THE TIE2 RTK: REGULATION AND DOWNSTREAM

SIGNALING

by

Celina Marie Sturk

A thesis submitted in conformity with the requirements

for the degree of Doctor of Philosophy

Graduate Department of Medical Biophysics

University of Toronto

© Copyright by Celina Marie Sturk 2009

THE TIE2 RTK: REGULATION AND DOWNSTREAM SIGNALING

Celina Marie Sturk Doctor of Philosophy, 2009

Department of Medical Biophysics University of Toronto

ABSTRACT

Tie2 is a receptor tyrosine kinase (RTK) involved in numerous aspects of both

normal and pathological angiogenesis. Proper functioning of this receptor is essential for

normal development of the vasculature in the embryo as well as vessel maintenance and

at sites of active angiogenesis in the adult. A growing list of pathological states has been

attributed to a disruption of the angiogenic ‘balance’ including psoriasis, arthritis,

atherosclerosis and diabetic retinopathy. Elucidating the molecular mechanisms behind

this important biological process will provide insight into the various molecules involved

as well as provide potential targets for novel angiogenic therapies.

In an attempt to better understand the signaling pathways downstream of the Tie2

receptor we have studied tyrosine residues on the receptor believed to play an important

role in Tie2 function. Of these, we have identified Y1111 as a negative regulatory site on

Tie2. Mutation of this site affects receptor phosphorylation and kinase activity.

Furthermore, protease digestion studies indicate that mutation of Y1111 may alter

receptor conformation and potentially relieve negative inhibition imparted by the C-tail

of Tie2.

As well, we examined potential Tie2 downstream binding partners, specifically

the novel Grb7 family of proteins. This work describes for the first time tyrosine

phosphorylation of Grb14, an adaptor molecule previously shown to bind Tie2 in vitro.

ii

Moreover, our data suggests a role for this adaptor in Tie2 signal transduction involving

two tyrosine residues in the receptor C-terminal tail; Y1100 and Y1106.

These studies provide important insight into both signal transduction downstream

of Tie2 as well as help us understand some of the molecular mechanisms behind the

intrinsic ability of this RTK to regulate its own activity.

iii

ACKNOWLEDGEMENTS

I would first like to thank my supervisor Dr. Dan Dumont for his expertise and

guidance over the years and especially for his time and dedication in the final stages of

my degree. I would also like to thank my committee members, Dr. Jane McGlade and

Dr. Jorge Filmus for their knowledge and advice throughout this process and Sue Santillo

for her administrative assistance.

I am also indebted to members of the Dumont Lab, past and present, as well as

numerous other graduate students at Sunny-B whose help, technical assistance and

camaraderie have proved invaluable. A special thank you to Nina Jones and Zubin

Master for ‘showing me the ropes’ in the early days and to Harold Kim and Paul Van

Slyke for their friendship and advice.

Of course there are so many people outside the lab without whom this journey

would not have been possible. To my entire family who have always shown me

unending support and love; especially to my parents for teaching me perseverance and

hard work. I couldn’t have done this without your encouragement and endless help.

Thank you to ‘Les femmes’, Isabelle and Amandine, for listening, commiserating and

making me laugh throughout all of life’s adventures.

Most importantly, thank you to my husband Dean Rutty for his unconditional

friendship, love and support and for being the ‘rock I cling to in the storm’. Also to my

daughter Madeline for the unending joy she brings to each and every day and to my little

boy who has been so patient before making his grand entrance into this world.

iv

TABLE OF CONTENTS

ABSTRACT II ACKNOWLEDGEMENTS IV TABLE OF CONTENTS V LIST OF ABBREVIATIONS VII LIST OF FIGURES XI LIST OF APPENDICES XII CHAPTER 1: GENERAL INTRODUCTION 1

1.1 VASCULAR DEVELOPMENT 2 1.1.1 Vasculogenesis 2 1.1.2 Angiogenesis 4

1.2 RECEPTOR TYROSINE KINASES (RTKS) 6 1.2.1 Structure 6 1.2.2 Activation 7 1.2.3 Regulation 8 1.2.4 Protein Binding Domains 11

SH2 Domain 12 PTB Domain 12 SH3 Domain 13 PH Domain 13

1.2.5 Intracellular signaling pathways 14 Enzymes 14 Adaptor proteins 15 Docking proteins 15

1.3 THE TIE RECEPTORS AND THE ANGIOPOIETINS 16 1.3.1 Structure 17 1.3.2 Expression 19 1.3.3 In Development 20 1.3.4 In the Adult 22 1.3.5 In Pathology 24

1.4 ANG/TIE SIGNAL TRANSDUCTION 26 1.4.1 Receptor Activation and Phosphorylation 26 1.4.2 Signal Transduction via Tie1 27 1.4.3 Cell survival 30 1.4.4 Cell migration 31 1.4.5 Proliferation 32 1.4.6 Tie2 context and signaling 33 1.4.7 Regulation of Tie2 activity 34

Receptor internalization 36 Inhibition by the C-terminal tail 36 Regulation by phosphatases 36

1.5 GROWTH FACTOR RECEPTOR BOUND PROTEINS (GRBS) 37 1.5.1 Structure 37

Proline Rich Region (PRR) 38 GM Region 40 RA domain 40 PH domain 40 BPS domain 41 SH2 domain 42

1.5.2 Biological role 42 1.5.3 Phosphorylation 44

v

CHAPTER 2: A NEGATIVE REGULATORY ROLE FOR Y1111 ON THE TIE2 RTK 47 2.1 ABSTRACT 48 2.2 INTRODUCTION 49 2.3 MATERIALS AND METHODS 52 2.4 RESULTS 55 2.5 DISCUSSION 71

CHAPTER 3: GRB 14 TYROSINE PHOSPHORYLATION IN TIE2 SIGNAL TRANSDUCTION78 3.1 ABSTRACT 79 3.2 INTRODUCTION 80 3.3 MATERIALS AND METHODS 83 3.4 RESULTS 86 3.5 DISCUSSION 95

CHAPTER 4: DISCUSSION AND FUTURE DIRECTIONS 99 4.1 NEGATIVE REGULATION OF TIE2 100 4.2 GRB PROTEINS AND TIE2 SIGNALING 106

4.2.1 Grb7 family phosphorylation 107 4.2.2 Grb/Tie2 interactions 110

4.3 CONCLUDING REMARKS 114 REFERENCES 116 APPENDICES 128

vi

LIST OF ABBREVIATIONS

°C Degree(s) Celsius

ABIN-2 A20 binding inhibitor of NF-kappaB activation-2

Ang angiopoietin

ATP adenosine 5’-triphosphate

BPS between PH and SH2

BSA bovine serum albumin

cDNA complementary DNA

CDR cysteinee-rich domain

CO2 carbon dioxide

COMP cartalige oligomeric matrix protein

CORT cloning of receptor targets

D aspartate

DMEM Dulbecco’s modified Eagle’s medium

DNA deoxyribonucleic acid

DOK downstream of tyrosine kinase

E glutamate

E. coli Escherichia coli

ECM extracellular matrix

EDTA ethylenediaminetetraacetic acid

EGF epidermal growth factor

EGFD epidermal growth factor-like domain

vii

EGTA ethylene glycol-bis (ß-aminoethyl ether) N, N, N’, N’ – tetracetdic acid

F phenylalanine

FAK focal adhesion kinase

FBS Fetal bovine serum

FGF fibroblast growth factor

FNIII fibronectin type III repeats

GF growth factor

GM Grb and Mig

Grb growth factor receptor bound protein

GST glutathione S-transferase

HEK human embryonic kidney

HEPES N-(2-hydroxyethyl) piperazine-N'-(2-ethanesulfonic acid)

HRP horseradish peroxidase

HSC hematopoietic stem cell

HUVEC human umbilical vascular endothelial cell

IAP inhibitor of apoptosis protein

IgD immunoglobulin-like domain

IGFR insulin growth factor receptor

IR insulin receptor

IRK insulin receptor kinase

IRS insulin receptor substrate

IUP intrinsically unstructured proteins

K lysine

viii

kDa kilodalton

M molar

MAPK mitogen activated protein kinase

mg milligram

MgCl2 magnesium chloride

Ml milliliter

mM millimolar

NaCl Sodium chloride

ng nanogram

NLS nuclear localization signal

PARP poly(ADP-ribose) polymerase

PBS phosphate buffered saline

PCR polymerase chain reaction

PDGF platelet-derived growth factor

PH pleckstrin homology

PI3K phosphatidylinositol 3’-kinase

PIR phosphorylated IR-interacting region

PKC protein kinase C

PRR proline rich region

PTB phosphotyrosine binding

PTP protein tyrosine phosphatase

pY phosphotyrosine

R arginine

ix

RA ras associating

RasGAP p21ras GTPase-activating protein

RTK receptor tyrosine kinase

SDS-PAGE sodium dodecyl sulphate-polyacrylamide gel electrophoresis

SH src homology

Shc src homolog and collagen homolog

Shp2 SH2 domain-containing tyrosine phosphatase 2

TBST tris buffered saline with Tween 20

Tek tunica interna endothelial cell kinase

Tie tyrosine kinase with immunoglogulin-like loops and epidermal growth factor

homology domains

VEGF vascular endothelial growth factor

VEGFR vascular endothelial growth factor receptor

VE-PTP vascular endothelial protein tyrosine phophatase

W tryptophan

WCL Whole cell lysate

WT wild type

Y tyrosine

μg microgram

μl microlitre

x

LIST OF FIGURES Figure 1.1 Vascular development

Figure 1.2 RTK activation

Figure 1.3 Structure of the Tie and Angiopoietin molecules

Figure 1.4 Tie2 mediated signal transduction

Figure 1.5 Grb7 family structure

Figure 2.1 Schematic representation of Tie2 point mutants

Figure 2.2 Phosphorylation of wild type Tie2 and Tie2 tyrosine to phenylalanine

point mutants

Figure 2.3 Mutation of Y1111 to Phenylalanine results in increased receptor kinase

activity

Figure 2.4 Mutation of Y1111 to Phenylalanine enhances downstream signaling

Figure 2.5 Tyrosine 1111 is important for structural negative regulation of the Tie-2

receptor

Figure 2.6 Mutation of Y1111 to aspartate (D) or glutamate (E) results in increased

receptor activity

Figure 2.7 Mutation of Y1111 on Tie2 alters its protease susceptibility

Figure 2.8 Schematic depicting Tie2 kinase domain and C-terminal tail

Figure 3.1 Grb14 is tyrosine phosphorylated in the presence of Tie2

Figure 3.2 Y1100 and Y1106 on Tie2 are important for Grb14 tyrosine

phosphorylation

Figure 3.3 The SH2 domain is important for Grb14 tyrosine phosphorylation

Figure 3.4 Grb14 in endothelial cells

xi

LIST OF APPENDICES

Appendix 1 Effect of pervanadate on Tie2 phosphorylation

Appendix 2 Shp2 phosphorylation in Tie2 signal transduction

Appendix 3 Grb14 identified by silver staining

Appendix 4 Grb7 SH2 domain plays a role in binding of pp85 and pp70

Appendix 5 Grb14 SH2 domain required for Tie2 mediated phosphorylation

Appendix 6 Mutation of Y1111 and receptor ubiquitination

Appendix 7 Mutation of Y1111 to Phenylalanine does not affect membrane receptor

levels

xii

CHAPTER 1: GENERAL INTRODUCTION

1

1.1 Vascular Development

During development of vertebrates, the cardiovascular system is the first organ

system to develop. This system encompasses the heart along with both blood and blood

vessels and is essential for supplying tissues and organs with nutrients and oxygen, and to

allow for adequate removal of metabolic waste. This system also functions as an

extensive communication 'highway' between distant organs and tissues. Proper

development and functioning of the vascular system, therefore, must be a tightly

regulated process and involves a large number of carefully regulated events. Often the

development of the vascular network is divided into two separate events: vasculogenesis,

that is the de novo formation of a primitive vascular network made up of essentially

uniform sized vessels, and angiogenesis, the remodeling of this initial vasculature into the

more complex network of vessels required in the mature organism (Figure 1.1).

1.1.1 Vasculogenesis

Early on in mammalian embryonic development, the process of gastrulation gives

rise to the three embryonic germ layers: endoderm, ectoderm and mesoderm. Of these, it

is the mesoderm layer that will go on to give rise to the vascular system. The first vessels

arise in the yolk-sac with the formation of blood islands, and the emergence of what is

believed to be a common hematopoietic and endothelial precursor, the hemangioblast 1.

In the embryo proper, vessel formation can occur in the absence of hematopoiesis,

suggesting that these cells are precursors unique to the endothelial lineage, called

angioblasts 2. In either case, it is the aggregation of the

2

Figure 1.1: Vascular development. Schmatic representation of the major stages of vascular development as discussed in the text.

3

angioblasts into tube like structures that gives rise to the primitive vascular network. The

differentiation of angioblast cells from mesoderm and subsequent arrangement of these

angioblasts into a primitive vascular plexus are collectively referred to as vasculogenesis

(reviewed in Risau and Flamme, 1995) 3.

1.1.2 Angiogenesis

Once a primary vascular plexus has been laid down, this immature and poorly

functioning network must be remodeled into the complex vasculature seen in the adult.

This occurs via the process of angiogenesis, originally described as the sprouting of new

vessels from the pre-existing, larger ones formed during vasculogenesis. Today,

however, the term angiogenesis is used in a more general sense to include the pruning

and remodeling of this primitive network including the increase in luminal size or repair

of a blood vessel by intercalated growth and the division of larger vessels into smaller

ones by intussusceptive growth (reviewed in Djonov et al., 2003) 4.

The best studied of these mechanisms, by far, is sprouting angiogenesis. This

process can be divided into a number of different stages and is initiated in response to

surrounding vascular endothelial growth factor (VEGF) gradients (see below) (reviewed

in Augustin et al., 2009) 5. Vasodilation and increased vessel permeability thereby allows

for extravasation of plasma proteins into the extravascular space, laying down a matrix

for the sprouting endothelial cells. Next, in order for cells to migrate into the

extracellular space, support cells (pericytes) must be loosened and attachment of

endothelial cells to the surrounding matrix must be disrupted. Enzymatic degradation of

the basement membrane subsequently clears a path for proliferating endothelial cells,

4

which loosely associate in a column that protrudes into the extravascular space. A model

has been proposed whereby invading ‘tip’ endothelial cells bearing numerous filopodia

are followed by a zone of proliferating and differentiating endothelial cells termed ‘stalk’

cells. Finally, these are proceeded by the quiescent ‘phalanx’ cells which are in constant

contact with pericytes and smooth muscle cells (reviewed in Augustin et al., 2009) 5.

Endothelial tubules usually develop initially as cords lacking a lumen, and

subsequent lumen formation is a process that is not well understood. However, mature

blood vessels do require recruitment of support cells, such as pericytes (for small vessels)

and smooth muscle cells (for large vessels) and this process helps stabilize the new vessel

and plays a role in inhibiting continued endothelial cell proliferation. The support layer

also provides endothelial survival signals to protect against vascular regression 6. For

larger arteries, the addition of a thick muscularized coat confers viscoelastic and

vasomotor properties. A new basal lamina is also formed around the vessel for support.

The final stages of new capillary growth, where vessels fuse to form closed loops and

deliver circulation to newly vascularized areas, are poorly understood.

Although it is convenient to divide vascular development into the two distinct

steps of vasculogenesis and angiogenesis, the complexity of the adult vasculature makes

us realize that things are not quite so simple. First, vessel growth must be spatially and

temporally regulated such that proliferation and regression is occurring simultaneously in

different locations of the body. Sites of endothelial cell migration and branching cannot

be random, but rather must follow a specific pattern in order to ensure proper

vascularization of all tissues of the organism. Additional considerations such as whether

or not to form capilleries or large vessels, arteries or veins, lymphatic or blood vessels

5

etc.; all of this must be taken into account and reminds us that vascular growth and

remodeling must involve the coordination of numerous diverse signaling pathways.

1.2 Receptor Tyrosine Kinases (RTKs)

1.2.1 Structure

In order to coordinate the activity of cells within a multicellular organism,

metazoans have evolved a variety of signaling networks involving cell surface protein

receptors which serve to transduce signasl from the cells extracellular environment to the

inside of the cell. Receptor Tyrosine Kinases (RTKs) are a large family of such

receptors; they span the cell membrane and are able to phosphorylate both themselves

and other cytoplasmic proteins on tyrosine residues, thereby initiating intricate

intracellular signal transduction cascades. These signaling pathways serve to modulate a

host of cell responses including cell proliferation, survival, differentiation and migration.

RTKs have been divided into at least 20 different families based on characteristic

structural features, often residing within the unique pattern of sequence motifs present in

their extracellular ligand binding domain. These include (among others) fibronectin type

III repeats (FNIII), immunoglobulin-like domain (IgD), epidermal growth factor-like

domain (EGFD) and cysteinee-rich domain (CRD) (reviewed in Hubbard and Till, 2000)

7. Other structural features include a hydrophobic helical transmembrane domain and an

intracellular domain which encompasses regulatory sequences and a conserved kinase

domain.

6

1.2.2 Activation

In the absence of ligand, RTKs are typically present at the cell surface in either

monomeric or dimeric form, but are maintained in an inactive state 8. Receptor activation

is initiated by binding of the cognate ligand to the extracellular domain. This results in

receptor conformational changes and/or clustering of receptor monomers into various

higher order multimers (reviewed in Hubbard and Till, 2000 and Schlessinger et al, 2000)

7,9. While ligand binding was initially believed to create receptor dimers, numerous

structural studies have since highlighted the diversity that exists in the number of receptor

subunits assembled for full receptor activation. Ultimately, however, it is believed that

ligand binding stabilizes receptor dimeric/multimeric form thereby promoting

transphosphorylation of tyrosine in the cytoplasmic domain 10. These often include

tyrosines located in the receptor activation loop whose phosphorylation contributes to

activation of receptor kinase activity. Tyrosines located in both juxtamembrane and C-

terminal tail regions can also become phosphorylated and serve as binding sites for

downstream signaling molecules containing specific phosphotyrosine binding motifs,

including src homology 2 (SH2) and phosphotyrosine binding (PTB) domains.

The exact mechanism surrounding how ligand binding increases receptor catalytic

activity is not well understood and is likely to differ among receptor subclasses. One

possibility is that ligand binding simply increases the concentration of receptor subunits

in a given area thereby facilitating the transphosphorylation events. It has also been

suggested that the stabilized direct interaction between receptor cytoplasmic domains

may be stimulatory for catalytic activity (reviewed in Hubbard, 2004) 8. For some

7

receptors, ligand binding may result in receptor conformational changes resulting in

increase kinase activity. For example, the epidermal growth factor (EGF) receptors have

been shown to form non-covalent dimers at the cell surface 11,12 and ligand binding

induces rotation of the transmembrane helices and thus realignment of the cytoplasmic

domains 12,13.

A key structural feature of RTKs is the activation loop/segment located within the

receptor kinase domain. This segment, which is approximately 20-25 residues long and

typically contains 1-3 tyrosine residues, has been shown to be essential for receptor

kinase activation. In an unphosphorylated state, the activation loop is not positioned

correctly for phosphoryl transfer 7. However, in the ligand-stabilized dimer, there exists

a low level basal kinase activity which is sufficient for transphosphorylation to occur.

Phosphorylation of the activation loop results in conformational changes which leave this

segment in a catalytically competent state 14 and typically results in overall enhancement

of receptor catalytic activity.

1.2.3 Regulation

Receptor activation can also be influenced by regions other than the kinase

domain. Specifically, both the juxtamembrane and C-terminal tail regions have been

implicated in inhibition of catalytic activity in a subset of RTKs (reviewed in Hubbard,

2004) 8 (Figure 1.2). This additional level of control over receptor activation is believed

to help prevent against naturally occurring mutations which produce receptors with

increased kinase activity. Many of these mutations do not appear to affect receptor

oligomerization, but rather involve inhibitory regions outside the kinase domain.

8

The juxtamembrane and C-terminal tail regions vary in length and composition

between RTKs. As mentioned above, they also typically contain tyrosine residues for

binding SH2 and PTB domain containing signaling proteins and a growing body of

evidence suggests that these tyrosine residues are also involved in receptor autoinhibition.

However, the way in which these regions exert their inhibitory effects appears to vary

between receptor subfamilies. For example, in the case of the ephrin (Eph) family of

receptors, the juxtamembrane region interacts with the kinase domain, and a tyrosine

residue within the juxtamembrane region, in the case of EphB2, (Tyr610) interferes with

salt bridge formation necessary for positioning of ATP and thus phosphoryl transfer 15.

The juxtamembrane region in the platelet derived growth factor (PDGF) receptors,

however, inhibits the activation segment in the kinase domain from adopting an ‘active’

conformation 16. Structural studies of the Tie2 receptor suggest that in an inactive state,

the C-terminal region of the receptor may interfere with substrate binding in the kinase

domain 17.

The prevalence of naturally occurring mutations in RTKs that lead to

constitutively active receptors and consequently various disease states (including cancers)

highlight the importance of maintaining these receptors at a low level of basal catalytic

activity.

9

C

N

Figure 1.2: (adapted from Hubbard, 2004) A) In the absence of ligand, the receptor tyrosine kinase domain of an RTK (green) remains in a basal low-activity state through inhibitory interactions imparted by the juxtamembrane region (orange) and/or the carboxyl terminal region (black) with the kinase domain. In addition, the activation domain (pink loop) is not optimally positioned for catalysis. B) Following ligand (large pink circles) binding and receptor multimerization, the cytoplasmic domains are positioned in a manner to allow transphosphorylation of tyrosine residues (small brown circles) in the juxtamembrane and C-terminal regions as well as the activation segment. C) After phosphorylation, inhibitory segments reconfigure and the kinase domains become fully active (red) creating a subset of phosphorylated tyrosine residues able to bind SH2 and PTB domain containing proteins.

A B C

10

1.3 Elements of Signal Transduction Pathways

Once RTKs have been activated, phosphorylation on receptor intracellular

tyrosine residues initiates sites for binding of signaling proteins and initiates downstream

signaling cascades. RTK signaling specificity is critical for proper transduction of a

particular signal. Since many protein kinases have somewhat broad substrate specificity,

it is important for these molecules to have evolved a way to organize this activation into a

distinct subset of signals. This is accomplished by using intracellular protein interactions

to form multiprotein signaling complexes. Many of these complexes are established by

non-catalytic motifs that mediate sequence specific protein-protein or protein-lipid

interactions.

1.2.4 Protein Binding Domains

Protein domains are able to fold independently to bring together their N and C

terminal segments and in doing so expose a ligand binding region (reviewed in Pawson

and Nash, 2003) 18. This region is often a short contiguous sequence, often no more than

10 amino acids in length. Furthermore, when isolated from the rest of the polypeptide,

these motifs retain their functional binding properties. It is possible for a given protein to

have more than one of such domains allowing for the formation of multiprotein

complexes. They also serve to localize proteins to a specific cellular location, recognize

protein post-translational modifications, link non-catalytic proteins to enzymes and alter

protein conformation and/or catalytic activity and substrate specificity.

11

SH2 Domain

This conserved protein domain was originally identified as a non-catalytic region

of Src family proteins 199. Approximately 100 amino acids in length, SH2 domains

recognize short phopho-tyrosine containing sequences thereby modulating

phosphorylation dependent interactions. While conserved residues of the SH2 domain

allows for interaction with phosphotyrosine, binding specificity is often determined by

the 3 or 4 amino acids (+1, +2 etc.) following the phosphotyrosine moiety (reviewed in

Kuriyan and Cowburn, 1997) 20. Because of their specificity for phosphotyrosine,

proteins harboring SH2 domains are often involved in signaling pathways mediated by

RTKs allowing these receptors to recruit distinct sets of proteins to their intracellular

domain.

A surprising discovery, in a handful of cases, SH2 domains have been shown to

bind to sequences in a non phosphotyrosine dependent manner. These include the SAP

SH2 domain which interacts with SLAM 21 and the Grb10 SH2 domain when interacting

with Nedd4 22.

PTB Domain

Phosphotyrosine binding (PTB) domains were first identified in the adaptor

protein Shc 23. While structurally unrelated to the SH2 domain, PTB domains also

recognize phosphotyrosine containing sequences as their name would imply. Unlike the

SH2 domain, however, binding specificity is conferred by residues N-terminal, not C-

terminal, to the phosphotyrosine on target sequences 24. Also unique to the PTB domain

is the lack of sequence homology between PTB domains of different proteins (reviewed

12

in Forman-Kay and Pawson, 1999) 25. Interestingly, not all PTB domains require

phosphorylation of tyrosine for binding, as their name would imply. Examples of

proteins whose PTB domain binds non-phosphorylated sequences include X11 26, Fe65

(Borg, 1996), Disabled (Dab) (Howell, 1999) and Numb (Dho, 1998).

SH3 Domain

Src homology 3 (SH3) domains are approximately 60 amino acids in length.

Structurally composed of two perpendicular beta-sheets, the hydrophobic ligand binding

domain recognizes short non-phosphorylated sequences rich in proline residues 27. Often

the recognition sequence is a variation of a PXXP motif, however there have been cases

where this is not the situation (reviewed in Pawson and Nash, 2003)18. As with many

other protein domains, SH3 domains often serve to assemble multiprotein signaling

complexes.

PH Domain

While structurally similar to PTB domains, pleckstrin homology (PH) domains

associate with charged polar headgroups of phosphoinositides 28,54. Some PH domains

bind with a particular specificity to phosphoinositides such as phosphatidyl-inositol (PI) -

4,5-bisphosphate or PI-3,4,5-P3. Structurally, it appears that a positively charged region

on the face of the PH domain allows it to interact with negatively charged phosphate

groups of the phosphatidylinositides. These domains of approximately 120 amino acids

in length are believed to function, at least some of the time, to sequester proteins at inner

surface of cell membrane (ie. in the vicinity of a receptor).

13

While the above mentioned protein modules are a few of the better known, it is

important to acknowledge that there exist many other domains (over 40 identified) which

are not mentioned here (reviewed in Pawson and Nash, 2003) 18.

1.2.5 Intracellular signaling pathways

The unique set of proteins recruited to the receptor upon receptor activation is key

in determining the biological specificity of that ligand/receptor complex and link the

receptor to various downstream intracellular signaling cascades.

Enzymes

Kinases and phosphatases are intracellular proteins that harbor intrinsic enzymatic

activity to either add (kinase) or remove (phosphatase) phosphate moieties on target

molecules. In response to a signal, these proteins are organized in various combinations

to elicit specific biological responses. The activity of these enzymes can be altered by the

protein’s phosphorylation state or allosteric changes in the protein. For example, an

important regulator of cell survival is phosphatidylinositol 3’-kinase (PI3K) which is

made up of two subunits: a p85 adaptor subunit and a p110 catalytic subunit. PI3K is

activated upon binding of the p85 subunit to phosphotyrosine which in turn elicits a

conformational change in the p110 subunit 29. Similarily, the protein tyrosine

phosphatase (PTP) Shp2 undergoes physical changes which promote activation of its

phosphatase activity. Shp2 protein is made up of two tandem SH2 domains followed by

a C-terminal PTP domain. In the absence of stimulation, Shp2 is maintained in a low-

14

activity state through interactions between its N-terminal SH2 domain and its PTP

domain. Activation of cell surface receptors creates binding sites for the Shp2 SH2

domains either directly or via scaffolding proteins and occupation of the Shp2 N-terminal

SH2 allows for conformational changes which relieve its self inhibition (reviewed in

Ostman et al., 2006) 30.

Adaptor proteins

Various enzymatic proteins must be organized into specific signaling pathways in

order to carry out the appropriate message. In RTK signaling, receptor activation leads to

creation of phospho-Y sites to bind SH2 or PTB containing molecules. Some of these

recruited proteins lack catalytic activity altogether and are made up entirely of interaction

regions such as SH2 and SH3 domains. These adaptor molecules are so named because

of there role in linking protein enzymes to tyrosine kinase signaling complexes 31. Grb2,

Shc, Nck, p85 and the Grb7 family members (Grb7/10/14) are all examples of adaptor

proteins.

Docking proteins

Docking or scaffolding proteins are a type of adaptor protein which, in addition to the

above mentioned domains, often also contain numerous phosphotyrosines. These

residues serve as binding sites for other SH2 domain containing signaling molecules.

Because these proteins often contain more SH2 binding sites than the RTK itself, they

serve as a scaffold to build and amplify signaling complexes downstream of the receptor.

15

The insulin receptor substrate (IRS) family and DOK family of proteins are perhaps the

best characterized in this category of signaling molecules.

By identifying which molecules are assembled and activated in response to specific

cellular signals we can better understand the intricacies of particular signaling pathways

and the role that these pathways play in various biological situations.

1.3 The Tie Receptors and the Angiopoietins

Many of the signals involved in the initial stages of vascular development (ie.

differentiation of pluripotent embryonic cells of the mesenchym into vascular precursor

cells), remain elusive. Fibroblast growth factor (FGF), along with a handful of

transcription factors, have been implicated in these early events (reviewed in Conway et

al., 2001)32. What is clear, however, is that the vascular endothelial growth factor

(VEGF) receptors, VEGFR-1 and -2 play an important role in early vasculogenesis

following this differentiation. In fact, VEGFR-2 expression appears to be an early

marker of developing endothelial lineage during vasculogenesis and appears as early as

E8.5-10.5 in blood islands of the yolk sac 33,34. VEGFR-2 is now thought to be a marker

of a common hematopoietic/endothelial precursor cell, the hemangioblast since it is

found before there is a distinction between endothelial and hematopoietic lineages in

blood islands 34.

Once endothelial cells of the yolk sac and embryo proper begin to coalesce and

form a honey comb like structure, this vascular network is remodeled and matured

16

through the process of angiogenesis. Numerous signaling pathways have been shown to

contribute to this second stage in vascular development including the VEGF/VEGFR and

Angiopoietin/Tie2 pathways.

While the importance of VEGF/VEGFR signaling in vascular development was

well documented early on, a second family of predominantly endothelial specific

receptors was identified in 1992; the tyrosine kinase with immunoglobulin (IG) and

epidermal growth factor (EGF) (Tie) family of RTKs. Two family members make up the

Tie family, Tie1 and Tie2. The Tie2 cognitive ligands, the angiopoietins (Angs), were

subsequently identified and are made up of four members, Ang1-4. To date, all the

angiopoietins appear to bind Tie2, while only Ang1 has been shown to activate Tie1 35,36

37. Whether of not Ang1 binds Tie1 still remains to be determined (see below).

Biologically, the Tie/Angiopoietin pathway(s) have been extensively implicated in

embryonic angiogenesis as well as in regulation of vascular homeostasis and remodeling

in the adult.

1.3.1 Structure

The Tie receptors are single transmembrane receptors found almost exclusively at

the cell surface of endothelial and haematopoietic cells. The two family members, Tie1

(Tie) and Tie2 (Tek) show a high degree of structural homology including an

extracellular domain (containing three immunoglobulin ( IG)-like loops, three epidermal

growth factor (EGF) homology motifs and three fibronectin type III (FN3) repeats), a

transmembrane domain and a split tyrosine kinase domain 38 39,40 (Figure 1.3). The

intracellular domains are highly conserved displaying 76% identity 41 while the

17

Ig

EGF

FNIII

tyrosine kinasedomain

Cell membrane

Figure 1.3: Structure of the Tie and Angiopoietin molecules. The structure of the tyrosine kinase with Ig and EGF homology domains (Tie) family in (A) including the extracellular domain composed of containing three immunoglobulin-like loops (Ig), three epidermal growth factor homology motifs (EGF) and three fibronectin type III (FN3) repeats, a transmembranedomain (black line) and a split tyrosine kinase domain (blue boxes). Their cognitive ligands, the Angiopoietins, are depicted in (B) including the N-terminal superclustering domain (SCD), followed by a coiled-coil domain (CCD), a linker region, and a C-terminal fibrinogen-like domain (FLD).

Ig

A B

SCD

CCD

Linker

FLD

Monomer

Oligomer

18

extracellular domains are only 33% identical. The solving of the crystal structure of the

Tie2 extracellular domain shows that the three Ig domains along with the EGF domains

fold into a compact, arrowhead-shaped structure for ligand binding 40.

All four members of the Angiopoietin family also share a conserved structure

comprising an N-terminal region responsible for dimerization or oligomerization, a short

linker region followed by a C-terminal fibrinogen-like domain involved in receptor

mediated interactions (Figure 1.3) 42,43. Structural analysis of Ang1 and Ang2 show that

the N-terminal domain is made up of a superclustering motif containing two cysteine

residues (cysteines 41 and 52) and a coil-coil domain responsible for the formation of

higher order oligomers 44,45. Electron microscopic rotary shadowing experiments

demonstrated a heterogeneous mix of trimeric, tetrameric and pentameric oligomers exist

for both Ang1 and Ang2 and it is believed that such higher order oligomers are required

for receptor activation 40,45.

The solving of the crystal structure of Tie2 extracellular domain in complex with

Ang2 suggests that the second Ig-like loop of the receptor extracellular domain appears to

be the binding site for the angiopoietin ligans 46,47. This detailed structural analysis of

the Ang2/Tie2 interaction reveals that the ligand/receptor interaction is somewhat unique

in that two complementary surfaces interact with no domain rearrangements and little

conformational changes in either molecule 40.

1.3.2 Expression

The Tie receptors are primarily found in blood and lymphatic ECs. In the

developing mouse embryo, expression of Tie1 could be detected as early as E9.5 48and in

19

the adult Tie1 mRNA expression has been shown to be upregulated in specific situations,

such as melanoma progression 49 and at sites of turbulent blood flow 50. The Tie

receptors have also been shown to be expressed by circulating haematopoietic cells such

as megakaryocytes, monocytes, neutrophils and haematopoietic stem cells in the bone

morrow 51. Tie2 expression has been detected in a handful of other non-endothelial cell

types such as keratinocytes 52, perivascular mesenchymal cells 53,54, neurons 55, fibroblast-

like cells of choroidal neovascular membranes 56 and endometrial epithelial and stromal

cells 57. This adds further to the complexity of the role of this receptor both in

development and in the adult.

Ang1 and 2 display their own distinct patterns of expression. Ang1 is expressed

by peri-endothelial mural cells (SMCs and pericytes), fibroblast cells and a number of

tumor cells 42,58,59. Ang1 has also been shown to bind the extracellular matrix (ECM) via

its linker peptide region 60. Ang2 is expressed primarily by endothelial cells 58,61,62 63 64,

especially those involved in angiogenic events , and can be stored in Wieble-Palade

bodies (endothelial cell organelle), presumably for rapid availability under specific

conditions 65.. Ang2 expression has been shown to be induced by hypoxia, shear stress

and the presence of the VEGF ligand 63,66.

1.3.3 In Development

During development of the mouse embryo, vascular remodeling and maturation

occurs between E9.5 and E12.5. Much of what we know about the role of of the Tie-

Angiopoietin pathway(s) during this time comes from loss of function and gain of

function studies in mice.

20

Mice engineered to lack the Tie receptor die in mid to late gestation (E13.5-18.5)

and display severe hemorrhaging and edema due to vessel wall defects 67. Because major

blood vessels are intact and the defects are in the microvasculature, it suggests that this

receptor plays a role in later stages of vessel remodeling and maturation 67,68. Further

supporting these data are chimeric studies which show that Tie -/- cells are able to

contribute to major vessels formed in vasculogenesis and early angiogenesis, but are not

found in capillaries formed later in development 69.

In contrast, Tie2 KO mice die earlier, between E10.5 and E12.5, due to cardiac

and vascular defects. In the heart, there is disrupted vascularization and trabeculation

with incomplete interdigitation of endocardial and cardiac cells. Furthermore, there

appears to be reduced interaction between the endocardial and myocardial cells. Also,

while the major vessels are present in these mice, there is a lack of vascular remodeling

and a suggested lack of smooth muscle cell recruitment to vessels, indicating a role for

this receptor in later stages of vascular development and vessel maturation and

stabilization 70. A paucity of endothelial cell numbers in these mice also points to a role

for Tie2 in endothelial cell survival. This is further corroborated by in vitro studies

which also suggest that this pathway protects against cell death through an Akt mediated

pathway (see below) 71-74.

Mice engineered to lack Ang1 die by embryonic day 12.5 due to defects in

vascular remodeling reminiscent of those seen in the Tie2 KO 75. These mice display a

similar lack of association between the endothelial cells and the surrounding support

cells, once again suggesting a role for the Tie-Ang pathway in pericyte recruitment.

21

In contrast, Ang2-/- mice do not display defects in embryonic vascular

development and appear to be born quite normal 61. While this suggests Ang2 is

dispensable during vascular development, there appears to be a role for Ang2 in distinct

vascular beds in the adult (see below). Interestingly, mice engineered to overexpress

Ang2 die embryonically with defects similar to Tie2 and Ang1 null mice 76, leading to the

speculation that Ang2 may in fact be a natural antagonist of the Ang/Tie2 pathway. This

is supported by the observation that Ang2 is expressed at sites of vascular remodeling and

regression such as in the regressing corpus luteum and during atresia of ovarian follicles

76,77.

1.3.4 In the Adult

While much of the initial focus on Tie2 revolved around its role during

development, it was also noticed that this receptor was expressed at sites of active

angiogenesis in the adult such as during follicular development and wound healing.

Interestingly, follow up studies demonstrated that Tie2 was expressed in virtually all

endothelial tissue in the adult 78. The fact that the receptor was also phosphorylated in

quiescent adult endothelium supported a role for this pathway in vascular maintenance in

the adult, as well as at sites of active angiogenesis. Complementing these expression

studies, Tie2 knock out mice were used in a conditional binary system where expression

of Tie2 was shown to partially rescue embryonic lethality 79 indicating Tie2 may also

function later in development despite the fact it is initially expressed quite early in

embryogenesis. Furthermore, chimeric studies of the Tie receptors show that endothelial

cells lacking both Tie1 and Tie 2 are only able to contribute to the vasculature until E15.5

22

80. Endothelial cells doubly heterozygous for Tie1 and 2 alleles, while able to contribute

to the vasculature throughout embryonic development, are absent from adult vasculature,

again suggesting a role for Tie2 in both late embryonic and adult vasculature.

Interestingly, it appears that Tie2 may also play a role in postnatal bone marrow

hematopoiesis. Chimeric analysis has shown that Tie1/2 deficient hematopoietic stem

cells (HSCs) are able to contribute to embryonic hematopoiesis, but in the adult, are not

able to expand and/or survive in the presence of wild type (WT) cells 81.

Angiopoietin-1 has also been shown to play a role in the adult, in this case as a

modulator of vessel leakiness. Transgenic overexpression of Ang1 in the skin of mice

results in the formation of larger, more numerous and more highly branched vessels

which are more resistant to vessel leakiness caused by permeability-inducing

inflammatory agents. Furthermore, systemic delivery of Ang1 into adult mice protects

vessels against VEGF induced leakiness implicating the Tie/Ang system as a modulator

of vessel permeability 82,83.

As mentioned above, Ang2 appears to be involved in the adult at sites of vascular

remodeling. In deletion studies, although Ang2 deficient mice do not show overt signs of

disrupted vasculature formation in the embryo, these mice display abnormal postnatal

vascular development. The vascular architecture in the retina is clearly perturbed in

Ang2-/- mice and normal regression of the hyaloid vasculature in the eye does not occur

61 84. This phenotype of the eye could not be rescued by Ang1 expression, suggesting a

unique role for Ang2 in postnatal angiogenesis and vascular remodeling.

Interestingly, Ang2 deficient mice suggest a role for the angiopoietins during

lymphatic angiogenesis. In the absence of Ang2, the mice develop chylous ascites

23

postnatally and either die 14 days after birth (in a 129/J genetic background) or live into

adulthood with accumulation of lymphatic fluid in the abdominal cavity (in a C5BL/6

background) 61,85. The lymphatics in these mice lack recruitment of support cells and

display severe patterning defects. Interestingly, insertion of Ang1 into the Ang2 locus

rescues these observed lymphatic defects suggesting an overlapping agonistic role for

Ang1 and Ang2 in the lymphatic endothelium 61.

Taken together, this data suggests an important role for the Ang/Tie system in

modulating endothelial and lymphatic vascular quiescence and remodeling in the adult.

1.3.5 In Pathology

Because angiogenesis is important not only during development, but also in

normal and pathological states in the adult, a long list of disorders is characterized or

caused by either excessive or insufficient angiogenesis. These include disease states such

as psoriasis, arthritis, atherosclerosis and diabetic retinopathy to name a few (reviewed in

Carmeliet, 2003) . B86 ecause Tie2 is expressed and activated in adult tissue, it would

stand to reason that this receptor may be involved in similar pathological states.

One avenue of research has been to look at the potential role of Tie2 in tumor

growth and metastasis. Early on, Tie2 expression was documented in human breast

tumors 87. Immunohistochemical analysis has demonstrated that expression of Tie-1,

Tie2, Ang-1 and Ang-2 is elevated in some but not all tumors indicating that this

ligand/receptor system may play a role in specific tumor micorenvironments.

Consistent with its role as vascular stabilizer in normal blood vessel physiology,

overexpression of Ang1 has been shown to produce anti-tumorigenic effects 88-90. While

24

there have been a handful of contradictory reports that suggest that Ang1 has a

stimulating effect of tumour growth 89, it is generally believed that Ang1 contributes to

maturation of tumour vasculature by recruitment of pericytes.

The role of Ang2 in tumour progression is even more enigmatic as overexpression

of Ang2 in tumour cells results in either hypervascularized and increasingly invasive

tumours (resulting in an increase in tumour growth) 89-92 93 or in disrupted angiogenesis,

increased cell death and suppression of tumour growth 90,91 93 94. These discrepancies can

most likely be attributed to the ‘context’ dependent nature of Ang2 effects in various

milieus and further study is necessary to unravel many of these conditions.

Inhibition of Tie2 itself has been accomplished using a soluble extracellular

domain of the receptor which has been shown to inhibit angiogenic growth and

metastases in tumor bearing mice supporting a role for this receptor in tumor

angiogenesis 95. Subsequently, agents such as synthetic peptides, intradiabodies and

other small molecule inhibitors have been examined for potential use in blockade of the

Tie2 pathway 96-98. Identification of successful Tie2 antagonists will no doubt prove

invaluable in deciphering the role of this receptor in various pathological states.

The Ang/Tie pathway has also been implicated in the process of inflammation.

Experiments in endothelial cells have shown that Ang1 is able to inhibit expression of

adhesion molecules such as ICAM1 and VCAM1 99. In vivo, this data is supported by

the observation that Ang1 overexpression in mice shows anti-inflammatory effects 82.

Conversely, Ang2, in conjuction with TNF alpha, potentiates expression of ICAM1 and

VCAM1. In vivo, Ang2 deficient mice do not elicit an inflammatory response, once

again supporting the hypothesis that Ang2 is an antagonist of Ang1/Tie2 signaling 65.

25

The lack of Ang2 appears to affect later steps of the inflammation cascade such as the

transition of leukocytes from rolling to firm adhesion, as well as affecting leukocyte

transmigration 65.

Finally, it is important to note that while signaling via the Ang1/Tie2 pathway

appears to be essential to maintenance of EC quiescence in adult, and thus generally

beneficial, excess signaling via this pathway has been shown to also have deleterious

effects. Tie2 overexpression in the skin of mice has been shown to produce a psoriasis-

like phenotype characterized by epidermal hyperplasia, inflammatory cell accumulation,

and altered dermal angiogenesis 52. Furthermore, there exist naturally occurring

mutations of the Tie2 receptor found in some familial vascular formations. An autosomal

dominant mutation resulting in an arginine to tryptophan (R to W) substitution in the

kinase domain of Tie2 has been found to co-segragate with venous malformations in

three unrelated families 54,100. This mutation is believed to confer constitutive activity to

the receptor 54,100 and potentially increase the survival capacity of endothelial cells found

in mural cell deficient vessels 101. In summary, the Ang/Tie2 system is essential for

many aspects of developmental and adult physiological vessel growth, maintenance and

remodeling, but is a process that must be carefully monitored as either the absence or

excessive signaling via this signaling pathway(s) can have serious repercussions on the

vasculature.

1.4 Ang/Tie Signal Transduction

1.4.1 Receptor Activation and Phosphorylation

26

Activation of Tie2 by the angiopoietins is believed to follow the pattern generally

proscribed to RTKs: ligand binding, receptor multimerization and autophosphorylation

resulting in phosphorylation of intracellular tyrosine residues (see above section on

RTKs).

In the case of the Angiopoitin/Tie system, we know that higher order ligand multimers

are required for receptor activation 40,42,44-46(Figure 1.3 and 1.4). Structural analysis by

Barton et al. has suggests that Tie2 receptor clustering occurs following interaction with

a preformed ligand multimer resulting in receptor activation 40.

Receptor autophosphorylation is also a crucial step in activation of signaling

pathways; this step functions to activate the kinase domain and create downstream

phosphorylated binding sites (Figure 1.4). Biochemical studies of Tie2 show that

phosphorylation of this receptor occurs first in the activation loop on tyrosine 992

(Y992), followed by residue(s) in the C-terminal tail 102.

Tie2 harbours 19 intracellular tyrosine residues, of which tyrosines 1100 (Y1100),

1106 (Y1106) and 1111 (Y1111) of the C-tail have generated interest to date as

modulators of intracellular signaling pathways. These pathways appear to influence a

number of cellular processes including survival, migration and proliferation (Figure 1.3).

1.4.2 Signal Transduction via Tie1

For many years after its discovery, Tie1 remained an orphan receptor. In the

absence of a ligand, studies were carried out using a chimeric Tie1 receptor composed of

the extracellular domain of macrophage colony-stimulating factor 1 (CSF1) receptor and

27

1. Ligand binding (preformed multimer)

2. Receptor multimerization

3. Transduction of signal to intracellular region

Y1100

pY992

Y1106

Y1111

4. Phosphorylation of activation loop and ↑

receptor kinase activity5. Phosphorylation of tyrosines in C-tail

7. Downstream signaling cascades and cellular

responsesSurvival Migration

Proliferation?

6. Recruitment of Tie2 binding proteins

Figure 1.4: Tek/Tie2 mediated signal transduction. Schematic representation of select signal transduction events upon activation of the Tek/Tie2 RTK (as described in text).

28

the intracellular Tie1 domain 103. These studies demonstrated that Tie1 may mediate

survival signals via a PI3K-AKT pathway similar to Tie2 (see below). Subsequent

reports eventually identified that Ang1 is in fact able to activate Tie1 in endothelial cells

35,36. The mode of action of Ang1 on Tie1, however, remains unclear as Tie1

phosphorylation by Ang1 is dependent on Tie2 35,36. Also unclear is the purpose of this

activation as siRNA studies indicate that Ang1 activation of Tie1 does not contribute to

known functions of Ang1 signaling such as anti-permeability and anti-apoptotic effects in

HUVEC cells 37.

Tie1/Tie2 hetero-oligomers have been observed previously 35 which may help

explain mechanism(s) of Tie1 signal transduction. It has been suggested that Tie1 may

affect Tie2 signal transduction. However these reports are conflicting as a study

conducted in ECs demonstrated that Tie1 can affect ligand binding to Tie2 the result of

which is enhanced Tie2 phosphorylation 104. In a second study using endothelial

progenitor cells, Tie1 phosphorylation appears to interfere with Tie2 phosphorylation and

subsequent downstream signaling 36. Once again the importance of cellular context in

this signaling pathway is highlighted and further study will be needed to clarify some of

these conflicting reports.

Finally, the Tie1 ectodomain has also been shown to get proteolytically cleaved

generating a 45kDa membrane anchored endodomain. This endodomain has been found

at sites of angiogenesis and vessel remodeling and its cleavage is mediated by phorbol

ester, VEGF, TNFalpha and changes in shear stress 105. VEGF mediated Tie1 cleavage

has been shown to induce Tie2 phosphorylation 106. However, the biological

consequence(s) of this receptor cleavage remains to be determined.

29

1.4.3 Cell survival

From the beginning, an important role in endothelial cell survival has been ascribed to

the Ang/Tie2 pathway. First noticed as a lack of endothelial cells in the Tie2 null

embryos 70, numerous in vitro experiments have since corroborated a strong role for the

Ang/Tie2 signaling pathway in maintenance of endothelial cell survival.

A key modulator of cell survival involves phosphoinositide 3-kinase, PI3K, which

mediates this cellular process through regulation of the serine-threonine kinase Akt

(PKB). Perhaps not surprisingly, then, came the early discovery that Tie2 could associate

with p85, the regulatory subunit of PI3K 71,107. Both PI3K and Akt are activated

downstream of Tie2. A growing number of studies show that Ang1 stimulation of

endothelial cells protects against cell death via a PI3K/Akt pathway 73,74,107-109. Inhibition

of cell death also appears to involve the inhibitor of apoptosis protein (IAP), survivin,

which is upregulated in endothelial cells in response to Ang1 stimulation108,110.

Other identified Tie2 binding proteins may also be involved in Angiopoietin

mediated cell survival via a PI3K pathway including the A20 binding inhibitor of NF-

kappaB activation-2 (ABIN-2) 111. In these studies it was shown that while PI3K

inhibitors were able to suppress ABIN-2 mediated inhibition of endothelial apoptosis, a

truncated ABIN-2 prevented Ang1 from inhibiting cell death suggesting these effects

were in fact downstream of Tie2 111.

It has also been suggested that Ang1 mediated endothelial cell survival may also

involve the MAPK pathway, where a balance of both pro and anti- apoptotic MAPK

signals are activated in response to Angiopoietin-1, resulting in a net attenuation of

30

apoptosis 110. Given the apparent importance of the angiopoietin/Tie2 pathway in

endothelial survival signals, it is not unusual to think that redundant signals downstream

of this receptor have evolved to carry out this important cellular function.

1.4.4 Cell migration

Mice lacking Tie2 signaling pathways lack proper vessel sprouting and remodeling

67,70,75, a process that is highly dependent on the ability of the endothelial cells to migrate

into the surrounding basement membrane. Consistent with these observations, numerous

in vitro experiments involving the angiopoietins have shown that these ligands are

involved in endothelial cell migration and sprouting 71,72,74,112-115. PI3K appears to play a

role in endothelial chemotaxis and inhibition of the PI3K pathway has been shown to

inhibit both Ang1 and Ang2 mediated endothelial cell migration 71,73,112,116. These effects

appear to be mediated by various pathways downstream of PI3K.

Focal adhesion kinase (FAK), for example, is a protein that is known to regulate

important changes in the actin cytoskeleton organization during cell migration and

adhesion 117 and its phosphorylation is induced by Ang1 in a PI3K dependent manner 74.

The Tie2 binding proteins, Grb7 and Shp2, have also been implicated in adhesion

dependent cell migration through associations via activated FAK 118 119, although their

role in Tie2 signaling remains to be determined.

The guanine nucleotide (GTP)-binding proteins rho and rac, also known mediators of

cytoskeletal changes and cell migration, have also been shown to be involved in Ang1

induced endothelial cell migration and inhibition of these by dominant negative

constructs suppresses Ang1 mediated cell migration in endothelial cells 116.

31

One of pathways best characterized downstream of Tie2 is mediated through Y1106

which binds the docking protein DokR (p56Dok2/FRIP) 71. DokR binds activated Tie2 via

its PTB domain and itself becomes phosphorylated. DokR’s PH domain also seems to

contribute to its activation by Tie2, presumably through recruitment to the cell membrane

via a PI3K dependent mechanism 120. The phosphorylated sites on DokR then serve as

binding sites for RasGAP and Nck, molecules which have both been shown to play a role

in cell motility 121. In fact, DokR expression has been shown to potentiate Ang1

stimulated cell migration through recruitment of an Nck/p21 activated kinase, Pak 122.

Finally, the SH2 containing protein, ShcA, has been identified as a Tie2 binding

molecule. Overexpression of a dominant negative ShcA reduced Ang1 induced cell

migration and sprouting, but not cell survival 115. This adaptor molecule has previously

been implicated in transmitting signals to the Ras/Mitogen-activated protein kinase

(MAPK) pathway, although whether or not this is its mode of action in Tie2 signaling

remains to be seen. Interestingly, Tournaire et al. have shown that a short synthetic

peptide that blocks Tie2 binding to Ang1 and Ang2 inhibits cell migration and Erk

activation in HUVEC cells 96. As in the case of cell survival, it is likely that a number of

pathways contribute to this important function of Tie2 signaling. The fact that disruption

of PI3K pathway only results in partial inhibition of angiopoietin mediated cell migration

further supports this hypothesis 71.

1.4.5 Proliferation

Because a number of angiogenic molecules, such as the VEGFs, play an

important role in stimulating endothelial mitogenesis, initial studies suggesting that the

32

Angiopoietin/Tie2 pathway was not involved in cell proliferation were surprising given

that the receptor had been found to interact with molecules upstream of the Ras/MAPK

pathway (such as Grb2 and Shp2). One possibility is that pathways such as those

mediated by DokR and another Tie2 binding partner, Grb14, may negatively regulate

mitogenic responses as both these molecules have been shown to attenuate proliferation

downstream of other cytokine and tyrosine kinase receptors 71,123-125.

Interestingly, a study by Kanda et al. was able to show that Ang1 is in fact able to

induce cell proliferation in cultured endothelial cells via a MAPK and p70 s6K dependent

pathway 126. A second study in endothelial cells isolated from bovine mesentery shows

that Ang1 stimulated a weak proliferative response in both lymphatic and aortic

endothelial cells 127. Interestingly, the proliferative response was absent from endothelial

cells of venous origin. This seemingly conflicting data supports the emerging idea that

context plays a significant role in Angiopoietin/Tie2 signaling.

1.4.6 Tie2 context and signaling

Much of this seemingly conflicting data surrounding Angiopoieint/Tie signaling

supports the notion that context plays an extremely significant role in this system.

Ligand specificity, endothelial cell origin and general surrounding milieu are all integral

aspects of Ang/Tie signaling not to be ignored. An interesting study conducted in

HUVEC cells demonstrated that even receptor location within any given cell may impact

which signal transduction pathways are promoted. In this study, when ECs were

confluent, Ang1 could mediate a Tie2-Tie2 trans-association between cells. Tie2 located

at such cell-cell contacts elicited Akt activation leading to inhibition of FOXO1 mediated

33

gene regulation and phosphorylation of eNOS 128. However, in the absence of cell-cell

contacts, ECM bound Ang1 localized Tie2 to cell-substratum contacts and preferentially

activated Erk 128. Thus, differential signaling proteins appear to be activated by Tie2

depending on context, in this case at sites of cell-cell vs. cell-matrix contacts.

1.4.7 Regulation of Tie2 activity

Negative regulation of RTK activity can occur at various steps of the signal

transduction cascade. Competing ligands, receptor conformation, recruitment of

negative regulatory molecules; these are just a few of the ways where RTK signaling can

be dampened or attenuated. As is the case with many receptors, very little is known

about Tie2 negative regulation.

Angiopoietins as antagonist ligands

Unique to the Angiopoietin/Tie system, the angiopoietins appear to have opposing

actions on endothelial cells. It was initially believed that along with Ang1, Ang4 acted as

receptor agonists, while Ang2 and Ang3 act as antagonists. Subsequent experiments,

however, suggest a dual role for Ang2 as both receptor agonist and antagonist.

While Ang1 is believed to be the main activating ligand for Tie2, Ang2 was first

described as an inhibitor of Tie2 signaling because mice overexpressing Ang2 die

embryonically with defects similar to Tie2 and Ang1 null mice 76 (see previous).

Furthermore, in this study the Ang2 ligand was unable to activate the Tie2 receptor in

endothelial cells. Since then, however, Ang2 has been shown to activate endothelial Tie2

under specific conditions, its actions influenced by various parameters such as dose,

source and exposure time. In a study conducted by Bogdanovic et al., Ang2 was reported

34

as a partial agonist of Tie2 signal transduction as Ang2 activation of Tie2 was

considerably weaker when compared to Ang1 stimulation of Tie2 in HUVEC cells 129.

Cell context may also be an important factor in Ang/Tie signaling as lymphatic

endothelial cells appear to be more responsive to Ang2 than Ang1 127. This is in line with

lymphatic defects observed in vivo in adult Ang2 KO mice 61. The activation state of the

ECs may also be a factor in angiopoietin signaling as Ang2 acts as an inhibitor of Ang1

signaling to destabilize the endothelium in the resting vasculature 65,98,130 35 while it acts

as a stimulator of Tie2 in activated or stressed endothelium 131. Finally, ligand

presentation may be an important factor in the role of Ang2 on Tie2 activation.

Biochemical analysis of recombinant Ang1 and Ang2 show differential mobility when

run in a denaturing gel whereby Ang1 appears as an oligomer while Ang2 as a dimer 46.

In line with these observations, it has been shown that higher order Ang oligomers act as

Tie2 agonists while dimers act as antagonists 44.

Mouse Ang3 and human Ang4 represent the mouse and human counterparts for

the same gene locus but are more divergent than the mouse and human counterparts of

Ang1 and Ang2 and are thus referred to as interspecies orthologues 132. Initial studies of

these ligands on human endothelial cells again supported opposing action for these two

ligands where Ang4, but not Ang3, was able to activate Tie2 signaling 132. Subsequent

studies, however, demonstrated that like Ang2, under certain conditions Ang3 was able to

activate Tie2 in mouse endothelial cells 133.

While the mechanisms surrounding these various observations are poorly

understood, it is clear context plays an essential role in the Ang/Tie pathway.

35

Receptor internalization

Receptor internalization and degradation is another way in which a cellular signal

can be attenuated 134. In the case of many RTKs, cell surface receptor levels have been

shown to be controlled by ligand binding which promotes rapid receptor internalization

and degradation 135. A study by Bogdanovic et al. has shown that binding of Ang1, and

to a lesser degree Ang2, induces Tie2 internalization and degradation, which may be

another way in which signal attenuation occurs in this system 129. Interestingly, Ang1

and Ang2 are not internalized with the receptor suggesting the ligands are released back

into the surrounding medium to be recycled.

Inhibition by the C-terminal tail

The solving of the Tie2 crystal structure has revealed unique features of Tie2 that may be

involved in its regulation17. In general, the activation loop in the tyrosine kinase domain

of RTKs is an important structural regulator of receptor kinase activity 8.

Phosphorylation of one or more tyrosines in the activation loop often enhances receptor

catalytic activity 14. In Tie2, the activation loop adopts an ‘active-like’ conformation in

absence of phosphorylation. Instead, it appears that residues in the C-tail may interact

with the substrate binding site in the kinase domain and interfere with substrate binding

17. The C-terminal tail may therefore provide an additional degree of control at the

receptor level.

Regulation by phosphatases

36

RTK activity can also be controlled through the addition and removal of

phosphate groups from the receptor. While ligand activation of the receptor normally

results in activation of receptor kinase activity and thus autophosphorylation, protein

phosphatases are responsible for the removal of phosphate groups. Tie2 has been shown

to interact not only with the tyrosine phosphatase Shp2, but also with the endothelial

specific vascular endothelial protein tyrosine phophatase (VE-PTP) (mouse homologue

of HPTP-ß) 136,137. While Shp2 does not appear to have an effect on Tie2

phosphorylation (our unpublished observations), VE-PTP was shown to be able to

attenuate Tie2 phosphorylation in overexpression studies 136.

Another possibility is that these phosphatases serve as signaling conduits in the

Tie2 pathway. For example, Shp2 possesses two SH2 domains and could easily serve to

link Tie2 to other phosphotyrosine binding signaling partners. Shp2 has been shown in

other signaling systems to be a positive regulator whereby it is required for the activation

of the Ras-ERK pathway (reviewed in 30). Further studies are required to determine the

role that either of these molecules play in Tie2 signal transduction.

1.5 Growth Factor Receptor Bound Proteins (Grbs)

Early on in the study of Tie2, yeast two-hybrid studies identified a number of

Tie2 binding partners {Jones, 1999; Kontos, 1999}. Included in these were two of the

three Grb7 family members, Grb7 and Grb14.

1.5.1 Structure

37

The Grb7 family of proteins are a group of intracellular adaptor proteins originally

identified by CORT (cloning of receptor targets) screens of cDNA expression libraries

using the autophosphorylated intracellular EGFR C-terminus 138-140. These proteins,

while lacking intrinsic enzymatic activity, have been implicated in various cellular

processes such as regulation of cellular growth and metabolism, cell migration and

apoptosis through their interactions with various receptor and non-receptor kinases and

other intracellular singnaling molecules. Grb7 family members (Grb 7/10 and 14) share a

highly conserved multidomain structure made up of an amino-terminal proline rich

region (PRR), a central segment called the GM region (Grb and Mig) and a carboxyl-

terminal SH2 domain (Figure 1.5).

Proline Rich Region (PRR)

Grb 7/10/14 posess a highly conserved P(S/A)IPNPFPEL motif in their N-

terminal region and have at least one other PXXP motif. Grb7 has five consensus PXXP

motifs which could potentially bind SH3 containing molecules, however no proteins have

been yet found to interact with this region of Grb7. In contrast, the Grb10 PRR has been

shown to bind the SH3 domain of c-Abl in vitro, but not the SH3 domains of PI3K, Grb2

or Fyn 141. The PRR has also been shown to bind proteins in an SH3 independent

manner. A study by Giovannone et al. showed that two novel GYF proteins (named

because of presence of glycine, tyrosine and phenylalanine residues), GIGYF (Grb10

interacting GYF protein) 1 and 2, can bind to tandem proline-rich segments of Grb10 via

GYF motifs 142. The N-terminal 110 amino acids of Grb14 have been shown to

38

Pro RA PH BPS SH2N C

Binding to SH3 domain containing proteins

Binding to Ras superfamily?

Binding to PtdInsPs

Binding to certain receptors and intracellular proteins

Binding to phosphotyrosineresidues of receptors and intracellular proteins

Figure 1.5: Grb7 family structure. Schematic representation of the growth factor receptor bound protein (Grb) family conserved structure including the N-terminal proline rich region (Pro), a Ras associating domain (RA), a central pleckstrin homology (PH) domain, between PH and SH2 (BPS) domain and C-terminal Src homology 2 (SH2) domain. (Adapted from Holt and Siddle, 2005)

GM Region

39

bind to ankyrin repeats of a novel human tankyrase, Tankyrase-2 143 although whether or

not the proline sequences play a direct role in this interaction remains to be determined.

GM Region

Early on it was noted that a region of approximately 300 amino acids on the Grb

proteins displayed high sequence homology (~50% amino acid identity) with a

Caenorhabditis elegans gene product involved in neuronal cell migration, Mig-10

138,144,145 139. In the Grb proteins, this so named GM (Grbs and Mig) region encompasses

a putative RA (Ras associating) domain and a pleckstrin homology (PH) domain 138,144,145

(Figure 1.5).

RA domain

The RA domain was proposed based on sequence homology analysis 145 indicating

that Grb7 family members may play a role in regulating Ras signaling pathways. While

initial studies failed to detect Grb7/G-protein interactions, a study by Rodriguez-Viciana

et al. demonstrate an interaction between Grb7 and N-Ras K-Ras and R-Ras suggesting

the RA domain may in fact link Grb7 family members to Ras signal transduction

pathways 146.

PH domain

Consistent with the role of the PH domain in other signaling molecules, it is believed

that the PH domain of Grbs7/10/14 serves to recruit theses proteins to the cell membrane

via interactions with membrane phospholipids. It has been shown that both Grb7 and

40

Grb14 PH domains preferentially bind D3 and D5-phosphoinositides, a phenomenon that

appears to be mediated by the PI3K pathway in the case of Grb7 147 148.

The Grb7 GM region has been shown to interact with one other protein, NIK

(Nuclear factor kappa B-inducing kinase), a member of the MAPKK family and a protein

involved in NFkappaB activation 149 150. In this study, a Grb7/NIK complex was

recruited to EGFR, ErbB2, ErbB3 and ErbB4 signaling complexes and was shown to

potentiate Grb7, ErbB2/ErbB4 and EGF-induced NFkappaB activation 150

BPS domain

Unique to the Grb7 family members is a conserved region of approximately 45

amino acids in length named the BPS (between pleckstrin homology and SH2) or PIR

(phosphorylated interacting region). Studies have discovered that this domain is

unstructured in solution and appears to belong to the IUP (intrinsically unstructured

proteins) class of proteins 151. Like other IUPs, the BPS domain exhibits little secondary

structure which has been proposed to allow flexibility in order to interact well with a

number of different target molecules 152. To date the BPS domain has been shown to

interact with the activated IR and IGFR 153 and functionally has been shown to inhibit IR

kinase activity 154 155 (see below for further info). The Grb14 BPS domain has also been

shown to bind to the ZZ domain of ZIP whereby the trimeric complex Grb14, ZIP and

PKCζ is believed to play a role in insulin signaling 156.

The solving of the BPS crystal structure with the IR kinase (IRK) shows that the

Grb14 BPS binds as a pseudosubstrate of the IRK and interferes with the phosphorylation

41

of exogenous substrates 157. This interaction also appears to be phosphotyrosine

dependent, although many other interactions are required to stabilize the association 157.

SH2 domain

This phosphotyrosine binding module is the most highly conserved region of Grb7 family

memebers. The Grb SH2 domains are type I SH2 domains and are responsible for

binding to a number of receptors and other signaling molecules via phosphotyrosine

mediated interactions. Interestingly, structural and biochemical studies have indicated

that the SH2 domains of both Grb10 and 14 may bind phosphotyrosine residues in a

unique fashion. Unlike the typical SH2-phosphotyrosine interaction which favors

binding of a phosphotyrosine contained in a region of extended conformation, Grb10 and

14 SH2 regions may favour binding to turn-containing phosphotyrosine sequences such

as those found in phosphorylated IR activation loop 157. As mentioned previously,

differential binding abilities of the Grb proteins may also be affected by specificity

imparted by the BPS domain, as is the case in insulin signaling.

1.5.2 Biological role

While all three Grb7 family members were originally cloned in screens using the

EGFR as bait, as with most other SH2 containing proteins it was quickly discovered that

this family of adaptors binds a number of other receptors (reviewed in 158). In some cases

more than one family member has been shown to bind the same receptor, however, each

Grb appears also to bind its own distinct set of proteins.

42

All three Grbs have been shown to associate with the IR. Initial studies focused on

Grb10 and provided somewhat conflicting data as to its role in this system as Grb10 was

shown to be both a positive and negative regulator of insulin signaling (reviewed in Holt

and Siddle., 2005) 158. A more recent look at Grb14 in insulin signaling suggests Grb7

family members may be negative regulators of this system. Biochemical studies show

that both Grb10 and Grb14 negatively impact IR kinase activity resulting in reduced

phosphorylation of downstream IR substrates such as IRS-1, IRS-2 , Shc and p62dok as

well as negatively impacting activation of various downstream IR biological functions

124,148,158-161. Some of the conflicting data regarding the role of these adaptors may in part

be due to differential tissue expression of the Grbs. This is corroborated by mouse

molecular studies that show tissue specific effects of Grb10 and 14 ablation on insulin

signaling 161. Furthermore, it has been proposed that Grb10 and 14 may compete with

other molecules for binding sites on the IR, and therefore affect other signaling pathways

indirectly. For example, both Grb10 and 14 have been shown to protect various IR sites

from dephosphorylation by phosphatases such as PTP1-B 155,158,161. Grb10 has been

shown to mediate insulin stimulated IR receptor degradation, presumably via a

ubiquitin/proteosomal degradation pathway 162. Together, these studies suggest a role for

Grb7 family members, especially Grb10 and 14, as negative regulators of some signaling

pathways.

Grb7 has also been shown to interact with IR 163, although less well than Grb10 and

14. Instead, Grb7 has been most defined as a mediator of cell migration via interactions

with FAK upon growth factor stimulation 164,165. This biochemical data corroborates the

observation that Grb7 has been found to be overexpressed in highly invasive tumours 139

43

166. A role for the Grb7 family members in cell migration should not be surprising given

that they share high homology with the C. elegans Mig10 gene which plays a role in

embryonic neuronal migration 144.

Other biological roles for Grb14 are also beginning to emerge. Grb14 was found to

bind tankyrase-2, a novel poly(ADP-ribose) polymerase (PARP) potentially involved in

vesicle trafficking 143. Grb14 has also been found to harbor a putative nuclear

localization signal (NLS) between amino acids 63 and 68 (RRKKD) and is located in

both rat retinal cytoplasmic and nuclear fractions suggesting the possibility that Grb14

may help localize some proteins to the nucleus 148. A likely candidate may be IRS-1

since it has been shown to bind Grb14 via Grb14 NPXY motif and PTB domain of IRS-1

in a phosphotyrosine independent manner 148. However, whether or not this is the case

remains to be determined.

.

1.5.3 Phosphorylation

While the biological role of the Grb proteins appears to be affected by their

binding specificity as well as their specific tissue distribution and expression levels, very

little is know about how they are regulated at the transcription, translation or

phosphorylation level.

The Grb7 family of proteins have been shown to be phosphorylated on serine,

threonine and tyrosine residues, although the functional significance of these

phosphorylations are still unclear in most cases. Both Grb10 and Grb14 were shown

early on to posess basal serine phosphoryltaion 125,138,140,167. Grb10 serine

phopshorylation can be further induced by GF stimulation such as EGF, PDGF, FGF or

44

insulin treatment 138,167, a phenomenon that may be mediated by PI3K and MAPK

pathways 167,168. Grb14 serine phosphorylation can also be induced by PDGF and FGF

but not EGF 125,140. PKC can also phosphorylate Grb14 on serine residues, primarily in

the BPS region 156.

Grb10 has also been shown to be tyrosine phosphorylated 141,169,170. In insulin

signaling, this phosphorylation appears to be a substrate of Src family kinases as opposed

to the IR’s kinase activity 169. This is also the case in endothelial cells where Grb10

tyrosine phosphorylated by VEGF is at least in part mediated by Src 171.

Grb7 tyrosine phosphorylation was noted early on in the presence of the Tie2

endothelial receptor 6, and by FAK at sites of focal adhesion 172. Grb7 is also tyroinse

phosphorylated in the presence of other growth factors such as EGF 173, ephrinB1 174 and

when binding to the Ret and HER2/erbB2 receptors 175 139.

To date, there are no reports of Grb14 tyrosine phosphorylation, although it is

predicted to do so under correct conditions given its interaction with numerous RTKs and

similar structure to Grb7 and 10.

45

1.6 Thesis Objectives and Organization

1. To examine the role of the Tie2 C-terminal tyrosine residues in Tie2 signal

transduction using previously established tyrosine to phenylalanine (Y to F) point

mutants of the Tie2 receptor. Y1111 presented with unique attributes and was

examined in an overexpression system using constructs where this residue was

mutated to various other amino acids to establish its role as a structural negative

regulator of Tie2 signaling. The results of these studies are presented in Chapter 2.

2. To further understand the role of the Grb adaptor proteins in Tie2 mediated signal

transduction. To this end we employed an overexpression system in HEK293T cells

and previously established Tie2 constructs where specific tyrosine residues (Y) were

mutated to phenylalanine (F). These constructs were used to elucidate key tyrosine

residues on Tie2 important for Grb signal transduction. The results of these studies

are presented in Chapter 3.

46

CHAPTER 2: A NEGATIVE REGULATORY ROLE FOR

Y1111 ON THE TIE2 RTK

Chapter 2 is a modified version of the following submitted manuscript to

Cellular Signaling:

A NEGATIVE REGULATORY ROLE FOR Y1111 ON THE TIE-2 RTK

Sturk C1,2, Kim H, Jones N, Dumont DJ* 1,2,3,4

Address: 1Molecular and Cellular Biology Research, Sunnybrook Research Institute, Sunnybrook Health Sciences Centre, Toronto, ON, Canada, 2Department of Medical Biophysics, University of Toronto, ON, Canada, 3Heart & Stroke/Richard Lewar Centre of Excellence, Faculty of Medicine, University of Toronto, Toronto, ON, Canada and 5R. Samuel McLaughlin Centre for Molecular Medicine, Toronto, ON, Canada

47

2.1 Abstract

Tie2 is a receptor tyrosine kinase (RTK) essential for aspects of both normal and

pathological angiogenesis. Understanding how this receptor is regulated is important for

development of therapeutic angiogenic agents. Evidence suggests the C-terminal tail of

the receptor plays a negative regulatory role in Tie2 signaling and function. Here we

investigated the role of a specific C-tail residue, Y1111, in Tie2 signaling by generating a

number of receptor point mutants. We found that mutation of this site to phenylalanine

(Y1111F) resulted in increased receptor phosphorylation and kinase activity, as well

increased downstream signaling. Furthermore, mutation of Y1111 to the highly charged

aspartate (Y1111D) or glutamate (Y1111E) resulted in an even more dramatic increase in

receptor phosphorylation and activity. Protease digestion studies indicated that these

mutations may alter receptor conformation and potentially relieve negative inhibition

imparted by the C-tail of Tie2. These studies suggest that Y1111 plays a key role in

negative regulation of Tie2 activity and they provide important insights into molecular

mechanisms behind the intrinsic ability of this RTK to regulate its own activity.

48

2.2 Introduction

Tie2 is a receptor tyrosine kinase (RTK) found primarily on the surface of

endothelial cells. Mouse molecular models have shown an essential role for this receptor

in aspects of angiogenesis. Tie2 deficient mice present with a plethora of vascular

abnormalities including a lack of proper sprouting and remodeling of the primitive vessel

network 67 70. In the adult, Tie2 has been implicated in both normal and pathological

states including wound healing, follicular development, diabetic retinopathy and

tumorigenesis. Given its importance in such a variety of processes, it has become

essential to decipher how this receptor functions in order to design therapeutic agents that

are able to target such angiogenic events.

Tie2 belongs to the Tie family of proteins. The two family members, Tie1/Tie

and Tie2/Tek, share a high degree of structural homology. Tie2 has been shown to bind

to a number of downstream signaling molecules including the adaptor proteins Grb2,

Grb7, Grb14, ShcA and DokR, the regulatory subunit of PI3K, p85, and the tyrosine

phosphatases, Shp2 and VE-PTP. A family of ligands known as the angiopoietins (Angs)

has been shown to play a role in Tie2 biology. Four angiopoietins have been identified to

date: Ang 1-4. Interestingly, these four members appear to have opposing actions in

endothelial cells as Ang1 and Ang4 act as receptor agonists, while Ang2 and Ang3 act as

context dependent antagonists. Currently, Ang1 remains the best characterized of the

angiopoietins and is believed to be the main activating ligand for Tie2. It has been shown

to activate the Tie2 receptor and contribute to such functions as endothelial cell survival

and migration 71,113. Ang1 has also been shown to counter blood vessel leakiness induced

49

either by vascular endothelial growth factor (VEGF) or inflammatory agents Thurston,

2000; Thurston, 1999} implicating these pathways in vessel maturation and stabilization.

Notwithstanding this evidence, many of the molecular mechanisms surrounding

Tie2 activation remain obscure. The solving of the Tie2 crystal structure has provided

initial insight into possible modes of receptor regulation 17. Specifically, unique

mechanisms appear to govern receptor auto-inhibition in the inactive receptor state. First,

it appears as if the activation loop adopts an ‘active-like’ conformation in the absence of

phosphorylation. Normally, RTK structure is such that the activation loop blocks either

the substrate or ATP binding site in the inactive state and phosphorylation of a specific

tyrosine residue within this region is required to relieve this inhibition. Instead, the Tie2

structure seems to suggest that the nucleotide binding loop is in an inhibitory

conformation in order to block access to ATP. Second, the C-terminal tail appears to also

play a role in receptor regulation as its positioning appears to block access to the

substrate binding site. This is postulated to be due to interactions involving C-tail

residues, including Y1100 and 1111. The suggestion that the C-tail inhibits receptor

activity is further supported by a study by Niu et al. which shows that deletion of the C-

terminal tail of Tie2 leads to an increase in receptor phosphorylation and kinase activity

176.

In this report we have examined the role of Y1111 in Tie2 activation.

Specifically, we generated point-mutations at this site that result in increased receptor

phosphorylation, activation and signaling, indicating that this one residue is very key in

maintenance of proper receptor regulation. Furthermore, protease digestion studies

support the model that a conformational change occurs within the receptor if Y1111

50

interactions are disrupted. This presents exciting possibilities with respect to

understanding how Tie2 is activated and how this activation is regulated.

51

2.3 Materials and Methods

Production of GST Fusion Proteins – Construction of GST DokRΔPH GST fusion was

described previously 122. Proteins were prepared from Escherichia coli (BL21)

(Invitrogen) using standard procedures, and the recombinant fusion proteins were purified

following immobilization on glutathione-Sepharose beads (Amersham Pharmacia

Biotech). Purified proteins were analyzed by SDS-polyacrylamide gel electrophoresis

followed by Coomassie Blue staining. The concentrations of the proteins were estimated

by comparison with bovine serum albumin standards.

Expression vectors and mutagenesis – Full-length Tie2 in pcDNA3.1 120 was used as a

template to generate point mutations with the QuikChange site-directed mutagenesis kit

(Stratagene). Primers used were as follows: for Y1111D 5’-

GAGAAGTTTACCGATGCAGGAATTGAC -3’, and for Y1111E 5’-

GAGAAGTTTACCGAGGCAGGAATTGAC-3’. All mutations were verified by

sequencing. The cDNAs representing wild type (WT), kinase-inactive (K853A) and

Y1111F Tie2 in pcDNA3.1 have been described previously {Jones, 2003 #66177. The

cDNA encoding Flag tagged full-length Shp2 was a kind gift from Gen-Sheng Feng.

Cell Culture, Antibodies and Production of Conditioned Medium - C166 and

HEK293T cells were grown on 10-cm plates in Dulbecco’s modified Eagle’s medium

(DMEM) (Sigma) supplemented with 10% FBS, 1% penicillin, 1% streptomycin, and

200mM L-glutamine. Antibodies used were as follows: polyclonal anti-Tie2 C-20 (Santa

Cruz), monoclonal anti-phosphotyrosine 4G10 (Upstate Biotechnology Inc.), monoclonal

52

anti-Tie2 (Pharmingen), monoclonal anti-Flag M2 (Sigma), polyclonal phospho-Tie2

(992) (Cell Signaling Technology). GST mixing experiments, coimmunoprecipitation

experiments, and Western blotting procedures have been previously described 71,177.

Transfection Procedures – HEK293T cells were cultured to ~75% confluency in 10-cm

cell culture plates and transfected with 2μg DNA using the lipofectamine reagent

(Invitrogen Life Technologies) and according to manufactures protocol. Cells were

harvested after 48 hours.

In Vitro Kinase Assay – Kinase assays were performed as follows: Cell lysates were

immunoprecipitated with anti-Tie2 antibodies for 2 h at 4 °C. Immunoprecipitates were

washed 3x in PLC lysis buffer (without sodium fluoride or sodium pyrophosphate) 177 and

2x in kinase buffer (2 mM MnCl2 + 50 mM HEPES pH 7.5 + 10 mM MgCl2 + 0.2 mM

dithiothreitol) and then incubated with 4 µg of either GST or GST-Dok-R PH as substrate

including 40 µCi of [ -32P]ATP (Amersham Biosciences), 20 µM ATP (Amersham

Biosciences) for 30 min at 30 °C 122. Kinase reactions were stopped by the addition of 2x

SDS-containing sample buffer and boiled for 10 min. Immunoprecipitates were

electrophoresed and half the gel was used to resolve Tie2 expression by Western analysis

using anti-Tie2 antibodies while the other half was dried and exposed to phosphorimager

analysis and quantification (ImageQuant). The representative values of Tie2 kinase

activity were reflected as the value of GST-Dok-R PH phosphorylation over the amount

of immunoprecipitated Tie2 in each of the samples. All experiments were performed

twice or more times with similar results.

53

Protease Digestions - Tie2 immunoprecipitates were washed 3X in 1% Triton lysis buffer

and 1X in either proetinase K digestion buffer (50mM NaCl, 20mM Tris-HCL pH7.5,

1mMDTT) or trypsin digestion buffer (100mM NaCl, 100mM Tris-HCL pH7.9, 1mM

CaCl2 and 1mMDTT). Digestons were carried out at 37°C in the presence of 10μg/ml of

either Proteinase K (Sigma) or sequencing grade Trypsin (Promega). Aliquots were

removed at the indicated time points, treated with PMSF and then boiled in SDS sample

buffer. Samples were analyzed by SDS-PAGE on 8, 10 or 12% gels and blotted for Tie2.

54

2.4 Results

Phosphorylation of wild type Tie2 and Tie2 tyrosine to phenylalanine point mutants. In

an attempt to elucidate signaling pathways downstream of the Tie2 receptor, our lab has

previously generated a number of tyrosine to phenylalanine point mutations within the C-

terminal tail of the receptor (shown schematically in Figure 2.1). Of the 19 tyrosine

residues on the intracellular portion of Tie2, C-terminal residues Y1100 and Y1106 have

been shown to be involved in downstream signaling events 71,102,115,120,122. Y1100 has

been implicated in signaling downstream to PI3K and as well appears to be a

multidocking site for the adaptor proteins ShcA, Grb7 and Grb2 71,116. Moreover , Y1106

was shown as an Angiopoietin-1 dependent autophosphorylation site on Tie2 that

mediates binding and phosphorylation of the docking protein Dok-R 120. Also postulated

to play an important role in receptor function is Y1111 (reviewed in 178). In an attempt to

further characterize the role of this tyrosine in Tie2 signal transduction, Y1111 was

disrupted by mutation to phenylalanine (Y1111F). This construct, along with two other

single point mutant constructs, Y1100F and Y1106F, were transiently expressed in 293T

cells. Also included were the wild type receptor (Tie2WT) and a kinase inactive mutant

of Tie2 (K853A). Immunoprecipitation of these lysates with a Tie2 specific antibody

followed by blotting with a phosphotyrosine specific antibody (4G10) revealed a distinct

phosphorylation pattern (Figure 2.2A). As anticipated, Tie2WT as well as the Y1100F

and Y1106F mutants became tyrosine phosphorylated while the kinase inactive mutant,

K853A did not (Figure 2.2A). Strikingly, mutation of tyrosine 1111 to phenylalanine

resulted in an increase in receptor phosphorylation (Figure 2.2A). This

55

Figure 2.1: Schematic representation of Tek/Tie2 point mutants. A number of tyrosine to phenylalanine point mutations were engineered by site directed mutagenesis (described previously in Jones et al., 2003).

56

Figure 2.2. Phosphorylation of wild type Tie2 and Tie2 tyrosine to phenylalanine point mutants. In A, HEK293T cells were either left untransfected (control) or were transfected with WT Tie2 , the kinase deficient Tie2 (K853A) or one of the Tie2 tyrosine mutatns (Y1100F, Y1106F or Y1111F). Lysates from these cells were then immunoprecipitated for Tie2, run on an SDS PAGE gel and subjected to Western blot using anti-phosphotyrosine (upper panel) or anti-Tie2 (lower panel). In B, adenoviral Tek/Tie2 constructs (WT, Y1100/1106F, Y1100/1111F, Y1106/1111F and Y3F) were expressed in C166 endothelial cells and lysates were subjected to immunoprecipitation and Western blotting as in A.

57

was consistently seen across many experiments and Figure 2.2A is a representative blot

of multiple experiments. Phosphorylation levels were subsequently quantified by

densitometric analysis and values were adjusted for loading (Figure 2.5C). These data

demonstrated that introducing a Y to F mutation at residue 1111 resulted in an increase in

receptor phosphorylation.

In order to assess the role of these tyrosine residues in the context of endothelial

cells, a number of adenoviral Tie2 mutant constructs were made (WT, Y1100/1106F,

Y1100/1111F, Y1106/1111F and Y3F) (Figure 2.1) and expressed in a Tie2 null

endothelial cell line (Figure 2.2B, lower panel). Phosphotyrosine blotting of Tie2

immunoprecipitated lysates from these cells showed that while WT and Y1100/Y1106

constructs show similar levels of phosphorylation, constructs that harbour a mutation at

Y1111 displayed increased phosphorylation (Figure 2.2B). This suggests that mutation

of Y1111 leads to an increase in receptor phosphorylation in the context of endothelial

cells.

Mutation of Y1111 to Phenylalanine results in increased receptor kinase activity. Tie2 is

a receptor tyrosine kinase and thus possesses intrinsic kinase activity. Previous studies

have shown that deletion of the C-terminal tail of Tie2 (murine residues 1108-1123)

results in an increased ability to phosphorylate synthetic peptide substrates in vitro 176.

This led us to ask the question 'does the increase in phosphorylation observed when

Y1111 is mutated to phenylalanine translate into a similar increase in receptor kinase

activity?'. To this end we carried out a kinase assay to

58

Figure 2.3: Mutation of Y1111 to Phenylalanine results in increased receptor kinase activity. HEK293T cells were transfected with wild type Tie2 (Tie2), a kinase deficient Tie2 (K853A) or the Y1111F mutant. Lysates were immunoprecipitated with an anti-Tie2 antibody and then either subjected to Western Blot using anti-Tie2 antibodies (A, top panel) or incubated in a kinasereaction with set amounts of glutathione-Sepharose purified GST-Dok-RΔPH and 32P labeled ATP (A, bottom panel). In B, results were quantified using densitometry and values were adjusted for receptor levels.

59

compare the kinase activity of Y1111F to that of the wild type receptor using a GST-

tagged Dok-R peptide as substrate. Dok-R is a docking protein that has been shown to be

tyrosine phosphorylated downstream of Tie2 signal transduction in vivo making this a

biologically relevant choice for a substrate. The DokR construct used herein lacks the

insoluble PH domain to allow for purification purposes and will be subsequently referred

to as GST-Dok-RΔPH (described previously) 122. The receptor was immunoprecipitated

from 293T cells expressing Tie2WT, K853A or Y1111F and incubated in a kinase

reaction with set amounts of glutathione-Sepharose purified GST-Dok-R and 32P labeled

ATP. Figure 2.3A is a representative blot of three experiments demonstrating the

enhanced ability of Y1111F to phosphorylate GST-Dok-RΔPH. Results were quantified

using densitometry and values were adjusted for receptor levels indicating an

approximate 7 fold increase in kinase activity (Figure 2.3B). These data show that

Y1111F has increased kinase activity over wild type Tie2.

Mutation of Y1111 to phenylalanine enhances downstream signaling. We next asked if

increased receptor phosphorylation and kinase activity of Y1111F translated into an

increase in signal transduction downstream of Tie2. To assess this, cell lysates were

prepared from 293T cells expressing Tie2WT, K853A or Y1111F and run on an SDS

polyacrylamide gel. Blotting with an antibody specific for phosphorylated tyrosine

revealed that not only is there a prominent increase in phosphorylation of the Y1111F

receptor over that of WT, but there is also a corresponding increase in phosphorylation of

other cellular proteins in the presence of the Y1111F mutant (Fig. 2.4A).

60

Figure 2.4: Mutation of Y1111 to Phenylalanine enhances downstream signaling. HEK 293T cells were transfected with TekWT, K853A or Y1111F either alone (A) or in conjunction with Grb14 (B), Shp2 (C) or p85 (D). In A, cell lysats were run on an SDS PAGE gel and blotted for phospho-tyrosine. In B through D, lysates were first immunoprecipitated with antiGrb14, anti-FLAG or anti-p85 (as indicated), run on an SDS PAGE gel and subjected to Western blot (antibodies used as indicated).

A B

C D

61

This suggests that Y1111F has the ability to increase tyrosine phosphorylation of other

cellular proteins over that of the wild-type receptor.

To specifically test the effect of the Y1111F mutation on a known downstream

target of Tie2 signaling, we looked at phosphorylation of Grb14 (an adaptor protein

previously shown to bind Tie2) in the presence of Y1111F. As shown in figure 2.4B,

Grb14 tyrosine phosphorylation is increased in the presence of the Y1111F mutant when

compared to Grb14 in the presence of WT Tie2.

We also examined the ability of Shp2 to become phosphorylated in the presence

of Y1111F. Shp2 has been shown to associate with the C-terminal tail of Tie2 via it's

SH2 domains 137,179 and has been postulated to play a role in Tie2 mediated MAPK

signaling (reviewed in Peters et al., 2004 178). While the mode of activation of Shp2

remains somewhat controversial, it has been shown that engaging Shp2 SH2 domain(s),

or phosphorylating specific tyrosine residues within the C-terminal tail, may be

responsible for activation of SHP2 catalytic activity (reviewed in Neel et al., 2003 180).

To look at phosphorylation of this phosphatase in our system, a FLAG-tagged construct

of Shp2 was transiently expressed in 293T cells in conjunction with Tie2WT, K853A or

Y1111F constructs. Lysates were immunoprecipitated for FLAG-tagged Shp2 and

Western blotted for phosphotyrosine. Figure 2.4C shows a dramatic increase in Shp2

phosphorylation in the presence of Y1111F versus the wild-type receptor (Tie2WT)

indicating that Y1111F mutation has the ability to 'hyper-activate' signaling pathways

downstream of this receptor.

Another important pathway downstream of Tie2 is endothelial cell survival via

PI3K/Akt activation. Previously it has been shown that Tie2 can bind to p85, the

62

regulatory subunit of PI3K. To examine the effect of Y1111F on this binding partner,

p85 was transiently expressed in 293T cells in conjunction with Tie2WT, K853A or

Y1111F constructs. Lysates were immunoprecipitated and Western blotted for

phosphotyrosine. Figure 2.4D shows an increase in p85 phosphorylation in the presence

of Y1111F versus the wild-type receptor (Tie2WT) indicating that Y1111F mutation has

the ability to 'hyper-activate' signaling pathways downstream of this receptor.

Tyrosine 1111 is important for structural negative regulation of the Tie2 receptor.

Activation of RTKs is a multi-step process that often involves receptor dimerization,

autophosphorylation and activation of the kinase domain. Evidence into the activation of

the Tie2 receptor has been provided by the solution of its crystal structure 17. Analysis of

this structure reveals a possible autoinhibitory mechanism involving the C-terminal tail of

the receptor which appears to block the substrate binding site. According to the crystal

structure, a number of C-terminal residues appear to be involved in stabilizing this

conformation, including Y1111 whose hydroxyl group appears to be hydrogen-bonded to

surrounding residues. Thus we postulated that mutation of Y1111 to F may be altering

receptor conformation by disrupting one or more of these hydrogen bonds, relieving C-

terminal autoinhibition and resulting in a more active form of the receptor (Figure 2.8).

To verify this hypothesis, we used site directed mutagenesis to generate two additional

mutants of Tie2, Y1111D and Y1111E whereby Y1111 was mutated to the highly

charged aspartate (D) and glutamate (E) residues respectively. It was predicted that

these residues would not be able to sit in the hydrophobic pocket where Y1111 is

normally positioned due to their highly charged side groups, and would

63

Figure 2.5: Tyrosine 1111 is important for structural negative regulation of the Tie-2 receptor. Lysates from HEK293T cells overexpressing wild type Tie2 (WT), the kinase deficient K853A or one of the Y1111 mutants (Y1111F, D or E) were immunoprecipitated for Tie2 and blotted for anti-phosphotyrosine (A, top panel) or phospho-Tie2 (B, top panel). Lysates subjected to SDS PAGE were blotted for Tie2 as a control for receptor levels (lower panel A and B). In C, results from 6 experiments were quantified using densitometry and values were adjusted for receptor levels. Results were expressed as the mean +/- S.E. p≤0.05

Relative Phosphorylation to WT

00.5

11.52

2.53

3.54

4.55

5.56

6.57

7.58

8.5

Vector WT K853A Y1100 Y1106 Y1111F Y1111D Y1111E

Construct

C

64

therefore be expected to induce a dramatic conformational change in the receptor. These

new mutants were expressed in 293T cells and immunoprecipitated from lysates as

previously indicated. When blotted for phosphotyrosine, Y1111D and Y1111E displayed

a dramatic increase in tyrosine phosphorylation above that of Tie2WT and even Y1111F

(Figure 2.5A). Similar results were also obtained by running out cell lysate and blotting

using a phospho-Tie2 specific antibody (Figure 2.5B). This suggests that replacing

Y1111 with different residues can alter the degree of receptor phosphorylation. It is our

belief that by introducing the highly charged residues D and E into the hydrophobic

pocket, receptor conformation has been significantly altered resulting in release of

autoinhibition of the C-terminal tail. In comparison, a Tyr to Phe mutation at residue

1111, while still more phosphorylated than WT, may be the result of the more minor

disruption (as predicted from the crystal structure) since phenylalanine would still be able

to fit into the hydrophobic pocket surrounding residue 1111 (see Figure 2.8 and

Discussion below).

Mutation of Y1111 to D or E results in increased receptor activity. Again, in order to

verify that this increase in receptor phosphorylation was also indicative of increased

receptor activation, we carried out a kinase assay to compare the kinase activity of

Y1111D and Y1111E (Figure 2.6A). An equal amount of receptor was

immunoprecipitated from 293T cells expressing Tie2WT, K853A or any of the Y1111

mutants (Y1111F/D/E) and incubated in a kinase reaction with set amounts of

glutathione-Sepharose purified GST-Dok-RΔPH and 32P labeled ATP. While Y1111F

displays an increase in kinase activity as seen previously (Figure 2.3A), Y1111D

65

Figure 2.6: Mutation of Y1111 to aspartate (D) or glutamate (E) results in increased receptor activity. HEK293T cells were transfected with wild type Tie2 (WT), a kinase deficient Tie2 (K853A) or one of the Y1111 mutants (Y1111F,D or E). Lysates were immunoprecipitated with an anti-Tie2 antibody and then either subjected to Western Blot using anti-Tie2 antibodies (A, top panel) or incubated in a kinase reaction with set amounts of glutothione-Sepharose purified GST-Dok-RΔPH and 32P labeled ATP (A, bottom panel). In B, lysates were run on an SDS PAGE gel and blotted for phospho-tyrosine (top panel) or Tie2 (lower panel).

66

and Y1111E mutants resulted in a dramatic increase in GST-DokRΔPH phosphorylation

(Figure 2.6A).

Figure 2.6B shows cell lysates from 293T cells overexpressing either of the

Y1111 mutants (Y1111D or E) show a sizable increase in overall protein phosphorylation

over WT , K853A and even Y1111F, again suggesting that mutation of Y1111 results in

an enhanced ability of these mutants to phosphorylate cellular proteins over that of the

WT receptor.

Mutation of Y1111 on Tie2 altered its protease susceptibility

The data from our Y1111D and E mutants support the hypothesis that mutation of Y1111

induces a conformational change in Tie2. Under this assumption, such conformational

changes may be reflected in the receptor’s susceptibility to proteases. To examine this

possibility we performed protease digestion studies on Tie2WT, K853A, and Y1111F,

Y1111D, Y1111E mutants. Constructs were expressed in 293T cells and receptors

immunoprecipitated using an antibody specific for the Tie2 extracellular domain were

digested with either proteinase K (Figure 2.7B) or trypsin (Figure 2.7A). Aliquots were

removed at various time points, analyzed by SDS-PAGE and blotted using an antibody

directed against the Tie2 C-tail. All constructs were the expected size of approximately

140 kDa prior to digestion (Figure 2.7A, first panel). Panels 2.7B depict the results of

digestion with proteinase K, a serine protease with broad cleavage specificity. After 2

minutes in the presence of this protease, it was apparent that the Y1111 mutants had an

altered pattern of digestion when compared to WT and K853A constructs. This was

67

Trypsin Digestion

Proteinase K Digestion

Figure 2.7.: Mutation of Y1111 on Tek/Tie2 alters its protease susceptibility. HEK293T cells were transfected with Tek/Tie2 WT receptor, the kinase deficient K853A receptor or one of the Y1111 mutatns (Y1111F,D or E). Lysates were immunoprecipitated with anti-Tek/Tie2 and immunoprecipitates were incubated in a digestion reaction with either trypsin (A) or proteinase K (B) at 37°C. Aliquotes were removed at various time points. Reactions were stopped by addition of PMSF followed by boiling in SDS sample buffer. Samples were analysed by SDS- PAGE followed by Western blot for Tek/Tie2.

B

A

68

Figure 2.8: Schematic depicting Tek/Tie2 kinase domain and C-terminal tail. In A, the C-terminal tail of Tek/Tie2 loops back on the kinase domain maintaining the receptor in a ‘closed’ like conformation. In B, hypothesised disruption of the ‘closed’ conformation by mutation of Y1111 to D or E resulting in a more ‘open’ receptor.

69

consistent across numerous time points and experiments (Figure 2.7B and data not

shown). Figures 2.7A show results from digestion with trypsin, a protease with increased

cleavage specificity. Again, Y1111 mutants present distinct cleavage patterns to that of

either WT or K853A. While trypsin appears to digest our constructs at a slower rate than

Proteinase K, in both cases it appears that the Y1111 mutants may be cleaved more

readily than WT or K853A. This data further suggests that mutation of Y1111 is

affecting receptor conformation, potentially transforming the receptor into a more ‘open’

conformation making it more accessible to protease digestion.

70

2.5 Discussion

Traditionally, much insight has been gleaned into the mechanism of activation of

RTKs by the solving of the crystal structure of the kinase domain. According to this

classic view, a large, flexible region known as the activation loop functions to inhibit

either substrate or ATP binding sites, thus curbing kinase activity. This inhibition is

relieved by phosphorylation of residues within this loop. Reasons for studying the

kinase domain in isolation of the remaining intracellular portion of the receptor include

the large size of these molecules, their association with the membrane, and the

complexity of the domains encompassed in the full length molecule. These realities have

made it difficult to look at the receptor in its entirety, and thus determine if other regions

may be affecting receptor activation. However, it has come to our attention that a

number of receptors appear to be regulated by regions outside the core catalytic domain.

The solving of the crystal structures for a number of RTKs, including MuSK,

EphB2, FLT3 and c-Kit, has highlighted the importance of the juxtamembrane region in

regulating receptor kinase activity 15,16,181. In all of these cases, the juxtamembrane

region appears to structurally inhibit the activation loop from adopting an active

conformation, thereby stabilizing the receptor in an inactive state. Accumulating

biochemical data also suggests that the C-terminal region of RTKs plays a role in kinase

regulation. In the case of the ErbB2, PDGFRβ and most RON tyrosine kinases,

truncation mutations in the C-terminal region results in an increase in receptor kinase

activation 182,183. In the case of PDFRβ, additional studies utilizing a conformational

specific antibody for the receptor suggest that the truncation mutants assume an 'active'

conformation in the absence of ligand, while full length receptors remained in the

71

inactive state under similar conditions 183. In the case of RON and Met RTKs, peptides

including C-terminal tail sequences were able to inhibit receptor activity, and in the case

of RON, pull down experiments demonstrated a direct interaction between the receptor

and these peptides. All of this points to a mechanism whereby the C-terminal region of

these receptors may contribute to structural regulation of receptor kinase activity.

Similar biochemical evidence now suggests that the Tie2 RTK may also be

regulated by analogous mechanisms. A study conducted by Niu et al, 176 demonstrate

that deletion of residues 1108-1123 (murine sequence) in the Tie2 C-terminal tail gives

rise to a receptor possessing increased kinase activity and signaling. This suggests a role

for this region in negative regulation of receptor activity. How the C-terminus imparts

this regulation, however, remains unclear.

The crystal structure of the Tie2 receptor, including the catalytic core, the kinase

insert domain, and the C-terminal tail has been solved 17. According to this structure, the

inactive Tie2 receptor adopts a somewhat unique conformation as the activation loop

appears to be in an 'active' conformation. Instead, it appears as though regions of the C-

terminal tail may be positioned in such a manner as to block access to the substrate

binding site, suggesting that structural inhibition of Tie2 may be imparted by regions

outside the catalytic core, namely the C-terminal tail. The structure predicts that this

conformation is stabilized in part by interactions that involve specific C-tail residues. Of

these, both Y1100 and Y1111 appear to form hydrogen bonds with residues within the

kinase domain. Furthermore, the phenyl ring of Y1111 appears to be buried in a

hydrophobic pocket. This body of work would suggest that residues such as Y1111

72

play a structural role in helping maintain the receptor in an 'inactive' conformation,

thereby exerting negative regulation.

In this manuscript, we provide biochemical evidence to support the idea that the

C-terminal tail of Tie2 is able to inhibit receptor kinase activity via a structural negative

regulatory mechanism. More specifically, we believe that a single tyrosine residue,

Y1111, plays an important role in this negative regulation. This can be first observed

using the Y1111F mutant which displays increased receptor phosphorylation and kinase

activity. Results from these studies support the hypothesis that important hydrogen

bonds that normally exist between the hydroxl group of Y1111 with the main chain NH

and carbonyl oxygen of L920 and A933 respectively (as shown in the Tie2 crystal

structure) are disrupted 17. We predict that this acts to destabilize the inhibitory

conformation imparted by the receptor C-tail on Tie2 kinase activity.

An alternate theory to describe this increase in receptor phosphorylation would

ascribe a role for Y1111 as a binding site for a negative regulatory molecule such as the

protein tyrosine phosphatase, Shp2 (SHPTP2). In fact, both SH2 domains of Shp2 have

been previously shown to associate with the Tie2 receptor in vitro 71,137. While Y1111

was initially proposed as the binding site for Shp2 on Tie2 137, subsequent binding

experiments revealed the ability of the Shp2 SH2 domains to associate with a number of

additional Tie2 sites in this in vitro context 71In vivo experiments where Shp2 was

overexpressed in Tie2-expressing 293T cells were unable to identify Tie2 as a substrate

for Shp2 phosphatase activity (our unpublished observations). Therefore, even if Y1111

is a true binding site for the Shp2 molecule, it is unlikely that a disruption at this residue

would affect receptor phosphorylation. While the mode of action of Shp2 activation

73

remains somewhat controversial, it has been shown that both engaging Shp2 SH2

domain(s), as well as phosphorylation of specific tyrosine residues within the C-terminal

tail of Shp2, may be responsible for activation of SHP2 catalytic activity (reviewed in

180). We have shown that Shp2 may be a substrate for Tie2 kinase activity as it is able to

be tyrosine phosphorylated downstream of this receptor (Figure 2.4C). Furthermore,

Shp2 phopshorylation appears to increase significantly in the presence of Y1111F,

suggesting that disruption of this site does not appear to impede Shp2 signal transduction

pathways downstream of Tie2. This suggests that disruption of Y1111 may not hinder

the role of Shp2, if there is one, downstream of Tie2. These last two points regarding the

role of Shp2 in Tie2 mediated signaling may indicate an adaptor role for Shp2 in this

case, possibly via an alternate Tie2 residue. Tie2 was identified by mass spectrometry as

a Shp2 binding partner after endothelial cell stimulation with 'flow' again adding another

dimension to the role of this molecule in Tie2 signal transduction 184. The exact functions

of Shp2 in this pathway, obviously, still remain to be determined.

Further evidence to suggest that mutation of Y1111 is doing more than just

disrupting binding of a negative regulatory molecule surfaces upon examination of the

Y1111D and E mutants. We have shown that mutation of Y1111 to a highly negatively

charged residue, such as D or E (Y1111D and Y1111E respectively) significantly

increases receptor phosphorylation and kinase activity, even above that of the Y1111F

mutant. Since Y1111 sits in a hydrophobic pocket according to the Tie2 crystal structure,

we hypothesize that while mutation of Tyr to Phe may disrupt hydrogen bonding with

regions of the kinase domain, moderately affecting Tie2 structure, changing this residue

to a highly negatively charged D or E that may no longer be able to fit in this

74

hydrophobic pocket disrupts receptor conformation even further. If mutation of Y1111

was simply disrupting a negative regulatory molecule, one would expect similar increases

in phosphorylation and kinase activity for all three Y1111 mutants (Y1111F, D and E).

Finally, we have also carried out digestion experiments to compare digestion

patterns of WT, kinase deficient (K853A), and Y1111 mutant receptors (Y1111F,D and

E) in the presence either trypsin or proteinase-K proteases. Figure 2.7 demonstrates the

differences that exist between the WT and K853A receptors versus the Y1111 mutants,

the latter group appearing to produce a smaller digestion fragment at the indicated time

points. These differences were observed across a number of different time points

indicating that the Y1111 mutants may have at least one alternate digestion site. We

speculate that additional digestion sites may be present on the more 'open' conformation

of the Y1111 mutants leading to the observed digestion patterns. Further structural

studies, such as the solving of the crystal structure of one or all of these Y1111 mutants,

would no doubt prove invaluable as to determining the exact changes induced by such

amino acid substitutions.

The Y1111D and E mutants raise an interesting possibility regarding changes

associated with activation of the Tie2 receptor. Previously, Y to E mutations have been

used to mimic pTyr at a particular site since the E residue is highly negatively charged

similar to the negatively charged phosphate of the pTyr 185. If this is true in the case of

Tie2, it would be interesting to determine whether or not Y1111 is a site that is

phosphorylated in vivo. A study involving mass spectrometry analysis was able to

identify both Y992 and Y1106 as phosphorylation sites on Tie2 102 but not 1100 or

Y1111. These studies, however, used in vitro methods to activate the receptor (ie. in the

75

absence of ligand), and may therefore not mimic the true state of the receptor in

endothelial cells. This is of particular interest if one considers the possibility that in

many cases receptor multimerization is not enough for full receptor activation, and that

binding of a receptor’s ligand may induce additional conformational changes that allow

full activation. For this reason, it would be interesting to compare phosphorylation

patterns of both overexpressed and ligand stimulated Tie2 in order to determine if

mutation of Y1111 mimics conformational changes induced by binding of the Tie2

ligand, angiopoietin. Preliminary studies conducted in our lab using angiopoietin-1,

suggest that while WT Tie2 phosphorylation is increased upon stimulation of

angiopoietin-1, phosphorylation of the Y1111 mutants is not affected by addition of

ligand (our unpublished observations). Whether or not our Y1111 mutants mimic full

receptor activation state or not remains to be determined.

Interestingly, naturally occurring activating Tie2 mutations have been identified

in cases of vascular malformations 54,100,101. While the exact mechanism behind how

these mutant receptors function is unclear, it appears that at least one of the mutations, a

substitution of arginine for tryptophan at position 849 in the kinase domain of Tie2

(R849W), results in a hyper-phosphorylated receptor harboring increased kinase activity

100. This mutation has subsequently been shown to protect endothelial cells from

apoptosis via the PI3K/Akt pathway and the authors suggest that the R849W mutation

allows for survival of endothelial cells in a mural-cell poor environment via ligand

independent Tie2 activation 101. It is interesting to note the similarities that exist between

this naturally occurring mutant and the Y1111 constructs both with respect to receptor

activity level and pathway activated. Whether or not the Y1111 occurs in a similar

76

pathological state in vivo, however, remains to be seen. In any case, the present study

provides the first biochemical evidence that mutation of a single C-terminal amino acid

on Tie2 may result in receptor conformational changes which are able to alter Tie2 kinase

activity. Gaining a better understanding surrounding the mechanism of activation of this

RTK will no doubt provide invaluable information into the role of Tie2 in both normal

and pathological situations in which it is involved.

77

CHAPTER 3: GRB 14 TYROSINE

PHOSPHORYLATION IN TIE2 SIGNAL

TRANSDUCTION

Chapter 3 is a modified version of the following manuscript submitted to

BMC Cell Biology:

Tyrosine Phosphorylation of Grb14 – Implications for Tie2 Signaling Celina Sturk 1,2 and Daniel J Dumont * 1,2,3,4

Address: 1Molecular and Cellular Biology Research, Sunnybrook Research Institute, Sunnybrook Health Sciences Centre, Toronto, ON, Canada, 2Department of Medical Biophysics, University of Toronto, ON, Canada, 3Heart & Stroke/Richard Lewar Centre of Excellence, Faculty of Medicine, University of Toronto, Toronto, ON, Canada and 5R. Samuel McLaughlin Centre for Molecular Medicine, Toronto, ON, Canada

e-mail: Celina Sturk sturk@srcl.sunnybrook.utoronto.ca; Daniel J Dumont*

dan.dumont@sri.utoronto.ca

78

3.1 Abstract

Growth factor receptor bound (Grb) proteins 7, 10 and 14 are a family of structurally

related multi-domain adaptor proteins involved in a variety of biological processes.

Grb7, 10 and 14 have been shown to become serine and/or threonine phosphorylated in

response to growth factor (GF) stimulation. Grb7 and 10 have also been shown to

become tyrosine phosphorylated under certain conditions. Grb7 is tyrosine

phosphorylated by the Tie2 angiogenic RTK. Grb14 has also been shown to interact with

Tie2, however tyrosine phosphorylation of this Grb7 family member has yet to be

reported.

Here we report for the first time tyrosine phosphorylation of Grb14. This

phosphorylation appears to require a kinase competent Tie2 as well as involves tyrosines

1100 and 1106 (Y1100 and Y1106) on the receptor. Furthermore, an intact SH2 domain

on Grb14 also appears to be required for Grb14 tyrosine phosphorylation by Tie2. Grb14

was also able to become tyrosine phosphorylated in primary endothelial cells when

treated with a soluble and potent variant of the Tie2 ligand, cartalige oligomeric matrix

protein (COMP) Ang1.

Our results show that Grb14, like its family members Grb7 and Grb10, is able to be

tyrosine phosphorylated. Furthermore, our data indicate a role for Grb14 in endothelial

signaling downstream of the Tie2 receptor.

79

3.2 Introduction

Signal transduction pathways are a series of highly coordinated events involving

numerous proteins, varying in both structure and function. Adaptor proteins play a

pivotal role in such molecular networks by allowing formation of protein complexes via

interactions involving their non-catalytic binding domains. The Growth factor Receptor

Bound (Grb) adaptor proteins are a group of structurally similar proteins beginning to

emerge as key players in a number of cellular functions including cell metabolism, cell

survival and cell migration.

The Grb7 family is made up of three members, Grb7,10 and 14. All family

members harbor an amino-terminal proline rich region (PRR), a putative Ras associating

(RA) domain, a pleckstrin homology (PH) domain and a carboxyl-terminal SH2 domain.

Furthermore, unique to this family of proteins is a novel interaction region, the BPS (for

Between PH and SH2) domain. To date the BPS domain has been shown to play a role in

certain Grb/receptor interactions including those involving the activated IR and IGFR 153.

While all three family members were originally cloned in screens using the EGFR

as bait, as with most other SH2 containing proteins it was quickly discovered that this

family of adaptors binds a number of other receptor and non-receptor proteins. Included

in these is the angiogenic Tie2 receptor tyrosine kinase (RTK). Tie2 is a receptor

tyrosine kinase (RTK) found primarily on the surface of endothelial cells. Mouse

molecular models have shown an essential role for this receptor in aspects of

angiogenesis. Tie2 deficient mice present with a plethora of vascular abnormalities

including a lack of proper sprouting and remodeling of the primitive vessel network 67 70.

80

In the adult, Tie2 has been implicated in both normal and pathological states including

wound healing, follicular development, diabetic retinopathy and tumorigenesis.

Tie2 belongs to the Tie family of proteins. While the only other family member,

Tie/Tie-1, remains an orphan receptor, a family of ligands known as the angiopoietins

(Angs) have been shown to play a role in Tie2 biology. Currently, Ang1 remains the best

characterized of the angiopoietins and is believed to be the main activating ligand for

Tie2. It has been shown to activate the Tie2 receptor and contribute to such functions as

endothelial cell survival and migration 71,113.

Jones et al. initially demonstrated Grb7 and Grb14 interactions with Tie2 in a

yeast two-hybrid screen using mouse heart and lung tissue 71. Overexpression studies

further demonstrated that the Grb7/Tie2 interaction was mediated by a multidocking site,

tyrosine 1100, on Tie2. Furthermore, Grb7 becomes tyrosine phosphorylated in the

presence of a kinase competent Tie2 receptor.

The Grb7 family of proteins have been shown to be phosphorylated on serine,

threonine and tyrosine residues, although the functional significance of these

phosphorylations is still unclear in most cases. Grb7 is phopshorylated on serine and

threonine residues in both quiescent and growth factor stimulated cells, although GF

stimulation does not appear to alter Grb7 phosphorylation state 139,186. Grb7 has also

been shown to be tyrosine phosphorylated in a number of instances including by FAK at

sites of focal adhesion , in the presence of certain growth factors such as EGF, and

ephrinB1 76and in the presence of receptors such as Tie2, Ret and HER2/erbB-2 71,172-175

139.

81

In contrast to Grb7, both Grb10 and Grb14 possess basal serine phosphorylation

which can be further induced by GF stimulation 125,138,140,167. Grb10 has also been shown

to be tyrosine phosphorylated in insulin signaling and in the presence of Tec 141,169,170. To

date, there are no previous reports of Grb14 tyrosine phosphorylation, although it has

been predicted to do so under correct conditions given its interaction with numerous

RTKs and its similar structure to Grb7 and 10.

Here we report for the first time Grb14 tyrosine phosphorylation. This

phosphorylation appears to be dependent on the presence of Tie2 kinase activity and

appears to involve tyrosine residues 1100 and 1106 of the Tie2 RTK as well as an intact

SH2 domain on Grb14.

82

3.3 Materials and Methods

Expression vectors– The cDNAs representing wild type (WT), kinase-inactive (K853A)

and Y1111F Tie2 in pcDNA3.1 have been described previously 120, 177. The cDNA

encoding Grb14 was a kind gift from Roger Daly and flag-tagged full-length Shp2 was a

kind gift from Gen-Sheng Feng and described previously 71. The cDNA encoding the full

length Grb14 140 was used as a template to generate an arginine-to-lysine point mutation

with the QuickChange site-directed mutagenesis kit (Stratagene). The mutation was

confirmed by sequencing. GST fusion proteins were described previously 71.

Production of GST fusion proteins – GST fusion proteins were prepared from

Escherichia coli using standard procedures. Recombinant proteins were immobilized on

glutathione-Sepharose beads (Amersham) at 4°C for 30 minutes and analyzed by SDS-

polyacrylamide gel elctrophoresis followed by Coomassie Blue staining. Bovine serum

albumin standards were used as a comparison to estimate protein concentrarion.

Cell Culture and Antibodies - Human umbilical vein endothelail (HUVEC) cells C166,

SVR, SVEC and HEK293T and HEK293Tie2 (a gift of Fu-Kuen Lin) cells were grown

on 10-cm plates in Dulbecco’s modified Eagle’s medium (DMEM) (Sigma)

supplemented with 10% FBS, 1% penicillin, 1% streptomycin, and 200mM L-glutamine.

EA.hy926 cells were further supplemented with hypoxanthine, aminopterin and

thymidine (Sigma) and HEK293Tie2 cells were further supplemented with 250 mg/ml

G418 (Life Technologies, Inc.), and 100 nM methotrexate

83

(Sigma).

Antibodies used were as follows: polyclonal anti-Tie2 C-20 (Santa Cruz), monoclonal

anti-phosphotyrosine 4G10 (Upstate Biotechnology Inc.), monoclonal anti-Tie2

(Pharmingen), monoclonal anti-Flag M2 (Sigma), polyclonal phospho-Tie2 (pY992)

(Cell Signaling Technology). GST mixing experiments, coimmunoprecipitation

experiments, and western blotting procedures have been previously described 179 71.

Transfection Procedures – HEK293T cells were cultured to ~75% confluency in 10-cm

cell culture plates and transfected with 2g DNA using the lipofectamine reagent

(Invitrogen Life Technologies) and according to manufactures protocol. Cells were

harvested after 48 hours.

In Vitro Kinase Assay – Kinase assays were performed as follows: Cell lysates were

immunoprecipitated with anti-Tie2 antibodies for 2 h at 4 °C. Immunoprecipitates were

washed 3x in PLC lysis buffer (without sodium fluoride or sodium pyrophosphate) 179 and

2x in kinase buffer (2 mM MnCl2 + 50 mM HEPES pH 7.5 + 10 mM MgCl2 + 0.2 mM

dithiothreitol) and then incubated with 4 µg of either GST or GST-Dok-R PH as substrate

including 40 µCi of [ -32P]ATP (Amersham Biosciences), 20 µM ATP (Amersham

Biosciences) for 30 min at 30 °C 122. Kinase reactions were stopped by the addition of 2x

SDS-containing sample buffer and boiled for 10 min. Immunoprecipitates were

electrophoresed and half the gel was used to resolve Tie2 expression by western analysis

using anti-Tie-2 antibodies while the other half was dried and exposed to phosphorimager

analysis and quantification (ImageQuant). The representative values of Tie2 kinase

activity were reflected as the value of GST-Dok-R PH phosphorylation over the amount

84

of immunoprecipitated Tie2 in each of the samples. All experiments were performed

twice or more times with similar results.

85

3.4 Results

Grb14 becomes tyrosine phosphorylated in the presence of Tie2

While both Grb7 and Grb10 have been shown to be tyrosine phosphorylated 166, 172, 174, 175,

71, 141, 168, 170, 139, Grb14 has only been found to become phosphorylated on serine and

threonine residues 125,140. Because Tie2 is a known tyrosine kinase, we set out to

determine whether or not Grb14 could become tyrosine phosphorylated in the presence of

this receptor. HEK293T cells were transfected with Grb14 either alone or with the wild

type receptor (WT) or a kinase inactive mutant of Tie2, K853A. Cell lysates from these

transfections were then subjected to immunoprecipitation using an anti-Grb14 antibody

and blotted for phosphotyrosine (4G10) (Figure 3.1). Grb14 becomes tyrosine

phosphorylated in the presence of the wild type receptor (WT) but not vector or K853A.

This shows for the first time that Grb14 can be tyrosine phosphorylated and in this case,

phosphorylation is dependent on the presence of the Tie2 kinase.

Y1100 and Y1106 on Tie2 are important for Grb14 tyrosine phosphorylation

RTKs function by becoming phosphorylated on tyrosine residues in their

intracellular domain upon activation, often providing binding sites to various proteins

containing phosphotyrosine binding domains. Grb14 contains a C-terminal SH2 domain

and is believed to bind to Tie2 via one of the receptor’s phosphotyrosine residues. We

used a number of Tie2 tyrosine to phenylalanine point mutants in order to determine

whether or not mutation of any of these tyrosine residues on Tie2 altered Grb14

86

Figure 3.1: Grb14 is tyrosine phosphorylated in the presence of Tie2. 293T kidney epithelial cells were transfected with Grb14 either alone or with the wild type receptor (Tie2) or a kinase inactive mutant of Tie2 (K853A). Cell lysates from these transfections were then subjected to immunoprecipitation using an anti-Grb14 antibody and blotted for phospho-tyrosine (4G10) or anti-FLAG to confirm the presence ofGrb14.

Tie2

Tie2

+G

rb14

K85

3A +

Grb

14

Vect

or +

Grb

14Blot: anti- FLAG(M2)

Blot: anti-pTyr

IP: Grb14

pGrb14

Grb14

87

Vect

or +

Grb

14

Tie2

+ G

rb14

K85

3A +

Grb

14

1100

+ G

rb14

1106

+ G

rb14

1111

+ G

rb14

1100

/110

6 +

Grb

14

IP: Grb14

Cell lysate

pGrb14

Grb14

Blot: anti-pTyr

Blot: anti-Grb14

Figure 3.2: Y1100 and Y1106 on Tie2 are important for Grb14 tyrosine phosphorylation. HEK293T cells were transfected with Grb14 alone or in combination with wild type Tie2, the kinase deficient Tie2 (K853A) or one of the Tie2 tyrosine mutants (Y1100F, Y1106F, Y1111F, Y1100/1106F). A)Lysates (100μg) from these cells were immunoprecipitated using an antibody specific for Grb14, run out on an SDS PAGE gel and blotted for phosphotyrosine (top panel). Cell lysates (20μg) were also run out on an SDS PAGE gel and subjected to Western blotting using an anti-Grb14 antibody as a control (bottom panel). B) Lysates were immunoprecipitated with an anti-Tie2 antibody and then either subjected to Western Blot using anti-Tie2 antibodies (top panel) or incubated in a kinase reaction with set amounts of glutothione-Sepharose purified GST-Dok-RΔPH and 32P labeled ATP (bottom panel).

vect

or

Tie2

K85

3A

1100

1106

1100

/110

6

IP: Tie2Blot: α -Tie2

GST-DokR

A

B

88

phosphorylation. 293T cells were transfected with Grb14 alone or in combination with

wild type Tie2, the kinase deficient Tie2 (K853A) or one of the Tie2 tyrosine mutants

(Y1100F, Y1106F, Y1111F, Y1100/1106F). While mutation of any one of the Tie2

tyrosine residues did not appear to alter Grb14 phosphorylation significantly, when both

Y1100 and Y1106 were mutated on the same receptor (Y1100/1106) there was a marked

decrease in Grb14 tyrosine phosphorylation (Figure 3.2A). This suggests that Y1100 and

Y1106 on Tie2 play a role in Grb14 mediated signal transduction downstream of this

receptor.

Previous studies have shown that mutation of receptor tyrosine residues to

phenylalanine (such the Tie2 mutants) can affect kinase activity of the receptor Mori,

1993}. This led us to ask the question 'is the observed decrease in Grb14

phosphorylation in the presence of the Y1100/1106F mutant was the result of decreased

kinase activity of this mutant receptor?'. To this end we carried out a kinase assay to

compare the kinase activity of the Y1100F, Y1106F and Y1100/1106F mutants to that of

the wild type receptor using a GST Dok-R fusion as substrate. Dok-R is a docking

protein that has been shown to be tyrosine phosphorylated downstream of Tie2 signal

transduction in vivo making this a biologically relevant choice for a substrate. The DokR

construct used herein lacks the insoluble PH domain to allow for purification purposes

and will be subsequently referred to as GST-Dok-RΔPH (described previously 179).

Tie2 receptor was immunoprecipitated from 293T cells expressing Tie2WT, K853A or

one of the Tie2 mutants (Y1100F Y1106F and Y1100/1106F) and incubated in a kinase

reaction with set amounts of glutathione-Sepharose purified GST-Dok-RΔPH and 32P

labeled ATP. Figure 3.2B is a representative blot of two separate experiments

89

demonstrating ability of Y1100/1106F to phosphorylate GST-Dok-RΔPH. These data

show that while there appears to be a slight decrease in Y1100/1106F kinase activity

when compared to Tie2WT, Y1100/1106F activity is comparable to that of the Y1106F

mutant. Recall from Figure 3.2A that Grb14 was phosphorylated in the presence of

Y1106F mutant to similar levels as when in the presence of the wild type receptor

indicating that the level of kinase activity of the Y1106 mutant is adequate for

phosphorylation of Grb14. By inference, therefore, we believe it is unlikely that the

decrease in kinase activity seen here for Y1100/1106F is the cause of the dramatic

decrease in Grb14 tyrosine phosphorylation observed in our overexpression studies.

Grb14 SH2 domain is important for Grb14 tyrosine phosphorylation

Both SH2 and BPS domains of the Grb7 family members have been shown to mediate

interactions with various receptor and non-receptor proteins. In order to investigate

whether or not the SH2 domain of Grb14 is able to bind Tie2 we performed an in vitro

pull down assay using the full length Tie2 receptor and the SH2 domain of Grb14. Grb14

and Grb2 (as a positive control) SH2 domains were purified from E. coli and

immobilized on GST beads. GST-SH2 domains were then incubated with lysates from

HEK293T cells transiently expressing either WT Tie2 or the kinase inactive K853A

mutant. The SH2 domains of both Grb2 and Grb14 were both able to pull down WT

Tie2, but not K853A (3.3A) indicating the ability of the Gr14 SH2 domain to interact

with Tie2. Cell lysate from transfected cells was run out to confirm transfection and

protein expression in 293T cells.

90

Figure 3.3: The SH2 domain is important for Grb14 tyrosine phosphorylation. In A), HEK 293T cells were transfected with Tie2, either WT or the kinase inactive K853A mutant. Lysateswere incubated with immobilized GST alone, GST-SH2 domains from Grb2 (control) or Grb14 (as indicated). Resulting complexes were resolved by SDS-PAGE and immunoblotted with anti-Tie2 antibodies. In B, HEK 293T cells were co-transfected with Tie2 (WT or kinase inactive) and FLAG tagged Grb14 (WT or the FLVRS mutant). Lysates subjected to immunoprecipitationby either Tie2 or FLAG antibodies and resolved by SDS-PAGE. Western Blotting with indicated antibodies was then carried out.

GST

GST

-SH

2 G

rb2

GST

-SH

2 G

rb14

Cel

l Lys

ate

A

B

K853A

Tie2

Blot: anti-Tie2

K85

3A

Tie2

IP: Tie2

Blot: pTyr

Blot: anti-Tie2

Blot: pTyr

Blot: anti-FLAG

K85

3A

Tie2

IP: FLAG

Grb14 FLVRS

91

Mutation of the SH2 domain FLVRS motif has been shown to disrupt SH2 domain

function. We generated an Arg to Lys (R646K) mutation in the Grb14 SH2 domain

within this conserved FLVRDS motif in order to determine the effect of disrupting Grb14

SH2 function on Grb14 tyrosine phosphorylation. HEk293T cells were transfected with

WT Tie2 or the K853A mutant along with either WT Grb14 or our Grb14FLVKDS

mutant. Lysates from these cells were immunoprecipitated with anti-FLAG antibody and

subjected to Western blot using a phosphotyrosine specific antibody (4G10) (Figure

3.3B). Figure 3.3B shows that mutation of the Grb14 FLVRDS motif reduced tyrosine

phosphorylation of Grb14 suggesting the functionality of this domain must be intact for

Tie2 mediated Grb14 tyrosine phosphorylation. Blots were stripped and reprobed using

anti- Tie2 and anti-FLAG antibodies as controls. Together these results suggest a role for

the Grb14 SH2 domain in Tie2 mediated Grb14 signaling.

Grb14 phosphorylation in endothelial cells

A yeast-two hybrid screen using a cDNA library derived from murine embryonic heart

and lung tissue and the intracellular portion of the Tie2 receptor initially identified Grb14

as a binding partner of Tie2 71. To date, Tie2 has been shown to be found predominantly

in endothelial cells. In order to verify that Grb14 is a potentially biologically relevant

player in the Tie2 signaling pathway, we examined Grb14 expression in a number of

endothelial cell lines. Four murine (C166, EOMA, SVR and SVEC) and one human

(EAhy92.6) endothelial cell lines as well as an HEK293 cell line that stably expresses the

full-length Tie2 receptor (HEK293Tie2) were immunoprecipitated and then subjected to

western blotting using an antibody specific for Grb14 (Figure 3.4A). Cell lysate from

92

Figure 3.4: Grb14 in endothelial cells. Lysates from C166, EOMA, SVR, SVEC and EAhy92.6 endothelial cell lines as well as an HEK293 cell line that stably expresses the full-length Tie2 receptor (HEK293Tie2) were immunoprecipitated and then subjected to Western Blotting using an antibody specific for Grb14 (A). Cell lysate from 293T cells overexpressingGrb14 was also run as a positive control and blotted for Grb14. In B) Human umbilical endothelial cells (HUVEC) were either left untreated or stimulated with a soluble and potent variant of the Tie2 ligand, cartalige oligomeric matrix protein (COMP) Ang1. Cell lysates were immunoprecipitated with anti-Grb14 and subjected to SDS PAGE and Western blot for phosphotyrosine. Cell lysate from 293T cells transfected with the Tie2 construct Y1111F and Grb14 was run as a positive control for Grb14 phosphorylation

Posi

tive

Con

trol

pGrb14

Uns

timul

ated

CO

MP

Ang1

Blot: pTyr

293T

+ G

rb14

HEK

293T

ie2

C16

6

SVR

EAhy

92.6

EOM

A

SVEC

Grb14IgG

Blot: α -Grb14

IP: a-Grb14CLA

B

50

kDa

61

IP: anti-Grb14

93

293T cells overexpressing Grb14 was also run as a positive control and blotted for

Grb14. The results show that Grb14 is present in three of the five endothelial cell lines

examined (EOMA, SVEC and EAhy92.6). These results were confirmed by RT-PCR

(data not shown). This data demonstrates that Grb14 is in fact present in endothelial cells

and therefore is a reasonable candidate for a player in the Tie2 signaling pathway.

Angiopoietin-1 (Ang1) is a known ligand of Tie2 in endothelial cells inducing a

number of the Tie2 pathways. In order to verify whether or not endogenous Grb14 is

able to be tyrosine phosphorylated in endothelial cells, we stimulated HUVEC cells with

a soluble and potent variant of the Tie2 ligand, cartilage oligomeric matrix protein

(COMP)Ang1 133. Cell lysate from either unstimulated cells or cells treated with COMP-

Ang1 was run on a gel and probed for phosphotyrosine. Lysate from 293T cells

transfected with Y1111F and Grb14 was run as a positive control for Grb14

phosphorylation. Figure 3.4B shows a tyrosine phosphorylated protein running at the

same size as our positive control in the COMP-Ang1 sitmulated lane, but to a much lesser

extent in the unstimulated sample. This data suggests that Grb14 is phosphorylated in

endothelial cells upon stimulation with the Ang1 variant, COMP-Ang1.

94

3.5 Discussion

Although the Grb7 family of proteins was described for the first time over a

decade ago, still much has to be learned about the role of these adaptors in various

signaling pathways. Most of what we know about these proteins has come from binding

studies and elucidation of the various domains involved in these interactions. Grb7, 10

and 14 have been shown to bind a number of receptors and intracellular proteins

primarily via SH2 mediated interactions, although the importance of the BPS domain is

also becoming apparent.

Grb14 was previously identified in a yeast two-hybrid screen as a binding partner

for the endothelial receptor Tie2 71. Tie2 is an angiogenic RTK involved in numerous

aspects of endothelial biology such as cell migration, cell survival and tubule formation.

We have now been able to show that Grb14 is endogenous to endothelial cells further

supporting a role for Grb14 in endothelial cell signaling. Furthermore, the results

reported herein are the first to describe tyrosine phosphorylation of Grb14. This

phosphorylation appears to require a kinase competent Tie2 and tyrosines 1100 and 1106

on the receptor.

Phosphorylation is the most widespread post-translational protein modification in

cell signaling 187. Phosphorylation of a protein can modulate its behavior in a multitude

of ways including its function, localization, half life and binding to other molecules 188.

In eukaryotes, phosphorylation is typically ‘carried out’ on serine, threonine or tyrosine

residues.

Since their discovery, Grb7 family members have been shown to become

phosphorylated under a number of conditions. The role of this phosphorylation,

95

however, remains somewhat enigmatic. All family memebers (Grb7, 10 and 14) have

been shown to possess both basal and growth factor induced serine phosphorylation

(reviewed in 158). Grb10 and Grb7 have also been shown to be tyrosine phosphorylated.

More specifically, Grb10 and Grb7 tyrosine phosphorylation has been previously

implicated in signaling downstream of endothelial receptors. Grb10 has been shown to

be tyrosine phosphorylated in endothelial cells in response to VEGF 171. Our lab has also

previously shown that Grb7 becomes tyrosine phosphorylated in the presence of the

endothelial receptor Tie2 71. Taken together, this data supports a role for Grb proteins in

endothelial biology.

The current study suggests that Grb14 tyrosine phosphorylation requires a kinase

competent Tie2 receptor. In a separate experiment, Grb14 was not tyrosine

phosphorylated in response to EGF stimulation suggesting some specificity for the Tie2

kinase (data not shown). Whether or not Grb14 is a direct substrate of the Tie2 kinase

activity, or rather is phosphorylated by an alternate kinase recruited to this receptor,

remains to be seen.

Grb proteins do not possess intrinsic kinase activity, but have been shown to bind

a number of receptor and non-receptor kinases (reviewed in 158,164,186). In insulin

signaling, Grb10 serine phosphorylation appears to involve the PI3K and MAPK

signaling pathways, while PKCζ seems to play a role in Grb14 phosphorylation 155,167.

Grb proteins have been found to bind a number of tyrosine kinases (reviewed in Holt and

Siddle, 2005 158). Interestingly, in the case of insulin signaling, although Grb10 and

Grb14 bind the IR, they do not appear to be direct substrates for the IR tyrosine kinase

activity 124,169. Instead, at least with respect to Grb10, Src/Fyn kinases were suggested to

96

be responsible. In general, how and why Grb proteins are phosphorylated remains to be

determined in most cases.

In the studies detailed in this report, Grb14 tyrosine phosphorylation was

abolished upon mutation of Y1100 and Y1106 to phenylalanine in the receptor double

mutant Tie21100/1106, suggesting these two residues may play a role in Grb14 tyrosine

phosphorylation downstream of Tie2. This should perhaps not be surprising given that

there is emerging evidence that the Grb7 family members may bind their target proteins

via two separate regions, the SH2 and BPS domains (Reviewed by Holt and Siddle, 2005

158).

This is particularly evident in the case of insulin signaling where both SH2 and

BPS domains of Grbs 7, 10 and 14 have been shown to bind the IR (reviewed in 158).

Specifically, structural studies have shown that the Grb14 SH2 and BPS domains bind

phosphorylated tyrosine residues within the IR activation loop 157. Whether tyrosines

1100 and 1106 bind Grb14 directly or play a more indirect role in Grb14 tyrosine

phosphorylation, however, remains to be determined.

It is interesting to note that in comparison with the other Grb7 family members,

Grb14 has been shown to interact with a relatively small number of receptors. In

accordance with this observation, we have found in vivo binding studies of Grb14 and

Tie2 have been particularly challenging (data not shown). This may be explained, at least

in part, by structural studies which have shown that Grb14 may have more difficulty

binding to phosphotyrosine containing ligands due to the presence of a non-glycyl residue

at the end of the BC loop and the lack of a P+3 binding pocket in the SH2 domain 189.

97

Further analysis of Grb14 tyrosine phosphorylation will no doubt provide

information that may help elucidate the role of this protein in Tie2 signaling. Mapping

the tyrosines on Grb14 which become tyrosine phosphorylated may give us a clue as to

what other proteins bind and are involved in Grb14 signaling. Tyr67 on Grb10 was

identified as major site of phosphorylation in response to insulin signaling. However,

this site is not conserved in Grb7 and 14. Interestingly, mutation of this site increased the

affinity of Grb10 for IR. This raises the possibility that tyrosine phosphorylation may be

involved in terminating Grb signaling at the receptor level. Alternatively, it may suggest

that Grb tyrosine phosphorylation recruits this adaptor for involvement in non-receptor

mediated signaling pathways. Further understanding of these sorts of post-translational

modifications seen in the Grb7 family will no doubt shed considerable insight into their

biological role.

98

CHAPTER 4: DISCUSSION AND FUTURE DIRECTIONS

99

This study set out to further characterize Tie2 signal transduction. We have

identified Y1111 on the receptor as an important regulator of Tie2 activity. Y1111

appears to play a negative regulatory role on Tie2 via structural mechanism(s).

Furthermore, mutation of this site results in increased phosphorylation of a number of

signaling molecules including the adaptor protein Grb14. While Grb14 has been shown

previously to be serine and threonine phosphorylated, this is the first report of Grb14

tyrosine phosphorylation. While the exact role of Grb14 in Tie2 signaling remains to be

determined, we believe Y1100 and Y1106 on the receptor are important regulators of this

pathway. A better understanding Tie2 regulation and signal transduction will hopefully

shed light on both normal and pathological conditions associated with this complex

receptor system.

4.1 Negative regulation of Tie2

Regulation of RTKs can occur via a number of different mechanisms. In the

course of our studies we have identified a negative regulatory site on the Tie2 RTK,

Y1111. Mutation of this site increases receptor phosphorylation and kinase activity and

appears to affect downstream signaling pathways.

Phosphatases often play an important role in modulating RTKs such as been

shown in the case of EGFR and PDGFR (reviewed in 30). Interestingly, treatment with

the phosphatase inhibitor pervanadate results in an increase in Tie2 phosphorylation (see

appendix 1). This suggests the possibility that Tie2 may be regulated by a phosphatase.

100

One of the initial proteins shown to interact with Tie2 in vitro was the PTP Shp2

8. This interaction was believed to occur via the N-terminal SH2 domain of Shp2 and

tyrosine 1111 on Tie2 137 179. We therefore initially postulated that the increase in

receptor phosphorylation seen when Y1111 was mutated on Tie2 may be due to

disruption of the Shp2 binding site. However, our data indicates that there is no visible

decrease in Tie2 phosphorylation when this receptor is overexpressed with Shp2 .

Therefore, while Shp2 may bind Tie2, the receptor may not be a substrate of Shp2’s

phosphatase activity. Interestingly, Shp2 phosphorylation appears to be increased when

the Y1111 binding site is disrupted (Chapter 2, figure 2.4 upper panel) which raises

questions surrounding the initial claims made surrounding Y1111 as a Shp2 binding site.

One explanation for these results is that Shp2 binds an alternate site (other than

Y1111) and acts as adaptor linking Tie2 to other molecules and pathways. Shp2 has in

fact been shown in a number of cases to positively regulate RTK signaling by interacting

with RTKs (reviewed in 180). In a study, Shp2 was shown to interact with Tie2 in

response to flow 184. It is possible that a role for Shp2 in Tie2 signaling may be very

situation and stimuli specific . In any case, Shp2 does not appear to negatively regulate

Tie2 signal transduction through Y1111 under the conditions used in our studies.

Another candidate negative regulatory molecule in Tie2 signaling is VE-PTP, a

receptor-type PTP expressed exclusively in blood vessel endothelial cells. When

overexpressed in Cos cells, VE-PTP was able to associate with Tie2 resulting in a

decrease in receptor tyrosine phosphorylation indicating that Tie2 may be a substrate for

VE-PTP phosphatase activity 136. While mouse molecular models suggest VE-PTP is

involved in vascular maintenance and remodeling 190, the VE-PTP KO has a similar

101

phenotype to the Tie2 and Ang1 knock outs. This result was somewhat puzzling since

one would expect the opposite phenotype if in fact VE-PTP was negatively regulating

Tie2 function 191. It is obvious that more studies are needed to understand the true role of

VE-PTP in Tie2 signaling. In any case, if this phosphatase is in fact negatively

regulating Tie2 phosphorylation, it is most likely not acting directly via tyrosine 1111

since VE-PTP is a transmembrane receptor-like protein (and does not harbor SH2 or PTB

domains).

While there is the possibility that an as of yet unidentified negative regulatory

molecule binds Y1111 on Tie2, emerging evidence suggests an alternate role for this

residue. A study conducted by Niu et al, 176demonstrates that deletion of residues 1108-

1123 (murine sequence) in the Tie2 C-terminal tail gives rise to a receptor possessing

increased kinase activity and signaling. This corroborated speculations that came from

the solving of the Tie2 crystal structure which suggested that the C-terminal tail may be

playing an inhibitory role in Tie2 activation 17. According to this structure, the inactive

Tie2 receptor adopts a conformation such that the activation loop appears to be in an

'active' conformation. Instead, regions of the C-terminal tail appear responsible for the

‘inactive’ state and are positioned in such a manner as to block access to the substrate

binding site. The structure predicts that this conformation is stabilized in part by

interactions that involve specific C-tail residues. Of these, both Y1100 and Y1111

appear to form hydrogen bonds with residues within the kinase domain. Furthermore, the

phenyl ring of Y1111 appears to be buried in a hydrophobic pocket. Together, this body

of work suggests that Y1111 plays a structural role in helping maintain the receptor in an

'inactive' conformation, thereby exerting negative regulation.

102

There is emerging evidence indicating that this is not such an unusual mechanism

of RTK regulation. The solving of the crystal structures for a number of RTKs,

including MuSK, EphB2, FLT3 and c-Kit, has highlighted the importance of the

juxtamembrane region in regulating receptor kinase activity 15,16,181. In all of these cases,

the juxtamembrane region appears to structurally inhibit the activation loop from

adopting an active conformation, thereby stabilizing the receptor in an inactive state.

Accumulating biochemical data also suggests that the C-terminal region of RTKs plays a

role in kinase regulation. In the case of the ErbB2, PDGFRβ and most RON tyrosine

kinases, truncation mutations in the C-terminal region results in an increase in receptor

kinase activation 182,183.

Our studies support the theory that Y1111 is more than a binding site for a negative

regulatory molecule, specifically when one considers the Y1111D and E mutants. We

have shown that mutation of Y1111 to a highly negatively charged residue, such as D or

E (Y1111D and Y1111E respectively) significantly increases receptor phosphorylation

and kinase activity, even above that of the Y1111F mutant (see Chapter 2, Figure 2.5 and

2.6). Since Y1111 sits in a hydrophobic pocket according to the Tie2 crystal structure,

we hypothesize that while mutation of Tyr to Phe may disrupt hydrogen bonding with

regions of the kinase domain, moderately affecting Tie2 structure, changing this residue

to a highly negatively charged D or E that may no longer be able to fit in this

hydrophobic pocket disrupts receptor conformation even further. If mutation of Y1111

was simply disrupting a negative regulatory molecule, one would expect similar increases

in phosphorylation and kinase activity for all three Y1111 mutants (Y1111F, D and E).

103

We have also compared digestion patterns of Tie2 and the Y1111 mutants. The

differences noticed in Y1111 mutant digestion patterns when compared to WT and

K853A receptors indicate that the Y1111 mutants may have at least one alternate

digestion site. One explanation for the presence the additional digestion site(s) is that the

receptor conformation created by mutating Y1111 opens up the C-terminal tail of Tie2

thereby exposing the additional site(s). Further structural studies, such as the solving of

the crystal structure of one or all of these Y1111 mutants, would no doubt prove

invaluable as to determining the exact changes induced by such amino acid substitutions.

Our data, however, lends further support to a model in which Y1111 plays a structural

regulatory role on Tie2.

One theory is that these alternate regulatory mechanisms are in place to prevent

undesirable (ie.ligand independent) RTK kinase activity which could result in severe, and

possibly undesirable, phenotypes (eg. from random meeting of receptors at cell surface

etc.). In fact it seems logical to have numerous, and sometimes overlapping, control

mechanisms in place in the case of RTKs such as Tie2 which plays such a critical role in

vascular biology. Many pathological states are in fact characterized by deregulated

vessel growth or regression.

Naturally occurring activating Tie2 mutations have been identified to date in

cases of vascular malformations 100, 54. Specifically, an autosomal dominant mutation has

been found in three unrelated families which segregates with a C to T transition at 2545

in Tie2 resulting in an argenine for tryptophan substitution (R849W) 54,100. While the

exact mechanism(s) behind how these mutant receptors result in such a phenotype is

unclear, it is interesting to note the similarities that exist between this naturally occurring

104

mutant and our Y1111 constructs. First, overexpression studies show R849W has

increased phopshorylation over WT Tie2 100. Second, R849W displays increased kinase

activity compared to WT Tie2 54,100,101. Finally, transfection of R849W into endothelial

cells results in an increase in phosphorylation of endothelial cell proteins 101. Recall that

the Y1111 mutants displayed increases in both tyrosine phosphorylation and kinase

activity when compare to WT Tie2. Cell lysates of cells overexpressing the Y1111

mutants also appeared to harbor an increased number of tyrosine phosphorylated proteins.

These mutants also resulted in an increase in tyrosine phosphorylation of specific

downstream signaling molecules such as p85,Shp2 and Grb14 (see below for further

discussion)

Despite these paralleles, whether or not the Y1111 mutation occurs in a similar

pathological state in vivo, however, remains to be seen. Interestingly, a study by Morris

et al. demonstrated that the R849W mutant Tie2 is constitutively phosphorylated in

HUVEC cells (ie. in the absence of ligand) while WT Tie2 requires ligand stimulation for

phosphorylation 101. The hypothesis put forth is that in vivo WT Tie2 is phosphorylated

by Ang1 produced by vessel support cells, but that the R849W mutant receptor is

phosphorylated even in the absence of ligand (and therefore support cells) resulting in the

venous malformations that are characterized by a disproportionate ratio of endothelial to

support cells. It would be interesting to see if the Y1111 mutants are also able to undergo

ligand independent phopshorylation in endothelial cells.

The Y1111D and E mutants also raise interesting questions regarding changes

associated with activation of the Tie2 receptor. Previously, Y to E mutations have been

used to mimic pTyr at a particular site since the E residue is highly negatively charged

105

similar to the negatively charged phosphate of the pTyr 185. If this is true in the case of

Tie2, it would be interesting to determine whether or not Y1111 is a site that is

phosphorylated in vivo. A study involving mass spectrometry analysis was able to

identify both Y992 and Y1106 as phosphorylation sites on Tie2 102 but not 1100 or

Y1111. These studies, however, used in vitro methods to activate the receptor (ie. in the

absence of ligand), and may therefore not mimic the true state of the receptor in

endothelial cells. This is of particular interest if one considers the possibility that in

many cases receptor multimerization is not enough for full receptor activation, and that

binding of a receptor’s ligand may induce additional conformational changes that allow

full activation. For this reason, it would be interesting to compare phosphorylation

patterns of both overexpressed and ligand stimulated Tie2 in order to determine if

mutation of Y1111 mimics conformational changes induced by binding of the Tie2

ligand, angiopoietin. Preliminary studies conducted in our lab using angiopoietin-1,

however, have failed to show that this is in fact the case (data not shown) and whether or

not other members of the angiopoietin family are able to do so remains to be determine.

While there is still much to be learnt about Tie2 regulation in vivo, the 1111

mutants raise some interesting possibilities regarding receptor activation and control. On

one hand it provides further support to the notion that this receptor is subject to structural

regulation via the C-terminal tail. Furthermore, given the similarities between the 1111

mutants and the naturally occurring Tie2 mutants, it may help explain the mechanism of

‘hyperactivation’ in Tie2 of VMS.

4.2 Grb proteins and Tie2 signaling

106

Previously in our lab, Grb7 and Grb14 were identified as Tie2 binding partners in

a yeast two-hybrid study 71. Grb7 and Grb14, along with a third family member, Grb10,

make up the Grb7 family of proteins. While the discovery of these proteins occurred

over a decade ago, elucidation of their role in various signaling pathways has been slow.

Our work suggests the Grb proteins may play a role downstream of Tie2. We have also

shown for the first time that Grb14 is tyrosine phosphorylated.

4.2.1 Grb7 family phosphorylation

Grb 7, 10 and 14 have been shown to become phosphorylated under a number of

conditions. The role of this phosphorylation, however, remains somewhat enigmatic.

Grb7 family members have been shown to be phosphorylated on serine/threonine

and/or tyrosine residues. Grb10 and 14 have been shown to possess both basal and

growth factor induced serine phopshorylation (reviewed in 158). In insulin signaling,

Grb10 serine phosphorylation appears to involve the PI3K and MAPK signaling

pathways, while PKCζ seems to play a role in Grb14 phosphorylation 167 156.

Grb 7 and 10 have also been shown to be tyrosine phosphorylated. More

specifically, Grb10 has been shown to be tyrosine phosphorylated in endothelial cells in

response to VEGF 171. Our lab has previously shown by overexpression studies in 293T

cells that Grb7 becomes tyrosine phosphorylated in the presence of the endothelial

receptor Tie2 71.

Unfortunately, examining the role of Grb7 in endothelial cell signaling has been

hampered by our inability to detect Grb7 in this cell type (data not shown). One

explanation for our failure to show Grb7 in endothelial cells may have been the lack of

107

sensitive enough reagents to detect the lower ‘endogenous’ levels of this protein.

Another interesting possibility is that the Grb7/Tie2 interaction occurs in non-endothelial

cell type(s) as additional studies have shown that Tie2 expression is not necessarily

restricted to endothelial cells as was once previously believed 52-54,56,57,192. It is possible,

therefore, that Grb7 is involved in Tie2 signaling pathways in these alternate cell types.

Grb14, however, appears to be a likely candidate for Tie2 signaling in endothelial

cells as we have been able to show by RT-PCR and Western analysis that Grb14 is found

in a number of endothelial cell lines as well as the primary endothelial cell line, HUVEC

(Chapter 3, Figure 3.4). Furthermore, we have shown for the first time that Grb14 can

become tyrosine phosphorylated. This phosphorylation appears to require a kinase

competent Tie2 receptor (Chapter 3, Figure 3.1). In a separate experiment, Grb14 was

not phosphorylated in the presence of activated EGFR suggesting some specificity of

Tie2 kinase (data not shown). Furthermore, Grb14 tyrosine phosphorylation was

abolished upon mutation of Y1100 and Y1106 to phenylalanine in the receptor double

mutant Tie21100/1106, suggesting these two residues may play a role in Grb14 tyrosine

phosphorylation downsream of Tie2. Interestingly, none of the Tie2 single mutants

appeared to have a significant diminishing effect on Grb14 tyrosine phosphorylation

when compared to the WT.

It is interesting to examine the conditions under which Grb14 becomes tyrosine

phosphorylated. First, Grb14 tyrosine phosphorylation was robust in the presence of the

WT Tie2 when cells had been treated with the phosphatase inhibitor pervanadate. One

reason that pervanadate may be required to see this phosphorylation is that the Tie2-

Grb14 interaction is somewhat weak and/or transient and is somehow stabilized by

108

pervanadate treatment (see below for further discussion). Pervanadate may also relieve

Grb14 dephosphorylation imparted by an as of yet unkown phosphatase. Also a

possibility is that overexpression of Tie2 is not sufficient for full receptor activation as

was previously believed and the use of pervanadate is required for complete receptor

activation in this setting, and thus for Grb14 phosphorylation. It is interesting to note that

the use of phosphatase inhibitors was also used to show Grb10 tyrosine phosphorylation

in insulin signaling 168.

The other situation in which Grb14 is robustly phosphorylated is in the presence of

the 1111 mutants. As we discussed previously, it is possible that the 1111 mutants

themselves represent a more ‘fully activated’ state of the receptor. Also possible is that

the 1111 mutants are a ‘hyperactivated’ Tie2 receptor similar to those seen in venous

malformations and Grb14 represents a pathway unique to this sort of pathological state.

It is interesting to note that another Tie2 binding partner, Shp2, appears to be

phosphorylated under the exact same conditions as Grb14, while the well studied

downstream signaling protein, DokR, does not (appendix 2 and data not shown). It is

possible that both Shp2 and Grb14 are involved in a common pathway downstream of

Tie2. Alternatively, they could both be involved in signaling in specific pathological

situations, as may be the case of ShcA in venous malformations.

Further analysis of Grb14 tyrosine phosphorylation will no doubt provide

information that may help elucidate the role of this protein in Tie2 signaling. For

example, mapping the tyrosines on Grb14 which become tyrosine phosphorylated may

give us a clue as to what other proteins bind and are involved in Grb14 signaling. While

we have been able to isolate Grb14 by silver staining (appendix 3), our attempts at

109

mapping these phosphorylated sites on Grb14 were unsuccessful, more sophisticated

methods of isolating phosphorylated proteins will not doubt be invaluable in this

endeavor.

One area of study that has made progress in Grb signaling is in the context of

insulin signaling. Grb10 and 14 both bind IR, but they do not appear to be substrates for

the receptors kinase activity. Instead, in the case of Grb10, tyrosine phosphorylation was

proposed to be regulated by Src/Fyn 169. In this study Tyr67 on Grb10 was identified as

major site of phosphorylation. However, this site is not conserved in Grb7 and 14.

Interestingly, mutation of this site increased affinity of Grb10 for IR. This raises the

possibility that tyrosine phosphorylation may be involved in terminating Grb signaling at

the receptor level. Alternatively, it may suggest that Grb tyrosine phosphorylation

recruits this adaptor for involvement in non-receptor mediated signaling pathways.

Further understanding of these sorts of post-translational modifications seen in the Grb7

family will no doubt shed considerable insight into their biological role.

4.2.2 Grb/Tie2 interactions

The Grb7 family members share a highly conserved multidomain structure

comprised of an amino-terminal proline rich region (PRR), a central segment called the

GM region (Grb and Mig) and a carboxyl-terminal SH2 domain (See Chapter 1, Figure

1.5) (see Introduction for a more detailed description of these domains). Both the SH2

and BPS domains have been shown to affect binding of the Grb proteins to RTKs. While

it was initially believed that Grb7 bound to Tie2 via its SH2 domain (since that was the

portion of the protein recovered in the yeast two-hybrid screen) 71, we have shown that

110

disruption of the Grb7 SH2 domain by mutation of the FLVRES sequence does not

completely abolish binding of Grb7 to Tie2 (appendix 4). This suggests that regions

other than the SH2 domain on Grb7 may be involved in binding the receptor. In insulin

signaling, both SH2 and BPS domains have been shown to contribute of binding of Grb7

to the IR 163. Further studies are needed to determine whether or not this is also the case

in the Grb7 interaction. Our attempts at isolating the domains proved unsuccessful as

many of the truncated proteins appears to be degraded when expressed in mammalian

cells (data not shown).

Grb7 has also been shown to bind non-receptor type proteins such as Shc 139 186,193

and Shp2 194via its SH2 domain. Our lab has previously shown that Grb7 also interacts in

a complex with Tie2 and two unidentified tyrosine phosphorylated proteins: pp70 and

pp85 71. Interestingly, we have now been able to show that pp70 and pp85 are no longer

present in the Grb7/Tie2 complex when the SH2 domain is disrupted (appendix 4). This

suggests that the Grb7 SH2 domain could also be involved in binding pp70 and pp85 in

Tie2 signaling. Also of interest would be to determine the identity of pp70 and pp85 as

they may be a clue in helping us determine what pathways lie downstream of Grb7 in this

and potentially other signaling systems.

Both the SH2 and BPS domains have also been shown to play a role in Grb14

mediated interactions 124,125,159,195. Interestingly, we have shown through preliminary

studies that mutation of Grb14 FLVRDS motif disrupts Grb14 tyrosine phosphorylation

in the presence of Tie2 (appendix 5 and data not shown). This suggests the SH2 domain

plays a role in Grb14 tyrosine phosphorylation. This is in accordance with data from

111

Jones et al. (1999) which isolated the Grb14 SH2 domain in the Tie2 yeast two-hybrid

screen.

Grb7 appears to have a broad binding specificity 164,165 imparted by binding of its

SH2 domain to pYXN motifs found in a number of RTKs and other tyrosine

phosphorylated proteins 139 186,193. Our lab has previously shown that Grb7 binds to Tie2

via the receptors multidocking site at Y1100 71.

Unlike Grb7, however, Grb14 appears to interact with a select number of

receptors. In fact, structural and biochemical studies have suggested that both Grb10 and

Grb14 may have difficulty binding to phosphotyrosine containing ligands due to the

presence of a non-glycyl residue at the end of the BC loop and the lack of a P+3 binding

pocket 196.

We have previously shown that Grb14 interacts with Tie2 in a number of pull

down assays 71. Co-immunoprecipitation studies between Tie2 and Grb14, however,

have thus far remained unsuccessful. Trouble co-immunoprecipitating this adaptor with

receptors appears to be a common difficulty among researchers in the field (Roger Daly -

personal communication). This may be in part due to some of the structural reasons

mentioned above. Furthermore, Grb 14 has been shown to exist in a dimeric form in

solution and structural data suggests that this dimer formation occludes phosphotyrosine

binding since the phosphate binding pocket is occluded by the other protomer 197.

Therefore, the interaction between Tie2 and Grb14 may be weak or transient and

therefore difficult to visualize by co-IP.

Also a possibility is that Grb14 may compete with other molecules binding to

Tie2. In insulin signaling, overexpression of either Grb10 or Grb14 inhibits

112

phosphorylation of a number of molecules including IRS-1, IRS-2, Shc and p62Dok 9. It

is also possible that the Grb14/Tie2 interaction is not direct since Grb14 has been shown

to bind other Tie2 binding molecules such as Shp2 and Shc.

Another possibility is that Grb14 is a negative regulator of Tie2 signaling. In

general, the mechanisms surrounding Tie2 negative regulation have not been well

documented. A study by Bogdanovic et al., 129 suggests that there may be a ligand

mediated pathway that served to negatively regulate Tie2 signaling. Our unpublished

data also suggests that Tie2 may be ubiquitinated (appendix 6). Interestingly, yeast two-

hybrid studies have found that Grb 10 and 14 interact with Nedd4, an E3 ubiquitin

protein ligase 22,143. Grb10 has also been suggested to promote IGFR and IR degradation,

presumably via a ubiquitin/proteosomal degradation pathway 162,189.

It is evident that there is still much to be learnt about the role the Grb proteins

play in Tie2 signal transduction. What is clear, however, is that a Grb14 mediated

pathway downstream of Tie2 appears to involve a distinct set of conditions. We now also

know that Grb14 can be tyrosine phosphorylated as was previously predicted, and that

this phosphorylation appears to involve multiple tyrosine residues on Tie2. Furthermore,

this tyrosine phosphorylation implies there may be a link between Grb14 and Tie2 and

other downstream SH2 and/or PTB containing molecules.

113

4.3 Concluding Remarks

Proper development, maintenance and regression of blood vessels are essential

components of the mammalian vasculature system. As we learn more about this complex

process, we realize the impact that context and cellular environment has on what signals

are elicited during these events. The interactions between the various signaling partners

(ligands, receptors, adaptors etc.) are both intricate and numerous; furthermore, they do

not exist in isolation. These environmental factors are not to be ignored and highlight the

need for the use of biologically relevant systems. This, of course, requires more

sophisticated reagents and techniques which are often the rate limiting factor behind

scientific advances.

Early on, mouse molecular models identified Tie2 as an important regulator of

angiogenesis. For many years following the cloning of the Tie2 receptor, their cognitive

ligands remained elusive. This initially hampered endogenous studies of this signaling

system. Therefore, until the discovery and isolation of the angiopoietins, much of the

work pertaining to Tie2 signal transduction has been carried out using overexpression

studies.

We are now just beginning to understand the role of different Tie2 ligands which

appear to be context specific. It is reasonable to assume the different ligands will affect

signal transduction pathways downstream of Tie2. In fact, the use of the angiopoietins

has allowed for the discovery of novel proteins involved in Tie2 signal transduction (eg.

ShcA and ABIN-2) 116,198.

114

Although described as Tie2 binding partners many years ago, the role of the Grb

proteins in this system have thus far remained unknown. And while there is still much to

be learnt about how Grb7 and 14 contribute to Tie2 signaling, it appears the role of these

adaptor proteins in Tie2 signaling is dependent on a distinct subset of conditions.

Our study of the Tie2 tyrosine mutants has also raised some interesting questions

regarding the important role of Y1111 in receptor activation. These findings also

highlight a need for additional information surrounding Tie2 activation and regulation.

Understanding how Tie2 is regulated and the various molecules and pathways involved in

eliciting this receptors signals is crucial if we are going to be able to target specific

molecules for angiogenic and anti-angiogenic therapies.

1

115

REFERENCES 1 Robb, L. and Elefanty, A. G., The hemangioblast--an elusive cell captured in culture. Bioessays 20 (8), 611 (1998). 2 Pardanaud, L. et al., Two distinct endothelial lineages in ontogeny, one of them related to hemopoiesis. Development 122 (5), 1363 (1996). 3 Risau, W. and Flamme, I., Vasculogenesis. Annu Rev Cell Dev Biol 11, 73 (1995). 4 Djonov, V. and Makanya, A. N., New insights into intussusceptive angiogenesis. EXS (94), 17 (2005). 5 Augustin, H. G., Koh, G. Y., Thurston, G., and Alitalo, K., Control of vascular morphogenesis and homeostasis through the angiopoietin-Tie system. Nat Rev Mol Cell Biol 10 (3), 165 (2009). 6 Carmeliet, P. and Collen, D., Molecular basis of angiogenesis. Role of VEGF and VE-cadherin. Ann N Y Acad Sci 902, 249 (2000). 7 Hubbard, S. R. and Till, J. H., Protein tyrosine kinase structure and function. Annu Rev Biochem 69, 373 (2000). 8 Hubbard, S. R., Juxtamembrane autoinhibition in receptor tyrosine kinases. Nat Rev Mol Cell Biol 5 (6), 464 (2004). 9 Schlessinger, J., Cell signaling by receptor tyrosine kinases. Cell 103 (2), 211 (2000). 10 Ullrich, A. and Schlessinger, J., Signal transduction by receptors with tyrosine kinase activity. Cell 61 (2), 203 (1990); Heldin, C. H., Dimerization of cell surface receptors in signal transduction. Cell 80 (2), 213 (1995). 11 Sako, Y., Minoghchi, S., and Yanagida, T., Single-molecule imaging of EGFR signalling on the surface of living cells. Nat Cell Biol 2 (3), 168 (2000). 12 Moriki, T., Maruyama, H., and Maruyama, I. N., Activation of preformed EGF receptor dimers by ligand-induced rotation of the transmembrane domain. J Mol Biol 311 (5), 1011 (2001). 13 Burke, C. L. and Stern, D. F., Activation of Neu (ErbB-2) mediated by disulfide bond-induced dimerization reveals a receptor tyrosine kinase dimer interface. Mol Cell Biol 18 (9), 5371 (1998). 14 Johnson, L. N., Noble, M. E., and Owen, D. J., Active and inactive protein kinases: structural basis for regulation. Cell 85 (2), 149 (1996). 15 Wybenga-Groot, L. E. et al., Structural basis for autoinhibition of the Ephb2 receptor tyrosine kinase by the unphosphorylated juxtamembrane region. Cell 106 (6), 745 (2001). 16 Griffith, J. et al., The structural basis for autoinhibition of FLT3 by the juxtamembrane domain. Mol Cell 13 (2), 169 (2004). 17 Shewchuk, L. M. et al., Structure of the Tie2 RTK domain: self-inhibition by the nucleotide binding loop, activation loop, and C-terminal tail. Structure 8 (11), 1105 (2000). 18 Pawson, T. and Nash, P., Assembly of cell regulatory systems through protein interaction domains. Science 300 (5618), 445 (2003). 19 Pawson, T. and Gish, G. D., SH2 and SH3 domains: from structure to function. Cell 71 (3), 359 (1992).

116

20 Kuriyan, J. and Cowburn, D., Modular peptide recognition domains in eukaryotic signaling. Annu Rev Biophys Biomol Struct 26, 259 (1997). 21 Sayos, J. et al., The X-linked lymphoproliferative-disease gene product SAP regulates signals induced through the co-receptor SLAM. Nature 395 (6701), 462 (1998). 22 Morrione, A. et al., mGrb10 interacts with Nedd4. J Biol Chem 274 (34), 24094 (1999). 23 Blaikie, P. et al., A region in Shc distinct from the SH2 domain can bind tyrosine-phosphorylated growth factor receptors. J Biol Chem 269 (51), 32031 (1994). 24 van der Geer, P. and Pawson, T., The PTB domain: a new protein module implicated in signal transduction. Trends Biochem Sci 20 (7), 277 (1995). 25 Forman-Kay, J. D. and Pawson, T., Diversity in protein recognition by PTB domains. Curr Opin Struct Biol 9 (6), 690 (1999). 26 Fiore, F. et al., The regions of the Fe65 protein homologous to the phosphotyrosine interaction/phosphotyrosine binding domain of Shc bind the intracellular domain of the Alzheimer's amyloid precursor protein. J Biol Chem 270 (52), 30853 (1995); Borg, J. P., Ooi, J., Levy, E., and Margolis, B., The phosphotyrosine interaction domains of X11 and FE65 bind to distinct sites on the YENPTY motif of amyloid precursor protein. Mol Cell Biol 16 (11), 6229 (1996). 27 Ren, R., Mayer, B. J., Cicchetti, P., and Baltimore, D., Identification of a ten-amino acid proline-rich SH3 binding site. Science 259 (5098), 1157 (1993). 28 Harlan, J. E., Hajduk, P. J., Yoon, H. S., and Fesik, S. W., Pleckstrin homology domains bind to phosphatidylinositol-4,5-bisphosphate. Nature 371 (6493), 168 (1994). 29 Shoelson, S. E. et al., Specific phosphopeptide binding regulates a conformational change in the PI 3-kinase SH2 domain associated with enzyme activation. EMBO J 12 (2), 795 (1993). 30 Ostman, A., Hellberg, C., and Bohmer, F. D., Protein-tyrosine phosphatases and cancer. Nat Rev Cancer 6 (4), 307 (2006). 31 Pawson, T. and Scott, J. D., Signaling through scaffold, anchoring, and adaptor proteins. Science 278 (5346), 2075 (1997). 32 Conway, E. M., Collen, D., and Carmeliet, P., Molecular mechanisms of blood vessel growth. Cardiovasc Res 49 (3), 507 (2001). 33 Millauer, B. et al., High affinity VEGF binding and developmental expression suggest Flk-1 as a major regulator of vasculogenesis and angiogenesis. Cell 72 (6), 835 (1993). 34 Yamaguchi, T. P. et al., flk-1, an flt-related receptor tyrosine kinase is an early marker for endothelial cell precursors. Development 118 (2), 489 (1993). 35 Saharinen, P. et al., Multiple angiopoietin recombinant proteins activate the Tie1 receptor tyrosine kinase and promote its interaction with Tie2. J Cell Biol 169 (2), 239 (2005). 36 Yuan, H. T. et al., Activation of the orphan endothelial receptor Tie1 modifies Tie2-mediated intracellular signaling and cell survival. FASEB J 21 (12), 3171 (2007). 37 Milner, C. S., Hansen, T. M., Singh, H., and Brindle, N. P., Roles of the receptor tyrosine kinases Tie1 and Tie2 in mediating the effects of angiopoietin-1 on endothelial permeability and apoptosis. Microvasc Res 77 (2), 187 (2009).

117

38 Dumont, D. J. et al., tek, a novel tyrosine kinase gene located on mouse chromosome 4, is expressed in endothelial cells and their presumptive precursors. Oncogene 7 (8), 1471 (1992). 39 Partanen, J. et al., A novel endothelial cell surface receptor tyrosine kinase with extracellular epidermal growth factor homology domains. Mol Cell Biol 12 (4), 1698 (1992). 40 Barton, W. A. et al., Crystal structures of the Tie2 receptor ectodomain and the angiopoietin-2-Tie2 complex. Nat Struct Mol Biol 13 (6), 524 (2006). 41 Schnurch, H. and Risau, W., Expression of tie-2, a member of a novel family of receptor tyrosine kinases, in the endothelial cell lineage. Development 119 (3), 957 (1993). 42 Davis, S. et al., Isolation of angiopoietin-1, a ligand for the TIE2 receptor, by secretion-trap expression cloning. Cell 87 (7), 1161 (1996). 43 Procopio, W. N., Pelavin, P. I., Lee, W. M., and Yeilding, N. M., Angiopoietin-1 and -2 coiled coil domains mediate distinct homo-oligomerization patterns, but fibrinogen-like domains mediate ligand activity. J Biol Chem 274 (42), 30196 (1999). 44 Davis, S. et al., Angiopoietins have distinct modular domains essential for receptor binding, dimerization and superclustering. Nat Struct Biol 10 (1), 38 (2003). 45 Kim, K. T. et al., Oligomerization and multimerization are critical for angiopoietin-1 to bind and phosphorylate Tie2. J Biol Chem 280 (20), 20126 (2005). 46 Fiedler, U. et al., Angiopoietin-1 and angiopoietin-2 share the same binding domains in the Tie-2 receptor involving the first Ig-like loop and the epidermal growth factor-like repeats. J Biol Chem 278 (3), 1721 (2003). 47 Macdonald, P. R. et al., Structure of the extracellular domain of Tie receptor tyrosine kinases and localization of the angiopoietin-binding epitope. J Biol Chem 281 (38), 28408 (2006). 48 Korhonen, J. et al., Enhanced expression of the tie receptor tyrosine kinase in endothelial cells during neovascularization. Blood 80 (10), 2548 (1992). 49 Kaipainen, A. et al., Enhanced expression of the tie receptor tyrosine kinase mesenger RNA in the vascular endothelium of metastatic melanomas. Cancer Res 54 (24), 6571 (1994). 50 Porat, R. M. et al., Specific induction of tie1 promoter by disturbed flow in atherosclerosis-prone vascular niches and flow-obstructing pathologies. Circ Res 94 (3), 394 (2004). 51 Arai, F. et al., Tie2/angiopoietin-1 signaling regulates hematopoietic stem cell quiescence in the bone marrow niche. Cell 118 (2), 149 (2004); De Palma, M., Venneri, M. A., Roca, C., and Naldini, L., Targeting exogenous genes to tumor angiogenesis by transplantation of genetically modified hematopoietic stem cells. Nat Med 9 (6), 789 (2003); De Palma, M. et al., Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell 8 (3), 211 (2005); Sturn, D. H. et al., Angiopoietin affects neutrophil migration. Microcirculation 12 (5), 393 (2005). 52 Voskas, D. et al., A cyclosporine-sensitive psoriasis-like disease produced in Tie2 transgenic mice. Am J Pathol 166 (3), 843 (2005). 53 Iurlaro, M. et al., Rat aorta-derived mural precursor cells express the Tie2 receptor and respond directly to stimulation by angiopoietins. J Cell Sci 116 (Pt 17), 3635

118

(2003); Tian, S., Hayes, A. J., Metheny-Barlow, L. J., and Li, L. Y., Stabilization of breast cancer xenograft tumour neovasculature by angiopoietin-1. Br J Cancer 86 (4), 645 (2002); Shahrara, S. et al., Differential expression of the angiogenic Tie receptor family in arthritic and normal synovial tissue. Arthritis Res 4 (3), 201 (2002). 54 Calvert, J. T. et al., Allelic and locus heterogeneity in inherited venous malformations. Hum Mol Genet 8 (7), 1279 (1999). 55 Kosacka, J. et al., Angiopoietin-1 promotes neurite outgrowth from dorsal root ganglion cells positive for Tie-2 receptor. Cell Tissue Res 320 (1), 11 (2005); Valable, S. et al., Angiopoietin-1-induced PI3-kinase activation prevents neuronal apoptosis. FASEB J 17 (3), 443 (2003). 56 Otani, A. et al., Expressions of angiopoietins and Tie2 in human choroidal neovascular membranes. Invest Ophthalmol Vis Sci 40 (9), 1912 (1999). 57 Hirchenhain, J. et al., Differential expression of angiopoietins 1 and 2 and their receptor Tie-2 in human endometrium. Mol Hum Reprod 9 (11), 663 (2003). 58 Stratmann, A., Risau, W., and Plate, K. H., Cell type-specific expression of angiopoietin-1 and angiopoietin-2 suggests a role in glioblastoma angiogenesis. Am J Pathol 153 (5), 1459 (1998). 59 Sugimachi, K. et al., Angiopoietin switching regulates angiogenesis and progression of human hepatocellular carcinoma. J Clin Pathol 56 (11), 854 (2003). 60 Xu, Y. and Yu, Q., Angiopoietin-1, unlike angiopoietin-2, is incorporated into the extracellular matrix via its linker peptide region. J Biol Chem 276 (37), 34990 (2001). 61 Gale, N. W. et al., Angiopoietin-2 is required for postnatal angiogenesis and lymphatic patterning, and only the latter role is rescued by Angiopoietin-1. Dev Cell 3 (3), 411 (2002). 62 Holash, J., Wiegand, S. J., and Yancopoulos, G. D., New model of tumor angiogenesis: dynamic balance between vessel regression and growth mediated by angiopoietins and VEGF. Oncogene 18 (38), 5356 (1999); Huang, Y. Q. et al., Thrombin induces increased expression and secretion of angiopoietin-2 from human umbilical vein endothelial cells. Blood 99 (5), 1646 (2002); Mandriota, S. J. and Pepper, M. S., Regulation of angiopoietin-2 mRNA levels in bovine microvascular endothelial cells by cytokines and hypoxia. Circ Res 83 (8), 852 (1998). 63 Zhang, L. et al., Tumor-derived vascular endothelial growth factor up-regulates angiopoietin-2 in host endothelium and destabilizes host vasculature, supporting angiogenesis in ovarian cancer. Cancer Res 63 (12), 3403 (2003). 64 Hackett, S. F. et al., Angiopoietin 2 expression in the retina: upregulation during physiologic and pathologic neovascularization. J Cell Physiol 184 (3), 275 (2000). 65 Fiedler, U. et al., Angiopoietin-2 sensitizes endothelial cells to TNF-alpha and has a crucial role in the induction of inflammation. Nat Med 12 (2), 235 (2006). 66 Oh, H. et al., Hypoxia and vascular endothelial growth factor selectively up-regulate angiopoietin-2 in bovine microvascular endothelial cells. J Biol Chem 274 (22), 15732 (1999). 67 Sato, T. N. et al., Distinct roles of the receptor tyrosine kinases Tie-1 and Tie-2 in blood vessel formation. Nature 376 (6535), 70 (1995). 68 Puri, M. C. et al., The receptor tyrosine kinase TIE is required for integrity and survival of vascular endothelial cells. EMBO J 14 (23), 5884 (1995).

119

69 Partanen, J. et al., Cell autonomous functions of the receptor tyrosine kinase TIE in a late phase of angiogenic capillary growth and endothelial cell survival during murine development. Development 122 (10), 3013 (1996). 70 Dumont, D. J. et al., Dominant-negative and targeted null mutations in the endothelial receptor tyrosine kinase, tek, reveal a critical role in vasculogenesis of the embryo. Genes Dev 8 (16), 1897 (1994). 71 Jones, N. et al., Identification of Tek/Tie2 binding partners. Binding to a multifunctional docking site mediates cell survival and migration. J Biol Chem 274 (43), 30896 (1999). 72 Papapetropoulos, A. et al., Direct actions of angiopoietin-1 on human endothelium: evidence for network stabilization, cell survival, and interaction with other angiogenic growth factors. Lab Invest 79 (2), 213 (1999). 73 Fujikawa, K. et al., Role of PI 3-kinase in angiopoietin-1-mediated migration and attachment-dependent survival of endothelial cells. Exp Cell Res 253 (2), 663 (1999). 74 Kim, I. et al., Angiopoietin-1 regulates endothelial cell survival through the phosphatidylinositol 3'-Kinase/Akt signal transduction pathway. Circ Res 86 (1), 24 (2000). 75 Suri, C. et al., Requisite role of angiopoietin-1, a ligand for the TIE2 receptor, during embryonic angiogenesis. Cell 87 (7), 1171 (1996). 76 Maisonpierre, P. C. et al., Angiopoietin-2, a natural antagonist for Tie2 that disrupts in vivo angiogenesis. Science 277 (5322), 55 (1997). 77 Goede, V. et al., Analysis of blood vessel maturation processes during cyclic ovarian angiogenesis. Lab Invest 78 (11), 1385 (1998). 78 Wong, A. L. et al., Tie2 expression and phosphorylation in angiogenic and quiescent adult tissues. Circ Res 81 (4), 567 (1997). 79 Jones, N. et al., Rescue of the early vascular defects in Tek/Tie2 null mice reveals an essential survival function. EMBO Rep 2 (5), 438 (2001). 80 Puri, M. C. and Bernstein, A., Requirement for the TIE family of receptor tyrosine kinases in adult but not fetal hematopoiesis. Proc Natl Acad Sci U S A 100 (22), 12753 (2003). 81 Puri, M. C., Partanen, J., Rossant, J., and Bernstein, A., Interaction of the TEK and TIE receptor tyrosine kinases during cardiovascular development. Development 126 (20), 4569 (1999). 82 Thurston, G. et al., Leakage-resistant blood vessels in mice transgenically overexpressing angiopoietin-1. Science 286 (5449), 2511 (1999). 83 Thurston, G. et al., Angiopoietin-1 protects the adult vasculature against plasma leakage. Nat Med 6 (4), 460 (2000). 84 Hackett, S. F., Wiegand, S., Yancopoulos, G., and Campochiaro, P. A., Angiopoietin-2 plays an important role in retinal angiogenesis. J Cell Physiol 192 (2), 182 (2002). 85 Pitera, J. E. et al., Dysmorphogenesis of kidney cortical peritubular capillaries in angiopoietin-2-deficient mice. Am J Pathol 165 (6), 1895 (2004). 86 Carmeliet, P., Angiogenesis in health and disease. Nat Med 9 (6), 653 (2003). 87 Peters, K. G. et al., Expression of Tie2/Tek in breast tumour vasculature provides a new marker for evaluation of tumour angiogenesis. Br J Cancer 77 (1), 51 (1998).

120

88 Hawighorst, T. et al., Activation of the tie2 receptor by angiopoietin-1 enhances tumor vessel maturation and impairs squamous cell carcinoma growth. Am J Pathol 160 (4), 1381 (2002); Hayes, A. J. et al., Expression and function of angiopoietin-1 in breast cancer. Br J Cancer 83 (9), 1154 (2000); Stoeltzing, O. et al., Angiogenesis and antiangiogenic therapy of colon cancer liver metastasis. Ann Surg Oncol 10 (7), 722 (2003). 89 Machein, M. R. et al., Angiopoietin-1 promotes tumor angiogenesis in a rat glioma model. Am J Pathol 165 (5), 1557 (2004). 90 Ahmad, S. A. et al., Differential expression of angiopoietin-1 and angiopoietin-2 in colon carcinoma. A possible mechanism for the initiation of angiogenesis. Cancer 92 (5), 1138 (2001). 91 Etoh, T. et al., Angiopoietin-2 is related to tumor angiogenesis in gastric carcinoma: possible in vivo regulation via induction of proteases. Cancer Res 61 (5), 2145 (2001). 92 Yoshiji, H. et al., Angiopoietin 2 displays a vascular endothelial growth factor dependent synergistic effect in hepatocellular carcinoma development in mice. Gut 54 (12), 1768 (2005). 93 Hu, B. et al., Angiopoietin-2 induces human glioma invasion through the activation of matrix metalloprotease-2. Proc Natl Acad Sci U S A 100 (15), 8904 (2003). 94 Yu, Q. and Stamenkovic, I., Angiopoietin-2 is implicated in the regulation of tumor angiogenesis. Am J Pathol 158 (2), 563 (2001). 95 Lin, P. et al., Antiangiogenic gene therapy targeting the endothelium-specific receptor tyrosine kinase Tie2. Proc Natl Acad Sci U S A 95 (15), 8829 (1998); Siemeister, G. et al., Two independent mechanisms essential for tumor angiogenesis: inhibition of human melanoma xenograft growth by interfering with either the vascular endothelial growth factor receptor pathway or the Tie-2 pathway. Cancer Res 59 (13), 3185 (1999). 96 Tournaire, R. et al., A short synthetic peptide inhibits signal transduction, migration and angiogenesis mediated by Tie2 receptor. EMBO Rep 5 (3), 262 (2004). 97 Jendreyko, N., Popkov, M., Rader, C., and Barbas, C. F., 3rd, Phenotypic knockout of VEGF-R2 and Tie-2 with an intradiabody reduces tumor growth and angiogenesis in vivo. Proc Natl Acad Sci U S A 102 (23), 8293 (2005); Popkov, M. et al., Targeting tumor angiogenesis with adenovirus-delivered anti-Tie-2 intrabody. Cancer Res 65 (3), 972 (2005). 98 Scharpfenecker, M., Fiedler, U., Reiss, Y., and Augustin, H. G., The Tie-2 ligand angiopoietin-2 destabilizes quiescent endothelium through an internal autocrine loop mechanism. J Cell Sci 118 (Pt 4), 771 (2005). 99 Kim, I. et al., Angiopoietin-1 reduces VEGF-stimulated leukocyte adhesion to endothelial cells by reducing ICAM-1, VCAM-1, and E-selectin expression. Circ Res 89 (6), 477 (2001); Gamble, J. R. et al., Angiopoietin-1 is an antipermeability and anti-inflammatory agent in vitro and targets cell junctions. Circ Res 87 (7), 603 (2000). 100 Vikkula, M. et al., Vascular dysmorphogenesis caused by an activating mutation in the receptor tyrosine kinase TIE2. Cell 87 (7), 1181 (1996). 101 Morris, P. N. et al., Functional analysis of a mutant form of the receptor tyrosine kinase Tie2 causing venous malformations. J Mol Med 83 (1), 58 (2005).

121

102 Murray, B. W., Padrique, E. S., Pinko, C., and McTigue, M. A., Mechanistic effects of autophosphorylation on receptor tyrosine kinase catalysis: enzymatic characterization of Tie2 and phospho-Tie2. Biochemistry 40 (34), 10243 (2001). 103 Kontos, C. D., Cha, E. H., York, J. D., and Peters, K. G., The endothelial receptor tyrosine kinase Tie1 activates phosphatidylinositol 3-kinase and Akt to inhibit apoptosis. Mol Cell Biol 22 (6), 1704 (2002). 104 Marron, M. B. et al., Regulated proteolytic processing of Tie1 modulates ligand responsiveness of the receptor-tyrosine kinase Tie2. J Biol Chem 282 (42), 30509 (2007). 105 Chen-Konak, L. et al., Transcriptional and post-translation regulation of the Tie1 receptor by fluid shear stress changes in vascular endothelial cells. FASEB J 17 (14), 2121 (2003); Yabkowitz, R. et al., Inflammatory cytokines and vascular endothelial growth factor stimulate the release of soluble tie receptor from human endothelial cells via metalloprotease activation. Blood 93 (6), 1969 (1999); Yabkowitz, R. et al., Regulation of tie receptor expression on human endothelial cells by protein kinase C-mediated release of soluble tie. Blood 90 (2), 706 (1997). 106 Singh, H. et al., Vascular endothelial growth factor activates the tie family of receptor tyrosine kinases. Cell Signal (2009). 107 Kontos, C. D. et al., Tyrosine 1101 of Tie2 is the major site of association of p85 and is required for activation of phosphatidylinositol 3-kinase and Akt. Mol Cell Biol 18 (7), 4131 (1998). 108 Papapetropoulos, A. et al., Angiopoietin-1 inhibits endothelial cell apoptosis via the Akt/survivin pathway. J Biol Chem 275 (13), 9102 (2000). 109 Kwak, H. J. et al., Angiopoietin-1 inhibits irradiation- and mannitol-induced apoptosis in endothelial cells. Circulation 101 (19), 2317 (2000). 110 Harfouche, R. et al., Mechanisms which mediate the antiapoptotic effects of angiopoietin-1 on endothelial cells. Microvasc Res 64 (1), 135 (2002). 111 Tadros, M., Rizk, S. S., Opher, E., and Thompson, L. D., Polymorphous hemangioendothelioma of the neck. Ann Diagn Pathol 7 (3), 165 (2003). 112 Mochizuki, Y., Nakamura, T., Kanetake, H., and Kanda, S., Angiopoietin 2 stimulates migration and tube-like structure formation of murine brain capillary endothelial cells through c-Fes and c-Fyn. J Cell Sci 115 (Pt 1), 175 (2002). 113 Kwak, H. J. et al., Angiopoietin-1 is an apoptosis survival factor for endothelial cells. FEBS Lett 448 (2-3), 249 (1999); Witzenbichler, B. et al., Chemotactic properties of angiopoietin-1 and -2, ligands for the endothelial-specific receptor tyrosine kinase Tie2. J Biol Chem 273 (29), 18514 (1998). 114 Hayes, A. J. et al., Angiopoietin-1 and its receptor Tie-2 participate in the regulation of capillary-like tubule formation and survival of endothelial cells. Microvasc Res 58 (3), 224 (1999); Koblizek, T. I. et al., Angiopoietin-1 induces sprouting angiogenesis in vitro. Curr Biol 8 (9), 529 (1998). 115 Audero, E. et al., Adaptor ShcA protein binds tyrosine kinase Tie2 receptor and regulates migration and sprouting but not survival of endothelial cells. J Biol Chem 279 (13), 13224 (2004). 116 Cascone, I. et al., Tie-2-dependent activation of RhoA and Rac1 participates in endothelial cell motility triggered by angiopoietin-1. Blood 102 (7), 2482 (2003).

122

117 Schaller, M. D. et al., Autophosphorylation of the focal adhesion kinase, pp125FAK, directs SH2-dependent binding of pp60src. Mol Cell Biol 14 (3), 1680 (1994). 118 Han, D. C. and Guan, J. L., Association of focal adhesion kinase with Grb7 and its role in cell migration. J Biol Chem 274 (34), 24425 (1999). 119 Yu, D. H. et al., Protein-tyrosine phosphatase Shp-2 regulates cell spreading, migration, and focal adhesion. J Biol Chem 273 (33), 21125 (1998). 120 Jones, N. et al., A unique autophosphorylation site on Tie2/Tek mediates Dok-R phosphotyrosine binding domain binding and function. Mol Cell Biol 23 (8), 2658 (2003). 121 Garrity, P. A. et al., Drosophila photoreceptor axon guidance and targeting requires the dreadlocks SH2/SH3 adapter protein. Cell 85 (5), 639 (1996). 122 Master, Z. et al., Dok-R plays a pivotal role in angiopoietin-1-dependent cell migration through recruitment and activation of Pak. EMBO J 20 (21), 5919 (2001). 123 Nelms, K., Snow, A. J., and Noben-Trauth, K., Dok1 encoding p62(dok) maps to mouse chromosome 6 and human chromosome 2 in a region of translocation in chronic lymphocytic leukemia. Genomics 53 (2), 243 (1998). 124 Kasus-Jacobi, A. et al., Identification of the rat adapter Grb14 as an inhibitor of insulin actions. J Biol Chem 273 (40), 26026 (1998). 125 Reilly, J. F., Mickey, G., and Maher, P. A., Association of fibroblast growth factor receptor 1 with the adaptor protein Grb14. Characterization of a new receptor binding partner. J Biol Chem 275 (11), 7771 (2000). 126 Kanda, S. et al., Angiopoietin 1 is mitogenic for cultured endothelial cells. Cancer Res 65 (15), 6820 (2005). 127 Nguyen, V. P. et al., Differential response of lymphatic, venous and arterial endothelial cells to angiopoietin-1 and angiopoietin-2. BMC Cell Biol 8, 10 (2007). 128 Fukuhara, S. et al., Differential function of Tie2 at cell-cell contacts and cell-substratum contacts regulated by angiopoietin-1. Nat Cell Biol 10 (5), 513 (2008). 129 Bogdanovic, E., Nguyen, V. P., and Dumont, D. J., Activation of Tie2 by angiopoietin-1 and angiopoietin-2 results in their release and receptor internalization. J Cell Sci 119 (Pt 17), 3551 (2006). 130 Reiss, Y. et al., Angiopoietin-2 impairs revascularization after limb ischemia. Circ Res 101 (1), 88 (2007). 131 Daly, C. et al., Angiopoietin-2 functions as an autocrine protective factor in stressed endothelial cells. Proc Natl Acad Sci U S A 103 (42), 15491 (2006). 132 Valenzuela, D. M. et al., Angiopoietins 3 and 4: diverging gene counterparts in mice and humans. Proc Natl Acad Sci U S A 96 (5), 1904 (1999). 133 Lee, H. J. et al., Biological characterization of angiopoietin-3 and angiopoietin-4. FASEB J 18 (11), 1200 (2004). 134 Sorkin, A. and Waters, C. M., Endocytosis of growth factor receptors. Bioessays 15 (6), 375 (1993). 135 Kasuga, M. et al., Insulin-induced receptor loss in cultured human lymphocytes is due to accelerated receptor degradation. Proc Natl Acad Sci U S A 78 (11), 6917 (1981); Keating, M. T. and Williams, L. T., Processing of the platelet-derived growth factor receptor. Biosynthetic and degradation studies using anti-receptor antibodies. J Biol Chem 262 (16), 7932 (1987).

123

136 Fachinger, G., Deutsch, U., and Risau, W., Functional interaction of vascular endothelial-protein-tyrosine phosphatase with the angiopoietin receptor Tie-2. Oncogene 18 (43), 5948 (1999). 137 Huang, L., Turck, C. W., Rao, P., and Peters, K. G., GRB2 and SH-PTP2: potentially important endothelial signaling molecules downstream of the TEK/TIE2 receptor tyrosine kinase. Oncogene 11 (10), 2097 (1995). 138 Ooi, J. et al., The cloning of Grb10 reveals a new family of SH2 domain proteins. Oncogene 10 (8), 1621 (1995). 139 Stein, D. et al., The SH2 domain protein GRB-7 is co-amplified, overexpressed and in a tight complex with HER2 in breast cancer. EMBO J 13 (6), 1331 (1994). 140 Daly, R. J., Sanderson, G. M., Janes, P. W., and Sutherland, R. L., Cloning and characterization of GRB14, a novel member of the GRB7 gene family. J Biol Chem 271 (21), 12502 (1996). 141 Frantz, J. D., Giorgetti-Peraldi, S., Ottinger, E. A., and Shoelson, S. E., Human GRB-IRbeta/GRB10. Splice variants of an insulin and growth factor receptor-binding protein with PH and SH2 domains. J Biol Chem 272 (5), 2659 (1997). 142 Giovannone, B. et al., Two novel proteins that are linked to insulin-like growth factor (IGF-I) receptors by the Grb10 adapter and modulate IGF-I signaling. J Biol Chem 278 (34), 31564 (2003). 143 Lyons, R. J. et al., Identification of a novel human tankyrase through its interaction with the adaptor protein Grb14. J Biol Chem 276 (20), 17172 (2001). 144 Manser, J., Roonprapunt, C., and Margolis, B., C. elegans cell migration gene mig-10 shares similarities with a family of SH2 domain proteins and acts cell nonautonomously in excretory canal development. Dev Biol 184 (1), 150 (1997). 145 Wojcik, J. et al., Sequence analysis identifies a ras-associating (RA)-like domain in the N-termini of band 4.1/JEF domains and in the Grb7/10/14 adapter family. Biochem Biophys Res Commun 259 (1), 113 (1999). 146 Rodriguez-Viciana, P., Sabatier, C., and McCormick, F., Signaling specificity by Ras family GTPases is determined by the full spectrum of effectors they regulate. Mol Cell Biol 24 (11), 4943 (2004). 147 Shen, T. L., Han, D. C., and Guan, J. L., Association of Grb7 with phosphoinositides and its role in the regulation of cell migration. J Biol Chem 277 (32), 29069 (2002). 148 Rajala, R. V. and Chan, M. D., Identification of a NPXY motif in growth factor receptor-bound protein 14 (Grb14) and its interaction with the phosphotyrosine-binding (PTB) domain of IRS-1. Biochemistry 44 (22), 7929 (2005). 149 Malinin, N. L., Boldin, M. P., Kovalenko, A. V., and Wallach, D., MAP3K-related kinase involved in NF-kappaB induction by TNF, CD95 and IL-1. Nature 385 (6616), 540 (1997). 150 Chen, D. et al., NIK is a component of the EGF/heregulin receptor signaling complexes. Oncogene 22 (28), 4348 (2003). 151 Moncoq, K. et al., The PIR domain of Grb14 is an intrinsically unstructured protein: implication in insulin signaling. FEBS Lett 554 (3), 240 (2003). 152 Tompa, P., Intrinsically unstructured proteins. Trends Biochem Sci 27 (10), 527 (2002).

124

153 He, W., Rose, D. W., Olefsky, J. M., and Gustafson, T. A., Grb10 interacts differentially with the insulin receptor, insulin-like growth factor I receptor, and epidermal growth factor receptor via the Grb10 Src homology 2 (SH2) domain and a second novel domain located between the pleckstrin homology and SH2 domains. J Biol Chem 273 (12), 6860 (1998). 154 Stein, E. G., Gustafson, T. A., and Hubbard, S. R., The BPS domain of Grb10 inhibits the catalytic activity of the insulin and IGF1 receptors. FEBS Lett 493 (2-3), 106 (2001). 155 Bereziat, V. et al., Inhibition of insulin receptor catalytic activity by the molecular adapter Grb14. J Biol Chem 277 (7), 4845 (2002). 156 Cariou, B. et al., The adapter protein ZIP binds Grb14 and regulates its inhibitory action on insulin signaling by recruiting protein kinase Czeta. Mol Cell Biol 22 (20), 6959 (2002). 157 Depetris, R. S. et al., Structural basis for inhibition of the insulin receptor by the adaptor protein Grb14. Mol Cell 20 (2), 325 (2005). 158 Holt, L. J. and Siddle, K., Grb10 and Grb14: enigmatic regulators of insulin action--and more? Biochem J 388 (Pt 2), 393 (2005). 159 Hemming, R. et al., Human growth factor receptor bound 14 binds the activated insulin receptor and alters the insulin-stimulated tyrosine phosphorylation levels of multiple proteins. Biochem Cell Biol 79 (1), 21 (2001). 160 Smith, F. M. et al., Mice with a disruption of the imprinted Grb10 gene exhibit altered body composition, glucose homeostasis, and insulin signaling during postnatal life. Mol Cell Biol 27 (16), 5871 (2007); Liu, F. and Roth, R. A., Grb-IR: a SH2-domain-containing protein that binds to the insulin receptor and inhibits its function. Proc Natl Acad Sci U S A 92 (22), 10287 (1995). 161 Cooney, G. J. et al., Improved glucose homeostasis and enhanced insulin signalling in Grb14-deficient mice. EMBO J 23 (3), 582 (2004). 162 Ramos, F. J. et al., Grb10 mediates insulin-stimulated degradation of the insulin receptor: a mechanism of negative regulation. Am J Physiol Endocrinol Metab 290 (6), E1262 (2006). 163 Kasus-Jacobi, A. et al., Evidence for an interaction between the insulin receptor and Grb7. A role for two of its binding domains, PIR and SH2. Oncogene 19 (16), 2052 (2000). 164 Han, D. C., Shen, T. L., and Guan, J. L., The Grb7 family proteins: structure, interactions with other signaling molecules and potential cellular functions. Oncogene 20 (44), 6315 (2001). 165 Shen, T. L. and Guan, J. L., Grb7 in intracellular signaling and its role in cell regulation. Front Biosci 9, 192 (2004). 166 Tanaka, S. et al., Coexpression of Grb7 with epidermal growth factor receptor or Her2/erbB2 in human advanced esophageal carcinoma. Cancer Res 57 (1), 28 (1997); Kishi, T. et al., Molecular cloning of human GRB-7 co-amplified with CAB1 and c-ERBB-2 in primary gastric cancer. Biochem Biophys Res Commun 232 (1), 5 (1997). 167 Dong, L. Q., Farris, S., Christal, J., and Liu, F., Site-directed mutagenesis and yeast two-hybrid studies of the insulin and insulin-like growth factor-1 receptors: the Src homology-2 domain-containing protein hGrb10 binds to the autophosphorylated tyrosine

125

residues in the kinase domain of the insulin receptor. Mol Endocrinol 11 (12), 1757 (1997). 168 Langlais, P. et al., Phosphorylation of Grb10 by mitogen-activated protein kinase: identification of Ser150 and Ser476 of human Grb10zeta as major phosphorylation sites. Biochemistry 44 (24), 8890 (2005). 169 Langlais, P., Dong, L. Q., Hu, D., and Liu, F., Identification of Grb10 as a direct substrate for members of the Src tyrosine kinase family. Oncogene 19 (25), 2895 (2000). 170 Mano, H. et al., Grb10/GrbIR as an in vivo substrate of Tec tyrosine kinase. Genes Cells 3 (7), 431 (1998). 171 Giorgetti-Peraldi, S., Murdaca, J., Mas, J. C., and Van Obberghen, E., The adapter protein, Grb10, is a positive regulator of vascular endothelial growth factor signaling. Oncogene 20 (30), 3959 (2001). 172 Han, D. C., Shen, T. L., and Guan, J. L., Role of Grb7 targeting to focal contacts and its phosphorylation by focal adhesion kinase in regulation of cell migration. J Biol Chem 275 (37), 28911 (2000). 173 Tanaka, S. et al., Grb7 signal transduction protein mediates metastatic progression of esophageal carcinoma. J Cell Physiol 183 (3), 411 (2000). 174 Han, D. C. et al., EphB1 associates with Grb7 and regulates cell migration. J Biol Chem 277 (47), 45655 (2002). 175 Pandey, A. et al., Direct association between the Ret receptor tyrosine kinase and the Src homology 2-containing adapter protein Grb7. J Biol Chem 271 (18), 10607 (1996). 176 Niu, X. L., Peters, K. G., and Kontos, C. D., Deletion of the carboxyl terminus of Tie2 enhances kinase activity, signaling, and function. Evidence for an autoinhibitory mechanism. J Biol Chem 277 (35), 31768 (2002). 177 Jones, N. and Dumont, D. J., Assignment of the Dokr gene to mouse chromosome 14D2-D3 by fluorescence in situ hybridization. Genomics 53 (3), 413 (1998). 178 Peters, K. G. et al., Functional significance of Tie2 signaling in the adult vasculature. Recent Prog Horm Res 59, 51 (2004). 179 Jones, N. and Dumont, D. J., The Tek/Tie2 receptor signals through a novel Dok-related docking protein, Dok-R. Oncogene 17 (9), 1097 (1998). 180 Neel, B. G., Gu, H., and Pao, L., The 'Shp'ing news: SH2 domain-containing tyrosine phosphatases in cell signaling. Trends Biochem Sci 28 (6), 284 (2003). 181 Till, J. H. et al., Crystal structure of the MuSK tyrosine kinase: insights into receptor autoregulation. Structure 10 (9), 1187 (2002); Chan, P. M. et al., Autoinhibition of the kit receptor tyrosine kinase by the cytosolic juxtamembrane region. Mol Cell Biol 23 (9), 3067 (2003). 182 Akiyama, T. et al., The transforming potential of the c-erbB-2 protein is regulated by its autophosphorylation at the carboxyl-terminal domain. Mol Cell Biol 11 (2), 833 (1991); Yokoyama, N., Ischenko, I., Hayman, M. J., and Miller, W. T., The C terminus of RON tyrosine kinase plays an autoinhibitory role. J Biol Chem 280 (10), 8893 (2005). 183 Chiara, F., Bishayee, S., Heldin, C. H., and Demoulin, J. B., Autoinhibition of the platelet-derived growth factor beta-receptor tyrosine kinase by its C-terminal tail. J Biol Chem 279 (19), 19732 (2004). 184 Tai, L. K. et al., Flow activates ERK1/2 and endothelial nitric oxide synthase via a pathway involving PECAM1, SHP2, and Tie2. J Biol Chem 280 (33), 29620 (2005).

126

185 Kassenbrock, C. K. and Anderson, S. M., Regulation of ubiquitin protein ligase activity in c-Cbl by phosphorylation-induced conformational change and constitutive activation by tyrosine to glutamate point mutations. J Biol Chem 279 (27), 28017 (2004). 186 Fiddes, R. J. et al., Inhibition of the MAP kinase cascade blocks heregulin-induced cell cycle progression in T-47D human breast cancer cells. Oncogene 16 (21), 2803 (1998). 187 Ubersax, J. A. and Ferrell, J. E., Jr., Mechanisms of specificity in protein phosphorylation. Nat Rev Mol Cell Biol 8 (7), 530 (2007). 188 Cohen, P., The regulation of protein function by multisite phosphorylation--a 25 year update. Trends Biochem Sci 25 (12), 596 (2000). 189 Vecchione, A. et al., The Grb10/Nedd4 complex regulates ligand-induced ubiquitination and stability of the insulin-like growth factor I receptor. Mol Cell Biol 23 (9), 3363 (2003). 190 Baumer, S. et al., Vascular endothelial cell-specific phosphotyrosine phosphatase (VE-PTP) activity is required for blood vessel development. Blood 107 (12), 4754 (2006). 191 Dominguez, M. G. et al., Vascular endothelial tyrosine phosphatase (VE-PTP)-null mice undergo vasculogenesis but die embryonically because of defects in angiogenesis. Proc Natl Acad Sci U S A 104 (9), 3243 (2007). 192 Hattori, K. et al., Establishment of bone marrow-derived endothelial cell lines from ts-SV40 T-antigen gene transgenic rats. Pharm Res 18 (1), 9 (2001); Takakura, N. et al., Critical role of the TIE2 endothelial cell receptor in the development of definitive hematopoiesis. Immunity 9 (5), 677 (1998). 193 Yokote, K., Margolis, B., Heldin, C. H., and Claesson-Welsh, L., Grb7 is a downstream signaling component of platelet-derived growth factor alpha- and beta-receptors. J Biol Chem 271 (48), 30942 (1996). 194 Keegan, K. and Cooper, J. A., Use of the two hybrid system to detect the association of the protein-tyrosine-phosphatase, SHPTP2, with another SH2-containing protein, Grb7. Oncogene 12 (7), 1537 (1996). 195 Cailliau, K. et al., Inhibition of FGF receptor signalling in Xenopus oocytes: differential effect of Grb7, Grb10 and Grb14. FEBS Lett 548 (1-3), 43 (2003). 196 Stein, E. G., Ghirlando, R., and Hubbard, S. R., Structural basis for dimerization of the Grb10 Src homology 2 domain. Implications for ligand specificity. J Biol Chem 278 (15), 13257 (2003). 197 Nouaille, S. et al., Interaction between the insulin receptor and Grb14: a dynamic study in living cells using BRET. Biochem Pharmacol 72 (11), 1355 (2006). 198 Hughes, D. P., Marron, M. B., and Brindle, N. P., The antiinflammatory endothelial tyrosine kinase Tie2 interacts with a novel nuclear factor-kappaB inhibitor ABIN-2. Circ Res 92 (6), 630 (2003). 199 Sadowski, I., Stone, J.C., Pawson, T., A noncatalytic domain conserved amoung cytoplasmic protein-tyrosine kinases modifies the kinase function and transformting activity of Fujinami sarcoma virus p130gag-fps. Mol Cel Biol 6 (12): 4396 (1992).

127

APPENDICES

128

Appendix 1: Effect of pervanadate on Tie2 phosphorylation. HEK293T cells were either left untransfected (control) or were transfected with WT Tie2 , the kinase deficient Tie2 (K853A) or the Tie2 tyrosine mutatn Y1111F. Immediately preceding harvesting, transfected cells were treated with either DMEM alone (top two panels) or with 10mM pervanadate (bottom two panels). Lysates from these cells were then immunoprecipitated for Tie2, run on an SDS PAGE gel and subjected to Western blot using anti-phosphotyrosine or anti-Tie2 (as indicated).

129

Appendix 2: Shp2 phosphorylation in Tie2 signal transduction. HEK293T cells were either left untransfected (control) or were transfected with WT Tie2 , the kinase deficient Tie2 (K853A) or one of the Tie2 tyrosine single mutants (Y1100F, Y11006F, Y1111F), double mutants (Y00/06F, Y00/11F, Y06/11F) or the triple mutant (Y3F). Lysates from these cells were then immunoprecipitated for Tie2 or FLAG (as indicated), run on an SDS PAGE gel and subjected to Western blot using anti-phosphotyrosine, anti-Tie2 , or FLAG (as indicated).

130

Grb14

1111

+ v

ecto

r

1111

+ G

rb14

1111

+ G

rb14

Phos

phat

ase

Appendix 3: Grb14 identified by silver staining. HEK293 cells were transfected with the Y1111F mutant in conjunction with either empty vector (lane 1) or Grb14 (lanes 2 and 3). In lane 3, cells were treated with pervanadate just before harvesting. Cells were lysed and subsequently subjected to SDS-PAGE. The gel was then subjected to silver staining.

131

IP: Grb7Blot: anti-pTyr

Appendix 4: Grb7 SH2 domain plays a role in binding of pp85 and pp70. HEK293T cells were transfected with Grb14 or Grb14FLVRS in combination with empty vector (vector), wild type Tie2 (WT), or the kinase deficient Tie2 (K853A). Lysates from these cells were immunoprecipitated using an antibody specific for Grb7 and blotted for phosphotyrosine (top panel) or Grb7 (bottom panel).

Vec

tor

WT

K85

3A

Vec

tor

WT

K85

3A

Grb7 Grb7FLVRS

pp85

Grb7pp70

IP: Grb7Blot: anti-Grb7Grb7

132

K85

3S

Y111

1F

WT

K85

3S

Y111

1F

WT

Grb14 Grb14FLVRS

Appendix 5: Grb14 SH2 domain required for Tie2 mediated phosphorylation. HEK293T cells were transfected with Grb14 or Grb14FLVRS in combination with empty vector, wild type Tie2 (WT), or the kinase deficient Tie2 (K853A). Lysates from these cells were immunoprecipitatedusing an antibody specific for FLAG (top panel) and blotted for phosphotyrosine (top panel). As a control for transfection of Grb14, cell lysate from transfected cells (20ug) was run on an SDS PAGE gel and blotted for FLAG.

Grb14

pGrb14IgG

Cell LysateBlot: FLAG

IP: FLAG Blot: anti-pTyr

133

Appendix 6: Mutation of Y1111 and receptor ubiquitination. HEK293T cells were transfectedwith an HA-tagged ubiquitin construct either alone (control) or in combination with Tie2 (Tek), the kinase deficient Tie2 (K853A) or the Y1111F mutant. Lysates from these cells were immunoprecipitated using an antibody specific for Tie2 and blotted either for the HA-tagged ubiquitin (top panel) or for Tie2 (bottom panel).

134

Appendix 7: Mutation of Y1111 to Phenylalanine does not affect membrane receptor levels. HEK293T cells were transfected with empty vector, wild type Tie2 (Tek), the kinase deficient Tie2 (K853A) or the Y1111F mutant receptor

135