Oxidative Stress as a Cardiovascular Risk Factor in...

Oxidative Stress as a Cardiovascular Risk Factor in Canadian Inuit

Dalal Usamah Zaid Alkazemi

Doctor of Philosophy

School of Dietetics and Human Nutrition

McGill University

Montréal, Québec, Canada

August 15, 2012

A thesis submitted to McGill University in partial fulfillment of the requirements

of the degree of Doctor of Philosophy

© Copyright Dalal UZ Alkazemi, 2012

All rights reserved

ii

Abstract

A decline in the traditional food consumption among the Inuit poses a nutritional

problem, as it provides nutrients associated with protection from cardiovascular

diseases (CVD). Controversy exists, however, regarding oxidative stress status

among Inuit due to possible higher tissue oxidizability of n‐3 polyunsaturated

fatty acids (PUFA) from the traditional Inuit n‐3 PUFA rich diet and their dietary

exposure to pro‐oxidative contaminants such as methylmercury and

polychlorinated biphenyls (PCBs). We proposed that determining whole body

oxidative stress status via the gold standard isoprostane biomarker that reflects

an overall response to the various metabolic and xenobiotic burdens could be an

indicator of the adaptations and predispositions of the Inuit to CVD. The

objectives of this thesis were: (i) to quantify plasma levels of F2‐isoprostanes, F3‐

isoprostanes, and isofurans and to assess their relationship to cardiometablic

risk factors; (ii) to evaluate the relationship between the selenium (Se) status

and plasma isoprostanes and to assess the interrelationship between dietary

methylmercury exposure and tissue selenium on oxidative stress status; (iii) to

examine the impact of decreased consumption of traditional foods on oxidative

stress parameters; and (iv) to evaluate the pro‐oxidant contribution of PCBs on

oxidative stress status. A series of cross‐sectional studies were conducted using

data from the International Polar Year Inuit Health Survey (2007‐2008). Cross‐

sectional analyses were conducted to examine the relationship between diet (Se,

n‐3 and n‐6 PUFAs, trans and saturated fat), contaminants (MeHg and PCBs),

obesity, and cardiovascular risk factors including hypertension, dyslipidemia,

smoking and C‐reactive protein in relation to oxidative stress using the gold

standard F2‐isoprostanes and the novel isoprostane biomarkers F3‐isoprostanes

and isofurans. Plasma levels of isoprostanes were determined by gas

chromatography/negative ion chemical ionization/mass spectrometry

methodology. Multivariate analyses were employed to determine final correlates

of isoprostanes. Results showed that the Inuit are protected from mercury‐

induced oxidative stress because of their high Se status and that highest

iii

consumers of traditional foods were most protected. The isoprostane isoforms,

F2‐isoprostanes and isofurans, were both associated with increased obesity. F2‐

isoprostanes and isofurans were additionally associated to C‐reactive protein

and blood pressure. The F3‐isoprostane isoform was shown to be quantifiable for

the first time in human plasma, which was likely due to the C20:5n‐3 rich Inuit

diet. F3‐isoprostanes and isofurans were more related to contaminants than F2‐

isoprostanes. Generally, this thesis demonstrates that the contaminants in the

Arctic do not appear to pose a significant oxidative stress risk among the Inuit;

however, shifting away from the traditional food is associated with obesity and

inflammatory‐induced oxidative stress. An important finding is that each

isoprostane isoform has a unique attribute to explain the potential CVD risk and

that simultaneous measurement of isoprostane species provides an

advantageous mechanistic insight into oxidative stress status not captured by F2‐

isoprostanes alone.

iv

Résumé

Une baisse dans la consommation des aliments traditionnels chez les Inuits pose

un problème nutritionnel car elle fournit des nutriments associés à la protection

contre les maladies cardio‐vasculaires (MCV). Il existe une controverse,

cependant, en ce qui concerne le stress oxydatif chez les Inuits en raison de

l’oxydabilité des tissus plus élevée possible de n‐3, des acides gras polyinsaturés

(AGPI) de l'alimentation traditionnelle des Inuits rich en AGPI n‐3, et leur

exposition alimentaire à des pro‐oxydants et des contaminants tels que le

méthylmercure et les biphényles polychlorés (BPC). Nous avons proposé que la

détermination de tout état de stress oxydatif du corps via la méthode de l'étalon‐

or isoprostanes, qui reflètent une réponse globale aux différentes charges

métaboliques et des xénobiotiques, puisse être un indicateur des adaptations et

des prédispositions des Inuits aux maladies cardiovasculaires. Les objectifs de

cette thèse sont les suivants: (i) de quantifier les concentrations plasmatiques de

F2‐isoprostanes, F3‐isoprostanes, et isofurans et d'évaluer leur relation avec les

facteurs de risque cardiometabolique (ii) d'évaluer le lieu entre le statut en

sélénium et les isoprostanes plasmatiques et d'évaluer la relation entre

l'exposition au méthylmercure et du sélénium alimentaire tissus sur un état de

stress oxydatif, (iii) d'examiner l'impact de la diminution de la consommation

d'aliments traditionnels sur les paramètres du stress oxydatif, et (iv) d'évaluer la

contribution des PCB sur le proxidant d’un état de stress oxydatif. Une série

d’études transversales ont été réalisées en utilisant les données de l'Enquête sur

l'Année Polaire Internationale de la Santé des Inuits (2007‐2008). Les analyses

transversales ont été menées pour étudier la relation entre l'alimentation

(sélénium (Se), n‐3 et n‐6 AF, gras trans et saturés), les contaminants (MeHg et

les PCB), l'obésité et les facteurs de risques cardiovasculaires comme

l'hypertension, la dyslipidémie, le tabagisme et la protéine C réactive par rapport

à un stress oxydatif avec l'étalon‐or F2‐isoprostanes et les nouveaux

biomarqueurs isoprostane F3‐isoprostanes et isofurans. Les concentrations

plasmatiques des isoprostanes ont été déterminées par chromatographie en

v

phase gazeuse, par l’ionisation négative chimique d'ions et par la méthode de

spectrométrie de masse. Des analyses variées ont été utilisées pour déterminer

les corrélats finals de isoprostanes. Les résultats ont montré que les Inuits sont

protégés par le mercure causé par le stress oxydatif en raison de leur statut en

sélénium élevée, et que les consommateurs d'aliments traditionnels ont été les

plus protégés. Tous les isoformes isoprostanes ont été associés à l'obésité. Les

F2‐isoprostanes et isofurans ont en outre été associés à la protéine C réactive et

de la pression artérielle. Aussi, les F3‐isoprostanes et isofurans sont davantage

liés aux contaminants que les F2‐isoprostanes. En règle générale, cette thèse

montre que les contaminants dans l'Arctique ne présentent aucun risque sur le

stress oxydatif, mais que l’abandon aujourd'hui de la nourriture traditionnelle

est associé à l'obésité et le stress induit par l'oxydation inflammatoire. Une

constatation importante est que chaque isoprostane possède un attribut unique

qui explique le risque potentiel de maladies cardio‐vaculaires, et une mesure

simultanée des isoprostanes donne un aperçu mécanistique avantageuse dans le

stress oxydatif pas capturé par les F2‐isoprostanes

vi

ADVANCE OF SCHOLARLY KNOWLEDGE

1. Original contribution to knowledge

This doctoral dissertation is the first study to comprehensively assess the

oxidative stress status among the Inuit using the gold standard biomarker F2‐

isoprostanes, and determine its correlates related to cardiometabolic risk factors

and the environmental contaminants methylmercury and polychlorinated

biphenyls (PCBs) that bioaccumulate in Arctic regions. Greater understanding of

oxidative stress status was achieved by quantifying novel isoprostanes isomers

that may competitively reduce F2‐isoprostanes generation under various

physiological conditions that may be associated with decreased oxygen tension

such as smoking; and by nutritional profiles associated with high n‐3

polyunsaturated fatty acids (PUFAs) intake. In addition, the simultaneous

measurement of these isoprostane isoforms provided mechanistic insights of the

“cardioprotective” advantage the Inuit may possess despite environmental

contaminant health risks. The quantification of plasma F2‐isoprostanes, F3‐

isoprostanes and isofurans, provided new evidence to support previous

suggestions that despite recurrent concerns over mercury and PCBs

environmental‐health risks, the traditional Inuit diet provides nutritional

antioxidant benefits that can counteract some environmental contaminant‐

induced health risks. In addition, we showed for the first time direct evidence

that the co‐presence of selenium and n‐3 PUFAs in the traditional Inuit foods

may be potential risk modifiers of cardiometabolic deterioration. Despite this

latter observation, we also showed that the Inuit might not be fully protected

from health risks associated with contaminants; especially this is for the younger

generation that continues to shift away from consuming traditional foods.

vii

2. Research publications in refereed scientific journals

Manuscript 1:

ISOPROSTANES AND ISOFURANS AS NON‐TRADITIONAL RISK FACTORS FOR

CARDIOVASCULAR DISEASE AMONG CANADIAN INUIT

Dalal Alkazemi, Grace M Egeland, L. Jackson Roberts II, Stan Kubow

In Press: Free Radic Res. doi:10.3109/10715762.2012.702900

[Online July 11, 2012].

3. Research publications to be submitted to refereed scientific journals

Manuscript 2:

NEW INSIGHTS REGARDING TISSUE SELENIUM AND MERCURY INTERACTIONS

ON OXIDATIVE STRESS FROM PLASMA ISOPROSTANE AND ISOFURAN

MEASURES IN CANADIAN INUIT

Dalal Alkazemi, Grace Egeland, L. Jackson Roberts II, H.M. Chan, Stan Kubow

Manuscript 3:

NOVEL EICOSAPENTAENOIC ACID‐DERIVED F3‐ISOPROSTANES AS

BIOMARKERS OF LIPID PEROXIDATION IN THE CANADIAN INUIT POPULATION

Dalal Alkazemi, Grace Egeland, L. Jackson Roberts II, H.M. Chan, Stan Kubow

viii

Contributions of Authors:

Thesis topic was conceived by the committee, and collaboration with Vanderbilt

University provided training for the laboratory protocol for isoprostane

biomarker extraction and derivatization. The laboratory work and data

statistical analyses were performed primarily by the candidate. In addition, the

candidate was the primary author and wrote the first draft for all manuscripts

with inputs from Dr. Kubow for the hypotheses generation. Dr. Kubow also

provided expertise in metabolism and oxidative stress, and he extensively edited

all manuscripts and supervised the scientific integrity of the work.

Dr. Grace M. Egeland provided the Inuit data and reviewed all manuscripts and

provided valuable inputs, edits, and her expertise in epidemiology and statistics.

In addition, Dr. Egeland supervised the socio‐cultural integrity of the work.

Dr. L. Jackson Roberts II, provided the technical training, expertise in free radical

and more specifically isoprostane biology and supervised GC‐MS analysis at his

laboratories. Dr. Roberts also reviewed all manuscripts and provided valuable

input and edits.

Dr. H.M. Chan provided input and edits on the manuscripts that he was listed as a

coauthor.

ix

Acknowledgments

I would like to thank my co‐supervisors, Dr. Stan Kubow and Dr.Grace Egeland.

I am grateful to Dr. Kubow for his support, mentorship, challenges, and hard

questions, that pushed me to grow and learn and become a critical thinker and

an independent researcher;

and to Dr. Egeland for introducing me to the Canadian Arctic research and for

her insights, good advice and for involving me in this unique research project.

I am ever grateful for Dr. Roberts who believed in me and generously supported

my work and gave me the honor to collaborate with a renowned scientist in the

field of free radical research.

My sincere gratitude for Mr. William Zackert at VU medical center and Ms. Donna

Leggee at McGill University’s CINE for their intense Laboratory training and their

help for all technical expertise.

My deepest gratitude goes out to my family for their endless support and

encouragement.

I am grateful to Dr. Katherine Gray‐Donald for introducing me to the world of

nutritional epidemiology; and who made this entire journey even more

rewarding with her wisdom and beautiful spirit.

I am ever grateful to have had the privilege to be surrounded by a pantheon of

nutritional experts in the School of Dietetics and Human Nutrition.

x

I would like to thank all of my fellow graduate students, especially Michele

Iskandar for her help in translating the thesis abstract to French; Sina Gallo, and

Mrs. Lise Grant, Helga Saudny‐Unterberger, the project manager for their advice,

encouragement and assistance; and all CINE and the department support staff.

I would like to acknowledge the coordinated efforts of all members of the IPY

study staff, coordinators, nurses, interviewers, coast guard crew, and specially

members of the community.

Lastly, I would like to acknowledge fund providers for this work: Government of

Canada Federal Program for International Polar Year, Canadian Institutes of

Health Research, Health Canada, Aboriginal Affairs and Northern Development

Canada, Government of Nunavut, University of Toronto, and Arctic Net.

xi

TABLE OF CONTENTS

ABSTRACT ii

RESUME iv

ADVANCE OF SCHOLARLY KNOWLEDGE vi

CONTRIBUTIONS OF AUTHORS viii

ACKNOWLEDGEMENTS ix

TABLE OF CONTENTS xi

LIST OF TABLES xvi

LIST OF ABBREVIATION xviii

CHAPTER 1: INTRODUCTION 1

1.1 Background 1

1.2 Rationale 3

1.3 Statement of purpose 7

1.4 Thesis objectives 8

CHAPTER 2: REVIEW OF THE LITERATURE 8

2.1 Health benefits on n‐3 fatty acids 8

2.2 Traditional diet and cardiovascular disease 10

2.3 N‐3 Fatty acids and cardiovascular risk factors 11

2.3.1 Glucose tolerance 11

2.3.2 Blood lipids 13

2.3.3 Blood pressure 14

xii

2.3.4 Obesity 14

2.4 Factors influencing n‐3 consumption 16

2.5 Modulators of cardiovascular risk 17

2.5.1 Metabolic syndrome 17

2.5.2 Genetic predisposition 19

2.6 Evidence of cardiovascular disease 20

2.7 C‐reactive protein 22

2.8 Oxidative Stress 24

2.8.1 F2‐isoprostanes 25

2.8.2 F3‐isoprostanes 27

2.8.3 Isofurans 29

2.,8.4 Isoprostanes Quantification 31

2.8.5 Selenium 33

2.8.6 Methylmercury 40

2.8.7 Persistent organic pollutants 41

2.8.8 Smoking status 44

BRIDGE 1 46

CHAPTER 3: MANUSCRIPT 1 47

3.1 ABSTRACT 48

3.2 INTRODUCTION 49

3.3 METHODS 50

3.3.1 Subjects recruitment 50

xiii

3.3.2 Anthropometric, measures and definitions 51

3.3.3 Laboratory methods 52

3.3.4 Plasma analysis of F2‐isoprostanes and isofurans 52

3.3.5 Statistical analysis 53

3.4 RESULTS 54

3.4.1 Subjects characteristics 54

3.4.2 Oxidative stress biomarkers 54

3.4.3 Isoprostanes correlates‐ CVD risk factors 55

3.4.4 Smoking 57

3.4.5 Inflammation 57

2.4.6 Predictors of isoprostanes 58

3.5 DISCUSSION 58

BRIDGE 2 69


4.1 ABSTRACT 71

4.2 INTRODUCTION 72

4.3 METHODS 75

4.31 Subjects recruitment 76

4.3.2 Anthropometric, measures, and definitions 76


4.3.4 Assessment of mercury and selenium exposure 76

xiv

4.3.5 Plasma analysis of F2‐isoprostanes and isofurans 77


4.4 RESULTS 78

4.4.1 Subjects characteristics 78

4.4.2 Isoprostanes, selenium and mercury 79

4.4.3 Comparison between selenium tertiles 80

4.4.4 Final variance predictors of plasma isoprostanes 80

4.5 DISCUSSION 81

BRIDGE 3 92


5.1 ABSTRACT 94

5.2 INTRODUCTION 95

5.3 METHODS 97

5.3.1 Subject recruitment 97

5.3.2 Anthropometric, measures, and definitions 97


5.3.4 Fatty acid analysis 98

5.3.5 Analysis of PCBs 99

5.3.6 Plasma analysis of isoprostanes 99


5.4 RESULTS 101

5.4.1 F3‐IsoPs relationship to F2‐IsoPs, selenium and mercury 102

xv

5.4.2 Isoprostanes and PCBs 102

5.4.3 Isoprostanes and relative erythrocyte fatty acids

concentrations

102

5.4.4 Comparison between n‐3 tertiles 103

5.4.5 Comparison between age categories 104

5.4.6 Final predictors of F3‐isoprostanes 104

5.5 DISCUSSION 104

CHAPTER 6: OVERALL SUMMARY AND CONCLUSIONS 120

6.1 Oxidative stress status and CVD risk 120

6.2 Sex differences in risk associated with oxidative stress 121

6.3 Oxidative stress in smokers 122

6.4 Oxidative stress and alcohol intake 123

6.5 Selenium protection against Hg‐induced oxidative 124

6.6 n‐3 PUFAs and isoprostanes 125

6.7 Strength and limitations 125

6.8 Future direction 127

6.9 Public health implications 127

6.10 Conclusion 128

7.00 REFERENCES 129

xvi

LIST OF TABLES

3.1. The anthropometrical and clinical characteristics for subjects with CRP<10 (n=233)

63

3.2 Pearson’s correlations between plasma biomarkers of oxidative stress and CV risk factors.

64

3.3 Geometric mean values and 95% CI of plasma biomarkers of oxidative stress for the total sample population and in comparison with categorical variables using student t‐test

65

3.4 Geometric mean values and 95% CI of plasma isoprostanes compared per smoking status and in relation to obesity.

66

3.5 Multiple comparisons of oxidative stress biomarkers levels between CRP categorical groups.

67

3.6 Multivariate associations showing the standardized regression coefficients (β) of plasma isoprostanes concentrations

68

4.1 Characteristics of study population (n=233) 87

4.2 Bivariate correlation between plasma IsoPs, Se and Hg 88

4.3 Characteristics of 223 subjects according to tertiles of blood Se concentrations

89

4.4 GLM adjusted Se‐Tertiles for plasma IsoPs concentrations 90

4.5 a‐d

Multivariate associations showing the regression coefficient (β) of plasma IsoPs concentrations

91

5.1 Plasma concentrations of isoprostanes1 and relative concentrations of fatty acids of study population

110

5.2 Plasma organochlorine concentrations (ug/L) and adjusted1 for total plasma lipids (ug/g lipids) of the study population

111

5.3 Unadjusted plasma isoprostanes, blood contaminants, Se and relative fatty acid concentrations according to tertiles of n‐3 PUFAs level

112

5.4 Bivariate correlations between plasma isoprostanes, contaminants, Se, and relative fatty acid concentrations

113

5.5 Partial correlation between plasma isoprostanes, contaminants, Se, and relative fatty acid concentrations adjusted for age, gender and waist circumference

114

xvii

5.6a Comparison of study characteristics according to age categories (<40 and ≥ 40 yrs)

115

5.6b Comparison of relative fatty acid concentrations according to age categories (<40 and ≥ 40 yrs)

116

5.6c Table 5.6‐c. Comparison of study characteristics according to age categories (<40 and ≥ 40 yrs)

117

5.7a Multivariate analysis showing the final correlates of F3‐IsoPs without toenail‐Se adjustment.

118

5.7b Multivariate analysis showing the final correlates of F2:F3 ratio without toenail‐Se adjustment.

118

5.7c Multivariate analysis showing the final correlates of F3‐IsoPs with toenail‐Se adjustment.

119

5.7d Multivariate analysis showing the final correlates of F2:F3 ratio with toenail‐Se adjustment.

119

xviii

LIST OF ABBREVIATIONS

ACE, angiotensin converting enzyme

ADD1, Alpha‐adducin 1

AGT, angiotensinogen

ASP, Alaskan Siberian Project

BAL, bronchoalveolar lavage

BF, body fat

BP, blood pressure

CAC, coronary artery calcium

CHD, coronary heart disease

CHF, congestive heart failure

CINE, Centre for Indigenous Peoples’ Nutrition and Environment

COX, cyclooxygenase

CoQ10, co‐enzyme Q10

CRISPS‐2, Cardiovascular Risk Factor Prevalence Study‐2

CRP, C‐reactive protein

CSF, cerebrospinal fluid

CV, cardiovascular

CVD, cardiovascular disease

DBP, diastolic blood pressure

EBC, exhaled breathe condensate

EDTA, ethylenediaminetetraacetic acid

F2‐IsoPs, F2‐isoprostanes

F3‐IsoPs, F3‐isoprostanes

FG, fasting glucose

GC‐NICI‐MS, gas chromatography‐negative‐ion chemical ionization mass

spectrometry

GNB3, guanine nucleotide‐binding protein G(I)/G(S)/G(T) subunit beta3

GOCADAN, genetics of coronart artery disease in Alaskan natives

GPx, glutathione peroxidase

xix

GSH, glutathione – reduced form

GSSG, glutathione – oxidized form

HB, hemoglobin

HDL‐C, high‐density lipoproteins cholesterol

Hg, mercury

hs‐CRP, high sensitivity C‐reactive protein

HTN, hypertension

ICAM, intercellular adhesion molecule

IDF, International Diabetes Federation

IGT, impaired glucose tolerance

IHD, ischemic heart disease

IMT, intima media thickness

IPY, International Polar Year

IsoFs, isofurans

ISR, Inuvialuit Settlement region

KHAS, Keewatin health assessment survey

LDL‐C, low‐density lipoprotein cholesterol

MetS, metabolic syndrome

MI, myocardial infarction

MUFA , monounsaturated fatty acids

NCEP, National Cholesterol Education Program In The Adult Treatment Panel III

NF‐κB, nuclear factor kappa‐light‐chain‐enhancer of activated B cells

NO, nitric oxide

Nrf2, nuclear factor (erythroid‐derived 2)

OGTT, oral glucose tolerance test

PCBs, polychlorinated biphenyls

PCDDs, polychlorinated dibenzo p‐dioxins

PCDFs, polychlorinated dibenzofurans

PFOS, perfluoroctanesufonate

xx

PG, prostaglandin

PLD, phospholipase D

POPs, persistent organic pollutants

PUFAs, polyunsaturated fatty acids

RR, relative risk

SBP, systolic blood pressure

SeP, selenoprotein

Se, selenium

SFAs, saturated fatty acids

SLHDP, Sandy Lake Health and Diabetes Project

SPE, solid‐phase extraction

T2DM, type 2 diabetes mellitus

T‐Chol, total cholesterol

TFAs, trans fatty acids

TG, triglycerides

TNF‐α, tumor necrosis factor–alpha

VCAM, vascular cell adhesion molecule

WC, waist circumference

WHO, World Health Organization

WHR, waist to hip ratio

1

CHAPTER 1: INTRODUCTION

1.1 Background

The Indigenous Peoples of Canada have gone through a major nutrition

transition in the last decade. A shift has been observed from their traditional

diet composed of sea mammals, land animals, fish, shellfish and plants, with

the majority of dietary energy coming from sea mammals to a more

westernized market food‐based diet (Kuhnlein, Receveur et al. 2001). The

market foods are comprised of the limited food choices available in their

local market, mainly highly processed food ingredients and canned non‐

perishable food items. Nutrient‐density comparisons of the traditional and

market portions of the diet revealed that traditional food had greater density

of proteins, iron, zinc, copper, magnesium, selenium (Se), and vitamin A

(Kuhnlein and Receveur 1996), monounsaturated and polyunsaturated fatty

acids (PUFA) of marine origin, the latter of which have been promoted for

their cardio‐protective properties. The shift to a Western type diet is

attributable in part to the greater availability of market foods in the

communities (Blanchet, Dewailly et al. 2000).

Environmental contaminants is another driving force for this dietary change

as heavy metals and organochlorines are found in many animal and plant

species in the region; consequently restricting their use as part of traditional

food by the locals (Kuhnlein and Receveur 1996). This shift away from the

traditional diet created a change in the Inuit diet composition marked by

increased saturated fat, sucrose, and alcohol consumption and a reduction in

the percent contribution of n‐3 PUFA‐rich marine foods to the diet (Kuhnlein

and Receveur 1996; Kuhnlein, Receveur et al. 2008). This dietary change has

compromised the nutritional value of the Inuit diet leading to micronutrient

deficiencies that are linked directly to common morbidities such as tooth

loss, anemia, and some types of cancer (Kuhnlein and Receveur 1996).

2

Moreover, in several indigenous populations, a shift away from traditional

hunting and gathering driven lifestyles and traditional diet is associated with

an increased prevalence of risk factors for cardiovascular diseases (CVD),

such as high blood pressure, elevated blood lipids, diabetes and obesity

(Jorgensen, Borch‐Johnsen et al. 2006; Jorgensen, Glumer et al. 2003;

Kuhnlein, Receveur et al. 2004). Inuit obesity prevalence rates are found to

be similar to those developed countries of Europe and North America (Young

2007). Consequently, increased rates of CVD incidences and diabetes

prevalence are reported throughout aboriginal communities (Young 2007).

This is a worrisome change from earlier studies, which found the Inuit to be

relatively free of ischemic heart disease (IHD) and lower age‐adjusted

cardiovascular disease and diabetes prevalence than the general population

(Pollex, Khan et al. 2004).

Currently, coronary heart disease (CHD) seems to be increasing in Alaskan

Inuit who were originally thought to be relatively immune to CHD because of

their n‐3 PUFAs consumption (Ebbesson, Risica et al. 2005). There also have

been increases in the rates of death from IHD and stroke among all

indigenous populations that are higher compared with whites (Bjerregaard,

Young et al. 2003). This finding indicates that despite several‐fold higher

intakes of n‐3 polyunsaturated fatty acids (n‐3 PUFAs) and Se in the Inuit

population that are related to improved antioxidant protection, they are not

necessarily protected from disease risk relative to the non‐indigenous

population. This latter evidence may be partly related to diminishing intake

of traditional n‐3 PUFA and Se rich foods, particularly in the younger age

groups. These latter nutrients may provide significant CVD benefits through

improvement of risk factors such as blood lipids and blood pressure

mediated partly via inhibitory effects on oxidative stress and inflammation.

3

The increased CVD incidence in the Inuit population might also be related to

lifestyle and environmental factors that may counteract some of the benefits

of the n‐3 PUFA and Se‐rich Inuit diet. For example, smoking, a rise in CVD

risk factors such as obesity, and exposure to environmental contaminants

that accumulate with age can all lead to increased oxidative stress and

inflammation.

1.2 Rationale

The divergence away from the traditional diet high in n‐3 PUFAs acids that is

accompanied by urbanization and westernization is introducing a

challenging situation for the Inuit to adapt on many levels. Because dramatic

changes in lifestyle and diet is occurring in the Inuit population that is in

rapid transition, systemic re‐evaluation of determinants of health and disease

is important in order to identify trends (time‐interval changes) that could

predict future increases in diabetes and cardiovascular disease.

Studies in Canadian and Alaskan Inuit (Young, Gerrard et al. 1999; Dewailly,

Blanchet et al. 2001; Ebbesson, Tejero et al. 2007) found that established

CVD risk factors such as blood lipoproteins, obesity, diabetes, hypertension,

and glucose tolerance are related to n‐3 PUFAs tissue concentrations.

Evidently, these CVD risk factors may be modulated by n‐3 PUFAs

consumption that increases with age. On the other hand, smoking and

exposure to environmental contaminants from the traditional food has

decreased among the young Inuit (Dewailly, Blanchet et al. 2003; Nobmann,

Ponce et al. 2005). Obesity in general and specifically abdominal obesity has

increased in the current wave of nutrition transition among the Inuit

(Kuhnlein, Receveur et al. 2004; Young, Bjerregaard et al. 2007). However,

there are some challenges in evaluating the metabolic risk profile of the Inuit

due to the lack of understanding of Inuit anthropometry and how it reflects

4

metabolic predisposition to disease (Young 2007). For example, when the

Inuit are compared to Euro‐Canadians, at each level of BMI or waist

circumference, the Inuit had lower blood pressure and lipid levels (Young

2007). In addition, Inuit studied in 1989‐1991, when compared to two

independent Caucasian samples studied at the same time, had lower

triglycerides and higher HDL‐cholesterol concentrations despite a trend for

an increased prevalence of higher waist circumference (WC) (Young 2007).

Such findings in Inuit have been consistently justified by their high intake of

n‐3 PUFAs from the traditional diet, which is speculated to be the modulator

of the relationship between increased obesity and health outcomes

(Nobmann, Ebbesson et al. 1999; Eilat‐Adar, Mete et al. 2009). Indeed,

Belanger et al. (2006) found high erythrocyte concentrations of n‐3 PUFAs in

adults of Nunavik that were more than three‐fold higher than previously

reported in white subjects. These high levels may provide anti‐oxidative

protection evident by an over‐expression of antioxidant enzymes and redox

status, which may provide protection against the high levels of

environmental contaminants found in this population such as methylmercury

(MeHg) and polychlorinated biphenyls (PCBs) (Belanger, Dewailly et al.

2006).

The antioxidative protection that the Inuit seem to possess could also be

partly due to high level of selenium (Se) found in their diet (Blanchet,

Dewailly et al. 2000), which is reflected in the two‐fold higher blood levels

when compared to Caucasians (Belanger, Dewailly et al. 2006). The degree of

protection of the Inuit diet could, however, be dependent upon the amount of

exposure to xenobiotic contaminants and on sufficient intake of specific

protective dietary elements in the Inuit diet. Sensitive and specific indices of

whole body oxidative stress that have been closely related to disease

occurrence such as isoprostanes (Montuschi, Barnes et al. 2004) are needed

5

to clarify better the relation between the Inuit diet, oxidative stress, and

specific disease risk factors; including environmental contaminants that can

exert deleterious health by additively inducing oxidative stress. In addition to

diet‐related factors, alcohol and smoking habits have also been speculated to

modulate differently both obesity development (Ordovas 2007) and disease

risk in each gender and age group.

The mechanistic role of n‐3 PUFAs in CVD development in the Inuit also

requires further study, including the possibility of antioxidant protection

provided by n‐3 PUFAs intake despite their highly unsaturated fatty acid

composition. It is unclear how the n3‐PUFAs mediate protection against

metabolic dysregulations associated with obesity. In addition, detailed study

of obesity determinants and how these factors convey CVD risk in the Inuit is

relatively unexplored in the Inuit population. For example, it is not known

whether abdominal obesity represented by the high WC in predominantly

intra‐abdominal (metabolically active) or subcutaneous tissue. It is also not

known how the body composition (e.g., percent body fat; %BF) of the Inuit

corresponds to various body mass index (BMI) or WC measurements. Direct

measurement of whole body oxidative stress status is lacking in the Inuit, as

previous work in Nunavik used proxy measures such as glutathione related

antioxidant enzymes, plasma concentrations and redox states of α‐

tocopherol and coenzyme Q10 (CoQ10) to assess indirectly oxidative stress

via these antioxidant status indicators. Also, it is not clear what the

determinants of oxidative stress in the Inuit are and how oxidative stress

relates to CVD risk factors such as obesity, hyperlipidemia and inflammation.

Transitional changes in diet and lifestyle among the Inuit are driving an

increased prevalence of diabetes and CVD, which requires more detailed

study, particularly in relation to oxidative stress. Emerging disease risk

6

factors include increased intakes of trans and saturated fats (Counil, Dewailly

et al. 2008), alcohol intake and tobacco smoke, which are dietary and lifestyle

factors that have been closely associated with increased oxidative stress.

Although certain predictors of oxidative stress have been well characterized

such as obesity, age, smoking, alcoholism and diet, the impact of Inuit

characteristics (such as increased abdominal obesity and advantageous

blood lipid profiles, such as low ratio of total cholesterol (T‐Chol) to high

density lipoprotein cholesterol (HDL‐C)) on whole body oxidative stress has

been sparsely investigated. Furthermore, Inuit metabolic adaptations as

affected by their diet, lifestyle and environment have never been fully

explored. It is possible that a decline in the traditional food intake in Inuit

populations may cause deleterious shifts in tissue fatty acid composition in

terms of decreased n‐3 PUFAs content (Bersamin, Luick et al. 2008) that may

promote increased CVD risk.

In that regard, this thesis work involved assessment of novel isoprostane

biomarkers (F3‐isoprostanes) that are n‐3 PUFAs oxidation products that

could be partly responsible for the cardioprotective effects associated with n‐

3 PUFAs rich marine food intake by the Inuit as these products may protect

against the pro‐inflammatory action of F2‐isoprostanes (Gao, Wang et al.

2007). Detailed study of the ratio of tissue F2‐isoprostanes to F3‐isoprostanes

could provide important insights regarding the cardioprotection associated

with the traditional n‐3 PUFAs rich Inuit diet, and would enable us to

investigate the association of these biomarkers to the dietary transition that

could predict future increases in diabetes and CVD.

Although previous literature has focused on the cardiovascular protective

effects of n‐3 PUFAs among the Inuit, relatively little attention has been given

to the possible cardioprotective effects of Se, particularly in relation to

7

protection against environmental contaminant health risks. Xenobiotic

contaminants can alone or synergistically modulate CVD risk factors such as

hypertension, insulin resistance, and dyslipidemia (Houston 2011), which

can involve deleterious changes in oxidative stress status in the Inuit and

their predisposition to CVD. Also, the thesis looked for a possible mechanistic

relationship in the Inuit regarding smoking with respect to the isoprostane

species, isofurans, as smoking is related with decreased tissue oxygen.

Decreased tissue oxygen tension has been associated with altered tissue

levels of isofurans as indicated in animal trials (Fessel and Jackson Roberts

2005).

1.3 Statement of Purpose

The overall thesis objective was to perform a comprehensive assessment of

whole body oxidative stress status among the Inuit using the gold standard

biomarker F2‐isoprostanes and the isoprostane isoforms F3‐isoprostanes,

and isofurans and subsequently identify correlates of these oxidative stress

factors in a subsample of the International Polar Year Inuit Health Survey

(2007‐2008) population. We hypothesized that F2‐isoprostanes are

associated with cardiometabolic risk factors and so may be used to evaluate

deleterious health risk associated with shift away from the Inuit traditional

diet beyond traditional CV risk factors. We propose that simultaneous

assessment of both F2‐isoprostanes and isofurans provides a better

understanding of oxidative stress than either analyte alone and that F3‐

isoprostanes may be related to n‐3 PUFA‐mediated protection against

oxidative stress.

8

1.4 Thesis Objectives:

1. To quantify plasma levels of F2‐isoprostanes, F3‐isoprostanes, and

isofurans in a random sample of the survey population.

2. To assess relationship between plasma isoprostanes and

cardiometabolic risk factors.

3. To evaluate the relationship between the Se status and plasma

isoprostanes.

4. To assess the interrelationship between dietary methylmercury

(MeHg) exposure and tissue Se on oxidative stress status.

5. To examine the impact of decreased consumption of traditional foods

on oxidative stress parameters.

CHAPTER 2: REVIEW OF THE LITERATURE

2.1 Health Benefits of n‐3 PUFAs

The original hypothesis that the Inuit are protected by their n‐3 PUFAs rich

diet from CVD was based on an observation made by Dyerberg and Bang in

1970’s who noted that there was an absence of coronary atherosclerosis

among Greenlandic Inuit living in their own environment and consuming

large amounts of marine mammals and fish. The investigators proposed that

this observed CHD protection was due to abundance of n‐3 PUFAs from

seafood and sea mammals in their diet (Dyerberg 1989). A large body of

observational data from epidemiological studies has demonstrated that the

consumption of fish is cardioprotecive (Daviglus, Stamler et al. 1997; Gillum,

Mussolino et al. 2000; Visioli, Rise et al. 2003; He, Song et al. 2004;

Mozaffarian 2008), which has been attributed to the long chain n‐3 PUFAs

found in fish, most notably eicosapentaenoic acid (EPA; 20:5n‐3), and

docosahexaenoic acid (DHA; 20:6n‐3) (Rupp, Wagner et al. 2004;

Mozaffarian, Ascherio et al. 2005; DeFilippis and Sperling 2006; Hooper,

9

Thompson et al. 2006). Large‐scale clinical trials using fish oil supplements

have suggested that the primary mechanisms through which n‐3 PUFAs

prevent cardiovascular disease are due to reductions in ventricular

fibrillation, arrhythmias (Wongcharoen and Chattipakorn 2005), and a

decrease in incidence of myocardial infarction (MI) (Siscovick, Raghunathan

et al. 1995; Albert, Hennekens et al. 1998; Marchioli, Schweiger et al. 2001).

Many studies suggest important anti‐atherogenic and anti‐thrombotic effects

mediated via improved lipoprotein metabolism (Harris, Lu et al. 1997), blood

pressure (Bao, Mori et al. 1998), endothelial function (Woodman, Mori et al.

2002), vascular reactivity (Mori, Watts et al. 2000), inflammation (Mori,

Woodman et al. 2003), platelet (Knapp HR 1997), fibrinolytic function

(Dunstan, Mori et al. 1999), cytokine production (Thies, Miles et al. 2001),

coagulation (Mori, Beilin et al. 1997) and oxidative stress (Barden, Mori et al.

2004). In earlier studies, the favorable effects of fish oils were often

attributed to EPA owing to the lack of sufficient data studying the individual

properties of both EPA and DHA (Mori and Woodman 2006). Newer set of

controlled trials using purified EPA and DHA oils demonstrated that both

EPA and DHA have important haemodynamic and anti‐atherogenic

properties (Mori and Woodman 2006); however, the effects of DHA may

differ from that EPA for some mechanisms of action. Both of these latter long

chain n‐3 PUFAs, however, are equally effective in reducing serum

triglycerides (Nestel, Shige et al. 2002; Mori and Woodman 2006) and

attenuating oxidative stress and cytokine production following cell

stimulation (Mori, Woodman et al. 2003). In contrast, high doses of both EPA

and DHA in subjects with diabetes may lead to impairment of glucose

tolerance by influencing hepatic glucose output (Woodman, Mori et al. 2002).

In addition, neither EPA nor DHA affects total cholesterol concentrations

(Grimsgaard, Bonaa et al. 1997). Data from human studies suggest that DHA

may be more favorable in lowering blood pressure and improving vascular

10

function, raising HDL‐C and attenuating platelet function (Mori, Beilin et al.

1997; Woodman, Mori et al. 2003). Other studies, however, have failed to

replicate such preferential properties of DHA versus EPA (Arterburn, Hall et

al. 2006). Contrary studies regarding independent effects of described for

EPA and DHA may be partly due to studies using a variety of high‐risk

individuals such as overweight, hyperlipidemic, or hypertensive subjects,

who have established altered metabolic profiles that may differentially affect

lipid metabolism. Variability in subject selection, dose differences, and

duration of interventions could also explain the observed inconsistencies in

these associations. Most of the above studies were performed in populations

with very low usual intake of fish in Western diet context (Mori and

Woodman 2006). Not all studies, however, have shown the beneficial effects

of fish intake on death resulting from CHD or on the risk of coronary disease,

and it may be that fish consumption is beneficial only in high‐risk

populations (Arterburn, Hall et al. 2006).

2.2 Traditional Inuit Diet and CVD

Studies regarding the CVD benefits of n‐3 PUFA rich diets in Inuit populations

have been contradictory in terms of markers of atherosclerosis. Data from

the Alaskan Siberian Project (ASP) involving a cross sectional study in 454

Inupiat Alaskan Eskimos showed no relation between n‐3 PUFAs

consumption and clinically verified CHD (Ebbesson, Risica et al. 2005). In

another study by the same ASP investigators involving 686 persons from the

same population, no differences in n‐3 PUFAs consumption was found among

those with and without atherosclerotic plaques. Also, there was no

relationship detected between n‐3 PUFAs intake and extent of plaque as

assessed by the plaque score representing the number of carotid arterial

segments with discrete plaque (Ebbesson, Roman et al. 2008). Significant

positive associations were seen, however, between intakes of saturated FAs

11

(SFAs), particularly palmitate, and to a lesser extent stearate, with carotid

plaque prevalence and extent. In addition, negative associations were found

between the consumption of EPA, DHA, palmitic acid, and stearic acid and the

mean intima media thickness (IMT) of the distal common carotid arteries in

the arterial segments that were unaffected by atherosclerotic plaque

(Ebbesson, Roman et al. 2008). Fatty acid intake, however, was measured in

grams per day but not as a percentage of kilocalories and could thereby be

reflective of subjects eating more of all foods including those higher in the

studied FAs. Nevertheless, these findings collectively suggest that high

intakes of n‐3 PUFAs in this Alaskan Inuit population does not protect against

plaque formation but is associated with decreased IMT. Importantly, the

above findings also signify that high consumption of SFAs, especially palmitic

acid, is an important risk factor for plaque formation, even in the presence of

large amounts of n‐3 PUFAs. A simultaneous increase in SFAs intake in

concert with n‐3 PUFAs consumption is less unlikely to occur from the

traditional Inuit diet, which is exemplified by observations as Inuit

consumers of more marketed foods had a significantly lower PUFAs:SFAs

ratio than consumers of diet rich in Inuit traditional foods (Bersamin, Luick

et al. 2008). Indeed, Kuhnlein et al. (Kuhnlein, Receveur et al. 2004)

compiling Inuit, Dane, and Metis data, found that on days without traditional

foods, higher percentage of SFAs, carbohydrate, fat and sucrose was

consumed as compared to days without traditional foods amongst Canadian

Arctic indigenous people (Kuhnlein, Receveur et al. 2004).

2.3. N‐3 PUFAs and Risk Factors of CVD

2.3.1 Glucose Intolerance

Early observations from a questionnaire study of Alaskan Eskimos showed

an inverse correlation between fish and seal consumption and abnormal

glucose tolerance and type 2 diabetes mellitus (T2DM); however, no

12

biomarker of FAs intake was measured in this study (Alder, Boyko et al.

1994). Both infrequent salmon and seal oil intake and high intake of non‐

indigenous food consumption among Alaskan Eskimos has been associated

with glucose intolerance (Alder, Boyko et al. 1994). ASP investigators

confirmed such observations showing an inverse association between plasma

concentrations of fasting and 2‐hrs insulin and plasma concentrations of EPA

and DHA (Ebbesson, Risica et al. 2005). The authors suggested that this latter

association could partly explain earlier observations that Eskimos have

significantly lower plasma concentrations of insulin compared to American

Indians and other population groups because of their greater insulin

sensitivity. In support of these associations, the results of the 4‐year ASP

intervention study where subjects were given nutrition education sessions

revealed that those with improved glucose tolerance had significantly higher

plasma concentrations of n‐3 PUFAs (Ebbesson, Ebbesson et al. 2005).

Further, ASP study showed positive associations between high plasma levels

of palmitic acid and impaired glucose tolerance (IGT) and T2DM (Ebbesson,

Adler et al. 2005). In Nunavik similar to ASP data (Dewailly, Blanchet et al.

2001), inverse associations were found between plasma insulin and an

increase in plasma EPA and the ratio of EPA to AA. However, in contrast to

ASP, studies in Nunavik showed that plasma levels of n‐3 PUFAs were

positively associated with plasma glucose (Dewailly, Blanchet et al. 2001).

The difference in direction of association with regards to n‐3 PUFAs with

plasma glucose might be reflective of a more disturbed metabolic profile

amongst Nunavik Inuit who were marked by higher levels of glucose

intolerance as compared to Alaskan Inuit. One limitation for confirmation of

such interpretation is the lack of an index reflective of insulin sensitivity from

Nunavik data to ascertain this discrepancy. It is noteworthy at the same level

of BMI, when Alaskan Inuit are compared to Nunavik; they have relatively

13

higher abdominal obesity and fasting glucose but have lower fasting insulin

and triglycerides in both men and women (Young, Bjerregaard et al. 2007).

2.3.2. Blood Lipids

Studies to date regarding the Inuit have shown that their higher n‐3 PUFAs

intake is consistently associated with improved lipid profile marked with low

triglycerides and high HDL‐C, which are considered cardioprotective assets

(Dewailly, Blanchet et al. 2003; Ebbesson, Adler et al. 2005). Moreover, blood

triglyceride concentrations and the tota cholesterol (T‐Chol): HDL‐C ratio did

not worsen with age amongst older Inuit as normally expected with aging,

which has been implicated as a positive health consequence associated with

higher traditional food consumption (Bersamin, Luick et al. 2008). On the

other hand, n‐3 PUFAs intake in the Inuit does not seem to have a direct

lowering effect on other blood lipid components such as low density

lipoprotein‐cholesterol (LDL‐C) and T‐Chol as seen in several studies

(Dewailly, Blanchet et al. 2003). For example, in Nunavik, both erythrocyte

EPA and DHA were positively associated with total and LDL‐C (Dewailly et al,

2001). However, when the ratio of EPA to AA was used, a negative

association was found with T‐Chol:HDL‐C, which suggests an anti‐

atherosclerotic effect. The paradoxical increase of LDL‐C was suggested to

reflect an increase in LDL particle size, which is less atherogenic (Dewailly,

Blanchet et al. 2001; Woodman, Mori et al. 2003). This latter suggestion was

based on earlier speculation that n‐3 PUFAs may change the composition of

LDL‐C, leading to less atherogenic LDL particles with lower phospholipid and

apolipoprotein B concentrations and a larger LDL particle size

(Ander, Dupasquier et al. 2003). Further confirmation with more direct

functional disease risk parameters such as IMT is needed to clarify these

lipoprotein relationships to disease risk.

14

2.3.3. Blood Pressure

With regards to arterial blood pressure, Inuit blood pressure ranks

intermediate among the global scale, but is low when compared with most

white populations (Bjerregaard, Young et al. 2004). Direct relationship

between n‐3 PUFAs intake and blood pressure is not consistent. Among

Alaskan Inuit, low blood pressure was found among highest n‐3 PUFAs

consumers and among those with highest quartiles of plasma n‐3 PUFAs

concentrations (Ebbesson, Adler et al. 2005). Diastolic pressure (DBP) rather

than systolic blood pressure (SBP) was found to be inversely associated with

n‐3 PUFAs consumption. The specificity of a relationship of n‐3 PUFAs to DBP

was also found in the ASP intervention study whereby significant lowering of

DBP but not SBP was observed among those that improved their glucose

tolerance without weight loss (Ebbesson, Ebbesson et al. 2005). In Nunavik,

investigators found no associations of n‐3 PUFAs with blood pressure

(Dewailly, Blanchet et al. 2001), which might be due to the low prevalence of

hypertension amongst Inuit from Nunavik (6%, 1992), versus the Alaskan

Eskimos (34%, 1994). Indeed, Moris et al reported that the hypotensive

effects of high doses of fish oils might be strongest in hypertensive subjects

and in those with clinical atherosclerotic disease or hypercholesterolemia

(Bao, Mori et al. 1998). Most studies that targeted healthy individuals with no

clinical manifestation of hypertension failed to detect hypotensive effects of

n‐3 PUFAs on blood pressure (Bao, Mori et al. 1998; Grimsgaard, Bonaa et al.

1998).

2.3.4. Obesity

Population studies have suggested that greater intakes of unsaturated fat

increase the resistance to obesity (Calder, Ahluwalia et al. 2011). In support

of such notion, lower plasma n‐3 PUFAs and lower consumption of n‐3 PUFAs

are associated with greater weight amongst Alaskan Inuit (Bersamin, Luick et

15

al. 2008). Moreover, in a preliminary report, plasma FAs profiles of

overweight Eskimos were indicated to be significantly different from those

with normal weight after adjustment for age, gender and IGT (Ebbesson,

Adler et al. 2005); however, the data was not published to compare the

compositional differences. Interestingly, abdominal obesity showed a

different trend with levels of n‐3 PUFAs consumption in Nunavik as Inuit

with higher waist circumference had higher concentrations of n‐3 PUFAs

compared to those with normal waist (Dewailly, Blanchet et al. 2003).

It has been well established that a cluster of metabolic abnormalities

including insulin resistance and dyslipidemia characterizes obese subjects

with abdominal obesity. It is not clear, however, how abdominal obesity

contributes to the development of the respective CVD risk factors in the Inuit,

particularly with respect to the relative relationship of abdominal obesity

versus general obesity as driving factors to the disturbed metabolic

abnormalities associated with CVD. In that regard, there is suggestion that

abdominal obesity may act as a buffer or a compensatory mechanism for the

added burden of increased insulin resistance.

Obese Inuit have higher values for CVD risk factors than non‐obese Inuit

(Dewailly, Blanchet et al. 2001; Young, Bjerregaard et al. 2007); however,

compared to obese Quebecers, obese Inuit have higher plasma

concentrations of n‐3 PUFAs and HDL‐C and lower concentrations of insulin

and triglycerides as well as a lower ratio of T‐Chol to HDL‐C (Dewailly,

Blanchet et al. 2001). Furthermore, Nunavik Inuit were found to have the

lowest risk status for age adjusted CVD as compared to Cree and Quebecers

despite having the highest prevalence of cigarette smoking and obesity

(Dewailly, Blanchet et al. 2003). When compared to Euro‐Canadians, at each

level of the BMI or waist circumference, Inuit had lower levels of most CVD

16

risk factors (Jorgensen, Glumer et al. 2003; Young 2007). Inuit men with WC

between 95‐100 cm, considered at high risk by the International Diabetes

Federation's criteria for the metabolic syndrome, had a mean triglyceride

levels of 1.02 mmol/L (95% CI=0.90‐1.15), which is equivalent to the mean

level of Euro Canadians with a WC of 75‐80 cm (mean=1.08 mmol/L; 95%

CI=0.94‐1.21) (Young, Bjerregaard et al. 2007). Thus, the metabolic impact of

different levels of obesity appears to be markedly less among the Inuit,

especially for plasma lipid indicators such as HDL‐C and triglycerides (Young,

Bjerregaard et al. 2007). Taken together, the above evidence appears to

suggest that n‐3 PUFAs might attenuate metabolic deterioration associated

with CVD in obese Inuit.

Interestingly, westernization has been shown to be associated with

decreased BMI and abdominal obesity amongst immigrant Greenlandic Inuit

women with a concomitant decreased HDL‐C and increased blood

triglyceride concentrations (Bjerregaard, Jorgensen et al. 2002; Bjerregaard,

Jorgensen et al. 2004). Indeed, Inuit women living in their native lands had

higher mean values for WC, waist to hip ratio (WHR), fasting insulin and

lower mean values for blood pressure, triglycerides and 2‐hrs insulin, when

compared to immigrant Inuit in Denmark, who consumed less seal and fish

and drank more alcohol than their kin at all levels of obesity (Jorgensen,

Borch‐Johnsen et al. 2006). Adjustment for other lifestyle factors such as

physical activity, socioeconomic status and level of education did not

attenuate the differences (Bjerregaard, Jorgensen et al. 2007).

2.4 Factors Influencing n‐3 PUFAs Consumption

Tissue concentrations of EPA, DHA and total n‐3 PUFAs among the Inuit

differ by regions. The n‐3 PUFAs concentrations of Nunavik were similar

overall to those observed among Alaskan river village Eskimos, but were

17

lower than those reported for Igloolik Inuit in Nunavut and Alaskan Coastal

village Eskimos (Dewailly, Blanchet et al. 2001). Differences between Arctic

regions may be attributed to the different laboratory methods used and also

to the territorial availability of fish species as populations in coastal regions

consume more marine mammals and fish than the land populations (Van

Oostdam, Donaldson et al. 2005). Furthermore, traditional food intake may

vary according to the degree of urbanization of Inuit communities, which can

include change in population density and administration services that, in

turn, could affect the degree of westernization of lifestyle and diet. A common

observation in northern indigenous populations has been that older Inuit

have had higher plasma concentrations of n‐3 PUFAs than younger Inuit

reflecting their higher intakes (Kuhnlein, Receveur et al. 2004; Nobmann,

Ponce et al. 2005). The FAs levels and ratios may also vary according to sex in

some regions as women in Nunavik were found to have higher n‐3 PUFA

values than Inuit men (Dewailly, Blanchet et al. 2001) although gender

differences are not always consistent. In Nunavik, plasma n‐3 PUFAs did not

vary according to smoking status, but alcohol abstainers had higher plasma

concentrations of EPA and DHA than did subjects who had ≥1 drink

(Dewailly, Blanchet et al. 2001). In addition, higher concentrations of n‐3

PUFAs and their ratios were found in Inuit who used medications of

hypertension, or had either hypercholesterolemia or diabetes than nonusers.

2.5 Modulators of CV Risk

2.5.1. Metabolic Syndrome (MetS)

The cluster of cardiovascular risk factors including abdominal obesity,

glucose intolerance, hypertension and dyslipidemia known as the metabolic

syndrome (MetS) has been investigated as a metabolic link towards the

development of CVD. MetS is associated with CHD, strokes, and CVD

mortality more than the individual components of the syndrome (Liese,

18

Mayer‐Davis et al. 1998). Moreover, MetS has been found to be a significant

predictor of incident of T2DM (Lorenzo, Okoloise et al. 2003). There is no

internationally agreed definition of the MetS, and estimates of the MetS vary

across populations because many different criteria have been used. The

World Health Organization (WHO) proposed a definition in 1998 and the

National Cholesterol Education Program Expert Panel (NCEP) published a

working definition in 2001. Data from the Sandy Lake Health and Diabetes

Project (SLHDP) which represented Oji Cree Indians showed moderate

agreement between the two definitions with a kappa value of 0.63 [95% CI

0.56‐0.70] (Liu, Hanley et al. 2006). Similar moderate agreement between the

two definitions was demonstrated in Greenland with a kappa value of 0.56

[95% CI 0.51‐0.61] (Jorgensen, Bjerregaard et al. 2004). The two definitions

yielded similar prevalence estimates for the population studied, although

with considerable disagreement in classification of risk factors as seen in

other ethnicities (Ford, Mokdad et al. 2003). The NCEP definition in both

studies seemed to overestimate some risk factors because of lower cutoffs for

triglycerides and blood pressure in both men and women (Ford, Mokdad et

al. 2003). On the other hand, using WHO definition abdominal obesity was

overestimated in men and HDL‐C was overestimated in women (Jorgensen,

Bjerregaard et al. 2004). Another shortcoming of the NCEP definition is the

reliance on fasting glucose to assess glycemic status, which may overlook a

large proportion of subjects with IGT and T2DM who are solely diagnosed by

2‐h glucose values.

Generally, a lower prevalence of the MetS has been found among Inuit when

compared to whites. Data from Keewatin Health Assessment Study (KHAS)

showed that the crude rate of the NCEP‐defined MetS was 3.5%, which was

lower than other native and non‐aboriginal populations in Canada (Liu,

Hanley et al. 2006). There was a gender difference in the prevalence of the

19

NCEP‐defined MetS as rates were found higher in women than in men with

the crude rates of MetS (18.8% vs. 6.7%) and age‐adjusted rates (22.0% vs.

8.2%), respectively. This latter observation could partly be explained by the

higher propensity of abdominal obesity in women than in men. When

compared to the Oji‐Cree and non‐aboriginal Canadian, Inuit had the lower

rates of hypertriglycemia, low HDL‐C, high and fasting glucose and they

ranked as intermediate in terms of abdominal obesity whereas native Cree

ranked higher than Inuit (Liu, Hanley et al. 2006). The rate of high blood

pressure in Inuit was shown to be similar to non‐aboriginal Canadians as

previously been demonstrated in Greenlandic Inuit (Bjerregaard, Young et al.

2003). All components of the MetS in Inuit increased with increased obesity.

Using logistic regression, the odds ratio for MetS was found to be 3.4 with

each 5 kg/m2 increase in BMI; however, the Inuit had more overall obesity

and abdominal obesity than non‐aboriginal subjects, yet had a much lower

rate of the MetS (Liu, Hanley et al. 2006). Only in Inuit were decreased odds

of the MetS achieved after adjustment of sex, age and BMI.

2.5.2. Genetic Predisposition

Evidence suggests that the Inuit have genetic differences vary their

susceptibilities to the CV risk factors (Lalouel, Rohrwasser et al. 2001). In this

regard, a genetic variant related to blood pressure have been identified such

as those encoding angiotensin converting enzyme (ACE), adducing (ADD1),

the G protein beta subunit (GNB3), and angiotensinogen (AGT) (Lalouel,

Rohrwasser et al. 2001). Canadian Inuit have one of the highest frequencies

of the AGT T235 allele of any population of the world; however, the allele

itself was not associated with elevated blood pressure or hypertension at the

population level (Hegele, Tully et al. 1997; Hegele, Young et al. 1997). Other

factors such as age, male gender and obesity were associated with high SBP

and DBP in the Inuit (Bjerregaard, Dewailly et al. 2003). Genetic research in

20

the frequencies of putative "deleterious alleles" from 13 candidate genes in

atherosclerosis and/or diabetes among three Canadian populations showed

that there were significant differences in the frequencies of five of the 13

alleles between Oji‐Cree and Inuit. Compared with the Oji‐Cree, the Inuit has

significantly lower AGT M174 and MTHFR 677T and higher frequency of HL‐

480C, ApoE E4, and FABP2 T54. When compared to whites, Inuit have excess

of deleterious alleles (Hegele 1999). Nevertheless, the differences in MetS

components prevalence between Inuit and whites cannot be explained by

differences in genetic factors alone, as susceptibility is particularly

modulated by environmental factors such as diet and physical activity.

2.6. Evidence of Cardiovascular Disease (CVD)

Cardiovascular disease (CVD) refers to any disease that affects the

cardiovascular system, principally cardiac disease, vascular diseases of the

brain and kidney, and peripheral arterial disease. The causes of CVD are

diverse but atherosclerosis and/or hypertension are the most common.

Atherosclerosis is an intricate, multifactorial vascular disease associated with

narrowing of the carotid, coronary, and femoral arteries by the formation of

stable and unstable plaques, whose formation depends on the grade of lipid

accumulation and inflammation. CVD is classified as a chronic inflammatory

disease associated with several common critical risk factors such as diabetes,

hypertension, obesity, dyslipidemias and smoking. These different pathologic

conditions eventually lead to IHD with clinical syndromes evidenced by

considerably reduced blood flow to the myocardium.

Primary mechanism(s) causing carotid atherosclerotic plaque to develop into

symptomatic disease are still uncertain; however, data suggests that the

mediators of inflammation and oxidative stress are not only the leading cause

of formation of plaque but also may be involved in rapid progression of

21

atheromatous lesions, plaque rupture, and intra‐luminal thrombosis

(DeGraba 1997). Several risk factors of atherosclerosis such as diabetes,

obesity, smoking, and thickening of the intima‐media of the carotid artery,

are associated with increased low‐grade inflammation, as evidenced by

moderate but significant increases of isoprostanes together with cytokines

(Interleukin‐6; IL‐6) and acute phase proteins such as high sensitivity C‐

reactive protein (hs‐CRP) in body fluids (Basu and Helmersson 2005;

Helmersson, Arnlov et al. 2005; Sinaiko, Steinberger et al. 2005; Wohlin,

Helmersson et al. 2007).

Atherosclerosis assessment by carotid IMT ultrasound examination is a

validated measure for the assessment of sub‐clinical atherosclerosis. Earlier

epidemiological studies showed that increases in the IMT are positively

associated with current (Burke, Evans et al. 1995) and future development of

CVD (O'Leary and Polak 2002). More recently in the Carotid Atherosclerosis

Progression study, carotid IMT independently predicted CVD events such as

strokes and MI in both young and old subjects covering the age range of 19 to

90 [hazard rate ratios per 1 SD common carotid artery (CCA) IMT increase

were 1.43 (95%CI: 1.35‐1.51) for MI, 1.47 (1.35‐1.60) for stroke, and 1.45

(1.38‐1.52) for MI, stroke, or death] (Lorenz, von Kegler et al. 2006).

Furthermore, data from the Osaka Follow‐up Study for Carotid

Atherosclerosis 2 showed that the predictive ability of carotid IMT for

vascular events was also independent in high‐risk patients whom risk factors

are managed clinically (taking medications). The relative risk (RR) of a CVD

event increased with increased IMT, even after adjustment of risk factors and

history of CVD [middle tertile RR, 2.5, 95% CI: 1.0‐6.3, to highest tertile RR,

3.6, 95% CI: 1.4‐ 9.0] (Kitagawa, Hougaku et al. 2007). Such findings imply

that medication intake and history of CVD do not hinder the predictive value

of IMT. Using multiple linear regressions, intakes of EPA and DHA were

22

significantly and inversely associated with IMT after the adjustments for age,

sex, and total energy intakes in Japanese subjects over 40 yrs old (Hino,

Adachi et al. 2004). In Nunavik, main determinants of atherosclerosis

assessed by IMT were age, sex, BMI, hypertension, and diabetes. There was

no association of any of the lipoproteins measured including T‐Chol, LDL‐C,

and HDL‐C (Noel, Dewailly et al. 2012).

2.7 C‐reactive protein (CRP)

CRP, an acute phase protein (APP), is present in various body fluids of normal

individuals. Any clinical disease characterized by tissue injury and/or

inflammation is accompanied by a significant elevation of serum CRP with

the concomitant stimulation of other APPs (Baumann and Gauldie 1994; Steel

and Whitehead 1994). The pro‐inflammatory cytokines interleukin‐1 (IL‐1),

IL‐6, and tumor necrosis factor‐α (TNF‐α) released at the site of tissue injury

initiate the acute phase reaction (APR) cascade: cytokines that activate and

are activated by nuclear factor (NF‐κB) that regulates APP gene‐expression in

the liver, which is their principal target organ (Libermann and Baltimore

1990; Barnes and Karin 1997). CRP mRNA transcription is induced

dramatically by IL‐6 in cooperation with IL‐1 (Ganter, Arcone et al. 1989).

Activators of the proinflammatory process facilitate disease and including

stress response, free radicals, oxidative stress, bacterial, and virus infections

(Elenkov, Iezzoni et al. 2005). Measurement of CRP is the most practical way

to assess the presence of an inflammatory state. CRP levels tend to be higher

than normal in patients with the metabolic syndrome (Grundy, Cleeman et al.

2004). An elevated CRP (> 3 mg/L) is an emerging risk factor for CVD (NCEP

Expert Panel on Detection and Treatment of High Blood Cholesterol, 2002).

Circulating hs‐CRP is recognized as one of the strongest independent

predictors of vascular death in a number of settings (Ridker, Stampfer et al.

23

2001; Rifai and Ridker 2001). CRP appears to be a stronger predictor than

LDL‐C, and it adds prognostic value to conventional Framingham risk

assessment (Blake and Ridker 2002). The link between CRP and

atherosclerosis is via the direct effect of CRP to promote atherosclerotic

processes and endothelial cell inflammation (Calabro, Willerson et al. 2003;

Verma, Wang et al. 2003; Wang, Li et al. 2003). CRP has been associated with

an increased risk of incident MI and stroke (Blake and Ridker 2002). Many

lines of evidence confirm that inflammation plays a role in the development

of hypertension, which suggests that inflammation as a key player in the

development of CVD, and is not a merely a biomarker. Cross‐sectional

evidence demonstrates higher CRP levels among those individuals with

elevated blood pressure (Giles, Croft et al. 2000; Bermudez and Ridker 2002;

Ford, Giles et al. 2002). In two large prospective cohort studies: Women’s

Health Study (Woodward, Rumley et al. 2003) and Hong Kong Cardiovascular

Risk Factor Prevalence Study‐2 (CRISPS‐2) (Cheung, Ong et al.), CRP levels

were shown to be associated with future development of hypertension in

both sexes, and plasma CRP was independently associated with the

development of hypertension in CRISPS‐2 (odds ratio per quartile=1.26, p <

0.010) (Cheung, Ong et al. 2011), which provides good evidence for critical

role of CRP in the development of hypertension.

Higher levels of CRP may increase BP by reducing nitric oxide (NO)

production in endothelial cells (Venugopal, Devaraj et al. 2002) resulting in

vasoconstriction and increased production of endothelin (Venugopal, Devaraj

et al. 2002). C‐reactive protein may also function as a proatherosclerotic

factor by upregulating angiotensin type 1 receptor expression (Wang, Li et al.

2003). Inflammation has been shown to correlate with endothelial

dysfunction (Yudkin, Stehouwer et al. 1999) and relate to the renin‐

angiotensin system (Brasier, Recinos et al. 2002).

24

2.8. Oxidative Stress

Oxidative stress status caused by the over production free radical species is a

well‐established indicator to assess CVD risk especially in the

epidemiological context (Block, Dietrich et al. 2002). Using the Framingham

Offspring Cohort to assess CVD risk, Keaney and al. (Keaney, Larson et al.

2003) found that oxidative stress was independently associated with

smoking, diabetes and obesity. Statistically significant correlations have been

found between lipoprotein susceptibility to peroxidation, the degree of

obesity, and the risk of developing CVD (Ghanim, Garg et al. 2001). In cross

sectional studies, subjects with established CVD had higher plasma

concentrations of lipid peroxidation products than controls (Stringer, Gorog

et al. 1989), and in prospective studies antibodies to oxidized LDL preceded

the later development of CVD (Salonen, Yla‐Herttuala et al. 1992).

Furthermore, established CVD risk factors such as hypertriglyceridemia,

hyperglycemia, and insulin resistance are closely linked to systematic

oxidative stress in the metabolic syndrome context (Hansel, Giral et al. 2004).

Thus, oxidative stress assessment may provide a tool to recognize at risk

populations such as the Inuit with an adverse metabolic profile for the

development of CVD. Additionally, oxidative stress status may help in the

identification of asymptomatic subjects at risk of atherosclerotic disease. As

was shown in otherwise healthy nonsmokers population, after the

adjustments of traditional factors and hs‐CRP, assessment of intracellular

oxidative stress by measuring glutathione redox state ‐ measured as the ratio

of reduced glutathione (GSH) to oxidized glutathione (GSSG), GSH/GSSG ‐

was an independent predictor of the presence of early atherosclerosis

measured by IMT (Ashfaq, Abramson et al. 2006), thereby indicating an early

role for oxidative stress in the pathogenesis of premature atherosclerosis.

Further, oxidative stress is hypothesized to be a mechanism by which

mercury (Hg) induces atherosclerosis (Lund, Miller et al. 1991; Lund, Miller

25

et al. 1993). Consumption of Hg‐contaminated fish was directly related to

induction of lipid peroxidation (Salonen, Seppanen et al. 2000) and to

interference with mitochondrial electron transport chain and depletion of

cellular GSH (Rissanen, Voutilainen et al. 2000).

2.8.1. F2‐Isoprostanes

Isoprostanes (IsoPs), also known as isoprostaglandins, are prostaglandin‐like

compounds that are formed non‐enzymatically from the free radical‐induced

oxidation of arachidonic acid (AA), an important PUFA of the (n‐6) series. AA

can be converted enzymatically through the cyclooxygenase (COX) pathway

to bioactive prostaglandins (F2a, E2, I2, among others) and thromboxanes.

Unlike COX‐derived prostaglandins, IsoPs are formed in situ esterified to

phospholipids and are subsequently released by the action of phospholipase

A2 and also platelet activating factor (PAF) acetylhydrolyses (Morrow, Awad

et al. 1992; Stafforini, Sheller et al. 2006). Thus iso‐prostaglandin F2a (F2‐

isoprostanes, F2‐IsoP), an isomer of prostaglandin F2a, is released from its

esterified form to free form into peripheral circulation, as opposed to

prostaglandins, which are generated directly from AA. The de‐esterification

phase is suggested to be one of the rate limiting steps for the release of free

IsoPs in the tissues and their additional availability in the circulation (Basu

2008). Metabolism of IsoPs has not been extensively studied; however, it has

been shown to occur essentially through oxidation‐reduction pathways as

enzymatically formed primary prostaglandins dependent on 15‐

prostaglandin dehydrogenase (15‐PGDH) and delta‐13‐reductase for their

degradation (Basu 2008). These and other hydrolytic, reductive, oxidative

enzymes are found ubiquitously in all tissues. Thus, basal F2‐IsoPs formation

in any tissue as well as its increased generation after induction of oxidant

stress may be followed by rapid degradation thereby preventing further

release into the circulation unless a deficit in de novo metabolizing enzyme

26

systems is present (Basu 2008). In a metabolic study, Roberts and colleague

administered labeled 8‐IsoPGF2a for over 1 hr into a male subject, and they

showed that 75% of the infused compound was excreted into the urine

during the following 4.5 hrs (Roberts, Moore et al. 1996). The half‐life of 8‐

IsoPGF2a has been found to be approximately 16 min in humans.

Elevation of F2‐IsoPs in human fluids and tissues has been found in obesity,

diabetes, smoking, atherosclerosis and many other disease states (Milne, Yin

et al. 2008). In the Framingham Heart Study using age‐ and sex‐adjusted

models, increased urinary creatinine‐indexed 8‐isoPGF2 levels were

positively associated with female sex, hypertension treatment, smoking,

diabetes, blood glucose, body mass index, and a history of cardiovascular

disease (Keaney, Larson et al. 2003). F2‐IsoPs have emerged as the gold

standard biomarker of whole body lipid peroxidation and oxidative stress in

both experimental and clinical studies (Kadiiska, Gladen et al. 2005). F2‐IsoPs

have been used extensively to quantify lipid peroxidation in association with

risk factors for atherosclerosis and other diseases (Milne and Morrow 2006).

Plasma F2‐IsoPs concentrations were found to be an independent predictor

for the presence of CAC and are differentiated between those with no CAC

and those with the generally low Agatston scores (calcium measure to

quantify calcification in any given artery used to calculate CAC) found in the

Cornonary Artery Risk Development in Young Adults (CARDIA) population

(Gross, Steffes et al. 2005). The association of F2‐IsoPs with CAC was

independent of the primary lipoproteins involved in atherosclerosis and the

CRP concentration, a marker of ongoing inflammatory processes, as well as

many other cardiovascular disease risk factors. Taken together, the above

observations indicate that plasma F2‐IsoPs may be an independent indicator

for the risk of CAC in generally healthy populations. In addition, using logistic

regression analysis revealed that urinary F2‐IsoP levels were an independent

27

predictor for both IMT and angiographic coronary artery disease in high

coronary artery disease risk patients (Basarici, Altekin et al. 2007). In

subjects with the MetS, urinary F2‐IsoPs concentration was positively related

to dietary intake (i.e., percent of energy) of total fat and various fats

(irrespective of degree of saturation); however, it was inversely related to

serum phospholipids levels of n‐3 PUFAs (Sjogren, Basu et al. 2005).

Supplementation with n‐3 PUFAs was shown to decrease plasma

concentration of F2‐IsoPs in healthy subjects (Nalsen, Vessby et al. 2006),

postmenopausal women (Higdon, Liu et al. 2000) and urine F2‐IsoPs

concentrations of treated hypertensive and diabetic subjects (Mori, Watts et

al. 2000).

2.8.2. F3‐Isoprostanes

F3‐isoprostanes (F3‐IsoPs) are formed from the oxidation of EPA in vivo. They

are newly isolated and structurally identified compounds that were

discovered in 2006 (Gao, Yin et al. 2006) and have been speculated to

modulate the beneficial biological effects of EPA and fish oil supplementation.

It has been proposed that the anti‐atherogenic and anti‐inflammatory

mechanisms of n‐3 PUFAs are partly linked to their interference with the AA

cascade that generates pro‐inflammatory eicosanoids (Kris‐Etherton, Harris

et al. 2002) and IsoPs (Davis, Gao et al. 2006). When F3‐IsoPs formation was

tested in vivo in mice, supplementation with EPA markedly increased its

quantities in heart tissue, and significantly decreased the levels of the pro‐

inflammatory F2‐IsoPs by up to 64% (Davis, Gao et al. 2006; Gao, Wang et al.

2007). The putative protective role of F3‐IsoPs in relation to human disease

risk has not been tested. EPA could not only replace AA in the phospholipid

bilayers of cell membranes but can also act as a competitive inhibitor of COX,

reducing the production of 2‐series prostaglandins and thromboxane, in

addition to reducing the 4‐series leukotrienes. The 3‐ and 5‐ series

28

eicosanoids that are derived from EPA are either less biologically active or

inactive as compared with the AA‐derived products and thus considered to

exert less inflammatory effects (Thies, Miles et al. 2001; Yang, Chan et al.

2004). There also evidence that a group of polyoxygenated DHA and EPA

derivatives termed resolvins that are produced in various tissues inhibit

cytokine expression and other inflammatory responses in microglia, skin

cells, and other cell types (Hong, Gronert et al. 2003; Serhan, Jain et al. 2003;

Serhan and Levy 2003). Further, oxidized EPA in the presence of Cu2+, but not

native EPA, significantly inhibited human neutrophil and monocyte adhesion

to endothelial cells (Sethi, Eastman et al. 1996; Sethi 2002). This effect was

induced via inhibition of endothelial adhesion receptor expression and was

modulated by the activation of the peroxisome proliferators activated

receptor α by EPA oxidation products. In addition, oxidized EPA markedly

reduced leukocyte rolling and adhesion to venular endothelium of

lipopolysaccharide treated mice in vivo, and the effect was not observed in

peroxisome proliferators activated receptor alpha deficient mice (Sethi

2002). With regards to the anti‐atherogenic properties, various aldehyde

oxidation products of EPA and DHA decrease the expression of the CD36

receptor in human macrophages, and up‐regulation of that receptor has been

linked to atherosclerosis (Vallve, Uliaque et al. 2002). However, none of these

latter reports identified the specific EPA and DHA peroxidation products

responsible for these reported effects.

Animal studies have shown that EPA and DHA supplementation reduced

urinary F2‐IsoP levels, as well as enhanced cellular antioxidant defense

systems (Sarsilmaz, Songur et al. 2003; Iraz, Erdogan et al. 2005). Although a

reduction of F2‐IsoPs levels by n‐3 PUFA can be attributed, in part, to a

decrease in membrane AA content (Calviello, Palozza et al. 1999; Davis, Gao

et al. 2006; Gao, Wang et al. 2007), the F2‐IsoPs reduction can also mediated

29

via the non‐enzymatic free‐radical peroxidation of n‐3 PUFAs to generate

reactive isoprostane species analogous to those formed from AA (Gao, Yin et

al. 2006). Also, animal studies have shown that J‐ring compounds generated

from in vivo oxidation of EPA and DHA can reach cellular concentrations

sufficiently high to induce Nrf2 [or NF‐E2‐related factor 2, Nuclear factor

(erythroid‐derived 2)‐like 2]‐based antioxidant and Phase II detoxification

defense systems (Gao, Wang et al. 2007). Nrf2 is a master transcription factor

shown to regulate expression of more than 200 genes, including those

involved in Phase II detoxification and antioxidant gene expression (Kwak,

Kensler et al. 2003; Kwak, Wakabayashi et al. 2003).

28.3. IsoFurans

Isofurans (IsoF) are novel products of free radical‐induced peroxidation of

AA that contain a substituted tetrahydrofuran ring. IsoF formation is

increasingly favored with increasing oxygen tension in vitro and in vivo

(Fessel and Jackson Roberts 2005). By contrast, F2‐IsoP formation is

increasingly disfavored with increasing oxygen tension (Fessel and Jackson

Roberts 2005). Quantification of IsoF with IsoPs would provide a more

complete picture of the extent of oxidant injury across tissue oxygen

concentrations than measurement of either analyte alone. Thus,

measurement of IsoFs provides unique information that complements

measurement of F2‐IsoPs as an index of lipid peroxidation reflective of tissue

oxygenation. Formation of IsoFs increases in a linear fashion from 1% to

100% O2, whereas the formation of IsoPs plateaus at 21% O2. These

disparate effects of ambient oxygen concentration on the formation of IsoFs

and IsoPs can be explained by the fact that oxygen concentration is the

critical determinant for the rates of mutually exclusive competing reactions

of a common intermediate involved in the mechanisms of IsoP and IsoF

formation. Using CCl4‐treated rats, an established animal model of oxidant

30

injury to the liver, Fessel et al. (Fessel, Porter et al. 2002) showed that levels

of esterified IsoFs and F2‐IsoPs in liver phospholipids increase markedly after

administration of CCl4 (Fessel, Porter et al. 2002). The hepatic levels of IsoFs

from both control and CCl4‐treated animals were much lower than levels of

F2‐IsoPs; however, the ratio of IsoFs to IsoPs in various tissues was shown to

vary in accordance with the level of tissue oxygenation in vivo. In that regard,

the ratio of IsoFs to IsoPs in two highly oxygenated organs, the kidney and

brain hippocampus showed that IsoFs exceeded IsoPs by ≈ 2.0‐ to 2.3‐fold,

which was the inverse of the observed hepatic ratios (Fessel, Porter et al.

2002). IsoFs are stable compounds that are present at readily detectable

levels in normal tissues and biological fluids. The levels of IsoFs detected in

normal biological fluids were as follows: rat urine 3.3 ±0.3 ng/ml (n=8), rat

plasma 334 ± 80 pg/ml (n= 8), mouse bronchoalveolar lavage fluid (210 ± 30

pg/ml, n=2), human plasma (71±10 pg/ml, n=2), and human urine (5.8 ±1.0

ng/ml, n= 5)(Fessel, Porter et al. 2002).

Potential utility for measuring IsoFs can include a wide variety of clinical

settings as diverse as oxidative damage to transplant organs during storage

in ambient air, retinopathy of prematurity, and the sequelae of hyperbaric

oxygen exposure (Fessel, Hulette et al. 2003). Additionally, mitochondrial

dysfunction can be a source of free radical generation and is a feature of a

number of disorders including neurodegenerative diseases such as

Parkinson’s disease (Fiskum, Starkov et al. 2003). Mitochondrial dysfunction

theoretically could also lead to increased cellular O2 concentration due to

impaired mitochondrial O2 utilization.

Interestingly, in that regard, levels of F2‐IsoPs in the substantia nigra from

patients with Parkinson’s disease were no different from levels in age‐

matched controls, whereas IsoF levels were significantly increased (Fessel,

31

Hulette et al. 2003). The potential utility of measuring IsoFs as an index of

oxidative stress has never been explored on a population level.

2.8.4 Isoprostanes Quantification

Several methods have been developed to quantify IsoPs in tissues and in

biological fluids such as blood and urine. Gas chromatographic (GC)/mass

spectromic (MS) approach using stable isotope dilution techniques has

several advantages over other approaches includes its high sensitivity and

specificity, which yields quantitative results in the low picogram range

(Morrow 2005). Its drawbacks are that it is labor intensive and requires

considerable expenditures on equipment. Alternative approaches have been

developed to quantify IsoPs using immunologic techniques (Fam and Morrow

2003). A potential drawback of these methods is that limited information is

currently available regarding their precision and accuracy (Morrow JD 2005).

In addition, little data exist comparing IsoPs levels determined by

immunoassay to mass spectrometry.

F2‐IsoPs are stable and robust molecules that are detectable in all human

tissues and biological fluids analyzed, including plasma, urine,

bronchoalveolar lavage (BAL) fluid, cerebrospinal fluid (CSF), and bile.

Several methods have been developed to analyze F2‐IsoPs; these include

chromatographic separation involving solid‐phase extraction (SPE) or

affinity chromatography with or without thin‐layer chromatography (TLC)

followed by final determination by gas chromatography–mass spectrometry

(GC‐MS), liquid chromatography (LC)–MS, or enzyme immunoassay (Mori,

Croft et al. 1999; Morrow and Roberts 1999; Yoshida, Hayakawa et al. 2007;

Liu, Morrow et al. 2009)(Yan, Byrd et al. 2007; Lee, Huang et al. 2008). GC–

negative‐ion chemical ionization mass spectrometric (GC/NICI‐MS)

employing stable isotope dilution is the preferred method for the

32

quantification of F2‐IsoPs and several alternative GC‐MS assays have been

developed by different investigators including FitzGerald and colleagues

(Pratico, Barry et al. 1998). For quantification purposes, Dr. Robert’s lab

measure 15‐F2t‐IsoP (also known as 8‐Iso PGF2α), and other F2‐IsoPs that

coelute with this compound. The advantages of GC‐MS include the high

resolution of GC separation on fused silica capillary columns and the

specificity and sensitivity of MS, which yield quantitative results in the low

picogram range (Morrow 2005; Nourooz‐Zadeh 2008).

Quantification of F2‐IsoPs by MS has distinct advantages compared to

analysis by immunoassay methodologies such as ELISA. Although ELISA

measurement offers high‐throughput analysis and does not require costly

instrumentation, the polyclonal antibodies used to bind F2‐IsoPs exhibit

cross‐reactivity with many other molecules similar in structure, including

COX‐derived PGF2α (Il'yasova, Morrow et al. 2004). This cross‐reactivity

results in the quantification of inflated concentrations of F2‐IsoPs.

Additionally, biological impurities can interfere with antibody binding. MS

offers high sensitivity and specificity yielding quantitative results in the

picogram per milliliter range (Milne, Gao et al. 2012).

The GC/MS protocol used in this study is a robust and specific methodology

for the quantification of F2‐IsoPs and IsoFs (and F3‐IsoPs) and has been

utilized for this purpose for more than 20 years. Importantly, this method

offers the lowest limit of quantitation of any reported mass spectrometric

methodology for F2‐IsoPs. This is particularly important in the quantification

of these molecules in plasma, CSF, EBC, and other biological fluids in which

low levels of F2‐IsoPs are found. Further, according to our knowledge, this is

the only reported methodology for the quantification of IsoFs. However, the

labor‐intensive purification and derivatization steps limit the throughput of

33

the assay as experienced personnel can analyze a maximum of only 20

samples in one day.

Due to the softer mode of ionization utilized, LC/MS methodologies for the

measurement of F2‐IsoPs offer a more rapid alternative to GC/MS as only

extraction from the biological matrix is required; the extensive chemical

derivatization steps necessary for GC/MS can be eliminated. For reference, a

summary of several MS‐based assays reported in the recent literature is

presented in the paper of Milne, Gao et al.(2012). The existence of these

multiple methodologies, however, is a major challenge to the field as the

specific F2‐IsoP isomers quantified differ from assay to assay and,

consequently, results cannot be directly compared between laboratories.

Further, several new techniques are being published each year. Within the

past year alone, at least five new methodologies have been reported in the

literature and each paper reports a unique way to extract and analyze

samples depending on the sample matrix and available equipment with no

validation or comparative studies.

2.8.5. Selenium (Se)

Selenium (Se) is an essential trace element that functions through its

association with selenoproteins (SePs) in the form of selenocysteine.

Through SePs (glutathione peroxidses (GPx), and thioredoxin reductase), Se

functions as a defense mechanism for oxidative stress, for the regulation of

thyroid hormone activity, and for the redox status to maintain vitamin C and

other molecules (Boosalis 2008). The role of Se as an antioxidant has been

well established, and a lowered Se status has been associated with an

increased risk of CVD and congestive heart failure (de Lorgeril, Salen et al.

2001; Witte, Clark et al. 2001). Se may have an impact on the course and

outcome of a number of etiologically inflammatory diseases and conditions.

34

In vitro studies demonstrate that there is strong indication that viral,

bacterial, or stress‐induced inflammation may be variably influenced by Se

availability (Rayman 2012).

In vivo studies in rats demonstrated that dietary Se influenced the

development of both genetic and age‐related hypertension and increased Se

intake was clearly associated with an increase in selenoantioxidant enzymes

activity and a decrease in cardiac oxidative injury (Lymbury, Marino et al.

2010). Observational studies demonstrated compromised Se status in

hypertensive patients (Nawrot, Staessen et al. 2007). In one meta‐analysis

that included 25 observational studies (14 cohort and 11 case‐control

studies) Flores‐Mato et al. (Flores‐Mateo, Navas‐Acien et al. 2006) observed a

50% increase in plasma Se concentrations was associated with a 24%

reduced risk of coronary events. Decreased serum Se levels have been

observed in acute and chronic inflammatory states with high CRP values

(Maehira, Luyo et al. 2002). High concentrations of serum Se predicted

reduced levels of oxidative stress (measured by F2‐IsoPs) and subclinical

COX‐mediated (but not cytokine‐mediated) inflammation in Swedish men in

a follow‐up study of 27 years (Helmersson, Arnlov et al. 2005). Men with the

highest quartiles of serum Se at baseline had decreased levels of urinary F2‐

IsoPs compared to all lower quartiles at follow‐up. These associations were

independent of BMI, diabetes, hyperlipidemia, hypertension, smoking, α‐

tocopherol and β‐carotene at baseline.

The interactive associations between Se, oxidative stress and inflammation

might be related to the proposed cardiovascular protective property of Se.

Both in vivo and in vitro studies suggest that Se may have cardioprotective

properties through alleviation of insulin resistance (Stapleton 2000). In

addition, Se is implicated in CVD on cardiac muscle integrity not only through

35

its direct role on the protection of endothelial cells against free radical

accumulation, but also in the biosynthesis of AA derivatives involved in

platelet and leukocyte functions (Cao, Reddy et al. 2000), and in the

regulation of cholesterol metabolism (Traulsen, Steinbrenner et al. 2004).

Moreover, Se prevents toxic effects of cadmium and Hg through its property

to form less toxic selenides. For example, MeHg can be converted to less toxic

inorganic Hg by the action of reduced selenite (selenide)(Masukawa, Kito et

al. 1982), or by diverting the binding of these cations to less critical proteins

protecting SeP systems against oxidative stress and free radicals, which are

the main pathways of toxicity (Lindh, Danersund et al. 1996). The role of Se

in reducing bioavailability or toxicity of MeHg is not adequately studied in

humans. Se has also been reported to affect homocysteine metabolism, with

increased levels of serum‐free reduced homocysteine levels being reported

in Se deficient rats (Uthus, Yokoi et al. 2002). In addition, Se modulates the

active transport of calcium in the heart muscle via affecting glutathione

peroxidase activity responsible for decrease of Ca(2+)‐ATPase and Ca2+

uptake activities in sarcoplasmic reticulum in Se‐deficient rats (Wang, Jia et

al. 1993). Animal studies suggest that Se may reduce CVD risk, and provide

putative insights into possible mechanisms. However, unlike animal models,

where experimental deficiencies of single nutrients can be produced, Se

deficiency in man is only one factor in a complex set of nutritional and other

variables, which may predispose or protect against disease (Alissa, Bahijri et

al. 2003). Some clinical investigations have underlined Se importance in the

cardiac function and the prevention of coronary atherosclerosis (Flores‐

Mateo, Navas‐Acien et al. 2006), and several recent prospective

epidemiological studies have attributed to Se deficiency a greater incidence

of CVD (Salonen, Alfthan et al. 1982; Virtamo, Valkeila et al. 1985; Suadicani,

Hein et al. 1992). Collectively, however, observational studies have provided

inconsistent results, and randomized controlled trials of Se supplements

36

have not demonstrated significant benefits, perhaps due to differences in

study design or sources and types of Se compounds used (Park and

Mozaffarian 2010).

Epidemiological studies investigated the consequences of perturbed Se status

suggest the possibility of a threshold effect, with no clear relationship with

coronary heart disease in a population with a high dietary intake, and a

modest association in populations with low to moderate Se intake (Rayman

2012). The association between Se and CVD has been studied by analyzing

serum and toenail samples. Levels of Se in toenails have been found to reflect

longer‐term intake than serum Se (Slotnick and Nriagu 2006). Increased Se is

also paradoxically associated with increased blood lipids and possible

toxicities, and individuals of adequate or high status could be affected

adversely (Rayman 2012). Significant increased risk of T2DM is a concern

with Se blood levels exceeding 121.6 µg/L (Stranges, Marshall et al. 2007).

High Se levels affect insulin signaling via overexpression of GPx1 and

consequently eliminating important hydrogen peroxides that act as second

messengers (McClung, Roneker et al. 2004) and thereby adversely affecting

insulin sensitivity. Alternatively, polymorphism of the various SeP including

GPx1 cannot be ruled out as a confounding factor that affects the health

benefits and risks of dietary Se (Schoenmakers, Agostini et al. 2011).

Se is a nutrient found in high amount of traditional Inuit diet (Blanchet,

Dewailly et al. 2000; Dewailly, Blanchet et al. 2001). Due to the absence of

accurate food composition data, a biomarker is required to assess Se status.

Inuit have blood Se levels that is two‐fold higher when compared to levels

reported in Caucasians (Belanger, Dewailly et al. 2006). Blood Se was found

to directly relate to consumption of traditional foods as it correlated

positively with erythrocyte membrane n‐3 PUFAs as well as other

37

environmental contaminants such as MeHg and polychlorinated biphenyls

(PCBs) (Belanger, Dewailly et al. 2006). The Inuit mean GPx activity was

reported to be much higher than level of GPx reported to be protective of

cardiovascular events (Inuit mean 77.5 ±1.2 µg/d hemoglobin; HB ranging

from 53.4‐111.4 µg/d HB, compared to >56.3 respectively. In addition, levels

of GPx and Se did not correlate in the Inuit (Belanger, Dewailly et al. 2006),

which is indicative of enzyme activity reaching maximal level. Indeed, a

threshold effect for Se has been described (Kok, De Bruijn et al. 1987). GPx is

an adaptive enzyme increasing in response to oxidative stress, aging, physical

activity, and iron‐deficiency anemia (Rayman 2009). Moreover, differences in

bioavailability of Se sources or heavy metals, which interact with Se, may

influence GPx activity; thus, there might be a critical threshold value that may

differ within persons and populations (Kok, De Bruijn et al. 1987; Battin and

Brumaghim 2009). Based on 2004 Nunavik Survey data, Se status provide

antioxidant protection as it was related to decreased homocysteine and

increased CoQ10 blood levels (Belanger, Dewailly et al. 2006; Belanger,

Mirault et al. 2008). Despite robust blood antioxidant defenses demonstrated

in Canadian Inuit, however, there is indication of oxidative stress that could

not be identified evidenced by an unusually elevated ubiquinone‐10 to total

CoQ10 ratio (Belanger, Mirault et al. 2008).

Cellular mechanisms for Se‐mediated cardiovascular protection as reviewed

in Duntas (2009) seem to be orchestrated via a variety of complex

mechanisms. Cardiovascular protection associated with Se has stemmed

from observations that Se regulates GPx activity, which decreases

intracellular ROS. Overexpressed GPx can decrease ROS and inflammation

partly by inhibiting IκB‐α phosphorylation and consequently the

translocation of NF‐ κB. GPx can double IκB‐α half‐life and so preserves its

degradation (Kretz‐Remy and Arrigo 2001). Therefore, increased Se levels

38

can impede the transactivation of genes that encode inflammatory cytokines,

thus inhibiting APR release (Maehira, Miyagi et al. 2003). Furthermore, the

role of Se as an anti‐inflammatory factor is linked to its effect on immune

cells, especially the macrophage signal transduction pathways. Se

supplementation was shown to result in a significant decrease in the bacteria

endotoxin lipopolysaccharide (LPS)‐induced expression of the main pro‐

inflammatory genes TNF‐α and COX‐2 by inhibiting the Mitogen‐activated

protein kinase pathways (Zamamiri‐Davis, Lu et al. 2002). Increased TNF‐α

may induce maximum activation of NF‐κB with suppressed tissue Se levels

while also increasing the secretion of CRP by hepatocytes. TNF‐α is a

powerful inducer of adhesion molecules such as intercellular adhesion

molecule‐1 (ICAM‐1), vascular cell adhesion molecule‐1 (VCAM‐1), and

endothelial leukocyte adhesion molecule‐1 (E‐selectin), which are required

to promote endothelial cell proinflammation by recruiting leukocytes across

the endothelium (Vunta, Belda et al. 2008). In addition to attenuated

inflammation via the inhibition NF‐κB via GPx mediated by elevated tissue Se,

an important anti‐inflammatory mechanism of Se is mediated by its role in

modulating monocyte adhesion to endothelial cells and migration. Monocytes

adhere to endothelium and differentiate into macrophages, which are the

main effectors of innate immunity in inflammation (Cao, Cohen et al. 2001).

The monocyte adhesion to the endothelial cells is modulated by L‐selectin, a

member of the selectin family, which facilitates neutrophil migration during

inflammatory response mediated by various ligands. L‐Selectin expression

can be markedly down regulated by metalloproteinases, which by cleaving its

receptor generates a soluble L‐selectin that may inhibit the adhesion of

lymphocytes to endothelial cells (Wang, Fuster et al. 2005).

Se was found to induce shedding of L‐selectin from monocytes, leading to

reduced differentiation into macrophages, while L‐selectin was considerably

increased. Alternatively, enhanced dietary Se status may modify lymphocyte

39

proliferation and immune response by altering the metabolism of AA and the

formation of eicosanoids (Cao, Reddy et al. 2000), via the modulation of

phospholipase D (PLD), which plays a crucial role in the signal transduction

in various cell types (Cao, Reddy et al. 2000). Lymphocytes from Se deficient

rats produce significantly lower prostaglandins than Se supplemented rats

that led to decreased activation of PLD, lower generation of phosphatidic acid

and diacylglycerol, and consequently lower activation of protein kinase C

(PKC) (Yamamoto, Endo et al. 1995). The addition of prostaglandins can

reverse these results and enhance PLD activity. Additionally, increased Se

may facilitate an adaptive response for redox regulation and cell protection

against proinflammatory gene expression.

In vitro a time‐dependent increase in 15‐deoxy‐Δ 12,14 ‐prostaglandin J2

(15d‐PGJ2) production, whose formation is mainly mediated by

cyclooxygenase‐1 (COX‐1) via Se supplemented macrophages stimulated by

LPS, has been described (Vunta, Belda et al. 2008). High doses of Se, however,

may impair other types of immunity such as antiparasitic or allergic asthma

responses, indicating that the levels of Se may differently affect various types

of immune response (Comstock, Alberg et al. 2008). Chronic inflammation is

influenced by genetic and environmental factors. Functional analysis of

selenoprotein S (SePS) polymorphism, 105G A, significantly impairs SePS

expression, which is followed by increased plasma levels of the cytokines IL‐

6, IL‐1 β and TNF‐α (Curran, Jowett et al. 2005). These latter results provide

good evidence of a link between SePS and cytokine production. On the other

hand, even more recent data has failed to document any role of SePS

polymorphisms in the susceptibility to develop immune‐mediated diseases

(Martinez, Santiago et al. 2008).

40

2.8.6 Methyl Mercury (MeHg)

MeHg is a contaminant found at high levels in the traditional Inuit diet

sources. Blood mercury concentrations of the Inuit were ten‐fold higher than

that of the Quebec population (reference level <0.1‐16 nmol/L) in the

Nunavut Inuit Health Survey 2004/Qanuippitaa (Dewailly, Ayotte, et al.

2007). Moreover, adults aged 45 to 74 had statistically higher Hg blood

concentrations compared with younger adults. This latter increasing trend in

relation to age probably reflects a higher intake of traditional food, as other

Inuit studies have found blood Hg concentrations increased significantly with

increased quartiles of annual consumption of mammals and fish (Mahaffey

and Mergler 1998; Johansen, Pars et al. 2000; Bjerregaard, Young et al. 2004).

Thus, blood Hg is a surrogate biomarker for traditional food intake reflecting

long‐term intake rather than only recent Hg intake. MeHg is shown to have

toxic effects on the central nervous system, and interferes with the normal

function of cardiovascular system. MeHg was reported to be associated with

increased risk of CVD and acute myocardial infarction in European

populations (Yoshizawa, Rimm et al. 2002). In a cross‐sectional study in

Nunavik, MeHg has been shown to be associated positively with SBP and

pulse pressure after adjustment for confounders including age, gender, waist

circumference, insulin sensitivity, LDL‐cholesterol, smoking habits, alcohol

consumptions, Se and others (Valera, Dewailly et al. 2009). In terms of MeHg

exposure monitoring methods for individuals, organic and inorganic Hg are

often measured in blood samples or in hair strands, the latter being by far the

best integrator of past exposure.

With knowledge of the MeHg kinetics in humans, the levels of the both

biomarkers of blood and hair can be related to MeHg body burden and

intakes. Studies have shown, however, that inorganic Hg is the major Hg

species in hair samples (91.74%), while inorganic and MeHg are both about

41

50% of total Hg in red blood cell and serum samples (Cernichiari, Brewer et

al. 1995). Inorganic tissue Hg measured in hair could be as a result of

environmental contamination through pollution or hair dyes, and amalgam

fillings, rather than dietary contamination such as fish. In addition, Hg levels

in hair do not always reflect the reported level of MeHg intake via its dietary

consumption (Canuel, de Grosbois et al. 2006). Since MeHg is of the

component of interest, which would reflect bioaccumulation through the

food chain, blood samples are the preferable choice for this assessment as

reflecting metabolism influenced by body burden. This consistent blood

measure, used previously in earlier Inuit surveys (Van Oostdam, Donaldson

et al. 2005), can be used to track trends of exposure across different Inuit

surveys.

2.8.7. Persistent Organic Pollutants (POPs)

POPs comprise polychlorinated dibenzo p‐dioxins (PCDDs), polychlorinated

dibenzofurans (PCDfs), and polychlorinated biphenyls (PCBs) and

chlorinated pesticides (toxaphene). New POPs of interest in the Arctic are

perfluoroctanesulfonate (PFOS), halogenated phenolic compounds (HPCs)

and polybrominated diphenyl ethers (PBDEs). Through their traditional diet,

the Inuit are exposed to a large amount and mixture of POPs, which

bioaccumulate in fatty tissues of marine mammals, fish, and terrestrial wild

game due to their lipophilic properties. Similarly in humans, POPs

accumulate in fatty moiety of tissues and so are in relatively high

concentrations in adipose tissues and breast milk.

Several studies indicated detrimental effects of POPs on several aspects of

cognitive development in utero. Others reported association with endocrine

disruptions and carcinogenic, hepatotoxic and immunotoxic effects (Van

Oostdam, Donaldson et al. 2005). In the Inuit populations, significant amount

42

of some of these compounds have been detected in the blood and adipose

tissue of adults and in cord blood from newborn studies. The data from

exposure assessment studies showed contradictory trends as to how POPs

are related to traditional food consumption. Surveys of dietary intake in

QikiqtaIjuaq (Broughton Island), Nunavut have shown that organochlorine

exposures in 1999 were higher than in l987 ‐1988, particularly among the

95th% consumers of narwhal muktuk and blubber in the community

(Kuhnlein, Receveur et al.1995).

On the other hand, data from Nunavik Inuit Health Survey 2004/Qanuippitaa

showed statistically significant declines in plasma levels for all

organochlorines observed between 1992 to 2004 (Dewailly, Dellaire et al.

2007). Surprisingly, the report on POPs from the 2004 survey concluded that

with the exception of PFOS, most compounds were not related to traditional

food consumptions, which is contrary to many previous reports

demonstrating such links in Canadian and Greenlandic Inuit (Van Oostdam,

Donaldson et al et al. 2005). PFOS plasma concentrations showed an age‐

related increase attributable to higher traditional food intake in people aged

45 to 74 yrs, which are found to be highest consumers of marine mammal fat

and fish (Dewailly, Dellaire et al. 2007). Such direct correlations were lacking

in relation to other POPs in the survey. The latter survey noted, however, that

tissue organochlorine levels in women of childbearing age still are of concern

since their blood POPs levels are reported to exceed the levels set by Health

Canada. This latter finding could be explained by an earlier report that

showed that women increase their consumption of traditional foods due to

changes in appetite and food preferences during pregnancy and the belief

that traditional food consumption would enhance health during pregnancy

(Muckle, Ayotte et al. 2001).

43

The above conflicting data could be due to effects modifiers other than age

such as BMI, serum cholesterol, fatty fish consumption, and lactation as seen

in other populations (Laden, Neas et al. 1999; Vaclavik, Tjonneland et al.

2006). In addition, statistical adjustments for these factors could be

complicated with sampling method adjustments in the calculation of variance

needed in a survey design. Thus, much of the data is interpreted crudely with

stratification of the survey population according to unadjusted tertiles of

exposures.

Furthermore, spatial variation in organochlorine levels among communities

could detect variation in exposure that might be overlooked in compiling

data of all communities in Nunavik for survey purposes. A smaller study

carried out the village of Salluit in Nunavik by Belanger and colleagues

(Belanger, Dewailly et al. 2006; Belanger, Mirault et al. 2008) showed that

PCBs correlated positively with age and BMI, suggesting its relation to longer

exposure and that their tissue concentrations increase with increased

adiposity and or fatty food intake. Further, PCBs correlated positively with

tissue levels of n‐3 PUFA, Se and MeHg (Belanger et al. 2006, 2008), which

suggest that all of these factors are present within the same matrix, most

likely in a dietary source. In addition, PCBs were found to be an independent

predictor of plasma concentrations of oxidized(ox)‐LDL, an important

component in the pathogenesis of atherosclerosis (Belanger, Mirault et al.

2008). Other identified predictors of ox‐LDL were both LDL‐C and HDL‐C,

which suggests a close relation of PCBs to all lipoprotein components

regardless of the lipoprotein oxidation level. However, since investigators

typically have not had a priori for the decision regarding the direction of the

association (dependent vs. independent), this may have resulted in

conflicting findings because high inter‐correlations between the variables

were not accounted for in the analyses.

44

2.8.8. Smoking Status

Smoking is a confounder that relates to the development of CVD through

many mechanisms including increasing oxidative stress, blood pressure and

blood lipids, and accumulation of atherosclerotic plaque. In the Genetics of

Coronary Artery Disease in Alaskan Natives (GOCADAN) study, current

smokers (OR=2.1; 95% CI=1.1‐3.8) and those who had quit < 5 yrs ago

(OR=1.6; 95% CI 1.1‐2.2) were more likely than nonsmokers to have carotid

plaques (Kaufman, Roman et al. 2008). Smoking also is an important

predictor of levels of Se in toenails (Hunter, Morris et al. 1990). Passive

smoking is also associated with increased risk of CHD. For instance,

increased cotinine levels in nonsmokers were associated with a 60% higher

CHD risk over a 20 yr period in the British Regional Heart Study (Whincup,

Gilg et al. 2004). Increased levels of F2‐IsoPs have been documented with

active cigarette smoking in blood (Morrow, Frei et al. 1995) and in urine

(Pilz, Oguogho et al. 2000). Similar findings were observed with passive

smoking. The prevalence of smoking is very high amongst the Inuit. Data

from Nunavik Inuit Health Survey 2004/Qanuippitaa showed that nearly

three quarters of Inuit aged 15 yrs and over were smokers (Plaziac 2007).

More women were found to be smokers than men; however, they smoked

fewer cigarettes per day and young adults aged 8 to 29 yrs smoked the most.

Further, the Inuit displayed a general knowledge on the health consequences

of passive smoking and 84% of participants reported smoking restrictions in

their homes.

Data on smoking has typically been obtained through an administered

lifestyle questionnaire that seeks information on the participants' current

smoking status, age at which first cigarette was smoked, age at which

smoking became a daily habit, number of cigarettes smoked, and finally

cessation of smoking (Block, Dietrich et al. 2006). Serum cotinine assay is

45

considered the 'gold standard' measure of exposure to cigarette smoke by

which current smoking status can be validated with high sensitivity and

specificity (Block, Dietrich et al. 2006). When assessing serum cotinine there

is substantial within‐person fluctuation in participants trying to quit or cut

down as a result of increased CVD risk in this age group. This fluctuation is

expected as was portrayed by the data from the Nunavik Health Survey 2004

whereby a high proportion of smokers had tried to stop smoking in the 12

months preceding the survey. As a result, cotinine measures alone may be of

limited use for validation of amount smoked, as they are informative only

about recent exposure, vary with individual smoking topography and are

dependent on time lapsed since the last cigarette smoked. Thus, for purposes

where timing, intensity and duration of exposure are critical, self‐reported

history of cigarette consumption may be a more relevant to atherosclerosis

development than current smoking status. For the substantial within‐person

variation (Block, Dietrich et al. 2006), numerous measures of cotinine would

be needed to characterize patterns of exposure; however, because of

practicality purposes in population studies an attenuation factor for serum

cotinine is usually applied in statistical analyses.

46

BRIDGE 1

From the preceding literature review, there is an indication that the Inuit

possess robust antioxidant status and this has been observed in few studies

mainly in the region of Nunavik (Belanger 2006, 2008). Assessment of

antioxidant status has its inherent limitation to fully explore whole body

oxidative stress status in the Inuit and how it relates to the emerging

increase in cardiometabolic abnormalities such as obesity, hypertension,

smoking, and the metabolic syndrome among the Inuit. We sought in this

study to assess oxidative stress using gold standard measure F2‐isoprostanes

and the novel biomarker, isofurans, in relation to the various cardiometabolic

risk factors.

47

CHAPTER 3: MANUSCRIPT 1

Isoprostanes and isofurans as non‐traditional risk factors for cardiovascular

disease among Canadian Inuit

Dalal Alkazemi1, Grace M Egeland1, L. Jackson Roberts II2, Stan Kubow1.

Authors affiliations:

1 School of Dietetics and Human Nutrition, & Centre for Indigenous Peoples’

Nutrition and Environment, McGill University, 21,111 Lakeshore Road, Ste‐

Anne‐de‐Bellevue, Quebec H9X3V9, Canada.

2 Departments of Pharmacology and Medicine, 522 RRB, Vanderbilt

University, Nashville, YN 37232‐6602, USA.

Corresponding Author: Stan Kubow, PhD

Corresponding Author’s Information: School of Dietetics and Human

Nutrition, McGill University, 21,111 Lakeshore Road, Ste‐Anne‐de‐Bellevue,

Quebec H9X3V9, Canada. Email: [email protected]

This published manuscript is reproduced here with permission of Informa

UK, Ltd. Article In Press: Free Radic Res. doi:10.3109/10715762.2012.702900

[Online July 11, 2012] Minor editorial changes were made to the original

publication for consistency with other thesis chapters.

48

3.1. Abstract

Objectives: The aim of the present study was to investigate the potential

importance of oxidative stress, measured by isoprostanes‐related

compounds, as non‐traditional risk factor for cardiovascular disease. We

planned to examine the relationship between concentrations of plasma F2‐

isoprostanes (F2‐IsoPs), isofurans (IsoFs), measures of obesity, and various

cardiometabolic risk factors. Materials and Methods: Cross‐sectional study

using a sub‐sample from the population of a survey conducted in the summer

and fall 2007 and 2008 by Canadian Coastguard Ship Amundsen in 36

Canadian Arctic Inuit communities. Subjects included a subset (n=233) of a

total study population (n=2595) with a mean age 42.56 ± 15.39 yr and body

mass index 27.78 ± 5.65 kg/m2. Plasma levels of F2‐IsoPs and IsoFs was

determined by gas chromatography/ negative iron chemical ionization/mass

spectrometry (GC/NICI/MS) method; and their relationships to waist

circumference, blood pressure C reactive proteins (CRP), blood lipids, and

fasting glucose were assessed by multivariate analyses. Results: Plasma F2‐

IsoPs correlated positively with CRP (r=.132, p =.048) and systolic blood

pressure (SBP) (r=.157, p=.024) after adjustment for age, sex and body mass

index. IsoFs correlated with waist circumference (WC) (r= .190, p= .005) and

SBP (r=.137, p =.048). F2‐IsoPs were not found elevated in smokers (p=.034),

whereas IsoFs were decreased in smokers (p=.001). WC, SBP and sex were

found to be major correlates of oxidative stress in Canadian Inuit.

Conclusions: Plasma measures of F2‐IsoPs and IsoFs increase with increased

obesity and associated cardiometabolic risk factors, including CRP and blood

pressure. Simultaneous measurement of IsoFs provides an advantageous

mechanistic insight into oxidative stress not captured by F2‐IsoPs alone.

49

3.2. Introduction

There is accumulating evidence that oxidative stress is a key mechanism

underlying the development of cardiovascular diseases and type 2 diabetes

mellitus (Morrow 2003; Stocker and Keaney 2004; Morrow 2005; Singh and

Jialal 2006). Specifically, oxidative stress has been hypothesized to contribute

to the development and progression of atherosclerosis via the chemical

modification of proteins, nucleic acids and lipids leading to production of

oxidative stress by‐products, such as lipid peroxides (Stocker and Keaney

2004). The oxidation of unsaturated lipids in cell membrane may modulate

diverse signal transduction pathways implicated in atherosclerosis leading to

increased expression of cell adhesion molecules, induction of pro‐

inflammatory pathways, activation of matrix metalloproteinase, vascular

smooth muscle cell proliferation, endothelial dysfunction and oxidation of

low density lipoprotein cholesterol (LDL‐C) (Stephens, Khanolkar et al. 2009)

Over the past decade, F2‐Isoprostane (F2‐IsoPs), which are prostaglandin‐like

products of the free radical‐catalyzed peroxidation of arachidonic acid, have

emerged as the “gold standard” in vivo assessment of oxidative stress (Milne,

Sanchez et al. 2007): associated with obesity and hypertension (Keaney Jr,

Larson et al. 2003) they promote inflammation and vasoconstriction (Hou,

Roberts et al. 2004; Basu 2006), the latter of which has been partly related to

their ability to induce thromboxane formation leading to contraction of

vascular smooth muscle and endothelial cell death (Minuz, Patrignani et al.

2002; Brault, Martinez‐Bermudez et al. 2003). In addition to F2‐IsoPs, the

ratio of F2‐IsoPs and isofurans (IsoFs) is thought to reflect ambient oxygen

concentrations within the environment in which lipid peroxidation occurs.

Thus, it gives a better‐integrated view of oxidative injury to measure both

products together (Roberts and Fessel 2004) IsoFs formation is favored and

occurs away from F2‐IsoPs formation as ambient oxygen concentration

50

increases. This can happen as a result of supplemental oxygen use, impaired

mitochondrial oxygen metabolism, and would be predicted to happen in any

other setting where ambient oxygen concentrations are increased above

baseline (Roberts and Fessel 2004).

Oxidative stress has rarely been evaluated among Indigenous Peoples

undergoing societal and nutrition transitions with consequences for

increased obesity and obesity‐related chronic diseases. Inuit across the

Arctic are undergoing rapid social, nutritional and health transitions that are

leading to deteriorating metabolic profiles with increased prevalence of

obesity, hypertension, diabetes, and heart disease (Bjerregaard, Dewailly et

al. 2003; Bjerregaard, Young et al. 2004). However, some evidence suggests

that the metabolic consequences of obesity may be markedly less among

Inuit than that observed in Euro‐Caucasians or other Indigenous Peoples

(Young, Bjerregaard et al. 2007). Thus, the objectives of the present study

were to examine the relationship between plasma levels of F2‐IsoPs and IsoFs

with demographic and cardiometabolic risk factors among Canadian Arctic

Inuit who have been experiencing an epidemiologic transition associated

with westernization (Bjerregaard, Dewailly et al. 2003; Bjerregaard, Young et

al. 2004; Young, Bjerregaard et al. 2007).

3.3. Subjects and Methods

3.3.1 Subject recruitment

The current study is based upon a random subsample of participants of a

population‐based International Polar Year Inuit (IPY) Health Survey details

of which are available elsewhere (Egeland, Cao et al. 2011). In brief, a cross‐

sectional survey was conducted in the summer and fall 2007 and 2008 for 33

coastal communities and for three non‐coastal communities representing all

communities in Inuvialuit Settlement region (ISR, Northwest Territories),

51

Nunavut and Nunatsiavut (Northern Labrador). Trained interviewers and

nurse staff collected information on subjects’ dietary habits, physical activity,

psychosocial factors, medical history, blood pressure, anthropometric

indices, fasting lipids, and various clinical indices. Fasting blood samples

were prepared and stored at ‐ 80°C for future analyses. Territorial research

licenses were obtained and the Ethical Review Board of the McGill University

Faculty of Medicine approved the study. Informed consent was obtained from

all participants prior to enrollment.

3.3.2 Anthropometric, physiologic measures, and definitions

Height, weight, and waist circumference (WC) and blood pressure were

measured during clinical session, performed by trained research nurse

according to the same standard protocol in survey as previously reported

(Egeland, Cao et al. 2011). A body mass index (BMI) of 25.0–29.9 kg/m2 was

considered overweight, and a BMI of 30 kg/m2 or greater was considered

obese. Because no cutoff for central obesity in aboriginal populations had yet

been defined, the obesegenic waist cut points for whites were used as 88 cm

or greater for women and 94 cm or greater for men (Alberti, Zimmet et al.

2005). Hypertension (HTN) defined as those presented with BP reading that

meet diagnostic cutoff of 140/90 mmHg calculated as an average of 3

independent readings on the day of the survey and/or nurse‐recorded use of

HTN medications. Diabetes were identified in self‐reports and/or the use of

anti‐diabetic medications and/or having elevated fasting or elevated 2‐hr

Oral glucose tolerance test (OGTT) when was available. While there have

been efforts to harmonize definitions of the MetS (Alberti, Eckel et al. 2009),

we present data for the most commonly used definitions. The prevalence for

the MetS was identified using both the US National Cholesterol Education

Program in the Adult Treatment Panel III (NCEP) and the International

Diabetes Federation (IDF) as presented in Table 3. CRP levels were

52

categorized according to the American Heart Association Criteria for

inflammatory state (Grundy, Hansen et al. 2004).

3.3.3 Laboratory methods

Fasting serum total cholesterol, high‐density lipoprotein cholesterol (HDL‐C)

and triglycerides were determined using enzymatic colorimetric tests and

low‐density lipoprotein cholesterol (LDL‐C) was calculated by Nutrasource

Diagnostics, Guelph, ON (Life Laboratories–Gamma Dynacare). Serum hs‐CRP

concentration was determined using immunoturbidimetric assay with

SYNCHRON® High Sensitivity CRPH reagent in conjunction with

SYNCHRON® Systems CAL 5 Plus (Beckman Coulter Inc., Fullerton, CA, USA)

in CINE at McGill University.

3.3.4 Plasma analysis of F2‐IsoPs and IsoFs

For plasma, samples were collected in vacutainer blood collection tubes

coated with ethylenediaminetetracetic acid (EDTA), and after centrifugation

plasma tubes were stored at ‐80°C until time of analysis. Purification,

derivatization, and analysis of F2‐IsoPs and IsoFs by stable isotope dilution

gas chromatography/negative ion chemical ionization mass spectrometry

(GC/NICI/MS) were performed as previously described (Morrow and

Roberts 1999). An Agilent 5973 Mass Spectrometer coupled to an Agilent

6890N Gas Chromatograph using a 15 mDB 1701 GC column was utilized

with an inlet temperature of 260°C. The helium carrier gas flow rate was 2

ml/min. For sample injection, the GC oven was programmed to run from 190

to 300°C at 20°C/minute for 9 min. Selective ion GC/NICI/MS monitoring was

569 m/z for F2‐IsoPs, 585 m/z for IsoFs, and 573 m/z for the internal

standard [2H4] 15‐F2t‐IsoP. Values are expressed in picograms per milliliter

of plasma (pg/mL). The precision of the assay is ± 6% and the accuracy is

96%.

53

3.3.5 Statistical analysis

Anthropometrics, clinical, and biochemical measures for all subjects were

reported as mean ± SD. Prevalences of preexisting medical conditions were

determined based on self‐reports and/or medication intake for the existing

condition. All variables were treated on a continuous scale in statistical

analyses. Skewed variables were logarithmically transformed. Results were

described as geometric mean ± 95% CI for log‐transformed data. The plasma

levels of F2‐IsoPs and IsoFs were compared using Student’s t test between

different categories of risk, i.e., male vs. female, obese vs. non‐obese,

hypertensive vs. non‐hypertensive, etc. Comparisons among subgroups were

performed by ANOVA and covariance analyses adjusting for age, sex, and

central obesity, which provided predicted mean oxidative stress markers by

smoking status, hsCRP‐level, and presence or absence of MetS. Boneferroni’s

post hoc test was used when a significant group effect was observed.

Correlation analysis was performed using Pearson’s correlation analysis to

assess the relationship between plasma concentrations of measures of

oxidative stress (F2‐Isops and IsoFs) and other study parameters including

obesity measures and cardiovascular risk factors. Partial correlation analysis

was performed accounting for age, sex and BMI. All p values were two tailed,

and p <0.05 was considered significant for all tests performed. To estimate

final predictors of the individual biomarkers variability and examine the

influence of confounding variables, multivariate analysis with stepwise

regression was used. For the stepwise regression, a α‐value of 0.05 was used

to exclude variables that had little or no influence on the biomarker under

analysis. All statistical analysis was performed using SPSS version 13.0

software (SPSS Inc., Chicago, IL).

54

3.4 Results

3.4.1 Subject characteristics

The current study is based on 294 subjects aged 18 yrs and older chosen

randomly from the IPY Inuit Health Survey of whom specimens were

available for full analysis. According to AHA recommendations (Grundy,

Hansen et al. 2004) CRP levels of 10 mg/L or greater represent evidence of

active infection, systematic inflammatory processes or trauma and thus those

individuals were excluded. A total of 60 subjects had CRP >10 were excluded

in the current study sample of 234 subjects; in addition, one outlier with a F2‐

IsoPs level much lower than the lower limit of 95% CI was also excluded,

which decreased the sample size to 233 subjects. The mean age of the subject

was 42.56±15.39 yr. The study sample characteristics are presented in Table

1. Of the 233 participants, (56% women), mean BMI was 27.78 kg/m2, 33.5%

were overweight and 30.4% were obese. Based on medical histories of

participants the prevalence of diabetes was 5.7%, hypertension 28.6%,

dyslipidemia 13.7%, cancer 8.5%, episode of heart attack 5.3% and stroke

3.1%. Higher prevalence of the MetS was observed using NCEP definition

with 56% subjects (69% women) identified with the MetS. Using IDF

definition lower prevalence was demonstrated, with 42% subjects (63%

women) identified with MetS. Seventy percent of participants were current

smokers with 41.8% of smokers reporting >10 cigarettes/day and 65% of all

participants reported drinking alcohol in the past year.

3.4.2 Oxidative stress biomarkers

Of the 294 individuals available for analyses, CRP was elevated for 20.75%.

All data presented represent 233 individuals with CRP < 10 mg/L. The

geometric mean (and interquartile ranges) for F2‐IsoPs was 27.33 (25.94‐

28.78) pg/ml; and for IsoF was 20.86 (18.90‐23.02) pg/ml (Table 1). F2‐

IsoPs correlated with IsoFs (r=.379, p=.001). IsoFs:F2 was determined to

55

study the interrelationship between the compounds as it has been suggested

that this ratio portray oxidative stress status of subjects more

comprehensively than any of the compounds alone (Roberts and Fessel

2004). IsoF:F2 positively correlated with IsoFs (r=.871, p=.001), but no

correlation was observed between IsoF:F2 and F2‐IsoPs. From these data, it

appears that only selected biomarkers are correlated.

3.4.3 Isoprostane correlates ‐ CVD risk factors

F2‐IsoPs and IsoFs correlated positively with obesity measures including

BMI, WC and %BF. Interestingly, IsoFs remained significantly positively

associated with WC after correcting for age, BMI and sex (Table 2). F2‐IsoPs

and IsoFs correlated positively with systolic blood pressure (SBP) even after

adjustment for age, sex and BMI. IsoFs correlated positively with both SBP

and diastolic blood pressure (DBP), however after adjustments correlation

with DBP was slightly attenuated from (r=.141, p=.036) to (r=.125, p=.053).

Further adjustment for smoking, alcohol intake, medication or supplement

intake did not change any of the correlations.

3.4.4 Oxidative stress biomarkers in relation to categorical variables

Levels of oxidative stress were studied among subjects who reported medical

conditions listed above as compared to the rest of the sample; no difference

was detected in any of the conditions. The exceptions were subjects with

history of heart attack and other heart disease related conditions and

subjects with diagnosed dyslipidemia (self‐identified according to health

survey questionnaire yes=with condition and/or taking medications for lipid

abnormality/no=absence of condition). The former condition group have

lower levels of F2‐IsoPs {yes, n=12, 20.56 (16.48‐25.67) vs. 27.83 (26.35‐

29.39), p=.012} and lower levels of IsoFs, {11.89 (5.81‐24.33) vs. 21.63

(19.58‐23.89), p=.008}; whereas the latter group (n=29) have higher levels of

56

IsoFs {32.28 (23.40‐44.55) vs. 19.59 (17.47‐21.68), p=.003}, and higher

IsoFs:F2 {1.04 (.95‐1.14) vs. .90 (.87‐.93), p=.006}. Adjustment of medication

use did not change the results in any of above stated conditions.

Further analyses were performed using various categorical variables

composed from the survey data that have relevance to oxidative stress

shown in Table 3. Subjects identified with MetS using both NCEP and IDF

definitions had higher IsoFs levels (Table 3), and in addition with IDF,

significantly higher levels of F2‐IsoPs were also observed. Using

hypertriglyceridimic waist as risk category, both F2‐IsoPs and IsoFs levels

were significantly higher in subjects recognized with the combination of

obesegenic waist and high levels of triglycerides {28% overall, 50% women}.

Subjects who reported taking any medications or consuming alcohol did not

show any significant difference in oxidative stress levels when compared to

their counterparts. Sensitivity analysis was performed to further explore

effects of alcohol intake on oxidative stress and no difference was observed

between subjects who reported alcohol consumption and those who did not.

The lack of alcohol intake on oxidative stress was also demonstrated by the

observation that overall mean isoprostane levels were within the normal

range. Subjects reported taking nutritional supplements had higher IsoFs

levels and higher ratio of IsoFs:F2‐IsoPs. Current smokers had lower levels of

IsoFs and IsoFs:F2‐IsoPs ratio as they were leaner. Indeed, higher levels of F2‐

Isops, IsoFs and IsoFs:F2‐IsoPs were observed in subjects with central obesity

and with higher body fat and among those who have higher blood pressure

(Table 3). In contrast, when levels of isoprostanes were compared using BMI

cutoffs there was no statistical significance in any of the biomarkers (data not

shown), which suggests abdominal obesity is a better indicator of risk among

Inuit than BMI alone. In terms of sex, women had higher levels of F2‐IsoPs

than men (Table 3), which could be attributed to higher adiposity found in

57

women demonstrated by higher BMI, body fat, and central adiposity (data

not shown). Of note, overall men showed significantly higher blood pressure

values and lower HDL‐C concentration when compared to women (data not

shown). Since the mean ranges were within normal levels, however, one

cannot infer deleterious or worse metabolic profile than women.

3.4.5 Smoking

Further analyses were performed to explain smoking status according to

duration of smoking (Table 4). Subjects were further categorized as: (a)

nonsmokers (never smoked); (b) ex‐smokers whom did not smoke but

answered yes to ever smoked; and (c) current smokers. Current smokers had

significantly lower levels of IsoFs than ex‐smokers and non‐smokers. Similar

results were obtained after adjustment of age, sex, WC and supplement use.

Further adjustment of medication or alcohol use did not change or attenuate

the significance. Stratifying according to BMI (not shown) showed most

smokers were lean (n=123) and that non‐obese smokers and obese smokers

(n=33) had lower IsoFs levels than both non‐obese non‐smokers (n=3) and

obese non‐smokers (n=7); however, this latter relationship did not reach

statistical significant.

3.4.6 Inflammation

Both F2‐IsoPs and IsoF both positively correlated with CRP and its

association with F2‐IsoPs remained significant after adjustment of age, sex

and BMI (Table 5). Subjects with CRP level <1 mg/L had significantly lower

levels of F2‐IsoPs and IsoFs than those in >3 mg/L even after adjustment of

age and sex. Further adjustment of BMI, %BF or WC did not affect the

significance of F2‐IsoPs levels but did attenuate difference associated with

IsoFs.

58

3.4.7 Predictors of Isoprostanes

The results of the stepwise multivariate regression models are contained in

Table 6. Same model was used for all biomarkers to correct for confounding

effects presented by the various independent variables. SBP, WC, age and sex

were found to be correlates of F2‐IsoPs (R2=.124). WC was the only

independent correlate for IsoFs; and WC was the final predictor for IsoF:F2‐

IsoPs (R2=3.5).

3.5 Discussion

A major finding from this study is that levels of both F2‐IsoPs and IsoFs were

elevated in Inuit categorized with the MetS. The isoprostane markers were

associated with abdominal obesity and with the hypertriglyceridemic waist

phenotype, which is considered an indirect indicator of visceral fat (Lemieux,

Poirier et al. 2007); and is associated with a deteriorated cardiometabolic

risk profile and an increased risk for coronary artery disease (Arsenault,

Lemieux et al. 2010). Visceral adipose tissue has been shown to be a unique

correlate of F2‐IsoPs even after adjustment of BMI and WC (Pou, Massaro et

al. 2007).

A novel finding of the present study was that abdominal obesity was

significantly related to both plasma F2‐IsoPs and IsoF while only IsoF

remained significantly correlated with WC even after adjustment for age, sex

and BMI. This latter observation signifies the importance of the assessment

of multiple entities of isoprostanes to detect pathophysiological associations

with oxidative stress, which might be missed via the measurement of F2‐

IsoPs alone. Prior literature has emphasized the important relationship of

visceral adiposity with inflammation and cardiometabolic risk (Forouhi,

Sattar et al. 2001; Brooks, Blaha et al. 2010). The present study has

demonstrated a positive association between isoprostanes and CRP,

59

consistent with findings in other high‐risk populations (Handelman, Walter

et al. 2001).

Both plasma F2‐IsoPs and IsoFs were associated with measures of blood

pressure, with F2‐IsoPs depicting increased oxidative stress among subjects

with elevated blood pressure. The relationship of F2‐IsoPs to both CRP and

blood pressure remained significant even after multivariate adjustment. Our

data support previous reports that F2‐IsoPs have an independent association

with hypertension (Keaney Jr, Larson et al. 2003) and that plasma CRP

predicts the development of hypertension (Cheung, Ong et al. 2011).

Increased generation of F2‐IsoPs may be responsible for early endothelial

dysfunction that is partly mediated by oxidative degradation of NO, which is

an important endothelium derived vasodilator (Lavi, Yang et al. 2008)

Likewise, CRP has been indicated to be a circulating biomarker of endothelial

dysfunction. CRP is associated with induction of pro‐inflammatory cytokines

and adhesion molecules leading to endothelial dysfunction and carotid

intima‐media thickening (Szmitko, Wang et al. 2003). Collectively, the

present data supports inflammation as a strong candidate as an underlying

mechanism for the pathogenesis of hypertension (Ghanem and Movahed

2007).

In this study, we found no association of smoking with F2‐IsoPs after

adjustment for important covariates. The majority of our population were

smokers with 95% of smokers reporting smoking >10 cigarettes per day with

exposure to second hand smoke in the nonsmokers being very likely, which

could partly account for the lack of a significant association of F2‐IsoPs with

smoking. While an increase in F2‐IsoPs has been documented with active

cigarette smoking (Morrow, Frei et al. 1995; Helmersson, Larsson et al.

2005), some reports have not found associations between plasma or urine

60

F2‐IsoPs and smoking status (Dietrich, Block et al. 2002; Ward, Hodgson et al.

2004). Other studies have reported lower levels of urinary F2‐IsoPs in

smokers compared to nonsmokers (Il'yasova, Morrow et al. 2005). Differing

features of the smoking populations has been postulated to account the

discrepancies in F2‐IsoP levels among studies such as the proportion of

higher intensity smokers (>2 packs/day) and exposure to smoke within 24‐

hr before blood draw (Dietrich, Block et al. 2002). A unique and apparently

contradictory finding was the clear gradient of diminishing levels of the IsoFs

with increasing levels of smoking. An oxygen deprived environment noted in

the tissues of smokers might have resulted in a lower production of IsoFs

since the production of IsoFs is highly disfavored as compared to F2‐IsoPs

formation in tissues with very low oxygen tension such as the liver (Roberts

and Fessel 2004). A significant degree of tissue hypoxia has been

demonstrated among smokers linked with vasoconstriction, arterial

baroreflex alterations, and endothelial dysfunction (Jensen, Goodson et al.

1991; Sørensen, Jõrgensen et al. 2009). Given the widespread smoking and

second hand smoke exposure in our population, we speculate that smokers

have lower levels of IsoFs due to tissue hypoxia and compromised lung

function due to chronic smoking exposures (Belanger, Dewailly et al. 2006).

As the present study is the first to examine for IsoFs status among smokers,

further investigations are needed among other populations to assess whether

low IsoFs could be a marker for smoking status.

Individual correlations with individual lipid and lipoprotein components

were not observed (Table 5). This latter finding could be due to that only

exhibited a mild form of dyslipidemia was noted in the Inuit population in

terms of moderately elevated triacylglycerol concentrations and normal

concentrations of LDL‐C. Although F2‐IsoPs are partly derived from oxidation

of LDL (oxLDL), the association between plasma ox‐LDL and plasma F2‐IsoPs

61

has been shown to be partly dependent upon the severity of dyslipidemia

(Hansel, Giral et al. 2004). In concert with the above, significantly higher F2‐

IsoPs levels were seen in subjects with diagnosed dyslipidemia when

compared to the rest of the Inuit as indicated above.

In our study sample, women showed higher level of F2‐IsoPs when compared

to men as has been observed previously in other populations (Dietrich, Block

et al. 2002; Keaney Jr, Larson et al. 2003; Gross, Steffes et al. 2005). This can

be partly attributed to a larger proportion of women with obesity and

metabolic cardiovascular risk factors as compared to men. Increased levels of

blood pressure, CRP and F2‐IsoPs, however, accompanied increased

abdominal obesity in both sexes; and controlling for BMI or WC did not

attenuate the effect of sex (data not shown). There is suggestion that women

are more protected than men against the effect of oxidative stress or lipid

abnormalities (Mosca, Appel et al. 2004). Longitudinal studies are needed to

evaluate whether such sex specific responses exist in the Inuit population

with respect to isoprostane status and CVD events.

In the present study, oxidative stress did not increase with age, which is in

agreement with several adult population studies (Dietrich, Block et al. 2002;

Keaney, Larson et al. 2003; Il'yasova, Morrow et al. 2005). Surprisingly, F2‐

IsoPs were negatively associated with age, which could be partly accounted

by a better metabolic profile in older subjects, as they were leaner. Also, in

another analyses of the current study population, striking age‐gradients in

long‐chain n‐3‐red blood cell fatty acids were observed in two geographic

areas of the survey (Zhou, Kubow et al.), indicating that older Inuit are

consuming more traditional foods as was also demonstrated in the Inuit

region of Nunavik (Quebec) (Belanger, Dewailly et al. 2006). Further

analyses, beyond the scope of the current study, are underway to evaluate

62

whether traditional nutritional intakes could account for the decreased in F2‐

IsoPs with age.

One limitation of the study is that the adiposity measures did not include

direct evaluation of visceral and hepatic fat. Thus, multivariate analyses

controlling for adiposity should be interpreted with caution. Further, self‐

reported alcohol data may not be accurate. There are many factors that can

cause variations in oxidative stress including medication intake, over‐the‐

counter‐antioxidant supplement use and other dietary factors; however, the

access of medical chart reviews facilitated adjustment for such cofactors in

the analysis when available and as required. Given that the majority of

participants smoked, that non‐smokers were likely passive smokers, and that

smoking was strongly related to a lower WC and BMI, we had insufficient

numbers and variability in smoking and adiposity status to fully evaluate the

relationship of smoking with oxidative stress in the current study.

In summary, the study showed that sex, central adiposity, inflammation,

hypertension and the metabolic syndrome were related to alterations in

plasma isoprostane levels, and that the relationships were specific to

isoprostane species. Only plasma IsoFs remained significantly correlated

with the indicator of visceral adiposity. Plasma F2‐IsoPs were associated with

elevated blood pressure and systemic inflammation. Smoking was associated

with lower plasma levels of IsoFs that could be the result of lower ambient

levels of tissue oxygen among smokers. Overall, these data suggest that

obesity‐induced inflammation is associated with a state of excess oxidative

stress that highlights a trend of metabolic deterioration in the Inuit

population undergoing a health transition.

63

Table 3.1 The anthropometrical and clinical characteristics for subjects with CRP<10 (n=233). Parameters Mean ± SD Age, yrs 42.56 ± 15.39 BMI 27.78 ± 5.65 WC, cm 92.45 ± 14.43 BF% 29.62 ± 10.64 FG 4.97 ± .66 T‐Chol, mmol/L 5.02 ± 1.15 LDL‐C, mmol/L 2.92 ± 1.00 HDL‐C, mmol/L 1.47 ± .48 T‐Chol: HDL‐C ratio 3.69 ± 1.39 TG, mmol/L 1.44 ± 1.41 SBP, mmHg 117.38 ± 16.59 DBP, mmHg 76.45 ± 10.63 hs‐CRP, mg/La 1.66 (1.46‐1.86) F2‐IsoPsb, pg/mL 27.33 (25.94‐28.78) IsoFsb, pg/mL 20.86 (18.90‐23.02) IsoF: F2b .92 (.89‐.95) n Males/ Females (% Females) 103/130 (55.8) Current smoking, % yes 69.7 Alcohol consumption, % yes 55.8 Medication intake any, % yes 38.4 Nutritional supplement anyc, % yes 9.6 a. Skewed data with values less than one were normalized by log transformtion after the addition of one. b. Data were log‐transformed to be able to use parametric tests, for both (a) and (b) geometric means and 95th% CIs are presented. c. Multivitamins, various B‐vitamins, salmon oil, cod liver oil, vit‐D, calcium and Ensure

Table 3.2 Pearson’s correlations between plasma biomarkers of oxidative stress and CV risk factors. CRP<10 (n=233) F2‐IsoPs IsoFs IsoF:F2 Age BMI WC BF% FG SBP DBP Tot‐C LDL HDL TG CRP

F2‐IsoPs 1.00 .379** ‐0.11 ‐.131* .219** .215** .211** 0.01 0.10 0.08 ‐0.04 ‐0.04 ‐0.07 0.09 .145* IsoFs .379** 1.00 .871** 0.03 .257** .303** .279** .136* .143* .141* 0.02 ‐0.01 ‐0.02 0.05 .159* IsoF:F2 ‐0.11 .871** 1.00 0.07 .151* .202** .177** 0.120† 0.13 .139* 0.00 ‐0.02 0.03 ‐0.00 0.07 Partial Correlations adjusted for age, sex and BMI

F2‐IsoPs 0.09 ‐0.04 0.03 .157* 0.09 ‐0.03 ‐0.05 ‐0.00 0.03 .132* IsoFs .190** 0.11 0.09 .123* 0.125† ‐0.03 ‐0.05 0.07 ‐0.14 0.09 IsoF: F2 .153* 0.12 0.07 0.08 0.11 ‐0.02 ‐0.04 0.07 ‐0.16 0.01 *. Correlation is significant at the 0.05 level (2‐tailed). **. Correlation is significant at the 0.01 level (2‐tailed). † trend at p>0.05 and p≤0.07

65

Table 3.3 Geometric mean values and 95% CI of plasma biomarkers of oxidative stress for the total sample population and in comparison with categorical variables using student t‐test n F2‐IsoPs IsoFs IsoFs: F2 Age (yrs) < 40 106 28.85(26.67‐31.21) 20.47(17.60‐23.82) .90(.86‐.95) ≥ 40 124 25.77(23.96‐27.72)* 21.10(18.53‐24.03) .94(.90‐.98) Sex

Male 102 25.16(23.47‐26.98) 19.19(16.29‐22.61) .92(.87‐.97) Female 128 28.83(26.68‐31.17)* 22.21(19.72‐25.03) .93(.89‐.97) Abdominal obesity

M <90, F <85 68 24.21(22.28‐26.30) 16.00(13.68‐18.72) .88(.83‐.93) M ≥90, F ≥85 105 29.68(27.66‐ 31.83)** 25.80 (22.86‐29.13)** .96(.93‐.99)**

Hypertensiona

HBP <140/90 179 27.10 (25.50‐28.79) 20.24(18.14‐22.59) .92(.88‐.95) HBP ≥140/90 14 36.04(29.55‐43.96)* 23.52(16.59‐33.35) .89 (.78‐.99) Current smoking

no 158 26.61(24.90‐28.43) 18.94(16.89‐21.24) .90(.87‐.94) yes 67 28.61(25.97‐31.53) 26.26(21.63‐31.89)** .97(.92‐1.02)* NCEP‐Metsb yes 101 28.98(26.87‐31.27) 22.81(19.68‐26.45) .93(.89‐.98) no 70 26.37(23.88‐29.12) 17.51(14.53‐21.11)* .88(.82‐.93) IDF‐Metsc yes 43 30.54(27.30‐34.15) 23.90(18.94‐30.16) .93(.86‐.99) no 60 25.64(23.21‐28.33)* 16.69(13.72‐20.33)* .87(.81‐.93) Hypertriglycerimic waistc

yes 38 29.10(26.23‐32.28) 24.25(18.79‐31.29) .95(.87‐1.03) no 97 25.09(23.12‐27.23)* 17.22(14.91‐19.90)* .89(.87‐.95) a. Hypertension defined as those presented with BP reading that meet diagnostic cutoff of 140/90 mmHg calculated as an average of 3 independent readings on the day of the survey, and/or nursed recorded use of HTN medications. b. ATP definition: any ≥ 3 of : abdominal obesity (cm) M≥102, F≥88, TG (mmol/L) ≥ 1.7, HDL‐C (mmol/L) M ≤ 1.03, F ≤ 1.3; BP (mmhg) ≥ 130/85, FG (mmol/L)≥6.1. c . IDF definition: abdominal obesity (cm) M≥94, F≥88, and any ≥ 3 of TG (mmol/L) ≥ 1.7, HDL‐C (mmol/L) M ≤ 1.03, F ≤ 1.29; BP (mmhg) ≥ 130/85, FG (mmol/L) ≥ 5 .6. Obesogenic waist with WC (cm) M≥102 and F≥88; and TG (mmol/L) ≥1.7. **. means difference is significant at the 0.01 level (2‐tailed); *. at the 0.05 level (2‐tailed).

66

Table 3.4 Geometric mean values and 95% CI of plasma isoprostanes compared per smoking status and in relation to obesity. n Unadjusted

Mean (95% CI) n Adjusteda

Mean (95% CI) F2‐IsoPs Current smoker

yes 158 26.61 (24.90‐28.43) 146 27.16 (25.41‐29.04) no 67 28.61 (25.97‐31.53) 57 28.77 (25.67‐32.06) Ex‐smoker 55 28.24 (25.51‐31.26) 47 28.18 (24.95‐31.77) Non‐smoker 10 33.26 (23.49‐47.10) 9 30.76 (23.50‐40.27)

IsoFs Current smoker

yes 158 18.94 (16.89‐21.24) a 146 19.41 (17.95‐22.86)a no 68 26.26 (21.63‐31.89)** b 58 24.89 (20.37‐ 30.34)

Ex‐smoker 56 24.20 (19.78‐29.61) 48 22.44 (18.07 ‐ 27.80) Non‐smoker 10 40.78 (20.47‐81.25)** b 9 39.17 (24.15 ‐ 63.53)*b

1‐way ANOVA with Bonferroni adjusted multiple comparisons where by the mean difference is significant at .05 level. a . adjusted means for age, sex, WC and Supplement use *. significant at the 0.05 level, **. significant at the 0.01 level.

67

Table 3.5 Multiple comparisons of oxidative stress biomarkers levels between CRP categorical groups. hs‐CRP, mg/L

n Mean (95%CI) n Mean (95%CI) n Mean (95%CI)

F2 Unadjusted adjusted for age and sex adjusted for age, sex,

and WC <1 87 24.87 (22.97‐26.93)a* 87 24.15 (22.23‐26.18)a 84 24.77 (22.75‐27.04)a 1‐3 80 28.64 (26.07‐31.48)b† 80 29.31 (26.92‐31.92)b** 73 29.79 (27.23‐32.51)b* >3‐<10 59 29.34 (26.53‐32.43)b* 59 29.71 (26.98‐32.81)b** 58 29.04 (26.24‐32.14)b† IsoF

<1 87 17.34 (14.91‐20.16)a* 87 17.10 (14.55‐20.09)a* 84 19.01 (16.14‐22.39) 1‐3 81 22.51 (19.06‐26.57)b† 81 22.75 (19.32‐26.85)b* 74 23.39 (19.82‐27.67) >3‐<10 59 23.87 (19.37‐29.40)b* 59 23.99 (19.77‐29.04)b* 58 21.83 (17.99‐26.49) IsoF: F2

<1 86 .89 (.85‐.95) 86 .90 (.85‐.95) 83 .93 (.88‐.98) 1‐3 80 .93 (.88‐.98) 80 .93 (.88‐.98) 73 .93 (.88‐.99) >3‐<10 59 .94 (.88‐.99) 59 .94 (.88‐.99) 58 .91 (.85‐.97) 1‐way ANOVA with Bonferroni adjusted multiple comparisons where by the mean difference is significant at .05 level. *.significant at the 0.05 level, **. significant at the 0.01 level, † trend at p>0.05 and p≤0.07.

68

Table 3.6 Multivariate associations showing the standardized regression coefficients (β) of plasma Isoprostanes concentrations Dependent variable F2‐IsoPs IsoFs IsoF: F2

R2x100 12.400 9.200 3.500

F (significance) 6.52 (.000) 18.95 (.000) 7.92 (.005)

Independent variables β p β p β p Average SBP 0.163 0.031

WC 0.205 0.004 0.303 0.000 0.202 0.005

age ‐0.203 0.006

sex 0.199 0.005

Variables entered into the stepwise linear regression analyses for all three dependent variables: hs‐CRP, Average SBP, Triaclyglycerols, WC, age, sex, current smoking, (yes=1, no=2), nutritional supplement, (yes=1, no=2), and alcohol intake, (yes=1, no=2)

69

BRIDGE 2

We have established in the first study that the Inuit appear to be relatively

protected from oxidative stress, as measured by plasma F2‐isoprostanes and

IsoFs. We also identified that both biomarkers are related to inflammation

and the cardiometabolic abnormalities cluster that increase the risk of CVD.

Evidently, oxidative stress increased in subjects with the metabolic

syndrome. We wished to explore the impact of environmental burdens,

mainly MeHg on oxidative stress status and investigate how Se can modulate

Hg‐induced oxidative stress. Previous observational and clinical studies

indicated some contradictory evidence in relation to Hg‐Se interactions on

risk of CVD (Park and Mozaffarian 2010) and oxidative stress is a strong

plausible mechanism that may explain further these interactions (Salonen,

Seppanen et al. 2000; Ralston and Raymond 2010). It is also important to

identify how Se and Hg relate to oxidative stress and other CVD risk factors

rather than being independent predictors of risk (Virtanen, Rissanen et al.

2007), and thus could reflect the status of these risk factors. More detailed

statistical analyses have been lacking regarding sources of contaminant

exposure and how diet modulates contaminant health and CVD risk factors.

We have utilized well‐validated tools to assess markers of atherosclerosis

and oxidative stress rather than proxy measures that have been performed

thus far. This area of research not been previously studied this extensively in

the Inuit population.

70


New insights regarding tissue selenium and mercury interactions on

oxidative stress from plasma isoprostane and isofuran measures in Canadian

Inuit

Authors:

Dalal Alkazemi1,2, Grace Egeland2,3, L. Jackson Roberts II4, H.M. Chan5, Stan

Kubow1

Affiliations: 1School of Dietetics and Human Nutrition, McGill University, 21,111

Lakeshore Road, Ste‐Anne‐de‐Bellevue, QC H9X3V9, Canada

2Centre for Indigenous Peoples’ Nutrition and Environment (CINE), McGill

University, 21,111 Lakeshore Road, Ste‐Anne‐de‐Bellevue, QC H9X 3V9 3Norwegian Institute of Public Health and Faculty of Medicine and Dentistry,

University of Bergen, Bergen, Norway.

4Department of Pharmacology and Medicine, 522 RRB, Vanderbilt

University, Nashville, TN 37232‐6602, USA 5Center for Advanced Research in Environmental Genomics, University of

Ottawa, 30 Marie Curie, Ottawa, ON K1N 6N5, Canada



Nutrition,

McGill University, 21,111 Lakeshore, Ste‐Anne‐de‐Bellevue, Quebec H9X3V9,

Canada.

Email: [email protected]

71

4.1 Abstract:

Objectives: The aim of the present study was to investigate in Inuit adults

the association of selenium (Se) and mercury (Hg) status with both F2‐

Isoprostanes (F2‐IsoPs) and isofurans (IsoFs) via assessment of plasma F2‐

IsoPs and IsoFs, whole blood Se and Hg, toenail Se, and cardiometabolic risk

factors. Materials and Methods: Cross‐sectional study using a random

subset (n=233) of a population‐based survey (n=2595) conducted in the

summer and fall 2007 and 2008 in 36 Canadian Arctic Inuit Communities.

Subjects had a mean age 42.56 ± 15.39 yr and body mass index 27.78 ± 5.65

kg/m2. The whole blood Hg mean was 90.75 nmol/L and the whole blood Se

mean was 3.85 µmol/L. The relationships of plasma isoprostane levels to Se

and Hg status indicators were assessed by multivariate analyses. Results:

Plasma F2‐IsoPs was inversely correlated with Se (r=‐.186, p=.005) and

toenail Se (r=‐.146, p=.044) but not with Hg or Hg:Se. IsoFs was inversely

correlated with Se (r=‐.164, p=.014) and positively with Hg (r=.228, p<.001)

and Hg:Se (r=.340, p<.001); and not with toenail Se. The strength of the

correlations remained unchanged even after multivariate adjustments of

cardiometabolic risk factors, including age and obesity although a positive

association of F2‐IsoPs with Hg:Se was shown. Conclusions: Se and mercury

status and their interactions are important factors modulating plasma F2‐

IsoPs and IsoFs levels such that the Inuit can be protected from Hg–induced

oxidative stress because of their high Se status.

72

4.2 Introduction:

Several, but not all, epidemiological studies have provided evidence that

elevated methylmercury (MeHg) exposure, as assessed by a higher body

burden of mercury (Hg), is associated with accelerated atherosclerosis,

increasing the risk of myocardial infarction and death from coronary heart

disease (Salonen, Seppanen et al. 2000; Guallar, Sanz‐Gallardo et al. 2002;

Yoshizawa, Rimm et al. 2002; Choi, Weihe et al. 2009). Human studies have

also shown that MeHg intake can promote the development of hypertension

and arrhythmias (Virtanen, Rissanen et al. 2007). Exposure to MeHg is

associated with oxidative stress and lipid peroxidation (Huang, Cheng et al.

1996; Chen, Qu et al. 2005) that has been related to direct antagonistic effects

of MeHg on Selenium (Se)‐dependant antioxidant enzymes (Gailer, George et

al. 2000). Additionally, sequestration of Se caused by Hg‐Se adducts can lead

to a functional deficit of tissue Se caused by decreased availability of Se for

incorporation into selenoproteins involved in antioxidant defense (Tapiero,

Townsend et al. 2003). MeHg‐induced oxidative stress has also been

associated with depressed tissue levels of sulfhydryl‐dependant antioxidant

proteins and the reduced form of glutathione (GSH) (Jin, Pan et al. 2008).

Increased dietary Se, however, can counteract the sequestration of Se by Hg‐

Se adducts (Ralston, Blackwell et al. 2007) and so maintain normal

selenoenzyme activities (Ralston and Raymond 2010). Animal studies have

generally indicated that Se intake can mitigate against dietary MeHg toxicity,

which appears to be related to both the absolute and relative amounts of

MeHg and Se present in the diet (Yoneda and Suzuki 1997; Ralston and

Raymond 2010). The examination of the diets and tissues of MeHg‐exposed

animals has shown that Se:Hg molar ratios above one are protective against

adverse effects associated with MeHg exposure (Ralston and Raymond

2010). Human studies, however, have shown mixed results regarding

associations between tissue Se status and adverse outcomes from MeHg

exposure and so their interactions remain unclear. High blood Se levels have

73

appeared to counter the cataractogenic risks (Lemire, Fillion et al. 2010) and

depressed motor function (Lemire, Fillion et al. 2011) seen in an Amazonian

fish eating population exposed to elevated dietary levels of MeHg. On the

other hand, large epidemiological case control studies such as the Health

Professionals Follow‐up Study (Yoshizawa, Rimm et al. 2002) have not

supported effect modification of MeHg exposure by tissue Se.

The Inuit population has high blood levels of Se and n‐3 polyunsaturated

fatty acids when compared to Caucasians (Dewailly, Ayotte et al. 2001;

Dewailly, Blanchet et al. 2003; Belanger, Dewailly et al. 2006) as the seafood‐

based Inuit traditional diet has a rich content of Se and n‐3 polyunsaturated

fatty acids (PUFA) that have been suggested to protect against cardiovascular

disease (O'Keefe and Harris 2000; Dewailly, Blanchet et al. 2001).

Additionally, the Inuit appear to be protected against oxidative stress‐

mediated complications associated with hyperlipidemia as shown by the

absence of elevated oxidized‐LDL in association with atherogenic blood lipid

values (Dallaire, Dewailly et al. 2003; Belanger, Dewailly et al. 2006). The

Inuit diet, however, is associated with MeHg exposure through the

consumption of traditional food. Elevated exposure of MeHg among the Inuit

may be associated with cardiovascular risk as it was positively and

independently associated with systolic blood pressure and pulse pressure

(Valera, Dewailly et al. 2008), which may indicate interference with the

normal functioning of cardiovascular system (Oka, Matsukura et al. 2002;

Grandjean, Murata et al. 2004; Pedersen, Jorgensen et al. 2005). It is unclear,

however, whether the high Se intake from the traditional Inuit diet protects

against MeHg‐ mediated oxidative stress. Inuit subjects from Nunavik were

recently shown to have elevated levels of plasma lipophilic antioxidants such

as α‐tocopherol, ubiquinone‐10 and coenzyme Q10 in comparison with

healthy Caucasian controls but no association between these plasma

biomarkers was seen with MeHg intake or other markers of traditional food

74

intake (Belanger, Mirault et al. 2008). Belanger et al. (2008) suggested that

an unusually elevated ratio of ubiquinone‐10 to total CoQ10 in the Inuit

plasma was reflective of oxidative stress and that the high levels of plasma

antioxidant components could reflect an adaptive response to an oxidative

stress of undetermined origin. Assessment of antioxidant‐related plasma

components, however, may have limited sensitivity to assess the

interrelationship between dietary MeHg exposure and tissue Se on oxidative

stress status.

In the present study, we measured F2‐isoprostanes (F2‐IsoPs) by mass

spectrometry, which is recognized as one of the most reliable approaches for

in vivo assessment of whole body oxidative stress (Kadiiska, Gladen et al.

2005). In a prospective study involving the Finish population, a negative

association was noted between blood Se concentrations and F2‐IsoPs.

However, the interrelationship between tissue Se status and dietary MeHg

exposure has not been previously studied with respect to F2‐IsoPs levels. To

obtain a more comprehensive view of possible oxidative injury, we also

measured the ratio of F2‐IsoPs to isofurans (IsoFs), which are products of

lipid peroxidation with a substituted tetrahydrofuran ring (Fessel and

Roberts 2005). The formation of F2‐IsoPs and IsoFs is differentially regulated

by oxygen tension; the formation of F2‐IsoPs is favored by low oxygen

tension whereas the formation of IsoFs is favored by elevated oxygen tension

that can occur in settings of mitochondrial dysfunction. Our previous work in

a sample of the International Polar Year (IPY) survey population indicated

that IsoFs and F2‐IsoPs showed a positive association to the pro‐

inflammatory C‐reactive protein (CRP), after adjustments for various

confounders (Alkazemi, Egeland et al. 2012). Oxidative stress and

inflammation are considered to be contributing components to the

progression of atherosclerosis (Uno and Nicholls 2010) and so could be

mediators of the adverse cardiovascular effects seen in MeHg‐exposed Inuit

subjects.

75

In this current study the association between F2‐IsoPs and IsoFs was

examined with respect to tissues reflecting short term (blood) and long term

(toenail) intake of MeHg and Se in order to assess these relationships with

CVD risk factors. An additional aim was to examine how much of the

variability in plasma F2‐IsoPs and IsoF levels can be explained by Hg body

burden, the MeHg‐Se interaction and cardiometabolic health outcomes. We

hypothesized that plasma levels of IsoPs and IsoFs were associated with Hg

exposure and that enhanced tissue Se status can protect against MeHg‐

related oxidative stress.

4.3 Subjects and Methods



population‐based IPY Health Survey, details of which are available elsewhere

(Egeland, Cao et al. 2011). In brief, a cross‐sectional survey was conducted in

the summer and fall 2007 and 2008 for 33 coastal communities and for three

non‐coastal communities representing all communities in Inuvialuit

Settlement Region (ISR, Northwest Territories), Nunavut and Nunatsiavut

Region (Northern Labrador). Trained interviewers and nurse staff collected

information on subjects’ dietary habits, physical activity, psychosocial

factors, medical history, blood pressure, anthropometric indices, fasting

lipids, and various clinical indices. Fasting blood samples were prepared and

stored at ‐80°C for future analyses. Territorial research licenses were

obtained and the Ethical Review Board of the McGill University Faculty of

Medicine approved the study. Informed consent was obtained from all

participants prior to enrollment.

76

4.3.2 Anthropometric, physiologic measures and definitions




(Egeland et al. 2011). A body mass index (BMI) of 25.0–29.9 kg/m2 was


obese.


Fasting serum total cholesterol, high‐density lipoprotein cholesterol (HDL‐C)

and triglycerides were determined using enzymatic colorimetric tests and

low‐density lipoprotein cholesterol (LDL‐C) was calculated by Nutrasource

Diagnostics, Guelph, Ont. (Life Laboratories–Gamma Dynacare). Serum high

sensitivity (hs)‐CRP concentration was determined using

immunoturbidimetric assay with SYNCHRON® High Sensitivity CRPH

reagent in conjunction with SYNCHRON® Systems CAL 5 Plus (Beckman

Coulter Inc., Fullerton, CA, USA) in the Centre for Indigenous Peoples'

Nutrition and Environment (CINE) at McGill University.

4.3.4 Assessment of Hg and Se exposure

Analyses for Hg and Se were performed at the Labouratoire de Toxicologie,

Institut national de santé publique, Quebéc, on whole blood. Briefly, blood

samples were diluted in a basic solution containing octylphenol ethoxylate

and ammonia, which was followed by inductively coupled plasma mass

spectrometry (ICP‐MS). Matrix matched calibration was performed using

blood from a non‐exposed individual. Toenails incorporate Se and may

reflect dietary intake over the past year (Longnecker, Stampfer et al. 1993).

Toenail Se analysis was performed at the laboratories of CINE and the

Department of Natural Resources Sciences at McGill University. Briefly,

samples were cleaned in acetone and distilled water, then digested in

77

concentrated nitric acid at 110°C for 4 h. Digests were dried at 160°C and

reconstituted in 2% nitric acid. Se concentrations were measured using a

Varian model ICP 820‐MS with a collision reactor interface.

4.3.5 Plasma analysis of F2‐IsoPs and IsoFs

Plasma samples were prepared from blood samples and stored at ‐80°C until

time of analysis. Purification, derivatization, and analysis of F2‐IsoPs and

IsoFs by stable isotope dilution gas chromatography/negative ion chemical

ionization mass spectrometry (GC/NICI/MS) were performed as previously

described (Kadiiska et al. 2005). An Agilent 5973 Mass Spectrometer coupled

to an Agilent 6890N Gas Chromatograph using a 15 mDB 1701 GC column

was utilized with an inlet temperature of 260°C. The helium carrier gas flow

rate was 2 ml/min. For sample injection, the GC oven was programmed to

run from 190 to 300°C at 20°C/min for 9 min. Selective ion GC/NICI/MS

monitoring was 569 m/z for F2‐IsoPs, 585 m/z for IsoFs, and 573 m/z for the

internal standard [2H4] 15‐F2t‐IsoP. Plasma values are expressed in pg/mL.

The precision of the assay is ± 6% and the accuracy is 96%.



reported as mean ± SD. The prevalence of preexisting medical conditions was




described as geometric mean ± 95% CI for log‐transformed data. The whole

blood Hg: Se ratio was calculated using the log‐transformed data for Hg and

Se. Correlation analysis was performed using Pearson’s correlation analysis

to assess the relationship between plasma concentrations of measures of

oxidative stress (F2‐IsoPs, IsoFs) and parameters of interest (Se, Hg, Hg:Se

ratio, and toenail Se), in addition to their relationship to other study

78

parameters including obesity measures and cardiovascular risk factors.

Baseline characteristics were compared between Se tertiles using ANOVA

and Boneferroni’s post‐hoc test was used when a significant group effect was

observed. All p‐values were two tailed, and p< 0.05 was considered

significant for all tests performed. General linear model was used to calculate

adjusted means of plasma IsoPs correcting for determined confounders

including age, WC, and sex. To estimate final predictors of the individual

biomarker variability and to examine the influence of confounding variables,

multivariate analysis with stepwise regression was used. Only variables that

were statistically different amongst Se tertiles were included in the model to

estimate their contribution to the variability in the IsoPs levels. Final models

were analyzed to verify if the assumptions of linear regression were met.

Colinearity between variables included in the final models was also assessed

to avoid two variables highly correlated in the same model. For the stepwise

regression, a α‐value of 0.05 was used to exclude variables that had little or

no influence on the biomarker under analysis. Sensitivity analyses were

performed in order to assess if exclusion of individuals with diabetes,

hypertension, stroke or any other cardiovascular disease was necessary for

IsoPs analyses. We also verified if exclusion of individuals taking medications

for hypertension changed the regression coefficients. All statistical analysis

was performed using SPSS version 13.0 software (SPSS Inc., Chicago, IL).

4.4 Results

4.4.1 Subjects characteristics


randomly from the IPY Inuit Health Survey from which specimens were

available for full analysis. Subjects with CRP ≥ 10 mg/L (n=60) were excluded

in the current study sample of 234 subjects. In addition, one outlier with F2‐

IsoPs levels with much lower levels than the lower limit of 95% CI was also

excluded decreasing the sample size to 233 subjects. The mean age of the

79

participants was 42.6 ±15.4 yrs. The study sample characteristics and mean

levels of F2‐IsoPs and IsoFs and their ratio IsoFs: F2‐IsoPs are presented in

Table 1, as previously reported to be with in normal range. Sixty‐six subjects

(28.3%) of our population subsample had plasma F2‐IsoPs levels greater ≥ 35

pg/ml, which is considered to be the upper value in the range for normal

human plasma levels (30 ±5 pg/ml) (Milne et al. 2007). Of the 233

participants, (56% women), mean BMI was 27.78 kg/m2; 33.5% were

overweight and 30.4% were obese. Based on medical histories of participants

the prevalence was 5.7% for diabetes, 28.6% for hypertension, 13.7%, for

dyslipidemia, 8.5% for cancer, 5.3% for episodes of heart attack and 3.1% for

stroke. Seventy percent of participants were current smokers with 41.8% of

smokers reporting >10 cigarettes/day and 65% of all participants reported

drinking alcohol in the past year. Of the 294 individuals available for

analyses, CRP (≥ 10 mg/L) was elevated for 20.8%. All data presented

represent 233 individuals with CRP < 10 mg/L.

4.4.2 IsoPs, Se and Hg

The assessment of the relationship between the various variables using

Pearson’s correlation showed that F2‐IsoPs were negatively associated with

both Se (r=‐.19, p=.005) and toenail Se (r=‐.146, p=.044) (Table 5). F2‐IsoPs

were not associated with either Hg (r=.057, p=.40) or Hg:Se (r=.11, p=.10).

IsoFs were negatively associated with Se (r=‐.14, p=.014) but were not

associated with toenail Se (r=‐.070, p=.34). IsoFs, however, were positively

associated with Hg (r=.23, p<.001) and with Hg:Se (r=.340, p<.001). The

IsoFs:F2‐IsoPs ratio was not associated with either Se (r=‐.070, p=.31) or

toenail Se (r=.014, p=.82); however, the IsoFs: F2‐IsoPs ratio was positively

associated with Hg (r=.27, p<.001) and with Hg:Se (r=.29, p<.001).

80

4.4.3 Comparison between Se tertiles

Participants were subgrouped into tertiles according to blood Se using the

tertile values of the entire group: <200, ≥200 & <340, and ≥340 µg/L (Table

6). In the remainder of all analyses, however, both Se and Hg were log‐

transformed as they were negatively skewed. The relationship between Se

and Hg was not linear across the Se tertiles. Blood Hg concentrations were

lower in the second Se tertile when compared to the first tertile. Conversely,

blood Se concentrations above 340 µg/L in the third tertile were associated

with elevated blood Hg concentrations comparable to those observed in the

first tertile. Regardless of the blood Hg:Se ratios, however, plasma

concentrations of both F2‐IsoPs and IsoFs decreased with increasing tertiles

of Se and the significant differences amongst the tertiles did not change after

correcting for confounders, including Hg.

4.4.4 Final variance predictors of plasma IsoPs

Multivariate analyses presented in Table 5 shows that the variance in F2‐

IsoPs concentrations was significantly predicted by blood Se (β=‐.138, p=.01)

and toenail Se (β=‐.096, p=.026), in addition to waist circumference (β=.003,

p<.001) and gender (β=.052, p=.038). F2‐IsoPs, however, were not associated

with Hg but were associated with the ratio of Hg:Se (β=.148, p=.021).

Variance in IsoF concentrations was predicted by both Se (β=‐.238, p=.016)

and Hg (β=.224, p<.001) and Hg:Se (β=.534, p<.001) in addition to waist

(β=.006, p<.001) and age (β=‐.004, p=.042). Variance of the IsoF: F2‐IsoPs

ratio was predicted by Hg (β=.118, p<.001), Hg:Se (β=.323, p<.001) and waist

(β=.003, p=.014) only.

81

4.5 Discussion

This study demonstrates that Inuit, who represent a unique population with

elevated blood concentrations of both Hg and Se, showed variations in

plasma IsoPs in relationship to both tissue Se and Hg. Despite elevated Hg

exposure compared to the Caucasian populations (Kingman, Albertini et al.

1998; Butler Walker, Houseman et al. 2006), the mean plasma concentration

of F2‐IsoPs for the population as a whole was within the normal range for

healthy humans (Milne, Sanchez et al. 2007) and just over quarter of the

sample had levels exceeding the upper range value (35 pg/mL). In

concordance with previous reports regarding Inuit of Nunavik (Belanger,

Dewailly et al. 2006), our findings confirm that Hg presence in the traditional

diet may not be of a major concern with respect to oxidative stress. Human

studies have indicated that Hg exposure is associated with an up‐regulation

in antioxidant activities (Ralston and Raymond 2010); however, the capacity

of the antioxidant defense system to prevent Hg‐induced oxidative stress

appears to vary according the type of exposure. High occupational Hg

exposure in mine workers in China has been associated with increased

oxidative DNA damage and higher serum malondialdehyde concentrations

(Chen, Qu et al. 2005) despite a concomitant increased expression of

selenoproteins such as GSH‐peroxidase (Px) and selenoprotein P (Chen, Qu et

al. 2005). In contrast, the Inuit from Nunavik and sports fisherman from

James Bay in Québec exposed to high dietary Hg exposure showed no adverse

effects of Hg on plasma oxidized LDL content as increased activity of blood

GSH redox cycle components, including GSH‐Px was observed (Chen, Qu et al.

2005; Belanger, Mirault et al. 2008). Despite the absence of Hg‐induced effect

on oxidized LDL in the latter two populations, a major rise in the ubiquinone‐

10 to ubiquinone‐10‐CoQ10 total redox ratio was observed that is suggestive

of oxidative stress. The biological significance of the above result is unclear in

view of the lack of an observed increase in whole body oxidative stress in the

Inuit population shown in the present study.

82

The observation of a positive association of Hg with F2‐IsoPs only when

tissue Hg was considered in the form of the Hg:Se ratio is novel since human

studies have not evaluated previously the interaction of tissue Hg and Se on

plasma IsoPs. A lack of relationship between Hg and plasma F2‐IsoPs and

other lipid peroxidation biomarkers has been previously reported in a group

of premenopausal women with low blood Hg concentrations {1.10 µg/L

(interquartile .58‐2.0)} but tissue Se was not evaluated (Pollack, Schisterman

et al. 2012). In further support of the interactive relationship of Hg and Se on

oxidative stress, we observed that Inuit at the two highest tertiles of plasma

of Se were most protected from oxidative stress independent of other

covariates that could influence Se status including tissue Hg, age,

cardiometabolic risk factors, smoking and blood lipids. In that regard, it is

noteworthy that the Inuit at the third tertile of blood Se was also associated

with the highest tissue Hg content, which signifies that high Se tissue content

protected against Hg‐mediated oxidative stress. To our knowledge, this is the

first human study to demonstrate protective effects of tissue Se status on

oxidative stress status in relation to MeHg exposure. The above findings

coincide with experimental animal studies showing that the co‐presence of

both Se and Hg within the same food matrix minimizes Hg‐induced oxidative

stress (Ralston and Raymond 2010). The modulation of Hg‐mediated

induction of F2‐IsoPs by tissue Se is also consistent with recent observations

that Se supplementation prevented the increase in urinary F2‐IsoPs induced

by dietary MeHg exposure in rats (Jin, Hidiroglou et al. 2012). Although Hg

and Se co‐accumulate in tissues, the mechanisms of their interaction are still

poorly understood. As oxidative stress is a major mechanism for Hg toxicity,

direct protection could be obtained increased levels of antioxidative

selenoproteins. Additionally, Se can provide protection by affecting the

kinetics and metabolism of Hg. In one small‐randomized trial, Seppanen et al.

(Seppanen, Soininen et al. 2004) found that Se supplementation in a Finnish

cohort with low Se intakes reduced hair Hg levels by one third over four

months. Li et al. (Li, Feng et al. 2008) showed that when five volunteers from

83

Wanshan Hg‐mining area were supplemented with Se‐enriched yeast for

three months, an increased urinary Hg excretion was observed.

Although both IsoFs and F2‐IsoPs were positively associated with the tissue

Hg:Se ratio in our study, only IsoFs and the IsoF: F2‐IsoPs ratio were

positively associated with tissue Hg content alone. The latter finding could be

explainable by observations that the ratio of IsoFs to F2‐IsoPs reflects

ambient oxygen concentrations within the environment in which lipid

peroxidation occurs (Fessel and Jackson Roberts 2005). Thus, as ambient

oxygen concentrations increase as a result of mitochondrial disruption, the

ratio is skewed toward IsoFs formation and away from F2‐IsoPs production

(Fessel and Jackson Roberts 2005). In that regard, impaired mitochondrial

oxygen consumption has been noted upon Hg exposure (Houston 2011),

which has been related to altered structural integrity of the mitochondrial

inner membrane (Lund, Miller et al. 1991). Addition of Hg to mitochondria

isolated from rat kidneys showed a dose‐related depolarization of the inner

mitochondrial membrane resulting in increased hydrogen peroxide

formation, GSH depletion and the formation of thiobarbituric acid reactive

substances (Lund, Miller et al. 1993). In vitro oxidative damage to

mitochondria from Hg exposure has also been observed in several other

tissues including myocardial tissues (Seppanen, Soininen et al. 2004).

Further population studies are needed to verify whether the causal

association of tissue Hg with a relative increase in IsoFs production can be

supported towards measurement of IsoFs as a sensitive biomarker of Hg

exposure.

Recent concerns have been raised that the high Hg content found in the

Northern traditional food may contribute to the development of

cardiometabolic disorders (Fontaine, Dewailly et al. 2008; Valera, Dewailly et

al. 2008). We have previously demonstrated that oxidative stress in the Inuit

is related to obesity‐induced inflammation and F2‐IsoPs and IsoFs were both

84

related significantly to systolic blood pressure (SBP) and CRP (Alkazemi,

Egeland et al. 2012). An interesting finding that emerged from the present

study was that the relationship between plasma IsoPs with either SBP or CRP

did not persist in the multivariate analyses when accounting for either Se or

Hg. The above results imply that the relationship of F2‐IsoPs and IsoFs with

either CRP or SBP in the Inuit population is Se and Hg co‐dependent and is

also suggestive of protective effects of Se on cardiometabolic disturbances

associated with IsoPs. We speculate that the latter observation is related to

the antioxidative and anti‐inflammatory properties of the different GSH‐Pxs

(Alissa, Bahijri et al. 2003) that could be induced by the high tissue content of

Se seen in this study. Further, Se may inhibit the activation of nuclear factor‐

κB by modulation of selenoprotein gene expression (Kretz‐Remy and Arrigo

2001) that, in turn, impedes the transactivation of genes that encode pro‐

inflammatory cytokines (Kretz‐Remy and Arrigo 2001). In addition, dietary

Se may inhibit the biotransformation of arachidonic acid towards the

formation of prostaglandins and thromboxanes that promote inflammation

(Hong, Li et al. 1989)

Our results differ from Valera et al. (Valera, Dewailly et al. 2009) who showed

a significant positive correlation between blood Hg concentrations and both

SBP and pulse pressure in Inuit adults from Nunavik, after considering the

confounding effects Se, n3‐PUFAs, and other co‐related variables. To

understand the differences in the two study findings, we constructed

sensitivity analysis by using the same confounders used in Valera et al.

(Valera, Dewailly et al. 2009) including n‐3 PUFAs but we did not find SBP to

be correlated with blood Hg, unless we included subjects presented with

acute inflammation (CRP ≥ 10 mg/L). Thus, adjustment for CRP may have

attenuated relationship between Hg and SBP. Also, our population had higher

mean blood Hg concentrations compared to the cohort in Nunavik [90.75

(78.95‐114.8) vs. 20.2 (46.6‐54.1) nmol/L] with comparable Se and n‐3

85

PUFAs concentrations, which should have allowed us to detect any adverse

effect of Hg in our cohort.

The Inuit represent a homogenous population in terms of intake of Se and Hg

primarily from traditional food sources; however, both Se and Hg

concentrations were skewed in our subsample indicating there are many

factors influencing their concentrations. Unlike the Nunavik studies

(Belanger, Dewailly et al. 2006; Valera, Dewailly et al. 2009), Hg and Se were

not highly correlated (r= .323, p<.001), which could indicate regional

differences of traditional food consumed in terms of wild animals such as

caribou as opposed to sea mammals. Factors shown to affect F2‐IsoPs such as

n‐3 PUFAs (Nalsen, Vessby et al. 2006) did not correlate with either of the

plasma IsoPs measured in this study and thus was not included in the further

analyses. While most Hg exposure among the Inuit is likely to be due to

intake of MeHg from traditional foods, we did not separately examine

different forms of Hg (inorganic, methyl), which may exert different effects

on oxidative stress biomarkers. In addition, many other dietary factors that

can further explain F2‐IsoPs variability were not included in this study,

including plasma antioxidants such as α‐tocopherol, ascorbic acid,

polyphenols and carotenoids. The nonlinear associations between Hg and Se

across Se tertiles could be attributable to residual confounding resulting

from not including other important constituents of traditional food. Thus,

residual confounding from dietary consumption of various antioxidants

components available in the survey season is also likely. Therefore, the

plasma IsoPs may be more strongly correlated with Hg than we observed in

our sample.

To our knowledge, this is the first human study to report a relationship

between tissue Hg and plasma F2‐IsoPs concentrations as new interactions

were demonstrated between the speciated forms of IsoPs with respect to

tissue Hg. Significantly, oxidative stress associated with Hg exposure was

86

highly modulated by tissue Se status, which emphasizes the importance of

concurrent tissue Se measurement for the assessment of Hg‐mediated

oxidative stress. We have also shown that measurement of Hg and Se can

provided new insights with respect to quantifying risk of oxidative stress as a

mechanistic link to the progression to CVD. More studies are needed to

confirm our observations using larger sample sizes in order to perform more

complex analyses dealing with Hg and Se interactions.

87

Table 4.1 Characteristics of study population (n=233) Mean (SEM) or 95%CI * Min‐Max Se (µg/L) 356.12 (16.14),

302.69 (281.84‐324.41)

140‐1400

Toenail‐Se (µg/g) .99 (.02) .39‐3.51 Hg (µg/L) 28.8 (1.96),

18.15 (15.79‐20.82)

.39‐200

Hg: Sea .50 (.13) ‐.19‐1.02 F2‐IsoPs (pg/ml)a 27.35(25.73‐28.64) 6.03‐81.28 IsoFs (pg/ml)a 20.81 (18.86‐22.96) .50‐173.78 IsoFs:F2‐IsoPs .92 (.89‐.95) .24‐1.45 Body Mass Index 27.78 (.38) Waist Circumference (CM) 92.43 (.97)

Body Fat % 29.63 (.71) SBP 117.36 (1.13) Diastolic Blood Pressure 76.54 (.73) Total‐Cholesterol 5.02 (.08) LDL‐C 2.92 (.07) HDL‐C 1.47 (.03) Triglycerides 1.44 (0.72) Fasting Glucose 4.97 (.04) CRPb 1.66 (.02) a. skewed data were log‐transformed to be able to use parametric tests, b. data with values less than one were normalized by log‐transforming after the addition of one. * for both a and b geometric means and 95th CIs are presented.

88

Table 4.2 Bivariate correlation between plasma IsoPs, Se and Hg Pearson correlations F2‐IsoPs IsoFs IsoFs:F2 Se r ‐.186** ‐.164* p .005 .014

Toenail Se r ‐.146* p .044

Hg r .288** .269** p <.001 <.001

Hg:Se r .340** .294** p <.001 <.001

CRP r .138* .158* p .037 .016

FG r .130 .115 p .094 .082 *. Correlation is significant at the 0.05 level (2‐tailed). **. Correlation is significant at the 0.01 level (2‐tailed).

89

Table 4.3 Characteristics of 223 subjects according to tertiles of blood Se concentrations T1<220

Mean(95%CI) T2≥200 & <340 Mean(95%CI)

T3 ≥340 Mean (95%CI)

P‐trend overall

n 72 77 78 Selenium (µg/l)

185.97 (181.46‐190.48)a

258.18 (250.63‐265.73)b*

609.87 (550.14‐669.60)b**,c**

<.01**

Toenail‐Se (µg/g)

.97 (.86‐1.08)

1.00 (.93‐1.07)

.99 (.94‐1.05)

0.83

Hg (µg/l)

29.32 (21.87‐36.72)

22.44 (15.92‐28.96)a

35.47(27.62)b* .026*

Hg:Se 0.53 (.47‐.59)a

0.45 (.41‐.50)b

0.52 (.49‐.54)

.037*

F2‐IsoPs (pg/ml)

29.64 (26.75‐32.85)a

27.54 (25.28‐30.22)

24.11 (22.07‐26.34)b**

<.01**

IsoFs (pg/ml)

28.98 (24.46‐34.33)a

17.53 (14.90‐20.63)b**

18.14 (15.24‐21.60)b**

<.01**

IsoFs:F2‐IsoPs 1.00 (.95‐1.05)a

.86 (.82‐.91)b**

.91 (.86‐.97) b#

<.01**

Age 39.28 (35.90‐42.66)a

38.25 (35.11‐41.38)a

49.92 (46.44‐53.41)b**

<.01**

BMI 28.37 (26.91‐29.84)

26.90 (25.65‐28.15)

28.07 (26.82‐29.33)

0.25

WC 94.22 (90.65‐97.80)

90.07 (86.93‐93.21)

93.33 (89.84‐96.81)

0.20

BF% 30.94 (28.54‐33.34)

27.87 (25.38‐30.37)

30.04 (27.54‐32.55)

0.19

SBP 116.25 (111.90‐121.13)

115.86 (112.52‐119.21)

119.52 (115.44‐123.59)

0.35

DBP 76.52 (73.69‐79.35)

76.85 (74.54‐79.17)

76.29 (73.74‐78.85)

0.94

T‐Chol 4.73 (4.48‐4.98)a

4.94 (4.71‐5.17)

5.36 (5.07‐5.66)b**

<.01**

LDL‐C 2.63 (2.42‐2.85)a

2.86 (2.66‐3.07)

3.18 (2.92‐3.43)b**

<.01**

HDL‐C 1.44 (1.34‐1.54)

1.47 (1.36‐1.58)

1.53 (2.92‐3.43)

0.48

Triglycerides 1.44 (1.18‐1.70)

1.34 (1.18‐1.50)

1.50 (1.18‐1.81)

0.68

Fasting Glucose

4.88 (4.74‐5.03)

4.90 (4.73‐5.05)

5.12 (4.96‐5.27)

.049*

CRP 1.83 (1.46‐2.26)

1.42 (1.14‐1.74)

1.77 (1.43‐2.18)

0.19

1‐way ANOVA with Bonferroni adjusted multiple comparisons where by the mean difference is significant at .05 level. *. significant at the 0.05 level, **. significant at the 0.01 level, # trend at p>0.05 and p≤0.07.

90

Table 4.4 GLM adjusted selenium tertiles for plasma IsoPs concentrations

T1 s‐Se <220 Mean (95%CI)

T2 s‐Se 200 & <340 Mean (95%CI)

T3 s‐Se 340 Mean (95%CI)

F‐test Effect of tertiles

Adjusted for: age, gender, WC, current smoking

n 60 70 72 F (p‐value)

F2‐IsoPs (pg/ml)

29.11a (26.30‐32.21)

28.31 (25.76‐31.19)

24.21b* (21.93‐26.73)

3.538 (.035)

IsoFs (pg/ml)

27.54a (22.86‐33.27)

19.14b* (16.07‐22.80)

17.18b** (14.35‐20.56)

6.890 (.001)

IsoFs:F2‐IsoPs .99a (.94‐1.05)

.88b* (.83‐ .94)

.90 (.84‐.95)

4.416 (.013)

Adjusted for age, gender, WC, current smoking, LDL, FG, and Mercury

n 59 68 70 F (p‐value)

F2‐IsoPs (pg/ml)

29.04a (26.24‐32.21)

28.91 (26.24‐31.77)

24.10b* (21.82‐26.55)

4.12 (.018)

IsoFs (pg/ml)

26.98a (22.59‐32.14)

19.82b* (16.83‐23.39)

17.14b** (14.45‐20.28)

6.678 (.002)

IsoFs:F2‐IsoPs .99a (.93‐1.04)

.89 b* (.84‐.94)

.90 (.84‐.95)

3.65 (.028)

1‐way ANOVA with Bonferroni adjusted multiple comparisons where by the mean difference is significant at .05 level. *. significant at the 0.05 level, **. significant at the 0.01 leve.

91

Table 4.5a‐d Multivariate associations showing the regression coefficient (β) of plasma IsoPs concentrations Stepwise Regression

F2‐IsoPs

β (SE) IsoFs

β (SE) IsoF:F2‐IsoPs β (SE)

Model (a) Constant 1.449 (.16) 1.264 (.28) .664 (.10)

Independent Variables:

Se ‐.138(.05)* ‐.238 (.02)* NS

WC .003(.001)** .007 (.002)** .003(.001)**

Sex .052(.63)**

R2 .104 .121 .038

F 7.13** 12.75** 7.31**

Model (b) Constant 1.16 (.09) .55 (.15) .528 (.11)


Mercury NS .222(.050)** .118(.03)**

WC .003 (.001)** .007 (.002)** .003 (.03)*

Sex .054 (.03)*

Age ‐.002 (.001)** ‐.003 (.002)*

R2 .101 .182 .102

F 6.90 ** 13.73 ** 10.56**

Model (c) Constant 1.25(.10) .68 (.15) .64 (.111)


Toenail Se ‐.096 (.04)* NS NS

WC .003(.001)** .007(.002)** .003 (.001)*

Gender .059 (.03)*

Age ‐.002(.001)*

R2 .13 .09 .04

F 5.99 ** 17.09** 6.53 *

Model (d) Constant 1.12 (.09) .46 (.15) .52(.11)


Hg:Se .15 (.07)* .56 (.11)** .32(.08)**

WC .003(.001)** .006(.002)** .003(.001)*

sex .053 (.07)*

Age ‐.003(.001)**

R2 .12 .19 .12

F 6.51** 22.22** 12.14**

In all models independent variables included: WC, sex, age, smoking, alcohol, DM or fasting glucose HTN or SBP, Lipidemia or T‐Chol on a continuos scale.

92

BRIDGE 3

We established the association of Se in the protection against Hg‐induced

oxidative stress in the Inuit. However, through exploratory analysis we

determined that n‐3 PUFAs did not appear to modulate the Se‐mediated

protective effects. The high n‐3 PUFA diet of the Inuit, however, could

engender increased biosynthesis of F3‐IsoPs. These latter IsoPs are oxidized

products of EPA that could exert cardioprotective effects since these

products are implicated in animal studies towards protection against

inflammation associated with F2‐IsoPs. In the third manuscript our aim was

to elucidate further the role of n3‐PUFAs in protection against inflammation

and oxidative stress as mediated by F3‐IsoPs.

93


Novel eicosapentaenoic acid‐derived F3‐isoprostanes as biomarkers of lipid

peroxidation in the Canadian Inuit population

Dalal Alkazemi1,2, Grace Egeland1,2, L. Jackson Roberts II3, H.M. Chan4, Stan1

Kubow

Authors affiliations: 1School of Dietetics and Human Nutrition, McGill University, 21,111

Lakeshore Road, Ste‐Anne‐de‐Bellevue, QC H9X3V9, Canada 2Centre for Indigenous Peoples’ Nutrition and Environment (CINE), McGill

University, 21,111 Lakeshore Road, Ste‐Anne‐de‐Bellevue, QC H9X 3V9 3Department of Pharmacology and Medicine, 522 RRB, Vanderbilt University,

Nashville, YN 37232‐6602, USA 4Center for Advanced Research in Environmental Genomics, University of

Ottawa, 30 Marie Curie, Ottawa, ON K1N 6N5, Canada



Nutrition,

McGill University, 21,111 Lakeshore, Ste‐Anne‐de‐Bellevue, Quebec H9X3V9,

Canada.

Email: [email protected]

94

5.1 Abstract:

Objectives: The aim of this study was to evaluate plasma F3‐isoprostanes

(F3‐IsoPs) as a biomarker of oxidative stress in Inuit adults relative to

previously measured the gold standard indicator of in vivo oxidative stress,

F2‐isoprostanes (F2‐isoPs), an important risk factor for cardiovascular

disease. We examined the relationship between concentrations of plasma F2‐

IsoPs, F3‐IsoPs and dietary constituents in the Inuit diet, namely n‐3 and n‐6

PUFAs, selenium (Se), mercury (Hg), polychlorinated biphenyls (PCBs), and

various cardiometabolic risk factors. Materials and Methods: Cross‐

sectional study using a subsample of a population‐based survey conducted in

the summer and fall 2007 and 2008 in 36 Canadian Arctic Inuit Communities.

Subjects included a random subset (n=233) of the total study population

(n=2595) with a mean age 42.56 ± 15.39 yr and body mass index 27.78 ±

5.65 kg/m2. Mean plasma F2‐IsoPs was 27.31 (95%CI 25.73‐28.64, µg/mL)

and F3‐IsoPs 3.71 (3.23‐4.23, µg/mL). Plasma levels of and F3‐IsoPs, products

of lipid peroxidation, were determined by gas chromatography/negative ion

chemical ionization/mass spectrometry and their relationships to n‐3 and n‐

6 PUFAs, whole blood Se and Hg, toenail Se, and PCBs were assessed by

multivariate analyses. Results: Plasma F3‐IsoPs was positively correlated F2‐

IsoPs (r=.191, p=.01), blood Se (r=.276, p<.001), blood Hg (r=.358, p<.001),

and was inversely correlated with toenail Se (r=‐.215, p=.008). In addition,

F3‐IsoPs positively correlated with all of congeners of PCBs measured

(p<.001); however, this was not associated with any metabolic measures

including blood lipids, CRP, blood pressure, or fasting glucose. F3‐IsoPs

concentrations was significantly predicted by Hg (β=.193, p=.004), smoking

(β= ‐.146, p=.026), and C20:5n‐3 (β=.067, p=.005) (R2=.283). When toenail Se

was included in the model, the F3‐IsoPs relationship to n‐3 PUFAs did not

persist, and the final correlates of F3‐IsoPs in the model were Hg (β=1.034,

p<.001), Se (β=‐.857, p<.001), PCB no.153 (β=‐.147, p=.028), and waist

circumference (β=.004, p=.029) (R2=.666). Conclusions: Plasma F3‐IsoPs is a

quantifiable oxidative stress biomarker in the Inuit because of their diet rich

95

in C20:5n‐3. Our findings indicate that environmental contaminants are

associated with plasma F3‐IsoPs and that F3‐IsoPs provide a mechanistic

explanation to the Inuit protection against oxidative stress.

5.2 Introduction:

Numerous epidemiological and interventional studies have shown that

regular consumption of fish containing a rich content of the n‐3

polyunsaturated fatty acids (PUFA) eicosapentaenoic (EPA, C20:5n3) and

docosahexaenoic (DHA, C22:6n3) reduces cardiovascular risk and mortality

(He, Song et al. 2004; Konig, Bouzan et al. 2005). Several traditional and non‐

conventional cardiovascular risk factors are influenced favorably by n‐3

PUFAs intake including triglyceride levels (Harris, Lu et al. 1997),

arrhythmias and thrombosis (Cheng and Santoni 2008), pro‐inflammatory

eicosanoids, and endothelial function (Lopez‐Garcia, Schulze et al. 2004). It is

well established that the 3‐, 5‐series eicosanoids derived from EPA are less

biologically active or inactive as compared with the 2‐, 4‐series eicosanoids

generated from the n‐6 PUFA arachidonic acid (AA) and thus are considered

as less inflammatory (Davis, Gao et al. 2006). Additionally, there is indication

that as EPA is highly unsaturated, it can readily undergo oxidation to several

bioactive compounds that may exert cardioprotective properties (Sethi 2002;

Hong, Gronert et al. 2003). In that regard, F3‐Isoprostanes (F3‐IsoPs), formed

from free radical‐induced peroxidation of EPA, may exert anti‐atherogenic

and anti‐inflammatory effects by acting as competitive inhibitors of

cyclooxygenase, reducing the production of the pro‐inflammatory 2‐series

prostaglandins and thromboxane generated from AA (Gao, Yin et al. 2006).

Also, oxidized EPA, but not native EPA, significantly inhibits human

neutrophil and monocyte adhesion to endothelial cells, which are linked to

the development of atherosclerosis and inflammation (Sethi 2002). Dietary

EPA may also diminish the formation of proinflammatory F2‐isoprostane (F2‐

IsoPs) derived from n‐6 PUFAs by channeling the free radical pathway away

from F2‐IsoPs. Mice receiving EPA supplementation showed markedly

96

increased heart tissue levels of F3‐IsoPs together with reduced levels of the

proinflammatory F2‐IsoPs by up to 64% (Davis, Gao et al. 2006); however, no

human studies to date have demonstrated a relationship between F2‐IsoPs

and F3‐IsoPs production.

The Inuit diet is high in n3 PUFAs, which is thought to contribute to the

historically low prevalence of chronic diseases in this population (Dewailly,

Blanchet et al. 2001). In recent years, however, a shift away from the

traditional diet toward an increasingly Western‐style based diet has been

observed that could increase the burden of chronic diseases including

obesity, cardiovascular disease, and type 2 diabetes (Bjerregaard, Jorgensen

et al. 2007; Chateau‐Degat, Dewailly et al. 2011). The health benefits of the

traditional Inuit diet are complicated by presence of pro‐oxidant

contaminants such as methylmercury and polychlorinated biphenyls (PCBs),

which could be partly counteracted by the high n‐3 PUFAs and selenium (Se)

intake of the Inuit (Grotto, Valentini et al. 2011; Park and Mozaffarian 2010).

Also, an adaptive response to oxidative stress induced by PCBs and MeHg has

been suggested by higher blood levels of Se, glutathione peroxidase,

glutathione reductase and ubiquinol in Inuit from Nunavik (Belanger,

Dewailly et al. 2006; Belanger, Mirault et al. 2008). Further, we demonstrated

that Inuit were protected against oxidative stress as assessed by plasma F2‐

IsoPs (Alkazemi, Egeland et al. 2012). Moreover, plasma levels of the

isoprostane species, isofurans, were inversely related to the presence of

selenium, indicating protective effects of Se intake from traditional Inuit

foods against mercury (Hg)‐induced oxidative stress (Alkazemi, Egeland et al.

2012).

Previous studies have highlighted a generational gap with traditional food

use, i.e., the elderly consume more traditional foods compared to the younger

generation (Zhou, Kubow et al. 2011). With the current trend of decreased

consumption of traditional foods in the younger generation, it is important to

97

examine the impact of such dietary change as assessed via fatty acid profiles

and oxidative stress parameters. We hypothesized that a decline in

traditional food intake observed between age groups (young Inuit <40 vs.

over 40 yrs) would be exhibited by increased levels of oxidative stress and an

altered red blood cell membrane fatty acid composition that reflects the shift

in the specific fatty acid composition of the diet. The objectives of the

present study were to: (a) determine F3‐IsoPs level and its relationship to F2‐

IsoPs and its relationship to metabolic and dietary parameters; (b) to

investigate the association between red blood cell fatty acid composition and

F2‐IsoPs and F3‐IsoPs; (c) to compare the two age categories (younger than

40 yrs vs. older) metabolic and lipid profile in relation to plasma

isoprostanes.

5.3 Subjects and Methods



population‐based IPY Health Survey details of which are available elsewhere

(Egeland, Cao et al. 2011). In brief, a cross‐sectional survey was conducted in

the summer and fall 2007 and 2008 for 33 coastal communities and for three

non‐coastal communities representing all communities in Inuvialuit

Settlement Region (ISR, Northwest Territories), Nunavut and Nunatsiavut

Region (Northern Labrador). Trained interviewers and nurse staff collected

information on subjects’ dietary habits, physical activity, psychosocial

factors, medical history, blood pressure, anthropometric indices, fasting

lipids, and various clinical indices. Fasting blood samples were prepared and

stored at ‐ 80°C for future analyses. Territorial research licenses were

obtained and the Ethical Review Board of the McGill University Faculty of

Medicine approved the study. Informed consent was obtained from all

participants prior to enrollment.

98

5.3.2 Anthropometric, physiologic measures and definitions




(Egeland, Cao et al. 2011). A body mass index (BMI) of 25.0–29.9 kg/m2 was


obese.


Fasting serum total cholesterol (T‐chol), high‐density lipoprotein (HDL)

cholesterol and triglycerides were determined using enzymatic colorimetric

tests and low‐density lipoprotein cholesterol (LDL‐C) was calculated by

Nutrasource Diagnostics, Guelph, Ont. (Life Laboratories–Gamma Dynacare.

Serum high sensitivity (hs)‐CRP concentration was determined using

immunoturbidimetric assay with SYNCHRON® High Sensitivity CRPH

reagent in conjunction with SYNCHRON® Systems CAL 5 Plus (Beckman

Coulter Inc., Fullerton, CA, USA) in CINE at McGill University.

5.3.4 Fatty acid analysis

Blood concentrations of fatty acids were determined in erythrocyte

membrane by gas‐liquid chromatography (Lipid Analytical Laboratories Inc,

Guelph, Canada). Fatty acid composition of red blood cells was determined

based on previous studies (Dewailly, et al., 2001)(Stark & Holub, 2004).

Lipids were extracted from the blood samples according to the method of

Folch et al. (Folch, et al., 1957). The fatty acid methyl esters were prepared by

the method of Morrison and Smith (Morrison & Smith, 1964) and were

analyzed on a Varian 2400 gas‐liquid chromatograph (Palo Alto, CA) with a

60‐metre DB‐23 capillary column (0.32 mm internal diameter).

99

5.3.5 Analysis of PCBs

Analysis of PCBs was performed at the laboratory of the Centre de

Toxicologie du Quebec, Canada, which is accredited by the Canadian

Association for Environmental Analytical Laboratories. Sixteen PCB

congeners [International Union for Pure and Applied chemistry; IUPAC) no.

28, 52, 99, 101, 105, 118, 128, 138, 153, 156, 163, 170, 180, 183, 187, and

Aroclor 1260 were measured in the purified extracts with an HP 5890 high

resolution GC equipped with dual capillary columns (HP Ultra I and Ultra II)

and dual Ni‐63 electron capture detectors (Hewlett‐Packard. Palo Alto, CA,

USA). PCBs were reported on a standardized lipid‐adjusted basis when

relying on blood specimens for quantifying concentration of lipophilic

environmental contaminants [Phillips, 1989]. Estimates of total serum lipids

were calculated by the following formula: Total lipids = 0.9+1.3*(cholesterol

+triglycerides) (Rylander, Nilsson‐Ehle et al. 2006). Lipids were measured

using the standard enzymatic procedures explained above.

5.3.6 Plasma analysis of isoprostanes

Plasma samples were prepared from blood samples and stored at ‐80°C until

time of analysis. Purification, derivatization, and analysis of F2‐IsoPs and F3‐

IsoPs by stable isotope dilution gas chromatography/negative ion chemical

ionization mass spectrometry (GC/NICI/MS) were performed as previously

described (Milne, Yin et al. 2007). An Agilent 5973 Mass Spectrometer

coupled to an Agilent 6890N Gas Chromatograph using a 15 mDB 1701 GC

column was utilized with an inlet temperature of 260°C. The helium carrier

gas flow rate was 2 ml/min. For sample injection, the GC oven was

programmed to run from 190 to 300°C at 20°C/min for 9 min. Selective ion

GC/NICI/MS monitoring was 569 m/z for F2‐IsoPs, 567 m/z for F3‐IsoPs, and

573 m/z for the internal standard [2H4] 15‐F2t‐IsoP. Values are expressed in

picograms per milliliter of plasma. The precision of the assay is ± 6% and the

accuracy is 96%.

100



reported as mean ± SEM. Prevalence’s of preexisting medical conditions were




described as geometric mean ± 95% CI for log‐transformed data. Fatty acid

content in erythrocyte membrane was expressed as a percentage of total

fatty acids. n‐3 PUFAs, n‐6 PUFAs were calculated by summing the

concentrations of individual fatty acids from the same class if they were

detectable and the ratio between the two parameters was calculated. Fatty

acid data are presented as mean and 95% CI. The plasma levels of F2‐IsoPs

and F3‐IsoPs were compared using Student’s t test between the two age

groups (< and ≥ 40 yrs). Plasma F2‐IsoPs and F3‐IsoPs, and the ratio of F2:F3,

and fatty acid concentrations, as a percent of total fatty acids were evaluated

by two age categories using age 40 yrs as a cutoff point. Correlation analysis

was performed using Pearson’s correlation analysis to assess the relationship

between plasma concentrations of measures of oxidative stress (F2‐IsoPs and

F3‐IsoPs) and other study parameters including obesity measures and

cardiovascular risk factors. Partial correlation analysis was performed

accounting for age, sex and WC; and for total lipids, mercury and selenium.

All p‐values were two tailed, and p< 0.05 was considered significant for all

tests performed. To estimate final predictors of the individual biomarkers

variability and examine the influence of confounding variables, multivariate

analysis with stepwise regression was used. For the stepwise regression, a α‐

value of 0.05 was used to exclude variables that had little or no influence on

the biomarker under analysis. All statistical analysis was performed using

SPSS version 13.0 software (SPSS Inc., Chicago, IL).

101

5.4 Results


randomly from the IPY Inuit Health Survey from which specimens were

available for full analysis. Subjects with CRP ≥ 10 mg/L (n=60) were excluded

in the current study sample of 234 subjects. In addition, one outlier with f2‐

IsoPs levels with much lower levels than the lower limit of 95% CI was also

excluded decreasing the sample size to 233 subjects. The mean age of the

participants was 42.6 ±15.4 yrs. The study sample characteristics and mean

levels of F2‐IsoPs and F3‐IsoPs and their ratio F2:F3‐IsoPs are presented in

Table 1, as previously reported to be within normal range (Alkazemi et al.,

2012). Sixty‐six subjects (28.3%) of our population subsample had plasma

F2‐IsoPs levels greater ≥ 35 pg/ml, which is considered to be the upper value

in the range for normal human plasma levels (30 ±5 pg/ml) [31]. Of the 233

participants, (56% women), mean BMI was 27.78 kg/m2; 33.5% were

overweight and 30.4% were obese. Based on medical histories of participants

the prevalence was 5.7% for diabetes, 28.6% for hypertension, 13.7%, for

dyslipidemia, 8.5% for cancer, 5.3% for episodes of heart attack and 3.1% for

stroke. Seventy percent of participants were current smokers with 41.8% of

smokers reporting >10 cigarettes/day and 65% of all participants reported

drinking alcohol in the past year. Of the 294 individuals available for

analyses, CRP was elevated (≥10 mg/L) for 20.8%. All data presented

represent 233 individuals with CRP < 10 mg/L.

5.4.1 F3‐IsoPs relationship to F2‐IsoPs, Se, and contaminants

The assessment of the relationship between the various variables using

Pearson’s correlations showed that F3‐IsoPs was positively correlated F2‐

IsoPs (r=.191, p=.01). Furthermore, F3‐IsoPs, similarly to F2‐IsoPs [Se (r=‐

.186, p=.005)], was inversely correlated with toenail Se (r=‐.215, p=.008).

Further, F3‐IsoPs remained inversely correlated to toenail Se even after

partial correction for age, gender, and waist (r=.225, p<.001), a relationship

to long‐term antioxidant status similarly observed with F2‐IsoPs (r=.164,

102

p=.047). In terms of blood Se, F3‐IsoPs relationship was the opposite to that

observed with F2‐IsoPs [toenail Se (r=‐.146, p=.044)], as it was positively

correlated with Se (r=.276, p<.001). In terms of contaminants, F3‐IsoPs was

positively correlated with Hg (r=.358, p<.001), and with Hg:Se (r=.305,

p<.001). The ratio of F2‐IsoPs:F3‐IsoPs was highly correlated to F3‐IsoPs(r= ‐

.809, p<.001) which indicates that it was modulated by F3‐IsoPs. F3‐IsoPs was

positively associated with age (r=.251, p=.001); however, not associated with

any metabolic measures including blood lipids, CRP, blood pressure, or

fasting glucose.

5.4.2 IsoPs and PCBs

Concentrations of the 16 PCBs measured in plasma samples from the 233

subjects are shown in Table 2. Our population levels of PCBs exposure were

found less than half of the levels reported in Nunavik (Belanger, Dewailly et

al. 2006). This could be due to regional difference indicating less reliance on

traditional foods and variations in the proportions of in‐land and coastal

communities covered in both surveys (Zhou, Kubow et al. 2011). The most

prominent congeners were Aroclor and no.153. F3‐IsoPs positively correlated

with all of congeners of PCBs measured, however when correlations were

adjusted for age, gender and weight circumference (WC) or % body fat (BF),

they did not persist (data not shown).

5.4.3 IsoPs and relative erythrocyte fatty acid concentrations

With regards to the fatty acids, F3‐IsoPs was positively associated with total

n‐3 PUFAs (r=.269, p<.001), and positively to the ratio of n‐3:n‐6 (r=.262,

p<.001) whereas F2‐IsoPs correlated negatively with n‐3 PUFAs (r=‐.138,

p=.036), and with n‐3:n‐6 (r=‐.184, p=.005). F3‐IsoPs was negatively

correlated with AA: EPA (r=‐.329, p<.001) and the ratio F2‐IsoPs:F3‐IsoPs was

positively correlated with AA:EPA (r=.357, p<.001); however, F2‐IsoPs was

not associated with AA:EPA. Further correlational analysis performed with

individuals fatty acids revealed that F3‐IsoPs was correlated positively with

103

C20:5n‐3 (r=.355, p<.001), C22:6n‐3 (r=.218, p=.003); and negatively to

C20:0 (r=‐.232, p=.002); and C20:2n‐6 (r=‐.246, p=.001). F2‐IsoPs were

positively correlated to C16:0 (r=.132, p=.046) and C16:1 (r=.150, p=.023)

whereas F2‐IsoPs: F3‐IsoPs correlated positively with C16:1 (r=.148, p=.016),

trans C16:1 (r=.197, p=.008), C18:0 (r=.212, p=.004), C20:0 (r=.229, p=.002),

C20:2n‐6 (r=.198, p=.007) and correlated negatively with C20:5n‐3 (r=‐.365,

p<.001).

5.4.4 Comparison between n‐3 tertiles

Participants were subgrouped into tertiles according to the relative total n‐3

PUFAs concentrations of the entire group: <4.36%; ≥4.36% and ≤6.90%; and

≥6.90% (Table 3) as the indicator of traditional food intake. For the

remainder of the analysis the congener no.153 was used as a proxy measure

for PCBs exposure, as it is a major component of environmental PCBs

(Twaroski TP et al. 2001) and among our study population it was the highest

in concentration (along with Aroclor); and found equally in all subjects tested

positive. In addition, PCB 153 is a higher chlorinated biphenyl that induces

cytochrome P450 and resists metabolism (Denomme et al. 1983).

Contaminant levels in terms of Hg and the Hg:Se ratio were significantly

higher in the third and second tertiles in comparison to the first tertile

whereas levels of PCB no. 153 was significantly (p<.001) higher in the third

tertile in comparison of the two lower tertiles that were not different

between each other. Even though Se level increased with increasing n‐3

tertiles, the Hg:Se ratio also increased. F3‐IsoPs concentrations increased

with increasing tertiles whereas F2‐IsoPs: F3‐IsoPs was lowest at the highest

n‐3 tertile. In addition, monounsaturated fatty acids (MUFA), saturated fatty

acid (SFA) and trans fatty acids (TFA) were all lowest levels at the two higher

n‐3 tertiles.

104

5.4.5 Comparison between age categories

Student t‐test assessing the difference between the two age categories (< 40

and ≥ 40 yrs) shows that the older Inuit had lower levels of F2‐IsoPs and

lower F2‐IsoPs: F3‐IsoPs, and higher Hg, PCBs and a deregulated

cardiometabolic profile evident by higher systolic blood pressure (SBP),

fasting glucose (FG), T‐cholesterol, and LDL‐C (Table 4). In addition, older

Inuit have higher F3‐IsoPs and n‐3 PUFAs and total PUFA intake. The younger

Inuit show higher F2‐IsoPs concentrations and significantly higher AA:EPA

and lower n‐3:n‐6.

5.4.6 Final predictors of F3‐IsoPs

Multivariate analyses presented in Table 9 shows that the variance of F3‐

IsoPs concentrations was significantly predicted by Hg (β=.193, p=.004),

smoking (β= ‐.146, p=.026), and n‐3 (β=.019, p=.048), C20:5n‐3 (β=.067,

p=.005), n‐6:n‐3 (β=‐.10, p=.039) and AA:EPA (β=‐.009, p=.035). F2:F3‐IsoPs

final predictors were Hg, (β= ‐.812, p=.015), Se (β= ‐1.95, p=.015), AA:EPA

(β= .056, p=.006), C20:5n‐3 (β=‐.365, p=.002) and smoking (β=.745, p=.025).

When replaced toenail Se with blood Se in the multivariate model, the EPA or

n‐3 PUFAs relationship to F3‐IsoPs did not persist; however, blood Hg

(β=1.034, p<.001) and Se (β=‐.857, p<.001), PCB (no.153) (β=‐.147, p=.028)

and waist (β=.004, p=.029) became major correlates of F3‐IsoPs

concentrations (R2= .666).

5.5 Discussion

This study is the first to report the presence of F3‐IsoPs in human plasma.

The F3‐IsoPs species are a novel class of F2‐isoP‐like compounds formed from

free radical‐induced peroxidation of EPA, the most abundant n‐3 PUFAs

present in fish oils and the fatty tissues of marine mammals. F3‐IsoPs have

previously only been observed in human urine or in the plasma of animals

supplemented with high levels of fish oils (Song, Paschos et al. 2009). Recent

human trials failed to observe F3‐IsoPs in plasma after supplementation with

105

n‐3 PUFAs (Mas, Woodman et al. 2010), which could be related to very low

baseline levels of EPA in animals and human tissues that may result in

plasma F3‐IsoPs concentrations to be below the limit of detection (30 pg/g of

tissue). The detection of F3‐IsoPs in the plasma of Inuit could be related to

their relatively high intakes of n‐3 PUFAs as compared with other

populations due to high consumption of marine mammals and fish (Muckle,

Ayotte et al. 2001). In that respect, the maximum daily intake of EPA and

DHA estimated from 24 h dietary recalls in Inuit populations has been

reported to be as high as 34.8 g and food frequency questionnaires have

shown mean annual daily intakes of EPA+DHA of 2.11 g (Dewailly, Blanchet

et al. 2001). In the present study, the Inuit population had much higher

erythrocyte n‐3 PUFAs concentrations (Table 1) than previously reported in

whites (Leeson et al. 2002); however, this was at a lower range compared to

Inuit from Nunavik {(mean ± SE: 6.05 ± .22, .18‐22.15) vs. (9.71± .23, 2.41‐

29.51) (Dewailly, Blanchet et al. 2001). Plasma F3‐IsoPs were detected in

80% of our population sample, and were directly related to erythrocyte EPA

concentrations, which accounted for 28% of total n‐3 (wt%). Thus, the high

cellular content of substrate likely allowed plasma F3‐IsoPs to be detectable.

Because n‐3 PUFAs are highly unsaturated, there has been concern that their

consumption at high levels might increase oxidative stress (McGrath,

Brennan et al. 1996). There are indications, however, that increased

formation of F3‐IsoPs from dietary n‐3 PUFAs may channel free radical‐

mediated oxidation away from n‐6 PUFA thereby interfering with the

formation of the highly inflammatory F2‐IsoPs peroxidation product (Gao, Yin

et al. 2006; Gao, Wang et al. 2007; Wada, DeLong et al. 2007).

In support of this concept, Inuit with highest tertile of erythrocyte

concentration of n‐3 PUFAs had the lowest plasma content of F2‐IsoPs:F3‐

IsoPs (Table 3). Indeed, we have previously reported that plasma F2‐IsoPs

concentrations in our Inuit population to be within normal ranges despite

106

their relatively high intake of n‐3 PUFAs (Alkazemi, Egeland et al. 2012).

Likewise, an assessment of urinary F2‐IsoPs demonstrated that consumption

of either EPA or DHA by subjects with hypertension and type 2 diabetes

reduced in vivo oxidant stress (Mori, Woodman et al. 2003; Barden, Mas et al.

2011). In a double blind, randomized, placebo‐controlled study, 59 patients

with type 2 diabetes were randomized to consume 4 g/day EPA, DHA, or

olive oil for 6 weeks (Mori, Woodman et al. 2003). Excretion of F2‐IsoPs was

reduced significantly by both EPA (19%) and DHA (20%) compared to olive

oil, with no change in inflammatory markers (Mori, Woodman et al. 2003).

Similarly, plasma F2‐IsoPs were reduced by 19% and 24% by EPA, and by

23% and 14% by DHA in treated hypertensive, type 2 diabetic patients and

hyperlipidemic men, respectively (Mas, Woodman et al. 2010). Despite the

reduction in plasma and urinary F2‐IsoPs, however, plasma F3‐IsoPs were not

detected in each of these studies, although Mas and colleagues employed GC‐

MS methodology similar to ours (Mas, Woodman et al. 2010). In addition, the

observed decreases in F2‐IsoPs in the above studies were unrelated to

changes in tissue content of EPA, DHA, AA, total n‐3 or n‐6 PUFAs (Mori,

Woodman et al. 2003; Mas, Woodman et al. 2010; Barden, Mas et al. 2011). It

has been suggested that the latter observations are due to changes in F2‐

IsoPs concentrations reflecting true reduction of oxidative stress rather than

resulting from supply or a substrate (Mas, Woodman et al. 2010). In

accordance with the above mentioned work, we also found that plasma F2‐

IsoPs were not related to erythrocyte fatty acid concentrations of any of the

measured fatty acids including EPA, DHA, AA, total n‐3s or n‐6 PUFAs, and

that F2‐IsoPs concentrations did not differ between the erythrocyte n‐3

tertiles. Our data indicated, however, that F2‐IsoPs concentrations inversely

correlated with n‐3 PUFAs and the n‐3: n‐6 ratio (p<.05) although this

relationship did not persist after partial correction for age, sex and WC.

Several animal studies have shown that supplementation of fish oil intake is

associated with increased tissue F3‐IsoPs concentrations and suppression of

the pro‐inflammatory F2‐IsoPs. Yin et al. (Yin, Liu et al. 2009) reported that

107

fish oil intake suppressed F2‐IsoPs and increased F3‐IsoPs in lung tissue in a

dose‐related manner in a murine model of ovalbumin‐induced lung

inflammation. An increase in F3‐IsoPs and a decrease in F2‐IsoPs were

demonstrated both in vitro and in vivo from heart tissues of mice fed EPA

and administered CCI4 (Gao, Yin et al. 2006).

In concordance to previously reported observations in northern native

populations (Bjerregaard, Jorgensen et al. 2004; Kuhnlein, Receveur et al.

2004; Nobmann, Ponce et al. 2005; Zhou, Kubow et al. 2011), our data also

demonstrate that older Inuit had higher erythrocyte concentrations of n‐3

PUFAs than younger Inuit, reflecting their higher intakes of traditional foods.

Decreased intake of traditional foods among younger Inuit has been noted,

especially consumption of marine mammals, due to greater availability of

market foods in the communities (Blanchet, Dewailly et al. 2000). Despite

older Inuit having a less favorable cardiometabolic profile (i.e., higher T‐

cholesterol, LDL‐C, FG, BF, and SBP), they appear to be protected against

whole body oxidative stress as shown by lower F2‐IsoPs. Likewise, regardless

of their higher blood concentrations of PCBs and Hg, older Inuit showed no

enhanced whole body oxidative stress, this could be partly related to their

higher Se status as shown via whole blood Se. In that regard, we have

reported protective effects in the same cohort from their high Se intake

against Hg‐mediated oxidative stress measured via plasma isofurans

(Alkazemi, Egeland et al. 2012). Previous studies have reported increased

oxidative stress in animals exposed to PCBs (Glauert, Tharappel et al. 2008).

PCB–induced lipid peroxidation has been associated with decreased hepatic

Se liver content and diminished SeGPx antioxidant activity (Twaroski,

O'Brien et al. 2001; Tawroski et al. 2001). Since blood Se was positively and

significantly associated with all PCBs (p<.001; data not shown), it is

conceivable that the Inuit are protected from PCB‐induced lipid peroxidation

by their high Se status. The higher F3‐IsoPs concentrations in older Inuit

could also provide metabolic protection as several in vivo and in vitro studies

108

have demonstrated that F3‐IsoPs exert anti‐inflammatory activities as

opposed to the potent pro‐inflammatory effects of F2‐IsoPs (Davis, Gao et al.

2006; Gao, Yin et al. 2006; Yin, Gao et al. 2007). Also, unlike F2‐IsoPs, F3‐IsoPs

do not affect adversely platelet shape or aggregation (Pratico, Murphy et al.

1997).

A study limitation is that it appears that plasma F3‐IsoPs are highly

modulated at the same level by the simultaneous presence of n‐3 PUFAs, Se,

PCBs, and Hg in the food matrix. Thus, these results should be interpreted

with caution, as all the above variables were highly correlated to each other

and at same level to F3‐IsoPs. The positive association of EPA to F3‐IsoPs is

consistent with this substrate‐product relationship; however, the positive

associations between F3‐IsoPs and PCBs, Hg or Se in the multivariate model

may reflect associations of F3‐IsoPs to the common food source of EPA in the

form of marine mammals (Butler Walker, Houseman et al. 2006).

Alternatively, the consistent positive relationship of F3‐IsoPs to PCBs, Hg and

WC might indicate that these pro‐inflammatory factors are promoting F3‐

IsoPs formation. Song et al. (2009) noted augmented excretion of F2‐IsoPs

and F3‐IsoPs in human urine following an acute inflammatory stimulus

induced by lipopolysaccharide treatment, indicating increased formation of

both isoprostane isoforms in response to a pro‐inflammatory response. Thus,

it is possible that F3‐IsoPs are indicative of chronic or acute exposure to

inflammation and oxidative stress. More studies needed to determine the

causal mechanisms for these observations.

In summary, detection of F3‐IsoPs in human plasma was demonstrated for

the first time, which can be related to the high erythrocyte EPA

concentrations in Inuit who are regular consumers of n‐3 PUFAs‐rich

seafood. Individuals with highest plasma concentrations of F3‐IsoPs also

showed the lowest erythrocyte percent content of n‐6: n‐3, AA:EPA,

saturated and trans fatty acids indicating higher consumption of traditional

109

foods and less market food intake. No difference in whole body oxidative

stress was noted among the n‐3 tertiles, despite a notable and significant

increase in plasma Hg and PCB concentrations in the highest tertile, which

indicates protection against contaminant‐induced oxidative stress. Also, the

elevated concentrations of F3‐IsoPs demonstrated in Inuit plasma could be a

significant novel factor to be considered in future evaluations of the health

benefits implicated with the traditional Inuit high n‐3 PUFAs diet.

110

Table 5.1 Plasma concentrations of isoprostanes1 and relative concentrations of fatty acids2 of study population N Mean 95%CI

F2‐IsoPs 230 27.31 25.73‐28.64

F3‐IsoPs 184 3.71 3.23‐4.23 F2‐IsoPs: F3‐IsoPs 183 2.74 2.50‐2.97 Total n‐33 233 6.05 5.61‐6.48 EPA 233 1.51 1.36‐1.66 DHA 233 2.57 2.38‐2.76 EPA + DHA 233 4.07 3.77‐4.38 Total n‐64 233 24.95 24.12‐25.77 AA 233 7.88 7.47‐ 8.28 n‐3: n‐6 233 .25 .23‐.27 n‐6: n‐3 233 6.27 5.42‐7.11 AA: EPA 233 8.65 7.62‐9.68 Total PUFA (n‐3+ n‐6 series) 233 30.99 29.97‐32.01 SFA5 233 43.04 42.36‐43.72 MUFA6 233 24.60 24.20‐25.00 TFA7 233 1.37 1.28‐1.46 1. F2‐IsoPs, F2‐ isoprostanes; F3‐IsoPs, F3‐Isoprostanes 2. n‐3, omega‐3; n‐6, omega‐6; AA, arachidonic acid (20:4n‐6); EPA, eicosapentaenoic acid (20:5n‐3); MUFA, monounsaturated fatty acids; PUFA, polyunsaturated Fatty acids; SFA, Saturated fatty acids; TFA, trans fatty acids. 3. C18:3 + 18:4 + 20:3 + 20:4 + 20:5 + 22:5 + 22:6 4. C18:2 + 18:3+ 20:2 +20:3 + 20:4 + 22:2 +22:4 +22:5 5. C14:0 +16:0 + 17:0 +18:0 + 20:0 +22:0 + 24:0 6. C14:1 +16:1 +18:1 + 20:1 + 22:1 + 24:1 7. C16:1t + 18:1t + 18:2t

111

Table 5.2 Plasma organochlorine concentrations (ug/L) and adjusted1 for total plasma lipids (ug/g lipids) of the study population n % Detected Geometric

Mean (ug/L)

95% CI Geometric Mean (ug/g lipids)

95% CI

Aroclor 115 51.57 9.02 7.12‐11.36 1.66 1.30‐2.11 no.101 32 14.35 .076 .061‐.092 .012 .009‐.015 no.105* 87 39.01 .069 .052‐.092 .008 .006‐.009 no.118* 114 51.12 .249 .191‐.310 .028 .023‐.035 no.128 49 21.97 .022 .017‐.027 .003 .009‐.004 no.138 115 51.57 .659 .523‐.808 .087 .069‐.109 no.153 115 52.57 1.599 1.26‐1.99 .229 .178‐.293 no.156* 105 47.09 .138 .106‐.170 .015 .012‐.019 no.163 115 51.57 .344 .264‐.428 .037 .029‐.048 no.170 112 50.22 .379 .292‐.473 .041 .032‐.053 no.180 115 51.56 .956 .749‐1.19 .118 .091‐.153 no.183 101 45.29 .105 .078‐.132 .012 .010‐.015 no.187 115 51.57 .316 .251‐.385 .038 .030‐.048 no.99 107 47.98 .299 .235‐.368 .039 .031‐.047 1. Adjusted for total serum lipids which were calculated by the following formula: Total lipids = 0.9+1.3*(cholesterol +triglycerides) [Rylander, 2006]. *. Dioxin‐like PCBs

112

Table 5.3 Unadjusted plasma isoprostanes1, blood contaminants, Se and relative fatty2 acid concentrations according to tertiles of n‐3 PUFAs level T1 (<4.36) T2 (>=4.36 &<6.90) T3 (>=6.90) 77

Mean (95%CI) 76 Mean (95%CI)

77 Mean (95%CI)

F2‐IsoPs 27.35 [24.96‐29.96]

28.85 [26.31‐31.64]

25.37 [23.05‐27.94]

NS

F3‐IsoPs 3.69a [2.70‐4.54]

4.93 ab [4.07‐5.97]

5.56b** [4.77‐6.49]

.005

F2‐IsoPs:F3‐IsoPs 3.39a [2.92‐3.86]

2.60b* [2.26‐2.94]

2.30b** [1.94‐2.65]

<.001

Hg 10.40a [7.71‐14.02]

17.56b** [14.00‐21.78]

28.44b** [23.74‐34.06]

<.001

Se 241.71a [222.95‐262.12]

270.46a [244.51‐299.16]

421.02b** [367.62‐482.06]

<.001

Hg:Se .42a [.37‐.48]

.51b* [.47‐.55]

.56b** [.53‐.59]

<.001

Adjusted No.153 109.60a [72.58‐165.50]

153.36a [111.48‐211.01]

646.99b** [463.77‐905.61]

<.001

n‐3 2.58a [2.29‐2.87]

5.78b** [5.62‐5.95]

9.69c** [9.14‐10.25]

<.001

n‐6 21.74a [19.95‐23.53]

27.37b** 26.23‐28.50]

25.72b** [24.71‐26.73]

<.001

n‐6:n‐3 11.25a [9.10‐13.40]

4.81b** [4.57‐5.05]

2.84b** [2.63‐3.04]

<.001

AA:EPA 12.95a [10.52‐15.39]

8.71b** [7.48‐9.94]

4.40c** [3.61‐5.19]

<.001

SFA 47.19a [45.77‐48.61]

41.75b** [40.98‐42.52]

40.24b** [39.66‐40.82]

<.001

MUFA 26.91a [26.21‐27.62]

23.91b** [23.34‐24.47]

23.03b** [22.53‐23.53]

<.001

PUFA 24.31a [22.31‐26.31]

33.14b** [31.98‐34.30]

35.41b** [34.51‐36.32]

<.001

TFA 1.58a [1.40‐1.76]

1.20b** [1.09‐1.32]

1.31b** [1.15‐1.48]

<.001

1.F2‐IsoPs, F2‐ isoprostanes; F3‐IsoPs, F3‐Isoprostanes 2.n‐3, omega‐3; n‐6, omega‐6; AA, arachidonic acid (20:4n‐6); EPA, eicosapentaenoic acid (20:5n‐3); MUFA, monounsaturated fatty acids; PUFA, polyunsaturated Fatty acids; SFA, Saturated fatty acids; TFA, trans fatty acids. 3. Adjusted for total serum lipids which were calculated by the following formula: Total lipids = 0.9+1.3*(cholesterol +triglycerides) [Rylander, 2006].

*. Correlation is significant at the 0.05 level (2‐tailed). **. Correlation is significant at the 0.01 level (2‐tailed).

113

Table 5.4 Bivariate correlation between plasma isoprostanes1, contaminants, Se, and relative fatty2 acid concentrations F2‐IsoPs F3‐IsoPs F2‐IsoPs:

F3‐IsoPs ToenailSe Se Hg Hg:Se No.1533

Toenail Se ‐.146* ‐.215** Se ‐.186** .276** ‐.362** Hg .358** ‐.315** ‐.160* .323** Hg:Se .305** ‐.238** ‐.171* n‐3 ‐.138* .269** ‐.320** ‐.142* .452** .307** .203** .446** n‐6 ‐.233** ‐.223** ‐.397* n‐3:n‐6 ‐.184** .262** ‐.277** .451** .332** .231** .479** AA:EPA ‐.329** .357** ‐.353** ‐.378** ‐.310** ‐.347** Saturated .135* MUFA .170* ‐.135* PUFA Trans .152* No.1533 .303** ‐.355** .537** .542** .482** 1.F2‐IsoPs, F2‐ isoprostanes; F3‐IsoPs, F3‐Isoprostanes 2.n‐3, omega‐3; n‐6, omega‐6; AA, arachidonic acid (20:4n‐6); EPA, eicosapentaenoic acid (20:5n‐3); MUFA, monounsaturated fatty acids; PUFA, polyunsaturated Fatty acids; SFA, Saturated fatty acids; TFA, trans fatty acids. 3.Adjusted for total serum lipids which were calculated by the following formula: Total lipids = 0.9+1.3*(cholesterol +triglycerides) [Rylander, 2006]. *.Correlation is significant at the 0.05 level (2‐tailed).

**.Correlation is significant at the 0.01 level (2‐tailed).

114

Table 5.5 Partial correlation between plasma isoprostanesa, contaminants, Se, and relative fattyb acid concentrations adjusted for age, gender and waist circumference (WC) F2‐IsoPs F3‐IsoPs F2‐IsoPs:

F3‐IsoPs Toenail Se Se Hg Hg:Se No.153 c

Toenail Se ‐.202 .078 ‐.255* Se .286** ‐.242* Hg .606** ‐.555** Hg:Se .635** ‐602** ‐.206 .07 n‐3 .234* ‐.289 ‐.602** .464** .486** .440** .278* n‐6 n‐3:n‐6 ‐.200 .07 .385** .367** .320** .250* AA:EPA ‐.280* .270* ‐.311** ‐.434** ‐.445** ‐.215 .05 SFA MUFA ‐.279* .257* .312** ‐.353** ‐.453** ‐.449** ‐.219* Total PUFA ‐.302 .221* .283* .279* TFA No.153c .402** .396** .358** a . F2‐IsoPs, F2‐ isoprostanes; F3‐IsoPs, F3‐Isoprostanes. b . n‐3, omega‐3; n‐6, omega‐6; AA, arachidonic acid (20:4n‐6); EPA, eicosapentaenoic acid (20:5n‐3); MUFA, monounsaturated fatty acids; PUFA, polyunsaturated Fatty acids; SFA, Saturated fatty acids; TFA, trans fatty acids. c . Adjusted for total serum lipids which were calculated by the following formula: Total lipids = 0.9+1.3*(cholesterol +triglycerides) [Rylander, 2006].

*. Correlation is significant at the 0.05 level (2‐tailed). **. Correlation is significant at the 0.01 level (2‐tailed).

115

Table 5.6‐a Comparison of study characteristics1 according to age categories (<40 and ≥ 40 yrs) N1/N2 Age < 40 years

Mean ± SEM Age ≥ 40 yrs Mean ± SEM

Significance (p<0.05)

Age 107/126 29.17 ± .55 53.94 ± 1.01 <.001 F2‐IsoPs 106/124 28.89

(26.67‐31.12) 25.77 (23.79‐31.21)

.038*

F3‐IsoPs 84/100 3.13 (2.49‐3.90)

4.25 (3.59‐34.99)

.027*

F2‐IsoPs:F3‐IsoPs 84/99 3.15 ± .19 2.39 ± 1.35 .001** Hg 104/123 11.39

(9.18‐14.14) 24.68 (20.68‐29.47)

<.001**

Se 104/123 250.49 (232.22‐270.15)

355.14 (320.25‐393.92)

<.001**

Toenail Se 92/103 .98 ± .036 .99 ± .027 NS Hg:Se 104/123 .44 ± .020 .55 ± .015 <.001** CRP 107/126 1.44

(1.13‐1.72) 1.85 (1.57‐2.16)

.042*

SBP 103/110 112.44 ± 1.38 122.00 ± 1.67 <.001** DBP 103/110 75.43 ± .98 77.41 ± 1.07 NS FG 107/125 4.71 ± .050 5.18 ± .06 <.001** T‐Chol 107/125 4.60 ± .088 5.40 ± .11 <.001** LDL‐C 106/122 2.58 ± .075 3.22 ± .10 <.001** HDL‐C 107/125 1.42 ± .045 1.51 ± .04 NS TG 107/123 1.32 ± .080 1.53 ± 1.30 NS WC 104/118 91.13± 1.46 93.61 ± 1.28 NS BMI 105/120 27.13 ± .54 28.34 ± .52 NS BF% 105/120 28.13 ± 1.03 30.92 ± 96 .049* 1 BF%, body fat percent; BMI, body mass index; CRP, C‐reactive protein; F2‐IsoPs, F2‐ isoprostanes; F3‐IsoPs, F3‐Isoprostanes; FG, fasting glucose; HDL‐C, high density lipoproteins cholesterol; Hg, mercury; HTN, hypertension; LDL‐C, low density lipoprotein cholesterol; SDP, diastolic blood pressure; Se, selenium; SBP, systolic blood pressure; T‐Chol, total cholesterol; TG, triglycerides; WC, waist circumference. *. Correlation is significant at the 0.05 level (2‐tailed). **. Correlation is significant at the 0.01 level (2‐tailed). NS‐ not significant

116

Table 5.6‐b Comparison of relative fatty acid1 concentrations according to age categories (<40 and ≥ 40 yrs) N1/N2 Age < 40 years

Mean ± SEM Age ≥ 40 yrs Mean ± SEM

Significance (p<0.05)

n‐3 107/126 4.49 ± .22 7.37 ± .32 <.001**

n‐6 107/126 25.38 ± .69 24.58 ± .51 NS n‐3:n‐6 107/126 .179 ± .01 .308 ± .015 <.001** AA:EPA 107/126 11.24 ± .91 6.45 ± .52 <.001** SFA 107/126 43.75 ± .56 42.43 ± 4.75 .06 MUFA 107/126 24.99 ± .32 24.28 ± .25 .08 Total PUFA 107/126 29.88 ± .84 31.94 ± .63 .047* TFA 107/126 1.38 ± .07 1.35 ± .06 NS

*. Correlation is significant at the 0.05 level (2‐tailed). **. Correlation is significant at the 0.01 level (2‐tailed). 1. n‐3, omega‐3; n‐6, omega‐6; AA, arachidonic acid (20:4n‐6); EPA, eicosapentaenoic acid (20:5n‐3); MUFA, monounsaturated fatty acids; PUFA, polyunsaturated Fatty acids; SFA, Saturated fatty acids; TFA, trans fatty acids.

117

Table 5.6‐c Comparison of study characteristics according to age categories (<40 and ≥ 40 yrs) Lipids1 Ng/g Lipids

N1/N2 Age < 40 years Mean (95%CI)

Age ≥ 40 yrs Mean (95%CI)

T‐test (p<0.05)

Aroclor 53/62 795.06 (615.60‐1026.60)

3100.99 (2255.28‐4562.85)

<.001

no.101 3/28 8.12 (3.15‐20.91)

12.57 (10.13‐15.60)

‐

no.105 31/56 3.81 (3.12‐4.66)

11.98 (9.16‐15.68)

<.001

no.118 52/62 14.15 (11.77‐16.99)

51.10 (38.1‐68.71)

<.001

no.128 9/40 2.57 (1.74‐3.78)

3.75 (3.09‐4.54)

‐

no.138 53/62 44.48 (84.29‐65.38)

155.81 (115.58‐210.09)

<.001

no.153 53/62 107.75 (82.49‐140.73)

435.01 (313.33‐603.95)

<.001

no.156 45/60 7.37 (5.75‐3.45)

26.93 (20.23‐35.83)

<.001

no.163 53/62 17.13 (13.16‐22.305)

73.01 (53.13‐100.35)

<.001

no.170 51/62 18.88 (14.13‐25.22)

78.20 (56.36‐108.49)

<.001

no.180 53/62 52.09 (38.82‐69.89)

237.57 (10.29‐331.40)

<.001

no.183 43/58 6.12 (4.96‐7.53)

20.98 (16.14‐27.31)

<.001

no.187 53/62 18.66 (14.46‐24.08)

69.57 (52.16‐92.41)

<.001

no.99 47/60 20.63 (16.99‐25.04)

63.17 (47.99‐83.12)

<.001

1. Adjusted for total serum lipids which were calculated by the following formula: Total lipids = 0.9+1.3 * (cholesterol +triglycerides) [Rylander, 2006].

118

Table 5.7 a Stepwise regression

Dependent variable: F3‐Isoprostanes

β (SE) β (SE) β (SE) β (SE) R2 F‐Value P‐Value

a Constant. Hg

.382 (.086)** .235 (.065)** .128 13.11 <.001

b Constant Hg Smoking

.540 (.116)** .249 (.064)** ‐.135(.068)* .166 8.78 <.001

c Constant Hg Smoking n‐3

.494 (.116)** .210 (.066)** ‐.151 (.067)* .019 (.01)* .203 7.39 <.001

d Constant Hg Smoking n‐6:n‐3

.639 (.123)** .232 (.064)** ‐.148 (.067)** ‐.010 (.005)* .206 7.54 <.001

e Constant Hg Smoking AA:EPA

.701 (.133)** .191 (.068)** ‐.141 (.066)* ‐.009 (.004)* .214 7.87 <.001

f Constant Hg Smoking C20:5 n‐3

.516 (.112)** .193 (.065)** ‐.146 (.065)** .067 (.023)* .238 9.057 <.001

All models included the following independent variables: Hg, Se, No.153, WC, age, sex, FG, T‐Chol, SBP or CRP, Smoking (1=yes, 2=no); and alcohol intake (1=yes, 2=no) Table 5.7b Stepwise regression Dependent variable: F2:F3

β (SE) β (SE) β (SE) β (SE) β (SE) R2 F‐Value P‐Value a Constant Hg Se Smoking 7.60

(1.69)** ‐.812 (.33)*

‐1.95 (.69)**

.745 (.33)*

.221 8.30 <.001

b Constant Hg Smoking C20:5n‐3 3.23

(.56)* ‐.804 (.32)*

.840 (.32)*

‐.365 (.12)**

.236 9.06 <.001

c Constant Se Smoking AA:EPA 5.80

(1.85)** ‐1.83

(.69)* .729 (.32)*

.056 (.02)**

.235 9.02 <.001

All models included the following independent variables: Hg, Se, No.153, WC, age, sex, FG, T‐Chol, SBP or CRP, Smoking (1=yes, 2=no); and alcohol intake (1=yes, 2=no)

119

Table 5.7c Stepwise regression Dependent variable: F3‐IsoPs

β (SE) β (SE) β (SE) β (SE) β (SE) R2 F‐Value P‐Value a Constant Hg .235

(.066)** .487 (.063)**

.485 59.25 <.001

b Constant Hg Se 1.78

(.40)** .826 (.10)**

‐.723 (.19)**

.587 44.01 <.001

c Constant Hg Se no.153 WC 1.89

(.38)** 1.034 (.11)**

‐.857 (.18)**

‐.147 (.065)*

.004 (.002)*

.666 29.97 <.001

All models included the following independent variables: Hg, Se, toenail Se, no.153, WC, age, sex, FG, T‐Chol, SBP or CRP, Smoking (1=yes, 2=no); and alcohol intake (1=yes, 2=no) Table 5.7d Stepwise regression Dependent variable: F2:F3

β (SE) β (SE) R2 F‐Value P‐Value a Constant Hg 4.74

(.334)** ‐2.31 (.319)*

.455 52.54 <.001

All models included the following independent variables: Hg, Se, toenail Se, No.153, WC, age, sex, FG, T‐Chol, SBP or CRP, Smoking (1=yes, 2=no); and alcohol intake (1=yes, 2=no)

120

CHAPTER 6: OVERALL SUMMARY AND CONCLUSIONS

6.1 Oxidative stress status and CVD risk

The relationship between oxidative stress biomarkers and cardiometabolic

risk factors has been sparsely addressed in the Inuit, which is an important to

study with respect to the protection against metabolic deteriorations

associated with CVD that is indicated in the Inuit population (Bjerregaard,

Young et al. 2003; Kuhnlein, Receveur et al. 2004; Young 2007). Also, recent

evidence suggests that there is a current trend in Inuit populations of an

increased prevalence of all cardiometabolic risk factors including obesity,

hypertension, hypertriglyceridemia, and smoking (Chateau‐Degat, Dewailly

et al. 2011; Alkazemi, Egeland et al. 2012). Such recent trends emphasize the

need to assess in the Inuit measures of oxidative stress, which is a potentially

important risk factor for CVD. This thesis research established that the Inuit

generally are protected from oxidative stress, however, not those with

evident clustering of metabolic risk factors associated with MetS. Inuit

categorized with hypertriglyceridemic waist, or those who met either NCEP

ATP or IDF definition for the MetS were shown to have higher plasma levels

of both F2‐IsoPs and IsoFs (Table 3.3). Nevertheless, the range of the

measured isoprostanes was well within normal ranges reported in healthy

Caucasians (Milne, Yin et al. 2007). This latter observation supports the

notion that the impact of obesity and associated metabolic parameters

common among obese Caucasians are less disrupted among the Inuit (Young,

Bjerregaard et al. 2007). Dysregulation in blood pressure and increase in

hypertension prevalence in the Inuit are emerging health concern (Valera,

Dewailly et al. 2009), and we showed that both isomers of isoprostanes, F2‐

IsoPs and IsoFs, were related to hypertension even after adjustments for WC,

sex, and age (Table 3.2). F2‐IsoPs remained significantly associated with

hypertension and CRP after multivariate adjustment. These findings

collectively indicate inflammation as a common theme as an underlying

pathogenesis of hypertension in this population (Ghanem and Movahed

2007). The above observations also supports the concept that oxidative

121

stress is an early event in the pathogenesis to cardiovascular disease as it

was highly correlated with abdominal obesity and as its generation seems to

precede other risk factors. Our data also indicated that early metabolic

dysregulation is independent of changes in blood lipids, as isoprostanes were

not related to individual lipid components (Table 5). We also observed that

Inuit with evident (diagnosed) dyslipidemia have significantly higher IsoFs

compared to counterparts (data not shown), suggesting that the Inuit lipid

profile is not completely protected against oxidative stress as previously

speculated and so perhaps there is a threshold whereby the protection

becomes limited. In addition, subjects with diagnosed abnormalities are

likely faced with multiple comorbidities and so it is likely the latter result can

be attributable to many residual confounding factors that were beyond the

scope of this thesis to further assess.

6.2 Sex difference associated with oxidative stress

We found no sex difference in obesity levels measured by BMI, and only a

clear distinction appeared between sexes when abdominal obesity was

considered, which was greater in women. Our data suggest that Inuit women

are at greater risk of oxidative stress due to their higher abdominal obesity,

and this warrant attention especially with increased rate of obesity in the

region (Egeland, Cao et al. 2011). In addition, Inuit women appeared to be

more susceptible to accumulate most cardiometabolic risk factors such as

higher triglycerides, CRP and SBP (of the 56% of subjects identified with

MetS according to NCEP definition, 69% were women). In that regard, it is

interesting to note that IsoP levels are generally similar or slightly higher in

healthy women compared to men (Keaney Jr, Larson et al. 2003; Katsuki,

Sumida et al. 2004). On the other hand, some authors have suggested that

women may generally be more protected than men against the adverse

health risks associated with oxidative stress or lipid abnormalities (Mosca,

Appel et al. 2004; Davies and Roberts II 2010), and that this protection may

be mediated via endogenous estradiols (Yagi 1997). Conversely, new

122

evidence showed oxidative stress measured via F2‐IsoPs increased with

increasing estrogen levels in premenopausal women (Sowers, McConnell et

al. 2008), and that F2‐IsoPs had an independent positive association with

estradiol after adjustment for age, race, age at menarche, γ‐tocopherol, ß‐

carotene, total cholesterol, and homocysteine (Schisterman, Gaskins et al.

2010). The results of these latter studies appear to question the commonly

held belief that women are protected from oxidative stress by endogenous

estradiol. This is a worrisome finding in light of reports of poor health status

among Inuit women (Ebbesson, Tejero et al. 2007), especially with the

increasing consumption of market foods and food insecurity issues faced by

women (Sharma, Gittelsohn et al. 2010). With the cumulative burden of

environmental factors such as exposure to heavy metals (Flora, Mittal et al.

2008), and poor nutrition (Schwedhelm, Maas et al. 2003), Inuit women may

be at an increased risk of oxidative stress, a key risk factor in chronic

diseases that affect women including CVD, cervical and breast cancer

(Goncalves, Erthal et al. 2005; Rossner, Gammon et al. 2006; Castelao and

Gago‐Dominguez 2008; Gago‐Dominguez and Castelao 2008).

6.3 Oxidative stress in smokers

We found that increased oxidative stress was not associated with smoking

status in the Inuit as observed by other investigators (Dietrich, Block et al.

2002; Ward, Hodgson et al. 2004), which could be likely due to smoking

status assessment method used in this study. Distinguishing smokers from

nonsmokers through questionnaires has discretional limitations in this

population given that the majority of our population was smokers (95%),

who reported smoking >10 cigarettes/day with the exposure to second hand

smoking is very likely. Previous studies of smoking and oxidative stress in

humans have been conflicting and difficult to interpret. Heterogeneity in

susceptibility to oxidative effects of smoking was purposed early to explain

the inconsistencies (Morrow, Frei et al. 1995). It has been demonstrated that

this susceptibility is modulated via polymorphisms of antioxidant genes

123

glutathione S‐transferase M1 (GSTM1), GSTT1, GSTP1, which was found

associated with susceptibility to accelerated decline of lung function in

smokers (He, Ruan et al. 2002; Muscat, Kleinman et al. 2004). Interestingly,

we found diminishing levels of plasma IsoFs with increasing levels of

smoking reflective of its decreased generation as it favors higher oxygen

tension compared to F2‐IsoPs. Indeed, smokers have been found to have a

significant degree of tissue hypoxia linked with physiological changes

associated with smoking such as vasoconstriction, arterial baroreflex

alterations, and endothelial dysfunction (Jensen, Goodson et al. 1991;

Sørensen, Jõrgensen et al. 2009). Thus, we speculate that IsoFs are present at

decreased levels in smokers due to above‐mentioned physiological changes

and decrease in lung capacity. The novelty of this finding, and the cross

sectional nature of this investigation precludes us from drawing definitive

conclusions. Further studies are needed to confirm this observation.

6.4 Oxidative stress and alcohol intake

We did not find association between oxidative stress and alcohol intake as

assessed in all exploratory analyses. Sixty‐five percent of the population sub‐

sample self‐reported alcohols use in the past year (vs. no use in the past

year), and heavy consumers were further identified by a positive response to

ever drinking alcohol to the point of losing consciousness in the past year.

There was likely non‐response bias in the self‐reporting of alcohol use as

there were missing values for this question on the health questionnaire that

might have affected the data interpretation. The role of cytochrome P450

enzymes in the metabolism of alcohol is well established. Isoenzymes

including CYP2E1, 1A2, and 3A4 are known to contribute to ethanol

metabolism (Ward, Puddey et al. 2005). Oxidative stress as measured by both

plasma and urinary F2‐IsoPs, was associated with alcohol intake as assessed

by biomarkers of alcohol intake such as γ‐glutamyl transpeptidase (Barden,

Zilkens et al. 2007), and with the pathogenesis of alcohol‐related

hypertension and endothelial dysfunction (Di Gennaro, Biggi et al. 2007).

124

Biomarkers for alcohol use would likely be more useful in investigating the

relationship between oxidative stress and alcohol in population surveys

because of reporting bias. Identification of alcoholism and the severity of

alcoholism would be helpful in considering the effects on oxidative stress as

an important factor in the haemodynamic, vascular and metabolic

abnormalities that indicate an unfavorable cardiovascular and metabolic risk

profile.

6.5 Selenium protection against mercury‐induced oxidative stress

Se as assessed via whole blood (reflective of short term intake) or toenail Se

(long‐term status), was associated with whole body oxidative stress evident

by inverse relationship to F2‐IsoPs concentrations. The inverse relationship

between Se status and F2‐IsoPs remained significant even after multivariate

adjustments. We found that plasma isofurans were more sensitive to

environmental burden than F2‐IsoPs, as their concentrations were

significantly associated to both blood Hg and Se (Table 6.5a and b). We

speculate this latter result is due to the proposed function of IsoFs as a

biomarker for conditions associated with mitochondrial dysfunction such as

MeHg exposure favoring its formation as mitochondrial damage can lead to

increased tissue O2 concentrations that promotes IsoF generation (Fessel and

Jackson Roberts 2005). Further studies are needed to verify the relationship

between Hg exposure and plasma IsoFs. Nevertheless, we demonstrated that

the presence of Hg in the traditional diet might not pose an immediate

concern with regards to oxidative stress. It should be noted, however, that we

showed that the Inuit protection from Hg‐induced oxidative stress varies

according to Se and Hg level of exposure. Not only higher antioxidant status

(Se) modulates oxidative stress, but also Hg‐Se interaction can influence

levels of both F2‐IsoPs and IsoFs (Table 6‐5d). This is an especially important

finding as both Hg and Se coexist in the same food matrix in the traditional

foods such as fish and sea mammals. Our above finding is also in support of

the observation that Se and Hg act antagonistically in experimental animals

125

(Khan and Wang 2009). Our results thus increase the potential validity of

assessing tissue Hg:Se ratios in relation to oxidative stress in toxicological

studies evaluating the nutritional risk versus benefit of traditional foods. This

thesis work also suggests that Hg may contribute to the development of

cardiovascular disease by increasing oxidative stress, and more importantly,

this relationship is modulated by not only Se alone, but the ratio of Hg:Se.

6.6 N‐3 PUFAs and Isoprostanes

We have quantified plasma levels of F3‐IsoPs for the first time in humans, and

they were present in 80% of all subjects tested. It is likely that the high

cellular concentration of the EPA substrate in the Inuit allowed F3‐IsoPs

detection as a clear gradient was demonstrated among tertiles of relative

concentration of n‐3 PUFAs. In addition, our work showed that F3‐IsoPs and

not F2‐IsoPs were directly modulated by tissue content of n‐3‐PUFAs.

According to our expectations and previous in vivo observations using

animal models, the thesis study showed that F3‐IsoPs and F2‐IsoPs have an

inverse relationship to each other and to Se status. This latter finding is

suggestive that the interaction of F3‐IsoPs and F2‐IsoPs can be a probable

mechanism for the reduced oxidative stress observed among our Inuit study

population despite the presence of factors that induce oxidative stress such

as increased age, cumulative cardiometabolic disturbances, environmental

pollutants (including both Hg and PCBs) and a high rate of smoking. Plasma

IsoFs however, were not associated with PCBs or n‐3 PUFAs and so were not

described in the third study (data not shown).

6.7 Strength and limitations

We used a biomarkers approach to assess dietary factors and to understand

health effects of constituents in traditional foods including n‐3 PUFAs, Hg,

and Se on oxidative stress in an insightful and careful approach. However,

here are several limitations to this work. First, because of the cross‐sectional

design and the highly inter‐correlated nature of the variables tested, results

126

should be interpreted with caution and this work does not allow for

conclusions for causality. We cannot rule out residual confounding from

variables not included in this thesis work. The lack of assessment of dietary

antioxidants, biomarkers for alcohol intake and smoking, specific markers for

tissue fat distribution or functional measures for atherosclerosis or

inflammation prevented us from further quantifying adverse risks associated

with oxidative stress. Inflammation status was defined by hs‐CRP alone and

therefore may have underestimated the prevalence of chronic inflammation

in the population. Additional cytokines of interest that were not measured

include IL‐6 and TNF‐α, which could have indicated propensity to increased

inflammation and associated with inflammatory‐mediated endothelial

dysfunction and other mechanistic links to the atherosclerotic process.

The lack of data on insulin limited further exploration of insulin sensitivity in

relation to oxidative stress and increased abdominal obesity. All

isoprostanes related to WC, which could be related to diminished glucose

sensitivity with greater visceral adiposity (D’Archivio, Annuzzi et al. 2012).

The greater sensitivity of plasma IsoFs to WC could be an indicator of an

ongoing mitochondrial impairment, which has been associated with diabetes

(Duarte, Moreira et al. 2012). Thus, IsoFs would be increased accordingly due

to an increase in ambient tissue oxygen concentrations. Whether IsoFs are a

mere biomarker or a mediator of adverse effects needs further exploration in

mechanistic studies.

This thesis provides good evidence in support of Hg‐induced oxidative stress

and highlights key factors that modulate oxidative stress in the Inuit. Overall,

the finding that many risk factors for CVD are associated with increased IsoP

levels supports the notion that lipid peroxidation is an important contributor

to the process of atherogenesis. The additive effect of various risk factors for

CVD can also be explained by this mechanism, because of their additive effect

on the extent of lipid peroxidation. F2‐IsoPs have emerged as one of the

127

most sensitive and reliable biomarkers of lipid peroxidation in vivo and

simultaneous measurements of the various IsoPs provided an advantage that

complemented the study of oxidative stress in the Inuit in relation to the

traditional diet in a comprehensive manner. The most reliable method for

measurement of IsoPs is stable isotope dilution mass spectrometry was

employed in this study (Milne, Yin et al. 2008).

6.8 Future directions

The body composition of the Inuit should be addressed in relation to

oxidative stress to further elucidate our observed sex differences in F2‐IsoPs

levels and to address the question whether women are more predisposed to

oxidative stress or are more protected. The emergence of the various IsoPs

isomers in relation to n‐3 PUFAs certainly is an exciting area of research that

could further increase understanding of essential fatty acid metabolism in

relation to CVD risk as well as to chronic diseases associated with oxidative

stress. In addition, it would be an exciting to evaluate the various

components of the traditional foods and establish a hierarchy or an index for

toxicological evaluation. This later approach could use probabilistic approach

similar to that proposed by Sioen et al. (2008) in order to be appropriate for

health campaigns or regional dietary recommendations in assessment of

risks and benefits of country foods. Such methodology could serve to provide

more appropriate evidence‐based recommendations.

6.9 Public Health Implications

Traditional foods appear to pose relatively little risk for oxidative stress for

the majority of the population. Young Inuit should be encouraged to include

various sources of traditional foods to maintain their advantageous

protection from CVD. Combined intake of n‐3 PUFA with Se via traditional

foods for the Inuit remains an important protective intervention against their

increasing risk of CVD.

128

6.10 Conclusions

Oxidative stress is an exciting area for research in nutritional and

toxicological studies with a growing understanding of its importance in

relation to the biological mechanisms involved in the pathogenesis of chronic

diseases including inflammatory asthma, chronic obstructive pulmonary

disease, cyctic fibrosis, cancer, CVD, diabetes, chronic fatigue syndrome and

neurological diseases. Isoprostanes can be quantified using GC‐MS

methodologies in biological fluids and the measurement of the novel

molecules F3‐IsoPs and IsoFs has great potential in studies involving

populations with high baseline intake of n‐3 PUFAs and organochlorine

contaminants.

129

7.00 References

Adler A.I., E.J. Boyko et al. (1994).”Lower prevalence of impaired glucose tolerance

and diabetes associated with daily seal oil or salmon consumption among

Alaska Natives.” Diabetes Care 17(12):1498‐501.

Albert, C. M., C. H. Hennekens, et al. (1998). "Fish consumption and risk of sudden

cardiac death." JAMA 279(1): 23‐8.

Alberti, K. G. M. M., R. H. Eckel, et al. (2009). "Harmonizing the metabolic

syndrome." Circulation 120(16): 1640‐1645.

Alberti, K. G. M. M., P. Zimmet, et al. (2005). "The metabolic syndrome ‐ a new

worldwide definition." Lancet 366(9491): 1059‐1062.

Alissa, E. M., S. M. Bahijri, et al. (2003). "The controversy surrounding selenium and

cardiovascular disease: a review of the evidence." Med Sci Monit 9(1): RA9‐

18.

Alkazemi, D., G. M. Egeland, et al. (2012). "Isoprostanes and isofurans as non‐

traditional risk factors for cardiovascular disease among Canadian Inuit."

Free Radic Res. (in preparation)

Arsenault, B. J., I. Lemieux, et al. (2010). "The hypertriglyceridemic‐waist

phenotype and the risk of coronary artery disease: results from the EPIC‐

Norfolk Prospective Population Study." CMAJ 182(13): 1427‐1432.

Arterburn, L. M., E. B. Hall, et al. (2006). "Distribution, interconversion, and dose

response of n‐3 fatty acids in humans." Am J Clin Nutr 83(6 Suppl): 1467S‐

1476S.

Ashfaq, S., J. L. Abramson, et al. (2006). "The relationship between plasma levels of

oxidized and reduced thiols and early atherosclerosis in healthy adults." J Am

Coll Cardiol 47(5): 1005‐11.

Bao, D. Q., T. A. Mori, et al. (1998). "Effects of dietary fish and weight reduction on

ambulatory blood pressure in overweight hypertensives." Hypertension

32(4): 710‐7.

130

Barden, A., E. Mas, et al. (2011). "The effects of oxidation products of arachidonic

acid and n3 fatty acids on vascular and platelet function." Free Radic Res

45(4): 469‐76.

Barden, A., R. R. Zilkens, et al. (2007). "A reduction in alcohol consumption is

associated with reduced plasma F2‐isoprostanes and urinary 20‐HETE

excretion in men." Free Radic Biol Med 42(11): 1730‐5.

Barden, A. E., T. A. Mori, et al. (2004). "Fish oil supplementation in pregnancy

lowers F2‐isoprostanes in neonates at high risk of atopy." Free Radic Res

38(3): 233‐9.

Barnes, P. J. and M. Karin (1997). "Nuclear factor‐kappaB: a pivotal transcription

factor in chronic inflammatory diseases." N Engl J Med 336(15): 1066‐71.

Basarici, I., R. E. Altekin, et al. (2007). "Associations of isoprostanes‐related

oxidative stress with surrogate subclinical indices and angiographic

measures of atherosclerosis." Coron Artery Dis 18(8): 615‐20.

Basu, S. (2006). "F2‐isoprostane induced prostaglandin formation in the rabbit."

Free Radic Res 40(3): 273‐277.

Basu, S. (2008). "F2‐isoprostanes in human health and diseases: from molecular

mechanisms to clinical implications." Antioxid Redox Signal 10(8): 1405‐34.

Basu, S. and J. Helmersson (2005). "Factors regulating isoprostane formation in

vivo." Antioxid Redox Signal 7(1‐2): 221‐35.

Battin, E. E. and J. L. Brumaghim (2009). "Antioxidant activity of sulfur and

selenium: a review of reactive oxygen species scavenging, glutathione

peroxidase, and metal‐binding antioxidant mechanisms." Cell Biochem

Biophys 55(1): 1‐23.

Baumann, H. and J. Gauldie (1994). "The acute phase response." Immunol Today

15(2): 74‐80.

Belanger, M. C., E. Dewailly, et al. (2006). "Dietary contaminants and oxidative

stress in Inuit of Nunavik." Metabolism 55(8): 989‐95.

131

Belanger, M. C., M. E. Mirault, et al. (2008). "Environmental contaminants and

redox status of coenzyme Q10 and vitamin E in Inuit from Nunavik."

Metabolism 57(7): 927‐33.

Bermudez, E. A. and P. M. Ridker (2002). "C‐reactive protein, statins, and the

primary prevention of atherosclerotic cardiovascular disease." Prev Cardiol

5(1): 42‐6.

Bersamin, A., B. R. Luick, et al. (2008). "Westernizing diets influence fat intake, red

blood cell fatty acid composition, and health in remote Alaskan Native

communities in the center for Alaska Native health study." J Am Diet Assoc

108(2): 266‐73.

Bjerregaard, P., E. Dewailly, et al. (2003). "Blood pressure among the Inuit

(Eskimo) populations in the Arctic." Scand J Public Health 31(2): 92‐9.

Bjerregaard, P., M. E. Jorgensen, et al. (2002). "Decreasing overweight and central

fat patterning with Westernization among the Inuit in Greenland and Inuit

migrants." Int J Obes Relat Metab Disord 26(11): 1503‐10.

Bjerregaard, P., M. E. Jorgensen, et al. (2007). "Cardiovascular risk amongst

migrant and non‐migrant Greenland Inuit in a gender perspective." Scand J

Public Health 35(4): 380‐6.

Bjerregaard, P., M. E. Jorgensen, et al. (2004). "Serum lipids of Greenland Inuit in

relation to Inuit genetic heritage, westernisation and migration."

Atherosclerosis 174(2): 391‐8.

Bjerregaard, P., T. K. Young, et al. (2004). "Indigenous health in the Arctic: an

overview of the circumpolar Inuit population." Scand J Public Health 32(5):

390‐5.

Bjerregaard, P., T. K. Young, et al. (2003). "Low incidence of cardiovascular disease

among the Inuit‐‐what is the evidence?" Atherosclerosis 166(2): 351‐7.

Blake, G. J. and P. M. Ridker (2002). "Inflammatory bio‐markers and cardiovascular

risk prediction." J Intern Med 252(4): 283‐94.

132

Blanchet, C., E. Dewailly, et al. (2000). "Contribution of Selected Traditional and

Market Foods to the Diet of Nunavik Inuit Women." Can J Diet Pract Res

61(2): 50‐59.

Block, G., M. Dietrich, et al. (2006). "Intraindividual variability of plasma

antioxidants, markers of oxidative stress, C‐reactive protein, cotinine, and

other biomarkers." Epidemiology 17(4): 404‐12.

Block, G., M. Dietrich, et al. (2002). "Factors associated with oxidative stress in

human populations." Am J Epidemiol 156(3): 274‐85.

Boosalis, M. G. (2008). "The role of selenium in chronic disease." Nutr Clin Pract

23(2): 152‐60.

Brasier, A. R., A. Recinos, 3rd, et al. (2002). "Vascular inflammation and the renin‐

angiotensin system." Arterioscler Thromb Vasc Biol 22(8): 1257‐66.

Brault, S., A. K. Martinez‐Bermudez, et al. (2003). "Selective neuromicrovascular

endothelial cell death by 8‐Iso‐prostaglandin F2alpha: possible role in

ischemic brain injury." Stroke 34(3): 776‐82.

Brooks, G. C., M. J. Blaha, et al. (2010). "Relation of C‐reactive protein to abdominal

adiposity." Am J Cardiol 106(1): 56‐61.

Burke, G. L., G. W. Evans, et al. (1995). "Arterial wall thickness is associated with

prevalent cardiovascular disease in middle‐aged adults. The Atherosclerosis

Risk in Communities (ARIC) Study." Stroke 26(3): 386‐91.

Butler Walker, J., J. Houseman, et al. (2006). "Maternal and umbilical cord blood

levels of mercury, lead, cadmium, and essential trace elements in Arctic

Canada." Environ Res 100(3): 295‐318.

Calabro, P., J. T. Willerson, et al. (2003). "Inflammatory cytokines stimulated C‐

reactive protein production by human coronary artery smooth muscle cells."

Circulation 108(16): 1930‐2.

Calder, P. C., N. Ahluwalia, et al. (2011). "Dietary factors and low‐grade

inflammation in relation to overweight and obesity." Br J Nutr 106 Suppl 3:

S5‐78.

133

Calviello, G., P. Palozza, et al. (1999). "Eicosapentaenoic acid inhibits the growth of

liver preneoplastic lesions and alters membrane phospholipid composition

and peroxisomal beta‐oxidation." Nutr Cancer 34(2): 206‐12.

Canuel, R., S. B. de Grosbois, et al. (2006). "New evidence on variations of human

body burden of methylmercury from fish consumption." Environ Health

Perspect 114(2): 302‐6.

Cao, Y. Z., Z. S. Cohen, et al. (2001). "Selenium modulates 1‐O‐alkyl‐2‐acetyl‐sn‐

glycero‐3‐phosphocholine (PAF) biosynthesis in bovine aortic endothelial

cells." Antioxid Redox Signal 3(6): 1147‐52.

Cao, Y. Z., C. C. Reddy, et al. (2000). "Altered eicosanoid biosynthesis in selenium‐

deficient endothelial cells." Free Radic Biol Med 28(3): 381‐9.

Castelao, J. E. and M. Gago‐Dominguez (2008). "Risk factors for cardiovascular

disease in women: relationship to lipid peroxidation and oxidative stress."

Med Hypotheses 71(1): 39‐44.

Cernichiari, E., R. Brewer, et al. (1995). "Monitoring methylmercury during

pregnancy: maternal hair predicts fetal brain exposure." NeuroToxicology

16(4): 705‐10.

Chateau‐Degat, M.L., E. Dewailly, et al. (2011). "Cardiovascular burden and related

risk factors among Nunavik (Quebec) Inuit: Insights from baseline findings in

the circumpolar Inuit Health in Transition cohort study." Can J Cardiol 26(6):

e190‐e196.

Chateau‐Degat, M.L., E. Dewailly, et al. (2011). "Obesity risks: towards an emerging

Inuit pattern." Int J Circumpolar Health 70(2): 166‐77.

Chen, C., L. Qu, et al. (2005). "Increased oxidative DNA damage, as assessed by

urinary 8‐hydroxy‐2'‐deoxyguanosine concentrations, and serum redox

status in persons exposed to mercury." Clin Chem 51(4): 759‐67.

Cheng, J. W. and F. Santoni (2008). "Omega‐3 fatty acid: a role in the management

of cardiac arrhythmias?" J Altern Complement Med 14(8): 965‐74.

134

Cheung, B. M. Y., K. L. Ong, et al. (2011). "C‐reactive protein as a predictor of

hypertension in the Hong Kong Cardiovascular Risk Factor Prevalence Study

(CRISPS) cohort." J Hum Hypertens 26(2): 108‐16.

Choi, A. L., P. Weihe, et al. (2009). "Methylmercury exposure and adverse

cardiovascular effects in Faroese whaling men." Environ Health Perspect

117(3): 367‐72.

Comstock, G. W., A. J. Alberg, et al. (2008). "The risk of developing lung cancer

associated with antioxidants in the blood: ascorbic acids, carotenoids, alpha‐

tocopherol, selenium, and total peroxyl radical absorbing capacity." Am J

Epidemiol 168(7): 831‐40.

Counil, E., E. Dewailly, et al. (2008). "Trans‐polar‐fat: all Inuit are not equal." Br J

Nutr: 4(1):1‐4.

Curran, J. E., J. B. Jowett, et al. (2005). "Genetic variation in selenoprotein S

influences inflammatory response." Nat Genet 37(11): 1234‐41.

Dallaire, F., E. Dewailly, et al. (2003). "Time trends of persistent organic pollutants

and heavy metals in umbilical cord blood of Inuit infants born in Nunavik

(Quebec, Canada) between 1994 and 2001." Environ Health Perspect

111(13): 1660‐4.

D’Archivio, M., G. Annuzzi, et al. (2012) “Predominant role of obesity/insulin

resistance in oxidative stress development.” Eur J Clin Invest 42 (1): 70–78

Davies, S. S. and L. J. Roberts Ii (2010). "F2‐isoprostanes as an indicator and risk

factor for coronary heart disease." Free Radic Biol Med 50(5): 559‐566.

Daviglus, M. L., J. Stamler, et al. (1997). "Fish consumption and risk of coronary

heart disease. What does the evidence show?" Eur Heart J 18(12): 1841‐2.

Davis, T. A., L. Gao, et al. (2006). "In vivo and in vitro lipid peroxidation of

arachidonate esters: the effect of fish oil omega‐3 lipids on product

distribution." J Am Chem Soc 128(46): 14897‐904.

135

de Lorgeril, M., P. Salen, et al. (2001). "Dietary and blood antioxidants in patients

with chronic heart failure. Insights into the potential importance of selenium

in heart failure." Eur J Heart Fail 3(6): 661‐9.

DeFilippis, A. P. and L. S. Sperling (2006). "Understanding omega‐3's." Am Heart J

151(3): 564‐70.

DeGraba, T. J. (1997). "Expression of inflammatory mediators and adhesion

molecules in human atherosclerotic plaque." Neurology 49(5 Suppl 4): S15‐9.

Dewailly, E., P. Ayotte, et al. (2001). "Exposure of the Inuit population of Nunavik

(Arctic Quebec) to lead and mercury." Arch Environ Health 56(4): 350‐7.

Dewailly, E., C. Blanchet, et al. (2003). "Fish consumption and blood lipids in three

ethnic groups of Quebec (Canada)." Lipids 38(4): 359‐65.

Dewailly, E., Ayotte, P., Pereg D, et al. (2007). “Exposure to environmental

contaminants in Nunavik: Metals.” Nunavik Inuit health survey 2004,

Qanuippitaa? How are we? Quebec: Institut national de santé publique de

Quebec (INSQ) & Nunavik regional board of health and social services

(NRBHSS). http://www.inspq.qc.ca

Dewailly, E., C. Blanchet, et al. (2001). "n‐3 Fatty acids and cardiovascular disease

risk factors among the Inuit of Nunavik." Am J Clin Nutr 74(4): 464‐73.

Dewailly, E., Dellaire R, Pereg D, et al. (2007). “Exposure to environmental

contaminants in Nunavik: Persistent organic pollutants and new

contaminants of concern.” Nunavik Inuit health Survey 2004, Qanuippitaa?

How are we? Quebec: Institut national de santé publique de Quebec (INSQ) &

Nunavik regional board of health and social services (NRBHSS).

http://www.inspq.qc.ca

Di Gennaro, C., A. Biggi, et al. (2007). "Endothelial dysfunction and cardiovascular

risk profile in long‐term withdrawing alcoholics." J Hypertens 25(2): 367‐73.

Dietrich, M., G. Block, et al. (2002). "Antioxidant supplementation decreases lipid

peroxidation biomarker F2‐isoprostanes in plasma of smokers." Cancer

Epidemiol Biomarkers Prev 11(1): 7‐13.

136

Duntas, L. H. (2009). "Selenium and inflammation: underlying anti‐inflammatory

mechanisms." Horm Metab Res 41(6): 443‐7.

Duarte, A.I., P.I. Moreira, et al. (2012) “Insulin in central nervous system: more

than just a peripheral hormone.” J Aging Res (in preparation)

Dyerberg, J. (1989). "Coronary heart disease in Greenland Inuit: a paradox.

Implications for western diet patterns." Arctic Med Res 48(2): 47‐54.

Ebbesson, S. O., A. I. Adler, et al. (2005). "Cardiovascular disease and risk factors in

three Alaskan Eskimo populations: the Alaska‐Siberia project." Int J

Circumpolar Health 64(4): 365‐86.

Ebbesson, S. O., L. O. Ebbesson, et al. (2005). "A successful diabetes prevention

study in Eskimos: the Alaska Siberia project." Int J Circumpolar Health 64(4):

409‐24.

Ebbesson, S. O., P. M. Risica, et al. (2005). "Eskimos have CHD despite high

consumption of omega‐3 fatty acids: the Alaska Siberia project." Int J

Circumpolar Health 64(4): 387‐95.

Ebbesson, S. O., P. M. Risica, et al. (2005). "Omega‐3 fatty acids improve glucose

tolerance and components of the metabolic syndrome in Alaskan Eskimos:

the Alaska Siberia project." Int J Circumpolar Health 64(4): 396‐408.

Ebbesson, S. O., M. J. Roman, et al. (2008). "Consumption of omega‐3 fatty acids is

not associated with a reduction in carotid atherosclerosis: the Genetics of

Coronary Artery Disease in Alaska Natives study." Atherosclerosis 199(2):

346‐53.

Ebbesson, S. O., M. E. Tejero, et al. (2007). "Fatty acid consumption and metabolic

syndrome components: the GOCADAN study." J Cardiometab Syndr 2(4):

244‐9.

Egeland, G. M., Z. Cao, et al. (2011). "Hypertriglyceridemic‐waist phenotype and

glucose intolerance among Canadian Inuit: the International Polar Year Inuit

Health Survey for Adults 2007‐2008." CMAJ 183(9): E553‐8.

137

Egeland, G. M., Z. Cao, et al. (2011). "Hypertriglyceridemic‐waist phenotype and

glucose intolerance among Canadian Inuit: the International Polar Year Inuit

Health Survey for Adults 2007‚Äì2008." CMAJ 183(9): E553‐E558.

Eilat‐Adar, S., M. Mete, et al. (2009). "Dietary patterns are linked to cardiovascular

risk factors but not to inflammatory markers in Alaska Eskimos." J Nutr

139(12): 2322‐8.

Elenkov, I. J., D. G. Iezzoni, et al. (2005). "Cytokine dysregulation, inflammation and

well‐being." Neuroimmunomodulation 12(5): 255‐69.

Fam, S. S. and J. D. Morrow (2003). "The isoprostanes: unique products of

arachidonic acid oxidation‐a review." Curr Med Chem 10(17): 1723‐40.

Fessel, J. P., C. Hulette, et al. (2003). "Isofurans, but not F2‐isoprostanes, are

increased in the substantia nigra of patients with Parkinson's disease and

with dementia with Lewy body disease." J Neurochem 85(3): 645‐50.

Fessel, J. P. and L. Jackson Roberts (2005). "Isofurans: novel products of lipid

peroxidation that define the occurrence of oxidant injury in settings of

elevated oxygen tension." Antioxid Redox Signal 7(1‐2): 202‐9.

Fessel, J. P., N. A. Porter, et al. (2002). "Discovery of lipid peroxidation products

formed in vivo with a substituted tetrahydrofuran ring (isofurans) that are

favored by increased oxygen tension." Proc Natl Acad Sci U S A 99(26):

16713‐8.

Fiskum, G., A. Starkov, et al. (2003). "Mitochondrial mechanisms of neural cell

death and neuroprotective interventions in Parkinson's disease." Ann N Y

Acad Sci 991: 111‐9.

Flora, S. J., M. Mittal, et al. (2008). "Heavy metal induced oxidative stress & its

possible reversal by chelation therapy." Indian J Med Res 128(4): 501‐23.

Flores‐Mateo, G., A. Navas‐Acien, et al. (2006). "Selenium and coronary heart

disease: a meta‐analysis." Am J Clin Nutr 84(4): 762‐73.

Fontaine, J., E. Dewailly, et al. (2008). "Re‐evaluation of blood mercury, lead and

cadmium concentrations in the Inuit population of Nunavik (Quebec): a

cross‐sectional study." Environ Health 7: 25.

138

Ford, E. S., W. H. Giles, et al. (2002). "Prevalence of the metabolic syndrome among

US adults: findings from the third National Health and Nutrition Examination

Survey." JAMA 287(3): 356‐9.

Ford, E. S., A. H. Mokdad, et al. (2003). "Trends in waist circumference among U.S.

adults." Obes Res 11(10): 1223‐31.

Forouhi, N. G., N. Sattar, et al. (2001). "Relation of C‐reactive protein to body fat

distribution and features of the metabolic syndrome in Europeans and South

Asians." Int J Obes Relat Metab Disord 25(9): 1327‐31.

Gago‐Dominguez, M. and J. E. Castelao (2008). "Role of lipid peroxidation and

oxidative stress in the association between thyroid diseases and breast

cancer." Crit Rev Oncol Hematol 68(2): 107‐14.

Gailer, J., G. N. George, et al. (2000). "Structural basis of the antagonism between

inorganic mercury and selenium in mammals." Chem Res Toxicol 13(11):

1135‐42.

Ganter, U., R. Arcone, et al. (1989). "Dual control of C‐reactive protein gene

expression by interleukin‐1 and interleukin‐6." EMBO J 8(12): 3773‐9.

Gao, L., J. Wang, et al. (2007). "Novel n‐3 fatty acid oxidation products activate Nrf2

by destabilizing the association between Keap1 and Cullin3." J Biol Chem

282(4): 2529‐37.

Gao, L., H. Yin, et al. (2006). "Formation of F‐ring isoprostane‐like compounds (F3‐

isoprostanes) in vivo from eicosapentaenoic acid." J Biol Chem 281(20):

14092‐9.

Ghanem, F. A. and A. Movahed (2007) "Inflammation in high blood pressure: a

clinician perspective." J Am Soc Hypertens 1(2): 113‐119.

Ghanim, H., R. Garg, et al. (2001). "Suppression of nuclear factor‐kappaB and

stimulation of inhibitor kappaB by troglitazone: evidence for an anti‐

inflammatory effect and a potential antiatherosclerotic effect in the obese." J

Clin Endocrinol Metab 86(3): 1306‐12.

139

Giles, W. H., J. B. Croft, et al. (2000). "Association between total homocyst(e)ine and

the likelihood for a history of acute myocardial infarction by race and

ethnicity: Results from the Third National Health and Nutrition Examination

Survey." Am Heart J 139(3): 446‐53.

Gillum, R. F., M. Mussolino, et al. (2000). "The relation between fish consumption,

death from all causes, and incidence of coronary heart disease. the NHANES I

Epidemiologic Follow‐up Study." J Clin Epidemiol 53(3): 237‐44.

Glauert, H. P., J. C. Tharappel, et al. (2008). "Role of oxidative stress in the

promoting activities of PCBs." Environ Toxicol Pharmacol 25(2): 247‐50.

Goncalves, T. L., F. Erthal, et al. (2005). "Involvement of oxidative stress in the pre‐

malignant and malignant states of cervical cancer in women." Clin Biochem

38(12): 1071‐5.

Grandjean, P., K. Murata, et al. (2004). "Cardiac autonomic activity in

methylmercury neurotoxicity: 14‐year follow‐up of a Faroese birth cohort." J

Pediatr 144(2): 169‐76.

Grimsgaard, S., K. H. Bonaa, et al. (1998). "Effects of highly purified

eicosapentaenoic acid and docosahexaenoic acid on hemodynamics in

humans." Am J Clin Nutr 68(1): 52‐9.

Grimsgaard, S., K. H. Bonaa, et al. (1997). "Highly purified eicosapentaenoic acid

and docosahexaenoic acid in humans have similar triacylglycerol‐lowering

effects but divergent effects on serum fatty acids." Am J Clin Nutr 66(3): 649‐

59.

Gross, M., M. Steffes, et al. (2005). "Plasma F2‐isoprostanes and coronary artery

calcification: the CARDIA Study." Clin Chem 51(1): 125‐31.

Grotto, D., J. Valentini, et al. "Evaluation of toxic effects of a diet containing fish

contaminated with methylmercury in rats mimicking the exposure in the

Amazon riverside population." Environ Res 111(8): 1074‐82.

Grundy, S. M., J. I. Cleeman, et al. (2004). "Implications of recent clinical trials for

the National Cholesterol Education Program Adult Treatment Panel III

Guidelines." J Am Coll Cardiol 44(3): 720‐32.

140

Grundy, S. M., B. Hansen, et al. (2004). "Clinical Management of Metabolic

Syndrome." Circulation 109(4): 551‐556.

Guallar, E., M. I. Sanz‐Gallardo, et al. (2002). "Mercury, fish oils, and the risk of

myocardial infarction." N Engl J Med 347(22): 1747‐54.

Handelman, G. J., M. F. Walter, et al. (2001). "Elevated plasma F2‐isoprostanes in

patients on long‐term hemodialysis." Kidney Int 59(5): 1960‐1966.

Hansel, B., P. Giral, et al. (2004). "Metabolic syndrome is associated with elevated

oxidative stress and dysfunctional dense high‐density lipoprotein particles

displaying impaired antioxidative activity." J Clin Endocrinol Metab 89(10):

4963‐71.

Harris, W. S., G. Lu, et al. (1997). "Influence of n‐3 fatty acid supplementation on

the endogenous activities of plasma lipases." Am J Clin Nutr 66(2): 254‐60.

He, J.‐Q., J. Ruan, et al. (2002). "Antioxidant Gene Polymorphisms and Susceptibility

to a Rapid Decline in Lung Function in Smokers." Am J Respir Crit Care Med

166(3): 323‐328.

He, K., Y. Song, et al. (2004). "Accumulated evidence on fish consumption and

coronary heart disease mortality: a meta‐analysis of cohort studies."

Circulation 109(22): 2705‐11.

Hegele, R. A. (1999). "Lessons from genetic studies in native Canadian

populations." Nutr Rev 57(5 Pt 2): S43‐9; discussion S49‐50.

Hegele, R. A., C. Tully, et al. (1997). "V677 mutation of methylenetetrahydrofolate

reductases and cardiovascular disease in Canadian Inuit." Lancet 349(9060):

1221‐2.

Hegele, R. A., T. K. Young, et al. (1997). "Are Canadian Inuit at increased genetic

risk for coronary heart disease?" J Mol Med (Berl) 75(5): 364‐70.

Helmersson, J., J. Arnlov, et al. (2005). "Serum selenium predicts levels of F2‐

isoprostanes and prostaglandin F2alpha in a 27 year follow‐up study of

Swedish men." Free Radic Res 39(7): 763‐70.

141

Helmersson, J., A. Larsson, et al. (2005). "Active smoking and a history of smoking

are associated with enhanced prostaglandin F(2alpha), interleukin‐6 and F2‐

isoprostane formation in elderly men." Atherosclerosis 181(1): 201‐207.

Higdon, J. V., J. Liu, et al. (2000). "Supplementation of postmenopausal women with

fish oil rich in eicosapentaenoic acid and docosahexaenoic acid is not

associated with greater in vivo lipid peroxidation compared with oils rich in

oleate and linoleate as assessed by plasma malondialdehyde and F(2)‐

isoprostanes." Am J Clin Nutr 72(3): 714‐22.

Hino, A., H. Adachi, et al. (2004). "Very long chain N‐3 fatty acids intake and carotid

atherosclerosis: an epidemiological study evaluated by ultrasonography."


Hong, S., K. Gronert, et al. (2003). "Novel docosatrienes and 17S‐resolvins

generated from docosahexaenoic acid in murine brain, human blood, and

glial cells. Autacoids in anti‐inflammation." J Biol Chem 278(17): 14677‐87.

Hong, Y., C. H. Li, et al. (1989). "The role of selenium‐dependent and selenium‐

independent glutathione peroxidases in the formation of prostaglandin F2

alpha." J Biol Chem 264(23): 13793‐800.

Hooper, L., R. L. Thompson, et al. (2006). "Risks and benefits of omega 3 fats for

mortality, cardiovascular disease, and cancer: systematic review." BMJ

332(7544): 752‐760.

Hou, X., L. J. Roberts, 2nd, et al. (2004). "Isomer‐specific contractile effects of a

series of synthetic f2‐isoprostanes on retinal and cerebral microvasculature."

Free Radic Biol Med 36(2): 163‐72.

Houston, M. C. (2011). "Role of mercury toxicity in hypertension, cardiovascular

disease, and stroke." J Clin Hypertens (Greenwich) 13(8): 621‐7.

Huang, Y. L., S. L. Cheng, et al. (1996). "Lipid peroxidation in rats administrated

with mercuric chloride." Biol Trace Elem Res 52(2): 193‐206.

Hunter, D. J., J. S. Morris, et al. (1990). "Predictors of selenium concentration in

human toenails." Am J Epidemiol 132(1): 114‐22.

142

Il'yasova, D., J. D. Morrow, et al. (2004). "Epidemiological marker for oxidant

status: comparison of the ELISA and the gas chromatography/mass

spectrometry assay for urine 2,3‐dinor‐5,6‐dihydro‐15‐F2t‐isoprostane." Ann

Epidemiol 14(10): 793‐7.

Il'yasova, D., J. D. Morrow, et al. (2005). "Urinary F2‐isoprostanes are not

associated with increased risk of type 2 diabetes." Obes Res 13(9): 1638‐44.

Iraz, M., H. Erdogan, et al. (2005). "Brief communication: omega‐3 essential fatty

acid supplementation and erythrocyte oxidant/antioxidant status in rats."

Ann Clin Lab Sci 35(2): 169‐73.

Jensen, J. A., W. H. Goodson, et al. (1991). "Cigarette Smoking Decreases Tissue

Oxygen." Arch Surg 126(9): 1131‐1134.

Jin, M. M., C. Y. Pan, et al. (2008). "[Association between metabolic syndrome and

the 10 years mortality of cerebro‐cardiovascular diseases in the senile

population]." Zhonghua Xin Xue Guan Bing Za Zhi 36(2): 118‐22.

Jin, X., N. Hidiroglou, et al. (2012). "Dietary selenium (Se) and vitamin E (VE)

supplementation modulated methylmercury‐mediated changes in markers of

cardiovascular diseases in rats." Cardiovasc Toxicol 12(1): 10‐24.

Johansen, P., T. Pars, et al. (2000). "Lead, cadmium, mercury and selenium intake

by Greenlanders from local marine food." Sci Total Environ 245(1‐3): 187‐94.

Jorgensen, M. E., P. Bjerregaard, et al. (2004). "Prevalence of the metabolic

syndrome among the Inuit in Greenland. A comparison between two

proposed definitions." Diabet Med 21(11): 1237‐42.

Jorgensen, M. E., K. Borch‐Johnsen, et al. (2006). "Lifestyle modifies obesity‐

associated risk of cardiovascular disease in a genetically homogeneous

population." Am J Clin Nutr 84(1): 29‐36.

Jorgensen, M. E., C. Glumer, et al. (2003). "Obesity and central fat pattern among

Greenland Inuit and a general population of Denmark (Inter99): Relationship

to metabolic risk factors." Int J Obes Relat Metab Disord 27(12): 1507‐1515.

143

Kadiiska, M. B., B. C. Gladen, et al. (2005). "Biomarkers of oxidative stress study II:

are oxidation products of lipids, proteins, and DNA markers of CCl4

poisoning?" Free Radic Biol Med 38(6): 698‐710.

Katsuki, A., Y. Sumida, et al. (2004). "Increased oxidative stress is associated with

serum levels of triglyceride, insulin resistance, and hyperinsulinemia in

Japanese metabolically obese, normal‐weight men." Diabetes Care 27(2):

631‐2.

Kaufman, D. J., M. J. Roman, et al. (2008). "Prevalence of smoking and its

relationship with carotid atherosclerosis in Alaskan Eskimos of the Norton

Sound region: the GOCADAN study." Nicotine Tob Res 10(3): 483‐91.

Keaney, J. F., Jr., M. G. Larson, et al. (2003). "Obesity and systemic oxidative stress:

clinical correlates of oxidative stress in the Framingham Study." Arterioscler

Thromb Vasc Biol 23(3): 434‐9.

Keaney, J. F., Jr., M. G. Larson, et al. (2003). "Obesity and systemic oxidative stress:

clinical correlates of oxidative stress in the Framingham Study." Arterioscler


Keaney Jr, J. F., M. G. Larson, et al. (2003). "Obesity and systemic oxidative stress:

Clinical correlates of oxidative stress in the Framingham study." Arterioscler


Khan, M. A. and F. Wang (2009). "Reversible dissolution of glutathione‐mediated

HgSe(x)S(1‐x) nanoparticles and possible significance in Hg‐Se antagonism."

Chem Res Toxicol 22(11): 1827‐32.

Kingman, A., T. Albertini, et al. (1998). "Mercury concentrations in urine and whole

blood associated with amalgam exposure in a US military population." J Dent

Res 77(3): 461‐71.

Kitagawa, K., H. Hougaku, et al. (2007). "Carotid intima‐media thickness and risk of

cardiovascular events in high‐risk patients. Results of the Osaka Follow‐Up

Study for Carotid Atherosclerosis 2 (OSACA2 Study)." Cerebrovasc Dis 24(1):

35‐42.

144

Kok, F. J., A. M. De Bruijn, et al. (1987). "Selenium status and chronic disease

mortality: Dutch epidemiological findings." Int J Epidemiol 16(2): 329‐32.

Konig, A., C. Bouzan, et al. (2005). "A quantitative analysis of fish consumption and

coronary heart disease mortality." Am J Prev Med 29(4): 335‐46.

Kretz‐Remy, C. and A. P. Arrigo (2001). "Selenium: a key element that controls NF‐

kappa B activation and I kappa B alpha half life." Biofactors 14(1‐4): 117‐25.

Kris‐Etherton, P. M., W. S. Harris, et al. (2002). "Fish consumption, fish oil, omega‐3

fatty acids, and cardiovascular disease." Circulation 106(21): 2747‐57.

Kuhnlein, H., O. Receveur, et al. (2001). "Energy, fat and calcium in bannock

consumed by Canadian Inuit." J Am Diet Assoc 101(5): 580‐1.

Kuhnlein, H. V. and O. Receveur (1996). "Dietary change and traditional food

systems of indigenous peoples." Annu Rev Nutr 16: 417‐42.

Kuhnlein, H. V., O. Receveur, et al. (1995). "Arctic indigenous women consume

greater than acceptable levels of organochlorines." J Nutr 125(10): 2501‐10.

Kuhnlein, H. V., O. Receveur, et al. (2008). "Unique patterns of dietary adequacy in

three cultures of Canadian Arctic indigenous peoples." Public Health Nutr

11(4): 349‐60.

Kuhnlein, H. V., O. Receveur, et al. (2004). "Arctic indigenous peoples experience

the nutrition transition with changing dietary patterns and obesity." J Nutr

134(6): 1447‐53.

Kwak, M. K., T. W. Kensler, et al. (2003). "Induction of phase 2 enzymes by serum

oxidized polyamines through activation of Nrf2: effect of the polyamine

metabolite acrolein." Biochem Biophys Res Commun 305(3): 662‐70.

Kwak, M. K., N. Wakabayashi, et al. (2003). "Antioxidants enhance mammalian

proteasome expression through the Keap1‐Nrf2 signaling pathway." Mol Cell

Biol 23(23): 8786‐94.

Laden, F., L. M. Neas, et al. (1999). "Predictors of plasma concentrations of DDE and

PCBs in a group of U.S. women." Environ Health Perspect 107(1): 75‐81.

145

Lalouel, J. M., A. Rohrwasser, et al. (2001). "Angiotensinogen in essential

hypertension: from genetics to nephrology." J Am Soc Nephrol 12(3): 606‐15.

Lee, C. Y., S. H. Huang, et al. (2008). "Measurement of F2‐isoprostanes,

hydroxyeicosatetraenoic products, and oxysterols from a single plasma

sample." Free Radic Biol Med 44(7): 1314‐22.

Liu, W., J. D. Morrow, et al. (2009). "Quantification of F2‐isoprostanes as a reliable

index of oxidative stress in vivo using gas chromatography‐mass

spectrometry (GC‐MS) method." Free Radic Biol Med 47(8): 1101‐7.

Lavi, S., E. H. Yang, et al. (2008). "The interaction between coronary endothelial

dysfunction, local oxidative stress, and endogenous nitric oxide in humans."

Hypertension 51(1): 127‐133.

Lemieux, I., P. Poirier, et al. (2007). "Hypertriglyceridemic waist: A useful

screening phenotype in preventive cardiology?" Can J Cardiol 23(Suppl. B):

23B‐31B.

Lemire, M., M. Fillion, et al. (2010). "Selenium and mercury in the Brazilian

Amazon: opposing influences on age‐related cataracts." Environ Health

Perspect 118(11): 1584‐9.

Lemire, M., M. Fillion, et al. (2011). "Selenium from dietary sources and motor

functions in the Brazilian Amazon." NeuroToxicology 32(6): 944‐953.

Li, P., X. Feng, et al. (2008). "Mercury exposures and symptoms in smelting

workers of artisanal mercury mines in Wuchuan, Guizhou, China." Environ

Res 107(1): 108‐14.

Libermann, T. A. and D. Baltimore (1990). "Activation of interleukin‐6 gene

expression through the NF‐kappa B transcription factor." Mol Cell Biol 10(5):

2327‐34.

Liese, A. D., E. J. Mayer‐Davis, et al. (1998). "Development of the multiple metabolic

syndrome: an epidemiologic perspective." Epidemiol Rev 20(2): 157‐72.

Lindh, U., A. Danersund, et al. (1996). "Selenium protection against toxicity from

cadmium and mercury studied at the cellular level." Cell Mol Biol (Noisy‐le‐

grand) 42(1): 39‐48.

146

Liu, J., A. J. Hanley, et al. (2006). "Characteristics and prevalence of the metabolic

syndrome among three ethnic groups in Canada." Int J Obes (Lond) 30(4):

669‐76.

Longnecker, M. P., M. J. Stampfer, et al. (1993). "A 1‐y trial of the effect of high‐

selenium bread on selenium concentrations in blood and toenails." Am J Clin

Nutr 57(3): 408‐13.

Lopez‐Garcia, E., M. B. Schulze, et al. (2004). "Consumption of (n‐3) fatty acids is

related to plasma biomarkers of inflammation and endothelial activation in

women." J Nutr 134(7): 1806‐11.

Lorenz, M. W., S. von Kegler, et al. (2006). "Carotid intima‐media thickening

indicates a higher vascular risk across a wide age range: prospective data

from the Carotid Atherosclerosis Progression Study (CAPS)." Stroke 37(1):

87‐92.

Lorenzo, C., M. Okoloise, et al. (2003). "The metabolic syndrome as predictor of

type 2 diabetes: the San Antonio heart study." Diabetes Care 26(11): 3153‐9.

Lund, B. O., D. M. Miller, et al. (1991). "Mercury‐induced H2O2 production and lipid

peroxidation in vitro in rat kidney mitochondria." Biochem Pharmacol 42

Suppl: S181‐7.

Lund, B. O., D. M. Miller, et al. (1993). "Studies on Hg(II)‐induced H2O2 formation

and oxidative stress in vivo and in vitro in rat kidney mitochondria." Biochem

Pharmacol 45(10): 2017‐24.

Lymbury, R. S., M. J. Marino, et al. (2009) "Effect of dietary selenium on the

progression of heart failure in the ageing spontaneously hypertensive rat."

Mol Nutr Food Res 54(10): 1436‐44.

Maehira, F., G. A. Luyo, et al. (2002). "Alterations of serum selenium concentrations

in the acute phase of pathological conditions." Clin Chim Acta 316(1‐2): 137‐

46.

147

Maehira, F., I. Miyagi, et al. (2003). "Selenium regulates transcription factor NF‐

kappaB activation during the acute phase reaction." Clin Chim Acta 334(1‐2):

163‐71.

Mahaffey, K. R. and D. Mergler (1998). "Blood levels of total and organic mercury in

residents of the upper St. Lawrence River basin, Quebec: association with

age, gender, and fish consumption." Environ Res 77(2): 104‐14.

Marchioli, R., C. Schweiger, et al. (2001). "Efficacy of n‐3 polyunsaturated fatty

acids after myocardial infarction: results of GISSI‐Prevenzione trial. Gruppo

Italiano per lo Studio della Sopravvivenza nell'Infarto Miocardico." Lipids 36

Suppl: S119‐26.

Martinez, A., J. L. Santiago, et al. (2008). "Polymorphisms in the selenoprotein S

gene: lack of association with autoimmune inflammatory diseases." BMC

Genomics 9: 329.

Mas, E., R. J. Woodman, et al. (2010). "The omega‐3 fatty acids EPA and DHA

decrease plasma F(2)‐isoprostanes: Results from two placebo‐controlled

interventions." Free Radic Res 44(9): 983‐90.

Masukawa, T., H. Kito, et al. (1982). "Formation and possible role of

bis(methylmercuric) selenide in rats treated with methylmercury and

selenite." Biochem Pharmacol 31(1): 75‐8.

McClung, J. P., C. A. Roneker, et al. (2004). "Development of insulin resistance and

obesity in mice overexpressing cellular glutathione peroxidase." Proc Natl

Acad Sci U S A 101(24): 8852‐7.

McGrath, L. T., G. M. Brennan, et al. (1996). "Effect of dietary fish oil

supplementation on peroxidation of serum lipids in patients with non‐insulin

dependent diabetes mellitus." Atherosclerosis 121(2): 275‐83.

Milne, G. L. and J. D. Morrow (2006). "Isoprostanes and related compounds: update

2006." Antioxid Redox Signal 8(7‐8): 1379‐84.

Milne, G. L., S. C. Sanchez, et al. (2007). "Quantification of F2‐isoprostanes as a

biomarker of oxidative stress." Nat Protoc 2(1): 221‐6.

148

Milne, G. L., H. Yin, et al. (2007). "Quantification of F2‐isoprostanes in biological

fluids and tissues as a measure of oxidant stress." Methods Enzymol 433:

113‐26.

Milne, G. L., H. Yin, et al. (2008). "Human biochemistry of the isoprostane

pathway." J Biol Chem 283(23): 15533‐7.

Milne, G. L., B. Gao, et al. (2012). "Measurement of F(2‐) isoprostanes and isofurans

using gas chromatography‐mass spectrometry." Free Radic Biol Med.

Available online 5 October 2012,

http://www.sciencedirect.com/science/article/pii/S0891584912011641

Minuz, P., P. Patrignani, et al. (2002). "Increased oxidative stress and platelet

activation in patients with hypertension and renovascular disease."

Circulation 106(22): 2800‐5.

Montuschi, P., P. J. Barnes, et al. (2004). "Isoprostanes: markers and mediators of

oxidative stress." FASEB J 18(15): 1791‐800.

Mori, T. A., L. J. Beilin, et al. (1997). "Interactions between dietary fat, fish, and fish

oils and their effects on platelet function in men at risk of cardiovascular

disease." Arterioscler Thromb Vasc Biol 17(2): 279‐86.

Mori, T. A., G. F. Watts, et al. (2000). "Differential effects of eicosapentaenoic acid

and docosahexaenoic acid on vascular reactivity of the forearm

microcirculation in hyperlipidemic, overweight men." Circulation 102(11):

1264‐9.

Mori, T. A. and R. J. Woodman (2006). "The independent effects of

eicosapentaenoic acid and docosahexaenoic acid on cardiovascular risk

factors in humans." Curr Opin Clin Nutr Metab Care 9(2): 95‐104.

Mori, T. A., R. J. Woodman, et al. (2003). "Effect of eicosapentaenoic acid and

docosahexaenoic acid on oxidative stress and inflammatory markers in

treated‐hypertensive type 2 diabetic subjects." Free Radic Biol Med 35(7):

772‐81.

149

Mori, T. A., K. D. Croft, et al. (1999). "An improved method for the measurement of

urinary and plasma F2‐isoprostanes using gas chromatography‐mass

spectrometry." Anal Biochem 268(1): 117‐25.

Morrow, J. D. (2003). "Is oxidant stress a connection between obesity and

atherosclerosis?" Arterioscler Thromb Vasc Biol 23(3): 368‐370.

Morrow, J. D. (2005). "Quantification of isoprostanes as indices of oxidant stress

and the risk of atherosclerosis in humans." Arterioscler Thromb Vasc Biol

25(2): 279‐286.

Morrow, J. D., J. A. Awad, et al. (1992). "Non‐cyclooxygenase‐derived prostanoids

(F2‐isoprostanes) are formed in situ on phospholipids." Proc Natl Acad Sci U

S A 89(22): 10721‐5.

Morrow, J. D., B. Frei, et al. (1995). "Increase in circulating products of lipid

peroxidation (F2‐isoprostanes) in smokers. Smoking as a cause of oxidative

damage." N Engl J Med 332(18): 1198‐203.

Morrow, J. D. and L. J. Roberts, 2nd (1999). "Mass spectrometric quantification of

F2‐isoprostanes in biological fluids and tissues as measure of oxidant stress."

Methods Enzymol 300: 3‐12.

Mosca, L., L. J. Appel, et al. (2004). "Evidence‐based guidelines for cardiovascular

disease prevention in women." J Am Coll Cardiol 43(5): 900‐921.

Mozaffarian, D. (2008). "Fish and n‐3 fatty acids for the prevention of fatal

coronary heart disease and sudden cardiac death." Am J Clin Nutr 87(6):

1991S‐6S.

Mozaffarian, D., A. Ascherio, et al. (2005). "Interplay between different

polyunsaturated fatty acids and risk of coronary heart disease in men."

Circulation 111(2): 157‐64.

Muckle, G., P. Ayotte, et al. (2001). "Determinants of polychlorinated biphenyls and

methylmercury exposure in inuit women of childbearing age." Environ

Health Perspect 109(9): 957‐63.

Muscat, J. E., W. Kleinman, et al. (2004). "Enhanced protein glutathiolation and

oxidative stress in cigarette smokers." Free Radic Biol Med 36(4): 464‐70.

150

Nalsen, C., B. Vessby, et al. (2006). "Dietary (n‐3) fatty acids reduce plasma F2‐

isoprostanes but not prostaglandin F2alpha in healthy humans." J Nutr

136(5): 1222‐8.

National Cholesterol Education Program Expert Panel on Detection, E. and A.

Treatment of High Blood Cholesterol in (2002). "Third Report of the National

Cholesterol Education Program (NCEP) Expert Panel on Detection,

Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult

Treatment Panel III) final report." Circulation 106(25): 3143‐421.

Nawrot, T. S., J. A. Staessen, et al. (2007). "Blood pressure and blood selenium: a

cross‐sectional and longitudinal population study." Eur Heart J 28(5): 628‐

33.

Nestel, P., H. Shige, et al. (2002). "The n‐3 fatty acids eicosapentaenoic acid and

docosahexaenoic acid increase systemic arterial compliance in humans." Am J

Clin Nutr 76(2): 326‐30.

Nobmann, E. D., S. O. Ebbesson, et al. (1999). "Associations between dietary factors

and plasma lipids related to cardiovascular disease among Siberian Yupiks of

Alaska." Int J Circumpolar Health 58(4): 254‐71.

Nobmann, E. D., R. Ponce, et al. (2005). "Dietary intakes vary with age among

Eskimo adults of Northwest Alaska in the GOCADAN study, 2000‐2003." J

Nutr 135(4): 856‐62.

Noel, M., E. Dewailly, et al. "Cardiovascular risk factors and subclinical

atherosclerosis among Nunavik Inuit." Atherosclerosis 221(2): 558‐64.

Nourooz‐Zadeh, J. (2008). "Key issues in F2‐isoprostane analysis." Biochem Soc

Trans 36(Pt 5): 1060‐5.

O'Keefe, J. H., Jr. and W. S. Harris (2000). "From Inuit to implementation: omega‐3

fatty acids come of age." Mayo Clin Proc 75(6): 607‐14.

O'Leary, D. H. and J. F. Polak (2002). "Intima‐media thickness: a tool for

atherosclerosis imaging and event prediction." Am J Cardiol 90(10C): 18L‐

21L.

151

Oka, T., M. Matsukura, et al. (2002). "Autonomic nervous functions in fetal type

Minamata disease patients: assessment of heart rate variability." Tohoku J

Exp Med 198(4): 215‐21.

Ordovas, J. M. (2007). "Gender, a significant factor in the cross talk between genes,

environment, and health." Gend Med 4 Suppl B: S111‐22.

Park, K. and D. Mozaffarian (2010). "Omega‐3 fatty acids, mercury, and selenium in

fish and the risk of cardiovascular diseases." Curr Atheroscler Rep 12(6):

414‐22.

Pedersen, E. B., M. E. Jorgensen, et al. (2005). "Relationship between mercury in

blood and 24‐h ambulatory blood pressure in Greenlanders and Danes." Am J

Hypertens 18(5 Pt 1): 612‐8.

Pilz, H., A. Oguogho, et al. (2000). "Quitting cigarette smoking results in a fast

improvement of in vivo oxidation injury (determined via plasma, serum and

urinary isoprostane)." Thromb Res 99(3): 209‐21.

Pollack, A. Z., E. F. Schisterman, et al. (2012). "Relation of Blood Cadmium, Lead,

and Mercury Levels to Biomarkers of Lipid Peroxidation in Premenopausal

Women." Am J Epidemiol 175(7): 645‐52

Pollex, R. L., H. M. Khan, et al. (2004). "The metabolic syndrome in Inuit." Diabetes

Care 27(6): 1517‐8.

Pou, K. M., J. M. Massaro, et al. (2007). "Visceral and subcutaneous adipose tissue

volumes are cross‐sectionally related to markers of inflammation and

oxidative stress: The Framingham Heart Study." Circulation 116(11): 1234‐

41.

Plaziac, C. (2007), “Tobacco Use.” Nunavik Inuit health Survey 2004, Qanuippitaa?

How are we? Quebec: Institut national de santé publique de Quebec (INSQ) &

Nunavik regional board of health and social services (NRBHSS).

http://www.inspq.qc.ca

Pratico, D., N. P. Murphy, et al. (1997). "Interaction of a thrombin inhibitor and a

platelet GP IIb/IIIa antagonist in vivo: evidence that thrombin mediates

152

platelet aggregation and subsequent thromboxane A2 formation during

coronary thrombolysis." J Pharmacol Exp Ther 281(3): 1178‐85.

Pratico, D., O. P. Barry, et al. (1998). "IPF2alpha‐I: an index of lipid peroxidation in

humans." Proc Natl Acad Sci U S A 95(7): 3449‐54.

Ralston, N. V., J. L. Blackwell, 3rd, et al. (2007). "Importance of molar ratios in

selenium‐dependent protection against methylmercury toxicity." Biol Trace

Elem Res 119(3): 255‐68.

Ralston, N. V. and L. J. Raymond (2010). "Dietary selenium's protective effects

against methylmercury toxicity." Toxicology 278(1): 112‐23.

Rayman, M. P. (2009). "Selenoproteins and human health: insights from

epidemiological data." Biochim Biophys Acta 1790(11): 1533‐40.

Rayman, M. P. (2012). "Selenium and human health." The Lancet 379(9822): 1256‐

1268.

Ridker, P. M., M. J. Stampfer, et al. (2001). "Novel risk factors for systemic

atherosclerosis: a comparison of C‐reactive protein, fibrinogen,

homocysteine, lipoprotein(a), and standard cholesterol screening as

predictors of peripheral arterial disease." JAMA 285(19): 2481‐5.

Rifai, N. and P. M. Ridker (2001). "High‐sensitivity C‐reactive protein: a novel and

promising marker of coronary heart disease." Clin Chem 47(3): 403‐11.

Rissanen, T., S. Voutilainen, et al. (2000). "Fish oil‐derived fatty acids,

docosahexaenoic acid and docosapentaenoic acid, and the risk of acute

coronary events: the Kuopio ischaemic heart disease risk factor study."

Circulation 102(22): 2677‐9.

Roberts, L. J., 2nd and J. P. Fessel (2004). "The biochemistry of the isoprostane,

neuroprostane, and isofuran pathways of lipid peroxidation." Chem Phys

Lipids 128(1‐2): 173‐86.

Roberts, L. J., 2nd, K. P. Moore, et al. (1996). "Identification of the major urinary

metabolite of the F2‐isoprostane 8‐iso‐prostaglandin F2alpha in humans." J

Biol Chem 271(34): 20617‐20.

153

Rossner, P., Jr., M. D. Gammon, et al. (2006). "Relationship between urinary 15‐F2t‐

isoprostane and 8‐oxodeoxyguanosine levels and breast cancer risk." Cancer

Epidemiol Biomarkers Prev 15(4): 639‐44.

Rupp, H., D. Wagner, et al. (2004). "Risk stratification by the "EPA+DHA level" and

the "EPA/AA ratio" focus on anti‐inflammatory and antiarrhythmogenic

effects of long‐chain omega‐3 fatty acids." Herz 29(7): 673‐85.

Rylander, L., P. Nilsson‐Ehle, et al. (2006). "A simplified precise method for

adjusting serum levels of persistent organohalogen pollutants to total serum

lipids." Chemosphere 62(3): 333‐6.

Salonen, J., G. Alfthan, et al. (1982). " Salonen, J., G. Alfthan, et al. (1982).

"Association between cardiovascular death and myocardial infarction and

serum selenium in a matched‐pair longitudinal study." The Lancet

320(8291): 175‐179.

Salonen, J. T., K. Seppanen, et al. (2000). "Mercury accumulation and accelerated

progression of carotid atherosclerosis: a population‐based prospective 4‐

year follow‐up study in men in eastern Finland." Atherosclerosis 148(2):

265‐73.

Salonen, J. T., S. Yla‐Herttuala, et al. (1992). "Autoantibody against oxidised LDL

and progression of carotid atherosclerosis." Lancet 339(8798): 883‐7.

Sarsilmaz, M., A. Songur, et al. (2003). "Potential role of dietary omega‐3 essential

fatty acids on some oxidant/antioxidant parameters in rats' corpus striatum."

Prostaglandins Leukot Essent Fatty Acids 69(4): 253‐9.

Schisterman, E. F., A. J. Gaskins, et al. (2010) "Influence of endogenous

reproductive hormones on F2‐isoprostane levels in premenopausal women:

the BioCycle Study." Am J Epidemiol 172(4): 430‐9.

Schoenmakers, E., M. Agostini, et al. (2011). "Mutations in the selenocysteine

insertion sequence‐binding protein 2 gene lead to a multisystem

selenoprotein deficiency disorder in humans." J Clin Invest 120(12): 4220‐35.

154

Schwedhelm, E., R. Maas, et al. (2003). "Clinical pharmacokinetics of antioxidants

and their impact on systemic oxidative stress." Clin Pharmacokinet 42(5):

437‐59.

Seppanen, K., P. Soininen, et al. (2004). "Does mercury promote lipid peroxidation?

An in vitro study concerning mercury, copper, and iron in peroxidation of

low‐density lipoprotein." Biol Trace Elem Res 101(2): 117‐32.

Serhan, C. N., A. Jain, et al. (2003). "Reduced inflammation and tissue damage in

transgenic rabbits overexpressing 15‐lipoxygenase and endogenous anti‐

inflammatory lipid mediators." J Immunol 171(12): 6856‐65.

Serhan, C. N. and B. Levy (2003). "Novel pathways and endogenous mediators in

anti‐inflammation and resolution." Chem Immunol Allergy 83: 115‐45.

Sethi, S. (2002). "Inhibition of leukocyte‐endothelial interactions by oxidized

omega‐3 fatty acids: a novel mechanism for the anti‐inflammatory effects of

omega‐3 fatty acids in fish oil." Redox Rep 7(6): 369‐78.

Sethi, S., A. Y. Eastman, et al. (1996). "Inhibition of phagocyte‐endothelium

interactions by oxidized fatty acids: a natural anti‐inflammatory

mechanism?" J Lab Clin Med 128(1): 27‐38.

Sharma, S., J. Gittelsohn, et al. "Addressing the public health burden caused by the

nutrition transition through the Healthy Foods North nutrition and lifestyle

intervention programme." J Hum Nutr Diet 23 Suppl 1: 120‐7.

Sinaiko, A. R., J. Steinberger, et al. (2005). "Relation of body mass index and insulin

resistance to cardiovascular risk factors, inflammatory factors, and oxidative

stress during adolescence." Circulation 111(15): 1985‐91.

Singh, U. and I. Jialal (2006). "Oxidative stress and atherosclerosis."

Pathophysiology 13(3): 129‐142.

Sioen, I., J. Van Camp et al. 2008 “Probabilistic intake assessment of multiple

compounds as a tool to quantify the nutritional‐toxicological conflict related

to seafood consumption.” Chemosphere 71(6): 1056–66

155

Siscovick, D. S., T. E. Raghunathan, et al. (1995). "Dietary intake and cell membrane

levels of long‐chain n‐3 polyunsaturated fatty acids and the risk of primary

cardiac arrest." JAMA 274(17): 1363‐7.

Sjogren, P., S. Basu, et al. (2005). "Measures of oxidized low‐density lipoprotein

and oxidative stress are not related and not elevated in otherwise healthy

men with the metabolic syndrome." Arterioscler Thromb Vasc Biol 25(12):

2580‐6.

Slotnick, M. J. and J. O. Nriagu (2006). "Validity of human nails as a biomarker of

arsenic and selenium exposure: A review." Environ Res 102(1): 125‐39.

Song, W. L., G. Paschos, et al. (2009). "Novel eicosapentaenoic acid‐derived F3‐

isoprostanes as biomarkers of lipid peroxidation." J Biol Chem 284(35):

23636‐43.

Sørensen, L. T., S. Jõrgensen, et al. (2009). "Acute Effects of Nicotine and Smoking

on Blood Flow, Tissue Oxygen, and Aerobe Metabolism of the Skin and

Subcutis." J Surg Res 152(2): 224‐230.

Sowers, M., D. McConnell, et al. (2008). "Oestrogen metabolites in relation to

isoprostanes as a measure of oxidative stress." Clin Endocrinol (Oxf) 68(5):

806‐13.

Stafforini, D. M., J. R. Sheller, et al. (2006). "Release of free F2‐isoprostanes from

esterified phospholipids is catalyzed by intracellular and plasma platelet‐

activating factor acetylhydrolases." J Biol Chem 281(8): 4616‐23.

Stapleton, S. R. (2000). "Selenium: an insulin‐mimetic." Cell Mol Life Sci 57(13‐14):

1874‐9.

Steel, D. M. and A. S. Whitehead (1994). "The major acute phase reactants: C‐

reactive protein, serum amyloid P component and serum amyloid A protein."

Immunol Today 15(2): 81‐8.

Stephens, J. W., M. P. Khanolkar, et al. (2009). "The biological relevance and

measurement of plasma markers of oxidative stress in diabetes and

cardiovascular disease." Atherosclerosis 202(2): 321‐329.

156

Stocker, R. and J. F. Keaney, Jr. (2004). "Role of oxidative modifications in

atherosclerosis." Physiol Rev 84(4): 1381‐478.

Stranges, S., J. R. Marshall, et al. (2007). "Effects of long‐term selenium

supplementation on the incidence of type 2 diabetes: a randomized trial."

Ann Intern Med 147(4): 217‐23.

Stringer, M. D., P. G. Gorog, et al. (1989). "Lipid peroxides and atherosclerosis." BMJ

298(6669): 281‐4.

Suadicani, P., H. O. Hein, et al. (1992). "Serum selenium concentration and risk of

ischaemic heart disease in a prospective cohort study of 3000 males."


Szmitko, P. E., C.‐H. Wang, et al. (2003). "New Markers of Inflammation and

Endothelial Cell Activation." Circulation 108(16): 1917‐1923.

Tapiero, H., D. M. Townsend, et al. (2003). "The antioxidant role of selenium and

seleno‐compounds." Biomed Pharmacother 57(3‐4): 134‐44.

Thies, F., E. A. Miles, et al. (2001). "Influence of dietary supplementation with long‐

chain n‐3 or n‐6 polyunsaturated fatty acids on blood inflammatory cell

populations and functions and on plasma soluble adhesion molecules in

healthy adults." Lipids 36(11): 1183‐93.

Traulsen, H., H. Steinbrenner, et al. (2004). "Selenoprotein P protects low‐density

lipoprotein against oxidation." Free Radic Res 38(2): 123‐8.

Twaroski, T. P., M. L. O'Brien, et al. (2001). "Effects of selected polychlorinated

biphenyl (PCB) congeners on hepatic glutathione, glutathione‐related

enzymes, and selenium status: implications for oxidative stress." Biochem

Pharmacol 62(3): 273‐81.

Uno, K. and S. J. Nicholls (2010). "Biomarkers of inflammation and oxidative stress

in atherosclerosis." Biomarkers in Medicine 4(3): 361‐373.

Uthus, E. O., K. Yokoi, et al. (2002). "Selenium deficiency in Fisher‐344 rats

decreases plasma and tissue homocysteine concentrations and alters plasma

homocysteine and cysteine redox status." J Nutr 132(6): 1122‐8.

157

Vaclavik, E., A. Tjonneland, et al. (2006). "Organochlorines in Danish women:

predictors of adipose tissue concentrations." Environ Res 100(3): 362‐70.

Valera, B., E. Dewailly, et al. (2008). "Cardiac autonomic activity and blood

pressure among Nunavik Inuit adults exposed to environmental mercury: a

cross‐sectional study." Environ Health 7: 29.

Valera, B., E. Dewailly, et al. (2009). "Environmental mercury exposure and blood

pressure among Nunavik Inuit adults." Hypertension 54(5): 981‐6.

Vallve, J. C., K. Uliaque, et al. (2002). "Unsaturated fatty acids and their oxidation

products stimulate CD36 gene expression in human macrophages."


Van Oostdam, J., S. G. Donaldson, et al. (2005). "Human health implications of

environmental contaminants in Arctic Canada: A review." Sci Total Environ

351‐352: 165‐246.

Venugopal, S. K., S. Devaraj, et al. (2002). "Demonstration that C‐reactive protein

decreases eNOS expression and bioactivity in human aortic endothelial cells."

Circulation 106(12): 1439‐41.

Verma, S., C. H. Wang, et al. (2003). "Hyperglycemia potentiates the proatherogenic

effects of C‐reactive protein: reversal with rosiglitazone." J Mol Cell Cardiol

35(4): 417‐9.

Virtamo, J., E. Valkeila, et al. (1985). "Serum selenium and the risk of coronary

heart disease and stroke." Am J Epidemiol 122(2): 276‐82.

Virtanen, J. K., T. H. Rissanen, et al. (2007). "Mercury as a risk factor for

cardiovascular diseases." J Nutr Biochem 18(2): 75‐85.

Visioli, F., P. Rise, et al. (2003). "Dietary intake of fish vs. formulations leads to

higher plasma concentrations of n‐3 fatty acids." Lipids 38(4): 415‐8.

Vunta, H., B. J. Belda, et al. (2008). "Selenium attenuates pro‐inflammatory gene

expression in macrophages." Mol Nutr Food Res 52(11): 1316‐23.

Wada, M., C. J. DeLong, et al. (2007). "Enzymes and receptors of prostaglandin

pathways with arachidonic acid‐derived versus eicosapentaenoic acid‐

derived substrates and products." J Biol Chem 282(31): 22254‐66.

158

Wang, C. H., S. H. Li, et al. (2003). "C‐reactive protein upregulates angiotensin type

1 receptors in vascular smooth muscle." Circulation 107(13): 1783‐90.

Wang, L., M. Fuster, et al. (2005). "Endothelial heparan sulfate deficiency impairs

L‐selectin‐ and chemokine‐mediated neutrophil trafficking during

inflammatory responses." Nat Immunol 6(9): 902‐10.

Wang, Y. Z., X. A. Jia, et al. (1993). "Effects of selenium deficiency on Ca transport

function of sarcoplasmic reticulum and lipid peroxidation in rat

myocardium." Biol Trace Elem Res 36(2): 159‐66.

Ward, N. C., J. M. Hodgson, et al. (2004). "Oxidative stress in human hypertension:

association with antihypertensive treatment, gender, nutrition, and lifestyle."

Free Radic Biol Med 36(2): 226‐232.

Ward, N. C., I. B. Puddey, et al. (2005). "Urinary 20‐hydroxyeicosatetraenoic acid

excretion is associated with oxidative stress in hypertensive subjects." Free

Radic Biol Med 38(8): 1032‐6.

Whincup, P. H., J. A. Gilg, et al. (2004). "Passive smoking and risk of coronary heart

disease and stroke: prospective study with cotinine measurement." BMJ

329(7459): 200‐5.

Witte, K. K., A. L. Clark, et al. (2001). "Chronic heart failure and micronutrients." J

Am Coll Cardiol 37(7): 1765‐74.

Wohlin, M., J. Helmersson, et al. (2007). "Both cyclooxygenase‐ and cytokine‐

mediated inflammation are associated with carotid intima‐media thickness."

Cytokine 38(3): 130‐6.

Wongcharoen, W. and N. Chattipakorn (2005). "Antiarrhythmic effects of n‐3

polyunsaturated fatty acids." Asia Pac J Clin Nutr 14(4): 307‐12.

Woodman, R. J., T. A. Mori, et al. (2003). "Effects of purified eicosapentaenoic acid

and docosahexaenoic acid on platelet, fibrinolytic and vascular function in

hypertensive type 2 diabetic patients." Atherosclerosis 166(1): 85‐93.

Woodman, R. J., T. A. Mori, et al. (2002). "Effects of purified eicosapentaenoic and

docosahexaenoic acids on glycemic control, blood pressure, and serum lipids

159

in type 2 diabetic patients with treated hypertension." Am J Clin Nutr 76(5):

1007‐15.

Woodman, R. J., T. A. Mori, et al. (2003). "Docosahexaenoic acid but not

eicosapentaenoic acid increases LDL particle size in treated hypertensive

type 2 diabetic patients." Diabetes Care 26(1): 253.

Woodward, M., A. Rumley, et al. (2003). "C‐reactive protein: associations with

haematological variables, cardiovascular risk factors and prevalent

cardiovascular disease." Br J Haematol 122(1): 135‐41.

Yagi, K. (1997). "Female hormones act as natural antioxidants‐‐a survey of our

research." Acta Biochim Pol 44(4): 701‐9.

Yamamoto, H., T. Endo, et al. (1995). "Activation of phospholipase D by

prostaglandin F2 alpha in rat luteal cells and effects of inhibitors of

arachidonic acid metabolism." Prostaglandins 50(4): 201‐11.

Yan, W., G. D. Byrd, et al. (2007). "Quantitation of isoprostane isomers in human

urine from smokers and nonsmokers by LC‐MS/MS." J Lipid Res 48(7): 1607‐

17.

Yang, P., D. Chan, et al. (2004). "Formation and antiproliferative effect of

prostaglandin E(3) from eicosapentaenoic acid in human lung cancer cells." J

Lipid Res 45(6): 1030‐9.

Yin, H., L. Gao, et al. (2007). "Urinary prostaglandin F2PG α is generated from the

isoprostane pathway and not the cyclooxygenase in humans." J Biol Chem

282(1): 329‐336.

Yin, H., W. Liu, et al. (2009). "Dietary supplementation of omega‐3 fatty acid‐

containing fish oil suppresses F2‐isoprostanes but enhances inflammatory

cytokine response in a mouse model of ovalbumin‐induced allergic lung

inflammation." Free Radic Biol Med 47(5): 622‐8.

Yoneda, S. and K. T. Suzuki (1997). "Equimolar Hg‐Se complex binds to

selenoprotein P." Biochem Biophys Res Commun 231(1): 7‐11.

160

Yoshida, Y., M. Hayakawa, et al. (2007). "Evaluation of lipophilic antioxidant

efficacy in vivo by the biomarkers hydroxyoctadecadienoic acid and

isoprostane." Lipids 42(5): 463‐72.

Yoshizawa, K., E. B. Rimm, et al. (2002). "Mercury and the risk of coronary heart

disease in men." N Engl J Med 347(22): 1755‐60.

Young, T. K. (2007). "Are the circumpolar Inuit becoming obese?" Am J Hum Biol

19(2): 181‐9.

Young, T. K., P. Bjerregaard, et al. (2007). "Prevalence of obesity and its metabolic

correlates among the circumpolar inuit in 3 countries." Am J Public Health

97(4): 691‐5.

Young, T. K., J. M. Gerrard, et al. (1999). "Plasma phospholipid fatty acids in the

central Canadian arctic: biocultural explanations for ethnic differences." Am J

Phys Anthropol 109(1): 9‐18.

Yudkin, J. S., C. D. Stehouwer, et al. (1999). "C‐reactive protein in healthy subjects:

associations with obesity, insulin resistance, and endothelial dysfunction: a

potential role for cytokines originating from adipose tissue?" Arterioscler


Zamamiri‐Davis, F., Y. Lu, et al. (2002). "Nuclear factor‐kappaB mediates over‐

expression of cyclooxygenase‐2 during activation of RAW 264.7 macrophages

in selenium deficiency." Free Radic Biol Med 32(9): 890‐7.

Zhou, Y. E., S. Kubow, et al. (2011). "Highly unsaturated n‐3 fatty acids status of

Canadian Inuit: International Polar Year Inuit Health Survey, 2007‐2008." Int

J Circumpolar Health 70(5): 498‐510.

Oxidative Stress as a Cardiovascular Risk Factor in...

Documents

Transcript of Oxidative Stress as a Cardiovascular Risk Factor in...