THE ROLE OF NOTCH SIGNALING IN VASCULAR DEVELOPMENT AND HOMEOSTASIS by Linda Ya-ting Chang B.Sc., University of British Columbia, 2003 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (Experimental Medicine) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) July 2010 © Linda Ya-ting Chang, 2010 ii ABSTRACT The vasculature is essential for the delivery of oxygen and nutrients and the removal of metabolic wastes from tissues of the body. The embryonic vasculature is developed through the processes of vasculogenesis, angiogenesis, and arteriogenesis. Once the vasculature is fully developed and stabilized, the adult vasculature shows very little proliferation or cell death. Nevertheless, the endothelium, which lines the lumen of the blood vessels, is actively involved in the control of vascular tone, permeability, blood flow, coagulation, inflammation and tissue repair. An injury to the endothelium is important for progression of diseases such as atherosclerosis and the sepsis syndrome. The Notch signaling pathway has emerged in the recent decade as an important player in multiple vascular processes and endothelial behaviors. This thesis examines the role of the Notch signaling pathway in embryonic arteriogenesis and endothelial survival signaling. The first part of this thesis investigates the developmental source of vascular smooth muscle cells. This study presents the first in situ observation of an immediate smooth muscle precursor cell present in all embryonic arteries. This Tie1+/CD31+/VE-cadherin- precursor requires Notch signaling to differentiate into vascular smooth muscle cells and to ensure vascular stability of newly formed arteries. However, Notch activation is not required in the precursor cells to maintain the medial layer of the arteries once the vessel is invested with vascular smooth muscle cells. In the second part of this thesis, the mechanism of Notch-induced endothelial survival signaling is examined. In endothelial cells, Notch signaling activates phosphotidylinositol-3 kinase (PI3K) through up-regulation of a secreted factor. Activity of PI3K is required to offset the parallel apoptotic signaling induced by Notch activation and to maintain endothelial survival through the up-regulation of Slug, a direct Notch target with anti-apoptotic activity. Upon treatment with apoptotic stimuli, Notch activation shows context-dependent effects on iii endothelial survival. Inhibition of PI3K activity and Slug expression by a stimulus abolishes Notch-induced endothelial survival and increases apoptotic death. The work presented in this thesis shows that the Notch signaling pathway is essential for the stability of the vasculature through regulation of vascular smooth muscle cell differentiation and endothelial cell survival. iv TABLE OF CONTENTS ABSTRACT … ....................................................................................................................... ii TABLE OF CONTENTS ........................................................................................................... iv LIST OF TABLES .................................................................................................................... viii LIST OF FIGURES................................................................................................................... ix LIST OF ABBREVIATIONS ...................................................................................................... xi ACKNOWLEDGEMENTS ...................................................................................................... xiv Chapter 1 INTRODUCTION ...................................................................................................... 1 1.1 Vascular development and homeostasis .................................................................... 2 1.1.1 Vasculogenesis and angiogenesis ...................................................................... 4 1.1.2 Embryonic arteriogenesis .................................................................................... 6 18.104.22.168 Sources of vascular smooth muscle cells during development .................... 8 1.1.3 Endothelial homeostasis in pathogenesis ......................................................... 13 22.214.171.124 Apoptosis .................................................................................................... 14 126.96.36.199 Diseases associated with endothelial apoptosis ........................................ 15 1.1.4 Signaling pathways in vascular development and homeostasis ........................ 17 1.2 Notch signaling pathway ........................................................................................... 18 1.2.1 Notch signaling overview ................................................................................... 18 1.2.2 Modifiers of Notch signaling .............................................................................. 22 1.2.3 Experimental methods of blocking Notch signaling in vivo ................................ 24 1.3 Notch signaling in the vasculature ............................................................................ 26 1.3.1 Notch in endothelial cell specification ................................................................ 26 188.8.131.52 Notch in endothelial arterial-venous specification ...................................... 27 1.3.2 Notch in angiogenesis and vascular remodeling ............................................... 30 184.108.40.206 Tip and stalk cell specification .......................................................................... 31 220.127.116.11 Notch in other endothelial functions ................................................................. 32 1.3.3 Notch signaling in smooth muscle cell development ......................................... 34 1.3.4 Notch and endothelial survival ........................................................................... 41 1.4 Aims of the studies ................................................................................................... 43 v Chapter 2 MATERIALS AND METHODS ............................................................................... 45 2.1 Cell culture..................................................................................................................... 45 2.1.1 Gene transfer .......................................................................................................... 45 2.2 Transgenic mice ............................................................................................................ 47 2.2.1 Timed matings ........................................................................................................ 48 2.3 Flow cytometry .............................................................................................................. 49 2.4 Immunofluorescence staining ........................................................................................ 50 2.4.1 Smooth muscle thickness quantification ................................................................. 51 2.5 β-galactosidase detection .............................................................................................. 53 2.6 Apoptosis/survival Assays ............................................................................................. 54 2.6.1 Annexin V binding assay ......................................................................................... 54 2.6.2 Neutral Red uptake assay ....................................................................................... 54 2.6.3 Activated caspase 3 detection ................................................................................ 55 2.7 Immunoblotting .............................................................................................................. 56 2.8 Real-time PCR ............................................................................................................... 57 2.9 Statistical analysis ......................................................................................................... 59 Chapter 3 VASCULAR SMOOTH MUSCLE DIFFERENTIATION FROM TIE1+ PRECURSORS REQUIRES NOTCH .............................................................. 60 3.1 Introduction .................................................................................................................... 60 3.2 VSMC are derived from a Tie1+CD31+VE-cadherin- precursor cell ............................... 63 3.3 A tissue-specific, inducible transgenic model for Notch inhibition ................................. 73 3.4 Blockade of Notch signaling in Tie1-positive precursors leads to hemorrhage localized to newly-forming vasculature ............................................................................................... 77 3.5 Notch signaling is required for differentiation of Tie1-positive precursors into vascular smooth muscle cells ............................................................................................................ 84 3.6 Notch activation is not required in Tie1+ progenitors after VSMC fate is acquired ....... 87 3.7 Discussion ..................................................................................................................... 91 vi Chapter 4 NOTCH ACTIVATION PROMOTES ENDOTHELIAL SURVIVAL THROUGH A PI3K-SLUG AXIS ............................................................................................. 94 4.1 Introduction .................................................................................................................... 94 4.2 Notch protects against LPS, but not homocysteine,-induced apoptosis ........................ 97 4.3 Notch signaling activates the PI3K pathway through a secreted factor ......................... 99 4.4 PI3K activity is essential for survival of Notch-activated endothelial cells ................... 101 4.5 PI3K activity is required for Notch-induced Slug expression ....................................... 103 4.6 Slug protects endothelial cells against LPS-induced apoptosis .................................. 109 4.7 Homocysteine induces apoptosis in Notch-activated cells by regulating PI3K and Slug .......................................................................................................................................... 111 4.8 Discussion ................................................................................................................... 113 4.8.1 Possible candidates for a Notch-induced pro-apoptotic signal ............................. 113 4.8.2 Possible mechanisms of Slug down-regulation by PI3K inhibition ........................ 115 Chapter 5 SUMMARY, PERSPECTIVES, AND FUTURE DIRECTIONS ............................. 117 5.1 Notch in arteriogenesis ................................................................................................ 117 5.1.1 Establishment of a new tool: tetOSdnMAML transgenic mouse ........................... 118 5.1.2 Characterization of the VSMC precursor .............................................................. 119 5.1.3 Clinical implication of Notch-induced VSMC differentiation .................................. 121 5.1.4 Possible involvement of other cell types in arteriogenesis .................................... 122 5.2 Notch and survival signaling ........................................................................................ 124 5.2.1 Mechanisms for Notch-induced PI3K activation ................................................... 124 5.2.2 Homocysteine and Slug expression ...................................................................... 125 5.2.3 Alternative methods of Notch activation ................................................................ 126 BIBLIOGRAPHY ................................................................................................................... 128 APPENDICES ....................................................................................................................... 143 vii Appendix A. Ethics approvals ............................................................................................ 143 Appendix B. List of publications ......................................................................................... 150 Appendix C. Previously published material ....................................................................... 152 viii LIST OF TABLES Table 1.1 Mammalian arterial and venous endothelial markers during vascular development ...................................................................................................... 5 Table 1.2 Signaling pathway activation by homocysteine and lipopolysaccharide .......... 16 Table 1.3 Vascular development processes affected in Notch pathway mutant mice ..... 40 Table 2.1 Primers for quantitative RT-PCR ..................................................................... 57 Table 3.1 Summary of embryo phenotypes ..................................................................... 79 ix LIST OF FIGURES Figure 1.1 Embryonic vascular development ...................................................................... 3 Figure 1.2 Mosaic origin of vascular smooth muscle cells ................................................ 11 Figure 1.3 A simplified view of Notch signaling pathway .................................................. 21 Figure 2.1 Smooth muscle thickness quantification .......................................................... 52 Figure 3.1 The tetracycline-inducible, endothelial-specific transgenic system .................. 64 Figure 3.2 Endothelial expression of β-galactosidase reporter ......................................... 65 Figure 3.3 Tie1 promoter, but not VE-cadherin (VE) promoter, is active in peri-endothelial cells .................................................................................................................. 67 Figure 3.4 Endogenous Tie1 promoter shows activity in peri-endothelial cells ................. 68 Figure 3.5 Tie1-positive cells show characteristics of VSMC precursor cells ................... 70 Figure 3.6 Precursor cells are enriched in the LacZ-/CD31+ population in the VEtTA:TetOSLacZ embryos ............................................................................ 72 Fiugre 3.7 Expression of dnMAML blocks Notch-induced target expression .................... 73 Figure 3.8 Inhibition of Notch in developing endothelium causes embryonic lethality ...... 74 Figure 3.9 Expression of dominant-negative Mastermind-like1 in endothelial cells leads to blockade of Notch signaling ............................................................................. 76 Figure 3.10 Blocking Notch signaling in Tie1-positive cells leads to localized hemorrhaging ......................................................................................................................... 78 Figure 3.11 Tie1tTA and VEtTA transgenic mice in C57BL/6J background behave similarly to the original albino strains ............................................................................. 80 Figure 3.12 Expression of dnMAML construct is comparable between Tie1tTA:dnMAML and VEtTA:dnMAML embryos ......................................................................... 82 Figure 3.13 Tie1 promoter drives dnMAML expression in endothelial cells and CD31dim perivascular cells ............................................................................................. 83 Figure 3.14 Blockade of Notch signaling Tie1+ precursor cells impedes vascular smooth muscle differentiation in vivo ............................................................................ 85 Figure 3.15 Expression of dnMAML in E10.5 embryos decreases the percentage of PDGFR-β positive cells .................................................................................... 86 x Figure 3.16 Blocking Notch signaling in Tie1+ precursor cells does not affect VSMC coverage of already established arteries ......................................................... 88 Figure 3.17 Effect of Notch blockade is more evident in distal portion of carotid artery compared to region proximal to the aorta ........................................................ 90 Figure 3.18 Proposed role of Notch activation in embryonic arteriogenesis ....................... 93 Figure 4.1 Notch activation protects endothelial cells against LPS-induced apoptosis while enhancing homocysteine-induced apoptosis ................................................... 98 Figure 4.2 Notch activation in endothelial cells leads to activation of PI3K signaling through a secreted factor ............................................................................... 100 Figure 4.3 PI3K pathway activity is necessary for the survival of Notch-activated endothelial cells ............................................................................................. 102 Figure 4.4 PI3K activity is required for Notch-induced Slug expression ......................... 104 Figure 4.5 Akt or mTOR activity is not required for Notch-induced Slug expression ..... 106 Figure 4.6 PI3K inhibition decreases both basal and Notch-induced Slug transcript level ................................................................................................ 108 Figure 4.7 Slug protects endothelial cells against LPS-induced apoptosis while exhibiting no effect on homocysteine-induced apoptosis ............................................... 110 Figure 4.8 Homocysteine blocks Notch-induced PI3K activation and Slug expression .. 112 Figure 4.9 Model of Notch-induced survival signaling ..................................................... 114 xi LIST OF ABBREVIATIONS ADAM a disintegrin and metelloprotease domain AGM aorta-gonad-mesonephros APH1 anterior pharynx-defective 1 AVC atrio-ventricular canal AVM arteriovenous malformation CADASIL cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy caspase cysteinyl aspartyl-directed proteases CD cluster of differentiation CDK8 cyclin-dependent kinase-8 CIR CBF1-interacting co-repressor COUP-TFII chicken ovalbumin upstream promoter transcription factor II CSL CBF-1, Suppressor of Hairless Su(H), Lag-1 Dll delta-like ligand dnMAML dominant-negative form of mastermind-like1 E10.5 embryonic day 10.5 ECM extracellular matrix EGF epidermal growth factor EndMT endothelial-to-mesenchymal transdifferentiation ER endoplasmic reticulum ES cell embryonic stem cell FACS fluorescence activated cell sorting FBXL14 F-box and leucine-rich repeat protein 14 FGF fibroblast growth factor Flk-1 fetal liver kinase 1 xii GAPDH glyceraldehyde 3-phosphase dehydrogenase GCN5 general control of amino-acid synthesis GFP green fluorescent protein GOF gain of function Grl gridlock GSI gamma-secretase inhibitor GSK-3β glycogen synthase kinase -3 beta HAT histone acetyl-transferases HDAC histone deacetylase complex Hes hairy/enhancer of Split Hey hairy/enhancer of Split related with YRPW HMEC human dermal microvascular endothelial cells HUVEC human umbilical vein endothelial cells IAP inhibitor of apoptosis protein ISV intersomitic vessel LDL low density lipoprotein LOF loss of function LPS lipopolysaccharide MAML mastermind-like MCM minichromosome maintenance Mdm2 mouse double minute 2 MEF2D myocyte-specific enhancer factor 2D Msx2 muscle segment homeobox 2 mTOR mammalian target of rapamycin Myf-5 myogenic factor 5 MyoD myogenic differentiation antigen 1 NFκB nuclear factor of kappa-light-chain in B-cells xiii NLS nuclear localization signal P10 post-natal day 10 PCAF p300/CBP-associated factor PDGFR platelet-derived growth factor receptor PEN2 presenilin enhancer 2 PI3K phosphotidylinositol-3 kinase Pofut protein o-fucosyl transferase Ppa partner of paired PTEN phosphatase and tensin homolog RNAi RNA interference ROS reactive oxygen species RT-PCR reverse transcription polymerase chain reaction SKIP ski-interacting protein SMA α-smooth muscle actin Smad3 mothers against decapentaplegic homolog 3 SMRT silencing mediator for retinoic acid and thyroid hormone receptor SRF serum response factor TACE tumor necrosis factor-a converting enzyme TGF-β transforming growth factor-beta TNF tumor necrosis factor VE-cadherin vascular endothelial cadherin VEGF vascular endothelial growth factor VEGFR-2 vascular endothelial growth factor receptor-2 VSMC vascular smooth muscle cells YFP yellow fluorescent protein xiv ACKNOWLEDGEMENTS I would like to thank my parents, Jei-seng and Chang-hua Chang, for their support and understanding. I would not be able to dedicate myself wholeheartedly to my studies if you haven’t been there for me, even if sometimes I was not there for you. I would also like to thank my sister Judy, for being my “lay-person” and editor through all the scholarship applications and her constant reminder that “science is cool”. I am grateful for my supervisor Dr. Aly Karsan for all these years of guidance, support and stimulating conversations. I have learnt so much about science in the environment that he has provided me. His excitement towards science is one of the many things that I will take away from this experience. I would like to thank our collaborators Drs. Daniel Dumont and Mira Puri for providing the transgenic mice used in this thesis and the many valuable comments. Thank you to Dr. Laura Benjamin also for providing the VEtTA transgenic mouse. I would also like to thank my supervisory committee members, Drs. Kelly McNagny, Catherine Pallen and Andrew Weng for their thought-provoking comments, useful suggestions, and dedication to my graduate school education and my projects. Many thanks to Denise MacDougal for her assistance with flow cytometry and friendship over the years. Also I would like to thank the staff at the BCCRC flow core, animal resource center and transgenic core facilities for their expert help. I am also grateful for the members of my laboratory, past and present, for their unfailing friendship, support, help and sessions of scientific musings through the years. I would not have enjoyed science and the graduate school experience without them. There are really memories that would last a lifetime. They have helped to build my skill set and confidence as a scientist-in-training. Thanks to Megan Fuller, Michelle Ly, Fred Wong and xv Michelle Higginson also for their help with my projects. I especially like to thank Dr. Michela Noseda, whose dedication and enthusiasm to science is something I marvel at. Without her inspiration, I would not be where I am today. Finally, I express my gratitude towards the Michael Smith Foundation of Health Research for providing the Graduate Studentships, which allow me to attend conferences and purchase useful reference books. Thanks to the Stem Cell Network, the Heart and Stroke Foundation of BC & Yukon, and the Canadian Institutes of Health Research for funding the research projects. 1 Chapter 1. INTRODUCTION The vasculature is one of the most important and complex organs in mammals and is essential for the delivery of oxygen and nutrients and the removal of metabolic wastes from tissues of the body. The vasculature is the first functional system to form during development and its disruption during development leads to early embryonic death. The development of vascular system is a carefully orchestrated event that results in a network of vessels with various calibers and functions. Multiple cell types are generated as the result of vascular development: the endothelial cells which form the inner vessel wall and the mural cells which associate with and coat the endothelial tubes. In the normal adult vasculature, there is very little endothelial cell proliferation or cell death. However, the homeostasis of the adult vasculature requires active maintenance. Endothelial survival signaling is important due to the constant assault from apoptotic stimuli in the blood stream encountered by the endothelium. One of the signaling pathways involved in regulating the vasculature is the Notch signaling pathway, whose role has been demonstrated in cell type specification, contact inhibition and homeostasis establishment (reviewed by (Iso, Hamamori et al. 2003; Karsan 2005; Gridley 2007)). This dissertation will discuss the role of Notch signaling in the establishment of the vasculature in the embryo as well as the maintenance of a stable vascular system in the adult. 2 1.1. Vascular development and homeostasis Formation of the cardiovascular system is an early and essential process during embryogenesis (Figure 1.1). The development of the vasculature initiates with the differentiation of endothelial cells, which are the specialized epithelial cells lining the lumen of the blood vessels, from mesodermal precursor cells called angioblasts. The cluster of endothelial cells then assembles into the primary capillary plexus, a network of uniformly sized endothelial tubes. This process is termed vasculogenesis (reviewed in (Flamme, Frolich et al. 1997; Drake, Hungerford et al. 1998)). This primitive plexus is then remodeled into a network of blood vessels of various diameters by the process of angiogenesis, during which vessels are pruned and new vessels are formed by either intussusception or sprouting. During angiogenic sprouting, the extracellular matrix (ECM) is broken down, cells sprout from the pre-existing endothelial tube, and endothelial cells then proliferate, migrate, and form into a new tubular structure (Folkman 1984). Finally, the new endothelial tube is stabilized by deposits of ECM and the recruitment of mural cells. In the case of artery formation, the mature arteries are surrounded by layers of vascular smooth muscle cells (VSMC) by the process of arteriogenesis (Carmeliet 2000). Interestingly, once stabilized, the adult vasculature is quiescent (Hobson and Denekamp 1984) with the exception of wound healing, menstruation (Reynolds, Killilea et al. 1992) or in pathological angiogenesis, such as in tumor growth (Hanahan and Folkman 1996) and rheumatoid arthritis (Colville-Nash and Scott 1992). The homeostasis of the mature vasculature is actively maintained by signaling pathways as the endothelium is constantly under stimulation by factors in the blood and the surrounding tissue. 3 Figure 1.1. Embryonic vascular development. Endothelial precursors, called angioblasts, are differentiated from mesodermal progenitors. Angioblasts are further specified to arterial or venous fate. Angioblasts differentiate into endothelial cells, which form the endothelial tubes of the primary vascular plexus. The primary plexus undergoes angiogenic remodeling to form a vascular system complete with arteries, veins, capillaries, and other vessels of various calibers. The nascent blood vessels are stabilized by the recruitment of mural cells such as pericytes or vascular smooth muscle cell. Mesodermal progenitor Angioblasts Arterial angioblasts Venous angioblasts Primary vascular plexus Mural cells recruitment Angiogenic remodeling Mural cells Endothelium 4 1.1.1. Vasculogenesis and angiogenesis The site of the first vascular structure formed during embryogenesis is the yolk sac, where mesodermal precursors from the primitive streak of the embryo migrate to form the blood island (Ferguson, Kelley et al. 2005). Angioblast differentiation in the blood island results in endothelial-lined lumen filled with primitive blood cells (Haar and Ackerman 1971). These endothelial cords then fuse to form the primary capillary plexus of the yolk sac. The extraembryonic vasculogenesis is temporally and spatially distinct from the intraembryonic vasculogenesis (Ferguson, Kelley et al. 2005). In the embryo proper, endothelial cell precursors also reside in the mesodermal layer in early embryonic development, more specifically in the paraxial and the lateral plate mesoderm (Wilting and Becker 2006). There is evidence of multipotent progenitor cells derived from the mesoderm that can differentiate into endothelial cells, such as the hemangioblasts (Vogeli, Jin et al. 2006) or fetal liver kinase-1 (Flk-1)-expressing cardiovascular progenitors (Yang, Soonpaa et al. 2008). These progenitors migrate to the site of vessel formation taking cues from the surrounding tissues. Endothelial specification of the angioblast is followed by lumen formation to generate an endothelial tube capable of containing the circulating blood (Coffin, Harrison et al. 1991). The first vessels formed through vasculogenesis in the embryo proper are the dorsal aorta and the cardinal vein (Flamme, Frolich et al. 1997). The arterial endothelium is phenotypically and functionally different from the venous endothelium. The arteries bring oxygenated blood at high pressure from the heart to the rest of the organism through smaller caliber arterioles and capillaries. The veins return the low- oxygen blood back to the pulmonary system. It was long believed that variation in hemodynamic force and oxygenation from the blood flow differentiates the arterial system from the venous system. Interestingly, “arterial” and “venous” angioblasts with differential arterial and venous markers (see Table 1.1) were observed from the mesoderm prior to the establishment of blood flow, suggesting that the specification between artery and veins is 5 regulated by genetic factors as well as blood flow (Wang, Chen et al. 1998; Adams, Wilkinson et al. 1999). Many of the markers specifying arteriovenous fate are components of signaling pathways, such as Notch/Delta-like ligand (Dll) and Eph receptor/Ephrin, proposing the involvement of those signaling pathways in the arteriovenous cell fate decision. Table 1.1. Mammalian arterial and venous endothelial markers during vascular development Arterial endothelial marker Venous endothelial marker EphrinB2 (Wang, Chen et al. 1998) EphB4 (Wang, Chen et al. 1998) Neuropilin1 (Herzog, Kalcheim et al. 2001) Neuropilin2 (Herzog, Kalcheim et al. 2001) Notch4 (Villa, Walker et al. 2001) COUP-TFII (You, Lin et al. 2005) Dll4 (Villa, Walker et al. 2001) CD44 (Wheatley, Isacke et al. 1993) The primary capillary plexus is a network of endothelial tubes with uniform diameters, which is further remodeled into a functional system of vessels of various calibers by angiogenesis. This process is readily observed in the formation of yolk sac vasculature and the embryonic cephalic vasculature (Coffin and Poole 1988). Vascular remodeling involves the regression of branches from large vessels like the artery, vessels splitting by intussusceptive growth and new capillary sprouting from pre-existing vessels. In the embryo proper, the intersomitic vessels are formed by new capillaries sprouting from the dorsal aorta (Poole and Coffin 1989). Angiogenic sprouting requires a sequence of highly coordinated endothelial activities, starting with the disruption of ECM around the formation of the new sprout, followed by the guided migration and proliferation of endothelial cells, the re- establishment of the ECM, and finally the stabilization of the nascent vessel with the recruitment of mural cells, such as pericytes and VSMC (Carmeliet 2003). The endothelial cell is an anchorage-dependent cell type, for which contact with ECM or neighboring cells is required for survival. During sprouting, additional survival signaling will be necessary as the endothelial cells obtain a migratory morphology and break away from their neighbors (Liu, 6 Ahmad et al. 2000). Angiogenic sprouting is regulated by cytokines from the surrounding tissue, bidirectional signaling between mural cells and the endothelium, and cell-cell contact dependent signaling between adjacent endothelial cells. The initiation of a new angiogenic sprout requires a specialized endothelial cell, called a tip cell. This cell is more “explorative” than the endothelial cells still residing in the monolayer of the original vessel. It is characterized by the appearance of filopodia, which are outstretched actin filament-containing protrusions that also contain vascular endothelial growth factor receptor-2 (VEGFR-2) (Gerhardt, Golding et al. 2003). The tip cell has the ability to respond to the changing gradient of vascular endothelial growth factor (VEGF) and guide the new sprout by migrating towards the area of high VEGF concentration. Following the tip cell is the stalk cell, which proliferates and elongates the sprout as the tip cell migrates (Gerhardt, Golding et al. 2003). The stalk cell also forms the lumen and later recruits mural cells. While tip cell formation is required for the generation of a new sprout, excessive tip cell specification may create a defective vascular network (Suchting, Freitas et al. 2007). In order for productive angiogenesis to occur, there must be signals present to direct the tip cell and stalk cell fate. These signals will be discussed in later sections. 1.1.2. Embryonic arteriogenesis The process of arteriogenesis is essential for the establishment of a proper embryonic vasculature. VSMC play an important role in the overall stability of the vasculature. During development, they provide structural support for the nascent arteries and are also involved in bidirectional signaling with the endothelium. In the absence of VSMC or pericytes, developing murine vasculature becomes unstable and exhibits a hemorrhagic phenotype along with embryonic lethality (Hungerford and Little 1999; Li, Sorensen et al. 1999). Diseases like hereditary hemorrhagic telangiectasia may be affected by poor VSMC development (Li, Sorensen et al. 1999). 7 VSMC development occurs between embryonic day 9.5 (E9.5) and E11.5 for major arteries (Takahashi, Imanaka et al. 1996). A detailed study on the differentiation of VSMC during early murine vascular development was performed with immunohistochemistry labeling of α-smooth muscle actin (SMA) (Takahashi, Imanaka et al. 1996). The first artery to associate with VSMC in the embryo proper is the dorsal aorta starting at E9.5. The ventral portion of the dorsal aorta is the first to recruit smooth muscle covering, followed by the dorsal region. Smooth muscle development is also more advanced in the thoracic region of the fused dorsal aorta compared to the paired dorsal aortae either rostral or caudal to it. VSMC development of the internal carotid artery initiates at E10.5 with only discontinuous VSMC coverage (Takahashi, Imanaka et al. 1996). The pharyngeal arch arteries also exhibit discontinuous SMA staining in the ventral region at E10.5, while there is still no VSMC around the endothelial tube in the extremities, such as the caudal artery of the tail (Takahashi, Imanaka et al. 1996). One day later, at E11.5, most arteries, including the vertebral artery and some veins are covered with at least a discontinuous layer of VSMC. VSMC of developing vessels present themselves as a heterogeneous population with a wide range of phenotypes at different stages of differentiation for different vascular beds (Owens, Kumar et al. 2004). As more studies are performed to examine the origin of embryonic VSMC, it becomes clear that the heterogeneity may be generated by the mosaic origin of the VSMC precursors (Majesky 2007). This observation adds layers of complexity in deciphering the molecular and cellular signaling pathways involved in VSMC differentiation as each developmental source may utilize different mechanisms to become mature VSMC. However, the work presented in this thesis proposes a local common VSMC precursor whose differentiation is regulated through Notch signaling. 8 18.104.22.168. Sources of vascular smooth muscle cells during development From lineage-tracing experiments using either a genetic reporter or cross-species- transplants, various embryonic tissues have been shown to be sources of VSMC during development (reviewed in (Gittenberger-de Groot, DeRuiter et al. 1999; Majesky 2007)) (Figure 1.2). While most of the early experiments were done in avian models, there has been an emergence of mammalian (mostly murine) models in recent years. In addition, the use of embryonic stem cells has also advanced our knowledge of progenitor cell differentiation and the molecular signaling pathways involved. The combination of both lines of investigation has shed light on the mosaic origin of VSMC. The first study to identify a tissue-specific source of VSMC focused on the avian neural crest cells, and found that this progenitor cell type incorporated into the cardiovascular system as VSMC of the ascending aorta, the aortic arch and the carotid arteries and the mesenchyme of the septum dividing the aorta and the pulmonary artery (Le Lievre and Le Douarin 1975; Kirby, Gale et al. 1983). Later studies using cell fate tracing during mammalian development confirmed the observations made in chick embryos (Jiang, Rowitch et al. 2000; Li, Chen et al. 2000). It is especially noteworthy that the boundary between neural crest- derived VSMC and VSMC from other sources is very distinct as no neural crest-derived VSMC are found in the neighboring subclavian arteries, coronary arteries and descending aorta (Jiang, Rowitch et al. 2000; Li, Chen et al. 2000). VSMC that arise from a distinct source are also found in the walls of coronary arteries. Cell fate tracing in avian development using cells transduced with β-galactosidase viral vector showed that progenitor cells from the pro-epicardium can give rise to cells of the mature epicardium, the coronary endothelium and also the VSMC of the coronary arteries (Mikawa and Gourdie 1996). This observation was confirmed in later studies using chick-quail chimera (Gittenberger-de Groot, Vrancken Peeters et al. 1998; Perez-Pomares, Macias et al. 1998) or 9 adenovirus tagged epicardium (Dettman, Denetclaw et al. 1998). Again, the boundary of pro- epicardium-derived VSMC is very distinct. The VSMC of the ascending aorta, which is physically connected to the coronary artery, are from another source. The cells of the vasculature come from the mesodermal germ layer of the developing embryo. During avian development, both the splanchnic mesoderm (Wiegreffe, Christ et al. 2009) and the somites (Pouget, Gautier et al. 2006) have been shown to differentiate into VSMC of the dorsal aorta. However, the processes are mutually exclusive and sequential. The splanchnic mesoderm contains progenitor cells that can differentiate into endothelial cells, smooth muscle cells and hematopoietic cells (Pardanaud, Luton et al. 1996). During early embryogenesis, the dorsal aorta is constructed by splanchnic mesoderm-derived endothelial cells and VSMC (Hungerford, Owens et al. 1996). As the vessel matures, the dorsal endothelium is replaced by somite-derived endothelial cells (Pouget, Gautier et al. 2006). Somite-derived endothelial and VSMC will proceed to replace splanchnic mesoderm- derived cells in the dorsal aorta as development advances. Cell fate tracing with either a lateral plate mesoderm-specific promoter or a paraxial mesoderm-specific promoter in murine embryos reveals similar observations (Wasteson, Johansson et al. 2008). Clonal analysis in murine embryogenesis also shows that VSMC of the descending aorta share common somite-derived precursor cells with the dorsal skeletal muscle (Esner, Meilhac et al. 2006). While the somitic precursors migrate ventrally towards the dorsal aorta, it does not migrate in the rostral-caudal axis (Esner, Meilhac et al. 2006), suggesting that VSMC of different segments of the dorsal aorta differentiate independently from each other. In the adult thoracic and abdominal aorta, cell tracing also shows that aortic VSMC are of somitic origin (Wasteson, Johansson et al. 2008). These studies suggest that in both murine and avian development, VSMC of the descending aorta first comes from the splanchnic mesoderm and are later replaced by somitic cells. 10 Examination of these tissue-derived VSMC showed a common theme: within a tissue, only a subset of cells has the ability to differentiate into VSMC. During avian dorsal aorta development, while cells from the mesoderm can migrate to the site of dorsal aorta formation, only a small percentage of mesoderm cells actually integrate as a part of the dorsal aorta (Pouget, Gautier et al. 2006), suggesting the existence of a vascular progenitor cell within the tissue that is predisposed towards a vascular fate. Several studies have been done to isolate and culture VSMC progenitor cells from mammalian sources. Among them, the mesoangioblasts (Minasi, Riminucci et al. 2002) and the embryonic stem cell (ES cell)-derived cardiovascular progenitors (Yamashita, Itoh et al. 2000; Yang, Soonpaa et al. 2008) both express the same marker, Flk-1, a receptor for VEGF. Mesoangioblasts are mesodermal progenitors isolated from E9.5 mouse embryonic dorsal aorta (Minasi, Riminucci et al. 2002). Mesoangioblast can incorporate into developing chick embryos as VSMC, skeletal muscle, cardiomyocytes and osteocytes, showing multipotency in the mesodermal lineages (Minasi, Riminucci et al. 2002). The mesoangioblasts also expresses hemo-angioblastic markers such as cluster of differentiation (CD)34, MEF2D (myocyte-specific enhancer factor 2D) and cKit and a subpopulation of mesoangioblasts is also positive for SMA expression (Minasi, Riminucci et al. 2002). 11 Figure 1.2. Mosaic origin of vascular smooth muscle cells. Vascular smooth muscle cells (VSMC) are derived from different tissue origin depending on the vascular bed. The arterial VSMC of the coronary artery comes from the pro-epicardial origin. The ascending aorta, the aortic arch and the carotid arteries are derived from the neural crest cells. The descending aorta is first populated by VSMC of the splanchnic mesoderm origin, which are later replaced by somite-derived VSMC. However, VSMC of the caudal region of the abdominal aorta remains to be of the splanchnic mesoderm origin. Pro-epicardium origin Neural crest origin Somitic origin Splanchnic mesoderm rostral caudal ventral dorsal 12 Flk-1 positive progenitor cells can also be generated from differentiating ES cells and can further differentiate into endothelial cells, mural cells and cardiomyocytes (Kattman, Huber et al. 2006). Flk-1 positive cells can also form vascular tubes in three-dimensional culture and be incorporated into the developing chick vasculature as both endothelial cells and mural cells (Yamashita, Itoh et al. 2000). When the Flk-1 promoter is used in lineage tracing experiments, endothelial cells, hematopoietic cells, cardiomyocytes and surprisingly, skeletal muscle cells, are found to come from Flk-1 positive progenitors; however, no VSMC was observed to have come from Flk-1+ cells (Motoike, Markham et al. 2003). A similar Flk-1 positive progenitor can be isolated from differentiating human ES cells by a cytokine- regulated protocol (Yang, Soonpaa et al. 2008). Both progenitor cells can be expanded ex vivo and are capable of incorporating into developing avian vasculature as endothelial and smooth muscle cells. However, there is still a lack of direct in vivo evidence and in situ tracking of an immediate vascular progenitor cell that can differentiate into VSMC within the mammalian development system. Anatomically, the endothelium is the cell type most closely associated with VSMC. Interestingly, the endothelium has been shown to be a possible source of VSMC through the process of endothelial-mesenchymal transdifferentiation (EndMT) in an avian developmental model (DeRuiter, Poelmann et al. 1997). Endothelial cell fate tracing with dye-uptake and co- localization of endothelial and smooth muscle markers suggests that the smooth muscle layer closest to the endothelium may come from an endothelial source (Arciniegas, Neves et al. 2005). In pulmonary hypertension, EndMT has also been suggested to be a mechanism for pathological arterial neointimal thickening (Arciniegas, Frid et al. 2007). However, there are no studies examining the endothelium as a possible source of smooth muscle progenitor cells during mammalian development. Experiments from our lab and others have shown that mammalian endothelial cells can be induced to undergo EndMT in vitro (Noseda, McLean et 13 al. 2004; Deissler, Lang et al. 2006; Kitao, Sato et al. 2009). This thesis will further investigate the in vivo significance of this observation. 1.1.3. Endothelial homeostasis in pathogenesis Once the cardiovascular system is fully developed, the adult vasculature remains seemingly quiescent with the exception of physiological processes like wound healing and menstruation. In adult macrovessels, only 0.1% of the endothelial cells are undergoing mitosis (Schwartz and Benditt 1977), while endothelial programmed cell death (apoptosis) is a rare event in healthy individuals (Alvarez, Gips et al. 1997). Still, the endothelium is actively involved in the control of vascular tone, permeability, blood flow, coagulation, inflammation and tissue repair (reviewed in (Cines, Pollak et al. 1998)). Endothelial dysfunction has been linked to a variety of cardiovascular diseases. Interestingly, when signaling pathways required for vascular development are disrupted in the adult endothelium, this can result in disturbances in vascular homeostasis and sometimes even death. For example, adult mice with endothelial-specific inactivation of the VEGF gene show increased mortality compared to littermate control mice (Lee, Chen et al. 2007). The VEGF mutants exhibit systemic vascular pathologies such as hemorrhages, abnormal accumulation of ECM and appearance of microinfarcts in multiple organs. Examination of the endothelium in VEGF mutants shows morphological changes typical of apoptosis (Lee, Chen et al. 2007). Disruption of vascular homeostasis from deregulated apoptosis and growth has systemic and detrimental effect on the overall health of the individual. This thesis will examine the mechanism of endothelial survival when stimulated with agents that damage the vasculature. 14 22.214.171.124. Apoptosis Apoptosis, often referred to as “programmed cell death”, can be morphologically identified by the plasma-membrane vesiculation (also called membrane blebbing), nuclear condensation, nuclear fragmentation and an appearance of overall cell shrinkage of an apoptotic cell (Kerr, Wyllie et al. 1972). The cellular material is encapsulated within vesicles (called apoptotic bodies) and is quickly phagocytosed without leading to an inflammatory response (Kerr, Wyllie et al. 1972). Apoptosis can be triggered by either an environmental source (extrinsic) or an intracellular stress (intrinsic) (reviewed by (Adams 2003)). The extrinsic apoptotic pathway often leads to activation of the intrinsic mechanism and both pathways result in the activation of a family of cysteinyl aspartyl-directed proteases (caspases), where a cascade of caspase activation leads to cleavages of many intracellular proteins during apoptosis (Thornberry and Lazebnik 1998). Pro-apoptotic stimuli can activate apoptotic pathways by interacting with the death receptor pathways, inducing expression or activity of proteins involved in the intrinsic pathway, or inhibiting survival signaling pathways within the cell. On the other hand, anti-apoptotic stimuli can activate survival pathways and/or induce expression of anti-apoptotic protein within the cell, such as the proteins in the Inhibitor of Apoptosis Protein (IAP) family or the Bcl-2 family. Maintenance of the adult vasculature is especially challenging because the endothelium is under constant assault of stimuli that are present in the blood. The survival of individual endothelial cells and the maintenance of homeostasis require a local balance of pro- and anti- apoptotic agents. Each of these stimuli has the capability to activate intracellular signaling events within the endothelial cells. Therefore, an intracellular balance of pro- and antiapoptotic signals must also be reached to prevent unwanted activation of the apoptotic cascade (Stefanec 2000). Endothelial apoptosis in the vasculature is associated with the initiation and/or progression of serious diseases, often leading to mortality. Thus, by broadening our understanding of the mechanisms by which different stimuli cause endothelial 15 apoptosis, we may be able to perturb the progression of vascular disease by maintaining endothelial integrity. 126.96.36.199. Diseases associated with endothelial apoptosis Many of the commonly known risk factors of cardiovascular disease are pro-apoptotic in endothelial cells. Therefore, it is not surprising that endothelial apoptosis exacerbates the progression of some diseases. This section introduces two distinct endothelial apoptotic stimuli examined in this thesis and their associated vascular disorders. Atherosclerosis is a disease that can lead to cardiac infarct, stroke or peripheral vascular disease. It is characterized by the build-up of the atherosclerotic plaque at the vessel wall. Over time, the plaque can grow and constrict the vessel, causing reduced blood flow. Endothelial apoptosis has been observed both in the atherosclerotic plaque and just downstream of the plaque in patients suffering from carotid atherosclerosis (Alvarez, Gips et al. 1997; Tricot, Mallat et al. 2000). Moreover, apoptotic endothelial cells show increased adherence to platelets and leukocytes (Bombeli, Schwartz et al. 1999; Schwartz, Karsan et al. 1999), which may contribute to the progression of plaque formation. One of the risk factors for atherosclerosis is hyperhomocysteinemia (Clarke, Daly et al. 1991; McCully 1996), where patients experience an elevated plasma concentration of total homocysteine. Homocysteine is a metabolic product in the conversion between methionine and cysteine. Serum homocysteine levels can be increased through genetic mutation of enzymes in the homocysteine metabolic pathway or through dietary deficiency of vitamin B’s required for homocysteine metabolism. High level of homocysteine can induce endothelial apoptosis by increasing intracellular reactive oxygen species (ROS) (Lee, Kim et al. 2005), upregulating p53 and Noxa, increasing endoplasmic reticulum (ER) stress (Zhang, Cai et al. 2001), activating the intrinsic apoptotic pathway through mitochondria destabilization (Tyagi, Ovechkin et al. 2006), and decreasing signaling through the phosphotidylinositol 3-kinase 16 (PI3K) pathway (Suhara, Fukuo et al. 2004). Homocysteine-induced endothelial apoptosis is most likely the result of interaction between multiple pro-apoptotic signaling pathways (see Table 1.2). Table 1.2. Signaling pathway activation by homocysteine and lipopolysaccharide. Intracellular reactive oxygen species p53 pathway Intrinsic apoptosis pathway PI3K pathway Homocysteine Increase ROS (+)1 Increase p53 activity (+) Activates intrinsic apoptotic pathway (+) Decrease PI3K activity (-) Lipopolysaccharide Increase ROS (+) Increase NO (-)2 Increase p53 acitivity (+) Activates intrinsic apoptotic pathway (+) Increase PI3K activity (+) Another human disease that is associated with endothelial death is sepsis. Sepsis is a systemic inflammatory disorder whose complications include systemic vascular collapse, multi-organ failure and acute respiratory distress (Bannerman and Goldblum 2003). Endotoxin, also known as lipopolysaccharide (LPS), is present on gram-negative bacteria and is a mediator of the sepsis syndrom (Parrillo 1993). LPS-stimulation induces endothelial apoptosis in vitro (Bannerman and Goldblum 2003). Endothelial apoptosis leads to detachment of cells from the vessel, activation of the coagulation pathway, and increase in vascular permeability. These endothelial defects may exacerbate the effect of LPS, especially in the lung, where respiratory distress can be caused by edema. When a caspase inhibitor is administrated after LPS injection, there is a reduced level of acute lung injury and decreased endothelial apoptosis (Kawasaki, Kuwano et al. 2000). LPS activates apoptosis in endothelial cells through generation of ROS and activation of both the extrinsic (Bannerman, Tupper et al. 2001) and intrinsic apoptotic pathways (Munshi, Fernandis et al. 2002; Wang, Akinci et al. 1 Increase in pathway activation 2 Decrease in pathway activation 17 2007). However, LPS-induced endothelial apoptosis is more readily observed with inhibition of endogenous anti-apoptotic proteins FLIP and Mcl-2 (Bannerman, Tupper et al. 2001). Interestingly, LPS stimulation in endothelial cells also activates pro-survival pathways such as nitric oxide synthesis (Huang, Kuo et al. 1998) and PI3K activation (Wong, Hull et al. 2004). The overall apoptotic phenotype in LPS-stimulated cells may require other pro-apoptotic signals to tip the balance between two opposing signals (Table 1.2). 1.1.4. Signaling pathways in vascular development and homeostasis Many signaling pathways are involved in generation of the complex vascular system, and some of these pathways are also required to maintain adult vessel homeostasis. The fibroblast growth factor (FGF) pathway is important for the specification of mesodermal progenitors and angioblasts (Cox and Poole 2000; Ciruna and Rossant 2001). Signaling through the VEGF pathway is required for multiple steps in vascular development, including angioblast migration, endothelial tube formation, and arterial endothelial cell specification (Coultas, Chawengsaksophak et al. 2005). In the adult, the VEGF signal is required for endothelial survival and regulation of vascular permeability (Lee, Chen et al. 2007). The Eph receptors and their ligands in the Ephrin family play integral roles in the arteriovenous specification of angioblast and endothelial cells (Wang, Chen et al. 1998; Adams, Wilkinson et al. 1999). The transforming growth factor-β (TGF-β) pathway is required for proper vessel patterning during angiogenesis and pericyte recruitment (Darland and D'Amore 2001; Pardali and ten Dijke 2009). The Tie/Angiopoietin signaling between mural cells and endothelial cells induces sprouting angiogenesis (Koblizek, Weiss et al. 1998) and enhances vessel stability. Lastly, examination of mutants in the Notch signaling pathways shows that Notch activation is essential for vascular development by regulating multiple endothelial processes (Iso, Hamamori et al. 2003). This thesis will examine the role of Notch signaling in both vascular smooth muscle development and endothelial cell survival. 18 1.2. Notch signaling pathway Development is the spatially and temporally controlled process whereby the complexity of a multi-cellular organism is built from one single cell. There are surprisingly few signaling pathways which govern the multifaceted processes of development (Gerhart 1999), and those that exist are often evolutionarily well-conserved (Pires-daSilva and Sommer 2003). One of these pathways is the Notch signaling pathway. Many components of this pathway have been shown to be conserved through the Metazoan lineage, from worms to humans (Gazave, Lapebie et al. 2009). At different stages of development, Notch signaling is found to be essential for processes that include asymmetric cell-fate decision, boundary formation, and lateral inhibition (reviewed by (Bray 1998; Artavanis-Tsakonas, Rand et al. 1999; Baron, Aslam et al. 2002; Hurlbut, Kankel et al. 2007)), all of which are involved in creating the diversity of cell types and their organization in an organism. 1.2.1. Notch signaling overview There are four identified Notch receptors in mammals, Notch1-4. The receptor is first translated as one single 300 kDa polypeptide, which is then processed by a furin-like convertase in the trans-Golgi network (Blaumueller, Qi et al. 1997; Logeat, Bessia et al. 1998). The processed receptor is expressed on the cell surface as a transmembrane heterodimer held together non-covalently by a calcium-dependent interaction (Rand, Grimm et al. 2000). Notch signaling is activated when transmembrane receptors interact with transmembrane ligands (Jagged1/2 and Dll1/3/4 in mammals) on neighboring cells. This physical interaction via the epidermal growth factor (EGF)-like repeats in the extraceullular domain of the receptor (Rebay, Fleming et al. 1991) is cell-cell contact dependent. Receptor- ligand interaction leads to the cleavage of the Notch receptor at an extracellular site (termed S2) by metalloprotease TACE (TNF-α converting enzyme; also known as ADAM17, a member of a disintegrin and metalloprotease domain family) (Brou, Logeat et al. 2000) or 19 Kuzbanian (Kuz, or ADAM10) (Rooke, Pan et al. 1996). Evidence supporting the subsequent transendocytosis of the extracellular domain of Notch receptor into the ligand-expressing cells exists (Parks, Klueg et al. 2000; Morel, Le Borgne et al. 2003), although it is unknown whether this process is required for Notch receptor activation. Following the cleavage at S2, the Notch receptor is further processed by the γ- secretase complex. Presenilin is the enzymatic subunit of γ-secretase, which also contains nicastrin, PEN2 (presenilin enhancer 2) and APH1 (anterior pharynx-defective 1) (Ray, Yao et al. 1999; Goutte, Tsunozaki et al. 2002; Hu, Ye et al. 2002; Lopez-Schier and St Johnston 2002; Kimberly, LaVoie et al. 2003). This second proteolytic cleavage in the transmembrane domain of Notch receptor (S3) releases the intracellular domain of Notch (NotchIC) into the cytoplasm (De Strooper, Annaert et al. 1999). NotchIC, which contains two putative nuclear localization signals (NLS) (Stifani, Blaumueller et al. 1992), translocates to the nucleus where it interacts with DNA-binding protein CSL (CBF-1, Suppressor of Hairless Su(H), Lag-1) (Tamura, Taniguchi et al. 1995). The nuclear localization of NotchIC has been shown to be required for its activity (Kopan, Nye et al. 1994; Schroeter, Kisslinger et al. 1998). In the absence of Notch activation, CSL is a transcriptional repressor (Dou, Zeng et al. 1994; Waltzer, Logeat et al. 1994) that binds to the DNA sequence 5’-C(T)GTGGGAA-3’ with high affinity (Tun, Hamaguchi et al. 1994) and recruits co-repressor proteins SMRT/N-CoR (silencing mediator for retinoic acid and thyroid hormone receptor/nuclear repressor), CIR (CBF1-interacting co-repressor), and SKIP (Ski-interacting protein) to target promoters. The CSL-co-repressor complex then recruits histone deacetylase complex (HDAC) to deacetylate histones at the promoter, leading to repression of target transcription (Kao, Ordentlich et al. 1998; Hsieh, Zhou et al. 1999). Interaction between NotchIC and CSL leads to the displacement of the co-repressors and the recruitment of co-activator Mastermind-like (MAML) (Jeffries, Robbins et al. 2002; Wu, Sun et al. 2002), which in turn recruits other co- activators such as CBP/p300 (Oswald, Tauber et al. 2001; Fryer, Lamar et al. 2002), PCAF 20 (p300/CBP-associated factor) and GCN5 (general control of animo-acid synthesis) (Kurooka and Honjo 2000), all of which have histone acetyl-transferase (HAT) activity. The co-activator complex opens up the chromatin and allows active transcription of target genes (summarized in Figure 1.3). The target genes of Notch signaling are highly cell type-dependent. There are two classically defined families of Notch targets, the Hes (hairy/enhancer of Split) and the Hey (hairy/enhancer of Split related with YRPW) transcription repressors (Iso, Sartorelli et al. 2001). From in vitro experiments in endothelial cells Hes1, Hey1, and Hey2 have all been shown to be Notch signaling targets (Shawber, Das et al. 2003; MacKenzie, Duriez et al. 2004). In addition, EphrinB2, an arterial specific marker, is also a target of Notch activation in endothelial cells (Shawber, Das et al. 2003). Recently, studies on the role of Notch activation in EndMT have also identified many mesenchymal genes as targets of Notch signaling (Noseda, Fu et al. 2006; Jin, Hansson et al. 2008; Niessen, Fu et al. 2008). I will discuss the Notch-induced mesenchymal transformation in more detail in later sections 21 Figure 1.3. A simplified view of the Notch signaling pathway. The receptor is processed by a furin-like convertase in the trans-Golgi network. The processed receptor is expressed on the cell surface as a transmembrane heterodimer. Notch signaling is activated when the receptors (Notch1-4 in mammals) interact with transmembrane ligands (Jagged1/2 and Dll1/3/4 in mammals) on neighboring cells through the extraceullular domain of the receptor. Receptor-ligand interaction leads to the cleavage of the Notch receptor by metalloprotease TACE. There is evidence supporting the subsequent trans-endocytosis of the extracellular domain of Notch receptor into the ligand-expressing cells. Following the cleavage at S2, Notch receptor is further processed by the γ-secretase complex. This second proteolytic cleavage releases the intracellular domain of Notch (NotchIC) into the cytoplasm. NotchIC translocates to the nucleus after it is untethered from the membrane. Once in the nucleus, NotchIC then interacts with DNA-binding protein CSL. CSL, in the absence of Notch activation, is a transcription repressor. Interaction between NotchIC and CSL leads to the displacement of the co-repressors and the recruitment of co-activator MAML. The co-activator complex opens up the chromatin and allows active transcription of target genes. There are two classically defined families of Notch targets, the Hes and the Hey transcription repressors. CSL CSL MAML CoA CoRFurin-like convertase TACE γ-secretase complex Jagged/Delta -like ligands Notch receptors NotchIC 22 1.2.2. Modifiers of Notch signaling Since Notch plays an important role in development where spatiotemporal control of signals is essential, activation of the pathway must be tightly regulated. Interestingly, the expression of the Notch receptors is found to be ubiquitous while receptor activation and downstream signaling show stringent regulation. One method by which Notch signaling is modulated is through ligand expression. At the initiation of angiogenesis, VEGF activates VEGF receptors, which in turn up-regulate transcription of Dll4 in endothelial cells. Thus, through activation of Notch receptor on neighboring endothelial cells by cell-to-cell contact, the Dll4-expressing cell can regulate the angiogenic process (Lobov, Renard et al. 2007). In addition to the cleavage by a furin-like convertase, other modifications are made to Notch receptors in the ER and the trans-Golgi network. The extracellular domain of Notch is o-fucosylated by protein o-fucosyl transferase (Pofut) in the ER (Okajima and Irvine 2002). Pofut expression is also required for proper protein folding and cellular membrane sorting of the receptor (Sasamura, Ishikawa et al. 2007). The Pofut1-null mouse shows phenotypes indicative of systemic Notch signaling defects (Okamura and Saga 2008). After the addition of the first fuctose, additional modifications can be made by extending the carbohydrate chain. Through glycosylation of EGF repeats, the Fringe family of glycosyl transferases can alter the affinity of Notch receptor to ligands and favor signals from Dll at the expense of Jagged ligands (Panin, Papayannopoulos et al. 1997; Bruckner, Perez et al. 2000). In this fashion, additional regulation of the Notch signal can be made without altering the expression of ligands or receptors. Another method of regulating ligand activity is through post-translational modification of the ligands. Expression of E3 ligases Mind bomb and Neuralized are required for Notch ligand activity (Itoh, Kim et al. 2003). Mind bomb and Neuralized are involved in the mono- ubiquitination and Epsin-mediated endocytosis of Notch ligands Jagged and Delta-like. This 23 post-translational modification is necessary for ligand-induced Notch activation. One possible mechanism is that some additional modifications to the ligands are necessary, and endocytosis drives the process by localizing the ligand and enzymes in the same endosome compartment to facilitate ligand modification. The second scenario is that ligand ubiquitination and endocytosis is required for creating a mechanical strain on the receptor upon ligand binding. This mechanical pulling force may lead to conformational changes in the receptor, revealing the S2 site for TACE cleavage. The transendocytosis of cleaved Notch receptor that follows may then enable γ-secretase cleavage of the remaining membrane-tethered receptor (Pratt, Wentzell et al. 2010). The membrane-localized receptors are also ubiquitinated and endocytosed. Endocytosis of Notch receptor can either lead to activation or de-activation of the receptor. Notch receptor is mono-ubiquitinated after TACE cleavage before it can be cleaved by the γ- secretase complex (Gupta-Rossi, Six et al. 2004). Impaired endocytosis blocks Notch signaling independent of ligand-activation (Vaccari, Lu et al. 2008). However, the requirement for this process in overall Notch activation is still unclear. Receptor endocytosis can also negatively regulate Notch signaling by the degradation of unused receptor in the proteasome. E3 ligases such as Numb have been implicated to ubiquitinate Notch, which leads to degradation of Notch receptor (McGill, Dho et al. 2009). Once the NotchIC-coactivator complex is assembled in the nucleus and transcription of the targets is activated, there must be a way to dampen the signal when it is no longer required. The co-activator complex recruits kinases such as cyclin-dependent kinase 8 (CDK8) (Fryer, White et al. 2004) and glycogen synthase kinase-3β (GSK-3β) (Espinosa, Ingles-Esteve et al. 2003); NotchIC has been shown to be a substrate for both kinases. Phosphorylation of NotchIC leads to its eventual ubiquitination and degradation by SEL10 E3 ligase (Oberg, Li et al. 2001; Wu, Lyapina et al. 2001). The presence of these modulator molecules spatiotemporally regulates a ligand-specific activation of Notch receptor. 24 1.2.3. Experimental methods of blocking Notch signaling in vivo Notch activation requires the relay of signals from the surface of neighboring cells to the promoters of the target genes in the nucleus. The multi-step process allows for the development of tools to interfere with the transduction of the signal at multiple points in order to study the endogenous function of this pathway. In the study of mammalian development and adult biological processes, transgenic and knock-out mouse models are widely used, although each has its advantages and pitfalls. These models will be discussed in a later section. Chemical or antibody-based inhibition of Notch signaling also holds promise for therapeutic application. However, the inevitable off-target effects of chemical inhibitors need to be closely examined. Both receptors and ligands of the Notch signaling pathway are transmembrane proteins whose functions are dependent on their extracellular domains. Recent animal studies have used antibodies against either Notch receptors or ligands to block the interaction and activation of the pathway. In mouse tumor models, Dll4 antibody is found to deregulate angiogenesis and block tumor growth in one model (Ridgway, Zhang et al. 2006), while decreasing the tumor-initiating population in another model (Hoey, Yen et al. 2009). Antibodies against Notch1 (Aste-Amezaga, Zhang et al. 2010) and Notch4 (Dontu, Jackson et al. 2004) receptors have been shown to block signaling in vitro. Antibody treatment is a specific method for signal inactivation, however, accessibility of the tissue and the systemic nature of the treatment may be concerns. Notch signaling can also be blocked by chemical inhibitors for γ-secretase (Geling, Steiner et al. 2002). Gamma-secretase inhibitors (GSI) were first developed as a treatment for Alzheimer’s disease. The use of GSI as a Notch blocker has been widely applied and accepted in in vitro experiments and has also been used in animal experiments involving rodent models. Since it requires systemic administration, Notch signaling in multiple systems 25 is affected. The most often observed side effect of GSI treatment is the expansion of intestinal goblet cells (Milano, McKay et al. 2004; Wong, Manfra et al. 2004). Overall, systemic GSI treatment is not a specific method of Notch inhibition, but small molecule inhibitors may provide better tissue penetration and diffusion compared to blocking antibodies. Several dominant negative molecules of the Notch pathway components have been developed. While they are widely used for in vitro experiments, few constructs are used for transgenic mouse generation. Truncated Dll1 is shown to have dominant negative activity in vertebrae formation (Cordes, Schuster-Gossler et al. 2004). Dominant-negative CSL and MAML1 are both shown to block CSL-dependent Notch signaling and cause accelerated differentiation of hematopoietic stem cells (Duncan, Rattis et al. 2005). One of the most widely used dominant negative constructs is the truncated MAML1 (Tu, Fang et al. 2005; Maillard, Tu et al. 2006; High, Zhang et al. 2007; Proweller, Wright et al. 2007; Santos, Sarmento et al. 2007), which only contains the Notch and CSL interaction domain, without the ability to bind to other co-activators (Maillard, Weng et al. 2004). MAML1, 2, and 3 have different levels of capability to interact with the intracellular domain of different Notch receptors (Wu, Sun et al. 2002). However, a dominant-negative form of MAML1 has the highest efficiency in blocking signals from all four Notch receptors and it has been shown to be a competent pan-notch inhibitor (Maillard, Weng et al. 2004). As a part of this thesis, an inducible, tissue-specific, dominant-negative MAML transgenic system is generated to inactivate Notch signaling in the endothelium. 26 1.3. Notch signaling in the vasculature Considering that the Notch phenotype was first observed in Drosophila as the appearance of notched wings in a loss-of-function (LOF) mutant followed by its identification as a neurogenic gene (Wharton, Johansen et al. 1985), it is interesting that many of the LOF mouse mutants of the Notch signaling pathway display cardiovascular defects (reviewed in (Iso, Hamamori et al. 2003; Alva and Iruela-Arispe 2004)). Notch receptors Notch1, Notch2, Notch3, and Notch4 and ligands Jagged1, Jagged2, Dll1, and Dll4 are all expressed in developing vasculature (Villa, Walker et al. 2001; Varadkar, Kraman et al. 2008; Sorensen, Adams et al. 2009). However, upon closer examination, the pattern and timing of expression are variable within the group of Notch pathway genes, suggesting that Notch signaling plays multiple roles in the establishment of a stable vasculature (summarized in Table 1.3). 1.3.1. Notch in endothelial cell specification Notch signaling has been shown, in various animal models, to be involved in endothelial specification in the dorsal aorta. In zebrafish development, blocking Notch signaling with GSI increases the endothelial population while decreasing the hematopoietic cell number (Lee, Vogeli et al. 2009), suggesting that Notch activation functions as a molecular switch between the two cell types, blocking endothelial differentiation in favor of hematopoietic lineages. In early avian development, Notch activation in the ventral mesoderm favors differentiation towards mural cell fate at the expense of the endothelial/hematopoietic lineages (Shin, Nagai et al. 2009). However, later in development, migration, integration and differentiation of somite-derived endothelial precursors into the dorsal aorta have all been shown to require Notch activation (Sato, Watanabe et al. 2008; Ohata, Tadokoro et al. 2009). These seemingly contradictory observations either represent the differences in molecular signaling between species or the ability of Notch signaling to regulate endothelial cell specification at different stages of differentiation. 27 In the mammalian system, there has been no detailed in vivo study of the role of Notch signaling in the specification of endothelial cells from precursor cells. In vitro ES cell differentiation studies have shown that Notch1 activation in ES cells blocks differentiation towards Flk-1 positive mesodermal progenitors (Schroeder, Meier-Stiegen et al. 2006). Activation of Notch4 in ES cell-derived Flk-1 positive mesodermal progenitor drives differentiation away from the hematopoietic fate and towards a cardiovascular fate (Chen, Stull et al. 2008). Both studies suggest a role for Notch signaling in the formation of endothelial cells in the mammalian system. However, in the LOF Notch mutants, the endothelial primary plexus is formed normally in the yolk sac (Alva and Iruela-Arispe 2004; Shawber and Kitajewski 2004). The main blood vessels, the dorsal aorta and the cardinal vein in the embryo proper, which are both established through the process of vasculogenesis, still contain endothelial cells. In the most severely affected Notch1/Notch4 double knockout mutant, although no observable lumen can be found in either the dorsal aorta or the cardinal vein, clusters of endothelial cells are found in the correct location (Krebs, Xue et al. 2000). It shows that the process of angioblast differentiation into endothelial cells is not affected during murine development. 188.8.131.52. Notch in endothelial arterial-venous specification Nevertheless, Notch activation still has an important role in endothelial cell specification in mammalian vascular development. The first clue that Notch signaling is regulating arterial- venous differentiation comes from the gridlock (grl) mutant in zebrafish. The grl gene is the zebrafish homolog of the mammalian hey2 gene, a target of Notch signaling. Gridlock is expressed in the mesoderm prior to the formation of the dorsal aorta and axial vein. Upon the formation of the vessels, the expression of Gridlock in the vasculature is restricted to the dorsal aorta, not in the axial vein (Zhong, Rosenberg et al. 2000) and the grl mutant contains defects only in the arteries, not veins (Weinstein, Stemple et al. 1995). Another Notch pathway mutant mindbomb also showed decreased expression of the arterial marker 28 EphrinB2 in the developing dorsal aorta and ectopic expression of the venous marker flt4 instead (Lawson, Scheer et al. 2001). The grl and mindbomb mutant phenotypes suggest that Notch signaling is required for arterial function of the newly formed dorsal aorta. In mammalian development, the expression of Notch receptors, Notch1 and Notch4, and ligands, Jagged1, Dll1 and Dll4, has been localized mainly in the arterial system, not in veins (Villa, Walker et al. 2001). Notch activation during embryonic vascular development has been restricted to the arteries as well (Souilhol, Cormier et al. 2006; Del Monte, Grego-Bessa et al. 2007). With LOF and gain-of-function (GOF) mutants of the arterially-restricted Notch receptors and ligands, the role of Notch in arterial-venous differentiation during mammalian development becomes evident. Notch1 null mice exhibit embryonic lethality before E11.5. The dorsal aorta appears collapsed (Swiatek, Lindsell et al. 1994) and there is decreased expression of arterial markers (Fischer, Schumacher et al. 2004). Conversely, Transgenic mice expressing activated Notch1 (Shawber, Funahashi et al. 2007; Venkatesh, Park et al. 2008) or Notch4 (Krebs, Xue et al. 2000) receptor exhibit aberrant expression of arterial markers in the veins and arteriovenous malformation (AVM), where there is fusion between the dorsal aorta and the anterior cardinal vein. These studies show that the two Notch receptors expressed in embryonic arteries are essential for the arterial identity and proper arterial-venous specification during development. Of the three Notch ligands expressed in embryonic arteries, each has different role in arterializations in vascular development. Dll4 heterozygous knock-in embryos die in utero at E10.5 (Duarte, Hirashima et al. 2004; Gale, Dominguez et al. 2004; Krebs, Shutter et al. 2004). Dll4+/- mutants exhibit a thinner dorsal aorta and an absence of internal carotid arteries while the venous plexus remains relatively normal (Gale, Dominguez et al. 2004). The arteries lack smooth muscle coverage and express reduced levels of arterial markers (Duarte, Hirashima et al. 2004; Gale, Dominguez et al. 2004). Interestingly, there is ectopic expression of venous marker EphB4 in the dorsal aorta (Duarte, Hirashima et al. 2004), 29 suggesting either that the venous specification is the default fate in the absence of arterial- driven signal or that Notch activation is actively suppressing venous fate. Conversely, overexpression of Dll4, either systemic or endothelial-specific, induces EphrinB2 in the anterior cardinal vein while down-regulation of EphB4 is observed (Trindade, Kumar et al. 2008). Due to the early lethality of the Dll4 mutant, arterializations in older embryos cannot be examined. Interestingly, the expression of another Delta-like ligand, Dll1, is induced in embryonic artery starting at E13.5 (Sorensen, Adams et al. 2009). Arteries in Dll1-/- mice also exhibit increased expression of the venous marker and decreased level of arterial markers (Sorensen, Adams et al. 2009), correlating with defects in arterialization even after most arteries are already established. The third arterial-restricted Notch ligand is Jagged1. Jagged1 expression is required for embryonic vascular development. However, unlike the Delta-like ligands, Jagged1 null embryos do not show reduced arterial marker in the dorsal aorta (High, Lu et al. 2008; Robert-Moreno, Guiu et al. 2008). Endothelial-specific ablation of Jagged1 fails to reduce the level of Notch1 activation in aortic endothelium (High, Lu et al. 2008), suggesting that Jagged1 is not involved in arterial-venous specification during vascular development. Jagged1 null mutants do exhibit other interesting vascular defects which will be discussed later. Murine mutants of other Notch signaling pathway components, not surprisingly, also exhibit defects in arterial-venous specification. Mindbomb null mutants are not able to generate functional Notch ligands and recapitulate the vascular defects of Dll4 or Notch1 mutants (Koo, Lim et al. 2005). DNA-binding partner of intracellular Notch, CSL, is also required for arterial identity. Both systemic and endothelial-specific ablation of CSL leads to reduced arterial diameter with decreases in EphrinB2 expression and smooth muscle recruitment (Krebs, Shutter et al. 2004). Finally, combined deletion of Notch targets Hey1 and Hey2 exhibits the same phenotype of decreased arterial markers in arterial endothelium and reduced smooth muscle coverage (Fischer, Schumacher et al. 2004). 30 These studies suggest that proper arterialization of mammalian vessels requires signals from Dll4 (Dll1 later in development)-expressing cells to activate receptors Notch1 and/or Notch4. Receptor activation will result in CSL-dependent transcription activation of targets Hey1, Hey2 and EphrinB2. 1.3.2. Notch in angiogenesis and vascular remodeling A predominant phenotype from most of the LOF Notch mutants is the disorganization of yolk sac vasculature (Gridley 2007; Roca and Adams 2007; Phng and Gerhardt 2009). Whereas the wildtype yolk sac has a system consisting of a large vitelline artery, smaller arteries and a network of capillaries at E9.5, the mutant vasculature exhibits a degenerating primary capillary plexus with vessels of uniform diameter. This complete lack of vascular remodeling points to defects in angiogenesis in the Notch LOF mutants. Both GOF and LOF mutants of Notch1 and Notch4 show that the two receptors are essential in driving the process of embryonic angiogenesis, although expression of Notch1 can compensate for the loss of Notch4 (Krebs, Xue et al. 2000; Uyttendaele, Ho et al. 2001; Limbourg, Takeshita et al. 2005; Venkatesh, Park et al. 2008). Of all the Notch ligands, only Jagged1 and Dll4 are implicated in the process of angiogenesis, suggesting that Jagged1 and Dll4 activates Notch1 and Notch4 receptor to regulate vascular remodeling (Xue, Gao et al. 1999; Duarte, Hirashima et al. 2004; Gale, Dominguez et al. 2004; Krebs, Shutter et al. 2004; High, Lu et al. 2008). As expected, mutants with genomic inactivation of CSL in the endothelium also show defect in angiogenesis, as Notch activation of downstream targets requires CSL (Krebs, Shutter et al. 2004). Finally, while no angiogenic defect is observed in Hey1 or Hey2-null mutants, the combination of Hey1 and Hey2 inactivation shows both targets are important for embryonic angiogenesis (Fischer, Schumacher et al. 2004). Angiogenic sprouting requires a sequence of highly coordinated endothelial activities, starting from the disruption of ECM around the formation of the new sprout to the final 31 stabilization of the nascent vessel. Notch activity appears to play a role in multiple steps of the angiogenic process. 184.108.40.206. Tip and stalk cell specification More detailed studies of the function of Dll4 and Notch in angiogenesis are made with the neonatal retina model. Murine retina has no vasculature during embryogenesis. Soon after birth, the retinal vascular system starts to develop as a sprout from the optic disc and initially forms a primitive vascular plexus which is rapidly remodeled into large and small vessels (Uemura, Kusuhara et al. 2006). In retinal vasculature, Dll4 is mostly expressed in tip cells (Hellstrom, Phng et al. 2007) while Notch activity is strongest in the stalk cells (Hofmann and Luisa Iruela-Arispe 2007). When Notch signaling in endothelial cells is blocked by either GSI treatment or genetic deletion of one copy of Dll4, angiogenesis is affected and exhibits excessive tip cell formation and branching (Hellstrom, Phng et al. 2007; Suchting, Freitas et al. 2007). In zebrafish vascular development, GSI treatment or Dll4 protein knockdown also causes excessive vessel sprouting and branching in the intersomitic vessels (Leslie, Ariza- McNaughton et al. 2007). These studies suggest that Dll4 expression suppresses Notch activation in the tip cells and enables the expression of VEGFR-2. Conversely, in the stalk cells, Notch activation suppresses VEGFR-2 expression and decreasing the response to migratory signal of surrounding VEGF molecules. Notch activation also reduces the cell’s ability to induce Dll4 expression through VEGFR-2 activation. One recent study examines the role of Notch ligand Jagged1 in angiogenic sprouting and finds that Jagged1 has an opposite effect to angiogenesis as compared to Dll4. Endothelial-specific inactivation of Jagged1 blocks retinal angiogenesis (Benedito, Roca et al. 2009). Jagged1 is found to be excluded from the tip cells in wild type vasculature. In Jagged1-null endothelial cells, Dll4 expression is induced in both the tip and stalk cells and Notch signaling is activated aberrantly throughout the vasculature (Benedito, Roca et al. 32 2009). Over-activation of Notch signaling then leads to a decreased sensitivity to VEGF- induced sprouting. The expression and interplay of Notch ligands and receptors is coordinated in order to form a functional vascular system. 220.127.116.11. Notch in other endothelial functions Notch has also been shown to affect angiogenesis through modulation of other endothelial behaviors. In addition to the specification of a tip cell, the extension of the new sprout needs controlled endothelial proliferation and migratory activity, which involves interaction between the endothelial cells and the ECM. Also, successful sprouting also requires anti-apoptotic signals in the absence of integrin-matrix interation (Brooks, Montgomery et al. 1994) and endothelial junction protein (Carmeliet, Lampugnani et al. 1999); both have been shown to be essential for endothelial survival. The role of Notch signaling in endothelial survival signaling will be reviewed in a later section. Activation of Notch signaling in endothelial cells leads to cell cycle arrest in vitro (Noseda, Chang et al. 2004; Liu, Xiao et al. 2006; Harrington, Sainson et al. 2008) and in mouse models (Trindade, Kumar et al. 2008). On the other hand, inhibition of Notch signaling results in increased endothelial cell proliferation in sprouting assays in vitro (Sainson, Aoto et al. 2005), in mouse and zebrafish development in vivo (Hellstrom, Phng et al. 2007; Suchting, Freitas et al. 2007), in the adult mouse (Dou, Wang et al. 2008), and during tumor angiogenesis (Noguera-Troise, Daly et al. 2006; Ridgway, Zhang et al. 2006). In mouse, increased endothelial cell proliferation of both tip and stalk cells may contribute to increased vessel diameter and branching after GSI treatment (Hellstrom, Phng et al. 2007), after neutralization of Dll4 activity by soluble Dll4 ligand (Lobov, Renard et al. 2007) and in Dll4 heterozygous mutants (Suchting, Freitas et al. 2007). Studies in endothelial cell cultures suggest that the inhibitory effect of Notch signaling on endothelial cell proliferation is mediated by the transcriptional regulation of downstream 33 targets of the NotchIC/CSL/MAML complex (Liu, Xiao et al. 2006). In endothelial cells, CSL- dependent Notch signaling regulates the expression of the cyclin-dependent kinase inhibitor p21CIP1 (Noseda, Chang et al. 2004). p21CIP1 expression is down-regulated by Notch1 and Notch4 activity, resulting in a reduction in the nuclear translocation of cyclinD and CDK4, in the down-regulation of Retinoblastoma protein phosphorylation, and, consequently, in cell cycle arrest (Noseda, Chang et al. 2004; Dou, Wang et al. 2008). Conversely, endothelial deletion of CSL in adult mice induced p21CIP1 and endothelial cell proliferation (Dou, Wang et al. 2008). The Notch-induced cell cycle arrest may also be brought on by down-regulation of the minichromosome maintenance (MCM) proteins (Noseda, Niessen et al. 2005; Emuss, Lagos et al. 2009). These proteins are required as part of the DNA replication process during the S phase. As the level of endogenous Notch activation increases with endothelial cell-cell contact, Notch-induced inhibition of proliferation may be a mechanism to maintain homeostasis in an endothelial sheet (Noseda, Chang et al. 2004). There is also some evidence that Notch signaling regulates the expression of ECM molecules. For example, there is increased transcription of fibronectin, laminin, and collagen in endothelial cells isolated from mouse embryos overexpressing Dll4 (Trindade, Kumar et al. 2008). As a result, these mutants show increased deposition of ECM around the dorsal aorta. Conversely, Dll4 heterozygous mouse embryos show decreased expression and irregular deposition of collagen IV and laminin (Benedito, Roca et al. 2009). Integrin expression is also regulated by Notch signaling in endothelial cells (Harrington, Sainson et al. 2008). Over- expression of the intracellular domain of Notch4 in endothelial cells results in a β1 integrin- mediated increase in adhesion to collagen, and these cells show a reduced sprouting response to VEGF both in vitro and in vivo (Leong, Hu et al. 2002). Together, these results illustrate that Notch can influence both matrix production and adhesive properties in the form of integrin receptor expression. 34 These studies showed that Notch is a multifaceted regulator of angiogenesis either in tumorgenesis or during development. The main role of Notch signaling in the vasculature lies in cell fate specification, both in vasculogenesis and angiogenesis. However, due to the early embryonic lethality of most Notch mutants, the role of Notch in the process of embryonic arteriogenesis has yet to be thoroughly investigated. 1.3.3. Notch signaling in smooth muscle cell development The process of myogenesis is regulated by a small group of transcription regulators, including MyoD (myogenic differentiation antigen), Myogenin, myogenic factor 5 (Myf-5), MEF2 and the Myocardin-serum response factor (SRF) complex. While MyoD and Myogenin activity is sufficient for skeletal muscle differentiation, Myocardin is the main transcription regulator of smooth muscle differentiation (Long, Creemers et al. 2007). MEF2C is expressed in developing VSMC and is required for vascular development (Lin, Lu et al. 1998). In vitro experiments in the role of Notch signaling in smooth muscle differentiation generated two opposing results. Activation of the Notch pathway by expression NotchIC or co-culturing with ligand- expressing cells is found to inhibit myogenic differentiation in C2C12 myoblastic cell line through MyoD or Myogenin (Shawber, Nofziger et al. 1996; Kuroda, Tani et al. 1999; Proweller, Pear et al. 2005). Hes1 transcription is rapidly induced by Notch activation in C2C12 cells, and this induction is correlated with an inhibition of MyoD-induced differentiation (Sasai, Kageyama et al. 1992). In addition to Hes1, Notch target Hey2 is also able to block myocardin-induced smooth muscle differentiation in 10T1/2 cells, blocking the binding of SRF to DNA by physically interacting with SRF (Doi, Iso et al. 2005). Activated forms of Notch inhibits the DNA binding of MEF2C and its cooperative activity with MyoD and myogenin to activate myogenesis by physical interaction (Wilson-Rawls, Molkentin et al. 1999). The interaction observed between NotchIC and MEF2C may be explained by the interaction 35 between MEF2C and Notch co-activation MAML1 (Shen, McElhinny et al. 2006). Notch may block MEF2C-induced myogenesis by pulling co-activator from the MEF2C trasactivation complex. In addition, Notch target Hey1 blocks MyoD-induced transcription of myogenin and MEF2C by binding to their promoters and hindering binding of MyoD (Buas, Kabak et al. 2010). Overall, Notch activation has been shown to inhibit MyoD or MEF2C-induced myogenesis. In mesenchymal stem cells, however, contact with immobilized Notch ligand Jagged1, co-culture with Dll1 expressing cell, or expression of NotchIC is sufficient to induce Hes1 expression and induce smooth muscle differentiation (Kurpinski, Lam et al. 2010). In another co-culture system, Notch3 receptor on mural cells is activated by Jagged1 on endothelial cells and this interaction is required for the expression of mesenchymal markers in the mural cells (Liu, Kennard et al. 2009). In 10T1/2 cells and aortic smooth muscle cells, expression of intracellular domain of Notch1, Notch3 or co-culture with Jagged1 (but not Dll4) expressing cells induces expression of smooth muscle markers with CSL-dependent mechanism (Doi, Iso et al. 2005). Notch activation in endothelial cells has been shown to directly induce transcription of SMA (Noseda, Fu et al. 2006) and platelet-derived growth factor receptor (PDGFR)-β (Jin, Hansson et al. 2008), among up-regulation of other mesenchymal markers (Noseda, McLean et al. 2004). However, a feedback mechanism exists for Notch-induced SMA expression. Hey1 or Hey2 expression in smooth muscle cells can inhibit CSL binding to SMA-promoter and reduce the endogenous level of mesenchymal markers (Tang, Urs et al. 2008). This feedback mechanism or the differential signal required for smooth muscle/skeletal muscle differentiation may be partially responsible for the conflicting reports on the role of Notch activation in myogenesis. In murine embryogenesis, the role of Notch signaling in vascular smooth muscle development is much less ambiguous. Lack of smooth muscle coverage is observed in multiple Notch pathway mutants, showing the requirement for Notch activation and 36 downstream signaling events in arteriogenesis. In Dll4 heterozygous knockout mutants, the dorsal aorta is constricted in diameter and showing reduced VSMC (Gale, Dominguez et al. 2004). In CSL-null mice, which lose all canonical Notch signaling capability, the expression of SMA is down-regulated (Krebs, Shutter et al. 2004). Genomic inactivation mutants of Hey2 showed thinner vessel wall in the ascending aorta, descending aorta and pulmonary artery (Sakata, Koibuchi et al. 2006), suggesting a VSMC defect. When both Hey1 and Hey2 are lost, there is a more apparent reduction of SM22-α positive cells around the dorsal aorta (Fischer, Schumacher et al. 2004). Finally, ligand modifier Mind bomb E3 ligase is also required for smooth muscle recruitment to the dorsal aorta (Koo, Lim et al. 2005). These observations point to an important role for Notch in VSMC development. However for most of the mutants general arterial-specification is altered, therefore the lack of smooth muscle may be secondary to arterial-venous defects. Using tissue-specific targeting of Notch signaling, some recent studies have elucidated the role of Notch signaling in VSMC differentiation in vivo. Blockade of Notch signaling through expression of a dominant-negative form of MAML1 (dnMAML) in neural crest cells is achieved using either the Wnt1 or Pax3 promoter (High, Zhang et al. 2007). Blocking Notch signaling in the neural crest cells leads to lethality in late gestation or neonatally depending on the promoter used. Mortality is due to various cardiovascular malformations including the lack of smooth muscle coverage to the pharyngeal arch arteries. The expression of dnMAML blocks the transcription of Notch target genes Hey1, Hey2 and HeyL in the VSMC of pharyngeal arch arteries, without affecting the endothelium (High, Zhang et al. 2007). Notch blockage does not alter the migration of the neural crest cells but inhibits the ability to differentiate towards the VSMC fate. Interestingly, dnMAML expression in mature smooth muscle cells does not give the same cardiovascular defects (High, Zhang et al. 2007). However, in another study, Notch signaling is found to not only drive the differentiation process in neural crest progenitors, but also the expansion of mature VSMC. VSMC lacking 37 Notch2 by smooth muscle targeted gene inactivation leads to increased postnatal mortality in the first three weeks (Varadkar, Kraman et al. 2008). The mice have smaller aorta and pulmonary arteries and show fewer proliferating VSMC. Previously in vitro experiment shows that expression of Notch target Hey1 promotes VSMC proliferation by down-regulation of p21CIP1 (Wang, Prince et al. 2003), which may be a possible mechanistic explanation for lack of cell growth in Notch2-null VSMC in vivo. Interestingly, another study may shed light on the origin of the Notch activation in the neural crest-derived VSMC. When Notch ligand Jagged1 is inactivated specifically in the endothelium using a Tie2-cre/lox system, the mutants die in utero due to cardiovascular defects (High, Lu et al. 2008). When examined in detail, the pharyngeal arch arteries and the dorsal aorta rostral to the heart all exhibit a loss of SMA or SM22-α positive mural cells. While arterial-venous specification and endothelial Notch activation are not affected in the mutants, Notch activation in the peri-endothelial cells and smooth muscle differentiation are both reduced (High, Lu et al. 2008). This study, in combination with the neural crest studies, suggests that Jagged1 expression in endothelial cells can stimulate Notch receptor on the VSMC precursors to drive differentiation. One caveat of the study is the use of the Tie2 promoter for endothelial-specific inactivation of Jagged1. Tie2 expression has been observed in cell types other than endothelial cells; monocytes (De Palma, Murdoch et al. 2007), hematopoietic stem cells (Hsu, Ema et al. 2000), as well as endothelial progenitor cells (Nowak, Karrar et al. 2004) have been shown to express Tie2 on the cell surface. Therefore, the effect of Jagged1 inactivation may not be attributed solely to the endothelium. Unexpectedly, in mutants with systemic Jagged1 inactivation, the dorsal aorta at the AGM (aorta-gonad-mesonephros) region does not exhibits reduced smooth muscle coverage (Robert-Moreno, Guiu et al. 2008), suggesting the Jagged1 induced smooth muscle differentiation observed by High and colleagues may be dependent on the vascular bed 38 examined, as VSMC of the AGM originate from the somites, not the neural crests (Wasteson, Johansson et al. 2008). Mutation of the Notch3 receptor has been associated with a human disease, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) (Joutel, Corpechot et al. 1996; Joutel, Corpechot et al. 1997), where cerebral vascular defects leads to stroke and dementia in patients. The arteriopathy is characterized by dismorphic intima due to VSMC degeneration (Ruchoux and Maurage 1997). The Notch3 mutation leads to an accumulation of the membrane-anchored extracellular domain of the receptor in VSMC (Joutel, Andreux et al. 2000) and this ectopic clustering of Notch3 receptors appears to be the cause of VSMC defects. However, the detailed mechanism of how Notch3 mutation leads to VSMC dysfunction is still under investigation. Interestingly, the maturation of arterial VSMC in mouse appears to require Notch3. In the absence of Notch3, the arterial smooth muscle cells are dysfunctional (Arboleda-Velasquez, Zhou et al. 2008). The arteries are enlarged with thinner vessel wall and disorganized tunica media. The arterial smooth muscle cells also appear to obtain a venous phenotype. As a result of these phenotypic defects, the Notch3-null mice are more susceptible to ischemic stroke when challenged (Arboleda-Velasquez, Zhou et al. 2008). In avian development models, somitic cells or extraembryonic mesoderm requires Notch signaling to form the smooth muscle cells. Using a CSL-binding-promoter reporter to illustrate Notch activation, one study shows that Notch activated cells are preferentially found in the dorsal aorta (Sato, Watanabe et al. 2008). Activation of Notch signaling in the somites increased the ventral migration of somitic cells and incorporation into the dorsal aorta, while inhibition of Notch signaling hinders both processes. Notch activation also predisposes somitic cells towards a VSMC fate in the dorsal aorta (Sato, Watanabe et al. 2008). In later stage of development, the somite-derived dermomyotome can also contribute to the dorsal aorta and Notch activation drives the cells towards a mural cell fate as oppose to becoming 39 part of the dermis or skeletal muscle system (Ben-Yair and Kalcheim 2008). Interestingly, at the site of primitive hematopoiesis in the extraembryonic mesoderm, Notch activated cells are also biased to become VSMC (Shin, Nagai et al. 2009). These studies suggest that Notch activation drives smooth muscle differentiation in different vascular progenitors in avian development, both in the embryo proper or in the extraembryonic tissues. Recently, a study in lung development yields interesting insight into arterial VSMC differentiation in the lung (Morimoto, Liu et al. 2010). Cell fate tracing reveals that the microvascular endothelium, the arterial endothelium and the arterial VSMC, but not bronchial VSMC, are all derived from progenitor cells that experienced Notch1 activation and are continuing to engage in Notch signaling at E14.5. While ablation of CSL in lung mesenchymal cells does not result in defects of lung function, there is a decrease in mesenchyme-derived arterial VSMC in the lung. However, EndMT may rescue the defect in VSMC differentiation and cause an overall normal phenotype (Morimoto, Liu et al. 2010). Notch signaling is one of the pathways implicated in the process of EndMT during cardiac cushion formation (Noseda, McLean et al. 2004; Timmerman, Grego-Bessa et al. 2004). During cardiac development, the atrio-ventricular canal (AVC) is formed by the invasion of endothelial cell-derived mesenchyme into the cardiac jelly. The first step of EndMT is the loss of endothelial-to-endothelial junctions followed by the gain of mesenchymal phenotype. Notch activation in endothelial cells in vitro suppresses the expression of vascular endothelial cadherin (VE-cadherin), an important endothelial junctional protein and up- regulates mesenchymal markers (Noseda, McLean et al. 2004). In CSL-null or Notch1-null embryos, the cardiac jelly is devoid of mesenchymal cells (Timmerman, Grego-Bessa et al. 2004). Notch activation induces expression of the Snail family of transcription repressors (Timmerman, Grego-Bessa et al. 2004; Niessen, Fu et al. 2008) which in turn reduces the expression of endothelial junction proteins to initiates EndMT and increases migration of the transdifferentiated endothelial cells in AVC formation. Therefore, Notch is essential for the 40 initiation of EndMT in mammalian cardiac development. However, the role of Notch in EndMT in the developing mammalian vasculature has yet been examined. Overall, studies involving Notch pathway mutant mice illustrate the importance of Notch activation in multiple processes during mammalian vascular development, as summarized in Table 1.3. Work done in this thesis will further delineate the role of Notch signaling in vascular smooth muscle development across all vascular beds in vivo. Table 1.3. Vascular development processes affected in Notch pathway mutant mice Vascular development process Notch Receptor Notch Ligand Notch Target Notch Co- activator/ modulator Arterio-venous specification Notch1 (Fischer, Schumacher et al. 2004; Venkatesh, Park et al. 2008) Notch4 (Shawber, Funahashi et al. 2007) Dll1 (Sorensen, Adams et al. 2009) Dll4 (Duarte, Hirashima et al. 2004; Krebs, Shutter et al. 2004; Trindade, Kumar et al. 2008) Hey1/Hey2 (Fischer, Schumacher et al. 2004) CSL (Krebs, Shutter et al. 2004) Mindbomb (Koo, Lim et al. 2005) Angiogenesis Notch1 (Krebs, Xue et al. 2000; Limbourg, Takeshita et al. 2005) Notch4 (Krebs, Xue et al. 2000; Uyttendaele, Ho et al. 2001) Jagged1 (Xue, Gao et al. 1999; Benedito, Roca et al. 2009) Dll4 (Duarte, Hirashima et al. 2004; Krebs, Shutter et al. 2004; Trindade, Kumar et al. 2008) Hey1/Hey2 (Fischer, Schumacher et al. 2004) CSL (Krebs, Shutter et al. 2004; Dou, Wang et al. 2008) Presenilin1 (Nakajima, Yuasa et al. 2003) Mindbomb (Koo, Lim et al. 2005) Vascular smooth muscle cell development Notch1 (Morimoto, Liu et al. 2010) Notch2 (Varadkar, Kraman et al. 2008) Notch3 (Domenga, Fardoux et al. 2004; Arboleda-Velasquez, Zhou et al. 2008) Jagged1 (High, Lu et al. 2008) Dll4 (Gale, Dominguez et al. 2004) Hey2 (Fischer, Schumacher et al. 2004; Sakata, Koibuchi et al. 2006) Hey1/Hey2 (Fischer, Schumacher et al. 2004) CSL (Krebs, Shutter et al. 2004; Morimoto, Liu et al. 2010) Mindbomb (Koo, Lim et al. 2005) Endothelial survival Notch1 (Limbourg, Takeshita et al. 2005) Dll4 (Trindade, Kumar et al. 2008) Presenilin1 (Nakajima, Yuasa et al. 2003) 41 1.3.4. Notch and endothelial survival There is conflicting evidence showing that Notch signaling promotes, blocks or has no effect towards endothelial apoptosis. During murine embryogenesis, over-expression of ligand Dll4 leads to vascular defects and increased expression of Notch targets (Trindade, Kumar et al. 2008). Despite the appearance of a dilated dorsal aorta, there is an increase in the proportion of apoptotic cells in the aortic endothelium. This study suggests that over- activation of Notch pathway leads to endothelial death. However, when Notch activation is inhibited by ablation of presenilin1, the catalytic unit of the γ-secretase complex, the capillary endothelium of the mutant embryo appears apoptotic (Nakajima, Yuasa et al. 2003). This observation, in contrast, shows that Notch activation is required for endothelial survival during embryogenesis. In addition, endothelial-specific genomic inactivation of receptor Notch1 also leads to aortic endothelial death (Limbourg, Takeshita et al. 2005), showing the discrepancy observed in the first two studies is not due to the difference in vascular bed examined. There is a possibility that the endothelial apoptosis observed is secondary to the disruption of the microenvironment. It could also represent the sensitivity of endothelial survival to the level of Notch activation during vascular development. Interestingly, genomic deletion of CSL in adult mouse does not lead to endothelial apoptosis (Dou, Wang et al. 2008), showing potential difference in the requirement for survival of Notch signaling in embryonic and adult endothelium. In vitro experiments show the differential effect of Notch receptors on endothelial apoptosis. Multiple studies have shown that Notch4 has an anti-apoptotic effect on cultured endothelial cells. In both primary venous endothelial cells and a transformed microvascular endothelial cell-line, expression of activated Notch4 protects cells against lipopolysaccharide (LPS)-induced apoptosis (MacKenzie, Duriez et al. 2004). Conversely, reduction of Notch4 or Hes1 protein level by RNA interference (RNAi) leads to apoptosis in cultured primary venous endothelial cells (Quillard, Coupel et al. 2008). Interestingly, the same study also showed 42 decreased Notch4 expression in the endothelium in a transplant arteriosclerosis animal model, showing a possible mechanism for endothelium apoptosis observed in atherosclerotic plaques. Disruption in Notch ligand Dll4 by RNAi in endothelial cells also decreases the viable cell fraction under serum starvation compared to RNAi control cells (Patel, Li et al. 2005). Inhibition of Notch signaling by GSI shows a similar increase in serum starvation-induced apoptosis (Takeshita, Satoh et al. 2007). These studies show that endogenous homotypic endothelial Notch signaling in vitro plays a role in maintaining monolayer homeostasis by blocking apoptosis. However, Quillard and colleagues show that tumor necrosis factor (TNF) induces apoptosis in endothelial cells through activation of Notch2 receptor (Quillard, Devalliere et al. 2009). Activation of Notch2 in primary venous endothelial cells leads to apoptosis by suppressing the expression of anti-apoptotic protein Survivin. Since the basal expression of Notch2 is low in primary venous endothelial cells (Noseda, Chang et al. 2004; Quillard, Devalliere et al. 2009), Notch2 may not be the primary receptor required for maintenance of endothelial homeostasis. However, it may still play a role in TNF-stimulated endothelium. The effect of Notch signaling on endothelial apoptosis is likely context-dependent. There is a lack of detailed studies on how Notch pathway can induce cell death or survival. In this thesis, I attempt to provide more insight into both the mechanism and context-dependent nature of the interplay between Notch and apoptosis. 43 1.4. Aims of the studies Notch signaling is instrumental in the regulation of many vascular processes. From both the standpoint of a “basic” biologist who aims to understand the fundamental processes of living things or a clinical researcher who aims to discover novel therapeutics for human disorders, there is an interest in a more detailed examination on the role of Notch in the vasculature. Notch signaling plays an important role in cell fate decision in the establishment of a functional vasculature both during developmental processes and under pathological conditions through the regulation of arterial-venous specification and tip-stalk cell designation. The work described in this thesis examined the role of Notch in two processes that have not been under the same kind of scrutiny: embryonic arteriogenesis and endothelial apoptosis. While the importance of smooth muscle cells in vascular stability during development and in the adult has been established, the characterization of VSMC precursors is still underway. Our lab has shown that Notch activation in cultured endothelial cells can induce EndMT, through both down-regulation of endothelial markers and up-regulation of smooth muscle markers (Noseda, McLean et al. 2004). While endothelial cells have been shown to give rise to VSMC in avian development, there is no study that directly shows the role of EndMT in mammalian arteriogenesis. Recently, several studies have been published that investigate the role of Notch signaling in mammalian smooth muscle development (reviewed by (Morrow, Guha et al. 2008)). The studies focus on neural crest-derived VSMC and show that Notch activation can facilitate VSMC differentiation and VSMC expansion. Using a binary inducible transgenic mouse system, we will examine whether EndMT occurs in mammalian vasculature and whether Notch activation can be a driving force in the process as observed in cell culture experiments. Our lab and others have shown that Notch may be essential for the maintenance of endothelial homeostasis by regulating proliferation, tip cell formation and apoptosis. Notch 44 activation shows either pro-apoptotic or anti-apoptotic activities context-dependently. We have shown that Notch activation protects endothelial cells against LPS-induced apoptosis by two independent mechanisms (MacKenzie, Duriez et al. 2004). In this thesis, the downstream signaling involved in this protective effect will be examined. From our study of the EndMT process, the zinc-finger protein Slug is identified as a direct Notch target in endothelial cells (Niessen, Fu et al. 2008). Interestingly, in irradiated mouse bone marrow, Slug showed anti- apoptotic effects (Inoue, Seidel et al. 2002). Notch signaling may provide the pro-survival effect through the induction of Slug. In cancer cells, Notch activation often exerts anti- apoptotic activity through the activation of the PI3K signaling pathway. Thus, the role of PI3K is also examined in Notch-induced endothelial survival signaling. Finally, I have examined the context-dependent nature of Notch-induced survival by examining the outcome of different apoptotic stimuli. Overall, two major hypotheses were tested in this thesis: (1) Endothelial cells can transdifferentiate into VSMC during murine artery development and the process is Notch- dependent; (2) Notch activation can protect endothelial cells against apoptosis by downstream signaling through Slug and/or PI3K. 45 Chapter 2. MATERIALS AND METHODS 2.1. Cell culture The human dermal microvascular endothelial cell line HMEC-1 (HMEC), is a cell line transformed with SV40 large T antigen (Ades, Candal et al. 1992). HMEC was provided by the Centers for Disease Control and Prevention in Atlanta, GA. HMEC were cultured in MCDB 131 medium (Sigma-Aldrich, St. Louis, MO) supplemented with 10% heat-inactivated calf serum (CS) (HyClone, Logan, Utah), 2 mM Glutamine and 100 U each of penicillin and streptomycin (Gibco, Invitorgen, Carlsbad, CA). Human umbilical vein endothelial cells (HUVEC) were isolated as previously described (Karsan, Yee et al. 1997), and maintained in MCDB 131 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS) (HyClone, Logan, Utah), 10% heat-inactivated CS, 20 ng/mL endothelial cell growth supplement (BD Bioscience, Bedford, MA), 16 U/mL heparin (Sigma-Aldrich, St. Louis, MO), 2 mM Glutamine and 100 U each of penicillin and steptomycin. Human aortic smooth muscle cells (HASMC) were purchased from Cascade Biologics (Invitrogen, Carlsbad, CA) and were cultured in Medium 231 supplemented with smooth muscle growth supplement (SMGS) (Cascade Biologics, Invitrogen, Carlsbad, CA) according to manufacturer’s instructions. Both HUVEC and HASMC were used between passages 1 to 5. The retroviral producer cell line AmphoPheonix was obtained from Dr. Gary Nolan (Stanford University, Pal Alto, CA) and cultured in Dulbecco’s modified Eagle’s medium (DMEM, Sigma-Aldrich, St. Louis, MO) supplemented with 10% heat-inactivated CS, 2 mM glutamine and 100 U each of penicillin and streptomycin. All cells were maintained at 37°C in 5% CO2. 2.1.1. Gene transfer Human cells (HMEC, HUVEC and HASMC) were transduced using the retroviral vectors pLNCX, pLNC-Notch1IC, pLNC-Slug (Niessen, Fu et al. 2008), pLNC-dnAKT 46 (Sakoda, Gotoh et al. 2003), MSCV-IRES-YFP (MIY), MIY-Notch1IC (Noseda, McLean et al. 2004), MSCV-GFP and MSCV-dnMAML (Maillard, Weng et al. 2004) as previously described (Noseda, McLean et al. 2004). Briefly, constructs were transiently transfected into the retroviral packaging cell line AmphoPhoenix using TransIT®-LT1 Transfection Reagent (Mirus Bio, Madison, WI). Retroviral supernatants were collected, filtered through a 0.45 µm filter, 8 µg/ml Polybrene (Sigma-Aldrich, St. Louis, MO) was added, and fresh medium was added back to the virus producing cells. The viral supernatant was then applied to target cells. This procedure was repeated twice over the next 48 hours. The pLNCX transduced cells were then selected for Neomycin resistance using 300 mg/mL G418 (Invitrogen, Carlsbad, CA). The MSCV-GFP, MSC-dnMAML, and the MIY transduced cells were sorted by fluorescent activated cell sorting (FACS) for yellow fluorescent protein (YFP) or green fluorescent protein (GFP) using a FACS-440 flow-sorter (BD, Franklin Lakes, NJ). Stable polyclonal cells were obtained to avoid artifects due to retroviral integration sites. The pLNC-dnAKT construct was a generous gift from Dr. Issei Komuro, Chiba University Graduate School of Medicine, Japan. The MSCV-dnMAML construct was a gift from Dr. Warren Pear, University of Pennsylvania, Philadelphia, PA. 47 2.2. Transgenic mice The Tie1tTA and tetOSLacZ mouse lines were generously provided by Dr. D. Dumont (Sunnybrook Health Sciences Centre, Toronto, ON) and were maintained on CD1 background as an outbred strain. The VEtTA mice were a gift from Dr. L. Benjamin (Harvard University, Boston, MA) and were maintained on FVB/NJ inbred background. All mouse lines were maintained as heterozygous transgenics. The tTA transgenics were identified with genotyping primers 5’ - CTC ACT TTT GCC CTT TAG AA - 3’ and 5’ - GCT GTA CGC GGA CCC ACT TT - 3’. The tetOSLacZ heterozygous mice were identified with genotyping primers 5’ – CTG GAT CAA ATC TGT CGA TCC TT - 3’ and 5’ – GCT GGA TGC GGC GTG CGG T - 3’. The TetOSdnMAML mice were generated in collaboration with the BCCRC Transgenic Core (Vancouver, BC, Canada). The transgenic animals were generated by pronuclear injections of linearized DNA coding for tetOS promoter followed by cDNA of the dominant-negative Mastermind-like1-GFP fusion construct. The injected blastocysts were transplanted into pseudo pregnant females. Transgenic mice were identified with genotyping primers 5’- CAT GCC ATG GAT GGT GAG CAA GGG CGA G – 3’ and 5’ - CCA TCG ATT TAC TTG TAC AGC TCG TCC A – 3’. Germline transmission was obtained for 3 transgenic lines. However, only one line showed expression matching that of reporter gene activity when crossed with driver transgenic mice. That line was propagated as the TetOSdnMAML transgenic mouse line and was maintained on CD1 background as an outbred strain. The VEtTA and Tie1tTA mice were also backcrossed to the C57BL/6J background. At the time of the experiment, both strains were in the fifth generation of C57BL/6J backcross. All animal-related experiments are approved by and conform to the guidelines of the Animal Care Committee of the University of British Columbia (Vancouver, British Columbia) 48 2.2.1. Timed matings Driver mouse (Tie1-tTA or VE-cadherin-tTA) were mated with responder mice (TetOSLacZ or TetOSdnMAML). The females were checked for vaginal plug in the morning. Noon of the day where plugs were observed was designated embryonic day 0.5 (E0.5). For tetracycline-treated embryos, plugged females were place on autoclaved water containing 50 mg/L tetracycline (Apotex Inc., North York, ON, Canada) at E0.5. The water bottle containing tetracycline was replaced daily. Tetracycline was withdrawn by replacing tetracycline- containing water with untreated water for the pregnant females at the specified days. Staging of the embryos were done by identification of stage-specific landmarks. For E10.5 embryos, the prominent division between the telencephalic vesicle, the mesencephalic vesicle and the fourth ventricle, the presence of the hindlimb buds and the elongated forelimb, the absence of footplates in the limbs, the presence of nasal processes and the presence of both the first and the second branchial arches were all used to stage the embryos. For E12.5 embryos, the presence of angular footplates and visible “rays” at the location of future interdigital zones without the presence of digits were observed for both the forelimbs and hindlimbs. For E14.5 embryos, there were an absence of hair follicles at the site of future whiskers; in addition, the fingers were separated and there were deep indentations between the toes. Only embryos of the correct stages were used for further analysis. 49 2.3. Flow cytometry Mouse embryos were dissected free of yolk sac tissue, minced and digested in a solution containing 1% BSA, 550 U/mL Collagenase Type II, 550 U/mL Collagenase Type IV, and 100 U/mL DNase I (Sigma-Aldrich, St. Louis, MO) in PBS for 15 minutes in a 37°C water bath. During digestion, the embryos were further broken down by repeated pipetting at 5 minute-intervals. Enzymatic reactions were inactivated by addition of DMEM with 5% CS. The embryonic cells were pelleted at 700 x G for 5 minutes at 4°C. The digested embryos were treated with red blood cell lysis buffer (0.8% NH4Cl, 0.1mM EDTA) (Sigma-Aldrich, St. Louis, MO) and resuspended in DMEM with 5% CS. The single-cell suspensions were either analyzed for GFP expression or stained for with Phycoerythrin (PE)-conjugated rat anti- mouse CD31 monoclonal antibody (5 µg/mL) (BD Bioscience, San Jose, CA) or a rat anti- mouse PDGFR-β antibody (10 µg/mL) (eBioscience, San Diego, CA). Rat IgG2a,κ (BD Bioscience, San Jose, CA) was used as isotype control. Cells were incubated with antibodies for one hour on ice before analysis. For PDGFR-β, secondary antibody goat anti-rat Alexa Fluor 633 (1:200) (Invitrogen, Carlsbad, CA) was used for detection. All flow cytometry analysis was done with the EPIC Elite flow cytometer (Beckman Coulter, Brea, CA) and FCS Express (De Novo Software, Los Angeles, CA). 50 2.4. Immunofluorescence staining For immunofluorescent staining of cultured cells, retroviral-transduced HMEC cells were plated at a density of 1.5 x 105 cells on a 4-well chamber slide (BD Biosciences, San Jose, CA). The cells were allowed to attach and grow until confluent. The cells were serum starved overnight and treated with DMSO (Sigma-Aldrich, St. Louis, MO) or LY294992 (Cell Signaling Technology, Danvers, MA) for the specified time. The cells were fixed in 4% paraformaldehyde (Sigma-Aldrich, St. Louis, MO), and blocked/permeablized in 4% FBS + 0.2% TritonX-100 (Sigma-Aldrich, St. Louis, MO) in PBS. Rabbit anti- Activated caspase 3 (BD Pharmingen, Franklin Lakes, NJ) was used at a dilution of 1:100 and the goat anti-rabbit Alexa594 conjugated secondary antibody was used at 1:200 (Molecular Probes, Invitrogen, Carlsbad, CA). The cells were counterstained with 4',6-diamidino-2-phenylindole (DAPI, 1 µg/mL) (Sigma-Aldrich, St. Louis, MO). For staining of sorted embryonic cells, 2.0 x 104 cells were spotted onto a slide using the Cytospin2 centrifuge (Shandon, Thermo Fisher, Waltham, MA). The cells were dried onto the slides, fixed in 4% paraformaldehyde and blocked/permeablized in 4% FBS + 0.2% TritonX-100 in PBS. The cells were stained with rat anti-mouse CD31 monoclonal antibody (1:200) (BD Bioscience, San Jose, CA), and rabbit anti-human α-smooth muscle actin (1:100) (NeoMarkers, Thermo Fisher, Fremont, CA). The fluorochrome-conjugated secondary antibodies goat anti–rat Alexa Fluor 488 (Molecular Probe, Invitrogen, Carlsbad, CA) and goat anti–rabbit Alexa Fluor 594 (Molecular Probe, Invitrogen, Carlsbad, CA) were used, and nuclei were counterstained with DAPI. Whole embryos were fixed in 2% paraformaldehyde in PBS (pH7.4) for 4 hours to overnight. Then the tissues were cryoprotected with overnight incubation in 30% sucrose (Sigma-Aldrich, St. Louis, MO), followed by 1 hour in Tissue-Tek® O.C.T. Compound (Opitimal Cutting Temperature; Sakura Finetek USA, Torrance, CA). The tissues were frozen 51 in O.C.T. compound at -80ºC and 10-µm-thick cryosections were made from frozen cryoblocks. The cryosections were blocked/permeablized with 4% goat serum + 0.2% TritonX-100 in PBS and stained with rat anti-mouse CD31 monoclonal antibody (1:200), rabbit anti-GFP antibody (1:300) (Molecular Probe, Invitrogen, Carlsbad, CA) and rabbit anti- human smooth muscle actin (1:100). The fluorochrome-conjugated secondary antibodies goat anti–rat Alexa Fluor 488 or Alexa Fluor 594 and goat anti–rabbit Alexa Fluor 488 or Alexa Fluor 594 were used, and nuclei were counterstained with DAPI. Immunofluorescent staining was detected with an imaging microscope (Axioplan II; Carl Zeiss, Inc.), and images were captured with a digital camera (1350EX; QImaging, Surrey, BC, Canada). 2.4.1. Smooth muscle thickness quantification Quantification of the smooth muscle thickness was done with the NIH Image software (Research Service Branch, National Institute of Health, Bethesda, MD) with additional user- defined algorithms. The algorithms were kindly provided by Dr. Alastair Kyle (BC Cancer Research Center, Vancouver, BC, Canada). Briefly, pixels with a value of 0 (white) were set to 1 and pixels with values of 255 (black) were set to 253. The lumen created by SMA staining was filled with black pixels (255) and the area outside of the SMA staining was set to be 254. For pixels that were between 1 and 253 and next to a 254 pixel (at the outer rim of the SMA staining), the distance between that pixel to the closest 255 pixel was recorded (the thickness of SMA staining at that point). Three thousand pixels were measured and the average SMA thickness was determined by calculating the mean for individual images. An example of the analysis is shown in Figure 2.1. 52 Figure 2.1. Smooth muscle thickness quantification. (A) Immunofluorescent image of a section of E12.5 descending aorta stained for α−smooth muscle actin (SMA). (B) The same image after undergoing processing to distinguish the stained region, the inner lumen and the area outside of the staining. (C) The shortest distance between a point on the outside rim of the staining to the inner lumen was tabulated for 2000 pixels along the staining (4 possible measurements are shown here). For this particular example, the SMA thickness is 17 ± 4 µm. processed A B C 53 2.5. β-galactosidase detection For wholemount X-gal staining, embryos were fixed in β-gal fixative solution (0.2% glutaraldehyde, 5 mM EGTA (Sigma-Aldrich, St. Louis, MO), 2 mM MgCl2 (Sigma-Aldrich, St. Louis, MO) in phosphate buffered saline PBS pH7.4) for 10 minutes, then washed in β-gal wash solution (2 mM MgCl2, 0.01% sodium deoxycholate (Sigma-Aldrich, St. Louis, MO), 0.02% NP-40 (Sigma-Aldrich, St. Louis, MO) in PBS pH7.4) three time for 15 minutes. Finally, the embryos were incubated with β-gal staining solution (1 mg/ml X-gal (Invitrogen, Carlsbad, CA), 5 mM potassium ferrocyanide (Sigma-Aldrich, St. Louis, MO), 5 mM potassium ferricyanide (Sigma-Aldrich, St. Louis, MO) in β-gal wash solution) for 2 hours at 37°C. The embryos were post-fixed in 2% paraformaldehyde and then embedded in OCT compound for cryosectioning. For staining on tissue cryosections, the staining was carried out as described, except the sections were incubated with staining solution overnight at 37°C. For staining for flow cytometry analysis, embryos were digested to single cell suspension as described. Cells were incubated with 0.5mM Fluorescein di-β-D- galactopyranoside (FDG) (Invitrogen, Carlsbad, CA) in 37°C waterbath for 2 minutes. Then enzymatic reaction was carried out on ice for 30 minutes before FACS analysis. 54 2.6. Apoptosis/survival Assays Cells were serum-starved overnight with MCDB 131 + 2% CS and were treated with 100 µg/mL bacterial lipopolysaccharide (LPS) from Escherichia coli (Sigma-Aldrich, St. Louis, MO) and 25 µM of ALLN (N-Acetyl-L-leucyl-L-leucyl-L-norleucinal) (EMD Chemicals, Gibbstown, NJ) or 7.5 mM homocysteine (Sigma-Aldrich, St. Louis, MO) for the specified time, unless otherwise stated. Untreated cells were used as control for homocysteine treatment, while cells treated with 25 µM ALLN were used as controls for the LPS+ALLN treatment to examine the effect of LPS. 2.6.1. Annexin V binding assay Medium from the plate with cells in suspension was collected and pooled with adherent cells trypsinized from the plate. The cells were washed twice with cold PBS and then resuspended in Binding Buffer (10 mM HEPES (Sigma-Aldrich, St. Louis, MO), pH 7.4; 140 mM NaCl (Sigma-Aldrich, St. Louis, MO); 2.5 mM CaCl2 (Sigma-Aldrich, St. Louis, MO)) at a concentration of ~1 x 106 cells/ml. Two and a half micro-litres of PE-conjugated Annexin V (Invitrogen, Carlsbad, CA) were added to 1 x 105 cells and the cells were incubated for 15 minutes at room temperature. 400 µL of ice-cold Binding Buffer was added to the cells and the cells were analyzed by flow cytometry. The percentages of AnnexinV positive cells were determined. 2.6.2. Neutral Red uptake assay HMEC were plated at the density of 3.0 x 104 cells per well in a 96-well plate. The cells were treated with LPS and ALLN or homocysteine as specified. The cells are incubated in MCDB 131 medium containing 2% CS and 0.0025% Neutral Red (Sigma-Aldrich, St. Louis, MO) for 4 hours under normal culture condition. After removing Neutral red-containing 55 medium carefully, cells were lysed and Neutral red was solubilized with 1% acetic acid (Sigma-Aldrich, St. Louis, MO) in 50% ethanol. The absorbance at 550 nm was quantified with Genios Microplate Reader (Tecan, Männedorf, Switzerland). 2.6.3. Activated caspase 3 detection HMEC were plated at the density of 1.5 x 105 cells per well in a four-well chamber slide (BD Biosciences, San Jose, CA). The cells were treated with DMSO or 40 mm of LY294002 for 8 hours. Immunofluorescence staining for activated caspase 3 was performed as described in section 2.4. The percentage of cells with activated caspase 3 staining was determined by calculating the proportion of the nuclei (DAPI-positive) in the field with activated caspase 3 co-staining. 56 2.7. Immunoblotting For immunoblotting, cells were lysed in RIPA buffer (PBS containing 1% NP-40 (Sigma-Aldrich, St. Louis, MO), 0.5% sodium deoxycholate, and 0.1% SDS (Sigma-Aldrich, St. Louis, MO)) with addition of fresh protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN). Fifty mg of total protein, as measured using Bio-Rad DC Protein Assay System (Bio-Rad Laboratories, Hercules, CA), were analyzed by sodium dodecyl sulfate- polyacrylamide gel electrophoresis, transferred to nitrocellulose membranes (Bio-Rad Laboratories, Hercules, CA), and developed by enhanced chemiluminescence (PerkinElmer Life Sciences, Boston, MA). Membranes were probed using the following antibodies: 1:1000 rabbit anti-Slug ( Cell Signaling Technology, Danvers MA), 1:1000 goat anti-Slug (clone G-18, Santa Cruz Biotechnology Inc., Santa Cruz, CA), 1:10000 mouse anti-Tubulin (Sigma-Aldrich, St. Louis, MO), 1:1000 rabbit anti-phospho-Akt (S473) (Cell Signaling Technology, Danvers MA), 1:5000 rabbit anti-Akt (Cell Signaling Technology, Danvers MA), 1:2000 goat anti- Notch1 (Santa Cruz Biotechnology Inc.), 1:1000 rabbit anti-human SMA, 1:1000 mouse anti- Myc-tag (Cell Signaling Technology, Danvers MA). 57 2.8. Real-time PCR Total RNA was isolated from cells using TriZOL reagent (Invitrogen, Carlsbad, CA) according to manufacturer instruction. 2.5 µg of total RNA was treated with DNase I (Invitrogen, Carlsbad, CA) and was reverse transcribed by using the Superscript II kit with random primers (Invitrogen, Carlsbad, CA). Alternatively, Cells-to-cDNA™ II Kit (Ambion, Applied Biosystems, Austin, TX) was used to generate cDNA from 1 x 104 to 1 x 105 mouse embryonic cells according to manufacturer’s instruction. For each PCR reaction 2.5 µl of the cDNA was used. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as the loading control for comparison across samples. Real time quantitative RT-PCR was performed on The Applied Biosystems 7900HT Fast Real-Time PCR System by using the Power SYBR® Green PCR kit (Applied Biosystems, Foster City, CA). Primers are described in Table 2.1. Table 2.1. Primers for quantitative RT-PCR Forward primer sequence 5’→ 3’ Reverse primer sequence 5’→ 3’ mouse CD31 AGCTAGCAAGAAGCAGGAAGGACA TAAGGTGGCGATGACCACTCCAAT mouse CD34 ATCATCTTCTGCTCCGAGTGCCAT AGCAAACACTCAGGCCTAACCTCA mouse cKit TGCCAACCAAGACAGACAAGAGGA AGGAGGATATTCCTGGCTGCCAAA mouse Flk-1 AGGCCCATTGAGTCCAACTACACA ATGTCTGCATGGTCCCATACTGGT mouse MEF2D GCTCCATGCAGTTCAGCAATCCAA AGGCTCCATTAGCACTGTTGAGGT mouse Msx2 TGAGGAAACACAAGACCAACCGGA TGACCTGGGTCTCTGTAAGGTTCA mouse SMA ATTGTGCTGGACTCTGGAGATGGT TGATGTCACGGACAATCTCACGCT mouse Tal1 GTTCACCAACAACAACCGGGTGAA AAGGCGGAGGATCTCATTCTTGCT mouse Tie1 TGAACACTCAGACCCACAGCAACT GCAGGTTGGCCAGCAATGTTAAGT mouse Tie2 ACACTGTCCTCCCAACAGCTTCTT TGATTCGATTGCCATCCAACGCAC mouse Hey1 CACGCCACTATGCTCAATGT TCTCCCTTCACCTCACTGCT 58 Forward primer sequence 5’→ 3’ Reverse primer sequence 5’→ 3’ mouse Hey2 TTCTGTCTCTTTCGGCCACT TTTGTCCCAGTGCTTGTCTG mouse p53 AAAGGATGCCCATGCTACAGAGGA AGGATTGTGTCTCAGCCCTGAAGT mouse p21 GGAATTGGAGTCAGGCGCAGAT GAAGAGACAACGGCACACTTTGCT mouse Bax ACAGCAATATGGAGCTGCAGAGGA TGTCCAGCCCATGATGGTTCTGAT mouse Lef1 TGGCATCCCTCATCCAGCTATTGT TAGCGTGCACTCAGCTACGACATT mouse Axin2 AAAGAAACTGGCAAGTGTCCACGC GGCAAATTCGTCACTCGCCTTCTT mouse MEF2C ACTTCCTGGAGAAGCAGAAAGGCA AACACGTTTCCTTCTTCAGCACGC mouse CSL TTGGTGTGTTCCTCAGCAAG GCTCCCCACTGTTGTGAACT mouse GAPDH TGCAGTGGCAAAGTGGAGAT TTTGCCGTGAGTGGAGTCATA human Hey1 AGAGTGCGGACGAGAATGGAAACT CGTCGGCGCTTCTCAATTATTCCT human Slug CCCTGAAGATGCATATTCGGAC CTTCTCCCCCGTGTGAGTTCTA human GAPDH GGACCTGACCTGCCGTCTAGAA GGTGTCGCTGTTGAAGTCAGAG 59 2.9. Statistical analysis Results were expressed as means ± standard error of mean (SEM). Data were analyzed using a two-tailed Student’s t-test or a paired Student’s t-test using the GraphPad Prism statistical program. 60 Chapter 3. VASCULAR SMOOTH MUSCLE DIFFERENTIATION FROM TIE1+ PRECURSORS REQUIRES NOTCH 3.1. Introduction VSMC play an important role in vascular homeostasis in the adult. During development, VSMC not only provide structural support for the nascent arteries, but are also involved in bidirectional signaling with the endothelium for the overall stability of the vasculature. However, the developmental origin of VSMC is still under investigation. Many embryonic tissues have been shown to be sources of VSMC during development [reviewed in (Gittenberger-de Groot, DeRuiter et al. 1999; Majesky 2007)]. Within a tissue, only a subset of cells has the ability to differentiate into VSMC. While cells from the mesoderm can migrate to the site of dorsal aorta formation during avian development, only a small percentage of mesoderm cells actually integrate as a part of the dorsal aorta (Sato, Watanabe et al. 2008), suggesting the existence of a local vascular progenitor cell within the tissue that is predisposed towards a vascular fate. Studies have been done to isolate and culture such progenitor cells from mammalian sources, among these progenitors, mesoangioblasts (Minasi, Riminucci et al. 2002) and ES cell-derived cardiovascular progenitors (Yamashita, Itoh et al. 2000; Yang, Soonpaa et al. 2008) have both been shown to expand and differentiate into VSMC ex vivo. However, there is still a lack of direct in vivo evidence and in situ tracking of an immediate precursor cell that can differentiate into VSMC within the mammalian development system. Anatomically, the endothelium is the most closely associated tissue to the VSMC, which makes it a prime candidate source of VSMC. Interestingly, the endothelium has been shown to be a possible source of VSMC through the process of EndMT in an avian developmental model (DeRuiter, Poelmann et al. 1997). In addition, we have previously observed EndMT in Notch-activated human endothelial cells (Noseda, McLean et al. 2004). However, there are 61 no studies examining whether the endothelium could be a source of smooth muscle progenitor cells during mammalian development. Two reports have suggested that murine ES cells differentiate into VSMC via an endothelial intermediate (Ema, Faloon et al. 2003; Hill, Obrtlikova et al. 2010). However, the ES cell-derived endothelial cells expressed endothelial markers at a much lower level and mesenchymal markers at a much higher level compared to mature endothelial cells (Hill, Obrtlikova et al. 2010). Therefore, the isolation of endothelial cells from differentiating ES cells with endothelial markers such as CD31 (Ema, Faloon et al. 2003) or Tie1 (Marchetti, Gimond et al. 2002) may also result in enrichment of immature vascular precursors, in addition to mature endothelial cells. The close association between vascular precursor cells and endothelial cells makes it difficult to distinguish between precursor-derived VSMC and endothelial cell-derived VSMC. Notch signaling is one of the evolutionarily conserved pathways implicated in the process of EndMT during cardiac cushion formation (Noseda, McLean et al. 2004). Disruption of Notch signaling during embryonic development leads to cardiac malformations as well as vascular defects (reviewed in (Phng and Gerhardt 2009)). A decrease in vascular smooth muscle coverage was observed in some Notch mutants, but the exact mechanism of the disruption and the cell type involved was not examined in detail. Using cell type specific targeting of Notch signaling, some recent studies have elucidated the role of Notch signaling in VSMC differentiation. Blockade of Notch signaling through expression of a dominant- negative form of MAML1 (dnMAML) in mouse embryonic neural crest cells leads to lethality in late gestation due to cardiovascular malformation (High, Zhang et al. 2007). There is a lack of smooth muscle coverage of the vessels where the VSMC has been shown to derive from a neural crest origin, such as the pharyngeal arch arteries (High, Zhang et al. 2007). In avian development models, enforced Notch activation in the somitic cells or extraembryonic mesoderm can be found to preferentially integrate into the smooth muscle layer of the dorsal 62 aorta or the arteries of the yolk sac (Ben-Yair and Kalcheim 2008; Sato, Watanabe et al. 2008; Shin, Nagai et al. 2009). These studies suggest that Notch signaling plays a role in VSMC differentiation from precursors that originate from either the neural crest or mesodermal sources. The effect of Notch signaling on VSMC differentiation from other sources is still not fully investigated. To determine whether Notch-induced EndMT is involved in the development of VSMC in a mammalian system by promoting differentiation from local precursors, we have utilized a binary tetracycline-inducible transgenic system to inactivate Notch signaling in endothelial cells during murine vascular development. Surprisingly, the two endothelial promoters we used generated different phenotypes. Our findings indicate that mature endothelium is not a source of VSMC during murine development. We have also observed that the Tie1 promoter is active in a local precursor population that can give rise to VSMC in a Notch-dependent fashion. Our results suggest that Notch signaling is essential for the differentiation of VSMC from Tie1+/CD31+/VE-cadherin- precursor cells. 63 3.2. VSMC are derived from a Tie1+CD31+VE-cadherin- precursor cell To determine whether endothelial cells can transdifferentiate into VSMC in mammalian development, we used a tetracycline-inducible binary transgenic system to track the endothelial cells. This system requires crossing driver mice expressing the transcription activator (tTA) under an endothelial cell-specific promoter with responder mice expressing a transgene under the tTA-activated tetOS promoter (Figure 3.1). We employed two different endothelial drivers, Tie1-tTA (Tie1tTA)(Sarao and Dumont 1998) and VE-cadherin-tTA (VEtTA)(Sun, Phung et al. 2005), to drive the expression of the β-galactosidase reporter. Tie1 is an orphan receptor tyrosine kinase involved in the regulation of the Tie2/Angiopoietin pathway and is expressed on the surface of endothelial cells (Partanen, Armstrong et al. 1992; Puri, Partanen et al. 1999) and immature hematopoietic cells (Rodewald and Sato 1996). VE-cadherin is an endothelial-specific junctional protein involved, as the name suggests, in the formation of the homotypic adherens junction between endothelial cells (Lampugnani, Corada et al. 1995). Both promoters are widely used for endothelial-specific expression of transgenes (Mukai, Rikitake et al. 2006; Rao, Lobov et al. 2007; Reiss, Droste et al. 2007; Lohela, Helotera et al. 2008; Wolfram, Diaconu et al. 2009). 64 Figure 3.1. The tetracycline-inducible, endothelial-specific transgenic system. This system requires crossing driver mice expressing tTA transcription activator under either the Tie1 (Tie1tTA) or the VE-cadherin (VE-tTA) promoter with responder mice expressing a transgene under the tTA-activated tetOS promoter. The two responder lines drive the expression of β-galactosidase reporter (tetOS-LacZ) or dnMAML (tetOS-dnMAML), an inhibitor of Notch signaling. Treatment of mice with tetracycline suppresses the expression of the transgenes, while tetracycline withdrawal induces endothelial expression of the transgenes. Both promoters were active in the embryonic vasculature at the stages when VSMC development occurs as demonstrated by wholemount X-gal staining (Figure 3.2A-D) (Takahashi, Imanaka et al. 1996). We confirmed that this system is capable of marking endothelial cell-derived mesenchymal cells, by showing β-galactosidase activity in the cardiac cushion mesenchyme of the atrio-ventricular canal at E10.5 (Figure 3.2E and F). 65 Figure 3.2. Endothelial expression of β-galactosidase reporter. (A-D) Both the Tie1 and VE-cadherin promoters drove reporter β-galactosidase expression, as detected by wholemount X-gal staining, in the developing heart and vasculature at E9.5 (A and B) and E10.5 (C and D). The endocardial cells (dashed lines) and endothelial-derived mesenchymal cells (arrows) in the atrio-ventricular canal (AVC) are labeled with β-galactosidase activity (E and F, bar = 50 µm). 66 We next examined the descending aorta at E12.5 to determine whether endothelial cells were a source of VSMC. VE-cadherin promoter driven β-gal activity was detected in the endothelial cells and some luminal cells, which may be endothelial cell-derived hematopoietic cells, but not in the surrounding peri-endothelial cells (Figure 3.3B). In contrast, the Tie1 promoter drove the expression of β-gal not only in endothelial cells, but also in the surrounding VSMC, which was verified by co-staining for α-smooth muscle actin (SMA) and β-gal activity (Figure 3.3A). This phenomenon was not restricted to the descending aorta, but was observed in all vascular beds examined (Figure 3.3 C to N), even though VSMC from the different arteries have been suggested to arise from different developmental sources (Majesky 2007). To rule out ectopic promoter activity from transgene insertion sites, endogenous Tie1 promoter activity was examined in embryos that were heterozygous for genomic LacZ knock-in at the Tie1 locus. At E10.5, β-gal activity was observed in the peri- endothelial cells of developing arteries of the knock-in mice (Figure 3.4 A and B). 67 Figure 3.3. The Tie1 promoter, but not the VE-cadherin (VE) promoter, is active in peri- endothelial cells. (A-B) β-gal activity in the descending aorta of E12.5 embryos was detected by X-gal staining. In Tie1tTA:tetOSLacZ (Tie1LacZ) embryos, β-galactosidase activity overlaps with both CD31 staining and SMA staining (A); VEtTA:tetOSLacZ (VELacZ) embryos show only endothelial and hematopoietic (arrow) β-galactosidase activity (B). The same observations are made in carotid arteries (C and D), vertebral arteries (E and F), different segments of the descending aorta (G vs. H, I vs. J and M vs. N) and the umbilical arteries (K and L).SMA: α−smooth muscle actin. 68 Since a close family member of Tie1, Tie2 is expressed on vascular progenitor cells and Tie1 can regulate Tie2 activity, Tie1 may also be expressed on precursor cells in addition to mature endothelial cells (Zengin, Chalajour et al. 2006; Foubert, Matrone et al. 2008). Using the Tie1 and VE-cadherin promoters, one may finally be able to discern between the role of precursor cells and the endothelium in VSMC differentiation. These findings suggested that Tie1 can be used to mark a subpopulation of VSMC or a VE-cadherin- precursor cell that is capable of VSMC differentiation. Further, we conclude that mature endothelial cells are not an embryonic source of VSMC. Figure 3.4. Endogenous Tie1 promoter shows activity in peri-endothelial cells. β-gal activity in E10.5 Tie1 LacZ knock-in embryos was detected by X-gal staining. β-gal activity was observed in the thin layer of peri-endothelial cells (arrow) of the umbilical artery (A) and the abdominal aorta (B). We subsequently attempted to isolate and further characterize the Tie1+ vascular cells that we identified in the embryonic vasculature. Cells positive for β-gal activity from the E10.5 Tie1tTA:TetOSLacZ or VEtTA:TetOSLacZ embryos were sorted using Fluorescein-based fluorescent β-gal substrate FDG (Figure 3.5A-C). At E10.5, VSMC development of the dorsal aorta is still in progress and VSMC development is initiating in other smaller arteries in the embryo (Takahashi, Imanaka et al. 1996). Therefore, while mature VSMC may be present at E10.5, there should still be a population of undifferentitated VSMC precursor cells. 69 As expected, examination of marker transcripts in E10.5 Tie1+ and VE-cadherin+ embryonic cells by quantitative reverse-transcription polymerase chain reaction (qRT-PCR) revealed an enrichment of endothelial markers (CD31, CD34, Flk-1, Tie1, Tie2 and VE- cadherin) compared to the whole unsorted embryo (Figure 3.5D). When comparing the Tie1+ to the VE-cadherin+ population, a similar level of endothelial markers expression was seen, suggesting the same level of enrichment for endothelial cells (Figure 3.5E). In addition, in the Tie1+ population, enrichment of the mesenchymal progenitor marker muscle segment homeobox (Msx)2 and SMA was also observed (Figure 3.5E). A similar marker expression profile has been described for mesoangioblasts (Minasi, Riminucci et al. 2002), which are embryo-derived mesodermal progenitors that can differentiate into smooth muscle cells in vitro. To determine the source of the increased level of mesenchymal markers in the Tie1+ cells, the sorted cells were cytospun onto a slide and marker expression of individual cells was examined by Immunofluorescence. The majority of the sorted cells from E10.5 Tie1tTA:TetOSLacZ embryos showed expression of the endothelial marker CD31, which may represent the mature endothelial cells (Figure 3.5F). There were also cells expressing the mesenchymal marker SMA and cells that showed co-expression of both CD31 and SMA, which may represent the undifferentiated precursor cells (Ferreira, Gerecht et al. 2007) (Figure 3.5F). This finding suggests that the Tie1+ population contains a precursor cell type that shares characteristics with progenitors capable of differentiating into VSMC. 70 Figure 3.5. Tie1-positive cells show characteristics of VSMC precursor cells. (A-C) E10.5 Tie1LacZ (B) and VELacZ (C) embryos were digested into single cells. Cells positive for β-galactosidase activity were isolated by fluorescence activated cell sorting (FACS). (D)Endothelial markers were enriched from both sorted population by qRT-PCR analysis compared to whole embryo control. (E) Tie1LacZ+ population expresses the same level of endothelial markers compared to VELacZ+ population, with additional mesenchymal marker expression (n = 3, *P < 0.05). Data presented as mean±SEM. (F) Tie1LacZ+ cells from E10.5 embryos (4 embryos pooled) showed expression of CD31 alone (green), SMA alone (red) and a population of cells co-expressing CD31 and SMA. DAPI counterstain is shown in blue. P-value was determined by Student’s t-test. 71 Co-expression of CD31 and SMA in the Tie1+ cells suggests that the potential precursor cells may express CD31. Interestingly, a CD31+ population negative for β-gal activity was observed with flow cytometry analysis in the VEtTA:TetOSLacZ embryos i.e. VE- cadherin-negative (Figure 3.6A). The VELacZ-/CD31+ cells also expressed mesenchymal markers Msx2 and SMA, but had low expression of endothelial markers compared to VELacZ+/CD31+ endothelial cells (Figure 3.6B). Analysis of the Tie1+ embryonic cells suggests that Tie1 promoter activity may be used to enrich a potential VSMC precursor cells that can give rise to smooth muscle cells in all embryonic arteries examined, while VE- cadherin promoter activity marked a mature endothelial population incapable of transdifferentiating into VSMC. 72 Figure 3.6. Precursor cells are enriched in the LacZ-/CD31+ population in VEtTA:TetOSLacZ embryos. CD31+ cells were sorted according to β-galactosidase activity (FITC) from VEtTA:TetOSLacZ E10.5 embryos (A). VELacZ-/CD31+ cells expressed mesenchymal progenitor markers Msx2 and SMA, while having low expression of endothelial markers (B) (n = 2). 73 3.3. A tissue-specific, inducible transgenic model for Notch inhibition To investigate the effect of Notch signaling on Tie1+ cells in vivo, we used the tetracycline-inducible, binary transgenic system to express an N-terminal dominant-negative mutant of Mastermind-like 1, dnMAML, which retains the Notch and CSL binding domains of MAML1, but lacks co-activator-binding capability (Weng, Nam et al. 2003), under the control of Tie1 promoter (Figure 3.1). In human endothelial cell culture, expression of a constitutively active form of Notch, NotchIC, induced the direct Notch mesenchymal target genes Slug (Niessen, Fu et al. 2008) and SMA (Noseda, Fu et al. 2006). Co-expression with the dnMAML construct successfully blocked induction of both Notch targets (Figure 3.7). Fiugre 3.7. Expression of dnMAML blocks Notch-induced target expression. Human microvascular endothelial cells (HMEC) were transduced with activated Notch (NotchIC) and dnMAML. Expression of Notch targets Slug and SMA were detected with immunoblotting. NotchIC expression in HMEC induced Slug and SMA, while co-expression of dnMAML blocked the induction. 74 We established the tetracycline-inducible system (TetOS-dnMAML) to bypass the cardiac defects and the embryonic lethality induced by endothelial-specific inhibition of Notch signaling (Limbourg, Takeshita et al. 2005) and to examine VSMC development in embryos ranging from mid (E10.5) to late (E14.5) gestation. The VEtTA mice were also analyzed to exclude effects due to blockade of Notch signaling in the mature endothelium. When dnMAML was expressed constitutively in the endothelium, the embryo died in utero at E11.5, with visible defects and developmental delay at E10.5 regardless of which promoter was used (Figure 3.8A-D). The cardiac defect-induced lethality was in agreement with previous studies done in murine embryos with Tie2-promoter-driven genomic inaction of Notch1 (Limbourg, Takeshita et al. 2005). Figure 3.8. Inhibition of Notch in developing endothelium causes embryonic lethality. (A-B) Developmental delay was observed in both Tie1 and VE-cadherin (VE) promoter driven expression of dnMAML at E10.5. Single Tg: single transgenic control littermate; Double Tg: double transgenic mutant. (C-D) Only necrotic double transgenic embryos were obtained at E11.5. 75 Addition of tetracycline to the drinking water of the mice enabled delay of dnMAML expression, which is fused to green fluorescent protein (GFP) for detection, to bypass embryonic lethality, and withdrawal of tetracycline induced the expression of dnMAML and down-regulated the expression of the Notch target gene Hey2 in embryonic endothelial cells within the first 24 hours (Figure 3.9A-B). The rapid induction of dnMAML after tetracycline withdrawal enabled precise timing of Notch inhibition unique to this transgenic system. Recent studies have suggested that MAML may interact with other transcription regulators to modify transcription of their target genes (McElhinny, Li et al. 2008). In the present transgenic system, expression of the dnMAML construct did not affect signaling through MEF2C, p53 and β-catenin as demonstrated by the transcript level of their respective targets (Figure 3.9B). Overall, the transgenic system provided an inducible method of blocking Notch signaling. 76 Figure 3.9. Expression of dominant-negative Mastermind-like1 in endothelial cells leads to blockade of Notch signaling. (A) One day after withdrawing tetracycline, there was observable expression of dnMAML-GFP fusion protein with both Tie1 and VE-cadherin promoters, with a further increase after two days. (B) Endothelial cells from E10.5 Tie1:dnMAML double transgenic embryos showed decreased level of Notch target Hey2 compared to wildtype embryos, while transcript levels of targets for other possible MAML binding partners (p53, β-catenin, or MEF2C) remained unaltered. * P = 0.01. P-value was determined by Student’s t-test. 77 3.4. Blockade of Notch signaling in Tie1-positive precursors leads to hemorrhage localized to newly-forming vasculature To examine the effect of blocking Notch in Tie1+ vascular precursors on embryonic arteriogenesis, embryos were treated with tetracycline starting at E0.5. The removal of tetracycline treatment and the induction of dnMAML expression at E10.5 for two days led to localized hemorrhages in both the forebrain and hindbrain region in the Tie1tTA:dnMAML double transgenic embryos (Figure 3.10C). However, VEtTA:dnMAML double transgenic embryos undergoing the same treatment did not show gross morphological defects (Figure 3.10D). Interestingly, when dnMAML was induced at E11.5 for two days in the Tie1tTA:dnMAML embryos, the E13.5 embryos displayed hemorrhage covering, in addition to the forebrain and hindbrain, the midbrain and the tip of the tail. When we examined the E14.5 Tie1tTA:dnMAML embryos after inducing dnMAML at E12.5, the hemorrhagic lesions were located at the tip of the snout and the interdigital zone of the forelimbs and hindlimbs. In contrast, the hindbrain appeared normal at this time point (Figure 3.10E and G). The pattern of hemorrhage was consistent across different litters of embryos at the same stages with the same schedule of tetracycline treatment, suggesting that the defects are specific and local. The hemorrhagic regions corresponded to areas undergoing significant morphological remodeling at the time of Notch blockade (Kaufman MH, 1992). Once the remodeling was complete, Tie1 promoter-driven dnMAML expression no longer caused hemorrhages in that location. The changes in the location of the hemorrhage demonstrate that the requirement for Notch signaling is both spatially and temporally regulated during embryonic vascular development. At all stages examined, there were no hemorrhagic defects in the VEtTA:dnMAML embryos with two-day induction of dnMAML (Figure 3.10F and H and Table 3.1). 78 Figure 3.10. Blocking Notch signaling in Tie1-positive cells leads to localized hemorrhaging. (A-H) Wholemount micrographs of embryos with dnMAML expression for 2 days prior to dissection. At E11.5, Tie1tTA:dnMAML embryos were necrotic, with visible developmental delay(A); whereas the VEtTA:dnMAML embryos were not hemorrhagic(B). At E12.5 Tie1tTA:dnMAML embryos showed hemorrhages localized in the forebrain region with sporadic hemorrhages in the hindbrain (C, red arrow indicates location of hemorrhages). At E13.5, there were widespread hemorrhages in both the forebrain and the hind brain with additional hemorrhagic lesions at the midbrain and the tip of the tail (E). At E14.5, the embryos displayed hemorrhaging in the tip of the snout and the interdigital zone of the developing limbs (G). At all the time points examined, there were no gross morphological defects in the VEtTA:dnMAML embryos. 79 Table 3.1.. Summary of embryo phenotypes Genotype Embryo stage TET withdraw Gross morphology Tie1tTA:dnMAML E9.5 n/a developmentally delayed, beating heart, n = 4 Tie1tTA:dnMAML E10.5 n/a developmentally delayed, hemorrhagic, n = 8 Tie1tTA:dnMAML E11.5 n/a necrotic, n = 3 Tie1tTA:dnMAML E12.5 E11.5 normal, GFP positive, n = 10 Tie1tTA:dnMAML E14.5 E13.5 normal, GFP positive, n = 5 Tie1tTA:dnMAML E11.5 E9.5 necrotic, n = 3 Tie1tTA:dnMAML E12.5 E10.5 hemorrhage in the head region, beating heart, n = 17 Tie1tTA:dnMAML E13.5 E11.5 hemorrhage in the head and tail, n = 4 Tie1tTA:dnMAML E14.5 E12.5 necrotic, hemorrhage in the facial region, limb and tail, n =12 VEtTA:dnMAML E9.5 n/a developmentally delayed, beating heart, n = 3 VEtTA:dnMAML E10.5 n/a developmentally delayed, beating heart, n = 4 VEtTA:dnMAML E11.5 n/a necrotic, n = 3 VEtTA:dnMAML E11.5 E9.5 normal, GFP positive, n = 2 VEtTA:dnMAML E12.5 E10.5 normal, GFP positive, n = 8 VEtTA:dnMAML E13.5 E11.5 normal, GFP positive, n = 3 VEtTA:dnMAML E14.5 E12.5 normal, GFP positive, n = 4 TET = tetracycline 80 Figure 3.11. Tie1tTA and VEtTA transgenic mice in C57BL/6J background behave similarly to the original albino strains. (A-B) β-gal activity in the descending aorta of E12.5 embryos was detected by X-gal staining. In B6Tie1tTA:tetOSLacZ embryos, β-galactosidase activity was detected in both the endothelium and the peri-endothelial cells (A); B6VEtTA:tetOSLacZ embryos showed only endothelial β-galactosidase activity (B). At E12.5 B6Tie1tTA:dnMAML embryos showed hemorrhages localized in the forebrain region with sporadic hemorrhages in the hindbrain (C, red arrow indicates location of hemorrhages). B6VEtTA:dnMAML embryos did not have hemorrhagic lesions. The B6 strain transgenics showed similar phenotype as the transgenics of albino background strains. 81 The two driver transgenic mice, Tie1tTA and VEtTA, were derived from and maintained in different background strains. To eliminate strain-dependent difference in phenotypes, we backcrossed both Tie1tTA and VEtTA transgenic mice into the C57BL/6J (B6) background. After five generations of backcrossing, the promoter activity of both transgenics was examined by crossing with the β-gal reporter mouse. Peri-endothelial β-gal acitivity was observed with the B6-Tie1tTA driver, but not the B6-VEtTA driver mouse (Figure 3.11A and B), showing that the phenotypic discrepancies between the Tie1 or VE-cadherin- promoter-driven transgenic mice are independent of the strain difference. Similarly, the hemorrhagic phenotype was obtained when B6Tie1tTA was used to drive dnMAML expression, but not in B6VEtTA:dnMAML embryos (Figure 3.11C and D). Therefore, the strain difference between the two driver transgenic mice did not cause the differences in phenotype observed. To further study the discrepancies of phenotype obtained by the two endothelial promoters, we examined the expression of the dnMAML-GFP fusion protein by flow cytometry in both Tie1tTA:dnMAML and VEtTA:dnMAML E12.5 embryos after two days of tetracycline withdrawal. For both strains of transgenic embryos, close to 90% of the CD31 positive endothelial cells expressed dnMAML-GFP after two days of induction (Figure 3.12B and C). Furthermore, the mean GFP fluorescence intensities for the two strains were comparable (Figure 3.12C), showing equal expression level of dnMAML-GFP with both promoters. Immunofluorescence analysis of the descending aorta for both strains also showed co-localization of CD31 and GFP in the endothelium (Figure 3.13). Interestingly, several CD31+ cells co-expressing dnMAML-GFP were observed in the peri-endothelial layers in the aorta, potentially representing the Tie1+CD31+ precursor cells isolated from Tie1tTA:LacZ embryos. The CD31 staining in these peri-endothelial cells appeared dimmer compared to the staining in the endothelium, suggesting they are not part of the mature endothelium (Figure 3.13). Blockade of Notch signaling in endothelial cells alone, as shown 82 by the VEtTA:dnMAML embryos, was not sufficient to generate the hemorrhagic phenotype observed. This suggested that the hemorrhagic defect is due to blockade of Notch signaling in a Tie1+CD31+VE-cadherin- population. Figure 3.12. Expression of the dnMAML construct is comparable between Tie1tTA:dnMAML and VEtTA:dnMAML embryos. E12.5 Tie1tTA:dnMAML and VEtTA:dnMAML embryos were digested into single cells and stained for the endothelial marker CD31. GFP and CD31 expression was analysed by flow cytometry(A) . Approximately 90% of the endothelial cells (CD31 positive) expressed dnMAML-GFP fusion protein two days after tetracycline withdrawal (B). The mean fluorescence intensity of GFP was 117 unit and 136 unit for Tie1 (solid) and VE-driven (dashed) expression respectively (C). Results shown are mean ± SEM. 83 Figure 3.13. Tie1 promoter drives dnMAML expression in endothelial cells and CD31dim perivascular cells. VEtTA:dnMAML double transgenic E12.5 embryos with 2-day induction of dnMAML shows correspondence of GFP expression and CD31 staining in the endothelial cells of the descending aorta. In Tie1tTA:dnMAML double transgenic embryos of the same stage, dnMAML-GFP expression is detected in both the endothelial cells and in perivascular cells (red arrows). These GFP-positive cells also show expression of CD31 that is lower than that of the endothelial cells. 84 3.5. Notch signaling is required for differentiation of Tie1-positive precursors into vascular smooth muscle cells Occurrence of hemorrhage often suggests defects in structural integrity of vessels. Since the Tie1+VE-cadherin- vascular cells represent a potential VSMC precursor cell, we examined smooth muscle coverage around the arteries by staining for SMA in the Tie1tTA:dnMAML double transgenic embryos. E12.5 embryos with two day induction of dnMAML were sectioned and the thickness of the SMA staining around the carotid arteries and descending aorta were quantified using the NIH image software. In both the descending aorta and carotid arteries, there were decreases in the thickness of VSMC in the double transgenic embryos compared to single transgenic littermates (Figure 3.14A to H). The vertebral artery in the Tie1tTA:dnMAML mutants appeared collapsed with no VSMC around the vessel. In the VEtTA:dnMAML double transgenic embryos, however, no difference was observed in smooth muscle coverage of the arteries (Figure 3.14I to P), suggesting that the effect was solely due to the blockade of Notch signal in the Tie1+ VE-cadherin- population, not due to a paracrine effect or heterotypic signaling from the endothelium to the VSMC. Figure 3.14. Blockade of Notch signaling in Tie1-positive precursor cells impedes vascular smooth muscle differentiation in vivo. (A-F) SMA expression in E12.5 embryonic arteries in Tie1tTA:dnMAML mice. In the descending aorta (A and B), vertebral arteries (C and D), and carotid arteries (E and F), there was a reduction of SMA staining in double transgenic embryos (Double Tg) compared to single transgenic littermates (Single Tg). When the thickness of the SMA staining was quantified, the difference was shown to be statistically significant (P = 0.009 for the descending aorta and P < 0.0001 for the carotid artery) (G and H). (I-N) SMA expression was analyzed for E12.5 embryonic arteries in VEtTA:dnMAML mice and showed no significant difference in VSMC thickness between Double Tg and Single Tg (P = 0.67 for the descending aorta and P = 0.10 for the carotid artery). Bar = 50 µm. P-value was determined by Student’s t-test. 85 86 We also examined the overall percentage of VSMC by surface PDGFR-β staining in E10.5 embryos after one day dnMAML induction. At E10.5, PDGFR-β expression has been described to be expressed mainly in endocardial cells and the peri-aortic mesenchyme (Shinbrot, Peters et al. 1994). In the Tie1tTA:dnMAML double transgenic embryos there was a significant decrease in the amount of PDGFR-β+ cells compared to littermate controls (Figure 3.15). Moreover, High et al. showed that when dnMAML was expressed in smooth muscle cells with the SM22-α promoter, there were no defects in smooth muscle cells of the developing aorta observed (High, Zhang et al. 2007). All together, our findings and observations by High and colleagues suggest that Notch activation is required in the Tie1+ precursor cells, but not in mature endothelial cells or mature VSMC, for VSMC development during murine embryogenesis. Figure 3.15. Expression of dnMAML in E10.5 embryos decreases the percentage of PDGFR-β positive cells. E10.5 Tie1tTA:dnMAML murine embryos were digested into single cells. Percentage of PDGFR-β positive cells were determined by flow cytometry. Result presented as mean ± SEM of 7 embryos. P-value was determined by Student’s t-test. 87 3.6 Notch activation is not required in Tie1+ precursors after VSMC fate is acquired When dnMAML expression was induced at different time points during development in the Tie1tTA:dnMAML mutant embryos, the pattern of hemorrhagic regions was altered (Figure 3.10). To determine whether VSMC coverage of the same vessels was affected by blocking Notch in Tie1+ precursors at a later time point, we examined the descending aorta in E14.5 Tie1tTA:dnMAML embryos with two day induction of dnMAML. Interestingly, we did not observe a difference in the VSMC coverage, although robust dnMAML-GFP expression was detected in the endothelium (Figure 3.16A and B). However, in the E14.5 descending aorta, no peri-endothelial expression of dnMAML-GFP was observed, suggesting either a decrease in the Tie1+VE-cadherin- population or the lack of involvement of this population in VSMC formation in the descending aorta at E14.5. At E12.5, when dnMAML was induced in these embryos, there are already multiple layers of VSMC investing the descending aorta (Figure 3.14A and I). We speculated that blocking Notch in Tie1+ cells does not affect the smooth muscle coverage of arteries that contained VSMC prior to induction of dnMAML expression. This further demonstrates that the Tie1+ population contains local VSMC precursors necessary for the de novo formation of VSMC, but not necessary for further VSMC expansion of more mature arteries. This would also suggest that Notch signaling was involved in the differentiation of nascent VSMC in newly established arteries. 88 Figure 3.16. Blocking Notch signaling in Tie1+ precursor cells does not affect VSMC coverage of already established arteries. (A) Tie1 promoter driven expression of dnMAML- GFP for two days did not decrease SMA staining in E14.5 descending aorta. (B) Quantification of smooth muscle thickness showed no difference between the single transgenic controls and the double transgenic embryos. P-value was determined by Student’s t-test. 89 Different regions of the same vessel may also mature at different times during development. We therefore examined smooth muscle coverage along the entire length of the carotid artery to determine whether Notch blockade had differential effects on smooth muscle generation depending on the distance away from the aortic arch. While both the proximal and distal halves of the carotid artery in Tie1tTA:dnMAML embryos showed significantly less smooth muscle coverage compared to littermate controls (Figure 3.17C to E), the decrease in VSMC in the distal region of the carotid artery was much more pronounced (Figure 3.17F). This finding suggests that as the embryo grows, the distal region of the artery is undergoing active remodeling, and blocking Notch signaling in Tie1+ precursor cells had a more apparent effect on VSMC development. The results also suggest that the Tie1+ cells represent local VSMC precursors and that there is minimal migration of the precursors along the same vessel in the rostro-caudal axis. Moreover, once the artery stabilizes, the requirement for Notch signaling in these precursor cells is reduced or eliminated. Taken together, these data illustrate the presence of a Tie1+ VE-cadherin- VSMC precursor that requires Notch signaling for VSMC differentiation. 90 Figure 3.17. Effect of Notch blockade is more evident in distal portion of carotid artery compared to region proximal to the aorta. (A) Smooth muscle actin staining was used to determine smooth muscle thickness around carotid arteries at different locations. SMA Immunofluorescence staining of representative carotid sections showed a more apparent decrease in smooth muscle coverage at distal end of the carotid artery. (B and C) Student t- test was performed on the carotid smooth muscle thickness between the single transgenic controls (Single Tg) and the double transgenic mutant (Double Tg) embryos. (D) The difference is greater in the distal carotid compared to the proximal carotid. P-value was determined by Student’s t-test. 91 3.7. Discussion VSMC can be derived from different embryonic tissues during development (Majesky 2007). Previous studies on the developmental origins of VSMC often focused on tissue defined population of progenitor cells, such as the neural crest cells (Jiang, Rowitch et al. 2000) and the somites (Pouget, Pottin et al. 2008). In this study, we have shown that during mouse vascular development, there is a Tie1+ precursor population that can give rise to VSMC in a variety of different arterial beds, including ones that were previously identified as neural crest or somite-derived. Whether this Tie1-promoter active progenitor represents a common subpopulation within the previously described VSMC origins is still under investigation. Other vascular progenitor cells, such as the mesoangioblasts (Minasi, Riminucci et al. 2002) and the ESC-derived Flk-1+ cells (Yang, Soonpaa et al. 2008), have been isolated and cultured with in vitro techniques. However, a direct visualization of the progenitor cell fate during normal vascular development using Flk1-cre knock-in mouse shows that Flk1+ progenitor cells can give rise to cardiac and skeletal muscle, but not VSMC (Motoike, Markham et al. 2003). A vascular progenitor can also be isolated from differentiating ES cells by selecting for cells with Tie1-promoter activity (Marchetti, Gimond et al. 2002). This finding strengthens our observation that Tie1 is a marker that can be used to purify VSMC precursor cells. Thus, this study shows for the first time the presence of an immediate vascular precursor that can differentiate into smooth muscle cells in vivo during mammalian embryogenesis. Many signaling pathways have been implicated in smooth muscle development (Yoshida and Owens 2005) and there have been several recent studies looking at the role of Notch signaling in VSMC phenotype (reviewed by (Morrow, Guha et al. 2008)). By in vitro experiments, it has been shown that Notch signaling can up-regulate transcription of mesenchymal markers, SMA (Noseda, Fu et al. 2006) and PDGFR-β (Jin, Hansson et al. 2008), and can drive mesenchymal transdifferentiation (Noseda, McLean et al. 2004). The 92 effect of Notch signaling in VSMC of the neural crest origin (High, Zhang et al. 2007) and the somitic origin during development (Ben-Yair and Kalcheim 2008; Sato, Watanabe et al. 2008; Shin, Nagai et al. 2009) has been studied using mammalian and avian models, respectively. In the avian dorsal aorta development models, somitic cells expressing activated Notch receptor are found to preferentially integrate into the dorsal aorta as both endothelial cells and VSMC. In this study, we have seen that Notch signaling is required in the differentiation of VSMC precursor cells that can give rise to VSMC of different embryonic arteries in a mammalian system. The Tie1+ precursor likely represents an immediate precursor predisposed for VSMC fate, regardless of where the cells originally migrated from (Figure 3.18). The tissue-specific requirement for Notch in VSMC differentiation shown in previous studies may simply be reflecting the effect of Notch activation on this distinct Tie1+ population. We also observed that sustained Notch activation is not necessary for further maturation of the vessel. Interestingly, Proweller and colleagues also showed that Notch activity is not required for maintenance of embryonic VSMC (Proweller, Wright et al. 2007). Taken together, the findings suggest that arteriogenesis in embryonic vasculature involves first a Notch- dependent differentiation of local Tie1+ VE-Cadherin- precursor cells into VSMC, followed by the expansion of VSMC, which may be Notch-independent. Failure of local progenitors to differentiate into VSMC also provides insight into the process of embryonic vascular development as the pattern of defects caused by the expression of dnMAML corresponds to sites of active arteriogenesis. Our results suggest that the differentiation of VSMC in the thoracic dorsal aorta is initiated prior to E10.5 as induction of dnMAML expression at E10.5 did not completely deplete, but only reduced, VSMC in descending aorta at E12.5. For the region of carotid artery distal to the aorta, however, dnMAML expression at E10.5 severely reduced the smooth muscle coverage of the vessel at E12.5, revealing that VSMC development occurs first in the thoracic region of the embryo, 93 Figure 3.18. Proposed role of Notch activation in embryonic arteriogenesis. The Tie1+ CD31+ VE-cadherin- population contains a local common vascular smooth muscle cell (VSMC) precursor. Notch activation is required for the differentiation of the precursor cells into mature VSMC at the onset of arteriogenesis of nascent arteries. Once the artery is invested with VSMC, Notch activation in the precursor is no longer required. then progresses cranially and caudally towards the extremities. This is also demonstrated by the later onset of hemorrhaging in the tail when Notch blockade is induced at E11.5. The same conclusion was drawn by Takahashi and colleagues (Takahashi, Imanaka et al. 1996) when they examined the expression of SMA during murine embryogenesis. They did not observe SMA expression around the vertebral arteries until E11.5, which supports our observation that VSMC differentiation of the vertebral artery is completely disrupted at E12.5 by Notch blockade initiated at E10.5. Using the Tie1tTA:dnMAML transgenic system, we can potentially map out the process of embryonic arteriogenesis in embryos of all stages. 94 Chapter 4. NOTCH ACTIVATION PROMOTES ENDOTHELIAL SURVIVAL THROUGH A PI3K-SLUG AXIS 4.1. Introduction Vascular homeostasis in the adult requires active maintenance by a balance of signals that leads to a non-proliferative, non-angiogenic, and non-apoptotic endothelial monolayer. Apoptosis is a form of death in which the cell participates in its own demise. It is a tightly regulated process that maintains the homeostasis of a biological system. Many injurious stimuli are present in the bloodstream, to which endothelial cells, the cells lining the inside of blood vessels, are constantly being exposed. Therefore, the endothelium must develop mechanisms to ensure its resistance to apoptotic agents without disturbing the homeostatic state. Endothelial apoptosis is associated with the initiation and progression of atherosclerosis (Alvarez, Gips et al. 1997; Tricot, Mallat et al. 2000) and attributed to the complication of sepsis (Bannerman and Goldblum 2003), among other cardiovascular diseases (Stefanec 2000). In atherosclerosis, plaques develop containing cholesterol deposits, leukocytes, platelets and smooth muscle cells, leading to the narrowing of vessels. Apoptotic endothelial cells have been observed in patients suffering from atherosclerosis (Alvarez, Gips et al. 1997; Tricot, Mallat et al. 2000). Platelets and leukocytes have increased adherence to apoptotic endothelial cells which may contribute to the progression of plaque formation (Bombeli, Schwartz et al. 1999; Schwartz, Karsan et al. 1999). In vitro experiments have also shown that apoptotic endothelial cells stimulate the survival and proliferation of VSMC, leading to intimal thickening and growth of plaque (Raymond, Desormeaux et al. 2004; Sakao, Taraseviciene-Stewart et al. 2007). One of the risk factors for atherosclerosis is hyperhomocysteinemia, where patients experience an elevated plasma concentration of total homocysteine (Clarke, Daly et al. 1991; McCully 1996). Homocysteine is a metabolic product 95 in the conversion between methionine and cysteine. Serum homocysteine levels can be increased through genetic mutation of enzymes in the homocysteine metabolic pathway or dietary deficiency of vitamin B’s required for homocysteine metabolism. Homocysteine stimulation has been shown to cause endothelial apoptosis both in vitro and in vivo (Zhang, Cai et al. 2001; Hossain, van Thienen et al. 2003; Wilson and Lentz 2005). Another human disease that is associated with endothelial death is sepsis. Sepsis is a systemic inflammatory disorder whose complications include systemic vascular collapse, multi-organ failure and acute respiratory distress (Bannerman and Goldblum 2003). LPS, present on gram-negative bacteria, has been shown to induce sepsis (Parrillo 1993). LPS stimulation induces endothelial apoptosis in vitro and in vivo (Haimovitz-Friedman, Cordon- Cardo et al. 1997; Bannerman and Goldblum 2003). Endothelial apoptosis leads to detachment of cells from the vessel, activation of coagulant, and increase in vascular permeability. These endothelial defects may exacerbate the sepsis syndrome, especially in the lung, where respiratory distress can be caused by edema. Inhibition of endothelial apoptosis has been shown to reduce the level of acute lung injury in LPS-challenged animals (Kawasaki, Kuwano et al. 2000). Induction of endothelial survival signaling, therefore, may protect against the effect of these injurious chemicals in the bloodstream and halt the progression of some cardiovascular diseases. Several signaling pathways have been implicated in the precise balance of survival and apoptosis in endothelial cells. Results from our lab and others have illustrated that Notch signaling plays a very important role in endothelial biology (MacKenzie, Duriez et al. 2004; Karsan 2005; Noseda, Niessen et al. 2005). Our lab has previously shown that activation of Notch1 and Notch4 can protect endothelial cells against LPS-induced apoptosis (MacKenzie, Duriez et al. 2004). The Notch- induced pro-survival effect is partially mediated though the induction of Bcl-2 and the maintenance of mitochondrial membrane integrity (MacKenzie, Duriez et al. 2004). However, 96 other mechanisms are involved in Notch-induced survival signaling. This section of the thesis focuses on the analysis of Notch-induced survival signaling, presents evidence that the survival function is context-dependent, and provides a possible mechanism to explain the context-dependent activity of the Notch pathway in apoptosis. 97 4.2. Notch protects against LPS, but not homocysteine,-induced apoptosis The conflicting observations on the effects of Notch activation in endothelial apoptosis in vivo, as discussed in section 1.3.4, suggest a context-dependent role for Notch in the induction of survival signals. To further examine the effect of Notch activation on endothelial survival, a human dermal microvascular endothelial cell line, HMEC, was treated with two different apoptotic stimuli known to affect endothelial cell function in the context of cardiovascular disorder. HMEC transduced with the constitutively active Notch construct or empty vector control were both stimulated with LPS and ALLN (N-Acetyl-L-leucyl-L-leucyl-L- norleucinal, a proteasome inhibitor) or homocysteine for 16 hours under serum-starved conditions. ALLN treatment blocks LPS-induced cell survival signals via inhibition of NFκB activation thereby revealing LPS-induced apoptosis (Zen, Karsan et al. 1998). Cell survival was monitored by neutral red uptake and was quantified as a percentage of the untreated cells. Notch activation in HMEC led to a higher proportion of live cells when stimulated with LPS and ALLN compared to control cells, but decreased cell survival when treated with homocysteine (Figure 4.1A). As neutral red uptake quantifies live cells, the increase in cell number could be caused by cell proliferation rather than survival. To confirm that apoptosis is affected in Notch activated cells, LPS or homocysteine-induced apoptosis was examined by the Annexin V binding assay. Fluorescent dye-conjugated annexin V binds to phosphatidylserine on the surface of early apoptotic cells, which enables quantification of apoptosis by flow cytometry (Koopman, Reutelingsperger et al. 1994; Vermes, Haanen et al. 1995). Upon LPS treatment, Notch activated cells showed a marked decrease in the apoptotic population compared to the vector control cells (Figure 4.1B), while Notch activation exacerbated the apoptotic effect of homocysteine (Figure 4.1C). Thus, in endothelial cells, Notch activation can act either as an anti-apoptotic factor or a pro-apoptotic factor depending on the stimulus, possibly through interaction with other pathways. 98 Figure 4.1. Notch activation protects endothelial cells against LPS-induced apoptosis while enhancing homocysteine-induced apoptosis. (A) Endothelial cells transduced with vector control or activated Notch (NotchIC) were treated with 100 µg/mL LPS and 25 µM of ALLN (n = 10) or 7.5 mM homocysteine (n = 7) for 16 hours. Cell survival was determined by neutral red uptake. NotchIC cells showed higher level of survival when treated with LPS, but lower level of survival when treated with homocysteine, compared to vector control cells. (B and C) The difference observed in cell survival is due to the effect of Notch activation on apoptosis as determined by the Annexin V binding assay. NotchIC and vector control HMEC were treated with 100 µg/mL LPS and 25 µM of ALLN (n = 5) or 7.5 mM homocysteine (n = 4) for 8 hours. Results shown are mean ± SEM. P-value was determined by Student’s t-test. 99 4.3 Notch signaling activates the PI3K pathway through a secreted factor In endothelial cells, homocysteine and LPS have the ability to activate some of the same apoptotic pathways (see Table 1.2). However, while LPS can also stimulate anti- apoptotic signaling through activation of the PI3K pathway (Wong, Hull et al. 2004), homocysteine interferes with PI3K signaling in endothelial cells (Suhara, Fukuo et al. 2004). To examine whether Notch activation can interact with the PI3K pathway in endothelial cells, two different types of human endothelial cells were examined. Since some signaling pathways may be altered in transformed cell lines such as HMEC, primary human umbilical vein endothelial cells (HUVEC) were used alongside HMEC to verify the findings. Both HMEC and HUVEC expressing NotchIC exhibited higher PI3K activity as shown by increased phosphorylation of its downstream effector Akt (Figure 4.2A). Since PI3K signaling is downstream of many growth factor receptors, we examined whether Notch activates PI3K through a cell-autonomous mechanism or through a secreted factor. Low-serum containing medium conditioned by NotchIC expressing HMEC or vector control HMEC was applied to parental HMEC. Only the medium conditioned with Notch activated HMEC was able to stimulate PI3K in the parental HMEC (Figure 4.2B). Densitometry was used to quantify the ratio between phosphorylated and total amount of Akt (Figure 4.2C). Statistical analysis showed a significant induction of Akt phosphorylation by medium conditioned with NotchIC-expressing HMEC. The phosphorylation of Akt was detected in the first 5 minutes after application of the medium (Figure 4.2B). This rapid response implicates a direct activation of PI3K, rather than a secondary effect through other pathways, supporting the presence of a secreted factor induced by Notch. The identity of the secreted factor remains under investigation. 100 Figure 4.2. Notch activation in endothelial cells leads to activation of PI3K signaling through a secreted factor. (A) Endothelial cells (HMEC or HUVEC) transduced with activated Notch (NotchIC) showed an increased level of Akt phosphorylation (pAkt) by immunoblotting with phospho-specific antibody, indicative of increased PI3K activity. (B) Medium conditioned by Notch activated cells or vector control cells was applied onto parental HMEC cells. Conditioned medium from Notch activated cells induced Akt phosphorylation while vector control conditioned medium did not. (C) The level of Akt phosphorylation was quantified by densitometry and the ratio of phosphorylated to total Akt was determined. Results shown are mean ± SEM of 4 experiments. P-value was determined by Student’s t- test. 101 4.4. PI3K activity is essential for survival of Notch-activated endothelial cells In many Notch-dependent cancer cell lines, the Notch pathway induces oncogenesis through the activation of PI3K which provides important cell survival signals in cancer cell lines (Mungamuri, Yang et al. 2006; Calzavara, Chiaramonte et al. 2008; Meurette, Stylianou et al. 2009). To examine the role of PI3K activation in Notch-activated endothelium, NotchIC- expressing endothelial cells were treated with a reversible chemical inhibitor of PI3K, LY294002. Following treatment with LY294002, there was an observable increase in cell death in NotchIC-expressing endothelial cells, but not in vector control cells. Immunofluorescence staining for activated effector Caspase 3 was used to determine the percentage of apoptotic cells in vector control or Notch-activated cells treated with LY294002. While there were few (around 10%) apoptotic cells in either untreated vector control endothelial cells, NotchIC-expressing cells, or LY294002-treated control cells, there was a marked increase of apoptosis in Notch-activated endothelial cells treated with the PI3K inhibitor (Figure 4.3A, B). Induction of apoptosis by PI3K pathway inhibition in Notch-activated cells was verified with another PI3K inhibitor, wortmannin, since inhibitor specificity may be a concern. Treatment with both LY294002 and wortmannin showed increased apoptosis in Notch-activated cells by the Annexin V binding assay (Figure 4.3C). Wortmannin is less stable when diluted to working concentration at physiological pH (Holleran, Egorin et al. 2003), which may explain the reduced effect of wortmannin-induced apoptosis compared to LY294002 treatment in NotchIC cells (Figure 4.3C). These results showed that PI3K activity is essential for survival of endothelial cells when Notch is activated. This requirement for PI3K was not observed in endothelial cells transduced with the empty vector, suggesting that Notch activation leads to induction of parallel pro-apoptotic pathways as well as a PI3K survival pathway. The opposing pathways 102 are balanced in Notch-activated cells, and the inhibition of PI3K tips the scales towards the apoptotic signals. Figure 4.3. PI3K pathway activity is necessary for the survival of Notch-activated endothelial cells. (A and B) HMEC transduced with vector control or NotchIC were treated with DMSO or 40µm PI3K inhibitor LY294002 for 8 hours. Apoptotic cells were quantified (n = 3) by immunofluorescence staining with antibody against activated caspase 3 (red) and counterstained with DAPI (blue). (C) HMEC transduced with vector control or NotchIC were treated with DMSO (vehicle), 40 µm LY294002 (LY) or 1 µM wortmannin (wort) for 16 hours. Percentage of apoptotic cells was determined by Annexin V binding. Results shown are mean ± SEM of 4 experiments. P-value was determined by the Student’s t-test. 103 4.5. PI3K activity is required for Notch-induced Slug expression We have shown that PI3K activity is required for endothelial survival when Notch is activated. Previous work from our lab has suggested that the expression of a direct transcriptional target of Notch, Slug, is also important for cell survival in Notch-activated endothelial cells (Niessen, Fu et al. 2008). To determine whether there is a link between the expression of Slug and PI3K activity, we evaluated the expression of Slug in Notch activated cells with and without PI3K inhibition. In HMEC, there was no observable basal level of Slug protein expression and Notch activation up-regulated Slug as previously observed (Niessen, Fu et al. 2008). Treatment with LY294002 decreased Notch-induced Slug expression (Figure 4.4A,B). Our lab has previously shown that Notch activation induces Slug through CSL- dependent transcriptional activation (Niessen, Fu et al. 2008). An additional level of regulation is shown here as Notch activates PI3K, which is in turn required for Slug expression. Since Slug expression is required for endothelial survival in Notch activated cells, the apoptosis induced by inhibition of PI3K observed may be through down-regulation of Slug. These data show that Notch activation induces a PI3K-Slug signaling axis important for endothelial survival. 104 Figure 4.4. PI3K activity is required for Notch-induced Slug expression. (A) HMEC transduced with NotchIC showed induction of Slug expression by immunoblotting. The expression of Slug was down-regulated by inhibition of PI3K with 40 µM of LY294002 (LY) for 4 hours. (B) The level of Slug expression was quantified by densitometry and the ratio of Slug to Tubulin was determined. Results shown are mean ± SEM of 5 experiments. P-value was determined by Student’s t-test. 105 The most studied downstream effector of the PI3K pathway is Akt, which is required for many of the previously described functions of PI3K. To determine whether PI3K modulates Slug expression through Akt, a kinase-dead Akt dominant-negative mutant (dnAKT) with three amino acid replacements (K179M, T308A, S473A) was used to inhibit Akt-dependent signaling (Sakoda, Gotoh et al. 2003). Expression of dnAKT did not decrease Notch-induced Slug expression (Figure 4.5A). In addition, treatment with the Akt inhibitor triciribine, which inhibits all three AKT family members, also did not reduce Slug expression in NotchIC-expressing cells (Figure 4.5B), suggesting that PI3K regulates Slug expression through an Akt-independent mechanism. High concentrations of LY294002 are also known to inhibit signaling through mTOR (mammalian target of rapamycin) (Brunn, Williams et al. 1996). To confirm that the effect of LY294002 treatment was not through inhibition of mTOR, NotchIC-expressing HMEC were treated with rapamycin, a chemical inhibitor of mTOR. Rapamycin did not inhibit Notch-induced Slug expression (Figure 4.5C). However, rapamycin only inhibits mTOR signaling through the raptor-mTOR complex without blocking signaling through the rictor-mTOR complex (Sarbassov, Ali et al. 2004). Therefore, we could not definitively conclude that Slug expression is not regulated by mTOR activity. 106 Figure 4.5. Akt or mTOR activity is not required for Notch-induced Slug expression. (A) Co-transduction of a dominant-negative form of Akt (dnAKT) (myc-tagged) with NotchIC did not down-regulate Notch-induced Slug protein as detected by immunoblotting. (B) Inhibition of Akt (25 µM triciribine) or (C) mTOR (1 µM rapamycin) did not alter Notch-induced Slug protein expression. The expression of Slug was down-regulated by inhibition of PI3K with 40 µM of LY294002 (LY) for 4 hours. 107 To further examine the mechanism of PI3K-dependent Slug expression, mRNA was isolated from HMEC transduced with either NotchIC or empty vector and treated with either LY294002 or DMSO vehicle. Transcript level of Slug was quantified using qRT-PCR. Notch activation up-regulated expression of its direct targets Hey1 and Slug in endothelial cells, confirming previous observations (Figure 4.6A). Interestingly, both basal and Notch-induced Slug transcripts were reduced by PI3K inhibition (Figure 4.6B). This decrease was not caused by a generalized inhibition of Notch-dependent transcription, as the mRNA level of Hey1 was not affected by PI3K inhibition (Figure 4.6C). This finding suggests that PI3K is able to regulate Slug transcription via a Notch-independent mechanism or that PI3K activity regulates Slug transcript stability. Further experiments are required to distinguish between the two mechanisms. 108 Figure 4.6. PI3K inhibition decreases both basal and Notch-induced Slug transcript level. (A) NotchIC expression in HMEC up-regulated mRNA level of both Hey1 and Slug. (B) Both endogenous and Notch-induced Slug transcription was reduced by PI3K inhibition. HMEC transduced with NotchIC or vector control were treated with 40 µM of LY294002 or DMSO vehicle for 4 hours. The mRNA for Slug (B) and Hey1 (C) were quantified by quantitative RT-PCR. Results shown are mean ± SEM of 5 experiments. P-value was determined by Student’s t-test. 109 4.6. Slug protects endothelial cells against LPS-induced apoptosis In other cell types, Slug has been shown to be an anti-apoptotic factor, protecting against DNA-damage-induced cell death (Wu, Heinrichs et al. 2005; Kurrey, Jalgaonkar et al. 2009). The requirement for Slug expression for cell survival in Notch activated endothelium suggests that Slug also has anti-apoptotic function in endothelial cells. To investigate the possibility that Notch imparts its protective activity against LPS stimulation through Slug, Slug transduced HMEC were treated with LPS and ALLN or homocysteine. Slug expression in HMEC led to a higher proportion of live cells when the cells were stimulated with LPS and ALLN, but cell survival was unchanged compared to control when cells were treated with homocysteine (Figure 4.7A). To confirm that apoptosis is affected in Slug-expressing cells, LPS or homocysteine-induced apoptosis was examined by the Annexin V binding assay. Upon LPS treatment, Slug-expressing cells showed a marked decrease in the apoptotic population compared to the vector control (Figure 4.7B), while Slug again did not inhibit homocysteine-induced apoptosis in HMEC (Figure 4.7C). The ability of Slug to convey an anti-apoptotic effect against LPS stimulation showed that Slug may be the downstream effector for Notch-induced protection. Slug did not offer the same protection against homocysteine-induced apoptosis, but unlike Notch, Slug did not exacerbate the effect of homocysteine treatment. The lack of Slug-induced protection against homocysteine treatment may indicate that the downstream signaling from Slug does not interfere with homocysteine- induced apoptotic signal; alternatively, homocysteine treatment may inhibit the pro-survival activity of Slug. This finding also shows that Slug does not induce parallel apoptotic pathways. 110 Figure 4.7. Slug protects endothelial cells against LPS-induced apoptosis while exhibiting no effect on homocysteine-induced apoptosis. (A) Endothelial cells transduced with vector control or Slug were treated with 100 µg/mL LPS and 25 µM of ALLN (n = 4) or 7.5 mM homocysteine (n = 4) for 16 hours. Cell survival was determined by neutral red uptake. Slug cells showed higher level of survival when treated with LPS compared to vector control cells. (B and C) The difference observed in cell survival was due to the effect of Slug expression on apoptosis as determined by Annexin V binding. Slug and vector control HMEC were treated with 100 µg/mL LPS and 25 µM of ALLN (n = 4) or 7.5 mM homocysteine (n = 3) for 8 hours. Results shown are mean ± SEM. P-value was determined by Student’s t-test. 111 4.7. Homocysteine induces apoptosis in Notch-activated cells by regulating PI3K and Slug Since both PI3K activity and Slug expression were important for the survival of Notch- activated cells, we hypothesized that homocysteine may interfere with either PI3K or Slug to induce an apoptotic phenotype in endothelial cells expressing NotchIC. Homocysteine treatment in Notch-activated cells reduced Akt phosphorylation, suggesting a decrease in PI3K activity (Figure 4.8A). Notch-induced Slug expression was also down-regulated by homocysteine treatment (Figure 4.8A). This result suggests that homocysteine stimulation leads to increased cell death in Notch-activated cells by blocking essential anti-apoptotic signals. On the other hand, LPS treatment of NotchIC-expressing HMEC did not change Slug protein expression (Figure 4.8A). These findings suggest that LPS stimulation may be able to enhance the endothelial survival signal through the PI3K-Slug axis in Notch activated cells, leading to increased cell survival compared to control cells. Homocysteine treatment, however, inhibits Notch-induced survival signaling by blocking PI3K activation and Slug expression, thereby amplifying Notch-induced apoptotic signals. Interestingly, homocysteine treatment also down-regulated Slug protein expression when Slug was expressed from an expression vector using a heterologous promoter (Figure 4.8B). This shows that homocysteine did not decrease Slug expression through transcription regulation or modulation of mRNA stability, since the mammalian expression construct did not include the endogenous Slug promoter or the 5’ and 3’ UTR region. Homocysteine may regulate Slug protein expression through modification of Slug protein stability thereby inhibiting Slug-induced anti-apoptotic activity. 112 Figure 4.8. Homocysteine blocks Notch-induced PI3K activation and Slug expression. (A) HMEC transduced with NotchIC or empty vector were treated with LPS or homocysteine (HCY) for the indicated time. The level of Akt phosphorylation and Slug expression was detected with immunoblotting. Notch1IC expression was also confirmed with immunoblotting. (B) HMEC transduced with flag-tagged Slug or empty vector were treated with LPS or HCY for 8 hours. The level of Slug expression was detected by immunoblotting. 113 4.8. Discussion This study presents a mechanism for the Notch-induced endothelial survival observed upon LPS-stimulation: the Notch-PI3K-Slug signaling axis (Figure 4.9). In this study, treatment of Notch-activated endothelial cells with the apoptotic agents, LPS or homocysteine resulted in a differential effect. The context-dependent effect of Notch depended on the interaction with the PI3K signaling pathway and the expression of Slug (Figure 4.9). In vivo studies have also shown a context-dependent effect of Notch signaling in endothelial survival (Nakajima, Yuasa et al. 2003; Limbourg, Takeshita et al. 2005; Dou, Wang et al. 2008; Trindade, Kumar et al. 2008). Whether the differential effect of Notch activation in vivo is dependent on the status of PI3K signaling still needs to be examined. 4.8.1. Possible candidates for a Notch-induced pro-apoptotic signal Our lab has previously shown that Notch activation leads to a decrease of p21CIP1 expression in endothelial cells (Noseda, Chang et al. 2004). In addition to the well-studied role as the inhibitor of cyclin-dependent kinases, p21CIP1 also acts as an anti-apoptotic factor in endothelial cells under serum-starvation (Bruhl, Heeschen et al. 2004; Mattiussi, Turrini et al. 2004). Although the downstream signals for p21CIP1 mediated protective activity are still unknown, Notch may still be inducing endothelial apoptosis through down-regulation of p21CIP1. Fig apo exp tran a s exp Not hom blo ure 4.9. Mo ptotic signa ression. N scription ac ecreted fac ression, wh ch activat ocysteine, ck Notch-ind del of Notc ling pathwa otch regula tivation via tor yet to b ich in turn ion reduce on the oth uced surviv h-induced ys. The ant tes anti-ap CSL and th e identified tips the bala s LPS-ind er hand, do al signals. survival sig i-apoptotic p optotic pro rough activa . Inhibition nce of dea uced apo wn-regulate naling. No athway dep tein Slug tion of PI3K of PI3K lea th and surv ptosis thro s PI3K act tch activates ends on PI3 by two m . Notch act ds to down ival signals ugh up-re ivation and both pro- a K activity a echanisms ivates PI3K -regulation towards ap gulation o Slug expre 114 nd anti- nd Slug : direct through of Slug optosis. f Slug; ssion to 115 Notch signaling in endothelial cells is known to interact with other pathways, most notably the TGF-β signaling pathway. Work from our lab has demonstrated synergistic interaction between Notch and TGF-β signaling through up-regulation of Smad3 (mothers against decapentaplegic homolog 3) (Fu, Chang et al. 2009). TGF-β signal can induce apoptosis in endothelial cells in a context-dependent mechanism (Tsukada, Eguchi et al. 1995; Pollman, Naumovski et al. 1999; Lu, Patel et al. 2009). Therefore, Notch-induced TGF- β signaling activity may drive the pro-apoptotic signaling observed in endothelial cells. 4.8.2. Possible mechanisms of Slug down-regulation by PI3K inhibition One recent study links PI3K activity to transcriptional regulation of Slug. Saegusa and colleagues showed that activation of Akt leads to increased nuclear β-catenin, which in turn activates transcription of Slug (Saegusa, Hashimura et al. 2009). NF-κB activity has also been shown to induce Slug transcription during frog development (Zhang, Carl et al. 2006). PI3K activity can lead to activation of NF-κB (Sizemore, Leung et al. 1999). However, both of the pathways require Akt activation. There is currently no study that shows the regulation of Slug transcription by a PI3K-dependent Akt-independent pathway. Our lab and others have shown that Slug is an unstable protein with a half-life of around 2 hours and is subjected to proteasomal degradation (Vernon and LaBonne 2006; Wang, Wang et al. 2009). Although a close family member Snail is phosphorylated by glycogen synthase kinase (GSK)-3β (which is inhibited by PI3K signaling) and targeted for proteosomal degradation through phosphorylation (Zhou, Deng et al. 2004), Slug is not a substrate of GSK-3β (Vernon and LaBonne 2006). Slug is, however, identified as a binding partner for a F-box protein Partner of paired (Ppa) in frog neural crest cells and Ppa can recruit E3 ubiquitin-ligase to Slug, leading to its degradation (Vernon and LaBonne 2006). Interestingly, a human homolog of Ppa, F-box and leucine-rich repeat protein 14 ( FBXL14), has recently been shown to induce proteasome degradation of Snail (Vinas-Castells, Beltran 116 et al. 2010). However, the degradation of Slug by FBXL14 was not examined and there is no established connection between Ppa or FBXL14 and PI3K signaling. PI3K may also act through p53 to regulate expression of Slug. The tumor-supressor p53 has been shown to regulate Slug expression by two opposing mechanisms. p53 can bind to the Slug promoter and activate Slug transcription (Wu, Heinrichs et al. 2005). However, p53 also negatively regulate Slug protein stability. Mdm2 (mouse double minute 2) is an E3 ubiquitin ligase, a p53 transcriptional target, and a substrate for Akt kinase activity (Ogawara, Kishishita et al. 2002). Mdm2 is phosphorylated by Akt, which enables interaction between Mdm2 and p53. Mdm2 then facilitates the ubiquitination and proteasomal degradation of p53 (Fuchs, Adler et al. 1998). A recent study showed that Slug is also ubiquinated by Mdm2 and degraded by the proteasome (Wang, Wang et al. 2009). Therefore, p53 activity can lead to down-regulation of Slug protein expression by induction of Mdm2, which can also be regulated by PI3K. A closer examination is required to elucidate the effect of PI3K inhibition and homocysteine treatment on these pathways to determine the mechanism of Slug transcript and protein regulation in endothelial cells. 117 Chapter 5. SUMMARY, PERSPECTIVES, AND FUTURE DIRECTIONS 5.1. Notch in arteriogenesis Data presented in Chapter 3 of this thesis suggests the presence of a Tie1+CD31+VE- cadherin- vascular smooth muscle precursor cell during embryonic vascular development. Using the transgenic mouse system that we generated, we also showed that Notch activation is required for the differentiation into VSMC from this Tie1+ precursor cell. While the focus of vascular smooth muscle development research has been on the heterogeneity of the tissue of origin (Gittenberger-de Groot, DeRuiter et al. 1999; Majesky 2007), little is known about the population of immediate precursors that are predisposed for differentiation within every vascular bed. This is the first in situ observation of an immediate vascular smooth muscle precursor cell in mammalian vascular development. We have also demonstrated that mature endothelial cells cannot be a source of VSMC in mammalian embryonic vasculature. Using a dnMAML transgenic mouse model, we showed that the requirement for Notch signaling in the Tie1+ vascular precursors depends on the maturity of the vessel. When Notch was blocked at the onset of vascular smooth muscle development, the vascular precursors fail to differentiate, causing a reduction or the complete absence of vascular smooth muscle coverage around the vessel. However, blocking Notch in the precursor population did not affect the growth of the smooth muscle layer around the more mature vessels when the vessel already contained differentiated smooth muscle cells prior to the indction of dnMAML. This suggests that development of the VSMC during embryonic arteriogenesis has at least two phases: one where smooth muscle cells are generated de novo from vascular precursor cells, followed by the second phase where the smooth muscle layer grows through the proliferation of existing cells (Varadkar, Kraman et al. 2008). We have shown that Notch is 118 required for phase one of vascular smooth muscle development, and Varadkar and colleague also showed a possible role of Notch in vascular smooth muscle expansion during embryogenesis (Varadkar, Kraman et al. 2008). 5.1.1. Establishment of a new tool: tetOSdnMAML transgenic mouse When the work presented in this thesis began, exploration of the role of Notch signaling in the field of mammalian vascular development had also just begun (Iso, Hamamori et al. 2003). The vascular phenotype of Notch1 null mutant mice demonstrated the requirement for Notch signaling in embryonic vascular remodeling and heart development (Swiatek, Lindsell et al. 1994). Similar phenotypes were recapitulated in other Notch pathway mutants (Xue, Gao et al. 1999; Krebs, Xue et al. 2000; Fischer, Schumacher et al. 2004; Gale, Dominguez et al. 2004; Krebs, Shutter et al. 2004). Endothelial-specific inactivation of Notch pathway components, using the Tie2 promoter, which is now known not to be specific to endothelial cells, confirmed the importance of Notch signaling in the vascular cell type during embryogenesis (Krebs, Shutter et al. 2004; Limbourg, Takeshita et al. 2005; High, Lu et al. 2008). However, due to early embryonic lethality of the Notch mutants at mid-gestation, the role of the Notch signaling pathway in the development of VSMC remained to be examined. To bypass the lethality caused by the cardiac development defects, we generated a tetracycline-inducible, tissue-specific transgenic mouse that expresses a pan-Notch signaling inhibitor, dnMAML. Through titration of the tetracycline treatment, Notch signaling can be blocked within one day of tetracycline withdrawal. The level of responsiveness of our transgenic system can only be rivaled by exogenous treatments such as GSI, which cannot be applied in a tissue-specific fashion. In addition, the tetracycline-inducible system provides a reversible system for Notch inhibition. Unlike the inducible Cre-Lox system, where genomic DNA is irreversibly excised (Zhang, Riesterer et al. 1996), temporal regulation of dnMAML 119 expression can be achieved by the addition or withdrawal of tetracycline from the drinking water. However, further optimization of the system is required for the efficient suppression of dnMAML. Overall, the tetOS-dnMAML transgenic mouse provides an opportunity to study the role of Notch signaling in a specific cell type at a defined time. 5.1.2. Characterization of the VSMC precursor While the studies described in Chapter 3 of this thesis provide strong evidence for the presence of the Tie1+ CD31+ VE-cadherin- VSMC precursor cells, a direct demonstration of this population’s ability to differentiate into VSMC is still lacking. Alternatively, the Tie1 promoter activity can be marking a subset of mature VSMC within all vascular beds, although the hemorrhagic defects and VSMC differentiation defects observed in the Tie1dnMAMLembryos cannot be explained by the blockade of Notch in a subset of VSMC. To further characterize the Tie1+CD31+VE-cadherin- VSMC precursor and dissect the mechanism of VSMC development, the cells will be isolated from mouse embryos and cultured ex vivo. In order to separate the precursor cells from contaminating mature endothelial cells and VSMC, single cell clones will be established and we will determine the ability of the clonal cells to differentiate into VSMC. Since the precursors express typical markers of the endothelial cells, the ability of the Tie1+CD31+VE-cadherin- precursor cells to differentiate into mature endothelial cells will also be examined. Clonal precursor cells may be examined for their capability to differentiate into other cell types. Once this VSMC precursor cell can be established ex vivo, different cytokines and chemical inhibitors can be used to examine the role of different signaling pathways in the differentiation and/or maintenance of vascular smooth muscle precursors. High and colleagues provided the first evidence for the role of Notch in mammalian vascular smooth muscle development during the course of this study. Using the same 120 dnMAML construct in a different transgenic system, they showed that blocking Notch in the neural crest cells inhibits smooth muscle differentiation in the pharyngeal arch arteries (High, Zhang et al. 2007). Here, we show the presence of Tie1+ cell-derived VSMC in neural crest- derived vessels such as the carotid artery and the aortic arch, as well as in other vascular beds. Therefore, by blocking Notch in all neural crest-derived tissue, Notch activation in the Tie1+ precursor may also be inhibited. We have shown that by blocking Notch in the Tie1+ precursor cell alone, without affecting other neural crest-derived cells, smooth muscle differentiation in the carotid arteries is blocked. It is possible that the smooth muscle phenotype of the neural crest-specific Notch inhibition is a result of blocking Notch in the Tie1+ VSMC precursors within the neural crest-derived population. Although the Tie1tTA:LacZ reporter embryos showed peri-endothelial staining in all arteries examined, and the hemorrhagic defects were observed in multiple vascular beds in the Tie1tTA:dnMAML embryos (suggesting the existence of a common VSMC precursors in different arteries) there is still no direct evidence for the existence of this precursor cell. To demonstrate that the Tie1+ population contains a common VSMC precursor, we will attempt to isolate Tie1+CD31+VE-cadherin- cells from different embryonic arteries. Using microdissection, different arteries undergoing the initiation of arteriogenesis can be isolated from Tie1tTA:tetOSLacZ embryos. If a common Tie1+ VSMC precursor cell exists, then we can isolate single cell colonies capable of differentiating into mature VSMC from different vascular beds. The presence of a common local VSMC precursor will not contradict the previously observed mosaic nature of VSMC. The precursors may still originate from different embryonic tissues, therefore may be regulated by different signaling pathways and require different mechanisms prior to arriving at their final arterial destinations. 121 5.1.3. Clinical implication of Notch-induced VSMC differentiation Blocking Notch signaling in the endothelium for two days does not affect smooth muscle development, showing that Notch activation in the endothelium is not necessary for providing any paracrine signals for the differentiation, proliferation or survival of embryonic VSMC. However, expression of the Notch ligand Jagged1 in the endothelium is required for vascular smooth muscle differentiation from neural crest progenitors (High, Lu et al. 2008). Endothelial cells may be driving vascular smooth muscle development through heterotypic signaling by providing the necessary ligands for Notch receptors on the vascular precursor cells. The human disorder, Alagille syndrome, is an autosomal dominant arteriodysplastic syndrome with multiple organ system involvement caused by mutations of the Jagged1 gene. Patients with Alagille syndrome suffer from increased incidence of intracranial bleeding, which contributes to the mortality of the disorder (Emerick, Rand et al. 1999). One can speculate that the expression of Jagged1 in the endothelium is required for stabilization of the brain vasculature through ensuring mural cell coverage of the vessels. However, the mural cells of the brain have not been closely examined to show whether vascular smooth muscle defects may play a role in the hemorrhaging observed. Moreover, missense mutations in the Notch1 gene have been associated with patients with thoracic aortic aneurysms (McKellar, Tester et al. 2007). Thoracic aortic aneurysms have previously been associated with mutations in the TGFβ receptors (Pannu, Fadulu et al. 2005), which are known to regulate VSMC differentiation and phenotypes (Bobik 2006). While the molecular and cellular implication of the Notch1 mutations are yet to be determined, it is possible that Notch-dependent VSMC differentiation/maturation plays a role in the development of aortic aneurysm. It would be interesting to examine the effect of these Notch1 mutations on VSMC phenotype. 122 In the adult, VSMC differentiation is involved in intimal thickening observed in atherosclerosis and arterial stenosis, although the origin of the VSMC precursors is still a subject of much debate (van Oostrom, Fledderus et al. 2009; Orlandi and Bennett 2010). Bone-marrow-derived intimal VSMC was observed after arterial injury in murine atherosclerosis model (Sata, Saiura et al. 2002), while a contribution from resident precursors was also demonstrated in neointima formation (Bentzon, Weile et al. 2006; Torsney, Mandal et al. 2007). Whether this resident VSMC precursor is Tie1-positive or shares any properties with the Tie1+ precursor described in this thesis is yet to be determined. Several studies have characterized the expression of components of the Notch signaling pathway following vascular injury. Hedgehog-induced Notch1 expression increases in intimal VSMC after vascular injury (Morrow, Cullen et al. 2009). Expressions of Notch1, Notch3, Jagged1 and Hey1 were also observed in bone marrow-derived cells in the neointima of injured vessels (Doi, Iso et al. 2009). Intimal hyperplasia after vascular injury was significantly reduced in Hey2 knock-out mice, suggesting a functional role of the Notch target gene in neointimal formation (Sakata, Xiang et al. 2004). However, the exact role of Notch activity during injury- induced intimal thickening requires further investigation. The dnMAML mouse model established in this thesis may be used to examine the potential role of the Notch signaling pathway in VSMC differentiation during neointimal formation. 5.1.4. Possible involvement of other cell types in arteriogenesis One caveat for the experiments using the Tie1tTA animal is the lack of specificity of the promoter. Expression of Tie1 is not restricted to the endothelium and the vascular precursor cells. Using a similar Tie1 promoter, Gustafsson and colleagues show that the Tie1 promoter is also active in around ten percent of hematopoietic cells and in several areas of the adult brain (Gustafsson, Brakebusch et al. 2001). Therefore, the phenotype and lethality observed 123 in the Tie1tTA:dnMAML embryos may have resulted from Notch inhibition in multiple cell types. Neverthelesss, the spatial and temporal specificities of the hemorrhagic phenotype in Tie1tTA:dnMAML embryos suggests that the effect is local and not caused by systemic defects such as abnormal platelet development. Using the VEtTA transgenic mouse, we show that the hemorrhagic phenotype is not caused by blocking Notch in the endothelium. In this thesis, we demonstrated a defect in VSMC development when Notch activation is blocked in Tie1+ cells. However, we could not prove that the effect is cell-autonomous. We also have to consider any possible paracrine effects that dysregulated hematopoietic cells (due to blockade of Notch) may have on the vascular phenotypes. Due to limitations of the in vivo system used, we cannot isolate the cellular source of the VSMC defect. A more detailed examination on the expression pattern of dnMAML in other cell types is necessary before we can attribute the phenotype to the smooth muscle differentiation defect. Alternatively, VSMC precursor cells expressing dnMAML may be isolated from the Tie1tTA:dnMAML embryos prior to arteriogenesis and the ability of the precursor to differentiate toward VSMC fate can be examined. 124 5.2. Notch and survival signalling In Chapter 4 of this thesis, the mechanism of Notch-induced endothelial survival signaling was examined. Surprisingly, we found that Notch activation can have conflicting effects on endothelial apoptosis that are dependent on the apoptotic stimulus. To understand this observation, we further probed the downstream mechanisms of Notch-induced cell survival. Notch activation in endothelial cells leads to the release of a secreted factor, which in turn activates PI3K. In Notch-activated endothelial cells, but not in control cells, PI3K activity is required for cell survival, revealing the presence of parallel apoptotic signals induced by Notch. PI3K activation imparts an endothelial cell survival effect in Notch activated cells by the induction of the anti-apoptotic protein Slug. Combined with previous data from our lab, we have shown that Notch regulates the expression of Slug through two mechanisms: direct transcriptional activation (Niessen, Fu et al. 2008) and additional transcriptional regulation through Notch-induced activation of PI3K. 5.2.1. Mechanisms for Notch-induced PI3K activation Notch can interact with other signaling pathways through PI3K. For example, Notch activates mTOR through PI3K signaling, and mTOR expression is able to down-regulate p53 protein expression (Mungamuri, Yang et al. 2006), thus providing a link between two major survival/apoptosis pathways. Therefore, by understanding the mechanism of Notch-induced PI3K activation, we can further our knowledge of the regulation of endothelial cell survival by Notch activation, which may be extended to other cell types such as cancer cells. The activation of PI3K signaling by Notch activation has been observed in various other normal or cancer cell lines (Ciofani and Zuniga-Pflucker 2005; Mungamuri, Yang et al. 2006; Gude, Emmanuel et al. 2008; Wang, Li et al. 2010) and likely involves Notch-induced transcriptional activation (Liu, Xiao et al. 2006). However, a non-canonical Notch signaling 125 (not through transcriptional activation with CSL/MAML complex) has also been suggested to regulate phosphorylation of Akt through the mTOR-rictor complex (Perumalsamy, Nagala et al. 2009). Recent studies have shown that Notch can activate Akt via the down-regulation of PTEN (Palomero, Sulis et al. 2007; Eliasz, Liang et al. 2010), although in the case of T-cell acute lymphoblastic leukemia (T-ALL), Notch activation imparts oncogenicity through other pathways in addition to PI3K pathway regulation (Medyouf, Gao et al. 2010). Induction of growth factor receptors, which signal through the PI3K pathway, by Notch has also been implicated in the oncogenic activity of Notch (Eliasz, Liang et al. 2010). In this study, we showed that Notch induces PI3K through a secreted factor. The same mechanism of PI3K activation has been observed in mammary epithelial cells (Meurette, Stylianou et al. 2009). The medium conditioned by mammary epithelial cells expressing NotchIC induced Akt phosphorylation in parental cells, while the conditioned medium from the vector control cells did not activate PI3K (Meurette, Stylianou et al. 2009). The identity of the Notch-induced secreted factor will be investigated. Our lab has performed a microarray analysis on Notch-activated endothelial cells; therefore a candidate gene approach to identify the secreted factor is possible. However, an upregulation of the mRNA may not always translate into an upregulation of protein, thus a global protein analysis on differentiatial level of secreted proteins between NotchIC-transduced or vector control-transduced endothelial cell will be preferable towards identifying the factor. 5.2.3. Homocysteine and Slug expression Regulation of Slug expression by homocysteine treatment may play a role in embryonic heart development. Maternal hyperhomocysteinemia has been shown to be a risk factor for congenital heart defects in human (Verkleij-Hagoort, Bliek et al. 2007). In animal models, hyperhomocysteinemia can cause cardiac defects in the neural crest-derived outflow tract, 126 which develops into the pulmonary artery and the aortic arch (Boot, Steegers-Theunissen et al. 2003). Interestingly, Slug expression is found to be important for proper formation of neural crest-derived tissues in frog development (LaBonne and Bronner-Fraser 2000). However, while Slug expression is present during cardiac outflow tract development, Slug is not essential for mammalian cardiac development because of the functional redundancy of a close family member Snail (Niessen, Fu et al. 2008). In the present study, we have shown that homocysteine treatment down-regulates the expression of Slug. It will be interesting to see whether the effect of homocysteine on cardiac development can be explained, at least partially, by the loss of Slug and/or Snail expression. 5.2.4. Alternative methods of Notch activation One caveat of the study is the strength of Notch activation by the NotchIC construct. NotchIC is considered a strong constitutive activator of the Notch pathway. Expression of NotchIC in endothelial cells not only affects the survival signaling pathway, but also inhibits endothelial proliferation (Noseda, Chang et al. 2004) and induces mesenchymal transdifferentiation (Noseda, McLean et al. 2004). It may become difficult to distinguish between the primary effects of Notch activation and the secondary effects brought on by the Notch-induced changes in the endothelial cells. Two possible alternative methods of Notch activation may offer a weaker, therefore, more physiologically relevant level of Notch activity. Inducible systems of Notch activation are available which will enable a regulated level and timing of the Notch signals. NotchIC has been fused to the ligand binding domain of the estrogen receptor (NotchIC-ER) providing a tamoxifen-induced nuclear translocation of NotchIC and transcription activation of Notch targets (Jeffries and Capobianco 2000). Also, a tetracycline inducible system, similar to the one used in our transgenic mouse, can also be used to induce expression of NotchIC in a more controlled way. Notch signaling can also be activated by co-culturing with ligand-expressing cells. In vivo, Jagged1 and Dll4 are the two 127 main ligands activating the Notch signaling pathway in the vasculature. Our lab has shown that co-culture with Notch ligand-expressing cells can recapitulate the effects of NotchIC on endothelial cells, albeit to a lesser degree (Noseda, Chang et al. 2004; Noseda, McLean et al. 2004). To confirm the observations obtained by NotchIC expression in the endothelial cells, experiments can be conducted with using endothelial cell lines co-cultured with Jagged1- or Dll4-expressing cells. In conclusion, the studies presented in this thesis have identified a role for Notch signaling in VSMC development and endothelial survival signaling. Importantly, we have identified a common vascular smooth muscle precursor cell and showed that Notch activation is required for its differentiation. Our finding also provides a possible mechanism for the stimulus-dependent activity of Notch-mediated cell survival. Together with our other studies (see Appendix C), our data shows that Notch activation plays a role in maintaining the quiescent adult endothelial monolayer, by suppressing angiogenic sprouting, inhibiting endothelial proliferation and promoting endothelial survival. However, during development, the Notch signaling pathway appears to be important for cell type specification, partially by promoting mesenchymal differentiation. 128 BIBLIOGRAPHY Adams, J. M. (2003). "Ways of dying: multiple pathways to apoptosis." Genes Dev 17(20): 2481‐95. Adams, R. H., G. A. Wilkinson, et al. (1999). "Roles of ephrinB ligands and EphB receptors in cardiovascular development: demarcation of arterial/venous domains, vascular morphogenesis, and sprouting angiogenesis." Genes Dev 13(3): 295‐306. Ades, E. W., F. J. Candal, et al. (1992). "HMEC‐1: establishment of an immortalized human microvascular endothelial cell line." J Invest Dermatol 99(6): 683‐90. Alva, J. A. and M. L. Iruela‐Arispe (2004). "Notch signaling in vascular morphogenesis." Curr Opin Hematol 11(4): 278‐83. Alvarez, R. J., S. J. Gips, et al. (1997). "17beta‐estradiol inhibits apoptosis of endothelial cells." Biochem Biophys Res Commun 237(2): 372‐81. Arboleda‐Velasquez, J. F., Z. Zhou, et al. (2008). "Linking Notch signaling to ischemic stroke." Proc Natl Acad Sci U S A 105(12): 4856‐61. Arciniegas, E., M. G. Frid, et al. (2007). "Perspectives on endothelial‐to‐mesenchymal transition: potential contribution to vascular remodeling in chronic pulmonary hypertension." Am J Physiol Lung Cell Mol Physiol 293(1): L1‐8. Arciniegas, E., C. Y. Neves, et al. (2005). "Endothelial‐mesenchymal transition occurs during embryonic pulmonary artery development." Endothelium 12(4): 193‐200. Artavanis‐Tsakonas, S., M. D. Rand, et al. (1999). "Notch signaling: cell fate control and signal integration in development." Science 284(5415): 770‐6. Aste‐Amezaga, M., N. Zhang, et al. (2010). "Characterization of Notch1 antibodies that inhibit signaling of both normal and mutated Notch1 receptors." PLoS One 5(2): e9094. Bannerman, D. D. and S. E. Goldblum (2003). "Mechanisms of bacterial lipopolysaccharide‐induced endothelial apoptosis." Am J Physiol Lung Cell Mol Physiol 284(6): L899‐914. Bannerman, D. D., J. C. Tupper, et al. (2001). "A constitutive cytoprotective pathway protects endothelial cells from lipopolysaccharide‐induced apoptosis." J Biol Chem 276(18): 14924‐32. Baron, M., H. Aslam, et al. (2002). "Multiple levels of Notch signal regulation (review)." Mol Membr Biol 19(1): 27‐38. Ben‐Yair, R. and C. Kalcheim (2008). "Notch and bone morphogenetic protein differentially act on dermomyotome cells to generate endothelium, smooth, and striated muscle." J Cell Biol 180(3): 607‐18. Benedito, R., C. Roca, et al. (2009). "The notch ligands Dll4 and Jagged1 have opposing effects on angiogenesis." Cell 137(6): 1124‐35. Bentzon, J. F., C. Weile, et al. (2006). "Smooth muscle cells in atherosclerosis originate from the local vessel wall and not circulating progenitor cells in ApoE knockout mice." Arterioscler Thromb Vasc Biol 26(12): 2696‐702. Blaumueller, C. M., H. Qi, et al. (1997). "Intracellular cleavage of Notch leads to a heterodimeric receptor on the plasma membrane." Cell 90(2): 281‐91. Bobik, A. (2006). "Transforming growth factor‐betas and vascular disorders." Arterioscler Thromb Vasc Biol 26(8): 1712‐20. Bombeli, T., B. R. Schwartz, et al. (1999). "Endothelial cells undergoing apoptosis become proadhesive for nonactivated platelets." Blood 93(11): 3831‐8. Boot, M. J., R. P. Steegers‐Theunissen, et al. (2003). "Folic acid and homocysteine affect neural crest and neuroepithelial cell outgrowth and differentiation in vitro." Dev Dyn 227(2): 301‐8. Bray, S. (1998). "Notch signalling in Drosophila: three ways to use a pathway." Semin Cell Dev Biol 9(6): 591‐7. 129 Brooks, P. C., A. M. Montgomery, et al. (1994). "Integrin alpha v beta 3 antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels." Cell 79(7): 1157‐64. Brou, C., F. Logeat, et al. (2000). "A novel proteolytic cleavage involved in Notch signaling: the role of the disintegrin‐metalloprotease TACE." Mol Cell 5(2): 207‐16. Bruckner, K., L. Perez, et al. (2000). "Glycosyltransferase activity of Fringe modulates Notch‐Delta interactions." Nature 406(6794): 411‐5. Bruhl, T., C. Heeschen, et al. (2004). "p21Cip1 levels differentially regulate turnover of mature endothelial cells, endothelial progenitor cells, and in vivo neovascularization." Circ Res 94(5): 686‐92. Brunn, G. J., J. Williams, et al. (1996). "Direct inhibition of the signaling functions of the mammalian target of rapamycin by the phosphoinositide 3‐kinase inhibitors, wortmannin and LY294002." EMBO J 15(19): 5256‐67. Buas, M. F., S. Kabak, et al. (2010). "The Notch effector Hey1 associates with myogenic target genes to repress myogenesis." J Biol Chem 285(2): 1249‐58. Calzavara, E., R. Chiaramonte, et al. (2008). "Reciprocal regulation of Notch and PI3K/Akt signalling in T‐ALL cells in vitro." J Cell Biochem 103(5): 1405‐12. Carmeliet, P. (2000). "Mechanisms of angiogenesis and arteriogenesis." Nat Med 6(4): 389‐95. Carmeliet, P. (2003). "Angiogenesis in health and disease." Nat Med 9(6): 653‐60. Carmeliet, P., M. G. Lampugnani, et al. (1999). "Targeted deficiency or cytosolic truncation of the VE‐ cadherin gene in mice impairs VEGF‐mediated endothelial survival and angiogenesis." Cell 98(2): 147‐57. Chen, V. C., R. Stull, et al. (2008). "Notch signaling respecifies the hemangioblast to a cardiac fate." Nat Biotechnol 26(10): 1169‐78. Cines, D. B., E. S. Pollak, et al. (1998). "Endothelial cells in physiology and in the pathophysiology of vascular disorders." Blood 91(10): 3527‐61. Ciofani, M. and J. C. Zuniga‐Pflucker (2005). "Notch promotes survival of pre‐T cells at the beta‐ selection checkpoint by regulating cellular metabolism." Nat Immunol 6(9): 881‐8. Ciruna, B. and J. Rossant (2001). "FGF signaling regulates mesoderm cell fate specification and morphogenetic movement at the primitive streak." Dev Cell 1(1): 37‐49. Clarke, R., L. Daly, et al. (1991). "Hyperhomocysteinemia: an independent risk factor for vascular disease." N Engl J Med 324(17): 1149‐55. Coffin, J. D., J. Harrison, et al. (1991). "Angioblast differentiation and morphogenesis of the vascular endothelium in the mouse embryo." Dev Biol 148(1): 51‐62. Coffin, J. D. and T. J. Poole (1988). "Embryonic vascular development: immunohistochemical identification of the origin and subsequent morphogenesis of the major vessel primordia in quail embryos." Development 102(4): 735‐48. Colville‐Nash, P. R. and D. L. Scott (1992). "Angiogenesis and rheumatoid arthritis: pathogenic and therapeutic implications." Ann Rheum Dis 51(7): 919‐25. Cordes, R., K. Schuster‐Gossler, et al. (2004). "Specification of vertebral identity is coupled to Notch signalling and the segmentation clock." Development 131(6): 1221‐33. Coultas, L., K. Chawengsaksophak, et al. (2005). "Endothelial cells and VEGF in vascular development." Nature 438(7070): 937‐45. Cox, C. M. and T. J. Poole (2000). "Angioblast differentiation is influenced by the local environment: FGF‐2 induces angioblasts and patterns vessel formation in the quail embryo." Dev Dyn 218(2): 371‐82. Darland, D. C. and P. A. D'Amore (2001). "TGF beta is required for the formation of capillary‐like structures in three‐dimensional cocultures of 10T1/2 and endothelial cells." Angiogenesis 4(1): 11‐20. De Palma, M., C. Murdoch, et al. (2007). "Tie2‐expressing monocytes: regulation of tumor angiogenesis and therapeutic implications." Trends Immunol 28(12): 519‐24. 130 De Strooper, B., W. Annaert, et al. (1999). "A presenilin‐1‐dependent gamma‐secretase‐like protease mediates release of Notch intracellular domain." Nature 398(6727): 518‐22. Deissler, H., G. K. Lang, et al. (2006). "TGFbeta induces transdifferentiation of iBREC to alphaSMA‐ expressing cells." Int J Mol Med 18(4): 577‐82. Del Monte, G., J. Grego‐Bessa, et al. (2007). "Monitoring Notch1 activity in development: evidence for a feedback regulatory loop." Dev Dyn 236(9): 2594‐614. DeRuiter, M. C., R. E. Poelmann, et al. (1997). "Embryonic endothelial cells transdifferentiate into mesenchymal cells expressing smooth muscle actins in vivo and in vitro." Circ Res 80(4): 444‐ 51. Dettman, R. W., W. Denetclaw, Jr., et al. (1998). "Common epicardial origin of coronary vascular smooth muscle, perivascular fibroblasts, and intermyocardial fibroblasts in the avian heart." Dev Biol 193(2): 169‐81. Doi, H., T. Iso, et al. (2009). "Notch signaling regulates the differentiation of bone marrow‐derived cells into smooth muscle‐like cells during arterial lesion formation." Biochem Biophys Res Commun 381(4): 654‐9. Doi, H., T. Iso, et al. (2005). "HERP1 inhibits myocardin‐induced vascular smooth muscle cell differentiation by interfering with SRF binding to CArG box." Arteriosclerosis Thrombosis and Vascular Biology 25(11): 2328‐2334. Domenga, V., P. Fardoux, et al. (2004). "Notch3 is required for arterial identity and maturation of vascular smooth muscle cells." Genes Dev 18(22): 2730‐5. Dontu, G., K. W. Jackson, et al. (2004). "Role of Notch signaling in cell‐fate determination of human mammary stem/progenitor cells." Breast Cancer Res 6(6): R605‐15. Dou, G. R., Y. C. Wang, et al. (2008). "RBP‐J, the transcription factor downstream of Notch receptors, is essential for the maintenance of vascular homeostasis in adult mice." FASEB J 22(5): 1606‐ 17. Dou, S., X. Zeng, et al. (1994). "The recombination signal sequence‐binding protein RBP‐2N functions as a transcriptional repressor." Mol Cell Biol 14(5): 3310‐9. Drake, C. J., J. E. Hungerford, et al. (1998). "Morphogenesis of the first blood vessels." Ann N Y Acad Sci 857: 155‐79. Duarte, A., M. Hirashima, et al. (2004). "Dosage‐sensitive requirement for mouse Dll4 in artery development." Genes Dev 18(20): 2474‐8. Duncan, A. W., F. M. Rattis, et al. (2005). "Integration of Notch and Wnt signaling in hematopoietic stem cell maintenance." Nat Immunol 6(3): 314‐22. Eliasz, S., S. Liang, et al. (2010). "Notch‐1 stimulates survival of lung adenocarcinoma cells during hypoxia by activating the IGF‐1R pathway." Oncogene. Ema, M., P. Faloon, et al. (2003). "Combinatorial effects of Flk1 and Tal1 on vascular and hematopoietic development in the mouse." Genes Dev 17(3): 380‐93. Emerick, K. M., E. B. Rand, et al. (1999). "Features of Alagille syndrome in 92 patients: frequency and relation to prognosis." Hepatology 29(3): 822‐9. Emuss, V., D. Lagos, et al. (2009). "KSHV manipulates Notch signaling by DLL4 and JAG1 to alter cell cycle genes in lymphatic endothelia." PLoS Pathog 5(10): e1000616. Esner, M., S. M. Meilhac, et al. (2006). "Smooth muscle of the dorsal aorta shares a common clonal origin with skeletal muscle of the myotome." Development 133(4): 737‐49. Espinosa, L., J. Ingles‐Esteve, et al. (2003). "Phosphorylation by glycogen synthase kinase‐3 beta down‐regulates Notch activity, a link for Notch and Wnt pathways." J Biol Chem 278(34): 32227‐35. Ferguson, J. E., 3rd, R. W. Kelley, et al. (2005). "Mechanisms of endothelial differentiation in embryonic vasculogenesis." Arterioscler Thromb Vasc Biol 25(11): 2246‐54. 131 Ferreira, L. S., S. Gerecht, et al. (2007). "Vascular progenitor cells isolated from human embryonic stem cells give rise to endothelial and smooth muscle like cells and form vascular networks in vivo." Circ Res 101(3): 286‐94. Fischer, A., N. Schumacher, et al. (2004). "The Notch target genes Hey1 and Hey2 are required for embryonic vascular development." Genes Dev 18(8): 901‐11. Flamme, I., T. Frolich, et al. (1997). "Molecular mechanisms of vasculogenesis and embryonic angiogenesis." J Cell Physiol 173(2): 206‐10. Folkman, J. (1984). "What is the role of endothelial cells in angiogenesis?" Lab Invest 51(6): 601‐4. Foubert, P., G. Matrone, et al. (2008). "Coadministration of endothelial and smooth muscle progenitor cells enhances the efficiency of proangiogenic cell‐based therapy." Circ Res 103(7): 751‐60. Fryer, C. J., E. Lamar, et al. (2002). "Mastermind mediates chromatin‐specific transcription and turnover of the Notch enhancer complex." Genes Dev 16(11): 1397‐411. Fryer, C. J., J. B. White, et al. (2004). "Mastermind recruits CycC:CDK8 to phosphorylate the Notch ICD and coordinate activation with turnover." Mol Cell 16(4): 509‐20. Fu, Y., A. Chang, et al. (2009). "Differential regulation of TGFbeta signaling pathways by notch in human endothelial cells." J Biol Chem. Fuchs, S. Y., V. Adler, et al. (1998). "Mdm2 association with p53 targets its ubiquitination." Oncogene 17(19): 2543‐7. Gale, N. W., M. G. Dominguez, et al. (2004). "Haploinsufficiency of delta‐like 4 ligand results in embryonic lethality due to major defects in arterial and vascular development." Proc Natl Acad Sci U S A 101(45): 15949‐54. Gazave, E., P. Lapebie, et al. (2009). "Origin and evolution of the Notch signalling pathway: an overview from eukaryotic genomes." BMC Evol Biol 9: 249. Geling, A., H. Steiner, et al. (2002). "A gamma‐secretase inhibitor blocks Notch signaling in vivo and causes a severe neurogenic phenotype in zebrafish." EMBO Rep 3(7): 688‐94. Gerhardt, H., M. Golding, et al. (2003). "VEGF guides angiogenic sprouting utilizing endothelial tip cell filopodia." J Cell Biol 161(6): 1163‐77. Gerhart, J. (1999). "1998 Warkany lecture: signaling pathways in development." Teratology 60(4): 226‐39. Gittenberger‐de Groot, A. C., M. C. DeRuiter, et al. (1999). "Smooth muscle cell origin and its relation to heterogeneity in development and disease." Arterioscler Thromb Vasc Biol 19(7): 1589‐94. Gittenberger‐de Groot, A. C., M. P. Vrancken Peeters, et al. (1998). "Epicardium‐derived cells contribute a novel population to the myocardial wall and the atrioventricular cushions." Circ Res 82(10): 1043‐52. Goutte, C., M. Tsunozaki, et al. (2002). "APH‐1 is a multipass membrane protein essential for the Notch signaling pathway in Caenorhabditis elegans embryos." Proc Natl Acad Sci U S A 99(2): 775‐9. Gridley, T. (2007). "Notch signaling in vascular development and physiology." Development 134(15): 2709‐18. Gude, N. A., G. Emmanuel, et al. (2008). "Activation of Notch‐mediated protective signaling in the myocardium." Circ Res 102(9): 1025‐35. Gupta‐Rossi, N., E. Six, et al. (2004). "Monoubiquitination and endocytosis direct gamma‐secretase cleavage of activated Notch receptor." J Cell Biol 166(1): 73‐83. Gustafsson, E., C. Brakebusch, et al. (2001). "Tie‐1‐directed expression of Cre recombinase in endothelial cells of embryoid bodies and transgenic mice." J Cell Sci 114(Pt 4): 671‐6. Haar, J. L. and G. A. Ackerman (1971). "A phase and electron microscopic study of vasculogenesis and erythropoiesis in the yolk sac of the mouse." Anat Rec 170(2): 199‐223. Haimovitz‐Friedman, A., C. Cordon‐Cardo, et al. (1997). "Lipopolysaccharide induces disseminated endothelial apoptosis requiring ceramide generation." J Exp Med 186(11): 1831‐41. 132 Hanahan, D. and J. Folkman (1996). "Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis." Cell 86(3): 353‐64. Harrington, L. S., R. C. Sainson, et al. (2008). "Regulation of multiple angiogenic pathways by Dll4 and Notch in human umbilical vein endothelial cells." Microvasc Res 75(2): 144‐54. Hellstrom, M., L. K. Phng, et al. (2007). "Dll4 signalling through Notch1 regulates formation of tip cells during angiogenesis." Nature 445(7129): 776‐80. Herzog, Y., C. Kalcheim, et al. (2001). "Differential expression of neuropilin‐1 and neuropilin‐2 in arteries and veins." Mech Dev 109(1): 115‐9. High, F. A., M. M. Lu, et al. (2008). "Endothelial expression of the Notch ligand Jagged1 is required for vascular smooth muscle development." Proc Natl Acad Sci U S A 105(6): 1955‐9. High, F. A., M. Zhang, et al. (2007). "An essential role for Notch in neural crest during cardiovascular development and smooth muscle differentiation." J Clin Invest 117(2): 353‐63. Hill, K. L., P. Obrtlikova, et al. (2010). "Human embryonic stem cell‐derived vascular progenitor cells capable of endothelial and smooth muscle cell function." Exp Hematol 38(3): 246‐257 e1. Hobson, B. and J. Denekamp (1984). "Endothelial proliferation in tumours and normal tissues: continuous labelling studies." Br J Cancer 49(4): 405‐13. Hoey, T., W. C. Yen, et al. (2009). "DLL4 blockade inhibits tumor growth and reduces tumor‐initiating cell frequency." Cell Stem Cell 5(2): 168‐77. Hofmann, J. J. and M. Luisa Iruela‐Arispe (2007). "Notch expression patterns in the retina: An eye on receptor‐ligand distribution during angiogenesis." Gene Expr Patterns 7(4): 461‐70. Holleran, J. L., M. J. Egorin, et al. (2003). "Use of high‐performance liquid chromatography to characterize the rapid decomposition of wortmannin in tissue culture media." Anal Biochem 323(1): 19‐25. Hossain, G. S., J. V. van Thienen, et al. (2003). "TDAG51 is induced by homocysteine, promotes detachment‐mediated programmed cell death, and contributes to the cevelopment of atherosclerosis in hyperhomocysteinemia." J Biol Chem 278(32): 30317‐27. Hsieh, J. J., S. Zhou, et al. (1999). "CIR, a corepressor linking the DNA binding factor CBF1 to the histone deacetylase complex." Proc Natl Acad Sci U S A 96(1): 23‐8. Hsu, H. C., H. Ema, et al. (2000). "Hematopoietic stem cells express Tie‐2 receptor in the murine fetal liver." Blood 96(12): 3757‐62. Hu, Y., Y. Ye, et al. (2002). "Nicastrin is required for gamma‐secretase cleavage of the Drosophila Notch receptor." Dev Cell 2(1): 69‐78. Huang, K. T., L. Kuo, et al. (1998). "Lipopolysaccharide activates endothelial nitric oxide synthase through protein tyrosine kinase." Biochem Biophys Res Commun 245(1): 33‐7. Hungerford, J. E. and C. D. Little (1999). "Developmental biology of the vascular smooth muscle cell: building a multilayered vessel wall." J Vasc Res 36(1): 2‐27. Hungerford, J. E., G. K. Owens, et al. (1996). "Development of the aortic vessel wall as defined by vascular smooth muscle and extracellular matrix markers." Dev Biol 178(2): 375‐92. Hurlbut, G. D., M. W. Kankel, et al. (2007). "Crossing paths with Notch in the hyper‐network." Curr Opin Cell Biol 19(2): 166‐75. Inoue, A., M. G. Seidel, et al. (2002). "Slug, a highly conserved zinc finger transcriptional repressor, protects hematopoietic progenitor cells from radiation‐induced apoptosis in vivo." Cancer Cell 2(4): 279‐88. Iso, T., Y. Hamamori, et al. (2003). "Notch signaling in vascular development." Arterioscler Thromb Vasc Biol 23(4): 543‐53. Iso, T., V. Sartorelli, et al. (2001). "HERP, a new primary target of Notch regulated by ligand binding." Mol Cell Biol 21(17): 6071‐9. Itoh, M., C. H. Kim, et al. (2003). "Mind bomb is a ubiquitin ligase that is essential for efficient activation of Notch signaling by Delta." Dev Cell 4(1): 67‐82. 133 Jeffries, S. and A. J. Capobianco (2000). "Neoplastic transformation by Notch requires nuclear localization." Mol Cell Biol 20(11): 3928‐41. Jeffries, S., D. J. Robbins, et al. (2002). "Characterization of a high‐molecular‐weight Notch complex in the nucleus of Notch(ic)‐transformed RKE cells and in a human T‐cell leukemia cell line." Mol Cell Biol 22(11): 3927‐41. Jiang, X., D. H. Rowitch, et al. (2000). "Fate of the mammalian cardiac neural crest." Development 127(8): 1607‐16. Jin, S., E. M. Hansson, et al. (2008). "Notch signaling regulates platelet‐derived growth factor receptor‐beta expression in vascular smooth muscle cells." Circ Res 102(12): 1483‐91. Joutel, A., F. Andreux, et al. (2000). "The ectodomain of the Notch3 receptor accumulates within the cerebrovasculature of CADASIL patients." J Clin Invest 105(5): 597‐605. Joutel, A., C. Corpechot, et al. (1996). "Notch3 mutations in CADASIL, a hereditary adult‐onset condition causing stroke and dementia." Nature 383(6602): 707‐10. Joutel, A., C. Corpechot, et al. (1997). "Notch3 mutations in cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), a mendelian condition causing stroke and vascular dementia." Ann N Y Acad Sci 826: 213‐7. Kao, H. Y., P. Ordentlich, et al. (1998). "A histone deacetylase corepressor complex regulates the Notch signal transduction pathway." Genes Dev 12(15): 2269‐77. Karsan, A. (2005). "The role of notch in modeling and maintaining the vasculature." Can J Physiol Pharmacol 83(1): 14‐23. Karsan, A., E. Yee, et al. (1997). "Fibroblast growth factor‐2 inhibits endothelial cell apoptosis by Bcl‐2‐ dependent and independent mechanisms." Am J Pathol 151(6): 1775‐84. Kattman, S. J., T. L. Huber, et al. (2006). "Multipotent flk‐1+ cardiovascular progenitor cells give rise to the cardiomyocyte, endothelial, and vascular smooth muscle lineages." Dev Cell 11(5): 723‐ 32. Kaufman, MH (1992). The Atlas of Mouse Development. Oxford, UK: Academic Press. p145,157,177. Kawasaki, M., K. Kuwano, et al. (2000). "Protection from lethal apoptosis in lipopolysaccharide‐ induced acute lung injury in mice by a caspase inhibitor." Am J Pathol 157(2): 597‐603. Kerr, J. F., A. H. Wyllie, et al. (1972). "Apoptosis: a basic biological phenomenon with wide‐ranging implications in tissue kinetics." Br J Cancer 26(4): 239‐57. Kimberly, W. T., M. J. LaVoie, et al. (2003). "Gamma‐secretase is a membrane protein complex comprised of presenilin, nicastrin, Aph‐1, and Pen‐2." Proc Natl Acad Sci U S A 100(11): 6382‐ 7. Kirby, M. L., T. F. Gale, et al. (1983). "Neural crest cells contribute to normal aorticopulmonary septation." Science 220(4601): 1059‐61. Kitao, A., Y. Sato, et al. (2009). "Endothelial to mesenchymal transition via transforming growth factor‐beta1/Smad activation is associated with portal venous stenosis in idiopathic portal hypertension." Am J Pathol 175(2): 616‐26. Koblizek, T. I., C. Weiss, et al. (1998). "Angiopoietin‐1 induces sprouting angiogenesis in vitro." Curr Biol 8(9): 529‐32. Koo, B. K., H. S. Lim, et al. (2005). "Mind bomb 1 is essential for generating functional Notch ligands to activate Notch." Development 132(15): 3459‐70. Koopman, G., C. P. Reutelingsperger, et al. (1994). "Annexin V for flow cytometric detection of phosphatidylserine expression on B cells undergoing apoptosis." Blood 84(5): 1415‐20. Kopan, R., J. S. Nye, et al. (1994). "The intracellular domain of mouse Notch: a constitutively activated repressor of myogenesis directed at the basic helix‐loop‐helix region of MyoD." Development 120(9): 2385‐96. Krebs, L. T., J. R. Shutter, et al. (2004). "Haploinsufficient lethality and formation of arteriovenous malformations in Notch pathway mutants." Genes Dev 18(20): 2469‐73. 134 Krebs, L. T., Y. Xue, et al. (2000). "Notch signaling is essential for vascular morphogenesis in mice." Genes Dev 14(11): 1343‐52. Kuroda, K., S. Tani, et al. (1999). "Delta‐induced Notch signaling mediated by RBP‐J inhibits MyoD expression and myogenesis." J Biol Chem 274(11): 7238‐44. Kurooka, H. and T. Honjo (2000). "Functional interaction between the mouse notch1 intracellular region and histone acetyltransferases PCAF and GCN5." J Biol Chem 275(22): 17211‐20. Kurpinski, K., H. Lam, et al. (2010). "TGF‐beta and Notch Signaling Mediate Stem Cell Differentiation into Smooth Muscle Cells." Stem Cells. Kurrey, N. K., S. P. Jalgaonkar, et al. (2009). "Snail and slug mediate radioresistance and chemoresistance by antagonizing p53‐mediated apoptosis and acquiring a stem‐like phenotype in ovarian cancer cells." Stem Cells 27(9): 2059‐68. LaBonne, C. and M. Bronner‐Fraser (2000). "Snail‐related transcriptional repressors are required in Xenopus for both the induction of the neural crest and its subsequent migration." Dev Biol 221(1): 195‐205. Lampugnani, M. G., M. Corada, et al. (1995). "The molecular organization of endothelial cell to cell junctions: differential association of plakoglobin, beta‐catenin, and alpha‐catenin with vascular endothelial cadherin (VE‐cadherin)." J Cell Biol 129(1): 203‐17. Lawson, N. D., N. Scheer, et al. (2001). "Notch signaling is required for arterial‐venous differentiation during embryonic vascular development." Development 128(19): 3675‐83. Le Lievre, C. S. and N. M. Le Douarin (1975). "Mesenchymal derivatives of the neural crest: analysis of chimaeric quail and chick embryos." J Embryol Exp Morphol 34(1): 125‐54. Lee, C. Y., K. M. Vogeli, et al. (2009). "Notch signaling functions as a cell‐fate switch between the endothelial and hematopoietic lineages." Curr Biol 19(19): 1616‐22. Lee, S., T. T. Chen, et al. (2007). "Autocrine VEGF signaling is required for vascular homeostasis." Cell 130(4): 691‐703. Lee, S. J., K. M. Kim, et al. (2005). "Nitric oxide inhibition of homocysteine‐induced human endothelial cell apoptosis by down‐regulation of p53‐dependent Noxa expression through the formation of S‐nitrosohomocysteine." J Biol Chem 280(7): 5781‐8. Leong, K. G., X. Hu, et al. (2002). "Activated Notch4 inhibits angiogenesis: role of beta 1‐integrin activation." Mol Cell Biol 22(8): 2830‐41. Leslie, J. D., L. Ariza‐McNaughton, et al. (2007). "Endothelial signalling by the Notch ligand Delta‐like 4 restricts angiogenesis." Development 134(5): 839‐44. Li, D. Y., L. K. Sorensen, et al. (1999). "Defective angiogenesis in mice lacking endoglin." Science 284(5419): 1534‐7. Li, J., F. Chen, et al. (2000). "Neural crest expression of Cre recombinase directed by the proximal Pax3 promoter in transgenic mice." Genesis 26(2): 162‐4. Limbourg, F. P., K. Takeshita, et al. (2005). "Essential role of endothelial Notch1 in angiogenesis." Circulation 111(14): 1826‐32. Lin, Q., J. Lu, et al. (1998). "Requirement of the MADS‐box transcription factor MEF2C for vascular development." Development 125(22): 4565‐74. Liu, H., S. Kennard, et al. (2009). "NOTCH3 expression is induced in mural cells through an autoregulatory loop that requires endothelial‐expressed JAGGED1." Circ Res 104(4): 466‐75. Liu, W., S. A. Ahmad, et al. (2000). "Endothelial cell survival and apoptosis in the tumor vasculature." Apoptosis 5(4): 323‐8. Liu, Z. J., M. Xiao, et al. (2006). "Notch1 signaling promotes primary melanoma progression by activating mitogen‐activated protein kinase/phosphatidylinositol 3‐kinase‐Akt pathways and up‐regulating N‐cadherin expression." Cancer Res 66(8): 4182‐90. Liu, Z. J., M. Xiao, et al. (2006). "Inhibition of endothelial cell proliferation by Notch1 signaling is mediated by repressing MAPK and PI3K/Akt pathways and requires MAML1." FASEB J 20(7): 1009‐11. 135 Lobov, I. B., R. A. Renard, et al. (2007). "Delta‐like ligand 4 (Dll4) is induced by VEGF as a negative regulator of angiogenic sprouting." Proc Natl Acad Sci U S A 104(9): 3219‐24. Logeat, F., C. Bessia, et al. (1998). "The Notch1 receptor is cleaved constitutively by a furin‐like convertase." Proc Natl Acad Sci U S A 95(14): 8108‐12. Lohela, M., H. Helotera, et al. (2008). "Transgenic induction of vascular endothelial growth factor‐C is strongly angiogenic in mouse embryos but leads to persistent lymphatic hyperplasia in adult tissues." Am J Pathol 173(6): 1891‐901. Long, X., E. E. Creemers, et al. (2007). "Myocardin is a bifunctional switch for smooth versus skeletal muscle differentiation." Proc Natl Acad Sci U S A 104(42): 16570‐5. Lopez‐Schier, H. and D. St Johnston (2002). "Drosophila nicastrin is essential for the intramembranous cleavage of notch." Dev Cell 2(1): 79‐89. Lu, Q., B. Patel, et al. (2009). "Transforming growth factor‐beta1 causes pulmonary microvascular endothelial cell apoptosis via ALK5." Am J Physiol Lung Cell Mol Physiol 296(5): L825‐38. MacKenzie, F., P. Duriez, et al. (2004). "Notch4‐induced inhibition of endothelial sprouting requires the ankyrin repeats and involves signaling through RBP‐Jkappa." Blood 104(6): 1760‐8. MacKenzie, F., P. Duriez, et al. (2004). "Notch4 inhibits endothelial apoptosis via RBP‐Jkappa‐ dependent and ‐independent pathways." J Biol Chem 279(12): 11657‐63. Maillard, I., L. Tu, et al. (2006). "The requirement for Notch signaling at the beta‐selection checkpoint in vivo is absolute and independent of the pre‐T cell receptor." J Exp Med 203(10): 2239‐45. Maillard, I., A. P. Weng, et al. (2004). "Mastermind critically regulates Notch‐mediated lymphoid cell fate decisions." Blood 104(6): 1696‐702. Majesky, M. W. (2007). "Developmental basis of vascular smooth muscle diversity." Arterioscler Thromb Vasc Biol 27(6): 1248‐58. Marchetti, S., C. Gimond, et al. (2002). "Endothelial cells genetically selected from differentiating mouse embryonic stem cells incorporate at sites of neovascularization in vivo." J Cell Sci 115(Pt 10): 2075‐85. Mattiussi, S., P. Turrini, et al. (2004). "p21(Waf1/Cip1/Sdi1) mediates shear stress‐dependent antiapoptotic function." Cardiovasc Res 61(4): 693‐704. McCully, K. S. (1996). "Homocysteine and vascular disease." Nat Med 2(4): 386‐9. McElhinny, A. S., J. L. Li, et al. (2008). "Mastermind‐like transcriptional co‐activators: emerging roles in regulating cross talk among multiple signaling pathways." Oncogene 27(38): 5138‐47. McGill, M. A., S. E. Dho, et al. (2009). "Numb regulates post‐endocytic trafficking and degradation of Notch1." J Biol Chem 284(39): 26427‐38. McKellar, S. H., D. J. Tester, et al. (2007). "Novel NOTCH1 mutations in patients with bicuspid aortic valve disease and thoracic aortic aneurysms." J Thorac Cardiovasc Surg 134(2): 290‐6. Medyouf, H., X. Gao, et al. (2010). "Acute T‐cell leukemias remain dependent on Notch signaling despite PTEN and INK4A/ARF loss." Blood 115(6): 1175‐84. Meurette, O., S. Stylianou, et al. (2009). "Notch activation induces Akt signaling via an autocrine loop to prevent apoptosis in breast epithelial cells." Cancer Res 69(12): 5015‐22. Mikawa, T. and R. G. Gourdie (1996). "Pericardial mesoderm generates a population of coronary smooth muscle cells migrating into the heart along with ingrowth of the epicardial organ." Dev Biol 174(2): 221‐32. Milano, J., J. McKay, et al. (2004). "Modulation of notch processing by gamma‐secretase inhibitors causes intestinal goblet cell metaplasia and induction of genes known to specify gut secretory lineage differentiation." Toxicol Sci 82(1): 341‐58. Minasi, M. G., M. Riminucci, et al. (2002). "The meso‐angioblast: a multipotent, self‐renewing cell that originates from the dorsal aorta and differentiates into most mesodermal tissues." Development 129(11): 2773‐83. Morel, V., R. Le Borgne, et al. (2003). "Snail is required for Delta endocytosis and Notch‐dependent activation of single‐minded expression." Dev Genes Evol 213(2): 65‐72. 136 Morimoto, M., Z. Liu, et al. (2010). "Canonical Notch signaling in the developing lung is required for determination of arterial smooth muscle cells and selection of Clara versus ciliated cell fate." J Cell Sci 123(Pt 2): 213‐24. Morrow, D., J. P. Cullen, et al. (2009). "Sonic Hedgehog induces Notch target gene expression in vascular smooth muscle cells via VEGF‐A." Arterioscler Thromb Vasc Biol 29(7): 1112‐8. Morrow, D., S. Guha, et al. (2008). "Notch and vascular smooth muscle cell phenotype." Circ Res 103(12): 1370‐82. Motoike, T., D. W. Markham, et al. (2003). "Evidence for novel fate of Flk1+ progenitor: contribution to muscle lineage." Genesis 35(3): 153‐9. Mukai, Y., Y. Rikitake, et al. (2006). "Decreased vascular lesion formation in mice with inducible endothelial‐specific expression of protein kinase Akt." J Clin Invest 116(2): 334‐43. Mungamuri, S. K., X. Yang, et al. (2006). "Survival signaling by Notch1: mammalian target of rapamycin (mTOR)‐dependent inhibition of p53." Cancer Res 66(9): 4715‐24. Munshi, N., A. Z. Fernandis, et al. (2002). "Lipopolysaccharide‐induced apoptosis of endothelial cells and its inhibition by vascular endothelial growth factor." J Immunol 168(11): 5860‐6. Nakajima, M., S. Yuasa, et al. (2003). "Abnormal blood vessel development in mice lacking presenilin‐ 1." Mech Dev 120(6): 657‐67. Niessen, K., Y. Fu, et al. (2008). "Slug is a direct Notch target required for initiation of cardiac cushion cellularization." J Cell Biol 182(2): 315‐25. Noguera‐Troise, I., C. Daly, et al. (2006). "Blockade of Dll4 inhibits tumour growth by promoting non‐ productive angiogenesis." Nature 444(7122): 1032‐7. Noseda, M., L. Chang, et al. (2004). "Notch activation induces endothelial cell cycle arrest and participates in contact inhibition: role of p21Cip1 repression." Mol Cell Biol 24(20): 8813‐22. Noseda, M., Y. Fu, et al. (2006). "Smooth Muscle alpha‐actin is a direct target of Notch/CSL." Circ Res 98(12): 1468‐70. Noseda, M., G. McLean, et al. (2004). "Notch activation results in phenotypic and functional changes consistent with endothelial‐to‐mesenchymal transformation." Circ Res 94(7): 910‐7. Noseda, M., K. Niessen, et al. (2005). "Notch‐dependent cell cycle arrest is associated with downregulation of minichromosome maintenance proteins." Circ Res 97(2): 102‐4. Nowak, G., A. Karrar, et al. (2004). "Expression of vascular endothelial growth factor receptor‐2 or Tie‐ 2 on peripheral blood cells defines functionally competent cell populations capable of reendothelialization." Circulation 110(24): 3699‐707. Oberg, C., J. Li, et al. (2001). "The Notch intracellular domain is ubiquitinated and negatively regulated by the mammalian Sel‐10 homolog." J Biol Chem 276(38): 35847‐53. Ogawara, Y., S. Kishishita, et al. (2002). "Akt enhances Mdm2‐mediated ubiquitination and degradation of p53." J Biol Chem 277(24): 21843‐50. Ohata, E., R. Tadokoro, et al. (2009). "Notch signal is sufficient to direct an endothelial conversion from non‐endothelial somitic cells conveyed to the aortic region by CXCR4." Dev Biol 335(1): 33‐42. Okajima, T. and K. D. Irvine (2002). "Regulation of notch signaling by o‐linked fucose." Cell 111(6): 893‐904. Okamura, Y. and Y. Saga (2008). "Pofut1 is required for the proper localization of the Notch receptor during mouse development." Mech Dev 125(8): 663‐73. Orlandi, A. and M. Bennett (2010). "Progenitor cell‐derived smooth muscle cells in vascular disease." Biochem Pharmacol 79(12): 1706‐13. Oswald, F., B. Tauber, et al. (2001). "p300 acts as a transcriptional coactivator for mammalian Notch‐ 1." Mol Cell Biol 21(22): 7761‐74. Owens, G. K., M. S. Kumar, et al. (2004). "Molecular regulation of vascular smooth muscle cell differentiation in development and disease." Physiol Rev 84(3): 767‐801. 137 Palomero, T., M. L. Sulis, et al. (2007). "Mutational loss of PTEN induces resistance to NOTCH1 inhibition in T‐cell leukemia." Nat Med 13(10): 1203‐10. Panin, V. M., V. Papayannopoulos, et al. (1997). "Fringe modulates Notch‐ligand interactions." Nature 387(6636): 908‐12. Pannu, H., V. T. Fadulu, et al. (2005). "Mutations in transforming growth factor‐beta receptor type II cause familial thoracic aortic aneurysms and dissections." Circulation 112(4): 513‐20. Pardali, E. and P. ten Dijke (2009). "Transforming growth factor‐beta signaling and tumor angiogenesis." Front Biosci 14: 4848‐61. Pardanaud, L., D. Luton, et al. (1996). "Two distinct endothelial lineages in ontogeny, one of them related to hemopoiesis." Development 122(5): 1363‐71. Parks, A. L., K. M. Klueg, et al. (2000). "Ligand endocytosis drives receptor dissociation and activation in the Notch pathway." Development 127(7): 1373‐85. Parrillo, J. E. (1993). "Pathogenetic mechanisms of septic shock." N Engl J Med 328(20): 1471‐7. Partanen, J., E. Armstrong, et al. (1992). "A novel endothelial cell surface receptor tyrosine kinase with extracellular epidermal growth factor homology domains." Mol Cell Biol 12(4): 1698‐707. Patel, N. S., J. L. Li, et al. (2005). "Up‐regulation of delta‐like 4 ligand in human tumor vasculature and the role of basal expression in endothelial cell function." Cancer Res 65(19): 8690‐7. Perez‐Pomares, J. M., D. Macias, et al. (1998). "The origin of the subepicardial mesenchyme in the avian embryo: an immunohistochemical and quail‐chick chimera study." Dev Biol 200(1): 57‐ 68. Perumalsamy, L. R., M. Nagala, et al. (2009). "A hierarchical cascade activated by non‐canonical Notch signaling and the mTOR‐Rictor complex regulates neglect‐induced death in mammalian cells." Cell Death Differ 16(6): 879‐89. Phng, L. K. and H. Gerhardt (2009). "Angiogenesis: a team effort coordinated by notch." Dev Cell 16(2): 196‐208. Pires‐daSilva, A. and R. J. Sommer (2003). "The evolution of signalling pathways in animal development." Nat Rev Genet 4(1): 39‐49. Pollman, M. J., L. Naumovski, et al. (1999). "Vascular cell apoptosis: cell type‐specific modulation by transforming growth factor‐beta1 in endothelial cells versus smooth muscle cells." Circulation 99(15): 2019‐26. Poole, T. J. and J. D. Coffin (1989). "Vasculogenesis and angiogenesis: two distinct morphogenetic mechanisms establish embryonic vascular pattern." J Exp Zool 251(2): 224‐31. Pouget, C., R. Gautier, et al. (2006). "Somite‐derived cells replace ventral aortic hemangioblasts and provide aortic smooth muscle cells of the trunk." Development 133(6): 1013‐22. Pouget, C., K. Pottin, et al. (2008). "Sclerotomal origin of vascular smooth muscle cells and pericytes in the embryo." Dev Biol 315(2): 437‐47. Pratt, E. B., J. S. Wentzell, et al. (2010). "The cell giveth and the cell taketh away: An overview of Notch pathway activation by endocytic trafficking of ligands and receptors." Acta Histochem. Proweller, A., W. S. Pear, et al. (2005). "Notch signaling represses myocardin‐induced smooth muscle cell differentiation." J Biol Chem 280(10): 8994‐9004. Proweller, A., A. C. Wright, et al. (2007). "Notch signaling in vascular smooth muscle cells is required to pattern the cerebral vasculature." Proc Natl Acad Sci U S A 104(41): 16275‐80. Puri, M. C., J. Partanen, et al. (1999). "Interaction of the TEK and TIE receptor tyrosine kinases during cardiovascular development." Development 126(20): 4569‐80. Quillard, T., S. Coupel, et al. (2008). "Impaired Notch4 activity elicits endothelial cell activation and apoptosis: implication for transplant arteriosclerosis." Arterioscler Thromb Vasc Biol 28(12): 2258‐65. Quillard, T., J. Devalliere, et al. (2009). "Notch2 signaling sensitizes endothelial cells to apoptosis by negatively regulating the key protective molecule survivin." PLoS One 4(12): e8244. 138 Rand, M. D., L. M. Grimm, et al. (2000). "Calcium depletion dissociates and activates heterodimeric notch receptors." Mol Cell Biol 20(5): 1825‐35. Rao, S., I. B. Lobov, et al. (2007). "Obligatory participation of macrophages in an angiopoietin 2‐ mediated cell death switch." Development 134(24): 4449‐58. Ray, W. J., M. Yao, et al. (1999). "Cell surface presenilin‐1 participates in the gamma‐secretase‐like proteolysis of Notch." J Biol Chem 274(51): 36801‐7. Raymond, M. A., A. Desormeaux, et al. (2004). "Apoptosis of endothelial cells triggers a caspase‐ dependent anti‐apoptotic paracrine loop active on VSMC." FASEB J 18(6): 705‐7. Rebay, I., R. J. Fleming, et al. (1991). "Specific EGF repeats of Notch mediate interactions with Delta and Serrate: implications for Notch as a multifunctional receptor." Cell 67(4): 687‐99. Reiss, Y., J. Droste, et al. (2007). "Angiopoietin‐2 impairs revascularization after limb ischemia." Circ Res 101(1): 88‐96. Reynolds, L. P., S. D. Killilea, et al. (1992). "Angiogenesis in the female reproductive system." FASEB J 6(3): 886‐92. Ridgway, J., G. Zhang, et al. (2006). "Inhibition of Dll4 signalling inhibits tumour growth by deregulating angiogenesis." Nature 444(7122): 1083‐7. Robert‐Moreno, A., J. Guiu, et al. (2008). "Impaired embryonic haematopoiesis yet normal arterial development in the absence of the Notch ligand Jagged1." EMBO J 27(13): 1886‐95. Roca, C. and R. H. Adams (2007). "Regulation of vascular morphogenesis by Notch signaling." Genes Dev 21(20): 2511‐24. Rodewald, H. R. and T. N. Sato (1996). "Tie1, a receptor tyrosine kinase essential for vascular endothelial cell integrity, is not critical for the development of hematopoietic cells." Oncogene 12(2): 397‐404. Rooke, J., D. Pan, et al. (1996). "KUZ, a conserved metalloprotease‐disintegrin protein with two roles in Drosophila neurogenesis." Science 273(5279): 1227‐31. Ruchoux, M. M. and C. A. Maurage (1997). "CADASIL: Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy." J Neuropathol Exp Neurol 56(9): 947‐64. Saegusa, M., M. Hashimura, et al. (2009). "Requirement of the Akt/beta‐catenin pathway for uterine carcinosarcoma genesis, modulating E‐cadherin expression through the transactivation of slug." Am J Pathol 174(6): 2107‐15. Sainson, R. C., J. Aoto, et al. (2005). "Cell‐autonomous notch signaling regulates endothelial cell branching and proliferation during vascular tubulogenesis." FASEB J 19(8): 1027‐9. Sakao, S., L. Taraseviciene‐Stewart, et al. (2007). "VEGF‐R blockade causes endothelial cell apoptosis, expansion of surviving CD34+ precursor cells and transdifferentiation to smooth muscle‐like and neuronal‐like cells." FASEB J 21(13): 3640‐52. Sakata, Y., N. Koibuchi, et al. (2006). "The spectrum of cardiovascular anomalies in CHF1/Hey2 deficient mice reveals roles in endocardial cushion, myocardial and vascular maturation." J Mol Cell Cardiol 40(2): 267‐73. Sakata, Y., F. Xiang, et al. (2004). "Transcription factor CHF1/Hey2 regulates neointimal formation in vivo and vascular smooth muscle proliferation and migration in vitro." Arterioscler Thromb Vasc Biol 24(11): 2069‐74. Sakoda, H., Y. Gotoh, et al. (2003). "Differing roles of Akt and serum‐ and glucocorticoid‐regulated kinase in glucose metabolism, DNA synthesis, and oncogenic activity." J Biol Chem 278(28): 25802‐7. Santos, M. A., L. M. Sarmento, et al. (2007). "Notch1 engagement by Delta‐like‐1 promotes differentiation of B lymphocytes to antibody‐secreting cells." Proc Natl Acad Sci U S A 104(39): 15454‐9. Sarao, R. and D. J. Dumont (1998). "Conditional transgene expression in endothelial cells." Transgenic Res 7(6): 421‐7. 139 Sarbassov, D. D., S. M. Ali, et al. (2004). "Rictor, a novel binding partner of mTOR, defines a rapamycin‐insensitive and raptor‐independent pathway that regulates the cytoskeleton." Curr Biol 14(14): 1296‐302. Sasai, Y., R. Kageyama, et al. (1992). "Two mammalian helix‐loop‐helix factors structurally related to Drosophila hairy and Enhancer of split." Genes Dev 6(12B): 2620‐34. Sasamura, T., H. O. Ishikawa, et al. (2007). "The O‐fucosyltransferase O‐fut1 is an extracellular component that is essential for the constitutive endocytic trafficking of Notch in Drosophila." Development 134(7): 1347‐56. Sata, M., A. Saiura, et al. (2002). "Hematopoietic stem cells differentiate into vascular cells that participate in the pathogenesis of atherosclerosis." Nat Med 8(4): 403‐9. Sato, Y., T. Watanabe, et al. (2008). "Notch mediates the segmental specification of angioblasts in somites and their directed migration toward the dorsal aorta in avian embryos." Dev Cell 14(6): 890‐901. Schroeder, T., F. Meier‐Stiegen, et al. (2006). "Activated Notch1 alters differentiation of embryonic stem cells into mesodermal cell lineages at multiple stages of development." Mech Dev 123(7): 570‐9. Schroeter, E. H., J. A. Kisslinger, et al. (1998). "Notch‐1 signalling requires ligand‐induced proteolytic release of intracellular domain." Nature 393(6683): 382‐6. Schwartz, B. R., A. Karsan, et al. (1999). "A novel beta 1 integrin‐dependent mechanism of leukocyte adherence to apoptotic cells." J Immunol 162(8): 4842‐8. Schwartz, S. M. and E. P. Benditt (1977). "Aortic endothelial cell replication. I. Effects of age and hypertension in the rat." Circ Res 41(2): 248‐55. Shawber, C., D. Nofziger, et al. (1996). "Notch signaling inhibits muscle cell differentiation through a CBF1‐independent pathway." Development 122(12): 3765‐73. Shawber, C. J., I. Das, et al. (2003). "Notch signaling in primary endothelial cells." Ann N Y Acad Sci 995: 162‐70. Shawber, C. J., Y. Funahashi, et al. (2007). "Notch alters VEGF responsiveness in human and murine endothelial cells by direct regulation of VEGFR‐3 expression." J Clin Invest 117(11): 3369‐82. Shawber, C. J. and J. Kitajewski (2004). "Notch function in the vasculature: insights from zebrafish, mouse and man." Bioessays 26(3): 225‐34. Shen, H., A. S. McElhinny, et al. (2006). "The Notch coactivator, MAML1, functions as a novel coactivator for MEF2C‐mediated transcription and is required for normal myogenesis." Genes Dev 20(6): 675‐88. Shin, M., H. Nagai, et al. (2009). "Notch mediates Wnt and BMP signals in the early separation of smooth muscle progenitors and blood/endothelial common progenitors." Development 136(4): 595‐603. Shinbrot, E., K. G. Peters, et al. (1994). "Expression of the platelet‐derived growth factor beta receptor during organogenesis and tissue differentiation in the mouse embryo." Dev Dyn 199(3): 169‐ 75. Sizemore, N., S. Leung, et al. (1999). "Activation of phosphatidylinositol 3‐kinase in response to interleukin‐1 leads to phosphorylation and activation of the NF‐kappaB p65/RelA subunit." Mol Cell Biol 19(7): 4798‐805. Sorensen, I., R. H. Adams, et al. (2009). "DLL1‐mediated Notch activation regulates endothelial identity in mouse fetal arteries." Blood 113(22): 5680‐8. Souilhol, C., S. Cormier, et al. (2006). "Nas transgenic mouse line allows visualization of Notch pathway activity in vivo." Genesis 44(6): 277‐86. Stefanec, T. (2000). "Endothelial apoptosis: could it have a role in the pathogenesis and treatment of disease?" Chest 117(3): 841‐54. Stifani, S., C. M. Blaumueller, et al. (1992). "Human homologs of a Drosophila Enhancer of split gene product define a novel family of nuclear proteins." Nat Genet 2(4): 343. 140 Suchting, S., C. Freitas, et al. (2007). "The Notch ligand Delta‐like 4 negatively regulates endothelial tip cell formation and vessel branching." Proc Natl Acad Sci U S A 104(9): 3225‐30. Suhara, T., K. Fukuo, et al. (2004). "Homocysteine enhances endothelial apoptosis via upregulation of Fas‐mediated pathways." Hypertension 43(6): 1208‐13. Sun, J. F., T. Phung, et al. (2005). "Microvascular patterning is controlled by fine‐tuning the Akt signal." Proc Natl Acad Sci U S A 102(1): 128‐33. Swiatek, P. J., C. E. Lindsell, et al. (1994). "Notch1 is essential for postimplantation development in mice." Genes Dev 8(6): 707‐19. Takahashi, Y., T. Imanaka, et al. (1996). "Spatial and temporal pattern of smooth muscle cell differentiation during development of the vascular system in the mouse embryo." Anat Embryol (Berl) 194(5): 515‐26. Takeshita, K., M. Satoh, et al. (2007). "Critical role of endothelial Notch1 signaling in postnatal angiogenesis." Circ Res 100(1): 70‐8. Tamura, K., Y. Taniguchi, et al. (1995). "Physical interaction between a novel domain of the receptor Notch and the transcription factor RBP‐J kappa/Su(H)." Curr Biol 5(12): 1416‐23. Tang, Y., S. Urs, et al. (2008). "Hairy‐related transcription factors inhibit Notch‐induced smooth muscle alpha‐actin expression by interfering with Notch intracellular domain/CBF‐1 complex interaction with the CBF‐1‐binding site." Circ Res 102(6): 661‐8. Thornberry, N. A. and Y. Lazebnik (1998). "Caspases: enemies within." Science 281(5381): 1312‐6. Timmerman, L. A., J. Grego‐Bessa, et al. (2004). "Notch promotes epithelial‐mesenchymal transition during cardiac development and oncogenic transformation." Genes Dev 18(1): 99‐115. Torsney, E., K. Mandal, et al. (2007). "Characterisation of progenitor cells in human atherosclerotic vessels." Atherosclerosis 191(2): 259‐64. Tricot, O., Z. Mallat, et al. (2000). "Relation between endothelial cell apoptosis and blood flow direction in human atherosclerotic plaques." Circulation 101(21): 2450‐3. Trindade, A., S. R. Kumar, et al. (2008). "Overexpression of delta‐like 4 induces arterialization and attenuates vessel formation in developing mouse embryos." Blood 112(5): 1720‐9. Tsukada, T., K. Eguchi, et al. (1995). "Transforming growth factor beta 1 induces apoptotic cell death in cultured human umbilical vein endothelial cells with down‐regulated expression of bcl‐2." Biochem Biophys Res Commun 210(3): 1076‐82. Tu, L., T. C. Fang, et al. (2005). "Notch signaling is an important regulator of type 2 immunity." J Exp Med 202(8): 1037‐42. Tun, T., Y. Hamaguchi, et al. (1994). "Recognition sequence of a highly conserved DNA binding protein RBP‐J kappa." Nucleic Acids Res 22(6): 965‐71. Tyagi, N., A. V. Ovechkin, et al. (2006). "Mitochondrial mechanism of microvascular endothelial cells apoptosis in hyperhomocysteinemia." J Cell Biochem 98(5): 1150‐62. Uemura, A., S. Kusuhara, et al. (2006). "Angiogenesis in the mouse retina: a model system for experimental manipulation." Exp Cell Res 312(5): 676‐83. Uyttendaele, H., J. Ho, et al. (2001). "Vascular patterning defects associated with expression of activated Notch4 in embryonic endothelium." Proc Natl Acad Sci U S A 98(10): 5643‐8. Vaccari, T., H. Lu, et al. (2008). "Endosomal entry regulates Notch receptor activation in Drosophila melanogaster." J Cell Biol 180(4): 755‐62. van Oostrom, O., J. O. Fledderus, et al. (2009). "Smooth muscle progenitor cells: friend or foe in vascular disease?" Curr Stem Cell Res Ther 4(2): 131‐40. Varadkar, P., M. Kraman, et al. (2008). "Notch2 is required for the proliferation of cardiac neural crest‐derived smooth muscle cells." Dev Dyn 237(4): 1144‐52. Venkatesh, D. A., K. S. Park, et al. (2008). "Cardiovascular and hematopoietic defects associated with Notch1 activation in embryonic Tie2‐expressing populations." Circ Res 103(4): 423‐31. 141 Verkleij‐Hagoort, A., J. Bliek, et al. (2007). "Hyperhomocysteinemia and MTHFR polymorphisms in association with orofacial clefts and congenital heart defects: a meta‐analysis." Am J Med Genet A 143A(9): 952‐60. Vermes, I., C. Haanen, et al. (1995). "A novel assay for apoptosis. Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled Annexin V." J Immunol Methods 184(1): 39‐51. Vernon, A. E. and C. LaBonne (2006). "Slug stability is dynamically regulated during neural crest development by the F‐box protein Ppa." Development 133(17): 3359‐70. Villa, N., L. Walker, et al. (2001). "Vascular expression of Notch pathway receptors and ligands is restricted to arterial vessels." Mech Dev 108(1‐2): 161‐4. Vinas‐Castells, R., M. Beltran, et al. (2010). "The hypoxia‐controlled FBXL14 ubiquitin ligase targets SNAIL1 for proteasome degradation." J Biol Chem 285(6): 3794‐805. Vogeli, K. M., S. W. Jin, et al. (2006). "A common progenitor for haematopoietic and endothelial lineages in the zebrafish gastrula." Nature 443(7109): 337‐9. Waltzer, L., F. Logeat, et al. (1994). "The human J kappa recombination signal sequence binding protein (RBP‐J kappa) targets the Epstein‐Barr virus EBNA2 protein to its DNA responsive elements." EMBO J 13(23): 5633‐8. Wang, H. L., I. O. Akinci, et al. (2007). "The intrinsic apoptotic pathway is required for lipopolysaccharide‐induced lung endothelial cell death." J Immunol 179(3): 1834‐41. Wang, H. U., Z. F. Chen, et al. (1998). "Molecular distinction and angiogenic interaction between embryonic arteries and veins revealed by ephrin‐B2 and its receptor Eph‐B4." Cell 93(5): 741‐ 53. Wang, S. P., W. L. Wang, et al. (2009). "p53 controls cancer cell invasion by inducing the MDM2‐ mediated degradation of Slug." Nat Cell Biol 11(6): 694‐704. Wang, W., C. Z. Prince, et al. (2003). "HRT1 modulates vascular smooth muscle cell proliferation and apoptosis." Biochem Biophys Res Commun 308(3): 596‐601. Wang, Z., Y. Li, et al. (2010). "Down‐regulation of Notch‐1 and Jagged‐1 inhibits prostate cancer cell growth, migration and invasion, and induces apoptosis via inactivation of Akt, mTOR, and NF‐ kappaB signaling pathways." J Cell Biochem 109(4): 726‐36. Wasteson, P., B. R. Johansson, et al. (2008). "Developmental origin of smooth muscle cells in the descending aorta in mice." Development 135(10): 1823‐32. Weinstein, B. M., D. L. Stemple, et al. (1995). "Gridlock, a localized heritable vascular patterning defect in the zebrafish." Nat Med 1(11): 1143‐7. Weng, A. P., Y. Nam, et al. (2003). "Growth suppression of pre‐T acute lymphoblastic leukemia cells by inhibition of notch signaling." Mol Cell Biol 23(2): 655‐64. Wharton, K. A., K. M. Johansen, et al. (1985). "Nucleotide sequence from the neurogenic locus notch implies a gene product that shares homology with proteins containing EGF‐like repeats." Cell 43(3 Pt 2): 567‐81. Wheatley, S. C., C. M. Isacke, et al. (1993). "Restricted expression of the hyaluronan receptor, CD44, during postimplantation mouse embryogenesis suggests key roles in tissue formation and patterning." Development 119(2): 295‐306. Wiegreffe, C., B. Christ, et al. (2009). "Remodeling of aortic smooth muscle during avian embryonic development." Dev Dyn 238(3): 624‐31. Wilson‐Rawls, J., J. D. Molkentin, et al. (1999). "Activated notch inhibits myogenic activity of the MADS‐Box transcription factor myocyte enhancer factor 2C." Mol Cell Biol 19(4): 2853‐62. Wilson, K. M. and S. R. Lentz (2005). "Mechanisms of the atherogenic effects of elevated homocysteine in experimental models." Semin Vasc Med 5(2): 163‐71. Wilting, J. and J. Becker (2006). "Two endothelial cell lines derived from the somite." Anat Embryol (Berl) 211 Suppl 1: 57‐63. 142 Wolfram, J. A., D. Diaconu, et al. (2009). "Keratinocyte but not endothelial cell‐specific overexpression of Tie2 leads to the development of psoriasis." Am J Pathol 174(4): 1443‐58. Wong, F., C. Hull, et al. (2004). "Lipopolysaccharide initiates a TRAF6‐mediated endothelial survival signal." Blood 103(12): 4520‐6. Wong, G. T., D. Manfra, et al. (2004). "Chronic treatment with the gamma‐secretase inhibitor LY‐ 411,575 inhibits beta‐amyloid peptide production and alters lymphopoiesis and intestinal cell differentiation." J Biol Chem 279(13): 12876‐82. Wu, G., S. Lyapina, et al. (2001). "SEL‐10 is an inhibitor of notch signaling that targets notch for ubiquitin‐mediated protein degradation." Mol Cell Biol 21(21): 7403‐15. Wu, L., T. Sun, et al. (2002). "Identification of a family of mastermind‐like transcriptional coactivators for mammalian notch receptors." Mol Cell Biol 22(21): 7688‐700. Wu, W. S., S. Heinrichs, et al. (2005). "Slug antagonizes p53‐mediated apoptosis of hematopoietic progenitors by repressing puma." Cell 123(4): 641‐53. Xue, Y., X. Gao, et al. (1999). "Embryonic lethality and vascular defects in mice lacking the Notch ligand Jagged1." Hum Mol Genet 8(5): 723‐30. Yamashita, J., H. Itoh, et al. (2000). "Flk1‐positive cells derived from embryonic stem cells serve as vascular progenitors." Nature 408(6808): 92‐6. Yang, L., M. H. Soonpaa, et al. (2008). "Human cardiovascular progenitor cells develop from a KDR+ embryonic‐stem‐cell‐derived population." Nature 453(7194): 524‐8. Yoshida, T. and G. K. Owens (2005). "Molecular determinants of vascular smooth muscle cell diversity." Circ Res 96(3): 280‐91. You, L. R., F. J. Lin, et al. (2005). "Suppression of Notch signalling by the COUP‐TFII transcription factor regulates vein identity." Nature 435(7038): 98‐104. Zen, K., A. Karsan, et al. (1998). "Lipopolysaccharide‐induced NF‐kappaB activation in human endothelial cells involves degradation of IkappaBalpha but not IkappaBbeta." Exp Cell Res 243(2): 425‐33. Zengin, E., F. Chalajour, et al. (2006). "Vascular wall resident progenitor cells: a source for postnatal vasculogenesis." Development 133(8): 1543‐51. Zhang, C., Y. Cai, et al. (2001). "Homocysteine induces programmed cell death in human vascular endothelial cells through activation of the unfolded protein response." J Biol Chem 276(38): 35867‐74. Zhang, C., T. F. Carl, et al. (2006). "An NF‐kappaB and slug regulatory loop active in early vertebrate mesoderm." PLoS One 1: e106. Zhang, Y., C. Riesterer, et al. (1996). "Inducible site‐directed recombination in mouse embryonic stem cells." Nucleic Acids Res 24(4): 543‐8. Zhong, T. P., M. Rosenberg, et al. (2000). "gridlock, an HLH gene required for assembly of the aorta in zebrafish." Science 287(5459): 1820‐4. Zhou, B. P., J. Deng, et al. (2004). "Dual regulation of Snail by GSK‐3beta‐mediated phosphorylation in control of epithelial‐mesenchymal transition." Nat Cell Biol 6(10): 931‐40. 143 APPENDICES Appendix A. Ethics approvals The following are the animal care certificates, human ethical approvals and biohazard approvals required during the work shown in this thesis. ANIMAL CARE CERTIFICATE The Animal Care Committee has examined and approved the use of animals for the above experimental project. This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the CCAC and some granting agencies. A copy of this certificate must be displayed in your animal facility. Office of Research Services and Administration 102, 6190 Agronomy Road, Vancouver, BC V6T 1Z3 Phone: 604-827-5111 Fax: 604-822-5093 THE UNIVERSITY OF BRITISH COLUMBIA Application Number: A06-0137 Investigator or Course Director: Aly Karsan Department: Pathology & Laboratory Medicine Funding Sources: Animals: Mice Tie1-tTA x TetOS:nlsLacZ 40 Start Date: July 1, 2006 Approval Date: June 23, 2009 Funding Agency: Heart and Stroke Foundation of British Columbia and Yukon Funding Title: Molecular mechanisms of endothelial survival/apoptosis Unfunded title: N/A Page 1 of 1 4/28/2010https://rise.ubc.ca/rise/Doc/0/04BBG2K1IQ1452V3HL9BQ0H4A7/fromString.html ANIMAL CARE CERTIFICATE THE UNIVERSITY OF BRITISH COLUMBIA Application Number: A07-0717 Investigator or Course Director: Aly Karsan Department: Pathology & Laboratory Medicine Funding Sources: Animals: Mice Rosa 101 Mice C57Bl/6J 253 Mice VE-tTA 266 Mice Top-NotchIC 248 Mice ScL-CRE-ERT 303 Mice eNOS-/- 65 Mice Tie1-tTA 266 Mice Notch1tm2Rko/GridJ 101 Mice VE-CRE 10 Mice SM22-rtTA 200 Mice TekCRE 10 Mice TetOS-dnMAML-GFP 266 Mice Rosa-YFP 121 Mice TetOS-LacZ 218 Start Date: December 14, 2007 Approval Date: March 26, 2009 Funding Agency: Canadian Institutes of Health Research (CIHR) Funding Title: Endothelial to mesenchymal transformation Funding Agency: Heart and Stroke Foundation of British Columbia and Yukon Funding Title: Dissecting gene regulatory networks in cardiac cushion development Funding Agency: Canadian Institutes of Health Research (CIHR) Funding Title: Endothelial to mesenchymal transformation Funding Agency: Genome British Columbia Funding Title: Dissecting gene regulatory networks in mammalian organogenesis Page 1 of 2 4/28/2010https://rise.ubc.ca/rise/Doc/0/3ELV8LB7INNKHBM3RI0ABFGA37/fromString.html The Animal Care Committee has examined and approved the use of animals for the above experimental project. This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the CCAC and some granting agencies. A copy of this certificate must be displayed in your animal facility. Office of Research Services and Administration 102, 6190 Agronomy Road, Vancouver, BC V6T 1Z3 Phone: 604-827-5111 Fax: 604-822-5093 Funding Agency: Genome Canada Funding Title: Dissecting gene regulatory networks in mammalian organogenesis Unfunded title: N/A Page 2 of 2 4/28/2010https://rise.ubc.ca/rise/Doc/0/3ELV8LB7INNKHBM3RI0ABFGA37/fromString.html The Principal Investigator/Course Director is responsible for ensuring that all research or course work involving biological hazards is conducted in accordance with the University of British Columbia Policies and Procedures, Biosafety Practices and Public Health Agency of Canada guidelines. This certificate is valid for one year from the above start or approval date (whichever is later) provided there are no changes. Annual review is required. A copy of this certificate must be displayed in your facility. Office of Research Services 102, 6190 Agronomy Road, Vancouver, V6T 1Z3 Phone: 604-827-5111 FAX: 604-822-5093 The University of British Columbia Biohazard Approval Certificate PROTOCOL NUMBER: B06-0065 INVESTIGATOR OR COURSE DIRECTOR: Aly Karsan DEPARTMENT: Pathology & Laboratory Medicine PROJECT OR COURSE TITLE: Molecular Mechanisms of Endothelial Survival/Apoptosis APPROVAL DATE: September 2, 2009 START DATE: June 29, 2006 APPROVED CONTAINMENT LEVEL: 2 UNFUNDED TITLE: N/A Page 1 of 1 4/28/2010https://rise.ubc.ca/rise/Doc/0/BI20KGEQN91KDANHGEFTLQR3F9/fromString.html The Principal Investigator/Course Director is responsible for ensuring that all research or course work involving biological hazards is conducted in accordance with the University of British Columbia Policies and Procedures, Biosafety Practices and Public Health Agency of Canada guidelines. This certificate is valid for one year from the above start or approval date (whichever is later) provided there are no changes. Annual review is required. A copy of this certificate must be displayed in your facility. Office of Research Services 102, 6190 Agronomy Road, Vancouver, V6T 1Z3 Phone: 604-827-5111 FAX: 604-822-5093 The University of British Columbia Biohazard Approval Certificate PROTOCOL NUMBER: B06-0124 INVESTIGATOR OR COURSE DIRECTOR: Aly Karsan DEPARTMENT: Pathology & Laboratory Medicine PROJECT OR COURSE TITLE: Endothelial-to-mesenchymal transformation APPROVAL DATE: September 2, 2009 START DATE: June 29, 2006 APPROVED CONTAINMENT LEVEL: 2 UNFUNDED TITLE: N/A Page 1 of 1 4/28/2010https://rise.ubc.ca/rise/Doc/0/7C5PLOT8ODS4VELQ730LF7GO39/fromString.html 150 Appendix B. List of publications The following is a list of publications achieved during my graduate school career. Publications listed here are results of collaborative work with other members of the lab examining the role of Notch in endothelial-mesenchymal transdifferentiation and endothelial cell cycle regulation. Each publication is summarized here and my contribution to the publication is also listed. Noseda M*, McLean G*, Niessen K, Chang L, Pollet I, Montpetit R, Shahidi R, Dorovini-Zis K, Li L, Beckstead B, Durand RE, Hoodless PA, Karsan A. (2004) Notch activation results in phenotypic and functional changes consistent with endothelial-to-mesenchymal transformation. Circ Res. 94, 910-917 • This peer-reviewed article is among the first to demonstrate in vitro Notch-induced endothelial-to-mesenchymal transdifferentiation, a process examined in vivo in Chapter 3 of this thesis • I was involved in generating the endothelial cell lines used in this study and in the generation of the following data: Figure 2. Notch activation induces mesenchymal transdifferentiation of endothelial cells from different vascular beds. Figure 5B. Jagged1 induces EMT. Supplemental Figure 1. Notch activation induced EMT in endothelial cells from different vascular beds Noseda M, Chang L, McLean G, Grim JE, Clurman BE, Smith LL, Karsan A. (2004) Notch Activation Induces Endothelial Cell Cycle Arrest and Participates in Contact Inhibition: Role of p21Cip1 Repression. Mol Cell Biol. 24, 8813-8822 • This peer-reviewed article shows a role of Notch in endothelial contact-inhibition. • I was involved in generating the endothelial cell lines used in this study and in the generation of the following data: Figure 5A. Tamoxifen-induced nuclear translocation of the Notch4IC-ER fusion protein. Figure 6B. Notch inhibits serum-induced cdk4-kinase activity. Figure 7A. Expression of p21 rescues the reduction of cdk4 nuclear localization. Figure 8C. Notch inhibition reduces endothelial contact inhibition of proliferation. Fu Y, Chang A, Chang L, Niessen K, Eapen S, Setiadi A, Karsan A. (2009) Differential regulation of transforming growth factor beta signaling pathways by Notch in human endothelial cells. J Biol Chem. 284(29):19452-62. 151 • This peer-reviewed article demonstrated a mechanism for direct interaction between the Notch and TGF-β signaling pathways in endothelial cells, which are both implicated in endothelial-to-mesenchymal transdifferentiation. • I was involved in generating the following data: Figure 4. Inhibition of Notch reduces Smad3 expression in the developing heart. Niessen K, Fu Y, Chang L, Hoodless PA, McFadden D, Karsan A. (2008) Slug is a direct Notch target required for initiation of cardiac cushion cellularization. J Cell Biol. 182(2):315- 25. • This peer-reviewed article demonstrates that Slug is a direct target for Notch and the role of Slug in endothelial-to-mesenchymal transdifferentiation during heart development. Chapter 4 of this thesis discussed an additional function for Notch- induced Slug expression. • I was involved in the unpublished data using the transgenic system I generated as described in this thesis. Noseda M, Fu Y, Niessen K, Wong F, Chang L, McLean G, Karsan A. (2006) Smooth Muscle alpha-actin is a direct target of Notch/CSL. Circ Res. 98(12), 1468-1470 • This peer-reviewed article shows that SMA is a direct target of Notch/CSL transcription regulation. • I was involved in generating the endothelial, fibroblast, and vascular smooth muscle cell lines used in this study and in the generation of the following data: Supplemental Figure 1. Notch induces transcript level of SMA MacKenzie F, Duriez P, Larrivee B, Chang L, Pollet I, Wong F, Yip C, Karsan A. (2004) Notch4-induced inhibition of endothelial sprouting requires the ankyrin repeats and involves signaling through RBP-Jkappa. Blood 104, 1760-1768 • This peer-reviewed article shows Notch-induced inhibition of angiogenesis is CSL- dependent and is partially through regulation of endothelial migration. • I was involved in generating a Notch mutant construct used in this study. Noseda M, Niessen K, McLean G, Chang L, Karsan A. (2005) Notch-dependent cell cycle arrest is associated with downregulation of minichromosome maintenance proteins. Circ Res. 97(2), 102-104. • This peer-reviewed article demonstrated another mechanism for Notch-induced endothelial growth inhibition. • I was involved in generating the endothelial, fibroblast, and vascular smooth muscle cell lines used in this study. 152 Appendix C. Previously published material The following is a compilation of the peer-reviewed publications listed in Appendix B: ISSN: 1524-4571 Copyright © 2004 American Heart Association. All rights reserved. Print ISSN: 0009-7330. Online TX 72514 Circulation Research is published by the American Heart Association. 7272 Greenville Avenue, Dallas, DOI: 10.1161/01.RES.0000124300.76171.C9 2004;94;910-917; originally published online Feb 26, 2004; Circ. Res. Ralph E. Durand, Pamela A. Hoodless and Aly Karsan Montpetit, Réza Shahidi, Katerina Dorovini-Zis, Linheng Li, Benjamin Beckstead, Michela Noseda, Graeme McLean, Kyle Niessen, Linda Chang, Ingrid Pollet, Rachel Endothelial-to-Mesenchymal Transformation Notch Activation Results in Phenotypic and Functional Changes Consistent With http://circres.ahajournals.org/cgi/content/full/94/7/910/DC1 Data Supplement (unedited) at: http://circres.ahajournals.org/cgi/content/full/94/7/910 located on the World Wide Web at: The online version of this article, along with updated information and services, is http://www.lww.com/reprints Reprints: Information about reprints can be found online at firstname.lastname@example.org 410-528-8550. E-mail: Fax:Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters http://circres.ahajournals.org/subscriptions/ Subscriptions: Information about subscribing to Circulation Research is online at by on April 28, 2010 circres.ahajournals.orgDownloaded from Notch Activation Results in Phenotypic and Functional Changes Consistent With Endothelial-to-Mesenchymal Transformation Michela Noseda,* Graeme McLean,* Kyle Niessen, Linda Chang, Ingrid Pollet, Rachel Montpetit, Réza Shahidi, Katerina Dorovini-Zis, Linheng Li, Benjamin Beckstead, Ralph E. Durand, Pamela A. Hoodless, Aly Karsan Abstract—Various studies have identified a critical role for Notch signaling in cardiovascular development. In this and other systems, Notch receptors and ligands are expressed in regions that undergo epithelial-to-mesenchymal transformation. However, there is no direct evidence that Notch activation can induce mesenchymal transdifferentiation. In this study we show that Notch activation in endothelial cells results in morphological, phenotypic, and functional changes consistent with mesenchymal transformation. These changes include downregulation of endothelial markers (vascular endothelial [VE]-cadherin, Tie1, Tie2, platelet-endothelial cell adhesion molecule-1, and endothelial NO synthase), upregulation of mesenchymal markers (-smooth muscle actin, fibronectin, and platelet-derived growth factor receptors), and migration toward platelet-derived growth factor-BB. Notch-induced endothelial-to-mesenchymal transformation does not seem to require external regulation and is restricted to cells expressing activated Notch. Jagged1 stimulation of endothelial cells induces a similar mesenchymal transformation, and Jagged1, Notch1, and Notch4 are expressed in the ventricular outflow tract during stages of endocardial cushion formation. This is the first evidence that Jagged1-Notch interactions induce endothelial-to-mesenchymal transformation, and our findings suggest that Notch signaling may be required for proper endocardial cushion differentiation and/or vascular smooth muscle cell development. (Circ Res. 2004;94:910-917.) Key Words: endothelial-to-mesenchymal transformation Notch Jagged1 endocardial cushion The Notch signaling pathway plays a critical role duringdevelopment. Four mammalian Notch receptors (Notch1 through 4) and 5 Notch ligands (Delta-like [Dll]-1, Dll3, Dll4, Jagged1, and Jagged2) have been identified. Notch receptor- ligand interaction results in a series of proteolytic cleavages of the Notch receptor, producing a C-terminal intracellular fragment (NotchIC) that translocates to the nucleus. In the nucleus, NotchIC binds to the transcriptional repressor CBF1/ RBP-J, thereby derepressing or coactivating the expression of various lineage-specific genes.1 Perturbation of the Notch pathway has been implicated in the pathogenesis of various cardiovascular diseases in hu- mans.2 Of interest, patients with Jagged1 mutations (Alagille syndrome) display congenital cardiovascular anomalies that seem to be secondary to faulty endocardial cushion forma- tion.3–6 In the mouse, Notch1-deficient embryos demonstrate severe vascular developmental defects, which are exacerbated in Notch1/Notch4 double-mutant embryos.7 Constitutive ac- tivation of Notch4 also causes defects in vascular remodel- ing.8,9 Mice that are rendered null for Jagged1 die from hemorrhage early during embryogenesis, whereas mice that are doubly heterozygous for a Jagged1-null allele and a Notch2 hypomorphic allele exhibit cardiac anomalies similar to those seen in Alagille syndrome.10,11 Genes that lie downstream of Notch activation, such as the basic helix-loop- helix factor, HRT2/HEY2, have also been implicated in cardiovascular development.12,13 Notch receptors and their ligands have been localized to the vasculature.14 Notch receptors have also been observed in the endocardium, and the Notch ligand Jagged1 is present on endocardial and periendocardial cells of the cardiac cush- ions.15,16 The endocardial cushion is a specialized embryonic Original received November 3, 2003; revision received February 6, 2004; accepted February 16, 2004. From the Department of Pathology and Laboratory Medicine (M.N., I.P., A.K.) and Experimental Medicine Program (G.M., K.N., L.C., A.K.), University of British Columbia; Department of Medical Biophysics (M.N., G.M., K.N., L.C., I.P., R.E.D., A.K.), Terry Fox Research Laboratories (R.M., P.A.H.), and Department of Pathology and Laboratory Medicine (A.K.), British Columbia Cancer Agency; and Department of Pathology (R.S., K.D.-Z.), Division of Neuropathology, University of British Columbia and Vancouver Hospital and Health Sciences Centre, Vancouver, British Columbia, Canada; Stem Cell Research Laboratory (L.L.), Stowers Institute for Medical Research, Kansas City, Mo; and Department of Bioengineering (B.B.), University of Washington, Seattle, Wash. *Both authors contributed equally to this study. Correspondence to Aly Karsan, Department of Pathology and Laboratory Medicine and Experimental Medicine Program, University of British Columbia, Vancouver, BC V6T 2B5, Canada. E-mail email@example.com © 2004 American Heart Association, Inc. Circulation Research is available at http://www.circresaha.org DOI: 10.1161/01.RES.0000124300.76171.C9 910 by on April 28, 2010 circres.ahajournals.orgDownloaded from tissue that gives rise to the cardiac valves and membranous septa. A critical event in cardiac cushion formation is a differentiation process referred to as endothelial-to- mesenchymal transformation (EMT), which is a specific form of epithelial-to-mesenchymal transformation.17,18 In the car- diovascular system of the adult, mesenchymal cells derived from the transformation of a subset of cardiac valve endothe- lial cells may also be necessary for maintenance of the leaflet architecture.19 Furthermore, EMT may play a role in the development of neointimal lesions in transplant atherosclero- sis and restenosis.20 Intercellular signaling between Notch receptors and li- gands is critical for cell fate determination by influencing cell proliferation, differentiation, and apoptosis.21 Notch members and their ligands are expressed in various regions that undergo EMT in order for development to proceed appropri- ately.22,23 Our studies demonstrate that Notch activation in endothelial cells promotes mesenchymal transformation and suggest that Jagged1-Notch interactions may participate in endocardial cushion formation by inducing EMT. Materials and Methods Cell Culture and Reagents The HMEC-1 microvascular endothelial cell line, hereafter referred to as HMEC, was provided by the Centers for Disease Control and Prevention (Atlanta, Ga) and cultured as previously described.9 Human umbilical vein endothelial cells (HUVECs) were isolated and cultured as previously described.24 Human aortic endothelial cells (HAECs) were purchased and cultured in supplemented endothelial growth media (Clonetics). Ovine endocardial cells (OECs) were isolated from sheep cardiac ventricles by treatment with collagenase (45 minutes at 37°C). DiI-acetylated LDL uptake and expression of endothelial markers (vascular endothelial [VE]-cadherin and plate- let-endothelial cell adhesion molecule [PECAM-1]) were confirmed. OECs were maintained in Waymouth’s media (Gibco) with 10% FBS and antibiotics. Gene Transfer Endothelial cells were transduced using the retroviral vector MSCV- IRES-YFP (MIY) (gift from R.K. Humphries, British Columbia Cancer Agency, Vancouver, BC). cDNA constructs encoding the C-terminal HA-tagged Notch4 intracellular domain, Notch1 intracel- lular domain (gift of S. Artavanis-Tsakonas, Harvard Medical School, Charlestown, Mass), and full-length Jagged1 were cloned into MIY. Endothelial cells were transduced as previously described.25 Transmission Electron Microscopy Endothelial cultures were processed as previously described.26 Briefly, cultures were washed with M199, fixed in 2.5% glutaralde- hyde/2% paraformaldehyde in 0.2 mol/L sodium cacodylate buffer for 1 hour, post-fixed in 1% OsO4 for 1 hour, stained en bloc with uranyl magnesium acetate, dehydrated, and embedded in Epon- Araldite. Blocks cut from the embedded cultures were reembedded for cross-sectioning. Thin sections were stained with uranyl acetate and lead citrate and viewed on a Zeiss EM 10 microscope. Immunoblotting Cells were lysed, and 50 g of total protein was analyzed by SDS-PAGE. The monoclonal antibody against the HA epitope was purchased from BAbCo, and anti–VE-cadherin, anti–PECAM-1, anti-Tie1, and anti-Tie2 antibodies were all from Santa Cruz Bio- technology. Anti-endothelial NO synthase (eNOS/NOS type III) and anti-fibronectin antibodies were purchased from Transduction Lab- oratories, and anti–-smooth muscle actin (SMA) antibody was obtained from Cymbus Biotechnology (Hampshire, UK). Anti- phospho-Smad2 and anti-total-Smad2 antibodies (gift from N. Kha- lil, University of British Columbia, Vancouver, BC) were manufac- tured by UBI. Migration Assay The ability of endothelial cells to migrate toward platelet-derived growth factor (PDGF)-BB (20 ng/mL) was measured by a modified Boyden chamber assay as previously described.9 Immunocytochemistry Cells were fixed in 4% paraformaldehyde and blocked/permeablized in 4% FCS/0.2% Triton X-100/PBS. Secondary antibodies were Alexa 594-conjugated. To quantitate the proportion of SMA-positive cells, at least five high-power fields (comprising at least 200 cells) were evaluated, and the proportion of SMA-positive cells expressed were as a percentage of the total number of DAPI-stained nuclei in each field. Flow Cytometry Cells were trypsinized, washed in PBS, and fixed in 4% paraformal- dehyde. Cells were blocked/permeabilized, stained with anti-SMA antibody, and analyzed on an EPICS ELITE-ESP flow cytometer (Beckman Coulter). Luciferase Assays Endothelial cells were transfected by electroporation (1.5106 cells) with 1 to 5 g of plasmid DNA as previously described.27 Forty- eight hours after transfection, dual-luciferase reporter assays were performed according to manufacturer’s recommendations (Promega Corporation). The CBF1-dependent reporter, 4xCBF1wt-LUC, was a gift from S.D. Hayward (Johns Hopkins School of Medicine, Baltimore, Md).28 The SMA promoter-reporter construct encompass- ing a 5.4-kb region comprising 2555 to 2813 of the rat SMA gene was a gift from F. Dandre and G.K. Owens (University of Virginia, Charlottesville, Va). Transfections were normalized by transfecting cells with 50 ng of the Renilla luciferase plasmid pRL-CMV (Promega). In Situ Hybridization The murine Jagged1 probe was a gift from S.E. Egan (Hospital for Sick Children, Toronto, Ontario), and the Notch1 and Notch4 probes were a gift from J. Rossant (Samuel Lunenfeld Research Institute, Toronto, Ontario). For whole-mount in situ hybridization, embryonic hearts were fixed overnight at 4°C in 4% paraformaldehyde in PBS, dehydrated in methanol, and stored at 20°C. For hybridizations, embryonic hearts were processed as described.29 Reverse Transcriptase–Polymerase Chain Reaction Total RNA was isolated using TRIzol Reagent (Invitrogen), DNase- treated, and reverse-transcribed to cDNA, followed by polymerase chain reaction (PCR) (see the online data supplement for primers and annealing conditions, available at http://circres.ahajournals.org). A control reaction, omitting reverse transcriptase (RT), was performed for each RNA sample to verify the absence of genomic DNA. Results Activated Notch Induces Endothelial-to- Mesenchymal Transformation During the course of a previous study, we noted that HMECs expressing Notch4IC (HMEC-Notch4IC) lost the character- istic cobblestone morphology of confluent endothelial cells.9 As seen in Figure 1A, HMEC-Notch4IC formed multilayered cultures suggesting loss of endothelial phenotype and poten- tial transformation to mesenchymal cells. Transmission elec- tron microscopy confirmed that HMEC-Notch4IC failed to form monolayers and proper cell-to-cell junctions and Noseda et al Notch Induces Mesenchymal Transformation 911 by on April 28, 2010 circres.ahajournals.orgDownloaded from showed marked overlapping with infrequent, rudimentary cell contacts (Figures 1B, panel i, and 1B, panel ii). In contrast, HMEC-vector cells retained their capacity to form junctions (Figure 1B, panel iii). VE-cadherin is a cell adhesion molecule that is localized to the interendothelial region and is required for the formation of adherens junctions.30 Immunoblotting for VE-cadherin demonstrated significant downregulation of this critical junc- tional molecule, suggesting mesenchymal transformation (Figure 1C). Furthermore, we noted a reduction in expression, to varying degrees, of several other endothelial-specific proteins (PECAM-1, Tie1, Tie2, and endothelial NOS) (Fig- ure 1C). In addition to the loss of endothelial phenotype, EMT implies the acquisition of mesenchymal markers.19,31–35 To determine whether Notch4-activated cells upregulate mes- enchymal proteins, we examined expression of SMA, fi- bronectin, and PDGF receptors in HMEC-Notch4IC. Immu- noblotting shows induction of all three proteins in HMEC- Notch4IC (Figure 1C). PDGF is a known chemotactic factor for mesenchymal cells, and in particular, PDGF-BB plays a role in recruitment of mesenchymal cells during vascular development.35,36 Hence, we examined the chemotactic response of HMEC- Notch4IC or HMEC-vector to PDGF-BB. HMEC-Notch4IC was able to migrate toward PDGF-BB in a modified Boyden chamber assay, whereas vector-transduced cells did not (Figure 1D). Thus, Notch4 activation in endothelial cells induces morphological, phenotypic, and functional changes observed during EMT. Several studies have shown that aortic endothelium can differentiate into mesenchymal-like cells in vitro.31,37 More recent work has demonstrated that HUVECs also retain the potential to differentiate into mesenchymal cells.19,38 In ad- dition to Notch4, Notch1 is also expressed in endothelial cells.14 Hence, we transduced HMECs, HAECs, and HUVECs with Notch4IC, Notch1IC, or empty vector. Our data demonstrated that activated Notch1, as well as Notch4, had the potential to induce EMT in endothelial cells from different vascular beds, as determined by morphological, immunophenotypic, and functional criteria (online Figure 1 and data not shown). Because of the critical requirement for EMT during cardiac cushion formation, we determined whether activated Notch was also able to induce EMT in cardiac endothelial cells. OECs were transduced with the empty vector, Notch4IC, or Notch1IC. OECs also underwent a morphological transformation, as witnessed by loss of uniform cell shape, loss of intercellular contacts, cellular polarization, formation of filopodia (Figure 2A), and the induction of SMA (Figure 2B). Downregulation of endocar- dial proteins such as PECAM-1 was also confirmed (data not shown). To determine the efficiency of Notch-induced EMT in different endothelial types, we quantitated SMA-positive cells by immunofluorescent staining 10 to 14 days after transduction. HAECs and OECs demonstrated the greatest capacity to undergo Notch-induced EMT (Figure 2C). Inter- estingly, in all endothelial types, Notch1 showed greater efficacy in inducing EMT. Taken together, these results indicate that activated Notch is able to induce mesenchymal transformation in endothelial cells from various vascular beds and suggest that different Notch members may play similar functional roles in EMT. Notch-Induced EMT Is Restricted to Cells Expressing Activated Notch Activated Notch may induce EMT in conjunction with activation of other signaling molecules. In this regard, trans- forming growth factor- (TGF-) has been suggested to play an important role in EMT in the endocardial cushion, as well as in ovine and human valvular and bovine aortic endothelial Figure 1. Activated Notch4 induces EMT. A, Phase-contrast micrographs of HMECs transduced with either the empty vector or Notch4IC. B, Transmission electron micrographs of cultures in A. HMEC-Notch4IC are arrayed in overlapping cell layers (i, magnification 51 000, bar1 m) with no evidence of intercel- lular junctions between adjacent cells (ii, magnification 18 000, bar2 m). HMEC-vector (iii, magnification 61 000, bar0.75 m) cells are focally apposed and form junctional complexes (arrows). C, Immunoblots probed for endothelial (E) and mesen- chymal (M) markers on cell lysates harvested from HMEC-vector or HMEC-Notch4IC. D, Migration of HMECs transduced with either th
UBC Theses and Dissertations
The role of Notch signaling in vascular development and homeostasis Chang, Linda Ya-ting 2010
Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.
- 24-ubc_2010_fall_chang_linda.pdf [ 24.13MB ]
- JSON: 24-1.0071086.json
- JSON-LD: 24-1.0071086-ld.json
- RDF/XML (Pretty): 24-1.0071086-rdf.xml
- RDF/JSON: 24-1.0071086-rdf.json
- Turtle: 24-1.0071086-turtle.txt
- N-Triples: 24-1.0071086-rdf-ntriples.txt
- Original Record: 24-1.0071086-source.json
- Full Text