fatty acid desaturase activities in metabolic syndrome and

ACTA

UNIVERSITATIS

UPSALIENSIS

UPPSALA

2007

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 292

Fatty Acid Desaturase Activities inMetabolic Syndrome andCardiovascular Disease

Special Reference to Stearoyl-CoA-Desaturase andBiomarkers of Dietary Fat

EVA WARENSJÖ

ISSN 1651-6206ISBN 978-91-554-7025-8urn:nbn:se:uu:diva-8312

�� ! "� �� #�� " �� " $�� %��" &��'( )�� *�� +#��(

��

,��-.� +( ��( �� /�� /�� &�� 0�� 1�� ( 0�� 2�"�� 0��31/3�� 4��5�� "�� ( /�� ( �� ( �� ( ��( �046 ��73� 3!!83��!37(

)�� " �� %&��0' �� ##�� "�� 9�� " �� "��( )�� /�� " �� 5�� " �� "�� 9��( �/� ��*�� #�� :�� #�� :�� #�� ( )�� 0��31/3�� %01�'�;<3�� %�<�' �� ;!3�� %�!�' �� #�� " �� 3�3�� / ��(� �� *� �� #� 01�� <� �� * �!� ��

�� &��0 �� *�� # � ��" ==(� ��( )�� *�� !� �� &��0 *�� " ��"�� 4&��*�� *�� 01�� <� �� &��0 *�� "�� 4&�( 0�� " �� % <��'� �� % <� ' �� 3>3�� *�� #�� " �� %?/' �*�� *� �� &��0(�� ?/ *�� *�� 3>3�� *�� *�� ( ,� �� "�� @��A � �� / �� %��# �� ' �� *�� *�� &��0� �� ( ,� �� 01� �� B <� C <��D ��#�� "�� 5�� " �� "�� ,�� #�3"�� ( �� #�� 01� #��*�� 5�� #�� *�� 01� ��"��()�� ##�� "�� 9�� #��

�� " �� #��(

� �� "�� 4��5�� &�� 0�� &�� E�� 0�� +��#�� +�� 0��31/3��3<3�� 3!3�� 0�#�� 6�� $�� 2�� 0�� "�� 01�

�� !" � �� #�$�� % �� " �� &�� $��" �� #��'" �� " ��()*+,* ��" ��

F +�� ,��-. ��

�006 <! 3<��<�046 ��73� 3!!83��!37��37= � %��CC��(5�(��C��G��H��37= �'

For Mira, Melker and Fredrik �

List of papers:

I. Warensjö E, Risérus U, Vessby B. Fatty acid composition of

serum lipids predicts the development of the metabolic syn-drome in men. Diabetologia. 2005; 48:1999-2005.*

II. Warensjö E, Ingelsson E, Lundmark E, Lannfelt L, Syvänen

AC, Vessby B, Risérus U. Polymorphisms in the SCD gene: associations with body fat distribution and insulin sensitivity. Obesity (Silver Spring). 2007 Jul;15(7):1732-40 †

III. Warensjö E, Sundström J, Vessby B, Cederholm T, Risérus

U. Serum fatty acid composition as a predictor of total and car-diovascular mortality: A population based prospective study in men. Submitted manuscript

IV. Warensjö E , Risérus U , Inga-Britt Gustafsson IB, Mohsen

R, Cederholm T, Vessby B. Effects of saturated and unsatu-rated fatty acids on estimated desaturase activities during a con-trolled dietary intervention. Submitted manuscript

Reprints were made with the permission of the publishers. * ©Springer-Verlag † ©NAASO, The Obesity Society Opponent: Associate Professor David Laaksonen, Department of Medicine, Kuopio University Hospital Supervisors: PhD Ulf Risérus, Professor emeritus Bengt Vessby and Pro-fessor Tommy Cederholm, Clinical Nutrition and Metabolism, Department of Public Health and Caring Sciences, Uppsala University

The lipoprotein particle on the front cover was reprinted with the permission from PeproTech Inc..

Contents

Introduction...................................................................................................11 Dietary fatty acids ....................................................................................11 Dietary fatty acids and cardiovascular disease .......................................13 Fatty acids as biomarkers ........................................................................14 Desaturases and elongases ......................................................................15

Stearoyl-CoA-desaturase .....................................................................16 Biomarker fatty acids, estimated desaturase activities and metabolic and cardiovascular diseases ...........................................................................17 Obesity and insulin resistance..................................................................17 Metabolic syndrome .................................................................................18 Genetic factors and variation...................................................................19

Aims..............................................................................................................21

Subjects and Methods ...................................................................................22 Study participants.....................................................................................22 Estimation of desaturase activities ...........................................................23 The ULSAM- cohort ................................................................................23 Methods in ULSAM-based studies ..........................................................24

Baseline investigations at 50 years ......................................................24 Investigations at 70 years.....................................................................25 Statistical Analysis ..............................................................................27 Study I..................................................................................................27 Study II ................................................................................................28 Study III...............................................................................................29

Study IV; Study subjects, design, diets and methods...............................31 Ethics........................................................................................................32

Results and discussion ..................................................................................33 Study I..................................................................................................33 Study II ................................................................................................36 Study III...............................................................................................40 Study IV...............................................................................................43

General Discussion .......................................................................................46 Biomarker fatty acids and metabolic and cardiovascular diseases .........46 Is the SCD [16:1/16:0] ratio a potential marker of saturated fat intake?47 Should SCD be estimated as 16:1/16:0 or 18:1/18:0?.............................48 Genetic variations in the SCD1-gene .......................................................48 Changes in desaturase activities, Cause or consequence? ......................49 The D6D/D5D pathway and eicosanoid signaling...................................50 How do estimated enzyme activities in plasma reflect corresponding activities in other tissues? ........................................................................51 Estimated desaturase activities in different serum lipid fractions ...........52 Are estimated desaturase activities valid markers of desaturase activities?..................................................................................................................53 Clinical utility and implications for public health ...................................54 Epidemiological causation and considerations .......................................54 Strengths and limitations..........................................................................55 Future directions ......................................................................................56

Conclusions...................................................................................................58

Sammanfattning på svenska (Summary in Swedish) ....................................59

Acknowledgements.......................................................................................61

References.....................................................................................................63

Abbreviations

CVD Cardiovascular disease TAG Triacylglycerol FA Fatty acid SFA Saturated fatty acid MUFA Monounsaturated fatty acid PUFA Polyunsaturated fatty acid LA Linoleic acid ALA �-linolenic acid TFA Trans fatty acids CLA Conjugated Linoleic acids CHD Coronary heart disease OA Oleic acid LDL Low density lipoprotein HDL High density lipoprotein EPA Eicosapentaenoic acid DHA Docosahexaenoic acid NEFA Non esterified fatty acid CE Cholesteryl esters PL Phospholipids AA Arachidonic acid SCD Stearoyl-CoenzymA-desaturase D6D �6 desaturase D5D �5 desaturase SREBP-1c Sterol regulatory element binding protein-1c PPAR Peroxisome proliferator activated receptors DHLA Dihomo-�-linolenic acid MetS Metabolic syndrome BMI Body mass index SNP Single nucleotide polymorphism WC Waist circumference RO-diet Rapeseed oil rich diet SAT-diet Butter rich diet HOMA Homeostasis model insulin resistance MAF Minor allele frequency HWE Hardy-Weinberg equilibrium

11

Introduction

Today the world is facing an obesity epidemic; over 1 billion adults are over-weight and over 300 million of them are clinically obese. Many of these people are prone to suffer from obesity related metabolic diseases such as cardiovascular disease (CVD) and type 2 diabetes 1, 2 and this is associated with premature death 3, 4.

Fat is the most energy dense nutrient and provides a large portion of the total energy intake for most individuals. Fat is also needed for cell structure and function, intercellular communication and genetic transcription 5. The development of obesity related metabolic diseases is influenced by the quan-tity and more so by the quality of fat in the diet 6, 7.

Dietary fatty acids Dietary fat is largely made up by triacylglycerol (TAG, “triglycerides”) con-sisting of three individual fatty acids (FAs), each linked by an ester bond to a glycerol backbone 8. Individual FAs have different biological effects and physical properties due to chain length, the degree of saturation and isomeric form. When the quality (type) of dietary fat is discussed it is the FA compo-sition of the fat that is considered. Most dietary FAs are uncomplicated in structure with a carboxyl group at one end and a methyl group at the other end of the carbon chain, but some FAs are branched 9. The number of carbon atoms varies between 4 and 24 and are usually even numbered 5. The most abundant FAs in the diet have 16 or 18 carbon atoms 8 and oleic acid (18:1), palmitic acid (16:0), linoleic acid (18:2 n-6) and stearic acid (18:0) are by quantity the most important individual FA in the Swedish diet 10. There are dietary FAs with an odd number of carbon atoms, e.g. pentadecanoic (15:0) and heptadecanoic acids (17:0), which are of ruminant origin 11 and mostly found in dairy products.

There are three major groups of FA; saturated fatty acids (SFA), monoun-saturated fatty acids (MUFA) and n-3 and n-6 polyunsaturated fatty acids (PUFA). The more double bonds a FA chain contains, the more unsaturated the FA is. More double bonds give the FA chain less regular shape, affecting the melting point. This gives unsaturated FA the ability to regulate metabolic processes in the cells; e.g. muscle cells with more unsaturated FAs in their membranes respond better to insulin 8 since the membrane is more “fluid”.

12

Two dietary PUFAs, linoleic acid (LA, 18:2 n-6) and �-linolenic acid (ALA, 18:3 n-3) cannot be endogenously synthesized in humans and are thus es-sential and must be provided with the diet. LA and ALA are precursors for eicosanoids, prostaglandins and leukotrienes, which are fast-acting effective and locally synthesized mediators which can acts as second messengers 12. Trans-fatty acids Lately much attention has been given to trans-fatty acids (TFA), which are unsaturated fatty acids with the double bond in trans-configuration, instead of the cis-configuration (Figure 1). Most TFAs are formed during the indus-trial process of partial hydrogenation when liquid fats are hardened and some TFAs are naturally formed in the rumen of dairy cows by an enzymatic reac-tion. TFAs stay solid even at room temperature and are more resistant to oxidation and spoilage. Natural TFAs are found in small amounts in some plants (pomegranates, peas and cabbage) 13. A special group of TFAs is the conjugated linoleic acids (CLA) 14.

Figure 1. Different fatty acids with 18 carbon atoms; stearic acid, oleic acid and elaidic acid. Dietary n-6/n-3 ratio There is a concern regarding a high dietary n-6/n-3 ratio, especially in the United States. A too high LA intake may lead to increased AA synthesis which is a precursor for pro-inflammatory and pro-thrombotic factors that may promote plaque formation. In the United States, this ratio is 10-12:1 15 but the recommendation in the Nordic Nutrition Recommendations 2004 is

13

to keep this ratio between 3-9:1, but there is no consensus about the optimal n-6/n-3 ratio in the diet. The n-6/n-3 ratio in the Swedish diet is 5:116.

Dietary fatty acids and cardiovascular disease A positive association between a diet high in SFAs and a negative associa-tion between a diet high in unsaturated fat, respectively, and the risk of CVD is documented 17, but not undisputed. In prospective studies within popula-tions, it has often been difficult to demonstrate a positive relationship be-tween intakes of SFA and coronary heart disease (CHD) 12, 18, 19, but cross-sectional studies have in general demonstrated a positive relationship be-tween intake of SFA and CHD 20. Mechanisms that link dietary fatty acids and CVD have been related to effects on blood cholesterol levels 20-22, insu-lin resistance23, inflammation and endothelial dysfunction 24.

SFAs are known to elevate low density lipoprotein (LDL) cholesterol, compared to carbohydrate. Myristic acid (14:0) is the most cholesterol-raising FA followed by lauric (12:0) and palmitic (16:0) acids. Stearic acid (18:0) is however neutral with respect to blood cholesterol levels 21, 25. LA is the most potent cholesterol lowering FA 15 and MUFAs, mainly oleic acid (OA, 18:1) are known to lower LDL-cholesterol levels 12.

MUFAs may induce anti-inflammation 12 and a MUFA rich diet im-proved insulin sensitivity, besides lowering plasma cholesterol and blood pressure in the KANWU-study 26-28. Substitution of saturated fat for unsatu-rated fat, mainly LA, reduced coronary heart disease events in several clini-cal studies (e.g. Finnish Mental Hospital Study 29 and Los Angeles Veteran Hospital Study 30) but addition of a large amount of LA in a diet low in satu-rated fat did not prevent, in primary prevention, cardiovascular or all-cause mortality in the Minnesota Coronary Survey 31 and some other studies (Re-viewed in 12, 18). CVD mortality was considerable lower with a higher intake of PUFA and LA after 15 years of follow-up in a population based cohort study in Finland 32.

N-3 PUFAs are known to lower TAG levels, to induce anti-inflammation and may improve endothelial dysfunction, but are known to slightly elevate LDL-cholesterol 33-35. The effect of n-3 FA on insulin sensitivity is uncertain 36. A cardioprotective effect of n-3 FA from fish exists, especially for fatal events, but the relation to non-fatal events is not entirely clear 37, 38. ALA is probably not as potent as n-3 fish FAs, eicosapentaenoic (EPA) and docosa-hexaenoic (DHA) to reduce cardiovascular risk 34. However, ALA as part of a Mediterranean style diet, significantly reduced cardiac events in the Lyon Diet heart Study 39 and in the Indo-Mediterranean Diet Heart Study 40. In the GISSI and DART trials n-3 fish FA as supplements or as part of a diet de-creased cardiovascular mortality 18, 41, 42.

14

TFAs raise LDL-cholesterol, reduce high density lipoprotein (HDL) cho-lesterol and promote inflammation and endothelial dysfunction 43 and may increase the risk of CHD 44.

Fatty acids as biomarkers It is known that dietary surveys often fail to adequately measure the in-

take of food and nutrients due to reporting bias and there is especially a problem with underreporting 45. It is recognized that it is particularly difficult to assess the intake of fatty foods with traditional methods and it is known that the degree of underreporting of especially dietary fat varies with an in-creasing BMI among the reporters 46-49. Instead, the use of biomarkers may provide an objective and more accurate way of estimating dietary intakes 5, 50.

FAs are transported in the blood in two ways; either as non esterified FA (NEFA) bound to albumin or in more complex structures known as lipopro-teins (see front cover). Lipoproteins are made up by TAG, cholesteryl esters (CE), phospholipids (PL), cholesterol and apolipoproteins 8.

The FA composition of the different fractions of the lipoprotein particle e.g. PL and CE, in platelets and erythrocyte membranes or other body

Figure 2. Endogenous fatty acid synthesis

15

tissues, such as skeletal muscle and adipose tissue, can be used as bio-markers of dietary fat quality. This composition reflects the FA composition (fat quality) of the diet at different time points after eating. The FA composi-tion in serum PL and CE reflects the habitual diet during the last days and weeks, respectively, while FAs of erythrocyte membranes and adipose tissue reflect the intake of those FA during the last months and year(s), respectively 5, 50-54.

FAs can be endogenously synthesized, elongated, desaturated or oxidized, which complicates the use of FA as biomarkers of dietary fat quality. SFAs can be synthesized de novo from carbohydrates, but this synthesis is rare in people who consume a Western diet (high fat diet) 55.

The correlation between dietary FA and plasma FA varies 56 and the best biomarkers are FA that cannot be endogenously synthesized i.e. LA, ALA and TFA 5, 50 . Tissue proportions of EPA and DHA may serve as biomarkers of fish intake 50 and the proportions of 15:0 and 17:0 in serum and adipose tissue have been validated as reliable biomarkers of milk fat 57-59.

FA composition in tissues and lipid fractions is influenced by genetic variation, age, sex, smoking, physical activity 50 and transport processes since different FAs compete for the incorporation in different lipid classes 60. Arachidonic acid (20:4 n-6, AA) is formed from LA and is the most impor-tant precursor for eicosanoids and is therefore tightly regulated in different membranes 5.

Desaturases and elongases The degree of unsaturation of a FA is a major determinant of the melting point and thus of the fluidity of biological membranes. The enzymes respon-sible for the unsaturation are the delta (�) desaturases. These enzymes are bound to the endoplasmic reticulum and introduce double bonds in specific positions of long chain FAs. Three �-desaturases are important for human FA metabolism; �9-desaturase also referred to as stearoyl-CoA-desaturase (SCD), �6-desaturase (D6D) and �5-desaturase (D5D). Elongases are the products of the Elovl1-6 (Elongation-of-very-long-chain-fatty-acids) genes. Elongases and desaturases work in concert to form long chain FAs in the endoplasmic reticulum 61, 62. The desaturation steps are believed to be rate limiting in the formation of FAs.

SCD introduces the first double bond at the 9, 10 position from the car-boxyl end of the SFA and represent the last step in the formation of MUFAs. SCD catalyzes the desaturation of FAs with 12 to 19 carbon atoms 63 and the preferential substrate is 18:0 64. D5D and D6D catalyze the synthesis of n-6 and n-3 long chain PUFAs. MUFAs are incorporated in membrane PL, adi-pose tissue TAG and CE. In addition, MUFAs can act as mediators in signal transduction and cellular differentiation 63, 65, 66. PUFAs are incorporated into

16

membrane PL, but are also needed for eicosanoid signaling, pinocytosis, ion channel modulation and regulation of gene expression 63.

PUFAs regulate desaturases via transcription factors that bind in the pro-moter regions of the different desaturase genes. Sterol regulatory element binding protein-1c (SREBP-1c) and peroxisome proliferator activated recep-tors (PPARs) are thought to play key roles in this regulation 63. Moreover, in a recent study human subcutaneous adipose tissue SCD, D6D and D5D tran-scripts were significantly reduced in response to acute weight loss in moder-ately overweight men 67 .

To study desaturase activities directly in sub-cellular microsomal (i.e. en-doplasmic reticulum) fractions is complicated in humans and not practicable in clinical studies. To measure mRNA/protein expression of desaturases raise ethical concerns in humans, since the most relevant tissue to study is the liver. Surrogate measures of desaturase activities, estimated as FA ratios in serum CE and PL, have instead been used in the studies of this thesis.

LA and ALA compete for the same elongases and desaturases. Although elongases and desaturases have a preference for n-3 FA, a high dietary intake of LA may lead to a relative deficiency of ALA and may also reduce the endogenous conversion of ALA to long chain n-3 FA68, 69.

Stearoyl-CoA-desaturase The SCD-gene is situated on chromosome 10 and two iso-forms of the hu-man SCD gene exists, SCD1 and SCD5. SCD1 is primarily expressed in the liver and adipose tissue, whereas SCD5 mainly is expressed in brain and pancreas. SCD1 share 85 % homology with murine SCD genes 70. SCD1 plays a major role in de novo synthesis of TAG, CE and wax esters and is needed for normal function of skin and eyelid 71. SCD is highly regulated and glucose, fructose, cholesterol, insulin, and estrogen are inducers, while PUFAs, alcohol and leptin are inhibitors of SCD-activity 72.

SCD has been proposed as a future therapeutic target for obesity and re-lated morbidities 73, 74 . It is known that SCD knock-out mice have better insulin sensitivity, higher energy metabolism and are leaner than their wild-type littermates 75. In addition, genes involved in �-oxidation are up regu-lated and genes involved in lipid synthesis are down regulated in the knock-out mice 76. Moreover, mice lacking the SCD gene are deficient in hepatic TAG and CEs despite the presence of other enzymes responsible for the synthesis of TAG and CEs and supplementation with dietary 18:1 and 16:1 does not restore lipid levels 77. Attie et al reported that human plasma TAG levels were closely correlated to estimated plasma SCD activity (desatura-tion index). In addition, the desaturation index was validated in various mice strains as an in vivo marker of SCD activity in this study 78. The relative mRNA expression level of SCD1 was three-fold in skeletal muscle from extremely obese compared to lean subjects and this increased mRNA ex-

17

pression corresponded to changes in estimated SCD activity 79. Risérus et al reported that increased mRNA expression of SCD1 adipose tissue in humans corresponded to increased estimated SCD activity measured in serum 80. A recent study reported that SCD1 mediates prolipogenic effects of dietary saturated fat in mice 81.

Biomarker fatty acids, estimated desaturase activities and metabolic and cardiovascular diseases FAs in serum CE and PL can be used as biomarkers of dietary fat quality and FA ratios as indicators of endogenous desaturation. Hence it is possible to use biomarker FAs and FA ratios (SCD, D5D and D6D) as the exposure variable in epidemiological studies of the association between dietary fat and disease outcome. A serum lipid FA composition characterized by high pro-portions of palmitic (16:0), palmitoleic (16:1) and dihomo-�-linolenic (20:3 n-6, DHLA) acids and a low proportion of LA, has in epidemiologic studies been related to obesity 82, insulin resistance 83, CVD and type 2 diabetes 84-86 and to individual components of the metabolic syndrome (MetS) 87. In a Finnish study, high serum proportions of LA decreased the risk of fasting glycemia and type 2 diabetes in middle-aged men 88. Harris et al reported in a recent meta analysis depressed levels of long chain n-3 FA (especially DHA) as a consistent marker of increased CHD risk 37. Moreover, estimated desaturase activities; higher SCD and D6D and lower D5D have been related to metabolic and cardiovascular diseases 89. In a lifestyle intervention study, an increase in estimated D5D and a decrease in estimated SCD and D6D activities were associated with an increase in insulin sensitivity, independent of lifestyle changes 90. Further, the estimated SCD-ratio [16:1/16:0], as a measure of dietary saturated fat 89, was established as an independent predic-tor of directly measured insulin sensitivity over 20 years 91.

Obesity and insulin resistance Obesity is a major risk factor for metabolic and cardiovascular diseases such as CVD and type 2 diabetes and is defined as the ratio between weight (kg) and height squared (m2), referred to as body mass index (BMI). Overweight is defined in subjects with BMI >25 kg/m2 and obesity is defined in subjects with a BMI >30 kg/m2. About half of the Swedish population is overweight and 10 % obese 92. Due to the current increase in childhood obesity the global life expectancy in the United States might decline for the first time in modern history 93.

18

Metabolic changes associated with obesity, especially abdominal obesity, are generally linked to insulin resistance 94. Insulin resistance is a situation when normal concentrations of insulin produce a biological response which is insufficient and the regulation of glucose decay, which eventually might lead to type 2 diabetes. Insulin resistant individuals have hyperinsulinemia together with normoglycemia or hyperglycemia 95, 96. Insulin sensitivity var-ies in healthy individuals, but obese individuals are very often insulin resis-tant 94. Insulin resistance develops because of genetic susceptibility, e.g. alterations in genes involved in the insulin signaling pathways, and lifestyle factors, such as intake of excess calories and decreased physical activity 97.

How is insulin resistance and obesity linked? Several mechanisms have been proposed. It is known that adipose tissue secrete NEFA into the circula-tion and elevated NEFA levels may induce many adverse metabolic and physiological effects such as insulin resistance, both in skeletal muscle and the liver. Adipose tissue act as an endocrine organ producing signals that may impair insulin sensitivity. Cytokines, such as the pro-inflammatory fac-tors TNF-� and IL-6 are released from adipose tissue and have been pro-posed as candidates that would promote adverse effects but the studies are inconclusive. Adiponectin, an adipokine, is another candidate that seems to be involved in the link between obesity and insulin resistance. Adiponectin is known to increase FA oxidation in skeletal muscle either by directly regu-lating genes involved in FA oxidation or indirectly by stimulating PPAR� expression. Adiponectin levels are reduced in obesity 94, 97. Insulin resistance in skeletal muscle is coupled with accumulation of intramyocellular lipids and intracellular-signaling molecules such as ceramides and diacylglycerol 97, 98.

Metabolic syndrome The MetS denotes the clustering of several metabolic risk factors in one in-dividual and usually includes disturbances in glucose and insulin metabo-lism, central obesity, dyslipidemia (high TAG levels, low HDL-cholesterol and high levels of small dense LDL-particles) and hypertension. Other dis-turbances often associated with MetS include impaired fibrinolysis and in-creased coagulation, signs of inflammation and endothelial dysfunction.

The concept of risk factor clustering was first introduced by Reaven in 1988 (although described as early as in the 1920s) and was referred to as Syndrome X 99. He meant that insulin resistance and compensatory hyperin-sulinemia were the driving forces behind the other metabolic disturbances in syndrome X and that the syndrome was a risk factor for cardiovascular dis-ease 100. Since then MetS has been a hot topic for researchers, but the defini-tion and its clinical value, beyond the risk associated with its individual components, remains controversial 101-105.

19

There are several definitions available e.g. the definitions according to WHO106, the European Group for the Study of Insulin Resistance (EGIR) 107, International Diabetes Federation (IDF)108 and the Adult Treatment Panel (ATP)III of the National Cholesterol Education Program (NCEP)109. The definitions are slightly different and thus depending on which definition you choose the prevalence of MetS will somewhat differ within the same study population 110. The most widely used definition is the one proposed by NCEP.

Studies have demonstrated that MetS predicts the risk to suffer and die from both CVD and type 2 diabetes but some studies have failed to confirm such a relationship111-115. Many factors, such as lifestyle (physical activity and diet), genetic predisposition and the “thrifty phenotype”, have been sug-gested to be involved in the etiology of MetS 99. The pathophysiology behind the risk factor clustering is complex and unlikely to be caused by one single underlying factor 102, but increased adipose tissue (and especially abdominal obesity) with subsequent insulin resistance might be the primary mecha-nisms that drive the syndrome 116, 117.

The prevalence of the MetS is increasing throughout the world (because of the current obesity epidemic) and increases with age and is higher in cer-tain ethnic groups 99, 117. In 60 year old men and women from Stockholm the prevalence of MetS was 13% among women and 19% among men (EGIR-definition) 118 and in the Atherosclerosis and Insulin Resistance study from Gothenburg the prevalence was 16% among 58-year old men (WHO defini-tion 1998) 119. In the US, in the Third National Health and Nutrition Exami-nation Survey, it was estimated that roughly 24% out of a of total 8608 men and women had MetS, defined according to NCEP ATPIII 110. In Finland, in the Kuopio Ischemic Heart Disease Risk Factor Study, the prevalence of MetS varied between 9 and 14% (NCEP ATPIII or WHO 1998 definition) in 1209 men (42 to 60 years), if those with diabetes and CVD were excluded 112.

Genetic factors and variation It is known that genetic factors in concert with the influences from the envi-ronment control and condition us to develop traits and diseases, including the different components of MetS such as insulin resistance, obesity and hypertension 117, 120. Diet has a special place among the external influences that affects human health and is believed to be one of the most important risk modulators for several diseases with multifactorial pathogenesis 121.

In order to investigate genetic determinants of multifactorial diseases such as CVD, polymorphic genes likely to be involved in the pathogenesis of the disease/phenotype of interest are selected and then analyzed as candidates121. The human genome holds the total genetic information and gives rise to mul-

20

tiple tissues. This is due to differential gene expression of the genetic code (DNA) in different cell types. DNA is packaged as chromosomes and a hu-man cell contains 22 pairs of autosomes and two sex chromosomes122. The DNA consists of about three billion basepairs coding for approximately 30 000 genes. Although more than 99% of the human DNA sequence is the same across populations, these DNA differences can have great impact on how humans respond to the environment and predisposes us for different diseases. Genetic variations are fairly common and are relatively evenly distributed across the human genome. The most common type (>90%) of variation found in the genome is the Single Nucleotide Polymorphism (SNP). SNPs are variations in the DNA sequence when one single nucleo-tide is changed to another (e.g. AAGGCTAA to ATGGCTAA) and occurs both in coding and non-coding regions of the genome. SNPs do not cause disease but may determine the likelihood for a disease. In study III of this thesis SNPs in our candidate gene, SCD1, have been analyzed in association with risk factors for metabolic disease 123. The study is a so-called genetic association study, in which different genotypes (SNPs) in a study population are tested with statistical techniques in association to phenotypic variation in order to identify genetic variations that may underlie disease 124, 125.

21

Aims

The overall aim of this thesis was to explore the serum proportions of indi-vidual FAs and estimated desaturase activities (SCD, D6D and D5D) as risk factors/markers for metabolic and cardiovascular diseases, to study genetic variations in the SCD1 gene and to investigate how dietary fat quality might influence estimated enzyme activities.

The specific aims were: I To examine the effect of estimated desaturase activities and indi-

vidual FAs in serum CE on the risk to develop the metabolic syndrome over 20 years.

II To examine the effect of estimated desaturase activities and se-

rum FAs in serum cholesteryl esters, on cardiovascular and total mortality risk with a maximum follow-up time of 33.7 years.

III To examine cross-sectional associations between genetic varia-

tions in the human SCD1-gene, defined as SNPs and haplotypes, and markers of obesity; BMI and waist circumference (WC), in-sulin sensitivity and estimated SCD activity.

IV To compare the effects of two diets with different fat qualities,

butter (SAT) vs. rapeseed oil (RO) diets, on FA composition and estimated desaturase activities in serum CE and PL.

22

Subjects and Methods

Study I, II and III were based on data from the Uppsala Longitudinal Study of Adult Men, ULSAM, and are observational in nature. Study IV is based on data collected during a strictly controlled dietary cross-over intervention study.

Study participants

I. 1558 men from the investigation at 50 years remained for analysis after the exclusion of 764 men in study I. Excluded were those who were hypertensive (supine diastolic BP >= 95 mmHg), were taking BP medication, had diabetes or were cur-rently taking medication for hyperlipidaemia. 706 of these par-ticipated in the 70 years investigation, had complete FA data, did not have MetS at baseline and were included in the logistic regression analyses.

II. 2009 of the men who participated in the baseline investigation

at 50 years had complete fatty acid data. Our analyses were carried out on this sample and on a healthy subsample (n=1558) excluding persons with previous myocardial infarc-tion (n= 7), stroke (n= 3), cancer (n=7), diabetes (n= 104) and those using lipid lowering drugs (n= 21).

III. Out of the 1221 participants at the investigation at 70 years,

1143 men had a valid DNA sample for genotyping of the SCD1 gene and these men made up the study population.

IV. Twenty subjects with moderate hyperlipidaemia were recruited

from an ongoing health survey at a local telephone company to participate in this study. Six females and fourteen males with a mean age of 50.9 ±10 (mean ± SD) years participated. They were all healthy and they did not take any drugs.

23

Estimation of desaturase activities The desaturase activities were estimated in all four studies as a product to precursor ratio of individual FA in serum CE, and in study IV in PL as well, according to the following: � SCD = 16:1 (n-7) /16:0 (SCD-16) and 18:1 (n-9) /18:0 (SCD-18) � D6D = 18:3 (n-6)/18:2 (n-6) � D5D = 20:4 (n-6)/20:3 (n-6) D6D in PL was in study III estimated as 20:3 (n-6)/18:2 (n-6), since the pro-portion of 18:3 (n-6) was too low for reliable quantification. It should be kept in mind that this ratio also measures the elongase activity but this step is regarded not to be rate-limiting (Figure 2). This ratio is thus considered to be a good estimate of the D6D activity.

The ULSAM- cohort The Uppsala Longitudinal Study of Adult Men, ULSAM, is a population-based cohort study that started in Uppsala, Sweden, in 1970. All men born 1920-1924 living in Uppsala at that time were invited to participate. The participants were examined at baseline at age 50 and reinvestigated at age 60, 70, 77 and 82. The five investigations were extensive and carried out in similar manners; however the number of investigations differed between the ages and the 60 years investigation was the most limited. All investigations were carried out after an overnight fast. Only data from the 50 and 70 years investigations were used in the studies of this thesis. At baseline, 2841 men were invited and 82% (2322) accepted to participate. At age 70 1681 men were invited and 73% (1221) participated. 126

Figure 2. A schematic overview of the ULSAM-study. All men born 1920-24 and living in Uppsala in 1970 were invited to the baseline investigation at 50 years. For the 60, 70 and 77 years investigation all men alive and still living in Uppsala were invited to participate. For the 82-years investi-gation those who participated in the 70 and/or the 77 years investigation were invited.

24

Methods in ULSAM-based studies

Baseline investigations at 50 years All investigations were carried out under standardized conditions and have been described in detail previously 127, 128. The investigations included a medical questionnaire and interview, as well as blood sampling, anthropom-etric measurements and blood pressures.

Anthropometry BMI was calculated as weight (kg) divided by height (m) squared and WC (cm) was measured midway between the lowest rib and the iliac crest in a supine position (measured only in a sub sample).

Blood Pressure Supine blood pressures were recorded after 10 minutes of rest in the right arm in a recumbent position with a Mercury manometer (Kifa Ercameter, wall-model). Systolic and diastolic blood pressures were read to the nearest 5 mmHg and the mean of two values was used in the analyses. Hypertension in study III was defined as a systolic blood pressure >= 140 mmHg or dia-stolic blood pressure >=90 mmHg or a regular use of antihypertensive medi-cation and used as a confounder in the Cox-regression.

Blood samples Blood samples were taken from the antecubital vein. Serum TAG and cho-lesterol concentrations were analyzed on a Technicon Auto Analyzer type II 129. HDL-cholesterol was assayed in the supernatant after precipitation with a sodium phosphotungstate and magnesium chloride solution 130. Serum fast-ing insulin were determined with the Phadebas Insulin test (Pharmacia, Upp-sala, Sweden) based on radioimmunosorbent technique 131. Fasting blood glucose was measured by spectrophotometry using the glucose oxidaze method. Homeostasis model insulin resistance (HOMA) index was calcu-lated as fasting insulin(mU/l) x fasting glucose(mmol/l) /22.5 according to Matthews 132.

Serum cholesteryl ester fatty acids The serum FA composition was analyzed as previously described 133, 134. The samples from the 50 years investigation had been stored in liquid nitrogen for about 15 years before analysis. Serum was extracted with a hexane-isopropanol solution (1+4) and CEs were separated from the extract by thin layer chromatography. Inter-esterification was carried out with acidic metha-nol at 85°C for two hours. Free cholesterol that had been liberated in the reaction was removed by aluminum oxide to avoid contamination of the

25

column. The percentage composition of methylated FAs from 14:0 to 22:6 was determined by gas chromatography (a 25 m NB-351 silica capillary column) with a flame ionization detector and helium as carrier gas. Every 25th sample was a serum control pool. The coefficient of variation varied between series analysis (n=35) from 2% (large peaks) to 10% (smaller peaks) and between successive gas chromatography runs 0.2-5 % (n=17).

Questionnaire and interview A self-administered questionnaire made according to Collen et al 135 was used to collect information on various factor including physical activity, medical treatment and previous and current disease. Smoking data was ob-tained from an interview.

Investigations at 70 years All investigations were carried out as previously described 83, 136.

Anthropometric measurements BMI was calculated as weight (kg) divided by height (m) squared and waist circumference (cm) was measured midway between the lowest rib and the iliac crest in a supine position.

Blood pressure Blood pressure was measured as described above.

Blood samples Blood samples were taken from the antecubital vein. Serum cholesterol and TAG concentrations were analyzed by enzymatic techniques using IL Test Cholesterol Trinders's Method and IL Test Enzymatic-colorimetric Method for use in a Monarch apparatus (Instrumentation Laboratories, Lexington, USA). HDL particles were separated by precipitation with magnesium chlo-ride/phosphotungstate. An oral glucose tolerance was carried out. 75 g glu-cose dissolved in 300 ml of water were given the subjects, and blood sam-ples for plasma glucose and insulin were drawn immediately before, and 30, 60, 90, and 120 min after ingestion of glucose. Plasma glucose was meas-ured by the glucose dehydrogenase method (Gluc-DH, Merck, Darmstadt, Germany). The intra-individual coefficient of variation for fasting plasma glucose was 3.2%. Plasma insulin was assayed using an enzymatic-immunological assay (Enzymmun, Boehringer Mannheim, Germany) per-formed in an ES300 automatic analyser (Boehringer Mannheim).

26

Serum cholesteryl ester fatty acids The FAs were analyzed within one year of collection and in a similar way as at the 50 years investigation. However, the lipids were extracted in chloro-form with 0.005% butylated hydroxy-toluene added as an antioxidant. After transmethylation the fatty acid methyl esters were separated by Gas Liquid Chromatography. The Gas Liquid Chromatograph system was made up by a Hewlett Packard (Avondale, PA, USA) GC 5890 with an automatic sampler 7671A and a 3392A integrator, and a 25 m Quadrex (New Haven, CT, USA) fused Silica capillary column (model OV-351). The FA were identified by comparing each peak’s retention time with the retention times of NuCheck Prep’s (Elysian, MN, USA) methyl standards (Gas Liquid Chromatograph reference standard GLC-68A).

Direct measurement of insulin sensitivity Insulin sensitivity was determined by the golden standard euglycemic hyper-insulinemic clamp technique according to DeFronzo 137, slightly modified. Briefly, insulin was infused in a primary dose during the initial 10 minutes and then as a continuous infusion for 110 minutes. The infusion rate was 56 mU (instead of 40 mU) per minute per body surface area (meters squared) to achieve nearly complete suppression of hepatic glucose output 138. The level of plasma glucose was maintained at 5.1 mmol/L by glucose infusion. The total amount of glucose infused serves as the sensitivity to the existing plasma insulin concentrations. The glucose disposal (M (mg x kg-1x min-1)) was calculated as the amount of glucose taken up during the last 60 minutes of the study. Tissue insulin sensitivity (amount of glucose metabolized per unit of plasma insulin) was also calculated as the insulin sensitivity index (M/I (mg x kg-1 x min-1 / 100 mgU/l)), where M is the glucose disposal and I is the mean insulin concentration during the clamp.

Questionnaire and interview A questionnaire was used to obtain information on physical activity, medical treatment and previous and current disease. The medical questionnaire was based on the questionnaires previously used at 50 and 60 years investiga-tions. During the clamp investigation, a nurse or a technician posed the ques-tion "Do you smoke?" and this forms the basis for the smoking variable.

Definition of the metabolic syndrome in study I MetS was defined according to the definition of the National Cholesterol Educational Program (NCEP) Adult Treatment Panel III (ATP III) 109, 139. MetS was considered present when three or more of the following risk fac-tors were met: Systolic blood pressure>=130 mmHg or diastolic blood pres-sure >=85mmHg, fasting plasma glucose >=6.1 mmol/l, TAG >=1.69 mmol/l, HDL-cholesterol <1.04 mmol/l or WC >102 cm. At the 50 years

27

investigation, WC was only recorded for those born in 1920 and 1921 (n=480) and to be able to define the obesity criteria a BMI cut-off was used instead. The BMI cut-off was 29.2 kg/m2, which corresponded to a waist circumference >102 cm (Linear regression equation: BMI [kg/m2] = 0.2863 x WC [cm] -0.034). This was derived from subjects with data on both BMI and WC. At the 70 years investigation, WC was measured for most men (n=1192) and was used to define the obesity criteria.

Follow up and outcome measures in study II Follow-up was from the examination date for the baseline investigation in 1970-73 and to December 31st 2003, with a maximum of 33.7 years of fol-low up (median 30.7 years rendering 61518 and 58149 person years at risk in the total sample and the health subsample, respectively). Information re-garding mortality was collected from the official Swedish Cause of Death registry held by the National Board of Health and Welfare (Socialstyrelsen) in Sweden. The register includes all Swedish citizens and the study partici-pants were linked to the registry by means of a unique personal identification code. The registry uses the codes of the International Classification of Dis-ease (ICD) and the outcomes in study III were defined a priori as cardiovas-cular mortality (ICD-9 codes 390-459, ICD-10 codes I00-I99, to comply with current European guidelines 140) and total mortality.

DNA preparation, SNP selection and genotyping in study III The DNA preparation was carried out as previously described 126. Eleven SNPs were selected from the dbSNP database 141 to be evenly distributed over a region including the SCD1 gene and about 10 kb upstream and 2 kb downstream of the gene. The SNPs were genotyped at the SNP technology platform at Uppsala University 142 using a 12-plex GenomeLab SNPStream system (Beckman Coulter)143. Three of the genotyped SNPs had a minor allele frequency (MAF) <0.05 and were therefore not further analyzed in study II. The quality of the genotype data was assessed by testing for Hardy-Weinberg equilibrium (HWE) using the �2distribution (p>0.05) for each as-say. The remaining eight SNPs conferred to HWE. The overall genotype call rate was 96.7%, and the accuracy was 99.96% according to duplicate analy-sis of on average 40% (4723/11 870) of the genotypes.

Statistical Analysis

Study I The statistical analyses were carried out using the software package STATA (version 6.0; STATA Corporation, TX, USA). Two tailed p-values and 95%

28

CI were given and p<0.05 considered significant. Continuous data are given as means ± SD. Smoking status was a categorical variable, where non-smoker=0, smoker=1 and ex-smoker=2 at 50 years and smoker=1 and non-smoker=0 at 70 years. Physical activity levels were graded 1 (lowest) to 4 (highest). The categorical variables were used as confounders in the multi-variate logistic regression models.

Distribution The normal distribution of continuous variables was examined with Shapiro-Wilk’s test. Variables non-normally distributed (W<0.95) were log-transformed.

Basic Statistics Student’s t-test was used to test differences in means between baseline esti-mated desaturases, individual FAs and clinical characteristics in individuals with and without the MetS at age 70.

Logistic regression analysis Logistic regression analysis was carried out to estimate the risk of having MetS at the age of 70 in relation to the estimated and standardized (SD=1, mean=0) desaturase activities at 50 years. Variables were standardized in order to be able to compare the effect of independent variables on the de-pendent variable. The logistic regression analysis was carried out univariate and multivariate. The following variables were included in the multivariate logistic regression models, either alone or in combination with each other: BMI, smoking behavior and /or physical activity.

Study II The statistical analysis was carried out with STATA, version 8.2 (College Station, TX, USA)

Distribution Continuous variables are presented as means (SD) or median (inter-quartile range) and categorical variables as number of individuals (%). The normal distribution of continuous variables was examined with Shapiro-Wilk’s test and variables non-normally distributed (W<0.95) were log-transformed (16:1, 18:0, 18:3 n-6, 20:5 n-3, 22:6 n-3, SCD and D6D).

Cox proportional hazard analysis Independent variables- individual fatty acids (14:0 -22:6) and estimated de-saturase activities (SCD, D6D and D5D) - were investigated in two ways: linear relations were investigated as the effect of 1 SD increments in con-

29

tinuous variables, and non-linear relations were examined using quartiles of the independent variables. Cox proportional hazard models were carried out to examine individual fatty acids and estimated desaturase activities in rela-tion to cardiovascular and total mortality. The analysis was carried out both on the total study sample and on the healthy subsample and for each sample and endpoint an unadjusted and a multivariable-adjusted (smoking status, physical activity, BMI, total cholesterol and hypertension) model were ex-amined. The proportional hazards assumption was checked using Schoenfeld’s test. Hazard ratios and two-tailed 95% CI are given.

Study III The statistical analysis was carried out with STATA, version 8.2 (College Station, TX, USA) and the adjusted �-level was calculated with a macro in SAS, version 8.0 (SAS Institute, Cary, NC, USA).

To account for possible false positive tests due to multiple testing an ad-justed significance level using Meff correction 144 was calculated. All test were two tailed and p-values were compared both with conventional signifi-cance level (p<0.05) and the adjusted �-level=0.0076.

Distribution Continuous, normally distributed variables are presented as mean ± standard deviation (SD). Skewed variables (fasting insulin, fasting glucose and SCD-16) were log-transformed before analysis and are presented with geometric mean and 95% CIs.

Linear regression models Associations between genotypes (SNPs and haplotypes) and phenotypes (insulin sensitivity, BMI and waist circumference and SCD-ratio), were ana-lyzed in linear regression models. General (codominant, step-wise-test) and additive (trend test) genetic models, using the most common homozygous allele (for each SNP) and the non- carrier of a specific haplotype as reference level, were carried out.

Multivariable linear models adjusting, the SNP (rs7849) that showed the strongest association with insulin sensitivity, for BMI and WC were carried out. Concomitant co-morbidity and drug use (hypertension treatment, lipid medication, diabetes prevalence and variables based on questionnaire ques-tions)as also adjusted for. To investigate whether the association between rs7849 and insulin sensitivity was modified by the level of obesity, the study population was stratified into tertiles of BMI and WC and a linear regression model was carried out on these strata, separately. The association between rs7849 and WC was also adjusted for insulin sensitivity.

30

Figure 3. Illustrates the pairwise linkage disequlibrium (LD) expressed as r2 be-tween the investigated Stearoyl-CoenzymeA-Desaturase Single Nucleotide Poly-morphisms (SNPs) in the Haploview software setting. Shading represents magni-tude and significance of the pairwise r2. Black, r2 > 0.8 (high r2), dark grey, r2 between 0.8 and 0.6, moderate grey, r2 between 0.4 and 0.6 and white, r2 < 0.2 (low r2). The haplotype group was constructed by rs10883463, rs7849, rs2167444 and rs508384.

Haplotype estimation Haplotypes were estimated using the software program PHASE 145. One group of haplotypes was estimated by selecting four SNPs in strong linkage disequilibrium (LD) based on pairwise marker comparisons (r2) in the Hap-loview software, version 3.2 146 (Figure 4). To account for uncertain phase in the haplotype estimations, only haplotypes with the probability of >0.98 were included in the statistical analysis.

31

Study IV; Study subjects, design, diets and methods

Study population Six women and fourteen men were recruited from an ongoing health survey at a local company. They were all healthy, but slightly overweight and had moderate hyperlipidaemia. None of the participants took any drugs.

Study design A two-period cross-over randomized study design was used. The interven-tion with each diet was three weeks with a wash-out period of four weeks in between. The subjects started either with a SAT-diet or a RO-diet. All were asked to maintain their usual physical activity and not to undertake any other lifestyle changes during and between the test periods.

Diets The diets were prepared at a metabolic ward kitchen, where all participants picked up all meals they consumed during the test periods. The energy re-quirements of the participants were estimated to 30(126) and 35(147) kcal (KJ)/ kg body weight for women and men, respectively. The menu was planned as a seven day menu including breakfast, lunch and dinner with some snacks in between. Both diets were intended to contain equal amounts of macronutrients, besides the differences in FA composition. Both diets were prepared with the same foods, except for the quality of fat. The diet rich in butter fat included butter, high fat cheese and cream, while the RO diet was prepared with a special spread (margarine type) high in rapeseed oil, a liquid margarine made of rapeseed oil for cooking and pure rapeseed oil. The rapeseed oil contained no erucic acid. The FA compositions of the diets were analyzed according to methods earlier reported 133. Investigations All samples and anthropometric measurements were carried out after an overnight fast. Weight (kg) and height (m) was recorded and BMI was calcu-lated as weight divided by height squared. TAG and cholesterol concentra-tions were measured in serum, using the IL Test Cholesterol Triander's method 181618-80 and IL Test Enzymatic-Colorimetric-Method 181709-00 for use in a Monarch 2000 centrifugal analyzer (Instrumentation Laboratory, Lexington, MA, USA.). The serum insulin concentrations were measured by an enzyme linked immunosorbent assay test from Boehringer Mannheim GmbH (Mannheim, Germany) and the blood glucose concentration by the glucose oxidase assay. The FA composition of CEs and PLs was determined by GLC as described previously 133, the same method used in the 70 years ULSAM investigation.

32

Statistical analysis The statistical analyses were carried out with the software SAS, version 9.1 (SAS institute Inc., NC, USA). All test were two-tailed and p <0.05 consid-ered significant.

The normality of the variables was tested with Shapiro Wilk’s test. Vari-ables not normally distributed (W<0.95) were log-transformed before analy-sis. The cross-over design of the study, the scales and the distribution of the variables were taken into account. To test for differences in proportions of FAs and estimated desaturase activities between the two test diets an analy-sis of variance (ANOVA) model was carried out, in which factors for diet, patient and sequence of intervention was accounted for. A test for carry-over effects was carried out according to Jones and Kenwood 147. Least square means (LSM), which take imbalances between groups into account, formed the basis for all tests and estimates in the analyses; however data is presented as means ± SD.

Ethics The studies were approved by the ethics committee at Uppsala University and all participants had given their informed consent.

33

Results and discussion

Study I

Baseline characteristics All studied clinical characteristics; blood pressures, fasting blood glucose and insulin levels, BMI, WC, HDL-cholesterol, serum TAG and HOMA-index, were at baseline higher (except for HDL-cholesterol which was lower) in subjects who had developed MetS 20 years later.

Prevalence of the metabolic syndrome The prevalence of MetS was relatively low at baseline (8.3%) but this was partly due to the fact that many individuals that could have been classified as having MetS at that time were excluded from the study. At the 70 years in-vestigation the prevalence was higher and more as expected, 16.9%, when those classified as having the MetS at age 50 were excluded, and 19.8% when the entire study population was considered.

Baseline fatty acid profile and estimated desaturase activities The difference in the relative proportions of individuals FA and estimated desaturase activities in cholesterol esters were as presented in Table 1 and in Figure 5 the difference in proportion (%) in the estimated desaturase activi-ties in those with and without MetS at baseline is presented.

Longitudinal analysis The risk of having MetS at age 70 (those having MetS at age 50 were ex-cluded) was approximately 30 percent higher for each SD increase of the estimated SCD [OR= 1.29, 95%CI= 1.0; 1.60] and D6D [1.35, 1.10; 1.65] ratios, while 30 percent lower for each SD increase of the estimated D5D activity [0.71, 0.77; 0.87]. In the multivariate analyses the risk of developing MetS over 20 years disappeared for estimated SCD and D6D activity after adjustment for BMI alone and in combination with physical activity and smoking. The relationship between estimated D5D activity and MetS re-mained after adjustment for all confounders and the estimated activities of SCD and D6D.

34

Table 1. The difference in the relative amount of fatty acids and estimated desatu-rase activities in serum cholesterol esters at the age of 50, in those men who had (n=119) or had not (n= 587) developed MS at age 70. The arrow indicates if the FA or ratio was higher (�) or lower (�) in subjects with MetS. Fatty acid (% of total FA) and fatty acid ratio

Subjects without MetS Mean ± SD

Subjects with MetS Mean ± SD

p- valuea

14:0 (Myristic) 1.09 ± 0.24 1.15 ± 0.23 0.012 � 16:0 (Palmitic) 11.4 ± 0.93 11.7 ± 0.95 0.012 � 16:1 n-7 (Palmitoleic) 3.5 ± 1.1 3.8 ± 1.1 0.003 � 18:0 (Stearic ) 1.14 ± 0.3 1.17 ± 0.3 0.34 18:1 n-9 (Oleic) 18.9 ± 2.5 19.4 ± 2.2 0.037 � 18:2 n-6 (Linoleic) 55.4 ± 4.7 54.0 ± 4.4 0.003 18:3 n-6 (�-linolenic ) 0.66 ± 0.16 0.73 ± 0.33 0.011 � 18:3 n-3 (�-linolenic ) 0.66 ± 0.16 0.67 ± 0.19 0.27 20:3 n-6 (Dihomo-�-linolenic)

0.54 ± 0.13 0.60 ± 0.15 0.000 �

20:4 n-6 (Arachidonic)

4.74 ± 0.94 4.79 ± 1.0 0.6

20:5 n-3 (Eicosapentaenoic)

1.30 ± 0.57 1.38 ± 0.74 0.28

22:6 n-3 (Docosahexaenoic)

0.70 ± 0.2 0.69 ± 0.2 0.31

�9-16 [16:1 n-7/16:0] 0.31 ± 0.09 0.33 ± 0.09 0.02 � �6 [18:3 (n-6)/18:2 (n-6)] 0.012 ± 6x10-3 0.014 ± 7x10-3 0.004 � �5 [20:4 (n-6)/20:3 (n-6)] 9.0 ± 2.1 8.3 ± 2.1 0.001

aThose subjects having MetS at baseline were excluded in the analyses.

Discussion In the present study the prevalence of MetS was low at baseline, 8.3% which is partly explained by the initial exclusions made. This is, however, in line with the results from a Finnish cohort consisting of 42-60 year old men. In that study, men with diabetes and CVD were excluded and the prevalence was between 9-14%, depending on the definition used 112. At the 70 years investigation the prevalence was higher, 16.9%, (those classified as having the MetS at age 50 were excluded), and 19.8% when the entire study popula-tion was considered, which accords with previous findings in Swedish men from Stockholm 118 and Gothenburg 119.

The serum lipid FA composition and the estimated desaturase ratios were already at baseline changed in those individuals that at the 70 years investi-gation were defined as having MetS. This changed FA composition was characterized by high proportions of palmitic (16:0), palmitoleic (16:1) and dihomo-�-linolenic (20:3 n-6) acids and a low proportion of LA.

35

Figure 4. The difference in proportion (%) in the estimated desaturase ac-tivities at the age of 50 in those who were defined as having (MetS, n=119) or not having (Not MetS, n=587) the metabolic syndrome at age 70. To be able to fit all the information in a reader-friendly figure the SCD and D6D ratio was multiplied by 10 and 1000, respectively. This means that a changed serum lipid FA composition with apparent high SCD and D6D and low D5D estimated activities might be an early sign of metabolic alterations preceding the risk factors clustering present in MetS. Also, the clinical characteristics (all of which can be associated with MetS) that were investigated in the present study had already started to deteriorate at baseline in those individuals who were defined as having MetS at the 70 years investigation. Thus, we can conclude that metabolic disturbances start early but it is not possible to say whether the changes in FA composition precedes or follows other disturbances from the results in this study.

In the logistic regression analysis, higher estimated activity of SCD and D6D and lower estimated D5D activity increased the risk of developing MetS over 20 years. The relationship between D5D and MetS was inde-pendent of BMI, smoking habit, physical activity and the estimated activities of SCD and D6D. The relationship between D6D and MetS was independent of smoking behavior, physical activity, as well as BMI in com-bination with smoking, whereas the effect of SCD activity was only inde-pendent of smoking. This suggests that the effect of D6D and D5D activity on the development of MetS is more or less independent of lifestyle factors, whereas the effect of SCD may be mediated via life style factors such as physical inactivity, diet or other factors that promote obesity. The associa-

36

tion between metabolic disease and life style is well known and the risk of diabetes in subjects with impaired glucose tolerance was reduced by 58% in a lifestyle intervention trial in Finland 148 and moderate and vigorous leisure time physical activity was reported to decrease the risk of MetS by two-thirds in a cohort of middle-aged Finnish men 149. The findings in the present study further establish serum lipid FA composition and estimated SCD, D6D and D5D activities as risk factors/markers for the development of metabolic disease. However it remains unclear how desaturases are involved in the etiology of MetS.

Study II Baseline characteristics The prevalence of smoking was similar in both samples (51%), while hyper-tension was slightly more prevalent in the healthy subsample (43% vs. 41%). Five percent of the total study population had diabetes. Total cholesterol (6.9 mmol/l) and BMI (25 kg/m2) did not differ between the two study samples.

Mortality data During follow up from the examination date at baseline until December 31st 2003, 1012 men died (rate 19.0/ 1000 person years at risk) and 461 from cardiovascular causes in the total sample and in the healthy subsample 931 men died (rate 18.5/ 1000 person years at risk) and 416 from cardiovascular causes. Due to the official hospital and cause-of-death registries in Sweden, there was no loss to follow up in this study.

Cox proportional hazard analysis The results from the Cox analysis did not substantially differ in the total and the healthy subsample and the results discussed below are only from the analysis in the total sample.

One SD increase of the proportions of serum myristic, palmitic, palmitoleic, oleic, �-linolenic, DHLA acids were associated with an in-creased risk of both cardiovascular and total mortality, while a high propor-tion of LA was associated with decreased mortality (table 2). The propor-tions of long chain n-3 fatty acids, ALA, EPA and DHA, as well as of stearic and arachidonic acids, were not associated with any mortality risk in the present study. For the majority of FA the associations followed the same pattern both for cardiovascular and total mortality and were most strongly associated with cardiovascular death. After adjustment for total cholesterol, BMI, smoking, physical activity and hypertension, the hazard ratios re-mained in the same ranges, but statistical significance was attenuated. The greatest risk was observed per 1 SD increment in the proportion of palmitoleic acid and the serum proportion of LA was most strongly and in-versely associated with mortality.

37

The adjusted hazard ratio, taking risk factors into account, for each SD in-crease of SCD-16 was 1.15 (1.04-1.27), for D6D 1.12 (1.0-1.24) and for D5D 0.88 (0.80-0.98) for cardiovascular mortality. If we estimated SCD as the 18:1[n-9]/18:0 ratio, this ratio was only associated with an increased risk of total mortality. In the adjusted model, the hazard of cardiovascular death in quartile IV compared to the referent level was 60% higher for SCD and for D6D and D5D the hazard was 50% higher and 24% lower, respectively (Figure 6).

Discussion This study is the first to relate indices of desaturase activities to mortality, and our results suggest a link between saturated fat and cardiovascular and total mortality and confirm a previous Finnish prospective study which linked high dietary and serum LA to decreased cardiovascular mortality 88.

Desaturases The specific role of estimated desaturase activities in cardiovascular disease and subsequent mortality is presently largely unknown.

Figure 5. . The adjusted hazard ratio (95% CI) for cardiovascular mortality (total study sample) in 4th quartile (4thQ) compared to the referent level of estimate de-saturase activity. Adjusted for total cholesterol, BMI, smoking, physical activity and hypertension.

However, in the present study a predictive value, beyond classical cardio-vascular risk factors, (including total cholesterol, BMI, smoking habit, physical activity and hypertension) was observed for estimated SCD and D6D activity. High D6D activity might lead to high serum proportions of

38

DHLA, which has been associated with insulin resistance 83 and obesity 82. The observed decrease in mortality a high estimated D5D activity might be linked to improved insulin sensitivity 150.

SCD and MUFA Each SD increase of estimated SCD activity and palmitoleic acid was asso-ciated with an approximate 30% mortality increase in the unadjusted model and in the adjusted model the increase was 15% and 20% for the SCD ratio and palmitoleic acid, respectively. High estimated SCD activity has previ-ously been strongly associated with cardiovascular and metabolic risks 84, possibly mediated by increased lipogenesis 151 It is possible that high esti-mated SCD1 activity is secondary to high intake of saturated fats as reflected by increased serum proportions of palmitoleic acid and to some extent oleic acid.

Palmitoleic acid content in the diet is in general low and most palmitoleic acid in serum is thus derived from endogenous metabolism of palmitic acid catalyzed by SCD. Palmitoleic acid is known to raise LDL-cholesterol as much as palmitic acid 152, potentially relating high serum proportions of this FA to CVD.

Table 2. Standardized Cox proportional hazard ratios for fatty acids in serum cho-lesteryl esters and estimated desaturase activities (FA ratios) in relation to Cardio-vascular mortality in the Total study sample (n= 2009). Fatty acid and estimated desaturase activities

Cardiovascular mortality Number of events= 461

Crude HR(95% CI)

Adjusteda HR(95% CI)

��b

Myristic acid (14:0) 1.16 (1.06-1.27) 1.12 (1.02-1.23) � Palmitic acid (16:0) 1.25 (1.14-1.37) 1.15 (1.04-1.26) � Palmitoleic acid (16:1) 1.32 (1.21-1.44) 1.18 (1.07-1.30) � Stearic acid (18:0) 1.07 (0.97-1.17) 1.04 (0.94-1.15) Oleic acid (18:1) 1.29 (1.18-1.41) 1.18 (1.07-1.30) � Linoleic acid (18:2 n-6) 0.76 (0.70-0.83) 0.85 (0.78-0.94) � �-Linolenic acid (18:3 n-6) 1.15 (1.05-1.27) 1.09 (0.98-1.21) � �-Linolenic acid (18:3 n-3) 1.08 (0.99-1.18) 1.10 (1.0-1.21) DH-�-linolenic acid (20:3 n-6) 1.16 (1.07-1.27) 1.06 (0.96-1.18) � Arachidonic acid (20:4 n-6) 1.00 (0.92-1.10) 0.95 (0.86-1.05) Eicosapentaenoic acid (20:5 n-3) 1.07 (0.98-1.17) 0.99 (0.90-1.09) Docosahexaenoic acid (22:6 n-3) 0.97 (0.89-1.07) 0.92 (0.84-1.02) SCD (16:1/16:0) 1.27 (1.16-1.39) 1.15 (1.04-1.27) � D6D (18:3 n-6 /18:2 n-6) 1.20 (1.10-1.32) 1.12 (1.0-1.24) � D5D (20:4 n-6/ 20:3-6) 0.84 (0.76-0.93) 0.88 (0.80-0.98) � DH indicates Dihomo, FA fatty acids, a The adjusted model included total cholesterol, BMI, smoking, physical activity and hypertension, b The arrows indicate if a high (�) or a low (�) proportion of the individual FAs or FA ratios was associated with mortality.

39

Figure 6. Nelson-Aalen plots of the cumulative hazard of cardiovascular mortality in the total study sample (n=2009) by above vs. below the median of the estimated (a) SCD (16:1/16:0), (b) D6D(16:3 n-6/ 18:2 n-6) and (c) D5D (20:4 n-6/ 20:3 n-6) indices in serum cholesteryl esters.

40

Oleic acid is possibly protective with regard to CVD 12, but was in the pre-sent study associated with increased mortality. However this might be caused by the fact that oleic acid was derived from the same dietary sources (dairy and meat products) as the SFAs when the samples were collected, as observed in a previous study from the United States 56.

Individual fatty acids High proportions of LA in serum CE were inversely related to cardiovascu-lar and total mortality in the present study. A one-unit increment in the pro-portion LA was associated with a 15% risk reduction of cardiovascular death, which is in line with previous findings32, 68, 153.

The serum proportions of long chain n-3 FA; ALA, EPA and DHA, also known to partly reflect the dietary intakes 51, were not associated with mor-tality in the present study. This might seem surprising since n-3 FA have been associated with cardioprotective effects 34, 35, but this might be due to a relatively high n-3 content of the background diet in the present study popu-lation (between 1 and 2 grams/day).

Palmitic acid, but not stearic acid was significantly associated with in-creased mortality, especially cardiovascular mortality. The finding may agree with experimental data suggesting that palmitic acid have unique ef-fects on several cellular function, such as apoptosis 154, endoplasmic reticu-lum stress 155, up-regulation of SCD-1 81 , and may accumulate in lipid me-tabolites such as ceramide and diacylglycerol 156, 157.

The results in the present study are in accordance with the findings in study I and to previous observational studies relating estimated desaturase activities and individual FA to cardiovascular and metabolic diseases 56, 68, 85,

86, 148, 158, 159 and as discussed in the background section of this thesis. In this study many associations between FA, estimated desaturase activities and mortality was independent of BMI and other risk factors for cardiovascular disease and this warrants further investigation.

Study III

SNPs and phenotype associations Markers of obesity (WC and BMI) and insulin sensitivity (M-value, n=986) were associated with the same SNPs, but with effects in opposing directions. rs10883463, rs7849, rs2167444 and rs508384 were all negatively associated with markers of obesity and positively associated with the M-value. The most pronounced effects on both WC and M-value was observed for subjects with the CC allele in rs7849, which corresponded to 4 cm less WC and 23% higher insulin sensitivity. The associations with BMI did not reach signifi-cance. If the M/I value was used instead of the M-value the ß-coefficients and p-values were in the same range.

41

When the association between the CC-allele (rs7849) and the M-value was adjusted for markers of obesity, BMI attenuated the relationship (ß = 0.75, p=0.034), while WC removed any significant effect (ß-coefficient= 0.62, p=0.084). Concomitant co-morbidity and drug use did not influence this association. The stratification on markers of obesity demonstrated that the association between rs7849 and insulin sensitivity was not associated with BMI. However, subjects in the lowest tertile of WC (85 ± 5 cm (mean ± SD)) homozygous for the rare allele of rs7849 had higher insulin sensitivity (ß = 1.21, p=0.023). The estimated SCD activity (16:1 [n-7]/16:0, n=489) was associated to CC in rs3071 and to GG in rs3793767 but the effects were minor.

Haplotypes and phenotype associations Three haplotypes (I, II and III) with a frequency of >0.05 were estimated. Subjects heterozygous (one copy) for haplotypes I and III were associated with higher obesity and lower obesity, respectively. The effect was the strongest between the haplotypes and WC, but the effect on BMI followed the same pattern, but did not reach significance. The direction of the associa-tion between the M-value and haplotypes I and III was opposite of that ob-served between obesity and haplotypes.

Table 3. Information regarding the eleven genotyped Single nucleotide polymorphisms in the SCD-gene and associations + (positive) or – (nega-tive) to the investigated phenotypes: M-value, Waist and SCD-ratio. Associations to:

db SNP ID Positiona Role MAF Alleles M-value

Waist SCD- ratio

rs655120b -6376 Promotor 0.002 A/G rs612472b 2863 Intron 0.015 C/T rs3870747 6417 Intron 0.069 C/T rs3071 7202 Intron,

Exon/intron boundaries

0.342 A/C -

rs3793766 7981 Intron 0.072 C/T rs3793767 8484 Intron 0.393 A/G + rs10883463 11692 Intron 0.068 C/T + - rs7849 15341 Exon 6 0.165 C/T + - rs2167444 17482 3’UTR 0.141 A/T + - rs508384 17499 3’UTR 0.166 A/C + - rs539480b 18590 3’UTR 0.023 C/T a Relative position from translation initiation site (ATG= +1) b SNPs with MAF <5% MAF minor allele frequency, Minor allele is bolded and in italics.

42

Discussion In the present study, each added copy of the rare allele of rs10883463, rs7849, rs2167444 and rs508384 was associated with lower BMI and WC and improved insulin sensitivity. Being homozygous for the rare allele, com-pared to being homozygous for the common allele of rs7849, demonstrated the strongest effect on both insulin sensitivity (ß=1.19, p=0.007) and WC (ß=-4.4, p=0.028), corresponding to 23% higher insulin sensitivity and 4 cm less waist line. These results accord well with animal data; mice with a dis-rupted SCD1 gene have reduced adiposity, increased insulin sensitivity and are resistant to diet induced weight gain 75.

The adjustment for and stratification on markers of obesity, that was car-ried out in the present study to find out whether the effect of rs7849 (CC) on insulin sensitivity was primary or secondary to obesity, indicated that the effect was secondary to abdominal obesity. This is in line with the results of study I, in which the estimated SCD activity was significantly associated with the development of MetS, but the association was lost after adjustment for obesity 151. Also, the key function of SCD is to regulate lipogenesis, whereas effects on glucose metabolism are believed to be secondary 74. Ad-justment for concomitant co-morbidity and drug use did not influence the association between rs7849 and insulin sensitivity either. In the present study the associations between haplotypes and phenotypes indicated that the rare alleles (Haplotype III) were associated with lower adiposity and higher insu-lin sensitivity further supporting the findings in the present study.

The effects of the gene polymorphisms on estimated SCD-16 were small, unsure and borderline significant and a bit disappointing (Table 2), since the estimated SCD activity can be regarded as a surrogate measure of SCD ac-tivity 89. The following has to be considered however: 1. FA composition was assessed in a sub-sample of the study population

leading to decreased power. 2. Estimated SCD activity is not a true enzyme activity, but a surrogate

measure. 3. SCD activity was estimated in serum CE and not in target tissues, i.e.

adipose tissue, liver or skeletal muscle.

If the 18:1[n-9]/18:0 ratio was used in the analysis no significant associa-tions were detected, in analogy with findings in study I and III. This might seem surprising since the favored substrate for SCD is 18:0 66. However, the lack of associations between SNPs in SCD and 18:1[n-9]/18:0 might be ex-plained by the higher dietary content of 18:1, compared to 16:1. This would lead to a “dilution” of the index by exogenous 18:1 and thus result in a worse reflection of the desaturation by SCD, which might have weakened the sta-tistical effects even more.

The present study may indicate that SCD-1 might be important for the development of obesity and consequently in the development of insulin re-

43

sistance. Thus it is possible that SCD1 might be a switch-point between obe-sity and insulin resistance. It is well known that metabolic diseases are asso-ciated with changes in intracellular lipid metabolism, which might lead to ectopic fat accumulation in e.g. liver and skeletal muscle. This involves the partitioning between de novo lipid synthesis and oxidation of lipids and SCD has been proposed as a key regulator of both these processes 74. It is thus possible that high SCD activity during a prolonged time leads to an accumu-lation of lipids in tissues other than adipose, thus promoting lipotoxicity and insulin resistance, which further aggravates metabolic disturbances 97, 160.

Study IV Fatty acid composition of the diets The fatty acid composition of the two test diets is presented in Figure 8.

Table 4. Baseline characteristics of study subjects

n=20, except for insulin when n=19 SD Standard deviation, IQR interquartile range

Clinical characteristics The subjects were moderately hyperlipidemic and on average slightly overweight at inclusion. Baseline characteristics are pre-sented in Table 4. There was no difference in body weight before, after or between each diet period. The concentrations of serum cho-lesterol (5.59 and 6.65 mmol/l) and triglycerides (1.77 and 2.03 mmol/l) were significantly differ-ent (p<0.001) after the RO- and SAT-diet, respectively.

Fatty acid composition and estimated desaturase activities in serum lipid esters The relative proportions of the individual FA in serum CE and PL were at the end of the test periods as follows. The proportion of 16:0 and 18:0 (only in CE) as well as of 16:1 and 20:3 n-6 were all significantly higher on the SAT-diet while the proportions of LA and ALA were significantly higher after the RO- diet. Docosapentadecanoic acid (DPA, 22:5 n-3) in PL was significantly higher (p<0.01) during the SAT-period. The milk fat bio-markers, 15:0 and 17:0 (only in CE), were significantly higher on the SAT-diet. The relative proportions of the estimated desaturase activities were as presented in Table 2. SCD-16 and D6D activities were significantly higher on the SAT-diet, while D5D significantly lower (p<0.001 for all).

Baseline characteristics a Mean(SD) or median (IQR)

Age 50.9 ± 10.0 Body weight (kg) 77.5 ± 9.9 BMI ( kg/m2) 26.3 ± 2.7 Serum cholesterol (mmol/l) 6.70 ± 1.20 Serum triglyerides(mmol/l) 1.99 (1.38-2.67)HDL-cholesterol (mmol/l) 0.88 (0.81-1.22)

Serum insulin (mU/l) 12.1(9.6-16.0)

Fasting glucose (mmol/l) 5.55(5.10-6.35)

44

Figure 7. The top pie-chart (a) illustrates the calculated FA content (%) ofthe RO diet and the lower pie-chart(b) illustrates the calculated FA con-tent (%) of the SAT-diet. LCFA Long chain fatty acids, MCFA medium chain fatty acid

Discussion In this study we wanted to study how surrogate measures of desaturase ac-tivities were affected by dietary fat quality. This is the first study to report how the estimated activities of FA desaturases change after a strictly con-trolled intervention with test diets representing two fat qualities; one diet rich in saturated fat and the other rich in unsaturated fat, based on rapeseed oil. The estimated SCD-16 (only CE) and D6D activities were significantly higher while D5D significantly lower in serum CE and PL on the SAT- compared to the RO-diet (Table 5). The SAT-diet induced a FA composi-tion, in the serum lipids, similar to that earlier reported in several epidemi-ologic studies to be associated with e.g. cardiovascular disease and type 2

45

diabetes 84 and reported in study I and II to be associated with MetS and cardiovascular and total mortality.

The content of palmitic (16:0) and stearic (18:0) acids was twice as high in the SAT-diet, compared to the RO-diet. In spite of the large differences in SFA content between the two test diets, differed the proportion of SFA in serum lipids only slightly after the intervention periods.

Table 5. The relative proportions (% of total) of estimated desaturase activi-ties in serum cholesteryl esters (CE) and phospholipids (PL) by diet group.

SAT- diet RO-diet Fatty acid ratio CE* PL* CE* PL* SCD-16 (16:1n-7/16:0) X 0.34 0.027 0.25*** 0.025 SD 0.10 0.006 0.12 0.007 SCD-18 (18:1 n-9/18:0) X 21.2 0.81 30.8*** 0.92*** SD 4.8 0.10 4.1 0.10 D6D* * X 0.017 0.18 0.013*** 0.13*** SD 0.005 0.028 0.005 0.029 D5D (20:4 n-6/20:3 n-6) X 7.9 2.3 10.5*** 2.9*** SD 1.5 0.42 2.9 0.70

Mean (X)± SD are presented, ** D6D-ratio was estimated in CE as 18:3 n-6/18:2 n-6 but in PL as 20:3 n-6/18:2 n-, *** P < 0.001 for difference between RO and SAT-diet. This might indicate that SCD is up-regulated as a consequence of a high intake of SFA, probably in order to modify intracellular levels of SFA and maintain membrane integrity. When SCD is up-regulated more 16:0 and 18:0 will be desaturated, thereby increasing the estimated desaturase indices in serum and this is in line with an observation in mice 161. An alternative possibility is that a diet high in SFA will counteract the inhibitory effect of a more PUFA rich diet on the expression of SCD 72, thus increasing desatura-tion and the SCD-16 ratio.

We propose in this study that the SCD-16-ratio (at least in CE) might be regarded as a sensitive and useful marker of dietary SFA, at least in high-fat Western populations. The SCD-18 ratio is not a very good marker, since the high dietary intake of OA will “dilute” the endogenous synthesized OA and thus affect the ratio in serum. The proportion of 16:1 was similar in the two test diets and the higher SCD-16 ratio in serum CE and PL must be affected by an increased desaturation driven by the higher SFA-content of the SAT-diet. Other inducers of SCD e.g. insulin, cholesterol and glucose 162 did not differ between the two diet-periods. However, there might be other factors involved in the up-regulation of SCD that we have not accounted for.

46

General Discussion

Biomarker fatty acids and metabolic and cardiovascular diseases In this thesis a serum FA composition, characterized by high proportions of palmitic (16:0), palmitoleic (16:1) and DHLA (20:3 n-6) acids and a low proportion of LA together with high estimated SCD and D6D and low esti-mated D5D activities, was reported to predict MetS over 20 years (study I) and cardiovascular and total mortality, independent of total cholesterol, BMI, smoking habit, physical activity and hypertension with a maximum of 33.7 years of follow up (study II).

The estimated D5D activity predicted MetS independently of BMI, physi-cal activity and smoking status, while the predictive value of the SCD [16:1/16:0] and D6D ratio was abolished by BMI. Study II suggested that endogenous FA desaturation might contribute to mortality risk since inde-pendent associations between estimated desaturase activity indices (high SCD and D6D and low D5D) and mortality were observed. In addition, the proportion of polyunsaturated fat, mainly LA, was inversely related to mor-tality, while FAs associated with a high intake of saturated fat, mirrored as high serum proportions of 16:0, 16:1 and 20:3 n-6, were directly associated with mortality. This is in analogy with previous studies 32, 68, 163.

The findings in study I and II corroborate previous studies investigating relations between biomarker FAs, estimated desaturase activities and meta-bolic and cardiovascular diseases. However, these studies add new informa-tion regarding estimated desaturase activities and the MetS and mortality in prospective settings with long follow-up. The findings also provide new information about the development of metabolic and cardiovascular diseases, since individual FAs and estimated desaturase indices were independently associated with the endpoints. The prevalence of the MetS is steadily climb-ing (as a consequence of the growing obesity epidemic) and cardiovascular diseases are the leading cause of death in the world 2, 164. The FA composi-tion and desaturase activities are modifiable by diet and this means that with the right type of dietary fat we may be able to prevent the progress of meta-bolic and cardiovascular diseases. However, the exact mechanisms behind these relationships are unknown and need to be further investigated.

47

Associations between FA composition and metabolic and cardiovascular risks have been reported in cross sectional and longitudinal epidemiological studies, conducted in different countries, age groups and in both men and women. Relations between low serum and tissue levels of LA and CHD have been reported since the 1950s and have been confirmed in later studies, but some studies have failed to demonstrate such a relationship 68. In Finland, subjects with low levels of LA and other PUFAs in serum lipids had greater risk to experience fatal or non-fatal MI 163 and a higher risk of cardiovascular death32. In another study from Finland, Salomaa et al reported serum CE proportions of 16:0, 16:1, 20:3 n-6, 18:3 n-6 and 20:4 to be higher and LA to be lower in serum CE in diabetics, compared to those with better glucose tolerance. In that study high estimated activity of SCD and D6D and low estimated activity of D5D were related to the degree of glucose intolerance 159. Correspondingly were the SCD[16:1/16:0] and D6D ratios positively, while the D5D ratio was inversely related to diabetes incidence in a recent Australian prospective case-cohort study 165. Lovejoy et al reported 14:0, 16:1 and 20:3 n-6 acids in serum CE and PL to be related to markers of insu-lin resistance in men and women in the US 158 and a similar FA composition has been associated with the risk of diabetes in several populations 85, 88, 166.

In study IV, saturated fat was able to “induce” the same serum FA pattern (including estimated desaturase activities) that we previously established as a risk marker in the development of MetS (study I) and cardiovascular disease and subsequent mortality (study II). The combined results from study I, II and IV underline the potential importance of dietary fat quality in the devel-opment of MetS, CVD and total mortality. Taken together, these results sup-port current dietary guidelines 16 and imply that replacing saturated fat with polyunsaturated fat may be advised in the prevention of metabolic and car-diovascular diseases.

Is the SCD [16:1/16:0] ratio a potential marker of saturated fat intake? In study IV we propose that the 16:1/16:0- ratio, especially in CE, might be regarded as a useful and sensitive marker of dietary SFA in Western high fat consuming populations. This marker could be potentially important and use-ful since it is widely recognized that dietary fat is difficult to assess with traditional dietary assessment methods 45. Why is it then better to use a ratio rather than an individual FA as a marker of SFA? Both the proportion of 16:1 and the 16:1/16:0 ratio have been related to metabolic and cardiovascu-lar risks with similar strength, as observed in study II. However, we believe that the SCD-16 ratio better mirrors the physiological function (adaptation)

48

of the “dietary stress” that a high intake of saturated fat causes. 16:0 is not a good marker since dietary 16:0 is rapidly desaturated (fast and effectively) to 16:1 in the body. Also, there is some endogenous production of 16:0 from carbohydrate but this contribution is small from a Western high fat diet 8. The possibility to use the 16:1/16:0 ratio as a marker of dietary SFA should be further explored in future controlled studies investigating biomarkers in relation to dietary fat intake.

Should SCD be estimated as 16:1/16:0 or 18:1/18:0? Estimated SCD activity can be estimated either as 18:1 (n-9)/18:0 or as 16:1 (n-7)/16:0. SCD is highly regulated by dietary, hormonal and other factors and regulates in turn the abundance of MUFA 72. The preferential substrate for SCD is 18:0 64 and one might wonder why the synthesis of the most abundant FA in the diet, oleic acid, needs to be so carefully regulated ? SCD1 is involved in FA partitioning and there is a requirement of endoge-nously synthesized MUFA for the synthesis of TAG 167 and cholesteryl es-ters 66 in the liver. In addition, endogenously synthesized MUFA mediates the induction of de novo lipogensis 81.

Both ratios have been associated with risks 78, 79, 90, 165, 168 but in our pro-spective studies the 18:1 (n-9)/18:0 ratio did not predict risk. This might be explained by a much higher dietary proportion of 18:1 compared to 16:1 and this will “dilute” the endogenously synthesized OA (18:0�18:1 n-9). This dilution affects the ratio in serum and a high ratio may be more compatible with a healthy diet28 and therefore does not predict risk in the same way as the 16:1 (n-7)/16:0.-ratio. In a recent Australian study, the 16:1 (n-7)/16:0 ratio was positively related to, while the 18:1 (n-9)/18:0 inversely related to the incidence of diabetes 165, in line with this interpretation and the findings in our studies and to previous studies in ULSAM.

Genetic variations in the SCD1-gene Study III of this thesis is one of the first studies completed in man investigat-ing genetic variations in the SCD1 gene and phenotypes. Study III demon-strated that subjects who were homozygous for the rare allele of rs7849 had on average 23% higher insulin sensitivity and 4 cm less WC, compared to those who were homozygous for the common allele. A finding that could potentially be clinically relevant. These results accord well with animal data 75. We tried to disentangle whether the effect of the SCD1 gene variations was primarily on obesity or insulin sensitivity. Our results suggested that the effect was primary on obesity and secondary on insulin sensitivity and this is consistent with the temporal relationship between obesity and insulin resis-

49

tance 97. It has previously been suggested that abdominal obesity is more genetically determined than general obesity 169, which the results in study III are in line with. The associations between the estimated SCD [16:1/16:0] activity and SNPs in the SCD1 gene were however a bit disappointing, but maybe not too surprising due to limited power in the analyses and since we did not measure true enzyme activities. It is possible that the FA desaturation by SCD1 is not under strong genetic control but more related to lifestyle (as suggested in study I). It is also possible that SCD1 affects metabolic proc-esses through other mechanism than by regulating FA composition.

Our data suggests that genetic variations in the SCD1 gene might be re-lated to obesity-induced phenotypes, as suggested in animal studies 76, 170. However when interpreting data from a genetic association study one has to remember that these associations might be driven by LD 171 to another func-tional polymorphism or to a nearby gene. In addition, we studied genotype-phenotype interactions in 70 year old men. This means that the influence of genetic effects on phenotypes might be less pronounced due to the healthy cohort effect and since we did not use a midlife sample (if we assume that the genetic effect would be more pronounced at a younger age).

Changes in desaturase activities, Cause or consequence? It is tempting to speculate about cause and consequence when trying to in-terpret the results from the two prospective studies in this thesis. It is plausi-ble that high SCD and D6D and low D5D activities lead to metabolic distur-bances that later is manifested as MetS (study I) and cardiovascular disease (study II).

Diet, degree of physical activity, genetics or perinatal “imprinting” (epi-genetic phenomenon 79) might cause changes in desaturase activities. This might lead to a deranged FA handling possibly influencing 1) ectopic fat accumulation (SCD) 74, 2) pro-inflammation and blood coagulation (D6D and D5D) 172 or 3) membrane FA composition that have an effect on trans-port processes and receptor binding affinities 173, 174. These mechanisms may in turn influence e.g. the development of insulin resistance and atherosclero-sis that will later be manifested as MetS, type 2 diabetes or CVD. It is also possible that the causative pathway is directed in the opposite way; meta-bolic disturbances lead to adaptive changes in desaturating enzymes in order to cope with an aberrant situation.

Thus, high SCD activity might be good in certain situations in order to maintain membrane structures, cellular communications and to keep intracel-lular levels of SFA and other metabolites (e.g. ceramide) within the normal range. An experimental study suggested that the induction of SCD1 in obese

50

insulin-resistant phenotypes may be a consequence of insulin resistance (rather than the cause) and may be initiated to counterbalance further insulin resistance 175. However, in another study, over-expression of SCD1 in rats induced insulin resistance 176 and yet another study in rats, examining the role of SCD1 in the onset of diet-induced hepatic insulin resistance, demon-strated that a decrease in SCD1 activity per se improved insulin action in the liver 177.

Risérus et al studied effects of rosiglitazone (a PPAR� receptor agonist and thus an insulin-sensitizing drug) on SCD1 mRNA expression in adipose tissue and estimated desaturase activities (SCD-16, D6D and D5D) in serum TAG 80. After rosiglitazone treatment SCD1 mRNA expression in adipose tissue had increased by 48% and the estimated plasma SCD and D5D activi-ties were increased, compared to placebo. The change in SCD-index was significantly correlated with the improvement in insulin sensitivity during the study. These results suggest that a high SCD expression might be benign when the body needs to activate de novo lipogenesis in subcutaneous adipose tissue in order to decrease the FA burden from other organs and in fact ame-liorate insulin sensitivity. In addition, this system probably needs to be care-fully regulated to be able to turn down the SCD activity when there is no need for de novo lipogenesis. Inflexibility of this system might lead to ec-topic lipid deposition and metabolic disturbances manifested as e.g. lipotox-icity, which might predispose for MetS and CVD 160, 178.

Hypothetically, a high intake of saturated fat (at the expense of unsatu-rated fat) may trigger SCD activity in order to inhibit accumulation of satu-rated fats, as observed in study IV. If this effect is persistent, including in-creased and persistent induction of SCD in other tissues, like abdominal fat depots, liver and skeletal muscle, this may be harmful. It is known that satu-rated fat may induce insulin resistance in humans 28 and this effect might be associated with changes in desaturase activities, as proposed in study IV. Mechanisms that link saturated fat intake to insulin resistance, include influ-ences on membrane fluidity, changes in lipogenic gene transcription, the type of FA within TAG and direct interference with insulin signaling via accumulation of lipid metabolites such as ceramide 157.

The D6D/D5D pathway and eicosanoid signaling AA is the major FA synthesized by the D6D/D5D pathway. Other important FAs synthesized by this pathway are DHLA, EPA and DHA. In many tissues AA is esterified at the sn-2 position of membrane PL and can be released from this position to participate in eicosanoid signaling. This release is stimulated by cell injury and other factors. AA is then enzymatically con-verted to eicosanoids 63. The eicosanoids is a group of closely related sub-

51

stances such as prostaglandins, thromboxanes, prostacyclins and leukotrienes that work as hormones and mediates local reactions such as inflammation and hemostasis and participates in the regulation of blood pressure, lipolysis and maintains the digestive tract epithelium. Eicosanoids are also derived from DHLA and EPA. Eicosanoids derived from AA and EPA can have opposite effects; e.g. leukotrienes derived from EPA have less pro-inflammatory properties, compared to those derived from AA 63, 172. An in-creased synthesis of eicosanoids from AA at the expense of eicosanoids from the n-3 family has been implicated in thrombus formation and inflammatory processes 9. An imbalance in the D6D/D5D pathway may thus be implicated in metabolic and cardiovascular diseases and may partly explain the results in study I and II.

A low proportion in serum lipids of LA has been related to the risk for type 2 diabetes 85, 159, 165, 166and has been attributed to a high intake of SFA and low intake of LA. An alternative explanation for a low LA in serum lipids could be that in the pre-diabetic state preceding manifest diabetes the high insulin levels might induce D6D thereby depleting LA and accumulat-ing DHLA in serum. This is consistent with a high estimated D6D activity, that dietary LA was positively related to diabetes risk and that the risk was most pronounced in those with higher insulin levels, as observed by Hodge et al 165. This has also been discussed by Wang et al 85 and Vessby et al 166. In these studies the associations between LA and diabetes were adjusted for baseline insulin levels and this did not affect the results, pointing at the first explanation to be the most adequate.

How do estimated enzyme activities in plasma reflect corresponding activities in other tissues?

Desaturase indices as evaluated in serum lipids probably reflect both he-patic and adipose tissue activities 67. We used fasting serum lipid ester com-position to estimate desaturase activities in all of our studies. This means that the systemic desaturase indices in our studies probably mainly reflect hepatic desaturation. This is very important to remember when interpreting data regarding estimated desaturase activities.

This also mean that, what is a true association between estimated serum desaturase activities (if mainly reflecting liver activities) and certain pheno-types might not be true for estimated desaturase activities in adipose tissue and the same phenotypes. In the physiological situation in Western societies with a relatively high dietary fat consumption, an increased SCD activity in adipose tissue may be positive and related to a normal adipo- and lipogene-sis. Contrary, an increased SCD activity in the liver might indicate an unde-

52

sirable situation with an increased lipid load and need for increased desatura-tion.

A recent study by Sjögren 179 et al highlights the complexity regarding de-saturase activities, estimated desaturase activities and insulin resistance. In this study, subcutaneous mRNA expression levels of SCD, D6D and D5D were analyzed in relation to adipose tissue desaturase indices and insulin resistance. The expression of SCD but not D6D and D5D was related to the estimated desaturase activities. Insulin resistant individuals had a higher adipose tissue 18:1/18:0 but not 16:1/16:0 (although tightly correlated), com-pared to insulin sensitive individuals. This relationship was independent of WC and TAG 179. Since this was a cross-sectional study it is not possible to say whether insulin resistance is the consequence or the cause of a high SCD activity. It is interesting that the 18:1/18:0 but not 16:1/16:0 in subcutaneous adipose tissue predict insulin resistance. This is in contrast to the results in study I, in which SCD was estimated in serum CE. To bear in mind is that oleic acid is the most abundant FA in adipose tissue (>50%) 63. It is thus possible that this FA pool is not as sensitive to dietary and metabolic changes, as other pools.

In the same study population as discussed above 179 , the correlation be-tween SCD[18:1/18:0] in serum PL and adipose tissue TAGs was not sig-nificant meaning that the short- and long-term reflections of SCD-18 are not analogous in this particular study population. In addition, the SCD-16 ratio was well correlated between fractions and tissues (NEFA, serum PL and adipose tissue TAG) (unpublished results).

Estimated desaturase activities in different serum lipid fractions

In this thesis, FAs in serum CE were measured in all studies, except in study IV where the FA proportions were measured in both CE and PL. The reason why we choose to study FAs in CE and not in any of the other lipid fractions is that this was the only fraction originally available in the UL-SAM-population. However, it is known that the proportion of FAs in all four serum lipid fractions (TAG, NEFA, CE and PL) is affected similarly by a dietary intervention; a diet high in LA will raise the amount of LA in all lipid fractions 68. As observed in study IV, qualitative changes in the proportions of individual FA in serum CE and PL due to the dietary interventions were analogous. In TAG, NEFA and PL these changes will only take a few days while changes in CE will take up to several weeks 68. Correlations between FAs in the diet and in CE are moderate 56, 68, but usually stronger than for those between FA of PLs and diet 68. In addition the short- and long-term reliability (repeatability) was higher for most major FA in CE than in PL 180.

53

This suggests that especially CEs may serve as a reasonably good biomarker of dietary fat quality, but PL may also be used 181. Also, the method variabil-ity was low in our studies (like in other studies), suggesting that the changes in FA composition we observed in study IV was due to dietary changes and endogenous metabolism. It is possible that the CE fraction is better to choose when studying the SCD index (as indicated in study IV), and PL when studying D5D and D6D indices, considering the functions of the product FA 63. However, this needs to be further explored in future intervention studies.

Are estimated desaturase activities valid markers of desaturase activities? In all studies of this thesis surrogate measures of desaturase activities have been used. This is of course a limitation. Ideal would be to measure mRNA levels or protein expression of the desaturase genes. However, this is diffi-cult to do in larger epidemiologic studies, but to some extent possible in smaller experimental studies (but not done in study IV). Nonetheless, there might be an ethical concern regarding the measurement in humans since the most relevant organ to study probably is the liver. Otherwise it is possible to measure desaturase expression in adipose tissue 80 and skeletal muscle 79. But we have to bear in mind that a high SCD activity might be good in adi-pose tissue during certain conditions but at the same time harmful in the liver.

To use an estimated desaturase activity as a surrogate measure of enzyme activity is widespread. It is known from studies using animal or in-vitro models that changes in SCD expression (mRNA or protein) is coupled with simultaneous changes in tissue FA ratios 71, 78, 79, 157, 177 and the cellular activ-ity of an SCD1 inhibitor was assessed by using the SCD index in human liver cells182. The relationship between D6D and D5D expression and FA ratios is not as widely investigated. However, it is known that SNPs and haplotypes in the FADS1 and FADS2 genes, that codes for D5D and D6D, respectively, are associated with the proportion of serum PL FAs belonging to the D5D and D6D pathways (Figure 2). The authors pointed out that their findings highlighted the importance of the desaturation pathways on n-6 and n-3 PUFA levels in PL and that this was under genetic control. The minor allele of several SNPs in these genes was associated with a decrease in the level of the desaturase product FA while the precursor FA was increased; indicating a decline in the desaturase activities (mRNA or protein) 183. This finding implies that D5D and D6D indices of may serve as a fairly good estimate of actual desaturase activities.

54

Clinical utility and implications for public health Biomarker FAs and estimated desaturase activities are independent risk fac-tors/markers in the development of Mets and in cardiovascular disease and subsequent mortality. This may indicate that the FA composition marks a state of pre-metabolic abnormalities that could determine future risk. This could have clinical implications. However, biomarker FAs are relative measures and therefore not very useful for the clinical setting, since this would require a cut-off value in order to determine the risk. Also, the predic-tive values of biomarker FAs and estimated desaturase activities were ob-tained on the group level, which means that they may not predict risk on the individual level.

To measure FA in serum fractions is cumbersome and involves several steps before the results are obtained 5. One way of facilitating the measure-ment of FA could be to use whole blood samples as suggested by Baylin et al 184 since this does not require separation of the different lipid fractions. From a public health perspective biomarker FAs and estimated desaturase activities are important since they are modifiable by diet and have been as-sociated with health outcomes in different populations. The serum FA com-position data itself could possibly be used as a preventive incentive to adhere to dietary recommendations. To use the 16:1/16:0 ratio as a marker of satu-rated fat intake may be potentially important in dietary surveys and interven-tion studies.

Epidemiological causation and considerations Epidemiology uses quantitative methods to study diseases in different popu-lations to provide information for prevention and control efforts in order to improve health 185.

The nature of a cohort study, such as the ULSAM study, is observational and causal inferences about the associations between FA biomarkers, desatu-rase activities and outcomes are hence difficult to make. However, prospec-tive cohort studies generally provide the best epidemiological evidence (be-sides experimental studies) about causation and the most direct measurement of the risk of developing disease. Before judging causality one has to ex-clude the possibility that bias, chance or confounding generated the results. 185 In our prospective studies (study I and II) differences in biomarker FA and desaturase activities were already manifest at baseline (in many times independent of confounding factors), fulfilling the temporal relationship for causality. FAs in serum and tissues and FA ratios might, however, be in the casual pathway where one factor leads to another factor until manifest dis-ease (including factors that have not been measured). This could be plausible when considering the multifactorial nature of cardiovascular disease and

55

MetS. To judge causality in an epidemiological study one has to take many factors, besides the temporal relationship, into consideration; plausibility of the results, consistency with previous studies, dose-response relationships, strengths of the associations, the type of study design and if there are several lines of evidence that leads to the same conclusion 185.

The role of genetic association studies in the uncovering and description of genes causative to common multifactorial traits remains controversial. There has been a problem to confirm findings from one population in an-other population. Genetic effects of individual candidate genes on multifac-torial traits are unlikely to be large and thus many studies have had limited power to detect true susceptibility effects. In order to overcome such prob-lems candidate genes and SNPs must be carefully chosen and the study population should be homogenous and sufficiently large and the results rep-licated in other populations of similar ethnic origin. A case-control study design is preferred for genetic association studies 125. However, we decided to use a cross-sectional design using all cohort participants instead of a nested case-control design in order not to loose power. However, in the cross-sectional study it is not possible to say anything about the causality.

Study IV was a dietary cross-over intervention study which has the ability to prove causality 185 , but selection of study participants, study power and duration influence the interpretation of the results from such a study.

Strengths and limitations ULSAM studies The strengths include that ULSAM is a large community based study sample with long follow up and with detailed phenotypic and genotypic characteri-zation of the study subjects and minimal losses to follow-up. To our knowl-edge ULSAM is the largest study sample with the gold standard measure-ment of insulin sensitivity, the euglycemic insulin clamp technique 137. The clamp measurements were done in 95% of those who participated in the 70 years investigation. The ULSAM population is homogenous (sex, age and geographically matched) and of the same ethnical background, decreasing the risk for population selection bias. Further, the analyses were specified a priori and hypothesis-driven and adjustments for multiple testing were made in study II in order to reduce the risk for type I errors.

The most obvious limitation of the ULSAM population is that it only con-sisted of men of the same age and ethnical background, decreasing the pos-sibility that the results can be generalized to other populations. When UL-SAM was initiated in 1970 the general believe was that risk for cardiovascu-lar disease was much greater in middle aged men than in women. Thus, only men were enrolled in several cohort studies including ULSAM.

56

To be able to detect effects of a genetic polymorphism in a genetic asso-ciation study (Study III) the sample size needs to be sufficient but the present study population was not very large 125, which might have driven the results toward the null hypothesis. It has, however, been suggested by Hattersley and McCarthy that small studies are likely to overestimate the true effect size 125. The SNPs genotyped in study III were able to capture most of the genetic variability in the SCD gene (75%), but there is a possibility that we might miss genetic variation.

Participants in a cohort study are subject to investigations and this might lead to a “healthy cohort effect” in terms of health and risk of disease, com-pared to the general population. This should be kept in mind when using data from a cohort study. For our prospective studies we used data from the baseline investigation unlikely to have been affected by this effect.

There is a possibility that cardiovascular mortality was misclassified but the accuracy of the Swedish cause-of-death registry is known to be high 186. Strengths of Study IV include the cross-over design and that the dietary in-tervention was strictly controlled and that all meals were provided to the participants. A limitation is that compliance was not measured in any other way than in changes in the serum FA composition, but effects on FA compo-sition indicated that the participants kept to the intervention diet. Also, the study was not very large and results should be validated in a longer term and larger intervention study. Last but not the least, in all four studies a limitation is the use of FA ratios as surrogate measures of enzyme activity. However, there are many studies that have indicated that changes in FA ratios, enzyme activities and mRNA ex-pression follow each other, as discussed above.

Future directions In this thesis we have shed light on the complex relationships between die-tary fat as mirrored in serum biomarker FAs, endogenous desaturation and the risk of metabolic and cardiovascular diseases. We have also investigated genetic variation in the SCD1 gene in association to obesity and insulin re-sistance, phenotypes that are the central drivers in the development of meta-bolic and cardiovascular diseases. In the intervention study we were able to demonstrate how estimated desaturase activities change in response to dif-ferent fat qualities. These four studies contribute to the present knowledge and support current dietary guidelines but there are still questions to be an-swered. The estimated desaturase activities should be validated with in vivo measurements of mRNA/protein expression. Estimated desaturase activities should in a longer term dietary study be analyzed in associations to meta-

57

bolic changes, e.g. insulin sensitivity, to be able to verify changes in enzyme activities. The mechanisms behind the independent associations between estimated desaturase activities and metabolic and cardiovascular risks should be investigated in future studies. It is possible that we will combat the devel-opment of metabolic and cardiovascular diseases or at least stop the progres-sion of them, with pharmacological interventions directed against desatu-rases 170, 172, 182. Pharmacological inhibitors of the SCD1 gene are already available and are currently tested in preclinical studies and others underway but the clinical utility have to be proven in the future.

58

Conclusions

In this thesis serum biomarker FAs and estimated SCD, �6 and �5 activities were explored in different settings. This thesis confirms and extends the role of serum biomarker FAs and estimated SCD, �6 and �5 activities as possible important players in metabolic and cardiovascular diseases and demonstrates that genetic variations in the SCD1 gene might be associated with obesity-induced phenotypes. The estimated SCD, D6D and D5D activity predicted MetS over 20 years. The relation between D5D and MetS was independent of BMI, physical ac-tivity and smoking status, while the predictive value of the SCD [16:1/16:0] and D6D was mainly explained by BMI. Endogenous FA desaturation might contribute to mortality risk on the group level since independent associations between estimated desaturase activity indices and mortality were observed. The proportion of LA in serum was inversely related to mortality, while FAs associated with a high intake of saturated fat (16:0, 16:1 and 20:3 n-6) were directly associated with mortal-ity. The rare allele of rs7849 of SCD1 demonstrated the strongest effect on both insulin sensitivity and waist circumference, corresponding to 23% higher insulin sensitivity and 4 cm less waist circumference. Saturated fat was able to “induce” the same serum FA pattern (including estimated desaturase activities) that predicted the development of MetS and mortality risk. We also propose that the SCD ratio [16:1/16:0], especially in CE, might be regarded as a useful and sensitive marker of dietary SFA, in Western populations with a high-fat intake. The combined results of this thesis emphasize the potential importance of dietary fat quality in the development of metabolic and cardiovascular dis-eases and support current dietary guidelines 16. Replacing saturated fat with polyunsaturated fat may be advised in the prevention of metabolic and car-diovascular diseases. However, this needs to be further evaluated in other observational studies and preferable formally tested in randomized con-trolled trials.

59

Sammanfattning på svenska (Summary in Swedish)

I dag är över en miljard människor i världen överviktiga (BMI >25 kg/m2) och av dessa är fler än 300 miljoner feta (BMI >30 kg/m2). Övervikt och fetma leder till ökad risk för sjukdomar såsom metabola syndromet (högt blodtryck, förändrade blodfetter, störd insulin- och glukosomsättning och fetma), hjärtkärlsjukdomar och typ 2 diabetes, och kan leda till för tidig död.

Förekomst av dessa sjukdomar kan till viss del tillskrivas fettet i kosten. För att utveckla dessa sjukdomar tycks typen (kvalitet) av fett i kosten större betydelse, än mängden. Fett är ett mycket viktigt näringsämne som behövs för att hålla membranstrukturer smidiga, för kommunikation mellan celler och reglering av genuttryck. Dessutom ger fett kroppen energi och bygger upp fettdepåer med många viktiga funktioner såsom isolering och produktion av hormonliknande substanser.

Fett i kosten består i huvudsak av triglycerider som i sin tur består av gly-cerol förbundet med tre fettsyror. Fettsyror kan vara mättade (t ex 16:0, pal-mitinsyra), enkelomättade (t ex 18:1, oljesyra; 16:1, palmitoljesyra) eller fleromättade (t ex 20:3 n-6, dihomo-�-linolensyra). Mättnadsgraden bestäms av antalet dubbelbindningar i fettsyran och ju fler dubbelbindningar desto mjukare är fettsyran. Omättade fettsyror har egenskaper som gör att de kan reglera flera viktiga processer inne i cellen. Linolsyra (18:2 n-6) och �-linolensyra (18:3 n-3) är livsnödvändiga fettsyror som måste tillföras krop-pen med kosten. Andra fettsyror kan kroppen själv bilda med hjälp av enzy-mer såsom desaturaser och elongaser.

Det är svårt att studera intaget av fett i kosten hos en grupp individer med hjälp av vanliga kostundersökningsmetoder eftersom risken är stor att perso-nerna, medvetet eller omedvetet, inte rapporterar vad de egentligen ätit. Fett-syror i blodlipider eller vävnadsfraktioner kan användas som biologiska mar-körer, så kallade biomarkörer, för att objektivt kunna studera typen av fettsy-ror i kosten. Dessa biomarkörer kan sedan användas för att studera sjuk-domsrisk i förhållande till kostens typ av fett.

I denna avhandling har jag studerat fettsyror i blod och fettsyrakvoter, som mått på desaturasaktivitet, i relation till 1) risken att utveckla det meta-bola syndromet under 20-års uppföljning (study I) och 2) risk för dödlighet under en uppföljning upp till 33.7 år (study II). Jag har dessutom studerat hur desaturaskvoter (stearoyl-CoA-desaturas (SCD), �6-desaturas (D6D) och �5-

60

desaturas (D5D)) och individuella fettsyror påverkas av två typer av kost där endast typen av fett skiljer sig. Vi jämförde effekter av en kost rik på mättat fett med en kost rik på omättat fett baserad på rapsolja (study IV). Slutligen studerades genetisk variation i SCD1 genen i relation till fetma, insulinkäns-lighet (hur bra kroppen svarar på insulinets verkan) och SCD-kvoten (study III).

Tre av studierna gjordes med data från Uppsala Longitudinal Study of Adult Men (ULSAM) som är en populationsstudie som startades 1970. Alla män som var födda 1920-24 och bodde i Uppsala med omnejd 1970 inbjöds att delta i undersökningen och hela 82 % deltog. Sedan 1970 när männen var 50 år har uppföljningar gjorts vid 60, 70, 77 och 82 års ålder. I april 2008 kommer 88-årsundersökningen att sättas igång. ULSAM är ett unikt forsk-ningsmaterial för att kunna studera risk för sjukdom och dödlighet.

I denna avhandling visar jag att ett visst mönster av fettsyror i blod, höga proportioner av 16:0, 16:1 och 20:3 n-6 och låga proportioner av 18:2 n-6, samt hög SCD och D6D och låg D5D kvot, är kopplat till risk för att utveck-la metabola syndromet och leder till ökad risk för dödlighet. Riskerna var oberoende av etablerade riskfaktorer för hjärtkärlsjukdom. Dessa resultat harmonierar med tidigare resultat men ger ny information om fettsyror och desaturaskvoter i relation till metabola syndromet och dödlighet i prospekti-va (framåtblickande) studier med lång uppföljningstid. Eftersom många av sambanden var oberoende av etablerade riskfaktorer kan detta tyda på att desaturas-enzymer tillsammans med typen av fett i kosten kan bidra till ut-vecklingen av metabola och hjärtkärl-sjukdomar på gruppnivå.

I koststudien, ”inducerades” det ovan nämnda mönstret av fettsyror och fettsyrakvoter av en kost rik på mättat fett. I koststudien lanseras även SCD-kvoten [16:1/16:0] som en möjlig biomarkör på mättat fett i kosten. Detta är potentiellt ett viktigt bidrag till kostundersökningar som ska verifieras i framtida studier.

Resultaten i dessa tre studier (I, II och IV) visar på betydelsen av rätt fett i kosten för att förhindra uppkomst av metabola och hjärtkärl-sjukdomar. Re-sultat i denna avhandling överensstämmer med dagens kostrekommendatio-ner som uppmanar oss att minska på mättat fett och att samtidigt öka intaget av omättat fett.

I studie IV visades att genetisk variation i SCD1 genen var kopplad till fetma och graden av insulinkänslighet, som överensstämmer med resultat från djurmodeller. Däremot var SCD-kvoten dåligt associerad till variationen i SCD1 genen. SCD enzymet ses som en möjlig kandidat för farmakologisk behandling för att förhindra/lindra utveckling av risker relaterat till fetma. Om detta är möjligt kommer framtiden att utvisa.

61

Acknowledgements

I would like to thank all of you, who made it possible to write this thesis and have contributed with scientific, everyday, computer and statistical knowl-edge and for making the “fika” and lunch breaks much more fun than with-out you! I wish to express my sincere gratitude especially to: Ulf Risérus, my principal supervisor for your scientific guidance and for your enormous research interest and sharing your knowledge in the field of diseases and nutrition and for your ability to remember journal references! Bengt Vessby, supervisor and the former Professor of section of Clinical Nutrition Research, who introduced me to fatty acids and desaturases and for sharing your vast knowledge in the field of nutrition, fatty acids, obesity related and cardiovascular diseases. I also appreciated all your constructive and timely comments on all my manuscripts and for being a good scientific mentor in general. Tommy Cederholm, assistant supervisor and current Professor of Clinical Nutrition and Metabolism for providing valuable comments, advice, support and for creating a stimulating and creative environment to work in. Lars Lannfelt, Professor of Geriatrics and head of the ICTUS group and co-author, for comments, valuable support and advice. Rawya Mohsen, co-author and for being an excellent database manager and for all help with computers, datasets, everyday stuff and everything else! I wish to thank my co-authors: Erik Ingelsson, Per Lundmark, Ann-Christine Syvänen, Johan Sundström and Inga-Britt Gustafsson for valuable comments and advice, statistical assistance, bioinformatics input and SNP genotyping. All the ULSAM men who participated in the five study investigations provid-ing so much data for my research!

62

My present and former colleagues at the sections of Clinical Nutrition and Metabolism and Geriatrics: Annika Smedman for being the perfect room mate and lecture-partner, for fun times when we attended “Årets Kock” and chats in general! Agneta Andersson for being a good and cheerful discussion partner and for sharing the room with me at various trips and for fun shopping together! Björn Zethelius for your proof-reading, fun and thoughtful discussions and providing scientific advice. Johanna Helmersson for your help with STATA, data and research questions in general and for all merry lunches (preferable salads and sushi!) together at Kålsängsgränd. Elisabet Rytter, for chats and being a good travel partner to Barcelona. Helena Peterson, for having mutual interests in the field of fatty acids and SCD1, for recipes and cheerful discussions!! Liisa Byberg for all help with web and statistics related questions, the fantas-tic shawl and everything else. Johan Sundelöf for taking the time to check my blood pressure and checking my red blood cell count when I was anaemic and nauseous and for all good laughs! Ulf Holmbäck, Anja Saletti, Wulf Becker, Brita Karlström, Samar Basu, Ann-Christin Åberg, Maria Lindau, Elina Rönnemaa, Johan Ärnlöv, Thomas Cars, Martin Wohlin, Bernice Wiberg, Lena Kilander, Cecila Nälsén, Johanna Törmä, Kristina Björklund-Bodegård, Christina Ström-Möller, Klara Edlund, Ylva Cedervall, Per Sjögren, Siv Tengblad, Barbro Simu, Eva Sejby, Tamanna Ferdous and the ICTUS-collaboration. I would also like thank all my friends, significant others and all your won-derful kids. My special thanks to the Uppsala-girls (Karin, Ninna, Louise, Josefin, Katti and Petra), the ÖG-team (gäng), Kristin and Anna for lunch dates and Anna Vallstedt, old “biomedicinare kompis” for fun lunches and work-outs in the gym! My gratitude to my second families in Ukiah! I would also like to express my deepest thanks to my family for total sup-port, discussions, and help with the kids, fun family dinners, “fikas”and good company! Pappa Arne, my twin-sister Anna, Rikard, my brother Lars, Eva, Julia, Ville, Emil, Nils, my brother Mats, Nilmeni, Ebba, Stina, Margareta, Jan, Karin P, Karin S, Alarik, Nora and Gustav.

All min kärlek till Fredrik, Mira och Melker � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �

63

References

1. Van Gaal LF, Mertens IL, De Block CE. Mechanisms linking obe-sity with cardiovascular disease. Nature. Dec 14 2006;444(7121):875-880.

2. Vasudevan AR, Ballantyne CM. Cardiometabolic risk assessment: an approach to the prevention of cardiovascular disease and diabetes mellitus. Clinical cornerstone. 2005;7(2-3):7-16.

3. Adams KF, Schatzkin A, Harris TB, Kipnis V, Mouw T, Ballard-Barbash R, Hollenbeck A, Leitzmann MF. Overweight, obesity, and mortality in a large prospective cohort of persons 50 to 71 years old. N Engl J Med. Aug 24 2006;355(8):763-778.

4. Sjostrom L, Narbro K, Sjostrom CD, Karason K, Larsson B, Wedel H, Lystig T, Sullivan M, Bouchard C, Carlsson B, Bengtsson C, Dahlgren S, Gummesson A, Jacobson P, Karlsson J, Lindroos AK, Lonroth H, Naslund I, Olbers T, Stenlof K, Torgerson J, Agren G, Carlsson LM. Effects of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med. Aug 23 2007;357(8):741-752.

5. Arab L. Biomarkers of fat and fatty acid intake. J Nutr. Mar 2003;133 Suppl 3:925S-932S.

6. Bray GA, Popkin BM. Dietary fat intake does affect obesity! Am J Clin Nutr. Dec 1998;68(6):1157-1173.

7. Willett WC. Dietary fat and obesity: an unconvincing relation. Am J Clin Nutr. Dec 1998;68(6):1149-1150.

8. Frayn KN. Metabolic Regulation -A human perspective. Second edition ed. Oxford, UK: Blackwell Science Ltd; 2003.

9. Abrahamsson L, Andersson A, Becker W, Nilsson G. Näringslära för högskolan [Nutrition for higher education]. 5th ed. Stockholm: LIBER AB; 2006.

10. Riksmaten. Kostvanor och näringsintag i Sverige. Metod och reultat analys. (National survey of dietary habits and nutrient intake in Sweden). Uppsala, Sweden: Livsmedelsverket (National food admi-nistration); 1997-98.

11. Wu Z, Palmquist DL. Synthesis and biohydrogenation of fatty acids by ruminal microorganisms in vitro. J Dairy Sci. Sep 1991;74(9):3035-3046.

12. Renaud S, Lanzmann-Petithory D. Coronary heart disease: dietary links and pathogenesis. Public Health Nutr. Apr 2001;4(2B):459-474.

13. Doyle E. Chemistry behind the News, Trans fatty acids. Journal of Chemical Education. 1997;74(9):1030-1032.

64

14. Riserus U. Trans fatty acids and insulin resistance. Atheroscler Suppl. May 2006;7(2):37-39.

15. Wijendran V, Hayes KC. Dietary n-6 and n-3 fatty acid balance and cardiovascular health. Annu Rev Nutr. 2004;24:597-615.

16. Nordic Nutrition Recommendadtions 2004. Integrating nutrition and physical activity. 4th ed. Copenhagen, Denmark: Nordic Coun-cil of Ministers; 2004.

17. Keys A, Menotti A, Karvonen MJ, Aravanis C, Blackburn H, Buzina R, Djordjevic BS, Dontas AS, Fidanza F, Keys MH, et al. The diet and 15-year death rate in the seven countries study. Am J Epidemiol. Dec 1986;124(6):903-915.

18. Hu FB, Manson JE, Willett WC. Types of dietary fat and risk of coronary heart disease: a critical review. J Am Coll Nutr. Feb 2001;20(1):5-19.

19. Pietinen P, Ascherio A, Korhonen P, Hartman AM, Willett WC, Albanes D, Virtamo J. Intake of fatty acids and risk of coronary heart disease in a cohort of Finnish men. The Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study. Am J Epidemiol. May 15 1997;145(10):876-887.

20. Schaefer EJ. Lipoproteins, nutrition, and heart disease. Am J Clin Nutr. Feb 2002;75(2):191-212.

21. Katan MB, Zock PL, Mensink RP. Effects of fats and fatty acids on blood lipids in humans: an overview. Am J Clin Nutr. Dec 1994;60(6 Suppl):1017S-1022S.

22. Ulbricht TL, Southgate DA. Coronary heart disease: seven dietary factors. Lancet. Oct 19 1991;338(8773):985-992.

23. Vessby B. Dietary fat and insulin action in humans. Br J Nutr. Mar 2000;83 Suppl 1:S91-96.

24. De Caterina R, Zampolli A, Del Turco S, Madonna R, Massaro M. Nutritional mechanisms that influence cardiovascular disease. Am J Clin Nutr. Feb 2006;83(2):421S-426S.

25. Kris-Etherton PM, Griel AE, Psota TL, Gebauer SK, Zhang J, Ether-ton TD. Dietary stearic acid and risk of cardiovascular disease: in-take, sources, digestion, and absorption. Lipids. Dec 2005;40(12):1193-1200.

26. Rivellese AA, Maffettone A, Vessby B, Uusitupa M, Hermansen K, Berglund L, Louheranta A, Meyer BJ, Riccardi G. Effects of dietary saturated, monounsaturated and n-3 fatty acids on fasting lipopro-teins, LDL size and post-prandial lipid metabolism in healthy sub-jects. Atherosclerosis. Mar 2003;167(1):149-158.

27. Rasmussen BM, Vessby B, Uusitupa M, Berglund L, Pedersen E, Riccardi G, Rivellese AA, Tapsell L, Hermansen K. Effects of die-tary saturated, monounsaturated, and n-3 fatty acids on blood pres-sure in healthy subjects. Am J Clin Nutr. Feb 2006;83(2):221-226.

28. Vessby B, Uusitupa M, Hermansen K, Riccardi G, Rivellese AA, Tapsell LC, Nalsen C, Berglund L, Louheranta A, Rasmussen BM, Calvert GD, Maffetone A, Pedersen E, Gustafsson IB, Storlien LH.

65

Substituting dietary saturated for monounsaturated fat impairs insu-lin sensitivity in healthy men and women: The KANWU Study. Diabetologia. Mar 2001;44(3):312-319.

29. Turpeinen O, Karvonen MJ, Pekkarinen M, Miettinen M, Elosuo R, Paavilainen E. Dietary prevention of coronary heart disease: the Finnish Mental Hospital Study. Int J Epidemiol. Jun 1979;8(2):99-118.

30. Dayton S, Pearce ML, Hashimoto S, Dixon W, Tomiyashi U. A controlled Clinical Trial of a Diet High in Unsaturated Fat in Pre-venting Complications of Atherosclerosis. Circulation. 1969;XL (July, Supplement No II).

31. Frantz ID, Jr., Dawson EA, Ashman PL, Gatewood LC, Bartsch GE, Kuba K, Brewer ER. Test of effect of lipid lowering by diet on car-diovascular risk. The Minnesota Coronary Survey. Arteriosclerosis. Jan-Feb 1989;9(1):129-135.

32. Laaksonen DE, Nyyssonen K, Niskanen L, Rissanen TH, Salonen JT. Prediction of cardiovascular mortality in middle-aged men by dietary and serum linoleic and polyunsaturated fatty acids. Arch In-tern Med. Jan 24 2005;165(2):193-199.

33. Brouwer IA, Zock PL, Camm AJ, Bocker D, Hauer RN, Wever EF, Dullemeijer C, Ronden JE, Katan MB, Lubinski A, Buschler H, Schouten EG. Effect of fish oil on ventricular tachyarrhythmia and death in patients with implantable cardioverter defibrillators: the Study on Omega-3 Fatty Acids and Ventricular Arrhythmia (SOFA) randomized trial. Jama. Jun 14 2006;295(22):2613-2619.

34. Seo T, Blaner WS, Deckelbaum RJ. Omega-3 fatty acids: molecular approaches to optimal biological outcomes. Curr Opin Lipidol. Feb 2005;16(1):11-18.

35. Schacky C, Harris WS. Cardiovascular benefits of omega-3 fatty acids. Cardiovasc Res. Jan 15 2007;73(2):310-315.

36. Giacco R, Cuomo V, Vessby B, Uusitupa M, Hermansen K, Meyer BJ, Riccardi G, Rivellese AA. Fish oil, insulin sensitivity, insulin secretion and glucose tolerance in healthy people: Is there any effect of fish oil supplementation in relation to the type of background diet and habitual dietary intake of n-6 and n-3 fatty acids? Nutr Metab Cardiovasc Dis. Nov 23 2006.

37. Harris WS, Poston WC, Haddock CK. Tissue n-3 and n-6 fatty acids and risk for coronary heart disease events. Atherosclerosis. Jul 2007;193(1):1-10.

38. Harris WS, Reid KJ, Sands SA, Spertus JA. Blood omega-3 and trans fatty acids in middle-aged acute coronary syndrome patients. The American journal of cardiology. Jan 15 2007;99(2):154-158.

39. de Lorgeril M, Salen P, Martin JL, Monjaud I, Delaye J, Mamelle N. Mediterranean diet, traditional risk factors, and the rate of cardio-vascular complications after myocardial infarction: final report of the Lyon Diet Heart Study. Circulation. Feb 16 1999;99(6):779-785.

66

40. Singh RB, Dubnov G, Niaz MA, Ghosh S, Singh R, Rastogi SS, Manor O, Pella D, Berry EM. Effect of an Indo-Mediterranean diet on progression of coronary artery disease in high risk patients (Indo-Mediterranean Diet Heart Study): a randomised single-blind trial. Lancet. Nov 9 2002;360(9344):1455-1461.

41. Dietary supplementation with n-3 polyunsaturated fatty acids and vitamin E after myocardial infarction: results of the GISSI-Prevenzione trial. Gruppo Italiano per lo Studio della Sopravvivenza nell'Infarto miocardico. Lancet. Aug 7 1999;354(9177):447-455.

42. Burr ML, Fehily AM, Gilbert JF, Rogers S, Holliday RM, Sweetnam PM, Elwood PC, Deadman NM. Effects of changes in fat, fish, and fibre intakes on death and myocardial reinfarction: diet and reinfarc-tion trial (DART). Lancet. Sep 30 1989;2(8666):757-761.

43. Mozaffarian D, Katan MB, Ascherio A, Stampfer MJ, Willett WC. Trans fatty acids and cardiovascular disease. N Engl J Med. Apr 13 2006;354(15):1601-1613.

44. Oh K, Hu FB, Manson JE, Stampfer MJ, Willett WC. Dietary fat intake and risk of coronary heart disease in women: 20 years of fol-low-up of the nurses' health study. Am J Epidemiol. Apr 1 2005;161(7):672-679.

45. Livingstone MB, Black AE. Markers of the validity of reported en-ergy intake. J Nutr. Mar 2003;133 Suppl 3:895S-920S.

46. Heitmann BL, Lissner L. Can adverse effects of dietary fat intake be overestimated as a consequence of dietary fat underreporting? Pub-lic Health Nutr. Dec 2005;8(8):1322-1327.

47. Heitmann BL, Lissner L. Dietary underreporting by obese individu-als--is it specific or non-specific? Bmj. Oct 14 1995;311(7011):986-989.

48. Lissner L, Troiano RP, Midthune D, Heitmann BL, Kipnis V, Subar AF, Potischman N. OPEN about obesity: recovery biomarkers, die-tary reporting errors and BMI. Int J Obes (Lond). Jun 2007;31(6):956-961.

49. Goris AH, Westerterp-Plantenga MS, Westerterp KR. Undereating and underrecording of habitual food intake in obese men: selective underreporting of fat intake. Am J Clin Nutr. Jan 2000;71(1):130-134.

50. Baylin A, Campos H. The use of fatty acid biomarkers to reflect dietary intake. Curr Opin Lipidol. Feb 2006;17(1):22-27.

51. Katan MB, Deslypere JP, van Birgelen AP, Penders M, Zegwaard M. Kinetics of the incorporation of dietary fatty acids into serum cholesteryl esters, erythrocyte membranes, and adipose tissue: an 18-month controlled study. J Lipid Res. Oct 1997;38(10):2012-2022.

52. Glatz JF, Soffers AE, Katan MB. Fatty acid composition of serum cholesteryl esters and erythrocyte membranes as indicators of li-noleic acid intake in man. Am J Clin Nutr. Feb 1989;49(2):269-276.

53. Zock PL, Mensink RP, Harryvan J, de Vries JH, Katan MB. Fatty acids in serum cholesteryl esters as quantitative biomarkers of die-

67

tary intake in humans. Am J Epidemiol. Jun 15 1997;145(12):1114-1122.

54. Baylin A, Kabagambe EK, Siles X, Campos H. Adipose tissue bio-markers of fatty acid intake. Am J Clin Nutr. Oct 2002;76(4):750-757.

55. Hellerstein MK. De novo lipogenesis in humans: metabolic and regulatory aspects. Eur J Clin Nutr. Apr 1999;53 Suppl 1:S53-65.

56. Ma J, Folsom AR, Shahar E, Eckfeldt JH. Plasma fatty acid compo-sition as an indicator of habitual dietary fat intake in middle-aged adults. The Atherosclerosis Risk in Communities (ARIC) Study In-vestigators. Am J Clin Nutr. Sep 1995;62(3):564-571.

57. Smedman AE, Gustafsson IB, Berglund LG, Vessby BO. Penta-decanoic acid in serum as a marker for intake of milk fat: relations between intake of milk fat and metabolic risk factors. Am J Clin Nutr. Jan 1999;69(1):22-29.

58. Wolk A, Furuheim M, Vessby B. Fatty acid composition of adipose tissue and serum lipids are valid biological markers of dairy fat in-take in men. J Nutr. Mar 2001;131(3):828-833.

59. Wolk A, Vessby B, Ljung H, Barrefors P. Evaluation of a biological marker of dairy fat intake. Am J Clin Nutr. Aug 1998;68(2):291-295.

60. Rise P, Eligini S, Ghezzi S, Colli S, Galli C. Fatty acid composition of plasma, blood cells and whole blood: Relevance for the assess-ment of the fatty acid status in humans. Prostaglandins Leukot Es-sent Fatty Acids. Jun 2007;76(6):363-369.

61. Jakobsson A, Westerberg R, Jacobsson A. Fatty acid elongases in mammals: their regulation and roles in metabolism. Prog Lipid Res. May 2006;45(3):237-249.

62. Jump DB. Fatty acid regulation of gene transcription. Crit Rev Clin Lab Sci. 2004;41(1):41-78

63. Nakamura MT, Nara TY. Structure, Function, And Dietary Regula-tion of Delta6, Delta5, and Delta9 Desaturases. Annu Rev Nutr. 2004;24:345-376.

64. Miyazaki M, Bruggink SM, Ntambi JM. Identification of mouse palmitoyl-coenzyme A Delta9-desaturase. J Lipid Res. Apr 2006;47(4):700-704.

65. Dobrzyn A, Ntambi JM. The role of stearoyl-CoA desaturase in the control of metabolism. Prostaglandins Leukot Essent Fatty Acids. Jul 2005;73(1):35-41.

66. Ntambi JM, Miyazaki M. Regulation of stearoyl-CoA desaturases and role in metabolism. Prog Lipid Res. Mar 2004;43(2):91-104.

67. Mangravite LM, Dawson K, Davis RR, Gregg JP, Krauss RM. Fatty acid desaturase regulation in adipose tissue by dietary composition is independent of weight loss and is correlated with the plasma triacyl-glycerol response. Am J Clin Nutr. Sep 2007;86(3):759-767.

68. Nikkari T. Serum fatty acids and coronary heart disease in Finnish populations. Prog Lipid Res. 1986;25(1-4):437-450.

68

69. Pan DA, Hulbert AJ, Storlien LH. Dietary fats, membrane phosphol-ipids and obesity. J Nutr. Sep 1994;124(9):1555-1565.

70. Wang J, Yu L, Schmidt RE, Su C, Huang X, Gould K, Cao G. Char-acterization of HSCD5, a novel human stearoyl-CoA desaturase unique to primates. Biochem Biophys Res Commun. Jul 8 2005;332(3):735-742.

71. Miyazaki M, Man WC, Ntambi JM. Targeted disruption of stearoyl-CoA desaturase1 gene in mice causes atrophy of sebaceous and meibomian glands and depletion of wax esters in the eyelid. J Nutr. Sep 2001;131(9):2260-2268.

72. Ntambi JM, Miyazaki M. Recent insights into stearoyl-CoA desatu-rase-1. Curr Opin Lipidol. Jun 2003;14(3):255-261.

73. Kusunoki J, Kanatani A, Moller DE. Modulation of Fatty Acid me-tabolism as a potential approach to the treatment of obesity and the metabolic syndrome. Endocrine. Feb 2006;29(1):91-100.

74. Sampath H, Ntambi JM. Stearoyl-coenzyme A desaturase 1, sterol regulatory element binding protein-1c and peroxisome proliferator-activated receptor-alpha: independent and interactive roles in the regulation of lipid metabolism. Curr Opin Clin Nutr Metab Care. Mar 2006;9(2):84-88.

75. Ntambi JM, Miyazaki M, Stoehr JP, Lan H, Kendziorski CM, Yan-dell BS, Song Y, Cohen P, Friedman JM, Attie AD. Loss of stearoyl-CoA desaturase-1 function protects mice against adiposity. Proc Natl Acad Sci U S A. Aug 20 2002;99(17):11482-11486.

76. Dobrzyn A, Ntambi JM. The Role of Stearoyl-CoA Desaturase in Body Weight Regulation. Trends Cardiovasc Med. Feb 2004;14(2):77-81.

77. Miyazaki M, Kim YC, Gray-Keller MP, Attie AD, Ntambi JM. The biosynthesis of hepatic cholesterol esters and triglycerides is im-paired in mice with a disruption of the gene for stearoyl-CoA desatu-rase 1. J Biol Chem. Sep 29 2000;275(39):30132-30138.

78. Attie AD, Krauss RM, Gray-Keller MP, Brownlie A, Miyazaki M, Kastelein JJ, Lusis AJ, Stalenhoef AF, Stoehr JP, Hayden MR, Ntambi JM. Relationship between stearoyl-CoA desaturase activity and plasma triglycerides in human and mouse hypertriglyceridemia. J Lipid Res. Nov 2002;43(11):1899-1907.

79. Hulver MW, Berggren JR, Carper MJ, Miyazaki M, Ntambi JM, Hoffman EP, Thyfault JP, Stevens R, Dohm GL, Houmard JA, Muoio DM. Elevated stearoyl-CoA desaturase-1 expression in skele-tal muscle contributes to abnormal fatty acid partitioning in obese humans. Cell Metab. Oct 2005;2(4):251-261.

80. Riserus U, Tan GD, Fielding BA, Neville MJ, Currie J, Savage DB, Chatterjee VK, Frayn KN, O'Rahilly S, Karpe F. Rosiglitazone in-creases indexes of stearoyl-CoA desaturase activity in humans: link to insulin sensitization and the role of dominant-negative mutation in peroxisome proliferator-activated receptor-gamma. Diabetes. May 2005;54(5):1379-1384.

69

81. Sampath H, Miyazaki M, Dobrzyn A, Ntambi JM. Stearoyl-CoA Desaturase-1 Mediates the Pro-lipogenic Effects of Dietary Satu-rated Fat. J Biol Chem. Jan 26 2007;282(4):2483-2493.

82. Warensjö E, Öhrvall M, Vessby B. Fatty acid composition and esti-mated desaturase activities are associated with obesity and lifestyle variables in men and women. Nutr Metab Cardiovasc Dis. Mar 2006;16(2):128-136.

83. Vessby B, Tengblad S, Lithell H. Insulin sensitivity is related to the fatty acid composition of serum lipids and skeletal muscle phosphol-ipids in 70-year-old men. Diabetologia. Oct 1994;37(10):1044-1050.

84. Vessby B. Dietary fat, fatty acid composition in plasma and the metabolic syndrome. Curr Opin Lipidol. Feb 2003;14(1):15-19.

85. Wang L, Folsom AR, Zheng ZJ, Pankow JS, Eckfeldt JH. Plasma fatty acid composition and incidence of diabetes in middle-aged adults: the Atherosclerosis Risk in Communities (ARIC) Study. Am J Clin Nutr. Jul 2003;78(1):91-98.

86. Wang L, Folsom AR, Eckfeldt JH. Plasma fatty acid composition and incidence of coronary heart disease in middle aged adults: the Atherosclerosis Risk in Communities (ARIC) Study. Nutr Metab Cardiovasc Dis. Oct 2003;13(5):256-266.

87. Tremblay AJ, Despres JP, Piche ME, Nadeau A, Bergeron J, Almeras N, Tremblay A, Lemieux S. Associations between the fatty acid content of triglyceride, visceral adipose tissue accumulation, and components of the insulin resistance syndrome. Metabolism. Mar 2004;53(3):310-317.

88. Laaksonen DE, Lakka TA, Lakka HM, Nyyssonen K, Rissanen T, Niskanen LK, Salonen JT. Serum fatty acid composition predicts development of impaired fasting glycaemia and diabetes in middle-aged men. Diabet Med. Jun 2002;19(6):456-464.

89. Vessby B, Gustafsson IB, Tengblad S, Boberg M, Andersson A. Desaturation and elongation of Fatty acids and insulin action. Ann N Y Acad Sci. Jun 2002;967:183-195.

90. Corpeleijn E, Feskens EJ, Jansen EH, Mensink M, Saris WH, de Bruin TW, Blaak EE. Improvements in glucose tolerance and insulin sensitivity after lifestyle intervention are related to changes in serum fatty acid profile and desaturase activities: the SLIM study. Diabe-tologia. Oct 2006;49(10):2392-2401.

91. Riserus U, Arnlov J, Berglund L. Long-Term Predictors of Insulin Resistance: Role of Lifestyle and Metabolic Factors in Middle-Aged Men. Diabetes Care. Jul 20 2007.

92. Information from WHO Global Infobase online. http://www.who.int/chp/en/english.html accessed on December 18 2006.

93. Olshansky SJ, Passaro DJ, Hershow RC, Layden J, Carnes BA, Brody J, Hayflick L, Butler RN, Allison DB, Ludwig DS. A poten-tial decline in life expectancy in the United States in the 21st cen-tury. N Engl J Med. Mar 17 2005;352(11):1138-1145.

70

94. Frayn KN. Obesity and metabolic disease: is adipose tissue the cul-prit? Proc Nutr Soc. Feb 2005;64(1):7-13.

95. DeFronzo RA, Ferrannini E. Insulin resistance. A multifaceted syn-drome responsible for NIDDM, obesity, hypertension, dyslipidemia, and atherosclerotic cardiovascular disease. Diabetes Care. Mar 1991;14(3):173-194.

96. Krentz AJ. Insulin resistance. Bmj. Nov 30 1996;313(7069):1385-1389.

97. Lebovitz HE. Insulin resistance--a common link between type 2 diabetes and cardiovascular disease. Diabetes Obes Metab. May 2006;8(3):237-249.

98. Adams JM, 2nd, Pratipanawatr T, Berria R, Wang E, DeFronzo RA, Sullards MC, Mandarino LJ. Ceramide content is increased in skele-tal muscle from obese insulin-resistant humans. Diabetes. Jan 2004;53(1):25-31.

99. Isomaa B. A major health hazard: the metabolic syndrome. Life Sci. Sep 26 2003;73(19):2395-2411.

100. Reaven GM. Banting lecture 1988. Role of insulin resistance in hu-man disease. Diabetes. Dec 1988;37(12):1595-1607.

101. Alberti KG, Zimmet P, Shaw J. Metabolic syndrome--a new world-wide definition. A Consensus Statement from the International Dia-betes Federation. Diabet Med. May 2006;23(5):469-480.

102. Kahn R, Buse J, Ferrannini E, Stern M. The metabolic syndrome: time for a critical appraisal: joint statement from the American Dia-betes Association and the European Association for the Study of Diabetes. Diabetes Care. Sep 2005;28(9):2289-2304.

103. Kahn R, Buse J, Ferrannini E, Stern M. The metabolic syndrome: time for a critical appraisal. Joint statement from the American Dia-betes Association and the European Association for the Study of Diabetes. Diabetologia. Sep 2005;48(9):1684-1699.

104. Reaven GM. The metabolic syndrome: is this diagnosis necessary? Am J Clin Nutr. Jun 2006;83(6):1237-1247.

105. Sundstrom J, Vallhagen E, Riserus U, Byberg L, Zethelius B, Berne C, Lind L, Ingelsson E. Risk associated with the metabolic syn-drome versus the sum of its individual components. Diabetes Care. Jul 2006;29(7):1673-1674.

106. Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classi-fication of diabetes mellitus provisional report of a WHO consulta-tion. Diabet Med. Jul 1998;15(7):539-553.

107. Balkau B, Charles MA. Comment on the provisional report from the WHO consultation. European Group for the Study of Insulin Resis-tance (EGIR). Diabet Med. May 1999;16(5):442-443.

108. Anonymous. The IDF consensus worldwide definition of the meta-bolic syndrome. International Diabetes Federation. http://www.idf.org/webdata/docs/metac_syndrome_def.pdf. Ac-cessed 19 December, 2006.

71

109. Executive Summary of The Third Report of The National Choles-terol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). Jama. May 16 2001;285(19):2486-2497.

110. Ford ES, Giles WH. A comparison of the prevalence of the meta-bolic syndrome using two proposed definitions. Diabetes Care. Mar 2003;26(3):575-581.

111. Isomaa B, Almgren P, Tuomi T, Forsen B, Lahti K, Nissen M, Taskinen MR, Groop L. Cardiovascular morbidity and mortality as-sociated with the metabolic syndrome. Diabetes Care. Apr 2001;24(4):683-689.

112. Lakka HM, Laaksonen DE, Lakka TA, Niskanen LK, Kumpusalo E, Tuomilehto J, Salonen JT. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. Jama. Dec 4 2002;288(21):2709-2716.

113. Sundstrom J, Riserus U, Byberg L, Zethelius B, Lithell H, Lind L. Clinical value of the metabolic syndrome for long term prediction of total and cardiovascular mortality: prospective, population based co-hort study. Bmj. Apr 15 2006;332(7546):878-882.

114. Wilson PW, D'Agostino RB, Parise H, Sullivan L, Meigs JB. Meta-bolic syndrome as a precursor of cardiovascular disease and type 2 diabetes mellitus. Circulation. Nov 15 2005;112(20):3066-3072.

115. Klein BE, Klein R, Lee KE. Components of the metabolic syndrome and risk of cardiovascular disease and diabetes in beaver dam. Dia-betes Care. Oct 2002;25(10):1790-1794.

116. Bjorntorp P. Visceral obesity: a "civilization syndrome". Obes Res. May 1993;1(3):206-222.

117. Moller DE, Kaufman KD. Metabolic syndrome: a clinical and mo-lecular perspective. Annu Rev Med. 2005;56:45-62.

118. Rosell M, De Faire U, Hellenius ML. Low prevalence of the meta-bolic syndrome in wine drinkers--is it the alcohol beverage or the lifestyle? Eur J Clin Nutr. Feb 2003;57(2):227-234.

119. Hulthe J, Bokemark L, Wikstrand J, Fagerberg B. The metabolic syndrome, LDL particle size, and atherosclerosis: the Atherosclero-sis and Insulin Resistance (AIR) study. Arterioscler Thromb Vasc Biol. Sep 2000;20(9):2140-2147.

120. Phillips C, Lopez-Miranda J, Perez-Jimenez F, McManus R, Roche HM. Genetic and nutrient determinants of the metabolic syndrome. Curr Opin Cardiol. May 2006;21(3):185-193.

121. Loktionov A. Common gene polymorphisms and nutrition: emerg-ing links with pathogenesis of multifactorial chronic diseases (re-view). J Nutr Biochem. Aug 2003;14(8):426-451.

122. Roche H, RP M. Ch 2. Molecular Aspects of Nutrition. In: Gibney M, Macdonald I, Roche H, eds. Nutrition and Metabolism: Black-well Publishing; 2003:6-29.

72

123. Anonymous. Genome variation and SNP information on the internet [Internet; Various webpages, http://www.genomenewsnetwirk.org, http://www.ornl.gov/sci/techresources/Human_Genome/]. Accessed April 3rd, 2006.

124. Cordell HJ, Clayton DG. Genetic association studies. Lancet. Sep 24-30 2005;366(9491):1121-1131.

125. Hattersley AT, McCarthy MI. What makes a good genetic associa-tion study? Lancet. Oct 8 2005;366(9493):1315-1323.

126. Information regarding the ULSAM cohort. Internet website: http://www.pubcare.uu.se/ULSAM/ (accessed January 2007).

127. Hedstrand H. A study of middle aged men with particular reference to risk factors for cardiovascular risk. Ups J Med Sci suppl. 1975;19:1-61.

128. Skarfors ET, Selinus KI, Lithell HO. Risk factors for developing non-insulin dependent diabetes: a 10 year follow up of men in Upp-sala. Bmj. Sep 28 1991;303(6805):755-760.

129. Rush R, Leon L, Turell J. Automated simultanous cholesterol and triglyceride determination on the Auto_Analyzer II instrument. Ad-vances in Automated Analysis. 1971;1:503-507.

130. Siegler L, Wu W. Separation of serum high-density lipoprotein for cholesterol determination: ultracentrifugation vs. precipitation with sodium phosphotungstate and magnesium chloride. Clinical Chemis-try. 1981(27):834-841.

131. Wide L, Axén R, Porath J. Radioimmunosobent assay of proteins. Chemical coupling of antibodies in insoluble dextran. Immunochem-istry. 1967;4:381-386.

132. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concen-trations in man. Diabetologia. Jul 1985;28(7):412-419.

133. Boberg M, Croon LB, Gustafsson IB, Vessby B. Platelet fatty acid composition in relation to fatty acid composition in plasma and to serum lipoprotein lipids in healthy subjects with special reference to the linoleic acid pathway. Clin Sci (Lond). May 1985;68(5):581-587.

134. Öhrvall M, Berglund L, Salminen I, Lithell H, Aro A, Vessby B. The serum cholesterol ester fatty acid composition but not the serum concentration of alpha tocopherol predicts the development of myo-cardial infarction in 50-year-old men: 19 years follow-up. Athero-sclerosis. Nov 15 1996;127(1):65-71.

135. Collen MF, Cutler JL, Siegelaub AB, Cella RL. Reliability of a self-administered medical questionnaire. Arch Intern Med. Jun 1969;123(6):664-681.

136. Byberg L, Siegbahn A, Berglund L, McKeigue P, Reneland R, Lithell H. Plasminogen activator inhibitor-1 activity is independ-ently related to both insulin sensitivity and serum triglycerides in 70-year-old men. Arterioscler Thromb Vasc Biol. Feb 1998;18(2):258-264.

73

137. DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol. Sep 1979;237(3):E214-223.

138. Pollare T, Vessby B, Lithell H. Lipoprotein lipase activity in skeletal muscle is related to insulin sensitivity. Arterioscler Thromb. Sep-Oct 1991;11(5):1192-1203.

139. 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. Dec 17 2002;106(25):3143-3421.

140. De Backer G, Ambrosioni E, Borch-Johnsen K, Brotons C, Cifkova R, Dallongeville J, Ebrahim S, Faergeman O, Graham I, Mancia G, Manger Cats V, Orth-Gomer K, Perk J, Pyorala K, Rodicio JL, Sans S, Sansoy V, Sechtem U, Silber S, Thomsen T, Wood D. European guidelines on cardiovascular disease prevention in clinical practice. Third Joint Task Force of European and Other Societies on Cardio-vascular Disease Prevention in Clinical Practice. Eur Heart J. Sep 2003;24(17):1601-1610.

141. Information on selected SNPs were obtained from the dbSNP data-base, http://www.ncbi.nlm.nih.gov/SNP/.

142. More information regarding the SNP technology platform at Upp-sala University, Sweden can be found at the following web-site: http://www.genotyping.se/.

143. Bell PA, Chaturvedi S, Gelfand CA, Huang CY, Kochersperger M, Kopla R, Modica F, Pohl M, Varde S, Zhao R, Zhao X, Boyce-Jacino MT, Yassen A. SNPstream UHT: ultra-high throughput SNP genotyping for pharmacogenomics and drug discovery. Biotech-niques. Jun 2002;Suppl:70-72, 74, 76-77.

144. Nyholt DR. A simple correction for multiple testing for single-nucleotide polymorphisms in linkage disequilibrium with each other. Am J Hum Genet. Apr 2004;74(4):765-769.

145. Stephens M, Smith NJ, Donnelly P. A new statistical method for haplotype reconstruction from population data. Am J Hum Genet. Apr 2001;68(4):978-989.

146. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visu-alization of LD and haplotype maps. Bioinformatics. Jan 15 2005;21(2):263-265.

147. Jones B, Kenwood M. Design and analysis of cross-over trials.: Champman and Hall; 1989.

148. Tuomilehto J, Lindstrom J, Eriksson JG, Valle TT, Hamalainen H, Ilanne-Parikka P, Keinanen-Kiukaanniemi S, Laakso M, Louheranta A, Rastas M, Salminen V, Uusitupa M. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glu-cose tolerance. N Engl J Med. May 3 2001;344(18):1343-1350.

149. Laaksonen DE, Lakka HM, Salonen JT, Niskanen LK, Rauramaa R, Lakka TA. Low levels of leisure-time physical activity and cardio-

74

respiratory fitness predict development of the metabolic syndrome. Diabetes Care. Sep 2002;25(9):1612-1618.

150. Pan DA, Lillioja S, Milner MR, Kriketos AD, Baur LA, Bogardus C, Storlien LH. Skeletal muscle membrane lipid composition is related to adiposity and insulin action. J Clin Invest. Dec 1995;96(6):2802-2808.

151. Warensjö E, Riserus U, Vessby B. Fatty acid composition of serum lipids predicts the development of the metabolic syndrome in men. Diabetologia. Oct 2005;48(10):1999-2005.

152. Nestel P, Clifton P, Noakes M. Effects of increasing dietary palmitoleic acid compared with palmitic and oleic acids on plasma lipids of hypercholesterolemic men. J Lipid Res. Apr 1994;35(4):656-662.

153. Sacks FM, Katan M. Randomized clinical trials on the effects of dietary fat and carbohydrate on plasma lipoproteins and cardiovas-cular disease. Am J Med. Dec 30 2002;113 Suppl 9B:13S-24S.

154. Listenberger LL, Ory DS, Schaffer JE. Palmitate-induced apoptosis can occur through a ceramide-independent pathway. J Biol Chem. May 4 2001;276(18):14890-14895.

155. Wei Y, Wang D, Pagliassotti MJ. Saturated fatty acid-mediated en-doplasmic reticulum stress and apoptosis are augmented by trans-10, cis-12-conjugated linoleic acid in liver cells. Mol Cell Biochem. Apr 11 2007.

156. Gaster M, Rustan AC, Beck-Nielsen H. Differential utilization of saturated palmitate and unsaturated oleate: evidence from cultured myotubes. Diabetes. Mar 2005;54(3):648-656.

157. Lee JS, Pinnamaneni SK, Eo SJ, Cho IH, Pyo JH, Kim CK, Sinclair AJ, Febbraio MA, Watt MJ. Saturated, but not n-6 polyunsaturated, fatty acids induce insulin resistance: role of intramuscular accumula-tion of lipid metabolites. J Appl Physiol. May 2006;100(5):1467-1474.

158. Lovejoy JC, Champagne CM, Smith SR, DeLany JP, Bray GA, Le-fevre M, Denkins YM, Rood JC. Relationship of dietary fat and se-rum cholesterol ester and phospholipid fatty acids to markers of in-sulin resistance in men and women with a range of glucose toler-ance. Metabolism. Jan 2001;50(1):86-92.

159. Salomaa V, Ahola I, Tuomilehto J, Aro A, Pietinen P, Korhonen HJ, Penttila I. Fatty acid composition of serum cholesterol esters in dif-ferent degrees of glucose intolerance: a population-based study. Me-tabolism. Dec 1990;39(12):1285-1291.

160. Lelliott C, Vidal-Puig AJ. Lipotoxicity, an imbalance between lipo-genesis de novo and fatty acid oxidation. Int J Obes Relat Metab Disord. Dec 2004;28 Suppl 4:S22-28.

161. Sampath H, Miyazaki M, Ntambi JM. Stearoyl CoA desaturase-1 mediates the prolipogenic effects of dietary saturated fat. Abstracts/ Chemistry and Physics of Lipids. 2006;143:63.

75

162. Miyazaki M, Ntambi JM. Role of stearoyl-coenzyme A desaturase in lipid metabolism. Prostaglandins Leukot Essent Fatty Acids. Feb 2003;68(2):113-121.

163. Miettinen TA, Naukkarinen V, Huttunen JK, Mattila S, Kumlin T. Fatty-acid composition of serum lipids predicts myocardial infarc-tion. Br Med J (Clin Res Ed). Oct 9 1982;285(6347):993-996.

164. Swahn E. A view from Sweden, January 31, 2006. Circulation. Jan 31 2006;113(4):f13-14.

165. Hodge AM, English DR, O'Dea K, Sinclair AJ, Makrides M, Gibson RA, Giles GG. Plasma phospholipid and dietary fatty acids as pre-dictors of type 2 diabetes: interpreting the role of linoleic acid. Am J Clin Nutr. Jul 2007;86(1):189-197.

166. Vessby B, Aro A, Skarfors E, Berglund L, Salminen I, Lithell H. The risk to develop NIDDM is related to the fatty acid composition of the serum cholesterol esters. Diabetes. Nov 1994;43(11):1353-1357.

167. Miyazaki M, Kim YC, Ntambi JM. A lipogenic diet in mice with a disruption of the stearoyl-CoA desaturase 1 gene reveals a stringent requirement of endogenous monounsaturated fatty acids for triglyc-eride synthesis. J Lipid Res. Jul 2001;42(7):1018-1024.

168. Stumvoll M. Control of glycaemia: from molecules to men. Min-kowski Lecture 2003. Diabetologia. May 2004;47(5):770-781.

169. Bouchard C, Tremblay A, Despres JP, Nadeau A, Lupien PJ, Theri-ault G, Dussault J, Moorjani S, Pinault S, Fournier G. The response to long-term overfeeding in identical twins. N Engl J Med. May 24 1990;322(21):1477-1482.

170. Dobrzyn A, Ntambi JM. Stearoyl-CoA desaturase as a new drug target for obesity treatment. Obes Rev. May 2005;6(2):169-174.

171. Palmer LJ, Cardon LR. Shaking the tree: mapping complex disease genes with linkage disequilibrium. Lancet. Oct 1 2005;366(9492):1223-1234.

172. Das UN. A defect in the activity of Delta6 and Delta5 desaturases may be a factor in the initiation and progression of atherosclerosis. Prostaglandins Leukot Essent Fatty Acids. May 2007;76(5):251-268.

173. Manco M, Calvani M, Mingrone G. Effects of dietary fatty acids on insulin sensitivity and secretion. Diabetes Obes Metab. Nov 2004;6(6):402-413.

174. Storlien LH, Baur LA, Kriketos AD, Pan DA, Cooney GJ, Jenkins AB, Calvert GD, Campbell LV. Dietary fats and insulin action. Dia-betologia. Jun 1996;39(6):621-631.

175. Pinnamaneni SK, Southgate RJ, Febbraio MA, Watt MJ. Stearoyl CoA desaturase 1 is elevated in obesity but protects against fatty acid-induced skeletal muscle insulin resistance in vitro. Diabetolo-gia. Dec 2006;49(12):3027-3037.

176. Voss MD, Beha A, Tennagels N, Tschank G, Herling AW, Quint M, Gerl M, Metz-Weidmann C, Haun G, Korn M. Gene expression pro-

76

filing in skeletal muscle of Zucker diabetic fatty rats: implications for a role of stearoyl-CoA desaturase 1 in insulin resistance. Diabe-tologia. Dec 2005;48(12):2622-2630.

177. Gutierrez-Juarez R, Pocai A, Mulas C, Ono H, Bhanot S, Monia BP, Rossetti L. Critical role of stearoyl-CoA desaturase-1 (SCD1) in the onset of diet-induced hepatic insulin resistance. J Clin Invest. Jun 2006;116(6):1686-1695.

178. Frayn KN. Adipose tissue as a buffer for daily lipid flux. Diabetolo-gia. Sep 2002;45(9):1201-1210.

179. Sjögren P, Sierra-Johnson J, Gertow K, Rosell M, Vessby B, deFaire U, Hamsten A, Hellenius M, Fisher RM. Fatty acid desaturases in human adipose tissue: relationships between gene expression, desau-ration indexes and insulin resistance. Diabetologia. 2007;Accepted for publication.

180. Ma J, Folsom AR, Eckfeldt JH, Lewis L, Chambless LE. Short- and long-term repeatability of fatty acid composition of human plasma phospholipids and cholesterol esters. The Atherosclerosis Risk in Communities (ARIC) Study Investigators. Am J Clin Nutr. Sep 1995;62(3):572-578.

181. Hodge AM, Simpson JA, Gibson RA, Sinclair AJ, Makrides M, O'Dea K, English DR, Giles GG. Plasma phospholipid fatty acid composition as a biomarker of habitual dietary fat intake in an ethni-cally diverse cohort. Nutr Metab Cardiovasc Dis. Jul 2007;17(6):415-426.

182. Liu G, Lynch JK, Freeman J, Liu B, Xin Z, Zhao H, Serby MD, Kym PR, Suhar TS, Smith HT, Cao N, Yang R, Janis RS, Krauser JA, Cepa SP, Beno DW, Sham HL, Collins CA, Surowy TK, Camp HS. Discovery of Potent, Selective, Orally Bioavailable Stearoyl-CoA Desaturase 1 Inhibitors. Journal of medicinal chemistry. Jun 28 2007;50(13):3086-3100.

183. Schaeffer L, Gohlke H, Muller M, Heid IM, Palmer LJ, Kompauer I, Demmelmair H, Illig T, Koletzko B, Heinrich J. Common genetic variants of the FADS1 FADS2 gene cluster and their reconstructed haplotypes are associated with the fatty acid composition in phos-pholipids. Human molecular genetics. Jun 1 2006;15(11):1745-1756.

184. Baylin A, Kim MK, Donovan-Palmer A, Siles X, Dougherty L, Tocco P, Campos H. Fasting whole blood as a biomarker of essential fatty acid intake in epidemiologic studies: comparison with adipose tissue and plasma. Am J Epidemiol. Aug 15 2005;162(4):373-381.

185. Bonita R, Beaglehole R, Kjellström T. Basic Epidemiology. 2nd ed: World Health Organization 2006.

186. Tunstall-Pedoe H, Kuulasmaa K, Amouyel P, Arveiler D, Rajakan-gas AM, Pajak A. Myocardial infarction and coronary deaths in the World Health Organization MONICA Project. Registration proce-dures, event rates, and case-fatality rates in 38 populations from 21 countries in four continents. Circulation. Jul 1994;90(1):583-612.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 292

Editor: The Dean of the Faculty of Medicine

A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-8312

ACTA

UNIVERSITATIS

UPSALIENSIS

UPPSALA

2007

fatty acid desaturase activities in metabolic syndrome and

Documents