoids may be of use for the selective killing to tumor
cells in vivo. Wine polyphenols, which include epicatechin, have been shown to
have a direct inhibitory eect on the proliferation of human prostate cancer cell
lines that was found to be mediated by the production of NO [119]. Furthermore,
two dietary avonoids isolated from the leaves of Morus alba (Moraceae),
quercetin-3-O-h-D-glucopyranoside and quercetin-3,7-di-O-h-D-glucopyranoside, were found to possess a signicant inhibitory eect on the growth of the
human promyelocytic leukemia cells (HL-60) [120]. Genistein and quercetin
have been found to inhibit the proliferation and migration of ras-oncogene-driven
tumor cells, rat breast adenocarcinoma, and human bladder carcinoma cell lines
[121]. Baicalin and baicalein have been shown to induce apoptosis in the
androgen-positive and -negative human prostatic carcinoma cell lines LNCaP
and JCA-1 [116].
Resveratrol (3,4V,5-trihydroxy-trans-stilbene) (Fig. 3), a common phytoalexin found in grape skins, peanuts, and red wine, has been speculated to
319
Figure 3 Structure of resveratrol. Highlighted areas show its structural similarity

with the flavonoid kaempferol in both the A-ring and the B-ring.
act as an antioxidant, promote nitric oxide production, inhibit platelet

aggregation, increase high-density lipoprotein cholesterol, and thereby serve
as a cardioprotective agent [122]. It has structural similarity to certain
avonoids such as kaempferol in that its structure mimics both the A-ring
and the B-ring of the avonoid (Fig. 3). In 2001, it was proposed that
resveratrol can function as a potent cancer chemopreventive agent, and there
has been a great deal of experimental eort directed toward dening this
eect [122,123]. Resveratrol has been shown to be a potent inducer of
apoptosis in many dierent cancer cell models [124132] and is also capable
of producing strong growth inhibitory eects on cancer cells, causing both
cell cycle arrest [132134] and inhibition of cell proliferation [135138].
However, in 2002 resveratrol was found to have no growth-inhibitory eects
on 4T1 breast cancer in vivo [139], even though it is a potent inhibitor of
4T1 breast cancer cells in vitro [139]. This lack of activity in vivo may be
due to the extensive metabolism of resveratrol that occurs on absorption to
produce glucuronidated [140,141] and sulfated [136,142] forms that may
possess dierent bioactivity against cancer cells. At present there are no data
to suggest that these metabolites are capable of cancer chemopreventive
eects similar to those of resveratrol.
320
C.
Spencer et al.
Depletion of Cellular Thiols
A possible mechanism of the cytotoxic nature of avonoids has been linked to

their potential intracellular interactions with thiols such as glutathione [143,144].
Catechin oxidation in the presence of cellular peroxidases and hydrogen peroxide
leads to the generation of monoglutathione conjugates (via quinone intermediates), which has the eect of lowering cellular GSH levels [144]. Quercetin [145]
and other avonoids such as taxifolin, luteolin, setin, and 3,3V,4V-trihydroxyavone have also been found to react with GSH to generate mono-and
diglutathionyl adducts [146148]. Interestingly, the conjugation of GSH with
catechin occurred on the B-ring [144], in contrast to the A-ring of quercetin, as a
result of the formation of quinone methidetype intermediates [145,147].
Quercetin exposure to both dermal broblasts and cortical neurons led to the
rapid formation of glutathionyl conjugates of quercetin and a depletion in cellular
thiols (Spencer et al., in preparation).
Similar chemical reactions are involved in the reaction of other catechols
such as the catecholamines, dopamine, and L-dihydroxyphenylalanine (LDOPA) with cysteine or GSH [149152] and can lead to the generation of
mitochondrial toxins with relevance to Parkinsons disease [153155]. As well
as possible cytotoxic eects of lowering cellular thiol levels or binding to
cysteine residues at the active site of specic enzymes, avonoids could act
benecially by acting to limit the formation of the potentially cytotoxic
catecholamine-thiol adducts, in a manner similar to that observed for dihydrolipoic acid [151]. Because of the structural similarity of the avonoid B-ring and
their ability to donate electrons efciently to form quinones, it is conceivable
that specic avonoids may be of use to prevent such neurotoxic compounds as
5-S-cysteinyl dopamine from forming in vivo.
IV.
ANTIOXIDANT EFFECTS INDEPENDENT OF

CLASSICAL REDUCING PROPERTIES
Many studies have demonstrated that avonoids have mechanisms of action

independent of their conventional hydrogen-donating free radical scavenging
properties by modulating enzyme activities, interfering with pathways of
intermediary metabolism, downregulating the expression of adhesion molecules,
acting at various sites within signal transduction cascades, and mimicking
substrates for various binding sites [4,156159]. Apigenin has been identied
as a avonoid that eectively blocks intercellular adhesion molecule 1 (ICAM1) upregulation and leukocyte adhesion in response to cytokines in vitro. Tumor
necrosis factor(TNF)-induced ICAM-1 upregulation in vivo eectively is
blocked by apigenin through a mechanism that is unrelated to free radical
321
scavenging or leukocyte function [156]. Furthermore, the adhesion of leukocytes

to cytokine-treated endothelial cells was blocked in cells cotreated with apigenin
[157]. Epigallocatechin strongly reduces ultraviolet-A-induced heme oxygenase
1 activation in skin-derived broblasts, however, it was also observed to activate
collagenase and cyclooxygenase expression (635), indicating that the eect of
this green tea polyphenol on cellular stress responses is complex and may
involve direct eects on signal transduction as well as changes that may be
associated with its antioxidant activity.
The potential role avonoids play in intracellular signaling events triggered
in response to oxidative stress is becoming increasingly clear, especially with
regard to the inuence they have on protein kinases, phosphatidylinositol-3
kinase, and nuclear factor-nB (NF-nB) [160,161]. Among the identied signal
transducers aected by avonoids are phosphoinositide 3-kinase (PI-3K) and
protein kinase C (PKC), which are key players in many cellular responses,
including cell multiplication, apoptosis, and transformation. The strong inhibitory
eects of quercetin on both NF-nB binding activity and oxidative deoxyribonucleic acid (DNA) damage suggest that its antioxidant activity may outweigh its
oxidative potential in the cellular environment, which might contribute to
quercetins reported anticarcinogenic and anti-inammatory eects. Quercetin
has also been reported to be an inhibitor of phospholipase A2 (PLA2) and
lipoxygenase and eectively inhibits the oxidative stressinduced expression of
heat shock protein 68 (Hsp68) [162]. Accumulating evidence also links avonoids to interactions within mitogen-activated protein kinase (MAPK) signaling
cascades [50,163166]. For example, quercetin downregulates both phorbol 12myristate 13-acetate (PMA) and tumor necrosis factor-a (TNF-a) expression in
human endothelial cell line ECV304 (ECV) via inhibiting both activator protein 1
(AP-1) activation and the Jun Kinase (JNK) pathway [163]. In addition, quercetin
has been observed to exert a signicant inhibitory eect on 4-hydroxy-2nonenal (HNE)-induced JNK activation [167]. It has been proposed that low
concentrations of avonoids, such as quercetin, may activate the MAPK pathway
[extracellular signal-regulated kinase (ERK2), JNK1, p38], leading to expression
of survival genes (c-Fos, c-Jun) and defensive genes (phase II detoxifying
enzymes; glutathione-S-transferase, quinone reductase), resulting in survival
and protective mechanisms (homeostasis response). However, increasing the
concentrations of these compounds additionally activates the caspase pathway,
leading to apoptosis [164]. The Gingko biloba extract, EGb 761, which is rich in
avonoids, has been observed to suppresses c-fos messenger ribonucleic acid
(mRNA) expression, followed by AP-1 DNA activation and in Jurkat T cells,
suggesting that the step in the signal transduction pathway for AP-1 activation
that is inhibited by EGb 761 is upstream to c-fos mRNA expression [168].
Recent interest has been directed to the use of avonoids and avonoid
derivatives as benzodiazepine receptor (BDZ-R) ligands [159]. Benzodiazepines
322
Spencer et al.
are the most widely prescribed class of psychoactive drugs in current therapeutic
use but have unwanted side eects such as sedation, myorelaxation, ataxia,
amnesia, and ethanol and barbiturate potentiation, and tolerance, and it is
postulated that a new family of ligands based on avonoids might prevent these
side eects. First isolated from plants and used as tranquilizers in folkloric
medicine, some natural avonoids have been shown to possess a selective and
relatively mild anity for BDZ-Rs and have a pharmacological prole compatible with a partial agonistic action. In a logical extension of this discovery, of
various synthetic derivatives of those compounds, such as 6,3V-dinitroavone,
have been found to have a very potent anxiolytic eect not associated with
myorelaxant, amnestic, or sedative actions [159]. Because of their selective
pharmacological prole and low intrinsic ecacy at the BDZ-Rs, avonoid
derivatives, such as those described, could represent an improved therapeutic tool
in the treatment of anxiety.
Catechins and specic avonols have been reported to interact with
proteins such as the mitochondrial adenosine triphosphatase (ATPase) [169],
calcium plasma membrane ATPase [170], protein kinase A [171173], and
protein kinase C [160,174176] through binding to the ATP binding site [177].
Quercetin was found to cause a 50% inhibition of calcium transport at a
concentration of 15 AM; morin and rutin had similar eects at concentrations
approximately 10-fold higher [170]. The order of inhibitory potency of the
avonoids on the Ca2+-transport ATPase from synaptosomal vesicles could be
linked to their solubility in the membrane lipid phase, since, in addition, quercetin
exhibited strong inhibition of phosphorylation of membrane proteins by ATP in
synaptosomal vesicles, whereas rutin and morin had no signicant eect [170].
Furthermore, certain avonoids have been shown to be very potent and highly
selective inhibitors of cyclin-dependent kinases (CDKs), which regulate the
initiation, progression, and completion of the cell cycle, and are thus critical
for cell growth [177]. As tumor development is closely associated with genetic
alteration and deregulation of CDKs and their regulators, these avonoid
inhibitors of CDKs may act as useful anticancer therapeutics in the future.
Epicatechin, epigallocatechin, epicatechin gallate, and epigallocatechin
gallate have been shown to reduce the ability of macrophages to bind oxidized
erythrocytes, suggesting that the activity of lectinlike receptors of macrophages
for oxidized erythrocytes may be regulated by oxidative mechanisms [178]. The
same green tea catechins have also been shown to aect pro- and anti-inammatory cytokines produced by human leukocytes in vitro. The production of
interleukin-1h was decreased and that of interleukin 10 enhanced in leukocytes
treated with EGC, ECG, and EGCG, but there was no eect on the production of
interleukin-6 or tumor necrosis factor-a [179]. Although these eects suggest
anti-inammatory properties of the tea-derived catechins, they were observed at
concentrations that were unlikely to be achievable in plasma in vivo and are
323
therefore unlikely to contribute to the protective eects of tea-derived avonoids

in inammatory diseases. The antioxidative activity of the tea polyphenols
theaavin, theaavin-3-gallate, theaavin-3,3V-digallate, (-)-epigallocatechin-3gallate, and gallic acid has been shown to be due not only to their ability to
scavenge ROS and superoxide in HL-60 cells but also to their ability to block
xanthine oxidase and related oxidative signal transducers [180].
It has also been suggested that the protective mechanism of avonoids may
include the inhibition of enzymatic functions other than oxidases, e.g., the
inhibition of lipoxygenase and thus prevention of the formation of leukotrienes
[181185]. Analysis of 12 aglycone avonoids showed that inhibitory potency
and selectivity against 5-lipoxygenase are conferred by the presence of a catechol
group in ring B as part of a 3,4-dihydroxycinnamoyl structure [186]. In contrast,
cross-over of inhibitory selectivity is observed in compounds containing few
hydroxyl substituents (with None in ring B), which are selective against cyclooxygenase. Other studies indicate that the avonoids such as taxifolin, eriodictyol, hesperetin, and luteolin can inhibit myeloperoxidase (MPO) release from
activated human neutrophils, and taxifolin and eriodictyol also strongly inhibit
MPO activity [187]. Furthermore, avonoids that are methylated at a single OH
group in the B-ring are only inhibitory when they react with activated neutrophils
in the presence of myeloperoxidase. Early investigations demonstrated that
quercetin, apigenin, and taxifolin (dihydroquercetin) inuence antiimmunoglobulin E(IgE)-induced histamine release and hydrogen peroxide generation
in basophil-containing leukocyte suspensions [188,189] and may impair the
oxidative burst of neutrophils to an extent dependent on their hydrophobicity
[190,191]. For example, quercetin inhibited the generation of superoxide anions
by neutrophils [191].
Flavonoids have also been investigated for their ability to inuence the
normal functioning of mitochondria and in particular the mitochondrial respiratory chain complexes [192195]. The avonoids robinetin, rhamnetin, eupatorin,
baicalein, 7,8-dihydroxyavone, and norwogonin all inhibited beef heart mitochondrial succinoxidase and NADH-oxidase activities and, in every case, the
concentration required to inhibit low-density lipoprotein oxidation by 50% (IC50)
observed for the NADH-oxidase enzyme system was lower than for succinoxidase activity, demonstrating a primary site of inhibition in the complex I reduced
nicotinamide-adenine dinucleotide(NADHcoenzyme Q reductase) portion of
the respiratory chain [192,193]. In addition, the avonoids with adjacent trihydroxyl, para-dihydroxyl, or ortho-dihydroxl (catechol) groups exhibited a
substantial rate of auto-oxidation, which was accelerated by the addition of
cyanide, demonstrating that the CN-/avonoid interaction can generate superoxide nonenzymatically.
Luteolin and quercetin have been shown to be potent antileishmanial agents
and may therefore be useful in the chemotherapy of leishmaniasis, which is a
324
Spencer et al.
major concern in developing countries. Both luteolin and quercetin inhibited the
growth of Leishmania donovani promastigotes and amastigotes in vitro, inhibited
DNA synthesis in promastigotes, and promoted topoisomerase IImediated
linearization of kDNA minicircles [196]. Quercetin has nonspecic eects on
normal human T cells; however, luteolin was nontoxic, suggesting it could be a
strong candidate for antileishmanial drug design. Another study suggested that
the skin cancer chemopreventive eects of silymarin (a avonoid drug from
Silybum marianum used in liver disease) are mediated via impairment of EGFR
signaling that ultimately leads to perturbation in cell cycle progression, resulting
in the inhibition of proliferation and induction of growth arrest [197].
V.
IN VIVO FLAVONOID METABOLITES
At present, studies investigating the bioactivity of biologically relevant avonoid

metabolites, which include glucuronides and O-methylated forms, are few
[54,55]. However, a number of investigations from 1998 to 2001 studied the
ability of the biologically relevant metabolites of quercetin to provide protection
against lipid peroxidation, in particular the oxidation of low-density lipoprotein
(LDL) [198201]. Both quercetin and quercetin 3-O-h-D-glucuronide were
found to protect against peroxynitrite-induced consumption of endogenous
lycopene, h-carotene, and a-tocopherol, suggesting that dietary quercetin, and
one of its major in vivo forms, may be capable of inhibiting peroxynitrite-induced
oxidative modication of LDL in the circulation [198]. Other studies have also
helped to strengthen the proposal that this conjugated quercetin metabolite
retains the ability to protect cellular and subcellular membranes from peroxidative attack by reactive oxygen species and peroxidative enzymes [199,201].
There are also a few studies that have studied the inuence of metabolism
on the bioactivity of avonoids in cell models [50,54,55,202,203]. The ability
of glucuronidated epicatechin (epicatechin-7-and epicatechin-5-O-h-D-glucuronides) to protect neurons and broblasts against oxidative stressinduced cell
death has been investigated [55]. A mixture of epicatechin-5-O-h-D-and
epicatechin-7-O-h-D-glucuronides exhibits no signicant protection against
peroxide-induced loss in neuronal or broblast viability and fails to prevent
peroxide-induced caspase-3 activation in both cell models [55] (Fig. 4). This
lack of activity may be based on the increased polarity of the glucuronide and
its reduced ability to partition, which would limit its access to cells. This
conclusion is consistent with the observation that uptake of epicatechin
glucuronide into both cortical neurons and dermal broblasts was not detectable
[55]. Another possibility is that the A-ring is the structurally important feature
for cell recognition or biological action within cells and in vivo, and presence
of a bulky glucuronide on the A-ring limits its activity. However, in vivo, the
325
Figure 4 Plot 1: Protection against peroxide-induced cell damage by epicatechin

(5), 3V-O-methyl epicatechin (.), and epicatechin glucuronide (E). Plot 2:
Increased activity of caspase-3-like protease activity induced by hydrogen
peroxide (50 AM) and inhibition by pretreatment with epicatechin, 3V-O-methyl
epicatechin, and epicatechin glucuronides.
possibility exists that epicatechin and other phenolic glucuronides may be

cleaved under local conditions of inammation and the free aglycone released
to express cellular eects. Indeed, h-glucuronidases are present in a number of
tissues within the body [204] and may be released by certain cells. For
example, histamine causes rapid exocytosis of h-glucuronidase from lung
macrophages [205] and luteolin monoglucuronide has been shown to be
cleaved to free luteolin by h-glucuronidase released from neutrophils stimulated
with ionomycin [206,207].
In contrast, epicatechin and 3V-O-methyl epicatechin have been shown to
elicit strong cytoprotective eects in broblasts and neurons [54,55] (Fig. 4) and are
associated with both cell types [55]. In addition, 3V-O-methyl-epicatechin was as
eective as epicatechin in protecting neurons against oxidized LDL(Ox-LDL)induced activation of JNK, c-jun, and procaspase-3 [50]. The ability of the Omethylated metabolite to exert protection similar to that of the native epicatechin is
interesting as H-donating potential of the two compounds is very dierent, as the Omethylated compound has a reduced capacity to donate electrons from its B-ring
[54]. This evidence would suggest that avonoids may function to protect cells
against death induced by oxidants by a mechanism independent of their classic
antioxidant properties, a concept that is beginning to receive much attention.
326
Spencer et al.
However, in vivo demethylation may occur intracellularly by the action of

cytochrome P450 enzymes that cleave the O-methylated metabolite to epicatechin.
The in vivo metabolites of eriocitrin (a avonoid glycoside present in
citrus) eriodictyol, homoeriodictyol, methylated eriodictyol, 3,4-dihydroxyhydrocinnamic acid and their conjugates were found in plasma and renal excreted
urine after oral ingestion of eriocitrin [208]. After administration of eriocitrin,
plasma exhibited elevated resistance to lipid peroxidation, therefore suggesting
that metabolites of eriocitrin may be capable of antioxidant action in vivo. 3-O-hD-Glucuronide has also been suggested to act as an eective antioxidant in blood
plasma low-density lipoprotein, because this conjugated metabolite was found to
have a substantial antioxidant eect on copper ioninduced oxidation of human
plasma low-density lipoprotein as well as 1,1-diphenyl-2-picrylhydrazyl radicalscavenging activity [203].
VI.
REDUCTION IN CELL BIOMARKERS OF

OXIDATIVE DAMAGE
As well as studies examining the role avonoids play in reducing cellular injury
and death induced by oxidative stress, there are many studies that have
examined their ability to reduce biomarkers of oxidative damage to cells.
Exposure of cells to an oxidative environment rapidly results in the increase
over time of damaged products of DNA [209], lipid [210], and protein [211], and
the amount and type of oxidative lesions induced are dependent on the oxidizing
species [212]. These oxidative products of cellular biomolecules are often used
as biomarkers of damage to cells by ROS and are also associated with the
viability of the cell in general. Increases in damaged products may in turn lead to
mitochondrial dysfunction and stimulation of signaling cascades, which may
lead to induction of repair of such damage, on the one hand, or to the death by
necrotic or apoptotic mechanisms, on the other hand (Fig. 5). The ultimate fate
of the cell is dependent on the level of damage, but it is generally assumed that a
reduction in these biomarkers of oxidative damage is benecial to the cell.
Flavonoids have been examined for their abilities to attenuate the oxidantinduced increase in damaged products; the section highlights some of the studies
in which they seem to have benecial eects.
A.
Cellular Lipid Oxidation
Flavonoid pretreatment of cells may act to prevent the peroxidation of cellular

membrane lipids. For example, the preincubation of cultured retinal cells with
eriodictyol, luteolin, quercetin, and taxifolin evoked protection against Fe2+
ascorbate-induced lipid peroxidation and increases in intracellular oxidative stress
327
Figure 5 The potential sites of action of flavonoids in the prevention of cell

damage. Oxidative stress results in the oxidation of lipids, proteins, and
deoxyribonucleic acid (DNA), which is in turn sensed by signaling pathways,
such as the MAPK signaling cascade and mitochondria. These events interlink to
cause induction of apoptosis/necrosis in response to the initial oxidative insult.
Flavonoids may act by direct scavenging of the oxidizing species or intermediate,
by direct protein interactions with signaling molecules, or by interaction with
mitochondria. MAPK, mitogen-activated protein kinase.
[213]. Addition of quercetin to culture cell medium enhanced the rate of growth
of baby hamster kidney (BHK-21) cells and also diminished levels of lipid
peroxidation breakdown products [214]. Furthermore, quercetin and three other
avonoids, hesperetin, naringenin, rutin, were found to be eective inhibitors of
lipid oxidation in two dierent in vitro experimental models: (1) Fe(2+)-induced
linoleate peroxidation, by detection of conjugated dienes, and (2) auto-oxidation
328
Spencer et al.
of rat cerebral membranes, by use of thiobarbaturic acid for assay of free

malondialdehyde production [215]. The ability to protect was observed to dier
in each system: rutin > hesperetin > quercetin > naringenin in system (1);
quercetin > rutin > hesperetin > naringenin in system (2), indicating the their
inhibitory nature is related not only to their structural characteristics but also to
their ability to interact with and penetrate the lipid bilayers.
The avonoids, 7-monohydroxyethylrutoside, setin, and naringenin also
signicantly inhibit lipid peroxidation in rat liver microsomes and may be
capable of replacing a-tocopherol as a chain-breaking antioxidant [216].
Catechin-rich extracts isolated from evening primrose seeds have also been
shown to attenuate the oxidation of L-a-phosphatidylcholine liposomes and
oxidation of leukemic L1210 murine cell membranes [217].
The Ginkgo biloba extract, EGb761, and its active constituents have been
shown to be eective in attenuating lipid peroxidation, changes in sulfydryl
group binding sites on the membrane proteins, and apoptosis induced by
hydroxyl radicals in rat cerebellar granule cells [218]. Specically, the total
avonoid component of EGb761 and a mixture of avonoids and terpenes
protected cells from oxidative damage and apoptosis; however, the total terpene
fraction of EGb761 was not eective. The avonoids baicalein and baicalin
eectively inhibit lipid peroxidation of rat brain cortex mitochondria induced by
Fe(2+)-ascorbic acid, AAPH, or reduced nicotinamide-adenine dinucleotide
phosphate (NADPH). Wogonin and wogonoside also show signicant eects
on NADPH-induced lipid peroxidation [69].
One study has reported that tea avonoids and other polyphenols (theaavin digallate, theaavin, epigallocatechin gallate, epigallocatechin, and gallic
acid) can inhibit arterial wall cell-mediated LDL oxidation [219]. Examination
of the mechanism of action indicated that compounds such as theaavin
digallate may act by decreasing superoxide production in macrophages and by
chelating iron ions.
In contrast to many studies that identify benecial eects of dietary
avonoids against cellular lipid oxidation, the action of avonoids on bovine
leukemia virustransformed lamb broblasts (line FLK) and HL-60 cells was
accompanied by lipid peroxidation [90]. Their toxicity was partly prevented by
iron chelator desferrioxamine and antioxidant N,NV-diphenyl-p-phenylene diamine, a result that pointed to the involvement of oxidative stress in their
cytotoxicity. Interestingly, the toxicity of quercetin was partly prevented by
nontoxic concentrations of other avonoids examined, thus suggesting potential
neutralization of quercetin cytotoxicity by intake of avonoid mixtures. In
another study, supplementation of rat hepatocyte cultures with the avonoid
myricetin led to the formation of phenoxyl radical intermediates, as detected in
intact cells by electron paramagnetic resonance (EPR) spectroscopy [220].
These phenoxyl radicals corresponded to one-electron oxidation products of
329
myricetin and have been suggested to be potential mediators of cellular toxicity.

However, myricetin was found to be able to inhibit lipid peroxidation induced
by iron in hepatocyte culture, suggesting that the intermediate generation of
phenoxyl radicals might also contribute to the antioxidant mechanism
of myricetin.
B.
Deoxyribonucleic Acid Oxidation
Exposure of cells to oxidative stress may lead to the generation of (1) oxidized
DNA base lesions, such as 8-OH-guanine, and/or (2) single and double strand
breaks, in both nuclear and mitochondrial DNA [212]. For example, exposure
of human bronchial epithelial cells to H2O2 causes rapid increases in purine
and pyrimidine base oxidation products, DNA strand breakage, and eventual
loss of cellular viability [221,222]. Flavonoids have been implicated as
eective inhibitors of DNA damage induced by a variety of oxidative
[23,25,223] and nitrative stresses [26]. As with other cell studies involving
avonoids, the most extensively studied are the avonols, in particular
quercetin. Quercetin and rutin have been observed to protect Caco-2 and
Hep G2 cells against hydrogen peroxideinduced DNA damage [25] as
measured by the comet assay [25,224226].
In agreement with these studies, quercetin (at concentrations above 10 AM)
and myricetin (>100 AM) decreased oxidant-induced DNA strand breakage and
oxidized pyrimidine bases in human lymphocytes exposed to hydrogen peroxide
[23]. Another avonol, myricetin, has been shown to protect against oxidative
base DNA modication induced by ferric nitrilotriacetate (Fe-NTA) in primary
rat hepatocyte cultures [227]. This reduction in DNA damage may result from an
induction of the DNA repair capacity of hepatocytes after exposure to myricetin,
which was suggested to be due to an activation of DNA excision repair enzymes,
in particular those that remove the more mutagenic purine oxidation products
from DNA of hepatocytes [227,228]. Furthermore, quercetin and myricetin
protect Caco-2 cells against DNA strand breakage induced by hydrogen peroxide,
although rutin and kaempferol were not eective in this cell system [225]. In
agreement with this, quercetin and rutin have also been found to reduce DNA
single strand breaks in Caco-2 cells stimulated by tert-butylhydroperoxide (tertBOOH) and menadione exposure [223,229], and quercetin inhibits both oxidative
DNA damage and NF-nB binding activity in HepG2 cells exposed to hydrogen
peroxide [161].
Data from an in vivo study also suggest that quercetin or its metabolites
may protect against DNA damage. Freshly collected lymphocytes from diabetic
patients treated for 2 weeks on a high avonol diet (mostly quercetin) were more
resistant to hydrogen peroxideinduced DNA damage than those from patients
on the unsupplemented diet [230]. Another in vivo investigation examined the
330
Spencer et al.
eects of green tea catechins on N-nitrosobis(2-oxopropyl)amine- nitrosobis(2oxopropyl)amine-(BOP)-induced oxidative stress in pancreas and liver. Hamsters injected with BOP showed increases in the concentration of lipid peroxides
and the amount of 8-OHdG (8-hydroxy-deoxyguanosine), which were signicantly depressed in those supplemented with a 0.1% solution of green tea
catechins as drinking water [231]. Furthermore, pretreatment of cultured human
lung cells with the green tea avonoid EGCG provided signicant protection
against the induction of DNA strand breakage and genetic damage induced by
tobacco-specic nitrosamines [4-(N-methyl-N-n-trosamino)-1-(3-pyridyl)-1butanone, a metabolite of nicotine] and stimulated human phagocytes [232].
One possible mechanism for the observed protection against DNA base
oxidation by avonoid might reside in their potent transition metal ion
binding abilities [32,33,35], as it is known that the generation of base
modication products by H2O2 is dependent on the presence of DNA bound
transition metal ions [233,234]. However, in contrast to the protective eects
observed for avonoids against oxidative insults, the intracellular steady state
of oxidized DNA bases, formed as a result of normal ongoing damage and
repair in human colon cells, was not altered by exposure to anthocyanins or
anthocyanidins [235]. This nding would suggest that avonoid may act
directly to scavenge the damaging oxidant but does not alter the normal
cellular level of DNA damage.
VII.
CONCLUSION
It is clear that avonoids elicit a number of dierent eects on a wide variety of

dierent cell types ranging from potent protective activities against oxidative
stress in some cells to strong pro-oxidant natures in others. At present there is no
clear pattern that emerges that helps dene a mechanism of protection or prooxidant nature, although it appears that the eects observed may be more
complex than the view that was held until recently that they acted only as potent
classical antioxidants. New data are emerging, including some studies that have
used the biologically relevant metabolite forms of avonoids, that suggest that
their bioactivity is also dependent on their ability to interact with redox-sensitive
intracellular signaling cascades and possibly with mitochondria (Fig. 6). An
overview of all the work since the 1970s would suggest that the mechanism of
action of these dietary agents in vivo is likely to involve both radical scavenging
properties and interactions within signaling cascades. Clearly, to determine the
precise mechanism, more studies using the avonoid metabolite forms in
appropriate cell systems are required. The days of direct addition of dietary
forms have passed, as the in vivo relevance of this type of study will always be
questioned in the light of the absorption and metabolism pathways. How
331
Figure 6 Possible cellular sites of action of flavonoids leading to protection

against cell damage and death by either apoptosis or necrosis: (A) Direct
scavenging of the radical or oxidizing species resulting in a reduction in oxidative
damage biomarkers; (B) specific interactions within intracellular signaling
cascades, for example, the MAPK pathway; reduction in oxidative stressinduced
phosphylation/activation of JNK and c-jun leading to eventual induction of
apoptosis; (C) interactions with mitochondria either at the level of ATP binding
sites or in respiratory chain complexes; (D) cell receptor interactions. MAPK,
mitogen-activated protein kinase; ATP, adenosine triphosphate; JNK, Jun kinase.
avonoids act to prevent disease and improve the quality of older life is still a
relative mystery, and further research is required before detailed information on
mechanism is obtained.
ACKNOWLEDGMENTS
We acknowledge the support of both the Biotechnology and Biological Sciences
Research Council and a European Union Fifth Framework RTD Programme
Grant (grant no. QLK4-1999-01590).
332
Spencer et al.
REFERENCES
1. Colic M, Pavelic K. Molecular mechanisms of anticancer activity of natural
dietetic products. J Mol Med 2000; 78:333336.
2. Eastwood MA. Interaction of dietary antioxidants in vivo: how fruit and vegetables
prevent disease? Q J Med 1999; 92:527530.
3. Galati G, Teng S, Moridani MY, Chan TS, OBrien PJ. Cancer chemoprevention
and apoptosis mechanisms induced by dietary polyphenolics. Drug Metabol Drug
Interact 2000; 17:311349.
4. Middleton EJ, Kandaswami C, Theoharides TC. The eects of plant avonoids on
Pharmacol Rev 2000; 52:673751.
5. Rice-Evans C. Implications of the mechanisms of action of tea polyphenols as
antioxidants in vitro for chemoprevention in humans. Proc Soc Exp Biol Med
1999; 220:262266.
6. Dragsted LO, Strube M, Leth T. Dietary levels of plant phenols and other nonnutritive components: could they prevent cancer? Eur J Cancer Prev 1997;
6:522528.
7. Riemersma RA, Rice-Evans CA, Tyrrell RM, Cliord MN, Lean ME. Tea
avonoids and cardiovascular health. Q J Med 2001; 94:277282.
8. Giugliano D. Dietary antioxidants for cardiovascular prevention. Nutr Metab
Cardiovasc Dis 2000; 10:3844.
9. Wedworth SM, Lynch S. Dietary avonoids in atherosclerosis prevention. Ann
Pharmacother 1995; 29:627628.
10. Manthey JA. Biological properties of avonoids pertaining to inammation.
Microcirculation 2000; 7:S29S34
11. Diplock AT, Charleux JL, Crozier-Willi G, Kok FJ, Rice-Evans C, Roberfroid M,
et al. Functional food science and defence against reactive oxidative species. Br J
Nutr 1998; 80 (suppl 1):S77S112.
12. Harborne JB, Williams CA. Advances in avonoid research since 1992.
13. Packer L, Rimbach G, Virgili F. Antioxidant activity and biologic properties of a
Biol Med 1999; 27:704724.
14. Surh Y. Molecular mechanisms of chemopreventive eects of selected dietary and
medicinal phenolic substances. Mutat Res 1999; 428:305327.
15. Sekher PA, Chan TS, OBrien PJ, Rice-Evans CA. Flavonoid B-ring chemistry and
antioxidant activity: fast reaction kinetics. Biochem Biophys Res Commun 2001;
282:11611168.
16. Rice-Evans C. Plant polyphenols: free radical scavengers or chain-breaking
antioxidants? Biochem Soc Symp 1995; 61:103116.
17. Rice-Evans CA, Miller NJ, Paganga G. Structure-antioxidant activity
relationships of avonoids and phenolic acids. Free Radic Biol Med 1996;
20:933956.
18. Green ES, Cooper CE, Davies MJ, Rice-Evans C. Antioxidant drugs and the
333
inhibition of low-density lipoprotein oxidation. Biochem Soc Trans 1993;

21:362366.
19. Heller FR, Descamps O, Hondekijn JC. LDL oxidation: therapeutic perspectives.
Atherosclerosis 1998; 137 (suppl):S25S31
20. Brown JE, Rice-Evans CA. Luteolin-rich artichoke extract protects low density
lipoprotein from oxidation in vitro. Free Radic Res 1998; 29:247255.
21. Fuhrman B, Aviram M. avonoids protect LDL from oxidation and attenuate
atherosclerosis. Curr Opin Lipidol 2001; 12:4148.
22. Hayek T, Fuhrman B, Vaya J, Rosenblat M, Belinky P, Coleman R, et al. Reduced
progression of atherosclerosis in apolipoprotein E-decient mice following
consumption of red wine, or its polyphenols quercetin or catechin, is associated
with reduced susceptibility of LDL to oxidation and aggregation. Arterioscler
23. Duthie SJ, Collins AR, Duthie GG, Dobson VL. Quercetin and myricetin protect
against hydrogen peroxide-induced DNA damage (strand breaks and oxidised
pyrimidines) in human lymphocytes. Mutat Res 1997; 393:223231.
24. Duthie SJ, Johnson W, Dobson VL. The eect of dietary avonoids on DNA
damage (strand breaks and oxidised pyrimdines) and growth in human cells. Mutat
Res Genet Toxicol Environ Mutagen 1997; 390:141151.
25. Aherne SA, OBrien NM. Protection by the avonoids myricetin, quercetin, and
rutin against hydrogen peroxide-induced DNA damage in Caco-2 and Hep G2
cells. Nutr Cancer 1999; 34:160166.
26. Zhao K, Whiteman M, Spencer JPE, Halliwell B. DNA damage by nitrite and
peroxynitrite: protection by dietary phenols. Methods Enzymol 2001; 335:296307.
27. Haenen GR, Paquay JB, Korthouwer RE, Bast A. Peroxynitrite scavenging by
avonoids. Biochem Biophys Res Commun 1997; 236:591593.
28. Heijnen CG, Haenen GR, van Acker FA, van D, V, Bast A. Flavonoids as
peroxynitrite scavengers: the role of the hydroxyl groups. Toxicol In Vitro 2001;
15:36.
29. Arteel GE, Schroeder P, Sies H. Reactions of peroxynitrite with cocoa procyanidin
oligomers. J Nutr 2000; 130:2100S2104S.
30. Arteel GE, Sies H. Protection against peroxynitrite by cocoa polyphenol
oligomers. FEBS Lett 1999; 462:167170.
31. Kerry N, Rice-Evans C. Inhibition of peroxynitrite-mediated oxidation of
dopamine by flavonoid and phenolic antioxidants and their structural relationships.
J Neurochem 1999; 73:247253.
32. Morel I, Lescoat G, Cillard P, Cillard J. Role of avonoids and iron chelation in
antioxidant action. Methods Enzymol 1994; 234:437443.
33. Brown JE, Khodr H, Hider RC, Rice-Evans CA. Structural dependence of
flavonoid interactions with Cu2+ ions: implications for their antioxidant properties. Biochem J 1998; 330(pt 3):11731178.
34. Sestili P, Guidarelli A, Dacha M, Cantoni O. Quercetin prevents DNA single strand
breakage and cytotoxicity caused by tert-butylhydroperoxide: free radical scavenging versus iron chelating mechanism. Free Radic Biol Med 1998; 25:196200.
35. Cheng IF, Breen K. On the ability of four avonoids, baicilein, luteolin,
334
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
Spencer et al.
naringenin, and quercetin, to suppress the Fenton reaction of the iron-ATP
complex. Biometals 2000; 13:7783.
Tournaire C, Croux S, Maurette MT, Beck I, Hocquaux M, Braun AM. Antioxidant
activity of avonoids: eciency of singlet oxygen (1 delta g) quenching. J
Photochem Photobiol B 1993; 19:205215.
Devasagayam TP, Subramanian M, Singh BB, Ramanathan R, Das NP.
Protection of plasmid pBR322 DNA by avonoids against single-stranded
breaks induced by singlet molecular oxygen. J Photochem Photobiol B 1995;
30:97103.
Scheline RR. Metabolism of oxygen heterocyclic compounds. In: CRC Handbook
of Mammalian Metabolism of Plant Compounds. Boca Raton, FL: CRC Press,
1999:243295.
Rice-Evans C, Spencer JPE, Schroeter H, Rechner AR. Bioavailability of
avonoids and potential bioactive forms in vivo. Drug Metabol Drug Interact
2000; 17:291310.
Spencer JPE, Schroeter H, Rechner A, Rice-Evans C. Bioavailability of avan-3ols and procyanidins: gastrointestinal tract inuences and their relevance to
bioactive forms in vivo. Antiox Redox Signal 2001; 3:10231040.
Spencer JPE, Chowrimootoo G, Choudhury R, Debnam ES, Srai SK, Rice-Evans
C. The small intestine can both absorb and glucuronidate luminal avonoids.
FEBS Lett 1999; 458:224230.
Scalbert A, Williamson G. Dietary intake and bioavailability of polyphenols. J
Nutr 2000; 130:2073S2085S.
Rice-Evans C. Flavonoid antioxidants. Curr Med Chem 2001; 8:797807.
Oliveira NG, Rodrigues AS, Chaveca T, Rueff J. Induction of an adaptive response
to quercetin, mitomycin C and hydrogen peroxide by low doses of quercetin in
V79 Chinese hamster cells. Mutagenesis 1997; 12:457462.
Ishige K, Schubert D, Sagara Y. avonoids protect neuronal cells from oxidative
stress by three distinct mechanisms. Free Radic Biol Med 2001; 30:433446.
Kim H, Kim YS, Kim SY, Suk K. The plant flavonoid wogonin suppresses death
of activated C6 rat glial cells by inhibiting nitric oxide production. Neurosci Lett
2001; 309:6771.
Levites Y, Weinreb O, Maor G, Youdim MB, Mandel S. Green tea polyphenol (-)epigallocatechin-3-gallate prevents N-methyl-4-phenyl-1,2,3,6-tetrahydropyridineinduced dopaminergic neurodegeneration. J Neurochem 2001; 78:10731082.
Chen C, Wei T, Gao Z, Zhao B, Hou J, Xu H, et al. Dierent eects of the
constituents of EGb761 on apoptosis in rat cerebellar granule cells induced by
hydroxyl radicals. Biochem Mol Biol Int 1999; 47:397405.
Yao Z, Drieu K, Papadopoulos V. The Ginkgo biloba extract EGb 761 rescues the
PC12 neuronal cells from beta-amyloid-induced cell death by inhibiting the
formation of beta-amyloid-derived diusible neurotoxic ligands. Brain Res 2001;
889:181190.
Schroeter H, Spencer JPE, Rice-Evans C, Williams RJ. avonoids protect neurons
from oxidized low-density-lipoprotein-induced apoptosis involving c-Jun Nterminal kinase (JNK), c-Jun and caspase-3. Biochem J 2001; 358:547557.

51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
335
Negre-Salvayre A, Salvayre R. Quercetin prevents the cytotoxicity of oxidized

LDL on lymphoid cell lines. Free Radic Biol Med 1992; 12:101106.
Negre-Salvayre A, Alomar Y, Troly M, Salvayre R. Ultraviolet-treated lipoproteins as a model system for the study of the biological eects of lipid peroxides
on cultured cells. III. The protective effect of antioxidants (probucol, catechin,
vitamin E) against the cytotoxicity of oxidized LDL occurs in two different ways.
Biochim Biophys Acta 1991; 1096:291300.
Negre-Salvayre A, Mabile L, Delchambre J, Salvayre R. alpha-Tocopherol,
ascorbic acid, and rutin inhibit synergistically the copper-promoted LDL oxidation
and the cytotoxicity of oxidized LDL to cultured endothelial cells. Biol Trace
Elem Res 1995; 47:8191.
Spencer JPE, Schroeter H, Kuhnle G, Srai SK, Tyrrell RM, Hahn U, Rice-Evans C.
fibroblasts from oxidative-stress-induced cell death involving caspase-3 activation.
Biochem J 2001; 354:493500.
Spencer JPE, Schroeter H, Crossthwaithe AJ, Kuhnle G, Williams RJ, Rice-Evans
hydrogen peroxide-induced cell death in neurons and fibroblasts. Free Radic Biol
Med 2001; 31:11391146.
Ullrich SE. Modulation of immunity by ultraviolet radiation: key effects on
antigen presentation. J Invest Dermatol 1995; 105:30S36S.
Pietta PG, Simonetti P, Gardana C, Brusamolino A, Morazzoni P, Bombardelli
E. Catechin metabolites after intake of green tea infusions. Biofactors 1998;
8:111118.
Kivits GAA, vanderSman FJP, Tijburg LBM. Analysis of catechins from green
and black tea in humans: a specic and sensitive colorimetric assay of total
catechins in biological uids. Int J Food Sci Nutr 1997; 48:387392.
Subirade I, Fernandez Y, Periquet A, Mitjavila S. Catechin protection of 3T3
Swiss fibroblasts in culture under oxidative stress. Biol Trace Elem Res 1995;
47:313319.
Nagata H, Takekoshi S, Takagi T, Honma T, Watanabe K. Antioxidative action of
avonoids, quercetin and catechin, mediated by the activation of glutathione
peroxidase. Tokai J Exp Clin Med 1999; 24:111.
Kaneko T, Matsuo M, Baba N. Inhibition of linoleic acid hydroperoxide-induced
toxicity in cultured human umbilical vein endothelial cells by catechins. Chem
Biol Interact 1998; 114:109119.
Formica JV, Regelson W. Review of the biology of quercetin and related
bioavonoids. Food Chem Toxicol 1995; 33:10611080.
Oyama Y, Noguchi S, Nakata M, Okada Y, Yamazaki Y, Funai M, et al. Exposure of
rat thymocytes to hydrogen peroxide increases annexin V binding to membranes:
inhibitory actions of deferoxamine and quercetin. Eur J Pharmacol 1999; 384:4752.
Plumb GW, Dupont MS, Williamson G. Modulation of AAPH-induced oxidative
stress in cell culture by avonoids. Biochem Soc Trans 1997; 25:S560
Nakayama T, Yamada M, Osawa T, Kawakishi S. Suppression of active oxygeninduced cytotoxicity by avonoids. Biochem Pharmacol 1993; 45:265267.
336
66.
Spencer et al.
Skaper SD, Fabris M, Ferrari V, Dalle CM, Leon A. Quercetin protects cutaneous
tissue-associated cell types including sensory neurons from oxidative stress
induced by glutathione depletion: cooperative effects of ascorbic acid. Free Radic
Biol Med 1997; 22:669678.
67. Gao D, Tawa R, Masaki H, Okano Y, Sakurai H. Protective effects of baicalein
against cell damage by reactive oxygen species. Chem Pharm Bull (Tokyo) 1998;
46:13831387.
68. Gao Z, Huang K, Xu H. Protective effects of avonoids in the roots of Scutellaria
baicalensis Georgi against hydrogen peroxide-induced oxidative stress in HSSY5Y cells. Pharmacol Res 2001; 43:173178.
69. Gao Z, Huang K, Yang X, Xu H. Free radical scavenging and antioxidant activities
of avonoids extracted from the radix of Scutellaria baicalensis Georgi. Biochim
Biophys Acta 1999; 1472:643650.
70. Villa P, Cova D, De Francesco L, Guaitani A, Palladini G, Perego R. Protective
effect of diosmetin on in vitro cell membrane damage and oxidative stress in
cultured rat hepatocytes. Toxicology 1992; 73:179189.
71. Kostyuk VA, Potapovich AI, Vladykovskaya EN, Korkina LG, Afanasev IB.
Inuence of metal ions on flavonoid protection against asbestos-induced cell
injury. Arch Biochem Biophys 2001; 385:129137.
72. Krutmann J, Grewe M. Involvement of cytokines, DNA damage, and reactive
oxygen intermediates in ultraviolet radiation-induced modulation of intercellular
adhesion molecule-1 expression. J Invest Dermatol 1995; 105:67S70S.
73. Tyrrell RM. Ultraviolet radiation and free radical damage to skin. Biochem Soc
Symp 1995; 61:4753.
74. Garssen J, Vandebriel RJ, van Loveren H. Molecular aspects of UVB-induced
immunosuppression. Arch Toxicol Suppl 1997; 19:97109.
75. Hart PH, Grimbaldeston MA, Finlay-Jones JJ. Sunlight, immunosuppression and
skin cancer: role of histamine and mast cells. Clin Exp Pharmacol Physiol 2001;
28:18.
76. de Gruijl FR, van Kranen HJ, Mullenders LH. UV-induced DNA damage, repair,
mutations and oncogenic pathways in skin cancer. J Photochem Photobiol B 2001;
63:1927.
77. Sarasin A. The molecular pathways of ultraviolet-induced carcinogenesis. Mutat
Res 1999; 428:510.
78. Godar DE. Light and death: photons and apoptosis. J Invest Dermatol Symp Proc
1999; 4:1723.
79. Takashima A, Bergstresser PR. Impact of UVB radiation on the epidermal
cytokine network. Photochem Photobiol 1996; 63:397400.
80. Kondo S, Sauder DN. Keratinocyte-derived cytokines and UVB-induced
immunosuppression. J Dermatol 1995; 22:888893.
81. Takashima A. UVB-dependent modulation of epidermal cytokine network: roles in
UVB-induced depletion of Langerhans cells and dendritic epidermal T cells. J
Dermatol 1995; 22:876887.
82. Katiyar SK, Mukhtar H. Green tea polyphenol (-)-epigallocatechin-3-gallate
treatment to mouse skin prevents UVB-induced inltration of leukocytes,
337
depletion of antigen-presenting cells, and oxidative stress. J Leukoc Biol 2001;

69:719726.
83. Katiyar SK, Elmets CA. Green tea polyphenolic antioxidants and skin photoprotection (review). Int J Oncol 2001; 18:13071313.
84. Katiyar SK, Matsui MS, Elmets CA, Mukhtar H. Polyphenolic antioxidant (-)epigallocatechin-3-gallate from green tea reduces UVB-induced inammatory
responses and inltration of leukocytes in human skin. Photochem Photobiol 1999;
69:148153.
85. Katiyar SK, Afaq F, Perez A, Mukhtar H. Green tea polyphenol (-)epigallocatechin-3-gallate treatment of human skin inhibits ultraviolet radiationinduced oxidative stress. Carcinogenesis 2001; 22:287294.
86. Srivastava PJ, Chandra S, Arif AJ, Singh C, Panday V. Metal chelators/
antioxidants: approaches to protect erythrocytic oxidative stress injury during
Plasmodium berghei infection in Mastomys coucha. Pharmacol Res 1999; 40:239
241.
87. Mitani H, Uchida N, Shima A. Induction of cyclobutane pyrimidine dimer
photolyase in cultured fish cells by UVA and blue light. Photochem Photobiol
1996; 64:943948.
88. Skibola CF, Smith MT. Potential health impacts of excessive flavonoid intake.
89. Sergediene E, Jonsson K, Szymusiak H, Tyrakowska B, Rietjens IM, Cenas N.
Prooxidant toxicity of polyphenolic antioxidants to HL-60 cells: description of
quantitative structure-activity relationships. FEBS Lett 1999; 462:392396.
90. Dickancaite E, Nemeikaite A, Kalvelyte A, Cenas N. Prooxidant character of
flavonoid cytotoxicity: structure-activity relationships. Biochem Mol Biol Int
1998; 45:923930.
91. Agullo G, Gamet-Payrastre L, Fernandez Y, Anciaux N, Demigne C, Remesy
C. Comparative effects of avonoids on the growth, viability and metabolism
of a colonic adenocarcinoma cell line (HT29 cells). Cancer Lett 1996;
105:6170.
92. Silva ID, Gaspar J, daCosta GG, Rodrigues AS, Laires A, Rueff J. Chemical
features of avonols affecting their genotoxicity. Potential implications in their use
as therapeutical agents. Chem Biol Interact 2000; 124:2951.
93. Halliwell B, Clement MV, Ramalingam J, Long LH. Hydrogen peroxide: ubiquitous in cell culture and in vivo? IUBMB Life 2000; 50:251257.
94. Miura YH, Tomita I, Watanabe T, Hirayama T, Fukui S. Active oxygens
generation by avonoids. Biol Pharm Bull 1998; 21:9396.
95. Sakagami H, Jiang Y, Kusama K, Atsumi T, Ueha T, Toguchi M, et al. Induction of apoptosis by avones, avonols (3-hydroxyavones) and isoprenoidsubstituted avonoids in human oral tumor cell lines. Anticancer Res 2000;
20:271277.
96. Weber LP, Kiparissis Y, Hwang GS, Niimi AJ, Janz DM, Metcalfe CD. Increased
cellular apoptosis after chronic aqueous exposure to nonylphenol and quercetin in
adult medaka (Oryzias latipes). Comp Biochem Physiol C Toxicol Pharmacol
2002; 131:5159.
338
97.
Spencer et al.
Pawlikowska-Pawlega B, Jakubowicz-Gil J, Rzymowska J, Gawron A. The effect

of quercetin on apoptosis and necrosis induction in human colon adenocarcinoma
cell line LS180. Folia Histochem Cytobiol 2002; 39:217218.
98. Rong Y, Yang EB, Zhang K, Mack P. Quercetin-induced apoptosis in the monoblastoid cell line U937 in vitro and the regulation of heat shock proteins expression. Anticancer Res 2000; 20:43394345.
99. Wei YQ, Zhao X, Kariya Y, Fukata H, Teshigawara K, Uchida A. Induction of
apoptosis by quercetin: involvement of heat shock protein. Cancer Res 1994;
54:49524957.
100. Csokay B, Prajda N, Weber G, Olah E. Molecular mechanisms in the
antiproliferative action of quercetin. Life Sci 1997; 60:21572163.
101. OBrien NM, Woods JA, Aherne SA, OCallaghan YC. Cytotoxicity, genotoxicity
and oxidative reactions in cell-culture models: modulatory effects of phytochemicals. Biochem Soc Trans 2000; 28:2226.
102. Rzymowska J, Gawron A, Pawlikowska-Pawlega B, Jakubowicz-Gil J, Wojcierowski J. The effect of quercetin on induction of apoptosis. Folia Histochem
Cytobiol 1999; 37:125126.
103. Iwashita K, Kobori M, Yamaki K, Tsushida T. Flavonoids inhibit cell growth and
induce apoptosis in B16 melanoma 4A5 cells. Biosci Biotech Biochem 2000;
64:18131820.
104. Wang IK, Lin-Shiau SY, Lin JK. Induction of apoptosis by apigenin and related
avonoids through cytochrome c release and activation of caspase-9 and caspase-3
in leukaemia HL-60 cells. Eur J Cancer 1999; 35:15171525.
105. Yamashita N, Kawanishi S. Distinct mechanisms of DNA damage in apoptosis
induced by quercetin and luteolin. Free Radic Res 2000; 33:623633.
106. Sahu SC, Gray GC. Lipid peroxidation and DNA damage induced by morin
and naringenin in isolated rat liver nuclei. Food Chem Toxicol 1997;
35:443447.
107. Sahu SC, Gray GC. Kaempferol-induced nuclear DNA damage and lipid
peroxidation. Cancer Lett 1994; 85:159164.
108. Knowles LM, Zigrossi DA, Tauber RA, Hightower C, Milner JA. avonoids
suppress androgen-independent human prostate tumor proliferation. Nutr Cancer
2000; 38:116122.
109. Kuo SM. Antiproliferative potency of structurally distinct dietary avonoids on
human colon cancer cells. Cancer Lett 1996; 110:4148.
110. Bhatia N, Agarwal C, Agarwal R. Differential responses of skin cancerchemopreventive agents silibinin, quercetin, and epigallocatechin 3-gallate on
mitogenic signaling and cell cycle regulators in human epidermoid carcinoma
A431 cells. Nutr Cancer 2002; 39:292299.
111. Choi JA, Kim JY, Lee JY, Kang CM, Kwon HJ, Yoo YD, et al. Induction of cell
cycle arrest and apoptosis in human breast cancer cells by quercetin. Int J Oncol
2001; 19:837844.
112. Asea A, Ara G, Teicher BA, Stevenson MA, Calderwood SK. Effects of the
flavonoid drug quercetin on the response of human prostate tumours to
hyperthermia in vitro and in vivo. Int J Hyperthermia 2001; 17:347356.

113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
339
Russo M, Palumbo R, Tedesco I, Mazzarella G, Russo P, Iacomino G, et al.

Quercetin and anti-CD95(Fas/Apo1) enhance apoptosis in HPB-ALL cell line.
FEBS Lett 1999; 462:322328.
Richter M, Ebermann R, Marian B. Quercetin-induced apoptosis in colorectal
tumor cells: possible role of EGF receptor signaling. Nutr Cancer 1999; 34:8899.
Xiao D, Zhu SP, Gu ZL. Quercetin induced apoptosis in human leukemia HL-60
cells. Zhongguo Yao Li Xue Bao 1997; 18:280283.
Chen S, Ruan Q, Bedner E, Deptala A, Wang X, Hsieh TC, et al. Effects of the
flavonoid baicalin and its metabolite baicalein on androgen receptor expression,
cell cycle progression and apoptosis of prostate cancer cell lines. Cell Prolif 2001;
34:293304.
Harris SR, Panaro NJ, Thorgeirsson UP. Oxidative stress contributes to the antiproliferative effects of avone acetic acid on endothelial cells. Anticancer Res
2000; 20:22492254.
Huynh HT, Teel RW. Selective induction of apoptosis in human mammary cancer
cells (MCF-7) by pycnogenol. Anticancer Res 2000; 20:24172420.
Kampa M, Hatzoglou A, Notas G, Damianaki A, Bakogeorgou E, Gemetzi C, et al.
Wine antioxidant polyphenols inhibit the proliferation of human prostate cancer
cell lines. Nutr Cancer 2000; 37:223233.
Kim SY, Gao JJ, Kang HK. Two avonoids from the leaves of Morus alba induce
differentiation of the human promyelocytic leukemia (HL-60) cell line. Biol
Pharm Bull 2000; 23:451455.
Lu HQ, Niggemann B, Zanker KS. Suppression of the proliferation and migration
of oncogenic ras-dependent cell lines, cultured in a three-dimensional collagen
matrix, by flavonoid-structured molecules. J Cancer Res Clin Oncol 1996; 122:
335342.
Bhat KPL, Kosmeder JW, Pezzuto JM. Biological effects of resveratrol. Antioxid
Redox Signal 2001; 3:10411064.
Gusman J, Malonne H, Atassi G. A reappraisal of the potential chemopreventive and chemotherapeutic properties of resveratrol. Carcinogenesis 2001;
22:11111117.
She QB, Bode AM, Ma WY, Chen NY, Dong Z. Resveratrol-induced activation of
p53 and apoptosis is mediated by extracellular-signal-regulated protein kinases and
p38 kinase. Cancer Res 2001; 61:16041610.
Park JW, Choi YJ, Suh SI, Baek WK, Suh MH, Jin IN, et al. Bcl-2 overexpression
attenuates resveratrol-induced apoptosis in U937 cells by inhibition of caspase-3
activity. Carcinogenesis 2001; 22:16331639.
Mahyar-Roemer M, Katsen A, Mestres P, Roemer K. Resveratrol induces colon
tumor cell apoptosis independently of p53 and precede by epithelial differentiation, mitochondrial proliferation and membrane potential collapse. Int J
Cancer 2001; 94:615622.
Dorrie J, Gerauer H, Wachter Y, Zunino SJ. Resveratrol induces extensive
apoptosis by depolarizing mitochondrial membranes and activating caspase-9 in
acute lymphoblastic leukemia cells. Cancer Res 2001; 61:47314739.
Ahmad N, Adhami VM, Afaq F, Feyes DK, Mukhtar H. Resveratrol causes WAF-
340
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
Spencer et al.
1/p21-mediated G(1)-phase arrest of cell cycle and induction of apoptosis in
human epidermoid carcinoma A431 cells. Clin Cancer Res 2001; 7:14661473.
Shih A, Davis FB, Lin HY, Davis PJ. Resveratrol induces apoptosis in thyroid
cancer cell lines via a MAPK-and p53-dependent mechanism. J Clin Endocrinol
Metab 2002; 87:12231232.
She QB, Huang C, Zhang Y, Dong Z. Involvement of c-jun NH(2)-terminal
kinases in resveratrol-induced activation of p53 and apoptosis. Mol Carcinog
2002; 33:244250.
Lee SH, Ryu SY, Kim HB, Kim MY, Chun YJ. Induction of apoptosis by 3,4Vdimethoxy-5-hydroxystilbene in human promyeloid leukemic HL-60 cells. Planta
Med 2002; 68:123127.
Joe AK, Liu H, Suzui M, Vural ME, Xiao D, Weinstein IB. Resveratrol induces
growth inhibition, S-phase arrest, apoptosis, and changes in biomarker expression
in several human cancer cell lines. Clin Cancer Res 2002; 8:893903.
Waffo-Teguo P, Hawthorne ME, Cuendet M, Merillon JM, Kinghorn AD, Pezzuto
JM, et al. Potential cancer-chemopreventive activities of wine stilbenoids and
avans extracted from grape (Vitis vinifera) cell cultures. Nutr Cancer 2002;
40:173179.
Wolter F, Akoglu B, Clausnitzer A, Stein J. Downregulation of the cyclin D1/Cdk4
complex occurs during resveratrol-induced cell cycle arrest in colon cancer cell
lines. J Nutr 2001; 131:21972203.
Sgambato A, Ardito R, Faraglia B, Boninsegna A, Wolf FI, Cittadini A.
Resveratrol, a natural phenolic compound, inhibits cell proliferation and prevents
oxidative DNA damage. Mutat Res 2001; 496:171180.
Schneider Y, Vincent F, Duranton B, Badolo L, Gosse F, Bergmann C, et al. Antiproliferative effect of resveratrol, a natural component of grapes and wine, on
human colonic cancer cells. Cancer Lett 2000; 158:8591.
Holian O, Wahid S, Atten MJ, Attar BM. Inhibition of gastric cancer cell
proliferation by resveratrol: role of nitric oxide. Am J Physiol Gastrointest Liver
Physiol 2002; 282:G809G816
Brakenhielm E, Cao R, Cao Y. Suppression of angiogenesis, tumor growth, and
wound healing by resveratrol, a natural compound in red wine and grapes. FASEB
J 2001; 15:17981800.
Bove K, Lincoln DW, Tsan MF. Effect of resveratrol on growth of 4T1 breast
cancer cells in vitro and in vivo. Biochem Biophys Res Commun 2002; 291:1001
1005.
de Santi C, Pietrabissa A, Mosca F, Pacifici GM. Glucuronidation of resveratrol, a
natural product present in grape and wine, in the human liver. Xenobiotica 2000;
30:10471054.
Kuhnle G, Spencer JPE, Chowrimootoo G, Schroeter H, Debnam ES, Srai SK, et
al. Resveratrol is absorbed in the small intestine as resveratrol glucuronide.
Biochem Biophys Res Commun 2000; 272:212217.
de Santi C, Pietrabissa A, Spisni R, Mosca F, Pacifici GM. Sulphation of
resveratrol, a natural compound present in wine, and its inhibition by natural
avonoids. Xenobiotica 2000; 30:857866.

143.
144.
145.
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
341
Moridani MY, Scobie H, Jamshidzadeh A, Salehi P, OBrien PJ. Caeic acid,

chlorogenic acid, and dihydrocaffeic acid metabolism: glutathione conjugate
formation. Drug Metab Dispos 2001; 29:14321439.
Moridani MY, Scobie H, Salehi P, OBrien PJ. Catechin metabolism: glutathione
conjugate formation catalyzed by tyrosinase, peroxidase, and cytochrome p450.
Chem Res Toxicol 2001; 14:841848.
Awad HM, Boersma MG, Vervoort J, Rietjens IM. Peroxidase-catalyzed formation
of quercetin quinone methide-glutathione adducts. Arch Biochem Biophys 2000;
378:224233.
Awad HM, Boersma MG, Boeren S, van Bladeren PJ, Vervoort J, Rietjens IM. The
regioselectivity of glutathione adduct formation with flavonoid quinone/quinone
methides is pH-dependent. Chem Res Toxicol 2002; 15:343351.
Rietjens IM, Awad HM, Boersma MG, van Iersel ML, Vervoort J, van Bladeren
PJ. Structure activity relationships for the chemical behaviour and toxicity of
electrophilic quinones/quinone methides. Adv Exp Med Biol 2002; 500:1121.
Awad HM, Boersma MG, Boeren S, van Bladeren PJ, Vervoort J, Rietjens IM.
Structure-activity study on the quinone/quinone methide chemistry of avonoids.
Chem Res Toxicol 2001; 14:398408.
Shen XM, Zhang F, Dryhurst G. Oxidation of dopamine in the presence of cysteine:
characterization of new toxic products. Chem Res Toxicol 1997; 10:147155.
Shen XM, Dryhurst G. Further insights into the inuence of L-cysteine on the
oxidation chemistry of dopamine: reaction pathways of potential relevance to
Parkinsons disease. Chem Res Toxicol 1996; 9:751763.
Spencer JPE, Jenner P, Daniel SE, Lees AJ, Marsden DC, Halliwell B. Conjugates
of catecholamines with cysteine and GSH in Parkinsons disease: possible
mechanisms of formation involving reactive oxygen species. J Neurochem 1998;
71:21122122.
Fremont L, Belguendouz L, Delpal S. Antioxidant activity of resveratrol and
alcohol-free wine polyphenols related to LDL oxidation and polyunsaturated fatty
acids. Life Sci 1999; 64:25112521.
Li H, Dryhurst G. Irreversible inhibition of mitochondrial complex I by 7-(2aminoethyl)-3,4-dihydro-5-hydroxy-2H-1,4-benzothiazine-3-carboxylic acid
(DHBT-1): a putative nigral endotoxin of relevance to Parkinsons disease.
J Neurochem 1997; 69:15301541.
Li H, Shen XM, Dryhurst G. Brain mitochondria catalyze the oxidation of 7-(2aminoethyl)-3,4-dihydro-5-hydroxy-2H-1,4-benzothiazine-3-carboxylic acid
(DHBT-1) to intermediates that irreversibly inhibit complex I and scavenge
glutathione, potential relevance to the pathogenesis of Parkinsons disease.
J Neurochem 1998; 71:20492062.
Zhang F, Dryhurst G. Effects of I-cysteine on the oxidation chemistry of
dopaminenew reaction pathways of potential relevance to idiopathic parkinsons
disease. J Med Chem 1994; 37:10841098.
Panes J, Gerritsen ME, Anderson DC, Miyasaka M, Granger DN. Apigenin
inhibits tumor necrosis factor-induced intercellular adhesion molecule-1 upregulation in vivo. Microcirculation 1996; 3:279286.
342
Spencer et al.
157.
Gerritsen ME, Carley WW, Ranges GE, Shen CP, Phan SA, Ligon GF, Percy CA.
Flavonoids inhibit cytokine-induced endothelial cell adhesion protein gene
expression. Am J Pathol 1995; 147:278292.
Soriani M, Rice-Evans C, Tyrrell RM. Modulation of the UVA activation of haem
oxygenase, collagenase and cyclooxygenase gene expression by epigallocatechin
in human skin cells. FEBS Lett 1998; 439:253257.
Medina JH, Viola H, Wolfman C, Marder M, Wasowski C, Calvo D, Paladini AC.
Overviewavonoids: a new family of benzodiazepine receptor ligands. Neurochem Res 1997; 22:419425.
Gamet-Payrastre L, Manenti S, Gratacap MP, Tulliez J, Chap H, Payrastre B.
Flavonoids and the inhibition of PKC and PI 3-kinase. Gen Pharmacol 1999;
32:279286.
Musonda CA, Chipman JK. Quercetin inhibits hydrogen peroxide (H2O2)-induced
NF-kappaB DNA binding activity and DNA damage in HepG2 cells. Carcinogenesis 1998; 19:15831589.
Gosslau A, Rensing L. Induction of Hsp68 by oxidative stress involves the
lipoxygenase pathway in C6 rat glioma cells. Brain Res 2000; 864:114
123.
Kobuchi H, Roy S, Sen CK, Nguyen HG, Packer L. Quercetin inhibits inducible
Physiol 1999; 277:C403C411
Kong AN, Yu R, Chen C, Mandlekar S, Primiano T. Signal transduction events
response and induction of apoptosis. Arch Pharm Res 2000; 23:116.
Nakahata N, Kyo R, Kutsuwa M, Ohizumi Y. Inhibition of mitogen-activated
protein kinase cascade by baicalein, a flavonoid of natural origin. Nippon
Yakurigaku Zasshi 1999; 114 (Suppl 1):215P219P.
Lo YY, Wong JM, Cruz TF. Reactive oxygen species mediate cytokine activation
of c-Jun NH2-terminal kinases. J Biol Chem 1996; 271:1570315707.
Uchida K, Shiraishi M, Naito Y, Torii Y, Nakamura Y, Osawa T. Activation of
stress signaling pathways by the end product of lipid peroxidation: 4-hydroxy-2nonenal is a potential inducer of intracellular peroxide production. J Biol Chem
1999; 274:22342242.
Mizuno M, Droy-Lefaix MT, Packer L. Gingko biloba extract EGb 761 is a
suppresser of AP-1 transcription factor stimulated by phorbol 12-myristate 13acetate. Biochem Mol Biol Int 1996; 39:395401.
ATPase: properties of puried and membrane-bound enzyme: effects of
avonoids. Biochimie 1975; 57:959967.
Barzilai A, Rahamimoff H. Inhibition of Ca2+-transport ATPase from synaptosomal vesicles by avonoids. Biochim Biophys Acta 1983; 730:245254.
Revuelta MP, Cantabrana B, Hidalgo A. Mechanisms involved in kaempferolinduced relaxation in rat uterine smooth muscle. Life Sci 2000; 67:251
259.
Revuelta MP, Hidalgo A, Cantabrana B. Involvement of cAMP and beta-
158.
159.
160.
161.
162.
163.
164.
165.
166.
167.
168.
169.
170.
171.
172.
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
183.
184.
185.
186.
343
adrenoceptors in the relaxing effect elicited by avonoids on rat uterine smooth

muscle. J Auton Pharmacol 1999; 19:353358.
Revuelta MP, Cantabrana B, Hidalgo A. Depolarization-dependent eect of
avonoids in rat uterine smooth muscle contraction elicited by CaCl2. Gen
Pharmacol 1997; 29:847857.
Rosenblat M, Belinky P, Vaya J, Levy R, Hayek T, Coleman R, et al. Macrophage
enrichment with the isoavan glabridin inhibits NADPH oxidase-induced cellmediated oxidation of low density lipoprotein. A possible role for protein kinase C.
J Biol Chem 1999; 274:1379013799.
Ursini F, Maiorino M, Morazzoni P, Roveri A, Pifferi G. A novel antioxidant
flavonoid (IdB 1031) affecting molecular mechanisms of cellular activation. Free
Radic Biol Med 1994; 16:547553.
Kantengwa S, Polla BS. avonoids, but not protein kinase C inhibitors, prevent
stress protein synthesis during erythrophagocytosis. Biochem Biophys Res
Commun 1991; 180:308314.
Fischer PM, Lane DP. Inhibitors of cyclin-dependent kinases as anti-cancer
therapeutics. Curr Med Chem 2000; 7:12131245.
Beppu M, Watanabe T, Yokota A, Ohmori S, Kikugawa K. Water-soluble
antioxidants inhibit macrophage recognition of oxidized erythrocytes. Biol Pharm
Bull 2001; 24:575578.
Crouvezier S, Powell B, Keir D, Yaqoob P. The effects of phenolic components of
tea on the production of pro- and anti-inammatory cytokines by human
leukocytes in vitro. Cytokine 2001; 13:280286.
Lin JK, Chen PC, Ho CT, Lin-Shiau SY. Inhibition of xanthine oxidase and
suppression of intracellular reactive oxygen species in HL-60 cells by theaavin3,3V-digallate, (-)-epigallocatechin-3-gallate, and propyl gallate. J Agric Food
Chem 2000; 48:27362743.
de Groot H, Rauen U. Tissue injury by reactive oxygen species and the protective
effects of avonoids. Fundam Clin Pharmacol 1998; 12:249255.
Dehmlow C, Erhard J, de Groot H. Inhibition of Kupffer cell functions as an
explanation for the hepatoprotective properties of silibinin. Hepatology 1996;
23:749754.
Dehmlow C, Murawski N, de Groot H. Scavenging of reactive oxygen species and
inhibition of arachidonic acid metabolism by silibinin in human cells. Life Sci
1996; 58:15911600.
Deak G, Muzes G, Lang I, Nekam K, Gonzalez-Cabello R, Gergely P, et al. Effects
of two bioavonoids on certain cellular immune reactions in vitro. Acta Physiol
Hung 1990; 76:113121.
Pignol B, Etienne A, Crastes dP, Deby C, Mencia-Huerta JM, Braquet P. Role of
avonoids in the oxygen-free radical modulation of the immune response. Prog
Clin Biol Res 1988; 280:173182.
Moroney MA, Alcaraz MJ, Forder RA, Carey F, Hoult JR. Selectivity of
neutrophil 5-lipoxygenase and cyclo-oxygenase inhibition by an anti-inammatory
avonoid glycoside and related aglycone avonoids. J Pharm Pharmacol 1988;
40:787792.
344
Spencer et al.
187.
T HB, Ip VCT Van Dijk H, Labadie RP. How avonoids inhibit the generation of
luminol-dependent chemiluminescence by activated human neutrophils. Chem
Biol Interact 1990; 73:323335.
Ogasawara H, Fujitani T, Drzewiecki G, Middleton EJ. The role of hydrogen
peroxide in basophil histamine release and the effect of selected avonoids. J
Allergy Clin Immunol 1986; 78:321328.
Middleton EJ. Effect of avonoids on basophil histamine release and other
secretory systems. Prog Clin Biol Res 1986; 213:493506.
Pagonis C, Tauber Al, Pavlotsky N, Simons ER. Flavonoid impairment of
neutrophil response. Biochem Pharmacol 1986; 35:237245.
Busse WW, Kopp DE, Middleton EJ. Flavonoid modulation of human neutrophil
function. J Allergy Clin Immunol 1984; 73:801809.
Hodnick WF, Duval DL, Pardini RS. Inhibition of mitochondrial respiration and
cyanide-stimulated generation of reactive oxygen species by selected avonoids.
Biochem Pharmacol 1994; 47:573580.
Hodnick WF, Milosavljevic EB, Nelson JH, Pardini RS. Electrochemistry of
1988; 37:26072611.
Hodnick WF, Kung FS, Roettger WJ, Bohmont CW, Pardini RS. Inhibition of
mitochondrial respiration and production of toxic oxygen radicals by avonoids: a
structure-activity study. Biochem Pharmacol 35:23452357.
Hodnick WF, Roettger WJ, Kung FS, Bohmont CW, Pardini RS. Inhibition of
mitochondrial respiration and production of superoxide and hydrogen peroxide
by avonoids: a structure activity study. Prog Clin Biol Res 1986; 213:249
252.
Mittra B, Saha A, Chowdhury AR, Pal C, Mandal S, Mukhopadhyay S, et al.
Luteolin, an abundant dietary component is a potent anti-leishmanial agent that
acts by inducing topoisomerase II-mediated kinetoplast DNA cleavage leading to
apoptosis. Mol Med 2000; 6:527541.
Ahmad N, Gali H, Javed S, Agarwal R. Skin cancer chemopreventive effects of a
flavonoid antioxidant silymarin are mediated via impairment of receptor tyrosine
kinase signaling and perturbation in cell cycle progression. Biochem Biophys Res
Commun 1998; 247:294301.
Terao J, Yamaguchi S, Shirai M, Miyoshi M, Moon JH, Oshima S, et al.
Protection by quercetin and quercetin 3-O-beta-D-glucuronide of peroxynitriteinduced antioxidant consumption in human plasma low-density lipoprotein. Free
Radic Res 2001; 35:925931.
Shirai M, Moon JH, Tsushida T, Terao J. Inhibitory effect of a quercetin
metabolite, quercetin 3-O-beta-D-glucuronide, on lipid peroxidation in liposomal
membranes. J Agric Food Chem 2001; 49:56025608.
Yamamoto N, Moon JH, Tsushida T, Nagao A, Terao J. Inhibitory effect of
quercetin metabolites and their related derivatives on copper ion-induced lipid
peroxidation in human low-density lipoprotein. Arch Biochem Biophys 1999;
372:347354.
188.
189.
190.
191.
192.
193.
194.
195.
196.
197.
198.
199.
200.

201.
345
da Silva EL, Piskula MK, Yamamoto N, Moon JH, Terao J. Quercetin metabolites
inhibit copper ion-induced lipid peroxidation in rat plasma. FEBS Lett 1998;
430:405408.
202. Koga T, Meydani M. Effect of plasma metabolites of (+)-catechin and quercetin on
73:941948.
203. Moon J, Tsushida T, Nakahara K, Terao J. Identication of quercetin 3-O-beta-Dglucuronide as an antioxidative metabolite in rat plasma after oral administration
of quercetin. Free Radic Biol Med 2001; 30:12741285.
204. Tukey RH, Strassburg CP. Human UDP-glucuronosyltransferases: metabolism,
expression, and disease. Annu Rev Pharmacol Toxicol 2000; 40:581616.
205. Triggiani M, Gentile M, Secondo A, Granata F, Oriente A, Taglialatela M, et al.
Histamine induces exocytosis and IL-6 production from human lung macrophages
through interaction with H1 receptors. J Immunol 2001; 166:40834091.
206. Shimoi K, Saka N, Nozawa R, Sato M, Amano I, Nakayama T, et al.
Deglucuronidation of a avonoid, luteolin monoglucuronide, during inammation.
Drug Metab Dispos 2001; 29:15211524.
207. Shimoi K, Saka N, Kaji K, Nozawa R, Kinae N. Metabolic fate of luteolin and its
functional activity at focal site. Biofactors 2000; 12:181186.
208. Miyake Y, Shimoi K, Kumazawa S, Yamamoto K, Kinae N, Osawa T. Identication
and antioxidant activity of avonoid metabolites in plasma and urine of eriocitrintreated rats. J Agric Food Chem 2000; 48:32173224.
209. Halliwell B, Aruoma OI. Dna damage by oxygen-derived speciesits mechanism
and measurement in mammalian systems. FEBS Lett 1991; 281:919.
210. Halliwell B, Chirico S. Lipid-peroxidationits mechanism, measurement, and
signicance. Am J Clin Nutr 1993; 57:S715S725
211. Berlett BS, Stadtman ER. Protein oxidation in aging, disease, and oxidative stress.
J Biol Chem 1997; 272:2031320316.
212. Halliwell B, Gutteridge MC. Free Radicals in Biology and Medicine. 3d ed.
Oxford: Clarendon Press, 1999.
213. Areias FM, Rego AC, Oliveira CR, Seabra RM. Antioxidant effect of avonoids
after ascorbate/Fe(2+)-induced oxidative stress in cultured retinal cells. Biochem
Pharmacol 2001; 62:111118.
214. Alliangana DM. Effects of beta-carotene, flavonoid quercitin and quinacrine on
cell proliferation and lipid peroxidation breakdown products in BHK-21 cells. East
Afr Med J 1996; 73:752757.
215. Saija A, Scalese M, Lanza M, Marzullo D, Bonina F, Castelli F. Flavonoids as
antioxidant agents: importance of their interaction with biomembranes. Free Radic
Biol Med 1995; 19:481486.
216. van Acker FA, Schouten O, Haenen GR, van D, V, Bast A. Flavonoids can replace
alpha-tocopherol as an antioxidant. FEBS Lett 2000; 473:145148.
217. Balasinska B, Troszynska A. Total antioxidative activity of evening primrose
(Oenothera paradoxa) cake extract measured in vitro by liposome model and
murine L1210 cells. J Agric Food Chem 1998; 46:35583563.
218. Xin W, Wei T, Chen C, Ni Y, Zhao B, Hou J. Mechanisms of apoptosis in rat
346
219.
220.
221.
222.
223.
224.
225.
226.
227.
228.
229.
230.
231.
232.
233.
Spencer et al.
cerebellar granule cells induced by hydroxyl radicals and the effects of EGb761
and its constituents. Toxicology 2000; 148:103110.
Yoshida H, Ishikawa T, Hosoai H, Suzukawa M, Ayaori M, Hisada T, et al.
Inhibitory effect of tea avonoids on the ability of cells to oxidize low density
lipoprotein. Biochem Pharmacol 1999; 58:16951703.
Morel I, Abalea V, Sergent O, Cillard P, Cillard J. Involvement of phenoxyl
radical intermediates in lipid antioxidant action of myricetin in iron-treated rat
hepatocyte culture. Biochem Pharmacol 1998; 55:13991404.
Spencer JPE, Jenner A, Chimel K, Aruoma OI, Cross CE, Wu R, et al. DNA strand
breakage and base modication induced by hydrogen-peroxide treatment of human
respiratory-tract epithelial-cells. FEBS Lett 1995; 374:233236.
Spencer JPE, Jenner A, Aruoma OI, Cross CE, Wu R, Halliwell B. Oxidative DNA
damage in human respiratory tract epithelial cells: time course in relation to DNA
strand breakage. Biochem Biophys Res Commun 1996; 224:1722.
Aherne SA, OBrien NM. Mechanism of protection by the avonoids, quercetin
and rutin, against tertbutylhydroperoxide-and menadione-induced DNA single
strand breaks in Caco-2 cells. Free Radic Biol Med 2000; 29:507514.
OBrien NM, Woods JA, Aherne SA, OCallaghan YC. Cytotoxicity, genotoxicity
and oxidative reactions in cell-culture models: modulatory effects of phytochemicals. Biochem Soc Trans 2000; 28:2226.
Duthie SJ, Dobson VL. Dietary avonoids protect human colonocyte DNA from
oxidative attack in vitro. Eur J Nutr 1999; 38:2834.
Noroozi M, Angerson WJ, Lean ME. Effects of avonoids and vitamin C on
oxidative DNA damage to human lymphocytes. Am J Clin Nutr 1998;
67:12101218.
Abalea V, Cillard J, Dubos MP, Sergent O, Cillard P, Morel I. Repair of ironinduced DNA oxidation by the flavonoid myricetin in primary rat hepatocyte
cultures. Free Radic Biol Med 1999; 26:14571466.
Morel I, Abalea V, Cillard P, Cillard J. Repair of oxidized DNA by the flavonoid
myricetin. Methods Enzymol 2001; 335:308316.
Sestili P, Guidarelli A, Dacha M, Cantoni O. Quercetin prevents DNA single
strand breakage and cytotoxicity caused by tert-butylhydroperoxide: free radical
scavenging versus iron chelating mechanism. Free Radic Biol Med 1998;
25:196200.
Lean ME, Noroozi M, Kelly I, Burns J, Talwar D, Sattar N, et al. Dietary avonols
protect diabetic human lymphocytes against oxidative damage to DNA. Diabetes
1999; 48:176181.
Takabayashi F, Harada N, Tahara S, Kaneko T, Hara Y. Effect of green tea
catechins on the amount of 8-hydroxydeoxyguanosine (8-OHdG) in pancreatic and
hepatic DNA after a single administration of N-nitrosobis(2-oxopropyl)amine
(BOP). Pancreas 1997; 15:109112.
Weitberg AB, Corvese D. The eect of epigallocatechin galleate and sarcophytol
A on DNA strand breakage induced by tobacco-specic nitrosamines and
stimulated human phagocytes. J Exp Clin Cancer Res 1999; 18:433437.
Dizdaroglu M, Rao G, Halliwell B, Gajewski E. Damage to the DNA bases in
234.
235.
347
mammalian chromatin by hydrogen-peroxide in the presence of ferric and cupric

ions. Arch Biochem Biophys 1991; 285:317324.
Aruoma OI, Halliwell B, Gajewski E, Dizdaroglu M. Damage to the bases in DNA
induced by hydrogen-peroxide and ferric ion chelates. J Biol Chem 1989;
264:2050920512.
Pool-Zobel BL, Bub A, Schroder N, Rechkemmer G. Anthocyanins are potent
antioxidants in model systems but do not reduce endogenous oxidative DNA
damage in human colon cells. Eur J Nutr 1999; 38:227234.
13
Understanding the Bioavailability
of Flavonoids Through Studies
in Caco-2 Cells
Thomas Walle, Richard A. Walgren, U. Kristina Walle,
Alema Galijatovic, and Jaya b. Vaidyanathan
Medical University of South Carolina
I.
INTRODUCTION
A large and steadily growing body of information, most from in vitro studies [1],
supports the benecial role of avonoids in the prevention of human disease. A
prerequisite for this proposition is that the avonoids need to be bioavailable to
exert these benecial eects in vivo. Yet, to date, the question of whether these
dietary compounds are indeed bioavailable has only limited answers [2,3].
Attempts to ascertain the bioavailability of the avonoids have been limited by
the complex behavior of these molecules in in vivo conditions. It is now known
that the bioavailability is restricted by such factors as limited absorption [4],
extensive metabolism [5], degradation due to the intestinal microora [6,7] as
well as intestinal enzymes [8,9], and, for some avonoids, limited chemical
stability [10,11].
To gain a better understanding of the various factors aecting the bioavailability of avonoids, there have been two general approaches: (1) isolated rodent
intestinal preparations and (2) human intestinal cell culture using Caco-2 cells. The
rst of these approaches has many proponents and has yielded much information
about mechanisms of transport as well as metabolism [12 18]. As these experiments are done with rat or other rodent tissues, the relevance of the information
obtained with this system to the human may, however, sometimes be questionable.
349
350
Walle et al.
The second approach, initiated rather recently, has also led to new information
about the mechanisms of both transport and metabolism. This approach has the
attractive advantage of using cultured human cells. Although highly relevant to the
clinical situation, the Caco-2 cell line has disadvantages as well.
The objectives of this chapter are to examine the utility of the Caco-2 cell
monolayer as a model for studies of the oral bioavailability of dietary avonoids
in humans and to outline the information that has been obtained with this system.
II.
Caco-2 CELL MONOLAYER
As a result of the complexity of the in vivo environment, we have adopted a

systematic and reductionist approach to understand the underlying factors aecting the bioavailability and mechanisms governing the absorption of avonoids.
Our studies have employed a model of the principal barrier of intestinal
absorption, the monolayer of columnar epithelial cells or enterocytes lining the
intestinal lumen. Our selected model of the intestinal epithelium is the human
Caco-2 cell line. Caco-2 cells are well-dierentiated human intestinal cells that
have been extensively characterized [19 22]. Caco-2 cells were originally
isolated from a human colonic adenocarcinoma [23]. Despite their origin, these
cells more closely resemble enterocytes than colonocytes both morphologically
and biochemically. Dierentiated Caco-2 monolayers express transport systems
for sugars, amino acids, bile acids, and dipeptides and express many of the brush
border membrane enzymes, including aminopeptidase, alkaline phosphatase,
sucrase, dipeptidyl aminopeptidase, and g-glutamyl transpeptidase.
Examination of colonocyte-and enterocyte-specic markers has demonstrated that fully dierentiated Caco-2 cells more closely resemble enterocytes.
For example, Caco-2 cells produce surfactantlike particles, a secreted membrane
produced in the colon and small intestine. Examination of the composition of the
Caco-2 surfactantlike particle [24] has shown that in early dierentiation these
cells produce particles with a mixed enteric and colonic composition containing
the colonic markers tissue-unspecic alkaline phosphatase (colonic), surfactant
protein A (colonic), and a1-antitrypsin (enteric). Three to six days after reaching
conuence (i.e., with further dierentiation), there are a loss of colonic marker
expression and an increase in enteric marker expression, including a transition
from production of the tissue-unspecic alkaline phosphatase to production of the
intestinal alkaline phosphatase.
In addition to the resemblance of Caco-2 monolayers to enterocytes, Caco2 monolayers are a well-validated and accepted preclinical model for predicting
human intestinal absorption of drug molecules. Apparent permeability values (see
Sec. III. B) obtained in Caco-2 cells have been shown to be predictive of human
in vivo permeability [25,26] and have been strongly correlated with data on the
351
oral absorption in humans [27 29]. In a 2001 study, it was also demonstrated that
the Caco-2 cell is a good model for the human intestinal ATP-binding cassette
(ABC) transporters, which are of great importance in regulation and absorption of
many foreign compounds [30].
III.
TRANSPORT EXPERIMENTS WITH

Caco-2 MONOLAYERS
A.
Cell Culture
The experimental setup is depicted in Figure 1. Caco-2 cells are grown as

monolayers in Eagles minimum essential medium with Earles salts, 10% fetal
calf serum, 1% nonessential amino acids, penicillin (100 U/mL), and streptomycin
(0.1 mg/mL) in a humidied 37 jC incubator with 5% carbon dioxide. Stock
cultures are split 1:12 when just conuent, using trypsin with ethylene diamine
tetra acetic acid (EDTA). For transport studies, the Caco-2 cells are seeded at a
density of 100,000 cells per 1-cm2 tissue culture insert containing 0.4-Am pore size
permeable polycarbonate membranes (Transwells, Corning Costar Corp., Cambridge, MA). The inserts are placed in 12-well tissue culture plates (Fig. 1). The
volume of cell culture medium within the inserts is 0.5 mL (apical or mucosal
side), and the surrounding wells contain 1.5 mL (basolateral or serosal side). The
medium on both sides of the cell layer is changed every 2 days. The integrity of the
cell monolayers is evaluated by measuring the transepithelial electrical resistance
(TEER) values with a volt/ohmmeter (Millicell-ERS, Millipore Corp., Bedford,
MA). Only cell inserts with resistance values exceeding 400 V cm2 are used for
transport experiments. The transport of [14C]mannitol, a marker of paracellular
transport, is also measured in the inserts. The cell monolayers are considered
tight or well formed when the mannitol transport is less than c0.3% of the dose
per hour, corresponding to an apparent permeability coecient ( Papp) of 0.5
106 cm/s (discussed later). The inserts are used for experiments at 18 30 days
after seeding. Before experiments, the cells are washed twice for 30 min with
Figure 1 Caco-2 cell monolayer in tissue culture insert (Transwell).
352
Walle et al.
warm Hanks balanced salt solution (HBSS) buered to pH 7.4 with 25-mM
HEPES. The buer is then replaced with fresh HBSS buer on one side of the cell
layer and avonoid in HBSS buer on the other side. Loading the flavonoid on the
apical side with sampling from the basolateral side mimics absorption, whereas
loading on the basolateral side with sampling from the apical side mimics eux.
[14C]Mannitol is added to the apical side of all inserts, and a basolateral sample is
assayed by liquid scintillation spectrometry at the end of the experiment. In
general the transport is linear for at least 3 h. However, when metabolism occurs,
as is common for flavonoids, it has been our experience that it is best to do the
transport experiments for no longer than 1 h.
B.
Calculations
The apparent permeability coecients ( Papp), expressed in centimeters per

second [27], are calculated as DQ/Dt 1/60 1/A 1/C0, where DQ/Dt is
the permeability rate (micrograms per minute), A is the surface area of the
membrane (square centimeters), and C0 is the initial concentration in the donor
chamber (micrograms per milliliter).
IV.
ABSORPTION OF FLAVONOID GLUCOSIDES
Most dietary avonoids are consumed as glycosides. For example, the most
prevalent dietary avonoid, quercetin, is present in the diet in a number of
Figure 2
Chemical structures of flavonoids.
353
glycosylated forms, the most abundant of which is quercetin 4V-h-O-glucoside

(QG) (Fig. 2). The great diversity of dietary avonoids has posed signicant
challenges to measuring the bioavailability of these compounds. It was originally thought that the glycosides were too polar to pass through the lipid
membranes of the enterocytes. It was thus presumed that the absorption of
avonoids required hydrolysis by the intestinal bacterial microora to release
the aglycone [31]. This was never demonstrated directly but served as a
reasonable hypothesis. In 1995, Hollman and associates put forward the
hypothesis that the glucosides of avonoids, e.g., QG, were absorbed intact
via the sodium-dependent glucose transporter 1 (SGLT1) [32]. Again, this was
never demonstrated but remained as another reasonable hypothesis.
In 1998, Walgren and colleagues demonstrated for the rst time, using the
human Caco-2 cell monolayer as a model, that whereas quercetin was reasonably
well absorbed, QG was not absorbed but rather was eciently eluxed across the
enterocytes [4]. This surprising nding led to further in-depth studies of the
mechanisms involved in avonoid transport using Caco-2 cells and resulted in two
key observations. The rst was that QG was in fact absorbed across the apical
(mucosal) membrane via the SGLT1 transporter and accumulated within the cell.
This was demonstrated by using appropriate tools for SGLT1 inhibition in the
Caco-2 cells and was rigorously conrmed in SGLT1-transfected CHO cells [33].
The second nding was that although QG accumulated within the enterocyte, it
was not further translocated across the basolateral membrane. This condition was
due, at least in part, to the fact that it was transported out of the cell across the
apical membrane by a transporter opposing SGLT1. Through kinetic studies as
well as use of a selective transport inhibitor, MK-571, this was strongly suggested
to be the multidrug resistance associated protein 2 (MRP2) transporter [34]. Only
when using a high concentration of this inhibitor together with high concentrations
of QG was it possible to override this elux and achieve some transcellular
absorption of QG [34]. These scenarios are summarized in Fig. 3.
It would appear that the conclusions based on these Caco-2 cell studies do
indeed apply to the human in vivo condition, as there are still a lack of evidence
for QG absorption, and only evidence for absorption of metabolites of quercetin
[9,35,36]. The Caco-2 cell studies have also pointed to an additional mechanism
governing the absorption of avonoids, i.e., hydrolysis of QG to quercetin [33],
which could then be absorbed (Fig. 3). The importance of this pathway has been
well supported by a study in ileostomy patients in which the administration of
QG resulted in complete hydrolysis to quercetin, presumably within the small
intestine, followed by eective absorption of quercetin [9]. The nature of the
glucosidase(s) involved and their exact location have been addressed [8,37] but
require much further study.
Observations with the isoavonoid genistin (genistein-7-glucoside) support
our ndings with QG. As for QG, this glucoside was not absorbed by the Caco-2
354
Walle et al.
Figure 3 Fluxes and fates of quercetin (Q) and quercetin-4V-glucoside (QG) in the
human enterocyte monolayer model Caco-2.
cells but instead eluxed. The MRP2 transport inhibitor reduced this eux
by 87% [38]. Also, as for QG, genistin was hydrolyzed to its aglycone in
this preparation.
The extent to which the ndings with QG and genistin apply to other
avonoid glucosides or avonoids conjugated with other sugar moieties is not yet
known. Further experiments in this area may benet from using the Caco-2 cell
as a model.
V.
ABSORPTION OF FLAVONOID AGLYCONES
Considering the preceding ndings, it becomes obvious that the absorption of

avonoid aglycones is of great importance. For quercetin, shown in Fig. 3, the
absorption appeared to be transcellular, although elux was greater than absorption [4]. This observation was not pursued further in the Caco-2 cell system,
because of very limited stability of this avonoid in cell culture systems [11]. We
thus selected a chemically stable avonoid for these studies, i.e., chrysin (see Fig.
2), which is a somewhat more lipid-soluble analog of quercetin with only two
hydroxyl groups. For chrysin, we expected to see a high rate of transcellular
absorption. However, this was not the case [5]. Instead, the transcellular
absorption of this avonoid seemed to be limited by ecient metabolism in the
Caco-2 cell monolayer. In further studies with chrysin, we could conclude that
glucuronidation via the uridine diphosphate (UDP)-glucuronosyltransferases
(UGTs) and sulfation via the sulfotransferases (SULTs) were highly ecient in
355
the Caco-2 cells [39] and, presumably, also in the normal human intestine. The
bioavailability of chrysin was thus highly impaired by glucuronidation and
sulfation, and the metabolites formed were eciently eluxed to the apical side,
presumably by the apical membrane transporter MRP2 [5]. This was the rst time
that enterocyte metabolism/transport was demonstrated to be the key determinant
of the bioavailability of avonoid aglycones (Fig. 4).
These studies with chrysin were extended into a clinical investigation in
which seven normal volunteers received a single oral dose of chrysin. Data
obtained from this study yielded a very low estimated bioavailability of 0.003
0.02% [40]. This was in agreement with the results of the Caco-2 cell study and,
extrapolating from the in vitro studies, was likely due to extensive presystemic
glucuronidation and sulfation. Interestingly, when Caco-2 cells were pretreated
with chrysin for several days, using concentrations that may be anticipated in a
clinical setting, one of the metabolic pathways, i.e., glucuronidation, was induced
(Fig. 4) [41]. The results were an upregulation of UGT1A1 in the Caco-2 cells [42]
and considerably more ecient glucuronidation and elimination of chrysin.
Whether this also applies to the in vivo condition in the human intestine remains
to be demonstrated.
Epicatechin is one of the avonoids in green tea and is present in tea leaves
as the aglycone. It has been the subject of additional transport studies in our
laboratory. Much to our surprise, we were unable to detect any apical to
basolateral absorption of this compound across the Caco-2 cell monolayers
[43]. Although this result may in part be due to limited detection sensitivity, it
is doubtful that a rate of transport below our level of detection would be
Figure 4 Fluxes and fates of chrysin (Chr) in the human enterocyte monolayer
model Caco-2. G, glucuronide conjugate; S, sulfate conjugate; UGT, uridine
diphosphate-glucuronosyltransferase; SULT, sulfotransferase.
356
Walle et al.
Figure 5 Fluxes and fates of epicatechin (EC) in the human enterocyte monolayer model Caco-2. S, sulfate conjugate; SULT, sulfotransferase.
signicant. On the other hand, epicatechin was eectively eluxed with a Papp
value of 1.3 106 cm/s (compared to mannitol transport in these cells of 0.3
106 cm/s). This observed unidirectional transport once again suggested the
presence of an elux transporter. Although verapamil, a P-glycoprotein transporter inhibitor, had no eect on this elux, MK-571, an MRP2 transporter
inhibitor, inhibited the elux by 50%. Of further importance was the observation
that MK-571, when added together with epicatechin on the apical side of the
monolayer, produced low but measurable absorption of epicatechin. In addition, as
with chrysin, a sulfate conjugate of epicatechin was formed by the Caco-2 cells.
This conjugate showed highly ecient eux to the apical side by the MRP2
transporter. These scenarios are summarized in Figure 5. These observations are
supported by clinical data, which have demonstrated that epicatechin and other tea
flavonoids have very low oral bioavailability in humans [44]. It remains to be seen
whether the interaction of epicatechin with MRP2 will hold true in vivo studies.
VI.
STUDIES OF ADDITIONAL FLAVONOIDS
A number of additional avonoids have been studied by using the Caco-2 cell
model. These include the simplest form of the avone class of avonoids, the
highly lipophilic unsubstituted avone, which has been shown to diuse readily
across the enterocyte monolayer [45]. It also includes the highly polar hesperidin
glycosides, which are suggested to be transported at a low rate via the paracellular pathway [46]. Another citrus avonoid, 7-geranyloxycoumarin, has
been shown to have a low transcellular permeation rate but was also shown to
357
accumulate in the Caco-2 cells [47]. Nobiletin, a lipophilic polymethoxylated

citrus avonoid, showed high accumulation in Caco-2 cells, in contrast to the
hydrophilic luteolin [48]. Proanthocyanidins and in particular their polymeric
forms have demonstrated low transport rates in Caco-2 cells [49].
VII.
POTENTIAL PROBLEMS WITH Caco-2 CELLS
One potential diculty of work with the Caco-2 cells is related to the tightness of
the monolayer. This is particularly important for compounds with very low net
transport rates. In the original development of this transport model, it was
recommended to measure the TEER value both before and after a transport
experiment. A TEER value z300 V cm2 in general meant a tight monolayer.
However, this may not necessarily be true. As have some other laboratories, we
have selected to test [14C]-mannitol transport as well. The relationship between
Papp for mannitol and the corresponding TEER values over a period of a year is
shown in Figure 6. On the basis of this information, we have in our laboratory
classied a tight monolayer to have a TEER value z 400 V cm2, which produces
a Papp for mannitol < 0.5 106 cm/s.
On rare occasions, we have observed changes in the Caco-2 cell transport
of certain compounds. At these times the TEER values have been normal, but
higher mannitol Papp values have been observed, changing from 0.2 0.5 to
Figure 6 Mannitol flux versus TEER values in confluent Caco-2 cells grown in
Transwells. TEER, transepithelial electrical resistance.
358
Walle et al.
0.7 1.8 106 cm/s. Of great concern was the observation that with these cells
certain avonoids that normally demonstrate a complete lack of absorption (e.g.,
QG and epicatechin) now clearly demonstrated absorption ( Papp values c0.6
106 cm/s). Although this nding may have been due to changes in transporter
expression or cell morphological characteristics, we do not know the exact reason
for this change in the Caco-2 cell transport. However, after use of a new batch of
frozen cells into culture, this problem was overcome.
VIII.
CONCLUSIONS
Transport studies with Caco-2 monolayers in conjunction with molecularly

specic analytical methodologies provided exciting opportunities to examine
systematically the steps involved in the transport of compounds across the
principal transport barrier of the intestine, the enterocyte. The ability to simplify
the number of confounding variables (e.g., the absence of intestinal microora
and pancreatic enzymes and lack of peristaltic mixing) is both a strength and a
limitation of the system. Using this system, we have learned that the degree of
absorption of the avonoids is not due only to glycosylation or nonglycosylation
of the molecule and degree of lipophilicity. Rather, absorption of the avonoids
is much more complicated and dependent on myriad interactions with numerous
transporters and enzymes contained in or on the surface of the membranes and
within the cytosol of enterocytes. To date several proteins have been shown to
play key roles, including (1) SGLT1, (2) MRP2, (3) P-glycoprotein, (4) UGTs,
and (5) SULTs. At present it is not known how the expression levels of these
proteins compare for this cell culture system and in vivo conditions; nor are we
fully cognizant of the potential interindividual variability of these proteins.
Further dening these dierences in expression will become key in understanding how these proteins regulate the bioavailability of the avonoids. At this
time it is also uncertain how other dietary constituents impact the absorption of
avonoids or, possibly of greater importance, how avonoids impact the
absorption of other ingested compounds (i.e., nutrients and pharmaceutical
agents). It is clear, though, that the Caco-2 cell culture system can be a powerful
tool in helping us address these questions and that the resulting information will
increase our understanding of the methods through which cellular transport
processes inuence the disposition and action of natural products and may aid in
the further development of new therapeutic agents.
ACKNOWLEDGMENTS
This research was supported by the National Institutes of Health grant GM55561
and the USDA grant CSREES 00-35200-9071.
359
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
Middleton EJ, Kandaswami C, Theoharides TC. The eects of plant avonoids on

Pharmacol Rev 2000; 52:673 751.
Graefe EU, Derendorf H, Veit M. Pharmacokinetics and bioavailability of the
flavonol quercetin in humans. Int J Clin Pharmacol Ther 1999; 37:219 233.
Rice-Evans C, Spencer JPE, Schroeter H, Rechner AR. Bioavailability of flavonoids
and potential bioactive forms in vivo. Drug Metab Drug Interact 2000; 17:291 310.
Walgren RA, Walle UK, Walle T. Transport of quercetin and its glucosides across
human intestinal epithelial Caco-2 cells. Biochem Pharmacol 1998; 55:1721 1727.
Walle UK, Galijatovic A, Walle T. Transport of the flavonoid chrysin and its
conjugated metabolites by the human intestinal cell line Caco-2. Biochem Pharmacol 1999; 58:431 438.
Bokkenheuser VD, Shackleton CHL, Winter J. Hydrolysis of dietary flavonoid
glycosides by strains of intestinal Bacteroides from humans. Biochem J 1987;
248:953 956.
Walle T, Walle UK, Halushka PV. Carbon dioxide is the major metabolite of
quercetin in humans. J Nutr 2001; 131:2648 2652.
Day AJ, DuPont MS, Ridley S, Rhodes M, Rhodes MJC, Morgan MRA, Williamson G. Deglycosylation of avonoid and isoflavonoid glycosides by human small
intestine and liver h-glucosidase activity. FEBS Lett 1998; 436:71 75.
Walle T, Otake Y, Walle UK, Wilson FA. Quercetin glucosides are completely
hydrolyzed in ileostomy patients before absorption. J Nutr 2000; 130:2658 2661.
Nordstrom CG. Autoxidation of quercetin in aqueous solution: an elucidation of the
autoxidation reaction. Suomen Kemistilehti 1968; B41:351 353.
Boulton DW, Walle UK, Walle T. Fate of the flavonoid quercetin in human cell
lines: chemical instability and metabolism. J Pharm Pharmacol 1999; 51:353 359.
Inoue H, Yokota H, Taniyama H, Kuwahara A, Ogawa H, Kato S, Yuasa A: 1Naphthyl h-D-glucuronide formed intraluminally in rat small intestine mucosa and
absorbed into the colon. Life Sci 1999; 65:1579 1588.
Kuhnle G, Spencer JPE, Schroeter H, Shenoy B, Debnam ES, Srai SKS, RiceEvans C, Hahn U. Epicatechin and catechin are O-methylated and glucuronidated
in the small intestine. Biochem Biophys Res Commun 2000; 277:507 512.
Gee JM, DuPont MS, Day AJ, Plumb GW, Williamson G, Johnson IT. Intestinal
transport of quercetin glycosides in rats involves both deglycosylation and interaction with the hexose transport pathway. J Nutr 2000; 130:2765 2771.
Donovan JL, Crespy V, Manach C, Morand C, Besson C, Scalbert A, Remesy C.
Catechin is metabolized by both the small intestine and liver of rats. J Nutr 2001;
131:1753 1757.
Andlauer W, Stumpf C, Furst P. Intestinal absorption of rutin in free and conjugated
forms. Biochem Pharmacol 2001; 62:369 374.
Spencer JP, Schroeter H, Shenoy B, Srai SK, Debnam ES, Rice-Evans C. Epicatechin is the primary bioavailable form of procyanidin dimers B2 and B5 after transfer
across the small intestine. Biochem Biophys Res Commun 2001; 285:588 593.
Crespy V, Morand C, Besson C, Manach C, Demigne C, Remesy C. Comparison
360
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
Walle et al.
of the intestinal absorption of quercetin, phloretin and their glucosides in rats. J Nutr
2001; 131:2109 2114.
Hidalgo IJ, Raub TJ, Borchardt RT. Characterization of the human colon carcinoma
cell line (Caco-2) as a model system for intestinal epithelial permeability. Gastroenterology 1989; 96:736 749.
Meunier V, Bourrie M, Berger Y, Fabre G. The human intestinal epithelial cell line
Caco-2: pharmacological and pharmacokinetic applications. Cell Biol Toxicol
1995; 11:187 194.
Gan L-SL, Thakker DR. Applications of the Caco-2 model in the design and
development of orally active drugs: elucidation of biochemical and physical barriers
posed by the intestinal epithelium. Adv Drug Delivery Rev 1997; 23:77 98.
Artursson P, Borchardt RT. Intstinal drug absorption and metabolism in cell
cultures: Caco-2 and beyond. Pharm Res 1997; 14:1655 1658.
Fogh J, Fogh JM, Orfeo T. One hundred and twenty-seven cultured human tumor
cell lines producing tumors in nude mice. J Natl Cancer Inst 1977; 59:221 225.
Engle MJ, Goetz GS, Alpers DH. Caco-2 cells express a combination of colonocyte
and enterocyte phenotypes. J Cell Physiol 1998; 174:362 369.
Lennernas H. Human jejunal eective permeability and its correlation with
preclinical absorption models. J Pharm Pharmacol 1997; 49:627 638.
Lennernas H. Human intestinal permeability. J Pharm Sci 1998; 87:403 410.
Artursson P. Epithelial transport of drugs in cell culture. I. A model for studying the
passive diusion of drugs over intestinal absorbtive (Caco-2) cells. J Pharm Sci
1990; 79:476 482.
Artursson P, Karlsson J. Correlation between oral drug absorption in humans and
apparent drug permeability coecients in human intestinal epithelial (Caco-2) cells.
Biochem Biophys Res Commun 1991; 175:880 885.
Yee S. In vitro permeability across Caco-2 cells (colonic) can predict in vivo (small
intestinal) absorption in man fact or myth. Pharm Res 1997; 14:763 766.
Taipalensuu J, Tornblom H, Lindberg G, Einarsson C, Sjoqvist F, Melhus H,
Garberg P, Sjostrom B, Lundgren B, Artursson P. Correlation of gene expression
of ten drug eux proteins of the ATP-binding cassette transporter family in normal
human jejunum and in human intestinal epithelial Caco-2 cell monolayers.
J Pharmacol Exp Ther 2001; 299:164 170.
Griths LA, Barrow A. Metabolism of flavonoid compounds in germ-free rats.
Biochem J 1972; 130:1161 1162.
Hollman PCH, de Vries JHM, van Leeuwen SD, Mengelers MJB, Katan MB.
Absorption of dietary quercetin glycosides and quercetin in healthy ileostomy
volunteers. Am J Clin Nutr 1995; 62:1276 1282.
Walgren RA, Lin J-T, Kinne RK-H, Walle T. Cellular uptake of dietary flavonoid
quercetin 4V-h-glucoside by sodium-dependent glucose transporter SGLT1.
Walgren RA, Karnaky KJ Jr, Lindenmayer GE, Walle T. Eux of dietary flavonoid
quercetin 4V-h-glucoside across human intestinal Caco-2 cell monolayers by apical
multidrug resistance-associated protein-2. J Pharmacol Exp Ther 2000; 294:830 836.
Morand C, Crespy V, Manach C, Besson C, Demigne C, Remesy C. Plasma
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
361
metabolites of quercetin and their antioxidant properties. Am J Physiol 1998;

275:R212 R219.
Graefe EU, Wittig J, Mueller S, Riethling A-K, Uehleke B, Drewelow B, Pforte H,
Jacobasch G, Derendorf H, Veit M. Pharmacokinetics and bioavailability of quercetin glycosides in humans. J Clin Pharmacol 2001; 41:492 499.
Day AJ, Canada FJ, Diaz JC, Kroon PA, Mclauchlan R, Faulds CB, Plumb GW,
Morgan MRA, Williamson G. Dietary flavonoid and isoavone glycosides are
hydrolysed by the lactase site of lactase phlorizin hydrolase. FEBS Lett 2000;
468:166 170.
Walle UK, French KL, Walgren RA, Walle T. Transport of genistein-7-glucoside by
human intestinal Caco-2 cells: potential role for MRP2. Res Comm Mol Pathol
Pharmacol 1999; 103:45 56.
Galijatovic A, Otake Y, Walle UK, Walle T. Extensive metabolism of the flavonoid
chrysin by human Caco-2 and Hep G2 cells. Xenobiotica 1999; 29:1241 1256.
Walle T, Otake Y, Brubaker JA, Walle UK, Halushka PV. Disposition and
metabolism of the flavonoid chrysin in normal volunteers. Br J Clin Pharmacol
2001; 51:143 146.
Galijatovic A, Walle UK, Walle T. Induction of UDP-glucuronosyltransferase by
the flavonoids chrysin and quercetin in Caco-2 cells. Pharm Res 2000; 17:21 26.
Galijatovic A, Otake Y, Walle UK, Walle T. Induction of UDP-glucuronosyltransferase UGT1A1 by the flavonoid chrysin in Caco-2 cells Potential role in
carcinogen bioinactivation. Pharm Res 2001; 18:374 379.
Vaidyanathan JB, Walle T. Transport and metabolism of the tea flavonoid
( )-epicatechin by the human intestinal cell line Caco-2. Pharm Res 2001; 18:
1420 1425.
Warden BA, Smith LS, Beecher GR, Balentine DA, Clevidence BA. Catechins are
bioavailable in men and women drinking black tea throughout the day. J Nutr 2001;
131:1731 1737.
Kuo S-M. Transepithelial transport and accumulation of flavone in human intestinal
Caco-2 cells. Life Sci 1998; 63:2323 2331.
Kim M, Kometani T, Okada S, Shimuzu M. Permeation of hesperidin glycosides
across Caco-2 monolayers via the paracellular pathway. Biosci Biotech Biochem
1999; 63:2183 2188.
Murakami A, Wada K, Ueda N, Sasaki K, Haga M, Kuki W, Takahashi Y, Yonei
H, Koshimizu K, Ohigashi H. In vitro absorption and metabolism of a citrus
chemopreventive agent, auraptene, and its modifying eects on xenobiotic enzyme
activities in mouse livers. Nutr Cancer 2000; 36:191 199.
Murakami A, Kuwahara S, Takahashi Y, Ito C, Furukawa H, Ju-ichi M, Koshimizu
K, Ohigashi H. In vitro absorption and metabolism of nobiletin, a chemopreventive
polymethoxyavonoid in citrus fruits. Biosci Biotech Biochem 2001; 65:194 197.
Deprez S, Mila I, Huneau J-F, Tome D, Scalbert A. Transport of proanthocyanidin
dimer, trimer and polymer across monolayers of human intestinal epithelial Caco-2
cells. Antiox Redox Signal 2001; 3:957 967.
14
Metabolism in the Small Intestine
and Gastrointestinal Tract
Jeremy P. E. Spencer and Catherine A. Rice-Evans
London, England
Surjit Kaila Singh Srai

Royal Free University College Medical School
London, England
I.
INTRODUCTION
Flavonoids have been the center of huge research interest over the last decade
[1 4]. They are the most abundant polyphenols in the human diet and are
divided into six main classes based on the degree of oxidation of the C-ring, the
hydroxylation pattern of the ring structure, and the substitution in the 3-position:
avanols (e.g., epicatechin), avonols (e.g., quercetin), avones (e.g., luteolin),
avanones (e.g., naringenin), isoavones (e.g., genistein) and anthocyanidins
(e.g., cyanidin) [3] (Fig. 1). A large number of in vitro studies have characterized them as powerful antioxidants against both reactive oxygen and reactive
nitrogen species [3,5 15]. Flavonoids with the highest antioxidant potential in
vitro contain a B-ring catechol group that readily donates a hydrogen (electron)
to stabilize a radical species [3]. Until recently, the ability of avonoids to act as
classical H-donating antioxidants was believed to underlie many of their
reported health eects [16 22]. However, the extent of their antioxidant
potential in vivo is dependent on the absorption, metabolism, distribution, and
excretion of these compounds within the body after ingestion and the reducing
properties of the resulting metabolites. An understanding of the processes
involved in the absorption and distribution of polyphenols is essential for
363
364
Spencer et al.
Figure 1 The structures of the five main classes of flavonoids. The major
differences between the individual groups reside in the hydroxylation pattern of
the ring structure, the degree of saturation of the C - ring, and the substitution in the
3 - position.
Metabolism
365
determining their bioactivities in vivo and their signicance. Since the late
1990s, much information has accumulated on the biotransformation of avonoids in the small intestine and gastrointestinal tract [1,2,17,23 28], as well as
the hepatic metabolism [29 32]. This chapter highlights the main sites of
biotransformation of avonoids within the gastrointestinal tract, the major
metabolites generated in the small and large intestine, and the implications of
this modification in determining how avonoids may act in vivo.
II.
MODIFICATION OF FLAVONOIDS IN THE MOUTH

AND STOMACH
Few studies have investigated the ability of saliva and gastric juice to alter the
avonoid structure. Saliva has been found to have little eect on the stability of
green tea catechins [33], however, degalloylation of avanol gallate esters, such
as epigallocatechin gallate, in human saliva has been observed [34]. Incubation
of procyanidin oligomers (dimer-hexamer) in human saliva for up to 30 min
does not result in modication of the compounds [28], suggesting that these
compounds remain intact on entering the stomach. The quercetin rutinoside rutin
has been shown to be hydrolysed by cell-free extracts of human salivary cultures
[35,36]. In contrast, the quercetin-3-rhamnoside quercitrin is not susceptible to
hydrolysis, suggesting that only rutin-glycosidase-elaborating organisms occur
in saliva [36]. Furthermore, oral streptococci isolated from the mouth of normal
individuals have been found to hydrolyse rutin to quercetin [37]. The streptococcal rutinase was found to be cytosolic and constitutive and to have a pH
optimum of 6.5, and its liberation of quercetin from rutin in the mouth was
hypothesized to be involved in intraoral cancer. The interaction of avanols and
procyanidins with salivary proteins has shown that (+) catechin has a higher
anity for proline-rich proteins than (-)-epicatechin, and C(4)-C(8)-linked
procyanidin dimers bind more strongly to them than their C(4)-C(6) counterparts
[38]. Polyphenol-protein binding in the form of adsorption with high-molecularweight salivary proteins, bacterial cells, and mucous materials may be one
explanation for the observed decrease in quercetin mutagenicity after incubation
with saliva [39].
Procyanidin oligomers ranging from a dimer to a decamer (isolated from
Theobroma cacao) have been observed to be unstable under conditions of low pH
similar to that present in the gastric juice of the stomach [40]. On incubation of the
procyanidins with simulated gastric juice, oligomers rapidly decompose essentially to epicatechin monomeric and dimeric units but also to other oligomeric
units [40]. Procyanidins may decompose in mild acidic environments as they are
readily cleaved to form avan-3-ol and quinone methide, which is in equilibrium
with a carbocation in stronger acidic conditions (Fig. 2) [41]. The carbocation is
converted to an anthocyanin on heating in alcoholic solutions, and the quinone
366
Spencer et al.
Figure 2 Decomposition of procyanidins in mild acidic environments (dimer as

example). Procyanidins are readily cleaved in mild acid solutions to form flavan - 3 ol and quinone methide, which is in equilibrium with a carbocation in stronger
acidic conditions. The carbocation is converted to an anthocyanin on heating in
alcoholic solutions. The quinone methide may be captured with a suitable
nucleophile (X).
methide may be captured with a nucleophile. During incubation in acid, other

oligomeric units, such as trimer and tetramer, are also formed and degraded, and
in some instances a time-dependent formation of larger oligomers occurs. Thus,
absorption of flavanols and procyanidins, for example, after consumption of
chocolate or cocoa, is likely to be influenced by preabsorption events in the
gastric lumen within the residence time. However, consideration needs to be
given to the food matrix, which may influence the pH environment of the
procyanidins and their subsequent decomposition. Monomeric flavonoid glycosides have been observed to be stable in the acidic environment of the stomach
and are not observed to undergo nonenzymatic deglycosylation [42]. Interestingly, the flavonoids ponciretin, hesperetin, naringenin, and diosmetin and
phenolic acids generated from flavonoids by human intestinal microflora (discussed later) have been observed to be eective inhibitors of the growth of
Helicobacter pylori, a bacterium known to cause problems in the stomach of
some patients [43].
Metabolism
III.
367
METABOLISM AND CONJUGATION

IN THE SMALL INTESTINE
Generally avonoids are present in plants conjugated to sugars, and therefore it

is these glycosides that are ingested in the diet and enter the gastrointestinal
tract. The exception to this rule are the avan-3-ols, such as the catechins and
procyanidins, which are almost always present in the diet in the nonglycosylated
form [3]. There are many factors that inuence the extent and rate of absorption
of ingested compounds by the small intestine [44]. These include physicochemical factors such as molecular size, lipophilicity, solubility, and pKa and
biological factors including gastric and intestinal transit time, lumen pH,
membrane permeability, and rst-pass metabolism [45,46].
A.
Effects of Intestinal Juice
On transfer from the stomach to the jejunum (the top two-fths of the small
intestine) the pH rises from about 2.0 to 7.0. It is well known that polyphenolic
compounds such as those with catechol structures can oxidize in neutral and
alkaline pH environments. Epigallocatechin gallate (EGCG) has been observed to
oxidize rapidly in authentic intestinal uid (measured pH of 8.5) with the amount
of EGCG decreasing 81.6% in only 5 min [47], whereas a similar incubation in
murine plasma (pH 7.4) resulted in only a 29.3% decrease in amount. However,
oxidation of EGCG resulted in the formation of dimerized products that were
observed to possess greater superoxide radical scavenging activity then EGCG
itself and have powerful iron chelating properties [47]. Another factor to consider
may be the relative abilities of these polyphenols to bind to proteins in the food
matrices in question. In complex food matrices, the pH is likely to be buered for
long periods, and, therefore, oxidation of the avanols may only occur to a
limited extent during intestinal transit. In addition, it has been suggested that
ascorbate signicantly increases the stability of avanols incubated in intestinal
uid [48], and therefore the presence of ascorbate in vivo may stabilize the
polyphenols in the neutral or alkaline environment of the small intestine.
B.
Jejunal and Ileal Conjugation and Metabolism
Many studies have indicated that signicant transfer of ingested avonoids

occurs from the lumen of the small intestine to the mesenteric circulation and
that extensive metabolism and conjugation of the avonoid occur during this
transfer [2,49 56]. Isolated preparations of rat small intestine [57] have been
utilized to study absorption and metabolism in the small intestine and can provide
information on events occurring in both the jejunum and the ileum [50,52,56,58
60]. This model (Fig. 3) allows study of the intestinal transfer of dietary
368
Spencer et al.
Figure 3 Diagrammatic representation of the isolated rat small intestine

perfusion model. Buffers and compound are perfused for up to 90 min at a rate
of 8 mL/min at 37 jC by using a peristaltic pump. Serosal fluid was collected from
the bottom of the paraffin chamber and analyzed by HPLC and mass
spectrometry. HPLC, high - performance liquid chromatography.
polyphenols (and their glycosides) and can be used to assess the rate of
absorption from the lumen. The solute under study appears on the serosal surface
in the same form as if it were transferred to the mesenteric circulation, and
therefore, enterocyte metabolism of flavanols and procyanidins, as well as their
rate of transfer across specic gut regions, may be studied [52]. Tissue viability is
assessed by measurement of glucose transfer [57], and viability is conrmed by
showing that fluid transfer continues at a constant rate for the 90-min collection
period and that glucose concentration in the absorbed fluid is more than double
that initially present in the perfused buer [52].
Absorption studies utilizing this model, with a wide range of avonoids
and their glycosides and hydroxycinnamates, show that there was in almost all
cases extensive metabolism of the polyphenol in the enterocyte during transfer
from the luminal to the serosal side [52] (Table 1, jejunum; Table 2, ileum). The
avonoid glycosides, luteolin-7-glucoside, kaempferol-3-glucoside, and quercetin-3-glucoside, were cleaved by rat jejunal or ileal mucosa, suggesting the
presence of h-glucosidase action before eux into the serosal uid. The major
products transferred across the small intestinal epithelium were glucuronides of
the parent aglycone or of the hydrolysed glycoside, although O-methylated
metabolites were also observed [52]. With an identical model, the major
37.8
31.0
10.3
45.4
23.2
36.2
27.9
8.9
27.2
35.0
13.1
11.6
6.0
Total %
absorbed
(90 min)
2.4
69.1
100.0
0.6
0.0
3.0
0.0
82.2
0.0
0.0
4.6
5.2
1.0
Perfused
compound
O - methylated glucuronide and O - methylated metabolites.
Quercetin
Quercetin-3-glucoside
Rutin
Kaempferol
Kaempferol-3-glucoside
Luteolin
Luteolin-7-glucoside
Hesperetin
Naringenin
Naringenin-7-glucoside
Catechin
Epicatechin
Resveratrol
Perfused
compound
(jejunum)
*
8.0
0.0
*
0.0
*
11.5
0.0
0.0
0.0
*
*
0.0
Aglycone of
perfused
compound
% of total absorbed
2.4
77.1
100.0
0.6
0.0
3.0
11.5
82.2
0.0
0.0
4.6
5.2
1.0
Total %
91.3
22.8
0.0
96.8
97.2
93.7
85.5
17.8
100.0
100.0
65.3
61.7
99.0
Glucuronides
6.3
0.0
0.0
2.5
2.8
3.3
3.0
0.0
0.0
0.0
30.1
33.1
0.0
Othera
% total absorbed
Table 1 Summary of the Absorption and Metabolism of Flavonoids in the Isolated Rat Jejunum Model
97.6
22.8
0.0
99.3
100.0
97.0
88.5
17.8
100.0
100.0
95.4
94.8
99.0
Total % of
metabolite
Metabolism
369
11.9
66.16
55.90
2.3
Hesperetin
Naringenin
Naringenin-7-glucoside
Catechin
Epicatechin
Resveratrol
0.5
98.3
69.3
66.3
17.1
31.8
100.0
32.7
0.0
12.9
0.0
Perfused
compound
O - methylated glucuronide and O - methylated metabolites.
19.3
8.8
0.6
59.1
11.4
25.6
9.2
Total %
absorbed
(90 min)
Quercetin
Rutin
Kaempferol
Kaempferol-3-glucoside
Luteolin
Luteolin-7-glucoside
Perfused
Compound
(ileum)
0.0
*
*
*
36.4
0.0
*
82.6
*
47.8
Aglycone of
perfused
compound
% of total absorbed
0.5
98.3
69.3
66.3
17.1
68.2
100.0
32.7
82.6
12.9
47.8
Total %
99.5
0.7
22.9
27.4
82.9
31.8
NS
67.3
18.4
87.1
52.2
Glucuronides
0.0
0.0
7.8
6.3
0.0
0.0
NS
0.0
0.0
0.0
0.0
Othera
% of total absorbed
Table 2 Summary of the Absorption and Metabolism of Flavonoids in the Isolated Rat Ileum Model
99.5
0.7
30.7
33.7
82.9
31.8
NS
67.3
18.4
87.1
52.2
Total % of
metabolite
370
Spencer et al.
Metabolism
371
metabolite of both naringenin and naringenin-7-glucoside was identied as

naringenin glucuronide, thus supporting the notion that glucuronidation as well
as hydrolysis can occur at the intestinal epithelium [61]. These perfusion studies
are consistent with those of Shimoi and associates [62], who studied transfer of
avonoids from the mucosa to the serosal side of a rat everted intestine model,
nding that both hydrolysis of luteolin-7-glucoside to luteolin and glucuronidation of luteolin occur during transfer across the small intestine. The extent of
glucuronidation in these experiments seemed dependent on the avonoid
structure, in that the avonoids with a substituted hydroxyl group on the Bring (i.e., hesperetin) were less predisposed to glucuronidation, whereas the
avonoids containing a 3V,4V-ortho-dihydroxy (or catechol) B-ring were transferred predominantly as glucuronides [52] (Tables 1 and 2). For example, the
jejunal transfer of quercetin resulted in its being almost totally glucuronidated
(97.6% of total transferred), whereas the absorption of hesperetin resulted in a
much lower level of glucuronidation (17.8%) (Table 1). Monophenolic B-ring
avonoids were also extensively glucuronidated, in particular naringenin, which
was only detected in serosal uid glucuronidated. Similar patterns of metabolism were observed in the ileum. However, in general glucuronidation occurred
to a lesser extent (Table 2), in line with studies that have recorded lower levels
of phase I and II enzymes present in the ileum compared to the jejunum.
Glucuronidation of these avonoids was observed to occur predominantly at the
5- and 7-positions on the A-ring, a process that would be expected to have little
inuence on the resulting antioxidant potential of the metabolite. Indeed, 1999
studies identied the 5-O-h-glucuronide of catechin and epicatechin excreted in
the urine of rats post ingestion, which does not interfere with their antioxidant
properties (as assessed by their ability to scavenge superoxide) [32,63].
Interestingly, two uridine diphosphate (UDP)-glucuronosyltransferase (UGT)
isoforms, UGT1A8 and UGT1A10, of human intestinal mucosa, which are
absent in liver, have been identied by reverse transcriptase-polymerase chain
reaction [49].
Similar studies with resveratrol led to the detection of the glucuronide of
resveratrol almost exclusively in the serosal uid [58] (Table 1), and another
avonoid, diosmetin, was also found to be rapidly glucuronidated in the rat [64].
Whereas the major metabolites observed on the serosal side after perfusion of
the jejunum with catechin or epicatechin were always glucuronidated, there
were also high levels of both O-methylated and O-methylated-glucuronide
forms [28,59] (Tables 1 and 2). 3V-O- and 4V-O-methylated derivatives of the
avanols were detected at high levels in the serosal uid (f30% of total
transferred) and O-methyl and O-methyl-glucuronidated catechins were the
predominant metabolites detected in the serosal uid (f50%), suggesting these
as the most bioavailable forms (Fig. 4). As with the other avonoids tested in
this model, there was a lower level of metabolism occurring in the ileum,
372
Spencer et al.
Figure 4 The structures of the main small intestinal metabolites of epicatechin

produced in the isolated rat small intestine perfusion model. (A) 3V - O - methyl
epicatechin; (B) 4V - O - methyl epicatechin; (C) 3V - O - methyl epicatechin - 5 glucuronide; (D) epicatechin - 5 - glucuronide; (E) epicatechin - 7 - glucuronide.
although the total amounts of both catechin and epicatechin absorbed were much
higher than in the jejunum (Tables 1 and 2). The greater susceptibility to
methylation of avanols than of other avonoids in the jejunum may reside in
the specificity of catechol-O-methyltransferase (COMT) for these compounds
[65]. These data were conrmed in 2001 in a similar model, in which the rat
jejunum and ileum were perfused with catechin [51]. In this study, catechin was
absorbed into intestinal cells and metabolized extensively to a point where no
native catechin could be detected in plasma from the mesenteric vein. Mesenteric
plasma contained glucuronide conjugates of catechin and 3V-O-methyl catechin,
indicating the intestinal origin of these conjugates and the large role the small
intestine plays in the biotransformation of avanols during absorption [51,59].
Although most studies have identied avanol metabolites as the main forms
Metabolism
373
found entering the hepatic portal vein after absorption from the small intestine,
the native avanols are detected in small amounts [59]. Oral administration of the
tea catechins, epicatechin, epigallocatechin, epicatechin gallate, and epigallocatechin gallate, to rats led to the detection of all four avanols in the portal blood
[66], clearly indicating that these avonoids may be absorbed intact to a small
degree. Studies have also shown that tannic acid and catechin may both interact
with the small intestine, but only catechin appears able to traverse the gut [56].
This nding may be due to the binding of tannic acid and catechin by endogenous
proteins in the intestinal lumen, limiting their absorption from the small intestine.
It should also be noted that catechins have bactericidal properties and may play
several roles in the digestive tract that may be linked to their protein-binding
properties. In the small intestine, catechins inhibit a-amylase activity but do not
aect lactic acid bacteria [67]. The inclusion of tea catechins in the diet for
several weeks had the eect of decreasing putrefactive products and at the same
time increasing organic acids by lowering pH [67].
Procyanidins have a high anity for proteins, and their absorption through
the gut barrier is most likely limited to lower oligomeric forms and to the
metabolites formed by the colonic microora. In 2001, perfusion of isolated
small intestine with the procyanidin dimers B2 and B5 extracted from cocoa
indicated that both forms of dimer are transferred to the serosal side of
enterocytes but only to a very small extent (<1% of the total transferred
avanol-like compounds) [68]. Perfusion of dimer mainly resulted in detection
of large amounts of unmetabolized/unconjugated epicatechin monomers on the
serosal side (f95.8%). Low levels of O-methylated dimer were also detected
(f3.2%), but no conjugates and metabolites of epicatechin, indicating that
metabolism of monomer and dimer is limited during dimer cleavage/translocation. Experiments with normal Caco-2 cells and radiolabeled procyanidins
suggested that dimer and trimer were transferred to the same extent as the
epicatechin monomer, whereas oligomers with an average degree of polymerization of 7 were not [29].
In addition to avan-3-ols, O-methylated derivatives of quercetin have
been shown to be generated in the small intestine [50]. Quercetin is taken up
into enterocytes and transferred to the plasma as glucuronidated, O-methylated,
and sulfated derivatives with a large fraction of the absorbed quercetin reexcreted into the lumen as conjugated derivatives both directly and via the
biliary duct [50]. This study indicates that sulfation of quercetin might occur in
intestinal cells; however, because of the lack of precise identication of
sulfated conjugates it remains unclear whether sulfation occurs in the small
intestinal tract. It has been observed that avonoids can inhibit the sulfation of
resveratrol in the duodenum [69] and human cytosolic sulfotransferases show a
high sulfating potential with avonoids and isoavones [70]. However, it is
unclear to what extent these enzymes are present in the small intestine, and
374
Spencer et al.
most studies indicate that the origin of most circulating sulfated avonoids is
the liver [26,51].
Another approach to obtain a better understanding of the bioavailability of
avonoids and their absorption and metabolism in the small intestine has used
cultured human caco-2 cells (see Chap. 13).
C.
Flavonoid Glycoside Processing
As there is little or no cleavage of dietary avonoid glycosides in the mouth and

stomach, the glycosidic forms of the avonoids must enter the small intestine,
where they are presented for absorption [1,29]. Because avonoid glycosides are
generally relatively polar in nature, their passive diusion across the membranes
of small intestinal brush border is limited. Many studies, however, have
suggested that avonoid glycosides are subject to the action of h-glucosidases
before their absorption in the jejunum and ileum [40,52,71 79], and it is
generally believed that the removal of the glycosidic moiety is necessary before
absorption of the avonoid can take place. The cleaved aglycone is thought then
to undergo passive diusion across the intestine brush border; however, the exact
mechanism of uptake of these compounds is still unknown. It has been suggested
that removal of the sugar and subsequent transport by proteins such as lactate
phloridzin hydrolase [75] may occur in the small intestine; although this process
may not occur with all avonoid glycosides. The ability of cell-free extracts from
human small intestine to deglycosylate various avonoid glycosides has been
investigated; it has been observed that quercetin-4V-glucoside, naringenin-7glucoside, apigenin-7-glucoside, genistein-7-glucoside, and daidzein-7-glucoside
are rapidly deglycosylated, whereas quercetin-3,4V-diglucoside, quercetin3-glucoside, kaempferol-3-glucoside, quercetin-3-rhamnoglucoside, and
naringenin-7-rhamnoglucoside remained unchanged [78]. In a similar study,
the hydrolysis of quercetin glucosides, including quercetin-3-glucoside and
rutin, by h-glucosidase isolated from the rat small intestine did occur, for the
activity of jejunal h-glucosidase was highest for quercetin-4V-glucoside,
whereas rutin was a poor substrate [79]. Furthermore, luteolin-7-glucoside,
kaempferol-3-glucoside, and quercetin-3-glucoside are cleaved by rat jejunal or
ileal mucosa, suggesting the presence of h-glucosidase in the enterocytes [52].
Most investigations and controversy have surrounded the absorption of
quercetin glucosides in the small intestine [1,71,75]. Initial investigations
pointed to the absorption of quercetin glucosides in the small intestine
[17,24 26,80 83]; however, more recently these observations have been
questioned. A number of studies reported the uptake of quercetin glucosides
into the circulation. Quercetin glucosides were reported to be absorbed from
onions fed to ileotomized volunteers [81], and investigations made since the late
1990s have postulated that avonoid glucosides may be absorbed by the small
Metabolism
375
intestine via the sodium-dependent glucose transporter (SGLT-1) [42,84 86].

However, a similar study in ileostomy patients, fed a meal containing high
concentrations of both quercetin mono-and diglucosides, resulted in no detection
of the these compounds in ileostomy uid [74]. In contrast, the amounts of the
aglycone quercetin were substantial, suggesting that both quercetin glycosides
are eciently hydrolyzed in the small intestine by h-glucosidases to quercetin
[74]. Observations of the absorption of quercetin-3-glucoside and other glucosides of quercetin may have been confused by the coelution of these compounds
with corresponding quercetin glucuronides on high-performance liquid chromatography (HPLC). With the introduction of new mass spectrometric techniques
for the detection of avonoids and their metabolites [87 89] it should be
possible to solve beyond doubt whether quercetin-3-glucoside is absorbed intact
in the small intestine or not. One argument against the uptake of intact quercetin
glucosides is that the metabolic capacity of h-glucosidase in the small intestine,
and of the liver, is too great for quercetin glucosides to escape deglycosylation
[71]. In support of this, an analysis of human plasma using HPLC with coularray
detection after oral administration of quercetin-3-glucoside or quercetin4V-glucoside determined that no intact quercetin glucosides were present [90].
The major components in plasma were detectable by coularray detection to be
quercetin glucuronides, as conrmed by the disappearance of the glucuronide
peaks after treatment of the plasma h-glucuronidase.
The absorption of quercetin glycosides in the small intestine has also been
investigated by using the Caco-2 cell model [86,91 93]. Initial observations
suggested the facile absorption of quercetin through the human intestinal
epithelium but did not support an active transport process for quercetin glucosides [91]. However, more recent investigations suggest that transport of one of
the predominant dietary forms of quercetin, quercetin-4V-h-glucoside, across the
apical membrane of enterocytes may involve both the apical multidrug resistance associated protein 2 (MRP-2) [92] and/or the SGLT-1 [84 86], meaning
that the transfer of quercetin glycosides in the small intestine might be possible.
However, addition of plasma on the basolateral side signicantly reduced the
eux of quercetin by 94%, and therefore the eect of plasma binding can result
in an overestimation of basolateral to apical eux and result in misleading net
ux calculations in these types of experiments [93]. Both quercetin-3-glucoside
(isoquercitrin) and quercetin-4V-glucoside (spiraeoside) signicantly inhibit
SGLT-1-mediated mucosal uptake of the glucose analog methyl-a-D-glucopyranoside (MDG), whereas the aglycone quercetin and quercetin-3-rhamnoglucoside (rutin) were ineective [85]. In addition, the transport activity of SGLT-1
was markedly inhibited by green tea polyphenols [84] and was most pronounced
with epicatechin gallate (ECG) and epigallocatechin gallate (EGCG). These
studies suggest that quercetin glucosides may be capable of interacting with
SGLT-1 in the mucosal epithelium and may therefore be absorbed by the small
376
Spencer et al.
intestine in vivo. Whether or not they may also escape deglycosylation in the
enterocytes and the liver is still to be addressed, although human feeding studies
would suggest that this is so.
Further evidence that confounds observations of the inability of glycosides
to cross the small intestine is derived from the many reports that the anthocyanidin glycosides (anthocyanins) are readily absorbed intact without initial
cleavage of the sugar groups in the lumen of the small intestine [82,94 96].
Cyanidin-3-O-h-D-glucoside rapidly appeared in the plasma of rats after
administerion; however, the cyanidin aglycone was not detected [95], although
it was present in the jejunum. Furthermore, both cyanidin-3-glucoside and
cyanidin-3,5-diglucoside were rapidly incorporated into the plasma of rats and
humans after oral dosage [94], again indicating that anthocyanins may be
absorbed from the digestive tract into the blood circulation system in mammals
without structural alteration of the glycoside forms. Other glycosides such as the
rhamnoglucoside of quercetin, rutin, are absorbed in the small intestine intact
[52,97,98]. With an isolated rat small intestine model, about 10% of the
administered rutin appeared on the vascular side, chiey as free rutin (5.6%),
but some rutin sulfate (2.5%) and glucuronide (2.0%) were also detected [97]. In
a similar investigation rutin was also observed on the serosal side after perfusion
of a rat jejunum and ileum perfusion model; however, metabolites of rutin were
not detectable [52].
IV.
COLONIC METABOLISM
Studies have suggested that the extent of absorption of dietary polyphenols in

the small intestine is relatively small (10 20%) [52,58,59]. The implications of
this low absorption in the small intestine are that the majority of ingested
polyphenols, including those absorbed and conjugated in the enterocytes and/or
the liver before transport back out into the lumen either directly or via the bile
[50], reach the large intestine, where they encounter colonic microora. The
colon contains approximately 1012 microorganisms/cm3, which have an enormous catalytic and hydrolytic potential, and this enzymatic degradation of
avonoids by the colonic microora results in a huge array of new metabolites.
For example, bacterial enzymes may catalyze many reactions, including hydrolysis, dehydroxylation, demethylation, ring cleavage, and decarboxylation, as
well as rapid deconjugation [30]. Unlike human enzymes, the microora
catalyze the breakdown of the avonoid backbone itself to simpler molecules
such as phenolic acids. Specic metabolites have been observed in urine after
consumption of a variety of phenolics. For example, the glycine conjugate of
benzoic acid, hippuric acid, is primarily derived from plant phenolics and
aromatic amino acids through the action of intestinal bacteria, and, conse-
Metabolism
377
quently, the level of hippuric acid would be expected to increase in the urine of
individuals consuming diets rich in avanols or polyphenols in general. It must
be noted, however, that hippuric acid could possibly derive from other sources
such as quinic acid or, in quantitative terms, more importantly from the aromatic
amino acids tryptophan, tyrosine, and phenylalanine, as well as from the use of
benzoic acid as a food preservative. To date, most studies looking at the
metabolism of avonoids in the large intestine have used either avanols or
avonols, and there are few data on the metabolism of other commonly
consumed avonoids and other polyphenols.
A.
Flavanols
The 5,7,3,3V,4V-hydroxylation pattern of avan-3-ols is believed to enhance ring

opening after hydrolysis [28,30], and metabolism of avanols by enzymes of the
microora of the large intestine results in many metabolites: 3,4-dihydrophenylacetic acid, 3-hydroxyphenylacetic acid, homovanillic acid, and their conjugates
derived from the B-ring [30] and phenolic acids from the C-ring. Flavanols,
because of their structures (no C-4 carbonyl group), can also degrade to the
specic metabolites phenylvalerolactones. Phenylpropionic acids (which may
undergo further metabolism to benzoic acids) may also be the products of
avanol metabolism in animal studies, which demonstrate ssion of the A-ring
[30]. Only 3.1% of the ingested catechin was extractable from feces after feeding
of rats, indicating that major absorption and/or degradation of catechin had
occurred in the gastrointestinal tract [99].
Such metabolites of avanols have been detected in human plasma and
urine after a single ingestion of green tea [100], suggesting that that there may
be signicant metabolism by gut microora in the colon. The two metabolites
(-)-5-(3V,4V,5V-trihydroxyphenyl)-g-valerolactone and (-)-5-(3V,4V-dihydroxyphenyl)-g-valerolactone were identied in urine by both LC-mass spectrometry
(LC-MS/MS) and nuclear magnetic resonance (NMR), appearing 7.5 13.5 h
after ingestion (after a 3-h lag time), whereas EC and EGC peaked at 2 h. As
well as their late excretion proles, the amounts of metabolite excreted were 8to 25-fold greater than those of epicatechin and epigallocatechin (EGC) excretion and accounted for 6 39% of the EC and EGC ingested. The late excretion
and high levels of these metabolites would suggest that they are generated from
the precursors epicatechin and EGC by the intestinal microorganisms. Before
this human study, similar observations were made in rats fed with labeled
catechin [101]. Here catechin glucuronides were observed in the bile after
dosage of rats with the catechin, and m- and p-hydroxyphenylproprionic acid, y(3-hydroxyphenyl)-g-valerolactone, and y-(3,4-dihydroxyphenyl)-g-valerolactone were identied as metabolites that arose through action of the colonic
microora [101].
378
Spencer et al.
In humans, studies on black tea consumption [38] suggest an association

between polyphenol intake and excreted amounts of hippuric acid found. The
formation of hippuric acid and hydroxyhippuric acids seems to be a possible
central metabolic pathway for dietary avonoids, in which the colon and the
liver are active metabolic sites [36]. Other hydroxybenzoic acid glycine
derivatives such as 4-hydroxyhippuric acid, vanilloylglycine, and isovanilloylglycine might also appear in reasonable amounts in urine after polyphenol
consumption in general.
The metabolism of procyanidins by incubated human colonic microora has
been studied in vitro under anoxic conditions, using nonlabeled and 14C-labeled
puried proanthocyanidin polymers [102]. Interestingly, the oligomers were
almost totally degraded after 48 h of incubation, and meta- or para-monohydroxylated-phenylacetic, phenylpropionic, and phenylvaleric acids were identied as
metabolites, providing the rst evidence that dietary avan-3-ol polymers can be
degraded to low-molecular-weight aromatic compounds in the body [102].
B.
Flavonols
Quercetin-3-rhamnoglucoside and quercetin-3-rhamnoside may be hydrolysed

by strains of colonic Bacteroides distasonis, B. uniformis, and B. ovatus, which
may cleave the sugar by using a-rhamnosidase and h-glucosidase to liberate
quercetin aglycone [103]. For example, a cell-free extract of B. distasonis,
containing h-glucosidase, displayed an enzymatic activity of 1 Amol/10 min/10
mg of protein [103]. Other bacteria, such as Enterococcus casseliavus, may
utilize the sugar to yield formate, acetate, and lactate but do not further
metabolize the aglycone [104]. Eubacterium ramulus occurs at numbers of
approximately 108/g dry feces in humans and has been observed to degrade
quercetin-3-glucoside [104], luteolin-7-glucoside, rutin, quercetin, kaempferol,
luteolin, eriodictyol, naringenin, taxifolin, and phloretin [105] to phenolic acids.
It may also hydrolyze kaempferol-3-sorphoroside-7-glucoside to kaempferol-3sorphoroside and transform 3,4-dihydroxyphenylacetic acid, a product of anaerobic quercetin degradation [104], to nonaromatic fermentation products [105].
E. ramulus is capable of degrading the aromatic ring system of quercetin,
producing the transient intermediate phloroglucinol. However, this bacterium
was observed not to grow on phloroglucinol or quercetin aglycone itself and
only to cleave the avonoid ring system when glucose was present as a
cosubstrate [104].
When quercetin-3-rhamnoside was incubated anaerobically with human
intestinal bacteria, quercetin, 3,4-hihydroxyphenylacetic acid, and 4-hydroxybenzoic acid were produced as metabolites [106]. Analysis of urinary metabolites
after orally administered rutin labeled with deuterium [(2V,5V,6V-2H]rutin led to
the detection of 3-hydroxyphenylacetic acid, 3-methoxy-4-hydroxyphenylacetic
Metabolism
379
acid, 3,4-dihydroxyphenylacetic acid, 3,4-dihydroxytoluene, and 3-(m-hydroxyphenyl)-propionic acid as rutin metabolites. Unmetabolized rutin and quercetin
were not present in the urine [102], suggesting the avonol had been metabolized to phenolic acid metabolites by colonic microorganisms.
C.
Other Studies
The avonoid glycosides rutin, hesperidin, naringin, and poncirin are also
metabolized to phenolic acids, via aglycones, by human intestinal microora that
produce a-rhamnosidase, exo-h-glucosidase, endo-h-glucosidase, and/or h-glucuronidase enzymes [107]. In addition, baicalin, puerarin, and daidzin were
transformed to their aglycones by the bacteria, producing h-glucuronidase, Cglycosidase, and h-glycosidase, respectively. h-Glucosidase (EC 3.2.1.21) has
been puried from Bacteroides JY-6, a human intestinal anaerobic bacterium [108].
Protocatechuic acid has been detected in the plasma of rats after administration of
cyanidin 3-O-h-D-glucoside, and it is proposed that this metabolite is produced
by degradation of cyanidin by the microora [95]. The metabolite was present in
the plasma at a concentration that was approximately eightfold higher than
that of cyanidin 3-O-h-D-glucoside, which had been absorbed intact in the
small intestine.
V.
CONCLUSIONS
Figure 5 summarizes the sites of biotransformation of dietary avonoids in the

body. It is clear that the gastrointestinal tract plays a very signicant role in the
metabolism and conjugation of these polyphenols before the liver is reached. In
the jejunum and ileum of the small intestine there is ecient glucuronidation of
nearly all avonoids to diering extents by the action of UDP-glucuronosyltransferase enzymes. In the case of catechol-containing B-ring avonoids there
is also extensive O-methylation by the action of COMT. Unabsorbed avonoids,
and those taken up, metabolized in the small intestine and liver, and transported
back into the intestinal lumen, reach the large intestine, where they are further metabolized by the gut microora to smaller phenolic acids. The extent to
which these phenolic acids are absorbed in the colon is unknown; however, they
are detected in the plasma and are often further conjugated and metabolized in
the liver. Remaining compounds derived from avonoid intake pass out in
the feces.
It is the action of these avonoid metabolites, in particular the Omethylated avonoids deriving from small intestinal absorption, that is of great
current interest. For example, the ability of 3V-O-methyl epicatechin and
epicatechin glucuronides to protect against apoptotic cell death induced by
380
Spencer et al.
Metabolism
381
hydrogen peroxide or oxidized low-density lipoprotein (LDL) has been investigated [109 111]; it is discussed further in Chapter 9. It is also possible that
avonoid glucuronides might be cleaved by the action of h-glucourosidases
located in human tissues such as the liver and neutrophils (Fig. 6). Various
quercetin glucuronides are deconjugated by liver cell-free extracts and by pure
recombinant human h-glucuronidase, indicating that the cleavage of glucuronides to free aglycones may occur in vivo [112]. Rat studies suggest that
metabolic conversion of avonoid does not inuence the antioxidative ability of
rat plasma, indicating that conjugated metabolites may participate in the antioxidant defense in blood plasma [55]. In terms of other possible modes of action,
studies suggest that the metabolism of avonoids by phase I and II enzymes in the
small intestine may aid detoxication of potential carcinogens. For example, anapthoavone increases the activity of P450/3A in human jejunal and ileal
microsomes [113], and chrysin causes an induction of intestinal UDP-glucuronosyltransferase (UGT1A1) in Caco-2-cells [114,115].
There is also a need to assess the role of the colonic microora in the overall
bioavailability and potential bioactivity of dietary avonoids. The amount of
absorption of colonic metabolites is unclear at this time, and there is growing
interest in the potential eects of the phenolic acids and their derivatives as
potentially benecial agents. For example, the human intestinal bacteria metabolites
of rutin and quercetin, 3,4-dihydroxyphenylacetic acid, and 4-hydroxyl-phenylacetic acid have been shown to possess more eective antiplatelet aggregation
activity than rutin and quercetin [107]. Furthermore, 2,4,6-trihydroxybenzaldehyde
and quercetin were more eective than rutin in their cytotoxicity against tumor cell
lines. The eects the phenolics themselves have on the microora are an emerging
eld, and it is possible that a avonoid-induced change in the rich colonic bacterial
population may have an inuence on the overall health of the individual.
Over recent years we have gained greater knowledge of the bioavailable
metabolites of dietary avonoids, and it is now essential to evaluate fully the role
of these conjugates and metabolites in disease prevention. It will be important to
assess whether the observed metabolism aids entry into cells and/or renders them
better or worse in providing protection against dierent stresses, such as
oxidative or nitrative stress. New data in the eld are already beginning to
Figure 5 Summary of the formation of metabolites and conjugates of flavonoids
in humans. Cleavage of procyanidins may occur in the stomach in environments of
low pH. All classes of flavonoids undergo extensive metabolism in the jejunum and
ileum of the small intestine, and resulting metabolites enter the portal vein and
undergo further metabolism in the liver. Colonic microflora degrade flavonoids into
smaller phenolic acids, which may also be absorbed. The fate of most of these
metabolites is renal excretion; however, the extent to which these compounds
enter cells and tissues is unknown.
382
Spencer et al.
Figure 6 Possible fate and biological action of flavanol metabolites/conjugates in

vivo.
suggest that avonoids may act to protect cells by more complex mechanisms
than was once thought [110]. Eventually it is hoped that these studies will allow
specic dietary recommendations that will increase general health in the
population to be made.
REFERENCES
1.
Rice-Evans C, Spencer JPE, Schroeter H, Rechner AR. Bioavailability of

flavonoids and potential bioactive forms in vivo. Drug Metabol Drug Interact
2000; 17:291 310.
2. Rice-Evans C. Flavonoid antioxidants. Curr Med Chem 2001; 8:797 807.
of flavonoids and phenolic acids. Free Radic Biol Med 1996; 20:933 956.
4. Rice-Evans C. Plant polyphenols: free radical scavengers or chain-breaking
antioxidants? Biochem Soc Symp 1995; 61:103 116.
Metabolism
383
Rice-Evans CA, Miller NJ. Antioxidant activities of avonoids as bioactive

components of food. Biochem Soc Trans 1996; 24:790 795.
6. Rice-Evans CA, Miller NJ, Bolwell PG, Bramley PM, Pridham JB. The relative
antioxidant activities of plant-derived polyphenolic avonoids. Free Radic Res
1995; 22:375 383.
7. Sekher PA, Chan TS, OBrien PJ, Rice-Evans CA. Flavonoid B-ring chemistry and
antioxidant activity: fast reaction kinetics. Biochem Biophys Res Commun 2001;
282:1161 1168.
8. Brown JE, Khodr H, Hider RC, Rice-Evans CA. Structural dependence of
avonoid interactions with Cu2+ ions: implications for their antioxidant properties. Biochem J 1998; 330 (pt 3):1173 1178.
9. Oldreive C, Zhao K, Paganga G, Halliwell B, Rice-Evans C. Inhibition of nitrous
acid-dependent tyrosine nitration and DNA base deamination by avonoids and
other phenolic compounds. Chem Res Toxicol 1998; 11:1574 1579.
10. Pannala AS, Rice-Evans C. Rapid screening method for relative antioxidant
activities of avonoids and phenolics. Methods Enzymol 1901; 335:266 272.
11. Packer L, Rimbach G, Virgili F. Antioxidant activity and biologic properties of a
Biol Med 1999; 27:704 724.
12. Lairon D, Amiot MJ. Flavonoids in food and natural antioxidants in wine. Curr
Opin Lipidol 1999; 10:23 28.
13. Gyorgy I, Foldiak G. Free-radical induced redox reactions of drugs a pulseradiolysis study. Magyar Kemiai Folyoirat 1992; 98:361 366.
14. deOliveira TT, Nagem TJ, daSilva MC, deMiranda LCG, Teixeira MA.
Antioxidant action of the avonoids derivatives. Pesquisa Agropecuaria
Brasileira 1999; 34:879 883.
15. Bohm H, Boeing H, Hempel J, Raab B, Kroke A. Flavonols, avones and anthocyanins as native antioxidants of food and their possible role in the prevention of
chronic diseases. Z Ernahrungswissenschaft 1998; 37:147 163.
16. Hollman PC, Hertog MG, Katan MB. Role of dietary avonoids in protection
against cancer and coronary heart disease. Biochem Soc Trans 1996; 24:785 789.
17. Hollman PC, Katan MB. Absorption, metabolism and health eects of dietary
avonoids in man. Biomed Pharmacother 1997; 51:305 310.
18. Hollman PC, Feskens EJ, Katan MB. Tea avonols in cardiovascular disease and
cancer epidemiology. Proc Soc Exp Biol Med 1999; 220:198 202.
19. Arts IC, Hollman PC, Bueno dMH, Feskens EJ, Kromhout D. Dietary catechins
and epithelial cancer incidence: the Zutphen elderly study. Int J Cancer 2001; 92:
298 302.
20. Arts IC, Hollman PC, Feskens EJ, Bueno dMH, Kromhout D. Catechin intake
might explain the inverse relation between tea consumption and ischemic heart
disease: the Zutphen Elderly Study. Am J Clin Nutr 2001; 74:227 232.
21. Hertog MG, Feskens EJ, Hollman PC, Katan MB, Kromhout D. Dietary
antioxidant avonoids and risk of coronary heart disease: the Zutphen Elderly
Study. Lancet 1993; 342:1007 1011.
22. Hertog MG, Hollman PC. Potential health eects of the dietary avonol quercetin.
Eur J Clin Nutr 1996; 50:63 71.
5.
384
23.
Spencer et al.
Hollman PCH, Katan MB. Absorption, metabolism and health eects of dietary
avonoids in man. Biomed Pharmacother 1997; 51:305 310.
24. Hollman PC, Katan MB. Health eects and bioavailability of dietary avonols.
Free Radic Res 1999; 31 (suppl):S75 S80
25. Hollman PC, Katan MB. Dietary avonoids: intake, health eects and
bioavailability. Food Chem Toxicol 1999; 37:937 942.
26. Hollman PC, Katan MB. Bioavailability and health eects of dietary avonols in
man. Arch Toxicol Suppl 1998; 20:237 248.
27. Hollman PC. Bioavailability of avonoids. Eur J Clin Nutr 1997; 51 (suppl 1):
S66 S69.
28. Spencer JPE, Schroeter H, Rechner A, Rice-Evans C. Bioavailability of avan-3ols and procyanidins: gastrointestinal tract inuences and their relevance to
bioactive forms in vivo. Antiox Redox Signal 2001; 3:1023 1040.
29. Scalbert A, Williamson G. Dietary intake and bioavailability of polyphenols.
J Nutr x2000; 130:2073S 2085S.
30. Scheline RR. Metabolism of oxygen heterocyclic compounds. In: CRC Handbook
of mammalian metabolism of plant compounds. Boca Raton, FL: CRC Press,
1999: 243 295.
31. Okushio K, Suzuki M, Matsumoto N, Nanjo F, Hara Y. Methylation of tea
catechins by rat liver homogenates. Bioscie Biotech Biochem 1999; 63:430 432.
32. Okushio K, Suzuki M, Matsumoto N, Nanjo F, Hara Y. Identification of
(-)-epicatechin metabolites and their metabolic fate in the rat. Drug Metab Dispos
1999; 27:309 316.
33. Tsuchiya H, Sato M, Kato H, Okubo T, Juneja LR, Kim M. Simultaneous
determination of catechins in human saliva by high-performance liquid
chromatography. J Chromatogr B 1997; 703:253 258.
34. Yang CS, Lee MJ, Chen L. Human salivary tea catechin levels and catechin
esterase activities: implication in human cancer prevention studies. Cancer
Epidemiol Biomarkers Prev 1999; 8:83 89.
35. Laires A, Pacheco P, Rueff J. Mutagenicity of rutin and the glycosidic activity of
cultured cell-free microbial preparations of human faeces and saliva. Food Chem
Toxicol 1989; 27:437 443.
36. Macdonald IA, Mader JA, Bussard RG. The role of rutin and quercitrin in
stimulating avonol glycosidase activity by cultured cell-free microbial preparations of human feces and saliva. Mutat Res 1983; 122:95 102.
37. Parisis DM, Pritchard ET. Activation of rutin by human oral bacterial isolates to
the carcinogen-mutagen quercetin. Arch Oral Biol 1983; 28:583 590.
38. de F, V, Mateus N. Structural features of procyanidin interactions with salivary
proteins. J Agric Food Chem 2001; 49:940 945.
39. Nishioka H, Nishi K, Kyokane K. Human saliva inactivates mutagenicity of
carcinogens. Mutat Res 1981; 85:323 333.
40. Spencer JPE, Chaudry F, Pannala AS, Srai SK, Debnam E, Rice-Evans C.
Decomposition of cocoa procyanidins in the gastric milieu. Biochem Biophys Res
Commun 2000; 272:236 241.
41. Porter IJ. Tannins. In: Harborne JB, ed. Plant Biochemistry. Vol. I. Plant
Phenolics. London: Academic Press, 2002:389 418.
Metabolism
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
385
Gee JM, Dupont MS, Rhodes MJC, Johnson IT. Quercetin glucosides interact
with the intestinal glucose transport pathway. Free Radic Biol Med 1998; 25:
19 25.
Bae EA, Han MJ, Kim DH. In vitro anti-Helicobacter pylori activity of some
avonoids and their metabolites. Planta Med 1999; 65:442 443.
Lin JH, Chiba M, Baillie TA. Is the role of the small intestine in rst-pass
metabolism overemphasized? Pharmacol Rev 1999; 51:135 158.
Higuchi WI, Ho NF, Park JY, Komiya I. Rate-limiting steps and factors in drug
absorption. In: Prescott LF, Nimno WS, eds. Drug Absorption. New York: ADIS
Press, 1981:35 60.
Ho NF, Park JY, Ni PF, Higuchi WI. Advancing quantitative and mechanistic
approaches in interfacing gastrointestinal drug absorption studies in animals and
humans. In: Crouthamel W, Sarapu AC, eds. Animal Models for Oral Drug
Delivery: In Situ and In Vivo Approaches. Washington DC: American Pharmaceutics Association, 1983:27 106.
Yoshino K, Suzuki M, Sasaki K, Miyase T, Sano M. Formation of antioxidants
from (-)-epigallocatechin gallate in mild alkaline uids, such as authentic intestinal juice and mouse plasma. J Nutr Biochem 1999; 10:223 229.
Chen ZY, Zhu QY, Wong YF, Zhang ZS, Chung HY. Stabilizing eect of ascorbic
acid on green tea catechins. J Agric Food Chem 1998; 46:2512 2516.
Cheng Z, Radominska-Pandya A, Tephly TR. Studies on the substrate specicity
of human intestinal UDP-lucuronosyltransferases 1A8 and 1A10. Drug Metab
Dispos 1999; 27:1165 1170.
Crespy V, Morand C, Manach C, Besson C, Demigne C, Remesy C. Part of
quercetin absorbed in the small intestine is conjugated and further secreted in the
intestinal lumen. Am J Physiol 1999; 277:G120 G126
131: 1753 1757.
Spencer JPE, Chowrimootoo G, Choudhury R, Debnam ES, Srai SK, Rice-Evans
C. The small intestine can both absorb and glucuronidate luminal avonoids.
FEBS Lett 1999; 458:224 230.
Windmill KF, McKinnon RA, Zhu X, Gaedigk A, Grant DM, McManus ME. The
role of xenobiotic metabolizing enzymes in arylamine toxicity and carcinogenesis:
functional and localization studies. Mutat Res 1997; 376:153 160.
Franski R, Bednarek P, Siatkowska D, Wojtaszek P, Stobiecki M. Application of
mass spectrometry to structural identication of avonoid monoglycosides isolated
from shoot of lupin (Lupinus luteus L.). Acta Biochim Pol 1999; 46:459 473.
Terao J. Dietary avonoids as antioxidants in vivo: conjugated metabolites of (-)epicatechin and quercetin participate in antioxidative defense in blood plasma. J
Med Invest 1999; 46:159 168.
Carbonaro M, Grant G, Pusztai A. Evaluation of polyphenol bioavailability in rat
small intestine. Eur J Nutr 2001; 40:84 90.
Fisher RB, Gardner ML. A kinetic approach to the study of absorption of solutes
by isolated perfused small intestine. J Physiol 1974; 241:211 234.
Kuhnle G, Spencer JPE, Chowrimootoo G, Schroeter H, Debnam ES, Srai SKS,
386
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
Spencer et al.
Rice-Evans C, Hahn U. Resveratrol is absorbed in the small intestine as resveratrol
glucuronide. Biochem Biophys Res Commun 2000; 272:212 217.
Kuhnle G, Spencer JPE, Schroeter H, Shenoy B, Debnam ES, Srai SK, Rice-Evans
small intestine. Biochem Biophys Res Commun 2000; 277:507 512.
Crevoisier C, Buri P, Boucherat J. The transport of three avonoids across articial
and biological membranes. 5. Transport in situ across the small intestine of the rat.
Pharm Acta Helv 1975; 50:231 236.
Choudhury R, Chowrimootoo G, Srai K, Debnam E, Rice-Evans CA. Interactions
of the avonoid naringenin in the gastrointestinal tract and the inuence of
glycosylation. Biochem Biophys Res Commun 1999; 265:410 415.
Shimoi K, Okada H, Furugori M, Goda T, Takase S, Suzuki M, Hara Y,
Yamamoto H, Kinae N. Intestinal absorption of luteolin and luteolin 7-O-betaglucoside in rats and humans. FEBS Letters 1998; 438:220 224.
Harada M, Kan Y, Naoki H, Fukui Y, Kageyama N, Nakai M. Identification of the
major antioxidative metabolites in biological uids of the rat with ingested (+)catechin and (-)-epicatechin. Biosci Biotechnol Biochem 1999; 63: 973 977.
Boutin JA, Meunier F, Lambert PH, Hennig P, Bertin D, Serkiz B. In-vivo and invitro glucuronidation of the avonoid diosmetin in rats. Drug Metab Dispos 1993;
21:1157 1166.
Mannisto PT, Kaakkola S. Catechol-O-methyltransferase (COMT): biochemistry,
molecular biology, pharmacology, and clinical efficacy of the new selective
COMT inhibitors. Pharmacol Rev 1999; 51:593 628.
Okushio K, Matsumoto N, Kohri T, Suzuki M, Nanjo F, Hara Y. Absorption of tea
catechins into rat portal vein. Biol Pharm Bull 1996; 19:326 329.
Hara Y. Inuence of tea catechins on the digestive tract. J Cell Biochem Suppl
1997; 27:52 58.
Spencer JPE, Schroeter H, Shenoy B, Srai SK, Debnam ES, Rice-Evans C.
Epicatechin is the primary bioavailable form of the procyanidin dimers B2 and B5
after transfer across the small intestine. Biochem Biophys Res Commun 2001;
285:588 593.
de Santi C, Pietrabissa A, Spisni R, Mosca F, Pacifici GM. Sulphation of
resveratrol, a natural compound present in wine, and its inhibition by natural
avonoids. Xenobiotica 2000; 30:857 866.
Pai TG, Suiko M, Sakakibara Y, Liu MC. Sulfation of avonoids and other
phenolic dietary compounds by the human cytosolic sulfotransferases. Biochem
Biophys Res Commun 2001; 285:1175 1179.
Day AJ, Williamson G. Biomarkers for exposure to dietary avonoids: a review of
the current evidence for identication of quercetin glycosides in plasma. Br J Nutr
2001; 86 (suppl 1):105 110.
Hollman PCH, van Trijp JMP, Buysman MNCP, VanderGaag MS, Mengelers
MJB, deVries JHM. Relative bioavailability of the antioxidant avonoid quercetin
from various foods in man. Febs Lett 1997; 418:152 156.
Hollman PC, Bijsman MN, van Gameren Y, Cnossen EP, de Vries JH, Katan MB.
The sugar moiety is a major determinant of the absorption of dietary avonoid
glycosides in man. Free Radic Res 1999; 31:569 573.
Metabolism
74.
387
Walle T, Otake Y, Walle UK, Wilson FA. Quercetin glucosides are completely
hydrolyzed in ileostomy patients before absorption. J Nutr 2000; 130: 2658 2661.
75. Day AJ, Canada FJ, Diaz JC, Kroon PA, Mclauchlan R, Faulds CB. Dietary
avonoid and isoavone glycosides are hydrolysed by the lactase site of lactase
phlorizin hydrolase. FEBS Lett 2000; 468:166 170.
76. Gee JM, Dupont MS, Day AJ, Plumb GW, Williamson G, Johnson IT. Intestinal
transport of quercetin glycosides in rats involves both deglycosylation and
interaction with the hexose transport pathway. J Nutr 2000; 130: 2765 2771.
77. Morand C, Manach C, Crespy V, Remesy C. Respective bioavailability of quercetin
aglycone and its glycosides in a rat model. Biofactors 2000; 12:169 174.
78. Day AJ, Dupont MS, Ridley S, Rhodes M, Rhodes MJ, Morgan MR.
Deglycosylation of avonoid and isoflavonoid glycosides by human small
intestine and liver beta-glucosidase activity. FEBS Lett 1998; 436:71 75.
79. Ioku K, Pongpiriyadacha Y, Konishi Y, Takei Y, Nakatani N, Terao J. Betaglucosidase activity in the rat small intestine toward quercetin monoglucosides.
Biosci Biotech Biochem 1998; 62:1428 1431.
80. Hollman PCH, vanderGaag M, Mengelers MJB, van Trijp JMP, deVries JH, Katan
MB. Absorption and disposition kinetics of the dietary antioxidant quercetin in
man. Free Radic Biol Med 1996; 21:703 707.
81. Hollman PC, de Vries JH, van Leeuwen SD, Mengelers MJ, Katan MB.
volunteers. Am J Clin Nutr 1995; 62:1276 1282.
82. Paganga G, RiceEvans CA. The identification of avonoids as glycosides in
human plasma. FEBS Lett 1997; 401:78 82.
83. Hollman PC, van Trijp JM, Buysman MN, van der Gaag MS, Mengelers MJ, de
Vries JHM, Katan MB. Relative bioavailability of the antioxidant avonoid
quercetin from various foods in man. FEBS Lett 1997; 418:152 156.
84. Kobayashi Y, Suzuki M, Satsu H, Arai S, Hara Y, Suzuki K, Miyamoto Y,
Shimizu M. Green tea polyphenols inhibit the sodium-dependent glucose
transporter of intestinal epithelial cells by a competitive mechanism. J Agric
Food Chem 2000; 48:5618 5623.
85. Ader P, Block M, Pietzsch S, Wolffram S. Interaction of quercetin glucosides
with the intestinal sodium/glucose co-transporter (SGLT-1). Cancer Lett 2001;
162: 175 180.
86. Walgren RA, Lin JT, Kinne RK, Walle T. Cellular uptake of dietary avonoid
quercetin 4V-beta-glucoside by sodium-dependent glucose transporter SGLT1.
87. Lommen A, Godejohann M, Venema DP, Hollman PC, Spraul M. Application of
directly coupled HPLC-NMR-MS to the identification and confirmation of
quercetin glycosides and phloretin glycosides in apple peel. Anal Chem 2000;
72: 1793 1797.
88. Justesen U. Collision-induced fragmentation of deprotonated methoxylated
avonoids, obtained by electrospray ionization mass spectrometry. J Mass
Spectrom 2001; 36:169 178.
89. Justesen U, Arrigoni E. Electrospray ionisation mass spectrometric study of
degradation products of quercetin, quercetin-3-glucoside and quercetin-3-rhamno-
388
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
Spencer et al.
glucoside, produced by in vitro fermentation with human faecal flora. Rapid
Commun Mass Spectrom 2001; 15:477 483.
Sesink AL, OLeary KA, Hollman PC. Quercetin glucuronides but not glucosides
are present in human plasma after consumption of quercetin-3-glucoside or
quercetin-4V-glucoside. J Nutr 2001; 131:1938 1941.
human intestinal epithelial Caco-2 cells. Biochem Pharmacol 1998; 55:1721 1727.
Walgren RA, Karnaky KJJ, Lindenmayer GE, Walle T. Elux of dietary flavonoid
quercetin 4V-beta-glucoside across human intestinal Caco-2 cell monolayers by
apical multidrug resistance-associated protein-2. J Pharmacol Exp Ther 2000; 294:
830 836.
Walgren RA, Walle T. The inuence of plasma binding on absorption/exsorption
in the Caco-2 model of human intestinal absorption. J Pharm Pharmacol 1999; 51:
1037 1040.
Miyazawa T, Nakagawa K, Kudo M, Muraishi K, Someya K. Direct intestinal
absorption of red fruit anthocyanins, cyanidin-3-glucoside and cyanidin-3,5diglucoside, into rats and humans. J Agric Food Chem 1999; 47:1083 1091.
Tsuda T, Horio F, Osawa T. Absorption and metabolism of cyanidin 3-O-betaD-glucoside in rats. FEBS Lett 1999; 449:179 182.
Lapidot T, Harel S, Granit R, Kanner J. Bioavailability of red wine anthocyanins
as detected in human urine. J Agric Food Chem 1998; 46:4297 4302.
Andlauer W, Stumpf C, Furst P. Intestinal absorption of rutin in free and
conjugated forms. Biochem Pharmacol 2001; 62:369 374.
Olthof MR, Hollman PC, Vree TB, Katan MB. Bioavailabilities of quercetin-3glucoside and quercetin-4V-glucoside do not differ in humans. J Nutr 2000; 130:
1200 1203.
Bravo L, Abia R, Eastwood MA, Saura-Calixto F. Degradation of polyphenols
(catechin and tannic acid) in the rat intestinal tract. Effect on colonic fermentation
and faecal output. Br J Nutr 1994; 71:933 946.
Li C, Lee MJ, Sheng SQ, Meng XF, Prabhu S, Winnik B, Huang BM, Chung JY, Yan
SQ, Ho CT, Yang CS. Structural identification of two metabolites of catechins and
their kinetics in human urine and blood after tea ingestion. Chem Res Toxicol 2000;
13:177 184.
Das NP, Griiths LA. Studies on flavonoid metabolism:metabolism of (+)-[14C]
catechin in the rat and guinea pig. Biochem J 1969; 115:831 836.
Deprez S, Brezillon C, Rabot S, Philippe C, Mila I, Lapierre C, Scalbert A.
Polymeric proanthocyanidins are catabolized by human colonic microflora into
low-molecular-weight phenolic acids. J Nutr 2000; 130:2733 2738.
Bokkenheuser VD, Shackleton CH, Winter J. Hydrolysis of dietary avonoid
glycosides by strains of intestinal Bacteroides from humans. Biochem J 1987; 248:
953 956.
Schneider H, Schwiertz A, Collins MD, Blaut M. Anaerobic transformation of
quercetin-3-glucoside by bacteria from the human intestinal tract. Arch Microbiol
1999; 171:81 91.
Schneider H, Blaut M. Anaerobic degradation of avonoids by Eubacterium
ramulus. Arch Microbiol 2000; 173:71 75.
Metabolism
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
389
Baba S, Furuta T, Fujioka M, Goromaru T. Studies on drug metabolism by use of

isotopes XXVII: urinary metabolites of rutin in rats and the role of intestinal
microora in the metabolism of rutin. J Pharm Sci 1983; 72:1155 1158.
Kim DH, Jung EA, Sohng IS, Han JA, Kim TH, Han MJ. Intestinal bacterial
metabolism of avonoids and its relation to some biological activities. Arch Pharm
Res 1998; 21:17 23.
Kim DH, Sohng IS, Kobashi K, Han MJ. Purification and characterization of betaglucosidase from Bacteroides JY-6, a human intestinal bacterium. Biol Pharma
Bull 1996; 19:1121 1125.
Spencer JPE, Schroeter H, Kuhnle G, Srai SKS, Tyrrell RM, Hahn U, Rice-Evans
C. Epicatechin and its in vivo metabolite, 3V-O-methyl epicatechin, protect human
Biochem J 2001; 354:493 500.
Schroeter H, Spencer JPE, Rice-Evans C, Williams RJ. Flavonoids protect neurons
from oxidized low-density-lipoprotein-induced apoptosis involving c-Jun Nterminal kinase (JNK), c-Jun and caspase-3. Biochem J 2001; 358:547 557.
Spencer JPE, Schroeter H, Crossthwaithe AJ, Kuhnle G, Williams RJ, Rice-Evans
hydrogen peroxide-induced cell death in neurons and fibroblasts. Free Radic Biol
Med 2001; 31:1139 1146.
OLeary KA, Day AJ, Needs PW, Sly WS, OBrien NM, Williamson G. Flavonoid
glucuronides are substrates for human liver beta-glucuronidase. FEBS Lett 2001;
503:103 106.
McKinnon RA, Burgess WM, Hall PM, Abdul-Aziz Z, McManus ME. Metabolism
of food-derived heterocyclic amines in human and rabbit tissues by P4503A
proteins in the presence of avonoids. Cancer Res 1992; 52:2108s 2113s.
Galijatovic A, Otake Y, Walle UK, Walle T. Induction of UDP-glucuronosyltransferase UGT1A1 by the avonoid chrysin in Caco-2 cells potential role in
carcinogen bioinactivation. Pharm Res 2001; 18:374 379.
Walle T, Otake Y, Galijatovic A, Ritter JK, Walle UK. Induction of UDPglucuronosyltransferase UGT1A1 by the avonoid chrysin in the human hepatoma
cell line hep G2. Drug Metab Dispos 2000; 28:1077 1082.
15
Absorption of Quercetin Glycosides
Andrea J. Day
University of Leeds
Leeds, England
Gary Williamson
Institute of Food Research
Norwich, England
I.
INTRODUCTION
Understanding dietary avonoid absorption is a fundamental requirement for the

determination of the biological activity of these compounds in humans. The
degree of absorption from the diet clearly aects the biological activities at
various avonoid-responsive sites within the body. Of the avonoid subclasses,
avonols are ubiquitous in plant food. Quercetin is the major avonol to be
consumed; the estimated average Dutch intake of quercetin was calculated as 16
mg/day (73% of the avonol intake) [1]. The main sources of quercetin in the
diet include tea, onion, broccoli, apple, and red wine. Quercetin is rarely found
in plants in a free form but is usually conjugated to sugar residues, and so most
foods contain quercetin glycosides and not the aglycone. Processing and cooking
of plant foods rarely cause deglycosylation [2], but treatments that release
hydrolytic enzymes from the plant tissue [3] or fermentation with microorganisms can result in an increased level of aglycone. For example, in some types of
red wine more than 50% of the quercetin exists in the free form [4]. The
conjugation of quercetin and other avonoids aects the mechanism by which
the compound is absorbed by altering the basic physicochemical properties and
thus the ability to enter cells, to interact with transporters, and to interact with
cellular (lipo)proteins.
Dietary avonoid glycosides, with the exception of anthocyanin glycosides
[5], are deglycosylated during the process of absorption from the lumen of the
391
392
Day and Williamson
intestine into the systemic circulation. The site of deglycosylation and uptake
through the epithelial cells of the intestine depends on the nature of the avonoid,
the nature of the attached sugar, and the position of attachment of the sugar on the
avonoid ring. Quercetin glucosides are deglycosylated in the small intestine by
endogenous h-glucosidases, whereas quercetin rhamnoglucosides are not substrates for these enzymes and are only deglycosylated by microora in the colon.
This distinction is illustrated by the dramatic dierences in pharmacokinetics
between quercetin-3-glucoside and quercetin-3-rhamnoglucoside. Deglycosylation in the small intestine occurs either in the gut lumen by lactase phlorizin
hydrolase (LPH) or after transport into epithelial cells [possibly by sodiumdependent glucose transporter 1 (SGLT-1) and other sugar transporters] by
cytosolic h-glucosidase. The avonoid aglycone generated by LPH diuses into
the epithelial cell as a result of its increased octanol/water partition coecient.
The intracellular aglycone, which results from either LPH/diusion or SGLT-1/
cytosolic h-glucosidase, is then conjugated with glucuronic acid and pumped out
of the cell by multidrug resistanceassociated protein (MRP) or p-glycoprotein
family transporters, either into the systemic circulation or back into the gut
lumen. The apparent result of these metabolic processes is that the quercetin
glucoside is eciently absorbed by the small intestine but the serosal product in
the hepatic portal vein is a quercetin glucuronide. The aim of this chapter is to
describe the current available experimental evidence to support the preceding
hypothesis and the consequences for health and disease.
II.
STUDIES OF QUERCETIN UPTAKE BEFORE 1995
Considerable research on absorption of quercetin was carried out between 1960

and 1985 [6]. Many of these studies involved feeding rats high doses of either
quercetin or quercetin-3-rhamnoglucoside (rutin), a major glycosylated form of
quercetin present in plant foods. The work demonstrated that quercetin was
absorbed to some extent in rats, but that rutin absorption was dependent on
intestinal microora activity; there was no signicant absorption in germ-free or
antibiotic-treated rats, suggesting that the glycoside conjugate required hydrolysis before the free quercetin could be absorbed. The high level of microbial
metabolism also resulted in degradation of the aglycone ring structure. Although
the phenolic acids produced from degradation can be absorbed, low levels of the
avonol aglycone in the urine would result. Several human studies were also
carried out during this period, again feeding either quercetin or rutin as pure
compounds rather than avonols from food. For example, Gugler and associates
[7] fed 4 g of quercetin aglycone (as solid) to human volunteers and detected no
quercetin in the urine (detection limit 0.04 g; 1% of dose). The low yield in urine
was also observed in more recent studies on quercetin absorption. Originally the
1. Pharmacokinetics parameters of uptake and

excretion after feeding foods or puried
compounds
2. Identication and quantication of plasma
metabolites formed by humans
1. Easier to conduct and control than human
intervention studies
2. Information on identication and quantication
of plasma metabolites in the animal
3. Eect of higher administered doses on absorption
4. Bilary excretion monitored easily
1. Intact tissue study possible
2. Eect of inhibitors able to be studied to gain
mechanistic information
1. Immortalized cell lines that allow repeated
studies with controlled conditions
2. Whole cell activity monitoring, allowing substrate
interaction within compartments to be followed
3. Regulation of protein expression or use of stably
transfected cell lines, and use of inhibitors to
provide mechanistic information
1. Substrate specicity and rates, along with inhibitor
studies to facilitate study of detailed mechanisms
Animal intervention
Animal tissue
(small intestine) ex vivo
Cell culture
(e.g., Caco-2 cell line)
Enzyme studies
Information gained/advantages
Human intervention
Study method
Limitations
1. No account of accessibility and cellular

compartmentalization of enzymes
1. Key proteins (e.g., enzymes and transporters)

potentially not expressed or expressed to
dierent levels than normal cells
2. No interaction with other food components
1. Blood ow or diusional sink potentially

not present
2. Tissue not of human origin
1. Nonhuman data not easily extrapolated

to humans
2. Dierent distribution of microora than human
3. Dierent prole of pharmacokinetic data
than human
4. Dierent metabolites from human metabolites
formed
1. No mechanistic information
2. Inability to use inhibitors
3. Absorption from specic sections of intestinal
tract and biliary excretion not easily measured
Table 1 Limitations of Approaches Used to Study Absorption of Flavonoid Glycosides

393
394
Day and Williamson
observations were interpreted as meaning that quercetin was not absorbed

eectively, but it is now known thatdespite the evidence of low urinary
yieldsthe quercetin in plasma can reach signicant levels [up to f7 Am (see
late discussion)].
Hollman and colleagues [8] developed a new analytical method that
allowed low levels of quercetin to be measured (detection limit of 0.15 ng/
mL). Flavonols were derivatised post column with aluminum, resulting in
uorescent compounds that could be measured at a high sensitivity. Using
ileostomized subjects who had had the colon surgically removed and so are
believed not to have the experimental complication of microbial metabolism),
Hollman and associates showed that quercetin was indeed bioavailable in humans
[9]. In the study, subjects were fed either fried onion (a rich source of quercetin4V-glucoside and quercetin-3,4V-glucoside), quercetin aglycone, or rutin supplements; urine and ileostomy eluent were collected. By calculating the amount of
quercetin remaining in the ileostomy bag, an indirect measure of absorbed
avonol was obtained. From onions, 52% of the quercetin was absorbed, whereas
from the quercetin aglycone or rutin supplement only 24% and 17% were
absorbed, respectively. Thus, it was suggested not only that the type of attached
sugar aected absorption, but that in the case of the glucoside the sugar actually
promoted uptake of quercetin.
Many dierent approaches have been used to study bioavailability of
quercetin glycosides. These include human and animal intervention studies,
which monitor appearance of compounds in plasma and urine. With animals,
particular sections of the gastrointestinal tract can be isolated or the bile duct can
be cannulated, to allow additional information that would not necessarily be
available from human studies to be generated. Enzyme and cell culture studies
provide further evidence of the mechanical processes involved in absorption,
metabolism, and distribution of avonol glycosides. The properties of enzymes
can also be studied by using the tools of molecular biology. Enzymes can be
expressed at high levels, to a high degree of purity, or in such a way that
interactions with other cellular components can be studied with less noise than is
present in vivo. With each of these approaches, there are signicant limitations of
the information gained from individual methods (Table 1). It is only by
combining data generated from complementary methods that we can start to
understand the processes involved in bioavailability.
III.
HUMAN INTERVENTION STUDIES
Studies investigating the absorption and disposition of avonol glycosides from

foods in human subjects entail measurement of quercetin in the plasma and
allow 24-h recovery in urine to be monitored after ingestion of dierent foods.
Resulting plasma concentrations of quercetin metabolites range from 0.1 AM to
395
7.6 AM in plasma, depending on the source and level of quercetin administered,

typically with 0.11.0% of dose excreted in urine. Table 2 summarizes reported
plasma and urine quercetin levels after consumption of quercetin-rich foods in
human subjects. Some of the studies show the dramatic eect that the nature of
the glycoside conjugate exerts on absorption of quercetin.
A study by Hollman and coworkers [10] showed that quercetin glucosides
from onions were rapidly absorbed* across the small intestine (time to reach
maximal plasma concentration, tmax < 0.7 h), whereas rutin exhibited a delayed
absorption (tmax 9.3 h), typical of absorption in the colon, in other studies [15,18]*,
pure quercetin-3-glucoside, quercetin-4V-glucoside, and rutin administered to
volunteers displayed no signicant dierence in the rate (tmax, 0.6 h) or extent
(maximal plasma concentration, Cmax, 5 AM) of absorption of the two glucosides.
However, rutin was absorbed much more slowly (tmax 6 h) and to a lesser extent
than quercetin glucosides in terms of peak plasma concentrations (Cmax 0.2 AM).
These were critical experiments since they demonstrated conclusively that the
nature of the sugar and not the position of attachment was the main factor in
determining the manner of absorption of quercetin glycosides.
Graefe and coworkers [21] in 2001 conducted a similar human study
comparing absorption from supplements with that from foods rich in the same
type of quercetin conjugate. Peak plasma concentrations of quercetin from either
onion or quercetin-4V-glucoside were similar (Cmax 7.6, 7.0 AM, respectively), as
were the times to reach maximal concentration (tmax 0.7 h for both). Even at twice
the dose, the peak plasma concentration of quercetin from either buckwheat tea or
puried rutin was signicantly lower (Cmax 2.0, 1.0 AM, respectively) and took a
longer time to be reached (tmax 4.3, 7.0 h, respectively) than for quercetin from
the other sources. A conclusion from the work is that the food matrix may
inuence absorption of the compounds to some extent, but that the most
signicant eect on absorption is the form of the conjugate attached to quercetin.
In a study comparing the absorption from dietary supplements of quercetin
aglycone with rutin at various doses [16], rutin showed a clear maximal plasma
concentration (of quercetin metabolites) at 6 h, with almost no quercetin in
plasma at 4 h, indicative of absorption in the distal parts of the small intestine or
in the colon. Late absorption was observed with all doses administered (16, 40,
100 mg rutin). In contrast, quercetin (8, 16, 50 mg aglycone) showed no clear
maximal plasma concentration (tmax 1.9, 2.7, 4.9 h, respectively), but some
absorption was observed from 30 min after ingestion, as is consistent with
absorption along the length of the small and large intestine.
Comparing the available data provides evidence that quercetin appears in
the plasma within a short period if present in the diet as the aglycone or a glucoside
* In the studies of Hollman and associates and Olthof and associates all samples were hydrolyzed
and only the aglycone was measured.
9
9
9
5
15
15
24
10
9
9
16
16
7
9
9
4
12
12
12
12
12
12
12
Nj of subject
68
98
100
186
49
13
114
87
100
100
8, 20, 50
8, 20, 50
6894
100
100
79
15.9
13.7
14.2
100
100
200
200
Amount ingested (mg)b

0.74
0.3
0.3
2.18
0.1e
0.07e
1.48f
0.37g
0.23
3.2
0.14, 0.22, 0.29
0.08, 0.16, 0.30
0.63h
4.5
5.0
0.6i
0.05j
0.03j
0.03j
7.6
7.0
2.0
1.0
Cmaxc (AM)
<0.7
2.5
9.3
1.31.9d
5.6
<0.6
1.9, 2.7, 4.9
6.5, 7.4, 7.5
0.5
0.6
nd
nd
nd
nd
0.7
0.7
4.3
7.0
Tmax (h)
1.39
0.44
0.35
0.8
0.5
1.1
nd
nd
nd
nd
nd
nd
3.1
3.6
nd
1.0
0.6
0.8
nd
nd
nd
nd
% Excreted
Cmax is maximal concentration of quercetin metabolites in plasma; Tmax is time to reach maximal plasma concentration.
Quercetin aglycone equivalent.
c
Maximal concentration unless otherwise stated.
d
Peak concentration for dierent compound analyzed.
e
Daily portion split over three occasions for a 7-day intervention period. Blood was sampled approximately 4 h after rst portion.
f
Blood sampled 90 min after intake on day 7 of intervention.
g
Blood sampled 3 h after meal.
h
Daily portion split over three occasions for a 7-day intervention period. Blood sampled after 10 h of fasting.
i
Average concentration 1.5 h after consumption of onion meal.
j
Daily portion split over two to three occasions for a 4-day intervention period. Blood was sampled on two occasions on day 4 and avonol levels were
found to be consistent; nd, not determined.
Quercetin glucosides
Quercetin glycosides
Rutin
Rutin
Rutin
Quercetin-4V-glucoside
Quercetin supplement
Rutin supplement
Rutin
Rutin
Rutin
Flavonol glycoside
Typical Quercetin Concentrations in Plasma and Urine After Dietary Supplementation in Humansa
Onion
Apple10
Supplement10
Onion11
Tea12
Onion12
Onion13
Flavonol-rich meal14
Supplement15
Supplement15
Onion16
Supplement16
Onion17
Supplement18
Supplement18
Onion19
Onion20
Tea20
Red Wine20
Onion21
Supplement21
Buckwheat tea21
Supplement21
10
Food
Table 2
396
Day and Williamson
397
conjugate rather than as a rhamnoglucoside (rutinoside) conjugate (Fig. 1).

Quercetin from apples reaches a maximal plasma concentration intermediate to
that of quercetin glucosides or rutin, probably reflecting that apples contain a
mixture of glycoside conjugates [10,22].
Apart from the studies of Hollman and associates [9] and Erlund and
colleagues [16], which measured uptake of quercetin from a supplement, there
have been no other studies investigating the absorption of quercetin in a free
form present within a food matrix. Quercetin is sparingly soluble in water, and
studies with rats investigating the eect of solubility on uptake have shown that
quercetin absorption is highly dependent on the solvent in which it is ingested
[23]. Although free quercetin is rarely found in food, there is still no
convincing evidence that quercetin glucosides are absorbed better than the
aglycone in humans.
One study investigated the bioavailability of quercetin after consumption of
red wine compared with consumption of either onions or tea over 3 days [20]. At
the time the blood samples were taken, the plasma concentration after red wine
consumption was between those observed after consumption of onions and
tea. However, there was no dierence in the 24-h urinary excretion after
consumption of quercetin from red wine and onions (although for tea excretion
Figure 1 Summary of published data showing the time taken to reach maximal
concentration of quercetin in plasma after consumption of various avonol-rich
foods or supplements. Q3G, quercetin-3-glucoside; Q4VG, quercetin-4V-glucoside. Amount of quercetin glucoside consumed varies with study (see Table 2
for details).
398
Day and Williamson
was signicantly lower since it contain quercetin predominantly in the form of

rutin). Red wine has been shown to contain signicant amounts of free quercetin
[4], but in the human study described, the conjugation pattern of quercetin was
not recorded. Therefore, it is not possible to conclude whether the similar
excretion (and hence absorption) of quercetin from onions and red wine is due
to the presence of similar quercetin conjugates or to comparable absorption of
free quercetin from a soluble matrix.
There is now unequivocal evidence that some quercetin glycosides are
absorbed from the upper parts of the gastrointestinal tract in humans. However,
there are conicting reports on whether quercetin glycosides are present in the
plasma. Some researchers have reported the presence of quercetin glycosides in
plasma or urine. Paganga and Rice-Evans [24] identied rutin along with several
other unidentied avonoid glycosides in human plasma. Aziz and associates
[11] identied quercetin-4V-glucoside and 3V-methylquercetin-4V-glucoside, and
Boyle and colleagues [25] identied quercetin-3-glucoside and 3V-methylquercetin-4V-glucoside in plasma after consumption of onions. All these reports utilized
methods that were based on high-performance liquid chromatography (HPLC)
under acidic conditions with identication based on retention time of avonol
glucoside standards after detection by diode array or uorescence. Mauri and
coworkers [26] identied rutin in plasma after consumption of tomato with
detection by LC-mass spectrometry (LC-MS).
In contrast, Moon and associates [17], Walle and colleagues [27], and Day
and coworkers [19] found no quercetin glucosides in plasma from subjects who
had consumed an onion meal. The methodology used was HPLC with detection
by diode array with pure standards as markers. These conditions would have
allowed detection of quercetin glucosides if present. Wittig and associates [28]
could not detect, using tandom LC-MS/MS analysis, quercetin glucosides in
plasma after subjects consumed onions. Graefe and colleagues [21], using
coularray detection, could not detect quercetin glucosides in the plasma or urine
of subjects consuming onions or buckwheat tea or consuming quercetin-4Vglucoside or rutin as supplements. Likewise, Sesink and coworkers [29], also
using coularray detection, could not detect quercetin-3-glucoside or quercetin-4Vglucoside in the plasma of subjects consuming these pure compounds. Erlund and
associates [16], using electrochemical detection, could not detect rutin after
supplementation. Furthermore, both Graefe and coworkers and Sesink and
coworkers observed identical metabolic proles in subjects after consumption
of the dierent quercetin glycoside sources. In the study of Morand and coworkers [30], the same quercetin metabolic prole was also observed after rats
were fed quercetin, rutin, or quercetin-3-glucoside. These results suggest that,
post deglycosylation, quercetin follows the same metabolic pathway regardless of
the form of quercetin glycosides administered and that plasma metabolites are
quercetin (and methylquercetin) sulfates and glucuronides [19].
399
The apparently contradictory resultsthe presence or absence of quercetin

glucosides in plasmaare probably explained by misidentication of quercetin
glucosides since quercetin glucosides can have very similar retention times to
those of the equivalent glucuronides under acidic conditions on HPLC and almost
identical ultraviolet (UV) absorption spectra [31].
IV.
ANIMAL INTERVENTION STUDIES
As described previously, many animal studies between 1950 and 1980

investigated the uptake and metabolism of quercetin and rutin in rats. More
recently, a few animal intervention studies also considered absorption of other
quercetin glycosides in order to gain more information about the site of
absorption from the gastrointestinal tract. Manach and colleagues [32] investigated the uptake of quercetin and rutin administered to rats either as a single
dose or after a 14-day dietary regimen. With the acute dose quercetin
metabolites were shown to be present in plasma within 2 h of administration
of quercetin, but, as with humans, quercetin metabolites formed from rutin
were not found in plasma until 4 h after ingestion, suggesting a site of
absorption at the more distal parts of the small intestine. The authors also
showed that progressively more quercetin accumulated in the cecal contents of
the rats fed rutin. With an extended feeding study of quercetin in the diet,
plasma quercetin levels remained high, but absorption of quercetin became less
ecient than with nonadapted rats. It is interesting to consider that most
studies involving absorption of avonols have used single meal (or dose)
intervention and have not considered the eect of typical diet on the absorption
of avonols.
Choudhury and coworkers [33] measured the amounts of quercetin,
quercetin-3-glucoside, or rutin in the urine after either an intravenous or an oral
dose. After intravenous administration the authors detected a proportion of each
of the compounds in the urine. However, after oral administration of the
avonol-free quercetin, quercetin metabolites or rutin could not be detected in the
urine. Only a small percentage of quercetin-3-glucoside was detected in the urine
after the glucoside was administered. In contrast, quercetin-3-glucoside was not
detected in plasma of rats fed quercetin-3-glucoside [30]. The study compared
absorption of quercetin, quercetin-3-glucoside, quercetin-3-rhamnoside, and
rutin. Quercetin was absorbed at approximately 33% of the level of quercetin3-glucoside when given at the same dose. At 4 h after administration, rutin
absorption was 10% of quercetin-3-glucoside absorption, and quercetin-3-rhamnoside absorption was not detected. This suggests that, at least in rats, quercetin3-glucoside is much more eciently absorbed than the aglycone or other
glycoside conjugates.
400
V.
Day and Williamson
ENZYME STUDIES
It was assumed originally that since secreted digestive juices do not possess
h-glucosidase activity, and since quercetin h-glycosides resist hydrolysis by
stomach acid and other pancreatic enzymes [34], avonoid glycosides would not
be hydrolysed until they reached the large intestine. In the colon, various
bacteria are able to release the aglycone but also further metabolize or degrade
the avonoid ring structure. However, high levels of endogenous h-glucosidase
activity have been demonstrated in both human and rat small intestine epithelial
cells [35,36] and in human liver [35]. The activity is attributed to a broadspecicity soluble cytosolic h-glucosidase [37] present in both tissues, and to
membrane-bound LPH [38] found on the brush border of the small intestine.
Only quercetin glucosides that are not conjugated through the 3-position of the
avonol are substrates for the broad-specicity cytosolic h-glucosidase [39].
Rutin is not a substrate for either LPH or cytosolic h-glucosidase, and the
absence of endogenous hydrolytic enzymes explains why rutin is not absorbed in
the small intestine but passes to the colon, where it is deglycosylated by
microbial a-rhamnosidases and h-glucosidases.
Table 3 shows the substrate specicity of cytosolic h-glucosidase and LPH.
It should be noted that the avonols quercetin-3-glucoside and quercetin-4Vglucoside were substrates for the lactase domain of LPH and not the phlorizin
hydrolase domain, as may have been expected, given that phlorizin has a
Table 3 Specic Activities of Purified Endogenous Enzymes on Quercetin

Glycosides and on Lactosea
Lactase phlorizin
Cytosolic
hydrolaseb
h-glucosidasec
Quercetin-4Vglucoside
Quercetin-3glucoside
Lactose
Rutin
Substrates
for human
small intestine Absorption in
extractd
small intestine
170
34
Yes
Yes
137
Yes
Yes
4
0
0
0
nd
No
Yese
No
Values are expressed as catalytic eciencies (Vmax/Km) at pH 7.4 and 37 jC; nd, not determined.
Lactase phlorizin hydrolase (LPH) was puried from sheep small intestine [40].
c
Recombinant human enzyme expressed in Pichia pastoris [39].
d
Cell-free extracts of human small intestine were assessed for activity [35].
e
Absorbed as glucose and galactose after hydrolysis.
a
401
avonoid structure [38]. The specicity of the two human enzymes is the key
factor in determining whether or not absorption occurs in the small intestine in
humans. The rat small intestine diers from the human small intestine in its
higher population of microbes. The relatively high content of microbes in the rat
gut may possess sucient activity to hydrolyse avonoid glycosides. In contrast,
the luminal contents of the human small intestine (which has only f104
microbes/mL in the duodenum compared to f1012 microbes/g in the colon
contents) probably do not exhibit signicant activity on avonoid glycosides.
However, it cannot be excluded that the human small intestine lumen may contain
endogenous proteolytically released LPH, cytosolic h-glucosidase from sloughed
o epithelial cells, or some cytosolic h-glucosidase from the bile [27].
The demonstrated endogenous mammalian h-glucosidase activity has
several implications for the absorption of quercetin glucosides. First, as LPH
is present on the luminal side of the small intestine, any hydrolysis of quercetin
glycosides would result in liberation of the aglycone that could readily diuse
across the epithelial cells as a result of increased hydrophobicity of the
aglycone and release within the unstirred layer. Thus LPH provides a plausible
explanation for absorption of quercetin in the small intestine and quercetin
glycosides would not have to enter epithelial cells for this to occur. Second, if
quercetin glycosides were transported across the intestinal wall, then the
cytosolic h-glucosidase in both the small intestine and the liver would be
active on the glycosidic bond of those avonols not conjugated in the 3position. Thus, if any avonol-3-glycoside were absorbed intact, it would not
be deglycosylated and would remain unchanged circulating in plasma until
possible resecretion in the bile. Third, populations with polymorphisms in these
enzymes would absorb and metabolize avonols dierently. LPH levels vary
widely: 75% of the worlds population show lactose maldigestion caused by
reduced levels of LPH in adulthood. This may have implications for avonol
absorption that have not been explored.
VI.
CELL CULTURE STUDIES
Caco-2 cells, derived from a human colon adenocarcinoma, dierentiate spontaneously when grown on plates or lters into a continuous monolayer of cells
presenting morphological and functional characteristics typical of normal ileal
enterocytes [4143] with a pseudobrush border membrane. In the dierentiated
form, the monolayer mimics in some respects the microvilli of the small intestine
epithelial cells. However, as for all immortalized cell lines, several key enzymes
present in the human small intestine in vivo either are not expressed or are
expressed to dierent extents between clones. The Caco-2 model has been used to
study the intestinal absorption of avonoids [4449]. The evidence for transport
402
Day and Williamson
of quercetin and quercetin glucosides across Caco-2 cells is confusing and is

sometimes contradictory to the results from isolated small intestine (discussed
later). Walgren and associates [44] demonstrated that although quercetin was
absorbed eciently across Caco-2 cells (by diusion [47]), quercetin-4V-glucoside was not. In subsequent experiments, Walgren and colleagues showed that
quercetin-4V-glucoside was pumped back out of Caco-2 cells by (MRP-2) (a
multidrug-resistant protein present on the apical side of the intestinal epithelium
[45]. By inhibiting MRP-2, the authors were able to demonstrate that quercetin4V-glucoside could be transported by the SGLT-1 and conrmed the result in a
nonintestinal cell line, Chinese hamster ovary cells, that had been stably transfected with SGLT-1 (G6C3 cells [46]). However, Caco-2 cells typically have
very low levels of LPH [50], and so deglycosylation as a route for absorption
would probably have been missing from these experiments.
In similar experiments, Murota and colleagues [49] demonstrated the
absorption and metabolism of quercetin, quercetin-3-glucoside, quercetin-4Vglucoside, and quercetin-3,4V-glucoside by using Caco-2 cells. Absorption of
quercetin was more ecient than that of any of the glucosides, although when
compared to the other glucosides, quercetin-4V-glucoside was absorbed and
metabolized to the greatest extent. Again, although not measured, the cells would
have been decient in LPH compared to that of normal small intestine, a
dierence that precludes quantitative comparison of the models. The authors
suggested that on the basis of octanol-water partition coecients of the quercetin
conjugates the lipophilicity was the determining factor in absorption of the
compound. Quercetin glucosides and quercetin conjugates (including methylated
quercetin metabolites) were measured in the basolateral solutions. Despite the
limitations, the results suggest that (1) quercetin glucosides can be transported
across intestinal cells, (2) quercetin glucosides are deglycosylated during transport across intestinal cells, and (3) quercetin is metabolized by Caco-2 cells.
Furthermore, as MRP-2 was not inhibited in these experiments, reduced transfer
of compound to the basolateral side may have occurred as a result of greater
apical elux of the glucoside than would normally be expected. Figure 2 summarizes the observations from Caco-2 cells.
If the Caco-2 cells are good experimental models for the human and the rat,
then the results with Caco-2 cells appear to contradict the results observed in the
human and rat studies. Quercetin glucosides are absorbed eciently in vivo.
Despite the elux eect of MRP-2 on net transfer of quercetin glucosides, and the
fact that none of the experiments incorporated a basolateral sink (the eect of
plasma protein binding [48,51,52]), a more ecient transfer of quercetin from
quercetin glucosides would have been expected than that observed. As described,
one possible explanation for a low percentage transfer in Caco-2 cells lies in the
level of expression of the h-glucosidase, LPH. Although Caco-2 cells express
LPH, the levels of the enzyme are signicantly lower than those found in vivo
403
Figure 2 Schematic representation of data obtained after incubation of quercetin

and quercetin glucosides with Caco-2 cells. Q, quercetin; Q3G, quercetin-3glucoside; Q4VG, quercetin-4V-glucoside; Q-conj, quercetin conjugated metabolites;
dotted arrows represent very low conversion to aglycone; thick arrows represent
major pathway of uptake and metabolism. , MRP-2/SGLT-1 or similar transporters
on the apical surface of the membrane; , unknown transporters on the basolateral
surface of the enterocyte. (Data summarized from Ref. 49). Quercetin glucosides,
metabolites, and free aglycone were measured in the basolateral solutions after
incubation with Caco-2 cells. (From Refs. 45,46.) Quercetin-4V-glucoside was
measured within the enterocyte, the MRP-2 transporter was shown to be involved in
the efflux of the glucoside back into the apical solution, and h-glucosidase activity
was demonstrated. (From Ref. 56.) A low level of apical deglycosylation was
found when quercetin-3-glucoside and quercetin-4V-glucoside were incubated with
Caco-2 cells. All researchers showed a preference of quercetin uptake over the
glucoside uptake. MRP-2, multidrug resistanceassociated protein 2; SGLT-1,
sodium-dependent glucose transporter 1.
(Table 4). As an example of quercetin metabolism by LPH-free cells, HepG2

cells, derived from human liver, contain only cytosolic h-glucosidase and do not
express LPH. HepG2 cells metabolize quercetin-4V-glucoside but do not metabolize quercetin-3-glucoside [55]. Because of the low LPH levels, Caco-2 cells
also rapidly metabolize quercetin-4V-glucoside [46] but metabolize quercetin-3glucoside very slowly if at all [56]. In marked contrast, rat small intestine
metabolizes both quercetin-4V- and 3-glucosides [57].
404
Day and Williamson
Table 4 Relative Activities of Lactase Phlorizin Hydrolase in Cells and Tissues

Typical specific activity
on lactose (mU/mg protein)
Caco2 cells (undierentiated)
Caco2 cells (dierentiated)
Caco2 cells clone PD7 (dierentiated)
Human small intestine cell-free extract
Human small intestine cell free extract
(from lactose-intolerant individuals)
Rat small intestine cell free extract
a
00.12a
0.1 (50)
0.3 (50)
2080 (53)
210 (53)
1020 (54)
Unpublished data.
VII.
RAT INTESTINAL STUDIES
Everted-jejunal sacs have been used to study uptake of quercetin glycosides in

vitro. Gee and associates [58] demonstrated, using elux of [14C]-galactose
across rat everted-jejunal sacs, that quercetin-3- and 4V-glucoside, but not rutin,
interacted with (but were not necessarily transported by) SGLT-1. Ader and
colleagues [59] have also shown an interaction of quercetin-3- and 4V-glucoside,
but not quercetin or rutin, with SGLT-1, by competitive Na+-dependent mucosal
uptake of methyl-a-D-glucopyranoside (a nonmetabolizable glucose analog). An
interaction, however, does not show that transport has occurred since phenolic
glucosides can interact by inhibiting sugar transport but at the same time are not
transported across the membrane. For example, phlorizin, the classical inhibitor
of SGLT-1, is not transported across the brush border because of the bulky
nature of the aglycone [60]. There is considerable interspecies variation in
properties of SGLT-1: the Ki for phlorizin is 10 AM for rat, 250 AM for human,
and 750 AM for rabbit; p-nitrophenyl h-glucoside is transported by the rabbit
form but inhibits the human and rat transporters [61], and these dierences make
comparisons of results from interspecies studies dicult.
Using a similar model, the rat perfused small intestine, Spencer and
colleagues [62] found that a substantial proportion of quercetin-3-glucoside
was transferred as the intact glucoside after a 90-min incubation, although some
hydrolysis had occurred during absorption, as evident by the presence of some
aglycone and of glucuronide metabolites. In contrast, Gee and coworkers [57]
and Crespy and associates [63] did not detect quercetin-3-glucoside in experiments with everted-jejunum or in situ perfused rat intestine for a shorter period
(using 15- to 30-min incubation, respectively). Transfer of quercetin had
occurred, since signicant concentrations of quercetin metabolites were found.
405
Rutin has been detected on the vascular side after incubation with perfused rat
intestine [62,64], although quercetin aglycone was not detected.
VIII.
COMPARISON OF EXPERIMENTAL MODELS
Quercetin was more eciently transferred across Caco-2 cells than quercetin
glucosides [49], but quercetin glucosides were more eciently transferred
across isolated rat small intestine than quercetin [57], and similar results were
seen in vivo in rats [30]. The conicting evidence may be a result of dierences
between the experimental models, either interspecies variation in transporter
properties (human model compared to rat model), or variation in expression of
proteins such as LPH (cell lines compared to normal tissue).
Quercetin-4V-glucoside can be transported by SGLT-1 into Caco-2 cells,
and interactions between quercetin-3- and 4V-glucoside and SGLT-1 have been
demonstrated in rat small intestine. LPH has activity toward avonol glycosides,
which may constitute the major role in the uptake of quercetin glycosides. Crespy
and associates [63] provided some evidence for the involvement of LPH in the
uptake of quercetin-3-glucoside. In the experiment, quercetin-3-glucoside was
incubated with perfused rat intestine in the presence or absence of phlorizin (an
inhibitor of SGLT-1). Quercetin-3-glucoside was hydrolysed by the lactase-active
site and not the phlorizin hydrolaseactive site of LPH [38], and so phlorizin
should not aect quercetin-3-glucoside hydrolysis. The composition of avonoids in the intestinal lumen contents was analyzed at the end of the 30-min
incubation. The results showed that in the presence of phlorizin there was no
signicant eect on quercetin aglycone formed from quercetin-3-glucoside by the
action of LPH, but also there was no dierence in the total quercetin-3-glucoside
absorbed (i.e., net absorption+ eluxed metabolites). However, there was a
dierence in the eluxed proportion of quercetin metabolites, probably due to
competition of quercetin-3-glucoside and phlorizin for both conjugating enzymes
and the elux transporters such as MRP-2 [65]. Crespy and colleagues [63]
showed that although twice the level of quercetin from quercetin-3-glucoside,
compared to the level from the aglycone, was absorbed to the serosal side of the
rat intestine (net absorption), the total absorption, including the metabolites that
were secreted to the luminal side, was the same for both compounds. Evidently
the glucoside does not help the transfer of quercetin into the enterocyte, but it
does slow the rate of elux, and so eventually a greater proportion of quercetin
metabolites from quercetin-3-glucoside enters the bloodstream.
In order to demonstrate further the role of LPH in absorption of quercetin3- and 4V-glucoside, N-(m-butyl)-deoxy-galactonojirimycin (NBDGM, an inhibitor of LPH), or phlorizin (an inhibitor of SGLT-1) was employed in the rat
everted-jejunum model. The results demonstrate that by inhibiting LPH in a rat
406
Day and Williamson
model, luminal deglycosylation of quercetin-3- and 4V-glucoside is signicantly

reduced. Consequently, transfer of quercetin-3- and 4V-glucoside to the vascular
side of the tissue is lower (Fig. 3). In the presence of phlorizin, luminal hydrolysis
is unaected, quercetin-4V-glucoside transfer is reduced, but interestingly there is
no eect on quercetin-3-glucoside. A plausible explanation of the data is that
quercetin-3- and 4V-glucoside are absorbed across the intestine by dierent
mechanisms: quercetin-3-glucoside is hydrolyzed by LPH, with SGLT-1 possibly
only contributing to a minor extent; quercetin-4V-glucoside, in contrast, has two
routes of passage into the enterocyte, luminal hydrolysis by LPH to quercetin
followed by diusion of the aglycone, or transport by SGLT-1 with subsequent
deglycosylation by cytosolic h-glucosidase.
The preceding hypothesis makes understanding the dierences observed in
the Caco-2 studies easier. As a result of low expression of LPH in Caco-2 cells,
SGLT-1 plays the dominant role in uptake. Hence, quercetin-4V-glucoside is
absorbed more eciently than quercetin-3-glucoside; however, overall absorption
of the glucosides is still lower than in the rat model, in which both SGLT-1 and LPH
are active. This does not exclude the possibility that higher expression of MRP-2
(due to the adenocarcinoma origin of the cells) contributes to the rapid elux of the
Figure 3 Effect of an SGLT-1 or lactase inhibitor on the absorption of quercetin3-glucoside or quercetin-4V-glucoside across rat everted-jejunal sacs. Solid bars,
quercetin-3-glucoside; open bars, quercetin-4V-glucoside; NBDGM, N-(m-butyl)deoxy-galactonojirimycin. Quercetin glucosides (100 AM) were incubated with rat
everted-jejunal sacs for 15 min as previously described [57] in the presence or
absence of phlorizin (25 AM) as an SGLT-1 inhibitor or NBDGM (250 AM) as a
lactase inhibitor. Total quercetin (as quercetin glucuronide metabolites) was
measured in the serosal solution at the end of the incubation and is expressed as
percentage absorbed. SGLT-1, sodium-dependent glucose transporter 1.
407
Figure 4 Potential mechanisms for quercetin glucoside absorption across the

human small intestine. Q, quercetin; Q3G, quercetin-3-glucoside; Q4VG, quercetin4V-glucoside; Q-conj, quercetin conjugated metabolites; LPH, lactase phlorizin
hydrolase; SGLT-1, sodium-dependent glucose transporter; MRP-2, multidrug
resistant protein; C-h-G, cytosolic h-glucosidase; P-II, phase 2 conjugating
enzymes; dotted arrows, very low transport; thick arrows, major pathway of
absorption and metabolism. We hypothesize that quercetin-4V-glucoside can be
transported by SGLT-1, but quercetin-3-glucoside is a poor substrate. Both
compounds are actively deglycosylated by LPH and uptake is significantly reduced
if lactase is inhibited. Only quercetin-4V-glucoside can be deglycosylated by
cytosolic h-glucosidase.
glucoside conjugate from the enterocyte. Figure 4 represents the proposed mechanism of uptake of quercetin glucosides in the human small intestine.
IX.
CONCLUSIONS
The absorption of quercetin glycosides is complex. Flavonol-rich foods contain

a mixture of many dierent types of conjugate, all of which may be absorbed in
a dierent manner. Quercetin rhamnoglucosides such as rutin, commonly found
in plant foods, appear only to be absorbed in the large intestine after hydrolysis
by the colon microora. Late absorption in the gastrointestinal tract leads to
408
Day and Williamson
signicantly lower overall absorption, probably as a result of degradation of the

avonoid ring structure. Quercetin glucosides, in contrast, are absorbed much
earlier in the small intestine, leading to higher overall absorption of the
compounds, and probably greater enterohepatic circulation. However, the
quantitative contribution of each step to overall absorption in vivo remains
unclear. LPH appears to play an important role in hydrolyzing the glucosidic
bond, releasing the aglycone within the intestinal lumen for absorption by a
diusion mechanism. LPH has broad specicity, excepting certain sugar types
(excluding rhamnosides) and positions of substitution on the avonoid structure.
SGLT-1, or another transporter, may also contribute to absorption of some
compounds. Little, if any, glucoside conjugate appears in the plasma, probably
because of ecient hydrolysis within the enterocyte or hepatocyte by cytosolic
h-glucosidase. Further studies on the substrate specicity of SGLT-1 to transport the various avonol glycoside moieties, particularly in populations with low
LPH expression, are required. The ability of other avonoid subclasses to
interact with LPH and SGLT-1 must also be determined as early absorption
leads to a greater bioavailablity of these bioactive compounds.
ACKNOWLEDGMENTS
We would like to thank Mike Morgan, Geo Plumb, Paul Kroon, Jean-Guy
Berrin, Karen Oleary, Susan DuPont, Jenny Gee, and Ian Johnson for valuable
contributions. We would also like to thank the Biotechnology and Biological
Sciences Research Council and the EU Framework V project Polybind QLK11999-00505 for contribution to funding in some of the experiments described.
REFERENCES
1.
2.
3.
4.
Hertog MGL, Hollman PCH, Katan MB, Kromhout D. Intake of potentially

Nutr Cancer 1993; 20:2129.
Price KR, Bacon JR, Rhodes, MJC. Effect of storage and domestic processing on
the content and composition of avonol glucosides in onion (Allium cepa). J Agric
Food Chem 1997; 45:938942.
Price KR, Rhodes MJC. Analysis of the major avonol glycosides present in four
varieties of onion (Alium cepa) and changes in composition resulting from autolysis.
J Sci Food Agric 1997; 74:331339.
McDonald MS, Hughes M, Burns J, Lean MEJ, Matthews D, Crozier A. Survey of
the free and conjugated myricetin and quercetin content of red wines of dierent
geographical origins. J Agric Food Chem 1998; 46:368375.

5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
409
Miyazawa T, Nakagawa K, Kudo M, Muraishi K, Someya K. Direct intestinal

absorption of red fruit anthocyanins, cyaniding-3-glucoside and cyaniding-3,5diglucoside, into rats and humans. J Agric Food Chem 1999; 47:10831091.
Scheline RR. Handbook of Mammalian Metabolism of Plant Compounds. Boca
Raton, FL: CRC Press, 1991:267287.
Gugler R, Leschik M, Dengler MJ. Disposition of quercetin in man after single oral
and intravenous doses. Eur J Clin Pharmacol 1975; 9:229234.
Hollman PCH, van Trijp JMP, Buysman, MNCP. Fluorescence detection of
avonols in HPLC by postcolumn chelation with aluminium. Anal Chem 1996;
68:35113515.
Hollman PCH, de Vries JH, van Leeuwen SD, Mengelers MJ, Katan MB.
volunteers. Am J Clin Nutr 1995; 62:12761282.
Hollman PCH, van Trijp JMP, Buysman MNCP, Gaag MS, Menglers MJB, deVries
JHM, Katan MB. Relative bioavailablity of the antioxidant quercetin from various
foods in man. FEBS Lett 1997; 418:152156.
Aziz AA, Edwards CA, Lean MEJ, Crozier A. Absorption and excretion of
conjugated avonols, including quercetin 4V-O-h-glucoside and isorhamnetin 4V-Oh-glucoside, by human volunteers. Free Radic Res 1998; 29:257269.
DeVries JHM, Hollman PCH, Meyboom S, Buysman, MNCP, Zock PL, van
Staveren WA, Katan MB. Plasma concentrations and urinary excretion of the
antioxidant avonols quercetin and kaempferol as biomarkers for dietary intake.
Am J Clin Nutr 1998; 68:6065.
Janssen PLTMK, Mensink RP, Cox FJJ, Harryvan JL, Hovenier R, Hollman PCH,
Katan MB. Eects of the avonoids quercetin and apigenin on hemostasis in
healthy volunteers: results from an in vitro and a dietary supplement study. Am J
Clin Nutr 1998; 67:255262.
Manach C, Morand C, Crespy V, Demigne C, Texier O, Regerat F, Remesy C.
Quercetin is recovered in human plasma as conjugated derivatives which retain
antioxidant properties. FEBS Lett 1998; 426:331336.
Hollman PCH, Buysman MNCP, van Gameren Y, Cnossen EPJ, deVries JHM,
Katan MB. The sugar moiety is a major determinant of the absorption of dietary
flavonoid glycosides in man. Free Radic Res 1999; 31:569573.
Erlund I, Kosonen T, Alfthan G, Maenpaa J, Perttunen K, Kenraali J, Parantainen J,
Aro A. Pharmacokinetics of quercetin from quercetin aglycone and rutin in healthy
volunteers. Eur J Clin Pharmacol 2000; 56:545553.
Moon J-H, Nakata R, Oshima S, Inakuma T, Terao J. Accumulation of quercetin
conjugates in blood plasma after short-term ingestion of onion by women. Am J
Physiol 2000; 279:R461R467.
Olthof MR, Hollman PCH, Vree TB, Katan MB. Bioavailabilities of quercetin-3glucoside and quercetin-4V-glucoside do not differ in humans. J Nutr 2000;
130:12001203.
Day AJ, Mellon F, Barron D, Sarrazin G, Morgan MRA, Williamson G. Human
metabolism of dietary avonoids: identication of plasma metabolites of quercetin.
Free Radic Res 2001. In press.
410
20.
Day and Williamson
de Vries JHM, Hollman PCH, van Amersfoort I, Olthof MR, Katan MB. Red
wine is a poor source of bioavailable avonols in men. J Nutr 2001; 131:745
748.
21. Graefe EU, Wittig J, Mueller S, Riethling AK, Uehleke B, Drewelow B, Pforte H,
Jacobasch G, Derendorf H, Veit M. Pharmacokinetics and bioavailability of
quercetin glycosides in humans. J Clin Pharmacol 2001; 41:492499.
22. Price KR, Prosser T, Richetin AMF, Rhodes MJC. A comparison of the avonol
content and composition in dessert, cooking and cider-making apples, distribution
within the fruit and eect of juicing. Food Chem 1999; 66:489494.
23. Piskula MK, Terao J. Quercetins solubility aects its accumulation in rat plasma
after oral administration. J Agric Food Chem 1998; 46:43134317.
24. Paganga G, Rice-Evans CA. The identication of avonoids as glycosides in human
plasma. FEBS Lett 1997; 401:7882.
25. Boyle SP, Dobson VL, Duthie SJ, Kyle JAM, Collins AR. Absorption and DNA
protective eects of avonoid glycosides from an onion meal. Eur J Nutr 2000;
39:213223.
26. Mauri PL, Iemoli L, Gardana C, Riso P, Simonetti P, Porrini M, Pietta PG. Liquid
chromatography electrospray ionization mass spectrometric characterization of
avonol glycosides in tomato extracts and human plasma. Rapid Commun Mass
Spectrom 1999; 13:924931.
27. Walle T, Otake Y, Walle UK, Wilson FA. Quercetin glucosides are completely
hydrolyzed in ileostomy patients before absorption. J Nutr 2000; 130:26582661.
28. Wittig J, Herderich M, Graefe EU, Veit M. Identification of quercetin glucuronides
in human plasma by high-performance liquid chromatography-tandem mass
spectrometry. J Chromatogr B 2001; 753:237243.
29. Sesink ALA, OLeary KA, Hollman PCH. Quercetin glucuronides but not
glucosides are present in human plasma after consumption of quercetin-3-glucoside
or quercetin-4V-glucoside. J Nutr 2001; 131:19381941.
30. Morand C, Manach C, Crespy V, Remesy C. Quercetin 3-O-beta-glucoside is better
absorbed than other quercetin forms and is not present in rat plasma. Free Radic Res
2000; 33:667672.
31. Day AJ, Williamson G. Biomarkers for exposure to dietary avonoids: a review of
the current evidence for identication of quercetin glycosides in plasma. Br J Nutr
2001; 86:S105S110.
32. Manach C, Morand C, Demigne C, Texier O, Regerat F, Remesy C. Bioavailability
of rutin and quercetin in rats. FEBS Lett 1997; 409:1216.
33. Choudhury R, Srai SK, Debnam E and Rice-Evans CA. Urinary excretion of
hydroxycinnamates and avonoids after oral and intravenous administration. Free
Radic Biol Med 1999; 27:278286.
34. DuPont MS, Gee JM, Price KR, Johnson IT. The availability of avonol glycosides
for small intestinal transport. Gut 1999; 44:TH517.
35. Day AJ, DuPont MS, Ridley S, Rhodes M, Rhodes MJC, Morgan MRA, Williamson
G. Deglycosylation of avonoid and isoavonoid glycosides by human small
intestine and liver h-glucosidase activity. FEBS Lett 1998; 436:7175.
36. Ioku K, Pongpiriyadacha Y, Konishi Y, Takei Y, Nakatani N, Terao J. h-
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
411
glucosidase activity in the rat small intestine towards quercetin monoglucosides.

Biosci Biotech Biochem 1998; 62:14281431.
Lambert N, Kroon PA, Faulds CB, Plumb GW, McLauchlan WR, Day AJ,
Williamson G. Purication of cytosolic beta-glucosidase from pig liver and its reactivity towards avonoid glycosides. Biochim Biophys Acta 1998; 1435:110116.
Day AJ, Canada FJ, Daz JC, Kroon PA, Mclauchlan R, Faulds CB, Plumb GW,
Morgan MRA, Williamson G. Dietary avonoid and isoavone glycosides are
hydrolysed by the lactase site of lactase phlorizin hydrolase. FEBS Lett 1998;
468:166170.
Berrin J-G, McLauchlan WR, Juge N, Williamson G, Kroon PA. The human cbg1-1
gene encodes a h-glucosidase which hydrolyses a broad range of xenobiotic
glycosides. Eur J Biochem. In press.
Rivera-Sagredo A, Canada FJ, Nieto O, Jimenez-Barbero J, Martn-Lomas M.
Substrate-specificity of small-intestinal lactaseassessment of the role of the
substrate hydroxyl-groups. Eur J Biochem 1992; 209:415422.
Hidalgo IJ, Raub TJ, Borchardt RT. Characterization of the human-colon carcinoma
cell-line (Caco-2) as a model system for intestinal epithelial permeability.
Gastroenterology 1989; 96:736749.
Pinto M, Robine-Leon S, Appay MD. Enterocyte-like differentiation and polarization of the human colon carcinoma cell line Caco-2 in culture. Biol Cell 1983;
47:323330.
Artursson P. Cell cultures as models for drug absorption across the intestinal
mucosa. Crit Rev Ther Drug Carrier Syst 1991; 8:305330.
human intestinal epithelial Caco-2 cells. Biochem Pharmacol 1998; 55:17211727.
Walgren RA, Karnaky KJ, Lindenmayer GE, Walle T. Efflux of dietary flavonoid
quercetin 4V-beta-glucoside across human intestinal Caco-2 cell monolayers by
apical multidrug resistance-associated protein-2. J Pharmacol Exp Ther 2000;
294:830836.
Walgren RA, Lin J-T, Kinne RK-H, Walle T. Cellular uptake of dietary flavonoid
quercetin 4V-h-glucoside by the sodium dependent glucose transporter, SGLT1.
J Pharmacol Exp Ther 2000; 294:837843.
Kuo SM. Transepithelial transport and accumulation of avone in human intestinal
Caco-2 cells. Life Sci 1998; 63:23232331.
Walgren RA, Walle T. The influence of plasma binding on absorption/exsorption in
the Caco-2 model of human intestinal absorption. J Pharm Pharmacol 1999;
51:10371040.
Murota K, Shimizu S, Chujo H, Moon JH, Terao J. Eciency of absorption and
metabolic conversion of quercetin and its glucosides in human intestinal cell line
Caco-2. Arch Biochem Biophys 2000; 384:391397.
Chantret I, Rodolosse A, Barbat A, Dussaulx E, Brotlaroche E, Zweibaum A,
Rousset M. Dierential expression of sucrase-isomaltase in clones isolated from
early and late passages of the cell-line caco-2evidence for glucose-dependent
negative regulation. J Cell Sci 1994; 107:213225.
Manach C, Morand C, Texier O, Favier ML, Agullo G, Demigne C, Regerat F,
412
Day and Williamson
Remesy C. Quercetin metabolites in plasma of rats fed diets containing rutin or

quercetin. J Nutr 1995; 125:19111922.
52. Boulton DW, Walle UK, Walle T. Fate of the flavonoid quercetin in human cell
lines: chemical instability and metabolism. J Pharm Pharmacol 1999; 51:353359.
53. Rossi M, Maiuri L, Fusco MI, Salvati VM, Fuccio A, Auricchio S, Mantei N, Zecca
L, Gloor SM, Semenza G. Lactase persistence versus decline in human adults:
multifactorial events are involved in down-regulation after weaning. Gastroenterology 1997; 112:15061514.
54. Goda T, Yasutake H, Tanaka T, Takase S. Lactase-phlorizin hydrolase and sucraseisomaltase genes are expressed dierently along the villus-crypt axis of rat jejunum.
J Nutr 1999; 129:11071113.
55. Day AJ. Human Absorption and Metabolism of avonoid glycosides Ph.D.
dissertation, University of East Anglia, England, 2000.
56. Plumb GW, McLauchlan R, Kroon PA, Day AJ, Faulds CB, Gee JM, DuPont MS,
Williamson G. Uptake and metabolism of quercetin glycosides in Caco-2 cell
culture and everted rat gut sacs. Polyphenol Communications 2000, FreisingWeihenstephan Germany, Sept 11-15 2000:401402.
57. Gee JM, DuPont MS, Day AJ, Plumb GW, Williamson G, Johnson IT. Intestinal
transport of quercetin glycosides in the rat involves both deglycosylation and
interaction with the hexose transport pathway. J Nutr 2000; 130:27652771.
58. Gee JM, DuPont MS, Rhodes MJC, Johnson IT. Quercetin glucosides interact with
the intestinal glucose transport pathway. Free Radic Biol Med 1998; 25:1925.
59. Ader P, Block M, Pietzsch S, Wolffram S. Interaction of quercetin glucosides with the
intestinal sodium/glucose co-transporter (SGLT-1). Cancer Lett 2001; 162:175180.
60. Lostao MP, Hirayama BA, Loo DDF, Wright EM. Phenylglucosides and the Na+/
glucose cotransporter (SGLT1)- analysis of interactions. J Membr Biol 1994;
142:161170.
61. Hirayama BA, Lostao MP, PanayotovaHeiermann M, Loo DDF, Turk E, Wright
EM. Kinetic and specificity differences between rat, human, and rabbit Na+-glucose
cotransporters (SGLT-1). Am J Physiol 1996; 33:G919G926.
62. Spencer JPE, Chowrimootoo G, Choudhury R, Debnam ES, Srai SK, Rice-Evans C.
The small intestine can both absorb and glucuronidate luminal avonoids. FEBS
Lett 1999; 458:224230.
63. Crespy V, Morand C, Besson C, Manach C, Demigne C, Remesy C. Comparison of
the intestinal absorption of quercetin, phloretin and their glucosides in rats. J Nutr
2001; 131:21092114.
64. Andlauer W, Stumpf C and Furst P Intestinal absorption of rutin in free and
conjugated forms. Biochem Pharmacol 2001; 62: 369374.
65. Mizuma T, Awazu S. Inhibitory effect of phloridzin and phloretin on
glucuronidation of p-nitrophenol, acetaminophen and 1-naphthol: kinetic demonstration of the inuence of glucuronidation metabolism on intestinal absorption in
rats. Biochim Biophys Acta 1998; 1425:398404.
16
Bioavailability of Flavanol Monomers
Jennifer L. Donovan
Medical University of South Carolina
Andrew L. Waterhouse
University of California, Davis
Davis, California, U.S.A.
I.
INTRODUCTION
Flavanols are one of the most abundant classes of avonoids, often referred to as
avan-3-ols or catechins. They are present as monomers, oligomers, and
polymers and are often esteried with gallic acid [1]. The focus of this chapter
are bioavailability and metabolism of the avanol monomers, catechin, epicatechin, and the green tea avanols epigallocatechin gallate, epigallocatechin, and
epicatechin gallate (Fig. 1).
Flavanols are abundant components of many foods and beverages. Red
wine, apples, tea, and chocolate are among the richest food sources [25]. The
intake of the avanol monomers was determined to be 50 mg/day in a Dutch
cohort with tea, chocolate, apples, and pears as the main sources [6]. The dietary
intake of avanols is likely to vary greatly among individuals and cultures.
Cultures that consume green tea and red wine by custom such as France or Japan
would be expected to have higher intakes.
Numerous in vivo studies have indicated that avanols have diverse
biological activities including antioxidant activity [711] and anticancer properties [1214]. Flavanols may inhibit platelet aggregation as well as other vascular
and inammatory processes that contribute to disease [1517]. These eects are
thought to be mediated by changes in eicosonoid synthesis after consumption of
avanol-containing foods [9,18]. Flavanol metabolites isolated from urine were
shown to reduce monocyte adhesion to endothelial cells in the inammatory
413
414
Donovan and Waterhouse
Figure 1 Chemical structures of monomeric flavanols.
process of atherosclerosis [7]. Feeding studies in several animal models have also
demonstrated reduced progression of fatty streaks or atherosclerosis [1921].
The study of avanol absorption and metabolism has been carried out for
some time, and there are a number of prior reviews on this subject [2225].
Studies on the absorption and metabolism of monomeric avanols are extensive,
going back to the 1950s, though all early work was conducted on pharmacological doses of catechin [2629]. The metabolic pathways of catechin elucidated
in these studies continue to be the models for other avonoids. They indicated
that the potential pathways of metabolism are glucuronidation and sulfation as
well as methylation, but the plasma and tissue levels of specic metabolites or
their proportions are not reective of those present after consumption of the same
avanols from foods [30].
Research in the last several years has focused on the bioavailability of avanols when consumed in green tea, red wine, and chocolate. Plasma levels of avanols and metabolites as well as some pharmacokinetic parameters have now been
documented in humans after short-term feeding studies with these foods. Additionally, animal and cell culture models as well as in vitro experiments have been
used to increase our understanding of the mechanisms of absorption, metabolism,
and elimination that ultimately regulate the bioavailability of avanols.
II.
MECHANISMS REGULATING THE BIOAVAILABILITY

OF FLAVANOLS
A.
Absorption
Flavanols are an unusual class of avonoids in that they are not present as
glycosides in the diet so a sugar moiety does not play a role in the site or
415
mechanism of absorption, as has been described for other avonoids [3133]

(see Chap. 15). Catechin, epicatechin, epigallocatechin, and epigallocatechin gallate are able to be absorbed in their intact forms and have been identied as
conjugated metabolites in plasma [8,30,34,35]. Epicatechin gallate may not be
absorbed intact as most studies nd negligible amounts in plasma and urine
after tea consumption [34,36]. Epicatechin gallate has been identied in the rat
portal vein as well as in humans after consumption of large doses of the pure
compound [37,38].
1.
Sources of Bioavailable Monomers
In addition to the common dietary sources of avanols [39,40], there may be

indirect routes to obtaining monomers or types of monomers. Studies indicate
that although epigallocatechin gallate has been identied in vivo, cleavage of
the gallate moiety from the avanol ring system may also occur. Amelsvoort
and colleagues [37] showed that 3% of epicatechin gallate and 5% of epigallocatechin gallate were in the hydrolyzed form in plasma after the administration of puried tea catechins to humans. Gallate esters that reach the large
intestine certainly have contact with bacterial esterases, although subsequent
absorption of the hydrolysis products has not been documented. A 1999 study
reported cleavage of epigallocatechin gallate in the mouth. The esterase
responsible was identied in saliva and is thought to be derived from human
epithelial cells (as opposed to oral bacteria), although this hypothesis remains
to be conrmed. The esterase was not reported to be present in human plasma
or liver [41]. Nevertheless, it is possible that at least some epicatechin and
epigallocatechin are produced from their respective gallate esters before
absorption.
Proanthocyanidins have also been investigated as a potential source of
monomeric avanols. In vitro experiments indicate that proanthocyanidin
dimers may be hydrolyzed into monomers by acid present in the stomach
[42]. The degree of polymerization of proanthocyanidins has been reported to
decrease in the rat intestine [43], and an energy-dependent cleavage of dimers
was described in the rat small intestine [44]. Authors have also suggested that
proanthocyanidins are cleaved by microora in the large intestine although
monomers have not been directly detected after incubation with intestinal
bacteria [4547]. Conversely, when rats were fed puried procyanidin B3, a
catechin dimer, no catechin or metabolites could be detected in plasma or urine
[48]. The study also showed that plasma levels of catechin are not dierent when
catechin is consumed alone or along with oligomeric procyanidins. These data
show that in vivo, oligomeric procyanidins are not cleaved into bioavailable
monomers at any point during the digestive process in the rat. Future experiments must be performed to determine whether proanthocyanidins are a source
of bioavailable monomers in humans.
416
2.
Mechanisms of Absorption
Most avanol absorption is thought to occur in the small intestine. The

mechanism of catechin and epicatechin absorption has been investigated by
using intestinal perfusion models in the rat. Both catechin and epicatechin
appear to be eciently absorbed by the jejunum and ileum. Extensive metabolism also occurs within the small intestine, demonstrating that these avanols
are absorbed into intestinal cells, and not by the paracellular route [49,50].
The eect of dose on catechin absorption was also studied, using an
intestinal perfusion [49]. In that study one-third of the catechin dose was
absorbed at all concentrations ranging from 1 to 100 AM, suggesting that
absorption of catechin by the small intestine is directly proportional to the dose
over a fairly wide range. The data indicate that catechin enters intestinal
epithelial cells by passive diusion, a mechanism that is generally proportional
to the dose. The process is relatively nonspecic but increases with the hydrophobicity of the molecule.
The mechanism of absorption of the gallate esters has not been studied
directly. A 1999 study demonstrated that absorption of epigallocatechin gallate
can begin to occur in the mouth [41]. The contrast between epigallocatechin
gallate and epicatechin gallate is intriguing as the molecules dier by only a
single hydroxyl group. Furthermore, the more hydrophilic molecule, epigallocatechin gallate, is absorbed, whereas epicatechin gallate appears not to be.
Thus, passive diusion is not the sole determinant of uptakea more specic
process must be involved. A plausible explanation is that a transporter with
specicity for one of these gallates is involved in their uptake or presystemic
elimination. Epicatechin gallate could also be completely hydrolyzed by an
esterase before reaching the circulating blood.
The avanols that enter epithelial cells likely undergo extensive metabolism within the small intestine (discussed later) before being delivered to the
liver and the circulating blood and tissues.
B.
Metabolism
As are most other avonoids, the avanols are metabolized by the phase II
metabolic processes glucuronidation, sulfation, and methylation. Metabolism
appears to be quite ecient as catechin and epicatechin are present exclusively
as metabolites in vivo after doses in foods [30,34]. Free catechins are only present
in plasma after large pharmaceutical doses are administered, suggesting the
metabolic enzymes are saturated [26]. A fraction of the tea catechins epigallocatechingallate and epigallocatechin exist in their native forms after green tea
consumption, the conjugated forms predominate [34]. The position and type of
conjugate certainly are major determinants of biological activity [7] as well as
417
their distribution and clearance in vivo. A general summary of major metabolites

is shown in Table 1. The classes of metabolites and the positions of the specic
conjugates are discussed later.
1.
Methylated Metabolites
Methylation may occur on avanols that contain ortho-hydroxy functional

groups. The position of methylation of catechin and epicatechin has been
identied as the 3- in humans and several other species [30,48,5155] although
small amounts of 4-O-methyl conjugates of catechin and epicatechin have been
described in both human and rat urine [55,56].
Gallocatechins, on the other hand, appear to be methylated predominantly
at the 4-position. After a 1.5-g dose of puried epigallocatechin in humans,
levels of 4-O-methylated metabolites were four- to sixfold higher than levels of
unmethylated forms in plasma [57]. Substitution at the 4-position indicates
dierent substrate specicity for the trihydroxylated functional group.
The gallic acid esters may be methylated on both the avonoid ring
and the gallic acid residue. Methylated metabolites of epigallocatechin gallate
have been identied in bile after a 100-mg dose was administered to the rat.
Methylation could occur at the 3- or 4-position of the avonoid ring and at
the 3- or 4-position of the gallic acid residue. It was also possible for a
single metabolite to be methylated on both the avonoid and the gallic acid
residue [58].
2.
Glucuronidated and Sulfated Metabolites
The majority of avanols that exist in vivo are conjugated with glucuronic acid
or sulfate. Mixed conjugates containing both glucuronide and sulfate have been
described, and methylated metabolites are generally further conjugated with
glucuronide or sulfate [30,52]. The addition of either of these groups results in
metabolites that are negatively charged at physiological pH and are more polar,
facilitating excretion in urine.
The major site of glucuronidation of catechin and epicatechin has been
identied in rats as the 5-position. Okushio and associates [55] and Harada and
colleagues [52] fed rats catechin or epicatechin, either puried or from red
wine or green tea. The major metabolites in urine and bile were identied as
3-O-methyl(epi)catechin 5-O-h-glucuronide and (epi)catechin 5-O-hglucuronide. Both research groups puried the metabolites from rat urine
and positively identied the positions of the methyl and glucuronide moieties
by nuclear Overhauser eect (NOE) and heteronuclear multiple bond connectivity (HMBC) spectroscopy experiments. Figure 2 shows the structure of
the major catechin and epicatechin metabolites in the rat. The position of
glucuronidation of avanols has not been studied in humans.
Epigallocatechin
Epicatechin gallate
Not detected except

after black tea
consumption
Phenylvalerolactones,
phenolic acids;
possible hydrolysis
of gallic acid residue
1228% of dose
3V- or 4V-position;
3- or 4-position of
gallic acid
Mainly sulfate
Epigallocatechin gallate
Metabolites have been identied and measured in human intervention studies [2630], animal studies [3437], and in vitro experiments [5267].
Urinary
excretion
Native
Hydrolysis
Methylated
Epicatechin
Mainly
Not detected except
Glucuronide
glucuronide
after black tea
sulfate
consumption
Mixed conjugates
Glucuronide
5-position
4 -VPosition
Not detected
20% of metabolites; 2030% of
metabolites;
3V-position; also
3V-position also
glucuronidated,
glucuronidated,
sulfated
sulfated
Phenylvalerolactones, Phenylvalerolactones, Phenylvalerolactones, Phenylvalerolactones,
phenolic acids
phenolic acids
phenolic acids
phenolic acids;
possible hydrolysis
of gallic acid residue
313% of dose
Not detected
Detected only after
Detected only after
consumption of
consumption of
foods that contain
foods that contain
large doses
large doses
310% of dose
57% of dose
24% of dose
Not detected
except after black
tea consumption
Catechin
Glucuronides Mainly mixed

Sulfates
conjugates
Glucuronide
5-position
Metabolites
Table 1 A Summary of Metabolites for Each of the Monomeric Flavanolsa
418
419
Figure 2 Glucuronide conjugates catechin and epicatechin identified in rats; R=H

or CH3. Glucuronidation occurs primarily at the 5-position. (From Refs. 52,55.)
There are no studies that have investigated the position of the sulfate
moiety. Donovan and colleagues [30] reported that 3 -O-methylcatechin was
not sulfated at all after red wine consumption, suggesting that conjugation with
one precludes conjugation with the other. However, catechin metabolites that are
both 3 -O-methylated and sulfated have been identied after larger doses in
humans as well as in several other species [26,53]. Methylated epicatechin has
also been identied as being predominantly in the sulfate form in humans [59].
3.
Ring-Fission Metabolites
Low-molecular-weight metabolites form from ssion of the heterocyclic ring

and the A-ring and then degradation into phenolic acids and lactones. Metabolites of avanol monomers are mainly hydroxylated derivatives of benzoic,
phenylacetic, and phenylpropionic acids. Hippuric acid and its hydroxylated
forms originate from the conjugation of glycine with benzoic acids. Phenylvalerolactones, derived from the B-ring and fragments of the C- and A-rings,
have also been identied. One important aspect of this metabolic pathway is the
deoxygenation of the B-ring, such that the 4 hydroxyl group from the B-ring is
lost. A few studies have also shown that carbon dioxide is formed. Structures of
some ring-ssion metabolites are shown in Figure 3.
Early animal studies on rats, mice, and guinea pigs were used to identify
the metabolites of catechin [60] and established a pathway for some of the steps
illustrated in Fig. 3. Subsequently, a human clinical study that administered 46
of catechin per person showed that these ring-ssion metabolites were absent in
the urine for the rst 12 h but predominated for the next 36 h [27]. They
identied 3-hydroxyphenylpropionic as well as several hydroxylated phenylvalerolactones as the major metabolites of catechin and showed that that the
420
Figure 3 Structures of some ring-fission metabolites of monomeric flavanols.

Phenolic acid metabolites may be conjugated with glycine to form hippuric acids
before excretion in urine. (Pathway adapted from Refs. 60,65.)
majority of these metabolites exist as glucuronide or sulfate conjugates. One

study by Kohri and coworkers [61] identied a specic glucuronic acid
conjugate of a new valerolactone yielded from epigallocatechin. This product
is shown in Figure 4.
A later study with a 2-g dose of catechin reported that the ring-ssion
metabolites accounted for only 10% of catechin excretion in urine over a 24-h
period. Among the products excreted were 3-hydroxybenzoic, 3-hydroxy hippuric acid, and 3-hydroxyphenylpropionic acid [26]. However, no valerolactones
were detected, and this study was carried out by using 14C-labeled catechin.
Much of the analysis used high-performance liquid chromatography (HPLC);
some used thin-layer chromatography (TLC). Dierences between these studies
were initially attributed to the dose, although the eect of dose on ring-ssion
metabolites has never been studied directly. However, a number of studies have
shown that the intestinal microora have a profound eect on the production of
ring-ssion products. For instance, Das and Griths were able to suppress their
production in the guinea pig completely after treatment with two antibiotics [62].
A later study with [14C]-catechin showed that antibiotics greatly suppressed
blood levels and urinary elimination of metabolites when the ring ssion was
arrested [63], apparently by interrupting the production of the lactones from the
lower gut. Nearly identical results were obtained in rats with epigallocatechin
421
Figure 4 Glucuronide conjugate of a valerolactone. (Valerolactone originates

from epigallocatechin identified in Ref. 61.)
[61]. A 2001 study with humans also showed that antibiotics drastically reduce
the appearance of the valerolactones in urine after catechin consumption [64].
Meselhy and colleagues [65] incubated epicatechin gallate and other
components of green tea with a human fecal or rat fecal suspension. The authors
identied several new compounds, including the ring-opened and deoxygenated
analog of the valerolactones. In particular they showed a progression of
avonoid ring A ssion and then, after opening the lactone ring, loss of acetate.
Another important reaction they reported was the conversion of the catechol
group to a phenol group by deoxygenation of the phenolic ring.
Few studies have reported levels of ring-ssion metabolites after doses
in common foods. After green tea was administered to humans at high doses
(400 mg total catechins), 4-hydroxy benzoic, 3,4-dihydihydroxybenzoic, and 3methoxy 4-hydroxy benzoic as well as 3-methoxy 4-hydroxyhippuric acid
were identied in plasma and urine, but it is not clear whether the valerolactones were investigated. The total amount of these metabolites accounted
for 15% of the dose in urine [66]. Li and colleagues [67] identied a dihydroxylated and a tri-hydroxylated phenylvalerolactone in human plasma and
urine after ingestion of a single dose of green tea. The metabolites were
presumed to originate from epicatechin and epigallocatechin and were present
at 825 times the concentration of their avonoid precursors in urine. They
accounted for 639% of the doses of epicatechin and epigallocatechin in
urine [67].
So, it appears to be very likely that the dierence between the earlier study
by Das and associates [27] and the 1983 study by Hackett and colleagues [26] is
due to dierent gut microora in the subject populations. This hypothesis is
strengthened by the results in 2000 of Li and coworkers [67], in that study the
variation in the levels of epicatechin and epigallocatechin conjugates excreted in
the urine varied fourfold between subjects, similarly to results of other reports,
but the levels of the valerolactones varied by more than 10 times.
422
Finally, a large proportion of catechin, in particular the A-ring, appears to

be totally catabolized by microbial action to carbon dioxide. Das and Griths
[60] observed 18% of the radioactivity in catechin appear in this form in animal
experiments, and Gott and Griths [63] observed that the proportion of
avonoid carbons released as carbon dioxide dropped from 45% to 6% (over
a 72-h period) when the rats were treated with antibiotics.
The ring-ssion process appears to be caused by microbial action, not
mammalian metabolism. It involves breakdown of the A-ring, possibly releasing
acetate or carbon dioxide. The remaining carbons of the C- and A-rings form a
g-lactone between the 3-hydroxy group and the newly formed carboxylic acid
group at the end of the remaining 5-carbon chain. The lactone can be broken and
suer a loss of one or more acetate 2-carbon fragments. Other important
reactions are the loss of one of the catechol oxygens as well as methylation. In
urine all these substance are present mostly as polar conjugates.
4.
Metabolism in Different Tissues
The phase II enzymes responsible for avanol metabolism are widely

distributed among tissues, including liver, lung, intestine, and kidney. The
levels of the enzymes and their distribution among tissues determine the
pattern of metabolites. Each of the reactions occurs within cells. Glucuronidation occurs on the luminal side of the endoplasmic reticulum by uridine-5Vdiphosphate glucuronosyltransferases (UGTs), a large family of enzymes.
Sulfation and methylation both occur in the cytosol by sulfotransferases
(SULTs), and catechol-O-methyltransferases (COMTs). The specic isoforms
involved have not been identied for avanols. The UGT1A family is thought
to be responsible for glucuronidation of other avonoids [68]. SULT1A1 and
SULT1A2 are implicated in the sulfation of phenol-type substrates and are
also known as the P-forms or thermostable (TS) forms [69]. Future recognition
of the specic isoforms involved in avanol metabolism will aid in our
understanding of interindividual variability and help predict interactions with
other xenobiotics and drugs [7072].
The metabolism of avanols by specic tissues has only been investigated
in rats. The small intestine is thought to be the major organ of glucuronidation of
avanols. Piskula and Terao [73] measured the activity of UGTs in preparations
of rat liver, kidney, lung, intestine, and plasma, using epicatechin as a substrate.
The small intestine had the highest capacity for glucuronidation and had
approximately 10 times the activity of the liver. Perfusion studies in the rat
also indicate that the small intestine is the most important organ of avanol
glucuronidation. Methylation may also occur in the small intestine but the process appears less ecient than glucuronidation [49,50,73]. Sulfation does not
appear to occur signicantly in the rat small intestine, although it should not be
423
discounted in humans, as intestinal SULT activity is characteristically much

higher in humans than in rats [74,75].
Donovan and coworkers [49] showed that glucuronides, and perhaps
methylated glucuronides, produced in the small intestine are able to enter
hepatocytes and become further metabolized therein. The mechanism of uptake
into hepatocytes is still unknown, but uptake of other glucuronides by the liver
has been previously documented [76,77].
Within the liver, SULT activity is predominant, but further methylation
may also occur. Catechin glucuronides were subsequently sulfated, as well as
methylated, in rat liver after in situ perfusion [49]. Piskula and Terao [73] also
reported that SULT activity for epicatechin was present exclusively in liver.
Within hepatocytes, it is also possible that catechin glucuronides are deglucuronidated and then reglucuronidated before reaching systemic circulation [49].
Further metabolism of avanols may occur in peripheral tissues. Piskula
and Terao [73] also showed that methylation was very active in the kidney.
Some UGT activity, although signicantly less than in the intestine, was also
present in kidney, lung, and plasma. A schematic representation of the proposed mechanisms of avanol metabolism is shown in Figure 5. Further
Figure 5 A schematic representation of the possible mechanisms of absorption,

metabolism, and elimination of flavanols. Absorbed avanols enter intestinal
epithelial cells, where they are extensively glucuronidated and sometimes methylated. Some of these metabolites may pass through the liver without actually
entering the hepatocytes. Other glucuronides are able to enter into hepatocytes
and are possibly de-re-glucuronidated. In the cytosol of hepatocytes, metabolites
may be methylated as well as sulfated before entry into circulation or elimination
in bile.
424
methylation and possible glucuronidation may occur in the kidneys, as suggested

by Piskula and Terao [73], although these metabolites are likely to be excreted
in urine thereafter.
C.
Elimination
Flavanols are quickly eliminated from plasma. The elimination half-lives of

avanols are far shorter than reported for the avonols, which are around 24 h
[78]. Half-lives in plasma after food doses range from 2 to 4 h for catechin and
epigallocatechin [30,36]. Epicatechin has a slightly longer half-life (36 h) [36].
The half-lives indicate that most avanols would be cleared from the body in 10
20 h. The half-life for epigallocatechingallate was reported to range from 5 to 7 h,
and so it may exist slightly longer in vivo [36]. Few studies are able to detect
appreciable levels in plasma 24 h after consumption [30,35,36,59].
Three major mechanisms of elimination of avanols have been reported:
elimination by the kidneys in urine, elimination by the liver in bile, and
elimination by the small intestine by an active elux transport mechanism.
The nal fate of most avanols after elimination by the intestine or liver is likely
to be through metabolism by microora in the colon.
1.
Elimination in Bile
Elimination in bile may be the most important mechanism of avanol elimination. After an intravenous injection of 14C-labeled catechin to rats, 3344% of
the dose was excreted in bile during the rst 24 h [28]. The major biliary
metabolites of catechin after oral administration of 10 or 100 mg of catechin to
the rat were two glucuronide conjugates of 3V -O-methylcatechin, which
accounted for 17% of the administered dose [54]. Both catechin and 3V -Omethylcatechin have also been reported to be present as glucuronide conjugates
as well as mixed glucuronide/sulfate conjugates in rat bile after intestinal
perfusion [49].
Elimination in bile has not been studied directly in humans; however,
in general, larger, extensively conjugated metabolites are more likely to be
eliminated in bile. Dierences in the proportions of metabolites in plasma and
urine also indicate that larger metabolites (i.e., glucuronide and mixed
glucuronide/sulfate conjugates) are excreted in bile [79]. Mixed glucuronide
and sulfate metabolites predominated in plasma after wine consumption but
made up a smaller fraction of metabolites in urine [56]. After green tea
consumption epigallocatechin was mainly in the glucuronide form in plasma
but was mostly sulfated in urine [34]. Epigallocatechin gallate is present at
high concentrations in plasma after green tea consumption but not detected in
urine [34,80].
2.
425
Elimination in Urine
After consumption of foods containing avanols 110% of the dose is typically

recovered in urine as forms containing the intact avonoid ring (Table 1). The
recovery of catechin ranged from 3% to 10% of the dose after red wine
consumption [56]. Epicatechin and epigallocatechin recoveries range from 2%
to 7% after green tea consumption [34,36]. One study reports 2530% recovery
of epicatechin after chocolate consumption, much of it in unconjugated form
[59]. The gallate esters are not detected in urine after green tea but small
amounts were detected after black tea consumption [3436].
Excretion of intact avanols in urine cannot be considered a reliable
indicator of absorption as it may be only a minor route of excretion. The
proportion excreted in urine may also vary with dose as well as the experimental
conditions. One study with three dierent doses of green tea catechins reported
that there was no correlation between the amount of epicatechin or epigallocatechin in urine and the areas under the curve (AUCs) in plasma [36].
The ring-ssion products constitute a large proportion of the urinary
metabolites in most, but not all studies [26,51,81]. The presence of these
components appears to be dependent on microbial action in the lower
intestine [62], so their elimination in the urine may be an indicator of gut
microora status.
Ethanol in red wine has been shown to enhance the elimination of
catechin in urine [56]. Twenty percent more catechin was excreted after wine
containing ethanol compared to dealcoholized wine. This study administered
only 12.4 g of ethanol in 120 mL of red wine. The eect of higher amounts of
ethanol may be more dramatic, although this possibility has not been studied.
Ethanol was suggested to enhance the elimination of avanols through a
diuretic eect, which would not be specic to wine. Thus, a similar increase in
urinary excretion may occur after the consumption of other alcoholic beverages along with avanol-containing foods.
3.
Elimination by the Small Intestine
Intestinal excretion is an important mechanism in the elimination of avonoid

conjugates [8284]. This is likely mediated by the multidrug resistance
associated protein (MRP) pumps. MRPs actively export conjugated metabolites
out of the small intestine back into the intestinal lumen and so prevent or
reduce systemic circulation [85]. An experiment using cultured Caco-2 cells
showed that two metabolites of epicatechin were excreted on the apical side of
the cells. Their elimination has been attributed to MRP-2, as elux was
signicantly reduced by a competitive MRP-2 inhibitor [86]. Conversely,
intestinal perfusion experiments with catechin indicate that catechin metabolites are not substrates for these transport proteins [49].
426
D.
Plasma Levels and Pharmacokinetics of Flavanols

in Humans
Human intervention studies are ultimately necessary to determine the bioavailability of avonoids from specic foods. A summary of recent studies including
the doses, maximal plasma concentrations, and urinary excretion is shown in
Table 2. There are large dierences in the reported plasma levels among studies
using seemingly similar food sources. The large interstudy variability cannot be
explained by the doses used but may be explained by dierences in experimental
conditions and analytical methodology.
1.
Analysis of Flavanols in Biological Samples
Early investigations of avanols in biological samples relied on colorimetric

measures of polyphenol content [87] or thin-layer chromatography to assess the
presence of specic metabolites [60,62]. Because it is dicult to separate the
polar conjugates chromatographically, they are typically hydrolyzed to release
the free forms, which are subsequently analyzed. Acid had been used to
hydrolyze the conjugates, but there is ample evidence of avanol degradation
by acid treatment, so enzymatic treatment is essential.
As a result of the low concentration of avanols in blood fractions such as
plasma when food level doses are consumed, there are only a few techniques
that have sucient sensitivity. The rst of these is gas chromatography coupled
to a mass spectrometer for detection [88]. This technique is quite sensitive and
provides the assurance of accurate identication of the analyte when the
molecular and major fragmentation ions are monitored. The drawbacks are that
it is necessary to derivatize the avanols to achieve sucient volatility and the
technique cannot be adapted for analysis of glucuronide and sulfate conjugates.
This method has been used to quantify the plasma levels of catechin and its
methylated derivatives after wine consumption, and a modied version is able to
detect catechin, quercetin, and resveratrol simultaneously [89].
Because of the nonvolatile nature of avanols or their metabolites, liquid
chromatography is ideally suited for their separation. Multielectrode electrochemical detection is a sensitive and selective technique that has been used to
measure avanols and other avonoids in biological samples [48,49,90]. The
method has been adapted to measure specic conjugate forms of avonols
[33,91] and could be adapted for avanol conjugates. Several other approaches
seem to have adequate sensitivity, including uorescence and [92] and chemiluminescence [93]. Both techniques are very selective and have little interference in biological samples. Chemiluminescence is dependent on the presence
of the catechol functional group, and sensitivity is lost after some known
metabolic conversions, such as methylation. Mass spectral detection has both
sensitivity and the generality to allow for the measurement of nearly all
35
164
137
35, 69, 104
220
111
32
82
88
33
38, 75, 113
102, 204, 306
110, 219, 329
33, 66, 99
7
24
33
6
13
52
6
36
146
62
67
124
118
105
225, 375, 525
8, 13, 18
Catechin
Epicatechin
Epicatechin
Epicatechin
Epicatechin
Epicatechin
Epicatechin
EGC
EGCG
ECG
Epicatechin
EGC
EGCG
ECG
Epicatechin
EGC
EGCG
ECG
Catechin
GC
CG
GCG
Epicatechin
EGC
EGCG
ECG
EGCG
EGCG
EGCG
EGC
Green tea
Green tea extract
Green tea extract
Black tea
Canned green tea
Reconstituted
green tea
Red wine
Chocolate
Chocolate
Chocolate
Chocolate
Chocolate
Green tea
Source
50176
700
257
133, 258, 355
4770
36
165275
268673
98572
ND
189, 651, 655
484, 1660, 1797
254, 696, 685
ND
NM
NM
NM
NM
NM
NM
NM
NM
174
145
20
51
5000
135303
657, 4300, 4410
35, 144, 255
Plasma Cmax
(nmol/L)
310
NM
NM
NM
2530
NM
57
34
ND
ND
4.7
2.6
ND
ND
6.1
3.7
ND
ND
1.8
0.4
ND
ND
2.6
3.7
0.14
0.12
NM
NM
NM
Urineb
(% of dose)
[99]
[107]
[100]
[35]
[108]
[36]
[30]
[97]
[8]
[8]
[59]
[9]
[34]
Reference
Reported levels represent the total of free, glucuronidated, and sulfated metabolites, as well as the methylated metabolites if measured. Unno et al. [107] and Nakagawa et al.
[100] report only free forms in plasma. EGC, epigallocatechin; EGCG, epigallocatechin gallate; CG, catechin gallate.
b
ND, not detected; NM, not measured. When several doses are shown, plasma levels are shown in respective order.
Dose
Flavanol
Table 2 A Summary of Human Clinical Studies Reporting Plasma Levels and Urinary Excretion After Consumption of
Flavanols from Foodsa

427
428
metabolites, including unanticipated ones. It would thus appear that liquid

chromatography-mass spectrometry (LC-MS) will be the most important tool in
future studies of avanol metabolism, and recent successes rely on this technology, including the identication and quantication of the new valerolactone
metabolites from epigallocatechin [61,67].
Careful consideration must be given to the analytical methodology
employed when avanols are reported in biological samples. Recovery and
stability experiments in plasma should be reported and hydrolysis techniques
should be veried. Detection and quantitation must be performed with a
method with some specicity for the analyte such as uorescence, multielectrode coulometry, or, ideally, mass spectrometry. Studies reporting very
high levels of free forms without any hydrolysis of conjugates after normal
doses, although potentially correct under certain conditions, must be regarded
with some skepticism.
2.
Pharmacological Doses
Earlier research on avanol absorption and metabolism was performed on

catechin as it was reported to be an eective treatment for viral hepatitis. These
studies used doses that exceed the amounts in foods by a factor of at least 50
[26,27,51,94]. As a result, large amounts of native catechin reported to be
present in plasma were likely due to the saturation of enzymatic pathways
involved in metabolism.
Hackett and coworkers [26] showed that after ingestion of 2 g of puried
14
C-labeled catechin, plasma radioactivity peaked at a level corresponding to
40 Amol/L catechin. Most of the catechin was present as metabolites; 12.5%
was in the native form. Half of the dose was excreted in urine, mainly as 3V O-methylcatechin glucuronide, 3V -O-methylcatechin sulfate, and catechin
glucuronide. Wermeille and colleagues [51] also showed that 60% of catechin
was methylated at the 3V -position and that catechin and 3V -O-methylcatechin
were present as sulfate and/or glucuronide conjugates in human urine after a
1-g dose.
3.
Red Wine
Red wine is a rich source of avanols as well as several other classes of

avonoids [95]. Plasma levels of catechin and metabolites were determined after
a single serving of red wine (120 mL) containing 35 mg of catechin [30].
Maximal levels were achieved at 1 h and ranged from 50 to 176 nM, representing some variability among subjects. Catechin was present mainly as a
mixed glucuronide/sulfate conjugate and to a much lesser extent a sulfate or
glucuronide conjugate. Methylated metabolites accounted for 20% of the total,
and 3-O-methylcatechin existed mostly as a glucuronide conjugate (Fig. 6). The
429
Figure 6 Proportions of the conjugate forms of catechin and 3V-O-methylcatechin

1 h after red wine consumption. (From Ref. 30.) Forms are free (F), sulfated (S),
glucuronidated (G), or glucuronidated and sulfated (GS).
elimination half-life of all metabolites ranged from 2 and 4 h; sulfated

metabolites were eliminated most rapidly. Recovery in urine ranged from 3%
to 10% of the dose and was enhanced by presence of ethanol presence in
wine [56].
4.
Chocolate
Chocolate is a rich source of epicatechin monomer in addition to its high

procyanidin content [96]. Several studies have fed chocolate samples and
determined the total amount of glucuronide and sulfate conjugates in plasma
[810,97]. Epicatechin levels in the experimental chocolate samples ranged
from 35 to 164 mg. Average maximal levels of epicatechin in plasma in these
studies ranged from 36 nM to 700 nM.
The specic forms of epicatechin metabolites as well as methylated forms
were measured after chocolate and cocoa consumption. This study reported
much higher maximal plasma levels (4.8 AM) than all other studies as well as a
2530% recovery rate in urine. Epicatechin was present as a mixed sulfate/
glucuronide, a sulfate conjugate, and a glucuronide conjugate. Methylated
epicatechin accounted for up to 40% of the total amount of metabolites and
was present as a sulfate or a mixed sulfate/glucuronide conjugate [59].
Multiple doses of chocolate were administered in a crossover study by
Wang and coworkers [10]. Subjects consumed chocolate samples containing 27,
53, or 80 mg of epicatechin. The total amount of glucuronide and sulfate conjugates in plasma were reported. Maximal plasma levels and the area under the
curve (AUC) increased proportionally after the three doses. The authors also
430
suggest that food (bread) delays the absorption of epicatechin but does not
signicantly aect the total amount absorbed, as reected by the AUCs.
5.
Green and Black Tea
Tea is the main dietary source of the gallic acid esters of avanols. Tea also
contains signicant amounts of gallocatechins (tri-hydroxylated B-ring). Green
tea contains higher levels of avanol monomers than black tea, which also
contains higher-molecular-weight products of avanol oxidation [98].
Although tea contains epicatechin monomer, it should be considered separately
in terms of bioavailability, as gallic acid esters may be hydrolyzed to form
epicatechin or epigallocatechin.
Flavanols from green tea have been studied most extensively of any
avanols (Table 2). Maximal plasma levels of epigallocatechin gallate, the
major avonoid in green tea, generally range from 100 to 600 nM after green tea
consumption; one study reports 5 AM epigallocatechin gallate after similar doses
[99]. Epicatechin generally ranges from 100 to 300 nM and epigallocatechin
from 200 to 700 nM. As discussed earlier, epicatechin gallate is never detected
in plasma after green tea consumption [34,36].
Only one study reported the dierent types of conjugate forms after tea
consumption [34]. Epicatechin was exclusively in the conjugated form: twothirds in the sulfate form and one-third in the glucuronide form. Epigallocatechin, conversely, was present mostly in the glucuronide form, followed by the
sulfate. As much as 313% was present in the free form in plasma. Epigallocatechin gallate was present mainly in the sulfate form, and the free form
accounted for 1228%. Mixed conjugates and methylated metabolites were not
identied in that study.
Three separate studies have investigated the eect of dose on plasma
levels of tea catechins. One reported that at the highest level, no further increase
in total plasma metabolites was observed [36]. Nakagawa and colleagues [100]
reported a similar trend, although that study measured only the native forms in
plasma. A third study, using a mixture of semipuried tea polyphenols, concluded the opposite and reported that AUCs at the highest doses were proportionally higher. The study also showed that the percentage of free EGCG
increased with increasing dose [80].
A 2001 experiment administered black tea four times during a 6-h period
[35]. Each cup contained approximately four times the amount of monomeric
avanols in a typical brew, and so subjects consumed the equivalent of 12 cups
of tea. Maximal levels of epicatechin, epigallocatechin, and epigallocatechin
gallate were achieved in 58 h. Levels of epigallocatechin gallate were
markedly lower than has been observed after consumption of green tea, even
considering the lower dose. Another dierence between the two types of teas
431
was that after consumption of black tea, epicatechin gallate was detected in
plasma. Epicatechin gallate level was reported to increase linearly in plasma for
24 h after consumption. The reason for its appearance in plasma after consumption of only black tea as well as its curious kinetic properties remain
unclear but may well be related to microbial action in the lower intestine. Small
amounts of both gallate esters were also detected in urine. The specic conjugate
forms of avanols and the methylated forms were not studied. As expected, only
small percentages of the initial doses were recovered in urine and less than 1%
of the dose could be recovered in feces, indicating that gut microora have
metabolized a large percentage of the ingested dose.
Flavanols have also been measured in plasma after green and black tea
consumption by using colorimetric detection after extraction and complexation
with 4-dimethylaminocinnamaldehyde (DMACA) [101]. This method is specic
for avanols, but it is unknown whether there are interferences in biological
samples. Additionally, the reaction does not occur if the A-ring is conjugated
[102] and the authors did not attempt to hydrolyze the conjugate forms before to
extraction. Nevertheless, the reported levels increased twice as much after green
tea compared to black tea consumption. The addition of milk had little eect on
plasma avanols, in agreement with an earlier study using similar methodology
[103]. The authors also suggested, on the basis of their assay, that approximately
half of the catechins in plasma were associated with lipoproteins, mainly highdensity lipoprotein (HDL) [101].
III.
CONCLUSIONS
The avanol monomers catechin, epicatechin, and epigallocatechin gallate are

absorbed in humans and animals. Epicatechin gallate is generally not identied
in vivo but may be present at low concentrations after black tea consumption.
Catechin and epicatechin are not present in their native forms but as combinations of methylated, glucuronidated, and sulfated metabolites. Epigallocatechin
and epigallocatechin gallate exist to some extent in their native forms but are
also mainly present as conjugates. The total amount of metabolites is generally
between 100 nM and 1 AM in plasma after consumption of foods very rich in
avanol content.
The conjugated forms of avanols are eliminated rather quickly with
reported half-lives ranging from 2 to7 h after consumption of wine, tea, or
chocolate. This indicates that dosage must occur several times per day to maintain
reasonable steady-state plasma concentrations throughout the day. The elimination of avanols is expected to be nearly complete after an overnight fast.
The ring-ssion products appear to persist for a much longer time, and thus
if these components have important physiological roles, the eect may be
432
persistent for about 24 h after consumption of foods high in avanols. Much

work remains to be done to characterize levels of these compounds in circulating
blood and other tissues, but the presence of these substances may be responsible
for the increase in the capacity of human plasma or LDL to resist oxidation when
wine, tea, or chocolate is ingested [8,19,104106], an increase that cannot be
justied by the increased level of the intact avonoids [66].
Mechanisms that regulate the bioavailability of avanols have been
further claried in the last several years. The general pathways of metabolismglucuronidation, sulfation, and methylationhave been identied, and
the importance of the small intestine and the liver in metabolism has been
recognized. Future identication of the isoforms of the enzymes involved in
avanol metabolism will be a key to predicting interactions with drugs and
xenobiotics as well as understanding the interindividual variability in levels of
avanol metabolites.
The avanol metabolites that exist in vivo need to be characterized further
as they are the major determinants of biological activity. Signicant amounts of
ring-ssion metabolites are likely present in vivo, but many of the specic
structures and levels in humans are not known. Additional information is also
needed on the positions of the glucuronide and sulfate moieties in human
metabolism. Determining the identities of specic metabolites and their distribution within tissues and cells will be a key to accessing their biological activity,
the major task at hand.
REFERENCES
1. Haslam E. Practical Polyphenolics: From Structure to Molecular Recognition and
Physiological Action. Cambridge: Cambridge University Press, 1998:422.
2. de Pascual-Teresa S, Santos-Buelga C, Rivas-Gonzalo J. Quantitative analysis of
avan-3-ols in Spanish foodstus and beverages. J Agric Food Chem 2000;
48(11):53315337.
3. Guyot S, Marnet N, Laraba D, Sanoner P, Drilleau J-F. Reversed-phase HPLC
following thiolysis for quantitative estimation and characterization of the four
main classes of phenolic compounds in dierent tissue zones of a French cider
apple variety (malus domestica var. Kermerrien). J Agric Food Chem 1998;
46(5):16981705.
4. Hammerstone JF, Lazarus SA, Mitchell AE, Rucker R, Schmitz HH. Identication
of procyanidins in cocoa (theobroma cacao) and chocolate using high-performance liquid chromatography mass spectrometry. J Agric Food Chem 1999;
47(2):490496.
5. Hammerstone JF, Lazarus SA, Schmitz HH. Proanthocyanidin content and
variation in some commonly consumed foods. J Nutr 2000; 130 (suppl 8):
2086S2092S.

6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
433
Arts I, Hollman P, Feskens E, de Mesquita H, Kromhout D. Catechin intake and

associated dietary and lifestyle factors in a representative sample of Dutch men
and women. Eur J Clin Nutr 2001; 55(2):7681.
Koga T, Meydani M. Eect of plasma metaboites of (+)-catechin and quercetin on
73(5):941948.
Rein D, Lotito S, Holt RR, Keen CL, Schmitz HH, Fraga CG. Epicatechin in
human plasma: in vivo determination and eect of chocolate consumption on
plasma oxidation status. J Nutr 2000; 130 (suppl 8):21092114.
Wan Y, Vinson JA, Etherton TD, Proch J, Lazarus SA, Kris-Etherton PM. Eects
of cocoa powder and dark chocolate on LDL oxidative susceptibility and prostaglandin concentrations in humans. Am J Clin Nutr 2001; 74(5):596602.
Wang JF, Schramm DD, Holt RR, Ensunsa JL, Fraga CG, Schmitz HH, Keen CL.
A dose-response eect from chocolate consumption on plasma epicatechin and
oxidative damage. J Nutr 2000; 130 (suppl 8):2115S2119S.
Whitehead TP, Robinson D, Allaway S, Syms J, Hale A. Eect of red wine
ingestion on the antioxidant capacity of serum. Clin Chem 1995; 41(1):3235.
Cliord AJ, Ebeler SE, Ebeler JD, Bills ND, Hinrichs SH, Teissedre PL,
Waterhouse AL. Delayed tumor onset in transgenic mice fed an amino acidbased diet supplemented with red wine solids. Am J Clin Nutr 1996;
64(5):748756.
Liao SS, Umekita Y, Guo JT, Kokontis JM, Hiipakka RA. Growth inhibition and
regression of human prostate and breast tumors in athymic mice by tea
epigallocatechin gallate. Cancer Lett 1995; 96(2):239243.
Yin PZ, Zhao JY, Cheng SJ, Hara Y, Zhu QF, Liu ZG. Experimental studies of
the inhibitory eects of green tea catechin on mice large intestinal cancers induced
by 1,2-dimethylhydrazine. Cancer Lett 1994; 79(1):3338.
Keevil J, Osman H, Reed J, Folts J. Grape juice, but not orange juice or grapefruit juice, inhibits human platelet aggregation. J Nutr 2000; 130(1):5356.
Putter M, Grotemeyer KH, Wurthwein G, Araghi-Niknam M, Watson RR,
Hosseini S, Rohdewald P. Inhibition of smoking-induced platelet aggregation by
aspirin and pycnogenol. Thromb Res 1999; 95(4):155161.
Rein D, Paglieroni TG, Pearson DA, Wun T, Schmitz HH, Gosselin R, Keen CL.
Cocoa and wine polyphenols modulate platelet activation and function. J Nutr
2000; 130 (suppl 8):21202126.
Schramm DD, Wang JF, Holt RR, Ensunsa JL, Gonsalves JL, Lazarus SA,
Schmitz HH, German JB, Keen CL. Chocolate procyanidins decrease the
leukotriene-prostacyclin ratio in humans and human aortic endothelial cells. Am
J Clin Nutr 2001; 73(1):3640.
Hayek T, Fuhrman B, Vaya J, Rosenblat M, Belinky P, Coleman R, Elis A,
Aviram M. Reduced progression of atherosclerosis in apolipoprotein E-decient
mice following consumption of red wine, or its polyphenols quercetin or catechin,
is associated with reduced susceptibility of LDL to oxidation and aggregation.
Arterioscler Thromb Vasc Biol 1997; 17(11):27442752.
Xu R, Yokoyama WH, Irving D, Rein D, Walzem R, German JB. Eect of dietary
434
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.

catechin and vitamin E on aortic fatty streak development in hypercholesterolemic
hamsters. Atherosclerosis 1998; 137(1):2936.
Yamakoshi J, Kataoka S, Koga T, Ariga T. Proanthocyanidin-rich extract from
grape seeds attenuates the developement of aortic atherosclerosis in cholesterolfed rabbits. Atherosclerosis 1999; 142(1):139149.
Hollman PC, Tijburg LB, Yang CS. Bioavailability of avonoids from tea. Crit
Rev Food Sci Nutr 1997; 37(8):719738.
Scalbert A, Williamson G. Dietary intake and bioavailability of polyphenols. J Nutr
2000; 130 (suppl 8):2073S2085S.
Hackett AM. The metabolism of avonoid compounds in mammals In: Cody V,
Middleton EJ, Harborne JB, eds. Plant Flavonoids in Biology and Medicine:
Biochemical, Pharmacological, and Structure-Activity Relationships. New York:
Alan R. Liss 1986:177194.
Hollman PCH, Arts ICW. Flavonols, avones and avanolsnature, occurrence
and dietary burden. J Sci Food Agric 2000; 80(7):10811093.
Hackett AM, Griths LA, Broillet A, Wermeille M. The metabolism and
excretion of (+)-[14C]cyanidol-3 in man following oral administration. Xenobiotica 1983; 13(5):279286.
Das NP. Studies on avonoid metabolism. Absorption and metabolism of (+)catechin in man. Biochem Pharmacol 1971; 20(12):34353445.
Das NP, Sothy SP. Studies on avonoid metabolism: biliary and urinary excretion
of metabolites of (+)-[U-14C]catechin. Biochem J 1971; 125(2):417423.
Watanabe H. The chemical structure of the intermediate metabolites of catechin
I-IV. Bull Agric Chem Soc Jpn 1959; 23(25):257271.
Donovan JL, Bell JR, Kasim-Karakas S, German JB, Walzem RL, Hansen RJ,
Waterhouse AL. Catechin is present as metabolites in human plasma after
consumption of red wine. J Nutr 1999; 129(9):16621668.
Hollman PCH, Devries JHM, Vanleeuwen SD, Mengelers MJB, Katan MB.
volunteers. Am J Clin Nutr 1995; 62(6):12761282.
Hollman PC, Bijsman MN, van Gameren Y, Cnossen EP, de Vries JH, Katan MB.
The sugar moiety is a major determinant of the absorption of dietary avonoid
glycosides in man. Free Radic Res 1999; 31(6):569573.
Morand C, Manach C, Crespy V, Remesy C. Quercetin-3-O-h-glucoside is better
absorbed than other quercetin forms and is not present in rat plasma. Free Radic
Res 2000; 33(5):667676.
Lee M-J, Wang Z-Y, Li H, Chen L, Sun Y, Gobbo S, Balentine DA, Yang CS.
Analysis of plasma and urinary tea polyphenols in human subjects. Cancer
Epidemiol Biomarkers Prev 1995; 4(4):393399.
Warden B, Smith L, Beecher G, Balentine D, Clevidence B. Catechins are
bioavailable in men and women drinking black tea throughout the day. J Nutr
2001; 131(6):17311737.
Yang C, Chen L, Lee M, Balentine D, Kuo M, Schantz S. Blood and urine levels of
tea catechins after ingestion of dierent amounts of green tea by human
volunteers. Cancer Epidemiol Biomarkers Prev 1998; 7(4):351354.

37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
435
Amelsvoort JMM, van het Hof KH, Mathot JNJJ, Mulder TPJ, Wiersma A,
Tijburg LBM. Plasma concentrations of individual tea catechins after a single
oral dose in humans. Xemobiotica 2001; 31(12):891901.
Okushio K, Matsumoto N, Kohri T, Suzuki M, Nanjo F, Hara Y. Absorption of tea
catechins into rat portal vein. Biol Pharm Bull 1996; 19(2):326329.
Arts I, van de Putte B, Hollman P. Catechin contents of foods commonly
consumed in the Netherlands. 1. Fruits, vegetables, staple foods, and processed
foods. J Agric Food Chem 2000; 48(5):17461751.
Arts I, van de Putte B, Hollman P. Catechin contents of foods commonly
consumed in the Netherlands. 2. Tea, wine, fruit juices, and chocolate milk. J Agric
Food Chem 2000; 48(5):17521757.
Yang C, Lee M, Chen L. Human salivary tea catechin levels and catechin esterase
activities: implication in human cancer prevention studies. Cancer Epidemiol
Biomark Prev 1999; 8(1):8389.
Spencer JP, Chaudry F, Pannala AS, Srai SK, Debnam E, Rice-Evans C.
Decomposition of cocoa procyanidins in the gastric milieu. Biochem Biophys Res
Commun 2000; 272(1):236241.
Abia R, Fry SC. Degradation and metabolism of 14c-labelled proanthocyanidins
from carob (ceratonia siliqua) pods in the gastrointestinal tract of the rat. J Sci
Food Agric 2001; 81(12):11561165.
Spencer JPE, Schroeter H, Shenoy B, Srai KS, Debnam ES, Rice-Evans C.
Epicatechin is the primary bioavailable form of the procyanidin dimers B2 and B5
after transfer across the small intestine. Biochem Biophys Res Commun 2001;
285(3):558593.
Deprez S, Brezillon C, Rabot S, Philippe C, Mila I, Lapierre C, Scalbert A.
Polymeric proanthocyanidins are catabolized by a human colonic microora into
low molecular weight phenolic acids. J Nutr 2000; 130(11):27332738.
Groenewoud G, Hundt HKL. The microbial metabolism of condensed (+)catechins by rat-caecal microora. Xenobiotica 1986; 16(2):99107.
Groenewoud G, Hundt HKL. The microbial metabolism of (+)-catechin to two
novel diarylpropan-2-ol metabolites in vitro. Xenobiotica 1984; 14(9):711717.
Donovan JL, Manach C, Rios L, Morand C, Scalbert A, Remesy C. Procyanidins
are not bioavailable in rats fed a single meal containing a grapeseed extract or the
procyanidin dimer B3. Br J Nutr 2002; 87(4):299306.
Catechin is metabolized by both the small intestine and liver in rats. J Nutr 2001;
131(6):17531757.
Kuhnle G, Spencer JPE, Schroeter H, Shenoy B, Debnam ES, Srai KS, Rice-Evans
small intestine. Biochem Biophys Res Commun 2000; 277(2):507512.
Wermeille M, Turin E, Griths LA. Identication of the major urinary metabolites
of (+)-catechin and 3-O-methyl-(+)-catechin in man. Eur J Drug Metab
Pharmacokinet 1983; 8(1):7784.
Harada M, Kan Y, Naoki H, Fukui Y, Kageyama N, Nakai M, Miki W, Kiso Y.
Identication of the major antioxidative metabolites in biological uids of the rat
436
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.

with ingested (+)-catechin and ()-epicatechin. Biosci Biotechnol Biochem 1999;
63(6):973977.
Hackett AM, Shaw IC, Griths LA. 3V-O-methyl-(+)-catechin glucuronide and 3VO-methyl-(+)-catechin sulphate: New urinary metabolites of (+)-catechin in the rat
and the marmoset. Experimentia 1982; 38(5):538540.
Shaw IC, Griths LA. Identication of the major biliary metabolite of (+)catechin in the rat. Xenobiotica 1980; 10(112):905911.
Okushio K, Suzuki M, Matsumoto N, Nanjo F, Hara Y. Identication of ()epicatechin metabolites and their metabolic fate in the rat. Drug Metab Dispos
1999; 27(2):309316.
Donovan JL, Kasim-Karakas S, German JB, Waterhouse AL. Urinary excretion of
catechin metabolites by human subjects after red wine consumption. Br J Nutr
2002; 87(1):3137.
Meng X, Lee M, Li C, Sheng S, Zhu N, Sang S, Ho C, Yang C. Formation and
identication of 4V-O-methyl-()-epigallocatechin in humans. Drug Metab Dispos
2001; 29(6):789793.
Kida K, Suzuki M, Matsumoto N, Nanjo F, Hara Y. Identication of biliary
metabolites of ()-epigallocatechin gallate in rats. J Agric Food Chem 2000;
48(9):41514155.
Baba S, Osakabe N, Yasuda A, Natsume M, Takizawa T, Nakamura T, Terao J.
Bioavailability of ()-epicatechin upon intake of chocolate and cocoa in human
volunteers. Free Radic Res 2000; 33(5):635641.
Das NP, Griths LA. Studies on avonoid metabolism: metabolism of (+)[14C]catechin in the rat and guinea pig. Biochem J 1969; 115(4):831836.
Kohri T, Matsumoto N, Yamakama M, Suzuki M, Nanjo F, Hara Y, Oku N.
Metabolic fate of ()-[4-3H] epigallocatechin gallate in rats after oral administration. J Agric Food Chem 2001; 49(8):41024112.
Das NP, Griths LA. Studies on avonoid metabolism: metabolism of (+)catechin in the guinea pig. Biochem J 1968; 110(3):449456.
Gott D, Griths L. Eects of antibiotic pretreatments on the metabolism and
excretion of [U14C](+)-catechin [(U14C](+)-cyanidanol-3) and its metabolite, 3VO-methyl-(+)-catechin. Xenobiotica 1987; 17(4):423434.
Li C, Meng X, Winnik B, Lee M, Lu H, Sheng S, Buckley B, Yang C. Analysis of
urinary metabolites of tea catechins by liquid chromatography/electrospray
ionization mass spectrometry. Chem Res Toxicol 2001; 14(6):702707.
Meselhy MR, Nakamura N, Hattori M. Biotransformation of ()-epicatechin 3-Ogallate by human intestinal bacteria. Chem Pharm Bull 1997; 45(5):888893.
Pietta P, Simonetti P, Gardana C, Brusamolino A, Morazzoni P, Bombardelli E.
Catechin metabolites after intake of green tea infusions. Biofactors 1998; 8(1
2):111118.
Li C, Lee M, Sheng S, Meng X, Prabhu S, Winnik B, Huang B, Chung J, Yan S,
Ho C, Yang C. Structural identication of two metabolites of catechins and their
kinetics in human urine and blood after tea ingestion. Chem Res Toxicol 2000;
13(3):177184.
Cheng Z, Radominska-Pandya A, Tephly TR. Studies on the substrate specicity
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
437
of human intestinal UDP-glucuronosyltransferases 1A8 and 1A10. Drug Metab

Dispos 1999; 27(10):11651170.
Ghazali RA, Waring RH. The eects of avonoids on human phenolsulphotransferases: potential in drug metabolism and chemoprevention. Life Sci 1999;
65(16):16251632.
Galijatovic A, Walle UK, Walle T. Induction of UDP-glucuronosyltransferase
by the avonoids chrysin and quercetin in Caco-2 cells. Pharm Res 2000; 17(1): 21
26.
Eaton EA, Walle UK, Lewis AJ, Hudson T, Wilson AA, Walle T. Flavonoids,
potent inhibitors of the human p-form phenolsulfotransferasepotential role in
drug metabolism and chemoprevention. Drug Metab Dispos 1996; 24(2):232237.
Raftogianis RB, Wood TC, Weinshilboum RM. Human phenol sulfotransferases
sult1A2 and sult1A1: genetic polymorphisms, allozyme properties, and human liver
genotype-phenotype correlations. Biochem Pharmacol 1999; 58(4):605616.
Piskula MK, Terao J. Accumulation of ()-epicatechin metabolites in rat plasma
after oral administration and distribution of conjugation enzymes in rat tissues.
J Nutr 1998; 128(7):11721178.
Pacici G, Franchi M, Bencini C, Repetti F, DiLascio N, Murano GB. Tissue
distribution of drug-metabolizing enzymes in humans. Xenobiotica 1988;
18(7):849856.
Dunn RTI, Klaassen CD. Tissue-specic expression of rat sulfotransferase
messenger RNAs. Drug Metab Dispos 1998; 26(6):598604.
Meier PJ, Eckhardt U, Schroeder A, Hagenbuch B, Steiger B. Substrate specicity
of sinusoidal bile acid and organic anion uptake systems in rat and human liver.
Hepatology 1997; 26(6):16671677.
Csala M, Banhegyi G, Braun L, Szirmai R, Burchell A, Burchell B, Benedetti A,
Mandl J. h-glucuronidase latency in isolated murine hematocytes. Biochem
Pharmacol, 2000; 59(7):801805.
Hollman PCH, van Trijp JMP, Buysman MNCP, Gaag MSvd, Mengelers MJB, de
Vries JHM, Katan MB. Relative bioavailability of the antioxidant avonoid
quercetin from various foods in man. FEBS Lett 1997; 418(12):152156.
Rozman KK, Klaassen CD. Absorption, distribution, and excretion of toxicants. In:
Klaassen CD, Amdur MO, Doull J, eds. Casarett and Doulls Toxicology. New
York: McGraw-Hill, 1996:91112.
Chow H-HS, Cai Y, Albert DS, Hakim I, Dorr R, Shahi F, Crowell J, Yang C, Hara
Y. Phase I pharmacokinetic study of tea polyphenols following single-dose
administration of epigallocatechin gallate and polyphenon E. Cancer Epidemiol
Biomark Prev 2001; 10(1):5358.
Griths LA. Studies on avonoid metabolism: identication of the metabolites of
(+)-catechin in rat urine. Biochem J 1964; 92(1):173179.
Crespy V, Morand C, Manach C, Besson C, Demigne C, Remesy C. Part of
quercetin absorbed in the small intestine is conjugated and further secreted in the
intestinal lumen. Am J Physiol 1999; 277(pt 1):G120G126.
Walgren RA, Karnaky KJ Jr, Lindenmayer GE, Walle T. Elux of dietary
avonoid quercetin 4V-h-glucoside across human intestinal Caco-2 cell monolayers
438
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.

by apical multidrug resistance-associated protein-2. J Pharmacol Exp Ther 2000;
294(3):830836.
Walle UK, Galijatovic A, Walle T. Transport of the avonoid chrysin and its
conjugated metabolites by the human intestinal cell line Caco-2. Biochem
Pharmacol 1999; 58(3):431438.
Borst P, Evers R, Kool M, Wijnholds J. The multi-drug resistance protein family.
Biochem Biophys Acta 1999; 1461(2):347357.
Vaidyanathan JB, Walle T. Transport and metabolism of the tea avonoid ()epicatechin by the human intestinal cell line Caco-2. Pharmaceut Res 2001;
18(10):14201425.
Bray HG, Thorpe WV. Analysis of phenolic compounds of interest in metabolism.
Methods Biochem Anal 1954; 1(1):2752.
Luthria DL, Jones AD, Donovan JL, Waterhouse AL. GC-MS determination of
catechin and epicatechin levels in human plasma. J High Resol Chromatogr 1997;
20(11):621623.
Soleas GJ, Yan J, Goldberg DM. Measurement of trans-resveratrol, (+)- catechin,
and quercetin in rat and human blood and urine by gas chromatography with mass
selective detection. Methods Enzymol 2001; 335(1):130145.
Manach C, Texier O, Morand C, Crespy V, Regerat F, Demigne C, Remesy C.
Comparison of the bioavailability of quercetin and catechin in rats. Free Radic
Biol Med 1999; 27(1112):12591266.
Sesink A, OLeary K, Hollman P. Quercetin glucuronides but not glycosides are
present in human plasma after consumption of quercetin-3-glucoside or quercetin4V-glucoside. J Nutr 2001; 131(7):19381941.
Donovan JL, Luthria DL, Stremple P, Waterhouse AL. Analysis of (+)- catechin,
()-epicatechin and their 3V- and 4V-O-methylated analogs: a comparison of
sensitive methods. J Chromatogr B 1999; 726(12):277283.
Nakagawa K, Miyazawa T. Chemiluminescence high-performance liquid chromatographic determination of tea catechin, ()-epigallocatechin 3-gallate, at
picomole levels in rat and human plasma. Anal Biochem 1997; 248(1):4149.
Giles AR, Gumma A. Biopharmaceutical evaluation of cyanidanol tablets using
pharmacokinetic techniques. Arzneim Forsch 1973; 23(1):98100.
Ritchey JG, Waterhouse AL. A standard red wine: monomeric phenolic analysis of
commercial cabernet sauvignon wines. Am J Enol Vitic 1999; 50(1):91100.
Waterhouse AL, Shirley JR, Donovan JL. Antioxidants in chocolate. Lancet 1996;
348(9030):834.
Richelle M, Tavazzi I, Enslen M, Oord EA. Plasma kinetics in man of
epicatechin from black chocolate. Eur J Clin Nutr 1999; 53(1):2226.
Balentine DA. Manufacturing and chemistry of tea. In: Ho C-T, Lee CY and
Huang M-T, eds. Phenolic Compounds in Foods and Their Eects on Health 1.
Analysis, Occurrence and Chemistry. Washington, DC: American Chemical
Society, 1992:102117.
Maiani G, Serani M, Salucci M, Azzini E, FerroLuzzi A. Application of a new
high-performance liquid chromatographic method for measuring selected polyphenols in human plasma. J Chromatogr B 1997; 692(2):311317.

100.
101.
102.
103.
104.
105.
106.
107.
108.
439
Nakagawa K, Okuda S, Miyazawa T. Dose-dependent incorporation of tea

catechins, ()-epigallocatechin-3-gallate and ()-epigallocatechin, into human
plasma. Biosci Biotech Biochem 1997; 61(12):19811985.
van het Hof K, Kivits G, Weststrate J, Tijburg L. Bioavailability of catechins from
tea: the eect of milk. Eur J Clin Nutr 1998; 52(5):356359.
Treutter D. Chemical reaction detection of catechins and proanthocyanidins with
4-dimethylaminocinnamaldehyde. J Chromatogr A 1989; 467(1):185193.
He YH, Kies C. Green and black tea consumption by humans: impact on
polyphenol concentrations in feces, blood and urine. Plant Foods Hum Nutr 1994;
46(3):221229.
Serani M, Maiani G, Ferro-Luzzi A. Alcohol-free red wine enhances plasma
antioxidant capacity in humans. J Nutr 1998; 128(6):10031007.
Serani M, Ghiselli A, Ferro-Luzzi A. In vivo antioxidant eect of green and black
tea in man. Eur J Clin Nutr 1996; 50(1):2832.
Serani M, Ghiselli A, Ferro-Luzzi A. Red wine, tea, and antioxidants. Lancet
1994; 344(8922):626.
Unno T, Kondo K, Itakura H, Takeo T. Analysis of ()-epigallocatechin gallate in
human serum obtained after ingesting green tea. Biosci Biotech Biochem 1996;
60(12):20662068.
Yang B, Arai K, Kusu FD. Determination of catechins in human urine subsequent
to tea ingestion by high-performance liquid chromatography with electrochemical
detection. Anal Biochem 2000; 238(1):7782.
17
Absorption and Metabolism
of Hydroxycinnamates
Andreas R. Rechner
London, England
I.
INTRODUCTION
The hydroxycinnamates are ubiquitous in the plant kingdom [1]. Derivatives of

the major hydroxycinnamates, such as p-coumaric acid (4-hydroxycinnamic
acid), caeic acid (3,4-dihydroxycinnamic acid), ferulic acid (4-hydroxy-3methoxycinnamic acid), and sinapic acid (4-hydroxy-3,5-dimethoxycinnamic
acid), are found in fruits, vegetables, grains, and coee [13]. They are also
central compounds to the polyphenol biosynthetic pathway of plants. The
shikimate pathway involves the metabolism of phenylalanine, which is converted to trans-cinnamic acid, followed by a hydroxylation at the 4-position of
the aromatic ring forming 4-hydroxycinnamic acid or p-coumaric acid. Further
hydroxylation at the 3-position yields caeic acid, with subsequent O-methylation of the 3-hydroxy group, resulting in ferulic acid. In fruits and vegetables
the hydroxycinnamates predominantly occur as esters with organic acids, such
as quinic acid, tartaric acid, or malic acid, or with sugars, such as glucose,
whereas glycosides are seldom found [13]. Oligomeric forms, especially of
ferulic acid, are found in grains [2]. The most extensively occurring hydroxycinnamic acid derivatives are esters of caeic acid with quinic acid, predominantly chlorogenic acid (IUPAC name: 5-caeoyl quinic acid) [4], the main
constituent in coee, apple juice, artichoke, eggplant, and peach. The 3caeoyl quinic acid ester is present as a major component of cherry, plum,
elderberry, and apricot, 4-caeoyl quinic acid and various dicaeoyl quinic
acids are minor constituents of some fruits and vegetables [13].
441
442
Rechner
The dietary intake of hydroxycinnamates is thought to range from 25 mg/

day to 1 g/day (predominantly caeic acid), comparing noncoee drinkers who
consume few fruits and vegetables with coee drinkers who have a high intake of
fruits and vegetables [3]. Caeic acid conjugates usually dominate the daily
dietary intake of hydroxycinnamates from the sources mentioned, but brancontaining food contains high levels of ferulic acid conjugates [3].
II.
ABSORPTION AND METABOLISM
The absorption, metabolism, and elimination of hydroxycinnamates, either from

dietary sources or in form of the pure compound, have been studied in humans
and animals. The pioneering work of Booth and DeEds in the 1950s and 1960s
in animals and humans largely contributed to the understanding of hydroxycinnamate metabolism. In oral administration studies in rats, rabbits, and
humans, as well as in in vitro experiments with isolated rat or rabbit liver, a
number of metabolic events have been identied. These include the cleavage of
ester bonds, O-methylation via catechol O-methyl transferase, dehydroxylation,
reduction of the residual double bond, h-oxidation, decarboxylation, and
glycination [516]. In nature as well as in the diet derivatives of p-coumaric,
caeic, ferulic, and sinapic acids represent the most commonly occurring
hydroxycinnamates. The following sections deal with the metabolic fate of
the free acids and their numerous derivatives, which are the forms of their
natural occurrence.
A.
Caffeic Acid and Its Derivatives
The metabolites and conjugates of caeic acid and caeic derivatives identied in
the dierent studies in humans and animals are summarized in Table 1. The
majority of studies in humans and animals on absorption and metabolism of
hydroxycinnamates have been undertaken with caeic acid and its main natural
conjugate, chlorogenic acid. The possession of a catechol moiety results in
extensive metabolism of caeic acid, which leads to the formation of a number of
metabolites. The metabolic pathways of caeic acid are schematically summarized in Figure 1 [6,16].
In early studies in humans and rats, Booth and associates [6] detected and
identied by paper chromatography 10 metabolites in human urine after the
ingestion of caeic acid, namely, caeic acid itself, ferulic acid, dihydrocaeic
acid, dihydroferulic acid, vanillic acid, m-coumaric acid, 3-(3-hydroxyphenyl)propionic acid, 3-hydroxyhippuric acid, feruloylglycine, and vanilloylglycine.
All identied metabolites appeared until 8 h post ingestion, except 3-hydroxyhippuric acid, which appeared between 8 and 24 h post ingestion [6]. Rates
Absorption and Metabolism of Hydroxycinnamates
443
of conjugation with glucuronic acid were not investigated. Interestingly,

the metabolism of caeic acid in rats diered signicantly, displaying
3-(3-hydroxyphenyl)-propionic acid as the major urinary metabolite. In contrast
to humans, rats also metabolized caeic acid to compounds possessing a
3-carbon side chain with only traces of 1-carbon side chain compounds. The
ingestion of 1 g of chlorogenic acid in humans resulted in the presence of caeic
acid, 3-hydroxyhippuric acid, m-coumaric acid, and dihydroferulic acid in urine,
whereas the ingestion of coee (equivalent to approximately 2 g of chlorogenic
acid) led only to the excretion of 3-hydroxyhippuric acid and m-coumaric acid.
As for caeic acid, a signicant dierence of chlorogenic acid metabolism was
observed between humans and rats: 3-(3-hydroxyphenyl)-propionic acid was the
major urinary metabolite in rats.
The urinary excretion of unchanged caeic acid post h-glucuronidase
treatment was also observed after the apllication of 1 g of pure caeic acid in
humans [17]. In addition, vanillic, ferulic, and isoferulic acids were also identied as key urinary elimination products after h-glucuronidase treatment. The
majority of the identied metabolites were excreted in the rst 4 h post
ingestion, and the total recovery of caeic acid and its identied metabolites
after 48 h was approximately 10% of the ingested dose. The metabolic fate of
the remaining caeic acid could not be determined as no caeic acid was
detected in fecal samples during the post-ingestion period of 48 h. Caeic acid
was also identied as a metabolite of ferulic acid in addition to ferulic acid itself
and two unidentied metabolites in urine post h-glucuronidase treatment. After
ingestion of 1 L of coee, only ferulic acid was identied in urine post hglucuronidase treatment.
In a human study, in which 2 cups of coee were ingested three times at
4-h intervals (149.7 mg hydroxycinnamates per cup equivalent to a total intake
of 898.4 mg hydroxycinnamates relative to chlorogenic acid) after 1 day of a
polyphenol-free diet, the methylated metabolites ferulic, isoferulic, and vanillic
acids were major elimination products detected in urine post h-glucuronidase
treatment [18]. No caeic acid or chlorogenic acid was found. Unconjugated
dihydroferulic acid was also identied, and 3-hydroxyhippuric acid (3-hydroxybenzoylglycine) expressed a strong association between the amounts excreted
and the ingestion of the coee. The latter two compounds were excreted in much
higher amounts than the conjugates ferulic, isoferulic, and vanillic acids but
showed maximal excretion more than 12 h after the rst coee ingestion,
indicating a prolonged metabolic route. The highest amounts of the conjugates
of the other metabolites were excreted 13 h after any of the three coee
ingestions, implying fast absorption, metabolism, and elimination. A representative time prole of the amounts excreted of the identied metabolites is shown
in Fig. 2 [18]. An association between the coee ingestion and the amounts
excreted of hippuric acid (benzoylglycine), a general aromatic metabolite in
444
Rechner
Table 1 Nature of Urinary and Plasma Metabolites of Hydroxycinnamates in

Humans and Rats
Humans
Reference
Caeic acida
[6]
[17]
[23]
[21]
Urine
cl
c,d,g,m
Rats
Plasma
Chlorogenic acid and caeic acid derivatives

c,f,h,j
[6]a,b
a,b
d
[17]
[24]
d,f,i,k
[20]b
(13)
a
[23]
c,n
[21]a
c,d,n
c,d,n
[22]b
d,f,g,j,m
[18]b
(13)
d,f,g,m
d,m
[19]b
Ferulic acid
[6]a
[17]b
[24]a
[28]b
[29]b
Urine
Plasma
d,f,g,i,j,l
e,f,hj
c,d
c,d
d,f,g,i,k,l
p-Coumaric acid
[9]a
[14]a
pt
Sinapic acid
[33]a
[34]a
[35]a
s, t, uy
c, d, g
Pure compound.
Natural or nutritional source.
c
Caeic acid (3,4-dihydroxycinnamic acid).
d
Ferulic acid (4-hydroxy-3-methoxycinnamic acid).
e
Dihydrocaeic acid (3-(3,4-dihydroxyphenyl)-propionic acid).
f
Dihydroferulic acid (3-(4-hydroxy-3-methoxyphenyl)-propionic acid.
g
Vanillic acid (4-hydroxy-3-methoxybenzoic acid).
h
m-Coumaric acid (3-hydroxycinammic acid).
b
pt
w
u, w
Figure 1 The metabolic pathways of caffeic acid. (From Refs. 6,16.)
Notes to Table 1:
i
3-(3-Hydroxyphenyl)-propionic acid.
j
3-Hydroxyhippuric acid (3-hydroxybenzoylglycine).
k
Feruloylglycine.
l
Vanilloylglycine.
m
Isoferulic acid (3-hydroxy-4-methoxycinnamic acid).
n
Chlorogenic acid (5-caeoyl quinic acid).
o
Hippuric acid (benzoylglycine).
p
p-Coumaric acid (4-hydroxycinnamic acid).
q
p-Coumaroylglycine.
r
4-Hydroxybenzoic acid.
s
3-(4-Hydroxyphenyl)-propionic acid.
t
4-Hydroxyhippuric acid (4-hydroxybenzoylglycine).
u
Sinapic acid (4-hydroxy-3,5-dimethoxycinnamic acid).
v
Dihydrosinapic acid (3-(4-hydroxy-3,5-dimethoxyphenyl)-propionic acid.
w
3-(3,5-Dihydroxyphenyl)-propionic acid.
x
3-Hydroxy-5-methoxycinnamic acid.
y
3-(3-Hydroxy-5-methoxyphenyl)-propionic acid.
445
446
Rechner
Figure 2 Time-amount profile of the excretion of the chlorogenic acid metabolites, ferulic, isoferulic, vanillic, dihydroferulic, and 3-hydroxyhippuric acid, in
single elimination samples after the consumption of coffee (three times in 4-h
intervals) for one volunteer. (From Ref. 18.)
urine, was also detected, leading to speculations about the metabolic fate of the
ingested hydroxycinnamates. Because of small amounts of ferulic acid derivatives in coee, the identied metabolite could not be accounted for exclusively
by caeic acid derivatives, apart from isoferulic acid, which most likely derives
from O-methylation of caeic acid.
In another human study with a similar design, artichoke extract capsules
were administered three times in 4-h intervals (total hydroxycinnamate intake
123.9 mg relative to chlorogenic acid) [19]. In contrast to coee, artichoke
447
contains only caeic acid esters but no ferulic acid derivatives or other
hydroxycinnamates. Investigation of urinary excretion of conjugates and metabolites post h-glucuronidase treatment revealed ferulic, isoferulic, vanillic, and
dihydroferulic acids as metabolites of the ingested caeic acid derivatives.
Many other studies have reported the appearance of methylated products
as metabolites of caeic acid derivatives. In particular, ferulic and dihydroferulic acids and additionally feruloylglycine were identied as urinary metabolites of caeic acid derivatives from a crude extract of Equisetum arvense in
a human study [20]. The rate of conjugation with glucuronic acid in this study
was > 90% for ferulic acid, 6080% for dihydroferulic acid, and 16%
for
feruloylglycine.
The absorption of caeic acid and chlorogenic acid has been examined in
healthy ileostomy (no colon) patients [21]. The extent of absorption was
calculated as the amount ingested minus the amount excreted in ileostomy
euent. Using this calculation, complete absorption of caeic acid and 33 F
17% absorption of chlorogenic acid were observed, whereas only 11% of the
ingested caeic acid and traces of chlorogenic acid were detected in urine. No
other metabolites were mentioned. Incubation with gastric juice, duodenal uid,
and ileostomy uid had no eect on both compounds, which were completely
recovered. The potential mechanism of degradation or metabolism of the
compounds in the small intestine and their transfer to the circulation remain
to be claried.
Caeic acid 4296 nmol/L, entirely as glucuronide) and ferulic (not
detectable78 nmol/L, more than 50% as free acid) were detected in plasma of
three volunteers post ingestion of 100 g of prunes after a 2-day lowhydroxycinnamate diet [22]. Furthermore, two of the three volunteers displayed basal plasma levels of caeic acid (as glucuronide) and ferulic acid
(more than 50% as free acid). Conjugated and free caeic (0.1460.496 Amol,
3687% as free acid), ferulic acid (0.0080.032 Amol), and chlorogenic acid
(0.0190.045 Amol as free acid) were identied post ingestion in the urine of
the three volunteers, of whom two showed basal urinary levels of the
compounds. Thus, there is considerable contradictory evidence for the potential absorption of chlorogenic acid per se.
In a study investigating the absorption of caeic acid and chlorogenic acid
in rats after oral administration, no absorption of chlorogenic acid and no
metabolism in the small intestine were observed [23]. In contrast, caeic acid
was absorbed and detected in plasma post enzyme treatment accompanied by
ferulic acid as a metabolite. Chlorogenic acid was only detected in urine of rats
after intravenous administration (50 mg/kg) but also not after oral administration; that nding is consistent with the hypothesis that major hydrolysis to
release free caeic acid occurs in the colon [24]. In an in vitro model using
isolated rat small intestine, chlorogenic acid was also not absorbed, but caeic
448
Rechner
acid was found to be absorbed through the intestinal lumen [25]. The perfused
caeic acid was glucuronidated to a high extent (approximately 60%), most
likely in the epithelium of the small intestine. Biliary excretion of glucuronidated caeic acid into small intestine was found to be only a minor metabolic
pathway in rats [26].
However, the administration of pure caeic acid, especially in the high
doses used in the studies, to humans or animals does not mirror the reality of the
diet. It more or less represents an exposure to an unusual, unknown compound
(i.e., in small intestine) in an overdose. This might result in a slightly distorted
picture of caeic acid metabolism, which does not start with caeic acid itself
but with caeic acid derivative. It is clear that much remains to be claried
concerning the bioavailability of chlorogenic acid and caeic acid deriving from
chlorogenic acid, and the mechanisms and sites of action. Dierences in study
design and specics of the volunteers studied, as well as the analytical system
applied, especially the mode of detection and sample preparation, might be
responsible for the present uncertainty.
The action of catechol O-methyl transferase might be a central metabolic
event after the absorption of free caeic acid or some of its metabolites with a
still intact catechol moiety, such as dihydrocaeic acid or protocatechuic acid.
In most studies administering chlorogenic acid or preparations rich in caeic
acid derivatives (i.e., coee), only O-methylated metabolites but no metabolites
with an intact catechol group were detected in urine, supporting the central role
of O-methylation of caeic acid post absorption [6,17,18]. Studying the Omethylation of caeic acid in vitro by using rat or rabbit liver slices or
preparations of liver, both possible O-methylation products, ferulic and isoferulic acids, were formed, and a meta/para ratio of 2.8:1 was recorded [13]. In
addition, the ability to reduce the residual double bond was also observed in
vitro with rat or rabbit liver slices [10].
Crucial for the understanding of potential bioactivities of caeic acid and
its metabolites and derivatives is knowledge about their intracellular metabolism
in the tissues after absorption and presence in the circulation. In vitro studies
have demonstrated the ability of caeic acid, chlorogenic acid, and dihydrocaeic acid to form quinones, when oxidized with peroxidase/H2O2 or tyrosinase/O2, which react with glutathione to form a number of conjugates [27]. The
glutathione conjugates were also formed in the presence of reduced nicotinamide-adenine dinucleotide phosphate (NADPH) in isolated rat hepatocyte
microsomes. This was prevented by a cytochrome P450 inhibitor, suggesting
the involvement of cytochrome P450 in the formation of the glutathione
conjugates. Furthermore, the cytotoxicity of three compounds to isolated rat
hepatocyates was enhanced by H2O2 or cumene hydroperoxidesupported
cytochrome P450 and was again prevented by a cytochrome P450 inhibitor.
These results suggest that cytochrome P450 metabolically activates hydroxy-
449
cinnamates and related compounds with a catechol moiety to form cytotoxic

quinons or quinoid metabolites.
This cytotoxicity of caeic acid, based on its catechol moiety and the
interaction with cytochrome P450, might be the reason for the extensive Omethylation and glucuronidation of caeic acid in humans observed in most
studies investigating the absorption, metabolism, and elimination of dietary
caeic acid derivatives.
B.
Ferulic Acid and Its Derivatives
The metabolites and conjugates of ferulic acid and ferulic derivatives identied
in the dierent studies in humans and animals are summarized in Table 1. The
metabolic fate of ferulic acid is qualitatively almost identical to that of caeic
acid (see Fig. 1). The same 3-hydroxyphenyl and 4-hydroxy-3-methoxyphenyl
acids were detected in human urine after the ingestion of ferulic acid and caeic
acid [7]. When ferulic acid was fed to rats, 3-(3-hydroxyphenyl)-propionic acid,
feruloylglycine, dihydroferulic acid, vanillic acid, and vanilloylglycine were
identied in urine, also showing the close relationship to caeic acid metabolism
[6]. Also in rats, 3-(3-hydroxyphenyl)-propionic acid was reported as the major
urinary metabolite of ingested ferulic acid, accompanied by smaller amounts of
vanillic acid [12]. These early studies indicate that dehydroxylation, reduction of
the side chain double bond, and h-oxidation are major metabolic events in ferulic
acid metabolism. In humans caeic acid was also detected alongside ferulic and
vanillic acids and three unidentied metabolites in urine post h-glucuronidase
treatment after the ingestion of 1 g of ferulic acid [17], suggesting the occurrence
of demethylation. Conjugation with glucuronic acid was reported to be approximately 3040% of the total amount excreted in urine after the consumption of a
single bolus of tomatoes equivalent to an ingestion of 30.1 mg ferulic acid in the
form of various derivatives [28]. The absorption rate of the ferulic acid from
tomatoes, based on the free and conjugated ferulic acid excreted, varied between
11% and 25%, showing maximal excretion around 7 h post consumption. After
the application of an extract from French maritime pine bark containing about
0.17% free ferulic acid and 0.47% ferulic acid in the form of ferulic acid
glucoside, which totals 13.97 mg and 2.35 mg total ferulic acid ingested in two
dierent administrations, in humans, an absorption rate of 3643% based on total
urinary excretion was observed [29]. The presence of free ferulic acid and the
form of application (gelatin capsules) are suggested to cause the higher rate of
absorption.
However, the O-methylation of the catechol group seems to inuence the
rate of absorption of ferulic acid compared to that of caeic acid. In rats 10.5 F
2.5% of an oral administered dose of 50 mg/kg was recovered in urine,
approximately 50% as the free acid and the other 50% conjugated with
450
Rechner
glucuronic acid [24]. Intravenous administration resulted in similar recoveries of

total ferulic acid in urine indicating extensive metabolism of ferulic acid in the
circulation and tissues. In an in vitro model using isolated rat small intestine, the
absorption of ferulic acid through the intestinal lumen was approximately four
times higher than the absorption of caeic acid [25]. The amount of conjugation
with glucuronic acid was around 20%, much lower than the 6070% observed for
caeic acid.
Ferulic acid was extensively excreted in the bile in the form of its
glucuronide (30% of the dose) after intravenous or intradermal ingestion in rats
[26]. In comparison, only small amounts of p-coumaric and caeic acid
(approximately 3% of the dose) were excreted in the bile. The biliary excretion
of ferulic acid glucuronide might result in its absorption in the small intestine or,
most likely, in its degradation in the colon. Dimers of ferulic acid, abundant
structural components of plant cell walls, especially cereal brans, were released
from the cell wall in the gastrointestinal tract of humans and rats by the colonic
microora expressing esterase activity [30]. In addition, diferulic acids were
detected in rat plasma post h-glucuronidase/sulfatase treatment after the administration of diferulic acids in sunower oil.
C.
p-Coumaric Acid and Its Derivatives
The metabolites and conjugates of p-coumaric acid identied in the dierent

studies in animals are summarized in Table 1. The metabolism of p-coumaric acid
is relatively straightforward and is reported to involve conjugation with glucuronic acid or glycine in the liver or epithelium of small intestine, reduction of
residual double bond to 3-(4-hydroxyphenyl)-propionic acid, and h-oxidation of
the reduction product to 4-hydroxybenzoic acid. These products and the unconjugated p-coumaric acid have been detected and identied in urine of rats after
the ingestion of p-coumaric acid [9,14]. The h-oxidation of p-coumaric acid
occurs in the liver and is carried out by adenosine triphosphate(ATP)-requiring
enzymes localized in the mitochondria [31]. The reduction of the side chain
double bond of p-coumaric acid was also identied as a metabolic action of
colonic microorganisms [11]. Dehydroxylation or O-methylation of the phenolic
hydroxyl group has not been described.
The absorption of p-coumaric acid was shown in the rat gut after the
feeding of spinach cell walls with 14C-labeled ferulic and p-coumaric acids to
rats [32]. The two free hydroxycinnamic acids were absorbed after release
from the cell wall in the cecum and colon, a nding that strongly suggested
an involvement of colonic bacteria. In an in vitro study of the absorption of
polyphenols using isolated small intestine from the rat, p-coumaric acid was
absorbed to a higher extent than caeic acid and to a lower extent than ferulic
acid [25]. Glucuronidation of p-coumaric acid in the epithelium of the rat
451
small intestine was not observed in this study. Biliary excretion of p-coumaric
acid (as glucuronide) was shown to be only a minor metabolic pathway in
rats [26]. The metabolism of p-coumaric acid is schematically summarized
in Figure 3.
D.
Sinapic Acid and Its Derivatives
The metabolites and conjugates of sinapic acid identied in the dierent studies
in animals are summarized in Table 1. The metabolism of sinapic acid has been
studied only in animals, rat and rabbit. The unchanged hydroxycinnamate, 3(3,5-dihydroxyphenyl)-propionic acid, dihydrosinapic acid (3-(4-hydroxy-3,
5-dimethoxyphenyl)-propionic acid), 3-hydroxy-5-methoxycinnamic acid, and
3-(3-hydroxy-5-methoxyphenyl)-propionic acid were detected and identied as
urinary metabolites in rats after an oral dose of 800 mg/kg [33,34].
Interestingly, the maximal excretion of the demethylated products occurred
2 or 3 days post ingestion, suggesting a prolonged metabolism. After the oral
administration of 100 mg/kg to rats, only the unchanged sinapic acid and 3(3,5-dihydroxyphenyl)-propionic acid were excreted, both partly conjugated
[35]. Sinapic acid was excreted during 24 h post administration, whereas 3(3,5-dihydroxyphenyl)-propionic acid excretion lasted 2 or 3 days. The 3-(3,5dihydroxyphenyl)-propionic acid was shown to be the major urinary metabolites in rabbits fed with sinapic acid (200 mg/kg). The intestinal microora of
the rabbit was identied as the metabolic site of action for the demethylation.
A more detailed study on the bacterial degradation revealed that sinapic acid
was reduced to dihydrosinapic acid and then O-methylated and dehydroxylated, nally forming 3-(3,5-dihydroxyphenyl)-propionic acid [36].
The metabolic fate of sinapic acid in humans as well as its derivatives,
which represent the naturally occurring form of sinapic acid, has not been
studied yet and remains uncertain. But in a study on the metabolic fate of
polyphenols from a polyphenol-rich diet, small amounts of sinapic acid were
detected in urine post h-glucuronidase treatment during a 5-day feeding period
[37]. This presence of sinapic acid (as glucuronide) in urine implies its
bioavailability from the diet in humans and a metabolic fate of dietary sinapic
acid derivatives similar to that of other hydroxycinnamic acid derivatives such
as chlorogenic acid.
III.
COLONIC METABOLISM
The colonic microora are a central site of hydroxycinnamate metabolism in

addition to the small intestine and liver. The majority of ingested hydroxycinnamates from dietary sources reach the colon, where they are exposed to the
452
Figure 3
Rechner
The metabolic pathways of p-coumaric acid.
453
colonic microora and their metabolic properties. Many metabolic events have
been suggested as actions of bacterial enzymes, including cleavage of the ester
bond, dehydroxylation, demethoxylation, reduction of the residual double
bond, and a- and h-oxidation [11]. In investigation of the possible site of
cleavage of the ester bond between caeic acid and quinic acid, the colon
microora were identied in an in vitro experiment as the only active extract,
whereas extracts of small intestine epithelium, plasma, and liver expressed no
esterase activity [38]. No cleavage of the ester bond of chlorogenic acids
occurred at the acidic pH of the gastric lumen, studied by incubating the pure
compound or extract rich in caeic acid esters with simulated gastric juice
[18,19,21]. Some human fecal microorganism expressing cinnamoyl esterase
activity were isolated and genotypically characterized as Escherichia coli
(three isolates) and Bidobacterium lactis and Lactobacillus gasseri (two strains)
[39]. The esterase activity of the characterized bacteria was essentially intracellular. Interestingly, the characterized bacteria did not express any other
phenol-degrading activities.
Other phenol-degrading activities of the human colonic microora were
identied in an in vitro incubation trial with chlorogenic acid [40]. After the
initial cleavage of the ester bond the reduction of the side chain double bond
of caeic and the dehydroxylation of the product dihydrocaeic acid resulted
in the formation of the endproduct 3-(3-hydroxyphenyl)-propionic acid. A
delay of approximately 5 h from the start of the incubation until the cleavage
of the ester bond occurred was observed, indicating a period of adaptation of
the colonic microora to the substrate. Further dehydroxylation or a- and hoxidation of the identied endproduct was not observed, but protocatechuic
acid (h-oxidation of dihydrocaeic acid) and m-coumaric acid (dehydroxylation of caeic acid) were detected as minor metabolites during the incubation.
O-Methylation of the catechol moiety of chlorogenic, caeic, or dihydrocaeic
acid also did not occur, pointing to other sites of action for this metabolic
event. The most likely site for O-methylation as well as glucuronidation is the
liver, but the epithelium of the colon might also be capable of this metabolic
action. The epithelium of the small intestine was able to glucuronidate and Omethylate avonoids and hydroxycinnamates in in vitro experiments with
isolated rat small intestine [25,41].
The colon seems to be the essential site for the release of free
hydroxycinnamic acids and their absorption. In a study of the uptake of 14Clabeled hydroxycinnamates bound to spinach cell walls in rats, the foregut was
localized as the site of absorption after the release of the labeled hydroxycinnamic acids form the cell wall by colonic microorganisms, when 25% of the
ingested dose of label was found to be associated with body tissue after only 2
h [32]. On the basis of the results on absorption, metabolism, degradation, and
elimination of chlorogenic acid in humans presented, its metabolic fate in
454
Rechner
Figure 4
The metabolic fate of dietary chlorogenic acid.
humans is schematically summarized in Figure 4. However, infused into the

rumen of the sheep, chlorogenic acid, caeic acid, and ferulic acid underwent
complete dehydroxylation and were mainly excreted as benzoic acid [15]. Small
amounts of urinary cinnamic and 3-(3-hydroxyphenyl)-propionic acids were
also detected. The complete dehydroxylation of free hydroxycinnamic acids
would result in the formation of phenylpropionic acid, which is a proposed
precursor of hippuric acid. An increase in hippuric acid excretion, a common
urinary metabolite, was associated with a high dietary intake of polyphenols
and/or hydroxycinnamates in humans and in rats [18,20,42,43]. Formed and
absorbed in the colon, 3-phenylpropionic acid might contribute to the hippuric
acid levels in plasma and urine after h-oxidation to benzoic acids and
subsequent glycination in the liver. However, hippuric acid also derives from
other dietary sources, such as aromatic amino acids, benzoates used as
preservatives, and quinic acid, which together with caeic acid forms chlorogenic acid [44,45].
IV.
CONCLUSIONS
In quantitative terms, colonic metabolism and the formed colonic metabolites

are more important than the absorption and conjugation of the hydroxycinnamic
acids. Whether these colonic metabolites express potential protective or preventive bioactivities in vivo remains to be studied. This also applies to the eect
of the dietary hydroxycinnamates and their metabolites on the composition of
the colonic microora.
455
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Macheix JJ, Fleuriet A, Billot J. Fruit Phenolics. Boca Raton; FL: CRC Press, 1990.
Herrmann K. Occurrence and content of hydroxycinnamic and hydroxybenzoic acid
compounds in foods. Crit Rev Food Sci Nutr 1989; 28:315347.
Cliord MN. Chlorogenic acid and other cinnamatesnature, occurrence and
International Union of Pure and Applied Chemistry. Nomenclature of cyclitols.
Biochem J 1976; 153:2331.
DeEds F, Booth AN, Jones FT. Methylation of phenolic hydroxyl groups by rabbit.
Fed Proc Fed Am Soc Exp Biol 1955, 14:189205, 332.
Booth AN, Emerson OH, Jones FT, Deeds F. Urinary metabolites of caeic and
chlorogenic acids. J Biol Chem 1957; 229:5159.
Shaw KNF, Trevarthen J. Exogenous sources of urinary phenol and indol acids.
Nature 1958; 182:797.
Hill GA, Ratcli J, Smith P. Urinary catechol ethers. Chem Ind 1959; 399:
Booth AN, Masri MS, Robbins DJ, Emerson OH, Jones FT, DeEds F. Urinary
phenolic acid metabolites of tyrosine. J Biol Chem 1960; 235:2649.
Masri MS, Booth AN, DeEds F. O-Methylation in vitro of dihydroxyl- and
trihydroxyl-phenolic compounds by liver slices. Biochim Biophys Acta 1962;
65:495.
Scheline RR. Metabolism of phenolic acids by the rat intestinal microora. Acta
Pharmacol Toxicol 1968; 26:189.
Teuchy H, Van Sumere CF. The metabolism of [1-14C]phenylalanin, [3-14C]cinnamic acid and [2-14C]ferulic acid in the rat. Arch Int Physiol Biochim 1971,
79:589597.
Creveling CR, Morris N, Shimizu H, Ong HH, Daly J. Catechol O-methyltransferase. IV. Factors aecting m-and p-mehylation of substituted catechols. Mol
Pharmacol 1972; 8:398409.
Griths LA, Smith GE, Metabolism of apigenin and related compounds in the rat.
Biochem J 1972; 128:901911.
Martin AK. The origin or urinary aromatic compounds excreted by ruminants. II.
The metabolism of phenolic cinnamic acids to benzoic acid. Br J Nutr 1982;
47:155164.
Scheline RR. Metabolism of acids, lactones, and esters. In: CRC Handbook of
Mammalian Metabolism of Plant Compounds. Boca Raton; FL: CRC Press,
1991:167175.
Jacobson EA, Newmark H, Baptista J, Bruce WR. A preliminary investigation of
the metabolism of dietary phenolics in humans. Nutr Rep Int 1983; 28:14091417.
Rechner AR, Spencer JPE, Kuhnle G, Hahn U, Rice-Evans CA. Novel biomarkers
of the bioavailability and metabolism of caeic acid derivatives in humans. Free
Radic Biol Med 2001; 30:12131222.
Rechner AR, Pannala AS, Rice-Evans CA. Caeic acid derivatives in artichoke
extract are metabolized to phenolic acids in vivo. Free Radic Res 2001; 35: 195
202.
456
20.
Rechner
Graefe EU, Veit M. Urinary metabolites of flavonoids and hydroxycinnamic acids

in humans after application of a crude extract from Equisetum arvense.
Phytomedicine 1999; 6:239246.
21. Olthof MR, Hollman PCH, Katan MB. Chlorogenic acid and caeic acid are
absorbed in humans. J Nutr 2001; 131:6671.
22. Cremin P, Kasim-Karakas S, Waterhouse AL. LC/ES-MS detection of
hydroxycinnamates in human plasma and urine. J Agric Food Chem 2001; 49:
17471750
23. Azuma K, Ippuushi K, Nakayama M, Ito H, Higashio H, Terao J. Absorption of
chlorogenic acid and caffeic acid in rats after oral administration. J Agric Food
Chem 2000; 48:54965500.
24. Choudhury R, Srai K, Debnam E, Rice-Evans CA. Urinary excretion of
hydroxycinnamates and avonoids after oral and intravenous administration. Free
Radic Biol Med 1999; 27:278286.
25. Spencer JPE, Chowrimootoo G, Choudhury R, Debnam ES, Srai SK, Rice-Evans
CA. The small intestine can both absorb and glucuronidate luminal avonoids.
FEBS Lett 1999; 458:224230.
26. Westendorf J, Czok G. Studies on the pharmacokinetics of C-14 cinnamic-acid
derivatives in rats. Z Ernahrungswiss 1978; 17:2636.
27. Moridani MY, Scobie H, Jamshidzadeh A, Salehi P, OBrien PJ. Caeic acid,
chlorogenic acid, and dihydrocaeic metabolism: glutathione conjugate formation.
Drug Metab Dispos 2001; 29:14321439.
28. Bourne LC, Rice-Evans CA. Bioavailability of ferulic acid. Biochem Biophys Res
Comm 1998; 253:222227.
29. Grosse Duweler K, Rohdewald P. Urinary metabolites of French maritime pine bark
extract in humans. Pharmazie 2000; 55:364368.
30. Andreasen MF, Kroon PA, Williamson G, Garcia-Conesa MT. Intestinal release
and uptake of phenolic antioxidant diferulic acids. Free Radic Biol Med 2001;
31:304314.
31. Ranganathan S, Ramasarma T. Enzymic formation of p-hydroxybenzoate from phydroxycinnamate. Biochem J 1971; 122:487493.
32. Buchanan CJ, Wallace G, Fry SC, Eastwood MA. In vivo release of 14C-labelled
phenolic groups from intact dietary spinach cell walls during passage through the rat
intestine. J Sci Food Chem 1996; 71:459469.
33. Griths LA. Metabolism of sinapic acid and related compounds in the rat. Biochem
J 1969; 113:603609.
34. Griths LA. 3,5-dihydroxyphenylpropionic acid, a further metabolite of sinapic
acid. Experientia 1970; 26:723724.
35. Meyer T, Scheline RR. 3,4,5-trimethoxycinnamic acid and related compounds. I.
Metabolism in the rat. Xenobiotica 1972; 2:391398.
36. Meyer T, Scheline RR. 3,4,5-trimethoxycinnamic acid and related compounds. I.
Metabolism in by the rat intestinal microora. Xenobiotica 1972; 2:383390.
37. Rechner AR, Kuhnle G, Bremner P, Hubbard GP, Moore KP, Rice-Evans CA. The
metabolic fate of dietary polyphenols in humans. Free Rad Biol Med 2002; 33:220
235.

38.
39.
40.
41.
42.
43.
44.
45.
457
Plumb GW, Garcia-Conesa MT, Kroon PA, Rhodes M, Ridley S, Williamson G.

Metabolism of chlorogenic acid by human plasma, liver, intestine and gut
microora. J Sci Food Agric 1999; 79:390392.
Coteau D, McCartney AL, Gibson GR, Williamson G, Faulds CB. Isolation and
characterization of human colonic bacteria able to hydrolyse chlorogenic acid.
J Appl Microbiol 2001; 90:873881.
Rechner AR, Watts WMP, Gibson GR, Rice-Evans CA. Colonic metabolism of
dietary phenolics: implications for the nature of the bioactive component. Free Rad
Biol Med 2001; 31:89S.
131:17531757.
Phibbs AN, Stewart J, Wright B, Wilson ID. Eect of diet on the urinary excretion of
hippuric acid and other dietary-derived aromatics in rat: a complex interaction
between diet, gut microora and substrate specity. Xenobiotica 1998; 28:527537.
Cliord MN, Copeland EL, Bloxsidge JP, Mitchell LA. Hippuric acid as a major
excretion product associated with black tea consumption. Xenobiotica 2000; 50:
317326.
Chesson A, Provan GJ, Russell WR, Scobbie L, Richardson AJ, Stewart C. Hydroxycinnamic acids in the digestive tract of livestock and humans. J Sci Food
Agric 1999; 79:373378.
Indahl SR, Scheline RR. Quinic acid aromatization in the rat: urinary hippuric acid
and catechol excretion following the singular or repeated administration of quinic
acid. Xenobiotica 1973; 3:549556.
Index
AAPH (see 2, 2V-azobis, 2-amidino

propane hydrochloride)
ABAP (see 2, 2V-azobis(2-methyl-propionamidine dihydrochloride))
Absorption, 291408, 422
ABTS (see 2,2V-azinobis-(3-ethylbenzothiazoline-6-sulphonicacid ABTS))
Acacetin, 146
Acylation, 105
Aging brain, 205219
Akt, 235, 237
Aldehyde, 166
AMVN (see 2,2V-azobis,
2,4-dimethylvaleronitrile)
Analysis, 97119
Angiotensin-converting-enzyme
(ACE), 182
ANT (adenine nucleotide
translocase), 279280
Anthocyanidin, 168169, 172, 175,
181, 249, 277
Anthocyanin, 11, 20, 2728, 32, 44,
46, 53, 73, 76, 78, 79, 81, 89,
102, 391
3-deoxyanthcyanins, 97119
diacylated, 24
Anthoxanthin, 168
Antioxidant activity, 26, 29, 30, 46,
7191, 171, 173, 176, 181,
184, 189, 206, 209
Antioxidant capacity (see Antioxidant

activity)
Antioxidant status (see Antioxidant
activity)
AP-1, 240, 253
Apigenidin, 98, 100
Apigenin, 6163, 88, 146, 288, 318,
320, 323, 364
Apolipoprotein B100, 166
Apoptosis, 234, 242, 254, 284293,
315, 316, 318
Apoptosis-inducing factor (AIF), 285
Apple, 7, 9, 10, 17, 21, 23, 26, 27, 73
76, 82, 83, 89, 396, 397, 413
Apricot, 9, 441
Arachidonic acid, 166
Artichoke, 11, 12, 17, 46, 441, 446
Ascorbate (see Vitamin C)
Asteraceae, 15
Atherosclerosis, 165190, 413
ATP-binding site, 255, 256
Auto-oxidation, 311
Avocado, 7
2,2V-azinobis-(3-ethylbenzothiazoline-6-sulphonic acid
ABTS), 71
2,2V-azobis, 2,4-dimethylvaleronitrile
(AMVN), 169
2, 2V-azobis, 2-amidinopropane
hydrochloride (AAPH), 169,
175, 178
459
460
2, 2V-azobis 2-methyl-propionamidine
dihydochloride (ABAP), 72
Bacterial esterase, 415
Baicelein, 313, 318, 323
Baicalin, 313, 316318
Banana, 6, 73, 76
BAX, 243, 246
Bcl-2, 237, 241, 242, 243, 285, 287
Behavior, 206, 208, 210, 213
Benzodiazepine
binding site, 280
receptor, 321
Benzoic acid, 419, 454
3-hydroxy, 420
4-hydroxy, 421
3,4 dihydroxy, 421
methoxy, 23
Benzoyl glycine (see Hippuric acid)
Bidobacterium lactis, 453
Bilberry, 46, 5354
Biliary excretion, 393
Bioavailability, 171, 349358,
413432
Biochanin A, 50, 56, 98, 99, 110
Blackberry, 5, 9, 17
Blackcurrant, 5, 9
Blood brain barrier (BBB), 208
215, 294
Blueberry, 5, 9, 17, 209210, 214
Branching ratios, 139
Brain aging, 205219
Brassica, 11, 15, 18
Breast cancer, 319
Broccoli, 12, 15, 7376, 81
Browning, 2729
Butein, 278
Cabbage, 11, 12, 15, 17, 7376,
8182
Cabernet Sauvignon, 175
Caco-2 cells, 187, 349358, 393,
401404
Caeic acid, 89, 12, 17, 22, 24, 26,
30, 32, 61, 170
Index
Caeoyl phenylethyl ester, 58
Caeoylquinic acid
3V-84, 441
4V, 84, 441
Caeoyl shikimic acid, 8, 28,
441449
Caeoyl tartaric acid, 15, 18
Caftaric acid, 8, 10, 25, 28
Calcium, 322
Calgranulin A, 305
Camellia sinensis, 5455, 99, 184
Capsicum frutescens, 18, 20
Carbon dioxide, 422
Carcinogenesis, 316
Carnosic acid, 61, 62
Carnosol, 61, 62
h-carotene, 168
Carrot, 11, 12
Caspase-3, 242, 243, 285, 288, 292
293, 312, 318
Caspase-9, 242
Catechin, 44, 56, 123, 124, 126, 170,
171, 172, 175177, 184, 185,
249, 277, 279, 283, 303, 312,
316, 322, 364, 369372,
413416, 420, 425, 428, 429
5V,7V-dimethyl, 125, 130
3V,4V-dimethyl, 126, 128, 130
epimerization, 125
fragmentation, 134
glucuronide, 428
O-methyl, 156
3V-O-methyl, 123, 130, 131, 429
4V-O-methyl, 130,131
3V-O-methyl glucuronide, 428
3V-O-methyl sulfate, 428
trimethyl, 126
Catechol-O-methyl transferase
(COMT), 371, 379, 422, 442
Catenin, 305
Cauliower, 7376
Cell cycle arrest, 287
Chalconaringenin, 84
Chalcone, 46, 277, 278
Chemiluminescence, 3, 426
Index
Cherry, 5, 7, 9, 10, 18, 28, 441
Chilling injury, 21
Chlorogenic acid, 7, 8, 15, 1720,
2224, 26, 28, 29, 31, 49, 81,
82, 441, 443444, 448
Chocolate, 90, 413, 427, 429
Cholesterol, 165190
Cholesterol linoleate hydroperoxide,
167
Cholesteryl ester, 173
Chrysin, 58, 146, 352, 355
Chrysoeriol 146
Citrus, 26, 80, 8889, 357
Clover, red, 56
Cocoa, 11, 289
Coee, 7, 17, 89, 441, 443, 446
Colonic metabolism, 454
Colonic microora, 145, 376,
451, 453
Columnidin, 100
Complex I (reduced nicotinamideadenine dinucleotide oxidase),
276278, 281, 283
Complex II (succino oxidase),
276278, 283
Complex III, 285
Complex IV (cytochrome oxidase),
285
Condensed polymers, 169
Conjugated diene, 186
Conjugation, 366367
M-coumaric acid, 442443
P-coumaric acid, 3, 8, 9, 11, 12, 17,
2123, 26, 30, 170, 441, 444,
450451
Coumarin, 61, 63
-7-geranyloxycoumarin, 356
P-coumaroyl,
glucose, 18, 19, 22, 24
-O-glucoside, 7, 8
malic ester, 84
quinic acid, 7, 10, 15, 19, 23
tartaric acid (ester), 18
Coumestrol, 181
CREB, 237, 246, 283, 423
461
Cryptochlorogenic acid, 7, 167
Cucumber, 7
Cucurbitaceae, 7, 17
Curcumin, 213
Cyanidin, 54, 172, 250, 364
3, 3-diglycoside, 181
-3-glucoside, 79, 181
-3-rutinoside, 79
-3-sophoroside, 79
Cyclo-oxygenase, 181
Cytochrome b, 558, 167
Cytochrome P450, 182, 448, 449
Cytoprotective, 309330
Cytotoxic, 309, 311, 315, 318
Daidzein, 181, 364
Decarboxylation, 442
Delphinidin, 11, 14, 24, 54, 364
-3-glucoside, 181
Dementia, 206
Derivatization, post-column, 394
Diabetes 89, 167
Dianella, 11, 24
Dicaeoyl quinic acid, 11, 15
Diferuloyl quinic acid, 11
Dihydrocaeic acid, 442, 444
Dihydroferulic acid, 20, 442, 444,
446449
Dihydroavanol, 277
Dihydroavonol, 277
3,4-dihydroxybenzoic aldehyde, 6
Dihydroxyphenylalanine, 10
3,4-dihydroxyphenylethanol-elenolic
acid, 187
Dimmer, 372
Dopamine, 233
Diosmetin, 313, 366, 367, 371
Diosmin, 89
O-diphenol, 28
DNA damage, 316, 321, 329
DPPH, 71
ECG (see Epicatechin gallate)
EGCG (see Epigallocatechin gallate)
EGC (see Epigallocatechin)
462
Eggplant, 7, 11, 17, 28, 441
Elderberry, 441
Ellagic acid, 181
Ellagitannin , 22, 172
Endive, 11, 12, 15, 17
Enteroyte, 350, 353, 355, 367, 374
Epicatechin, 55, 146, 171, 184, 185,
249250, 253, 255, 279, 289,
293, 303, 312313, 316, 318,
322, 324325, 352, 355, 356,
358, 364, 369371, 377, 413,
415, 416, 425, 427, 429, 430
O-glucuronide 156
5-O-glucuronide, 151, 155, 419
7-O-glucuronide, 155
3V-O-methyl 155, 293
4V-O-methyl, 155
3V-O-methyl 5-0-glucuronide, 417
Epicatechin gallate (ECG), 4, 6, 25,
55, 184, 185, 278, 288, 289,
292, 364, 375, 413416, 421,
427, 430, 431
Epigallocatechin (EGC), 55, 171, 185,
279, 288, 289, 364, 377, 413
416, 420, 421, 425, 427, 430
Epigallocatechin gallate (EGCG), 55,
124, 184185, 207, 288, 289,
292, 314, 315, 321, 322, 364,
367, 375, 413416, 427, 430
Epoxyisoprostane, 166
Equol, 56
Eriodictyol, 323, 326
ERK (see Extracellular signal-related
kinases)
Escherichia coli, 453
Extracellular signal-related kinases
(ERK 1/2), 235237, 244
247, 252, 283, 290, 293, 321
Everted sac, 404, 406
Ferulic acid, 89, 11, 12, 15, 16, 19, 20,
2226, 61, 82, 170, 307, 441
442, 444, 446447, 449450
Feruloyl esterase, 30
Feruloyl glycine, 442
Index
Feruloylquinic acid, 10, 15, 84
Fibroblasts, 312313, 324
Fisetin, 316317, 320, 328
Fish oil, 186
Flash photolysis, 130
Flavan, 168
Flavanol, 77, 86, 123, 124, 168170,
172, 367, 373, 377, 413432
methylated, 123140
monomer, 413432
Flavan-3-ol (see Flavanol)
Flavanone, 21, 46, 77, 79, 86,
8889, 169, 280
Flavone, 21, 44, 89, 168169, 277,
280, 356
Flavonol, 44, 81, 8384, 86, 8889,
168170, 172, 277, 280, 399,
315, 322
Foam cell, 172, 182, 187, 188
Food processing, 27
Formononetin, 56, 9899
Fragment nomenclature, 149
Fragmentation reactions, 146, 148
FRAP (ferric reducing ability of
plasma), 7273, 75, 86, 87
French Maritime bark extract, 278
French paradox, 171
Galangin, 58, 146
Gallic acid, 36, 17, 26, 30, 171,
181, 185, 413
Gallotannin, 172
Garlic extract, 213
Gastric juice, 365
Gastrointestinal tract, 171, 363, 398
metabolism in, 363382
Genistein, 98, 99, 181, 278, 279, 287,
288, 316, 318, 353, 364
Gene modulation, 303307
Genomics, 303
Gentisic acid, 4
Gentobioside, 82
Ginkgo biloba extract (EGb 761), 45,
46, 50, 51, 216, 217, 248
250, 289, 328
Index
Ginkgolide, 51, 217
Ginseng, 216217
Ginsenosides, 218
Glabridin, 170, 172, 178180
Glucogallin, 6
Glucosidase, 56, 353, 374, 375,
400402, 407
Glucuronidase, 381, 443, 447
Glucuronidation, 250, 368, 371, 379,
414, 416, 418419, 422, 423,
448, 450, 453
Glucuronide, 324, 368, 371, 372
Glutathione (GSH), 167, 182, 209,
287, 289, 320
peroxidase, 89
-S-transferase, 292293
Glutathionyl conjugates, 320
Glycination, 442
Glycyrrihizin, 46
Graminaeae, 15, 16
Grape, 57, 9, 10, 11, 17, 2628,
7376, 81, 171, 175, 210
Grapefruit, 7, 7377, 79, 80
Green tea, 312, 314, 321, 330
Glycation, 89, 90
HaCaT cells, 303307
Helicobacter pylori, 366
HepG2, 403
Hepatic portal vein, 372, 380
Herb, 4365, 217
Hesperetin, 61, 80, 323, 327, 328,
364, 366, 369371
O-glucuronide, 156
Hesperidin, 79, 80, 89, 356
High-density lipoprotein (HDL),
167, 168
High-performance capillary electrophoresis, 2, 45
High-performance liquid chromatography (HPLC), 2, 45, 49, 57,
78, 80, 82, 84, 101104, 108,
398, 420
Hippuric acid, 376, 377, 419, 420, 443
3-hydroxy, 420, 442, 443, 446
463
[Hippuric acid]
3-methoxy 4-hydroxy, 421
HL-60 cells, 288
Horseradish, 5
peroxidase, 306
HT29 human colon carcinoma cells,
289, 305
Hydrogen peroxide, 292, 312313,
315316, 318
Hydroxybenzoic acid (HBA), 17,
10, 17, 19, 23, 30, 31, 169
Hydroxycinnamate (see Hydroxycinnamic acid)
Hydroxycinnamic acid (HCA), 13,
7, 1012, 1421, 23, 26, 30
31, 79, 8284, 88, 104, 169,
441454
Hydroxycinnamoyl quinate transferase, 19, 22
Hydroxyethylrutinoside, 279
3-(3-hydroxyphenyl) propionic acid,
442443, 449, 453
Hydroxytyrosol, 172, 187
Hypercholesterolaemia, 167, 178, 184
Hypericum perforatum, 45, 47, 50,
5153
Ileostomy, 394, 447
Ileum, 370, 380, 416
Intestinal juice, 367
Iron-sulfur cluster, 276
Irradiation,
gamma 21
ultraviolet, 26, 175, 304, 306, 313, 314
Ischemic damage, 215216
Isochlorogenic acid, 7
Isoferulic acid, 446447
Isoavan, 168, 170
Isoavone, 44, 50, 97119, 168,
169, 181, 274, 276
Isoavonoid, 273
Isoprostane, 166, 175
Isoquercitrin, 53
Isorhamnetin, 146
-3-O-rutinoside 51
464
Jejunum, 367, 369, 371, 380, 416
JNK (see c-Jun N-terminal kinases)
Juice
apple, 188, 441
cranberry, 188
grape, 188
grapefruit,188
orange, 89,
C-Jun N-terminal kinases (JNK
1/2/3), 235, 238244, 246,
247, 252, 283, 290293
Kaempferol, 61, 63, 79, 81, 82, 88,
146, 170, 216, 250, 253, 288,
289, 293, 364, 369, 370, 312,
316317, 319
O-glucuronide, 156
-3-O-rutinoside, 48, 51
rhamnoside, 48
Kale, 15
Keratinocyte, 304
Kiwi fruit, 7
Lactobacillus gasseri, 453
Lactose phlorizin hydrolase (LPH),
400401, 404406
LC-MS, 398
LC-MS/MS, 398
Learning, 206, 209210, 214
Leek, 7376, 81, 82
Lemon, 27
Lettuce, 12, 15, 7376, 81, 89
Licochalcone, 178
Linoleic acid, 185
Lipid peroxide, 173
Lipoxygenase, 167, 169, 181, 187, 306
Licorice (Glycyrrhiza glabra), 178180
Low density lipoprotein (LDL),
165190
Lumen, 367
Luteolin, 46, 6163, 88, 146, 170,
320, 323, 357, 364, 369371
Luteolinidin, 98, 100
Malvidin, 11, 14, 364
Index
[Malvidin]
3-O-glucoside 54
Macrophage, 166, 167, 175, 176,
179, 180, 182, 186
Manganese superoxide dismutase
(Mn-SOD), 281, 283, 291
MAP kinase (see mitogen activated
protein kinases)
Mass spectrometry, 2, 45, 118119
electrospray ionization (ESI-MS),
45, 48, 50, 5157, 65, 118, 148
atmospheric pressure ionization
(APCI-MS), 45, 58, 65, 148
ion trap (ITMS), 45, 4849
tandem, 155, 159
thermospray (TSP-MS), 45
Melon, 7
Memory, 206, 209, 213215, 218
Metabolism, 366367, 442
Metal chelation, 44, 168, 169
O-methylated, 250, 324325
O-methylated glucuronide, 371, 380
Methylation, 105, 414, 416, 417
419, 422, 423, 449, 450, 453
Microora, 349, 376377, 379, 415
Mitochondria, 254, 273294
Mitochondria coupling, 279281
Mitochondrial action, 273294
Mitochondrial ATPase, 279281
Mitochondrial permeability transition
(MPT), 280281, 288
Mitochondrial respiratory chain,
273294
Mitogen-activated protein kinases
(MAPK), 214, 233257, 274,
321, 327, 331, 286, 290293
MK-571, 353, 356
Monocyte, 165, 413
Morin, 278, 316
Mouth, 365
MPT, 254, 288
MRP, 353, 355356, 402403,
405406, 423, 425
Myricetin, 146, 171, 176, 216, 288,
315317, 329
Index
NADH oxidase, 43, 276, 277
NADH-ubiquinone reductase, 277
NADP oxidase, 167, 169, 180,
182, 189
Naringenin, 58, 80, 146, 250, 316,
327328, 364, 366, 369371
O-glucuronide 156
Naringin, 79
Narirutin, 79, 80
Necrosis, 316
Neochlorogenic acid, 7
Neohesperidin, 79
Neurodegeneration, 247
Neuron, 312, 324
hippocampal, 215
striatal, 208
Neuroprotective, 51, 206207, 209,
215217, 233
Neutral loss scan, 159
NF-nB, 235, 321
Nitric oxide, 245246
Nobiletin, 356
Nuclear magnetic resonance (NMR),
2, 115117
Oleic acid, 185
Oleuropein, 172, 187
Olive oil, 185187
Onion, 5, 16, 7376, 81, 89, 395398
ORAC (oxygen radical absorbance
capacity), 7273, 75, 8687
Orange, 18, 7377, 80
h-oxidation, 442, 450, 453
Oxidative stress, 311, 327
Oxidized low density lipoprotein,
165190, 289, 293
Oxygenase, 167
Oxysterol, 166, 179
P38, 235, 238, 290292
Paracellular transport, 351, 356, 416
Paraoxonase (PON1), 167168, 169,
176, 179
Parent ion scan, 159
Parsley, 61, 6364
465
Patuletin, 82
Pea, 7376
Peach, 7, 9, 18, 23, 73, 7476, 82,
83, 441
Pear, 9, 18, 7376, 8283, 413
Peripheral vascular disease, 186
Perlargonidin-3-glucoside, 79
Permeability coecient, 352, 358
Permeability transition pore complex
(PTPC), 280, 281, 288
Peroxynitrite, 292, 309
Petunidin, 11, 54
3-O-glucoside 54
P-glycoprotein, 356, 358
Phenylacetic acid, 419
Phenylalanine ammonia-lyase, 19,
2122
Phenolic acid, 132
Phenoxyl-radical-phenate
coupling, 132
Phenylpropionic acid, 419
3-hydroxy, 420
Phenylvalerolactone, 418, 419, 421
Phloridzin, 84
Phospholipase A2, 167
Phospholipase D, 167
Photoionization, 130
Photo-oxidation, 130
Phycoerythrin, 72
Phytoalexin, 22, 24
Phytoestrogen (see Isoavone)
PI3 kinase, 235, 237, 321
Pineapple, 7, 1718, 22
Pine bark extract, 303307
Pinobanksin, 57, 58, 280, 282
Pinus maritima, 303
Pka, microscopic, 125, 131
Plum, 5, 6, 9, 17, 18, 7376, 79, 441
Poly (ADP-ribose) polymerase
(PARP), 286, 288
Polymeric anthocyanidin, 171
Polyphenoloxidase (PPO), 23, 28, 29
Polysaccharide, 23
Pomegranate, 180184, 188
PON1 (see Paraoxonase)
466
Potato, 5, 11, 12, 16, 89
Precursor ion, 157
Proanthocyanidin, 211,357, 415
Procyanidin, 28, 52, 54, 59, 90, 303,
318, 365367, 373, 378, 415
Procyanidin oligomer, 279
Product ion spectrum, 157
Pro-oxidant, 315
Propolis, 56, 57
Protection
selective,124
dichlorodiphenyl-methane,126
A-ring,128
Protein kinase, 180, C 209, 321
Protocatechuic acid, 37, 17, 187,
448449
Proton anity, 137
MNDO calculations, 137
Proton pump, 278
Prune, 30
Puncirchin, 366
Pycnogenol, 278, 318
Quercetin, 61, 63, 79, 81, 88, 146,
170, 171, 172, 176, 177, 216,
250, 255, 278, 279, 280, 282,
288, 311, 312, 315318, 320,
321, 323, 324, 327329, 352,
364, 365, 369371, 378, 394
395, 397399, 402, 403
-3,4V-diglucoside, 81
glucoside, 396397
-3-glucoside, 81, 84, 397400,
403406
-4V-glucoside, 81, 352, 353, 358,
395396, 398, 400, 402, 403
405, 406
O-glucuronide, 151, 406
glycosides, 391408
3V-O-methyl 156
3V-O-methyl, 4-O-glucoside, 398
rhamnoglucoside, 397, 399
Quercitrin, 49, 52
Quinone methide, 365
O-quinone, 23
Index
Radish, 12
Raspberry, 17, 7377, 79
Reduced nicotinamide-adenine
dinucleotide oxidase
(see NADH oxidase)
Redox potential, 132, 275279
Resveratrol, 279, 318319, 369, 371
Retro-Diels-Alder reaction, 134, 149
Rosaceae, 17, 25
Rosemary, 62, 64
Rosmarinic acid, 14, 15, 26, 61, 62
Rotenone, 277
Rutin, 49, 84, 90, 170, 309, 313,
327329, 370, 376, 379,
394400, 405
Salicylic acid, 46
Saliva, 365
Selected ion monitoring (SIM), 157
Selected reaction monitoring (SRM),
158
Seven Countries Study, 185
SGLT-1 transporter, 353, 358, 402,
403404, 406
Shikimate, 16
Signaling, 44, 311
Signal transduction, 233257
Sinapic acid, 8, 9, 12, 22, 82, 441,
442, 444, 451
Singlet oxygen, 168
Small intestine, 363, 366, 367, 371,
372, 375, 380, 448
Solanaceae, 7
Soybean, 56, 57
Spinacetin, 82
Spinach, 12, 15, 17, 26, 7376, 81,
82, 210
St. Johns wort (see Hypericum
perforatum),
Stomach, 365
Strawberry, 5, 9, 26, 7379, 80, 210
Succinoxidase (see Complex II)
Sulfation, 373, 380, 414, 416 ,
418419, 422423
Sulfotransferase, 356, 358, 422423
Index
Superoxide anion (see Superoxide
radical)
Superoxide radical, 71, 179, 182,
187, 277, 281
Synthesis
enantioselective, 124
hemisynthesis, 124
Syringic acid, 36
Tannin, 6, 32, 54, 82, 169
condensed, 171
hydrolysable, 25, 181
Taxifolin, 278, 303, 312, 320, 323, 364
Tea, 5455, 88, 89, 181, 184185,
207, 211, 212, 214, 215, 218,
396, 397, 413, 415, 416, 419,
421, 425, 427, 429, 430
TEAC (Trolox equivalent antioxidant
capacity), 7274, 8687
Terpenoid, 217
Theaavin, 185, 288
Tomato, 5, 7, 9, 10, 17, 18, 20, 22,
7376, 83, 89
Thioredoxin peroxidase, 304305
Transepithelial electrical resistance, 357
Transferrin receptor protein, 305
Tricetinidin, 100
3, 5, 7-trihydroxy avone, 277
Trolox equivalent antioxidant
capacity (see TEAC)
467
Tubulin, 304305
Turmeric, 213
Tyrosinase, 448
Tyrosine ammonia-lyase, 20
Tyrosol, 187
UGT (uridine diphosphate
glucuronosyl transferase),
355, 358, 422
UGT1A1, 355
Vaccinium myrtillus, 53, 54
Valerolactone, 377, 420
Vanillic acid, 36, 442, 444,
446447, 449
Vanilloyl glycine, 442, 449
Verbascoside, 14
P-vinyl guaiacol 25, 30
Vitamin C, 71, 8586, 88, 89
Vitamin E, 168169, 185
Vitis Vinifera, 18, 2021, 5960, 61
Wine, 26, 89, 171178, 181, 188,
207, 214216, 218, 396, 413,
425, 427429, 391
Wogonin, 313
Xanthine oxidase, 43, 306

Flavoids in Health and Disease (Catherine Rice Evans-Lester Packer) PDF

Uploaded by

Copyright:

Available Formats

Flavoids in Health and Disease (Catherine Rice Evans-Lester Packer) PDF

Uploaded by

Document Information

Original Title

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

Flavoids in Health and Disease (Catherine Rice Evans-Lester Packer) PDF

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

What topics are covered in the book?

What topics are covered in the book?

flavonoids in

Catherine fl. Rice-Euans

MARCEL DEKKER, INC.

NEW YORK BASEL

All Rights Reserved.

OXIDATIVE STRESS AND DISEASE

LESTER PACKER, PH.D.

Oxidative Stress in Cancer, AIDS, and Neurodegenerative Diseases,

Biothiols in Health and Disease, edited by Lester Packer and Enrique

Additional Volumes in Preparation

Flavonoids and avonoid-rich botanical extracts increasingly have

Series Introduction (Lester Packer and Enrique Cadenas)

Occurrence and Analysis

1. Phenolic Acids in Fruits and Vegetables

Characterization, Chemical, and Biological Properties

5. Synthesis, Identication, Quantication and Chemical

6. Investigation of Flavonoids and Their In Vivo Metabolite

7. Eects of Flavonoids on the Oxidation of Low-Density

Cell Interactions In Vivo and In Vitro

8. Phytochemicals and Brain Aging: A Multiplicity of Eects

Mitochondrial Actions of Flavonoids and Isoavonoids

11. Gene Modulation of HaCaT Cells Induced by Pine Bark

Absorption, Metabolism, and Bioavailability

13. Understanding the Bioavailability of Flavonoids Through

14. Metabolism in the Small Intestine and Gastrointestinal Tract

15. Absorption of Quercetin Glycosides

16. Bioavailability of Flavanol Monomers

17. Absorption and Metabolism of Hydroxycinnamates

Michael Aviram, D.Sc. Lipid Research Laboratory, Technion Faculty of

Lille University of Science and Technology, Villeneuve

Andrea J. Day Procter Department of Food Science, University of Leeds,

James A. Joseph, Ph.D. Jean Mayer USDA Human Nutrition Center on

Farmacia Dr. Carlo Bravi, Brescia, Italy

Piergiorgio Pietta, Ph.D. Institute of Biomedical TechnologiesNational

Teaching Hospital of Brabois, Vandoeuvre, France

Lille University of Science and Technology, Villeneuve

Jaya b. Vaidyanathan Department of Cell and Molecular Pharmacology

Unilever Health Institute, Vlaardingen,

Richard A. Walgren, M.D., Ph.D. Department of Cell and Molecular

Institute of Food Research, Norwich, England

Sheila Wiseman, Ph.D.

Unilever Health Institute, Vlaardingen, The Neth-

Kuresh A. Youdim, Ph.D. Wolfson Center for Age-Related Diseases,

Fleuriet and Macheix

Phenolic Acids in Fruits and Vegetables

OCCURRENCE IN FRUITS AND VEGETABLES

Fleuriet and Macheix

Figure 1 Chemical structure of hydroxybenzoic acids and some derivatives

Contents of Hydroxybenzoic Acids in Vegetables and Ripe Fruitsa

Phenolic Acids in Fruits and Vegetables

Fleuriet and Macheix

Phenolic Acids in Fruits and Vegetables

fruits of certain Solanaceae (tomato, eggplant, pepper), Cucurbitaceae (melon,

Fleuriet and Macheix

Figure 2 Chemical structure of hydroxycinnamic acids and some common