Kruppel-Like Transcription Factors: Master

KRUPPEL-LIKE TRANSCRIPTION FACTORS: MASTER

REGULATORS OF VASCULAR ENDOTHELIUM

PANJAMAPORN SANGWUNG

Submitted in partial fulfillment of the requirements

For the degree of Doctor of Philosophy

Dissertation Advisor: Mukesh K. Jain, MD.

Department of Physiology and Biophysics

CASE WESTERN RESERVE UNIVERSITY

August, 2017 CASE WESTERN RESERVE UNIVERSITY

SCHOOL OF GRADUATE STUDIES

We hereby approve the thesis/dissertation of

Panjamaporn Sangwung

candidate for the degree of Doctor of Philosophy*.

Committee Chair

George R. Dubyak, Ph.D.

Committee Member

Thomas N. Nosek, Ph.D.

Committee Member

Julian E. Stelzer, Ph.D.

Committee Member

Jonathan D. Smith, Ph.D.

Dissertation Advisor

Mukesh K. Jain, M.D.

Date of Defense

July 6th, 2017

*We also certify that written approval has been obtained for any proprietary material contained therein. DEDICATION

This dissertation is dedicated to my family, especially my parents (Boonma and

Usanee Saengwang) and my grandparents (Anan and Aonkaew Intana) for their constant source of support and love.

TABLE OF CONTENTS

LIST OF TABLES………………………………………………………….…………..v

LIST OF FIGURES.………………..…………………………..………………………vi

ACKNOWLEDGEMENTS………………………… … ……..………….……….….viii

LIST OF ABBREVIATIONS………………… …………..…….…………………….xii

ABSTRACT……………………………………………….……...……….……..……. 1

CHAPTER 1. SIGNIFICANCE AND INTRODUCTION……………..……………. 3

1.1 Significance ……………………………………………………………….……. 3

1.2 Structure and functional importance of vascular endothelium …..…… 5

Basic structure of blood vessels……………………………………...……... 5

Functional Importance of vascular endothelium…………………………… 7

1.3 Dysfunction of endothelial cells and diseases ………………………..… 14

1.4 Regulation of endothelial function ……………………………………….... 16

Biochemical stimuli……………………………………..………………….… 16

Biomechanical stimuli……………………………………………..…….…… 17

1.5 Kruppel-like factor (KLF) ………………………………………….……..….. 18

Structure of KLFs……………………………………………………………….. 18

Identification and characterization of endothelial KLFs………………....….. 19

KLF2 – Identification and characterization………………..……………..... 19

Role of endothelial KLF2…………………………………….………….…... 21

Regulation of endothelial KLF2………………………………………….…. 25

KLF4 – Identification and Characterization……………………………….. 30

Role of endothelial KLF4………………………….…………………….…... 31

Regulation of endothelial KLF4…………………………………….….…… 32

KLF6 – Identification, characterization, and functions………………...... 33

1.6 Endothelial hemoglobin …………………………………..……………….... 35

Functional importance of hemoglobin …………………………………..….... 35

Hemoglobin and disorders…………………………………………………..… 36

Structure of hemoglobin ……………………………………………….……..... 37

Regulation of hemoglobin………………………………………………….…... 37

Hemoglobin in non-erythroid cells…………………………………………..… 38

CHAPTER 2. KLF2 AND KLF4 CONTROL ENDOTHELIAL IDENTITY AND

VASCULAR INTEGRITY…..…………………………..………………………...... 41

2.1 Abstract………..…………………………………………………….……….….. 41

2.2 Introduction………..…………………………………………………..……...… 42

2.3 Materials and methods…………………………………………….………...... 44

2.4 Results…….……………………………………………………...…..………..... 52

Endothelial-specific Klf2 and Klf4 deletion leads to rapid death of adult

Mice..………………………………………………………………………...... 52

Endothelial-specific deletion of Klf2 and Klf4 leads to vascular leak

and systemic coagulopathy……………………………………………...... 59

Endothelial-specific Klf2 and Klf4 deletion results in profound

alterations in the EC transcriptome…………………………….…...……… 67

2.5 Discussion and conclusion...………………………………………...……… 69

2.6 Author contributions……...…………………………………………...……… 73

2.7 Acknowledgments…………...………………………………………...………. 73

CHAPTER 3. REGULATION OF ENDOTHELIAL HEMOGLOBIN ALPHA

EXPRESSION BY KRUPPEL-LIKE FACTORS……………..…………………... 74

3.1 Abstract………..………………………………………………...………...……. 74

3.2 Introduction………..………………………………………………...……..…… 75

3.3 Materials and methods………………………………………………..….…… 77

3.4 Results…….…………………………………………………………………..…. 85

KLF2 and KLF4 induce HBA expression in the endothelium…………………… 85

The transactivation of the human hemoglobin alpha gene promoter

in ECs by KLF2/KLF4………………………………………………………... 93

Direct binding of KLF4 to the endogenous hemoglobin alpha

promoter in the EC…………………………………………………...... 98

3.5 Discussion…….…………………………………………………..…………… 100

3.6 Conclusion…….………………………...…………………………..………… 102

3.7 Acknowledgment……………………...…………..………………..………… 102

CHAPTER 4. DISCUSSION AND SUMMARY STATEMENTS OF

THE THESIS PROJECTS …………………………………………………………. 103

4.1 Discussion …….………………………...…………………………..……….... 103

4.2 Summary statements of the thesis projects…..…..…………....…..….... 106

iii

CHAPTER 5. FUTURE DIRECTIONS…….………………...... …...... 107

BIBLIOGRAPHY.…………………....…………………………...……………….... 109

LIST OF TABLES

Table 2.1 Primer sequences for qPCR (TaqMan) analysis………………….….. 46

Table 2.2 Differentially expressed genes (q<0.05) in the Hallmark

coagulation pathway (EC-DKO vs CRE) at day 6 post-tamoxifen……… 72

Table 3.1 Primer sequences for qPCR (TaqMan) analysis……………….…….. 79

Table 3.2 The HBA promoter mutagenic oligonucleotide primers (5’-3’)………. 83

Table 3.3 Identified peptides of HBA protein in the MEJ of KLF4

overexpressing HCAECs using LC-MS/MS………………..…….……….. 92

LIST OF FIGURES

Figure 1.1 Basic structure of normal artery…………………………………………. 5

Figure 1.2 Schematic representation of both tight and adherens

junctions in ECs…………………………………………………...…...... 9

Figure 1.3 The TM/protein C/protein S anticoagulation system….……………… 11

Figure 1.4 NO production by eNOS………………………………………….…….. 13

Figure 1.5 A schematic representation of common structure and

functional domains for Krüppel-like factors (KLFs)………………………... 18

Figure 1.6 Human hemoglobin genes located on chromosome 11and 16 ….… 38

Figure 1.7 Possible mechanism of NO control at the MEJ………………….…… 40

Figure 2.1 Expression of Klf2, Klf4, and their targets in primary cardiac

microvascular EC……………………………….………………………….…. 54

Figure 2.2 Endothelial-specific Klf2 and Klf4 deletion leads to rapid

death of adult mice……………….……………………….….……………….. 55

Figure 2.3 Electrocardiogram (EKG) recording………….……………….……….. 56

Figure 2.4 Echocardiographic analysis at day 6 after tamoxifen

administration……………………….……………..…………………….……. 57

Figure 2.5 Gross post-mortem examination in EC-Klf2-KO and

EC-Klf4-KO mice…………………………………..……………..…………… 58

Figure 2.6 Endothelial-specific deletion of Klf2 and Klf4 leads to

vascular leak and systemic coagulopathy………………..….…...………… 62

Figure 2.7 Endothelial-specific deletion of Klf2 and Klf4 results in

vascular leak…………………..……………………..…………..…….……… 64

Figure 2.8 Representative electron microscopic (EM) images of brain

indicate degeneration of EC and extravascular erythrocytes in

the EC-DKO mice at day 6 post-tamoxifen……………………..………….. 65

Figure 2.9 Expression of F3, Serpine1, and F2rl3 mRNA in primary

cardiac microvascular EC at day 6 post-tamoxifen……………...………… 66

Figure 2.10 Endothelial-specific Klf2 and Klf4 deletion results in profound

alterations in the EC transcriptome……………………………………….… 68

Figure 3.1 Hemoglobin alpha mRNA expression in the EC by KLF2/KLF4…… 87

Figure 3.2 Expression of KLF2 or KLF4 mRNA in ECs………………………….. 88

Figure 3.3 KLF2/KLF4-mediated hemoglobin alpha protein expression

in the EC cultured alone on a petri dish…………………………...... 89

Figure 3.4 Hemoglobin alpha protein expression in the EC by KLF2/KLF4…... 90

Figure 3.5 KLF2/KLF4-mediated hemoglobin alpha protein expression

in the EC and the MEJ obtained from the co-cultured model…………….. 91

Figure 3.6 KLF2/KLF4 mediate a transactivation of the HBA promoter………... 95

Figure 3.7 KLF2/KLF4 fail to mediate a transactivation of the mutant

HBA promoter………………………………………………….……………… 96

Figure 3.8 Critical KLF-binding sites on the HBA promoter……………………... 97

Figure 3.9 Direct binding of KLF4 on the HBA promoter in the EC…………...... 99

vii

ACKNOWLEDGEMENTS

I would like to express my wholehearted thanks to all those who helped me to make my PhD study and dissertation work a success. This dissertation would not have been possible without help of many people.

First and foremost, I would like to express my grateful thanks to my mentor, Dr. Mukesh Jain, for giving me the opportunity to join the lab and for giving me mental and technical advice on my professional and personal life.

Dr.Jain is one of the best mentors I have ever had in my life. Dr.Jain provided leadership development, useful guidance, tutelage, consultation, and feedback to me. He also provided me with opportunity to explore research, full material supports, and expert. He connected me to resources to ensure progress of the projects. He motivated, supported, and challenged me to push myself beyond the limits to achieve the goals and success in my profession. Dr.Jain is an outstanding physician scientist, a great speaker, and creative writer. I have learned a tremendous amount from him.

I am pleased to acknowledge and thank my dissertation committees (Drs.

George Dubyak, Jonathan Smith, Thomas Nosek, and Julian Stelzer). I met my dissertation committees at least once a year. My committees helped me directing my research and troubleshooting problems. Dr. Dubyak is one of the best teachers I have ever met. He has tremendous amount of knowledge in vascular biology and cell signaling. He was always available for me when needed. Dr.

Dubyak provided guidance to access to internal and external resources, and

viii

served as troubleshooters. He helped me plan a graduation timeline and made sure that things moved along. Dr. Smith encouraged me to think critically by challenging me with logical questions, and to become better problem solvers. He has massive amount of knowledge in molecular and cell biology and other related areas. I learned by talking to him and listening to his thoughtful questions.

Dr. Nosek served as a resource on graduate school policies and procedures. He moved things along the schedule and provided guidance on research projects and balance in life. Dr. Stelzer encouraged me to gain useful and valuable experience while in graduate school. He asked questions that helped steer my reflections in a productive way.

I am deeply thankful to a number of faculty members and the administrative staffs at the Department of Physiology and Biophysics (DPB). I would like to thank Dr. Walter Boron and the DPB admissions team for accepting me to the program. Although I missed the first PhD interview appointment, Drs.

Corey Smith, William Schilling, Andrea Romani, and Thomas Nosek gave me a second chance for the interview. They also provided guidance on general issues in a grad school. Other members at the DPB who provided me with great supports and suggestions include Dr. Tingwei Mu, Dr. Shyue-An Chan, Dr.

Sudha Chakrapani, Morley Schwebel, and Micheal Little. Jean Davis also helped me from the very beginning of my PhD application and during a PhD study.

I also thank a number of faculty members, members, and the administrative staffs of Case Cardiovascular Research Institute and Institute for

Transformative Molecular Medicine for supporting and helping me. These include

Drs. Jonathan Stamler, Shi Can, Yunmei Wang, Ganapati Mahabeleshwar,

Diana Ramirez-Bergeron, Aaron Proweller, Douglas Hess, Alfred Hausladen,

Fabio Fonseca, and Divya Seth. I also thank Susan Dowhan, Kimberly Loftus, and Alisa Alfaro for helping me with scheduling meetings and Kristen Camputaro for helping me move things forward.

My grateful thanks also go to colleagues in Dr. Jian laboratory for helping me with all of my research projects and for keeping the atmosphere in the lab conducive for good research. I value and appreciate Dr. Xudong Liao and Dr.

Yuan Lu for helping and being with me throughout any difficulties and uncertainty in science and life in general, and for keeping my motivation strong. They are very knowledgeable in several areas of researches: they provided a number of technical guidance and helped me with troubleshooting. I also thank Dr. Lalitha

Nayak for teaching me some lab skills, for motivating me, and for supplying me with snacks. I thank Dr. Guangjin Zhou for teaching me molecular lab skills.

Other former and current members of the Jain lab who provided guidance on my research and personal life include Drs. Lilei Zhang, Neelakantan Vasudevan,

Rongli Zhang, and Domenick Prosdocimo. I also thank Stephanie Lapping for helping me with lab supplies.

I also would like to thank my former advisors Dr. Samaisukh Sophasan and Dr. Yasuko Iwakiri for providing consultation and suggestions to help me progress in my research and career development.

Finally, I would like to express my heartfelt gratitude to my family and friends for a tremendous and constant amount of supports, motivation, and

encouragement. These friends include Pennapa Ranprai, Suttira Intapad, Petra

Hirsova, Jittima Weerachayaphorn, Paratakorn Veethong, Ruechakorn

Chantawee, Noppawan Charususin, and Pawonmon Klaphajon Kolly. I also thank Danniel Harris, Caroline Sanadi, Robert Stanek, and Prattana Samasilp for their friendship and advice.

LIST OF ABBREVIATIONS

AGEs Advanced glycosylation end products

AP-1 Activator protein-1

APC Activated protein C

ChIP Chromatin immunoprecipitation

CRE Cdh5(PAC)-Ert2cre

EC Endothelial cell

EC-DKO Endothelial-specific Klf2/Klf4 double knockout

EC-Klf2-KO Endothelial-specific Klf2 knockout

EC-Klf4-KO Endothelial-specific Klf4 knockout

ECG Electrocardiogram

ECHO Echocardiography

EF Ejection fraction

EPCR Endothelial protein C receptor

ERK5 Extracellular signal regulated kinase 5 eNOS Endothelial nitric oxide synthase

GAPDH Glyceraldehyde 3-phosphate dehydrogenase

GC Guanylate cyclase

HBA Hemoglobin subunit alpha

ICAM-1 Intercellular cell adhesion molecule-1

IL1β Interleukin 1 beta

JAM-1 Junctional adhesion molecule-1

KLF Krupple like factor

xii

NO Nitric oxide mRNA Messenger RNA

MEF2 Myocyte enhancing factor 2

MEJ Myoendothelial junction

MEK Mitogen-activated protein kinase kinase

MMP Matrix metalloproteinase

NF-κB Nuclear factor kappa-light-chain-enhancer of activated B cells

PAI-1 Plasminogen activator inhibitor-1

PECAM-1 Platelet and endothelial cell adhesion molecule-1

PI3K Phosphatidylinositol 3-kinase

ROS Reactive oxygen species

SMC Smooth muscle cell

TF Tissue factor

TFPI Tissues factor pathway inhibitor

TGF-β Transforming growth factor-beta

TM Thrombomodulin

TNF-α Tumor necrosis factor alpha

tPA Tissue plasminogen activator

uPA Urokinase plasminogen activator

VCAM-1 Vascular cell adhesion molecule-1

VEGF Vascular endothelial growth factor

vWF von Willebrand factor

xiii

Kruppel-Like Transcription Factors: Master Regulators of Vascular

Endothelium

Abstract

PANJAMAPORN SANGWUNG

Endothelial cells (ECs) are central determinants of vascular health. Dysfunction

of ECs, characterized in part by reduced nitric oxide (NO) signaling, contributes

to various cardiovascular diseases worldwide. Multiple lines of evidence support an essential role for KLF2 and KLF4 in endothelial cell biology. However, the

precise in vivo role of both factors in orchestrating EC health and vascular

hemostasis remains undetermined. The central hypotheses of this work are (i)

KLF2 and KLF4 serve as vital transcriptional regulators that maintain endothelial

health and vascular hemostasis in adult animal, and (ii) these factors govern

endothelial hemoglobin subunit alpha (HBA) expression, which is critical to fine-

tune NO diffusion from EC to vascular smooth muscle cells (SMC). In vivo

studies using inducible-genetic deficiency of both endothelial KLF2 and KLF4

(double knockout; EC-DKO) in adult mice caused acute deterioration of vascular

permeability and disruption of the coagulation system, leading to heart failure,

stroke, and death within days upon gene deletion. Mechanistically, abrogation of

both KLF2 and KLF4 profoundly altered the EC transcriptome, suggesting that

these factors regulate a broad range of endothelial genes and thus determine EC

homeostasis. Further, KLF2 and KLF4 transcriptionally regulated HBA gene expression in ECs. Collectively, this study identifies KLF2 and KLF4 as essential determinants of endothelial health and vascular homeostasis.

CHAPTER 1

SIGNIFICANCE AND INTRODUCTION

1.1 SIGNIFICANCE

Cardiovascular disorders are a major cause of mortality and morbidity worldwide, and endothelial dysfunction is a central event in the pathophysiology of these diseases (Feletou & Vanhoutte, 2006; S. S. Lim et al., 2012). Hence, elucidating nodal regulators and the underlying mechanisms that are critical for healthy endothelial hemostasis is of great importance in the search of therapeutic targets.

Furthermore, identification and cataloging the expression profiles of the direct targets of KLF2 and/or KLF4 will provide the solid platform for further investigating novel players or gene regulatory programs mediated by KLF2/KLF4 that are important for maintaining vascular endothelial functions. The observation will advance our understanding of the relative role of each factor in regulating endothelial gene expression and will provide a basic insight for understanding the transcriptional pathways governing endothelial health.

Further, defining a link between KLF2 and/or KLF4 to their target gene products such as HBA is an exciting area of investigation. Evaluating KLF2/KLF4 functions in regulating the expression of HBA in ECs will provide groundbreaking information for understanding endothelial function and its communication to SMC, and vascular hemostasis. Collectively, the results of these studies may provide the foundation for developing novel therapies aimed at ameliorating cardiovascular disease. KLF2 and KLF4 confer vasculoprotective properties to

the vascular endothelium, positioning these transcription factors as an ideal

target for the therapeutic treatment of cardiovascular disease.

OUTLINE

This research is dedicated to the identification of critical transcriptional regulators

of endothelial function and vascular homeostasis, and the underlying

mechanisms. To begin with, I introduce the important physiological functions of

the vascular endothelium and its regulation. This is followed by the identification, characterization, functions, and regulation of endothelial KLF2, KLF4, and KLF6.

Next, the functional relevance of hemoglobin, disorders, and regulation of

hemoglobin are described. Chapter 2 depicts our in vitro and in vivo findings of

the absolute requirement of KLF2 and KLF4 for endothelial function and vascular

integrity, which was published in the Journal of Clinical Investigation Insights in

2017 (Sangwung et al., 2017). Chapter 3 describes our finding of the

transcriptional regulators mediating the expression of HBA in the endothelium,

which was accepted for publication in Vascular Medicine in 2017 (Sangwung et

al., In press). Chapter 4 includes discussion and summary statements of thesis project. Finally, I wrap up my dissertation with future directions in Chapter 5.

INTRODUCTION

1.2 STRUCTURE AND FUNCTIONAL IMPORTANCE OF VASCULAR

ENDOTHELIUM

Basic structure of blood vessels

An extensive network of blood vessels maintains functions of the organs throughout the body by supplying oxygen and nutrients to tissues and individual cells, and removing cellular and metabolic waste products. The basic structure of blood vessels is composed of a continuous monolayer of the vascular endothelial cells (ECs) that resides at the inner layer, or the intima, of the entire vascular system at the interface between blood and all tissues. The middle layer termed the media consists of muscle cells-termed vascular smooth muscle cells. The outer layer, or adventitia, contains a mixture of extracellular matrix (ECM), connective tissue, fibroblasts, and nerve cells (Figure 1.1) (Aird, 2007b; Libby,

Ridker, & Hansson, 2011).

Figure 1.1 Basic structure of normal artery. The normal artery contains three layers. The inner layer, the tunica intima, is lined by a single layer of endothelial cells. The middle layer, or tunica media, contains smooth muscle cells embedded in a complex extracellular matrix. The adventitia, the outer layer of arteries, contains mast cells, nerve endings and microvessels.

The vascular system is comprised of arteries, arterioles, capillaries, venules, and veins that serve as conduits. Oxygenated blood from the heart moves along arteries and arterioles into capillaries where exchange of gases, nutrients, and waste products between blood and tissues takes place.

Deoxygenated blood flows in venules and veins, and returns to the cardiopulmonary system to be reoxygenated. The largest arteries like the aorta are thick, as several layers of smooth muscle cells intertwined with an elaborate elastin and collagen-based matrix. Veins are thin and have valves: they function as a low pressure reservoir and facilitate the flow of deoxygenated blood from the organs to the heart. Venous vessels have less smooth muscle cells and less complex extracellular matrix component. The smallest vessels have less smooth muscle cell coverage. Capillaries, the major exchange vessels in the circulation, are extraordinarily thin; the diameter is less than 10 um. They contain flat ECs surrounded to a variable extent by occasional pericytes and extracellular matrix

(Aird, 2007b; Lilly, 2014).

ECs in the arteries have long and narrow shape and align in direction of undisturbed blood flow. In the veins, the shape of ECs is shorter and wider and the cells do not align in the direction of blood flow (Aird, 2007b). A portion of ECs extends and protrudes through the internal elastic lamina toward vascular SMCs to form a junction. This junction is termed myoendothelial junction (MEJ). The formation of the MEJ is normally seen in small (e.g. small arteries, veins or arterioles) as opposed to large diameter vessels (Straub, Zeigler, & Isakson,

2014). This specialized junction acts as a conduit for communication between

ECs and SMCs. An example is the passage of NO from ECs to SMCs, a process

critical for the regulation of vascular tone and function (Feletou, 2011; Straub et

al., 2012).

Functional importance of vascular endothelium

Vascular endothelium was initially considered as an inert cellophane-like

membrane that lines the circulatory system with its function to maintain vascular

permeability (Cines et al., 1998). This concept was, however, changed due to

enormous advances in studying the complex functions of EC since the 1980’s.

Vascular endothelium is considered to be an organ in the body that lines the

entire circulatory system from head to toe (W. C. Aird, 2004). Endothelium

functions are achieved through the presence of membrane-bound receptors for

numerous molecules, and through the specific junctional proteins and receptors

that govern cell-cell and cell-matrix interactions. Molecules that act on the

receptors include proteins (e.g. growth factors, coagulant, and anticoagulant

proteins), lipid transporting particle (e.g. low-density lipoprotein), metabolites (e.g.

nitrous oxide and serotonin), and hormones (e.g. endothelin-1) (Cines et al.,

1998). The vascular endothelium actively participates in a variety of physiological processes, most of which are performed by specific subsets of blood vessel types or vascular bed (Aird, 2007a). These include permeability, leukocyte transmigration, hemostasis, vasomotor tone, humidification, thermoregulation, sieve function, scavenging, immune tolerance, and proliferation/angiogenesis.

Four representative examples are discussed below.

Vascular permeability

ECs tightly control transport of fluids and solutes into and out of the blood. A passive movement of fluids and small solutes across the endothelial barrier is processed via the paracellular route. Macromolecules move across the barriers via the transcellular route by membrane-bound vesicular carries (e.g. caveolae) or by vesicle-derived transendothelial channels (Aird, 2007a). Basal permeability of ECs is determined by differences in junctional properties, EC fenestration, and/or differential activity of the transcytotic machinery. Increased permeability of fluids and small solutes, but not macromolecules, is proportional to the presence of EC fenestrae.

The coordinated opening and closure of cell–cell junctions control, to a large extent, permeability. ECs express both tight and adherens junctions (Figure

1.2), transmembrane adhesive proteins that promote homophilic interactions and form a pericellular zipper-like structure along the cell border (Dejana, 2004). Tight junctions mediate ion and solute passage through paracellular routes, and may serve as a membrane fence, limiting the free movement of lipids and proteins.

Adherens junctions initiate cell-to-cell contacts and promote maturation and maintenance of cell-to-cell contacts (Dejana, Tournier-Lasserve, & Weinstein,

2009). The importance of the endothelial tight junction (e.g. claudin-5) and adherens junction (e.g. VE-cadherin) in mediating permeability has been evidenced in mice. Claudin-5 null mice exhibit impaired blood brain barrier function. Depletion of VE-cadherin in mice alters permeability in the lung and heart (Aird, 2007a). In addition, both tight and adherens junctions regulate

8 endothelial cell functions. They bind, through their cytoplasmic tails, to cytoskeletal and signaling proteins that maintain anchoring of junctions to actin microfilaments and transfer intracellular signals to the inside of the cells.

Therefore, the organization of the intercellular junctions is an important process

for the control of vascular permeability.

Figure 1.2 Schematic representation of both tight and adherens junctions in ECs. The cytoplasmic components of tight junctions include ZO proteins, cingulin, ZONAB, and others. Cell-cell adhesion at adherens junctions is promoted by cadherins (VE-cadherin and N-cadherin) which directly bind to p120, β-catenin, and plakoglobin. Adherens junctions interact with a large set of actin binding proteins. In addition, phosphatases and kinases are directly or indirectly associated to adherens junction components. Growth factor receptors including VEGF receptor 2 and TGFβ receptor complex can also bind to VE-cadherin complex.

Hemostasis and anticoagulation

The endothelium controls a wide range of homeostatic functions. ECs mediate the expression of binding sites for anticoagulant and procoagulant factors on the cell membrane (Rajendran et al., 2013). ECs also produce and express anticoagulant and procoagulant molecules that are unevenly distributed throughout the vascular system (Aird, 2007a). The anticoagulant factors include tissue factor pathway inhibitor (TFPI), heparin, thrombomodulin, endothelial protein C receptor (EPCR), tissue-type plasminogen activator (t-PA), ecto-

ADPase, prostacyclin, and nitric oxide (NO). The procoagulant molecules synthesized by ECs include tissue factor, plasminogen activator inhibitor (PAI-I), von Willebrand factor (vWF), and protease activated receptors.

A common function of healthy ECs is to maintain blood fluidity and inhibit coagulation of blood by enhancing the activity of multiple anticoagulant pathways

(Rajendran et al., 2013). At the endothelial cell surface, TFPIs are displayed;

TFPIs block the actions of the factor-VIIa-tissue-factor complex, preventing the initiation of coagulation. Moreover, ECs produce and display heparin sulphate proteoglycans (HS) which bind to anti-thrombin III, and subsequently inhibit thrombin generated by the coagulation cascade. ECs also sequester vWF within storage granules named Weibel-Palade bodies (WPB). NO generated in ECs inhibits platelet activation (and therefore coagulation) (Pober & Sessa, 2007).

Figure 1.3 The TM/protein C/protein S anticoagulation system.

Among multiple anticoagulant pathways, the thrombomodulin (TM)/protein

C/protein S anticoagulant system is physiologically vital, as disruption of this

system is associated with many diseases such as severe thromboembolic

disease and severe sepsis (Dahlback & Villoutreix, 2005). This system inhibits the procoagulant function of activated factors VIII (FVIIIa) and V (FVa) (Esmon,

1995; Weiler, 2004). This pathway is initiated by a binding of TM, a membrane glycoprotein abundantly expressed on ECs, with thrombin which is the enzyme responsible for clot formation and a potent platelet activator. TM-thrombin complex rapidly cleaves protein C zymogen that binds to endothelial protein C receptor (EPCR) or circulating protein C to form anticoagulant, activated protein

C (APC). APC, working in concert with its cofactor protein S, proteolytically degrades FVa and FVIIIa of the coagulation cascade, thereby suppressing further thrombin generation (Figure 1.3). The TM-thrombin complex also can

11 activate the latent inhibitor of fibrinolysis, TAFI. The activated TAFI inhibits the interaction of fibrin with fibrinolytic enzymes, preventing fibrinolysis, a breakdown of fibrin clots. In addition, the TM/protein C/protein S system modifies receptor- mediated signaling processes that are initiated by coagulation factors by three different mechanisms. These include an activation of protease-activated receptor

(PAR)-1 and PAR2 via the APC-EPCR complex, a direct competition between

TM and PAR1, 3, and 4 for the ligand thrombin, and a prevention of the generation of signaling-competent coagulation proteases.

Vascular tone

ECs regulate vasomotor tone by producing several vasodilator (e.g. NO and prostacyclin) and vasoconstrictor substances (e.g. endothelin [ET-1]) in response to both humoral and mechanical stimuli (Galley & Webster, 2004; Kazmi, Boyce,

& Lwaleed, 2015). These vasoactive substances can profoundly influence structure and function of the neighboring smooth muscle cells. NO and prostacyclin are released from ECs in all vascular beds. NO is synthesized in

ECs by endothelial nitric oxide synthase (eNOS) which catalyses the conversion of L-arginine to NO (Figure 1.4). Changes in intracellular calcium (Ca2+) in

response to shear force or via a receptor-mediated process activate eNOS.

Subsequently, NO diffuses into SMC and activates guanylate cyclase (GC) which

converts GTP to cGMP. A protein kinase that is activated by cGMP inhibits Ca2+ influx into SMC, leading to decreased calcium-calmodulin stimulation of myosin light chains. This reduces smooth muscle tension development and causes

12 vasodilation. Furthermore, ECs synthesize prostacyclin from arachidonic acid in response to inflammatory factors (e.g. interleukin 1 and platelet-derived growth factor). Prostacyclin serves as a vasodilator and inhibits platelet aggregation and thrombosis. Vasoconstrictor actions of ET-1 are initiated by binding of ET-1 with

ETA receptors in vascular smooth muscle. In some blood vessels, ETB receptors

contribute to vasoconstriction. However, ET-1 can exert vasodilator actions

through stimulation of ETB receptors in ECs. Thus, the balance in endothelial

production of vasoactive substances and cross-talk between these substances

control vascular tone.

Figure 1.4 NO production by eNOS.

Inflammation/Leukocyte trafficking

Cytokines including chemokines, colony stimulating factors (CSF), IL, growth

factors, and interferons (IFN) are proteins that control both the amplitude and duration of the immune and inflammatory responses. ECs function as key mediators in host defense and inflammation; they produce and react to a number

13 of cytokines and other factors (Aird, 2007a; Galley & Webster, 2004). ECs also participate in passage of leukocytes from blood to underlying tissues at sites of inflammation or infection. Leukocyte trafficking takes place most exclusively in postcapillary venules and involves several steps which include initial attachment, rolling, firm adhesion, and transmigration through activated ECs (Aird, 2007a).

ECs present several chemokines to circulating leukocytes. ECs in an activated state present cell surface E- and P-selectin which interact with leukocyte carbohydrate-based ligands. This mediates rolling of leukocytes. Firm adhesion of leukocytes with the vessel wall is caused by an interaction between leukocyte integrins and endothelial intercellular adhesion molecule (ICAM)-1 and vascular cell adhesion molecule (VCAM)-1. Mechanism of transmigration is less well known, but it seems to involve PECAM-1, junctional adhesion molecule (JAM)-1, and CD99.

1.3 DYSFUNCTION OF ENDOTHELIAL CELLS AND DISEASES

Dysfunction of ECs is characterized by an alteration of the endothelial actions toward impaired vasodilation, pro-inflammation, and pro-thrombosis (Rajendran et al., 2013). Risk factors that are associated with endothelial dysfunction include smoking, aging, hypercholesterolemia, hypertension, hyperglycemia, obesity, elevated C-reactive protein, and chronic systemic Infection (Widlansky, Gokce,

Keaney, & Vita, 2003). Most of these cardiovascular risk factors can increase a number of free radicals in the body which can disrupt the balance of NO.

A shift in signaling from an NO-mediated silencing of cellular processes toward activation by redox signaling activates molecular machinery in ECs, leading to the expression of chemokines, cytokines, and adhesion molecules. NO which is produced in the quiescent ECs inhibits the expression of these molecules by targeting cysteine groups in key regulatory factors such as NF-κB

(a redox-sensitive transcription factor) and the mitochondria. As a consequence, the cellular processes are silenced. In contrast, reactive oxygen signaling predominates in the activated state of ECs. Reactive oxygen species, like NO, target key regulatory proteins such as NF-κB and phosphatases, promoting endothelial cell activation. A growing body of evidence has shown that endothelium dysfunction encompasses the development and progression of a broad spectrum of human disease states (W. C. Aird, 2004; Feletou & Vanhoutte,

2006; Rajendran et al., 2013). The examples are atherogenesis, heart failure, thrombosis, cancer, hemoglobinopathies, coagulation, sepsis, pulmonary hypertension, diabetes, inflammatory bowel disease, and stroke. Impairments of the organization of endothelial cell junctions have been related to human pathologies such as vascular malformations, hemorrhagic stroke, and edema

(Dejana et al., 2009).

Maintenance of healthy endothelial cell function and vascular integrity will assure delivery of blood and nutrients to all organs in the body. Dysfunction of

ECs plays a critical role in a variety of human diseases. However, critical regulators of endothelial cell functions have not been fully characterized in animal model. Identification of master regulators of endothelial cells will lead to the

15 prevention and more effective treatments of the diseases related with dysfunction of ECs.

1.4 REGULATION OF ENDOTHELIAL FUNCTION

The structure and function of vascular endothelial cells can be regulated by

various forms of biochemical (i.e. proinflammatory cytokine) and biomechanical

(i.e. laminar shear stress) stimuli (DePaola, Gimbrone, Davies, & Dewey, 1992;

Jain, Sangwung, & Hamik, 2014).

Biochemical stimuli

Exposure of ECs to cytokines leads to profound changes in gene expression and

function, allowing the cells to participate actively in the immune and inflammatory

reactions, hemostasis (Mantovani, Bussolino, & Dejana, 1992). For example,

high concentrations of tumor necrosis factor (TNF)-alpha suppress eNOS mRNA expression in human ECs (Yoshizumi, Perrella, Burnett, & Lee, 1993) and promote oxygen-derived free radical production (Madge & Pober, 2001).

Accordingly, ROS decreases NO bioavailability, which leads to decreased vosorelaxation. TNF-alpha also suppresses the expression of TM, a key endothelial anticoagulant cofactor, through NF-κB activation (Sohn et al., 2005) and induces the expression of tissue factor, which is a procoagulant cofactor

(Kirchhofer, Tschopp, Hadvary, & Baumgartner, 1994).

Advanced glycosylation end products (AGEs), which are proteins modified

by glucose that accumulate in the vasculature at a rapid rate in diabetic subjects,

have been shown to influence endothelial gene expressions and functions

(Stirban, Gawlowski, & Roden, 2014). For instance, AGEs increase vascular endothelial growth factor (VEGF) expression which in turn induces endothelial expression of monocyte adhesion and intercellular cell adhesion molecule-1

(ICAM-1) (Mamputu & Renier, 2004). This leads to increased monocyte adhesion to ECs. AGE-induced monocyte adhesion and ICAM-1 expression is oxidative stress-sensitive and involves PKC and NF-κB-dependent signaling pathways. In addition, AGEs promote oxidative stress, nitrotyrosine formation, NF-κB activation, resulting in capillary cell apoptosis (Kowluru, 2005).

Biomechanical stimuli

Laminar blood flow has been showed to promote key endothelial genes such as eNOS and TM which promote antithrombotic, antiadhesive, and antiinflammatory phenotypes of the endothelium. In contrast, disturbed blood flow leads to endothelial dysfunction and proinflammatory gene expression. Laminar sheer stress is observed at unbranched regions of vessels while turbulent sheer stress can be seen in branched and curvature portions. The earliest lesions in atherosclerosis occur in vessel sites with curvature and vessel branches where turbulent blood flow patterns are found. Expression of kruppel-like transcription factors (KLFs) 2 and 4 are decreased at branch points and major curves of vessels and are suppressed by turbulent blood flow (Atkins & Jain, 2007).

Accumulating data from cell-based studies in the past several years has identified KLF2 and KLF4 as key mediators for the expression of multiple endothelial gene products and endothelial cell functions.

1.5 KRUPPEL-LIKE FACTORS

Structure of KLFs

Kruppel-like factors (KLFs) are members of the C2H2-type zinc-finger family

(SP/KLF family) of DNA-binding transcription factors that play important roles in diverse biological events including cell stemness, differentiation, growth and functions (McConnell & Yang, 2010). The structure of KLF proteins consists of the transactivation and transrepression domains at the N-terminus and three consecutive zinc-finger moieties at the extreme C-terminus, each of which contains two conserved cysteine residues and two conserved histidine residues for zinc binding (Figure 1.5) (Jain et al., 2014). The linkers between the zinc- fingers contain a highly conserved 7-aa sequence (TGEKP(Y/F)X). KLFs bind to

GC-, GT-, and CACCC-box motifs which are found in gene promoters and other

regulatory elements. Residues that contribute to the DNA-binding specificity are

located in the zinc-fingers. Indeed, crystal structure analysis of a complex

between KLF4 zinc-fingers and its DNA target reveals that guanidinium groups of

three invariant arginines form critical hydrogen bonds with three guanine bases

and contribute the most to its DNA-binding specificity (Schuetz et al., 2011).

Figure 1.5 A schematic representation of common structure and functional domains for Kruppel-like factors (KLFs).

A total of 18 mammalian family members (KLF1-KLF18) including a predicted KLF18 have been identified to date (Pei & Grishin, 2013). The first mammalian Kruppel named erythroid Kruppel-like factor (EKLF/KLF1) was initially discovered in 1993 and subsequently shown to regulate β-globin synthesis and erythrocyte development. KLFs have influential effects on many cell types. In the endothelium, expression of three members of the KLF family ̶

KLF2, KLF4, and KLF6 ̶ has been documented (Botella et al., 2002; Kojima et al.,

2000; Kuo, Veselits, Barton, et al., 1997; Yet et al., 1998). To date, KLF2 and

KLF4 have emerged as important regulators of endothelial biology.

Identification and characterization of endothelial KLFs

KLF2 – Identification and characterization

KLF2, a 354-aa protein, was initially identified using a homology screening strategy by Lingrel and colleagues (Anderson, Kern, Crable, & Lingrel, 1995) and termed lung kruppel-like factor owing to its high expression in this tissue. Human and mouse KLF2 genes are >85% identical, and the homology in primary structure between human and mouse KLF2 proteins is greater than 90% (Wani,

Conkright, Jeffries, Hughes, & Lingrel, 1999). Subsequent studies revealed high expression of KLF2 in immune and endothelial cells, and their essential role in cellular quiescence (H. Das et al., 2006; Kuo, Veselits, & Leiden, 1997;

Mahabeleshwar et al., 2011). Further, studies from our group have identified endothelial KLF2 as a flow-mediated molecular switch that promotes flow- dependent induction of specific targets (e.g. TM and eNOS) and inhibition of

19 endothelial activation by cytokines (e.g. TNFα-induced IL1β) (Lin et al., 2005). As such, endothelial KLF2 conveys a pro-quiescence and anti-inflammation function.

Expression of KLF2 during embryonic development is observed at embryonic day 7 (E7) following by decreased expression at E11 and a subsequent increase at E15 (Anderson et al., 1995). Studies in rodents and humans show that while KLF2 is highly expressed in the endothelial layer of linear segments of the aorta, its expression is decreased at branch points

(Dekker et al., 2002). Consistent with KLF2’s pro-quiescence role in EC, these areas have been identified as the earliest artheroprone regions of the human vasculature (Gimbrone, Topper, Nagel, Anderson, & Garcia-Cardena, 2000).

Despite a number of in vitro observations on KLF2’s roles in ECs, in vivo validation of endothelial KLF2 action was proven to be difficult due to lethality of

KLF2 deficiency. Systemic KLF2-deficient mice die from cardiovascular abnormalities such as intraembryonic and intraamniotic hemorrhage, and exhibit impaired blood vessel maturation due to the lack of SMC recruitment and tunica media formation (Basu et al., 2005; Kuo, Veselits, Barton, et al., 1997).

Constitutive EC-specific deficiency or overexpression of KLF2 is also embryonic lethal ((Lee et al., 2006) and unpublished observation). Endothelial KLF2 deficiency leads to embryonic heart failure secondary to a high-cardiac-output state which is caused by a profound loss of peripheral vascular resistance (Lee et al., 2006).

Role of endothelial KLF2

Anti-inflammation and Inhibition of endothelial cell activation

Inflammatory stimuli (cytokines, pathogens) activate ECs by enhancing expression of chemokines and adhesion molecules on the cell surface to recruit immune cells to the vessel wall. The initial study from our group delineated for the first time that KLF2 serves as a key transcriptional regulator that inhibits endothelial proinflammatory activation. Specifically, KLF2 suppresses the induction of VCAM-1 and E-selectin expression following proinflammatory cytokine stimulation, leading to an attenuation of immune cell attachment to and rolling on ECs (Lin, Hamik, Jain, Kumar, & Jain, 2006; SenBanerjee et al., 2004).

Mechanistically, KLF2 inhibits NF-κB signaling at multiple levels. NF-κB pathway is a prototypical proinflammatory signaling pathway in which activation leads to induction of many proinflammatory genes (Lawrence, 2009). KLF2 can inhibit

NF-κB-mediated transcription by competing for its key transcriptional coactivator,

CBP/p300, damping the expression of NF-κB target genes such as VCAM-1 and

E-selectin (SenBanerjee et al., 2004). KLF2 can also suppress the expression of

PAR1, the principal receptor of thrombin, which subsequently prevents thrombin- mediated nuclear translocation of NF-κB and DNA binding (Lin et al., 2006).

Finally, genome-wide studies using microarray approach verifies the role of KLF2 in inhibiting endothelial activation and inflammation (Dekker et al., 2006; Parmar et al., 2006).

In addition, KLF2 inhibits the TGF-β signaling pathway: this pathway normally has proatherogenic effects on endothelium. KLF2 regulates the TGF-β

21 pathway via two different mechanisms. KLF2 induces expression of Smad7, a

known inhibitor of TGF-β signaling which competes with receptor-regulated

Smads (R-Smads) for association with the type I TGF-β receptor and induces

degradation of the type I receptor (Boon et al., 2007). As a result, the

phosphorylation and subsequent nuclear accumulation of Smad2 is inhibited,

leading to a suppression of the TGF-β-induced Smad4-mediated transcriptional

activity. Additionally, KLF2 inhibits an activator protein 1 (AP-1), an essential

cofactor for TGF-β-dependent transcription of many genes. KLF2 also attenuates

expression of activating transcription factor 2 (ATF2), which is among the heterodimeric components of AP-1 (Fledderus et al., 2007).

Anti-thrombotic function

ECs produce factors with anticoagulant, antiplatelet, and profibrinolytic effects in order to maintain blood fluidity. Studies by our group using gain and loss of KLF2 previously showed that KLF2 regulates key factors involved in the antithrombotic property of endothelial cell surface (Lin et al., 2005). Indeed, KLF2 induces TM, a key endothelial cell surface factor in the protein C anticoagulant pathway, by directly binding to the promoter region of TM. KLF2 also suppresses expression of plasminogen activator inhibitor-1 (PAI-1), a key modulator of the fibrinolytic pathway that controls the uPA/tPA/plasmin/MMP proteolytic activities and therefore maintains the tissue homeostasis (Ghosh & Vaughan, 2012).

Furthermore, KLF2 prevents cytokine-mediated induction of tissue factor (TF), the most powerful activator of the coagulation system (Grandaliano, Gesualdo,

Ranieri, Monno, & Schena, 2000). Additional players in the thrombotic pathway identified by genome-wide microarray studies are also under regulation of KLF2

(Dekker et al., 2006; Parmar et al., 2006). Thus, these observations position

KLF2 as a critical regulator of endothelial thrombotic function.

Anti-Proliferation, Migration, and Angiogenesis

Normally quiescent vascular ECs can receive input from their surrounding environment and undergo processes of new blood vessel formation including cell proliferation, cytoskeletal reorganization, migration and, finally, differentiation and formation of a new vessel (Munoz-Chapuli, Quesada, & Angel Medina, 2004).

This blood vessel formation occurs under both physiological (e.g. menstruation and placenta formation) and pathological (e.g. tumor growth and diabetic retinopathy) conditions (Hoeben et al., 2004). Multiple studies demonstrate that

KLF2 negatively regulates angiogenic stimulators. Studies using KLF2 overexpression and the nude ear mouse model of angiogenesis reveal the ability of KLF2 to inhibit VEGF‐A-mediated angiogenesis (Bhattacharya et al., 2005).

Mechanistically, KLF2 attenuates the expression of VEGFR2/KDR, the key

VEGF receptor, by competing with SP1 for promoter binding, resulting in decreased VEGF-mediated calcium flux, proliferation and induction of proinflammatory factors in ECs. Further, a wounding assay of HUVECs overexpressing KLF2 shows a significant reduction of migration of KLF2- overexpressing ECs (Dekker et al., 2006). This can be explained by the ability of

KLF2 to affect multiple genes including VEGFR2 and the potent antimigratory

23 semaphorin-3F (SEMA3F). Hence, KLF2 reduces cell proliferation and migration that collectively repress angiogenesis.

Vasoreactivity and vascular tone regulation by KLF2

ECs produce vasodilatory (e.g. eNOS and C-type natriuretic peptide) and vasoconstrictive (e.g. endothelin-1) factors that mediate vessel tone. Under pathological conditions, the imbalance between these factors shifts toward a vasoconstrictive state. Studies by our group initially demonstrated that KLF2 directly binds to and activates eNOS promoter. KLF2 recruits the transcriptional coactivator CBP/p300 to the eNOS promoter, leading to eNOS gene expression

(SenBanerjee et al., 2004). In addition to transcription, KLF2 also regulates eNOS enzymatic activity. Studies by Parmar et al showed that statin treatment, via KLF2 action, induces the gene expression of C-type natriuretic peptide and argininosuccinate synthetase, a limiting factor for eNOS substrate bioavailability

(Parmar et al., 2005). Moreover, overexpression of KLF2 decreases the expression of caveolin-1 (Parmar et al., 2006), a cell membrane protein that serves as a major negative regulator of eNOS activity (Ju, Zou, Venema, &

Venema, 1997). Besides the NO-mediated vasodilatory signaling, KLF2 can restrain the expression of vasoconstrictive mediators including endothelin, adrenomedullin, and angiotensin-converting enzyme (Dekker et al., 2006; Dekker et al., 2005). Altogether, KLF2 mediates expression of genes known to play important roles in vasoreactivity and vessel tone.

Maintenance of endothelial barrier function by KLF2

ECs form a barrier between blood and the tissues. Movement of fluid and solutes across the vessel wall is selectively regulated by endothelial tight junctions and adherens junctions. Our group asserts that KLF2 has a protective effect on the endothelial barrier, as validated by an increase in mustard oil-induced vascular leakage in the ear of KLF2 heterozygous mice. In vitro gain- and loss-of KLF2 studies also reveal the protective effect of KLF2 on EC monolayer permeability under diverse stimuli. The underlying mechanisms involve the ability of KLF2 to induce a key tight junction protein, occludin, and to reduce phosphorylation of myosin light chain, a key event in cell contraction (Lin et al., 2010). Furthermore, studies in mice with inducible global deletion or overexpression of KLF2 identify endothelial KLF2 as a key regulator of the blood brain barrier (BBB) function.

KLF2 regulates several key EC factors involved in BBB tight junction formation such as occludin, claudin-12, junction adhesion molecule-1 (JAM-1), and AF6

(Shi et al., 2013). Collectively, these observations mark KLF2 as key regulator of endothelial barrier function.

Regulation of endothelial KLF2

Laminar flow-mediated induction of KLF2

Given the importance of KLF2 in cell-based studies, how KLF2 expression is regulated in the endothelium has been the subject of considerable interest.

Laminar shear stress has atheroprotective effects on the endothelium, while non- laminar blood flow (turbulent flow) observed at branch point regions of the

25 vascular tree shows pro-atherogenic actions (Nayak, Lin, & Jain, 2011).

Expression of KLF2 is relatively high in the linear segments of the vessel, but attenuated at branch points (Dekker et al., 2002), suggestive of a flow-mediated expression pattern. Indeed, KLF2 is exclusively induced in ECs exposed to shear stress in vitro. Moreover, knockdown of KLF2 prevents flow-mediated induction of eNOS and flow-mediated reduction of endothelin-1, supporting the role of

KLF2 in regulating flow-mediated effects (Dekker et al., 2005). Subsequent studies reported that approximately 50% of the most highly flow regulated genes are dependent on upregulation of KLF2 (Parmar et al., 2006).

Studies to date suggest that the laminar flow-mediated increase in KLF2 expression is under regulation of a MEK5/ERK5/MEF2 signaling pathway. This is evidenced by prevention of flow-mediated endothelial KLF2 induction in ECs overexpressing a dominant negative MEF2 or mutant MEK5 (Parmar et al., 2006).

In addition to this pathway, laminar flow-mediated KLF2 induction is also regulated through phosphatidylinositol 3-kinase (PI3K) signaling pathway.

Specifically, components of a PI3K-dependent shear stress regulatory complex including p300/CBP-associated factor, heterogeneous nuclear ribonucleoprotein

D, and nucleolin bind to KLF2 promoter and thus promote KLF2 expression

(Huddleson, Ahmad, & Lingrel, 2006; Huddleson, Ahmad, Srinivasan, & Lingrel,

2005). Moreover, prolonged sheer stress, through PI3K signaling, can stabilize

KLF2 mRNA (van Thienen et al., 2006).

Another player that mediates flow-induced KLF2 expression is microRNA-

92a (miR-92a) (Wu et al., 2011). The expression levels of KLF2 and its targets

26 are reduced in ECs overexpressing miR-92a, but enhanced in ECs treated with miR-92a inhibitor. Pulsatile flow downregulates miR-92a precursor, and thus enhances KLF2 expression. Furthermore, Rhee et al found that RNA-binding protein HuR is increased in ECs exposed to oscillatory shear flow, and that siRNA knockdown of HuR increased KLF2 expression. However, regulation of

KlF levels by HuR is unclear and is not caused by changes of KLF2 mRNA stability despite a direct binding of HuR protein (Rhee et al., 2010) . Studies by

Jin and colleagues identified histone deacetylase 5 (HDAC5) as a negative regulator of flow-induced KLF2 and eNOS expression (W. Wang et al., 2010).

Shear stress causes phosphorylation of HDAC5, which in turn promotes HDAC5 nuclear export through a calcium/calmodulin-dependent pathway. Accordingly, association between HDAC5 and MEF2 is disrupted, leading to enhanced MEF2 transcriptional activity to induce KLF2 expression. Subsequent studies by the same group discovered that HDAC5 itself can directly bind to endothelial KLF2, and thus suppresses transcriptional activity of KLF2 (Kwon, Wang, Xu, & Jin,

2014). Similar to laminar flow-mediated dissociation of HDAC5 and MEF2, laminar flow-induced phosphorylation of HDAC5 suppresses its binding with

KLF2. Although pathways that regulate flow-mediated KLF2 induction has been revealed as mentioned above, the specific cell surface receptors or sensors that are responsible for initiating these pathways have not yet been uncovered.

Cytokine-mediated inhibition of KLF2

KLF2 expression is inhibited by many proinflammatory cytokines such as interleukin-1β (IL-1β) and TNF-α (Kumar, Lin, SenBanerjee, & Jain, 2005;

SenBanerjee et al., 2004). The TNF-α-mediated reduction of KLF2 involves both

NF-κB and HDAC pathways. TNF-α induces formation of a trimolecular complex

(p65, HDACs, and MEF2) that inhibits MEF2 functions and subsequently suppresses KLF2 expression.

Statin-mediated induction of KLF2

Statins, 3-hydroxyl-3-methyglutaryl-coenzyme A reductase (HMG-CoA reductase) inhibitors, are atheroprotective and has been widely used for treatments of cardiovascular diseases (Taylor et al., 2013). The link between statins and KLF2 was demonstrated by our group and others, showing statins robustly enhance the expression of KLF2 (Parmar et al., 2005; Sen-Banerjee et al., 2005). KLF2 induction by statins is dependent on the inhibition of the Rho pathway, as validated by a reduction of statin-induced KLF2 expression in response to geranylgeranyl pyrophosphate (GGPP) treatment. GGPP is a downstream product of HMG-CoA reductase and is the key isoprenoid intermediate for posttranslational modifications of Rho protein. In addition, siRNA knockdown of

KLF2 strongly abrogates mevastatin-mediated increase of downstream targets, eNOS and TM. Mutation of the MEF binding site on the KLF2 promoter region completely prevents the mevastatin-mediated induction of the KLF2 promoter.

Thus, these observations identify KLF2 as a nuclear mediator of statin effects in

ECs.

Other inducers of KLF2 expression

Resveratrol, a constituent of red wine, exerts protective effects across several diseases including cardiovascular diseases, cancer, and etc (Baur & Sinclair,

2006). It has been shown that resveratrol enhanced expression of KLF2 and its targets in ECs. The ability of resveratrol to mediate KLF2 induction is inhibited in

ECs treated with the non-specific sirtuin inhibitor sirtinol or transfected in siRNA

SIRT1, suggesting that resveratrol-induced KLF2 is dependent on SIRT1.

KLF2 induction by resveratrol is also dependent on MEK5/MEF2 signaling pathway, as evidenced by reduction in KLF2 induction by resveratrol in ECs expressing either a dominant-negative MEK5 or MEF2 mutant (Gracia-Sancho,

Villarreal, Zhang, & Garcia-Cardena, 2010).

Proteasome inhibitors are used for the treatment of multiple myeloma and potentially other types of malignancies. Proteasome inhibitors effectively prevent the proteasomal degradation of NF-κB inhibitor (IκB) and thus block NF-κB activation. Bortezomib, a dipeptide boronic acid derivative is the only proteasome inhibitor approved for human use (Orlowski & Kuhn, 2008). Interestingly,

Bortezomib induced KLF2 expression and KLF2-dependint gene targets such as

TM and eNOS, while suppressing levels of the pro-coagulant molecules in dose- dependent manner (Hiroi et al., 2009; Nayak et al., 2014). Cytokine-induced NF-

κB activation and induction of E-selectin and vascular adhesion cell molecule-1

(VCAM-1) are inhibited by Bortezomib. Knockdown of the RelA (p65) subunit of

NF-κB gene expression prevents cytokine-induced NF-κB activation but does not diminish TM upregulation by bortezomib, suggesting a signaling pathway independent of NF-κB, a principal target of proteasome inhibitors.

KLF4 – Identification and characterization

KLF4 was first identified in 1996 by Yang and colleagues (Shields, Christy, &

Yang, 1996), and named gut kruppel-like factor owing to its high expression in the epithelium of the gastrointestinal tract. KLF4-knockout mice are perinatal lethal because of dehydration due to skin barrier defects (Segre, Bauer, & Fuchs,

1999). KLF4 is vital for normal gastric epithelial homeostasis as conditional tissue-specific KLF4 deletion in mice results in abnormal differentiation of the gastric epithelial cells (Katz et al., 2005). Further, KLF4 is highly expressed in the mouse cornea, and conditional ablation of KLF4 in the surface ectoderm-derived structures of the eye leads to corneal epithelial fragility, stromal edema, and loss of conjunctival goblet cells (Swamynathan et al., 2007). In cancer biology, KLF4 has been shown to inhibit cell growth (Ghaleb et al., 2005). In stem cell biology,

KLF4 is one of the four Yamanaka factors that serve a role in maintaining pluripotent state of stem cells (Takahashi & Yamanaka, 2006).

Role of endothelial KLF4

KLF4 was identified in ECs and cloned from a human vascular EC cDNA library in 1998 (Yet et al., 1998). Initial insight regarding KLF4’s role in the endothelium was established by our laboratory in 2007 (Hamik et al., 2007). KLF4 is a flow inducible factor that, like KLF2, is able to augment expression of key endothelial targets such as eNOS and TM, and to inhibit a diverse set of pro-inflammatory factors (e.g PAI-1). Knockdown of KLF4 markedly increases the expression of

TNF-α-mediated pro-coagulant and pro-inflammatory mediators TF and VCAM-1.

These studies indicate that KLF4 modulates the inflammatory and coagulant state of the endothelium. Allied with KLF2, KLF4 regulates endothelial barrier function, as evidenced by increased transendothelial permeability in KLF4- deficient ECs and augmented lipopolysaccharide-induced lung injury and pulmonary edema following KLF4 depletion in mice (Cowan et al., 2010). This is explained by the ability of KLF4 to transcriptionally induce VE-cadherin expression by binding to its promoter. Our group unequivocally demonstrated that animals bearing endothelial-specific overexpression of KLF4 are resistant to atherothrombosis while an opposite effect is seen in mice bearing endothelial- specific deficiency of KLF4 (Zhou et al., 2012). Further, mice with KLF4 deficiency in endothelial and hematopoietic cells show increased injury-induced neointimal formation. KLF4 negatively regulates neointimal formation after carotid injury by counteracting the actions of a central regulator of inflammation, NF-κB

(Yoshida, Yamashita, Horimai, & Hayashi, 2014).

Similar to the function of KLF2, KLF4 regulates angiogenesis but through distinct mechanisms (Hale et al., 2014). KLF4 differentially regulates expression of Notch receptors, ligands, and target genes, and limits cleavage-mediated activation of Notch1. Recent studies using endothelial KLF4 conditional knockout

(cKO) mice denote an enhancement of neutrophil and lymphocyte accumulation and an elevation of cell adhesion molecule expression in injured kidneys, suggestive of a pro-inflammatory state of KLF4-dificient endothelium (Yoshida,

Yamashita, Iwai, & Hayashi, 2016). Additionally, the ability of statins to ameliorate renal ischemia-reperfusion injury is abrogated in KLF4 cKO mice.

Mechanistically, statins-induced KLF4 mediates the suppressive effect of statins on TNF-α-induced VCAM1 expression by reducing NF-κB binding to the VCAM1 promoter. Thus, these findings underscore the importance of KLF4 in endothelial biology.

Accumulating evidence from cell-based and in vivo studies suggest overlapping functions between KLF2 and KLF4. Both factors are increasingly appreciated as important regulators in endothelial biology. However, differences and similarities in the roles of endothelial KLF2 and KLF4 in vascular physiology and pathology in vivo remain uncharacterized.

Regulation of endothelial KLF4

Similar to KLF2, KLF4 expression is enhanced by laminar sheer stress, vasoprotective statins, and resveratrol (Hamik et al., 2007; Ohnesorge et al.,

2010; Villarreal et al., 2010; Yoshida et al., 2016; Zhou et al., 2012). Studies

32 assessing endothelial KLF4 regulation have revealed that the same

MEK/ERK5/MEF2 axis also induces KLF4. Bortezomib also promotes KLF4 expression (Hiroi et al., 2009; Nayak et al., 2014). In addition to KLF2, miR-92a suppresses KLF4 expression in arterial endothelium (Fang & Davies, 2012). Low expression levels of both KLFs are observed in swine aortic arch endothelium, a site with high atherosusceptibility where miR-92a expression is elevated, relative to protected thoracic aorta. However, unlike KLF2 that is reduced following cytokine stimulation, KLF4 expression is induced indicating differential expression in response to inflammatory stimuli (Hamik et al., 2007; SenBanerjee et al., 2004). The full repertoire of endothelial cell gene expression regulated by

KLF2 and KLF4 in response to cytokines has not been determined.

KLF6 – Identification, characterization, and functions

KLF6 was initially cloned from placenta and liver cDNA libraries in 1997 by two independent groups (Koritschoner et al., 1997; Lalazar, Wong, Yamasaki, &

Friedman, 1997), and identified in human peripheral blood leukocytes in 1998

(Suzuki et al., 1998). In the study by Ratziu et al, KLF6 was induced in activated hepatic stellate cells after liver injury, and functioned to transactivate collagen

α1(I), the key component of the hepatic scar (Ratziu et al., 1998). In addition,

KLF6 transactivates genes (e.g. TGF-β1, and types I and II TGF-β receptors) that are directly involved in the injury response. As such, KLF6 is considered as a damage–response factor (Kim et al., 1998). Furthermore, deletion of KLF6 in

33 mice results in embryonic lethality due to a failure of erythropoiesis and yolk sac vascularization (Matsumoto et al., 2006).

Endothelial KLF6 expression is induced in response to vascular injury

(Kojima et al., 2000). KLF6 subsequently activates the synthesis of urokinase

plasminogen activator (uPA) ̶ a key enzyme implicated in tissue remodeling,

tumor metastasis, and apoptosis ̶ in endothelial cells which leads to increased

bioactive transforming growth factor-beta (TGF-β). Upon vascular injury, KLF6 together with specificity protein 1 (SP1) transactivates expression of many

components of the TGF-β receptor complex. These include an endoglin (Botella

et al., 2002), and an activin receptor-Like Kinase-1 (ALK1), which is an

endothelial TGF-β receptor involved in angiogenesis (Garrido-Martin et al., 2013).

In addition, inhibition of the KLF6-SP2 complex formation enhances the

functional expression of matrix metalloproteinase-9 (MMP-9), which promotes

endothelial cell migration and mobility (A. Das et al., 2006). Recent in vitro and in

vivo studies show that up-regulation of KLF6 by vascular wounding promotes

MMP14 expression; MMP14 co-localizes with endoglin in ECs surrounding the

injured area (Gallardo-Vara et al., 2016). Many other genes (e.g. E-cadherin,

tissue factor pathway inhibitor-2, IL6 and VEGF) that are also involved in cell

motility and invasion during angiogenesis and vascular remodeling are under

regulation of KLF6 (McConnell & Yang, 2010). Thus, a body of evidence

supports a key role of endothelial KLF6 in regulating vascular development,

angiogenesis and vascular remodeling in response to vascular injury.

While much more needs to be understood, accumulating data support our hypothesis that KLF2 and KLF4 serve as nodal regulators that coordinate endothelial gene expression, which subsequently maintains healthy endothelial cell function and vascular hemostasis.

1.6 ENDOTHELIAL HEMOGLOBIN

Functional importance of hemoglobin

In each erythrocyte, around 200-300 million molecules of hemoglobin are found.

Hemoglobin concentration in blood is 13.5 – 18.0 g/dl in men and 11.5 – 16.0 g/dl in women (Thomas & Lemb, 2012). The functions of erythrocyte-encapsulated hemoglobin have been extensively explored. The main function of hemoglobin is to bind to and distribute O2 from the lung to all other tissues; each hemoglobin

carries up to four O2 molecules. Besides O2 transport, hemoglobin functions to

carry CO2 as carbaminohaemoglobin from tissues back to the lung for disposal.

In addition, hydrogen ions that form in the erythrocyte from the conversion of CO2 into bicarbonate are buffered by hemoglobin. Lastly, hemoglobin plays an additional role in NO metabolism (Thomas & Lemb, 2012). NO binds to the heme group or reacts with sulphydryl groups of cysteine residues in the beta-globin chain, forming nitrosothiols on the hemoglobin molecule. While binding of NO to heme is mostly seen in venous blood, nitrosohemoglobin is mostly formed in arterial blood. In a systemic capillary where hemoglobin saturation is decreased,

NO is released from its cysteine-bound position and may either be transferred to the heme group or be released from erythrocyte (Gaston, Singel, Doctor, &

Stamler, 2006; Thomas & Lemb, 2012). A conserved cysteine 93 within the β- chain (βCys93) which is oxidatively modified by NO to form an S-nitrosothiol

(SNO) is critical for the release of NO bioactivity. A mouse line in which βCys93 is replaced by alanine has intolerance to hypoxia, develops cardiac failure, and dies during transient hypoxia (Zhang et al., 2015; Zhang, Hess, Reynolds, &

Stamler, 2016).

Hemoglobin and disorders

Hemoglobin has been studied for more than 50 years and more than 1,000 hemoglobin disorders are known. Disorders of hemoglobin are inherited blood diseases that affect how O2 is carried in the body. Some of the most common

human genetic diseases worldwide are hemoglobinopathies (e.g., sickle cell

disease) and thalassemias (Forget & Bunn, 2013). Sickle-cell disease is caused

by abnormalities of hemoglobin structure in which a single-base mutation

(substitution of valine for glutamate) in beta-globin chain occurs. This leads to the

production of the unstable and relatively insoluble hemoglobin S, an inherited

variant of normal adult hemoglobin (hemoglobin A), which shortens red cell

survival and blocks small blood vessels (Wilson, Forsyth, & Whiteside, 2010).

Impaired hemoglobin production is found in thalassemia. A defect in the

production of either alpha or beta-globin chain leads to excess production of the

other chain, which in turn may initiate precipitation and eventually causes

hemolysis and anemia. Alpha-thalassemia is caused by deletion of one to four

36 alpha-globin genes and beta-thalassemia by a single mutation in beta-globin gene (Wilson et al., 2010).

Structure of hemoglobin

In adult human, most of the hemoglobin is a heterotetrameric complex (64.458

Da) which consists of four polypeptides: two alpha-globin and two beta-globin

(Schechter, 2008). Each alpha-chain and beta-chain contains 141 and 146 amino acids, respectively. Each globin chain binds to a heme prosthetic group with an iron atom by noncovalent forces. The iron atoms are primarily in the physiologic

2+ ferrous (Fe ) chemical valence state where oxygen (O2), carbon monoxide (CO), and nitric oxide (NO) can bind. Carbon dioxide (CO2) also interacts with hemoglobin at its amino-terminal residues as a weak carbamino complex, not by binding to the iron atoms.

Regulation of Hemoglobin

In erythroid cells, the expression of globin genes is tightly regulated. Human alpha-globin and beta-globin are encoded by the duplicated HBA1 and HBA2 genes and by the HBB genes, respectively. Four genes for alpha-like chains are clustered on chromosome 16; zeta-globin genes are at the 5’ end of the cluster, followed by two of the alpha-chain. Five genes for the beta-like chains are located on chromosome 11. At the 5’ of the cluster is one for epsilon, followed by two for gamma, one for delta, and one for beta-globin (Figure 1.6) (Thomas &

Lemb, 2012). Expression of the alpha-globin genes is governed by the HS-40

37 enhancer element that is located 40 kb upstream of zeta-globin. The beta-gene is controlled by the locus control region that is positioned 6-20 kb upstream of epsilon-globin (Pace & Makala, 2012). Multiple studies have identified transcription factors such as KLF1, KLF4, GATA-1, and NFE2 that govern the expression of hemoglobin genes in erythrocytes (Funnell et al., 2014; Kalra,

Alam, Choudhary, & Pace, 2011; Li, Ding, Li, Story, & Pace, 2012; Marini et al.,

2010).

Figure 1.6 Human hemoglobin genes located on chromosome 11 (Beta-like globin gene locus) and chromosome 16 (alpha -like globin gene locus). The functional beta-like genes are shown in gray. The functional alpha-like genes are shown in blue and the pseudogenes are in white. The expression of genes in both loci is regulated in a coordinated manner during the different stages of development.

Hemoglobin in non-erythroid cells

In addition to erythrocytes, studies over the past years identified the expression

of hemoglobin in non-erythroid cells including ECs, macrophages, and epithelial

cells (Saha et al., 2014). The landmark studies by Straub et al document that monomeric hemoglobin subunit alpha (HBA) is found in human and mouse arterial ECs, enriched in the MEJ, and regulates NO signaling (Straub et al.,

2012). Permissive NO diffusion to SMCs or NO scavenging by HBA is assigned

38 by the oxidation state of the HBA heme iron. This state is controlled by the activity of the cytochrome B5 reductase (CyB5R3) to convert the HBA heme iron from Fe3+ (NO diffusion) to Fe2+ state (NO scavenging). A regulation of NO

signaling by HBA also requires its interaction with eNOS in addition to binding to

CyB5R3 (Figure 1.7) (Straub, Butcher, et al., 2014). While eNOS acts as a

producer promoting NO synthesis in the ECs, HBA functions as NO scavenger

preventing NO diffusion to adjacent SMCs. Disruption of the macromolecular

complex of HBA and eNOS leads to an increase in bioavailable NO for

vasodilation. This is supported by in vivo studies using an HBA mimetic peptide

that demonstrate an increase in capillary oxygenation and arteriole blood flow,

and a sustained decrease in systolic blood pressure in normal and angiotensin II-

induced hypertensive mice (Keller et al., 2016). The post-translational regulation

and functions of endothelial hemoglobin subunit alpha has been extensively

characterized. However, its direct transcriptional regulation, and the molecular link that coordinates endothelial HBA expression, regulation, and NO signaling remain to be revealed.

The focus of this study is directed toward assessing the transcriptional regulation of hemoglobin expression in ECs. Accumulating studies demonstrate that KLF2 and KLF4 confer a healthy phenotype to ECs, at least in part by their ability to regulate expression of key endothelial genes such as eNOS. On the basis of these observations, we postulate the central hypothesis that KLF2 and

KLF4 play distinct and redundant roles in regulation of endothelial health and

39 vascular hemostasis. Further, these two factors control the expression of endothelial HBA which has a role in fine-tuning the regulation of vascular tone.

Figure 1.7 Possible mechanism of NO control at the MEJ. Hemoglobin subunit alpha (HBA) forms a complex with eNOS and CyB5R3 at the MEJ. When HBA resides in the Fe3+ state, the eNOS-mediated NO is diffused to the SMCs. Reduction of HBA from Fe3+ to Fe2+ state by the activity of CyB5R3 lead to scavenging of NO by HBA, blocking diffusion of NO.

CHAPTER 2

KLF2 AND KLF4 CONTROL ENDOTHELIAL IDENTITY AND VASCULAR

INTEGRITY

Sangwung, P., Zhou, G., Nayak, L., Chan, E.R., Kumar, S., Kang, D.W., Zhang,

R., Liao, X., Lu, Y., Sugi, K., Fujioka, H., Shi, H., Lapping, S.D., Ghosh, C.C.,

Higgins, S.J., Parikh, S.P., Jo, H., Jain, M.K.

JCI Insight. 2017 Feb 23; 2(4): e91700. doi:10.1172/jci.insight.91700.

Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700 2.1 ABSTRACT

Maintenance of vascular integrity in the adult animal is needed for survival, and it is critically dependent on the endothelial lining, which controls barrier function, blood fluidity, and flow dynamics. However, nodal regulators that coordinate endothelial identity and function in the adult animal remain poorly characterized.

Here, we show that endothelial KLF2 and KLF4 control a large segment of the endothelial transcriptome, thereby affecting virtually all key endothelial functions.

Inducible endothelial-specific deletion of Klf2 and/or Klf4 reveals that a single allele of either gene is sufficient for survival, but absence of both (EC-DKO) results in acute death from myocardial infarction, heart failure, and stroke. EC-

DKO animals exhibit profound compromise in vascular integrity and profound dysregulation of the coagulation system. Collectively, these studies establish an absolute requirement for KLF2/4 for maintenance of endothelial and vascular integrity in the adult animal.

2.2 INTRODUCTION

The maintenance of an intact vascular network to deliver oxygen and nutrients to all tissues is critical for organismal survival. The endothelium is critical to vascular integrity by virtue of its ability to control fundamental properties such as permeability, blood fluidity, and vasomotor tone (Aird, 2007a). The endothelium is also a dynamic and highly responsive tissue whose function can be altered by biomechanical (e.g., blood flow) and biochemical (e.g., cytokine) stimuli (W. C.

Aird, 2004). For example, laminar blood flow alters cellular gene expression in a manner that promotes a healthy endothelium and maintains vascular integrity while disturbed flow confers antiparallel effects (Heo, Fujiwara, & Abe, 2011; Jain et al., 2014). Further, biochemical stimuli, such as cytokines, can activate the endothelium, as seen in acute (e.g., sepsis) or chronic disease states (e.g., coronary artery disease), culminating in leakage of fluid from the intravascular space and tissue edema, formation of blood clots that impair flow, and altered vascular tone resulting in blood pressure dysregulation (W. C. Aird, 2004). Given the importance of vascular integrity in organismal survival, the identification of nodal regulators is of interest.

Studies over the past decade have led to the appreciation that members of the Kruppel-like family of transcription factors (KLFs) regulate endothelial biology (Atkins & Jain, 2007). In particular, two members of this family — namely

KLF2 and KLF4 — are enriched in the endothelium, regulated by flow and cytokines, and have been shown in cell-based studies to directly regulate key endothelial genes, such as endothelial nitric oxide synthase (Nos3) and

42 thrombomodulin (Thbd) (Dekker et al., 2002; Hamik et al., 2007; Lin et al., 2005;

Parmar et al., 2006). In vivo studies using systemic deletion of KLF2 and/or KLF4 result in death during embryonic development or shortly after birth (Basu et al.,

2005; Chiplunkar et al., 2013; Kuo, Veselits, Barton, et al., 1997; Segre et al.,

1999). Hemizygous or conditional global deletion of KLF2 alters vascular blood- brain barrier functions (Lin et al., 2010; Shi et al., 2013).

Finally, endothelial deficiency of KLF4 was found to render animals susceptible to atherosclerosis and thrombosis and injury-induced neointimal formation, underscoring the importance of this gene family in vessel biology

(Yoshida et al., 2014; Zhou et al., 2012). However, the importance of having two

KLFs with seemingly overlapping functions in the adult animal has not been illuminated and remains a major question in the field. Here, using inducible gene deletion approaches we have deleted endothelial Klf2, Klf4, or both in the adult organism. Our results demonstrate an absolute requirement for both factors in control of cardinal endothelial functions and maintenance of an intact vasculature.

2.3 MATERIALS AND METHODS

Animal models. All animals are of the C57BL6/J mouse strain. Endothelial- specific Klf2 and/or Klf4 knockout mice were generated by breeding floxed mice

(Klf2 f/f, Klf4 f/f, or Klf2 f/f-Klf4 f/f) with CRE mice (originally from R. Adams,

University of Münster, Münster, Germany). Endothelial-specific deletion of Klf2 plus one allele of Klf4 (EC-Klf2-KO/Klf4+/–) mice were generated by mating

Cdh5(PAC)-Ert2cre-Klf2 f/f-Klf4 f/f mice with Cdh5(PAC)-Ert2cre-Klf2 f/f mice.

Endothelial-specific deletion of Klf4 plus one allele of Klf2 (EC-Klf4-KO/Klf2+/–) mice were generated by mating Cdh5(PAC)-Ert2cre-Klf2 f/f-Klf4 f/f mice with

Cdh5(PAC)-Ert2cre-Klf4 f/f mice. To trigger endothelial-specific deletion of Klf2 and/or Klf4 gene deletion, 8- to 10-week-old mice were intraperitoneally injected with tamoxifen (2 mg/25 g) (MP Biomedicals). Cdh5(PAC)-Ert2cre mice were used as a control group. The survival study was conducted in both male and female mice. Other studies were performed in male mice.

EC isolation. Primary microvascular ECs were isolated from the heart tissue of mice using a standard technique as previously described with a minor modification (Y. C. Lim et al., 2003; Zhou et al., 2012). Briefly, the hearts were washed in cold PBS, minced with blades, and digested in PBS containing

1%BSA, collagenase type I, 1 mM CaCl2, and 1 mM MgCl2 at 37°C for 45 minutes. ECs were purified by using Dynabeads (Thermo Fisher Scientific) conjugated with anti-CD31 antibody (catalog 553370, BD Biosciences). The cells were immediately subjected to further experiments without culturing them on a

44 petri dish. We isolated ECs by mechanical mincing and enzymatic digestion to obtain single-cell suspension of the tissues. ECs were separated from the rest of the cells and were isolated by CD31-coated magnetic beads. Although CD31 is expressed on platelets and subsets of leukocytes in addition to ECs, its expression on ECs is higher than on platelets and subsets of leukocytes (Y.

Wang & Sheibani, 2002). In addition, CD31 is one of the two most specific antibodies used for isolation of ECs, and purity of the EC fraction sorted by CD31 antibody is >95% (van Beijnum, Rousch, Castermans, van der Linden, &

Griffioen, 2008).

RNA isolation and qPCR. Total RNA from ECs was isolated by using the High

Pure RNA Isolation Kit and reverse transcribed into cDNA using the iScript cDNA

Synthesis Kit (Bio-Rad). Quantitative real-time PCR (qPCR) was carried out using TaqMan Master Mix (Applied Biosystems), and the reactions were performed using the StepOnePlus Real-Time PCR System (Applied Biosystems).

Endogenous GAPDH was used as a normalizer for gene expression. The primer sequences used are listed in Table 2.1 Relative mRNA expression levels (fold changes) between groups were calculated using the δ-δ Ct method.

Table 2.1 Primer sequences for qPCR (TaqMan) analysis.

Gene Forward (5’ to 3’) Reverse (5’ to 3’) Probe number Klf2 CTAAAGGCGCATCTGCGTA TAGTGGCGGGTAAGCTCGT 48 Klf4 CGGGAAGGGAGAAGACACT GAGTTCCTCACGCCAACG 62 Nos3 ATCCAGTGCCCTGCTTCA GCAGGGCAAGTTAGGATCAG 12 Thbd ATGCGTGGAGCATGAGTG CTGGCATCGAGGAAGGTC 81 Serpine1 AGGATCGAGGTAAACGAGAGC GCGGGCTGAGATGACAAA 69 Gapdh TGTCCGTCGTGGATCTGAC CCTGCTTCACCACCTTCTTG 80 F3 TTCTCCAGGAAAACTAACCAAAA CCACAATGATGAGTGTTTCTCC 38 F2rl3 TGCTGTATCCTTTGGTGCTG TAGATGCTGGGGGTCTGG 64 Esam GGTTTTGTTCCTGGGACTGA CGTGCAACTCCATCTGAGC 7 Ceacam1 TGTGAAATCTCGAATCCAGTCA TCTGAGAGGCCTCCTTGTGT 63 Jam2 GAACCTGCAGGAAGATAAAGTCA GCACTTCACAGGCAGGAAC 92 Jam3 GAACTCGGAGACAGGCACTC TACTGCCCAGAGTCGTCCTT 56 Cgnl1 GAGTGTGGAGGAGGCAACC TGGCTCTGCTCTGAGGAGTT 18 Cldn5 ACGGGAGGAGCGCTTTAC GTTGGCGAACCAGCAGAG 66 Cldn11 GCCTGGAGTGGCCAAGTA AGATGGTGGCGACAATGG 20 Cldn12 CTCTGTCGCAGGCCTCTTT AATGTGATCAGCCGCAGTTT 38 Tjp1 TTTGAGAGCAAGCCTTCTGC AGCATCAGTTTCGGGTTTTC 4 Afdn GAGAATCCCAAGGACTACTGTATTG GCACCTCTCTCATCAGAATGC 1 Cgn GACCGACTCAACAAAGAGCTG GGCCTTTTCCTTGTAATCTTCC 1 Ocln GTCCGTGAGGCCTTTTGA GGTGCATAATGATTGGGTTTG 10 Vcl AGGAGACTTGCGAAGACAGG GCCGTCGCCACTTGTTTA 110 Actn2 TTTGCTGCTCAGGCCAAT AGCTCCGAGCGATCTCCT 81 Ctnnal1 CGCTGCTAAGGGAAATCAGT TGGGCTTTGGAAGTGAAAGA 91 Ctnna3 TGCCATATTTCATGAGCACAC TTGACATGGAACAAGCCAGA 10 Cdh5 GTTCAAGTTTGCCCTGAAGAA GTGATGTTGGCGGTGTTGT 56

RNA sequencing and bioinformatics analysis. The quality of RNA isolated from cardiac microvascular ECs (n = 4 per group, each n was pooled from 2 mice) was measured by Agilent 2200 TapeStation, and all samples have an RNA integrity number >7.4. Total RNA samples were submitted to the UCSF genomics core laboratories for RNA sequencing. Library preparation was employed using the Illumina TruSeq Stranded Total RNA kit. Sequencing was performed by using the Illumina SE 50-bp platform. Reads were aligned to the mouse genome

(Ensembl Mouse GRCm38) using tophat. Reads that mapped to known mRNAs were subjected to differential expression analysis. A total of 6,094 genes were determined to be differentially expressed between the DKO and control group using a q value of less than 0.05 (FDR-corrected P value). This list was filtered for genes with a >2-fold change, which narrowed our list to 2,743 genes. This list was used as the input for our gene set enrichment analysis using the

GenePattern toolset (Broad Institute). Pathways with a P value correction with

FDR < 0.05 were considered significant. RNA sequencing data have been deposited in the NCBI’s Gene Expression Omnibus and are accessible through

GEO series accession number GSE92965 (https://www.ncbi.nlm.nih. gov/geo/query/acc.cgi?acc=GSE92965).

Echocardiogram. At days 6 after tamoxifen administration, animals under anesthesia were placed on a temperature-controlled platform with heart rate and respiratory rate monitoring. Ultrasound gel was applied on the furless chest, and noninvasive images were captured when the heart rate was stable (500

47 beats/minute) by using a 14-MHz ultrasound probe that was connected to a

VisualSonic Vevo 770 machine.

Telemetry measurement and ST wave analysis. Animals were anesthetized by isoflurane. A sterilized DSI wireless telemetric transmitter (DSI) was implanted into the abdominal cavity. Animals were monitored on a telemetry receiver, and a baseline measurement (before tamoxifen injection) was conducted. Two days following the implantation, animals were injected with tamoxifen, and data were perpetually recorded for 9 days. ST wave analysis was conducted using

LabChart7 software.

Evans blue vascular permeability assay. After tamoxifen injection for 4 to 6 days, animals were anesthetized with ketamine/xylazine (170 mg/kg and 5 mg/kg, respectively) via intraperitoneal injection. One hundred μl of PBS containing 1%

EBD (Sigma-Aldrich) was administered via retro-orbital injection, and EBD was allowed to circulate in the body for 2 hours. Excess EBD in the vasculature and blood was removed via transcardial perfusion of PBS containing 2 mM EDTA for

8 minutes. The peristaltic pump was set at 4 ml/minute. The excised tissues were homogenized in formamide (Sigma-Aldrich), and EBD was extracted from the homogenized tissues by incubating at 70°C for 24 hours. Following a centrifugation at 12,000 g for 30 minutes, the absorbance of EBD in the supernatant was measured at 620 nm (the absorbance maximum for EBD) and

48 corrected for a contamination of heme pigments (OD740 nm) as follows: OD620

(corrected) = OD620 − (1.326 × OD740) + 0.03.

Peripheral blood count, blood smear, coagulation, and fibrinogen assays. Blood was collected from the inferior vena cava using 3.2% sodium citrate as an anticoagulant, and peripheral blood count was carried out by using an automated analyzer (Hemavet 950FS). A blood smear was stained with Wright’s stain, and the images were captured by a Leica DM2000 LED microscope (Leica

Microsystems Inc.). Whole blood was centrifuged at 2,000 g for 15 minutes at room temperature, and plasma was collected. Coagulation assays for the extrinsic and intrinsic pathways (prothrombin time and activated partial thromboplastin time, respectively) were carried out as described in the manufacturer’s instructions (Helena Laboratories). Plasma fibrinogen was measured by using the fibrinogen assay kit (Helena Laboratories).

ELISA. Plasma cardiac troponin I (Life Diagnostics Inc.), D-Dimer (Cloud-Clone

Corp), and Angpt-2 (Abcam) levels were determined as described in the manufacturers’ instructions.

Western blot analysis. Primary cardiac microvascular ECs were homogenized and lysed for 10 minutes at 4°C in RIPA buffer (Thermo Fisher Scientific) containing proteinase inhibitor cocktail. The lysed cells were centrifuged at

15,871 g for 10 minutes at 4°C, and the supernatant was collected. Protein was

49 separated on 4%–20% Tris-Glycine gel and transferred onto Nitrocellulose membrane. The membrane was blotted with primary antibodies against KLF4

(catalog AF3640, R&D Systems Inc.), eNOS (catalog 610296, BD Biosciences),

TM (catalog sc-7097, Santa Cruz Biotechnology Inc.), and GAPDH (catalog

G9545, Sigma-Aldrich). Secondary antibodies include horseradish peroxidase– conjugated anti-rabbit IgG (catalog 7074), anti-mouse IgG (catalog 7076,

Cell Signaling Technology Inc.), and anti-goat IgG (catalog sc-2020, Santa Cruz

Biotechnology Inc.)

Immunohistochemistry. Formalin-fixed, paraffin-embedded tissues were cut in 8-

μm-thick sections and mounted on coverslips for eosin staining. Apoptosis staining of ventricles (5-μm-thick sections) was assessed using the ApopTag

Peroxidase In Situ Apoptosis Detection Kit (Millipore) as described in manufacturer’s instructions. Tissue images were taken on a Leica DM2000 LED microscope.

Transmission electron microscopy. Animals were anesthetized and 1% lidocaine was injected through abdominal aorta toward the left to open the valves and relax the smooth muscle cells. Animals were perfused transcardially with triple aldehyde-DMSO at a flow rate of 10 ml/minute for 10 minutes. Brain and lungs were isolated from CRE and EC-DKO mice at day 6 after tamoxifen injection.

Small pieces of the brain and lung tissue were fixed for 2 hours at room temperature by immersion in the freshly prepared triple aldehyde-DMSO (Fujioka,

Tandler, & Hoppel, 2012). After rinsing in 0.1 M HEPES buffer (pH 7.3), they were postfixed in ferrocyanide-reduced osmium tetroxide. Another water rinse was followed by an overnight soak in acidified uranyl acetate. After again rinsing in distilled water, the tissue blocks were dehydrated in ascending concentrations of ethanol, passed through propylene oxide, and embedded in Poly/Bed resin

(Polysciences). Thin sections were sequentially stained with acidified uranyl acetate followed by a modification of Sato’s triple lead stain(Hanaichi et al., 1986).

These sections were examined in a FEI Tecnai Spirit (T12) transmission electron microscope with a Gatan US4000 4k × 4k CCD.

Statistics. Data are presented as mean ± SEM. Statistical analyses were performed using 2-tailed Student’s t test and ANOVA with post-hoc test for multiple comparisons to analyze the difference between 2 groups and among the groups, respectively. A P value equal to or less than 0.05 was considered significant.

Study approval. All animal studies were approved by the Case Western Reserve

University IACUC.

2.4 RESULTS

Endothelial-specific Klf2 and Klf4 deletion leads to rapid death of adult mice.

To evaluate the importance of endothelial KLF2 and KLF4 in adult vasculature, we generated mouse models that allow for inducible endothelial-specific deletion of Klf2 (EC-Klf2-KO), Klf4 (EC-Klf4-KO), Klf2 plus one allele of Klf4 (EC-Klf2-

KO/Klf4+/–), Klf4 plus one allele of Klf2 (EC- Klf4-KO/Klf2+/–), or both genes

(EC-DKO) through the tamoxifen-mediated activation of Cdh5(PAC)-Ert2cre

(CRE). Deletion of Klf2 and Klf4 as well as reduced expression of well- established downstream targets, such as Nos3 and Thbd, was confirmed in primary microvascular endothelial cells (EC) isolated from the heart at day 4 to 6 after tamoxifen injection (Figure 2.1, A–D). Strikingly, following tamoxifen injection, EC-DKO mice succumbed starting from day 6 after injection (40% of mortality rate) and reached a 100% mortality rate at day 9 (Figure 2.2A). By contrast, expression of one allele of Klf2 or Klf4 was sufficient for survival. Thus, we focused our analysis on EC-DKO mice.

To glean insights into the cause of acute death, we performed continuous telemetry monitoring before and after tamoxifen injection. Electrocardiogram

(EKG) recording revealed that EC-DKO animals experienced a progressive reduction in heart rate and elevation of T wave area (an indicator of ischemic heart injury) (Figure 2.2, B and C, and Figure 2.3). Echocardiographic analysis in

EC-DKO mice at day 6 revealed a significant decrease in cardiac function, as evidenced by reduced left ventricular fractional shortening, ejection fraction, and

52 cardiac output (Figure 2.2D). In addition, significant left ventricular hypertrophy and dilation (Figure 2.4, B and C), increased circulating cardiac troponin I

(indicative of myocyte injury; Figure 2.2E), and cardiomyocyte death (Figure

2.2F) were observed in EC-DKO mice. By contrast, the EKG and cardiac function were not significantly altered in EC-Klf2-KO and EC-Klf4-KO mice (Figures 2.3 and 2.4). Finally, gross postmortem examination revealed small-vessel hemorrhages in the brain, lungs, heart, and subcutaneous tissues of EC-DKO animals but not those of CRE or single-knockout animals (Figure 2.2G and

Figure 2.5). Consonant with these observations, EC-DKO mice exhibited stroke- like symptoms (e.g., partial loss of voluntary movement and weakness in the limbs). These results indicate that endothelial deletion of Klf2 and Klf4 led to compromised vasculature, which manifested as cardiac failure, neurologic dysfunction, and death.

Figure 2.1 Expression of Klf2, Klf4, and their targets in primary cardiac microvascular EC Relative mRNA expression levels of Klf2 (Kruppel-like family of transcription factor 2), Klf4, Nos3 (endothelial nitric oxide synthase), and Thbd (thrombomodulin) in primary cardiac microvascular EC isolated from mice at day 6 (A), day 5 (B), and day 4 (C) following tamoxifen (n=3-5 per genotype, each sample was pooled from 2 mice). (D) Representative western blot of primary cardiac microvascular EC using antibodies against KLF4, eNOS (endothelial nitric oxide synthase), TM (thrombomodulin), and GAPDH (endogenous glyceraldehyde 3-phosphate dehydrogenase). CRE: Cdh5(PAC)-Ert2cre, EC-DKO: EC-specific Klf2 and Klf4 double knockout, EC-Klf2-KO: EC-specific Klf2 knockout, EC-Klf4-KO: EC-specific Klf4 knockout. Sample in each group was pooled from 7-8 mice. *: P < 0.05, * *: P < 0.01, NS: Not significant. One-way ANOVA with Bonferroni's post hoc test. Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700 54

Figure 2.2 Endothelial-specific Klf2 and Klf4 deletion leads to rapid death of adult mice (A) Survival curve of tamoxifen-induced endothelial-specific Klf2 and/or Klf4 gene deletion in adult mice. n = 37 EC-specific Klf2 and Klf4 double knockout (EC-DKO); n = 14 Cdh5(PAC)-Ert2cre (CRE); n = 19 EC-specific Klf2 knockout (EC-Klf2-KO); n = 13 EC-specific Klf4 knockout (EC- Klf4-KO), n = 8 EC-specific knockout of Klf4 plus one allele of Klf2 (EC-Klf4-KO/Klf2+/–); n = 11, EC-specific knockout of Klf2 plus one allele of Klf4 mice (EC-Klf2-KO/Klf4+/–). (B) Reduction of heart rate and (C) ST wave segment elevation were observed in EC-DKO mice (n = 4–5 per genotype). (D) Fractional shortening (FS), ejection fraction (EF), and cardiac output (CO) in CRE (n = 3) and EC-DKO mice (n = 5) at day 6 after tamoxifen. (E) Circulating cardiac troponin I level (n = 11–16 per genotype) and (F) representative images of TUNEL staining in the heart showing massive cardiomyocyte death in EC-DKO mice at day 6 after tamoxifen (n = 3 per genotype). Scale bar: 100 μm. (G) Representative gross anatomy images of brain, lungs, heart, and subcutaneous tissues at day 6 after tamoxifen indicate spontaneous hemorrhage in EC-DKO mice (n = 3–4 per genotype). EC, endothelial cell. Data are presented as mean ± SEM values. *P < 0.05, **P < 0.01. 2-tailed Student’s t test. Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700

Figure 2.3 Electrocardiogram (EKG) recording. Representative EKG of CRE, EC-DKO, EC-Klf2-KO, and EC-Klf4-KO mice at baseline without tamoxifen (left) and day 6 post-tamoxifen injection (right). n=4-5 per genotype. CRE: Cdh5(PAC)- Ert2cre, EC-DKO: EC-specific Klf2 and Klf4 double knockout, EC-Klf2-KO: EC-specific Klf2 knockout, EC-Klf4-KO: EC-specific Klf4 knockout. Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700

Figure 2.4 Echocardiographic analysis at day 6 after tamoxifen administration. (A) Fractional shortening (FS), ejection fraction (EF), and cardiac output (CO) in CRE (n=3), EC- Klf2-KO (4), EC-Klf4-KO (n=4), and EC-DKO mice (n=5). (B) Left ventricle (LV)-to-body weight (BW) ratio. (C) LV internal dimension at end diastole (LVIDd), and at end systole (LVIDs). CRE: Cdh5(PAC)-Ert2cre, EC-DKO: EC-specific Klf2 and Klf4 double knockout, EC-Klf2-KO: EC- specific Klf2 knockout, EC-Klf4-KO: EC-specific Klf4 knockout. Data are presented as mean ± SEM. *: P < 0.05, * *: P < 0.01. NS: Not significant. One-way ANOVA with Bonferroni's post hoc test. Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700

Figure 2.5 Gross post-mortem examination in EC-Klf2-KO and EC-Klf4-KO mice. Representative gross anatomy pictures of brain, lungs, heart, and subcutaneous tissues from EC- Klf2-KO (EC-specific Klf2 knockout) and EC-Klf4-KO (EC-specific Klf4 knockout) mice at day 6 post-tamoxifen (n=3-4 per genotype). Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700

Endothelial-specific deletion of Klf2 and Klf4 leads to vascular leak and systemic coagulopathy.

To understand the basis for widespread small-vessel hemorrhage in multiple organs, we hypothesized that alterations in fundamental properties, such as barrier function and/or blood coagulation, may be operative. Histological analyses of brain and lung tissues revealed extravasation of blood into adjacent tissues (Figure 2.6A, top row, and Figure 2.7A). Electron micrographs demonstrated loss of EC contiguity (Figure 2.6A, middle row; Figure 2.7A, and

Figure 2.8). To gain a quantitative assessment of vascular permeability, we injected control and EC-DKO mice with Evans Blue dye (EBD) and examined the tissues for extravasation at day 4–6 after tamoxifen injection. Significant EBD accumulation in the extravascular tissues of the lungs and kidneys was obvious in EC-DKO mice at day 4–5 (Figure 2.7, B and C) and the brain, lungs, kidneys, and heart at day 6 (Figure 2.6A, bottom row; and Figure 2.7, A and D).

Consistent with these observations, qPCR analyses revealed a marked reduction in a number of tight junction and adherens junction genes in ECs from EC-DKO mice (Figure 2.6B). Finally, previous studies have shown that increased angiopoietin-2 (Angpt-2) contributes to vascular leak (Benest et al., 2013;

Gallagher et al., 2007; Roviezzo et al., 2005). As shown in Figure 2.6C, circulating Angpt-2 levels were markedly increased in EC-DKO mice.

Another etiology for widespread, multiorgan hemorrhage is alteration in the hemostatic and coagulation systems. To assess the coagulation system, we first performed a complete blood count after tamoxifen injection. While no effect

59 was seen on white blood cell or red blood cell counts, a significant reduction in platelets (thrombocytopenia) was observed in EC-DKO animals (Figure 2.6D).

Further, microscopic examination of the peripheral blood smear revealed the presence of fragmented red blood cells (i.e., schistocytes) in EC-DKO animals at day 6 after tamoxifen injection (Figure 2.6E). Given the combination of low platelet counts and fragmented red blood cells, we posited that the precipitous drop in platelets was likely secondary to excessive consumption, a scenario seen in coagulopathic states such as disseminated intravascular coagulation (Levi,

Toh, Thachil, & Watson, 2009). Indeed, blood coagulation studies revealed significant elevations in prothrombin, activated partial thromboplastin times, and plasma D-dimer levels in EC-DKO mice (Figure 2.6, F and G), indicating dysregulation of coagulation.

Intact blood flow depends upon a fine balance between circulating and endothelial procoagulant and anticoagulant factors. Our group previously reported that both KLF2 and KLF4 induced the expression of thrombomodulin

(Hamik et al., 2007; Lin et al., 2005), the key blood-clotting regulator in the protein C anticoagulation pathway that is expressed in the endothelium (Healy,

Rayburn, Rosenberg, & Weiler, 1995; Weiler-Guettler, Aird, Husain, Rayburn, &

Rosenberg, 1996). We observed a near-complete loss of thrombomodulin expression both in mRNA and protein levels (Figure 2.1, A–D) in the ECs isolated from EC-DKO mice at day 6 after tamoxifen. We also looked at other factors in procoagulation and anticoagulation pathways. While expression of tissue factor (F3) was not significantly changed, levels of the serine (or cysteine)

60 peptidase inhibitor, clade E, member1 (Serpine1) and coagulation factor II receptor-like 3 (F2rl3) were increased in the EC-DKO mice (Figure 2.9). The above results provide evidence for the presence of a coagulopathic condition on day 6 after tamoxifen injection in EC-DKO mice that is likely secondary to the decline in known antithrombotic factors such as thrombomodulin. This acquired deficiency in thrombomodulin is similar to observations in genetic models of thrombomodulin deficiency (Isermann et al., 2001; Weiler-Guettler et al., 1996) and suggests that loss of this critical factor on ECs likely contributes to development of a prothrombotic milieu, resulting in widespread consumption of coagulation factors and platelets, leading to generalized bleeding.

Figure 2.6 Endothelial-specific deletion of Klf2 and Klf4 leads to vascular leak and systemic coagulopathy. (A, top) Representative pictures of brain Eosin staining show erythrocyte leakage (white arrow) in extravascular tissues at day 6 post-tamoxifen (n=3 per genotype). Scale bar is 50 µm. (A, middle) Representative electron microscopic (EM) images of brain indicate discontinuity of endothelial monolayer in the EC-DKO mice at day 6 post-tamoxifen (n=3-4 per genotype). Scale bar is 2 µm. Endothelial cells (EC): red arrowhead, extravascular erythrocytes: white arrow, vessel lumen: black arrowhead. Red outlined inset is a high magnification image of the indicated region. Scale bar is 1 µm. (A, bottom) Representative pictures of Evans blue vascular permeability assay for the brain at day 6 post-tamoxifen (n=7-8 per genotype). (B) Relative mRNA expression levels of endothelial tight and adherens junction genes in primary cardiac microvascular EC 6 days after tamoxifen injection (n=3-5 per genotype, each sample was pooled from 2 mice). (C) Circulating Angiopoietin 2 protein levels pre- and post- tamoxifen (n=6-8 per genotype). (D) White blood cell (WBC), red blood cell (RBC), and blood platelet (PLT) counts at day 6 post-tamoxifen (n=25-29 per genotype). (E) Representative

62 pictures of blood smear in CRE and EC-DKO mice (n=3 per genotype). White arrows indicate erythrocyte fragmentation. Scale bar is 50 µm. (F) Extrinsic and intrinsic pathway coagulation assay (n=11-13 per genotype). (G) Plasma D-Dimer levels at day 6 after tamoxifen (CRE, n=5; EC-DKO, n=6). CRE: Cdh5(PAC)-Ert2cre, EC-DKO: EC-specific Klf2 and Klf4 double knockout. Data are presented as mean ± SEM. *: P < 0.05; * *: P < 0.01; NS: Not significant. Two-way ANOVA with Tukey’s post hoc test and 2-tailed Student’s t test. Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700

Figure 2.7 Endothelial-specific deletion of Klf2 and Klf4 results in vascular leak. (A, top) Representative pictures of lung Eosin staining show erythrocyte leakage (white arrow) in extravascular tissues at day 6 post-tamoxifen (n=3-5 per genotype). Scale bar is 200 µm. (A, middle) Representative electron microscopic (EM) images of lung indicate discontinuity of endothelial monolayer at day 6 post-tamoxifen (n=3-4 per genotype). Scale bar is 2 µm. Endothelial cells (EC): red arrowhead, extravascular erythrocytes: white arrow, leucocytes: blue arrow. (A, bottom) Representative pictures of Evans blue vascular permeability assay for the lungs and kidneys at day 6 post-tamoxifen (n=7-8 per genotype). (B-D) Quantification of extravasated Evans Blue dye in the lungs, kidneys, and heart at day 4 (B, n=6-8 per genotype), day 5 (C, n=6-9 per genotype), and day 6 (D, n=7-8 per genotype) post-tamoxifen. CRE: Cdh5(PAC)-Ert2cre, EC-DKO: EC-specific Klf2 and Klf4 double knockout. Data are presented as mean ± SEM. *: P < 0.05, * *: P < 0.01, NS: Not significant. 2-tailed Student’s t test. Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700

Figure 2.8 Representative electron microscopic (EM) images of brain indicate degeneration of EC and extravascular erythrocytes in the EC-DKO mice at day 6 post- tamoxifen. Scale bar is 2 µm. Endothelial cells (EC): red arrowhead, extravascular erythrocytes: white arrow, leucocytes: blue arrow, vessel lumen: black arrowhead (n=3-4 per genotype). EC- DKO: EC-specific Klf2 and Klf4 double knockout. Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700

Figure 2.9 Expression of F3, Serpine1, and F2rl3 mRNA in primary cardiac microvascular EC at day 6 post-tamoxifen. n=3-4 per genotype, each sample was pooled from 2 mice. F3: tissue factor, Serpine1: serine (or cysteine) peptidase inhibitor, clade E, member 1, F2rl3: coagulation factor II receptor-like 3, CRE: Cdh5(PAC)-Ert2cre, EC-DKO: EC-specific Klf2 and Klf4 double knockout. *: P < 0.05, * *: P < 0.01, NS: Not significant. 2-tailed Student’s t test. Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700 .

Endothelial-specific Klf2 and Klf4 deletion results in profound alterations in the EC transcriptome.

Given that dual loss of KLF2 and KLF4 led to a profound compromise of the most fundamental properties of the vasculature, we posited that the alteration in gene expression may be quite significant. Unbiased transcriptomic analyses in mouse cardiac microvascular ECs derived from CRE and EC-DKO mice at the beginning of day 6 after tamoxifen injection demonstrated extensive differences in the endothelial transcriptome (6,094 genes with q < 0.05; Figure 2.10, A and B; lists of genes are deposited in the NCBI’s GEO, accession number GSE92965). By contrast, we identified a small number of differentially expressed genes in EC-

Klf2-KO (45 genes with q < 0.05) and EC-Klf4-KO (30 genes with q < 0.05) mice when compared with CRE animals. Gene set enrichment analysis of ECs from

DKO mice revealed marked alterations (q < 0.05) in 28 of the 50 hallmark gene pathways (Figure 2.10C), while no significant changes were noted in single- knockout mice. These observations indicate that loss of both KLF2 and KLF4 profoundly altered the endothelial transcriptomic landscape.

Figure 2.10 Endothelial-specific Klf2 and Klf4 deletion results in profound alterations in the EC transcriptome. (A) A volcano plot showing 6,094 significantly differentially expressed genes with q<0.05 (red dots) in primary cardiac microvascular endothelial cells (EC) obtained from EC- DKO when compared to CRE. Significantly differentially expressed genes were determined by Cuffdiff. (B) Heatmap of top 200 genes differentially expressed in primary cardiac microvascular ECs of CRE and that of EC-DKO (n=4 per genotype, each n was pooled from 2 mice). (C) Normalized Enrichment Score (NES) represents the distribution of 28 of the 50 hallmark gene pathways. Positive and negative values indicate up-regulation of genes in the CRE in comparison to EC-DKO and up regulation of genes in EC-DKO relative to CRE, respectively. CRE: Cdh5(PAC)-Ert2cre, EC-DKO: EC-specific Klf2 and Klf4 double knockout. Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700 .

2.5 DISCUSSION AND CONCLUSION

Evolutionary studies suggest that the vascular system arose approximately 600 million years ago to overcome the time-distance constraints of diffusion critical advance for metazoan life (Monahan-Earley, Dvorak, & Aird, 2013). The endothelium arose to provide barrier function, localize immune/coagulation functions, and optimize flow to help overcome diffusion constraints to ensure adequate supply of oxygen and nutrients to all tissues. In this study, we advance that view that two transcription factors, KLF2 and KLF4, govern these ancient and fundamental properties of the endothelium and vascular integrity.

We demonstrated, for the first time to our knowledge, that loss of both endothelial KLF2 and KLF4 expression in adult animals resulted in acute vascular dysfunction and death. The profound vascular defects seen in EC-DKO mice are likely a combined result of vascular barrier disruption as well as a generalized state of consumptive coagulopathy. Although widespread damage to the endothelium or vascular barrier (secondary to junction proteins) can trigger consumptive coagulation, our results demonstrating early, substantial loss of thrombomodulin suggest that acquired thrombomodulin deficiency is an important contributing factor to the development of a consumptive coagulopathy.

This is consistent with previous murine studies wherein endothelial-specific thrombomodulin deletion resulted in cerebellar thrombohemorrhagic strokes, generalized thrombosis, and disseminated intravascular coagulation (Isermann et al., 2001; Weiler-Guettler et al., 1996). Further, RNA sequencing data showed changes of genes (q < 0.05) in the coagulation pathway (Table 2.2), and

69 examination of other factors important for hemostasis revealed significant elevation in Serpine1 and F2rl3 expression, predisposing to the development of a prothrombotic milieu. Thus, both EC-autonomous function and altered vascular barrier function likely contribute to the full phenotype observed in EC-DKO mice.

We also note that our findings are consistent with but add significantly to previous findings. Specifically, Chiplunkar et al. (Chiplunkar et al., 2013) reported the role of both factors in regulating vascular integrity using systemic knockout embryos of both Klf2 and Klf4 genes. However, their study used a global deletion of Klf2 and Klf4 genes in the embryos and focused on early development. In contrast, our study specifically abrogated both Klf2 and Klf4 genes in ECs in adult animals. Thus, the present study definitively highlights the absolute requirement of these two KLF factors to maintain normal EC function and consequently vessel integrity in the adult animal.

Another important aspect of our work relates to the observation that a single allele of either Klf2 or Klf4 is sufficient for maintenance of intact vasculature and survival. These data provide insights into a long-standing question in the field regarding redundancy between KLF2 and KLF4. Indeed, we and others have reported that both factors are induced by the same stimuli (e.g., flow and statins) and that both regulate similar key endothelial targets (e.g., Nos3 and Thbd) to confer antiinflammatory and antithrombotic effects to the vessel wall

(Dekker et al., 2002; Jain et al., 2014; Lin et al., 2005; SenBanerjee et al., 2004;

Zhou et al., 2012). Given the obvious requirement for an intact vasculature, it is likely that the redundancy of KLF2/4 provides an important safety net to ensure

70 survival. However, the fact that one allele is sufficient for survival does not necessarily mean that the vasculature is healthy. Indeed, endothelial loss of

KLF4 alone renders animals susceptible to chronic inflammation and atherothrombosis and injury-induced neointimal formation (Yoshida et al., 2014;

Zhou et al., 2012). Further, previous studies showed that loss of either KLF2 or

KLF4 renders the vasculature susceptible to leakage in response to inflammatory or ischemic injury (Cowan et al., 2010; Lin et al., 2010). Thus, it is likely that mice lacking 3 of 4 alleles will be susceptible to vascular dysfunction under stress conditions.

Another key finding relates to the breadth of the endothelial transcriptome under control of KLF2/4. It is hypothesized that, while the genome consists of approximately 20,000 genes (International Human Genome Sequencing, 2004), most cells express approximately 50%–60% of these at any given time. The large transcriptomic alterations seen in ECs deficient in KLF2 and KLF4 suggest that these 2 factors control a large segment of the active endothelial transcriptome, affecting numerous cellular processes, indicating a profound landscape-level change in EC identity and function. A broad change in gene expression was observed at the beginning of day 6 after tamoxifen, reflective of changes before catastrophic functional collapse. That said, we cannot rule out the possibility that the interaction of ECs with other cells and plasma components may contribute. Future studies assessing RNA sequencing at day 3–5 may be helpful in this regard. Collectively, our findings establish endothelial KLF2 and

KLF4 as master regulators of endothelial biology and vascular integrity.

Table 2.2 Differentially expressed genes (q<0.05) in the Hallmark coagulation pathway (EC-DKO vs CRE) at day 6 post-tamoxifen Ensembl_ID Gene Description EC-DKO vs EC-DKO vs CRE.RawP EC-DKO vs CRE.log2FC CRE.FDR ENSMUSG00000027875 Hmgcs2 3-hydroxy-3-methylglutaryl-Coenzyme A synthase 2 -2.044163402 3.82E-05 0.000360995 ENSMUSG00000064023 Klk8 kallikrein related-peptidase 8 -2.184932988 2.64E-19 3.55E-17 ENSMUSG00000030054 Gp9 glycoprotein 9 (platelet) -1.494656684 0.003454284 0.018829565 ENSMUSG00000074743 Thbd thrombomodulin -1.987964929 4.29E-29 1.49E-26 ENSMUSG00000031066 Usp11 ubiquitin specific peptidase 11 -1.016095878 1.84E-06 2.45E-05

ENSMUSG00000018593 Sparc secreted acidic cysteine rich glycoprotein 1.92362113 4.01E-43 4.51E-40 ENSMUSG00000037411 Serpine1 serine (or cysteine) peptidase inhibitor, clade E, member 1 2.968612577 7.75E-10 2.15E-08 ENSMUSG00000001131 Timp1 tissue inhibitor of metalloproteinase 1 3.617368817 9.90E-15 6.84E-13 ENSMUSG00000020689 Itgb3 integrin beta 3 2.676751127 5.99E-08 1.13E-06 ENSMUSG00000000957 Mmp14 matrix metallopeptidase 14 (membrane-inserted) 3.346851409 1.60E-14 1.06E-12 ENSMUSG00000015533 Itga2 integrin alpha 2 2.426126163 9.20E-21 1.48E-18 ENSMUSG00000029664 Tfpi2 tissue factor pathway inhibitor 2 1.178684483 8.38E-12 3.48E-10 ENSMUSG00000040152 Thbs1 thrombospondin 1 2.513729725 1.48E-08 3.12E-07 ENSMUSG00000043613 Mmp3 matrix metallopeptidase 3 3.179156931 2.70E-05 0.000265317 ENSMUSG00000030162 Olr1 oxidized low density lipoprotein (lectin-like) receptor 1 1.933678157 1.66E-07 2.83E-06 ENSMUSG00000001249 Hpn hepsin 1.084598753 0.00194077 0.011415364 ENSMUSG00000042622 Maff v-maf musculoaponeurotic fibrosarcoma oncogene family, protein 1.201299954 1.78E-05 0.000184435 F (avian) ENSMUSG00000026715 Serpinc1 serine (or cysteine) peptidase inhibitor, clade C (antithrombin), 1.343104001 3.88E-08 7.58E-07 member 1 ENSMUSG00000021822 Plau plasminogen activator, urokinase 2.243213549 5.79E-49 9.58E-46 ENSMUSG00000031538 Plat plasminogen activator, tissue 1.588270362 4.34E-05 0.000403984 ENSMUSG00000021190 Lgmn legumain 1.257164738 5.28E-10 1.51E-08 ENSMUSG00000026421 Csrp1 cysteine and glycine-rich protein 1 1.241158903 3.64E-19 4.76E-17 ENSMUSG00000024659 Anxa1 annexin A1 1.54903684 7.43E-27 1.99E-24 ENSMUSG00000021877 Arf4 ADP-ribosylation factor 4 1.013203713 2.50E-12 1.17E-10 Republished with permission of American Society for Clinical Investigation, from KLF2 and KLF4 control endothelial identity and vascular integrity, Sangwung et al., Volume 2, Issue 4, e91700, doi:10.1172/jci.insight.91700,February 23, 2017, Copyright (2017); permission conveyed through 72 Copyright Clearance Center, Inc. https://insight.jci.org/articles/view/91700 2.6 AUTHOR CONTRIBUTIONS

MKJ, LN, HJ, and PS designed the studies. PS, GZ, LN, SK, DWK, YL, HF, HS,

SDL, CCG, and SJH performed the experiments. RZ performed animal surgeries.

KS performed echocardiogram. PS, GZ, RZ, ERC, LN, XL, YL, KS, HF, SK, CCG, and SJH analyzed the data. MKJ, XL, YL, SMP, and HJ provided research guidance. PS prepared the figures and drafted the manuscript. MKJ edited the manuscript and approved final version of the manuscript.

2.7 ACKNOWLEDGMENTS

This work was supported by NIH grants R01HL110630-01, R01HL112486,

R01HL086548, and R01HL119195 (to MKJ); R01HL093234 and R01HL125275

(to SMP); and K08HL121131-01 (to LN) as well as by American Heart

Association grants 12SDG12050558 (to YL), 12SDG12070077 (to XL), and

16POST31200017 (to SJH).

CHAPTER 3

REGULATION OF ENDOTHELIAL HEMOGLOBIN ALPHA EXPRESSION BY

KRUPPEL-LIKE FACTORS

Sangwung, P., Zhou, G., Lu, Y., Liao, X., Wang, B., Mutchler, SM., Miller, M.,

Chance, MR., Straub, AC., and Jain, MK.

Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like

Reprinted by permission of SAGE Publications, Ltd.

3.1 ABSTRACT

Hemoglobin subunit alpha (HBA) expression in endothelial cells (ECs) has

recently been shown to control vascular tone and function. We sought to

elucidate the transcriptional regulation of HBA expression in the EC. Gain-of-

KLF2 or KLF4 function studies led to significant induction of HBA in ECs. An

opposite effect was observed in ECs isolated from animals with endothelial-

specific ablation of Klf2, Klf4 or both. Promoter reporter assays demonstrated

that KLF2/KLF4 transactivated the hemoglobin alpha promoter, an effect that

was abrogated following mutation of all 4 putative KLF binding sites. Fine

promoter mutational studies localized three out of four KLF binding sites (sites 2,

3, and 4) as critical for the transactivation of the HBA promoter by KLF2/KLF4.

Chromatin immunoprecipitation studies showed that KLF4 bound to the HBA

promoter in ECs. Thus, KLF2/KLF4 serve as important regulators that promote

HBA expression in the endothelium.

3.2 INTRODUCTION

Hemoglobin is expressed in both erythroid cells and a variety of non-erythroid cells such as endothelial cells (ECs), macrophages, and epithelial cells (Saha et al., 2014). In erythroid cells, two alpha globin chains assemble with two beta globin chains to form adult hemoglobin heterotetramers with four heme prosthetic groups which function to carry oxygen to tissues and transport carbon dioxide away from tissues to the lungs (Schechter, 2008). Transcription factors such as

KLF1, GATA1, and NFE2 play a role in regulation of hemoglobin genes in erythroid cells (Funnell et al., 2014; Kalra et al., 2011; Li et al., 2012; Marini et al.,

2010). However, the mechanisms of hemoglobin regulation in non-erythroid cells remain unknown.

ECs are critical constituents of the blood vessel wall (Aird, 2007a) and their dysfunction is a proximate event in the development of numerous cardiovascular diseases (William C. Aird, 2004; Feletou & Vanhoutte, 2006; S. S.

Lim et al., 2012). Previous studies have documented that hemoglobin subunit alpha (HBA) is expressed in ECs and enriched in the myoendothelial junction

(MEJ), a portion of ECs that protrudes through the internal elastic lamina and juxtaposes vascular smooth muscle cells (SMCs) (Rahaman et al., 2015; Straub,

Butcher, et al., 2014; Straub et al., 2012). These specialized structures serve as a conduit for communication between these two cell types such as the passage of NO from EC to SMC, a process critical for the regulation of vascular tone and function (Feletou, 2011; Straub et al., 2012; Straub, Zeigler, et al., 2014). The

MEJ is typically observed in smaller diameter vessels (e.g. small arteries, veins

75 or arterioles) as opposed to larger ones. Endothelial HBA functions to fine-tune

NO diffusion to SMCs (Straub et al., 2012). Efficiency of NO scavenging by HBA requires its coupling to eNOS and active heme iron redox cycling by cytochrome

B5 reductase 3 (CYB5R3), which reduces the HBA heme iron from the Fe3+,

nonreactive form, to the Fe2+ state capable of blocking NO diffusion (Rahaman et

al., 2015; Straub, Butcher, et al., 2014; Straub et al., 2012). While such

posttranslational regulation of HBA in ECs is well described, the transcriptional

regulation of endothelial HBA remains unknown. Hence, elucidating the

molecular mechanisms of how endothelial HBA is regulated may provide a

deeper understanding of vascular homeostasis and how alterations in this

process lead to cardiovascular disease.

Studies over the past decade from our group and others have identified

kruppel-like factors (KLFs) termed KLF2 and KLF4 as essential regulators of

endothelial homeostasis, endothelial identity, and vascular integrity (Atkins & Jain,

2007; Jain et al., 2014; Sangwung et al., 2017). KLFs are members of the zinc-

finger family of DNA-binding transcription factors of which a total of 18

mammalian family members have been identified to date (McConnell & Yang,

2010). In ECs, KLF2 and KLF4 are induced by laminar flow and regulate the

expression of many flow-dependent endothelial gene products (e.g. eNOS) that

are critical for endothelial function (Hamik et al., 2007; Lin et al., 2005;

SenBanerjee et al., 2004; Zhou et al., 2012). Furthermore, both factors have

been shown to inhibit EC pro-inflammatory activation by virtue of their ability to

76 inhibit key inflammatory mediators such as NF-κB. Here we demonstrate that

KLF2 and KLF4 serve as potent regulators of HBA expression in ECs.

3.3 MATERIALS AND METHODS

Animals. All animals used were C57BL6/J mouse strain. Endothelial-specific Klf2 and/or Klf4 knockout mice were generated by breeding floxed mice (Klf2/Klf4) with tamoxifen-inducible Cdh5(PAC)-CreERT2 mice (originally from R. Adams,

University of Münster, Münster, Germany). Cdh5(PAC)-CreERT2 mice were used as control. Eight to ten-week-old male mice were intraperitoneally injected with tamoxifen (2mg/25g) to trigger endothelial-specific gene deletion. All animal protocols carried out in this study were approved by the Case Western Reserve

University Institutional Animal Care and Use Committee.

Cell culture and a vascular cell co-culture model. Human umbilical vein endothelial cells (HUVECs), human coronary artery endothelial cells (HCAECs) and human coronary artery smooth muscle cells (HCASMCs) were cultured in

EBM-2 supplemented with an EGM-2 bullet kit, EBM-2 supplemented with an

EGM-2MV bullet kit, and SmGM-2 supplemented with SmGM-2 bullet kit (Lonza,

Basel, Switzerland), respectively. Human umbilical vein smooth muscle cells

(HUVSMCs) and bovine aortic endothelial cells (BAECs) were obtained from Cell

Applications, Inc. (San Diego, CA, USA) and cultured in growth medium according to the manufacturer’s instruction. All cells were cultured at 37oC and

5% CO2. For a vascular cell co-culture model, ECs were co-cultured with

77 vascular SMCs on a Transwell (24 mm) (Corning Inc., NY, USA) to allow a formation of MEJs as previously described (Heberlein et al., 2010; Straub et al.,

2012). SMCs were first seeded on the lower part of the transwell insert and cultured for 24 hours. In the same transwell insert, ECs were cultured on the upper part of the insert and grown for additional 24 hours. Following this initial period of growth, KLF2 or KLF4 expression was manipulated by adenoviral infection.

Adenoviral infection and plasmid DNA transfection. For overexpression in human

ECs, cells at passage 3–9 were infected with control or adenovirus (10 multiplicity of infection or less) carrying mouse Klf2 or human KLF4 genes. For transient plasmid DNA expression, BAECs (passage 3–9) were transfected with

50 ng of control vector, KLF2 or KLF4 plasmid along with 200 ng of a luciferase construct driven by control, wild type or mutant HBA promoter by using FuGENE

6 transfection reagent (Roche Applied Science, Indianapolis, IN, USA) according to the manufacturer’s instruction with minor modifications. Experiments were performed 48 hours after infection or transfection.

EC isolation. ECs were isolated from the heart of mice using a standard technique as previously described with a minor modification (Y. C. Lim et al.,

2003). Briefly, the hearts were washed in cold PBS, minced with blades, digested in PBS containing 1%BSA, collagenase type I, 1mM CaCl2, and 1 mM MgCl2 at

37oC for 45 minutes. ECs were purified using Dynabeads-conjugated with anti-

CD31antibody (catalog 553370, BD Biosciences, San Jose, CA, USA).

RNA isolation and Real time-PCR. Total RNA from cells was isolated by using

High Pure RNA Isolation Kit and reverse transcribed into cDNA using iScript

cDNA Synthesis Kit (Bio-Rad, Hercules, CA, USA). Quantitative real-time PCR

(qPCR) was carried out using TaqMan Master Mix (Applied Biosystems Inc.,

Foster City, CA, USA) and the reactions were performed using the StepOnePlus

Real-Time PCR System (Applied Biosystems Inc.). Endogenous glyceraldehyde

3-phosphate dehydrogenase (GAPDH) was used as reference (normalizer) for the gene expression. Primers are listed in Table 3.1. Relative mRNA expression levels (fold changes) between groups were calculated using the delta-delta Ct method.

Table 3.1 Primer sequences for qPCR (TaqMan) analysis.

Gene Forward (5’ to 3’) Reverse (5’ to 3’) Probe number KLF2 CATCTGAAGGCGCATCTG CGTGTGCTTTCGGTAGTGG 11 KLF4 GGGAGAAGACACTGCGTCA GGAAGCACTGGGGGAAGT 52 NOS3 GCATCCCTACTCCCACCAG TTCTTCACACGAGGGAACTTG 23 HBA CCGACAAGACCAACGTCAA GGGGAAGGACAGGAACATC 81 GAPDH TGTTCGTCATGGGTGTGAAC GGTGCTAAGCAGTTGGTGGT 90 Klf2 CTAAAGGCGCATCTGCGTA TAGTGGCGGGTAAGCTCGT 48 Klf4 CGGGAAGGGAGAAGACACT GAGTTCCTCACGCCAACG 62 Gapdh TGTCCGTCGTGGATCTGAC CCTGCTTCACCACCTTCTTG 80 Hba CTCTGGGGAAGACAAAAGCA GTGGGGAAGCTAGCAAACAT 27

KLF2: Human kruppel-like factor 2; KLF4: Human kruppel-like factor 4; NOS3: Human nitric oxide synthase 3; HBA: Human hemoglobin subunit alpha; GAPDH: Human Glyceraldehyde 3- phosphate dehydrogenase; Klf2: Mouse kruppel-like factor 2; Klf4: Mouse kruppel-like factor 4; Gadph: Mouse glyceraldehyde 3-phosphate dehydrogenase; Hba: Mouse hemoglobin subunit alpha; qPCR: Quantitative real-time PCR. Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like Factors. Vascular Medicine, published ahead of print. Copyright © 2017 by the Authors. Reprinted by permission of SAGE Publications, Ltd.

Western blot. EC, MEJ, and SMC were harvested and lysed with ice-cold RIPA lysis buffer (Thermo Fisher Scientific, Waltham, MA, USA) containing protease and phosphatase inhibitor cocktail. Protein concentration was measured by using

BCA Protein Assay Reagent Kit (Thermo Fisher Scientific). Protein lysates were subjected to protein electrophoresis using 4-12% Bis-Tris gels (Life Technologies,

Carlsbad, CA) and transferred to nitrocellulose membrane. The membrane was incubated overnight at 4oC with primary antibodies: goat anti-hemoglobin α (V-

13) (catalog sc-31109, Santa Cruz Biotechnology, Inc., Santa Cruz, CA), rabbit

anti-GAPDH (catalog G9545, Sigma-Aldrich, St. Louis, MO), mouse anti-

eNOS/NOS Type III (catalog 610297, BD Biosciences, Franklin Lakes, NJ),

rabbit anti-KLF2 (catalog 09-820, EMD Millipore, Billerica, MA), and goat anti-

KLF4 (catalog AF3640, R&D Systems, Inc., Minneapolis, MN, USA). Secondary

antibodies (LI-COR Biosciences, Lincoln, NE, USA) are donkey anti-goat IgG,

IRDye 800CW conjugated (catalog 926-32214), donkey anti-rabbit IgG, IRDye

680RD conjugated (catalog P/N 925-68073), goat anti-mouse IgG, IRDye 680RD

conjugated (catalog 926-68070), and goat anti-rabbit IgG, IRDye 800CW

conjugated (catalog 926-32211). Protein was detected using Odyssey imaging

systems (LI-COR Biosciences). Protein quantification was performed using Bio-

Rad Quantity One software.

Protein Digestion. Gel pieces cut from SDS-polyacrylamide gels were first

destained with 50% acetonitrile in 100 mM ammonium bicarbonate, and 100%

acetonitrile. Then, the protein was reduced by 20 mM DTT at room temperature

80 for 60 min followed by the alkylation using 50 mM idoacetamide for 30 min in the dark. The reaction reagents were removed and the gel pieces were washed with

100 mM ammonium bicarbonate and dehydrated in acetonitrile. Sequencing grade modified trypsin (Promega, Madison, WI) in 50 mM ammonium bicarbonate was added to the dried gel pieces and incubated at 37°C for overnight. Proteolytic peptides extracted from the gel with 50% acetonitrile in 5% formic acid were then dried and re-suspended in 0.1% formic acid.

Mass Spectrometry. Liquid chromatography-tandem mass spectrometry analysis of the resulting peptides was performed on Orbitrap Elite Hybrid Mass

Spectrometer (ThermoFisher Scientific, Waltham, MA) coupled with a Waters nanoAcquity UPLC system (Waters, Taunton, MA). The spectra were acquired in the positive ionization mode by data-dependent methods consisting of a full MS scan in high-mass accuracy FT-MS mode at 120 000 resolution, and MS/MS on the twenty most abundant precursor ions in CID mode with the normalized collision energy of 35%. Mascot Daemon (version 2.4.0; Matrix Science, London,

UK) was used to identify the peptides, and the data were searched against

SwissProt human database. Mass tolerance was set at 10 ppm for precursor ions and 0.8 Da for product ions. Carbamidomethylation of cysteines was set as a fixed modification, and oxidation of methionine as a variable modification. The significance threshold p value was set to < 0.05. Proteins hits with at least two unique peptides at Mascot score > 20 were considered to be identified.

Luciferase assay. Expression plasmid containing the luciferase reporter driven by a 953 bp (counting from the transcription start site) fragment of the human hemoglobin alpha gene (HBA) promoter or a control empty vector (SwitchGear

Genomics, Carlsbad, CA, USA) was co-transfected with KLF2, KLF4 or both expression plasmids into BAECs. The cells were grown for 48 hours to allow sufficient expression of luciferase protein, and luciferase activity

(bioluminescence) was measured on a luminometer according to the manufacturer’s instruction with a minor modification. Background signal was determined from wells without cells. The luciferase activity was normalized with protein concentration and values were presented in % Relative Light Units (RLU).

Site-directed mutagenesis. KLF binding sites in the HBA promoter were generated by QuikChange Lightning Site-Directed Mutagenesis Kit (Agilent

Technologies, Santa Clara, CA) using the luciferase reporter driven by a 953 bp fragment of the HBA promoter plasmid as a template. Forward and reverse site- directed mutagenesis primers containing the desired mutation flanked by unmodified nucleotide sequences, were designed to anneal on opposite strands of DNA (Table 3.2).

Table 3.2 The HBA promoter mutagenic oligonucleotide primers (5’-3’)

HBA-KLF site 1 Sense CCGGGCGTGCCCCCGCGCCAAAAGCATAAACCCTGGCGC Antisense GCGCCAGGGTTTATGCTTTTGGCGCGGGGGCACGCCCGG HBA-KLF site 2 Sense CGGTCCAGGCCGCGCGACGGGCTCCGCGCCAG Antisense CTGGCGCGGAGCCCGTCGCGCGGCCTGGACCG HBA-KLF site 3 Sense CCGCGCAGGCCCCGGACGGGACTCCCCTGCG Antisense CGCAGGGGAGTCCCGTCCGGGGCCTGCGCGG HBA-KLF site 4 Sense CGTCCTGGCCCCCGGACCGCGTGCACCCCCAG Antisense CTGGGGGTGCACGCGGTCCGGGGGCCAGGACG

HBA: Hemoglobin subunit alpha; KLF: Kruppel-like factor. Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like

Reprinted by permission of SAGE Publications, Ltd.

Chromatin immunoprecipitation. DNA-protein complexes were isolated from

HUVECs overexpressing KLF4 after fixation with formaldehyde. Antibody against

KLF4 (catalog AF3640, R&D Systems Inc., Minneapolis, MN, USA) was used to capture KLF4-DNA protein complexes. These immune-complexes were precipitated by protein A/G magnetic beads. The immunoprecipitated DNA was reverse crosslinked and analyzed by qPCR for enrichment of KLF4 on the promoter region of HBA. Target gene enrichment was determined by qPCR

(SYBR Green). The qPCR primer sequences for the promoter region of HBA are

5’- GGGTGGAGGGTGGAGACG -3’ (Forward) and 5’-

CGCCAGGGTTTATGCTTGGG -3’ (Reverse). Five percent of DNA input was used as the normalization. Relative levels (fold changes) between groups were calculated using the delta-delta Ct method.

Statistics. All data are presented as mean ± SEM. Statistical analyses were done using 2-tailed Student’s t test and One-way ANOVA with Bonferroni's post hoc test to analyze the difference between two groups and among the groups, respectively. A p value of 0.05 or less was considered significant.

3.4 RESULTS

KLF2 and KLF4 induce HBA expression in the endothelium.

Given the importance of endothelial KLF2/KLF4 in maintaining endothelial homeostasis, endothelial identity, and vascular integrity, we first investigated whether KLF2/KLF4 mediated endothelial HBA expression. Adenoviral overexpression of KLF2 or KLF4 strongly induced HBA mRNA in human coronary artery ECs (HCAECs) (Figure 3.1). An increase in Klf2 and Klf4 mRNA expression were also confirmed in HCAECs (Figure 3.2, A and B). As a positive control (Sangwung et al., 2017), mRNA expression of NOS3 was examined and found to be significantly enhanced (Figure 3.1A). Further, the effect of KLF2 or

KLF4 deficiency on Hba mRNA expression was examined in ECs isolated from adult mice with tamoxifen-inducible endothelial Klf2 (EC-Klf2-KO), Klf4 (EC-Klf4-

KO) or both Klf2 and Klf4 (EC-DKO) knockout. While Hba levels were not significantly affected by loss of either KLF2 or KLF4, loss of both factors resulted in a marked decrease of Hba expression levels compared to CRE control mice

(Figure 3.1B).

Next, we confirmed the induction of HBA as well as eNOS proteins by

KLF2 or KLF4 in ECs which were either cultured alone (Figure 3.3, A and B) or co-cultured with vascular smooth muscle cells (Figure 3.4, A and B, and Figure

3.5, A-C). For the latter study, human primary coronary artery ECs (HCAECs) were co-cultured with coronary artery vascular smooth muscle cells (HCASMCs) on a transwell insert to allow a formation of the MEJ (Straub et al., 2012). These cell types were chosen as they have previously been shown to allow for optimal

MEJ formation. Adenovirus carrying KLF2, KLF4 or empty vector control was then introduced into the ECs on the top of the transwell insert. EC, MEJ, and

SMC proteins were collected according to a previously reported method

(Heberlein et al., 2010; Straub et al., 2012). A protein at the molecular weight of approximately 32 kDa was recognized by anti-hemoglobin alpha antibody in the

EC and MEJ that were infected with adenovirus carrying KLF2 or KLF4, but not empty vector (Figure 3.4, A and B, and Figure 3.5, A-C). The size of this protein band is close to that of dimeric hemoglobin and mass spectrometry analyses confirmed the presence of HBA (Table 3.3). In addition, expression of hemoglobin beta protein was not detectable in the EC and MEJ (data not shown), suggesting a specificity of endothelial hemoglobin alpha expression by

KLF2/KLF4. Collectively, these observations support a role for KLF2 and KLF4 to promote HBA expression in the EC and the MEJ.

Figure 3.1 Hemoglobin alpha mRNA expression in the EC by KLF2/KLF4.

(A) Overexpression of either KLF2 or KLF4 induced an expression of hemoglobin subunit alpha

(HBA) and endothelial nitric oxide synthase (NOS3) mRNA in human coronary artery ECs. A representative data of three independent experiments is shown. (B) Cardiac microvascular ECs were isolated from EC-specific deletion of Klf2 (EC-Klf2-KO), Klf4 (EC-Klf4-KO), both Klf2 and

Klf4 (EC-DKO), and CRE control (n=8-9 per genotype, each n was pooled form 2 mice). EC-DKO

showed attenuated mRNA expression levels of Hba. Ad-EV: Control (empty) adenovirus; Ad-

KLF2: Adenovirus carrying Kruppel-like factor 2; Ad-KLF4: Adenovirus carrying Kruppel-like factor

4; CRE: Cdh5(PAC)-CreERT2; EC: Endothelial cell. Data are presented as mean ± SEM values.

* *: P < 0.01. NS: Not significant.

Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like

Reprinted by permission of SAGE Publications, Ltd.

Figure 3.2 Expression of KLF2 or KLF4 mRNA in ECs.

(A-B) Human coronary artery endothelial cells were infected with adenovirus carrying empty vector (Ad-EV), mouse KLF2 (Ad-KLF2), or human KLF4 (Ad-KLF4) and cultured on the petri dishes for 48 hours. Expression of mRNA was measured by TagMan real-time PCR. A representative data of three independent experiments is shown. (C-D) Cardiac microvascular

ECs were isolated from EC-specific deletion of Klf2 (EC-Klf2-KO), Klf4 (EC-Klf4-KO), both Klf2

and Klf4 (EC-DKO), and CRE control (n=8-9 per genotype, each n was pooled form 2 mice).

CRE: Cdh5(PAC)-CreERT2; EC: Endothelial cell; KLF: Kruppel-like factor. Data are presented as

mean ± SEM values. * *: P < 0.01. NS: Not significant.

Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like

Reprinted by permission of SAGE Publications, Ltd.

Figure 3.3 KLF2/KLF4-mediated hemoglobin alpha protein expression in the EC cultured alone on a petri dish.

(A) Representative western blot analysis of HBA, eNOS, KLF2, KLF4, and GAPDH when human coronary artery endothelial cells were cultured alone (n=3 independent experiments). The first

band in HBA panel was likely non-specific due to its higher molecular weight. (B) Quantification of

western blot analysis (normalized to GAPDH). Ad-EV: Control (empty) adenovirus; Ad-KLF2:

Adenovirus carrying Kruppel-like factor 2; Ad-KLF4: Adenovirus carrying Kruppel-like factor 4;

HBA: Hemoglobin subunit alpha; eNOS: Endothelial nitric oxide synthase; GAPDH:

Glyceraldehyde 3-phosphate dehydrogenase. Data are presented as mean ± SEM values. *: P <

0.05, * *: P < 0.01. NS: Not significant.

Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like

Reprinted by permission of SAGE Publications, Ltd.

Figure 3.4 Hemoglobin alpha protein expression in the EC by KLF2/KLF4.

(A) Schematic picture showing a transwell co-culture of HCAECs/HCASMCs. (B) Representative

western blot analysis obtained from HCAECs/HCASMCs co-culture where the endothelial cells

were overexpressed with either KLF2 or KLF4 (n=3-4 independent experiments). Ad-EV: Control

(empty) adenovirus; Ad-KLF2: Adenovirus carrying Kruppel-like factor 2; Ad-KLF4: Adenovirus

carrying Kruppel-like factor 4; HBA: Hemoglobin subunit alpha; eNOS: Endothelial nitric oxide synthase; GAPDH: Glyceraldehyde 3-phosphate dehydrogenase; HCAECs: Human coronary

artery endothelial cells; HCASMCs: Human coronary artery smooth muscle cells.

Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like

Reprinted by permission of SAGE Publications, Ltd.

Figure 3.5 KLF2/KLF4-mediated hemoglobin alpha protein expression in the EC and the

MEJ obtained from the co-cultured model.

(A-C) Quantification of western blot analysis (normalized to GAPDH) from HCAECs/HCASMCs co-culture where ECs were overexpressed with either KLF2 or KLF4 (n=3-4 independent experiments). Ad-EV: Control (empty) adenovirus; Ad-KLF2: Adenovirus carrying Kruppel-like factor 2; Ad-KLF4: Adenovirus carrying Kruppel-like factor 4; HBA: Hemoglobin subunit alpha; eNOS: Endothelial nitric oxide synthase; GAPDH: Glyceraldehyde 3-phosphate dehydrogenase;

HCAECs: Human coronary artery endothelial cell; HCASMCs: Human coronary artery smooth muscle cells; EC: Endothelial cells; MEJ: Myoendothelial junction; SMC: Smooth muscle cell.

Data are presented as mean ± SEM values. *: P < 0.05, * *: P < 0.01. NS: Not significant.

Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like

Reprinted by permission of SAGE Publications, Ltd.

Table 3.3 Identified peptides of HBA protein in the MEJ of KLF4 overexpressing HCAECs using LC-MS/MS

Score Number Peptide Mr (expt) Mr (calc) ppm Monomera Dimera 2-12 VLSPADKTNVK 1170.6596 1170.6608 -1 45 13-32 AAWGKVGAHAGEYGAEALER 2041.9962 2041.9969 0 36 18-32 VGAHAGEYGAEALER 1528.7267 1528.7270 0 59 50

18-41 VGAHAGEYGAEALERMFLSFPTTK 2597.2592 2597.2584 0 67 33-41 MFLSFPTTK 1086.5415 1086.5420 0 40 40 42-57 TYFPHFDLSHGSAQVK 1832.8844 1832.8846 0 68 63-91 VADALTNAVAHVDDMPNALSALSDLHAHK 3011.4750 3011.4771 -1 27 92-100 LRVDPVNFK 1086.6184 1086.6186 0 46 34 94-100 VDPVNFK 817.4324 817.4334 -1 27 24

a The monomer (16 kDa) and the dimer (32 kDa)-like bands on a gel. HBA: Hemoglobin subunit alpha; MEJ: Myoendothelial junction; KLF4: Kruppel-like factor 4; HCAECs: Human coronary artery endothelial cells; Mr (expt): Experimental molecular weight of each peptide; Mr (calc): Calculated molecular weight of each peptide; ppm: parts per million.

Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like Factors. Vascular Medicine, published ahead of

92 The transactivation of the human hemoglobin alpha gene promoter in ECs by KLF2/KLF4.

To determine if KLF2/KLF4 could directly regulate the HBA promoter, we performed a luciferase reporter assay in bovine aortic endothelial cells (BAECs) which have high transfection efficiency when compared to human ECs. As shown in Figure 3.6, co-transfection of either KLF2 or KLF4 or both KLFs significantly increased the HBA promoter activity. Previous reports identified KLFs as DNA- binding transcription factors that bind to GC-, GT-, and CACCC-box motifs in gene promoters and other regulatory elements in order to mediate gene expression (Hamik et al., 2007; Jain et al., 2014; Lin et al., 2005; SenBanerjee et al., 2004; Zhou et al., 2012). We therefore aimed to pin-point the KLF recognition sites (GC-, GT-, and CACCC) of the HBA promoter that are critical for

KLF2/KLF4 binding to augment the transactivation of the promoter in the endothelium. We first focused on four previously reported (Funnell et al., 2014;

Marini et al., 2010) KLF binding sites of the HBA promoter (Figure 3.7A, site 1, -

33 to -41 bp; site 2, -86 to -94 bp; site 3: -115 to -123 bp; site 4, -172 to -180 bp), which were mutated to prevent KLF binding. In our study, the luciferase reporter was driven by a 953 bp fragment of the HBA promoter plasmid. We found that the promoter induction by KLF2 (Figure 3.7B) or KLF4 (Figure 3.7C) was significantly diminished with mutation of all four KLF binding sites (Mutant 1234).

Next, to pinpoint the site (s) that is critical for KLF-mediated transactivation, we generated single and multiple mutations within the four KLF binding sites on the HBA promoter. In the presence of KLF2 (Figure 3.8A) or KLF4 (Figure 3.8B),

93 the HBA promoter mutants lacking at least two of the three sites (sites 2, 3, and

4) had the lowest luciferase activity when compared to other mutants, suggesting that three (sites 2, 3 and 4) out of the four KLF binding sites on the HBA promoter are critical for KLF2 and KLF4 to transactivate the expression of endothelial HBA. Interestingly, a co-expression of KLF4 with the luciferase reporter driven by the HBA promoter mutant 14 led to a significant increase in the luciferase signals when compared to a co-expression of KLF4 with the reporter driven by the wild-type HBA promoter (Figure 3.8B). These observations suggest that the binding sites 1 and 4 together are not required for KLF4 ̶ but suppress the ability of KLF4 ̶ to activate the endothelial HBA expression. Altogether,

KLF2/KLF4 promote HBA expression in the endothelium by transactivating the

HBA promoter.

Figure 3.6 KLF2/KLF4 mediate a transactivation of the HBA promoter.

Luciferase reporter assay in bovine aortic endothelial cells (n=3-4 independent experiments). The

HBA promoter activity was significantly enhanced in the cells that were co-transfected the HBA promoter with KLF2 and/or KLF4. KLF2: Kruppel-like factor 2 plasmid; KLF4: Kruppel-like factor 4 plasmid; HBA: Hemoglobin subunit alpha. Data are presented as mean ± SEM values.

* *: P < 0.01.

Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like

Reprinted by permission of SAGE Publications, Ltd.

Figure 3.7 KLF2/KLF4 fail to mediate a transactivation of the mutant HBA promoter.

(A) Schematic drawing of the luciferase reporter driven by a fragment (953 bp, counting from the transcription start site) the human HBA promoter. Four positions of consensus recognition sites for KLFs are indicated by white rectangles. (B-C) The mutated HBA promoter lacking four KLF binding sites was co-expressed with either KLF2 (B) or KLF4 (C) (n=3-4 independent

experiments). Gray box denotes mutated KLF binding site on the HBA promoter. KLF2: Kruppel-

like factor 2 plasmid; KLF4: Kruppel-like factor 4 plasmid; HBA: Hemoglobin subunit alpha; Luc:

Luciferase reporter. Data are presented as mean ± SEM values. * *: P < 0.01. NS: Not significant.

Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like

Reprinted by permission of SAGE Publications, Ltd.

Figure 3.8 Critical KLF-binding sites on the HBA promoter.

Luciferase activity of wild type or mutated HBA promotor constructs co-expressed with either

KLF2 (A) or KLF4 (B) were observed in bovine aortic endothelial cells (n=3-4 independent

experiments). Gray box represents mutated KLF binding site on the HBA promoter. KLF2:

Kruppel-like factor 2 plasmid; KLF4: Kruppel-like factor 4 plasmid; HBA: Hemoglobin subunit

alpha; Luc: Luciferase reporter. Data are presented as mean ± SEM values. *: P < 0.05,

* *: P < 0.01 when compared with a co-expression of Wild-Type HBA promoter with KLF2 or

KLF4. NS: Not significant.

Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like

Reprinted by permission of SAGE Publications, Ltd.

Direct binding of KLF4 to the endogenous hemoglobin alpha promoter in the EC.

Next, we assessed whether KLF4 directly bound to the hemoglobin alpha promoter in the endothelium by chromatin immunoprecipitation (ChIP). We focused on KLF4 in ChIP studies because of the availability of ChIP validated antibody. We found a significant enrichment of KLF4 (Figure 3.9A) on the promoter region of HBA in HUVECs overexpressing KLF4. No enrichment was observed with control IgG. The specificity and efficiency of qPCR primers were confirmed by agarose gel electrophoresis showing single amplicon of 190 bp

(Figure 3.9B). These results indicated that KLF4 directly bound to the HBA promoter in the ECs.

Figure 3.9 Direct binding of KLF4 on the HBA promoter in the EC. (A) Enrichment of KLF4 on the HBA promoter was observed by chromatin Immunoprecipitation

(ChIP) coupled with quantitative PCR in human umbilical vein endothelial cells that were infected with KLF4 adenovirus for 48 hours (n=3 independent experiments). (B) Representative image of

agarose gel electrophoresis showed the efficiency and specificity of the qPCR HBA primer pair

which produced single amplicon of 190 bp (n=3 independent experiments). Ad-EV: Control

(empty) adenovirus; Ad-KLF4: Adenovirus carrying Kruppel-like factor 4; HBA: Hemoglobin subunit alpha. Data are presented as mean ± SEM values. * *: P < 0.01. NS: Not significant.

Sangwung, P., et al. Regulation of Endothelial Hemoglobin Alpha Expression by Kruppel-Like

Reprinted by permission of SAGE Publications, Ltd.

3.5 DISCUSSION

The fact that hemoglobins can exist outside of red blood cells has been

appreciated for many years with recent studies suggesting that their functions

extend beyond acting as oxygen carriers to include regulating nitric oxide

scavenging and blood pressure. An important gap in our understanding centers

on how the expression of hemoglobin is controlled in non-erythroid cell types,

including endothelial cells. Here we show that KLF2 and/or KLF4 regulate

endothelial HBA expression by binding directly to and transactivating the HBA

promoter.

Transcriptional regulation of HBA in ECs by KLFs is most likely similar to

mechanisms of regulating in erythroid cells. KLFs have been shown to be

responsible for hemoglobin expression in erythroid and non-erythroid cells. In

erythroid cells, KLF4 is shown to recognize a CACCC motif in the promoter

region of HBA and gamma-globin (HBG) genes (Kalra et al., 2011; Marini et al.,

2010). Specifically, Marini et al. report four KLF binding sites on the HBA promoter that are important for KLF4-mediated hemoglobin alpha expression in

erythroid cells (Marini et al., 2010). Previous study using microarray analysis in

ECs shows that HBA is the top 5 genes regulated by KLF2 (Parmar et al., 2006).

Our study showed for the first time that both KLF2 and KLF4 transactivated HBA expression in the EC, and that three of the four KLF binding sites (sites 2, 3, and

4) are critical for KLF2/KLF4 mediated HBA expression. Besides KLF2 and KLF4,

KLF3 has also been shown to promote the adult alpha-globin genes in erythroid

cells; however it has been shown to suppress HBA in non-erythroid cells (Funnell

100 et al., 2014). KLF1 and KLF2 promote beta-globin gene expression in embryos

(Alhashem et al., 2011). This raises the possibility that a member of the KLF family activates expression of hemoglobin in other non-erythroid cells in addition to ECs.

Free alpha hemoglobin molecule is unstable and precipitates (Abdullah

UYH, 2104). In erythroid cells, alpha and beta hemoglobin subunits are held together by noncovalent bonds (Bunn, 1987). In our study, we observed a protein of 32 kDa in the EC and the MEJ which represents dimeric HBA protein. A dimerization of HBA observed in our study may be caused by aggregation of excess amount of free hemoglobin proteins in the cells. It will be interesting to further explore how dimeric hemoglobin is assembled and functional importance of hemoglobin dimer. Furthermore, we detected monomeric HBA form in EC,

MEJ, and SMC, which is similar to observations by Straub et al (Straub et al.,

2012). However, significant changes in expression levels of dimeric, but not monomeric HBA, were observed. We speculate that, when HBA protein levels reach a high enough concentration in ECs, it may spontaneously form dimers. An additional explanation is that KLFs induce proteins that confer a post- translational modification that facilitates dimerization that is resistant to denaturation. This also explains why the monomeric HBA bands appeared similar between EV and KLF-infected samples as the extra amount of HBA has formed dimers.

KLF2 and KLF4 promote eNOS expression which subsequently leads to increased nitric oxide (NO) production and vasodilation (Atkins & Jain, 2007).

101

Here we show that KLF2 and KFL4 induce endothelial HBA expression. Previous reports demonstrate that endothelial HBA can capture NO and thereby prevent diffusion of NO from ECs to SMCs, promoting vasoconstriction (Straub, Butcher, et al., 2014). Thus, by virtue of their ability to increase eNOS and HBA, KLFs may form an autoregulatory loop to control vasoreactivity and key physiologic functions such as blood pressure and tissue perfusion. The results of this study provide a better understanding of transcriptional regulation of endothelial HBA, a fine-tuner of NO diffusion to SMCs that contributes to maintenance of vascular homeostasis. Hence, our findings have the potential to provide the foundation for developing novel therapies aimed at ameliorating cardiovascular diseases.

3.6 CONCLUSION

Collectively, our work provides novel information regarding transcriptional regulation of endothelial expression of hemoglobin with implications for vascular function and homeostasis.

3.7 ACKNOWLEDGMENT

We thank Dr. Alfred Hausladen for providing technical advice and troubleshooting experiments. This work was supported by NIH Grants

R01HL110630-01, R01HL112486, R01HL086548, R01HL119195 (to MKJ),

R00HL-11290402, R01HL-133864, R01HL-128304 (to ACS), by AHA grants

12SDG12050558 (to YL), 12SDG12070077 (to XL) and 16GRNT27250146 (to

ACS).

102

CHAPTER 4

DISCUSSION AND SUMMARY STATEMENTS OF THE THESIS PROJECTS

4.1 DISCUSSION

Vasculature arose around 600 million years ago and is maintained to distribute blood, nutrients, and oxygen throughout the body and remove waste products

(Monahan-Earley et al., 2013). Nature also created a monolayer of vascular endothelial cells with multi-tasking ability to actively maintain integrity and functions of the vasculature (Aird, 2007a). Based on multiple cell-based studies,

KLF2 and KLF4 has been appreciated as key regulators of endothelial cell function (Atkins & Jain, 2007; Jain et al., 2014), but little is known about the in vivo role of both factors in endothelial biology. Here, we unequivocally demonstrated fundamental functions of both KLF2 and KLF4 in adult mice and provided unbiased genome-wide data on how endothelial transcriptional landscape is controlled by KLF2 and KLF4.

We revealed a profound effect of both KLF2 and KLF4 on animal survival.

Deficiency of both KLF2 and KLF4 in the endothelium for 6 days resulted in death of adult mice, and none of these animals lived beyond day 9 after tamoxifen administration. The presence of a single allele of either KLF2 or KLF4 was enough to sustain the life of animal. This addresses an unknown question in the field of endothelial biology regarding the functional redundancy of KLF2 and

KLF4 in the endothelium. Our data suggest that adult mice cannot afford to lose both KLF2 and KLF4 in the endothelium and these factors backup each other.

103

We observed that while the expression of KLF2 was declined, KLF4 expression tended to be increased, and vice versa. Accordingly, one factor compensates for the other as an effective backup system to protect against endothelial dysfunction. The backup system with the functional redundancy of transcription factors is also observed in other cells ̶ snail2/snail3 transcription factors in hematopoietic cells and myf5/myogenin in myotomal cells (Pioli & Weis, 2014; Y.

Wang, Schnegelsberg, Dausman, & Jaenisch, 1996). Furthermore, endothelial specific-deletion of both KLF2 and KLF4 in adult mice led to heart failure, DIC- like condition, and stroke secondary to barrier disruption and dysregulation of coagulation system. These indicate that loss of both KLF2 and KLF4 immediately impairs EC function, as evidenced by a defect in both barrier function and coagulation system, which leads to death. Without both KLF2 and KLF4, multiple nodes of gene expression involved in hemostasis, inflammation, apoptosis, cell cycle and other important pathways were altered. This finding lands KLF2 and

KLF4 as master regulators of endothelial gene expression and function.

One of the well-known functions of the endothelium is to produce NO (Galley &

Webster, 2004). Diffusion of NO to SMCs underneath ECs leads to vasodilation.

Studies in the past few years claimed NO diffusion to SMC through the MEJ

(Straub, Butcher, et al., 2014; Straub et al., 2012). Hemoglobin alpha (HBA) localized at the MEJ acts as a gatekeeper blocking diffusion of NO to SMC.

While function of HBA is well characterized in the EC, its transcriptional regulation is unknown. We discovered that KLF2 and/or KLF4 transcriptionally

104 regulated the expression of endothelial HBA by directly binding to and transactivating its promoter region. We detected significant changes in expression levels of dimeric, but not monomeric HBA. We speculate that a dimerization of HBA may be caused by aggregation of excess amount of free

HBA proteins when HBA protein levels reach a high enough concentration in ECs.

Peroxynitrite, a strong oxidant formed intravascularly in vivo, is the product of the reaction between NO and superoxide. Romero et al detected a dimerization of hemoglobin chains after peroxynitrite treatment and revealed the formation of two types of cross-linked species between hemoglobin subunits ̶ disulfide

(mercaptoethanol reducible) and dityrosine (mercaptoethanol non-reducible) bridges (Romero et al., 2003). The non-reducible dimerization of HBA observed in our study may be formed via the dityrosine bridges which are catalyzed by peroxynitrite. KLFs may also induce proteins that confer a post-translational modification that facilitates dimerization. Induction of eNOS by KLF2/4 leads to production of NO, a potent vasodilator. To fine-tune balance between vasodilation and vasoconstriction, a body has its own way to counter balance NO production. Although the functions of KLF2/4-mediated endothelial HBA were not investigated in our study, it may fine-tune NO diffusion to SMC, as a mechanism to counterbalance increased NO production.

105

4.2 SUMMARY STATEMENTS OF THE THESIS PROJECTS

In conclusion, these findings certainly position endothelial KLF2 and KLF4 on the top of the transcription hierarchy as critical mediators of the endothelial function and vascular hemostasis. Our studies provide adequate grounds to consider both

KLF2 and KLF4 as future therapeutic targets for treatment of cardiovascular and other diseases that are related with endothelial dysfunction.

106

CHAPTER 5

FUTURE DIRECTIONS

In the absence of either KLF2 or KLF4, adult animals do not show any sign of sickness. Moreover, as long as one copy of KLF2 or KLF4 gene remains, the animals appear normal. However, these single KLF mutant mice may not be healthy, as Zhou et al show that EC-Klf4-KO mice are more atherogenic (Zhou et al., 2012). Future studies may address whether EC-Klf2-KO or EC-Klf4-KO are susceptible to other stress conditions such as exercise or transverse aortic constriction (TAC). It is also interesting to investigate whether or not the EC-DKO phenotype can be rescued by heparin, warfarin, and soluble thrombomodulin.

RNA sequencing and Bioinformatics studies were performed in microvascular

ECs. It is worth further identifying KLF2/4-dependent genes and pathways in medium-and/or large-sized blood vessels. Moreover, it is not known whether or not the interaction of ECs with other cells and plasma components instead of the endothelium standalone contributes to dysregulation of coagulation. Thus, RNA sequencing at day 3–5 in EC-DKO may be further assessed. Pro-inflammatory cytokines such asIL-1β and TNF-α suppress MEF2-mediated KLF2 expression

(Kumar et al., 2005). In contrast to KLF2, KLF4 expression is induced after cytokine stimulation, indicating differential expression in response to inflammatory stimuli (Hamik et al., 2007). In addition, both KLF2 and KLF4 have been identified as direct regulators of laminar flow-mediated endothelial gene products (Atkins & Jain, 2007; Jain et al., 2014). A comprehensive understanding of each factor’s role in regulating the endothelial gene expression program in

107 response to laminar flow or cytokines has not been enumerated: future studies using chromatin immunoprecipitation sequencing (ChIP-seq) will characterize this issue. It is also worth further investigating post-translational modification of

KLF as it is not fully characterized. Furthermore, statins promote expression of

KLF2 and KLF4 and are commonly used to reduce LDL cholesterol and treat vascular diseases. Alternative drugs or small molecules that specifically and potently mediate optimal levels of endothelial KLF2 and KLF4 expression may be further identified. In addition, EC-targeted nanoparticles containing soluble KLF2 and KLF4 may be developed.

We discovered that KLF2 and/or KLF4 bind directly to and transactivate the HBA promoter region. Future studies using DNase I digestion and high-throughput sequencing (DNase-seq) in ECs can be performed to map chromatin accessibility or the DNase I hypersensitive sites (DHSs) which indicate regions of an open chromatin state (Madrigal & Krajewski, 2012). Although beyond the scope of the current studies, we anticipate that future studies will need to evaluate the molecular mechanism of HBA dimerization and its function. Future studies will also address whether or not KLF2/KLF4-mediated HBA alters blood pressure in vivo. These studies will include mice with tamoxifen-inducible endothelial-specific deletion of Klf2, Klf4 or both, or mice with endothelial-specific constitutive Klf2 or Klf4 overexpressing.

108

BIBLIOGRAPHY

Abdullah UYH, A. A., Ibrahim HM, Alwi ZB, Baig AA, et al. (2104). The Free

Alpha-Hemoglobin: A Promising Biomarker for β-Thalassemia. Journal of

Molecular Biomarkers & Diagnosis, 5(5). doi: 10.4172/2155-9929.1000197

Aird, W. C. (2004). Endothelium as an organ system. Critical Care Medicine,

32(Supplement), S271-S279. doi: 10.1097/01.ccm.0000129669.21649.40

Aird, W. C. (2004). Endothelium as an organ system. Crit Care Med, 32(5 Suppl),

S271-279.

Aird, W. C. (2007a). Phenotypic heterogeneity of the endothelium: I. Structure,

function, and mechanisms. Circ Res, 100(2), 158-173. doi:

10.1161/01.RES.0000255691.76142.4a

Aird, W. C. (2007b). Phenotypic heterogeneity of the endothelium: II.

Representative vascular beds. Circ Res, 100(2), 174-190. doi:

10.1161/01.RES.0000255690.03436.ae

Alhashem, Y. N., Vinjamur, D. S., Basu, M., Klingmuller, U., Gaensler, K. M., &

Lloyd, J. A. (2011). Transcription factors KLF1 and KLF2 positively

regulate embryonic and fetal beta-globin genes through direct promoter

binding. J Biol Chem, 286(28), 24819-24827. doi:

10.1074/jbc.M111.247536

Anderson, K. P., Kern, C. B., Crable, S. C., & Lingrel, J. B. (1995). Isolation of a

gene encoding a functional zinc finger protein homologous to erythroid

109

Kruppel-like factor: identification of a new multigene family. Mol Cell Biol,

15(11), 5957-5965.

Atkins, G. B., & Jain, M. K. (2007). Role of Kruppel-like transcription factors in

endothelial biology. Circ Res, 100(12), 1686-1695. doi:

10.1161/01.RES.0000267856.00713.0a

Basu, P., Morris, P. E., Haar, J. L., Wani, M. A., Lingrel, J. B., Gaensler, K. M., &

Lloyd, J. A. (2005). KLF2 is essential for primitive erythropoiesis and

regulates the human and murine embryonic beta-like globin genes in vivo.

Blood, 106(7), 2566-2571. doi: 10.1182/blood-2005-02-0674

Baur, J. A., & Sinclair, D. A. (2006). Therapeutic potential of resveratrol: the in

vivo evidence. Nat Rev Drug Discov, 5(6), 493-506. doi: 10.1038/nrd2060

Benest, A. V., Kruse, K., Savant, S., Thomas, M., Laib, A. M., Loos, E. K., . . .

Augustin, H. G. (2013). Angiopoietin-2 is critical for cytokine-induced

vascular leakage. PLoS One, 8(8), e70459. doi:

10.1371/journal.pone.0070459

Bhattacharya, R., Senbanerjee, S., Lin, Z., Mir, S., Hamik, A., Wang, P., . . . Jain,

M. K. (2005). Inhibition of vascular permeability factor/vascular endothelial

growth factor-mediated angiogenesis by the Kruppel-like factor KLF2. J

Biol Chem, 280(32), 28848-28851. doi: 10.1074/jbc.C500200200

Boon, R. A., Fledderus, J. O., Volger, O. L., van Wanrooij, E. J., Pardali, E.,

Weesie, F., . . . Horrevoets, A. J. (2007). KLF2 suppresses TGF-beta

signaling in endothelium through induction of Smad7 and inhibition of AP-

110

1. Arterioscler Thromb Vasc Biol, 27(3), 532-539. doi:

10.1161/01.ATV.0000256466.65450.ce

Botella, L. M., Sanchez-Elsner, T., Sanz-Rodriguez, F., Kojima, S., Shimada, J.,

Guerrero-Esteo, M., . . . Bernabeu, C. (2002). Transcriptional activation of

endoglin and transforming growth factor-beta signaling components by

cooperative interaction between Sp1 and KLF6: their potential role in the

response to vascular injury. Blood, 100(12), 4001-4010. doi:

10.1182/blood.V100.12.4001

Bunn, H. F. (1987). Subunit assembly of hemoglobin: an important determinant

of hematologic phenotype. Blood, 69(1), 1-6.

Chiplunkar, A. R., Curtis, B. C., Eades, G. L., Kane, M. S., Fox, S. J., Haar, J. L.,

& Lloyd, J. A. (2013). The Kruppel-like factor 2 and Kruppel-like factor 4

genes interact to maintain endothelial integrity in mouse embryonic

vasculogenesis. BMC Dev Biol, 13, 40. doi: 10.1186/1471-213X-13-40

Cines, D. B., Pollak, E. S., Buck, C. A., Loscalzo, J., Zimmerman, G. A., McEver,

R. P., . . . Stern, D. M. (1998). Endothelial cells in physiology and in the

pathophysiology of vascular disorders. Blood, 91(10), 3527-3561.

Cowan, C. E., Kohler, E. E., Dugan, T. A., Mirza, M. K., Malik, A. B., & Wary, K.

K. (2010). Kruppel-like factor-4 transcriptionally regulates VE-cadherin

expression and endothelial barrier function. Circ Res, 107(8), 959-966.

doi: 10.1161/CIRCRESAHA.110.219592

Dahlback, B., & Villoutreix, B. O. (2005). The anticoagulant protein C pathway.

FEBS Lett, 579(15), 3310-3316. doi: 10.1016/j.febslet.2005.03.001

111

Das, A., Fernandez-Zapico, M. E., Cao, S., Yao, J., Fiorucci, S., Hebbel, R.

P., . . . Shah, V. H. (2006). Disruption of an SP2/KLF6 repression complex

by SHP is required for farnesoid X receptor-induced endothelial cell

migration. J Biol Chem, 281(51), 39105-39113. doi:

10.1074/jbc.M607720200

Das, H., Kumar, A., Lin, Z., Patino, W. D., Hwang, P. M., Feinberg, M. W., . . .

Jain, M. K. (2006). Kruppel-like factor 2 (KLF2) regulates proinflammatory

activation of monocytes. Proc Natl Acad Sci U S A, 103(17), 6653-6658.

doi: 10.1073/pnas.0508235103

Dejana, E. (2004). Endothelial cell-cell junctions: happy together. Nat Rev Mol

Cell Biol, 5(4), 261-270. doi: 10.1038/nrm1357

Dejana, E., Tournier-Lasserve, E., & Weinstein, B. M. (2009). The control of

vascular integrity by endothelial cell junctions: molecular basis and

pathological implications. Dev Cell, 16(2), 209-221. doi:

10.1016/j.devcel.2009.01.004

Dekker, R. J., Boon, R. A., Rondaij, M. G., Kragt, A., Volger, O. L., Elderkamp, Y.

W., . . . Horrevoets, A. J. (2006). KLF2 provokes a gene expression

pattern that establishes functional quiescent differentiation of the

endothelium. Blood, 107(11), 4354-4363. doi: 10.1182/blood-2005-08-

3465

Dekker, R. J., van Soest, S., Fontijn, R. D., Salamanca, S., de Groot, P. G.,

VanBavel, E., . . . Horrevoets, A. J. (2002). Prolonged fluid shear stress

induces a distinct set of endothelial cell genes, most specifically lung

112

Kruppel-like factor (KLF2). Blood, 100(5), 1689-1698. doi: 10.1182/blood-

2002-01-0046

Dekker, R. J., van Thienen, J. V., Rohlena, J., de Jager, S. C., Elderkamp, Y. W.,

Seppen, J., . . . Horrevoets, A. J. (2005). Endothelial KLF2 links local

arterial shear stress levels to the expression of vascular tone-regulating

genes. Am J Pathol, 167(2), 609-618. doi: 10.1016/S0002-

9440(10)63002-7

DePaola, N., Gimbrone, M. A., Jr., Davies, P. F., & Dewey, C. F., Jr. (1992).

Vascular endothelium responds to fluid shear stress gradients. Arterioscler

Thromb, 12(11), 1254-1257.

Esmon, C. T. (1995). Thrombomodulin as a model of molecular mechanisms that

modulate protease specificity and function at the vessel surface. FASEB J,

9(10), 946-955.

Fang, Y., & Davies, P. F. (2012). Site-specific microRNA-92a regulation of

Kruppel-like factors 4 and 2 in atherosusceptible endothelium. Arterioscler

Thromb Vasc Biol, 32(4), 979-987. doi: 10.1161/ATVBAHA.111.244053

Feletou, M. (2011) The Endothelium: Part 1: Multiple Functions of the Endothelial

Cells-Focus on Endothelium-Derived Vasoactive Mediators. San Rafael

(CA).

Feletou, M., & Vanhoutte, P. M. (2006). Endothelial dysfunction: a multifaceted

disorder (The Wiggers Award Lecture). Am J Physiol Heart Circ Physiol,

291(3), H985-1002. doi: 10.1152/ajpheart.00292.2006

113

Fledderus, J. O., van Thienen, J. V., Boon, R. A., Dekker, R. J., Rohlena, J.,

Volger, O. L., . . . Horrevoets, A. J. (2007). Prolonged shear stress and

KLF2 suppress constitutive proinflammatory transcription through

inhibition of ATF2. Blood, 109(10), 4249-4257. doi: 10.1182/blood-2006-

07-036020

Forget, B. G., & Bunn, H. F. (2013). Classification of the disorders of hemoglobin.

Cold Spring Harb Perspect Med, 3(2), a011684. doi:

10.1101/cshperspect.a011684

Fujioka, H., Tandler, B., & Hoppel, C. L. (2012). Mitochondrial division in rat

cardiomyocytes: an electron microscope study. Anat Rec (Hoboken),

295(9), 1455-1461. doi: 10.1002/ar.22523

Funnell, A. P., Vernimmen, D., Lim, W. F., Mak, K. S., Wienert, B., Martyn, G.

E., . . . Crossley, M. (2014). Differential regulation of the alpha-globin

locus by Kruppel-like Factor 3 in erythroid and non-erythroid cells. BMC

Mol Biol, 15, 8. doi: 10.1186/1471-2199-15-8

Gallagher, D. C., Bhatt, R. S., Parikh, S. M., Patel, P., Seery, V., McDermott, D.

F., . . . Sukhatme, V. P. (2007). Angiopoietin 2 is a potential mediator of

high-dose interleukin 2-induced vascular leak. Clin Cancer Res, 13(7),

2115-2120. doi: 10.1158/1078-0432.CCR-06-2509

Gallardo-Vara, E., Blanco, F. J., Roque, M., Friedman, S. L., Suzuki, T., Botella,

L. M., & Bernabeu, C. (2016). Transcription factor KLF6 upregulates

expression of metalloprotease MMP14 and subsequent release of soluble

114

endoglin during vascular injury. Angiogenesis, 19(2), 155-171. doi:

10.1007/s10456-016-9495-8

Galley, H. F., & Webster, N. R. (2004). Physiology of the endothelium. Br J

Anaesth, 93(1), 105-113. doi: 10.1093/bja/aeh163

Garrido-Martin, E. M., Blanco, F. J., Roque, M., Novensa, L., Tarocchi, M., Lang,

U. E., . . . Bernabeu, C. (2013). Vascular injury triggers Kruppel-like factor

6 mobilization and cooperation with specificity protein 1 to promote

endothelial activation through upregulation of the activin receptor-like

kinase 1 gene. Circ Res, 112(1), 113-127. doi:

10.1161/CIRCRESAHA.112.275586

Gaston, B., Singel, D., Doctor, A., & Stamler, J. S. (2006). S-nitrosothiol signaling

in respiratory biology. Am J Respir Crit Care Med, 173(11), 1186-1193.

doi: 10.1164/rccm.200510-1584PP

Ghaleb, A. M., Nandan, M. O., Chanchevalap, S., Dalton, W. B., Hisamuddin, I.

M., & Yang, V. W. (2005). Kruppel-like factors 4 and 5: the yin and yang

regulators of cellular proliferation. Cell Res, 15(2), 92-96. doi:

10.1038/sj.cr.7290271

Ghosh, A. K., & Vaughan, D. E. (2012). PAI-1 in tissue fibrosis. J Cell Physiol,

227(2), 493-507. doi: 10.1002/jcp.22783

Gimbrone, M. A., Jr., Topper, J. N., Nagel, T., Anderson, K. R., & Garcia-

Cardena, G. (2000). Endothelial dysfunction, hemodynamic forces, and

atherogenesis. Ann N Y Acad Sci, 902, 230-239; discussion 239-240.

115

Gracia-Sancho, J., Villarreal, G., Jr., Zhang, Y., & Garcia-Cardena, G. (2010).

Activation of SIRT1 by resveratrol induces KLF2 expression conferring an

endothelial vasoprotective phenotype. Cardiovasc Res, 85(3), 514-519.

doi: 10.1093/cvr/cvp337

Grandaliano, G., Gesualdo, L., Ranieri, E., Monno, R., & Schena, F. P. (2000).

Tissue factor, plasminogen activator inhibitor-1, and thrombin receptor

expression in human crescentic glomerulonephritis. Am J Kidney Dis,

35(4), 726-738.

Hale, A. T., Tian, H., Anih, E., Recio, F. O., 3rd, Shatat, M. A., Johnson, T., . . .

Hamik, A. (2014). Endothelial Kruppel-like factor 4 regulates angiogenesis

and the Notch signaling pathway. J Biol Chem, 289(17), 12016-12028.

doi: 10.1074/jbc.M113.530956

Hamik, A., Lin, Z., Kumar, A., Balcells, M., Sinha, S., Katz, J., . . . Jain, M. K.

(2007). Kruppel-like factor 4 regulates endothelial inflammation. J Biol

Chem, 282(18), 13769-13779. doi: 10.1074/jbc.M700078200

Hanaichi, T., Sato, T., Iwamoto, T., Malavasi-Yamashiro, J., Hoshino, M., &

Mizuno, N. (1986). A stable lead by modification of Sato's method. J

Electron Microsc (Tokyo), 35(3), 304-306.

Healy, A. M., Rayburn, H. B., Rosenberg, R. D., & Weiler, H. (1995). Absence of

the blood-clotting regulator thrombomodulin causes embryonic lethality in

mice before development of a functional cardiovascular system. Proc Natl

Acad Sci U S A, 92(3), 850-854.

116

Heberlein, K. R., Straub, A. C., Best, A. K., Greyson, M. A., Looft-Wilson, R. C.,

Sharma, P. R., . . . Isakson, B. E. (2010). Plasminogen activator inhibitor-1

regulates myoendothelial junction formation. Circ Res, 106(6), 1092-1102.

doi: 10.1161/CIRCRESAHA.109.215723

Heo, K. S., Fujiwara, K., & Abe, J. (2011). Disturbed-flow-mediated vascular

reactive oxygen species induce endothelial dysfunction. Circ J, 75(12),

2722-2730.

Hiroi, T., Deming, C. B., Zhao, H., Hansen, B. S., Arkenbout, E. K., Myers, T.

J., . . . Rade, J. J. (2009). Proteasome inhibitors enhance endothelial

thrombomodulin expression via induction of Kruppel-like transcription

factors. Arterioscler Thromb Vasc Biol, 29(10), 1587-1593. doi:

10.1161/ATVBAHA.109.191957

Hoeben, A., Landuyt, B., Highley, M. S., Wildiers, H., Van Oosterom, A. T., & De

Bruijn, E. A. (2004). Vascular endothelial growth factor and angiogenesis.

Pharmacol Rev, 56(4), 549-580. doi: 10.1124/pr.56.4.3

Huddleson, J. P., Ahmad, N., & Lingrel, J. B. (2006). Up-regulation of the KLF2

transcription factor by fluid shear stress requires nucleolin. J Biol Chem,

281(22), 15121-15128. doi: 10.1074/jbc.M513406200

Huddleson, J. P., Ahmad, N., Srinivasan, S., & Lingrel, J. B. (2005). Induction of

KLF2 by fluid shear stress requires a novel promoter element activated by

a phosphatidylinositol 3-kinase-dependent chromatin-remodeling pathway.

J Biol Chem, 280(24), 23371-23379. doi: 10.1074/jbc.M413839200

117

International Human Genome Sequencing, C. (2004). Finishing the euchromatic

sequence of the human genome. Nature, 431(7011), 931-945. doi:

10.1038/nature03001

Isermann, B., Hendrickson, S. B., Zogg, M., Wing, M., Cummiskey, M., Kisanuki,

Y. Y., . . . Weiler, H. (2001). Endothelium-specific loss of murine

thrombomodulin disrupts the protein C anticoagulant pathway and causes

juvenile-onset thrombosis. J Clin Invest, 108(4), 537-546. doi:

10.1172/JCI13077

Jain, M. K., Sangwung, P., & Hamik, A. (2014). Regulation of an inflammatory

disease: Kruppel-like factors and atherosclerosis. Arterioscler Thromb

Vasc Biol, 34(3), 499-508. doi: 10.1161/ATVBAHA.113.301925

Ju, H., Zou, R., Venema, V. J., & Venema, R. C. (1997). Direct interaction of

endothelial nitric-oxide synthase and caveolin-1 inhibits synthase activity.

J Biol Chem, 272(30), 18522-18525.

Kalra, I. S., Alam, M. M., Choudhary, P. K., & Pace, B. S. (2011). Kruppel-like

Factor 4 activates HBG gene expression in primary erythroid cells. Br J

Haematol, 154(2), 248-259. doi: 10.1111/j.1365-2141.2011.08710.x

Katz, J. P., Perreault, N., Goldstein, B. G., Actman, L., McNally, S. R., Silberg, D.

G., . . . Kaestner, K. H. (2005). Loss of Klf4 in mice causes altered

proliferation and differentiation and precancerous changes in the adult

stomach. Gastroenterology, 128(4), 935-945.

118

Kazmi, R. S., Boyce, S., & Lwaleed, B. A. (2015). Homeostasis of Hemostasis:

The Role of Endothelium. Semin Thromb Hemost, 41(6), 549-555. doi:

10.1055/s-0035-1556586

Keller, T. C. t., Butcher, J. T., Broseghini-Filho, G. B., Marziano, C., DeLalio, L. J.,

Rogers, S., . . . Isakson, B. E. (2016). Modulating Vascular

Hemodynamics With an Alpha Globin Mimetic Peptide (HbalphaX).

Hypertension, 68(6), 1494-1503. doi:

10.1161/HYPERTENSIONAHA.116.08171

Kim, Y., Ratziu, V., Choi, S. G., Lalazar, A., Theiss, G., Dang, Q., . . . Friedman,

S. L. (1998). Transcriptional activation of transforming growth factor beta1

and its receptors by the Kruppel-like factor Zf9/core promoter-binding

protein and Sp1. Potential mechanisms for autocrine fibrogenesis in

response to injury. J Biol Chem, 273(50), 33750-33758.

Kirchhofer, D., Tschopp, T. B., Hadvary, P., & Baumgartner, H. R. (1994).

Endothelial cells stimulated with tumor necrosis factor-alpha express

varying amounts of tissue factor resulting in inhomogenous fibrin

deposition in a native blood flow system. Effects of thrombin inhibitors. J

Clin Invest, 93(5), 2073-2083. doi: 10.1172/JCI117202

Kojima, S., Hayashi, S., Shimokado, K., Suzuki, Y., Shimada, J., Crippa, M. P., &

Friedman, S. L. (2000). Transcriptional activation of urokinase by the

Kruppel-like factor Zf9/COPEB activates latent TGF-beta1 in vascular

endothelial cells. Blood, 95(4), 1309-1316.

119

Koritschoner, N. P., Bocco, J. L., Panzetta-Dutari, G. M., Dumur, C. I., Flury, A.,

& Patrito, L. C. (1997). A novel human zinc finger protein that interacts

with the core promoter element of a TATA box-less gene. J Biol Chem,

272(14), 9573-9580.

Kowluru, R. A. (2005). Effect of advanced glycation end products on accelerated

apoptosis of retinal capillary cells under in vitro conditions. Life Sci, 76(9),

1051-1060. doi: 10.1016/j.lfs.2004.10.017

Kumar, A., Lin, Z., SenBanerjee, S., & Jain, M. K. (2005). Tumor necrosis factor

alpha-mediated reduction of KLF2 is due to inhibition of MEF2 by NF-

kappaB and histone deacetylases. Mol Cell Biol, 25(14), 5893-5903. doi:

10.1128/MCB.25.14.5893-5903.2005

Kuo, C. T., Veselits, M. L., Barton, K. P., Lu, M. M., Clendenin, C., & Leiden, J. M.

(1997). The LKLF transcription factor is required for normal tunica media

formation and blood vessel stabilization during murine embryogenesis.

Genes Dev, 11(22), 2996-3006.

Kuo, C. T., Veselits, M. L., & Leiden, J. M. (1997). LKLF: A transcriptional

regulator of single-positive T cell quiescence and survival. Science,

277(5334), 1986-1990.

Kwon, I. S., Wang, W., Xu, S., & Jin, Z. G. (2014). Histone deacetylase 5

interacts with Kruppel-like factor 2 and inhibits its transcriptional activity in

endothelium. Cardiovasc Res, 104(1), 127-137. doi: 10.1093/cvr/cvu183

120

Lalazar, A., Wong, L., Yamasaki, G., & Friedman, S. L. (1997). Early genes

induced in hepatic stellate cells during wound healing. Gene, 195(2), 235-

243.

Lawrence, T. (2009). The nuclear factor NF-kappaB pathway in inflammation.

Cold Spring Harb Perspect Biol, 1(6), a001651. doi:

10.1101/cshperspect.a001651

Lee, J. S., Yu, Q., Shin, J. T., Sebzda, E., Bertozzi, C., Chen, M., . . . Kahn, M. L.

(2006). Klf2 is an essential regulator of vascular hemodynamic forces in

vivo. Dev Cell, 11(6), 845-857. doi: 10.1016/j.devcel.2006.09.006

Levi, M., Toh, C. H., Thachil, J., & Watson, H. G. (2009). Guidelines for the

diagnosis and management of disseminated intravascular coagulation.

British Committee for Standards in Haematology. Br J Haematol, 145(1),

24-33. doi: 10.1111/j.1365-2141.2009.07600.x

Li, B., Ding, L., Li, W., Story, M. D., & Pace, B. S. (2012). Characterization of the

transcriptome profiles related to globin gene switching during in vitro

erythroid maturation. BMC Genomics, 13, 153. doi: 10.1186/1471-2164-

13-153

Libby, P., Ridker, P. M., & Hansson, G. K. (2011). Progress and challenges in

translating the biology of atherosclerosis. Nature, 473(7347), 317-325. doi:

10.1038/nature10146

Lilly, B. (2014). We have contact: endothelial cell-smooth muscle cell interactions.

Physiology (Bethesda), 29(4), 234-241. doi: 10.1152/physiol.00047.2013

121

Lim, S. S., Vos, T., Flaxman, A. D., Danaei, G., Shibuya, K., Adair-Rohani, H., . . .

Memish, Z. A. (2012). A comparative risk assessment of burden of

disease and injury attributable to 67 risk factors and risk factor clusters in

21 regions, 1990-2010: a systematic analysis for the Global Burden of

Disease Study 2010. Lancet, 380(9859), 2224-2260. doi: 10.1016/S0140-

6736(12)61766-8

Lim, Y. C., Garcia-Cardena, G., Allport, J. R., Zervoglos, M., Connolly, A. J.,

Gimbrone, M. A., Jr., & Luscinskas, F. W. (2003). Heterogeneity of

endothelial cells from different organ sites in T-cell subset recruitment. Am

J Pathol, 162(5), 1591-1601. doi: 10.1016/S0002-9440(10)64293-9

Lin, Z., Hamik, A., Jain, R., Kumar, A., & Jain, M. K. (2006). Kruppel-like factor 2

inhibits protease activated receptor-1 expression and thrombin-mediated

endothelial activation. Arterioscler Thromb Vasc Biol, 26(5), 1185-1189.

doi: 10.1161/01.ATV.0000215638.53414.99

Lin, Z., Kumar, A., SenBanerjee, S., Staniszewski, K., Parmar, K., Vaughan, D.

E., . . . Jain, M. K. (2005). Kruppel-like factor 2 (KLF2) regulates

endothelial thrombotic function. Circ Res, 96(5), e48-57. doi:

10.1161/01.RES.0000159707.05637.a1

Lin, Z., Natesan, V., Shi, H., Dong, F., Kawanami, D., Mahabeleshwar, G. H., . . .

Jain, M. K. (2010). Kruppel-like factor 2 regulates endothelial barrier

function. Arterioscler Thromb Vasc Biol, 30(10), 1952-1959. doi:

10.1161/ATVBAHA.110.211474

122

Madge, L. A., & Pober, J. S. (2001). TNF signaling in vascular endothelial cells.

Exp Mol Pathol, 70(3), 317-325. doi: 10.1006/exmp.2001.2368

Madrigal, P., & Krajewski, P. (2012). Current bioinformatic approaches to identify

DNase I hypersensitive sites and genomic footprints from DNase-seq data.

Front Genet, 3, 230. doi: 10.3389/fgene.2012.00230

Mahabeleshwar, G. H., Kawanami, D., Sharma, N., Takami, Y., Zhou, G., Shi,

H., . . . Jain, M. K. (2011). The myeloid transcription factor KLF2 regulates

the host response to polymicrobial infection and endotoxic shock.

Immunity, 34(5), 715-728. doi: 10.1016/j.immuni.2011.04.014

Mamputu, J. C., & Renier, G. (2004). Advanced glycation end-products increase

monocyte adhesion to retinal endothelial cells through vascular endothelial

growth factor-induced ICAM-1 expression: inhibitory effect of antioxidants.

J Leukoc Biol, 75(6), 1062-1069. doi: 10.1189/jlb.0603265

Mantovani, A., Bussolino, F., & Dejana, E. (1992). Cytokine regulation of

endothelial cell function. FASEB J, 6(8), 2591-2599.

Marini, M. G., Porcu, L., Asunis, I., Loi, M. G., Ristaldi, M. S., Porcu, S., . . . Moi,

P. (2010). Regulation of the human HBA genes by KLF4 in erythroid cell

lines. Br J Haematol, 149(5), 748-758. doi: 10.1111/j.1365-

2141.2010.08130.x

Matsumoto, N., Kubo, A., Liu, H., Akita, K., Laub, F., Ramirez, F., . . . Friedman,

S. L. (2006). Developmental regulation of yolk sac hematopoiesis by

Kruppel-like factor 6. Blood, 107(4), 1357-1365. doi: 10.1182/blood-2005-

05-1916

123

McConnell, B. B., & Yang, V. W. (2010). Mammalian Kruppel-like factors in

health and diseases. Physiol Rev, 90(4), 1337-1381. doi:

10.1152/physrev.00058.2009

Monahan-Earley, R., Dvorak, A. M., & Aird, W. C. (2013). Evolutionary origins of

the blood vascular system and endothelium. J Thromb Haemost, 11 Suppl

1, 46-66. doi: 10.1111/jth.12253

Munoz-Chapuli, R., Quesada, A. R., & Angel Medina, M. (2004). Angiogenesis

and signal transduction in endothelial cells. Cell Mol Life Sci, 61(17),

2224-2243. doi: 10.1007/s00018-004-4070-7

Nayak, L., Lin, Z., & Jain, M. K. (2011). "Go with the flow": how Kruppel-like

factor 2 regulates the vasoprotective effects of shear stress. Antioxid

Redox Signal, 15(5), 1449-1461. doi: 10.1089/ars.2010.3647

Nayak, L., Shi, H., Atkins, G. B., Lin, Z., Schmaier, A. H., & Jain, M. K. (2014).

The thromboprotective effect of bortezomib is dependent on the

transcription factor Kruppel-like factor 2 (KLF2). Blood, 123(24), 3828-

3831. doi: 10.1182/blood-2014-01-547448

Ohnesorge, N., Viemann, D., Schmidt, N., Czymai, T., Spiering, D., Schmolke,

M., . . . Schmidt, M. (2010). Erk5 activation elicits a vasoprotective

endothelial phenotype via induction of Kruppel-like factor 4 (KLF4). J Biol

Chem, 285(34), 26199-26210. doi: 10.1074/jbc.M110.103127

Orlowski, R. Z., & Kuhn, D. J. (2008). Proteasome inhibitors in cancer therapy:

lessons from the first decade. Clin Cancer Res, 14(6), 1649-1657. doi:

10.1158/1078-0432.CCR-07-2218

124

Pace, B. S., & Makala, L. H. (2012). The Hemoglobin Regulatory Regions. In N.

Ahituv (Ed.), Gene Regulatory Sequences and Human Disease (pp. 19-

40). New York, NY: Springer New York.

Parmar, K. M., Larman, H. B., Dai, G., Zhang, Y., Wang, E. T., Moorthy, S. N., . . .

Garcia-Cardena, G. (2006). Integration of flow-dependent endothelial

phenotypes by Kruppel-like factor 2. J Clin Invest, 116(1), 49-58. doi:

10.1172/JCI24787

Parmar, K. M., Nambudiri, V., Dai, G., Larman, H. B., Gimbrone, M. A., Jr., &

Garcia-Cardena, G. (2005). Statins exert endothelial atheroprotective

effects via the KLF2 transcription factor. J Biol Chem, 280(29), 26714-

26719. doi: 10.1074/jbc.C500144200

Pei, J., & Grishin, N. V. (2013). A new family of predicted Kruppel-like factor

genes and pseudogenes in placental mammals. PLoS One, 8(11), e81109.

doi: 10.1371/journal.pone.0081109

Pioli, P. D., & Weis, J. H. (2014). Snail transcription factors in hematopoietic cell

development: a model of functional redundancy. Exp Hematol, 42(6), 425-

430. doi: 10.1016/j.exphem.2014.03.002

Pober, J. S., & Sessa, W. C. (2007). Evolving functions of endothelial cells in

inflammation. Nat Rev Immunol, 7(10), 803-815. doi: 10.1038/nri2171

Rahaman, M. M., Reinders, F. G., Koes, D., Nguyen, A. T., Mutchler, S. M.,

Sparacino-Watkins, C., . . . Straub, A. C. (2015). Structure Guided

Chemical Modifications of Propylthiouracil Reveal Novel Small Molecule

Inhibitors of Cytochrome b5 Reductase 3 That Increase Nitric Oxide

125

Bioavailability. J Biol Chem, 290(27), 16861-16872. doi:

10.1074/jbc.M114.629964

Rajendran, P., Rengarajan, T., Thangavel, J., Nishigaki, Y., Sakthisekaran, D.,

Sethi, G., & Nishigaki, I. (2013). The vascular endothelium and human

diseases. Int J Biol Sci, 9(10), 1057-1069. doi: 10.7150/ijbs.7502

Ratziu, V., Lalazar, A., Wong, L., Dang, Q., Collins, C., Shaulian, E., . . .

Friedman, S. L. (1998). Zf9, a Kruppel-like transcription factor up-

regulated in vivo during early hepatic fibrosis. Proc Natl Acad Sci U S A,

95(16), 9500-9505.

Rhee, W. J., Ni, C. W., Zheng, Z., Chang, K., Jo, H., & Bao, G. (2010). HuR

regulates the expression of stress-sensitive genes and mediates

inflammatory response in human umbilical vein endothelial cells. Proc Natl

Acad Sci U S A, 107(15), 6858-6863. doi: 10.1073/pnas.1000444107

Romero, N., Radi, R., Linares, E., Augusto, O., Detweiler, C. D., Mason, R. P., &

Denicola, A. (2003). Reaction of human hemoglobin with peroxynitrite.

Isomerization to nitrate and secondary formation of protein radicals. J Biol

Chem, 278(45), 44049-44057. doi: 10.1074/jbc.M305895200

Roviezzo, F., Tsigkos, S., Kotanidou, A., Bucci, M., Brancaleone, V., Cirino, G., &

Papapetropoulos, A. (2005). Angiopoietin-2 causes inflammation in vivo

by promoting vascular leakage. J Pharmacol Exp Ther, 314(2), 738-744.

doi: 10.1124/jpet.105.086553

126

Saha, D., Patgaonkar, M., Shroff, A., Ayyar, K., Bashir, T., & Reddy, K. V. (2014).

Hemoglobin expression in nonerythroid cells: novel or ubiquitous? Int J

Inflam, 2014, 803237. doi: 10.1155/2014/803237

Sangwung, P., Zhou, G., Lu, Y., Liao, X., Wang, B., Mutchler, S., . . . Jain, M. (In

press). Regulation of endothelial hemoglobin alpha expression by kruppel-

like factors. Vascular Medicine.

Sangwung, P., Zhou, G., Nayak, L., Chan, E. R., Kumar, S., Kang, D. W., . . .

Jain, M. K. (2017). KLF2 and KLF4 control endothelial identity and

vascular integrity. JCI Insight, 2(4), e91700. doi: 10.1172/jci.insight.91700

Schechter, A. N. (2008). Hemoglobin research and the origins of molecular

medicine. Blood, 112(10), 3927-3938. doi: 10.1182/blood-2008-04-078188

Schuetz, A., Nana, D., Rose, C., Zocher, G., Milanovic, M., Koenigsmann, J., . . .

Carstanjen, D. (2011). The structure of the Klf4 DNA-binding domain links

to self-renewal and macrophage differentiation. Cell Mol Life Sci, 68(18),

3121-3131. doi: 10.1007/s00018-010-0618-x

Segre, J. A., Bauer, C., & Fuchs, E. (1999). Klf4 is a transcription factor required

for establishing the barrier function of the skin. Nat Genet, 22(4), 356-360.

doi: 10.1038/11926

Sen-Banerjee, S., Mir, S., Lin, Z., Hamik, A., Atkins, G. B., Das, H., . . . Jain, M. K.

(2005). Kruppel-like factor 2 as a novel mediator of statin effects in

endothelial cells. Circulation, 112(5), 720-726. doi:

10.1161/CIRCULATIONAHA.104.525774

127

SenBanerjee, S., Lin, Z., Atkins, G. B., Greif, D. M., Rao, R. M., Kumar, A., . . .

Jain, M. K. (2004). KLF2 Is a novel transcriptional regulator of endothelial

proinflammatory activation. J Exp Med, 199(10), 1305-1315. doi:

10.1084/jem.20031132

Shi, H., Sheng, B., Zhang, F., Wu, C., Zhang, R., Zhu, J., . . . Atkins, G. B. (2013).

Kruppel-like factor 2 protects against ischemic stroke by regulating

endothelial blood brain barrier function. Am J Physiol Heart Circ Physiol,

304(6), H796-805. doi: 10.1152/ajpheart.00712.2012

Shields, J. M., Christy, R. J., & Yang, V. W. (1996). Identification and

characterization of a gene encoding a gut-enriched Kruppel-like factor

expressed during growth arrest. J Biol Chem, 271(33), 20009-20017.

Sohn, R. H., Deming, C. B., Johns, D. C., Champion, H. C., Bian, C., Gardner, K.,

& Rade, J. J. (2005). Regulation of endothelial thrombomodulin

expression by inflammatory cytokines is mediated by activation of nuclear

factor-kappa B. Blood, 105(10), 3910-3917. doi: 10.1182/blood-2004-03-

0928

Stirban, A., Gawlowski, T., & Roden, M. (2014). Vascular effects of advanced

glycation endproducts: Clinical effects and molecular mechanisms. Mol

Metab, 3(2), 94-108. doi: 10.1016/j.molmet.2013.11.006

Straub, A. C., Butcher, J. T., Billaud, M., Mutchler, S. M., Artamonov, M. V.,

Nguyen, A. T., . . . Isakson, B. E. (2014). Hemoglobin alpha/eNOS

coupling at myoendothelial junctions is required for nitric oxide scavenging

128

during vasoconstriction. Arterioscler Thromb Vasc Biol, 34(12), 2594-2600.

doi: 10.1161/ATVBAHA.114.303974

Straub, A. C., Lohman, A. W., Billaud, M., Johnstone, S. R., Dwyer, S. T., Lee, M.

Y., . . . Isakson, B. E. (2012). Endothelial cell expression of haemoglobin

alpha regulates nitric oxide signalling. Nature, 491(7424), 473-477. doi:

10.1038/nature11626

Straub, A. C., Zeigler, A. C., & Isakson, B. E. (2014). The myoendothelial

junction: connections that deliver the message. Physiology (Bethesda),

29(4), 242-249. doi: 10.1152/physiol.00042.2013

Suzuki, T., Yamamoto, T., Kurabayashi, M., Nagai, R., Yazaki, Y., & Horikoshi, M.

(1998). Isolation and initial characterization of GBF, a novel DNA-binding

zinc finger protein that binds to the GC-rich binding sites of the HIV-1

promoter. J Biochem, 124(2), 389-395.

Swamynathan, S. K., Katz, J. P., Kaestner, K. H., Ashery-Padan, R., Crawford, M.

A., & Piatigorsky, J. (2007). Conditional deletion of the mouse Klf4 gene

results in corneal epithelial fragility, stromal edema, and loss of

conjunctival goblet cells. Mol Cell Biol, 27(1), 182-194. doi:

10.1128/MCB.00846-06

Takahashi, K., & Yamanaka, S. (2006). Induction of pluripotent stem cells from

mouse embryonic and adult fibroblast cultures by defined factors. Cell,

126(4), 663-676. doi: 10.1016/j.cell.2006.07.024

Taylor, F., Huffman, M. D., Macedo, A. F., Moore, T. H., Burke, M., Davey Smith,

G., . . . Ebrahim, S. (2013). Statins for the primary prevention of

129

cardiovascular disease. Cochrane Database Syst Rev(1), CD004816. doi:

10.1002/14651858.CD004816.pub5

Thomas, C., & Lemb, A. B. (2012). Physiology of haemoglobin. Contin Educ

Anaesth Crit Care Pain, 12(5), 251-256. doi:

https://doi.org/10.1093/bjaceaccp/mks025

van Beijnum, J. R., Rousch, M., Castermans, K., van der Linden, E., & Griffioen,

A. W. (2008). Isolation of endothelial cells from fresh tissues. Nat Protoc,

3(6), 1085-1091.

van Thienen, J. V., Fledderus, J. O., Dekker, R. J., Rohlena, J., van Ijzendoorn,

G. A., Kootstra, N. A., . . . Horrevoets, A. J. (2006). Shear stress sustains

atheroprotective endothelial KLF2 expression more potently than statins

through mRNA stabilization. Cardiovasc Res, 72(2), 231-240. doi:

10.1016/j.cardiores.2006.07.008

Villarreal, G., Jr., Zhang, Y., Larman, H. B., Gracia-Sancho, J., Koo, A., & Garcia-

Cardena, G. (2010). Defining the regulation of KLF4 expression and its

downstream transcriptional targets in vascular endothelial cells. Biochem

Biophys Res Commun, 391(1), 984-989. doi: 10.1016/j.bbrc.2009.12.002

Wang, W., Ha, C. H., Jhun, B. S., Wong, C., Jain, M. K., & Jin, Z. G. (2010). Fluid

shear stress stimulates phosphorylation-dependent nuclear export of

HDAC5 and mediates expression of KLF2 and eNOS. Blood, 115(14),

2971-2979. doi: 10.1182/blood-2009-05-224824

130

Wang, Y., Schnegelsberg, P. N., Dausman, J., & Jaenisch, R. (1996). Functional

redundancy of the muscle-specific transcription factors Myf5 and

myogenin. Nature, 379(6568), 823-825. doi: 10.1038/379823a0

Wang, Y., & Sheibani, N. (2002). Expression pattern of alternatively spliced

PECAM-1 isoforms in hematopoietic cells and platelets. J Cell Biochem,

87(4), 424-438. doi: 10.1002/jcb.10321

Wani, M. A., Conkright, M. D., Jeffries, S., Hughes, M. J., & Lingrel, J. B. (1999).

cDNA isolation, genomic structure, regulation, and chromosomal

localization of human lung Kruppel-like factor. Genomics, 60(1), 78-86.

doi: 10.1006/geno.1999.5888

Weiler-Guettler, H., Aird, W. C., Husain, M., Rayburn, H., & Rosenberg, R. D.

(1996). Targeting of transgene expression to the vascular endothelium of

mice by homologous recombination at the thrombomodulin locus. Circ Res,

78(2), 180-187.

Weiler, H. (2004). Mouse models of thrombosis: thrombomodulin. Thromb

Haemost, 92(3), 467-477. doi: 10.1160/TH04-05-0307

Widlansky, M. E., Gokce, N., Keaney, J. F., Jr., & Vita, J. A. (2003). The clinical

implications of endothelial dysfunction. J Am Coll Cardiol, 42(7), 1149-

1160.

Wilson, M., Forsyth, P., & Whiteside, J. (2010). Haemoglobinopathy and sickle

cell disease. Contin Educ Anaesth Crit Care Pain, 10(1), 24-28. doi:

https://doi.org/10.1093/bjaceaccp/mkp038

131

Wu, W., Xiao, H., Laguna-Fernandez, A., Villarreal, G., Jr., Wang, K. C., Geary,

G. G., . . . Shyy, J. Y. (2011). Flow-Dependent Regulation of Kruppel-Like

Factor 2 Is Mediated by MicroRNA-92a. Circulation, 124(5), 633-641. doi:

10.1161/CIRCULATIONAHA.110.005108

Yet, S. F., McA'Nulty, M. M., Folta, S. C., Yen, H. W., Yoshizumi, M., Hsieh, C.

M., . . . Lee, M. E. (1998). Human EZF, a Kruppel-like zinc finger protein,

is expressed in vascular endothelial cells and contains transcriptional

activation and repression domains. J Biol Chem, 273(2), 1026-1031.

Yoshida, T., Yamashita, M., Horimai, C., & Hayashi, M. (2014). Deletion of

Kruppel-like factor 4 in endothelial and hematopoietic cells enhances

neointimal formation following vascular injury. J Am Heart Assoc, 3(1),

e000622. doi: 10.1161/JAHA.113.000622

Yoshida, T., Yamashita, M., Iwai, M., & Hayashi, M. (2016). Endothelial Kruppel-

Like Factor 4 Mediates the Protective Effect of Statins against Ischemic

AKI. J Am Soc Nephrol, 27(5), 1379-1388. doi: 10.1681/ASN.2015040460

Yoshizumi, M., Perrella, M. A., Burnett, J. C., Jr., & Lee, M. E. (1993). Tumor

necrosis factor downregulates an endothelial nitric oxide synthase mRNA

by shortening its half-life. Circ Res, 73(1), 205-209.

Zhang, R., Hess, D. T., Qian, Z., Hausladen, A., Fonseca, F., Chaube, R., . . .

Stamler, J. S. (2015). Hemoglobin betaCys93 is essential for

cardiovascular function and integrated response to hypoxia. Proc Natl

Acad Sci U S A, 112(20), 6425-6430. doi: 10.1073/pnas.1502285112

132

Zhang, R., Hess, D. T., Reynolds, J. D., & Stamler, J. S. (2016). Hemoglobin S-

nitrosylation plays an essential role in cardioprotection. J Clin Invest,

126(12), 4654-4658. doi: 10.1172/JCI90425

Zhou, G., Hamik, A., Nayak, L., Tian, H., Shi, H., Lu, Y., . . . Jain, M. K. (2012).

Endothelial Kruppel-like factor 4 protects against atherothrombosis in mice.

J Clin Invest, 122(12), 4727-4731. doi: 10.1172/JCI66056

133