The role of TAM receptors in brain tumour cell signalling ... · in brain tumour cell signalling...

The role of TAM receptors

in brain tumour

cell signalling and behaviour

By

Mikaella Vouri

The thesis is submitted in partial fulfilment of the requirements of

the award of the degree of Doctor of Philosophy of the University

of Portsmouth

May 2016

Word count: 38 170

ii

Abstract

Tumour invasion is the key element in the high rate of mortality and

morbidity in glioma patients. Increased expression, in neoplastic cells, of

receptor tyrosine kinases (RTKs) of the TAM family (comprising Tyro3, Axl and

MerTK), has been reported in several cancers including gliomas, in which they

play a key role in tumour invasion. Axl RTK, with molecular weight of 120-140

kDa, mediates glioma cell adhesion and invasion through a variety of ways.

Within the scope of this thesis, Western blotting, quantitative polymerase chain

reaction (qRT-PCR) and glioblastoma multiforme (GBM) tissue microarray

revealed an upregulation of Axl and Tyro3 in brain tumours and two adult GBM

cell lines, SNB-19 and UP007. Both cell lines were treated with the TAM ligand

Gas6 and/or the specific Axl small molecule inhibitor BGB324, and analysed in

assays for survival, 3D colony growth, motility, migration and invasion. Western

blotting was used to detect protein expression and signal protein

phosphorylation. In both cell lines, BGB324 inhibited specifically

phosphorylation of Axl as well as Akt kinase further downstream. BGB324 also

inhibited survival and proliferation of both cell lines in a concentration-

dependent manner, as well as completely suppressing migration and invasion.

Axl inhibition by BGB324 also sensitised GBM cells to golden standard

chemotherapeutic agent temozolomide. Furthermore, novel, unconventional

activation mechanisms for the TAMs in human GBM cells were investigated.

With the use of Western blotting, co-immunoprecipitation and in vitro kinase

assays, Axl was shown to both interact with and be activated by the RTK EGFR.

With the aid of qRT-PCR screens, EGFR was shown to promote GBM cell

invasion through the Axl/TIMP1/MMP9 signalling axis. Additionally,

iii

heterodimerisation of Axl with its sister RTK, Tyro3, in GBM cells was confirmed

using co-immunoprecipitation assays; the functional significance of this complex

was determined to be promotion of GBM cell survival. In conclusion, this thesis

demonstrates the importance of TAM signalling in GBM, identifies novel

molecular pathways employed by GBM cells for their survival, growth and

spread, and thereby further strengthens the case for targeting TAM receptors

as a novel therapeutic approach to combat both primary tumours as well as

secondary tumours arising from drug resistance.

iv

Table of Contents

Abstract ..................................................................................................... ii

LIST OF TABLES ........................................................................................ ix

LIST OF FIGURES ....................................................................................... x

DECLARATION ......................................................................................... xv

ACKNOWLEDGEMENTS ............................................................................. xvi

DEDICATION ........................................................................................... xvii

DISSEMINATION .................................................................................... xviii

ABBREVIATIONS ...................................................................................... xx

General introduction .................................................................................. 1

1.1 The hallmarks of cancer ............................................................................. 2

1.2 The normal brain ....................................................................................... 4

1.3 Brain cancer .............................................................................................. 5

1.3.1 Glioblastoma classification ....................................................................... 6

1.3.2 GBM treatment ....................................................................................... 8

1.3.3 GBM biological characteristics and targeted therapy .................................. 10

1.3.3.1 Proliferation ........................................................................................ 10

1.3.3.1.1 Targeted kinase inhibition with small molecule inhibitors .................. 15

1.3.3.2 Apoptosis evasion ............................................................................ 19

1.3.3.3 Necrosis .......................................................................................... 21

1.3.3.4 Angiogenesis ................................................................................... 22

1.3.3.5 Cell invasion .................................................................................... 23

1.3.4 The TAM family of RTKs ................................................................... 26

1.3.4.1 The structure of TAMs ......................................................................... 27

1.3.4.2 The TAMs ligands ................................................................................ 29

1.3.4.3 The TAM expression pattern ................................................................. 30

1.3.4.4 The TAM function ................................................................................ 31

1.3.4.5 TAM signalling .................................................................................... 32

1.3.4.6 TAMs and the central nervous system ................................................... 33

1.3.4.7 TAMs and GBM ................................................................................... 36

1.3.4.8 TAMs in other cancers ......................................................................... 38

v

1.3.4.9 TAMs as key targets in cancer .............................................................. 41

1.3.5 Project significance .......................................................................... 43

Materials and Methods ............................................................................ 44

2.1 Cell culture ............................................................................................... 45

2.1.1 Human cell lines ................................................................................. 45

2.1.2 Maintenance of cell lines ..................................................................... 45

2.1.3 Recovery of cryopreserved cells ........................................................... 46

2.1.4 Cell counting ...................................................................................... 47

2.1.5 Cell treatments ................................................................................... 47

2.2 Western Blotting ....................................................................................... 48

2.2.1 Cell lysis and protein extraction ............................................................ 48

2.2.2 Protein quantification .......................................................................... 48

2.2.3 Sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS–PAGE)

................................................................................................................. 49

2.2.4 Protein transfer .................................................................................. 50

2.3 Immunohistochemistry .............................................................................. 51

2.4 Molecular biology techniques ..................................................................... 53

2.4.1 Transient knockdown with short interfering RNA (siRNA) ........................ 53

2.4.2 Stable knockdown with short hairpin RNA (shRNA) ................................ 54

2.4.4 Synthesis of complementary DNA (cDNA) ............................................. 55

2.4.5 Real time quantitative real-time polymerase chain reaction-(qRT-PCR) .... 56

2.4.6 Determination of most stable reference genes for signalling studies ........ 57

2.4.7 Immunoprecipitation ........................................................................... 59

2.4.8 Luciferase reporter assay ..................................................................... 59

2.5 Functional assays ...................................................................................... 60

2.5.1 Cell survival/growth assay ................................................................... 60

2.5.1 Apoptosis assay .................................................................................. 61

2.5.2 Scratch wound assay .......................................................................... 62

2.5.3 Cell motility assay ............................................................................... 62

2.5.4 3D colony growth assay ...................................................................... 63

2.5.5 Cell invasion assay .............................................................................. 63

2.5.6 Cell cycle analysis ............................................................................... 64

2.5.7 In vitro kinase activity assay ................................................................ 64

2.5.8 Assessment of –tyrosine kinase phosphorylation status using an antibody

array system ............................................................................................... 65

vi

2.6 Statistical analyses .................................................................................... 66

Investigating TAM receptors and their ligands in GBM ................................. 67

3.0 Introduction ....................................................................................... 68

3.1 Results ............................................................................................... 69

3.1.1 The TAMs are overexpressed in GBM cells ................................................ 69

3.1.2 Gas6 activates Axl signalling in GBM cells ................................................. 74

3.1.3 Effect of novel proposed TAM ligands on Axl activation ............................. 76

3.1.4 TAM activation promotes cell growth/survival ........................................... 78

3.2 Discussion .......................................................................................... 80

3.2.1 TAM expression in the brain .................................................................... 80

3.2.2 RTK activation by classic TAM ligands ...................................................... 82

3.2.3 RTK activation by novel TAM ligands ........................................................ 84

3.2.4 TAM activation promotes cell growth ....................................................... 88

3.2.5 Summary .............................................................................................. 89

Axl blockade inhibits GBM cell invasion ...................................................... 90

4.0 Introduction ....................................................................................... 91

4.1 Results ............................................................................................... 92

4.1.1 Establishment of stable Axl knockdown cell lines ....................................... 92

4.1.2 Axl knockdown reduces cell growth/viability ............................................. 94

4.1.3 Axl knockdown inhibits cell migration ....................................................... 95

4.1.4 Selective Axl inhibition by two SMIs reduces cell viability ........................... 97

4.1.5 BGB324 selectively inhibits Axl activity in GBM .......................................... 99

4.1.6 BGB324 inhibits GBM cell growth and colony formation ........................... 101

4.1.7 Inhibition of GBM cell migration and invasion by Axl small molecule inhibition

.................................................................................................................. 106

4.2 Discussion ........................................................................................ 109

4.2.1 Establishment of stable Axl knockdown cell lines ..................................... 109

4.2.2 Genetic Axl knockdown inhibits proliferation and cell migration in GBM cells

.................................................................................................................. 111

4.2.3 Axl inhibition by a small molecule inhibitor ............................................. 112

vii

4.2.4 The Axl SMI BGB324 inhibits GBM cell growth and colony formation ......... 115

4.2.5 BGB324 inhibits cell migration and invasion in GBM ................................. 117

4.2.6 Summary ............................................................................................ 120

EGFR heterodimerises with Axl to promote GBM cell invasion .................... 121

5.0 Introduction ..................................................................................... 122

5.1 Results ............................................................................................. 124

5.1.1 EGF activates Axl via EGFR in SNB-19 cells ............................................. 124

5.1.2 EGFR kinase directly activates Axl kinase in vitro .................................... 127

5.1.3 EGFR induces upregulation of pro-invasive genes via Axl ......................... 129

5.1.4 EGFR positively regulates cell invasion signalling via Axl .......................... 131

5.1.5 EGFR stimulates cell cycle progression independently of Axl..................... 134

5.2 Discussion ........................................................................................ 137

5.2.1 EGF activates Axl via EGFR in SNB-19 cells ............................................. 137

5.2.2 EGFR induces pro-invasive genes upregulation via Axl ............................. 142

5.2.3 Summary ............................................................................................ 145

TAM downstream signalling in GBM ....................................................... 147

6.0 Introduction ..................................................................................... 148

6.1 Results ............................................................................................. 150

6.1.1 Tyro3 physically interacts with Axl ......................................................... 150

6.1.2 Axl and Tyro3 protein expression are interdependent .............................. 152

6.1.3 Tyro3 knockdown induces a cell cycle G2/M arrest................................... 154

6.1.4 Axl downstream signalling in GBM ......................................................... 157

6.1.5 EGFR-Axl downstream signalling in GBM ................................................ 162

6.2 Discussion ........................................................................................ 165

6.2.1 Tyro3 physically interacts with Axl ......................................................... 165

6.2.2 Axl and Tyro3 protein expression are interdependent .............................. 167

6.2.3 Axl-Tyro3 related signalling ................................................................... 169

6.2.3.1 Upregulated genes ......................................................................... 170

6.2.3.2 Downregulated genes ........................................................................ 173

6.2.3.3 NF-κB signalling ................................................................................ 177

viii

6.2.4 Tyro3 promotes GBM cell proliferation and survival ................................. 178

6.2.5 Axl downstream signalling in GBM ......................................................... 180

6.2.6 EGFR-Axl downstream signalling in GBM ................................................ 188

6.2.7 Summary ............................................................................................ 189

General Discussion ................................................................................ 191

7.1 Summary of key findings ......................................................................... 192

7.2 The TAMs as therapeutic targets .............................................................. 193

7.3 Concerns about TAM inhibition ................................................................. 196

7.4 Future directions ..................................................................................... 199

REFERENCES .......................................................................................... 201

APPENDIX .............................................................................................. 221

ix

LIST OF TABLES

Table 1: Cell lines used throughout the duration of project………………………….45

Table 2: List of primary antibodies used for immunoblotting..........................51

Table 3: List of secondary antibodies used for immunoblotting......................51

Table 4: Axl targeting sequences................................................................54

Table 5: Primers used in qRT-PCR……………………………………………………………57

Table 6: Housekeeping gene stability ranking…………………………………………...58

Table 7: Effects of Tyro3 knockdown on the expression tumour suppressors and

oncogenes…………………………………………………………………………………………155

Table 8: Axl inhibitors currently in development……………………………………….200

x

LIST OF FIGURES

Chapter 1- General introduction

Figure 1.1: The hallmarks of cancer……………………………………………………………3

Figure 1.2.: Glial function……………………………………….…………………………………5

Figure 1.3: Schematic representation of the genetic changes observed in glioma

subtype pathogenesis. ……………………………………………………………………………………7

Figure 1.4: New glioma classification…………………………………………………………8

Figure 1.5: Summary of PI3K and MAPK pathways…………………………………..…13

Figure 1.6: EGFR signalling……………………………………………………………………..15

Figure 1.7: EGFR activation…………………………………………………………………….16

Figure 1.8: Kinase inhibitor–protein interactions depicted by ribbon and

chemical structures……………………………………………………………………………..…17

Figure 1.9: Schematic representation of the intrinsic and extrinsic apoptotic

pathways……………………………………………………………………………..………………20

Figure 1.10: Schematic of GBM cell invasion………………………………………………24

Figure 1.11: Schematic representation of the 20 known RTK

subfamilies.…………………………………………………………………………………………..27

Figure 1.12: The TAM RTK family structure……………………………………………….28

Figure 1.13: Gas6 structure…………………………………………………………….………29

Figure 1.14: Gas6-LG /Axl-Ig complex………………………………………………………30

Figure 1.15: Summary of Axl signalling pathways………………………………………32

Figure 1.16: TAM signalling in cells of the CNS…………………………………………..35

Chapter 2- Materials and methods

Figure 2.1: Schematic representation of haemocytometer square indicating

xi

counting margins………………………………………………………………………..…………47

Figure 2.2: GL805a microarray panel display……………………………………………..53

Figure 2.3: T175a microarray panel display……………………………………………….53

Chapter 3-Investigating TAM receptors and their ligands in GBM

Figure 3.1: Cell line characteristics…………………………………………………………..70

Figure 3.2: TAM and ligand screen…………………….…………..…….……….…………71

Figure 3.3: Axl screening …………………………………………..……………………….….72

Figure 3.4: Tyro3 screening…………………………………………………………………….73

Figure 3.5: Axl tumour formation abilities …………………………………………………73

Figure 3.6: Effect of Gas6 stimulation on TAM phosphorylation and signalling in

GBM cells……………………………….………………………………………………………….…75

Figure 3.7: Effect of ligand stimulation on TAM phosphorylation and in GBM

cells……………………………………………………………………………………………………..77

Figure 3.8: MTS assay for Gas6 stimulation……………………………………………….79

Figure 3.9: Binding of Gas6 on Axl receptor vs Tyro3/MerTK…………………….…82

Figure 3.10: Galectin-3 structure……………………………………………………………..84

Figure 3.11: Presumed role of galectins in angiogenesis and

chemoresistance……………………………………………………………………………………85

Figure 3.12: Tub proteins structure……………………………………………………….…86

Figure 3.13: Tulp-1 facilitates phagocytosis as MerTK-dependent bridging

molecules…………………………………………….……………………………………………….86

Chapter 4-Axl blockade inhibits GBM cell invasion

Figure 4.1: Puromycin killing curve…………………………………………………………..93

Figure 4.2: Western blot for selected cell clones confirming Axl

knockdown………………………………………….……………………………………………….94

Figure 4.3: MTS assay for Axl knockdown clones……………………………….……..95

Figure 4.4: Effect of Axl knockdown on cell migration…………………………........96

Figure 4.5: MTS assays for cells treated with BGB324…………………………........98

xii

Figure 4.6: Comparative efficacies of BGB324 for inhibition of Axl, Tyro3 and

Akt phosphorylation in GBM cells…………………………..……………………………….100

Figure 4.7: Effect of BGB324 on long-term growth of GBM cell colonies in

3D……………………………………………………………………………………………………..101

Figure 4.8: Axl blockade affects cell division ………………………..………………….102

Figure 4.9: Annexin V vs PI staining……………………………………………………….103

Figure 4.10: Effect of TMZ on viability of SNB-19 and UP007 cell lines.………..104

Figure 4.11: Effect of BGB324 on GBM cell growth in combination with

TMZ……………………………………………………………………………………………………105

Figure 4.12: Effect of BGB324 on GBM cell migration, motility and matrix

invasion………………………………………………………………………………………………107

Figure 4.13: Invasion assay for SNB-19 and UP007 cells following with BGB324

using Gas6 as chemoattractant………………………………………………………………108

Figure 4.14: Diagrammatic summary of lipid mediated transfection……………109

Figure 4.15: Simplified schematic representation of viral shRNA…………………110

Figure 4.16: Schematic representation of inhibitor binding………………………..113

Figure 4.17: Chemical structures of BGB324 and BMS777607…………………….114

Chapter 5- EGFR heterodimerises with Axl to promote GBM cell

invasion

Figure 5.1: Western blot of time-course of EGFR, Axl and Tyro3 phosphorylation

by EGF………………………………………………………………………………………………..125

Figure 5.2: Western blot of Axl and EGFR phosphorylation by EGF and its

inhibition by gefitinib and BGB324………………………………….……………………...126

Figure 5.3: Hetero-interaction of EGFR and Axl RTKs………………………………..127

Figure 5.4: In vitro kinase activity of recombinant Axl and EGFR, with Axl Y779

peptide as substrate…………………………………………………………………………….128

Figure 5.5: qRT-PCR analysis…………………………………………………………………130

Figure 5.6: Western blot showing STAT3 phosphorylation…………………………130

Figure 5.7: EGFR-Axl hetero-interaction regulates the balance between

expression for invasive and proliferative signalling in SNB-19 cells……………132

Figure 5.8: qRT-PCR analysis of MMP9, MMP2 and CD44 genes………………….133

xiii

Figure 5.9: Quantitative RT-PCR analysis of CCND1 gene expression in SNB-19

cells altered by EGFR activation, with and without the influence of

Axl…………………………………………………………………………………………………..…134

Figure 5.10: Effect of gefitinib on viability of SNB-19 and UP007 cell lines as

shown by MTS assay…………………………………………………………………………….135

Figure 5.11: Effect of gefitinib on GBM cell growth in combination with

BGB324………………………………………………………………………………………………136

Figure 5.12: Schematic summary of this study, showing hetero-interaction

between EGFR and Axl............................................................................146

Chapter 6 –TAM downstream signalling in GBM

Figure 6.1: Hetero-interaction of Axl and Tyro RTKs……………………..………….151

Figure 6.2: Effect of gene silencing by RNAi on TAM phosphorylation and in

GBM cells…………………………………………………………………………………………….153

Figure 6.3: Cell cycle analysis after siTyro3 knockdown…………………………….156

Figure 6.4: Effect of Tyro3 knockdown on cell survival………………………………157

Figure 6.5: Relative pixel densitometry bar graph indicating changes in

phosphorylation levels of kinases……………………………………………………………158

Figure 6.6: Relative pixel densitometry bar graph indicating changes in

phosphorylation levels of kinases……………………………………………………………159

Figure 6.7: NF-κB expression levels in SNB-19 and UP007 cells………………….161

Figure 6.8: Relative pixel densitometry bar graph indicating phosphorylation

levels of kinases……………………………………………………………………………….….163

Figure 6.9: Relative pixel densitometry bar graph indicating levels of

kinases……………………………………………………………………………………….………164

Figure 6.10: Proposed model of the mechanism of post-translational Tyro3

regulation by Axl…………………………………………………….……………………………169

Figure 6.11: Proposed mechanism of Tyro3-mediated MOS kinase in GBM cell

survival and proliferation………………………………………………………………………171

Figure 6.12: Cyclin D1 function………………………………………………………………175

Figure 6.13: Schematic representation of NF-κB activation………………………..177

xiv

Figure 6.14: Proposed mechanism of Tyro3 mediated GBM cell survival and

proliferation………………………………………………………………………………………..179

Figure 6.15: Schematic representation of the canonical Wnt/β-catenin signalling

pathway……………………………………………………………………….…………………….182

Figure 6.16: Schematic representation of AMPK’s subunits and its activation

process………………………………………………………………………………………….……184

Figure 6.17: PI3K activation…………………………………………………………………..185

Figure 6.18: Summary of Axl-mediated activation of downstream signalling

effectors resulting in increased, invasion, survival and proliferation of GBM

cells……………………………………………………………………………………………………187

Figure 6.19: Summary of EGFR-Axl mediated activation of intracellular

downstream signalling effectors resulting in increased, invasion, survival and

proliferation of GBM cells………………………………………………………………………189

Chapter 7-General discussion

Figure 7.1: Schematic drawing: pre-treatment and during treatment with

angiogenic agents in GBM……………………………………………………….……….……195

Figure 7.2: A TAM-Regulated Cycle of Inflammation….……………………………..198

xv

DECLARATION

Whilst registered as a candidate for the above degree, I have not been

registered for any other research award. The results and conclusions embodied

in this thesis are the work of the named candidate and have not been submitted

for any other academic award.

xvi

ACKNOWLEDGEMENTS

Although it is not the traditional way, I want to first of all, thank God for

reasons He knows best.

Secondly, I want to thank my supervisor and mentor Dr. Sassan Hafizi,

for his scientific, but also personal, support and advice. He has always

encouraged my initiatives and ideas, and through our very helpful discussions

I have learned a lot about the exciting field of cancer research. His guidance

and will to help were critical for the fulfilment of this work, and I will always be

grateful for his contribution in turning a student into a scientist.

I greatly appreciate the advice and guidance of my second supervisor

Dr. Qian An, my third supervisor Professor Geoffrey Pilkington and the entire

Neuro-oncology group for taking the time to regularly monitor and discuss the

progress of this work, but also for their genuine interest and encouragement.

My life in Portsmouth would have been much more difficult without the

support of my friends, whom I owe the greatest thank you. I have shared the

best moments of the past 4 years with many lovely and just-weird-enough-to-

like-me people, so here I’d like to thank: Maria Papanikolaou and Francesca

Pieropan for putting up with my daily rants and freak outs, as well as for

standing by me during the most difficult times of my life. I would also like to

thank Andrea Rivera for unhealthy food, movie nights and random words of

wisdom, as well as Irene De La Rocha, Louise Kelly, Ashik Kalam, Rebecca

Mather, Kathleen Keatley, Emily Pinkstone, Andrienne Iacovou, Andrea

Georgiou and Salman Goudarzi for just being awesome. I also would like to

thank my best friend Mike Tochnitis for his undivided support throughout the

years. I wouldn’t be the person I am today if it wasn’t for his friendship

My family has given me everything a family can give; I am forever

grateful to my rock aka my dad, my sister, my brother in law and my lovely

Godson for their love and support, and for always being there for me in happy

and difficult times, despite the geographical distance.

I acknowledge here the financial support of this work by Headcase

Cancer Trust.

xvii

DEDICATION

I dedicate this thesis to my angel up in heavens always shining her

light down me.

You were my inspiration, my moral compass and the reason I

decided to become a scientist.

You are always with me in mind, spirit and soul,

my guiding light

forever.

I love you mum.

xviii

DISSEMINATION

Meetings attended:

ESMO, Signalling in cancer symposium, March 2016.

IBBS, May 2015. Oral Presentation: “Axl as a novel therapeutic target in

brain tumour cell progression”.

AACR annual meeting, April 2015. Poster: “Hetero-activation amongst TAM

receptor tyrosine kinases and via EGFR in human glioma cells”.

IBBS, May 2014. Poster: “Axl signalling promotes glioma cell invasion”

EMBO meeting, May 2014. Poster: “Axl signalling promotes glioma invasion”

Biochemical society meeting, January 2014.

University of Portsmouth Science Faculty Poster Presentation Day, July, 2013.

Poster: “The role of Gas6-Axl signalling in glioblastoma multiforme”.

BNOS meeting, July 2013. Poster: “Ligand-dependent and –independent Axl

signalling in Glioblastoma Multiforme”.

Hammer out brain tumour patient conference, March, 2013. Poster: “The

role of a pro-survival signalling pathway in the survival of glioma cells”.

6th International Summer School ISS 2013-Part A in Piran, Slovenia.

BNOS meeting, June 2012.

xix

Full publications:

Vouri, M., An, Q., Pilkington, G., & Hafizi, S. (2016). Axl-EGFR receptor

tyrosine kinase hetero-interaction provides EGFR with access to pro-invasive

signalling in human cancer cells. (submitted)

Vouri, M., An, Q., Pilkington, G., & Hafizi, S. (2016) Tyro3 promotes cancer

cell survival through Axl signalling. (In Preparation)

Vouri, M., An, Q., Birt, M., Pilkington, G., & Hafizi, S. (2015). Small molecule

inhibition of Axl receptor tyrosine kinase potently suppresses multiple

malignant properties of glioma cells. Oncotarget.

An Q, Filmore.L.H, Vouri.M, Pilkington.G (2014) Brain tumour cell line

authentication, an efficient alternative to capillary electrophoresis by using a

microfluidics-based system. Neuro-oncology.

xx

ABBREVIATIONS

APS Ammonium persulfate ATO Arsenic trioxide BBB Blood brain barrier BMI1 B cell-specific Moloney murine leukaemia virus integration site 1 CDK Cyclin dependent kinase CML Chronic myelocytic leukaemia FDA Food and Drug Administration CNS Central nervous system co-IP Co-immunoprecipitation DN Dominant negative ECM Extracellular matrix EGF Epidermal growth factor EGFR Epidermal growth factor receptor EMT Epithelial to mesenchymal transction eNOS Endothelial nitrix oxide synthase EZH2 Enhancer of zeste homolog 2 GBM Glioblastoma multiforme GSK3β Glycogen synthase kinase 3β HIF Hypoxia induced factor IDH1 Isocitrate dehydrogenase 1 Ig Immunoglobulin LG Laminin G-like MAPK Mitogen-activated protein kinases MITF Microphthalmia-associated transcription factor MMP Matrix metalloproteinases NF-κB Nuclear factor -κB NPC Neural progenitor cell GSC Glioblastoma stem cell NSC Neural stem cell NSCL Non-small cell lung cancer PDGFRα Platelet-derived growth factor receptor α PI3K Phosphatidylinositol-4,5-bisphosphate 3-kinase PLCγ Phosphoinositide phospholipase Cγ PTEN Phosphatase and tensin homolog RB Retinoblastoma ROS Reactive oxygen species RTK Receptor tyrosine kinase SH2 Src Homology 2 shRNA Short hairpin RNA TAM Tyro3, Axl , MerTK TIMP1 Tissue inhibitor of matrix metalloproteinases 1 TMZ Temozolomide TNF Tumour necrosis factor VEGF Vascular endothelial growth factor

https://en.wikipedia.org/wiki/B_cell

https://en.wikipedia.org/wiki/Murine_leukemia_virus

1

Chapter 1

General introduction

2

1.1 The hallmarks of cancer

The word “cancer” is used to describe a condition where the body’s cells

begin to grow and reproduce uncontrollably, are able to avoid cell death and

proceed to invade and destroy healthy tissue, including organs (1). Cancer is

one of the most studied diseases of our century as it poses the leading cause

of death. Evidently, cancer is a multi-process and multi-gene disease that drives

normal cells to become progressively malignant. Hanahan and Weinberg first

reviewed a specific series of steps that cells must undergo to transform and

become cancerous so-called the ‘Hallmarks of cancer’ (2). Initially, it was

thought that there are six essential alterations in cell physiology that dictate

malignant growth, which were later on revised to ten (3). These include: self-

sufficiency in growth signals, insensitivity to anti-growth signals, evasion of

programmed cell death (apoptosis), limitless replicative potential, genomic

instability, sustained angiogenesis, tissue invasion and metastasis,

reprogramming of energy metabolism, tumour-promoting inflammation and

evading immune destruction (2, 3)(summarised in Figure 1.1).

Cancer cells can achieve the above hallmarks in many ways (Figure 1.1).

Firstly, growth autonomy can be achieved either through the overexpression of

growth receptors and secretion of growth factors leading to autocrine

activation, or acquired mutations that lead to the constitutive activation of

growth pathways, or downregulation of negative regulators. Furthermore,

insensitivity to anti-growth signals is achieved through the permanent

suppression of genes related to cell cycle inhibition and cell death, which in turn

leads to increased genomic instability and acquisition of further mutations.

Cancer cells are also able to escape shortening of telomeres and hence

3

senescence/death due to their ability to upregulate telomerase, an enzyme that

maintains telomeric length at the end of each cell cycle. Another way cancer

cells evade cell death is by acquiring mutations that disable DNA damage

recognition and induction of programmed cell death. Due to the high energetic

demands of cancer cells, they acquire the ability to orchestrate production of

new vasculature around the tumour to increase supply of oxygen and nutrients.

Furthermore, they deregulate metabolic pathways in order to maximize energy

generation. Moreover, cancer cells develop mutations which allow them to

excrete enzymes that breakdown the extracellular matrix and cell-cell adhesions

thereby enabling them to invade surrounding tissue or metastasise to other

organs. Lastly, it has become apparent that cancer cells are able to evade

immune recognition, as well as induce tumour inflammation that recruits

seemingly normal cells that contribute to the acquisition of hallmark traits by

creating the tumour microenvironment (2, 3).

Figure 1.1: The hallmarks of cancer. Acquired capabilities of cancer cells providing them with an evolutionary advantage as depicted by Hanahan and Weinberg (3).

4

1.2 The normal brain

The human brain is the main organ of the central nervous system (CNS). It

is located in the head, protected by the skull. The human brain consists of about

1012 cells, of which around 1011 are considered to be neurons with the remaining

9 × 1011 being glia. Neurons are electrically excitable cells composed of a cell

body, dendrites and an axon. Neurons form networks that transmit information

to the rest of the body through electrical and chemical signals (4). There are

three main types of glial cells in the mature CNS: astrocytes, oligodendrocytes

and microglia (5). Astrocytes are star-shaped cells found in the brain and spinal

cord. Astrocytic processes envelop neuron-made synapses, thus regulating

homeostatic maintenance, neurogenesis, neural cell migration and control of

brain blood flow (5) (Figure 1.2). Oligodendrocytes are also restricted to the

brain and produce myelin (fatty white substance) which surrounds the axon of

neuronal cells to form an electrically insulating layer essential for the rapid

conduction of action potentials in the CNS (6) (Figure 1.2). Microglia are the

resident immune cells of the CNS and play important roles in combating brain

infections and inflammation, which includes engulfing dead cells and debris (7)

(Figure 1.2). Other important cell types present in the brain include the

ependymal cells which are cuboidal and multi-ciliated structures (8). Ependymal

cells line the interface between the brain parenchyma and the ventricular

cavities and regulate the movement of cerebrospinal fluid (CSF) through the

ventricular system (8) (Figure 1.2).

5

Figure 1.2: Glial function. Glia interact with neurons and the surrounding blood vessels. Oligodendrocytes wrap myelin around axons. Astrocytes extend processes that ensheath blood vessels and synapses. Microglia keep the brain under surveillance for damage or infection. Ependymal cells line the ventricles thus regulating CSF flow (4).

1.3 Brain cancer

Brain cancer refers to the intracranial mass created by abnormal and

uncontrolled cell division, arising either from brain cells or spread from cancers

primarily located in other organs also referred as metastatic tumours.

Glioblastoma multiforme (GBM) is the most common malignant primary brain

tumour constituting 45.2% of all malignant CNS tumours and 80% of all primary

malignant CNS tumours (9). Astrocytomas (WHO grade I to IV) are the most

common type of glioma with the major associated tumours being pilocytic

astrocytomas (Grade I), diffuse astrocytoma (Grade II), anaplastic astrocytoma

(Grade III) and glioblastoma multiforme (Grade IV) (10). GBMs primarily affect

adults, with a peak incidence ranging between 45 and 75 years and are

6

preferentially located within the cerebral hemispheres. Most GBMs arise rapidly

de novo, with no previous lesions and are termed primary GBM (9).

1.3.1 Glioblastoma classification

Gliomas are classified morphologically to the glial cell type they

resemble. In addition to the morphological classification they are usually

distinguished by the WHO grading system. Lower grade tumours (e.g. grade I

and II) are well differentiated and resemble normal tissue. Higher grade

tumours (e.g. grade III and IV) are anaplastic and experience high cell atypia

and mitotic activity (11).

Genomic analyses of glioma samples led to the molecular

characterisation of four different subtypes termed Neural, Proneural, Classical

and Mesenchymal (summarised in Figure 1.3) (12).

The classical subtype has been widely associated with amplification of

chromosome 7 and deletion of chromosome 9p21.3 and 10. These result in

amplification of the epidermal growth factor receptor (EGFR) and disruption of

the retinoblastoma (RB) pathway. Glioblastoma cells belonging to the classical

subtype also express high levels of neural precursor and stem cell marker

Nestin. Notch and Sonic hedgehog signalling pathways are also over expressed

(12, 13).

The proneural subtype is characterised by mutations in isocitrate

dehydrogenase 1 (IDH1) and amplification of the platelet–derived growth factor

receptor (PDGFRα). Tumours belonging to the proneural type also show high

levels of p53 mutations, loss of heterogeneity, amplification of chromosome 7,

loss of chromosome 10 and high expression of genes involved in the

7

development of oligodendrocytes such as NKX-2, OLIG2, SOX and PDGFRα. In

cases where no amplification of PDGFRα was observed, PIK3CA mutations were

present (12, 13).

The Neural subtype is characterised by high levels of neuronal markers

such as NEFL, GABRA1, SYT1 and SLC12A5 (12, 13).

The Mesenchymal subtype is characterised by hemizygous deletion of

Neurofibromin (NF) 1 (a negative regulator of the Ras signalling pathway) and

mutations in phosphatase and tensin homologue (PTEN). High expression of

the mesenchymal markers YKL40, MET, CD44 and genes belonging to the

Tumour necrosis factor (TNF) and Nuclear Factor-Kappa B (NF-κB) pathways is

observed. The mesenchymal type is highly associated with necrosis and

inflammatory infiltrates (12, 13).

Figure 1.3: Schematic representation of the genetic changes observed in glioma subtype pathogenesis. EGFR, epidermal growth factor receptor; PTEN, phosphatase and tensin homolog; TNF, tumour necrosis factor; PDGFRA, platelet-derived growth factor receptor–α; IDH, isocitrate dehydrogenase; PI3K, phosphoinositol 3–kinase; HIF, hypoxia-inducible factor; TIC, tumour initiating cell; BCPC, brain cancer propagating cell (14).

8

Recently a new categorisation was proposed, splitting gliomas into 3

molecular subgroups based on the tumour mutation status of the IDH gene and

the co-deletion or loss of heterozygosity at chromosome 1p/19q combined with

a mutation in telomerase reverse transcriptase (TERT) or Alpha

Thalassemia/Mental Retardation Syndrome X-Linked (ATRX), conferring

telomere maintenance by telomerase or through the lengthening of telomeres,

respectively (15, 16) (Figure 1.4).

Figure 1.4: New glioma classification. Schematic representation of the genetic changes observed in glioma subtype pathogenesis according to the new classification. IDH, isocitrate dehydrogenase; PTEN, phosphatase and tensin homolog; TERT, telomerase reverse transcriptase; ATRX, Alpha Thalassemia/Mental Retardation Syndrome X-Linked (17).

1.3.2 GBM treatment

The current gold standard therapy for the treatment of GBM includes

surgical tumour resection followed by the use of temozolomide (TMZ), an oral

cytotoxic DNA-alkylating chemotherapeutic adjuvant, and radiotherapy. Studies

by Stupp and colleagues demonstrated that administration of TMZ in

9

combination with radiation therapy followed by adjuvant TMZ treatment for 6

months improved median overall survival compared to radiation alone (14.6

months compared to 12.1 months) and increased twofold the 2-year survival

from 10.4% to 26.1% (18). Despite recent advancements in the understanding

of glioma pathology, efforts to develop a more effective treatment for GBM

patients have not been successful. Gliomas are highly heterogeneous tumours,

both phenotypically and genetically, thus making single targeted therapies

unlikely to be viable. Moreover, treatment failure has also been attributed to

the heterogeneity of gliomas which is thought to arise from the malignant

transformation of neural progenitor cells (NPCs) present in the subventricular

zone, the lining of the lateral ventricles, the dentate gyrus and the subcortical

white matter of the adult brain. The first evidence supporting this theory derived

from the isolation of glioblastoma stem cells (GSCs) from tumours, which were

shown to share the same characteristics with NPCs: high motility, association

with blood vessels and white matter tracts, possession of immature antigenic

phenotypes, and activation of developmental signalling pathways (10).

Subsequently, several theories were formed to explain how the transformation

of NPCs results in tumour formation. The maturation arrest theory states that

a glial progenitor cell after it becomes transformed generates abnormal progeny

which accumulate additional mutations and impaired differentiation potential

leading to tumorigenesis (10). These cells are thought to be resistant to therapy

through a variety of ways including more effective DNA damage repair,

autophagy (protein degradation system which protects cells from death)

activation, as well as the use of differential signalling pathways (19). Another

major barrier to successful GBM treatment is the blood-brain barrier (BBB). The

10

BBB is an extensive network of capillaries formed by endothelial cells enforced

by astrocytes and pericytes. The BBB acts a filter for the brain preventing the

entrance of various substances from the blood stream (20). Consequently, the

identification of essential regulatory pathways and multifaceted targeted

therapy is required to effectively improve GBM patient cell survival. As well as

the design of BBB permissive therapies.

1.3.3 GBM biological characteristics and targeted therapy

Gliomas exhibit high heterogeneity phenotypically resulting from a range

of cellular features such as: unrestricted proliferation, extensive invasion,

neovascularisation, necrosis, and resistance to apoptosis. Gliomas have the

unique ability to diffusely invade and infiltrate surrounding brain tissue,

intertwining themselves among native cells and so making complete surgical

resection impossible (21). These hallmark biological processes and their links to

specific genetic aberrations and associated signalling pathways will be discussed

in the following subsections in relation to attempts for targeted therapies.

1.3.3.1 Proliferation

The first GBM hallmark is uncontrolled proliferation. Normal cells activate

mitogenic signalling pathways upon growth factor binding, cell–cell adhesion,

and/or contact with extracellular matrix (ECM) components. These signals

activate transmembrane receptors which in turn promote gene expression

leading to cell division. GBM cells overcome normal impositions on the control

of growth signals through multiple mechanisms which mainly include activation

of receptor tyrosine kinases (RTKs).

11

Constitutive activation of receptors is achieved through overexpression

or genetic amplification of growth factor receptor genes, as well as through

gene mutations that lead to ligand-independent signalling. Alternatively,

intracellular signalling pathways can be constitutively activated through

activating or suppressing mutations in propagating signalling proteins (14, 21).

Amongst the most deregulated signalling pathways are the mitogen-activated

protein kinase (MAPK) and the phosphatidylinositol-4,5-bisphosphate 3-kinase

(PI3K) pathways (summarised in Figure 1.5).

The MAPK pathway is activated by growth factor signalling, mutation, or

overexpression and involves the sequential activation of Ras, Raf, MEK and ERK

kinases which leads to the relocation of ERK kinase to the nucleus where it

phosphorylates nuclear transcription factors that induce the expression of genes

promoting cell cycle progression (22). Mutations that constitutively activate Ras

are the most common in human cancer; however very few have been reported

in glioma, thus Ras over-activation is believed to be achieved primarily through

RTK overexpression/constitutive activation (23). One important MAPK

downstream transcription target is cyclin D1, which a crucial cell cycle

progression protein (24). Under normal conditions, the cell cycle is halted by

the RB protein which binds and sequesters the E2F family of transcription

factors, thus preventing transactivation of genes essential for progression of

the cell cycle. Growth factor stimulation of the MAPK cascade leads to the

induction of cyclin D1 which associates with the cyclin-dependent kinases

(CDK)4 and CDK6 which in turn phosphorylate RB, thus enabling E2F mediated

gene expression which allows entry to the S-phase and cell cycle progression

(25, 26). In GBM the cell cycle checkpoint is deregulated in a variety of ways

12

allowing uncontrolled division. The Rb1 gene is found mutated in ∼25% of GBM

patients (27). RB activity is also lost through the inactivation of p16Ink4a, a

negative regulator of CDK4 and CDK6, or amplification of CDK4 (28).

The PI3Ks are heterodimers comprised of a catalytic and a regulatory

subunit and are activated in response to mitogenic signals. The PI3Ks catalyse

the phosphorylation of phosphatidylinositol-4,5-bisphosphate [PtdIns(4,5)P2] to

produce phosphatidylinositol (3,4,5)-trisphosphate [PtdIns(3,4,5)P3], thus

creating docking sites for a multitude of signalling proteins (29). PIK3CA gain-

of-function point mutations have been detected with 15% frequency in GBM

(30). Furthermore, PI3K is directly antagonised by PTEN, which is found

inactivated in 50% of high-grade gliomas either by mutations or epigenetic

mechanisms, each resulting in uncontrolled PI3K signalling (31, 32). The

phosphoinositide-dependent kinase (PDK1) and Akt are then recruited to the

membrane where Akt is activated via phosphorylation of two key residues, T308

and S473, respectively (33). Assessment of the phosphorylation status of these

residues is often the method of choice for monitoring PI3K pathway activity in

cell lines and primary tumours, including GBM samples, 85% of which have

been reported to display activated Akt. Akt phosphorylates many proteins

involved in the regulation of cell growth, proliferation, metabolism, and

apoptosis (34).

13

Figure 1.5: Summary of PI3K and MAPK pathways. Upon PI3K activation by RTKs, PIP2 is converted into PIP3 thus leading to PDK1 recruitment and downstream activation of Akt. PTEN negatively regulates PI3K signalling by dephosphorylating PIP3, converting it back to PIP2. Akt activation leads to both positive and negative regulation of a wide range of target proteins (not all shown). The Ras signalling pathway is also activated by growth factors binding cell membrane RTKs through GRB2 (growth factor receptor-bound protein 2), SOS, Ras, Raf, MEK (mitogen-activated ERK kinase) and ERK, leading to cell proliferation. BAD, BCL-2-associated agonist of cell death; FOXO, forkhead box O; RHEB, Ras homologue enriched in brain; TSC2, tuberous sclerosis 2 (35).

As mentioned earlier, the MAPK and PI3K pathways are mainly activated

through binding of growth factor ligands to RTKs. Cancer cells upregulate

autocrine loops, undergo genomic amplification, and/or mutation of the

receptor leading to constitutive activation in the absence of ligand. The most

prominent deregulated receptor identified to be essential for gliomagenesis is

EGFR. EGFR gene amplification indeed characterises the classical subgroup of

GBM, which accounts for 25%–30% of cases. Further characterisation showed

that EGFR mutations often occur simultaneously with EGFR rearrangement

and/or focal amplification; these occur in 57% of GBM cases, making it the most

14

prevalent deregulated receptor in GBM (36). Wild-type EGFR is classically

activated through ligand binding such as epidermal growth factor (EGF),

transforming growth factor-, heparin-binding EGF-like growth factor,

amphiregulin, epiregulin, betacellulin, and epigen (37). In GBM, however,

activation of EGFR occurs independently of ligand through cross-talk between

RTKs at the cell surface or between EGFR mutants and the wild-type receptor

(38, 39). EGFR mutations in GBM are common with the leading variant being

EGFR variant (v)III (EGFRvIII) which results from truncation of exons 2-7 at the

extracellular domain rendering it constitutively active (40). Moreover, EGFRvIII,

unlike the wildtype EGFR, fails to internalise upon activation and therefore has

increased stability in the plasma membrane, contributing further to its sustained

tumorigenic signalling (41) (Figure.1.6).

EGFRvIII is a highly validated glioma target as evidenced by the capacity

of activated EGFR mutants to enhance tumorigenic behaviour of human GBM

cells by reducing apoptosis and increasing proliferation (42, 43). Consistent with

enhanced apoptosis resistance by EGFRvIII mediated through upregulation of

Bcl-XL (a pro-survival member of the Bcl-2 family)(43), activated EGFR has also

been shown to confer radio- and chemo-resistance to GBM cells by sequential

activation of Ras and both the MAPK and PI3K cascades (44). Moreover, it has

been recently identified that EGFR and EGFRvIII are often co-expressed and are

able to heterodimerise resulting in enhanced downstream signalling (45).

15

Figure 1.6: EGFR signalling. Binding of EGF to EGFR, activates downstream signalling pathways including the MAPK and PI3K/AKT. The EGFRvIII mutant is constitutively active and can induce signal transduction in the absence of EGF by heterodimeric association with WT EGFR. Activated EGFR phosphorylates EGFRvIII triggering nuclear transport of EGFRvIII, enhanced phosphorylation of STAT3, and increased EGFRvIII-STAT3 interaction in the nucleus, which drives transformation (45).

1.3.3.1.1 Targeted kinase inhibition with small molecule inhibitors

Generally, protein kinases catalyse the transfer of the terminal

phosphate of ATP to substrates that contain either serine, threonine or tyrosine

residues (46, 47). All kinases typically share a conserved arrangement of

secondary structure elements that are arranged into 12 subdomains that fold

into a bi-lobed catalytic core structure with ATP binding taking place in a deep

cleft located between the lobes (46, 47). ATP binds the cleft and forms

hydrogen bonds with the kinase 'hinge', the fragment that connects the amino-

and carboxy-terminal kinase domains (46, 47). The ribose and triphosphate

groups of ATP bind in a hydrophilic channel extending to the substrate binding

site that features conserved residues that are essential to catalysis. All kinases

have a conserved activation loop, which regulates the kinase activity and

16

contains conserved DFG (Asp-Phe-Gly) and APE (Ala-Pro-Glu) motifs at the start

and end of the loop, respectively (46, 47) (example for EGFR is given in Figure

1.7). The activation loop is fluid and can adopt a number of conformations with

the extremes being catalytically competent and usually phosphorylated form,

and an 'inactive' form in which the activation loop blocks the substrate binding

site (46, 47).

Figure 1.7: EGFR activation. The kinase domain of the EGF receptor is kept in an inactive state by a leucine wedge. ATP binds the inactive EGFR kinase causing the displacement of the activation fragment thus causing receptor heterodimerisation and subsequent autophosphorylation (48).

There are four main types of small molecule inhibitors that can block

kinase activity depending on their binding site. Type 1 inhibitors recognise the

active (DFG-in) conformation of the kinase and compete with ATP for the

binding site. Type 1 inhibitors typically consist of a heterocyclic ring system that

17

occupies the purine binding site, where it serves as a scaffold for side chains

that occupy the adjacent hydrophobic regions I and II (49) (Figure 1.8A).

Type 2 inhibitors recognise the inactive kinase conformation (‘DFG-out’)

of the kinase. Movement of the activation loop to the DFG-out conformation

results in the exposure of an additional hydrophobic binding site directly

adjacent to the ATP binding site, which the compound binds thus blocking

kinase activation (49) (Figure 1.8B).

Figure 1.8: Kinase inhibitor–protein interactions depicted by ribbon (left) and chemical structures (right). The chemical structures depict hydrophobic regions I and II of ABL1 forming hydrogen bonds with the kinase inhibitor. The DFG motif (pink), hinge and the activation loop of ABL1 are indicated in the ribbon representations. The kinase inhibitors are shown in light blue. A. ABL1 in complex with the type I ATP-competitive inhibitor PD166326. B. ABL1 in complex with the type II inhibitor imatinib. The allosteric pocket exposed in the DFG-out conformation is indicated by the blue shaded area (right)(49).

18

Allosteric inhibitors bind outside the ATP-binding site and thereby

modulate kinase activity. These inhibitors present the highest degree of kinase

selectivity due to the exploitation of binding sites and regulatory mechanisms

that are unique to a particular kinase (49).

Lastly, covalent inhibitors form an irreversible, covalent bond to the

kinase active site by reacting with a nucleophilic cysteine residue and

irreversibly blocking binding of ATP to the kinase, thereby rendering the kinase

inactive (49).

Due to the high prevalence of EGFR amplification and the EGFRvIII in

GBM, several attempts have been made to target them clinically using both

reversible and irreversible ATP-competitive SMIs. Some of the reversible

inhibitors include erlotinib (Genentech/Roche/OSI), gefitinib (AstraZeneca),

lapatanib (GlaxoSmithKline) and PKI166 (Novartis), and the irreversible

inhibitors include canertinib (Pfizer/Warner-Lambert) and pelitinib (Wyest-

Ayerst) (50). Though a number of EGFR inhibitors have been developed,

gefitinib, an ATP mimetic of the anilinoquinazoline family that covalently inhibits

the EGFR kinase, represents the best explored inhibitor in the clinic for the

treatment of GBM (51). The anti-tumour activity of gefitinib is independent of

the expression level of EGFR, but is heavily determined by its ability to inhibit

anti-apoptotic signals (52). Moreover, in in vitro studies gefitinib reduced the

proliferation of glioma cells by autophagic mechanisms involving AMPK

activation (51). Clinically as determined in phase II clinical trials gefitinib is well

tolerated and displays anti-tumour activity, but the median overall survival time

in GBM patients was only 38.4 weeks from treatment initiation (50). Moreover,

19

it was shown that gefitinib is able to reach high concentrations in the tumour

tissue and can efficiently dephosphorylate EGFR; however, it was unable to shut

down regulatory circuits that promote sustained downstream signal

transduction (53).

Even though gefitinib is well tolerated and effective in GBM patients, the

recurrent problem of resistance arising from various mechanisms such as

concomitant PTEN loss or MET activation (54) limits its efficacy in the clinic.

Additionally, recent studies show that plasma concentrations of gefitinib

following therapy were only 6–11% of the starting dose indicating that the poor

response observed in the clinic could also be due to insufficient delivery of the

SMIs to the tumour (50).

1.3.3.2 Apoptosis evasion

Another hallmark of GBMs is resistance to death-inducing stimuli such as

radiotherapy and chemotherapy. GBM cells have shown to evade cell death

through the upregulation of molecules involved in mitogenic signalling such as

RTKs, the PI3K–PTEN–Akt signalling axis (55), as well as regulatory and effector

molecules residing in classical cell death networks of both extrinsic (death

receptor-mediated) and intrinsic (mitochondria-dependent) mediated apoptosis

(56). The intrinsic apoptotic pathway is regulated by the balance of pro-

apoptotic and anti-apoptotic members of the Bcl-2 family of proteins.

Upregulation of anti-apoptotic Bcl-2 family members (BAK, BAD, BID, BAX, Bcl-

XL, Mcl-1) protects the mitochondrial membrane integrity and inhibits

cytochrome c release, thus negatively affecting the caspase cascade and the

20

apoptotic program (56) (The extrinsic and intrinsic pathways are summarised

in Figure 1.9).

Figure 1.9: Schematic representation of the intrinsic and extrinsic apoptotic pathways. The extrinsic apoptotic pathway is activated by induction of the death receptor on the cell membrane. The intrinsic pathway is usually activated by the loss of the mitochondrial membrane integrity. Chemotherapy and radiotherapy-induced apoptosis is executed via the intrinsic pathway. , activation; , inhibition (57).

Anti-apoptotic members Bcl-2 and Mcl-1 have been shown to be

upregulated in glioma and their expression to correlate with tumour grade (58).

Additionally, EGFRVIII can upregulate Bcl-XL in glioma cells conferring resistance

21

to the chemotherapeutic agent cisplatin (59). Also, EGFR and EGFRvIII can

interact with Puma, resulting in its cytoplasmic sequestration and inactivation

(60). In addition, Bcl-2 family members may contribute to gliomagenesis by

enhancing migration and invasion through alteration of the expression levels of

metalloproteinases and their inhibitors (61-63).

Due to their central role and importance in apoptosis, several attempts

to target them have been made. BH3 mimetics emerged as potent

antineoplastic therapeutics, either alone or in combination with other drugs,

however they exhibit low permeability by the BBB and are often associated with

significant cytotoxic effects to non-cancer cells (56).

1.3.3.3 Necrosis

Necrosis presents another hallmark of GBM tumours, with studies

supporting its presence in up to 88% of cases (64). Necrotic cell death is often

referred to as unscheduled cell death, implying that within a multicellular

organism it is an unregulated process, which is now known not to be true.

During necrosis the plasma membrane is disrupted leading to the spillage of

intracellular proteins which activates a damage response from the host immune

system. Necrosis occurs when blood supply is limited or when energy supplies

(ATP) are depleted (65). Recent research, shows that necrosis is a regulated

process and can occur in a variety of ways. In glioma, necrosis is thought to be

driven by BCL2L12, a member of the Bcl-2 family of proteins, which blocks

caspace-3 and caspase-7 downstream of mitochondrial membrane breakdown,

thus shifting the cell death balance from apoptosis to necrosis. BCL2L12 inhibits

caspase-7 processing through direct physical interaction, whilst caspase-3

22

inhibition occurs through the transcriptional upregulation of the small heat

shock protein alpha β-crystallin (66).

1.3.3.4 Angiogenesis

GBMs are highly vascular and are characterised by irregular and dilated

vessels presenting vascular leakage (67). In addition to the poor vascular

architecture, endothelial cells associated with the tumour vasculature fail to

form tight junctions and have few associated pericytes or astrocytic foot

processes leaving the integrity of the BBB compromised, resulting in increased

interstitial oedema. Interstitial oedema may further compromise regional blood

flow and intensify tumour hypoxia leading to areas of necrosis (68). Recent

evidence suggests angiogenesis may occur following a vascular collapse in

glioma. Hypoxia is a critical aspect of the glioma microenvironment and it has

been associated with poor prognosis, increased angiogenesis, tumour growth,

radio-and chemotherapy resistance, and tumour invasiveness (69, 70). Hypoxia

induces hypoxia induced factor (HIF)-1, a heterodimeric transcriptional factor

(71), which induces expression of pro-angiogenic and vascular permeability

factors such as: (vascular endothelial growth factor) VEGF, endothelial nitric

oxide synthase (eNOS), angiopoietin, ephrin and others such as: glycolytic

enzymes, glucose transporter-1, inducible nitric oxide synthase (72) and Twist

(73, 74). In addition to oxygen levels, EGFR or loss of tumour suppressors can

also induce HIF-1 expression (74, 75). In human glioma biopsies, it has been

shown that VEGF-A overexpression correlates directly to proliferation,

vascularisation, and degree of malignancy, and therefore corresponds inversely

to prognosis (76, 77).

23

Several types of angiogenesis inhibitors have been developed for

therapeutic use. Because it is the main driver of angiogenesis, most approaches

target VEGF signalling (78). Bevacizumab (Avastin®), is a humanised

monoclonal antibody that inhibits VEGFR by sequestering VEGF-A (its natural

ligand). To date, it has been the most successful drug in the clinic and received

accelerated food and drug administration (FDA) approval for use in patients

with recurrent GBM in 2009. Initial studies with bevacizumab in recurrent GBM

showed patient response rates of 28–40% and 6-month progression free

survival rate of 40–50% (79, 80). Despite the improved survival rates observed

with bevacizumab, all patients eventually relapsed, thus combination therapies

are currently investigated. Combination of bevacizumab with irinotecan, a

chemotherapy drug, increased response up to 50.2% and the overall

progression free survival rates to 8.9 months (79, 80). Currently, there are two

ongoing randomised phase III clinical trials exploring bevacizumab in

combination with radiation and temozolomide in newly diagnosed GBM but

results from these studies are not yet available.

1.3.3.5 Cell invasion

One of the biggest challenges in treating GBM patients is the highly

invasive nature of the tumour, which results in recurrent tumour development

adjacent to the resection margin or within centimetres of the resection cavity.

Invasion by glioma cells into regions of normal brain is driven by a multifactorial

process involving cell interactions with the ECM and with adjacent cells, as well

as accompanying biochemical processes supportive of proteolytic degradation

of ECM and active cell movement. The most frequent route of invasion of glial

24

tumour cells is along white matter tracts and basement membranes of blood

vessels as they lack the ability to penetrate the basement membrane of blood

vessels thus promoting local spread (81) (Figure 1.10).

Figure 1.10: Schematic of GBM cell invasion. GBM cells invade along white-matter tracts, around neurons and blood vessels, and in the subpial region. The immunohistostaining panel on the bottom right demonstrates individual elongated, hyperchromatic tumour nuclei oriented along myelinated axons. The panel at the top right illustrates molecular events relating to the invasion of single cells (82).

Glioma invasiveness has been linked to the regulation of members of the

family of metalloproteases (MMP) and their endogenous tissue inhibitors

(TIMPs). In GBM, Akt signalling has been shown to increase MMP2 and MMP9

activity thus promoting the tumour cell proteolytic capability to invade normal

brain (83). The tumour suppressor PTEN has also been reported to suppress

GBM invasion by suppressing proteolysis of the extracellular matrix by MMP9

and MMP2 and by inactivating two Rho-family GTP-binding proteins, Rac and

25

Cdc42 (84). Moreover, overexpression of Bcl-w, a pro-survival Bcl-2 protein,

has been reported to promote invasion of GBM by activating Akt signalling,

which leads to the phosphorylation of GSK3β and allows subsequent nuclear

translocation of β-catenin. Once in the nucleus, β-catenin upregulates target

genes, such as MMP2, TWIST1 and Snail to promote invasion (85).

Other proteins involved in GBM invasion include angiopoietin-2, which in

addition to its involvement in angiogenesis also plays a role in inducing tumour

cell infiltration by activating MMP2 (86). Ephrin receptors which mediate

neurodevelopmental processes such as axon guidance and cell migration in

glioma have been shown to regulate migration and invasion (21). Moreover,

several integrins such as αvβ3 and αvβ5 are found overexpressed in GBM and

have been shown to interact with osteopontin, a natural ligand in normal brain,

to mediate tumour cell migration in brain tissue (21).

The effects of several Akt-targeted drugs on GBM invasion have been

investigated pre-clinically and clinically. Sulindac (Merck), a non-steroidal anti-

inflammatory drug (NSAID), has been shown to inhibit GBM invasion in vitro by

dephosphorylating Akt at Ser473 (87). Arsenic trioxide (ATO), which has been

approved by the FDA for the treatment of promyelocytic leukaemia, has been

shown to accumulate more in brain tumours than in normal human brain

tissues. In phase I studies, ATO was well-tolerated with TMZ and radiotherapy

against malignant gliomas (88). BKM120 (Novartis) is a pan-class I PI3K

inhibitor which is able to cross the BBB and has been shown to effectively inhibit

the PI3K/Akt pathway in vitro and in vivo in glioma. Furthermore, treatment of

various GBM cell lines with BKM120 showed a dose-dependent inhibition of

26

growth while treatment of mice with intracranial U87MG xenografts increased

the median survival without any obvious adverse effects (89). Currently,

BKM120 is being studied in a phase II trial in patients with recurrent GBMs

(NCT01339052). There are also studies investigating the efficacy of this drug in

combination with current first-line GBM treatments. There is an ongoing phase

I study for newly diagnosed GBMs that combines BKM120 with radiotherapy

and TMZ (NCT01473901). BKM120 is also combined with bevacizumab in a

phase I/II study for relapsed/refractory GBMs (NCT01329660). In addition,

there are studies verifying the efficacy of BKM120 in combination with other

inhibitors, include: a phase Ib/II study of INC280, a c-Met inhibitor, and

BKM120 in recurrent GBM (NCT01870726) and a phase Ib/II study of BKM120

and one of the alkylating agents, carboplatin or lomustine, also in recurrent

GBM (NCT0193461).

1.3.4 The TAM family of RTKs

RTKs are high-affinity cell surface receptors that are activated by growth

factors, cytokines, and hormones to regulate a variety of biological processes

both in normal and pathogenic conditions (90). To date, 58 different human

RTKs have been identified and divided into twenty subfamilies (Figure 1.11). All

RTKS share a similar molecular architecture comprised of a ligand-binding

region at the N-terminus, a single transmembrane helix, and a cytoplasmic

region that contains the tyrosine kinase as well as C- terminal regulatory regions

(90). The TAMs are a relatively recently identified subfamily of RTKs. TAM

receptors differ from other RTK subfamilies because they harbour a conserved

homologous amino acid sequence within the kinase domain, KW(I/L)A(I/L)ES,

https://en.wikipedia.org/wiki/Dissociation_constant#Protein-ligand_binding

https://en.wikipedia.org/wiki/Cell_surface_receptor

https://en.wikipedia.org/wiki/Growth_factor

https://en.wikipedia.org/wiki/Growth_factor

https://en.wikipedia.org/wiki/Cytokine

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2914105/figure/F1/

27

not present in other RTKs as well as adhesion molecule-like domains in the

extracellular region (91, 92).

Figure 1.11: Schematic representation of the 20 known RTK subfamilies(90).

1.3.4.1 The structure of TAMs

The TAM family of RTKs is so named according to its members: Tyro3

(Sky), Axl (UFO) and MerTK. The TAMs possess an N-terminal extracellular

region that is formed by two immunoglobulin (Ig)-like domains and two

fibronectin type III repeats and an intracellular region with intrinsic tyrosine

kinase enzyme activity (93) (Figure 1.12). The TAM receptor genes share similar

genomic structures, encoding transcripts which range in size from 3 to 5 kb (94-

96). Tyro3 and Axl appear to have the most similar genomic structure (91, 97,

98). In contrast, Axl and MerTK have the most similar amino acid sequences for

the tyrosine kinase domain (99). Overall, the protein sequences of the human

28

TAM receptors share 31–36% amino acid sequence identity within the

extracellular region. The intracellular domains share 54–59% sequence identity

with greatest homology in the tyrosine kinase domain (99). The full length

human Tyro3, Axl, and MerTK proteins contain 890, 894, and 999 amino acids,

respectively. Although the predicted protein sizes are 97, 98, and 110 kDa for

Tyro3, Axl, and MerTK, respectively, the actual molecular weights range from

100 to 140 kDa for Axl and Tyro3 and 165–205 kDa for MerTK due to post-

translational modifications, including glycosylation, phosphorylation, and

ubiquitination (97, 100, 101).

Figure 1.12: The TAM RTK family structure. The TAMS share two extracellular FNIII and two (Ig)-like domains, followed by the intracellular tyrosine kinase. Also, in the diagram are depicted the tyrosine autophosphorylation sites by the asterisks and known SH2 domain-docking sites. (102).

29

1.3.4.2 The TAMs ligands

The natural ligands for the TAMs are two homologous vitamin K-

dependent proteins, Gas6 (for all three TAMs) and Protein S (Tyro3 and MerTK

only) (103, 104). Gas6 is a 678-amino acid protein comprised of 11 -

carboxyglutamic acid (Gla) residues at the N-terminal, a loop region and four

EGF-like repeats, and a sex hormone-binding globulin (SHBG)-like structure

(105) at the C-terminus, which is composed of two V-shaped globular laminin

G-like (LG) domains with calcium-binding sites (106) (Figure 1.13).

Figure 1.13: Gas6 structure , showing the Gla domain, EGFR-like repeats followed by the LG domains at the C terminus (107).

Protein S has the same characteristics as Gas6 but additionally contains

a thrombin-sensitive cleavage site at the loop region, which accounts for its role

as a negative regulator of blood coagulation (93, 108). The glutamate residues

in the Gla region are post-translationally modified by -glutamyl carboxylase in

a vitamin K-dependent manner. The Gla residues mediate calcium-dependent

binding to negatively charged membrane phospholipids on apoptotic cells (108),

such as phosphatidylserine (PtdSer) (103). The C-terminal LG domains mediate

binding of the protein to the TAM receptors. Specifically, the functional Gas6/Axl

signalling complex is a heterotetramer formed by 2 receptor molecules and 2

ligand molecules in a circular structure (109). The major binding site within the

ligand is fully contained within the LG1 domain which interacts with both the

Ig1 and Ig2 domains of the receptor, with no direct interaction in the dimer

30

between the 2 receptor molecules or the 2 ligand molecules (Figure 1.14).

Moreover, MerTK and Tyro3 have been shown to lack the Gas6 binding major

groove and bind Gas6/Protein S in a 1:1 stoichiometry (109).

Figure 1.14: Gas6-LG /Axl-IG complex. A. Top view of Gas6-LG (Cyan)/Axl IG (IG1 in yellow and IG2 in brown) complex towards the cell membrane. B. Side view of Gas6/Axl complex with the cell membrane at the bottom. C. Front view of Gas6/Axl interaction, in the direction indicated by the arrow in B (109).

In 2009, Tubby, Tulp-1 and Galectin-3 were identified as novel TAM

ligands, although their physiological significance is still unknown. Tulp-1 was

shown to interact with Axl, Tyro3 and MerTK in co-immunoprecipitation

experiments, whilst Tubby interacted with only MerTK. Co-immunoprecipitation

experiments also confirmed Galectin-3 association with MerTK, but the affinity

for either Axl or Tyro3 has not been reported. Tubby, Tulp-1 and Galectin-3

have been shown to promote MerTK mediated phagocytosis (110, 111).

1.3.4.3 The TAM expression pattern

In adult tissues, Tyro3, Axl, and MerTK exhibit widespread distribution

with overlapping but unique expression profiles. Tyro3 is most abundantly

31

expressed in the CNS and is also found in ovary, testis, breast, lung, kidney,

osteoclasts, and retina as well as a number of hematopoietic cell lines including

monocytes/macrophages and platelets. Axl is ubiquitously expressed with

noTable levels found in monocytes/macrophages, platelets, endothelial cells,

heart, skeletal muscle, liver, kidney, and testis (100). MerTK is expressed in

monocytes/macrophages, dendritic cells, NK cells, NKT cells, megakaryocytes,

and platelets. High levels of MerTK expression are also detected in ovary,

prostate, testis, lung, retina, and kidney, whereas lower levels are found in

heart, brain, and skeletal muscle (112).

1.3.4.4 The TAM function

Tyro3 is the least studied of the TAM receptors; nonetheless, a handful

of studies have provided clues as to its function in the CNS (discussed in more

detail in Section 1.3.4.6) as well as mediation of platelet aggregation (113),

osteoclastic bone resorption (114, 115) and normal reproductive system

development (97, 116, 117). Tyro3 was also shown to negatively regulate type

2 immunity through its expression on dendritic cells (118). Even though Axl

expression has been characterised, the receptor function has been mostly

studied in malignancy and its contribution in normal conditions is still not known

for most cases. Axl signalling is involved in many processes ranging from the

differentiation of cells in the erythroid lineage, protecting vascular vessels from

injury, clearance of apoptotic cells, angiogenesis, haematopoiesis platelet

aggregation, cell survival, proliferation, regulation of pro-inflammatory cytokine

production and regulation of the actin cytoskeleton (119, 120). MerTK functions

have been mostly limited in the regulation of the immune system as it has been

32

shown to mediate phagocytosis of apoptotic cells and cytokine production, as

well as to be required for prevention of systemic autoimmune disease (112).

1.3.4.5 TAM signalling

The TAMs have been shown to mediate their effects through the

activation of a variety of signalling pathways. Axl activation has been shown to

activate the MAPK pathway, the PI3K pathway, as well as phospholipase C-γ

(PLC-γ). Further Axl binding partners include SOCS-1, Nck2, RanBPM, and C1-

TEN through their Scr homology 2 (SH2) domains (93, 120) (Summarised in

Figure 1.15). MerTK activation has also been linked to PI3K and PLC-γ

signalling, as well as to the STAT transcription factor pathway and vav1. Tyro3

has been shown to activate the PI3K and MAPK signalling (93, 120).

Figure 1.15: Summary of Axl signalling pathways leading to platelet aggregation, cell survival, proliferation, regulation of pro-inflammatory cytokine production, and regulation of the actin cytoskeleton (120).

33

1.3.4.6 TAMs and the central nervous system

In the nervous system, the TAM receptor and their ligands have been

shown to be expressed by both neurons and glial cells. Tyro3 particularly is

highly expressed in human brain tissue whilst the expression of both Axl and

MerTK within the CNS is more restricted with overall expression being lower

(121). Tyro3 expression was confirmed in the olfactory bulbs, cerebral cortex,

piriform cortex, all cortical layers of the cerebellum, the amygdala, hippocampus

and the cerebellum (122). Axl and MerTK are expressed at low levels in the

Purkinje cell layer of the cerebellum, endothelial cells , oligodendrocytes,

Schwann cells, and astrocytes (123), with the most important site of Axl

expression in the brain being the vasculature (124). Gas6 has been shown to

be expressed in the hippocampus, midbrain, the cerebellum, as well as the

ventral spinal cord (125) . In contrast, protein S is expressed only at low levels

in the CNS, in the locus coeruleus, choroid plexus, as well as in astrocytes and

in retinal pigment epithelial cells (126).

Even though TAM expression is well characterised in the brain little is

known about their function (summary of known function presented in Figure

1.16). Axl has been shown to promote survival and growth of GnRH neuronal

through the induction of the MAPK/ERK1/2 pathway as well as PI3K/Akt and

PI3K/Rac/Rho pathway to regulate migration (127, 128). Tyro3 was detected

over the surface of dendrites and axons and has been shown to modulate

synaptic plasticity through the MAPK and PI3K signalling pathways in the

dendritic compartment of hippocampal and cortical neurons (129). Moreover,

Tyro3 when activated by Protein S protects against excitotoxic injury by

activating Akt which in turn increases Bcl-2 and Bcl-XL levels (130). Protein S-

34

Tyro3 signalling has also been shown to be involved in maintaining the integrity

of the BBB although the exact mechanisms are still unknown (124). Gas6

activation of Tyro3 also protects oligodendrocytes from TNF-mediated toxicity

and promotes myelination by oligodendrocytes (131, 132).

The TAM receptors have also been shown to regulate the brain immune

system. All three TAMs are expressed by microglia and negatively regulate

dendritic cells and macrophages in the brain. Studies in triple knockout mice

revealed that without the TAM receptors, activated microglia produce increased

amounts of pro-inflammatory cytokines, leading to increased brain

inflammation (133, 134).

Amongst their other various functions in the brain, the TAMs have also

been shown to be important for normal brain development and most particularly

in the maintenance of the stem cell pool. Wang and colleagues showed that,

following genetic tagging and isolation of NSCs, the TAMs are expressed in

accordance with stem cell markers and introduction of a dominant negative

form of Axl and a double Axl/MerTK knockout promoted neural stem cell

differentiation (135). The importance of the TAMs and their ligands in the stem

cell pool maintenance is also supported by studies showing that loss of Gas6

reduces the stem-like cell populations, while Protein S inhibits subventricular

cell proliferation. In the same study, they also report that NPCs maintain a high

level of reactive oxygen species (ROS), which has been shown to promote

neural stem cell renewal; moreover, ROS is also produced during the γ-

carboxylation of Gas6 and Protein S (136).

35

Figure 1.16: TAM signalling in cells of the CNS. The schematic represents a generalization of the important TAM signalling pathways. A. Protein S mediates survival of cortical and hippocampal neurons through Tyro3. B. Gas6 stimulates both migration and survival through Axl in GnRH neuronal cells. C. Gas6-Tyro3 signalling regulates neuronal repair and synaptic plasticity. D. Axl mediates survival signalling via Gas6. E. Microglia express Axl and MerTK. Gas6 has been shown to bind both Axl and MerTK, limiting inflammatory responses and promoting phagocytosis in microglia, while Tubby reportedly binds MerTK and promotes phagocytosis in macrophages and retinal epithelial pigment cells (128).

36

1.3.4.7 TAMs and GBM

Additional to their role in normal CNS function, Hutterer and colleagues

have demonstrated by mRNA expression studies that Axl and Gas6 are

overexpressed in high grade gliomas. High Axl/Gas6 expression in patients with

GBM correlated with poor overall survival, indicating that Axl may significantly

contribute to glioma progression. Immunohistochemical studies also revealed

that Axl/Gas6 are not only expressed in tumour cells but also in the tumour

vasculature (137). Given the significance of Axl expression observed in many

other tumours, the role of the TAMs in brain cancer has become a topic of great

interest.

Transfection of glioma cells with a truncated Axl protein, which served

as a dominant negative (DN) inhibitor of the kinase, resulted in decreased

growth. This growth disadvantage was independent of Gas6 stimulation.

Introduction of Axl-DN cells into nude mice correlated with in vitro experiments,

showing reduced tumour growth compared to control cells without affecting the

tumour vasculature but instead by decreasing invasive potential. Detailed

assessment of the effect of Axl on cell migration revealed that Axl-DN cells have

a reduced locomotor activity and disrupted cell-cell adhesion. Magnetic

resonance imaging (MRI) scans of injected mice also revealed that Axl-DN mice

had smaller tumours compared to Axl-WT mice (138).

Further studies by Wang and colleagues show that MerTK is also

overexpressed in high grade glioblastomas. Additionally, knockdown of MerTK

resulted in morphological changes with GBM cells appearing elongated and

forming more compact spheres compared to the rounded and amoeboid shaped

control cells. Treatment of shMerTK and control cells with etoposide ,

37

topoisomerase inhibitor chemotherapeutic drug, also revealed that MerTK

knockdown sensitised cells to apoptosis, while control cells showed elevated

MerTK activation upon DNA damage and increase resistance to etoposide (139).

In agreement with the previous study, knockdown of MerTK or Axl was also

demonstrated to increase apoptosis and chemosensitivity in astrocytoma upon

serum starvation (140).

B cell-specific Moloney murine leukaemia virus integration site 1 (BMI1)

is a master regulator of NPCs stemness and self-renewal through the negative

regulation of the INK4A locus, which encodes p16, p15 and p14, and p21 genes.

BMI1 is often found overexpressed in brain cancers. BMI1 was shown to

regulate self-renewal in CD133+ populations and proliferation and cell fate

determination in CD133– populations (141). Recently, Gas6 was shown to be a

novel target gene for BMI1 although the exact mechanism remains elusive

(142). BMI1 belongs to the multiprotein complex PRC1. The PRC1 complex in

cooperation with the initiation complex PRC2-enhancer of zeste homologue 2

(EZH2) mediate methylation of histone proteins, primarily at lysine 27 of

histone H3 (H3K27) to induce gene silencing (143). Interestingly, Axl has been

shown to be directly regulated by EZH2 in glioma. EZH2 was shown to be up-

regulated and co-expressed with Axl in glioblastoma cells. Further studies

revealed EZH2 positively regulates Axl expression independently of methylation

(144). Furthermore, knockdown of EZH2 in glioblastoma significantly reduced

the invasive character of the cells and this was shown to be mediated by Axl.

The authors also point out that EZH2 is required during EMT for E-cadherin

repression by Snail. Axl has been extensively associated with Snail mediated

EMT in breast cancer and this could represent another potential link between

38

EZH2 and Axl (145). Given the importance of the polycomb proteins in NSC and

GSC maintenance and regulation, it could hint at Axl/Gas6 as a novel regulatory

signalling pathway in this process.

Furthermore, Axl inhibition by the small molecule BMS-777607 was

shown to result in increased apoptosis, decreased proliferation and migration

in vitro and ex vivo in SF126 and U118 GBM cells. In vivo, in xenografts tumour

volume was reduced by 56%. Moreover Axl blockade had anti-angiogenic

effects and expression of Axl was localised in hyper cellular tumour regions, the

migratory front of tumour cells in pseudo-palisades, and in vascular proliferates

within the tumour (146), indicating that Axl may be involved in the regulation

of many GBM hallmark features.

TAM signalling has only been started to be investigated in glioma and

the full repercussions of expression and downstream signalling need to be

investigated in more detail. Therefore, lessons from how the TAMs are involved

in glioma progression can undoubtedly inform on their involvement in other

cancers.

1.3.4.8 TAMs in other cancers

The TAMs were linked to cancer at the outset of their discovery. Axl was

originally isolated from chronic myelogenous leukaemia (CML) cells where it

was also found to possess transforming abilities (94) and since then has been

linked to several other leukaemia subtypes such as B-cell chronic lymphocytic

leukaemia, where its overexpression promotes survival through its association

with other kinases (147). Interestingly, Axl overexpression has been also shown

to increase Imatinib (tyrosine kinase inhibitor) and doxorubicin

39

(chemotherapeutic drug) chemoresistance in CML (148) and acute myeloid

leukaemia (149) respectively. Increased resistance to Imatinib due to Axl and

MerTK overexpression has also been reported in gastrointestinal stromal

tumours (150) and non-small lung carcinoma (151). Additionally, studies using

knockdown of Axl by short hairpin (sh)RNA revealed that Axl promotes the

survival of cutaneous squamous carcinoma cells through Akt activation and

inhibition of the pro-apoptotic factors of the Bcl-2 family of proteins. In the

same study, it was also demonstrated that Axl knockdown results in increased

cytochrome c release upon UV treatment (152). Recently, Tyro3 has been

identified to regulate expression of Microphthalmia-associated transcription

factor (MITF), a master regulator of melanocyte development, in melanoma

and to promote tumorigenesis in spite of B-Raf-induced senescence (153).

Tyro3 was also shown to promote cell proliferation and chemoresistance in

breast cancer (154).

Along with increasing chemo resistance and promoting cell survival the

TAMs have been also shown to promote tumour growth and invasion/migration.

Global gene expression profiling revealed Axl and Tyro3 to be up regulated in

thyroid cancer and their expression to be essential for cell proliferation and

invasion. Inhibition of either of these kinases induced apoptosis and inhibition

of Axl signalling decreased tumour cell invasion and migration. Silencing of the

Axl receptor additionally impaired tumour vasculature, indicating a possible

involvement in tumour angiogenesis (155). Studies by Gjerdrum and colleagues

showed that Axl overexpression in metastatic breast cancer is induced by

epithelial-to-mesenchymal (EMT) mediators Twist, Snail and Zeb2, in a process

in which cells lose their adhesive properties and increase their migration

40

potential. In the same study, they show that Axl can positively impact on the

expression of these mediators and is possibly involved in a positive feedback

loop (156). In a similar study, Vuoriluoto and colleagues show that Slug and H-

Ras induced-EMT in basal-like breast cancer cells is dependent on the up-

regulation of vimentin, a type III intermediate filament protein, which in turn

mediates invasion and migration by up-regulating Axl. Axl signalling was shown

to be able to drive migration and extravasation of breast cancer cells from the

blood stream even in the absence of vimentin, indicating that Axl is a highly

significant regulator of metastasis (157).

Axl has also been demonstrated to regulate migration/invasion through

induction of MMP9 by the activation of the ERK/NF-κB pathway and by

mediating chromatin remodelling by Brg-1. Additionally, Axl activation by Gas6

stimulated the expression of the pro-survival Bcl-2 family of proteins Bcl-2 and

Bcl-XL and activated NF-κB thus suppressing apoptosis. MMP9 expression by Axl

signalling was found to be independent from Gas6 stimulation, implicating that

possibly Axl activation is mediated by distinct pathways as well (158). The high

importance of Axl signalling in metastasis was elegantly demonstrated also in

prostate cancer. Axl was found to be overexpressed in castrate resistant

prostate cancer and to significantly increase migration, invasion and

proliferation by activation of the PI3K/ NF-κB pathway. Moreover, Axl was

shown to promote the secretion of IL-6 and paracrine STAT3 activation

increasing prostate cancer proliferation (159).

41

1.3.4.9 TAMs as key targets in cancer

Axl has been at the forefront of scientific research due to its involvement

in tumour chemoresistance. As with many RTKs several small molecule drugs

were developed in order to block activity.

The most noteworthy Axl inhibitor to date is BGB324 (R428; Rigel

Pharmaceuticals, BergenBio). BGB324 has the highest specificity and activity

towards Axl compared to other inhibitors available (IC50=14 nmol/L). BGB324

is clinical phase I for aggressive and metastatic cancers (160).

Other drugs inhibiting Axl though non exclusively are available in the

market and include Bosutinib (SKI606, PF5208763, Bosulif; Pfizer) which mainly

targets Bcr-Abl1 and Src (IC50=1 and 3.5 nmol/L, respectively) but can also

block Axl with an IC50 of 174 nmol/L. Bosutinib was authorized by the FDA and

the European Medicines Agency (EMA) in 2012 for the treatment of patients

with Philadelphia chromosome-positive CML. Bosutinib is currently in a phase II

clinical trial for glioblastomas and breast cancer (161). Cabozantinib (XL184,

Cometriq; Exelixis) targets a variety of kinases including VEGFR-2 (IC50=0.035

nmol/L), Met (IC50=1.3 nmol/L), Ret (IC50=5.2 nmol/L) as primary targets, as

well as Kit (IC50=14.3 nmol/L), Flt-3 (IC50=11.3 nmol/L), and Axl (IC50=7

nmol/L) and was authorized by the FDA and the EMA in 2012 for the treatment

of medullary thyroid cancer and is currently in a phase III clinical trial for

hepatocellular, renal, and prostate cancers (162). Sunitinib (SU11248, Sutent;

Pfizer) was authorized by the FDA in 2006 for treatment of renal cell carcinomas

and imatinib-resistant gastrointestinal stromal tumours, and in 2011 for

pancreatic neuroendocrine tumours. It is an inhibitor targeting Flt-3 (IC50=0.5

nmol/L), VEGFR-2 (IC50=20 nmol/L), Kit (IC50=22nmol/L), as well as Axl

42

(IC50=5 nmol/L) and currently also in clinical trials for NSCLC (163). Further Axl

inhibitors in Phase II clinical trials include Foretinib (XL880, GSK1363089;

Exelixis, GlaxoSmithKline) which primarily targets Met, VEGFR-2, Ron, Tie- 2,

and Kit (IC50=0.4, 0.86, 3, 1.1, and 3.7 nmol/L, respectively) as well as Axl

(IC50=11 nmol/L) and is a type II inhibitor presently in a phase II clinical trial

for NSCLC and in phase I/II for breast cancer (164). MGCD265 (Mirati

Therapeutics) is a potent inhibitor of Met (IC50=1 nmol/L), Ron (IC50=2 nmol/L),

VEGFR 1/ 2/3 (IC50=3, 3, and 4 nmol/L, respectively), Tie-2 (IC50=7 nmol/L),

and Axl (IC50=10 nmol/L). It is currently in phase I/II clinical trials for NSCLC

and in phase I for advanced malignancies (163). Early stages of phase I clinical

trials inhibitors include BMS777607 (ASLAN002; Aslan Pharmaceuticals and

Bristol-Myers Squibb), a type II Met inhibitor (IC50=3.9 nmol/L) that also acts

on Ron (IC50=1.8 nmol/L), Flt-3 (IC50=16 nmol/L), Tyro3 (IC50=4.3 nmol/L),

MerTK (IC50=14 nmol/L), and Axl (IC50=1.1 nmol/L). BMS777607 is currently in

a phase I clinical trial for advanced or metastatic solid tumours (146).

LY2801653 (Eli Lilly Company) strongly inhibits Ron (IC50=0.8 nmol/L), Met

(IC50=4.7 nmol/L), Axl (IC50=11 nmol/L), and Flt-3 (IC50=31 nmol/L) (59) and

is currently in a phase I clinical trial for advanced cancers (165). Lastly, S49076

(Servier) targets Met (IC50=2 nmol/L), FGFR-1 (IC50=68 nmol/L), FGFR-2

(IC50=95 nmol/L), FGFR-3 (IC50=200 nmol/L), and Axl (IC50=56 nmol/L) is

currently in a phase I clinical trial for advanced solid tumours (166).

The issue with all these inhibitors though is that they are not highly

specific and can inhibit a variety of other RTKs with higher affinity. Given the

emerging role of Axl in malignancy novel specific inhibitors or neutralising

antibodies should be developed and tested in the clinic.

43

1.3.5 Project significance

Gliomas are highly heterogeneous, invasive tumours that do not respond

to traditional therapies. One of the biggest challenges in treating glioma is

targeting all tumour hallmarks effectively so that cancer cells undergo a shock

and eventual cell death. Axl was recently shown to be overexpressed in glioma

and studies up to now have together demonstrated its importance in regulating

major GBM hallmarks, although the mechanisms and the full extent of its

overexpression still remain unknown. The purpose of this PhD was to

investigate the expression of TAMs in GBM, to map out key TAM signalling

pathways that contribute to GBM tumorigenesis, and to assess the anti-cancer

effects of novel TAM inhibitors in GBM cells. The results of this study would

provide a better understanding of how we can utilise Axl blockade as a novel

improved approach in GBM patient treatment.

44

Chapter 2

Materials and Methods

45

2.1 Cell culture

2.1.1 Human cell lines

The human GBM cell lines used in this study were UP007 (established in-

house from GBM biopsy resected at King’s College Hospital, London under ethics

permission LREC00-173/11/SC/0048) and SNB-19 (DSMZ German Brain

Tumour Bank). Both cell lines were authenticated in-house as described

previously (167) and were mycoplasma tested. Cell lines used in screens or

confirmation of principles are summarized in Table 1.

Table 1: Cell lines used throughout the duration of project

Cell line Type Source

HA-c Human astrocytes ATCC

hcMEK/D3 Human cerebral microvascular endothelial cells ATCC

Neurons Human neuroblastoma cells ATCC

SNB-19 Human GBM cells DSMZ

UP007 Human GBM cells In house

MDA-MB-231 Human breast cancer cells ATCC

SCC-25 Human head and neck cancer cells ATCC

Detroit Human head and neck cancer cells ATCC

ATCC, American type culture collection; DSMZ, Deutsche Sammlung von Mikroorganismen und Zellkulturen.

2.1.2 Maintenance of cell lines

All cell culture was carried out in a Class II biological safety cabinet

(Herasafe, Heraeus, Thermo Scientific). All consumables used for cell culture,

such as glass Pasteur pipettes, cell culture flasks and pipette tips, were either

sterile as purchased (plasticware) or through autoclaving (glassware). Prior to

46

use, they were sterile and sprayed with 70% ethanol and Mycozap (Lonza,

Slough, UK) before being placed in the hood. All solid waste was autoclaved,

while liquid waste was aspirated into a flask containing a working concentration

of Virkon (Lonza) disinfectant solution.

Cells were normally cultured in “complete” medium, comprising

Dulbecco’s Modified Eagle Medium (Fisher Scientific, Loughborough, UK)

supplemented with 10% foetal bovine serum (FBS; Lonza), 2 mM L-glutamine

(Life technologies, Paisley, UK) and 1% penicillin/streptomycin (Fisher

Scientific). Cells were grown in a humidified incubator with 5% CO2 at 37◦C, and

were typically passaged into a desired vessel once they reached 80%

confluency. Cells were washed with 10 ml warm phosphate buffered saline

(PBS) and incubated with trypsin/EDTA solution (Lonza) for 5 minutes at 37oC.

The detached cells were re-suspended in 10 ml complete medium to inactivate

the trypsin, then centrifuged at 1150 rpm for 5 minutes and the pellet re-

suspended in complete medium and plated at 1:10 dilution.

2.1.3 Recovery of cryopreserved cells

Vials of frozen cells were rapidly defrosted in 370C water bath to reduce

dimethyl sulfoxide (DMSO) contact. 10 ml media was applied in a drop wise

manner to the cells to avoid shock and to dilute the DMSO. Cells were

centrifuged 1150 rpm for 5 minutes, the supernatant discarded and pellets re-

suspended in 10 ml media. The cells were then placed in a 75 cm2 flask (“T75”).

The medium was replaced after 24 hours to remove any remaining DMSO

before experimental use.

47

2.1.4 Cell counting

Cells were suspended in 10 ml media and counted using a

haemocytometer, Glasstic® slide (Hycor Biomedical Ltd, Edinburgh). Ten µl of

cell suspension was loaded onto the counting chamber, which was mounted

onto a light microscope to visualise the cells. Two squares were counted as

shown in the diagram bellow (Figure 2.1). The cell concentration was

determined by dividing the counted cells by the number of squares and

multiplying by 104 to derive the number of cells per ml.

Figure 2.1: Schematic representation of haemocytometer square indicating counting margins

2.1.5 Cell treatments

Cells were treated either with ligands or SMI at concentrations and time

points as indicated throughout the text. All drugs were diluted in DMSO at stock

concentrations of 100 µM, aliquoted and stored at -80oC until usage. Ligands

48

were diluted in dH2O or PBS (Life technologies) according to the manufacturers

protocol, aliquoted and stored at -80oC until usage.

2.2 Western Blotting

2.2.1 Cell lysis and protein extraction

For cell lysis, the media was aspirated and the cells were washed twice

in ice-cold PBS. For Western (immuno-) blotting, cells were lysed in ice-cold

RIPA lysis buffer (150 mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholate,

0.1% SDS, 50 mM Tris, pH 8.0) supplemented with the respective

serine/threonine and tyrosine phosphatase inhibitors calyculin A (Sigma,

Dorset, UK) and sodium orthovanadate (Fisher scientific) as well as a protease

inhibitor cocktail (Fisher scientific). Cells were further shaken in lysis buffer at

4oC on a shaker for 30 minutes before collection. Lysates were clarified by

centrifugation at 14 000 x g for 10 minutes at 4oC. The soluble supernatant was

transferred into new Eppendorf tubes and stored at -200C until required. For

immunoprecipitations, cells were lysed in Pierce™ IP Lysis Buffer (Thermo

Scientific) supplemented with the same phosphatase and protease inhibitors

(Fisher scientific). Lysates were used immediately for immunoprecipitation

experiments.

2.2.2 Protein quantification

Usually, cells that had been plated in equal numbers in multi-well plates

were expected to produce similar amounts of total protein extract for

downstream assays. However, where protein concentration in samples was

quantified, the BCA (bicinchoninic acid) protein assay was employed. Bovine

serum albumin (BSA; 1 mg/ml stock; Sigma) was diluted to protein standard

49

concentrations of 10, 20, 30, 40 and 50 µg/ml, and 25 µl pipetted into a 96-

well plate in duplicate wells. For each sample, the whole cell lysate was diluted

1:10 in distilled water and 10 µl of diluted sample was pipetted into the same

96 well plate. 200 µl of 1x BCA working reagent (1:50 reagent B: reagent A)

was added to standards and samples. The plate was incubated at 37oC for 30

minutes before absorbance at 592 nM was measured on a spectrophotometric

microplate reader (Synergy; BioTek, Potton, UK). The absorbance readings of

the BSA standards were utilised to generate a standard curve to determine the

protein concentrations of each sample.

2.2.3 Sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS–PAGE)

The apparatus glass plates were cleaned with distilled water and 70%

ethanol prior to mixing the gel. The gel was mixed and poured into the gel

apparatus, a comb was inserted and the gel was left to polymerise. Once set,

the comb was removed and the wells washed with distilled water and running

buffer (25 mM Tris base, 190 mM glycine and 0.1% SDS, pH 8.3). The cell

lysate was mixed with 4x loading (“Laemmli”) buffer (0.313 M Tris-HCl (pH 6.8),

10% SDS, 0.05% bromophenol blue, 50% glycerol; all purchased from Fisher

scientific) supplemented with 5 mM DTT (Fisher scientific). Samples were boiled

for 5 minutes at 95oC. Samples were loaded onto a 10% gel (10% Protogel

(National Diagnostics), 0.1 M Tris pH 8.8, 3% Sucrose, 0.05% sodium dodecyl

sulphate (SDS), 0.05% N,N,N,N – tetramethylethyl-enediamine (TEMED; Fisher

scientific), 0.05% ammonium persulphate (APS)], and stacking gel (3%

acrylamide for electrophoresis (Protogel; National diagnostics, Hessle, UK), 0.05

M Tris, 3% Sucrose. 0.05% SDS, 0.025% TEMED, 0.05% APS). PageRuler Plus

50

Prestained Protein Ladder, a mixture of nine blue-, orange- and green-stained

proteins (10 to 250 kDa), was used as size standards.

2.2.4 Protein transfer

The separated proteins were transferred onto an activated

polyvinylidene fluoride membrane (PVDF; Millipore, Nottingham, UK) using a

wet transfer method (transfer buffer: 25 mM Tris base, 200 mM Glycine, 0.04%

SDS and 0.005% Tween-20). Transfer occurred by applying 350 mA for 2 hours

at 4oC. Non–specific protein binding was blocked by incubating PVDF blots for

1 hour in 5% non–fat milk/TBST when blotting for total protein and in 5%

BSA/TBST when blotting for phospho–proteins. Membranes were incubated

with primary antibodies (Table 2) overnight at 40C in 5% milk or 5% BSA/TBST.

The membranes were washed 3 times for 10 minutes in TBST, then incubated

with the corresponding secondary antibody (Table 3) for 1 hour at room

temperature, and again rinsed three times in TBST. The presence of protein–

bound antibody was detected with enhanced chemiluminescence development

reagent (Luminata Forte; Millipore) for 3 minutes and visualised with a high

sensitivity CCD camera imaging platform (Chemidoc MP; Bio-Rad, Hemel

Hempstead, UK). The software ImageJ (168) was used for densitometric

quantification of Western blot band intensities.

51

Table 2: List of primary antibodies used for immunoblotting

Antibody Species Dilution Source Axl (C-20) Goat polyclonal 1:1000 Santa Cruz

Tyro3 (C-20) Goat polyclonal 1:1000 Santa Cruz MerTK (B-10) Mouse monoclonal 1:1000 Santa Cruz EGFR (D38B1) Rabbit polyclonal 1:1000 Cell Signalling

β-Actin Rabbit polyclonal 1:10000 Sigma Akt 1/2/3 (S473) Rabbit polyclonal 1:1000 Santa Cruz

ERK1/2 Mouse monoclonal 1:1000 Santa Cruz pAxl Rabbit polyclonal 1:500 R&D systems

pSky/Mer Rabbit polyclonal 1:1000 Cell Signalling pEGFR Goat polyclonal 1:1000 Santa Cruz

Akt1/2/3 Rabbit polyclonal 1:1000 Santa Cruz pERK Mouse monoclonal 1:1000 Santa Cruz

Table 3: List of secondary antibodies used for immunoblotting

Antibody Dilution Source

Anti-rabbit HRP 1:2000 Dako

Anti-rabbit AP 1:5000 Cell Signalling

Anti-Goat HRP 1:5000 Promega

Anti-Goat AP 1:2000 Santa Cruz

Anti-mouse HRP 1:2000 Dako

2.3 Immunohistochemistry

Brain tumour tissue microarray with normal tissue as control, containing

35 cases of glioblastoma, 2 adjacent brain tissue and 3 normal brain tissue with

core diameter 1.5 mm and 5 µm thickness (GL805a; US Biomax, Rockville, MD,

Figure 2.2) and brain tumour tissue array with brain tissue as control, containing

12 cases with core diameter 1.5 mm and 5 µm thickness (T175a; US Biomax,

Rockville, MD, Figure 2.3) were deparaffinised using histology grade xylene

(Sigma-Aldrich, Dorset, UK) for 5 minutes and rehydrated using decreasing

52

concentrations (100%, 70%, 50%, 30%) of molecular grade ethanol (Fisher

Scientific) for 1 minute and 30 seconds respectively. Slides were then rinsed for

2 minutes in water. Antigen unmasking was performed in 0.01 M sodium citrate

(Fisher Scientific), pH 6 at 95ºC with two changes for 5 minutes each. The slides

were washed twice for 2 minutes in deionised water. Endogenous peroxidase

activity was blocked by incubating with 0.1% hydrogen peroxide (Sigma-

Aldrich) at room temperature. Following three 5 minute washes with PBS, the

slides were blocked with 1.5% rabbit serum in PBS for 1 hour. The slides were

then incubated with anti-Axl antibody at 1:100 dilution (goat polyclonal; R&D

Systems) in 1.5% rabbit serum/PBS for 2 hours. The slides were then washed

with PBS 3 times for 5 minutes and incubated with biotinylated anti-goat IgG

secondary antibody (Sigma) in 1.5% rabbit serum/PBS for 1 hour at room

temperature. The slides were then incubated for 30 minutes with avidin-biotin

enzyme reagent VECTASTAIN ABC Kit (Vector laboratories, Peterborough, UK)

and washed with PBS three times for 5 minutes. The slides were then developed

using ImmPACT DAB Peroxidase (HRP) Substrate (Vector laboratories) for 10

seconds and washed with deionising water 3 times for 5 minutes. The tissues

were counter-stained with haematoxylin and dehydrated through increasing

concentrations of alcohols (50%, 70%, 95%, 100%) for 30 seconds and 1

minute respectively, followed by xylene for 1 minute, after which they were

mounted with DPX mountant (Sigma-Aldrich). Images were captured by a Zeiss

Axiophot brightfield microscope at 20x magnification.

53

A

B

C

D

E

F

G

H

1

2

3

4

5

6

7

8

9

10

Figure 2.2: GL805a microarray panel display. Cer Cerebrum Malignant tumor,

NAT, Normal tissue.

1

2

3

4

5

6

7

8

Figure 2.3: T175a microarray panel display. Bra Brain Benign tumor,

Malignant tumor, Normal tissue

2.4 Molecular biology techniques

2.4.1 Transient knockdown with short interfering RNA (siRNA)

Cells were transfected with a pool of four siRNA nucleotides targeting

Axl, Tyro3 or TIMP1 (Santa Cruz, CA, USA), depending on the particular

investigation, at concentrations indicated. Control cells were transfected with a

pool of non-targeting siRNA at the same concentration. Transfection was

achieved using INTERFERin reagent (Polyplus, Reading, UK). INTERFERin and

siRNA complexes were prepared in serum free DMEM (Life Technologies),

vortexed and incubated at room temperature for 15 minutes before the total

54

mixture was added to cells in culture medium. Cells were then incubated with

the mixture for 48 or 72 hours before cell lysis or mRNA extraction for either

confirmation of knockdown or cell signalling investigation by either Western blot

or qRT-PCR.

2.4.2 Stable knockdown with short hairpin RNA (shRNA)

SNB-19 and UP007 cells were transfected with 1 µg Axl shRNA plasmids

(HuSH-29mer, OriGene) summarised in Table 4 over a period of 3 days, after

which cells were put under 1.5 µg/ml for SNB-19 and 1 µg/ml for UP007

puromycin selection for two weeks before clone selection. Clonal expansion was

performed under continuous puromycin selection throughout the project.

Table 4: Axl targeting sequences

Construct Construct sequence Origene catalogue number

Empty N/A TR20003

Scramble N/A TR30012

shAxl #1 GAACAGGATGACTGGATAGTGGTCAGCCA TR320269A / TI378296

shAxl #2 GCCGCTGCCTGTGTCCTCATCTTGGCTCT TR320269B / TI378297

shAxl #3 TGTGATGTTCTAAGGCTCTGAGAGTCTAG TR320269C / TI378298

shAxl #4 GGCACTGTAGTTCTAAGACTCAAATGTTC TR320269D / TI378299

55

2.4.3 Purification of RNA

Cellular total RNA was isolated using GeneJET RNA purification kit (Life

Technologies) according to the manufacturer’s protocol; RNA was finally eluted

with 100 µl RNA–free sterile water. The purity and concentration of isolated

total RNA was assessed using a spectrophotometer (ND-1000; NanoDrop

Technologies, Wilmington, DE, USA) to measure the UV absorption of samples.

The absorbance at 260 nm (A260) was used to determine the concentration of

RNA in undiluted samples. The ratio of absorbance at 260 nm and 280 nm

(A260/A280 ratio) was used to determine the samples’ purity, where a ratio of

>1.8 indicated high RNA purity. As RNA optimally absorbs light at a wavelength

of 260 nm due to its high content of nucleotides, while proteins and other

impurities that might be contaminating the isolated RNA absorb best at 280 nm,

the A260/A280 ratio is the preferred way to assess if an RNA sample is pure.

2.4.4 Synthesis of complementary DNA (cDNA)

cDNA was synthesised using a reverse transcription kit (nanoScript2RT;

PrimerDesign, Southampton, UK). Initially a mastermix consisting of 2 µg RNA,

sterile water and Oligo dT was heated at 65oC for 5 minutes. Following

immediate transfer on ice, the mixture was supplemented with nanoScript2 4X

Buffer, 10 mM dNTP mix, water and nanoScript2 enzyme. The total 20 µl

reaction was then briefly vortexed followed by a pulse spin and incubated at

room temperature for 5 minutes and then at 42oC for 20 minutes. The reaction

was heat-inactivated by incubation at 75oC for 10 minutes. cDNA was used for

amplification or stored at -20OC.

56

2.4.5 Real time quantitative real-time polymerase chain reaction-(qRT-PCR)

qRT-PCR amplification was performed in 96-well plates in a mastermix

for probes (Roche, Burgess Hill, UK) and run on a LightCycler® 96 System

(Roche). The genes investigated using pre-designed primers/probes were

purchased from Integrated DNA Technologies (Leuven, Belgium). The

primer/probe genes were labelled with FAM and VIC fluorophores which

enabled a duplex reaction for simultaneous determination of expression of both

gene of interest and housekeeping gene respectively. The amplification

procedure entailed 45 cycles of 95oC for 10 seconds (melting) followed by 60oC

for 30 seconds (annealing and extension). For the human EMT gene expression

screen, the qRT-PCR amplifications were performed using the Human CSC and

EMT Biomarker RT ArrayTM Kit (NanoCinna Technologists, Tehran, Iran) and a

SYBR Green system. For each reaction β-actin and GAPDH were utilised as

endogenous control genes. The amplification procedure entailed 95°C for 15

seconds (1 cycle), followed by 60°C for 1 minute (45 cycles). Relative

expression analysis was performed using the equation N=N0 x 2Cp

(LightCycler®96 software; Roche), normalising against the gene for ATP

synthase subunit beta (ATP5B), which was determined in-house as the best

internal reference gene out of 12 genes tested (geNorm kit, Primerdesign,

Southampton, UK). The primers used are summarised in the Table below:

57

Table 5: Primers used in qRT-PCR

Primer Sequence Source

Axl NM_021913(2) IDT

Tyro3 NM_006293(1) IDT

MerTK NM_006343(1) IDT

Gas6 NM_000820(3) IDT

CD44 NM_000610(8) IDT

CCND1 NM_053056(1) IDT

MMP9 NM_004994(1) IDT

MMP2 NM_001127891(2) IDT

TIMP1 NM_003254(1) IDT

ATP5B NM_001686.3 Life technologies

GAPDH NM_001686.3/ NM_001256799.1 Life technologies

IDT, Integrated DNA technologies

2.4.6 Determination of most stable reference genes for signalling

studies

The double-dye (hydrolysis) probe geNorm 12 gene kit qRT-PCR Array

contained multiple reference genes. The geNorm module in qbase+ version 2.5

(Biogazelle, http://www.qbaseplus.com) was used to compute expression

stability values for all reference targets. geNorm analysis was performed in Cq

values directly exported from the Roche LightCycler 96 software. geNorm

calculates the gene expression stability measure M (M-value) for a reference

gene as the average pairwise variation V for that gene with all other tested

reference genes. Stepwise exclusion of the gene with the highest M value allows

http://www.ncbi.nlm.nih.gov/nuccore/NM_001686.3

http://www.ncbi.nlm.nih.gov/nuccore/NM_001256799.1

58

ranking of the tested genes according to their expression stability (24). The

geNorm algorithm determines the pairwise variation Vn/n+1, between two

sequential normalisation factors containing an increasing number of genes. A

large variation means that the added gene has a significant effect and should

preferably be included for calculation of a reliable normalisation factor.

Additional reference genes are not required if the cut-off value is below 0.15 as

set by Vandesompele et al. (2002) (24) Rank aggregation analysis was

performed in the R statistical programming environment (version 3.0.2) using

the Rankaggreg package (version 0.4–3) (25). To determine the best

housekeeping gene for our study, we treated SNB-19 and UP007 cells with

BGB324 IC50 concentrations, as well as Axl shRNA clones and vehicle controls

and screen for gene stability. The 7 most stable genes for each cell line under

the two conditions as shown in Table 6 were compared and ATP5B was

determined to be the best housekeeping gene under all conditions.

Table 6: Housekeeping gene stability ranking

SNB-19 RNAi

SNB-19 BGB324

UP007 RNAi

UP007 BGB324

UBC RPL13A B2M UBC

ACTB GAPDH SDHA ATP5B

CYC1 EIF4A2 18S CYC1

GAPDH SDHA EIF4A2 EIF4A2

YWHAZ B2M RPL13A TOP1

SDHA UBC UBC RPL13A

TOP1 ATP5B ATP5B ACTB

59

2.4.7 Immunoprecipitation

Cells were grown to confluence and treated as indicated. To extract

proteins, cells were washed with ice-cold PBS and lysed in Pierce IP lysis buffer

(Thermo Fisher) (See section 2.2.4). Following a 10-minute centrifugation at

10,000 x g, the lysates were pre-cleared by incubation with 10 µl protein A/G-

agarose beads (Santa Cruz) for 1 hour at 4°C with constant rotation. Lysates

were then incubated with 2 µg immunoprecipitation (IP) antibodies for 2 hours

at room temperature, with constant rotation. Antibodies against GAPDH or

Tensin2 (Santa Cruz) were used as species-aligned negative control IP

antibodies. Following incubation, 20 µl of protein A/G-agarose beads was added

to each sample and incubated further for 2 hours. The beads were then

separated by centrifugation at 8,000 x g for 1 minute and washed once with

ice-cold IP lysis buffer and thrice with PBS (Life Technologies). Finally, 4x SDS-

PAGE loading buffer was added to the beads and solubilised samples underwent

SDS-PAGE and Western blotting.

2.4.8 Luciferase reporter assay

50 000 cells were seeded in complete medium and allowed to adhere

overnight. The next day, 500 ng of the reporter plasmids containing promoter

fragments and a 1:100 dilution of the positive transfection control

(pGL4.74[hRluc/TK] vector expressing Renilla luciferase, Promega) were mixed

with JetPrime transfection reagent (Polyplus) according to the manufacturers

protocol, and added to the cells. After 4 hours of incubation at 37ºC, the

transfection medium was replaced with fresh medium and expression was

allowed to occur for further 48 hours. Cells were passively lysed in luciferase

60

assay lysis buffer and analysed for luciferase levels and activity as follows: 20

µl of crude cell lysate was added to 100 µl LARSII solution (luciferase assay

substrate in assay, for firefly luminescence) and the luminescence level was

measured for 10 seconds using a luminometer (1450 microbeta, Perkin Elmer,

Waltham, MA). Following measurement, 100 µl Stop and Glo reagent (for

Renilla luminescence) was added and the luminescence was read again. The

luminescence values measured were subtracted from background readings

(untransfected control lysates) and then normalised against the Renilla

luciferase positive transfection control.

2.5 Functional assays

2.5.1 Cell survival/growth assay

The MTS assay measures the reduction of 3-(4,5-dimethylthiazol-2-yl)-

5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS)

compound in the presence of phenazine methosulphate (PMS), which relies on

NAD(P)H-dependent oxidoreductase enzymes and therefore reflects the

metabolic activity of the cell; this correlates with cell proliferation and viability

(169). This assay therefore served as an indicator of cell growth following

treatment with Gas6 and cytotoxicity following treatment with small molecule

inhibitors and TMZ (170). 1,500 cells per well were seeded in 96-well plates and

incubated overnight, prior to various treatments (see above). Three or seven

days’ post drug or Gas6 treatment respectively, MTS reagent (Fisher Scientific)

was added to cells at 0.4 µM in the presence of 0.3 nM PMS (Sigma) and

incubated for a further 2 hours, after which absorbance was measured at 492

nm using a spectrophotometric microplate reader (Synergy; BioTek, Potton,

61

UK). The calculation used to calculate the IC50 concentrations is summarised in

the following formula:

where AbsIC50 is the absorbance at 492nm that is expected to be generated with

the 50% inhibitory concentration of the inhibitor, AbsIC100 is the absorbance at

492 nm generated with complete cell death following addition of inhibitor

representing the 100% inhibitory concentration, AbsIC0 is the absorbance at 492

nm generated with cell culture as control representing the 0% inhibitory

concentration and AbsBG is the background absorbance at 492 nm generated

with the detection reagents at non-toxic concentrations and without interaction

with the test compounds (170).

2.5.1 Apoptosis assay

Cell membrane changes occur early in apoptosis with externalisation of

the phospholipid phosphatidylserine (PS) from the inner membrane to the outer

leaflet. The calcium–dependent phospholipid–binding protein Annexin V binds

strongly to PS on the external membrane. Detection of fluorochrome–labelled

Annexin V by flow cytometry has been exploited as a marker of early apoptosis.

This assay was used assess the cytotoxicity of the Axl SMI BGB324. 30,000 cells

per sample were treated with 2.5 µM and 5 µM (SNB-19) or 1 µM and 2 µM

BGB324 (UP007) or vehicle (DMSO) for 24 hours. Following drug treatments,

the cells were centrifuged at 400 x g for 5 minutes and re-suspended in Annexin

V binding buffer and incubated with Annexin V-CF488A conjugate and Hoechst

62

33342 (Insight Biotechnology, Wembley, UK) for 15 minutes at 37oC using a

heating block. The cells were then spun down at 400 x g for 5 minutes and

washed with Annexin V binding buffer twice by repeating the centrifugation and

resuspension. Finally, the cells were resuspended in 100 µl supplemented with

10µg/ml propidium iodide (Insight Biotechnology) and 30 µl was loaded onto a

special chamber slide (NC-Slide A2™) and cell populations were analysed for

Annexin V/PI fluorescence (NucleoCounter® NC-3000™; Chemometec A/S,

Allerød, Denmark) according to the manufacturer’s protocol.

2.5.2 Scratch wound assay

Linear cell migration along a surface was measured by scratch wound

assay, where a linear scratch was made in a confluent cell monolayer with a

200 µl yellow pipette tip. Following injury, wound closure was monitored using

an inverted Zeiss Axiovert 200M microscope housed in a live cell imaging

chamber under normal cell culture conditions (37°C, 5% CO2, humidified

atmosphere). Images of 4 points per well were captured every 3 hours over a

total period of 21 hours. Image analysis following the experiment was

performed using ImageJ, and cell migration rates (area change/hour) calculated

thereafter.

2.5.3 Cell motility assay

GBM cells were seeded in 24-well plates at low density (1,500 cells/well).

Following treatment with BGB324 or vehicle (DMSO), live cell microscopic

images of 4 points per well were taken every 20 minutes for 24 hours under

normal culture conditions (37°C, 5% CO2, humidified atmosphere) using a Zeiss

Axiovert 200M inverted live cell microscope with time-lapse imaging at 10x

63

magnification. Quantification of cell tracking, measuring distance and trajectory

were performed using ImageJ with its ‘Cell Tracking’ plug-in. In total, 60 cells

per condition were tracked randomly for a period of 3 hours, with total distance

(µm) and average velocity (µm/min) being calculated.

2.5.4 3D colony growth assay

Wells of 24-well plates were first coated with 0.5 mL of complete

medium/0.5% agar, and then overlaid with GBM cells suspended in complete

medium/0.35% agar at low density (1,500 cells/well). Cells were incubated for

2 weeks in total, with media and treatments being replenished weekly. Colonies

present at the end of the incubation period were stained by incubation with

0.05% crystal violet solution (Fisher Scientific) for 1 hour. Images were taken

of each well using a Zeiss Stemi SVG dissecting microscope (1.5x

magnification), and colonies were counted in 5 different fields of view.

2.5.5 Cell invasion assay

Modified Boyden transwell chambers were used to assess cell invasion

through extracellular matrix. Briefly, polycarbonate hanging inserts (Corning,

Amsterdam, The Netherlands) were coated with 60 µg/ml Geltrex™ LDEV-free

reduced growth factor basement membrane matrix (Life Technologies).

Invasion in the cell culture incubator was allowed to occur over 6 hours for

UP007 cells (20,000 cells/well) and 16 hours for SNB-19 cells (10,000

cells/well). Non-invading cells on the upper surface of the membrane were

removed with a cotton swab and cells that had invaded and adhered to the

lower side of the membrane were fixed for 15 minutes with 37% formaldehyde

solution containing 10-15% methanol (Sigma). Following fixation and brief

64

wash with PBS, the adherent cells were stained with 0.5% crystal violet. Images

were captured by a Zeiss Axiophot brightfield microscope and cells were

counted in 5 random fields at a 10x magnification. Invasion was expressed as

mean fold±standard error of the mean (SEM) of the number of total cells

counted per well.

2.5.6 Cell cycle analysis

SNB-19 cells were serum starved for 24 hours prior to being treated with

50 ng/ml EGF alone or in combination with 2.5 μM BGB324, or vehicle (DMSO)

for 24 hours. Following drug treatments, the cells were centrifuged at 400 x g

for 5 minute, washed with PBS and resuspended in 70% ethanol/PBS (Thermo

Fisher Scientific) for 2 hours on ice. Following a PBS wash, the cells were

incubated with 1 µg/ml DAPI, 0.1% Triton X-100 in PBS (Chemometec A/S,

Allerød, Denmark) for 5 minutes at 37oC. Thirty μl of stained cells was loaded

onto a special chamber slide (NC-Slide A2™) and cell populations were analysed

for DAPI fluorescence (NucleoCounter® NC-3000™; Chemometec) according

to the manufacturer’s protocol.

2.5.7 In vitro kinase activity assay

The Axl Kinase Enzyme System (Promega) was employed according to

the manufacturer’s protocol with minor modifications. Briefly, each of

recombinant Axl or EGFR kinases (Stratech Scientific, Suffolk, UK) was

incubated for 30 minutes with the appropriate substrate peptide

(DCLDGLYALMSRC, 0.2 µg/µl; Pepceuticals Ltd, Leicester, UK), which contained

Tyrosine 779 (Y779) of human Axl within it, in the presence or absence of the

EGFR inhibitor gefitinib (10 µM) or Axl inhibitor BGB324 (10 µM), in Kinase assay

65

reaction buffer containing ATP (25 μM). DMSO was used as vehicle control

instead of inhibitor. ADP-GloTM reagent (Promega) was then added to the wells

and incubated at room temperature for 60 minutes. Finally, the Kinase detection

reagent was added for 30 minutes at room temperature. Luminescence was

read using a microplate luminometer (BMG Labtech Fluorstar Optima,

Offenburg, Germany). To determine the kinase activity of native Axl pulled

down SNB-19 cells by IP, the beads pellet obtained from Axl IP (described

above) was washed a final time with Axl reaction buffer (8,000 x g for 1 minute).

The Kinase assay reaction buffer, containing ATP (25 μM) and the Axl substrate

peptide (0.2 µg/µl) (Promega) were added directly to the pellet, and the kinase

reaction incubated for 30 minutes at room temperature. The reaction was

stopped, developed and detected in the same way as for recombinant kinases

above.

2.5.8 Assessment of –tyrosine kinase phosphorylation status using

an antibody array system

The simultaneous assessment of phosphorylated tyrosine kinases was

assessed using the R&D Proteome Profiler TM Array Human Phospho-kinase

Array kit (R&D). Each array system has antibodies directed against 43

phosphorylated tyrosine kinases and 2 related total proteins spotted in duplicate

on a nitrocellulose membrane. Positive and negative internal controls were also

spotted onto the membranes. Protein determination was performed and 50 µg

of whole cell lysate incubated with each array overnight at 4OC, hybridising to

the arrays via the membrane–bound antibody. The membranes were washed 6

times in TBST for 5 minutes each wash to remove excess lysate. A secondary

66

pan–phosphor-tyrosine antibody was applied for 4 hours which bound to the

phosphorylated tyrosine residues. The membranes were washed 6 times in

TBST for 5 minutes each wash to remove antibody. Signals were detected by

chemiluminescence provided by each kit. The pixel density of each array spot

was measured and expressed as optical density (OD)/mm2. To account for

variation in exposure, 6 blank readings and readings over the negative control

markers were performed. All signal readings were exported to Microsoft Excel.

The mean signal for the pair of duplicate spots representing each TK was

determined and the median blank and negative reading was subtracted to give

the final reading.

2.6 Statistical analyses

All data are expressed as mean±SEM, obtained from a minimum of 3

independent experiments, each constituting multiple replicates per condition.

Quantitative data were analysed by Analysis of Variance (ANOVA) with post hoc

Tukey test for multiple comparisons with one control group or multiple time

points/treatments or paired t-test for comparisons of control with treatment

Statistical analyses and graphical representations were performed using Prism

(GraphPad Software Inc). The level of statistical significance is indicated in the

figures and accompanying legends. Western blot film processing was performed

using Adobe Photoshop CC 2014 software (Adobe Systems Incorporated, CA,

USA).

67

Chapter 3

Investigating TAM receptors

and their ligands in GBM

68

3.0 Introduction

Brain cancer is often associated with high mortality rates and is one of

the most frequent tumours of the economically active population. Glioblastoma

multiforme (GBM), the most common type of brain cancer in adults, is highly

heterogeneous and aggressive, with patients facing a year of survival after

diagnosis (171). Despite intensive research, no real progress has been made in

treating this malignant type of tumour.

Vajkoczy et al, first reported Axl overexpression in gliomas and noted

that high Axl/Gas6 expression in GBM patients correlated with poor overall

survival, indicating that Axl may significantly contribute to glioma progression.

Experimental blockade of Axl signalling with either siRNA or introduction of

dominant negative Axl protein resulted in decreased growth and invasive

potential of GBM cells in culture (137, 138, 140). Axl and Gas6 mRNAs were

then demonstrated to be overexpressed in high grade gliomas and more

specifically, glioma cells of pseudopalisades and reactive astrocytes (137).

Moreover, Axl/Gas6 co-expression occurred in tumour vessels with no

expression detected in non-neoplastic surrounding tissue (137). Interestingly,

Axl inhibition by siRNA resulted in a better response to the current mainstay

glioma chemotherapeutic agent temozolomide, as well as more recent

combination chemotherapeutic drugs carboplatin, and vincristine (140, 172,

173). Together, these observations suggest the TAMs to be an attractive new

target for GBM therapy.

The primary aim of this section is to provide an in depth characterisation

of the TAMs in two human GBM cell lines in terms of TAM expression and

activation by established as well as novel ligands, with the aim of forming a

69

better understanding of the role of TAM signalling in promoting GBM

progression.

3.1 Results

3.1.1 The TAMs are overexpressed in GBM cells

To begin with, an expression screen of all TAMs in different human brain

cell types, as well as the GBM cell lines involved in this study, was conducted.

The commercially available SNB-19 and in-house UP007 GBM cell lines were

selected for this study. First, in advance of the first TAM experiments, a

population doubling time analysis was performed for both cell lines by plating

an equal number of cells (10 000) and then counting them using a

haemocytometer every 24 hours for 3 days. The values were plotted as shown

in Figure 3.1A and the population doubling time was calculated using the

equation derived from the slope. SNB-19 cells were calculated to double every

27 hours and UP007 cells every 30 hours. In addition, clonogenic assays

confirmed that the SNB-19 cell line has a greater proliferative capacity as

indicated by the number of colonies produced after two weeks in soft agar

(Figure 3.1B). However, the UP007 cell line was the more invasive of the two,

as more cells were able to invade through a matrix compared to SNB-19 cells

over 6 hours (Figure 3.1C).

70

Figure 3.1: Cell line characteristics. A. Cell number plotted versus time graph. Population doubling time was calculated by the line equation (N=3). B. Bar graph showing the number of colonies counted for each cell line after two weeks in soft agar (N=3). C. Representative pictures of cell invasion. Cells were allowed to invade through matrix for 6 hours before staining using PDGF-AA as a chemoattractant (N=1 experiment); scale bar=55 µm. Unpaired two-tailed student t-test. Data are mean±SEM *p<0.05. PDT, population doubling time.

After establishing these fundamental phenotypic characteristics of the

cells, a comparative expression analysis of the TAM receptors across the various

cell types was performed. As shown by Western blotting in Figure 3.2A, all three

TAMs were expressed in human brain microvascular endothelial cells

(hcMEK/D3), whilst only Axl and Tyro3 were expressed in normal human

astrocytes (HA-c). Both Axl and Tyro3 were found to be highly expressed in

both of the GBM cell lines SNB-19 and UP007, whilst MerTK was expressed at

71

low levels in SNB-19 cells. qRT-PCR confirmed the Western blot results, showing

similar expression patterns at the mRNA level (Figure 3.2B-E). Additionally,

Gas6 mRNA was found at high levels in cultured normal human cerebellar

astrocytes and the UP007 cell line (Figure 3.2E). As MerTK levels were minimal

in the SNB-19 cell line and not detectible in the UP007 cell line, the rest of the

studies focused only on Axl and Tyro3 receptors.

Figure 3.2: TAM and ligand screen. A. Western blot screen of TAM receptors in protein extracts from a panel of human brain cell cultures: microvascular endothelial cells (hCMEC/D3), astrocytes (HA-c) and two GBM cell lines, SNB-19 and UP007. B. Quantitative PCR analysis of mRNA expression of the genes for Axl, MerTK, Tyro3 and Gas6 in extracts from human whole brain (WB), endothelial cells, astrocytes, SNB-19 and UP007 cells (N=3 separate experiments).

72

In conjunction with previous studies on Axl expression in GBM available

on Gene Expression Omnibus (GEO) profiles, a public functional genomics data

repository, slides with a brain tumour tissue microarray (TMA), including normal

brain sections, were immunostained for Axl and Tyro3. Staining confirmed

expression of Axl in reactive astrocytes in 14 out of 73 GBM samples (Figure

3.3), whilst Tyro3 expression varied in samples with high and low expression in

the normal brain and GBM tissues (Figure 3.4).

Figure 3.3: Axl screening and tumour formation abilities. Left panel shows representative images for Axl staining in normal brain, middle panel presents Axl negative GBM and Axl positive GBM is presented in right panel. Arrows represent Axl positive endothelial cells in normal brain vasculature and positive astrocytes in Axl positive GBM tissue. Scale bar=68 µm.

73

Figure 3.4: Tyro3 screening. Left panel shows representative images for Tyro3 staining in normal brain, middle panel presents low Tyro3 expression in GBM tissue and high Tyro3 expression in GBM is presented in right panel. Arrows represent Axl positive endothelial cells in normal brain vasculature and positive astrocytes in Axl positive GBM tissue. Scale bar=53 µm.

Moreover, in collaboration with researchers at the University of Bergen,

orthotopic implantation of 500 000 SNB-19 and UP007 cells into nude rats was

performed. The SNB-19 cells were able to produce tumours whilst the UP007

cell line failed (Figure 3.5).

Figure 3.5: Axl tumour formation abilities. Left: representative image of cell implantation into rat brain in a stereotactic frame. Middle: Coronal section showing MRI detection of tumour mass (red arrow) of SNB-19 cells 9 months after implantation. Right: Coronal section MRI scan of rat brain injected with UP007 cells, with no sign of tumour.

74

3.1.2 Gas6 activates Axl signalling in GBM cells

Gas6 and Protein S are the natural ligands for the TAMs in both normal

and tumorigenic environments. However, studies have indicated that

overexpression of Axl in cancer can lead to ligand-independent activation.

Having established the overexpression of Axl in glioblastoma cell lines, the next

phase focused on receptor activation. Gas6 is the highest affinity ligand for all

three TAM receptors and has also been found to be overexpressed in glioma

(137, 138). The cultured GBM cells were stimulated with exogenous

recombinant human Gas6 over a time course period and cell lysates were

analysed by Western blot in order to detect receptor activation through its

phosphorylation state. Addition of Gas6 resulted in a significant increase in

phosphorylation of Axl in SNB-19 cells after 15 minutes of stimulation (Figure

3.6A, p<0.0061, ANOVA with Tukey’s multiple comparison post-hoc analysis)

but failed to do the same in UP007 cells even after 1 hour (Figure 3.6B,

p<0.9627, ANOVA with Tukey’s multiple comparison post-hoc analysis).

Additionally, Gas6 was unable to stimulate Tyro3 phosphorylation in both cell

lines (Figure 3.6E-F, p<0.8504 and p<0.7441 for SNB-19 and UP007

respectively, ANOVA with Tukey’s multiple comparison post-hoc analysis).

However, Gas6 rapidly induced downstream intracellular Akt phosphorylation,

a well-known downstream Axl target, in both cell lines by approximately two-

fold, therefore supporting the activation of Axl signalling in these cells (Figure

3.6C-D, p<0.0294 and p<0.0156 for SNB-19 and UP007 cells respectively,

ANOVA with Tukey’s multiple comparison post-hoc analysis) (Total Axl Western

blots are shown in Appendix 1).

75

Figure 3.6: Effect of Gas6 stimulation on TAM phosphorylation and signalling in GBM cells. Western blot showing time-course of Axl phosphorylation by Gas6 (400 ng/ml) in SNB-19 and UP007 cells (A; N=5 blots for both cell lines), of Tyro3 phosphorylation in SNB-19 and UP007 cells (B; N=3 blots for both cell lines), and of pAkt levels (C) in SNB-19 (N=3 blots) and UP007 (N=4 blots) cells. Data are mean±SEM protein expression normalised against loading control protein; ANOVA with Tukey’s multiple comparison post-hoc analysis. ***p<0.001, *p<0.05 and ns, not significant, versus time 0.

76

3.1.3 Effect of novel proposed TAM ligands on Axl activation

While both Gas6 and Protein S are well established ligands for the TAM

receptors, less is known about Tulp-1, Tubby and Galectin-3; these three

molecules were recently suggested to activate MerTK to enhance its role in

efferocytosis (111, 174). Therefore, having investigated the effect of Gas6 on

Axl activation, the focus turned on Tulp-1 and Galectin-3 as they have been

shown to be overexpressed in many cancers but had hitherto not been studied

in brain tumour cells. Comparing with the effects of Gas6 and Protein S as

controls, the response of TAMs to Galectin-3 and Tulp-1 was tested by Western

blotting for the phosphorylated receptors. Axl and Tyro3 were activated by both

novel ligands in a dose-dependent manner in both cell lines. Choosing the

higher concentrations of 500 ng/ml for Galectin-3 and 1000 ng/ml Tulp-1

respectively (Figure 3.7A-B), time course stimulation experiments confirmed

activation of the receptors. The SNB-19 cell line was found to be more

responsive to both novel ligands, with Axl activation occurring within two

minutes of ligand addition. The UP007 cell line however, did not respond to

Tulp-1, and an increase in Axl phosphorylation was observed only after 60

minutes of Galectin-3 ligand addition (Figure 3.7C-D).

77

Figure 3.7: Effect of ligand stimulation on TAM phosphorylation and signalling in GBM cells. Western blot showing dose response of Axl (A) and Tyro3 (B) phosphorylation by Gas6 (400 ng/ml), Protein S (1000-10 ng/ml), Galectin-3 (500-5 ng/ml) and Tulp-1 (1000-1 ng/ml) (N=1 blot). C. Western blot showing time-course of Axl phosphorylation by Galectin-3 (500 ng/ml) in SNB-19 and UP007 cells (N=2 blots). D. Western blot showing time-course of Axl phosphorylation by Tulp-1 (1000 ng/ml) in SNB-19 and UP007 cells (N=2 blots).

78

3.1.4 TAM activation promotes cell growth/survival

The 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-

sulfophenyl)-2H-tetrazolium (MTS) assay has been long used in cell culture to

measure cell growth (175). The assay enables the quantification of cell viability

through the reduction of MTS tetrazolium compound by viable cells to generate

a coloured formazan product that is soluble in cell culture medium (176). In

order to assess the effect of TAM activation by traditional and novel ligands,

cells were plated in equal numbers into 96-well plates and serum starved for

24 hours prior to addition of ligands at increasing concentrations for 3 days. As

shown in Figure 3.8A, Gas6 at 600 ng/ml significantly increased cell

growth/viability after 3 days in SNB-19 cells (Figure 3.8A, p<0.0281, ANOVA

with Tukey’s multiple comparison post-hoc analysis), whereas 1000 ng/ml of

Gas6 was required for UP007 cells to show increased cell growth/viability

(Figure 3.8A, p<0.0313, ANOVA with Tukey’s multiple comparison post-hoc

analysis). In addition, Galectin-3 significantly increased growth/viability of SNB-

19 cells even at the lowest concentration of 0.5 ng/ml (Figure 3.8B, p<0.0028,

ANOVA with Tukey’s multiple comparison post-hoc analysis). An increase in cell

growth/viability was also observed in the UP007 cells, which required the higher

dose of 500 ng/ml (Figure 3.8B, p<0.0374, ANOVA with Tukey’s multiple

comparison post-hoc analysis). Tulp-1 stimulation had no effect on cell

growth/viability as shown in figure 3.8C (p<0.9799 and p<0.7533 for SNB-19

and UP007 cells respectively, ANOVA with Tukey’s multiple comparison post-

hoc analysis).

79

Figure 3.8: Effect of TAM activation on cell proliferation. MTS assay for serum starved SNB-19 and UP007 cells treated with recombinant Gas6 (0-1000 ng/ml; A), Galectin-3 (0-500 ng/ml; B) and Tulp-1 (0-1000 ng/ml; C) for 3 days. Data are mean±SEM; N=3 separate experiments, with each treatment in quadruplicate wells. ANOVA with Tukey’s multiple comparison post-hoc analysis *p<0.05; ***p<0.001 vs untreated; ns, not significant.

80

3.2 Discussion

The related RTKs Axl and MerTK have recently been associated with

glioma pathophysiology (137, 140), and therefore have emerged as attractive

novel therapeutic targets for brain tumours as part of individualised treatment

approaches. In this section, the TAM and TAM ligand profile for two

glioblastoma cell lines was provided.

3.2.1 TAM expression in the brain

The TAMs are expressed ubiquitously in the body and have a wide variety

of functions. Expression levels vary between tissues depending on function.

Tyro3 is most highly expressed the nervous system, ovaries, testis, breast, lung,

kidney, osteoclasts, retina, monocytes/macrophages, and platelets (120). Axl

has the broadest expression profile of all the TAMs, being expressed in a wide

variety of tissues including the hippocampus, cerebellum, heart, skeletal

muscle, liver, kidney, testis, brain, monocytes, macrophages, platelets,

endothelial cells and dendritic cells (120). MerTK expression is found in the

ovary, prostate, testis, lung, retina, and kidney, and macrophages, dendritic

cells, NK cells, NKT cells, megakaryocytes and platelets; low levels are also

found in the heart and skeletal muscle cells (120). Of particular relevance to

this study is the finding that the normal brain expresses very low to

undetectable levels of MerTK and Axl (121).

In agreement with previous studies, Axl was shown to be highly

expressed in human normal brain microvascular endothelial cells whilst at low

levels in cultured normal human astrocytes (123). Both qRT-PCR and Western

blot analyses revealed that Axl is overexpressed in glioblastoma tumours (Figure

81

3.1). In accordance with previous findings by Hutterer and colleagues, Axl

expression was also confirmed by human glioblastoma tissue staining and found

to be mostly present in reactive astrocytes (137) (Figure 3.3). In contrast to

previous reports showing MerTK to be overexpressed in glioma and promote

survival and invasion (177, 178), it was found to only be expressed in human

microvascular endothelial cells, but was absent from astrocytes and was

detected only at low levels in the SNB-19 cell line. Tyro3 was as expected

present in both the human microvascular endothelial cells and astrocytes (121);

however, interestingly, it was also found to be overexpressed in both GBM cell

lines. This indicates a novel avenue for study, as Tyro3 overexpression has not

been previously reported in brain cancer. However, given that only two GBM

cell lines were tested, it is not yet clear whether Tyro3 overexpression is more

widespread in GBM or if it has any wider significance in tumorigenesis. An

immunohistochemical analysis for Tyro3 expression in a 96 human brain cancer

tissue microarray screen turned out futile due to lack of an effective antibody

for immunostaining. Two different anti-human Tyro3 antibodies were tested,

both of which appeared to stain the entire tissue non-specifically, thus making

it impossible to make any conclusions (Figure 3.4). Ideally, both Western blot

and qRT-PCR analysis should be performed on fresh biopsy tissues in order to

overcome this issue. Similarly, Gas6 was only detectable by qRT-PCR analysis,

also due to the lack of a specific antibody. Nevertheless, Gas6 was present in

all the samples tested at the mRNA level, with highest expression in human

astrocytes and the UP007 cell line, as also confirmed by the transcriptome

database of acutely isolated purified astrocytes, neurons, and oligodendrocytes,

82

which was assembled by the Barres group to provide cell-type-specific markers

(123).

3.2.2 RTK activation by classic TAM ligands

The TAMs are known to be activated by binding of their natural ligands

Gas6 and Protein S, both homologous vitamin K-dependent proteins. Both

ligands contain an N-terminal region comprised of the vitamin K-dependent set

of post-translationally modified Gla residues, a loop region, four EGF-like

repeats, and a SHBG-like structure that is composed of two LG domains. The

loop region of Protein S contains thrombin-sensitive cleavage sites which are

not present in Gas6, setting the two ligands apart (108).

Even though Gas6 shows 43% amino-acid sequence identity with protein

S (108), studies have shown that Axl is activated exclusively by Gas6 but not

by Protein S (103). In contrast, Tyro3 preferentially binds Protein S compared

to Gas6, whereas MertK binds both ligands weakly (103). Additionally, it has

been shown that structurally, the two LG domains of Gas6 possess specific

epitopes that crosslink the two Axl Ig domains with a 2:2 stoichiometry (Figure

3.9), whereas Gas6 binds Tyro3 and MerTK with a 1:1 stoichiometry due to the

lack of a major Gas6 binding site on the latter receptors (103).

Figure 3.9: Binding of Gas6 on Axl receptor vs Tyro3/MerTK. Figure shows Gas6-induced Axl homodimerisation on the left and weak Gas6 binding to either Tyro3 or MerTK. Figure adapted from (103).

83

Studies have shown that Axl overexpression can involve self-activation

even in the absence of ligand (179). Indeed, in both the SNB-19 and UP007 cell

lines, a basal Axl activation was observed, which only in the case of the SNB-

19 cell line was augmented by exogenous Gas6 (Figure 3.6). A basal level of

Tyro3 phosphorylation was also detected by Western blot; however, exogenous

Gas6 had no additional effect on receptor activation (Figure 3.6). Consequently,

TAM receptor activation may occur in two ways: (i) Gas6 can activate the TAMs

in an autocrine and/or paracrine manner. Both of our cell lines express Gas6 at

the mRNA level, thus being able to activate the receptors fully to reach a

saturated state. Indeed, Gas6 was present in the SNB-19 cells at lower levels

than UP007, and therefore this could explain why this cell line was responsive

to Gas6 stimulation, as the exogenous ligand was abler to activate the cell line

that possessed less of the ligand and therefore was more sensitive to it. (ii)

Alternatively, Axl and Tyro3 overexpression leads to ligand-independent

activation, therefore addition of the ligand has no dramatic effect. Additionally,

the independence of Gas6 observed in the case of Tyro3 could be a by-product

of receptor heterodimerisation which is discussed in detail in Chapter 6. The

TAMs have been speculated to heterodimerise and form a signalling complex.

Indeed, Axl and Tyro3 have been shown to co-immunoprecipitate in rat cells

and able to cross-phosphorylate each other (180).

Gas6 had no effect on Tyro3 phosphorylation in both SNB-19 and UP007

cells. Given the preferential binding of Gas6 to Axl, the particular ligand-receptor

binding stoichiometry, as well as a greater overexpression of Axl vs Tyro3 in

the GBM cells, Axl could act as a quencher and thereby prevent any Tyro3

activation by the ligand.

84

3.2.3 RTK activation by novel TAM ligands

Galectin-3 and Tulp-1 have recently been identified as novel ligands for

the TAMs (111, 174). Galectin-3 is a member of the lectin family and is

comprised of two protein domains, a C-terminal carbohydrate-binding domain

and an N-terminal oligomerisation domain (181) (Figure 3.10).

Figure 3.10: Galectin-3 structure. Figure was adapted by (182).

Galectin-3 has been linked to many different cellular processes, with the

most relevant being cell adhesion, cell growth and differentiation, cell cycle

regulation and regulation of apoptosis (183). Moreover, Galectin-3 expression

has been shown to be highest in GBM compared to other types of brain cancer

and more specifically in neoplastic astrocytes, macrophages/microglial cells,

endothelial cells and some B- and T-lymphocytes (184). Galectin-3 expression

has been linked to glioma cell motility, angiogenesis, survival and invasion

(Figure 3.11) but the full repercussions and mechanisms of expression are still

unclear (185, 186). Galectin-3 expression in GBM coincides with the one for Axl,

indicating that receptor activation by Galectin-3 could potentiate a novel

activation mechanism for Axl. Interestingly, Galectin-3 biological functions are

85

modulated by MMP2 and MMP9 (184), both of which are downstream Axl

targets (187). Hence, Axl may modulate Galectin-3 to promote GBM cell

invasion and survival. Moreover, Galectin-3 overexpression could serve as to

inhibit microglia activation through TAM activation and thus limit tumour

induced inflammation. The functions of Galectin-3 in GBM could be numerous

and diverse thus making it a promising target along with the TAMs.

Figure 3.11: Presumed role of galectins in angiogenesis and chemoresistance. Necrotic foci in glioblastoma are typically surrounded by hypercellular zones referred to as pseudopalisades. It has been shown that pseudopalisades are hypoxic and express high amounts of the HIF-1. Galectin-3 (Gal-3) expression is stimulated by hypoxia via HIF-1 and stimulates angiogenesis in vitro and in vivo. Gal-3 has also been shown to interact with NG2 proteoglycan, a component of microvasculature pericytes, which stimulates endothelial cell motility and morphogenesis. In addition, Gal-3 is involved in chemoresistance, a process that is increased in hypoxic conditions (184).

Tulp-1 belongs to novel family of tubby proteins which share the highly

conserved C-terminal (Figure 3.12). Tulp-1 is selectively expressed in the retina

and is considered to be a phagocytosis-stimulating molecule in retinal pigment

epithelium cells (188).

86

Figure 3.12: Tub proteins structure. Figure shows the conserved c-terminal tubby domain which contains the DBD and the phosphatidylinositol-binding region and the less conserved N-terminus is (189).

Information on Tulp-1 function and signalling is still scarce. Nevertheless,

from a handful of studies, Tulp-1 was shown to interact with all three TAM

receptors in co-immunoprecipitation experiments (111), whilst an interaction

with Galectin-3 was only confirmed for MerTK, with its affinity for Axl and Tyro3

being unknown in that study (174). Tulp-1 was shown to act as a bridging

molecule through its N-terminal minimal phagocytic determinants domain for

MerTK binding and C-terminal phagocytosis prey-binding domain for prey

binding (111). Functionally, both ligands were found to be associated with

MerTK associated phagocytosis (Figure 3.13) (111, 174); however, their

physiological significance in other settings is yet to be determined.

Figure 3.13: Tulp-1 facilitates phagocytosis as MerTK-dependent bridging molecule. Tulp1 bridges phagocytosis preys, through the C-terminal PPBD and MerTK through the N-terminal MPD(s) leading to phagocytosis (111).

87

In this study, both Galectin-3 and Tulp-1 activated Axl and Tyro3

phosphorylation in both GBM cell lines (Figure 3.7), with a more pronounced

effect in SNB-19 cells. Tulp-1 seems to activate Tyro3 more strongly than Axl

in both cell lines, whilst Galectin-3 showed no preference in receptor activation.

Interestingly, Tulp-1 increased Tyro3 phosphorylation as much as Protein S,

which is the preferred Tyro3 ligand (108); however, Galectin-3 failed to do so.

Moreover, Galectin-3 led to Axl phosphorylation in Gas6-independent UP007

cells whilst it had a less pronounced effect on Tyro3, indicating a stronger

affinity for Axl, as well as a higher receptor affinity than Gas6. Even though

phosphorylation of Axl was observed by both ligands in GBM cells, further

devoted studies are required to determine the level of activation and affinity for

the receptors. Since these two ligands are structurally different from Gas6 and

Protein S, crystallographic and modelling studies can shed light on the exact

binding sites and how these differ from canonical ligands. Moreover, it still

unclear whether binding of Galectin-3 and Tulp-1 to the TAMs diversifies

downstream signalling. Microarray studies following stimulation with all four

ligands and bioinformatics analysis could indicate common pathways as well as

point out proposed functions. Furthermore, kinase activity assays following

ligand stimulation can indicate the degree of receptor activation, as well as

receptor kinase functionality. In the case of glioma, it would be interesting to

perform functional assays such as scratch wound, invasion and proliferation

assays to identify their role. Moreover, knockdown of the ligands should be

investigated in glioma in order to determine ligand compensation, as well as

antagonism. It would also be interesting to investigate the effect of

overexpression of all ligands and how their expression affects TAM receptor

88

activation and downstream signalling. Given their diverse roles, it is possible

that ligand binding determines TAM receptor downstream response and

predetermines the biological function regulated. For example, Galectin-3

binding can affect the TAMs on microglia more than endothelial cells, whereas

Gas6 may act more potent as an angiogenesis promoting factor. In this study,

the activation of both Axl and Tyro3 receptors has been investigated however

these results are only preliminary and were not pursued further due to time

constraints and other priorities of the project. Thus, a more detailed

investigation should ensue.

3.2.4 TAM activation promotes cell growth

In this study Galectin-3 was shown to drastically increase cell

growth/survival after serum starvation of the cells whilst Tulp-1 had no effect.

Even though Gas6 has previously been shown to protect cells of the CNS from

serum deprivation-induced cell death (190), such an investigation had not

hitherto been carried out for Galectin-3. Treatment of cells with Galectin-3

drastically increased cell viability/growth as shown by MTS assay (Figure 3.8),

indicating for the first time that it may act as a mitogenic or pro-survival factor.

Moreover, Galectin-3 was able to increase cell viability/growth at lower

concentrations than Gas6. Together with the observation that Galectin-3

activated Axl more readily than Gas6 in time-course experiments, these results

may suggest that Galectin-3 maybe a stronger ligand for Axl binding. However,

this cannot be proven until biochemical and biophysical investigations are

carried out to determine the absolute affinity, kinetics and stoichiometry of the

protein-protein interaction between Galectin-3 and Axl. Also, as the MTS assay

89

cannot distinguish between increased proliferation or reduced cell death, to

further investigate the comparative effects of Gas6 and Galectin-3 on GBM cells,

a detailed cell cycle analysis should be performed after treatment in order to

identify how each cell cycle phase is affected. This technique is based on the

ability of a fluorescent dye to stain DNA. The fluorescence intensity correlates

with the amount of DNA in the cell which changes as the cell exits G1 to enter

the S and progress to G2 and M phases (191). Moreover, the effect observed

could be the result of an alternative mechanism, as Galectin-3 has been shown

to bind and activate other receptors as well. In order to distinguish between Axl

and other receptors a blocking antibody for Axl can be used to prevent ligand

binding and then stimulate cells with exogenous Galectin-3. If Galectin-3 acts

via Axl then no change in cell viability would be observed.

3.2.5 Summary

This chapter focuses on the functional and molecular characterisation of

two GBM cell lines, SNB-19 and UP007, and their responses, and

responsiveness, to the TAM ligand Gas6 in both ligand-dependent and –

independent manners. This will enable us to further investigate any signalling

pathways involved downstream of Axl in these cell lines as well as to elucidate

the functional importance of Axl in GBM.

90

Chapter 4

Axl blockade inhibits GBM cell

invasion

91

4.0 Introduction

Novel SMIs of RTKs are increasingly being developed as promising new

targeted molecular therapies for a variety of cancers. Axl RTK has become an

increasingly attractive target in the past years, especially since the discovery of

its overexpression and related increased invasiveness and chemoresistance in

a large number of cancers (102, 192, 193). BGB324 (also known as R428) [(1-

(6,7-dihydro-5H-benzo[6,7]cyclohepta[1,2-c]pyridazin-3-yl)-N3-((7-pyrrolidin-

1-yl)-6,7,8,9-tetrahydro-5H-benzo[7]annulene-2-yl)-1H-1,2,4-triazole-3,5-di-

amine)] is a highly selective SMI of Axl (out of 133 tyrosine and serine/threonine

kinases tested) (160). In MTS assays, BGB324 was shown to affect cell viability

with IC50 values ranging from 0.79–2.13 μmol/L in Ewing sarcoma (194) and to

exhibit minimal off target effects with concentrations varying from 0.95 to 3.39

µM in a variety of cell lines (160). Treatment of mesenchymal carcinoma cells

with BGB324 abolished cell invasion and increased chemosensitivity (195),

whilst it nullified the tumorigenic effect of triple negative breast cancer as well

as blocking their acquired resistance to erlotinib (anti-EGFR SMI). BGB324 is

currently in Phase 1b clinical trials for acute myeloid leukaemia and non-small

cell lung cancer, and has so far been shown to be safely tolerated in clinical

safety studies in healthy volunteers at doses up to 1.5 g daily

(http://www.bergenbio.com/BGB324). Additionally, in breast cancer mouse

models, twice daily dosing of 25, 50, and 100 mg/kg BGB324 resulted in mouse

plasma concentrations of 2.4, 6.8, and 9.0 µM BGB324 respectively (160).

BGB324 represents the most promising Axl SMI due to its high selectivity for

Axl compared to other SMIs currently in trials (163)

92

In this chapter, a comprehensive study of the cell biological and

signalling effects of BGB324 on two distinct patient-derived GBM cell lines is

presented. Additionally, confirmation of the effects of BGB324 as being Axl-

selective was enabled through parallel experiments on the same GBM cell lines

with Axl stably knocked down.

4.1 Results

4.1.1 Establishment of stable Axl knockdown cell lines

In the previous chapter, it was established that Axl is overexpressed in

two human GBM cell lines. To determine whether removal of Axl expression

influences survival, proliferation and invasion in glioma cells, stable Axl

knockdown GBM cell lines were created using 4 different short hairpin (sh) RNA

vectors, along with additional cell clones incorporating one scramble shRNA

vector and one empty vector. The chosen shRNA containing vectors contained

a puromycin resistance cassette for selection. Initially, a killing curve was

performed for one week with puromycin concentrations ranging from 0 to 1.5

µg/ml as suggested by the manufacturer’s protocol, to determine the specific

sensitivity of the two GBM cell lines. Cell death was observed in both cell lines

at the higher concentrations of puromycin, and the effective, highest

concentration of puromycin at which cells could survive was determined to be

1 µg/ml for the SNB-19 cell line and 0.75 µg/ml for UP007 cells (Figure 4.1).

93

Figure 4.1: Puromycin killing curve for SNB-19 (left) and UP007 (right). 250 000 cells were plated in 6-well plates and treated with puromycin concentrations 0-1.5 µg/ml for two weeks. Images obtained using the x5 objective under the phase contrast microscope.

Following puromycin titration, cells were plated in 10 cm dishes and

transfected with 1 µg of 4 unique 29 mer shRNA OriGene constructs in retroviral

untagged vector and control vectors (Cambridge Bioscience; Cambridge, UK).

After three days, puromycin was added at predetermined concentrations and

selection was allowed to occur over two weeks. Cells that had taken up the

vector would be resistant to puromycin and thus stayed alive. After this period,

transfected cells remaining alive in the dish were trypsinised and seeded into

new 10 cm dishes at 1:10 dilution, which allowed single cells to adhere apart

from each other and expand clonally thereafter.

After 4 weeks, each of 12 clear colonies for each vector were picked with

a yellow tip and transferred into a well of a 24-well plate, whilst remaining under

94

puromycin selection. For each cell line, 24 individual colonies were picked and

expanded. The clones were screened for Axl expression by Western blot. For

the SNB-19 cells, we successfully obtained a clone for each construct with

varying Axl knockdown levels, whilst, notably, the UP007 cells only produced

one viable clone that was able to be expanded (Figure 4.2).

Figure 4.2: Western blot for selected cell clones confirming Axl knockdown for SNB-19 (left) and UP007 (right) cells, followed by densitometric semi-quantitation of Axl expression normalised to β-actin. (N=1)

4.1.2 Axl knockdown reduces cell growth/viability

Following the successful establishment of stable Axl knockdown clones

for both cell lines, the MTS assay was employed to determine the effect on cell

growth/viability. The SNB-19 cell line was not particularly affected by Axl

knockdown as shown in Figure 4.3 (p<0.3371, ANOVA followed by Tukey’s

multiple comparisons test). In stark contrast, establishing viable clones with Axl

knockdown for the UP007 cell line was challenging. The clones were not viable

and the MTS assay also confirmed significantly decreased cell viability and

95

growth rate for the single UP007 clone lacking Axl (Figure 4.3, p<0.0001,

ANOVA followed by Tukey’s multiple comparisons test. This suggested that

knockdown of Axl deleteriously affected the survival of these cells.

Figure 4.3: MTS assay for Axl knockdown clones. Data are mean ±SEM; N=3 separate

experiments with each treatment in quadruplicate wells. ANOVA with Tukey multiple

comparison post-hoc analysis *p<0.05; ****p<0.0001; ns, not significant, for

comparisons indicated.

4.1.3 Axl knockdown inhibits cell migration

The scratch wound, or wound healing, assay is used to study cell

migration in vitro. An artificial gap is made on a confluent cell monolayer, which

is filled by migrating cells on the gap edge. Capture of images at regular

intervals can then enable quantitative determination of the rate of cell migration

(196). Near-complete Axl knockdown in SNB-19 cells severely impaired cell

migration (Figure 4.4A, p<0.0022, ANOVA followed by Tukey’s multiple

comparisons test). However, even a low, residual level of Axl expression in

shRNA clones was sufficient to counteract this anti-migratory effect of Axl

knockdown in SNB-19 cells (Figure 4.4A). A decreased migration rate was also

observed in the single UP007 Axl knockdown clone (Figure 4.4B, p<0.0001,

ANOVA followed by Tukey’s multiple comparisons test).

96

Figure 4.4: Effect of Axl knockdown on cell migration. Representative images from scratch wound assay at 0 and 21 hours followed by graphical representation of area change per hour for (A) SNB-19 shAxl clones and control and (B) UP007 shAxl clone and control. Data are mean ±SEM; N=3 separate experiments with 2 scratches per cell clone.; ANOVA with Tukey multiple comparison post-hoc analysis ****p<0.0001, **p<0.01, * p<0.05 for comparisons indicated.

97

4.1.4 Selective Axl inhibition by two SMIs reduces cell viability

For this study, the effect on cell viability of two SMIs currently in phase

I clinical trials, BGB324 (BergenBio, Bergen, Norway) and BMS777607 (Santa

Cruz, CA, USA) was evaluated. Cells were treated with BGB324 (5 µM-0.1 µM)

or BMS777607 (2-100 µM) and cell viability was assessed after three days by

MTS assay (Figure 4.5). BGB324 significantly reduced cell viability from 2 µM in

SNB-19 cells (Figure 4.5A, p<0.0001, ANOVA with Tukey multiple comparison

post-hoc analysis) whilst this was apparent from 1 µΜ in UP007 cells (Figure

4.5B, p<0.0001, ANOVA with Tukey multiple comparison post-hoc analysis).

BMS777607 on the other hand reduced cell viability from 20 µM for both SNB-

19 and UP007 cells (Figure 4.5C-D, both p<0.0001, ANOVA with Tukey multiple

comparison post-hoc analysis). The half maximal inhibitory concentration (IC50)

measures the concentration of an inhibitor which is required for it to reduce the

measured response by half. In this way, the IC50 is usually used to describe the

potency of a drug. In order to directly compare the two SMIs used in this study,

the IC50 values were calculated from the concentration–response curves

generated from the MTS assay.

In the MTS assay, the IC50 for BGB324 was calculated to be 2.5 μM for

SNB-19 cells and 1 μM for UP007 cells, whereas for BMS 777607, it was

calculated to be 118.5 µM for SNB-19 and 35 µM for UP007 cells (Figure 4.5).

These results indicate that the UP007 cell line is roughly 2.5-fold more

susceptible to Axl inhibition than SNB-19, indicating that it may be more

dependent on Axl signalling. Since BGB324 was more potent than BMS777607,

we decided to utilise exclusively BGB324 thereafter as a research tool to inhibit

Axl for the rest of the study.

98

Figure 4.5: MTS assays for SNB-19 (A and C) and UP007 (B and D) cells treated with varying concentrations of BGB324 and BMS777607 respectively followed by % inhibition vs Log BGB324/BMS777607 concentration graphs with IC50 values. Data are mean ±SEM; N=3; ANOVA with Tukey multiple comparison post-hoc analysis ****p<0.0001, ***p<0.001, **p<0.01, * p<0.05 vs untreated (0) cells.

99

4.1.5 BGB324 selectively inhibits Axl activity in GBM

The effect of BGB324 on Gas6-stimulated activity of Axl and Tyro3 in the

GBM cell lines was investigated by Western blotting of the phosphorylated

(activated) states of both RTKs. Pre-incubation with BGB324 for 1 hour, at up

to 10 µM, did not inhibit Gas6-induced Axl phosphorylation in SNB-19 cells

(Figure 4.6A, p<0.4875, ANOVA with Tukey multiple comparison post-hoc

analysis). However, in UP007 cells, BGB324 markedly inhibited Axl

phosphorylation (which was Gas6-independent) at 10 µM to below baseline

levels (Figure 4.6A, p<0.0202, ANOVA with Tukey multiple comparison post-

hoc analysis). BGB324 had no effect on Tyro3 phosphorylation levels in SNB-19

cells (Figure 4.6B, p<0.8387, ANOVA with Tukey multiple comparison post-hoc

analysis), whereas it showed a trend towards significance in inhibition in UP007

cells (Figure 4.6B, p<0.7173, ANOVA with Tukey multiple comparison post-hoc

analysis). At 100 µM, BGB324 was toxic to all cells, as reflected by the absence

of a band for β-actin in the Western blots (Figure 4.6A-C).

Next, the effect of BGB324 on activation of downstream signalling was

investigated in the GBM cells. BGB324 pre-treatment significantly inhibited

downstream Akt phosphorylation in a concentration-dependent manner in both

cell lines, independently of Gas6 stimulation (Figure 4.6C, p<0.0187 and

p<0.0202 for SNB-19 and UP007 respectively, ANOVA with Tukey multiple

comparison post-hoc analysis). Therefore, the Akt signalling pathway appears

to emanate from Axl activation in both GBM cell lines. (Total Axl Western blots

are shown in appendix 2).

100

Figure 4.6: Comparative efficacies of BGB324 for inhibition of Axl, Tyro3 and Akt phosphorylation in GBM cells. Western blots showing inhibition by BGB324 (0.1–100 µM) of phosphorylation of Axl (A), Tyro3 (B) and Akt kinase downstream (C) stimulated by Gas6 (400ng/ml) in SNB-19 and UP007 cells. Data are mean ±SEM (N=3 separate experiments); ANOVA with Tukey multiple comparison post-hoc analysis **p<0.01, *p<0.05, ns, not significant, for comparisons indicated.

101

4.1.6 BGB324 inhibits GBM cell growth and colony formation

Having observed that Gas6 stimulation of Axl promoted GBM cell

growth/survival, and that BGB324 robustly inhibited Axl signalling in the GBM

cells, the effect of BGB324 on cell colony formation in 3D was investigated. The

soft agar assay was thus employed, which allows for a valid assessment of drug

effect on cell growth and its ability to undergo unlimited division (197, 198). A

solution of single cells was plated in soft agar and allowed to grow over the

period of two weeks, either in the presence or absence of BGB324 at a range

of concentrations. Colonies formed were stained with crystal violet and counted

at 5 different fields of view; each qualifying colony was defined to consist of at

least 50 cells (198). BGB324 significantly reduced colony formation at 0.75 µM

in UP007 cells (p<0.0001, two-tailed student t-test) and inhibition of SNB-19

colony formation was apparent at 2 µM of BGB324 (p <0.0001, ANOVA with

Tukey multiple comparison post-hoc analysis) with complete inhibition being

achieved, without cell death, at 10 µM BGB324 (Figure 4.7).

Figure 4.7: Effect of BGB324 on long-term growth of GBM cell colonies in 3D. Representative micrographs of SNB-19 colonies in soft agar at 1.5x magnification (above; i and ii) and 15x magnification (below; iii and iv) untreated (left) or treated with 10 µM BGB324 (right). Histograms show average colony counts of 5 different fields per treatment for SNB-19 and UP007. ANOVA with Tukey multiple comparison post-hoc analysis for SNB-19 cells and two tailed student t-test for UP007 cells. Data are mean ±SEM (N=3 separate experiments); ****p<0.0001 versus untreated.

102

Due to the significant decrease in the number of colonies formed in the

presence of BGB324 and the observation of single viable cells being present

after two weeks, the pro-apoptotic vs cytotoxic effect of BGB324 was evaluated

by Annexin V/PI staining with flow cytometry. Annexin V binds to

phosphatidylserine (PS) which is exposed on the outer membrane of apoptotic

cells, whilst propidium iodide (PI) stains dead cells. This combination of staining

is widely used to determine the proportion of cells within a population that are

viable, apoptotic, or necrotic (199). No significant increase in the number of

apoptotic or necrotic cells was observed in both cell lines following 24-hour

treatment with IC50 and double-IC50 concentrations of BGB324 (Figure 4.9,

p<0.5413 and p<0.6359 for UP007 and SNB-19 cells respectively, Two-way

ANOVA with Tukey multiple comparison post-hoc analysis). Therefore, the

functional cell assays revealed that SMI inhibition of Axl at non-toxic

concentrations specifically affected the functional parameters measured, and

that these functions are therefore at least in part mediated by Axl. Indeed, a

closer look at the cells through tracking under live cell imaging revealed an

impairment in cell division upon treatment with BGB324 rather than apoptosis

(Figure 4.8).

Figure 4.8: BGB324 effect on cell division. Effect of BGB324 on GBM cell division shown by representative pictures for SNB-19 cells taken every 30 minutes when untreated and treated with 2 µM BGB324. Arrows mark cell of interest, which is undergoing division over a period of 3 hours (N=3 independent experiments) at 10X magnification.

103

Figure 4.9: Representative propidium iodide (PI) vs Annexin V fluorescence intensity in SNB-19 cells treated with vehicle, 2.5 µM and 5 µM BGB324. Quadrants and markers in the displayed plots were used to demarcate the various cell populations (top panel). Quantified, comparative cell counts (%) are shown for healthy, apoptotic and necrotic cells following 24-hour BGB324 treatment of SNB-19 and UP007 cells (bottom panel). Data are mean±SEM (N=3 separate experiments); ANOVA with Tukey multiple comparison post-hoc analysis ns, not significant versus untreated.

Temozolomide (TMZ) is a chemotherapeutic adjuvant to radiotherapy

and is currently the only mildly successful chemotherapeutic agent for GBM,

extending median survival by 3-4 months (200). Therefore, a possible

combinatorial effect of Axl inhibition (by BGB324) and TMZ in GBM cells was

explored. A concentration killing curve was initially conducted to identify sub-

maximal TMZ concentrations to be used with BGB324 (Figure 4.10, p<0.0001

for both cell lines, ANOVA with Tukey multiple comparison post-hoc analysis).

Treatment with TMZ resulted in concentration-dependent decreases in viability

104

of both UP007 and SNB-19 cells. From the killing curves, the IC50 for TMZ was

calculated to be 24 µM for SNB-19 and 32 µM for UP007 (Figure 4.10).

Figure 4.10: Effect of temozolomide (TMZ) on viability of SNB-19 and UP007 cell lines, Data are mean±SEM (N=3 separate experiments); ANOVA with Tukey multiple comparison post-hoc analysis ****p<0.0001, ***p<0.001 versus untreated (UT).

The concentrations of 10, 25 and 50 µM of TMZ were then used in

conjunction with 1 µM BGB324. The inhibitory effect of BGB324 at 1 µM was

significantly enhanced by co-incubation with 10 µM of TMZ, versus either agent

alone. Combinations of BGB324 with 50 µM TMZ did not significantly add to this

effect, possibly due to the increased toxicity of TMZ at such high a concentration

(Figure 4.11, p<0.0001, ANOVA with Tukey multiple comparison post-hoc

analysis). However, all concentrations of BGB324 exhibited enhanced inhibitory

effects when combined with sub-toxic doses 10 µM and 25 µM of TMZ (Figure

4.11, p<0.0001, ANOVA with Tukey multiple comparison post-hoc analysis).

105

Figure 4.11: Effect of BGB324 on GBM cell growth in combination with temozolomide (TMZ). MTS assay showing cell growth/survival of SNB-19 and UP007 cells treated with varying concentrations of TMZ alone or in combination with BGB324. Data are mean±SEM (N=3 separate experiments); ANOVA with Tukey multiple comparison post-hoc analysis, ****p<0.0001, ***p<0.001 versus untreated (UT). # indicates significance when compared to 10 µM of TMZ.

106

4.1.7 Inhibition of GBM cell migration and invasion by Axl small

molecule inhibition

One of the biggest challenges in GBM treatment is the highly invasive

nature of the tumours, which leads to multiple relapses. Therefore, the

effectiveness of BGB324 against the migration, motility and invasive capacity of

the GBM cells was tested. BGB324 inhibited cell migration in both SNB-19 and

UP007 cells in a concentration-dependent manner (Figure 4.12A, p<0.0086 and

p<0.0001 respectively, ANOVA with Tukey multiple comparison post-hoc

analysis). Moreover, cell tracking experiments with live cell imaging, where

single cells were monitored for 3 hours and movement was tracked, revealed

that sub-IC50 doses of BGB324 profoundly inhibited the motility of cells from

both lines, affecting both average velocity of movement and total distance

travelled (Figure 4.12B, p<0.0001 for both cell lines, unpaired two tailed

student t-test). This indicates that BGB324 can effectively block cell motility and

migration at doses below those that can inhibit cell growth and viability. In

addition, Axl inhibition with BGB324 also abolished invasion of both GBM cell

lines through extracellular matrix, when stimulated with PDGF-AA as glioma cell

chemoattractant (Figure 4.12C, p<0.0020 and p<0.0006 for SNB-19 and UP007

cells respectively, ANOVA with Tukey multiple comparison post-hoc analysis).

In addition, Gas6, when itself used as a chemoattractant, also significantly

increased cell invasion in both cell lines, an effect that was inhibited by BGB324

(Figure 4.13A, p<0.0510 and p<0.0858 for SNB-19 and UP007 respectively,

ANOVA with Tukey multiple comparison post-hoc analysis). Also, Gas6

stimulation was able to increase cell migration in SNB-19 cells whereas it has

no effect on UP007 cells (Figure 4.13B).

107

Figure 4.12: Effect of BGB324 on GBM cell migration, motility and matrix invasion. A. Representative images from scratch wound assay for SNB-19 cells at 0 and 21 hours (scale bar=390 µm) followed by graphical representation of migration rate (area change per hour) for SNB-19 and UP007 cells treated with varying concentrations of BGB324. B. Cell tracking experiments yielded data for average total distance travelled and average velocity for SNB-19 and UP007 cells following treatment with BGB324. C. Invasion assay for SNB-19 (N=3) and UP007 (N=4) cells following treatment with BGB324 using PDGF-AA as chemoattractant. Data are mean ±SEM (N=3 independent experiments); ANOVA with Tukey multiple comparison post-hoc analysis, ***p<0.001, **p<0.01, * p<0.05 compared to untreated (UT) or as indicated.

108

Figure 4.13: A. Invasion assay for SNB-19 and UP007 cells following treatment with BGB324 using Gas6 as chemoattractant. B. Effect of Gas6 on GBM cell migration shown by graphical representation of migration rate (area change per h) for SNB-19 and UP007 cells. Data are mean ±SEM (N=3 independent experiments); ANOVA with Tukey multiple comparison post-hoc analysis (A) and unpaired two tailed student t-test (B), ***p<0.001, **p<0.01, * p<0.05

109

4.2 Discussion

4.2.1 Establishment of stable Axl knockdown cell lines

Establishment of stable Axl knockdown was performed via plasmid

insertion into the cell nucleus by transfection. This process comprises the

formation of a complex between the nucleic acids of interest and a positively

charged lipidic transfection reagent, which then interacts with heparan sulphate

proteoglycans on the cell membrane. This leads to endocytosis of the

transfection complex and its sequestration into intracellular vesicles, which

facilitates entry of the DNA plasmid into the nucleus, leading to permanent

genome integration (201) (Figure 4.14).

Figure 4.14: Diagrammatic summary of lipid mediated transfection mechanism (202) Briefly, the DNA plasmid complexes with the lipofection reagent which in turn enters the cell membrane via endocytosis.

Even though cationic lipid transfection remains the most popular

transfection vehicle option for many scientists, this process results in low

transfection efficiency due to being dependent on nucleic acid/chemical ratios,

plasmid size, solution pH, and cell membrane conditions (203). Moreover, it is

110

time consuming and lengthy when aiming for permanent DNA plasmid

integration. Ideally, stable knockdown is achieved rapidly with a higher

efficiency using a lentiviral delivery system. A virus is produced with virulence

genes env, vif, vpr, vpu, and nef deleted, containing the plasmid of interest,

which is then used to infect the cells and integrate the plasmid into the genome

(204) (Figure 4.15).

Figure 4.15: Simplified schematic representation of viral shRNA delivery. Diagram adapted from (http://www.biocat.com/genomics/lentivirus-product-line).

Due to the nature of this technique, a Class II, Type A2 Biological Safety

Cabinet is required, as well as extra precautions which are not available at the

University of Portsmouth. Therefore, the cationic lipid transfection method was

the best option given the facilities behind this study. A major potential caution

to this approach came from the fact that Axl expression appeared to be restored

http://www.biocat.com/genomics/lentivirus-product-line

111

in knockdown cell clones after a period of roughly two months, thus limiting the

number of experiments that could be performed. Given that the whole process

of transfection, selection, clone picking and expansion as well as Axl screening

takes about 4-5 months, no more new stable clones were established following

detection of Axl re-expression due to time constraints. Instead, short term

experiments requiring Axl knockdown were performed using transient

transfection with siRNA pools. Nonetheless, shAxl knockdown clones were still

successfully established for SNB-19 cells with knockdown levels of 40, 70, 90

and 80% respectively whilst 80% Axl knockdown was achieved for UP007 cells

(Figure 4.2); this enabled us to study key biological functions regulated by Axl.

4.2.2 Genetic Axl knockdown inhibits proliferation and cell migration

in GBM cells

In the first report of Axl overexpression in GBM, one of the first

experimental observations was that Axl knockdown by either RNAi or by

dominant negative inhibition resulted in decreased proliferation, cell migration

and invasion (138, 140), thus making Axl an attractive target for therapy.

In this study, the MTS assay was used to evaluate the effect of stable

Axl knockdown on cell growth/viability. Notably, UP007 shAxl clone viability was

drastically reduced compared to control and empty vector clones; this effect

was apparent already during the clone selection process, as all but one of the

UP007 shAxl cells clones were not viable. In contrast, although the SNB-19

shAxl clones exhibited a trend towards reduced viability, quantitative analysis

however revealed no statistically significant changes (Figure 4.3). This

phenomenon could either be due to sufficient levels of Axl remaining in those

112

clones that could adequately support viability as normal, or the presence of

MerTK in SNB-19, which could compensate for the loss of Axl. A double Axl-

MerTK knockdown would thus reveal the role of MerTK in growth/viability of

the SNB-19 cell line. Cell cycle analysis would also provide more information on

whether Axl is involved in viability or proliferation of these cells and how

removal of Axl affects the cell cycle stages. Additionally, soft agar assays would

provide information on the ability of these cells to expand clonally after Axl

removal and potentiate the significance of Axl in GBM tumourigenesity.

In accordance with previous studies, Axl knockdown significantly

reduced cell migration as observed by wound healing assays in both cell lines

(Figure 4.4). In SNB-19 cells, reduced migration was observed only where 90%

knockdown was achieved, indicating that levels of Axl expression below 90%

are sufficient to counteract this anti-migratory effect of Axl knockdown. The

effect of Axl knockdown on cell invasion was not studied with the shAxl clones

due to restoration of Axl expression in the stable knockdown clones at the time

the experiments were due to be performed. Ideally, the invasion abilities of Axl

knockdown clones should be investigated with the use of a modified Boyden

chamber assay, where cells move through pores and enter into a

fibronectin/matrigel matrix. Following confirmation of invasion impairment, an

Axl plasmid transfection should rescue the phenotype, thus establishing Axl as

a major regulator of glioma invasion.

4.2.3 Axl inhibition by a small molecule inhibitor

The focus of this study was to evaluate the effect of two small molecule

inhibitors, BGB324 and BMS777607, as novel treatments for glioma, with a

113

focus on the more Axl-specific BGB324. BGB324 (Figure 4.16A) is a novel Type

I, ATP-competitive Axl SMI, with more than 40-fold Axl selectivity compared to

MerTK and Tyro3 and 100 fold to other kinases (205). Type I inhibitors bind the

DFG-in conformation with a U-shaped geometry as shown in Figure 4.16B.

Figure 4.16: Schematic representation of inhibitor binding. a) normal substrate binding, b) substrate binding inhibition by ATP competitive inhibitor, c) substrate binding inhibition by attachment of ATP non-competitive inhibitor to alternative pocket (206).

BGB324 has been shown to effectively block Axl-dependent events,

including downstream Akt phosphorylation and invasiveness of breast cancer

cells (160). BMS777607 (Figure 4.17B), an originally designed Type II MET

inhibitor with 3-fold higher affinity for Axl compared to MET binds the ATP

inactivated DFG-out conformation of kinases (205, 207). Even though Type II

inhibitors are suggested to be more selective due to their ability to adopt an

extended conformation via interactions with DFG residues of the activation loop

to enter an allosteric region (adjacent to the ATP binding site) that is only

accessible in the inactive DFG-out conformation of the kinase (205) (Figure

4.16C), both in our hands and others, BGB324 has been shown to be more

114

selective for Axl than BMS777607 (146, 160). Of course the RTK activity assays

performed with the inhibitors in vitro do not correspond to inhibitor specificity

or affinity. Whilst BMS777607 (1.1 nM) performed better than BGB324 (14 nM)

in cell free in vitro assays (205) it appears to be more specific and has higher

affinity.

Figure 4.17: A. Chemical structure of Type I Axl inhibitor BGB324. B. Chemical structure of Type II Axl inhibitor BMS777607 (205)

First, the effect of BGB324 and BMS777607 on viability of two GBM cell

lines was investigated by MTS assay. The IC50 for BMS777607 in the MTS assay

was calculated to be 118 µM for SNB-19 and 35 µM for UP007 cells (Figure

4.5C-D). These values are at least tenfold higher than the IC50 values for

BGB324, which are 2.5 µM and 1 µM for SNB-19 and UP007 cells respectively

(Figure 4.5A-B). Other studies validate this difference in inhibitory potency

between the two molecules measured by us. Holland et al. determined the IC50

for BGB324 to be 14 nM in in vitro biochemical kinase assays using recombinant

Axl protein (160), around a 100-fold more potent in this system than

BMS777607 with its IC50 of 12.5 µM (146). Also, in vivo studies showed that

BGB324 successfully reduced tumour growth in breast cancer at dosages

ranging from 7–25 mg/kg twice daily (160) whilst BMS777607 reduced

115

glioblastoma tumour growth in mice at doses ranging from 30–100 mg/kg twice

daily (146). Even though both inhibitors can effectively inhibit Axl, BGB324

remains the most selective SMI as studies have shown 100-fold greater

selectivity for Axl against other kinases, whilst BMS777607 has also been shown

to inhibit Met with an IC50 of 3.9 nM (120). Taking these data collectively,

BGB324 was the preferred SMI for the rest of the study, as it has shown

promising results in breast and lung cancer clinical trials (160, 208) and it had

not been studied in glioma previously.

The results from the experiments on BGB324 and cell viability mirrored

those with stable knockdown of Axl, showing the UP007 cell line to be about

two-fold more sensitive to Axl inhibition than SNB-19 cells. Following this, the

inhibitory effect of the SMI on the activity of the Axl molecule was investigated

by Western blot (Figure 4.6). Pre-incubation with BGB324 reduced Axl activation

by Gas6 in both cell lines as well as downstream target Akt. Moreover, Tyro3

showed a trend towards a significant decrease in phosphorylation in the UP007

cell line indicating a possible interaction between Axl and Tyro3. This interaction

is a focus of study in greater depth in Chapter 6.

4.2.4 The Axl SMI BGB324 inhibits GBM cell growth and colony

formation

The significance of Axl in glioma was first shown by the introduction of

a DN form of Axl in glioma cells, which reduced cell proliferation by 30%

compared to wildtype (138). Furthermore, Axl shRNA knockdown significantly

decreased astrocytoma cell proliferation in soft agar (140). In agreement with

previous studies, treatment with BGB324 significantly impaired the ability of

116

SNB-19 and UP007 cells to form colonies in soft agar assays in a dose-

dependent manner. Further analysis by Annexin V/PI staining and quantification

by flow cytometry revealed that this effect was not due to cell death caused by

cytotoxicity, as doses of up to double the IC50 concentrations over a period of

24 hours did not increase the number of apoptotic or necrotic cells but impaired

the ability of cells to divide (Figure 4.8 and 4.9).

Even though several studies report Axl knockdown or blockade to reduce

cell growth in a variety of cancers including ovarian (209) and pancreatic (119)

cancers, the mechanisms behind this phenomenon remain to be determined.

Prior to this study, only one report had shown that pharmacologic Axl inhibition

or siRNA targeting inhibits cell growth and induces cell-cycle arrest and

apoptosis. This study was in acute myeloid leukaemia (AML), where Axl

blockade inhibited FLT3 RTK hyperactivity brought about as a result of internal

tandem duplication (210). This finding indicated an important role for Axl in

regulation of the cell cycle in AML. To explore this role further, a comprehensive

study can be carried out on the cytoskeletal changes occurring during cell

division, e.g. by fluorescent microscopy, as well as investigation of genes

altered during the cell cycle by qRT-PCR following Axl inhibition to pinpoint the

exact mechanism. Moreover, each cell cycle phase should be analysed

separately in order to identify during which specific phase Axl is most involved.

Temozolomide (TMZ) is currently the only mildly successful

chemotherapeutic agent for GBM, extending median survival by 3-4 months

(200). In our study, we also observed an increase in sensitivity to TMZ when

Axl was concomitantly blocked with BGB324. From the killing curves, the IC50

for TMZ was calculated to be 24 µM for SNB-19 and 32 µM for UP007 (Figure

117

4.10). When used in conjunction with 1 µM BGB324, a significant decrease was

observed with as low as 10 µM of TMZ; this indicates a possible synergy

between the two treatments (Figure 4.11). In agreement with our results, a

study by Keating and colleagues showed Axl knockdown in astrocytoma cells to

significantly reduce the IC50 of TMZ from 68.6 μmol/L to a range of 1 to 18.2

μmol/L (140). However, ideally, TMZ-resistant cell lines should be used to

evaluate accurately the effect of Axl knockdown to TMZ chemosensitivity.

4.2.5 BGB324 inhibits cell migration and invasion in GBM

Axl has been established as a major regulator of invasion in glioma. In

three different studies, Axl knockdown (137, 140), introduction of a dominant

negative Axl form (138) and pharmacological inhibition with the less potent Axl

SMI BMS777607 (146) resulted in reduced GBM cell invasion. More specifically,

in a study by Vajkoczy et al., introduction of Axl-DN cells into nude mice

correlated with in vitro experiments, showing reduced tumour growth compared

to control cells without affecting the tumour vasculature but instead by

decreasing invasive potential (138). Detailed assessment of the effect of Axl on

cell migration revealed that Axl-DN cells have a reduced locomotor activity and

disrupted cell-cell adhesion (138). At the molecular level, Axl has been

associated with BMI1, a master regulator of NPCs stemness and self-renewal.

BMI1 is often found overexpressed in brain cancers (141). Recently, Gas6 was

shown to be a novel target gene for BMI1 although the exact mechanism

remains elusive (142). BMI1 belongs to the multiprotein complex PRC1. The

PRC1 complex in cooperation with the initiation complex EZH2 mediate

methylation of histone proteins to induce gene silencing (143). Axl has been

118

shown to be directly regulated by EZH2 in glioma, where EZH2 was shown to

be up-regulated and co-expressed with Axl in glioblastoma cells, furthermore,

EZH2 appeared to positively regulate Axl expression independently of changes

in methylation (144). In addition, knockdown of EZH2 in glioblastoma

significantly reduced the invasive character of the cells and this was shown to

be mediated by Axl. The authors also point out that EZH2 is required during

EMT for E-Cadherin repression by Snail. Axl has been extensively associated

with Snail-mediated EMT in breast cancer; this therefore could represent

another potential link between EZH2 and Axl (145).

In agreement with previous studies we have shown that inhibition of Axl

either by knockdown or inhibition with BGB324 significantly reduces cell

migration as observed in scratch wound assays in a dose-dependent manner.

Moreover, by cell tracking under live cell imaging microscopy, the calculated

total distance travelled and the velocity of cell migration were shown to be

reduced significantly at sub-IC50 concentrations (MTS assay) for SNB-19 and at

the IC50 concentration (MTS assay) for UP007 cells. Also, cell invasion

stimulated by PDGF-AA as chemoattractant was also reduced to basal levels by

BGB324 in both cell lines (Figure 4.12), indicating that Axl mediates the invasive

character of GBM cells in response to diverse humoral cues. Indeed, studies

have shown that Axl mediates tumour invasion and chemoresistance through

the activation of the PI3K/Akt pathway in breast cancer (211, 212) as well as

ovarian cancer (187). The BGB324 inhibition of GBM cell migration and invasion

we observed could be the result of suppressed Akt signalling. Indeed, Akt has

been shown to enhance MMP2 in mammary epithelial cells, MMP9 production

via NF-κB in fibrosarcoma cells and fibroblast motility by phosphorylating Girdin,

119

an actin-binding protein that promotes stress fibre formation and lamellipodia

(213). The exact mechanism present in glioma requires further investigation.

In addition, Gas6 was also an effective direct chemoattractant for SNB-19 cells,

which likely was mediated via Axl as these cells are more responsive to Axl

stimulation by Gas6 than UP007. However, in slight contrast, when itself used

as chemoattractant, Gas6 significantly increased cell invasion of both SNB-19

and UP007 cell lines (Figure 4.13). Furthermore, Axl inhibition by BGB324

inhibited invasion only marginally, perhaps indicating that Gas6 may promote

invasion only partly via Axl kinase.

In conclusion, the data in this chapter indicate that targeting Axl for

inhibition by a specific SMI such as BGB324 has robust anti-tumour effects on

GBM cells and could potentially be effective in restraining tumour invasion in

vivo and, in combination with other agents, could effectively prolong glioma

patient survival. Further studies will be needed to explore the anti-GBM tumour

efficacy of Axl SMI BGB324 in vivo and ultimately in glioma patients. However,

despite the promising initial signs of success with BGB324 in other cancers, it

as yet unknown whether the drug can cross the blood-brain barrier, which

represents one of the greatest challenges in drug delivery to the brain (214).

This question can be answered with the aid of animal models. Specifically,

glioma patient biopsies can be orthotopically injected in the brain of

immunocompromised nude rats and allowed to grow before injecting BGB324

intraperitoneally. Depending on tumour volume of treatment compared to

vehicle control and positive controls, it can be demonstrated if BGB324 is

suitable for treating glioma patients. Moreover, by using a fluorescent tagged

120

version of the drug, following rat sacrifice, immunostaining can reveal blood

brain carrier permeability to BGB324.

4.2.6 Summary

Two cultured adult GBM cell lines, SNB-19 and UP007, were treated with

the novel Axl small molecule inhibitor BGB324, and analysed in assays for

survival, 3D colony growth, motility, migration and invasion. Western blot was

used to detect protein expression and signal protein phosphorylation. In both

cell lines, BGB324 inhibited specifically phosphorylation of Axl as well as Akt

kinase further downstream. BGB324 also inhibited survival and proliferation of

both cell lines in a concentration-dependent manner, as well as completely

suppressing migration and invasion. In conclusion, small molecule inhibitor-led

targeting of Axl may be a promising therapy for GBM progression.

121

Chapter 5

EGFR heterodimerises with Axl

to promote GBM cell invasion

122

5.0 Introduction

Acquired resistance to conventional and targeted therapies is becoming

a major hindrance in cancer management. It is increasingly clear that cancer

cells are able to evolve and rewire canonical signalling pathways to their

advantage, thus evading cell death and promoting cell invasion, amongst other

cancer hallmarks (3). Overexpression of Axl RTK has been correlated with poor

prognosis in patients, promotion of increased cancer cell invasiveness,

epithelial-to-mesenchymal transition (EMT) and chemoresistance in both breast

and lung cancers (215, 216), as well as with poor overall survival of GBM

patients and increased cell invasiveness (137). In addition, in the previous

chapter, inhibition of Axl with a small molecule inhibitor was shown to hinder

glioma cell survival, migration and invasion (217), thus highlighting the viability

of targeting Axl specifically in cancer therapy.

In addition to its principal role in invasion, numerous studies suggest that

Axl overexpression mediates secondary resistance to both conventional and

targeted therapies. The most noteworthy observation to date is acquired

resistance to erlotinib (a reversible tyrosine kinase inhibitor) in EGFR-mutated

lung adenocarcinoma by induction of an EMT-like phenotype through Axl

activation both in vitro and in vivo (218). Similarly, in another study Axl

overexpression correlated with acquired resistance to gefitinib (EGFR SMI)

(219). Recently, expression of Axl has been identified as a predictor of lack of

response to the EGFR–targeted inhibitors lapatinib and erlotinib in triple-

negative breast cancer cells (220). Interestingly, it has been found that EGFR

signalling trans-activates Axl and that this ligand-independent Axl activity,

through possible hetero-interaction amongst different RTKs, could then

123

diversify downstream signalling pathways beyond those triggered by EGFR

alone. Activation of EGFR by EGF stimulation of MDA-MB-231 breast cancer cells

was shown to lead to both Axl and MET phosphorylation but not vice versa

(220). The EGFR-mediated Axl activation led to widespread downstream

signalling changes that were blocked by Axl siRNA, whilst Gas6 stimulation had

lesser effects. This pivotal study therefore suggests that Axl can serve as an

alternative conduit for EGFR signalling, at least when it comes to Akt activation

(220).

In the present study, results are presented that show the identification

of a novel protein-protein interaction between EGFR and Axl. This hetero-

interaction involves Axl as a gateway for EGFR to activate invasive signalling

pathways in GBM cells, which are known for their highly invasive and

proliferative phenotype (200), and which overexpress both RTKs (137) (221).

We show that EGF, via EGFR, indirectly activates invasive signalling via Axl in

human cancer cells in a unidirectional manner, through a protein-protein

interaction between EGFR and Axl. The EGFR-Axl hetero-interaction promotes

cell invasion through the TIMP1-MMP9 invasion regulatory axis whilst also

inhibiting cell proliferation. This phenomenon does not occur via EGFR signalling

alone but exclusively via the EGFR-Axl hetero-interaction. These findings reveal

a greater diversity to the various functions of RTKs than previously thought and,

in cancer, have implications for guiding molecular targeted therapies as well as

for managing secondary drug resistance.

124

5.1 Results

5.1.1 EGF activates Axl via EGFR in SNB-19 cells

In order to examine the relationship between EGFR and Axl in GBM cells,

Axl activation was investigated; this was done through analysing residue-

specific phosphorylation in serum-starved SNB-19 cells in response to

stimulation by EGF over a time course of 0-60 minutes, with PDGF and FGF as

controls. EGF, in addition to causing a rapid increase in EGFR phosphorylation

in under 2 minutes, also rapidly stimulated Axl phosphorylation within the same

time period (Figure 5.1A). Interestingly, EGF stimulated phosphorylation of only

the mature fully glycosylated 140 kDa Axl but not the partially glycosylated 120

kDa form (both forms are activated by Gas6) (Figure 5.1A). The same EGF-

stimulated phosphorylation of Axl was also observed in the UP007 cell line;

however, this effect was not unique to cells of this cancer type, but was also

observed in human breast and head and neck cancer cell lines (Figure 5.1C-D).

In contrast to the effects of EGF, no significant change in Axl phosphorylation

occurred after treatment with the other distinct RTK ligands PDGF or FGF

(Figure 5.1B). In addition, a similar response to EGF stimulation was observed

for the Tyro3 receptor, with EGF causing a rapid phosphorylation of Tyro3 in

under 2 minutes, the same time period as for EGFR and Axl (5.1A). Also, neither

PDGF nor FGF had such an effect on Tyro3, and moreover, as with Axl, the

effect was observed in cells from several different cancer types (Figure 5.1B).

125

Figure 5.1: EGFR activates Axl. A. Western blot of time-course of EGFR, Axl and Tyro3 phosphorylation by EGF (50 ng/ml) (N=3) in SNB-19 cells. B. Axl (N=3 blots) and Tyro3 (N=2 blots) phosphorylation by PDGF (10ng/ml) and FGF (10 ng/ml) in SNB-19 cells. C. Western blot of time-course of Axl and Tyro3 by EGF (50 ng/ml) in UP007 (N=1). D. Western blot of Axl and Tyro3 stimulated with Gas6 (400 ng/ml) and EGF (50 ng/ml) for 10 minutes in breast cancer cells (MDA-MB-231) and head and neck cancer cells (SCC-25 and Detroit) (N=1 blot).

To probe the specificity of the kinase-substrate relationship, pre-

treatment with the highly selective Axl SMI BGB324, which blocks the kinase

activity of Axl (15, 21), did not affect the phosphorylation stimulated by EGF

(15), whereas pre-treatment with the EGFR-specific SMI gefitinib blocked Axl

phosphorylation following EGF treatment. EGFR in cells was rapidly activated by

EGF stimulation within 15 minutes, an effect that was abolished by gefitinib,

but which was not affected by pre-incubation with the Axl-specific BGB324

(Figure 5.2A, Western blot semi-quantification for both SNB-19 and UP007 cell

126

is shown in appendix 3). Therefore, Axl activation in response to EGF is specific

and relies on EGFR activity and not Axl activity, suggesting a direct

phosphorylation of Axl by EGFR. To investigate whether this cross

phosphorylation observed in SNB-19 cells occurs in both directions, we tested

EGFR phosphorylation following stimulation by Gas6, the natural ligand for Axl.

Gas6 had no effect on EGFR phosphorylation, further indicating a unidirectional

relationship between EGFR and Axl (Figure 5.2B). Additionally, Western blotting

of pERK and pAkt, two well-known downstream targets of RTK signalling

pathways, showed no change in phosphorylation upon EGF stimulation in the

presence of the Axl inhibitor BGB324, indicating that the two intracellular

kinases are regulated predominantly by EGFR signalling in this context (Figure

5.2C).

Figure 5.2: A. Western blot of Axl and EGFR phosphorylation by EGF and its inhibition by gefitinib and BGB324. B. Western blot of EGFR phosphorylation by EGF and Gas6 and its inhibition by gefitinib. C. Western blot of pERK and pAkt levels after EGF stimulation and influence of gefitinib and BGB324. Each blot is representative of three separate blots carried out for that experiment.

127

5.1.2 EGFR kinase directly activates Axl kinase in vitro

Having observed that EGF stimulation leads to native Axl phosphorylation

in living cells, we investigated whether the observed unidirectional

transactivation of Axl by EGFR occurs through a physical interaction between

EGFR and Axl. Indeed, EGFR protein co-immunoprecipitated (co-IPed) with Axl

protein even in the absence of EGFR stimulation by EGF (Figure 5.3).

Figure 5.3: Hetero-interaction of EGFR and Axl RTKs. Co-IP of Axl and EGFR, followed by Western blot probing for presence in the complexes of Axl (left blot) and EGFR (right blot). Control co-IP antibodies were anti-GAPDH (Ctrl Ig1).

The potential of EGFR to activate Axl in a cell-free in vitro system using

Kinase-Glo Luminescent Kinase Assay (Promega) was then investigated (222).

Both recombinant active Axl and EGFR kinases were able to phosphorylate the

Axl substrate peptide containing Y779, a residue that is a well-known docking

site in Axl for multiple signalling proteins and marker of Axl activation (22).

Furthermore, the in vitro phosphorylation of Axl Y779 peptide by Axl and EGFR

kinases was inhibited by BGB324 and gefinitib respectively (Figure 5.4A,

p<0.0001 and p<0.0003 for EGFR and Axl respectively, ANOVA with Tukey

multiple comparison post-hoc analysis).

128

In addition, the influence of EGF-EGFR on kinase activity of native Axl in

cells was investigated. Axl that was IPed from SNB-19 cells following 5-minute

incubation with EGF exhibited a greater in vitro activity, as reflected by

phosphorylation of the Axl Y779 peptide. Therefore, Axl is a direct substrate of

EGFR kinase in vitro, as well as in vivo in intact cells following activation of EGFR

by EGF (Figure 5.4B, p<0.0268, ANOVA with Tukey multiple comparison post-

hoc analysis).

Figure 5.4: A. In vitro kinase activity of recombinant Axl and EGFR kinases, with Axl Y779 peptide as substrate. Kinases were tested either alone or in the presence of the specific Axl or EGFR inhibitors, BGB324 (10µM) and gefitinib (10 µM), respectively. Values are normalised against background Y779 peptide luminescence. B. In vitro kinase activity of native Axl immunoprecipitated from SNB-19 cells after EGF stimulation, with or without 1 hour gefinitib (10 µM) pre-treatment. Sec15B and GAPDH IPs following EGF stimulation were used as negative control pulldowns (data not shown). Data are mean±SEM (N=3 separate experiments); ANOVA with Tukey multiple comparison post-hoc analysis, ****p<0.0001, **p<0.01, *p<0.05 versus untreated.

129

5.1.3 EGFR induces upregulation of pro-invasive genes via Axl

The observed hetero-interaction between EGFR and Axl prompted us to

explore the potential discrete downstream signalling pathways that emanated

from EGFR either via Axl or not. Therefore, SNB-19 cells were treated with EGF

to activate the receptors, with some treatments also including BGB324 to block

Axl kinase, thus concentrating the effect of EGF towards the EGFR receptor

only. Axl has been shown to promote invasion as well as chemoresistance to

EGFR SMIs through the induction of EMT genes (216, 218, 220, 223).

Therefore, a qRT-PCR analysis of the expression of 15 EMT-related genes

(CD44, CD24, CD133, ALDH1, CCND1, CDK4, KRT19, NOTCH1, TWIST, MMP9,

AKT2, TIMP1, AKT1, MMP2, PI3K) was carried out. Out of these, we observed

a significant change in the mRNA levels of CCND1, TIMP1, CD44, MMP2 and

MMP9 genes (Figure 5). TIMP1 and CD44 mRNA levels were significantly

increased when Axl kinase was blocked during EGF stimulation (Figure 5.5A.

p<0.0029 and D, p<0.0040, ANOVA with Tukey multiple comparison post-hoc

analysis), whereas CCND1 mRNA was significantly increased by EGF stimulation

irrespective of Axl kinase activity (Figure 5.5A, p<0.0028, ANOVA with Tukey

multiple comparison post-hoc analysis). Interestingly, EGF stimulation increased

MMP9 mRNA levels, whilst concomitant Axl inhibition by BGB324 blocked this,

despite presence of EGF (Figure 5.5B, p<0.0165, ANOVA with Tukey multiple

comparison post-hoc analysis). In contrast, MMP2 was significantly reduced by

EGF stimulation only when Axl signalling was inhibited (Figure 5.5C, p<0.0352,

ANOVA with Tukey multiple comparison post-hoc analysis).

130

Figure 5.5: qRT-PCR analysis of TIMP1 (A), MMP9 (B), MMP2 (C) and CD44 (D) genes in SNB-19 cells treated for 24 hours with 50 ng/ml EGF, with or without 2 µM BGB324. Data are mean±SEM (N=3 independent experiments); ANOVA with Tukey multiple comparison post-hoc analysis, **p<0.01, *p<0.05 compared to either untreated/controls or as indicated.

Moreover, Axl stimulation activated STAT3 signalling which was inhibited

by BGB324 as seen in Figure 5.6 which can potential indicate a regulatory

mechanism for MMP9 and MMP2 expression.

Figure 5.6: Western blot showing STAT3 phosphorylation following 15 minute Gas6 stimulation followed by BGB324 inhibition in UP007 and SNB-19 cell lines. (N=1)

131

5.1.4 EGFR positively regulates cell invasion signalling via Axl

In Chapter 2, Axl signalling was demonstrated to regulate the migration

and invasion of the GBM cells used in this study (15), therefore identifying genes

involved in invasion to be altered by EGFR-Axl signalling was of particular

interest. In order to functionally evaluate the importance of this interaction,

cells were treated with EGF with or without Axl kinase blockade by BGB324, or

EGFR blockade by gefitinib, and subsequently observed its effect on GBM cell

invasion, compared to untreated and the inhibitors on their own. EGFR

activation significantly increased cell invasion, an effect that was counteracted

by Axl blockade. Most importantly, the combination of EGFR stimulation with

Axl inhibition led to a greater decrease in invasion compared to Axl inhibition

alone (Figure 5.7C, p<0.0001, ANOVA with Tukey multiple comparison post-

hoc analysis). Moreover, gefitinib was able to reduce invasion back to basal

levels (Figure 5.7C). Furthermore, TIMP1 knockdown by siRNA (Figure 6.5A,

p<0.000218765, p<0.0200244, p<0.00491499, multiple t-test) was also able

to increase invasion regardless of EGFR-Axl stimulation/inhibition, indicating

that they may act via TIMP1 to regulate invasion (Figure 5.7B, p<0.001, two-

way ANOVA with Tukey multiple comparison post-hoc analysis).

132

Figure 5.7: EGFR-Axl hetero-interaction regulates the balance between gene expression for invasive and proliferative signalling in SNB-19 cells. A. qRT-PCR confirmation of TIMP1 gene knockdown in SNB-19 cells treated for 24 hours with 50 ng/ml EGF with or without 2 µM BGB324. B. Effect of siRNA TIMP1 knockdown in combination with EGF stimulation on SNB-19 cell invasion through matrix, using PDGF-AA as chemoattractant. C. Invasion assay for SNB-19 cells following treatment with EGF, BGB324 and gefinitib using PDGF-AA as chemoattractant. Data are mean±SEM (N=3 independent experiments); ANOVA with Tukey multiple comparison post-hoc analysis, ****p<0.0001, ***p<0.001, **p<0.01, * p<0.05 compared to either untreated/controls or as indicated. # indicates significance when compared to EGF treatment.

In order to further probe the mechanisms behind this behaviour, the

mRNA levels of the other altered genes under these conditions and upon TIMP1

knockdown were monitored. MMP9 mRNA levels were increased upon EGF

stimulation (p<0.0001, two-way ANOVA with Tukey multiple comparison post-

hoc analysis), an effect that was enhanced by TIMP1 knockdown (p<0.268691,

p<0.0540195, p<0.725338, multiple t-test). Interestingly, Axl kinase inhibition

blocked the EGF-induced upregulation of MMP9, irrespective of TIMP1

133

knockdown (Figure 5.8A). In contrast to MMP9, TIMP1 knockdown had no effect

on the EGF-induced changes in CD44 (p<0.860447, p<0.8284, p<0.831766,

multiple t-test) and MMP2 expression (p<0.312455, p<0.854102, p<0.953393,

multiple t-test) (Figure 5.8B-C). Therefore, these results indicate that Axl

positively regulates cancer cell invasion via the TIMP1-MMP9 axis, that EGF also

activates this pathway only via Axl, and moreover that discrete EGFR signalling

alone in fact directly counteracts Axl invasive signalling.

Figure 5.8: qRT-PCR analysis of MMP9, MMP2 and CD44 genes in SNB-19 cells with siRNA knockdown of TIMP1 over 3 days, in the presence of EGF (50 ng/ml) with or without Axl blockade by BGB324 (2 µM). Data are mean±SEM (N=3 independent experiments); two-way ANOVA with Tukey multiple comparison post-hoc analysis, followed by multiple t-test, ***p<0.001, **p<0.01, * p<0.05 compared to either untreated/controls or as indicated.

134

5.1.5 EGFR stimulates cell cycle progression independently of Axl

The observed increase in the CCND1 gene, which codes for the cell cycle

regulator cyclin D1, in response to EGF stimulation was then further

investigated. Cell cycle analysis after treatment with EGF alone or in

combination with BGB324 revealed a decrease of cells in the G1 phase of the

cell cycle with a concomitant increase in the G2/M population (Figure 5.10B).

Therefore, EGFR signalling alone stimulates cell cycle progression in SNB-19

cells in addition to attenuating Axl invasive signalling.

The effect of EGFR blockade on cell proliferation/survival was also

investigated by MTS assays. A concentration killing curve was initially conducted

to identify sub-optimal gefitinib concentrations to be used with BGB324.

Treatment with gefitinib resulted in concentration-dependent decreases in

viability of both UP007 and SNB-19 cells from 5 µM (Figure 5.9, p<0.0001 and

p<0.0016 for SNB-19 and UP007 cells respectively, ANOVA with Tukey multiple

comparison post-hoc analysis).

Figure 5.9: Effect of gefitinib on viability of SNB-19 and UP007 cell lines as shown by MTS assay. Data are mean±SEM (N=3 separate experiments); ANOVA with Tukey multiple comparison post-hoc analysis, ****p<0.0001, ***p<0.001 versus untreated (UT).

135

Figure 5.10: Quantitative RT-PCR analysis of CCND1 gene expression in SNB-19 cells altered by EGFR activation, with and without the influence of Axl. (A) qRT-PCR analysis of CCND1 gene in SNB-19 cells treated for 24 hours with 50 ng/ml EGF, with or without 2 µM BGB324. (B) Cell cycle analysis of cells treated for 24 hours with EGF alone or in EGF in combination with BGB324. Data are mean±SEM (N=3 independent experiments); ANOVA with Tukey multiple comparison post-hoc analysis ****p<0.0001, ***p<0.001, **p<0.01, * p<0.05 compared to either untreated/controls or as indicated.

136

The concentrations of 2, 5 and 10 µM of gefitinib were then used in

conjunction with 1.5 µM BGB324 for UP007 and 2 µM for SNB-19 cells. The

inhibitory effect of gefitinib at 2 and 10 µM for SNB-19 was significantly

enhanced by co-incubation with 2 µM of BGB324. For the UP007 cells, the

inhibitory effect of gefitinib at 2 µM was significantly enhanced by co-incubation

with 1.5 µM BGB324 (Figure 5.11, p<0.0001 for both cell lines, two-way ANOVA

with Tukey multiple comparison post-hoc analysis).

Figure 5.11: Effect of gefitinib on GBM cell growth in combination with BGB324. MTS assay showing cell growth/survival of SNB-19 and UP007 cells treated with varying concentrations of gefitinib alone or in combination with BGB324. Data are mean±SEM (N=3 separate experiments); ANOVA with Tukey multiple comparison post-hoc analysis **p<0.01, *p<0.05.

137

5.2 Discussion

5.2.1 EGF activates Axl via EGFR in SNB-19 cells

EGFR is a transmembrane glycoprotein that is a member of the RTK

superfamily, and is activated by EGF. EGFR has been shown to regulate cell

proliferation, differentiation, migration and survival in cancer (224). Moreover,

EGFR activation in cancer has been shown to promote resistance to

chemotherapy and radiation treatment (225).

Amplification and overexpression of the EGFR gene presents a hallmark

for GBM tumours. The most common EGFR mutant is named EGFRvIII which

results from the deletion of exons 2-7 of the EGFR gene. EGFRvIII is unable to

bind ligand and is rendered constitutively active (226). In GBM, it is usually co-

expressed with the wild type receptor. Fan and colleagues showed that EGFR

and EGFRvIII liaise in cellular transformation assays, with the combination being

more potent than either expressed alone. EGF stimulation was able to further

enhance their effect (39). Presence of the EGFRvIII mutation has been reported

in breast and lung cancers, but several of these reports remain controversial

due to technical problems with EGFRvIII detection. These are discussed in a

detailed review by Hui and colleagues, who point out that reports of EGFRvIII

expression in other cancers such as ovarian, prostate, breast and lung are

inconsistent with studies reporting no expression whilst others report low levels

of expression compared to GBM, in which it is most highly expressed (227). The

EGFRvIII mutant has been closely associated with MET RTK and cross-activation

has been linked to increased chemoresistance (228). In the same study, they

report that increased Axl phosphorylation was observed in EGFRvIII cells by

tandem liquid chromatography MS (228), although no other reports of

138

interaction between the two receptors have been published. Additionally,

EGFRvIII was found to be co-expressed with ELMO1 and Dock180, two proteins

that interact to activate Rac GTPase to promote invasion in glioma (229, 230).

Moreover, Axl was recently found to be a direct binding partner of ELMO and

to promote cell proliferation and invasion through its phosphorylation (231).

These findings imply an indirect interaction between EGFRvIII and Axl to regulate

glioma cell invasion. The exact relationship between EGFRvIII and Axl remains

to be established; however, it represents a new avenue of research that could

potentially produce a more beneficial treatment for patients harbouring the

mutation.

Nonetheless, overexpression of wildtype EGFR has been widely

associated with Axl overexpression, and several studies have demonstrated Axl

transactivation by EGFR to promote an EMT phenotype which leads to

resistance to therapy (192). However, even though the relationship is now

increasingly indicated as a mediator of chemoresistance, a direct physical

interaction between the two receptors has hitherto not been shown. The first

part of this chapter focuses on demonstrating a physical interaction between

the receptors and therefore direct activation of Axl by EGFR.

Treatment of SNB-19 cells with EGF, and not other growth factors,

stimulated activation of Axl directly through EGFR-Axl hetero-interaction in a

unidirectional way, i.e. that EGFR activates Axl but not vice versa (Figure 5.1).

The activation of Axl was shown to occur through direct phosphorylation by

EGFR of Axl tyrosine 779, one of the key residues within Axl that serves as a

multi-substrate docking site for further downstream signalling (22) (Figure 5.1).

Moreover, EGFR stimulated activation of native Axl protein in SNB-19 cells, an

139

effect that was blocked by specific inhibition of EGFR. In agreement with our

findings, Meyer and colleagues have shown that EGF stimulation trans-activated

Axl in breast cancer cells (which we also observed here), which intensified

downstream EGFR signalling (18). However, in GBM cells, EGF stimulation and

Axl kinase inhibition had no effect on EGF-stimulated activation of the

intracellular signalling kinases Akt and ERK. This indicates that Axl cannot be

the sole mediator of the EGFR signals to these kinases in these cells. Tyrosine

779 of Axl is a major kinase target and docking site for several intracellular

signalling adaptors and kinases (2, 22). Therefore, Axl phosphorylation by EGFR

could be required for protein docking and activation of alternative downstream

signalling pathways that would not be canonically regulated by either receptor

alone through homo-dimerisation. Recently, Elkabets et al showed that Axl

activates the EGFR/PKC/mTOR axis in head and neck and oesophageal

squamous cell carcinomas, in order to arbitrate resistance to PI3Kα inhibitors

(23). However, here we did not observe any effect on EGFR phosphorylation by

Gas6-induced Axl activation. Furthermore, specific Axl kinase small molecule

inhibition did not hinder phosphorylation by EGFR (15), indicating that the event

occurs independently of Axl kinase.

Most importantly, a direct interaction between EGFR and Axl by co-IP

was shown for the first time in this study (Figure 5.3). The co-IP assay is used

to determine whether two proteins of interest are physically associated in a

protein complex. Briefly, this is achieved through lysis of cells under conditions

that preserve protein-protein interactions. The target protein is specifically IPed

from the cell extracts by binding of a specific antibody that is then immobilised

by Protein A/G beads. Any protein that is not precipitated is washed away. The

140

separate proteins present in the IP complexes are resolved by SDS-PAGE and

each protein of interest is detected by Western blotting using a specific antibody

(232). Even though co-IPs are straightforward in principle, each IP experiment

relies on the strength and stability of the particular protein association complex

under study for successful detection. In the case of EGFR-Axl pulldown, this

presented a technical challenge. A large amount of protein was required to pull

down enough protein as the interaction was apparently weak for IP signal

purposes. Ideally, the physical interaction between the receptors would be

better demonstrated as well as investigated by Förster resonance energy

transfer (FRET). FRET is a technique based on a donor fluorophore, which in an

excited electronic state transfers its energy to a nearby acceptor through non-

radiative dipole–dipole coupling. The proteins of interest are tagged and

fluorescence is captured by confocal microscope imaging and only occurs if

proteins physically interact (233). Using this method, the live interaction of

EGFR and Axl could be studied in real time. Moreover, since only the

phosphorylation of Y799 residue, which presents one of many phosphorylation

sites on Axl, was investigated, creation of specific Axl tyrosine residue mutations

would shed light on the exact mechanism of activation by EGFR. This could

easily be achieved by transfection of cells with plasmids harbouring the

mutations, followed by activation of EGFR and subsequent Western blotting of

lysates and probing for phosphor-Axl. The mutation prohibiting Axl

phosphorylation by EGFR would thus be indicated to be involved in their

interaction.

Another interesting observation was that EGFR was only able to

141

phosphorylate the fully glycosylated form of Axl. Axl is a transmembrane

receptor of molecular weight between 100 and 140kDa (Figure 5.1). Axl is

highly glycosylated and is present on the cell membrane as fully and partially

glycosylated protein species (234). The differences in signalling between Axl

glycosylated forms is still unknown. Moreover, the functional significance of

different Axl forms remains to be elucidated. One report shows that Axl

deglycosylation reduces cell proliferation and invasion in mouse hepatocellular

carcinoma cell lines (235). It is possible that differential glycosylation burden

alters protein-protein interactions to diversify signalling. Efforts during this PhD

to study the interaction between EGFR and the different glycosylated forms of

Axl were futile. Treatment with tunicamycin, which blocks N-linked glycosylation

to obtain partially glycosylated Axl forms, was unsuccessful and therefore I was

not able to perform EGFR co-IP or activation experiments. The preferential

activation of the 140 KDa Axl by EGFR could be due to the spatial conformation

of the receptors. However, a multifaceted study is required so as to probe

further this biophysical selectivity. Determination of the X-ray crystal structures

of EGFR and the alternatively glycosylated forms of Axl would enable

mathematical modelling to address this question. Moreover, a detailed analysis

of downstream signalling for each glycosylated Axl form should be performed

with microarrays to determine if downstream signalling differs depending on

glycosylation levels.

Interestingly, Tyro3 was also activated by EGFR within two minutes of

stimulation by EGF (Figure 5.1). To date, there are no reports of any interaction

between EGFR and Tyro3; hence the significance of this novel interaction

remains to be investigated. Tyro3 is highly expressed in the brain, with regions

142

of high expression including the olfactory bulbs, cerebral cortex, piriform cortex

(ventrolateral edge of the cerebral hemispheres), all cortical layers of the

cerebellum, the amygdala, hippocampus and the cerebellum (128). Even

though the function of Tyro3 in the brain is still not fully understood, reports

show that Tyro3 activation by Gas6 activates downstream ERK and PI3K

signalling pathways in cortical neurons, likely modulating strength of synaptic

connection (128). Moreover, Tyro3 was demonstrated to promote neuronal

survival through Akt phosphorylation (128) and to maintain the integrity of the

blood-brain barrier (128). Given the pro-survival abilities of Tyro3, it is

conceivable that EGFR partners up with the receptor to increase survival of

cancer cells. Furthermore, the interaction observed may not be through direct

phosphorylation of Tyro3 by EGFR but via Axl. We and others have shown that

EGFR physically interacted with and activated Axl, which may in turn activate

Tyro3 through interaction. Indeed, in a recent study, Brown and colleagues

showed Tyro3 to co-IP with Axl and enhance Axl activation by their common

ligand Gas6, while Axl was shown to phosphorylate a kinase-dead form of Tyro3

(180). These results indicate that heterodimers of Axl and Tyro3 may exist on

the cell surface, and which interact with EGFR. These findings are still

preliminary, but portray an exciting avenue for investigation.

5.2.2 EGFR induces pro-invasive genes upregulation via Axl

The EGFR-Axl interaction came into the spotlight after the discovery of

Axl overexpression as a mediator of increased chemoresistance to EGFR SMIs.

Axl has been shown to promote chemoresistance in gastrointestinal stromal

tumours (236), head and neck cancer (237), lung cancer (218, 219) and breast

143

cancer (220, 238). Several studies have proposed Axl signalling pathways to be

controlled by EGFR in order to promote EMT-mediated chemoresistance in

breast and lung cancers (220, 223, 239).

Canonically, EMT is the process by which epithelial cells lose their cell

polarity and cell-cell adhesion, and gain migratory and invasive properties to

become mesenchymal stem cells; these are multipotent stromal cells that can

differentiate into a variety of cell types. EMT is essential for numerous

developmental processes including mesoderm formation and neural tube

formation. EMT has also been shown to occur in wound healing, in organ fibrosis

and in the initiation of metastasis for cancer progression (240). However, recent

findings challenge the role of EMT in metastasis and highlight an unexpected

role in chemoresistance. Indeed, Fischer and colleagues showed that in lung

cancer, EMT induction does not promote metastasis but contributes to increase

resistance to therapy (241); this observation was also supported by similar

findings in pancreatic (242) and colorectal cancer (243).

In light of these findings, the rest of the study in this chapter focused on

investigating the significance of EGFR-Axl EMT related signalling in glioma. Five

EMT related genes were identified to be specifically regulated by Axl-EGFR

signalling but not by Axl alone. More specifically, the anti-invasive genes TIMP1,

CD44 and CCND1 genes were found to be upregulated when EGFR was

activated together with Axl kinase activity shut off (Figure 5.5). This indicates

that Axl actively suppresses their expression when activated by EGFR.

A novel observation is the regulation of expression of the matrix

metalloproteinase inhibitor TIMP1 by the EGFR-Axl hetero-complex. In order to

investigate the effect of EGFR-Axl-TIMP1 signalling on cell behaviour, invasion

144

assays where TIMP1 expression was transiently knocked down and stimulated

with EGF, or EGF in combination with BGB324 to block Axl kinase, were

performed. As expected, TIMP1 knockdown on its own resulted in a significant

increase in cell invasion. EGF stimulation in combination with BGB324 led to a

significant increase in cell invasion compared to TIMP1 siRNA alone. Moreover,

EGF stimulation led to a significant increase in invasion, an effect which was

counteracted by Axl kinase inhibition. Axl kinase inhibition in concert with EGFR

activation completely diminished any baseline invasion to a greater extent than

Axl inhibition alone. Indeed, we observed an increase in TIMP1 mRNA upon EGF

stimulation concomitantly with Axl inhibition, which can account for this

phenomenon. EGFR inhibition alone was able to restore the number of invaded

cells back to baseline, indicating that EGFR alone cannot account for the

invasive potential of these cells. With the purpose of identifying why EGFR

activation further increased cell invasion under these conditions, we also

investigated and confirmed increases in MMP2 and MMP9 genes, with no further

change in CD44. Interestingly, previous studies have established MMP9 to be

directly regulated by Axl in breast cancer (11) and MMP2 in ovarian cancer (24)

independently of TIMP1 activity. Given our results, we found that in GBM cells,

the EGFR-Axl-TIMP1 axis promotes invasion through upregulation of MMP9,

whilst Axl regulates MMP2 expression independently of EGFR or TIMP1. Our

results are supported by previous studies that reported that MMP9 binds to

TIMP1 whilst failing to bind MMP2 (25). Furthermore, Axl has been reported to

regulate MMP2 at the transcriptional level in ovarian cancer through the PI3K

pathway (24). In Chapter 1, Axl was shown to activate the PI3K pathway in

glioma and in Chapter 2 Axl inhibition was shown to suppress it (15), indicating

145

that the same regulation exists in glioma. Additionally, recent reports place

MMP2 and MMP9 under the regulation of STAT3 signalling in breast cancer

(244), a known downstream Axl effector (245). Indeed, not only Gas6

stimulates STAT3 phosphorylation in both SNB-19 and UP007 cell lines but

BGB324 effectively inhibits Gas6-mediated stimulation (Figure 5.6) indicating

another possible regulatory mechanism.

In addition, the gene for cyclin D1, CCND1, was upregulated upon EGF

stimulation, irrespective of Axl kinase activity. Coupled with this, cell cycle

analysis revealed that EGF stimulation, with or without Axl blockade, drove cells

to enter mitosis. Taken together, these data show that EGF-EGFR signalling

promotes cell proliferation through upregulation of TIMP1 and cyclin D1 (26),

whilst EGFR can also exploit Axl signalling to promote cell invasion through

downregulation of TIMP1 and cyclin D1 with concomitant upregulation of MMP9.

Moreover, combination of gefitinib and BGB324 reduces cell viability more

potently than either agent alone (Figure 5.11). Conclusively, inhibition of Axl

and EGFR potentiates a more effective GBM treatment as it presents a dual

targeting approach in which the proliferation and invasion of cancer cells is

inhibited.

5.2.3 Summary

In this chapter, a specific relationship between EGFR and Axl RTKs was

demonstrated, enabled through a direct protein-protein interaction, which

diversifies the signalling pathways available to EGFR. Having previously shown

that Axl signalling is required for GBM cell invasion (15), in this chapter it is

shown for the first time that EGFR can also signal via Axl to promote cell

invasion (Figure 5.12), a role for which EGFR, being a major mitogenic growth

146

factor, is not well known as a single entity. Furthermore, it is likely that EGFR-

Axl signalling balances invasion and proliferation through the regulation of key

genes such as CCND1, TIMP1 and CD44; indeed, EGFR alone promotes

expression of these genes whilst EGFR-Axl supresses their expression. Given

these observations, it is conceivable that a combination of EGFR and Axl

inhibition can be more effective than monotherapy for treating some cancer

patients, in particular as a means to prevent or delay secondary resistance to

targeted therapy.

Figure 5.12: Schematic summary of this study, showing hetero-interaction between EGFR and Axl, with unidirectional activation of Axl by EGFR, leading to activation of pro-invasive signalling pathways, involving TIMP1 as a central regulator.

147

Chapter 6

TAM downstream signalling in

GBM

148

6.0 Introduction

Receptor tyrosine kinases are cell surface receptors known to regulate

major biological processes such as growth and survival of cells. RTKs have been

studied extensively throughout the years both in normal and pathological cell

behaviour. Indeed, they were established early on as oncogenes, promoting

cancer progression and metastasis (246), and therefore presented as promising

new selective targets for therapy using small molecule inhibitors (247). The

TAM subfamily of RTKs has generated much interest in recent years, especially

due to the observation that overexpression of Axl, through its ability to interact

with EGFR, is linked to cancer cell chemoresistance (248) as well as induction

of EMT in breast (216, 220) and lung (218) cancers.

Axl and MerTK have been shown to be overexpressed in glioblastoma

multiforme (GBM) (137, 138). GBMs are highly invasive in nature and do not

respond to current therapies (200). Preliminary studies in GBM show that

inhibition of Axl or MerTK signalling decreases cell growth, invasion and

chemoresistance (138, 144, 177, 178), making the TAMs an attractive target

for GBM therapy. However, while Axl and MerTK have become very popular

cancer research topics, Tyro3 has been left in the shadows. Only recently, Tyro3

overexpression was linked to increased proliferation in hepatocellular carcinoma

(249) as well as acquired resistance to taxol (chemotherapeutic drug) in ovarian

cancer (250). Even less is known of the function of Tyro3 in GBM. Tyro3

overexpression (217) in GBM cell lines was shown in Chapter 3 although its

exact role still remains to be unveiled.

As Tyro3 has been shown to have neuro-protective functions in the CNS

(130), I speculate that it is probably not a sole promoter/driver of glioma but

149

rather may exert pro-tumorigenic properties through association with other

RTKS. Indeed downregulation of both Axl and Tyro3 was able to reverse taxol

resistance in ovarian cancer (209, 251). Also, studies in rat fibroblasts

demonstrated that Tyro3 overexpression leads to increased cell proliferation in

the presence of Axl. Additionally, Tyro3 co-IPed with Axl and enhanced Axl

activation by their common ligand Gas6, while Axl was shown to phosphorylate

a mutated, kinase-dead form of Tyro3 (180). In previous chapters, the

possibility of interaction between the two receptors in glioblastoma was

proposed, although their relationship was not investigated further. The purpose

of this chapter is dual: firstly, the Axl-Tyro3 interaction is explored, as well as

downstream signalling in GBM. A second aim was to investigate Axl signalling

in GBM independently of any other interactions.

150

6.1 Results

6.1.1 Tyro3 physically interacts with Axl

In Chapter 3, Tyro3 and Axl receptor as well as their common natural

ligand Gas6 were shown to be overexpressed in glioblastoma (217). To

investigate the relationship between Tyro3 and Axl, co-IP experiments were

performed in SNB-19 cells, which confirmed a physical interaction between the

two endogenous RTKs (Figure 6.1A). Addition of human recombinant Gas6 for

20 minutes only slightly increased the co-IP association. Having previously

observed high levels of Gas6 in this cell line (217) (which can stimulate Axl and

Tyro3 both in an autocrine and paracrine matter (108) (Figure 6.1B), the

importance of Gas6 in this interaction was further investigated. Cells were either

left untreated, treated with Gas6 for 20 minutes or treated with the vitamin K

antagonist warfarin for 24 hours to deactivate any endogenous Gas6. Co-IP

experiments revealed that Gas6 enhances the complex formation resulting in a

greater pulldown of the Axl-Tyro3 complex. In confirmation, warfarin treatment

resulted in a lower amount of the complex, thus indicating that Gas6 is required

for heterodimerisation (Figure 6.1C).

151

Figure 6.1: Hetero-interaction of Axl and Tyro RTKs. A. Co-IP of Axl and Tyro3, followed by Western blot probing for presence in the complexes of Axl (left blot) and Tyro3 (right blot). B. Co-IP of Axl and Tyro3, followed by Western blotting for presence in the complexes of Axl (left blot) and Tyro3 (right blot) following 15-minute stimulation with Gas6 (400 ng/ml). C. Co-IP of Axl and Tyro3, followed by Western blotting for presence in the complexes of Axl (left blot) and Tyro3 (right blot) following 24-hour pre-treatment with warfarin and 15-minute stimulation with Gas6 (400 ng/ml). Control co-IP antibody (Ctrl Ig1) was anti-GAPDH. N=3 separate experiments.

152

6.1.2 Axl and Tyro3 protein expression are interdependent

Initially, in order to investigate signalling downstream of the Axl-Tyro3

hetero-interaction, expression of each receptor was transiently silenced in turn.

Surprisingly, Axl knockdown resulted in a significant decrease in Tyro3 protein

levels, whilst Tyro3 knockdown had no effect on Axl expression (Figure 6.2A,

p<0.0043 and p<0.0053 for Axl and Tyro3 respectively, ANOVA with Tukey

multiple comparison post-hoc analysis). Analysis of the mRNA levels by qRT-

PCR showed no significant difference in gene expression of Tyro3 after Axl

knockdown (Figure 6.2B, p<0.4260 unpaired two tailed student t-test); this

indicated that the results observed were not due to off-target effects on Tyro3

mRNA by the Axl siRNA, but rather that there is some kind of reciprocal

regulation by both receptors on each other’s protein expression levels at the

post-transcriptional level. Alternatively, Tyro3 was less stable without Axl

present, resulting in faster degradation. Consequently, due to the observation

of this co-dependent expression amongst Axl and Tyro3, which was a novel

finding, as well as the fact that they share a common ligand (Gas6), further

investigation of this phenomenon and its downstream signalling became very

challenging. Moreover, in Chapter 4, Axl inhibition by the highly specific SMI

BGB324 was able to reduce phosphorylation of Tyro3 in UP007 GBM cells (217),

also indicating that Tyro3 activation and signalling is dependent on Axl.

153

Figure 6.2: Effect of gene silencing by RNAi on TAM phosphorylation and signalling in GBM cells. Western blot showing protein knockdown by increasing concentrations of Axl siRNAs (A, N=3) and Tyro3 siRNAs (B, N=3) in SNB-19 cells, followed by semi-quantification of the results in bar graphs. C. qRT-PCR data showing effect of Axl siRNA knockdown (15 nM) on Axl and Tyro3 gene expression. Data are mean±SEM protein expression normalised against loading control protein; ANOVA with Tukey multiple comparison post-hoc analysis (A) and unpaired two tailed student t-test (B), **p<0.01, *p<0.05 and ns, not significant.

154

6.1.3 Tyro3 knockdown induces a cell cycle G2/M arrest

The primary aim of this section was to identify genes targeted by Tyro3

in GBM. Since Axl and Tyro3 share the same ligand, Gas6, (92) and Tyro3

expression is Axl co-dependent, Tyro3 knockdown was utilised to identify

changes in gene expression. Briefly, cells were treated with 20 nM of Tyro3 and

control siRNAs, and after 48 hours they were treated with Gas6 for an extra 24

hours in order to activate gene expression. Total RNA was extracted and

reversed transcribed, and cDNA derived from an original RNA concentration of

50 ng/µl was loaded into each well of the Qiagen RT2 profiler PCR array and

gene expression analysed according to manufacturer’s protocol (SABiosciences)

(252). The manufacturer’s RT2 profiler PCR array data analysis package was

employed to derive fold-regulation, presented in Table 7. Positive fold-change

values indicate an up-regulation whilst negative values indicate a down-

regulation. Out of the 84 genes in the array, the 3 most upregulated and 3 most

downregulated genes (with a cut-off of ≥1.30 fold-change) were selected.

Notably, the cell cycle regulatory gene CCND1 was downregulated, whilst MOS,

S1004A and CDH1 genes were upregulated. Additionally, cell survival and DNA

damage repair related genes JUN and BCL2L1 were downregulated.

155

Table 7: Effects of Tyro3 knockdown on tumour suppressor and oncogenes expressed by qRT-PCR using the Qiagen ‘Tumour suppressor and oncogene’ RT2 profiler PCR array. Data are fold-change (FC) determined in SNB-19 cells incubated for 24 hours with Gas6 (400 ng/ml), compared to SNB-19 cells in the absence or presence of Tyro3 knockdown (N=1 experiment).

Symbol Fold Change Type

MOS ↑ 1.815 Serine/threonine kinase

S100A4 ↑ 1.4845 Signalling molecule

CDH1 ↑ 1.366 Adhesion molecule

JUN ↓ -1.4845 Transcription factor

BCL2L1 ↓ -1.4743 Signalling molecule

CCND1 ↓ -1.3379 Kinase activator

Having measured changes in genes relevant to the cell cycle and cell

survival, the functional role of Tyro3 knockdown was then investigated through

cell cycle analysis and annexin V/PI staining. Prior to cell cycle analysis, SNB-

19 cells were treated with Tyro3 and control siRNAs and Gas6 for 24 hours. Cell

cycle analysis revealed that following Tyro3 knockdown 52.4% of the cells

(Figure 6.3B) analysed were in the S-phase of the cell cycle compared to 37.3%

in control cells (Figure 6.3A). Moreover, in the case of Tyro3 knockdown only

12.1% of cells (Figure 6.3B) were able to progress to the M phase compared to

27.3% for control cells (Figure 6.3A), indicating a defect in the G2/M transition

upon Tyro3 knockdown.

156

Figure 6.3: Cell cycle analysis of SNB-19 cells treated for 24 hours with Gas6 following siTyro3 knockdown (B) and control (A). Markers in the displayed plots were used to demarcate G1, S ad M phases of the cell cycle (N=1 experiment).

157

Annexin V/PI staining also revealed that cells with Tyro3 knocked down

were less healthy compared to control. Indeed, 54% of Tyro3 knockdown cells

were in the early apoptotic phase compared to 29% for control cells. In

addition, treatment with staurosporine, a well-established inducer of apoptosis,

increased early apoptosis in 55% of control cells, but had no further effect on

Tyro3 knockdown cells, indicating an already existing commitment to apoptosis

in these cells (Figure 6.4).

Figure 6.4: Effect of Tyro3 knockdown on cell survival. Propidium iodide (PI) vs Annexin V fluorescence intensity in SNB-19 cells for incubated with Tyro3 (siTyro3) and control (siControl) siRNAs and treated with either Gas6 or combination of Gas6 + staurosporine. Quadrants and markers in the displayed plots were used to demarcate the various cell populations (N=1).

6.1.4 Axl downstream signalling in GBM

The activation/deactivation of TAM network-related intracellular

signalling kinases in SNB-19 cells was evaluated using the human phospho-RTK

array, a semi-quantitative tool which allows for the simultaneous identification

of the phosphorylation status of 42 kinases in a rapid and sensitive manner.

SNB-19 cells were serum-starved overnight and then stimulated with either 400

158

ng/ml Gas6 for 15 minutes or pre-incubated with 5 µM BGB324 for 1 hour.

Quantification by optical densitometry of pixel density (OD/mm2) confirmed

activation (through phosphorylation state) of 33 kinases following Gas6

stimulation. Pre-incubation with BGB324 either completely inhibited the

activation state of some kinases, or reduced the effect of Gas6, or had no effect.

Specifically, a complete inhibition of Gas6-induced phosphorylation of β-catenin,

AMPK1α, Akt 1/2/3 (S473), c-jun, PLC-γ1, STAT5 α/β and PYK2 kinases was

observed, indicating exclusive Axl regulation (Figure 6.6). Moreover, whilst Gas6

was able to induce phosphorylation of ERK1/2, GSK3 α/β, Akt 1/2/3 (T308),

STAT6 and HcK kinases, BGB324 was unable to block this activation, indicating

that another TAM such as Tyro3 or MerTK likely mediated those Gas6 effects

(Figure 6.5). In addition, interestingly, PRAS40 and Fgr phosphorylation were

completely abolished upon Gas6 stimulation, whilst treatment with BGB324

significantly increased their phosphorylation above basal levels (Figure 6.6).

Figure 6.5: Relative pixel densitometry bar graph indicating changes in phosphorylation

levels of kinases for untreated, treated with Gas6 (400 ng/ml) or Gas6 in combination

with 5 µM BGB324 for SNB-19 cells (N=1 experiment (membrane)).

159

Figure 6.6: Relative pixel densitometry bar graph indicating changes in phosphorylation levels of kinases for untreated, treated with

Gas6 (400 ng/ml) or Gas6 in combination with 5 µM BGB324 in SNB-19 cells (N=1 experiment (i.e. one membrane probed)).

160

As an additional investigation, the activation of the NF-κB signalling

pathway was investigated by measuring its downstream gene transcriptional

activity through a luciferase assay system. The Dual-Luciferase® Reporter

(DLR™) Assay System utilizes two reporter assays. Plasmids are transfected

into the cell and after lysis the activities of firefly and Renilla luciferases are

measured sequentially from a single sample. The Renilla luciferase vector acts

as an internal control for plasmid transfection, whilst the firefly luciferase vector

is used to measure transcriptional activity for the gene of interest. The

luminescence values are then normalised to the activity of the Renilla control

to minimise experimental variability. (253). Using this system, Gas6 treatment

for 24 hours was not observed to activate NF-κB signalling (Figure 6.7A,

p<0.8447, p<0.5322 for SNB-19 and UP007 cells respectively, unpaired two

tailed student t-test), nor did Axl blockade with BGB324 affect the pathway in

both SNB-19 and UP007 cells (Figure 6.7C, p<0.2497 and p<0.7088

respectively, ANOVA with Tukey multiple comparison post-hoc analysis).

However, treatment with BMS777607 significantly reduced NF-κB pathway

activity at both at 15 and 24 hours after treatment in both cell lines (Figure

6.7D, p<0.0083 and p<0.0017 for SNB-19 and UP007 respectively, ANOVA with

Tukey multiple comparison post-hoc analysis). In addition, SNB-19 cells with

Axl expression stably knocked down through shRNA did not show any significant

difference in NF-κB activity as compared to control knockdown cells, whilst a

significant decrease in NF-κB activity was observed upon Axl knockdown in the

UP007 cell line (Figure 6.7B, p<0.1515 ANOVA with Tukey multiple comparison

161

post-hoc analysis for SNB-19 cells and p<0.0064 unpaired two tailed student t-

test for the UP007 cells).

Figure 6.7: NF-κB expression levels in SNB-19 and UP007 cells as indicated by percentage LARII/Renilla after A. Gas6 (400 ng/ml) treatment for 24 hours, B. Stable Axl knockdown clones, C. treatment with 1 µM BGB324 for 15 and 24 hours and D. treatment with 100 µM and 35 µM BMS777607 for SNB-19 and UP007 cells respectively for 15 and 24 hours. Data are mean±SEM; N=3, ANOVA with Tukey multiple comparison post-hoc analysis or student t-test **p<0.01, * p<0.05, ns; not significant vs control.

162

6.1.5 EGFR-Axl downstream signalling in GBM

In Chapter 5, the EGFR-Axl heterodimerisation was investigated. Axl was

shown to act a gateway for EGFR to invasive signalling through regulation of

the TIMP1-MMP9 axis. In order to further investigate EGFR-Axl downstream

signalling, the human phospho-RTK array was employed, which allows for the

simultaneous detection of relative phosphorylation of 49 different RTKs by

chemiluminescence detection. SNB-19 cells were serum-starved overnight and

then stimulated with either 50 ng/ml EGF for 10 minutes or pre-incubated with

5 µM BGB324 for 1 hour before EGF stimulation. Quantification by optical

densitometry of pixel density (OD/mm2) confirmed activation of 34 signalling

kinases following EGF stimulation. Pre-incubation with BGB324 was used to

block Axl kinase activity and pinpoint which kinases are phosphorylated by EGFR

only or in cooperation with Axl. Interestingly, Akt 1/2/3 (S743), eNOS, β-catenin

and HSP60 were strongly phosphorylated after EGF stimulation, whilst this was

completely inhibited in the presence of BGB324, indicating a unique EGFR-Axl

signalling pathway. Moreover, WINK1 was deactivated by EGF stimulation and

completely reactivated above basal levels when Axl kinase was blocked. An

increase in phosphorylation was also observed after EGF stimulation of PI3K

pathway-related kinases such as Akt 1/2/3 (T308), TOR and CREB, which was

kept at basal levels by Axl kinase blockade. Interestingly, phosphorylation of

FAK and PRAS40 as well as Fgr were completely inhibited following EGF

addition, with Axl inhibition having no effect on this, thus indicating a direct

regulation by EGFR independently of Axl (Figure 6.9). Furthermore, GSK3 α/β,

STAT3 (S727) and STAT5 α/β were all increased in their phosphorylation states

by EGF stimulation with a lack of influence of BGB324 on this (Figure 6.8).

163

H u m an P h o sp h o k in a se a s sa y

Re

lati

ve

Pix

el

De

nsit

om

etr

y (

OD

/mm

2)

GS

K-3

/b

ST

AT

5a

p7

0 S

6 K

ina

se

(T

42

1/4

24

)

Fy

n

ST

AT

5b

ST

AT

3 (

Y7

05

)

FA

K

ST

AT

5

/b

ST

AT

3 (

S7

27

)

PR

AS

40

EG

FR

0 .0

0 .1

0 .2

0 .3

U n tre a te d

E G F

E G F + B G B 3 24

Figure 6.8: Relative pixel densitometry bar graph indicating phosphorylation levels of

kinases for untreated, treated with 50 ng/ml EGF or EGF in combination with 5µM

BGB324 in SNB-19 cells (N=1 experiment).

164

Figure 6.9: Relative pixel densitometry bar graph indicating phosphorylation levels of kinases for untreated, treated with 50 ng/ml EGF

or EGF in combination with 5µM BGB324 in SNB-19 cells (N=1 experiment (one membrane probed))

165

6.2 Discussion

6.2.1 Tyro3 physically interacts with Axl

The TAM receptor genes share similar genomic structures, with Axl and

Tyro3 sharing more similarity than MerTK as well as having a higher affinity for

their common ligand Gas6 (97, 119, 254, 255). A heterodimerisation between

the receptors has been speculated for years although supporting evidence was

not shown until recently. In one study, Axl co-immunoprecipitated with Tyro3

in gonadotropin-releasing hormone (GnRH) neuronal cells and was shown to

promote survival (117). In Rat2 fibroblasts, Gas6 treatment induced Tyro3-

dependent Axl phosphorylation, which further lead to the trans-phosphorylation

of Tyro3 (180). Furthermore, downregulation of both Axl and Tyro3 were able

to reverse taxol resistance in ovarian cancer (209, 251). In Chapter 3, Tyro3

overexpression in GBM was reported for the first time, whilst in Chapter 4 a hint

of such hetero-interaction was presented when BGB324, a highly specific Axl

inhibitor, was observed to also affect Tyro3 phosphorylation. The aim of the

first part of this chapter was to investigate the Axl-Tyro3 protein-protein

interaction and downstream signalling in GBM cells. Indeed, co-IP experiments

confirmed a physical interaction between the two RTKs in GBM cells on

endogenous levels (Figure 6.1). Importantly, this interaction was shown to be

dependent on the presence of Gas6 in the system (Figure 6.2), indicating that

ligand presence may be essential for receptor interaction, countering the notion

of the TAMs being able to homodimerise even in the absence of their ligand

(179, 256).

166

A study by Tsou and colleagues on ligand specificity and TAM receptor

activation that Gas6 alone was sufficient to induce Axl homodimerisation, which

occurred with 2:2 stoichiometry (103). However, this was not observed with

MerTK and Tyro3. The same investigators also reported that only one minor

Gas6 binding site is highly conserved in all TAMs, whereas the second major

Gas6 binding site in MerTK and Tyro3 is not (103, 109). Furthermore,

computational modelling performed by Sasaki and colleagues shows that

monomeric Tyro3 is readily accommodated at the major Axl binding site of Gas6

(109). Moreover, co-precipitation studies and competition studies with soluble

TAMs revealed that Axl interaction with Gas6 was more conspicuous compared

with MerTK and Tyro3 interactions with Gas6, pointing out that if all three

receptors were present on a single cell, Axl would be the preferred receptor for

ligand binding (103). These results together indicate that Axl heterodimerisation

with MerTK or Tyro3 may also occur in addition to the canonical

homodimerisation of each RTK, in order to promote Gas6-induced signalling

and/or ligand-independent signalling. To further investigate this phenomenon,

we performed co-IP experiments after 15-minutes stimulation with exogenous

Gas6 and observed a slight increase in the amount of Tyro3 that co-IPed with

Axl. Furthermore, pre-treatment of cells with warfarin, which blocks

endogenous vitamin K epoxide reductase and therefore prevents -

carboxylation of synthesised Gas6, rendering it unable to activate the receptor

(103), reduced the Axl-Tyro3 co-IP, indicating that Gas6 also has an influence

on the heterodimerisation and that it is not entirely ligand-independent. In order

to further understand this interaction, structural analysis of the three-

dimensional structures of Axl and Tyro3 receptors with Gas6 binding by X-ray

167

crystallography is needed to understand the functional architecture of this

interaction. Furthermore, mutational analysis can delve further into the ligand-

receptors interaction as well as how heterodimerisation is induced through

ligand binding.

6.2.2 Axl and Tyro3 protein expression are interdependent

In an effort to dissect Axl-Tyro3 signalling, a gene knockdown approach

was employed. Surprisingly, knockdown of either receptor resulted in a

reduction of protein expression of the other. At the mRNA level, Tyro3

knockdown did not have any effect on Axl (Figure 6.3), whereas Axl knockdown

significantly reduced Tyro3 levels, indicating a possible regulation of expression

by Axl. Little is known about the regulation of these receptors. One mechanism

suggests endocytosis as a mechanism of Axl receptor downregulation after

Gas6 stimulated interaction of Axl with the ubiquitin ligase c-Cbl which leads to

ubiquitination and eventual degradation (101). At the post-transcriptional level,

Axl was shown to be regulated by two miRNAs, miR-34a and miR199a/b,

identified by a bioinformatics screen using non-small cell lung, breast and

colorectal cancer cell lines (119). The Axl gene promoter harbours binding sites

for transcription factors Sp1/Sp3, MZF1, CREB and AP2 (257). MerTK was found

to also harbour the same Sp1/Sp3 binding sites as Axl in Sertoli cells (258).

Therefore, given the homology of the TAM genes, it is possible that Tyro3 also

contains the Sp1/Sp3 regulatory binding sites. Sp1 activity has been shown to

be induced by the MAPK and PI3K pathways in response to growth factors

(259). Therefore, Axl can possibly regulate Sp1 activity through a feedback loop

in either a MAPK- or PI3K-dependent manner. Upon Axl knockdown, the loop is

disrupted and Tyro3 expression is also downregulated.

168

At the post-translational level, proper folding of the Axl protein, and its

final expression, was shown to be dependent on the molecular chaperone

HSP90, which functions to maintain conformational maturation and structural

integrity of a variety of cellular proteins (260). Indeed, inhibition of HSP90 was

shown to result in CHIP E3 ligase-dependent Axl degradation (260).

Furthermore, the TAMs are post-translationally modified through N-

glycosylation. Sasaki and colleagues pointed out that, although all three TAMs

possess homologous domains with conserved glycosylation sites, Tyro3 and Axl

nevertheless possess major differences in their conformations, in particular with

respect to the position of the Ig domains, where Axl-Ig1 is fully extended

whereas Tyro3-Ig1 appears to bend, thus making a glycosylation side

inaccessible (109). This indicates that proper protein folding and modification

is essential for receptor expression. Furthermore, Axl could possibly regulate

Tyro3 expression through a variety of ways: for instance, Src, which is a non-

receptor tyrosine kinase protein linked to cancer progression by promoting

other RTK signals (147) possesses SH2, SH3, and tyrosine kinase domains, has

been shown to bind Axl in B-cell chronic lymphocytic leukaemia cells (147).

Moreover, Src is able to phosphorylate HSP90 and thereby affect the folding

and activation of certain proteins. The same regulation of Axl expression by

Tyro3 can be expected as Tyro3 can also bind and activate the Src kinase (261).

Interestingly, Axl may also regulate HSP90 through eNOS, whose product nitric

oxide (NO) can mediate S-nitrosylation of HSP90, which is a reversible covalent

modification that inhibits HSP90 ATPase activity (262). Furthermore, eNOS is

negatively regulated through sequestration by caveolin-1 scaffolding domain

(263), which is an Axl binding partner (264), therefore removal of Axl would

169

promote activation of eNOS and NO production that leads to HSP90 inhibition

and therefore incorrect Tyro3 folding and subsequent degradation (Figure

6.10).

Figure 6.10: Proposed model of the mechanism of post-translational Tyro3 regulation by Axl. On the left, the sequestration of eNOS by caveolin-1 allows correct Tyro3 folding and expression by HSP90. On the right, upon Axl knockdown eNOS is released and inhibits HSP90 function through S-nitrosylation.

Currently, there are no reports on regulation of Tyro3 by Axl or vice

versa. Given the limited amount of information there is on TAM gene regulation

as well as post-translational modification, no direct links can be drawn to explain

the phenomenon observed in this study. However, an exciting new field of

investigation is presented and more research should be done on the how the

TAMs are regulated.

6.2.3 Axl-Tyro3 related signalling

Due to their common ligand, a technical challenge was presented in

terms of probing further the nature of the Axl-Tyro3 interaction in order to e.g.

map both their discrete and concerted signalling effects. As an attempt, I

decided to knockdown Tyro3, as Axl expression was not affected by its

170

knockdown, and stimulate the cells with exogenous Gas6 followed by a qRT-

PCR expression screen of well-known tumour suppressors and oncogenes

(Table 7).

6.2.3.1 Upregulated genes

6.2.3.1.1 The MOS gene

From the screen, we identified MOS, the product of which is a kinase

first discovered to control the meiotic cell cycle in vertebrate oocytes (265). Mos

kinase functions via activating maturation promoting factor (MPF), a universal

G2/M regulator in eukaryotic cell division (265). Mos has been reported to act

as an oncogene and to be a potent activator of the MAPK cascade(266).

Interestingly, constitutive activation of the MOS/MAPK pathway, results in S-

phase-induced apoptosis or G1 cell cycle arrest (265). Moreover, cells in G2 enter

the M phase but are unable to undergo cytokinesis and instead arrest as

binucleated cells in a G1-like state (267, 268). Additionally, MOS has been shown

to disrupt the M phase through its localisation and phosphorylation of a

kinetochore component, thus increasing chromosome instability and preventing

normal congregation (269). There are no reports of direct suppression of MOS

kinase by Tyro3 in the literature, making this a novel finding. The MOS gene is

a CREB responsive gene, which can act as an activator or a suppressor of a

gene depending on its phosphorylation state (270-272). CREB is usually

activated by the MAPK pathway, which is downstream of Tyro3 (92) and could

present a novel negative regulatory pathway for MOS by Tyro3. This

suppression enables cancer cells to progress through the cell cycle even though

they are characterised by genomic instability and would otherwise enter

171

apoptosis. Moreover, MOS kinase is negatively regulated by Casein Kinase 2

(CK2) (273) which is activated by the Src family of kinases (274), which are in

turn activated by Tyro3 (261) pointing to a second negative regulatory

mechanism mediated by Tyro3. Moreover, CK2 acts as a positive modulator of

Wnt/β-catenin/Lef-Tcf pathway, suppressing β-catenin degradation and β-

catenin binding to APC (275), a known proliferation inducer in glioma (276).

Therefore, the data outlined above collectively indicate that Tyro3 may suppress

Mos kinase whilst upregulating pro-proliferative signalling, thus establishing its

role as a propagator of GBM cell growth (Figure 6.11).

Figure 6.11: Proposed mechanism of Tyro3-mediated MOS kinase suppression in GBM cell survival and proliferation.

6.2.3.1.2 The S100A4 gene

The second most upregulated gene was S100A4, which encodes for a

Ca+2-binding protein. Even though no known enzymatic activity has been

assigned to this protein, it is considered to promote its effect through interaction

with other proteins (277). The S100A4 protein has been shown to regulate cell

172

cycle G2-M transition by sequestering p53 (277). The S100A4 gene promoter

has been shown to contain sequences recognised by Sp1, AP-1 and CBF

transcription factors (278) Even though no known direct interaction has been

identified between S100A4 and Tyro3, Tyro3 is known to activate ERK1/2

downstream (92), which regulates AP1-mediated transcription (279). Once

again, the effect of Tyro3 presence in cancer cells seems to promote the

successful entry and progression through the cell cycle, enabling evasion of

checkpoint mediated arrest and apoptosis.

6.2.3.1.3 The CDH1 gene

The third most upregulated gene was CDH1, which encodes for E-

cadherin. E-cadherin is known to interact with APC to regulate mitotic

progression (280). Activation of APC/Cdh1 by genotoxic stress has been linked

to abrogated G2/M progression, spindle body assembly and chromatid

segregation (281). Furthermore, E-cadherin regulates the cell cycle on many

levels: it controls the turnover of the CDK inhibitors p27 and p21; it inhibits

cyclin D transcription by degradation of Ets2 and degradation of the CDK1

activator Cdc25A; it modulates Chk1 activity by regulating claspin stability;

lastly, it regulates microtubules dynamics and mitotic spindle formation by

targeting JNK for degradation during G2/M transition (280-283). Moreover, E-

cadherin is a well-established tumour suppressor, as its loss usually leads to

EMT induction and increased cell migration and invasion (284). E-cadherin has

been shown to be negatively regulated in cancer through the recruitment of

HDACs to remodel the chromatin structure by Twist (285). Twist expression is

activated by Akt, STAT3, MAPK, Ras, and Wnt signalling (286), and Tyro3 has

been shown to activate Akt and MAPK (92). Therefore, these observations

173

together suggest that Tyro3 signalling leads to downregulation of E-cadherin

whilst its upregulation would occur upon Tyro3 signalling inhibition or

knockdown (as we observed here). The importance of E-cadherin regulation by

Tyro3 is dual: firstly, E-cadherin suppression is required for EMT induction,

which has been shown to be mediated by Axl to increase chemoresistance.

Additionally, Axl signalling promotes invasion, which requires the loss of cell

adhesion. Secondly, one of the hallmarks of tumour cells is uncontrolled

proliferation which is usually achieved through the loss of cell cycle regulatory

checkpoints. Suppression of E-cadherin by Tyro3 relieves cancer cells from the

G2/M transition checkpoint, thus allowing proliferation and survival.

6.2.3.2 Downregulated genes

6.2.3.2.1 The JUN gene

The JUN gene codes for the c-Jun, a transcription factor which together

with fos transcription factor forms the AP-1 transcription factor complex (287).

AP-1 controls expression of genes involved in proliferation, differentiation, cell

death and the response to genotoxic agents or stress (288). c-Jun acts a cell

proliferation promoter by activating cyclin D1 transcription and by inhibiting p21

accumulation through repression of p53 expression (288). AP-1 was also shown

to regulate apoptosis, although its role is very complex as it can trigger both

pro- or anti-apoptotic signals depending on cell type or the death-inducing

treatments (288). Moreover, c-Jun activation by JNK2 has been implicated with

chemoresistance in cancer through transcription of Multi-Drug Resistance

(MDR1) gene (289). c-Jun can be regulated by degradation and gene

expression, and the main regulators of its expression include MAPK family

174

proteins (279, 290). Jun N-terminal kinase (JNK) binds and phosphorylates the

c-Jun transactivation domain, increasing its ability to activate the transcription

of target genes. c-Jun can also be phosphorylated and form a heterodimer with

ATF-2 which activated the c-jun promoter. Moreover, ERK is found to increase

c-jun transcription and stability through CREB, ATF-2 and GSK3 respectively

(279, 290). Tyro3 has been known to activate ERK1/2 downstream (92),

therefore promoting c-jun gene expression. Given the significance of c-Jun in

chemoresistance, survival and proliferation, activation by Tyro3 in cancer cells

provides them with an additional advantage. Axl has already been shown to

mediate EMT-related chemoresistance to EGFR SMIs in several other cancers.

This effect though could be mediated via Tyro3 through heterodimerisation, or

it could provide an additive effect where Tyro3 induces pro-survival gene

expression. Additionally, it can present another potential mechanism by which

promotes neuronal survival.

6.2.3.2.2 The BCL2L1 gene

The second most downregulated gene was BCL2L1, which codes for BCL-

XL protein, a potent pro-survival factor. BCL-XL exerts its pro-survival effects in

two main ways: firstly, it inhibits pro-apoptotic BAX and BAK action by

competing with them for binding to membranes, thus preventing the formation

of lethal pores in the mitochondrial outer membrane (291). Secondly, BCL-XL

binds to the voltage-dependent anion channel 1 (VDAC1), a porin ion channel

located on the outer mitochondrial membrane, and inhibits mitochondrial

permeability transition and thus cell death (292, 293). Overexpression of BCL-

XL in cancer is very common and has been linked to increased multidrug

chemoresistance (294). Moreover, it has been shown to aid cancer cells

175

overcome mitotic spindle damage checkpoint and genomic instability (295,

296). Upregulation of BCL-XL by Tyro3 was also observed by Protein S

stimulation in an excitotoxic injury neuronal model. Moreover, BCL-XL

upregulation was shown to be mediated by the PI3K-Akt pathway and to protect

neurons from cell death (130). In GBM, Tyro3 could utilise BCL-XL expression

to protect the cancer cell from chemotherapy-induced cell death, as well as

increased genomic instability-induced cell death.

6.2.3.2.3 The CCND1 gene

The third most downregulated gene upon Tyro3 knockdown in GBM cells

was CCND1, which encodes for cyclin D1. Cyclin D1 promotes cell proliferation

by binding to and activating CDK4 and CDK6; the complex formed then

phosphorylates the tumour suppressor protein pRB, which regulates the E2F

transcription factors, promoting gene expression required for entry of cells into

the S phase of cell cycle (297, 298) (Figure 6.12).

Figure 6.12: Cyclin D1 function. Entrance into S phase is characterised by phosphorylation of pRb by cyclin D1–cyclin-dependent kinase (CDK)-4, cyclin D1–CDK6 and cyclin E–CDK2 complexes. The phosphorylation of pRb is associated with release of the E2F1–3 transcription factors, which then activate genes that are required for cell-cycle progression (299).

176

Besides its major role in the cell cycle, cyclin D1 can bind the promoter

and initiate transcription of genes involved in migration and invasion, such as

thrombospondin and the Rho effector ROCK2 (297, 298). Moreover, cyclin D1

regulates the expression of genes that are involved in DNA replication and the

DNA damage checkpoint, and can directly bind proteins involved in the DNA

damage response (297, 298). Cyclin D1 has been shown to promote

homologous recombination-mediated DNA repair by forming a complex with

recombinase RAD51 and BRCA2 and recruiting them to the damage site (297,

298). Cyclin D1 is often overexpressed in cancer and aids cells to mount an

enhanced DNA damage response and increased genomic instability (298). Cyclin

D1 gene expression has been previously associated with Tyro3 in cancer. Tyro3

knockdown in breast cancer and hepatocellular carcinoma resulted in decreased

proliferation which correlated with a decrease in cyclin D1 (154, 249, 300). The

cyclin D1 gene is regulated by a variety of transcription factors. Studies have

shown sp1, AP-1 as well as the ATF2-CREB complex of transcription factors to

promote its transcription (301). Even though no direct link has yet been

reported between Tyro3 and these transcription factors, in melanoma Tyro3

has been shown to regulate MITF gene expression through the transcription

factor SOX10 (153). SOX10 has been shown to interact with sp1, synergistically

activate the MBP promoter in oligodendrocytes (302). Therefore, Tyro3 may be

able to promote cyclin D1 expression through SOX10. Additionally, cyclin D1

expression in glioma has been shown to be highly dependent on CREB, which

was shown to be activated by the PI3K and MAPK pathways (303), both of

which are downstream targets of Tyro3 activity (92). During gene screening, I

also showed that c-Jun, which belongs to the Ap-1 complex, is positively

177

regulated by Tyro3 and can promote cyclin D1 expression directly. This finding

has major implications for cancer cells as Tyro3 promotes entry through the cell

cycle and increases cell division.

6.2.3.3 NF-κB signalling

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

a protein complex that controls transcription of DNA to promote cell survival.

Normally, NF-κB dimers are sequestered in the cytoplasm by IκBs (Inhibitor of

κB). When a stimulus activates the pathway, IκB kinase (IKK) phosphorylates

and marks the IκB molecules for ubiquitination, thus freeing NF-κB and allowing

nuclear translocation to occur (Figure 6.13). NF-κB has been shown to be

activated by Akt (304, 305).

Figure 6.13: Schematic representation of NF-κB activation. NF-κB dimers are

sequestered in the cytoplasm by IκBs. Upon stimulation, IκB kinase (IKK) phosphorylates and marks the IκB molecules for ubiquitination, thus freeing NF-κB and

allowing nuclear translocation to occur (306).

178

Recently, it has been shown in schwannoma that activation of the

Gas6/Axl signalling pathway recruits Src, focal adhesion kinase (FAK) and NF-

κB, and that NF-κB mediates Gas6/Axl-mediated overexpression of survivin,

cyclin D1 and FAK, leading to enhanced survival, cell-matrix adhesion and

proliferation (307). In order to investigate the role of TAMs on the regulation of

NF-κB in GBM cells, we utilised a luciferase assay to detect changes in specific

NF-κB pathway-regulated gene expression under different conditions.

Stimulation of TAM signalling by exogenous Gas6 had no effect on NF-κB

pathway transcriptional regulatory activity. Also, stable Axl knockdown did not

affect NF-κB activity in SNB-19 cells, whilst it was reduced in UP007 cells. This

difference could be due to the expression of MerTK in SNB-19 cells (absent in

UP007 cells) which may compensate for the loss of Axl. Additionally, inhibition

of Axl showed a time-dependent decrease in NF-κB expression which was not

significant. However, the SMI BMS777607 completely inhibited NF-κB activity,

which can be explained by the fact that BMS777607 not only inhibits Axl but

also Tyro3 and MET receptors, all of which have been shown to regulate NF-κB

in other contexts.

6.2.4 Tyro3 promotes GBM cell proliferation and survival

Following the genetic alterations observed after Tyro3 knockdown cell

cycle analysis confirmed an increase of cells in the S and G2-M phases (Figure

6.3), which could be due to the aforementioned upregulation of MOS, S1004A

and CHD1 genes with concomitant downregulation of BCL2L1, CCND1 and JUN

genes. Moreover, Annexin V/PI staining indicated increased apoptosis in cells

lacking Tyro3 (Figure 6.4). Collectively these results indicate that Tyro3

179

signalling positively promotes the cell cycle, thus increasing cancer cell

proliferation, whilst protecting cells from cell death (Figure 6.14). These results

identify a new exciting role for Tyro3 in cancer. Due to Tyro3 being the least

studied TAM receptor, the mechanisms behind these phenomena certainly

warrants further investigation.

Figure 6.14: Proposed mechanism of Tyro3 mediated GBM cell survival and proliferation. Tyro3 activates Akt and ERK signalling which leads to MOS, S100A4 and CDH1 suppression, whilst upregulating BCL1L2, JUN and CCND1 transcription, thus providing GBM cells increased survival and cell cycle checkpoint evasion.

Additionally, it is still unclear whether Tyro3 achieves these functions

through regulation of certain gene sets alone or in co-operation with Axl. It

would be interesting to determine whether Axl takes advantage of Tyro3 to

exert pro-proliferative and survival signalling or else whether Tyro3 can fulfil

this role alone. What is becoming more clear from these studies is that Axl and

Tyro3 have distinctive roles, at least in GBM cells but most likely in other cells

180

in which they are co-expressed. Whilst Axl mediates GBM cell invasion, Tyro3

predominantly promotes cell proliferation and cell survival, thus making GBM

therapy very difficult and currently unsuccessful. These findings indicate that a

combinatory inhibition of both receptors may be more effective than targeting

either alone.

6.2.5 Axl downstream signalling in GBM

In order to dissect the Axl related intracellular signalling kinase network

in SNB-19 cells, a phosphokinase screen was performed after Gas6 stimulation

alone or in the presence of BGB324 (Figure 6.5 and 6.6). Pre-incubation of cells

with BGB324 either completely inhibited activation or reduced the effect of

Gas6, or had no effect. Specifically, we observed a complete inhibition of Gas6-

induced phosphorylation of the kinases β-catenin, AMPK1α, Akt 1/2/3 (S473),

c-Jun, PLC-γ1, STAT5 α/β and PYK2 in the presence of BGB324, indicating their

exclusive regulation by Axl out of the TAMs. In addition, Gas6 induced

phosphorylation of a further set of kinases, ERK1/2, GSK3 α/β, Akt 1/2/3

(T308), STAT6 and HcK, which BGB324 pre-incubation was unable to affect;

this therefore indicates that another TAM such as Tyro3 or MerTK is the main

RTK for their activation by Gas6, further supported by their having previously

been shown to be activated downstream (120). Conversely, Gas6 stimulation

actually completely abolished phosphorylation of PRAS40, a downstream target

of Akt and a negative regulator of the mTOR pathway, which is involved in the

control of cell growth, survival, proliferation (308). Phosphorylation was also

abolished of Fgr, a member of the Src family of protein tyrosine kinases, which

acts a negative regulator of cell migration and adhesion (309). Treatment with

181

BGB324 significantly increased phosphorylation of both kinases above basal

levels indicating that Axl is responsible for their negative regulation. As Axl is

now well-established as a pro-invasion signalling molecule, its inhibition of Fgr

and PRAS40 is therefore reasonable in the context of cancer as they are both

negative regulators of the hallmarks of cancer cells: invasion and proliferation.

The Wnt signalling pathway is centred around the regulation of β-catenin

phosphorylation/degradation. Canonically, in the absence of Wnt, β-catenin

protein is sequestered and degraded through the actions of a complex of

proteins formed by the scaffolding protein Axin, the tumour suppressor APC,

casein kinase 1 (CK1), and GSK3. CK1 and GSK3 sequentially phosphorylate the

amino terminal region of β-catenin, resulting in β-catenin recognition by β-Trcp,

an E3 ubiquitin ligase subunit, and subsequent β-catenin ubiquitination and

proteasomal degradation. When Wnt ligand is present, the Frizzled (fz) receptor

and its co-receptor, low-density lipoprotein receptor related protein 6 (LRP6)

are activated and form a complex that recruits Dishevelled (Dvl). The completed

complex then promotes LRP6 phosphorylation and inhibition of the inhibitory

Axin complex, thus stabilising β-catenin, which then translocates to the nucleus

to form complexes with TCF/LEF to activate gene expression (310) (Figure

6.15).

182

Figure 6.15: Schematic representation of the canonical Wnt/beta-catenin signalling pathway. (a) In the absence of Wnt signal, beta-catenin is recruited into the APC/Axin/GSK3β complex, and phosphorylated by GSK3β. The phosphorylated beta-catenin binds to the proteosome machinery and is targeted for degradation, suppressing transcription (b) Wnt binds to its Fz receptor and LRP5/6 co-receptor and activates Dvl, leading to the inhibition of APC/Axin/GSK3β-mediated β-catenin degradation leading to gene transcription (311).

β-catenin signalling has been shown to be crucial for GBM cell

proliferation and invasion (276). Moreover, it has been shown to be under the

regulation of the RTK MET in GBM (312). In the present study, we observed

increased phosphorylation of β-catenin after stimulation with Gas6 and its

complete inhibition upon Axl blockade with BGB324, indicating a complete

regulation by Axl signalling. Indeed, Gas6 been shown to activate Akt and

concomitantly inhibit GSK3 activity, promoting β-catenin stabilisation in breast

cancer cells (313). Moreover, as Axl is a strong activator of Akt, it therefore

183

makes sense that Axl thereby controls GSK3 activity (92). MET could also

regulate the Wnt signalling pathway via Axl. Indeed, MET was found to co-IP

with Axl in GnRH neuronal cells; MET signalling acted via Axl to trigger p38

MAPK/Akt and induce neuronal migration. The activation of MET by its ligand

HGF was only possible when Axl was present, indicating that cross-talk between

MET and Axl can exist and may be essential for signal diversification (314).

Additionally, Gas6 was shown to induce β-catenin stabilisation and subsequent

TCF/Lef transcriptional activation in a mammary system through Axl activation

but not Tyro3 (313).

AMPK was also phosphorylated upon Gas6 stimulation. AMPK is a

regulator of cellular energy homeostasis and increases cellular energy levels by

inhibiting anabolic energy-consuming pathways and stimulating energy-

producing, catabolic pathways (Figure 6.16). AMPK is a complex of three

proteins, STE-related adaptor (STRAD), mouse protein 25 (MO25), and LKB1,

a serine/threonine kinase that activates AMPK (315). Recently, LKB1 was shown

to be under the regulation of Protein kinase C (PKC) (316), which is also

regulated by Axl (317). Even though no direct relationship was found between

Axl and AMPK regulation, in our system Axl exclusively regulates AMPK

activation, probably to manage increasing energy demands of invading cancer

cells.

184

Figure 6.16: Schematic representation of AMPK’s subunits and its activation process. AMPK detects changes in the cellular energy state that occur in response to nutrient variations or metabolic stress caused by changes in the AMP/ATP ratio. The activation of AMPK triggers key enzymes of glucose metabolism and fatty acids (318).

The Akt serine/threonine kinase is part of the PI3K pathway, which

promotes survival and growth in response to extracellular signals. Growth factor

stimulation caused activation of the cell surface receptor PI3K. PI3K in turns

phosphorylates lipids on the cell membrane forming the second messenger

phosphatidylinositol (3,4,5)-trisphosphate (PIP3). Akt is then recruited to the

membrane containing PIP3 through its lipid docking sites, where it becomes

activated. Axl has been shown to interact with the p85β subunit of PI3K and

the adaptor protein Grb2 through tyrosine 821, leading to their activation (319).

Full activation of Akt requires phosphorylation by both kinases PDK1 and

mTORC2 on T308 and S473, respectively (320). In our cells, phosphorylation

of S473 seems to be exclusively regulated by Axl as both Gas6 and EGF induced

185

phosphorylation was completely abolished upon Axl blockade by BGB324. Axl

signalling through PI3K/Akt has been firmly established to regulate cell

migration (321), growth, angiogenesis (322), and apoptosis (323, 324) in many

circumstances, processes which are all taken advantage of by cancer cells to

spread, grow and survive.

The primary function of phospholipase C (PLC) is to catalyse the

hydrolysis of phosphatidylinositol 4,5-bisphophate (PIP2) to generate inositol

(1,4,5) trisphosphate (IP3) and 1,2-diacylglycerol (DAG). IP3 initiates an

increase in intracellular calcium, whereas DAG activates protein kinase C (PKC).

Axl has already been shown to mediate PLCγ signalling through the latter’s

interaction with tyrosines 821 and 866 in Axl (120). PLC-γ signalling initiates a

signalling cascade (Figure 6.17) that results in cytokine production, activation

of effector function, cell proliferation and cell migration (325, 326).

Figure 6.17: PI3K activation. Activated PI3K phosphorylates PIP2, thereby creating PIP3 at the plasma membrane. Both Akt and PDK1 are then recruited to the membrane by binding PIP3. This positions PDK1 and Akt such that PDK1 can phosphorylate Akt on threonine 308 (T308) and activate downstream signalling (327).

186

Axl has been shown to activate PLC-γ in CLL B cells and increase their

survival (147). PLC-γ has been shown to mediate HIF1α-induced

chemoresistance in glioma (328). It would thus be reasonable to propose that

Axl activates PLC-γ in GBM cells in order to counteract the effects of

chemotherapy and promote cell survival.

Signal transducer and activator of transcription 5 (STAT5) protein exists

in two forms, STAT5A and STAT5B which are 90% identical. STAT5 proteins

are involved in cytosolic signalling and in mediating the expression of specific

genes. The STAT5B complex has been shown to be instrumental in glioma cell

survival by direct activation of the BCL-XL promoter (329). Activation of STAT5

by Axl has been described in natural killer cell development (330) but there

their association has not been documented in any other context. It is possible

that Axl promotes glioma cell survival through BCL-XL transcription by STAT5

but more studies are required to elucidate the repercussions of such regulation.

STAT5 has also been to promote transcription of invasion-related genes in

glioma through a pathway dependent on STAT5 DNA binding (331). It is

therefore also conceivable that Axl may directly activate STAT5 to promote

glioma invasion.

PYK2 is a non-receptor protein-tyrosine kinase that regulates

reorganisation of the actin cytoskeleton, cell polarisation, cell migration,

adhesion, spreading and bone remodelling primarily by activating the MAPK

pathway (332). Moreover, activated Pyk2 acts as a scaffold for Src-dependent

phosphorylation of PDK1, which in turn phosphorylates paxillin and thus

regulates integrity of focal adhesions (333). PYK2 has been shown to partner

up with FAK to promote the Wnt/β-catenin pathway through phosphorylating

187

GSK3β (334). Signalling via Gas6/Axl in schwannoma resulted in FAK/Src/NFκB

activation and increased cell-matrix adhesion and survival (307). PYK2

activation has also been demonstrated to occur through PKC (335) which can

be activated through the interaction of Axl with PLC-γ (120). Given the invasive

role of Axl in many cancers, PYK2 regulation may add to its pro-invasion role

through the regulation of focal adhesions.

In conclusion, we have shown Axl to act as a promoter of cell invasion

through the exclusive regulation of β-catenin, AMPK1α, Akt 1/2/3 (S473), c-jun,

PLC-γ1, STAT5 α/β and PYK2 (Figure 6.18).

Figure 6.18: Summary of Axl-mediated activation of downstream intracellular signalling effectors resulting in increased migration, invasion, survival and proliferation of GBM cells.

188

6.2.6 EGFR-Axl downstream signalling in GBM

As shown in the previous section, Axl was found to exclusively

phosphorylate Akt 1/2/3 (S473) and regulate β-catenin even when activated by

EGFR, making these proteins direct Axl targets. Moreover, eNOS and HSP60

were strongly phosphorylated after EGF stimulation, which was blocked

completely by BGB324, indicating a unique EGFR-Axl hetero-interactive

signalling. Moreover, we show for the first time eNOS phosphorylation by Axl,

which confirms indirect HSP90 regulation by Axl and its probable involvement

in protein folding. WINK1 was deactivated by EGF stimulation and completely

reactivated above basal levels when Axl kinase was blocked. WINK1 kinase has

been shown to be phosphorylated and activated by EGFR to promote ERK5

MAPK cascades to promote neuronal growth (336); therefore this novel Axl

regulation may diversify EGFR signalling in GBM. Furthermore, an increase in

phosphorylation of PI3K-pathway related kinases such as Akt 1/2/3 (T308), TOR

and CREB was observed after EGF stimulation, which was reduced to basal

levels by Axl kinase blockade. When considering these together with the kinase

assay results described in the previous section, it is apparent that Axl is not the

main regulator of these proteins but amplifies the EGFR response. Interestingly,

FAK and PRAS40 as well as Fgr were completely inhibited following EGF

addition, while Axl inhibition had no effect on their phosphorylation. These

results therefore indicate that both EGFR and Axl negatively regulate these

proteins, as Axl activation by Gas6 had the same effect. As with Gas6

stimulation, GSK3 α/β, STAT3 (S727) were also phosphorylated by EGF

stimulation. BGB324 mediated Axl blockade was unable to counteract the EGF

189

effect indicating that Axl is not the primary regulator of these proteins but

another TAM receptor (Figure 6.19).

Figure 6.19: A. Summary of EGFR-Axl mediated activation of intracellular downstream signalling effectors resulting in increased migration, invasion, survival and proliferation of GBM cells. B. EGFR specific downstream targets.

6.2.7 Summary

In summary, this chapter demonstrates that Axl interacts with its “sister”

RTK, Tyro3, to promote proliferation and survival of GBM cells. Moreover,

whereas Axl mainly regulates, either on its own or via interaction with EGFR,

genes and kinases involved in cell invasion and chemoresistance, the Axl-Tyro3

hetero-interaction could diversify Axl signalling to additionally regulate cell

survival and proliferation. This is in keeping with the notion that EGFR utilises

Axl to access signalling pathways for cell migration and invasion (Chapter 4).

Furthermore, as we have shown EGFR to also phosphorylate Tyro3, a

supramolecular complex of these proteins may exist which gives a significant

190

advantage to cancer cells. Therefore, it is possible that targeted molecular

therapies aimed at inhibiting all 3 RTKs in this complex might be in order for

the successful treatment of a subset of GBM patients in whose tumours this

complex and associated signalling has been detected.

191

Chapter 7

General Discussion

192

7.1 Summary of key findings

In this study, I began with examining the expression levels of TAM

receptors in GBM. Axl overexpression in reactive astrocytes was confirmed in a

GBM tissue array as well as in GBM cell lines, in which Tyro3 overexpression

was also identified. Moreover, activation by traditional (Gas6 and Protein S) as

well as novel TAM ligands (Galectin-3 and Tulp-1) was shown through phospho-

RTK Western blotting.

A key novel finding was the identification of RTK hetero-complexes

involving Axl and/or Tyro3 with EGFR. The three receptors seem to co-operate

in order to promote invasion, survival and growth of GBM cells respectively.

Gliomas are characterised by excessive growth, invasion, anti-apoptotic

signalling, angiogenesis and areas of necrosis. During this study, Axl was first

shown to promote invasion of GBM cells, following which EGFR was revealed to

also signal via Axl, a process that diversifies its role beyond cell proliferation to

cell migration and invasion. Axl inhibition by both small molecule inhibition and

RNAi reduced cell invasion, whilst EGFR inhibition by gefitinib did not affect

basal invasion levels. Importantly, EGF-activated EGFR significantly increased

invasion only in the presence of an active Axl kinase, indicating that EGFR

invasive potential is mediated by Axl. In addition, RNAi and qRT-PCR gene

screening also revealed Tyro3 to provide GBM cells with alternative signalling

pathways, in this case the ability to evade cell cycle checkpoint-induced death.

Most importantly, EGFR was able to induce activation of both Axl and Tyro3

receptors, indicating that perhaps some co-operation between all three

receptors is required for GBM cells to become resilient to therapy. Interestingly,

more than 57% of GBM tumours either overexpress EGFR or possess an

193

overactive mutant version, which achieves constitutive signalling. However,

blockade of EGFR sooner than later leads to resistance to small molecule

inhibitors, thus making therapy ineffective for patients. In cancers such as

breast, lung and head and neck cancers, resistance to EGFR inhibition arises

from the overexpression of Axl, which activates downstream EMT signalling,

thus promoting chemoresistance. Tyro3 has not been implicated in this thus

far, probably because it has been the least studied of the three TAM receptors;

it would therefore be interesting to further investigate TAM signalling, both via

canonical homo-dimer formation as well as the novel potential heterodimer with

EGFR, or even possible hetero-trimer/multimers with other RTKs. Additionally,

in this study only the effect of EGF ligand was investigated. Taking into

consideration the existence of 6 additional ligands that are able to activate EGFR

their significance for this novel interaction needs further investigation. It is still

not clear if Axl can also be activated by EGFR activation via its alternative ligands

and how this affects downstream signalling.

7.2 The TAMs as therapeutic targets

The TAMs present promising targets for the treatment of cancers and

specifically GBM as they have been established as regulators of anti-apoptotic

pathway activation, cell growth and survival in a variety of malignancies. In this

thesis, as well as in a series of other studies, Axl inhibition results in decreased

invasion and proliferation. More importantly, Axl inhibition sensitises cells to

chemotherapy, indicating that its effects are multifaceted and the aim of

“shutting down” multiple key biological processes in cancer cells can be

achieved through a single target.

194

Even though the angiogenic switch observed in brain cancers as well as

necrosis was not investigated in this study, it would be reasonable to speculate

that TAM signalling positively regulates those processes in GBM as well. Tyro3

signalling has been shown to be critical in the regulation of the BBB integrity

(124) although the mechanisms are not known. Of significance, Axl signalling

has been shown to downregulate angiopoietin-2, an antagonist of angiopoietin-

1, a master promoter of vascular development and angiogenesis, in HUVEC cells

(337). Moreover, Axl-dependent activation of Akt is thought to promote tumour

angiogenesis in conjunction with VEGF-A and VEGFR2 (322). Importantly,

phosphorylated Axl was localised in vascular proliferates within GBM xenograft

tumours; in the same study Axl inhibition significantly decreased angiogenesis

both in vitro and in vivo (146). Gas6 was also demonstrated to induce

angiogenesis in the human retinal endothelial cells by ERK signalling (338).

Indeed, in this study Gas6 stimulation of Axl resulted in eNOS phosphorylation,

which positively regulates angiogenesis (339). Collectively these findings

indicate that TAM signalling is important for tumour angiogenesis and they

would make an attractive therapeutic target. By targeting the TAMs, the tumour

blood supply, survival and invasive potential would be interrupted thus making

them susceptible to chemotherapeutic induced death.

Previous efforts to target angiogenesis in glioma with bevacizumab failed

despite an initial response, leading to decreased abnormal morphology and

organisation of tumour-related vasculature because tumours become more

invasive (Figure 7.1) (340). Therefore, Axl inhibition may represent a more

promising therapy as it would target both tumour angiogenesis and GBM cell

invasion (138, 146, 217).

195

Figure 7.1: Schematic drawing: pre-treatment and during treatment with angiogenic agents in GBM. A. Contrast leakage (white) occurs around leaky tumour vessels enhancing the tumour area on MRI. Capillaries in surrounding tissue are not leaky before treatment. B. Contrast-enhanced area is strongly reduced under anti-VEGF treatment. Tumour cells migrate furtively into the surrounding tissue and co-opt existing vasculature after treatment (341).

196

7.3 Concerns about TAM inhibition

Targeting of the TAMs in cancer is now widely believed to results in high

therapeutic potential. Nonetheless there are some concerns that need to be

weighed with regards to their normal function. For example, MerTK mutations

lead to defective phagocytosis of photoreceptor outer segments by the retinal

pigment epithelium (RPE) resulting in retinal degeneration. However, studies in

rodents suggest that retinal degeneration only occurs after prolonged MerTK

inhibition, which is further supported by clinical data in humans (342). Whilst

this example concerns specifically the consequences of inhibition of MerTK, the

same potential concerns may also apply for Axl and Tyro3.

Tyro3 signalling is essential for normal development and function (132)

of the brain as well as that of the BBB (124). Therefore, Tyro3 blockade over

an extended period of time could result in BBB collapse as well as the

progression of neurological disorders resulting from synaptic rigidity and

demyelination.

Another important aspect that needs to be considered is the role of the

TAMs in immunity. Not only are the TAMs expressed in microglia to regulate

inflammation but they also seem to be essential for systemic immunity (Figure

7.2). Deficiencies in TAM signalling have been shown to result in sustained

immune activation and chronic inflammation (343). Additionally, cytokine

receptor signalling systems, such as signalling via the type 1 interferon receptor,

are co-dependent on TAM family receptors (344). Both Axl and MerTK are

expressed on multiple immune cells, such as dendritic cells and macrophages

and have an essential immune-suppressive role via the abrogation of Toll-like

receptor and cytokine receptor signalling. In a model of colorectal cancer,

197

deficiency of Axl and MerTK in mice led to increased production of inflammatory

cytokines favouring a tumour-promoting environment, leading to enhanced

formation of colonic adenomatous polyps (345). TAM targeting could therefore

have adverse effects for tumour promotion, in which case the development of

targeting strategies that avoid inciting adverse pro-inflammatory effects might

be optimal for future therapy development.

On the other hand, the TAMs drive natural killer (NK) cell functional

maturation and normal expression of inhibitory and activating NK cell receptors

(346). NK cells are anti-tumorigenic and conduct immune surveillance through

their ability to monitor cell surfaces of autologous cells for aberrant expression

of MHC class I molecules and cell stress markers, thus differentiating between

normal and malignant cells (347). A recent study, demonstrated a role for TAM

receptors in the regulation of cancer metastases by the modulation of NK cell

activity. In this study, mice with targeted inactivation of the E3 ubiquitin ligase,

Cbl-B, demonstrated enhanced NK cell activity against metastatic tumour

models and this unique activity was found to be related to the regulation of

TAM family receptors that were found to be ubiquitination substrates for Cbl-B.

Additional studies showed that TAM small molecule inhibition confers treatment

potential through enhancing in vivo NK cell activity. Moreover, Gas6 inactivation

by warfarin yielded similar results, indicating that TAM inhibition can have

additional anti-cancer activity through NK cell-mediated action against

metastatic disease (348).

198

Figure 7.2: TAM-Regulated Cycle of Inflammation. A. Schematic representation of APC signalling showing the engagement of TLR (green), cytokine receptor (blue), and TAM receptor (red) signalling pathways. B. Schematic representation of the inflammatory cycle initiated by TLR activation (green pathway) which results in the activation of cytokine receptors and Axl. The final stage of the cycle (red pathways) results in TAM signalling activation, the transcription of SOCS genes, and the inhibition of both cytokine receptor and TLR signalling pathways (343).

199

7.4 Future directions

Although, as discussed above, chronic and sustained inhibition of TAM

family RTKs may have deleterious side effects such as autoimmunity and

retinitis, they can still be effective targets in the treatment of cancer in the

short-term. The TAM kinases present a multifaceted grouping of anti-cancer

targets through their involvement in tumour invasion, proliferation, survival and

angiogenesis. Therapeutic TAM inhibition may sensitise tumour cells to killing

by chemotherapy, radiation or other targeted agents and, in doing so, may

create an even more robust innate immune response, which may enhance

immunotherapeutic efficacy in combination with immune checkpoint inhibitors.

A key challenge is the development of highly specific and effective TAM

inhibitors in order to limit toxicity and adverse effects resulting from pan-kinase

inhibition. Currently the most specific Axl inhibitor present is BGB324 (R428 ),

which potently blocks autophosphorylation of Axl at the multiple docking site

Tyr821 within the tyrosine kinase domain, with consequent inhibition of

activation of Akt, Src family of proteins phosphorylation with low nanomolar

affinity (160). BGB324 entered phase I clinical trials in 2013 for breast and lung

cancers but its effectiveness in clinical glioma is still unknown. In addition to

BGB324, new drugs are currently being developed which on the bench-side

seem to be very potent and hopefully they will enter into clinical trials sometime

in future (Table 8), amongst which at least one will be proven to be suiTable

for GBM and cross the BBB.

.

200

Table 8: Axl inhibitors currently in development

DP3975 Small

molecule

inhibitor

AXL Preclinical Inhibited cell migration and

proliferation in mesotheliomas (349).

LDC1267 Small

molecule

inhibitor

AXL,

Tyro3,

Mer

Preclinical Induces NK cells to kill tumour cells in

mouse metastatic breast cancer and

melanoma model (348)

NA80x1 Small

molecule

inhibitor

AXL Preclinical Inhibits AXL phosphorylation, cell

motility, and invasion in MDA-MB-435

cells (350).

YW327.6S2 Antibody AXL Preclinical Anti-AXL monoclonal antibody (351).

GL21.T Aptamer Axl Preclinical Binds the extracellular domain of Axl at

high affinity (352)

NPS-1034 Small

molecule

inhibitor

Axl

MET

Preclinical Inhibits cell proliferation and induces

cell death in EGFR resistant lung cancer

cell lines (353)

201

REFERENCES

1. Alberts DS, Hess LM. Introduction to Cancer Prevention. Fundamentals of Cancer Prevention. Berlin, Heidelberg: Springer Berlin Heidelberg; 2008. p. 1-12. 2. Hanahan D, Weinberg RA. The Hallmarks of Cancer. Cell. 2000;100(1):57-70. 3. Hanahan D, Weinberg Robert A. Hallmarks of Cancer: The Next Generation. Cell. 2011;144(5):646-74. 4. Tortora GJ, Derrickson B, Tortora GJ. Principles of anatomy and physiology. Hoboken, N.J.: Wiley; 2009. 5. Kimelberg HK, Nedergaard M. Functions of astrocytes and their potential as therapeutic targets. Neurotherapeutics : the journal of the American Society for Experimental NeuroTherapeutics. 2010;7(4):338-53. 6. Bradl M, Lassmann H. Oligodendrocytes: biology and pathology. Acta Neuropathologica. 2010;119(1):37-53. 7. Wake H, Moorhouse AJ, Nabekura J. Functions of microglia in the central nervous system – beyond the immune response. Neuron Glia Biology. 2011;7(Special Issue 01):47-53. 8. Spassky N, Merkle FT, Flames N, Tramontin AD, García-Verdugo JM, Alvarez-Buylla A. Adult Ependymal Cells Are Postmitotic and Are Derived from Radial Glial Cells during Embryogenesis. The Journal of Neuroscience. 2005;25(1):10-8. 9. Ostrom QT, Gittleman H, Stetson L, Virk SM, Barnholtz-Sloan JS. Epidemiology of Gliomas. In: Raizer J, Parsa A, editors. Current Understanding and Treatment of Gliomas. Cham: Springer International Publishing; 2015. p. 1-14. 10. Sanai N, Alvarez-Buylla A, Berger MS. Neural Stem Cells and the Origin of Gliomas. N Engl J Med. 2005;353(8):811-22. 11. Grzmil M, Hemmings BA. Deregulated signalling networks in human brain tumours. Biochimica et Biophysica Acta (BBA) - Proteins & Proteomics. 2010;1804(3):476-83. 12. Verhaak RGW, Hoadley KA, Purdom E, Wang V, Qi Y, Wilkerson MD, et al. Integrated Genomic Analysis Identifies Clinically Relevant Subtypes of Glioblastoma Characterized by Abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell. 2010;17(1):98-110. 13. Brennan C, Momota H, Hambardzumyan D, Ozawa T, Tandon A, Pedraza A, et al. Glioblastoma Subclasses Can Be Defined by Activity among Signal Transduction Pathways and Associated Genomic Alterations. PLoS ONE. 2009;4(11):e7752. 14. Van Meir EG, Hadjipanayis CG, Norden AD, Shu H-K, Wen PY, Olson JJ. Exciting New Advances in Neuro-Oncology: The Avenue to a Cure for Malignant Glioma. CA: a cancer journal for clinicians. 2010;60(3):166-93. 15. Foote Michael B, Papadopoulos N, Diaz Luis A, Jr. Genetic Classification of Gliomas: Refining Histopathology. Cancer Cell. 2015;28(1):9-11. 16. Yang P, Cai J, Yan W, Zhang W, Wang Y, Chen B, et al. Classification based on mutations of TERT promoter and IDH characterizes subtypes in grade II/III gliomas. Neuro-Oncology. 2016. 17. Wen PY, Reardon DA. Neuro-oncology in 2015: Progress in glioma diagnosis, classification and treatment. Nat Rev Neurol. 2016;12(2):69-70. 18. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJB, et al. Radiotherapy plus Concomitant and Adjuvant Temozolomide for Glioblastoma. N Engl J Med. 2005;352(10):987-96. 19. Schonberg DL, Lubelski D, Miller TE, Rich JN. Brain tumor stem cells: molecular characteristics and their impact on therapy. Molecular aspects of medicine. 2014;0:82-101.

202

20. van Tellingen O, Yetkin-Arik B, de Gooijer MC, Wesseling P, Wurdinger T, de Vries HE. Overcoming the blood-brain tumor barrier for effective glioblastoma treatment. Drug Resistance Updates. 2015;19:1-12. 21. Furnari FB, Fenton T, Bachoo RM, Mukasa A, Stommel JM, Stegh A, et al. Malignant astrocytic glioma: genetics, biology, and paths to treatment. Genes & Development. 2007;21(21):2683-710. 22. Molina JR, Adjei AA. The Ras/Raf/MAPK Pathway. Journal of Thoracic Oncology. 2006;1(1):7-9. 23. Guha A, Feldkamp MM, Lau N, Boss G, Pawson A. Proliferation of human malignant astrocytomas is dependent on Ras activation. Oncogene. 1997;15(23):2755-65. 24. Lavoie JN, L'Allemain G, Brunet A, Müller R, Pouysségur J. Cyclin D1 Expression Is Regulated Positively by the p42/p44MAPK and Negatively by the p38/HOGMAPK Pathway. Journal of Biological Chemistry. 1996;271(34):20608-16. 25. Aktas H, Cai H, Cooper GM. Ras links growth factor signaling to the cell cycle machinery via regulation of cyclin D1 and the Cdk inhibitor p27KIP1. Molecular and Cellular Biology. 1997;17(7):3850-7. 26. Weinberg RA. The retinoblastoma protein and cell cycle control. Cell. 1995;81. 27. Henson JW, Schnitker BL, Correa KM, Vondeimling A, Fassbender F, Xu HJ, et al. THE RETINOBLASTOMA GENE IS INVOLVED IN MALIGNANT PROGRESSION OF ASTROCYTOMAS. Annals of Neurology. 1994;36(5):714-21. 28. Schmidt EE, Ichimura K, Reifenberger G, Collins VP. CDKN2 (p16/MTS1) Gene Deletion or CDK4 Amplification Occurs in the Majority of Glioblastomas. Cancer Research. 1994;54(24):6321-4. 29. Hawkins PT, Anderson KE, Davidson K, Stephens LR. Signalling through class I PI3Ks in mammalian cells. Biochemical Society Transactions. 2006;34:647-62. 30. Samuels Y, Wang Z, Bardelli A, Silliman N, Ptak J, Szabo S, et al. High Frequency of Mutations of the PIK3CA Gene in Human Cancers. Science. 2004;304(5670):554-. 31. Maehama T, Dixon JE. The Tumor Suppressor, PTEN/MMAC1, Dephosphorylates the Lipid Second Messenger, Phosphatidylinositol 3,4,5-Trisphosphate. Journal of Biological Chemistry. 1998;273(22):13375-8. 32. Knobbe CB, Reifenberger G. Genetic alterations and aberrant expression of genes related to the phosphatidyl-inositol-3 '-kinase/protein kinase B (Akt) signal transduction pathway in glioblastomas. Brain Pathology. 2003;13(4):507-18. 33. Mora A, Komander D, van Aalten DMF, Alessi DR. PDK1, the master regulator of AGC kinase signal transduction. Seminars in Cell & Developmental Biology. 2004;15(2):161-70. 34. Wang HM, Wang H, Zhang W, Huang HJ, Lia WSL, Fuller GN. Analysis of the activation status of Akt, NF kappa B, and Stat3 in human diffuse gliomas. Laboratory Investigation. 2004;84(8):941-51. 35. Knowles MA, Hurst CD. Molecular biology of bladder cancer: new insights into pathogenesis and clinical diversity. Nat Rev Cancer. 2015;15(1):25-41. 36. Brennan CW, Verhaak RGW, McKenna A, Campos B, Noushmehr H, Salama SR, et al. The Somatic Genomic Landscape of Glioblastoma. Cell. 2013;155(2):462-77. 37. Schneider MR, Wolf E. The epidermal growth factor receptor ligands at a glance. Journal of Cellular Physiology. 2009;218(3):460-6. 38. Li L, Chakraborty S, Yang CR, Hatanpaa KJ, Cipher DJ, Puliyappadamba VT, et al. An EGFR wild type-EGFRvIII-HB-EGF feed-forward loop regulates the activation of EGFRvIII. Oncogene. 2014;33(33):4253-64. 39. Fan Q-W, Cheng Christine K, Gustafson WC, Charron E, Zipper P, Wong Robyn A, et al. EGFR Phosphorylates Tumor-Derived EGFRvIII Driving STAT3/5 and Progression in Glioblastoma. Cancer Cell. 2013;24(4):438-49.

203

40. Humphrey PA, Wong AJ, Vogelstein B, Friedman HS, Werner MH, Bigner DD, et al. Amplification and Expression of the Epidermal Growth Factor Receptor Gene in Human Glioma Xenografts. Cancer Research. 1988;48(8):2231-8. 41. Thorne AH, Zanca C, Furnari F. Epidermal growth factor receptor targeting and challenges in glioblastoma. Neuro-Oncology. 2016. 42. Nishikawa R, Ji XD, Harmon RC, Lazar CS, Gill GN, Cavenee WK, et al. A mutant epidermal growth factor receptor common in human glioma confers enhanced tumorigenicity. Proceedings of the National Academy of Sciences. 1994;91(16):7727-31. 43. Nagane M, Coufal F, Lin H, Bögler O, Cavenee WK, Huang H-JS. A Common Mutant Epidermal Growth Factor Receptor Confers Enhanced Tumorigenicity on Human Glioblastoma Cells by Increasing Proliferation and Reducing Apoptosis. Cancer Research. 1996;56(21):5079-86. 44. Chakravarti A, Chakladar A, Delaney MA, Latham DE, Loeffler JS. The Epidermal Growth Factor Receptor Pathway Mediates Resistance to Sequential Administration of Radiation and Chemotherapy in Primary Human Glioblastoma Cells in a RAS-dependent Manner. Cancer Research. 2002;62(15):4307-15. 45. Zadeh G, Bhat Krishna PL, Aldape K. EGFR and EGFRvIII in Glioblastoma: Partners in Crime. Cancer Cell. 2013;24(4):403-4. 46. Johnson LN, Lowe ED, Noble MEM, Owen DJ. The structural basis for substrate recognition and control by protein kinases 1. FEBS Letters. 1998;430(1-2):1-11. 47. Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. The Protein Kinase Complement of the Human Genome. Science. 2002;298(5600):1912-34. 48. Foreman C. John , Johansen Torben, Alasdair GJ. Textbook of Receptor Pharmacology. Third edition ed. United States of America: Taylor and Francis group, LCC; 2010. 49. Zhang J, Yang PL, Gray NS. Targeting cancer with small molecule kinase inhibitors. Nat Rev Cancer. 2009;9(1):28-39. 50. Taylor TE, Furnari FB, Cavenee WK. Targeting EGFR for Treatment of Glioblastoma: Molecular Basis to Overcome Resistance. Current cancer drug targets. 2012;12(3):197-209. 51. Rich JN, Reardon DA, Peery T, Dowell JM, Quinn JA, Penne KL, et al. Phase II Trial of Gefitinib in Recurrent Glioblastoma. Journal of Clinical Oncology. 2004;22(1):133-42. 52. Mellinghoff IK, Wang MY, Vivanco I, Haas-Kogan DA, Zhu S, Dia EQ, et al. Molecular Determinants of the Response of Glioblastomas to EGFR Kinase Inhibitors. N Engl J Med. 2005;353(19):2012-24. 53. Hegi ME, Diserens A-C, Bady P, Kamoshima Y, Kouwenhoven MCM, Delorenzi M, et al. Pathway Analysis of Glioblastoma Tissue after Preoperative Treatment with the EGFR Tyrosine Kinase Inhibitor Gefitinib—A Phase II Trial. Molecular Cancer Therapeutics. 2011;10(6):1102-12. 54. Jun HJ, Acquaviva J, Chi D, Lessard J, Zhu H, Woolfenden S, et al. Acquired MET expression confers resistance to EGFR inhibition in a mouse model of glioblastoma multiforme. Oncogene. 2012;31(25):3039-50. 55. Pareja F, Macleod D, Shu C, Crary JF, Canoll PD, Ross AH, et al. PI3K and Bcl-2 Inhibition Primes Glioblastoma Cells to Apoptosis through Down-regulation of Mcl-1 and phospho-BAD. Molecular Cancer Research. 2014. 56. Kouri FM, Jensen SA, Stegh AH. The Role of Bcl-2 Family Proteins in Therapy Responses of Malignant Astrocytic Gliomas: Bcl2L12 and Beyond. The Scientific World Journal. 2012;2012:8. 57. de Vries EGE, Gietema JA, de Jong S. Tumor Necrosis Factor–Related Apoptosis-Inducing Ligand Pathway and Its Therapeutic Implications. Clinical Cancer Research. 2006;12(8):2390-3.

204

58. Rieger L, Weller M, Bornemann A, Schabet M, Dichgans J, Meyermann R. BCL-2 family protein expression in human malignant glioma: a clinical-pathological correlative study. Journal of the Neurological Sciences. 1998;155(1):68-75. 59. Nagane M, Levitzki A, Gazit A, Cavenee WK, Huang HJS. Drug resistance of human glioblastoma cells conferred by a tumor-specific mutant epidermal growth factor receptor through modulation of Bcl-X(L) and caspase-3-like proteases. Proceedings of the National Academy of Sciences of the United States of America. 1998;95(10):5724-9. 60. Zhu H, Cao X, Ali-Osman F, Keir S, Lo H-W. EGFR and EGFRvIII interact with PUMA to inhibit mitochondrial translocalization of PUMA and PUMA-mediated apoptosis independent of EGFR kinase activity. Cancer Letters. 2010;294(1):101-10. 61. Wick W, Wagner S, Kerkau S, Dichgans J, Tonn JC, Weller M. BCL-2 promotes migration and invasiveness of human glioma cells. FEBS Letters. 1998;440(3):419-24. 62. Wick W, Grimmel C, Wild-Bode C, Platten M, Arpin M, Weller M. Ezrin-Dependent Promotion of Glioma Cell Clonogenicity, Motility, and Invasion Mediated by BCL-2 and Transforming Growth Factor-β2. The Journal of Neuroscience. 2001;21(10):3360-8. 63. Wick W, Wild-Bode C, Frank B, Weller M. BCL-2-induced glioma cell invasiveness depends on furin-like proteases. Journal of Neurochemistry. 2004;91(6):1275-83. 64. Barker FG, Davis RL, Chang SM, Prados MD. Necrosis as a prognostic factor in glioblastoma multiforme. Cancer. 1996;77(6):1161-6. 65. Zeh HJI, Lotze MT. Addicted to Death: Invasive Cancer and the Immune Response to Unscheduled Cell Death. Journal of Immunotherapy. 2005;28(1):1-9. 66. Stegh AH, Kesari S, Mahoney JE, Jenq HT, Forloney KL, Protopopov A, et al. Bcl2L12-mediated inhibition of effector caspase-3 and caspase-7 via distinct mechanisms in glioblastoma. Proceedings of the National Academy of Sciences. 2008;105(31):10703-8. 67. Baluk P, Hashizume H, McDonald DM. Cellular abnormalities of blood vessels as targets in cancer. Current Opinion in Genetics & Development. 2005;15(1):102-11. 68. Hashizume H, Baluk P, Morikawa S, McLean JW, Thurston G, Roberge S, et al. Openings between Defective Endothelial Cells Explain Tumor Vessel Leakiness. The American Journal of Pathology. 2000;156(4):1363-80. 69. Méndez O, Zavadil J, Esencay M, Lukyanov Y, Santovasi D, Wang S-C, et al. Knock down of HIF-1α in glioma cells reduces migration in vitro and invasion in vivo and impairs their ability to form tumor spheres. Molecular Cancer. 2010;9(1):1-10. 70. Zagzag D, Zhong H, Scalzitti JM, Laughner E, Simons JW, Semenza GL. Expression of hypoxia-inducible factor 1α in brain tumors. Cancer. 2000;88(11):2606-18. 71. Wenger RH, Gassmann M. Oxygen(es) and the hypoxia-inducible factor-1. Biol Chem. 1997;378(7):609-16. 72. Fong G-H. Mechanisms of adaptive angiogenesis to tissue hypoxia. Angiogenesis. 2008;11(2):121-40. 73. Yang M-H, Wu K-J. TWIST activation by hypoxia inducible factor-1 (HIF-1): Implications in metastasis and development. Cell Cycle. 2008;7(14):2090-6. 74. Kaur B, Khwaja FW, Severson EA, Matheny SL, Brat DJ, Van Meir EG. Hypoxia and the hypoxia-inducible-factor pathway in glioma growth and angiogenesis. Neuro Oncol. 2005;7. 75. Goldman CK, Kim J, Wong WL, King V, Brock T, Gillespie GY. Epidermal growth factor stimulates vascular endothelial growth factor production by human malignant glioma cells: a model of glioblastoma multiforme pathophysiology. Molecular Biology of the Cell. 1993;4(1):121-33.

205

76. Ke LD, Shi Y-X, Im S-A, Chen X, Yung WKA. The Relevance of Cell Proliferation, Vascular Endothelial Growth Factor, and Basic Fibroblast Growth Factor Production to Angiogenesis and Tumorigenicity in Human Glioma Cell Lines. Clinical Cancer Research. 2000;6(6):2562-72. 77. Lafuente JV, Adán B, Alkiza K, Garibi JM, Rossi M, Cruz-Sánchez FF. Expression of vascular endothelial growth factor (VEGF) and platelet-derived growth factor receptor-β (PDGFR-β) in human gliomas. Journal of Molecular Neuroscience. 1999;13(1):177-85. 78. Gerstner ER, Batchelor TT. Antiangiogenic Therapy for Glioblastoma. Cancer Journal (Sudbury, Mass). 2012;18(1):45-50. 79. Kreisl TN, Kim L, Moore K, Duic P, Royce C, Stroud I, et al. Phase II Trial of Single-Agent Bevacizumab Followed by Bevacizumab Plus Irinotecan at Tumor Progression in Recurrent Glioblastoma. Journal of Clinical Oncology. 2009;27(5):740-5. 80. Friedman HS, Prados MD, Wen PY, Mikkelsen T, Schiff D, Abrey LE, et al. Bevacizumab Alone and in Combination With Irinotecan in Recurrent Glioblastoma. Journal of Clinical Oncology. 2009;27(28):4733-40. 81. Bernstein JJ, Woodard CA. GLIOBLASTOMA CELLS DO NOT INTRAVASATE INTO BLOOD-VESSELS. Neurosurgery. 1995;36(1):124-32. 82. Louis DN. Molecular pathology of malignant gliomas. Annual Review of Pathology-Mechanisms of Disease. Annual Review of Pathology-Mechanisms of Disease. 12006. p. 97-117. 83. Chang L, Zhao DAN, Liu H-B, Wang Q-S, Zhang P, Li C-L, et al. Activation of sonic hedgehog signaling enhances cell migration and invasion by induction of matrix metalloproteinase-2 and -9 via the phosphoinositide-3 kinase/AKT signaling pathway in glioblastoma. Molecular Medicine Reports. 2015;12(5):6702-10. 84. Furukawa K, Kumon Y, Harada H, Kohno S, Nagato S, Teraoka M, et al. PTEN gene transfer suppresses the invasive potential of human malignant gliomas by regulating cell invasion-related molecules. International Journal of Oncology. 2006(29):73-81. 85. Lee WS, Woo EY, Kwon J, Park M-J, Lee J-S, Han Y-H, et al. Bcl-w Enhances Mesenchymal Changes and Invasiveness of Glioblastoma Cells by Inducing Nuclear Accumulation of β-Catenin. PLoS ONE. 2013;8(6):e68030. 86. Hu B, Guo P, Fang Q, Tao H-Q, Wang D, Nagane M, et al. Angiopoietin-2 induces human glioma invasion through the activation of matrix metalloprotease-2. Proceedings of the National Academy of Sciences. 2003;100(15):8904-9. 87. Lee H-C, Park I-C, Park M-J, An S, Woo S-H, Jin H-O, et al. Sulindac and its metabolites inhibit invasion of glioblastoma cells via down-regulation of Akt/PKB and MMP-2. Journal of Cellular Biochemistry. 2005;94(3):597-610. 88. Cohen KJ, Gibbs IC, Fisher PG, Hayashi RJ, Macy ME, Gore L. A phase I trial of arsenic trioxide chemoradiotherapy for infiltrating astrocytomas of childhood. Neuro-Oncology. 2013;15(6):783-7. 89. Koul D, Fu J, Shen R, LaFortune TA, Wang S, Tiao N, et al. Antitumor Activity of NVP-BKM120—A Selective Pan Class I PI3 Kinase Inhibitor Showed Differential Forms of Cell Death Based on p53 Status of Glioma Cells. Clinical Cancer Research. 2012;18(1):184-95. 90. Lemmon MA, Schlessinger J. Cell signaling by receptor-tyrosine kinases. Cell. 2010;141(7):1117-34. 91. Lewis PA, Crosier KE, Wood CR, Crosier PS. Analysis of the murine Dtk gene identifies conservation of genomic structure within a new receptor tyrosine kinase subfamily. Genomics. 1996;31:13-9. 92. Hafizi S, Dahlbäck B. Signalling and functional diversity within the Axl subfamily of receptor tyrosine kinases. Cytokine & Growth Factor Reviews. 2006;17(4):295-304.

206

93. Hafizi S, Dahlbäck B. Signalling and functional diversity within the Axl subfamily of receptor tyrosine kinases. Cytokine & Growth Factor Reviews. 2006;17(4):295-304. 94. O'Bryan JP, Frye RA, Cogswell PC, Neubauer A, Kitch B, Prokop C, et al. axl, a transforming gene isolated from primary human myeloid leukemia cells, encodes a novel receptor tyrosine kinase. Molecular and Cellular Biology. 1991;11(10):5016-31. 95. Mark MR, Scadden DT, Wang Z, Gu Q, Goddard A, Godowski PJ. rse, a novel receptor-type tyrosine kinase with homology to Axl/Ufo, is expressed at high levels in the brain. Journal of Biological Chemistry. 1994;269(14):10720-8. 96. Graham D, Dawson T, Mullaney D, Snodgrass H, Earp H. Cloning and mRNA expression analysis of a novel human protooncogene, c-mer [published erratum appears in Cell Growth Differ 1994 Sep;5(9):1022]. Cell Growth Differ. 1994;5(6):647-57. 97. Lu Q, Gore M, Zhang Q, Camenisch T, Boast S, Casagranda F, et al. Tyro-3 family receptors are essential regulators of mammalian spermatogenesis. Nature. 1999;398(6729):723-8. 98. Schulz AS, Schleithoff L, Faust M, Bartram CR, Janssen JWG. The genomic structure of the human UFO receptor. Oncogene. 1993;8:509-13. 99. Graham DK, Dawson TL, Mullaney DL, Snodgrass HR, Earp HS. Cloning and mRNA expression analysis of a novel human protooncogene, c-mer. Cell Growth Differ. 1994;5:647-57. 100. O'Bryan JP. Axl, a transforming gene isolated from primary human myeloid leukemia cells, encodes a novel receptor tyrosine kinase. Mol Cell Biol. 1991;11:5016-31. 101. Valverde P. Effects of Gas6 and hydrogen peroxide in Axl ubiquitination and downregulation. Biochemical and Biophysical Research Communications. 2005;333(1):180-5. 102. Graham DK, DeRyckere D, Davies KD, Earp HS. The TAM family: phosphatidylserine-sensing receptor tyrosine kinases gone awry in cancer. Nat Rev Cancer. 2014;14(12):769-85. 103. Tsou W-I, Nguyen K-QN, Calarese DA, Garforth SJ, Antes AL, Smirnov SV, et al. Receptor Tyrosine Kinases, TYRO3, AXL and MER, Demonstrate Distinct Patterns and Complex Regulation of Ligand-Induced Activation. Journal of Biological Chemistry. 2014. 104. Nagata K. Identification of the product of growth arrest-specific gene 6 as a common ligand for Axl, Sky, and Mer receptor tyrosine kinases. J Biol Chem. 1996;271:30022-7. 105. Mark MR, Chen J, Hammonds RG, Sadick M, Godowsk PJ. Characterization of Gas6, a Member of the Superfamily of G Domain-containing Proteins, as a Ligand for Rse and Axl. Journal of Biological Chemistry. 1996;271(16):9785-9. 106. Sasaki T, Knyazev PG, Cheburkin Y, Göhring W, Tisi D, Ullrich A, et al. Crystal Structure of a C-terminal Fragment of Growth Arrest-specific Protein Gas6. Journal of Biological Chemistry. 2002;277(46):44164-70. 107. Verma A, Warner SL, Vankayalapati H, Bearss DJ, Sharma S. Targeting Axl and Mer Kinases in Cancer. Molecular Cancer Therapeutics. 2011;10(10):1763-73. 108. Hafizi S, Dahlbäck B. Gas6 and protein S. FEBS Journal. 2006;273(23):5231-44. 109. Sasaki T, Knyazev PG, Clout NJ, Cheburkin Y, Göhring W, Ullrich A, et al. Structural basis for Gas6–Axl signalling. The EMBO Journal. 2006;25(1):80-7. 110. Caberoy NB, Maiguel D, Kim Y, Li W. Identification of tubby and tubby-like protein 1 as eat-me signals by phage display. Exp Cell Res. 2010;316. 111. Caberoy NB, Zhou Y, Li W. Tubby and tubby-like protein 1 are new MerTK ligands for phagocytosis. The EMBO Journal. 2010;29(23):3898-910.

207

112. Behrens EM, Gadue P, Gong S-y, Garrett S, Stein PL, Cohen PL. The mer receptor tyrosine kinase: expression and function suggest a role in innate immunity. European Journal of Immunology. 2003;33(8):2160-7. 113. Angelillo-Scherrer A, Burnier L, Flores N, Savi P, DeMol M, Schaeffer P, et al. Role of Gas6 receptors in platelet signaling during thrombus stabilization and implications for antithrombotic therapy. The Journal of Clinical Investigation. 2005;115(2):237-46. 114. Katagiri M, Hakeda Y, Chikazu D, Ogasawara T, Takato T, Kumegawa M, et al. Mechanism of Stimulation of Osteoclastic Bone Resorption through Gas6/Tyro 3, a Receptor Tyrosine Kinase Signaling, in Mouse Osteoclasts. Journal of Biological Chemistry. 2001;276(10):7376-82. 115. Nakamura YS, Hakeda Y, Takakura N, Kameda T, Hamaguchi I, Miyamoto T, et al. Tyro 3 Receptor Tyrosine Kinase and its Ligand, Gas6, Stimulate the Function of Osteoclasts. STEM CELLS. 1998;16(3):229-38. 116. Pierce A, Xu M, Bliesner B, Liu Z, Richards J, Tobet S, et al. Hypothalamic but not pituitary or ovarian defects underlie the reproductive abnormalities in Axl/Tyro3 null mice. Molecular and cellular endocrinology. 2011;339(1-2):151-8. 117. Pierce A, Bliesner B, Xu M, Nielsen-Preiss S, Lemke G, Tobet S, et al. Axl and Tyro3 Modulate Female Reproduction by Influencing Gonadotropin-Releasing Hormone Neuron Survival and Migration. Molecular Endocrinology. 2008;22(11):2481-95. 118. Chan PY, Silva EAC, De Kouchkovsky D, Joannas LD, Hao L, Hu D, et al. The TAM family receptor tyrosine kinase TYRO3 is a negative regulator of type 2 immunity. Science. 2016;352(6281):99-103. 119. Axelrod H, Pienta KJ. Axl as a mediator of cellular growth and survival. Oncotarget. 2014;5(19):8818-52. 120. Linger RMA, Keating AK, Earp HS, Graham DK. TAM Receptor Tyrosine Kinases: Biologic Functions, Signaling, and Potential Therapeutic Targeting in Human Cancer. Advances in Cancer Research. Volume 100: Academic Press; 2008. p. 35-83. 121. Lai C, Lemke G. An extended family of protein-tyrosine kinase genes differentially expressed in the vertebrate nervous system. Neuron. 1991;6(5):691-704. 122. Prieto AL, Weber JL, Lai C. Expression of the receptor protein-tyrosine kinases Tyro-3, Axl, and Mer in the developing rat central nervous system. The Journal of Comparative Neurology. 2000;425(2):295-314. 123. Cahoy JD, Emery B, Kaushal A, Foo LC, Zamanian JL, Christopherson KS, et al. A Transcriptome Database for Astrocytes, Neurons, and Oligodendrocytes: A New Resource for Understanding Brain Development and Function. The Journal of Neuroscience. 2008;28(1):264-78. 124. Zhu D, Wang Y, Singh I, Bell RD, Deane R, Zhong Z, et al. Protein S controls hypoxic/ischemic blood-brain barrier disruption through the TAM receptor Tyro3 and sphingosine 1-phosphate receptor. Blood. 2010;115(23):4963-72. 125. Prieto AL, Weber JL, Tracy S, Heeb MJ, Lai C. Gas6, a ligand for the receptor protein-tyrosine kinase Tyro-3, is widely expressed in the central nervous system1. Brain Research. 1999;816(2):646-61. 126. Stitt TN. The anticoagulation factor protein S and its relative, Gas6, are ligands for the Tyro 3/Axl family of receptor tyrosine kinases. Cell. 1995;80:661-70. 127. Allen MP, Zeng C, Schneider K, Xiong X, Meintzer MK, Bellosta P, et al. Growth Arrest-Specific Gene 6 (Gas6)/Adhesion Related Kinase (Ark) Signaling Promotes Gonadotropin-Releasing Hormone Neuronal Survival via Extracellular Signal-Regulated Kinase (ERK) and Akt. Molecular Endocrinology. 1999;13(2):191-201. 128. Pierce AM, Keating AK. TAM receptor tyrosine kinases: Expression, disease and oncogenesis in the central nervous system. Brain Research. 2014;1542(0):206-20.

208

129. Prieto AL, O’Dell S, Varnum B, Lai C. Localization and Signaling of the Receptor Protein Tyrosine Kinase Tyro3 in Cortical and Hippocampal Neurons. Neuroscience. 2007;150(2):319-34. 130. Zhong Z, Wang Y, Guo H, Sagare A, Fernández JA, Bell RD, et al. Protein S Protects Neurons from Excitotoxic Injury by Activating the TAM Receptor Tyro3–Phosphatidylinositol 3-Kinase–Akt Pathway through Its Sex Hormone-Binding Globulin-Like Region. The Journal of Neuroscience. 2010;30(46):15521-34. 131. Binder MD, Cate HS, Prieto AL, Kemper D, Butzkueven H, Gresle MM, et al. Gas6 Deficiency Increases Oligodendrocyte Loss and Microglial Activation in Response to Cuprizone-Induced Demyelination. The Journal of Neuroscience. 2008;28(20):5195-206. 132. Binder MD, Xiao J, Kemper D, Ma GZM, Murray SS, Kilpatrick TJ. Gas6 Increases Myelination by Oligodendrocytes and Its Deficiency Delays Recovery following Cuprizone-Induced Demyelination. PLoS ONE. 2011;6(3):e17727. 133. Weinger J, Brosnan C, Loudig O, Goldberg M, Macian F, Arnett H, et al. Loss of the receptor tyrosine kinase Axl leads to enhanced inflammation in the CNS and delayed removal of myelin debris during Experimental Autoimmune Encephalomyelitis. Journal of Neuroinflammation. 2011;8(1):49. 134. Fourgeaud L, Través PG, Tufail Y, Leal-Bailey H, Lew ED, Burrola PG, et al. TAM receptors regulate multiple features of microglial physiology. Nature. 2016;532(7598):240-4. 135. Wang J, Zhang H, Young AG, Qiu R, Argalian S, Li X, et al. Transcriptome Analysis of Neural Progenitor Cells by a Genetic Dual Reporter Strategy. STEM CELLS. 2011;29(10):1589-600. 136. Gely-Pernot A, Coronas V, Harnois T, Prestoz L, Mandairon N, Didier A, et al. An Endogenous Vitamin K-Dependent Mechanism Regulates Cell Proliferation in the Brain Subventricular Stem Cell Niche. STEM CELLS. 2012;30(4):719-31. 137. Hutterer M, Knyazev P, Abate A, Reschke M, Maier H, Stefanova N, et al. Axl and Growth Arrest–Specific Gene 6 Are Frequently Overexpressed in Human Gliomas and Predict Poor Prognosis in Patients with Glioblastoma Multiforme. Clinical Cancer Research. 2008;14(1):130-8. 138. Vajkoczy P, Knyazev P, Kunkel A, Capelle H-H, Behrndt S, von Tengg-Kobligk H, et al. Dominant-negative inhibition of the Axl receptor tyrosine kinase suppresses brain tumor cell growth and invasion and prolongs survival. Proceedings of the National Academy of Sciences. 2006;103(15):5799-804. 139. Wang Y, Moncayo G, Morin P, Jr., Xue G, Grzmil M, Lino MM, et al. Mer receptor tyrosine kinase promotes invasion and survival in glioblastoma multiforme. Oncogene. 2012. 140. Keating AK, Kim GK, Jones AE, Donson AM, Ware K, Mulcahy JM, et al. Inhibition of Mer and Axl Receptor Tyrosine Kinases in Astrocytoma Cells Leads to Increased Apoptosis and Improved Chemosensitivity. Molecular Cancer Therapeutics. 2010;9(5):1298-307. 141. Venugopal C, Li N, Wang X, Manoranjan B, Hawkins C, Gunnarsson T, et al. Bmi1 marks intermediate precursors during differentiation of human brain tumor initiating cells. Stem Cell Research. 2012;8(2):141-53. 142. Molofsky AV, Pardal R, Iwashita T, Park I-K, Clarke MF, Morrison SJ. Bmi-1 dependence distinguishes neural stem cell self-renewal from progenitor proliferation. Nature. 2003;425(6961):962-7. 143. Sparmann A, van Lohuizen M. Polycomb silencers control cell fate, development and cancer. Nat Rev Cancer. 2006;6(11):846-56. 144. Ott M, Litzenburger UM, Sahm F, Rauschenbach KJ, Tudoran R, Hartmann C, et al. Promotion of Glioblastoma Cell Motility by Enhancer of Zeste Homolog 2 (EZH2) Is Mediated by AXL Receptor Kinase. Plos One. 2012;7(10).

209

145. Ott M, Litzenburger UM, Sahm F, Rauschenbach KJ, Tudoran R, Hartmann C, et al. Promotion of Glioblastoma Cell Motility by Enhancer of Zeste Homolog 2 (EZH2) Is Mediated by AXL Receptor Kinase. PLoS ONE. 2012;7(10):e47663. 146. Onken J, Torka R, Korsing S, Radke J, Kremenetskaia I, Nieminen M, et al. Inhibiting receptor tyrosine kinase AXL with small molecule inhibitor BMS-777607 reduces glioblastoma growth, migration, and invasion in vitro and in vivo2016. 147. Ghosh AK, Secreto C, Boysen J, Sassoon T, Shanafelt TD, Mukhopadhyay D, et al. The novel receptor tyrosine kinase Axl is constitutively active in B-cell chronic lymphocytic leukemia and acts as a docking site of nonreceptor kinases: implications for therapy. Blood. 2011;117(6):1928-37. 148. Dufies M, Jacquel A, Belhacene N, Robert G, Cluzeau T, Luciano F, et al. Mechanisms of AXL overexpression and function in Imatinib-resistant chronic myeloid leukemia cells2011. 149. Hong C-C, Lay J-D, Huang J-S, Cheng A-L, Tang J-L, Lin M-T, et al. Receptor tyrosine kinase AXL is induced by chemotherapy drugs and overexpression of AXL confers drug resistance in acute myeloid leukemia. Cancer Letters. 2008;268(2):314-24. 150. Mahadevan D, Cooke L, Riley C, Swart R, Simons B, Della Croce K, et al. A novel tyrosine kinase switch is a mechanism of imatinib resistance in gastrointestinal stromal tumors. Oncogene. 2007;26(27):3909-19. 151. Linger RMA, Cohen RA, Cummings CT, Sather S, Migdall-Wilson J, Middleton DHG, et al. Mer or Axl receptor tyrosine kinase inhibition promotes apoptosis, blocks growth and enhances chemosensitivity of human non-small cell lung cancer. Oncogene. 2012. 152. Papadakis ES, Cichon MA, Vyas JJ, Patel N, Ghali L, Cerio R, et al. Axl Promotes Cutaneous Squamous Cell Carcinoma Survival through Negative Regulation of Pro-Apoptotic Bcl-2 Family Members. J Invest Dermatol. 2011;131(2):509-17. 153. Zhu S, Wurdak H, Wang Y, Galkin A, Tao H, Li J, et al. A genomic screen identifies TYRO3 as a MITF regulator in melanoma. Proceedings of the National Academy of Sciences. 2009;106(40):17025-30. 154. EKYALONGO RC, MUKOHARA T, FUNAKOSHI Y, TOMIOKA H, KATAOKA Y, SHIMONO Y, et al. TYRO3 as a Potential Therapeutic Target in Breast Cancer. Anticancer Research. 2014;34(7):3337-45. 155. Avilla E, Guarino V, Visciano C, Liotti F, Svelto M, Krishnamoorthy G, et al. Activation of TYRO3/AXL Tyrosine Kinase Receptors in Thyroid Cancer. Cancer Research. 2011;71(5):1792-804. 156. Gjerdrum C, Tiron C, Høiby T, Stefansson I, Haugen H, Sandal T, et al. Axl is an essential epithelial-to-mesenchymal transition-induced regulator of breast cancer metastasis and patient survival. Proceedings of the National Academy of Sciences. 2009. 157. Vuoriluoto K, Haugen H, Kiviluoto S, Mpindi JP, Nevo J, Gjerdrum C, et al. Vimentin regulates EMT induction by Slug and oncogenic H-Ras and migration by governing Axl expression in breast cancer. Oncogene. 2011;30(12):1436-48. 158. Tai KY, Shieh YS, Lee CS, Shiah SG, Wu CW. Axl promotes cell invasion by inducing MMP-9 activity through activation of NF-[kappa]B and Brg-1. Oncogene. 2008;27(29):4044-55. 159. Paccez JD, Vasques GJ, Correa RG, Vasconcellos JF, Duncan K, Gu X, et al. The receptor tyrosine kinase Axl is an essential regulator of prostate cancer proliferation and tumor growth and represents a new therapeutic target. Oncogene. 2012. 160. Holland SJ, Pan A, Franci C, Hu Y, Chang B, Li W, et al. R428, a Selective Small Molecule Inhibitor of Axl Kinase, Blocks Tumor Spread and Prolongs Survival in Models of Metastatic Breast Cancer. Cancer Research. 2010;70(4):1544-54.

210

161. Khoury HJ, Cortes JE, Kantarjian HM, Gambacorti-Passerini C, Baccarani M, Kim D-W, et al. Bosutinib is active in chronic phase chronic myeloid leukemia after imatinib and dasatinib and/or nilotinib therapy failure. Blood. 2012;119(15):3403-12. 162. Nagilla M, Brown RL, Cohen EEW. Cabozantinib for the Treatment of Advanced Medullary Thyroid Cancer. Advances in Therapy. 2012;29(11):925-34. 163. Feneyrolles C, Spenlinhauer A, Guiet L, Fauvel B, Daydé-Cazals B, Warnault P, et al. Axl Kinase as a Key Target for Oncology: Focus on Small Molecule Inhibitors. Molecular Cancer Therapeutics. 2014;13(9):2141-8. 164. Qian F, Engst S, Yamaguchi K, Yu P, Won K-A, Mock L, et al. Inhibition of Tumor Cell Growth, Invasion, and Metastasis by EXEL-2880 (XL880, GSK1363089), a Novel Inhibitor of HGF and VEGF Receptor Tyrosine Kinases. Cancer Research. 2009;69(20):8009-16. 165. Yan SB, Peek VL, Ajamie R, Buchanan SG, Graff JR, Heidler SA, et al. LY2801653 is an orally bioavailable multi-kinase inhibitor with potent activity against MET, MST1R, and other oncoproteins, and displays anti-tumor activities in mouse xenograft models. Investigational New Drugs. 2013;31(4):833-44. 166. Burbridge MF, Bossard CJ, Saunier C, Fejes I, Bruno A, Léonce S, et al. S49076 Is a Novel Kinase Inhibitor of MET, AXL, and FGFR with Strong Preclinical Activity Alone and in Association with Bevacizumab. Molecular Cancer Therapeutics. 2013;12(9):1749-62. 167. An Q, Fillmore HL, Vouri M, Pilkington GJ. Brain tumor cell line authentication, an efficient alternative to capillary electrophoresis by using a microfluidics-based system. Neuro-Oncology. 2014;16(2):265-73. 168. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Meth. 2012;9(7):671-5. 169. Berridge MV, Herst PM, Tan AS. Tetrazolium dyes as tools in cell biology: New insights into their cellular reduction. Biotechnol Annu Rev. 2005;11:127-52. 170. Malich G, Markovic B, Winder C. The sensitivity and specificity of the MTS tetrazolium assay for detecting the in vitro cytotoxicity of 20 chemicals using human cell lines. Toxicology. 1997;124(3):179-92. 171. Wen PY, Kesari S. Malignant Gliomas in Adults. New England Journal of Medicine. 2008;359(5):492-507. 172. Mrugala MM, Crew LK, Fink JR, Spence AM. Carboplatin and bevacizumab for recurrent malignant glioma. Oncology Letters. 2012;4(5):1082-6. 173. AYDIN B, PATIL M, BEKELE N, WOLFF JEA. Vincristine in High-grade Glioma. Anticancer Research. 2010;30(6):2303-10. 174. Caberoy NB, Alvarado G, Bigcas J-L, Li W. Galectin-3 is a new MerTK-specific eat-me signal. Journal of cellular physiology. 2012;227(2):401-7. 175. Cory AH, Owen TC, Barltrop JA, Cory JG. Use of an aqueous soluble tetrazolium/formazan assay for cell growth assays in culture. Cancer communications. 1991;3(7):207-12. 176. Barltrop JA, Owen TC, Cory AH, Cory JG. 5-(3-carboxymethoxyphenyl)-2-(4,5-dimethylthiazolyl)-3-(4-sulfophenyl)tetrazolium, inner salt (MTS) and related analogs of 3-(4,5-dimethylthiazolyl)-2,5-diphenyltetrazolium bromide (MTT) reducing to purple water-soluble formazans As cell-viability indicators. Bioorganic & Medicinal Chemistry Letters. 1991;1(11):611-4. 177. Rogers AEJ, Le JP, Sather S, Pernu BM, Graham DK, Pierce AM, et al. Mer receptor tyrosine kinase inhibition impedes glioblastoma multiforme migration and alters cellular morphology. Oncogene. 2012;31(38):4171-81. 178. Wang Y, Moncayo G, Morin P, Jr., Xue G, Grzmil M, Lino MM, et al. Mer receptor tyrosine kinase promotes invasion and survival in glioblastoma multiforme. Oncogene. 2013;32(7):872-82.

211

179. Burchert A, Attar EC, McCloskey P, Fridell YW, Liu ET. Determinants for transformation induced by the Axl receptor tyrosine kinase. Oncogene. 1998;16(24):3177-87. 180. Brown JE, Krodel M, Pazos M, Lai C, Prieto AL. Cross-Phosphorylation, Signaling and Proliferative Functions of the Tyro3 and Axl Receptors in Rat2 Cells. PLoS ONE. 2012;7(5):e36800. 181. Zhuang T, Lee H-S, Imperiali B, Prestegard JH. Structure determination of a Galectin-3–carbohydrate complex using paramagnetism-based NMR constraints. Protein Science : A Publication of the Protein Society. 2008;17(7):1220-31. 182. de Boer RA, van der Velde AR, Mueller C, van Veldhuisen DJ, Anker SD, Peacock WF, et al. Galectin-3: A Modifiable Risk Factor in Heart Failure. Cardiovascular Drugs and Therapy. 2014;28(3):237-46. 183. Ochieng J, Furtak V, Lukyanov P. Extracellular functions of galectin-3. Glycoconjugate Journal. 2002;19(7):527-35. 184. Le Mercier M, Fortin S, Mathieu V, Kiss R, Lefranc F. Galectins and Gliomas. Brain Pathology (Zurich, Switzerland). 2010;20(1):17-27. 185. Ikemori RY, Machado CML, Furuzawa KM, Nonogaki S, Osinaga E, Umezawa K, et al. Galectin-3 Up-Regulation in Hypoxic and Nutrient Deprived Microenvironments Promotes Cell Survival. PLoS ONE. 2014;9(11):e111592. 186. Debray C, Vereecken P, Belot N, Teillard P, Brion J-P, Pandolfo M, et al. Multifaceted role of galectin-3 on human glioblastoma cell motility. Biochemical and Biophysical Research Communications. 2004;325(4):1393-8. 187. Rankin EB, Fuh KC, Taylor TE, Krieg AJ, Musser M, Yuan J, et al. AXL is an Essential Factor and Therapeutic Target for Metastatic Ovarian Cancer. Cancer research. 2010;70(19):7570-9. 188. Mukhopadhyay S, Jackson PK. The tubby family proteins. Genome Biology. 2011;12(6):1-9. 189. Carroll K, Gomez C, Shapiro L. Tubby proteins: the plot thickens. Nat Rev Mol Cell Biol. 2004;5(1):55-64. 190. Ferland G. Vitamin K and the Nervous System: An Overview of its Actions. Advances in Nutrition: An International Review Journal. 2012;3(2):204-12. 191. Krishan A. Rapid flow cytofluorometric analysis of mammalian cell cycle by propidium iodide staining. The Journal of Cell Biology. 1975;66(1):188-93. 192. Huang L, Fu L. Mechanisms of resistance to EGFR tyrosine kinase inhibitors. Acta Pharmaceutica Sinica B. 2015;5(5):390-401. 193. Wu X, Liu X, Koul S, Lee CY, Zhang Z, Halmos B. AXL kinase as a novel target for cancer therapy. Oncotarget; Vol 5, No 20. 2014. 194. Fleuren EDG, Hillebrandt-Roeffen MHS, Flucke UE, te Loo DMWM, Boerman OC, van der Graaf WTA, et al. The role of AXL and the in vitro activity of the receptor tyrosine kinase inhibitor BGB324 in Ewing sarcoma. Oncotarget. 2014;5(24):12753-68. 195. Wilson C, Ye X, Pham TQ, Lin E, Chan SM, McNamara E, et al. AXL inhibition sensitizes mesenchymal cancer cells to anti-mitotic drugs. Cancer Research. 2014. 196. Liang C-C, Park AY, Guan J-L. In vitro scratch assay: a convenient and inexpensive method for analysis of cell migration in vitro. Nat Protocols. 2007;2(2):329-33. 197. Alley MC, Lieber MM. Improved optical detection of colony enlargement and drug cytotoxicity in primary soft agar cultures of human solid tumour cells. British Journal of Cancer. 1984;49(2):225-33. 198. Franken NAP, Rodermond HM, Stap J, Haveman J, van Bree C. Clonogenic assay of cells in vitro. Nat Protocols. 2006;1(5):2315-9. 199. van Genderen H, Kenis H, Lux P, Ungeth L, Maassen C, Deckers N, et al. In vitro measurement of cell death with the annexin A5 affinity assay. Nat Protocols. 2006;1(1):363-7.

212

200. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJB, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 2005;352(10):987-96. 201. Felgner PL, Gadek TR, Holm M, Roman R, Chan HW, Wenz M, et al. Lipofection: a highly efficient, lipid-mediated DNA-transfection procedure. Proceedings of the National Academy of Sciences. 1987;84(21):7413-7. 202. Zuris JA, Thompson DB, Shu Y, Guilinger JP, Bessen JL, Hu JH, et al. Cationic lipid-mediated delivery of proteins enables efficient protein-based genome editing in vitro and in vivo. Nat Biotech. 2015;33(1):73-80. 203. Kim TK, Eberwine JH. Mammalian cell transfection: the present and the future. Analytical and Bioanalytical Chemistry. 2010;397(8):3173-8. 204. Zufferey R, Nagy D, Mandel RJ, Naldini L, Trono D. Multiply attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat Biotech. 1997;15(9):871-5. 205. Myers SH, Brunton VG, Unciti-Broceta A. AXL Inhibitors in Cancer: A Medicinal Chemistry Perspective. Journal of Medicinal Chemistry. 2015. 206. Reece J, Urry L, Cain M, Wasserman S, Minorsky P, Jackson R. Campbell Biology (9th Edition): Benjamin Cummings; 2010. 207. Schroeder GM, An Y, Cai Z-W, Chen X-T, Clark C, Cornelius LAM, et al. Discovery of N-(4-(2-Amino-3-chloropyridin-4-yloxy)-3-fluorophenyl)-4-ethoxy-1-(4-fluorophenyl)-2-oxo-1,2-dihydropyridine-3-carboxamide (BMS-777607), a Selective and Orally Efficacious Inhibitor of the Met Kinase Superfamily. Journal of Medicinal Chemistry. 2009;52(5):1251-4. 208. Sheridan C. First Axl inhibitor enters clinical trials. Nat Biotech. 2013;31(9):775-6. 209. Kim N-Y, Lee H-Y, Lee C. Metformin targets Axl and Tyro3 receptor tyrosine kinases to inhibit cell proliferation and overcome chemoresistance in ovarian cancer cells. International Journal of Oncology. 2015;47:353-60. 210. Park I-K, Mishra A, Chandler J, Whitman SP, Marcucci G, Caligiuri MA. Inhibition of the receptor tyrosine kinase Axl impedes activation of the FLT3 internal tandem duplication in human acute myeloid leukemia: implications for Axl as a potential therapeutic target. Blood. 2013;121(11):2064-73. 211. Li Y, Jia L, Ren D, Liu C, Gong Y, Wang N, et al. Axl mediates tumor invasion and chemosensitivity through PI3K/Akt signaling pathway and is transcriptionally regulated by slug in breast carcinoma. IUBMB Life. 2014;66(7):507-18. 212. Li Y, Jia L, Liu C, Gong Y, Ren D, Wang N, et al. Axl as a downstream effector of TGF-β1 via PI3K/Akt-PAK1 signaling pathway promotes tumor invasion and chemoresistance in breast carcinoma. Tumor Biol. 2015;36(2):1115-27. 213. Chin YR, Toker A. Function of Akt/PKB Signaling to Cell Motility, Invasion and the Tumor Stroma in Cancer. Cellular signalling. 2009;21(4):470-6. 214. Pardridge WM. The Blood-Brain Barrier: Bottleneck in Brain Drug Development. NeuroRx. 2005;2(1):3-14. 215. Gjerdrum C, Tiron C, Høiby T, Stefansson I, Haugen H, Sandal T, et al. Axl is an essential epithelial-to-mesenchymal transition-induced regulator of breast cancer metastasis and patient survival. Proceedings of the National Academy of Sciences of the United States of America. 2010;107(3):1124-9. 216. Asiedu MK, Beauchamp-Perez FD, Ingle JN, Behrens MD, Radisky DC, Knutson KL. AXL induces epithelial to mesenchymal transition and regulates the function of breast cancer stem cells. Oncogene. 2014;33(10):1316-24. 217. Vouri M, An Q, Birt M, Pilkington GJ, Hafizi S. Small molecule inhibition of Axl receptor tyrosine kinase potently suppresses multiple malignant properties of glioma cells. Oncotarget; Vol 6, No 18. 2015.

213

218. Zhang Z, Lee JC, Lin L, Olivas V, Au V, LaFramboise T, et al. Activation of the AXL kinase causes resistance to EGFR-targeted therapy in lung cancer. Nat Genet. 2012;44(8):852-60. 219. Bae SY, Hong J-Y, Lee H-J, Park HJ, Lee SK. Targeting the degradation of AXL receptor tyrosine kinase to overcome resistance in gefitinib-resistant non-small cell lung cancer. Oncotarget. 2015;6(12):10146-60. 220. Meyer AS, Miller MA, Gertler FB, Lauffenburger DA. The Receptor AXL Diversifies EGFR Signaling and Limits the Response to EGFR-Targeted Inhibitors in Triple-Negative Breast Cancer Cells2013 2013-08-06 00:00:00. ra66-ra p. 221. Padfield E, Ellis HP, Kurian KM. Current Therapeutic Advances Targeting EGFR and EGFRvIII in Glioblastoma. Frontiers in Oncology. 2015;5:5. 222. Koresawa M, Okabe T. High-Throughput Screening with Quantitation of ATP Consumption: A Universal Non-Radioisotope, Homogeneous Assay for Protein Kinase. ASSAY and Drug Development Technologies. 2004;2(2):153-60. 223. Byers LA, Diao L, Wang J, Saintigny P, Girard L, Peyton M, et al. An epithelial-mesenchymal transition (EMT) gene signature predicts resistance to EGFR and PI3K inhibitors and identifies Axl as a therapeutic target for overcoming EGFR inhibitor resistance. Clinical cancer research : an official journal of the American Association for Cancer Research. 2013;19(1):279-90. 224. Normanno N, De Luca A, Bianco C, Strizzi L, Mancino M, Maiello MR, et al. Epidermal growth factor receptor (EGFR) signaling in cancer. Gene. 2006;366(1):2-16. 225. Herbst RS. Review of epidermal growth factor receptor biology. International Journal of Radiation Oncology • Biology • Physics. 2004;59(2):S21-S6. 226. Hatanpaa KJ, Burma S, Zhao D, Habib AA. Epidermal Growth Factor Receptor in Glioma: Signal Transduction, Neuropathology, Imaging, and Radioresistance. Neoplasia (New York, NY). 2010;12(9):675-84. 227. Gan HK, Cvrljevic AN, Johns TG. The epidermal growth factor receptor variant III (EGFRvIII): where wild things are altered. FEBS Journal. 2013;280(21):5350-70. 228. Huang PH, Mukasa A, Bonavia R, Flynn RA, Brewer ZE, Cavenee WK, et al. Quantitative analysis of EGFRvIII cellular signaling networks reveals a combinatorial therapeutic strategy for glioblastoma. Proceedings of the National Academy of Sciences of the United States of America. 2007;104(31):12867-72. 229. Jarzynka MJ, Hu B, Hui K-M, Bar-Joseph I, Gu W, Hirose T, et al. ELMO1 and Dock180, a Bipartite Rac1 Guanine Nucleotide Exchange Factor, Promote Human Glioma Cell Invasion. Cancer research. 2007;67(15):7203-11. 230. Feng H, Hu B, Jarzynka MJ, Li Y, Keezer S, Johns TG, et al. Phosphorylation of dedicator of cytokinesis 1 (Dock180) at tyrosine residue Y722 by Src family kinases mediates EGFRvIII-driven glioblastoma tumorigenesis. Proceedings of the National Academy of Sciences of the United States of America. 2012;109(8):3018-23. 231. Abu-Thuraia A, Gauthier R, Chidiac R, Fukui Y, Screaton RA, Gratton J-P, et al. Axl Phosphorylates Elmo Scaffold Proteins To Promote Rac Activation and Cell Invasion. Molecular and Cellular Biology. 2015;35(1):76-87. 232. Isono E, Schwechheimer C. Co-immunoprecipitation and Protein Blots T Plant Developmental Biology. Methods in Molecular Biology. 6552010. p. 377-87. 233. Sekar RB, Periasamy A. Fluorescence resonance energy transfer (FRET) microscopy imaging of live cell protein localizations. The Journal of Cell Biology. 2003;160(5):629-33. 234. O'Bryan JP, Fridell Y-W, Koski R, Varnum B, Liu ET. The Transforming Receptor Tyrosine Kinase, Axl, Is Post-translationally Regulated by Proteolytic Cleavage. Journal of Biological Chemistry. 1995;270(2):551-7. 235. Li J, Jia L, Ma Z-H, Ma Q-H, Yang X-H, Zhao Y-F. Axl glycosylation mediates tumor cell proliferation, invasion and lymphatic metastasis in murine hepatocellular carcinoma. World Journal of Gastroenterology : WJG. 2012;18(38):5369-76.

214

236. Mahadevan D, Theiss N, Morales C, Stejskal AE, Cooke LS, Zhu M, et al. Novel receptor tyrosine kinase targeted combination therapies for imatinib-resistant gastrointestinal stromal tumors (GIST). Oncotarget. 2015;6(4):1954-66. 237. Giles KM, Kalinowski FC, Candy PA, Epis MR, Zhang PM, Redfern AD, et al. Axl Mediates Acquired Resistance of Head and Neck Cancer Cells to the Epidermal Growth Factor Receptor Inhibitor Erlotinib. Molecular Cancer Therapeutics. 2013;12(11):2541-58. 238. Liu L, Greger J, Shi H, Liu Y, Greshock J, Annan R, et al. Novel Mechanism of Lapatinib Resistance in HER2-Positive Breast Tumor Cells: Activation of AXL. Cancer Research. 2009;69(17):6871-8. 239. Wu F, Li J, Jang C, Wang J, Xiong J. The role of Axl in drug resistance and epithelial-to-mesenchymal transition of non-small cell lung carcinoma. International Journal of Clinical and Experimental Pathology. 2014;7(10):6653-61. 240. Wang J, Wei Q, Wang X, Tang S, Liu H, Zhang F, et al. Transition to resistance: An unexpected role of the EMT in cancer chemoresistance. Genes & Diseases. 2016;3(1):3-6. 241. Fischer KR, Durrans A, Lee S, Sheng J, Li F, Wong STC, et al. Epithelial-to-mesenchymal transition is not required for lung metastasis but contributes to chemoresistance. Nature. 2015;advance online publication. 242. Zheng X, Carstens JL, Kim J, Scheible M, Kaye J, Sugimoto H, et al. Epithelial-to-mesenchymal transition is dispensable for metastasis but induces chemoresistance in pancreatic cancer. Nature. 2015;527(7579):525-30. 243. Martino A, San K, Daniel N, Ezekiel U. Chemoresistance-Induced Epithelial-Mesenchymal Transition of a Colorectal Cancer Cell Line. The FASEB Journal. 2015;29(1 Supplement). 244. Zhang F, Wang Z, Fan Y, Xu Q, Ji W, Tian R, et al. Elevated STAT3 Signaling-Mediated Upregulation of MMP-2/9 Confers Enhanced Invasion Ability in Multidrug-Resistant Breast Cancer Cells. International Journal of Molecular Sciences. 2015;16(10):24772-90. 245. Yanagita M, Arai H, Nakano T, Ohashi K, Mizuno K, Fukatsu A, et al. Gas6 Induces Mesangial Cell Proliferation via Latent Transcription Factor STAT3. Journal of Biological Chemistry. 2001;276(45):42364-9. 246. Ullrich A, Schlessinger J. Signal transduction by receptors with tyrosine kinase activity. Cell. 1990;61(2):203-12. 247. Hojjat-Farsangi M. Small-Molecule Inhibitors of the Receptor Tyrosine Kinases: Promising Tools for Targeted Cancer Therapies. International Journal of Molecular Sciences. 2014;15(8):13768-801. 248. Scaltriti M, Elkabets M, Baselga J. Molecular Pathways: AXL, a Membrane Receptor Mediator of Resistance to Therapy. Clinical Cancer Research. 2016. 249. Duan Y, Wong W, Chua SC, Wee HL, Lim SG, Chua BT, et al. Overexpression of Tyro3 and its implications on hepatocellular carcinoma progression. International Journal of Oncology. 2016;48:358-66 250. Lee C. Overexpression of Tyro3 receptor tyrosine kinase leads to the acquisition of taxol resistance in ovarian cancer cells. Molecular Medicine Reports. 2015;12(1485-1492. ). 251. Suh Y-A, Jo S-Y, Lee H-Y, Lee C. Inhibition of IL-6/STAT3 axis and targeting Axl and Tyro3 receptor tyrosine kinases by apigenin circumvent taxol resistance in ovarian cancer cells. International Journal of Oncology. 2015;46:1405-11. 252. Gaj S, Eijssen L, Mensink RP, Evelo CTA. Validating nutrient-related gene expression changes from microarrays using RT(2) PCR-arrays. Genes & Nutrition. 2008;3(3-4):153-7. 253. Muthumani K. Transient transfection and luciferase assay. 2006.

215

254. Lewis PM, Crosier KE, Wood CR, Crosier PS. Analysis of the MurineDtkGene Identifies Conservation of Genomic Structure within a New Receptor Tyrosine Kinase Subfamily. Genomics. 1996;31(1):13-9. 255. Heiring C, Dahlbäck B, Muller YA. Ligand Recognition and Homophilic Interactions in Tyro3: STRUCTURAL INSIGHTS INTO THE Axl/Tyro3 RECEPTOR TYROSINE KINASE FAMILY. Journal of Biological Chemistry. 2004;279(8):6952-8. 256. Konishi A, Aizawa T, Mohan A, Korshunov VA, Berk BC. Hydrogen Peroxide Activates the Gas6-Axl Pathway in Vascular Smooth Muscle Cells. Journal of Biological Chemistry. 2004;279(27):28766-70. 257. Paccez JD, Vogelsang M, Parker MI, Zerbini LF. The receptor tyrosine kinase Axl in cancer: Biological functions and therapeutic implications. International Journal of Cancer. 2014;134(5):1024-33. 258. Wong CCS, Lee WM. The proximal cis-acting elements Sp1, Sp3 and E2F regulate mouse mer gene transcription in Sertoli cells. European Journal of Biochemistry. 2002;269(15):3789-800. 259. Chu S, Ferro TJ. Sp1: Regulation of gene expression by phosphorylation. Gene. 2005;348:1-11. 260. Krishnamoorthy GP, Guida T, Alfano L, Avilla E, Santoro M, Carlomagno F, et al. Molecular Mechanism of 17-Allylamino-17-demethoxygeldanamycin (17-AAG)-induced AXL Receptor Tyrosine Kinase Degradation. The Journal of Biological Chemistry. 2013;288(24):17481-94. 261. Toshima J, Ohashi K, Iwashita S, Mizuno K. Autophosphorylation Activity and Association with Src Family Kinase of Sky Receptor Tyrosine Kinase. Biochemical and Biophysical Research Communications. 1995;209(2):656-63. 262. Li J, Soroka J, Buchner J. The Hsp90 chaperone machinery: Conformational dynamics and regulation by co-chaperones. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research. 2012;1823(3):624-35. 263. Shiojima I, Walsh K. Role of Akt Signaling in Vascular Homeostasis and Angiogenesis. Circulation Research. 2002;90(12):1243-50. 264. Laurance S, Aghourian MN, Jiva Lila Z, Lemarié CA, Blostein MD. Gas6-induced tissue factor expression in endothelial cells is mediated through caveolin-1–enriched microdomains. Journal of Thrombosis and Haemostasis. 2014;12(3):395-408. 265. Okazaki K, Sagata N. The Mos/MAP kinase pathway stabilizes c-Fos by phosphorylation and augments its transforming activity in NIH 3T3 cells. The EMBO Journal. 1995;14(20):5048-59. 266. Sagata N. What does mos do in oocytes and somatic cells? BioEssays. 1997;19(1):13-21. 267. Yew N, Strobel M, Woude GFV. Mos and the cell cycle: the molecular basis of the transformed phenotype. Current Opinion in Genetics & Development. 1993;3(1):19-25. 268. Fukasawa K, Vande Woude GF. Mos overexpression in Swiss 3T3 cells induces meiotic-like alterations of the mitotic spindle. Proceedings of the National Academy of Sciences. 1995;92(8):3430-4. 269. Wang XM, Yew N, Peloquin JG, Vande Woude GF, Borisy GG. Mos oncogene product associates with kinetochores in mammalian somatic cells and disrupts mitotic progression. Proceedings of the National Academy of Sciences of the United States of America. 1994;91(18):8329-33. 270. Wang G, Cheng Z, Liu F, Zhang H, Li J, Li F. CREB is a key negative regulator of carbonic anhydrase IX (CA9) in gastric cancer. Cellular Signalling. 2015;27(7):1369-79. 271. Nichols M, Weih F, Schmid W, DeVack C, Kowenz-Leutz E, Luckow B, et al. Phosphorylation of CREB affects its binding to high and low affinity sites: implications for cAMP induced gene transcription. The EMBO Journal. 1992;11(9):3337-46.

216

272. Lamph WW, Dwarki VJ, Ofir R, Montminy M, Verma IM. Negative and positive regulation by transcription factor cAMP response element-binding protein is modulated by phosphorylation. Proceedings of the National Academy of Sciences. 1990;87(11):4320-4. 273. Yue J, Ferrell JE. Mechanistic Studies of the Mitotic Activation of Mos. Molecular and Cellular Biology. 2006;26(14):5300-9. 274. Donella-Deana A, Cesaro L, Sarno S, Ruzzene M, Brunati AM, Marin O, et al. Tyrosine phosphorylation of protein kinase CK2 by Src-related tyrosine kinases correlates with increased catalytic activity. Biochemical Journal. 2003;372(Pt 3):841-9. 275. Gao Y, Wang H-y. Casein Kinase 2 Is Activated and Essential for Wnt/β-Catenin Signaling. Journal of Biological Chemistry. 2006;281(27):18394-400. 276. Nager M, Bhardwaj D, Cant, C, Medina L, Nogu, et al. B-Catenin Signalling in Glioblastoma Multiforme and Glioma-Initiating Cells. Chemotherapy Research and Practice. 2012;2012:7. 277. Sherbet GV. Metastasis promoter S100A4 is a potentially valuable molecular target for cancer therapy. Cancer Letters.280(1):15-30. 278. Cohn MA, Hjelmsø I, Wu L-C, Guldberg P, Lukanidin EM, Tulchinsky EM. Characterization of Sp1, AP-1, CBF and KRC binding sites and minisatellite DNA as functional elements of the metastasis-associated mts1/S100A4 gene intronic enhancer. Nucleic Acids Research. 2001;29(16):3335-46. 279. Kayahara M, Wang X, Tournier C. Selective Regulation of c-jun Gene Expression by Mitogen-Activated Protein Kinases via the 12-O-Tetradecanoylphorbol-13-Acetate- Responsive Element and Myocyte Enhancer Factor 2 Binding Sites. Molecular and Cellular Biology. 2005;25(9):3784-92. 280. Qiao X, Zhang L, Gamper AM, Fujita T, Wan Y. APC/C-Cdh1: From cell cycle to cellular differentiation and genomic integrity. Cell Cycle. 2010;9(19):3904-12. 281. Fukushima H, Ogura K, Wan L, Lu Y, Li V, Gao D, et al. SCF-Mediated Cdh1 Degradation Defines a Negative Feedback System that Coordinates Cell-Cycle Progression. Cell reports. 2013;4(4):10.1016/j.celrep.2013.07.031. 282. Gao D, Inuzuka H, Korenjak M, Tseng A, Wu T, Wan L, et al. Cdh1 Regulates Cell Cycle through Modulating the Claspin/Chk1 and the Rb/E2F1 Pathways. Molecular Biology of the Cell. 2009;20(14):3305-16. 283. Gutierrez GJ, Tsuji T, Chen M, Jiang W, Ronai ZeA. Interplay between Cdh1 and JNK activity during the cell cycle. Nature cell biology. 2010;12(7):686-95. 284. Berx G, van Roy F. Involvement of Members of the Cadherin Superfamily in Cancer. Cold Spring Harbor Perspectives in Biology. 2009;1(6):a003129. 285. Liu Y-N, Lee W-W, Wang C-Y, Chao T-H, Chen Y, Chen JH. Regulatory mechanisms controlling human E-cadherin gene expression. Oncogene. 2005;24(56):8277-90. 286. Khan MA, Chen H-c, Zhang D, Fu J. Twist: a molecular target in cancer therapeutics. Tumor Biol. 2013;34(5):2497-506. 287. Chiu R, Boyle WJ, Meek J, Smeal T, Hunter T, Karin M. The c-fos protein interacts with c-JunAP-1 to stimulate transcription of AP-1 responsive genes. Cell. 1988;54(4):541-52. 288. Eferl R, Wagner EF. AP-1: a double-edged sword in tumorigenesis. Nat Rev Cancer. 2003;3(11):859-68. 289. Xia Y, Yang W, Bu W, Ji H, Zhao X, Zheng Y, et al. Differential Regulation of c-Jun Protein Plays an Instrumental Role in Chemoresistance of Cancer Cells. The Journal of Biological Chemistry. 2013;288(27):19321-9. 290. Morton S, Davis RJ, McLaren A, Cohen P. A reinvestigation of the multisite phosphorylation of the transcription factor c-Jun. The EMBO Journal. 2003;22(15):3876-86.

217

291. Billen LP, Kokoski CL, Lovell JF, Leber B, Andrews DW. Bcl-XL Inhibits Membrane Permeabilization by Competing with Bax. PLoS Biol. 2008;6(6):e147. 292. Michels J, Kepp O, Senovilla L, Lissa D, Castedo M, Kroemer G, et al. Functions of BCL-XL at the Interface between Cell Death and Metabolism. International Journal of Cell Biology. 2013;2013:10. 293. Czabotar PE, Lessene G, Strasser A, Adams JM. Control of apoptosis by the BCL-2 protein family: implications for physiology and therapy. Nat Rev Mol Cell Biol. 2014;15(1):49-63. 294. Minn A, Rudin C, Boise L, Thompson C. Expression of bcl-xL can confer a multidrug resistance phenotype. Blood. 1995;86(5):1903-10. 295. Minn AJ, Boise LH, Thompson CB. Expression of Bcl-xL and loss of p53 can cooperate to overcome a cell cycle checkpoint induced by mitotic spindle damage. Genes & Development. 1996;10(20):2621-31. 296. Janumyan YM, Sansam CG, Chattopadhyay A, Cheng N, Soucie EL, Penn LZ, et al. Bcl‐xL/Bcl‐2 coordinately regulates apoptosis, cell cycle arrest and cell cycle entry. The EMBO Journal. 2003;22(20):5459-70. 297. Choi Yoon J, Li X, Hydbring P, Sanda T, Stefano J, Christie Amanda L, et al. The Requirement for Cyclin D Function in Tumor Maintenance. Cancer Cell. 2011;22(4):438-51. 298. Musgrove EA, Caldon CE, Barraclough J, Stone A, Sutherland RL. Cyclin D as a therapeutic target in cancer. Nat Rev Cancer. 2011;11(8):558-72. 299. Coller HA. What's taking so long? S-phase entry from quiescence versus proliferation. Nat Rev Mol Cell Biol. 2007;8(8):667-70. 300. Ekyalongo RC, Mukohara T, Kataoka Y, Funakoshi Y, Tomioka H, Kiyota N, et al. Mechanisms of acquired resistance to insulin-like growth factor 1 receptor inhibitor in MCF-7 breast cancer cell line. Investigational New Drugs. 2012;31(2):293-303. 301. Klein EA, Assoian RK. Transcriptional regulation of the cyclin D1 gene at a glance. Journal of Cell Science. 2008;121(23):3853-7. 302. Wei Q, Miskimins WK, Miskimins R. Sox10 acts as a tissue-specific transcription factor enhancing activation of the myelin basic protein gene promoter by p27Kip1 and Sp1. Journal of Neuroscience Research. 2004;78(6):796-802. 303. Daniel P, Filiz G, Brown DV, Hollande F, Gonzales M, D'Abaco G, et al. Selective CREB-dependent cyclin expression mediated by the PI3K and MAPK pathways supports glioma cell proliferation. Oncogenesis. 2014;3(6):e108. 304. Madrid LV, Mayo MW, Reuther JY, Baldwin AS. Akt Stimulates the Transactivation Potential of the RelA/p65 Subunit of NF-κB through Utilization of the IκB Kinase and Activation of the Mitogen-activated Protein Kinase p38. Journal of Biological Chemistry. 2001;276(22):18934-40. 305. Kane LP, Shapiro VS, Stokoe D, Weiss A. Induction of NF-κB by the Akt/PKB kinase. Current Biology. 1999;9(11):601-S1. 306. Umezawa K. Inhibition of tumor growth by NF-κB inhibitors. Cancer Science. 2006;97(10):990-5. 307. Ammoun S, Provenzano L, Zhou L, Barczyk M, Evans K, Hilton DA, et al. Axl/Gas6/NF[kappa]B signalling in schwannoma pathological proliferation, adhesion and survival. Oncogene. 2014;33(3):336-46. 308. Wang L, Harris TE, Roth RA, Lawrence JC. PRAS40 Regulates mTORC1 Kinase Activity by Functioning as a Direct Inhibitor of Substrate Binding. Journal of Biological Chemistry. 2007;282(27):20036-44. 309. Vines CM, Potter JW, Xu Y, Geahlen RL, Costello PS, Tybulewicz VL, et al. Inhibition of β2 Integrin Receptor and Syk Kinase Signaling in Monocytes by the Src Family Kinase Fgr. Immunity. 2001;15(4):507-19. 310. MacDonald BT, Tamai K, He X. Wnt/β-catenin signaling: components, mechanisms, and diseases. Developmental cell. 2009;17(1):9-26.

218

311. Luo J, Chen J, Deng Z-L, Luo X, Song W-X, Sharff KA, et al. Wnt signaling and human diseases: what are the therapeutic implications? Lab Invest. 2007;87(2):97-103. 312. Kim KH, Seol HJ, Kim EH, Rheey J, Jin HJ, Lee Y, et al. Wnt/β-catenin signaling is a key downstream mediator of MET signaling in glioblastoma stem cells. Neuro-Oncology. 2013;15(2):161-71. 313. Goruppi S, Chiaruttini C, Ruaro ME, Varnum B, Schneider C. Gas6 Induces Growth, β-Catenin Stabilization, and T-Cell Factor Transcriptional Activation in Contact-Inhibited C57 Mammary Cells. Molecular and Cellular Biology. 2001;21(3):902-15. 314. Salian-Mehta S, Xu M, Wierman ME. AXL and MET crosstalk to promote gonadotropin releasing hormone (GnRH) neuronal cell migration and survival. Molecular and Cellular Endocrinology. 2013;374(1–2):92-100. 315. Carling D, Sanders MJ, Woods A. The regulation of AMP-activated protein kinase by upstream kinases. Int J Obes. 2008;32(S4):S55-S9. 316. Xie Z, Dong Y, Scholz R, Neumann D, Zou M-H. Phosphorylation of LKB1 at Serine 428 by Protein Kinase C-ζ Is Required for Metformin-Enhanced Activation of the AMP-Activated Protein Kinase in Endothelial Cells. Circulation. 2008;117(7):952-62. 317. Elkabets M, Pazarentzos E, Juric D, Sheng Q, Pelossof Raphael A, Brook S, et al. AXL Mediates Resistance to PI3Kα Inhibition by Activating the EGFR/PKC/mTOR Axis in Head and Neck and Esophageal Squamous Cell Carcinomas. Cancer Cell. 2015;27(4):533-46. 318. Veronica H, Isabel R, Enrique B, Elvira A, Carmen S. Glucagon-Like Peptide-1 and Its Implications in Obesity. Fedele DM, editor2013. 319. Weinger JG, Gohari P, Yan Y, Backer JM, Varnum B, Shafit-Zagardo B. In brain, Axl recruits Grb2 and the p85 regulatory subunit of Pl3 kinase; in vitro mutagenesis defines th requisite binding sites for downstream Akt activation. Journal of neurochemistry. 2008;106(1):134-46. 320. Gao Y, Moten A, Lin H-K. Akt: a new activation mechanism. Cell Res. 2014;24(7):785-6. 321. Nielsen-Preiss SM, Allen MP, Xu M, Linseman DA, Pawlowski JE, Bouchard RJ, et al. Adhesion-Related Kinase Induction of Migration Requires Phosphatidylinositol-3-Kinase and Ras Stimulation of Rac Activity in Immortalized Gonadotropin-Releasing Hormone Neuronal Cells. Endocrinology. 2007;148(6):2806-14. 322. Ruan G-X, Kazlauskas A. Axl is essential for VEGF-A-dependent activation of PI3K/Akt. The EMBO Journal. 2012;31(7):1692-703. 323. Liu R, Gong M, Li X, Zhou Y, Gao W, Tulpule A, et al. Induction, regulation, and biologic function of Axl receptor tyrosine kinase in Kaposi sarcoma. Blood. 2010;116(2):297-305. 324. Demarchi F, Verardo R, Varnum B, Brancolini C, Schneider C. Gas6 Anti-apoptotic Signaling Requires NF-κB Activation. Journal of Biological Chemistry. 2001;276(34):31738-44. 325. Carpenter G, Ji Q-s. Phospholipase C-γ as a Signal-Transducing Element. Experimental Cell Research. 1999;253(1):15-24. 326. DeBell K, Graham L, Reischl I, Serrano C, Bonvini E, Rellahan B. Intramolecular Regulation of Phospholipase C-γ1 by Its C-Terminal Src Homology 2 Domain. Molecular and Cellular Biology. 2007;27(3):854-63. 327. Buchkovich NJ, Yu Y, Zampieri CA, Alwine JC. The TORrid affairs of viruses: effects of mammalian DNA viruses on the PI3K-Akt-mTOR signalling pathway. Nat Rev Micro. 2008;6(4):266-75. 328. Gouaze-Andersson V, Delmas C, Taurand M, Martinez-Gala J, Evrard S, Mazoyer S, et al. FGFR1 induces glioblastoma radioresistance through the PLCγ/HIF1α pathway. Cancer Research. 2016.

219

329. Latha K, Li M, Chumbalkar V, Gururaj A, Hwang Y, Dakeng S, et al. Nuclear EGFRvIII-STAT5b complex contributes to glioblastoma cell survival by direct activation of the Bcl-XL promoter. International journal of cancer Journal international du cancer. 2013;132(3):509-20. 330. Park I-K, Trotta R, Yu J, Caligiuri MA. Axl/Gas6 pathway participates in human natural killer cell development by positively regulating FLT3 activation. European journal of immunology. 2013;43(10):10.1002/eji.201243116. 331. Cao S, Wang C, Zheng Q, Qiao Y, Xu K, Jiang T, et al. STAT5 regulates glioma cell invasion by pathways dependent and independent of STAT5 DNA binding. Neuroscience Letters. 2011;487(2):228-33. 332. Lev S, Moreno H, Martinez R, Canoll P, Peles E, Musacchio JM, et al. Protein tyrosine kinase PYK2 involved in Ca2+-induced regulation of ion channel and MAP kinase functions. Nature. 1995;376(6543):737-45. 333. Taniyama Y, Weber DS, Rocic P, Hilenski L, Akers ML, Park J, et al. Pyk2- and Src-Dependent Tyrosine Phosphorylation of PDK1 Regulates Focal Adhesions. Molecular and Cellular Biology. 2003;23(22):8019-29. 334. Gao C, Chen G, Kuan S-F, Zhang DH, Schlaepfer DD, Hu J. FAK/PYK2 promotes the Wnt/β-catenin pathway and intestinal tumorigenesis by phosphorylating GSK3β. eLife. 2015;4:e10072. 335. Sabri A, Govindarajan G, Griffin TM, Byron KL, Samarel AM, Lucchesi PA. Calcium- and Protein Kinase C–Dependent Activation of the Tyrosine Kinase PYK2 by Angiotensin II in Vascular Smooth Muscle. Circulation Research. 1998;83(8):841-51. 336. Xu B-e, Lee B-H, Min X, Lenertz L, Heise CJ, Stippec S, et al. WNK1: analysis of protein kinase structure, downstream targets, and potential roles in hypertension. Cell Res. 2005;15(1):6-10. 337. Li Y, Ye X, Tan C, Hongo JA, Zha J, Liu J, et al. Axl as a potential therapeutic target in cancer: role of Axl in tumor growth, metastasis and angiogenesis. Oncogene. 2009;28(39):3442-55. 338. Kim YS, Jung S-H, Jung DH, Choi S-J, Lee Y-R, Kim JS. Gas6 Stimulates Angiogenesis of Human Retinal Endothelial Cells and of Zebrafish Embryos via ERK1/2 Signaling. PLoS ONE. 2014;9(1):e83901. 339. Fukumura D, Gohongi T, Kadambi A, Izumi Y, Ang J, Yun C-O, et al. Predominant role of endothelial nitric oxide synthase in vascular endothelial growth factor-induced angiogenesis and vascular permeability. Proceedings of the National Academy of Sciences of the United States of America. 2001;98(5):2604-9. 340. Thompson EM, Frenkel EP, Neuwelt EA. The paradoxical effect of bevacizumab in the therapy of malignant gliomas. Neurology. 2011;76(1):87-93. 341. Verhoeff JJ, van Tellingen O, Claes A, Stalpers LJ, van Linde ME, Richel DJ, et al. Concerns about anti-angiogenic treatment in patients with glioblastoma multiforme. BMC Cancer. 2009;9(1):1-9. 342. Gal A, Li Y, Thompson DA, Weir J, Orth U, Jacobson SG, et al. Mutations in MERTK, the human orthologue of the RCS rat retinal dystrophy gene, cause retinitis pigmentosa. Nat Genet. 2000;26(3):270-1. 343. Rothlin CV, Ghosh S, Zuniga EI, Oldstone MBA, Lemke G. TAM Receptors Are Pleiotropic Inhibitors of the Innate Immune Response. Cell. 2007;131(6):1124-36. 344. Sharif MN, Šošić D, Rothlin CV, Kelly E, Lemke G, Olson EN, et al. Twist mediates suppression of inflammation by type I IFNs and Axl. The Journal of Experimental Medicine. 2006;203(8):1891-901. 345. Bosurgi L, Bernink JH, Delgado Cuevas V, Gagliani N, Joannas L, Schmid ET, et al. Paradoxical role of the proto-oncogene Axl and Mer receptor tyrosine kinases in colon cancer. Proceedings of the National Academy of Sciences of the United States of America. 2013;110(32):13091-6.

220

346. Caraux A, Lu Q, Fernandez N, Riou S, Di Santo JP, Raulet DH, et al. Natural killer cell differentiation driven by Tyro3 receptor tyrosine kinases. Nat Immunol. 2006;7(7):747-54. 347. Waldhauer I, Steinle A. NK cells and cancer immunosurveillance. Oncogene. 2008;27(45):5932-43. 348. Paolino M, Choidas A, Wallner S, Pranjic B, Uribesalgo I, Loeser S, et al. The E3 ligase Cbl-b and TAM receptors regulate cancer metastasis via natural killer cells. Nature. 2014;507(7493):508-12. 349. Ou WB, Corson JM, Flynn DL, Lu WP, Wise SC, Bueno R, et al. AXL regulates mesothelioma proliferation and invasiveness. Oncogene. 2011;30(14):1643-52. 350. Zhang Y-X, Knyazev PG, Cheburkin YV, Sharma K, Knyazev YP, Őrfi L, et al. AXL Is a Potential Target for Therapeutic Intervention in Breast Cancer Progression. Cancer Research. 2008;68(6):1905-15. 351. Ye X, Li Y, Stawicki S, Couto S, Eastham-Anderson J, Kallop D, et al. An anti-Axl monoclonal antibody attenuates xenograft tumor growth and enhances the effect of multiple anticancer therapies. Oncogene. 2010;29(38):5254-64. 352. Cerchia L, Esposito CL, Camorani S, Rienzo A, Stasio L, Insabato L, et al. Targeting Axl With an High-affinity Inhibitory Aptamer. Mol Ther. 2012;20(12):2291-303. 353. Rho JK, Choi YJ, Kim SY, Kim TW, Choi EK, Yoon S-J, et al. MET and AXL Inhibitor NPS-1034 Exerts Efficacy against Lung Cancer Cells Resistant to EGFR Kinase Inhibitors Because of MET or AXL Activation. Cancer Research. 2014;74(1):253-62.

221

APPENDIX

222

Appendix 1

Western blot showing timecourse of Axl phosphorylation by Gas6 (400 ng/ml)

(A), of Tyro3 phosphorylation (B) and of Akt levels (C) in SNB-19 and UP007

cells

223

Appendix 2

Western blots showing inhibition by BGB324 (0.1-100 µM) of phosphorylation

of Axl (A), Tyro3 (B) and Akt kinase downstream (C) stimulated by Gas6 (400

ng/ml) in SNB-19 an UP007 cells.

224

Appendix 3

Western blot of EGFR phosphorylation by EGF and inhibition by gefitinib and

BGB324 in SNB-19 and UP007 cells. Data are mean±SEM (n=3 separate

experiments); ANOVA with Tukey’s multiple comparisons post hoc analysis,

**p<0.01, *p<0.05, ns, not significant, for comparisons indicated

The role of TAM receptors in brain tumour cell signalling ... · in brain tumour cell signalling...

Documents

Transcript of The role of TAM receptors in brain tumour cell signalling ... · in brain tumour cell signalling...