VASCULOPROTECTIVE EFFECTS OF INSULIN AND RESVERATROL IN VIVO

Transcription

1 VASCULOPROTECTIVE EFFECTS OF INSULIN AND RESVERATROL IN VIVO By Danna Marie Breen A thesis submitted in conformity with the requirements For the degree of Doctor of Philosophy Graduate Department of Physiology University of Toronto Copyright by Danna Marie Breen (2010)

2 Title: Vasculoprotective effects of insulin and resveratrol in vivo. Degree: Ph.D. Year of Convocation: 2010 Full Name: Danna Marie Breen Department: Physiology, University of Toronto General Abstract Atherosclerosis is a leading cause of morbidity and mortality worldwide and type 2 diabetes and obesity-associated metabolic syndrome, both characterized by insulin resistance, are potent risk factors. These conditions also increase the risk for restenosis after revascularization procedures used for treatment of atherosclerosis. Studies have shown that insulin and resveratrol (RSV), a red wine polyphenol, decrease neointimal growth after vessel injury in models of restenosis, demonstrating a protective effect on the vasculature. However, oral glucose and sucrose were used in insulin studies to maintain normoglycemia, and their effect on neointimal formation was not assessed. Several studies have shown that nitric oxide (NO) production is stimulated by insulin and RSV, and since NO can decrease neointimal growth, the objective of this thesis was to address the mechanism of action of insulin or RSV to protect against restenosis, and determine whether NO production mediates these effects. To examine this, we treated rats with insulin or RSV and performed arterial balloon injury. In Study 1, insulin reduced neointimal area after injury in rats receiving oral glucose but not oral sucrose. Oral glucose alone had no effect on neointimal formation or insulin sensitivity whereas oral sucrose increased neointimal growth and induced insulin resistance. In Study 2, insulin decreased neointimal area and cell migration, and increased re-endothelialization. These effects were abolished by nitric oxide synthase (NOS) inhibition. In addition, insulin increased enos protein expression in the vessel. In Study 3, RSV reduced neointimal growth, cell proliferation, and migration after injury, without affecting re-endothelialization. Most of these effects were abolished by NOS inhibition, except for the decrease in cell migration. Insulin sensitivity and systolic blood pressure were not affected by RSV. ii

3 Together, the results demonstrate that insulin, independent of glycemic effects, and RSV have a protective effect on the vessel against restenosis, which is mediated by NO. Since both insulin and RSV decrease neointimal formation without negatively impacting reendothelialization, insulin or RSV treatment could provide some advantage over anti-mitogenic agents currently used in drug-eluting stents, which delay re-endothelialization. These studies suggest that insulin or RSV may have clinical potential in the prevention of restenosis after angioplasty. iii

4 Acknowledgment This thesis is the result of five years of work whereby I have been supported by numerous people and I am pleased to now have the opportunity to express my appreciation for all of them. First and foremost I offer my sincerest gratitude to my supervisor, Dr Adria Giacca, who has supported me throughout my doctoral studies with her motivation and knowledge, allowing me the room to work in my own way. I simply could not wish for a friendlier or more generous supervisor, and your dedication to research and teaching is an inspiration. I would like to thank my co-supervisor Dr Evangelia Tsiani, and committee members Dr Michelle Bendeck, and Dr George Fantus who monitored my work and took tremendous effort in reading and providing me with valuable comments on several versions of this thesis. I thank you all. In my daily work I have been blessed with a friendly and entertaining group of colleagues and friends; Loretta Lam, Dr Edward Park, Dr Andrei Oprescu, Sandra Pereira, Khajag Koulajian, Christina Dirlea, and especially Christine Tang and June Guo. I will fondly remember our fantastic and somewhat extravagant conference trips! My family was always a pillar of support and I cannot begin to express how important they are to me. I am extremely grateful for my husband Mario, for his love and patience during this time, which enabled me to complete this work. My gratitude would definitely be incomplete if I would forget to thank God for my life and His blessings through all tests during the past five years. My deepest and sincere gratitude for inspiring and guiding me. iv

5 Table of Contents 1.0 Introduction Atherosclerosis Mechanism of Atherosclerosis Development Restenosis Animal Models of Restenosis Mechanism of the Arterial Response to Balloon Catheter Injury Platelet Activation Leukocyte Infiltration Vascular Smooth Muscle Cell Phenotype Switch Vascular Smooth Muscle Cell Proliferation and Apoptosis Vascular Smooth Muscle Cell Migration Matrix Metalloproteinases (MMPs) Extracellular Matrix Endothelial Regeneration Progenitor Cell Contribution Type 2 Diabetes and Metabolic Syndrome Nitric Oxide Insulin General effects of insulin Effects of insulin on atherosclerosis Effects of insulin on restenosis Mechanism of Insulin Action in the Vasculature General features of insulin signal transduction pathways Insulin stimulates NO production and induces vasodilation Insulin stimulates ET- 1 and induces vasoconstriction Insulin inhibits platelet function Insulin has anti- inflammatory effects Effect(s) of insulin on proliferation and migration Resveratrol General effects of RSV Effects of resveratrol on atherosclerosis Effects of resveratrol on restenosis Mechanism of Resveratrol Action in the Vasculature General resveratrol signal transduction Resveratrol stimulates NO production and induces vasodilation Resveratrol inhibits platelet function and thrombus formation Resveratrol has anti- oxidant effects Resveratrol has anti- inflammatory effects Resveratrol inhibits proliferation and migration Summary Rationale and Significance of the Studies General Hypothesis Specific Aims General Methods Animals Surgical Procedures Subcutaneous Implant Procedure v

6 2.2.2 Intraperitoneal Catheterization for Glucose Infusion Osmotic Pump Implant Procedure Carotid Balloon Injury Aortic Balloon Injury Vessel Cannulation Blood and Vessel Sample Collection Fixed Vessel Collection Frozen Vessel Collection Morphometric Measurements and Calculations Histomorphometry/Matrix Staining Cell Migration Assay BrdU- Labelling Terminal dutp nick- end labeling (TUNEL) Assay Extracellular Matrix Staining with Movat s Pentachrome and Picrosirius Red Evan s Blue Staining Vessel Homogenization Protein Assay Western Blot Analysis Flow Cytometry Hyperinsulinemic- Euglycemic Clamp Blood Pressure Analysis Metabolic Parameters Plasma Insulin Plasma Free Fatty Acid Plasma Triglyceride Plasma Glucose Determination Statistical Analysis Study Abstract Introduction Materials and Methods Animal Models Surgical Procedures Vessel and Blood Sample Collection Metabolic Parameters Histomorphometry/Matrix Staining Vessel Cannulation Hyperinsulinemic- Euglycemic Clamp Statistical Analysis Results Metabolic Parameters Morphometric Measurements Insulin Sensitivity Measurements Discussion Effect of insulin and oral glucose Effect of insulin and oral sucrose Study Abstract...90 vi

7 4.2 Introduction Materials and Methods Animal Models Surgical Procedures Vessel and Blood Sample Collection Metabolic Parameters Histomorphometry/Matrix Staining Evan s Blue Staining Western Blot Analysis Flow Cytometry Blood Pressure Analysis Statistical Analysis Results Metabolic Parameters Morphometric Measurements Discussion Study Abstract Introduction Materials and Methods Animal Models Vessel and Blood Sample Collection Metabolic Parameters Histomorphometry/Matrix Staining Evan s Blue Staining Vessel Cannulation Hyperinsulinemic- Euglycemic Clamp Blood Pressure Analysis Statistical Analysis Results Metabolic Parameters Morphometric Measurements Insulin Sensitivity Measurements Discussion Summary and Conclusions Summary of Each Study in the Thesis General Summary General Conclusion General Discussion Is the Effect of Insulin Protective or Detrimental in a Model of Restenosis? The Protective Effects of Insulin or Resveratrol Against Mechanically- Induced Restenosis: How do They Differ? The Role of NO in Insulin and RSV Action: What is the Source of NO? Limitations of the Studies Future Directions References 152 vii

8 List of Abbreviations AhR aryl hydrocarbon receptor AICAR 5-aminoimidazole- 4- carboxamide- 1- β- D- ribofuranoside AMPK 5 adenosine monophosphate- activated protein kinase ANOVA analysis of variance ApoE apoplipoprotein E AP-1 activating protein- 1 BH4 tetrahydrobiopterin BSA bovine serum albumin CamKinaseII calcium/ calmodulin- dependent protein kinase II CDK cyclin dependent kinase cgmp cyclic guanylate monophosphate COX cyclooxygenase CVD cardiovascular disease DAB 3,3 - diabminobenzidine DCCT Diabetes Control and Complications Trial ECM extracellular matrix EGF epidermal growth factor enos endothelial nitric oxide synthase EPC endothelial progenitor cell ERK 1/ 2 extracellular regulated kinase ET-1 endothelin- 1 FAD flavin adenine dinucleotide FFA free fatty acid FGF fibroblast growth factor FMN flavin mononucleotide FOXO1 forkhead transcription factor- 1 Grb2 growth factor receptor- binding protein 2 HDL high- density lipoprotein HSC hematopoietic stem cell HSP90 heat shock protein 90 ICAM-1 intracellular adhesion molecule- 1 IGF-1 insulin- like growth factor- 1 IL-1β interleukin- 1 beta IL-6 interleukin- 6 inos inducible nitric oxide synthase IRS insulin receptor substrate JNK c- jun NH2- terminal kinase LCCA left common carotid artery LDL low- density lipoprotein LDLR low- density lipoprotein receptor L-NAME L- N G - Nitroarginine methyl ester MAPK mitogen- activated protein kinase MCP-1 monocyte chemoattractant protein- 1 MMP matrix metalloproteinase MSC mesenchymal stem cell MT-MMP membrane type- matrix metalloproteinase mtor mammalian target of rapamycin viii

9 NADPH nicotinamide adenine dinucleotide phosphate hydrogen NFκB nuclear factor kappa B NMDA N- methyl D- aspartate nnos neuronal nitric oxide synthase NO nitric oxide PAGE polyacrylamide gel electrophoresis PAI-1 plasminogen activator inhibitor- 1 PBS phosphate buffered saline PDGF platelet- derived growth factor PDK-1 phosphoinositide- dependent protein kinase- 1 PE polyethylene PIP2 phosphatidylinositol 4, 5- bisphosphate PIP3 phosphatidylinositol- 3, 4, 5- triphosphate PI3K phosphatidylinositol- 3 kinase PKC protein kinase C PKG protein kinase G PSR picrosirius red PTCA percutaneous transluminal coronary angioplasty PTEN phosphatase and tensin homolog PTP1B protein tyrosine phosphatase 1B PVDF polyvinylidene difluoride RIA radioimmunoassay ROS reactive oxygen species RSV resveratrol SDS sodium dodecyl sulfate SHIP-2 SH2- domain containing inositol phosphatase 2 SIRT sirtuin SMC smooth muscle cell SM-MHC smooth muscle- myosin heavy chain SM22α smooth muscle 22 alpha Sos son of sevenless STZ streptozotocin S6K1 p70 S6 kinase- 1 TBST tris- buffered saline tween TG triglyceride TIMP tissue inhibitor of metalloproteinase TNFα tumour necrosis factor alpha TUNEL terminal dutp nick- end labeling UKPDS United Kingdom Prospective Diabetes Study UV ultraviolet VCAM-1 vascular cell adhesion molecule- 1 VEGF vascular endothelial growth factor VLDL very low- density lipoprotein VSMC vascular smooth muscle cell ZDF Zucker diabetic fatty ix

10 List of Tables Table 3.1. Neointimal formation with or without an i.p. catheter Table 3.2. Daily fasting and fed plasma glucose levels over the treatment period...80 Table 3.3. Fasting plasma levels of insulin, triglycerides, and free fatty acids at 14 days after arterial injury...81 Table 3.4. Fed plasma levels of insulin, triglycerides, and free fatty acids at 14 days after arterial injury Table 3.5. Intimal cell number and cell density per unit area measured at 14 days after vessel injury.83 Table 4.1. Daily fasting and fed plasma glucose levels over the treatment period.101 Table 4.2. Fed plasma levels of insulin, triglycerides, and free fatty acids at 28 days after arterial injury.102 Table 4.3. Percentage of white blood cells stained positive for c-kit, sca-1, and VEGFR Table 5.1. Daily food, fluid and total caloric intake, final weight gain, fed plasma glucose and insulin levels over the treatment periods x

11 List of Figures Figure 1.1. Signal transduction by the insulin receptor Figure 1.2. Proposed signal transduction by resveratrol Figure 3.1. The effect of insulin, oral glucose, and oral sucrose on neointimal formation Figure 3.2. The effect of insulin, oral glucose, and oral sucrose on intimal area Figure 3.3. The effect of insulin, oral glucose, and oral sucrose on medial area and external elastic lamina perimeter Figure 3.4. The effect of insulin, oral glucose, and oral sucrose on cell proliferation Figure 3.5. The effect of oral glucose and oral sucrose on whole body insulin sensitivity Figure 4.1. The effect of insulin and NOS inhibition on intimal area Figure 4.2. The effect of insulin and NOS inhibition on elastin and collagen accumulation Figure 4.3. The effect of insulin and NOS inhibition on cell migration Figure 4.4. The effect of insulin on SMC differentiation markers and enos expression Figure 4.5. The effect of insulin and NOS inhibition on re-endothelialization Figure 4.6. The effect of insulin and NOS inhibition on systolic blood pressure Figure 5.1. The effect of RSV and NOS inhibition on intimal area Figure 5.2. The effect of RSV and NOS inhibition on cell proliferation Figure 5.3. The effect of RSV on cell apoptosis Figure 5.4. The effect of RSV and NOS inhibition on elastin and collagen content Figure 5.5. The effect of RSV and NOS inhibition on cell migration Figure 5.6. The effect of RSV on re-endothelialization Figure 5.7. The effect of RSV and NOS inhibition on systolic blood pressure Figure 5.8. The effect of RSV on whole body insulin sensitivity Figure 7.1. Summary of the effects of insulin and RSV in the vessel after arterial balloon injury xi

12 Published Manuscripts Arising from Completion of this Thesis Introduction Chapter) Breen DM, Giacca A. Effects of Insulin on the Vasculature. (Current Vascular Pharmacology, In Revision #183846). Permission to reproduce portions of the above manuscript has been obtained from the copyright owner, Bentham Science Publishers. Study 1) Breen DM, Dhaliwall JK, Chan KK, Guo J, Lam L, Bendeck MP, Giacca A. Insulin inhibits and oral sucrose increases neointimal growth after arterial injury in rats. Journal of Vascular Research. 2010; 47: Permission to reproduce portions of the above manuscript has been obtained from the copyright owner, S. Karger AG, Basel. Study 2) Breen DM*, Chan KK*, Dhaliwall JK, Ward MR, Koudsi NA, Lam L, De Souza M, Ghanim H, Dandona P, Stewart D, Bendeck MP, Giacca A. Insulin Increases Reendothelialization and Inhibits Cell Migration and Neointimal Growth After Arterial Injury. Arteriosclerosis, Thrombosis & Vascular Biology. 2009; 29(7): *Equal contribution. Permission to reproduce portions of the above manuscript has been obtained from the copyright owner, American Heart Association. Note: For Study 2, Table 4.3 and Figure 4.4A-F have been published in the manuscript listed above. The remaining data presented in Study 2 are currently unpublished. Study 3) Breen DM, Guo J, Mroziewicz M, Tsiani E, Bendeck MP, Giacca A. Resveratrol Inhibits Neointimal Formation After Arterial Injury Through A Nitric Oxide Synthase- Dependent Mechanism. (American Journal of Pathology, In Revision #AJP ). Permission to reproduce portions of the above manuscript is not yet required from the copyright owner, the American Journal of Pathology. xii

13 1 Introduction 1.1 Atherosclerosis Atherosclerotic cardiovascular disease (CVD) (coronary heart disease, myocardial infarction, stroke, peripheral vascular disease) is the leading cause of morbidity and mortality in the developed world, accounting for approximately 35.3% of all deaths in The Framingham Heart Study, which analyzed the 10 year incidence of atherosclerotic CVD and death in men, found a decline in both morbidity and mortality due to cardiovascular diseases 2. This may be explained by improved primary prevention and medical interventions 2. However, due to the recent epidemic of obesity and obesity- associated diabetes, diabetes- associated cardiovascular mortality is increasing and is therefore a major and growing public health concern 3. Atherosclerosis develops over a lifetime, and often symptoms or complications are not present. However, when clinical complications occur, they are highly variable and range from chronic stable symptoms, such as coronary heart disease, angina or arteriosclerosis, to fatal acute events, such as myocardial infarction or stroke. Non- modifiable risk factors for atherosclerosis include age, gender, and family history of CVD. Modifiable risk factors include smoking, alcohol consumption, hypertension, hyperlipidemia, diabetes, insulin resistance, and obesity. Methods for prevention have included identification of risk factors for an atherosclerotic event, lifestyle modifications, and pharmacologic intervention to lower plasma cholesterol concentrations. 1

14 1.2 Mechanism of Atherosclerosis Development Atherosclerosis represents a chronic inflammatory response of the arterial wall 4. The atherosclerotic lesion is divided into three stages: 1) the fatty streak; 2) the intermediate lesion; and 3) the fibrous plaque. The fatty streak is formed by the accumulation of macrophages and T- lymphocytes within the arterial wall, which may begin in childhood. The intermediate lesion is characterized by the accumulation of lipids, vascular smooth muscle cells (VSMC), and extracellular matrix (ECM). The lesion progressively increases in size and as a result the lumen of the vessel narrows. Over time an advanced fibrous plaque develops, which is characterized by a lipid- rich core containing necrotic tissue covered by a fibrous cap. An unstable plaque eventually ruptures and leads to thrombosis, which is partial or complete occlusion of the blood vessel. The pathogenesis of atherosclerosis is described by a more recent version of the response to injury hypothesis that emphasizes endothelial dysfunction and is widely accepted 4. Endothelial dysfunction is defined as the loss of endothelial integrity and/ or function. The normal healthy endothelium inhibits vasoconstriction, provides an anti- thrombotic surface to prevent platelet and leukocyte adhesion, and inhibits VSMC growth. Endothelial dysfunction is caused by mechanical or chemical injury; for example by oxidized low- density lipoproteins (LDL), chemical agents, toxins (including byproducts of cigarette smoking), hyperglycemia, and hyperhomocystinemia. The result is compensatory responses that change the normal homeostatic properties of the endothelium transforming it to a pro- thrombotic state, including impaired endothelial- dependent vasodilation. The earliest stage of atherogenesis is characterized by proteoglycan- mediated lipoprotein retention in the subendothelium; where proteoglycans bind lipoproteins through ionic interaction and lipoproteins permeate the intact endothelium and become oxidized. This process initiates an inflammatory response where membrane adhesion molecule expression increases on endothelial cells, thereby promoting leukocyte and platelet adhesion and increased permeability to lipoproteins (reviewed in 4 ). Leukocytes migrate into the intima and become macrophages that engulf modified 2

15 forms of lipoproteins, especially LDL, thus becoming foam cells that form the fatty streak. The endothelium, macrophages, and platelets release cytokines and growth factors, which stimulate migration and proliferation of VSMCs that mix with the area of inflammation forming an intermediate lesion. Progression of inflammation results in focal necrosis. VSMC proliferation and matrix deposition results in fibrosis. At this point the lesion is composed of a lipid and necrotic tissue core covered by a fibrous cap and the lesion protrudes into the lumen and alters blood flow (reviewed in 4 ). Collagen provides strength and stability to the fibrous cap, however thin, macrophage- rich caps are unstable and can promote plaque rupture. The majority of sudden deaths from myocardial infarcts are attributed to ruptures or fissures resulting in hemorrhage into the plaque, thrombosis, and occlusion of the artery Restenosis In contrast to atherosclerosis, restenosis is initiated by acute mechanical injury, such as balloon angioplasty, and the development is rapid 5. Percutaneous transluminal coronary angioplasty (PTCA) is a well- established technique for myocardial revascularization of patients with coronary artery disease due to atherosclerosis. However, PTCA alone is limited by restenosis in 30-60% of cases 6. Therefore, stenting after PTCA was introduced using the balloon- expandable bare- metal stent as it provided distinct advantages over angioplasty alone, such as a drastically reduced rate of restenosis (20-30% in humans) 6. Subsequently, drug- eluting stents were designed to release pharmacological agents after insertion to inhibit VSMC proliferation and in turn restenosis. First- generation drug- eluting stents were coated with either rapamycin or paclitaxel, which reduced the restenosis rate to less than 10% as compared to the bare- metal stents 7. These agents primarily inhibit VSMC migration and proliferation, which represent crucial events in the development of in- stent restenosis 8. Stent thrombosis has emerged as a severe complication 9 ; despite this, PTCA with stenting is currently the preferred method of 3

16 treatment. Although drug- eluting stents have greatly decreased the rate of restenosis, results in diabetic individuals are still worse than results in non- diabetic people 10. Restenosis is the renarrowing of the blood vessel after balloon angioplasty and is a function of early elastic recoil, negative vessel remodeling, and neointimal hyperplasia 11. Elastic recoil is the rapid loss of lumen gain after balloon angioplasty due to the elastin fibers that comprise the internal and external elastic lamina. Vessel remodelling is a gradual process by which lumen size is structurally modified in response to changes in blood flow 11. For example, in initial stages of atherosclerosis the vessel enlarges (positive remodeling) to increase lumen size in response to the growing intima. However, severe atherosclerosis in vessels promotes negative or constrictive remodeling, which further decreases lumen size. Neointimal hyperplasia is an exaggerated healing response that occurs in the vessel wall after injury. In humans, the neointima after balloon injury stabilizes in three to six months 12. Elastic recoil and constrictive remodeling can be eliminated with a stent, therefore restenosis after stenting is mainly characterized by neointimal formation, which is a complex process resulting from VSMC proliferation and deposition of ECM Animal Models of Restenosis Animal models are important tools in understanding the vascular response to arterial injury following interventions, especially balloon angioplasty. The most commonly used animal models for the study of restenosis after balloon angioplasty are the primate, pig, rabbit, rat, and mouse. Primate models best resemble human restenosis and thrombotic activity since they are prone to acute stent thrombosis within the first three days following injury 14. In the porcine model both balloon angioplasty and stenting result in extensive neointimal formation. In addition, this model is useful to study the kinetics of restenosis as they develop human- like lesions 14. The rabbit model has been used to study dyslipidemia and restenosis as hypercholesterolemia can be readily induced using atherogenic diets, resulting in reproducible 4

17 atherosclerotic- like lesions 15. Transgenic mouse models are also used to study the molecular mechanisms involved in neointimal growth, however balloon catheters are too large to be used in mice, so arterial injury is induced with small guidewires 16. The kinetics of neointimal formation are similar to the rat model, however neointimal formation in mice is often difficult to reproduce due to the nature of the injury with the irregular shape of the guidewire and also the different mouse strains 16. Although the rat model has minimal inflammatory and thrombotic components, and does not allow the study of long- term vessel remodelling, it is the best characterized model of arterial injury and restenosis (reviewed in 17 ). In this thesis both aortic and left common carotid arterial angioplasty models were used, which involve inducing fibroproliferative lesions within a long segment of the artery using a 2 French Fogarty catheter. Balloon angioplasty induces intimal injury by denuding the endothelial layer, resulting in platelet activation and secretion of growth factors from platelets and damaged VSMCs. Inflammation can also occur due to monocyte recruitment to the site of injury. Subsequently, there is VSMC migration from media to intima, and VSMC proliferation and ECM deposition in the neointima, resulting in lumen narrowing. Bone marrow derived smooth muscle precursors can also contribute to neointimal growth 18, as can adventitial cells 19. Re- endothelialization occurs starting from endothelium at the margins of the denuded area and is contributed by bone marrow derived endothelial precursor cells 20. The time course of neointimal growth has been well established after experimental balloon angioplasty in the rat (reviewed in 17 ). Medial VSMC proliferation peaks two days after balloon angioplasty and is followed by a peak in VSMC migration to the intima at four days. VSMCs replicate in the neointima and reach maximum proliferation by seven days. Neointimal growth, mainly due to ECM deposition, continues until 28 days at which time lesion growth plateaus The rat arterial injury model is limited when compared with human restenosis because; 1) inflammation and thrombosis are implicated in human restenosis and are much less after vessel 5

18 injury in the rat 21, 2) in humans there is compensatory enlargement of the vessel to maintain constant lumen diameter, however this does not occur in rats 17, and 3) there are no pre- existing lesions in the rat model of balloon injury as in humans at the time of PTCA. As compounds that reduce restenosis in the rat model are not always effective in humans, particularly anti-oxidant and anti-inflammatory treatments, interpretation of results obtained in the rat model of balloon injury should be viewed with caution. 1.5 Mechanism of the Arterial Response to Balloon Catheter Injury Platelet Activation The role of platelets in promoting neointimal formation in the rat model of balloon injury has been demonstrated 22. Endothelial denudation of the vessel exposes the subendothelial components, mainly collagen and fibronectin, to the circulating blood, thereby promoting platelet adhesion and aggregation within seconds after endothelial loss and up to seven days after injury 22. Binding of platelets to the components of subendothelium matrix proteins, results in degranulation of the platelets and release of prothrombotic factors such as thrombin, ADP, serotonin, thromboxane A 2, fibrinogen, and von Willebrand factor, which promote additional adhesion and aggregation 11. Platelet- derived growth factor (PDGF) is also released at the site of injury and may stimulate VSMC activation Leukocyte Infiltration Leukocytes and monocytes are also recruited to the vessel wall following balloon injury and have been implicated in neointimal formation 24. Leukocytes secrete many cytokines and growth factors that promote neointimal growth 25. Monocyte- derived macrophages, which secrete growth factors and cytokines that promote VSMC proliferation 26, comprise approximately 1% of the total neointimal cell number 21. Taken together, these studies suggest a role of inflammation in the vessel response to balloon injury leading to neointimal formation. 6

19 1.5.3 Vascular Smooth Muscle Cell Phenotype Switch Immediately following arterial injury, the majority of medial VSMCs (up to 70%) undergo apoptosis, which will be further discussed in the following section. In the 48 hours following arterial injury, VSMC dedifferentiation is promoted by exposure to growth factors such as PDGF, fibroblast growth factor (FGF), epidermal growth factor (EGF), and TGFβ 27. In contrast, studies suggest that insulin- like growth factor- 1 (IGF- 1) may promote VSMC differentiation as it also strongly activates phosphatidylinositol- 3 kinase (PI3K)/ Akt signalling 28 in addition to mitogen- activated protein kinase (MAPK), which is the main mechanism of action for the other growth factors 29. Normally quiescent and differentiated VSMCs become active, enter the cell cycle, and divide (between 10-30% of surviving VSMCs). The phenotype switch to an activated secretory state results in increased synthesis of cytokines and growth factors that act as paracrine mediators to activate additional VSMCs, facilitate leukocyte chemotaxis and infiltration into the vessel wall 30, and stimulate production of collagen and elastin 31. Activated VSMCs also proliferate and migrate to the intima where the inflammatory components further promote the activated state 32. Activated VSMCs are characterized by changes in morphology with decreased cytoskeletal myofilaments, reduced levels of contractile proteins, and increased synthetic organelles. Smooth muscle myosin heavy chain (SM-MHC), smooth muscle 22 alpha (SM22α), and SMC α- actin are protein markers of differentiation and their protein expression decreases as VSMCs dedifferentiate. VSMC activation and phenotypic switch is reversible 33. Activation of the mammalian target of rapamycin (mtor)/ p70s6 kinase- 1 (S6K1) pathway, which is well known to promote proliferation and protein synthesis, stimulates VSMC dedifferentiation by decreasing contractile protein expression (SM-MHC, calponin, and SMC α- actin), resulting in a more active proliferative state 34. 7

20 1.5.4 Vascular Smooth Muscle Cell Proliferation and Apoptosis As mentioned in the previous section, during the initial 48 hours following arterial injury, stationary and quiescent VSMCs enter the cell cycle and divide, resulting in restoration of the number of VSMCs in the medial layer where eventually they resume their quiescent state. Medial VSMC proliferation begins as early as 24 hours after arterial injury and peaks at two days, followed by VSMC migration to the intima (which will be discussed in the next section) where VSMCs replicate and reach maximum proliferation by seven days (reviewed in 17 ). VSMCs proliferate under the influence of various mediators including oxidative stress 35, growth factors released from leukocytes and platelets, and interactions of VSMCs with the ECM 36. Initial medial VSMC proliferation is mediated by FGF in this rat arterial injury model 37. FGF is found in VSMCs and endothelial cells 37 and also targets fibroblasts and endothelial cells to stimulate proliferation in contrast to vascular endothelial growth factor (VEGF), which stimulates proliferation primarily in endothelial cells. This will be described in more detail in section In addition, expression of IGF and its receptor 39 are increased in the rat aorta after balloon injury, greater in the media compared to the intima, which suggests that IGF- 1 promotes early medial VSMC proliferation. Furthermore, the pro- inflammatory cytokines tumour necrosis factor- alpha (TNF-α), and interleukin- 1β (IL-1β), are upregulated after arterial injury and have been shown to induce intimal VSMC proliferation 40, possibly through upregulation of PDGF from endothelial cells. As previously mentioned, apoptosis is an integral part of the response to arterial injury. Immediately following arterial injury extensive medial VSMC death occurs initiated by medial VSMC stretching and growth factors secreted by VSMCs when injury breaks cell- ECM interactions (30 minutes to four hours 41 after injury). The remaining VSMCs proliferate and migrate to the neointima. Repair of the vessel after injury is also associated with apoptosis in both the medial and intimal layers and occurs between seven and 21 days after arterial injury in 8

21 the carotid artery and aorta 42. Several mediators that promote VSMC proliferation also stimulate apoptosis, including oxidative stress (ROS) and cytokines secreted by macrophages (TNFα and interleukin- 6 (IL- 6)) and T- lymphocytes (interferon- γ) 43. These mediators stimulate protooncogenes like the Bcl- 2 proteins that regulate VSMC apoptosis 44. Activation of pro- apoptotic proteins (Bcl- x s, Bad, Bid, Bax, Back) induces their translocation to the mitochondria where they release mitochondrial mediators of apoptosis such as caspases 44. The overall effect of apoptosis may be beneficial since it balances VSMC proliferation and subsequent neointimal formation Vascular Smooth Muscle Cell Migration As previously mentioned, the subpopulation of VSMCs that escape growth inhibition continue to replicate and migrate from the medial layer across the internal elastic lamina to the luminal surface. This event reaches maximum at four days after arterial injury 45. These cells colonize the intimal surface and begin to form a new layer called the neointima. The movement of VSMCs to the luminal surface follows a gradient of chemoattractants in the wall, depends upon the production of matrix degrading proteases, and requires the synthesis of new ECM proteins. Migration of VSMCs involves a dominant plasma membrane leading edge protruding in contact with an extracellular substrate, and binding integrin transmembrane receptors to form focal complexes and secure focal adhesions 46. The cell is propelled forward by actin filament alignment and myosin contraction, as well as the disengagement of the focal adhesions from the cell surface 47. The release of growth factors and ECM proteins serve as potent chemoattractants and enhance VSMC migration. PDGF is the most potent chemoattractant for VSMCs and the PDGF ligand and receptor mrna are upregulated in the neointima after arterial injury 48. PDGF-BB infusion in rats has been shown to increase neointimal growth after arterial injury mainly through 9

22 an increase in VSMC migration 23. In addition, IGF-1 is secreted by VSMCs and endothelial cells in vitro 49 and has also been reported to stimulate migration of human VSMCs 50. The ECM is also involved in the VSMC migratory response to vascular injury. Acute arterial injury increases the expression of ECM components including collagen, hyaluronic acid, and glycoproteins. Studies have shown that type VIII collagen regulates VSMC migration after arterial injury 51 by promoting VSMC attachment and chemotaxis. Hyaluronic acid levels are also upregulated in the neointima after injury 52 and have been associated with both endothelial and VSMC migration. The glycoproteins fibronectin, vitronectin, and thrombospondin-1 are upregulated in response to injury and are associated with VSMC migration 53. Fibronectin has many functions, which include VSMC migration, cellular attachment, and wound healing, and is locally elevated after vessel injury 53. Osteopontin is another glycoprotein upregulated after injury. It has been shown that neutralizing antibodies against osteopontin decrease neointimal growth after arterial injury 54. Osteopontin also stimulates matrix metalloproteinase (MMP) activity in VSMCs 55. Tenascin, another ECM protein, is also increased after arterial injury 56 and regulates cell adhesion and migration 57. ECM degrading enzymes also play a role in VSMC migration. For example, elastase activity is increased after arterial injury in rabbits 58, which could promote greater ECM turnover and facilitate VSMC migration. Cathepsins S and K, proteases that express potent elastolytic and collagenolytic activities, are also increased after arterial injury 59. PDGF activates the enzyme calcium/ calmodulin- dependent protein kinase II (CamKinaseII) 60, which affects actin filament formation and interaction with small GTP- binding proteins to stimulate VSMC migration. Activation of CamKinaseII in growth- arrested VSMCs restores the ability of these cells to migrate in response to PDGF 61. Furthermore, calcium channel blockers, amlodipine 62 for example, inhibit VSMC migration, demonstrating that calcium is required for VSMC migration, as low calcium levels inhibit CamKinaseII activation. PDGF- 10

23 induced VSMC adhesion and migration is MAPK- dependent 63. Furthermore, PDGF has also been shown to activate PI3K to stimulate Rac to promote VSMC migration 64, which may be upstream of MAPK. PDGF- induced VSMC migration is also partially dependent on integrin receptors, suggesting potential interaction with ECM proteins to further promote VSMC migration 65. Small GTPases, including Ras, Rho, Cdc42, and Rac 66, 67, are also involved in VSMC migration. The GTPases are activated by soluble factors via tyrosine kinase and G proteincoupled receptors and by cell adhesion 67. Ras activates the MAPK enzyme, extracellular regulated kinase 1/2 (ERK1/2), by activating Raf and MAPK kinase 68. Furthermore, MAPKs phosphorylate myosin- like light chain kinases to promote cell movement and migration 69. Rho is also a GTP- binding protein and is involved in mediating cytoplasmic actin filament organization that occurs in migrating cells stimulated by chemoattractants 67. Rho promotes formation of actin stress fibers and focal adhesions with integrin clusters, and inhibition of Rho has been shown to attenuate VSMC migration 70. Cdc42 promotes actin polymerization, through Rac activation, to form leading edge lamellipodia and membrane ruffles to promote cell migration. Cdc42 and Rac also activate JNK1/2 and p38 MAPK 64. Doanes et al 71 demonstrated that Rac is required for PDGF- stimulated VSMC migration. NFκB is also activated by arterial injury 72 and is linked to the inflammatory response (upregulates VCAM- 1 and MCP- 1 expression) associated with increased VSMC migration and neointimal formation 73. As mentioned in the previous paragraph, PI3K, a member of the insulin signalling pathway, is involved in PDGF- induced VSMC migration 74 as studies suggest that PI3K is necessary for PDGF- induced activation of Rac 64, 75. PI3K does not appear to be required for cell adhesion but is required for cell spreading and migration 74. PI3K may also be involved in regulating ECM protein- induced VSMC migration 65. In addition, Akt phosphorylation and subsequent mtor/ S6K1 activation correlate with increased VSMC migration 76. For example, 11

24 naringin, a compound found in citrus fruit, decreased TNFα- induced PI3K/Akt/mTOR/S6K1 signalling and VSMC migration 77. Furthermore, rapamycin, an mtor inhibitor, decreases VSMC migration 78, supporting the role of mtor in VSMC migration Matrix Metalloproteinases (MMPs) MMPs, zinc- dependent endopeptidases produced by VSMCs, macrophages, and lymphocytes, are involved in ECM degradation, which is necessary for VSMC migration from the medial layer to the intima 79. There are four types of MMPs: 1) interstitial collagenases (substrate: interstitial collagen), 2) stromelysins (substrates: laminin and fibronectin), 3) gelatinases (substrate: type IV collagen and gelatin), and 4) membrane type-mmps (MT-MMPs). The proteolytic activities of MMPs are regulated at three levels: 1) gene expression and synthesis of the inactive pro- enzyme MMP, 2) activation of the pro- enzyme by tissue/ plasma proteinases like plasmin, cell- associated MT- MMPs and by oxidative stress, and 3) inhibition by endogenous tissue inhibitors of metalloproteinases (TIMPs 1-4) (reviewed in 79 ). MMPs enable VSMC migration after carotid balloon injury in rats 45. The role of MMPs in VSMC migration has been demonstrated by studies where VSMC migration was inhibited by administration of nonselective MMP inhibitors 45, 80 and by gene transfection of TIMPs into balloon injured vessels in rats 81. Various proteinases are induced following balloon injury in various animal models, however in rats mainly MMP- 2 activity and MMP- 9 expression and activity are increased 82. Increased MMP- 2 and MMP- 9 activity is also associated with VSMC migration 82. MMPs are secreted in a pro- enzyme form and are activated by limited proteolysis 83. Total MMP activity is mainly determined by MMP expression but is also regulated independently by TIMPs and ROS 79. MMP expression may be regulated by ECM proteins, cytokines (IL- 1 and TNFα), and growth factors (bfgf, PDGF, TGFβ, and EGF) 84. Several of these activate transcription factors such as activating protein- 1 (AP- 1) (through MAPK 12

25 signalling) and nuclear factor kappa B (NFκB) that regulate MMP expression 85. Type VIII collagen, a matrix protein that is upregulated in the rat carotid artery after balloon injury 51, has been shown to bind to integrin receptors on VSMCs to stimulate MMP synthesis and VSMC migration 86. For example, αvβ3 integrin receptors are upregulated in VSMCs following balloon injury and mediate VSMC migration by stimulating MMP (MMP- 2 and MMP- 9) synthesis 55. Conversely, nitric oxide (NO) decreases MMP- 2 and MMP- 9 expression and also increases TIMP-2 production to inhibit VSMC migration 87, perhaps as a result of inhibition of the NFκB signalling cascade Extracellular Matrix Normally, the ECM acts as a barrier to VSMC migration from the media to the intima, however after arterial injury it is modified to allow the movement of cells. The ECM constitutes the majority of the neointima as cellular components make up only approximately 11% 17. The ECM in the artery is composed of the scaffolding elements of collagens (types I, III, IV, and VI) and elastin embedded in a mixture of glycoproteins (e.g. fibronectin) and proteoglycans (e.g. heparan sulphate). Both collagen and elastin are synthesized by endothelial cells, intimal and medial VSMCs, and adventitial fibroblasts 88. Collagens make up the major structural component of ECM in the artery and their expression is increased after balloon injury 88. Collagen exists as either a meshwork surrounding individual VSMCs or as interstitial dense fibers that occupy a large volume of the tissue 89. Elastin is one of the most abundant proteins of large arteries and provides elasticity 89. The synthesis of the ECM proteins collagen (type VIII collagen in particular 51 ) and elastin is increased after arterial balloon injury reaching significant elevation at 21 and 60 days after injury 90, despite the increased production of MMPs, plasminogen activators and plasmin that degrade matrix proteins 45, 91. This may be due to a transient induction of proteases that are mainly involved in VSMC migration and not ECM degradation 90. Increases in elastin synthesis 13

26 appears to be more significant than collagen synthesis, which may be explained by elastin s sensitivity to mechanical forces of blood flow or balloon expansion attributed to the greater distensibility of elastin 90. The ECM interacts with cells via cell surface receptors like integrins, which are upregulated after balloon injury in rats. Binding of collagen to integrins initiates intracellular signalling cascades by recruiting cytoplasmic tyrosine kinases and adapter proteins to focal adhesion sites at the cell membrane 92. As previously discussed, dynamic adhesive interactions between cell surface integrins and the ECM play a central role in VSMC migration after balloon injury by providing cell- substrate contact and cell motility 75. Re- expression of the embryonic forms of fibronectin occurs in the media and adventitia of rabbit arteries 24 to 48 hours after injury; two weeks after balloon injury when the neointima is formed, fibronectin mrna and protein accumulates in the luminal layers of the neointima 93. Osteopontin is also elevated after arterial injury 94 and as previously mentioned, neutralizing antibodies against osteopontin decrease neointimal growth after injury 54. TGF-β1 increases the synthesis of fibronectin, fibrillar collagens, elastin, thrombospondin, and proteoglycans, all of which are increased after arterial injury Endothelial Regeneration Arterial balloon angioplasty results in complete endothelial denudation, however soon after injury neighbouring endothelial cells begin to replicate in response to growth factors such as VEGF 96 and bfgf 97, to restore the protective endothelial layer and prevent neointimal formation. VEGF is well known to stimulate endothelial cells, however it does have effects on monocytes, macrophages, hematopoietic stem cells, and VSMCs (reviewed in 98 ). The mitogenic action of VEGF primarily affects endothelial cells, thus providing an advantage of VEGF application in the vessel. Phosphorylation of endothelial nitric oxide synthase (enos) at Serine 1177 by Akt is required for VEGF- induced endothelial cell migration 99. Furthermore, VEGF binding to VEGFR2 has been shown to initiate autophosphorylation of the receptor and lead to 14

27 Grb2 activation through Src binding. This leads to ERK1/2 and p38 MAPK activation, which is involved in endothelial cell proliferation and migration respectively 100. Despite this, the role of VEGF in reducing neointimal formation remains unclear as VEGF treatment was also shown to promote restenosis 101. In addition, a clinical trial of local VEGF delivery for restenosis did not show any benefit 102. Regeneration of the endothelium begins from the leading edge of the denuded area within 24 hours of arterial injury and ceases approximately six weeks later 103, however, depending on the extent of injury, it is seldom complete in the carotid artery 104. If the denuded area is a relatively short distance between adjacent sources of endothelium, more complete endothelial regeneration occurs 104. Neointimal thickness is decreased by this re- endothelialization process as the endothelial layer decreases the proliferation of underlying VSMCs through NO production 104, which is discussed further in section 1.7. Therefore, delays in re- endothelialization result in greater neointimal growth. In vivo, administration of L- arginine, which is catalyzed by NOS to release NO, prevents neointimal growth 105. enos transfection into the vessel wall after balloon injury in rats increased local NO production and decreased VSMC proliferation, migration, and ECM production, which are involved in neointimal formation 106. Additionally, in bovine and porcine models of carotid injury, perivascular implantation of endothelial cell grafts reduced neointimal growth by increasing NO availability 107. Neointimal growth is also increased in enos knockout mice after carotid ligation despite an intact endothelium 108. These studies provide evidence that restoration of the endothelium and increased NO availability are important for limiting neointimal formation after arterial injury Progenitor Cell Contribution Stem cells are capable of self- renewal and differentiation while progenitor cells are more differentiated. Together they constitute approximately 0.1% of total bone marrow cells in humans. Stem cells have been broadly classified as either hematopoietic (HSC) or mesenchymal 15

28 (non- hematopoietic) (MSC) stem cells. HSCs differentiate into hematopoietic progenitor cells and then into various blood cell lines, whereas MSCs differentiate into adipocytes, chondrocytes, osteoblasts, and muscle cells (reviewed in 109 ). Endothelial progenitor cells (EPC), a subset of bone marrow- derived progenitor cells, are thought to have the same precursor as hematopoietic stem cells 110. In addition to the progenitor cells residing in the bone marrow, there is a small population of progenitor cells maintained in the circulation 111, which can differentiate into either endothelial or VSMCs. There is evidence in studies using bone marrow transplant in combination with mouse 112 or rat 18 models of vessel injury, that several types of progenitor cells are also a source of newly differentiated VSMCs and increase neointimal formation. However, other studies in mice suggest that the new population of VSMCs at the site of vessel injury originate as either progenitor cells that permanently reside in the vessel wall 113 or as VSMCs resident in the vessel wall 19 and not from bone marrow- derived progenitor cells. The controversy is a result of the methodology chosen to identify the cells as being double- positive for a bone marrow- derived cell and VSMC. Markers for VSMCs (SMCα- actin in particular) can be expressed by other cell lineages and also the low resolution microscopy used in some studies can lead to false positive identification of cells (reviewed in 114 ). As mentioned in the previous section, restoration of the endothelium decreases neointimal formation and it was previously thought that the endothelial cells at the perimeter of the denuded area were the primary source of endothelial cell replication, however reendothelialization may also occur from circulating or bone marrow- derived progenitor cells. Recent studies have shown that circulating and bone marrow- derived EPC recruitment accelerates re- endothelialization and decreases neointimal formation after vessel injury 20. Furthermore, Fadini et al 115 demonstrated that diabetes impairs EPC mobilization after 16

29 ischaemia- induced injury in rats and that lowering of hyperglycemia with insulin partially recovered EPC mobilization. The majority of studies were performed in mice and the progenitor cell contribution to neointimal growth appears to vary by the type of vessel injury 116. Tanaka et al 116 showed that mechanical injury using a straight wire to denude and dilate the mouse femoral artery resulted in the greatest progenitor cell recruitment to the arterial wall compared to a polyethylene tube placed around the femoral artery or ligation of the common carotid artery proximal to the bifurcation. Bone marrow- derived progenitor cell recruitment after arterial injury to the neointima occurs in three stages; 1) cell mobilization from the bone marrow, 2) bone marrow cell migration and recruitment to the site of injury, and 3) bone marrow cell differentiation into vascular cells. These stages are closely related as several signalling molecules are involved in multiple stages. Studies suggest that bone marrow progenitor cells are mobilized to the peripheral circulation in response to injury- induced stress signals 117. Exit from the bone marrow, which is rich in stromal cells and ECM proteins that attach to the progenitor cells through integrin interaction, requires migration of progenitor cells through a vascular barrier that separates the hematopoietic compartment from the circulation 117. Granulocyte Colony Stimulating Factor (G-CSF) stimulates progenitor cell mobilization 118 by increasing proteases to cleave adhesion molecules. G-CSF also stimulates MMP- 9, which releases soluble c- kit and c- kit ligand (KitL) (also known as Stem Cell Factor) so that c-kit positive progenitor cells can be recruited 119. Similarly, stromal derived factor- 1α (SDF- 1α) also increases progenitor cell mobilization in the bone marrow by activating MMP- 9, which subsequently cleaves KitL 119. Once in the peripheral circulation, progenitor cells migrate to the site of injury. SDF- 1α stimulates progenitor cell recruitment by increasing platelet adhesion to the site of injury and 17

30 stimulating platelet p- selectin to promote progenitor cell adhesion and recruitment 120. VEGF/ VEGFR also increase progenitor cell migration and recruitment to the site of injury 101. At the site of arterial injury, bone marrow- derived progenitor cells differentiate into endothelial cells or VSMCs to integrate into the neointima. VEGF promotes endothelial cell differentiation 121, whereas TGF- β and PDGF- BB have been demonstrated to induce differentiation into VSMCs 122. Direct cell- to- cell contact may also stimulate differentiation 122. The origin and contribution of progenitor cells to the vessel response to injury is very controversial as much of our understanding is from a small number of animal studies. Even less is known regarding the specific contribution of progenitor cells to vascular disease in humans, however stem cell therapy for vascular disease is currently being tested in clinical trials Type 2 Diabetes and Metabolic Syndrome Type 2 diabetes has become a worldwide epidemic where macrovascular complications are the major cause of mortality in diabetic patients 3. Type 2 diabetes is characterized by insulin resistance, defined as the inability of insulin to exert its metabolic actions, and a defect in insulin secretion from the pancreatic β cells, which is inadequate to compensate for insulin resistance 3. More than 80% of type 2 diabetic subjects are obese and obese persons often have elevated plasma levels of free fatty acids (FFAs) due to their expanded and more lipolytically active adipose tissue stores. It is now well established that insulin resistance itself is a potent risk factor for atherosclerosis and restenosis after percutaneous interventions 124, and the risk increases when hyperglycemia is present 125. As a result, diabetic patients account for approximately one quarter of coronary revascularization procedures 126 and have a greater restenosis rate than non- diabetic subjects after percutaneous interventions such as angioplasty and stenting and require repeat revascularization 127. Insulin resistance is a typical feature of a cluster of risk factors for both atherosclerosis and type 2 diabetes, which has been termed the metabolic syndrome. 18

31 The metabolic syndrome affects approximately 25% of the population 128 and is characterized by a cluster of abnormalities which include central obesity, insulin resistance, compensatory hyperinsulinemia, dyslipidemia, and hypertension and confer a high risk for type 2 diabetes and atherosclerosis. Currently, the most accepted hypothesis to describe the pathophysiology of metabolic syndrome is insulin resistance, which can be caused by obesity via the release of fatty acids and other secretory products such as cytokines from the expanded fat tissue 128. Hyperinsulinemia is generally considered to be secondary to insulin resistance but may also be stimulated by excess circulating nutrients. Furthermore, excess circulating insulin can aggravate insulin resistance, contributing to a vicious cycle. The metabolic and cardiovascular features of metabolic syndrome (i.e. dyslipidemia, hypertension) greatly contribute to atherosclerosis; however there are vascular effects of insulin by itself, which are both atherogenic and vasculoprotective. The atherogenic effect of insulin appears to be MAPKdependent 129 while the vasculoprotective effect appears to involve the PI3K pathway 130. Insulin has mostly protective effects in the endothelium and platelets 131, which are present in the intact vessel, but are absent in VSMC cultures. Therefore, the prevailing effect of insulin on atherogenesis in vivo is now considered to be protective and instead, insulin resistance is considered to be harmful, because of the unresponsiveness to insulin action, including its vascular anti- inflammatory properties 132. Studies from our laboratory support a protective role of insulin against restenosis 133, 134 and will be discussed further in this thesis. The obesity- associated pro- inflammatory state due to cytokine release from adipose tissue and to ROS generation by excess nutrients has recently been suggested to be the major cause of insulin resistance 128. Oxidative stress and associated inflammatory damage are also mediators of vascular injury in atherosclerosis and restenosis. Therefore, it is possible that in obesity, both insulin resistance and atherosclerosis are caused, at least in part, by inflammation. 19

32 1.7 Nitric Oxide NO plays a pivotal role in the regulation of vascular homeostasis. NO is produced through the conversion of L- arginine to L- citrulline by NOS and exerts its effects by activating guanylate cyclase, which increases cyclic guanylate monophosphate (cgmp) formation, and then activates protein kinase G (PKG) 135. NO bioavailability in the vasculature is greatly influenced by the presence of superoxide. Superoxide degrades NO and forms peroxynitrite 136, which plays a role in the pathogenesis of several diseases. The family of NOS enzymes consists of three members: enos, inducible NOS (inos), and neuronal NOS (nnos). Although these isoforms are very different, they all require the cofactors nicotinamide adenine dinucleotide phosphate hydrogen (NADPH), flavin adenine dinucleotide (FAD), flavin mononucleotide (FMN), and tetrahydrobiopterin (BH 4 ) to catalyze the enzyme reaction 137. enos and nnos are constitutively present in many tissues and NO production is mainly regulated by intracellular calcium fluxes that permit calmodulin binding, which activates the enzyme. enos gene expression can also be up- or down- regulated by various stimuli 138 and regulated through phosphorylation of the enzyme. enos can be phosphorylated on multiple serine, threonine, and tyrosine residues, however Serine1177 is the major positive regulatory site (reviewed in 139 ) and is phosphorylated by Akt (activated by insulin 140, VEGF 141 ) and 5 adenosine monophosphate- activated protein kinase (AMPK) (activated by resveratrol (RSV) 142, metformin 143 ). In contrast, inos is not normally produced by most cells as it is transcriptionally regulated and is induced by cytokines 144. NO output differs between each NOS isoform where enos produces the least and inos produces the greatest amount of NO. Reduced bioavailability of the main substrate L- arginine or depletion of the NOS cofactor BH 4 by peroxynitrite- induced oxidation 145 can compromise NOS function and result in uncoupling of the NOS dimer. The term NOS uncoupling has been used to describe this 20

33 event whereby electron flow through the enzyme in interrupted and superoxide is produced instead of NO. As previously mentioned, superoxide can then react with NO to form peroxynitrite, thus promoting a vicious cycle to further deplete NO. In general, NOS uncoupling is detrimental to the vasculature as it results in both an increase in ROS production and loss of the protective effects of NO. In the rat arterial balloon injury model, the main sources of NO are endothelial cells (in an injured vessel there are endothelial cells along the perimeter of the de- endothelialized area 146 ), circulating EPCs 20, platelets 147 (all of which express enos), and VSMCs (express inos and nnos) 148. It is well established that NO diminishes neointimal growth after injury 105, 149, whereas NOS inhibition aggravates restenosis 150. Gene transfer studies have shown that upregulation of enos diminishes neointimal formation after arterial injury in rats 151, 152. In turn, enos deficiency results in greater neointimal growth after cuff injury in mice 153. The primary role of inos on neointimal formation is controversial, as studies in knockout mice indicate that inos accelerates restenosis 154, whereas several gene transfer studies indicate that it decreases it Based on the latter studies, the prevalent view is that inos induction after injury is beneficial, although its effect may be variable because inos can produce superoxide when substrates and cofactors for synthesis of NOS are lacking 158. There is one report of induction of nnos in VSMC after arterial injury and aggravation of neointimal formation after nnos inhibition 159, however another study reported that nnos was not present in vessel cross- sections before or after arterial injury in rabbits 160. Consistent results have yet to be demonstrated in humans, although numerous animal studies have demonstrated a protective effect of NO on restenosis after vessel injury. In the Angioplastic Coronaire Corvasal Diltiazem study, patients that underwent angioplasty and received NO from i.v. linsidomine followed by oral molsidomine for six months had a decreased occurrence of restenosis 161. However in a separate study, there was no effect of NO in patients 21

34 that received high- dose oral molsidomine for six months after angioplasty 162. In addition, the delivery method of NO is limiting as systemic administration can cause hypotension, headaches, and increased bleeding. This highlights an advantage of treatment with compounds such as insulin or RSV that may also stimulate NO production to inhibit restenosis and could be administered locally (i.e. in a drug- eluting stent). NO acts through several mechanisms to protect the vasculature. Specifically, NO induces vasorelaxation, and inhibits several of the processes associated with atherosclerosis development following endothelial cell damage, including leukocyte- endothelial adhesion, VSMC migration and proliferation, as well as platelet aggregation (reviewed in 163 ). In the rat arterial balloon injury model NO inhibits platelet aggregation and adhesion, and increases blood flow 164. Furthermore, in the same study inos expression was increased in the innermost layer of the media and in the neointima, suggesting that inos may be a major source of NO when the endothelium is not present. NO has also been shown to inhibit leukocyte accumulation as an increase in adherent leukocytes was found in inos knockout mice 165. As discussed in section 1.5.8, the effect of NO on the endothelial layer is protective, as NO stimulates endothelial regeneration after vessel injury in the rat 166. It appears that VEGF, when added to coronary endothelial cells in culture, increases cytoplasmic calcium levels to stimulate enos and NO production and subsequently cgmp, resulting in MAPK activation and increased cell proliferation 167. In this model, the mechanism of cgmp to promote MAPK activation was not examined. Furthermore, in enos knockout mice, the angiogenic response (which requires endothelial cell proliferation) to hindlimb ischemia was diminished and could not be stimulated by VEGF 168. NO also suppresses endothelial cell apoptosis to maintain the integrity of the endothelium. For example, cultured endothelial cells subjected to shear stress 169 or lipopolysaccharide 170 did not undergo apoptosis in the presence of increased NO. 22

35 In contrast to endothelial cells, studies have shown that NO plays a role in decreasing VSMC proliferation and the mechanism(s) are more clear 171, 172. NO inhibits the G1/ S phase transition to induce cell cycle arrest by upregulating the expression/ activity of cell cycle regulatory proteins like p21 or cyclin- dependent kinase NO also activates cgmp and PKG to decrease VSMC growth, as NO donors inhibit DNA synthesis in VSMCs 171. This effect involves PKG to decrease intracellular calcium levels by inhibiting calcium release from the sarcoplasmic reticulum, decreasing calcium influx across the plasma membrane, and promoting calcium efflux. The decrease in intracellular calcium reduces calcineurin activation and thus inhibits proliferation 174. Furthermore, the growth inhibitory effect of NO, possibly from endothelial cells, may maintain VSMC quiescence 171. Xu et al 175 demonstrated that endothelial cells stimulate thioredoxin, an endogenous anti- oxidant, through a NOS- and NADPH oxidaseindependent mechanism to decrease ROS in VSMCs and inhibit VSMC proliferation. Studies also indicate that NO induces apoptosis in rabbit VSMCs 176. inos overexpression in transfected VSMCs promoted apoptosis, whereas L-NMMA, a NOS inhibitor, decreased apoptosis in inostransfected VSMCs 177. This increase in NO production by inos was associated with an increase in the tumour suppressor gene p53, which induces apoptosis in many cell types. NO also inhibits VSMC migration, as NO donors inhibited the number and distance of migrating VSMCs 178 and also angiotensin- II- induced VSMC migration measured in vitro 179. As previously discussed, migration of VSMCs depends upon growth factor- induced increases in cytosolic calcium, which may be inhibited by NO by preventing the increase in calcium 180. Furthermore, VSMC migration is prevented by the ECM barrier, whereas MMPs degrade the basement membrane and ECM to allow VSMC migration. enos transfection into rat VSMCs decreased MMP- 2 and MMP- 9 activity and increased TIMP- 2 activity 181, suggesting that NO may inhibit VSMC migration in part by regulating MMPs. NO also affects matrix elements. L- arginine, the substrate for NO production, decreased hyaluronan synthase 23

36 expression 182. There are contradictory reported effects concerning the effect of NO on ECM proteins. NO has been shown to inhibit basal type I collagen levels 183 and collagen synthesis in vitro 184 whereas NO donors increased collagen I and III synthesis in human coronary VSMCs 185. Furthermore, NO treatment decreased osteopontin expression in human saphenous vein segments ex vivo 186. Taken together, these studies suggest that NO has a protective effect to prevent neointimal hyperplasia after vessel injury as it inhibits several key processes, including VSMC migration and proliferation and ECM production, involved in the formation of the intima. 1.8 Insulin General effects of insulin. Insulin action results in a variety of biological outcomes, including control of glucose, fat, and protein metabolism, cellular growth and development 187. In the liver, insulin increases glycogen synthesis and suppresses hepatic glucose production. Insulin promotes energy storage in adipocytes by increasing glucose and FFA uptake, and stimulating lipogenesis while inhibiting lipolysis. In skeletal muscle insulin stimulates glucose and amino acid uptake and promotes glycogen and protein synthesis. The role of insulin in metabolism is well defined and established; however the effect of insulin in the vasculature has only more recently been considered and is less clear. As previously mentioned, it is debated whether insulin is protective or detrimental in the vasculature. Protective effects of insulin include stimulation of NO production (through PI3K activation) and subsequent vasodilation, inhibition of platelet adhesion and aggregation, and anti- inflammatory effects, whereas the detrimental effects include growth- promotion (through MAPK activation) and vasoconstriction. 24

37 1.8.2 Effects of insulin on atherosclerosis. Initially, insulin was thought to promote atherosclerosis, based on epidemiological studies which showed an association between hyperinsulinemia and atherosclerotic CVD 188, 189. However, this concept has been challenged as hyperinsulinemia is often accompanied by other risk factors such as dyslipidemia and hypertension 190 and usually coexists with (and is a marker of) insulin resistance. Epidemiological studies have described an association of atherosclerotic CVD and insulin resistance 124, 191. In the Insulin Resistance Atherosclerosis Study (IRAS) 124, the association was greater than that with hyperinsulinemia. The Diabetes Control and Complications Trial (DCCT) 192 and United Kingdom Prospective Diabetes Study (UKPDS) 193 studies have demonstrated that intensive treatment of diabetes to increase insulin levels or action does not increase but actually decreases atherosclerotic CVD. These results are likely due to the effect of insulin to improve glycemic control, nevertheless they suggest that insulin is, at least, not detrimental. Whether insulin is atherogenic or vasculoprotective has not been adequately assessed in animal models of atherosclerosis, in part due to the limitations of the non- transgenic models. Early studies showed that insulin treatment aggravated atherosclerosis in cholesterol- fed alloxan- diabetic 194 and non- diabetic rabbits 195, and delayed the regression of atherosclerosis in cholesterol- fed chickens re-fed a normal diet 196. In contrast, in later studies insulin treatment did not promote atherogenesis in cholesterol- fed rabbits 197. Instead, massive doses of oral insulin absorbed systemically protected against atherogenesis (decreased both size and number of lesions) in Apolipoprotein E (ApoE) knockout mice 198 (ApoE knockout mice develop 199, 200 spontaneous atherosclerotic lesions without high- fat feeding), whereas systemic knockout and disruption of only one allele 201 of insulin receptor substrate- 2 (IRS- 2) aggravated atherogenesis in the same model compared to the ApoE knockout mice with intact IRS- 2. In addition to the species differences, the discrepancy over whether insulin is protective or 25

38 atherogenic may be explained by the different diets used in these studies. For example, in the rabbit and chicken studies high- cholesterol diets were used, whereas either normal chow or high- fat diets were used in the studies with mice. Both high- cholesterol and high- fat diets are known to induce insulin resistance, which was not measured in the studies mentioned, and in the insulin resistant state VLDL production by the liver is increased (reviewed in 202 ). Insulin is lipogenic and normally has an acute suppressive effect on VLDL production from the liver. However, it is not clear whether in conditions of insulin resistance the increase in VLDL production is due to insulin resistance or hyperinsulinemia per se (reviewed in 202 ). Therefore, it is possible that high- cholesterol feeding in combination with insulin treatment may have increased lipid deposition in the vessel wall resulting in promotion of atherosclerosis. The latter findings can be taken as evidence for a protective effect of insulin signalling, unless compensatory and possibly atherogenic pathways (e.g. Shc, IRS- 1) were potentiated. In the mouse, complete disruption of IRS- 2 also results in systemic effects which in turn can influence atherogenesis, such as hypertension and hyperglycemia, in addition to insulin resistance and hyperinsulinemia 203 and similarly in doubly deficient mice (ApoE/ IRS- 2- null) increased glucose intolerance, hyperinsulinemia, and insulin resistance are present 200. Partial disruption of IRS- 2 (single allele) in ApoE- null mice fed standard chow aggravates atherosclerosis but does not affect body weight, circulating glucose levels, fasting plasma insulin levels, glucose tolerance or glucose- stimulated insulin secretion, however when fed a high- fat diet, heterozygous IRS- 2 null mice show increased fasting plasma insulin levels and increased glucose intolerance 201. Although systemic IRS- 2 knockout aggravates atherosclerosis, the effect of selective disruption in insulin signaling is controversial. Recent studies have reported opposite results on atherosclerosis with selective knockout of insulin receptor or IRS- 2 in myeloid cells in low- density lipoprotein receptor (LDLR) knockout 204 (LDLR knockout mice develop atherosclerotic lesions with high- fat feeding) and ApoE knockout mice 205, thus rekindling the 26

39 debate about the atherogenic vs. vasculoprotective effects of insulin. However, preliminary studies in ApoE- null mice with selective endothelial knockout of insulin receptor indicate that the action of insulin in endothelium is indeed protective 206. IGF- 1 shares similar signaling with that of insulin but preferentially stimulates MAPK versus PI3K and along with other growth factors like PDGF, traditionally has been considered to promote the development of atherosclerosis by stimulating VSMC migration and proliferation 207. Furthermore, its expression was found to be increased in aortic lesions of ApoE knockout mice compared to non- lesion areas 208. However, a recent study reported that low- dose IGF- 1 decreased plaque burden in ApoE knockout mice while increasing phosphorylated Akt, enos, and circulating EPCs 209, suggesting a beneficial role of IGF- 1 in atherosclerosis, similar to the role of insulin Effects of insulin on restenosis. A few studies have evaluated the effect of insulin on the arterial response to injury. Some studies suggest that insulin treatment aggravates intimal thickening in models of type 1 diabetes, perhaps by increasing IGF , 211. In models of insulin resistance, such as the Zucker Fatty rat 212, the fructose- fed rat 213, and the Zucker diabetic fatty (ZDF) rat 214, neointimal growth after injury is increased, however it is not clear whether this is due to insulin resistance or to hyperinsulinemia. Insulin treatment of non- diabetic rats did not affect DNA synthesis in injured aortas during normoglycemia or hyperglycemia but decreased DNA synthesis in the presence of hypoglycemia 215, however, intimal area was not measured. Prior to our studies, there were only a few reports on the effect of insulin on intimal thickness after balloon injury in non- diabetic models and their results were opposite: in one study by Indolfi et al 216, insulin treatment via islet transplantation increased neointimal growth, whereas in another study by Kubota et al 217 neointimal growth after cuff- injury was greater in mice with a defect in insulin signalling (IRS- 2 null mice). Thus the effect of insulin on restenosis remains a controversial issue. 27

40 A major goal of our laboratory is to examine the effects of insulin on restenosis and explore the potential mechanisms involved. In the initial study by my colleagues, the effect of insulin on vascular growth after balloon catheter injury in rats fed a control (12% fat in calories) or a high- fat diet (60% fat) was determined 133. Plasma insulin levels were elevated (levels equivalent to those seen in insulin resistant Zucker Fatty rats) by s.c. insulin implants (5U/day in a g rat) given two days before carotid balloon injury. The insulin- treated groups received glucose in drinking water to avoid frank hypoglycemia (<3.3 mm), however plasma glucose levels declined to the non- hypoglycemic range (~ 4 mm). Intimal area evaluated 14 days after injury was decreased in insulin- treated rats on the control diet, however the effect of insulin was diminished in the rats on the high- fat diet. Intimal cell proliferation was greater than control in all treated groups. Thus, a mitogenic effect of insulin was observed at this dose, however the predominant effect of insulin was to decrease neointimal growth despite increasing proliferation. These results are the first evidence for a protective effect of subcutaneous insulin treatment in a model of restenosis and are in accordance with those of the previously mentioned mouse study with increased neointimal growth when insulin signaling was disrupted (IRS- 2 deficiency) 217. In a subsequent study, we defined the effect of insulin on VSMC kinetics over the time course of neointimal formation 134. A lower dose of insulin (3U/day in a g rat) and i.p. glucose infusion soon after surgery allowed us to investigate the effect of insulin concentrations that can be seen also in non- insulin resistant rats after an oral glucose load, while avoiding glucose- lowering. Insulin decreased neointimal area at its peak (28 days), reduced elastin content and tended to decrease collagen content. Intimal proliferation and apoptosis were not changed by this dose of insulin. However, insulin markedly inhibited cell migration evaluated at its peak (four days after injury). 28

41 In preliminary studies we found that insulin inhibits MMP- 2 expression and activity and MMP- 9 activity, which is evidence for a mechanism whereby insulin could inhibit VSMC migration 218. However, in contrast to our studies, Foster et al 219 reported that subcutaneous insulin (~4U/day) given by an osmotic pump enhanced neointimal formation after vascular injury in normal rats. Instead of the 40% glucose water used in our studies 133, 10% sucrose water was given to maintain normoglycemia, which could have induced some insulin resistance 220. Thus, it is possible that the increase in neointimal growth observed when insulin was combined with oral sucrose was due to sucrose and not to insulin, which is addressed in Study 1 of this thesis. In agreement, Pu et al 221 has also shown that insulin infusion (~1.6U/day in a g rat equivalent to ~3.6U/day when compared to a g rat), using the same s.c. insulinreleasing pellet as in our studies 133, 134, increases neointimal formation possibly via suppression of protein tyrosine phosphatase 1B (PTP1B) function leading to enhanced PDGF receptor signaling. Most recently the insulin analog glargine (equivalent to ~3.5U/day) was shown to reduce neointimal formation in Zucker fatty rats 222, which exhibit hypertriglyceridemia, obesity, hyperinsulinemia, and insulin resistance. The attenuation of neointimal formation was associated with a decrease in tissue oxidative stress and an increase in plasma IGF- 1 levels. Therefore, it is possible that the beneficial effect of glargine is due to insulin and/ or IGF- 1. As discussed above, there are suggestions of a dual effect of IGF- 1 on atherosclerosis, similar to insulin. IGF- 1 is overexpressed in the rat aorta after arterial injury 38, however its major effect on restenosis does not seem to be protective. For example, a D- peptide analogue of IGF- 1 inhibited VSMC proliferation but did not decrease neointimal formation after carotid balloon injury in the rat 223. In the same model, blocking IGF- 1 signalling with a dominant negative IGF- 1 receptor 29

42 adenovirus decreased neointimal formation 224. Subsequent studies in mice have also shown that IGF- 1 increased intimal thickness after arterial injury Mechanism of Insulin Action in the Vasculature General features of insulin signal transduction pathways. Insulin signal transduction involves two major pathways- 1) the PI3K signalling pathway (primarily responsible for metabolic functions) and 2) the MAPK signalling pathway (important for cell growth and development) (please see Figure 1.1 for a summary of insulin signalling). Insulin binding to its receptor results in autophosphorylation of tyrosine residues (reviewed in 226 ). The resulting tyrosine kinase activity enables the insulin receptor to phosphorylate the tyrosine residues of IRS- 1 or IRS- 2 and Shc, which serve as docking proteins for downstream signalling molecules. Phosphorylated IRS is then able to associate with the p85 regulatory subunit of PI3K, which activates the catalytic subunit p110. Once activated, PI3K phosphorylates the substrate phosphatidylinositol 4,5- bisphosphate (PIP 2 ) to generate the lipid phosphatidylinositol 3,4,5 trisphosphate (PIP 3 ), which in turn activates phosphoinositidedependent protein kinase- 1 (PDK- 1) and PDK- 2. PDK- 1 activates Akt (phosphorylates the Threonine 308 residue) and atypical protein kinase C isoforms ζ and λ (PKCζ and PKCλ). The role of PDK- 2 is not clear and it is suggested that PDK- 2 is actually rictor- mtor (TORC2), a protein complex of mtor, rictor, and mammalian stress- activated protein kinase interacting protein 1, that regulates the cytoskeleton. PDK- 2 phosphorylates Akt at the Serine 473 residue, a separate site from PDK- 1, and it is suggested to mediate mtor- S6K1 binding (reviewed in 227 ). Akt phosphorylates many substrates including; enos to stimulate NO production 228 and forkhead transcription factor (FOXO1) to inhibit gluconeogenesis. Insulininduced Akt phosphorylation also activates mtor to stimulate S6K1 and/or phosphorylate eukaryotic initiation factor- 4E binding protein- 1 (4E-BP1) to promote protein synthesis 30

43 (reviewed in 229 ). Akt also phosphorylates the pro- apoptotic Bcl- 2 family member Bad to prevent apoptosis 230. As previously mentioned, the docking protein Shc interacts with the insulin receptor resulting in phosphorylation of tyrosine residues on Shc, thus enabling it to bind to growth factor receptor- binding protein 2 (Grb2). Grb2 can also bind to IRS- 1 for activation. Grb2 brings the pre- associated GTP exchange factor Son of sevenless (Sos) to the proximity of Ras. This puts Ras in the GTP bound form (the activated form), thus enabling Ras to stimulate Raf, which initiates a kinase phosphorylation cascade involving MEK1/2 and ERK1/2 MAPK 231. MAPK activation regulates growth and mitogenesis through gene transcription and also increases endothelin- 1 (ET- 1) secretion to induce vasoconstriction. Protein tyrosine phosphatases, such as PTP1B, SHP- 1, and SHP- 2, dephosphorylate the insulin receptor and IRS- 1 and lipid phosphatases, such as SHIP- 2 and PTEN, dephosphorylate PIP 3 to inhibit insulin signalling 232. Very recently, insulin- stimulated NO production was shown to inactivate protein tyrosine phosphatases to enhance insulin responsiveness in endothelial cells Insulin stimulates NO production and induces vasodilation. One of the most important protective vascular effects of insulin is stimulation of enos 234, to increase NO production, via PI3K and Akt in endothelial cells Insulin has also been reported to upregulate nnos in neurons 238, and inos in VSMCs 239. Furthermore, insulin treatment was also shown to augment NO- stimulated guanylate cyclase activity in VSMCs 240. enos has many phosphorylation sites, however it is widely accepted that the Serine 1177 residue (equivalent to Serine 1179 in bovine enos 235 ) is the major regulatory site of insulinstimulated enos activation 241, 242. However, a very recent study suggests that enos phosphorylation at Serine 615 also contributes to insulin- stimulated NO production 243. This phosphorylation- dependent mechanism to activate enos is calcium- independent 237. Even though Akt is necessary for insulin- stimulated activation of enos, Akt activation alone is not 31

44 sufficient for enos activation 235. A possible explanation for this specificity is that association of heat shock protein 90 (HSP90) with enos is required for enos- mediated NO production 244. Takahashi et al 244 demonstrated that insulin stimulation causes HSP90-Akt-eNOS association and results in subsequent enos phosphorylation, whereas in the absence of HSP90, Akt was not observed in the enos complex. The interactions between HSP90, Akt, and enos are not clear, however HSP90 may facilitate Akt activation of enos by acting as a scaffold to promote kinasesubstrate association. As previously mentioned, insulin does not affect intracellular calcium levels in endothelial cells, suggesting that insulin- stimulated NO production is calcium- independent 237 and therefore dissimilar to classical vasodilators such as acetylcholine. NO diffuses into the VSMCs and endothelium to promote vasodilation 235. Insulin administered i.v. in humans increases total limb blood flow 245. Furthermore, impaired vasodilation is a characteristic of individuals with type 2 diabetes and increases the risk of atherosclerosis development. Insulin therapy in type 2 diabetic patients stimulates vasodilation and increases blood flow Insulin stimulates ET- 1 and induces vasoconstriction. On the other hand, insulin also induces vasoconstriction 247. Insulin stimulates the production and secretion of ET- 1, a vasoconstrictor released from the vascular endothelium, in endothelial cells through increases in ET- 1 gene expression 248. In vivo, insulin infusion increased plasma ET- 1 levels and induced hypertension in rats 249. It has also been reported that insulin stimulates vasoconstriction via ET- 1 during NOS inhibition 250 and that this response is mediated by MAPK 251. Furthermore, obese and insulin resistant individuals have excess circulating levels of ET Taken together, these studies suggest that insulin resistance and the associated hyperinsulinemia may stimulate ET- 1 production by endothelial cells. In the insulin resistant state the balance between NO and ET- 1 production by insulin is altered in favour of ET- 1 and results in vasoconstriction. Therefore, as insulin has opposing vasodilator and 32

45 vasoconstrictor actions, the net effect of insulin on blood pressure is minimal in healthy individuals Insulin inhibits platelet function. Human platelets contain insulin receptors, which are involved in the regulation of platelet function. Furthermore, enos is also present in human platelets and can be stimulated by insulin to increase NO production 254, which decreases platelet adhesion and aggregation 131. Studies have also shown that insulin inhibits platelet aggregation, perhaps in part through stimulation of NO production, in non- obese individuals, however this effect was not present in insulin resistant obese and hypertensive individuals 255. Therefore, an inhibitory effect on platelet aggregation may contribute to a protective effect of insulin on the vasculature Insulin has anti- inflammatory effects. Initially, from epidemiological studies, insulin was thought to increase the risk of atherosclerotic CVD 188, 189. It was suggested that insulin stimulation of the MAPK signalling pathway promoted inflammation and thrombosis by stimulating monocyte migration and vascular cell expression of PAI However, the majority of studies have demonstrated that insulin has important anti- inflammatory properties (reviewed in 257 ) and that it is insulin resistance that is detrimental. Insulin administration in mice suppressed macrophage cholesterol synthesis and content, macrophage lipid peroxide content and ROS generation 198. Furthermore, in other animal studies insulin suppressed endotoxin- 258 and thermal injury- 259 induced pro- inflammatory transcription factors and genes. In endotoxin- treated human macrophage- 1 cells, the stimulated increase in apoptosis, TNFα, and IL-1β secretion was decreased by insulin 260. Insulin decreased the expression of intracellular adhesion molecule- 1 (ICAM- 1) in human aortic endothelial cells through a NOS- dependent mechanism 261. Recent studies in humans have highlighted the protective effect of insulin in vivo on blood markers of inflammation and thrombosis. Insulin 33

46 infusion decreased plasma levels of ICAM , monocyte chemoattractant protein- 1 (MCP- 1) 262, tissue factor 263, plasminogen activator inhibitor- 1 (PAI- 1) 262, 263, and C- reactive protein 264. Insulin infusion also inhibited NFκB by stimulating IκB in mononuclear cells 262.These changes could be seen in obese subjects (who are generally insulin resistant) and were not due to insulin- induced changes in glucose as normoglycemia was maintained Effect(s) of insulin on proliferation and migration. Insulin- stimulated VSMC proliferation may be explained by activation of the MAPK signalling pathway as insulin and several other growth factors are known to activate this pathway in VSMCs 265. MAPKs phosphorylate other kinases, including regulatory transcription factors required for the expression of genes involved in proliferation. Insulin has a mitogenic effect in VSMCs 266. On the other hand, the majority of in vivo studies report that insulin does not affect 215, 267 or even decreases 268 VSMC proliferation. For example, hyperinsulinemia (10U/day), without changes in glycemia, in normal rats did not affect DNA synthesis two and four days after aortic injury 215, whereas in another study insulin (6U/day) treatment for four days after aortic injury decreased DNA synthesis in normal rats 268. The mechanism of this effect was not addressed but an effect of hypoglycemia was suggested as an explanation. However, in contrast, our laboratory has previously shown that insulin (5U/day) administration in normal rats increases VSMC proliferation after carotid balloon injury at seven days post- injury 133. The effect of insulin on VSMC migration is controversial, as studies report that insulin either stimulates 269, 270 or inhibits VSMC migration in vitro. However, the effect of insulin appears to be time- and dose- dependent; acute treatment at lower doses 271, 272 was inhibitory whereas chronic 269 or supraphysiological 270 doses were stimulatory. This may be explained by the fact that high levels of insulin stimulate IGF- 1 receptors to mediate VSMC migration 50. Chronic insulin treatment can also induce insulin resistance 274, which may prevent insulin s 34

47 inhibitory effect on VSMC migration. Insulin s inhibitory effect on VSMC migration was abolished by NOS inhibitors 272, suggesting a NO- dependent mechanism Resveratrol General effects of RSV. RSV (trans-3,5,4 -trihydroxystilbene) is a natural phytochemical found in several plant species, including grapes, and is subsequently present in high concentrations in red wine. Microbial infections, ultraviolet radiation, and chemical stressors induce RSV synthesis (reviewed in 275 ). Currently, RSV is commercially available as a dietary supplement. The presence of RSV in red wine lead to speculation that RSV may partly explain the French Paradox, which is the observation that mortality from cardiovascular disease is relatively low in France, despite high consumption of saturated fats, perhaps in relation to moderate red wine consumption. Numerous studies have been performed and have demonstrated that RSV, at least in vitro and in animal studies, has many properties that would support this hypothesis. Furthermore, from these studies, several beneficial health effects of RSV have been reported 275, including anti- oxidant activity, cancer prevention, lifespan extension, and anti- diabetic and anticardiovascular disease properties. However, very few studies were performed in humans. The cardiovascular benefits of RSV include protection against LDL oxidation, increasing high- density lipoprotein (HDL) cholesterol, inhibition of platelet aggregation and/or adhesion, and suppression of proliferation or hypertrophy of VSMCs and ECM synthesis (reviewed in 275 ), which will be discussed in the following sections. RSV has both estrogen- receptor agonist and antagonistic activity depending on the system, therefore its estrogen- like effect could also explain its cardio- protective ability as estrogen stimulates NO release 276. RSV is more widely known for its anti- cancer properties. Studies have shown inhibition of cellular events associated with cancer initiation, promotion, and progression (reviewed in 277 ). RSV is recognized as an activator of AMPK 278, 279, an energy sensor and suppressor of proliferation in many different cell 35

48 types. This suggests that RSV s anti- cancer and anti- atherosclerotic activity (i.e. suppression of VSMC proliferation) may involve AMPK activation. There are fewer studies examining the possible anti- diabetic effects of RSV, however interest has been rising in recent years. Oral RSV in Streptozotocin- induced diabetic rats resulted in a significant lowering of plasma glucose and lipid levels and stimulated glucose uptake in skeletal muscle, suggesting an insulin- like effect 280. In a very recent study, RSV increased insulin sensitivity and AMPK activity in mice fed a high- calorie diet 281, 282. We have shown that RSV also stimulates glucose uptake in L6 skeletal muscle cells in a SIRT1- and AMPK- dependent manner 279. Sirtuins are a family of histone/ protein deacetylases shown to be involved in lifespan extension following caloric restriction in many organisms, from yeast to mammals. Many studies suggest that RSV is an activator of SIRT1 (reviewed in 283 ), however this concept has been challenged as very recent studies using new biochemical assays suggest that RSV is not a direct activator of SIRT Effects of resveratrol on atherosclerosis. Generally the effect of RSV on atherosclerosis is thought to be protective, however one early study reported that RSV (0.6mg/kg/day; gavage) promoted atherosclerosis in diet- induced hypercholesterolemic rabbits without changes in plasma lipoprotein- cholesterol (VLDL, LDL, HDL) concentrations 285. In contrast, another study in hypercholesterolemic rabbits demonstrated that RSV (3mg/kg/day) given in the drinking water suppresses atherosclerosis 286, however as in the previous study, plasma lipid levels were not affected. The difference between the studies may be explained by the different doses of RSV given or the frequency of RSV administration. Subsequent reports in transgenic mouse models also showed that RSV has a protective effect against atherosclerosis. In Fukao et al 287 RSV (~1.5mg/kg/day; in food) decreased atherosclerotic plaque formation and laser- induced thrombus size in ApoE/ LDLR knockout mice, without changes in plasma lipid (total cholesterol) levels. Atherogenic lesion formation 36

49 was also decreased in ApoE knockout mice given RSV (~32mg/kg/day; in food), however in this study total- and LDL-cholesterol levels were decreased while HDL- cholesterol levels were increased in the RSV group 288. Vascular cell adhesion molecule- 1 (VCAM- 1) and ICAM- 1 levels in the atherosclerotic vessel were also decreased with RSV. Polyphenol mixtures containing low levels of RSV also inhibit atherosclerosis in ApoE knockout mice 289, 290. Plasma lipids were measured only in Norata et al 289, however no changes were found. From these studies it seems that RSV is vasculoprotective and that changes in plasma lipid levels are only observed with the highest doses of RSV Effects of resveratrol on restenosis. RSV (4mg/kg/day; intragastric) was initially shown to decrease neointimal growth after arterial injury in rabbits 291. In the same study RSV decreased VSMC proliferation in vitro, which was suggested as the mechanism to decrease neointimal formation. In a subsequent study Gu et al 292 demonstrated that RSV (10mg/kg/day; gavage) reduced neointimal growth after aortic balloon injury in rats. RSV also accelerated re- endothelialization and increased enos mrna expression in the aorta after injury; however, at a dose of 50mg/kg/day, RSV only decreased neointimal growth, indicating a dose- dependent effect. More recently, RSV (1mg/kg/day) was shown to reduce neointimal growth after carotid balloon injury in rats when administered in an emulsion system via i.p. injection 293. Furthermore, other AMPK activators (adiponectin 294, 5- aminoimidazole-4-carboxamide-1-β-d-ribofuranoside (AICAR) 295 ), have been reported to inhibit restenosis after angioplasty in animals and humans Mechanism of Resveratrol Action in the Vasculature General resveratrol signal transduction. RSV s molecular mechanism of action is not entirely clear. RSV is lipophilic and can easily cross the blood brain barrier and affect the central nervous system. RSV can enter the cell through simple diffusion 296 and more recent studies have identified several receptors for RSV 37

50 binding, including an aryl hydrocarbon receptor (AhR) 297, a sulfonylurea receptor (RSV binding blocks SUR1/K(IR)6.2 K(ATP) channels) 298, and the estrogen receptor in the vascular endothelium 299. As previously mentioned, RSV is most well known to act through intracellular targets such as SIRT1, AMPK, and MAPK (please see Figure 1.2 for a summary of RSV signalling). A major difference between insulin and RSV signalling is that PI3K regulates a large portion of insulin action whereas AMPK appears to be a key component of RSV action 281. A growing number of studies also support a role for RSV in stimulating NO production, which will be addressed in the following sections. A classical RSV signal transduction pathway has not been defined as RSV affects countless signalling molecules with its effect varying with cell type. Therefore, I have limited the discussion to the most relevant signalling molecules in the vasculature as they pertain to each of the following sections Resveratrol stimulates NO production and induces vasodilation. RSV exerts several cardiovascular protective effects by stimulating NO production. It has been reported that RSV enhances expression and activity of enos in endothelial cells 300, EPCs 292, and in vivo in the heart 301. RSV decreased inos expression in macrophages 302, however it also enhanced cytokine- induced inos expression in rat VSMCs 303. RSV has been shown to increase NO production through an estrogen receptor and calcium dependent mechanism in endothelial cells. Furthermore, RSV activates AMPK 282 and AMPK has been reported to activate enos through serine phosphorylation 306. For example, in endothelial cells, RSV induced enos activation and subsequent NO production through an AMPK- dependent mechanism 142. Another possible mediator of RSV- induced NO production is SIRT1 as RSV increased enos expression through a SIRT1- dependent manner in both cultured mouse arteries and human coronary arterial endothelial cells 307. The vasorelaxant properties of RSV are, in part, mediated by NO. For example, chronic oral administration of RSV in rats induced significant vasodilation through a NO- dependent 38

51 mechanism as L-N G -nitro-l-arginine methyl ester (L-NAME) (a NOS inhibitor) coadministration prevented the effect of RSV 308. RSV also induces vasorelaxation in the aorta of hypertensive rats by decreasing superoxide production 308 (superoxide inactivates NO). RSV has also been shown to inhibit ET- 1- induced MAPK activity through upregulation of cgmp/ PKG activity in VSMCs 309, down- regulate ET- 1 evoked protein tyrosine phosphorylation to inhibit MAPK in porcine coronary arteries 310, and induce vasorelaxation 311. Similar to insulin, RSV improves blood pressure in models of hypertension 312, however RSV treatment alone may not have an effect on systolic blood pressure in normal rats Resveratrol inhibits platelet function and thrombus formation. RSV is also an effective inhibitor of platelet adhesion and aggregation (reviewed in 275 ). RSV inhibited platelet aggregation in human whole blood 313, in platelets isolated from human plasma 314, and in experimentally induced hypercholesterolemic rabbits 314. RSV modulates platelet activity through multiple mechanisms; 1) inhibition of platelet adhesion to collagen 315, 2) inhibition of calcium influx into thrombin- stimulated platelets 314, and 3) reduction of tissue factor (a cell surface glycoprotein that initiates blood coagulation) activity in endothelial cells and monocytes 316. RSV treatment also decreased TNFα- induced VCAM- 1 expression 317, which could potentially reduce leukocyte adhesion to the vascular endothelium. Recently, RSV was shown to activate enos in platelets isolated from human plasma to stimulate NO production and inhibit platelet function 318, which could represent a source of NO after arterial injury. The mechanism of the anti- platelet effect of RSV may involve cyclooxygenase (COX) inhibition 319. The COX pathway promotes the formation of prostaglandins from arachidonic acid, which initiate VSMC contraction of blood vessels and promote fever and pain as part of the inflammatory response to injury. RSV is a competitive inhibitor of COX 320 and prevents the release of prostaglandins and thromboxanes 321, 322. Thromboxanes work with prostaglandins to promote vasoconstriction and activate platelets in blood vessels to form a thrombus. RSV 39

52 inhibits thrombus formation by inhibiting thromboxane B2/ 12- hydroxyeicosatraenoate, which is specifically responsible for platelet aggregation induction and contraction of arteries 322. RSV also reduced laser- induced thrombosis in ApoE/ LDLR knockout mice Resveratrol has anti- oxidant effects. ROS are essential in cell signalling and regulation at low concentrations, and affect gene transcription, protein synthesis, and cell function. However, an overabundance of ROS causes oxidative stress, which is one of the most important contributors to vascular damage and endothelial dysfunction. An excellent example of this oxidative stress is the production of superoxide and NO by endothelial cells, activated macrophages, and neutrophils during the inflammatory response. As previously discussed, superoxide and NO react to produce the oxidant peroxynitrite, which is involved in the development of atherosclerosis. Therefore, inhibition of vascular ROS can enhance the bioavailability of NO and protect against vascular disease. LDL oxidation is a main cause of endothelial injury. RSV protects lipids from peroxidative degradation 323 and prevents oxidized LDL uptake into the vascular wall 324. Protection of LDLs from peroxidation by RSV may be mediated through chelation of copper ions 325 and by free radical scavenging 326. Ferrylmyoglobin and peroxynitrite also oxidize LDLs and RSV has been shown to protect against their oxidative effects 327. In addition, RSV induces intracellular anti- oxidant enzymes such as glutathione reductase and superoxide dismutase to decrease ROS levels Resveratrol has anti- inflammatory effects. Numerous studies have demonstrated a significant anti- inflammatory effect of RSV. RSV s anti- inflammatory properties are partly explained by its anti- oxidant properties, which were addressed in the previous section. RSV decreases the expression of pro- inflammatory markers (monocyte chemoattractant protein- 1 (MCP-1), MIP-1, endothelin- 1 (ET-1)) and 40

53 markers for macrophages and lymphocytes in the vascular wall 289. RSV also suppresses IL- 6 gene expression and protein synthesis in macrophages 329. RSV decreases the expression of the adhesion molecules ICAM-1 and VCAM More recently, in humans, RSV decreased NFκB, p47 phox protein expression, and c- jun NH 2 - terminal kinase- 1 (JNK-1) mrna expression in mononuclear cells 331, demonstrating that RSV has a suppressive effect on inflammation as well as oxidative stress. Furthermore, the anti- inflammatory and anti- oxidant effects of RSV may both be explained, at least in part, by activation of AMPK. As an AMPK activator 282, RSV could decrease fat accumulation in insulin sensitive tissues, thereby quenching the source of inflammation Resveratrol inhibits proliferation and migration. Several studies report that RSV inhibits the proliferation of VSMCs 310, 332, 333. Mainly, RSV decreases cell proliferation by inhibiting cyclin- dependent kinases (CDKs), (CDKs regulate cell cycle progression and DNA synthesis) and activating CDK inhibitors and tumour suppressor genes (p53 and p21). For example, RSV decreases VSMC proliferation by blocking the cell cycle at the G1/ S transition 332 and increasing p53 and p21 levels 334. It appears that RSV utilizes different signalling pathways to inhibit proliferation, however it is important to remember that these pathways must converge upon common downstream regulators of the cell cycle, such as CDKs and CDK inhibitors. RSV s inhibitory effect on VSMC proliferation has been reported to be mediated by the estrogen receptor and to be NOdependent (associated increase in inos expression) 333. Moreover, ET-1- evoked VSMC proliferation and MAPK phosphorylation were inhibited by RSV through a cgmp- dependent mechanism 309. MAPK induces VSMC proliferation and contraction in blood vessels 335 and RSV has been shown to inhibit ERK1/2, JNK1, and p38 MAPK kinase phosphorylation and activity in porcine coronary arteries 310. Moreover, RSV has also been shown to inhibit cell proliferation through AMPK activation 336. Thus MAPK inhibition or AMPK activation may represent NO- 41

54 independent mechanisms to decrease cell proliferation. Cells that initiate apoptosis pathways do not undergo proliferation and thus RSV may affect proliferation indirectly, by stimulating apoptosis. Studies show that RSV induces 334, 337 or does not affect 334 VSMC apoptosis, which may depend on the concentration of RSV. In contrast to VSMCs, RSV stimulates proliferation in endothelial cells 338. Klinge et al 339 reported that RSV activated MAPK and enos through an estrogen receptor- dependent mechanism in human umbilical vein endothelial cells. RSV also increased proliferation, migration, and enos expression in EPCs 292. However, at very high concentrations, RSV induced apoptosis in endothelial cells 340. RSV also inhibits VSMC 341 migration while promoting endothelial 340 and EPC migration 342. The expression of MMPs, as previously mentioned, are essential for VSMC migration and contribute to the intimal thickening that is characteristic of restenosis 45. MMP expression is under MAPK and NFκB control. RSV inhibits MMP and MMP expression in human glioblastoma and vascular smooth muscle cells respectively. More specifically, in human VSMCs treated with TNFα, RSV decreased MMP- 9 expression through a reduction in the binding of the transcription factors NFκB and AP-1 344, suggesting that RSV may decrease VSMC migration by NFκB- mediated inhibition of MMPs. As discussed earlier, RSV activates AMPK and there is some evidence that AMPK activation decreases cell migration 345, 346, in addition to proliferation. Furthermore, maintenance of the differentiated state of VSMCs prevents their transition to a migratory phenotype and results in decreased VSMC migration. As mentioned previously, inhibition of the mtor/ S6K1 pathway by rapamycin (mtor is downstream of AMPK) has been shown to promote VSMC differentiation 34. RSV is also an inhibitor of mtor signalling 347, thus representing yet another mechanism by which RSV may decrease VSMC migration. 42

55 1.12 Summary Type 2 diabetes and obesity- associated metabolic syndrome, which are characterized by insulin resistance, are potent risk factors for atherosclerosis. Both conditions are also risk factors for restenosis after revascularization procedures used for treatment of atherosclerosis. The overall effect of insulin on the vasculature has been a point of controversy, however the prevailing effect of insulin on atherogenesis in vivo is now considered to be protective (PI3K mediated) and instead, insulin resistance is considered to be harmful (MAPK mediated) because of the unresponsiveness to insulin action, including its vascular anti- inflammatory properties. Previous studies from our laboratory support a protective role of insulin against restenosis. Furthermore, studies from other authors support that RSV may also protect against restenosis and atherosclerosis. The mechanisms underlying the protective effects of insulin or RSV are not fully understood. NO is a critical regulator of vascular homeostasis and its production has been reported to be stimulated by both insulin and RSV in numerous studies, thus representing a likely target of insulin or RSV action Rationale and Significance of the Studies Atherosclerotic cardiovascular disease is a leading cause of morbidity and mortality worldwide 1 and revascularization procedures, such as angioplasty and stenting, often result in restenosis. Type 2 diabetes has also become a global epidemic 348 and macrovascular complications are the major cause of mortality in diabetic patients 349. Type 2 diabetes and the related metabolic syndrome, a cluster of risk factors for both atherosclerosis and type 2 diabetes, are both characterized by insulin resistance, and it is now well established that insulin resistance itself is a potent risk factor for atherosclerosis and restenosis after percutaneous interventions 124. Although drug- eluting stents have reduced the rate of restenosis in non- diabetic individuals, results are still worse in persons with type 2 diabetes

56 In vitro, insulin has mitogenic effects on VSMCs 265, however previous studies in our laboratory 133, 134, as well as by others 222, have shown that insulin decreases neointimal growth after rat carotid balloon injury, a model of restenosis, thereby demonstrating a protective effect of insulin on the vasculature. NO production is important for vascular function and has been shown to diminish neointimal growth after injury 105, 149. Furthermore, insulin has protective effects on the endothelium by stimulating enos and NO production 235, 236. NOS inhibition has also been reported to abolish the effect of insulin to inhibit VSMC migration in vitro 272. However, whether the mechanism of action of insulin to limit neointimal growth involves NO has not been elucidated. Another promising compound to prevent restenosis is RSV, an AMPK activator with both anti- CVD and anti- diabetic properties. RSV also decreases neointimal formation after arterial injury , similar to insulin 133, 134, and has been shown to enhance the expression of enos in injured aortas 292, however NO was not demonstrated to be causally linked to the inhibition of neointimal growth. Moreover, these studies did not focus on investigating the mechanism or signalling molecules involved in RSV action in vivo. In addition, reports suggest that RSV action is Akt- independent, suggesting that RSV would remain effective in a state of insulin resistance, thus providing an advantage over insulin. Therefore, it is important that the mechanism of action of RSV to decrease neointimal formation is addressed. The three studies in this thesis were performed to specifically address the mechanism of action of insulin or RSV to protect against restenosis, and whether NO production mediates these effects using the NOS inhibitor L-NAME. To do this, we utilized an in vivo experimental model of restenosis, carotid balloon angioplasty, in normal rats in all three studies. Although the rat model has minimal inflammatory and thrombotic components when compared with human restenosis 21, and does not allow the study of long- term vascular remodelling, it is the best characterized model of arterial injury and restenosis 17 and facilitates the investigation of the 44

57 mechanism of insulin s and RSV s effect on restenosis. These studies may provide initial evidence for the clinical potential of insulin or RSV or their downstream mechanisms in the prevention of restenosis after angioplasty General Hypothesis Insulin and RSV have protective effects against restenosis after arterial injury and their protective effects are mediated by NO Specific Aims This thesis consists of three studies that examined the protective effects of insulin or RSV in a model of restenosis in rats, and the role of NO as a possible mediator of these effects. Study 1. To investigate the effects of insulin treatment on neointimal growth after arterial injury, independent of glucose- lowering and oral glucose or sucrose. Study 2. To examine the consequences of NOS inhibition, with the in vivo inhibitor L-NAME, on the effect of insulin to inhibit neointimal formation. Study 3. To determine the effect of RSV on the processes involved in neointimal formation after arterial injury, and to investigate the consequences of NOS inhibition on these effects of RSV. 45

58 Figure 1.1. Signal transduction by the insulin receptor. Insulin binding to its receptor results in tyrosine phosphorylation (-Tyr-PO32) of the insulin receptor, IRS (Insulin Receptor Substrates)1-4 and Shc (Src- homology collagen- like). Phosphatidyl Inositol-3 Kinase (PI3K) and GRB bind to phosphorylated IRS, which activate the PI3K and MAPK (Mitogen Activated Protein Kinase) pathways. GRB also binds to phosphorylated Shc. Akt is downstream of PI3K. 46

59 Figure 1.2. Proposed signal transduction by resveratrol. Resveratrol action in the vasculature may involve binding to the estrogen receptor to stimulate nitric oxide (NO) production 304. The increase in NO production is possibly mediated by activation of SIRT1 307 and/or 5 - AMP- Activated Protein Kinase (AMPK) 142. Furthermore, activation of AMPK may lead to inhibition of Mitogen Activated Protein Kinase (MAPK) pathway 295, however there are also reports that resveratrol can stimulate MAPK to improve endothelial function through an estrogen receptor- dependent mechanism

60 2 General Methods 2.1 Animals For all studies, normal male Sprague- Dawley rats (Charles River, Quebec, Canada) weighing g were used for experiments. Male rats were studied since estrogen has been shown to attenuate intima thickening following carotid injury in rats and consequently the vessel response to injury is greater in males 350. The rats were housed in the University of Toronto s Department of Comparative Medicine. They were exposed to a 12h light/dark cycle. The rats used in Study 1 were fed a normal rodent diet (Dyets Inc ; 18% Protein diet with 70% carbohydrate and 12% fat; powder form; Dyets Inc., Bethlehem, PA) to allow comparison with the study by Kim et al 133. The rats used in Studies 2 and 3 were also fed a normal rodent diet that differed slightly from Study 1 (Teklad Global 2018; 21% Protein Diet with 64% carbohydrate and 14% fat; pellet form; Harland Teklad Global Diets, Madison, WI). The animal care committee of the University of Toronto approved all procedures. 2.2 Surgical Procedures Subcutaneous Implant Procedure. For Studies 1 and 2 rats were implanted with either insulin- releasing implants (5U/day for Study 1 and 3U/day for Study 2; 10-14% bovine insulin, 86-90% palmitic acid; 26 ± 2mg/ capsule, LinShin Canada Inc.) or blank implants (100% palmitic acid; 26 ± 2mg/ capsule) two days (Study 1) or three days (Studies 2 and 3) before balloon angioplasty using isoflurane gas. 48

61 The implants were placed at the back of the rat s neck using a hypodermic needle. Rats receiving insulin capsules were also given subcutaneous (s.c.) 0.25ml glucagon and 2.5ml 50% glucose solution i.p. to prevent hypoglycemia caused by tiny pieces that broke away from the capsule during implantation. Previous reports have demonstrated sustained release and action of the insulin implants (>40 days) after implantation 351. The blood glucose of the rats was monitored frequently (every 2 hours) throughout the day using a glucometer (Glucometer Elite, Bayer Inc.). The insulin- treated rats were also provided with 40% glucose in their drinking water to maintain normoglycemia. Analgesic (Buprenorphine) was given at the end of each surgical procedure Intraperitoneal Catheterization for Glucose Infusion. In Study 1, where the rats received high- dose insulin (5U/day), an incision was made on the dorsal surface of the neck and also through the abdominal muscles on the mid- lateral side of the rat abdomen after the carotid balloon injury procedure. Sterile tubing (polyethylene- 50 (PE- 50); Cay Adams, Boston, MA) extended with a segment of silastic tubing (length of 2cm, internal diameter of 0.02inches; Dow Corning, Midland, MI) was tunneled subcutaneously from the neck incision, along the side, and through the abdominal muscles into the peritoneal cavity of the rat. The tubing was secured with a suture in the abdominal muscle wall and subcutaneous tissue, and exteriorized at the back of the neck. Saline was flushed through the line to ensure proper infusion and the incisions were closed. The insulin- treated rats received variable infusions of 25% glucose solution (50% dextrose diluted in saline (0.9% sodium chloride) to maintain euglycemia, where the infusion rate was adjusted daily based on blood glucose levels. Control rats received a saline infusion (0.9% sodium chloride) Osmotic Pump Implant Procedure. In Studies 2 and 3 rats were implanted with osmotic pumps for RSV (4mg/kg/day dissolved in polyethylene glycol- 300 and dimethyl sulfoxide (9:1)) or L-NAME (2mg/kg/day or 10mg/kg/day dissolved in saline) treatments three days before balloon angioplasty using 49

62 isoflurane gas. Osmotic pumps allow continuous delivery of RSV or L-NAME. An incision was made on the dorsal side of the rat, at a different site from the capsule implant, and a pocket was made s.c. with a blunt dissector to insert the osmotic pump, and the wound was closed Carotid Balloon Injury. Arterial injury was performed by balloon angioplasty only on rats with blood glucose levels greater than 4.0mM. Injury to the left common carotid artery (LCCA) of the rat was induced by means of an inflated balloon catheter. Rats were anesthetized with a ketamine: xylazine: acepromazine cocktail (87:1.7:0.4mg/ml, 1ul/g body weight) (Study 1) or isoflurane (Studies 2 and 3). A midline incision of the neck was made to expose the left external carotid artery and rats underwent balloon injury where a 2F Fogarty balloon catheter (Baxter) was introduced into the LCCA and then inflated (30µl of saline) and withdrawn while rotating to denude the vessel of endothelium. This was performed four times. Then the LCAA was ligated and the wound was closed Aortic Balloon Injury. Injury to the aorta of the rat was induced by means of an inflated balloon catheter. In brief, after undergoing isoflurane anesthesia, rats underwent balloon injury of the thoracic aorta. A 2F Fogarty balloon catheter (Baxter) was introduced via the LCCA into the aorta and then inflated and withdrawn three times. Then the LCAA was ligated and the wound was closed Vessel Cannulation. Three days before the hyperinsulinemic- euglycemic clamp, rats were anesthetized with a ketamine: xylazine: acepromazine cocktail (87:1.7:0.4mg/ml, 1ul/g body weight) (Study 1) or isoflurane (Study 3). Indwelling catheters were inserted into the right internal jugular vein for infusions and the LCCA for sampling. Polyethylene catheters (PE-50; Cay Adams, Boston, MA), each extended with a segment of silastic tubing (length of 3cm, internal diameter of 0.02inches; Dow Corning, Midland, MI), were used. The venous catheter was extended to the level of the 50

63 right atrium and the arterial catheter was advanced to the level of the aortic arch. Both catheters were tunneled subcutaneously and exteriorized. The catheters were filled with a mixture of saline and heparin (1000 U/ml) to maintain patency and were closed at the end with a metal pin. The rats recovered for three days before experiments. 2.3 Blood and Vessel Sample Collection After arterial injury rats were sacrificed at the appropriate time- point in either the fed or fasted state in order to obtain blood samples in different metabolic states. Blood glucose was monitored throughout the treatment using a glucometer and measured at the same time of the day (~10a.m.). Random fed glucose levels were also measured to ensure that normoglycemia was maintained throughout the fed state. These values were converted to plasma glucose assuming a normal hematocrit. To obtain fasting blood glucose levels, rats were fasted once for ~16 h (tap water/ 40% glucose solution/ 10% sucrose solution was given without food). Under general anesthesia, blood samples were collected via cardiac puncture for insulin, triglyceride (TG), and FFA measurements. Following blood sample collections, rats were sacrificed via anesthetic overdose Fixed Vessel Collection. To obtain fixed vessels for morphometric analysis and staining, rats were subjected to retrograde perfusion through the abdominal aorta with Krebs- Ringer solution and then with 10% buffered formalin at the physiologic pressure ( mmHg). A 2cm length vessel was excised from the middle of each of the left and right common carotid arteries to ensure that the vessel cross- sections used for analysis had the most even vessel damage and were within the arterial region of the vessel which never completely re- endothelializes. Immediately following removal, the vessel was immersed in 10% buffered formalin for 24 hours and then transferred to phosphate buffered saline (PBS). All vessels were embedded in paraffin and divided into two blocks by cutting the midsection of the artery. Cross- sections were taken from the midsection. 51

64 2.3.2 Frozen Vessel Collection. To obtain fresh vessel samples for Western blot, a 2cm length vessel was excised from the middle of each of the left and right common carotid arteries, rinsed in saline to remove excess blood and immediately frozen in liquid nitrogen and stored at -80ºC. 2.4 Morphometric Measurements and Calculations For all analyses in this thesis, a blinded coding system was used to avoid experimenter bias. Intimal and medial cross- sectional areas were measured on hematoxylin and eosin stained slides (prepared in the Histology/ Pathology Core laboratory at The Centre for Modeling Human Disease (Toronto Centre for Phenogenomics, Toronto, ON, Canada)) and analyzed using a computer- assisted morphometric system (Simple PCI, Compix Inc.). Images of the vessels were obtained using a Nikon E600 microscope (Nikon, Mississauga, ON, Canada), digitized using a camera (model C NRB; Hamamatsu, Inc.). Two separate cross- sections were measured on each carotid artery and averaged. This average value was used to represent each individual rat. Intimal area was measured as the area encompassed by the internal elastic lamina minus the lumen area (in this case lumen area was determined by tracing around the inside edge of the vessel and calculating the area inside). Medial area was measured as the area encompassed by the external elastic lamina minus the area encompassed by the internal elastic lamina (including lumen area). Lumen area was calculated by subtracting the intimal area and the medial area from the total area encompassed by the external elastic lamina (in this case total area was calculated using the external elastic lamina perimeter). This method of obtaining lumen area avoids artifacts due to paraffin embedding. 52

65 2.5 Histomorphometry/Matrix Staining Cell Migration Assay. Cell migration into the intima was measured four days after injury and intimal cells on the lumen surface were immunostained with an antibody against histone H1 (MAB1276; Chemicon) which does not permeate the internal elastic lamina and as a result stains only intimal cell nuclei 45, 82. This migration assay takes advantage of the fact that the first cells appear in the intima three to four days after injury 82 and do not complete a round of replication until 24 hours later. An approximately 2cm centre section of the LCCA was cut lengthwise and pinned open on a Teflon card with the lumen side up. The arteries were rinsed with PBS and bathed in a solution of 0.3% hydrogen peroxide and cold methanol for 30 minutes to block endogenous peroxidase. Non- specific protein binding was blocked with 2% normal horse serum (diluted with 1% PBS: bovine serum albumin (BSA)) for 30 minutes. The arteries were then incubated with the primary antibody histone H1 monoclonal antibody (MAB1276, mouse anti- human nuclei and chromosomes, Chemicon International Inc) (diluted 1:100 in 1% PBS/BSA) for 30 minutes. This was followed by incubation with the secondary antibody, 1% biotinylated horse anti- mouse IgG (diluted in 1% PBS/BSA), for 30 minutes. Peroxidase labeling was carried out using an avidinbiotin complex (ABC Elite, Vector Laboratories) during a 30 minute incubation. Arteries were incubated in Tris (6.1g/L of distilled water, ph 7.6) for 5 minutes and then stained using 0.05% 3,3 -diabminobenzidine (DAB) in 0.05mmol/L Tris-HCl (ph 7.6) with 50µl of 30% hydrogen peroxide for 5 minutes. All steps were followed by PBS rinses in triplicate. At the end of the assay the arteries were mounted on glass slides and coverslips were applied with a 50:50 mixture of PBS: glycerol. The artery was visualized with light microscopy and the entire luminal surface of the vessel was counted with a minimum of 12 fields taken at a magnification of 200X. Each nucleus that was oriented parallel to the long axis of the vessel was counted, thus discounting medial cells which are perpendicular. 53

66 2.5.2 BrdU- Labelling. To label proliferating neointimal cells, rats were injected i.p. with three doses of 25 mg/kg 5-bromo-2 -deoxyuridine (BrdU; Sigma) at 17, 9, and 1 hour before sacrifice. The LCCAs embedded in paraffin were sliced into 4µm thick sections and mounted on slides in the Histology/ Pathology Core laboratory at The Centre for Modeling Human Disease (Toronto Centre for Phenogenomics, Toronto, ON, Canada). The vessel cross- sections were deparaffinized in xylene washes and rehydrated in 100% ethanol. To block endogenous peroxidase activity, the slides were immersed in a solution of 3% hydrogen peroxide dissolved in cold methanol. After repeated washes in distilled water, proteinase K (DAKO Corporation) was added to the cross- sections to digest proteins and expose DNA. Again the slides were rinsed in distilled water and immersed in 4N hydrochloric acid to denature DNA so that the antibodies could bind to the BrdU incorporated into the DNA. The slides were then rinsed in distilled water, followed by PBS for 10 minutes, and then incubated in the primary antibody (rat anti- BrdU antibody diluted in 1% normal horse serum and PBS solution, 1:40 dilution; DAKO Corporation) at 37ºC in a humidified chamber for 60 minutes to stain BrdU incorporated into the DNA of proliferating intimal VSMCs. Following the incubation, the slides were washed with PBS and incubated with the secondary antibody (biotinylated anti- MoIgG HAM antibody diluted in PBS, 1:40 dilution; Vector Laboratories) at 37ºC in a humidified chamber for 30 minutes. The slides were rinsed with PBS and incubated in the tertiary antibody (ABC Elite diluted in PBS, 1 drop/ 2.5ml PBS; Vector Laboratories) at 37ºC in a humidified chamber for 30 minutes. Subsequently, the slides were washed with PBS and then a Tris solution (50mM ph 7.6 in distilled water) for 5 minutes, then immersed in DAB (Sigma) solution for 10 minutes. Immediately following, the reaction was stopped by immersing the slides in 70% ethanol for 5 minutes. The slides were again washed with PBS and counterstained with hematoxylin (hematoxylin diluted 1: 2 in distilled water; DAKO Corporation) solution for 3 minutes. The slides were dipped in acid 54

67 alcohol (10ml distilled water in 190ml 95% ethanol plus 8 drops of glacial acetic acid) followed by development with Scott s Tap Water for 5 minutes. The slides were rinsed with PBS and dehydrated with successive washes of increasing concentrations of ethanol. Xylene was applied to the slides for 5 minutes (2 times) and then slides were mounted and coverslipped with Permount. The total number of BrdU- positive stained cells was counted in the intima of the vessel cross- sections under 400X magnification Terminal dutp nick- end labeling (TUNEL) Assay. The rate of cell apoptosis in the intima was measured using terminal dutp nick- end labeling (TUNEL) on cross- sections. The staining was prepared in the Histology/ Pathology Core laboratory at The Centre for Modeling Human Disease (Toronto Centre for Phenogenomics, Toronto, ON, Canada). The vessel cross- sections were deparaffinized in xylene and rehydrated with decreasing concentrations of ethanol, followed by washes with distilled water and PBS. Proteinase K (10µg/ml in PBS; DAKO Corporation) was added to digest proteins and expose DNA. The vessels were then washed with distilled water and PBS, followed by incubation in a solution of 3% hydrogen peroxide dissolved in methanol for 30 minutes to block endogenous peroxidase activity. The slides were again washed in PBS, pre- incubated with 1X One- Phor- All Buffer (10X stock diluted to 1X with 0.1% TritonX- 100 in distilled water; Pharmacia) for 10 minutes, and then incubated in TdT solution (10µl 10X One- Phor- All Buffer + 6µl dilute biotin- 16- d UTP + 1µl 100 µ M d ATP + 1µl TdT enzyme + 82µl 0.1% TritonX- 100 in distilled water) in a humidified chamber for 2 hours at 37ºC. The slides were washed with PBS and incubated in the ABC Complex solution (Vectastain) for 90 minutes at 37ºC, followed by a wash with a solution of 0.1% TritonX- 100 in PBS. The slides were immersed in DAB (Sigma) solution for 15 minutes to develop the colour and counterstained with hematoxylin (hematoxylin diluted 1:2 in distilled water; DAKO Corporation) for 20 seconds, then washed with water and dipped in acid alcohol (10ml distilled water in 190ml 95% ethanol 55

68 plus 8 drops of glacial acetic acid) followed by development with Scott s Tap Water for 1 minute. The slides were rinsed with PBS and dehydrated with successive washes of increasing concentrations of ethanol. Xylene was applied for 1 minute (3 times) and then slides were mounted and coverslipped with Permount. The total number of TUNEL- positive stained cells was counted in the intima of the vessel cross- sections under 400X magnification Extracellular Matrix Staining with Movat s Pentachrome and Picrosirius Red. Cross- sections taken 14 and 28 days after injury were also stained with Movat s pentachrome or picrosirius red (PSR) dye for elastin and collagen respectively. PSR stains collagen fibers red and Movat s pentachrome stains elastin fibers deep purple. The Movat s Pentachrome and PSR staining was prepared in the Histology/ Pathology Core laboratory at The Centre for Modeling Human Disease (Toronto Centre for Phenogenomics, Toronto, ON, Canada). For the Movat s Pentachrome stain, the vessel cross- sections were deparaffinized in xylene washes and rehydrated in 95% ethanol. The slides were immersed in Bouin s solution for 1 hour at 60ºC, washed well under running water, rinsed in 0.5% acetic acid, and then stained with alcian blue for 5 minutes, followed by an additional rinse in 0.5% acetic acid. The slides were then washed in water and stained with Working Elastin stain for 8 minutes. The slides were incubated in 3% sod thiosulfate for 1 minute and then washed in distilled water. Biebrich Scarlet- Acid Fuchsin was used to stain the slides for 90 seconds followed by another wash under running water. 0.5% acetic acid was used to rinse the slides with subsequent differentiation in phosphotungstic acid for 5 minutes and an addition rinse with 0.5% acetic acid and then absolute ethanol. The slides were incubated in saffron stain for 5 minutes and followed by successive washes of absolute ethanol. Xylene was applied to the slides for 5 minutes (3 times) and then slides were mounted and coverslipped with Permount. 56

69 The protocol for PSR staining was similar to Movat s Pentachrome with the following changes; after the deparaffinization and rehydration steps the slides were stained in Celestine blue for 5 minutes and then washed in water and stained in Harris haematoxylin for 5 minutes. The slides were again rinsed with water and dipped in acid alcohol (5 dips). Scott s Tap Water was used to develop the slides for 1 minute. The slides were washed under running water and stained with PSR for 30 minutes and blotted dry, followed by subsequent dehydration, xylene application, and mounting. To quantify the degree of collagen or elastin stain in the cross- sections of arteries, the colour/ intensity of the positively stained areas in the media and intima of the vessel was analyzed using a computer laboratory imaging system (NIS Elements BR 3.0, Nikon). A selected intensity of the stain was set as positive staining and used as a threshold. The software uses this threshold to identify the area of the vessel stained positively within a defined total area. The amount of collagen or elastin positively stained area was then expressed as a percentage of total area measured Evan s Blue Staining. To measure the re- endothelialized area at 14 days after aortic balloon injury, Evan s Blue dye solution was introduced into the rat circulation via the abdominal aorta. Evan s Blue binds to negatively charged albumin and circulates throughout the vasculature. While in circulation, this dye penetrates any vessel without an intact endothelial cell layer and stains that area blue, leaving the intact endothelium unstained (remains white). 1.0ml of 2.5% Evans Blue solution (diluted in distilled water) was injected with a syringe into rats via the abdominal aorta. Ten minutes after the injection, rats were sacrificed via anesthetic overdose and the aorta was perfusion- fixed for 15 minutes with 4% buffered formalin at physiologic pressure ( mmHg). To observe Evans Blue uptake macroscopically, the thoracic aorta was dissected 57

70 out, opened longitudinally, and an image was captured with a digital camera. The reendothelialized area was defined as the area not stained with Evans Blue dye. To quantify the area stained with Evans Blue on the longitudinal sections of the aorta, the colour/ intensity of the stained area on the luminal surface of the vessel was analyzed using a computer laboratory imaging system (NIS Elements BR 3.0, Nikon). A selected intensity of the stain was set as positive staining and used as a threshold. The software uses this threshold to identify the area of the vessel stained positively within a defined total area. The amount of positively stained area was then expressed as a percentage of total area measured. Cross- sectional areas were measured on hematoxylin and eosin stained slides from unopened sections of these aortas as described in the Morphometric Measurements and Calculations section Vessel Homogenization The frozen vessels were pulverized using a mortar and pestle in liquid nitrogen. The powdered vessel was placed into 150µl of cell lysis buffer (1% sodium dodecyl sulfate (SDS), 1mM phenylmethylsulfonyl fluoride, and 10 mg/ml leupeptin in 50 mm Tris buffer, ph 7.6). The sample was centrifuged (13000 rpm, 10 minutes, RT) and the supernatant was aliquoted for either protein assay or Western blot. The sample for Western blot was diluted 1: 1 with electrophoresis sample buffer (4% SDS, 20% glycerol, 10% 2-mercaptoethanol, 0.004% bromophenol blue, M Tris HCl). 2.7 Protein Assay The Biorad Detergent- Compatible Protein Assay (Biorad, ON, Canada) was used to measure the protein concentration of each sample with BSA used as a standard. The assay is a colorimetric assay for protein concentration following detergent solubilization. The aliquoted samples for protein assay were thawed, vortexed, and kept on ice. Then the samples were diluted (1: 3) with distilled water (in this case only 5µl of sample was needed) in a glass tube. Stock 58

71 BSA (1 mg/ml) was diluted accordingly with cell lysis buffer to prepare a standard curve ranging from 0 to 1 mg/ml. Then 20 µl of these BSA standard samples were transferred to glass tubes. 100 µl of the Reagent A solution (20 µl of Reagent S/ 1 ml of Reagent A) was added to each test tube and then vortexed, followed by the addition of 800 µl of Reagent B and vortexing. After 15 minutes samples were transferred to cuvettes and absorbance was read at 750 nm using a spectrophotometer. 2.8 Western Blot Analysis Carotids were removed seven days after injury, homogenized, and processed as previously described in section 2.6. Protein samples of the appropriate concentration (25µg was used for all antibodies except total enos for which 60µg was used) were thawed, vortexed, and then subjected to sodium dodecyl sulfate- polyacrylamide gel electrophoresis (SDS-PAGE) electrophoresis (10% polyacrylamide gel) for 90 minutes. Following electrophoresis separation, protein was transferred to polyvinylidene difluoride (PVDF) membranes in transfer buffer on ice for 90 minutes. The membranes were then blocked for 1 hour at room temperature in Trisbuffered saline- Tween (TBST) containing 5% non-fat dried milk, ph 7.4, followed by overnight incubation at 4ºC with the appropriate primary antibody diluted in the blocking solution (anti- SM-MHC (Santa Cruz Biotechnology sc-6956 (1:1000)), anti- SM22α (Santa Cruz Biotechnology sc (1:500)), anti- SMC alpha- actin (Sigma A2547 (1:1000)), anti- total enos antibody (Santa Cruz Biotechnology sc-654 (1:500)), and anti- alpha tubulin (Cell Signalling Technology #2144 (1:1000)) antibodies, all specific for rat). After three washes with TBST (20 minutes each) to remove excess primary antibody, membranes were incubated with the secondary antibody horseradish peroxidase- conjugated goat anti- rabbit IgG (Santa Cruz sc (1:2000)) for 1 hour at room temperature. The membranes were then washed three times (5 minutes each) with distilled water and developed using enhanced chemiluminescence (Thermo 59

72 Scientific) and exposed to x-ray film. The bands obtained from immunoblotting were quantified by densitometry using Image J (National Institute of Health Research). 2.9 Flow Cytometry Rat blood was obtained at three and seven days after carotid injury in separate rats by cardiac puncture. Flow cytometry was performed as in Fadini et al 115 by our collaborator Dr. M.R. Ward in Dr. D.J. Stewart s laboratory at St. Michael s Hospital (Toronto, ON). Blood was drawn into Sodium Citrate (9:1 v/v) Vacutainer tubes (BD Biosciences). For each staining sample, 200 µl of blood was mixed with 1.0 ml RBC lysis buffer (ebioscience) and incubated for 10 min at room temperature in the dark. Tubes were centrifuged and the supernatant discarded. The remaining cells (pellet) were suspended in 100 µl of PBE staining buffer, and appropriate primary antibody (mouse anti-rat VEGFR2 (Novus Biologicals), rabbit anti-rat c-kit (Neuromics), and rabbit anti-rat sca-1(novus Biologicals) was incubated with the cells at room temperature in the dark for 30 minutes. Following washing (2x 1.0 ml PBE buffer) and centrifugation, cells were exposed to secondary antibody (FITC donkey anti-rabbit IgG) or Straptavidin-APC for 30 minutes in the dark. Following two additional wash and centrifugations steps, cells were suspended in PBS. The number of gated (defined) cells positive for the three markers was quantified by flow cytometry (Beckman Coulter Cytomics FC500) using FL1 (FITC for c-kit and sca-1) and FL4 (for VEGFR2) filters Hyperinsulinemic- Euglycemic Clamp The rats were fasted overnight (~5pm to 9am). At -20 minutes and immediately before the infusion (Basal Period), blood samples were taken for measurements of plasma glucose and insulin. At the onset of the clamp, insulin (Humulin R, 5mU/kg.min) and glucose (variable rate) were infused through the jugular vein for two hours to maintain euglycemia while blood samples were taken from the LCAA for plasma glucose measurements. The total volume of blood 60

73 withdrawn was ~4 ml. After plasma separation, red blood cells were diluted at 1: 1 in heparinized saline (4U/ml) and re- infused into the rats. At the end of the experiments, the rats were anesthetized with intravenous administration of ketamine: xylazine: acepromazine cocktail (87:1.7:0.4 mg/ml, 1ul/g body weight) Blood Pressure Analysis Systolic blood pressure was measured indirectly from the tail via the cuff technique, using the Visitech BP-2000 Blood Pressure Analysis System. This system determines the cuff pressure at which changes in blood flow occur during occlusion or release of the cuff 352. Each rat was placed in a restrainer on the heated platform (37ºC) with the blood pressure cuff placed around the base of the tail. The restrainer was covered with a towel to avoid distraction for the rat. This was followed by a rest period of approximately 15 minutes to lessen the effects of stress. Then 20 preliminary systolic blood pressure measurements were taken but not recorded to allow the rat to become accustomed to the increasing pressure of the tail cuff. Following the preliminary measurements, 10 systolic blood pressure measurements were recorded and the average was taken as the final value. Systolic blood pressure was measured before treatment and throughout at 4, 14, and 28 days after arterial injury Metabolic Parameters Plasma Insulin. Plasma insulin levels were determined by radioimmunoassay (RIA) using kits specific for rat insulin (Linco Research Inc, St. Charles, MO). The rat insulin RIA is a double antibody radioimmunoassay. Insulin in the sample competes with a fixed amount of 125 I- labeled insulin for binding sites on the specific antibodies. Bound and free insulin were separated by the addition of a second antibody immunoabsorbent followed by centrifugation and aspiration of the supernatant. The radioactivity of the pellet was then measured and was inversely proportional to the amount of insulin in the sample. A standard curve was first constructed using insulin 61

74 standards with known concentrations (0.1, 0.2, 0.5, 1.0, 2.0, 5.0, 10.0 ng/ml) in duplicate. After samples were pipetted into appropriate tubes, 125 I-insulin was added (100 µl to all tubes) followed by the rat insulin antibody (100 µl to all tubes). The tubes were then vortexed to ensure mixing and were incubated overnight (18-24 hrs) at 4ºC. On the next day, 1.0 ml of precipitating reagent was added to all tubes followed by vortexing and incubating for 20 minutes at 4ºC. The tubes were then centrifuged at 2500 rpm for 40 minutes. The supernatant was aspirated and the radioactivity in the pellet was counted for 4 minutes in a gamma counter (Beckman Instruments, Fullerton, CA, USA). The counts (B) for each of the standards and unknown were expressed as a percentage of the mean counts of the 0- standard (Bo). % Activity Bound = B (Standard or sample) x 100% Bo The %B/Bo for each standard was plotted on the y- axis and the known concentration of the standard on the x- axis on log-log graph paper and a standard curve was constructed. The concentration of the unknown samples was determined by interpolation of the standard curve. The coefficient of interassay variation determined on reference plasma was less than 10% Plasma Free Fatty Acid. Plasma levels of FFA were analyzed using a colorimetric kit under enzymatic reaction (Wako Industrials, Neuss, Germany). The method relies upon the acylation of coenzyme A by the fatty acids in the presence of added acyl- CoA synthetases (ACS). The acyl- CoA produced is oxidized by adding acyl- CoA oxidase (ACOD) which generates hydrogen peroxide. Hydrogen peroxide, in the presence of peroxidase (POD) allows the oxidative condensation of 3-methyl-Nethyl-N-(B-hydroxyethyl)-aniline (MEHA) with 4- aminophenazone to form the final reaction product, which is a purple coloured adduct. This can be measured colorimetrically at 550 nm. The results are correct to within 1.1%. The reactions of this assay are listed below: 62

75 ACS Free fatty acids + ATP + CoA Acyl-CoA + AMP + PPi ACOD Acyl-CoA + O , 3-trans-Enoyl- CoA + H 2 O 2 POD 2H 2 O aminophenazone + MEHA Final Reaction Product + 3H Plasma Triglyceride. Plasma TG levels were measured using a colorimetric enzymatic kit from Roche Diagnostics GmbH (Mannheim, Germany) corrected for free glycerol concentration. The preliminary reaction removes free glycerol from the sample prior to the hydrolysis of TG. This is followed by hydrolysis of the TG to glycerol by using a lipase. Free and released glycerol is detected by an enzymatic colorimetric reaction utilizing glycerol kinase and glycerol 3- phosphate oxidase. The reactions were as follows: Preliminary Reaction: Glycerol Kinase Free glycerol + ATP Glycerol 3-phosphate + ADP Glycerol 3-phosphate oxidase Glycerol 3-phosphate + O Dihydroxyacetonephosphate + H 2 O 2 Peroxidase H 2 O chlorophenol oxidation product 63

76 Assay Reaction: Lipase TG + 3H 2 O Glycerol + Fatty Acids Glycerol Kinase Glycerol + ATP Glycerol 3-phosphate + ADP Glycerol 3- phosphate oxidase Glycerol 3-phosphate + O Dihydroxyacetone phosphate + H 2 O 2 Peroxidase H 2 O chlorophenazone + 4-chlorophenol (p-benzoquilone)-monoiminol-phenazone (4-PBMP) + 2H 2 O + HCl The absorbance of 4-PBMP at 500 nm determines the amount of glycerol liberated from TG. The coefficient of variation for this assay was ~ 1.4% Plasma Glucose Determination. Plasma glucose concentrations were measured by the glucose oxidase methods using a Beckman Glucose Analyzer II (Beckman, Fullerton, USA). A 10 µl sample of plasma containing D- glucose is pipetted into a solution containing oxygen and glucose oxidase. The glucose reacts with oxygen in the following reaction catalyzed by glucose oxidase: D-glucose + O 2 + H 2 O gluconic acid + H 2 O 2 In the reaction, oxygen is used at the same rate of glucose to form gluconic acid. A polarographic oxygen sensor is used to detect oxygen consumption, which is directly proportional to the glucose concentration in the sample. Results are obtainable within 30 seconds following sample addition. Plasma sample was re-analyzed until repeated measurements were within a difference of 3 mg/dl. The analyzer was calibrated before use and frequently during the 64

77 experiment with the 150/50 glucose/urea nitrogen standard (Beckman, Fullerton, USA) that accompanied each kit Statistical Analysis Values are expressed as mean ± SEM. Values are expressed as mean ± SEM. Statistical analysis was performed using analysis of variance (ANOVA) to test for significant differences between groups and the post hoc comparisons were performed with Tukey s t- test. Statistical calculations were performed using SAS software (Statistical Analysis System, Cary, NC). Significance was accepted at P<

78 3 Study 1 Insulin Inhibits and Oral Sucrose Increases Neointimal Growth After Arterial Injury in Rats Modified from: Breen DM, Dhaliwall JK, Chan KK, Guo J, Lam L, Bendeck MP, Giacca A. Insulin inhibits and oral sucrose increases neointimal growth after arterial injury in rats. Journal of Vascular Research. 2010; 47: Permission to reproduce portions of the above manuscript has been obtained from the copyright owner, S. Karger AG, Basel. 66

79 3.1 Abstract Background/Aims: In our previous studies, rats on insulin treatment (5U/day) and oral glucose to avoid hypoglycemia had reduced neointimal growth after arterial injury. However, plasma glucose in the insulin- treated rats was lower than normal and the effect of oral glucose remained undetermined. In this study the effects of insulin, in the presence of normal plasma glucose levels, and oral glucose or sucrose were investigated in the same model. Methods: Rats were divided into six groups: 1) control implants and tap water; 2) insulin implants (5U/day) and oral glucose+i.p. glucose to avoid any glucose lowering; 3) insulin implants (4U/day) and oral glucose; 4) insulin implants (4U/day) and oral sucrose; 5) control implants and oral glucose; and 6) control implants and oral sucrose. Results: Insulin treatment at both doses reduced neointimal area (P<0.001) 14 days after injury in rats receiving oral glucose but not in those receiving oral sucrose. Oral glucose, without insulin, had no effect on neointimal formation whereas oral sucrose increased neointimal growth (P<0.05). Oral sucrose (P<0.05) but not oral glucose decreased insulin sensitivity measured with hyperinsulinemic clamps. Conclusions: (1) Insulin decreases neointimal growth after arterial injury independent of glucose- lowering or oral glucose administration and (2) oral sucrose per se affects neointimal growth. 67

80 3.2 Introduction Metabolic syndrome is a common disorder characterized by the development of a cluster of abnormalities which include insulin resistance, compensatory hyperinsulinemia, glucose intolerance (type 2 diabetes, impaired glucose tolerance, or impaired fasting glucose), central obesity, dyslipidemia, hypertension, and microalbuminuria 353. The metabolic syndrome affects up to 25% of the population 353 and is associated with an increased risk of atherosclerosis and restenosis 354 in part because of insulin resistance/ hyperinsulinemia per se 124, 188. In the metabolic syndrome, hyperinsulinemia and insulin resistance coexist, therefore it is unclear which is the main risk factor for atherosclerosis and restenosis. In our previous study, we investigated the effect of insulin on neointimal hyperplasia after balloon catheter injury, a model of restenosis. Normal rats on insulin treatment (5U/day) and given a control diet had significantly reduced neointimal growth at 14 days after arterial injury 133. However, the beneficial effect of insulin was diminished after a high fat diet, which induces insulin resistance similar to metabolic syndrome. The decrease in neointimal area induced after insulin treatment in rats fed the control diet occurred despite an increase in neointimal cell proliferation 133. However, one limitation of the study was that the plasma glucose levels in the insulin- treated rats were lower than normal, and it has been shown in previous studies that hypoglycemia induced by insulin can decrease DNA synthesis in the rat aorta up to four days after arterial injury 215. Although in our study the plasma glucose levels were not in the hypoglycemic range, the effect of insulin was not studied in the presence of normal plasma glucose. In another study, we investigated the effects of insulin on neointimal growth after balloon catheter injury of the carotid artery in rats; however a lower dose of insulin was used. We have shown that insulin treatment (3U/day) in normal rats given oral glucose and standard diet, did not significantly decrease neointimal area at 14 days after arterial injury but decreased neointimal area at 28 days after arterial injury 134. Taken together, the results of the two studies suggest that 68

81 the action of insulin on neointimal growth may be dose- dependent, however it is also possible that in the first study neointimal growth decreased significantly because of glucose lowering rather than the higher insulin dose. A common limitation of the two studies is that the insulin- treated rats were given oral glucose to help maintain plasma glucose levels within the normal range and the effect of oral glucose was not assessed. It is theoretically possible that the decrease in neointimal growth is caused by the oral glucose rather than insulin. Glucose intake orally stimulates gastrointestinal hormones, such as GLP- 1 and GIP, whose receptors are present in vascular cells. GLP- 1 has been reported to induce vasodilation 355 and improve endothelial function 356 whereas GIP was found to affect portal venous and hepatic arterial blood flow in dogs 357. Thus these hormones can have vascular actions, although their effect on neointimal growth has not been investigated. The purpose of this study was to investigate the effects of insulin treatment at the higher dose of 5U/day on neointimal growth after arterial injury, while maintaining the blood glucose levels within the normal physiological range. We have also determined the effect of oral glucose, without insulin treatment, on neointimal growth after arterial injury. In addition, we wished to investigate the effect of oral sucrose alone and in combination with insulin, as a recently published study found that neointimal growth was actually increased in insulin- treated rats after balloon catheter injury while normal plasma glucose levels were maintained with 10% oral sucrose 219. This is in contrast to our studies, where we used 40% glucose in water to maintain plasma glucose and decreased neointimal growth. Sucrose contains fructose, which is a known inducer of insulin resistance 220, thus we speculate that sucrose alone (the effect of which was not evaluated in the previous study) may increase neointimal growth. 69

82 3.3 Materials and Methods Animal Models Male Sprague- Dawley rats weighing between g were used. The rats were divided into six groups: 1) rats with blank implants and access to tap water (Control = C); 2) rats with insulin implant (5U/day) and access to glucose in their drinking water as a 40% (w/v) solution (Insulin (5U/day) + 40% oral glucose = I5G) but with i.p. glucose infusion (25%) to avoid any glucose lowering and correspond to Kim et al 133 ; 3) rats with insulin implant (4U/day) and access to glucose in their drinking water as a 40% (w/v) solution (Insulin (4U/day) + 40% oral glucose = I4G) to compare to Foster et al 219 ; 4) rats with insulin implant (4U/day) and access to sucrose in their drinking water as a 10% (w/v) solution as in Foster et al 219 (Insulin (4U/day) + 10% oral sucrose = I4S); 5) rats with blank implants and access to glucose in their drinking water as a 40% (w/v) solution (Control-Glucose = CG); and 6) rats with blank implants and access to sucrose in their drinking water as 10% (w/v) solution (Control-Sucrose = CS). The dose of insulin at 5U/day resulted in plasma insulin levels similar to that observed in Zucker fatty rats. The animals were maintained as described in the General Methods section 2.1. The rats were implanted with either insulin- releasing implants (5U/day or 4U/day; LinShin Canada Inc.) or blank implants two days before balloon angioplasty as described in the General Methods section Surgical Procedures Refer to the General Methods section for details regarding surgical procedures. Rats were anesthetized and the LCCA was injured with a balloon catheter to denude the vessel of endothelium. After balloon injury, an intraperitoneal catheter was inserted for infusion of 25% glucose to maintain normal blood glucose levels (control rats received continuous infusion of saline) after surgery (when rats do not eat or drink) and the rate of infusion was adjusted 70

83 according to frequent glycemic determination (every 2 hours). After recovery from surgery the insulin (5U/day)- treated rats still received continuous glucose infusion for 14 days to ensure the blood glucose levels remained in the normal range and the infusion rate was monitored on a daily basis, whereas the control rats received saline infusion. In control rats we compared those receiving saline infusion via the i.p. catheter to rats without the i.p. catheter and found no difference in neointimal area (P>0.05) (Table 3.1). Therefore data for these animals was pooled as a single control group (C). In the rats receiving insulin at 4U/day, an i.p. catheter for glucose infusion was not used. Oral glucose or sucrose was used alone to maintain plasma glucose levels within the normal range Vessel and Blood Sample Collection Two weeks after arterial injury the rats were sacrificed where approximately half of the rats were sacrificed in the fed state and half in the fasted state in order to obtain blood samples in different metabolic states. Under general anesthesia, blood samples were collected via cardiac puncture for insulin, FFA, and TG measurements. Rats were then sacrificed via anesthetic overdose and subjected to retrograde perfusion through the abdominal aorta with Krebs-Ringer solution and then with 10% formalin at physiologic pressure ( mmHg). The carotids were removed, immersed in 10% formalin for >24h, and embedded in paraffin (refer to the General Methods section for details) Metabolic Parameters Blood glucose levels were measured daily either in the fasted or fed state as previously described in General Methods section 2.3. Assays for plasma insulin, FFA, and TG were performed as described in the General Methods section 2.12 of this thesis. 71

84 3.3.5 Histomorphometry/Matrix Staining Neointimal and medial cross- sectional areas were measured on hematoxylin and eosin stained slides and analyzed using a computer- assisted morphometric system (Simple PCI, Compix Inc). Neointimal cell proliferation was measured 14 days after carotid balloon injury with an anti-brdu antibody 133. Refer to the General Methods section for a detailed description of staining and analysis Vessel Cannulation Three days before the hyperinsulinemic- euglycemic clamp, rats were anesthetized with isoflurane and indwelling catheters were inserted into the right internal jugular vein for infusion and the LCCA for blood sampling, as described in the General Methods section Hyperinsulinemic- Euglycemic Clamp After an overnight fast, insulin (5mU/kg.min) and glucose (variable rate) were infused through the jugular vein for 2 hours to maintain euglycemia while blood samples were taken from the LCCA for plasma glucose measurements. Refer to the General Methods section 2.10 for a detailed description of the clamp procedure Statistical Analysis Values are expressed as mean ± SEM. Statistical analysis was performed using analysis of variance (ANOVA) to test for significant differences between groups and the post hoc comparisons were performed with Tukey s t- test. Significance was accepted at P< Results Metabolic Parameters Daily food intake, caloric intake, body weight change, as well as the fasting and fed plasma glucose levels was measured over the treatment period. Food intake was significantly less in all rats given oral glucose compared to the control group (C: 24±1g/day; I5G: 9±2g/day; I4G: 72

85 18±1g/day; I4S: 22±1g/day; CG: 14±2g/day; CS: 21±1g/day, P<0.01 I4G vs. C, P<0.001 I5G and CG vs. C) because of the 40% glucose in water. Average daily caloric intake was not significantly different among groups (C: 112±5kcal/day; I5G: 111±9kcal/day; I4G: 137±2kcal/day; I4S: 133±8kcal/day; CG: 101±9kcal/day; CS: 120±5kcal/day). The change in body weight (data not shown) was also not significantly different among groups. Neither fasting nor fed plasma glucose (Table 3.2) was different between control and insulin- treated rats receiving oral glucose; whereas fasting plasma glucose was greater in rats receiving oral sucrose alone (P<0.01). However, fasting plasma glucose was lower in insulin- treated rats receiving oral sucrose (P<0.001), suggesting that the 10% oral sucrose given to the insulin- treated rats was inadequate to maintain normal plasma glucose levels, despite insulin resistance, since these rats received less total glucose than the insulin- treated rats receiving 40% oral glucose. Fasting and fed levels of plasma insulin, TG, and FFA were measured at 14 days after arterial injury. FFA and TG levels were assessed because dyslipidemia is present in the metabolic syndrome and we have shown they have an effect on neointimal thickness 133. As expected, the fasting insulin levels were greater in all insulin- treated groups than in the control groups (P<0.001) (Table 3.3). The fasting TG levels tended to be higher in the control-sucrose group (P<0.10) (Table 3.3). The fasting FFA levels in the insulin- treated rats receiving oral glucose were significantly decreased (P<0.05), whereas in the insulin- treated rats receiving oral sucrose FFA only tended to decrease (P<0.10) (Table 3.3). In the fed state, the insulin levels were greater in the insulin- treated groups than in all other groups (P<0.05), as expected (Table 3.4). The control-sucrose group tended to have elevated fed TG levels (P<0.10) (Table 3.4). The fed FFA levels were not significantly changed in any group (Table 3.4). 73

86 3.4.2 Morphometric Measurements Figure 3.1 shows representative cross- sections of neointimal formation 14 days after balloon catheter injury of the carotid artery in all groups. Insulin treatment at a dose of 5U/day and 4U/day decreased neointimal thickening in rats receiving oral glucose compared to the control rats and those receiving oral glucose alone (P<0.001) (Figures 3.1, 3.2A). In contrast, rats receiving insulin treatment at 4U/day and oral sucrose had a similar neointimal area to control rats or those receiving oral glucose alone. Oral glucose alone did not have any effect on neointimal thickening versus control, however oral sucrose increased neointimal thickening (P<0.05) (Figures 3.1, 3.2A). We also calculated the ratio of neointimal to medial thickness, which confirm our neointimal area results (Figure 3.2B). The calculated lumen area was significantly higher in both the 5U/day and 4U/day insulin- treated groups receiving oral glucose, 0.24 ± 0.02 mm 2 and 0.22 ± 0.02 mm 2 respectively, compared to control, 0.15 ± 0.02 mm 2 (P<0.05) whereas the lumen area in the insulin (4U/day)-treated rats receiving oral sucrose (0.18 ± 0.03 mm 2 ) and the control-glucose group (0.17 ± 0.02 mm 2 ) were not significantly different than control. Lumen area had a slight tendency to decrease in the control-sucrose group (0.12 ± 0.01 mm 2 ) compared to control (P<0.20). The medial area was similar among all groups. The external elastic lamina perimeter for the rats given oral sucrose had a very modest increase (P<0.05) (Figures 3.3A, 3.3B). Neointimal cell proliferation, evaluated as % BrdU incorporation, was increased to a similar degree in the insulin- treated groups and was greater than control (P<0.01) (Figure 3.4). Neointimal cell proliferation in the control-glucose and control-sucrose group was similar to control. The total number of cells was counted in the neointima and was significantly decreased in both insulin-treated groups receiving oral glucose (P<0.05) (Table 3.5). Neointimal cell density was calculated as the total number of cells divided by neointimal cross- sectional area. 74

87 Neointimal cell density tended to be greater in the insulin- treated rats receiving oral glucose compared to control (P<0.10) (Table 3.5) Insulin Sensitivity Measurements To test for the possibility that the sucrose- fed rats were insulin resistant and subsequently insulin was not able to exert its inhibitory effect on neointimal growth, we used a hyperinsulinemiceuglycemic clamp to measure insulin sensitivity after 16 days (equivalent to 2 days before and 14 days after carotid balloon injury) of treatment with oral glucose or oral sucrose. A 5 mu/kg.min insulin infusion rate was chosen to assess half- maximum glucose disposal (glucose infusion rate). The glucose infusion rate during the last 30 minutes of a 2 hour hyperinsulinemic- euglycemic clamp is an indication of whole body insulin sensitivity. Plasma insulin levels during the clamp were not significantly different (C: 699±41pM; CG: 702±27pM; CS: 812±53pM). The glucose infusion rate was unchanged in rats receiving oral glucose treatment compared to those receiving tap water (P>0.05) (Figure 3.5), indicating that oral glucose did not affect insulin sensitivity. However, oral sucrose significantly decreased the glucose infusion rate (P<0.05), suggesting that indeed oral sucrose induced insulin resistance in this model. 3.5 Discussion. In the present study the in vivo effects of insulin, without changes in glycemia, and oral glucose or sucrose on neointimal formation in the rat carotid balloon injury model were investigated. Insulin treatment combined with oral glucose but not oral sucrose decreased neointimal area after injury. Oral glucose alone had no such effect, however, oral sucrose had opposite effects to those of insulin on neointimal area, which may be attributed to a decrease in insulin sensitivity. 75

88 3.5.1 Effect of insulin and oral glucose. In the fasted state, FFA levels of the I5G and I4G group were lower than the control group. This was expected since insulin prevents lipolysis by inhibiting hormone sensitive lipase 358 and suggest that these rats remained insulin sensitive. In the current study, normal plasma glucose levels, not significantly different from control, were achieved in both the fasted and fed states in the insulin (5U/day)- treated group by infusing glucose i.p. and by providing glucose in the drinking water. At the lower dose of insulin (4U/day) oral glucose was sufficient to maintain normal plasma glucose levels. The insulin treatment in rats receiving oral glucose resulted in a decrease in neointimal thickness despite an increase in neointimal cell proliferation, whereas oral glucose alone had no effect. Also, the increase in lumen area with insulin was not a result of outward remodeling since the external elastic lamina perimeter measured in these groups was the same. Taken together, these data indicate that insulin treatment reduces neointimal growth after arterial injury independent of oral glucose administration and confirms that the decrease in neointimal thickness observed in Kim et al 133 was due to a direct effect of insulin and not to an insulin- induced decrease in plasma glucose. The extent of the percentage decrease was also comparable to Kim et al 133, even though our absolute neointimal thickness is greater than in the previous study, since a different experimenter was performing the surgery. Also notable and in accordance with our previous study 133 was the effect of insulin to increase neointimal cell proliferation even though neointimal area was decreased. This suggests that insulin may have both growth- promoting and vasculoprotective effects, however in our conditions the vasculoprotective effect of insulin to decrease neointimal area prevailed. An inhibitory effect on VSMC migration or stimulation of apoptosis could explain why the total number of neointimal cells did not increase with insulin despite the increase in proliferation. We have shown that a lower dose of insulin does inhibit cell migration without affecting apoptosis 134. Furthermore, in the present study insulin + oral glucose also tended to increase 76

89 neointimal cell density despite decreasing total neointimal cell number, suggesting that insulin inhibits ECM accumulation, which we have observed previously 134. A decrease in ECM accumulation in addition to insulin s inhibition of cell migration would also result in a decrease in neointimal growth Effect of insulin and oral sucrose. We also examined the effect of oral sucrose alone and in combination with insulin on neointimal growth since insulin was shown to increase neointimal hyperplasia by Foster et al 219 in rats where 10% sucrose water was used to maintain normoglycemia. We found that in insulintreated rats receiving oral sucrose, the inhibitory effect of insulin on neointimal growth was not observed while oral sucrose alone increased neointimal thickness after arterial injury. Interestingly, however, oral sucrose did not block the effect of insulin to increase neointimal cell proliferation suggesting the development of selective insulin resistance. Sucrose is a disaccharide of glucose and fructose. Both high- sucrose 359 and high- fructose 360 feeding are commonly used to induce insulin resistance in rats and studies have shown that fructose is the nutrient mediator of sucrose- induced insulin resistance 220. Glucose feeding however, did not affect insulin sensitivity 361. The oral sucrose group in our study had elevated fasting plasma glucose levels, indicating that these rats were insulin resistant, which was confirmed with the hyperinsulinemic- euglycemic clamp as oral sucrose decreased insulin sensitivity compared to the tap water and oral glucose groups. In addition, the oral sucrose group tended to have increased fasting and fed TG levels, which are also associated with insulin resistance 362 and are observed after sucrose feeding as fructose bypasses the glucokinase checkpoint and also the phosphofructokinase checkpoint in the liver, immediately entering glycolysis and fatty acid synthesis 363. Sucrose- induced protein glycation may also be involved as advanced glycation endproducts (AGEs) damage endothelial cells and ECM proteins, which could promote neointimal growth or explain the increased external elastic perimeter (an 77

90 indication of vessel remodeling). Therefore, it seems that in Foster et al 219, the oral sucrose given to the insulin- treated rats reduced insulin sensitivity, thus blocking insulin s action to inhibit neointimal growth. However, insulin sensitivity at the level of the vessel was not assessed. Future studies could measure markers of insulin sensitivity, for example serine phosphorylation of IRS-1, via Western blot analysis. Why in Foster s study insulin + oral sucrose actually increased neointimal growth whereas in our study it only failed to decrease it compared to control is not clear. The rats in Foster s study are older than ours and perhaps more insulin resistant, thus it is possible that the effect of insulin was enough to counteract the effect of oral sucrose in our study. In conclusion, we demonstrate that insulin treatment reduces neointimal growth at 14 days after arterial injury due to an effect of insulin and not because of insulin- induced glucose lowering or oral glucose ingestion, and that oral sucrose should not be used to maintain normoglycemia while giving insulin because it increases neointimal growth. Thus, these results further support a vascular protective effect of insulin in vivo. 78

91 Table 3.1. Neointimal formation with or without an i.p. catheter. Cross-sectional areas of the intima of carotid arteries from control rats with an i.p. catheter and control rats without an i.p. catheter measured at 14 days after vessel injury. 14 days Treatment Groups Control With i.p. Catheter (n=5) Control Without i.p. Catheter (n=5) Intimal Area (mm²) 0.23± ±0.04 Values are means ± SEM. No significant differences were found. 79

92 Table 3.2. Daily fasting and fed plasma glucose levels over the treatment period. 14 days Treatment Groups C (n=10) I5G (n=9) I4G (n=8) I4S (n=8) CG (n=10) CS (n=10) Fasting Plasma Glucose (mm) ± ± 0.3*** ± 0.1** Fed Plasma Glucose (mm) ± ± ± 0.1 C=Control; I5G=Insulin (5U/day) + oral glucose; I4G=Insulin (4U/day) + oral glucose; I4S=Insulin (4U/day) + oral sucrose; CG=Oral Glucose; CS=Oral Sucrose. Data are expressed as means + SEM. **P<0.01, ***P<0.001 vs. C. Glucose levels were measured daily for 14 days. Data are means + SEM of individual average glucose levels taken over the treatment period after an overnight 16 hour fast. Data are means + SEM of individual average glucose levels taken over the treatment period. 80

93 Table 3.3. Fasting plasma levels of insulin, triglycerides, and free fatty acids at 14 days after arterial injury. 14 days Treatment Groups C (n=5) I5G (n=4) I4G (n=3) I4S (n=3) CG (n=5) CS (n=4) Insulin (pm) 82±8 1442±223*** 927±114*** 704±83*** 88±14 109±7 Triglycerides (mm) 0.72± ± ± ± ± ±0.03 Free Fatty Acids (ueq/l) 632± ±84* 105±38* 342±27 329±75 476±145 C=Control; I5G=Insulin (5U/day) + oral glucose; I4G=Insulin (4U/day) + oral glucose; I4S=Insulin (4U/day) + oral sucrose; CG=Oral Glucose; CS=Oral Sucrose. Data are expressed as means + SEM and represent values taken at sacrifice. *P<0.05, ***P<0.001 vs. C. 81

94 Table 3.4. Fed plasma levels of insulin, triglycerides, and free fatty acids at 14 days after arterial injury. 14 days Treatment Groups C (n=5) I5G (n=5) I4G (n=5) I4S (n=5) CG (n=5) CS (n=6) Insulin (pm) 170± *** 1006±167* 969±100* ±20 Triglycerides (mm) 1.23± ± ± ±0.26 Free Fatty Acids (ueq/l) 230± ±23 256± ±38 C=Control; I5G=Insulin (5U/day) + oral glucose; I4G=Insulin (4U/day) + oral glucose; I4S=Insulin (4U/day) + oral sucrose; CG=Oral Glucose; CS=Oral Sucrose. Data are expressed as means + SEM and represent values taken at sacrifice. *P<0.05, ***P<0.001 vs. C. 82

95 Table 3.5. Intimal cell number and cell density per unit area measured at 14 days after vessel injury. 14 days Treatment Groups C (n=10) I5G (n=7) I4G (n=8) I4S (n=8) CG (n=9) CS (n=10) Intimal Cell Number 1276±93 914±47* # 864±47* ## 1238± ± ±113 Intimal Cell Density (cells/mm²) 5272± ± ± ± ± ±146 C=Control; I5G=Insulin (5U/day) + oral glucose; I4G=Insulin (4U/day) + oral glucose; I4S=Insulin (4U/day) + oral sucrose; CG=Oral Glucose; CS=Oral Sucrose. Data are expressed means ± SEM. *P<0.05 vs. C. # P<0.05, ## P<0.01 vs. CG. 83

96 L L I I M C I5G M L L I I M M L I4G I4S L I I M CG M CS Figure 3.1. The effect of insulin, oral glucose, and oral sucrose on neointimal formation. Representative photomicrographs of cross-sections taken at 14 days after carotid injury (x400). L=Lumen; I=Intima; M=Media. 84

97 A Intimal Area (mm 2 ) *** ### *** ### * 0.00 C I5G I4G I4S CG CS B 2.00 * IA :MA *** ### ** ## 0.00 C I5G I4G I4S CG CS Figure 3.2. The effect of insulin, oral glucose, and oral sucrose on intimal area. (A) Crosssectional areas of the intima of carotid arteries measured at 14 days after carotid injury (C: n=10; I5G: n=9; I4G: n=8; I4S: n=8; CG: n=10; CS: n=10). (B) Ratio of intima: medial area. Values are means ± SEM. *P<0.05, **P<0.01, ***P<0.001 vs. C. ## P<0.01, ### P<0.001 vs. CG. P<0.001 vs. CS. 85

98 A 0.20 Medial Area (mm 2 ) C I5G I4G I4S CG CS B External Elastic Lamina Perimeter (mm) C I5G I4G I4S CG CS * ** Figure 3.3. The effect of insulin, oral glucose, and oral sucrose on medial area and external elastic lamina perimeter. Cross-sectional areas of the (A) media and (B) external elastic lamina of carotid arteries measured at 14 days after carotid injury (C: n=10; I5G: n=9; I4G: n=8; I4S: n=8; CG: n=10; CS: n=10). Values are means ± SEM. *P<0.05, **P<0.01 vs. C. 86

99 BrdU- Labelled Cells (%) ** # ** # *** 0 C I5G I4G I4S CG CS Figure 3.4. The effect of insulin, oral glucose, and oral sucrose on cell proliferation. Cell proliferation expressed as % BrdU-labeled cells in the intima at 14 days after carotid injury (C: n=10; I5G: n=7; I4G: n=8; I4S: n=8; CG: n=9; CS: n=10). Values are means ± SEM. **P<0.01, ***P<0.001 vs. C. # P<0.05 vs. CG. P<0.001 vs. CS. 87

100 Glucose Infusion Rate (umol/kg/min) * C CG CS Figure 3.5. The effect of oral glucose and oral sucrose on whole body insulin sensitivity. Whole body insulin sensitivity is indicated by the glucose infusion rate during the last 30 min of the 2h hyperinsulinemic- euglycemic clamp (C: n=10; CG: n=9; CS: n=8). Values are expressed as mean ± SEM. *P<0.05 vs. C. 88

101 4 Study 2 The Vasculoprotective Effects of Insulin are Nitric Oxide Synthase- Dependent The data presented in Study 2 are currently unpublished except for Table 4.3 and Figure 4.4A-F, which are published in and modified from: Breen DM*, Chan KK*, Dhaliwall JK, Ward MR, Koudsi NA, Lam L, De Souza M, Ghanim H, Dandona P, Stewart D, Bendeck MP, Giacca A. Insulin Increases Re-endothelialization and Inhibits Cell Migration and Neointimal Growth After Arterial Injury. Arteriosclerosis, Thrombosis & Vascular Biology. 2009; 29(7): *Equal contribution. Permission to reproduce portions of the above manuscript has been obtained from the copyright owner, American Heart Association. 89

102 4.1 Abstract Background/Aims: Metabolic syndrome, characterized by insulin resistance, is a risk factor for atherosclerosis and restenosis after percutaneous revascularization procedures. Insulin has protective effects on the endothelium by stimulating nitric oxide (NO) production and endothelial nitric oxide synthase (enos) expression but also has mitogenic effects on vascular smooth muscle cells (VSMC) in vitro. Furthermore, NOS inhibition has been reported to abolish the effect of insulin to inhibit VSMC migration in vitro. We have previously shown that insulin decreases neointimal growth and cell migration to the neointima after carotid angioplasty in normal rats. Since insulin can stimulate NOS and NO can decrease neointimal growth, we investigated the consequences of NOS inhibition, with the in vivo inhibitor L-NAME, on the inhibitory effects of insulin. Methods: Rats were given subcutaneous implants and divided into four groups: 1) control; 2) insulin (3U/day); 3) insulin+l-name (2mg/kg/day); and 4) control+l-name. Treatment began three days before arterial (carotid or aortic) balloon catheter injury and was continued postinjury. Results: Insulin decreased both neointimal area (P<0.05) and cell migration (P<0.001), and increased re- endothelialization (P<0.05). The addition of L-NAME abolished insulin s effect on neointimal growth, cell migration, and also re- endothelialization. Furthermore, insulin increased expression of VSMC differentiation markers (P<0.05) and the number of circulating progenitor cells (P<0.05). enos protein expression tended to increase with insulin treatment (P=0.08). Conclusions: These results show that insulin has a protective effect on both the VSMC and endothelium in vivo, and that NO plays a role in mediating these effects. 90

103 4.2 Introduction Type 2 diabetes and metabolic syndrome are risk factors for atherosclerotic CVD, partly because of associated abnormalities such as hypertension, dyslipidemia, glucose intolerance or overt hyperglycemia, and partly because of the atherogenic risk of insulin resistance and/ or hyperinsulinemia per se. There is evidence for both growth promoting and protective vascular effects of insulin. For example, insulin increases proliferation of VSMCs in vitro 266. However, insulin also stimulates endothelial cell production of NO 130, which inhibits VSMC growth and migration 272, and has anti- inflammatory effects 261. In addition to the increased risk for atherosclerosis, patients with diabetes and metabolic syndrome have an increased risk for restenosis after PTCA and stenting 10. After balloon angioplasty, neointimal injury results in endothelial denudation, followed by platelet activation and secretion of growth factors from platelets and damaged VSMCs 5. Subsequently, there is VSMC migration and proliferation, and ECM deposition resulting in neointimal growth and lumen narrowing 5. Although the effect of insulin on restenosis remains a point of debate, we have shown 133, 134, that insulin treatment is protective and decreases neointimal formation after arterial injury 364. In addition, we observed that insulin, without changes in glycemia, decreases cell migration and accelerates re- endothelialization, resulting in decreased neointimal growth 134. Early re- endothelialization of the injured vessels inhibits lesion development, facilitates vascular repair, and improves long term vessel patency 365. Vascular endothelial cells produce the vasodilator molecule NO, which has significant protective effects on several steps of the restenosis process as outlined in the Introduction section 1.7. For example, increased enos expression via gene transfer diminishes neointimal formation after arterial injury 152 and enos deficiency results in greater neointimal growth after cuff injury 104, 195. Furthermore treatment of 91

104 ApoE knockout mice with AVE9488, a compound that increases enos mrna and protein expression, reduced neointimal formation after cuff- injury 366. The primary role of inos on neointimal formation is less clear, with studies indicating that it accelerates restenosis 154, and others indicating that it decreases it 156, 157. More recently, adenoviral- mediated gene delivery of enos and inos following endothelial injury in the rabbit carotid artery showed that both enos and inos reduced neointimal formation, however enos gene delivery also accelerated reendothelialization whereas inos gene delivery actually inhibited re- endothelialization, highlighting a possible advantage of enos 367. Insulin activates and upregulates enos via PI3K and Akt 235, 236. Insulin has also been reported to upregulate inos 239 and enhance NO- induced guanylate cyclase activity in VSMCs 240. Furthermore, insulin s inhibitory effect on VSMC migration in vitro was abolished by NOS inhibitors 272. Thus, NOS activation represents a likely mechanism for insulin to exert its inhibition of neointimal formation. Therefore, the objective of our study is to investigate the consequences of NOS inhibition on the effects of insulin. 4.3 Materials and Methods Animal Models Four groups of male Sprague- Dawley rats weighing g and fed normal rat chow were studied: 1) control (C); 2) insulin (3U/day; s.c. capsule (bovine insulin))- treated (LinShin Inc.) (I); 3) insulin + NG-nitro-L-arginine methyl ester (L-NAME) (2mg/kg/day) (I+L2); and 4) control + L-NAME (C+L2). Insulin was given at 3U/day, which is lower than the dose used in Study 1, as 40% oral glucose was previously shown to maintain plasma glucose levels similar to control without the additional i.p. glucose infusion 134. This dose of L-NAME has been previously shown to inhibit NO production in the rat aorta, based on vessel cgmp assays 368, without affecting blood pressure 369. All insulin- treated rats were given 40% glucose in drinking water to 92

105 maintain normoglycemia, which we have found in Study 1 to have no effect on neointimal formation. The animals were maintained as described in the General Methods section Surgical Procedures Three days after onset of treatment, injury to either the LCCA or the aorta of the rat was induced by means of an inflated balloon catheter 133. Refer to the General Methods section 2.2.4/ for details regarding surgical procedures Vessel and Blood Sample Collection The rats were sacrificed in either the fed or fasted state at various time points after injury, based on previous studies investigating the kinetics of vascular injury. Blood samples were collected by cardiac puncture under general anesthesia. Thereafter, rats were sacrificed via anesthetic overdose and the carotids were perfusion- fixed for four minutes with 10% buffered formalin at physiologic pressure ( mmHg). The carotids were removed, immersed in 10% formalin for >24h and embedded in paraffin. Refer to General Methods section for a complete description Metabolic Parameters Blood glucose was monitored throughout the treatment using a glucometer (see General Methods section 2.3). Plasma insulin was determined by radioimmunoassay using kits specific for rat insulin. Plasma FFA and TG were measured using specific colorimetric kits. Refer to General Methods section 2.12 for details Histomorphometry/Matrix Staining Cell migration into the neointima was measured four days after injury and neointimal cells on the lumen surface were immunostained with an antibody against histone H1 and counted 82. Refer to General Methods section for complete details. Cross- sectional areas were measured on hematoxylin and eosin stained slides at 14 and 28 days after injury. Crosssections taken at 28 days after injury were also stained with Movat s pentachrome or PSR dye 93

106 for elastin and collagen respectively. Refer to the General Methods section for a detailed description of staining and analysis Evan s Blue Staining Re- endothelialized area was measured at 14 days after balloon injury and visualized with Evan s Blue dye, which does not stain endothelial cells. Following perfusion, the thoracic aorta was dissected out and opened longitudinally where the area on the luminal surface stained with Evan s Blue was measured (General Methods section 2.5.5). Cross- sectional areas were measured on hematoxylin and eosin stained slides from unopened sections of these aortas as described in the General Methods section Western Blot Analysis Carotids were removed seven days after injury, homogenized, and processed as previously described in the General Methods section 2.6. Protein was separated by SDS-PAGE and analyzed with anti- SM-MHC, anti- SM22α, anti-smc α-actin, anti- total enos, and antialpha tubulin antibodies. Refer to the General Methods section 2.8 for a detailed description Flow Cytometry Rat blood was obtained 3 and 7 days after carotid injury in separate rats by cardiac puncture and flow cytometry was performed as in Fadini et al 115 (see General Methods section 2.9 for details) Blood Pressure Analysis Systolic blood pressure was measured indirectly from the tail via the cuff technique, using the Visitech BP-2000 Blood Pressure Analysis System (see General Methods section 2.11 for details). Blood pressure was measured before treatment and throughout at 4, 14, and 28 days after arterial injury. 94

107 Statistical Analysis Values are expressed as mean ± SEM. Statistical analysis was performed using analysis of variance (ANOVA) to test for significant differences between groups and the post hoc comparisons were performed with Tukey s t- test. Significance was accepted at P< Results Metabolic Parameters Daily food intake, caloric intake, body weight change, as well as the fasting and fed plasma glucose levels were measured over the treatment period. Food intake was less for insulin (I)- and I+L-treated rats (C: 29±1g/day; I: 19±1g/day; I+L2: 20±1g/day; C+L2: 29±1g/day, P<0.001 I and I+L vs. C), however despite this, the insulin and I+L group had greater total caloric intake compared to controls due to the 40% glucose in drinking water (C: 93±3kcal/day; I: 125±3kcal/day; I+L2: 126±2kcal/day; C+L2: 95±3kcal/day, P<0.001 I and I+L2 vs. C). There were no significant changes in body weight, however the insulin- and I+L2-treated rats tended to gain more weight than controls (C: 91±9g; I: 121±13g; I+L2: 114±10g; C+L2: 84±11g, P>0.20 I vs. C; P>0.30 I+L2 vs. C). Both fasting and fed plasma glucose levels taken over the treatment period were similar in all treatment groups (Table 4.1). Fed levels of plasma insulin, TG, and FFA were measured at 28 days after arterial injury. FFA and TG levels were assessed because dyslipidemia is present in the metabolic syndrome and we have shown they have an effect on neointimal thickness 133. As expected, plasma insulin levels were greater in the insulin- treated groups than in all other groups (P<0.01) (Table 4.2). Fed TG and FFA levels were not significantly different from control. 95

108 4.4.2 Morphometric Measurements Treatment with insulin for 28 days after balloon injury resulted in decreased neointimal area (Figure 4.1A,B) (P<0.05). This effect was completely blocked by co- treatment with the NOS inhibitor L-NAME, indicating an NO- dependent effect of insulin. We also calculated the ratio of neointimal to medial thickness, which confirm our neointimal area results (Figure 4.1C). By contrast, there was no change in medial area compared to control (C=0.15±0.01mm 2, I=0.15±0.01mm 2, I+L2=0.16±0.01mm 2, C+L2=0.16±0.01mm 2 ). There were no significant differences in external elastic lamina perimeter compared to control (C=2.8±0.1mm, I=2.6±0.2mm, I+L2=3.0±0.1mm, C+L2= 3.0±0.1mm), indicating that outward remodeling did not likely contribute to the changes in neointimal area with treatment. ECM accumulation is a major determinant of neointimal size. Therefore, vessel crosssections were stained with Movat s pentachrome for elastin or PSR for collagen to assess the ECM contribution. Insulin significantly decreased elastin accumulation by 50.7% (P<0.001) and had a slight tendency to decrease collagen accumulation by 31.8% (P=0.30) in the intima of the carotid at 28 days compared to control, and this effect was prevented by co- administration of L- NAME (Figure 4.2). To investigate whether insulin decreased cell migration through a NOS- dependent mechanism, we measured cell migration from the media to the neointima four days after vessel injury. Insulin markedly inhibited migration by ~62%, reducing the number of neointimal cells from 534±55cells/mm 2 in control rats to 204±55cells/mm 2 in the insulin group (P<0.001) (Figure 4.3A,B). The addition of L-NAME completely abolished the effect of insulin to decrease migration and resulted in even greater cell migration than control (I+L2=787±87cells/mm 2 ) (P<0.05). VSMC dedifferentiation from the contractile to the synthetic migratory/ proliferative phenotype is associated with increased VSMC migration and proliferation. In vitro studies have shown that insulin promotes VSMC differentiation 370, therefore we investigated the in vivo effect 96

109 of insulin on the VSMC phenotype by measuring three markers of differentiation seven days after injury (Figure 4.4A-F). Western blots revealed an increase in SM-MHC (P<0.05), SM22α (P<0.05), and SMC α-actin (P<0.01) expression with insulin. Since recruitment of circulating progenitor cells contributes to neointima formation 18, we used flow cytometry to detect circulating progenitor cells, which were defined by sca-1 or c-kit positivity 115, 153. Insulin increased the number of circulating cells positive for sca-1 (P<0.05) (Table 4.3) and tended to increase the number of c-kit positive cells at 3 days after injury (P=0.18). There were no significant differences observed with insulin treatment for cells positive for VEGFR2, an endothelial cell marker. Only a small quantity of cells were double positive for c-kit and VEGFR2 (double positivity would indicate an endothelial progenitor cell), however there was no difference with insulin treatment In addition, we could not detect any cells double positive for sca-1 and VEGFR2, which is similar to Fadini et al 115. Circulating precursors also contribute to re- endothelialization and it is possible that neointimal formation is reduced by accelerated re- endothelialization. To address this, we used a model of aortic injury where re- endothelialization is more consistent and complete 371. Analysis of rat aorta specimens after Evan s Blue staining at 14 days showed that insulin treatment accelerated re- endothelialization of the balloon-injured arterial segments by ~22% compared to control (P<0.05) (Figure 4.5A,B), which was completely prevented by co- administration of L- NAME. This was accompanied by a significant reduction in neointimal thickening in the aorta after balloon injury in the insulin-treated group (P<0.01), which was prevented by L-NAME, (Figure 4.5C) as also observed in the carotid artery. In an attempt to begin to address the specific NOS isoform that may be involved in insulin action, we used Western blot to measure changes in enos protein expression after balloon injury in the vessel. Insulin treatment tended to increase enos expression compared to control (P=0.08) (Figure 4.4G,H). 97

110 Furthermore, no changes were observed in systolic blood pressure with either insulin or L-NAME administration over the course of treatment (Figure 4.6). 4.5 Discussion The present study focused on the in vivo effect of insulin on the kinetics of neointimal growth in the rat arterial injury model and the possible involvement of NO in insulin action. Insulin accelerated re- endothelialization and had a marked inhibitory effect on injury- induced cell migration, leading to decreased neointimal area. The inhibitory effect of insulin on cell migration and neointimal formation, as well as the stimulatory effect on re- endothelialization, was completely abolished by L-NAME. Therefore, our data provide evidence for a protective effect of insulin on vascular growth that is mediated by NO. We did not measure neointimal cell proliferation or apoptosis after insulin treatment as we have previously found no effect of subcutaneous insulin (3U/day) treatment on proliferation or apoptosis in the same model of carotid balloon injury 134. Insulin increases VSMC differentiation as indicated by increased VSMC differentiation markers. As mentioned above, medial VSMCs contribute to neointimal growth by dedifferentiating and migrating to the intima. Insulin could have affected migration via an effect on cell motility 273 and/ or through the regulation of the medial VSMC phenotype 370. In vitro studies have implicated a PI3K- inos- NO- cgmp- dependent pathway for insulin to inhibit CamKinaseII activation and VSMC migration 239, 273, and promote the maintenance of a quiescent phenotype. As previously mentioned in the Introduction section 1.7, studies have demonstrated that NO deceases the synthesis of ECM matrix components 182, 184. In agreement with these studies, we found a NO- mediated decrease in ECM accumulation after insulin treatment. This decrease in ECM production likely contributes to the decrease in neointimal area and is consistent with 98

111 insulin preventing the VSMC switch from a contractile to synthetic migratory/ proliferative phenotype. Of the three isoforms of NOS, enos is the most likely to be involved in insulin action as enos- derived NO induces vasorelaxation, inhibits leukocyte- endothelial adhesion, platelet aggregation, VSMC migration and proliferation, and increases re- endothelialization 163. As previously mentioned, upregulation of enos diminishes neointimal formation after arterial injury 152, 372 and enos deficiency promotes neointimal growth after cuff injury 104, 195. Carotid balloon injury, contrary to cuff injury 373, results in complete denudation of the endothelium and re- endothelialization is seldom complete 104, 374. Since there is no endothelium after balloon injury, it might be argued that insulin- induced changes in enos should not be important. However, enos can be activated in the endothelial cells bordering the lesion to inhibit neointimal thickness and circulating EPCs expressing enos contribute to reendothelialization 20. In addition, enos is expressed in platelets 147 and insulin has a potent effect on stimulating platelet enos 375 and NO production 376, thereby decreasing platelet adhesion and aggregation 131, 377. A reduction in the number of platelets at the site of injury would also have an additional benefit of reducing VSMC migration, as platelets have a key role in promoting VSMC migration through PDGF secretion. Consistent with our results using L-NAME, L- arginine, the precursor of NO, has been shown to inhibit neointimal formation after arterial balloon injury in normal rats 378 and rabbits 105, although the specific NOS isoform activated was not addressed. Insulin has also been reported to stimulate inos 239, 240 in VSMCs, which do not express enos after injury 148, 379. However, as discussed in the introduction, the role of inos on neointimal formation is not clear. Re- endothelialization begins in the endothelium (expresses enos) at the margin of the lesion 380 and from circulating precursors 20, and is stimulated by NO 166. As previously mentioned, gene delivery of enos increased re- endothelialization whereas inos gene delivery inhibited re- 99

112 endothelialization after endothelial injury in the rabbit carotid artery, although both resulted in decreased neointimal formation 367. Furthermore, we have shown that insulin increases the number of circulating progenitor cells, which are known to express enos 20. Since insulin only tended to increase enos expression in the vessel, insulin may also increase enos activity as insulin is known to activate 235 and upregulate 236 enos via PI3K. In summary, insulin stimulated endothelial regeneration, tended to increase enos protein expression, and also suppressed neointimal thickening. Taken together, these data suggest that enos is the specific NOS isoform involved in insulin action to decrease neointimal formation as activation of inos may have likely resulted in decreased re- endothelialization. Unfortunately, studies using an enos or inos specific in vivo inhibitor are limited as there is no specific enos inhibitor and studies using prolonged inos inhibition are not feasible due to the high cost of the inhibitors. Not much is known about the role of nnos in the vascular system, although there is one report of induction of nnos in VSMC after arterial injury and aggravation of neointimal formation after nnos inhibition 159. Insulin has also been shown to upregulate nnos in neurons 238. Hypertension is among the cluster of disorders present in the metabolic syndrome 353 and available studies have demonstrated that insulin may improve blood pressure in models of hypertension 381, 382, however in the present study blood pressure was not affected by insulin. L- NAME, at specific doses, is known to be hypertensive as it blocks NO- mediated vasodilatation. However, although not measured in this study, the dose of L-NAME used was previously shown to inhibit NO production based on vessel cgmp assays 368 without affecting blood pressure 369. In conclusion, we have shown for the first time that the effect of insulin to decrease neointimal formation after arterial injury is NO- dependent. These results provide evidence for inhibition of the growth response to vessel injury by insulin in vivo and suggest that in metabolic syndrome insulin resistance may result in loss of vascular protection. 100

113 Table 4.1. Daily fasting and fed plasma glucose levels over the treatment period. 28 days Treatment Groups C (n=10) I (n=8) I+L2 (n=8) C+L2 (n=8) Fasting Plasma Glucose (mm) 5.0± ± ± ±0.1 Fed Plasma Glucose (mm) 6.5± ± ± ±0.1 C=Control; I=Insulin; L2=L-NAME (2mg/kg/day). Data are expressed as means + SEM. No significant differences were found compared to control. Glucose levels were measured daily for 28 days. Data are means + SEM of individual average glucose levels taken over the treatment period after an overnight 16 hour fast. Data are means + SEM of individual average glucose levels taken over the treatment period. 101

114 Table 4.2. Fed plasma levels of insulin, triglycerides, and free fatty acids at 28 days after arterial injury. 28 days Treatment Groups C (n=7) I (n=6) I+L2 (n=4) C+L2 (n=4) Insulin (pm) 128±16 834±90** 1017±355** 141±20 Triglycerides (mm) 1.87± ± ± ±0.45 Free Fatty Acids (ueq/l) 311±69 233±49 417±40 503±42 C=Control; I=Insulin; L2=L-NAME (2mg/kg/day). Data are expressed as means + SEM and represent values taken at sacrifice. **P<0.01 vs. C. Note: The rats used to measure neointimal formation in Figure 4.1 were randomly sacrificed in either the fed or fasting state, and as a result there are fewer rats in each treatment group in the above table compared to Figure 4.1. This may have limited the power to detect differences between treatment groups. 102

115 Table 4.3. Percentage of white blood cells stained positive for c-kit, sca-1, and VEGFR2. Basal 3 days 7 days Treatment Groups Untreated/ Uninjured (n=5) C (n=7) I (n=7) C (n=9) I (n=8) c-kitª 4.8 ± ± ± ± ± 0.3 sca-1ª 0.16 ± ± ± 0.03* 0.15 ± ± 0.02 VEGFR2ª ± ± ± ± ± c-kit/ VEGFR2ª ± ± ± sca-1/ VEGFR2ª C=Control; I=Insulin. Data are expressed as means + SEM and represent values taken at sacrifice. *P<0.05 vs. C. ªExpressed as % of all cells defined (gated) as white blood cells. Note: In the Fadini et al 115 paper sca-1 + cells separated by fluorescence- activated cell sorting also showed very little positivity for VEGFR2. 103

116 A C I C+L2 I+L2 B Intimal Area (mm 2 ) C I I+L2 C+L2 * # C IA:MA C I I+L2 C+L2 * # days 28 days Figure 4.1. The effect of insulin and NOS inhibition on intimal area. (A) Representative photomicrographs of cross-sections taken at 28 days after carotid injury (x400). (B) Crosssectional areas of the intima of carotid arteries measured at 28 days after injury. (C) Ratio of intimal: medial area (C: n=10; I: n=8; I+L2: n=8; C+L2: n=8). Values are means ± SEM. *P<0.05 vs. C, # P<0.05 vs. I+L2. 104

117 A C C I C I C+L2 I+L2 C+L2 I+L2 B D % Intimal Area Elastin Positive C I I+L2 C+L2 *** # 28 days % Intimal Area Collagen Positive C I I+L2 C+L2 28 days Figure 4.2. The effect of insulin and NOS inhibition on elastin and collagen accumulation. Representative images of staining with (A) Movat s pentachrome to show elastin and (C) PSR to show collagen in the intima at 28 days after carotid injury. Quantitative analysis of cross-sections stained for elastin (B) or collagen (D) (C: n=10; I: n=8; I+L2: n=8; C+L2: n=8). Values are means ± SEM. ***P<0.001 vs. C, # P<0.05 vs. I+L2. 105

118 A C I B Number of cells/mm C+L2 C I I+L2 C+L2 *** ### * I+L2 0 4 days Figure 4.3. The effect of insulin and NOS inhibition on cell migration. (A) Representative photomicrographs showing cells on the lumen surface at 4 days after carotid injury (x400). (B) Quantitative analysis of cell migration (C: n=7; I: n=10; I+L2: n=8; C+L2: n=6). Values are means ± SEM. *P<0.05, ***P<0.001 vs. C, ### P<0.001 vs. I+L2. 106

119 A SM-MHC alpha-tubulin C SM22alpha alpha-tubulin B SM-MHC Expression (Arbitrary Densitometric Units Normalized by Alpha-tubulin) Control Insulin * D SM22alpha Expression (Arbitrary Densitometric Units Normalized by Alpha-tubulin) Control Insulin * 7 days 7 days E alpha-actin alpha-tubulin G enos F Alpha-actin Expression (Arbitrary Densitometric Units Normalized by Alpha-tubulin) Control Insulin H enos Expression (Arbitrary Densitometric Units Normalized by Alpha-tubulin) Control Insulin alpha-tubulin P= days 7 days Figure 4.4. The effect of insulin on SMC differentiation markers and enos expression. Representative Western blot showing (A) SM-MHC, (C) SM22α, (E) SMC α-actin, and (G) enos. Densitometric analysis of Western blots to detect (B) SM-MHC, (D) SM22α, (F) SMC α- actin, and (H) enos (C: n=7-8; I: n=7-8). Values are means ± SEM. *P<0.05, P<0.01 vs. C, P=0.08 vs. C. 107

120 A C C+L2 B C % Re-endothelialization Intimal Area (mm 2 ) I * ## 14 days ** ## 14 days I+L2 Figure 4.5. The effect of insulin and NOS inhibition on re-endothelialization. (A) Representative images of Evan s Blue dye staining of aorta at 14 days after vessel injury. (B) Quantification of re-endothelialized area (% of total vessel area). (C) Cross-sectional areas of the intima in the aorta at 14 days after vessel injury (C: n=7; I: n=8; I+L2: n=8; C+L2: n=7). Values are means ± SEM. *P<0.05, **P<0.01 vs. C, ## P<0.01 vs. I+L C I I+L2 C+L2 C I I+L2 C+L2

121 Systolic Blood Pressure (mmhg) C C+L2 I I+L Days Figure 4.6. The effect of insulin and NOS inhibition on systolic blood pressure. Systolic blood pressure at 3 days before and 4, 14, and 28 days after carotid injury (corresponding to day 0, 7, 17, and 31 of treatment respectively). (C: n=10; I: n=8; I+L2: n=8; C+L2: n=8). Values are means ± SEM. No significant differences were found. 109

122 5 Study 3 Resveratrol Inhibits Neointimal Formation After Arterial Injury Through A Nitric Oxide Synthase- Dependent Mechanism Modified from: Breen DM, Guo J, Mroziewicz M, Tsiani E, Bendeck MP, Giacca A. Resveratrol Inhibits Neointimal Formation After Arterial Injury Through A Nitric Oxide Synthase- Dependent Mechanism. (American Journal of Pathology, In Revision #AJP ). Permission to reproduce portions of the above manuscript is not yet required from the copyright owner, the American Journal of Pathology. 110

123 5.1 Abstract Background/Aims: Revascularization procedures used for treatment of atherosclerosis often result in restenosis. Resveratrol (RSV), a red wine polyphenol, decreases vascular smooth muscle cell (VSMC) proliferation and increases endothelial nitric oxide synthase (enos) expression in vitro. RSV has been shown to decrease neointimal formation after arterial injury, however our understanding of the mechanism is still incomplete. Our main objective was to address the NO- dependence of this effect. Methods: RSV (4mg/kg/day,s.c.) alone or in combination with the NOS inhibitor N-Nitro-L-arginine methyl ester (L-NAME) (2mg/kg/day,s.c.) was given to Sprague-Dawley rats beginning at 3 days before arterial (carotid or aortic) injury. Vessels (carotid or aorta) were collected at 4, 7, 14, and 28 days after injury. Results: RSV reduced neointimal formation by 50% at 28 days (P<0.05) and decreased neointimal cell proliferation by 37% at 7 days (P<0.01) after injury. RSV also inhibited cell migration by 77% at 4 days after injury (P<0.001) but did not affect re-endothelialized area at 14 days in the aortic injury model. The RSV- induced decrease in neointimal area and cell proliferation was abolished by co- treatment with L-NAME while the RSV- induced decrease in cell migration was not. Conclusions: This is the first demonstration of RSV decreasing neointimal cell proliferation and neointimal growth through a NOS- dependent mechanism. These results support a vasculoprotective role for RSV and suggest that RSV may have clinical potential in the prevention and treatment of restenosis after angioplasty. 111

124 5.2 Introduction Vascular injury leads to pathologic repair and remodeling that involves VSMC migration and proliferation and ECM deposition, subsequently resulting in neointimal hyperplasia (reviewed in 17 ). Endothelial cell loss is a major contributing factor to the pathologic repair of the injured vessel. The disruption of endothelial integrity leads to an associated reduction in the production of vasculoprotective mediators and increased vasoconstrictor and growth- promoting substances, resulting in elevated vascular tone, platelet adhesion, enhanced inflammation, and medial VSMC proliferation 17. The resulting neointimal hyperplasia is the basis for restenosis after revascularization procedures such as angioplasty and stenting 13. Vascular endothelial cells produce NO, a vasodilator and the metabolic product of L- arginine by NOS, which plays a critical role in the regulation of vascular homeostasis. Because endothelial cell loss plays a key role in the pathogenesis of neointimal hyperplasia after vascular injury, a therapeutic strategy that promotes re- endothelialization of the injured vessels would inhibit lesion formation and thus improve vessel patency. Epidemiologic studies have demonstrated that the incidence of coronary artery disease in France is markedly lower as compared with other western countries with a fat containing diet. The French Paradox has been attributed to consumption of red wine in France (an inverse correlation between red wine consumption and incidence of cardiovascular disease) 324. RSV, a polyphenolic compound found in grape skins and red wine in high concentrations, has been suggested as one of the agents responsible for the French Paradox. Several biological activities of RSV have been reported, including anti- oxidant activity, cancer chemoprevention, anti- CVD and anti- diabetic properties, and most recently lifespan extension. RSV exerts cardiovascular protection by decreasing LDL oxidation, improving HDL cholesterol levels, inhibiting platelet aggregation and/ or adhesion, and decreasing VSMC proliferation or hypertrophy and ECM synthesis (reviewed in Bertelli et al 275 ). In particular, RSV has been shown to decrease 112

125 neointimal thickening in animal models of arterial injury , however only Gu et al 292 performed mechanistic studies, thus the mechanism(s) and signaling pathways by which RSV decreases neointimal formation are still incomplete. As discussed in the Introduction section 1.7, NO provides significant vascular protection and is produced by different isoforms of NOS enzymes. enos and nnos expression is constant and inos is induced by cytokines 144. enos activation diminishes neointimal formation after arterial injury 152 and as expected, enos deficiency results in greater neointimal growth 104, 195. NOS activation represents a likely mechanism for RSV to exert its inhibition on neointimal formation, since RSV enhances expression and activity of enos in endothelial cells 338 and EPCs 292. However, NO has not been demonstrated to play a direct role in the effect of RSV to decrease neointimal growth in vivo. Therefore, the objective of our study is to determine the effect of RSV on VSMC growth kinetics over the time course of neointimal formation after balloon catheter injury of the carotid artery in rats. Furthermore, since RSV can stimulate NOS and NO can decrease neointimal growth, we investigated the consequences of NOS inhibition on the effects of RSV. 5.3 Materials and Methods Animal Models Six groups of male Sprague- Dawley rats weighing g and fed normal chow were studied: 1) control (C); 2) RSV (4mg/kg/day, s.c.)-treated (R); 3) RSV + NG-nitro-L-arginine methyl ester (L-NAME) (2mg/kg/day, s.c.) (R+L2); 4) RSV + L-NAME (10mg/kg/day, s.c.) (R+L10); 5) control + L-NAME (2mg/kg/day, s.c.) (C+L2); and 6) control + L-NAME (10mg/kg/day, s.c.) (C+L10). Both doses of L-NAME have been demonstrated to inhibit NO production in the rat aorta, based on vessel cgmp assays 368, where only the 10mg/kg/day dose of 113

126 L-NAME elevated blood pressure 369. The animals were maintained as described in the General Methods section Surgical Procedures Three days after onset of treatment, injury to either the LCCA or the aorta of the rat was induced by means of an inflated balloon catheter 133. Refer to the General Methods section 2.2.4/ for details regarding surgical procedures Vessel and Blood Sample Collection The rats were sacrificed in either the fed or fasted state at various time points after injury, based on previous studies investigating the kinetics of vascular injury. Blood samples were collected by cardiac puncture under general anesthesia. Thereafter, rats were sacrificed via anesthetic overdose and the carotids were perfusion- fixed for four minutes with 10% buffered formalin at physiologic pressure ( mmHg). The carotids were removed, immersed in 10% formalin for >24h, and embedded in paraffin. Refer to General Methods section for a complete description Metabolic Parameters Blood glucose was monitored throughout the treatment (see General Methods section 2.3). Plasma insulin was determined by radioimmunoassay using kits specific for rat insulin. Refer to General Methods section 2.12 for details Histomorphometry/Matrix Staining Cell migration into the neointima was measured four days after injury and neointimal cells on the lumen surface were immunostained with an antibody against histone H1 82. The number of cells on the luminal surface of the vessel was counted. Refer to General Methods section for complete details. Neointimal cell proliferation was measured at seven days after injury with an anti-brdu antibody, as previously described 82 as neointimal cell proliferation 114

127 reaches a maximum one week after vascular injury 383. See General Methods section for details. Neointimal cell apoptosis (14 days after injury) was measured by terminal dutp nickend labeling (TUNEL) as described in General Methods section Cross- sectional areas were measured on hematoxylin and eosin stained slides at 14 and 28 days after injury. Crosssections taken 28 days after injury were also stained with Movat s pentachrome or PSR dye for elastin and collagen respectively. Refer to the General Methods section for a detailed description of staining and analysis Evan s Blue Staining Re- endothelialized area was measured at 14 days after balloon injury and visualized with Evan s Blue dye. Following perfusion, the thoracic aorta was dissected out and opened longitudinally to observe Evan s Blue uptake macroscopically (endothelial cells do not stain) (for details see General Methods section 2.5.5). Cross- sectional areas were measured on hematoxylin and eosin stained slides from unopened sections of these aortas as described in the General Methods section Vessel Cannulation Three days before the hyperinsulinemic- euglycemic clamp, rats were anesthetized with isoflurane and indwelling catheters were inserted into the right internal jugular vein for infusion and the LCCA for blood sampling, as described in the General Methods section Hyperinsulinemic- Euglycemic Clamp After an overnight fast, insulin (5mU/kg.min or 20mU/kg.min) and glucose (variable rate) were infused through the jugular vein for two hours to maintain euglycemia while blood samples were taken from the LCCA for plasma glucose measurements. Refer to the General Methods section 2.10 for a detailed description of the clamp procedure. 115

128 5.3.9 Blood Pressure Analysis Systolic blood pressure was measured indirectly from the tail via the cuff technique, using the Visitech BP-2000 Blood Pressure Analysis System (see General Methods section 2.11 for details). Blood pressure was measured before treatment and throughout at 4, 14, and 28 days after arterial injury Statistical Analysis Values are expressed as mean ± SEM. Statistical analysis was performed using analysis of variance (ANOVA) to test for significant differences between groups and the post hoc comparisons were performed with Tukey s t- test. Significance was accepted at P< Results Metabolic Parameters Food, fluid, and total caloric intake were unchanged for RSV- treated rats compared to controls at all time points (Table 5.1). There was no overall effect of treatment on weight change although the C+L2 group gained more weight than controls at seven days. Plasma glucose levels taken over the treatment period were the same in RSV- treated rats and control rats, however plasma glucose was decreased in the R+L2 group at 14 days compared to RSV alone and in the R+L2 and C+L2 groups at 28 days compared to control or RSV alone (Table 5.1). Plasma insulin levels taken at sacrifice were similar in all groups (Table 5.1) Morphometric Measurements Treatment with RSV for both 14 and 28 days after balloon injury resulted in decreased neointimal formation (Figure 5.1). This effect was completely blocked by co- treatment with the NOS inhibitor L-NAME at both doses, indicating an NO- dependent effect of RSV. By contrast, there was no change in medial area between the groups (14 days: C=0.17±0.01mm 2, R=0.17±0.01mm 2, R+L2=0.16±0.01mm 2, C+L2=0.17±0.02mm²; 28 days: C=0.15±0.01mm 2, 116

129 R=0.15±0.01mm 2, R+L2=0.17±0.01mm 2, R+L10=0.17±0.01mm², C+L2=0.17±0.01mm 2, C+L10=0.17±0.01mm²). In addition, there were no differences in external elastic lamina perimeter between the groups at 14 (C=3.0±0.1mm, R=2.8±0.2mm, R+L2=3.0±0.1mm, C+L2=3.0±0.2mm) and 28 days (C=2.8±0.1mm, R=2.8±0.2mm, R+L2=3.1±0.2mm, R+L10=3.0±0.1mm, C+L2= 3.0±0.1mm, C+L10=2.9±0.1mm), indicating that there is no vessel remodeling. Neointimal cell proliferation measured at seven days after vessel injury was decreased with RSV treatment (P<0.01) (Figure 5.2). NOS inhibition with L-NAME abolished the effect of RSV to decrease neointimal cell proliferation. There was no significant difference in the percentage of apoptotic cells in the neointima between control and RSV- treated rats at 14 days after vessel injury (Figure 5.3). To determine whether the inhibitory effects of RSV on intimal area at 28 days was due to an effect on ECM accumulation, vessel cross- sections were stained with Movat s pentachrome for elastin or PSR for collagen. RSV significantly decreased elastin accumulation by ~30% (P<0.05) and had a slight tendency to decrease collagen accumulation by ~22% (P=0.40) in the intima of the carotid at 28 days compared to control (Figure 5.4). Furthermore, this effect was prevented by co- administration with L-NAME. To investigate whether RSV decreased neointimal growth by inhibiting migration, cell migration to the neointima was measured four days after vessel injury. RSV markedly inhibited cell migration by 76%, reducing the number of neointimal cells from 555±59cells/mm 2 in control rats to 131±27cells/mm 2 (P<0.001) (Figure 5.5). L-NAME did not alter the effect of RSV to decrease migration (R+L2=70±12cells/mm 2 ). Analysis of rat aorta specimens after Evan s Blue staining at 14 days showed that RSV treatment did not affect re- endothelialization of the balloon- injured arterial segments compared 117

130 to control (Figure 5.6A, B). Despite this, RSV reduced neointimal thickening in the aorta after balloon injury (Figure 5.6C) (P<0.05), as was also observed in the carotid artery. Furthermore, no changes were observed in systolic blood pressure with either RSV or low dose L-NAME (2mg/kg/day) treatment over the course of treatment (Figure 5.7). However, as expected high dose L-NAME (10mg/kg/day) alone and in combination with RSV elevated systolic blood pressure during the course of treatment to 28 days after carotid injury (P<0.05) Insulin Sensitivity Measurements RSV has insulin- like and insulin- sensitizing effects. For example, RSV stimulates glucose uptake 279 in a skeletal muscle cell line and in vivo RSV has been shown to protect against high- fat diet induced insulin resistance in mice 282. Preliminary data from our laboratory show that RSV abolishes FFA- induced insulin resistance on glucose uptake in muscle and liver. Furthermore we have previously shown that insulin decreases neointimal growth after carotid balloon injury 133. Taken together these data suggest that RSV may decrease neointimal growth indirectly through an increase in insulin sensitivity. Therefore we used a hyperinsulinemiceuglycemic clamp to measure insulin sensitivity in overnight fasted rats after the 28 day treatment with RSV. The 20 mu/kg.min and 5 mu/kg.min insulin infusion rates were used to assess maximum and half- maximum glucose disposal (glucose infusion rate) respectively. The glucose infusion rate was unchanged with RSV treatment (Figure 5.8), indicating that RSV did not affect insulin sensitivity. 5.5 Discussion The present study focused on the in vivo effect of RSV on the kinetics of neointimal growth in the rat arterial injury model and the possible involvement of NO in RSV action. In our study, RSV had a marked inhibitory effect on injury- induced migration and proliferation of 118

131 neointimal cells, leading to decreased neointimal area. Furthermore, NO appears to play a causal role in the protective effect of RSV on vascular growth. The inhibitory effect of RSV on both proliferation and subsequent neointimal formation appear to be NO- dependent. However, the decrease in cell migration by RSV was unaffected by L-NAME. Another mechanism by which RSV could inhibit cell migration is by inactivating MAPK, one of the major intracellular signaling pathways critical for migration, as RSV is well known to inhibit MAPK activation in many types of cells (reviewed in 384 ). However, as the reduction in neointimal growth and cell proliferation by RSV is completely abolished by NOS inhibition, the inhibitory effect of RSV on cell proliferation may be the primary mechanism by which RSV reduces intimal hyperplasia, regardless of the inhibitory effect on cell migration. In addition to suppression of proliferation, RSV can also induce apoptosis 337, 385, which could result in a decrease in neointimal formation. However, we did not observe any changes in apoptosis with RSV treatment at 14 days. The ECM, which is primarily composed of elastin and collagen, is a major component of the neointimal layer and influences neointimal thickening. RSV has been shown to inhibit AGEinduced collagen synthesis in VSMCs 386, so we wished to see if RSV affected matrix deposition in our model. Elastin staining was significantly decreased with RSV treatment compared to control at the 28 day time point, and there was a tendency for RSV to decrease collagen staining, suggesting that a decrease in matrix deposition may contribute to the net reduction in neointimal growth. Furthermore, NO also affects matrix elements. For example, L- arginine, the substrate for NO production, decreased hyaluronan synthase expression which is an enzyme responsible for making hyaluronan (a component of the ECM) 182. NO has also been shown to inhibit basal type I collagen levels 183 and collagen synthesis in vitro 184. In our study the effect of RSV to decrease elastin and collagen deposition was prevented with NOS inhibition, again supporting a role of NO in mediating RSV s protective effects on the vasculature. 119

132 Re- endothelialization can decrease neointimal growth and RSV has been shown to accelerate re- endothelialization at 10mg/kg/day and have no effect at 50mg/kg/day administered by oral gavage 292. Given the ~40% bioavailability of RSV in rats 387 and the unknown activity of RSV metabolites after oral administration 387, our effective dose was possibly lower than the effective dose provided by 10mg/kg/day oral gavage. However, this could explain why we did not see effects on re- endothelialization. Since RSV decreases neointimal growth without delaying re- endothelialization or increasing re- endothelialization, depending on dose, RSV treatment could provide some advantage over the anti- mitogenic agents (rapamycin and paclitaxel) currently used which delay re- endothelialization 8. Insulin resistance may be a pathogenic factor for endothelial dysfunction possibly through impaired enos activity and increased oxidative breakdown of NO 388. Several diabetic animal models have shown that RSV can activate AMPK and improve insulin sensitivity 281, 282. Furthermore, we have previously shown subcutaneous insulin treatment to decrease neointimal formation after arterial injury 133. Therefore, we wished to address the possibility of RSV improving insulin sensitivity in order to decrease neointimal growth. We found no effect on whole body insulin sensitivity as indicated by the glucose infusion rate during a hyperinsulinemic- euglycemic clamp. Nonetheless, as RSV has been shown to improve insulin sensitivity 281, 282 in models of insulin resistance and its mechanism of action is Akt- independent, it is likely that the inhibitory effect of RSV on neointimal formation would remain in an insulin resistant state. This would provide an advantage over insulin as insulin s inhibitory effect on neointimal growth was shown to be prevented by high- fat feeding 133. Hypertension is among the cluster of disorders present in the metabolic syndrome 353 and available studies have demonstrated that RSV may improve blood pressure in models of hypertension 312. However, RSV (10mg/kg/day; gavage) treatment alone for up to six weeks did not have an effect on systolic blood pressure 312, which is in agreement with our study. As 120

133 previously mentioned, L-NAME, at specific doses, blocks NO- mediated vasodilatation. However, we purposely used a dose (2mg/kg/day) which was previously shown not to affect blood pressure 369. Although not measured in this study, both the 2mg/kg/day and 10mg/kg/day dose of L-NAME have been demonstrated to inhibit NOS based on vessel cgmp assays 368. In addition, high dose L-NAME (10mg/kg/day) alone or in combination with RSV, increased systolic blood pressure, as expected, and also prevented RSV s inhibitory effect on neointimal growth. In our study we did not investigate the specific NOS isoform involved in RSV action as L-NAME is a general NOS inhibitor. In previous in vitro studies RSV increased enos expression 301, 312, 385, and in vivo enos expression was enhanced with an oral dose of RSV ( mg/kg/day) 292 that should result in comparable effects to our s.c. dose. Therefore, we hypothesize that enos is likely the isoform activated by RSV to inhibit neointimal formation. Unfortunately, the currently available NOS inhibitors do not have specificity for enos. The signaling pathway upstream of NOS by which RSV decreases neointimal formation is not clear. RSV activates AMPK 282 and AMPK has been demonstrated to phosphorylate and activate enos in endothelial cells 306, 389 and cardiac myocytes 306. AICAR, a well known AMPK activator, decreases neointimal formation after arterial injury 295. Taken together, this suggests a potential role of AMPK activation in our model, possibly upstream of NOS. Although this has recently been disputed 284, many studies indicate that RSV is an activator of sirtuins, mainly SIRT1, which are histone/protein deacetylases shown to be involved in lifespan extension 390. Our previous studies in vitro demonstrate that RSV stimulates SIRT1 to activate AMPK, since SIRT1 inhibition abolished RSV- stimulated AMPK phosphorylation 279. There is also evidence suggesting that SIRT1 is upstream of enos as SIRT1 has been shown to deacetylate and activate the enos enzyme to promote endothelium- dependent vascular relaxation 391. Therefore, activation of SIRT1 may help to explain the action of RSV. 121

134 In conclusion, we have shown for the first time that the effect of RSV to decrease neointimal formation after arterial injury is NOS- dependent. These results provide evidence for inhibition of the growth response to vessel injury by RSV in vivo and suggest that RSV treatment may offer protection against restenosis. 122

135 Table 5.1. Daily food, fluid and total caloric intake, final weight gain, fed plasma glucose and insulin levels over the treatment periods. Treatment Groups 4 days C R R+L2 C+L2 (n=6) (n=9) (n=9) (n=7) 7 days C R R+L2 C+L2 (n=6) (n=7) (n=6) (n=6) 14 days C R R+L2 C+L2 (n=7) (n=7) (n=7) (n=7) 28 days C R R+L2 C+L2 (n=9 ) (n=10) (n=10) (n=9) Food Intake (g/day) 27±1 25±1 24±1 22±1 29±1 28±1 27±1 33±3 27±1 28±1 27±1 29±1 28±1 28±1 28±1 29±1 Fluid Intake (g/day) 32±2 32±2 28±1 28±1 34±2 37±4 29±2 44±3 33±1 34±1 33±2 35±3 37±2 35±3 32±2 31±2 Total Caloric Intake (kcal/g/day) 89±3 83±3 79±4 73±3 94±4 92±7 87±4 109±8 89±2 92±2 89±3 96±2 92±3 92±2 92±2 96±3 Weight Change (g) -1±6 9±5 6±6-3±3 5±5 7±5 12±8 29±5* 34±6 46±6 36±4 44±13 88±11 81±10 86±9 83±10 Fed Plasma Glucose (mm) 6.2± ± ± ± ± ± ± ± ± ± ±0.2# 6.2± ± ± ±0.1**## 6.1±0.1*# Fed Plasma Insulin (pm)¹ 89±9 132±10 99±12 146±32 150±12 178±23 159±16 132±29 142±17 176±18 171±30 102±8 108±17 132±17 138±17 149±17 C=Control; R=Resveratrol-treated; L2=L-NAME (2mg/kg/day)-treated. Data are expressed as means ± SEM. **P<0.01, *P<0.05 vs. C; ##P<0.01, #P<0.05 vs. R. Data are means ± SEM of individual average glucose levels taken over the treatment period. ¹Data are expressed as means ± SEM and represent values taken at sacrifice. 123

136 A C R B Intimal Area (mm 2 ) C+L2 C R R+L2 C+L2 R+L10 C+L10 R+L2 # ### * 14 days 28 days Figure 5.1. The effect of RSV and NOS inhibition on intimal area. (A) Representative photomicrographs of cross-sections taken at 28 days after carotid injury (x400). (B) Crosssectional areas of the intima of carotid arteries measured at 14 (C: n=7; R: n=7; R+L2: n=7; C+L2: n=7) and 28 (C: n=9; R: n=10; R+L2: n=10; R+L10: n=8; C+L2: n=9; C+L10: n=8) days after carotid injury. Values are means ± SEM. *P<0.05 vs. C, ### P<0.001 vs. R+L2, # P<0.05 vs. R+L10 at 28 days. 124

137 % BrdU-Labelled Cells C R R+L2 C+L2 ** ### 7 days Figure 5.2. The effect of RSV and NOS inhibition on cell proliferation. Cell proliferation expressed as % BrdU-labeled cells in the intima at 7 days after carotid injury (C: n=6; R: n=7; R+L2: n=6; C+L2: n=6). Values are means ± SEM. **P<0.01 vs. C, ### P<0.001 vs. R+L2. 125

138 % TUNEL Positive Cells Control Resveratrol 14 days Figure 5.3. The effect of RSV on cell apoptosis. Intimal cell apoptosis expressed as % TUNEL positive cells in the intima at 14 days after carotid injury (C: n= 6; R: n=7). No significant differences were found. 126

139 A C C R C R C+L2 R+L2 C+L2 R+L2 B D % Intimal Area Elastin Positive C R R+L2 C+L2 * ### * 28 days % Intimal Area Collagen Positive C R R+L2 C+L2 28 days Figure 5.4. The effect of RSV and NOS inhibition on elastin and collagen content. Representative images of staining with (A) Movat s pentachrome to show elastin and (C) PSR to show collagen in the intima at 28 days after carotid injury. Quantitative analysis of cross-sections stained for elastin (B) or collagen (D) (C: n=9; R: n=10; R+L2: n=10; C+L2: n=9) days after carotid injury. Values are means ± SEM. *P<0.05 vs. C, ### P<0.001 vs. R+L2 at 28 days. 127

140 A C R C+L2 R+L2 B Number of cells/mm *** C R R+L2 C+L2 *** 0 4 days Figure 5.5. The effect of RSV and NOS inhibition on cell migration. (A) Representative photomicrographs showing cells on the lumen surface at 4 days after carotid injury (x400). (B) Quantitative analysis of cell migration (C: n=6; R: n=9; R+L2: n=9; C+L2: n=7). Values are means ± SEM. ***P<0.001 vs. C. 128

141 A B C Control % Re-endothelialization Intimal Area (mm 2 ) Control Resveratrol 14 days 14 days Resveratrol Control Resveratrol * Figure 5.6. The effect of RSV on re-endothelialization. (A) Representative images of Evan s Blue dye staining of aorta at 14 days after vessel injury. (B) Quantification of re-endothelialized area (% of total vessel area). (C) Cross-sectional areas of the intima in the aorta at 14 days after vessel injury (C: n=8; R: n=9). Values are means ± SEM. *P<0.05 vs. C. 129

142 Systolic Blood Pressure (mmhg) *, ## R+L10 vs C, R # C+L10 vs R *** C+L10 vs C * R+L10 vs C C R R+L2 R+L10 C+L2 C+L10 **, ## C+L10 vs C, R *, # R+L10 vs C, R Days Figure 5.7. The effect of RSV and NOS inhibition on systolic blood pressure. Systolic blood pressure at 3 days before and 4, 14, and 28 days after carotid injury (corresponding to day 0, 7, 17, and 31 of treatment respectively) (C: n=10; R: n=10; R+L2: n=10; R+L10: n=8; C+L2: n=8; C+L10: n=9). Values are expressed as mean ± SEM. Values for each treatment group were compared at each time- point. *P<0.05, **P<0.01, ***P<0.001 vs. C, # P<0.05, ## P<0.01 vs. R. 130

143 Glucose Infusion Rate (umol/kg/min) Control Resveratrol 5mU.kg -1 min -1 20mU.kg -1 min -1 Insulin Infusion Rate Figure 5.8. The effect of RSV on whole body insulin sensitivity. Whole body insulin sensitivity is indicated by the glucose infusion rate during the last 30 min of the 2h hyperinsulinemic- euglycemic clamp (C: n=5; R: n=6). Values are expressed as mean ± SEM. No significant differences were found. 131

144 6 Summary and Conclusions 6.1 Summary of Each Study in the Thesis In our previous studies 133, 134, rats on insulin treatment and oral glucose to avoid hypoglycemia had reduced neointimal growth after arterial injury. However, plasma glucose in the insulin- treated rats was lower than normal 133 and the effect of oral glucose remained undetermined. In a separate study by another group 219 oral sucrose was used instead of glucose to maintain euglycemia with insulin treatment, however under these conditions neointimal growth was increased, unlike in our studies. In Study 1 the effects of insulin, where normal plasma glucose levels were maintained, and oral glucose or sucrose were investigated. Insulin treatment reduced neointimal area after injury in rats receiving oral glucose but not in those receiving oral sucrose. Oral glucose, without insulin, had no effect on neointimal formation whereas oral sucrose alone increased neointimal growth. Furthermore, oral sucrose but not oral glucose decreased insulin sensitivity measured with hyperinsulinemic clamps. Insulin has protective effects on the endothelium by stimulating NO production and enos expression 235, 236 but also has mitogenic effects on VSMCs in vitro 266. Furthermore, NOS inhibition has been reported to abolish the effect of insulin to inhibit VSMC migration in vitro 272. In Study 2, we examined whether NOS was a mediator of insulin s inhibitory effect on neointimal formation. Therefore, we investigated the consequences of NOS inhibition, with the in vivo inhibitor L-NAME, on the effects of insulin. Insulin decreased neointimal area, ECM 132

145 production, and cell migration. These effects were abolished by the addition of L-NAME. Furthermore, insulin increased re- endothelialization, which was also completely prevented by co- administration of L-NAME. In addition, insulin increased the expression of VSMC differentiation markers, the number of circulating progenitor cells, and tended to increase enos protein expression in the vessel. RSV has many vasculoprotective effects, including decreasing LDL oxidation, increasing HDL cholesterol, inhibiting platelet aggregation and/ or adhesion, and suppressing VSMC proliferation and ECM synthesis (reviewed in Bertelli et al 275 ), several of which are NOdependent (see Introduction section 1.11 for details). Furthermore, RSV has been shown to decrease neointimal formation after arterial injury , however the mechanism of the effect or the signalling molecules involved remain unclear. In Study 3, our objective was to investigate the mechanism of RSV action and specifically address the NO- dependence of RSV s inhibitory effect on neointimal formation. RSV reduced neointimal growth, ECM production, intimal cell proliferation, and migration after injury. However, RSV did not affect apoptosis or reendothelialization. The RSV- induced decrease in intimal area, ECM production, and cell proliferation was abolished by co- treatment with L-NAME while the RSV- induced decrease in cell migration was not. Furthermore, no changes in insulin sensitivity or systolic blood pressure were observed with RSV treatment. 6.2 General Summary By using the carotid balloon injury model of restenosis, the studies in this thesis have demonstrated that insulin (independent of any glycemic effects) and RSV have a protective effect on the vessel against restenosis and that their effects are mediated by NOS. 133

146 6.3 General Conclusion The results of the studies in this thesis indicate that NO plays a role in mediating the protective effect of insulin and RSV on VSMCs and endothelium in vivo to decrease neointimal growth. These results suggest that systemic RSV may have clinical potential in the prevention of restenosis. Systemic insulin cannot be given to non- diabetic patients outside of a supervised facility (i.e. hospital) because of the risk of hypoglycemia. Since both insulin and RSV decrease neointimal formation without negatively impacting re- endothelialization, insulin or RSV treatment could provide some advantage over the anti- mitogenic agents (rapamycin and paclitaxel) currently used in drug- eluting stents, which delay re- endothelialization. 134

147 7 General Discussion 7.1 Is the Effect of Insulin Protective or Detrimental in a Model of Restenosis? Although our understanding of the effect of insulin on restenosis and atherosclerosis has improved, it remains a point of controversy whether insulin itself is vasculoprotective. Several different models have been used in rats to simulate the effect of hyperinsulinemia/ insulin resistance and examine neointimal formation; the fructose- fed rat 213, the obese Zucker rat 212, and the ZDF rat 214. The general consensus from these studies was that these rats had greater neointimal formation, however it is not clear whether this is due to the elevated level of endogenous insulin or to insulin resistance (insulin resistance is thought to be primary and cause hyperinsulinemia), in these models as indicated by the tendency of blood glucose to increase. As a result, our laboratory has focused on examining the relatively short- term effect of exogenous insulin, without insulin resistance, as indicated by the need for treatment with oral glucose. An earlier study in our laboratory determined the effect of insulin on vascular growth after arterial injury in rats and found that intimal area was decreased in insulin- treated rats on a normal diet, however the effect of insulin was diminished by high- fat feeding 133. These data suggest that the predominant effect of insulin was to decrease neointimal growth and that insulin resistance, not insulin, is detrimental in the pathogenesis of restenosis. In contrast to our studies, insulin has also been reported to enhance neointimal formation after vascular injury in normal rats 219. These rats were given sucrose water to maintain 135

148 normoglycemia instead of glucose water as in our studies. Therefore, in Study 1 we attempted to determine the reasons for these differences in results. We eliminated any glucose lowering by i.p. glucose infusion while using a high insulin dose (5U/day) (decreased neointimal growth), and included a group receiving oral glucose alone, which did not affect neointimal growth. Oral sucrose is commonly used to induce insulin resistance in rats 359, 360, which was confirmed in our study, and not surprisingly, insulin treatment given with oral sucrose was not effective in decreasing neointimal growth. Thus, we believe that the increase in neointimal growth observed when insulin was combined with oral sucrose in Foster s paper 219 was due to sucrose, not to insulin. In Study 2, and also in a recently published study by our laboratory 134, we used a lower dose of insulin, which allowed us to investigate the effect of insulin concentrations that can be seen also in non- insulin resistant rats after an oral glucose load, while avoiding glucoselowering. Similar to the results of Study 1, insulin decreased neointimal area. This would be in keeping with a vasculoprotective effect of insulin. However, Pu et al 221 have shown that insulin increases neointimal formation. Furthermore, in Indolfi et al 216 insulin treatment via islet transplantation increased neointimal growth. One of the major differences between our studies and those of Pu s 221 and Indolfi s 216, is that in the latter studies insulin was given long before arterial injury (3-4 weeks) whereas in our studies insulin treatment prior to injury was relatively short- term (2-3 days). Therefore, a possible explanation of the discrepancy with our findings is that the prolonged treatment with insulin may have resulted in some insulin resistance, as sustained high levels of insulin have been shown to upregulate a negative feedback loop resulting in inhibition of IRS signalling by mtor/ S6K Thus insulin resistance could then block insulin s vasculoprotective effects. Furthermore, insulin also activates a growth- factor like pathway mediated by MAPK, which remains intact in insulin resistant states. It is possible that the down- regulation of the PI3K- 136

149 mediated signalling pathways and continued activity of the insulin- stimulated MAPK signalling pathway, may explain the increase in neointimal formation in some studies. In Study 1 and 2 we did not measure insulin sensitivity in our insulin treatment groups because basal plasma insulin levels were significantly elevated (close to those obtained during the clamp), which interferes with the interpretation of the data from the hyperinsulinemic- euglycemic clamp. In future studies the clamp could be performed soon after the insulin implant is removed, which would reduce basal insulin levels to control. It is likely that the insulin treatment induced some insulin resistance in these groups, however fasting plasma glucose levels were similar in our insulintreated groups to controls despite oral glucose/ sucrose, suggesting a prevalent state of excess rather than deficiency of insulin action. Taken together, it appears that under conditions of normal insulin sensitivity, insulin treatment decreases neointimal growth, whereas under conditions of insulin resistance (including insulin- induced insulin resistance) neointimal growth is increased. In our two models of insulin resistance (high fat diet 133 and sucrose feeding) insulin given at the time of arterial injury was not detrimental. However, in Study 1 we have observed an effect of insulin to increase neointimal cell proliferation which remained intact under conditions of insulin resistance. Therefore, we cannot exclude that in other models of insulin resistance and under different experimental conditions, this effect may prevail and lead to neointimal hyperplasia. Nevertheless, the results of Study 1 and 2 support a protective role of insulin in the vasculature, independent of any glucose- lowering effect, as indicated by a reduction in neointimal formation after arterial injury with insulin administration. Furthermore, oral glucose used to maintain normoglycemia did not affect neointimal formation whereas oral sucrose increased it, which may be explained by sucrose- induced insulin resistance. 137

150 7.2 The Protective Effects of Insulin or Resveratrol Against Mechanically- Induced Restenosis: How do They Differ? In Study 2 and Study 3 we investigated the role of NO in the effect of insulin and RSV to decrease neointimal formation, respectively. The results are summarized in Figure 7.1. In these studies, both insulin and RSV inhibited neointimal formation and cell migration. However, only RSV decreased cell proliferation as in a previous study 134 we found that insulin did not affect proliferation. Insulin and RSV both inhibit neointimal growth through a NO- dependent mechanism, despite this, the role of NO in their mechanism of action appears to differ. For example, insulin s inhibitory effect on cell migration is NO- dependent whereas RSV s is not. In addition to NO, MAPK is also associated with increased VSMC migration. Therefore, it is not surprising that insulin action involves NO, as insulin is generally thought to activate MAPK or at least not inhibit it. In contrast, as RSV is well known to inhibit MAPK activation in many types of cells (reviewed in 384 ), it is likely that RSV could inhibit cell migration by inactivating MAPK. NFκB is also associated with increased VSMC migration 73. Therefore, insulin and/ or RSV may also decrease cell migration through an anti- inflammatory effect. Previously, we found that insulin did not decrease plasma cytokines 134, however an effect on inflammatory markers at the level of the vessel has yet to be determined. As previously mentioned, insulin decreases cell migration and neointimal growth but not cell proliferation 134 in this model. This suggests a significant role of cell migration inhibition by insulin to decrease neointimal formation. However, unlike insulin, RSV had an inhibitory effect on both cell migration and proliferation, and subsequent neointimal formation. Furthermore, RSV s inhibitory effect on cell migration was NO- independent whereas the reduction in neointimal growth and cell proliferation by RSV is completely abolished by NO inhibition, implying that RSV s effect on proliferation is more important than migration. Taken together, 138

151 this suggests that inhibition of cell migration may be more important for insulin action than for RSV, and that the inhibitory effect of RSV on cell proliferation may be the primary mechanism by which RSV reduces neointimal growth, regardless of the inhibitory effect on cell migration. Insulin and RSV also have differing effects on re- endothelialization. Insulin stimulated re- endothelialization, whereas RSV did not affect re- endothelialization. However, other authors have shown that RSV accelerates re- endothelialization at a dose higher than that used in Study 3. Thus, RSV at least does not delay re- endothelialization. After angioplasty, reendothelialization also occurs from circulating precursors 393, 394, and the fact that insulin increased the circulating precursors in Study 2 and RSV has been reported to increase circulating EPCs 292, suggests an additional benefit for both. These data taken together suggest that insulin or RSV could provide some advantage over the anti- mitogenic agents currently used in stents, which delay re- endothelialization 8. As an additional benefit, both insulin 131, 377, at physiological concentrations, and RSV 314 have been reported to have inhibitory effects on platelet aggregation in vivo, which initiates thrombosis. Therefore, it seems that insulin and RSV have potential for experimentation in a drug- eluting stent, which ideally would inhibit restenosis and thrombosis while at the same time protecting the re- endothelialization process. The upstream signaling molecules activated by insulin and RSV are also likely to be distinct. It is well accepted that stimulation of enos 234 and NO production by insulin is PI3K/ Akt- dependent 235, 236. Less is known regarding the signaling molecules upstream of enos in RSV action. RSV has been reported to activate AMPK 282 and AMPK has been demonstrated to phosphorylate and activate enos 306, 389. RSV also activates SIRT1 and studies suggest that SIRT1 may be upstream of AMPK 279 and/ or enos 391. The difference between insulin and RSV concerning their signaling pathways may be most relevant in a state of reduced insulin sensitivity. In this state, the insulin- stimulated PI3K/ Akt signaling pathway (which is generally thought to be responsible for the vasculoprotective effects of insulin) is blocked. As RSV 139

152 signalling is likely not to depend on Akt activation, instead probably involving AMPK, RSV could still be effective in states of reduced insulin sensitivity, thereby offering an advantage over insulin in a clinical setting, which will be further addressed in the Future Directions section The Role of NO in Insulin and RSV Action: What is the Source of NO? In the rat arterial balloon injury model used in this thesis, there are several types of cells near the site of injury that express NOS and could therefore be potential sources of NO, however the specific NOS isoform expressed differs between cell types. The cell types that primarily express enos include endothelial cells (in an injured vessel there are always endothelial cells along the perimeter of the injured area 146 and as the repair response progresses the endothelial cell layer is restored), circulating EPCs 20, and platelets 147. In addition, monocyte- derived macrophages express inos 395 and VSMCs express both inos and nnos 148, 379. It is generally accepted that enos and inos induction after arterial injury is beneficial 155, although the effect of inos may be more variable because it produces the greatest amount of NO and thus can produce superoxide, which is detrimental, when substrates and cofactors for synthesis of NO are lacking 158. The primary finding of Study 2 and Study 3, through the use of the in vivo NOS inhibitor L-NAME, is that stimulation of NO production is the mechanism through which insulin and RSV decrease neointimal formation, respectively. Therefore, the question arises as to which NOS isoform is specifically being affected by insulin or RSV. Insulin stimulates inos 239 in VSMCs. VSMCs are one of the most abundant cell types present particularly during the early stages of the repair response and they express inos 148, 379. Thus representing a potential significant source for insulin- stimulated NO production. Inflammatory cells also express significant amounts of inos, and insulin exerts antiinflammatory effects in vitro and in vivo 132 but we found very few cells positive for CD68 (macrophage marker) or CD45 (general hematopoietic cell marker) after carotid injury 134. This 140

153 suggests that inflammatory cells are not a significant source of NO in our study. Furthermore, we were unable to detect inos in the carotid artery via Western blot analysis or RT-PCR. However, insulin is also well known to stimulate enos 234 to increase NO production in endothelial cells 235, 236. We have demonstrated that insulin increases the number of circulating precursor cells (which could potentially differentiate into endothelial cells) and promotes re- endothelialization. EPCs and endothelial cells are sites of enos- dependent NO production. In addition, we have also shown that enos protein expression tended to be increased with insulin treatment in the injured vessel. Taken together, these data indicate that enos is the isoform most likely involved in insulin action, however we cannot rule out the possibility of inos- dependent NO production in the VSMCs. The effect of RSV on the different NOS isoforms is similar to that of insulin. RSV increases the expression and activity of enos in endothelial cells 300 and EPCs 292. RSV decreased inos expression in macrophages 302, but also enhanced cytokine- induced inos expression in rat VSMCs 303. Furthermore, increased expression of both enos and inos in the heart 301 was observed in mice treated with RSV. According to the data in Gu et al 292 there may be a stronger case for RSV action to involve enos activation. In the same rat arterial balloon injury model used in this thesis, Gu et al 292 demonstrated that a high dose of RSV increased re- endothelialization and the number of circulating EPCs, both sites of enos- mediated NO production. Furthermore, enos mrna and protein expression were enhanced with RSV in the aorta after arterial injury. However, our studies (Study 3) may support a role of inos instead of enos, as RSV did not increase re- endothelialization at the dose used in our study. To address the specific NOS isoform involved in insulin or RSV action, immunohistochemistry for enos or inos on carotid artery cross- sections could be performed followed by studies in a mouse arterial injury model using enos or inos knockout mice. 141

154 Alternatively, perhaps insulin- or RSV- stimulated NO production is not limited to activation of a single NOS isoform. It is possible that during the earlier stages of the repair response, before EPC recruitment and re- endothelialization occur, insulin or RSV primarily activate inos in VSMCs to increase NO production and then over time switch to enosdependent NO production in endothelial cells. Ultimately, regardless of the source of NO, the main message is that insulin and RSV decrease neointimal formation through a NO- dependent mechanism. Subsequent studies could specifically address the isoform involved and will be discussed in the Limitations of the Studies section

155 A B Figure 7.1. Summary of the effects of insulin and RSV in the vessel after arterial balloon injury. (A) Insulin (3U/day) decreased neointimal growth, ECM accumulation, and cell migration, and promoted reendothelialization, all through a NO- dependent mechanism. VSMC differentiation and the number of circulating progenitor cells were increased, however whether these effects are mediated by NO are not yet determined. (B) RSV decreased neointimal growth, ECM accumulation, and cell proliferation through a NO- dependent mechanism. RSV also decreased cell migration, however this effect was NO- independent. 143

156 8 Limitations of the Studies 1. The cell migration assay used in Study 2 and 3 allows us to measure intimal cell migration in vivo, however it cannot distinguish cell origin or type 45. In a previous study 134, we identified very few macrophages or hematopoietic cells in general, in the medial and intimal layers of control rats after arterial injury, which is consistent with studies by other groups 17. Therefore, inflammatory cells are not likely to play a significant role in this model. However, we cannot rule out the possibility of inflammatory cells adhering to the luminal surface immediately following arterial injury. Furthermore, in the same study 134 we also stained for alpha- actin, a VSMC marker, and found a significant amount of positive stain in the media and intima, suggesting that those cells are primarily VSMCs. Currently, there is much debate concerning the origin of VSMCs present in the intima after arterial injury. As previously mentioned, the medial layer was thought to be the primary source of VSMCs, however other possibilities include adventitial fibroblasts migrating from the adventitia to the intima 19 and also progenitor cells from either the bone marrow 112 or within the vessel wall itself 113. We have previously found that insulin increases the number of circulating progenitor cells 134, which may affect neointimal growth. 2. In addition to VSMC migration and proliferation, ECM accumulation plays an important role in neointimal formation. In Study 2 and 3, we measured collagen and elastin content using immunohistochemical staining for picrosirus red (collagen) and Movat s pentachrome (elastin) 144

157 on vessel cross- sections under light microscopy. The accuracy of this method is limited. More sensitive measurements of collagen and elastin content could be performed using biochemical analysis where 14 C-proline is used to label collagen and elastin, and measure their accumulation and turnover rate 90. Furthermore, additional components of the ECM could also be measured. Similar to collagen and elastin, proteoglycans (i.e. heparan sulphate) are abundant in the late stages of ECM formation and their accumulation in the neointima could be analyzed. Hyaluronic acid and glycoprotein (i.e. fibronectin, tenascin, thrombospondin, and osteopontin) deposition in the neointima, which are part of the earlier response to arterial injury compared to elastin and collagen, could also be examined in future studies. Immunohistochemical staining of carotid artery cross- sections, as used for analysis of collagen and elastin, or Western blot analysis would be effective techniques to look for changes in several of these ECM components. 3. In Study 2 and 3 we did not identify which isoform of the NOS enzyme is involved in insulin or RSV action. Unfortunately, specific enos inhibitors are not currently available and prolonged studies with inos specific inhibitors are not feasible due to the high cost. However, in the General Discussion section 7.3, evidence from this thesis and other authors was presented in support of enos as the most likely candidate involved in insulin and RSV action to inhibit neointimal formation. This point could be addressed using immunohistochemical staining of vessel cross- sections for enos or inos, however this method is not quantitative and would not establish a causal role for enos or inos. Thus, an arterial injury model in enos or inos knockout mice could be used to see if the protective effect of insulin or RSV is abolished. These studies are currently being performed in our laboratory. The knockout mouse model has limitations as both mouse strains exhibit a phenotype that could interfere with our treatment and results. enos knockout mice are hypertensive, insulin resistant, and develop greater dietinduced atherosclerosis and increased neointimal formation (this indicates that enos is important in restenosis) after cuff injury compared to wild- type mice. In contrast, inos 145

158 knockout mice are sensitive to infection, more insulin sensitive, and develop less diet- induced atherosclerosis and decreased neointimal formation after cuff injury than wild- type (reviewed in 396 ). Thus, it may be difficult to observe changes with insulin or RSV treatment related specifically to the knockout. 4. In Study 2 and 3 the effect of insulin and RSV on systolic blood pressure was measured, however carotid blood flow and diameter was not assessed. Neointimal formation after arterial injury is inhibited by increased blood flow and is mediated, at least in part, by NO 369. Insulin has been reported to increase carotid blood flow in uninjured carotids in the absence of changes in blood pressure 397. Similarly, RSV 286 also increased flow- mediated dilation. Therefore, insulin and/ or RSV could increase blood flow through a NOS- dependent mechanism to inhibit neointimal growth. Furthermore, it is also possible that in addition to a direct effect of NOS increasing flow, a secondary effect of flow- induced activation of NOS contributes to the effect of insulin or RSV. In the future blood flow could be measured using ultrasound techniques. 5. The beneficial health effects of RSV in humans have been attributed mainly to moderate red wine consumption over a long period of time. Furthermore, studies in humans use oral RSV supplements. In Study 3, as in the majority of studies investigating the effects of RSV, RSV was not given orally, which continues to be a major criticism of most studies with RSV. We used subcutaneous implants to administer RSV, as this allowed a more constant and accurate determination of RSV dose, and allowed us to avoid the confounding effects of ultraviolet (UV) light exposure to RSV given in food or water, as UV light isomerizes RSV to an inactive form. Furthermore, oral administration of RSV requires a much higher dose as it is readily metabolized in the digestive tract (the bioavailability of RSV in rats is ~40% 387 ) and also the activity of RSV metabolites after oral administration is unknown. Studies could be repeated with RSV given orally to see if it is still effective, however our preliminary data shows that RSV also decreases neointimal formation after arterial injury in mice when given orally in food. 146

159 6. The rat arterial injury model is limited when compared with human restenosis. Inflammation and thrombosis are implicated in human restenosis and are much less after vessel injury in the rat 21. For example, we found few macrophages or hematopoietic cells via immunohistochemical staining on carotid artery cross- sections after balloon injury 134. This model is also limited because in humans there is compensatory enlargement of the vessel to maintain constant lumen diameter, which does not occur in rats 17, and our studies were carried out in normal arteries rather than atherosclerotic arteries. Therefore, we must be cautious in generalizing our conclusions from this model to restenosis in humans. The larger rabbit and pig models mimic human restenosis more closely and it would be interesting to investigate the effect of insulin on neointimal formation after balloon injury in those models. However, the rat model of carotid artery injury is the most well characterized and is very useful for studying kinetics and mechanisms of response to arterial injury

160 9 Future Directions 1. In these studies the effects of RSV and insulin were examined in normal rats. In a previous study done by our laboratory as well as in the oral sucrose model of Study 1, the effect of insulin to decrease neointimal formation was abolished when the rats were fed a high- fat diet or oral sucrose to induce insulin resistance 133. Ultimately, it is in this insulin resistant state, where insulin is less effective, that RSV may have the greatest potential for use clinically. Therefore, it is of clinical importance to determine whether RSV is still effective and can even restore the effect of insulin in this model. Recent preliminary data from our laboratory shows that RSV is still effective to decrease neointimal growth after arterial injury when rats were fed a high- fat diet. Furthermore, RSV also improved whole body insulin sensitivity in these rats as assessed by a hyperinsulinemic- euglycemic clamp, suggesting that perhaps also at the level of the vessel, insulin action may be enhanced. 2. In our studies subcutaneous implants were used, however systemic insulin treatment cannot be given to non- diabetic people due to the risk of hypoglycemia. Furthermore, systemic RSV treatment in humans (mainly orally) may not be as effective due to its rapid conversion to various conjugates by the digestive system and concentrated absorption in various organs (stomach, liver, kidney, intestine), which may limit exposure of the vasculature to RSV. Alternatively, a local delivery of insulin at the level of the vessel, perhaps in a drug- eluting stent, may be beneficial. We have preliminary data showing that insulin applied to the carotid 148

161 artery, via a pluronic gel releasing system, decreases cell migration and neointimal growth after balloon injury (see Appendix A). These data suggest that perhaps insulin would be effective in a drug- eluting stent to decrease restenosis, which will be investigated by our laboratory in the future in the rabbit arterial injury model. Since insulin absorbed from pluronic gels is transient (<24 hours) but was still effective, insulin may be infused i.v. as an alternative treatment. This method of delivery is not feasible for chronic treatment but could be performed under supervised hospital conditions at the time of percutaneous intervention in humans. Therefore, it would be interesting to determine if a short duration (<18 hours) of insulin i.v. infusion would have protective effects on the vasculature in the rat restenosis model following balloon injury. 3. As previously discussed in the Introduction section in detail, it is well known that insulin activates PI3K/ Akt resulting in enos stimulation, however this has not yet been examined in the rat arterial injury model. To address the signalling pathway upstream of enos, a PI3K inhibitor (i.e. LY294002) could be applied locally on the vessel at the time of injury using the pluronic gel system to see if insulin s effect on neointimal formation is prevented. Additionally, to implicate PI3K upstream of both Akt and enos, Western blot analysis could be performed on carotid arteries from LY treated rats for phosphorylated Akt and enos. Akt phosphorylation could also be measured in carotid arteries from L-NAME- treated rats to further provide evidence that Akt is upstream of enos. The signaling pathway upstream of enos by which RSV decreases neointimal formation is not clear, however, as discussed in the General Discussion section 7.2, studies indicate a possible role for AMPK and/ or SIRT1. To identify whether RSV s action to decrease neointimal formation in our model is mediated by AMPK or SIRT1, the respective inhibitors, Compound C and nicotinamide, could be used. Compound C cannot be administered systemically due to toxicity, however it could be applied locally at the time of injury using the pluronic gel releasing system. Furthermore, our laboratory has preliminary data showing that nicotinamide prevents the 149

162 RSV- induced decrease in neointimal growth, suggesting that RSV s effect is also SIRT1- dependent. Alternatively, AMPK or SIRT1 knockout mice could be used to see if RSV is still effective to reduce neointimal formation after arterial wire injury. 4. As mentioned previously, atherosclerosis is a chronic inflammatory response of the arterial wall and both insulin and RSV have anti- inflammatory properties. Furthermore, as we have found that insulin and RSV are protective against restenosis, it would be interesting to expand our investigation to a model of atherosclerosis as well. It is important to investigate models of atherosclerosis in addition to models of restenosis, because, although occlusive arterial disease confers a high risk for CVD events, the majority of myocardial infarctions result from atheroma, the major cause of stenosis. It is debated whether in animal models of atherosclerosis the prevailing effect of insulin is atherogenic or protective, although the recent studies favour a protective effect due to insulin s action on the endothelium 206. Oral insulin, which lowered plasma glucose levels, was shown to decrease plaque formation in ApoE knockout mice 198. However, insulin treatment still needs to be analyzed where normal glucose levels are maintained and insulin is administered parenterally, as in the treatment of diabetes. In addition, the effect of insulin on plaque stability and the NO dependence of insulin s effect remain to be investigated. We have preliminary data showing that treatment of chow- fed ApoE knockout mice, a commonly used model of atherosclerosis, for 16 weeks with 0.05U/day insulin significantly decreases lesion area in the descending aorta as shown by staining with oil red O. The same effect is also observed in the proximal aorta and aortic sinus, which are sites of earlier lipid deposition. This is not due to insulin- induced glucose- lowering as glycemia was actually increased in this group of insulin- treated mice given 40% oral glucose. Additional experiments should be completed to maintain glycemia at the same level as in the control group to eliminate any glycemic effect of insulin. However, these studies do support a protective role of insulin in the vasculature. 150

163 Studies in models of atherosclerosis, like the hypercholesterolemic rabbit 286, and the ApoE 288 and ApoE/ LDLR knockout 287 mouse, have shown that RSV is protective against atherosclerosis, as plaque formation was reduced (see Introduction section for details). However, the mechanism of the effect of RSV has not been elucidated, thus it would be interesting to investigate its possible NO- dependence. 151

164 10 References (1) Lloyd-Jones D, Adams R, Carnethon M, De SG, Ferguson TB, Flegal K, Ford E, Furie K, Go A, Greenlund K, Haase N, Hailpern S, Ho M, Howard V, Kissela B, Kittner S, Lackland D, Lisabeth L, Marelli A, McDermott M, Meigs J, Mozaffarian D, Nichol G, O'Donnell C, Roger V, Rosamond W, Sacco R, Sorlie P, Stafford R, Steinberger J, Thom T, Wasserthiel-Smoller S, Wong N, Wylie-Rosett J, Hong Y. Heart disease and stroke statistics update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation 2009 January 27;119(3):e (2) Sytkowski PA, Kannel WB, D'Agostino RB. Changes in risk factors and the decline in mortality from cardiovascular disease. The Framingham Heart Study. N Engl J Med 1990 June 7;322(23): (3) Stolar M. Glycemic control and complications in type 2 diabetes mellitus. Am J Med 2010 March;123(3 Suppl):S3-11. (4) Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J Med 2005 April 21;352(16): (5) Orford JL, Selwyn AP, Ganz P, Popma JJ, Rogers C. The comparative pathobiology of atherosclerosis and restenosis. Am J Cardiol 2000 August 24;86(4B):6H-11H. (6) Winslow RD, Sharma SK, Kim MC. Restenosis and drug-eluting stents. Mt Sinai J Med 2005 March;72(2):81-9. (7) Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR, O'Shaughnessy C, Caputo RP, Kereiakes DJ, Williams DO, Teirstein PS, Jaeger JL, Kuntz RE. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003 October 2;349(14): (8) Luscher TF, Steffel J, Eberli FR, Joner M, Nakazawa G, Tanner FC, Virmani R. Drug-eluting stent and coronary thrombosis: biological mechanisms and clinical implications. Circulation 2007 February 27;115(8): (9) Lanzer P, Sternberg K, Schmitz KP, Kolodgie F, Nakazawa G, Virmani R. Drug-eluting coronary stent very late thrombosis revisited. Herz 2008 July;33(5):

165 (10) Stettler C, Allemann S, Egger M, Windecker S, Meier B, Diem P. Efficacy of drug eluting stents in patients with and without diabetes mellitus: indirect comparison of controlled trials. Heart 2006 May;92(5): (11) Bennett MR, O'Sullivan M. Mechanisms of angioplasty and stent restenosis: implications for design of rational therapy. Pharmacol Ther 2001 August;91(2): (12) Kimura T, Kaburagi S, Tamura T, Yokoi H, Nakagawa Y, Yokoi H, Hamasaki N, Nosaka H, Nobuyoshi M, Mintz GS, Popma JJ, Leon MB. Remodeling of human coronary arteries undergoing coronary angioplasty or atherectomy. Circulation 1997 July 15;96(2): (13) Hoffmann R, Mintz GS, Dussaillant GR, Popma JJ, Pichard AD, Satler LF, Kent KM, Griffin J, Leon MB. Patterns and mechanisms of in-stent restenosis. A serial intravascular ultrasound study. Circulation 1996 September 15;94(6): (14) Bayes-Genis A, Kantor B, Keelan PC, Altman JD, Lubbe DF, Kang JH, Schwartz RS. Restenosis and Hyperplasia: Animal Models. Curr Interv Cardiol Rep 2000 November;2(4): (15) Fukuyama J, Ichikawa K, Hamano S, Shibata N. Tranilast suppresses the vascular intimal hyperplasia after balloon injury in rabbits fed on a high-cholesterol diet. Eur J Pharmacol 1996 December 30;318(2-3): (16) Sata M, Maejima Y, Adachi F, Fukino K, Saiura A, Sugiura S, Aoyagi T, Imai Y, Kurihara H, Kimura K, Omata M, Makuuchi M, Hirata Y, Nagai R. A mouse model of vascular injury that induces rapid onset of medial cell apoptosis followed by reproducible neointimal hyperplasia. J Mol Cell Cardiol 2000 November;32(11): (17) Ferns GA, Avades TY. The mechanisms of coronary restenosis: insights from experimental models. Int J Exp Pathol 2000 April;81(2): (18) Religa P, Bojakowski K, Maksymowicz M, Bojakowska M, Sirsjo A, Gaciong Z, Olszewski W, Hedin U, Thyberg J. Smooth-muscle progenitor cells of bone marrow origin contribute to the development of neointimal thickenings in rat aortic allografts and injured rat carotid arteries. Transplantation 2002 November 15;74(9): (19) Siow RC, Mallawaarachchi CM, Weissberg PL. Migration of adventitial myofibroblasts following vascular balloon injury: insights from in vivo gene transfer to rat carotid arteries. Cardiovasc Res 2003 July 1;59(1): (20) Iwakura A, Luedemann C, Shastry S, Hanley A, Kearney M, Aikawa R, Isner JM, Asahara T, Losordo DW. Estrogen-mediated, endothelial nitric oxide synthase-dependent mobilization of bone marrow-derived endothelial progenitor cells contributes to reendothelialization after arterial injury. Circulation 2003 December 23;108(25): (21) Ferns GA, Reidy MA, Ross R. Balloon catheter de-endothelialization of the nude rat carotid. Response to injury in the absence of functional T lymphocytes. Am J Pathol 1991 April;138(4):

166 (22) Fingerle J, Johnson R, Clowes AW, Majesky MW, Reidy MA. Role of platelets in smooth muscle cell proliferation and migration after vascular injury in rat carotid artery. Proc Natl Acad Sci U S A 1989 November;86(21): (23) Jawien A, Bowen-Pope DF, Lindner V, Schwartz SM, Clowes AW. Platelet-derived growth factor promotes smooth muscle migration and intimal thickening in a rat model of balloon angioplasty. J Clin Invest 1992 February;89(2): (24) Hansson GK. Immunological control mechanisms in plaque formation. Basic Res Cardiol 1994;89 Suppl 1:41-6. (25) Mitra AK, Del Core MG, Agrawal DK. Cells, cytokines and cellular immunity in the pathogenesis of fibroproliferative vasculopathies. Can J Physiol Pharmacol 2005 August;83(8-9): (26) Boyle EM, Jr., Lille ST, Allaire E, Clowes AW, Verrier ED. Endothelial cell injury in cardiovascular surgery: atherosclerosis. Ann Thorac Surg 1997 March;63(3): (27) Owens GK, Kumar MS, Wamhoff BR. Molecular regulation of vascular smooth muscle cell differentiation in development and disease. Physiol Rev 2004 July;84(3): (28) Hayashi K, Saga H, Chimori Y, Kimura K, Yamanaka Y, Sobue K. Differentiated phenotype of smooth muscle cells depends on signaling pathways through insulin-like growth factors and phosphatidylinositol 3-kinase. J Biol Chem 1998 October 30;273(44): (29) Hayashi K, Takahashi M, Kimura K, Nishida W, Saga H, Sobue K. Changes in the balance of phosphoinositide 3-kinase/protein kinase B (Akt) and the mitogen-activated protein kinases (ERK/p38MAPK) determine a phenotype of visceral and vascular smooth muscle cells. J Cell Biol 1999 May 17;145(4): (30) Mukaida N, Harada A, Matsushima K. Interleukin-8 (IL-8) and monocyte chemotactic and activating factor (MCAF/MCP-1), chemokines essentially involved in inflammatory and immune reactions. Cytokine Growth Factor Rev 1998 March;9(1):9-23. (31) Madri JA, Bell L, Marx M, Merwin JR, Basson C, Prinz C. Effects of soluble factors and extracellular matrix components on vascular cell behavior in vitro and in vivo: models of deendothelialization and repair. J Cell Biochem 1991 February;45(2): (32) Owens GK. Regulation of differentiation of vascular smooth muscle cells. Physiol Rev 1995 July;75(3): (33) Kocher O, Gabbiani F, Gabbiani G, Reidy MA, Cokay MS, Peters H, Huttner I. Phenotypic features of smooth muscle cells during the evolution of experimental carotid artery intimal thickening. Biochemical and morphologic studies. Lab Invest 1991 October;65(4): (34) Martin KA, Rzucidlo EM, Merenick BL, Fingar DC, Brown DJ, Wagner RJ, Powell RJ. The mtor/p70 S6K1 pathway regulates vascular smooth muscle cell differentiation. Am J Physiol Cell Physiol 2004 March;286(3):C507-C

167 (35) Tulis DA, Durante W, Liu X, Evans AJ, Peyton KJ, Schafer AI. Adenovirus-mediated heme oxygenase-1 gene delivery inhibits injury-induced vascular neointima formation. Circulation 2001 November 27;104(22): (36) Merrilees MJ, Lemire JM, Fischer JW, Kinsella MG, Braun KR, Clowes AW, Wight TN. Retrovirally mediated overexpression of versican v3 by arterial smooth muscle cells induces tropoelastin synthesis and elastic fiber formation in vitro and in neointima after vascular injury. Circ Res 2002 March 8;90(4): (37) Lindner V, Lappi DA, Baird A, Majack RA, Reidy MA. Role of basic fibroblast growth factor in vascular lesion formation. Circ Res 1991 January;68(1): (38) Cercek B, Fishbein MC, Forrester JS, Helfant RH, Fagin JA. Induction of insulin-like growth factor I messenger RNA in rat aorta after balloon denudation. Circ Res 1990 June;66(6): (39) Khorsandi MJ, Fagin JA, Giannella-Neto D, Forrester JS, Cercek B. Regulation of insulin-like growth factor-i and its receptor in rat aorta after balloon denudation. Evidence for local bioactivity. J Clin Invest 1992 November;90(5): (40) Wainwright CL, Miller AM, Wadsworth RM. Inflammation as a key event in the development of neointima following vascular balloon injury. Clin Exp Pharmacol Physiol 2001 November;28(11): (41) Perlman H, Maillard L, Krasinski K, Walsh K. Evidence for the rapid onset of apoptosis in medial smooth muscle cells after balloon injury. Circulation 1997 February 18;95(4): (42) Bochaton-Piallat ML, Gabbiani F, Redard M, Desmouliere A, Gabbiani G. Apoptosis participates in cellularity regulation during rat aortic intimal thickening. Am J Pathol 1995 May;146(5): (43) Geng YJ, Wu Q, Muszynski M, Hansson GK, Libby P. Apoptosis of vascular smooth muscle cells induced by in vitro stimulation with interferon-gamma, tumor necrosis factor-alpha, and interleukin-1 beta. Arterioscler Thromb Vasc Biol 1996 January;16(1): (44) McCarthy NJ, Bennett MR. The regulation of vascular smooth muscle cell apoptosis. Cardiovasc Res 2000 February;45(3): (45) Bendeck MP, Zempo N, Clowes AW, Galardy RE, Reidy MA. Smooth muscle cell migration and matrix metalloproteinase expression after arterial injury in the rat. Circ Res 1994 September;75(3): (46) Willis AI, Pierre-Paul D, Sumpio BE, Gahtan V. Vascular smooth muscle cell migration: current research and clinical implications. Vasc Endovascular Surg 2004 January;38(1): (47) Horwitz AR, Parsons JT. Cell migration--movin' on. Science 1999 November 5;286(5442): (48) Majesky MW, Reidy MA, Bowen-Pope DF, Hart CE, Wilcox JN, Schwartz SM. PDGF ligand and receptor gene expression during repair of arterial injury. J Cell Biol 1990 November;111(5 Pt 1):

168 (49) Delafontaine P, Bernstein KE, Alexander RW. Insulin-like growth factor I gene expression in vascular cells. Hypertension 1991 May;17(5): (50) Bornfeldt KE, Raines EW, Nakano T, Graves LM, Krebs EG, Ross R. Insulin-like growth factor-i and platelet-derived growth factor-bb induce directed migration of human arterial smooth muscle cells via signaling pathways that are distinct from those of proliferation. J Clin Invest 1994 March;93(3): (51) Bendeck MP, Regenass S, Tom WD, Giachelli CM, Schwartz SM, Hart C, Reidy MA. Differential expression of alpha 1 type VIII collagen in injured platelet-derived growth factor- BB--stimulated rat carotid arteries. Circ Res 1996 September;79(3): (52) Riessen R, Wight TN, Pastore C, Henley C, Isner JM. Distribution of hyaluronan during extracellular matrix remodeling in human restenotic arteries and balloon-injured rat carotid arteries. Circulation 1996 March 15;93(6): (53) Dubin D, Peters JH, Brown LF, Logan B, Kent KC, Berse B, Berven S, Cercek B, Sharifi BG, Pratt RE,. Balloon catheterization induced arterial expression of embryonic fibronectins. Arterioscler Thromb Vasc Biol 1995 November;15(11): (54) Liaw L, Lombardi DM, Almeida MM, Schwartz SM, deblois D, Giachelli CM. Neutralizing antibodies directed against osteopontin inhibit rat carotid neointimal thickening after endothelial denudation. Arterioscler Thromb Vasc Biol 1997 January;17(1): (55) Bendeck MP, Irvin C, Reidy M, Smith L, Mulholland D, Horton M, Giachelli CM. Smooth muscle cell matrix metalloproteinase production is stimulated via alpha(v)beta(3) integrin. Arterioscler Thromb Vasc Biol 2000 June;20(6): (56) Hedin U, Holm J, Hansson GK. Induction of tenascin in rat arterial injury. Relationship to altered smooth muscle cell phenotype. Am J Pathol 1991 September;139(3): (57) LaFleur DW, Chiang J, Fagin JA, Schwartz SM, Shah PK, Wallner K, Forrester JS, Sharifi BG. Aortic smooth muscle cells interact with tenascin-c through its fibrinogen-like domain. J Biol Chem 1997 December 26;272(52): (58) Barolet AW, Nili N, Cheema A, Robinson R, Natarajan MK, O'Blenes S, Li J, Eskandarian MR, Sparkes J, Rabinovitch M, Strauss BH. Arterial elastase activity after balloon angioplasty and effects of elafin, an elastase inhibitor. Arterioscler Thromb Vasc Biol 2001 August;21(8): (59) Cheng XW, Kuzuya M, Sasaki T, Arakawa K, Kanda S, Sumi D, Koike T, Maeda K, Tamaya- Mori N, Shi GP, Saito N, Iguchi A. Increased expression of elastolytic cysteine proteases, cathepsins S and K, in the neointima of balloon-injured rat carotid arteries. Am J Pathol 2004 January;164(1): (60) Bilato C, Pauly RR, Melillo G, Monticone R, Gorelick-Feldman D, Gluzband YA, Sollott SJ, Ziman B, Lakatta EG, Crow MT. Intracellular signaling pathways required for rat vascular smooth muscle cell migration. Interactions between basic fibroblast growth factor and plateletderived growth factor. J Clin Invest 1995 October;96(4):

169 (61) Pauly RR, Bilato C, Sollott SJ, Monticone R, Kelly PT, Lakatta EG, Crow MT. Role of calcium/calmodulin-dependent protein kinase II in the regulation of vascular smooth muscle cell migration. Circulation 1995 February 15;91(4): (62) Schachter M. Vascular smooth muscle cell migration, atherosclerosis, and calcium channel blockers. Int J Cardiol 1997 December 31;62 Suppl 2:S85-S90. (63) Kingsley K, Plopper GE. Platelet-derived growth factor modulates rat vascular smooth muscle cell responses on laminin-5 via mitogen-activated protein kinase-sensitive pathways. Cell Commun Signal 2005 January 31;3(1):2. (64) Mackay DJ, Hall A. Rho GTPases. J Biol Chem 1998 August 14;273(33): (65) Nelson PR, Yamamura S, Kent KC. Platelet-derived growth factor and extracellular matrix proteins provide a synergistic stimulus for human vascular smooth muscle cell migration. J Vasc Surg 1997 July;26(1): (66) Schlaepfer DD, Hanks SK, Hunter T, van der GP. Integrin-mediated signal transduction linked to Ras pathway by GRB2 binding to focal adhesion kinase. Nature 1994 December 22;372(6508): (67) van Nieuw Amerongen GP, van H, V. Cytoskeletal effects of rho-like small guanine nucleotide-binding proteins in the vascular system. Arterioscler Thromb Vasc Biol 2001 March;21(3): (68) Campbell SL, Khosravi-Far R, Rossman KL, Clark GJ, Der CJ. Increasing complexity of Ras signaling. Oncogene 1998 September 17;17(11 Reviews): (69) Klemke RL, Cai S, Giannini AL, Gallagher PJ, de Lanerolle P, Cheresh DA. Regulation of cell motility by mitogen-activated protein kinase. J Cell Biol 1997 April 21;137(2): (70) Seasholtz TM, Majumdar M, Kaplan DD, Brown JH. Rho and Rho kinase mediate thrombinstimulated vascular smooth muscle cell DNA synthesis and migration. Circ Res 1999 May 28;84(10): (71) Doanes AM, Irani K, Goldschmidt-Clermont PJ, Finkel T. A requirement for rac1 in the PDGF-stimulated migration of fibroblasts and vascular smooth cells. Biochem Mol Biol Int 1998 June;45(2): (72) Cercek B, Yamashita M, Dimayuga P, Zhu J, Fishbein MC, Kaul S, Shah PK, Nilsson J, Regnstrom J. Nuclear factor-kappab activity and arterial response to balloon injury. Atherosclerosis 1997 May;131(1): (73) Landry DB, Couper LL, Bryant SR, Lindner V. Activation of the NF-kappa B and I kappa B system in smooth muscle cells after rat arterial injury. Induction of vascular cell adhesion molecule-1 and monocyte chemoattractant protein-1. Am J Pathol 1997 October;151(4): (74) Yamboliev IA, Chen J, Gerthoffer WT. PI 3-kinases and Src kinases regulate spreading and migration of cultured VSMCs. Am J Physiol Cell Physiol 2001 August;281(2):C709-C

170 (75) Abedi H, Zachary I. Signalling mechanisms in the regulation of vascular cell migration. Cardiovasc Res 1995 October;30(4): (76) Rice KM, Kakarla SK, Mupparaju SP, Paturi S, Katta A, Wu M, Harris RT, Blough ER. Shear stress activates Akt during vascular smooth muscle cell reorientation. Biotechnol Appl Biochem 2010 January 7. (77) Lee EJ, Kim DI, Kim WJ, Moon SK. Naringin inhibits matrix metalloproteinase-9 expression and AKT phosphorylation in tumor necrosis factor-alpha-induced vascular smooth muscle cells. Mol Nutr Food Res 2009 December;53(12): (78) Poon M, Marx SO, Gallo R, Badimon JJ, Taubman MB, Marks AR. Rapamycin inhibits vascular smooth muscle cell migration. J Clin Invest 1996 November 15;98(10): (79) Galis ZS, Khatri JJ. Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. Circ Res 2002 February 22;90(3): (80) Bendeck MP, Conte M, Zhang M, Nili N, Strauss BH, Farwell SM. Doxycycline modulates smooth muscle cell growth, migration, and matrix remodeling after arterial injury. Am J Pathol 2002 March;160(3): (81) Forough R, Koyama N, Hasenstab D, Lea H, Clowes M, Nikkari ST, Clowes AW. Overexpression of tissue inhibitor of matrix metalloproteinase-1 inhibits vascular smooth muscle cell functions in vitro and in vivo. Circ Res 1996 October;79(4): (82) Bendeck MP, Irvin C, Reidy MA. Inhibition of matrix metalloproteinase activity inhibits smooth muscle cell migration but not neointimal thickening after arterial injury. Circ Res 1996 January;78(1): (83) Parsons SL, Watson SA, Brown PD, Collins HM, Steele RJ. Matrix metalloproteinases. Br J Surg 1997 February;84(2): (84) Sternlicht MD, Werb Z. How matrix metalloproteinases regulate cell behavior. Annu Rev Cell Dev Biol 2001;17: (85) Moon SK, Cha BY, Kim CH. ERK1/2 mediates TNF-alpha-induced matrix metalloproteinase- 9 expression in human vascular smooth muscle cells via the regulation of NF-kappaB and AP- 1: Involvement of the ras dependent pathway. J Cell Physiol 2004 March;198(3): (86) Hou G, Mulholland D, Gronska MA, Bendeck MP. Type VIII collagen stimulates smooth muscle cell migration and matrix metalloproteinase synthesis after arterial injury. Am J Pathol 2000 February;156(2): (87) Katsuyama K, Shichiri M, Marumo F, Hirata Y. NO inhibits cytokine-induced inos expression and NF-kappaB activation by interfering with phosphorylation and degradation of IkappaB-alpha. Arterioscler Thromb Vasc Biol 1998 November;18(11): (88) Strauss BH, Robinson R, Batchelor WB, Chisholm RJ, Ravi G, Natarajan MK, Logan RA, Mehta SR, Levy DE, Ezrin AM, Keeley FW. In vivo collagen turnover following experimental balloon angioplasty injury and the role of matrix metalloproteinases. Circ Res 1996 September;79(3):

171 (89) Wight TN. The extracellular matrix and atherosclerosis. Curr Opin Lipidol 1995 October;6(5): (90) Nili N, Zhang M, Strauss BH, Bendeck MP. Biochemical analysis of collagen and elastin synthesis in the balloon injured rat carotid artery. Cardiovasc Pathol 2002 September;11(5): (91) Reidy MA, Irvin C, Lindner V. Migration of arterial wall cells. Expression of plasminogen activators and inhibitors in injured rat arteries. Circ Res 1996 March;78(3): (92) Franco CD, Hou G, Bendeck MP. Collagens, integrins, and the discoidin domain receptors in arterial occlusive disease. Trends Cardiovasc Med 2002 May;12(4): (93) Bauters C, Marotte F, Hamon M, Oliviero P, Farhadian F, Robert V, Samuel JL, Rappaport L. Accumulation of fetal fibronectin mrnas after balloon denudation of rabbit arteries. Circulation 1995 August 15;92(4): (94) Giachelli CM, Bae N, Almeida M, Denhardt DT, Alpers CE, Schwartz SM. Osteopontin is elevated during neointima formation in rat arteries and is a novel component of human atherosclerotic plaques. J Clin Invest 1993 October;92(4): (95) Batchelor WB, Robinson R, Strauss BH. The extracellular matrix in balloon arterial injury: a novel target for restenosis prevention. Prog Cardiovasc Dis 1998 July;41(1): (96) Asahara T, Bauters C, Pastore C, Kearney M, Rossow S, Bunting S, Ferrara N, Symes JF, Isner JM. Local delivery of vascular endothelial growth factor accelerates reendothelialization and attenuates intimal hyperplasia in balloon-injured rat carotid artery. Circulation 1995 June 1;91(11): (97) Lindner V, Majack RA, Reidy MA. Basic fibroblast growth factor stimulates endothelial regrowth and proliferation in denuded arteries. J Clin Invest 1990 June;85(6): (98) Breen EC. VEGF in biological control. J Cell Biochem 2007 December 15;102(6): (99) Dimmeler S, Dernbach E, Zeiher AM. Phosphorylation of the endothelial nitric oxide synthase at ser-1177 is required for VEGF-induced endothelial cell migration. FEBS Lett 2000 July 21;477(3): (100) Ushio-Fukai M, Alexander RW, Akers M, Griendling KK. p38 Mitogen-activated protein kinase is a critical component of the redox-sensitive signaling pathways activated by angiotensin II. Role in vascular smooth muscle cell hypertrophy. J Biol Chem 1998 June 12;273(24): (101) Ohtani K, Egashira K, Hiasa K, Zhao Q, Kitamoto S, Ishibashi M, Usui M, Inoue S, Yonemitsu Y, Sueishi K, Sata M, Shibuya M, Sunagawa K. Blockade of vascular endothelial growth factor suppresses experimental restenosis after intraluminal injury by inhibiting recruitment of monocyte lineage cells. Circulation 2004 October 19;110(16): (102) Hedman M, Hartikainen J, Syvanne M, Stjernvall J, Hedman A, Kivela A, Vanninen E, Mussalo H, Kauppila E, Simula S, Narvanen O, Rantala A, Peuhkurinen K, Nieminen MS, Laakso M, Yla-Herttuala S. Safety and feasibility of catheter-based local intracoronary vascular 159

172 endothelial growth factor gene transfer in the prevention of postangioplasty and in-stent restenosis and in the treatment of chronic myocardial ischemia: phase II results of the Kuopio Angiogenesis Trial (KAT). Circulation 2003 June 3;107(21): (103) Reidy MA, Clowes AW, Schwartz SM. Endothelial regeneration. V. Inhibition of endothelial regrowth in arteries of rat and rabbit. Lab Invest 1983 November;49(5): (104) Clowes AW, Reidy MA, Clowes MM. Mechanisms of stenosis after arterial injury. Lab Invest 1983 August;49(2): (105) McNamara DB, Bedi B, Aurora H, Tena L, Ignarro LJ, Kadowitz PJ, Akers DL. L-arginine inhibits balloon catheter-induced intimal hyperplasia. Biochem Biophys Res Commun 1993 May 28;193(1): (106) Ferguson JJ. Meeting highlights: nitric oxide donors. Circulation 1994 July;90(1):4. (107) Nugent HM, Rogers C, Edelman ER. Endothelial implants inhibit intimal hyperplasia after porcine angioplasty. Circ Res 1999 March 5;84(4): (108) Yogo K, Shimokawa H, Funakoshi H, Kandabashi T, Miyata K, Okamoto S, Egashira K, Huang P, Akaike T, Takeshita A. Different vasculoprotective roles of NO synthase isoforms in vascular lesion formation in mice. Arterioscler Thromb Vasc Biol 2000 November;20(11):E96- E100. (109) Chamberlain G, Fox J, Ashton B, Middleton J. Concise review: mesenchymal stem cells: their phenotype, differentiation capacity, immunological features, and potential for homing. Stem Cells 2007 November;25(11): (110) Ribatti D. Hemangioblast does exist. Leuk Res 2008 June;32(6): (111) Zhao Y, Glesne D, Huberman E. A human peripheral blood monocyte-derived subset acts as pluripotent stem cells. Proc Natl Acad Sci U S A 2003 March 4;100(5): (112) Han CI, Campbell GR, Campbell JH. Circulating bone marrow cells can contribute to neointimal formation. J Vasc Res 2001 March;38(2): (113) Sainz J, Al Haj ZA, Caligiuri G, Demerens C, Urbain D, Lemitre M, Lafont A. Isolation of "side population" progenitor cells from healthy arteries of adult mice. Arterioscler Thromb Vasc Biol 2006 February;26(2): (114) Hoofnagle MH, Thomas JA, Wamhoff BR, Owens GK. Origin of neointimal smooth muscle: we've come full circle. Arterioscler Thromb Vasc Biol 2006 December;26(12): (115) Fadini GP, Sartore S, Schiavon M, Albiero M, Baesso I, Cabrelle A, Agostini C, Avogaro A. Diabetes impairs progenitor cell mobilisation after hindlimb ischaemia-reperfusion injury in rats. Diabetologia 2006 December;49(12): (116) Tanaka K, Sata M, Hirata Y, Nagai R. Diverse contribution of bone marrow cells to neointimal hyperplasia after mechanical vascular injuries. Circ Res 2003 October 17;93(8):

173 (117) Nervi B, Link DC, DiPersio JF. Cytokines and hematopoietic stem cell mobilization. J Cell Biochem 2006 October 15;99(3): (118) Cavallaro AM, Lilleby K, Majolino I, Storb R, Appelbaum FR, Rowley SD, Bensinger WI. Three to six year follow-up of normal donors who received recombinant human granulocyte colony-stimulating factor. Bone Marrow Transplant 2000 January;25(1):85-9. (119) Heissig B, Hattori K, Dias S, Friedrich M, Ferris B, Hackett NR, Crystal RG, Besmer P, Lyden D, Moore MA, Werb Z, Rafii S. Recruitment of stem and progenitor cells from the bone marrow niche requires MMP-9 mediated release of kit-ligand. Cell 2002 May 31;109(5): (120) Zernecke A, Schober A, Bot I, von HP, Liehn EA, Mopps B, Mericskay M, Gierschik P, Biessen EA, Weber C. SDF-1alpha/CXCR4 axis is instrumental in neointimal hyperplasia and recruitment of smooth muscle progenitor cells. Circ Res 2005 April 15;96(7): (121) Oswald J, Boxberger S, Jorgensen B, Feldmann S, Ehninger G, Bornhauser M, Werner C. Mesenchymal stem cells can be differentiated into endothelial cells in vitro. Stem Cells 2004;22(3): (122) Muto A, Fitzgerald TN, Pimiento JM, Maloney SP, Teso D, Paszkowiak JJ, Westvik TS, Kudo FA, Nishibe T, Dardik A. Smooth muscle cell signal transduction: implications of vascular biology for vascular surgeons. J Vasc Surg 2007 June;45 Suppl A:A15-A24. (123) Miglionico M, Patti G, D'Ambrosio A, Di SG. Percutaneous coronary intervention utilizing a new endothelial progenitor cells antibody-coated stent: a prospective single-center registry in high-risk patients. Catheter Cardiovasc Interv 2008 April 1;71(5): (124) Howard G, O'Leary DH, Zaccaro D, Haffner S, Rewers M, Hamman R, Selby JV, Saad MF, Savage P, Bergman R. Insulin sensitivity and atherosclerosis. The Insulin Resistance Atherosclerosis Study (IRAS) Investigators. Circulation 1996 May 15;93(10): (125) Willfort-Ehringer A, Ahmadi R, Gessl A, Gschwandtner ME, Haumer A, Lang W, Minar E, Zehetmayer S, Ehringer H. Neointimal proliferation within carotid stents is more pronounced in diabetic patients with initial poor glycaemic state. Diabetologia 2004 March;47(3): (126) Flaherty JD, Davidson CJ. Diabetes and coronary revascularization. JAMA 2005 March 23;293(12): (127) Seven-year outcome in the Bypass Angioplasty Revascularization Investigation (BARI) by treatment and diabetic status. J Am Coll Cardiol 2000 April;35(5): (128) Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2005 April 16;365(9468): (129) Agazie YM, Bagot JC, Trickey E, Halenda SP, Wilden PA. Molecular mechanisms of ATP and insulin synergistic stimulation of coronary artery smooth muscle growth. Am J Physiol Heart Circ Physiol 2001 February;280(2):H795-H

174 (130) Zeng G, Nystrom FH, Ravichandran LV, Cong LN, Kirby M, Mostowski H, Quon MJ. Roles for insulin receptor, PI3-kinase, and Akt in insulin-signaling pathways related to production of nitric oxide in human vascular endothelial cells. Circulation 2000 April 4;101(13): (131) Westerbacka J, Yki-Jarvinen H, Turpeinen A, Rissanen A, Vehkavaara S, Syrjala M, Lassila R. Inhibition of platelet-collagen interaction: an in vivo action of insulin abolished by insulin resistance in obesity. Arterioscler Thromb Vasc Biol 2002 January;22(1): (132) Dandona P, Chaudhuri A, Mohanty P, Ghanim H. Anti-inflammatory effects of insulin. Curr Opin Clin Nutr Metab Care 2007 July;10(4): (133) Kim T, Chan KK, Dhaliwall JK, Huynh N, Suen R, Uchino H, Naigamwalla D, Bendeck MP, Giacca A. Anti-atherogenic effect of insulin in vivo. J Vasc Res 2005 November;42(6): (134) Breen DM, Chan KK, Dhaliwall JK, Ward MR, Al KN, Lam L, De SM, Ghanim H, Dandona P, Stewart DJ, Bendeck MP, Giacca A. Insulin increases reendothelialization and inhibits cell migration and neointimal growth after arterial injury. Arterioscler Thromb Vasc Biol 2009 July;29(7): (135) Ignarro LJ. Physiology and pathophysiology of nitric oxide. Kidney Int Suppl 1996 June;55:S2- S5. (136) Szabo C, Ischiropoulos H, Radi R. Peroxynitrite: biochemistry, pathophysiology and development of therapeutics. Nat Rev Drug Discov 2007 August;6(8): (137) Forstermann U, Closs EI, Pollock JS, Nakane M, Schwarz P, Gath I, Kleinert H. Nitric oxide synthase isozymes. Characterization, purification, molecular cloning, and functions. Hypertension 1994 June;23(6 Pt 2): (138) Li H, Wallerath T, Forstermann U. Physiological mechanisms regulating the expression of endothelial-type NO synthase. Nitric Oxide 2002 September;7(2): (139) Fleming I. Molecular mechanisms underlying the activation of enos. Pflugers Arch 2009 December 13. (140) Kobayashi T, Taguchi K, Nemoto S, Nogami T, Matsumoto T, Kamata K. Activation of the PDK-1/Akt/eNOS pathway involved in aortic endothelial function differs between hyperinsulinemic and insulin-deficient diabetic rats. Am J Physiol Heart Circ Physiol 2009 November;297(5):H1767-H1775. (141) Fulton D, Gratton JP, McCabe TJ, Fontana J, Fujio Y, Walsh K, Franke TF, Papapetropoulos A, Sessa WC. Regulation of endothelium-derived nitric oxide production by the protein kinase Akt. Nature 1999 June 10;399(6736): (142) Xu Q, Hao X, Yang Q, Si L. Resveratrol prevents hyperglycemia-induced endothelial dysfunction via activation of adenosine monophosphate-activated protein kinase. Biochem Biophys Res Commun 2009 October 16;388(2): (143) Davis BJ, Xie Z, Viollet B, Zou MH. Activation of the AMP-activated kinase by antidiabetes drug metformin stimulates nitric oxide synthesis in vivo by promoting the association of heat shock protein 90 and endothelial nitric oxide synthase. Diabetes 2006 February;55(2):

175 (144) Nathan C. Inducible nitric oxide synthase: what difference does it make? J Clin Invest 1997 November 15;100(10): (145) Crabtree MJ, Tatham AL, Al-Wakeel Y, Warrick N, Hale AB, Cai S, Channon KM, Alp NJ. Quantitative regulation of intracellular endothelial nitric-oxide synthase (enos) coupling by both tetrahydrobiopterin-enos stoichiometry and biopterin redox status: insights from cells with tet-regulated GTP cyclohydrolase I expression. J Biol Chem 2009 January 9;284(2): (146) Kipshidze N, Dangas G, Tsapenko M, Moses J, Leon MB, Kutryk M, Serruys P. Role of the endothelium in modulating neointimal formation: vasculoprotective approaches to attenuate restenosis after percutaneous coronary interventions. J Am Coll Cardiol 2004 August 18;44(4): (147) Muruganandam A, Mutus B. Isolation of nitric oxide synthase from human platelets. Biochim Biophys Acta 1994 May 25;1200(1):1-6. (148) Yan Z, Hansson GK. Overexpression of inducible nitric oxide synthase by neointimal smooth muscle cells. Circ Res 1998 January 9;82(1):21-9. (149) Chaux A, Ruan XM, Fishbein MC, Ouyang Y, Kaul S, Pass JA, Matloff JM. Perivascular delivery of a nitric oxide donor inhibits neointimal hyperplasia in vein grafts implanted in the arterial circulation. J Thorac Cardiovasc Surg 1998 March;115(3): (150) Cayatte AJ, Palacino JJ, Horten K, Cohen RA. Chronic inhibition of nitric oxide production accelerates neointima formation and impairs endothelial function in hypercholesterolemic rabbits. Arterioscler Thromb 1994 May;14(5): (151) Kong D, Melo LG, Mangi AA, Zhang L, Lopez-Ilasaca M, Perrella MA, Liew CC, Pratt RE, Dzau VJ. Enhanced inhibition of neointimal hyperplasia by genetically engineered endothelial progenitor cells. Circulation 2004 April 13;109(14): (152) Janssens S, Flaherty D, Nong Z, Varenne O, Van Pelt N, Haustermans C, Zoldhelyi P, Gerard R, Collen D. Human endothelial nitric oxide synthase gene transfer inhibits vascular smooth muscle cell proliferation and neointima formation after balloon injury in rats. Circulation 1998 April 7;97(13): (153) Zhang LN, Wilson DW, da C, V, Sullivan ME, Vergona R, Rutledge JC, Wang YX. Endothelial NO synthase deficiency promotes smooth muscle progenitor cells in association with upregulation of stromal cell-derived factor-1alpha in a mouse model of carotid artery ligation. Arterioscler Thromb Vasc Biol 2006 April;26(4): (154) Chyu KY, Dimayuga P, Zhu J, Nilsson J, Kaul S, Shah PK, Cercek B. Decreased neointimal thickening after arterial wall injury in inducible nitric oxide synthase knockout mice. Circ Res 1999 December 3;85(12): (155) der Leyen HE, Dzau VJ. Therapeutic potential of nitric oxide synthase gene manipulation. Circulation 2001 June 5;103(22): (156) Shears LL, Kibbe MR, Murdock AD, Billiar TR, Lizonova A, Kovesdi I, Watkins SC, Tzeng E. Efficient inhibition of intimal hyperplasia by adenovirus-mediated inducible nitric oxide 163

176 synthase gene transfer to rats and pigs in vivo. J Am Coll Surg 1998 September;187(3): (157) Barbato JE, Zuckerbraun BS, Overhaus M, Raman KG, Tzeng E. Nitric oxide modulates vascular inflammation and intimal hyperplasia in insulin resistance and the metabolic syndrome. Am J Physiol Heart Circ Physiol 2005 July;289(1):H228-H236. (158) Vasquez-Vivar J, Kalyanaraman B, Martasek P, Hogg N, Masters BS, Karoui H, Tordo P, Pritchard KA, Jr. Superoxide generation by endothelial nitric oxide synthase: the influence of cofactors. Proc Natl Acad Sci U S A 1998 August 4;95(16): (159) Morishita T, Tsutsui M, Shimokawa H, Horiuchi M, Tanimoto A, Suda O, Tasaki H, Huang PL, Sasaguri Y, Yanagihara N, Nakashima Y. Vasculoprotective roles of neuronal nitric oxide synthase. FASEB J 2002 December;16(14): (160) Arthur JF, Yin ZL, Young HM, Dusting GJ. Induction of nitric oxide synthase in the neointima induced by a periarterial collar in rabbits. Arterioscler Thromb Vasc Biol 1997 April;17(4): (161) Lablanche JM, Grollier G, Lusson JR, Bassand JP, Drobinski G, Bertrand B, Battaglia S, Desveaux B, Juilliere Y, Juliard JM, Metzger JP, Coste P, Quiret JC, Dubois-Rande JL, Crochet PD, Letac B, Boschat J, Virot P, Finet G, Le BH, Livarek B, Leclercq F, Beard T, Giraud T, Bertrand ME,. Effect of the direct nitric oxide donors linsidomine and molsidomine on angiographic restenosis after coronary balloon angioplasty. The ACCORD Study. Angioplastic Coronaire Corvasal Diltiazem. Circulation 1997 January 7;95(1):83-9. (162) Wohrle J, Hoher M, Nusser T, Hombach V, Kochs M. No effect of highly dosed nitric oxide donor molsidomine on the angiographic restenosis rate after percutaneous coronary angioplasty: a randomized, placebo controlled, double-blind trial. Can J Cardiol 2003 April;19(5): (163) Naseem KM. The role of nitric oxide in cardiovascular diseases. Mol Aspects Med 2005 February;26(1-2): (164) Yan ZQ, Yokota T, Zhang W, Hansson GK. Expression of inducible nitric oxide synthase inhibits platelet adhesion and restores blood flow in the injured artery. Circ Res 1996 July;79(1): (165) Hickey MJ, Sharkey KA, Sihota EG, Reinhardt PH, Macmicking JD, Nathan C, Kubes P. Inducible nitric oxide synthase-deficient mice have enhanced leukocyte-endothelium interactions in endotoxemia. FASEB J 1997 October;11(12): (166) Guo JP, Panday MM, Consigny PM, Lefer AM. Mechanisms of vascular preservation by a novel NO donor following rat carotid artery intimal injury. Am J Physiol 1995 September;269(3 Pt 2):H1122-H1131. (167) Parenti A, Morbidelli L, Cui XL, Douglas JG, Hood JD, Granger HJ, Ledda F, Ziche M. Nitric oxide is an upstream signal of vascular endothelial growth factor-induced extracellular signalregulated kinase1/2 activation in postcapillary endothelium. J Biol Chem 1998 February 13;273(7):

177 (168) Murohara T, Asahara T, Silver M, Bauters C, Masuda H, Kalka C, Kearney M, Chen D, Symes JF, Fishman MC, Huang PL, Isner JM. Nitric oxide synthase modulates angiogenesis in response to tissue ischemia. J Clin Invest 1998 June 1;101(11): (169) Dimmeler S, Haendeler J, Nehls M, Zeiher AM. Suppression of apoptosis by nitric oxide via inhibition of interleukin-1beta-converting enzyme (ICE)-like and cysteine protease protein (CPP)-32-like proteases. J Exp Med 1997 February 17;185(4): (170) Tzeng E, Kim YM, Pitt BR, Lizonova A, Kovesdi I, Billiar TR. Adenoviral transfer of the inducible nitric oxide synthase gene blocks endothelial cell apoptosis. Surgery 1997 August;122(2): (171) Garg UC, Hassid A. Nitric oxide-generating vasodilators and 8-bromo-cyclic guanosine monophosphate inhibit mitogenesis and proliferation of cultured rat vascular smooth muscle cells. J Clin Invest 1989 May;83(5): (172) Ishida A, Sasaguri T, Kosaka C, Nojima H, Ogata J. Induction of the cyclin-dependent kinase inhibitor p21(sdi1/cip1/waf1) by nitric oxide-generating vasodilator in vascular smooth muscle cells. J Biol Chem 1997 April 11;272(15): (173) Kibbe MR, Li J, Nie S, Watkins SC, Lizonova A, Kovesdi I, Simmons RL, Billiar TR, Tzeng E. Inducible nitric oxide synthase (inos) expression upregulates p21 and inhibits vascular smooth muscle cell proliferation through p42/44 mitogen-activated protein kinase activation and independent of p53 and cyclic guanosine monophosphate. J Vasc Surg 2000 June;31(6): (174) Li SJ, Sun NL. Regulation of intracellular Ca2+ and calcineurin by NO/PKG in proliferation of vascular smooth muscle cells. Acta Pharmacol Sin 2005 March;26(3): (175) Xu S, He Y, Vokurkova M, Touyz RM. Endothelial cells negatively modulate reactive oxygen species generation in vascular smooth muscle cells: role of thioredoxin. Hypertension 2009 August;54(2): (176) Pollman MJ, Yamada T, Horiuchi M, Gibbons GH. Vasoactive substances regulate vascular smooth muscle cell apoptosis. Countervailing influences of nitric oxide and angiotensin II. Circ Res 1996 October;79(4): (177) Iwashina M, Shichiri M, Marumo F, Hirata Y. Transfection of inducible nitric oxide synthase gene causes apoptosis in vascular smooth muscle cells. Circulation 1998 September 22;98(12): (178) Sarkar R, Meinberg EG, Stanley JC, Gordon D, Webb RC. Nitric oxide reversibly inhibits the migration of cultured vascular smooth muscle cells. Circ Res 1996 February;78(2): (179) Dubey RK, Jackson EK, Luscher TF. Nitric oxide inhibits angiotensin II-induced migration of rat aortic smooth muscle cell. Role of cyclic-nucleotides and angiotensin1 receptors. J Clin Invest 1995 July;96(1): (180) Cohen RA, Weisbrod RM, Gericke M, Yaghoubi M, Bierl C, Bolotina VM. Mechanism of nitric oxide-induced vasodilatation: refilling of intracellular stores by sarcoplasmic reticulum 165

178 Ca2+ ATPase and inhibition of store-operated Ca2+ influx. Circ Res 1999 February 5;84(2): (181) Gurjar MV, Sharma RV, Bhalla RC. enos gene transfer inhibits smooth muscle cell migration and MMP-2 and MMP-9 activity. Arterioscler Thromb Vasc Biol 1999 December;19(12): (182) Dattilo JB, Dattilo MP, Crane JT, Yager DR, Makhoul RG. The nitric oxide precursor L- arginine reduces expression of hyaluronan synthase in experimental vein bypass grafts. J Surg Res 1998 January;74(1): (183) Rizvi MA, Myers PR. Nitric oxide modulates basal and endothelin-induced coronary artery vascular smooth muscle cell proliferation and collagen levels. J Mol Cell Cardiol 1997 July;29(7): (184) Kolpakov V, Gordon D, Kulik TJ. Nitric oxide-generating compounds inhibit total protein and collagen synthesis in cultured vascular smooth muscle cells. Circ Res 1995 February;76(2): (185) Schmidt A, Geigenmueller S, Voelker W, Seiler P, Buddecke E. Exogenous nitric oxide causes overexpression of TGF-beta1 and overproduction of extracellular matrix in human coronary smooth muscle cells. Cardiovasc Res 2003 June 1;58(3): (186) Colotti C, Vittorini S, Ottaviano V, Maltinti M, Angeli V, Del RS, Giannessi D. Nitric oxide treatment reduces neo-intimal formation and modulates osteopontin expression in an ex-vivo human model of intimal hyperplasia. Cytokine 2009 May;46(2): (187) Saltiel AR, Kahn CR. Insulin signalling and the regulation of glucose and lipid metabolism. Nature 2001 December 13;414(6865): (188) Ruige JB, Assendelft WJ, Dekker JM, Kostense PJ, Heine RJ, Bouter LM. Insulin and risk of cardiovascular disease: a meta-analysis. Circulation 1998 March 17;97(10): (189) Pyorala M, Miettinen H, Laakso M, Pyorala K. Plasma insulin and all-cause, cardiovascular, and noncardiovascular mortality: the 22-year follow-up results of the Helsinki Policemen Study. Diabetes Care 2000 August;23(8): (190) Yarnell JW, Sweetnam PM, Marks V, Teale JD, Bolton CH. Insulin in ischaemic heart disease: are associations explained by triglyceride concentrations? The Caerphilly prospective study. Br Heart J 1994 March;71(3): (191) Shinozaki K, Naritomi H, Shimizu T, Suzuki M, Ikebuchi M, Sawada T, Harano Y. Role of insulin resistance associated with compensatory hyperinsulinemia in ischemic stroke. Stroke 1996 January;27(1): (192) Nathan DM, Cleary PA, Backlund JY, Genuth SM, Lachin JM, Orchard TJ, Raskin P, Zinman B. Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes. N Engl J Med 2005 December 22;353(25):

179 (193) Stratton IM, Cull CA, Adler AI, Matthews DR, Neil HA, Holman RR. Additive effects of glycaemia and blood pressure exposure on risk of complications in type 2 diabetes: a prospective observational study (UKPDS 75). Diabetologia 2006 August;49(8): (194) DUFF GL, BRECHIN DJ, FINKELSTEIN WE. The effect of alloxan diabetes on experimental cholesterol atherosclerosis in the rabbit. IV. The effect of insulin therapy on the inhibition of atherosclerosis in the alloxan-diabetic rabbit. J Exp Med 1954 October 1;100(4): (195) Marquie G. Effect of insulin in the induction and regression of experimental cholesterol atherosclerosis in the rabbit. Postgrad Med J 1978 February;54(628):80-5. (196) STAMLER J, PICK R, KATZ LN. Effect of insulin in the induction and regression of atherosclerosis in the chick. Circ Res 1960 May;8: (197) Nordestgaard BG, Agerholm-Larsen B, Stender S. Effect of exogenous hyperinsulinaemia on atherogenesis in cholesterol-fed rabbits. Diabetologia 1997 May;40(5): (198) Shamir R, Shehadeh N, Rosenblat M, Eshach-Adiv O, Coleman R, Kaplan M, Hamoud S, Lischinsky S, Hayek T. Oral insulin supplementation attenuates atherosclerosis progression in apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol 2003 January 1;23(1): (199) Clough MH, Schneider DJ, Sobel BE, White MF, Wadsworth MP, Taatjes DJ. Attenuation of accumulation of neointimal lipid by pioglitazone in mice genetically deficient in insulin receptor substrate-2 and apolipoprotein E. J Histochem Cytochem 2005 May;53(5): (200) Gonzalez-Navarro H, Vila-Caballer M, Pastor MF, Vinue A, White MF, Burks D, Andres V. Plasma insulin levels predict the development of atherosclerosis when IRS2 deficiency is combined with severe hypercholesterolemia in apolipoprotein E-null mice. Front Biosci 2007;12: (201) Gonzalez-Navarro H, Vinue A, Vila-Caballer M, Fortuno A, Beloqui O, Zalba G, Burks D, Diez J, Andres V. Molecular mechanisms of atherosclerosis in metabolic syndrome: role of reduced IRS2-dependent signaling. Arterioscler Thromb Vasc Biol 2008 December;28(12): (202) Lewis GF, Carpentier A, Adeli K, Giacca A. Disordered fat storage and mobilization in the pathogenesis of insulin resistance and type 2 diabetes. Endocr Rev 2002 April;23(2): (203) Kubota N, Tobe K, Terauchi Y, Eto K, Yamauchi T, Suzuki R, Tsubamoto Y, Komeda K, Nakano R, Miki H, Satoh S, Sekihara H, Sciacchitano S, Lesniak M, Aizawa S, Nagai R, Kimura S, Akanuma Y, Taylor SI, Kadowaki T. Disruption of insulin receptor substrate 2 causes type 2 diabetes because of liver insulin resistance and lack of compensatory beta-cell hyperplasia. Diabetes 2000 November;49(11): (204) Han S, Liang CP, DeVries-Seimon T, Ranalletta M, Welch CL, Collins-Fletcher K, Accili D, Tabas I, Tall AR. Macrophage insulin receptor deficiency increases ER stress-induced apoptosis and necrotic core formation in advanced atherosclerotic lesions. Cell Metab 2006 April;3(4):

180 (205) Baumgartl J, Baudler S, Scherner M, Babaev V, Makowski L, Suttles J, McDuffie M, Tobe K, Kadowaki T, Fazio S, Kahn CR, Hotamisligil GS, Krone W, Linton M, Bruning JC. Myeloid lineage cell-restricted insulin resistance protects apolipoproteine-deficient mice against atherosclerosis. Cell Metab 2006 April;3(4): (206) Rask-Madsen C. Loss of insulin signalling in vascular endothelium accelerates atherosclerosis in apolipoprotein E knockout mice. Diabetes 56 (Suppl 1), A Ref Type: Abstract (207) Ross R. Atherosclerosis--an inflammatory disease. N Engl J Med 1999 January 14;340(2): (208) Harrington SC, Simari RD, Conover CA. Genetic deletion of pregnancy-associated plasma protein-a is associated with resistance to atherosclerotic lesion development in apolipoprotein E-deficient mice challenged with a high-fat diet. Circ Res 2007 June 22;100(12): (209) Sukhanov S, Higashi Y, Shai SY, Vaughn C, Mohler J, Li Y, Song YH, Titterington J, Delafontaine P. IGF-1 reduces inflammatory responses, suppresses oxidative stress, and decreases atherosclerosis progression in ApoE-deficient mice. Arterioscler Thromb Vasc Biol 2007 December;27(12): (210) Bornfeldt KE, Arnqvist HJ, Capron L. In vivo proliferation of rat vascular smooth muscle in relation to diabetes mellitus insulin-like growth factor I and insulin. Diabetologia 1992 February;35(2): (211) Schiller NK, McNamara DB. Balloon catheter vascular injury of the alloxan-induced diabetic rabbit: the role of insulin-like growth factor-1. Mol Cell Biochem 1999 December;202(1-2): (212) Park SH, Marso SP, Zhou Z, Foroudi F, Topol EJ, Lincoff AM. Neointimal hyperplasia after arterial injury is increased in a rat model of non-insulin-dependent diabetes mellitus. Circulation 2001 August 14;104(7): (213) Chajara A, Raoudi M, Delpech B, Leroy M, Basuyau JP, Levesque H. Increased hyaluronan and hyaluronidase production and hyaluronan degradation in injured aorta of insulin-resistant rats. Arterioscler Thromb Vasc Biol 2000 June;20(6): (214) Jonas M, Edelman ER, Groothuis A, Baker AB, Seifert P, Rogers C. Vascular neointimal formation and signaling pathway activation in response to stent injury in insulin-resistant and diabetic animals. Circ Res 2005 September 30;97(7): (215) Ridray S, Ktorza A, Picon L, Capron L. In vivo effect of insulin on the acute proliferative response of the rat aorta to injury. Arterioscler Thromb 1992 May;12(5): (216) Indolfi C, Torella D, Cavuto L, Davalli AM, Coppola C, Esposito G, Carriero MV, Rapacciuolo A, Di Lorenzo E, Stabile E, Perrino C, Chieffo A, Pardo F, Chiariello M. Effects of balloon injury on neointimal hyperplasia in streptozotocin-induced diabetes and in hyperinsulinemic nondiabetic pancreatic islet-transplanted rats. Circulation 2001 June 19;103(24):

181 (217) Kubota T, Kubota N, Moroi M, Terauchi Y, Kobayashi T, Kamata K, Suzuki R, Tobe K, Namiki A, Aizawa S, Nagai R, Kadowaki T, Yamaguchi T. Lack of insulin receptor substrate- 2 causes progressive neointima formation in response to vessel injury. Circulation 2003 June 24;107(24): (218) Dhaliwall JK, Chan KK, Al Koudsi N, Lam L, Madadi G, Bendeck MP, Giacca A. In vivo effect of insulin to decrease matrix metalloproteinase (MMP)-2 activity. Diabetes (Suppl.1) 54, A Ref Type: Abstract (219) Foster E, Zhang S, Kahn AM. Insulin stimulates arterial neointima formation in normal rats after balloon injury. Diabetes Obes Metab 2006 May;8(3): (220) Thresher JS, Podolin DA, Wei Y, Mazzeo RS, Pagliassotti MJ. Comparison of the effects of sucrose and fructose on insulin action and glucose tolerance. Am J Physiol Regul Integr Comp Physiol 2000 October;279(4):R1334-R1340. (221) Pu Q, Chang Y, Zhang C, Cai Y, Hassid A. Chronic insulin treatment suppresses PTP1B function, induces increased PDGF signaling, and amplifies neointima formation in the ballooninjured rat artery. Am J Physiol Heart Circ Physiol 2009 January;296(1):H132-H139. (222) Murthy SN, Sukhanov S, McGee J, Greco JA, Chandra S, Delafontaine P, Kadowitz PJ, McNamara DB, Fonseca VA. Insulin glargine reduces carotid intimal hyperplasia after balloon catheter injury in Zucker fatty rats possibly by reduction in oxidative stress. Mol Cell Biochem 2009 April 10. (223) Hayry P, Myllarniemi M, Aavik E, Alatalo S, Aho P, Yilmaz S, Raisanen-Sokolowski A, Cozzone G, Jameson BA, Baserga R. Stabile D-peptide analog of insulin-like growth factor-1 inhibits smooth muscle cell proliferation after carotid ballooning injury in the rat. FASEB J 1995 October;9(13): (224) Lim HJ, Park HY, Ko YG, Lee SH, Cho SY, Lee EJ, Jameson JL, Jang Y. Dominant negative insulin-like growth factor-1 receptor inhibits neointimal formation through suppression of vascular smooth muscle cell migration and proliferation, and induction of apoptosis. Biochem Biophys Res Commun 2004 December 17;325(3): (225) Zhu B, Zhao G, Witte DP, Hui DY, Fagin JA. Targeted overexpression of IGF-I in smooth muscle cells of transgenic mice enhances neointimal formation through increased proliferation and cell migration after intraarterial injury. Endocrinology 2001 August;142(8): (226) Taniguchi CM, Emanuelli B, Kahn CR. Critical nodes in signalling pathways: insights into insulin action. Nat Rev Mol Cell Biol 2006 February;7(2): (227) Dong LQ, Liu F. PDK2: the missing piece in the receptor tyrosine kinase signaling pathway puzzle. Am J Physiol Endocrinol Metab 2005 August;289(2):E187-E196. (228) Dimmeler S, Fleming I, Fisslthaler B, Hermann C, Busse R, Zeiher AM. Activation of nitric oxide synthase in endothelial cells by Akt-dependent phosphorylation. Nature 1999 June 10;399(6736):

182 (229) Proud CG. Regulation of protein synthesis by insulin. Biochem Soc Trans 2006 April;34(Pt 2): (230) Brazil DP, Hemmings BA. Ten years of protein kinase B signalling: a hard Akt to follow. Trends Biochem Sci 2001 November;26(11): (231) Skolnik EY, Batzer A, Li N, Lee CH, Lowenstein E, Mohammadi M, Margolis B, Schlessinger J. The function of GRB2 in linking the insulin receptor to Ras signaling pathways. Science 1993 June 25;260(5116): (232) Vinciguerra M, Foti M. PTEN and SHIP2 phosphoinositide phosphatases as negative regulators of insulin signalling. Arch Physiol Biochem 2006 April;112(2): (233) Hsu MF, Meng TC. Enhancement of Insulin Responsiveness by Nitric Oxide-mediated Inactivation of Protein-tyrosine Phosphatases. J Biol Chem 2010 March 12;285(11): (234) Gonzalez M, Flores C, Pearson JD, Casanello P, Sobrevia L. Cell signalling-mediating insulin increase of mrna expression for cationic amino acid transporters-1 and -2 and membrane hyperpolarization in human umbilical vein endothelial cells. Pflugers Arch 2004 July;448(4): (235) Montagnani M, Chen H, Barr VA, Quon MJ. Insulin-stimulated activation of enos is independent of Ca2+ but requires phosphorylation by Akt at Ser(1179). J Biol Chem 2001 August 10;276(32): (236) Montagnani M, Golovchenko I, Kim I, Koh GY, Goalstone ML, Mundhekar AN, Johansen M, Kucik DF, Quon MJ, Draznin B. Inhibition of phosphatidylinositol 3-kinase enhances mitogenic actions of insulin in endothelial cells. J Biol Chem 2002 January 18;277(3): (237) Hartell NA, Archer HE, Bailey CJ. Insulin-stimulated endothelial nitric oxide release is calcium independent and mediated via protein kinase B. Biochem Pharmacol 2005 March 1;69(5): (238) Yuan ZR, Liu B, Zhang Y, Yuan L, Muteliefu G, Lu J. Upregulated expression of neuronal nitric oxide synthase by insulin in both neurons and astrocytes. Brain Res 2004 May 15;1008(1):1-10. (239) Begum N, Ragolia L, Rienzie J, McCarthy M, Duddy N. Regulation of mitogen-activated protein kinase phosphatase-1 induction by insulin in vascular smooth muscle cells. Evaluation of the role of the nitric oxide signaling pathway and potential defects in hypertension. J Biol Chem 1998 September 25;273(39): (240) Kahn AM, Allen JC, Seidel CL, Lichtenberg DS, Song T, Zhang S. Insulin increases NOstimulated guanylate cyclase activity in cultured VSMC while raising redox potential. Am J Physiol Endocrinol Metab 2000 April;278(4):E627-E633. (241) Salt IP, Morrow VA, Brandie FM, Connell JM, Petrie JR. High glucose inhibits insulinstimulated nitric oxide production without reducing endothelial nitric-oxide synthase Ser1177 phosphorylation in human aortic endothelial cells. J Biol Chem 2003 May 23;278(21):

183 (242) Kim F, Gallis B, Corson MA. TNF-alpha inhibits flow and insulin signaling leading to NO production in aortic endothelial cells. Am J Physiol Cell Physiol 2001 May;280(5):C1057- C1065. (243) Ritchie SA, Kohlhaas CF, Boyd AR, Yalla KC, Walsh K, Connell JM, Salt IP. Insulinstimulated phosphorylation of endothelial nitric oxide synthase at serine-615 contributes to nitric oxide synthesis. Biochem J 2010 February 15;426(1): (244) Takahashi S, Mendelsohn ME. Synergistic activation of endothelial nitric-oxide synthase (enos) by HSP90 and Akt: calcium-independent enos activation involves formation of an HSP90-Akt-CaM-bound enos complex. J Biol Chem 2003 August 15;278(33): (245) Clark MG, Wallis MG, Barrett EJ, Vincent MA, Richards SM, Clerk LH, Rattigan S. Blood flow and muscle metabolism: a focus on insulin action. Am J Physiol Endocrinol Metab 2003 February;284(2):E241-E258. (246) Vehkavaara S, Makimattila S, Schlenzka A, Vakkilainen J, Westerbacka J, Yki-Jarvinen H. Insulin therapy improves endothelial function in type 2 diabetes. Arterioscler Thromb Vasc Biol 2000 February;20(2): (247) Schroeder CA, Jr., Chen YL, Messina EJ. Inhibition of NO synthesis or endothelium removal reveals a vasoconstrictor effect of insulin on isolated arterioles. Am J Physiol 1999 March;276(3 Pt 2):H815-H820. (248) Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi M, Mitsui Y, Yazaki Y, Goto K, Masaki T. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature 1988 March 31;332(6163): (249) Juan CC, Shen YW, Chien Y, Lin YJ, Chang SF, Ho LT. Insulin infusion induces endothelin- 1-dependent hypertension in rats. Am J Physiol Endocrinol Metab 2004 November;287(5):E948-E954. (250) Eringa EC, Stehouwer CD, Merlijn T, Westerhof N, Sipkema P. Physiological concentrations of insulin induce endothelin-mediated vasoconstriction during inhibition of NOS or PI3-kinase in skeletal muscle arterioles. Cardiovasc Res 2002 December;56(3): (251) Eringa EC, Stehouwer CD, van Nieuw Amerongen GP, Ouwehand L, Westerhof N, Sipkema P. Vasoconstrictor effects of insulin in skeletal muscle arterioles are mediated by ERK1/2 activation in endothelium. Am J Physiol Heart Circ Physiol 2004 November;287(5):H2043- H2048. (252) Irving RJ, Noon JP, Watt GC, Webb DJ, Walker BR. Activation of the endothelin system in insulin resistance. QJM 2001 June;94(6): (253) Westerbacka J, Wilkinson I, Cockcroft J, Utriainen T, Vehkavaara S, Yki-Jarvinen H. Diminished wave reflection in the aorta. A novel physiological action of insulin on large blood vessels. Hypertension 1999 May;33(5): (254) Randriamboavonjy V, Schrader J, Busse R, Fleming I. Insulin induces the release of vasodilator compounds from platelets by a nitric oxide-g kinase-vamp-3-dependent pathway. J Exp Med 2004 February 2;199(3):

184 (255) Trovati M, Mularoni EM, Burzacca S, Ponziani MC, Massucco P, Mattiello L, Piretto V, Cavalot F, Anfossi G. Impaired insulin-induced platelet antiaggregating effect in obesity and in obese NIDDM patients. Diabetes 1995 November;44(11): (256) Hsueh WA, Quinones MJ. Role of endothelial dysfunction in insulin resistance. Am J Cardiol 2003 August 18;92(4A):10J-7J. (257) Dandona P, Aljada A, Dhindsa S, Garg R. Insulin as an anti-inflammatory and antiatherosclerotic hormone. Clin Cornerstone 2003;Suppl 4:S13-S20. (258) Jeschke MG, Klein D, Bolder U, Einspanier R. Insulin attenuates the systemic inflammatory response in endotoxemic rats. Endocrinology 2004 September;145(9): (259) Jeschke MG, Einspanier R, Klein D, Jauch KW. Insulin attenuates the systemic inflammatory response to thermal trauma. Mol Med 2002 August;8(8): (260) Cuschieri J, Bulger E, Grinsell R, Garcia I, Maier RV. Insulin regulates macrophage activation through activin A. Shock 2008 February;29(2): (261) Aljada A, Saadeh R, Assian E, Ghanim H, Dandona P. Insulin inhibits the expression of intercellular adhesion molecule-1 by human aortic endothelial cells through stimulation of nitric oxide. J Clin Endocrinol Metab 2000 July;85(7): (262) Dandona P, Aljada A, Mohanty P, Ghanim H, Hamouda W, Assian E, Ahmad S. Insulin inhibits intranuclear nuclear factor kappab and stimulates IkappaB in mononuclear cells in obese subjects: evidence for an anti-inflammatory effect? J Clin Endocrinol Metab 2001 July;86(7): (263) Aljada A, Ghanim H, Mohanty P, Kapur N, Dandona P. Insulin inhibits the pro-inflammatory transcription factor early growth response gene-1 (Egr)-1 expression in mononuclear cells (MNC) and reduces plasma tissue factor (TF) and plasminogen activator inhibitor-1 (PAI-1) concentrations. J Clin Endocrinol Metab 2002 March;87(3): (264) Koskenkari JK, Kaukoranta PK, Rimpilainen J, Vainionpaa V, Ohtonen PP, Surcel HM, Juvonen T, la-kokko TI. Anti-inflammatory effect of high-dose insulin treatment after urgent coronary revascularization surgery. Acta Anaesthesiol Scand 2006 September;50(8): (265) Begum N, Song Y, Rienzie J, Ragolia L. Vascular smooth muscle cell growth and insulin regulation of mitogen-activated protein kinase in hypertension. Am J Physiol 1998 July;275(1 Pt 1):C42-C49. (266) Pfeifle B, Ditschuneit H. Effect of insulin on growth of cultured human arterial smooth muscle cells. Diabetologia 1981 February;20(2): (267) Ledet T. Growth hormone stimulating the growth of arterial medial cells in vitro. Absence of effect of insulin. Diabetes 1976 November;25(11): (268) Capron L, Jarnet J, Kazandjian S, Housset E. Growth-promoting effects of diabetes and insulin on arteries. An in vivo study of rat aorta. Diabetes 1986 September;35(9):

185 (269) Nakao J, Ito H, Kanayasu T, Murota S. Stimulatory effect of insulin on aortic smooth muscle cell migration induced by 12-L-hydroxy-5,8,10,14-eicosatetraenoic acid and its modulation by elevated extracellular glucose levels. Diabetes 1985 February;34(2): (270) Gockerman A, Prevette T, Jones JI, Clemmons DR. Insulin-like growth factor (IGF)-binding proteins inhibit the smooth muscle cell migration responses to IGF-I and IGF-II. Endocrinology 1995 October;136(10): (271) Yang M, Yang Y, Zhang S, Kahn AM. Insulin-stimulated hydrogen peroxide increases guanylate cyclase activity in vascular smooth muscle. Hypertension 2003 October;42(4): (272) Kahn AM, Allen JC, Seidel CL, Zhang S. Insulin inhibits migration of vascular smooth muscle cells with inducible nitric oxide synthase. Hypertension 2000 January;35(1 Pt 2): (273) Zhang S, Yang Y, Kone BC, Allen JC, Kahn AM. Insulin-stimulated cyclic guanosine monophosphate inhibits vascular smooth muscle cell migration by inhibiting Ca/calmodulindependent protein kinase II. Circulation 2003 March 25;107(11): (274) Huang C, Somwar R, Patel N, Niu W, Torok D, Klip A. Sustained exposure of L6 myotubes to high glucose and insulin decreases insulin-stimulated GLUT4 translocation but upregulates GLUT4 activity. Diabetes 2002 July;51(7): (275) Bertelli AA, Das DK. Grapes, wines, resveratrol, and heart health. J Cardiovasc Pharmacol 2009 December;54(6): (276) Deroo BJ, Korach KS. Estrogen receptors and human disease. J Clin Invest 2006 March;116(3): (277) Aggarwal BB, Bhardwaj A, Aggarwal RS, Seeram NP, Shishodia S, Takada Y. Role of resveratrol in prevention and therapy of cancer: preclinical and clinical studies. Anticancer Res 2004 September;24(5A): (278) Zang M, Xu S, Maitland-Toolan KA, Zuccollo A, Hou X, Jiang B, Wierzbicki M, Verbeuren TJ, Cohen RA. Polyphenols Stimulate AMP-Activated Protein Kinase, Lower Lipids, and Inhibit Accelerated Atherosclerosis in Diabetic LDL Receptor-Deficient Mice. Diabetes 2006 August;55(8): (279) Breen DM, Sanli T, Giacca A, Tsiani E. Stimulation of muscle cell glucose uptake by resveratrol through sirtuins and AMPK. Biochem Biophys Res Commun 2008 September 12;374(1): (280) Su HC, Hung LM, Chen JK. Resveratrol, a red wine antioxidant, possesses an insulin-like effect in streptozotocin-induced diabetic rats. Am J Physiol Endocrinol Metab 2006 January 24. (281) Um JH, Park SJ, Kang H, Yang S, Foretz M, McBurney MW, Kim MK, Viollet B, Chung JH. AMPK-deficient mice are resistant to the metabolic effects of resveratrol. Diabetes 2009 November 23. (282) Baur JA, Pearson KJ, Price NL, Jamieson HA, Lerin C, Kalra A, Prabhu VV, Allard JS, Lopez-Lluch G, Lewis K, Pistell PJ, Poosala S, Becker KG, Boss O, Gwinn D, Wang M, 173

186 Ramaswamy S, Fishbein KW, Spencer RG, Lakatta EG, Le CD, Shaw RJ, Navas P, Puigserver P, Ingram DK, de CR, Sinclair DA. Resveratrol improves health and survival of mice on a high-calorie diet. Nature 2006 November 16;444(7117): (283) Baur JA, Sinclair DA. Therapeutic potential of resveratrol: the in vivo evidence. Nat Rev Drug Discov 2006 June;5(6): (284) Pacholec M, Bleasdale JE, Chrunyk B, Cunningham D, Flynn D, Garofalo RS, Griffith D, Griffor M, Loulakis P, Pabst B, Qiu X, Stockman B, Thanabal V, Varghese A, Ward J, Withka J, Ahn K. SRT1720, SRT2183, SRT1460, and Resveratrol Are Not Direct Activators of SIRT1. J Biol Chem 2010 March 12;285(11): (285) Wilson T, Knight TJ, Beitz DC, Lewis DS, Engen RL. Resveratrol promotes atherosclerosis in hypercholesterolemic rabbits. Life Sci 1996;59(1):L15-L21. (286) Wang Z, Zou J, Cao K, Hsieh TC, Huang Y, Wu JM. Dealcoholized red wine containing known amounts of resveratrol suppresses atherosclerosis in hypercholesterolemic rabbits without affecting plasma lipid levels. Int J Mol Med 2005 October;16(4): (287) Fukao H, Ijiri Y, Miura M, Hashimoto M, Yamashita T, Fukunaga C, Oiwa K, Kawai Y, Suwa M, Yamamoto J. Effect of trans-resveratrol on the thrombogenicity and atherogenicity in apolipoprotein E-deficient and low-density lipoprotein receptor-deficient mice. Blood Coagul Fibrinolysis 2004 September;15(6): (288) Do GM, Kwon EY, Kim HJ, Jeon SM, Ha TY, Park T, Choi MS. Long-term effects of resveratrol supplementation on suppression of atherogenic lesion formation and cholesterol synthesis in apo E-deficient mice. Biochem Biophys Res Commun 2008 September 12;374(1):55-9. (289) Norata GD, Marchesi P, Passamonti S, Pirillo A, Violi F, Catapano AL. Anti-inflammatory and anti-atherogenic effects of cathechin, caffeic acid and trans-resveratrol in apolipoprotein E deficient mice. Atherosclerosis 2006 June 24. (290) Lefevre J, Michaud SE, Haddad P, Dussault S, Menard C, Groleau J, Turgeon J, Rivard A. Moderate consumption of red wine (cabernet sauvignon) improves ischemia-induced neovascularization in ApoE-deficient mice: effect on endothelial progenitor cells and nitric oxide. FASEB J 2007 December;21(14): (291) Zou J, Huang Y, Cao K, Yang G, Yin H, Len J, Hsieh TC, Wu JM. Effect of resveratrol on intimal hyperplasia after endothelial denudation in an experimental rabbit model. Life Sci 2000 December 1;68(2): (292) J G, Cq W, Hh F, Hy D, Xl X, Ym X, By W, Dj H. Effects of resveratrol on endothelial progenitor cells and their contributions to reendothelialization in intima-injured rats. J Cardiovasc Pharmacol 2006 May;47(5): (293) Hung CF, Fang CL, Liao MH, Fang JY. The effect of oil components on the physicochemical properties and drug delivery of emulsions: tocol emulsion versus lipid emulsion. Int J Pharm 2007 April 20;335(1-2):

187 (294) Nishimura M, Hashimoto T, Kobayashi H, Yamazaki S, Okino K, Fujita H, Inoue N, Takahashi H, Ono T. Association of the circulating adiponectin concentration with coronary instent restenosis in haemodialysis patients. Nephrol Dial Transplant 2006 June;21(6): (295) Nagata D, Takeda R, Sata M, Satonaka H, Suzuki E, Nagano T, Hirata Y. AMP-activated protein kinase inhibits angiotensin II-stimulated vascular smooth muscle cell proliferation. Circulation 2004 July 27;110(4): (296) Lancon A, Delma D, Osman H, Thenot JP, Jannin B, Latruffe N. Human hepatic cell uptake of resveratrol: involvement of both passive diffusion and carrier-mediated process. Biochem Biophys Res Commun 2004 April 16;316(4): (297) Singh NP, Hegde VL, Hofseth LJ, Nagarkatti M, Nagarkatti P. Resveratrol (trans-3,5,4'- trihydroxystilbene) ameliorates experimental allergic encephalomyelitis, primarily via induction of apoptosis in T cells involving activation of aryl hydrocarbon receptor and estrogen receptor. Mol Pharmacol 2007 December;72(6): (298) Hambrock A, de Oliveira Franz CB, Hiller S, Grenz A, Ackermann S, Schulze DU, Drews G, Osswald H. Resveratrol binds to the sulfonylurea receptor (SUR) and induces apoptosis in a SUR subtype-specific manner. J Biol Chem 2007 February 2;282(5): (299) Yu HP, Hwang TL, Hwang TL, Yen CH, Lau YT. Resveratrol prevents endothelial dysfunction and aortic superoxide production after trauma hemorrhage through estrogen receptor-dependent hemeoxygenase-1 pathway. Crit Care Med 2010 January 14. (300) Taubert D, Berkels R. Upregulation and activation of enos by resveratrol. Circulation 2003 March 25;107(11):e78-e79. (301) Das S, Alagappan VK, Bagchi D, Sharma HS, Maulik N, Das DK. Coordinated induction of inos-vegf-kdr-enos after resveratrol consumption: a potential mechanism for resveratrol preconditioning of the heart. Vascul Pharmacol 2005 April;42(5-6): (302) Tsai SH, Lin-Shiau SY, Lin JK. Suppression of nitric oxide synthase and the down-regulation of the activation of NFkappaB in macrophages by resveratrol. Br J Pharmacol 1999 February;126(3): (303) Cignarella A, Minici C, Bolego C, Pinna C, Sanvito P, Gaion RM, Puglisi L. Potential proinflammatory action of resveratrol in vascular smooth muscle cells from normal and diabetic rats. Nutr Metab Cardiovasc Dis 2006 July;16(5): (304) Klinge CM, Wickramasinghe NS, Ivanova MM, Dougherty SM. Resveratrol stimulates nitric oxide production by increasing estrogen receptor alpha-src-caveolin-1 interaction and phosphorylation in human umbilical vein endothelial cells. FASEB J 2008 July;22(7): (305) Takahashi S, Uchiyama T, Toda K. Differential effect of resveratrol on nitric oxide production in endothelial f-2 cells. Biol Pharm Bull 2009 November;32(11): (306) Chen ZP, Mitchelhill KI, Michell BJ, Stapleton D, Rodriguez-Crespo I, Witters LA, Power DA, Ortiz de Montellano PR, Kemp BE. AMP-activated protein kinase phosphorylation of endothelial NO synthase. FEBS Lett 1999 January 29;443(3):

188 (307) Csiszar A, Labinskyy N, Pinto JT, Ballabh P, Zhang H, Losonczy G, Pearson K, de CR, Pacher P, Zhang C, Ungvari Z. Resveratrol induces mitochondrial biogenesis in endothelial cells. Am J Physiol Heart Circ Physiol 2009 July;297(1):H13-H20. (308) Rush JW, Quadrilatero J, Levy AS, Ford RJ. Chronic resveratrol enhances endotheliumdependent relaxation but does not alter enos levels in aorta of spontaneously hypertensive rats. Exp Biol Med (Maywood ) 2007 June;232(6): (309) El-Mowafy AM, Alkhalaf M, Nassar NN. Resveratrol reverses ET-1-evoked mitogenic effects in human coronary arterial cells by activating the kinase-g to inhibit ERK-enzymes. Int J Cardiol 2009 August 21;136(3): (310) El Mowafy AM, White RE. Resveratrol inhibits MAPK activity and nuclear translocation in coronary artery smooth muscle: reversal of endothelin-1 stimulatory effects. FEBS Lett 1999 May 14;451(1):63-7. (311) Liu Z, Song Y, Zhang X, Liu Z, Zhang W, Mao W, Wang W, Cui W, Zhang X, Jia X, Li N, Han C, Liu C. Effects of trans-resveratrol on hypertension-induced cardiac hypertrophy using the partially nephrectomized rat model. Clin Exp Pharmacol Physiol 2005 December;32(12): (312) Miatello R, Vazquez M, Renna N, Cruzado M, Zumino AP, Risler N. Chronic administration of resveratrol prevents biochemical cardiovascular changes in fructose-fed rats. Am J Hypertens 2005 June;18(6): (313) Kirk RI, Deitch JA, Wu JM, Lerea KM. Resveratrol decreases early signaling events in washed platelets but has little effect on platalet in whole food. Blood Cells Mol Dis 2000 April;26(2): (314) Wang Z, Huang Y, Zou J, Cao K, Xu Y, Wu JM. Effects of red wine and wine polyphenol resveratrol on platelet aggregation in vivo and in vitro. Int J Mol Med 2002 January;9(1):77-9. (315) Godichaud S, Krisa S, Couronne B, Dubuisson L, Merillon JM, Desmouliere A, Rosenbaum J. Deactivation of cultured human liver myofibroblasts by trans-resveratrol, a grapevine-derived polyphenol. Hepatology 2000 April;31(4): (316) Pendurthi UR, Williams JT, Rao LV. Resveratrol, a polyphenolic compound found in wine, inhibits tissue factor expression in vascular cells : A possible mechanism for the cardiovascular benefits associated with moderate consumption of wine. Arterioscler Thromb Vasc Biol 1999 February;19(2): (317) Bertelli AA, Baccalini R, Battaglia E, Falchi M, Ferrero ME. Resveratrol inhibits TNF alphainduced endothelial cell activation. Therapie 2001 September;56(5): (318) Gresele P, Pignatelli P, Guglielmini G, Carnevale R, Mezzasoma AM, Ghiselli A, Momi S, Violi F. Resveratrol, at concentrations attainable with moderate wine consumption, stimulates human platelet nitric oxide production. J Nutr 2008 September;138(9): (319) Khanduja KL, Bhardwaj A, Kaushik G. Resveratrol inhibits N-nitrosodiethylamine-induced ornithine decarboxylase and cyclooxygenase in mice. J Nutr Sci Vitaminol (Tokyo) 2004 February;50(1):

189 (320) MacCarrone M, Lorenzon T, Guerrieri P, Agro AF. Resveratrol prevents apoptosis in K562 cells by inhibiting lipoxygenase and cyclooxygenase activity. Eur J Biochem 1999 October 1;265(1): (321) Martinez J, Moreno JJ. Effect of resveratrol, a natural polyphenolic compound, on reactive oxygen species and prostaglandin production. Biochem Pharmacol 2000 April 1;59(7): (322) Pace-Asciak CR, Rounova O, Hahn SE, Diamandis EP, Goldberg DM. Wines and grape juices as modulators of platelet aggregation in healthy human subjects. Clin Chim Acta 1996 March 15;246(1-2): (323) Fremont L, Belguendouz L, Delpal S. Antioxidant activity of resveratrol and alcohol-free wine polyphenols related to LDL oxidation and polyunsaturated fatty acids. Life Sci 1999;64(26): (324) Fremont L. Biological effects of resveratrol. Life Sci 2000 January 14;66(8): (325) Zou JG, Huang YZ, Chen Q, Wei EH, Hsieh TC, Wu JM. Resveratrol inhibits copper ioninduced and azo compound-initiated oxidative modification of human low density lipoprotein. Biochem Mol Biol Int 1999 June;47(6): (326) Fauconneau B, Waffo-Teguo P, Huguet F, Barrier L, Decendit A, Merillon JM. Comparative study of radical scavenger and antioxidant properties of phenolic compounds from Vitis vinifera cell cultures using in vitro tests. Life Sci 1997;61(21): (327) Brito P, Almeida LM, Dinis TC. The interaction of resveratrol with ferrylmyoglobin and peroxynitrite; protection against LDL oxidation. Free Radic Res 2002 June;36(6): (328) Jang M, Pezzuto JM. Cancer chemopreventive activity of resveratrol. Drugs Exp Clin Res 1999;25(2-3): (329) Zhong M, Cheng GF, Wang WJ, Guo Y, Zhu XY, Zhang JT. Inhibitory effect of resveratrol on interleukin 6 release by stimulated peritoneal macrophages of mice. Phytomedicine 1999 May;6(2): (330) Ahn KS, Kim JH, Oh SR, Ryu SY, Lee HK. Inhibitory activity of stilbenes from medicinal plants on the expression of cell adhesion molecules on THP1 cells. Planta Med 2000 October;66(7): (331) Dandona P, Mohanty P, Ghanim H, Marumganti A, Sia CL, Abuaysheh S, Rajeswari J. Resveratrol suppresses pro-inflammatory genes causing insulin resistance. Diabetes 57 (Suppl 1), A Ref Type: Abstract (332) Zou J, Huang Y, Chen Q, Wang N, Cao K, Hsieh TC, Wu JM. Suppression of mitogenesis and regulation of cell cycle traverse by resveratrol in cultured smooth muscle cells. Int J Oncol 1999 October;15(4): (333) Ekshyyan VP, Hebert VY, Khandelwal A, Dugas TR. Resveratrol inhibits rat aortic vascular smooth muscle cell proliferation via estrogen receptor dependent nitric oxide production. J Cardiovasc Pharmacol 2007 July;50(1):

190 (334) Mnjoyan ZH, Fujise K. Profound negative regulatory effects by resveratrol on vascular smooth muscle cells: a role of p53-p21(waf1/cip1) pathway. Biochem Biophys Res Commun 2003 November 14;311(2): (335) Jin N, Siddiqui RA, English D, Rhoades RA. Communication between tyrosine kinase pathway and myosin light chain kinase pathway in smooth muscle. Am J Physiol 1996 October;271(4 Pt 2):H1348-H1355. (336) Lin JN, Lin VC, Rau KM, Shieh PC, Kuo DH, Shieh JC, Chen WJ, Tsai SC, Way TD. Resveratrol Modulates Tumor Cell Proliferation and Protein Translation via SIRT1-Dependent AMPK Activation. J Agric Food Chem 2009 November 20. (337) Poussier B, Cordova AC, Becquemin JP, Sumpio BE. Resveratrol inhibits vascular smooth muscle cell proliferation and induces apoptosis. J Vasc Surg 2005 December;42(6): (338) Szende B, Tyihak E, Kiraly-Veghely Z. Dose-dependent effect of resveratrol on proliferation and apoptosis in endothelial and tumor cell cultures. Exp Mol Med 2000 June 30;32(2): (339) Klinge CM, Blankenship KA, Risinger KE, Bhatnagar S, Noisin EL, Sumanasekera WK, Zhao L, Brey DM, Keynton RS. Resveratrol and estradiol rapidly activate MAPK signaling through estrogen receptors alpha and beta in endothelial cells. J Biol Chem 2005 March 4;280(9): (340) In K, Park J, Park H. Resveratrol at high doses acts as an apoptotic inducer in endothelial cells. Cancer Res Treat 2006 February;38(1): (341) Venkatesan B, Valente AJ, Reddy VS, Siwik DA, Chandrasekar B. Resveratrol blocks interleukin-18-emmprin cross-regulation and smooth muscle cell migration. Am J Physiol Heart Circ Physiol 2009 August;297(2):H874-H886. (342) Wang XB, Huang J, Zou JG, Su EB, Shan QJ, Yang ZJ, Cao KJ. Effects of resveratrol on number and activity of endothelial progenitor cells from human peripheral blood. Clin Exp Pharmacol Physiol 2007 November;34(11): (343) Gagliano N, Moscheni C, Torri C, Magnani I, Bertelli AA, Gioia M. Effect of resveratrol on matrix metalloproteinase-2 (MMP-2) and Secreted Protein Acidic and Rich in Cysteine (SPARC) on human cultured glioblastoma cells. Biomed Pharmacother 2005 August;59(7): (344) Lee B, Moon SK. Resveratrol inhibits TNF-alpha-induced proliferation and matrix metalloproteinase expression in human vascular smooth muscle cells. J Nutr 2005 December;135(12): (345) Nagata D, Mogi M, Walsh K. AMP-activated protein kinase (AMPK) signaling in endothelial cells is essential for angiogenesis in response to hypoxic stress. J Biol Chem 2003 August 15;278(33): (346) Kanellis J, Kandane RK, Etemadmoghadam D, Fraser SA, Mount PF, Levidiotis V, Kemp BE, Power DA. Activators of the energy sensing kinase AMPK inhibit random cell movement and chemotaxis in U937 cells. Immunol Cell Biol 2006 February;84(1):

191 (347) Jiang H, Shang X, Wu H, Gautam SC, Al-Holou S, Li C, Kuo J, Zhang L, Chopp M. Resveratrol downregulates PI3K/Akt/mTOR signaling pathways in human U251 glioma cells. J Exp Ther Oncol 2009;8(1): (348) Saltiel AR. New perspectives into the molecular pathogenesis and treatment of type 2 diabetes. Cell 2001 February 23;104(4): (349) Hsueh WA, Jackson S, Law RE. Control of vascular cell proliferation and migration by PPARgamma: a new approach to the macrovascular complications of diabetes. Diabetes Care 2001 February;24(2): (350) Akishita M, Ouchi Y, Miyoshi H, Kozaki K, Inoue S, Ishikawa M, Eto M, Toba K, Orimo H. Estrogen inhibits cuff-induced intimal thickening of rat femoral artery: effects on migration and proliferation of vascular smooth muscle cells. Atherosclerosis 1997 April;130(1-2):1-10. (351) Wang PY. The reliability of a compressed mixture of insulin and palmitic acid to sustain a reduction in hyperglycemia in rodents. ASAIO Trans 1987 July;33(3): (352) Kurtz TW, Griffin KA, Bidani AK, Davisson RL, Hall JE. Recommendations for blood pressure measurement in humans and experimental animals. Part 2: Blood pressure measurement in experimental animals: a statement for professionals from the subcommittee of professional and public education of the American Heart Association council on high blood pressure research. Hypertension 2005 February;45(2): (353) Groop L, Orho-Melander M. The dysmetabolic syndrome. J Intern Med 2001 August;250(2): (354) Grundy SM. Metabolic syndrome: a multiplex cardiovascular risk factor. J Clin Endocrinol Metab 2007 February;92(2): (355) Nystrom T, Gonon AT, Sjoholm A, Pernow J. Glucagon-like peptide-1 relaxes rat conduit arteries via an endothelium-independent mechanism. Regul Pept 2005 February 15;125(1-3): (356) Nystrom T, Gutniak MK, Zhang Q, Zhang F, Holst JJ, Ahren B, Sjoholm A. Effects of glucagon-like peptide-1 on endothelial function in type 2 diabetes patients with stable coronary artery disease. Am J Physiol Endocrinol Metab 2004 December;287(6):E1209-E1215. (357) Kogire M, Inoue K, Sumi S, Doi R, Yun M, Kaji H, Tobe T. Effects of gastric inhibitory polypeptide and glucagon on portal venous and hepatic arterial flow in conscious dogs. Dig Dis Sci 1992 November;37(11): (358) McTernan PG, Harte AL, Anderson LA, Green A, Smith SA, Holder JC, Barnett AH, Eggo MC, Kumar S. Insulin and rosiglitazone regulation of lipolysis and lipogenesis in human adipose tissue in vitro. Diabetes 2002 May;51(5): (359) Brown AT, Smith TP, Cruz CP, Poirier LA, Simmons D, Williams DK, Wang Y, Eidt JF, Moursi MM. Intimal hyperplasia following carotid endarterectomy in an insulin-resistant rat model. Metabolism 2003 July;52(7):

192 (360) Lee MK, Miles PD, Khoursheed M, Gao KM, Moossa AR, Olefsky JM. Metabolic effects of troglitazone on fructose-induced insulin resistance in the rat. Diabetes 1994 December;43(12): (361) Thorburn AW, Storlien LH, Jenkins AB, Khouri S, Kraegen EW. Fructose-induced in vivo insulin resistance and elevated plasma triglyceride levels in rats. Am J Clin Nutr 1989 June;49(6): (362) Koyama K, Chen G, Lee Y, Unger RH. Tissue triglycerides, insulin resistance, and insulin production: implications for hyperinsulinemia of obesity. Am J Physiol 1997 October;273(4 Pt 1):E708-E713. (363) Kazumi T, Vranic M, Steiner G. Triglyceride kinetics: effects of dietary glucose, sucrose, or fructose alone or with hyperinsulinemia. Am J Physiol 1986 March;250(3 Pt 1):E325-E330. (364) Breen DM, Dhaliwall JK, Chan KK, Guo J, Lam L, Bendeck MP, Giacca A. Insulin Inhibits and Oral Sucrose Increases Neointimal Growth after Arterial Injury in Rats. J Vasc Res 2010 February 6;47(5): (365) Young JJ. Neointimal formation following drug-eluting stents: physiology, timeline, and the influence of drug delivery systems. Rev Cardiovasc Med 2007;8 Suppl 1:S3-10. (366) Wohlfart P, Xu H, Endlich A, Habermeier A, Closs EI, Hubschle T, Mang C, Strobel H, Suzuki T, Kleinert H, Forstermann U, Ruetten H, Li H. Antiatherosclerotic effects of smallmolecular-weight compounds enhancing endothelial nitric-oxide synthase (enos) expression and preventing enos uncoupling. J Pharmacol Exp Ther 2008 May;325(2): (367) Cooney R, Hynes SO, Sharif F, Howard L, O'Brien T. Effect of gene delivery of NOS isoforms on intimal hyperplasia and endothelial regeneration after balloon injury. Gene Ther 2007 March;14(5): (368) Arnal JF, Warin L, Michel JB. Determinants of aortic cyclic guanosine monophosphate in hypertension induced by chronic inhibition of nitric oxide synthase. J Clin Invest 1992 August;90(2): (369) Ellenby MI, Ernst CB, Carretero OA, Scicli AG. Role of nitric oxide in the effect of blood flow on neointima formation. J Vasc Surg 1996 February;23(2): (370) Wang CC, Gurevich I, Draznin B. Insulin affects vascular smooth muscle cell phenotype and migration via distinct signaling pathways. Diabetes 2003 October;52(10): (371) Schwartz SM, Stemerman MB, Benditt EP. The aortic intima. II. Repair of the aortic lining after mechanical denudation. Am J Pathol 1975 October;81(1): (372) Hayashi T, Sumi D, Juliet PA, Matsui-Hirai H, Asai-Tanaka Y, Kano H, Fukatsu A, Tsunekawa T, Miyazaki A, Iguchi A, Ignarro LJ. Gene transfer of endothelial NO synthase, but not enos plus inducible NOS, regressed atherosclerosis in rabbits. Cardiovasc Res 2004 February 1;61(2):

193 (373) Moroi M, Zhang L, Yasuda T, Virmani R, Gold HK, Fishman MC, Huang PL. Interaction of genetic deficiency of endothelial nitric oxide, gender, and pregnancy in vascular response to injury in mice. J Clin Invest 1998 March 15;101(6): (374) Fishman JA, Ryan GB, Karnovsky MJ. Endothelial regeneration in the rat carotid artery and the significance of endothelial denudation in the pathogenesis of myointimal thickening. Lab Invest 1975 March;32(3): (375) Fleming I, Schulz C, Fichtlscherer B, Kemp BE, Fisslthaler B, Busse R. AMP-activated protein kinase (AMPK) regulates the insulin-induced activation of the nitric oxide synthase in human platelets. Thromb Haemost 2003 November;90(5): (376) Trovati M, Anfossi G. Insulin, insulin resistance and platelet function: similarities with insulin effects on cultured vascular smooth muscle cells. Diabetologia 1998 June;41(6): (377) Trovati M, Anfossi G, Cavalot F, Massucco P, Mularoni E, Emanuelli G. Insulin directly reduces platelet sensitivity to aggregating agents. Studies in vitro and in vivo. Diabetes 1988 June;37(6): (378) Taguchi J, Abe J, Okazaki H, Takuwa Y, Kurokawa K. L-arginine inhibits neointimal formation following balloon injury. Life Sci 1993;53(23):L387-L392. (379) Gonzalez-Fernandez F, Lopez-Farre A, Rodriguez-Feo JA, Farre J, Guerra J, Fortes J, Millas I, Garcia-Duran M, Rico L, Mata P, de Miguel LS, Casado S. Expression of inducible nitric oxide synthase after endothelial denudation of the rat carotid artery: role of platelets. Circ Res 1998 November 30;83(11): (380) Hayashi K, Nakamura S, Morishita R, Moriguchi A, Aoki M, Matsumoto K, Nakamura T, Kaneda Y, Sakai N, Ogihara T. In vivo transfer of human hepatocyte growth factor gene accelerates re-endothelialization and inhibits neointimal formation after balloon injury in rat model. Gene Ther 2000 October;7(19): (381) Dela F, Larsen JJ, Mikines KJ, Galbo H. Normal effect of insulin to stimulate leg blood flow in NIDDM. Diabetes 1995 February;44(2): (382) Steinberg HO, Brechtel G, Johnson A, Fineberg N, Baron AD. Insulin-mediated skeletal muscle vasodilation is nitric oxide dependent. A novel action of insulin to increase nitric oxide release. J Clin Invest 1994 September;94(3): (383) Clowes AW, Reidy MA, Clowes MM. Kinetics of cellular proliferation after arterial injury. I. Smooth muscle growth in the absence of endothelium. Lab Invest 1983 September;49(3): (384) Signorelli P, Ghidoni R. Resveratrol as an anticancer nutrient: molecular basis, open questions and promises. J Nutr Biochem 2005 August;16(8): (385) Hsieh TC, Juan G, Darzynkiewicz Z, Wu JM. Resveratrol increases nitric oxide synthase, induces accumulation of p53 and p21(waf1/cip1), and suppresses cultured bovine pulmonary artery endothelial cell proliferation by perturbing progression through S and G2. Cancer Res 1999 June 1;59(11):

194 (386) Mizutani K, Ikeda K, Yamori Y. Resveratrol inhibits AGEs-induced proliferation and collagen synthesis activity in vascular smooth muscle cells from stroke-prone spontaneously hypertensive rats. Biochem Biophys Res Commun 2000 July 21;274(1):61-7. (387) Marier JF, Vachon P, Gritsas A, Zhang J, Moreau JP, Ducharme MP. Metabolism and disposition of resveratrol in rats: extent of absorption, glucuronidation, and enterohepatic recirculation evidenced by a linked-rat model. J Pharmacol Exp Ther 2002 July;302(1): (388) Miatello R, Risler N, Castro C, Gonzalez S, Ruttler M, Cruzado M. Aortic smooth muscle cell proliferation and endothelial nitric oxide synthase activity in fructose-fed rats. Am J Hypertens 2001 November;14(11 Pt 1): (389) Morrow VA, Foufelle F, Connell JM, Petrie JR, Gould GW, Salt IP. Direct activation of AMPactivated protein kinase stimulates nitric-oxide synthesis in human aortic endothelial cells. J Biol Chem 2003 August 22;278(34): (390) Borra MT, Smith BC, Denu JM. Mechanism of human SIRT1 activation by resveratrol. J Biol Chem 2005 April 29;280(17): (391) Mattagajasingh I, Kim CS, Naqvi A, Yamamori T, Hoffman TA, Jung SB, DeRicco J, Kasuno K, Irani K. SIRT1 promotes endothelium-dependent vascular relaxation by activating endothelial nitric oxide synthase. Proc Natl Acad Sci U S A 2007 September 11;104(37): (392) Ueno M, Carvalheira JB, Tambascia RC, Bezerra RM, Amaral ME, Carneiro EM, Folli F, Franchini KG, Saad MJ. Regulation of insulin signalling by hyperinsulinaemia: role of IRS-1/2 serine phosphorylation and the mtor/p70 S6K pathway. Diabetologia 2005 March;48(3): (393) Urao N, Okigaki M, Yamada H, Aadachi Y, Matsuno K, Matsui A, Matsunaga S, Tateishi K, Nomura T, Takahashi T, Tatsumi T, Matsubara H. Erythropoietin-mobilized endothelial progenitors enhance reendothelialization via Akt-endothelial nitric oxide synthase activation and prevent neointimal hyperplasia. Circ Res 2006 June 9;98(11): (394) Griese DP, Ehsan A, Melo LG, Kong D, Zhang L, Mann MJ, Pratt RE, Mulligan RC, Dzau VJ. Isolation and transplantation of autologous circulating endothelial cells into denuded vessels and prosthetic grafts: implications for cell-based vascular therapy. Circulation 2003 November 25;108(21): (395) Akyurek LM, Fellstrom BC, Yan ZQ, Hansson GK, Funa K, Larsson E. Inducible and endothelial nitric oxide synthase expression during development of transplant arteriosclerosis in rat aortic grafts. Am J Pathol 1996 December;149(6): (396) Liu VW, Huang PL. Cardiovascular roles of nitric oxide: a review of insights from nitric oxide synthase gene disrupted mice. Cardiovasc Res 2008 January;77(1): (397) Chaudhuri A, Kanjwal Y, Mohanty P, Rao S, Sung BH, Wilson MF, Dandona P. Insulininduced vasodilatation of internal carotid artery. Metabolism 1999 November;48(11):

195 APPENDIX A Local Insulin Application on the Carotid Artery Inhibits Neointimal Formation. Insulin and Neointimal Formation Danna M. Breen 1, Michelle P. Bendeck PhD 2,3 and Adria Giacca MD 1,2. Department of Physiology 1, Medicine 2, Laboratory Medicine and Pathobiology 3, University of Toronto, Toronto M5S 1A8, Canada. Address for correspondence: Dr. Adria Giacca, University of Toronto, Department of Physiology, Medical Science Building, 1 King s College Circle, Room 3336, Toronto, Ontario Canada M5S 1A8 Tel: (416) Fax: (416) adria.giacca@utoronto.ca Funding provided by a grant to A. Giacca from the Heart and Stroke Foundation of Canada. Danna Breen was supported by a Heart & Stroke Foundation of Ontario Graduate Scholarship, and an Ontario Graduate Scholarship. Word Count Body: 4239 Word Count Abstract: 220 Total # Figures/Tables: 4 183

196 ABSTRACT OBJECTIVE: Anti-mitogenic agents currently used to prevent restenosis in drug-eluting stents (DES) delay re-endothelialization. Delayed re-endothelialization is now considered as the main cause of late stent thrombosis with DES, which emphasizes the need for new treatments. We have shown that systemic insulin treatment decreases neointimal growth and accelerates reendothelialization after arterial injury in a rat model of restenosis. However, systemic insulin treatment cannot be given to non-diabetic individuals because of the risk of hypoglycemia. Therefore, we herein investigated whether local insulin treatment is also effective in reducing neointimal growth after arterial injury. METHODS and RESULTS: Rats were given local vehicle (V) or local insulin (I) delivered via pluronic gel applied around the carotid artery immediately following balloon injury. Vessels were collected at 4 and 28 days after injury. Plasma glucose and systemic insulin levels were not affected by local insulin treatment. Insulin decreased intimal area at 28 days (P<0.05) and also inhibited cell migration at 4 days (P<0.05). CONCLUSIONS: These results are the first demonstration that insulin applied locally on the carotid artery is effective to inhibit the repair response to vessel injury. Therefore, local insulin treatment, possibly via insulin- eluting stents, may have therapeutic potential against restenosis. Keywords: insulin, pluronic gel, migration, neointima, angioplasty 184

197 Percutaneous revascularization procedures, such as angioplasty and stenting, are increasingly used to treat atherosclerotic vascular disease, however these procedures are often followed by restenosis. With balloon- expandable bare- metal stent rates of restenosis are approximately 30%. Drug-eluting stents (DES) were designed to release pharmacological agents after insertion to inhibit vascular smooth muscle cell (VSMC) proliferation and thus restenosis. First-generation DES are coated with either rapamycin or paclitaxel, which have reduced the restenosis rate to less than 10% as compared to the bare-metal stents 1,2. However, stent thrombosis has emerged as an infrequent but severe complication 3,4. These agents primarily inhibit VSMC migration and proliferation, which represent crucial events in the development of in-stent restenosis 5. However, both rapamycin and paclitaxel have been shown to decrease migration and proliferation of mature endothelial cells 6,7, as well as proliferation, differentiation, and homing of endothelial progenitor cells 8,9, and delay re-endothelialization 5, Furthermore, rapamycin and paclitaxel induce the expression of tissue factor, a key factor in the initiation of coagulation and thrombus formation 13,14. These effects may be responsible for the increased risk of late stent thrombosis with DES, which has emphasized the need for new treatments. We have found that subcutaneous insulin treatment decreases neointimal formation 15 and VSMC migration, while accelerating re-endothelialization in a rat model of restenosis 16. Thus insulin if effective locally (systemic insulin treatment cannot be given to non-diabetic people due to the risk of hypoglycemia) could provide some advantage over the anti-mitogenic agents currently used in stents because of this effect on re-endothelialization. As endothelial progenitor cells (EPC) are involved in re-endothelialization after angioplasty 17,18, EPC trapping stents were developed, which affect VSMC via re-endothelialization 19,20. Over these stents insulin-eluting stents could have the advantage of an additional direct effect to inhibit VSMC migration as 185

198 suggested by our studies with subcutaneous insulin 16, thus potentially being more effective on restenosis. Furthermore, insulin, at physiological concentrations, has been shown to have inhibitory effects on platelet aggregation, which initiates thrombosis, in vivo Therefore, it seems that insulin has potential for experimentation in a DES, which ideally would inhibit restenosis while at the same time promoting re-endothelialization. However, before the development of insulin-eluting stents, the effectiveness of local insulin treatment needs to be addressed. Therefore, the objective of this study is to determine whether local treatment of the injured vessel with insulin is effective in decreasing neointimal growth. METHODS Animal Models. Two groups of male Sprague-Dawley rats (Charles River) fed rat chow (Purina 5001) weighing g were studied: vehicle control and insulin (human; Humulin, Lilly)- treated. Insulin was administered via pluronic gel. Under light isoflurane anesthesia, the left common carotid artery was injured with a 2F balloon catheter (Baxter) which was inflated and withdrawn 4 times, followed by application of 0.5ml pluronic gel solution around the vessel. The solution contained insulin (10nM)/vehicle in 30% gel F-127 in saline 25,26. This concentration corresponds to approximately 10 times the plasma concentration of insulin in Chan et al 16. Analgesic (Buprenorphine) was given at the end of each surgical procedure. Blood glucose was monitored throughout the treatment using a glucometer (Glucometer Elite, Bayer Inc.). On the first day of treatment fed blood glucose was measured approximately once per hour and after that once per day at the same time of the day (~10a.m.). Fasting blood glucose was measured once per week. The Animal Care Committee of the University of Toronto approved all procedures. Blood and Vessel Sample Collection. The rats were sacrificed at various time points after injury, based on previous studies investigating the kinetics of vascular injury. Blood samples 186

199 were collected by cardiac puncture under general anesthesia. Thereafter, rats were sacrificed via anesthetic overdose and the carotids were perfusion-fixed for 4 minutes with 4% buffered paraformaldehyde at physiologic pressure (110 mmhg). The carotids were removed, immersed in 4% paraformaldehyde for >48h, embedded in paraffin, and divided into two blocks by cutting the midsection of the artery. Cross-sections were taken from the midsection. Metabolic Parameters. Plasma insulin was determined by radioimmunoassay (Linco) using kits specific for rat insulin and with 100% cross-reactivity with human insulin (Humulin R; Lilly) (used for treatment). Plasma FFA was measured using a colorimetric kit from Wako. Plasma triglycerides were also measured using colorimetric kits from Boehringer. Histomorphometry/Matrix Staining. To measure cell migration into the intima, carotids were collected 4 days after injury and intimal cells on the lumen surface were immunostained with an antibody against histone H1 (MAB1276; Chemicon) which does not permeate the internal elastic lamina 27. The entire luminal surface of the vessel was counted with a minimum of 12 fields taken at a magnification of 20X. Each nucleus that was oriented parallel to the long axis of the vessel was counted, whereas medial cells were perpendicular. A blinded coding system was used to avoid experimenter bias. This migration assay takes advantage of the fact that the first VSMCs appear in the intima 3 to 4 days after injury 27 and do not complete a round of replication until 24h later. Cross-sectional areas were measured on hematoxylin and eosin stained slides at 28 days after injury. Images were analyzed using a computer-assisted morphometric system (Simple PCI, Compix Inc). Intimal area was measured as the area encompassed by the internal elastic lamina minus the lumen area (in this case lumen area was determined by tracing around the inside edge of the vessel and calculating the area inside). Medial area was measured as the area 187

200 encompassed by the external elastic lamina minus the area encompassed by the internal elastic lamina (including lumen area). Lumen area was calculated by subtracting the intimal area and the medial area from the total area encompassed by the external elastic lamina (in this case total area was calculated using the external elastic lamina perimeter), assuming circular geometry of the vessel to avoid fixation artifacts. Statistics. Values are mean ± SEM. Means for vehicle versus insulin treatment at each timepoint were compared by the two-tailed Student s t-test. Calculations were performed using SAS (Statistical Analysis System, Cary, NC). RESULTS Fluid and food intake, change in body weight, and both fasting and fed plasma glucose levels taken over the treatment period were the same in local insulin-treated rats and vehicle control rats (Table 1). Plasma hormone and metabolite levels were taken at sacrifice and there were no changes in fed plasma insulin, triglyceride, and free fatty acid levels (Table 2). Treatment with local insulin immediately following balloon injury resulted in decreased intimal area after 28 days (P<0.05) (Figures 1A and 1B). By contrast, there was no change in medial area between the groups (V=0.16±0.01mm 2, I=0.14±0.01mm 2 ). Furthermore, lumen area was greater in the insulin-treated group at 28 days (P<0.05) (Figure 1C). There were no differences in external elastic lamina perimeter between the groups at 28 days (V=2.9±0.2mm, I=2.7±0.1mm), indicating that outward remodeling did not likely contribute to the increase in lumen area with local insulin treatment. To investigate whether local insulin, as previously seen with systemic insulin, inhibited cell migration, cell migration from the media to the intima was measured 4 days after vessel injury. Local insulin markedly inhibited cell migration by 85%, reducing the number of intimal 188

201 VSMCs from 495±98cells/mm 2 in vehicle control rats to 74±24cells/mm 2 (P<0.05) (Figure 2A and 2B). DISCUSSION The present study focused on the in vivo effect of local insulin application on the injured vessel in the rat carotid injury model of restenosis. Insulin had a marked inhibitory effect on injury-induced cell migration, leading to decreased neointimal formation. We observed no changes in plasma insulin or glucose levels. Therefore, our data provide evidence for a protective effect of insulin on vascular growth, in the absence of any systemic effect. The pathogenesis of restenosis after revascularization procedures is characterized by many processes that also occur in atherogenesis, and both diabetes and metabolic syndrome, which are characterized by insulin resistance, are associated not only with atherosclerosis but also with restenosis 28,29. However, apart from the risk provided by minimal or overt hyperglycemia, it is unclear whether insulin resistance or hyperinsulinemia provides additional risk. A few studies have evaluated the effect of systemic insulin on the arterial response to injury. Some studies suggest that insulin treatment aggravates intimal thickening in models of type 1 diabetes In models of insulin resistance, such as the Zucker Fatty rat 33-35, the fructosefed rat 36, the pre-diabetic OLETF rat 37, and the diabetic ZDF rats 38, neointimal growth after injury in increased, however it is not clear whether this is due to insulin resistance or to hyperinsulinemia. Insulin treatment of non-diabetic rats did not affect DNA synthesis in injured aortas during normoglycemia or hyperglycemia but decreased DNA synthesis in the presence of hypoglycemia 39,40 ; however intimal area was not measured. There are currently few reports on the effect of insulin on intimal thickness after balloon injury in non-diabetic models and their results are opposite. Chronic insulin treatment via islet transplantation 31 or subcutaneous insulin 189

202 infusion resulting in glucose- lowering increased neointimal growth 41. In contrast, neointimal growth was greater in mice with a defect in insulin signaling (IRS-2 null mice) 42, however it is not apparent whether this was due to decreased insulin action on the vessel or to the abnormal metabolic milieu. We have previously found a similar effect with subcutaneous insulin treatment which decreased neointimal formation after balloon angioplasty in rats 15. The discrepancy between our studies 15,16 and other studies in hyperinsulinemic rodents 31,41 may be explained by some degree of insulin resistance induced by chronic insulin itself or by the oral sucrose given to the insulin-treated rats in one study 43. Therefore, a major finding of our current study is that locally applied insulin on the carotid artery itself at the time of injury decreases neointimal area under euglycemic conditions, without any indication of systemic insulinization or insulin resistance. The rat carotid balloon injury model is currently the best characterized model of restenosis. The time course of neointimal growth has been well established after balloon angioplasty of the carotid artery in the rat 44. Medial VSMC proliferation peaks 2 days after angioplasty, followed by a peak in VSMC migration from the media to the intima at 4 days. Vascular VSMCs replicate in the neointima and reach a maximum proliferation rate by 7 days 44. Neointimal growth continues through 14 days after vessel injury and peaks and plateaus at 28 days 44. We found that insulin markedly inhibited cell migration into the intima, similar to what we have observed in a previous study with subcutaneous insulin treatment 16. Several studies have examined the effect of insulin on VSMC migration in vitro, and the results are controversial, as insulin at supraphysiological concentrations stimulated VSMC migration 45 in some studies but not in others 46. At physiological concentrations, insulin by itself did not affect migration 47, but inhibited the migration of VSMC stimulated by PDGF 48 and wounding in the 190

203 presence of inducible nitric oxide synthase (inos) induction 47. The latter finding may have pathophysiological implications since inos is upregulated in VSMC after balloon angioplasty 49. The migration assay measures histone H1 positive nuclei, thus we cannot definitively distinguish migrated VSMC and infiltrating hematopoietic or progenitor cells. We focused on VSMC migration rather than proliferation because we have previously found no effect of subcutaneous insulin (3 U/day) treatment on VSMC proliferation in the same model of carotid balloon injury 16 and thus, reduced proliferation was unlikely to account for the effect of local insulin. At a higher dose (5 U/day), we previously found that insulin increases VSMC proliferation in accordance with in vitro studies 50,51, despite decreasing neointimal growth 15. Since the local insulin levels were likely higher in this study than in Chan et al 16, it is possible that proliferation is increased and this will have to be determined. Systemic insulin treatment increased re-endothelialization in an aortic balloon injury model 52. Carotid balloon injury is not the best model to assess re-endothelialization, however it will have to be used to assess the effect of local insulin in future studies. In the systemic insulin treatment model, our preliminary data suggest that the effect of insulin to increase reendothelialization, decrease VSMC migration, and neointimal growth are NO- dependent, as they are abolished by a nitric oxide synthase inhibitor 52. Interestingly, a hormone similar to insulin, insulin- like growth factor- 1 (IGF-1), the systemic administration of which modestly increased neointimal growth in one study 53, was reported to decrease neointimal growth when applied locally in pluronic gels, perhaps via decreasing VSMC migration 54. Recently, IGF- 1 was also found to activate enos 55. With local treatment in pluronic gel, sustained release is brief 56 and release is periadventitial. Despite this, the effect of local insulin was marked, which suggests that sustained 191

204 release at an endovascular site as in stents may be even more effective, alone or perhaps in combination with commonly used drugs. In particular, a combination with paclitaxel may be promising as insulin may interfere with the inhibitory effect of rapamycin on mtor and subsequent VSMC proliferation. Larger animal models, such as the iliac artery injury in rabbits and coronary artery injury in pigs, should be studied once a stent is developed because their arteries are more similar to the human coronary artery. In conclusion, we have shown for the first time that local insulin treatment applied directly on the carotid artery is effective to decreased neointimal thickness and reduce VSMC migration after arterial injury. These results provide evidence for inhibition of the growth response to vessel injury by insulin in vivo and suggest that local insulin treatment, possibly in the form of a drug- coated stent, may offer protection against restenosis. 192

205 FIGURE LEGENDS Figure 1. Representative photomicrographs of cross-sections taken at 28 days after carotid injury (x400) (A). Cross-sectional areas of the intima (B) and lumen (C) of carotid arteries measured at 28 (Local Vehicle: n=9; Local Insulin: n=10) days after injury. Values are means ± SEM. *P<0.05 vs. Local Vehicle. Figure 2. Representative photomicrographs showing intimal cell on the lumen surface at 4 days after carotid injury (x 400) (A). Quantitative analysis of cell migration (Local Vehicle: n=3; Local Insulin: n=5) (B). Values are means ± SEM. *P<0.05 vs. Local Vehicle. 193

206 References 1. Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban HE, Perin M, Colombo A, Schuler G, Barragan P, Guagliumi G, Molnar F, Falotico R. A randomized comparison of a sirolimuseluting stent with a standard stent for coronary revascularization. N Engl J Med 2002;346: Moses JW, Leon MB, Popma JJ, Fitzgerald PJ, Holmes DR, O'Shaughnessy C, Caputo RP, Kereiakes DJ, Williams DO, Teirstein PS, Jaeger JL, Kuntz RE. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med 2003;349: Lanzer P, Sternberg K, Schmitz KP, Kolodgie F, Nakazawa G, Virmani R. Drug-eluting coronary stent very late thrombosis revisited. Herz 2008;33: Stettler C, Allemann S, Egger M, Windecker S, Meier B, Diem P. Efficacy of drug eluting stents in patients with and without diabetes mellitus: indirect comparison of controlled trials. Heart 2006;92: Luscher TF, Steffel J, Eberli FR, Joner M, Nakazawa G, Tanner FC, Virmani R. Drugeluting stent and coronary thrombosis: biological mechanisms and clinical implications. Circulation 2007;115: Akselband Y, Harding MW, Nelson PA. Rapamycin inhibits spontaneous and fibroblast growth factor beta-stimulated proliferation of endothelial cells and fibroblasts. Transplant Proc 1991;23: Prasad CK, Resmi KR, Krishnan LK, Vaishnav R. Survival of endothelial cells in vitro on Paclitaxel-loaded coronary stents. J Biomater Appl 2005;19: Butzal M, Loges S, Schweizer M, Fischer U, Gehling UM, Hossfeld DK, Fiedler W. Rapamycin inhibits proliferation and differentiation of human endothelial progenitor cells in vitro. Exp Cell Res 2004;300: Inoue T, Sata M, Hikichi Y, Sohma R, Fukuda D, Uchida T, Shimizu M, Komoda H, Node K. Mobilization of CD34-positive bone marrow-derived cells after coronary stent implantation: impact on restenosis. Circulation 2007;115: Finn AV, Kolodgie FD, Harnek J, Guerrero LJ, Acampado E, Tefera K, Skorija K, Weber DK, Gold HK, Virmani R. Differential response of delayed healing and persistent inflammation at sites of overlapping sirolimus- or paclitaxel-eluting stents. Circulation 2005;112: Joner M, Finn AV, Farb A, Mont EK, Kolodgie FD, Ladich E, Kutys R, Skorija K, Gold HK, Virmani R. Pathology of drug-eluting stents in humans: delayed healing and late thrombotic risk. J Am Coll Cardiol 2006;48:

207 12. Kotani J, Awata M, Nanto S, Uematsu M, Oshima F, Minamiguchi H, Mintz GS, Nagata S. Incomplete neointimal coverage of sirolimus-eluting stents: angioscopic findings. J Am Coll Cardiol 2006;47: Steffel J, Latini RA, Akhmedov A, Zimmermann D, Zimmerling P, Luscher TF, Tanner FC. Rapamycin, but not FK-506, increases endothelial tissue factor expression: implications for drug-eluting stent design. Circulation 2005;112: Stahli BE, Camici GG, Steffel J, Akhmedov A, Shojaati K, Graber M, Luscher TF, Tanner FC. Paclitaxel enhances thrombin-induced endothelial tissue factor expression via c-jun terminal NH2 kinase activation. Circ Res 2006;99: Kim T, Chan KK, Dhaliwall JK, Huynh N, Suen R, Uchino H, Naigamwalla D, Bendeck MP, Giacca A. Anti-atherogenic effect of insulin in vivo. J Vasc Res 2005;42: *Chan KK, *Breen D, Dhaliwall JK, Ward MR, Koudsi NA, Lam L, De Souza M, Ghanim H, Dandona P, Stewart D, Bendeck MP, Giacca A. Insulin Increases Re-endothelialization and Inhibits Cell Migration and Neointimal Growth After Arterial Injury.*Equally contributed to work. In Revision for Atherosclerosis, Thrombosis, and Vascular Biology Urao N, Okigaki M, Yamada H, Aadachi Y, Matsuno K, Matsui A, Matsunaga S, Tateishi K, Nomura T, Takahashi T, Tatsumi T, Matsubara H. Erythropoietin-mobilized endothelial progenitors enhance reendothelialization via Akt-endothelial nitric oxide synthase activation and prevent neointimal hyperplasia. Circ Res 2006;98: Griese DP, Ehsan A, Melo LG, Kong D, Zhang L, Mann MJ, Pratt RE, Mulligan RC, Dzau VJ. Isolation and transplantation of autologous circulating endothelial cells into denuded vessels and prosthetic grafts: implications for cell-based vascular therapy. Circulation 2003;108: Aoki J, Serruys PW, van BH, Ong AT, McFadden EP, Sianos G, van der Giessen WJ, Regar E, de Feyter PJ, Davis HR, Rowland S, Kutryk MJ. Endothelial progenitor cell capture by stents coated with antibody against CD34: the HEALING-FIM (Healthy Endothelial Accelerated Lining Inhibits Neointimal Growth-First In Man) Registry. J Am Coll Cardiol 2005;45: Blindt R, Vogt F, Astafieva I, Fach C, Hristov M, Krott N, Seitz B, Kapurniotu A, Kwok C, Dewor M, Bosserhoff AK, Bernhagen J, Hanrath P, Hoffmann R, Weber C. A novel drug-eluting stent coated with an integrin-binding cyclic Arg-Gly-Asp peptide inhibits neointimal hyperplasia by recruiting endothelial progenitor cells. J Am Coll Cardiol 2006;47: Hiramatsu K, Nozaki H, Arimori S. Reduction of platelet aggregation induced by euglycaemic insulin clamp. Diabetologia 1987;30:

208 22. Kahn NN, Bauman WA, Hatcher VB, Sinha AK. Inhibition of platelet aggregation and the stimulation of prostacyclin synthesis by insulin in humans. Am J Physiol 1993;265:H2160- H Trovati M, Anfossi G, Cavalot F, Massucco P, Mularoni E, Emanuelli G. Insulin directly reduces platelet sensitivity to aggregating agents. Studies in vitro and in vivo. Diabetes 1988;37: Westerbacka J, Yki-Jarvinen H, Turpeinen A, Rissanen A, Vehkavaara S, Syrjala M, Lassila R. Inhibition of platelet-collagen interaction: an in vivo action of insulin abolished by insulin resistance in obesity. Arterioscler Thromb Vasc Biol 2002;22: Parfyonova Y, Plekhanova O, Solomatina M, Naumov V, Bobik A, Berk B, Tkachuk V. Contrasting effects of urokinase and tissue-type plasminogen activators on neointima formation and vessel remodelling after arterial injury. J Vasc Res 2004;41: Huang SL, Ding B, Shan B, Yu QS, Guo ZG. Prevention of intima hyperplasia by mitogenactivated protein kinase antisense oligodeoxynucleotide. Acta Pharmacol Sin 2000;21: Bendeck MP, Irvin C, Reidy MA. Inhibition of matrix metalloproteinase activity inhibits smooth muscle cell migration but not neointimal thickening after arterial injury. Circ Res 1996;78: Osanai H, Kanayama H, Miyazaki Y, Fukushima A, Shinoda M, Ito T. Usefulness of enhanced insulin secretion during an oral glucose tolerance test as a predictor of restenosis after direct percutaneous transluminal coronary angioplasty during acute myocardial infarction in patients without diabetes mellitus. Am J Cardiol 1998;81: Takagi T, Yoshida K, Akasaka T, Kaji S, Kawamoto T, Honda Y, Yamamuro A, Hozumi T, Morioka S. Hyperinsulinemia during oral glucose tolerance test is associated with increased neointimal tissue proliferation after coronary stent implantation in nondiabetic patients: a serial intravascular ultrasound study. J Am Coll Cardiol 2000;36: Bornfeldt KE, Arnqvist HJ, Capron L. In vivo proliferation of rat vascular smooth muscle in relation to diabetes mellitus insulin-like growth factor I and insulin. Diabetologia 1992;35: Indolfi C, Torella D, Cavuto L, Davalli AM, Coppola C, Esposito G, Carriero MV, Rapacciuolo A, Di Lorenzo E, Stabile E, Perrino C, Chieffo A, Pardo F, Chiariello M. Effects of balloon injury on neointimal hyperplasia in streptozotocin-induced diabetes and in hyperinsulinemic nondiabetic pancreatic islet-transplanted rats. Circulation 2001;103: Schiller NK, McNamara DB. Balloon catheter vascular injury of the alloxan-induced diabetic rabbit: the role of insulin-like growth factor-1. Mol Cell Biochem 1999;202:

209 33. Park SH, Marso SP, Zhou Z, Foroudi F, Topol EJ, Lincoff AM. Neointimal hyperplasia after arterial injury is increased in a rat model of non-insulin-dependent diabetes mellitus. Circulation 2001;104: Haudenschild CC, Van SW, Chobanian AV. Response of the aorta of the obese Zucker rat to injury. Arteriosclerosis 1981;1: Ridray S, Heudes D, Michel O, Penicaud L, Ktorza A. Increased SMC proliferation after endothelial injury in hyperinsulinemic obese Zucker rats. Am J Physiol 1994;267:H1976- H Chajara A, Raoudi M, Delpech B, Leroy M, Basuyau JP, Levesque H. Increased hyaluronan and hyaluronidase production and hyaluronan degradation in injured aorta of insulin-resistant rats. Arterioscler Thromb Vasc Biol 2000;20: Tamura K, Kanzaki T, Tashiro J, Yokote K, Mori S, Ueda S, Saito Y, Morisaki N. Increased atherogenesis in Otsuka Long-Evans Tokushima fatty rats before the onset of diabetes mellitus: association with overexpression of PDGF beta-receptors in aortic smooth muscle cells. Atherosclerosis 2000;149: Jonas M, Edelman ER, Groothuis A, Baker AB, Seifert P, Rogers C. Vascular neointimal formation and signaling pathway activation in response to stent injury in insulin-resistant and diabetic animals. Circ Res 2005;97: Capron L, Jarnet J, Kazandjian S, Housset E. Growth-promoting effects of diabetes and insulin on arteries. An in vivo study of rat aorta. Diabetes 1986;35: Ridray S, Ktorza A, Picon L, Capron L. In vivo effect of insulin on the acute proliferative response of the rat aorta to injury. Arterioscler Thromb 1992;12: Pu Q, Chang Y, Zhang C, Cai Y, Hassid A. Chronic insulin treatment suppresses PTP1B function, induces increased PDGF signaling, and amplifies neointima formation in the balloon-injured rat artery. Am J Physiol Heart Circ Physiol 2009;296:H132-H Kubota T, Kubota N, Moroi M, Terauchi Y, Kobayashi T, Kamata K, Suzuki R, Tobe K, Namiki A, Aizawa S, Nagai R, Kadowaki T, Yamaguchi T. Lack of insulin receptor substrate-2 causes progressive neointima formation in response to vessel injury. Circulation 2003;107: Foster E, Zhang S, Kahn AM. Insulin stimulates arterial neointima formation in normal rats after balloon injury. Diabetes Obes Metab 2006;8: Clowes AW, Reidy MA, Clowes MM. Mechanisms of stenosis after arterial injury. Lab Invest 1983;49: Gockerman A, Prevette T, Jones JI, Clemmons DR. Insulin-like growth factor (IGF)- binding proteins inhibit the smooth muscle cell migration responses to IGF-I and IGF-II. Endocrinology 1995;136:

210 46. Grotendorst GR, Chang T, Seppa HE, Kleinman HK, Martin GR. Platelet-derived growth factor is a chemoattractant for vascular smooth muscle cells. J Cell Physiol 1982;113: Zhang S, Yang Y, Kone BC, Allen JC, Kahn AM. Insulin-stimulated cyclic guanosine monophosphate inhibits vascular smooth muscle cell migration by inhibiting Ca/calmodulin-dependent protein kinase II. Circulation 2003;107: Jacob A, Molkentin JD, Smolenski A, Lohmann SM, Begum N. Insulin inhibits PDGFdirected VSMC migration via NO/ cgmp increase of MKP-1 and its inactivation of MAPKs. Am J Physiol Cell Physiol 2002;283:C704-C Yan Z, Hansson GK. Overexpression of inducible nitric oxide synthase by neointimal smooth muscle cells. Circ Res 1998;82: Cruzado M, Risler N, Castro C, Ortiz A, Ruttler ME. Proliferative effect of insulin on cultured smooth muscle cells from rat mesenteric resistance vessels. Am J Hypertens 1998;11: Pfeifle B, Ditschuneit H. Effect of insulin on growth of cultured human arterial smooth muscle cells. Diabetologia 1981;20: Breen DM, Guo J, Mroziewicz M, Bendeck MP, Giacca A. Insulin accelerates reendothelialization and decreases neointimal formation after arterial injury through a nitric oxide synthase-dependent mechanism. Diabetes 57[Suppl.1], A Ref Type: Abstract 53. Li H, Dimayuga P, Yamashita M, Yano J, Fournier M, Lewis M, Cercek B. Arterial injury in mice with severe insulin-like growth factor-1 (IGF-1) deficiency. J Cardiovasc Pharmacol Ther 2002;7: Wong AH, Amabile PG, Yuksel E, Waugh JM, Dake MD. Perivascular release of insulinlike growth factor-1 limits neointima formation in the balloon-injured artery by redirecting smooth muscle cell migration. J Vasc Interv Radiol 2001;12: Sukhanov S, Higashi Y, Shai SY, Vaughn C, Mohler J, Li Y, Song YH, Titterington J, Delafontaine P. IGF-1 reduces inflammatory responses, suppresses oxidative stress, and decreases atherosclerosis progression in ApoE-deficient mice. Arterioscler Thromb Vasc Biol 2007;27: Simons M, Edelman ER, DeKeyser JL, Langer R, Rosenberg RD. Antisense c-myb oligonucleotides inhibit intimal arterial smooth muscle cell accumulation in vivo. Nature 1992;359:

211 4 days 28 days Treatment Groups V (n=3) I (n=5) V (n=9) I (n=10) Fluid Intake (g/day) Food Intake (g/day) Weight Change (g) Fasting Plasma Glucose (mm) Fed Plasma Glucose (mm) * Table 1. Daily fluid and food intake, final weight change, fasting and fed plasma glucose levels over the treatment period. V=Local Vehicle; I=Local Insulin. Data are expressed as means + SEM. *P<0.05 vs. Local Vehicle. Data are means + SEM of individual average glucose levels measured over the treatment period. On the first day of treatment blood glucose was measured approximately once per hour and after that both fasting and fed blood glucose were measured once per day at the same time of the day (~10a.m.). 199

212 4 days 28 days Treatment Groups V (n=3) I (n=5) V (n=9) I (n=10) Insulin (pm) Triglycerides (mm) Free Fatty Acids (ueq/l) Table 2. Fed plasma levels of insulin, triglycerides, and free fatty acids at 4 and 28 days after arterial injury. V=Local Vehicle; I=Local Insulin. Data are expressed as means + SEM and represent values taken at sacrifice. 200

213 A B Intimal Area (mm 2 ) Local Vehicle Local Insulin * 0.00 Local Vehicle (n=9) Local Insulin (n=10) C 0.20 * Lumen Area (mm 2 ) Figure Local Vehicle Local Insulin (n=9) (n=10) 201

214 A Local Vehicle Local Insulin B Number of cells/mm * Local Vehicle Local Insulin (n=3) (n=5) Figure 2 202