The Role of the Carotenoid Lycopene as an Antioxidant to

Decrease Osteoporosis Risk in Women: Clinical and in

vitro Studies

by

Erin Shea Mackinnon

A thesis submitted in conformity with the requirements

for the degree of Doctor of Philosophy
Institute of Medical Science
University of Toronto

© Copyright by Erin Shea Mackinnon 2010

 The Role of the Carotenoid Lycopene as an Antioxidant to
Decrease Osteoporosis Risk in Women: Clinical and in vitro
Studies

Erin Shea Mackinnon

Doctor of Philosophy

Institute of Medical Science

University of Toronto

2010

Abstract
Lycopene is a potent carotenoid antioxidant shown to decrease the risk of chronic diseases

associated with oxidative stress and has recently begun to be studied in relation to osteoporosis.

However, studies specifically associating intervention with lycopene and a decreased risk for

osteoporosis have not yet been conducted, and the mechanisms by which lycopene affects bone

have yet to be elucidated. The purpose of this thesis was to explore the hypotheses that

supplementation with lycopene would increase antioxidant capacity while decreasing oxidative

stress parameters; subsequently decreasing bone turnover markers, and thus the risk of

osteoporosis in postmenopausal women. Specifically, experiments were designed to determine

whether lycopene acts in its antioxidant capacity to improve bone health, and to delineate the

mechanisms of these effects. These hypotheses were investigated through a cross-sectional

study, a randomized controlled clinical study, and in vitro studies on human osteoblast cells. The

results presented in this thesis demonstrate that intervention with the potent antioxidant lycopene

significantly increased concentrations of the 5-cis isomer and resulted in significantly decreased

oxidative stress parameters in postmenopausal women. This decrease in oxidative stress

parameters resulted in significantly decreased concentrations of the bone resorption marker

ii
crosslinked N-telopeptides of type I collagen (NTx). The typical diet of participants included a

relatively low intake of lycopene, and the corresponding serum lycopene concentrations were not

as effective in decreasing biomarkers of oxidative stress and bone resorption as those obtained

from supplementation with lycopene to increase 5-cis serum lycopene. Studies on the

paraoxonase enzyme suggest that lycopene is most effective in quenching oxidative stress to

decrease bone turnover markers when the internal antioxidant defenses are insufficient or

decremented. Mechanisms demonstrated by the in vitro findings suggest that cis lycopene is

capable of both preventing and repairing the damaging effects of oxidative stress in osteoblasts.

Overall, this thesis provides evidence that lycopene acts through its antioxidant capacity to

decrease oxidative stress parameters and bone turnover markers, and may, therefore, reduce the

risk for osteoporosis. Based on these findings, the consumption of lycopene by women to

improve overall bone health should be considered.

iii
 Acknowledgements

I would like to thank Dr. Leticia Rao (aka “Letty”), my supervisor. Thank you for being the greatest teacher I

have ever had; you have taught me so much. Your endless support and commitment to this project were

essential to the completion of this thesis. Your dedication to your work is truly inspirational.

I would also like to extend my gratitude to Dr. A. Venket Rao, my co-supervisor and the lycopene

guru. Your thoughtful comments and suggestions provided me with great incentive to answer the important

questions. Your encouragement and contribution to this work is deeply appreciated. Thank you as well to the

members of my committee, Drs. A. El-Sohemy and R.G. Josse for all of their help with the completion of this

thesis. Dr. El-Sohemy, your advice and input, particularly on the polymorphism work of this thesis was

invaluable. Dr. Josse, thank you for always making time to provide guidance and support when needed and for

sharing valuable insights on the work embodied in this thesis.

Funding is shared by the Canadian Institutes of Health Research (CIHR) and the Research and

Development Departments of Genuine Health Inc., the H.J. Heinz Co, Millenium Biologix Inc. (Canada),

Kagome Co. (Japan) and LycoRed, Ltd. (Israel). We gratefully acknowledge Amy Strauss, Dr. C Derzko and

Karl BruckMueller, for allowing us access to their list of patients who had signed consent forms indicating

interest in clinical studies together with the following students who assisted with participant recruitment: A.

Dias, T. Huang, M. Maksimowski, M. Simms and K. Zarudny. Thank you to S. Ho for assistance with the

dietary analysis and Ms. Hyeon-Joo Lee for lending her teaching expertise on allelic discrimination. Thank you

to Dr. Bhavnani at the core lab of St. Michael's Hospital for the use of the Fluorskan Ascent machine and the

StepOnePlus™ Real-Time PCR system. Thank you to Steve Allen at WaterSolutions Inc., for his generous

donation of the cis lycopene and for analyzing the lycopene used for tissue culture techniques with HPLC. A

very special thanks to Dr. Ru Li, not only for sharing your lab space but especially for your valued friendship,

advice and the opportunities you have afforded me over the past 3 years, it was a pleasure to work with you.

A special and grateful thank you to all of the family and friends who donated much more than their

support: T. Mackinnon, J. Clark, S. Mackinnon, D. Anderson, K. Connor, L. Brooking, N. Solman, S. Soberg,

K. Clark and S. Ware.

iv
 Thank you to the entire extended Collins and Mackinnon clans – to have so many family members

who are such excellent cheerleaders really is such a privilege. Have you ever seen such a big cheerleading

team?! What a great support system.

To KC, my dear, dear friend- one thing I know for sure is this journey would have been very different

if you were not a part of it. The breakfast therapy sessions, pedicures, relaxing dinners and now the marathon

phone calls from NZ…thank you for being such a wonderful friend and kindred spirit. Rebecca, thank you for

always making time for coffee talks and last-minute lunches, and for William who always makes me happy!

Ali, thank you for always understanding me so well- having you beside me during this always made things

better. Emma and Matthew, your smiling faces, hugs, infectious giggles and time spent with you was the best

stress relief an Auntie could hope for! Thank you to Julie, Ali, Elizabeth and Karissa, for stress-relieving

dinners out and girls-night-in.

For my mother- and father-in-law, Stan and Sherry, you are both extremely supportive and kind.

Thank you both for being there for the triumphs, now matter how small. Stan, the biggest lycopene believer

and advocate – thank you for having as much faith in the “powers” of this molecule as I do.

Elizabeth, having a sister really is one of the greatest things in life -thank you for back rubs, concert

dates, offering endless sympathy and for always being on my side.

For my Mom and Dad, who have been with me on this every step of the way- thank you for being

such wonderful role models and for always believing in me. You taught me how hard work and perseverance

can take you far in life and helped me to never give up. Thank you for always, always being there - comforting

me, offering encouragement, listening for hours on end and showing me that dreaming big and aiming high is

important.

This thesis is dedicated with much love to my husband, Christopher. You have worked so hard, in so

many different ways, so I could pursue this goal for us and for our future. Thank you for taking such care of

me - you always know when I need you and when I need to relax! Thank you for understanding why I wanted

to do this, and for supporting me in all things big and small. You are giving me such a beautiful life – you are

an amazing man and best friend, a true partner. We did it!

v
 TABLE OF CONTENTS

PAGE

ABSTRACT ii

ACKNOWLEDGEMENTS iv

LIST OF TABLES xi

LIST OF FIGURES xiv

LIST OF APPENDICES xvii

LIST OF ABBREVIATIONS xviii

CHAPTER 1 1

INTRODUCTION

SECTION I: POSTMENOPAUSAL OSTEOPOROSIS 2

A. General introduction 2
B. Bone biology: remodelling and bone turnover markers 3
C. Role of bone turnover markers as determinants of
fracture risk and response to therapy 8
D. Risk factors for osteoporosis 12
E. Pharmaceutical and nutritional recommendations 13

SECTION II: OXIDATIVE STRESS 15

A. Chemistry and mechanisms of action 15
B. Oxidative stress and the risk of osteoporosis 19
C. The ability of antioxidants to decrease oxidative stress
and subsequently the risk of osteoporosis 21

SECTION III: LYCOPENE 26

A. Structure and sources of lycopene 26
B. Safety and toxicity 29
C. Bioavailability and metabolism 30
D. Mechanisms of lycopene action 34
E. Lycopene and its effect on chronic disease 35
F. Lycopene and its effect on bone 37

vi
SECTION IV: POLYMORPHISMS AND NUTRACEUTICAL
INTERVENTIONS 40
A. The relevance of polymorphism studies in nutritional
research 40
B. Polymorphism studies related to oxidative stress and
the effect on osteoporosis risk 41
C. The paraoxonase enzyme: its antioxidative capacity and
the role of its polymorphisms in the development of
osteoporosis 43

CHAPTER 2 47

HYPOTHESES, OBJECTIVES AND THESIS OVERVIEW

Hypotheses 48
Objectives and specific aims 49
Overview of thesis chapters 3-9 and their application to the
hypotheses 50

CHAPTER 3 53

IN-DEPTH ANALYSIS OF LYCOPENE INTAKE AND SERUM

LYCOPENE CONCENTRATIONS IN WOMEN AGED 25-70

Abstract 54
Introduction 56
Methods 57
Results and discussion 61

CHAPTER 4 77

DIETARY RESTRICTION OF LYCOPENE FOR A PERIOD OF ONE

MONTH RESULTED IN SIGNIFICANTLY INCREASED BIOMARKERS
OF OXIDATIVE STRESS AND BONE RESORPTION IN
POSTMENOPAUSAL WOMEN

Abstract 78
Introduction 79
Methods 81
Results 85
Discussion 90

vii
CHAPTER 5 95

SUPPLEMENTATION WITH THE ANTIOXIDANT LYCOPENE

SIGNIFICANTLY DECREASED OXIDATIVE STRESS PARAMETERS
AND THE BONE RESORPTION MARKER CROSSLINKED N-
TELOPEPTIDE OF TYPE I COLLAGEN IN POSTMENOPAUSAL
WOMEN AT RISK OF OSTEOPOROSIS

Abstract 96
Introduction 97
Methods 98
Results 102
Discussion 112

CHAPTER 6 119

SUPPLEMENTATION WITH LYCOPENE IN POSTMENOPAUSAL

WOMEN RESULTED IN SERUM CONCENTRATIONS OF 5-CIS
LYCOPENE WHICH WERE HIGHER THAN USUALLY OBTAINED
THROUGH THE DAILY DIET AND WERE ASSOCIATED WITH
LOWER BIOMARKERS OF LIPID PEROXIDATION AND BONE
RESORPTION

Abstract 120
Introduction 121
Methods 122
Results 125
Discussion 130

CHAPTER 7 135

PARAOXONASE 1 POLYMORPHISMS 172TA AND 584AG

MODIFIED THE ASSOCIATION BETWEEN SERUM
CONCENTRATIONS OF THE ANTIOXIDANT LYCOPENE AND BONE
TURNOVER MARKERS AND OXIDATIVE STRESS PARAMETERS IN
WOMEN 25-70 YEARS OF AGE

Abstract 136
Introduction 137
Methods 139
Results 142
Discussion 151

viii
CHAPTER 8 156

SUPPLEMENTATION WITH LYCOPENE RESULTED IN DECREASED

LIPID PEROXIDATION PARAMETERS, WHICH INTERACTED WITH
THE PON1 172TA GENOTYPE TO DECREASE THE BONE
RESORPTION MARKER NTx IN POSTMENOPAUSAL WOMEN

Abstract 157
Introduction 158
Methods 160
Results 163
Discussion 170

CHAPTER 9 174

CIS LYCOPENE ISOMERS FOUND IN HIGH CONCENTRATIONS IN

HUMAN SERUM AND WHICH POSSESS THE GREATEST
ANTIOXIDANT CAPACITY, WERE CAPABLE OF PREVENTING AND
REPAIRING THE DAMAGING EFFECTS OF REACTIVE OXYGEN
SPECIES IN HUMAN OSTEOBLAST CELLS

Abstract 175
Introduction 177
Materials and Methods 179
Results 185
Discussion 196

CHAPTER 10 202

GENERAL DISCUSSION AND CONCLUSIONS

Summary of experimental findings 203

Discussion of experimental findings and their relevance to
the hypothesis and the current body of knowledge 208
Implications of experimental findings 214
Limitations and difficulties encountered 215
Future Directions 216
Conclusions 220

LIST OF REFERENCES 221

APPENDICES 237

APPENDIX I 237

Informed Consent package for clinical studies

ix
APPENDIX II 247

Summary of intervention effects on oxidative stress

parameters, bone turnover markers and antioxidant
capacity resulting from supplementation with regular
tomato juice, lycopene-rich tomato juice and
tomato lycopene capsules

APPENDIX III 249

Example diagrams of HPLC chromatogram and allelic

discrimination plots used to determine PON1 genotype

APPENDIX IV 253

Trend graphs for oxidative stress parameters,

antioxidant capacity, antioxidant enzymes and NTx
from baseline through to the end of the lycopene
intervention study period: combined cross-sectional and
intervention study data

APPENDIX V 259

Comments on the PON1 polymorphisms of the

human CD34+ cells used for the in vitro work
presented in Chapter 9

x
 List of Tables

TABLES PAGE

CHAPTER 1

Table 1.1: Biochemical markers of bone turnover including bone

formation and bone resorption markers. 7

Table 1.2: List of factors resulting from increased bone turnover

which contribute to loss of BMD, increased fragility and risk of fracture. 12

Table 1.3: Lycopene content of foods rich in lycopene, shown in g/g and
mg per serving. 29

CHAPTER 3

Table 3.1: Summary statistics for study participants stratified into two
groups according to age. 62

Table 3.2: Percentage of study participants in each age group who smoked
cigarettes, consumed vitamins or medications and were menopausal. 62

Table 3.3: Consumption of lycopene, calories, caffeine and alcoholic

beverages for each group. 63

Table 3.4: Summary statistics of lycopene intake and serum lycopene for
each age group. 67

Table 3.5: Serum lycopene and consumption of lycopene foods for each
age group. 74

CHAPTER 4

Table 4.1: Participant characteristics as determined at the beginning of the

study. 85

Table 4.2: Lycopene restriction decreased serum carotenoid

concentrations. 87

Table 4.3: Carotenoid content of most commonly consumed lycopene-

containing foods, demonstrating the amount of individual carotenoids
present in g/g. 92

xi
CHAPTER 5

Table 5.1: Participant characteristics and baseline values for oxidative

stress parameters, bone turnover markers and antioxidant capacity for
each supplement group. 105

Table 5.2: Carotenoid content of the supplements given to participants. 106

Table 5.3: Mean AUC as a measure of lycopene absorption for all

three sources of lycopene and placebo supplements. 106

Table 5.4: Mean serum carotenoid concentrations for all participants for
the duration of the study. 107

CHAPTER 6

Table 6.1: Characteristics for each participant group. 127

Table 6.2: Comparison of lycopene values, oxidative stress parameters,

and bone turnover markers between women who were supplemented with
lycopene to those who obtained a low lycopene intake from their usual
daily diet. 128

Table 6.3: Comparison of lycopene values, oxidative stress parameters,

and bone turnover markers between women who were supplemented with
lycopene to those who obtained a higher than average lycopene intake from
their usual daily diet. 129

CHAPTER 7

Table 7.1: Participant characteristics for each genotype showing values

for bone turnover markers, oxidative stress parameters and other
descriptive statistics. 144

Table 7.2: Association between serum lycopene concentration and

bone turnover markers, oxidative stress parameters and antioxidant
capacity for PON1 172TA and 584AG genotypes. 145

Table 7.3: Distribution of participants according to combined genotypes

of the 172TA and 584AG polymorphisms. 148

Table 7.4: Participant characteristics for each combined genotype

showing values for bone turnover markers, oxidative stress parameters
and other descriptive statistics. 148

xii
 Table 7.5: Association between serum lycopene and bone turnover
markers, oxidative stress parameters and antioxidant capacity for
PON1 172TA and 584AG genotypes 149

CHAPTER 8

Table 8.1: Participant characteristics for 172TA genotype showing

values for bone turnover markers, oxidative stress parameters and
other descriptive statistics. 166

Table 8.2: Number of participants in each supplement group, sorted

according to 172TA genotype. 166

Table 8.3: Mean serum carotenoid concentrations for all participants for
the duration of study. 167

Table 8.4: Changes in antioxidant capacity, oxidative stress parameters,

and bone turnover markers for participants supplemented with lycopene, 168
grouped according to 172TA genotype.

Table 8.5: Association between change in TBARS and change in NTx

according to 172TA genotype. 169

CHAPTER 9

Table 9.1: Isomeric content of the three types of lycopene used in this
study. 180

Table 9.2: The effect of varying concentrations of H2O2 on cell viability

after 3 hours. 187

Table 9.3: Concentration of lycopene present in the medium and

treated cells as determined from HPLC analysis. 190

APPENDICES

Table II.1: Summary of individual supplement effects of regular tomato

juice, lycopene rich tomato juice, or tomato lycopene capsules. 248

xiii
 List of Figures

FIGURES PAGE

CHAPTER 1

Figure 1.1: Schematic of formation of reactive oxygen species formation,

including points of antioxidant enzyme action. 18

Figure 1.2: Major isomers of the antioxidant lycopene. 28

CHAPTER 3

Figure 3.1: Lycopene intake was significantly and positively associated

with total serum lycopene in female participants between the ages of 25 to
70 years. 67

Figure 3.2: Frequency of lycopene consumption for each age group. Raw
tomatoes were the most common source of lycopene. 69

CHAPTER 4

Figure 4.1: all-trans, 5-cis and other-cis lycopene were significantly decreased
after lycopene restriction. 86

Figure 4.2: The endogenous antioxidant enzymes, CAT and SOD, were
significantly decreased in postmenopausal participants who refrained from
lycopene consumption for a period of 1 month. 88

Figure 4.3: The endogenous antioxidant enzyme GPx was significantly

increased in postmenopausal participants who refrained from lycopene
consumption for a period of 1 month. 88

Figure 4.4: Concentrations of the bone resorption marker, NTx, were

significantly increased in postmenopausal participants who refrained
from lycopene consumption for a period of 1 month. 89

CHAPTER 5

Figure 5.1: LYCOPENE-supplemented participants had significantly

increased total antioxidant capacity after 4 months of supplementation. 107

Figure 5.2: The increased total antioxidant capacity for the

LYCOPENE-supplemented participants was opposite that seen in participants
supplemented with placebo capsules. 108

xiv
 Figure 5.3: LYCOPENE-supplemented participants had significantly
decreased protein oxidation after four months. 108

Figure 5.4: LYCOPENE-supplemented participants had significantly

decreased lipid peroxidation after four months. 109

Figure 5.5: After 4 months of supplementation, the change in protein thiols

and lipid peroxidation seen in LYCOPENE-supplemented participants was
significantly different from the change seen in placebo-supplemented
participants. 110

Figure 5.6: LYCOPENE-supplemented participants had significantly

decreased in NTx after 2 and 4 months. 111

Figure 5.7: The decreased NTx in LYCOPENE-supplemented participants

was significantly opposite from the increased NTx in participants
supplemented with placebo after 2 and 4 months. 111

CHAPTER 7

Figure 7.1: Participants with the 0/1 genotype in the high serum lycopene
group had significantly lower concentrations of the lipid peroxidation
marker TBARS and the bone turnover markers BAP and NTx than
participants in the low serum lycopene group. 150

CHAPTER 8

Figure 8.1: Participants with the 172TT genotype had significantly

lower NTx concentrations after lycopene supplementation than carriers
of the 172A allele. 169

CHAPTER 9

Figure 9.1: H2O2 generated the production of ROS. 187

Figure 9.2: Treatment with H2O2 for 3 hours significantly decreased number
and area of mineralized nodules. 188

Figure 9.3: Treatment with H2O2 for 9 days significantly decreased number
and area of mineralized nodules. 188

Figure 9.4: Pre-treatment with 45:55 and 28:72 cis:trans lycopene

significantly decreased the H2O2 –stimulated generation of ROS. 192

Figure 9.5: Pre-treatment with 45:55 and 28:72 cis:trans lycopene prevented
the negative effect of decreased area of mineralized nodules by H2O2. 193

xv
 Figure 9.6: Post-treatment with 45:55 or 28:72 cis:trans lycopene decreased
the generation of ROS stimulated H2O2. 194

Figure 9.7: Post-treatment with 45:55 or 28:72 cis:trans lycopene repaired

the negative effect on number of mineralized nodules caused by H2O2. 195

Figure 9.8: Post-treatment with 45:55 or 28:72 cis:trans repaired

the negative effect on total area of mineralized nodules caused by H2O2. 195

CHAPTER 10

Figure 10.1: Summary diagram of combined experimental findings presented

in this thesis. Red arrows illustrate points of lycopene action shown to prevent
the risk of osteoporosis. 207

APPENDICES

Figure IIIa: Example chromatogram showing peaks for the individual

carotenoids. 250

Figure III b: Example of allelic discrimination plot used for genotyping

the 172TA polymorphism. 251

Figure IIIc: Example of allelic discrimination plot used for genotyping

the 584AG polymorphism. 252

Figure IVa: Trend graphs showing change in antioxidant capacity from

baseline, through to the end of the lycopene supplement period. 256

Figure IVb: Trend graphs showing change in oxidative stress parameters,

Trolox and NTx from baseline, through to the end of the lycopene
supplement period. 257

Figure IVc: Trend graphs showing change in antioxidant enzymes

from baseline, through to the end of the lycopene supplement period. 258

xvi
 List of Appendices

APPENDICES PAGE

APPENDIX I

Informed Consent package for clinical studies 236

APPENDIX II

Summary of intervention effects on oxidative stress parameters, bone turnover

markers and antioxidant capacity resulting from supplementation with regular
tomato juice, lycopene-rich tomato juice and tomato lycopene capsules 247

APPENDIX III

Example diagrams of HPLC chromatogram and allelic discrimination plots used

to determine PON1 genotype 249

APPENDIX IV

Trend graphs for oxidative stress parameters, antioxidant capacity, antioxidant

enzymes and NTx from baseline through to the end of the lycopene
intervention study period: combined cross-sectional and intervention study data 253

APPENDIX V

Comments on the PON1 polymorphisms of the human CD34+ cells used

for the in vitro work presented in Chapter 9 259

xvii
 List of Abbreviations

A
AAP ascorbic acid phosphate
ABTS 2, 2, azinobis (3-ethylbenzothiazoline-6-sulfonic acid)
ABTS 2, 2, azinobis (3-ethylbenzothiazoline-6-sulfonic acid) radical cation
ALP alkaline phosphatase
ANOVA One-way Analysis of Variance
Arg arginine
AUC area under curve

B
BAP bone alkaline phosphatase
BMD bone mineral density
BCE bone collagen equivalents
GP -glycerophosphate
BHT butylated hydroxy toluene
BMI body mass index
BMU(s) basic multicellular unit(s)
BSP bone sailoprotein
BTM(s) bone turnover marker(s)

C
CAT catalase
CM-H2DCFDA 5-(and 6) chloromethyl-2’7’-dichlorodihydrofluorescein diacetate acetyl
ester
CMOII carotenoid monooxygenase II
CO2 carbon monoxide
COLIA1 collagen 1  1
CTx crosslinked C-telopeptides of type I collagen
CuZn-SOD copper zinc superoxide dismutase

D, E, F, G

ddH2O double distilled water

Dex dexamethasone
D-Pyr deoxypyridinoline
DTNB 5-5’-dithio-bis (2-nitrobenzoic acid)
DXA dual X ray absorptiometry
EDTA disodium ethylene diamine tetraacetic acid
ELISA enzyme-linked immunosorbent assay
FBS fetal bovine serum
FDA Food and Drug Administration

xviii
Gln glutamine
GLM general linear model
GPx glutathione peroxidase
GSH glutathione
GSSG oxidized glutathione disulfide

H, I, J, K
H2O water
H2O2 hydrogen peroxide
Hb hemoglobin
HCl hydrochloric acid
HDL high-density lipoprotein
HMG-CoA hydroxymethylglutaryl coenzyme A
HPLC high-performance liquid chromatography
HRP horseradish peroxidase
HRT hormone replacement therapy
IL-1/6 interleukin-1/6
IRMA immunoradiometric assay
KRB Kreb’s ringer buffer

L, M

LDL low-density lipoprotein

Leu leucine
MBNF mineralized bone nodules
MDA malondialdehyde
MeOH absolute methanol
Met methionine
MAPK mitogen-activated protein kinase
Mn-SOD manganese superoxide dismutase

Nf nuclear factor-

NADPH nicotinamide adenine dinucleotide phosphate
nM BCE nanomoles Bone Collagen Equivalents per litre
Nrf2 nuclear factor E2 p45-related factor 2
NTx crosslinked N-telopeptides of type I collagen

O, P

O2 superoxide radical

OB osteoblasts
OCN osteocalcin
OD optical density

xix
OH hydroxyl radical
OSL Observed Safe Level
OPG osteoprotegerin
PBS phosphate buffered saline
PICP procollagen type I C-terminal propeptide
PINP procollagen type I N-terminal propeptide
PON1 paraoxonase 1
PON -/- paraoxonase 1 knockout
PON1 Tg paraoxonase 1 transgenic
ppm parts per million
PTH parathyroid hormone
Pyr pyridinoline

Q, R, S

RANK receptor activator of nuclear factor 

RANKL receptor activator of nuclear factor  ligand
RBC red blood cells
RFU relative fluorescence units
RIA radioimmunoassay
ROS reactive oxygen species
Runx2 runt-related transcription factor 2
SaOS2 human osteosarcoma osteoblast-like cells
SERM(s) selective estrogen receptor modulator(s)
SNP single nucleotide polymorphism
SOD superoxide dismutase
SH sulfhydryl

TAC total antioxidant capacity

TBA thiobarbituric acid
TBA-MDA thiobarbituric acid-malondialdehyde
TBARS thiobarbituric acid reactive substances
TEAC Trolox equivalent antioxidant capacity
THF tetrahydrofuran
TNF tumor necrosis factor
TRAP5b type 5 tartrate resistant acid phosphatase

U, V, W, X, Y, Z

USDA United States Department of Agriculture

VDR vitamin D receptor
VLDL very low-density lipoprotein
WHO World Health Organization
Zn-SOD zinc superoxide dismutase

xx
CHAPTER 1

Introduction

1
SECTION I: Postmenopausal osteoporosis

IA: General introduction

Osteoporosis is a systemic disease characterized by low bone mass and structural deterioration of

the microarchitecture of bone, resulting in an increased risk of fracture. Loss of bone usually

occurs without symptoms, which is why osteoporosis is often referred to as the “silent thief”.

Osteoporotic fractures are associated with chronic, immobilizing pain and greatly impact the

quality of life and the ability to perform daily functions. The probability that a Caucasian women

over the age of 50 will fracture a hip is 14%, and 1 out of 5 patients do not survive one year after

having had a hip fracture 1. Early postmenopausal women lose bone at an average rate of 0.5-2%

per year 2.

Measurement of bone mineral density (BMD) is typically performed by dual X ray

absorptiometry (DXA), which provides a static, complete measure of the skeletal status. BMD is

measured at various skeletal sites, including the spine, hip and wrist 3. Central DXA of the hip

and spine provides the best predictive evidence for fracture risk 2. Using DXA measurements,

BMD is expressed in relation to the value expected for the patient’s age and sex, referred to as

the Z-score, and in relation to the expected value for young adults of the same sex and race,

referred to as the T-score 3. The World Health Organization (WHO) has established the

following diagnostic for osteoporosis criteria based on BMD T-scores: normal BMD (T-score

between +2.5 and -1.0), osteopenia/low BMD (T-score between -1.0 and -2.5, inclusive),

osteoporosis (T-score lower than -2.5) and severe osteoporosis (T-score lower than -2.5 with the

presence of one or more fragility fractures) 4.

2
IB: Bone biology: remodelling and bone turnover markers

Bone is a dynamic tissue that is continuously being remodeled as old bone is removed and new

bone is formed. This process of bone turnover maintains mineral homeostasis, skeletal size and

mechanical strength, and provides a mechanism by which damaged bone can be repaired 5, 6. In

normal, healthy states apart from growth, this process does not result in a net loss or gain of

bone. Remodelling occurs in discrete microscopic units of bone called basic multicellular units

(BMUs) 7. The basic components of bone include an organic matrix of protein, which has been

produced by osteoblasts, newly formed matrix mineralized over time by calcium and

hydroxyapatite 6. The major cells in BMUs are the osteoblasts, osteoclasts and osteocytes. The

osteoclasts are responsible for resorbing old bone, the osteoblasts for synthesizing new bone and

the osteocytes for transduction of signals necessary to sustain mechanical loads. The process of

bone turnover is closely coupled and mediated primarily by osteoblasts which are activated by

various hormones, growth factors and cytokines 7. Systemic hormones which control the

turnover process include estrogen, androgen, parathyroid hormone (PTH), follicle-stimulating

hormone and thyroid-stimulating hormone 6. Osteoblasts signal osteoclasts to differentiate and

mature, subsequently stimulating resorption of bone within the BMU via the receptor activator of

nuclear-factor-kappa (RANK) and its ligand (RANKL) 7. RANKL is a tumor necrosis factor

(TNF) cytokine produced by osteoblasts which binds to RANK on the surface of osteoclasts.

Another member of the TNF family, osteoprotegerin (OPG), acts as a decoy receptor to block

RANKL from binding to its receptor and thus moderates and quenches osteoclast resorption

activity 6.

Osteoclasts resorb bone by secreting hydrogen ions and lysosomal proteases, such as

cathepsin K, which are capable of degrading the bone matrix 6. The process of bone resorption

occurs for approximately 2-3 weeks, whereupon osteoclasts undergo apoptosis, and osteoblasts

3
migrate to the resorption cavity to lay down a cement line and begin the synthesis of new matrix.

As this new bone is laid down, some of the osteoblasts become buried within the matrix and

mature into osteocytes. Osteocytes communicate with each other and the bone surface via fluid-

filled channels. The turnover cycle of a BMU takes several months for primary mineralization,

with a secondary stage of mineralization which may occur over several decades 7.

Bone formation markers reflect osteoblast enzyme activity or by-products of collagen

synthesis. These include procollagen I extension peptides, osteocalcin (OCN) and bone specific

alkaline phosphatase (BAP). Bone resorption markers are typically type I collagen degradation

products, including telopeptides of type I collagen and pyridinoline (Pyr) and deoxypyridinoline

(D-Pyr) crosslinks 7. Other resorption markers are indicative of osteoclast activity, such as

RANKL and acid phosphatase, typically measured by type 5 tartrate resistant acid phosphatase

(TRAP5b) concentration 8. The physiological functions of the common bone turnover markers

are listed in Table 1.1. For the purpose of this thesis, BAP and NTx will be described in further

detail as they were the markers used for the studies reported throughout this work.

Alkaline phosphatase (ALP) is an enzyme that catalyzes the hydrolysis of

monophosphate ester groups and is found throughout the body 7. The four main isoenzymes of

ALP include germ cell, placental, intestinal and tissue non-specific, including the bone, kidney

and liver forms 7. ALP is attached to the outside of cell membranes in kidney tubules, intestinal

epithelium, liver, placenta and bone 9. In serum, 95% of ALP is comprised of the bone and liver

isoforms 7, and in tissues, 40-50% of ALP is comprised of the bone isoform 8. BAP is

functionally similar to other types of ALP, but is antigenically different due to post-translational

modification 9.

BAP is an osteoblast enzyme which acts in several ways: (1) as a calcium-binding protein

(2) to increase the concentration of inorganic phosphate and (3) to destroy pyrophosphate which

4
can inhibit crystallization of minerals 7. More specifically, it catalyzes the hydrolysis of

phosphate esters at the surface of the osteoblast, providing high concentrations of phosphate for

the bone mineralization process. It is cleaved by phospholipase to enter the circulation where it is

measured as a bone formation marker indicative of osteoblast activity 9. The use of BAP as a

marker of bone formation in clinical settings provides several advantages as it possesses a low

intra-individual variability (10% in the long term). In addition, BAP has a half-life of 1-2 days in

the circulation which decreases the sensitivity to circadian rhythms and increases the stability of

collected samples. BAP does, however, offer some disadvantage, particularly in patients with

liver diseases, as it demonstrates a cross-reactivity with the liver isoform of approximately 15%
7, 8
.

The major collagen present in bone is type I collagen. In bone, it has a helical structure

that contains an amino (N)-telopeptide and a carboxy (C)-telopeptide 7. These peptides are the

non-helical region of type I collagen where the crosslinks attach 8. During bone resorption, the

collagen crosslinks bound to either the N (NTx)- or C (CTx)-telopeptide, are hydrolyzed by the

osteoclast enzyme, cathepsin K, and released from the bone matrix into the circulation as a
10
marker of bone resorption . The asparatate residue of the N-telopeptide is vulnerable to

racemisation or isomerization, resulting in either -L, -L, -D, -D aspartyl 7. The degree to

which collagen is isomerized can reflect skeletal age; typically the  aspartyl forms are present

in newly-formed, young bone, whereas the isomerized  aspartyl forms are present in older bone
5, 7
. NTx is measured by detecting an epitope of the N-terminal telopeptide of the -2 chain of

type I collagen 8.

NTx is one of the most commonly used measures of bone resorption in clinical settings

and compared to other bone resorption markers offers several advantages, particularly sample

stability as samples can be stored for many years without significant degradation. It is not as

5
sensitive to the diet as other markers of bone resorption, such as CTx, but it is sensitive to

circadian rhythms so its measurement needs to be standardized among participants and between

participant visits. Variability tends to be higher in urine samples 7, which need to be normalized

to creatinine concentrations.

All markers of bone turnover are subject to variability, some of which can be controlled

by careful modification of collection and handling techniques, such as collecting samples at a

specific time of day in the fasting state 10, 11. Analytical variability can be up to 5%, particularly

if special care of the aforementioned controllable variables, such as collection time, is not taken
8
. Seasonal variation can occur particularly during winter months when reduced exposure to

sunlight decreases serum vitamin D status. Other controllable factors include menstrual cycle

and limiting alcohol consumption and exercise prior to collection 10. Sources of variability that

cannot be controlled include age, menopausal status, sex, use of medications, presence of disease
8, 10, 11
and prior or present fracture . Typically, inter-individual variability can be up to 10% for

bone turnover markers measured in serum and 15-25% for markers measured in the urine, which

is why careful standardization of techniques is vital in order to obtain accurate results using

markers of bone turnover 8.

6
Table 1.1: Biochemical markers of bone turnover including bone formation and bone resorption
markers 10-12

Biochemical Markers of Bone Formation

Name
Bone alkaline phosphatase (BAP)
Procollagen type I C-terminal
propeptide (PICP)
Procollagen type I N-terminal
propeptide (PINP)
Osteocalcin (OCN) (bone
gamma-carboxyglutamic protein
{BGP})
Biochemical Markers of Bone Resorption
Name
Cathepsin K
Hydroxyproline1
crosslinked C-telopeptide of type
I collagen (CTx)
crosslinked N-telopeptide of type
I collagen (NTx)
RANKL
Total & free pyridinoline (Pyr)
Total & free deoxypyridinoline
(D-Pyr)
Tartrate resistant acid
phosphatase 5b (TRAP5b)
1
No longer a preferred measure of bone resorption.
Mechanism
Osteoblast enzyme, indicative of mineralization
Collagen based, generated during osteoblastic synthesis of
bone matrix, reflects changes in bone collagen synthesis
Collagen based, generated during osteoblastic synthesis of
bone matrix
Noncollagenous protein, synthesized by osteoblasts as part
of bone matrix
Mechanism
Osteoclast enzyme which cleaves helical and telopeptide
regions of type I collagen
Amino acid unique to collagenous proteins, cleaved during
bone resorption
Collagen based, cleaved by peptidases during bone
resorption, telopeptide fragments generated by cathepsin K,
degradation products of osteoclastic activity
Collagen based, cleaved by peptidases during bone
resorption, telopeptide fragments generated by cathepsin K,
degradation products of osteoclastic activity
Osteoclast formation and function
Collagen degradation product released during bone
resorption
Collagen degradation product released during bone
resorption
Osteoclast enzyme, reflects number and activity of
osteoclasts

7
IC: Role of bone turnover markers as determinants of

fracture risk and response to therapy

Bone turnover is balanced in healthy young adults, with as much bone being formed as is being

reabsorbed. However, during menopause, the process of bone turnover increases and becomes

uncoupled, resulting in a faster rate of bone resorption which cannot be compensated for by bone

formation, resulting in an overall loss of bone 7. Menopause and increasing age are associated

with higher concentrations of bone turnover markers (BTMs) because of sex hormone

deficiency, which can inhibit both 25-hydroxyvitamin D production and calcium absorption from
6, 13
the intestine, resulting in accelerated bone loss . Bone cells express estrogen receptors that,

when bound by estrogen, decrease the production of cytokines responsible for bone resorption
14
and osteoclastogenesis, such as interleukin-1 (IL-1) and –6 (IL-6) . In postmenopausal

osteoporosis, decreased estrogen results in higher circulating levels of cytokines, resulting in an

5, 14
uncoupling of bone resorption and formation, in favour of the former . At the cellular level,

this loss of sex steroids increases the number of remodelling events resulting in resorption

cavities that are not being properly refilled. This causes a negative calcium balance followed by a

net loss of bone from each BMU. Overall, this is evidenced by several factors, such as increased

cortical porosity (Table 1.2), which lead to the rapid loss of BMD, increased fragility and

significantly augmented risk of fracture 5, 6. BTMs greatly increase during menopause, peak one

to three years after ovarian function stops and plateau after approximately eight to ten years 9, 15.
8, 16
In some cases, elevated BTMs can persist up to thirty or forty years postmenopause . It is

reported that estrogen deficiency associated with menopause can increase the loss of trabecular
14
bone by as much as 5% per year . Increases in bone resorption markers associated with

menopause can range from 0-150% coupled with concomittant increases in bone formation

8
ranging from 0-100% 8. Multiple studies have shown an association between menopause,

osteoporosis and elevated BTMs. Typically the concentration of BTMs increases in the rank

order premenopausal<postmenopausal<osteoporotic, with a higher amount of uncoupling

17,
between bone resorption and formation in favour of the former, in women with osteoporosis
18
.

BMD measurements are considered the best method to determine changes in bone health

and osteoporosis risk. However, measurements of BMD by DXA are not dynamic and cannot

predict changes that may occur post-measurement 7, nor do they provide an accurate depiction of
19
whole bone strength . On the other hand, repeated measures of BTMs provide a dynamic
11
analysis of ongoing changes occurring within the skeletal system . In addition, changes in

BMD occur slowly and can take up to one to two years after treatment initiation to be detected 8,
20
. In fact, treatment with antiresorptive medications show a less than 3% change in BMD in 3-6

months, with an increase of only 2-5% per year 8. Furthermore, results from a prospective study

following incident fractures in 668 women, showed that 47% of women who suffered from a

fracture had a BMD which was not considered to be within the osteoporotic range according to

the WHO diagnostic criteria (T-scores of <-2.5) 19.

Thus, the use of bone turnover markers has become increasingly popular in clinical trials

as they can detect changes much earlier in the course of therapy and can be easily measured as

either formation or resorption markers in the urine or serum of participants 8. In addition, they

have been significantly linked to fracture risk in multiple studies 21-23 due to a significant positive

correlation between high BTMs and loss of BMD 7. This correlation is strengthened with

increasing age, such that in women who are  30 years post menopause, bone turnover accounts
16
for 40-50% of the variance in BMD of the whole skeleton . In a longitudinal study,

postmenopausal women classified as “fast bone losers” lost more than 3% of their bone mass per

9
year. After 15 years of follow-up these “fast bone losers” had higher values of the bone turnover

markers OCN and CTx, and decreased body weight and bone mass, as well as significantly
22
increased risk of fracture . A prospective cohort study of women aged 50-88 showed that

participants who had the highest bone turnover (classified as above the upper limit of the

premenopausal range) lost 4 to 6 times more bone from the mid radius and approximately 2

times more bone from the distal radius, than participants who had lower bone turnover (classified

as within the premenopausal range) 24. These findings demonstrate that a fast rate of bone loss 22,
24
and low BMD are important determinants in the risk of osteoporotic fracture. Elevated BTMs

have been shown to be significantly associated with fracture risk above a certain threshold level.

This provided evidence that increased resorption, together with an increased activation frequency
25
of BMUs, could increase fragility over and above that already present due to low BMD .

Garnero et al. have shown that postmenopausal women in the highest quartile of BTM

concentration (D-Pyr, CTx and NTx) had approximately double the risk of fracture compared

with women in lower quartiles. This relative risk is doubled again when high bone resorption is

coupled with the presence of fragility fracture and low BMD, and is significantly higher than the

risk associated with low BMD alone 23. These findings suggest that the combination of BTM and

BMD measurements, together with clinical outcomes such as existing fracture, provide a better

predictor of fragility fracture risk.

BTMs also provide an excellent indicator of the therapeutic response 8. A large

randomized controlled trial by Delmas et al. reported that in postmenopausal osteoporotic

women intervention with intermittent or daily ibandronate (a bisphosphonate medication)

resulted in significantly reduced concentrations of NTx (44.8% and 51.0%, respectively) and

CTx (49.2% and 58.9%, respectively) after the first 3 months of administration. This remained

10
constant for the 3 year study duration and was correlated with significantly decreased risk of

vertebral fracture and significantly increased BMD of the lumbar spine and total hip 21.

The goal of many osteoporosis therapies is to reduce BTMs to within the lower half of

premenopausal reference ranges because premenopausal women between the ages of 30 to 45 do

not usually experience significant bone loss and can be considered to have optimal bone health.

Reference ranges for premenopausal bone turnover markers can be useful tools to identify

women with a higher fracture risk and to monitor the course of pharmaceutical therapies. In

postmenopausal women with osteoporosis, treatment with antiresorptive therapies results in

rapidly decreased concentrations of bone resorption markers, followed by a plateau after 3-6
7, 20
months . As a result of coupling, bone formation markers typically take longer to show a

marked effect and plateau after 6-12 months of treatment. Decreases seen in bone formation

markers are typically 50% of those seen for bone resorption markers 7.

For treatment with anabolics, increases in bone formation are expected after 1-3 months
7
. Premenopausal reference ranges for BTMs have recently been reported to be 9.0-28.8 U/L for

BAP and 13.6-62.7 nmol BCE/mmol creatinine for NTx in Caucasian women aged 30-45 years
20
(N=200) . The values for BAP are similar to those reported in the manufacturer’s protocol of

the ELISA kits used for detection of bone turnover markers in this thesis project, reported at

11.6-29.6 U/L for BAP (N=126) in women of similar age. The NTx reference ranges reported in

the manufacturer’s protocol of the ELISA kits reported the serum range to be 6.2-19.0 nM BCE,

which tends to be lower than that reported by Glover et al. in the urine 20.

11
Table 1.2: List of factors resulting from increased bone turnover which contribute to loss of
BMD, increased fragility and risk of fracture 25
Factors contributing to increased fragility, loss of BMD and risk of fracture as a result
of increased bone turnover
Loss of connectivity, decreased trabecular interconnectivity
Increase in number of unfilled resorption lacunae causing weakened trabeculae
Cortex erosion on endosteal surface resulting in reduced moment of inertia
Increased cortical porosity
Decreased cortical thickness
Negative bone balance at each BMU leads to net increase in bone loss

ID: Risk factors for osteoporosis

Osteoporosis is multi-factorial disease. The major risk factors include age (>65 years), low

BMD, presence of vertebral compression fracture, glucocorticoid use, malabsorption syndrome,

tendency to fall, primary hyperparathyroidism, hypogonadism, early menopause and prior

fragility fractures or family history of fracture. Minor contributors include low dietary calcium,

smoking, excessive alcohol or caffeine intake, weight (<57 kg), weight loss (>10% of weight at

age 25, can be associated with anorexia nervosa or bulimia), rheumatoid arthritis, history of

hyperthyroidism and long-term use of anticonvulsant or heparin therapy 2. Some of the

contributing risk factors, such as smoking and strenuous exercise are associated with increased

oxidative stress, a concept which will be addressed in the upcoming Section IIB 26, 27.

A fragility fracture occurs as a result of minimal trauma 2. The official WHO definition of

fragility fracture is “a fracture caused by injury that would be insufficient to fracture normal

bone: the result of reduced compressive and/or torsional strength of bone” 28. The most common

sites of fragility fracture are the spine, hip and distal forearm 13. The four main factors which can

predict the risk of fracture include a low BMD, age, family history of osteoporosis and prior

fragility fracture. For each standard deviation of BMD below the reference level, the risk of

fracture at the forearm, hip and spine approximately doubles 2, 8.

12
IE: Pharmaceutical and nutritional recommendations

There are two major classes of osteoporosis pharmaceuticals, the anabolic and antiresorptive

treatments. Anabolic agents act to increase bone formation, and consequently bone turnover,

stimulating bone formation. Antiresorptive agents act in the opposite manner to decrease bone
3
resorption, and thus bone turnover, resulting in prevention of overall bone loss .

Bisphosphonates directly target the hydroxyapatite bone mineral surfaces within bone

remodelling sites to inhibit the resorption activity of osteoclasts. Selective estrogen receptor

modulator (s) (SERMs), an estrogen agonist, and hormone replacement therapy (HRT) neutralize

or overcome the effects of bone loss associated with estrogen deficiency 5. Currently approved

antiresorptive treatments for use by the Food and Drug Administration (FDA) include the

bisphosphonates, Alendronate (Fosamax), Ibandronate (Boniva), Risedronate (Actonel) and

Zoledronic acid (Aclasta), salmon calcitonin (Miacalcin, Calcimar, Fortical), the SERM,

Raloxifene (Evista) and HRT (Climara, Estrace, Estraderm, Estratab, Ogen, Ortho-Est, Pramarin,

Vivells, Activella, Femhrt, Premphase, Prempro). Parathyroid hormone, PTH1-34 or teriparatide

(Forteo), is the only anabolic agent currently approved for use by the FDA 3. The new class of

osteoporosis medications currently being investigated for use is a fully human monoclonal

antibody (Denosumab) which would bind to RANKL, imitating the effects of OPG and acting as

an inhibitor of RANKL 3, 5.

Currently there are limited official nutritional recommendations to offset the loss of bone

associated with postmenopause. They include a daily intake of 1500 milligrams of calcium and

800 IU of vitamin D 2. Adequate calcium intake is necessary to achieve peak bone mass and
13
modify the rate of bone turnover . Bone stores calcium and when there is decreased dietary

intake or absorption of calcium, it is mobilized from bone by the action of PTH 6. This leads to

decreased bone mass and increased fragility. Vitamin D is essential for calcium absorption,

13
 13
particularly in cases where dietary calcium intake is limited , and higher intake has been

strongly correlated with decreased fracture risk 3, 13. Less than 50% of women meet the required

calcium intake, and vitamin D insufficiency is common in many women, because it is found in a

very limited number of foods and therefore must be obtained through supplements or exposure to

sunlight 13. Vitamin D deficiency is common in those who have an existing fracture, live in areas

with long winter months during which exposure to sunlight is limited 3 and/or in those who are

elderly due to a reduced ability to produce the active form of vitamin D 6. Vitamin D deficiency

is a well documented risk factor for osteoporosis, as it results in less efficient intestinal calcium

absorption, increased bone loss, muscle weakness and weakened bone microstructure 13.

Although other important micronutrients such as strontium, magnesium, potassium and

vitamin K are emerging in the literature as potential components beneficial to bone 29, 30, further

attention to the nutritional factors which influence bone turnover in the form of randomized

clinical trials are required to suggest appropriate additional methods of osteoporosis prevention
13, 31
. It is possible that a combination of these dietary factors, such as magnesium and zinc,

together with the recommended calcium and vitamin D, would interact to enhance the individual

effects and improve BMD, subsequently reducing the risk of osteoporosis 13.

Recently, epidemiological studies have examined the benefits of consuming a diet rich in

fruits and vegetables, providing high potassium and bicarbonates. Lin et al. 32 have shown that a

1 month intervention with a diet high in bicarbonates and potassium, in the form of fruits and

vegetables, resulted in significantly reduced OCN and CTx. This may be due to an overall

alteration of acid-base homeostasis in favour of alkali producing components to decrease acidity.

It may also be due to components such as -carotene, vitamin C and vitamin K, acting alone or
33
in combination . The lowest incidence of osteoporosis in Europe is reported in the

Mediterranean area. It has been suggested that this is due to the dietary patterns in this region,

14
which typically include a wide variety of bioactive molecules found in fruits and vegetables with
34
anti-inflammatory, antioxidant and alkalinizing properties which are beneficial to bone .

However, thus far, there are limited data from randomized controlled trials to support the idea

that increased consumption of fruits and vegetables in general would be beneficial in reducing

the risk of osteoporosis 6, 33.

SECTION II: Oxidative stress

IIA: Chemistry and mechanisms of action

Reactive oxygen species (ROS) are free radicals containing one or more unpaired electrons. This

makes them highly reactive, because as they seek out another electron to fill their orbital and

stabilize their electron balance, they essentially attack cellular components. The superoxide

radical (O2-) is generated when a single electron is added to the ground state oxygen molecule,

typically generated via electron transport chains in the mitochondria, endoplasmic reticulum and

nuclear membranes 35. Free radicals such as hydrogen peroxide (H2O2), hydroxyl (OH) and O2-

are continuously generated during normal cellular metabolism as well as during acute or chronic

immune responses 36
. For example, the O2- radical is generated in the liver by the desaturase
35
enzyme or by activated leukocytes during the immune response . Other factors which can

generate ROS include cigarette smoke, physical stress, ultraviolet radiation, ischemia-

reperfusion, chemicals, deep fried foods, environmental pollution and toxins and the aging

process 35-37.

In the absence of adequate antioxidant defenses, ROS such as H2O2 and OH are highly

reactive and can cause oxidative damage to macromolecules resulting in oxidation of lipids,

proteins and DNA 38. Oxidative stress is caused by either an increase in the production of ROS
15
and free radicals, or a decrease in the antioxidant defence system 39. In a state of oxidative stress,

antioxidants are low and markers of oxidative damage are abundant. This can lead to cellular

damage and the development of age-related chronic diseases. Free radicals, either

environmentally or internally produced, can be neutralized by antioxidants 35.

There are three lines of defense against ROS and oxidative stress. These include, in order

of use: the preventative antioxidants, the radical scavenging antioxidants and the repair/de novo

antioxidants. The preventative antioxidants include the metal chelating proteins and the

endogenous antioxidant enzymes catalase (CAT), glutathione peroxidase (GPx) and superoxide

dismutase (SOD). The concentrations and locations of these antioxidants are highly regulated

because their main function is to suppress the formation of free radicals to promote cell survival
35
. CAT is an antioxidant enzyme which acts to detoxify H2O2 by converting it into water and
40
oxygen (Figure 1.1). It has one of the highest turnover rates of all enzymes and can convert

millions of H2O2 molecules per second 41. Depressed CAT is associated with an increased risk of

chronic diseases associated with oxidative stress such as atherosclerosis, diabetes,

40
neurodegenerative diseases and postmenopausal osteoporosis . GPx is an antioxidant enzyme

associated with selenium, which is required for its optimal function 42. It is found in the plasma
35
and the cytosolic or membrane compartment of tissues . It catalyzes the conversion of H2O2

into water and oxygen, using reduced glutathione (GSH) as a substrate which is oxidized to

glutathione disulfide (GSSG) in the process (Figure 1.1). SOD catalyzes the dismutation of O2-

to oxygen and H2O2, the latter of which is subsequently neutralized by CAT or GPx into water

and oxygen (Figure 1.1). SOD exists in three isoforms: manganese (Mn-SOD), copper zinc

(CuZn-SOD) and extracellular. SOD is localized to the intracellular space of tissues and is

filtered through the kidney before being reabsorbed and catabolized within the proximal tubules
42
. Free radicals which are not effectively suppressed by these preventative antioxidants initiate

16
peroxidative chain reactions which must then be terminated by the chain-breaking, radical-

scavenging antioxidants 35.

The radical-scavenging antioxidants include the dietary antioxidants, which can act as

singlet oxygen quenchers, radical scavengers, reducing agents, or radical chain-suppressors,

breakers or terminators 35, 42. Examples include the carotenoids, flavonoids and vitamins C and E.

These antioxidants break the chain of oxidation at the site of attack by being preferentially

oxidized themselves, essentially attacking the free radical to prevent oxidation of nearby cellular

components such as proteins, lipids and DNA 35.

The repair/de novo antioxidants include DNA repair enzymes, lipase, protease and

transferase, and they are responsible for repairing damage and reconstituting tissues in cases

where the above preventative and radical scavenging antioxidants were not able to adequately

prevent chain oxidation and oxidative damage 35.

Under normal physiological conditions, the cell can prevent free radical attack by

upregulating the antioxidant defenses used. However, when these defenses are unable to fully

prevent the chain oxidation, tissue damage to biomolecules such as protein, lipids and DNA
35
occurs . This can occur by the induction of apoptosis, reduction of cellular proliferation, cell
43
cycle arrest and modulation of cellular differentiation . Oxidative stress weakens the entire

antioxidant system because all three forms of antioxidants act in complex redox reactions in an

attempt to combat oxidative stress. In so doing, they interact with each other through a series of

non-enzymatic and enzymatic reactions and are decremented or inactivated in the process.

Oxidative stress can also propagate pre-existing damage from disease, infection, or trauma. The

imbalance between oxidation-reduction processes can contribute to disease development by the

35
upregulation of gene expression involved in pathology . Diseases associated with oxidative

stress include cardiovascular disease, cancer, diabetes, neurological diseases and osteoporosis 44,

17
the latter of which is the subject of this thesis and will be addressed in further detail in the

upcoming Sections IIB and IIC.

Many studies show an increase in oxidative stress with age and it is known to be a natural
35, 37
process of aging . However, it is possible that the elevated oxidative stress seen in

postmenopausal women is a direct result of decreased estrogen, as estrogen is known to provide

antioxidant benefits 45. We and others have previously shown that menopause is associated with

higher levels of oxidative stress 39, 46.

Smoking and strenuous exercise are other known causes of oxidative stress. Case-control

studies indicate that markers of DNA, protein and lipid oxidation are elevated in people who
26
smoke cigarettes . Research suggests that strenuous exercise can generate oxidative stress by

the following mechanisms: anoxia-reoxygenation, mechanical damage to muscles, increased

respiration of pollutants, oxidation of catecholamines and leakage of single electrons from the

mitochondria to form O2- 27.

O2 SOD GSH GSSG

GPx
H2O2 H2 O
CAT
O2

OH

Cellular damage to
DNA, proteins, lipids

Figure 1.1: Schematic of formation of reactive oxygen species formation, including points of
antioxidant enzyme action.

18
IIB: Oxidative stress and the risk of osteoporosis

Many research studies have demonstrated that oxidative stress induced by ROS is a risk factor

for osteoporosis, as it increases the rate of bone loss 47-50. At the molecular level ROS affect both

osteoclasts and osteoblasts. In fact, several of the intracellular signals which are involved in

osteoclast formation and osteoblast differentiation and apoptosis are sensitive to ROS, including

nuclear factor- (Nf), phosphatidylinositol 3-kinase, p38 mitogen-activated protein kinase

(MAPK) and c-Jun aminoterminal kinase 47, 51, 52.

ROS can stimulate differentiation and bone resorption by osteoclasts 51, 53. During normal

physiological processes, O2_ generated on the outer membrane of osteoclasts is localized to the
54
ruffled border . Here, it assists with bone resorption by oxidizing calcium binding sites or

acidifying the environment near the osteoclasts, while simultaneously inhibiting the function of
54, 55
the antioxidant enzymes SOD and GPx . The osteoclast enzyme, TRAP, which is used to

digest the bone matrix, generates ROS, which in turn can destroy the organic matrix of bone 54.

Osteoclasts also contain the enzyme nicotinamide adenine dinucleotide phosphate (NADPH)
51
reduced oxidase which is involved in the generation of ROS . However, during states of

oxidative stress, this physiological process becomes pathological, resulting in increased

osteoclastic bone resorption. In rat bone, ROS have been shown to stimulate osteoclast

differentiation and bone resorption activity in the same manner as regulatory hormones IL-1 and

PTH 53. Specifically, O2_, which is converted in vivo to H2O2, has been shown to enhance bone

resorption through degradation of the bone matrix 53. It may also break down OCN into smaller

peptides 54. The lipid peroxidation marker, malondialdehyde (MDA) has been used as a measure

of osteoclast function 50
because osteoclasts generate O2_ which can result in the oxidation of

lipids, manifesting as increased serum concentrations of MDA (or other lipid peroxidation

markers) 41.
19
 High concentrations of ROS can damage osteoblast cells to prevent normal growth and

development 56 and have been shown to induce osteoblast death 52. In osteoblasts, H2O2 has been

shown to decrease cell growth, ALP activity, calcification, mineralization and gene expression of

osteogenic markers such as ALP, bone sailoprotein (BSP) and runt-related transcription factor 2
43, 56-58
(Runx2) . ROS can also activate Nf, resulting in the down-regulation of osteoblast

differentiation 58.

Epidemiological studies have shown that oxidative stress is an independent risk factor for
49
osteoporosis and that there is an inverse correlation between oxidative stress biomarkers and
47, 48
BMD . Concentrations of the isoprostane lipid peroxidation biomarker, 8-iso-PGF2, have

been inversely correlated with BMD in men and women, thereby demonstrating a connection
47
between increased oxidative stress and reduced BMD . In postmenopausal women with

osteoporosis, there is a significant, negative correlation between femoral BMD and the lipid
41
peroxidation marker MDA . Compared to healthy postmenopausal controls, MDA is

significantly elevated in cases with postmenopausal osteoporosis or bone fractures. A bone

fracture results in a high generation of ROS, thought to occur when the broken collagen strands
50
react with oxygen to yield oxygen metabolites . In fracture patients oxidative stress is

significantly elevated two to three weeks after fracture, as evidenced by increased concentrations
48
of MDA, an effect which is compounded by multiple fractures . This increase in oxidative
54
stress after fracture may facilitate bone healing , particularly during the stage of callus

formation, when fibroblasts and inflammatory cells increase the production of ROS, however it
48
can also cause oxidative injury similar to reperfusion injury . These epidemiological studies,

together with research carried out in vitro, demonstrate a significant correlation between

oxidative stress and postmenopausal osteoporosis.

20
IIC: The ability of antioxidants to decrease oxidative stress

and subsequently the risk of osteoporosis

In vitro studies suggest an important role for antioxidants in abrogating the effects of oxidative

stress on bone. Osteoblasts have been shown to produce antioxidants such as GPx which can
59
protect against the damaging effects of ROS . In MC3T3-E1 osteoblast-like cells, treatment

with H2O2 to induce oxidative stress was associated with the prolonged up-regulation in gene

expression of the transcription factor nuclear factor E2 p45-related factor 2 (Nrf2), which

regulates antioxidant enzymes by assisting with recognition of the antioxidant-response element

57
. Using xanthine/xanthine oxidase to generate ROS, Fatokun et al. showed that damage induced

by ROS, as evidenced by decreased cell viability, was prevented by CAT in MC3T3-E1

59
osteoblast-like cells. This effect is attributed to the ability of CAT to neutralize H2O2 . The

addition of metallothionein, a metal-chelating preventative antioxidant, to primary mouse bone

marrow stromal cells impaired H2O2-stimulated Nf signalling, consequently preventing any

inhibition of osteoblast differentiation 58. Trolox, a water soluble vitamin E analogue, was shown

to enhance ALP activity in MC3T3-E1 osteoblast-like cells, thus enhancing osteoblast

differentiation by decreasing the generation of ROS 56.

The activation of Nf is essential for the formation of osteoclasts and research shows

that this can be enhanced by ROS and suppressed by antioxidants. GPx has been shown to be the

predominant antioxidant enzyme in osteoclasts (RAW 264.7) and its expression is increased by

RANKL and 17 -estradiol. Estrogen deficiency results in decreased concentration of GPx in

osteoclasts, but the loss of bone typically associated with estrogen deficiency can be prevented

by CAT 51.

21
 Several case-control studies have compared the concentrations of antioxidant enzymes in

healthy postmenopausal women to those seen in women with osteopenia or postmenopausal

60
osteoporosis. A clinical study by Hahn et al found that in postmenopausal women with

osteopenia, GPx activity was significantly higher than that of postmenopausal women with a

normal BMD. GPx is most likely increased as a primary defense against the high levels of ROS,

particularly H2O2, being generated in osteopenic women. The activation of the signaling cascade

that increases internal antioxidant defenses such as GPx activity is a mechanism by which the

differentiation of osteoclasts and estrogen-deficiency bone loss could be down-regulated.

However, with the progression to osteoporosis, the primary defenses become overwhelmed and

cannot entirely combat the oxidative stress to prevent bone loss and may lead to the eventual
60
weakening or inhibition of antioxidant enzymes such as GPx . This is evidenced by findings

which consistently report that osteoporotic women have significantly depressed activities of
49, 50, 61
CAT, GPx and SOD, compared to the concentrations found in healthy control women .
49
Sanchez-Rodriguez et al. determined that the ratio of SOD/GPx was significantly higher in

osteoporotic women than that seen in healthy controls and was negatively correlated with BMD.

They postulate that this imbalance between GPx and SOD is due to increased concentrations of

H2O2, resulting in elevated oxidative stress in participants with osteoporosis. Furthermore,

49 61
concentrations of the antioxidant enzymes GPx , SOD and CAT have been positively

correlated with BMD, as well as total antioxidant capacity 49, which demonstrates a link between

antioxidant status and BMD in postmenopausal women. Prasad et al. demonstrated that in

fracture patients total antioxidant capacity is significantly elevated one to two weeks following
48
fracture compared to baseline values and healthy controls . These findings suggest that

antioxidant capacity is elevated post-fracture as a compensatory mechanism to attempt to repair

the damage caused by oxidative stress. This defense mechanism may be overwhelmed while

22
attempting to quench the increased O2_ production by osteoclasts during fracture healing, as is
50
evidenced later by significantly depressed SOD and GPx .

Recently, studies have emerged demonstrating that medications used to treat

45, 62-64
postmenopausal osteoporosis may act in part to decrease oxidative stress . It may be that

these medications act to some extent as antioxidants to decrease parameters of oxidative stress,
45, 62
particularly, HRT and Raloxifene . The depressed antioxidant enzymes and increased

concentration of lipid peroxidation typically seen in women with postmenopausal osteoporosis 49,
50, 61, 63
were mitigated by a 3 month course of Raloxifene therapy, after which CAT activity was
63
significantly increased and lipid peroxidation was significantly decreased . Similar findings

were reported by Misra et al. who found that in postmenopausal women given HRT (estradiol

valerate) for 6 months, lipid peroxidation was decreased by 16.3%, and GSH was increased by
64
5.9% . Osteoporosis medications may also decrease not only lipid, but protein oxidation as

well. Treatment with HRT for 6 months significantly decreased protein oxidation in

postmenopausal women, while simultaneously increasing GSH 45. Treatment with Raloxifene for

12 months also significantly decreased protein oxidation in osteoporotic participants compared to

concentrations found in matched, non-osteoporotic controls 62.

These studies on osteoporosis medications provide further proof that oxidative stress is a

risk factor for osteoporosis. The fact that they can act as antioxidants to decrease oxidative stress,

in addition to their known effects on BMD and fracture risk 2, suggests that intervention with

other, dietary antioxidants may enhance the effects of osteoporosis medications, or may stand

alone with similar beneficial effects. For example, in women between the ages of 50-79 from the

Women’s Health Initiative a significant interaction was found between HRT and a high intake of

vitamin C on the BMD of the femoral neck, spine and total hip 31.

23
 Many studies have provided evidence that dietary antioxidants on their own are capable

of counteracting the detrimental effects of oxidative stress and have been demonstrated to be
65-67
important in decreasing the risk of osteoporosis . Research suggests that a low intake of

vitamin C, a potent antioxidant capable of decreasing oxidative stress, is associated with a

decreased BMD and increased loss of bone in postmenopausal women 68-71. Specifically, in men

and women participating in a 17-year long prospective study, a daily intake of 308 mg/day was

associated with a decreased risk of hip and non-vertebral fractures 71. In smokers, a low intake of
67
vitamins C or E was associated with a higher risk of hip fracture and in non-smokers a low

intake of vitamins C or E was associated with an elevation of the bone resorption marker, CTx 72.

In a case control study, Maggio et al. found that postmenopausal women with osteoporosis had

significantly lower serum concentrations of vitamins A, C and E, as well as significantly lower

GPx and SOD than controls 65.

Studies have also examined the effects of carotenoids on bone, particularly those which

are vitamin A (or retinol) precursors, including -cryptoxanthin and - and -carotene.

Postmenopausal women with osteoporosis have been shown to have markedly reduced serum

concentrations of retinol, -cryptoxanthin, zeaxanthin and - and -carotene, compared to

healthy postmenopausal women 66. Overall carotenoid intake has been inversely associated with
73
fracture risk in a study by Sahni et al. . This study showed that a high carotenoid intake of

23.71 mg/day (including -cryptoxanthin, lutein, lycopene, zeaxanthin and - and -carotene)

was associated with a significantly lower risk of hip fracture compared to participants with a

lower intake of 7.30 mg/day. In vitro, -cryptoxanthin has been shown to stimulate bone

formation, specifically by increasing gene expression of Runx2, -1 (I) collagen and ALP,

resulting in increased cell number, calcium content, ALP activity and mineralization in bone

tissues 74, 75.

24
 The data on -carotene and risk of osteoporosis are still controversial. Some studies

suggest that there may be a beneficial effect of -carotene on bone 76-80, such as that reported by

Wattanapenpaiboon et al. who found -carotene intake to be positively associated with bone

mass in postmenopausal women 80. Other studies suggest a null or even detrimental effect, most

probably due to its association with vitamin A 31, 81-83. For example, in the Nurse’s Health Study,

a vitamin A intake of  3000 µg RE per day was shown to significantly increase the risk of
81
fracture in postmenopausal women . This detrimental effect was attributed to an increase in

retinol associated with vitamin A intake. In this study, the risk of fracture was not significantly

affected by -carotene, most likely due to the fact that a high intake of -carotene does not

increase bioavailability, and conversion to retinol is only in the range of approximately 8-16% of

the dietary -carotene consumed 82.

These studies provide evidence of a link between antioxidants and bone health.

Specifically, they show that intake of antioxidants can act to decrease oxidative stress to reduce

the risk of osteoporosis in postmenopausal women.

25
SECTION III: Lycopene

IIIA: Structure and sources of lycopene

Lycopene is a 40-carbon, highly unsaturated, open straight chain hydrocarbon, which is
84
comprised of 11 conjugated and 2 unconjugated double bonds . Although it does not possess

the -ionic ring structure and provitamin A activity, it is similar to other carotenoids with its

isoprenoid polyene structure with bilateral symmetry around the central double bond 44. It is the

longest carotenoid and rotation of the conjugated bonds results in the isomerization of lycopene
85
to form cis-geometrical isomers . Seven of the eleven conjugated bonds are stereochemically
86
active resulting in a total of 72 different isomers of lycopene . The structure of the most

common lycopene isomers can be found in Figure 1.2. The isomerization of all-trans to cis can

be induced in part by light, thermal energy and chemical reactions. However, overall it is a

highly stable molecule, with the individual isomers varying in their stability in the following

order of magnitude: 5-cis>all-trans>9-cis>13-cis>15-cis>7-cis>11-cis 44, 85.

Lycopene is a natural pigment produced by plants and micro-organisms but not animals.

The plant sources of lycopene are comprised primarily of the all-trans isomer of lycopene, on
86
average containing 79-95% of the all-trans isomer of lycopene . The main sources of dietary

lycopene are tomatoes and tomato-based products, which account for over 80% of the lycopene
85
in the daily diet . Other foods rich in lycopene include watermelon, papaya, guava, pink
87-89
grapefruit and rosehip The quantity of lycopene found in common lycopene-rich foods is

shown in Table 1.3. Limited data exists on the average daily intake of lycopene in Canadians 90,
91
. However, we have shown that intake is highly variable for women aged 25-70, with the

average intake being 6.14 ± 5.35 mg/day, which is higher than that reported for other countries.

26
The majority of dietary lycopene was obtained primarily through raw tomatoes, but in those with

higher intake, significantly more cooked and/or processed tomato products were consumed 91.

Lycopene comprises 75-90% of the total carotenoids found in tomatoes and tomato

products. In the tomato fruit, the majority of lycopene is located within the insoluble and fibrous

parts, particularly the skin, which contains approximately 5 times more lycopene than the pulp 86.

There are also modest amounts of other carotenoids such as -carotene (1-2%), -carotene (1-

1.3%), neurosporene (7-9%), lutein (0.01-1.1%), phytoene (5.6-10%), phytofluene (2.5-3%) and

- (0.03%) and -carotene (3-7%) 92. Tomatoes also contain moderate amounts of flavonoids in

their skin, such as kaempferol, naringenin and quercitin, and hydrocinnamic acids such as

caffeic, chlorogenic, p-coumaric and ferulic acids. Tomatoes are also a good source of fibre, both

soluble and insoluble, including celluloses and pectins. -carotene in tomatoes contributes to the

overall vitamin A content. Other vitamins and minerals found in tomatoes include biotin, folic

acid, niacin, pantothenic acid, potassium, thiamine, riboflavin and vitamins B6, C and E.

Processing decreases the concentrations of these vitamins, particularly vitamin E, which is found

solely in the seed of the tomato fruit. Carotenoid content of tomatoes is also affected by the

geographical site of production, climate, harvest and post-harvest handling, seasonal variation,
86
processing and storage . It is possible that lycopene acts in combination with these other

carotenoids, vitamins and minerals in tomatoes and tomato products to exert beneficial effects 93.

27
 all-trans

Common lycopene isomers 5-cis

9-cis

13-cis

15-cis

Figure 1.2: Major isomers of the antioxidant lycopene

28
Table 1.3: Lycopene content of foods rich in lycopene, shown in g/g and per average serving
size, as published by the United States Department of Agriculture (USDA) 89 and other research
87, 88

Food Average amount of lycopene

1
By weight (g/g) Per average serving size (mg/serving)
Grapefruit, pink/red, raw 14.19 3.26
Ketchup 167.07 2.51b
Pasta sauce 126.55 31.64
Papaya, raw 36.50 5.11
Pink guava, raw 54.00 8.91
Pizza sauce 127.10 7.75a
Rosehip 240.50 4.81b
Salsa 105.13 6.52a
Tomato, canned (puree) 217.54 54.39
Tomato, canned, stewed 40.88 10.42
Tomato juice 90.37 21.96
Tomato, paste 287.64 4.71b
Tomato, red, ripe, raw 25.73 4.63
Tomato soup 53.45 13.04
Vegetable juice 96.60 23.38
Watermelon, raw 45.32 6.89
1
Serving sizes were calculated based on weight per 1 cup measure, unless indicated by a for ¼ cup or b for 1
tablespoon.

IIIB: Safety and toxicity

The consumption of lycopene, either through food or supplements, has greatly increased over the

past few years, not only as a result of the great number of research studies demonstrating its

beneficial effects, but also due to an increase in the number of petitions to the FDA for the

approval of Qualified Health Claim language regarding the use of lycopene 94.
95
To date , the longest reported duration of lycopene supplementation was 20 weeks at

13.3 mg/day 96, and the highest reported dosage used was 150 mg/day for 7 days 97. Considering

the low usual daily intake of lycopene 91, these and other epidemiological studies 96-99, in which

the given dose of lycopene is much higher than the usual daily intake, provide convincing

evidence of no adverse effect of lycopene consumption in the short term. There have been two

29
extensive reviews encompassing both animal and human data which demonstrated that there are

no significant adverse effects of consumption of lycopene as either a natural or synthetic source

100, 101
. Overall, research suggests that the Observed Safe Level (OSL) of lycopene intake is 75

mg/day 86 for long-term supplementation.

IIIC: Bioavailability and metabolism

Of the carotenoids, lycopene is found in the highest concentration in human serum, and of the
44
tissues studied, it concentrates highest in adipose , but also in the prostate, liver and adrenal

gland 84. Average concentrations reported in human serum range from as low as 80 nM 102 to as
103
high as 3980 nM , with a wide standard deviation. This variation may be due to analytical

differences, such as the number of lycopene isomers measured or sensitivity of high-performance

liquid chromatography (HPLC) methods, or may simply be due to differences in dietary intake of

the study population. Approximately 10-30% of lycopene from the diet is absorbed into the

serum and tissues, with a reported half-life varying from as little as 2-3 days 44, 104, to as high as
105, 106
12-33 days . Studies suggest that it is the cis isomers of lycopene which are preferentially
85
accumulated in the tissues and serum . Serum consists of approximately 50% cis lycopene,

while tissues contain 75-90% cis lycopene 84-86, 92. In serum and tissues, lycopene may exist in as

many as 20-30 different isomers, however, the all-trans and 5-cis isomers contribute the most to

overall lycopene content 85, 86.

Lycopene consumed in the diet is absorbed with lipids and lipid-soluble vitamins 86. It is

separated from the food matrix and its protein complexes are broken down before being
86, 95
solubilized into lipid droplets by bile salts and fats . In the duodenum, lycopene is

incorporated into the inner portion of bile acid micelles, then absorbed through the brush border

30
membrane of enterocytes through passive diffusion, where they are packaged into chylomicrons

and secreted via the mesenteric lymph system into the blood and to the liver 44, 85, 86. The enzyme

lipoprotein lipase allows passive uptake of a small amount of lycopene by tissues prior to the
85
clearance of chylomicron remnants by liver . In the liver, lycopene can accumulate, or after

partial storage be repackaged into low-density lipoprotein (LDL) or very-low density lipoprotein

(VLDL) (VLDL is metabolized to form LDL) for recirculation in the blood and further uptake
85, 86
into tissues via the LDL receptor . Tissues containing high LDL receptor activity, such as

the prostate, adrenals and liver, tend to contain higher concentrations of lycopene 85.

Lycopene may be metabolized by the enzyme carotenoid monooxygenase II (CMOII),

resulting in asymmetric cleavage into three or more forms of apo-lycopenals including: apo-8’-,

apo-10’-, or apo-12’-lycopenal. The CMOII enzyme is found in moderate concentrations in the

adrenals, pancreas and prostate, and high concentrations in the eye, heart, liver, skeletal muscle

and small intestine. CMOII metabolizes both cis and all-trans lycopene, with a preference for cis
95
. Lycopene is oxidized within the system to 5,6-epoxide which is further metabolized to 5,6-

dihydroxy-5,6-dihydrolycopene (cis or all-trans) or 2,6-cyclolycopene-1,5- diol 107. The oxidized

form, 2,6-cyclolycopene-1,5-diol, is detectable in both human serum and breast milk 108 and 5,6-

dihydroxy-5,6-dihydrolycopene is the predominant lycopene metabolite detectable in human

107
plasma . Apo-lycopenals are also generated from the oxidative cleavage of lycopene epoxide

as well as by free radical oxidation, lipoxygenase activity, cytochrome P450 enzymes or other

carotenoid cleavage enzymes. These metabolites themselves may be physiologically active and

have been shown in vitro to provide some anti-carcinogenic properties 95, however these effects

have not been clearly delineated at this time.

Research suggests that processed tomato products are the most bioavailable, because

processing reduces particle size and divides the plant cell wall and fibrous pieces, releasing

31
lycopene from its matrix into the lipid component of the food or meal 84, 109. This is verified by a

study which has shown that the consumption of tomato puree for two weeks results in higher
110
serum lycopene than that resulting from consumption of cooked whole tomatoes . It has also

been suggested that the heating involved in processing increases the concentration of cis

lycopene, although these findings are still conflicting and indicate that only a small amount of
85, 95
isomerization occurs as a result of heating . Overall, processing increases the concentration

of cis lycopene by less than 10%, suggesting that the isomerization is due in large part to

physiological processes occurring during digestion and metabolism 85.

Cis isomers are thought to be better absorbed than the all-trans form of lycopene because

they are more soluble in the lipid phase, which means they can be more efficiently incorporated

into micelles and passed across the intestinal barrier 111. Although the exact mechanisms for the

selective uptake of cis isomers by micelles have not been completely elucidated, it is possible

that the presence of one or more cis double bonds reduces molecule length, allowing easier

uptake 85. In addition, cis isomers have higher polarity and do not aggregate and form crystals to

the same degree as the all-trans form of lycopene, this reduces the size of particles and leads to
85, 86
enhanced cis absorption . A recent bioavailability study by Failla et al. has shown that

differences in digestive processing of ingested cis and all-trans isomers of lycopene result in the

increased ratio of cis:trans isomers noted in tissues compared with foodstuffs, an effect attributed

to preferential incorporation of cis into micelles and more efficient uptake of cis by intestinal
112
epithelial cells . It is also possible that all-trans to cis isomerization occurs while lycopene

passes through the intestinal mucosa, during its incorporation into chylomicrons 86, or once it is

within the liver or tissues 98.

The bioavailability of lycopene can be also be improved by cooking or consumption with

84, 95
dietary fat . Lycopene is a lipid soluble molecule, and consumption of dietary fat, such as

32
triacylglycerols and cholesterols, stimulates the production of bile, allowing for more efficient

incorporation into micelles, thus enhancing absorption. Consumption of 5-10 grams of fat in a

meal is required for the optimal absorption of lycopene, a typical healthy diet in which 40% of

calories are obtained by fat, should provide enough fat for the optimal absorption of lycopene 85,
95
.
36, 84,
Lycopene bioavailability may also be affected by the presence of other carotenoids
113
. Lycopene may compete with other carotenoids of similar structure for incorporation into

micelles, intestinal uptake, incorporation into chylomicrons, or LDL binding or transport.

Research suggests that despite this potential for competition between lycopene and other

carotenoids, such as -carotene and lutein, it does not actually interfere with their overall

bioavailability and may in fact have a synergistic effect 113.

On the other hand, plant sterols, cholesterol-lowering drugs, fibre and fat substitutes, such

as sucrose polyester, can interfere with the incorporation of lycopene into micelles, reducing
85, 95
absorption . One mechanism by which these materials are thought to interfere is that they

make a hydrophobic sink in the lumen of the small intestine which binds to lycopene, preventing

uptake by enterocytes 85.

An inverse association between smoking and serum lycopene has also been found.

However, rather than smoking decreasing lycopene absorption itself, it could be that absorbed

serum lycopene is utilized faster in people who smoke as an attempt to counteract the associated
95
oxidative damage . This could explain the lower concentrations detected in the serum of

smokers.

33
IIID: Mechanisms of lycopene action

There are many proposed and studied mechanisms of lycopene action, both oxidative and non-

oxidative. The non-oxidative mechanisms involve non-covalent binding interactions between

lycopene and proteins and cell membranes, including growth factor signalling, modulation of
95
hormone and immune responses and enhancing gap junctional intercellular communication .

These mechanisms may affect mutagenesis, carcinogenesis and/or cell differentiation and
114
proliferation . Specific examples include: (1) carcinogen metabolism, (2) inhibition of

hydroxymethylglutaryl coenzyme A (HMG-CoA) reductase and up-regulation of LDL receptor

activity in macrophages to decrease cholesterol, (3) repression of insulin-like growth factors to

inhibit cellular proliferation, (4) gene function regulation, (5) modulation of the liver

metabolizing enzyme cytochrome P450 2E1, (6) modulation of expression of cell cycle

regulatory proteins to stop cell division at G0-G1 phase, (7) suppression of carcinogen induced

phosphorylation of regulatory proteins (p53 and RB anti-carcinogens), (8) inhibition of IL-6,

androgens and 5-lipoxygenase, and (9) upregulation of connexion 43 resulting in increased gap-

junction communication 44, 84, 86, 95, 114.

Of particular interest to this thesis, however, are the well documented antioxidative

mechanisms of lycopene which allow it to protect cellular biomolecules such as lipids, proteins,
36, 97, 115
enzymes, lipoproteins and DNA from oxidative modification , thus preventing

development of chronic diseases resulting from oxidative stress. Lycopene is a potent

antioxidant. It has a total antioxidant capacity of 2.9  0.15 mM Trolox equivalents, which is

31-55% higher than the other carotenoids, such as -cryptoxanthin, - carotene, or -carotene 36.

Of the lycopene isomers, the cis forms have the highest antioxidant capacity, in the following

order of magnitude: 5-cis>9-cis>7-cis>13-cis>11-cis>all-trans 44, 116.

34
 Lycopene, although an acyclic isomer of -carotene, is unique compared to the other

carotenoids as it does not possess the -ionone ring structure, thus it is not metabolized to

vitamin A in the system 84. It is the polyene structure of carotenoids that gives them the capacity
36
to quench and inactivate singlet oxygen and free radicals . In terms of singlet oxygen

quenching ability, it is at least twice as high as -carotene and ten times as high as -tocopherol
117
. It has also been shown to quench a wide range of other free radicals such as reactive nitrogen
36, 95
species, H2O2, nitrogen dioxide, peroxyl radicals, thyl and sulphonyl . Lycopene interacts

with free radicals and is itself oxidized in the process. This may occur by electron capture,

cleavage of the polyene chain as peroxyl radicals are added, or hydrogen abstraction from allylic

positions. Due to its lipophilic nature, lycopene is embedded within the non-polar inner-

environment of the cell membrane or associated with lipoproteins, allowing for a greater

scavenging of free radicals in hydrophobic environments. This location within the lipophilic

component allows lycopene to provide lipids and lipoproteins with powerful resistance against

oxidative damage 36.

IIIE: Lycopene and its effect on chronic disease

Lycopene has been extensively studied because of its potent ability to decrease oxidative stress,

and has been shown to significantly enhance total antioxidant capacity, while simultaneously

decreasing oxidation of lipids, proteins and DNA. Studies have associated lycopene with a

decreased risk of chronic diseases including cardiovascular disease, cancer, hypertension, male

infertility, neurodegenerative disorders and diabetes (for comprehensive review, see 44). Perhaps

two of the most extensively studied lycopene relationships to date are those of lycopene and

cancer and lycopene and cardiovascular disease.

35
 Several studies support the role of lycopene in the prevention of cardiovascular disease.

Many studies provide evidence that this effect of lycopene on cardiovascular disease may be

attributed to its ability to prevent oxidative modification of LDL 99, 118, 119. It has also been shown

to significantly decrease total serum cholesterol, triglycerides and LDL while simultaneously

increasing high-density lipoprotein (HDL) 119-121. These findings are strengthened when lycopene

is consumed with olive oil or as a capsule in an oil matrix. A significant association between

high lycopene and higher serum HDL, lower ratio of total cholesterol to HDL, lower serum LDL
64, 122
and lower triacylglycerols in healthy participants has been reported . Lycopene may also

affect the risk of myocardial infarction in a dose-dependent manner, studies suggest that high

concentrations of lycopene in adipose tissue decrease the risk of myocardial infarction by

decreasing intimal wall thickness 123, 124.

High consumption of tomatoes has been associated with a lower risk of many types of
95, 114, 125
cancer . In vitro, lycopene was shown to be an effective inhibitor of growth of human
126
endometrial, lung and mammary cancer cells lines by inhibiting proliferation . High intake of

tomatoes or tomato products was associated with a 40% decrease in the risk of lung cancer and a
95
60% reduction in the risk of developing colorectal cancer . Reviews suggest that lycopene is

most effective at the anatomical sites of prostate, pancreas, esophagus, lung, stomach and colon
95, 114, 125
. There is also some evidence to suggest that lycopene may also be capable of decreasing
95, 127
the risk of oral/laryngeal/pharyngeal, cervical, breast and ovarian cancer , however results
95
in these areas are still conflicting and need to be substantiated further . In light of the vast

amount of scientific research demonstrating a potential link between lycopene and decreased risk

of cancer, the FDA performed an evidence-based review and concluded that there was some

limited evidence to support an association between tomato consumption and reduced risk of

36
gastric, ovarian, pancreatic and prostate cancers, although at this time, the existing evidence is

not powerful enough for the FDA to make official qualified health claims 94.

IIIF: Lycopene and its effect on bone

Despite the well known antioxidant properties of lycopene and its ability to decrease the risk of
44
chronic diseases associated with oxidative stress such as cancer and cardiovascular disease ,

studies examining the effects of lycopene on bone and the risk of osteoporosis are limited and are

only starting to emerge in the literature. This could be due to the fact that the effects of vitamin A

and its precursors such as -carotene have been well documented both in vitro and in vivo and

the contradictory findings suggest 76-80 that the possibility of a detrimental effect of vitamin A on
81, 83
bone cannot be ruled out . However, lycopene is an acyclic isomer of -carotene and is not
84
converted to vitamin A during metabolism ; consequently there is no relevant basis for

comparison of lycopene to these previous findings on vitamin A and its precursors.

There are a limited number of in vitro and animal studies examining the effects of

lycopene on osteoblasts and osteoclasts. In vitro, lycopene has been shown to inhibit the

formation of osteoclasts and their production of ROS, resulting in an overall inhibition of

128, 129
resorption in the presence or absence of PTH . In both human and mouse osteoblasts,
130 77
lycopene significantly increased cell number, ALP activity , and gene expression of BSP ,

suggesting that lycopene stimulates the growth and differentiation of osteoblasts. In a murine

model of osteoporosis, animals who were fed lycopene had significantly increased trabecular

number and decreased trabecular separations, resulting in a significantly inhibited loss of bone in
131
the proximal femur and tibia, compared to control-fed animals . Although the exact

37
mechanism by which lycopene exerts these effects on bone has not yet been delineated, it has

been proposed that these changes occur by means of its potent antioxidant activity 128-130.

Recently, lycopene has also been shown to be associated with the risk reduction of

osteoporosis in postmenopausal women. Case-control studies have demonstrated that in women

with osteoporosis, serum lycopene concentrations are lower than in healthy women of the same

age. A study by Maggio et al. showed that in a sample of 45 osteoporotic women, serum

lycopene was 22% lower than serum lycopene in 45 healthy postmenopausal controls 66. Similar

findings were reported by Yang et al., who found that in 27 osteoporotic women, serum lycopene

concentrations were 38% lower than those found in matched healthy controls, despite having a

similar daily intake 103.

We have shown that in postmenopausal women who consumed an average of 7.4 mg of

lycopene per day, the resultant serum lycopene is associated with significantly lower biomarkers

of protein oxidation and bone resorption compared to women who consumed less daily lycopene

and had lower concentrations of serum lycopene 132. These findings suggest that the mechanism

by which lycopene exerts its effects on bone is an antioxidative one. Misra et al. have shown that

lycopene has the same antioxidative effects as HRT in postmenopausal women. Lipid

peroxidation significantly decreased while GSH significantly increased, and although the effects

of lycopene on bone turnover or BMD were not reported by these authors, the findings provide

further proof to the mechanistic effect of lycopene as an antioxidant and promote further

investigation 64. A positive association between vitamin C and HRT on BMD has been reported

(as mentioned in Section IIC above) 31, so it is entirely possible that lycopene may also interact

with HRT or other osteoporosis medications to exert a beneficial effect on bone in

postmenopausal women.

38
 78-80
Studies have also examined the relationship between lycopene and BMD . Dietary

lycopene intake has been shown to account for 10% of the variance in bone mineral content, an

effect not seen with other dietary carotenoids. In premenopausal women, lycopene intake has
80
been shown to be positively associated with total BMD and in postmenopausal women

lycopene intake was significantly associated with a protective effect on BMD at the lumbar spine
78
. On the other hand, radial BMD was not correlated with serum lycopene in postmenopausal

Japanese participants, while there was a weak correlation between radial BMD and -
79
cryptoxanthin and -carotene . The discrepancy between study findings could be due to

differences in the race of participants, differences in BMD measurement sites and differences in

correlation parameters (i.e. BMD correlated with lycopene intake 78, 80 and BMD correlated with

serum lycopene concentrations 79). Therefore, further in-depth study into the effect of lycopene

on BMD is necessary to solidify the correlation between intake of lycopene and the

corresponding serum concentrations to BMD.

The positive association between lycopene and BMD may contribute to a decrease in the

risk of fragility fracture related to osteoporosis. Recently, a longitudinal dietary study by Sahni et

al. showed that participants with a high intake of 12.7 mg per day of lycopene had significantly

lower incidence of non-vertebral osteoporotic fracture compared to participants who consumed a

73
lower intake of 2.7 mg/day . Participants consuming  4.4 servings of lycopene-containing

foods per week, including tomatoes, tomato juice/sauce, grapefruit, pizza and watermelon, had

significantly fewer hip fractures than participants who consumed <4.4 servings per week.

Furthermore, these associations were not found for -cryptoxanthin, lutein/zeaxanthin, or - or

-carotene 73, suggesting that lycopene may provide the most benefit for decreased osteoporotic

fracture risk.

39
SECTION IV: Polymorphisms and nutraceutical

interventions

IVA: The relevance of polymorphism studies in nutritional

research

Functional foods provide benefits to human health over and above that found in traditional food

sources, thus the study of nutritional factors important to human health and disease risk is an

ever-expanding field. However results from clinical studies on functional foods and dietary
133
nutrients are often conflicting due to varying responses in human studies . Clinical trials on

dietary nutrients are similar to those of pharmaceutical products in that they establish a portion of

the population which responds as expected, with a number of other participants who experience
134
either adverse or null effects . These varying responses may be due to differences in genetic

variability which affect the response to specific dietary nutrients. The field of Nutrigenetics

addresses how genetic variation can affect the interaction between dietary nutrients and health

and disease prevention and the mechanisms by which this interaction occurs 135.

Single nucleotide polymorphisms (SNPs) are DNA sequence variations that occur when a

single nucleotide in the genome sequence is altered similarly in at least 1% of the population.

These sequence variations occur at approximately every 100-300 bases along the human genome

and account for 90% of all human genetic variation. SNPs can occur in either the coding or non-

coding region of the genome and the majority involve the replacement of cytosine (C) with

thymine (T) 134. It is estimated that the human genome contains approximately 10 million SNPs,

with just over 3 million classified to date 133, 136. Because SNPs can affect the response to dietary

nutrients by moderating their absorption, metabolism, bioavailability and circulation 137, or their

40
 134
effectiveness in disease prevention , their inclusion in clinical and epidemiological study

protocols contributes greatly to further understanding the role of nutrition in human health and

disease 137.

IVB: Polymorphism studies related to oxidative stress and

the effect on osteoporosis risk

Approximately 25-35% of the variance in susceptibility to fracture is attributed to genetic factors

138
. Genetic studies on osteoporosis have mainly examined BMD as a phenotype due to its

association with fracture risk and clinical diagnosis of osteoporosis. However, approximately

50% of women who suffer fractures have a BMD that is not within the osteoporotic range as
19
outlined by the WHO , providing further emphasis to the fact that osteoporosis is a

multifactorial disease with a number of contributing risk factors. As outlined in Section IC

above, examination of BTMs can contribute important information regarding osteoporosis and

fracture risk. However, a limited number of genetic studies on osteoporosis have used BTMs as

phenotypes. Candidate genes for osteoporosis study which are known to be important to the

biochemical and physiological pathways of bone include the vitamin D receptor (VDR), estrogen

receptor ( or ), collagen 1  1 (COLIA1), transforming growth factor 1, androgen receptor,

IL-6, apolipoprotein E, PTH, calcitonin receptor, OCN and calcium sensing receptor 138.

To date, approximately 40 gene variants have been found to be associated with regulation

of BMD or fracture risk. Research suggests that BMD is determined by multiple genes, and each
138
associated polymorphism contributes a small to moderate effect on BMD overall . With

respect to pharmacological interventions for treatment of osteoporosis, evidence suggests that

certain genetic factors can mediate the pharmacological response. For example, in a study on the

41
bisphosphonate etidronate in 108 osteopenic women, those carriers of the SP1 SS genotype

(dbSNP rs 1800012, polymorphism of COLIA1) had an increased BMD in the femoral neck as a

result of therapy, while those with the SP1 s allele had decreased BMD as a result of therapy 139.

These findings are just one example of how genetic variation can affect the response to

intervention, in this case pharmaceutical, thus modifying the effect on bone.

The majority of studies on polymorphisms relate to the interaction between SNPs and

BMD, and the response to pharmacological intervention. Despite the well known association
47-50
between oxidative stress and osteoporosis , a very limited number of studies have examined

polymorphisms which may affect the oxidative stress pathway to moderate the risk of

osteoporosis and their interaction with dietary nutrients. The VDR receptor has been studied and

interactions between VDR and (a) calcium intake on BMD, (b) calcium absorption, (c) dietary

vitamin D and BMD have been found (for review, see Cusak and Cashman 140). A recent study

by Oh et al. has shown an association between polymorphisms of the CAT gene and BMD in

postmenopausal women 40. The +22348CT polymorphism on exon 9 (rs17880449) of the CAT

gene was associated with BMD at the lumbar spine and femoral neck in participants with the

dominant T allele. Participants with the T allele also had significantly higher OCN activity than

participants homozygous for the C allele 40. These findings suggest that the CAT gene may be an

important genetic determinant of BMD and bone metabolism. Together, these studies provide

further support to the idea that genetic variation can affect the response to dietary intervention

and antioxidant action on bone.

Nevertheless, studies exploring the relationship between polymorphisms, oxidative stress

and osteoporosis are limited and require additional research. Interactions between

polymorphisms and the serum concentrations of nutrients, together with their effects on

oxidative stress, are particularly important in the study of osteoporosis prevention and treatment

42
because although genotype cannot be modified to optimize prevention or treatment methods, diet

and nutrient intake can be 140. The vast array of metabolic pathways by which nutrients can affect

bone by modulating oxidative stress suggest that there are many more gene-nutrient interactions

which have yet to be discovered and researched. The incorporation of genotyping into the

science of nutritional studies on osteoporosis may lead to the development of dietary

recommendations which are specific to individual needs and may provide greater protection

against the risk of developing this disease. These polymorphism studies will be important in

delineating a comprehensive approach to prevention and treatment of osteoporosis which will be

optimally beneficial for individuals.

IVC: The paraoxonase enzyme: its antioxidative capacity and

the role of its polymorphisms in the development of

osteoporosis

Human paraoxonase (PON1) is an HDL-associated, calcium-dependent esterase enzyme 141. It is

142
synthesized by the liver but is present in high concentrations in serum . One of the main

physiological functions of the PON1 enzyme is that it contributes to the antioxidant properties of

HDL 143. As an HDL-associated enzyme, PON1 retards or prevents the oxidative modification of
144, 145
LDL by hydrolyzing lipid peroxides and cholesteryl-ester hydroperoxides on LDL . It has
146
also been shown to hydrolyze hydrogen peroxide , a potent ROS. These mechanisms prevent
147
the accumulation of damaging ROS, particularly the lipid hydroperoxides , and decrease

oxidative stress in the system.

The standard measurement of PON1 activity is to determine its ability to hydrolyze

148
paraoxon . Studies show that there is an inverse correlation between the ability of PON1 to

43
hydrolyze paraoxon and its ability to hydrolyze lipid peroxides 149. This paradoxical relationship

may be explained by the fact that the sulfhydryl group on cysteine-284 is necessary for PON1 to

protect LDL from oxidation but is not required in the hydrolysis of paraoxon 150.

Antioxidants can affect PON1 activity. PON1 acts as an antioxidant to inhibit LDL

oxidation, which can result in the degradation of PON1. In states of oxidative stress, where the

concentration of lipid peroxidation by-products is elevated, the PON1 enzyme is significantly

143
inactivated . Antioxidants, which can scavenge these free radicals before oxidization occurs,
151
may prevent decremented PON1 resulting from oxidative stress . It has been suggested that

antioxidants may preserve PON1 activity, resulting in a moderate effect on PON1 activity after

supplementation 141, 150, 152.

The PON1 gene has two SNPs in the coding region that result in amino acid substitutions

at positions 55 and 192. At position 55, leucine is replaced by methionine (172TA or

Leu55Met) and at position 192, glutamine is replaced by arginine (584AG or Gln192Arg).

These SNPs are largely responsible for modulating the activity of PON1, and the 584A allele and

172T allele have each been associated with higher PON1 activity 144. PON1 activity according to

genotype is in the rank order 584GG>584AG>584AA and 172TT>172TA>172AA. Analysis of

combined genotypes has demonstrated that individuals with the 172AA genotype have lower

PON1 activity regardless of the 584AG polymorphism 144, 147, it is also possible that there may

be a cumulative effect of these polymorphisms on PON1 activity 153.

The association between diet and serum PON1 activity has been examined 145, 151, 154, 155.

While some studies report that PON1 is positively correlated with intake of antioxidants such as

carotenoids 156 and vitamins C and E 151, other studies report a null 155, 157 or opposite 154 effect. It

is possible that these conflicting results may be attributed to PON1 polymorphisms. Dalgard et

al. found no effect of antioxidant rich orange and blackcurrant juice or vitamin E

44
supplementation on PON1 activity prior to genotyping which revealed an increase in PON1

activity after juice consumption in carriers of the 172T allele only 155. However further research

into this mechanism is required as findings remain inconsistent 145, 158, 159.

The effect of PON1 polymorphisms on lycopene intervention has been recently studied
160, 161
. Bub et al. reported that in older participants with increased oxidative stress who were

carriers of the 584G allele, consumption of tomato juice for a period of 8 weeks resulted in

significantly decreased LDL-oxidation and thiobarbituric acid reactive substances (TBARS) and
160
increased antioxidant capacity, while those with the 584AA genotype showed no effect .
161
Similar findings were reported in healthy participants with a shorter supplement period . The

uptake of lycopene occurs via LDL receptors and tissues containing high LDL receptor activity,

such as the prostate, adrenals and liver, tend to contain higher concentrations of lycopene 85. Due

to its location in the lipoprotein component, it has been correlated with the ability to decrease

LDL oxidation and overall lipid peroxidation 99, 118, 119. This, together with its potent antioxidant

capacity, suggests that lycopene may have a powerful sparing capacity of PON1.

Because of the known antioxidant actions of PON1, the 172TA and 584AG

polymorphisms have been extensively studied in relation to chronic diseases associated with

oxidative stress, such as atherosclerosis and diabetes 147-149, 153. Since oxidative stress is a known
47-50, 56
risk factor for osteoporosis , PON1 polymorphisms in relation to osteoporosis have

received recent attention. Lipid peroxidation by-products, such as lipid hydroperoxides, are
162
potent reactive oxygen species, mutagenic to cells in vivo . LDL oxidation products stimulate

bone loss by inducing bone marrow progenitors to an adipogenic, rather than osteogenic fate 163.

These by-products can inhibit differentiation of osteoblasts by changing mineral content,

163, 164
decreasing formation and inhibiting mineralization . Further, oxidized LDL have recently

been shown to increase the generation of ROS by mouse osteoclasts (RAW 264.7), and affect the

45
differentiation and activity of osteoclasts 165. Because PON1 prevents the oxidation of LDL, it is

possible that it could prevent the negative effects of oxidative stress on bone. And although there

is no confirmed association between bone mineral density (BMD) and the serum activity of
166, 167
PON1 activity at this time , the PON1 SNP may nonetheless be important in the

development of osteoporosis due to its antioxidant properties and ability to alter the

susceptibility of lipids to oxidation. To date only one study has assessed the effects of PON1
166
polymorphisms on the risk of developing osteoporosis. Yamada et al. showed the relation

between BMD and the 172TA and 584AG polymorphisms in postmenopausal women. For

172TA, participants with the TT genotype had significantly lower BMD in the lumbar spine

and femoral neck than participants with the TA or AA genotype. For 584AG, participants with

the GG genotype had a significantly lower BMD of the femoral neck than participants with the

GA or AA genotype. Another study has shown that the 172TT and 584G genotypes are the least
148
effective in protecting LDL against oxidation . These facts, together with the association

between lipid peroxidation and osteoporosis outlined above, may suggest an important

mechanism of the PON1 polymorphisms in decreasing the risk of osteoporosis.

46
 CHAPTER 2

Hypotheses, objectives and thesis overview

47
 Hypotheses
Oxidative stress is a known contributing risk factor for osteoporosis as it causes an increased loss

of bone. Lycopene is a potent antioxidant capable of decreasing oxidative stress and may

therefore be beneficial in reducing the risk of osteoporosis. The hypotheses explored in this

thesis are as follows:

I. Supplementation with lycopene will increase antioxidant capacity of the serum while

decreasing oxidative stress parameters, such as protein and lipid oxidation, subsequently

decreasing bone turnover markers, and thus reducing the risk of osteoporosis in

postmenopausal women.

II. Although there are several isomers of lycopene, those possessing the highest antioxidant

capacity, the cis isomers, will provide the most benefit to decrease oxidative stress

parameters and bone turnover markers.

III. Interactions between polymorphisms and serum lycopene may moderate the effect of

lycopene on oxidative stress parameters and bone turnover markers. Specific

polymorphisms in the antioxidative enzyme PON1 will interact with lycopene to modify

oxidative stress and the response to lycopene in bone.

48
 Objectives and specific aims
The following are the main objectives of this thesis work, conducted in order to explore the

above hypotheses:

I. Conduct a cross-sectional study in women aged 25-70 to explore whether lycopene

obtained through the daily diet is capable of decreasing oxidative stress parameters and

bone turnover markers. Specifically, to determine whether the response to lycopene is

moderated by polymorphisms on the PON1 enzyme.

II. Conduct a randomized, controlled, intervention study in postmenopausal women 50-60

years of age to (a) determine whether lycopene supplementation directly results in

increased antioxidant capacity, decreased oxidative stress parameters and subsequently

decreased bone turnover markers, (b) establish whether lycopene supplementation

provides more benefit on the aforementioned parameters than the lower quantity of

lycopene typically obtained in the daily diet, (c) ascertain whether the polymorphisms on

the PON1 enzyme interact with lycopene to affect the response to intervention in

postmenopausal women.

III. Delineate the mechanisms by which lycopene can prevent or repair the damaging effects

of ROS in human osteoblast cells and whether different lycopene isomers provide more

benefit in this regard than others.

49
 Overview of thesis Chapters 3-9 and their application to the

hypotheses

CHAPTER 3: In-depth analysis of lycopene intake and serum lycopene concentrations in

women aged 25-70

In this chapter, an in-depth analysis of lycopene intake and serum concentrations of lycopene in

women aged 25-70 was conducted. Correlations were made between intake and serum

concentrations of lycopene and other dietary nutrients, as well as participant characteristics such

as age and body mass index (BMI). The most common sources of lycopene-containing foods and

the frequency of their consumption were analyzed for different age groups. This chapter is

important to the objectives of this thesis because it sets a basis for comparison of typical

lycopene intake in women and establishes whether lycopene intake and resultant serum lycopene

concentrations are affected by other parameters.

CHAPTER 4: Dietary restriction of lycopene for a period of one month resulted in

significantly increased biomarkers of oxidative stress and bone resorption in

postmenopausal women

This chapter examined the effects of a diet without lycopene in postmenopausal women aged 50-

60. Due to the fact that lycopene is found in a limited number of foods it is important to

determine whether a conscious consumption of these foods is beneficial to bone health. Further,

many intervention studies on lycopene include a washout period during which no lycopene-

containing foods are consumed, this chapter provides evidence of the effects of lycopene

restriction either through food choices or as participants in clinical studies.

50
CHAPTER 5: Supplementation with the antioxidant lycopene significantly decreased

oxidative stress parameters and the bone resorption marker crosslinked N-telopeptide of

type I collagen (NTx) in postmenopausal women at risk of osteoporosis

In this intervention study, 60 postmenopausal women were supplemented with tomato lycopene

capsules, regular tomato juice or lycopene-rich tomato juice, providing 30-70 mg of lycopene a

day for a period of 4 months, to determine the direct effects of lycopene on oxidative stress

parameters, antioxidant capacity and bone turnover markers. Results from this chapter provide

the first clinical evidence demonstrating that intervention with lycopene can decrease oxidative

stress parameters and bone turnover markers and thus the risk of osteoporosis.

CHAPTER 6: Supplementation with lycopene in postmenopausal women resulted in serum

concentrations of 5-cis lycopene which were higher than usually obtained through the daily

diet, and were associated with lower biomarkers of lipid peroxidation and bone resorption

In this chapter, data from postmenopausal women who participated in the randomized, controlled

intervention study was compared to data from the cross-sectional study to determine whether

supplementation with lycopene (30-70 mg/day) would be more beneficial in decreasing oxidative

stress parameters and bone turnover markers than an intake of lycopene usually obtained through

the daily diet. A comparison between these groups of participants determined how serum

lycopene affects oxidative stress parameters and bone turnover markers. This chapter provides

evidence that it is the antioxidant capacity of lycopene which is beneficial in decreasing

oxidative stress parameters and bone resorption markers.

CHAPTER 7: Paraoxonase 1 polymorphisms 172TA and 584AG modified the

association between serum concentrations of the antioxidant lycopene and bone turnover

markers and oxidative stress parameters in women 25-70 years of age

51
This chapter investigated whether polymorphisms of the PON1 enzyme modified the association

between serum lycopene and oxidative stress parameters and bone turnover markers in women

aged 25-70. This cross-sectional study provides evidence of an in interaction between PON1

polymorphisms and serum lycopene absorbed from the usual daily diet in women, and elucidates

potential mechanisms of lycopene action.

CHAPTER 8: Supplementation with lycopene resulted in decreased lipid peroxidation

parameters, which interacted with the PON1 172TA genotype to decrease the bone

resorption marker NTx in postmenopausal women

This chapter further explored the interaction between PON1 polymorphisms and the response of

oxidative stress parameters and bone turnover markers to lycopene intervention in

postmenopausal women aged 50-60. The experimental work in this chapter delineates the

mechanisms by which lycopene supplementation may decrease the risk of osteoporosis.

CHAPTER 9: Cis lycopene isomers found in high concentrations in human serum, and

which possess the greatest antioxidant capacity, were capable of preventing and repairing

the damaging effects of reactive oxygen species in human osteoblast cells

This chapter investigated the mechanisms of action of lycopene at the cellular level. Human

osteoblast cells were treated with varying isomeric concentrations of lycopene to delineate the

mechanisms by which lycopene acts in these cells. Specifically, human osteoblasts were treated

with varying concentrations of lycopene isomers that are typically found in foods (high all-trans)

and human serum (high cis) to determine whether they were capable of preventing and repairing

the damaging effects of ROS on formation of mineralized bone nodules by osteoblasts. This

chapter provides further mechanistic evidence of lycopene action in bone cells.

52
 CHAPTER 3

In-depth analysis of lycopene intake and serum lycopene

concentrations in women aged 25-70

This chapter was originally published in the Journal of Medicinal Food 12 (4), 2009 as a paper
by Mackinnon, E.S., Rao A.V., and Rao L.G., entitled “Lycopene Intake by Canadian Women is
Variable, Similar Among Different Ages, but Greater than That Reported for Women in Other
Countries”. It has been updated to include additional recent findings and reformatted for this
thesis with the publisher’s permission.

53
 ABSTRACT

Lycopene is an antioxidant associated with a reduced risk of chronic diseases common in

women, such as osteoporosis and cancer. However, no official recommendation for lycopene

consumption exists, and intake data on Canadian women are limited. This study was designed to

generate more information about average lycopene intake in a sub-population of Canadian

women of different ages. A cross-sectional study was conducted at St. Michael’s Hospital in

Toronto, Ontario, Canada. One hundred and three women, between the ages of 25 and 70 years,

who were not on any medications, were recruited to record their diet for 7 days and provide a

serum sample. Statistical analyses were performed to compare the types of lycopene-containing

foods consumed, and ascertain associations between intake of lycopene and

macro/micronutrients, and whether participant characteristics, such as BMI, could predict

lycopene intake. These results showed that the average lycopene intake of participants in this

study was 6.03 ± 5.35 mg/day, which is higher than that reported in other countries, and

corresponds to an average serum lycopene concentration of 1346 ± 60.97 nM. Intake was similar

among age groups but was highly variable. In all participants, raw tomatoes were consumed

more frequently than all other sources of lycopene combined (aged 25-49, p<0.0005 and aged

50-70, p<0.0001), and those with the highest lycopene intake consumed more cooked/processed

tomato products than those with lower intake (p<0.005). High consumption of cooked/processed

tomato products was associated with high cis lycopene concentrations (p<0.0001). Participants

25–49 years old consumed more cooked/processed lycopene sources overall (p<0.02),

particularly dried/powdered tomatoes (p<0.05), pizza (p<0.002) and ketchup (not significant,

p<0.10), while 50–70 year olds consumed more tomato juice (p<0.05). Intake or serum lycopene

could not be predicted by any participant characteristics, such as BMI. In older participants,

lycopene intake was positively correlated with intake of niacin and vitamins A, D and K (p<0.01,

54
p<0.05, p<0.05, p<0.001, respectively). Serum lycopene was positively correlated with -

cryptoxanthin (younger participants, p<0.005), lutein/zeaxanthin (older, p<0.05), - (older,

p<0.0001) and -carotene (older, p<0.002). These findings are significant to women’s health and

may contribute to the establishment of nutritional and health recommendations regarding

consumption of lycopene by Canadian women to prevent chronic diseases.

55
 INTRODUCTION

The antioxidant lycopene is a member of the carotenoid family of compounds naturally present

in a limited number of fruits and vegetables 84. Among all carotenoids, lycopene has the highest

singlet oxygen quenching ability; it is two to three times higher than -carotene and 10 times
117
higher than -tocopherol . In the diet, the main sources of lycopene are tomatoes and tomato
88
products . However, it is also found in watermelons, papayas, guavas, pink grapefruits and
87
rosehips . In raw tomatoes, lycopene is present predominantly in its all-trans isomeric form.

However, in cooked and processed tomatoes higher concentrations of the cis isomeric form of
168
lycopene are present, accounting for a more efficient absorption of lycopene . Lycopene is a

highly lipid-soluble carotenoid, and its absorption is improved when ingested with a small

amount of fat 169.

Oxidative stress is now recognized as an important factor in the development of several

chronic diseases 26. Consumption of lycopene-containing foods results in significantly decreased

oxidation of lipids 118, proteins and DNA 99, 170 in men and women of all ages, and recent studies

have specifically associated lycopene intake with a decreased risk of age-related chronic diseases
44
. A growing body of evidence suggests that lycopene intake in women is of particular

importance because of its association with a decreased risk for breast, cervical and ovarian
125
cancers . We have recently shown high lycopene intake to be associated with lower bone

resorption markers in postmenopausal women, suggesting a beneficial effect of lycopene on

bone health by lowering the risk of osteoporosis 132.

Although levels of lycopene in the range of 6–60 mg per day have been reported to be

beneficial against chronic diseases, no specific recommendation for its daily intake has been

published. A very limited number of studies have reported the intake levels of lycopene in
171-176 90
women , with only one study addressing its intake in Canadian women . Therefore, the
56
present study was conducted with the objective of assessing lycopene intake and serum lycopene

concentrations in a cohort of the female population representative of different age groups. This

article reports findings from a systematic evaluation of lycopene intake according to age in a

female participant population in Canada.

METHODS

Experimental participants

This study protocol was approved by the research ethics board at St. Michael's Hospital, Toronto,

and followed the guidelines of good clinical practices. A total of 108 healthy women, 25 to 70

years old, were recruited mainly between the end of September to the beginning of May in the

years 2003–2007. The following recruitment methods were used: advertisements in local

papers/magazines and on the University of Toronto website, as well as posters placed in local

hospitals, fitness centers and health food stores. Additionally, patients who had recently attended

St. Michael’s Hospital for a BMD measurement, and had signed a form indicating their interest

in osteoporosis research, were contacted directly by telephone. Any women who were on

medications for heart disease, high blood pressure, cholesterol or lipid lowering, diabetes and/or

osteoporosis were excluded from participating.

Research design

This was a cross-sectional study in which participants were screened over the telephone to

determine study eligibility. Those interested in participating came to the hospital for an

appointment at which informed consent was obtained and participants were given detailed verbal

and written instructions on how to complete the 7-day estimated food records (Please refer to

Appendix I for the official informed consent package distributed to participants). In brief,

57
participants were asked to record everything that they consumed for 7 days in typical household

measures or grams/ounces if they had access to a food scale. They were also asked to record

dietary/vitamin supplement information, including amount and frequency of consumption and

the brand name. The participant’s height, weight and blood pressure were also taken, and

information was obtained from the participants regarding menopausal status, medication and

dietary/vitamin supplement use, smoking status and intake of alcohol and caffeine.

Dietary analyses

Food records were analyzed using NutriBase 5™ Clinical Edition software (version 5, released

2004, CyberSoft, Inc., Phoenix, AZ). This software generated a daily report for each participant

that included the amount of macro- and micronutrients obtained based on their consumption of

food, beverages and dietary/vitamin supplements. Included in this output was a daily energy

intake which was averaged for the 7-day period. Any days that were not within 30% of the

average were considered unusual and removed from the analysis. The remaining days were used

to calculate the average daily intake of all of the macro- and micronutrients for each participant.

Lycopene intake could not be assessed by this software and was therefore analyzed

separately using the USDA national nutrient database for lycopene as a reference, which lists the

content of lycopene in each food in µg/measure 177. Using this information, the lycopene content

was calculated in milligrams for each food, and an average of the total daily lycopene consumed

was calculated for each participant.

Analysis of serum lycopene

The serum concentrations of the antioxidants lycopene and other carotenoids were measured
178
using HPLC according to previously published methods , with minor modifications (Z. Liu,

unpublished observations). All chemicals were HPLC grade and were purchased from Sigma

Aldrich Canada, Oakville, ON, Canada. In brief, serum carotenoids were triple extracted with a

58
solution containing 0.0625% butylated hydroxy toluene (BHT)-Ethanol, 0.005% BHT-Hexane

and echinenone (internal standard, CaroteNature, Switzerland). Following nitrogen evaporation,

the residues were reconstituted in 100 L ethanol-BHT (0.0625%). Carotenoids were eluted with

acetonitrile and methanol (65:35, v/v) and 0.065% triethylamine, at a flow rate of 1.5 mL per

minute, using the Waters 2690 Alliance HPLC System and a Waters 996 PDA detector (Milford,

MA, USA). The analytical column used was a Waters Spherisorb 3 M ODS2, 4.6 x 250 mm.

The carotenoids lycopene, -cryptoxanthin, lutein/zeaxanthin and -/-carotene, were analyzed

at a wavelength of 450 nm. Lycopene was detected as all-trans, 5-cis and other-cis isomers (2

peaks). All carotenoids were determined using external standard calibration curves on the Waters

Millennium data management software, 4.0 edition (Milford, MA, USA) (for example

chromatogram, please refer to Appendix III, Figure IIIa).

Statistical analyses

All statistical analyses were performed using GraphPad (San Diego, CA) PRISM™ version 4

and SigmaStat® (Systat Software, San Jose, CA) statistical software version 2. Participants were

stratified according to age into two groups: group 1, 25–49 years old; and group 2, 50–70 years

old. Summary statistics of participant demographics, such as age and BMI, were generated.

These summary statistics were presented with mean and standard error of the mean (SEM)

values. Lycopene intake and serum lycopene were analyzed to determine the distribution of

intake, and results were presented with mean, standard deviation (SD), median, 25th and 75th

percentiles, 95% confidence interval (CI) of the mean and coefficient of variation (CV). All other

data were expressed as mean  SEM. Unpaired t-tests were used to compare types of lycopene-

containing foods consumed among age groups and high/low intake groups.

Pearson correlation was carried out to determine the associations between intake of

lycopene and the following macro- and micro-nutrients: carbohydrates (g), protein (g), fat (g),
59
saturated fat (g), fibre (g), cholesterol (mg), calcium (mg), folate (g), iron (mg), iodine (g),

magnesium (mg), niacin (mg), phosphorus (mg), riboflavin (mg), selenium (g), sodium (mg),

thiamin (mg), zinc (mg) and vitamins A (IU), B-12 (g), C (mg), D (IU), E (IU) and K (g).

Pearson correlation was also used to determine the association between serum lycopene,

lycopene intake and concentrations of other serum carotenoids. In cases where data were not

normally distributed, Spearman rank correlation was used as a substitute. Multiple linear

regressions were performed to determine whether intake of lycopene or serum lycopene could be

predicted by the following participant characteristics: age, BMI (see Table 3.1), smoking habits,

use of medications or dietary/vitamin supplements (see Table 3.2) and/or consumption of

calories, caffeine, wine, liquor, or beer (see Table 3.3). All other statistical methods used are

stated in the text.

60
 RESULTS AND DISCUSSION

The importance of this study is that it presents information about previously under-reported

lycopene intake in a sub-population of Canadian women of different age groups. The results will

be presented and discussed in Sections I to VI below.

I. Summary statistics for participants

In total, 108 recruited participants completed the study. Five of these participants were excluded

from analyses due to irregular dietary patterns during the week of the study which may have

interfered with the dietary portion of the analysis. Thus, the final sample size included 103

participants. Summary statistics for participants are given in Tables 3.1-3.3. Table 3.1 shows the

average age of the participants, as well as their average BMI, pulse rate and blood pressure,

which were taken to assess the overall health of the participant population. Overall, the groups of

participants had normal pulse rates and blood pressures for their respective ages. The BMI for

group 1 was in the healthy range; for the older group 2, the average BMI was a little higher, in

the overweight range, but this difference was not statistically significant. Information regarding

smoking and menopausal status as well as intake of dietary/vitamin supplements and medications

is shown in Table 3.2. In both age groups, the majority of participants were non-smokers (aged

25-49, 76.7% and aged 50-70, 66.7%) and did not take any medications (67.8% and 65.3%,

respectively, Table 3.2). However, dietary/vitamin supplement intake was quite common, with

more than 50% of the participants in each group consuming supplements daily or occasionally

(Table 3.2). In group 1, 16.1% of participants were menopausal, whereas 27.8% of participants

were menopausal in group 2 (Table 3.2). The parameters presented in Table 3.3 yielded further

information on the population of participants in this study and were used in correlation analyses

to determine if they could predict lycopene intake or serum lycopene in either age group (see

below).
61
Table 3.1: Summary statistics for study participants stratified into two groups according to age

Parameter Participant group

Group 1 (25–49 years old) Group 2 (50–79 years old)
N = 31 N = 72
Mean ± SEM Range Mean ± SEM Range
Age (years) 32.97 ± 1.41 25–49 56.01 ± 0.39 50–67
2
BMI (kg/m ) 23.93 ± 0.85 16.9–37.3 25.26 ± 0.52 18.0–41.4
Pulse rate (beats/minute) 71.64 ± 2.87 47–106 73.36 ± 1.29 52–98
Blood pressure (mm Hg) Systolic 117.8 ± 2.54 90–141 120.1 ± 2.05 88–153
Diastolic 73.35 ± 1.81 60–97 76.83 ± 1.31 55–96

Table 3.2: Percentage of study participants in each age group who smoked cigarettes, consumed
vitamins or medications and were menopausal

Parameter Percentage
Group 1 Group 2
(25–49 years old) (50–79 years old)
N =31 N =72
Smokers Never 76.7 66.7
Current 13.3 9.7
Previous (1 year)/social 10.0 23.6
Supplement use None 40.0 31.9
Occasional 10.0 4.2
Daily 50.0 63.9
Medication None 67.8 65.3
Birth control 29.0 0.0
HRT 0.0 8.3
Other 3.2 26.4
Menopausal status Pre 83.9 2.8
Menopausal (≤2 years) 16.1 27.8
Post (>2 years) 0.0 69.4

62
Table 3.3: Consumption of lycopene, calories, caffeine and alcoholic beverages for each group
of study participants

Product consumed Mean ± SEM values

Group 1 (25–49 years Group 2 (50–79 years
old) old)
N = 31 N =72
Lycopene (mg/day) 5.59 ± 0.93 6.23 ± 0.64
Energy (kcal/day) 1776 ± 89.17 1870 ± 78.75
Caffeine, coffee, tea (black and green), soda 1.32 ± 0.21 2.49 ± 0.21
(cups/day)
Wine (servings/week)1 1.47 ± 0.39 2.32 ± 0.39
Liquor (servings/week)2 0.62 ± 0.32 0.22 ± 0.12
Beer (servings/week)3 0.15 ± 0.10 0.49 ± 0.18
1
One serving = 6 oz = 177 g.
2
One serving = 1.5 fl oz = 42 g.
3
One serving = 341 mL bottle = 343 g.

II. Lycopene intake and serum lycopene of participants

The average lycopene intake for the 103 participants in this study was 6.03 ± 5.35 mg/day.

22.3% of these participants consumed what was considered to be a negligible lycopene intake

(<1.0 mg consumed/day), while only 8.9% consumed greater than 15 mg/day (data not shown).

These average intake values were higher than the reported values of 1.1–5.6 mg/day in the
90, 173, 174 171
United States and 1.64–5.01 mg/day throughout Europe . Some studies do report

intakes as high as 7–11 mg/day; however, these findings were limited to the older population

(ages 50 and above) 175, 179.

The average lycopene intake of 6.03 ± 5.35 mg/day in these participants corresponded to

an average total serum lycopene of 1346 ± 615.7 nM. The total lycopene serum concentration

included concentrations of 661.6 ± 331.3 nM all-trans, 425.9 ± 204.8 nM 5-cis and 258.6 ±

118.4 nM other-cis, which conferred a ratio of approximately 50:50 cis:trans lycopene. This ratio

of cis:trans in the serum is consistent with previous reports 92.

63
 In all participants, intake of lycopene was significantly and positively correlated with

total serum lycopene (r = 0.58, p<0.0001, Figure 3.1). Furthermore, there was a significant,

positive correlation between total cis lycopene in the serum and the number of processed/cooked

lycopene-containing foods consumed per week (r = 0.43, p<0.0001, data not shown). There was

no correlation between serum cis lycopene concentration and the number of raw lycopene-

containing foods consumed per week. This was to be expected, since cooked and processed

tomatoes contain greater quantities of total lycopene and higher concentrations of cis lycopene,

resulting in a more efficient absorption of total lycopene 168.

For this population of participants, the average lycopene intake was 5.59 ± 5.17 mg/day

for women 25–49 years old and 6.23 ± 5.45 mg/day for women 50–70 years old. These intakes

corresponded to average serum lycopene values of 1271 ± 544.0 nM and 1377 ± 644.2 nM,

respectively. Table 3.4 shows the statistical data for lycopene intake and serum lycopene in these

two groups. For both age groups, the intake of lycopene was highly variable (see high standard

deviations, Table 3.4), indicating a wide difference in lycopene consumption. However, the

distribution of lycopene intake was quite similar between the age groups, as can be seen by the

percentile distribution and CV. These observations may be explained by the fact that lycopene is

only found in a select group of foods 90. This intake data supports the only other Canadian study

that examined lycopene intake in approximately the same age groups of women, where average

intakes of 5.55 ± 10.41 mg/day were reported for younger women and intakes of 5.26 ± 9.77

mg/day were reported for older women; serum lycopene concentrations were not reported by

these authors 90. Some studies suggest that lycopene intake is much lower in the older population
171, 180
. In the present study, there was no significant difference in lycopene intake between the

two age groups (Table 3.3), which is similar to results previously reported for Canada 90. Despite

this high variation in lycopene intake, the serum lycopene followed a normal pattern of

64
distribution and had much lower coefficients of variation, suggesting that there may be a
181
saturation effect with respect to lycopene absorption . Just as there was no significant

difference between lycopene intakes among age groups, there was no difference in serum

lycopene.
90
The present study differs from the previously published Canadian study in that the

participant population was taken only from Toronto, while the previous Canadian study reported

intakes from five different geographical areas in Canada. Indeed, a few American studies have
182, 183
reported that geographical area affects lycopene intake , which may be attributed to
90
differences in food choices and/or availability. Although the Canadian study did not report

serum lycopene concentrations, those reported here are similar to those previously reported in the

United States 184, 185.

American studies which examined similar age groups, report average intakes with ranges
90, 173, 174, 176
of 1.1–5.6 mg/day . These previous studies used 24-hour recall or food frequency

questionnaires to report lycopene intake, which tends to introduce more error because
182
participants may not accurately remember intake of specific foods over time . A strength of

this study was that 7-day dietary records were used, and these have previously been established

as an accurate method of assessment due to the fact that they yield strong correlations between
132
intake and serum concentrations of lycopene . Further, these dietary records require

participants to provide exact quantities of lycopene-containing foods consumed, which allows for

a precise calculation of lycopene intake from each food. Error may be introduced if the

participants do not maintain their usual daily diets over the study period. In order to minimize

this source of error in the present study, participants were instructed to maintain their usual

dietary habits and indicate whether there were any unusual days so they could be excluded from

the dietary analyses.

65
 In terms of beneficial effects of lycopene consumption, the average intake of lycopene

reported here may seem low, especially when considering that most intervention trials supply at

least 15 mg/day of lycopene and that “therapeutic” levels are considered to be as high as 60
186
mg/day . However, population studies show that actual lycopene intakes provide much less
172, 173, 176
lycopene than what is typically supplied in intervention studies . In fact, in North

America, approximately 50% of the population consume less than 2 mg of lycopene per day 44.

Furthermore, in the Italian population, known to frequently consume lycopene-rich foods, and in

which lycopene is the most commonly consumed carotenoid, a recent study reported that the

average lycopene intake was only 7.4 mg/day 187. Overall, the intake levels reported in this study
90
and those previously reported in Canada tended to be higher than those of other European

countries such as Spain, Ireland, France, the Netherlands and the United Kingdom, where the

average daily intake was reported to be 1.64, 4.43, 4.75, 4.86 and 5.01 mg/day, respectively 171.

66
 4000

Total serum lycopene (nM)

3000

2000

1000

0
0 5 10 15 20 25
Lycopene intake (mg/day)

Figure 3.1: Correlation between lycopene intake and total serum lycopene in 103 female
participants between the ages of 25 to 70 years who provided fasting blood samples and 1 week
dietary records.

Table 3.4: Summary statistics of lycopene intake and serum lycopene for each age group of
study participants

Statistics Lycopene intake and total serum lycopene for participants grouped
according to age
Group 1 Group 2
aged 25–49 (N = 31) aged 50–79 (N = 72)
Lycopene intake Total serum Lycopene intake Total serum
(mg/day) lycopene (nM) (mg/day) lycopene (nM)
Mean ± SD 5.59 ± 5.17 1271 ± 544.0 6.23 ± 5.45 1377 ± 644.2
25th 0.84 792.8 2.06 907.9
percentile
Median 4.92 1333 5.02 1245
75th 8.96 1631 8.12 1803
percentile
Maximum 21.70 2543 21.43 3001
95% CI of (3.69, 7.48) (1068, 1474) (4.94, 7.51) (1226, 1529)
mean
CV (%) 92.49 42.81 87.60 46.77

67
III. Analysis of lycopene consumption by participants

Figure 3.2 shows the average consumption of lycopene-containing foods per week for

participants 25–49 and 50–70 years old. The type of lycopene-containing food that was

consumed most often in this population, for both age groups, was raw tomatoes. In fact,

participants in both age groups consumed significantly more raw tomatoes than all other sources

of lycopene combined (aged 25-49, p<0.0005 and aged 50-70, p<0.0001, Figure 3.2). Similar

findings have been reported for the United States 182, France, Spain and the United Kingdom 171,

whereas in Ireland and the Netherlands, processed tomato products tended to be consumed more
171
frequently . The Canadian study on lycopene intake in a female population showed that

cooked/processed tomato products, such as tomato sauce, were consumed more frequently 90. It

may be that this discrepancy is caused by seasonal variation in lycopene consumption by

participants in these Canadian studies. However, reports suggest that the effect of seasonal
171
variation on lycopene intake tends to be minor and primarily occurs in the summer months,

when the consumption of watermelon and raw tomatoes might be increased because of

availability 188.

These findings showed that participants 25–49 years old consumed significantly more

dried tomatoes/tomato powder (p<0.05), pizza (p<0.002) and slightly more ketchup (not

significant, p<0.10) than participants 50–70 years old. On the other hand, participants 50–70

years old consumed on average more tomato juice (p<0.05) than the younger participants. These

findings are consistent with previous reports 182 and may be because of the perception among the

older age group that products such as pizza and ketchup are associated with higher calorie, fat

and salt intake, and should therefore be consumed less frequently to reduce the risk of chronic

diseases, such as hypertension 189, 190. This was further demonstrated by the fact that, on average,

participants 25-49 years old consumed almost twice as much lycopene from cooked/processed

68
sources (5.81 ± 0.69 servings per week), than participants aged 50-70 (3.65 ± 0.38 servings per

week) (p<0.005).

Figure 3.2: Frequency of lycopene

consumption by 103 participants for a 1-

To
m
week period, showing all sources of Average number of

at
o
lycopene consumed. The types of times consumed/week

dr To
ie
d/
lycopene-containing foods are listed,

3
po o
w
m
from left to right, from highest to lowest

de ast ehi ne uc hu izz als uic uic ou elo ke ra ng ou fru uc sin ra mi

at

*
r
concentration of lycopene in g/g. Data

p
T

R
om To

e
os a
shown are grouped according to

at ma
o
participant age and are mean  SEM. An

p
c
n
unpaired t-test was used to test for

to

d
sa
differences in type of food consumed
Type of lycopene-containing food

e
e
(*p<0.05, p<0.002, and  p<0.0005 for

tc
p
group 1, and p<0.0001 for group 2).


Ve

a
ge To To

S
ta m m
b

a
le
j
a

e
to
j
at

e
o

*
W

s
at
T
Pa

e
om

p
r
pa

m
a
ya

to om od tab gr s

n
c
gu

o
T

o
av
a

d
a
bl eg

t o
o
V


w
o r
e

a
Pi Be sa Re
C

e
le
nk an uc d
hi BQ

s
li/
B

p
p e
in dr pp ltiv

it
sa
e Pe

e
es
g
e
M

r ita
u

w n

Group 2
Group 1

69
IV: Correlation of lycopene intake and serum lycopene with

participant characteristics, serum carotenoids and intake of macro-

and micronutrients

Statistical analyses were performed to determine whether daily lycopene intake correlated with

daily intake of macro- and micronutrients and whether lycopene intake or serum lycopene could

be predicted by other participant characteristics. For example, BMI may be associated with

lycopene intake due to a higher caloric intake over time. No association between lycopene

intake or serum lycopene and age, BMI, smoking and/or use of medications or dietary/vitamin

supplements was found in the present study; a finding that differs from other studies that
182
focussed on serum lycopene only , in which associations were found. However, because
132
previous studies, including ours , found a correlation between lycopene intake and serum
99, 170
lycopene , this discrepancy is unclear at present. In addition, no significant association

between lycopene intake and intake of other macro- and micronutrients in participants aged 25-

49 was found. However, in participants 50–70 years old, lycopene intake was positively

correlated with intake of niacin (r = 0.31, p<0.01), vitamins A (r = 0.26, p<0.05), D (r = 0.27,

p<0.05) and K (r = 0.43, p<0.001) (data not shown). The correlation between lycopene and

vitamin A intake in older participants is to be expected because many foods high in lycopene are

also high in vitamin A. Alternatively, the absence of correlation between lycopene intake and

micronutrients in the younger group may have occured because participants in the younger group

tended to have a more varied diet, choosing different foods for each meal and more day-to-day

variety, whereas older participants in this population tended to have the same meal for breakfast

and lunch at least three to four times per week (data not shown).

In participants aged 25-49, serum lycopene was positively correlated with -

cryptoxanthin (r = 0.52, p<0.005). In participants aged 50-70, there were significant, positive
70
correlations between lutein/zeaxanthin (r = 0.27, p<0.05), - (r = 0.53, p<0.0001) and -carotene

(r = 0.38, p<0.002). Of the many types of carotenoids found in tomatoes and tomato products,

lycopene is the most abundant, accounting for as much as 90% of the carotenoid concentration

overall 93. However these foods do contain moderate concentrations of other carotenoids such as
89, 191
-carotene and lutein/zeaxanthin and their consumption has been shown to significantly
192
increase serum concentrations of these other carotenoids . The increase in -carotene seen in

the older participants only could be due to the fact that older participants consumed significantly

more tomato juice per week (compared to younger participants, p<0.05, Figure 3.2), which

contains 2.6 g/g of -carotene 89.

V: Comparison of lycopene consumption between low and high

quartiles of intake

Table 3.5 presents data on lycopene consumption and serum lycopene for the lowest and highest

quartiles of intake for each age group. In both age groups, participants with the highest quartile

of lycopene intake consumed significantly more cooked/processed lycopene foods than

participants in the low quartile of intake. This was to be expected because cooked/processed

lycopene-containing foods have higher concentrations of lycopene than the raw food sources 168.

As expected, this high quartile of intake corresponded to significantly higher concentrations of

serum lycopene, including all-trans, 5-cis and other-cis isomers (Table 3.5). Cooking or

processing of tomato products enhances bioavailability of lycopene because processing reduces

particle size and divides the plant cell wall and fibrous pieces 109 and releases lycopene from its
85
food matrix into the lipid component of the food . It can also cause isomerization to the cis

form, which may result in better absorption of lycopene 193. This was demonstrated by the results

showing that a high consumption of cooked/processed lycopene-containing foods was

71
significantly and positively associated with high serum concentrations of total cis lycopene

(p<0.0001, Section II above).

For participants 25–49 years old (group 1), those with low lycopene intake (according to

quartiles) had an average intake of 0.67 ± 0.17 mg/day (Table 3.5) and did not consume

lycopene-containing foods on a daily basis. A comparison between types of lycopene-containing

foods consumed among these participants showed that there was no significant difference

between consumption of raw or cooked/processed lycopene-containing foods, and intake levels

of both were low (0.88 ± 0.64 and 0.75 ± 0.71 servings per week, respectively). Participants in

this age group with a high lycopene intake (according to quartiles) had an average intake of

13.01 ± 4.4 mg/day (Table 3.5); 100% of these participants consumed lycopene-containing foods

on a daily basis. A comparison between types of lycopene-containing foods consumed among

these participants showed that consumption of cooked/processed lycopene-containing foods

(6.57 ± 2.57 servings per week) was significantly higher than consumption of raw foods (1.71 ±

1.80 servings per week) (p<0.005). When compared to participants of the same age who were

considered to have a low intake of lycopene, there was no significant difference in the number of

servings of raw lycopene-containing foods; however, those with a high intake consumed a

significantly higher number of servings of cooked/processed lycopene-containing foods than

those with a low intake (p<0.0001, Table 3.5). As was expected, participants with the highest

quartile of lycopene intake had significantly higher all-trans, 5-cis and other-cis lycopene (952.9

± 199.2 nM, 722.3 ± 298.1 nM, 370.4 ± 128.0 nM) than participants with low intake (448.1 ±

221.2 nM, 290.7 ± 113.9 nM, 205.6 ± 102.1 nM) (p<0.005, p<0.005 and p<0.02, respectively,

Table 3.5).

For participants 50–70 years old (group 2) those with low lycopene intake (according to

quartiles) had an average intake of 0.59 ± 0.65 mg/day (Table 3.5); only 5% of these participants

72
consumed lycopene-containing foods on a daily basis. A comparison between types of lycopene-

containing foods consumed among these participants showed that consumption of raw lycopene-

containing foods (1.33 ± 1.50 servings per week) was significantly higher than that of

cooked/processed lycopene-containing foods (0.28 ± 0.46 servings per week) (p<0.01).

Participants in this age group with a high lycopene intake had an average intake of 14.14 ± 3.95

mg/day (Table 3.5); 94% of these participants consumed lycopene-containing foods on a daily

basis. A comparison between types of lycopene-containing foods consumed among these

participants showed that consumption of both raw and cooked/processed lycopene-containing

foods was high (4.44 ± 3.37 and 5.94 ± 2.34 servings per week, respectively). In this age group

there was no significant difference between consumption of either raw or cooked/processed

lycopene-containing foods for high-intake participants. Compared to participants of the same age

who were considered to have a low lycopene intake, those with high lycopene intake consumed

significantly more servings per week of both raw (p<0.005) and cooked/processed lycopene-

containing foods (p<0.0001). As with participants in the younger age group, participants in the

highest quartile of lycopene intake had significantly higher all-trans, 5-cis and other-cis lycopene

(966.5 ± 303.6 nM, 608.9 ± 180.1 nM, 358.5 ± 89.45 nM) than participants with low intake

(429.2 ± 181.4 nM, 290.7 ± 113.9 nM, 205.6 ± 102.1 nM) (all: p<0.0001, Table 3.5).

To determine differences in age for those considered to have both low and high lycopene

intake (according to quartiles), group 1 (25–49 years old) was compared to group 2 (50–70 years

old). There were no significant differences between average daily intake for those with low

lycopene intake (0.67 ± 0.17 for group 1, 0.59 ± 0.65 for group 2) or for those with high

lycopene intake (13.01 ± 4.4 for group 1, 14.14 ± 3.95 for group 2, Table 3.5). For participants

with a low lycopene intake, group 1 consumed significantly more cooked/processed lycopene-

containing foods than group 2 (p<0.05), whereas group 2 tended to consume more raw lycopene-

73
containing foods (effect not significant). For participants with a high lycopene intake,

consumption of cooked/processed lycopene-containing foods was similar between age groups,

whereas the older group consumed slightly more raw lycopene foods (not significant, p= 0.06).

Table 3.5: Consumption of lycopene for the lowest and highest quartiles of intake, showing
number of servings of raw or cooked/processed lycopene foods for each age group and serum
lycopene

Lycopene (mean ± SD) Lowest Quartile of Highest Quartile of

Intake Intake
Group 1 Group 2 Group 1 Group 2
(Age 25-49) (Age 50-79) (Age 25-49) (Age 50-79)
Average lycopene intake (mg/day) 0.67 ± 0.17 0.59 ± 0.65 13.01 ± 4.4 14.14 ±
3.95
Number of servings raw lycopene 0.88 ± 0.64 1.33 ± 1.50a 1.71 ± 1.80 4.44 ± 3.37b
foods per week1
Number of servings of 0.75 ± 0.71c 0.28 ± 0.46 6.57 ± 5.94 ± 2.34d
cooked/processed lycopene foods 2.57d,e
per week2
Total serum lycopene (nM) 944.4 ± 883.5 ± 1909 ± 1934 ±
448.1 350.1 399.5 f 553.3 d
all-trans lycopene (nM) 448.1 ± 429.2 ± 952.9 ± 966.5 ±
221.2 181.4 199.2 e 303.6 d
Total cis lycopene (nM) 496.3 ± 454.3 ± 956.4 ± 967.4 ±
207.0 173.9 233.8 e 264.8 d
5-cis lycopene (nM) 290.7 ± 279.0 ± 722.3 ± 608.9 ±
113.9 110.8 298.1 e 180.1 d
other-cis lycopene (nM) 205.6 ± 175.3 ± 370.4 ± 358.5 ±
102.1 69.06 128.0 g 89.45 d
1
Includes raw tomatoes, pink grapefruit, watermelon and papaya.
2
Includes cooked tomatoes, ketchup, tomato juice/sauce, etc.
a
Significantly higher consumption than cooked/processed foods in same age/quartile group (p<0.01).
b
Significantly higher consumption than in lowest quartile group of same age (p<0.005).
c
Significantly higher consumption cooked/processed foods than group 2, low quartile of intake (p<0.05).
d, e, f , g
Significantly higher than in lowest quartile group of same age (d p<0.0001, e p<0.005, f p<0.001, g p<0.02).

74
VI: Summary and Conclusions

In summary, this chapter showed that: (1) participants had an average lycopene intake of 6.03 ±

5.35 mg/day, which is similar to previously reported results on lycopene intake in Canadian

women 90 and higher than women in other countries in North America and Europe; (2) this intake

of lycopene corresponded to an average serum lycopene concentration of 1346 ± 60.97 nM; (3)

lycopene intake is highly variable, as demonstrated by the wide standard deviation of the intake,

but is similar among age groups; (4) in both age groups, the most frequently consumed lycopene-

containing food was raw tomatoes, and participants with the highest intake of lycopene

consumed significantly more cooked/processed tomato products than those with lower lycopene

intake; (5) participants 25–49 years old consumed significantly more cooked/processed sources

of lycopene, particularly pizza and dried tomato/tomato powder, than participants 50–70 years

old, who consumed significantly more tomato juice; (6) there was no association between

lycopene intake or serum lycopene and age, BMI, smoking, caloric intake and/or use of

medications or dietary/vitamin supplements in either age group. However, in older participants,

50–70 years old, lycopene intake was positively correlated with intake of niacin and vitamins A,

D and K; and (6) serum lycopene was positively correlated with serum -cryptoxanthin in

younger participants and lutein/zeaxanthin, - and -carotene in older participants.

The present study provides important information on intake of lycopene and resulting

serum lycopene concentrations in a sub-population of Canadian women using a 7-day estimated

food record. Prior to this study, information on average lycopene intake and serum
90
concentrations in Canadian women has been limited . These results showed that lycopene

intake is similar among different age groups of women, although the types of lycopene-

containing foods consumed differ. Compared to populations in America and parts of Europe,

women in this study tended to have a higher lycopene intake, which could be explained by a

75
wide variety of lycopene-containing foods, particularly tomatoes and tomato products, which

were consumed on a fairly regular basis.

As of yet, there is no official, established recommendation for lycopene intake or how

many servings of lycopene-containing foods are necessary to provide beneficial serum lycopene

concentrations. Studies suggest that an intake as little as 6-8 mg per day may result in serum
132, 194, 195
lycopene concentrations capable of exerting moderate protective antioxidant effects .

To obtain these intake levels, one could consume 1.5 cups of raw tomato, 3 tablespoons of

ketchup or ¼ cup of tomato sauce/juice per day 89. However, other intervention studies suggest a

higher intake may be necessary to appreciably reduce the risk of developing chronic diseases

such as cancer, cardiovascular disease and hypertension, particularly in people considered to be

at “high risk” 44, 97, 99, 115, 120, 196. This idea will be investigated further in Chapters 5 and 6 of this

thesis.

Health benefits of lycopene intake include the risk reduction of diseases important in
44, 125, 132
women’s health , thus the findings reported here are noteworthy because they provide

important information on lycopene consumption in a sub-population of women in Canada. This

information may be useful in promoting and formulating the establishment of nutrient and health

recommendations regarding lycopene intake for the prevention of chronic diseases prevalent in

women’s health, which can lead to maintaining an excellent quality of life.

76
 CHAPTER 4

Dietary restriction of lycopene for a period of one month

resulted in significantly increased biomarkers of oxidative
stress and bone resorption in postmenopausal women

This chapter has been submitted for publication to the Journal of Nutrition, Health and Aging as
a research paper entitled: “Dietary restriction of lycopene for a period of one month significantly
increased biomarkers of oxidative stress and bone resorption in postmenopausal women” by E.S.
Mackinnon, A.V. Rao, and L.G. Rao
77
 ABSTRACT

Background & Aims: Lycopene is a carotenoid commonly found in tomatoes and tomato

products which acts as an antioxidant to decrease oxidative stress and osteoporosis risk. This

paper set out to determine the effects of a lycopene-restricted diet on oxidative stress parameters

and bone turnover markers in postmenopausal women. Methods: Twenty-three healthy

postmenopausal women, 50-60 years old, provided blood samples at baseline and following a

one-month lycopene-depletion period. Serum samples were analyzed for carotenoids; the

oxidative stress parameters protein thiols and TBARS; the antioxidant enzymes SOD, CAT and

GPx, and the bone turnover markers BAP and NTx. A paired t-test measured changes after the

lycopene restricted diet. Results: Lycopene restriction resulted in significantly decreased serum

lycopene (p<0.0001), lutein/zeaxanthin (p<0.01) and -/-carotene (both: p<0.05). GPx

(p<0.01), lipid and protein oxidation increased (not significant, p=0.05 and p<0.10, respectively),

while CAT and SOD were significantly depressed (p<0.05 and p<0.005, respectively). These

changes coincided with significantly increased NTx (p<0.05). Conclusions: These findings

emphasize the importance of daily lycopene consumption as an antioxidant functioning to

decrease bone resorption in postmenopausal women. This study provides further proof that

lycopene may be beneficial in reducing the risk of osteoporosis.

78
 INTRODUCTION
197
Lycopene is a 40-carbon, acyclic isomer of -carotene . Among the carotenoid family, it is
117
credited with the highest singlet oxygen quenching capacity , which makes it a powerful

antioxidant. It is a highly unsaturated, open, straight chain hydrocarbon, containing 11

44, 197
conjugated and 2 unconjugated double bonds . Lycopene is the most predominant
197
carotenoid found in human serum . Over 80% of lycopene consumed in the diet is obtained
85
through consumption of tomatoes and tomato products . However, it is also found in

watermelon, pink grapefruit, rosehips and pink guava 44, 87, 93.

Lycopene exists in different isomeric forms. There is the all-trans form of lycopene, and

several different cis isomers, which are formed after rotation of any of the 11 conjugated double

bonds. Cis isomers are credited with the highest antioxidant capacity, with 5-cis lycopene having
44, 116
the highest antioxidant capacity and all-trans lycopene having the lowest . In lycopene-

containing foods such as raw tomatoes, lycopene exists primarily in the all-trans form, but in
85
human serum a higher concentration of cis isomers exists . This increase in isomer

concentrations could result from exposure to high temperatures during food processing, but more

likely it is a result of selective uptake and/or differential bioavailability of cis isomers 44, 111, 197.

Lycopene is well documented for its ability to decrease biomarkers of oxidative stress.

Research shows that it is capable of decreasing several lipid, protein and DNA markers 97, 99, 115,
118
. In some studies, it has also demonstrated a capacity to increase markers of antioxidant
198-200
capacity and endogenous antioxidant enzymes . The antioxidant properties of lycopene

have been credited with its ability to decrease the risk of age-related chronic diseases often

attributed to oxidative stress. Research suggests that through its antioxidant capacity, lycopene

may decrease the risk of infertility, diabetes, dementia, cardiovascular disease and several types
44, 95 132
of cancer . Our previous and current studies described in this thesis, provide further

79
evidence that lycopene may also decrease the risk of osteoporosis in postmenopausal women.

This link between lycopene and chronic diseases has been demonstrated through an inverse

association of lycopene with biomarkers of oxidative stress, not only through epidemiological

studies, but also through dietary intervention studies. Many of these intervention studies require

a washout period, during which lycopene is restricted in the diet. Lycopene restriction is

necessary to obtain a clinically relevant response to lycopene, as it has been shown that higher

baseline lycopene concentrations result in a moderate to null response to lycopene 201. However,

studies on the effects of a carotenoid-depleted diet are limited and those that exist examined

small study populations (N<10), very few biomarkers of oxidative stress and did not report on
202-204
how this type of diet would affect the risk of developing age-related chronic diseases .

Studies on lycopene depletion are important not only because they help identify crucial

functions, but they also help to elucidate mechanisms which may not be accurately demonstrated

in intervention studies where absorption saturation occurs 205.

We have previously shown a correlation between high serum lycopene and decreased
132
protein oxidation and bone resorption markers in postmenopausal women . This chapter

reports findings on lycopene restriction in postmenopausal women, showing that refraining from

consuming lycopene containing foods for a period of just one month has significant effects on

not only on oxidative stress parameters, but also antioxidant capacity and bone resorption

markers, further illustrating the importance of lycopene in the daily diet to maintain overall

health, particularly in bones.

80
 METHODS

Participant Recruitment and Blood Sample Collection

The protocol was approved by the Research Ethics Board at St. Michael's Hospital, Toronto,

Ontario, Canada and followed the guidelines of good clinical practices. Please refer to Appendix

I for the official informed consent package distributed to participants. Female participants

between 50-60 years old, who were at least one year postmenopausal, were recruited by

telephone and advertisements as part of a larger two-part clinical study described in Chapters 3

and 5 of this thesis. Any participants who were on medications for heart disease, high blood

pressure, diabetes and/or osteoporosis were excluded as were participants who smoked

cigarettes. Participants submitted dietary records outlining foods, beverages and nutritional

supplements consumed over the previous 7 days and provided a baseline 12-hour fasting blood

sample. Another set of dietary records and a fasting blood sample was collected following a one-

month washout period during which no lycopene-containing foods were consumed. Please refer

to Form F, Appendix I, for a composite list of foods participants were asked to avoid.

Serum Analyses

Measurements of antioxidant capacity

Unless otherwise specified, all materials were obtained from Sigma Aldrich Canada, Oakville,
178, 206
ON, Canada. Two methods reported in the literature were used to determine the

antioxidant capacity of the serum. The first, HPLC, was performed to determine the

concentrations of serum antioxidants lycopene and other carotenoids according to previously

178
published methods , with minor modifications (Z. Liu, unpublished observations). All

chemicals were HPLC grade. Serum carotenoids were triple extracted using a solution of

0.0625% BHT-Ethanol, 0.005% BHT-Hexane and echinenone internal standard (CaroteNature,

81
Switzerland). Following evaporation under a stream of nitrogen gas, the residues were

reconstituted in 100 L ethanol-BHT (0.0625%). Carotenoids were eluted with acetonitrile and

methanol (65:35, v/v) and 0.065% triethylamine, at a flow rate of 1.5 mL per minute, using the

Waters 2690 Alliance HPLC System and a Waters 996 PDA detector (Milford, MA, USA). The

analytical column used was a Waters Spherisorb 3 M ODS2, 4.6 x 250 mm. Separation was

performed by elution with a solvent containing acetonitrile and methanol (65:35, v/v) and

0.065% triethylamine, at a flow rate of 1.5 mL per minute. The carotenoids lycopene, -

cryptoxanthin, lutein/zeaxanthin and -/-carotene, were analyzed at a wavelength of 450 nm.

Lycopene was detected as all-trans, 5-cis and other-cis isomers. All carotenoids were determined

using external standard calibration curves on the Waters Millennium data management software,

4.0 edition (Milford, MA, USA) (for example chromatogram, please refer to Appendix III,

Figure IIIa).

Total antioxidant capacity (TAC) was also measured, using the Trolox equivalent
206
antioxidant capacity assay (TEAC) to determine the overall capacity of antioxidants in vitro.

Antioxidant molecules quench the radical cation 2,2, azinobis (3-ethylbenzothiazoline-6-sulfonic

acid) (ABTS), a blue-green chromophore. The absorbance was read at 734 nm to determine

decolorization (Milton Roy Spectronic 1001 Plus, PA, USA), and compared to the decolorization

produced by Trolox, a vitamin E analogue.

Measurements of oxidative stress parameters

207
Protein oxidation was determined by estimating protein-sulfhydryl groups (thiols) in serum .

The optical density was read at 412 nm (Milton Roy Spectronic 1001 Plus, PA, USA) and

protein thiols were calculated using an absorptivity of 13600 cm-1 M-1. A high concentration of

protein thiols corresponds to a lower protein oxidation.

82
 Lipid peroxidation was measured in the serum using the thiobarbituric acid-
208
malondialdehyde (TBA-MDA) assay and was reported as TBARS . The optical density was

read at 535 nm (Milton Roy Spectronic 1001 Plus, PA, USA) and the concentration of TBARS

was calculated using an absorptivity of 156 mM-1cm-1.

Measurements of endogenous antioxidant enzymes

Hemolysates were prepared using 0.5 mL of red blood cells (RBC) mixed with 2 mL double

distilled water (ddH2O); total hemoglobin (Hb) and CAT were measured immediately after

hemolysate preparation, and GPx and SOD were measured after an overnight freeze.

Total Hb was detected in RBC hemolysates using the Drabkin’s method (Sigma Aldrich,

Canada), according to the manufacturer’s protocol and using a standard curve generated from

cyanmethemoglobin.

CAT was measured using the H2O2 decomposition method, which measures the change

in the optical density (OD) of the sample as CAT converts H2O2 into water and oxygen 209. The

decrease in OD of H2O2 was measured at 20ºC at 240 nm (Milton Roy Spectronic 1001 Plus, PA,

USA) for 30 seconds. CAT activity was expressed as K/g Hb, where K = 2.3/(Δ time)log OD15

seconds/OD 30seconds.

210
SOD was determined by the auto-oxidation of epinephrine . SOD acts as a potent

inhibitor of the auto-oxidation of epinephrine at an elevated pH of 10.2 at 30ºC. The OD was

measured at 480 nm (Milton Roy Spectronic 1001 Plus, PA, USA) for 10 minutes and SOD

enzyme activity (U) was defined as 50% inhibition of the rate of epinephrine oxidation compared

to a negative control sample. SOD was normalized to milligrams of Hb and expressed as U/mg

Hb.

GPx activity was measured by the decrease in NADPH absorbance at 340 nm at 37ºC in

the presence of H2O2 211. One enzyme unit of GPx was defined as the number of micromoles of

83
NADPH oxidized per minute, where OD1M340 =6.2. GPx activity was normalized to grams Hb

and expressed as U/g Hb.

Measurements of bone turnover markers

Commercially purchased ELISA kits were used to measure the serum concentrations of the bone

turnover markers crosslinked N-telopeptide of type I collagen (NTx), a measure of bone

resorption (INTER MEDICO, Ontario) and bone-specific alkaline phosphatase (BAP), a marker

of bone formation (ESBE Scientific, Ontario). NTx values were expressed as nanomoles of bone

collagen equivalents per litre (nM BCE) and BAP was expressed as U/L, where one Unit

represented 1 mol of p-nitrophenyl phosphate hydrolyzed per minute at 25C in 2-amino-2-

methyl-1-propanol buffer.

Statistical Analyses

All statistical analyses were performed using GraphPad PRISM 5.00 for Windows (GraphPad

Software, California). Summary statistics of participant demographics such as age and BMI were

generated and presented as means ± standard errors of the mean (SEM). A paired t-test was used

to test for significant differences in bone turnover markers, oxidative stress parameters and

antioxidant status after lycopene restriction. For data that were not normally distributed, the

Grubb’s test for outliers was used to exclude the offending outlier. If data were not normally

distributed after this test, then the Wilcoxon matched pairs test was used. Statistical significance

was considered at p<0.05. Percent change in bone turnover markers, oxidative stress parameters

and antioxidant status were also calculated to determine trends of the change. Any other statistics

used are specified in the text.

84
 RESULTS

Participant characteristics

Twenty-three postmenopausal women participated in this study. The characteristics for these

participants are shown in Table 4.1, including age, BMI and years since menopause. The average

lycopene intake for participants prior to the restrictive diet was 3.5 ± 0.6 mg per day (Table 4.1),

which is lower than averages we and others have previously published for this age group (Please
90, 132
refer to Chapter 3) . The range of lycopene intake was quite wide, from 0 to 12.3 mg per

day, with 39% of participants consuming a low lycopene intake which provided 2 mg or less per

day.

Table 4.1: Participant characteristics as determined at the beginning of the study for 23
postmenopausal, female participants, aged 50-60.

Baseline participant characteristic Mean ± SEM

Age (years) 54.4 ± 0.6
BMI (kg/m2) 24.7 ± 0.9
Years since menopause 4.1 ± 0.6
Daily lycopene intake (mg/day) 3.5 ± 0.6
Total serum lycopene (nM) 1171.0 ± 111.1

Lycopene restriction significantly decreased all-trans and total cis

serum lycopene concentrations

After one-month during which participants refrained from consuming lycopene-containing foods,

mean lycopene intake significantly decreased from 3.50 ± 0.60 to 0.13 ± 0.06 mg/day (p<0.001).

This corresponded to a significant decrease in total serum lycopene of 54.86 ± 3.59% (p<0.0001,

Table 4.2). Concentrations of all-trans and cis isomers were significantly lower in the serum of

participants after lycopene restriction (p<0.0001, Figure 4.1). In fact, all-trans, 5-cis and other-

cis lycopene decreased by averages of 59.00 ± 4.45%, 53.37 ± 3.64% and 46.84 ± 4.32%,

respectively (all: p<0.0001).

85
 In addition to significant decreases of 54.86 ± 3.59% in serum lycopene, there were also

significant decreases of 12.77 ± 5.03% for serum lutein/zeaxanthin (p<0.01), 13.03 ± 6.88 for

serum - carotene (p<0.05) and 22.84 ± 5.09 for -carotene (p<0.0005) after lycopene restriction

(Table 4.2) compared to baseline values. However, the overall percent change in these serum

carotenoids was not as high as that seen for lycopene. In fact, an unpaired t-test was performed

and the percent change in lycopene (-54.86%) was significantly greater than that of each of the

carotenoids (p<0.0001 for decrease in serum lycopene vs. decrease in each carotenoid),

demonstrating that the greatest decrease in serum carotenoids after restriction was seen for

lycopene. There was no effect on TAC of samples after lycopene restriction (data not shown).

800
Baseline
Serum lycopene (nM)

Lycopene restricted
600

400
*
*
200 *

0
all-trans 5-cis other-cis

Figure 4.1: Serum concentrations of all-trans, 5-cis and other-cis lycopene in 23

postmenopausal female participants who refrained from consuming lycopene-containing foods
for a period of one month. Values are mean  SEM; baseline and lycopene restricted values were
compared using a paired t-test (*p<0.0001).

86
Table 4.2: Change in serum carotenoid concentrations after postmenopausal female participants
refrained from consuming lycopene containing foods for a period of 1 month.

Carotenoid Concentration in serum (nM) Results of Average %

Baseline Lycopene paired t-test1 change
(mean ± SEM) restricted (mean ± SEM)
(mean ± SEM)
-carotene 408.4 ± 131.4 334.4 ± 110.6 p<0.05 -13.03 ± 6.88
-carotene 1443.0 ± 278.9 1035.0 ± 221.7 p<0.0005 -22.84 ± 5.09
-cryptoxanthin 403.2 ± 58.4 367.6 ± 47.3 p = 0.229 -1.29 ± 8.21
Lycopene 1171.0 ± 111.1 494.9 ± 48.46 p<0.0001 -54.86 ± 3.59a

Lutein/zeaxanthin 516.4 ± 49.57 443.0 ± 47.04 p<0.01 -12.77 ± 5.03

1
Wilcoxon matched pairs test used for these non-normally distributed data sets.
a
Average percent change in lycopene was significantly higher than that seen for all of the other carotenoids
(p<0.0001), as determined by unpaired t-test or Mann-Whitney test.

Lycopene restriction increased oxidative stress parameters and

affected endogenous antioxidant enzyme activities

Lycopene restriction for a period of one month resulted in marginally decreased protein

thiols from 423.7 ± 19.31 M to 392.3 ± 14.22 M, conferring an apparent increase in protein

oxidation (not significant, p=0.05). There was also a marginal increase in TBARS, from a

concentration of 8.08 ± 0.44 nmol/ml to 9.18 ± 0.76 nmol/ml, indicating an apparent increase in

lipid peroxidation (not significant, p<0.10). Protein oxidation and lipid peroxidation increased by

5.45 ± 3.29% and 14.50 ± 7.08%, respectively.

Concentrations of the endogenous antioxidant enzymes CAT and SOD were significantly

decreased after lycopene restriction (p<0.05 and p<0.005, respectively, Figure 4.2). The average

decrease was 8.43 ± 9.30% for CAT and 22.71 ± 11.84% for SOD. Conversely, concentrations

of GPx were significantly increased (p<0.01, Figure 4.3) after lycopene restriction with an

average increase of 114.7 ± 35.82% in enzyme activity. There was a significant, positive linear

87
relationship between this change in GPx and the change in TBARS after lycopene restriction

(slope = 0.050, p<0.05, data not shown).

100
100

80
* 80

SOD (U/mg Hb)

CAT (K/g Hb)

60 60

40 40
**
20 20

0 0
Baseline Lycopene restricted Baseline Lycopene restricted

Figure 4.2: Enzyme activity of the endogenous antioxidant enzymes, CAT and SOD, at baseline
and after 1 month of lycopene restriction in 23 postmenopausal participants. Values are mean 
SEM; baseline and lycopene restricted values were compared using a paired t-test (*p<0.05 and
** p<0.005, respectively).

40
Glutathione peroxidase

*
30
(U/g Hb)

20

10

0
Baseline Lycopene restricted

Figure 4.3: Activity of the endogenous antioxidant enzyme GPx in 23 postmenopausal

participants who refrained from lycopene consumption for a period of 1 month. Values are mean
 SEM; baseline and lycopene restricted values were compared using a paired t-test (*p<0.01).

88
Lycopene restriction significantly increased the bone resorption

marker NTx

Postmenopausal participants who refrained from consuming lycopene for a period of 1

month had significantly increased concentrations of the bone resorption marker, NTx (p<0.05,

Figure 4.4). The average increase in NTx was 20.60 ± 9.81%. The bone formation marker, BAP,

remained stable throughout the study period and there were no changes as a result of lycopene

restriction (data not shown).

30
*
NTx (nM BCE)

20

10

0
Baseline Lycopene restricted

Figure 4.4: Concentrations of the bone resorption marker, NTx, in 23 postmenopausal

participants who refrained from lycopene consumption for a period of 1 month. Values are mean
 SEM; baseline and lycopene restricted values were compared using a paired t-test (*p<0.05).

89
 DISCUSSION

This chapter presented data showing that lycopene restriction, for a period of only one

month, resulted in important changes in biomarkers of oxidative stress and bone resorption

markers. Refraining from consuming lycopene-containing foods resulted in significantly lower

serum carotenoids, particularly lycopene, which coincided with apparent increases in oxidative

stress parameters, significant increases in the bone resorption marker NTx and GPx enzyme

activity, and depressed CAT and SOD enzyme activities. This is the first study reporting solely

on the effects of dietary lycopene restriction on these parameters in postmenopausal women who

are at higher risk for osteoporosis.

The average lycopene intake at baseline for this subset of participants was lower, at 3.5
91
mg per day, compared to the 6.23 mg/day reported by us (p<0.05) and the 5.26 mg/day
90
reported by others for Canadian women in this age group. The fact that participants in this

study, with an initial lower than average lycopene intake, demonstrated such significant and

important changes in biomarkers of oxidative stress and bone resorption after only one month of

refraining from consuming lycopene-containing foods, further illustrates that even a small daily

intake of lycopene is biologically important and may offer protection against the effects of

oxidative stress, particularly on bones.

The reported half-life of lycopene in human serum ranges from as little as 2-3 days 104, to
105, 106
as high as 12-33 days . Restriction of lycopene-containing foods for only 12-14 days has

been reported to decrease serum lycopene by 50% 106, 212, which is similar to what was reported

here after 30 days (Table 4.2). This change in serum lycopene after restriction, reported by us
106, 212
and others , suggests that in the present study, the simultaneous effects shown on

biomarkers of oxidative stress and bone resorption occur as a direct result of lycopene depletion.

90
 Despite the significant changes in serum carotenoid concentrations after lycopene

depletion, there was no change in overall TAC of the serum, contrary to what has been
192
previously reported . Many intervention studies on carotenoids demonstrate that their

consumption does not always increase TAC 213, and therefore it is not unexpected that restriction

of these foods would not affect TAC. Further, it is possible that to compensate for their lack of

lycopene-containing foods, participants substituted their diet with other fruits and vegetables

high in antioxidants. This would nullify any possible effect of lycopene depletion on TAC, and is

an idea which requires further exploration in future studies on lycopene restriction.

The apparent changes seen on biomarkers of oxidative stress, antioxidant enzymes and

bone resorption may not be solely due to a lack of lycopene. While lycopene accounts for as
93
much as 90% of the carotenoids found in tomatoes and tomato products , these foods do
89,
contain modest concentrations of other carotenoids, such as -carotene and lutein/zeaxanthin
191
(Table 4.3). Supplementation with tomato products has been shown to significantly increase
113, 192
not only lycopene but other carotenoids , and these carotenoids have been shown to work

in synergy to exert beneficial effects 214. Significant decreases were shown in not only lycopene,

but lutein/zeaxanthin and - and -carotene, which is consistent with other studies on carotenoid
106
depletion . Although the percent change in serum lycopene in these participants was

significantly higher than the change in the other carotenoids after depletion (p<0.0001), the

possibility that there is an effect of restriction of other carotenoids present in tomatoes and

tomato products on the parameters studied cannot be completely discounted.

91
Table 4.3: Carotenoid content of most commonly consumed lycopene-containing foods,
demonstrating the amount of individual carotenoids present in g/g according to the US
Department of Agriculture 89.

Carotenoid Concentration (g/g)

Tomato, raw Tomato Sauce Tomato Juice
-carotene 1.0 0.0 0.0
-carotene 4.5 2.6 2.6
-cryptoxanthin 00 0.03 0.0
Lutein/zeaxanthin 1.2 0.2 0.6
Lycopene 25.7 139.8 90.4

The participants in this study had apparent increases in lipid and protein oxidation

parameters after 30 days of lycopene restriction, but this effect was not significant. This is

consistent with other published findings in which depletion for 23 days did not significantly

increase lipid peroxidation. Significant increases in lipid peroxidation have been reported after

longer periods of carotenoid depletion, up to 120 days 202, 203, 215, suggesting that a longer period

of depletion may have magnified the reported effect of lycopene restriction on lipid and protein

oxidation. However, in addition to the apparent effects seen on lipid and protein oxidation, there

were also significant increases in the bone resorption marker NTx (p<0.05), corresponding to an

increase of 20.60 ± 9.81%. This significant increase in NTx may lead to a long-term decrease in

BMD and increased fracture risk 2, suggesting that a longer restriction period may be detrimental

to bone health, particularly in this group of postmenopausal women who were already at higher

risk for osteoporosis 1. It is for this reason that a longer period of lycopene depletion for the

purpose of clinical intervention trials, particularly in postmenopausal women, should be avoided.

It was to be expected that antioxidant enzyme concentrations would decrease after

lycopene restriction, since it has been suggested that consumption of antioxidants, particularly
64, 199, 202, 216
lycopene, may increase the activities of these endogenous enzymes . It has been

reported that 68 days of a carotenoid depleted diet resulted in significantly lower SOD activity

92
202
. The decrease in SOD and CAT after restriction of lycopene in this study may be explained by

the fact that since tomatoes and tomato products contain copper, manganese, selenium and zinc,

which are cofactors of antioxidant enzymes 192, these essential minerals were also lacking during

the lycopene-restriction diet. During states of oxidative stress, the endogenous and dietary

antioxidants work in conjunction to quench free radicals. When dietary antioxidants are high,

they may act as electron donors and direct scavengers of superoxide radicals, thus preserving the

concentrations of antioxidant enzymes in the system 35, 202. However, when dietary antioxidants

are low, as demonstrated in the present study with the lycopene restricted diet, the antioxidant
35
enzymes may act to decrease oxidative stress as a compensatory mechanism , and are

decremented in the process. This explains the depressed activities of CAT and SOD shown in

these participants. However, these findings show that while lycopene restriction did significantly

decrease SOD and CAT activity, GPx activity significantly increased as a result of lycopene

restriction (Figures 4.2 and 4.3). A similar finding on GPx is reported in a study on -carotene
204
depletion in premenopausal women . All three enzymes are involved in the conversion of the

superoxide anion radical into water and oxygen. However, GPx is also involved in the reduction
162
of lipid hydroperoxides, a by-product of lipid peroxidation, into non-detrimental alcohols .

This role of GPx in reducing lipid peroxidation, suggests that the increased GPx shown after

lycopene restriction may be due to the concomitant increase in lipid peroxidation. In fact, there

was a significant, positive, linear relationship found between change in GPx activity and TBARS

(p<0.05), indicating that the increase in GPx was a mechanism to protect against the increasing

lipid peroxidation which occurs after lycopene restriction. A positive association between

concentration of TBARS and GPx activity has previously been reported in a large clinical study

of 903 women (p<0.001) 39. The compensatory mechanisms of CAT, SOD and GPx associated

with a lycopene-deprived diet may be the reason why the increase in oxidative stress was not

93
significant. It is possible that these antioxidant enzymes acted in place of lycopene, moderating

the resultant increases in oxidative stress biomarkers, resulting in non-significant increases of

protein oxidation and lipid peroxidation of 5.45 ± 3.29% and 14.50 ± 7.08% only.

The implications of the present findings are important and may assist in establishing

guidelines for future clinical trials in which a period of lycopene restriction is required, by

yielding information on the effects of restriction, particularly on bone health. In addition,

89
lycopene is present in a select number of foods ; therefore not consuming these products as a

part of the regular daily diet may result in negative health consequences in women, particularly

with respect to bone health. Although at present lycopene is not considered to be an essential

nutrient, and as such no formal daily intake levels are specified, based on human intervention

studies an intake of 8-10 mg per day is recommended by health professionals. Results from this

study also provide further proof of the importance of consuming tomatoes and tomato products,

as a source of lycopene in the daily diet, to maintain overall health and decrease the risk for age-

related chronic diseases, particularly osteoporosis, which is associated with oxidative stress.

94
 CHAPTER 5

Supplementation with the antioxidant lycopene significantly

decreased oxidative stress parameters and the bone
resorption marker crosslinked N-telopeptide of type I
collagen in postmenopausal women

This chapter has been submitted for publication to Osteoporosis International as a research paper
entitled: “Supplementation with the antioxidant lycopene significantly decreases oxidative stress
parameters and the bone resorption marker N-telopeptide of type I collagen in postmenopausal
women” by E.S. Mackinnon, A.V. Rao, R.G. Josse, and L.G. Rao
95
 ABSTRACT

We have previously shown that in vitro and in vivo lycopene is associated with a protective

effect on bone. However, intervention studies directly investigating the ability of lycopene to

decrease osteoporosis risk have not been reported. Sixty postmenopausal women, 50-60 years

old, were recruited for a randomized controlled intervention study to determine whether

lycopene would decrease osteoporosis risk. Following a one-month washout without lycopene

consumption, participants consumed either (N=15/group): (1) regular tomato juice, (2) lycopene-

rich tomato juice, (3) tomato lycopene capsules or (4) placebo capsules, twice daily for total

lycopene intakes of 30, 70, 30 and 0 mg/day respectively for 4 months. Serum collected after the

washout, 2 and 4 months of supplementation, was assayed for carotenoid content, BAP, NTx,

TAC, CAT, SOD, GPx, lipid and protein oxidation. Repeated-measures ANOVA showed that

lycopene-supplementation for 4 months significantly increased serum lycopene compared to

placebo (p<0.001). This increase was similar with all three supplements; therefore, participants

were pooled into a “LYCOPENE-supplemented” group for further statistical analyses.

LYCOPENE-supplementation for 4 months resulted in significantly increased TAC (p<0.05),

and decreased lipid peroxidation (p<0.001), protein oxidation (p<0.001) and NTx (p<0.001) and

these changes were significantly different from placebo consumption (not significant, p=0.06,

p<0.05, p<0.005 and p<0.02, respectively). LYCOPENE supplementation did not significantly

change BAP, CAT, SOD or GPx. These findings suggest that lycopene in tomato juice or capsule

form may be beneficial in reducing the risk of osteoporosis in postmenopausal women owing to

its potent antioxidant properties.

96
 INTRODUCTION

Osteoporosis is a systemic disease characterized by low bone mass and deterioration of the

microarchitecture of bone, resulting in an increased risk of fracture. Osteoporotic fractures are

associated with chronic pain and often immobility, which greatly impacts the quality of life and

ability to perform daily activities. The probability that a Caucasian women over the age of 50

will fracture a hip is 14%, and there is an excess 20% mortality one year after a hip fracture 1.

Early postmenopausal women lose bone at an accelerated rate. Nutritional recommendations to

offset this loss of bone include a daily intake of 1500 milligrams of calcium and 800 IU of

vitamin D 2. Although other important micronutrients such as strontium, magnesium, potassium

and vitamin K are emerging in the literature as components shown to be beneficial to bone 29, 30,

further assessment of other nutritional factors which influence bone turnover are required which

may suggest a benefit for osteoporosis prevention 31.

Studies show that oxidative stress is associated with the development of osteoporosis as it

causes a loss of bone. An inverse correlation between oxidative stress biomarkers and BMD has
47
been shown and in patients with fractures, oxidative stress parameters are higher than those
48
found in healthy controls . Antioxidants capable of counteracting this effect, by quenching

ROS, have been demonstrated to be important in decreasing the risk of osteoporosis 65.
117
Lycopene is a potent carotenoid with the highest singlet oxygen quenching ability .

Because of its potent ability to decrease oxidative stress, it has been associated with a decreased

risk of chronic diseases (for review, see 44). Recently, we have shown that a high serum lycopene

is associated with decreased protein oxidation parameters and bone resorption markers in
132
postmenopausal women , suggesting a potential benefit of lycopene in reducing the risk of

osteoporosis 73, 78, 103. At the cellular level, lycopene has been shown to stimulate the growth and

differentiation of osteoblasts 77, 130 and inhibit the formation and resorption activity of osteoclasts

97
128
. However, to date, there have been no clinical intervention studies demonstrating a direct

association between lycopene supplementation and decreased risk for osteoporosis. This chapter

reports findings of an intervention study in which postmenopausal women were given different

lycopene supplements to determine whether these supplements could decrease oxidative stress

parameters and bone turnover markers, thus decreasing the risk of osteoporosis in

postmenopausal women.

METHODS

Participant Recruitment and Blood Sample Collection

This protocol was approved by the Research Ethics Board at St. Michael's Hospital, Toronto,

Ontario, Canada (approved August 2, 2001) and followed the guidelines of good clinical

practices in accordance with the Declaration of Helsinki. Participant recruitment was conducted

from 2003-2007. Please refer to Appendix I for the official informed consent package distributed

to participants. Female participants between 50-60 years old, who were at least one year

postmenopausal, were recruited by telephone and advertisements. Any participants who smoked

cigarettes or were on medications which may affect bone metabolism, including those for heart

disease, high blood pressure, diabetes and/or osteoporosis were excluded from participating.

Participants were asked to refrain from consuming vitamins containing antioxidants, nutritional

supplements or foods containing lycopene for the duration of the study.

After obtaining informed consent, participants were randomly assigned to one of the four

following lycopene supplement groups, to be taken twice daily: (1) 15 mg lycopene in the form

of regular tomato juice (30 mg/day), (Heinz, Canada) (2) 35 mg lycopene in the form of

lycopene-rich tomato juice (70 mg/day), (Kagome, Japan), (3) 15 mg lycopene in the form of

tomato lycopene (Lyc-O-Mato®) capsules (30 mg/day) (LycoRed Ltd., Israel) and (4) zero
98
lycopene in the form of placebo capsules (0 mg/day) (LycoRed Ltd., Israel). Following a one-

month washout period during which no lycopene-containing foods were consumed, a baseline

12-hour fasting blood sample was collected. Height, weight and blood pressure were also taken

at this time. Participants were then instructed to begin taking their supplement, twice daily with

meals, while continuing to avoid all other lycopene-containing foods in their daily diet.

Additional blood samples were collected after 2 and 4 months of supplementation. Participants

submitted their 7-day dietary records at the time of each blood sample collection, giving details

of foods, beverages and nutritional supplements consumed over the previous 7 days. Compliance

was determined using supplement counting at the final study meeting.

Supplement Analysis

Total lycopene in milligrams for each of the juices was provided by Kagome, Japan and Heinz,

Canada. To determine the different isomers present in the juice as well as other carotenoids
217
HPLC was performed as previously described , with modifications (Z. Liu, unpublished

methods). HPLC was carried out using the Waters 2690 Alliance HPLC System and a Waters

996 PDA detector (Milford, MA, USA). The analytical column used was a Waters Spherisorb 3

M ODS2, 4.6 x 250 mm. Separation was performed by elution with a solvent containing

acetonitrile and methanol (65:35, v/v) and 0.065% triethylamine, at a flow rate of 1.5 mL per

minute. The carotenoids lycopene, lutein, -cryptoxanthin, zeaxanthin and -/-carotene, were

analyzed at a wavelength of 450 nm. Lycopene was detected as all-trans, 5-cis and other-cis

isomers. Other-cis isomers consist of two undetermined lycopene isomers. All carotenoids were

determined using external standard calibration curves on the Waters Millennium data

management software, 4.0 Edition (Milford, MA, USA) (for example chromatogram, please refer

to Appendix III, Figure IIIa). Carotenoid content of the capsules, including lycopene isomers,

was supplied by LycoRed Ltd., Israel.

99
Serum Analyses

Unless otherwise specified, all materials were obtained from Sigma Aldrich Canada, Oakville,

ON, Canada. For specific details on the following methodology, please refer to Chapter 4. The

concentrations of serum lycopene, lutein, -cryptoxanthin, zeaxanthin and -/-carotene were

measured using HPLC, according to previously published methods 178, with minor modifications.

Lycopene was detected as all-trans, 5-cis and other-cis isomers. TAC was measured using the
206
TEAC method . Protein oxidation was determined by estimating protein-sulfhydryl groups
207
(thiols) in serum . A high concentration of protein thiols corresponds to a lower protein

oxidation. Lipid peroxidation was measured in the serum using the TBA-MDA assay and was
208 209
reported as TBARS . CAT was measured using the H2O2 decomposition method and its

activity was expressed as K/g Hb, where K = 2.3/(Δ time)log OD15 seconds/OD 30seconds. SOD was

determined by the auto-oxidation of epinephrine 210 and was expressed as U/mg Hb. GPx activity

was measured by the decrease in NADPH absorbance at 340 nm at 37ºC in the presence of H2O2
211
and was expressed as U/g Hb. Commercially purchased ELISA kits were used to measure the

serum concentrations of the bone turnover markers NTx, a measure of bone resorption (INTER

MEDICO, Ontario) and BAP, a marker of bone formation (ESBE Scientific, Ontario). NTx was

expressed as nanomoles of bone collagen equivalents per litre (nM BCE) and BAP was

expressed as U/L, where one Unit represented 1 mol of p-nitrophenyl phosphate hydrolyzed per

minute at 25C in 2-amino-2-methyl-1-propanol buffer.

100
Statistical Analyses

All statistical analyses were performed using GraphPad PRISM 5.02 for Windows (GraphPad

Software, California). Summary statistics of participant demographics such as age and BMI were

generated for each group and presented as means ± standard errors of the mean (SEM). Baseline

characteristics of participants were analyzed with one-way ANOVA and the Tukey multiple

comparison test to determine whether there were any significant differences. The area under the

curve (AUC) was plotted using serum lycopene concentration over time and analysis was

performed using the trapezoid rule, without baseline subtraction, as a representation of lycopene

absorption. Data were expressed as nMmonth. Throughout the text, the participants who

consumed either regular tomato juice, lycopene-rich tomato juice or tomato lycopene capsules

were pooled and referred to as the “LYCOPENE-supplemented” group for the following reasons:

(1) there were no significant differences among the three sources of lycopene in the baseline

characteristics of the participants, or in their effects on total serum lycopene and (2) repeated-

measures two-way ANOVA to test for significant differences between groups over time and

sources of lycopene supplement with respect to bone turnover markers, oxidative stress

parameters and antioxidant status, was performed and the results showed that there were no

significant differences in effect on these parameters by the different sources of lycopene.

Repeated-measures one-way ANOVA, with Tukey’s multiple comparison test, was then

performed to determine the overall effect of supplementation with LYCOPENE or placebo

capsules over time of treatment on the above clinical end-point markers and parameters. In cases

where data were not normally distributed, the Friedman test with Dunn’s multiple comparison

test was substituted. Percent change after 2 and 4 months was also calculated and compared to

the change in participants supplemented with placebo using an unpaired t-test, or the Mann-

101
Whitney test for data that were not normally distributed. Any other statistical analyses performed

are specified in the text. Significance was considered at p<0.05.

RESULTS

A total of 60 participants, 15 per supplement group, completed this study. Participant

characteristics are described in Table 5.1. There were no significant differences among groups

with respect to age, BMI, years since menopause, or blood pressure (Table 5.1). Similarly, there

were no significant differences at baseline for average serum lycopene, bone turnover markers,

oxidative stress parameters, antioxidant enzymes, or TEAC (Table 5.1).

To determine the different isomers present in the juice as well as the concentrations of

other carotenoids, HPLC analyses were carried out and details of the results are given in Table

5.2. To determine the absorption of the lycopene provided in each of the supplements, the AUC

was examined, which is generally used to represent the absorption of lycopene over the
212
supplemental period . Despite having a lower concentration of cis isomers as shown in Table

5.2, the tomato lycopene capsules resulted in a higher absorption of cis isomers than the regular

tomato juice (p<0.05, Table 5.3). However, overall, the total lycopene absorption for the 4 month

study duration (all-trans + total cis) was not significantly different among the three sources of

lycopene (regular tomato juice, AUC = 5543 ± 500.7 nMmonth, lycopene-rich tomato juice,

AUC = 6983 ± 500.7 nMmonth, tomato lycopene capsules, AUC = 6943 ± 420.7 nMmonth).

On the other hand, all three sources of lycopene resulted in significantly higher lycopene

absorption (AUC) than placebo supplements (all: p<0.001, Table 5.3).

The total serum lycopene significantly increased after 2 and 4 months for the

LYCOPENE-supplemented group (repeated-measures ANOVA, p<0.0001, Table 5.4), an effect

102
not shown in participants consuming placebo. Serum lycopene reached a maximum level after 2

months of LYCOPENE supplementation (1915 ±113.1 nM), which was maintained up to the end

of the 4-month supplementation period (2012 ± 88.56 nM) (Table 5.4). At the end of the

supplementation period, participants who had consumed LYCOPENE had a serum lycopene

concentration that was more than three times higher than that of participants who had consumed

placebo capsules (p<0.0001).

Serum -carotene also significantly increased over time for the LYCOPENE–

supplemented group (p<0.001 after 2 and 4 months; refer to Table 5.4 for change in serum -

carotene). Small, insignificant increases were shown in serum lutein for participants consuming

both types of tomato juice (data not shown). Although there was a significant increase in serum

-carotene in the LYCOPENE-supplemented group, a t-test showed that over the 4 month

supplementation period, the average increase in serum lycopene (345.4 ± 46.8%) was more than

three times higher than the average increase in -carotene (104.0 ± 15.67% ) (p<0.0001).

LYCOPENE-supplementation significantly increased TAC from the baseline

concentration of 1.56 ± 0.05 mM to a concentration of 1.66 ± 0.05 mM after 4 months (p<0.05,

Figure 5.1). Consumption of placebo capsules did not have an effect on TAC; after 4 months of

supplementation the concentration was 1.59 ± 0.08 mM and not significantly different from the

baseline value of 1.56 ± 0.09 mM (Figure 5.1). In fact, placebo supplementation for four months

resulted in a decrease of 3.22 ± 4.82% in TAC, compared to the increase of 8.56 ± 3.04% which

was shown in the LYCOPENE-supplemented participants (not significant, p=0.06, Figure 5.2).

The LYCOPENE-supplemented group showed a significant decrease in protein

(repeated-measures ANOVA, p<0.0001) and lipid oxidation (repeated-measures ANOVA,

p<0.0001). After 4 months, the LYCOPENE–supplemented group had a 15.56 ± 3.75%

significant increase in protein thiols from the baseline concentration of 405.6 ± 14.69 M to

103
455.2 ± 15.50 M (p<0.001, Figure 5.3), indicating a decreased protein oxidation. This was

significantly opposite to the 5.09 ± 3.19% decrease in protein thiols seen in placebo-

supplemented participants (p<0.005, Figure 5.5). Decreased lipid peroxidation after LYCOPENE

supplementation was evident as shown by an 11.93 ± 2.16% significant decrease in TBARS from

the baseline concentration of 7.91 ± 0.46 nmol/ml to 6.80 ± 0.35 nmol (p<0.001, Figure 5.4),

which was significantly greater than the negligible 0.37 ± 5.25% decrease seen in placebo-

supplemented participants (p<0.05, Figure 5.5). There was no significant effect of lycopene

supplementation on CAT, SOD, or GPx (data not shown).

The LYCOPENE-supplemented group had significantly decreased NTx over time

(repeated-measures ANOVA, p<0.0001). After 2 months of LYCOPENE supplementation, NTx

decreased from the baseline concentration of 24.21 ± 1.14 nM BCE to 21.22 ± 0.92 nM BCE

(p<0.01, Figure 5.6) or 8.35 ± 3.36% (Figure 5.7), and further to 19.62 ± 0.88 nM BCE

(p<0.001, Figure 5.6) or 15.09 ± 3.49% after 4 months (Figure 5.7). There was no significant

change in participants consuming placebo from baseline (20.25 ± 1.76 nM BCE) to after the four

month supplement period (20.78 ± 2.35 nM BCE) (Figure 5.6). On the other hand, consumption

of placebo capsules resulted in the significantly opposite effect of a 9.98 ± 8.45% increase in

NTx after 2 months (p<0.02), which was maintained throughout the duration of the study

(increase of 9.17 ± 13.28% after 4 months, p<0.02 relative to the decrease in NTx of

LYCOPENE-supplemented participants, Figure 5.7). LYCOPENE supplementation did not

result in any changes in the bone formation marker BAP (data not shown).

104
Table 5.1: Participant characteristics and baseline values for oxidative stress parameters, bone
turnover markers and antioxidant capacity for each supplement group.

Parameter Measured 1 Summary statistics for supplement group (mean ± SE)

Regular Lycopene-rich Tomato Placebo
tomato tomato juice lycopene capsules
juice (N=15) (N=15) capsules (N=15)
(N=15)
Age (yrs) 55.20 ± 0.84 56.13 ± 0.64 54.33 ± 0.74 54.90 ±
0.72
Years since menopause 6.08 ± 1.10 6.89 ± 1.75 3.36 ± 0.65 3.73 ± 0.62
Blood pressure-diastolic (mm 72.09 ± 2.10 79.0 ± 2.93 74.40 ± 2.45 76.38 ±
Hg) 4.15
Blood pressure-systolic (mm 110.6 ± 3.72 128.0 ± 4.51 116.2 ± 4.75 103.2 ±
Hg) 9.79
BMI (kg/m2) 24.37 ±1.43 26.36 ± 1.50 26.49 ± 1.40 26.19 ±
0.89
Pulse rate (beats/min) 68.0 ± 2.99 78.2 ± 3.69 69.10 ± 2.26 76.62 ±
4.49
Total serum lycopene (nM) 602.2 ± 766.2 ± 106.8 574.5 ± 101.8 579.9 ±
93.9 68.7
Total antioxidant capacity 1.56 ± 0.09 1.41 ± 0.08 1.72 ± 0.07 1.56 ± 0.09
(TEAC) (mM)
Antioxidant CAT (K/g 60.35 ± 2.30 61.03 ± 4.15 51.70 ± 2.96 56.60 ±
enzymes Hb) 3.16
GPx (U/g 29.12 ± 5.16 22.66 ± 3.41 26.24 ± 4.9 25.55 ±
Hb) 4.33
SOD (U/mg 31.52 ± 4.44 32.28 ± 3.67 32.03 ± 5.0 30.84 ±
Hb) 4.21
Bone turnover BAP (U/L) 22.29 ± 2.33 25.08 ± 1.98 23.11 ± 1.59 24.03 ±
markers 2.36
NTx (nM 25.33 ± 2.16 22.64 ± 1.70 24.65 ± 2.12 20.25 ±
BCE) 1.76
Oxidative Protein 408.8 ± 425.4 ± 24.69 382.6 ± 28.00 426.6 ±
stress thiols (M) 23.95 23.75
parameters TBARS 7.18 ± 0.56 7.02 ± 0.51 8.85 ± 0.87 7.13 ± 0.67
(nmol/mL)
1
No significant differences between groups for any of the measured baseline parameters.

105
Table 5.2: Carotenoid content of the supplements given to participants, showing quantities in
milligrams per day, as provided by either 2 cans of tomato juice or 2 tomato lycopene capsules.
Carotenoid Milligrams per day (Mean ± SEM)
Regular tomato Lycopene-rich Tomato lycopene
1 1
juice tomato juice capsules2
(2 cans per day) (2 cans per day) (2 capsules per day)
Total lycopene 25.50 ± 4.89 68.03 ± 5.81 30.00 ± 0.0
all-trans lycopene 20.76 ± 3.98 53.83 ± 8.78 28.80 ± 0.0
a b
Total cis lycopene 4.74 ± 0.91 14.20 ± 3.44 1.20 ± 0.0
-carotene 1.21 ± 0.25 3.19 ± 0.82 1.00 ± 0.0
Lutein 0.30 ± 0.07 0.76 ± 0.31 not specified 2
1
As determined by HPLC outlined in methods section
2
Based on information provided by supplier
a, b
Significantly lower total cis than lycopene rich tomato juice (a p<0.05, b p<0.02)

Table 5.3: Mean AUC as a measure of lycopene absorption, for serum concentration of lycopene
and lycopene isomers versus time for all three sources of lycopene and placebo supplements.

Supplement group Mean ± SEM of AUC, serum concentration vs. time (nMmonth)
all-trans 5-cis other-cis Total
lycopene lycopene lycopene lycopene
Regular tomato juice 2731 ± 258.7 1821 ± 991.1 ± 88.96a 5543 ± 500.7
165.4a
Lycopene-rich tomato 3473 ± 250.2 2314 ± 175.4 1196 ± 95.58 6983 ± 500.7
juice
Tomato lycopene 3188 ± 200.2 2494 ± 161.9 1260 ± 75.81 6943 ± 420.7
capsules
Placebo capsules1 923.3 ± 109.8 702.2 ± 71.61 416.8 ± 40.72 2049 ± 214.5
1
Placebo values had significantly lower AUC than all three lycopene supplement groups (Tukey, p<0.001)
a
Significantly lower than AUC for tomato lycopene capsules (Tukey, p<0.05)

106
Table 5.4: Mean serum carotenoid concentrations for all participants for the duration of study,
showing changes in -carotene and individual lycopene isomers as well as total cis and total
lycopene.

Supplement Time Serum carotenoids (nM)

group (months) mean ± SEM
all-trans 5-cis other-cis Total cis Total -
lycopene lycopene lycopene lycopene lycopene carotene
LYCOPENE1 0 288.3 ± 226.6 ± 132.7 ± 359.3 ± 647.6 ± 1073 ±
28.87 20.11 11.29 30.57 58.38 117.4
2a 931.8 ± 649.1 ± 334.3 ± 983.4 ± 1915 ± 1633 ±
56.53 42.19 19.72 59.38 113.1 150.2
4a 979.0 ± 685.0 ± 347.7 ± 1033 ± 2012 ± 1751 ±
48.48 30.22 17.36 45.01 88.56 181.9
Placebo 0 263.9 ± 198.9 ± 117.2 ± 316.0 ± 579.9 ± 950.9 ±
capsules 37.2 21.8 12.5 32.6 68.7 145.9
2 266.4 ± 206.1 ± 119.8 ± 325.8 ± 592.3 ± 929.6 ±
42.9 31.4 18.2 47.8 89.6 143.5
4 280.7 ± 200.3 ± 124.1 ± 324.4 ± 605.0 ± 1020 ±
77.1 52.8 27.3 77.7 154.2 159.1
1
Lycopene supplements provided as regular tomato juice, lycopene-rich tomato juice or tomato lycopene capsules
a
LYCOPENE-supplemented group had significantly increased serum carotenoids after 2 (Tukey’s/Dunn’s multiple
comparison test, p<0.001) and 4 months (Tukey’s/Dunn’s multiple comparison test, p<0.001).

2.0
*
1.5
Trolox (mM)

1.0

0.5

0.0
0 2 4 0 2 4
LYCOPENE-supplemented Placebo-supplemented
Time period for intervention (mos.)

Figure 5.1: Total antioxidant capacity in the serum, as determined using the TEAC assay, of 60
postmenopausal female participants supplemented with LYCOPENE (N=45) or placebo capsules
(N=15) for a period of 4 months. Values are mean  SEM and were compared within supplement
group using repeated-measures ANOVA (*p<0.05).

107
 15
LYCOPENE-supplemented
Change in Trolox (%) 10 Placebo-supplemented

-5

-10
2 months 4 months
Supplement group

Figure 5.2: Change in total antioxidant capacity over the 4 month supplement period. 60
postmenopausal women aged 50-60 were supplemented with LYCOPENE (N=45) or placebo
capsules (N=15) and the change in TAC was measured relative to baseline concentration. Values
are mean % change  SEM for each supplement group and were compared using an unpaired t
test.

500 *
Protein thiols ( M)

400

300

200

100

0
0 2 4 0 2 4
LYCOPENE-supplemented Placebo-supplemented
Time period for intervention (mos.)

Figure 5.3: Concentration of protein oxidation in the serum over time, as determined by protein-
sulfhydryl groups (thiols), of 60 postmenopausal female participants aged 50-60, supplemented
with LYCOPENE (N=45) or placebo capsules (N=15) for a period of 4 months Values are mean
 SEM and were compared within supplement group using repeated-measures ANOVA
(*p<0.001).

108
 10

8 *

TBARS (nmol/ml)
6

0
0 2 4 0 2 4
LYCOPENE-supplemented Placebo-supplemented
Time period for intervention (mos.)

Figure 5.4: Concentration of lipid peroxidation in the serum over time, as determined by
TBARS, of 60 postmenopausal female participants aged 50-60, supplemented with LYCOPENE
(N=45) or placebo capsules (N=15) for a period of 4 months. Values are mean  SEM and were
compared within supplement group using repeated-measures ANOVA (*p<0.001).

109
 30
LYCOPENE-supplemented
*

Change in protein
Placebo-supplemented
20

thiols (%) 10

-10

0
Change in TBARS (%)

-5

-10

-15 
2 months 4 months

Supplement group

Figure 5.5: Change in protein and lipid peroxidation over the 4 month supplement period. 60
postmenopausal women aged 50-60 were supplemented with LYCOPENE (N=45) or placebo
capsules (N=15) and the change in protein thiols and TBARS was measured relative to the
baseline concentration. Values are mean % change  SEM for each supplement group and were
compared using an unpaired t test (*p<0.005 and p<0.05).

110
 30

*
**
NTx (nM BCE)
20

10

0
0 2 4 0 2 4
LYCOPENE-supplemented Placebo-supplemented
Time period for intervention (mos.)

Figure 5.6: Concentration of the bone resorption marker NTx in the serum over time, of 60
postmenopausal female participants aged 50-60, supplemented with LYCOPENE (N=45) or
placebo capsules (N=15) for a period of 4 months. Values are mean  SEM and were compared
within supplement group using repeated-measures ANOVA (*p<0.01 and **p<0.001).

30
LYCOPENE-supplemented
Change in NTx (%)

Placebo-supplemented
20

10

-10
*
-20 2 months
*
4 months
Supplement group

Figure 5.7: Change in the bone resorption marker NTx over the 4 month supplement period. 60
postmenopausal women aged 50-60 were supplemented with LYCOPENE (N=45) or placebo
capsules (N=15) and the change in NTx was measured relative to the baseline concentration
Values are mean % change  SEM for each supplement group and were compared using an
unpaired t test (*p<0.02).

111
 DISCUSSION

The present intervention study showed that in 50-60 year old postmenopausal women, lycopene

intake from either tomato lycopene capsules or 2 types of tomato juice was directly and

significantly associated with total, all-trans, 5-cis and other-cis serum lycopene, with a

corresponding significant increase in TAC. Of particular interest, the data presented in this

chapter also showed that women consuming lycopene had significantly decreased concentrations

of the oxidative parameters lipid peroxidation and protein oxidation and the bone resorption

marker NTx over the supplement period. These results were significant compared to the

inconsequential results achieved after participants consumed placebo capsules. Trials on dietary

supplements are continuously being carried out and are critical to delineate the current
1, 29-31
understanding of factors which can maintain skeletal health . The findings reported in the

present study contribute to this knowledge and provide a basis for further trials of carotenoids,

particularly the role of lycopene in bone health. This is the first clinical intervention study to

report an association between lycopene supplementation and the potential decrease in the risk of

osteoporosis based on changes in the bone resorption marker NTx.

The finding that consumption of lycopene supplements had a significant and direct

correlation with total serum lycopene suggests that lycopene from these supplements is well

absorbed. The absorption data indicated that participants consuming tomato lycopene capsules

had significantly higher absorption of cis lycopene than participants consuming regular tomato

juice. Both the tomato lycopene capsules and the regular tomato juice had significantly less

quantity (expressed in mg) of total cis lycopene per daily serving than the lycopene-rich tomato

juice. Despite this difference in quantity (mg) of total cis, participants who consumed tomato

lycopene capsules had a significantly higher absorption of total cis lycopene compared to

participants who consumed regular tomato juice and similar absorption to that for the lycopene-

112
rich tomato juice. This finding may suggest that the matrix in which lycopene is provided affects
212
the absorption of cis isomers of lycopene and is consistent with previous findings . Several

studies suggest that lycopene capsules show enhanced bioavailability of cis lycopene compared

to tomato juice 212, 218.

The fact that consumption of all three forms of lycopene supplementation gave similar

AUC values suggests that there is a maximum absorption concentration of lycopene at which

saturation occurs. Similarly, a recent dose-response study on lycopene consumption showed that
115
serum lycopene significantly increased independently of dose . In a recent clinical study on

the pharmacokinetics of lycopene consumption, participants were given a single, low dose of 10

mg, or a higher dose ranging from 60-120 mg of lycopene. The findings showed that overall the

amount of lycopene absorbed by participants was the same at all doses, and regardless of the
181
dose, 80% of participants absorbed less than 6 mg of lycopene . The present study found a

similar saturation effect in that the absorption of lycopene was the same for participants

consuming either 30 or 70 mg of lycopene for a period of 4 months. However, the total serum

lycopene concentrations resulting from these doses (2012  88.56 nM) were 31.6% higher than

those resulting from an intake of 6.23 mg per day (1377  75.92 nM, Chapter 3), suggesting that
181
absorption saturation occurs at higher doses than previously reported . These findings also

demonstrated that serum lycopene did not increase further between 2 and 4 months of

supplementation, which is supported by other intervention studies in which peak lycopene

concentrations have been reported as early as 1-2 weeks after the beginning of supplementation
219
. However, it remains to be seen whether a dose-dependent absorption of lycopene would be

found if participants were given a lycopene supplement providing concentrations ranging from 0

to 30 mg per day.

113
 In addition to changes in serum lycopene, there were changes in -carotene as a result of

consumption of the three supplements provided. While lycopene accounts for as much as 90% of
93
the carotenoids found in tomatoes and tomato products , -carotene is also present, and the
199
possibility that it exerts positive effects of its own should be taken into consideration . Since

the change in serum lycopene was significantly higher after 4 months of supplementation than

the change in serum -carotene, and since lycopene is reported to be a more potent antioxidant
117
than -carotene , the present findings suggest that lycopene is more responsible than -

carotene for the effects of decreased oxidative stress parameters and bone resorption markers.

Some studies suggest a beneficial effect of -carotene on bone 76-79, although these data are still

controversial, most probably due to its association with vitamin A 31, 81. A Vitamin A intake of 

3000 µg RE per day was shown to significantly increase the risk of fracture in postmenopausal

women. This increased risk was attributed to the increased retinol associated with vitamin A

intake 81. -carotene was not associated with an increased risk of fracture, this is most likely due

to reduced absorption and conversion to vitamin A; it is estimated that only 8-16% of dietary -
81, 82
carotene is converted to retinol . Furthermore, while most clinical intervention studies
220-222
typically use 12-24 mg/day of -carotene, the supplements used in this study provided a

relatively lower dose; the daily serving of lycopene-rich tomato juice provides the same -

carotene as approximately 1/3 cup of raw carrots 89. However, carotenoids have been shown to
220
work in synergy , and therefore the possibility that both lycopene and -carotene are exerting

the positive antioxidant effects seen in this study cannot be completely ruled out.

The increase in TAC with lycopene supplementation may stem from the fact that

tomatoes contain other vitamins and antioxidants, such as polyphenols and carotenoids, which
191, 192
contribute to its overall antioxidant capacity and ability to combat oxidative stress .

114
Indeed, the TEAC method demonstrates the concentration of not only the carotenoids, but all

other types of antioxidants in vivo, both water- and lipid-soluble. The TEAC method provides a

measure of the in vivo balance between ROS and the antioxidants which act to quench them,
223
including those supplied in the diet . Thus, the lycopene supplements consumed by the

participants increased the serum concentrations of individual carotenoids (lycopene and -

carotene), resulting in an increased ability to effectively combat oxidative stress. This provides

further proof to the idea that it is the potent antioxidant capacity of lycopene which is responsible

for the decreased oxidative stress and NTx shown in this study.

The finding that the LYCOPENE-supplemented group had significantly decreased

protein oxidation (increased protein thiols) after 4 months is supported by other studies on

lycopene and oxidative stress 44, 198 and suggests that both capsules and juices are reliable sources

of lycopene when the goal is the reduction of protein oxidation. Cellular and animal studies
224, 225
showed that lycopene can reduce lipid peroxidation . Moreover, intervention studies have

shown that lycopene consumption is correlated with decreased lipid peroxidation 217, particularly

in women consuming greater than the population’s average daily intake of approximately 5
171
mg/day of lycopene . The present findings corroborate those previously reported, with a

statistically significant decrease in lipid peroxidation after 4 months of lycopene supplementation

with tomato juice or capsules at a concentration of 30 mg per day or higher.

Some studies suggest that supplementation with lycopene increases concentrations of the

endogenous antioxidant enzymes 199, 200, 226. Tomatoes contain copper, manganese, selenium and
192
zinc, which are cofactors of the antioxidant enzymes, GPx and SOD . Consumption of

lycopene has been shown to significantly increase activities of GPx and SOD in participants with

chronic diseases, such as hypertension. However, these diseases are typically associated with
199, 200, 226
depressed activities of antioxidant enzymes , thus in healthy individuals such as the

115
participants from this study, where activities are within the normal range, it is not surprising that

there is no effect of lycopene on the antioxidant enzymes examined. This is verified by the

majority of other studies, in which healthy individuals supplemented with tomato products did
227 192, 228
not have significant changes in CAT, SOD, or GPx activity . An animal study on

antioxidant enzymes after lycopene consumption showed that GPx and SOD were stimulated

only at lower doses of lycopene, while higher doses resulted in concentrations similar to baseline
229
values . The authors propose that this effect is due to an as of yet undetermined secondary

mechanism. A human intervention study, by Misra et al. 64, showed increased concentrations of

lycopene resulted in reduced glutathione (an endogenous antioxidant metabolite important to

199
GPx activity ) after 6 months of supplementation with the same brand of tomato lycopene

capsules used in this study, except at the lower dose of 4 mg per day compared to the 30 mg per
229
day given to our participants. If the speculated dose response effect by Breinholt et al. is

correct, it would provide further explanation of the null effect seen in our participants who

consumed relatively high doses of lycopene over a long period of time.

The present intervention study demonstrated an association between increased serum

lycopene and decreased NTx in postmenopausal women. Clinical trials on osteoporosis

treatments have frequently used bone turnover markers to measure the effects of treatments in
230
postmenopausal women because they can predict the risk of fracture in two ways: (I) a high
11
bone turnover is correlated with a low BMD and (II) an increased bone turnover negatively

affects bone fragility and microarchitecture 231. These findings showed that supplementation with

LYCOPENE resulted in a significant 8.35  3.36% decrease in NTx after 2 months, which

decreased further to a total of 15.09 ± 3.49% after 4 months. This is in keeping with another

dietary intervention study, in which supplementation with phylloquinone (Vitamin K) for

approximately 2 months resulted in a significant 9% decrease in NTx in postmenopausal women

116
232
. More importantly, these changes in bone resorption markers are comparable to those seen in
233, 234
postmenopausal women supplemented with calcium and vitamin D , which are currently

recommended for the prevention of osteoporosis 2. A study in which 40 postmenopausal women

consumed a milk beverage containing 1200 mg of calcium and 5.7 g of vitamin D daily for a

period of 6 months showed significant reductions of approximately 24% and 17% in Pyr and D-
234
Pyr, respectively . Similarly, supplementation with 596 mg/day of calcium significantly

decreased serum CTx by 13% after 6 months in postmenopausal women 233. However, although

most participants in this study had a marked decrease in NTx, 24.4% of participants were

considered to be “non-responders”, in that their change in NTx was -1.0%. This could be due to

SNPs which affect the response to lycopene; an idea which will be further addressed in Chapters

7 and 8.

There was no significant change found in BAP, a frequently used marker of bone

formation, in these participants. Although serum BAP gives an accurate description of changes

in bone turnover, particularly in postmenopausal osteoporosis 235, it is evident from other studies

on osteoporosis medications that changes in BAP tend to be smaller in magnitude compared to

236
those seen in NTx, and can take up to 6 months to be detected . This intervention study was

carried out for only 4 months, and perhaps a longer period of lycopene intervention would have

resulted in significant changes in bone formation markers. Results from work on osteoblast cells

to determine whether lycopene is capable of preventing and repairing the damaging effects of

ROS on the formation of mineralized bone nodules in vitro are presented in Chapter 9.

Nevertheless, in the present study the effect on NTx is still biologically important regardless of

the lack of effect on BAP. Many current therapies for osteoporosis target the imbalance in bone

turnover markers by attempting to decrease bone resorption 2. Studies with the anti-resorptive

bisphosphonate, alendronate, showed that a decrease in NTx was associated not only with long-

117
term increases in BMD but also with restoring the balance between bone turnover markers that is
237, 238
necessary for bone health . More specifically, a 15% decrease in serum NTx within the

first 6 months of therapy was shown to predict significant increases in BMD at the total hip,
239
greater trochanter, intertrochanteric region and spine after 2.5 years of therapy (p<0.05) . In

this study, a similar anti-resorptive effect of lycopene supplementation for four months resulted

in an average decrease in NTx of 15.09 ± 3.49%. Although the short duration of this 4-month

study did not allow us to assess the BMD of our participants, the present results on NTx suggest

that long-term supplementation with lycopene may result in increased BMD. Based on these

findings, the results support the hypothesis that lycopene may decrease the risk of osteoporosis

as evidenced by the reduction of the bone resorption marker NTx.

In summary, the results of the present study showed both time and treatment effects of

lycopene supplementation by demonstrating a significant increase in serum lycopene after

supplementation with tomato juice or capsules, which resulted in a decrease in the bone

resorption marker NTx in postmenopausal women. This reduction in NTx may be due to the

ability of the absorbed lycopene to reduce the oxidative stress parameters in these women. These

findings are the first to show that lycopene intervention, given in capsule or juice form,

supplying at least 30 mg/day, may decrease the risk of osteoporosis by decreasing biomarkers of

oxidative stress and bone resorption. Therefore, this novel study suggests that lycopene may be

used as a natural complementary or alternative supplement for the prevention and treatment of

osteoporosis in postmenopausal women.

118
 CHAPTER 6

Supplementation with lycopene in postmenopausal women

resulted in serum concentrations of 5-cis lycopene which
were higher than usually obtained through the daily diet, and
were associated with lower biomarkers of lipid peroxidation
and bone resorption

119
 ABSTRACT

We have previously shown that a high serum lycopene corresponds to a lower bone resorption
132
marker and protein oxidation in postmenopausal women . In addition, this thesis has shown

that supplementation with high doses of lycopene (30 or 70 mg/day) directly results in decreased

oxidative stress parameters and the bone resorption marker, crosslinked N-telopeptide of type I

collagen (NTx). However, to date, no information exists about exactly how much lycopene is

necessary to decrease bone turnover markers in postmenopausal women. In the present study,

data from postmenopausal female participants who were supplemented with lycopene (N=45,

43.33 ± 2.84 mg/day) was compared to data from participants who obtained both a low (N=48,

2.59 ± 0.32 mg/day) and high daily intake (N=27, 11.83 ± 0.89 mg/day) of lycopene to

determine whether the elevated dose obtained through supplementation was more beneficial in

reducing bone turnover markers than intakes typically obtained from the usual daily diet. Blood

samples were assayed for NTx, BAP, carotenoid content, TAC, CAT, SOD, GPx and lipid

(TBARS) and protein oxidation (protein thiols). Participants supplemented with lycopene had

significantly lower TBARS and NTx than participants who obtained a low intake of 2.59 ± 0.32

mg/day (p<0.0001 and p<0.05, respectively) or high intake of 11.83 ± 0.89 mg/day (p<0.01 and

not significant, p=0.05, respectively) of lycopene through their usual daily diets. These

differences in NTx and TBARS may be attributed to a significantly higher concentration of

serum 5-cis in lycopene-supplemented participants (p<0.0001 compared to low usual daily intake

participants and p<0.05 compared to high usual daily intake participants). This suggests that it is

the 5-cis isomer, with the most potent antioxidant capacity which, at higher concentrations,

decreases bone turnover markers due to its ability to provide the greatest protection against

oxidative stress.

120
 INTRODUCTION
117
Lycopene is a carotenoid exhibiting the highest singlet oxygen quenching capacity , which

makes it a powerful antioxidant. In lycopene-containing foods, lycopene exists primarily in the

lower antioxidant capacity all-trans form. However, in cooked and processed tomatoes, such as

canned tomatoes and tomato soup, slightly higher concentrations of the cis isomeric form of

lycopene are present 85, 168. Isomerization of all-trans to cis lycopene also occurs during digestion

and may result from several factors including selective uptake and/or differential bioavailability
44, 111, 197
, resulting in a high concentration of cis isomers in human serum. The cis isomers of

lycopene are recognized as having the highest antioxidant capacity, with 5-cis lycopene having

the highest antioxidant capacity 44, 116.

Research shows that lycopene is capable of decreasing several lipid, protein and DNA
97, 99, 115, 118
oxidation markers and increasing markers of antioxidant capacity and endogenous
198-200
antioxidant enzymes . These antioxidant properties have been explored in many clinical

studies, resulting in evidence suggesting that lycopene may decrease the risk of infertility,

diabetes, dementia, cardiovascular disease and several types of cancer 44, 95. In a cross-sectional

study, we have previously shown that lycopene obtained through the daily diet decreased the
132
serum levels of the bone resorption marker NTx in postmenopausal women . This thesis

provided evidence that supplementation with lycopene for a period of 4 months significantly

increased antioxidant capacity and significantly decreased biomarkers of bone resorption and

lipid and protein oxidation (Chapter 5). Postmenopausal women in this intervention study were

supplemented daily with 30 or 70 mg of lycopene per day in the form of tomato juice or

capsules, and results showed that all three supplements resulted in similar beneficial effects. As

shown in Chapter 3, the intake of lycopene in postmenopausal women is quite variable, ranging

from a minimum intake of 0.0 mg/day to a maximum quantity of 21.70 mg/day, with a mean

121
intake of 6.23  5.45 mg/day (Table 3.4, Chapter 3). Overall, however, the doses of lycopene

given in our intervention study are much higher than the reported quantities obtained from the

average daily intake for women of this age group (mean daily intake ± SD = 6.23 ± 5.45 mg/day,

Chapter 3) 91.

Restriction of dietary lycopene for a period of one month resulted in increases in

oxidative stress parameters and detrimental increases in the bone resorption marker NTx

(Chapter 4). The effects of lycopene restriction and intervention in postmenopausal women

prompted further exploration into whether the reported effects of a high daily dose of lycopene,

as given in the intervention study (Chapter 5), were significantly better than the effects of

lycopene obtained through the daily diet, which typically provide a lower quantity. In the present

study, data from participants in the intervention study, in which bone turnover markers,

antioxidant capacity and oxidative stress parameters were significantly affected, was compared

to data from participants in our cross-sectional study of the same age who consumed their usual

lycopene intake (Chapter 3). The findings reported in this chapter provide further evidence that

supplementation with lycopene may be important in postmenopausal women who are at risk for

osteoporosis.

METHODS

Participant Recruitment and Study Design

The study protocol was approved by the Research Ethics Board at St. Michael's Hospital,

Toronto, Ontario, Canada and followed the guidelines of good clinical practices. Female

participants were between 50-60 years old and at least one year postmenopausal. Any

122
participants who were on medications for heart disease, high blood pressure, diabetes and/or

osteoporosis were excluded.

Volunteers were recruited to participate in one or both of the following: (Study Design 1)

Cross-sectional study in which participants maintained their usual diet, recorded their diet for 7

days including all foods, drinks and vitamin or mineral supplements, and provided a fasting

blood sample on the 8th day, and (Study Design 2) Intervention study in which participants

refrained from consuming lycopene-containing foods other than the provided supplements of

either regular tomato juice (30 mg of lycopene/day), lycopene-rich tomato juice (70 mg of

lycopene/day), or tomato lycopene capsules (30 mg of lycopene/day). Blood samples collected

after 4 months of lycopene-supplementation, together with 7-day dietary records, were used for

the analyses listed below. 7-day dietary records were used to estimate usual daily intake in Study

Design 1 and data obtained from them can be extrapolated to assume nutrient intake over time.

Full study details can be found in Chapters 3 and 5, respectively.

Blood sample analyses

Unless otherwise specified, all materials were obtained from Sigma Aldrich Canada, Oakville,

ON, Canada. For specific details on the following methodology, please refer to Chapter 4. The

concentrations of serum lycopene, lutein, -cryptoxanthin, zeaxanthin and -/-carotene were

measured using HPLC, according to previously published methods 178, with minor modifications.

Lycopene was detected as all-trans, 5-cis and other-cis isomers. TAC was measured using the
206
TEAC method . Protein oxidation was determined by estimating protein-sulfhydryl groups
207
(thiols) in serum . A high concentration of protein thiols corresponds to a lower protein

oxidation. Lipid peroxidation was measured in the serum using the TBA-MDA assay and was
208 209
reported as TBARS . CAT was measured using the H2O2 decomposition method and its

activity was expressed as K/g Hb, where K = 2.3/(Δ time)log OD15 seconds/OD 30seconds. SOD was

123
determined by the auto-oxidation of epinephrine 210 and was expressed as U/mg Hb. GPx activity

was measured by the decrease in NADPH absorbance at 340 nm at 37ºC in the presence of H2O2
211
and was expressed as U/g Hb. Commercially purchased ELISA kits were used to measure the

serum concentrations of the bone turnover markers NTx, a measure of bone resorption (INTER

MEDICO, Ontario) and BAP, a marker of bone formation (ESBE Scientific, Ontario). NTx was

expressed as nanomoles of bone collagen equivalents per litre (nM BCE) and BAP was

expressed as U/L, where one Unit represented 1 mol of p-nitrophenyl phosphate hydrolyzed per

minute at 25C in 2-amino-2-methyl-1-propanol buffer.

Statistical Analyses

All statistical analyses were performed using GraphPad PRISM 5.00 for Windows (GraphPad

Software, California). In order to determine whether the effects of lycopene supplementation

were better than the effects resulting from average daily consumption, data on oxidative stress

parameters, antioxidant capacity and bone turnover markers measured from the serum of

participants supplemented with lycopene for four months (N=45) (as described in Chapter 5),

was compared to data from serum samples of women of the same age group who participated in

our cross-sectional study (N=75).

The group of participants from the cross-sectional study were referred to as the “usual

daily intake” group (N=75) (Study Design 1) and participants who were supplemented with

lycopene (Study Design 2) were referred to as the “lycopene-supplemented group” (N=45). The

“usual daily intake group” was further subdivided into two groups according to their average

daily intake of lycopene (mean intake 6.23 mg/day, Chapter 3). Participants who consumed an

average of  6.23 mg/day were considered as the “low usual daily intake group” (N=48) and

participants who consumed > 6.24 mg/day were considered as the “high usual daily intake

group” (N=27).
124
 Summary statistics of participant demographics such as age and BMI were generated and

presented as means ± standard errors of the mean (SEM). One-way ANOVA was used to test for

significant differences between groups with respect to these summary statistics. An unpaired t-

test was used to test for significant differences in bone turnover markers, oxidative stress

parameters and antioxidant status between each of the usual daily intake groups and the

lycopene-supplemented group. For data that were not normally distributed, the Grubb’s test for

outliers was used to exclude the offending outlier. If data were not normally distributed after this

test, then the Mann-Whitney test was used. Statistical significance was considered at p<0.05.

RESULTS

A total of 75 female volunteers were recruited to participate in the cross-sectional component of

this study (Study Design 1) and 45 participants were recruited to participate in the intervention

study (Study Design 2). Summary statistics for participant characteristics are shown in Table 6.1.

There was no significant difference between any of the groups for any of the parameters

measured, with the exception of vitamin C intake, which was significantly lower in the lycopene-

supplemented group compared to the low (p<0.05) and high usual daily intake groups (p<0.001).

In the lycopene-supplemented group intake of lycopene was significantly higher (43.33 ± 2.84

mg/day) than both the low (2.59 ± 0.32 mg/day, p<0.0001, Table 6.2) and high usual daily intake

groups (11.83 ± 0.89 mg/day, p<0.0001, Table 6.3). However, when comparing the

corresponding total serum lycopene concentrations, there was only a significant difference

between the low usual daily intake group (1094 ± 80.24 nM) and the lycopene-supplemented

group (2012 ± 88.56 nM) (p<0.0001, Table 6.2). This significant difference applied to each of

the lycopene isomers, including all-trans, 5-cis and other-cis lycopene (all: p<0.0001, Table 6.2).

125
 The lack of significant difference between the total serum lycopene of the high usual

daily intake group and the lycopene-supplemented group, suggests that there is a maximal

absorption of lycopene, as indicated in Chapter 5. Although not significant, there was an 11.5%

difference in total lycopene which may be due to differences in the concentrations of the

different lycopene isomers. Specifically, these results showed that the lycopene-supplemented

group had 16.3% higher 5-cis lycopene (685.0 ± 30.22 nM) than the high usual daily intake

group (573.5 ± 41.08 nM) (p<0.05, Table 6.3). The lycopene-supplemented group did not have

significantly higher serum concentrations of -cryptoxanthin, lutein, zeaxanthin, or -/-

carotene than either usual daily intake group (data not shown).

The lycopene-supplemented group had 29.2% lower TBARS (p<0.0001) and 12.7% NTx

(p<0.05) than the low usual daily intake group (Table 6.2). There were no significant differences

in the concentrations of protein thiols, TAC or BAP between these two groups (Table 6.2).

However, the lycopene-supplemented group had 28.1% lower CAT activity (p<0.0001), 17.6%

lower SOD activity (p<0.001) and 53.0% higher GPx activity (p<0.0001) than the low usual

daily intake group (Table 6.2).

The lycopene-supplemented group had 22.0% lower TBARS (p<0.001) and 15.3% lower

NTx (not significant, p=0.05) than the high usual daily intake group. There were no significant

differences in the concentrations of protein thiols, TAC or bone formation between these two

groups (Table 6.3). However, the lycopene-supplemented group had 28.4% lower CAT activity

(p<0.001), 18.4% lower SOD activity (p<0.001) and 27.1% higher GPx activity (p<0.05) than

the high usual daily intake group (Table 6.3).

Participants in low usual daily intake group had significantly lower lycopene intake,

corresponding to a significantly lower total serum lycopene, all-trans, 5-cis and other-cis

lycopene, than the high usual daily intake group (all: p<0.0001). In addition, the high usual daily

126
intake group had 35.5% higher GPx (23.92 ± 3.26 U/g Hb) than the low usual daily intake group

(15.44 ± 1.40 U/g Hb) (p<0.05). However, despite the significant difference in serum lycopene

between these two groups, there were no other significant differences in oxidative stress

parameters, antioxidant capacity or bone turnover markers between women who consumed low

lycopene and women who consumed high lycopene in their usual daily diets (Tables 6.2 and

6.3).

Table 6.1: Characteristics for each participant group, demonstrating no significant differences in
any of the measured parameters.

Parameter Summary statistics (mean ± SE)

Measured
Low usual daily High usual daily Lycopene-
intake group intake group supplemented group
(N=48) (N=27) (N=45)
Age (years) 55.60 ± 0.45 56.37 ± 0.54 55.22 ± 0.43
BMI (kg/m2) 25.55 ± 0.70 25.39 ± 0.79 25.71 ± 0.83
Pulse rate 73.46 ± 1.52 74.14 ± 2.12 71.65 ± 1.89
(beats/min)
Systolic BP (mm 121.5 ± 2.76 121.3 ± 2.92 118.0 ± 2.74
Hg)
Diastolic BP (mm 77.95 ± 1.82 76.52 ± 1.63 75.06 ± 1.49
Hg)
Years since 6.30 ± 0.88 6.59 ± 1.05 5.29 ± 0.73
menopause
Energy (kcal/day) 1816 ± 78.74 2032 ± 137.8 2307 ± 537.5

Calcium intake 1019 ± 81.40 1345 ± 229.8 2307 ± 537.5

(mg/day)
Vitamin C intake 265.4 ± 48.54 460.5 ± 116.5 118.7 ± 34.07a
(mg/day)
Vitamin D intake 221.1 ± 38.79 444.2 ± 108.4 215.8 ± 48.64
(IU/day)
a
Significantly lower than low (p<0.05) and high usual daily intake groups (p<0.001).

127
Table 6.2: Comparison of lycopene values, oxidative stress parameters and bone turnover
markers between women who were supplemented with lycopene with those who obtained a low
lycopene intake from their usual daily diet, using an unpaired t-test.

Parameter measured Mean ± SEM p

Low usual daily Lycopene- value1
intake group supplemented group
(N=48) (N=45)
Lycopene intake (mg/day) 2.59 ± 0.32a 43.33 ± 2.84 <0.0001
Serum lycopene Total 1094 ± 80.24 2012 ± 88.56 <0.0001
(nM) all-trans 539.0 ± 40.07 979.0 ± 48.48 <0.0001
5-cis 342.4 ± 26.13 685.0 ± 30.22 <0.0001
other-cis 212.6 ± 15.96 347.7 ± 17.36 <0.0001
Total antioxidant capacity (mM) 1.65 ± 0.03 1.66 ± 0.05 0.190
Bone turnover NTx (nM 21.97 ± 1.11 19.19 ± 0.79 0.047
markers BCE)
BAP (U/L) 24.42 ± 1.40 23.56 ± 1.02 0.900
Oxidative stress Protein thiols 454.7 ± 15.37 455.2 ± 15.50 0.982
parameters (M)
TBARS 9.06 ± 0.35 6.80 ± 0.35 <0.0001
(nmol/mL)
Antioxidant CAT (K/g Hb) 81.50 ± 4.13 58.59 ± 2.06 <0.0001
enzymes SOD (U/mg 47.60 ± 2.44 39.22 ± 4.72 0.001
Hb)
GPx (U/g Hb) 15.44 ± 1.40 32.82 ± 2.85 <0.0001
1
Data that were not normally distributed were compared using the Mann-Whitney test
a
The range of lycopene intake for the low usual daily intake group is 0.0-6.07 mg/day.

128
Table 6.3: Comparison of lycopene values, oxidative stress parameters and bone turnover
markers between women who were supplemented with lycopene to those who obtained a higher
than average lycopene intake from their usual daily diet, using an unpaired t-test.

Parameter measured Mean ± SEM p

High usual daily Lycopene- value1
intake group supplemented group
(N=27) (N=45)
Lycopene intake (mg/day) 11.83 ± 0.89a 43.33 ± 2.84 <0.0001
Serum lycopene Total 1780 ± 121.66 2012 ± 88.56 0.122
(nM) all-trans 870.5 ± 62.09 979.0 ± 48.48 0.174
5-cis 573.5 ± 41.08 685.0 ± 30.22 0.030
other-cis 335.5 ± 22.98 347.7 ± 17.36 0.617
Total antioxidant capacity (mM) 1.73 ± 0.04 1.66 ± 0.05 0.981
Bone turnover NTx (nM 22.66 ± 1.85 19.19 ± 0.79 0.053
markers BCE)
BAP (U/L) 23.01 ± 1.73 23.56 ± 1.02 0.772
Oxidative stress Protein thiols 455.4 ± 22.16 455.2 ± 15.50 0.993
parameters (M)
TBARS 8.72 ± 0.55 6.80 ± 0.35 0.006
(nmol/mL)
Antioxidant CAT (K/g Hb) 81.87 ± 5.60 58.59 ± 2.06 0.001
enzymes SOD (U/mg 48.08 ± 3.51 39.22 ± 4.72 0.003
Hb)
GPx (U/g Hb) 23.92 ± 3.26 32.82 ± 2.85 0.044
1
Data that were not normally distributed were compared using the Mann-Whitney test
a
The range of lycopene intake for the high usual daily intake group is 6.37-21.43 mg/day.

129
 DISCUSSION
In this clinical study, data from women who were supplemented with lycopene for four months,

resulting in significantly decreased oxidative stress parameters and decreased NTx (Chapter 5),

were compared to data from the cross-sectional study of women of the same age who obtained

lycopene in their usual daily diets (Chapter 3). The present study demonstrated a direct

association between high serum lycopene, as obtained through supplementation, to low TBARS

and NTx.

Supplementation with lycopene for four months resulted in significantly increased TAC

(for results on TAC please refer to Chapter 5). However, in this study, the lycopene-

supplemented group did not have significantly higher TAC than women who consumed lower

amounts of lycopene in their daily diets (low and high usual daily intake groups). The TEAC

assay measures the overall capacity of all antioxidants in vitro, including dietary antioxidants,
223, 240
thiol groups of proteins, uric acid and antioxidant enzymes . The lack of significant

difference could be due to dietary differences or differences in multivitamin intake between these

groups. The lycopene-supplemented group only obtained lycopene through their supplements

and refrained from consuming other lycopene-containing foods, such as raw tomatoes, which

may contain higher or different amounts of other polyphenols, vitamins and carotenoids than the
191, 192
provided supplements . In addition, since multivitamins contain antioxidants, participants

in the intervention study were asked to refrain from consuming them throughout the study

duration as they may have interfered with the effects of lycopene supplementation. Conversely,

participants from the low and high usual daily intake group maintained their usual habits, which

in many cases included the consumption of multivitamins or antioxidant supplements (i.e.

vitamins C or E). This was evidenced by the fact that in the low usual daily intake group 50.0%

of participants consumed multivitamins or other antioxidant vitamins on a daily basis, and in the

130
high usual daily intake group 51.9% of participants did so as well. As shown in the results,

participants from the low usual daily intake and high usual daily intake groups had significantly

higher vitamin C intake than participants in the lycopene-supplemented group (p<0.05 and

p<0.001, respectively, Table 6.1). It is highly probable that these factors are responsible for the

lack of significant difference in TAC between the lycopene-supplemented group and the usual

daily intake groups.

Research has shown that lycopene consumption significantly increases activities of GPx

and SOD. However, these studies are limited and were only seen in participants with chronic

diseases, such as hypertension, which are usually associated with lower activities of antioxidant
199, 200, 226
enzymes . The findings reported in this chapter showed that in healthy participants a

higher lycopene was associated with a higher GPx and lower SOD and CAT. A study on how
229
lycopene consumption affects antioxidant enzymes in female Wistar rats showed that GPx

and SOD were stimulated only at lower doses of lycopene, while higher doses resulted in

activities similar to baseline values. The authors propose that this effect is due to an as of yet

undetermined secondary mechanism. Based on this study by Breinholt et al. 229, together with the

findings of this chapter, it is possible that this mechanism may be the result of a U-shaped dose

response between antioxidant enzyme activity and lycopene concentration, similar to that
241
described for phytoestrogens . Regardless, the relationship between lycopene and antioxidant

enzymes is apparently quite complex and requires further in-depth study.

We have shown in a previous cross-sectional study that postmenopausal women

consuming high levels of lycopene in their daily diets (7.35 ± 0.80 mg/day) had significantly

lower biomarkers of protein oxidation and bone resorption than women consuming a lower daily
132
lycopene intake ( 3.68 ± 0.94 mg/day) (N=33) . In the present study, with a larger total

sample size (N=120) this association was not found. This discrepancy may be due to SNPs

131
which affect the response to lycopene, particularly at these lower intakes. This idea will be

further explored in Chapters 7 and 8. The fact that women consuming a higher dose of lycopene

(provided by lycopene supplements, intake ranging from 30 to 70 mg/day), had significantly

lower TBARS and NTx, than women who obtained lycopene through their daily diets (intake

ranging from 0.0 to 21.43 mg/day) suggests that supplementation with higher doses of lycopene

may be necessary to significantly decrease oxidative stress, and its associated loss of bone, to

decrease the risk of osteoporosis.

The difference in TBARS (p<0.01) and NTx (not significant, p=0.05) between the

lycopene-supplemented group and high usual daily intake group is of particular interest due to

the fact that the total serum lycopene was not significantly different between the groups (11.5%

higher in lycopene-supplemented group, p=0.12, Table 6.3). This similarity between total serum

lycopene concentrations was expected because pharmacokinetics show that the amount of

lycopene absorbed is the same from single doses of 10-120 mg, suggesting a saturation of
181
absorption . There was however, a significant difference between serum concentrations of 5-

cis lycopene (p<0.05, Table 6.3); those participants in the lycopene-supplemented group had

16.3% higher 5-cis serum lycopene than those participants in the high usual daily intake group.

This suggests that lower TBARS and NTx in the lycopene-supplemented group are due in large

part to a high serum concentration of 5-cis lycopene. Because there were no significant

differences in the other carotenoids between the lycopene-supplemented group and the usual

daily intake groups, particularly -carotene and lutein which were present in the lycopene

supplements given (Table 5.2, Chapter 5), one can assume that the differences in NTx and

TBARS can be attributed to lycopene alone. The fact that this isomer of lycopene contains the

highest antioxidant capacity of all of the lycopene isomers 44 lends further proof to the idea that it

132
is the antioxidant capacity of lycopene which has the ability to decrease biomarkers of oxidative

stress and bone resorption.

These findings suggest that increasing consumption of foods rich in the cis isomers of

lycopene would result in the beneficial lowering of TBARS and NTx. The treatments consumed

by the lycopene-supplemented group provided 1.20 ± 0.0 (tomato lycopene capsules), 4.74 ±

0.91 (regular tomato juice) and 14.20 ± 3.44 (lycopene-rich tomato juice) mg per day of total cis

lycopene (Table 5.2, Chapter 5). For participants in the usual daily intake groups, the most

frequently consumed source of lycopene was raw tomatoes (Chapter 3). Because more than 95%

of lycopene in standard raw tomatoes is in the all-trans configuration, it stands to reason that

even participants who consumed a high amount of lycopene in their daily diets would be

consuming less than 0.5 mg of cis lycopene per day if the majority of lycopene was provided as

raw tomato. Processed tomato products, such as the tomato juice consumed by the lycopene-

supplemented group, are more readily absorbed because processing reduces particle size and

divides the plant cell wall and fibrous pieces. Although research suggests that some all-trans to

cis isomerization occurs during digestion, and that the cis isomers of lycopene are more readily
85
absorbed , in order to have significantly increased 5-cis serum lycopene over and above that

usually provided by the daily diet, increased consumption of foods rich in cis, such as tomato
85, 168
juice or soup should be considered . These foods typically provide not only a higher

concentration of cis isomers in a matrix that enhances absorption, but a higher overall quantity of

total lycopene which would contribute to a higher 5-cis concentration in the serum. However, the

high serum concentrations of 5-cis shown to be beneficial would require one to consume quite a

large quantity of tomato products on a daily basis. For example, 1.25 cup tomato sauce, 3.5 cups

tomato soup, 6 cups of watermelon, or 13 medium-sized raw tomatoes, would each provide
89
approximately 43 mg of lycopene per day . Thus, a more realistic intake goal may be an

133
increased daily consumption of processed tomato products, together with either lycopene

provided in capsule (1 capsule contains 15 mg lycopene, LycoRed Ltd., Israel) or juice form

(156 mL serving contains 15 mg lycopene, Heinz, Canada).

In conclusion, these results have shown that in postmenopausal women lycopene

supplementation (43.33 ± 2.84 mg/day) was associated with significantly lower biomarkers of

lipid peroxidation and bone resorption than would be obtained if lycopene was consumed solely

through the daily diet. This may be attributed to the significantly higher serum concentration of

5-cis lycopene associated with consumption of lycopene supplements such as tomato juice or

capsules, which are not frequently consumed in great quantities in the usual daily diet. This study

provides further evidence that supplementation with lycopene is beneficial for reducing bone

turnover markers in postmenopausal women and may contribute towards establishing specific

recommendations for daily lycopene supplementation with tomato juice or capsules.

134
 CHAPTER 7

Paraoxonase 1 polymorphisms 172TA and 584AG

modified the association between serum concentrations of
the antioxidant lycopene and bone turnover markers and
oxidative stress parameters in women 25-70 years of age

This chapter has been submitted to the Journal of Nutrigenetics and Nutrigenomics as an original
article entitled: “Paraoxonase 1 polymorphisms 172TA and 584AG modified the association
between serum concentrations of the antioxidant lycopene and bone turnover markers and
oxidative stress parameters in women 25-70 years of age” by E.S. Mackinnon, A. El-Sohemy,
A.V. Rao, and L.G. Rao.

135
 ABSTRACT

Background/Aims: Polymorphisms of the paraoxonase 1 (PON1) enzyme affect the ability to

protect LDL from oxidation. Oxidative stress is a risk factor for osteoporosis and antioxidants

may be beneficial for prevention. The aim of this study was to determine whether PON1

genotypes modified the association between lycopene and bone turnover markers and oxidative

stress parameters. Methods: Blood samples from 107 women 25-70 years of age were analyzed

for serum carotenoid concentrations, BAP, NTx and oxidative stress parameters. Subjects were

genotyped for the 172TA and 584AG polymorphisms of PON1. Results: The PON1

polymorphism modified the association between lycopene and NTx and BAP (p<0.02 and

p<0.05 for interaction). In the combined 172TT and 584G genotype, high serum lycopene was

associated with decreased BAP (p<0.01) and NTx (p<0.05). Among those with the combined

172A and 584G genotype, however, increased serum lycopene was associated with increased

BAP (p<0.05) and NTx (p<0.05). Conclusions: These findings show that PON1 polymorphisms

modified the association between serum concentrations of lycopene and oxidative stress

parameters and bone turnover markers and may, therefore, moderate the risk of osteoporosis.

136
 INTRODUCTION
141
Human paraoxonase 1 (PON1) is an HDL-associated, calcium-dependent esterase enzyme .

One of the main physiological functions of the PON1 enzyme is that it contributes to the
143
antioxidant properties of HDL . As an HDL-associated enzyme PON1 retards or prevents the

oxidative modification of LDL by hydrolyzing lipid peroxides and cholesteryl-ester

144, 145 146
hydroperoxides on LDL . It has also been shown to hydrolyze hydrogen peroxide ,a

potent reactive oxygen species (ROS). These mechanisms act to prevent the accumulation of the
147
damaging ROS, particularly the lipid hydroperoxides , and decrease oxidative stress. PON1
148
activity is typically measured by determining the ability of PON1 to hydrolyze paraoxon .

There is an inverse correlation between the ability of PON1 to hydrolyze paraoxon and its ability
149
to hydrolyze lipid peroxides . The PON1 gene has two single nucleotide polymorphisms

(SNPs) that result in amino acid substitutions which can modify its activity. At position 55,

leucine is replaced by methionine (172TA or Leu55Met) and at position 192, glutamine is

replaced by arginine (584AG or Gln192Arg). The genotypes 172TT and 584GG have been

shown to be the least effective in protecting LDL from oxidation 148, 242.

Oxidative stress is a potential risk factor for osteoporosis, because it can increase loss of
54
bone . Lipid peroxidation by-products, such as lipid hydroperoxides, are potent ROS,
162
mutagenic to cells in vivo . LDL oxidation products stimulate bone loss by inducing bone
163
marrow progenitors to an adipogenic, rather than osteogenic fate . These by-products can

inhibit differentiation of osteoblasts by changing mineral content, decreasing formation and

163, 164
inhibiting mineralization . Due to the antioxidant capacity of PON1, it is possible that it

could prevent these effects of oxidative stress on bone development. To date only one study has

assessed the effects of PON1 polymorphisms on the risk of developing osteoporosis. Yamada et
166
al. examined the relation between BMD and the 172TA and 584AG polymorphisms in

137
postmenopausal women. For 172TA, participants with the TT genotype had significantly

lower BMD in the lumbar spine and femoral neck than carriers of the A allele. For 584AG,

participants with the GG genotype had significantly lower BMD of the femoral neck than

carriers of the A allele. These genotypes associated with lower BMD (172TT and 584GG) are
148
the least effective in protecting LDL against oxidation and this observation suggests a

potential mechanism by which PON1 affects the risk of osteoporosis.

Lycopene is a potent carotenoid antioxidant found in tomatoes and tomato products, as

well as other red-pigmented fruits, such as watermelon and pink grapefruit. The uptake of

lycopene occurs via LDL receptors and tissues containing high LDL receptor activity, such as

the prostate, adrenals and liver, tend to contain higher concentrations of lycopene 85. Due to its

location in the lipoprotein component, it has been correlated with the ability to decrease LDL
119
oxidation and overall lipid peroxidation . This, together with its potent antioxidant capacity,
150
suggests that lycopene may have a powerful sparing capacity of PON1 . Many lycopene

intervention studies have associated lycopene with decreased risk for chronic diseases associated
243
with oxidative stress . However, only two studies have assessed the effect of PON1
160, 161
polymorphisms on lycopene intervention . Bub et al. reported that in older participants

with increased oxidative stress who were carriers of the 584G allele, consumption of tomato

juice for a period of 8 weeks resulted in significantly decreased LDL-oxidation and TBARS, and
160
increased antioxidant capacity, while those with the 584AA genotype showed no effect .

Similar findings were reported in healthy participants with a shorter period of intervention with
161
tomato juice . The effect of the 172TA polymorphism on lycopene intervention was not

reported by these authors.

Studies examining whether the 172TA and 584AG polymorphisms moderate the

association between lycopene and oxidative stress parameters and bone turnover markers have

138
not yet been carried out. This chapter investigates the effects of serum lycopene on bone turnover

markers, oxidative stress parameters and antioxidant capacity and whether the 172TA or

584AG polymorphisms modulate these effects in women between the ages of 25-70 years.

METHODS

Participant Recruitment and Blood Sample Collection

This protocol was approved by the Research Ethics Board at St. Michael's Hospital, Toronto,

Ontario, Canada and followed the guidelines of good clinical practices. 108 female participants

25-70 years of age were recruited by telephone and advertisements. Any participants who were

on medications that could alter bone metabolism were excluded. This was a cross-sectional study

design in which participants submitted dietary records outlining foods, beverages and nutritional

supplements consumed over the previous 7 days and provided a baseline 12-hour fasting blood

sample. Blood pressure, height and weight were taken.

Dietary analyses

Food records were analyzed using NutriBase 5™ Clinical Edition software (version 5, released

2004, CyberSoft, Inc., Phoenix, AZ). This software generated a daily energy output and using the

average energy intake for the 7-day period, any days that were not within 30% of the average

were considered unusual and were removed from the analysis. The remaining days were used to

calculate the average daily intake of all of the macro- and micronutrients for each participant.

Lycopene intake could not be assessed by this software and was, therefore, analyzed separately

using the USDA national nutrient database for lycopene as a reference, which lists the content of
177
lycopene in each food as µg/measure . Using this information, the lycopene content was

139
calculated in milligrams for each food, and an average of the total daily lycopene consumed was

calculated for each participant.

Blood sample analyses

Unless otherwise specified, all materials were obtained from Sigma Aldrich Canada, Oakville,

ON, Canada. DNA was isolated from the whole blood of participants using the Blood Genomic

DNA Isolation Kit (NORGEN Biotek Corporation, Thorold, ON). Allelic discrimination of the

PON1 polymorphisms 584AG (rs662) and 172TA (rs854560) was performed using

TaqMan® SNP Genotyping Assays and the StepOnePlus™ Real-Time PCR system (Applied

Biosystems, Canada). For an example of the StepOnePlus™ Real-Time PCR system’s allelic

discrimination plot used to determine genotype, please refer to Appendix III, Figures IIIb and

IIIc.

For specific details on the following methodology, please refer to Chapter 4. The

concentrations of serum lycopene, lutein, -cryptoxanthin, zeaxanthin and -/-carotene were

measured using HPLC, according to previously published methods 178, with minor modifications.

Lycopene was detected as all-trans, 5-cis and other-cis isomers. TAC was measured using the
206
TEAC method . Protein oxidation was determined by estimating protein-sulfhydryl groups
207
(thiols) in serum . A high concentration of protein thiols corresponds to a lower protein

oxidation. Lipid peroxidation was measured in the serum using the TBA-MDA assay and was
208 209
reported as TBARS . CAT was measured using the H2O2 decomposition method and its

activity was expressed as K/g Hb, where K = 2.3/(Δ time)log OD15 seconds/OD 30seconds. SOD was

determined by the auto-oxidation of epinephrine 210 and was expressed as U/mg Hb. GPx activity

was measured by the decrease in NADPH absorbance at 340 nm at 37ºC in the presence of H2O2
211
and was expressed as U/g Hb. Commercially purchased ELISA kits were used to measure the

serum concentrations of the bone turnover markers NTx, a measure of bone resorption (INTER

140
MEDICO, Ontario) and BAP, a marker of bone formation (ESBE Scientific, Ontario). NTx was

expressed as nanomoles of bone collagen equivalents per litre (nM BCE) and BAP was

expressed as U/L, where one Unit represented 1 mol of p-nitrophenyl phosphate hydrolyzed per

minute at 25C in 2-amino-2-methyl-1-propanol buffer.

Statistical Analyses

All statistical analyses were performed using GraphPad PRISM 5.00 for Windows (GraphPad

Software, California) or SPSS Statistics 17.0 (SPSS, Inc., Chicago, Illinois). Summary statistics

were generated for participant characteristics of each genotype and presented as means ±

standard error of the mean (SEM). Differences between genotypes were detected by unpaired t-

test or the Mann-Whitney test for data that were not normally distributed. Analyses were first

performed separately for the 172TA and 584AG SNPs, then participants were grouped

according to their combined genotypes for 172TA and 584AG. Alleles of each were

designated with numbers such that for the 172TA genotype, TT = 0 and carriers of the A allele

= 1 and for the 584AG genotype, AA = 0 and carriers of the G allele = 1. This resulted in 4

groups for the combined polymorphisms 172/584 designated as: 0/0 (N=18), 0/1 (N=26), 1/0

(N=37) and 1/1 (N=26), respectively. Multiple linear regressions were performed to determine

whether there were linear relationships between serum lycopene and bone turnover markers,
244
oxidative stress parameters and antioxidant capacity among the different genotypes . Data

presented below were adjusted for age and BMI. If there was a significant linear relationship

between serum lycopene and oxidative stress parameters or bone turnover markers, an age and

BMI-adjusted general linear model (GLM) was used to test for a significant interaction between

genotypes and serum lycopene on that parameter. The model included NTx, BAP, or TBARS as

the dependent variable, with genotype, serum lycopene and their interaction included as fixed

factors. The covariates age and BMI were included in the model as they are significant predictors

141
of oxidative stress parameters and bone turnover markers. When multiple linear regression

showed significant effects, slope comparisons of the linear regression lines were also performed.

Within combined genotypes, participants were grouped by median serum lycopene into “high”

serum lycopene and “low” serum lycopene participants. An unpaired t-test was performed to

determine whether there were differences in bone turnover markers, oxidative stress parameters

and antioxidant capacity within these genotype groups 244.

RESULTS

In total, 108 participants were recruited and completed the study. One participant was excluded

from the following analyses as genotyping of the sample was unsuccessful, resulting in a total

sample size of 107 participants. Participants were grouped according to their genotype for the

172TA and 584AG polymorphisms. The genotype distributions were in Hardy-Weinberg

equilibrium for 172TA (TT = 44, TA = 45 and AA = 18) and 584AG (AA = 55, AG = 42

and GG = 10). The minor allele frequencies of the two SNPs were as follows: 172TA; 0.38

and 584AG; 0.29. Due to the small number of participants with the genotype 584GG (N=10),

these participants were combined with 584AG heterozygotes and classified as carriers of the

584G allele (N=52). Similarly, participants with the 172AA genotype were grouped with 172TA

heterozygotes and classified as carriers of the 172A allele (N=63). All characteristics of the

participants are outlined in Table 7.1. For the 172TA genotype, participants with the 172A

allele had significantly higher BAP than participants with the 172TT genotype (p<0.05, Table

7.1). This finding was unexpected considering the significantly higher BMI in that genotype

(p<0.01, Table 7.1) and the known negative correlation between BMI and bone turnover markers
20
.

142
 Multiple linear regression analyses of 172TA polymorphism demonstrated that in

participants with the 172TT genotype, there was a negative relationship between NTx and serum

lycopene (unadjusted, r = -0.33, p<0.05). This effect remained significant after adjusting for age

and BMI (adjusted, r = -0.35, p<0.02), demonstrating that each 1 nM increase in serum lycopene

was significantly associated with a 0.35 nM BCE decrease in NTx (Table 7.2). This effect was

not seen in carriers of the 172A allele. A significant interaction between the 172TA genotype

and lycopene was observed on NTx concentration (p<0.02; adjusted for age and BMI, Table 7.2).

In this model, the 172TA genotype was a significant predictor of NTx concentration (p<0.05;

adjusted for age and BMI, data not shown).

Multiple linear regression analyses of the 584AG polymorphism demonstrated that in

participants with the 584AA genotype there was a significant positive association between

TBARS and serum lycopene (adjusted, r = 0.23, p<0.05). This relationship demonstrated that

each 1 nM increase in serum lycopene was significantly associated with a 0.23 nmol/ml increase

in TBARS (Table 7.2). There was no significant interaction between 584AG genotype and

serum lycopene on TBARS (p=0.13, Table 7.2).

These results suggested that these two SNPs may have contrasting effects on bone

turnover markers and oxidative stress parameters in relation to serum lycopene, particularly for

the 172TA genotype. For these reasons, the participants were grouped according to combined

genotype, as described in the Methods, Statistical Analyses section, and the multiple linear

regression analysis was performed again 244.

143
Table 7.1: Participant characteristics for each genotype showing values for bone turnover
markers, oxidative stress parameters and other descriptive statistics.

Participant characteristic 172TA 584AG

TT TA+AA AA AG+GG
(N=44) (N=63) (N=55) (N=52)
Age (years) 48.27 ± 1.95 50.00 ± 1.34 49.76 ± 1.58 48.87 ± 1.61
BMI (kg/m2) 23.73 ± 0.56 26.01 ± 0.62a 25.26 ± 0.62 24.88 ± 0.64
Lycopene intake (mg/day) 5.65 ± 0.66 5.82 ± 0.66 5.49 ± 0.65 6.02 ± 0.83
Total serum lycopene (nM) 1251 ± 96.20 1376 ± 79.18 1397 ± 96.41 1248 ± 73.78
NTx (nM BCE) 20.48 ± 1.13 21.04 ± 1.13 20.92 ± 1.02 20.76 ± 1.16
BAP (U/L) 20.61 ± 1.20 24.17 ± 1.02a 23.98 ± 1.17 21.34 ± 1.03
Protein thiols (M) 490.3 ± 16.39 471.1 ± 13.04 492.8 ± 15.06 464.5 ± 13.49
TBARS (nmol/ml) 7.62 ± 0.46 7.64 ± 0.33 7.93 ± 0.40 7.31 ± 0.37
Trolox (mM) 1.65 ± 0.04 1.68 ± 0.03 1.63 ± 0.03 1.70 ± 0.03
CAT (K/g Hb) 90.25 ± 4.33 81.42 ± 3.75 82.74 ± 3.91 87.52 ± 4.17
SOD(U/mg Hb) 46.94 ± 2.40 49.15 ± 2.36 46.00 ± 2.05 50.63 ± 2.72
GPx (U/g Hb) 19.64 ± 2.18 17.71 ± 1.57 17.04 ± 1.81 20.04 ± 1.83
Energy (kcal/day) 1804 ± 66.12 1847 ± 84.79 1771 ± 65.26 1838 ± 80.07
Calcium (mg/day) 957.0 ± 74.70 1185 ± 116.1 1093 ± 81.98 942.4 ± 72.5
Vitamin C (mg/day) 301.0 ± 61.6 280.5 ± 53.2 337.1 ± 63.1 237.9 ± 48.2
Vitamin D (IU/day) 260.6 ± 47.82 317.1 ± 55.14 327.3 ± 57.52 258.6 ± 48.88
a
Significantly higher than 172TT genotype (t-test, p<0.05).

144
Table 7.2: Association between serum lycopene concentration (nM) and bone turnover markers,
oxidative stress parameters and antioxidant capacity for PON1 172TA and 584AG
genotypes.
Dependent variables Genotype
172TA
TT TA+AA
NTx  regression coefficients1 -0.352 0.202
2
R 0.472 0.357
p3 0.016 0.107
4
p 0.013

Dependent variables Genotype

584AG
AA AG+GG
TBARS  regression coefficients1 0.233 -0.090
R2 0.650 0.614
p3 0.036 0.445
p4 0.130
1
 regression coefficients were generated using an age and BMI-
adjusted multiple linear regression model
2
Multiple correlation coefficient
3
p value for the relation between serum lycopene and NTx or TBARS
4
p value for the interaction between serum lycopene and PON1
genotype on NTx or TBARS

145
Combined genotypes

The number of participants for each combined genotype is shown in Table 7.3. There were no

significant differences between genotypes for age, BMI, intake or serum lycopene, bone turnover

markers, oxidative stress parameters, antioxidant enzymes or capacity (Table 7.4).

There were significant positive correlations between lycopene intake and serum lycopene

in all genotypes (0/1: r = 0.75, p<0.001, 1/0: r = 0.68, p<0.001, 1/1: r = 0.49, p<0.02) except for

0/0 which was not significant (r = 0.46, p<0.08). It has been shown that serum lycopene values
183
may be affected by body weight , but in the present study these associations remained

significant after adjusting for weight, except for the 0/0 group (p=0.13). The lack of correlation

between intake and serum lycopene in the 0/0 group could be due to differences in carotenoid

bioavailability 160, however, a more likely explanation is the smaller number of participants with

this genotype (N=18). There were no significant associations between the PON1 genotype and

other serum carotenoids on any of the aforementioned dependent variables (such as BAP and

NTx), suggesting that the effects discussed below on bone turnover markers and oxidative stress

parameters can be attributed to serum lycopene alone.

Among participants with the 0/1 genotype, every 1 nM increase in serum lycopene was

significantly associated with a 0.40 nM BCE decrease in NTx (p<0.05, Table 7.5) and a 0.51 U/L

decrease in BAP (p<0.01, Table 7.5). A non-significant association between high serum

lycopene and lower concentrations of the lipid oxidation marker TBARS was also found

(p=0.07). Conversely, in participants with the 1/1 genotype a 1 nM increase in serum lycopene

was significantly associated with a 0.50 nM BCE increase in NTx (p<0.05) and a 0.50 U/L

increase in BAP (p<0.05). Slope comparisons of the linear regression lines demonstrated the

opposite effects of serum lycopene on NTx (p<0.001), BAP (p<0.002) and TBARS (p<0.01)

146
between the 0/1 and 1/1 genotypes (data not shown). There were no significant effects on these

parameters for the 0/0 or 1/0 genotypes (Table 7.5).

According to the GLM analysis, the combined PON1 genotype on its own was a

predictor of NTx (p<0.05; adjusted for age and BMI, data not shown). There was a similar, non-

significant effect on BAP (not significant, p=0.06; adjusted for age and BMI, data not shown).

However the association between genotype and NTx and BAP was modified by serum lycopene.

There was a significant interaction between serum lycopene and PON1 combined genotype on

NTx (p<0.02; adjusted for age and BMI, Table 7.5) and BAP (p<0.05; adjusted for age and BMI,

Table 7.5). An interaction between serum lycopene and PON1 combined genotypes on TBARS

was not observed (unadjusted, p<0.10, adjusted for age and BMI, p=0.30).

In order to verify these findings, participants were grouped according to median serum

lycopene within genotypes and characterized as having a “low” serum lycopene or a “high”

serum lycopene. In participants with the 0/1 genotype, those with a high serum lycopene had

27.0% lower lipid peroxidation than participants with low serum lycopene (p<0.05, Figure 7.1).

This genotype also showed differences in the bone turnover markers, BAP and NTx. In 0/1

participants with a high serum lycopene, BAP was 32.8% lower (p<0.02, Figure 7.1) and NTx

was 27.8% lower (p<0.02, Figure 7.1) than that found in participants with low serum lycopene.

Conversely, in 1/1 participants those with a high serum lycopene had a 35.9% higher NTx than

participants with a low serum lycopene (p<0.02, Figure 7.1). That the 0/1 and 1/1 genotypes

responded differently to lycopene is further illustrated by comparison of bone turnover markers

between those with high serum lycopene. Despite having the same concentrations of serum

lycopene (0/1: 1672 ± 127.8, 1/1: 1649 ± 99.54 nM), BAP and NTx were significantly lower in

participants with the 0/1 genotype group than in participants with the 1/1 genotype (unpaired t-

test, p<0.01 and p<0.05, respectively, Figure 7.1). These results were in keeping with the

147
findings from the multiple linear regressions as described above (Table 7.5). Non-significant

differences were also seen with respect to antioxidant enzymes for participants with the 0/1

genotype; high serum lycopene was associated with higher CAT (17%), SOD (16%) and GPx

(29%) than participants considered to have low serum lycopene (data not shown).

Table 7.3: Distribution of participants according to combined genotypes of the 172TA and
584AG polymorphisms.

Combined genotypes 172 584 N Total N per combined genotype

0/0 TT AA 18 18
0/1 TT AG 17 26
TT GG 9
1/0 AA AA 16 37
TA AA 21
1/1 AA AG 2 26
TA AG 23
AA GG 1
TA GG 0

Table 7.4: Participant characteristics for each combined genotype showing values for bone
turnover markers, oxidative stress parameters and other descriptive statistics.

Participant characteristic Combined genotype

0/0 0/1 1/0 1/1
Age (years) 47.67 ± 3.17 48.73 ± 2.50 50.78 ± 1.78 49.0 ± 2.08
BMI (kg/m2) 23.98 ± 1.06 23.55 ± 0.63 25.88 ± 0.76 26.2 ± 1.06
Lycopene intake (mg/day) 4.60 ± 0.95 6.37 ± 1.29 5.93 ± 0.84 5.67 ± 1.06
Total serum lycopene (nM) 1285 ± 168.8 1227 ± 116.2 1452 ± 118.1 1269 ± 93.10
NTx (nM BCE) 20.62 ± 1.33 20.39 ± 1.36 21.06 ± 1.39 21.00 ± 1.92
BAP (U/L) 21.20 ± 1.70 20.19 ± 1.67 25.66 ± 1.79 22.50 ± 1.20
Protein thiols (M) 509.3 ± 30.10 476.6 ± 18.18 484.8 ± 17.08 453.0 ± 20.02
TBARS (nmol/ml) 8.02 ± 0.85 7.34 ± 0.52 7.89 ± 0.43 7.28 ± 0.53
Trolox (mM) 1.62 ± 0.06 1.67 ± 0.05 1.64 ± 0.04 1.73 ± 0.04
CAT (K/g Hb) 79.23 ± 6.18 97.89 ± 5.55 84.50 ± 5.02 77.16 ± 5.61
SOD (U/mg Hb) 41.03 ± 2.75 51.20 ± 3.41 48.49 ± 2.69 50.07 ± 4.27
GPx (U/g Hb) 17.08 ± 3.70 21.41 ± 2.67 17.02 ± 2.03 18.66 ± 2.53
Energy (kcal/day) 1933 ± 129.0 1823 ± 105.7 1810 ± 110.5 1792 ± 107.2
Calcium (mg/day) 817.2 ± 120.9 989.3 ± 86.37 1180 ± 120.7 1054 ± 126.2
Vitamin C (mg/day) 316.0 ± 90.43 154.9 ± 37.39 407.2 ± 90.64 234.1 ± 66.14
Vitamin D (IU/day) 270.7 ± 102.4 254.8 ± 62.11 385.6 ± 79.73 215.8 ± 49.93

148
Table 7.5: Association between serum lycopene (nM) and bone turnover markers, oxidative
stress parameters and antioxidant capacity for PON1 172TA and 584AG genotypes.

Dependent variables Combined genotypes

0/0 0/1 1/0 1/1
NTx  regression coefficients1 -0.312 - 0.403 0.150 0.501
R2 0.328 0.525 0.378 0.499
p3 0.266 0.033 0.376 0.024
p4 0.012
BAP  regression coefficients1 0.344 -0.512 -0.081 0.495
R2 0.353 0.282 0.263 0.448
p3 0.218 0.008 0.645 0.030
p4 0.044
TBARS  regression coefficients1 0.291 -0.313 0.252 0.090
R2 0.820 0.615 0.650 0.541
p3 0.095 0.070 0.076 0.658
p4 0.298
1
 regression coefficients were generated using an age and BMI-adjusted multiple
linear regression model
2
Multiple correlation coefficient
3
p value for the relation between serum lycopene and NTx, BAP or TBARS
4
p value for the interaction between serum lycopene and PON1 genotype on NTx,
BAP or TBARS

149
 10
Low serum lycopene

8 High serum lycopene

TBARS (nmol/ml)
a
6

30

20 b
BAP (U/L)

10

30 b
NTx (nM BCE)

20 b

10

0
0/0 0/1 1/0 1/1
Combined PON1 genotype

Figure 7.1: Genotypes were stratified by median serum lycopene concentrations to compare
differences in the oxidative stress parameter TBARS and the bone turnover markers BAP and
NTx. Genotypes were grouped according to 172/584 and designated as: 0/0 = TT/AA, 0/1 =
TT/GG+GA, 1/0 = TA+AA/AA and 1/1 = TA+AA/GG, respectively. Median values of serum
lycopene were 1189 nM, 1164 nM, 1502 nM and 1208 nM for 0/0, 0/1, 1/0 and 1/1, respectively.
Values shown are means  SEM. Means were compared using an unpaired t-test (a p<0.05 and b
p<0.02).

150
 DISCUSSION

This chapter reported the effects of the 172TA and 584AG polymorphisms on the

association between the antioxidant lycopene and bone turnover markers and oxidative stress

parameters in women between the ages 25-70 years. A significant interaction was observed

between the combined genotype of 172TA and 584AG and lycopene on the bone turnover

markers BAP and NTx. These results demonstrated that in participants with the combination of

the 172TT genotype for the 172TA SNP and carriers of the 584G allele for the 584AG SNP,

there was a significant negative correlation between serum lycopene and bone formation and

resorption markers. The opposite effect was seen in participants with the combined 172A allele

carriers and 584G allele carriers, where a positive association between serum lycopene and NTx

was observed.

PON1 activity is depressed in diseases associated with oxidative stress, such as diabetes

and cardiovascular disease. This is probably due to the inactivation of the enzyme which occurs
150
as PON1 hydrolyzes the lipid peroxidation by-products typically elevated during states of

oxidative stress. Limited studies have suggested that there is no association between PON1
166, 167
concentrations and osteoporosis . However, an association between the 172TA and

584AG genotypes and BMD has been reported. Yamada et al. found that in postmenopausal

women, the 172TT genotype was associated with significantly lower BMD in the lumbar spine

and femoral neck than participants with the 172TA or 172AA genotypes, and the 584GG

genotype was associated with significantly lower BMD of the femoral neck than participants
166
with the 584GA or 584AA genotypes . BMD was not measured in the present study, rather,
236
the bone turnover markers BAP and NTx were used as predictors of bone health . In the

present study, findings showed that the PON1 combined genotype was an independent predictor

of concentrations of BAP and NTx. In addition, we found that those with the combined 172TT
151
and 584G allele carriers were affected favourably by high serum lycopene, with lower bone

turnover markers and lipid peroxidation. These findings suggest a role of lycopene in genotypes
166
which confer a higher risk for osteoporosis, as determined by reduced BMD . High bone

turnover markers result in injurious effects on bone composition and fragility 236, and a decrease

in bone turnover markers is strongly correlated with improved BMD and decreased fracture risk
11, 245
. The fact that high serum lycopene was associated with lower bone turnover markers in

participants with the combined 172TT and 584G allele carriers suggests that lycopene may

decrease the risk of osteoporosis for individuals with this genotype.

Bub et al., are the only other group to have assessed the effect of the PON1
160, 161
polymorphism and lycopene on oxidative stress parameters and antioxidant status .

Although they did not examine the effects of the 172TA polymorphism, their intervention

study showed that in participants carrying the 584G allele, consumption of tomato juice, which is

a rich source of lycopene, resulted in decreased LDL-oxidation and TBARS and increased serum

antioxidant capacity. This effect was not seen in 584AA participants. They suggest that this

effect could be due to the ability of 584AG to modulate PON1 activity since carriers of the

584G allele have higher PON1 activity than participants with the 584AA genotype 160. However,

in their other intervention study, the 584GG genotype also had a significantly higher baseline

lipid peroxidation level which could also explain the difference in magnitude of response to

lycopene 161. In the present study, when the association between serum lycopene, resulting from

usual daily lycopene intake, and oxidative stress parameters was analyzed according to the

584AG genotype, participants with the 584AA genotype had a significant, positive association

between serum lycopene and TBARS. More importantly, results showed that in the combined

genotype analysis of the 172TA and 584AG genotypes, a high lycopene was associated with

a lower TBARS, BAP and NTx in participants with the combined 172TT and 584G genotype

152
(Figure 7.1). These findings suggest that the 172TA polymorphism also makes an important

contribution to the mechanistic effect of lycopene. The significant interaction between genotype

and serum lycopene on NTx and BAP suggests that the PON1 genotype may be an important

factor in the ability of lycopene to decrease bone turnover markers, and thus may decrease the

risk of osteoporosis.

PON1 activity, as determined by the ability of PON1 to hydrolyze paraoxon, was not

measured in the present study. However, activity of PON1 is modulated by the 172TA and

584AG polymorphisms and the resulting differences in activity have been previously reported
144, 148
. There appears to be an additive effect of the 172TA and 584AG genotypes resulting

in a significantly higher PON1 activity, occurring in the order of magnitude

TT/GG>TT/AG>TT/AA 142. Based on previously reported findings 142, 145, 166 it can be estimated

that the PON1 activity of the combined 172TT and 584G genotype would be significantly higher

than the other combined genotypes. Studies suggest that dietary antioxidants may preserve PON1
141, 150
activity . Studies on human lipoproteins show that the flavonoid, glabridin, has a more

potent ability to reduce the amount of cholesteryl linoleate hydroperoxides, thus decreasing

oxidative stress, when it is added together with PON1 150. This could explain, in part, why there

was a significant association of lycopene in the combined 172TT and 584G genotype

participants, as this genotype confers the highest PON1 activity.

The combined 172TT and 584G genotype, shown to have the highest PON1 hydrolytic

activity, has the lowest protective effect against LDL oxidation 148, 242. Thus, dietary antioxidants

may be more important to assist with decreasing oxidative stress. Lycopene has the highest
117
antioxidant capacity among the carotenoids and is particularly effective in decreasing
44, 84
oxidative stress in disease states as is evidenced by many clinical studies . However, in

some cases, where participants are not in elevated states of oxidative stress, the effect of daily

153
lycopene intake may only be marginal. In particular, the effects on lipid peroxidation vary

considerably depending on the age and disease state of the study population 115, 122, 170. Recently,

it has been suggested that lycopene consumption can decrease oxidative stress only in those who

are genetically predisposed to low PON1 antioxidative ability (carriers of the 584G allele),
161
particularly in healthy participants . As discussed above, the association between the PON1

polymorphism and the response to lycopene has previously been reported in only two studies 160,
161
, but another antioxidant intervention study on 172TA and 584AG in which participants

were supplemented with antioxidant-rich walnut paste, has presented similar findings 159. Thus, it

may be that the different magnitudes of lycopene effect in previous studies, particularly in

healthy populations, may be attributed to unexplored PON1 polymorphisms. The present

findings suggest that lycopene, as obtained through a typical diet, is more beneficial in healthy

participants when PON1 antioxidative capacity is low (i.e. in the combined 172TT and 584G

genotype). This is an idea which requires further exploration with clinical intervention studies, in

which participants are supplemented with lycopene and the changes in oxidative stress

parameters and bone turnover markers are measured according to PON1 polymorphisms, and

will be addressed in Chapter 8.

A significant interaction between PON1 genotype and serum lycopene was observed on BAP

and NTx, suggesting that the antioxidant activities of PON1 and lycopene may act in conjunction

to decrease oxidative stress parameters and bone turnover markers. As described above this

would explain the association between lycopene and bone turnover markers in the combined

172TT and 584G genotype, as it has previously been associated with the highest PON1 activity,
148, 166, 242
lowest antioxidative capacity and decreased BMD . Research suggests that 584AG
159
may predict antioxidant status better than the 172TA polymorphism . Thus, the 584AG

SNP may have a greater effect on antioxidant mechanisms. This, together with the depressed

154
PON1 activity associated with the 172A allele, could explain the opposite association seen in the

combined 172A and 584G genotype and the absence of association seen in the other genotypes

(combined 172 TT and 584AA and combined 172A and 584AA). The detailed mechanisms by

which these effects occur require further study to confirm whether the combined 172TT and

584G genotype confers the best response to lycopene intervention with decreased bone turnover

markers and oxidative stress parameters. However, these novel findings suggest that in the

combined 172TT and 584G genotype, consumption of a diet rich in lycopene is associated with

lower bone turnover markers and lipid peroxidation, and may result in an overall decreased risk

for osteoporosis in women.

155
 CHAPTER 8

Supplementation with lycopene resulted in decreased lipid

peroxidation parameters, which interacted with the PON1
172TA genotype to decrease the bone resorption marker
NTx in postmenopausal women

This chapter has been submitted to Genes and Nutrition as an original article entitled:
“Supplementation with lycopene resulted in decreased lipid peroxidation parameters, which
interacted with the PON1 172TA genotype to decrease the bone resorption marker NTx in
postmenopausal women” by E.S. Mackinnon, A. El-Sohemy, A.V. Rao, and L.G. Rao.
156
 ABSTRACT

Lycopene is a potent antioxidant which may decrease the risk of osteoporosis. This effect is

attributed to the antioxidative properties of lycopene; however, the exact mechanism by which

this occurs has not yet been elucidated. A significant interaction between lycopene and the PON1

genotype has been shown. This chapter assesses whether the PON1 172TA polymorphism

affects the response to dietary intervention with lycopene. Following a one-month washout

period, 45 postmenopausal women, between 50-60 years of age, were supplemented for 4

months with either regular tomato juice, lycopene-rich tomato juice, or tomato lycopene

capsules, providing 30-70 mg of lycopene per day. Serum samples were analyzed for lycopene;

total antioxidant capacity; the oxidative stress parameters, protein thiols and TBARS; the bone

turnover markers, BAP and NTx; and the antioxidant enzymes, CAT, SOD and GPx. Genotyping

of the 172TA polymorphism was performed and repeated-measures ANOVA was used to

determine change over time for each genotype. Serum lycopene significantly increased as a

result of supplementation in the TT genotype and carriers of the A allele (both: p<0.0001), as

well as significantly decreased protein (p<0.005 and p<0.05, respectively) and lipid peroxidation

(p<0.005 and p<0.0005, respectively). However, participants with the TT genotype responded

more favourably to lycopene, with corresponding increased TAC (p<0.01) and significantly

decreased NTx (p<0.001). This effect was not significant in carriers of the A allele. A significant

interaction between PON1 genotype and change in TBARS (p<0.05) suggests that

supplementation with lycopene resulted in decreased lipid peroxidation, with interacted with the

PON1 172TT genotype to decrease bone resorption markers in postmenopausal women. These

findings provide mechanistic evidence of how intervention with lycopene may act to decrease

lipid peroxidation and thus the risk of osteoporosis in postmenopausal women.

157
 INTRODUCTION

Lycopene is an antioxidant which has been strongly associated with an ability to decrease

oxidative stress. Due to its location in the lipoprotein component of cells, it has been correlated

with the ability to decrease LDL oxidation and overall lipid peroxidation 99, 118, 119. Many studies

have attributed the antioxidant capacity of lycopene with a decrease in the risk of age-related
243
chronic diseases associated with oxidative stress, such as cancer and cardiovascular disease .
47, 56
Due to the fact that oxidative stress is a contributing risk factor for osteoporosis , lycopene

has recently been brought into focus as a possible beneficial supplement for bone. Increased
73
intake of lycopene has been associated with decreased risk of hip and non-vertebral fracture

and increased BMD in women 78, 80. Serum concentrations of lycopene tend to be decremented in
66, 103
women with postmenopausal osteoporosis . We were the first to show that a high serum

lycopene is associated with a decreased protein oxidation and the bone resorption marker NTx in
132
postmenopausal women , suggesting that the mechanism by which lycopene can decrease

osteoporosis risk and improve BMD is an antioxidative one.

Lipid peroxidation by-products have been associated with increased bone resorption and

decreased BMD. In vitro, studies have shown that LDL-oxidation products can stimulate
163
osteoprogenitor cells to an adipogenic, rather than osteogenic fate . Oxidized LDL can also

inhibit osteoblastic differentiation by decreasing formation, changing mineral content and

163, 164
inhibiting mineralization . Lycopene is well documented for its ability to specifically
99, 118, 119
protect LDL from oxidation , which lends further proof to the idea that lycopene may

benefit bone by its antioxidative properties.

141
PON1 is a calcium dependent esterase enzyme associated with HDL . The PON1

enzyme provides HDL with antioxidant properties 143 and is capable of slowing or abrogating the

oxidative modification of LDL. It accomplishes this by hydrolyzing lipid peroxides and

158
cholesteryl-ester hydroperoxides on LDL 144, 145. Because PON1 prevents the oxidation of LDL,

it is possible that it could prevent the effects of oxidative stress on bone. It is also capable of

neutralizing H2O2 167, which is highly reactive, and has been shown to enhance osteoclastic bone
43, 53, 56-58, 246
resorption and prevent differentiation and activity of osteoblasts . All of these

functions contribute to an ability of PON1 to decrease oxidative stress 147. PON1 can be modified

by antioxidants and may also have an effect on bone through its antioxidative mechanisms,

although studies in this specific area are limited 166, 167.

The PON1 gene has two SNPs that result in amino acid substitutions which can modify

its activity. At position 55, leucine is replaced by methionine (172TA or Leu55Met) and at

position 192, glutamine is replaced by arginine (584AG or Gln192Arg). Bub et al have shown

that the 584AG polymorphism affects the response to tomato juice intervention both in healthy
161 160
young men , and in older participants who typically have higher oxidative stress .

Specifically, in carriers of the 584G allele, consumption of tomato juice resulted in significantly

increased antioxidant capacity, together with decreased LDL-oxidation and TBARS, an effect

not seen in participants with the 584AA genotype 160, 161. These studies showed that the genotype

which is known to have the least antioxidative capacity of PON1 to protect LDL from oxidation
148, 242
(584GG) , was associated with decreased oxidative stress in response to lycopene

intervention. This suggests that lycopene may act as a secondary defense to combat oxidative

stress when internal antioxidant defenses are low. The effects of lycopene intervention on the

172TA polymorphism have not yet been reported.

Yamada et al, have examined the effects of PON1 polymorphisms on BMD in a large

population of 816 Japanese postmenopausal participants 166. For 172TA, participants with the

TT genotype had significantly lower BMD in the lumbar spine and femoral neck than

participants with the TA or AA genotype. For 584AG, participants with the GG genotype had

159
a significantly lower BMD of the femoral neck than participants with the GA or AA genotype.

These genotypes associated with lower BMD (172TT and 584GG) are also the least effective in
148
protecting LDL against oxidation which may suggest an important mechanism of the PON1

polymorphisms in decreasing the risk of osteoporosis. Interestingly, one of the genotypes

associated with decreased BMD, 584GG, was also associated with a more favourable response to

lycopene intervention with resulting decreased oxidative stress parameters and increased

antioxidant capacity 160, 161. However, studies specifically associating how PON1 polymorphisms

affects the response of oxidative stress and bone turnover markers to lycopene intervention have

not yet been conducted. This chapter examines whether the PON1 172TA polymorphism

affected the response to lycopene and attempted to delineate the mechanisms by which this can

occur.

METHODS

Participant Recruitment and Blood Sample Collection

Please refer to Chapter 5 of this thesis for detailed descriptions on participant recruitment, study

design and methodology. In brief, female participants aged 50-60 years old, at least one year

postmenopausal, non-smoking and not on medications that may affect bone metabolism were

recruited. This protocol was approved by the Research Ethics Board at St. Michael's Hospital,

Toronto, Ontario, Canada and followed the guidelines of good clinical practices. Participants

were randomly assigned to one of the four following lycopene supplement groups, taken twice

daily: (1) regular tomato juice (30 mg/day, Heinz, Canada) (2) lycopene-rich tomato juice (70

mg/day, Kagome, Japan) and (3) tomato lycopene capsules (30 mg/day, LycoRed Ltd., Israel).

Following a one-month washout period during which no lycopene-containing foods were

consumed, a baseline 12-hour fasting blood sample and 7-day dietary records were collected.
160
Additional blood samples and dietary records were collected after 2 and 4 months of

supplementation. Dietary analyses were carried out as described previously (Chapter 3) using

NutriBase 5™ Clinical Edition software (version 5, released 2004, CyberSoft, Inc., Phoenix,

AZ).

Blood sample analyses

Unless otherwise specified, all materials were obtained from Sigma Aldrich Canada, Oakville,

ON, Canada. Genotyping was performed as described in Chapter 7. Allelic discrimination of the

PON1 polymorphism 172TA (rs854560) was performed using TaqMan® SNP Genotyping

Assays and the StepOnePlus™ Real-Time PCR system (Applied Biosystems, Canada). For an

example of the StepOnePlus™ Real-Time PCR system’s allelic discrimination plot used to

determine genotype, please refer to Appendix III, Figures IIIb and IIIc.

The concentrations of serum lycopene, lutein, -cryptoxanthin, zeaxanthin and -/-

carotene were measured using HPLC, according to previously published methods 178, with minor

modifications. Lycopene was detected as all-trans, 5-cis and other-cis isomers. TAC was
206
measured using the TEAC method . Protein oxidation was determined by estimating protein-
207
sulfhydryl groups (thiols) in serum . A high concentration of protein thiols corresponds to a

lower protein oxidation. Lipid peroxidation was measured in the serum using the TBA-MDA
208
assay and was reported as TBARS . CAT was measured using the H2O2 decomposition
209
method and its activity was expressed as K/g Hb, where K = 2.3/(Δ time)log OD15 seconds/OD

30seconds. SOD was determined by the auto-oxidation of epinephrine 210 and is expressed as U/mg

Hb. GPx activity was measured by the decrease in NADPH absorbance at 340 nm at 37ºC in the

presence of H2O2 211 and was expressed as U/g Hb. Commercially purchased ELISA kits were

used to measure the serum concentrations of the bone turnover markers NTx, a measure of bone

resorption (INTER MEDICO, Ontario) and BAP, a marker of bone formation (ESBE Scientific,

161
Ontario). NTx was expressed as nanomoles of bone collagen equivalents per litre (nM BCE) and

BAP was expressed as U/L, where one Unit represented 1 mol of p-nitrophenyl phosphate

hydrolyzed per minute at 25C in 2-amino-2-methyl-1-propanol buffer. For specific details on

the above methodology, please refer to Chapter 4.

Statistical Analyses

All statistical analyses were performed using GraphPad PRISM 5.00 for Windows (GraphPad

Software, California) or SPSS Statistics 17.0 (SPSS, Inc., Chicago, Illinois). Summary statistics

of participant characteristics were generated for each genotype and presented as means ±

standard error of the mean (SEM). Repeated-measures ANOVA, with Tukey’s post hoc analysis,

was performed to determine the change in antioxidant capacity, oxidative stress parameters and

bone turnover markers with lycopene supplementation for each genotype. Average daily

lycopene intake provided by regular tomato juice (30 mg/day), lycopene-rich tomato juice (70

mg/day) or tomato lycopene capsules (30 mg/day) was calculated for each genotype group. As

described in Chapter 5 of this thesis, all three supplements resulted in similar serum lycopene

concentrations; therefore, participants were pooled for analyses and referred to as “lycopene-

supplemented”. Multiple linear regressions were performed to determine the correlation between

change in oxidative stress parameters, antioxidant capacity and bone turnover markers. Change

after 4 months of lycopene supplementation was expressed as a percentage of the baseline value.

Data presented below were adjusted for BMI. If there was a significant linear relationship found,

a test for a significant interaction was performed using GLM (SPSS Statistics 17.0). The model

included change in NTx as the dependent variable, with genotype, change in TBARS and their

interaction included as fixed factors. The covariate BMI was included in the model as it is a

significant predictor of oxidative stress parameters and bone turnover markers.

162
 RESULTS

In total, 45 participants supplemented with lycopene completed this intervention study. The

distribution of the 172TA genotype was in Hardy-Weinberg equilibrium (TT = 19, TA = 17

and AA = 9). The minor allele frequency was 0.39. Due to the small number of participants with

the AA genotype (N=9), this genotype was combined with TA heterozygotes and classified as

carriers of the A allele (N=26).

The participant characteristics after the washout are outlined in Table 8.1. There were no

significant differences among any of the participant characteristics at baseline, including

oxidative stress parameters and bone turnover markers (Table 8.1). There was a non-significant

difference in the daily lycopene intake to be provided by the supplements. As a result of

randomization of supplements assigned to participants, those with the TT genotype consumed a

quantity of 48.95 ± 4.71 mg of lycopene per day from regular tomato juice, lycopene-rich tomato

juice, or tomato lycopene capsules, while carriers of the A allele consumed a quantity of 39.15 ±

3.38 per day (not significant, p<0.10, Table 8.1). The supplements given to each of the 45

participants are shown in Table 8.2, including the percentage of participants within each

genotype who were given either regular tomato juice, lycopene-rich tomato juice or tomato

lycopene capsules. Of the supplements given, more participants with the TT genotype were

randomly assigned to consume lycopene-rich tomato juice (34.6%) while more carriers of the A

allele were randomly assigned to consume tomato lycopene capsules (35.3%, Table 8.2). This

accounts for the difference in mg quantities of lycopene consumed between the genotypes, as the

lycopene-rich juice provided 70 mg of lycopene per day compared to the tomato lycopene

capsules which provided 30 mg per day. The change in serum carotenoids after supplementation

is shown in Table 8.3. Irrespective of genotype or the amount of lycopene consumed, serum

lycopene significantly increased over the supplement period for all-trans, total cis and total

163
lycopene after 2 (p<0.001) and 4 months (p<0.001). Similarly, -carotene significantly increased

after 2 (p<0.05) and 4 months (p<0.01). There was no significant difference in the magnitude of

serum carotenoid increase between the genotypes (Table 8.3), suggesting that regardless of the

supplement given or the genotype of participants, the serum carotenoids all-trans, total cis, total

lycopene or -carotene significantly and similarly increased with lycopene supplementation.

There was no difference in absorption of lycopene among 172TA genotypes; the AUC for total

lycopene was 6429  549.0 nMmonth for the TT genotype and 6534  302.2 nMmonth for

carriers of the A allele (unpaired t-test, not significant, data not shown).

Supplementation with lycopene resulted in significantly lower protein oxidation and lipid

peroxidation in all participants regardless of the 172TA genotype (Table 8.4). However, there

was only a significant increase in TAC (ANOVA, p<0.01, Table 8.4) and significant decrease in

NTx (ANOVA, p<0.001, Table 8.4) in participants with the TT genotype. These participants also

showed a non-significant increase in GPx (ANOVA, p<0.10, Table 8.4). Thus, despite the

decrease in oxidative stress seen in all participants, only in participants with the TT genotype did

this decreased oxidative stress result in significant decreases in the bone resorption marker NTx

after 4 months of lycopene supplementation (decreased by 19.82 ± 4.97%). This average

decrease was almost twice as high as that seen in carriers of the A allele (decreased by 11.64 

4.79%).

While most carriers of the A allele had decreased NTx after lycopene supplementation,

this was not significant by repeated-measures ANOVA (Table 8.4). This could be due to the fact

that there was a higher percentage of “non-responders” associated with this genotype. “Non-

responders” were categorized according to the decrease, or lack thereof, in NTx after lycopene

supplementation, whereby a participant was considered a “non-responder” if the decrease in NTx

was -1.0%. Participants with the TT genotype tended to respond better to lycopene

164
supplementation with decreased NTx than carriers of the A allele did. For carriers of the A allele,

30.8% were considered non-responders, while only 15.8% of participants with the TT genotype

were considered non-responders. This indicates that the risk of being a non-responder was

approximately doubled in carriers of the A allele.

There was no significant difference in baseline concentrations of NTx between genotypes

(Figure 8.1). However, after 4 months of supplementation with lycopene, participants with the

TT genotype had an average NTx concentration of 17.02  0.92 nM BCE, which was

significantly lower than the average concentration of 21.52  1.27 nM BCE in carriers of the A

allele (p<0.02, Figure 8.1). There was no significant difference in the concentration of protein

thiols, TBARS, Trolox, BAP, CAT, SOD, or GPx between genotype groups after 4 months of

supplementation.

The data presented in this study suggested that women with the TT genotype responded

more favourably to lycopene supplementation, with significantly increased TAC and

significantly decreased NTx (Table 8.4), than carriers of the A allele. To examine possible

mechanisms of this effect further, multiple linear regression models of the change in oxidative

stress parameters and bone turnover markers for each genotype were performed. These data

suggested that there was a direct relationship between the change in the lipid peroxidation

parameter TBARS, and the change in the bone resorption marker NTx in participants with the

TT genotype. For participants with this genotype, every 1% decrease in TBARS was associated

with a 0.53% decrease in NTx (p<0.05; adjusted for BMI, Table 8.5). A significant interaction

between the PON1 172TA genotype and change in TBARS was observed on change in NTx

concentration (p<0.05, Table 8.5). This significant interaction (p<0.05), together with the data

from the repeated-measures ANOVA, suggest that lycopene acted to decrease lipid peroxidation

165
(TT, p<0.005 and A allele, p<0.0005, Table 8.5), which interacted with the PON1 172TA

genotype to decrease the bone resorption marker NTx in 172 TT participants.

Table 8.1: Baseline participant characteristics for 172TA genotype showing values for bone
turnover markers, oxidative stress parameters and other descriptive statistics after the washout
period.

Participant characteristic 172TA Genotype

TT TA+AA
(N=19) (N=26)
Age (years) 55.79 ± 0.67 54.81 ± 0.56
BMI (kg/m2) 25.69 ± 1.16 26.15 ± 1.21
Antioxidant capacity Serum lycopene (nM) 642.7 ± 651.2 ±
98.87 72.38
Trolox (mM) 1.50 ± 0.06 1.60 ± 0.07
Antioxidant enzymes CAT (K/g Hb) 55.66 ± 4.28 57.60 ± 2.06
SOD(U/mg Hb) 31.95 ± 3.86 34.17 ± 3.83
GPx (U/g Hb) 26.30 ± 4.17 25.82 ± 3.38
Oxidative stress parameters Protein thiols (M) 402.1 ± 408.2 ±
24.97 18.12
TBARS (nmol/ml) 7.04 ± 0.49 8.54 ± 0.69
Bone turnover markers NTx (nM BCE) 22.63 ± 1.68 25.36 ± 1.53
BAP (U/L) 23.30 ± 1.68 23.64 ± 1.57
Daily lycopene intake to be provided by supplements 48. 95 ± 4.71 39.15 ± 3.38
(mg/day)

Table 8.2: Number of participants in each supplement group, sorted according to 172TA
genotype.
Supplement group 172TA Genotype
TT TA+AA allele
(N=19) (N=26)
N % N %
Regular tomato juice 7 26.9 8 23.5
Lycopene-rich tomato juice 9 34.6 6 17.7
Tomato lycopene capsules 3 11.6 12 35.3

166
Table 8.3: Mean serum carotenoid concentrations for all participants for the duration of study,
showing increases in -carotene and individual lycopene isomers, including total cis and total
lycopene.

Serum carotenoids (nM) Time (months) PON1 172TA genotype

Mean ± SEM TT TA+AA
all-trans lycopene1 0 286.6 ± 46.26 289.5 ± 37.58
2 926.2 ± 110.2 935.9 ± 57.83
4 1022 ± 82.27 947.3 ± 59.19
Total cis lycopene1 0 356.1 ± 53.22 361.7 ± 36.82
2 929.4 ± 109.9 1023 ± 65.10
4 1053 ± 78.97 1018 ± 78.97
Total lycopene1 0 642.7 ± 98.87 651.2 ± 72.38
2 1856 ± 216.6 1959 ± 110.2
4 2075 ± 154.9 1965 ± 105.1
-carotene2 0 1237 ± 221.6 952.7 ± 121.7
2 1833 ± 308.4 1486 ± 128.9
4 2112 ± 384.2 1487 ± 128.5
1
All isomers of serum lycopene significantly increased after 2 (Tukey’s/Dunn’s multiple
comparison test, p<0.001) and 4 months (Tukey’s/Dunn’s multiple comparison test, p<0.001)
2
-carotene significantly increased after 2 (Dunn’s multiple comparison test, p<0.05) and 4
months (Dunn’s multiple comparison test, p<0.01)

167
Table 8.4: Results from repeated-measures ANOVA analysis demonstrating significant changes
over the course of the 4 month study period for participants supplemented with lycopene,
grouped according to 172TA genotype.

Parameter Statistical p value for lycopene-

measurement supplemented 172TA genotype
TT TA+AA
(N=19) (N=26)
Antioxidant Total serum % change after 4 350.4  71.15 301.9  50.19
capacity lycopene months
ANOVA1 ↑ p<0.0001 ↑ p<0.0001
0 vs. 4 months ↑ p<0.001 ↑ p<0.001
TAC % change after 4 14.32  5.63 3.01  2.96
months
ANOVA ↑ p<0.01 ns
0 vs. 4 months ↑ p<0.01 ns
GPx % change after 4 154.4  65.94 127.9  60.70
months
ANOVA ns (p=0.08) ns
0 vs. 4 months ns ns
Oxidative stress Protein thiols % change after 4 16.83  5.52 14.63  5.18
parameters months
ANOVA ↑ p<0.005 ↑ p<0.05
0 vs. 4 months ↑ p<0.05 ↑ p<0.05
TBARS % change after 4 -10.95  3.62 -12.64  2.70
months
ANOVA ↓ p<0.005 ↓ p<0.0005
0 vs. 4 months ↓ p<0.01 ↓ p <0.001
Bone resorption marker NTx % change after 4 -19.82  4.97 -11.64  4.79
months
ANOVA ↓ p<0.001 ns
0 vs. 4 months ↓ p<0.001 ns
1
Repeated-measures ANOVA with Tukey’s post hoc comparisons showing change after 4 months when ANOVA
significant (ns=not significant).

168
 TT
30
A allele

NTx (nM BCE)

20 *

10

0
Washout 4 months
Time period of intervention

Figure 8.1: Concentration of NTx in 45 postmenopausal participants at baseline (washout) and

after four months of supplementation with lycopene. Participants were grouped according to the
172TA genotype. Values are means  SEM and were compared using an unpaired t-test
(*p<0.02).

Table 8.5: Association between change in TBARS and change in NTx according to the 172TA
genotype
Dependent variable Genotype
172TA
TT TA+AA
NTx  regression coefficients1 0.525 -0.379
R2 0.428 0.229
3
p 0.02 0.063
p4 0.044
1
 regression coefficients were generated using an age and
BMI-adjusted multiple linear regression model
2
Multiple correlation coefficient
3
p value for the relation between change in TBARS and
change in NTx
4
p value for the interaction between change in TBARS and
PON1 genotype on NTx

169
 DISCUSSION
This chapter presented evidence suggesting that the PON1 172 TA polymorphism moderated

the response to lycopene intervention in postmenopausal women. Specifically, the TT genotype

conferred a more favourable response, with a corresponding decrease in the bone resorption

marker NTx. This was evidenced by results showing that while both genotypes had significantly

increased serum lycopene, with corresponding decreases in oxidative stress biomarkers, only the

TT genotype had significantly increased TAC and significantly decreased NTx. These effects

were not significant in carriers of the A allele. Despite the fact that there was no significant

difference in the baseline concentrations of NTx (Table 8.1), participants with the TT allele had

significantly lower NTx than carriers of the A allele following four months of lycopene

supplementation (Figure 8.1). Furthermore, the risk of being a non-responder to lycopene

intervention with decreased NTx was twice as high in carriers of the A allele.

The findings reported in this study are the first to provide evidence of a mechanism by

which lycopene can decrease the bone resorption marker NTx. A significant, linear correlation

between decreased TBARS and decreased NTx in participants with the TT genotype was shown.

A significant interaction indicated that the elevated serum lycopene associated with

supplementation resulted in decreased lipid peroxidation, which interacted with the PON1

polymorphism to affect the bone resorption marker NTx. Based on these findings we propose

that lipid peroxidation begins to decrease almost immediately following lycopene intervention,

which then contributes to the decrease in NTx in TT genotype participants. This is supported by

research showing that significant changes in lipid peroxidation markers can occur early with

lycopene supplementation. A randomized cross-over study by Agarwal et al found that lycopene

supplementation with doses similar to those used here (ranging from 39.2 to 75.0 mg/day)

resulted in significantly decreased TBARS and oxidized LDL after just one week of

170
supplementation 99. Significant changes in NTx after dietary intervention tend to take longer as

research suggests it may take anywhere from 1 to 2 months to note changes. For example,

restriction of lycopene for a 1 month period was shown to significantly increase NTx (Chapter

4), and lycopene supplementation was shown to significantly decrease NTx after 2 months

(Chapter 5). This finding is similar to a dietary intervention study with phylloquinone (Vitamin

K) in postmenopausal women, which demonstrated that it took approximately 2 months to note a

232
small, but significant decrease of 9% in serum NTx . Therefore it is entirely possible that the

change in TBARS started to occur immediately and this decrease in lipid peroxidation allowed

for the concomitant decrease in NTx (Table 8.5). Further, these findings support those previously

reported in this thesis, that in participants with the TT genotype, a higher serum lycopene was

significantly associated with a lower NTx (Chapter 7). This chapter expanded on these findings

to show that an increase in serum lycopene resulted in the decreased oxidative stress parameter

TBARS, resulting in the direct decrease of NTx in participants with the TT genotype.

These results also showed that the TAC significantly increased only in the TT genotype

group, suggesting that this genotype results in a better response to lycopene supplementation

with concomitant elevated antioxidant capacity. It is possible that this is not due to a function of

this SNP, but rather due to the fact that a higher number of participants with this genotype

consumed lycopene-rich tomato juice as part of the randomized intervention (34.6%, Table 8.2).

Further, carriers of the A allele consumed a higher percentage of tomato lycopene capsules,

which does not independently increase TAC (Appendix II). These findings may stem from the

fact that tomato juice contains other vitamins, polyphenols and carotenoids which contribute to
191, 192
its overall antioxidant capacity and ability to combat oxidative stress . However, another

possible explanation for this finding is that it may be a result of the PON1 172TA

polymorphism. This genotype (TT) may respond better to lycopene intervention with increased

171
 148, 242
TAC, as it is the genotype shown to possess the lowest antioxidative capacity of PON1 .

Interestingly, the TT genotype has also been associated with lower BMD of the lumbar spine and
166
femoral neck compared to the TA and AA genotypes by Yamada et al . PON1, as an

endogenous enzyme with antioxidative capacities, would be considered a part of the internal first

line of defense against oxidative stress 35. Because the TT genotype is associated with a lower

antioxidative ability of PON1, this suggests that the internal antioxidant defenses are not as

powerful, allowing for some oxidative damage to go unchecked 159. This may result in a build-up

of LDL-oxidation and lipid peroxidation by-products 163, 164 which can damage bone, resulting in

the decreased BMD previously associated with this genotype 166. In states such as this, where the

internal preventative antioxidants are not sufficient, a dietary antioxidant would be relied upon to
35
quench free radicals and decrease oxidative stress . Thus, the increased oxidative stress and

decreased BMD previously associated with the TT genotype could be compensated for by

lycopene. This may also account for the marginal increase in GPx associated with lycopene

supplementation in individuals with this genotype (Table 8.4). The mechanisms underlying this

concept warrant further investigation.

The lack of significant effect on antioxidant enzymes is not surprising. Many studies that

show improved antioxidant enzymes with lycopene supplementation are demonstrated in

populations known to have depleted internal defenses, such as those with cardiovascular disease
199, 200, 226
and diabetes . Chapter 5 showed that there was no significant change in antioxidant

enzymes in a healthy group of postmenopausal women. A large variability in the magnitude of

effects on antioxidant enzymes when participants are grouped according to the 172TA

genotype, indicated that some participants have increased, and some decreased, antioxidant

enzymes with lycopene supplementation. This may be due to the presence of other SNPs in the

172
antioxidant enzymes which may affect the response to antioxidant supplementation, an idea

which is beyond the scope of this chapter.

Data from the cross-sectional study suggested that there was a significant interaction

between the combined genotypes of 584AG and 172TA on markers of both bone formation

and resorption (Chapter 7). These two genotypes could not be combined for analyses in this

chapter, as the number of participants in each combined genotype group was too low (0/0, N=9,

0/1, N=10, 1/1, N=15 and 1/0, N=11, data not shown). A limitation of this study was the

relatively small sample size of 45 participants. However, based on the currently presented

findings, this is an idea which warrants further investigation using a larger, randomized

controlled trial, perhaps one in which participants are pre-selected based on their combined

PON1 genotype.

In summary, this chapter was the first to show that the PON1 polymorphism may affect

the response to lycopene, in that the TT genotype, which is associated with a low antioxidative

capacity of PON1, was correlated with a better response to lycopene with decreased NTx. The

present results have shown that the increased serum lycopene associated with long-term lycopene

supplementation decreased lipid peroxidation, which interacted with the PON1 polymorphism to

affect the bone resorption marker NTx. These findings showed a mechanism by which lycopene

can affect NTx in postmenopausal women, providing further proof that lycopene acts as a potent

antioxidant to decrease oxidative stress, and may therefore result in a decreased risk of

osteoporosis.

173
 CHAPTER 9

Cis lycopene isomers found in high concentrations in human

serum, and which possess the greatest antioxidant capacity,
were capable of preventing and repairing the damaging
effects of reactive oxygen species in human osteoblast cells

This chapter has been submitted to Free Radical Research as an original article entitled: “Cis
lycopene isomers found in high concentrations in human serum, and which possess the greatest
antioxidant capacity, were capable of preventing and repairing the damaging effects of reactive
oxygen species in human osteoblast cells” by E.S. Mackinnon and L.G. Rao.

174
 ABSTRACT

Research suggests that oxidative stress caused by ROS causes bone loss. In vitro, ROS can

stimulate bone resorption by osteoclasts and inhibit osteoblastic cell differentiation and

maturation, resulting in decreased bone formation. Lycopene is an antioxidant capable of

quenching ROS to decrease oxidative stress. Recent dietary studies have associated a high intake

of lycopene with significantly lower bone resorption markers and risk for hip and non-vertebral

fracture, and a positive 4 year change in lumbar spine BMD. Lycopene-containing foods consist

of primarily the all-trans lycopene isomer, while the isomeric ratio of lycopene present in human

serum is approximately 50:50 cis:trans lycopene. Cis isomers have higher antioxidant capacity

and are more readily absorbed during digestion than all-trans isomers. Despite this, current in

vitro studies typically utilize the all-trans form of lycopene. This chapter explores the hypothesis

that the cis isomers of lycopene, which are present in high concentrations in human serum and

which have higher antioxidant capacity than the all-trans isomer, would be more beneficial in

preventing and repairing the damaging effects of ROS in human osteoblast cells. To determine

whether lycopene could prevent the damaging effects of ROS on mineralized bone nodule

formation (MBNF) in osteoblasts, cloned human CD34+ osteoblasts were pre-treated with 1 M

of 45:55, 28:72 or 5:95 cis:trans lycopene for 48 hours prior to induction of oxidative stress with

250 M of H2O2 for 3 hours. Additionally, to determine whether lycopene could repair the

effects of ROS, lycopene was added after H2O2 treatment. Generation of ROS was determined

immediately using 5-(&6) chloromethyl-2’7’-dichlorodihydrofluorescein diacetate acetyl ester

dye, and MBNF were measured on day 17 of culture using the von Kossa technique. The present

findings demonstrated that the addition of H2O2 resulted in significantly increased generation of

ROS (p<0.001), which long-term resulted in a decreased number and area of mineralized bone

nodules (both: p<0.001). Pre- and post-treatment with 45:55 or 28:72 cis:trans lycopene resulted
175
in significantly lower ROS generation (p<0.001) and higher nodule area (p<0.05), compared to

treatment with H2O2 alone. This effect was not seen in vehicle or 5:95 cis:trans lycopene treated

cells. HPLC analysis revealed that cis:trans isomerization occurs in osteoblastic medium and that

the lycopene containing a high concentration of cis lycopene was better absorbed by osteoblast

cells. These findings support the hypothesis that the cis isomers of lycopene are capable of

preventing and repairing the damaging effects of H2O2-induced oxidative stress on the formation

of mineralized bone nodules. Further, these results provide evidence of a mechanistic effect of

lycopene in human osteoblast cells suggesting that the reported effects of lycopene on bone are

due to the potent antioxidant action of lycopene to prevent and repair the damaging effects of

ROS.

176
 INTRODUCTION

ROS are free radicals which, in the absence of adequate antioxidant defenses, can cause

oxidative damage to macromolecules resulting in oxidation of lipids, proteins and DNA. H2O2 is

a ROS which can be produced in vivo during normal cellular enzymatic processes, but is
38
particularly high during states of oxidative stress . H2O2 is an especially effective ROS

molecule, since it is permeable to cellular membranes and tissues and has a long half-life 63, 247. It

is also readily converted to the hydroxyl radical (OH) 247

, which in itself is highly reactive and

capable of causing extensive cellular damage.

Research suggests that oxidative stress induced by ROS causes bone loss and may

therefore contribute to the risk of osteoporosis 47-50

. The superoxide radical (O2_), which is

converted in vivo to H2O2, has been shown to enhance bone resorption through degradation of
53, 246
the bone matrix . Studies show that ROS affect both the bone resorbing cells, the

osteoclasts, and the bone forming cells, the osteoblasts. ROS can stimulate differentiation and
51, 248
bone resorption by osteoclasts and damage osteoblast cells to prevent normal growth and

development 56. High concentrations of ROS have also been shown to induce osteoblast death 52.

In particular, studies in mouse osteoblast cell cultures that have been challenged with
43, 56-58
H2O2 resulted in decreased cell growth, ALP activity and mineralization . Research also

suggests that in MC3T3-E1 osteoblast cells, a single exposure to H2O2 is enough to decrease

mineralization and gene expression of osteogenic markers such as ALP, BSP and Runx2, while

simultaneously increasing gene expression of transcription factors regulating antioxidant

enzymes 57, indicating a damaging state of oxidative stress.

Lycopene is an antioxidant with the highest singlet oxygen capacity among the

carotenoid family and an overall quenching ability that is about three times higher than -
117
carotene and ten times higher than -tocopherol . Owing to this potent antioxidant capacity,
177
numerous studies have demonstrated the ability of lycopene to decrease oxidative stress and the

risk of developing age-related chronic diseases 44, 84, 93.

The known potent antioxidant capacity of lycopene, together with the knowledge that

oxidative stress is an important risk factor for osteoporosis, has prompted research investigating

the relationship between lycopene and osteoporosis. We were the first to show that in

postmenopausal women, a high serum lycopene is associated with lower bone resorption markers
132
. Since that time, a high intake of lycopene has been associated with a positive 4 year change

in lumbar spine BMD 78 and a decreased risk for hip and non-vertebral fracture 73. Furthermore,

serum concentrations of lycopene have been shown to be significantly lower in women with

osteoporosis compared to age-matched, healthy controls 66, 103.

In a murine model of osteoporosis, animals who were fed lycopene had significantly

increased trabecular number and decreased trabecular separations, resulting in a significantly

inhibited loss of bone in the proximal femur and tibia, compared to control-fed animals 131. There

are a limited number of in vitro studies examining the effects of lycopene on osteoblasts and

osteoclasts. Lycopene has been shown to inhibit the formation of osteoclasts and their production

of ROS, resulting in an overall inhibition of resorption in the presence or absence of PTH 128, 129.

In both human and mouse osteoblasts, lycopene significantly increased cell number, ALP
130 77
activity , and gene expression of BSP , suggesting that lycopene stimulates osteoblastic

differentiation.

To date, the lycopene used for in vitro studies was comprised primarily of the all-trans
85
form of lycopene similar to those found in lycopene-containing foods. However, in vivo, the
197
all-trans isomer of lycopene is not as well absorbed as the cis isomer . Indeed, the isomeric

forms found in high concentrations in human serum are the cis isomers, which have the higher
44, 116
antioxidant capacity . It has been reported that the concentration of cis lycopene is as high

178
as 50-88% in human serum and tissues 84, 85, 92. This high concentration in human serum may be
197
due to isomerization of all-trans to cis during digestion . It may also be due to preferential

absorption of cis isomers. Cis isomers have less of a tendency to aggregate and form crystals

thus resulting in smaller particles that may be more easily absorbed 85. In addition, cis isomers

are more soluble in the lipophilic phase than the all-trans isomer, resulting in a more efficient
111
incorporation into micelles during digestion . Despite the fact that lycopene present in serum

and tissues has a higher concentration of cis isomers, the majority of current in vitro studies on

lycopene continue to apply lycopene containing primarily all-trans, which may affect the ability

to extrapolate research findings to true biological situations.

This chapter presents results from in vitro studies on human clonal CD34+ osteoblast

cells using varying isomeric ratios of cis:trans lycopene to determine whether specific lycopene

isomers were capable of preventing and repairing the damaging effects of H2O2-induced

oxidative stress.

MATERIALS AND METHODS

Materials

Lycopene was generously donated by LycoRed Ltd., Israel (5:95) and WaterSolutions, Inc.

(45:55 and 28:72). Lycopene from WaterSolutions was provided in a solution containing

tetrahydrofuran (THF) with 0.05% BHT as a vehicle, and the crystalline lycopene from LycoRed

Ltd. was dissolved in a similar manner. Table 9.1 illustrates the isomeric content of the types of

lycopene used in this study (information provided by WaterSolutions Inc., ID, USA).

Concentration of lycopene in THF was determined using the standard spectrophotometric

217
method . Sterile stock solutions of lycopene isomers were created just prior to use by

179
dissolving the original lycopene in medium at the required concentration, vortexing for 1 minute

at high speed, and then incubating the solution at 37ºC for 15 minutes.

Ham’s F-12 medium was purchased from Central Technical Services, University of

Toronto (Toronto, ON, Canada). Fetal bovine serum was purchased from CanSera, International

(Rexdale, ON, Canada). Antibiotic-antimycotic was purchased from Gibco (Burlington, ON,

Canada). Trypsin was purchased from Worthington Biochemical Corporation (Lakewood, NJ,

USA). Unless otherwise specified, all other materials were purchased from Sigma-Aldrich

Canada (Oakville, ON, Canada).

Table 9.1: Isomeric content of the three types of lycopene used in this study as determined by
HPLC, showing total cis and all-trans and the percentage composition of cis isomers.

Type of lycopene (isomeric Lycopene isomer composition (%)

ratio) all- total 5-cis 9-cis 13- 15-cis other-
trans ci s ci s ci s
45:55 55.05 44.95 26.92 3.87 3.93 1.24 8.99
28:72 71.84 28.16 18.46 2.96 2.99 0.0062 3.74
5:95 94.93 5.07 2.46 0.067 1.51 0.25 0.783

Cell culture

Human haematopoietic cells with osteoblast-like characteristics (CD34+) cells were purchased

from Lonza (formerly Cambrex, Walkersville, MD, USA). The cells were then cloned by

limiting dilution to obtain various clones that have the property to mineralize in culture in the

presence of -glycerophosphate and dexamethasone (Rao, L.G., personal communication). These

cloned CD34+ cells were used for this study. Cells were maintained in a 5% CO2 incubator at

37ºC in 75 cm2 flasks. Medium supporting osteogenic differentiation was comprised of the

following: Ham’s F-12, 10% FBS, 28 mM HEPES buffer, 1.1 mM calcium chloride, 2.0 mM
249
glutamine and 1% antibiotic-antimycotic . Cells were passaged weekly by trypsinization as
249
previously described and were plated in 24-well dishes for analysis of MBNF and cell

180
viability (5000 cells per well), or 96-well dishes for analysis of cell number (1000 cells/well) or

ROS (3000 cells/well). Medium was supplemented with 50 g/mL ascorbic acid phosphate and

0.1 M dexamethasone and on day 8 of culture, 10 mM of -glycerophosphate was also added.

To determine the effects of H2O2 and lycopene on osteoblast cells, three different treatment

schedules were performed as follows:

(Treatment 1) To determine the effects of varying concentrations of H2O2, cells were

treated on day 8 of culture with H2O2 in the range of 0-500 M, with sterile water as a

vehicle, for a period of three hours. Following this induction of oxidative stress, medium

was removed and replaced with fresh medium free of treatments.

(Treatment 2) To determine whether lycopene could prevent the effects of ROS, cells

were pre-treated with 1 M of 45:55, 28:72 or 5:95 cis:trans lycopene or THF vehicle on

day 8 for a period of 48 hours. Medium was removed and replaced with medium

containing 250 M H2O2 for a period of three hours. Following this induction of

oxidative stress, medium was removed and replaced with fresh medium free of

treatments.

(Treatment 3) To determine whether lycopene could repair the effects of ROS, cells

were treated on day 8 with 150-300 M H2O2 for a period of three hours. Following this

induction of oxidative stress, medium was removed and replaced with fresh medium

containing 1 M of 45:55, 28:72 or 5:95 cis:trans lycopene or THF vehicle. After 48

hours, medium was removed and replaced with fresh medium free of treatments.

Cell viability and cell number

Cell viability was measured using the alamarBlue® Assay (BIOSOURCE™, Invitrogen,

Canada) according to the manufacturer’s protocol. In brief, dye was added to the medium at 10%

of the culture volume and cells were placed in the CO2 incubator for 3 hours. Following
181
incubation, medium was transferred to a 96-well plate and the absorbance was measured at

wavelengths of 520 nm and 640 nm (Titertek Multiskan MCC/340, Huntsville, AB, USA).

Percent reduction of the dye (indicating growth and viability of cells) was determined according

to the manufacturer’s protocol. Cell viability was expressed as percent negative control. To

determine the effect of H2O2 and/or lycopene on cell viability this assay was performed

immediately after the 3 hour treatment period with H2O2 (Treatment 1 and 2 above) or

immediately after the 48 hour lycopene treatment period (Treatment 3 above).

250
Cell number was determined using the Methylene Blue method . Dye was comprised

of 1% methylene blue dissolved in 0.01 M borate buffer, pH 8.5. On day 13 of culture, cells were

fixed overnight at 4ºC with 4% paraformaldehyde. Following fixation, cells were washed three

times with phosphate buffered saline (PBS) (pH 7.4). Cells were stained with dye for 5 minutes

and rinsed 3 times with 0.01 M borate buffer, followed by solubilisation of the dye that stained

the nuclei with ethanol: 0.01 M HCl at a ratio of 1:1. Optical density was measured at a

wavelength of 620 nm (Titertek Multiskan MCC/340, Huntsville, AB, USA) and data were

expressed as percent negative control.

Generation of ROS

Generation of intracellular ROS was measured using the fluorescent probe 5-(and

6)chloromethyl—2’7’-dichlorodihydrofluorescein diacetate acetyl ester (CM-H2DCFDA)

(Molecular Probes™, Invitrogen, Canada) according to the manufacturer’s protocol and

251
previously published methods , with minor modifications. The fluorescent probe was

desiccated with 100% high-grade ethanol then diluted with Kreb’s Ringer Buffer (KRB) for

addition to cells at a final concentration of 25 M. Cells were incubated with the dye for 20

minutes at room temperature under reduced light conditions and replaced with fresh KRB. To

determine the effects of H2O2 on generation of ROS (Treatment 1 above), baseline measurement

182
of fluorescence was taken at an excitation wavelength of 485 nm and an emission wavelength of

527 nm (Fluorskan Ascent, Thermo Scientific, Waltham, MA, United States). H2O2 was then

added in concentrations of 150, 200, 250 and 300 M and the fluorescence was measured

kinetically for 2.5 hours. To determine whether pre-treatment with lycopene could inhibit the

generation of ROS by H2O2 (Treatment 2), cells were pre-treated with 1 M of each type of

lycopene for 48 hours, at which point the above method was repeated. Data were expressed in

relative fluorescence units (RFU). To determine whether lycopene could decrease the generation

of ROS in H2O2-treated cells (Treatment 3), cells which had been pre-treated with H2O2 were

processed according to the above method. Following a baseline measurement of fluorescence,

lycopene was added at a concentration of 1 M and cells were incubated for 3 hours; at which

point fluorescence was measured again (data expressed as percent negative control).

Quantification of mineralized bone nodule formation (MBNF)

Cells were stained in situ to determine MBNF using the von Kossa technique as previously

described 249. In brief, on day 17 of culture, cells were washed twice with PBS (pH 7.4) and fixed

with 4% paraformaldehyde. Fixed cells were kept at 4ºC until the time of staining, but no less

than 12 hours. Following fixation, cells were washed once with PBS and twice with double

distilled water (ddH2O). Cells were stained with 0.5 ml of 5% silver nitrate and exposed to

ultraviolet light for one hour. Following 3 more washes with ddH2O, 0.5 ml of 5% sodium

thiosulphate was applied for a period of 5-10 minutes and the cells were washed again. 50%

glycerol was applied to each well and MBNF were quantified according to number and total area

of nodules per well using the FluorChem 8900 image analyzer (Alpha Innotech Corporation, San

Leandro, CA, USA). Data were expressed as percent negative control.

Determination of medium and cellular lycopene by HPLC

183
For determination of cellular lycopene, cells were plated in 75 cm2 flasks at concentrations of

300 000 cells/flask. On day 8 of culture, cells were treated with 1 M of lycopene. On day 10, 48

hours after the application of lycopene, medium was collected and cells were harvested by

trypsinization. The resulting cellular pellet was disrupted by the addition of saturated sodium

hydroxide and ethanol containing BHT and incubated overnight at 37ºC. 100 L each of ethanol

and methanol were added and the remainder of the procedure was performed as outlined below.

Medium and cellular uptake of the three types of lycopene used was determined by
178
HPLC analysis as previously described, with minor modifications . To ensure that lycopene

was dissolved in the medium prior to application to cells, stock solutions of 1 M were prepared

directly in medium as described above. 2-propanol was added in a ratio of 1:1 and lycopene was

extracted using hexane: methylene chloride (5:1) with BHT and evaporated under a stream of

nitrogen gas. The residues were reconstituted in 100 L ethanol-BHT (0.0625%) and internal

standard (echinenone, CaroteNature, Switzerland). Lycopene remaining in the medium of cells

treated for 48 hours with lycopene was extracted by this same method.

HPLC was carried out using the Waters 2690 Alliance HPLC System and a Waters 996

PDA detector (Milford, MA, USA). The analytical column used was a Waters Spherisorb 3 M

ODS2, 4.6 x 250 mm. Separation was performed by elution with a solvent containing acetonitrile

and methanol (65:35, v/v) and 0.065% triethylamine, at a flow rate of 1.5 mL per minute.

Lycopene was analyzed at a wavelength of 450 nm and was determined using external standard

calibration curves on the Waters Millennium data management software, 4.0 edition (Milford,

MA, USA).

Cellular protein concentration was determined using Coomassie Blue dye reagent (Bio-
130
Rad, Mississauga, Ontario, Canada) as previously described , and cellular lycopene

184
concentrations were normalized to the µg protein content of the matching sample (units of

cellular lycopene were expressed in M/g protein).

Statistical analyses

Data were expressed as mean ± SEM, with at least 4 replicates in each group generated from 2 or

more experiments, with the exception of the HPLC analysis, which was performed in duplicate.

Differences between the treatment groups were measured using one-way ANOVA and Tukey’s

multiple comparison test for post hoc analysis, using GraphPad Prism version 5.02 for Windows

(GraphPad Software, San Diego, CA, USA). Significance was considered at p<0.05.

RESULTS

Effects of H2O2 on cell viability, generation of ROS and the formation

of mineralized bone nodules (Treatment 1)

To determine the concentrations of H2O2 to use during experimentation which would induce

oxidative stress without damaging the growth and function of CD34+ cells, cellular viability was

analyzed at concentrations of H2O2 ranging from 5-500 M. Addition of H2O2 for three hours

resulted in significantly decreased cell viability at concentrations of 350-500 M H2O2 (p<0.001,

Table 9.2). Lower concentrations of 5-100 M H2O2 had no effect (data not shown). Since

concentrations of 350 M H2O2 or higher significantly decreased cellular viability, those

concentrations were excluded from further analyses.

Treatment with H2O2 resulted in significantly higher production of ROS, a sign of

oxidative stress. After only 5 minutes, cells treated with concentrations of 150-300 M H2O2

inclusive, demonstrated a significantly higher generation of ROS than that generated by the

control cells (p<0.001, Figure 9.1). This effect continued at every time point measured (p<0.001)
185
up to 150 minutes, where the fluorescence values (in RFU) for H2O2–treated cells were as

follows: 150 µM H2O2, 59.8 ± 2.2, 200 µM H2O2, 63.9 ± 3.2, 250 µM H2O2, 80.7 ± 7.1 and 300

µM H2O2, 73.1 ± 6.8. These values were significantly higher than the control value of 38.1 ± 2.5

RFU (all: p<0.001). Although there appeared to be a dose-dependent increase in the generation

of ROS after 90 minutes of incubation with H2O2, there was no significant difference in the

generation of ROS among different concentrations of H2O2 at 90-150 minutes, suggesting that

intracellular induction of oxidative stress occurred maximally at the lower concentration of 150

M H2O2.

Treatment with H2O2 to induce oxidative stress, resulted in significantly decreased

number (concentrations 250-300 M, p<0.001) and area (concentrations of 150-300 M,

p<0.001) of mineralized nodules (Figure 9.2). Furthermore, an inhibitory dose-response effect on

total nodule area was demonstrated at higher concentrations of H2O2 from 200 to 300 M (Figure

9.2). The use of a continuous, lower dose of H2O2 (50 or 75 µM) resulted in similar effects on

MBNF (Figure 9.3) as seen from a one-time high dose of H2O2 (Figure 9.2). Continuous

treatment with 50 or 75 M resulted in both significantly decreased number and area of nodules

(p<0.001, Figure 9.3), which was comparable to the effects seen with a one-time high dose of

250-300 M (p<0.001, Figure 9.2).

186
Table 9.2: The effect of varying concentrations of H2O2 (150-500 µM) on cell viability after 3
hour treatment. Following H2O2 treatment, alamarBlue® was applied and cell viability is
expressed as % reduced dye in treated cells compared to untreated control cells (N8 for all
treatments).

Concentration H2O2 (M) Cell viability (% control) ± SEM

0 100.00 ± 2.72
150 100.9 ± 6.27
200 104.8 ± 5.85
250 94.25 ± 4.00
300 92.96 ± 7.65
350 72.00 ± 3.85a
400 72.73 ± 2.29 a
450 70.52 ± 2.65 a
500 73.34 ± 3.39 a
a
Significantly lower than control (p<0.001)

100
* Control
150 M H2 O2
80
*
200 M H2 O2
*
Fluroscence (RFU)

250 M H2O2
*
60 300 M H2 O2
*
*
*
40
*
*
20

0
0 25 50 75 100 125 150 175
Time (minutes)

Figure 9.1: Cells were stained with CM-H2DCFDA dye and oxidative stress was induced by the
addition of varying concentrations H2O2 (150-300 M). Production of ROS was then determined
at various periods of time up to 150 minutes. Values are mean  SEM and were compared to
control cells using one-way ANOVA (*p<0.001). N8 for all treatments.

187
 150

125

Number of nodules
(% control)
100

75 *
50

25
*

0
0 50 100 150 200 250 300 350
Concentration H 2O2 (M)

150

125
Total nodule area
(% control)

100

75 *
50

25

0
0 50 100 150 200 250 300 350
Concentration H 2O2 (M)

Figure 9.2: Cells were treated for 3 hours with varying concentrations of H2O2 to induce
oxidative stress on day 8 of culture and analyzed for MBNF using the von Kossa technique on
day 17. Values are mean  SEM and were compared to control cells using one-way ANOVA
(*p<0.001). N8 for all treatments.

125 125

100 100
Total nodule area
Nodule number

(% control)
(% control)

75 75
* *
50 50

25 25

0 0
0 20 40 60 80 0 20 40 60 80
Concentration H 2O2 (M) Concentration H 2O2 (M)

Figure 9.3: Cells were treated continuously with 50 or 75 M H2O2 from day 8 to 17 of culture
and were analyzed for MBNF using the von Kossa technique on day 17. Values are mean  SEM
and were compared to control cells using one-way ANOVA (*p<0.001). N8 for all treatments.

188
Cellular uptake of different lycopene isomers

Table 9.3 shows that the concentrations of lycopene detected in the medium by HPLC were not

statistically different to what was dissolved in the medium (compared to 1 M, t-test not

significant, Table 9.3), nor were the concentrations of the total lycopene in medium statistically

different from each other (45:55, 1.02 ± 0.03 M, 28:72, 1.08 ± 0.36 M, 5:95, 1.24 ± 0.61 M).

However, the cis lycopene as detected in the medium was higher than that detected in the

original lycopene (in THF), which suggests that all-trans to cis isomerization may have occurred

within the medium. The cis concentration increased to the following ratios of cis:trans upon

dissolution in the medium: 28:72 increased to 58:42, 45:55 increased to 55:45, and 5:95

increased to 21:79.

After 48 hours, the concentration of total lycopene remaining in the medium of cells

treated with 5:95 cis:trans lycopene was significantly lower than the lycopene remaining in the

medium of cells treated with 45:55 (p<0.05) or 28:72 (p<0.01). Cellular uptake of all-trans

lycopene was similar for cells treated with all three isomeric forms of lycopene (45:55, 3.22 ±

0.47 x 10-3 M/g, 28:72, 3.46 ± 1.06 x 10-3 M/g, 5:95, 2.84 ± 0.68 x 10-3 M/g, Table 9.3).

The uptake of cis lycopene was significantly lower for cells treated with 5:95 compared to cells

treated with 45:55 or 28:72 with average cellular uptake being 0.10 x 10-3 ± 0.10 M/g, 2.56 x

10-3 ± 0.44 M/g and 3.01 x 10-3 ± 0.79 M/g, respectively (both: p<0.05, Table 9.3). This

significantly higher uptake of cis lycopene in cells treated with either 45:55 or 28:72 lycopene

resulted in a non-significantly higher uptake of total lycopene overall compared to cells treated

with 5:95 lycopene.

189
Table 9.3: Concentration of lycopene present as determined from HPLC analysis (N2 for
each).

Lycopene concentration as determined from Lycopene isomer added to medium

HPLC (M) (1 M total)
45:55 28:72 5:95
Medium: as added to cells at the start all- 0.46 ± 0.05 0.45 ± 0.15 0.98 ± 0.38
of incubation (day 8) trans
total 0.56 ± 0.02 0.63 ± 0.21 0.26 ± 0.26
cis
Total 1.02 ± 0.03 1.08 ± 0.36 1.24 ± 0.61
Medium: present after 48 hours of all- 0.18 ± 0.02 0.19 ± 0.01 0.089 ± 0.04
incubation (day 10) trans
total 0.17 ± 0.03 0.17 ± 0.01 not detected
cis
Total 0.35 ± 0.01 0.36 ± 0.00 0.089 ± 0.04a
Cells: lycopene uptake after 48 hours all- 3.22 ± 0.47 3.46 ± 1.06 2.84 ± 0.68 x
of incubation (day 10)1 trans x 10-3 x 10-3 10-3
total 2.56 ± 0.44 3.01 ± 0.79 0.10 ± 0.10b x
cis x 10-3 x 10-3 10-3
Total 5.79 ± 0.91 6.47 ± 1.86 2.94 ± 0.70 x
x 10-3 x 10-3 10-3
1
Cellular lycopene concentration was normalized according to individual g protein level and was expressed as
M/g.
a
After 48 hours, concentration of total lycopene remaining in cellular medium treated with 5:95 was significantly
lower than that in cells treated with 45:55 (p<0.05) and 28:72 (p<0.01).
b
Uptake of total cis lycopene in 5:95 treated cells was significantly lower than 45:55 and 28:72 treated cells (p<0.05
for both).

190
Ability of lycopene to prevent damaging effects of ROS on formation

of mineralized bone nodules (Treatment 2)

As demonstrated in Figure 9.1, addition of H2O2 resulted in significantly higher generation of

ROS than in control cells at all time points (p<0.001, from 5 to 150 minutes). However, in cells

pre-treated with 45:55 and 28:72 cis:trans lycopene, followed by the induction of oxidative stress

with 250 M H2O2, there was a significantly lower generation of ROS than those treated with

H2O2 alone (p<0.05, from 90 to 150 minutes). Pre-treatment with either 45:55 and 28:72

cis:trans lycopene, followed by induction of oxidative stress with 250 M H2O2, resulted in

significantly lower generation of ROS than that of cells pre-treated with vehicle prior to 250 M

H2O2 (p<0.05 at 150 minutes, Figure 9.4). There were no significant effects in cells pre-treated

with vehicle.

Induction of oxidative stress with H2O2 resulted in a significantly decreased area of

nodules. Additionally, total nodule area was significantly lower in cells treated with H2O2

compared to negative control, THF, 45:55 or 28:72 cis:trans lycopene treated cells (p<0.05,

Figure 9.5). Treatment with 45:55 or 28:72 cis:trans lycopene alone resulted in a significant

increase in the area of mineralized nodules (p<0.05), an effect not seen in vehicle-treated cells.

This beneficial effect of lycopene is also demonstrated in cells pre-treated with 45:55 and 28:72

cis:trans lycopene, prior to the induction of oxidative stress with H2O2, where there was a

significantly higher area of nodules than those treated with H2O2 alone (p<0.05) or those pre-

treated with vehicle (p<0.01 compared to pre-treatment with 45:55 and p<0.05 compared to pre-

treatment with 28:72 cis:trans, Figure 9.5). There was no such effect in cells pre-treated with

vehicle (Figure 9.5). There were no significant effects on the number of nodules (data not

shown).

191
 125
Control
Vehicle
100 45:55 lycopene
28:72 lycopene
Fluoroscence (RFU)

H2 O2
75
*
45:55 lycopene + H2 O2
*
28:72 lycopene + H2 O2
* Vehicle + H2 O2
50

25

0
0 25 50 75 100 125 150
Time (minutes)

Figure 9.4: Cells were pre-treated with lycopene for 48 hours (days 8-10 of culture), then
stained with CM-H2DCFDA dye. Oxidative stress was then induced by the addition of 250 M
of H2O2 and the generation of ROS was determined kinetically from 1 to 150 minutes. Values
are mean  SEM and were compared using one-way ANOVA (*p<0.05). N8 for all treatments.

192
 200 
Total nodule area (% control)

Control

* * Vehicle
150
45:55 lycopene
28:72 lycopene
H2 O2
100  Vehicle + H2 O 2
45:55 lycopene + H 2 O2
50 28:72 lycopene + H 2 O2

0
Treatment Group

Figure 9.5: Cells were pre-treated for 48 hours with different lycopene isomers (day 8-10),
followed by a one-time high dose of 250 M of H2O2 (day 10) to induce oxidative stress. On day
17, cells were analyzed for MBNF using the von Kossa technique. Values are mean  SEM and
were compared using one-way ANOVA ( p<0.05 compared to negative, vehicle and/or
lycopene controls and *p<0.05 compared to H2O2-treated). N8 for all treatments.

Ability of lycopene to repair damaging effects of ROS on formation of

mineralized bone nodules (Treatment 3)

As demonstrated above, treatment with 200 and 250 M H2O2 resulted in significantly increased

generation of ROS compared to control (Figure 9.1). Conversely treatment with lycopene alone

significantly decreased ROS generation (p<0.001, Figure 9.6). The addition of 1 M 45:55 or

28:72 cis:trans lycopene, after H2O2-induced oxidative stress resulted in significantly lower

generation of ROS compared to cells treated with H2O2 only (p<0.001, Figure 9.6).

Treatment with H2O2 resulted in a significant decrease in the total number (250-300 M.

p<0.05, Figure 9.7) and area of mineralized nodules (200-300 M H2O2, p<0.01, Figure 9.8).

However, treatment with 45:55 or 28:72 cis:trans lycopene after the induction of oxidative stress

193
with H2O2 resulted in a significantly higher total number (at 200 M H2O2, p<0.05, Figure 9.7)

and area of mineralized nodules (at 200 and 250 M H2O2, p<0.05, Figure 9.8). In fact, at these

concentrations of H2O2, treatment with lycopene after the induction of oxidative stress resulted in

total nodule number and area which was similar to those seen in control cells. There was no

effect of vehicle treatment (Figures 9.7 and 9.8) after the induction of oxidative stress.

There was no significant effect on cell viability, as determined immediately after

Treatments 2 and 3, or on cell number, as determined on day 13 of culture (data not shown). In

addition, there was no significant effect of 5:95 cis:trans lycopene as added in Treatments 2 and

3.

125

100

* *
** **
ROS (% control)

75 **
**

50

25

0
ol

e
e

M 2

2
2O

2O

2O

2O

2O

2O

2O
2
en
cl

en

2O
tr

hi
on

H
p
p

H
Ve

co
co

M
C

M


0

0

0

0

0
ly
ly

00
0

0
5

20
20

25

25

25
20

25
:5

:7

2
45

28

+
+
+

+
h.

2
5
h.
5

:7

:7
:5
Ve

:5

Ve
28

45

28
45

Treatment group

Figure 9.6: Cells were pre-treated with either 200 or 250 µM H2O2 for 3 hours on day 8 of
culture to induce a state of oxidative stress and were then stained with CM-H2DCFDA dye. 1 µM
45:55 or 28:72 cis:trans lycopene was added and the production of ROS was determined after 3
hours. Values are mean % control  SEM and were compared using one-way ANOVA
(*p<0.001 compared to control and **p<0.001 compared to H2O2-treated). N10 for all
treatments.

194
 125
** Control
Total nodule number

45:55 lycopene
100
28:72 lycopene
(% control)

Vehicle
75

50

*
25
0 50 100 150 200 250 300 350 400
Concentration H 2O2 (M)

Figure 9.7: Cells were pre-treated with 150-300 µM H2O2 for 3 hours on day 8 of culture to
induce a state of oxidative stress, followed by treatment with 1 µM of 45:55 or 28:72 cis:trans
lycopene for a period of 48 hours. Cells were analyzed for MBNF using the von Kossa technique
on day 17 of culture. Values are mean % control  SEM and were compared using one-way
ANOVA (*p<0.05 for H2O2 compared to control and **p<0.05 for post-treatment with lycopene
compared to H2O2-treated). N=8 for all treatments.
Total nodule area (% control)

150 **
Control
125 45:55 lycopene
28:72 lycopene
100 Vehicle

75

50

25 *
0 50 100 150 200 250 300 350 400
Concentration H 2O2 (M)

Figure 9.8: Cells were pre-treated with 150-300 µM H2O2 for 3 hours on day 8 of culture to
induce a state of oxidative stress, followed by treatment with 1 µM of 45:55 or 28:72 cis:trans
lycopene for a period of 48 hours. Cells were analyzed for MBNF using the von Kossa technique
on day 17 of culture. Values are mean % control  SEM and were compared using one-way
ANOVA (*p<0.01 for H2O2 compared to control and **p<0.05 for post-treatment with lycopene
compared to H2O2-treated). N=8 for all treatments.

195
 DISCUSSION

To summarize, this chapter reported the following findings:

1) H2O2 significantly increased the intracellular generation of ROS at concentrations of 150

to 300 M which, long-term, significantly decreased the number and area of mineralized

bone nodules, suggesting that H2O2-induced oxidative stress, which resulted in damaging

effects to CD34+ osteoblast cells.

2) Regardless of isomeric type of lycopene added to the cells, lycopene was detectable in

CD34+ osteoblast cells 48 hours after treatment. Uptake of cis lycopene by osteoblast

cells was significantly higher in cells treated with the 28:72 and 45:55 cis:trans lycopene

than cells treated with 5:95 lycopene, resulting in a non-significantly higher concentration

of total lycopene.

3) Pre-treatment with lycopene (28:72 and 45:55 cis:trans) prior to the induction of

oxidative stress with H2O2 resulted in a significantly lower generation of ROS and

subsequently, significantly higher area of MBNF, suggesting the lycopene was capable of

preventing the damaging long-term effects of ROS in osteoblast cells.

4) After the induction of oxidative stress with H2O2, post-treatment with lycopene (28:72

and 45:55 cis:trans) resulted in a significantly decreased generation of ROS and

subsequently, significantly increased number and area of MBNF, suggesting that

lycopene was capable of repairing the damaging long-term effects of ROS in osteoblast

cells.

5) There was no significant effect of 5:95 cis:trans lycopene, suggesting that this lower

antioxidant capacity lycopene was not capable of preventing or repairing the damaging

long-term effects of ROS in osteoblast cells.

196
 Previously published studies have used concentrations of 500 M H2O2 or higher to

induce oxidative stress and exert effects on function of osteoblast cells 56, despite the fact that it
52
has been shown that these higher concentrations of H2O2 caused osteoblast cell death . The

time period of H2O2 treatment used in these studies was also higher, ranging from 12-96 hours 52,
56
. However, more recent studies have demonstrated that just 3 hours of treatment with similar

doses of H2O2 was enough to induce ROS effects, without damaging or killing the osteoblasts 57,
252
. Since concentrations of 350 M or higher applied for just 3 hours were found in the present

study to significantly reduce the cellular viability in clonal human CD34+ osteoblast cells (Table
52, 56
9.2), lower concentrations were used than previously reported to induce ROS effects in

CD34+ cells. The present results showed that concentrations of 150-300 for just 3 hours

significantly increased intracellular induction of ROS and decreased long-term formation of

mineralized nodules (Figures 9.1 and 9.2). Use of a continuous, lower dose of H2O2 (50 or 75

M) resulted in similar effects on MBNF (Figure 9.3) as that from a one-time high dose of H2O2

(Figure 9.2), therefore the one-time high dose of H2O2 was used in the remaining experiments

(Figures 9.4-9.7). The present findings on treatment with a one-time high dose of 150-300 M

H2O2 for a 3 hour period support those of Arai et al., who reported that a single, 3 hour exposure

to 400 M significantly decreased mineralization in MC3T3-E1 osteoblast cells 57.

These results showed that treatment with lycopene for only 48 hours was enough to

prevent and repair the long-term damaging effects of ROS on osteoblasts, specifically MBNF.

Typically, antioxidants are added to osteoblast cultures continuously and for longer periods of

time to induce effects on mineralization and ALP activity 74, 253. However it has been shown that

antioxidants, particularly carotenoids, can affect gene expression and synthesis of proteins

important in osteoblast differentiation and mineralization, such as BSP, Runx2, ALP and 1(I)

collagen, after only 24-72 hours of treatment 74, 77, 253. This suggests that a short treatment period

197
with an antioxidant, such as lycopene, would be enough to prevent or undo long-term negative

effects on mineralization, as shown in the present study. Previous studies have shown that brief

pre-treatment (less than 24 hours) with antioxidants, such as metallothionein and Trolox, prior to

the induction of oxidative stress with H2O2, can inhibit the decreased ALP activity associated

with H2O2 treatment alone 56, 58. However the long-term effect on osteoblast differentiation after

such treatments, as determined by the assessment of MBNF, has not been shown prior to this

chapter.

The finding that all-trans lycopene isomerized when added to the medium prior to its use

in cell culture is in agreement with a previously reported study which suggests that it is possible

for lycopene to rapidly isomerize in solutions such as medium, to an equal mixture of cis:trans
254 92
. It is postulated that this equilibrium of lycopene isomers provides more stability . Not all

types of lycopene used in this study isomerized in the medium to an equal mixture of cis:trans.

The 28:72 cis:trans did increase to a ratio of 58:42, but the 5:95 cis:trans only increased to a

ratio of 21:79 cis:trans (Table 9.3). The 45:55 cis:trans remained relatively similar once in the

medium (a ratio of 55:45 cis:trans), but this could be because the concentration of cis isomers

was already quite high. Nevertheless, the concentrations of lycopene isomers detected in the

medium could explain the fact that similar positive effects were seen with the lycopene

containing the higher ratios of cis (28:72 and 45:55) (Figures 9.4-9.8), since the concentration of

cis in the medium added to cells was similar for both of these types of lycopene (detected as

58:42 and 55:45 cis:trans by HPLC, respectively, Table 9.3).

The concentration of lycopene in human tissues, such as kidney and ovary, has been well
93, 178
documented but has never been reported for bone tissue. Similarly, although there have
112, 254-256
been many published studies on cellular uptake of lycopene , there have been no

published studies on lycopene uptake by osteoblast cells. The uptake of cis was expected to be

198
greater than all-trans, since bioavailability studies suggest that cis lycopene is better absorbed 85,
111 112
, and research on cellular uptake shows that absorption of cis is higher than all-trans . As

shown in Table 9.3, the isomeric forms of lycopene containing higher concentration of cis

resulted in significantly higher uptake of cis by osteoblast cells. Overall, total lycopene uptake

was also higher in cells treated with 28:72 and 45:55 cis:trans than cells treated with 5:95

cis:trans (6.47 ± 1.86 x 10-3, 5.79 ± 0.91 x 10-3, 2.94 ± 0.70 x 10-3 M/g protein) although this

effect was not a significant one. It is possible that after 48 hours the small amount of cis

lycopene present in the 5:95 cis:trans had either already been absorbed and metabolized, or it

had been degraded. Indeed, some cellular uptake studies measured lycopene in 24 hours or less
255, 256
.
254
A lycopene uptake study by Liu et al. has shown that after 50 hours approximately

60% of lycopene remained in the medium of their treated human prostate cancer cells (LNCap).

However the present findings showed that after 48 hours the percent of total lycopene remaining

in the medium was lower, at approximately 30% remaining for cells treated with either 45:55 and

28:72 and 10% for cells treated with 5:95 cis:trans. This suggests that the lycopene in the

medium has either been degraded or that it has been metabolized at a higher rate by these CD34+
254
cells than that described for LNCap cells in the other study . This may explain the lower

concentration of cellular lycopene in cells treated with 5:95, as there was significantly less

lycopene remaining in this lycopene-treated medium after 48 hours (Table 9.3). It is possible that

the all-trans lycopene was degraded or metabolized faster than the cis. The HPLC method

currently used was not capable of detecting lycopene metabolites to determine whether or not

this was the case. In this study, uptake of lycopene in cells which were not exposed to oxidative

stress was examined. A future area of exploration which would be of interest would be to

199
determine the concentration of cellular lycopene after the induction of oxidative stress, to

determine just how much of the lycopene is utilized to quench the ROS in the osteoblast cells.

For the current studies on oxidative stress and mineralization, lycopene was added to

osteoblast cells at a concentration of 1 M, which is similar in concentration to that typically

257
found in the serum of healthy humans consuming a regular diet . In fact, data from the cross-

sectional study, in which serum from 108 women 25-70 years of age was assayed for lycopene

concentration, showed an average serum concentration of 1.3 M total lycopene (Chapter 3).

Research shows that isomeric ratios in human serum are generally about 50:50 cis:trans
92
lycopene . Although lycopene-containing foods are comprised of 79-95% of the all-trans
86
isomer of lycopene , due to preferential uptake, isomerization during digestion and more
85, 86, 111,
efficient absorption, the resultant serum concentrations of lycopene contain higher cis
112
. This is verified by data from the cross-sectional study (Chapter 3), in which the average

percent of lycopene isomers in the serum was 31.66 ± 0.33% 5-cis, 19.18 ± 0.44% other-cis and

49.16 ± 0.46% all-trans lycopene, conferring a ratio of approximately 50:50 cis:trans lycopene.

The findings reported in this study have shown that the lycopene isomeric forms which are most

similar to those concentrations found in human serum (conferring a 50:50 ratio of approximately

cis:trans) have significant effects on preventing and repairing the damaging effects of H2O2-

induced oxidative stress.

As shown in Table 9.3, the 45:55 and 28:72 cis:trans lycopene isomerized to similar

isomer concentrations of 55:45 and 58:42, respectively, in the medium. The fact that these two

types of lycopene with the higher concentration of cis isomers in the medium were more

effective in preventing and repairing the damaging effects of ROS is not surprising (Figures 9.4-

9.8). Not only are cis isomers present in higher concentrations in human serum and tissues in

vivo, but they also possess higher antioxidant capacity than the all-trans configuration 116. It has

200
been postulated that the order of antioxidant capacity for the various lycopene isomers is as
44
follows: 5-cis>9-cis>7-cis>13-cis>11-cis>all-trans . Conversely, minimal effects on

osteoblasts treated with primarily all-trans lycopene (5:95 lycopene) were found (data not

shown). Currently, studies in our laboratory are investigating whether a longer, continuous

treatment period of up to 9 days with all-trans lycopene (5:95) would be more effective in

preventing and repairing the damaging effects of ROS induced by H2O2. The present study

findings however, suggest that the ability of lycopene to prevent and repair the effects of ROS as

shown in these results can be attributed to the potent antioxidant capacity of the cis isomer.

In conclusion, these findings demonstrated that the lycopene isomeric forms which are

commonly found in high concentrations in human serum and exhibit the highest antioxidant

capacity were capable of preventing and repairing the damaging long-term effects of H2O2-

induced oxidative stress on the formation of mineralized bone nodules in osteoblasts. This is the

first report of the uptake and effects of varying lycopene isomers on human osteoblast cells.

These novel in vitro studies establish a basis for further in vitro research using cis lycopene

isomers to more accurately mimic true biological situations. This chapter provides evidence of a

mechanistic effect of lycopene in human osteoblast cells, suggesting that the reported effects of

lycopene on bone are due to the potent antioxidant action of lycopene to prevent and repair the

damaging effects of ROS at the cellular level.

201
 CHAPTER 10

General discussion and conclusions

202
 Summary of Experimental Findings

Lycopene is a potent carotenoid antioxidant with the highest singlet oxygen quenching ability

and the highest overall total antioxidant capacity among other carotenoids 36. Recently, lycopene

has received attention in the field of bone health. In vitro studies have shown that lycopene
77,128, 129,130
affects osteoblasts and osteoclasts , and clinical studies have suggested a positive
66, 103
effect of lycopene on bone . Previous studies in our laboratory have shown that a high

serum lycopene corresponds to a lower protein oxidation and NTx in postmenopausal women,

providing the first in vivo evidence that the effect of lycopene on bone may be due to its

antioxidant capacity 132, 258. However, studies specifically associating intervention with lycopene

to a decreased risk of osteoporosis have not yet been reported. In addition, the mechanisms by

which lycopene exerts a protective effect on bone have not yet been delineated. Therefore,

experiments were conducted which were designed to determine whether lycopene could act as an

antioxidant to decrease bone turnover markers and therefore, the risk of osteoporosis, in

postmenopausal women. Specifically, to determine whether direct supplementation with

lycopene would increase antioxidant capacity, resulting in decreased oxidative stress parameters

and bone turnover markers. These experiments also sought to explore whether cis lycopene,
116
which possesses the highest antioxidant capacity of the lycopene isomers , would provide the

most benefit to decreased oxidative stress parameters and bone turnover markers. It is possible

that interactions between PON1 polymorphisms and serum lycopene may moderate the effect of

lycopene on oxidative stress and osteoporosis risk. In order to delineate the potential mechanisms

by which lycopene acts to decrease osteoporosis risk, an investigation of polymorphisms in the

antioxidative enzyme PON1 were performed.

In summary, the key findings of this thesis, as they relate to the original hypotheses are as

follows:

203
I. The main source of lycopene in the diet of this female population was raw tomatoes. A

correlation between lycopene intake and serum lycopene concentrations was

demonstrated, indicating that dietary lycopene was well absorbed. Lycopene intake was

variable and the average daily intake was quite low (Chapter 3). The average daily intake

of 6.03  0.53 mg per day may not be enough to provide sufficient serum concentrations

of lycopene to decrease oxidative stress parameters and bone resorption markers.

II. Dietary restriction of lycopene for a period of one month resulted in an increase in

oxidative stress parameters in postmenopausal women. The antioxidant enzymes CAT

and SOD were significantly decremented as they acted to quench ROS while trying to

compensate for a lack of dietary antioxidants, while GPx was significantly increased in

an attempt to quench the lipid hydroperoxides generated during oxidative stress. These

factors contributed to an increase in NTx, suggesting that the compensatory mechanisms

are not sufficient to cope with the lack of quenching ability usually provided by dietary

lycopene. The findings described in Chapter 4 showed that the restriction of lycopene in

the daily diet increased oxidative stress parameters and the bone resorption marker NTx

in postmenopausal women, which may contribute towards an increased risk of

osteoporosis in this cohort of women.

III. Postmenopausal women supplemented with lycopene in the form of tomato juice or

capsules for a period of four months had significantly increased serum lycopene and total

antioxidant capacity, with a concomitant decrease in the oxidative stress biomarkers for

protein and lipid oxidation. This decrease corresponded to a significant decrease in the

bone resorption marker NTx, which may subsequently reduce the risk of osteoporosis.

Doses of lycopene given in 30 or 70 mg per day resulted in similar serum lycopene

concentrations, suggesting there was a maximal absorption of lycopene. Both doses

204
IV. Participants supplemented with lycopene (30-70 mg per day) had lower biomarkers of

lipid peroxidation and bone resorption than participants who obtained lycopene from

their usual daily diets (6.23  0.64 mg per day). This effect may be attributed to the

significantly greater concentration of the potent antioxidant 5-cis isomeric form of

lycopene in the serum of participants supplemented daily with tomato juice or lycopene

capsules (Chapter 6).

V. Due to the duration of the study period, there was no statistically significant effect of

lycopene intervention on the bone formation marker BAP. However, results of the in

vitro study presented in Chapter 9 demonstrated that lycopene was well absorbed in

osteoblasts, particularly the cis isomer. Further, the isomeric forms of lycopene

containing higher concentrations of cis lycopene were capable of significantly preventing

and repairing the damaging effects caused by the ROS H2O2. These effects were

manifested by a decreased generation of ROS and improved formation of mineralized

bone nodules and were directly attributed to the high antioxidant capacity of the cis

isomer (Chapter 9).

VI. There was a significant interaction between PON1 172TA and 584AG

polymorphisms and serum lycopene on bone turnover markers in women. In women with

the combined 172TT and 584G genotype there was a correlation between a high serum

lycopene concentration and low TBARS, BAP and NTx. This effect was not seen in

participants with other genotypes. These findings are the first to provide evidence that the

PON1 polymorphism interacts with lycopene to affect bone health in women (Chapter 7).

205
VII. Specifically, in postmenopausal women the 172TA polymorphism affected the change

in bone resorption markers in response to lycopene intervention. This was confirmed in

the clinical intervention study in which the 172TT genotype was associated with a better

response to lycopene supplementation than carriers of the 172A allele, as manifested by

significantly increased total antioxidant capacity and decreased NTx (Chapter 8).

VIII. Investigation of the 172TA polymorphism provided mechanistic evidence of lycopene

action in bone; lycopene supplementation significantly decreased lipid peroxidation,

which interacted with the PON1 polymorphism to decrease the bone resorption marker

NTx (Chapter 8). These findings suggest that lycopene intervention is most beneficial to

decrease bone resorption markers when internal antioxidant defenses are insufficient or

decremented.

Overall, the results presented in this thesis show the various mechanisms by which

lycopene acts to decrease oxidative stress parameters and bone turnover markers to prevent

osteoporosis. These findings support the hypothesis that lycopene may be beneficial in

decreasing the risk of osteoporosis, and supplementation with lycopene-rich foods, such as

tomato juice or capsules, provides greater benefit than is usually achieved through the daily diet

alone. The key findings of this thesis, as outlined above, are demonstrated in Figure 10.1. This

figure outlines a proposed mechanism of lycopene action, based on the experimental results

presented in Chapters 3 through 9 of this thesis, illustrating the various points of lycopene action

in the bone turnover pathway leading to osteoporosis. In brief, lycopene acts to prevent the build-

up of lipid and protein oxidation by-products which, if left unchecked, would increase bone

turnover in favour of bone resorption. Thus, lycopene prevents the loss of bone which can lead to

the development of osteoporosis.

206
 Intrinsic factors: Lifestyle factors:
aging Generation of smoking
menopause ROS strenuous exercise

Endogenous primary defenses

↑ serum cis lycopene = ↓ 30-70 mg Altered PON1, CAT, SOD, GPx

oxidative stress lycopene/day

Insufficient/decremented primary
defenses

↑ Lipid and protein oxidation by-

products cis lycopene

Prevents ROS damage

↑ Osteoclast differentiation &

resorption
↓ Osteoblast differentiation
& function

cis lycopene
↑ Turnover in favour of bone
Repairs ROS damage
resorption

30-70 mg lycopene/day

↑ serum cis lycopene = ↓ NTx

Loss of bone mass

OSTEOPOROSIS

Figure 10.1: Summary diagram of combined experimental findings from this thesis. Red arrows

illustrate points of lycopene action as determined from experimental results shown to prevent the

development of osteoporosis.
207
 Discussion of experimental findings and their relevance to

the hypotheses and the current body of knowledge

I. Findings of this thesis contribute new clinical knowledge on

how lycopene affects bone and demonstrate the importance of

lycopene consumption to decrease bone turnover markers.

This thesis is the first to provide direct clinical evidence that intervention with lycopene can

decrease oxidative stress parameters and bone resorption markers, and may subsequently

decrease the risk for osteoporosis in postmenopausal women. To date, other epidemiological

studies were either cross-sectional or prospective analyses and have focussed on lycopene intake

and/or serum lycopene concentrations as obtained through the usual daily diet. One of the main

limitations of these previous studies includes the fact that often serum lycopene was not
73, 78, 80
measured and used as part of the analyses . As shown repeatedly throughout this thesis,

concentrations of cis isomers of lycopene appear to be largely responsible for the effects on

oxidative stress parameters and bone resorption and formation markers (Chapters 4-6 and 9).

However, differences in bioavailability and/or absorption of lycopene from the diet may

introduce the potential for error or inaccuracy in analyses of the effect on bone. In addition, there

is an acute lack of a standardized method by which to measure dietary lycopene intake and it

therefore has to be calculated manually for each food using currently published references on

lycopene content in foods. An example is the lycopene content database provided by the USDA
89
. However, dietary lycopene contained in foods can vary greatly, depending on a variety of

factors including type of processing, cooking methods and the ripeness and cultivar of raw

tomatoes 86, 168. A heavy reliance on participants to recall frequency of consumption of lycopene,

particularly after a long period of time, also limits accuracy. Without quantification of serum

208
lycopene concentrations to support the daily intake data provided by participants, there may be a

wide margin of error in clinical study findings.

Furthermore, as demonstrated by the lycopene intake data provided in Chapter 3, the

usual daily intake of 6.03  0.53 mg of lycopene was significantly lower than the quantity
96-99
usually given in intervention trials . In fact, 22.3% of participants consumed what was

considered to be a negligible lycopene intake (<1.0 mg consumed/day), while only 8.9%

consumed greater than 15 mg/day (Chapter 3). The low intake of lycopene as shown in this

participant population may not be enough to provide the significant benefit of decreased

oxidative stress parameters and bone resorption markers (Chapter 5 and 6). Therefore, this thesis

provides incentive for daily supplementation with foods rich in lycopene, such as tomato juice or

capsules, to elevate daily intake and subsequent serum concentrations.

The clinical intervention study described in this thesis is the longest reported intervention

study to date in which the daily dose of lycopene provided was  30 mg of lycopene per day 95,
96
. The consumption of lycopene-rich tomato juice, providing 70 mg of lycopene per day for a

period of 4 months was well tolerated and did not demonstrate any adverse effects. Further, the

serum concentrations at this high dose of 70 mg demonstrate that there is an absorption

saturation of lycopene and that a daily dose of 30 mg may be sufficient in order to obtain the

associated beneficial effects of lycopene such as decreased oxidative stress.

II. Findings in this thesis provide evidence that it is the antioxidant

capacity of lycopene which is responsible for the effects on bone.

This thesis provides evidence that it is the antioxidant capacity of lycopene which decreases

oxidative stress parameters and may consequently decrease the risk of osteoporosis in women.

Specifically, the present studies suggest that it is the 5-cis isomer of lycopene which is largely

responsible for this effect. The 5-cis isomer of lycopene possesses the highest antioxidant
209
capacity of all of the lycopene isomers 44. At the in vitro level, this thesis has shown that the cis

isomer was best absorbed by human CD34+ osteoblast cells and demonstrates a significant

capacity to both prevent and repair the damaging effects of ROS on the formation of mineralized

bone nodules, and thus bone formation by osteoblasts (Chapter 9). This effect was not seen when

a high concentration of all-trans was used, suggesting that the beneficial effects of lycopene can

be attributed to the higher antioxidant capacity of the cis lycopene.

At the in vivo level, supplementation with lycopene resulted in significantly increased

total antioxidant capacity which corresponded to significantly lower oxidative stress parameters

and the bone resorption marker NTx. The increase in total antioxidant capacity may be due to an

overall increase in all of the isomers of lycopene, as well as -carotene, and other vitamins and

polyphenols contained in the supplements. Of particular interest, were the significantly increased

concentrations of 5-cis serum lycopene (p<0.001, Chapter 5). The supplements used in this study

provided 1.20-14.20 mg of cis lycopene per day, and in particular, the tomato lycopene capsules,

resulted in the highest serum concentration of 5-cis lycopene. This verifies many reports that the

matrix in which lycopene is consumed enhances the uptake of cis isomers 85, 212, 218.

Participants with a high daily intake of lycopene consumed on average, more cooked and

processed tomato products (p<0.0001), which usually contain higher amounts of lycopene

(Chapter 3). The consumption of cooked/processed lycopene-containing foods showed a

significant, positive correlation with total serum cis lycopene (p<0.0001), that the consumption

of raw lycopene products did not (Chapter 3). These data suggest that in order to increase serum

concentrations of the antioxidant-rich 5-cis lycopene, an increased consumption of foods rich in

bioavailable lycopene, such as processed tomato products, should be considered. However, a

conscious consumption of these foods on a daily basis may not provide as much benefit as that

given by daily supplementation. As shown in Chapter 6, those participants with a daily intake of

210
11.83  0.89 mg of lycopene per day had corresponding 5-cis serum concentrations that were

16.3% lower than those found in lycopene-supplemented participants, which resulted in 22.0%

higher TBARS and 15.3% higher NTx than participants supplemented with 43.33  2.84 mg per

day of lycopene. Taken together, these findings suggest that daily supplementation with a

lycopene-rich food, such as the tomato lycopene capsules or juices provided in this experimental

work, is necessary to increase serum concentrations of the potent antioxidant 5-cis lycopene.

This increase in 5-cis serum lycopene is directly attributed to its ability to decrease oxidative

stress parameters and bone turnover markers and may consequently decrease the risk of

osteoporosis, as shown in the clinical and in vitro experimental results presented in this thesis.

III. Findings in this thesis provide evidence of a mechanism by

which lycopene acts on bone that is moderated by genetic variation in

the paraoxonase 1 enzyme.

Individual responses to nutraceutical intervention can vary greatly depending on genetic factors

which influence how the nutrient is digested, absorbed, metabolized, distributed to tissues, and

the mechanism by which it acts. This makes the study of nutraceutical intervention a complex

endeavour, as responses to the nutrient of interest are often optimal in some study participants,

null in others, and can even demonstrate the opposite of the intended effect. This thesis is the

first to provide clinical evidence of an association between lycopene intervention and decreased

bone turnover markers. However, it has also shown that lycopene interacted with genetic factors,

specifically the polymorphism on the PON1 antioxidant enzyme, to affect the response in bone.

The results presented in this thesis provide mechanistic evidence by which lycopene acts

to decrease oxidative stress, interacting with the PON1 polymorphism to decrease the bone

turnover markers BAP and NTx. This decrease in bone turnover markers is beneficial as it has
21-23
been associated with an increased BMD and decreased fracture risk . In women who
211
consumed their usual daily diets (6.03  0.53 mg of lycopene per day) a significant interaction

between serum lycopene and the PON1 polymorphisms was observed on BAP and NTx. This

was evidenced by an association between high serum lycopene and low TBARS, BAP and NTx,

in participants with the combined 172TT and 584G genotype only. The opposite effect on BAP

and NTx was seen in participants with the combined 172TT and 584AA genotype, suggesting

that the PON1 genotype is an important predictor of the response to lycopene in bone.

The effect of serum lycopene on NTx is most likely attributed to the 172A genotype; a

significant interaction was shown between serum lycopene and PON1 172A genotype on NTx.

In participants with the 172TT genotype a high serum lycopene was associated with a lower

NTx. Analysis of results obtained from the clinical intervention study strengthened these

findings. After intervention with lycopene, postmenopausal participants with the 172TT

genotype had significantly lower concentrations of NTx than participants carrying the 172A

allele. Oxidative stress parameters decreased after intervention in all genotypes; however, only in

the 172TT genotype was there a significant decrease in NTx. Research shows that the 172TT
166
genotype is associated with a lower BMD and lower antioxidative capacity of PON1 to
148, 242
prevent LDL oxidation . That this genotype was shown to respond more favourably to

lycopene intervention provides evidence that lycopene is particularly effective when internal

defenses against oxidative stress are not as high. A significant interaction between the change in

TBARS and the PON1 genotype provides evidence that the mechanism by which lycopene acts

on bone is to decrease lipid peroxidation, interacting with the PON1 genotype to decrease bone

resorption markers. These are the first findings to demonstrate a mechanism by which lycopene

may decrease the risk of osteoporosis. They also reveal a genotype which may respond better to

lycopene intervention as a method to decrease osteoporosis risk, as this genotype possesses a

greater need for dietary antioxidants to compensate for insufficient endogenous antioxidant

212
defenses. This genotype contains the major allele of the Caucasian, African-American and

Chinese populations (major allele frequencies: 0.63, 0.80 and 0.98, respectively).

IV. Findings in this thesis provide a new, updated method for use of

lycopene in cell culture techniques.

Typically in vitro studies use the all-trans form of lycopene for cell culture techniques. The all-

trans form of lycopene is the isomer which is present in the highest concentration in foods, at
86
approximately 79-95% of total lycopene . However, the process of digestion and absorption

results in the physiological isomerization of all-trans to cis lycopene. Therefore, in true

biological situations, the cis isomer of lycopene is present in serum and delivered to cells and

tissues in a ratio of approximately 50:50 cis:trans. Due to preferential uptake, the ratio in tissues

is as high as 75-90% in favour of cis 84-86, 92. This mechanism was verified in this thesis (Chapter

9), which demonstrated that the use of lycopene isomers containing higher concentrations of cis

(45:55 and 28:72 cis:trans) resulted in a more efficient delivery and absorption of total lycopene

to the cell. Furthermore, the use of this higher cis lycopene demonstrated significant effects on

mineralization that were not significant in cells treated with a higher concentration of all-trans

(5:95 cis:trans). Therefore, this thesis provides new methodology which employs cis lycopene,

the isomeric form most similar to that occurring in true biological situations. The use of this

isomer will most likely result in a more efficient delivery of lycopene and may provide more

accurate information of the effects of lycopene at the cellular level.

213
 Implications of Experimental Findings

There are several implications resulting from the experimental findings presented in this thesis.

The greatest impact of this thesis is that it provides a good rationale that supplementation with

lycopene may decrease the risk of osteoporosis by decreasing oxidative stress parameters and

bone turnover markers; contributors to osteoporosis risk. The decrease in the bone resorption

marker NTx associated with lycopene supplementation was similar to that previously shown in
233, 234
postmenopausal women supplemented with calcium and vitamin D . Based on these

findings, the consumption of lycopene by women to improve overall bone health should be

considered. These findings justify further investigation with randomized controlled trials to

determine whether an official recommendation for lycopene supplementation in women is

warranted, with particular attention given to the PON1 genotype. Furthermore, the evidence

presented here emphasizes the importance of antioxidants in general for the prevention of

osteoporosis and should encourage a greater number of randomized, controlled trials

investigating these effects and their use as preventative measures. This experimental work

delineated mechanisms by which oxidative stress can affect bone resorption and formation

markers to increase osteoporosis risk. These findings should prompt further investigation into

this relationship, by specifically targeting the decrease of the contributing risk factor oxidative

stress to prevent and treat osteoporosis. The interaction between PON1 polymorphisms and

serum lycopene shown in this thesis provides further evidence that targeting individual genetic

make-up to determine which nutrients would improve bone health is an area requiring further

attention. Future study examining the interactions between polymorphisms and nutraceutical

interventions and their relation to oxidative stress should be investigated to advance the

understanding of osteoporosis risk, prevention and treatment. The incorporation of genotyping

into the science of nutritional studies on osteoporosis may lead to the development of dietary

214
recommendations which are specific to individual needs and may provide greater protection

against the risk of developing this disease. These polymorphism studies will be important in

delineating a comprehensive approach to prevention and treatment of osteoporosis which will be

optimally beneficial for individuals.

Limitations and difficulties encountered

The most challenging problem encountered during the experimental work described in this thesis

was participant recruitment, particularly for the randomized clinical intervention study. The

sample of participants required was very specific, and this limited participant eligibility. For

example it proved difficult to recruit women between 50-60 years of age who were not on any

medications. However, what proved to be a more difficult challenge was to find volunteers who

were willing to change their diets for a period of 5 months in order to participate in the study;

specifically to refrain from taking vitamins and supplements, and to restrict lycopene-containing

foods from their diet. Many eligible potential participants did not enrol for this reason, which

limited the sample size to 60. A larger sample size may have contributed more information to the

polymorphism analysis on how lycopene and PON1 interact to affect bone. Although many

clinical intervention studies use a lycopene-washout period, the data presented in this thesis

suggest that limitation of lycopene-containing foods during the supplemental study period may

not have been necessary. Firstly, lycopene intake data, as shown in Chapter 3, demonstrated that

the usual daily intake of 6.23  0.64 mg per day (range 0.0 – 21.43 mg/day) for female

participants aged 50 and over was 85.4% lower than the average of 43.33  2.84 mg per day

(range 30.0 – 70.0 mg/day) provided by the supplements for the intervention study (Mann-

Whitney test, p<0.0001). This usual daily intake corresponded to a total serum lycopene

215
concentration of 1377  75.92 nM that was 31.6% lower than that of 2012  88.56 nM obtained

by lycopene supplementation with tomato juice or capsules (unpaired t-test, p<0.0001). More

importantly, this corresponded to a 5-cis serum concentration of 436.4  25.37 nM that was

36.3% lower than that of 685.0  30.22 nM obtained by lycopene supplementation (unpaired t-

test, p<0.0001). Secondly, the absorption saturation demonstrated in Chapter 5, whereby

supplementation with 70 mg of lycopene per day resulted in similar serum lycopene

concentrations as those resulting from 30 mg per day (Table 5.3), suggests that there is a

maximal absorption of lycopene occurring at such higher doses. Therefore, a clinical study in

which participants supplement their usual daily diet with tomato juice or capsules to achieve a

maximal serum lycopene concentration similar to that obtained in the clinical intervention study

(approximately 2012  88.56 nM total serum lycopene) could be easily achieved. This would

eliminate the need of participants to restrict their diets and may yield a higher recruitment for the

participant population.

Future directions

A longitudinal prospective study in which postmenopausal women are supplemented with

lycopene and changes in BMD and fracture risk are monitored would further support the present

findings. Although there is a strong correlation between BTMs, BMD and fracture risk, this

would solidify the findings shown in this thesis. Sahni et al, in a longitudinal 17-year study,

reported that lycopene intake was correlated with increased BMD and decreased risk of fracture
73, 78
at the lumbar spine and hip . However, they did not report serum lycopene concentrations.

As shown in Chapter 6, the serum lycopene concentrations resulting from intervention with

lycopene provide more benefit than those typically obtained through the diet alone, particularly

216
with respect to NTx. Sahni et al reported that 4.4 servings was associated with decreased risk of

hip fracture, but they did not clarify between raw tomatoes and cooked/processed tomato
73
products . As discussed previously, processed tomato products provide a higher quantity of

lycopene and result in a better absorption of lycopene. Thus despite the correlation shown in this

thesis between intake and serum concentrations (Chapter 3), and for the aforementioned reasons,

the studies by Sahni et al did not provide enough information about what mg quantities of

lycopene and types of lycopene-containing foods would result in decreased BMD and risk of

fracture. A randomized, clinical intervention trial supplying 30 mg of lycopene per day, in which

increased serum lycopene is directly associated with the change in BMD after 1 to 3 years would

confirm the findings reported in this thesis.

Now that the use of lycopene for the prevention of osteoporosis has been demonstrated in

a randomized, clinical intervention trial, further trials examining the effects of lycopene (1) in

conjunction with FDA-approved medication to treat osteoporosis and (2) in conjunction with the

intake of recommended quantities of calcium and vitamin D to prevent osteoporosis, are

warranted. The determination of whether lycopene can provide added benefit to currently

approved methods for prevention and treatment would result in a more comprehensive picture of

the role of lycopene in bone health.

The in vitro work reported in Chapter 9 demonstrated that cis lycopene was capable of

preventing and repairing the damaging effects of ROS in human osteoblast cells. Clinical

evidence showed that lycopene can decrease bone resorption markers, however, the effect of cis

lycopene on osteoclasts in vitro was not investigated at this time. ROS can stimulate

differentiation of, and resorption by osteoclasts, therefore further in vitro studies delineating

whether lycopene can inhibit or repair these effects of ROS would provide mechanistic evidence

217
of lycopene action on osteoclasts, and verify the present clinical findings showing that lycopene

decreases the bone resorption marker NTx.

Due to smaller sample sizes, the genotype frequencies of the minor alleles for both PON1

172TA and 584AG were low and hence were grouped with the heterozygous participants

for statistical analyses. However, the effects shown in this thesis may have been more clearly

delineated in a larger sample size which did not require combining those homozygous for minor

alleles with heterozygotes. Indeed, the PON1 polymorphisms are associated stepwise increases

in PON1 activity, i.e. 172TT>172TA>172AA 144, 147, suggesting that there could be intermediate

effects which may have dulled the significance of particular results seen here. For example, this

may account for the non-significant association observed between serum lycopene on lipid

peroxidation parameters (Table 7.5, Chapter 7). Therefore, a larger study, in which participants

are pre-selected based on their genotypes to ensure both equal numbers and sturdy representation

of genotypes, may strengthen the present findings and further elucidate the mechanisms by

which PON1 modulates the effect of lycopene on bone. A potential alternative to further explore

the interaction between the PON1 SNPs and lycopene on bone would be the study of PON1

knockout (PON -/-) and transgenic (PON1 Tg) mice 259-263. PON -/- mice have been shown to have

increased levels of oxidative stress and decreased internal antioxidant defenses as well as an
261, 262
increased susceptibility to both HDL and LDL oxidation compared to wildtypes . Several

PON1 Tg mouse models have been developed, which have shown an increased ability to protect
263
LDL from oxidation , inhibit lipid hydroperoxide formation on HDL and protect HDL

integrity and function 260, 261. PON -/- and PON Tg mice, together with wildtype controls, fed with

either lycopene or control diets to determine the effects on BMD and/or histomorphometric

indices, such as trabecular number, may provide further proof that lycopene acts to improve bone

health in situations where individuals are genetically predisposed to low internal antioxidant

218
defenses. Further, tissue culture work targeted at delineating the mechanisms by which PON1

interacts with lycopene to promote bone health would strengthen the clinical findings reported in

this thesis. Use of osteoblasts or osteoclasts representing each of the genotypes for 172TA and

584AG would provide verification of the interactions and mechanisms reported in Chapters 7

and 8.

219
 Conclusions

Osteoporosis is a serious health concern affecting approximately 1 in 4 Canadian women 2. As a

high proportion of the population are over the age of 65, the incidence of this disease is

increasing and optimization of prevention and treatment methods are necessary to counteract this

debilitating disease. Dietary components are important contributors to prevention and their

administration into the usual lifestyle is relatively simple and inexpensive. Understanding the

mechanisms by which this disease manifests itself is also an important proponent towards

prevention and treatment. This thesis provides mechanistic evidence demonstrating that lycopene

acts in its potent antioxidant capacity to decrease oxidative stress parameters and bone turnover

markers and may, therefore, reduce the risk of developing osteoporosis, particularly in cases

where there is a genetic predisposition to low internal antioxidant defenses. Preliminary evidence

suggests that it may also be beneficial in conjunction with currently approved medications to

treat the disease by reinstating the balance between bone formation and resorption. The novel

results presented in this thesis provide incentive for women to supplement their diets with

lycopene to decrease oxidative stress and improve overall bone health.

220
 List of References

1. NIH Consensus Development Panel on Osteoporosis Prevention, D., and Therapy.

Osteoporosis prevention, diagnosis, and therapy. JAMA 285(6): 785-95 (2001).
2. Brown, J. P. & Josse, R. G. 2002 clinical practice guidelines for the diagnosis and
management of osteoporosis in Canada. CMAJ 167S1-S34 (2002).
3. Hegge, K. A., Fornoff, A.S., Gutierres, S.L., Haack, S.L. New therapies for osteoporosis.
Journal of Pharmacy Practice 22(1): 53-64 (2009).
4. WHO. ( Geneva, 1994).
5. Byrjalsen, I., Leeming, D. J., Qvist, P., Christiansen, C. & Karsdal, M. A. Bone turnover
and bone collagen maturation in osteoporosis: effects of antiresorptive therapies.
Osteoporos Int 19(3): 339-48 (2008).
6. de Villiers, T. J. Bone health and osteoporosis in postmenopausal women. Best Pract Res
Clin Obstet Gynaecol 23(1): 73-85 (2009).
7. Brown, J. P. et al. Bone turnover markers in the management of postmenopausal
osteoporosis. Clin Biochem 42(10-11): 929-42 (2009).
8. Bergmann, P. et al. Evidence-based guidelines for the use of biochemical markers of
bone turnover in the selection and monitoring of bisphosphonate treatment in
osteoporosis: a consensus document of the Belgian Bone Club. Int J Clin Pract 63(1): 19-
26 (2009).
9. Eastell, R., Baumann, M., Hoyle, N. R. & Wieczorek, L. Bone Markers: Biochemical and
Clinical Perspectives (Martin Dunitz Ltd., London, 2001).
10. Garnero, P. Biomarkers for osteoporosis management: utility in diagnosis, fracture risk
prediction and therapy monitoring. Mol Diagn Ther 12(3): 157-70 (2008).
11. Camacho, P. M. & Lopez, N. A. Use of biochemical markers of bone turnover in the
management of postmenopausal osteoporosis. Clin Chem Lab Med 46(10): 1345-57
(2008).
12. Calvo, M. S., Eyre, D. R. & Gundberg, C. M. Molecular basis and clinical application of
biological markers of bone turnover. Endocr Rev 17(4): 333-68 (1996).
13. Cashman, K. D. Diet, Nutrition, and Bone Health. J. Nutr. 137(11): 2507S-2512 (2007).
14. Fischer, M. & Raue, F. Measurements of bone mineral density. Mineral density in
metabolic bone disease. Q J Nucl Med 43(3): 233-40 (1999).
15. Keen, R. W. et al. Can biochemical markers predict bone loss at the hip and spine?: a 4-
year prospective study of 141 early postmenopausal women. Osteoporos Int 5(6): 399-
406 (1996).
16. Garnero, P., Sornay-Rendu, E., Chapuy, M. C. & Delmas, P. D. Increased bone turnover
in late postmenopausal women is a major determinant of osteoporosis. J Bone Min Res
11(3): 337-49 (1996).
17. Kushida, K., Takahashi, M., Kawana, K. & Inoue, T. Comparison of markers for bone
formation and resorption in premenopausal and postmenopausal subjects, and
osteoporosis patients. J Clin Endocrinol Metab 80(8): 2447-50 (1995).
18. Reginster, J. Y. et al. Bone resorption in post-menopausal women with normal and low
BMD assessed with biochemical markers specific for telopeptide derived degradation
products of collagen type I. Calcif Tissue Int 69(3): 130-137 (2001).
19. Garnero, P. & Delmas, P. D. Contribution of bone mineral density and bone turnover
markers to the estimation of risk of osteoporotic fracture in postmenopausal women. J
Musculoskelet Neuronal Interact 4(1): 50-63 (2004).

221
20. Glover, S. J., Garnero, P., Naylor, K., Rogers, A. & Eastell, R. Establishing a reference
range for bone turnover markers in young, healthy women. Bone 42(4): 623-30 (2008).
21. Delmas, P. D. et al. Daily and intermittent oral ibandronate normalize bone turnover and
provide significant reduction in vertebral fracture risk: results from the BONE study.
Osteoporos Int 15(10): 792-8 (2004).
22. Riis, B. J., Hansen, M. A., Jensen, A. M., Overgaard, K. & Christiansen, C. Low bone
mass and fast rate of bone loss at menopause: equal risk factors for future fracture: a 15-
year follow-up study. Bone 19(1): 9-12 (1996).
23. Garnero, P., Sornay-Rendu, E., Claustrat, B. & Delmas, P. D. Biochemical Markers of
Bone Turnover, Endogenous Hormones and the Risk of Fractures in Postmenopausal
Women: The OFELY Study. J Bone Miner Res 15(8): 1526-1536 (2000).
24. Garnero, P., Sornay-Rendu, E., Duboeuf, F. & Delmas, P. D. Markers of bone turnover
predict postmenopausal forearm bone loss over 4 years: the OFELY study. J Bone Miner
Res 14(9): 1614-21 (1999).
25. Recker, R., Lappe, J., Davies, K. M. & Heaney, R. Bone remodeling increases
substantially in the years after menopause and remains increased in older osteoporosis
patients. J Bone Miner Res 19(10): 1628-33 (2004).
26. Frei, B. Efficacy of Dietary Antioxidants to Prevent Oxidative Damage and Inhibit
Chronic Disease. J Nutr 1343196S-3198S (2004).
27. Singh, V. N. A current perspective on nutrition and exercise. J Nutr 122(Suppl 3): 760-5
(1992).
28. WHO. ( Geneva, 1998).
29. Genuis, S. J. & Schwalfenberg, G. K. Picking a bone with contemporary osteoporosis
management: nutrient strategies to enhance skeletal integrity. Clin Nutr 26(2): 193-207
(2007).
30. New, S. A. et al. Dietary influences on bone mass and bone metabolism, further evidence
of a positive link between fruit and vegetable consumption and bone health? Am J Clin
Nutr 71(1): 142-151 (2000).
31. Wolf, R. L. et al. Lack of relation between vitamin and mineral antioxidants and bone
mineral density: results from the Women's Health Initiative. Am J Clin Nutr 82(3): 581-
588 (2005).
32. Lin, P. H. et al. The DASH diet and sodium reduction improve markers of bone turnover
and calcium metabolism in adults. J Nutr 133(10): 3130-6 (2003).
33. Lanham-New, S. A. The balance of bone health: tipping the scales in favor of potassium-
rich, bicarbonate-rich foods. J Nutr 138(1): 172S-177S (2008).
34. Puel, C., Coxam, V. & Davicco, M. J. [Mediterranean diet and osteoporosis prevention].
Med Sci (Paris) 23(8-9): 756-60 (2007).
35. Willcox, J. K., Ash, S. L. & Catignani, G. L. Antioxidants and prevention of chronic
disease. Crit Rev Food Sci Nutr 44(4): 275-95 (2004).
36. Shi, J., Qu, Q., Kakuda, Y., Yeung, D. & Jiang, Y. Stability and synergistic effect of
antioxidative properties of lycopene and other active components. Crit Rev Food Sci Nutr
44(7-8): 559-73 (2004).
37. Harman, D. Aging: a theory based on free radical and radiation chemistry. J Gerontol
11(3): 298-300 (1956).
38. Griffiths, H. R. et al. Biomarkers. Mol Aspects Med 23(1-3): 101-208 (2002).
39. Trevisan, M. et al. Correlates of markers of oxidative status in the general population. Am
J Epidemiol 154(4): 348-56 (2001).

222
40. Oh, B. et al. Associations of catalase gene polymorphisms with bone mineral density and
bone turnover markers in postmenopausal women. J Med Genet 44(1): e62 (2007).
41. Sendur, O. F., Turan, Y., Tastaban, E. & Serter, M. Antioxidant status in patients with
osteoporosis: A controlled study. Joint Bone Spine 76(5): 514-8 (2009).
42. Schwedhelm, E., Maas, R., Troost, R. & Boger, R. H. Clinical pharmacokinetics of
antioxidants and their impact on systemic oxidative stress. Clin Pharmacokinet 42(5):
437-59 (2003).
43. Lee, D. H., Lim, B. S., Lee, Y. K. & Yang, H. C. Effects of hydrogen peroxide (H2O2)
on alkaline phosphatase activity and matrix mineralization of odontoblast and osteoblast
cell lines. Cell Biol Toxicol 22(1): 39-46 (2006).
44. Rao, A. V. & Rao, L. G. Carotenoids and human health. Pharmacol Res 55(3): 207-16
(2007).
45. Telci, A. et al. Postmenopausal hormone replacement therapy use decreases oxidative
protein damage. Gynecol Obstet Invest 54(2): 88-93 (2002).
46. Mackinnon, E. S., Rao, A.V., Josse, R.G., Rao, L.G. Menopausal and postmenopausal
women have significantly higher bone resorption and oxidative stress parameters than
premenopausal women. Journal of Bone and Mineral Research 23(Suppl. 1): S326
(2008).
47. Basu, S., Michaelsson, K., Olofsson, H., Johansson, S. & Melhus, H. Association
between oxidative stress and bone mineral density. Biochem Biophys Res Commun
288275-9 (2001).
48. Prasad, G., Dhillon, M. S., Khullar, M. & Nagi, O. N. Evaluation of oxidative stress after
fractures: a preliminary study. Acta Orthop Belg 69(6): 546-551 (2003).
49. Sanchez-Rodriguez, M. A., Ruiz-Ramos, M., Correa-Munoz, E. & Mendoza-Nunez, V.
M. Oxidative stress as a risk factor for osteoporosis in elderly Mexicans as characterized
by antioxidant enzymes. BMC Musculoskelet Disord 8124 (2007).
50. Sontakke, A. N. & Tare, R. S. A duality in the roles of reactive oxygen species with
respect to bone metabolism. Clin Chim Acta 318(1-2): 145-8 (2002).
51. Lean, J. M., Jagger, C. J., Kirstein, B., Fuller, K. & Chambers, T. J. Hydrogen peroxide is
essential for estrogen-deficiency bone loss and osteoclast formation. Endocrinology
146(2): 728-35 (2005).
52. Park, B. G., Yoo, C. I., Kim, H. T., Kwon, C. H. & Kim, Y. K. Role of mitogen-activated
protein kinases in hydrogen peroxide-induced cell death in osteoblastic cells. Toxicology
215(1-2): 115-125 (2005).
53. Garrett, I. R. et al. Oxygen-derived free radicals stimulate osteoclastic bone resorption in
rodent bone in vitro and in vivo. J Clin Invest 85(3): 632-9 (1990).
54. Banfi, G., Iorio, E. L. & Corsi, M. M. Oxidative stress, free radicals and bone
remodeling. Clin Chem Lab Med 46(11): 1550-5 (2008).
55. Sheweita, S. A. & Khoshhal, K. I. Calcium metabolism and oxidative stress in bone
fractures: role of antioxidants. Curr Drug Metab 8(5): 519-25 (2007).
56. Mody, N., Parhami, F., Sarafian, T. A. & Demer, L. L. Oxidative stress modulates
osteoblastic differentiation of vascular and bone cells. Free Radic Biol Med 31509-519
(2001).
57. Arai, M., Shibata, Y., Pugdee, K., Abiko, Y. & Ogata, Y. Effects of reactive oxygen
species (ROS) on antioxidant system and osteoblastic differentiation in MC3T3-E1 cells.
IUBMB Life 59(1): 27-33 (2007).

223
58. Liu, A. L., Zhang, Z. M., Zhu, B. F., Liao, Z. H. & Liu, Z. Metallothionein protects bone
marrow stromal cells against hydrogen peroxide-induced inhibition of osteoblastic
differentiation. Cell Biol Int 28(12): 905-911 (2004).
59. Fatokun, A. A., Stone, T. W. & Smith, R. A. Responses of differentiated MC3T3-E1
osteoblast-like cells to reactive oxygen species. Eur J Pharmacol 587(1-3): 35-41 (2008).
60. Hahn, M. et al. Effects of bone disease and calcium supplementation on antioxidant
enzymes in postmenopausal women. Clin Biochem 41(1-2): 69-74 (2008).
61. Ozgocmen, S., Kaya, H., Fadillioglu, E., Aydogan, R. & Yilmaz, Z. Role of antioxidant
systems, lipid peroxidation, and nitric oxide in postmenopausal osteoporosis. Mol Cell
Biochem 295(1-2): 45-52 (2007).
62. Korucuoglu, U. et al. Assessment of protein oxidation in women using raloxifene. Mol
Cell Biochem 290(1-2): 97-101 (2006).
63. Ozgocmen, S., Kaya, H., Fadillioglu, E. & Yilmaz, Z. Effects of calcitonin, risedronate,
and raloxifene on erythrocyte antioxidant enzyme activity, lipid peroxidation, and nitric
oxide in postmenopausal osteoporosis. Arch Med Res 38(2): 196-205 (2007).
64. Misra, R. et al. LycoRed as an alternative to hormone replacement therapy in lowering
serum lipids and oxidative stress markers: a randomized controlled clinical trial. J Obstet
Gynaecol Res 32(3): 299-304 (2006).
65. Maggio, D. et al. Marked decrease in plasma antioxidants in aged osteoporotic women:
results of a cross-sectional study. J Clin Endocrinol Metab 881523-1527 (2003).
66. Maggio, D. et al. Low levels of carotenoids and retinol in involutional osteoporosis. Bone
38(2): 244-8 (2006).
67. Melhus, H., Michaelsson, K., Holmberg, L., Wolk, A. & Jungall, L. Smoking, antioxidant
vitamins, and the risk of hip fracture. J Bone Min Res 14129-35 (1999).
68. Hall, S. L. & Greendale, G. A. The relation of dietary vitamin C intake to bone mineral
density: results from the PEPI study. Calcif Tissue Int 63(3): 183-189 (1998).
69. Kaptoge, S. et al. Effects of dietary nutrients and food groups on bone loss from the
proximal femur in men and women in the 7th and 8th decades of age. Osteoporos Int
14(5): 418-428 (2003).
70. Macdonald, H. M., New, S. A., Golden, M. H., Campbell, M. K. & Reid, D. M.
Nutritional associations with bone loss during the menopausal transition: evidence of a
beneficial effect of calcium, alcohol, and fruit and vegetable nutrients and of a
detrimental effect of fatty acids. Am J Clin Nutr 79(1): 155-165 (2004).
71. Sahni, S. et al. Protective effect of total and supplemental vitamin C intake on the risk of
hip fracture-a 17-year follow-up from the Framingham Osteoporosis Study. Osteoporos
Int 20(11): 1853-61 (2009).
72. Pasco, J. A. et al. Antioxidant vitamin supplements and markers of bone turnover in a
community sample of nonsmoking women. J Womens Health (Larchmt) 15(3): 295-300
(2006).
73. Sahni, S. et al. Protective Effect of Total Carotenoid and Lycopene Intake on the Risk of
Hip Fracture: A 17-Year Follow-Up From the Framingham Osteoporosis Study. J Bone
Miner Res (2009).
74. Uchiyama, S. & Yamaguchi, M. beta-cryptoxanthin stimulates cell differentiation and
mineralization in osteoblastic MC3T3-E1 cells. J Cell Biochem 95(6): 1224-34 (2005).
75. Yamaguchi, M. & Uchiyama, S. Effect of carotenoid on calcium content and alkaline
phosphatase activity in rat femoral tissues in vitro: the unique anabolic effect of beta-
cryptoxanthin. Biol Pharm Bull 26(8): 1188-1191 (2003).

224
76. Zhang, J. et al. Antioxidant Intake and Risk of Osteoporotic Hip Fracture in Utah: An
Effect Modified by Smoking Status. Am J Epidemiol 163(1): 9-17 (2006).
77. Park, C. K., Ishimi, Y., Ohmura, M., Yamaguchi, M. & Ikegami, S. Vitamin A and
carotenoids stimulate differentiation of mouse osteoblastic cells. J Nutr Sci and Vitaminol
43281-296 (1997).
78. Sahni, S. et al. Inverse association of carotenoid intakes with 4-y change in bone mineral
density in elderly men and women: the Framingham Osteoporosis Study. Am J Clin Nutr
89(1): 416-24 (2009).
79. Sugiura, M. et al. Bone mineral density in post-menopausal female subjects is associated
with serum antioxidant carotenoids. Osteoporos Int 19(2): 211-9 (2008).
80. Wattanapenpaiboon, N., Lukito, W., Wahlqvist, M. L. & Strauss, B. J. Dietary carotenoid
intake as a predictor of bone mineral density. Asia Pac J Clin Nutr 12(4): 467-73 (2003).
81. Feskanich, D., Singh, V., Willett, W. C. & Colditz, G. A. Vitamin A intake and hip
fractures among postmenopausal women. Jama 287(1): 47-54 (2002).
82. Barker, M. E. et al. Serum Retinoids and -Carotene as Predictors of Hip and Other
Fractures in Elderly Women. J Bone Min Res 20(6): 913-920 (2005).
83. Melhus, H. et al. Excessive dietary intake of vitamin A is associated with reduced bone
mineral density and increased risk for hip fracture. Ann Intern Med 129(10): 770-8
(1998).
84. Agarwal, S. & Rao, A. V. Tomato lycopene and its role in human health and chronic
disease. CMAJ 163(6): 739-44 (2000).
85. Boileau, T. W., Boileau, A. C. & Erdman, J. W., Jr. Bioavailability of all-trans and cis-
isomers of lycopene. Exp Biol Med (Maywood) 227(10): 914-9 (2002).
86. Preedy, V. R., Watson, R.R. (ed.) Tomatoes and Tomato Products: Nutritional,
Medicinal, and Therapeutic Properties (Science Publishers, Enfield, 2008).
87. Bohm, V., Frohlich, K. & Bitsch, R. Rosehip -- a "new" source of lycopene? Mol Aspects
Med 24(6): 385-9 (2003).
88. Tonucci, L. H. et al. Carotenoid Content of Thermally Processed Tomato-Based Food
Products. J Agric Food Chem 43579-586 (1995).
89. U.S. Department of Agriculture, A. R. S. (United States Department of Agriculture,
2009).
90. Johnson-Down, L., Saudny-Unterberger, H. & Gray-Donald, K. Food habits of
Canadians: lutein and lycopene intake in the Canadian population. J Am Diet Assoc
102(7): 988-91 (2002).
91. Mackinnon, E. S., Rao, A. V. & Rao, L. G. Lycopene intake by Canadian women is
variable, similar among different ages, but greater than that reported for women in other
countries. J Med Food 12(4): 829-35 (2009).
92. Shi, J. & Le Maguer, M. Lycopene in tomatoes: chemical and physical properties
affected by food processing. Crit Rev Biotechnol 20(4): 293-334 (2000).
93. Khachik, F. et al. Chemistry, distribution, and metabolism of tomato carotenoids and
their impact on human health. Exp Biol Med (Maywood) 227(10): 845-51 (2002).
94. Kavanaugh, C. J., Trumbo, P. R. & Ellwood, K. C. The U.S. Food and Drug
Administration's evidence-based review for qualified health claims: tomatoes, lycopene,
and cancer. J Natl Cancer Inst 99(14): 1074-85 (2007).
95. Preedy, V. R., Watson, R.R. (ed.) Lycopene: Nutritional, Medicinal, and Therapeutic
Properties (Science Publishers, Enfield, 2008).

225
96. Olmedilla, B. et al. A European multicentre, placebo-controlled supplementation study
with alpha-tocopherol, carotene-rich palm oil, lutein or lycopene: analysis of serum
responses. Clin Sci (Lond) 102(4): 447-56 (2002).
97. Rao, A. V. & Agarwal, S. Bioavailability and in vivo antioxidant properties of lycopene
from tomato products and their possible role in the prevention of cancer. Nutr Cancer
31199-203 (1998).
98. Frohlich, K., Kaufmann, K., Bitsch, R. & Bohm, V. Effects of ingestion of tomatoes,
tomato juice and tomato puree on contents of lycopene isomers, tocopherols and ascorbic
acid in human plasma as well as on lycopene isomer pattern. Br J Nutr 95(4): 734-41
(2006).
99. Agarwal, S. & Rao, A. V. Tomato lycopene and low density lipoprotein oxidation: a
human dietary intervention study Lipids 33(10): 981-984 (1998).
100. Matulka, R. A., Hood, A. M. & Griffiths, J. C. Safety evaluation of a natural tomato
oleoresin extract derived from food-processing tomatoes. Regul Toxicol Pharmacol
39(3): 390-402 (2004).
101. Michael McClain, R. & Bausch, J. Summary of safety studies conducted with synthetic
lycopene. Regul Toxicol Pharmacol 37(2): 274-85 (2003).
102. (ed. Institute of Medicine, F. a. N. B.) 325-382 and 448-457 (National Academy Press,
Washington, D.C, 2000).
103. Yang, Z., Zhang, Z., Penniston, K. L., Binkley, N. & Tanumihardjo, S. A. Serum
carotenoid concentrations in postmenopausal women from the United States with and
without osteoporosis. Int J Vitam Nutr Res 78(3): 105-11 (2008).
104. Stahl, W. & Sies, H. Lycopene: a biologically important carotenoid for humans? Arch
Biochem Biophys 336(1): 1-9 (1996).
105. Burri, B. J., Chapman, M. H., Neidlinger, T. R., Seo, J. S. & Ishida, B. K. Tangerine
tomatoes increase total and tetra-cis-lycopene isomer concentrations more than red
tomatoes in healthy adult humans. Int J Food Sci Nutr1-16 (2008).
106. Rock, C. L., Swendseid, M. E., Jacob, R. A. & McKee, R. W. Plasma Carotenoid Levels
in Human Subjects Fed a Low Carotenoid Diet. J. Nutr. 122(1): 96-100 (1992).
107. Khachik, F., Beecher, G. R. & Smith, J. C., Jr. Lutein, lycopene, and their oxidative
metabolites in chemoprevention of cancer. J Cell Biochem Suppl 22236-46 (1995).
108. Khachik, F. et al. Identification, quantification, and relative concentrations of carotenoids
and their metabolites in human milk and serum. Anal Chem 69(10): 1873-81 (1997).
109. Porrini, M., Riso, P. & Testolin, G. Absorption of lycopene from single or daily portions
of raw and processed tomato. Br J Nutr 80(4): 353-61 (1998).
110. Holloway, D. E., Yang, M., Paganga, G., Rice-Evans, C. A. & Bramley, P. M.
Isomerization of dietary lycopene during assimilation and transport in plasma. Free Radic
Res 32(1): 93-102 (2000).
111. Bohm, V. & Bitsch, R. Intestinal absorption of lycopene from different matrices and
interactions to other carotenoids, the lipid status, and the antioxidant capacity of human
plasma. Eur J Nutr 38(3): 118-25 (1999).
112. Failla, M. L., Chitchumroonchokchai, C. & Ishida, B. K. In Vitro Micellarization and
Intestinal Cell Uptake of cis Isomers of Lycopene Exceed Those of All-trans Lycopene.
J. Nutr. 138(3): 482-486 (2008).
113. Tyssandier, V. et al. Vegetable-borne lutein, lycopene, and {beta}-carotene compete for
incorporation into chylomicrons, with no adverse effect on the medium-term (3-wk)
plasma status of carotenoids in humans. Am J Clin Nutr 75(3): 526-534 (2002).

226
114. Seren, S. et al. Lycopene in cancer prevention and treatment. Am J Ther 15(1): 66-81
(2008).
115. Devaraj, S. et al. A dose-response study on the effects of purified lycopene
supplementation on biomarkers of oxidative stress. J Am Coll Nutr 27(2): 267-73 (2008).
116. Bohm, V., Puspitasari-Nienaber, N. L., Ferruzzi, M. G. & Schwartz, S. J. Trolox
equivalent antioxidant capacity of different geometrical isomers of alpha-carotene, beta-
carotene, lycopene, and zeaxanthin. J Agric Food Chem 50(1): 221-6 (2002).
117. Di Mascio, P., Kaiser, S. & Sies, H. Lycopene as the most efficient biological carotenoid
singlet oxygen quencher. Arch Biochem Biophys 274(2): 532-8 (1989).
118. Hadley, C. W., Clinton, S. K. & Schwartz, S. J. The consumption of processed tomato
products enhances plasma lycopene concentrations in association with a reduced
lipoprotein sensitivity to oxidative damage. J Nutr 133(3): 727-32 (2003).
119. Silaste, M. L., Alfthan, G., Aro, A., Kesaniemi, Y. A. & Horkko, S. Tomato juice
decreases LDL cholesterol levels and increases LDL resistance to oxidation. Br J Nutr
98(6): 1251-8 (2007).
120. Shen, Y. C., Chen, S. L. & Wang, C. K. Contribution of tomato phenolics to
antioxidation and down-regulation of blood lipids. J Agric Food Chem 55(16): 6475-81
(2007).
121. Fuhrman, B., Elis, A. & Aviram, M. Hypocholesterolemic effect of lycopene and beta-
carotene is related to suppression of cholesterol synthesis and augmentation of LDL
receptor activity in macrophages. Biochem Biophys Res Commun 233(3): 658-62 (1997).
122. Ahuja, K. D., Pittaway, J. K. & Ball, M. J. Effects of olive oil and tomato lycopene
combination on serum lycopene, lipid profile, and lipid oxidation. Nutrition 22(3): 259-
65 (2006).
123. Arab, L. & Steck, S. Lycopene and cardiovascular disease. Am J Clin Nutr 71(6 Suppl):
1691S-5S; discussion 1696S-7S (2000).
124. Rissanen, T., Voutilainen, S., Nyyssonen, K., Salonen, R. & Salonen, J. T. Low plasma
lycopene concentration is associated with increased intima-media thickness of the carotid
artery wall. Arterioscler Thromb Vasc Biol 20(12): 2677-81 (2000).
125. Giovannucci, E. Tomatoes, tomato-based products, lycopene, and cancer: review of the
epidemiologic literature. J Natl Cancer Inst 91(4): 317-31 (1999).
126. Levy, J. et al. Lycopene is a more potent inhibitor of human cancer cell proliferation than
either alpha-carotene or beta-carotene. Nutr Cancer 24(3): 257-66 (1995).
127. Cramer, D. W., Kuper, H., Harlow, B. L. & Titus-Ernstoff, L. Carotenoids, antioxidants
and ovarian cancer risk in pre- and postmenopausal women. Int J Cancer 94(1): 128-134
(2001).
128. Rao, L. G., Krishnadev, N., Banasikowska, K. & Rao, A. V. Lycopene I - Effect of
osteoclasts: Lycopene inhibits basal and parathyroid hormone-stimulated osteoclast
formation and mineral resorption mediated by reactive oxygen species in rat bone
marrow cultures. J Med Food 669-78 (2003).
129. Ishimi, Y., Ohmura, M., Wang, X., Yamaguchi, M. & Ikegami, S. Inhibition by
carotenoids and retinoic acid of osteoclast-like cell formation induced by bone resorbing
agents in vitro. J Clin Biochem Nutr 27113-122 (1999).
130. Kim, L., Rao, A. V. & Rao, L. G. Lycopene II-- Effect on osteoblasts: The carotenoid
lycopene stimulates cell proliferation and alkaline phosphatase activity of SaOS-2 cells. J
Med Food 679-86 (2003).
131. Yamane, M., Nakamura, N., Inakuma, T., Ishimi, Y. in 8th World Congress and 11th
ISHS Symposium (Toronto, ON, 2008).
227
132. Rao, L. G. et al. Lycopene consumption decreases oxidative stress and bone resorption
markers in postmenopausal women. Osteoporos Int 18109-115 (2007).
133. Ferguson, L. R. Nutrigenomics approaches to functional foods. J Am Diet Assoc 109(3):
452-8 (2009).
134. Ferguson, L. R. Nutrigenomics: integrating genomic approaches into nutrition research.
Mol Diagn Ther 10(2): 101-8 (2006).
135. DeBusk, R. M., Fogarty, C. P., Ordovas, J. M. & Kornman, K. S. Nutritional genomics in
practice: where do we begin? J Am Diet Assoc 105(4): 589-98 (2005).
136. Williams, C. M. et al. The challenges for molecular nutrition research 1: linking genotype
to healthy nutrition. Genes Nutr 3(2): 41-9 (2008).
137. Mutch, D. M., Wahli, W. & Williamson, G. Nutrigenomics and nutrigenetics: the
emerging faces of nutrition. Faseb J 19(12): 1602-16 (2005).
138. Nguyen, T. V., Center, J. R. & Eisman, J. A. Pharmacogenetics of osteoporosis and the
prospect of individualized prognosis and individualized therapy. Curr Opin Endocrinol
Diabetes Obes 15(6): 481-8 (2008).
139. Qureshi, A. M., Herd, R. J., Blake, G. M., Fogelman, I. & Ralston, S. H. COLIA1 Sp1
polymorphism predicts response of femoral neck bone density to cyclical etidronate
therapy. Calcif Tissue Int 70(3): 158-63 (2002).
140. Cusack, S. n. & Cashman, K. D. Impact of genetic variation on metabolic response of
bone to diet. Proceedings of the Nutrition Society 62(04): 901-912 (2003).
141. Gong, M. et al. Quercetin up-regulates paraoxonase 1 gene expression with concomitant
protection against LDL oxidation. Biochemical and Biophysical Research
Communications 379(4): 1001-1004 (2009).
142. Mackness, B., Durrington, P. N. & Mackness, M. I. Human serum paraoxonase. Gen
Pharmacol 31(3): 329-36 (1998).
143. Jaouad, L., Milochevitch, C. & Khalil, A. PON1 paraoxonase activity is reduced during
HDL oxidation and is an indicator of HDL antioxidant capacity. Free Radic Res 37(1):
77-83 (2003).
144. Mackness, B., Mackness, M. I., Arrol, S., Turkie, W. & Durrington, P. N. Effect of the
molecular polymorphisms of human paraoxonase (PON1) on the rate of hydrolysis of
paraoxon. Br J Pharmacol 122(2): 265-8 (1997).
145. Rantala, M. et al. Dietary modifications and gene polymorphisms alter serum
paraoxonase activity in healthy women. J Nutr 132(10): 3012-7 (2002).
146. Aviram, M. et al. Paraoxonase inhibits high-density lipoprotein oxidation and preserves
its functions. A possible peroxidative role for paraoxonase. J Clin Invest 101(8): 1581-90
(1998).
147. Mackness, B. et al. Serum paraoxonase (PON1) 55 and 192 polymorphism and
paraoxonase activity and concentration in non-insulin dependent diabetes mellitus.
Atherosclerosis 139(2): 341-9 (1998).
148. Mackness, B., Mackness, M. I., Arrol, S., Turkie, W. & Durrington, P. N. Effect of the
human serum paraoxonase 55 and 192 genetic polymorphisms on the protection by high
density lipoprotein against low density lipoprotein oxidative modification. FEBS Lett
423(1): 57-60 (1998).
149. Mackness, B. et al. Paraoxonase activity in two healthy populations with differing rates
of coronary heart disease. Eur J Clin Invest 30(1): 4-10 (2000).
150. Aviram, M. et al. Human serum paraoxonase (PON 1) is inactivated by oxidized low
density lipoprotein and preserved by antioxidants. Free Radic Biol Med 26(7-8): 892-904
(1999).
228
151. Jarvik, G. P. et al. Vitamin C and E intake is associated with increased paraoxonase
activity. Arterioscler Thromb Vasc Biol 22(8): 1329-33 (2002).
152. Aviram, M. et al. Pomegranate juice consumption reduces oxidative stress, atherogenic
modifications to LDL, and platelet aggregation: studies in humans and in atherosclerotic
apolipoprotein E-deficient mice. Am J Clin Nutr 71(5): 1062-76 (2000).
153. Agachan, B., Yilmaz, H., Karaali, Z. & Isbir, T. Paraoxonase 55 and 192 polymorphism
and its relationship to serum paraoxonase activity and serum lipids in Turkish patients
with non-insulin dependent diabetes mellitus. Cell Biochem Funct 22(3): 163-8 (2004).
154. Kleemola, P. et al. Dietary determinants of serum paraoxonase activity in healthy
humans. Atherosclerosis 160(2): 425-32 (2002).
155. Dalgard, C. et al. No influence of increased intake of orange and blackcurrant juices and
dietary amounts of vitamin E on paraoxonase-1 activity in patients with peripheral
arterial disease. Eur J Nutr 46(6): 354-63 (2007).
156. Blum, S., Aviram, M., Ben-Amotz, A. & Levy, Y. Effect of a Mediterranean meal on
postprandial carotenoids, paraoxonase activity and C-reactive protein levels. Ann Nutr
Metab 50(1): 20-4 (2006).
157. Ferre, N. et al. Regulation of Serum Paraoxonase Activity by Genetic, Nutritional, and
Lifestyle Factors in the General Population. Clin Chem 49(9): 1491-1497 (2003).
158. Tomas, M. et al. Interaction between the Gln-Arg 192 variants of the paraoxonase gene
and oleic acid intake as a determinant of high-density lipoprotein cholesterol and
paraoxonase activity. Eur J Pharmacol 432(2-3): 121-8 (2001).
159. Nus, M. et al. Arylesterase Activity and Antioxidant Status Depend on PON1-Q192R and
PON1-L55M Polymorphisms in Subjects with Increased Risk of Cardiovascular Disease
Consuming Walnut-Enriched Meat. J. Nutr. 137(7): 1783-1788 (2007).
160. Bub, A. et al. Paraoxonase 1 Q192R (PON1-192) polymorphism is associated with
reduced lipid peroxidation in R-allele-carrier but not in QQ homozygous elderly subjects
on a tomato-rich diet. Eur J Nutr 41(6): 237-43 (2002).
161. Bub, A., Barth, S. W., Watzl, B., Briviba, K. & Rechkemmer, G. Paraoxonase 1 Q192R
(PON1-192) polymorphism is associated with reduced lipid peroxidation in healthy
young men on a low-carotenoid diet supplemented with tomato juice. Br J Nutr 93(3):
291-7 (2005).
162. Valko, M. et al. Free radicals and antioxidants in normal physiological functions and
human disease. The International Journal of Biochemistry & Cell Biology 39(1): 44-84
(2007).
163. Parhami, F. et al. Atherogenic Diet and Minimally Oxidized Low Density Lipoprotein
Inhibit Osteogenic and Promote Adipogenic Differentiation of Marrow Stromal Cells.
Journal of Bone and Mineral Research 14(12): 2067-2078 (1999).
164. Parhami, F., Garfinkel, A. & Demer, L. L. Role of Lipids in Osteoporosis. Arterioscler
Thromb Vasc Biol 20(11): 2346-2348 (2000).
165. Mazière, C. et al. Oxidized low density lipoprotein decreases Rankl-induced
differentiation of osteoclasts by inhibition of Rankl signaling. Journal of Cellular
Physiology 221(3): 572-578 (2009).
166. Yamada, Y., Ando, F., Niino, N., Miki, T. & Shimokata, H. Association of
polymorphisms of paraoxonase 1 and 2 genes, alone or in combination, with bone
mineral density in community-dwelling Japanese. J Hum Genet 48(9): 469-75 (2003).
167. Verit, F. F., Celik, H., Yazgan, P., Erel, O. & Geyikli, I. Paraoxonase-1 activity as a
marker of atherosclerosis is not associated with low bone mineral density in healthy
postmenopausal women. Arch Gynecol Obstet 275(5): 353-9 (2007).
229
168. Agarwal, A., Shen, H., Agarwal, S. & Rao, A. V. Lycopene Content of Tomato Products:
Its Stability, Bioavailability and In Vivo Antioxidant Properties. J Med Food 4(1): 9-15
(2001).
169. Weisburger, J. H. Evaluation of the evidence on the role of tomato products in disease
prevention. . Proc Soc Exp Biol Med. 218140-3 (1998).
170. Riso, P., Visioli, F., Erba, D., Testolin, G. & Porrini, M. Lycopene and vitamin C
concentrations increase in plasma and lymphocytes after tomato intake. Effects on
cellular antioxidant protection. Eur J Clin Nutr 58(10): 1350-8 (2004).
171. O'Neill, M. E. et al. A European carotenoid database to assess carotenoid intakes and its
use in a five-country comparative study. Br J Nutr 85(4): 499-507 (2001).
172. Michaud, D. S. et al. Associations of plasma carotenoid concentrations and dietary intake
of specific carotenoids in samples of two prospective cohort studies using a new
carotenoid database. Cancer Epidemiol Biomarkers Prev 7(4): 283-290 (1998).
173. Nebeling, L. C., Forman, M. R., Graubard, B. I. & Snyder, R. A. Changes in carotenoid
intake in the United States: the 1987 and 1992 National Health Interview Surveys. J Am
Diet Assoc 97(9): 991-6 (1997).
174. Slattery, M. L. et al. Carotenoids and colon cancer. Am J Clin Nutr 71(2): 575-82 (2000).
175. Tucker, K. L. et al. Carotenoid intakes, assessed by dietary questionnaire, are associated
with plasma carotenoid concentrations in an elderly population. J Nutr 129(2): 438-45
(1999).
176. Chug-Ahuja, J. K. et al. The development and application of a carotenoid database for
fruits, vegetables, and selected multicomponent foods. J Am Diet Assoc 93(3): 318-23
(1993).
177. U.S. Department of Agriculture, A. R. S. (United States Department of Agriculture,
2005).
178. Stahl, W., Schwarz, W., Sundquist, A. R. & Sies, H. cis-trans isomers of lycopene and
beta-carotene in human serum and tissues. Arch Biochem Biophys 294(1): 173-177
(1992).
179. Michaud, D. S. et al. Associations of plasma carotenoid concentrations and dietary intake
of specific carotenoids in samples of two prospective cohort studies using a new
carotenoid database. Cancer Epidemiol Biomarkers Prev 7(4): 283-90 (1998).
180. Carroll, Y. L., Corridan, B. M. & Morrissey, P. A. Carotenoids in young and elderly
healthy humans: dietary intakes, biochemical status and diet-plasma relationships. Eur J
Clin Nutr 53(8): 644-53 (1999).
181. Diwadkar-Navsariwala, V. et al. A physiological pharmacokinetic model describing the
disposition of lycopene in healthy men. J. Lipid Res. 44(10): 1927-1939 (2003).
182. Ganji, V. & Kafai, M. R. Population determinants of serum lycopene concentrations in
the United States: data from the Third National Health and Nutrition Examination
Survey, 1988-1994. J Nutr 135(3): 567-72 (2005).
183. Casso, D. et al. Correlates of serum lycopene in older women. Nutr Cancer 36163-169
(2000).
184. Mayne, S. T. et al. Plasma lycopene concentrations in humans are determined by
lycopene intake, plasma cholesterol concentration and selected demographic factors. J
Nutr 129(4): 849-854 (1999).
185. Ford, E. S. Variations in serum carotenoid concentrations among United States adults by
ethnicity and sex. Ethn Dis 10(2): 208-17 (2000).
186. Monograph. Lycopene. Altern Med Rev 8(3): 336-342 (2003).

230
187. Lucarini, M., Lanzi, S., D'Evoli, L., Aguzzi, A. & Lombardi-Boccia, G. Intake of vitamin
A and carotenoids from the Italian population--results of an Italian total diet study. Int J
Vitam Nutr Res 76(3): 103-9 (2006).
188. Watanabe, S., Zhuo, X. G. & Kimira, M. Food safety and epidemiology: new database of
functional food factors. Biofactors 22(1-4): 213-9 (2004).
189. Fodor, J. G., Whitmore, B., Leenen, F. & Larochelle, P. Lifestyle modifications to
prevent and control hypertension. 5. Recommendations on dietary salt. Canadian
Hypertension Society, Canadian Coalition for High Blood Pressure Prevention and
Control, Laboratory Centre for Disease Control at Health Canada, Heart and Stroke
Foundation of Canada. Cmaj 160(9 Suppl): S29-34 (1999).
190. Khan, N. A. et al. The 2009 Canadian Hypertension Education Program
recommendations for the management of hypertension: Part 2--therapy. Can J Cardiol
25(5): 287-98 (2009).
191. Canene-Adams, K., Campbell, J. K., Zaripheh, S., Jeffery, E. H. & Erdman, J. W., Jr. The
tomato as a functional food. J Nutr 135(5): 1226-30 (2005).
192. Tyssandier, V. et al. Effect of Tomato Product Consumption on the Plasma Status of
Antioxidant Microconstituents and on the Plasma Total Antioxidant Capacity in Healthy
Subjects. J Am Coll Nutr 23(2): 148-156 (2004).
193. Allen, C. M. et al. Changes in plasma and oral mucosal lycopene isomer concentrations
in healthy adults consuming standard servings of processed tomato products. Nutr Cancer
47(1): 48-56 (2003).
194. Porrini, M. & Riso, P. What Are Typical Lycopene Intakes? J. Nutr. 135(8): 2042S-2045
(2005).
195. Rao, A. V. & Shen, H. Effect of low dose lycopene intake on lycopene bioavailability
and oxidative stress. Nutrition Research 22(10): 1125-1131 (2002).
196. Paran, E. in Tomatoes, Lycopene & Human Health, Preventing Chronic Disease (ed. Rao,
A. V.) 169-182 (Caledonian Science Press, 2006).
197. Rao, A. V. & Agarwal, S. Role of antioxidant lycopene in cancer and heart disease. J Am
Coll Nutr 19(5): 563-9 (2000).
198. Rao, A. V. Processed tomato products as a source of dietary lycopene: bioavailability and
antioxidant properties. Can J Diet Pract Res 65161-165 (2004).
199. Bose, K. S. & Agrawal, B. K. Effect of lycopene from tomatoes (cooked) on plasma
antioxidant enzymes, lipid peroxidation rate and lipid profile in grade-I hypertension.
Ann Nutr Metab 51(5): 477-81 (2007).
200. Bose, K. S. & Agrawal, B. K. Effect of lycopene from cooked tomatoes on serum
antioxidant enzymes, lipid peroxidation rate and lipid profile in coronary heart disease.
Singapore Med J 48(5): 415-20 (2007).
201. Hoppe, P. P., Kramer, K., van den Berg, H., Steenge, G. & van Vliet, T. Synthetic and
tomato-based lycopene have identical bioavailability in humans. Eur J Nutr 42(5): 272-8
(2003).
202. Dixon, Z. R. et al. Effects of a carotene-deficient diet on measures of oxidative
susceptibility and superoxide dismutase activity in adult women. Free Radic Biol Med
17(6): 537-44 (1994).
203. Dixon, Z. R., Shie, F.-S., Warden, B. A., Burri, B. J. & Neidlinger, T. R. The Effect of a
Low Carotenoid Diet on Malondialdehyde-Thiobarbituric Acid (MDA-TBA)
Concentrations in Women: A Placebo-Controlled Double-Blind Study. J Am Coll Nutr
17(1): 54-58 (1998).

231
204. Omaye, S. T. et al. Blood antioxidants changes in young women following beta-carotene
depletion and repletion. J Am Coll Nutr 15(5): 469-474 (1996).
205. Burri, B. J., Neidlinger, T. R. & Clifford, A. J. Serum Carotenoid Depletion Follows
First-Order Kinetics in Healthy Adult Women Fed Naturally Low Carotenoid Diets. J.
Nutr. 131(8): 2096-2100 (2001).
206. Re, R. et al. Antioxidant activity applying an improved ABTS radical cation
decolorization assay. Free Radic Biol Med 26(9-10): 1231-7 (1999).
207. Hu, M. L. Measurement of Protein Thiol Groups and Glutathione in Plasma. Methods
Enzymol 233380-385 (1994).
208. Draper, H. H. et al. Comparative Evaluation of Thiobarbituric Acid Methods for the
Determination of Malondialdehyde in Biological Materials. Free Radic Biol Med 15(4):
353-363 (1993).
209. Aebi, H. E. Catalase (ed. Bergmeyer, H. U.) (Verlag Chemie, Weinheim, 1983).
210. Misra, H. P. & Fridovich, I. The Role of Superoxide Anion in the Auto-oxidation of
Epinephrine and a Simple Assay for Superoxide Dismutase. J. Biol. Chem. 247(10):
3170-3175 (1972).
211. Pleban, P. A., Munyani, A. & Beachum, J. Determination of selenium concentration and
glutathione peroxidase activity in plasma and erythrocytes. Clin Chem 28(2): 311-6
(1982).
212. Cohn, W. et al. Comparative multiple dose plasma kinetics of lycopene administered in
tomato juice, tomato soup or lycopene tablets. Eur J Nutr 43(5): 304-12 (2004).
213. Pellegrini, N., Riso, P. & Porrini, M. Tomato consumption does not affect the total
antioxidant capacity of plasma. Nutrition 16(4): 268-271 (2000).
214. Stahl, W. et al. Carotenoid mixtures protect multilamellar liposomes against oxidative
damage: synergistic effects of lycopene and lutein. FEBS Lett 427(2): 305-8 (1998).
215. Lin, Y. et al. Estimating the concentration of beta-carotene required for maximal
protection of low-density lipoproteins in women. Am J Clin Nutr 67(5): 837-845 (1998).
216. Hsu, Y. M. et al. Characterizing the lipid-lowering effects and antioxidant mechanisms of
tomato paste. Biosci Biotechnol Biochem 72(3): 677-85 (2008).
217. Rao, A. V., Waseem, Z. & Agarwal, S. Lycopene content of tomatoes and tomato
products and their contribution to dietary lycopene. Food Research International 31(10):
737-741 (1998).
218. Paetau, I. et al. Chronic ingestion of lycopene-rich tomato juice or lycopene supplements
significantly increases plasma concentrations of lycopene and related tomato carotenoids
in humans. Am J Clin Nutr 68(6): 1187-95 (1998).
219. Paetau, I. et al. Chronic ingestion of lycopene-rich tomato juice or lycopene supplements
significantly increases plasma concentrations of lycopene and related tomato carotenoids
in humans. Am J Clin Nutr 68(6): 1187-1195 (1998).
220. Zhao, X. et al. Modification of lymphocyte DNA damage by carotenoid supplementation
in postmenopausal women. Am J Clin Nutr 83(1): 163-169 (2006).
221. Heinrich, U. et al. Supplementation with beta-carotene or a similar amount of mixed
carotenoids protects humans from UV-induced erythema. J Nutr 133(1): 98-101 (2003).
222. Collins, A. R., Olmedilla, B., Southon, S., Granado, F. & Duthie, S. J. Serum carotenoids
and oxidative DNA damage in human lymphocytes. Carcinogenesis 19(12): 2159-62
(1998).
223. Ghiselli, A., Serafini, M., Natella, F. & Scaccini, C. Total antioxidant capacity as a tool
to assess redox status: critical view and experimental data. Free Radic Biol Med 29(11):
1106-14 (2000).
232
224. Matos, H. R., Mascio, P. D. & Medeiros, M. H. Protective effect of lycopene on lipid
peroxidation and oxidative DNA damage in cell culture. Arch Biochem Biophys 383(1):
56-59 (2000).
225. Bhuvaneswari, V., Velmurugan, B., Abraham, S. K. & Nagini, S. Tomato and garlic by
gavage modulate 7,12-dimethylbenz[a]anthracene-induced genotoxicity and oxidative
stress in mice. Braz J Med Biol Res 37(7): 1029-1034 (2004).
226. Bose, K. S. & Agrawal, B. K. Effect of long term supplementation of tomatoes (cooked)
on levels of antioxidant enzymes, lipid peroxidation rate, lipid profile and glycated
haemoglobin in Type 2 diabetes mellitus. West Indian Med J 55(4): 274-8 (2006).
227. Madrid, A. E., Vasquez, Z.D., Leyton, A.F., Mandiola, C., Escobar, F. J.A. Short-term
Lycopersicum esculentum consumption may increase plasma high density lipoproteins
and decrease oxidative stress. Rev Med Chili 134(7): 855-62 (2006).
228. Collins, J. K. et al. Lycopene from two food sources does not affect antioxidant or
cholesterol status of middle-aged adults. Nutr J 315 (2004).
229. Breinholt, V. M. et al. Effects of dietary antioxidants and 2-amino-3-methylimidazo[4,5-
f]- quinoline (IQ) on preneoplastic lesions and on oxidative damage, hormonal status, and
detoxification capacity in the rat. Food Chem Toxicol 41(10): 1315-1323 (2003).
230. Sellmeyer, D. E., Schloetter, M. & Sebastian, A. Potassium Citrate Prevents Increased
Urine Calcium Excretion and Bone Resorption Induced by a High Sodium Chloride Diet.
J Clin Endocrinol Metab 872008-2012 (2002).
231. Seeman, E. & Delmas, P. D. Bone quality--the material and structural basis of bone
strength and fragility. N Engl J Med 354(21): 2250-61 (2006).
232. Martini, L. A., Booth, S. L., Saltzman, E., do Rosario Dias de Oliveira Latorre, M. &
Wood, R. J. Dietary phylloquinone depletion and repletion in postmenopausal women:
effects on bone and mineral metabolism. Osteoporos Int 17(6): 929-35 (2006).
233. Meunier, P. J. et al. Consumption of a high calcium mineral water lowers biochemical
indices of bone remodeling in postmenopausal women with low calcium intake.
Osteoporos Int 16(10): 1203-9 (2005).
234. Palacios, S. et al. Changes in bone turnover markers after calcium-enriched milk
supplementation in healthy postmenopausal women: a randomized, double-blind,
prospective clinical trial. Menopause 12(1): 63-8 (2005).
235. Brazier, M. et al. Markers of bone remodeling in the elderly subject: effects of vitamin D
insufficiency and its correction. J Bone Miner Res 10(11): 1753-61 (1995).
236. Eastell, R. & Hannon, R. A. Biomarkers of bone health and osteoporosis risk. Proc Nutr
Soc 67(2): 157-62 (2008).
237. Greenspan, S. L. et al. Early changes in biochemical markers of bone turnover predict the
long-term response to alendronate therapy in representative elderly women: a randomized
clinical trial. J Bone Miner Res 131431-1438 (1998).
238. Stepan, J. J. & Vokrouhlicka, J. Comparison of biochemical markers of bone remodelling
in the assessment of the effects of alendronate on bone in postmenopausal osteoporosis.
Clin Chim Acta 288(1-2): 121-135 (1999).
239. Greenspan, S. L., Rosen, H. N. & Parker, R. A. Early changes in serum N-telopeptide and
C-telopeptide cross-linked collagen type 1 predict long-term response to alendronate
therapy in elderly women. J Clin Endocrinol Metab 85(10): 3537-40 (2000).
240. Psotova, J., Zahalkova, J., Hrbac, J., Simanek, V. & Bartek, J. Determination of total
antioxidant capacity in plasma by cyclic voltammetry. Two case reports. Biomedical
Papers 145(2): 81-83 (2001).

233
241. Almstrup, K. et al. Dual effects of phytoestrogens result in u-shaped dose-response
curves. Environ Health Perspect 110(8): 743-8 (2002).
242. Li, H. L., Liu, D. P. & Liang, C. C. Paraoxonase gene polymorphisms, oxidative stress,
and diseases. J Mol Med 81(12): 766-79 (2003).
243. Rao, A. V., Ray, M. R. & Rao, L. G. Lycopene. Adv Food Nutr Res 5199-164 (2006).
244. Fontaine-Bisson, B. et al. Genetic polymorphisms of tumor necrosis factor-alpha modify
the association between dietary polyunsaturated fatty acids and fasting HDL-cholesterol
and apo A-I concentrations. Am J Clin Nutr 86(3): 768-74 (2007).
245. Hochberg, M. C. et al. Changes in bone density and turnover explain the reductions in
incidence of nonvertebral fractures that occur during treatment with antiresorptive agents.
J Clin Endocrinol Metab 87(4): 1586-92 (2002).
246. Key, L. L., Jr., Ries, W. L., Taylor, R. G., Hays, B. D. & Pitzer, B. L. Oxygen derived
free radicals in osteoclasts: the specificity and location of the nitroblue tetrazolium
reaction. Bone 11(2): 115-9 (1990).
247. Halliwell, B., Clement, M. V. & Long, L. H. Hydrogen peroxide in the human body.
FEBS Lett 486(1): 10-3 (2000).
248. Bax, B. E. et al. Stimulation of osteoclastic bone resorption by hydrogen peroxide.
Biochem Biophys Res Commun 183(3): 1153-8 (1992).
249. Rao, L. G., Liu, L. J., Murray, T. M., McDermott, E. & Zhang, X. Estrogen added
intermittently, but not continuously, stimulates differentiation and bone formation in
SaOS-2 cells. Biol Pharm Bull 26(7): 936-45 (2003).
250. Genty, C., Palle, S., Vanelle, L., Bourrin, S. & Alexandre, C. Colorimetric evaluation of
cultured osteoblast-like cells (ROS 17/2.8). Biotech Histochem 69(3): 160-4 (1994).
251. Lee, K. J. & Jeong, H. G. Protective effects of kahweol and cafestol against hydrogen
peroxide-induced oxidative stress and DNA damage. Toxicol Lett 173(2): 80-7 (2007).
252. Fatokun, A. A., Stone, T. W. & Smith, R. A. Hydrogen peroxide-induced oxidative stress
in MC3T3-E1 cells: The effects of glutamate and protection by purines. Bone 39(3): 542-
51 (2006).
253. Vali, B., Rao, L. G. & El-Sohemy, A. Epigallocatechin-3-gallate increases the formation
of mineralized bone nodules by human osteoblast-like cells. J Nutr Biochem 18(5): 341-7
(2007).
254. Liu, A. et al. Absorption and subcellular localization of lycopene in human prostate
cancer cells. Mol Cancer Ther 5(11): 2879-85 (2006).
255. O'Sullivan, L., Ryan, L. & O'Brien, N. Comparison of the uptake and secretion of
carotene and xanthophyll carotenoids by Caco-2 intestinal cells. Br J Nutr 98(1): 38-44
(2007).
256. Chichili, G. R. et al. Protective effects of tomato extract with elevated -carotene levels
on oxidative stress in ARPE-19 cells. British Journal of Nutrition 96(04): 643-649
(2006).
257. Lowe, G. M., Booth, L. A., Young, A. J. & Bilton, R. F. Lycopene and -carotene protect
against oxidative damage in HT29 cells at low concentrations but rapidly lose this
capacity at higher doses. Free Radical Research 30(2): 141 - 151 (1999).
258. Collins, E. S. in Institute of Medical Science 119 (University of Toronto, Toronto, 2005).
259. Jansen, K. L., Cole, T. B., Park, S. S., Furlong, C. E. & Costa, L. G. Paraoxonase 1
(PON1) modulates the toxicity of mixed organophosphorus compounds. Toxicol Appl
Pharmacol 236(2): 142-53 (2009).
260. Oda, M. N. et al. Paraoxonase 1 overexpression in mice and its effect on high-density
lipoproteins. Biochem Biophys Res Commun 290(3): 921-7 (2002).
234
261. Rozenberg, O., Shiner, M., Aviram, M. & Hayek, T. Paraoxonase 1 (PON1) attenuates
diabetes development in mice through its antioxidative properties. Free Radic Biol Med
44(11): 1951-9 (2008).
262. Shih, D. M. et al. Mice lacking serum paraoxonase are susceptible to organophosphate
toxicity and atherosclerosis. Nature 394(6690): 284-7 (1998).
263. Tward, A. et al. Decreased atherosclerotic lesion formation in human serum paraoxonase
transgenic mice. Circulation 106(4): 484-90 (2002).

235
 APPENDIX I

Informed consent package for clinical studies

(Most recent REB-approved version, 02/26/2007)

236
 FORM A
Selection, Inclusion and Exclusion Criteria

Oxidative Stress and the Risk of Osteoporosis: the Role of Antioxidant Lycopene
Selection and Inclusion Criteria:

Selection and Inclusion Criteria:

Study 1:
 Healthy women aged 25-79
Study 2:
 Healthy women whose menses have ceased at least one year prior to entry
 Aged 50-60

The women who are willing to participate should agree to provide fasting blood and urine
samples and maintain their dietary records when needed.

Exclusion Criteria:

Study 1:
 Those who take medications for osteoporosis, coronary heart disease, high
blood pressure and diabetes
 Those with other metabolic bone diseases.

Study 2:
 Those who smoke
 Those who are on hormone replacement therapy and on estrogen treatment
 Those who are on multi-vitamin supplements and herbal preparations
 Those who take medications, particularly for osteoporosis, coronary heart
disease, high blood pressure and diabetes
 Those with other metabolic bone diseases

237
 FORM B
Information Sheet for Participants

Oxidative Stress and the Risk of Osteoporosis: the Role of Antioxidant Lycopene
(CIHR/University-Industry Program)

Studies suggest that oxidative stress may play a role in the development of osteoporosis.
Although these studies suggest that the antioxidants vitamin C, E and beta-carotene may reduce
the risk of osteoporosis, little is known of the role played by lycopene, one of the most potent
dietary antioxidants found in tomatoes and tomato products. Recent studies have shown that high
dietary intake of lycopene is associated with a lower incidence of several chronic diseases.
However, there is a virtual lack of information concerning the status of oxidative stress and
antioxidants in persons at high risk for developing osteoporosis or those who already have the
disease. Dietary intervention strategies cannot be developed until such information is available.
Because women with high bone turnover markers have an increased risk of osteoporosis, finding
that their oxidative stress markers are also high would suggest that oxidative stress contributes to
this risk. Our laboratory is studying whether the antioxidant lycopene is beneficial to
postmenopausal women who are at risk of osteoporosis. Our major objectives are: Study 1:
To measure oxidative stress parameters and antioxidant status in fasting blood and urine samples
of healthy women aged 25 to 79 years and correlate these parameters with their bone turnover
markers (a measure of how fast their bone forms and gets removed). Study 2: To correlate the
subjects’ intake of tomato products and blood lycopene levels, with oxidative stress parameters
and bone turnover markers. Study 1 and Study 2: To determine whether genetic variations in
enzymes that protect against oxidative stress can explain any of the associations that are
observed.

Healthy women between 25 to 79 years will be asked to participate in study 1, and

postmenopausal women between 50 to 60 years will be asked to participate in study 2. In Study
1, the participants will be asked to keep a 7-day food record, fast overnight on day 7 and then
have their blood and urine collected and used to measure the above parameters. In Study 2, the
participants will be asked to refrain from consuming lycopene-rich products (list to be provided)
for 4 weeks prior to and during the duration of the study and then given regular tomato juice,
lycopene-rich tomato juice, placebo capsules or tomato lycopene capsules twice daily for a
period of 4 months. Fasting blood and urine samples will be taken after a 4-week wash-out
period and at 2 and 4-months after the start of treatment.

This novel pilot study should provide initial clinical data on which to base future studies of the
effects of dietary lycopene on the development of osteoporosis.

If you have any questions, please call the following telephone numbers:

1. Dr. Leticia G. Rao, Assistant Professor of Medicine and Co-Director, Calcium Research
Laboratory, Division of Endocrinology and Metabolism, St. Michael’s Hospital, 38 Shuter
St. Annex, Toronto, Ontario M5B 1A6. Tel #: (416) 864-5838, Mon – Fri, 9:00 AM – 5:00
PM.

238
2. Dr. A.V. Rao, Professor of Nutrition and Director, Program in Food Safety, Department of
Nutritional Sciences, University of Toronto, Toronto, Ontario, Tel #: (416) 978-3621, Tue –
Thurs, 9:00 AM – 12:00 PM.
3. Ms. Erin Mackinnon, PhD student at the Institute of Medical Science, University of Toronto.
Supervisor: Dr. L.G. Rao. Tel #: (416) 439-5838, Mon – Fri, 9:00 AM – 5:00 PM

The following investigators can also be consulted if necessary:

4. Dr. T.M. Murray, Professor, University of Toronto, 38 Shuter St. Annex, Toronto, Ontario
M5B 1A6.
5. Dr. R.G. Josse, Professor and Director of the Osteoporosis Centre and Associate Physician-
In-Chief, St Michael’s Hospital and Department of Medicine, University of Toronto. 61
Queen St East. Toronto, Ontario.
6. Dr. A. El-Sohemy, Assistant Professor, Department of Nutritional Sciences, University of
Toronto, Toronto, Ontario.
7. Dr. C. Derzko, Associate Professor of Medicine, University of Toronto and Head,
Multidisciplinary Menopause Program, St. Michael’s Hospital, 61 Queen St East Toronto,
Ontario M5C 2T2
8. Dr. S. Jamal, Assistant Professor of Medicine, University of Toronto and Endocrinologist, St.
Michael’s Hospital. 61 Queen St. East, Toronto, Ontario M5C 2T2.

The Sponsors for the study are as follows:

1. Canadian Institute of Health Research (CIHR)

2. Kagome Co. Ltd, Japan
3. LycoRed Natural Products Industries Ltd, Israel
4. H.J. Heinz Company, Canada
5. Millenium Biologix Inc, Canada

239
 FORM C
Consent to Participate in a Research Study

Title of Research: Oxidative stress and bone health: clinical studies on the role of the
antioxidant lycopene

For information on investigators and study sponsor, please see Form B

Purpose of the Research: St. Michael’s Hospital and the University of Toronto are carrying out
research in order to understand the role of oxidative stress and antioxidants in osteoporosis.

Description of the Research:

Healthy women who are between the ages of 25-79 can participate in study 1. For study 2,
women who are at least 1 year postmenopausal and are between the age of 50 and 60 years can
participate. Please see Form B- Information Sheet for Participants which describes the study in
greater detail.

Study 1: If you chose to participate, you will be required to attend 2 visits of

approximately ½ hour each, either at 61 Queen St E, 7th floor or 3 Shuter Wing, suite 3-046. On
first visit, you will be given several forms to take home and will be given instructions on how to
complete them. You will be asked to maintain a record of your food intake for a period of 7 days
on the forms provided. You will be given an empty vial for urine collection. On day 7, you will
be asked to fast overnight (twelve hours prior to your appointment). The following day you will
be asked to come to the clinic with your second void morning urine sample (this can be collected
at the clinic if necessary) and the completed food records. Body weight, height, pulse rate and
blood pressure will be taken and fasting blood samples will be collected (approximately 6
tablespoons).

Study 2: If you chose to participate, you will be required to attend 3 visits of

approximately ½ hour each, either at 61 Queen St E, 7th floor or 3 Shuter Wing, suite 3-046.
You will be asked to refrain from eating foods that are good sources of lycopene, particularly
tomatoes or tomato-based products (list in Appendix 3, Form F) during a wash-out period of 4
weeks prior to and throughout the 4-month study period. You will be asked to keep a diary of
your daily food intake during the last 7 days of your 4-week wash-out period prior to the start of
intervention, 7 days prior to the 2-month-visit and 7 days prior to the 4-month visit. (Appendix 3,
Forms E1-2). You will be randomly assigned to 4 groups: group 1 will be asked to take placebo
capsules and group 2, 15-mg lycopene capsules (Lyc-O-Mato, LycoRed, Israel); group 3 will be
asked to drink tomato juice (H.J. Heinz, Canada) containing the regular amount of 15 mg
lycopene and group 4, tomato juice containing 35 mg lycopene (Kagome Co, Japan). All
treatments to be taken twice daily for the next 4 months.

Fasting blood and urine samples will be collected basally (at the end of the wash-out period) and
after 2 and 4 months. At each visit to the clinic, food intake records will be collected.
The blood and urine samples will be used for the analyses of antioxidants, oxidative
stress markers and bone turnover markers. The remainder of the blood and urine samples will be
frozen at –80oC at St. Michael’s Hospital for potential future studies related to oxidative stress,
including the study of genetic markers related to antioxidant enzymes. These samples will be
kept for 5 years and the specimens will be destroyed after the 5 year storage period.
240
Potential Harms (Injury, Discomforts or Inconveniences): When blood is taken from the
vein, a slight discomfort or redness may be experienced which will usually disappear in a few
days.

Potential Benefits: There are no direct benefits to you from this study other than the satisfaction
of participating in this research on diet and risk of osteoporosis.

Reimbursement/Compensation: Parking fee, bus or TTC fare at the day of visit will be
reimbursed. There is no compensation for those who only complete Study 1. Because of the
length of time and dietary restrictions involved in Study 2, participants will be provided with
$50.00 of compensation for every appointment in which blood, urine and dietary records are
collected for a total of $150.00 compensation.

Confidentiality and Privacy: Confidentiality will be respected and no information that

discloses your identity will be released or published without your consent unless required by law.
For genetic studies, records of data collected during this research relating to you and your care
will be kept confidential. No information that discloses your identity will be released or
published without your consent. Information regarding the results of this research can be sent to
your family physician at your request. The genetic material from the blood and urine samples
isolated as a result of this research will be stored anonymously so that, as new genes are
discovered which are implicated in osteoporosis, it will possible for research in this area to
continue.

Publication of Results: The results of this study will be published in scientific journals and/or
presented at conferences, seminars or other public forums without breaking the confidentiality
and privacy as stated above. Your identity will not be disclosed in any presentations or
publications of the results of the study.

Participation and Withdrawal: Participation in research is voluntary. If you choose not to

participate, you will continue to have access to customary care at St Michael’s Hospital. If you
choose to participate in this study you can withdraw from the study at any time without any
effect on the care you will receive at St Michael’s Hospital. The research findings can be made
available to you upon completion of the research. If you wish to withdraw your consent after you
have completed the study you may contact the Principal Investigator, Dr. Leticia Rao, at (416)
864-5838, and your samples and all other information you provided during your participation
will be destroyed immediately.

Research Ethics Board Contact: If you have any questions as a research participant you may
contact Dr. Julie Spence, Chair of the Research Ethics Board at (416) 864-6060 Ext 2557

241
 FORM C - continued

Signatures of Participants, Person Obtaining the Consent and Investigator

Title of Research: Oxidative stress and the risk of osteoporosis: the role of the antioxidant
lycopene

Consent: I acknowledge that the research study described above and the nature, purpose and
possible adverse effects of this study have been explained to me. I have read the consent form
and understand the procedures to be undertaken. I agree to participate in the study with the
knowledge that my participation is voluntary and that I may withdraw from the study at any
time.

I hereby consent to participate as shown below, and have been given a copy of this consent form:

I agree to participate in Study 1: Yes No

I agree to participate in Study 2: Yes No
I may be contacted for other future studies related to bone health Yes No
Date: I.D. Code:___________________

Name of Participant (Print) Signature Date

______ _____________

Name of Person obtaining the consent Signature Date

Ms. Erin Mackinnon, MSc. ____________________ _____________

PhD candidate,
Institute of Medical Science

Name of Investigator Signature Date

Dr. L.G. Rao, Ph.D. ____________________ _____________

Associate Professor of Medicine
University of Toronto and
Co-Director, Calcium Research Laboratory, St. Michael’s Hospital

242
 FORM D
Participants Source Document

Oxidative Stress and the Risk of Osteoporosis: the Role of Antioxidant Lycopene

Date _____________

Name I.D. No. _______________

Address ________________________ Tel. No. ________________

_________________________________
_________________________________

Health Status

Age Weight Height Pulse Rate _______ Blood Pressure _______

Record of consumption or use of the following items:

Please check the appropriate line
Caffeine (coffee/tea/soda consumption in Supplements: vitamins, herbals
cups/day): ___ No
___ 0 ___ Yes
___ 1-2 if yes, please include the amount, frequency
___ 3-4
___ 5+ and brand in the 7-day food record

Smoking: Alcohol consumption:

___ Non-smoker Beer ____ glass/week
___ Ex-smoker. When did you stop Wine ___ glass/week
smoking? ________ Hard liquor _____ oz/week
____ Smoker ___
____ ½ pack/day Hormone Replacement Therapy
____ 1 pack/day ______ Never ____ Stopped, ______when
____ 2+ packs/day _______ Yes ___________date started

_____ Occasional. When did you Other Estrogen users (including oral
last smoke?_________ contraceptive pill) :
______ date started

6. Notes _________________________________________________

243
 FORM E-1
Instructions for Completing
The Food Record FORM E-2

Oxidative Stress and the Risk of Osteoporosis: the Role of Antioxidant Lycopene

Please record all food consumed for the period of 7 days. It is important to maintain your usual
eating habit. If you have any questions, please call either Erin Mackinnon at (416) 864-5805, Dr.
L.G. Rao at (416) 864-5838, or Dr. A.V. Rao at (416) 978-3621 or leave a message and your call
will be returned promptly.

1. Ideally, foods and beverages should be recorded as soon as possible after consumption.

2. A complete description of the food will increase the accuracy of your assessment.

e.g., baked/boiled/pan-fried/deep-fried
Raw/steamed/canned
McDonald’s/Kellogg’s/Lean Cuisine

3. Please record the amount of the food or beverage consumed.

e.g., 1 slice of white bread 1 cup of Kellogg’s corn flakes

2 heaping tsp of brown sugar 8 oz 2% milk
2 creamers

Meat portions are often difficult to estimate. If you are having difficulty, measure the item’s
dimension (e.g., roast beef, 5x1/2 “x 41/2" x 1/4"). Also, 1 deck of playing cards is the same
size as 3 oz of cooked meat.

4. For combination dishes, please detail the individual ingredients or provide recipes.
Correct Incorrect
1 medium plain bagel 1 cheese sandwich
1 tsp margarine
2 oz cheddar cheese
1-1/2 cups cooked white rice Chicken stir fry
1/2cup broccoli, boiled
2 oz chicken breast
1 tsp soy sauce

5. For vitamins and other supplements please record the type of supplement, amount and
frequency taken, and the brand name. This information can be attached separately if
desired.

244
 FORM E-2 (day 1)1
Food Record

Oxidative Stress and the Risk of Osteoporosis: the Role of Antioxidant Lycopene

Date ________ Name_______________ I.D. No__________________

Day 1
Time Food/Beverage and Description (one item/line) Include Quantity For Official
eaten vitamin supplements when applicable (cups, ml, g, Use Only
tsp, etc)

Is this a usual day? Yes No_ _

If “No”, please explain_____________________________________________________________

1
This form E-2 provided similar pages for days 2 through to 7, which have been omitted from
this Appendix

245
 FORM F
Foods to Avoid by Participants of Study 2

Oxidative Stress and the Risk of Osteoporosis: the Role of Antioxidant Lycopene

Please avoid the following lycopene-containing foods during the treatment and wash-out phases
of Study 2:

Chilli sauce
Chinese (blood) orange
Clam cocktail
Pink guava
Pizza sauce
Red grapefruit
Rosehip (puree and tea)
Salsa
Spaghetti sauce
Tomato juice
Tomato juice in Bloody Mary mix
Tomato ketchup
Tomato paste
Tomato puree
Tomato sauce
Tomato soup
Tomato-containing BBQ sauce
Tomatoes (canned and fresh)
Vegetable juice (V8)
Watermelon

246
 APPENDIX II

Summary of intervention effects on oxidative stress

parameters, bone turnover markers and antioxidant capacity
resulting from supplementation with regular tomato juice,
lycopene-rich tomato juice and tomato lycopene capsules

247
Table II.1: Summary of effects on serum lycopene, antioxidant capacity, oxidative stress and the
bone resorption marker NTx, in participants supplemented with either regular tomato juice,
lycopene rich tomato juice or tomato lycopene capsules for four months.

Parameter Statistical test p value for supplement group

measured 1 Regular Lycopene-rich Tomato
tomato juice tomato juice lycopene
capsules
↑ Total serum % change after 248.2  45.58 305.5  77.22 351.9  65.83
lycopene (all-trans + 4 months
cis)
ANOVA 2 <0.0001 <0.0001 <0.0001
0 vs. 4 mos. 3 <0.001 <0.001 <0.001
↑ Total antioxidant % change after 8.34  6.39 18.14  4.41 -1.48  3.60
capacity 4 months

ANOVA not significant <0.05 not significant

0 vs. 4 mos. not significant <0.01 not significant
↑Protein thiols % change after 17.87  6.34 9.04  6.79 19.76  6.47
(↓ protein oxidation) 4 months

ANOVA <0.05 not significant <0.05

0 vs. 4 mos. <0.05 not significant <0.05
↓Lipid peroxidation % change after -14.94  4.33 -7.63  3.60 -13.21  3.19
(TBARS) 4 months

ANOVA <0.001 not significant <0.05

(<0.10)
0 vs. 4 mos. <0.01 not significant <0.05
↓ Bone resorption % change after -15.37  7.02 -10.44  4.36 -19.46  6.61
marker (NTx) 4 months

ANOVA <0.005 <0.0005 <0.005

0 vs. 4 mos. <0.05 <0.01 <0.01
1
Arrow indicates direction of effect: ↑ indicates an overall increase, ↓ indicates an overall decrease.
2
Repeated-measures ANOVA was performed for each individual supplement group.
3
as determined by Tukey post tests as part of repeated measures ANOVA.

248
 APPENDIX III

Example diagrams of HPLC chromatogram and allelic

discrimination plots used to determine PON1 genotype

249
 AU

-0.0010

-0.0005

0.0000

0.0005

0.0010

0.0015

0.0020

0.0025

0.0030
2.00
4.00
4.905 4.708
6.00
8.00
10.00

11.348
12.00

12.815
14.00
16.00

17.163
Minutes

17.628
18.00

18.085
19.073
20.00
22.00
24.00
26.00

26.177
28.00

28.519
30.00
32.00
34.00

Figure IIIa: Example chromatogram showing peaks for the carotenoids lutein, zeaxanthin,
-cryptoxanthin, echinenone standard, all-trans lycopene, 5-cis lycopene, other-cis lycopene
(2 peaks), -carotene, and -carotene represented by the numbers 4.708, 4.905, 11.348,
12.815, 17.163, 17.628, 18.085, 19.073, 26.177 and 28.519, respectively.

250
Figure III b: Example of allelic discrimination plot used for participant genotyping for the
172TA polymorphism

251
Figure IIIc: Example of allelic discrimination plot used for participant genotyping for the
584AG polymorphism

252
 APPENDIX IV

Trend graphs for oxidative stress parameters, antioxidant

capacity, antioxidant enzymes and NTx from baseline,
through to the end of the through to the end of the lycopene
intervention study period: combined cross-sectional and
intervention study data

253
 INTRODUCTION, OBJECTIVES and METHODS

Dietary restriction of lycopene resulted in significantly increased NTx and GPx, with

significantly depressed concentrations of CAT and SOD and non-significant decreases in protein

and lipid oxidation (Chapter 4). Lycopene restriction is necessary to obtain a clinically relevant

response to lycopene, as it has been shown that higher baseline lycopene concentrations result in

a moderate to null response to lycopene 201. This thesis has also shown that supplementation with

lycopene, in the form of tomato lycopene capsules, regular tomato juice, or lycopene-rich tomato

juice resulted in significantly increased total antioxidant capacity, while simultaneously

decreasing biomarkers of protein and lipid oxidation and NTx (Chapter 5). A high dose of

lycopene not usually obtained through the daily diet was shown to result in lower TBARS and

NTx (Chapter 6). This is an investigation to determine the trends of these changes from baseline,

at which the normal daily diet was consumed, through the washout period, until the end of the 4

month lycopene-supplemented period. Further, since these findings were unable to show any

significant effects of lycopene supplementation on antioxidant enzymes (Chapter 5), it is

important to examine the trends which occur from baseline, when the daily diet is consumed, to

the end of the lycopene intervention study, to garner more information about the changes that are

taking place.

Seventeen postmenopausal women completed both the cross-sectional study and the

intervention study, providing us with information about their usual daily intake prior to

participating in the intervention study. The usual daily intake of these participants was 3.54 

0.80 mg of lycopene per day. Following a washout period, during which the average lycopene

intake was 0.14  0.08 mg per day, participants consumed 43.53  4.89 mg of lycopene for a

period of 4 months. Serum samples were used to measure total antioxidant capacity, NTx,

254
protein thiols, TBARS, CAT, SOD and GPx as described in Chapters 4-6. Figures IVa and IVb

show the trends of these changes from baseline through to the end of the supplement period.

RESULTS AND DISCUSSION

As seen in Figure IVa, serum lycopene was markedly decreased after the one month washout

period, but increased to a concentration over and above that of the baseline value (about 126%

higher than the baseline). Similarly, antioxidant capacity was approximately 4.7% higher after

supplementation relative to baseline. As expected from findings reported in Chapters 4 and 5,

from baseline to washout, NTx and protein and lipid peroxidation parameters noticeably

increased. As serum lycopene was replenished with the intervention, oxidative stress parameters

and NTx began to decrease to concentrations less than those at baseline, with NTx at a

concentration 5.6% lower and protein and lipid oxidation at 7.8% and 8.8% lower than that of

baseline. These findings suggest that the higher concentrations of serum lycopene resulting from

intervention provided a greater benefit of decreased biomarkers of oxidative stress and bone

resorption than those obtained through the daily diet alone. As outlined in Chapter 6, this is most

likely due to the significant increase in the serum concentration of the high antioxidant capacity

5-cis lycopene which occurs with lycopene supplementation.

The trends occurring in antioxidant enzymes suggest that the proposed compensatory

mechanisms as presented in Chapter 4 may not be reversed upon replenishment with lycopene

for 4 months. This may explain why there was no significant change in GPx after lycopene

intervention (Chapter 5), but in participants who have maintained their usual daily lycopene

intake, the high daily intake group had significantly higher GPx (23.92  3.26) than the low daily

intake group (15.44  1.40) (p<0.05, Chapter 6). Further, the changes in the antioxidant enzymes

255
over time showed quite a bit of variability. This may be due to the relatively small sample size.

However, similar variability was found from the data presented in Chapter 5, suggesting that

there may be different factors contributing to the concentrations of antioxidant enzymes which

were eliminating any potential effects lycopene supplementation for 4 months. Specifically,

polymorphisms in the antioxidant enzymes themselves may affect their response to lycopene and

could account for the lack of significant effect shown here. This is an idea which may warrant

further investigation as it is beyond the scope of this thesis.

Total serum lycopene (nM)

2400 2.00

Trolox (mM)
1800 1.75

1200
1.50

600
1.25
e

ut

os
.
ut
e

os

os
os

in
in

ho
ho

m
m

el

m
m
el

as
as
as

4.
as

4.

2
2

W
B
W
B

Study period Study period

Figure IVa: Trend graphs showing change in antioxidant capacity including serum lycopene and
total antioxidant capacity (Trolox), for 17 postmenopausal women, from baseline, at which the
normal daily diet was consumed, providing an average of 3.54  0.80 mg of lycopene per day,
through to the end of the lycopene supplement period, during which an average of 43.53 ± 4.89
mg/day was provided in the form of tomato lycopene capsules, regular tomato juice, or lycopene
rich tomato juice, from the washout study period to the 4 month study period.

256
 500

Protein thiols ( M)

450

400

350

ut

os
.
os
in

ho

m
el

m
as
as

4.
2
W
B
Study period
12.5
TBARS (nmol/ml)

10.0

7.5

5.0
e

ut

os
.
os
in

ho

m
el

m
as
as

4.
2
W
B

Study period
30
NTx (nM BCE)

25

20

15
e

ut

os
.
os
in

ho

m
el

m
as
as

4.
2
W
B

Study period

Figure IVb: Trend graphs showing change in protein thiols, TBARS and NTx, for 17
postmenopausal women, from baseline, at which the normal daily diet was consumed, providing
an average of 3.54  0.80 mg of lycopene per day, through to the end of the lycopene supplement
period, during which an average of 43.53 ± 4.89 mg/day was provided in the form of tomato
lycopene capsules, regular tomato juice, or lycopene rich tomato juice, from the washout study
period to the 4 month study period.

257
 40

GPx (U/g Hb)

30

20

10

ut

os
os
in

ho

m
el

m
as
as

4.
2
W
B
100
CAT (K/g Hb)

75

50
e

ut

os
os
in

ho

m
el

m
as
as

4.
2
W
B

60
SOD (U/mg Hb)

40

20
e

os
ut

.
os
in

ho

m
m
el

as
as

4.
2
W
B

Study period

Figure IVc: Trend graphs showing change in antioxidant enzymes for 17 postmenopausal
women, from baseline, at which the normal daily diet was consumed, providing an average of
3.54  0.80 mg of lycopene per day, through to the end of the lycopene supplement period,
during which an average of 43.53 ± 4.89 mg/day was provided in the form of tomato lycopene
capsules, regular tomato juice, or lycopene rich tomato juice, from the washout study period to
the 4 month study period.

258
 APPENDIX V

Comments on the PON1 polymorphisms of the human CD34+

cells used for the in vitro work presented in Chapter 9

259
 OBJECTIVES AND METHODS

There was a significant interaction between lycopene and PON1 polymorphisms on oxidative

stress parameters and bone turnover markers in women (Chapters 7 and 8). The in vitro work

presented in Chapter 9 of this thesis suggested that cis lycopene was capable of preventing and

repairing the damaging effects of ROS in haematopoietic cells with osteoblast-like

characteristics (CD34+). Genotyping of these osteoblasts was performed to determine whether

the PON1 polymorphisms may substantiate the findings reported in Chapter 9 as well as the

interaction between PON1 polymorphisms and lycopene as reported in Chapters 7 and 8.

The following methods were used: (1) Cells were plated at a density of 5 x 105 cells in a

75 cm2 flask. Medium was changed every 48 hours, and on day 7, cells were passaged by

trypsinization and the resultant cellular pellet was recovered for DNA extraction (for full details

on cell culture techniques please refer to Chapter 9), (2) DNA was extracted using the Genomic

DNA Isolation Kit according to the manufacturer’s protocol for cultured cells (NORGEN Biotek

Corporation, Thorold, ON), and (3) Allelic discrimination of the PON1 polymorphisms

584AG (rs662) and 172TA (rs854560) was performed using TaqMan® SNP Genotyping

Assays and the StepOnePlus™ Real-Time PCR system (Applied Biosystems, Canada).

260
 RESULTS AND CONCLUSIONS

The human CD34+ cells used for the in vitro work presented in Chapter 9 of this thesis possess

the following PON1 172TA and 584AG genotypes: 172TA and 584AG. The combined

genotype, as outlined in Chapter 7, is 1/1. A significant interaction between the combined PON1

genotype and lycopene in women was described in Chapter 7, suggesting that this genotype may

not respond as well to lycopene with decreased oxidative stress parameters and bone turnover

markers. This could explain the lack of significant effect of lycopene alone on osteoblast cells

(cells treated with lycopene from days 8-10 of culture, Figures 9.7 and 9.8, Chapter 9). Perhaps

use of a cell line with the combined genotype 0/1, which was shown to have the preferential

response to lycopene (Table 7.5, Chapter 7), would demonstrate a significant effect of 48 hour

treatment of lycopene alone on both the number and area of mineralized nodules in human

osteoblast cells. The ability of lycopene to prevent and repair the damaging effects of ROS also

may have been significantly better if the genotype 0/1 (shown to respond best to lycopene) were

used. This was not investigated using multiple cell lines possessing different genotypes to

determine whether or not this was the case, however, this is an idea warranting further

exploration in future studies.

261
 Life is an occasion
Rise to it.

262