Angiopoietin-1/Tie2 Signaling Axis in Endothelial And

ANGIOPOIETIN-1/TIE2 SIGNALING IN ENDOTHELIAL CELLS

Sharon Harel

McGill University Department of Physiology Montreal, Quebec, Canada

August 2016

A thesis submitted to the Faculty of Graduate Studies and Research in partial fulfillment of the requirements of the degree of Doctor of Philosophy

 Sharon Harel 2016

I. ABSTRACT ...... 6

I. RESUME ...... 8

II. ACKNOWLEDGMENTS ...... 10

III. LIST OF PUBLICATIONS ...... 11

IV. CONTRIBUTION OF AUTHORS...... 12

V. LIST OF ABBREVIATIONS ...... 13

VI. LIST OF FIGURES ...... 19

VII. LIST OF TABLES ...... 21

CHAPTER 1: Literature Review ...... 22

1.1 Vascular endothelium: structure, function and development ...... 23 1.1.1 Introduction ...... 23 1.1.2 Structure and function of the vascular endothelium ...... 25 1.1.2.1 Structure and function of blood vessels ...... 25 1.1.2.2 Structure of the vascular endothelium ...... 26 1.1.2.3 The quiescent endothelium ...... 27 1.1.2.4 The activated endothelium and inflammation...... 28 1.1.2.5 The activated endothelium and angiogenesis ...... 30 1.1.3 The formation of blood vessels ...... 31 1.1.3.1 Vasculogenesis ...... 31 1.1.3.2 Angiogenesis ...... 32 1.1.3.3 Vessel maturation...... 34 1.2 Biological function of Ang-1/Tie-2 axis ...... 35 1.2.1 Pathological angiogenesis and clinical aspects ...... 35 1.2.2 Ang-1/Tie-2 in embryonic development (in vivo) ...... 37 1.2.3 Ang-1/Tie-2 and in vitro angiogenesis ...... 40 1.2.4 Ang-1/Tie-2 signaling pathways ...... 40 1.2.4.1 Tie receptors...... 41 1.2.4.2 Angiopoietins ...... 44 1.2.4.2.1 Angiopoietin-2 ...... 45 1.2.4.2.2 Angiopoietin-3/4 ...... 46 1.2.4.3 The phosphoinositide 3-kinase pathway ...... 47 1.2.4.4 NFκB and the inflammatory phenotype ...... 49 1.2.4.5 The MAPKs ...... 50 1.2.4.5.1 Dual specifity phosphatases as modulators of the MAPKs ...... 55 1.2.4.6 The downstream of kinase related protein pathway ...... 56 1.2.4.7 The role of integrins in angiopoietin signaling ...... 57

1.2.4.8 Nuclear modulators of Ang-1/Tie2 induced angiogenesis...... 58 1. 3 Reactive oxygen species and NADPH oxidase in angiogenesis ...... 59 1.3.1 Introduction ...... 59 1.3.2 NADPH oxidase: an overview ...... 60 1.3.2.1 Vascular NADPH oxidase structure, localization and function...... 61 1.3.2.2 Regulation of NOXs in endothelium ...... 63 1.3.2.3 NOX2 and NOX4 in angiogenesis ...... 64 1.3.3 Mitochondrial derived ROS in angiogenesis ...... 65 1.3.3.1 Regulation of mitochondrial ROS ...... 66 1.3.3.2 Mitochondrial derived ROS in cell signaling ...... 67 1.4.1 Dual role of reactive oxygen species in angiogenesis ...... 67 1.4.2 The balance between oxidant and antioxidants ...... 68 1.4 MicroRNAs in the regulation of Angiogenesis ...... 70 1.4.1 Discovery of miRNA ...... 70 1.4.2 Mechanisms of microRNA biogenesis and silencing by the RNA-induced silencing complex...... 71 1.4.3 MicroRNAs as modulators of signaling pathways ...... 73 1.4.4 MicroRNAs as modulator of Angiogenesis ...... 74 1.4.5 miR-640 ...... 75 1.4.6 ZFP91 ...... 76 1.5 ETS family of transcription Factors in the regulation of angiogenesis ...... 77 1.5.1 ETS’s family of transcription factors an overview ...... 77 1.5.2 The Angiopoietin-1/Tie-2 signaling and ETS...... 78 1.5.3 ETS in angiogenesis ...... 79 1.6 General objectives and specific aims ...... 81 Preface to Chapter 2 ...... 82

Chapter 2: NOX2, NOX4 and Mitochondrial-Derived Reactive Oxygen Species (mtROS)

Contribute to Angiopoietin-1 Angiogenic Responses in Endothelial Cells ...... 83

2.1 Abstract ...... 85

2.2. Introduction ...... 88

2.3 Materials and Methods ...... 92

2.4 Results ...... 97

2.5 Discussion ...... 102

2.6 Acknowledgement ...... 110

2.7 References ...... 112

2.8 Figures...... 118

2.9 Figure legends ...... 132

2.10 Tables ...... 138

Preface to chapter 3 ...... 139

Chapter 3: Regulation of angiogenesis by Angiopoietin-1: Roles of miR-640 and ZFP91

...... 140

3.1 Abstract ...... 143

3.2 Introduction ...... 144

3.3 Material and Methods ...... 146

3.4 Results ...... 151

3.5 Discussion ...... 158

3.6 Acknowledgments...... 161

3.7 References ...... 162

3.8 Figures...... 167

3.9 Figure legends ...... 184

3.10 Tables ...... 192

Preface to chapter 4 ...... 193

Chapter 4: ETS1, ELK1 and ETV4 Contribute to Angipoietin-1 Angiogenic Responses in

Endothelial Cells ...... 194

4.1 Abstract ...... 196

4.2 Introduction ...... 198

4.3 Material and Methods ...... 199

4.4 Results ...... 204

4.5 Discussion ...... 208

4.6 Acknowledgments...... 213

4.7 References ...... 214

4.8 Figures...... 220

4.9 Figure Legends...... 229

4.10 Tables ...... 233

Chapter 5: Original Contribution to Scientific Knowledge ...... 238

Chapter 6: Discussion and Opportunities for Future Research ...... 242

Clinical Relevance ...... 254 Reference List ...... 263

I. ABSTRACT

Angiogenesis, the formation of new blood vessels from pre-existing ones, is a normal physiological process required for embryonic development, growth and wound healing.

Aberrant angiogenesis is a hallmark of several pathologies, including cancer, asthma, diabetic retinopathies and pulmonary hypertension. The Angiopoietins and Tie receptors have emerged as alternative targets for therapeutic intervention due to their function as regulators of angiogenesis. Angiopoietin-1 (Ang-1), the main ligand of Tie-2 receptors, promotes vessel growth, inhibits inflammation and maintains vessel stability. Although important advances have been made in understanding the function of Ang-1 in the vasculature, the intracellular signaling pathways downstream of Ang-1 remain largely unexplored.

In our first study, we explored the importance of reactive oxygen species (ROS) derived from NADPH oxidases (NOX2 and NOX4) as well as the mitochondria in the signaling and biological responses elicited by Ang-1 in cultured endothelial cells. Our results indicate that NOX2, NOX4 and mitochondrial-derived ROS contribute distinctively to

Ang-1-induced activation of mitogen-activated protein kinases (ERK1/2, p38 and

SAPK/JNK), the PI-3 kinase/AKT pathway and to Ang-1-induced endothelial cell survival, migration and differentiation.

In our second study, we explored the functional importance of micro RNAs (miRNAs) in the regulation of signaling and biological responses elicited by Ang-1 in cultured endothelial cells. Our study indicates that exposure to Ang-1 leads to significant decrease in the expression of microRNA miR-640 and that this microRNA exerts strong inhibitory effects on the pro-angiogenic responses elicited by Ang-1 including regulation of

6 endothelial cell migration and differentiation. We have also identified ZFP91 as a direct target of miR-640 and that Ang-1-induced cell migration and differentiation requires upregulation of ZFP91 expression and that this upregulation is achieved through removal of the inhibitory effect of miR-640 on ZFP91 expression.

Previous studies from our labs have identified 58 genes to be significantly induced in endothelial cells exposed to Ang-1. In silico analyses of the promoters of these genes predicted that the transcription factors ETS1, ETV4 and ELK1 may be involved in the upregulation of several genes in cells exposed to Ang-1. Our third study have confirmed that exposure of endothelial cells to Ang-1 significantly increases ETS1 expression and significantly enhances the DNA binding activities of ETS, ELK1 and ETV4 and that these responses are accomplished through activation of the p38 and SAPK/JNK as well as the PI-3 kinase/AKT pathways. Moreover, we found that selective silencing of ETS1,

ELK1 or ETV4 expression abolished Ang-1-induced angiogenic responses and eliminated the upregulation by Ang-1 of a subset of pro-angiogenesis genes.

We conclude that Ang-1 induces ROS production through NOX2, NOX4 and the mitochondria to coordinate its angiogenic responses and that these responses of Ang-1 are mediated in part through modulation of ZFP91 expression and that this modulation is achieved through downregulation of miR-640 expression. Finally, we concluded that exposure of endothelial cells to Ang-1 leads to the activation of ETS1, ELK1 and ETV4 transcription factors and that these factors are involved in upregulation of a set of pro- angiogenic genes that are required for the promotion of endothelial cell survival, migration and differentiation by Ang-1.

I. RESUME

L’angiogenèse, la formation de nouveaux vaisseaux sanguins à partir de vaisseaux pré- existants, est un processus physiologique normal nécessaire pour le développement embryonnaire, la croissance et la cicatrisation des plaies. Une angiogénèse aberrante est caractéristique de nombreuses pathologies, dont le cancer, l'asthme, les rétinopathies diabétiques et de l'hypertension pulmonaire. Les angiopoiétines et ses récepteurs Tie ont

émergé comme des cibles alternatives d'intervention thérapeutique en raison de leur fonction en tant que régulateurs de l'angiogenèse. Angiopoïétine-1 (Ang-1), l'agoniste principal du récepteur Tie-2 favorise la croissance des vaisseaux, inhibe l'inflammation et maintient la stabilité des vaisseaux sanguins. Ang-1 régule une constellation de réponses biologiques, y compris l'inhibition de l'apoptose et l'inflammation et la promotion de la migration et la différenciation cellulaire. Bien que des progrès importants aient été réalisés dans la compréhension de la fonction d’Ang-1 dans le système vasculaire, les voies de signalisation intracellulaire en aval d’Ang-1 restent largement inexplorées.

En utilisant des cellules endothéliales de la veine ombilicale humaine (HUVEC), nous avons identifié qu’Ang-1 induit la production d’espèces réactives oxygénées dérivées de

NOX2, NOX4 et de la mitochondrie. Plus précisément, nous avons constaté que NOX2,

NOX4 et ROS mitochondriaux contribuent distinctement à l'activation des voies des

MAPKs et PI3K/AKT induite par Ang-1 ainsi qu’aux réponses angiogéniques induites par Ang-1.

Étant donné qu'une grande partie du génome est sous le contrôle des microARNs, nous avons supposé qu’ Ang-1 induit des microARNs qui ciblent les gènes pro-angiogéniques et réprime des microARNs qui ciblent les gènes anti-angiogéniques. Nous avons constaté

8 qu’Ang-1 régule positivement ZFP91 et que cette protéine est nécessaire pour induire la migration et la différenciation. Nos résultats suggèrent que miR-640 est réprimé par Ang-

1 comme un mécanisme pour induire la migration et la différenciation des cellules endothéliales par surexpression de ZFP91.

Parce que nous avions déjà identifié 58 gènes surexprimés par Ang-1, nous avons ensuite

étudié à travers quels facteurs de transcription Ang-1 régule son transcriptome. Nous avons trouvé qu’Ang-1 augmente l’expression de l'ARNm de ETS1 via les voies de signalisation de p38, SAPK / JNK et PI3K / AKT et qu'il augmente le pouvoir de transcription de ETS1, ELK1 et ETV4. De plus, nous avons observé que l’inhibition de

ETS1, ELK1 ou ETV4 a permis de supprimer les réponses angiogéniques induites par

Ang-1 ainsi que la surexpression d'un sous-ensemble de gènes, dont plusieurs gènes pro- angiogéniques.

Nous concluons qu’Ang-1 induit la production d’espèces réactives oxygénées dérivées de

NOX2, NOX4 et des mitochondries afin de coordonner ses réponses angiogéniques. Ang-

1 module également l'expression ZFP91 par régulation négative de miR-640 afin d'induire la migration et la différenciation des cellules endothéliales. Enfin, Ang-1 conduit à l'activation de ETS1, ELK1 et ETV4 pour réguler un ensemble de gènes pro- angiogéniques impliqués dans la survie, la migration et la différenciation induite par

Ang-1.

II. ACKNOWLEDGMENTS

First and foremost I want to thank Dr Hussain for welcoming me in is laboratory, for creating great opportunities for me and for being so understanding. I wish I had the opportunity to spend more time working in collaboration with him. He still had so much to teach me. I now understand that feeling of so many of his past graduate students who didn’t want to leave his laboratory, he has so much knowledge to transfer and six years go way to fast to even grasp a glimpse of that knowledge.

I would like to thank all my colleagues with whom completion of my doctoral studies would not have been possible. A particular thank you to Raquel Echavarria for inspiring me both as a great scientist and an amazing baker: I can’t recall the number of time, thinking about your achievements, helped me stay on track with my own goals.

Thanks Hodan Ismail for being such a great friend inside and outside of the laboratory, you left too quickly. Thanks to Flavia Stana and Veronica Sanchez, our most recent generation, for making the laboratory such a pleasant environment but also for being there to save my experiments when Zachary was sick. I will miss you both dearly, we were a nice pack. Thanks to Dominique Mayaki for his patience, time and technical help.

Special acknowledgment goes to my committee for their useful input, Dr Sheldon

Magder, Dr Alvin Shrier, Dr John White and Dr Jean-Phillipe Gratton. As well as the

NSERC Canadian Graduate Schorlarship program and the research institute scholarship program for financing my studies.

Finally, I would like to thank my family, my son and my husband, for their patience.

Are you done with your thesis? Yes I am, yes I am for real this time.

III. LIST OF PUBLICATIONS

1. Harel S, Mayaki D, Hussain S, NOX2, NOX4 and Mitochondrial-Derived Reactive Oxygen Species (mtROS) Contribute to Angiopoietin-1 Angiogenic Responses in Endothelial Cells Submitted to Vascular Pharmacology

2. Harel S, Sanchez-Gonzalez V, Echavarria R, Mayaki D, Hussain S, Regulation of angiogenesis by Angiopoietin-1: Roles of miR-640 and ZFP91. To be submitted to Blood

3. Harel S, Sanchez-Gonzalez V, Sanchez-Galan J, Ismail H, Mayaki D, Huck L, Blanchette M, Hussain S, ETS1, ELK1 and ETV4 Contribute to Angiopoietin-1 Angiogenic Responses in Endothelial Cells to be submitted to ATVB

4. Harel S, Mofarrahi M, Kim H, Mayaki D, Vildhard F, Hussain S, Regulation of migration of skeletal muscle precursor cells by angiopoietins: role of NADPH oxidase. In preparation for submission to JAP

5. Echevarria R, Mayaki D, Neel J-C, Harel S, Sanchez V, Lebrun J-J, Hussain SN. Angiopoietin-1 inhibits Toll-like receptor 4 signaling in endothelial cells: Role of miR-146b-5p. Cardiovascular Research, 2015 Jun 1;106(3):465-77

6. Mofarrahi M, McClung JM, Kontos CD, Davis EC, Tappuni B, Moroz N, Pickett AE, Huck L, Harel S, Danialou G, Hussain SN. Angiopoietin-1 enhances skeletal muscle regeneration in mice. Am J Physiol Regul Integr Comp Physiol. 2015 Apr 1;308(7):R576-89

7. Rahman M., Mofarrahi M., Kristof A., Nkengfac B., Harel S., Hussain, S.N., Reactive Oxygen Species Regulation Of Autophagy In Skeletal Muscles. Autophagy, 2014 Jan 20;20(3):443-59

8. Mofarrahi M, Sigala I, Vassilakopoulos T, Harel S, Guo Y, Debigare R, Maltais F, Hussain SN. Angiogenesis related factors in skeletal muscles of COPD patients: roles of angiopoietin-2. J Appl Physiol. 2013 May;114(9):1309-18

9. Ismail, H., Mofarrahi, M., Echavarria, R., Harel, S., Verdin, E., Lim, H.W., Jin, Z.G., Sun, J., Hussain, S., Angiopoietin-1 and vascular endothelial growth factor regulation of leukocyte adhesion to endothelial cells: role of nur77 receptors Arterioscler Thromb Vasc Biol. 2012 Jul;32(7):1707-16

10. Fielhaber, J. A., S. F. Carroll, Dydensborg, A.B., Shourian, M., Triantafillopoulos, A., Harel, S., Hussain, S., Bouchard, M., Qureshi, S.T., Kristof, A.S., (2012). "Inhibition of Mammalian Target of Rapamycin Augments Lipopolysaccharide-Induced Lung Injury and Apoptosis." Journal of immunology. 2012 May 1;188(9):4535-42

IV. CONTRIBUTION OF AUTHORS

Dr. Sabah N. Hussain was instrumental in the interpretation of data and the generation of all the manuscripts. I am responsible for most of the experiments, the analysis carried out in the works comprised in this thesis and the generation of the manuscripts (Manuscript 1 to 3). Dominique Mayaki provided technical assistance for quantitative real-time PCR experiments throughout the work presented in this thesis. The co-authors contributed in the following ways:

Manuscript 2: Regulation of angiogenesis by Angiopoietin-1: Roles of miR-640 and ZFP91

Echavarria Raquel (McGill University) – Provided data from micro-array identifying miR-640 as a miRNA regulated by Ang-1.

Sanchez-Gonzalez Veronica (McGill University) – executed qRT-PCR and miRNA pull down experiments

Manuscript 3: ETS1, ELK1 and ETV4 Contribute to Angiopoietin-1 Angiogenic Responses in Endothelial Cells

Sanchez-Galan Javier (McGill University) - Generated in silico predictive tool, guided us with is expertise in bioinformatics and edited manuscript

Sanchez-Gonzalez Veronica (McGill University) – Contributed assistance with Western Blots, qRT-PCR and blind analysis

Hodan Ismail- Contributed ETS EMSAs

Huck Laurent (McGill University) – Performed immunofluorescence analysis

Blanchette Matthieu (McGill University)- Guided us with is expertise in bioinformatics

V. LIST OF ABBREVIATIONS

ABIN-2: A20 binding inhibitor of NF- cPLA2: cytosolic phospholipases A2

B activation 2 CSF-1: colony stimulating factor 1

AF: activation function CT: comparative threshold

Ago-2: argonaute 2 Ang: angiopoietin CXCR4: (C-X-C motif) receptor 4

AP-1: activator protein 1 DAPK: death-associated protein kinase

ARE: adenylate-uridylate (AU)-rich DBD: DNA-binding domain elements DGCR8: DiGeorge syndrome critical

ASK1: apoptosis signal-regulating region gene 8 kinase 1 DLL4: delta-like ligand 4

ATF: activating transcription factor DMEM: Dulbecco's Modified Eagle

BAD: Bcl-2-associated death promoter Medium protein DNA: deoxyribonucleic acid

Bax: Bcl-2-associated X protein DN: dominant negative

BMDC: bone marrow-derived cell DN-IKK: dominant-negative IkB

BMP: bone morphogenetic protein kinase beta bp: base pair dNTP: deoxyribonucleotide triphosphate

BRCA1: breast cancer 1 Dok-R: downstream of kinase-related

BSA: bovine serum albumin DUSP: dual-specificity phosphatase

CD: common docking E2: estradiol

CH2: Cdc25 homology EC: endothelial cell

Clo: cloche ECL: enhanced chemiluminescence

CMV: cytomegalovirus ECM: extracellular matrix

EDTA: ethylenediaminetetraacetic acid FITC: fluorescein isothiocyanate

EGF: epidermal growth factor FOXO1: forkhead box protein O1

EGFR: epidermal growth factor receptor GAP: GTPase-activating protein

Egr-1: early growth response 1 GAPDH: glyceraldehyde phosphate

ETS1:ets protooncogene 1 dehydrogenase

ELK1: elk1 ets transcription factor GARG16: glucocorticoid attenuated

ETV4: ets variant 4 response gene 16

ELISA: enzyme-linked immunosorbent GDP: guanosine-5'-diphosphate GEF: assay guanine exchange factor

EMT: epithelial-mesenchymal transition GFP: green fluorescent protein eNOS: endothelial nitric oxide synthase GnRH: gonadotropin-releasing hormone

EPC: endothelial precursor cell GPCR: G protein coupled receptor

Eph: ephrin GR: glucocorticoid receptor

ERK1/2: extracellular signal regulated GRE: glucocorticoid responsive element kinase 1 and 2 Grb2: growth factor receptor-bound

ET-1: endothelin 1 protein 2

FADD: Fas-associated protein with GSK3: glycogen synthase kinase-3 beta death domain GTP: guanosine-5'-triphosphate

FAK: focal adhesion kinase HB-EGF: heparin-bound epidermal

FBS: fetal bovine serum HBSS: Hank’s balanced solution

FCS: fetal calf serum HEPES: growth factor 4-(2-hydroxy

FGF: fibroblast growth factor ethyl)- 1-piperazineethanesulfonic acid

HER2: human epidermal growth factor IP10: interferon gamma-induced protein receptor 2 10

HIF1: hypoxia inducible factor 1 IRAK1: interleukin-1 receptor-

HMEC: human mammary epithelial cells associated kinase 1

HGF: hepatocyte growth factor IRF3: interferon regulatory factor 3

HRP: horseradish peroxidase JNK1/2: c-jun nuclear kinase 1 and 2

HSPG: heparin sulfate proteoglycan KIM: kinase interaction motif

HUVEC: human umbilical vein KLF2: Kruppel-like factor 2 endothelial cell LBD: ligand-binding domain iNOS: inducible nitric oxide synthase LBP: lipopolysaccharide binding protein

ICAM1: intercellular adhesion molecule LHRH: luteinizing hormone-releasing

1 hormone 16

Ig: immunoglobulin LLC: Lewis lung carcinoma

IB: IkappaBalpha LPS: lipopolysaccharide

IKK: IkappaB kinase MAPK: mitogen-activated protein

IL-1: interleukin 1 beta kinase

IL-1R: interleukin 1 receptor MAPKK: mitogen-activated protein

IL-2: interleukin 2 kinase kinase

IL-6: interleukin 6 MAPKKK: mitogen-activated protein

IL-8: interleukin 8 kinase kinase kinase

IL-33: interleukin 33 mDIA: mammalian diaphanous

INF: interferon

MEF2: myocyte enhancer factor 2 NOX2: NADPH oxidase 2

MEMα: minimum essential medium NOX3: NADPH oxidase 3 alpha NOX4: NADPH oxidase 4 miRNA: microRNA NOX5: NADPH oxidase 5

MK2: mitogen activated protein kinase NSCLC: non-small-cell lung carcinoma associated protein kinase 2 Nur77: nuclear receptor 77

MKP: mitogen-activated protein kinase oxLDL: oxidized low-density phosphatase lipoprotein

MLC: myosin light chain p85β: PI-3 kinase regulatory subunit 2

MLCK: myosin light chain kinase p120-RasGAP: p120-Ras GTPase

MMP-2: matrix metalloproteinase 2 activating protein

MMP-16: matrix metalloproteinase 16 PACT: protein activator of PKR

MOI: multiplicity of infection PAF: platelet activating factor

NFAT: nuclear factor of activated T PAI-1: plasminogen activator inhibitor-1 cells PAK1: p21-activating kinase

NF-B: nuclear factor kappa b PAMPs: pathogen-associated molecular

NGF: nerve growth factor patterns

NLS: nuclear localization signal PAR: protease activated receptor

NO: nitric oxide PBS: phosphate buffer saline

NOD: nucleotide-binding PCR: polymerase chain reaction oligomerization domain PDGF: platelet derived growth factor

NOS: nitric oxide synthase PDGFR: platelet derived growth factor

NOX1: NADPH oxidase 1 receptor

PGI2: prostacyclin RIG-I: retinoid acid inducible gene I

PH: pleckstrin homology RLU: relative luminescence units

PI: phosphoinositide RNA: ribonucleic acid RNA Pol II:

PI3-K: phosphoinositide 3-kinase RNA polymerase II

PKB: protein kinase B ROS: reactive oxygen species

PKC: protein kinase C zeta RTK: receptor tyrosine kinase S1P:

PLC: phospholipase C gamma PLZF: sphingosine-1-phosphate promyelocytic leukaemia zinc finger SAPK/JNK: stress-activated protein

PMA: phorbol 12-myristate 13-acetate kinase/Jun-amino-terminal kinase

PMSF: phenylmethanesulfonylfluoride SDS: sodium dodecyl sulfate SDS- p-NPP: p-nitrophenyl phosphate PAGE: sodium dodecyl sulfate

PPAR: peroxisome proliferator-activated polyacrylamide gel electrophoresis receptor SE: standard error

PR: progesterone receptor 17 Ser: serine sFBS: charcoal-stripped FBS

PRR: pattern recognition receptors SFK: Src family kinase

Pro: proline SH2: Src homology 2

Pre-miRNA: precursor microRNA SNP: short nucleotide polymorphism

Pri-miRNA: primary microRNA shRNA: short hairpin ribonucleic acid

PTB: phosphotyrosine binding siRNA: small interfering ribonucleic

PTP: protein tyrosine phosphatase acid

PVDF: polyvinylidene difluoride SOD1: superoxide dismustase 1

RGD: Arginine-Glycine-Aspartic acid SOD2: superoxide dismustase 2

RISC: RNA-induced silencing complex SOS: son of sevenless

STAT: signal transducer and activator of VCAM1: vascular cell adhesion transcription molecule 1 sTie2: soluble Tie-2 VEGF: vascular endothelial growth

TAK1: TGF-beta activated kinase 1 factor

TAM: tamoxifen VEGFR: vascular endothelial growth

TF: tissue factor factor receptor

TGF-: transforming growth factor beta VE-cadherin: vascular endothelial

Thr: threonine cadherin

TIR: Toll-IL-1 receptor VE-PTP: vascular endothelial protein

TLR: toll-like receptor tyrosine phosphatase

TNF-: tumor necrosis factor alpha vSMC: vascular smooth muscle cell

TRAF: TNF receptor associated factor WM: wortmannin

TSC2: tuberous sclerosis protein 2 WPB: Weibel-Palade bodies

Tyr: tyrosine WT: wild-type

UTR: untranslated region UV: XPO5: exportin 5 ultraviolet

VI. LIST OF FIGURES

Chapter 2

Figure 2.1: Ang-1 Regulation of ROS in HUVECs ...... 118 Figure 2.2: NOX2 and NOX4 Contributions to ROS Production ...... 119 Figure 2.3: Ang-1 Regulation of mtROS ...... 120 Figure 2.4: NOX2 Regulation of Ang-1 Signaling ...... 121 Figure 2.5: NOX4 Regulation of Ang-1 Signaling ...... 122 Figure 2.6: mtROS Regulation of Ang-1 Signaling...... 123 Figure 2.7: Roles of NOX2 and NOX4 in Cell Survival ...... 124 Figure 2.8: mtROS Effects on Cell Survival ...... 125 Figure 2.9: ROS Effects on Cell Migration ...... 126 Figure 2.10: Roles of ROS in Cell Differentiation ...... 127 Supplementary Figure S2.1: Copie number of NOXs and DUOX isoforms in HUVECs ...... 128 Supplementary Figure S2.2: Verification of NOXs knockdown ...... 129 Supplementary Figure S2.3: Ang-1 induced Mitochondrial ROS in NOXs knockdown 130

Chapter 3

Figure 3.1: Ang-1 regulation of miR-640 in HUVECs...... 167 Figure 3.2: Effect of mature miR-640 overexpression on Ang-1-induced angiogenesis 168 Figure 3.3: Effect of miR-640 inhibition on Ang-1-induced angiogenesis ...... 169 Figure 3.4: Identification of miR-640 targets ...... 170 Figure 3.5 ZFP91 is a direct target of miR-640 ...... 171 Figure 3.6: Ang-1 regulation of ZFP91 ...... 172 Figure 3.7: Effect of ZFP91 inhibition on Ang-1-induced angiogenesis...... 173 Supplementary Figure 3.1: Abundance of miR-640 in HUVECs...... 174 Supplementary Figure 3.2: Angiogenesis factor regulation of miR-211-3p...... 175 Supplementary Figure 3.3: Transfection efficiency of miR-640 in HUVECs...... 176 Supplementary Figure 3.4: Effect of miR-640 on the 3rd predicted seed in ZFP91 3’UTR ...... 177 Supplementary Figure 3.5: Effect of inhibition of miR-640 on Luciferase expression.. 178 Supplementary Figure 3.6: Verification of ZFP91 knock down with siRNA ...... 179 Supplementary Figure 3.7: Effect of ZFP91 silencing on Ang-1-induced cell survival 180 Supplementary Figure 3.8: Effect of ZFP91 silencing on Ang-1-induced inhibition of apoptosis ...... 181 Supplementary Figure 3.9: Effect of ZFP91 silencing pro-angiogenic gene targets ...... 182 Supplementary Figure 3.10: Expression of ZFP91 in a model of hind limb ischemia ... 183

Chapter 4

Figure 4.1: Predicting transcription factor of Ang-1 regulated gene set ...... 220 Figure 4.2: Ang-1 regulates ETSs expression and DNA binding activity ...... 221

Figure 4.3: Ang-1 induces ETS-1 accumulation in the nucleus ...... 222 Figure 4.4: Ang-1 induces ELK1 accumulation in the nucleus ...... 223 Figure 4.5: Ang-1 induces ETV4 accumulation in the nucleus ...... 224 Figure 4.6: Regulation of ETS1, ELK1 and ETV4 by MAPK, AKT and mTOR pathway ...... 225 Figure 4.7: Role of ETS1, ELK1 and ETV4 in Ang-1 angiogenic responses ...... 226 Figure 4.8: ETS1, ELK1 and ETV4 mediate Ang-1 transcriptome ...... 227 Supplementary Figure 4.1: ETS1, ELK1 and ETV4 effect on basal level of Ang-1- induced genes ...... 228

VII. LIST OF TABLES

Chapter 2

Table 2.1: Primers used for quantitative real-time PCR experiments ...... 138

Chapter 3

Table 3.1: Primers used for quantitative real-time PCR experiments ...... 192

Chapter 4

Table 4.1: Z-scores for the Ang-1 dataset over TRANSFAC profiles (ETS-type Family) ...... 233 Supplementary Table 4.1: Primer List ...... 234 Supplementary Table 4.2: Description of luciferase reporter ...... 236 Supplementary Table 4.3: Enrichment in biological process for a subset of 22 Ang-1-upregulated gene obtain with NOA ...... 237

CHAPTER 1: Literature Review

1.1 Vascular endothelium: structure, function and development

1.1.1 Introduction

The cardiovascular system comprises the heart and the network of vessels circulating blood throughout our body. Blood circulation in the vascular network allows for an efficient transport of gases, nutrients, signaling molecules, cells and metabolic waste between tissues and organs. Every vessel has a luminal surface formed of a single layer of endothelial cells (ECs), called endothelium. The endothelium is an interface between the blood and the extravascular space [1]. Capillaries where most exchanges occur are simply composed of an endothelium, a basal lamina and pericytes.

The endothelium plays an important physiological role in vascular biology: it controls the vascular tone, coagulation, platelet and leukocyte interactions, blood flow, and angiogenesis [2]. Angiogenesis is the formation of new blood vessels from pre-existing ones, during embryonic development, bone morphogenesis, wound healing, the menstrual cycle and pregnancy, but also growth [3]. Altered angiogenesis is involved in more than

70 pathologies arising from excessive or insufficient blood vessel formation [4]. For this reason, understanding the molecular mechanisms that regulate EC biology and blood vessel formation is important to identify new therapeutic strategies for diseases that involve vessel growth. The angiopoietin-Tie family of proteins, an EC-specific Receptor

Tyrosine Kinase (RTK) system, emerges as a target for therapeutical intervention in tumor progression, but also in other vascular and inflammatory pathologies [5;6].

Targeting Ang-1/Tie-2 signaling axis therapeutic potential has been proved in pre-clinical studies. However its broad spectrum of biological effects limits its use and fine tuning

23 intracellular signals to promote the desired therapeutic actions would be a better approach. Angiopoietin-1 (Ang-1), an agonist of Tie-2 receptors, promotes vessel growth in sites of active remodeling while maintaining quiescence in adult tissues by promoting vessel integrity [6]. Important progress has been made in understanding the molecular mechanisms responsible for Ang-1 functions in the vasculature. The role of reactive oxygen species in fine tuning Ang-1/Tie-2 induced angiogenesis have been reported [7-

9]. Still, not much as been done to identify which NOX is activated downstream of Ang-

1/Tie-2 and the respective contribution of NOX and Mitochondrial derived ROS to Ang-1 induced angiogenesis. Moreover, little is known about the nuclear modulators of Ang-

1/Tie-2 signalling axis and their signaling mechanism. Only a handful of transcription factors (TFs) have been identified to be regulated by Ang-1: KLF2 [10], FOXO1 [11],

AP1 [12], GATA3 [13], NFκB [14], STAT [15], EGR1 [16] and 3 members of ETS family: NERF2 [17], ELF1 [13] and Elk1 [18], many of which are also redox sensitive.

We are still far from a complete image of Ang-1/Tie-2 signaling network and we need to identify their gene target and their upstream regulatory mechanisms. It is primordial to understand how the tight balance between pro-angiogenic and anti-angiogenic signaling is maintained and the cellular mechanisms underlying Ang-1/Tie-2 induced angiogenesis, in order to eventually develop specific and efficient therapies to treat angiogenesis- related diseases. The overall objective of this thesis is to understand the molecular mechanisms through which Ang-1 fine tune its biological function in ECs.

1.1.2 Structure and function of the vascular endothelium

1.1.2.1 Structure and function of blood vessels

The vascular system is composed of the heart and an extended network of blood vessels, comprising arteries, arterioles, capillaries, venules and veins. The integral function of the vasculature depends on both, the structural composition of each vessel and the heterogeneity of ECs [19]. The structure of a vessel correlates with its function and the level of pressure to which it is exposed [19]. Blood vessels are formed by three distinct layers: tunica intima, tunica media and tunica externa. The tunica intima is the internal layer of a vessel, mainly comprised of ECs, it is found thoughout the vascular network and capillaries only have tunica intima. The tunica media is formed by concentric layers of smooth muscle and connective tissue that surround ECs and control vascular tone. The tunica media is enriched in smooth muscles cells in arteries and veins. The tunica externa is an outer layer of connective tissue that anchors the vessel to the adjacent tissues [19].

Tunica media and tunica externa comprise the mural cells. Mural cells are mesenchymal cells that closely associate with ECs and are required for homeostasis and organ function.

They are commonly subdivided into vascular smooth muscle cells (vSMCs) and pericytes based on their morphology, location and the expression of specific markers [20]. In general vSMCs form multiple concentric layers around arteries and veins, while pericytes usually interact with ECs in smaller vessels such as arterioles, capillaries and venules

[20]. The vasculature doesn’t undergo peristalsis and only delivers oxygen and nutrients to the tissues in the body in response to the rhythmic contractions of the heart. The oxygenated blood travels away from the heart through the arteries, thick-walled vessels able to expand when the blood enters under pressure, and then circulates from the arteries

25 to the arterioles [20]. The arterioles through contraction of the muscular layer regulate the amount of flow that reaches the tissues and guide the blood into the capillaries. The capillaries are very small vessels that interact closely with the tissues and interconnect arteries and veins. The deoxygenated blood from the capillary beds is returned to the heart through venules and veins.

1.1.2.2 Structure of the vascular endothelium

The endothelium contributes to the control of blood fluidity, platelet aggregation and vascular tone and it participates in the regulation angiogenesis and inflammation. The endothelium is lining the vascular system and its single unit is the EC. ECs are derived from a common mesodermal angioblast. Throughout development, their immediate environment and neighboring cells will determine their heterogeneity and specificity.

There is a great heterogeneity among ECs, depending on their function and location, but most can be characterized by the presence of Weibel-Palade bodies (WPB), fenestra and large amounts of caveolae for the transport of biological material across the endothelial barrier and ECs all form tight monolayers by establishing adherens and tight junctions with neighboring ECs [19;21]. At the molecular level ECs express several endothelium- specific promoters: flk-1, Flt-1, Tie-2, von Willebrand factor, and endothelin-1 [22-26].

Moreover, the promoters of these endothelial genes are enriched in transcription factor binding site (TFBS) for Sp1, EGR-1, ets, and GATA transcription factors [27]. The type of permeability required for a specific tissue largely determines the characteristics of the

ECs lining their capillaries. For example, the blood brain barrier is a very specialized endothelium that requires low permeability and is formed by microvascular ECs with a large number of tight and adherens junctions and significantly reduced number of

26 caveolae compared to other capillaries bed, whereas the sinusoidal capillaries of the liver form pores or fenestrae to facilitate the exchange of fluid and proteins [19].

1.1.2.3 The quiescent endothelium

In normal physiologic conditions, the endothelium contributes to the regulation of blood pressure and microcirculatory flow through its effects on tone of arterioles. It also contributes to regulation of blood vessel permeability and is vital for the maintenance of an anti-coagulative environment. Moreover, it plays a major role in the regulation of wound healing, physiological angiogenesis, and normal host local tissue inflammatory responses including regulation of blood cell migration into tissues

The endothelium regulates blood flow and vascular tone by releasing vasodilators such as nitric oxide (NO) and prostacyclin (PGI2), or vasoconstrictors like endothelin-1 (ET-1) to increase or decrease the tone of the surrounding layers of vSMCs [28-30]. NO is an important vasorelaxant by the family of NO synthases (NOS). The endothelial isoform of

NOS (eNOS) can be activated by VEGF, shear stress and agonists of heterotrimeric G- protein-coupled receptors (GPCRs) like bradykinin and estrogen [31]. The effects of NO on the vasculature are pleiotropic and include vSMC relaxation, regulation of platelet aggregation and inhibition of leukocyte adhesion [32;33]. PGI2 is a byproduct of arachidonic acid metabolism released by ECs in response to disturbances in vascular function [34]. Similarly to NO, PGI2 limits vasoconstriction and inhibits platelet activation by stimulating cell surface prostacyclin receptors and intracellular peroxisome proliferator-activated receptors (PPAR) β/δ on platelets and vSMCs [34]. In contrast, ET-

1 is produced by ECs and vSMCs as a rapid mechanism to reduce vascular tone. At low

27 concentrations ET-1 can also act as a pro-inflammatory mediator. ET-1 binds to Type A endothelin receptors and this interaction mainly mediates the effect of ET-1 on vasoconstriction [35].

In the resting endothelium, the anti-coagulative environment is maintained by preventing the activation of thrombin, a serine protease important in the coagulation cascade due to its role in the cleavage of fibrinogen to fibrin [36]. Thrombin can also affect permeability, vasomotor tone, leukocyte trafficking, migration, angiogenesis and hemostasis; and some of these effects are mediated by the family of protease-activated receptors (PARs) [36].

Endothelial dysfunction arise when the endothelium fails to perform its physiological function, including blood flow maintenance and regulation of vascular tone, coagulation and vascular permeability, leading to an array of vascular related disease [2].

1.1.2.4 The activated endothelium and inflammation

Endothelial cells are considered to be very plastic in the adult due to their ability to respond to various stimuli and reversibly change their functions. This process is termed endothelial activation. EC activation can be divided into type I activation, rapid responses independent of new gene expression; and type II activation, slower responses that require new gene expression [37]. The five core changes of endothelial cell activation are loss of vascular integrity; expression of leucocyte adhesion molecules; change in phenotype from antithrombotic to prothrombotic; cytokine production; and upregulation of Human Leukocyte Antigen (HLA) molecules [38]. In other words, EC activation leads to changes allowing the endothelium to be a key player in the inflammatory process.

Inflammatory mediators such as histamine and leukotrienes activate type I responses on

ECs via heterotrimeric GPCRs [37]. Type I activation generates NO and PGI2 which synergize to create a vasodilator effect that increases blood flow and leukocyte delivery

[39]. The release of Ca2+ from the endoplasmic reticulum into the cytosol also plays an important role in this response. Calmodulin-bound Ca2+ activates myosin light chain kinase (MLCK) which then phosphorylates myosin light chain (MLC) to induce contraction of actin filaments, disruption of tight and adherens junctions, and vascular leakage [40]. Phosphorylated MLC also initiates the exocytosis of P-Selectin from WPB which further promotes leukocyte extravasation [41]. At rest the endothelium inhibits coagulation by preventing the activation of thrombin through the expression of tissue factor pathway inhibitor, thrombomodulin and heparan sulphate proteoglycans (HSPGs)

[42]. However, in response to injury, following type I ECs activation, the endothelium will acquire pro-coagulant activity and in this case the expression of tissue factor (TF) is critical. TF promotes pro-coagulant effects on ECs through the activation of factor IX, factor X and pro-thrombinase [42]. Type I activation has a very limited effect on immune cell extravasation, particularly because after a very short period of time the receptors that mediate this response become desensitized [43].Type II activation also increases blood flow, vascular permeability and leukocyte recruitment [37]. In contrast with type I activation, type II activation is necessary for a more sustained inflammatory response and requires a longer time to be initiated [37]. Pro-inflammatory cytokines secreted by activated leukocytes, such as tumor necrosis factor alpha (TNF-) and interleukin 1 beta (IL1-), stimulate signaling pathways on ECs able to activate the transcription factors nuclear factor kappa B (NF-κB) and activator protein 1 (AP-1) that

29 initiate the transcription of pro-inflammatory proteins [37;44;45]. Among these proteins are cytokines and adhesion molecules that promote leukocyte rolling and extravasation, including E-selectin, intracellular adhesion molecule 1 (ICAM-1) and vascular cell- adhesion molecule 1 (VCAM-1) [46].

In the case of chronic inflammation, angiogenesis also occurs within the sites of sustained injury to support the survival of infiltrated inflammatory cells [37]. Angiogenesis promotes the transformation of the initial matrix into a more permanent stroma comprised of connective tissue, a characteristic feature of certain chronic diseases such as rheumatoid arthritis and atherosclerosis [47].

1.1.2.5 The activated endothelium and angiogenesis

Various stimuli can also activate quiescent ECs towards a pro-angiogenic phenotype.

Angiogenic factors excreted by surrounding hypoxic tissue can lead to ECs activation.

ECs will start degrading the extracellular matrix, enter in a proliferative state and migrate toward these angiogenic factors to form tubular structure to establish the foundation of new blood vessels. The cascade is initiated by the destabilization of the blood vessel.

Angipoietin-2 is believed to be an important contributor. Vessel destabilization is followed by extracellular matrix degradation (ECM) which involves many proteases, including the matrix metalloproteases (MMPs), leading to the release of FGF-2, TGF-β,

VEGF, IGF-1 and TNF-α from the matrix. These growth factors will further promote ECs migration and proliferation.

1.1.3 The formation of blood vessels

The heart is the first functional organ that forms during development, ensuring that the embryo will receive enough oxygen and nutrients as it develops [48]. The heart and the blood vessels originate independently from the mesoderm and it is only at later stages of development that they connect to each other to create a functional circulatory system

[48]. Two distinct processes, vasculogenesis and angiogenesis, are responsible for the formation of blood vessels [48].

1.1.3.1 Vasculogenesis

Vasculogenesis is the formation of new blood vessels from angioblasts [49]. During gastrulation mesodermal cells are induced by bone morphogenetic proteins (BMPs) and fibroblast growth factors (FGFs) to become hemangioblasts, which are precursors of two distinct cell lineages: the angioblast and the pluripotent hematopoietic stem cell [49].

Vasculogenesis occurs in the blood islands, located in the distal part of the yolk sac, where hemangioblasts aggregate and differentiate into hematopoietic stem cells (inner cells) and angioblasts (outer cells) [50]. Signals from the endoderm and mesoderm including VEGF, retinoic acid and BMP4 determine the commitment of hemangioblasts to an either endothelial or hematopoietic fate [49;50]. The angioblasts further differentiate into ECs, expand throughout the entire yolk sac and form a primitive vascular network called the primary capillary plexus [48]. The primary capillary plexus eventually interconnects with the dorsal aorta and the cardinal vein in the embryo proper to create a functional network that allows blood flow as the heart begins to beat [48].

After vasculogenesis the capillary plexus is actively remodeled into a highly branched network containing capillaries, arteries and veins [51]. Vascular remodeling involves the generation of new capillaries as well as changes in vessel diameter, lumen formation and stabilization of the vasculature through the recruitment of mural cells [48;51].

1.1.3.2 Angiogenesis

Angiogenesis is the formation of new capillaries from pre-existing ones. It is required during embryogenesis and for physiological and pathological processes of vascular remodeling that occur in the adult [48]. There are two types of angiogenesis involved in the vascularization of tissues: sprouting and non-sprouting angiogenesis, also known as intussusception [52]. Intussusception, is the process of generating new vessels by splitting pre-existing ones [52]. In capillaries intussusception occurs when opposing vascular walls protrude into the lumen creating a bridge between ECs that will later become a transluminal pillar with an interstitial core [52]. Vessels formed by this type of angiogenesis are less leaky [52]. Newly formed sprouts transform into functional blood vessels via sprout extension and the formation of new vascular connections [53]. Sprout extension occurs through the migration and proliferation of ECs located behind the tip.

Upon finding tips of other sprouts or capillaries, they suppress their motile behavior, establish adhesive interactions and form a lumen to allow blood flow [53].

Sprouting angiogenesis, the formation of new blood vessel from pre-existing ones, is of paramount importance for normal embryonic vascular development and vascular tissue regeneration in adults. This process is regulated by a tight balance between pro- and anti- angiogenic agents and involves a cascade of events of which ECs migration is an essential component, together with vessel stabilization and remodeling. The migratory

32 process is directionally regulated by chemotactic, haptotactic, and mechanotactic stimuli and further involves dynamic and coordinated changes in cell adhesion, signal transduction and cytoskeleton reorganization to enable progression of the migrating cells.

Angiogenesis is a coordinated ordered series of events leading to neovascularization and requires precise spatio-temporal regulation of several processes. It all starts with low

oxygen levels and most transcriptional responses to low O2 are mediated by hypoxia- inducible factors (HIFs). HIF are transcription factors controlling the expression of numerous angiogenic genes. Accordingly, HIF1 is a major regulator of angiogenesis.

Hypoxia promotes through HIFs, vessel growth by upregulating multiple pro-angiogenic pathways that mediate key aspects of endothelial cell biology. Activation of endothelial cells hypoxic factors such as Hypoxia Inducible Factor 1 (HIF1), induces the production of angiogenic growth factor (VEGF, Angiopoietins, PDGF), degradation of basement membranes, proliferation, migration, adhesion, and remodeling of endothelial cells.

VEGFA is of paramount importance in the first step of the angiogenic program and is a gene target of HIF1. During sprouting angiogenesis some ECs within the vessel wall, called the tip cells, are selected by a VEGF-A gradient to lead the growth of the sprout

[54;55]. VEGF-A together with notch receptors and their Delta-like-4 (DLL4) ligand control vessel morphogenesis [56]. The tips of the sprouts express high levels of VEGFR-

2 and respond to spatial concentration gradients of VEGF-A that guide the growth of the sprout [56]. Meanwhile, DLL4 expression is induced in selected tip cells in response to

VEGF-A and activates the notch signaling pathway in neighboring ECs, making them less responsive to pro-angiogenic stimuli by suppressing VEGFR-2 expression [57]. In

33 addition to VEGF-A other molecules including semaphorins, plexins and netrins contribute to sprout guidance [58].

1.1.3.3 Vessel maturation

Vessel maturation occurs when the vasculature changes from a state of active remodeling into a fully functional, quiescent network [59]. Maturation requires the suppression of proliferation and sprouting, followed by the stabilization of existing structures through interaction with mural cells [59]. One mechanism of paracrine regulation between mural cells and ECs is the platelet-derived growth factor (PDGF)/platelet-derived growth factor receptor (PDGFR)  signaling pathway. PDGF secreted by the endothelium acts as a chemoattractant to pericytes who express PDGFR, induces differentiation of mesenchymal cells into mural cells and promotes proliferation of vSMCs [60].

Sphingosine-1-phosphatase (S1P) signaling also regulates the recruitment of mural cells through the activation of a group of endothelial differentiation gene receptors [61]. In addition, transforming growth factor beta (TGF-) induces the differentiation of mesenchymal cells into pericytes and stimulates them to produce VEGF [62;63].

Importantly, the angiopoietin/Tie receptor system is another paracrine signaling pathway associated with pericyte recruitment [64]. Activation of Tie-2 receptors induces the expression of endothelial heparin-binding epidermal-like growth factor (HB-EGF) and hepatocyte growth factor (HGF) which stimulate migration of vSMC and induce their association with ECs [65;66]. Additionally, serotonin has also been described as a mediator of vSMCs recruitment to ECs by Ang-1 in the context of pulmonary hypertension [42]

1.2 Biological function of Ang-1/Tie-2 axis

The Angiopoietin-Tie system is comprised by two type I tyrosine kinase receptors expressed primarily in ECs (Tie-1 and Tie-2) and three secreted ligands (Angiopoietin-1

(Ang-1), Angiopoietin-2 (Ang-2) and Angiopoietin-3 (Ang-3) in mice, and its human orthologue Angiopoietin-4 (Ang-4)) [67]. Several years after the discovery of VEGF,

Tie-1 and Tie-2 receptors were described as orphan receptors critically involved in the formation of the vasculature and the search for their ligands led to the discovery of a growth factor family consisting of both, receptor agonists (Ang-1) and receptor antagonists (Ang-2) [64;68]. This section focuses on the Angiopoietin-Tie system, with a special emphasis on Ang-1 and its regulation of vascular biology.

1.2.1 Pathological angiogenesis and clinical aspects

Various missense hyperactivating mutation in Tie-2 result in venous malformation, demonstrating that Tie-2 signaling pathway is critical in venous morphogenesis [69;70].

No genetic mutations for Ang-1 to 4 or Tie-1 have been found to be involved in human genetic disorder so far. However altered expression of angiogenic factors are involved in various angiogenesis-related diseases. The utmost studied diseases resulting from excessive angiogenesis are cancer, diabetic blindness and hypertension, age-related macular degeneration, rheumatoid arthritis, psoriasis, pulmonary hypertension and asthma while impaired angiogenesis can lead to coronary artery disease, stroke and atherosclerosis-induced cardiovascular and cerebrovascular ischemia, chronic wounds neurodegeneration, hypertension, pre-eclampsia, respiratory distress, osteoporosis and uterine bleeding [4]. Interestingly Ang-1,-2 and Tie-2 expression are altered in many of

35 these angiogenesis-related disease [71]. For example Ang-1 expression is enhanced in rheumatoid arthritis and pulmonary hypertension [72], while Ang-2 levels are increased in diabetic patients [73]. Obviously, many cancer cell lines and tumor tissues, including breast cancer cells and tissues, small cell lung carcinoma [74], cervical cancer [75], and prostate cancer [76], produce Ang-1. Inhibition of Ang-1/Tie-2 signaling results in attenuation of tumor growth in many xenograph models [77-79]. Ang-1, Ang2 and Tie-2 are all overexpressed and involved psiorasis skin and antipsoriatic treatment is accompanied by significant reduction of Ang2 expression [80]. Although impaired angiogenesis can lead to atherosclerosis, atherosclerotic plaques exhibit increase angiogenesis and vascular densities. Hypoxia is one of the most potent stimuli for angiogenesis, and zones of hypoxia are present within thickened atherosclerotic plaque.

Hypoxia-induced angiogenesis is regulated by hypoxia HIF-1 transcription factor.

It is interesting to notice inflammation and angiogenesis are frequently coupled in pathologies such as atherosclerosis, diabetes, asthma and arthritis. As a matter of fact

Ang-1 is a modulator of inflammation and innate immune response. Acute exposure of

EC to Ang-1 leads to a slight but rapid increase in P- selectin expression, resulting in small increases in leukocyte adhesion [81]. Chronic exposure to Ang-1 and VEGF or

TNFα, strongly inhibit leukocyte adhesion to endothelial cells, migration of leukocytes across endothelial cells, and expressions of adhesion molecules and pro-inflammatory mediators [82-87]. The anti-inflammatory properties of Ang-1 also exert a protective effect on airway inflammation and hyperreactivity in an asthma animal model and in diabetic retinopathy [86;88]. Current potential therapies targeting components of the

Ang-1/Tie-2 pathway include: inhibition of Ang2 and administration of Ang-1 in asthma,

Ang2 as an inhibitor of angiogenesis in cancer therapy [89], and Ang-1 in the treatment of ischemia and sepsis [90]. Moreover administration of Ang-1 reduces infarction and attenuates the progression of cardiac dysfunction in a rat model [91]. Ang-1 was shown to prevent and reverse diabetic retinal vascular changes in both new and established diabetes [73]. It appears angiopoietins may be therapeutically useful in a number of pathologies, however they are implicated in vessel remodeling, induction of angiogenesis and inflammation, indicating that strategies to minimize any side effects while optimizing for specificity are likely to be needed. Considering Ang-1/Tie2 signaling axis activates different cross-talking pathways it can explain the difficulty to use Ang-1 or Tie2 as direct therapeutic target or agent. Therapeutic approaches will be more efficient if we have a precise understanding of the mechanisms of action of Ang-1/Tie2 downstream signaling and target the specific pathway responsible for the pathological outcome.

1.2.2 Ang-1/Tie-2 in embryonic development (in vivo)

Targeted disruption of Tie-2 or Ang-1 genes is associated with extensive hemorrhage, detachment of the endothelium, loss of endothelial cell integrity, and failure to recruit pericytes, causing embryonic lethality at days E10-12 [64;92;93]. Ang-1 appears indispensable during early development, but its actions are apparently dispensable in adulthood: specifically, it was found that a deletion of Ang-1 after E13.5 does not cause any overt phenotypes; the animals are viable, fertile and healthy. Cardiac-specific knockout of Ang-1 reproduced the phenotype of the conventional knockout, demonstrating that the early vascular abnormalities arise from flow-dependent defects

[94]. However, when combined with injury or microvascular stress, Ang-1 deficiency

37 resulted in profound organ damage, accelerated angiogenesis, and fibrosis. These findings redefine our understanding of the biological roles of Ang-1: it is dispensable in quiescent vessels but has a powerful ability to modulate the vascular response after injury

[94]. Interestingly Tie-2 mRNA levels are substantially reduced in Ang-1-/- embryos, suggesting that Ang-1 modulates the expression of its own receptor [64].

The specific roles of Ang2 in embryonic vascular development are still undefined.

Transgenic mice overexpressing Ang2 are phenotypically similar Tie-2-/- and Ang-1-/- mice suggests that Ang2 plays a role as an Ang-1 antagonist during vascular development

[95]. Genetic deletion of Ang2 triggers major lymphatic vascular defects, the development of post-natal chylous ascites, and subsequent death [96], suggesting Ang2 plays a critical role in lymphatic vascular development. Ang-1 overexpression is able to rescue lymphatic, but not angiogenic, defects of Ang2-/- mice, suggesting both angiopoietins might function as agonists for Tie-2 receptors in lymphangiogenesis.

Transgenic mice revealed that overexpression of Ang-1 leads to unleaky blood vessels, which were also resistant to leaks cause by inflammatory agents [82]. Moreover overexpression of Ang-1 in mice skin produces larger, more numerous, and more highly branched vessels [97]. In contrast other studies suggest that in mature, quiescent vessels,

Ang-1 production is not required to regulate permeability [94]. In mice embryo, localized

Ang-1 overexpression in the liver elicits significant enlargement and sprouting in the hepatic arterial circulation and induces portal vein dilation [98]. In developed mice, Ang-

1 stimulates in vivo vascular remodeling, vascular enlargement, enhanced wound healing, and increased lymphangiogenesis [82;97;99-102]. Engineered soluble Ang-1, where the

N-terminal region was replaced by Cartilage Oligomeric Matrix Protein coiled-coiled

38 domain (COMP-Ang), has been reported to enhance wound healing in a diabetes mouse model through increased angiogenesis and lymphangiogenesis [100]. COMP-Ang-1 also significantly enhances angiogenesis in a corneal micropocket assay [103]. In vivo, Ang-1 exerts anti-angiogenic activity. In model of cancer the role of Ang-1 also remains controversial. Studies in mice have shown Ang-1 contributes to reduction of xenograft tumor size of various cancerous cell line due to increased coverage and reduced angiogenesis [104;105]. These results are in contradiction with studies in which Ang-1 overexpression increases the growth of human cervical cancers in mice and in a rat glioma model by promoting angiogenesis [75;106]. Ang-1 exact contribution to physiological and pathological angiogenesis is still highly debated and systemic context has to be considered.

Studies in mice carrying mutations on the Tie1 and Tie2 genes revealed that these receptors have non-redundant roles in the vasculature. In mice disruption of the Tie1 gene leads to embryonic lethality by E13.5 due to edema, hemorrhage and defects on vascular integrity [70;102;103]. Furthermore the phenotype of Tie1-/- mice also depends on their genetic strain [102;103]. Tie1-/- mice also show severe defects in the lymphatic vasculature at very early stages of development, suggesting that Tie1 is not only required for normal vascular development but also for embryonic lymphangiogenesis [107]. In contrast, mice carrying targeting null mutations on the Tie2 gene have an earlier lethal phenotype and die by E10.5 [101;103]. These mice have less ECs and distended blood vessels, exhibit abnormal heart development and fail to remodel the primary capillary plexus into a functional vascular network [101;103].

1.2.3 Ang-1/Tie-2 and in vitro angiogenesis

In vitro reductionist models, helped established that Ang-1 inhibits apoptosis triggered by stimuli such as serum deprivation, TNFα, and oxidized low density lipoprotein (ox-LDL)

[7;108-110]. Ang-1 also functions as a chemoattractant promoting endothelial cell migration, as measured by the classical Boyden chamber assay and the wound-healing assay [111-114]. Ang-1 promotes endothelial cell sprouting and differentiation into tube- like structures in 2D and 3D matrices [113;115-117]. The influence of Ang-1 on endothelial cell proliferation is controversial. In murine brain capillary endothelial cells,

Ang-1 elicits strong proliferative responses, whereas in adrenal-cortex-derived microvascular endothelial and human umbilical vein endothelial cells, responses are milder [116;118]. Studies also report no improvement in the proliferative capacity of endothelial cells in the presence of Ang-1 [111;112].

Ang-1 is a well establish and efficient regulator of ECs vascular leakage. Ang-1 triggers mobilization of PECAM-1 to junction areas, inhibition of VE-cadherin phosphorylation, and tightening of EC junctions, resulting in attenuation of VEGF- and thrombin-induced endothelial cell permeability [84]. Ang-1 also suppresses VEGF-induced protein and mRNA levels of adhesion molecule responsible for monocyte adhesion (ICAM-1,

VCAM-1, and E-selectin) [83].

1.2.4 Ang-1/Tie-2 signaling pathways

Ang-1 is able to differentially induce angiogenic responses in sites of vascular remodeling while maintaining stability in mature, quiescent vessels. These opposing responses seem to originate from the differential activation of signaling pathways

40 downstream of Tie-2 receptors between sparse and confluent cells [119;120]. It has been demonstrated that Tie-2 localizes to cell–matrix contacts in sparse cells and to cell–cell junctions in contacting cells upon Ang-1 stimulation [119;120]. In the absence of cell-cell junctions Ang-1/Tie-2 complexes accumulate at the cell edges and in sites of contact with the ECM [120]. Ang-1 co-localizes with Tie-2 at the rear end of sparse cells and preferentially promotes migration through the activation of the downstream of kinase related (Dok-R) signaling pathway [119;120]. In contrast, confluent cells form Ang-

1/Tie-2 complexes in cell-cell contacts allowing Tie-2 receptors to interact with other

Tie-2 receptors located in neighboring ECs. Localization of Tie-2 at cell-cell contacts leads to activation of a particular set of downstream proteins, mainly the pro-survival

PI3-K/AKT signaling pathway [119]. Ang-1 stimulation also increases the expression of vascular endothelial protein tyrosine phosphatase (VE–PTP) at cell–cell junctions and, through the formation of Tie-2/VE–PTP complexes, inhibits vascular permeability [119].

The fact that Ang-1 is able to elicit various cellular responses depending on the context makes it a very interesting regulator of EC biology. Thus, the purpose of this section is to explain what we know about the activation of intracellular signaling pathways by Ang-1 and their differential contribution to survival, angiogenesis, vascular quiescence, inflammation and permeability.

1.2.4.1 Tie receptors

Upon Ang-1 binding, Tie-2 receptors dimerize and cross-phosphorylate multiple tyrosine residues in its cytoplasmic domain. Phosphorylated tyrosines are docking site that can bind to SH2 domains of various scaffold proteins. These docking site are allowing specific activation of different pathways leading to Ang-1 induced angiogenesis or Ang-1

41 anti-inflammatory effects. Vascular endothelial protein tyrosine phosphatase (VE-PTP), is an endothelial cell-specific phosphatase that directly interacts with and regulates the activity of Tie-2 [121;122]. Exposure to Ang-1 protein has been shown to trigger activation of several pathways, the majority of which are mediated through Tie-2 receptors. However, selective pathways are also activated as a result of ligation of various integrins present in Tie-2 microenvironment by Ang-1.

Tie-1 and Tie-2, also known as Tek, are type I transmembrane RTKs first identified as orphan receptors predominantly expressed in ECs. Tie-1 and Tie-2 are also found in primitive hematopoietic cells, outside the vascular compartment of gliomas and skeletal muscle cells [123-126]. The name Tie, tyrosine kinase with immunoglobulin (Ig) and epidermal growth factor (EGF) homology domains, describes the structural properties of their extracellular domain comprised of two amino-terminal Ig-like loops, followed by three EGF homology motifs, a third Ig-like loop and three fibronectin type III repeats close to the transmembrane region [123;127]. Both receptors also contain a carboxyl- terminal tyrosine kinase domain responsible for the activation of intracellular signaling pathways [123;127].

To date, Tie-1 has no known ligands and evidence of its function in ECs is just beginning to emerge. Studies using chimeric receptors formed by the ectodomain of the RTK TrkA fused to the transmembrane and intracellular domains of Tie-1 or Tie-2 have shown that in contrast to TrkA/Tie-2, the TrkA/Tie-1 receptor is unable to autophosphorylate or phosphorylate other intracellular proteins [128]. However, Ang-1 and Ang-4 have been shown to stimulate Tie-1 phosphorylation in EA.hy926 cells and the use of a chimeric receptor formed by the extracellular domain of colony stimulating factor-1 (CSF-1)

42 receptor fused to the intracellular domain of Tie-1 demonstrated that activated Tie-1 can associate with phosphoinositide 3-kinase (PI3-K), phosphorylate AKT and protect cells from apoptosis [128;129].

Current evidence suggests that Tie-1 does not signal via ligand-induced kinase activation, but instead Tie-1 associates with Tie-2 and modulates its activity [128-132]. First it was hypothesized that Tie-1 and Tie-2 are present as pre-formed complexes in ECs and their association is mediated by their intracellular domains. However, the formation of Tie-

1/Tie-2 heteromeric complexes upon treatment with COMP-Ang-1, a soluble Ang-1 chimeric protein more potent than native Ang-1 has also been described [128]. More recently, it has been proposed that Tie-1 and Tie-2 are associated on the cell surface prior to ligand binding and that the ability of individual angiopoietins to effectively destabilize

Tie-1/Tie-2 complexes define their respective agonistic or antagonistic roles [132].

Strikingly, reducing Tie-1 protein levels in human umbilical vein endothelial cells

(HUVECs) using small hairpin RNA (shRNA) increased Ang-1 mediated Tie-2 phosphorylation and allowed Ang-2 to strongly activate Tie-2 receptors [132].

Tie-2 receptors are able to bind to all angiopoietins [64;95;133]. Mutagenesis and binding studies originally identified the first Ig loop and EGF repeats of the Tie-2 extracellular domain as necessary for angiopoietin recognition and binding [134].

However, more recent crystallographic analysis of the Tie-2 receptor ectodomain and

Ang-2/Tie-2 complexes revealed that the ligand binding site is the second Ig domain

[135]. Despite being able to bind all angiopoietins with similar affinities, the effects of these ligands on Tie-2 receptor activation are different [64;95]. Ang-1 consistently activates Tie-2 receptors, whereas Ang-2 seems to act in a context-dependent manner as

43 either agonist or antagonist of Tie-2 [64;95;136-138]. The less studied Ang-4 seems to activate Tie-2 similarly to Ang-1 [133]. Tie-2 activation requires ligand multimerization and the formation of ligand-receptor clusters at sites of cell-cell or cell-extracellular matrix (ECM) contacts, which induce Tie-2 receptor autophosphorylation in tyrosine

(Tyr) residues near its catalytic carboxyl-terminal end and create high affinity binding sites for signaling molecules and adaptor proteins that recognize phosphotyrosine motifs

[119;139].

1.2.4.2 Angiopoietins

Ang-1 is an agonist of Tie-2 receptors [64;140]. The DNA sequences of both human and mouse Ang-1 cDNA clones share 97.6% identity and encode a glycosylated protein of

498 amino acids that specifically binds to Tie-2 receptors in ECs and induces their

Tyrosine phosphorylation [141]. Ang-1 is constitutively expressed in many adult tissues by mural cells, fibroblasts, ECs, smooth muscle cells, pericytes, and their precursors and exerts paracrine effects on Tie-2-expressing cells [64;141].

Structurally, Ang-1 is a glycoprotein comprised of a carboxyl-terminal fibrinogen-like domain, a central coiled-coiled domain, a short amino-terminal superclustering domain and a secretory signal sequence [141-143]. The fibrinogen-like domain is responsible for receptor binding, whereas the coiled-coiled domain is responsible for oligomerization of the fibrinogen-like domains and the superclustering domain allows these oligomers to organize into multimers of variable size [139;142;143]. The proper formation of Ang-1 oligomers and multimers by disulfide links involving the cysteines 41 and 54, located in the superclustering domain, is crucial for Tie-2 receptor binding and activation [139]. The minimal unit able to activate Tie-2 receptors is a tetrameric form of Ang-1, and oligomers

44 unable to form multimers have a reduced activation of Tie-2 receptors [139;143]. Binding of Ang-1 to Tie-2 receptors induces their autophosphorylation and allows them to interact with different protein effectors and activate several downstream signaling pathways

[6;53]. Ang-1 has several distinct functions in the vasculature of the embryo and in adult tissues [6]. The intracellular signaling pathways activated by Ang-1, as well as their functional significance for EC biology are discussed in detail in the following section.

1.2.4.2.1 Angiopoietin-2

Angiopoietin-2 (Ang-2) has been described as an antagonist of Tie-2 receptors, but it can also act as an agonist depending on the context [95;131;137;138]. The angiopoietins are structurally similar and Ang-2 conserves ~60% amino acid identity with Ang-1 [95].

Both proteins have an amino-terminal coiled-coil domain and a carboxyl-terminal fibrinogen-like domain, although Ang-2 lacks a cysteine between the coiled-coil and fibrinogen-like domains [95;140;142;143]. Ang-2 is usually found in dimeric form, and the differences in protein structure compared to Ang-1 restrict the protein to the formation of less order multimers [143]. Ang-2 is secreted by ECs and in the normal adult its expression is restricted to sites of active vascular remodeling such as the ovary, placenta and uterus [95].

Embryos of transgenic mice that specifically overexpress Ang-2 in their vasculature die at E9.5–10.5 and exhibit an abnormal vascular phenotype similar to mice lacking Ang-1 or Tie-2, although their defects appear to be more severe [64;92;95]. Ang-2 expression leads to heart abnormalities and to the formation of a discontinuous vascular network

[95]. The defects observed in these mice suggest that Ang-2 has antagonistic functions to

Ang-1, which is further supported by the pattern of expression of Ang-2 in adult tissues at

45 sites of active vascular remodeling and by the fact that Ang-2 can also block Ang-1- induced activation of Tie-2 receptors [95]. However, Ang-2 can also act as an agonist and activate Tie-2 receptors to induce similar signaling pathways and cellular responses as

Ang-1 depending on the context [136-138].

In addition, Ang-2 promotes destabilization of the vasculature by antagonizing Ang-1- induced recruitment of mural cells to ECs [144;145]. In diabetic retinopathy Ang-2 causes loss of pericyte coverage and destabilizes the retinal capillaries in the absence of

VEGF [145]. Furthermore, in a transgenic model in which Ang-2 is expressed in ECs the restoration of blood flow after limb ischemia is drastically impaired, an effect mainly attributed to defective recruitment of vSMC in Ang-2 transgenic mice [144].

Ang-2 also primes the vasculature to potentiate its response to angiogenic and inflammatory stimuli. The local cytokine environment largely determines the outcome of

Ang-2 signaling [146-148]. In the presence of VEGF, Ang-2 induces migration, proliferation and sprouting of blood vessels; while in the absence of VEGF Ang-2 favors cell death and vessel regression [146-148]. In the case of inflammation, Ang-2 deficient mice are unable to elicit a rapid pro-inflammatory response in response to TNF- and fail to induce adhesion molecule expression [149]. These findings stress the antagonistic mode of action of Ang-2 on the vasculature, which clearly oppose the quiescent and anti- inflammatory effects of Ang-1 in ECs [82;149] .

1.2.4.2.2 Angiopoietin-3/4

Angiopoietin-4 (Ang-4) and its mouse orthologue angiopoietin-3 (Ang-3) share all the structural characteristics of angiopoietins and are able to bind to Tie-2 receptors, but not

46 to Tie-1 [133]. Ang-4 was first identified as an agonist of Tie-2 receptors, whereas Ang-3 was described as an antagonist [133;150]. Evidence later demonstrated that Ang-3 can also phosphorylate Tie-2 receptors on ECs of its own species [150]. Ang-4 and Ang-3 phosphorylate AKT on Serine 473, promote survival of ECs and induce corneal angiogenesis in vivo [150]. Additionally, Ang-4 has been found to protect against serum deprivation-induced apoptosis, to increase migration and to promote capillary tube formation in cultured ECs through the activation of Tie-2 receptors [151]. Studies carried out in Lewis lung carcinoma (LLC) cells have demonstrated that Ang-3 can bind to the cell surface or localize to the basement membrane via perlecan, a HSPG [152].

Furthermore, cell bound Ang-3 is able to induce retraction and loss of integrity in bovine pulmonary artery EC monolayers [152]. Ang-3 and Ang-4 are expressed in response to hypoxia, highlighting their role in vascular remodeling. Moreover, it has been proposed that Ang-4 functions similarly to Ang-1 during hypoxia-induced angiogenesis [151].

1.2.4.3 The phosphoinositide 3-kinase pathway

PI3-kinases are a ubiquitous family of lipid kinases that generate 3'-phosphorylated phosphoinositides (PIs) [153]. PI3- kinases are categorized into three main classes (I, II, and III), based on sequence homologies, regulation, and lipid specificities [153]. PI3- kinases are heterodimeric proteins containing a regulatory subunit and a catalytic subunit

[154]. Four functionally different lipid products, namely PtdINs- 3-P, PtdIns-3,4-P2,

PtdIns-3,5-P2, and PtdIns-3,4,5-P3, act as docking sites for proteins containing a PH domain and are involved in an array of cellular functions including inflammation, migration, proliferation, survival, and glucose metabolism [155;156]. In response to growth factors, class I enzymes generate PtdIns-3,4,5-P3 [157]. These enzymes are

47 composed of a p85 regulatory subunit and a p110 subunit that contains a lipid-directed catalytic domain. The p85 subunit is recruited to receptor phosphotyrosines through its

SH2 domain and in turn recruits the catalytic subunit to activate tyrosine kinase receptors

[154]. Class I PI3-kinase activity is terminated through dephosphorylation of PtdIns-

3,4,5-P3 mediated by two inositol phosphatases, SH2-containing inositol polyphosphate

5-phosphatase (SHIP) and phosphoinositide-lipid 3-phosphatase (PTEN) [158]. Class I enzymes can activate the mammalian target of rapamycin (mTOR)/p70S6 kinase (S6K) signaling pathway, which is involved in the regulation of cell growth, insulin metabolism, and protein synthesis [159;160]. In endothelial cells the class I PI3-kinase pathway is activated downstream from Tie-2 receptors in response to Ang-1 and Ang-2, and this pathway promotes Ang-1-triggered endothelial cell migration, adhesion, differentiation, and survival [109;110;161]. Ang-1 via PI3K/AKT(PKB)/FKHR initiates expression of

Survivin and promotes endothelial cells survival and represses expression of Ang-2 to further stabilize the vasculature [11;110]. The activation of the PI3-K/AKT signaling pathway is largely responsible for the anti-apoptotic effect of Ang-1 in ECs [108-110]. In

ECs activation of the PI3-K/AKT pathway by Ang-1 prevents apoptosis caused by serum deprivation through the induction of survivin and the inhibition of caspase-3, -7 and -9 activities [108;110] . Additionally, Ang-1 activation of Tie-2 in confluent ECs induces the expression of Krüppel-like factor 2 (KLF2), a zinc finger transcription factor associated with vascular quiescence, downstream of PI3-K/AKT pathway [10;162-164].

Ang-1 induced activation of the PI3-K/AKT pathway stimulates transcriptional activity of myocyte enhancer factor 2 (MEF2) to induce the expression of KLF2 in ECs [10].

Although the activation of PI3-K in quiescent cells favors survival, its activation can also

48 contribute to the angiogenic cascade in activated ECs. Ang-1 is able to induce endothelial sprouting, partly through cytoskeletal changes and proteinase secretion dependent on the activation of PI3-K [116;117]. Ang-1 treatment in porcine pulmonary artery ECs induces phosphorylation of focal adhesion kinase (FAK), and secretion of plasmin and matrix metalloproteinase-2 (MMP-2); both inhibited by pharmacological inhibitors of PI3-K

[117]. Ang-1 treatment also promotes migration, capillary tube formation and sprouting partly through eNOS phosphorylation and NO production [113]. Pharmacological inhibition of the PI3K/AKT pathway reduces Ang-1-stimulated eNOS phosphorylation and NO production; and completely abrogates the effect of Ang-1 on angiogenesis [113].

1.2.4.4 NFκB and the inflammatory phenotype

The inflammatory activation of ECs leads to the recruitment of leukocytes to the sites of injury due to an increase in pro-inflammatory mediators and adhesion molecules expressed largely in response to the activation of the NFκB signaling pathway [5680]. It is suggested that Ang-1 exerts its anti-inflammatory effect by selective inhibition NFκB through Tie-2 receptors and that this effect is meditated by direct protein-protein interactions between Tie-2 receptors, A20 binding inhibitor of NFκB activation-2 (ABIN-

2), and NFκB protein subunits [14;165]. In turn, inhibition of NFκB will prevent transcription of pro-inflammatory genes. VEGF is known to promote leukocyte adhesion through the induction of E-selectin, VCAM1 and ICAM1 [166]. In contrast to the pro- inflammatory effects of VEGF, Ang-1 inhibits leukocyte adhesion, the expression of adhesion molecules and, as mentioned earlier, VEGF-induced vascular leakage

[10;83;167;168]. Some of the mechanisms through which Ang-1 exerts these effects are

49 beginning to emerge, particularly the role of nuclear receptor-77 (Nur77) and the transcription factor KLF2 in mediating this anti-inflammatory response [10;169]

Nur77, a member of the nuclear receptor subfamily of ligand-independent nuclear receptors, is rapidly expressed in response to growth factors and cytokines [170]. Nur77 interacts with the p65 subunit of NF-κB and interferes with its ability to bind promoters of pro-inflammatory cytokines, particularly interleukin 2 (IL-2) [171]. In ECs the expression of Nur77 increases in response to TNF- as a negative feedback mechanism of the pro-inflammatory response and leads to inhibition of NF-κB activation by transcriptionally upregulating IkappaB alpha (IκBα) [172]. Furthermore, inhibition of

Nur77 increases TNF-α induced adhesion molecule expression and leukocyte adhesion

[172]. In ECs Ang-1 also induces Nur77 expression and when combined with VEGF, its expression is potentiated [169;173]. Ang-1 induction of Nur77 is mediated by the PI3-K and ERK1/2 signaling pathways, which differs from the mechanism described for VEGF- induced Nur77 [169;173]. The inhibition of VEGF-induced NF-κB, VCAM1 and E- selectin expression, and leukocyte adhesion by Ang-1 disappears when Nur77 expression is disrupted, stressing the role of this protein in the anti-inflammmatory effects of Ang-1 on the vasculature [169].

Ang-1 induces the expression of the transcription factor KLF2 in ECs [10]. KLF2 is not only involved in vascular quiescence but also mediates the inhibitory effect of Ang-1 on

VEGF-induced expression of VCAM1 and monocyte adhesion to ECs [10].

1.2.4.5 The MAPKs

The mitogen activated protein kinases (MAPKs) are a family of conserved protein kinases controlling a large number of cellular processes including proliferation,

50 migration, survival and differentiation in response to extracellular stimuli [174;175].

MAPK signaling cascades are modules comprised of three kinases sequentially activated by phosphorylation: MAPK kinase kinase (MAPKKK), MAPK kinase (MAPKK) and

MAPK [174;175]. These MAPK modules can be activated upon stimulation of extracellular receptors or in response to physical stimuli by small guanosine triphosphate

(GTP)-binding proteins, such as Ras and Rho GTPases, or by STE20 kinases [174-176].

The spatiotemporal activation of MAPK signaling determines the type of cellular response obtained, stressing the importance of the molecular mechanisms involved in the inactivation of these signaling cascades [177]. MAPK signaling can be terminated by inhibiting Ras or Rho activity through GTPase-activating proteins (GAPs) which catalyze the exchange of GTP for guanosine diphosphate (GDP) [178]. Another mechanism through which MAPK signaling can be terminated is the dephosphorylation of Tyr and

Ser/Thr residues mediated by phosphatases [179].

There are four typical MAPK signaling pathways in mammals: extracellular signal- regulated kinase 1 and 2 (ERK1/2), c-Jun N-terminal kinase 1, 2 and 3 (SAPK/JNK), p38

(,,,) and ERK5 [174;175]. ERK1 and 2 are activated mainly by cell surface receptors, such as RTKs and GPCRs, in response to growth factors and cytokines [180].

In the case of RTKs, ligand binding induces receptor clustering and autophosphorylation of Tyr residues located in their cytoplasmic domain. These phosphorylated residues act as binding sites for scaffolding and signaling proteins containing Src homology 2 (SH2) or

PTB domains including growth factor receptor-bound protein 2 (Grb2) which recruits son of sevenless (SOS) to the plasma membrane [181]. SOS is a guanine nucleotide exchange factor (GEF) that stimulates the exchange of GDP bound to the small GTPase Ras to

GTP, which is required for a positive regulation of Ras activity [178]. Activated Ras is able to interact directly with its effectors, the MAPKKKs A-Raf, B-Raf, and Raf-1, to induce their activation [174]. Activated MAPKKKs phosphorylate the MAPKKs MEK1 and MEK2, which in turn phosphorylate the MAPKs ERK1/2 within a conserved

Threonine-Glutamate-Tyrosine (TEY) motif located in their activation loop [174;175].

Phosphorylated ERK1/2 accumulates in the nucleus, where it can phosphorylate a large number of transcription factors such as nuclear factor of activated T cells (NFAT), Elk-1,

MEF2, c-Fos, c-Myc and signal transducer and activator of transcription (STAT) 3

[181;182]. ERK1/2 can also phosphorylate cytoplasmic substrates such as death- associated protein kinase (DAPK) and tuberous sclerosis complex 2 (TSC2); and substrates associated with the cell membrane like Syk, and calnexin, or the cytoskeleton such as paxillin [181;183]. ERK1/2 mediates cell growth, proliferation, differentiation, survival and cytoskeletal changes through the activation of its multiple substrates

[181;183].

Unlike ERK1/2, the MAPKs p38 and SAPK/JNK are mainly activated by inflammatory cytokines and environmental stresses such as DNA damage, hypoxia, oxidative stress and ultraviolet (UV) light [181;184;185]. These kinases are activated by Rac and Cdc42,

GTPases of the Rho family, downstream of RTKs and GPCRs; and by the recruitment of

TNF receptor associated factor (TRAF) proteins in response to pro-inflammatory cytokines such as TNF- and IL-1 [181;184;186]. The recruitment of TRAF proteins promotes the activation of downstream MAPKKKs, many of which are shared by the p38 and SAPK/JNK pathways [181].

Four genes encode the different isoforms of the p38 family of MAPKs [184]. p38α and p38β are widely expressed in cell lines and tissues, whereas p38γ and p38δ show a more restricted pattern of expression [181]. The major MAPKKs upstream of p38 are MKK3 and MKK6 [181;187]. MKK3 and MKK6 are activated by several MAPKKKs, including

MEKK1-3, ASK1 and TAK1 [184]. Upon stimulation, p38 isoforms phosphorylate a large number of substrates including calcium-dependent phospholipase A2 (cPLA2),

MNK1/2 and Bcl-2-associated X protein (Bax) in the cytoplasm; and activating transcription factor (ATF) 1 and 2, MEF2, Elk-1, Ets1 and p53 in the nucleus [184].

Activation of p38 plays a key role in immune and inflammatory responses, cell cycle progression and apoptosis [184;188].

There are three known isoforms of SAPK/JNK encoded by three distinct genes:

JNK1/SAPKγ, JNK2/SAPKα and JNK3/SAPKβ [181;185]. JNK1 and JNK2 exhibit a wide tissue distribution, whereas JNK3 localizes mainly in neurons, testis, and cardiac myocytes [189]. The major MAPKKs that activate SAPK/JNK isoforms are MKK4 and

MKK7 [181;185]. MKK4/7 are phosphorylated and activated by several MAPKKKs, including MEKK1, ASK1/2 and TAK1 [181]. Upon activation SAPK/JNK accumulates in the nucleus where it can interact with its substrates [190]. Some of the transcription factors that have been shown to be phosphorylated by SAPK/JNK include c-Jun, p53,

ATF2, NFAT, Elk-1, STAT3, c-Myc and JunB [181;191]. SAPK/JNK plays an important role in the control of cell proliferation via c-Jun and the transcription of genes that control the cell cycle, such as cyclin D1 [192]. SAPK/JNK phosphorylation of c-Jun on Ser63 and/or Ser73 increases c-Jun-dependent transcription of genes containing AP-1-binding

53 sites [181;192]. SAPK/JNK has also been implicated in the differentiation of hematopoietic cells and in apoptosis induced by cellular stresses [193].

Several pro-angiogenic factors including VEGF, FGF and PDGF are known to activate

ERK1/2 and induce proliferation in ECs [194;195]. VEGFR signaling can also activate the p38 and SAPK/JNK pathways to promote endothelial migration in various types of

ECs including microvascular, aortic and umbilical vein [196]. Similarly, Ang-1 signaling activates the MAPK signaling pathways ERK1/2, p38, and SAPK/JNK simultaneously in

ECs [6;7;12;18;109;118;197]. In contrast to VEGF signaling, the mitogenic effect of

ERK1/2 activation by Ang-1 is relatively modest [118]. In Ang-1 signaling ERK1/2 activation is more a pro-survival pathway involved in the anti-apoptotic effect of Ang-1 on ECs, rather than an inducer of proliferation [18]. Ang-1 also elicits activation of p38 and the inhibition of this pathway using pharmacological inhibitors results in a potentiated effect of Ang-1 inhibition of apoptosis, which confirms the pro-apoptotic role of p38 in this context [18]. The activation of SAPK/JNK by Ang-1 is also a pro-apoptotic pathway [7]. However, despite the activation of the pro-apoptotic pathways p38 and

SAPK/JNK, Ang-1 strongly inhibits endothelial apoptosis induced in response to serum deprivation. This suggests that the effect of the anti-apoptotic pathways ERK1/2 and PI-

3K predominate over the pro-apoptotic MAPKs [7;18].

In addition to its effects on cell survival, MAPK activation in response to Ang-1 affects angiogenic responses [12;16]. Ang-1-induced activation of ERK1/2 and SAPK/JNK, in combination with PI3-K activation, stimulates production of interleukin 8 (IL-8) in ECs

[12]. IL-8 is a chemokine not only associated with the infiltration of inflammatory cells at sites of injury, but also with angiogenesis [198]. Ang-1 signaling promotes the

54 transcription of the IL8 gene through the AP-1 transcription factor by inducing phosphorylation of c-Jun on both Ser63 and Ser73 downstream of ERK1/2 and

SAPK/JNK. IL-8 production partly mediates the migratory and proliferative effect of

Ang-1 on ECs [12]. In addition to AP-1, MAPK activation in response to Ang-1 also leads to the induction of another transcription factor termed early growth response-1

(Egr-1) [16]. Egr-1 is a zinc finger transcription factor and an immediate early response gene rapidly induced by various stimuli including growth factors, cytokines, hypoxia and shear stress [199;200]. The Egr-1 transcription factor has been involved in regulating angiogenesis by inducing expression of various growth factors including FGF [201;202].

In the case of Ang-1, Egr-1 is transiently induced in ECs downstream from ERK1/2 and

PI3-K signaling pathways and plays a role in Ang-1-mediated endothelial migration and proliferation of ECs, albeit the molecular mechanisms behind these effects remain largely unknown [16].

1.2.4.5.1 Dual specifity phosphatases as modulators of the MAPKs

Controlled activation of intracellular signaling cascades largely depends on the reversible phosphorylation of proteins, lipids and other small molecules [203]. Phosphatases are enzymes with the ability to hydrolyze the phosphoester bonds on their substrates.

Phosphatases were discovered several years after protein kinases and although less studied, they are also powerful controllers of intracellular signaling cascades [204;205].

Dual-specificity phosphatases (DUSPs) are a subclass of phosphatases able to specifically dephosphorylate both serine/threonine and tyrosine residues on their targets, which include MAPKs [206;207]. Previously, we described how Ang-1 mediates some of its

55 biological effects through the activation of ERK1/2, p38 and SAPK/JNK including survival, migration and proliferation of ECs [6;7;18;197]. More recently Dr Hussain’s laboratory identified DUSPs involved in Ang-1 induced regulation of MAPKs in an effort to find new ways to manipulate MAPK-dependent responses in the context of angiogenesis. This study was the first to show Ang-1 differentially induces DUSP1,

DUSP4, and DUSP5 in ECs through activation of the PI-3 kinase, ERK1/2, p38, and

SAPK/JNK pathways. This study also revealed by lack-of-function siRNA screening that

DUSP1 preferentially dephosphorylates p38 protein and is involved in Ang-1-induced cell migration and differentiation. DUSP4 preferentially dephosphorylates ERK1/2, p38, and SAPK/JNK proteins and, under conditions of serum deprivation, is involved in Ang-

1-induced cell migration, several antiapoptotic effects, and differentiation. DUSP5 preferentially dephosphorylates ERK1/2 proteins and is involved in cell survival and inhibition of permeability [208].

1.2.4.6 The downstream of kinase related protein pathway

In ECs Dok-R has been described as a protein associated with Tie-2 receptors involved in further amplifying Ang-1 signaling by providing various docking sites for downstream signaling proteins R [1433]. Dok-R, has been shown to interact with the tyrosine residue of Tie-2 receptors leading to recruitment of the Nck adaptor protein, RasGAP, PAK-1, resulting in increased EC migration [209;210]. The recruitment of PAK-1 to the receptor results in the reorganization of the actin cytoskeleton and is required for Ang-1-mediated EC migration [209]. Ang-1 exposure in HUVECs triggers significant increases in PAK-1 phosphorylation and this phosphorylation is associated with induction of endothelial cell

56 migration, further confirming involvement of the DOK-R/Nck/PAK-1 pathway in the biological response to Ang-1 [7]. Other than the role of Dok-R in cell migration, the function of this signaling pathway in relation to the Ang-1/Tie-2 signaling in ECs remains largely unknown. However, it has been suggested that Dok-R activation could be switching the EC response more towards a migratory state, rather than a pro-survival and quiescent state; which correlates with its localization at rear end of migrating cells alongside activated Tie-2 receptors [119].

1.2.4.7 The role of integrins in angiopoietin signaling

In addition to Tie receptors, Ang-1 and Ang-2 can also bind to different integrins including α2β1, α5β1, αvβ3, and αvβ5 [211-214]. Integrins are heterodimeric membrane glycoproteins responsible for outside-in and inside-out signaling that modulate cell adhesion and migration [215;216]. The integrin family is comprised of structurally related receptors for ECM proteins and Ig superfamily proteins [216]. Some integrins can also bind cell surface receptors like VCAM1 to promote cell–cell adhesion and the activation of intracellular signaling pathways [217;218].

Ang-1 can bind directly to α5β1 integrins through its fibrinogen-like domain, although the precise peptide sequence mediating this interaction remains elusive [213].

Interestingly, Ang-1 interaction with α5β1 is inhibited by RGD peptides despite the fact that the fibrinogen-like domain of Ang-1 lacks a RGD motif, or any known binding sites to integrin [211;212]. It has been suggested that the conserved peptide sequence

Glutamine-Histidine-Arginine-Glutamate-Aspartate-Glycine-Serine (QHREDGS) within the fibrinogen-like domain of Ang-1 could mediate its binding to integrins due to its

57 similarity to integrin-binding sequences found in fibrinogen and fibronectin [212]. Ang-2 also binds and activates integrins, particularly in non-vascular cells such as myocytes, glioma and breast cancer cells [89;211;219]. In ECs, Ang-2 has been found to co- immunoprecipitate with α5β1 integrin after stimulation with TNF-α [220]. Furthermore, in activated endothelial tip cells Ang-2 induces FAK phosphorylation on Tyr397 in an integrin-dependent manner and stimulates EC migration and sprouting [214].

1.2.4.8 Nuclear modulators of Ang-1/Tie2 induced angiogenesis

Little is known regarding the nature of various transcription factors activated by the

Ang-1/Tie-2 pathway. KLF2 [10], FOXO1 [11], AP1 [12], GATA3 [13], NFκB [14],

STAT [15], EGR1 [16] and 3 members of ETS family: NERF2 [17], ELF1 [13] and Elk1

[18], are TFs reported to be regulated by Ang-1/Tie2 axis. Even less is known on the signaling mechanism by which these transcription factors are transducing Ang-1 signal.

Elk-1 is a member of ETS-oncogene family that forms a ternary complex by binding serum response factors and serum response elements in various promoters. Ang-1 induces Elk1 phosphorylation downstream from Erk1/2. A study linked the PI-3 kinase,

ERK1/2, and SAPK/JNK pathways to the activation of AP-1 and associated this transcription factor with induction of IL8 release in endothelial cells exposed to Ang-1

[12]. Activation of KLF2 in endothelial cells exposed to Ang-1 is linked to the PI3-kinase pathway. Moreover, it was found that deletion of KLF2 eliminates inhibitory effects of

Ang-1 on VEGF-induced adhesion molecule expression and monocyte adhesion, suggesting KLF2 mediates anti-inflammatory effects of Ang-1 [10]. Additionally,

HDAC7 and Nur77 are nuclear modulator of Ang-1 induced inflammation [169]. Ang-1

58 was shown to potentiate VEGF-induced Nur77 expression. Ang-1 induces Nur77 through the PI3K and ERK 1/2 pathways, promoting the anti-inflammatory effects of Ang-1 by initiating a negative feedback loop on VEGF induced ECs activation [169]

1. 3 Reactive oxygen species and NADPH oxidase in angiogenesis

1.3.1 Introduction

Accumulating evidence suggest that intracellular reactive oxygens species (ROS) regulate cytokine and growth factor signaling [221-223]. ROS are highly reactive molecules containing oxygen: hydrogen peroxide (H2O2), superoxide (O2),

. 1 hydroxyl radical ( OH) and singlet oxygen ( O2). In a cellular context, ROS are produced in the cytoplasm by various enzymatic system, including Nadph oxidase (NOX) and xanthine oxidase but also the mitochondrial electron transport chain [224]. ROS act as signaling molecules in growth factor mediated physiological responses by interacting with proteins, lipids, carbohydrates and nucleic acid and altering the function of the target molecules. Pro-angiogenic growth factors such as platelet derived growth factor (PDGF), epidermal growth factor (EGF), vascular endothelial growth factor (VEGF) and Ang-1 ligand binding stimulate a burst of ROS production required for normal MAPK signaling downstream of these receptor tyrosine kinases (RTKs) [7;8;225-228].

NOX-derived ROS are key molecular players in angiogenesis through the regulation of different biological systems such as VEGF, Ang-1, angiotensin II and AP-1 [7;225;229-

231]. The main focus of this study, therefore, was to characterize the ways in which

NOX2, NOX4 and mtROS regulate Ang-1 induced MAPK signaling and to investigate how they influence Ang‐1‐induced EC survival, migration and differentiation. This study

59 further expands our understanding of how Ang-1 fine tunes it’s signaling and how it specifically contributes to the different step of the angiogenic paradigm. This study should also provide significant insight on new effective strategies to modulate different outcomes of Ang-1/Tie-2 signaling pathway.

1.3.2 NADPH oxidase: an overview

The proper activation of intracellular signaling cascades largely depends on the reversible modification of proteins, lipids and other small molecules. ROS oxidative modifications on proteins, lipids RNA and DNA were for a long time, considered toxic for the cells and tissues. The concept of ROS being involved in regulating cell signaling events is now widely accepted in the scientific community. In large amount and during chronic exposure, ROS can be toxic for the biological systems. When ROS are released in discreet amount they contribute to the oxydo-reductive balance in the cells. The cells comprise various Redox systems to maintain this oxydo-reductive homeostasis. ROS are produced by various metabolic reaction and are a by-product of cellular respiration. They

− comprise the superoxide anion (O2 ), hydrogen peroxide (H2O2), and the hydroxyl radical

(OH ) [232]. They contribute to a wide array of signaling pathway and biological outcome, including differentiation, proliferation, autophagy, migration to name a few.

ROS targeting specificity arises from the properties of the different ROS generating enzymes and the specific ROS they produce. They showcase different reactivity, stability, diffusion capacity and localization [232]. More precisely, The core member of the NOX family, NOX2, referred to as gp91phox or phagocytic NOX, is a transmembrane protein responsible for transporting electrons across biological membranes, which leads to the

− reduction of oxygen into O2 [233]. Endothelial cells predominantly express NOX4, and

NOX2 to a lower extent [234;235].

1.3.2.1 Vascular NADPH oxidase structure, localization and function

NADPH oxidase (NOX) is a multi-subunit enzyme comprising membrane and cytosolic components, which actively communicate upon a wide variety of stimuli, including growth factor receptor activation. Initially thought to be express only in phagocytic cells

7 new members of the NOX family have been identified and are express in various tissue:

NOX1 to NOX5 and DUOX1 and DUOX2. Except for NOX3, all other NOXs and

DUOXs are expressed in the vasculature and in VSMCs. These enzyme have a common function; the reduction of oxygen to superoxide which can be attributed to the presence of conserved structural domain. All NOX enzymes comprise a NADPH-binding site, a

Flavin adenine dinucleotide-binding region, six conserved transmembrane domain and four conserved heme-binding histidines [236]. Typically NOX2 and p22, two integral membrane protein assemble to form the heteromeric subunit referred to as

phox phox flavocytochrome b558 (cyt b558). The other regulatory subunits, p40 , p47 and p67phox, exist in the cytosol as a complex. Upon stimulation, p47phox is phosphorylated, and the regulatory complex translocate to the membrane and associates with cyt b558 to activate the enzyme [237]. The functional enzyme transfers electrons from its substrate, oxygen, through a prosthetic group, flavin and a heme group, carrying the electrons. The complex require two small GTPases, Rac2 and Rap1A [238]. Under unstimulated conditions, Rap1A is a membrane protein and Rac2 is maintained inactive in the cytosol by its interaction with Rho-GDI (guanine nucleotide dissociation inhibitor) [239;240].

Upon activation Rac2 binds GTP and translocates to the membrane with the other cytosolic complex.

NOX2 requires interaction with five regulatory subunits for optimal activation. NOX2 associates with p22phox, p47phox, p67phox, p40phox and Rac1/2 [241], it localizes at the plasma membrane, in phagosome, in redoxisomes and at the leading edge of lamelipodia [235;242;243]. In vascular cells, NOX2 activity is believed to be inducible: it is activated by Ang II, endothelin-1, VEGF, TNFα, and mechanical forces

[241;244;245].

NOX4 differs from typical NOXs. It localizes at focal adhesion and intracellular membranes of nucleus, mitochondria and endoplasmic reticulum [235;246]. Although it was shown that NOX4 activity was required for VEGF induced cell migration [244],

NOX4 is believed to be constitutively active and regulated at the gene level and only requires association with p22phox for its activity [247]. More recent evidence showed that NOX4 might interact with other regulatory protein for optimal activity or for specific localization. It has been shown that NOX4 interacts with Poldip2. Association with

Poldip2 increases NOX4 activity by 3-fold. Nox4-Poldip2-derived ROS activate RhoA to strengthen focal adhesions and increase stress fiber formation. Interestingly, the overexpression or the depletion of Poldip2 inhibits migration of cultured smooth muscle cells, suggesting a central role for Nox4-Poldip2 in synchronizing the process of directed cell migration [248]. NOX4 differs from the other members by another characteristic.

NOX4 mainly produces H2O2 as a byproduct of oxygen reduction with only 10% of superoxide [249]. Another difference with NOX4 is that most NOXs enzymes are localized at least in part at the cell surface where they can release ROS into the

62 extracellular milieu, while NOX4 is localized to internal membranes such as the endoplasmic reticulum, nuclear membrane and mitochondria [235;250-252].

1.3.2.2 Regulation of NOXs in endothelium

Increase production and release of NOX-derived ROS, associated with the up-regulation of different NOX subtypes, contributes to vascular pathologies, including cardiovascular diseases, diabetes, obesity, cancer, and neurodegeneration. Therefore, understanding the regulation mechanisms of NOX enzymes should contribute to find ways to control the formation of NOX-derived ROS in various pathological states.

It is well established that the expression as well as activity of various NOX leading to increase in ROS production are upregulated by pro-inflammatory cytokines, growth factors, hormones, vasoactive agents, metabolic intermediates, modified lipids and lipoproteins in different cardiovascular cell. First and foremost NOXs can be activated in several ways, including phosphorylation of cystosolic regulatory subunits by PKC, PKA or PI3K, MAPK and non receptor associated protein kinases (JAK and SRC) [253-258].

Anothoer regulatory mechanism of the NOXs enzymes is through protein–protein interactions among NOXs members and members of the thioredoxin family. More over transient oscillations in intracellular concentration of various ions may trigger NOXs activation [259-261].

Secondly, enhanced level of NOXs expression is an essential mechanisms responsible for excessive and sustained release of ROS in the ECs. Upregulation of NOXs expression depends on several transcription factors. Activated NF-kB, AP-1, and STAT have been correlated with enhanced expression of NOXs within atherosclerotic lesions and in the

63 vascular wall of diabetic and hypertensive patients as well as in various animal models [262-265]. Interestingly these transcription factors participate in a signaling loop and their activity is regulated in part through NOX-derived ROS.

In addition to activation and transcriptional regulation, NOXs can be modulated through epigenetic mechanism. Epigenetic pathways are implicated in the regulation of NOXs levels and function in various cell types [266]. More specifically, it has been demonstrated that inhibition of HDAC reduces the transcriptional activity and expression of NOX in human ECs [267]. In this study, increased acetylation of histones following the inhibition of HDACs prevented the binding of AP-1 transcription factor and RNA polymerase 2A to NOX4 promoter. Others reported that the gene and protein expression levels of NOX4 were significantly reduced by HDAC class I inhibitors.

More recently, it has been shown that miRNA participates in the regulation of NOXs. It has been demonstrated that miRNA-25 as well as miR-21a-3p target the 3′UTR of

NOX4, leading to NOX down-regulation and consequent reduction in ROS production [268;269]. Aside from direct post-transcriptional regulation of NOX by miRNAs, recent studies emphasized that ROS, possibly NOX-derived ROS regulate the expression and specificity of several miRNAs [270;271].

Besides these regulatory mechanism, NOXs are also sensitive to redox regulation. Both

NOX2 and NOX4 ROS release is regulated by hyperoxia as well has hypoxia [272]

1.3.2.3 NOX2 and NOX4 in angiogenesis

NOX4 and NOX2 have been found to be involved in angiogenesis. NOX2 is activated in endothelial cells by pro-angiogenic factors including VEGF and Ang-1. NOX2-derived

ROS have been implicated in endothelial cell proliferation and migration. We have

64 shown previously that Ang-1 triggers NOX derived production of ROS which in turns promotes endothelial cell migration while negatively regulating Erk1/2 phosphorylation

[7]. Other studies showed Ang-1 triggered NOX-derived ROS production was required for Ang-1 induction of EC migration and sprouting while positively regulating AKT and

ERK1/2 [8]. NOX4 has been shown to be involved in hypoxia-induced angiogenesis and promotes capillary tube formation, EC migration and improved hind limb ischemia recovery [273]. The release of proangiogenic cells from the bone marrow has been shown to be dependent on ROS signaling, in NOX2 knockout mice with impaired ischemia-induced blood flow recovery [274]. NOX2 and NOX4-derived ROS promote proliferation and survival of endothelial cells via p38, extracellular signal regulated kinase, and Akt signaling [234;273]. IQGAP1 a NOX2 molecular partner is involved in the different stages of angiogenic processes, including ROS-mediated disruption of VE- cadherin–containing adherence junctions and VEGF2 signaling, both of which are known to promote endothelial migration [275].

1.3.3 Mitochondrial derived ROS in angiogenesis

Emerging evidence also suggest that mitochondrial-derived ROS (mtROS) are important in the regulation of endothelial function. Recent studies reported VEGF triggered mtROS release that was required for VEGF induced migration [276] and for VEGF induced tube formation [230]. To date the role of mtROS in angiogenesis remains controversial: it was shown to promote endothelial cell migration and sprouting [277], while another study showed that suppressing mtROS maintained the angiogenic capacity of endothelial cells

[278]. We have previously shown a modest contribution of mtROS in Ang-1 induced

ROS release [7], but the role of mtROS during Ang-1 induced cell signaling remains unclear and largely undefined.

1.3.3.1 Regulation of mitochondrial ROS

The main ROS producer in the cell is the respiratory chain. At least eight components of the respiratory chain/mitochondria can generate ROS, but it is ROS generated from complex I, II and III that have been involved in ROS-induced cell signaling events [279].

Therefore, mtROS are being recognized as integral in physiological and cellular regulation. Complexes I and II release ROS into the mitochondrial matrix while, complex III release ROS to both sides of the mitochondrial inner membrane [280]. ROS release to the inner membrane space allows access to cytosolic targets. Complex III- derived ROS have specifically been shown to be required for many biological processes including oxygen sensing, cell differentiation, and adaptive immunity [281].

Suprisingly mtROS are highly regulated. Multiple stimuli, including immunoreceptor ligation, cytokine stimulation, and hypoxia have been reported to increase mtROS levels.

Increased cytosolic calcium concentration and activation of PI3K also lead to increase mtROS [281]. Negative feedback signaling loop with increase in antioxidant protein expression, through Sirt3 and FOXO TFs activation, decrease mtROS [281]. Finally, mitochondrial membrane potential directly correlate with mtROS production; therefore decreases mitochondrial membrane potential and decreases mtROS levels. Mitochondrial membrane potential is mainly regulated through uncoupling proteins [281].

1.3.3.2 Mitochondrial derived ROS in cell signaling

ROS inhibition of protein phosphatase have been well described. More specifically,

PTP1B was the first phosphatase demonstrated to be inhibited by ROS [282]. ROS also have the ability to inhibit other classes of phosphatases, including phosphatase and tensin homolog (PTEN, lipid phosphatase) and MAPK phosphatases [283;284]. Therefore, mitochondrial-derived ROS are believed to also be involved in the regulation of protein phosphatases.

Mitochondrial-derived ROS are required for hypoxia-mediated HIF stabilization. Indeed, hypoxia leads to activation of signaling cascade promoting a transcriptional program to adapt to low level of oxygen, which in turn generate an increase in mtROS production.

This increase in mtROS production is important in the regulation of several of the cellular responses to hypoxia. Treatment of cells with TNFα thus leads to mitochondrial ROS production, oxidation of JNK phosphatase catalytic cysteines, and activation of JNK signaling and cell death. SOD2 provides a protective effect by lowering cellular ROS levels and decreasing JNK activation after TNFα treatment [285]. Although our knowledge in the regulation of ROS production and the effects of ROS on cell signaling pathways as well as their biological outcome as expanded tremendously in the past, investigating the source or ROS and their specificity of action is of primordial importance.

1.4.1 Dual role of reactive oxygen species in angiogenesis

The various forms of ROS when released in discreet amount appeared to exert cell signaling functions and to be necessary for normal cellular function. It should be

67 considered that excess ROS production or excess ROS exposure can be detrimental to the cells and the execution of their biological program. In those instance we are in presence of oxidative stress. In fact, several studies reported that excess ROS or overactivation of

NOXs can lead to the inhibition of angiogenesis in pathological conditions.

Hypertension –induced oxidative stress inhibits angiogenesis and leads to microvascular rarefication and capillaries disappearance [6485]. ROS dose and timing are important during wound healing. Initially, high level of ROS are produced at the wound site to prevent infection. As ROS levels decrease, the wound healing process starts. In situation where the antioxidant defenses are impaired, or excessive ROS is produced as observe in the diabetic patients, it creates an extended oxidative stress, leading to chronic, non- healing wound and impaired angiogenesis [6484] Studies using H2O2 have shown that at a high concentration, H2O2 causes endothelial injury; while at a low concentration, it stimulates angiogenesis [1502]

1.4.2 The balance between oxidant and antioxidants

The shift in the balance between oxidants and antioxidants in favor of oxidants is termed oxidative stress. Oxidants are naturally produced by the cells and are required for normal biological processes. In order to avoid oxidative damage, organisms have integrated antioxidant systems, which include enzymatic and non-enzymatic antioxidants that are effective in neutralizing excess of ROS. As described in section 1.3.1, various species of

ROS are produce at different sites within the cells. Similarly, the human body produces various species of antioxidants. Six type of enzymatic antioxidant contribute to maintain the Redox balance: 1) Superoxide Dismustase (SOD) converts superoxide to H2O2, SOD1

68 in the cytoplasm , SOD2 is localized at the mitochondria and SOD3 at the extracellular membrane, Catalase converts H2O2 to H2O, glutathione peroxidase , thioredoxines dispose of H2O2 and has cytoplasmic, and mitochondrial forms and glutathione transferase, inactivate secondary metabolites, such as unsaturated aldehydes, epoxides, and hydroperoxides, it is found in the cytoplasm, the membrane and the mitochondria..

Vitamin A, C, E and glutathione are also produced by the cell as antioxidant defense and to prevent cellular damage. Interestingly they have different localization within the cell.

Vitamin E is concentrated in the hydrophobic interior site of cell membrane and is the principal defense against oxidant-induced membrane injury. Vitamin C provides intracellular and extracellular antioxidant by scavenging oxygen free radical. GSH is highly abundant in all cell compartments and is a major soluble antioxidant.

1.4 MicroRNAs in the regulation of Angiogenesis

Micro RNAs (miRNAs) are small endogenously expressed non-coding RNA molecules that regulate gene expression at the post-transcriptional level. They are key regulators of several physiological and pathophysiological processes in the vascular wall. The importance of miRNAs to vascular homeostasis has been demonstrated by genetic silencing of Dicer and Drosha, which results in significant inhibition of angiogenesis in vivo and in vitro [286]. EC-selective deletion of Dicer results in significant attenuation of angiogenic responses to a variety of stimuli (tumours, VEGF administration) and results in significant attenuation of the expressions of key regulators of angiogenesis, including

Tie-2 receptors, VEGFR2, and endothelial NOS [287;288]. Tie-2 and Ang-1 are very important regulators of angiogenesis, vascular integrity, and normal repair processes. The functional roles of miRNAs in the regulation of Ang-1-induced angiogenesis are unknown. We hypothesize that in endothelial cells (ECs) the Ang-1/Tie-2 axis promotes angiogenesis through upregulation of specific sets of pro-angiogeneic miRNAs and downregulation of the expression of a specific set of anti-angiogenic miRNAs.

1.4.1 Discovery of miRNA In 1993 the studies of Victor Ambros, Gary Ruvkun and colleagues described the first microRNA (miRNA) gene, lin-4, and later in 2000, the second miRNA gene let-7, which control development in Caenorhabditis elegans by regulating gene expression at the post- transcriptional level [289;290]. Our knowledge of miRNAs and their mechanisms of regulation quickly expanded to plants and animals [291;292]. Furthermore, it has been

70 estimated that miRNAs could potentially regulate a very large portion of the human genome, as much as 70%, emphasizing their importance as regulators of gene expression

[293]. microRNAs (miRNAs) are short, 22-nt long non-coding RNAs that play a major role in gene regulation in many biological processes, development and diseases

[294;295]. They have tissue-specific expression patterns and down regulate the expression of multiple mRNA genes thereby modulating complex networks such as regulatory loops and signaling pathways [296]. miRNAs also work in concert to regulate efficiently a target gene or a group of target genes. miRNAs use short regions of complementarities, around 6-8 nucleotides within the 3’ UTR of their target mRNAs, to mediate their repressor functions [297]. The extent of pairing between the 3’UTR of the target mRNA and the miRNA “seed”, a 5′ region of the miRNA located on nucleotides

2–7, is important for miRNA target recognition and determines the mechanism of repression [293;297]. Two mechanisms through which miRNAs regulate gene expression at the post-transcriptional level have been described [297;298]. The first occurs when the base pairing between miRNAs and mRNA targets is by precise or nearly precise complementarity, and leads to direct cleavage and destruction of the target mRNA through a process involving the machinery of RNA interference [297;298]. The second mechanism occurs when miRNAs are only partially complementary to their mRNA targets, and results in inhibition of protein synthesis while the stability of the mRNA target is maintained [297;298].

1.4.2 Mechanisms of microRNA biogenesis and silencing by the RNA-induced silencing complex.

The biogenesis of miRNAs starts in the nucleus, where miRNA genes are transcribed by

RNA polymerase II or III into primary miRNA transcripts (pri-miRNA) [299;300]. The expression of miRNAs can be modulated by transcription factors such as NFκB, c-Myc and p53, ELK1, ETV4, or by epigenetic changes in their promoter sequences [271;301-

306]. The genes of different miRNAs with multiple mRNA targets can also be found clustered in the genome in tandem and in several cases these miRNAs can be coordinately transcribed as a common polycistronic precursor transcript [307;308].

The pri-miRNA is then cleaved by a nuclear microprocessor complex comprised of

Drosha, a RNase III enzyme, and the DiGeorge critical region 8 (DRGCR8) protein

[309]. In general the structure of a pri-miRNA includes a double-stranded hairpin stem of

33 nuceotides, a terminal loop and two single-stranded regions upstream and downstream of the hairpin [310]. The hairpin stem and the single-stranded regions flanking the pri- miRNA are critical for binding to DGCR8 and subsequent cleaving by Drosha at a particular site located 11 nucleotides away from the single-stranded/double-stranded junction at the base of the hairpin stem to generate the precursor miRNA (pre-miRNA)

[310;311]. Drosha-mediated processing of pri-miRNAs into pre-miRNAs is not required by miRNAs derived from introns, called mirtrons, that have been released from their transcripts after splicing and their size resembles a pre-miRNA [312].

Subsequently the pre-miRNAs are exported from the nucleus into the cytoplasm by exportin-5 (XPO5) in a complex with Ran-GTP [313]. Once in the cytoplasm, the pre- miRNAs are processed by Dicer, a RNAse III enzyme, to generate a short RNA duplex that will give origin to a mature, single-stranded miRNA of around 20–30 nucleotides in length [298;310]. Although the miRNA duplex could give rise to two different mature miRNAs, only one strand is usually incorporated into the RNA-induced silencing complex

(RISC) to guide the repression of its target mRNAs, while the other strand is degraded [314].

Based on the thermodynamic stability of the base pairs at the two ends of the duplex, the miRNA strand with the less stable base pair at its 5' end is loaded into the RISC [315]. The

RISC is comprised of the RNase Dicer, the double-stranded RNA-binding domain protein

TRBP, the protein activator of PKR (PACT) and the core component Argonaute-2 (Ago-2)

[298]. The RISC silences the expression of its target genes at the post-transcriptional level, which are selected through interactions between the miRNA loaded into the RISC and the mRNA target [297;298]. Then, depending on their degree of complementarity, partial or full, the RISC proceeds to repress translation or induces mRNA degradation, respectively

[297;298].

1.4.3 MicroRNAs as modulators of signaling pathways

Several biological processes including inflammation, angiogenesis and cardiovascular development require miRNA regulation [193;316]. miRNAs are often found as families of redundant genes with numerous possible targets, which makes it difficult to dissect their physiological functions [317]. Furthermore, the degree of target repression by miRNAs is usually low, suggesting that few miRNA targets could be relevant enough to exert a phenotypical change [317]. Signaling pathways are excellent candidates for miRNA regulation due to their dynamic nature and their sensitivity to slight protein variations [317]. Various miRNAs that target modulators of key signaling pathways have been identified and some of their functions include signal amplification, crosstalk between different pathways and signal robustness [317]

Activation of a quiescent endothelium leads to endothelial cell proliferation, degradation of extracellular matrix, modification of cell adhesion properties, migration, cell survival, forming tube-like structures, and eventually maturation into new blood vessels.

Therefore, angiogenesis is a complex process comprising multiple molecular and cellular events that require tight spatio-temporal regulation. miRNAs offer a unique advantage by regulating gene expression: they can be upregulated rapidly in cells; it is believed that miRNA regulated feed forward loops are more efficient than transcriptional repressors

[205]. Moreover, miRNAs can regulate multiple targets, often several of them in a single pathway, a characteristic contributing to the efficient modulation of their target in response to specific stimuli.

1.4.4 MicroRNAs as modulator of Angiogenesis miRNAs that regulate angiogenesis are divided into two groups, pro-angiogenic miRNAs and anti-angiogenic miRNAs. In ECs, pro-angiogenic miR-126 is thought of as a

“guardian” miRNA due to its abundance and its importance to the regulation of various cell functions. Deletion of miR-126 impairs vascular integrity and significantly attenuates migration, proliferation, and other angiogenic activities [318;319]. miR-126 also promotes VEGF signalling and maintains vascular integrity [318;319]. The miR-23-27-

24 cluster is another pro-angiogenic miRNA. It inhibits several angiogenesis inhibitors, including thrombospondin-1, sprouty-2, semaphorin 6A, and semaphorin 6D [320].

Inhibition of miR-27b attenuates sprout formation [286]. miR-17-92 cluster (miR-17, miR-18a, miR-19a, miR-19b-1, miR-20, and miR-92), is anti-angiogenic. Overexpression of miR-92a inhibits angiogenesis through selective inhibition of integrin 5[321].

Similarly, overexpressions of miR-17, miR-18a, miR-19a, and miR-20a attenuate EC sprouting while inhibitions of miR-17 and miR-20a result in increased capillary formation [322]. miR-217 represses the NAD+-dependent deacetylase SIRT1 and blocks

EC migration and tube formation [323]. It should be emphasized that despite recent progress in the identification of important roles for some miRNAs in the vascular system, the functional significance most miRNAs is as yet unknown

1.4.5 miR-640 hsa-miR-640 is a poorly conserved microRNA belonging to an inter-species family comprising one gene conserved in Homo sapiens, Macaca mulata, Pongo pygmaeus and

Pan troglodytes. The mature sequence of miR-640 in the four mentioned species share the exact same seed sequence but differ by one or two nucleotides either in their 3’ or 5’ region.

The miR-640 gene is located in an intronic region on chromosome 19, in intron 1 of

GATAD2A gene and it is not part of any miRNA clusters in that region of Chromosome

19. miR-640 is expressed in various tissue including skeletal muscle, liver, heart and to a lesser extent in kidneys. A recent study showed hydrogen sulfide (H2S)-induced angiogenesis depends on downregulation of the expression of miR-640 [324]. MiR-640, a low-abundance and anti-angiogenic miRNA is down-regulated in cholangiocarcinoma, ovarian carcinoma and lymphocytic leukemia [325-327]. So far only one target of miR-

640 has been identified. Zhou et al. recently demonstrated miR-640 targets HIF1a a pro- angiogenic factor [324]. In the present study we examined whether repression of miR-

640 is also required for Ang-1-induced angiogenesis and we aim at identifying direct targets of miR-640 responsible for Ang-1 pro-angiogenic potential.

1.4.6 ZFP91 Several algorithm predict three binding site for miR-640 in Zinc finger protein 91

(ZFP91) 3’UTR, making it a potential target of this miRNA. ZFP91 is an atypical E3 ligase activating NF-κB-inducing Kinase (NIK) via Lys63-linked ubiquitination in the non-canonical NF-κB signaling pathway [328]. Recent studies have shown ZFP91 expression is upregulated in prostate cancer, it promotes proliferation and tumorigenesis of colon cancer and its upregulated in acute myelogenous leukemia [329-331].

Interestingly overexpression of ZFP91 was shown to enhance proliferation of colon cancer cells and to promote tumor growth in vivo [328]. Moreover, ZFP91 presents secondary structure typical of transcription factors; five zinc-finger domains, one leucine- zipper pattern, one coiled-coil structure, and several nuclear localization signals [332]. In a recent study ZFP91 was reported to act as a transcription factor upregulating HIF1 through the NFκB/p65 pathway. ZFP91 appears as a major oncogenic protein with the ability to promote the activation of two proangiogenic pathways. ZFP91 stabilizes and activates NIK and HIF-1α [328;329]. NIK oncogenic properties have been documented in lung cancer, pancreatic cancer, ovarian cancer and colon cancer [333-335], while HIF-1α promotes angiogenesis responses including: proliferation, angiogenesis, invasion, metastasis, and cell cycle progression [336]. However, the role of ZFP91 in angiogenesis is not well defined, even less so, its role in growth factor induced angiogenesis or physiological angiogenesis.

1.5 ETS family of transcription Factors in the regulation of angiogenesis

ETS (E-twenty-six specific) family of transcription factors was discovered in 1983 with its first member being ETS1. Based on sequence homology a whole family of TFs was subsequently discovered. The ETS factor family is highly conserved across metazoan; from the sponge to Homo sapiens, including Drosophila and C. elegans. To this day the family as extended to 27 members in vertebrates. Interestingly, most ECs expressed genes show an enrichment in TFBS for members of the ETS factor family, hinting investigator that ETS factors are potentially essential in ECs biology.

1.5.1 ETS’s family of transcription factors an overview

The ETS factor family codes for TFs with the ability to either activate or repress transcription. ETS factors are classified into twelve groups, according to their sequence homology [337]. The main characteristic of this family of TFs, is the presence of a highly conserved DNA binding domain, the ETS domain [338]. This domain is specific to the ETS factor family and is not present in any other group of TFs. The ETS domain recognize specific elements in the DNA named EBS (Ets binding site). The EBS element comprise 9bp, including a central consensus sequence 5’-GGAA/T-3’ [339]. This consensus sequence is present in several promoters and enabled ETS factor to positively or negatively regulate more than 200 genes, depending on the presence of a transactivating or transrepressing domain [337]. Another important domain within the

ETS factor family is the pointed domain (PNT), involved in protein-protein interactions.

ETS1 secondary structure comprises a PNT domain, but ELK1 and ETV4 only have the

ETS domain.

Interestingly ETS factors rarely exert their function in solo. They often cooperate with other transcription factors. Moreover, ETS factors are nuclear targets of several cell signaling pathways, and they are integrating extracellular signals as transcriptional regulators. Being at the core of several cell signaling pathway the ETS factors are involved in multiple physiologic and pathologic processes. They regulate angiogenesis, embryonic development, cell morphogenesis and are involved in pathologies such as cancer or rheumatoid arthritis [340]. Surprisingly, the various ETS factor are not redundant in their transcriptional regulation and they specifically regulate genes, despite the highly conserved ETS domain and the consensus TFBS [341]. It is believed ETS factors achieve their functional diversity and their DNA binding specificity, through various mechanisms, including divergent properties of the conserved ETS and PNT domains, the involvement of flanking structured and unstructured regions appended to these dynamic domains, posttranslational modifications, and protein partnerships with other DNA-binding proteins and coregulators [342]

1.5.2 The Angiopoietin-1/Tie-2 signaling and ETS

Little is currently know on ETS factor regulation by Ang-1. ETS factors have been reported to regulate transcription of Ang-2 [343]. Ang-2 plays an important role in vessel destabilization in the initiation of angiogenesis. Both ETS1 and ETS2 increase Ang-2 promoter activity downstream of VEGF signaling [343]. ESE1, another ETS factor was also shown to transcriptionally regulate Ang-1. ESE-1 binds to specific ETS sites within the Ang-1 promoter that are functionally important for transactivation by ESE-1. ESE-1 and Ang-1 are induced in synovial fibroblasts in response to inflammatory cytokines, with ESE-1 induction slightly preceding that of Ang-1 [344]. NERF2, a member of ETS

78 factors binds to the promoter region of Tie2 and transactivates its expression. NERF2 expression is increase during hypoxia as well as upon treatment with Ang-1. This study suggest that Ang-1 regulates the expression of its own receptor through upregulation of

NERF2 during hypoxia [343]. ELF1 is yet another ETS factor involved in angiopoietin signaling. ELF1 is enriched in the developing vasculature of the embryo, where it regulates the expression of the Tie2 gene. Interestingly blocking ELF1 DNA binding activity prevented Ang-1 induced migration. This study highlighted the role of ELF1 in

Ang-1 induced angiogenesis [13;345]. Our group has previously shown that Ang-1 had the potential to induce ELK1 DNA binding activity through post-translational modification. Indeed, Ang-1 can phosphorylate ELK1 through ERK1/2 activation in ECs

[7]. Taken together these studies are indicating that ETS factor are indeed very important regulators of the angiogenic process, and might be key player in Ang-1 induced angiogenesis.

1.5.3 ETS in angiogenesis

ETSs have been implicated in both physiological and pathological angiogenesis. They regulate a wide variety of biological process and several ETS factors have been shown to play major role in the regulation of angiogenesis and vascular development. In ECs, 19

ETS factors are expressed, and only a handful of them have been studied for their role in angiogenesis. None of the ETS factors are highly expressed in the quiescent endothelium, but activation of the endothelium leads to increase in their expression and/or DNA binding activity [346]. Several of ETS factors gene targets are pro- angiogenic genes, but to this date, few of the ETS factors have been directly characterized for their biological function as pro-angiogenic TFs. In this thesis we aimed

79 at identifying new ETS factors involved in Ang-1 signaling, but more importantly defined their biological role in Ang-1-induced angiogenic responses.

ETS1 is the most studied ETS factor. It has been shown to regulate about 90 genes, including: Ang-2 [343], BRACA1 [347], several pro-angiogenic cytokines [347-350], several MMPs [351-354], PLAU [355], RhoC [356], Sprouty (SPRY2) [357], Flt1 [358], and EGR1 [359], most genes enriched in cancer or angiogenesis pathways.

ELK1 has been shown to regulate the expression of less than 20 genes. Among these genes several are key player in angiogenesis and are enriched in cancer pathways. For example ELK1 regulates PDGF, Erbb2, which encodes the human epidermal growth factor receptor 2 and Fos, [360-362] and they are inducer of angiogenesis. PDGF stimulates angiogenesis in vitro and in vivo, Erbb2 overexpression in in human tumor cells is closely associated with increased angiogenesis and expression of VEGF, while

Jun-Fos heterodimers regulate a large variety of biological processes including cell differentiation, proliferation, apoptosis, and oncogenic transformation [363-365].

ETV4, one of the homolog in the PEA3 family of ETS factor is even less studied but it has been involved in the pathology of cancer [366]. ETV4 has been reported to regulate approximately twenty genes. Again several of the ETV4 regulated genes are pro- angiogenic genes, including ETS1 [367], TGFR2 [368], several MMPs; MMP-1,-2, -7 and -14 [369-372] which are typically associated with poor prognosis in tumor expressing

PEA3 transcription factor [373-375] and cyclins; CCND2 and CCND3 [376;377], cell cycle regulators also associated with poor prognosis of cancer [376].

1.6 General objectives and specific aims

The family of angiopoietins and Tie receptors regulates vascular homeostasis, angiogenesis and inflammation under both, physiological and pathological conditions.

Hence, unraveling the molecular mechanism through which angiopoietins differentially mediate their effects represents a unique opportunity to find new targets for therapeutical intervention against vascular pathologies. Although important advances have been made in understanding the various biological function of Ang-1, the main agonist of Tie-2 receptors, little is yet known about the regulation of the intracellular signaling pathways that it activates in ECs. In general the objective of this thesis is to characterize the molecular mechanisms through which Ang-1 regulates EC biology and to study its role in angiogenesis.

To that end, we have the following specific aims:

1- Characterize the source of Ang-1-induced ROS production and elucidate the

specific roles of NOX2, NOX4-derived and mitochondrial-derived ROS in Ang-1

signaling and Ang-1-induced angiogenesis

2- Identify the functional importance of miRNAs in Ang-1-induced angiogenesis: in

particular the functional importance of miR-640 and identify putative targets

through which miR-640 regulates Ang-1-induced angiogenesis.

3- Identify the functional importance of ETS transcription factors in Ang-1-induced

angiogenesis and identify their putative gene targets.

Preface to Chapter 2

Accumulating evidence suggest that intracellular reactive oxygens species (ROS) regulate cytokine and growth factor signaling [221-223]. More precisely, it has been shown that Ang-1 induced phenotypes rely in part on intracellular ROS generated by

NADPH oxidase (NOX) [7;8;223]. Recent evidence suggest that mitochondrial production of ROS is also a tightly controlled process, and plays a role in the propagation of cellular signaling pathways [378] and they are involved in endothelial function [379].

However to date, the exact sources of Ang-1 induced ROS and its function in Ang-1 induced angiogenesis remain largely undefined. Therefore, we hypothesized that Ang-1 could induce ROS production from either NOX2 or NOX4 isoforms, the two most abundant NOXs enzyme expressed in ECs or from the mitochondria as a mechanism to control MAPK signaling to differentially affect Ang-1-induced survival, migration and differentiation of ECs.

Chapter 2: NOX2, NOX4 and Mitochondrial-Derived Reactive Oxygen Species (mtROS) Contribute to Angiopoietin-1 Angiogenic Responses in Endothelial Cells

NOX2, NOX4, and Mitochondrial-Derived Reactive Oxygen Species Contribute to Angiopoietin-1 Signaling and Angiogenic Responses in Endothelial Cells

Sharon Harel, Dominique Mayaki and Sabah NA Hussain

Meakins-Christie Laboratories and Translational Research in Respiratory Diseases Program, Research Institute of the McGill University Health Centre; Department of Critical Care Medicine, McGill University Health Centre, Montréal, Québec, Canada.

Corresponding author: Dr. Sabah Hussain Room EM2.2224 Research Institute of the McGill University Health Centre 1001 Décarie Blvd., Montréal, Québec, Canada H4A 3J1 Tel: 514-934-1934 x34645 E-mail: [email protected]

2.1 Abstract Angiopoietin-1 (Ang-1) is a ligand of Tie-2 receptors that promotes survival, migration, and differentiation of endothelial cells. Several studies have linked reactive oxygen species (ROS) to Ang-1 signaling and distinct angiogenic responses, but the molecular sources of these ROS have never been clearly identified. In this study, we have identified source-specific contributions of ROS to Ang-1/Tie 2 signaling and angiogenic responses in human umbilical vein endothelial cells (HUVECs), specifically the differential contributions of mitochondrial ROS (mtROS) and ROS from two isoforms of NADPH oxidase (NOX2, NOX4). We demonstrate that: 1) Ang-1 induces significant increases in mtROS production under normal conditions but does not when cells are pre-incubated with mitochondrial antioxidants; 2) Ang-1 induces rapid Tie-2-dependent increases in cytosolic ROS production but does not when NOX2 and NOX4 are knocked down; 3)

Ang-1 induces simultaneous increases in phosphorylation of AKT, ERK1/2, p38, and

SAPK/JNK proteins within a few minutes of exposure, but this response is strongly and selectively attenuated when NOX2 and NOX4 are knocked down or cells are pre-treated with mitochondrial antioxidants; 4) Ang-1 exerts a strong effect on HUVEC survival in serum-deprived medium and enhances cell migration and capillary tube formation, but the survival response is inhibited by NOX2 knockdown and the migration and tube formation responses are entirely absent with NOX4 knockdown or pre-treatment with mitochondrial antioxidants. We conclude that Ang-1 triggers NOX2, NOX4, and the mitochondria to release ROS and that ROS derived from these sources play distinct roles in the regulation of the Ang-1/Tie 2 signaling pathway and pro-angiogenic responses.

Keywords: angiogenesis, angiopoietin-1, NADPH oxidase, reactive oxygen species, mitochondrial reactive oxygen species, endothelial cells, apoptosis, mitogen-activated protein kinases

Abbreviations: Ang-1: angiopoietin-1; ROS: reactive oxygen species; mtROS: mitochondrial reactive oxygen species; ECs: endothelial cells; HUVECs: human umbilical vein endothelial cells; PDGF: platelet derived growth factor; EGF, epidermal growth factor; VEGF: vascular endothelial growth factor; PBS: phosphate buffered saline; FBS: fetal bovine serum; shRNA: short hairpin RNA; SEM: standard errors of the mean.

2.2. Introduction The endothelial-specific receptor tyrosine kinase 2 (Tie-2) and its main ligand angiopoietin 1 (ANGPT1 or Ang-1) are very important regulators of angiogenesis, vascular integrity, and inflammation. Ang-1 promotes endothelial cell (EC) migration [1], sprouting, and differentiation into tube-like structures [2-4] and strongly inhibits apoptosis [5]. It also stimulates vascular remodeling, promotes lymphangiogenesis [6,7], and induces trans-autophosphorylation of Tie-2 receptors to activate several downstream signaling pathways, including the PI-3 kinase/AKT, DOK-R/PAK1, and mitogen- activated protein kinase (MAPK) pathways (ERK1/2, p38, and SAPK/JNK)[8].

Furthermore, Ang-1 promotes the release of secondary mediators of inflammation, such as interleukin-8, and activates several transcription factors, including activating protein-1

(AP-1), early growth response-1 (Egr-1), and KLF-2 [2,3,9].

- - Hydrogen peroxide (H2O2), superoxide (O2 ), and hydroxyl (OH ) ions are reactive oxygen species that are produced both by the mitochondrial electron transport chain and by enzymatic sources, such as by NADPH (NOX) and dual (DUOX) oxidases and nitric oxide synthases (NOS). The primary function of NOX and DUOX enzymes is to produce

ROS. ROS serve as second-messenger molecules and control cell signal transduction through post-translational modifications of Cys residues in targeted proteins. In the vascular system, for example, a burst of ROS production derived primarily from NOX activates MAPK pathways downstream from receptors of pro-angiogenic growth factors such as platelet-derived growth factor (PDGF) and vascular endothelial growth factor

(VEGF) [10,11].

NOX2 and NOX4 are the most abundantly expressed isoforms in ECs, while

NOX1 and NOX5 are expressed to a lesser extent [12]. NOX1, 2, and 5 generate superoxide ions that can cause endothelial dysfunction and contribute to vascular disease under pathological conditions, but also have crucial roles in cell signaling. NOX4 produces H2O2, which acts as both an intracellular messenger and a cell-signaling molecule [13]. NOX2 is localized on the plasma membrane, phagosomes, endosomes, and at the leading edge of lamellipodia [12] and requires interaction with five regulatory subunits (p22phox, p47phox, p67phox, p40phox, and Rac1/2) for optimal activation. In vascular cells, its activity is induced by vasoactive ligands such as VEGF and angiotensin II [14].

NOX4 is localized on the plasma membrane, focal adhesions, nucleus, mitochondria, and endoplasmic reticulum [12] and promotes migration of VEGF-stimulated ECs [14]. In contrast to NOX2, it is constitutively active, which means that no interaction with regulatory subunits is necessary for activation to occur.

Cell signaling in angiogenesis has long been a focus of the work in our laboratory and we were the first to report that acute exposure to Ang-1 elicits important release of

ROS in ECs [15]. Using a peptide that blocks interaction between NOX2 and p47phox, we observed that Ang-1-induced ROS release is dependent on NOX2 activation and demonstrated its importance to Ang-1-induced EC migration [15]. Subsequent studies have confirmed these results and have pointed to the general role that NAPDH oxidase- derived ROS play in Ang-1-mediated angiogenesis and EC differentiation [16,17]. Yet, despite the progress that has been made so far in linking NOX in general to the Ang-

1/Tie-2 pathway, it remains unclear as to precisely how each NOX isoform contributes to receptor signaling.

To as substantial degree, mitochondria also contribute to the ROS that have an impact upon the normal biological action of the Ang-1/Tie-2 axis. Mitochondrial ROS

(mtROS) had long been thought to be a mere by-product of aerobic respiration, but recent evidence has suggested that they also actively participate in cellular signaling. Integrin, for instance, triggers the generation of mtROS within a few minutes of ligation, leading to the oxidation of SHP2 [18], a phosphatase that is crucial to angiogenic cell function. mtROS have also been linked to PDGF signaling in fibroblasts [19] and have been shown to play important roles in VEGF-induced migration and capillary tube formation in ECs

[20]. Furthermore, as their production increases in ECs, PI-3 kinase activity accelerates, which, in general, results in significant impacts on cell growth, proliferation, differentiation, motility, and survival. Endothelial sprouting, in particular, has been closely linked to the presence of mtROS in the vasculature [21].

Overall, these studies indicate that mtROS do, indeed, play distinct regulatory roles in cell signaling. However, the precise manner in which they interact with the Ang-

1/Tie 2 signaling pathway and exert pro-angiogenic responses in ECs is, as yet, unknown.

Two reports have provided indirect evidence of their contribution in response to Ang-1 exposure. They show that pre-treatment of cells with rotenone, which inhibits Complex I of the mitochondrial oxidative phosphorylation pathway, attenuates Ang-1-induced ROS production [15,16]. However, source-specific ROS measurements were not performed in these studies, nor were any assessments of the functional significance of mtROS to angiogenesis or Ang-1/Tie signaling. The main focus of the present study, therefore, is to identify the differential contributions of mtROS, NOX2, and NOX4 to Ang-1-induced

ROS generation in ECs and to determine what their functional roles are in relation to

Ang-1/Tie-2 axis signaling and Ang-1-induced pro-angiogenic responses.

2.3 Materials and Methods All experiments involving human tissues were approved by the Research Ethics Board of the Research Institute of the McGill University Health Centre and conform to the principles outlined in the Declaration of Helsinki on Ethical Principles for Medical

Research Involving Human Subjects.

Materials: Recombinant human Ang-1, neutralizing goat IgG polyclonal anti-Tie-2 receptor antibody, and control goat IgG antibodies were purchased from R&D Systems

(Minneapolis, MN). Antibodies for p38, phospho-p38, ERK1/2, phospho-ERK1/2, AKT, phospho-AKT, SAPK/JNK, phospho-SAPK/JNK, and cleaved Caspase-3 were purchased from Cell Signaling Technology (Danvers, MA). Mito-TEMPOL was purchased from

Sigma-Aldrich (Oakville, ON). Adenoviral vectors expressing short hairpin sequences targeting NOX2 (Ad-shNOX2), NOX4 (Ad-shNOX4), and GFP (control, Ad-shGFP) were constructed as previously described [22]. In brief, 21-base-pair short hairpin RNA sequences directed against NOX2, NOX4, or enhanced GFP are placed under the control of a Tie-2-dependent murine U6 promoter. A separate CMV promoter drives the expression of a reporter gene (GFP for shNOX2 and shNOX4 constructs; LacZ for shGFP construct).

Cell Culture: Human umbilical vein endothelial cells (HUVECs) were isolated from umbilical cords, as previously described [3,9]. Informed consent was obtained from subjects prior to inclusion in the study. HUVECs were grown in MCDB 131 full medium containing 20% fetal bovine serum (FBS), endothelial cell growth supplement, 2 mM/l glutamine, heparin, and gentamicin (full medium). Some experiments were performed in

MDCB 131 basal medium containing 2% FBS.

Adenovirus Infection of HUVECs: When HUVECS reached 60-70% confluency they were washed twice with phosphate buffered saline (PBS) then infected for 6 h with serum-free medium containing 100 mutiplicity of infection (MOI) virus units of Ad- shGFP, Ad-shNOX2, or Ad-shNOX4. At the end of the infection period, virus- containing medium was replaced with complete medium and cells were left to recover for 48-72 h. mtROS Attenuation: Two experimental approaches were used to selectively attenuate mtROS production: 1) Mitochondria-targeted antioxidant peptide SS31 is a cell- permeable tetrapeptide (d-Arg-2′, 6′-dimethyltyrosine-Lys-Phe-NH2) that exerts strong antioxidant capabilities. SS31 attenuates mtROS production, prevents mitochondrial swelling, and guards against oxidative cell death [23]. SS20, a structurally-related tetrapeptide (Phe-D-Arg-Phe-Lys-NH2) that does not exert antioxidant activity, was used as a control [23]. Both were generously provided by Dr. Peter W. Schiller

(Université de Montréal); 2) Mito-TEMPOL is a mitochondria-targeted antioxidant agent that is a derivative of TEMPOL, a SOD mimetic membrane permeable superoxide scavenger [24]. Prior to experimentation, HUVECs were maintained for 24 h in full medium containing 200 nM of SS31 or SS20, MitoTempol (10 M), or dimethyl sulfoxide (DMSO, 0.001%). Antioxidants and corresponding controls were replenished as needed throughout the experimental process.

H2O2 Measurement: Amplex® Red reagent (Life Technologies, Burlington, ON) in combination with horseradish peroxidase detects hydrogen peroxide (H2O2) release from biological samples. Adenovirus-infected or antioxidant-treated HUVECs and their controls were grown to confluence in 96-well flat clear-bottom black polystyrene

93 microplates. To evaluate the role of Tie-2 receptors in Ang-1-induced ROS production, cells were treated with control IgG or neutralizing Tie-2 IgG antibodies (37.5 g/ml) for

4 h then exposed to PBS or Ang-1 (300 ng/ml) and Amplex Red. The intensity of fluorescence was continuously monitored for 60 min using a SpectraMax® M2

Microplate reader (Molecular Devices, Sunnyvale, CA). The slope of fluorescence over time was quantified for each preparation and expressed as relative fluorescence units per minute. mtROS Measurement: mtROS levels were measured with MitoSOX™, a fluorescent

- dye that permeates live cells and is rapidly oxidized by O2 inside the mitochondria.

Antioxidant-treated HUVECs and their controls were grown to confluence in 96-well plates as above. Cells were treated with MitoSOX reagent for 30 min then exposed to

PBS or Ang-1 (300 ng/ml) for 15 min. To evaluate the role of Tie-2 receptors in Ang-1- induced mtROS production, cells were treated with control IgG or neutralizing Tie-2 IgG antibodies (37.5g/ml) for 4 h, exposed to MitoSOX reagent for 30 min, then exposed to

PBS or Ang-1 for 15 min. MitoSOX fluorescence was measured using a microplate reader, as described above.

Immunoblotting: Total cell lysates (30 µg of total protein) loaded onto Tris-glycine

SDS-polyacrylamide gels along with a molecular-weight size marker were boiled for 5 min. RIPA buffer (Santa Cruz Biotechnology, Dallas, TX) was used to lyse cells and protein concentrations were measured using a Bio-Rad protein assay (Saint-Laurent,

QC). Proteins were then electrophoretically transferred to polyvinylidene difluroide

(PVDF) membranes and blocked with 5% non-fat dry milk. Membranes were incubated

94 with antibodies overnight at 4°C. Horseradish peroxidise-conjugated secondary antibodies and ECL reagents were used to detect proteins, as previously described [25].

Cell Counting: Forty-eight hours post-infection, adenovirus-infected HUVECs were seeded in 12-well plates at a density of 8x104/cm2. Cells were maintained for 24 h in full

(20% FBS), basal (2% FBS), or basal medium containing Ang-1 (300 ng/ml). Cells were counted using a hemocytometer.

Caspase-3 Activity Assays: Forty-eight hours post-infection, adenovirus-infected

HUVECs were seeded in 12-well plates at a density of 8x104/cm2. Cells were maintained for 24 h in full (20% FBS), basal (2% FBS), or basal medium containing Ang-1 (300 ng/ml). Caspase-3 activation was determined by measuring the intensity of cleaved

Caspase-3 protein with immunoblotting.

Wound Healing Assay: HUVEC monolayers were wounded with a 200 μl pipette tip and maintained for 8 h in basal (2% FBS) or basal medium containing Ang-1 (300 ng/ml). Wounded areas were visualized using an Olympus inverted microscope and quantified using Image-Pro Plus™ software (Media Cybernetics, Bethesda, MD). Values are reported as % wound healing equal [1 – (wound area at t8/wound area at t0)] x 100, where t8 is the time (8 h) over which cells were maintained in media and t0 is the time immediately after wounding, as previously described [25].

Capillary Tube Formation Assay: HUVECs were seeded in 96-well plates pre-coated with growth factor-reduced Matrigel at a density of 1x104 cells per well. Cells were maintained for 24 h in basal (2% FBS) or basal medium containing Ang-1 (300 ng/ml).

Whole-well images were captured using an Olympus inverted microscope (40X) and analyzed using Image-Pro Plus™ software. Angiogenic tube formation was determined

95 by counting branching points of formed tubes and total tube length in each field, as previously described [25].

Real-Time PCR (qPCR): Total RNA was extracted using a PureLink® RNA Mini Kit and Invitrogen™ SuperScript® II RNase H-Reverse Transcriptase (Life Technologies).

RNA (2 μg) was reverse transcribed for 50 min at 42°C then for 5 min at 90°C. To carry out qPCR reactions, specific primers for human NOX1, NOX2, NOX3, NOX4, NOX5,

DUOX1, and DUOX2 (Supplementary Materials Table 1) and an Applied Biosystems

7500 qPCR System (Foster City, CA) were used. -ACTIN was used as a control. Qiagen

SYBR® Green PCR Master Mix (25 μl) (Hilden, Germany) was added to a mix of reverse transcriptase (1 μl) and primers (3.5 μl) for a final concentration of 10 μM. The thermal profile was as follows: 10 min at 95°C, 40 cycles of 15 s at 95°C, 30 s at 57°C, and 34 s at 72°C. Primer-dimer formation and contamination were assessed using melt analyses for each PCR experiment. A single melt peak for each set of primers was used to confirm that a single PCR reaction product was generated. Results were analyzed using the comparative threshold cycle method, which determines the value (CT) at which the amplification curve crosses the threshold line. CT values were used to calculate absolute copy numbers from standard curves generated by plasmids containing full-length coding sequences of NOX1, NOX2, NOX3, NOX4, NOX5, DUXO1, DUOX2 and -ACTIN

(Supplementary Materials Table 1). All qPCR experiments were performed in triplicate.

Data Analysis: Data are expressed as means ± SEM. Differences between experimental groups were determined using a Two-Way Analysis of Variance followed by a Student–

Newman–Keuls post-hoc test. P values <0.05 were considered statistically significant.

2.4 Results

Ang-1 Regulation of ROS: Over the 60 min monitoring period, a linear rate of H2O2 release was observed when cells were exposed to PBS alone (basal ROS production)

(Figure 1). In cells pre-treated with IgG antibody, exposure to Ang-1 increased ROS levels relative to PBS alone (Figure 1). This response was absent when cells were pre- treated with neutralizing Tie-2 antibody, indicating that Tie-2 binding is required for the

Ang-1-induced effect on ROS production to occur (Figure 1).

Roles of NOX2 and NOX4: mRNA levels of various NOX and DUOX isoforms in

HUVECs were measured. NOX4 and NOX2 are the most abundantly-expressed NOX isoforms, with averages of 361 and 16 copies per 105 copies of -ACTIN, respectively

(Fig.A.1). Relatively low levels of NOX1, NOX3, DUOX1, and DUOX2 were detected, with averages of 0.1, 0.8, 0.01, and 1 copies per 105 copies of -ACTIN, respectively

(Supplementary Materials Figure 1). NOX5 was not detected. To evaluate the individual roles of NOX2 and NOX4 in Ang-1-induced ROS production, adenoviral vectors expressing short hairpin sequences targeting NOX2 (Ad-shNOX2), NOX4 (Ad- shNOX4), and GFP (Ad-shGFP, control) were used. Infection with Ad-shNOX2 attenuated NOX2 mRNA and protein levels without affecting NOX4 mRNA levels

(Supplementary Materials Figure 2). Similarly, infection with Ad-shNOX4 attenuated

NOX4 mRNA without affecting NOX2 (Supplementary Materials Figure 2). This confirms the efficacy and selectivity of Ad-shNOX2 and Ad-shNOX4 as selective knockdown vectors. We were unable to verify knockdown of NOX4 protein with Ad- shNOX4 because of poor quality of NOX4 antibodies, although four different commercially available products were tested.

Exposure to Ang-1 resulted in increased ROS production relative to exposure to

PBS alone (Figure 2). When NOX2 was knocked down, ROS levels in cells exposed to

PBS alone were lower than in cells exposed to Ad-shGFP, and exposure to Ang-1 had no effect on ROS production (Figure 2). Similar responses were observed with NOX4 knockdown (Figure 2). Combined NOX2/NOX4 knockdown resulted in diminished ROS production as compared to Ad-shGFP-infected cells, and exposure to Ang-1 decreased

ROS levels relative to exposure to PBS alone (Figure 2). These results indicate that

NOX2 and NOX4 contribute to basal ROS levels and that both isoforms are required for significant Ang-1-induced production of ROS to occur.

Ang-1 Regulation of mtROS: In cells pre-treated with SS20 (control), Ang-1 exposure increased the fluorescent MitoSOX signal as compared to exposure to PBS alone (Figure

3B). This response was not seen in cells pre-treated with SS31 (Figure 3B). To confirm the requirement for Tie-2 binding, neutralizing Tie-2 IgG antibody was used. In cells pre- incubated with control IgG antibody, Ang-1 increased the fluorescent signal as compared to PBS alone (Figure 3C). This response was not seen in cells pre-incubated with neutralizing Tie-2 antibody, indicating that Tie-2 receptors are required for significant

Ang-1 induced production of mtROS to occur. To determine NOX2 and NOX4 contributions to mtROS levels, fluorescence was measured in Ad-shGFP-, Ad-shNOX2-, and Ad-shNOX4-infected cells. Ang-1 increased mtROS levels in Ad-shGFP-infected cells, thereby excluding their involvement in the process (Supplementary Materials

Figure 3).

Ang-1/Tie-2 Signaling: Within a few minutes, Ang-1 increased AKT, ERK1/2, p38, and

SAPK/JNK phosphorylation in Ad-shGFP-infected cells, (Figure 4). NOX2 knockdown

98 significantly increased basal phosphorylation levels of ERK1/2, p38 and SAPK/JNK and negated the Ang-1 effect on AKT, p38, and SAPK/JNK phosphorylation (Figure 4). In addition, NOX2 knockdown attenuated the relative increase in ERK1/2 phosphorylation elicited by Ang-1 from about 6-fold in cells infected with Ad-shGFP to about 3-fold

(Figure 4). In comparison, NOX4 knockdown significantly increased basal phosphorylation levels of AKT and ERK1/2 and negated the Ang-1 effect on SAPK/JNK phosphorylation (Figure 5). Exposure to Ang-1 increased AKT, ERK1/2, p38, and

SAPK/JNK phosphorylation in SS20 pre-treated cells (Figure 6). Pre-treatment with

SS31 significantly increased basal levels of ERK1/2, p38 and SAPK/JNK phosphorylation and negated the effect of Ang-1 on p38 and SAPK/JNK phosphorylation

(Figure 6). This suggests that NOX2, NOX4, and mtROS exert significant and overlapping effects on AKT and MAPK activation downstream from Tie-2 in response to

Ang-1 exposure.

Cell Survival and Caspase-3 Cleavage: Serum deprivation alone decreased cell counts by 50% and increased cleaved Caspase-3 intensity 3.5-fold while serum deprivation in combination with Ang-1 increased cell counts and decreased cleaved Caspase-3 intensity

(Figure 7), confirming that Ang-1 exerts a pro-survival effect on endothelial cells. NOX2 knockdown alone enhanced the serum deprivation-induced effect on cell death, reducing the cell count by 78% and increasing cleaved Caspase-3 intensity 3-fold, but in combination with Ang-1 no ameliorative response was seen (Figure 7). This suggests that

NOX2 is required for the pro-survival effect of Ang-1 to occur. NOX4 knockdown alone reduced the cell count by 40% and increased cleaved Caspase-3 intensity 2-fold, but in combination with Ang-1, cell counts increased and cleaved Caspase-3 intensity decreased

99 relative to serum deprivation alone (Figure 7). This suggests that NOX4 is not required for the pro-survival effect of Ang-1 to occur. In cells pre-treated with SS20 or DMSO

(vehicle), Ang-1 increased cell counts and decreased cleaved Caspase-3 intensity relative to serum deprivation alone (Figure 8). The same observations were made when cells were pre-treated with SS31 or MitoTEMPOL, suggesting that mtROS are not required for the pro-survival effect of Ang-1 to occur (Figure 8).

Wound Healing Assays: In Ad-shGFP-infected cells, exposure to Ang-1 increased EC migration relative to exposure to PBS alone (basal migration) (Figure 9A-B). NOX2 knockdown alone increased the degree of migration relative to basal migration, but in combination with Ang-1 no appreciable effect on cell migration was observed (Figure

9B). NOX4 knockdown alone exerted no effect on the degree of migration observed relative to basal migration, but, in combination with Ang-1, increased migration was observed (Figure 9B). Ang-1 increased cell migration relative to basal levels in cells pre- treated with SS20 or DMSO, but the pro-migration effect of Ang-1 was absent in cells pre-treated with SS31 or MitoTEMPOL (Figure 9C-D). Taken together, these results indicate that NOX2 inhibits EC migration and that both NOX4 and mtROS are required for Ang-1-induced migration to occur.

Capillary Tube Formation: Exposure to Ang-1 and NOX2 knockdown both increased capillary tube length relative to basal levels in Ad-shGFP-infected cells (Figure 10A-B).

NOX2 and NOX4 knockdown negated this Ang-1-induced effect on differentiation

(Figure 10B). Ang-1 increased tube length relative to basal values in cells pre-treated with SS20 or DMSO, but pre-treatment with SS31 or Mito-TEMPOL negated this effect

(Figure 10C-D). These results indicate that NOX2 inhibits EC differentiation into tubes

100 and that NOX4 and mtROS are required for Ang-1-induced capillary tube formation to occur.

101

2.5 Discussion The primary findings of this study are:

1) Ang-1 induces rapid Tie-2-dependent increases in ROS production in HUVECs and

NOX2, NOX4, and mtROS contribute to this response; 2) ROS derived from NOX2,

NOX4, and mitochondria act as messenger intermediates in Ang-1-induced cell signaling: NOX2 is involved in phosphorylation of AKT, p38, and SAPK/JNK proteins,

NOX4 is involved in SAPK/JNK phosphorylation, and mtROS are involved in p38 and

SAPK/JNK phosphorylation; 3) NOX2-, NOX4-, and mitochondrial-derived ROS regulate Ang-1-induced angiogenic responses of endothelial cells; NOX2 is required for cell survival and inhibition of apoptosis while NOX4 and mtROS are required for cell migration and capillary tube formation

Ang-1 Regulation of ROS: Many studies have confirmed that Ang-1 triggers increases in ROS production in ECs although the molecular sources of these ROS have not been clearly identified. In one of our previous studies, a peptide (gp91ds-tat) that inhibits the association of gp91phox (NOX2) subunits with p47phox subunits of NADPH oxidase and a dominant-negative form of Rac-1 (Rac1N17) were used to demonstrate that NOX2 contributes to enhanced ROS release in Ang-1-exposed ECs [15]. In two subsequent studies, non-specific NOX inhibitors were used to suggest that at least two isoforms of

NADPH oxidase may be the chief ROS regulators of Ang-1/Tie-2 signaling in ECs

[16,17]. In one of these studies, ECs derived from p47phox–deficient mice were used to confirm its importance to Ang-1-induced ROS release [17]. In the present study, we explore source-specific contributions of ROS to Ang-1/Tie-2 signaling and angiogenic responses. A loss-of-function approach that utilized adenoviruses expressing shRNA for

102

NOX2 and NOX4 and two independent mitochondria-selective ROS inhibitors, SS31 peptide and Mito-TEMPOL, was selected as the most efficacious method of determining

ROS specificity. Results suggest that at least three sources of ROS, including NOX2,

NOX4, and the mitochondria, contribute to ROS release in response to Ang-1 exposure and that each plays a distinct role in signal regulation and in eliciting angiogenic responses in endothelial cells.

The assertion that NOX4 and mtROS make distinct contributions can be challenged on the basis that NOX4 protein has been detected in the mitochondria of various cell types and that depletion of NOX4 or its regulatory subunit p22phox results in significant attenuation of mtROS production [26-28]. However, van Buul and colleagues have determined that NOX4 protein is found in focal adhesions, along stress fibers, and in the nucleus of ECs but not in the mitochondria [12]. Our study shows that basal mtROS production declines in cells with NOX4 knockdown but not in those with NOX2 knockdown (Supplementary Materials Figure 2.3). This confirms that NOX4 contributes to ROS release into the mitochondria, but we also found that Ang-1 is still capable of inducing mtROS production despite NOX4 knockdown (Supplementary Materials Figure

2.3). This suggests that the source of increased mtROS in these cells was not NOX4 but that they originated from oxidative phosphorylation complexes.

ROS Regulation of Ang-1/Tie-2 Signaling: The significance of the PI-3 kinase/AKT,

ERK1/2, p38, and SAPK/JNK pathways to Ang-1/Tie-2 signaling and angiogenic function in ECs has been confirmed by several studies (for review, see [29]). Here we report for the first time that the Ang-1/Tie-2 axis utilizes ROS generated by NOX2,

NOX4, and the mitochondria to trigger distinct effects on the AKT, ERK1/2, p38, and

103

SAPK/JNK pathways. We have demonstrated that NOX2 knockdown leads to negation of the stimulatory effects of Ang-1 on AKT, p38, and SAPK/JNK phosphorylation, results that are congruent with previous studies that show that Rac1- and p47phox- dependent NOX2 activity is required for Ang-1-induced activation of the these pathways to occur in HUVECs [15,17]. We have also demonstrated that NOX4 knockdown attenuates Ang-1-induced SAPK/JNK protein phosphorylation. Finally, we have shown that mitochondrial antioxidants attenuate Ang-1-induced p38 and SAPK/JNK but not

AKT and ERK1/2 phosphorylation. This is in line with results from a study using hypoxic ECs that indicates that mtROS are required for p38 activation to occur [30].

The primary mechanism through which ROS regulate growth factor cellular signaling is through direct oxidation and inactivation of critical cysteine residues of several protein tyrosine phosphatases (PTPs) whose function is to specifically regulate tyrosine kinase receptor activation and downstream pathways. We advance the view that regulation of AKT phosphorylation by NOX2 is likely mediated through oxidative inactivation of PTEN, PP2, and PP1 phosphatases that negatively regulate PI-3 kinase activity [31,32]. With respect to regulation of MAPKs, it has been well established that

ROS act upon them through four different mechanisms including: a) Direct oxidative inactivation of phosphatases that selectively target tyrosine kinase receptors upstream from MAPKs [33]; b) Direct oxidation of regulatory kinases such as MAP2K and

MAP3K, which leads to activation of downstream kinases such as p38 and SAPK/JNK

[34]; c) Direct oxidation and activation of Ras small GTPases that regulate MAPKs downstream [35]; and d) Inactivation of specific dual-specificity phosphatases (DUSPs) that target ERK1/2, p38, and SAPK/JNK proteins and maintain them in an inactive state

104

[36]. We have recently demonstrated in ECs that three of these DUSPs (DUSP1, DUSP4, and DUSP5) regulate the activities of the ERK1/2, p38, and SAPK/JNK pathways downstream from Tie-2 receptors [25]. It should also be emphasized, though, that differences in the localization, abundance, and kinetics of ROS production, along with the status of antioxidant pools and cellular redox conditions, are all important factors that determine specific effects of NOX2, NOX4, and mtROS on MAPK activation.

ROS Regulation of Cell Survival: Regulation of cell survival and cell death (apoptosis) by NOX-derived ROS has been investigated in several different cell types under various conditions with contrasting results. Both NOX2 and NOX4 have been shown to promote cell death through direct and indirect induction of pro-apoptosis pathways [37,38] whereas, under different conditions, pro-survival effects have been attributed to them

[39,40]. In this study, we used Ang-1 exposure as a stimulus for cell survival and serum deprivation as a stimulus for cell death, since it is well-known that Ang-1 promotes survival and strongly inhibits serum deprivation-induced apoptosis in ECs [5,41,42].

These pro-survival and anti-apoptotic effects of Ang-1 are evidenced in this study by increased cell counts and decreased Caspase-3 cleavage in AD-shGFP-infected ECs.

We also made two new observations regarding the regulatory roles of ROS in EC survival. First, when cells were serum-deprived, counts were lower under NOX2 knockdown conditions and higher under NOX4 knockdown conditions as compared to full-medium Ad-shGFP-infected cells (Figure 7A). Second, Ang-1 failed to induce an increase in cell counts or to inhibit Caspase-3 cleavage in NOX2 knockdown cells. In comparison, NOX4 knockdown had no effect on the pro-survival and anti-apoptotic effects of Ang-1 (Figure 7A-C). These findings suggest that NOX2 is essential to EC

105 survival and apoptosis inhibition whether Ang-1 is present or not, and that in the absence of Ang-1, NOX4 promotes apoptosis in serum-deprived cells. These conclusions concur with those of Jeon and Boo regarding the roles of NOX2 and NOX4 in serum-deprived and shear stress-exposed ECs [43,44]. Furthermore, a pro-survival role for NOX2 has also been described in cardiac myocytes in relation to 1 integrin signaling and in epithelial cells in relation to peroxisome proliferator-activated receptor gamma (PPAR-γ) signaling [45,46], and a pro-apoptosis role for NOX4 has been shown in ECs exposed to

TGF [47].

The mechanisms underlying the differing effects of NOX2 and NOX4 on EC survival remain speculative. One possible mechanism is related to the nature of the ROS

- being generated by the two isoforms. NOX2 generates O2 anions, which mediate one- electron signaling, whereas NOX4 generates H2O2, which mediates two-electron

- signaling [48]. The intracellular balance between O2 and H2O2 is a major determinant of

- cell fate that is tied to cytosolic pH level [49]. An increase in O2 causes an increase in intracellular pH through a specific effect on the Na+/H+ antiporter [50]. Maintenance of pH in the alkaline range leads to resistance to apoptotic stimuli. Conversely, an increase in H2O2 triggers intracellular acidosis through an indirect inhibitory effect on membrane

Na+/H+ exchange and renders the cell more amenable to apoptotic stimuli [51].

In addition to the cellular pH mechanism, we suggest that NOX2 promotes cell survival in Ang-1-exposed cells through simultaneous activation of the AKT, and

SAPK/JNK pathways downstream from Tie-2 receptors. These pathways have previously been shown to promote survival and to inhibit apoptosis in Ang-1-exposed serum- deprived ECs [5,8,41,42]. These pathways are believed to promote survival through the

106 activation of several transcription factors, including activating protein -1 (AP-1), early response factor 1 (EGR-1), and KLF2 [2,3,9]. Finally, it is possible that NOX2 also regulates EC survival through the action of its molecular partners, including p47phox, p67phox, and p40phox [52].

ROS Regulation of Migration and Differentiation: The individual roles of NOX2,

NOX4, and mtROS in Ang-1-induced EC migration and capillary tube formation have heretofore been unknown. This study reveals that Ang-1 elicits increases in migration and capillary tube formation in Ad-shGFP-infected ECs and in those pre-treated with SS20 or

DMSO (Figures 9-10). We also found that Ang-1 exerts no effect on migration or tube formation when NOX4 is knocked down or when cells have been pre-treated with SS31 or Mito-TEMPOL, suggesting that NOX4 and mtROS are required for Ang-1-induced pro-angiogenesis responses to occur. The capillary tube formation assay is designed to detect several pro-angiogenic responses of ECs, including activation, proliferation, migration, cell attachment to the extra-cellular matrix, protease production, and cell alignment. Our demonstration that NOX4 promotes cell migration and tube formation downstream from Tie-2 receptors concurs with previous work on full medium-grown microvascular and macrovascular ECs that confirms the regulatory importance of NOX4 to migration and differentiation [53].

It has been shown that NOX4 overexpression in endothelial cells of transgenic mice enhances in-vivo angiogenesis [54]. Correspondingly, our assertion that mtROS are required for Ang-1-indued EC migration and differentiation to occur is consonant with recent evidence from 3D in-vitro angiogenesis assays that suggests that mitochondrial- derived H2O2 promotes EC sprouting [21]. Furthermore, it has been reported that

107 angiogenesis factors like VEGF trigger increases in mitochondrial metabolism and mtROS production, and that inhibition of mtROS by mitochondria-targeted vitamin E

(Mito-Vit-E) strongly attenuates VEGF-induced migration in-vitro and carotid artery re- endothelialization in-vivo [20].

The regulatory roles of NOX2 on angiogenesis factor-induced migration and differentiation remain debatable. For instance, it has been shown to be involved in

VEGF-induced migration of human ECs [55,56], but not of primary murine ECs [57].

Three studies have reported that Ang-1 strongly induces EC migration and tube formation through enhanced NOX2-derived ROS release [15-17], based on experiments where non- specific antioxidants, a peptide that inhibits the association of p47phox with NOX2, and

ECs derived from p47phox deficient mice were used. Our study used NOX2 knockdown to demonstrate that EC migration and tube formation increased in the absence of Ang-1, suggesting that NOX2 exerts an inhibitory effect on basal migration and differentiation

(Figures 9-10). This conclusion is in accordance with the observations of von Löhneysen et al. [58], who reported that overexpression of NOX2 in Cos cells inhibits migration, whereas the opposite effect occurs with NOX4. These results can be interpreted to mean that NOX2 is a necessary component in pro-angiogenesis responses to Ang-1 although, alternatively, it might be argued that Ang-1 failed to induce migration and differentiation of NOX2-knockdown ECs because basal levels of migration and tube formation had already reached their peak prior to Ang-1 exposure.

Amongst the possible mechanisms through which NOX and mtROS regulate EC migration and differentiation is the process of oxidative inactivation. Several phosphatases that regulate the activity of Focal Adhesion Kinase (FAK), itself a major

108 regulator of cell migration, can be inhibited by ROS. This has been demonstrated in cells exposed to various angiogenesis factors [19,59]. Another means by which ROS may promote EC migration is through glutathiolation of sarcoplasmic reticulum Ca2+-ATPase

(SERCA), an enzyme complex that regulates reticular Ca2+ levels [14], since release of

Ca2+ from the sarcoplasmic reticulum is an important event in growth factor induction of cell migration. It is also possible that NOX4 and mtROS modulate the activities of phosphatases such as PTEN and DUSPs, which regulate the PI-3 kinase/AKT, ERK1/2, and SAPJ/JNK pathways. These pathways are known to promote EC migration and differentiation downstream from Tie-2 receptors [1,3,60-62]. Finally, ROS activation of

RhoA and Rac1, GTPases that are involved in pro-migration responses to Ang-1 [61,63], and/or direct oxidation and stabilization of the F-actin filament may also be possible mechanisms through which ROS regulate cell migration.

Conclusions: Ang-1 triggers rapid but transient Tie-2-dependent ROS increases in endothelial cells. NOX2, NOX4, and the mitochondria contribute to Ang-1-induced increases in ROS. NOX2, NOX4, and mtROS selectively regulate the AKT, ERK1/2, p38, and SAPK/JNK pathways downstream from Tie-2 receptors. NOX2 is necessary to

Ang-1-induced survival of serum-deprived endothelial cells. NOX4 and mtROS are necessary to Ang-1-induced endothelial cell migration and tube formation.

109

2.6 Acknowledgement This work was supported by the Canadian Institutes of Health Research. The authors are grateful to Ms. Anne Gatensby for editing the manuscript.

110

Conflict of interest

The authors declare that they no conflicts of interest.

111

2.7 References

[1] B. Witzenbichler, P.C. Maisonpierre, P. Jones, G.D. Yancopoulos, J.M. Isner, Chemotactic properities of angiopoietin-1 and -2, ligands for the endothelial- specific tyrosine kinase Tie2. J.Biol Chem. 273 (1998) 18514-18521.

[2] N.A. Abdel-Malak, C.B. Srikant, A.S. Kristof, S.A. Magder, J.A. Di Battista, S.N.A. Hussain, Angiopoietin-1 promotes endothelial proliferation and migration through AP-1 dependent autocrine production of interleukin-8. Blood 111 (2008) 4145-4154.

[3] N.A. Abdel-Malak, M. Mofarrahi, D. Mayaki, L.M. Khachigian, S.N. Hussain, Early Growth Response-1 Regulates Angiopoietin-1-Induced Endothelial Cell Proliferation, Migration, and Differentiation. Arterioscler.Thromb.Vasc.Biol. 29 (2008) 209-216.

[4] I. Kim, H.G. Kim, S.O. Moon, S.W. Chae, J.N. So, K.N. Koh, B.C. Ahn, G.Y. Koh, Angiopoietin-1 induces endothelial cell sprouting through the activation of focal adhesion kinase and plasmin secretion. Circ.Res. 86 (2000) 952-959.

[5] R. Harfouche, H.M. Hassessian, Y. Guo, V. Faivre, C.B. Srikant, G.D. Yancopoulos, S.N.A. Hussain, Mechanisms which mediate the anti-apoptotic effects of angiopoietin-1 on endothelial cells. Microvasc.Res. 64 (2002) 135-147.

[6] C.H. Cho, H.K. Sung, K.T. Kim, H.G. Cheon, G.T. Oh, H.J. Hong, O.J. Yoo, G.Y. Koh, COMP-angiopoietin-1 promotes wound healing through enhanced angiogenesis, lymphangiogenesis, and blood flow in a diabetic mouse model. Proc.Natl.Acad.Sci.U.S.A 103 (2006) 4946-4951.

[7] C. Suri, J. McClain, G. Thurston, D.M. McDonald, H. Zhou, E.H. Oldmixon, T.N. Sato, G.D. Yancopoulos, Increased vascularization in mice overexpressing angiopoietin-1. Science 282 (1998) 468-471.

[8] R. Harfouche, J.P. Gratton, G.D. Yancopoulos, S.N.A. Hussain, Angiopoietin-1 activates both anti- and pro-apoptotic mitogen activated protein kinases. FASEB J. 17 (2003) 1523-1525.

[9] N. Abdel-Malak, R. Harfouche, S.N. Hussain, Transcriptome of angiopoietin 1- activated human umbilical vein endothelial cells. Endothelium 14 (2007) 285- 302.

[10] Y.S. Bae, S.W. Kang, M.S. Seo, I.C. Baines, E. Tekle, P.B. Chock, S.G. Rhee, Epidermal growth factor (EGF)-induced generation of hydrogen peroxide. Role in EGF receptor-mediated tyrosine phosphorylation. J.Biol.Chem. 272 (1997) 217- 221.

112

[11] M. Ushio-Fukai, VEGF signaling through NADPH oxidase-derived ROS. Antioxid.Redox.Signal. 9 (2007) 731-739.

[12] J.D. van Buul, M. Fernandez-Borja, E.C. Anthony, P.L. Hordijk, Expression and localization of NOX2 and NOX4 in primary human endothelial cells. Antioxid.Redox.Signal. 7 (2005) 308-317.

[13] L. Serrander, L. Cartier, K. Bedard, B. Banfi, B. Lardy, O. Plastre, A. Sienkiewicz, L. Forro, W. Schlegel, K.H. Krause, NOX4 activity is determined by mRNA levels and reveals a unique pattern of ROS generation. Biochem.J. 406 (2007) 105-114.

[14] A.M. Evangelista, M.D. Thompson, V.M. Bolotina, X. Tong, R.A. Cohen, Nox4- and Nox2-dependent oxidant production is required for VEGF-induced SERCA cysteine-674 S-glutathiolation and endothelial cell migration. Free Radic.Biol.Med. 53 (2012) 2327-2334.

[15] R. Harfouche, N.A. Malak, R.P. Brandes, A. Karsan, K. Irani, S.N. Hussain. Roles of reactive oxygen species in angiopoietin-1/tie-2 receptor signaling. FASEB J. 19[12], 1728-1730. 2005.

[16] Y.M. Kim, K.E. Kim, G.Y. Koh, Y.S. Ho, K.J. Lee, Hydrogen peroxide produced by angiopoietin-1 mediates angiogenesis. Cancer Res. 66 (2006) 6167-6174.

[17] J.X. Chen, H. Zeng, M.L. Lawrence, T.S. Blackwell, B. Meyrick, Angiopoietin-1- induced Angiogenesis is Modulated by Endothelial NADPH Oxidase. Am.J.Physiol Heart Circ.Physiol 291 (2006) H1563-H1572.

[18] M.L. Taddei, M. Parri, T. Mello, A. Catalano, A.D. Levine, G. Raugei, G. Ramponi, P. Chiarugi, Integrin-mediated cell adhesion and spreading engage different sources of reactive oxygen species. Antioxid.Redox.Signal. 9 (2007) 469-481.

[19] J. Frijhoff, M. Dagnell, M. Augsten, E. Beltrami, M. Giorgio, A. Ostman, The mitochondrial reactive oxygen species regulator p66Shc controls PDGF-induced signaling and migration through protein tyrosine phosphatase oxidation. Free Radic.Biol.Med. 68 (2014) 268-277.

[20] Y. Wang, Q.S. Zang, Z. Liu, Q. Wu, D. Maass, G. Dulan, P.W. Shaul, L. Melito, D.E. Frantz, J.A. Kilgore, N.S. Williams, L.S. Terada, F.E. Nwariaku, Regulation of VEGF Induced Endothelial Cell Migration by Mitochondrial Reactive Oxygen Species. Am.J Physiol Cell Physiol2011).

[21] K.M. Connor, S. Subbaram, K.J. Regan, K.K. Nelson, J.E. Mazurkiewicz, P.J. Bartholomew, A.E. Aplin, Y.T. Tai, J. Aguirre-Ghiso, S.C. Flores, J.A. Melendez, Mitochondrial H2O2 regulates the angiogenic phenotype via PTEN oxidation. J.Biol.Chem. 280 (2005) 16916-16924. 113

[22] P. Sinnayah, T.E. Lindley, P.D. Staber, M.D. Cassell, B.L. Davidson, R.L. Davisson, Selective gene transfer to key cardiovascular regions of the brain: comparison of two viral vector systems. Hypertension 39 (2002) 603-608.

[23] K. Zhao, G.M. Zhao, D. Wu, Y. Soong, A.V. Birk, P.W. Schiller, H.H. Szeto, Cell-permeable peptide antioxidants targeted to inner mitochondrial membrane inhibit mitochondrial swelling, oxidative cell death, and reperfusion injury. J Biol.Chem. 279 (2004) 34682-34690.

[24] H.L. Liang, F. Sedlic, Z. Bosnjak, V. Nilakantan, SOD1 and MitoTEMPO partially prevent mitochondrial permeability transition pore opening, necrosis, and mitochondrial apoptosis after ATP depletion recovery. Free Radic.Biol.Med. 49 (2010) 1550-1560.

[25] R. Echavarria, S.N. Hussain, Regulation of angiopoietin-1/Tie-2 receptor signaling in endothelial cells by dual-specificity phosphatases 1, 4, and 5. J Am Heart Assoc. 2 (2013) e000571.

[26] A.J. Case, S. Li, U. Basu, J. Tian, M.C. Zimmerman, Mitochondrial-localized NADPH oxidase 4 is a source of superoxide in angiotensin II-stimulated neurons. Am.J.Physiol Heart Circ.Physiol 305 (2013) H19-H28.

[27] R. Koziel, H. Pircher, M. Kratochwil, B. Lener, M. Hermann, N.A. Dencher, P. Jansen-Durr, Mitochondrial respiratory chain complex I is inactivated by NADPH oxidase Nox4. Biochem.J. 452 (2013) 231-239.

[28] K. Block, Y. Gorin, H.E. Abboud, Subcellular localization of Nox4 and regulation in diabetes. Proc.Natl.Acad.Sci.U.S.A 106 (2009) 14385-14390.

[29] N.P. Brindle, P. Saharinen, K. Alitalo, Signaling and functions of angiopoietin-1 in vascular protection. Circ.Res. 98 (2006) 1014-1023.

[30] B.M. Emerling, L.C. Platanias, E. Black, A.R. Nebreda, R.J. Davis, N.S. Chandel, Mitochondrial reactive oxygen species activation of p38 mitogen-activated protein kinase is required for hypoxia signaling. Mol.Cell Biol. 25 (2005) 4853- 4862.

[31] S.D. Hingtgen, X. Tian, J. Yang, S.M. Dunlay, A.S. Peek, Y. Wu, R.V. Sharma, J.F. Engelhardt, R.L. Davisson, Nox2-containing NADPH oxidase and Akt activation play a key role in angiotensin II-induced cardiomyocyte hypertrophy. Physiol Genomics 26 (2006) 180-191.

[32] C.X. Santos, A.D. Hafstad, M. Beretta, M. Zhang, C. Molenaar, J. Kopec, D. Fotinou, T.V. Murray, A.M. Cobb, D. Martin, S.M. Zeh, N. Anilkumar, K. Schroder, C.M. Shanahan, A.C. Brewer, R.P. Brandes, E. Blanc, M. Parsons, V. Belousov, R. Cammack, R.C. Hider, R.A. Steiner, A.M. Shah, Targeted redox inhibition of protein phosphatase 1 by Nox4 regulates eIF2alpha-mediated stress signaling. EMBO J. 35 (2016) 319-334.

114

[33] D. Meng, X. Shi, B.H. Jiang, J. Fang, Insulin-like growth factor-I (IGF-I) induces epidermal growth factor receptor transactivation and cell proliferation through reactive oxygen species. Free Radic.Biol.Med. 42 (2007) 1651-1660.

[34] K.Z. Guyton, Y. Liu, M. Gorospe, Q. Xu, N.J. Holbrook, Activation of mitogen- activated protein kinase by H2O2. Role in cell survival following oxidant injury. J.Biol.Chem. 271 (1996) 4138-4142.

[35] L. Mitchell, G.A. Hobbs, A. Aghajanian, S.L. Campbell, Redox regulation of Ras and Rho GTPases: mechanism and function. Antioxid.Redox.Signal. 18 (2013) 250-258.

[36] H. Kamata, S. Honda, S. Maeda, L. Chang, H. Hirata, M. Karin, Reactive oxygen species promote TNFalpha-induced death and sustained JNK activation by inhibiting MAP kinase phosphatases. Cell 120 (2005) 649-661.

[37] J.M. Li, L.M. Fan, V.T. George, G. Brooks, Nox2 regulates endothelial cell cycle arrest and apoptosis via p21cip1 and p53. Free Radic.Biol.Med. 43 (2007) 976- 986.

[38] S. Basuroy, S. Bhattacharya, C.W. Leffler, H. Parfenova, Nox4 NADPH oxidase mediates oxidative stress and apoptosis caused by TNF-alpha in cerebral vascular endothelial cells. Am.J.Physiol Cell Physiol 296 (2009) C422-C432.

[39] S.S. Deshpande, P. Angkeow, J. Huang, M. Ozaki, K. Irani, Rac1 inhibits TNF- alpha-induced endothelial cell apoptosis: dual regulation by reactive oxygen species. FASEB J. 14 (2000) 1705-1714.

[40] T. Mochizuki, S. Furuta, J. Mitsushita, W.H. Shang, M. Ito, Y. Yokoo, M. Yamaura, S. Ishizone, J. Nakayama, A. Konagai, K. Hirose, K. Kiyosawa, T. Kamata, Inhibition of NADPH oxidase 4 activates apoptosis via the AKT/apoptosis signal-regulating kinase 1 pathway in pancreatic cancer PANC-1 cells. Oncogene 25 (2006) 3699-3707.

[41] A. Papapetropoulos, D. Fulton, K. Mahboubi, R.G. Kalb, D.S. O'Connor, F. Li, D.C. Altieri, W.C. Sessa, Angiopoietin-1 inhibits endothelial cell apoptosis via the Akt/Survivin pathway. J.Biol Chem. 275 (2000) 9102-9105.

[42] I. Kim, H.G. Kim, J.N. So, J.H. Kim, H.J. Kwak, Y. Koh, Angiopoietin-1 regulates endothelial cell survival through the phosphatidylinsitol 3'-kinase/AKT signal transduction pathway. Cir.Res. 86 (2000) 24-29.

[43] H. Jeon, Y.C. Boo, Laminar shear stress enhances endothelial cell survival through a NADPH oxidase 2-dependent mechanism. Biochem.Biophys.Res.Commun. 430 (2013) 460-465.

115

[44] B.H. Jeon, Y.C. Boo, Roles of NADPH oxidase 2 and 4 in endothelial cell survival and death under serum depletion. Advances in Biological Chemistry 4 (2014) 10-19.

[45] B.I. Rosc-Schluter, S.P. Hauselmann, V. Lorenz, M. Mochizuki, F. Facciotti, O. Pfister, G.M. Kuster, NOX2-derived reactive oxygen species are crucial for CD29-induced pro-survival signalling in cardiomyocytes. Cardiovasc.Res. 93 (2012) 454-462.

[46] J. Tickner, L.M. Fan, J. Du, D. Meijles, J.M. Li, Nox2-derived ROS in PPARgamma signaling and cell-cycle progression of lung alveolar epithelial cells. Free Radic.Biol.Med. 51 (2011) 763-772.

[47] F. Yan, Y. Wang, X. Wu, H.M. Peshavariya, G.J. Dusting, M. Zhang, F. Jiang, Nox4 and redox signaling mediate TGF-beta-induced endothelial cell apoptosis and phenotypic switch. Cell Death.Dis. 5 (2014) e1010.

[48] K.D. Martyn, L.M. Frederick, L.K. von, M.C. Dinauer, U.G. Knaus, Functional analysis of Nox4 reveals unique characteristics compared to other NADPH oxidases 3655. Cell Signal. 18 (2006) 69-82.

[49] M.V. Clement, S. Pervaiz, Intracellular superoxide and hydrogen peroxide concentrations: a critical balance that determines survival or death. Redox.Rep. 6 (2001) 211-214.

[50] M. Shibanuma, T. Kuroki, K. Nose, Superoxide as a signal for increase in intracellular pH. J.Cell Physiol 136 (1988) 379-383.

[51] D.S. Kaufman, M.S. Goligorsky, E.P. Nord, M.L. Graber, Perturbation of cell pH regulation by H2O2 in renal epithelial cells. Arch.Biochem.Biophys. 302 (1993) 245-254.

[52] D. Italiano, A.M. Lena, G. Melino, E. Candi, Identification of NCF2/p67phox as a novel p53 target gene. Cell Cycle 11 (2012) 4589-4596.

[53] S.R. Datla, H. Peshavariya, G.J. Dusting, K. Mahadev, B.J. Goldstein, F. Jiang, Important role of Nox4 type NADPH oxidase in angiogenic responses in human microvascular endothelial cells in vitro. Arterioscler.Thromb.Vasc.Biol. 27 (2007) 2319-2324.

[54] S.M. Craige, K. Chen, Y. Pei, C. Li, X. Huang, C. Chen, R. Shibata, K. Sato, K. Walsh, J.F. Keaney, Jr., NADPH oxidase 4 promotes endothelial angiogenesis through endothelial nitric oxide synthase activation. Circulation 124 (2011) 731- 740.

[55] M. Ushio-Fukai, Y. Tang, T. Fukai, S.I. Dikalov, Y. Ma, M. Fujimoto, M.T. Quinn, P.J. Pagano, C. Johnson, R.W. Alexander, Novel role of gp91(phox)-

116

containing NAD(P)H oxidase in vascular endothelial growth factor-induced signaling and angiogenesis. Circ.Res. 91 (2002) 1160-1167.

[56] S. Ikeda, M. Yamaoka-Tojo, L. Hilenski, N.A. Patrushev, G.M. Anwar, M.T. Quinn, M. Ushio-Fukai, IQGAP1 regulates reactive oxygen species-dependent endothelial cell migration through interacting with Nox2. Arterioscler.Thromb.Vasc.Biol. 25 (2005) 2295-2300.

[57] P. Haddad, S. Dussault, J. Groleau, J. Turgeon, S.E. Michaud, C. Menard, G. Perez, F. Maingrette, A. Rivard, Nox2-containing NADPH oxidase deficiency confers protection from hindlimb ischemia in conditions of increased oxidative stress. Arterioscler.Thromb.Vasc.Biol. 29 (2009) 1522-1528.

[58] L.K. von, D. Noack, M.R. Wood, J.S. Friedman, U.G. Knaus, Structural insights into Nox4 and Nox2: motifs involved in function and cellular localization. Mol.Cell Biol. 30 (2010) 961-975.

[59] M.A. Abdelsaid, A.B. El-Remessy, S-glutathionylation of LMW-PTP regulates VEGF-mediated FAK activation and endothelial cell migration. J.Cell Sci. 125 (2012) 4751-4760.

[60] Z. Master, N. Jones, J. Tran, J. Jones, R.S. Kerbel, D.J. Dumont, Dok-R plays a pivotal role in angiopoietin-1-dependent cell migration through recruitment and activation of Pak. EMBO J. 20 (2001) 5919-5928.

[61] N. Jones, S.H. Chen, C. Sturk, Z. Master, J. Tran, R.S. Kerbel, D.J. Dumont, A unique autophosphorylation site on Tie2/Tek mediates Dok-R phosphotyrosine binding domain binding and function. Mol.Cell Biol. 23 (2003) 2658-2668.

[62] I. Cascone, E. Audero, E. Giraudo, L. Napione, F. Maniero, M.R. Philips, J.G. Collard, G. Serini, F. Bussolino, Tie-2-dependent activation of RhoA and Rac1 participates in endothelial cell motility triggered by angiopoietin-1. Blood 102 (2003) 2482-2490.

[63] E. Audero, I. Cascone, F. Maniero, L. Napione, M. Arese, L. Lanfrancone, F. Bussolino, Adaptor ShcA protein binds tyrosine kinase Tie2 receptor and regulates migration and sprouting but not survival of endothelial cells. J.Biol.Chem. 279 (2004) 13224-13233.

117

2.8 Figures

Figure 2.1: Ang-1 Regulation of ROS in HUVECs

118

Figure 2.2: NOX2 and NOX4 Contributions to ROS Production

119

Figure 2.3: Ang-1 Regulation of mtROS

120

Figure 2.4: NOX2 Regulation of Ang-1 Signaling

121

Figure 2.5: NOX4 Regulation of Ang-1 Signaling

122

Figure 2.6: mtROS Regulation of Ang-1 Signaling

123

Figure 2.7: Roles of NOX2 and NOX4 in Cell Survival

124

Figure 2.8: mtROS Effects on Cell Survival

125

Figure 2.9: ROS Effects on Cell Migration

126

Figure 2.10: Roles of ROS in Cell Differentiation

127

Supplementary Figure S2.1: Copie number of NOXs and DUOX isoforms in HUVECs

128

Supplementary Figure S2.2: Verification of NOXs knockdown

129

Supplementary Figure S2.3: Ang-1 induced Mitochondrial ROS in NOXs knockdown

130

131

2.9 Figure legends

Figure 2.1: Ang-1 Regulation of ROS in HUVECs

Top: Amplex Red fluorescent detection of ROS in goat IgG antibody-treated (4 h)

HUVECs (control) exposed post-treatment to PBS or Ang-1 (300 ng/ml) for 60 min. N=6 per group. *p<0.05, compared to PBS.

Bottom: Amplex Red fluorescent detection of ROS in neutralizing goat Tie-2 antibody- treated (4 h) HUVECs exposed post-treatment to PBS or Ang-1 (300 ng/ml) for 60 min.

N=6 per group. *p<0.05, compared to PBS.

Figure 2.2: NOX2 and NOX4 Contributions to ROS Production

Amplex Red fluorescent detection of ROS in Ad-shGFP- or Ad-shNOX2-, or Ad- shNOX4-infected HUVECs exposed 72 h post-infection to PBS or Ang-1 (300 ng/ml) for

60 min. N=12 per group. Values are means  SEM. *p<0.05, compared to PBS alone.

p<0.05, compared to Ad-shGFP-infected cells in PBS.

Figure 2.3: Ang-1 Regulation of mtROS

A: MitoSOX fluorescent detection of mtROS in SS31- or SS20-treated HUVECs permeated 4 h post-treatment with MitoSOX reagent for 30 min then exposed to PBS or

Ang-1 (300 ng/ml) for 15 min. N=12 per group.

B: MitoSOX fluorescent detection of mtROS in IgG- or neutralizing Tie-2 IgG antibody- treated HUVECs permeated 4 h post-treatment with MitoSOX reagent for 30 min then exposed to PBS or Ang-1 (300 ng/ml) for 15 min. N=6 per group. Values are means 

SEM. *p<0.05, compared to cells exposed to PBS alone.

132

Figure 2.4: NOX2 Regulation of Ang-1 Signaling

Representative immunoblots and optical densities (OD) of phosphorylated AKT,

ERK1/2, p38, and SAPK/JNK proteins in Ad-shGFP- or Ad-shNOX2-infected HUVECs exposed 72 h post-infection to Ang-1 (300 ng/ml) for 5 or 15 min. Time 0 equals control

(no Ang-1). N=4 per group. Values are means  SEM and are expressed as fold change from control values measured in cells infected with Ad-shGFP. *p<0.05, compared to controls.

Figure 2.5: NOX4 Regulation of Ang-1 Signaling

Representative immunoblots and optical densities (OD) of phosphorylated AKT,

ERK1/2, p38, and SAPK/JNK proteins in Ad-shGFP- or Ad-shNOX4-infected HUVECs exposed 72 h post-infection to Ang-1 (300 ng/ml) for 5 or 15 min. Time 0 equals control

(no Ang-1). N=4 per group. Values are means  SEM and are expressed as fold change from control values measured in cells infected with Ad-shGFP. *p<0.05, compared to controls.

Figure 2.6: mtROS Regulation of Ang-1 Signaling

Representative immunoblots and optical densities (OD) of phosphorylated AKT,

ERK1/2, p38, and SAPK/JNK proteins in SS20- or SS31-pre-treated HUVECs exposed

24 h post-treatment to Ang-1 (300 ng/ml) for 5 or 15 min. Time 0 equals control (no

Ang-1). N=3 per group. Values are means  SEM and are expressed as fold change from control values measured in cells infected with Ad-shGFP. *p<0.05, compared to controls.

133

Figure 2.7: Roles of NOX2 and NOX4 in Cell Survival

A: Cell number of Ad-shGFP-, Ad-shNOX2-, and Ad-shNOX4-infected HUVECs seeded 48 h post-infection in 12-well plates (8x104/cm2) then incubated in full (20%

FBS), basal (2% FBS), or basal medium containing Ang-1 (300 ng/ml) for 24 h. N=12 per group. Values are means  SEM. *p<0.05, compared to full medium. #p<0.05, compared to basal medium. p<0.05, compared to Ad-shGFP-infected cells in basal medium.

B-C: Representative immunoblots of cleaved Caspase-3 and -Actin and optical densities

(OD) of cleaved Caspase-3 in Ad-shGFP-, Ad-shNOX2-, and Ad-shNOX4-infected

HUVECs incubated 48 h post-infection in full (20% FBS), basal (2% FBS), or basal medium containing Ang-1 (300 ng/ml) for 24 h. N=4 per group. Values are means 

SEM *p<0.05, compared to full medium. #p<0.05, compared to basal medium.

Figure 2.8: mtROS Effects on Cell Survival

A, B: Cell number and optical densities (OD) of cleaved Caspase-3 in SS20 and SS31 pre-treated HUVECs seeded 24 h post-treatment in 12-well plates (8x104/cm2) then incubated in full (20% FBS), basal (2% FBS), or basal medium containing Ang-1 (300 ng/ml) for 24 h. N=12 per group for cell counting, N=4 for OD. Values are means 

SEM. *p<0.05, compared to full medium. #p<0.05, compared to basal medium, p<0.05, compared to SS20-treated cells in basal medium.

134

C, D: Cell number and optical densities (OD) of cleaved Caspase-3 in vehicle and Mito-

TEMPOL pre-treated HUVECs seeded 24 h post-treatment in 12-well plates (8x104/cm2) then incubated in full (20% FBS), basal (2% FBS), or basal medium containing Ang-1

(300 ng/ml) for 24 h. N=12 per group for cell counting, N=4 for OD. Values are means 

SEM. *p<0.05, compared to full medium. #p<0.05, compared to basal medium, p<0.05, compared to SS20-treated cells in basal medium.

Figure 2.9: ROS Effects on Cell Migration

A: Representative micrographs of wound healing experiments. Ad-shGFP-infected

HUVECs were wounded (t=0 h) then maintained in basal medium containing PBS or

Ang-1 (300 ng/ml) for 8 h. Dotted lines delineate the width of wound areas at t=0. Note that wound healing at t=8 was more advanced in Ang-1 exposed cells.

B: Percent wound healing in Ad-shGFP-, Ad-shNOX2-, and Ad-shNOX4-infected

HUVECs wounded (t=0 h) then maintained in basal medium containing PBS or Ang-1

(300 ng/ml) for 8 h. N=12 per group. Values are means ± SEM. *p<0.05, compared to

PBS alone. p<0.05, compared to Ad-shGFP-infected cells in PBS.

C, D: Percent wound healing in SS20-, SS31-, DMSO (vehicle)-, and Mito-TEMPOL- pre-treated HUVECs wounded (t= 0 h) then maintained in basal medium containing PBS or Ang-1 (300 ng/ml) for 8 h. N=12 per group. Values are means ± SEM. *p<0.05, compared to PBS alone.

Figure 2.10: Roles of ROS in Cell Differentiation

135

A-B: Representative micrograph of capillary tube formation and total tube length of Ad- shGFP-, Ad-shNOX2-, and Ad-shNOX4-infected HUVECs maintained 48 h post- infection in growth factor-reduced Matrigel-pre-coated plates with basal medium containing PBS or Ang-1 (300 ng/ml) for 24 h. N=12 per group. Values are means 

SEM. *P<0.05 compared to PBS alone. p<0.05, compared Ad-shGFP-infected cells in

PBS.

C, D: Total tube length of SS20-, SS31-, DMSO (vehicle)-, and Mito-TEMPOL-pre- treated HUVECs HUVECs maintained 24 h post-treatment in growth factor-reduced

Matrigel-pre-coated plates with basal medium containing PBS or Ang-1 (300 ng/ml) for 24 h. N=12 per group. Values are means  SEM.*P<0.05, compared to PBS alone.

Supplementary Figure S2.1: Copie number of NOXs and DUOX isoforms in

HUVECs Means SEM of mRNA copy number of NOX and DUOX isoforms measured with qPCR in HUVECs. Values are normalized per 105 copies of -ACTIN. Notice the relatively high abundance of NOX4 and NOX2 mRNA expression in these cells.

Supplementary Figure S2.2: Verification of NOXs knockdown A-B: mRNA expression of NOX2 and NOX4 (A) and NOX2 protein (B) in HUVECs infected with

Ad-shGFP, Ad-shNOX2 or Ad-shNOX4. mRNA expression was measured with qPCR while NOX2 and -ACTIN proteins were detected with immunoblotting. *p<0.05, as compared with cells infected with Ad-shGFP.

Supplementary Figure S2.3: Ang-1 induced Mitochondrial ROS in NOXs knockdown. Means SEM of MitoSOX fluorescence measured in HUVECs infected 48h

136 earlier with Ad-GFP, Ad-shNOX2 or Ad-shNOX4 and were exposed to PBS or Ang-1

(300 ng/ml). Fluorescence was measured 15 min after PBS or Ang-1 exposure. *p<0.05, as compared to corresponding PBS treatment.  p<0.05, as compared to PBS treatment in

Ad-shGFP-infected cells.

137

2.10 Tables

Table 2.1: Primers used for quantitative real-time PCR experiments

Type Sequence (5’ 3’) Accession Expected Size Number (bp)

-Actin F: AGAAAATCTGGCACCACACC NM_001101 126 R: GGGGTGTTGAAGGTCTCAAA

DUOX1 F: TTCACGCAGCTCTGTGTCAA NM_017434.3 96 R: AGGGACAGATCATATCCTGGCT

DUOX2 F: ACGCAGCTCTGTGTCAAAGGT NM_014080 70 R: TGATGAACGAGACTCGACAGC

NOX1 F: CTTCCTCACTGGCTGGGATA NM_013955 199 R: TGACAGCATTTGCGCAGGCT

NOX2 F: CCAGTGAAGATGTGTTCAGCT NM_000397 154 R: GCACAGCCAGTAGAAGTAGAT

NOX3 F: GGGAAGCCCTTACTAACGACC NM_015718.1 708 R: CCAACTGGGAGCTCATATCAA

NOX4 F: TTGTCTTCTACATGCTGCTG NM_016931.4 285 R: AGGCACAAAGGTCCAGAAAT

NOX5 F: CACTGTGCTGCAGTCGTGTC NM_024505 240 R: GAGCTCCTCGAAGGTGATGG

138

Preface to chapter 3 miRNAs have recently been recognized as key regulators of vascular diseases, inflammation, and angiogenesis[380;381]. Specific sets of miRNAs have been shown to regulate the expressions and actions of particular angiogenic factors in ECs, for example vascular endothelial growth factor (VEGF) [382]. No information is yet available regarding the nature of microRNAs that are regulated by the Ang-1/Tie-2 axis in ECs.

We, therefore, conducted pilot experiments in which we measured the expressions of miRNAs in HUVECs exposed to Ang-1 (300 ng/ml) for 24hrs and 48 hrs. Expressions of

1,105 human mature miRNAs and 1,105 human pre-miRNAs were measured using

Affymetrix miRNA Arrays 2.0 (Genome Quebec). FlexArray software was used to determine statistical significance which was set at changes  2.5-fold and  0.4-fold.

Ang-1 triggered significant upregulation of ten mature miRNAs two of which (miR-126 and miR-146b) are known to be expressed in ECs and their roles in angiogenesis (miR-

126) and inflammation (miR-146b) have been investigated. In addition, exposure to

Ang-1 triggered significant downregulation of ten mature miRNAs. Among these was miR-640, a miRNA not well characterized in terms of expression and functional importance in the regulation of angiogenesis. In this study, we hypothesized miR-640 miRNAs negatively regulate critical processes involved in angiogenesis such as EC proliferation, survival, migration an differentiation and that downregulation of these microRNAs may be necessary for Ang-1/Tie-2 signalling processes to proceed.

139

Chapter 3: Regulation of angiogenesis by Angiopoietin- 1: Roles of miR-640 and ZFP91

140

Regulation of angiogenesis by Angiopoietin-1: Roles of miR-640 and ZFP91

Sharon Harel, Veronica Sanchez-Gonzalez, Raquel Echavarria, Dominique Mayaki, and

Sabah NA Hussain

Meakins Christie Laboratories, and Translational Research in Respiratory Diseases Program, McGill University Health Centre and Research Institute; Department of Critical Care, McGill University Health Centre, Montreal, Québec, Canada.

Corresponding author: Dr. Sabah Hussain Room EM2.2224 Research Institute of McGill University Health Centre 1001 Decarie Blvd, Montreal, Quebec H4A 3J1 Canada Tel: 514-934-1934 x34645 Email: [email protected]

141

Key Points

 Angipoitein-1 upregulates ZFP91, by downregulation of miR-640 to promote

endothelial cell migration and tube formation

142

3.1 Abstract

Angiopoietin-1 (Ang-1) is a ligand of Tie-2 receptors that promotes survival, migration, and differentiation of endothelial cells. Several studies have established the existence of regulatory loops between angiogenic growth factors and distinct pro or anti-angiogenic miRNAs, but to date Ang-1-regulated miRNAs and their function in angiogenesis have never been clearly identified. In this study, we have identified miR-640, as an Ang-1- regulated miRNA involved in Ang-1 angiogenic responses in human umbilical vein endothelial cells (HUVECs). We demonstrate that: 1) Ang-1 induces significant decrease in miR-640 levels 2) Increase in miR-640 levels inhibit Ang-1-induced cell migration and tube formation and inhibition of miR-640 enhances these responses 3) miR-640 directly targets Zinc Finger Protein 91 (ZFP91) an atypical E3-ubiquitin ligase, also acting as a transcription factor 4) Ang-1 induces ZFP91 through down-regulation of miR-640 and 5) ZFP91 silencing inhibits Ang-1-induced cell migration and tube formation. We conclude that Ang-1 triggers an upregulation of ZFP91 through transcriptional down-regulation of miR-640 and both miR-640 and ZFP91 play important roles in the regulation of the Ang-1/Tie 2 signaling pathway and pro-angiogenic responses.

143

3.2 Introduction

Angiopoietin-1 (Ang-1) and its receptor, Tie-2, are important modulators of physiological and pathological angiogenesis. In mice embryos, deletion of the Ang-1 gene leads to major defects in the vascular endothelium 1. In cultured endothelial cells

(ECs), Ang-1 inhibits apoptosis and inflammatory responses and promotes differentiation, sprouting, and migration 2. In vivo, Ang-1 enhances collateral vessel formation in ischemia-induced angiogenesis 3. Exposure of ECs to Ang-1 triggers the autophosphorylation of Tie-2 receptors and the activation the PI-3 kinase and the mitogen activated protein kinases (MAPKs) (Erk1/2, p38, and SAPK/JNK) pathways 2;4. Ang-1 therapeutic potential has been proved in pre-clinical studies. However its broad spectrum of biological effects limits its use and fine tuning intracellular signals to promote the desired therapeutic actions would be a better approach. MiRNAs have recently been recognized as key regulators of vascular diseases and angiogenesis 5;6. Their manipulation can reveal new targets in the settings of pathological vascularization.

Signal-induced miRNAs are typically pro-angiogenic, contributing to the angiogenic action of different factors such as VEGF or FGF 7;8. To date only one miRNA, miR-

146b-5p, has been shown to be regulated by Ang-1 and to promote Ang-1 anti- inflammatory effects 9. It remains unclear whether or not Ang-1-regulated miRNAs, can modulate Ang-1 effects on EC proliferation, apoptosis, migration and differentiation.

We hypothesized, in ECs the Ang-1/Tie-2 axis promotes angiogenesis through upregulation of specific sets of pro-angiogenic miRNAs and/or downregulation of the expression of a specific set of anti-angiogenic miRNAs. A recent study showed hydrogen sulfide (H2S)-induced angiogenesis depends on downregulation of the expression of miR-

144

640 10. MiR-640, a low-abundance and anti-angiogenic miRNA is down-regulated in cholangiocarcinoma, ovarian carcinoma and lymphocytic leukemia 11-13. So far only one target of miR-640 has been identified. Zhou et al. recently demonstrated miR-640 targets

HIF1 a pro-angiogenic factor 10. In the present study we examined whether repression of miR-640 is also required for Ang-1-induced angiogenesis and we aim at identifying direct targets of miR-640 responsible for Ang-1 pro-angiogenic potential.

145

3.3 Material and Methods

Materials: Recombinant human Ang-1 and recombinant human Ang-2 were purchased from R&D Systems (Minneapolis, MN). Vascular endothelial growth factor (VEGF), fibroblast growth factor 2 (FGF-2) or transforming growth factor beta (TGF-) were purchased from Bioshop (Burlington, ON). Antibodies for ZFP91 and -ACTIN were purchased from Sigma-Aldrich (Oakville, ON). Antibody for cleaved CASPASE-3 was purchased from Cell Signaling Technology (Danvers, MA).

Cell culture: Human umbilical vein endothelial cells were purchased from Lonza (Basel,

Switzerland). Cells were grown in MCDB 131 medium (Life Technologies, Rockville,

MD, USA) supplemented with 20% fetal bovine serum (FBS), EC growth supplement, 2 mM glutamine, heparin, and gentamicin sulfate (complete culture medium), and incubated at 37°C and 5% CO2. HUVECs were pre-incubated for various times with PBS or Ang-1 (300ng/ml). For serum deprivation experiments, cells were incubated in MCDB

131 medium supplemented with 2%FBS, 2 mM glutamine, heparin, and gentamicin sulfate. miRNA analysis: Total RNA was extracted using Qiazol® and miRNeasy Mini Kit

(Qiagen, Hilden, Germany) according to the manufacturers’ protocols. miRNA was detected using an NCode™ miRNA qRT-PCR Kit (Invitrogen) and real-time PCR with specific primers (Supplementary material online, Materials), SYBR® green, and a 7500

Real-Time PCR System. Pri-miR-640 was detected using TaqMan® assays (Applied

Biosystems). All experiments were performed in triplicate. Relative miRNA expression was determined using the CT method where CT values of individual miRNA data were

14 normalized to CT values of U6 snRNA as previously described .

146

Adenoviral Transfection: When HUVECS reached 60-70% confluency they were then infected for 6 h with serum-free medium containing 100 mutiplicity of infection (MOI) virus units of adenoviruses expressing GFP or EX-TEK (Vector Biolabs, Philadelphia,

PA) 15. Cells were left to recover for 48 h prior to experimental treatment with PBS or

Ang-1 (300ng/ml).

Exosome isolation: HUVECs were grown to confluence, starved for 6 hours and treated with or without Ang-1 (300ng/ml) for 24 h. Exosomes excreted from cells into the medium were isolated using miRCURY™ Exosome Isolation Kit (Exiqon, Vedbaek,

Denmark). miRNAs were extracted from the purified exosomes as stated above.

Transfection with siRNA, miRNA mimics and LNA-inhibitors: HUVECs were transfected with 10 nM of a synthetic siRNA against ZFP91 or 15 nM of a synthetic mature miRNA mimic (Ambion) or 25 nM LNA- miRNA Inhibitor (Exiqon, Vedbaek,

Denmark) using Lipofectamine™ RNAiMAX (Invitrogen) according to the manufacturers’ instructions. All experimental treatments were performed 48 h after transfection.

Cell counting: HUVECs were seeded in 12-well plates at a density of 8x104/cm2. Cells were maintained for 24 h in complete medium (containing 20% FBS), serum deprivation medium (containing 2% FBS) in the absence and presence of Ang-1 (300 ng/ml). Cells were counted using a hemocytometer.

Caspase-3 activity assay: HUVECs were seeded at a density of 8x104cells/cm2 in 12 well plates and maintained in complete medium or serum deprivation medium in the absence and presence of Ang-1 (300ng/ml). Caspase-3 activation in cell lysate was

147 analyzed 24 h later by performing immunoblotting using antibody selective to cleaved

Caspase-3.

Scratch Wound Healing Assay: HUVEC monolayers were wounded with a 200 μl pipette tip and maintained for 8 h in serum deprivation medium in the presence of PBS or

Ang-1 (300 ng/ml). Wounded areas were visualized using an Olympus inverted microscope and quantified using Image-Pro Plus™ software (Media Cybernetics,

Bethesda, MD). Values are reported as % wound healing equal [1 – (wound area at t8/wound area at t0)] x 100, where t8 is the time (8 h) over which cells were maintained in media and t0 is the time immediately after wounding, as previously described 16.

Capillary Tube Formation: HUVECs were seeded in 96-well plates pre-coated with growth factor-reduced Matrigel at a density of 1x104 cells per well. Cells were maintained for 24 h in serum deprivation medium in the presence of PBS or Ang-1 (300 ng/ml). Whole-well images were captured using an Olympus inverted microscope (40X) and analyzed using Image-Pro Plus™ software. Angiogenic tube formation was determined by counting total branches in each field, as previously described 16. mRNA analysis: Total RNA was extracted using an Ambion PureLink RNA mini kit

(Life Technologies, Inc, Burlington, Ontario, Canada) according to the manufacturer’s protocol. mRNA was detected using real-time PCR (qPCR) using specific primers

(Supplementary Table 1), SYBR® green, and a 7500 Real-Time PCR System (Applied

Biosystems, Foster City, CA). -ACTIN was used as control gene. All experiments were performed in triplicate. Relative mRNA expression was determined using the

-CT 17 comparative threshold (CT) method (2 ) as previously described

148

Immunoblotting: Total cell lysates were separated using SDS–PAGE, transferred onto polyvinylidene difluoride (PVDF) membranes, blocked with 5% milk and were then probed overnight at 4°C with specific primary antibodies. Proteins were detected using

HRP-conjugated secondary antibodies and ECL reagents as previously described 18.

3’UTR luciferase constructs: p-miTarget™-miRNA 3’UTR Target plasmids containing

2033 bp of wild-type 3’UTRs of ZFP91 (-19-2113 bp, ZFP91-wt-1) and 2013 pb of wild type 3’UTR of ZFP91 (1844-3856 bp, ZFP91-wt-2) cloned downstream of Sv-40-driven

Firefly Luciferase cassette were purchased from GeneCopoeia™ Inc. (Rockville,

MD)(supplementary materials). ZFP91-wt-1 plasmid contains two miR-640 target sites

(seeds 1 and 2) while ZFP91-wt-2 plasmid contains one miR-640 target site (seed 3).

Both plasmids have a CMV-driven Renilla luciferase sequence. Mutated versions of

ZFP91-wt-1 construct (ZFP91-seed1-mut and ZFP91-seed2-mut) carrying 3-bp substitutions in the miR-640 target sites were constructed by site-directed mutagenesis using InFusion HD Cloning kit (Clonetech) (Figure 5). HUVECs were co-transfected with one of these plasmids and control mimic, miR-640 mimic, LNA-scrambled or LNA- anti-miR-640. Cells were allowed to recover for 48 h, starved for 6 h in MCDB 131 medium, and then maintained for 24 h in medium containing PBS or Ang-1. Firefly and

Renilla luciferase activities were quantified using a Dual-Luciferase® Reporter Assay

System and Firefly luciferase activity was normalized for Renilla activity to control for transfection efficiency differences.

Biotin-labeled Pull-down Assays: Biotinylated miR-640 (Exiqon, Denmark) pull-down assay with target mRNAs was performed as described earlier 7;19. Briefly, HUVECs were transfected with 50nM of biotin-control mimic or biotin-miR-640 mimic. After 48 h,

149 whole-cells were lysed. The lysates were mixed with the pre-coated Streptavidin-Dyna beads (Dyna beads M-280 Streptavidin, #11205D, Invitrogen) overnight at 4°C. Beads were then pelleted and washed in lysis buffer. RNA was then isolated with TRIzol

(Invitrogen). RNA was precipitated using standard chloroform-isopropanol method and was then subjected for qPCR detection of specific transcripts.

Data analysis: Data are expressed as means ± SEM. Differences between experimental groups were determined using a Two-Way Analysis of Variance followed by a Student–

Newman–Keuls post-hoc test. P values <0.05 were considered statistically significant.

150

3.4 Results Ang-1 regulation of miR-640: To assess the effect of Ang-1 on miR-640 expression,

HUVECs were pre-incubated for 24, 48 or 72 h with PBS or Ang-1 (300 ng/ml), then expressions of mature miR-640 was measured with qPCR. As compared to results observed with PBS, Ang-1 significantly decreased miR-640 expression at the three time points (Figure 1A). To assess whether these changes in miR-640 expression are due to decreased transcription, precursor (pri-miR-640) levels were measured after 2, 4, 6, and

12 h. Pri-miR-640 transcript levels significantly increased after 2 h of Ang-1 exposure but then declined significantly after 6 and 12 h (Figure 1B). In comparison, mature miR-

640 levels didn’t change after 2 and 4 h but significantly decreased after 6 and 12 h of

Ang-1 exposure (Figure 1B). Exosomes are mediators of intercellular communication and incorporate and transfer mRNAs and miRNAs. To verify if mature miR-640 was secreted as an exosome-associated molecule, we measured mature miR-640 levels in exosomes isolated from HUVECs pre-incubated with Ang-1 for 24 h. Ang-1 exposure triggered significant decrease in exosome miR-640 levels (Figure 1C). These results suggest that the decrease in cellular miR-640 levels triggered by Ang-1 exposure is not due to increased exertion of miR-640 in the exosomes. To assess the importance of Tie-2 receptors in the regulation of miR-640 by Ang-1, cells were infected with Ad-GFP or Ad-

ExTek. Ad-ExTek adenoviruses deliver a recombinant soluble Tie-2 receptor capable of blocking Tie-2 receptor activation in response to Ang-1 exposure. Ang-1 significantly decreased miR-640 levels in cells infected with Ad-GFP. Ang-1 did not alter miR-640 expression in cells infected with Ad-ExTek, indicating that Tie-2 receptor activation is necessary for Ang-1 to have an inhibitory effect on miR-640 expression (Figure 1D).

151

MiR-640 gene is located within an intron of GATAD2A gene. To assess whether Ang-1- mediated decreases in miR-640 expression correlate with that of GATAD2A, we measured the expression of GATAD2A after 2, 4, 6 and 12 h of PBS or Ang-1 exposure.

Ang-1 significantly decreased GATAD2A levels after 12 h suggesting that at that time point, miR-640 expression correlates with that of GATAD2A (Figure 1E). To determine the effects of other angiogenesis factors in the regulation of miR-640 expression,

HUVECs were pre-incubated for 24 hr with PBS, Ang-2, FGF-2, VEGF or TGF-. All these angiogenesis factors significantly decreased miR-640 levels (Figure 1F). This inhibitory effect appears to be specific to miT-640 since the expression of another miRNA downregulated by Ang-1 (miR-211-3p) was increased by Ang-2, decreased by

VEGF and FGF-2 and was not altered by TGF- (supplementary Figure 2).

Regulation of cell survival, migration and differentiation by miR-640: HUVECs were transfected with control mimic or miR-640 mimic. After 48 h of recovery, cells were then exposed to PBS or Ang-1. Experiments were performed to verify that transfection with miR-640 mimic results in significant increases in miR-640 (Supplementary Figure 3).

Ang-1 has been shown to exert pro-survival, pro-migratory and pro-differentiation effect on endothelial cells. In cells transfected with control mimic, serum deprivation alone significantly decreased cell count and significantly increased cleaved Caspase-3 intensity while serum deprivation in combination with Ang-1 increased cell counts and decreased cleaved Caspase-3 intensity (Figure 2), confirming that Ang-1 exerts a pro-survival effect on endothelial cells. This pro-survival effect of Ang-1 was still present in cells transfected with miR-640 mimic (Figure 2). In cells transfected with control mimic, exposure to Ang-1 increased EC migration and capillary tube formation relative to PBS

152 alone (basal migration and differentiation)(Figure 2). In cell transfected with miR-640 mimic, Ang-1 failed to increase cell migration and capillary tube formation (Figure 2).

Taken together these results indicate that over-expression of miR-640 resulted in significant inhibition of the pro-migration and pro-differentiation effects of Ang-1 but had no effect on the pro-survival effects of Ang-1.

To further confirm the roles of miR-640 in Ang-1-induced responses, HUVECs were transfected with LNA-scrambled or LNA-anti-miR-640. After 48 h of recovery cells were exposed to PBS or Ang-1. In cells transfected with LNA-scrambled or LNA-anti- miR-640, exposure to Ang-1 increased cell number and decreased cleaved Caspase-3 intensity as compared to serum deprivation alone (Figure 3). In cells transfected with

LNA-scrambled, exposure to Ang-1 significantly increased cell migration and capillary tube formation relative to PBS (Figure 3). Inhibition of miR-640 with LNA-anti-miR-

640 significantly increased basal cell migration and capillary tube formation (PBS exposure) as compared to cells transfected with LNA-scrambled (Figure 3). In cells transfected with LNA-anti-miR-640, exposure to Ang-1 significantly increased cell migration and capillary tube formation as compared to PBS (Figure 3). Taken together these results suggest that inhibition of miR-640 had no effect on EC survival but significantly increased basal cell migration and tube formation. miR-640 targets ZFP91: To identify potential targets for miR-640 in ECs, we interrogated three databases, the TargetScan (www.targetscan.org), miRSearch3.0

(www.exiqon.com/mirsearch) and miRanda (www.microrna.org)20-22(Figure 4). These three databases predicted ZFP91 as a target for miR-640 because of the presence of three miR-640 binding sites in the 3’ UTR of the ZFP91 mRNA. In cells transfected with

153 miR-640 mimic, ZFP91 mRNA and protein levels were significantly decreased as compared to cells transfected with control mimic (Figure 4). Transfection with LNA- anti-miR-640 significantly increased ZFP91 mRNA and protein levels as compared to cells transfected with LNA-scrambled (Figure 4). To confirm that miR-640 selectively targets ZFP91, HUVECs were transfected with empty reporter, ZFP91-wt-1 or ZFP91- wt-2 luciferase reporter plasmids (Figure 5). These reporter plasmids were transfected along with control mimic, miR-640 mimic, scrambled-LNA or LNA-anti-miR-640. In cells transfected with ZFP91-wt-1 plasmid and control mimic or LNA-scrambled, relative luciferase activity was significantly lower than that measured in cells transfected with empty reporter vector (Figure 5 and Supplementary Figure 3). These responses were not observed in cells transfected with ZFP91-wt-2 (Supplementary Figures 3 and 4).

Transfection with ZFP91-wt-1 plasmid and miR-640 mimic resulted in significantly lower relative luciferase activity as compared to transfection with the same plasmid and control mimic (Figure 5). Transfection with ZFP91-wt-1 plasmid and LNA-anti-miR-

640 resulted in significantly higher relative luciferase activity as compared to transfection with the same plasmid and LNA-scrambled (Supplementary Figure 3). These effects of miR-640 mimic and LNA-anti-miR-640 were not observed in cells transfected with

ZFP91-seed1-mut or ZFP91-seed2-mut (Figure 5). Taken together, these results suggest that endogenous ZFP91-wt1 3’UTR level is regulated by direct binding of miR-640 to seed 1 and 2 but not to seed 3 binding sites and that over-expression of miR-640 leads to a decrease in this level while inhibition of endogenous miR-640 causes a significant increase in ZFP91-wt1 3’UTR levels. To further confirm direct binding of miR-640 to

ZFP91 3’UTR, we performed a miRNA pull down using biotin-labeled miR-640 mimic.

154

Biotin-labeled control mimic served as a control in these experiments. ZFP91 mRNA in the pull-down material were significantly higher and in cell lysates were significantly lower in HUVECs transfected with biotin-labeled miR-640 as compared to those transfected with biotin-labeled control mimic (Figure 5). These results indicate that miR-

640 directly binds to ZFP91 mRNA and that pulling down ZFP91 mRNA with biotin- labeled miR-640 mimic depleted total cellular levels of ZFP91 mRNA.

Ang-1 regulation of ZFP91 expression: To evaluate whether Ang-1 exposure regulates

ZFP91 expression, HUVECs were pre-incubated for 12 and 24 h with PBS or Ang-1 and mRNA and protein levels of ZFP91 were measured. Ang-1 exposure significantly increased ZFP91 mRNA and ZFP91 protein levels (Figure 6). To confirm that these effects of Ang-1 exposure on ZFP91 expression are mediated through modulation of miR-640 levels, HUVECs were transfected with ZFP91-wt-1, ZFP91-seed1-mut or

ZFP91-seed2-mut plasmids and were then exposed for 24 h to PBS or Ang-1. In cells transfected with ZFP91-wt-1, Ang-1 exposure significantly increased relative luciferase activity as compared to PBS exposure (Figure 6). This effect of Ang-1 was not observed in cells transfected with ZFP91-seed1-mut and ZFP91-seed2-mut plasmids (Figure 6).

These results confirm that Ang-1 exposure increases ZFP91 expression through downregulation of miR-640 expression and removal of the inhibitory effect of miR-640 on ZFP91 mRNA expression. We observed that several angiogenesis factors including

Ang-2, VEGF and FGF-2 decrease miR-640 in ECs. To confirm if these angiogenesis factors also regulate ZFP91 expression, HUVECs were pre-incubated for 24 h with PBS,

Ang-2, VEGF or FGF-2. Exposure to Ang-2 and VEGF but not to FGF-2 significantly increased ZFP91 protein levels (Figure 6). These results indicate that ZFP91 expression

155 is regulated by other angiogenesis factors including Ang-2 and VEGF in addition to Ang-

ZFP91 role in Ang-1-induced angiogenesis: HUVECs were transfected with a scrambled siRNA or ZFP91 siRNA. After 48 h of recovery cells were exposed to PBS or

Ang-1 and angiogenic responses (cell survival, migration and differentiation) were then measured. Experiments were performed to verify that transfection with ZFP91 siRNA results in significant decrease of its transcript (Supplementary Figure 5). Exposure to

Ang-1 significantly increases cell number and significantly decreases cleaved Caspase-3 intensity in cells transfected with scrambles siRNA or ZFP91 siRNA oligos suggesting that ZFP91 doesn’t play a major role in the pro-survival effects of Ang-1 (Supplementary

Figures 6 and 7). Ang-1 exposure significantly increases cell migration in cells transfected with scrambled siRNA oligos and in cells co-transfected with scrambled siRNA oligos and LNA-scrambled (Figure 7). Ang-1 exposure fails to increase cell migration in cells transfected with ZFP91 siRNA oligos and cells co-transfected with

ZFP91 siRNA oligos and LNA-anti-miR-640 (Figure 7). In addition, basal cell migration was significantly decreased in cells co-transfected with ZFP91 oligos and LNA-anti-miR-

640 as compared to cells co-transfected with scrambled siRNA oligos and LNA- scrambled (Figure 7).

Ang-1 exposure significantly increases capillary tube formation in cells transfected with scrambled siRNA oligos and in cells co-transfected with scrambled siRNA oligos and LNA-scrambled (Figure 7). Ang-1 exposure fails to increase cell migration in cells transfected with ZFP91 siRNA oligos and cells co-transfected with

ZFP91 siRNA oligos and LNA-anti-miR-640 (Figure 7). In addition, basal capillary tube

156 formation was significantly decreased in cells transfected with ZFP91 siRNA oligos and cells co-transfected with ZFP91 oligos and LNA-anti-miR-640 (Figure 7). These results indicate that silencing ZFP91 significantly decreases basal cell migration and tube formation and prevented the effects of Ang-1 on cell migration and capillary tube formation.

To evaluate the functional effect of ZFP91 on the expression of the pro- angiogenic gene target of the non-canonical NFκB, HUVECs were transfected with scrambled siRNA and ZFP91 siRNA and the expression of genes known to be regulated by non-canonical NFB was measured with qPCR. ZFP91 knockdown with siRNA oligos significantly decreased Ang-1, VEGF, NIK and VCAM1 levels but had had no effects on CCL19 and BAFF expression (supplementary Figure 8). These results suggest that ZFP91 regulates EC migration and differentiation through the regulation of pro- angiogenic gene targets of the non-canonical NFκB pathway.

157

3.5 Discussion The main findings of this study are: 1) Ang-1 induces a Tie-2-dependent decrease in miR-640 levels; 2) miR-640 prevents Ang-1-induced cell migration and tube formation but has no influence on the pro-survival effects of Ang-1; 3) ZFP91 is a direct target of miR-640; and 4) ZFP91 is required for Ang-1-induced endothelial cell migration and tube formation.

We report here, for the first time, that Ang-1 downregulates miR-640 in ECs.

This response appears to be mediated through transcriptional inhibition since the pri- miR-640 is significantly down-regulated together with its encoded target gene

GATAD2A, and miR-640 did not accumulate in exosome in response to Ang-1 exposure

(Figure 1). Our results clearly indicate that miR-640 has important anti-angiogenic biological effects since its expression is downregulated by several pro-angiogenesis factors including VEGF and FGF-2 (Figure 1) and that over-expression of miR-640 using selective mimic results in significant decrease in cell migration and differentiation

(Figures 2-3). These findings are in accordance with a recent study which reported that miR-640 has inhibitory effects on hydrogen sulfide (H2S)-induced cell migration and

10 tube formation . In that study, Zhou et al. concluded miR-640 effect on H2S-induced angiogenesis are mediated through targeting of HIF1. Indeed, silencing HIF abolished

H2S-induced cell migration and tube formation. Interestingly, inhibition of miR-640 led to an increase in basal cell migration and tube formation in Zhou et al. study and in our current study indicating that miR-640 represses these angiogenic responses at basal level.

Others studies, have correlated the decreased levels of miR-640 with poor prognosis in ovarian carcinoma and lymphocytic leukemia 11;13. Our study emphasizes the role of

158 miR-640 as a key player in the regulation of angiogenesis, and that miR-640 expression can be modulated by various angiogenesis factors.

There is no current evidence Ang-1 regulates HIF1, the only known target of miR-640. Therefore, to understand how the pro-angiogenic signals of Ang-1 were transduced downstream of miR-640, we had to identify additional targets of miR-640.

Computational analyses predicted three putative miR-640 binding sites in the 3′-UTR of

ZFP91. In the current study, we confirm that miR-640 directly targets the 3′-UTR of

ZFP91 and that 2 out of 3 predicted seed sequence (position 112-121 and 335-347) are the main binding sites for miR-640 on ZFP91 3’UTR (Figure 5, 6D and supplementary

Figure 4, 5). Moreover miR-640 expression is negatively associated with the expression of ZFP91 (Figure 4): significant downregulation of the level of ZFP91 mRNA and protein were observed after miR-640 overexpression, whereas miR-640 inhibition upregulated the level of ZFP91.

Zinc finger protein 91 (ZFP91), is an atypical E3 ligase activating NF-κB- inducing Kinase (NIK) via Lys63-linked ubiquitination in the non-canonical NF-κB signaling pathway 23. Recent studies have shown ZFP91 expression is upregulated in prostate cancer, it promotes proliferation and tumorigenesis of colon cancer and its upregulated in acute myelogenous leukemia 24-26. Interestingly, ZFP91 presents secondary structure typical of transcription factors; five zinc-finger domains, one leucine- zipper pattern, one coiled-coil structure, and several nuclear localization signals 27. In a recent study ZFP91 was reported to act as a transcription factor upregulating HIF1 through the NFκB/p65 pathway. ZFP91 appears as a major oncogenic protein with the ability to promote the activation of two proangiogenic pathways. ZFP91 stabilizes and

159 activates NIK and HIF-1α 23;24. NIK oncogenic properties have been documented in lung cancer, pancreatic cancer, ovarian cancer and colon cancer 28-30, while HIF-1α promotes angiogenesis responses including: proliferation, angiogenesis, invasion, metastasis, and cell cycle progression 31. However, the role of ZFP91 in angiogenesis is not well defined, even less so, its role in growth factor induced angiogenesis or physiological angiogenesis.

In the present study, we found that mRNA and protein levels of ZFP91 were upregulated by Ang-1 treatment (Figure 6). Interestingly ZFP91 protein levels are also increased by treatment with Ang-2 and VEGF. Moreover, ZFP91 silencing with siRNA abolished Ang-1-induced cell migration. ZFP91 silencing also decreased tube formation and abolished Ang-1-induced tube formation (Figure 7). These results indicate Ang-1 might promote angiogenesis, more specifically cell migration and tube formation by repressing miR-640 and leading to an increase of ZFP91 protein. We also showed that

ZFP91 silencing concomitantly with miR-640 inhibition, abolished the increase in basal migration and tube formation observe with miR-640 inhibition alone. Finally we showed

ZFP91 expression is increase during hindlimb ischemia (supplementary Figure 10).

Interestingly overexpression of ZFP91 was shown to enhance proliferation of colon cancer cells and to promote tumor growth in vivo 23. This is the first time ZFP91 is associated with ECs migration and differentiation, and our study confirms ZFP91 is capable of inducing angiogenic responses outside of a pathological context, but also that it contributes to basal level of ECs differentiation potential. Although no clear relationship has been established between Ang-1 and HIF1, in light of the most recent studies 10;23 we could speculate that ZFP91 effect on Ang-1 induced angiogenesis could be in part mediated by an increase in HIF and VEGF expression. Indeed,

160 overexpression of ZFP91 upregulated the expression of VEGF through upregulation of

HIF1 23. It was also demonstrated that HIF1 knockdown or miR-640 overexpression downregulated VEGF levels, interestingly we showed ZFP91 silencing also leads to

VEGF downregulation as well as Ang-1, NIK and VCAM1 (supplementary Figure 9).

These results are positioning ZFP91 at the center of a feedforward signaling loop where growth factors such as VEGF or Ang-1 induces its expression through downregulation of miR-640 and ZFP91 further increase production of VEGF or Ang-1. ZFP91 effect on

VEGF and Ang-1 mRNA levels could in part be explained by the ability of HIF1 to upregulate VEGF. Moreover, it has been shown that ZFP91 and HIF1 are both upregulated in colon cancer. Thus it is possible to speculate that down-regulation of miR-

640 can lead to upregulation of several targets of the same pathway to modulate pro- angiogenic responses in the setting of cancer. It remains to be clarified whether ZFP91 effect on Ang-1-induced angiogenesis is dependent on its activation of the non-canonical

NFκB pathway through its ubiquitin activity and/or through its transcriptional activation of pro-angiogenic gene targets or both.

In conclusion, we revealed a new role for miR-640 and ZFP91 in mediating Ang-

1 pro-angiogenic responses. Ang-1 transcriptionally regulates miR-640, while the 3′-UTR of ZFP91 serves as its direct target, which subsequently transduces the proangiogenic signals of Ang-1 through a feedforward signaling mechanism. The study sheds a light on how Ang-1 fine tunes its biological responses.

3.6 Acknowledgments

We would like to thank Dr Rivard and his team for providing samples of ischemic hindlimb.

161

3.7 References

1. Suri C, Jone PF, Patan S et al. Requisite role of angiopoietin-1, a ligand for the

TIE-2 receptor, during embryonic angiogenesis. Cell 1996;87:1171-1180.

2. Brindle NP, Saharinen P, Alitalo K. Signaling and functions of angiopoietin-1 in

vascular protection. Circ.Res. 2006;98:1014-1023.

3. Shyu K-G, Manor O, Magner M, Yancopoulos GD, Isner JM. Direct

intramuscular injection of Plasmid DNA encoding angiopoietin-1 but not

angiopoietin-2 augments revascularization in the rabbit ischemic hindlimb.

Circulation 1998;98:2081-2087.

4. Master Z, Jones N, Tran J et al. Dok-R plays a pivotal role in angiopoietin-1-

dependent cell migration through recruitment and activation of Pak. EMBO J.

2001;20:5919-5928.

5. Urbich C, Kuehbacher A, Dimmeler S. Role of microRNAs in vascular diseases,

inflammation, and angiogenesis. Cardiovasc.Res. 2008;79:581-588.

6. Bonauer A, Boon RA, Dimmeler S. Vascular microRNAs. Curr.Drug Targets.

2010;11:943-949.

7. Phatak P, Byrnes KA, Mansour D et al. Overexpression of miR-214-3p in

esophageal squamous cancer cells enhances sensitivity to cisplatin by targeting

survivin directly and indirectly through CUG-BP1. Oncogene 2016;35:2087-

2097.

162

8. Suarez Y, Fernandez-Hernando C, Yu J et al. Dicer-dependent endothelial

microRNAs are necessary for postnatal angiogenesis. Proc.Natl.Acad.Sci.U.S.A

2008;105:14082-14087.

9. Echavarria R, Mayaki D, Neel JC et al. Angiopoietin-1 inhibits toll-like receptor 4

signalling in cultured endothelial cells: role of miR-146b-5p. Cardiovasc.Res.

2015;106:465-477.

10. Zhou Y, Li XH, Zhang CC et al. Hydrogen sulfide promotes angiogenesis by

downregulating miR-640 via the VEGFR2/mTOR pathway. Am.J.Physiol Cell

Physiol 2016;310:C305-C317.

11. Voigtlander T, Gupta SK, Thum S et al. MicroRNAs in Serum and Bile of

Patients with Primary Sclerosing Cholangitis and/or Cholangiocarcinoma.

PLoS.ONE. 2015;10:e0139305.

12. Li X, Lu Y, Chen Y, Lu W, Xie X. MicroRNA profile of paclitaxel-resistant

serous ovarian carcinoma based on formalin-fixed paraffin-embedded samples.

BMC.Cancer 2013;13:216.

13. Visone R, Rassenti LZ, Veronese A et al. Karyotype-specific microRNA

signature in chronic lymphocytic leukemia. Blood 2009;114:3872-3879.

14. Supinski GS, Wang L, Song XH, Moylan JS, Callahan LA. Muscle-specific

calpastatin overexpression prevents diaphragm weakness in cecal ligation

puncture-induced sepsis. J Appl.Physiol (1985.) 2014;117:921-929.

163

15. Lin P, Buxton JA, Acheson A et al. Antiangiogenic gene therapy targeting the

endothelium-specific receptor tyrosine kinase Tie2. Proc.Natl.Acad.Sci.

1998;95:8829-8834.

16. Echavarria R, Hussain SN. Regulation of angiopoietin-1/Tie-2 receptor signaling

in endothelial cells by dual-specificity phosphatases 1, 4, and 5. J Am Heart

Assoc. 2013;2:e000571.

17. Abdel-Malak N, Harfouche R, Hussain SN. Transcriptome of angiopoietin 1-

activated human umbilical vein endothelial cells. Endothelium 2007;14:285-302.

18. Abdel-Malak NA, Srikant CB, Kristof AS et al. Angiopoietin-1 promotes

endothelial proliferation and migration through AP-1 dependent autocrine

production of interleukin-8. Blood 2008;111:4145-4154.

19. Orom UA, Lund AH. Isolation of microRNA targets using biotinylated synthetic

microRNAs. Methods 2007;43:162-165.

20. Malinda KM. In vivo matrigel migration and angiogenesis assay. Methods

Mol.Biol. 2009;467:287-294.

21. van RE, Kauppinen S. Development of microRNA therapeutics is coming of age.

EMBO Mol.Med. 2014;6:851-864.

22. Skovseth DK, Kuchler AM, Haraldsen G. The HUVEC/Matrigel assay: an in vivo

assay of human angiogenesis suitable for drug validation. Methods Mol.Biol.

2007;360:253-268.

164

23. Jin X, Jin HR, Jung HS et al. An atypical E3 ligase zinc finger protein 91

stabilizes and activates NF-kappaB-inducing kinase via Lys63-linked

ubiquitination. J.Biol.Chem. 2010;285:30539-30547.

24. Ma J, Mi C, Wang KS, Lee JJ, Jin X. Zinc finger protein 91 (ZFP91) activates

HIF-1alpha via NF-kappaB/p65 to promote proliferation and tumorigenesis of

colon cancer. Oncotarget. 2016

25. Paschke L, Rucinski M, Ziolkowska A et al. ZFP91-a newly described gene

potentially involved in prostate pathology. Pathol.Oncol.Res. 2014;20:453-459.

26. Unoki M, Okutsu J, Nakamura Y. Identification of a novel human gene, ZFP91,

involved in acute myelogenous leukemia. Int.J.Oncol. 2003;22:1217-1223.

27. Saotome Y, Winter CG, Hirsh D. A widely expressed novel C2H2 zinc-finger

protein with multiple consensus phosphorylation sites is conserved in mouse and

man. Gene 1995;152:233-238.

28. Qu LL, He L, Zhao X, Xu W. Downregulation of miR-518a-3p activates the NIK-

dependent NF-kappaB pathway in colorectal cancer. Int.J.Mol.Med.

2015;35:1266-1272.

29. Thu YM, Richmond A. NF-kappaB inducing kinase: a key regulator in the

immune system and in cancer. Cytokine Growth Factor Rev. 2010;21:213-226.

165

30. Uno M, Saitoh Y, Mochida K et al. NF-kappaB inducing kinase, a central

signaling component of the non-canonical pathway of NF-kappaB, contributes to

ovarian cancer progression. PLoS.ONE. 2014;9:e88347.

31. Krock BL, Skuli N, Simon MC. Hypoxia-induced angiogenesis: good and evil.

Genes Cancer 2011;2:1117-1133.

166

3.8 Figures

Figure 3.1: Ang-1 regulation of miR-640 in HUVECs

167

Figure 3.2: Effect of mature miR-640 overexpression on Ang-1-induced angiogenesis

168

Figure 3.3: Effect of miR-640 inhibition on Ang-1-induced angiogenesis

169

Figure 3.4: Identification of miR-640 targets

170

Figure 3.5 ZFP91 is a direct target of miR-640

171

Figure 3.6: Ang-1 regulation of ZFP91

172

Figure 3.7: Effect of ZFP91 inhibition on Ang-1-induced angiogenesis

173

Supplementary Figure 3.1: Abundance of miR-640 in HUVECs

174

Supplementary Figure 3.2: Angiogenesis factor regulation of miR-211-3p

175

Supplementary Figure 3.3: Transfection efficiency of miR-640 in HUVECs

176

Supplementary Figure 3.4: Effect of miR-640 on the 3rd predicted seed in ZFP91 3’UTR

177

Supplementary Figure 3.5: Effect of inhibition of miR-640 on Luciferase expression

178

Supplementary Figure 3.6: Verification of ZFP91 knock down with siRNA

179

Supplementary Figure 3.7: Effect of ZFP91 silencing on Ang-1-induced cell survival

180

Supplementary Figure 3.8: Effect of ZFP91 silencing on Ang-1-induced inhibition of apoptosis

181

Supplementary Figure 3.9: Effect of ZFP91 silencing pro-angiogenic gene targets

182

Supplementary Figure 3.10: Expression of ZFP91 in a model of hind limb ischemia

183

3.9 Figure legends

Figure 3.1: Ang-1 regulation of miR-640 in HUVECs

A: Mature miR-640 expression measured in HUVECs pre-incubated for 24, 48 or 72 h with PBS or Ang-1 (300ng/ml).

B: pri-miR-640 and mature miR-640 expression measured in HUVECs pre-incubated for

2, 4, 6 and 12 h with PBS or Ang-1 (300ng/ml).

C: Mature miR-640 expression measured in exosomes secreted by HUVECs pre- incubated for 24 h with PBS or Ang-1 (300ng/ml).

D: Mature miR-640 expression measured in HUVECs transfected with Ad-GFP or Ad-

ExTEK and then incubated for 24 h with PBS or Ang-1 (300ng/ml).

E: GATAD2A mRNA expression measured in HUVECs pre-incubated for 2, 4, 6 and 12 h with PBS or Ang-1 (300ng/ml).

F: Mature miR-640 expression measured in HUVECs pre-incubated for 24 h with PBS or

Ang-2 (300ng/ml), FGF-2 (10ug/ml), VEGF (40ng/ml) and TGF- (2ng/ml). Values are means± SEM and are expressed as fold change from values measured in cells pre- incubated with PBS. *P < 0.05, compared with pre-incubation with PBS, n = 6.

Figure 3.2: Effect of mature miR-640 overexpression on Ang-1-induced angiogenesis

A: Cell number of control mimic- and miR-640 mimic-transfected HUVECs seeded 24 h post-transfection in 12-well plates (8x104/cm2) then incubated in full (20% FBS), basal

(2% FBS), or basal medium containing Ang-1 (300 ng/ml) for 24 h. N=12 per group.

184

Values are means  SEM. *p<0.05, compared to full medium. #p<0.05, compared to basal medium.

B: Optical densities (OD) of cleaved Caspase-3 in control mimic- and miR-640 mimic- transfected HUVECs seeded 24 h post-transfection in full (20% FBS), basal (2% FBS), or basal medium containing Ang-1 (300 ng/ml) for 24 h. N=4 per group. Values are means  SEM *p<0.05, compared to full medium. #p<0.05, compared to basal medium.

C: Percent wound healing in control mimic- and miR-640 mimic-transfected HUVECs wounded (t=0 h) then maintained in basal medium containing PBS or Ang-1 (300 ng/ml) for 8 h. N=12 per group. Values are means ± SEM. *p<0.05, compared to PBS alone.

D: Total tube numbers of control mimic- and miR-640 mimic-transfected HUVECs maintained 24 h post-transfection in growth factor-reduced Matrigel-pre-coated plates with basal medium containing PBS or Ang-1 (300 ng/ml) for 24 h. N=12 per group.

Values are means  SEM. *P<0.05 compared to PBS alone.

Figure 3.3: Effect of miR-640 inhibition on Ang-1-induced angiogenesis

A: Cell number of LNA-scrambled- and LNA-anti-miR-640-transfected HUVECs seeded

24 h post-transfection in 12-well plates (8x104/cm2) then incubated in full (20% FBS), basal (2% FBS), or basal medium containing Ang-1 (300 ng/ml) for 24 h. N=12 per group. Values are means  SEM. *p<0.05, compared to full medium. #p<0.05, compared to basal medium.

B: Optical densities (OD) of cleaved Caspase-3 in LNA-scrambled- and LNA-anti-miR-

640-transfected HUVECs seeded 24 h post-transfection in full (20% FBS), basal (2%

FBS), or basal medium containing Ang-1 (300 ng/ml) for 24 h. N=4 per group. Values

185 are means  SEM *p<0.05, compared to full medium. #p<0.05, compared to basal medium.

C: Percent wound healing in LNA-scrambled- and LNA-anti-miR-640-transfected

HUVECs wounded (t=0 h) then maintained in basal medium containing PBS or Ang-1

(300 ng/ml) for 8 h. N=12 per group. Values are means ± SEM. *p<0.05, compared to

PBS alone.

D: Total tube numbers of LNA-scrambled- and LNA-anti-miR-640-transfected HUVECs maintained 24 h post-transfection in growth factor-reduced Matrigel-pre-coated plates with basal medium containing PBS or Ang-1 (300 ng/ml) for 24 h. N=12 per group.

Values are means  SEM. *P<0.05 compared to PBS alone.

Figure 3.4: Identification of miR-640 targets

A: Venn diagram displaying computationally predicted targets of hsa-miR-640 by

TargetScan, miRanda, and miRsearch3.0. Target mRNAs predicted by the three algorithms are listed.

B: ZFP91 mRNA expression in miR-640 mimic- or LNA-anti-miR-640-transfected

HUVECs. N=6 per group. Values are means  SEM. *P<0.05 compared to control.

C-D: Representative immunoblots and optical densities (OD) of ZFP91 in control mimic- or miR-640 mimic-transfected HUVECs 48 h post-transfection. N=6 per group. Values are means  SEM and are expressed as fold change from control mimic. *p<0.05, compared to controls.

D-E: Representative immunoblots and optical densities (OD) of ZFP91 in LNA- scrambled- or LNA-anti-miR-640-transfected HUVECs 48 h post-transfection. N=6 per

186 group. Values are means  SEM and are expressed as fold change from control mimic.

*p<0.05, compared to controls.

Figure 3.5: ZFP91 is a direct target of miR-640

A: Sequence alignment of miR-640 and its target sites in 3′UTRs of ZFP91.

B: Relative luciferase activity in HUVECs transfected with control mimic or miR-640 mimic and luciferase reporter plasmid expressing WT or mutated 3’UTRs of ZFP91.

N=12. Values are means ± SEM expressed as RLU *p < 0.05 as compared to control mimic, # p < 0.05 as compared to empty vector.

C: Levels of miR-640 48 hours following transfection of 50 nM biotin-control mimic or biotin-miR-640 mimic as measured by q-PCR analysis

D: Levels of U6 RNA 48 hours following transfection of 50 nM biotin-control mimic or biotin- miR-640 mimic as measured by q-PCR analysis

E: Levels of ZFP91 mRNA in the materials pulled down by biotin-miR-640

F: Levels of total input mRNAs measured by q-PCR. The enrichment of miR was calculated as follows: miR pull-down/control pull-down (X), miR input/control input (Y),

Fold binding = X/Y. N=3 Values are mean ±SEM expressed as fold change from biotin- control mimic. *p < 0.05

Figure 3.6: Ang-1 regulation of ZFP91

A: ZFP91 mRNA expression measured in HUVECs transfected with luciferase reporter plasmid expressing WT 3’UTR of ZFP91 followed by incubation for 12 and 24 h with

PBS or Ang-1 (300ng/ml).

B, C: Representative immunoblots and optical densities (OD) of ZFP91 in HUVECs pre- incubated for 12 and 24 h with PBS or Ang-1 (300ng/ml).

187

D: Relative Luciferase activity in HUVECs transfected with luciferase reporter plasmid expressing wild type or mutated 3’UTRs of ZFP91 then incubated for 24 h with PBS or

Ang-1 (300ng/ml)

E, F: Representative immunoblots and optical densities (OD) of ZFP91 in HUVECs pre- incubated for 24 h with PBS or Ang-2 (300ng/ml) or FGF-2 10ng/ml) or VEGF

(40ng/ml). N=6 per group. Values are means  SEM and are expressed as fold change from PBS. *p<0.05, compared to controls.

Figure 3.7: Effect of ZFP91 inhibition on Ang-1-induced angiogenesis

A: Percent wound healing in scrambled-siRNA- and ZFP91 siRNA-transfected HUVECs wounded (t=0 h) then maintained in basal medium containing PBS or Ang-1 (300 ng/ml) for 8 h. N=12 per group. Values are means ± SEM. *p<0.05, compared to PBS alone.

B: Percent wound healing in scrambled-siRNA- and LNA-scrambled-co-transfected or

ZFP91 siRNA- and LNA-anti-miR-640-co-transfected HUVECs wounded (t=0 h) then maintained in basal medium containing PBS or Ang-1 (300 ng/ml) for 8 h. N=12 per group. Values are means ± SEM. *p<0.05, compared to PBS alone.

C: Total tube numbers in scrambled-siRNA- and ZFP91 siRNA-transfected HUVECs maintained 24 h post-transfection in growth factor-reduced Matrigel-pre-coated plates with basal medium containing PBS or Ang-1 (300 ng/ml) for 24 h. N=12 per group.

Values are means  SEM. *P<0.05 compared to PBS alone. # p<0.05 compared to scrambled-siRNA.

D: Total tube numbers in scrambled-siRNA- and LNA-scrambled-co-transfected or

ZFP91 siRNA- and LNA-anti-miR-640-co-transfected HUVECs maintained 24 h post-

188 transfection in growth factor-reduced Matrigel-pre-coated plates with basal medium containing PBS or Ang-1 (300 ng/ml) for 24 h. N=12 per group. Values are means 

SEM. *p<0.05 compared to PBS alone. # p<0.05 compared to scrambled-siRNA.

Supplementary Figure 3.1: Abundance of miR-640 in HUVECs. miR-640 and miR-126-star copies per L. HUVECs were lysed and tRNA extracted using Qiagen miRNA extraction kit. RT was done using AB taqman miRNA RT kit specific miR640 and miR-126-star. ddQPCR was done using quanta3D with 1L of cDNA. N=3. Mean ±SEM.

Supplementary Figure 3.2: Angiogenesis factor regulation of miR-211-3p

Mature miR-211-3p expression measured in HUVECs pre-incubated for 24 h with PBS or Ang-2 (300ng/ml), FGF-2 (10ug/ml), VEGF (40ng/ml) and TGF- (2ng/ml). Values are means± SEM and are expressed as fold change from values measured in cells pre- incubated with PBS. *P < 0.05, compared with pre-incubation with PBS, n = 6.

Supplementary Figure 3.3: Transfection efficiency of miR-640 in HUVECs miR-640 levels in Control mimic- or miR640 mimic-transfected HUVECs 48 h post transfection. N=6. Mean ±SEM, *p<0.05 compared to control mimic.

Supplementary Figure 3.4: Effect of miR-640 on the 3rd predicted seed in ZFP91

3’UTR

Relative luciferase activity in HUVECs transfected with control mimic or miR-640 mimic and luciferase reporter plasmid expressing WT 3’UTR of ZFP91 containing a third predicted seed sequence.

Supplementary Figure 3.5: Effect of inhibition of miR-640 on Luciferase expression

189

Relative luciferase activity in HUVECs transfected with LNA-scrambled or LNA-anti- miR-640 mimic and luciferase reporter plasmid expressing WT 3’UTRs of ZFP91.

Values are means ± SEM expressed as RLU. N=12

Supplementary Figure 3.6: Verification of ZFP91 knock down with siRNA

ZFP91 mRNA expression in scrambled siRNA- or ZFP91 siRNA-transfected HUVECs,

48 h post-transfection.

Supplementary Figure 3.7: Effect of ZFP91 silencing on Ang-1-induced cell survival

Cell number of scrambled siRNA- and ZFP91 siRNA-transfected HUVECs seeded 24 h post-transfection in 12-well plates (8x104/cm2) then incubated in full (20% FBS), basal

(2% FBS), or basal medium containing Ang-1 (300 ng/ml) for 24 h. N=12 per group.

Values are means  SEM. *p<0.05, compared to full medium. #p<0.05, compared to basal medium.

Supplementary Figure 3.8: Effect of ZFP91 silencing on Ang-1-induced inhibition of apoptosis. Optical densities (OD) of cleaved Caspase-3 in LNA-scrambled- and LNA- anti-miR-640-transfected HUVECs seeded 24 h post-transfection in full (20% FBS), basal (2% FBS), or basal medium containing Ang-1 (300 ng/ml) for 24 h. N=4 per group.

Values are means  SEM *p<0.05, compared to full medium. #p<0.05, compared to basal medium.

Supplementary Figure 3.9: Effect of ZFP91 silencing pro-angiogenic gene targets.

Ang-1, VEGF, CCL19, NIK, VCAM1 and BAFF mRNA expression measured in

HUVECs transfected with scrambled-siRNA or ZFP91 siRNA. N=8. *p<0.05, compared to controls.

190

Supplementary Figure 3.10: Expression of ZFP91 in a model of hind limb ischemia

ZFP91 mRNA levels measured by q-PCR in ischemic and non-ischemic adductor muscle of C57BL6 mice. N=4 per group. Values are means  SEM and are expressed as fold change from non-ischemic. *p<0.05

191

3.10 Tables

Table 3.1: Primers used for quantitative real-time PCR experiments Type Sequence (5’ 3’) Accession Expected Size Number (bp)

-actin F: AGAAAATCTGGCACCACACC NM_001101 126 R: GGGGTGTTGAAGGTCTCAAA

GAPDH F: AAGAAGGTGGTGAAGCAGGCG NM_023964.1 166 R: ACCAGGAAATGAGCTTGACAA

ZFP91 F: CTCGCTATTTGCAGCACCAC NM_053023.4 165 R: GCCCGAGCACAATATTCACA

NIK/MAP3K14 F: ATCAGCTGTGAATCGTCAGG NM_003954.1 212 R: CCTGGAAACATTTGTCATCACC CCL19 F: GAGTCCGAGTCAAGCATTGT NM_006274 198 R: TGGTAGCATTGCAATCTGGG BAFF F: GGTCCAGAAGAAACAGGATCT NM_006573 213 R: TTCATCCCCAAAGACATGGAC VEGF F: CTACCTCCACCATGCCAAGT NM_001025366.1 281 R: CACACAGGATGGCTTGAAGA hsa-miR-640 GATCCAGGAACCTGCCTCT MIMAT0003310

U6 ACTAAAATTGGAACGATACAGAGA NR_004394.1

5S ATACCGGGTGCTGTAGGCTT D14867

pre-miR-640 F: TGGGCAAGTTCCTGAAGATCAGA MIMAT0003310 R: AGAGGCAGGTTCCTGGATCAT

192

Preface to chapter 4

Ang-1-induced angiogenic responses are orchestrated through the activation of proangiogenic genes or inhibition antiangiogenic gene transcription. It is well established that Ang-1 promotes survival, migration and differentiation of ECs through the activation of MAPK signaling pathways ERK1/2, p38 and SAPK/JNK as well as through the activation of PI3K/AKT pathway [6;7;12;18;109;118;197]. MAPKs and PI3K/AKT pathways activation lead to gene transcription. Previous studies carried out in our lab on gene expression profiling of ECs exposed to Ang-1 revealed 58 genes were upregulates by Ang-1 treatment [383]. In silico analysis of Ang-1 transcriptome predicted the ETS family of transcription factors would play an important role in the regulation of Ang-1 transcriptome. More specifically, ETS1, ELK1 and ETV4 were predicted to regulate

Ang-1-induced genes. The ETS factors are important markers of the endothelium and are well known for their role in vascular development and angiogenesis and are regulated by the MAPKs pathways[384;385]. Therefore we hypothesized that ETS1, ELK1 and ETV4 could be induced by Ang-1 downstream of the MAPKs to induce transcriptional regulation of proangiogenic genes affecting Ang-1-dependent survival, migration and differentiation in ECs.

193

Chapter 4: ETS1, ELK1 and ETV4 Contribute to Angipoietin-1 Angiogenic Responses in Endothelial Cells

194

ETS1, ELK1, and ETV4 Contribute to Angiopoietin-1 Signaling and Angiogenic Responses in Endothelial Cells Sharon Harel1, Veronica Sanchez1, Javier E Sanchez Galan2, Hodan Ismail1, Dominique

Mayaki1, Laurent Huck1, Mathieu Blanchette2 and Sabah NA Hussain1

1 Meakins-Christie Laboratories and Translational Research in Respiratory Diseases Program, Research Institute of the McGill University Health Centre; Department of Critical Care Medicine, McGill University Health Centre, Montréal, Québec, Canada.

2 School of Computer Science, McGill University, McConnell Engineering Bldg., Rm. 318, 3480 Rue University, Montréal, Québec H3A 0E9 Canada ; McGill Centre for Bioinformatics, McGill University, Montréal, Québec Canada

195

4.1 Abstract

Objectives: Angiopoietin-1 (Ang-1) is the main ligand for endothelial cell-specific tyrosine kinase (Tie-2) receptors. Ang-1 is known to promote migration and differentiation while also inhibiting apoptosis and vascular leakage. Few transcription factors (TFs) have been identified for their role in transducing the biological effects of

Ang-1. Our general objective is to identify the TFs and corresponding targets involved in the transduction of Ang-1-induced events such as cell survival, migration and differentiation.

Methods and Results: In an effort to identify new TFs that mediate cellular signaling downstream of Ang-1, we used an in silico tool to detect enrichment in transcription factor binding sites (TFBS) for 86 genes, previously found to be regulated upon exposure to Ang-1. The in silico analysis revealed enrichment in TFBS for the ETS family (ETS1, ELK1 and ETV4). Expression of ETS1, ELK1 and ETV4 was evaluated with real-time PCR, whereas DNA binding activity was monitored with electrophoretic mobility shift assays, luciferase assay and nuclear localization by immunofluorescence.

Cell migration was measured with wound healing assays, whereas cell survival and differentiation of cells into capillary-like tube structures were monitored with cell counting and Matrigel assay. To selectively inhibit ETS1, ELK1 and ETV4 expression, we used siRNA oligonucleotides. ETS1 mRNA expression increased within 30 min of

Ang-1 exposure and declined after 3 h. Ang-1 exposure resulted in an increase in nuclear mobilization and augmented DNA binding activity for ETS1, ELK1 and ETV4.

196

Knockdown of ETS TFs expression completely abrogated Ang-1-induced endothelial survival, tube formation and capillary-like tube formation of HUVECs.

Conclusion : Ang-1 triggers a significant induction of ETS1, ELK1 and ETV4 DNA binding activity which in turn contributes to Ang-1-induced endothelial cell survival, migration and capillary-like tube formation through regulation of pro-angiogenic genes..

Keywords: Angiopoietin-1, Transcription Factors, Angiogenesis, Endothelial cells

197

4.2 Introduction

The receptor tyrosine kinase (Tie-2) and its associated ligands, the angiopoietins, have emerged as important regulators of angiogenesis, both in adults and in embryos. In adult mice, Ang-1 stimulates in vivo vascular remodeling, vascular enlargement, wound healing, and lymphangiogenesis 1. In addition, Ang-1 inhibits endothelial cell (EC) apoptosis and stimulates migration, proliferation, and differentiation of these cells 1.

Despite the importance of the Ang-1/Tie-2 receptor pathway to vascular homeostasis and angiogenesis, little progress has been made toward the identification of new transcription factors that are activated downstream from it. It has been shown that Ang-1 inhibits the transcriptional activity of FoxO-1 downstream of the PI-3 kinase/AKT pathway, thereby reducing the expression of several proapoptotic proteins 2. KLF2, a transcription factor involved in vascular quiescence is activated by Ang-1 through PI-3K/AKT pathway and might be involved in Ang-1 anti-inflammatory effect 3. Elk-1 is activated by Ang-1 downstream from the Erk1/2 pathway, and forms a complex by binding serum response factor and serum response element in various promoters 2. Our own group showed that

Ang-1 triggers significant and transient induction of Egr-1, and Egr-1 contributes to Ang-

1-induced endothelial cell migration and proliferation 4. We also showed that Ang-1- induced production of IL-8 depends on AP-1 DNA binding activity 5. Ang-1 also induces induced expression of NERF2, an ETS factor, in both quiescent and hypoxic cells 6.

However, most of these studies don’t address the importance of these TFs in mediating the biological functions of the Ang-1/Tie-2 pathway.

198

The ETS family of TFs has been identified as a significant contributor to a variety of biological functions, including cell differentiation, cell cycle, cell migration, inflammation, proliferation, apoptosis and angiogenesis 7. ETS factors are induced by pro-angiogenic factors such as vascular endothelial growth factor (VEGF) and fibroblast growth (FGF-1) 8, 9. On activation they bind promoter regions of several proangiogenesis genes and endothelial cell markers including Angiopoietin-2, Tie-1/Tie-2,

Flt1, early growth response 1 (EGR1), several matrix metallopeptidase (MMPs), von

Willebrand Factor (VWF) 10-16. The aim of this study is to identify ETS factors induced downstream from the Ang-1/Tie-2 pathway and to define their contribution to the regulation of Ang-1-induced angiogenic responses.

4.3 Material and Methods

Identification of Transcription factor putatively regulating Ang-1 transcriptome

Briefly, a set of 86 possibly co-regulated genes was obtained from a set of microarray experiments (differential expression) 17. A computer program called GATOR (Sanchez

Galan et al., in prep) then identifies likely common regulators of these genes, together with the location of their putative binding sites, via a combination of a comparative genomics approach and a statistical over-representation analysis. Briefly, the candidate regulatory regions of a given gene are defined as the non-coding conserved regions

(NCS; obtained by running the PhastCons program 18 on a 28-way multiple alignment of vertebrate genomes 19 located within a fixed distance (typically 10kb, possibly more) away from the transcription start site of the gene. The set of NCS from the corresponding set of co-regulated genes is then scanned for putative transcription factor binding sites for more than 400 TFs, with position-weight matrices in the Transfac 20 and Jaspar 21

199 databases. TFs whose sites are found unexpectedly often in the set of NCS (based on a new statistical test taking into consideration genome-wide binding site frequency, distribution, and GC content biases) are reported as putatively involved in the regulation of the gene set. Genes are then clustered based on similarity in TFBS content (Figure

1A), resulting in the prediction of a small number of transcription factors specifically associated to each gene. Related approaches have been previously shown by us 22, 23 and others 24-27 to be very effective at narrowing down the list of candidate transcription factors involved in the process under study.

Material

ETS antibody was purchased from Novus Biologicals. ELK1 and ETV4 antibody were from Santa Cruz. Recombinant Angiopoietin-1 was purchased from R&D.

Cell Culture

HUVECs were purchased from Glycotech and used between passage 4 to 7. HUVECs were grown in MCDB131 medium supplemented with 20% fetal bovine serum (FBS), endothelial cell growth supplement, 2 mmol/L glutamine, heparin, and gentamicin.

HUVECs were starved 6 h in basal media before receiving Ang-1 treatment at a concentration of 300ng/ml.

Real‐Time PCR

Total RNA was extracted using an Ambion PureLink RNA mini kit (Life Technologies,

Inc, Burlington, Ontario, Canada). Real‐time PCR amplifications using specific primers were performed using SYBR green and a 7500 Real‐Time PCR System (Applied

Biosystems, Foster City, CA) (Supplementary Table 2). All experiments were performed

200 in triplicate and relative mRNA expression was analyzed using the ΔCT method and expressed as 2−ΔΔCT.

Electromobility shift assay

Electrophoretic mobility shift assays for ETS1 and ETV4 were performed with a Gelshift

ETS1/ETV4 kit (Panomics, Fremont, CA) using double-stranded DNA probes for human

ETS1 and ETV4 and 20 μg of nuclear extracts, according to the manufacturer’s instructions.

Luciferase Assay

ETS1 and ELK1 luciferase reporter were kindly provided by Dr Jiwaji (University of

Glasgow, United Kingdom). DNA encoding a multiple cloning cassette and thymidine kinase promoter (PTK) was inserted between Kpn I and Hind III upstream of the firefly luciferase gene (Fluc) in pGL3 Basic (Promega) generating pMN2. DNA encoding TFBS sequences were inserted upstream of PTK. Fluc was then replaced with a unique DNA reporter sequence (UR) such that each TFBS was attached to a different UR 28. Briefly,

HUVECs were pelleted and resuspended in 100 μL Optimem at a density of 1x107 cells/ml and mixed with 1 μg of pRL-SV40 (Promega) and 1 μg of each of the TFBS encoding vectors (see Supplementary Table 1). Cells were electroporated using the appropriate nucleofection program (Amaxa). Nucleofected cells were incubated at 37°C for 24 h before Ang-1 treatment. HUVECs were treated with PBS or Ang-1 for 6 h, lysed, and mRNA was extracted as described above. UR and Rluc analysis was conducted as described above. A 10-fold dilution series of UR or Rluc linear dsDNA was created and a standard curve was generated. Unknown samples were compared to the standard curve and the copy number was calculated. Transfection efficiency was

201 accounted for by normalizing the UR copy numbers to that of Rluc in each sample.

Changes in gene expression were quantified by comparing the log2 ratio for treated cells to untreated cells.

Nuclear staining by immunofluorescence

HUVECs were grown in NUNC LabTek chambered slide coated with fibronectin. At confluence, cells were starved 6 h in basal media and treated with PBS or Ang-1

(300ng/ml) for 1 h, 3 h or 6 h, then fixed for 10 minutes in 4% paraformaldehyde, and permeabilized for 10 minutes in 0.5% Triton X-100 in PBS. Cells were then incubated with primary antibodies against Ets1 (1:100), Elk1 (1:800) or Etv4 (1:800) overnight at

4ºC. Cells were washed and treated with Rhodamin-conjugated goat anti-rabbit secondary antibody (Molecular Probes, Eugene, OR) for 1 h at RT and then incubated with 4′,6′- diamidino-2-phenylindole (DAPI) for 5 minutes at RT. Images were obtained using a confocal microscope.

Transfection with siRNA Oligos

HUVECs were transfected with siRNA (siGENOME SMARTpool) directed against either ETS1, ELK1, ETV4, or a nontargeting siRNA pool (siCONTROL; Dharmacon,

Lafayette, CO), using Lipofectamine RNAiMAX reagent (Life Technologies, Inc,

Burlington, Ontario, Canada). All experiments were performed 48 hours post- transfection. Knockdown was verified with real-time polymerase chain reaction and immunoblotting.

Cell Count

Forty-eight hours post-infection, siRNA-transfected HUVECs were seeded in 12-well plates at a density of 8x104 cells/cm2. Cells were maintained for 24 h in full medium

202

(20% FBS), basal medium (2% FBS), or basal medium containing Ang-1 (300 ng/ml).

Cells were counted using a hemocytometer.

Wound Healing Assay

HUVEC monolayers were wounded with a 200 μl pipette tip and maintained for 8 h in basal medium (2% FBS) or basal medium containing Ang-1 (300 ng/ml). Wounded areas were visualized using an Olympus inverted microscope and quantified using Image-Pro

Plus™ software (Media Cybernetics, Bethesda, MD). Values are reported as % wound healing equals [1 – (wound area at t8/wound area at t0)] x 100, where t8 is the time (8 h) over which cells were maintained in media and t0 is the time immediately after wounding, as previously described 29.

Capillary Tube Formation

HUVECs were seeded in 96-well plates pre-coated with growth factor-reduced

Matrigel at a density of 1x104 cells per well. Cells were maintained for 24 h in basal

(2% FBS) or basal medium containing Ang-1 (300 ng/ml). Whole-well images were captured using an Olympus inverted microscope (40X) and analyzed using Image-Pro

Plus™ software. Angiogenic tube formation was determined by counting branching points of formed tubes and total tube length in each field, as previously described 29

Data Analysis

Data are expressed as means ± SEM. Differences between experimental groups were determined using a Two-Way Analysis of Variance followed by a Student–Newman–

Keuls post-hoc test. P-values < 0.05 were considered statistically significant.

203

4.4 Results

ETS family of transcription factor regulates Ang-1 transcriptome

Using GATOR we were able to identify ETS1, ELK1 and ETV4 as transcription factors potentially involved in Ang-1-induced gene regulation. TFBS enrichment for ETS1,

ELK1 and ETV4 generated significantly high Z-scores for the Ang-1 up-regulated gene set (Table 1). To identify the genes that generated those high Z-scores, gene-by-gene scoring analysis was performed. The analysis resulted in a two-dimensional matrix of Z- scores with 62 columns (PWM) and 58 rows for the Ang-1 up-regulated gene set. Using the heatmap 2 function of the gplots package of R. Associations in the heatmap have a biological meaning: a cluster reflects how groups of genes are regulated by a TF or a family of TFs. We observed several clusters corresponding to the ETSs factors (Figure

1).

In order to validate the relationship between Ang-1 and the ETSs, we verified the mRNA levels of ETS1, ELK1 and ETV4 in HUVECs. We observed that ETS1 mRNA levels were lower compared to the mRNA levels of ELK1 and ETV4 (Figure 2A). To establish a direct link between Ang-1 and ETS1, ELK1 or ETV4, a time course of the expression of these ETSs was assessed by treating HUVECs with Ang-1 (300 ng/mL) for 30 minutes, 1, 3, and 6 hours. Ang-1 elicits a significant induction of ETS1 mRNA in a time-dependent fashion at 30 minutes, 1 h and 3 h (Figure 2B). Ang-1 doesn’t elicit an increase in ELK1 or ETV4 mRNA, suggesting that they might be regulated post- translationally. In addition, Ang-1 significantly increases ETS1 and ETV4 DNA binding activity, as detected by electrophoretic mobility shift assays (Figure 2 C-D). We also

204 evaluated ETS1 and ELK1 DNA binding activity with Luciferase reporter and observed that Ang-1 significantly increases ETS1 and ELK1 transcriptional activity (Figure 2 E-F)

Mobilization of ETS proteins to the nucleus was evaluated by immunofluorescence in

HUVECs treated with PBS or Ang-1 for 1, 3 or 6 h. Ang-1 enhances nuclear accumulation of ETS1, ELK1 and ETV4 protein at 1 h (Figure 3-5). We observed that

ETS1 was strictly expressed in the nucleus and its expression was induced by Ang-1 treatment (Figure 3). ELK1 and ETV4 appeared to be present in both the cytoplasm and the nucleus, but Ang-1 enhanced nuclear localization of ELK1 and ETV4 at 1 h (Figure

4-5). These results suggests that Ang-1 regulates ETS1 at the transcriptional level; it is possible that Ang-1 regulates ELK1 and ETV4 activity post-translationally and enhances the DNA binding activity of ETS1, ELK1 and ETV4.

The involvement of the PI-3 kinase, Erk1/2, p38, SAPK/JNK, and mTOR pathways in

Ang-1-induced ETS1, ELK1 and ETV4 expression was evaluated by using selective inhibitors of these pathways. When HUVECs are incubated with inhibitors alone, no significant induction or repression of ELK1 or ETV4 mRNA is observed (Figure 6B-C).

However, inhibition of the p38, SAPK/JNK and PI-3 kinase/AKT pathways results in significant induction of ETS1 mRNA (Figure 6A). Moreover, inhibition of the p38,

SAPK/JNK and PI-3 kinase/AKT pathways results in a significant reduction of Ang-1- induced ETS1 mRNA expression (Figure 6A). Inhibition of p38 MAPKs by BIRB796 also results in augmentation of ELK1 mRNA expression in Ang-1-treated cells. These results suggest that the p38, SAPK/JNK and PI-3 kinase/AKT pathways promote Ang-1-

205 induced ETS1 expression, whereas ELK1 and ETV4 do not appear to be significantly regulated at the transcriptional level.

To evaluate the role of ETS1, ELK1 and ETV4 in mediating biological responses to Ang-

1 in ECs, we used siRNA oligos to attenuate their expression. We used cell count, scratch wound healing and tube formation to evaluate the importance of ETS1, ELK1 and ETV4 in Ang-1-induced cell survival, migration and differentiation. In scrambled siRNA- transfected cells, Ang-1 significantly increased cell count number, wound healing and tube formation (Figure 7 A-C). In contrast, Ang-1 fails to promote cell survival in cells that are transfected with ETS1, ELK1 or ETV4 siRNA (Figure 7A). Similarly, Ang-1 fails to increase cell migration in HUVECs transfected with ETS1 and ETV4 siRNA oligos; however, ELK1 silencing has no effect on Ang-1 induced cell migration (Figure

7B). Finally, Ang-1 fails to increase tube numbers in cells transfected with ETS1, ELK1 or ETV4 siRNA. Moreover, ELK1 and ETV4 silencing also leads to a significant decrease in basal tube formation. Taken together these results indicate that ETS1, ELK1 and ETV4 play a role in mediating various angiogenic responses initiated by Ang-1.

Finally, we evaluated whether ETS1, ELK1 or ETV4 regulate the expression of proangiogenesis genes in HUVECs. First we selected a subset of genes that were upregulated by Ang-1, that were predicted to be targets of the ETSs, and that showed enrichment in angiogenesis related biological processes based on GO annotation. To select the subset of genes we used Network Ontology Analysis 30 (Supplementary Table

3). We began by investigating the ability of Ang-1 to induce the selected subset of genes, extracted from the microarray 17. To evaluate this, HUVECs were treated with PBS or

206

Ang-1 for 4 h. Ang-1 significantly increased the expression of 16 out of 20 genes selected for the study, namely: Flt1, PLAU, EGR1, DUSP4, DIPA, Cdc42ep2, SH3BP5,

ANGPTL4, BHLHB2, SLC4A7, SMTN, RapGEF5, KLF2, HK2, Jag1, STC1 and

DUSP5 (Figure 8A). VEGFC, Cyclin D1, TRIB1 and HMGA2 were not significantly induced by Ang-1 and might have been false positive results from the microarray; we therefore used them as controls in the subsequent experiments. We also included VEGFA as a control. VEGFA is induced by Ang-1 at 4 h (Figure 8B), but is not regulated by

ETS1, ELK1 or ETV4.

To confirm whether ETS1, ELK1 or ETV4 regulate this subset of the Ang-1 transcriptome, HUVECs were transfected with either scrambled, ETS1, ELK1 or ETV4 siRNA. Transfection with ETS1 siRNA prevented Ang-1-induced expression of most genes, except for DUSP5, and the controls (VEGFA, VEGFC, TRIB1, HMGA2) (Figure

8B). These results indicate that ETS1 makes a significant contribution to the regulation of the Ang-1 transcriptome, either directly or indirectly. Transfection with ELK1 siRNA prevented Ang-1-induced expression of ANGPTL4, with no influence on the other genes

(Figure 8C). Transfection with ETV4 siRNA prevented Ang-1-induced expression of

DUSP4 and DIPA, with no effect on the other genes (Figure 8D). These results indicate that although ELK1 and ETV4 are involved in Ang-1-induced gene regulation, ETS1 is the major contributor. Interestingly transfection with either ETS1, ELK1 or ETV4 resulted in a decrease in the expression of DUSP4 at basal level (supplementary Figure

1). These results also indicate that the effects of ETS1, ELK1 and ETV4 on Ang-1- induced angiogenic responses could be mediated by their impact on the expression levels of several pro-angiogenic genes.

207

4.5 Discussion

The main findings of this study include: (1) in ECs, Ang-1 induces a significant increase in the expression and DNA binding of ETS1, through the p38, SAPK/JNK and PI-3 kinase/AKT pathways; (2) Ang-1 induces DNA binding and nuclear translocation of

ELK1 and ETV4; (3) ETS1, ELK1 and ETV4 play an important role in Ang-1 induced cell survival and capillary-like tube formation; (4) ETS1 and ETV4 contribute to Ang-1- induced migration; (5) inhibition of ETS1, ELK1 or ETV4 expression in ECs significantly attenuates the expression of Ang-1-induced pro-angiogenic genes.

Regulation of ETSs Transcription factors

In ECs, little is known about transcription factors that are involved in regulating gene expression downstream from Tie-2 receptors. We report here that exposure of ECs to

Ang-1 elicits significant induction of ETS1 expression and enhanced ETS1, ELK1 and

ETV4 DNA binding activity. Previous reports have confirmed that ETS factors are activated by many proangiogenesis growth factors, including VEGF, FGF, PDGF and endothelin-1 8, 9, 31. ETS-1 expression was also reported to be induced by Ang-1 in peripheral blood stem cells 32. We also reported previously that Ang-1-induced ERK1/2 activation lead to increased activation of ELK1 33.

The MAPK and the PI-3 kinase/AKT pathways have emerged as major signaling pathways through which Tie-2 receptors regulate EC functions such as survival, migration, adhesion, and proliferation 1. Moreover, reports have confirmed that the

Erk1/2 pathway plays an important role in mediating ETS1 and ELK1 responses to many mediators, including VEGF 34. ELK1 is activated by ERK1/2, p38 and SAPK/JNK

MAPKs 35. SAPK/JNK and p38 MAPKs have been reported as activators of ETV4, 36

208 while ETS1 is a known target of ERK1/2 MAPK 37. The present results indicate that

Ang-1-mediated induction of ETS1 transcription in HUVECs is dependent on the p38,

SAPK/JNK and PI-3 kinase/AKT pathways. It was previously reported that c Jun activates ETS1, which correlates with our observation that inhibition of SAPK/JNK prevented Ang-1-induced ETS1 regulation 38. Although ELK1 and ETV4 mRNA expression was not induced by Ang-1 treatment, inhibition of p38 MAPK led to an Ang-

1-induced increase of ELK1 mRNA levels. We observed here that the p38 MAPK pathway inhibits Ang-1-induced ELK1 transcription, and we speculate that this effect is achieved by suppressing the activation of the Erk1/2 pathway. It has been previously reported that the modulatory effect of the p38 MAPKs on Erk1/2 activity is mediated through the inhibitory effect of PP2A phosphatase on MEK1/2 phosphorylation 39. We also reported a similar effect of p38 MAPK on Ang-1-induced EGR1 transcription 4.

Several previous studies suggest that the ERK1/2 pathway contributes positively to Ang-

1-induced survival, migration, and proliferation of ECs 4, 5, 17, 33. In contrast, the p38

MAPK pathway acts as an inhibitory biological switch through which other stimuli regulate the degree to which the Ang-1/Tie-2 axis modulates angiogenic processes. Here we report that inhibition of ERK1/2 and SAPK/JNK significantly decreases the basal transcription of ETV4. Very little is known about the interactions between ETV4 and the

MAPKs pathway, although it has been demonstrated that ERK1/2 induces the transcription and activation of ETV4 40, 41. There are no reports of ex vivo or in vivo activation of ETV4 by SAPK/JNK, although it appears that ETV4 can be activated by

SAPK/JNK in vitro 35. Our study reveals that Ang-1 treatment leads to increased

209 transcription of ETS1, and induces nuclear accumulation of ETS1, ELK1, ETV4 and ETS factors in ECs.

Regulation of ECs Survival, Migration, and Capillary-Like Tube Formation by ETS factors

The present study reveals that Ang-1 stimulates ECs survival, migration and capillary- like tube formation, and knocking down ETS1, ELK1 or ETV4 expression using siRNA attenuates these effects. All three ETS factors are required for Ang-1-induced cell survival and capillary-like tube formation, while ELK1 is the only factor with no effect on Ang-1-induced migration. Moreover ELK1 and ETV4 knockdown decreased basal level of capillary-like tube formation. However, the mechanisms through which ETS1,

ELK1 and ETV4 modulates EC survival, migration, and capillary-like tube formation remain unclear. One possibility is that ETS factors upregulates the transcription of proangiogenesis genes and that these factors act to stimulate EC survival, migration and differentiation. This hypothesis is supported by reports that document the induction of proangiogensis genes by ETS factors 10-16. The functional importance of ETS1, ELK1 and ETV4 in Ang-1-induced angiogenic responses through the regulation of proangiogenesis genes was confirmed using siRNA. The present study confirmed that

Ang-1 induces16 out of 20 genes that were previously reported as being upregulated by

Ang-1 in a differential expression microarray (Figure 8A) 17. We show that ETS1 is required for Ang-1 induction of FLT1, PLAU, EGR1, DUSP4, DIPA, Cdc42ep2,

SH3BP5, ANGPTL4, BHLHB2, SLC4A7, SMTN, RapGEF5, KLF2, HK2, JAG1, STC1.

Several other genes are poorly studied and interestingly, most of these genes have been associated with angiogenic processes. FLT1 codes for a VEGF receptor and its role in angiogenesis is well defined 42. PLAU gene encodes a protease involved in degradation

210 of the extracellular matrix and possibly tumor cell migration and proliferation 43.

ANGPTL4 acts as a proangiogenic factor in various pathologies, including diabetes and arthritis 44-46. JAG1 promotes proliferation, survival and epithelial-to-mesenchymal transition 47. Our own studies reported that EGR1 activation is required for Ang-1- induced EC migration and differentiation 4. We also reported that Ang-1-induced EC survival, migration and differentiation depends on Ang-1 induction of DUSP4 29. This observation raises the possibility that ETS1 contributes to Ang-1-induced angiogenic responses through the upregulation of DUSP4 and EGR1. Interestingly, PLAU, FLT1 and EGR1 have been reported as direct gene targets of ETS1 11, 12, 15, 48. Therefore we speculate that FLT1, EGR1 and PLAU are direct gene targets of the Ang-1/ETS1 axis.

Although previous studies have reported the role that ETS1 plays in EC survival, migration and differentiation 49, 50, this is the first time that a transcriptional regulatory relationship has been established between Ang-1 and ETS1; importantly, a relationship which results in the alteration of the various EC responses to Ang-1. We also identified several new genes induced by Ang-1, which may be contributing to the angiogenic responses downstream of Ang-1, and to the Ang-1/ETS-axis; however, their function in this axis should be addressed in future studies.

Knocking down ELK1 abolished the induction of ANGPTL4 that results from Ang-1 treatment. We speculate that the observed effects of ELK1 on tube formation and on

Ang-1-induced tube formation and survival, are mediated at least in part through up regulation of ANGPTL4. The promoter region of ANGPTL4 contains several TFBS for

ETS factors. It is possible that ELK1 targets the ANGPTL4 gene, although no report

211 clearly defines this transcriptional regulatory relationship. Moreover, in vitro studies showed that overexpression of ANGPTL4 enhanced cell migration and inhibited cell death 51. Several other studies report the role of ELK1 in survival 52, 53 and differentiation of various cell types. ELK1 silencing was reported to abolish connective tissue growth factor-induced migration and capillary-like tube formation of fibroblasts 54. These reports are in correlation with our own observation of the role of ELK1 in Ang-1-induced angiogenic responses.

Finally, ETV4 silencing abolished Ang-1-induction of DUSP4 and DIPA, respectively.

DIPA is poorly studied, but has been implicated in adipocyte differentiation 55 and although its contribution to Ang-1-induced responses might be minimal, it’s worth investigating further. We speculate that the effect of ETV4 on Ang-1-induced angiogenic responses may be mediated primarily through DUSP4. As previously mentioned, Ang-1 upregulation of DUSP4 is required for Ang-1-induced survival, migration and differentiation of ECs 29. Furthermore, we hypothesize that our previous observation on the role of DUSP4 in Ang-1-induced responses requires the activation of both ETS1 and

ETV4 TFs, which likely work in conjunction with c Jun. Further consideration should be given to the reported and potential cooperative interaction between the various ETS factors. Specifically, ETV4 was reported to activate ETS1 56, while c Jun, ELK1 and

ETV4 are potential cooperating transcription factors, as determined by the TRRUST algorithm 57. ETS1 was also reported to be activated by c Jun 38. Ang-1 activation of c

Jun has already been reported by our group, and taken together with our current results we could speculate that the Ang-1 transcriptome depends on the relationship of transcriptional regulation between c Jun and the ETS factors, ETS1, ELK1 and ETV4;

212 namely, instead of regulating the Ang-1 transcriptome individually, they form a complex network of interacting TFs. Indeed, it was reported that ETS factors do not exert their specificity through differences in their TFBSs, since ETS factors bind very similar DNA sequences. ETS factors vary considerably outside of their DNA binding domains, allowing them to have diverse trans-regulatory roles. It’s the regulation and activation of these factors by distinct classes of MAPKs that give rise to the specificity of their DNA binding activity 35.

Additional studies will clearly be needed to investigate the precise downstream pathways through which ETS factors regulate Ang-1-induced survival, migration and differentiation of ECs. The use of chromatin immunoprecipitation (ChIP) would be necessary to identify the direct gene targets of ETS1, ELK1 and ETV4.

In summary we report, as an original finding, that Ang-1 induces the expression and activation of the transcription factor ETS1, ELK1 and ETV4 and that this effect is mediated through the p38, SAPK/JNK and PI-3 kinase/AKT pathways for ETS1. Our results also indicate that these ETS factors contribute significantly to Ang-1-induced survival, migration and capillary-like tube formation of ECs, through the regulation of

Ang-1-induced proangiogenic genes.

4.6 Acknowledgments

We would like to thank Dr Jiwaji for the kind gift of pMN vectors to asses transcriptional activity of ETS1 and ELK1.

213

4.7 References

(1) Brindle NP, Saharinen P, Alitalo K. Signaling and functions of angiopoietin-1 in vascular protection. Circ Res 2006 April 28;98(8):1014-23.

(2) Daly C, Wong V, Burova E, Wei Y, Zabski S, Griffiths J, Lai KM, Lin HC, Ioffe E, Yancopoulos GD, Rudge JS. Angiopoietin-1 modulates endothelial cell function and gene expression via the transcription factor FKHR (FOXO1). Genes Dev 2004 May 1;18(9):1060-71.

(3) Sako K, Fukuhara S, Minami T, Hamakubo T, Song H, Kodama T, Fukamizu A, Gutkind JS, Koh GY, Mochizuki N. Angiopoietin-1 induces Kruppel-like factor 2 expression through a phosphoinositide 3-kinase/AKT-dependent activation of myocyte enhancer factor 2. J Biol Chem 2008 December 23;284(9):5592-601.

(4) Abdel-Malak NA, Mofarrahi M, Mayaki D, Khachigian LM, Hussain SN. Early Growth Response-1 Regulates Angiopoietin-1-Induced Endothelial Cell Proliferation, Migration, and Differentiation. Arterioscler Thromb Vasc Biol 2008 December 26;29(2):209-16.

(5) Abdel-Malak NA, Srikant CB, Kristof AS, Magder SA, Di Battista JA, Hussain SNA. Angiopoietin-1 promotes endothelial proliferation and migration through AP-1 dependent autocrine production of interleukin-8. Blood 2008;111(8):4145- 54.

(6) Christensen RA, Fujikawa K, Madore R, Oettgen P, Varticovski L. NERF2, a member of the Ets family of transcription factors, is increased in response to hypoxia and angiopoietin-1: a potential mechanism for Tie2 regulation during hypoxia. J Cell Biochem 2002;85(3):505-15.

(7) Sharrocks AD. The ETS-domain transcription factor family. Nat Rev Mol Cell Biol 2001 November;2(11):827-37.

(8) Heo SH, Choi YJ, Ryoo HM, Cho JY. Expression profiling of ETS and MMP factors in VEGF-activated endothelial cells: role of MMP-10 in VEGF-induced angiogenesis. J Cell Physiol 2010 September;224(3):734-42.

(9) Squarzoni P, Parveen F, Zanetti L, Ristoratore F, Spagnuolo A. FGF/MAPK/Ets signaling renders pigment cell precursors competent to respond to Wnt signal by directly controlling Ci-Tcf transcription. Development 2011 April;138(7):1421- 32.

(10) Hasegawa Y, Abe M, Yamazaki T, Niizeki O, Shiiba K, Sasaki I, Sato Y. Transcriptional regulation of human angiopoietin-2 by transcription factor Ets-1 1. Biochem Biophys Res Commun 2004 March 26;316(1):52-8.

214

(11) Valter MM, Hugel A, Huang HJ, Cavenee WK, Wiestler OD, Pietsch T, Wernert N. Expression of the Ets-1 transcription factor in human astrocytomas is associated with Fms-like tyrosine kinase-1 (Flt-1)/vascular endothelial growth factor receptor-1 synthesis and neoangiogenesis. Cancer Res 1999 November 1;59(21):5608-14.

(12) Lee SH, Bahn JH, Choi CK, Whitlock NC, English AE, Safe S, Baek SJ. ESE- 1/EGR-1 pathway plays a role in tolfenamic acid-induced apoptosis in colorectal cancer cells. Mol Cancer Ther 2008 December;7(12):3739-50.

(13) Schwachtgen JL, Janel N, Barek L, Duterque-Coquillaud M, Ghysdael J, Meyer D, Kerbiriou-Nabias D. Ets transcription factors bind and transactivate the core promoter of the von Willebrand factor gene. Oncogene 1997 December 18;15(25):3091-102.

(14) Ghosh S, Basu M, Roy SS. ETS-1 protein regulates vascular endothelial growth factor-induced matrix metalloproteinase-9 and matrix metalloproteinase-13 expression in human ovarian carcinoma cell line SKOV-3. J Biol Chem 2012 April 27;287(18):15001-15.

(15) Igarashi T, Abe M, Oikawa M, Nukiwa T, Sato Y. Retinoic acids repress the expression of ETS-1 in endothelial cells. Tohoku J Exp Med 2001 May;194(1):35- 43.

(16) Dube A, Thai S, Gaspar J, Rudders S, Libermann TA, Iruela-Arispe L, Oettgen P. Elf-1 is a transcriptional regulator of the Tie2 gene during vascular development. Circ Res 2001 February 2;88(2):237-44.

(17) Abdel-Malak N, Harfouche R, Hussain SN. Transcriptome of angiopoietin 1- activated human umbilical vein endothelial cells. Endothelium 2007 November;14(6):285-302.

(18) Siepel A, Bejerano G, Pedersen JS, Hinrichs AS, Hou M, Rosenbloom K, Clawson H, Spieth J, Hillier LW, Richards S, Weinstock GM, Wilson RK, Gibbs RA, Kent WJ, Miller W, Haussler D. Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Res 2005 August;15(8):1034-50.

(19) Miller W, Rosenbloom K, Hardison RC, Hou M, Taylor J, Raney B, Burhans R, King DC, Baertsch R, Blankenberg D, Kosakovsky Pond SL, Nekrutenko A, Giardine B, Harris RS, Tyekucheva S, Diekhans M, Pringle TH, Murphy WJ, Lesk A, Weinstock GM, Lindblad-Toh K, Gibbs RA, Lander ES, Siepel A, Haussler D, Kent WJ. 28-way vertebrate alignment and conservation track in the UCSC Genome Browser. Genome Res 2007 December;17(12):1797-808.

(20) Matys V, Kel-Margoulis OV, Fricke E, Liebich I, Land S, Barre-Dirrie A, Reuter I, Chekmenev D, Krull M, Hornischer K, Voss N, Stegmaier P, Lewicki-Potapov B, Saxel H, Kel AE, Wingender E. TRANSFAC and its module TRANSCompel: 215

transcriptional gene regulation in eukaryotes. Nucleic Acids Res 2006 January 1;34(Database issue):D108-D110.

(21) Portales-Casamar E, Thongjuea S, Kwon AT, Arenillas D, Zhao X, Valen E, Yusuf D, Lenhard B, Wasserman WW, Sandelin A. JASPAR 2010: the greatly expanded open-access database of transcription factor binding profiles. Nucleic Acids Res 2010 January;38(Database issue):D105-D110.

(22) Shnyreva M, Weaver WM, Blanchette M, Taylor SL, Tompa M, Fitzpatrick DR, Wilson CB. Evolutionarily conserved sequence elements that positively regulate IFN-gamma expression in T cells. Proc Natl Acad Sci U S A 2004 August 24;101(34):12622-7.

(23) Blanchette M, Tompa M. Discovery of regulatory elements by a computational method for phylogenetic footprinting. Genome Res 2002 May;12(5):739-48.

(24) Defrance M, Touzet H. Predicting transcription factor binding sites using local over-representation and comparative genomics. BMC Bioinformatics 2006;7:396.

(25) Bulyk ML. Computational prediction of transcription-factor binding site locations. Genome Biol 2003;5(1):201.

(26) Hestand MS, van GM, Villerius MP, van Ommen GJ, den Dunnen JT, 't Hoen PA. CORE_TF: a user-friendly interface to identify evolutionary conserved transcription factor binding sites in sets of co-regulated genes. BMC Bioinformatics 2008;9:495.

(27) Marstrand TT, Frellsen J, Moltke I, Thiim M, Valen E, Retelska D, Krogh A. Asap: a framework for over-representation statistics for transcription factor binding sites. PLoS ONE 2008;3(2):e1623.

(28) Jiwaji M, Daly R, Gibriel A, Barkess G, McLean P, Yang J, Pansare K, Cumming S, McLauchlan A, Kamola PJ, Bhutta MS, West AG, West KL, Kolch W, Girolami MA, Pitt AR. Unique reporter-based sensor platforms to monitor signalling in cells. PLoS ONE 2012;7(11):e50521.

(29) Echavarria R, Hussain SN. Regulation of angiopoietin-1/Tie-2 receptor signaling in endothelial cells by dual-specificity phosphatases 1, 4, and 5. J Am Heart Assoc 2013;2(6):e000571.

(30) Wang J, Huang Q, Liu ZP, Wang Y, Wu LY, Chen L, Zhang XS. NOA: a novel Network Ontology Analysis method. Nucleic Acids Res 2011 July;39(13):e87.

(31) Naito S, Shimizu S, Maeda S, Wang J, Paul R, Fagin JA. Ets-1 is an early response gene activated by ET-1 and PDGF-BB in vascular smooth muscle cells. Am J Physiol 1998 February;274(2 Pt 1):C472-C480.

216

(32) Kim MS, Lee CS, Hur J, Cho HJ, Jun SI, Kim TY, Lee SW, Suh JW, Park KW, Lee HY, Kang HJ, Lee DS, Koh GY, Nakagami H, Morishita R, Park YB, Kim HS. Priming with angiopoietin-1 augments the vasculogenic potential of the peripheral blood stem cells mobilized with granulocyte colony-stimulating factor through a novel Tie2/Ets-1 pathway. Circulation 2009 December 1;120(22):2240- 50.

(33) Harfouche R, Gratton JP, Yancopoulos GD, Hussain SNA. Angiopoietin-1 activates both anti- and pro-apoptotic mitogen activated protein kinases. FASEB J 2003;17(11):1523-5.

(34) Watanabe D, Takagi H, Suzuma K, Suzuma I, Oh H, Ohashi H, Kemmochi S, Uemura A, Ojima T, Suganami E, Miyamoto N, Sato Y, Honda Y. Transcription factor Ets-1 mediates ischemia- and vascular endothelial growth factor-dependent retinal neovascularization. Am j Pathol 2004 May;164(5):1827-35.

(35) Yang SH, Whitmarsh AJ, Davis RJ, Sharrocks AD. Differential targeting of MAP kinases to the ETS-domain transcription factor Elk-1. EMBO J 1998 March 16;17(6):1740-9.

(36) Selvaraj N, Kedage V, Hollenhorst PC. Comparison of MAPK specificity across the ETS transcription factor family identifies a high-affinity ERK interaction required for ERG function in prostate cells. Cell Commun Signal 2015;13:12.

(37) Yang BS, Hauser CA, Henkel G, Colman MS, Van BC, Stacey KJ, Hume DA, Maki RA, Ostrowski MC. Ras-mediated phosphorylation of a conserved threonine residue enhances the transactivation activities of c-Ets1 and c-Ets2. Mol Cell Biol 1996 February;16(2):538-47.

(38) Oka T, Rairkar A, Chen JH. Structural and functional analysis of the regulatory sequences of the ets-1 gene. Oncogene 1991 November;6(11):2077-83.

(39) Westermarck J, Li SP, Kallunki T, Han J, Kahari VM. p38 mitogen-activated protein kinase-dependent activation of protein phosphatases 1 and 2A inhibits MEK1 and MEK2 activity and collagenase 1 (MMP-1) gene expression. Mol Cell Biol 2001 April;21(7):2373-83.

(40) Keld R, Guo B, Downey P, Gulmann C, Ang YS, Sharrocks AD. The ERK MAP kinase-PEA3/ETV4-MMP-1 axis is operative in oesophageal adenocarcinoma. Mol Cancer 2010;9:313.

(41) Fontanet P, Irala D, Alsina FC, Paratcha G, Ledda F. Pea3 transcription factor family members Etv4 and Etv5 mediate retrograde signaling and axonal growth of DRG sensory neurons in response to NGF. J Neurosci 2013 October 2;33(40):15940-51.

(42) Shibuya M. Vascular endothelial growth factor receptor-1 (VEGFR-1/Flt-1): a dual regulator for angiogenesis. Angiogenesis 2006;9(4):225-30.

217

(43) Huang CY, Chang MC, Huang WY, Huang CT, Tang YC, Huang HD, Kuo KT, Chen CA, Cheng WF. Urokinase-type plasminogen activator resulting from endometrial carcinogenesis enhances tumor invasion and correlates with poor outcome of endometrial carcinoma patients. Sci Rep 2015;5:10680.

(44) Mousavizadeh R, Scott A, Lu A, Ardekani GS, Behzad H, Lundgreen K, Ghaffari M, McCormack RG, Duronio V. Angiopoietin-like 4 promotes angiogenesis in the tendon and is increased in cyclically loaded tendon fibroblasts. J Physiol 2016 June 1;594(11):2971-83.

(45) Hermann LM, Pinkerton M, Jennings K, Yang L, Grom A, Sowders D, Kersten S, Witte DP, Hirsch R, Thornton S. Angiopoietin-like-4 is a potential angiogenic mediator in arthritis. Clin Immunol 2005 April;115(1):93-101.

(46) Babapoor-Farrokhran S, Jee K, Puchner B, Hassan SJ, Xin X, Rodrigues M, Kashiwabuchi F, Ma T, Hu K, Deshpande M, Daoud Y, Solomon S, Wenick A, Lutty GA, Semenza GL, Montaner S, Sodhi A. Angiopoietin-like 4 is a potent angiogenic factor and a novel therapeutic target for patients with proliferative diabetic retinopathy. Proc Natl Acad Sci U S A 2015 June 9;112(23):E3030- E3039.

(47) Pedrosa AR, Trindade A, Carvalho C, Graca J, Carvalho S, Peleteiro MC, Adams RH, Duarte A. Endothelial Jagged1 promotes solid tumor growth through both pro-angiogenic and angiocrine functions. Oncotarget 2015 September 15;6(27):24404-23.

(48) Kitange G, Shibata S, Tokunaga Y, Yagi N, Yasunaga A, Kishikawa M, Naito S. Ets-1 transcription factor-mediated urokinase-type plasminogen activator expression and invasion in glioma cells stimulated by serum and basic fibroblast growth factors. Lab Invest 1999 April;79(4):407-16.

(49) Iwasaka C, Tanaka K, Abe M, Sato Y. Ets-1 regulates angiogenesis by inducing the expression of urokinase-type plasminogen activator and matrix metalloproteinase-1 and the migration of vascular endothelial cells. J Cell Physiol 1996 December;169(3):522-31.

(50) Sato Y, Teruyama K, Nakano T, Oda N, Abe M, Tanaka K, Iwasaka-Yagi C. Role of transcription factors in angiogenesis: Ets-1 promotes angiogenesis as well as endothelial apoptosis. Ann N Y Acad Sci 2001 December;947:117-23.

(51) Li X, Chen T, Shi Q, Li J, Cai S, Zhou P, Zhong Y, Yao L. Angiopoietin-like 4 enhances metastasis and inhibits apoptosis via inducing bone morphogenetic protein 7 in colorectal cancer cells. Biochem Biophys Res Commun 2015 November 6;467(1):128-34.

(52) Sun NK, Huang SL, Chang TC, Chao CC. Sorafenib induces endometrial carcinoma apoptosis by inhibiting Elk-1-dependent Mcl-1 transcription and

218

inducing Akt/GSK3beta-dependent protein degradation. J Cell Biochem 2013 August;114(8):1819-31.

(53) Vickers ER, Kasza A, Kurnaz IA, Seifert A, Zeef LA, O'donnell A, Hayes A, Sharrocks AD. Ternary complex factor-serum response factor complex-regulated gene activity is required for cellular proliferation and inhibition of apoptotic cell death. Mol Cell Biol 2004 December;24(23):10340-51.

(54) Liu SC, Chuang SM, Hsu CJ, Tsai CH, Wang SW, Tang CH. CTGF increases vascular endothelial growth factor-dependent angiogenesis in human synovial fibroblasts by increasing miR-210 expression. Cell Death Dis 2014;5:e1485.

(55) Bezy O, Elabd C, Cochet O, Petersen RK, Kristiansen K, Dani C, Ailhaud G, Amri EZ. Delta-interacting protein A, a new inhibitory partner of CCAAT/enhancer-binding protein beta, implicated in adipocyte differentiation. J Biol Chem 2005 March 25;280(12):11432-8.

(56) Chen JH, Wright CD. PEA3, Oct 1 and Oct 2 positively regulate the human ETS1 promoter. Oncogene 1993 December;8(12):3375-83.

(57) Han H, Shim H, Shin D, Shim JE, Ko Y, Shin J, Kim H, Cho A, Kim E, Lee T, Kim H, Kim K, Yang S, Bae D, Yun A, Kim S, Kim CY, Cho HJ, Kang B, Shin S, Lee I. TRRUST: a reference database of human transcriptional regulatory interactions. Sci Rep 2015;5:11432.

219

4.8 Figures

Figure 4.1: Predicting transcription factor of Ang-1 regulated gene set

220

Figure 4.2: Ang-1 regulates ETSs expression and DNA binding activity

221

Figure 4.3: Ang-1 induces ETS-1 accumulation in the nucleus

222

Figure 4.4: Ang-1 induces ELK1 accumulation in the nucleus

223

Figure 4.5: Ang-1 induces ETV4 accumulation in the nucleus

224

Figure 4.6: Regulation of ETS1, ELK1 and ETV4 by MAPK, AKT and mTOR pathway

225

Figure 4.7: Role of ETS1, ELK1 and ETV4 in Ang-1 angiogenic responses

226

Figure 4.8: ETS1, ELK1 and ETV4 mediate Ang-1 transcriptome

227

Supplementary Figure 4.1: ETS1, ELK1 and ETV4 effect on basal level of Ang-1- induced genes

228

4.9 Figure Legends

Figure 4.1: Predicting transcription factor of Ang-1 regulated gene set

Heatmap representing 58 genes up-regulated by Ang-1 and clustered according to similar

Z-scores obtained for 62 PMW from Transfac, for flanking region of 10kb upstream and downstream of each genes. The color scale represents the value of the Z-score.

Figure 4.2: Ang-1 regulates ETSs expression and DNA binding activity

A) Time course of ETS1, ELK1 and ETV4 mRNA in HUVECs exposed to Ang-1

(n=6 per group) *p<0.05 compared with PBS.

B) Immunoblot of ETS1 DNA binding activity.

C) Immunoblot of ETV4 DNA binding activity.

D) HUVECs transfected with Luciferase reporter containing ETS1 TFBS, and then

treated with Ang-1 for 4 h (n=6 per group), Values are mean±SE *p<0.05

compared with PBS.

E) HUVECs transfected with Luciferase reporter containing ELK1 TFBS, and then

treated with Ang-1 for 4 h (n=6 per group). Values are mean±SE *p<0.05

compared with PBS.

Figure 4.3: Ang-1 induces ETS-1 accumulation in the nucleus

A) Immunostaining for ETS1 protein in HUVECs exposed to PBS or Ang-1

(300ng/ml) for 1 h, 3 h or 6 h.

B) Quantification of nuclear staining (n>130 per group). Values are mean±SE

*p<0.05 compared to PBS.

229

Figure 4.4: Ang-1 induces ELK1 accumulation in the nucleus

A) Immunostaining for ELK1 protein in HUVECs exposed to PBS or Ang-1

(300ng/ml) for 1 h, 3 h or 6 h.

B) Quantification of nuclear staining (n>130 per group). Values are mean±SE

*p<0.05 compared to PBS.

Figure 4.5: Ang-1 induces ETV4 accumulation in the nucleus

A) Immunostaining for ETV4 protein in HUVECs exposed to PBS or Ang-1

(300ng/ml) for 1 h, 3 h or 6 h.

B) Quantification of nuclear staining (n>130 per group). Values are mean±SE

*p<0.05 compared to PBS.

Figure 4.6: Regulation of ETS1, ELK1 and ETV4 by MAPK, AKT and mTOR pathway

A to C: Effect of pretreatment with vehicle; U0126 (ERK1/2 inhibitor); BIRB796 (p38 inhibitor); SP600125 (SAPK/JNK inhibitor); Wortmannin (WM) (PI-3K inhibitor) or

Rapamycin (mTOR inhibitor) on mRNA expression of A) ETS1, B) ELK1 and C) ETV4.

(n=6 per group) Values are mean±SE *p<0.05 compared to PBS, #p<0.05 compared to vehicle with PBS and ɸ p<0.05 compared to vehicle with Ang-1.

Figure 4.7: Role of ETS1, ELK1 and ETV4 in Ang-1 angiogenic responses

A) Cell count in HUVECs transfected with scrambled, ETS1, ELK1 or ETV4 siRNA

and maintained for 24 h in medium containing 20% FBS, 2% FBS or 0.2% FBS

with Ang-1. Values are mean±SE (n=12 per group), *p<0.05 compared to

20%FBS, # p<0.05 compared to 2% FBS.

230

B) Percent wound healing in HUVECs transfected with scrambled, ETS1, ELK1 or

ETV4 siRNA and maintained for 8 h with PBS or Ang-1 (n=12 per group).

Values are mean±SE, *p<0.05 compared to PBS.

C) Number of tubes formed in HUVECs transfected with scrambled, ETS1, ELK1 or

ETV4 siRNA and maintained for 24 h with PBS or Ang-1 (n=12 per group).

Values are mean±SE, *p<0.05 compared to PBS, #p<0.05 compared to

scrambled.

Figure 4.8: ETS1, ELK1 and ETV4 mediate Ang-1 transcriptome

A) Expression of genes induced by Ang-1 at 4 h (n=12 per group). Values are

mean±SE, *p<0.05 compared to PBS.

B) Expression of genes induced by Ang-1 at 4 h in HUVECs transfected with

scrambled or ETS1 siRNA (n=12 per group). Values are mean±SE, *p<0.05

compared to PBS.

C) Expression of genes induced by Ang-1 at 4 h in HUVECs transfected with

scrambled or ELK1 siRNA (n=6 per group). Values are mean±SE, *p<0.05

compared to PBS.

D) Expression of genes induced by Ang-1 at 4 h in HUVECs transfected with

scrambled or ETV4 siRNA (n=6 per group). Values are mean±SE, *p<0.05

compared to PBS.

231

Supplementary Figure 4.1: ETS1, ELK1 and ETV4 effect on basal level of Ang-1- induced genes

A) Basal level of expression of DUSP4 and Cyclin D1 in HUVECs transfected with

scrambled or ETS1 siRNA (n=12 per group). Values are mean±SE, *p<0.05

compared to PBS.

B) Basal level of expression of DUSP4 and DIPA in HUVECs transfected with

scrambled or ELK1 siRNA (n=6 per group). Values are mean±SE, *p<0.05

compared to PBS

C) Basal level of expression of DUSP4 and TRIB1 in HUVECs transfected with

scrambled or ETV4 siRNA (n=6 per group). Values are mean±SE, *p<0.05

compared to PBS.

232

4.10 Tables

Table 4.1: Z-scores for the Ang-1 dataset over TRANSFAC profiles (ETS-type Family)

Matrix Transcription Factor Z-score

M00025 ELK-1 6.97

M00771 ETS-1 7.43

M00655 ETV-4 6.15

M00340 c-ETS-2 4.90

M00497 STAT3 1.94

M00499 STAT5A 1.70

M00175 AP-4 1.57

233

Supplementary Table 4.1: Primer List Type Sequence (5’ 3’) Accession Expected Size Number (bp)

-actin F: AGAAAATCTGGCACCACACC NM_001101 126 R: GGGGTGTTGAAGGTCTCAAA GAPDH F: AAGAAGGTGGTGAAGCAGGCG NM_023964.1 166 R: ACCAGGAAATGAGCTTGACAA VEGFA F: CTACCTCCACCATGCCAAGT NM_001025366.1 281 R: CACACAGGATGGCTTGAAGA VEGFC F: TGTACAAGTGTCAGCTAAGG NM_005429 187 R:CCACATCTATACACACCTCC ELK1 F: ATCTGTGACGCTGTGGCAGT NM_001114123.2 100 R: CCAGCTTGAATTCACCACCA ETV4 F: CCTGAGATCCTCTGGCACCT NM_001079675 116 R: CCCCTCCCTGAGATGTGAAG ETS1 F: CAAGCCTGTCATTCCTGCTG NM_005238 140 R: TGAATTCCCAGCCATCTCCT TRIB1 F: GAGGAGAGAACCCAGCTTAGA NM_025195 179 R: GAGGATCTCAGGGCTCACGTA PLAU F: GTGAGCGACTCCAAAGGCA NM_025195 126 R: GCAGTTGCACCAGTGAATGTT Cyclin D1 F: TTGCACACCTCTGGCTCTGT NM_007631.2 227 R: ACCCTCCAATAGCAGCGAAA HMGA2 F: ACCCAGGGGAAGACCCAAA NM_003483 92 R: CCTCTTGGCCGTTTTTCTCCA DUSP4 F: AGCATCATCTCGCCCAACTT NM_001394 169 R: AGACCGGAAAGCTGAAGACG DIPA F: CTCATGCAGGAGGTGAATCG NM_006848 124 R: AGTCCAGGAAGCAGCAGAGG Cdc42ep2 F: ACCAAGGTGCCCATCTATCTG NM_006779 128 R: CCACTGCCAATATGAATGGTGTG Flt1 F: CACTGGGCAGCAGACAAATC NM_002019 109 R: TCACACCTTGCTTCGGAATG SH3BP5 F: GAGCGAGCTGGTGCATAAGG NM_004844 109 R: TGGACTTGTTGATGGCTCTCT BHLHB2 F: AGAGACGTGACCGGATTAAC NM_003670 102 R: GAACCACTGCTTTTTCCAAG SLC4A7 F: AGTCCTCCTTCTAGCCCTAGC NM_003615 123 R:GGCGGATGAATTACTACTGTGG SMTN F: CTTAGCCCTCAGAACCGACG NM_134269 103 R: TGGACCAGACAGCGGTAGAA

234

Type Sequence (5’ 3’) Accession Expected Size Number (bp) ANGPTL4 F: CCACTTGGGACCAGGATCAC NM_139314 115 R: CGGAAGTACTGGCCGTTGAG RapGEF5 F: AAGACCTGGCGGACACTTTG NM_012294 142 R: GCTCTTGCTCGTGAATTGAATTG EGR1 F: CACGCCGAACACTGACATTT NM_001964 135 R: TAGTCGGGGATCATGGGAAC KLF2 F: GCACGCACACAGGTGAGAAG NM_016270.2 269 R: ACCAGTCACAGTTTGGGAGGG HK2 F: TTTGACCACATTGCCGAATGC NM_000189 117 R: GGTCCATGAGACCAGGAAACT JAG1 F: CGGGATTTGGTTAATGGTTATC 153 R: ATAGTCACTGGCACGGTTGTAGC DUSP5 F: GGATCCCTGTGGAAGACAGC NM_004419 221 R: GACCATGCTCCTCCTCTGCT STC1 F:TGGTGATCAGTGCTTCTGCAAC NM_003155.2 169 R:CTCAGTGATGGCTTCAGGGTTC

235

Supplementary Table 4.2: Description of luciferase reporter

Plasmid TF Sequence Repe Sequ Refere Unique Specific F qPCR primer ats ence nce reporter sequence ID sequence pMN191 Elk CCATGGA T002 [48- CCCGCGTG GTGCCCGTGTTCCTTT -1 GGG 50 49] CCCGTGTTC CTTTTT pMN141 Ets GGAGGAA 3x R040 [12] GCGCTCCC CTCTGTTGCGCTCCCT GT 51 TCTGTTGCG CTCCCT

236

Supplementary Table 4.3: Enrichment in biological process for a subset of 22 Ang- 1-up-regulated gene obtain with NOA GO: Term Term Name Nb genes

GO:0030334 Regulation of cell migration 7

GO:0001525 Angiogenesis 6

GO:00512703 Regulation of cellular component 7 movement

GO:0040012 Regulation of locomotion 7

GO:0043536 Positive regulation of blood vessel 3 endothelial cell migration Regulation of VEGFR signaling GO:0030947 pathway 3 Response to chemical stimulus GO:00422211 11 GO:0001666 Response to Hypoxia 5 GO:00704823 Response to oxygen levels 5 GO:00435354 Regulation of blood vessel endothelial cell migration 3

237

Chapter 5: Original Contribution to Scientific Knowledge

238

The research carried out to produce this thesis has generated several original contributions:

Chapter 2: NOX2, NOX4 and Mitochondrial-Derived Reactive Oxygen Species

(mtROS) Contribute to Angiopoietin-1 Angiogenic Responses in Endothelial Cells

1- In ECs Angiopoietin-1 triggers rapid increases in reactive oxygen species (ROS)

production in endothelial cells that is dependant on Tie-2 receptor.

2- NOX2, NOX4 and mitochondrial ROS are required for Angiopoietin-1-induced

ROS production

3- NOX2, NOX4 and mitochondrial ROS are required for Angiopoietin-1-induced

phosphorylation of AKT, ERK1/2, p38, and SAPK/JNK proteins in endothelial

cells.

4- NOX2 promotes survival of serum-deprived endothelial cells.

5- NOX4 and mitochondrial ROS are positive regulator angiopoietin-1-induced

migration and tube formation in endothelial cells.

Chapter 3: Regulation of angiogenesis by Angiopoietin-1: Roles of miR-640 and ZFP91

1. Ang-1 transcriptionally downregulates miR-640 and its gene target GATAD2A

2. miR-640 has anti-angiogenic properties. miR-640 overexpression prevents Ang-1-

induced cell migration and differentiation while miR-640 inhibition as the

opposite effect.

3. miR-640 directly targets ZFP91 3’UTR through binding to at least 2 potential

seed sequences.

239

4. Ang-1 modulates ZFP91 expression in ECs and ZFP91 is required for Ang-1

induced migration and differentiation.

Chapter 4: ETS1, ELK1 and ETV4 Contribute to Angiopoietin-1 Angiogenic Responses in Endothelial Cells

1- Angiopoeitin-1 induces an increase in ETS1 mRNA levels and increases its DNA

binding activity

2- Angiopoietin-1 triggers an increase of ELK1 and ETV4 DNA binding activity

3- ETS1, ELK1 and ETV4 play an important role in Ang-1 induced survival, and

tube formation

4- ETS1 and ETV4 play an important role in Ang-1 induced migration.

5- Angiopoietin-1 transcriptionally regulates FLT1, PLAU, DIPA, Cdc42ep2,

SH3BP5, ANGPTL4, BHLHB2, SLC4A7, SMTN, RapGEF5, HK2, JAG1, and

STC1.

6- ETS1 directly or indirectly regulates the levels of Ang-1 induced FLT1, PLAU,

DUSP4, EGR1, DIPA, Cdc42ep2, SH3BP5, ANGPTL4, BHLHB2, SLC4A7,

SMTN, RapGEF5, HK2, JAG1, and STC1.

7- ELK1 directly or directly regulates the levels of Ang-1 induced ANGPTL4

8- ETV4 directly or indirectly regulates the levels of Ang-1 DIPA and DUSP4

240

241

Chapter 6: Discussion and Opportunities for Future Research

242

NOX2, NOX4 and Mitochondrial-Derived Reactive Oxygen Species (mtROS)

Contribute to Angiopoietin-1 Angiogenic Responses in Endothelial Cells

Ang-1 simultaneously activates the MAPK signaling pathways ERK1/2, p38 and

SAPK/JNK, as well as the PI3K/AKT pathway to exert its biological function in the vasculature [6;7;12;18;109;118;197]. Previous studies have shown the role of ROS in regulating the activity of the MAPKs and AKT pathway and our laboratory highlighted the contribution of Ang-1 in the induction of ROS production in ECs [7;225]. ROS regulation of MAPKs and AKT pathway are in part responsible for ROS effects on Ang-

1-induced angiogenesis. Our results show that Ang-1 triggers in a Tie-2 dependent manner, a transient release of ROS in ECs, relying on the presence of functional forms of

NOX2, NOX4 and unaltered mitochondrial ROS release. Moreover, each source of ROS showed distinct specificity in the activation of the MAPKs and PI3K/AKT pathway and their biological function was found to be non-redundant. NOX2 is involved in phosphorylation of AKT, p38, and SAPK/JNK proteins, NOX4 is involved in SAPK/JNK phosphorylation, and mtROS are involved in p38 and SAPK/JNK phosphorylation.

The induction of ROS release as modulators of MAPK and PI3K/AKT signaling pathways activated by angiogenic factors has been described before for VEGF and PDGF

[225;228]. Our observations that Ang-1-induced released of ROS from NOX2 is required for AKT, p38, and SAPK/JNK and is involved in cell survival and inhibition of apoptosis correlates with observations from previous studies [7;8]. Our results also show that Ang-1-induced NOX4 and mitochondrial ROS release is involved in the modulation of cell migration and tube formation. These results highlight the distinct contribution of

243 different ROS sources to Ang-1-induced angiogenic responses and correlates with previous reports on the role of NOX4 and mitochondrial-derived ROS [386].

The interplay between NOX4 and mitochondrial ROS represents an interesting findings of this study. Our study shows that basal mtROS production declines in cells with NOX4 knockdown. This confirms that NOX4 contributes to ROS release into the mitochondria, but we also found that Ang-1 is still capable of inducing mtROS production despite

NOX4 knockdown. This indicate that NOX4-derived ROS are not the direct source of the observed mtROS release and our results in agreement with the concept that NOX4 may contribute to upregulation of mtROS release through ROS feed-forward mechanism.

Indeed, it was reported that depletion of p22phox an essential regulatory subunit for both

NOX2 and NOX4 function, as well as NOX4 specific knockdown, led to a significant decrease in ROS production in mitochondria [387-389]. More specifically NOX4 was shown to inhibit complex I in the mitochondria, and inhibition of complex I, leads to increase mtROS production [389]. However, additional experimentation is required to elucidate how NOX4 contributes to basal mtROS production, including localization of

NOX4 and localization of NOX4-derived ROS.

Additional experimentation is also required to elucidate the exact signaling mechanisms behind NOXs and mitochondrial ROS modulation of the MAPKs and AKT pathways. We need to identify the targets of NOX2, NOX4 and mitochondrial derived

ROS. We do not exclude the possibility that post-translational modification of the

DUSPs are part of the mechanism at least in the maintenance of basal level of activation of the MAPKs. DUSPs are very sensitive to oxidative post-translational modifications

[390]. DUSPs contain a redox-sensitive cysteine motif in their catalytic domain and can

244 undergo oxidation by ROS which results in the loss of phosphatase activity [390].

Perhaps, oxidation and inhibition of DUSPs phosphatase activity by ROS could explain in part how NOX2, NOX4 or the mitochondrial-derived ROS contribute to the basal level of MAPK activation. Moreover, we believe the distinct function of NOX2, NOX4 and mitochondrial-derived ROS depends on the nature of ROS and their sub-cellular localization. Finally, it would be important to identify how Ang-1 is activating NOX2,

NOX4 and the specific mechanism of how it induces the release of ROS from the mitochondria. NOX2, is known to be activated through assembly of its regulatory subunits. NOX4 on the other hand, is believe to be constitutively active. Our results clearly show that NOX4 ROS release can be induced by Ang-1. It is worth investigating how Ang-1 increases NOX4 ROS production, through post-translational modification or yet unknown protein-protein interactions that would also contribute to its specific localization.

In Summary, our results highlight the importance of ROS in the regulation of

MAPK and AKT signaling pathways activated by Ang-1 and demonstrate that the magnitude and the balance of MAPK and AKT phosphorylation is largely responsible for the coordinated angiogenic and pro-survival responses that we observed in ECs stimulated with Ang-1.

The major limitation in this study is the absence of identified mechanisms. It would be important to identify how Ang-1/Tie-2 activates NOXs or lead to mtROS release.

Similarly, the mechanism of regulation of the MAPK wasn’t resolved. Investigation should be pursued to identify the mechanism by which NOXs or mtROS affect the level of activation of the MAPK whether direct or indirect. Finally, this study would have been

245 more complete if we had measured the different species of ROS that were generated and their localization upon Ang-1 stimulation. The limitation preventing specific ROS measurement is due to the nature of ROS and their very short half-life. The current probes available might not allow us to confidentially distinguish ROS species or their exact location. Another limitation was the absence of reliable NOX4 antibody. Having a

NOX4 antibody would have allowed us in part to better understand NOX4 involvement in mitochondrial ROS release. It would be important to determine how NOX2 and NOX4 are regulated. We observed that both NOX2 and NOX4 are not transcriptionally regulated by Ang-1, which left us with the hypothesis that they must be post- translationally regulated, through modification or association with molecular partners. In order to verify this assumption, it will be necessary to perform experiments demonstrating how Ang-1/Tie-2 activation leads to the activation of NOX2. NOX2 regulation has been well studied during neutrophil stimulation in which p47phox, p67phox, p40phox, gp91phox and p22phox are phosphorylatedphox . NOX4 activity, too, is potentially regulated but by unknown molecular partners. This could potentially be done by the use of immunoprecipitation and mass spectroscopy to identify which proteins associate with

NOX4 upon Ang-1 stimulation. Through use of live imaging, one could potentially observe where NOX4 is located within the cell and if it is translocated upon Ang-1 stimulation. Reports have indicated that NOX4 phosphorylation is required for its recruitment at membrane scaffold and localized ROS release [391], highlighting the importance of verifying if Ang-1 leads to NOX4 phosphorylation and through which kinase. Mutation of the phosphorylation site on NOX4 would allow confirmation that

Ang-1 induced phosphorylation is required for the recruitment to its site of action.

246

Regulation of angiogenesis by Angiopoietin-1: Roles of miR-640 and ZFP91

There is increasing evidence implicating miRNAs in the control of Angiogenesis.

In this study we were interested in finding miRNAs whose expression are modulated by

Ang-1 signaling in ECs. Data obtained from miRNA affymetrix arrays previously performed in our laboratory showed that Ang-1 was able to induce and or downregulate several miRNAs at 12, 24 and 48 h: emphasizing the idea that Ang-1 regulates the expression levels of miRNAs to carry out its functions. The large majority of Ang-1 modulated miRNAs have unknown functions and targets, opening interesting horizon for the study of miRNAs as downstream effectors of Ang-1 signaling in the vasculature. In this study, we tried to elucidate how survival, migration and differentiation of ECs are influence by Ang-1-downregulated miRNAs.

miR-640 is the sole member of its family and is downregulated in various cancer.

At the time this study was carried, it remained poorly studied. More recently, its role in

H2S-induced angiogenesis through direct targeting of HIF1 has been shown [6187]. In this study we describe a new target of miR-640, ZFP91 and its function in the endothelium. Interestingly we observed that several angiogenic growth factors, namely

Ang-1, VEGF, FGF-2, Ang-2 and TGF, downregulate miR-640, suggesting miR-640 might be a key player in the maintenance of a quiescent endothelium and its downregulation is required to activate angiogenesis. Simultaneously, only Ang-1, Ang-2 and VEGF lead to increase in ZFP91 expression, indicating specificity in miR-640 targeting could be mediated by the initial signal. Indeed, it has been reported that several miRNAs might be required for efficient targeting and we are speculating, that Ang-1,

247

Ang-2 and VEGF, are regulating an array of miRNA that are targeting ZFP91, that is different from the array of miRNAs regulated by FGF-2 or TGF-. Moreover, the specificity of targeting could be due in part to 5’ or 3’ post-biogenesis modification of the miRNAs. 3′ modifications can influence miRNA stability and efficiency of target repression [392]. Reports have shown miRNAs are subject to oxidative modification which regulate their target specificity [271]. Therefore identifying post-biogenesis modification of miR-640 upon stimulation with Ang-1, Ang-2 and VEGF, could contribute to a better understanding of its target specificity. In our study, we did not elucidate the signaling pathways activated downstream of Ang-1 that are responsible for miR-640 downregulation, nor the transcription factors implicated in that process. The promoter region of miR-640 encoded target gene, GATAD2A, contains several transcription factor binding sites for ETS transcription factors. We have established that

ETS factors are important modulators of Ang-1 signaling and it would be relevant to study their role in Ang-1 regulation of miRNAs expression. Interestingly ELK1 and

ETV4, have already been reported to transcriptionally regulate several miRNAs

[301;393]. It is also interesting to note that Ang-1 regulates GATAD2A mRNA levels. It is a novel finding in itself and GATAD2A promoter contains several TFBS for ETS factors including ETV4. Studying the role of GATAD2A in Ang-1 signaling and in relationship to miR-640 would be a promising avenue for better understanding of Ang-

1/miRNA/encoded gene target signaling loops.

In this study, we found that Ang-1 transcriptionally downregulates the expression of miR-640 as a mechanism through which Ang-1 signaling modulates the accumulation of ZFP91 and influences ECs migration and differentiation. Ang-1 increased the levels

248 of ZFP91 via downregulation of miR-640, which results in the typical Ang-1-induced cell migration and tube formation. MiR-640 role in ECs migration and tube formation correlates previous study [324]. ZFP91 role in ECs migration and differentiation is reported here for the first time. Although previous studies hinted its pro-angiogenic function [329-331].

Our study also raises the question as to whether Ang-1-induced ZFP91 contributes to both canonical and non-canonical NFκB pathways and the role of these signaling axis in angiogenesis. The involvement of Ang-1/miR-640/ZFP91 in NFκB signaling also suggest miR-640 and ZFP91 might be involved in inflammation. Ang-1 has been described as an anti-inflammatory agent [6]. Inflammatory activation of ECs leads to recruitment of leukocytes to sites of injury due to an increase in pro- inflammatory mediators and adhesion molecules expressed largely in response to activation of the NFκB signaling pathway [37]. Ang-1 decreases inflammation through the recruitment of the intracellular protein A20 binding inhibitor of NFκB activation 2

(ABIN-2) to Tie-2 receptors which interferes with the activation of the NFκB pathway

[165]. More recently we showed Ang-1 modulates Toll-Like-Receptor-4 (TLR4) signaling and influences innate immune responses. Ang-1 reduces the levels of IRAK1 and TRAF6 in toll like receptor signaling through induction of miR-146b-5p, which lead to decrease leukocyte adhesion [394]. miR-640 and ZFP91 role in Ang-1 anti- inflammatory effect is worth investigating in the near future. Finally, ZFP91 has been reported to exhibit dual function. It can act as a transcription factor or as an E3-ubiquitin ligase. It would be worth investigating if Ang-1 transcriptional regulation of ZFP91 leads to its activation as an E3-ubiquitin ligase, a transcription factor or both.

249

Several studies have suggested Ang-1 or Tie-2 agonist, could have a potential in cancer therapy [395;396]. Our finding suggest a potential role for miR-640 targets as a novel approach in the treatment of aberrant angiogenesis. Targeting molecular players regulating specific angiogenic responses being a better approach than targeting the initiation of the Ang-1/Tie-2 signaling cascade, which will lack specificity of action.

Although we identified a new miRNA regulated by Ang-1 and targeting ZFP91, this study didn’t show how ZFP91 might be involved in Ang-1 induced angiogenesis.

For a more complete approach it would have been worthwhile to identify the signaling pathway regulated by ZFP91 and its relationship to Ang-1. Indeed, we could have performed pulldown for mass spectrometry analysis and identify ZFP91 targets for ubiquitination.

Major limitation in this study is the absence of in vivo data in the determination of miR-640 biological function. miR-640 is not conserved and therefore we couldn’t study its role in angiogenesis in a murine model. Knowing that miRNA functions sometime differs in vivo and in vitro. Thus, evaluating miRNA functions in vitro only is inappropriate. The closer we could have come to study its function in human, would have been to correlate its expression and the expression of its targets in tissues from disease exhibiting aberrant angiogenesis. Functional validation is a limiting step in miRNA studies. MiRNA regulatory networks only complicates this process. Although we characterized the relationship between miR-640 and ZFP91, one could wonder which other miRNAs or which combination of miRNA is associated with the altered migration and tube formation; which other miRNAs or which combination of miRNA target ZFP91.

Moreover, is miR-640 targeting other members of a signaling pathway or regulatory

250 network involved in migration and tube formation? miRNA regulation, is also under the control of other transcriptional and post-transcriptional regulations. We did not identify if miR-640 was subjected to modification altering its target specificity.

ETS1, ELK1 and ETV4 Contribute to Angiopoietin-1 Angiogenic Responses in

Endothelial Cells

Previously an analysis of the Ang-1 transcriptome performed in our lab suggested that Ang-1 induces the expression of immediate-early genes as a mechanism to coordinate pro-angiogenic signaling in ECs downstream of the MAPKs [397]. Ang-1 elicited coordinated responses designed to improve EC survival, proliferation and differentiation and to attenuate apoptosis and inflammation. Ang-1 induced the upregulation of several known positive regulators of angiogenesis including FLT1,

DUSP4, DUSP5 and ANGPTL4. In silico analysis, revealed these genes could be regulated by ETS1, ELK1 and ETV4. We decided to evaluate the importance of these three transcription factors in the biological responses of Ang-1 in ECs. Our results showed that Ang-1 elicits enhanced production of ETS1 as well as enhanced DNA binding activity for ETS1, ELK1 and ETV4. We further showed that ETS1, ELK1 and

ETV4 orchestrate the induction of several Ang-1-induced proangiogenic genes such as

EGR1, DUSP4, ANGPTL4, FLT1.

To evaluate the functional role of ETS factors in the biological effects of Ang-1, we used siRNAs to knock down ETS1, ELK1 and ETV4 expression in HUVECs. Our results showed that Ang-1-induced ETS1, ELK1 and ETV4 play a role in Ang-1-induced

ECs survival and differentiation while only ETS1 and ETV4 are involved in Ang-1

251 induced migration. The mechanisms through which these ETS factors mediates Ang-1 effects are still unclear. The most likely mechanism by which ETS factors enhances

Ang-1-induced survival, migration and differentiation is through the regulation of several proangiogenic genes. In our study, we demonstrated that ETS factors mediates the expression of several proangiogenic genes, including DUSP4 and EGR1. We have previously shown the involvement of DUSP4 and EGR1 in Ang-1-induced angiogenic responses [16;208]. However, future experiments should identify all ETS target genes in

Ang-1-stimulated ECs to help us further understand the ETS-dependent mechanisms behind Ang-1-induced angiogenesis. For that end, one would use the ChlP-on-chip technique whereby one immunoprecipitates DNA sequences in Ang-1-treated HUVECs by using antibodies specific to ETS1, ELK1 or ETV4 then perform DNA microarray analysis on the immunoprecipitates. In addition, it would be interesting to show in vivo whether Ang-1-induced ETS factors plays a role in Ang-1-induced angiogenesis. This could be accomplished by using ETS1, ELK1 or ETV4 knockout mice in an already established in vivo model of vascularization such as in experimentally induced mouse hind limb ischemia. Then, we could check whether ETS factors knock-out mice show less collateral vessels formation in response to intramuscular Ang-1 gene delivery.

In summary, our results highlight the contribution of ETS1, ELK1 and ETV4 as transcriptional regulators of Ang-1/Tie-2 signaling and their importance in Ang-1- induced angiogenic responses.

Limitation in this study, include the identification of Ang-1 induced post- translational modification and how they regulate the activity of ETS1, ELK1 and ETV4.

During the course of this study we used two different antibodies and phospho-antibodies

252 for the various ETS transcription factors and they lacked specificity and didn’t allow us to obtain reliable results. Moreover, ETV4 post-translational modification required for the activation of its transcriptional activity were unavailable at the time. This was also the major impairment for the execution of ChIP assay.

Integration

Although the research on NOXs and mtROS, miR-640 and ZFP91 and the ETSs TF was carried independently, there is a potential relationship between Ang1- induced ROS and miR-640 and ETS transcription factors. Interestingly, miRNA are sensitive to ROS- induced modification and these modifications dictate the target specifity of miRNA

[6296]. Therefore we could speculate that Ang-1 induced ROS could lead to oxidative modification of miR-640 and of other miRNA dictating Ang-1 pro-angiogenic program.

Ang-1 induced ROS might not modulate miRNA specificity by direct oxidative modification, but it could regulate miRNA transcription through regulation of Redox sensitive transcription factors. Ang-1 leads to the activation of several of these Redox sensitive TFs, including AP1, NFκB, Nrf2, and the ETSs. For example, ETS1 is activated by low dose of H2O2 [4658]. Preliminary results showed that Ang-1-induced ROS, including mtROS were also required for ETS-1 DNA binding activity (data not shown).

Therefore, we could speculate that Ang-1 induce ROS could potentially transcriptionaly regulate miRNA, but also other members of the ETS family if they have antioxidant responsive elements in their structure.

We have collected many pieces of the puzzle and they could all be put together to give us an image of how ROS are central in Ang-1 signaling.

253

Clinical Relevance

Translational research is the art of translating current knowledge, in this case in endothelium molecular and cellular physiology, into tools that can be used to cure human diseases. Our contribution to the knowledge of Ang-1 signaling can give important information to translational researchers pursuing the noble goal of curing various disease arising from endothelial dysfunction. Endothelial dysfunction is characterized by a dysregulation in the production of a wide range of factors that regulate 1) vascular tone,

2) cellular adhesion, 3) smooth muscle cell proliferation, and 4) vessel wall inflammation

[398]. Various environmental factors such as smoking, aging, hypercholesterolemia, hypertension, hyperglycemia, and premature atherosclerotic disease are all associated with endothelial dysfunction [399;400]. Altered endothelial function leads to a chronic inflammatory process in addition to abnormal vasoreactivity, which elevate the risk of cardiovascular events [401]. ROS have a central role in normal and abnormal endothelial function and vascular function. ROS contribute to vascular homeostasis by their role in signaling processes. On the other hand, in cardiovascular disease states, the release of

ROS is increased to a level that overcomes the natural antioxidant systems in the cell.

This leads to uncontrolled ROS production which precedes impaired endothelial function and vascular dysfunction Importantly, ROS are generated at sites of inflammation. The vascular endothelium, is situated between the blood and the vascular smooth muscle and is thus a major target of oxidative stress, and plays a critical role in the pathophysiology of several vascular diseases [402].

254

Impaired endothelium-dependant vascular dilation is an early indicator of subsequent negative cardiovascular events. The endothelium controls its state of constriction/dilation by secreting endothelium-derived relaxing factor nitric oxide (NO) and several other vasodilating agents, including prostacyclin and the endothelium-dependent hyperpolarizing factor (EDHF). The endothelium also produces contracting factors, such as endothelin-1 (ET-1) and thromboxane [403]. Superoxide has been classified as a vasoconstrictor agent [403]. NO interacts with O2, forming peroxynitrite (ONOO−).

Under physiological conditions, ROS production is low and the antioxidant systems can maintain the balance of the redox state, thus preventing the conversion of of NO to

ONOO-. In disease states in which the production of ROS is increased or the antioxidant capacity of the vessel is decreased, NO is readily transformed to ONOO−, resulting in inhibition of endothelial-dependent relaxation [404]. Ang-1 has been reported to increase vasoconstriction in the setting of sepsis[405]. Moreover, Ang-1 pro-angiogenic effects were shown to be dependent on eNOS and subsequent NO production. It appears that

NO plays a role in vasodilation as well as in angiogenesis. In light of our observation, we could propose that increased Ang-1 in disease states could lead to increased vasoconstriction, since Ang-1 induces increased ROS production from NOX2, NOX4 and mtROS. This increase could overwhelm a system already coping with oxidative stress and lead to the formation of ONOO-. Interestingly, in physiological conditions,

Ang-1 was reported to lead to vasodilation in a NO dependent manner [406], supporting the idea that in disease states Ang-1 induced ROS might be responsible for the observed vasoconstriction. Our findings are offering new avenues to explore. Although antioxidant therapy has failed in the past, it could be interesting to try to specifically

255 target NOX2, NOX4 and mtROS as the actual sources of ROS. Further studies need to be performed to confirm the role of these ROS sources in the regulation of vascular tone.

This could easily be achieved using intravital microscopy and measuring the vascular tone in knock out mice under physiological and disease states.

Cellular adhesion

Proliferation of vascular smooth muscle cells (VSMCs) is a hallmark of vascular dysfunction. Signaling between endothelial and smooth muscle cells is essential for maintaining tone in mature vessels. A primary feature of atherosclerotic plaques is the transition of smooth muscle cells into a synthetic state in which they become proliferative and secrete excess ECM contributing to plaque buildup [406]. It is widely thought that endothelial cell injury is the trigger for smooth muscle proliferation and ECM synthesis.

More relevant to our studies is the fact that ROS have been shown to promote VSMCs proliferation upon arterial injury. The effect of excess ROS leads to increased proliferation in VSMCs which is mediated through inhibition of NO, an inhibitor of

VSMCs proliferation [406;407]. Moreover it has been demonstrated increased VSMCs proliferation is mediated by ROS activation of ERK1/2e [407]. NOXs also contribute to

VSMCs proliferation and the development of cardiovascular disease [408] and Ang-1 level of expression in VSMCs correlate with their proliferative state [409]. As previously stated, our observations, lead us to believe that increased Ang-1 in disease states could lead to increased VSCMs proliferation, since Ang-1 induces increased ROS production from NOX2, NOX4 and mtROS, leading to the neutralization of NO. NOX2 was shown to cause eNOS uncoupling and activation of xanthine/xanthine oxidase, further increasing ROS production. Our results, particularly the regulation of regulation

256 of AKT pathway and MAPKs by NOXs and mtROS, are promising new potential therapeutic targets. It appears that NOX2 is an important contributor to VSMCs proliferation, but future investigations should try to discriminate the contribution of

NOX4 and mtROS. Moreover, future studies should try to identify the gene targets of the

MAPKs and AKT pathway regulated by the NOXs to confirm which proteins specifically contribute to VSMCs proliferation. It is noteworthy to mention that increased proliferation can be of help when it occurs in an area of tissue injury and repair. That is likely why these pathways exist. It is uncontrolled and dysregulated proliferation that cause harm. It will be important to identify the molecular player in uncontrolled VSMCs proliferation and make sure that therapeutic targeting will not temper with the beneficial effects of VSMCs proliferation.

Adhesion of leukocyte to the endothelium is a hallmark of vascular inflammation.

Endothelial cell activation, which is defined by the endothelial expression of cell-surface adhesion molecules, such as VCAM-1, ICAM-1, and E-selectin, can lead to endothelial dysfunction. Endothelial cell activation is induced by proinflammatory cytokines, such as

TNF-α, and facilitates the recruitment and attachment of circulating leukocytes to the vessel wall. NO is an important modulatorof leukocyte adhesion. Inhibition of eNOS leads to increased leukocyte adhesion, whereas NO donors decrease the expression of

VCAM-1, ICAM-1, E-selection and leukocyte adhesion in response to a variety of proinflammatory cytokines. This means that NO acts as an important anti-inflammatory molecule. The anti-inflammatory effects of NO occur through inhibition of NF-κB. Ang-

1 also has anti-inflammatory effects in the vasculature. Hypercholesterolemia, smoking, or oxidative stress, can enhance endothelial cell activation of adhesion molecules, leading

257 to increased vasoconstriction, smooth muscle proliferation, platelet aggregation, leukocyte adhesion. Superoxide produced by Ang-1 can react with NO to render it biologically inactive and promotes leukocyte adherence, whereas superoxide dismustase

(SOD), a superoxide scavenger limits the inactivation of NO and inhibits leukocyte adhesion. Neutrophils that are firmly adherent to vascular ECs are also activated, which further results in the production and release of ROS, fueling the pool or ROS and increasing inflammation

Ang-1 decreases inflammation by recruiting the intracellular protein ABIN-2 to Tie-2 receptors which interferes with the activation of the NF-κB pathway and inhibits the expression of adhesion molecules, leukocyte adhesion and vascular leakage. In the context of vascular inflammation, Ang-1 appears to have a protective role. It would be interesting to verify whether Ang-1 induced ROS contribute to this protective effect or, to the contrary, do they fueling vascular inflammation and contribute to disease. One approach to investigating this could be to use intravital microscopy to observe leukocyte adherence and extravasation in the microvasculature of mice. The first step would be to verify whether Ang-1 induces or diminishes leukocyte adhesion. Next observe if Ang-1 still exert its effect in NOX knock out animals. In vivo we would have to discriminate between the contribution of vascular NOXs and leukocyte NOXs. Chimeric animals could be used f in which either the vascular NOXs are knockout while the leukocyte

NOXs are functional. Ang-1 appears to contribute in VSMCs to increased proliferation and increased vascular tone, and endothelial dysfunction whereas in the context of vascular inflammation Ang-1 appears to have a protective effect. The experiments suggested above thus should be performed in normal animals as well as in animal models

258 of various vascular disease that comprise the four hallmarks of vascular dysfunction to confirm the overall contribution of Ang-1 and to verify in which direction it tips the balance. Our studies give hints on how to alter one or the other phenotypes. For example, reducing Ang-1 induced ROS might provide benefits to VSMCs proliferation and vascular tone, whereas ZFP91, which regulates VCAM1, might be a target for enhancing

Ang-1 protective effects in vascular inflammation

In summary, oxidative stress increases vascular endothelial permeability and promotes leukocyte adhesion, coupled with alterations in endothelial signal transduction and redox sensitive transcription factors [402] . Therefore understanding the specific roles of NOX- derived ROS and mitochondrial derived ROS and identifying their targets could help develop new therapeutic strategies. Targeting redox sensitive transcription factors or signaling molecule, could prove to be a more specific approach. This approach might not be the most specific since TFs regulate several genes. We also have identified around 20 gene targets for the redox sensitive ETS factors and targeting these genes might prove to be the most specific approach. ETS factors have a role in leukocyte adhesion and endothelial permeability and identifying which genes contribute to these phenotype would increase our pool of potential gene targets.

Dysfunctional endothelium also directly stimulates cancer inflammation and metastasis

[410]. Ang-1/Tie-2 signaling has wide-ranging effects on tumor behavior that includes angiogenesis and, inflammation. Tie-2 receptors are up-regulated in tumor vasculature and play a role in tumor progression. Although in vitro Ang-1 pro-angiogenic effects are undebatable, its effects in vivo are still disputed. Ang1 is expressed in many human

259 breast cancer cell lines but the gene is not expressed accordingly in breast cancer clinical specimens, even though its receptor, Tie2, is abundant in the vasculature of all of these tumours. Also overexpression of Ang1 did not enhance tumoor growth, but instead caused retardation of tumour growth. Ang1 induces stabilization of newly formed blood vessels. This characteristic limits the otherwise continuous angiogenesis in the tumor, and consequently give rise to inhibition of tumor growth. The delineation of the Ang-1/Tie-2 pathways performed in vitro will not provide potential target for cancer but should allow us to identify targets for specific effects, including migration, proliferation, survival, differentiation, leukocyte adhesion and vessel permeability. Ang-1 targeting in therapeutic approach lead to somewhat deceiving effects, mainly due to its regulation of several pathways leading to an array of different phenotypes. For example delivery of Ang-1 stabilized peritubular capillaries in folic acid nephropathy but this was accompanied by profibrotic and inflammatory effects [411]. Our study provides several potential therapeutic target downstream of Ang-1. The miR-640 and ZFP91 are interesting because they specifically alter migration and differentiation, without affecting cell survival. However, miRNA per se are not good potential targets. First, there are currently three approaches for miRNA targeting: miRNA sponges, chemically modified oligonucleotides and inhibitory small molecules. These approaches assume that most commonly human disease is caused by upregulation of miRNAs and therefore the focus has been on inhibiting miRNAs involved in the disease. These inhibitory methods also require the design of vehicles for their efficient delivery in vivo in vivo. Attacking multiple genes relevant to human disease at once is viewed as a powerful ability of therapeutic miRNA mimics. However, this can lead to potential toxicity in normal

260 tissues, especially under conditions in which the therapeutic delivery of miRNA mimics will also lead to an accumulation of exogenous miRNA in normal cells [391]. In light of these facts, miRNA will likley remain a tool to study pathways, until further progress is made. Altering miRNA gene targets is likely a better option.

Our study of the ETS family of TFs confirms they are somewhat redundant in function, but interestingly they have distinct gene targets. The genes we have identified, can further be studied to confirm how they impact the angiogenic paradigm.

High levels of ROS have been detected in several cancer and they contribute to the development of tumors and their progression. Cancer cells also express increased levels of antioxidant protein to balance the levels of intracellular ROS. The challenge here is to deprive cells from ROS-induced tumor promoting events and promote ROS- induced apoptosis. Our observation that NOX2 exert an anti-apoptotic role while

NOX4 exerts a pro-apoptotic role in endothelial cells, might enable us to use this knowledge as a tool to tip the balance in favor of ROS-induced apoptosis in cancer cells. Antioxidants have not proved to be efficient, sometime even contributing to tumor progression and invasiveness. Our study identified the source of Ang-1-induced

ROS and we observed that Ang-1 induces ROS production from 3 different sources.

Interestingly they contribute to the regulation of different phenotypes. NOX2-derived

ROS inhibit migration and differentiation, while NOX4 and mtROS are promoting

Ang-1 induced migration and differentiation. Our next step would be to verify in cancer cells or tumors how we could modulate these phenotype through over expression

261 of NOX2 or inhibition of NOX4 and mtROS. We provided evidence that blunting the expression of specific source of ROS can lead to different phenotype, highlighting the importance to specifically target a source of ROS instead of using general antioxidants.

262

Reference List

[1] K.K. Griendling, R.W. Alexander, Endothelial control of the cardiovascular system: recent advances. FASEB J. 10 (1996) 283-292.

[2] P.M. Vanhoutte, Endothelium and control of vascular function. State of the Art lecture. Hypertension 13 (1989) 658-667.

[3] A.S. Chung, N. Ferrara, Developmental and pathological angiogenesis. Annu.Rev.Cell Dev.Biol. 27 (2011) 563-584.

[4] P. Carmeliet, Angiogenesis in health and disease. Nat.Med. 9 (2003) 653-660.

[5] H. Huang, A. Bhat, G. Woodnutt, R. Lappe, Targeting the ANGPT-TIE2 pathway in malignancy. Nat.Rev.Cancer 10 (2010) 575-585.

[6] N.P. Brindle, P. Saharinen, K. Alitalo, Signaling and functions of angiopoietin-1 in vascular protection. Circ.Res. 98 (2006) 1014-1023.

[7] R. Harfouche, N.A. Malak, R.P. Brandes, A. Karsan, K. Irani, S.N. Hussain. Roles of reactive oxygen species in angiopoietin-1/tie-2 receptor signaling. FASEB J. 19[12], 1728-1730. 2005. Ref Type: Journal (Full)

[8] J.X. Chen, H. Zeng, M.L. Lawrence, T.S. Blackwell, B. Meyrick, Angiopoietin-1- induced Angiogenesis is Modulated by Endothelial NADPH Oxidase. Am.J.Physiol Heart Circ.Physiol 291 (2006) H1563-H1572.

[9] Y.M. Kim, K.E. Kim, G.Y. Koh, Y.S. Ho, K.J. Lee, Hydrogen peroxide produced by angiopoietin-1 mediates angiogenesis. Cancer Res. 66 (2006) 6167-6174.

[10] K. Sako, S. Fukuhara, T. Minami, T. Hamakubo, H. Song, T. Kodama, A. Fukamizu, J.S. Gutkind, G.Y. Koh, N. Mochizuki, Angiopoietin-1 induces Kruppel-like factor 2 expression through a phosphoinositide 3-kinase/AKT- dependent activation of myocyte enhancer factor 2. J.Biol.Chem. 284 (2008) 5592-601.

[11] C. Daly, V. Wong, E. Burova, Y. Wei, S. Zabski, J. Griffiths, K.M. Lai, H.C. Lin, E. Ioffe, G.D. Yancopoulos, J.S. Rudge, Angiopoietin-1 modulates endothelial cell function and gene expression via the transcription factor FKHR (FOXO1). Genes Dev. 18 (2004) 1060-1071.

[12] N.A. Abdel-Malak, C.B. Srikant, A.S. Kristof, S.A. Magder, J.A. Di Battista, S.N.A. Hussain, Angiopoietin-1 promotes endothelial proliferation and migration

263

through AP-1 dependent autocrine production of interleukin-8. Blood 111 (2008) 4145-4154.

[13] H. Song, J.I. Suehiro, Y. Kanki, Y. Kawai, K. Inoue, H. Daida, K. Yano, T. Ohashi, P. Ottegen, W.C. Aird, T. Kodama, T. Minami, A critical role for GATA3 in mediating Tie2 expression and function in large vessel endothelial cells. J.Biol.Chem.2009).

[14] D.P. Hughes, M.B. Marron, N.P. Brindle, The antiinflammatory endothelial tyrosine kinase Tie2 interacts with a novel nuclear factor-kappaB inhibitor ABIN- 2. Circ.Res. 92 (2003) 630-636.

[15] E.I. Korpelainen, M. Kärkkäinen, Y. Gunji, M. Vikkula, K. Alitalo, Endothelial receptor tyrosine kinases activate the STAT signaling pathway: mutant Tie-2 causing venous malformations signals a distinct STAT activation response. Oncogene 18 (1999) 1-8.

[16] N.A. Abdel-Malak, M. Mofarrahi, D. Mayaki, L.M. Khachigian, S.N. Hussain, Early Growth Response-1 Regulates Angiopoietin-1-Induced Endothelial Cell Proliferation, Migration, and Differentiation. Arterioscler.Thromb.Vasc.Biol. 29 (2008) 209-216.

[17] R.A. Christensen, K. Fujikawa, R. Madore, P. Oettgen, L. Varticovski, NERF2, a member of the Ets family of transcription factors, is increased in response to hypoxia and angiopoietin-1: a potential mechanism for Tie2 regulation during hypoxia. J.Cell Biochem. 85 (2002) 505-515.

[18] R. Harfouche, J.P. Gratton, G.D. Yancopoulos, S.N.A. Hussain, Angiopoietin-1 activates both anti- and pro-apoptotic mitogen activated protein kinases. FASEB J. 17 (2003) 1523-1525.

[19] W.C. Aird, Phenotypic heterogeneity of the endothelium: II. Representative vascular beds. Circ.Res. 100 (2007) 174-190.

[20] A. Armulik, A. Abramsson, C. Betsholtz, Endothelial/pericyte interactions. Circ.Res. 97 (2005) 512-523.

[21] G. Sowa, Caveolae, caveolins, cavins, and endothelial cell function: new insights. Front Physiol 2 (2012) 120.

[22] C. Patterson, M.A. Perrella, C.M. Hsieh, M. Yoshizumi, M.E. Lee, E. Haber, Cloning and functional analysis of the promoter for KDR/flk-1, a receptor for vascular endothelial growth factor. J.Biol.Chem. 270 (1995) 23111-23118.

[23] K. Morishita, D.E. Johnson, L.T. Williams, A novel promoter for vascular endothelial growth factor receptor (flt-1) that confers endothelial-specific gene expression. J.Biol.Chem. 270 (1995) 27948-27953.

264

[24] T.M. Schlaeger, Y. Qin, Y. Fujiwara, J. Margram, T.N. Sato, Vascular endothelial cell lineage-specific promoter in transgenic mice. Development 121 (1995) 1089- 1098.

[25] W.C. Aird, N. Jahroudi, H. Weiler-Guettler, H.B. Rayburn, R.D. Rosenberg, Human von Willebrand factor gene sequences target expression to a subpopulation of endothelial cells in transgenic mice. Proc.Natl.Acad.Sci.U.S.A 92 (1995) 4567-4571.

[26] M. Paul, M. Zintz, W. Bocker, M. Dyer, Characterization and functional analysis of the rat endothelin-1 promoter. Hypertension 25 (1995) 683-693.

[27] N. Almendro, T. Bellon, C. Rius, P. Lastres, C. Langa, A. Corbi, C. Bernabeu, Cloning of the human platelet endothelial cell adhesion molecule-1 promoter and its tissue-specific expression. Structural and functional characterization. J.Immunol. 157 (1996) 5411-5421.

[28] D.D. Rees, R.M. Palmer, S. Moncada, Role of endothelium-derived nitric oxide in the regulation of blood pressure. Proc.Natl.Acad.Sci.U.S.A. 86 (1989) 3375-3378.

[29] P. Kubes, D.N. Granger, Nitric oxide modulates microvascular permeability. Am.J.Physiol. 262 (19920) H611-H615.

[30] P.W. Shaul, Regulation of endothelial nitric oxide synthase: location, location, location. Annu.Rev.Physiol 64 (2002) 749-774.

[31] E.J. Filardo, J.A. Quinn, K.I. Bland, A.R. Frackelton, Jr., Estrogen-induced activation of Erk-1 and Erk-2 requires the G protein-coupled receptor homolog, GPR30, and occurs via trans-activation of the epidermal growth factor receptor through release of HB-EGF. Mol.Endocrinol. 14 (2000) 1649-1660.

[32] R.M. Palmer, D.S. Ashton, S. Moncada, Vascular endothelial cells synthesize nitric oxide from L-arginine. Nature 333 (1988) 664-666.

[33] P. Kubes, M. Suzuki, D.N. Granger, Nitric oxide: an endogenous modulator of leukocyte adhesion. Proc.Natl.Acad.Sci.U.S.A 88 (1991) 4651-4655.

[34] Z.S. Katusic, A.V. Santhanam, T. He, Vascular effects of prostacyclin: does activation of PPARdelta play a role? Trends Pharmacol.Sci. 33 (2012) 559-564.

[35] F. Bohm, J. Pernow, The importance of endothelin-1 for vascular dysfunction in cardiovascular disease. Cardiovasc.Res. 76 (2007) 8-18.

[36] T. Minami, A. Sugiyama, S.Q. Wu, R. Abid, T. Kodama, W.C. Aird, Thrombin and phenotypic modulation of the endothelium. Arterioscler.Thromb.Vasc.Biol. 24 (2004) 41-53.

265

[37] J.S. Pober, W.C. Sessa, Evolving functions of endothelial cells in inflammation. Nat.Rev.Immunol. 7 (2007) 803-815.

[38] B.J. Hunt, K.M. Jurd, Endothelial cell activation. A central pathophysiological process. BMJ 316 (1998) 1328-1329.

[39] M.M. Teixeira, T.J. Williams, P.G. Hellewell, Role of prostaglandins and nitric oxide in acute inflammatory reactions in guinea-pig skin. Br.J.Pharmacol. 110 (1993) 1515-1521.

[40] T. Stevens, J.G. Garcia, D.M. Shasby, J. Bhattacharya, A.B. Malik, Mechanisms regulating endothelial cell barrier function. Am.J.Physiol Lung Cell Mol.Physiol 279 (2000) L419-L422.

[41] K.A. Birch, B.M. Ewenstein, D.E. Golan, J.S. Pober, Prolonged peak elevations in cytoplasmic free calcium ions, derived from intracellular stores, correlate with the extent of thrombin-stimulated exocytosis in single human umbilical vein endothelial cells. J.Cell Physiol 160 (1994) 545-554.

[42] K.A. Tanaka, N.S. Key, J.H. Levy, Blood coagulation: hemostasis and thrombin regulation. Anesth.Analg. 108 (2009) 1433-1446.

[43] R.R. Gainetdinov, R.T. Premont, L.M. Bohn, R.J. Lefkowitz, M.G. Caron, Desensitization of G protein-coupled receptors and neuronal functions. Annu.Rev.Neurosci. 27 (2004) 107-144.

[44] L.N. Lopez-Bojorquez, F. Arechavaleta-Velasco, F. Vadillo-Ortega, D. Montes- Sanchez, J.L. Ventura-Gallegos, A. Zentella-Dehesa, NF-kappaB translocation and endothelial cell activation is potentiated by macrophage-released signals co- secreted with TNF-alpha and IL-1beta. Inflamm.Res. 53 (2004) 567-575.

[45] T. Martin, P.M. Cardarelli, G.C. Parry, K.A. Felts, R.R. Cobb, Cytokine induction of monocyte chemoattractant protein-1 gene expression in human endothelial cells depends on the cooperative action of NF-kappa B and AP-1. Eur.J.Immunol. 27 (1997) 1091-1097.

[46] B. Engelhardt, H. Wolburg, Mini-review: Transendothelial migration of leukocytes: through the front door or around the side of the house? Eur.J.Immunol. 34 (2004) 2955-2963.

[47] R. Khurana, M. Simons, J.F. Martin, I.C. Zachary, Role of angiogenesis in cardiovascular disease: a critical appraisal. Circulation 112 (2005) 1813-1824.

[48] I. Flamme, T. Frolich, W. Risau, Molecular mechanisms of vasculogenesis and embryonic angiogenesis. J.Cell Physiol 173 (1997) 206-210.

266

[49] J.D. Coffin, J. Harrison, S. Schwartz, R. Heimark, Angioblast differentiation and morphogenesis of the vascular endothelium in the mouse embryo. Dev.Biol. 148 (1991) 51-62.

[50] J.W. Xiong, Molecular and developmental biology of the hemangioblast. Dev.Dyn. 237 (2008) 1218-1231.

[51] E.M. Conway, D. Collen, P. Carmeliet, Molecular mechanisms of blood vessel growth. Cardiovasc.Res. 49 (2001) 507-521.

[52] H. Kurz, P.H. Burri, V.G. Djonov, Angiogenesis and vascular remodeling by intussusception: from form to function. News Physiol Sci. 18 (2003) 65-70.

[53] H.G. Augustin, G.Y. Koh, G. Thurston, K. Alitalo, Control of vascular morphogenesis and homeostasis through the angiopoietin-Tie system. Nat.Rev.Mol.Cell Biol. 10 (2009) 165-177.

[54] C. Ruhrberg, H. Gerhardt, M. Golding, R. Watson, S. Ioannidou, H. Fujisawa, C. Betsholtz, D.T. Shima, Spatially restricted patterning cues provided by heparin- binding VEGF-A control blood vessel branching morphogenesis. Genes Dev. 16 (2002) 2684-2698.

[55] H. Gerhardt, M. Golding, M. Fruttiger, C. Ruhrberg, A. Lundkvist, A. Abramsson, M. Jeltsch, C. Mitchell, K. Alitalo, D. Shima, C. Betsholtz, VEGF guides angiogenic sprouting utilizing endothelial tip cell filopodia. J.Cell Biol. 161 (2003) 1163-1177.

[56] M. Hellstrom, L.K. Phng, J.J. Hofmann, E. Wallgard, L. Coultas, P. Lindblom, J. Alva, A.K. Nilsson, L. Karlsson, N. Gaiano, K. Yoon, J. Rossant, M.L. Iruela- Arispe, M. Kalen, H. Gerhardt, C. Betsholtz, Dll4 signalling through Notch1 regulates formation of tip cells during angiogenesis. Nature 445 (2007) 776-780.

[57] A.F. Siekmann, N.D. Lawson, Notch signalling limits angiogenic cell behaviour in developing zebrafish arteries. Nature 445 (2007) 781-784.

[58] B. Larrivee, C. Freitas, S. Suchting, I. Brunet, A. Eichmann, Guidance of vascular development: lessons from the nervous system. Circ.Res. 104 (2009) 428-441.

[59] R.K. Jain, Molecular regulation of vessel maturation. Nat.Med. 9 (2003) 685-693.

[60] P. Lindahl, B.R. Johansson, P. Leveen, C. Betsholtz, Pericyte loss and microaneurysm formation in PDGF-B-deficient mice. Science 277 (1997) 242- 245.

[61] Y. Liu, R. Wada, T. Yamashita, Y. Mi, C.X. Deng, J.P. Hobson, H.M. Rosenfeldt, V.E. Nava, S.S. Chae, M.J. Lee, C.H. Liu, T. Hla, S. Spiegel, R.L. Proia, Edg-1, the G protein-coupled receptor for sphingosine-1-phosphate, is essential for vascular maturation. J.Clin.Invest 106 (2000) 951-961.

267

[62] K.K. Hirschi, S.A. Rohovsky, P.A. D'Amore, PDGF, TGF-beta, and heterotypic cell-cell interactions mediate endothelial cell-induced recruitment of 10T1/2 cells and their differentiation to a smooth muscle fate. J.Cell Biol. 141 (1998) 805-814.

[63] D.C. Darland, L.J. Massingham, S.R. Smith, E. Piek, M. Saint-Geniez, P.A. D'Amore, Pericyte production of cell-associated VEGF is differentiation- dependent and is associated with endothelial survival. Dev.Biol. 264 (2003) 275- 288.

[64] C. Suri, P.F. Jone, S. Patan, S. Bartunkova, P.C. Maisonpierre, S. Davis, T. Sato, G.D. Yancopoulos, Requisite role of angiopoietin-1, a ligand for the TIE-2 receptor, during embryonic angiogenesis. Cell 87 (1996) 1171-1180.

[65] E. Iivanainen, L. Nelimarkka, V. Elenius, S.M. Heikkinen, T.T. Junttila, L. Sihombing, M. Sundvall, J.A. Maatta, V.J. Laine, S. Yla-Herttuala, S. Higashiyama, K. Alitalo, K. Elenius, Angiopoietin-regulated recruitment of vascular smooth muscle cells by endothelial-derived heparin binding EGF-like growth factor. FASEB J. 17 (2003) 1609-1621.

[66] H. Kobayashi, L.M. DeBusk, Y.O. Babichev, D.J. Dumont, P.C. Lin, Hepatocyte Growth Factor Mediates Angiopoietin-induced Smooth Muscle Cells Recruitment. Blood 108 (2006) 1260-1266.

[67] N.W. Gale, G. Yancopoulos, Growth factors acting via endothelial cell-specific receptor tyrosine kinases: VEGF, angiopoietins, and ephrins in vascular development. Genes & Development 13 (1999) 1055-1066.

[68] D.J. Dumont, G. Gradwohl, F. Guo-Hua, M.C. Puri, M. Gertsenstein, A. Auerbach, M.L. Breitman, Dominant-negative and targeted null mutations in the endothelial receptor tyrosine kinase,Tek, reveal a critical role in vasculogenesis of the embryo. Genes & Development 8 (1994) 1897-1909.

[69] M. Vikkula, L.M. Boon, K.L. Carraway, III, J.T. Calvert, A.J. Diamonti, B. Goumnerov, K.A. Pasyk, D.A. Marchuk, M.L. Warman, L.C. Cantley, J.B. Mulliken, B.R. Olsen, Vascular dysmorphogenesis caused by an activating mutation in the receptor tyrosine kinase TIE2. Cell 87 (1996) 1181-1190.

[70] V. Wouters, N. Limaye, M. Uebelhoer, A. Irrthum, L.M. Boon, J.B. Mulliken, O. Enjolras, E. Baselga, J. Berg, A. Dompmartin, S.A. Ivarsson, L. Kangesu, Y. Lacassie, J. Murphy, A.S. Teebi, A. Penington, P. Rieu, M. Vikkula, Hereditary cutaneomucosal venous malformations are caused by TIE2 mutations with widely variable hyper-phosphorylating effects. Eur.J.Hum.Genet. 18 (2010) 414-420.

[71] E.M. Gravallese, A.R. Pettit, R. Lee, R. Madore, C. Manning, A. Tsay, J. Gaspar, M.B. Goldring, S.R. Goldring, P. Oettgen, Angiopoietin-1 is expressed in the synovium of patients with rheumatoid arthritis and is induced by tumour necrosis factor alpha. Ann.Rheum.Dis. 62 (2003) 100-107.

268

[72] L. Du, C.C. Sullivan, D. Chu, A.J. Cho, M. Kido, P.L. Wolf, J.X. Yuan, R. Deutsch, S.W. Jamieson, P.A. Thistlethwaite, Signaling molecules in nonfamilial pulmonary hypertension. N.Engl.J.Med. 348 (2003) 500-509.

[73] H.S. Lim, G.Y. Lip, A.D. Blann, Angiopoietin-1 and angiopoietin-2 in diabetes mellitus: relationship to VEGF, glycaemic control, endothelial damage/dysfunction and atherosclerosis. Atherosclerosis 180 (2005) 113-118.

[74] M. Takahama, M. Tsutsumi, T. Tsujiuchi, K. Nezu, K. Kushibe, S. Taniguchi, Y. Kotake, Y. Konishi, Enhanced expression of Tie2, its ligand angiopoietin-1, vascular endothelial growth factor, and CD31 in human non-small cell lung carcinomas. Clin.Cancer Res. 5 (1999) 2506-2510.

[75] W.S. Shim, M. Teh, A. Bapna, I. Kim, G.Y. Koh, P.O. Mack, R. Ge, Angiopoietin 1 promotes tumor angiogenesis and tumor vessel plasticity of human cervical cancer in mice. Exp.Cell Res. 279 (2002) 299-309.

[76] G.J. Caine, A.D. Blann, P.S. Stonelake, P. Ryan, G.Y. Lip, Plasma angiopoietin- 1, angiopoietin-2 and Tie-2 in breast and prostate cancer: a comparison with VEGF and Flt-1. Eur.J.Clin.Invest 33 (2003) 883-890.

[77] P. Lin, J.A. Buxton, A. Acheson, C. Radziejewski, P.C. Maisonpierre, G.D. Yancopoulos, K.M. Channon, L.P. Hale, M.W. Dewhirst, S.E. George, K.G. Peters, Antiangiogenic gene therapy targeting the endothelium-specific receptor tyrosine kinase Tie2. Proc.Natl.Acad.Sci.U.S.A 95 (1998) 8829-8834.

[78] P. Lin, P. Polverini, M. Dewhirst, S. Shan, P.S. Rao, K. Peters, Inhibition of tumor angiogenesis using a soluble receptor establishes a role for Tie2 in pathologic vascular growth. J.Clin.Invest 100 (1997) 2072-2078.

[79] A.H. Wagner, M.R. Schroeter, M. Hecker, 17beta-estradiol inhibition of NADPH oxidase expression in human endothelial cells. FASEB J. 15 (2001) 2121-2130.

[80] K. Kuroda, A. Sapadin, T. Shoji, R. Fleischmajer, M. Lebwohl, Altered expression of angiopoietins and Tie2 endothelium receptor in psoriasis. J.Invest Dermatol. 116 (2001) 713-720.

[81] C. Lemieux, R. Maliba, J. Favier, J.F. Theoret, Y. Merhi, M.G. Sirois, Angiopoietins can directly activate endothelial cells and neutrophils to promote proinflammatory responses. Blood 105 (2005) 1523-1530.

[82] G. Thurston, C. Suri, K. Smith, J. McClain, T.N. Sato, G.D. Yancopoulos, D.M. McDonald, Leakage-resistant blood vessels in mice transgenically overexpressing angiopoietin-1. Science 286 (1999) 2511-2514.

[83] I. Kim, S.O. Moon, S.K. Park, S.W. Chae, G.Y. Koh, Angiopoietin-1 reduces VEGF-stimulated leukocyte adhesion to endothelial cells by reducing ICAM-1, VCAM-1, and E-selectin expression. Circ.Res. 89 (2001) 477-479.

269

[84] J.R. Gamble, J. Drew, L. Trezise, A. Underwood, M. Parsons, L. Kasminkas, J. Rudge, G. Yancopoulos, M.A. Vadas, Angiopoietin-1 is an antipermeability and anti-inflammatory agent in vitro and targets cell junctions. Circ.Res. 87 (2000) 603-607.

[85] I. Kim, J.L. Oh, Y.S. Ryu, J.N. So, W.C. Sessa, K. Walsh, G.Y. Koh, Angiopoietin-1 negatively regulates expression and activity of tissue factor in endothelial cells. FASEB J. 16 (2002) 126-128.

[86] A.M. Joussen, V. Poulaki, A. Tsujikawa, W. Qin, T. Qaum, Q. Xu, Y. Moromizato, S.E. Bursell, S.J. Wiegand, J. Rudge, E. Ioffe, G.D. Yancopoulos, A.P. Adamis, Suppression of diabetic retinopathy with angiopoietin-1. Am.j.Pathol. 160 (2002) 1683-1693.

[87] B. Witzenbichler, D. Westermann, S. Knueppel, H.P. Schultheiss, C. Tschope, Protective role of angiopoietin-1 in endotoxic shock. Circulation 111 (2005) 97- 105.

[88] D.C. Simoes, T. Vassilakopoulos, D. Toumpanakis, K. Petrochilou, C. Roussos, A. Papapetropoulos, Angiopoietin-1 protects against airway inflammation and hyperreactivity in asthma. Am.J.Respir.Crit Care Med. 177 (2008) 1314-1321.

[89] B. Hu, M.J. Jarzynka, P. Guo, Y. Imanishi, D.D. Schlaepfer, S.Y. Cheng, Angiopoietin 2 induces glioma cell invasion by stimulating matrix metalloprotease 2 expression through the alphavbeta1 integrin and focal adhesion kinase signaling pathway. Cancer Res. 66 (2006) 775-783.

[90] N.M. Novotny, T. Lahm, T.A. Markel, P.R. Crisostomo, M. Wang, Y. Wang, J. Tan, D.R. Meldrum, Angiopoietin-1 in the treatment of ischemia and sepsis. Shock 31 (2009) 335-341.

[91] K. Takahashi, Y. Ito, M. Morikawa, M. Kobune, J. Huang, M. Tsukamoto, K. Sasaki, K. Nakamura, H. Dehari, K. Ikeda, H. Uchida, S. Hirai, T. Abe, H. Hamada, Adenoviral-delivered angiopoietin-1 reduces the infarction and attenuates the progression of cardiac dysfunction in the rat model of acute myocardial infarction. Mol.Ther. 8 (2003) 584-592.

[92] T.N. Sato, Y. Tozawa, U. Deutsch, K. Wolburg-Buchholz, Y. Fujiwara, M. Gendron-Maguire, T. Gridley, H. Wolburg, W. Risau, Y. Qin, Distinct roles of the receptor tyrosine kinases tie-1 and Tie-2 in blood vessel formation. Nature 376 (1995) 70-74.

[93] T.M. Schlaeger, S. Bartunkova, J.A. Lawitts, G. Teichmann, W. Risau, U. Deutsch, T.N. Sato, Uniform vascular-endothelial-cell-specific gene expression in both embryonic and adult transgenic mice. Proc.Natl.Acad.Sci. 94 (1997) 3058- 3063.

270

[94] M. Jeansson, A. Gawlik, G. Anderson, C. Li, D. Kerjaschki, M. Henkelman, S.E. Quaggin, Angiopoietin-1 is essential in mouse vasculature during development and in response to injury. J.Clin.Invest 121 (2011) 2278-2289.

[95] P.C. Maisonpierre, C. Suri, P.F. Jones, S. Bartunkova, S.J. Wiegand, C. Radziejewski, D. Compton, J. McClain, T.H. Aldrich, N. Papadopoulos, T.J. Daly, S. Davis, T. Sato, G.D. Yancopoulos, Angiopoietin-2, a natural antagonist for Tie-2 that disrupt in vivo angiogenesis. Science 277 (1997) 55-62.

[96] N.W. Gale, G. Thurston, S.F. Hackett, R. Renard, Q. Wang, J. McClain, C. Martin, C. Witte, M.H. Witte, D. Jackson, C. Suri, P.A. Campochiaro, S.J. Wiegand, G.D. Yancopoulos, Angiopoietin-2 is required for postnatal angiogenesis and lymphatic patterning, and only the latter role is rescued by Angiopoietin-1. Dev.Cell 3 (2002) 411-423.

[97] C. Suri, J. McClain, G. Thurston, D.M. McDonald, H. Zhou, E.H. Oldmixon, T.N. Sato, G.D. Yancopoulos, Increased vascularization in mice overexpressing angiopoietin-1. Science 282 (1998) 468-471.

[98] N.L. Ward, A.L. Haninec, S.P. Van, J.G. Sled, C. Sturk, R.M. Henkelman, I.R. Wanless, D.J. Dumont, Angiopoietin-1 causes reversible degradation of the portal microcirculation in mice: implications for treatment of liver disease. Am.j.Pathol. 165 (2004) 889-899.

[99] K.G. Shyu, O. Manor, M. Magner, G.D. Yancopoulos, J.M. Isner, Direct intramuscular injection of plasmid DNA encoding angiopoietin-1 but not angiopoietin-2 augments revascularization in the rabbit ischemic hindlimb. Circulation 98 (1998) 2081-2087.

[100] C.H. Cho, H.K. Sung, K.T. Kim, H.G. Cheon, G.T. Oh, H.J. Hong, O.J. Yoo, G.Y. Koh, COMP-angiopoietin-1 promotes wound healing through enhanced angiogenesis, lymphangiogenesis, and blood flow in a diabetic mouse model. Proc.Natl.Acad.Sci.U.S.A 103 (2006) 4946-4951.

[101] C.H. Cho, K.E. Kim, J. Byun, H.S. Jang, D.K. Kim, P. Baluk, F. Baffert, G.M. Lee, N. Mochizuki, J. Kim, B.H. Jeon, D.M. McDonald, G.Y. Koh, Long-term and sustained COMP-Ang1 induces long-lasting vascular enlargement and enhanced blood flow. Circ.Res. 97 (2005) 86-94.

[102] J.K. Chae, I. Kim, S.T. Lim, M.J. Chung, W.H. Kim, H.G. Kim, J.K. Ko, G.Y. Koh, Coadministration of angiopoietin-1 and vascular endothelial growth factor enhances collateral vascularization. Arterioscler.Thromb.Vasc.Biol. 20 (2000) 2573-2578.

[103] C.H. Cho, R.A. Kammerer, H.J. Lee, M.O. Steinmetz, Y.S. Ryu, S.H. Lee, K. Yasunaga, K.T. Kim, I. Kim, H.H. Choi, W. Kim, S.H. Kim, S.K. Park, G.M. Lee, G.Y. Koh, COMP-Ang1: a designed angiopoietin-1 variant with nonleaky angiogenic activity. Proc.Natl.Acad.Sci.U.S.A 101 (2004) 5547-5552. 271

[104] F. Maltais, J. Jobin, M.J. Sullivan, S. Bernard, F. Whittom, K.J. Killian, M. Desmeules, M. Belanger, P. LeBlanc, Metabolic and hemodynamic responses of lower limb during exercise in patients with COPD. J.Appl.Physiol 84 (1998) 1573-1580.

[105] A.J. Hayes, W.Q. Huang, J. Yu, P.C. Maisonpierre, A. Liu, F.G. Kern, M.E. Lippman, S.W. McLeskey, L.Y. Li, Expression and function of angiopoietin-1 in breast cancer. Br.J.Cancer 83 (2000) 1154-1160.

[106] W.S. Shim, M. Teh, A. Bapna, I. Kim, G.Y. Koh, P.O. Mack, R. Ge, Angiopoietin 1 promotes tumor angiogenesis and tumor vessel plasticity of human cervical cancer in mice. Exp.Cell Res. 279 (2002) 299-309.

[107] H.S. Lim, A.D. Blann, A.Y. Chong, B. Freestone, G.Y. Lip, Plasma vascular endothelial growth factor, angiopoietin-1, and angiopoietin-2 in diabetes: implications for cardiovascular risk and effects of multifactorial intervention. Diabetes Care 27 (2004) 2918-2924.

[108] R. Harfouche, H.M. Hassessian, Y. Guo, V. Faivre, C.B. Srikant, G.D. Yancopoulos, S.N.A. Hussain, Mechanisms which mediate the anti-apoptotic effects of angiopoietin-1 on endothelial cells. Microvasc.Res. 64 (2002) 135-147.

[109] I. Kim, H.G. Kim, J.N. So, J.H. Kim, H.J. Kwak, Y. Koh, Angiopoietin-1 regulates endothelial cell survival through the phosphatidylinsitol 3'-kinase/AKT signal transduction pathway. Cir.Res. 86 (2000) 24-29.

[110] A. Papapetropoulos, D. Fulton, K. Mahboubi, R.G. Kalb, D.S. O'Connor, F. Li, D.C. Altieri, W.C. Sessa, Angiopoietin-1 inhibits endothelial cell apoptosis via the Akt/Survivin pathway. J.Biol Chem. 275 (2000) 9102-9105.

[111] B. Witzenbichler, P.C. Maisonpierre, P. Jones, G.D. Yancopoulos, J.M. Isner, Chemotactic properities of angiopoietin-1 and -2, ligands for the endothelial- specific tyrosine kinase Tie2. J.Biol Chem. 273 (1998) 18514-18521.

[112] H. Sun, R. Lesche, D.M. Li, J. Liliental, H. Zhang, N. Gavrilova, B. Mueller, X. Liu, H. Wu, PTEN modulates cell cycle progression and cell survival by regulating phosphatidylinositol 3, 4, 5, -trisphosphate and Akt/ protein kinase B signaling pathway. Proc.Natl.Acad.Sci. 96 (1999) 6199-6204.

[113] J.X. Chen, M.L. Lawrence, G. Cunningham, B.W. Christman, B. Meyrick, HSP90 and Akt modulate Ang-1-induced angiogenesis via NO in coronary artery endothelium. J.Appl.Physiol 96 (2004) 612-620.

[114] I. Cascone, E. Audero, E. Giraudo, L. Napione, F. Maniero, M.R. Philips, J.G. Collard, G. Serini, F. Bussolino, Tie-2-dependent activation of RhoA and Rac1 participates in endothelial cell motility triggered by angiopoietin-1. Blood 102 (2003) 2482-2490.

272

[115] E. Audero, I. Cascone, F. Maniero, L. Napione, M. Arese, L. Lanfrancone, F. Bussolino, Adaptor ShcA protein binds tyrosine kinase Tie2 receptor and regulates migration and sprouting but not survival of endothelial cells. J.Biol.Chem. 279 (2004) 13224-13233.

[116] T.I. Koblizek, C. Weiss, G.D. Yancopoulos, Deutsch, Risau, Angiopoietin-1 induces sprouting angiogenisis in vitro. Curr.Biol. 8 (1998) 529-532.

[117] I. Kim, H.G. Kim, S.O. Moon, S.W. Chae, J.N. So, K.N. Koh, B.C. Ahn, G.Y. Koh, Angiopoietin-1 induces endothelial cell sprouting through the activation of focal adhesion kinase and plasmin secretion. Circ.Res. 86 (2000) 952-959.

[118] S. Kanda, Y. Miyata, Y. Mochizuki, T. Matsuyama, H. Kanetake, Angiopoietin 1 is mitogenic for cultured endothelial cells. Cancer Res. 65 (2005) 6820-6827.

[119] P. Saharinen, L. Eklund, J. Miettinen, R. Wirkkala, A. Anisimov, M. Winderlich, A. Nottebaum, D. Vestweber, U. Deutsch, G.Y. Koh, B.R. Olsen, K. Alitalo, Angiopoietins assemble distinct Tie2 signalling complexes in endothelial cell-cell and cell-matrix contacts. Nat.Cell Biol. 10 (2008) 527-537.

[120] S. Fukuhara, K. Sako, T. Minami, K. Noda, H.Z. Kim, T. Kodama, M. Shibuya, N. Takakura, G.Y. Koh, N. Mochizuki, Differential function of Tie2 at cell-cell contacts and cell-substratum contacts regulated by angiopoietin-1. Nat.Cell Biol. 10 (2008) 513-526.

[121] S. Baumer, L. Keller, A. Holtmann, R. Funke, B. August, A. Gamp, H. Wolburg, K. Wolburg-Buchholz, U. Deutsch, D. Vestweber, Vascular endothelial cell- specific phosphotyrosine phosphatase (VE-PTP) activity is required for blood vessel development. Blood 107 (2006) 4754-4762.

[122] M. Winderlich, L. Keller, G. Cagna, A. Broermann, O. Kamenyeva, F. Kiefer, U. Deutsch, A.F. Nottebaum, D. Vestweber, VE-PTP controls blood vessel development by balancing Tie-2 activity. J.Cell Biol. 185 (2009) 657-671.

[123] J. Partanen, E. Armstrong, T.P. Makela, J. Korhonen, M. Sandberg, R. Renkonen, S. Knuutila, K. Huebner, K. Alitalo, A novel endothelial cell surface receptor tyrosine kinase with extracellular epidermal growth factor homology domains. Mol.Cell.Biol. 12 (1992) 1698-1707.

[124] D.J. Dumont, T.P. Yamaguchi, R.A. Conlon, J. Rossant, M.L. Breitman, Tek, a novel tyrosine kinase gene located on mouse chromose 4, is expressed in endothelial cells and their presumptive precursors. Oncogene 7 (1992) 1471-1480.

[125] A. Iwama, I. Hamaguchi, M. Hashiyama, Y. Murayama, K. Yasunaga, T. Suda, Molecular cloning and characterization of mouse tie and tek receptor tyrosine kinase genes and their expression in hematopoietic stem cells. Biochem.Biophys.Res.Commun. 195 (1993) 301-309.

273

[126] O.H. Lee, J. Xu, J. Fueyo, G.N. Fuller, K.D. Aldape, M.M. Alonso, Y. Piao, T.J. Liu, F.F. Lang, B.N. Bekele, C. Gomez-Manzano, Expression of the receptor tyrosine kinase Tie2 in neoplastic glial cells is associated with integrin beta1- dependent adhesion to the extracellular matrix. Mol.Cancer Res. 4 (2006) 915- 926.

[127] D.J. Dumont, G.L. Gradwohl, G.H. Fong, R. Auerbach, M.L. Breitman, The endothelial-specific recptor tyrosine kinase, Tek, is a member of a new subfamily of receptors. Oncogene 8 (1993) 1293-1301.

[128] P. Saharinen, K. Kerkela, N. Ekman, M. Marron, N. Brindle, G.M. Lee, H. Augustin, G.Y. Koh, K. Alitalo, Multiple angiopoietin recombinant proteins activate the Tie1 receptor tyrosine kinase and promote its interaction with Tie2. J.Cell Biol. 169 (2005) 239-243.

[129] C.D. Kontos, E.H. Cha, J.D. York, K.G. Peters, The endothelial receptor tyrosine kinase Tie1 activates phosphatidylinositol 3-kinase and Akt to inhibit apoptosis. Mol.Cell Biol. 22 (2002) 1704-1713.

[130] R.A. Frost, G.J. Nystrom, L.S. Jefferson, C.H. Lang, Hormone, cytokine, and nutritional regulation of sepsis-induced increases in atrogin-1 and MuRF1 in skeletal muscle. Am.J.Physiol Endocrinol.Metab 292 (2007) E501-E512.

[131] K.L. Kim, I.S. Shin, J.M. Kim, J.H. Choi, J. Byun, E.S. Jeon, W. Suh, D.K. Kim, Interaction between Tie receptors modulates angiogenic activity of angiopoietin2 in endothelial progenitor cells. Cardiovasc.Res. 72 (2006) 394-402.

[132] T.C. Seegar, B. Eller, D. Tzvetkova-Robev, M.V. Kolev, S.C. Henderson, D.B. Nikolov, W.A. Barton, Tie1-Tie2 interactions mediate functional differences between angiopoietin ligands. Mol.Cell 37 (2010) 643-655.

[133] D.M. Valenzuela, J.A. Griffiths, J. Rojas, T.H. Aldrich, P.F. Jones, H. Zhou, J. McClain, N.G. Copeland, D.J. Gilbert, N.A. Jenkins, T. Huang, N. Papadopoulos, P.C. Maisonpierre, S. Davis, G.D. Yancopoulos, Angiopoietin-3 and 4: Diverging gene counterparts in mice and humans. Proc.Natl.Acad.Sci.U.S.A. 96 (1999) 1904-1909.

[134] U. Fiedler, T. Krissl, S. Koidl, C. Weiss, T. Koblizek, U. Deutsch, G. Martiny- Baron, D. Marme, H.G. Augustin, Angiopoietin-1 and angiopoietin-2 share the same binding domains in the Tie-2 receptor involving the first Ig-like loop and the epidermal growth factor-like repeats. J.Biol.Chem. 278 (2003) 1721-1727.

[135] W.A. Barton, D. Tzvetkova-Robev, E.P. Miranda, M.V. Kolev, K.R. Rajashankar, J.P. Himanen, D.B. Nikolov, Crystal structures of the Tie2 receptor ectodomain and the angiopoietin-2-Tie2 complex. Nat.Struct.Mol.Biol. 13 (2006) 524-532.

274

[136] H.T. Yuan, E.V. Khankin, S.A. Karumanchi, S.M. Parikh, Angiopoietin 2 is a partial agonist/antagonist of Tie2 signaling in the endothelium. Mol.Cell Biol. 29 (2009) 2011-2022.

[137] H.Z. Kim, K. Jung, H.M. Kim, Y. Cheng, G.Y. Koh, A designed angiopoietin-2 variant, pentameric COMP-Ang2, strongly activates Tie2 receptor and stimulates angiogenesis. Biochim.Biophys.Acta 1793 (2009) 772-780.

[138] I. Kim, J.H. Kim, S.-O. Moon, H.J. Kwak, N.G. Kim, G.Y. Koh, Angiopoietin-2 at high concentration can enhance endothelial cell survival through the phosphatidylinsitol 3'-kinase/AKT signal transduction pathway. Oncogene 19 (2000) 4549-4552.

[139] K.T. Kim, H.H. Choi, M.O. Steinmetz, B. Maco, R.A. Kammerer, S.Y. Ahn, H.Z. Kim, G.M. Lee, G.Y. Koh, Oligomerization and multimerization are critical for angiopoietin-1 to bind and phosphorylate Tie2. J.Biol.Chem. 280 (2005) 20126- 20131.

[140] S. Davis, T.H. Aldrich, P.F. Jone, A. Acheson, D. Compton, V. Jain, T.E. Ryan, J. Bruno, C. Radziejewski, P.C. Maisonpierre, G.D. Yancopoulos, Isolation of angiopioetin-1, a ligand for the tie2 receptor, by secretion-trap expression cloning. Cell 87 (1996) 1161-1169.

[141] R. Lucas, L. Holmgren, I. Garcia, B. Jim en ez, S.J. Mandriota, F. Borlat, B.K.L. Sim, Z. Wu, G.E. Grau, Y. Shing, G.A. Soff, N. Bouck, M.S. Pepper, Multiple forms of angiostatin induce apoptosis in endothelial cells. Blood 92 (1998) 4730- 4741.

[142] W.N. Procopio, P.I. Pelavin, W.M. Lee, N.M. Yeilding, Angiopoietin-1 and -2 coiled coil domains mediate distinct homo-oligomerization patterns, but fibrinogen-like domains mediate ligand activity. J.Biol Chem. 274 (1999) 30196- 30201.

[143] S. Davis, N. Papadopoulos, T.H. Aldrich, P.C. Maisonpierre, T. Huang, L. Kovac, A. Xu, R. Leidich, E. Radziejewska, A. Rafique, J. Goldberg, V. Jain, K. Bailey, M. Karow, J. Fandl, S.J. Samuelsson, E. Ioffe, J.S. Rudge, T.J. Daly, C. Radziejewski, G.D. Yancopoulos, Angiopoietins have distinct modular domains essential for receptor binding, dimerization and superclustering. Nat.Struct.Biol. 10 (2003) 38-44.

[144] Y. Reiss, J. Droste, M. Heil, S. Tribulova, M.H. Schmidt, W. Schaper, D.J. Dumont, K.H. Plate, Angiopoietin-2 impairs revascularization after limb ischemia. Circ.Res. 101 (2007) 88-96.

[145] H.P. Hammes, J. Lin, P. Wagner, Y. Feng, H.F. Vom, T. Krzizok, O. Renner, G. Breier, M. Brownlee, U. Deutsch, Angiopoietin-2 causes pericyte dropout in the normal retina: evidence for involvement in diabetic retinopathy. Diabetes 53 (2004) 1104-1110. 275

[146] I.B. Lobov, P.C. Brooks, R.A. Lang, Angiopoietin-2 displays VEGF-dependent modulation of capillary structure and endothelial cell survival in vivo. Proc.Natl.Acad.Sci.U.S.A 99 (2002) 11205-11210.

[147] Y. Oshima, T. Deering, S. Oshima, H. Nambu, P.S. Reddy, M. Kaleko, S. Connelly, S.F. Hackett, P.A. Campochiaro, Angiopoietin-2 enhances retinal vessel sensitivity to vascular endothelial growth factor. J.Cell Physiol 199 (2004) 412-417.

[148] Y. Oshima, S. Oshima, H. Nambu, S. Kachi, K. Takahashi, N. Umeda, J. Shen, A. Dong, R.S. Apte, E. Duh, S.F. Hackett, G. Okoye, K. Ishibashi, J. Handa, M. Melia, S. Wiegand, G. Yancopoulos, D.J. Zack, P.A. Campochiaro, Different effects of angiopoietin-2 in different vascular beds: new vessels are most sensitive. FASEB J. 19 (2005) 963-965.

[149] U. Fiedler, Y. Reiss, M. Scharpfenecker, V. Grunow, S. Koidl, G. Thurston, N.W. Gale, M. Witzenrath, S. Rosseau, N. Suttorp, A. Sobke, M. Herrmann, K.T. Preissner, P. Vajkoczy, H.G. Augustin, Angiopoietin-2 sensitizes endothelial cells to TNF-alpha and has a crucial role in the induction of inflammation. Nat.Med. 12 (2006) 235-239.

[150] H.J. Lee, C.H. Cho, S.J. Hwang, H.H. Choi, K.T. Kim, S.Y. Ahn, J.H. Kim, J.L. Oh, G.M. Lee, G.Y. Koh, Biological characterization of angiopoietin-3 and angiopoietin-4. FASEB J. 18 (2004) 1200-1208.

[151] M. Yamakawa, L.X. Liu, T. Date, A.J. Belanger, K.A. Vincent, G.Y. Akita, T. Kuriyama, S.H. Cheng, R.J. Gregory, C. Jiang, Hypoxia-inducible factor-1 mediates activation of cultured vascular endothelial cells by inducing multiple angiogenic factors 7. Circ.Res. 93 (2003) 664-673.

[152] Y. Xu, Y.J. Liu, Q. Yu, Angiopoietin-3 is tethered on the cell surface via heparan sulfate proteoglycans. J.Biol.Chem. 279 (2004) 41179-41188.

[153] L.E. Rameh, L.C. Cantley, The role of phosphoinositide 3-kinase lipid products in cell function. J.Biol.Chem. 274 (1999) 8347-8350.

[154] J. Domin, M.D. Waterfield, Using structure to define the function of phosphoinositide 3-kinase family members. FEBS Lett. 410 (1997) 91-95.

[155] T. Fukao, S. Koyasu, PI3K and negative regulation of TLR signaling. Trends Immunol. 24 (2003) 358-363.

[156] J.M. Kavran, D.E. Klein, A. Lee, M. Falasca, S.J. Isakoff, E.Y. Skolnik, M.A. Lemmon, Specificity and promiscuity in phosphoinositide binding by pleckstrin homology domains. J.Biol.Chem. 273 (1998) 30497-30508.

276

[157] P.T. Hawkins, T.R. Jackson, L.R. Stephens, Platelet-derived growth factor stimulates synthesis of PtdIns(3,4,5)P3 by activating a PtdIns(4,5)P2 3-OH kinase. Nature 358 (1992) 157-159.

[158] L.R. Rohrschneider, J.F. Fuller, I. Wolf, Y. Liu, D.M. Lucas, Structure, function, and biology of SHIP proteins. Genes Dev. 14 (2000) 505-520.

[159] N. Pullen, P.B. Dennis, M. Andjelkovic, A. Dufner, S.C. Kozma, B.A. Hemmings, G. Thomas, Phosphorylation and activation of p70s6k by PDK1. Science 279 (1998) 707-710.

[160] D.D. Sarbassov, S.M. Ali, D.M. Sabatini, Growing roles for the mTOR pathway. Curr.Opin.Cell Biol. 17 (2005) 596-603.

[161] K. Fujikawa, I. De Aos Scherpenseel, S.K. Jain, E. Presman, L. Varticovski, Role of Pl 3-Kinase in angiopoietin-1-mediated migration and attachment-dependent survival of endothelial cells. Exp.Cell Res. 253 (1999) 663-672.

[162] R.J. Dekker, R.A. Boon, M.G. Rondaij, A. Kragt, O.L. Volger, Y.W. Elderkamp, J.C. Meijers, J. Voorberg, H. Pannekoek, A.J. Horrevoets, KLF2 provokes a gene expression pattern that establishes functional quiescent differentiation of the endothelium. Blood 107 (2006) 4354-4363.

[163] S. SenBanerjee, Z. Lin, G.B. Atkins, D.M. Greif, R.M. Rao, A. Kumar, M.W. Feinberg, Z. Chen, D.I. Simon, F.W. Luscinskas, T.M. Michel, M.A. Gimbrone, Jr., G. Garcia-Cardena, M.K. Jain, KLF2 Is a novel transcriptional regulator of endothelial proinflammatory activation. J.Exp.Med. 199 (2004) 1305-1315.

[164] Z. Lin, A. Kumar, S. SenBanerjee, K. Staniszewski, K. Parmar, D.E. Vaughan, M.A. Gimbrone, Jr., V. Balasubramanian, G. Garcia-Cardena, M.K. Jain, Kruppel-like factor 2 (KLF2) regulates endothelial thrombotic function. Circ.Res. 96 (2005) e48-e57.

[165] A. Tadros, D.P. Hughes, B.J. Dunmore, N.P. Brindle, ABIN-2 protects endothelial cells from death and has a role in the antiapoptotic effect of angiopoietin-1. Blood 102 (2003) 4407-4409.

[166] I. Kim, S.O. Moon, S.H. Kim, H.J. Kim, Y.S. Koh, G.Y. Koh, Vascular endothelial growth factor expression of intercellular adhesion molecule 1 (ICAM- 1), vascular cell adhesion molecule 1 (VCAM-1), and E-selectin through nuclear factor-kappa B activation in endothelial cells 13. J.Biol.Chem. 276 (2001) 7614-7620.

[167] J. Gavard, V. Patel, J.S. Gutkind, Angiopoietin-1 prevents VEGF-induced endothelial permeability by sequestering Src through mDia. Dev.Cell 14 (2008) 25-36.

277

[168] M. Oubaha, J.P. Gratton, Phosphorylation of endothelial nitric oxide synthase by atypical PKC zeta contributes to angiopoietin-1-dependent inhibition of VEGF- induced endothelial permeability in vitro. Blood 114 (2009) 3343-3351.

[169] H. Ismail, M. Mofarrahi, R. Echavarria, S. Harel, E. Verdin, H.W. Lim, Z.G. Jin, J. Sun, H. Zeng, S.N. Hussain, Angiopoietin-1 and vascular endothelial growth factor regulation of leukocyte adhesion to endothelial cells: role of nuclear receptor-77. Arterioscler.Thromb.Vasc.Biol. 32 (2012) 1707-1716.

[170] J. Martinez-Gonzalez, L. Badimon, The NR4A subfamily of nuclear receptors: new early genes regulated by growth factors in vascular cells. Cardiovasc.Res. 65 (2005) 609-618.

[171] H. Harant, I.J. Lindley, Negative cross-talk between the human orphan nuclear receptor Nur77/NAK-1/TR3 and nuclear factor-kappaB. Nucleic Acids Res. 32 (2004) 5280-5290.

[172] B. You, Y.Y. Jiang, S. Chen, G. Yan, J. Sun, The orphan nuclear receptor Nur77 suppresses endothelial cell activation through induction of IkappaBalpha expression. Circ.Res. 104 (2009) 742-749.

[173] S. Wang, X. Li, M. Parra, E. Verdin, R. Bassel-Duby, E.N. Olson, Control of endothelial cell proliferation and migration by VEGF signaling to histone deacetylase 7. Proc.Natl.Acad.Sci.U.S.A 105 (2008) 7738-7743.

[174] L. Chang, M. Karin, Mammalian MAP kinase signalling cascades. Nature 410 (2001) 37-40.

[175] G.L. Johnson, R. Lapadat, Mitogen-activated protein kinase pathways mediated by ERK, JNK, and p38 protein kinases. Science 298 (2002) 1911-1912.

[176] I. Dan, N.M. Watanabe, A. Kusumi, The Ste20 group kinases as regulators of MAP kinase cascades. Trends Cell Biol. 11 (2001) 220-230.

[177] J. Pouyssegur, V. Volmat, P. Lenormand, Fidelity and spatio-temporal control in MAP kinase (ERKs) signalling. Biochem.Pharmacol. 64 (2002) 755-763.

[178] T. Matozaki, H. Nakanishi, Y. Takai, Small G-protein networks: their crosstalk and signal cascades. Cell Signal. 12 (2000) 515-524.

[179] S.M. Keyse, Protein phosphatases and the regulation of MAP kinase activity. Semin.Cell Dev.Biol. 9 (1998) 143-152.

[180] T.G. Boulton, S.H. Nye, D.J. Robbins, N.Y. Ip, E. Radziejewska, S.D. Morgenbesser, R.A. DePinho, N. Panayotatos, M.H. Cobb, G.D. Yancopoulos, ERKs: a family of protein-serine/threonine kinases that are activated and tyrosine phosphorylated in response to insulin and NGF. Cell 65 (1991) 663-675.

278

[181] M. Cargnello, P.P. Roux, Activation and function of the MAPKs and their substrates, the MAPK-activated protein kinases. Microbiol.Mol.Biol.Rev. 75 (2011) 50-83.

[182] R.H. Chen, C. Sarnecki, J. Blenis, Nuclear localization and regulation of erk- and rsk-encoded protein kinases. Mol.Cell Biol. 12 (1992) 915-927.

[183] S. Yoon, R. Seger, The extracellular signal-regulated kinase: multiple substrates regulate diverse cellular functions. Growth Factors 24 (2006) 21-44.

[184] A. Cuadrado, A.R. Nebreda, Mechanisms and functions of p38 MAPK signalling. Biochem.J. 429 (2010) 403-417.

[185] R.J. Davis, Signal transduction by the JNK group of MAP kinases. Cell 103 (2000) 239-252.

[186] S. Bagrodia, B. Derijard, R.J. Davis, R.A. Cerione, Cdc42 and PAK-mediated signaling leads to Jun kinase and p38 mitogen-activated protein kinase activation. J.Biol.Chem. 270 (1995) 27995-27998.

[187] J. Han, J.D. Lee, Y. Jiang, Z. Li, L. Feng, R.J. Ulevitch, Characterization of the structure and function of a novel MAP kinase kinase (MKK6). J.Biol.Chem. 271 (1996) 2886-2891.

[188] T.M. Thornton, M. Rincon, Non-classical p38 map kinase functions: cell cycle checkpoints and survival. Int.J.Biol.Sci. 5 (2009) 44-51.

[189] A.M. Bode, Z. Dong, The functional contrariety of JNK. Mol.Carcinog. 46 (2007) 591-598.

[190] Y. Mizukami, K. Yoshioka, S. Morimoto, K. Yoshida, A novel mechanism of JNK1 activation. Nuclear translocation and activation of JNK1 during ischemia and reperfusion. J.Biol.Chem. 272 (1997) 16657-16662.

[191] M.A. Bogoyevitch, K.R. Ngoei, T.T. Zhao, Y.Y. Yeap, D.C. Ng, c-Jun N- terminal kinase (JNK) signaling: recent advances and challenges. Biochim.Biophys.Acta 1804 (2010) 463-475.

[192] K. Sabapathy, K. Hochedlinger, S.Y. Nam, A. Bauer, M. Karin, E.F. Wagner, Distinct roles for JNK1 and JNK2 in regulating JNK activity and c-Jun-dependent cell proliferation. Mol.Cell 15 (2004) 713-725.

[193] D.N. Dhanasekaran, E.P. Reddy, JNK signaling in apoptosis. Oncogene 27 (2008) 6245-6251.

[194] L.W. Wu, L.D. Mayo, J.D. Dunbar, K.M. Kessler, M.R. Baerwald, E.A. Jaffe, D. Wang, R.S. Warren, D.B. Donner, Utilization of distinct signaling pathways by

279

receptors for vascular endothelial cell growth factor and other mitogens in the induction of endothelial cell proliferation. J.Biol.Chem. 275 (2000) 5096-5103.

[195] Y. Yu, J.D. Sato, MAP kinases, phosphatidylinositol 3-kinase, and p70 S6 kinase mediate the mitogenic response of human endothelial cells to vascular endothelial growth factor. J.Cell Physiol 178 (1999) 235-246.

[196] S. Rousseau, F. Houle, H. Kotanides, L. Witte, J. Waltenberger, J. Landry, J. Huot, Vascular endothelial growth factor (VEGF)-driven actin-based motility is mediated by VEGFR2 and requires concerted activation of stress-activated protein kinase 2 (SAPK2/p38) and geldanamycin-sensitive phosphorylation of focal adhesion kinase. J.Biol.Chem. 275 (2000) 10661-10672.

[197] M.J. Yoon, C.H. Cho, C.S. Lee, I.H. Jang, S.H. Ryu, G.Y. Koh, Localization of Tie2 and phospholipase D in endothelial caveolae is involved in angiopoietin-1- induced MEK/ERK phosphorylation and migration in endothelial cells. Biochem.Biophys.Res.Commun. 308 (2003) 101-105.

[198] D.J. Waugh, C. Wilson, The interleukin-8 pathway in cancer. Clin.Cancer Res. 14 (2008) 6735-6741.

[199] V.P. Sukhatme, X.M. Cao, L.C. Chang, C.H. Tsai-Morris, D. Stamenkovich, P.C. Ferreira, D.R. Cohen, S.A. Edwards, T.B. Shows, T. Curran, ., A zinc finger- encoding gene coregulated with c-fos during growth and differentiation, and after cellular depolarization. Cell 53 (1988) 37-43.

[200] L.M. Khachigian, V. Lindner, A.J. Williams, T. Collins, Egr-1-induced endothelial gene expression: a common theme in vascular injury. Science 271 (1996) 1427-1431.

[201] G.J. Delbridge, L.M. Khachigian, FGF-1-induced platelet-derived growth factor- A chain gene expression in endothelial cells involves transcriptional activation by early growth response factor-1. Circ.Res. 81 (1997) 282-288.

[202] R.G. Fahmy, C.R. Dass, L.Q. Sun, C.N. Chesterman, L.M. Khachigian, Transcription factor Egr-1 supports FGF-dependent angiogenesis during neovascularization and tumor growth. Nat.Med. 9 (2003) 1026-1032.

[203] P. Cohen, The origins of protein phosphorylation. Nat.Cell Biol. 4 (2002) E127- E130.

[204] N.K. Tonks, B.G. Neel, From form to function: signaling by protein tyrosine phosphatases. Cell 87 (1996) 365-368.

[205] N.K. Tonks, Protein tyrosine phosphatases: from genes, to function, to disease. Nat.Rev.Mol.Cell Biol. 7 (2006) 833-846.

280

[206] R.J. Dickinson, S.M. Keyse, Diverse physiological functions for dual-specificity MAP kinase phosphatases. J.Cell Sci. 119 (2006) 4607-4615.

[207] K.L. Jeffrey, M. Camps, C. Rommel, C.R. Mackay, Targeting dual-specificity phosphatases: manipulating MAP kinase signalling and immune responses. Nat.Rev.Drug Discov. 6 (2007) 391-403.

[208] R. Echavarria, S.N. Hussain, Regulation of angiopoietin-1/Tie-2 receptor signaling in endothelial cells by dual-specificity phosphatases 1, 4, and 5. J Am Heart Assoc. 2 (2013) e000571.

[209] Z. Master, N. Jones, J. Tran, J. Jones, R.S. Kerbel, D.J. Dumont, Dok-R plays a pivotal role in angiopoietin-1-dependent cell migration through recruitment and activation of Pak. EMBO J. 20 (2001) 5919-5928.

[210] N. Jones, D.J. Dumont, The Tek/Tie2 receptor signals through a novel Dok- related docking protein, Dok-R. Oncogene 17 (1998) 1097-1108.

[211] T.R. Carlson, Y. Feng, P.C. Maisonpierre, M. Mrksich, A.O. Morla, Direct cell adhesion to the angiopoietins mediated by integrins. J.Biol Chem. 276 (2001) 26516-26525.

[212] S.M. Dallabrida, N. Ismail, J.R. Oberle, B.E. Himes, M.A. Rupnick, Angiopoietin-1 promotes cardiac and skeletal myocyte survival through integrins. Circ.Res. 96 (2005) e8-24.

[213] I. Cascone, L. Napione, F. Maniero, G. Serini, F. Bussolino, Stable interaction between 51 integrin and Tie2 tyrosine kinase receptor regulates endothelial cell response to Ang-1. J.Cell Biol. 170 (2005) 993-1004.

[214] M. Felcht, R. Luck, A. Schering, P. Seidel, K. Srivastava, J. Hu, A. Bartol, Y. Kienast, C. Vettel, E.K. Loos, S. Kutschera, S. Bartels, S. Appak, E. Besemfelder, D. Terhardt, E. Chavakis, T. Wieland, C. Klein, M. Thomas, A. Uemura, S. Goerdt, H.G. Augustin, Angiopoietin-2 differentially regulates angiogenesis through TIE2 and integrin signaling. J.Clin.Invest 122 (2012) 1991-2005.

[215] G. Serini, L. Napione, M. Arese, F. Bussolino, Besides adhesion: new perspectives of integrin functions in angiogenesis. Cardiovasc.Res. 78 (2008) 213-222.

[216] C.J. Avraamides, B. Garmy-Susini, J.A. Varner, Integrins in angiogenesis and lymphangiogenesis. Nat.Rev.Cancer 8 (2008) 604-617.

[217] Q. Chen, J. Massague, Molecular pathways: VCAM-1 as a potential therapeutic target in metastasis. Clin.Cancer Res. 18 (2012) 5520-5525.

[218] H. Jin, J. Varner, Integrins: roles in cancer development and as treatment targets. Br.J.Cancer 90 (2004) 561-565.

281

[219] Y. Imanishi, B. Hu, M.J. Jarzynka, P. Guo, E. Elishaev, I. Bar-Joseph, S.Y. Cheng, Angiopoietin-2 stimulates breast cancer metastasis through the alpha(5)beta(1) integrin-mediated pathway. Cancer Res. 67 (2007) 4254-4263.

[220] Y.J. Koh, H.Z. Kim, S.I. Hwang, J.E. Lee, N. Oh, K. Jung, M. Kim, K.E. Kim, H. Kim, N.K. Lim, C.J. Jeon, G.M. Lee, B.H. Jeon, D.H. Nam, H.K. Sung, A. Nagy, O.J. Yoo, G.Y. Koh, Double antiangiogenic protein, DAAP, targeting VEGF-A and angiopoietins in tumor angiogenesis, metastasis, and vascular leakage. Cancer Cell 18 (2010) 171-184.

[221] J.M. Li, A.M. Mullen, S. Yun, F. Wientjes, G.Y. Brouns, A.J. Thrasher, A.M. Shah, Essential role of the NADPH oxidase subunit p47(phox) in endothelial cell superoxide production in response to phorbol ester and tumor necrosis factor- alpha. Circ.Res. 90 (2002) 143-150.

[222] T. Finkel, Signal transduction by mitochondrial oxidants. J Biol.Chem. 287 (2012) 4434-4440.

[223] T. Finkel, Signal transduction by reactive oxygen species. J.Cell Biol. 194 (2011) 7-15.

[224] K.M. Holmstrom, T. Finkel, Cellular mechanisms and physiological consequences of redox-dependent signalling. Nat.Rev.Mol.Cell Biol. 15 (2014) 411-421.

[225] M. Ushio-Fukai, VEGF signaling through NADPH oxidase-derived ROS. Antioxid.Redox.Signal. 9 (2007) 731-739.

[226] K.C. Chen, Y. Zhou, W. Zhang, M.F. Lou, Control of PDGF-induced reactive oxygen species (ROS) generation and signal transduction in human lens epithelial cells. Mol.Vis. 13 (2007) 374-387.

[227] Y.S. Bae, S.W. Kang, M.S. Seo, I.C. Baines, E. Tekle, P.B. Chock, S.G. Rhee, Epidermal growth factor (EGF)-induced generation of hydrogen peroxide. Role in EGF receptor-mediated tyrosine phosphorylation. J.Biol.Chem. 272 (1997) 217- 221.

[228] Y.S. Bae, J.Y. Sung, O.S. Kim, Y.J. Kim, K.C. Hur, A. Kazlauskas, S.G. Rhee, Platelet-derived growth factor-induced H(2)O(2) production requires the activation of phosphatidylinositol 3-kinase. J.Biol.Chem. 275 (2000) 10527- 10531.

[229] C.A. Nguyen Dinh, A.C. Montezano, D. Burger, R.M. Touyz, Angiotensin II, NADPH oxidase, and redox signaling in the vasculature. Antioxid.Redox.Signal. 19 (2013) 1110-1120.

[230] Y.F. Pung, W.J. Sam, K. Stevanov, M. Enrick, C.L. Chen, C. Kolz, P. Thakker, J.P. Hardwick, Y.R. Chen, J.R. Dyck, L. Yin, W.M. Chilian, Mitochondrial

282

oxidative stress corrupts coronary collateral growth by activating adenosine monophosphate activated kinase-alpha signaling. Arterioscler.Thromb.Vasc.Biol. 33 (2013) 1911-1919.

[231] H. Okuno, A. Akahori, H. Sato, S. Xanthoudakis, T. Curran, H. Iba, Escape from redox regulation enhances the transforming activity of Fos. Oncogene 8 (1993) 695-701.

[232] C.R. Reczek, N.S. Chandel, ROS-dependent signal transduction. Curr.Opin.Cell Biol. 33 (2015) 8-13.

[233] A.R. Cross, A.W. Segal, The NADPH oxidase of professional phagocytes-- prototype of the NOX electron transport chain systems. Biochim.Biophys.Acta 1657 (2004) 1-22.

[234] A. Petry, T. Djordjevic, M. Weitnauer, T. Kietzmann, J. Hess, A. Gorlach, NOX2 and NOX4 mediate proliferative response in endothelial cells. Antioxid.Redox.Signal. 8 (2006) 1473-1484.

[235] J.D. van Buul, M. Fernandez-Borja, E.C. Anthony, P.L. Hordijk, Expression and localization of NOX2 and NOX4 in primary human endothelial cells. Antioxid.Redox.Signal. 7 (2005) 308-317.

[236] R.M. Touyz, A.M. Briones, M. Sedeek, D. Burger, A.C. Montezano, NOX isoforms and reactive oxygen species in vascular health. Mol.Interv. 11 (2011) 27-35.

[237] R.M. Touyz, X. Chen, F. Tabet, G. Yao, G. He, M.T. Quinn, P.J. Pagano, E.L. Schiffrin, Expression of a functionally active gp91phox-containing neutrophil- type NAD(P)H oxidase in smooth muscle cells from human resistance arteries: regulation by angiotensin II. Circ.Res. 90 (2002) 1205-1213.

[238] B.A. Diebold, G.M. Bokoch, Molecular basis for Rac2 regulation of phagocyte NADPH oxidase. Nat.Immunol. 2 (2001) 211-215.

[239] S. Grizot, J. Faure, F. Fieschi, P.V. Vignais, M.C. Dagher, E. Pebay-Peyroula, Crystal structure of the Rac1-RhoGDI complex involved in nadph oxidase activation. Biochemistry 40 (2001) 10007-10013.

[240] M. Takahashi, T.J. Dillon, C. Liu, Y. Kariya, Z. Wang, P.J. Stork, Protein kinase A-dependent phosphorylation of Rap1 regulates its membrane localization and cell migration. J.Biol.Chem. 288 (2013) 27712-27723.

[241] R.P. Brandes, K. Schroder, Composition and functions of vascular nicotinamide adenine dinucleotide phosphate oxidases. Trends Cardiovasc.Med. 18 (2008) 15- 19.

283

[242] M. Ushio-Fukai, Localizing NADPH oxidase-derived ROS. Sci.STKE. 2006 (2006) re8.

[243] F.D. Oakley, D. Abbott, Q. Li, J.F. Engelhardt, Signaling components of redox active endosomes: the redoxosomes. Antioxid.Redox.Signal. 11 (2009) 1313- 1333.

[244] A.M. Evangelista, M.D. Thompson, V.M. Bolotina, X. Tong, R.A. Cohen, Nox4- and Nox2-dependent oxidant production is required for VEGF-induced SERCA cysteine-674 S-glutathiolation and endothelial cell migration. Free Radic.Biol.Med. 53 (2012) 2327-2334.

[245] R. Dworakowski, S.P. Alom-Ruiz, A.M. Shah, NADPH oxidase-derived reactive oxygen species in the regulation of endothelial phenotype. Pharmacol.Rep. 60 (2008) 21-28.

[246] A.J. Case, S. Li, U. Basu, J. Tian, M.C. Zimmerman, Mitochondrial-localized NADPH oxidase 4 is a source of superoxide in angiotensin II-stimulated neurons. Am.J.Physiol Heart Circ.Physiol 305 (2013) H19-H28.

[247] K.D. Martyn, L.M. Frederick, L.K. von, M.C. Dinauer, U.G. Knaus, Functional analysis of Nox4 reveals unique characteristics compared to other NADPH oxidases 3655. Cell Signal. 18 (2006) 69-82.

[248] F.J. Miller, Jr., NADPH oxidase 4: walking the walk with Poldip2. Circ.Res. 105 (2009) 209-210.

[249] Y. Nisimoto, B.A. Diebold, D. Cosentino-Gomes, J.D. Lambeth, Nox4: a hydrogen peroxide-generating oxygen sensor. Biochemistry 53 (2014) 5111-5120.

[250] I. Takac, K. Schroder, L. Zhang, B. Lardy, N. Anilkumar, J.D. Lambeth, A.M. Shah, F. Morel, R.P. Brandes, The E-loop is involved in hydrogen peroxide formation by the NADPH oxidase Nox4. J.Biol.Chem. 286 (2011) 13304-13313.

[251] A. Sturrock, T.P. Huecksteadt, K. Norman, K. Sanders, T.M. Murphy, P. Chitano, K. Wilson, J.R. Hoidal, T.P. Kennedy, Nox4 mediates TGF-beta1-induced retinoblastoma protein phosphorylation, proliferation, and hypertrophy in human airway smooth muscle cells. Am.J.Physiol Lung Cell Mol.Physiol 292 (2007) L1543-L1555.

[252] J. Kuroda, K. Nakagawa, T. Yamasaki, K. Nakamura, R. Takeya, F. Kuribayashi, S. Imajoh-Ohmi, K. Igarashi, Y. Shibata, K. Sueishi, H. Sumimoto, The superoxide-producing NAD(P)H oxidase Nox4 in the nucleus of human vascular endothelial cells. Genes Cells 10 (2005) 1139-1151.

[253] T. Yamamori, O. Inanami, H. Nagahata, M. Kuwabara, Phosphoinositide 3-kinase regulates the phosphorylation of NADPH oxidase component p47(phox) by

284

controlling cPKC/PKCdelta but not Akt. Biochem.Biophys.Res.Commun. 316 (2004) 720-730.

[254] L.E. Kilpatrick, S. Sun, H. Li, T.C. Vary, H.M. Korchak, Regulation of TNF- induced oxygen radical production in human neutrophils: role of delta-PKC. J.Leukoc.Biol. 87 (2010) 153-164.

[255] Y.S. Kim, M.J. Morgan, S. Choksi, Z.G. Liu, TNF-induced activation of the Nox1 NADPH oxidase and its role in the induction of necrotic cell death 1. Mol.Cell 26 (2007) 675-687.

[256] D. Gianni, B. Bohl, S.A. Courtneidge, G.M. Bokoch, The involvement of the tyrosine kinase c-Src in the regulation of reactive oxygen species generation mediated by NADPH oxidase-1. Mol.Biol.Cell 19 (2008) 2984-2994.

[257] I.M. Fenyo, I.C. Florea, M. Raicu, A. Manea, Tyrphostin AG490 reduces NAPDH oxidase activity and expression in the aorta of hypercholesterolemic apolipoprotein E-deficient mice. Vascul.Pharmacol. 54 (2011) 100-106.

[258] F. Chen, Y. Yu, S. Haigh, J. Johnson, R. Lucas, D.W. Stepp, D.J. Fulton, Regulation of NADPH oxidase 5 by protein kinase C isoforms. PLoS.ONE. 9 (2014) e88405.

[259] M. Janiszewski, L.R. Lopes, A.O. Carmo, M.A. Pedro, R.P. Brandes, C.X. Santos, F.R. Laurindo, Regulation of NAD(P)H oxidase by associated protein disulfide isomerase in vascular smooth muscle cells. J.Biol.Chem. 280 (2005) 40813-40819.

[260] F.J. Miller, Jr., M. Filali, G.J. Huss, B. Stanic, A. Chamseddine, T.J. Barna, F.S. Lamb, Cytokine activation of nuclear factor kappa B in vascular smooth muscle cells requires signaling endosomes containing Nox1 and ClC-3. Circ.Res. 101 (2007) 663-671.

[261] D. Pandey, J.P. Gratton, R. Rafikov, S.M. Black, D.J. Fulton, Calcium/calmodulin-dependent kinase II mediates the phosphorylation and activation of NADPH oxidase 5. Mol.Pharmacol. 80 (2011) 407-415.

[262] R. Gareus, E. Kotsaki, S. Xanthoulea, d.M. van, I, M.J. Gijbels, R. Kardakaris, A. Polykratis, G. Kollias, M.P. de Winther, M. Pasparakis, Endothelial cell-specific NF-kappaB inhibition protects mice from atherosclerosis. Cell Metab 8 (2008) 372-383.

[263] S.A. Manea, A. Manea, C. Heltianu, Inhibition of JAK/STAT signaling pathway prevents high-glucose-induced increase in endothelin-1 synthesis in human endothelial cells. Cell Tissue Res. 340 (2010) 71-79.

[264] C. Recio, A. Oguiza, I. Lazaro, B. Mallavia, J. Egido, C. Gomez-Guerrero, Suppressor of cytokine signaling 1-derived peptide inhibits Janus kinase/signal

285

transducers and activators of transcription pathway and improves inflammation and atherosclerosis in diabetic mice. Arterioscler.Thromb.Vasc.Biol. 34 (2014) 1953-1960.

[265] E. Sozen, B. Karademir, B. Yazgan, P. Bozaykut, N.K. Ozer, Potential role of proteasome on c-jun related signaling in hypercholesterolemia induced atherosclerosis. Redox.Biol. 2 (2014) 732-738.

[266] P. Hayes, U.G. Knaus, Balancing reactive oxygen species in the epigenome: NADPH oxidases as target and perpetrator. Antioxid.Redox.Signal. 18 (2013) 1937-1945.

[267] D. Siuda, U. Zechner, H.N. El, D. Prawitt, D. Langer, N. Xia, S. Horke, A. Pautz, H. Kleinert, U. Forstermann, H. Li, Transcriptional regulation of Nox4 by histone deacetylases in human endothelial cells. Basic Res.Cardiol. 107 (2012) 283.

[268] Y. Fu, Y. Zhang, Z. Wang, L. Wang, X. Wei, B. Zhang, Z. Wen, H. Fang, Q. Pang, F. Yi, Regulation of NADPH oxidase activity is associated with miRNA- 25-mediated NOX4 expression in experimental diabetic nephropathy. Am.J.Nephrol. 32 (2010) 581-589.

[269] G.M. Gordillo, A. Biswas, S. Khanna, X. Pan, M. Sinha, S. Roy, C.K. Sen, Dicer knockdown inhibits endothelial cell tumor growth via microRNA 21a-3p targeting of Nox-4. J.Biol.Chem. 289 (2014) 9027-9038.

[270] R. Ranjan, Y.G. Lee, M. Karpurapu, M.A. Syed, S. Chung, J. Deng, J.J. Jeong, G. Zhao, L. Xiao, R.T. Sadikot, M.J. Weiss, J.W. Christman, G.Y. Park, p47phox and reactive oxygen species production modulate expression of microRNA-451 in macrophages. Free Radic.Res. 49 (2015) 25-34.

[271] J.X. Wang, J. Gao, S.L. Ding, K. Wang, J.Q. Jiao, Y. Wang, T. Sun, L.Y. Zhou, B. Long, X.J. Zhang, Q. Li, J.P. Liu, C. Feng, J. Liu, Y. Gong, Z. Zhou, P.F. Li, Oxidative Modification of miR-184 Enables It to Target Bcl-xL and Bcl-w. Mol.Cell 59 (2015) 50-61.

[272] S. Pendyala, I.A. Gorshkova, P.V. Usatyuk, D. He, A. Pennathur, J.D. Lambeth, V.J. Thannickal, V. Natarajan, Role of Nox4 and Nox2 in hyperoxia-induced reactive oxygen species generation and migration of human lung endothelial cells. Antioxid.Redox.Signal. 11 (2009) 747-764.

[273] S.M. Craige, K. Chen, Y. Pei, C. Li, X. Huang, C. Chen, R. Shibata, K. Sato, K. Walsh, J.F. Keaney, Jr., NADPH oxidase 4 promotes endothelial angiogenesis through endothelial nitric oxide synthase activation. Circulation 124 (2011) 731- 740.

[274] N. Urao, H. Inomata, M. Razvi, H.W. Kim, K. Wary, R. McKinney, T. Fukai, M. Ushio-Fukai, Role of nox2-based NADPH oxidase in bone marrow and

286

progenitor cell function involved in neovascularization induced by hindlimb ischemia. Circ.Res. 103 (2008) 212-220.

[275] S. Ikeda, M. Yamaoka-Tojo, L. Hilenski, N.A. Patrushev, G.M. Anwar, M.T. Quinn, M. Ushio-Fukai, IQGAP1 regulates reactive oxygen species-dependent endothelial cell migration through interacting with Nox2. Arterioscler.Thromb.Vasc.Biol. 25 (2005) 2295-2300.

[276] Y. Wang, Q.S. Zang, Z. Liu, Q. Wu, D. Maass, G. Dulan, P.W. Shaul, L. Melito, D.E. Frantz, J.A. Kilgore, N.S. Williams, L.S. Terada, F.E. Nwariaku, Regulation of VEGF Induced Endothelial Cell Migration by Mitochondrial Reactive Oxygen Species. Am.J Physiol Cell Physiol2011).

[277] K.M. Connor, S. Subbaram, K.J. Regan, K.K. Nelson, J.E. Mazurkiewicz, P.J. Bartholomew, A.E. Aplin, Y.T. Tai, J. Aguirre-Ghiso, S.C. Flores, J.A. Melendez, Mitochondrial H2O2 regulates the angiogenic phenotype via PTEN oxidation. J.Biol.Chem. 280 (2005) 16916-16924.

[278] M. Schleicher, B.R. Shepherd, Y. Suarez, C. Fernandez-Hernando, J. Yu, Y. Pan, L.M. Acevedo, G.S. Shadel, W.C. Sessa, Prohibitin-1 maintains the angiogenic capacity of endothelial cells by regulating mitochondrial function and senescence. J.Cell Biol. 180 (2008) 101-112.

[279] M.P. Murphy, How mitochondria produce reactive oxygen species. Biochem.J. 417 (2009) 1-13.

[280] F.L. Muller, Y. Liu, R.H. Van, Complex III releases superoxide to both sides of the inner mitochondrial membrane. J.Biol.Chem. 279 (2004) 49064-49073.

[281] L.A. Sena, N.S. Chandel, Physiological roles of mitochondrial reactive oxygen species. Mol.Cell 48 (2012) 158-167.

[282] T.C. Meng, T. Fukada, N.K. Tonks, Reversible oxidation and inactivation of protein tyrosine phosphatases in vivo. Mol.Cell 9 (2002) 387-399.

[283] J. Kwon, S.R. Lee, K.S. Yang, Y. Ahn, Y.J. Kim, E.R. Stadtman, S.G. Rhee, Reversible oxidation and inactivation of the tumor suppressor PTEN in cells stimulated with peptide growth factors. Proc.Natl.Acad.Sci.U.S.A 101 (2004) 16419-16424.

[284] D.J. Levinthal, D.B. Defranco, Reversible oxidation of ERK-directed protein phosphatases drives oxidative toxicity in neurons. J.Biol.Chem. 280 (2005) 5875- 5883.

[285] H. Kamata, S. Honda, S. Maeda, L. Chang, H. Hirata, M. Karin, Reactive oxygen species promote TNFalpha-induced death and sustained JNK activation by inhibiting MAP kinase phosphatases. Cell 120 (2005) 649-661.

287

[286] A. Kuehbacher, C. Urbich, A.M. Zeiher, S. Dimmeler, Role of Dicer and Drosha for endothelial microRNA expression and angiogenesis. Circ.Res. 101 (2007) 59- 68.

[287] Y. Suarez, C. Fernandez-Hernando, J.S. Pober, W.C. Sessa, Dicer dependent microRNAs regulate gene expression and functions in human endothelial cells. Circ.Res. 100 (2007) 1164-1173.

[288] Y. Suarez, C. Fernandez-Hernando, J. Yu, S.A. Gerber, K.D. Harrison, J.S. Pober, M.L. Iruela-Arispe, M. Merkenschlager, W.C. Sessa, Dicer-dependent endothelial microRNAs are necessary for postnatal angiogenesis. Proc.Natl.Acad.Sci.U.S.A 105 (2008) 14082-14087.

[289] R.C. Lee, R.L. Feinbaum, V. Ambros, The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell 75 (1993) 843-854.

[290] B. Wightman, I. Ha, G. Ruvkun, Posttranscriptional regulation of the heterochronic gene lin-14 by lin-4 mediates temporal pattern formation in C. elegans. Cell 75 (1993) 855-862.

[291] B.J. Reinhart, E.G. Weinstein, M.W. Rhoades, B. Bartel, D.P. Bartel, MicroRNAs in plants. Genes Dev. 16 (2002) 1616-1626.

[292] M. Lagos-Quintana, R. Rauhut, W. Lendeckel, T. Tuschl, Identification of novel genes coding for small expressed RNAs. Science 294 (2001) 853-858.

[293] E. Huntzinger, E. Izaurralde, Gene silencing by microRNAs: contributions of translational repression and mRNA decay. Nat.Rev.Genet. 12 (2011) 99-110.

[294] E. Wienholds, R.H. Plasterk, MicroRNA function in animal development. FEBS Lett. 579 (2005) 5911-5922.

[295] I. Alvarez-Garcia, E.A. Miska, MicroRNA functions in animal development and human disease. Development 132 (2005) 4653-4662.

[296] Q. Cui, Z. Yu, E.O. Purisima, E. Wang, Principles of microRNA regulation of a human cellular signaling network. Mol.Syst.Biol. 2 (2006) 46.

[297] D.P. Bartel, MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116 (2004) 281-297.

[298] A.J. Pratt, I.J. MacRae, The RNA-induced silencing complex: a versatile gene- silencing machine. J.Biol.Chem. 284 (2009) 17897-17901.

[299] Y. Lee, M. Kim, J. Han, K.H. Yeom, S. Lee, S.H. Baek, V.N. Kim, MicroRNA genes are transcribed by RNA polymerase II. EMBO J. 23 (2004) 4051-4060.

288

[300] G.M. Borchert, W. Lanier, B.L. Davidson, RNA polymerase III transcribes human microRNAs. Nat.Struct.Mol.Biol. 13 (2006) 1097-1101.

[301] X. Zhang, B. Zhang, J. Gao, X. Wang, Z. Liu, Regulation of the microRNA 200b (miRNA-200b) by transcriptional regulators PEA3 and ELK-1 protein affects expression of Pin1 protein to control anoikis. J.Biol.Chem. 288 (2013) 32742- 32752.

[302] K.D. Taganov, M.P. Boldin, K.J. Chang, D. Baltimore, NF-kappaB-dependent induction of microRNA miR-146, an inhibitor targeted to signaling proteins of innate immune responses. Proc.Natl.Acad.Sci.U.S.A 103 (2006) 12481-12486.

[303] K.A. O'Donnell, E.A. Wentzel, K.I. Zeller, C.V. Dang, J.T. Mendell, c-Myc- regulated microRNAs modulate E2F1 expression. Nature 435 (2005) 839-843.

[304] L. He, X. He, L.P. Lim, S.E. de, Z. Xuan, Y. Liang, W. Xue, L. Zender, J. Magnus, D. Ridzon, A.L. Jackson, P.S. Linsley, C. Chen, S.W. Lowe, M.A. Cleary, G.J. Hannon, A microRNA component of the p53 tumour suppressor network. Nature 447 (2007) 1130-1134.

[305] Y. Saito, G. Liang, G. Egger, J.M. Friedman, J.C. Chuang, G.A. Coetzee, P.A. Jones, Specific activation of microRNA-127 with downregulation of the proto- oncogene BCL6 by chromatin-modifying drugs in human cancer cells. Cancer Cell 9 (2006) 435-443.

[306] B. Brueckner, C. Stresemann, R. Kuner, C. Mund, T. Musch, M. Meister, H. Sultmann, F. Lyko, The human let-7a-3 locus contains an epigenetically regulated microRNA gene with oncogenic function. Cancer Res. 67 (2007) 1419-1423.

[307] Y. Hayashita, H. Osada, Y. Tatematsu, H. Yamada, K. Yanagisawa, S. Tomida, Y. Yatabe, K. Kawahara, Y. Sekido, T. Takahashi, A polycistronic microRNA cluster, miR-17-92, is overexpressed in human lung cancers and enhances cell proliferation. Cancer Res. 65 (2005) 9628-9632.

[308] C.P. Bracken, P.A. Gregory, N. Kolesnikoff, A.G. Bert, J. Wang, M.F. Shannon, G.J. Goodall, A double-negative feedback loop between ZEB1-SIP1 and the microRNA-200 family regulates epithelial-mesenchymal transition. Cancer Res. 68 (2008) 7846-7854.

[309] Y. Lee, C. Ahn, J. Han, H. Choi, J. Kim, J. Yim, J. Lee, P. Provost, O. Radmark, S. Kim, V.N. Kim, The nuclear RNase III Drosha initiates microRNA processing. Nature 425 (2003) 415-419.

[310] J. Winter, S. Jung, S. Keller, R.I. Gregory, S. Diederichs, Many roads to maturity: microRNA biogenesis pathways and their regulation. Nat.Cell Biol. 11 (2009) 228-234.

289

[311] Y. Zeng, B.R. Cullen, Efficient processing of primary microRNA hairpins by Drosha requires flanking nonstructured RNA sequences. J.Biol.Chem. 280 (2005) 27595-27603.

[312] J.G. Ruby, C.H. Jan, D.P. Bartel, Intronic microRNA precursors that bypass Drosha processing. Nature 448 (2007) 83-86.

[313] R. Yi, Y. Qin, I.G. Macara, B.R. Cullen, Exportin-5 mediates the nuclear export of pre-microRNAs and short hairpin RNAs. Genes Dev. 17 (2003) 3011-3016.

[314] D.S. Schwarz, G. Hutvagner, T. Du, Z. Xu, N. Aronin, P.D. Zamore, Asymmetry in the assembly of the RNAi enzyme complex. Cell 115 (2003) 199-208.

[315] A. Khvorova, A. Reynolds, S.D. Jayasena, Functional siRNAs and miRNAs exhibit strand bias. Cell 115 (2003) 209-216.

[316] Y. Suarez, W.C. Sessa, MicroRNAs as novel regulators of angiogenesis. Circ.Res. 104 (2009) 442-454.

[317] M. Inui, G. Martello, S. Piccolo, MicroRNA control of signal transduction. Nat.Rev.Mol.Cell Biol. 11 (2010) 252-263.

[318] S. Wang, A.B. Aurora, B.A. Johnson, X. Qi, J. McAnally, J.A. Hill, J.A. Richardson, R. Bassel-Duby, E.N. Olson, The endothelial-specific microRNA miR-126 governs vascular integrity and angiogenesis. Dev.Cell 15 (2008) 261- 271.

[319] J.E. Fish, M.M. Santoro, S.U. Morton, S. Yu, R.F. Yeh, J.D. Wythe, K.N. Ivey, B.G. Bruneau, D.Y. Stainier, D. Srivastava, miR-126 regulates angiogenic signaling and vascular integrity. Dev.Cell 15 (2008) 272-284.

[320] Q. Zhou, R. Gallagher, R. Ufret-Vincenty, X. Li, E.N. Olson, S. Wang, Regulation of angiogenesis and choroidal neovascularization by members of microRNA-23~27~24 clusters. Proc.Natl.Acad.Sci.U.S.A 108 (2011) 8287-8292.

[321] A. Bonauer, G. Carmona, M. Iwasaki, M. Mione, M. Koyanagi, A. Fischer, J. Burchfield, H. Fox, C. Doebele, K. Ohtani, E. Chavakis, M. Potente, M. Tjwa, C. Urbich, A.M. Zeiher, S. Dimmeler, MicroRNA-92a controls angiogenesis and functional recovery of ischemic tissues in mice. Science 324 (2009) 1710-1713.

[322] C. Doebele, A. Bonauer, A. Fischer, A. Scholz, Y. Reiss, C. Urbich, W.K. Hofmann, A.M. Zeiher, S. Dimmeler, Members of the microRNA-17-92 cluster exhibit a cell-intrinsic antiangiogenic function in endothelial cells. Blood 115 (2010) 4944-4950.

[323] M. Potente, L. Ghaeni, D. Baldessari, R. Mostoslavsky, L. Rossig, F. Dequiedt, J. Haendeler, M. Mione, E. Dejana, F.W. Alt, A.M. Zeiher, S. Dimmeler, SIRT1

290

controls endothelial angiogenic functions during vascular growth. Genes Dev. 21 (2007) 2644-2658.

[324] Y. Zhou, X.H. Li, C.C. Zhang, M.J. Wang, W.L. Xue, D.D. Wu, F.F. Ma, W.W. Li, B.B. Tao, Y.C. Zhu, Hydrogen sulfide promotes angiogenesis by downregulating miR-640 via the VEGFR2/mTOR pathway. Am.J.Physiol Cell Physiol 310 (2016) C305-C317.

[325] T. Voigtlander, S.K. Gupta, S. Thum, J. Fendrich, M.P. Manns, T.O. Lankisch, T. Thum, MicroRNAs in Serum and Bile of Patients with Primary Sclerosing Cholangitis and/or Cholangiocarcinoma. PLoS.ONE. 10 (2015) e0139305.

[326] X. Li, Y. Lu, Y. Chen, W. Lu, X. Xie, MicroRNA profile of paclitaxel-resistant serous ovarian carcinoma based on formalin-fixed paraffin-embedded samples. BMC.Cancer 13 (2013) 216.

[327] R. Visone, L.Z. Rassenti, A. Veronese, C. Taccioli, S. Costinean, B.D. Aguda, S. Volinia, M. Ferracin, J. Palatini, V. Balatti, H. Alder, M. Negrini, T.J. Kipps, C.M. Croce, Karyotype-specific microRNA signature in chronic lymphocytic leukemia. Blood 114 (2009) 3872-3879.

[328] X. Jin, H.R. Jin, H.S. Jung, S.J. Lee, J.H. Lee, J.J. Lee, An atypical E3 ligase zinc finger protein 91 stabilizes and activates NF-kappaB-inducing kinase via Lys63- linked ubiquitination. J.Biol.Chem. 285 (2010) 30539-30547.

[329] J. Ma, C. Mi, K.S. Wang, J.J. Lee, X. Jin, Zinc finger protein 91 (ZFP91) activates HIF-1alpha via NF-kappaB/p65 to promote proliferation and tumorigenesis of colon cancer. Oncotarget.2016).

[330] L. Paschke, M. Rucinski, A. Ziolkowska, T. Zemleduch, W. Malendowicz, Z. Kwias, L.K. Malendowicz, ZFP91-a newly described gene potentially involved in prostate pathology. Pathol.Oncol.Res. 20 (2014) 453-459.

[331] M. Unoki, J. Okutsu, Y. Nakamura, Identification of a novel human gene, ZFP91, involved in acute myelogenous leukemia. Int.J.Oncol. 22 (2003) 1217-1223.

[332] Y. Saotome, C.G. Winter, D. Hirsh, A widely expressed novel C2H2 zinc-finger protein with multiple consensus phosphorylation sites is conserved in mouse and man. Gene 152 (1995) 233-238.

[333] L.L. Qu, L. He, X. Zhao, W. Xu, Downregulation of miR-518a-3p activates the NIK-dependent NF-kappaB pathway in colorectal cancer. Int.J.Mol.Med. 35 (2015) 1266-1272.

[334] Y.M. Thu, A. Richmond, NF-kappaB inducing kinase: a key regulator in the immune system and in cancer. Cytokine Growth Factor Rev. 21 (2010) 213-226.

291

[335] M. Uno, Y. Saitoh, K. Mochida, E. Tsuruyama, T. Kiyono, I. Imoto, J. Inazawa, Y. Yuasa, T. Kubota, S. Yamaoka, NF-kappaB inducing kinase, a central signaling component of the non-canonical pathway of NF-kappaB, contributes to ovarian cancer progression. PLoS.ONE. 9 (2014) e88347.

[336] B.L. Krock, N. Skuli, M.C. Simon, Hypoxia-induced angiogenesis: good and evil. Genes Cancer 2 (2011) 1117-1133.

[337] B.J. Graves, J.M. Petersen, Specificity within the ets family of transcription factors. Adv.Cancer Res. 75 (1998) 1-55.

[338] A.D. Sharrocks, The ETS-domain transcription factor family. Nat.Rev.Mol.Cell Biol. 2 (2001) 827-837.

[339] J. Dittmer, The biology of the Ets1 proto-oncogene. Mol.Cancer 2 (2003) 29.

[340] J. Dittmer, A. Nordheim, Ets transcription factors and human disease. Biochim.Biophys.Acta 1377 (1998) F1-11.

[341] K. Redlich, H.P. Kiener, G. Schett, M. Tohidast-Akrad, E. Selzer, I. Radda, G.H. Stummvoll, C.W. Steiner, M. Groger, P. Bitzan, P. Zenz, J.S. Smolen, G. Steiner, Overexpression of transcription factor Ets-1 in rheumatoid arthritis synovial membrane: regulation of expression and activation by interleukin-1 and tumor necrosis factor alpha. Arthritis Rheum. 44 (2001) 266-274.

[342] P.C. Hollenhorst, L.P. McIntosh, B.J. Graves, Genomic and biochemical insights into the specificity of ETS transcription factors. Annu.Rev.Biochem. 80 (2011) 437-471.

[343] Y. Hasegawa, M. Abe, T. Yamazaki, O. Niizeki, K. Shiiba, I. Sasaki, Y. Sato, Transcriptional regulation of human angiopoietin-2 by transcription factor Ets-1 1. Biochem.Biophys.Res.Commun. 316 (2004) 52-58.

[344] C. Brown, J. Gaspar, A. Pettit, R. Lee, X. Gu, H. Wang, C. Manning, C. Voland, S.R. Goldring, M.B. Goldring, T.A. Libermann, E.M. Gravallese, P. Oettgen, ESE-1 is a novel transcriptional mediator of angiopoietin-1 expression in the setting of inflammation. J.Biol.Chem. 279 (2004) 12794-12803.

[345] X. Huang, C. Brown, W. Ni, E. Maynard, A.C. Rigby, P. Oettgen, Critical role for the Ets transcription factor ELF-1 in the development of tumor angiogenesis. Blood 107 (2006) 3153-3160.

[346] P.C. Hollenhorst, D.A. Jones, B.J. Graves, Expression profiles frame the promoter specificity dilemma of the ETS family of transcription factors. Nucleic Acids Res. 32 (2004) 5693-5702.

[347] J.S. Yordy, O. Moussa, H. Pei, D. Chaussabel, R. Li, D.K. Watson, SP100 inhibits ETS1 activity in primary endothelial cells. Oncogene 24 (2005) 916-931.

292

[348] M. Arman, N. Aguilera-Montilla, V. Mas, A. Puig-Kroger, M. Pignatelli, R. Guigo, A.L. Corbi, F. Lozano, The human CD6 gene is transcriptionally regulated by RUNX and Ets transcription factors in T cells. Mol.Immunol. 46 (2009) 2226- 2235.

[349] Z. Shi, Z. Cai, S. Wen, C. Chen, C. Gendron, A. Sanchez, K. Patterson, S. Fu, J. Yang, D. Wildman, R.H. Finnell, D. Zhang, Transcriptional regulation of the novel Toll-like receptor Tlr13. J.Biol.Chem. 284 (2009) 20540-20547.

[350] S. Nimer, J. Zhang, H. Avraham, Y. Miyazaki, Transcriptional regulation of interleukin-3 expression in megakaryocytes. Blood 88 (1996) 66-74.

[351] A. Grajkowski, J. Pedras-Vasconcelos, V. Wang, C. Ausin, S. Hess, D. Verthelyi, S.L. Beaucage, Thermolytic CpG-containing DNA oligonucleotides as potential immunotherapeutic prodrugs. Nucleic Acids Res. 33 (2005) 3550-3560.

[352] S.H. Heo, Y.J. Choi, H.M. Ryoo, J.Y. Cho, Expression profiling of ETS and MMP factors in VEGF-activated endothelial cells: role of MMP-10 in VEGF- induced angiogenesis. J.Cell Physiol 224 (2010) 734-742.

[353] S. Ghosh, M. Basu, S.S. Roy, ETS-1 protein regulates vascular endothelial growth factor-induced matrix metalloproteinase-9 and matrix metalloproteinase-13 expression in human ovarian carcinoma cell line SKOV-3. J.Biol.Chem. 287 (2012) 15001-15015.

[354] W. Erl, W. Siess, Sphingosine-1-phosphate and the leading Edg-1 of vascular smooth muscle cells. Circ.Res. 89 (2001) 474-476.

[355] T. Igarashi, M. Abe, M. Oikawa, T. Nukiwa, Y. Sato, Retinoic acids repress the expression of ETS-1 in endothelial cells. Tohoku J.Exp.Med. 194 (2001) 35-43.

[356] B. Spangler, M. Kappelmann, B. Schittek, S. Meierjohann, L. Vardimon, A.K. Bosserhoff, S. Kuphal, ETS-1/RhoC signaling regulates the transcription factor c- Jun in melanoma. Int.J.Cancer 130 (2012) 2801-2811.

[357] W. Ding, S. Bellusci, W. Shi, D. Warburton, Functional analysis of the human Sprouty2 gene promoter. Gene 322 (2003) 175-185.

[358] M.M. Valter, A. Hugel, H.J. Huang, W.K. Cavenee, O.D. Wiestler, T. Pietsch, N. Wernert, Expression of the Ets-1 transcription factor in human astrocytomas is associated with Fms-like tyrosine kinase-1 (Flt-1)/vascular endothelial growth factor receptor-1 synthesis and neoangiogenesis. Cancer Res. 59 (1999) 5608- 5614.

[359] S.H. Lee, J.H. Bahn, C.K. Choi, N.C. Whitlock, A.E. English, S. Safe, S.J. Baek, ESE-1/EGR-1 pathway plays a role in tolfenamic acid-induced apoptosis in colorectal cancer cells. Mol.Cancer Ther. 7 (2008) 3739-3750.

293

[360] I. Ferreiro, M. Barragan, A. Gubern, E. Ballestar, M. Joaquin, F. Posas, The p38 SAPK is recruited to chromatin via its interaction with transcription factors. J.Biol.Chem. 285 (2010) 31819-31828.

[361] G.C. Kujoth, D.F. Robinson, W.E. Fahl, Binding of ETS family members is important for the function of the c-sis/platelet-derived growth factor-B TATA neighboring sequence in 12-O-tetradecanoylphorbol-13-acetate-treated K562 cells. Cell Growth Differ. 9 (1998) 523-534.

[362] G.K. Scott, J.C. Daniel, X. Xiong, R.A. Maki, D. Kabat, C.C. Benz, Binding of an ETS-related protein within the DNase I hypersensitive site of the HER2/neu promoter in human breast cancer cells. J.Biol.Chem. 269 (1994) 19848-19858.

[363] N. Sato, J.G. Beitz, J. Kato, M. Yamamoto, J.W. Clark, P. Calabresi, A. Raymond, A.R. Frackelton, Jr., Platelet-derived growth factor indirectly stimulates angiogenesis in vitro. Am.j.Pathol. 142 (1993) 1119-1130.

[364] R. Kumar, R. Yarmand-Bagheri, The role of HER2 in angiogenesis. Semin.Oncol. 28 (2001) 27-32.

[365] A. Hamik, B. Wang, M.K. Jain, Transcriptional regulators of angiogenesis. Arterioscler.Thromb.Vasc.Biol. 26 (2006) 1936-1947.

[366] S. Oh, S. Shin, R. Janknecht, ETV1, 4 and 5: an oncogenic subfamily of ETS transcription factors. Biochim.Biophys.Acta 1826 (2012) 1-12.

[367] J.H. Chen, C.D. Wright, PEA3, Oct 1 and Oct 2 positively regulate the human ETS1 promoter. Oncogene 8 (1993) 3375-3383.

[368] J.L. Kopp, P.J. Wilder, M. Desler, J.H. Kim, J. Hou, T. Nowling, A. Rizzino, Unique and selective effects of five Ets family members, Elf3, Ets1, Ets2, PEA3, and PU.1, on the promoter of the type II transforming growth factor-beta receptor gene. J.Biol.Chem. 279 (2004) 19407-19420.

[369] C.J. Lockwood, G. Krikun, V.A. Hausknecht, C. Papp, F. Schatz, Matrix metalloproteinase and matrix metalloproteinase inhibitor expression in endometrial stromal cells during progestin-initiated decidualization and menstruation-related progestin withdrawal. Endocrinology 139 (1998) 4607-4613.

[370] Y. Izumiyama, Y. Ohiro, F. Higashino, K. Yoshida, K. Taguchi, S. Todo, T. Kohgo, Y. Totsuka, M. Shindoh, E1AF expression is closely correlated with malignant phenotype of tongue squamous cell carcinoma through activation of MT1-MMP gene promoters. Oncol.Rep. 13 (2005) 715-720.

[371] Z. Ma, M.J. Chang, R. Shah, J. Adamski, X. Zhao, E.N. Benveniste, Brg-1 is required for maximal transcription of the human matrix metalloproteinase-2 gene. J.Biol.Chem. 279 (2004) 46326-46334.

294

[372] H.C. Crawford, B. Fingleton, M.D. Gustavson, N. Kurpios, R.A. Wagenaar, J.A. Hassell, L.M. Matrisian, The PEA3 subfamily of Ets transcription factors synergizes with beta-catenin-LEF-1 to activate matrilysin transcription in intestinal tumors. Mol.Cell Biol. 21 (2001) 1370-1383.

[373] H. Habelhah, F. Okada, M. Kobayashi, K. Nakai, S. Choi, J. Hamada, T. Moriuchi, M. Kaya, K. Yoshida, K. Fujinaga, M. Hosokawa, Increased E1AF expression in mouse fibrosarcoma promotes metastasis through induction of MT1-MMP expression. Oncogene 18 (1999) 1771-1776.

[374] S. Horiuchi, H. Yamamoto, Y. Min, Y. Adachi, F. Itoh, K. Imai, Association of ets-related transcriptional factor E1AF expression with tumour progression and overexpression of MMP-1 and matrilysin in human colorectal cancer. J.Pathol. 200 (2003) 568-576.

[375] C. Cai, C.L. Hsieh, J. Omwancha, Z. Zheng, S.Y. Chen, J.L. Baert, L. Shemshedini, ETV1 is a novel androgen receptor-regulated gene that mediates prostate cancer cell invasion. Mol.Endocrinol. 21 (2007) 1835-1846.

[376] J. Jiang, Y. Wei, D. Liu, J. Zhou, J. Shen, X. Chen, S. Zhang, X. Kong, J. Gu, E1AF promotes breast cancer cell cycle progression via upregulation of Cyclin D3 transcription. Biochem.Biophys.Res.Commun. 358 (2007) 53-58.

[377] D.Y. Jun, M.K. Kim, I.G. Kim, Y.H. Kim, Characterization of the murine cyclin D2 gene: exon/intron organization and promoter activity. Mol.Cells 7 (1997) 537- 543.

[378] R.B. Hamanaka, N.S. Chandel, Mitochondrial reactive oxygen species regulate cellular signaling and dictate biological outcomes. trends biochem.sci. 35 (2010) 505-513.

[379] D.X. Zhang, D.D. Gutterman, Mitochondrial reactive oxygen species-mediated signaling in endothelial cells. Am.J.Physiol Heart Circ.Physiol 292 (2007) H2023-H2031.

[380] C. Urbich, A. Kuehbacher, S. Dimmeler, Role of microRNAs in vascular diseases, inflammation, and angiogenesis. Cardiovasc.Res. 79 (2008) 581-588.

[381] A. Bonauer, R.A. Boon, S. Dimmeler, Vascular microRNAs. Curr.Drug Targets. 11 (2010) 943-949.

[382] C. Stahlhut, Y. Suarez, J. Lu, Y. Mishima, A.J. Giraldez, miR-1 and miR-206 regulate angiogenesis by modulating VegfA expression in zebrafish. Development 139 (2012) 4356-4364.

[383] N. Abdel-Malak, R. Harfouche, S.N. Hussain, Transcriptome of angiopoietin 1- activated human umbilical vein endothelial cells. Endothelium 14 (2007) 285- 302.

295

[384] N. Selvaraj, V. Kedage, P.C. Hollenhorst, Comparison of MAPK specificity across the ETS transcription factor family identifies a high-affinity ERK interaction required for ERG function in prostate cells. Cell Commun.Signal. 13 (2015) 12.

[385] Y. Sato, Role of ETS family transcription factors in vascular development and angiogenesis. Cell Struct.Funct. 26 (2001) 19-24.

[386] B.M. Emerling, L.C. Platanias, E. Black, A.R. Nebreda, R.J. Davis, N.S. Chandel, Mitochondrial reactive oxygen species activation of p38 mitogen-activated protein kinase is required for hypoxia signaling. Mol.Cell Biol. 25 (2005) 4853- 4862.

[387] R. Koziel, H. Pircher, M. Kratochwil, B. Lener, M. Hermann, N.A. Dencher, P. Jansen-Durr, Mitochondrial respiratory chain complex I is inactivated by NADPH oxidase Nox4. Biochem.J. 452 (2013) 231-239.

[388] A.K. Doughan, D.G. Harrison, S.I. Dikalov, Molecular mechanisms of angiotensin II-mediated mitochondrial dysfunction: linking mitochondrial oxidative damage and vascular endothelial dysfunction. Circ.Res. 102 (2008) 488- 496.

[389] R. Koziel, H. Pircher, M. Kratochwil, B. Lener, M. Hermann, N.A. Dencher, P. Jansen-Durr, Mitochondrial respiratory chain complex I is inactivated by NADPH oxidase Nox4. Biochem.J. 452 (2013) 231-239.

[390] R.M. Liu, J. Choi, J.H. Wu, K.A. Gaston Pravia, K.M. Lewis, J.D. Brand, N.S. Mochel, D.M. Krzywanski, J.D. Lambeth, J.S. Hagood, H.J. Forman, V.J. Thannickal, E.M. Postlethwait, Oxidative modification of nuclear mitogen- activated protein kinase phosphatase 1 is involved in transforming growth factor beta1-induced expression of plasminogen activator inhibitor 1 in fibroblasts. J.Biol.Chem. 285 (2010) 16239-16247.

[391] G. Xi, X.C. Shen, C. Wai, D.R. Clemmons, Recruitment of Nox4 to a plasma membrane scaffold is required for localized reactive oxygen species generation and sustained Src activation in response to insulin-like growth factor-I. J.Biol.Chem. 288 (2013) 15641-15653.

[392] S.K. Wyman, E.C. Knouf, R.K. Parkin, B.R. Fritz, D.W. Lin, L.M. Dennis, M.A. Krouse, P.J. Webster, M. Tewari, Post-transcriptional generation of miRNA variants by multiple nucleotidyl transferases contributes to miRNA transcriptome complexity. Genome Res. 21 (2011) 1450-1461.

[393] A. Chamorro-Jorganes, M.Y. Lee, E. Araldi, S. Landskroner-Eiger, M. Fernandez-Fuertes, M. Sahraei, R.M. Quiles Del, S.C. van, J. Yu, C. Fernandez- Hernando, W.C. Sessa, Y. Suarez, VEGF-Induced Expression of miR-17-92 Cluster in Endothelial Cells Is Mediated by ERK/ELK1 Activation and Regulates Angiogenesis. Circ.Res. 118 (2016) 38-47. 296

[394] R. Echavarria, D. Mayaki, J.C. Neel, S. Harel, V. Sanchez, S.N. Hussain, Angiopoietin-1 inhibits toll-like receptor 4 signalling in cultured endothelial cells: role of miR-146b-5p. Cardiovasc.Res. 106 (2015) 465-477.

[395] G. Bergers, D. Hanahan, Modes of resistance to anti-angiogenic therapy. Nat.Rev.Cancer 8 (2008) 592-603.

[396] A. Abdollahi, J. Folkman, Evading tumor evasion: current concepts and perspectives of anti-angiogenic cancer therapy. Drug Resist.Updat. 13 (2010) 16- 28.

[397] R. Schreck, K. Albermann, P.A. Baeuerle, Nuclear factor kappa B: an oxidative stress-responsive transcription factor of eukaryotic cells (a review). Free Radic.Res.Commun. 17 (1992) 221-237.

[398] J.E. Deanfield, J.P. Halcox, T.J. Rabelink, Endothelial function and dysfunction: testing and clinical relevance. Circulation 115 (2007) 1285-1295.

[399] K.E. Sorensen, D.S. Celermajer, D. Georgakopoulos, G. Hatcher, D.J. Betteridge, J.E. Deanfield, Impairment of endothelium-dependent dilation is an early event in children with familial hypercholesterolemia and is related to the lipoprotein(a) level. J.Clin.Invest 93 (1994) 50-55.

[400] N. Gokce, J.F. Keaney, Jr., L.M. Hunter, M.T. Watkins, J.O. Menzoian, J.A. Vita, Risk stratification for postoperative cardiovascular events via noninvasive assessment of endothelial function: a prospective study. Circulation 105 (2002) 1567-1572.

[401] P.O. Bonetti, L.O. Lerman, A. Lerman, Endothelial dysfunction: a marker of atherosclerotic risk. Arterioscler.Thromb.Vasc.Biol. 23 (2003) 168-175.

[402] H. Lum, K.A. Roebuck, Oxidant stress and endothelial cell dysfunction. Am.J.Physiol Cell Physiol 280 (2001) C719-C741.

[403] P.M. Vanhoutte, Say NO to ET. J.Auton.Nerv.Syst. 81 (2000) 271-277.

[404] V.B. O'Donnell, P.H. Chumley, N. Hogg, A. Bloodsworth, V.M. Darley-Usmar, B.A. Freeman, Nitric oxide inhibition of lipid peroxidation: kinetics of reaction with lipid peroxyl radicals and comparison with alpha-tocopherol. Biochemistry 36 (1997) 15216-15223.

[405] E. Hall, Z.L. Brookes, Angiopoietin-1 increases arteriolar vasoconstriction to phenylephrine during sepsis. Regul.Pept. 131 (2005) 34-37.

[406] N.D. Tsihlis, C.S. Oustwani, A.K. Vavra, Q. Jiang, L.K. Keefer, M.R. Kibbe, Nitric oxide inhibits vascular smooth muscle cell proliferation and neointimal hyperplasia by increasing the ubiquitination and degradation of UbcH10. Cell Biochem.Biophys. 60 (2011) 89-97.

297

[407] G.N. Rao, B.C. Berk, Active oxygen species stimulate vascular smooth muscle cell growth and proto-oncogene expression. Circ.Res. 70 (1992) 593-599.

[408] B. Lassegue, M.A. San, K.K. Griendling, Biochemistry, physiology, and pathophysiology of NADPH oxidases in the cardiovascular system. Circ.Res. 110 (2012) 1364-1390.

[409] N.R. Nayak, C.J. Kuo, T.A. Desai, S.J. Wiegand, B.L. Lasley, L.C. Giudice, R.M. Brenner, Expression, localization and hormonal control of angiopoietin-1 in the rhesus macaque endometrium: potential role in spiral artery growth. Mol.Hum.Reprod. 11 (2005) 791-799.

[410] J.W. Franses, N.C. Drosu, W.J. Gibson, V.C. Chitalia, E.R. Edelman, Dysfunctional endothelial cells directly stimulate cancer inflammation and metastasis. Int.J.Cancer 133 (2013) 1334-1344.

[411] D.A. Long, K.L. Price, E. Ioffe, C.M. Gannon, L. Gnudi, K.E. White, G.D. Yancopoulos, J.S. Rudge, A.S. Woolf, Angiopoietin-1 therapy enhances fibrosis and inflammation following folic acid-induced acute renal injury. Kidney Int. 74 (2008) 300-309.

298