Targeting the Notch-1/IGF-1R/Akt Axis in at Orthotopic Model of Advanced Non-Small Cell Lung Cancer

Loyola University Chicago Loyola eCommons

Dissertations Theses and Dissertations

2013

Shuang Liang Loyola University Chicago

Follow this and additional works at: https://ecommons.luc.edu/luc_diss

Part of the Molecular Biology Commons

Recommended Citation Liang, Shuang, "Targeting the Notch-1/IGF-1R/Akt Axis in at Orthotopic Model of Advanced Non-Small Cell Lung Cancer" (2013). Dissertations. 675. https://ecommons.luc.edu/luc_diss/675

This Dissertation is brought to you for free and open access by the Theses and Dissertations at Loyola eCommons. It has been accepted for inclusion in Dissertations by an authorized administrator of Loyola eCommons. For more information, please contact [email protected].

LOYOLA UNIVERSITY CHICAGO

TARGETING THE NOTCH-1/IGF-1R/AKT AXIS IN AN ORTHOTOPIC

MODEL OF ADVANCED NON-SMALL CELL LUNG CANCER

A DISSERTATION SUBMITTED TO

THE FACULTY OF THE GRADUATE SCHOOL

IN CANDIDACY FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

PROGRAM IN MOLECULAR BIOLOGY

SHUANG LIANG

CHICAGO, ILLINOIS

AUGUST 2013

ACKNOWLEDGEMENTS

I would like to express the deepest appreciation to my mentor Dr. Maurizio

Bocchetta, who has been a tremendous mentor for me. I would like to thank him for encouraging my research and for allowing me to growth as a research scientist. Without his guidance and persistent help this dissertation would not have been possible.

I would also like to thank my dissertation committee, Dr. Caroline Le Poole, Dr.

William Simmons, Dr. Clodia Osipo, Dr. Manuel Diaz and Dr. Maurizio Bocchetta for serving as my committee members even at hardship. I also want to thank them for letting my defense be an enjoyable moment, and for their brilliant comments and suggestions.

I would especially like to thank Dr. Bocchetta’s lab members: Sandra Eliasz,

Sylvia Skucha, Paola Galluzzo, Anna Sobol, and Brittany Rambo. They have been always there to support me both scientifically and mentally during my Ph.D. study.

Finally, I would like to thank my family. Words cannot express how grateful I am to my mother-in law, father-in-law, my mother, and father for all of the sacrifices that they’ve made on my behalf. I would also like to thank to my beloved husband, Zhenyu

Zhong. Thanks for supporting me for everything, and especially I can’t thank him enough for encouraging me throughout this experience. To my beloved son Eric Haotian

Zhong, I would like to express my thanks for being such a good boy always cheering me up.

iii

To my family

TABLE OF CONTENTS

ACKNOWLEDGMENTS iii

LIST OF TABLES vii

LIST OF FIGURES viii

LIST OF ABBREVIATIONS x

ABSTRACT xv

CHAPTER 1: INTRODUCTION 1 Lung cancer and treatment 1 Hypoxic tumor microenvironment 6 Hypoxia-induced signaling 6 Hypoxia and cancer 7 The Notch signaling pathway 10 The Notch signaling cascade 10 Regulation of Notch by hypoxia 14 The role of the Notch pathway in lung cancer 14 The IGF-1R signaling pathway 16 Ligands and receptor signaling 16 The role of the IGF1-R pathway in NSCLC and other tumors 19 IGF-1R as a target for cancer treatment 21 The EGF-R signaling pathway 23 An overview of the EGF-R pathway 23 Endocytic sorting of EGF-R 24 Effects of ligands on EGF-R endocytic sorting 25

CHAPTER 2: RATIONALE AND DISSERTATION OUTLINE 27

CHAPTER 3: MATERIALS AND METHODS 36 Cell culture 36 Reagents 36 Mouse procedures 37 Cell Lysis, Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and Western Blotting 39 Chromatin Immunoprecipitation (ChIP) 40 PCR and quantitative PCR (qPCR) for Chromatin Immunoprecipitation assay 42 Immunofluorescent Microscopy 43 Plasmids and transfections 44 Lentiviral systems 45 Receptor tyrosine kinase (RTK) arrays 46 v RNA isolation and cDNA preparation 47 Quantitative RT-PCR (qRT-PCR) 48 Histological study (Haematoxylin and eosin) 50 Protein phosphorylation expression and analysis 51 Cell surface biotinylation 51 Cell viability and proliferation 52 Statistical analysis 53

CHAPTER 4: RESULTS 56 Notch-1 directly regulates IGF-1R transcription through its association with the +1478 DNA region 57 Establishment of mouse orthotopic models with advanced NSCLC 62 Hypoxia/Notch-1/IGF-1R/Akt-1 axis also takes place in vivo 65 Aim1: Determine whether inhibition of the Notch-1/IGF-1R/Akt axis is a suitable therapeutic tool to target hypoxic NSCLC 68 Survival analysis of mice treated with GSI, IGF-1R inhibitory antibody or pan Akt inhibitor, alone or in combination 68 GSI treatment resulted in the effective killing of tumors in hypoxic areas 80 GSI treatment reduced brain and liver metastasis 83 GSI treatment decreased the expression of glycolytic enzymes 84 Aim2: Delineate the detailed mechanism by which tumor evasion occurred during IGF-1R inhibitory antibody treatment 85 MK-0646 promoted IGF-1R degradation in a proteasome dependent manner 85 Tumor evasion occurred after IGF-1R inhibitory antibody (MK- 0646) treatment 86 Phosphorylated EGF-R (pEGF-R) and EGF-R total protein levels were increased upon MK-0646 treatment in vivo and in vitro 87 Polyubiquitinated EGF-R was reduced and EGF-R protein accumulated in the plasma membrane upon MK-0646 treatment 91 MK-0646 treatment increased the expression of EGF-R ligand amphiregulin by activating Akt-1 signaling 92 Aim3: Study if administration of IGF-1R inhibitory antibody sensitizes NSCLC cells to EGF-R inhibition 96 MK-0646 treated NSCLC cells became sensitive to EGF-R inhibiton in vitro 96 Administration of MK-0646 followed by EGF-R inhibitor (erlotinib) had improved efficacy in treating advanced NSCLC in vivo 98

CHAPTER 5: DISCUSSION 100

BIBLIOGRAPHY 111

VITA 131

LIST OF TABLES

Table Page

1.1. The TNM classification for staging of non-small cell lung cancer (NSCLC) 3 1.2. Standard treatment options for NSCLC and 5-year survival rates according to stage 5 3.1. Primers used for human and mouse specific GAPDH 38 3.2. Primers used for Chromatin Immunoprecipitation (ChIP) analysis 42 3.3. Primers used for quantitative RT-PCR (qRT-PCR) analysis 48 3.4. List of all antibodies used in this dissertation work 54 4.1. Quantitative PCR analysis of the human/mouse GAPDH ratio in the whole left lungs of 12 representative animals 5-weeks after tail vein injection 64 4.2. Statistical analysis of the mice survival curves shown in Figure 4.8 72

vii

LIST OF FIGURES

Figure Page

1.1. The Notch signaling pathway 13 1.2. The IGF-1R signaling pathway 18 2.1. Dissertation outline 35 4.1. Schematic of Notch-1 activation in hypoxia, which provides survival signals to NSCLC cells through activating Akt-1 56 4.2. Schematic representation of the promoter region of the IGF-1R gene 59 4.3. Notch-1 associates with the +1478 region of the IGF-1R gene 60 4.4. Notch-1, MAML-1 and p300 associate with DNA region +1478 of IGF-1R, and transfection of NSCLC cells with plasmid DN- MAML-1 disrupts these associations 61 4.5. Characterization of lung tumors in advanced NSCLC mice model at the time of therapy initiation (5-weeks after tail vein injection) 63 4.6. GLUT-1 is an effective hypoxia marker in NSCLC cells 66 4.7. Activated Notch-1 (Notch-1IC) is expressed in hypoxic tumor cells, together with IGF-1R and phosphorylated Akt-1 (Ser473) but not with PTEN 67 4.8. Survival curves of mice injected with A549 and H1437 cells treated with indicated agents 71 4.9. GSI treatment caused significant reduction in the expression of Notch target genes, hypoxic markers and IGF-1R 81 4.10. GSI treatment led to apoptosis of hypoxic NSCLC cells 82 4.11. GSI treatment caused significant reduction in brain and liver metastasis 83 4.12. GSI treatment decreased the expression of glycolytic enzymes 84 viii

4.13. MK-0646 promoted IGF-1R degradation in a proteasome dependent manner 86 4.14. Bioluminescence quantification of the tumor burden in the lungs of mice treated with IGF1-R inhibitory antibody (MK-0646) 87 4.15. Phosphorylated EGF-R (pEGF-R) was increased in IGF-1R inhibitory antibody (MK-0646) treated mouse lung compared to control 89 4.16. MK-0646 treatment increased total EGF-R protein and phosphorylated EGF-R (pEGF-R) levels both in vitro and in vivo 90 4.17. MK-0646 treated NSCLC cells showed increased pEGF-R in the plasma membrane and cytosol, increased EGF-R total protein only in the plasma membrane, and reduced EGF-R polyubiquitination 92 4.18. MK-0646 treatment led to Akt activation and increased ampheregulin expression 94 4.19. The expression level of amphiregulin is regulated by Akt signaling 95 4.20. NSCLC cells pre-treated with MK-0646 became sensitive to EGF- R inhibitor (erlotinib) 97 4.21. NSCLC cells pre-treated with MK-0646 became sensitive to EGF- R downregulation 98 5.1. EGF-R surface expression remained the same one-week after sorting 109

LIST OF ABBREVIATIONS

7-AAD 7-aminoactinomycin D

ACL Adenocarcinoma of the lung

ADAM10 A disintegrin and metalloproteinase domain-containing protein 10

ALDOA Aldolase-A

APH1 Anterior pharynx-defective 1

AR Amphiregulin

ARNT Aryl hydrocarbon receptor nuclear translocator

BTC β-cellulin

CBF-1 C-promoter binding factor 1

Cbl Casitas b-lineage lymphoma

ChIP Chromatin immunoprecipitation

ECL Enhanced chemiluminescence

EGF Epidermal growth factor

EGF-R Epidermal growth factor receptor

EMT Epithelial-mesenchymal transition

EPI Epiregulin

EPS15 Epidermal growth factor receptor substrated 15

ERK Extracellular signal regulated kinase

ESCRT Endosomal sorting complexes required for transport

FACS Fluorescence activated cell sorting

FBS Fetal bovine serum

FGF-R3 Fibroblast growth factor receptor 3

FIH-1 Factor inhibiting HIF-1

GAPDH Glyceraldehyde 3-phosphate dehydrogenase

GFP Green fluorescent protein

GLUT-1 Glucose transporter 1

GPCRs G-protein coupled receptors

Grb2 Growth factor receptor-bound protein 2

GSI γ-secretase inhibitor

HB-EGF Heparin-binding EGF

HER2 Human epidermal growth factor receptor 2

HES Hairy enhancer of split

HIF-1α Hypoxia inducible factor-1α

HIFs Hypoxia inducible factors

HK-2 Hexokinase 2 hr Hour

HRE Hypoxia response elements

HRS Hepatocyte growth factor-regulated tyrosine kinase substrate

HRT Hairy-related transcription factor i.p. Intraperitoneal

xi IGF-1 Insulin like growth factor-1

IGF-1R Insulin like growth factor-1 receptor

IGFBPs IGF binding proteins

ILVs Intraluminal vesicles

IR Insulin receptor

IRS Insulin receptor substrate

MAML-1 Mastermind like-1

MAPK Mitogen activated protein kinase min Minute ml Milli-liter mM Milli-molar

MMR Mismatch repair mTOR Mammalian target of rapamycin mTORC-1 Mammalian target of rapamycin kinase complex 1

MVBs Multivesicular bodies

NaCl Sodium chloride

NCSTN Nicastrin

NH4Cl Ammonium chloride

NICD Notch intracellular domain nm Nanometer nM Nanomolar

NSCLC Non-small cell lung cancer

Oct-4 Octamer-binding transcription factor 4

xii PAGE Poly-acrylamide gel electrophoresis

PBS Phosphate buffered saline

PDK-1 Phosphoinositide-dependent kinase-1

PDK-2 Pyruvate dehydrogenase kinase 2 pEGF-R Phosphorylated epidermal growth factor receptor

PEN2 Presenilin enhancer 2

PHD Prolyl hydroxylase-domain protein

PI3K Phosphatidylinositol 3-kinase

PSEN Presenilin

PTEN Phosphatase and tensin homolog

Raf Rapidly accelerated fibrosarcoma

Ras Rat sarcoma

REDD-1 Regulated in development and DNA damage responses 1

ROS Reactive oxygen species

RT Room temperature

RT-PCR Reverse transcription polymerase chain reaction

RTK Receptor tyrosine kinase s Second

SCLC Small cell lung cancer

SDS Sodium dodecyl sulfate

SHC Src homology 2 domain containing transforming protein

SOS Son of Sevenless

STAT Signal transducers and activators of transcription

xiii SV40 Simian virus 40

T-ALL T-cell acute lymphoblastic leukemia/lymphoma

TACE Tumor necrosis factor-α-converting enzyme

TBS Tris-buffered saline

TBS-T Tris-buffered saline with Tween-20

TGFα Transforming growth factor α

TGFβ Transforming growth factor β

TKI Tyrosine kinase inhibitor

TUNEL Terminal deoxynucleotidyl transferase dUTP nick end labeling

VEGF-A Vascular endothelial growth factor A

VHL Von Hippel-Lindau

µg Microgram

µl Microliter

µM Micromolar

xiv

ABSTRACT

Lung cancer is the leading cause of cancer death in the U.S. and worldwide. The most frequent type of lung cancer is non-small cell lung cancer (NSCLC). Among the three major histological types of NSCLC, adenocarcinoma of the lung (ACL) is the most prevalent. NSCLC is mostly diagnosed at advanced stages (stage IIIB 18% of cases, stage

IV 40% of cases) due to the lack of effective early detection methods. The median survival of such patients is only 12.3 months when treated with carboplatinum plus taxol and the antiangiogenic agent bevacizumab (Sandler et al., 2006). Thus, the discovery of alternative therapeutic strategies is of extreme importance.

Others and we have previously found that Notch signaling plays a crucial role in

NSCLC. Inhibition of Notch-1 signaling in NSCLC cells causes cell death specifically under hypoxia. Our preliminary results indicate that Notch-1 provides necessary survival signals to NSCLC cells by positively regulating insulin-like growth factor-1 receptor

(IGF-1R) to activate the Akt-1 pathway. This effect is exacerbated by the fact that Notch-

1 represses phosphatase and tensin homolog (PTEN) expression (Eliasz et al., 2010).

Therefore, the overall hypothesis is that an inhibitor of Notch activation, such as γ- secretase inhibitor (GSI) MRK-003; or a fully humanized antibody against the human

IGF-1R that promotes the receptor degradation (IGF-1R inhibitory antibody) MK-0646; or a pan-Akt inhibitor MK-2206, alone or in combinations, could improve the efficacy of chemotherapy for advanced NSCLC by specifically targeting hypoxic NSCLC tumor

xv environment. We tested this hypothesis in mouse preclinical models of orthotopic, advanced (metastatic) NSCLC. Our results indicate that all treatments (with the exception of MK-2206, which caused lethal glucose intolerance), significantly improved the survival of mice compared to controls. The combination therapies of GSI plus cisplatin and GSI plus IGF-1R inhibitory antibody significantly improved median survival compared to single agents. The expression levels of hypoxic markers were reduced in

GSI treated mice lungs, indicating that GSI treatment caused specific cell death of hypoxic tumor areas. GSI treatment also significantly reduced cancer metastasis to the liver and brain. MK-0646 had tumoristatic effect at the beginning of the treatment.

However, tumor cells resumed growth and eventually led to the death of the experimental animals. We found that tumor evasion from MK-0646 treatment was mediated by overactivated epidermal growth factor receptor (EGF-R) signaling pathway, which is due to stabilization of the total EGF-R protein and Akt-1 mediated upregulation of

Amphiregulin, an EGF-R ligand. Of relevance, the sensitivity of NSCLC cells to erlotinib

(EGF-R inhibitor) was increased after MK-0646 treatment. Sequential therapy of MK-

0646 followed by erlotinib significantly prolonged survival of the mice compared to each single agent or simultaneous administration of two drugs.

Together, these data indicate the following: Notch inhibition seems to specifically target tumor hypoxic regions in vivo and to significantly inhibit tumor metastasis. We also found serendipitously that the administration of IGF-1R inhibitory antibody sensitizes NSCLC to erlotinib, a second-line therapeutic agent whose sensitivity is

xvi difficult to predict, with the notable exception of NSCLC with activating mutations in the

EGF-R itself.

xvii CHAPTER 1

INTRODUCTION

LUNG CANCER AND TREATMENT

Being the second most common type of cancer, lung cancer is the leading cause of cancer related deaths in the U.S. and worldwide, in both men and women. Each year, the number of people who have died from lung cancer is more than that of colon, breast, and prostate cancers combined. For the United States in 2013, it is estimated that there will be 228,190 new cases and 159,480 deaths from lung cancer, which will account for about 27% of all cancer related deaths (Siegel et al., 2013). Based on major histological features, lung cancer is divided into two major types: small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). Accounting for about 87.5% of diagnosed lung cancer, NSCLC is the most common type of lung cancer. NSCLC is further grossly sub- characterized into three major histological types: squamous cell carcinoma, large-cell carcinoma and adenocarcinoma of the lung. Adenocarcinoma of the lung (ACL) is the most common type of NSCLC, which accounts for 40% of all lung cancer cases. It is also the most prevalent lung cancer in women and in individuals who have never smoked

(Shigematsu et al., 2005). It seems to originate at the broncho-alveolar junction of the lungs, and often spreads to spaces between the lungs and the chest wall, which makes it more difficult to detect at an early stage (Galluzzo and Bocchetta, 2011). Early stage lung cancer often does not cause any symptoms. In most cases, the tumor cells have

1 2 already spread to other parts of the body before it can be detected with an imaging test such as a chest x-ray. Therefore, most NSCLC cases are diagnosed at advanced disease stages. Based on the size of the lung tumor, and whether cancer cells have spread to lymph nodes or other tissues, lung cancer can be diagnosed at four different stages (from stage I to IV, see Table 1.1 of NSCLC staging with clinical definition). The overall 5- year survival rate for NSCLC cancer is only 15%. The NSCLC standard treatment strategies and survival rates for each stage are summarized in Table 1.2 (Ries, 2007). Due to the lack of effective early detection methods, 18% of the NSCLC patients are diagnosed at stage IIIB and 40% of patients are diagnosed at stage IV (Wisnivesky et al.,

2005). Stage IIIB and IV are the advanced stages (metastatic) in which cancer has spread from the chest to other parts of the body, such as the brain, bones, liver and the other lung

(Goldstraw et al., 2007). The median survival of such patients is only 12.3 months when treated with platinum plus taxol and Avastin (Bevacizumab), a monoclonal antibody against vascular endothelial growth factor A (VEGF-A) (Sandler et al., 2006). Moreover, recent studies indicated that inhibitors targeting the VEGF pathway might elicit tumor adaptation, increasing tumor invasiveness, and hence promote tumor progression to higher malignant stages (Paez-Ribes et al., 2009). Therefore, the introduction of effective screening methods and finding of alternative therapeutic strategies are of extreme importance.

Table 1.1 The TNM classification for staging of non-small cell lung cancer (NSCLC). The staging system was developed by the American Joint Commission on Cancer (Goldstraw et al., 2007).

N (Regional lymph M (Distant Stage T (Primary tumor) nodes) metastasis) T1a: Tumor ≤ 2 cm in the N0: No regional node M0: No distant greatest dimension metastasis metastasis IA T1b: Tumor > 2 cm but ≤ 3 N0 M0 cm in the greatest dimension T2a: Tumor > 3 cm but ≤ 5 IB N0 M0 cm in the greatest dimension N1: Metastasis in ipsilateral peribronchial and/or ipsilateral hilar T1a lymph nodes and M0 intrapulmonary nodes, IIA including involvement by direct extension T1b N1 M0 T2a N1 M0 T2b: Tumor > 5 cm but ≤ 7 N0 M0 cm in the greatest dimension T2b N1 M0 T3: Tumor > 7 cm or one that directly invades any of the following: • Chest wall, diaphragm, phrenic nerve, mediastinal pleura, or parietal pericardium IIB Tumor in the main • N0 M0 bronchus < 2 cm distal to the carina but without involvement of the carina • Associated atelectasis/obstructive pneumonitis of the entire lung or separate tumor nodule(s) in the same lobe T1: Tumor ≤ 3 cm in greatest N2: Metastasis in the IIIA dimension, surrounded by ipsilateral mediastinal M0 lung or visceral pleura, no and/or subcarinal lymph

bronchoscopic evidence of node(s) invasion, more proximal than the lobar bronchus; superficial spreading of tumor in the central airways T2: Tumor > 3 cm but ≤ 7 cm or tumor with any of the following: • Invades visceral pleura • Involves the main bronchus ≥ 2 cm distal to N2 M0 the carina • Associated with atelectasis/obstructive pneumonitis extending to hilar region but not involving the entire lung T3 N2 M0 T3 N1 M0 T4: Tumor of any size that invades any of the following: mediastinum, heart, great vessels, trachea, recurrent N0 M0 laryngeal nerve, esophagus, vertebral body, or carina; separate tumor nodule(s) in a different ipsilateral lobe T4 N1 M0 T4 N2 M0 N3: Metastasis in the contralateral mediastinal, contralateral hilar, T1 ipsilateral or contralateral M0 IIIB scalene, or supraclavicular lymph nodes T2 N3 M0 T3 N3 M0 T4 N3 M0 M1a: Separate tumor nodule(s) IV T Any N Any in a contralateral lobe; tumor with

pleural nodules or malignant pleural (or pericardial) effusion M1b: Distant T Any N Any metastasis

Table 1.2 Standard Treatment Options for NSCLC and 5-year survival rates according to stage. The treatment strategies for NSCLC are obtained from the NIH website. The 5-year survival rates were calculated from the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) database, based on people who were diagnosed with non-small cell lung cancer between 1998 and 2000 (Ries, 2007).

Survival Stage Standard Treatment Options Rate (5Y) Surgery IA: 49% IA and IB Radiation therapy IB: 45% Surgery IIA: 30% Chemotherapy followed by surgery IIA and IIB Surgery followed by chemotherapy IIB: 31% External radiation therapy Surgery followed by chemotherapy Chemotherapy followed by surgery Resectable Surgery followed by chemotherapy combined with disease radiation therapy Surgery followed by radiation therapy Chemotherapy and radiation therapy Unresectable External radiation therapy alone disease Internal radiation therapy or laser surgery III Radiation therapy alone 14% A Superior Radiation therapy followed by surgery sulcus tumors Chemotherapy and radiation followed by surgery Surgery alone Surgery Tumors that Surgery and radiation therapy invade the Radiation therapy alone chest wall Chemotherapy combined with radiation therapy and/or surgery Chemotherapy followed by external radiation IIIB therapy. 5% Chemotherapy and radiation therapy Chemotherapy

followed by surgery External radiation therapy alone External or internal radiation therapy Combination chemotherapy Maintenance therapy with an anticancer drug after combination chemotherapy Combination chemotherapy and targeted therapy IV 1% with a monoclonal antibody Targeted therapy with a tyrosine kinase inhibitor External radiation therapy Laser therapy and/or internal radiation therapy

HYPOXIC TUMOR MICROENVIRONMENT

Hypoxia-induced signaling

Hypoxia-induced signaling is primarily mediated by hypoxia-inducible factors

(HIFs), including HIF-1 and HIF-2 (Bruick, 2003; Semenza, 1999). HIF is a heterodimer containing a constitutively expressed HIF-β subunit (also known as ARNT, the aryl hydrocarbon receptor nuclear translocator) and HIF-α subunit, which is regulated and stabilized by low level of oxygen (Wang and Semenza, 1995). When HIF-α is stabilized, it subsequently dimerizes with the β subunit, and the heterodimer functions as a transcription factor that binds to the specific hypoxia response elements (HREs) on the promoters of its target genes, mediating the transcription of hypoxia responsive targets.

The regulation of HIF-1α expression by oxygen is at the level of protein degradation.

HIF-1α can be recognized by E3 ubiquitin ligase Von Hippel-Lindau (VHL) tumor- suppressor protein and targeted for proteasomal degradation only when proline residues

402 and 564 are hydroxylated (Ivan et al., 2001). The hydroxylation reaction requires the presence of oxygen, and is catalyzed by the PHD 1-3 enzymes (prolyl hydroxylase-

7 domain protein 1-3) (Bruick and McKnight, 2001). In addition to HIF-1α stabilization, the level of oxygen also regulates the interaction between HIF-1α and its transcriptional coactivators P300 and CBP. In the presence of oxygen, FIH-1 (factor inhibiting HIF-1) mediated hydroxylation of asparagine residue 803 of HIF-1α blocks the binding between

HIF-1α and P300/CBP, therefore inhibiting the transcription of HIF-1α downstream target genes (Lando et al., 2002; Mahon et al., 2001). Similar to HIF-1α, the protein level of HIF-2α and its transcriptional activity are also regulated by hydroxylation of proline and asparagine residues. The regulation of HIF-1α and HIF-2α by the level of oxygen is universal to almost all types of mammalian cells. Both HIF-α and HIF-β have been linked with the molecular mechanisms that regulate carbohydrate and lipid metabolism

(Pescador et al., 2010), erythropoiesis and iron metabolism (Haase, 2010), and angiogenesis (Pugh and Ratcliffe, 2003). Furthermore, HIF-1 and HIF-2 are also important in regulating embryonic stem cells in early embryo development, which is a naturally occurring hypoxic environment due to the lack of fully established vasculature

(Simon and Keith, 2008). Due to a complete lack of cephalic vascularization, HIF-1α knockout mice are embryonic lethal (Ryan et al., 1998).

Hypoxia and cancer

In solid tumors, there are always regions with low levels of oxygen, creating a hypoxic tumor microenvironment. This is due to the fact that tumor cells proliferate quickly and the vasculature is not completely established. Hypoxia is often associated

8 with malignant progression and poor clinical outcomes in advanced cancers (Brizel et al.,

1999; Brizel et al., 1996; Hockel et al., 1996; Nordsmark and Overgaard, 2004). This is at least partially due to the low level of oxygen counteracting the effects of chemotherapy and radiation therapy, which kill tumor cells by targeting highly proliferating cells and generating reactive oxygen species, respectively (Brown, 2007).

Hypoxia induces a wide range of biological changes that contribute to the progression of malignant tumor cells. It has been shown that hypoxia is associated with activation of a number of genes that promote cell proliferation, survival, motility, cell adhesion and angiogenesis (Harris, 2002). Hypoxia has been implicated in enhancing genomic instability, which contributes to tumorigenesis and malignant progression, owing to diminished DNA repair, elevated mutagenesis and genomic rearrangements.

HIF-1α is able to interfere with the DNA repair mechanism by downregulating

MutSalpha expression, which recognizes base mismatches (Koshiji et al., 2005).

Moreover, the expression of DNA mismatch repair (MMR) gene Mlh1 is specifically reduced in cells under hypoxia due to histone deacetylation (Mihaylova et al., 2003).

Furthermore, hypoxia leads to genome rearrangements and mutagenesis due to the induction of chromosome fragile sites (Coquelle et al., 1998; Yuan et al., 2000). The induction of the expression level of drug resistant genes, such as multidrug resistance transporter P-glycoprotein (P-gp) by HIF-1α can ultimately lead to tumor resistance to chemotherapeutics (Comerford et al., 2002; Wartenberg et al., 2003). More importantly, recent studies reported that hypoxia also plays an important role in expending or maintaining cancer stem cell or the tumor initiating cell population eventually leading to

9 tumor progression (Keith and Simon, 2007). Hypoxia, through the mediation of HIF-1α or HIF-2α, was shown to induce the expression of genes that regulate the differentiation and homing of stem or progenitor cells, including Notch (Gustafsson et al., 2005; Yuan et al., 2000), beta-catenin (Kaidi et al., 2007), Oct-4 (Covello et al., 2006) and c-Myc

(Gordan et al., 2007). The Notch intracellular domain (NICD) has been shown to physically interact with and stabilized by HIF-1α under hypoxia, hence activating the downstream signaling that is critical in maintaining an undifferentiated cell status

(Gustafsson et al., 2005). Additionally, HIF-1α also regulates the expression of the γ- secretase component APH-1A, suggesting another mechanism that leads to the enhanced

Notch activity (Wang et al., 2006). In addition to the direct impact of hypoxia on stem and progenitor cells, hypoxia also influences the function of stem cell niches. It has been shown that hypoxia decreases the potential of mesenchymal progenitor cells to undergo myogenic differentiation and maintains the undifferentiated state of niche stromal cells

(Lin and Yun, 2010). Furthermore, hypoxia promotes epithelial to mesenchymal transition (EMT), either by enhancing the production of transforming growth factor β

(TGFβ) (Nishi et al., 2004; Orphanides et al., 1997) or increasing the activity of the

Notch intracellular domain (Chen et al., 2007; Gustafsson et al., 2005). Both TGF

(Miettinen et al., 1994) and Notch (Zavadil et al., 2004) are known regulators that induce

EMT. EMT is a process through which epithelial cells lose their phenotype and transit to a mesenchymal phenotype, considered as a crucial event that promotes cancer progression and metastasis. Although HIF-1α and HIF-2α are major regulators that mediate chronic cellular responses to hypoxia, there are also HIF-independent pathways

10 contributing to acute hypoxic responses. For examples, hypoxia induced expression of the protein “regulated in development and DNA damage responses 1” (REDD-1) suppresses the mammalian target of rapamycin kinase complex 1 (mTORC-1), which contributes both to normal cellular homeostasis and tumor suppression (DeYoung et al.,

2008). However, in many cases, tumor cells can evade this metabolic checkpoint due to inactivation of the REDD-1 pathway, ultimately leading to tumor development and progression (Kucejova et al., 2011; Schwarzer et al., 2005). All of the above evidence suggests that hypoxia is a critical regulator in enhancing tumorigenesis, increasing tumor metastasis and facilitating cancer progression, resulting in an aggressive tumor phenotype. Additionally, hypoxia promotes tumor cell resistance to conventional chemo/radiotherapy. Therefore, it is of extreme importance to identify novel therapeutic strategies to target hypoxic tumor regions for cancer treatment.

THE NOTCH SIGNALING PATHWAY

The Notch signaling cascade

The Notch signaling pathway is an evolutionarily conserved signaling pathway, which plays an important role during embryonic development and cell differentiation

(Artavanis-Tsakonas et al., 1999). The mammalian Notch family consists of four

(Notch1-4) type I transmembrane receptors and five Notch ligands, Jagged-1 (JAG-1),

JAG-2, Delta-like-1 (DLL-1), DLL-3 and DLL-4 (Lai, 2004). Notch receptors are produced as precursor forms. In order to form the active Notch, Notch precursors must undergo three sequential proteolytic cleavages (Figure 1.1). The first cleavage is

11 mediated by furin-like convertase in the Golgi (S1 cleavage), which cleaves Notch precursor protein to generate the mature form of Notch before its transport to the cell surface (Logeat et al., 1998). After the furin-like convertase mediated cleavage (S1 cleavage), the extracellular and intracellular domains are no longer covalently linked.

Notch signaling is a cell-cell communication process, which allows the crosstalk between the receptor expressing and ligand expressing cells. Since both Notch receptors and ligands are transmembrane proteins, cell-cell contact becomes a prerequisite to trigger downstream signaling events. Upon the binding of Notch ligand that is expressed on one cell to the extracellular domain of the mature Notch receptor on another cell, the ligand is trans-endocytosed into the ligand-expressing cell (Parks et al., 2000). This event exposes the extracellular cleavage site of Notch to the metalloproteinases ADAM10 (A

Disintegrin and metalloproteinase domain-containing protein 10) or tumor necrosis factor-α-converting enzyme (TACE; also know as ADAM17) (Brou et al., 2000).

ADAM17/TACE cleaves Notch in vitro, but only the ADAM10 KO mice show the Notch deficiency phenotype (Hartmann et al., 2002; Zhao et al., 2001). Hence, probably both

ADAMs cleave Notch in vivo. This second cleavage mediated by ADAM10 (or

ADAM17) (S2 cleavage) exposes the binding sites of the transmembrane region of Notch to the γ-secretase complex. γ-secretase is a protease complex that contains presenilin

1(PSEN1), PSEN2, nicastrin (NCSTN), presenilin enhancer 2 (PEN2) and anterior pharynx-defective 1 (APH1) proteins (Kaether et al., 2006). The cleavage mediated by γ- secretase complex (S3 cleavage) results in the release of the Notch intracellular domain

(NICD), also called activated Notch, to the cytosol and subsequent translocation into the

12 nucleus (Schroeter et al., 1998). There are small molecule compounds that inhibit the γ- secretase complex mediated cleavage, called γ-secretase inhibitors (GSI). Once activated,

Notch intracellular domain (NICD) translocates into the nucleus, and binds to C-promoter binding factor 1 (CBF-1) (Tamura et al., 1995), resulting in the replacement of the co- repressors that were bound to CBF-1 with the co-activator protein mastermind-like 1

(MAML1) and other transcription factors (Oswald et al., 2001). This eventually converts the transcriptional repressor complex into a transcriptional activator complex (Jeffries et al., 2002; Wu et al., 2000). This transcriptional activator complex then promotes the transcription of many downstream target genes, such as HES (the transcriptional repressors of the hairy enhancer of split), HEY (the hairy-related transcription factor,

HRT, also known as HEY) (Iso et al., 2003), p21/Waf1 (Rangarajan et al., 2001), NFkB

(Oswald et al., 1998), cyclin D1 (Ronchini and Capobianco, 2001) and c-MYC (Klinakis et al., 2006).

Figure 1.1 The Notch signaling pathway. Notch receptors are produced as precursor forms. In order to form the active Notch, Notch precursors must undergo three sequential proteolytic cleavages: S1 cleavage is mediated by furin-like convertase in the Golgi, which cleaves Notch precursor protein to generate the mature form of Notch before its transport to the cell surface; S2 cleavage mediated by ADAM10 (or ADAM17) exposes the binding sites of the transmembrane region of Notch to the γ-secretase complex; S3 cleavage mediated by γ-secretase complex results in the release of the Notch intracellular domain (NICD) to the cytosol and subsequent translocation into the nucleus. Once NICD translocates into the nucleus, it binds to CBF-1, converting the transcriptional repressor complex into a transcriptional activator complex, which promotes the transcription of many downstream Notch target genes.

Regulation of Notch by hypoxia

Accumulating evidence has suggested that Notch signaling is activated by hypoxia. In primary neural stem cells and colorectal cancer-derived cell lines, hypoxia prevents cell differentiation and maintains a stem-like cell phenotype in a Notch dependent manner (Gustafsson et al., 2005; Yeung et al.). Using neurogenic rat embryonic carcinoma cells, the authors were able to demonstrate that in hypoxia, the

Notch intracellular domain (NICD) interacts with, and is stabilized by HIF-1α, and the complex is recruited to the Notch responsive promoters bound to CBF-1 to activate the

Notch downstream target genes Hes-1 and Hey-2 (Gustafsson et al., 2005). Additionally,

Notch signaling has been shown to be required for the hypoxia mediated epithelial- mesenchymal transition (EMT), as well as for tumor cell migration and invasion

(Sahlgren et al., 2008). Moreover, hypoxia stimulates pulmonary arterial hypertension by activating the NOTCH3-HES-5 signaling pathway (Li et al., 2009). Recently, Lanner and colleagues reported that hypoxia upregulated the expression of the Notch ligand DLL-4 and increased Notch signaling, which promoted arterial differentiation. Since Notch signaling also upregulates DLL-4 expression, a positive feedback loop is formed to sustain DLL-4 expression and Notch signaling (Lanner et al., 2013).

The role of the Notch pathway in lung cancer

The important role of the Notch signaling pathway has been implicated in cancer development. In many different solid tumors, aberrant activation of Notch signaling can lead to tumor cell proliferation and tumorigenesis, such as in breast (Klinakis et al.,

2006), colorectal (van Es et al., 2005), and pancreatic (Miyamoto et al., 2003) cancers. In contrast, Notch signaling seems to have a tumor suppressor role in keratinocytes

(Rangarajan et al., 2001). Many studies suggest that Notch drives transformation and tumor progression mostly by promoting cell proliferation and inhibiting apoptosis. It has been shown that inhibition of Notch-3 reduced tumor cell proliferation and caused apoptosis of lung cancer cell lines in vitro (Konishi et al., 2007). We have found that

Notch-1 signaling plays a crucial role in mediating the survival of NSCLC under hypoxia

(Chen et al., 2007). In addition, another study also indicates that Notch-1 signaling might play an important role in the carcinogenesis and progression of NSCLC, since the protein levels of Notch-1, Jagged-1 and VEGF were closely correlated with the pathological type and lymph node metastasis (Jiang et al., 2007). Furthermore, we have shown that inhibition of Notch-1 signaling (either through genetic siRNA downregulation of Notch-1 or using GSI) in NSCLC cells caused cell death specifically under hypoxia condition.

This specific effect could be rescued by overexpressing intracellular Notch-1 (NICD).

Further immunofluorescent staining analysis of intracellular Notch-1 and the hypoxic marker Glucose transporter-1 (GLUT-1) indicated that the activated Notch-1 is only present in hypoxic areas. Furthermore, we found that the activated Notch-1 under hypoxic areas can promote tumor growth through two mechanisms – upregulating IGF-1 and its receptor IGF-1R and downregulating the expression of phosphatase and tensing homolog (PTEN) (Eliasz et al., 2010). Therefore, using small molecule inhibitors, such as

GSI, to target Notch-1 signaling could be a potential therapeutic strategy to treat NSCLC.

THE IGF-1R SIGNALING PATHWAY

Ligand and receptor signaling

Insulin-like growth factor 1 receptor (IGF-1R) is a member of receptor tyrosine kinase family proteins, which has 70% amino acid homology to the insulin receptor (IR)

(Ullrich et al., 1986). IGF-1R is ubiquitously expressed. There is one single copy of the

IGF-1R gene, which is located on chromosome 15. The mRNA of IGF-1R is transcribed as a single peptide and then cleaved into two subunits: the α subunit and the β subunit.

The α subunit is completely extracellular, while the β subunit consists of extracellular, transmembrane and intracellular domains. The intracellular domain of the β subunit contains the kinase activity. Both the α subunit and β subunit are glycosylated, and they can associate with each other to form a monomer through disulfide bonds. Two IGF-1R monomers can associate and form an IGF-1R homodimer, which includes two α and two

β subunits(Ward and Garrett, 2004). The homodimer can be bound by IGF-1R ligands, including IGF-1, IGF-2 and insulin. IGF-1 and IGF-2 bind to IGF-1R with high affinity, while insulin can also bind to IGF-1R but with 100-1,000 fold lower affinity (Singh and

Rubin, 1993). IGF-1 and IGF-2 are mostly produced in the liver and secreted to the blood stream, but other tissues can also produce IGFs, including the lungs. The local bioavailability of IGFs in the peripheral tissues is regulated by IGF binding proteins

(IGFBPs 1-6). These proteins have opposite functions in regulating IGF-1R signaling.

They can either have an inhibitory function by trapping IGF-1 and IGF-2 in the circulation, preventing them from binding to their receptors, or they can promote IGF-1R signaling by prolonging the half-life of IGF-1 and IGF-2 (Pollak et al., 2004). IGF-2 also

17 binds to IGF-2R, which does not have intracellular kinase activity and does not promote signaling. However, IGF-2R is able to bind to IGF-1R and inhibit its downstream signaling, therefore acting as a negative regulator of the IGF-1R signaling pathway (Yee,

2006). Due to IGF-1R and insulin receptor (IR) sharing a lot of homology, heterodimers containing half IGF-1R monomer and half IR monomer can also be found (Pandini et al.,

1999).

Upon ligand binding to the IGF-1R receptor, autophosphorylation of the receptor occurs, leading to the activation of downstream signaling components (Figure 1.2). There are two major downstream signaling pathways of IGF-1R signaling: one involves PI3K,

AKT and mTOR activation whereas the other is mediated by RAS and MAPK (Tao et al.,

2007). The kinase activity of the receptors leads to phosphorylation of the insulin receptor substrate (IRS) family of proteins, which in turn activate PI3K-AKT and several other downstream signaling pathways. The consequences of this pathway activation are inhibition of apoptosis, cell survival and cell growth (Testa and Tsichlis, 2005). The second downstream signaling mediated by RAS and MAPK is triggered by binding of

Shc to the IGF-1R β subunit, which in turn recruits Grb2 and SOS, resulting in the activation of the Ras-Raf1-MEK-ERK signaling pathway. Another mechanism that leads to RAS-MAPK activation by IGF-1R is mediated by phosphorylated IRS, which can also provide docking sites for Grb2 (Ward et al., 1996). The consequences of ERK signaling activation include the stimulation of proliferation, increased motility and survival (Howe et al., 2002).

Figure 1.2 The IGF-1R signaling pathway. Upon ligand binding to the IGF-1R receptor, autophosphorylation of the receptor occurs, leading to the activation of two major downstream signaling pathways: one involves activation of PI3K, AKT and mTOR signaling whereas the other is mediated by RAS and MAPK. The consequences of activation of both of these signaling pathways include increased protein synthesis, cell survival and proliferation, and decreased apoptosis.

The role of the IGF-1R pathway in NSCLC and other tumors

The expression of IGF-1R is ubiquitous. Almost all tissues express certain amounts of IGF-1R (Abbott et al., 1992). In many types of cancer, including osteosarcoma (Burrow et al., 1998), colorectal (Hakam et al., 1999), breast (Happerfield et al., 1997; Railo et al., 1994), prostate (Hellawell et al., 2002), lung (Kaiser et al.,

1993), and gastric cancer (Pavelic et al., 2003), IGF-1R has been found to be overexpressed, but mutations and gene amplifications of IGF-1R are rarely identified.

IGF-1R signaling has been linked to tumorigenesis and cancer progression.

Overexpression of wild type IGF-1R is sufficient to transform both human NIH-3T3 fibroblast cells and mouse embryo fibroblasts in vitro. In addition, it is been shown that mouse fibroblasts that do not have functional IGF-1R are much more difficult to be transformed with different viral and cellular oncogenes, such as simian virus (SV) 40 large T antigen (Sell et al., 1993), Ras (Sell et al., 1994), c-src527 (mutation at tyrosine

527 to phenylalanine) (Valentinis et al., 1997) and ETV6-NTRK3 (Morrison et al., 2002).

In addition, in a transgenic mouse model, overexpressing IGF-1 in the basal layer of the skin epidermis increased the incidence of squamous papilloma formation. There are also studies suggesting that activation of the IGF-1R signaling pathway leads to increased tumor cell proliferation and an increased incidence of cancer metastasis. In a mouse model of colon cancer, treatment with recombinant IGF-1 significantly increased the rates of tumor progression and liver metastasis. Clinical reports also suggested that there is a correlation between high level of serum IGF-1 and increased risk of developing different types of cancer, including prostate, breast, colorectal, ovarian, bladder and lung

20 cancer in humans. Besides the role of tumorigenesis and metastasis, IGF-1R also has a significant anti-apoptotic signaling capacity. Overexpression of IGF-1R has been shown to protect fibroblasts from UV-B light induced apoptosis (Kulik et al., 1997). Activation of the IGF-1R signaling pathway has also been shown to have an inhibitory role in chemical (D'Ambrosio et al., 1997) or cytokine (Wu et al., 1996) induced apoptosis. In addition, Trastuzumab (Herceptin), a HER2/neu receptor inhibitory antibody, can only inhibit breast cancer cell proliferation when the level of IGF-1 is minimal (Lu et al.,

2001), suggesting activation of IGF-1R signaling can rescue cells from Trastuzumab (a monoclonal antibody that interferes with the HER2/neu receptor) mediated apoptosis.

Many studies have shown that IGF-1, IGF-2 and IGF-1R are expressed in NSCLC cell lines, as well as human lung tumor tissues of different histological sub-types (Jaques et al., 1988; Kaiser et al., 1993; Minuto et al., 1986; Nakanishi et al., 1988; Rotsch et al.,

1992). Moreover, IGF-1 and -2 treatments of human small cell and non-small cell lung cancer cell lines promote cell growth and proliferation (Favoni et al., 1994). To study the role of IGF-1 and IGF-2 on tumor formation in the lung, an IGF-1 transgenic mouse model, where IGF-1 is specifically overexpressed in the alveolar epithelial cells, was generated. Compared to their nontransgenic littermates, these mice are more prone to have adenomatous hyperplasia (the over-growth of the endometrium) and to form spontaneous pulmonary adenomas in older mice (>1yr) (Frankel et al., 2005). In addition, overexpression of IGF-2 in the lung epithelium significantly increased the incidence of developing lung tumors, especially in mice order than 18 months (Moorehead et al.,

2003). Several studies have suggested that high plasma levels of IGF-I and low levels of

IGFBP-3 were associated with an increased risk of lung cancer (Renehan et al., 2004;

Spitz et al., 2002; Yu et al., 1999). In lung cancer as well as other solid tumors, the crosstalk between IGF-1R and epidermal growth factor (EGF) receptor (EGF-R) has been suggested. IGF-1R/EGF-R heterodimerization has been found in erlotinib treated NSCLC cells, leading to the activation of the IGF-1R downstream signaling components

PI3K/Akt and MAPK, which provides survival signals to the cells (Morgillo et al., 2006).

Besides IGF-1R/EGF-R herterodimerization, another possible mechanism that leads to interaction between these two receptors is through activating G-protein coupled receptors

(GPCRs). It has been shown that activation of IGF-1R turns on many GPCRs (Delcourt et al., 2007). On the other hand, activation of GPCRs can induce metalloprotease

(ADAM) protein family mediated shedding of EGF-R ligands (Sahin et al., 2004) and lead to receptor activation. Additionally, in a tamoxifen-resistant breast cancer cell line, an EGF-R tyrosine kinase inhibitor (TKI) can disrupt the interaction between EGF-R and

IRS-1 and promote association of IGF-1R and IRS-1, thus promoting IGF-1R signaling

(Knowlden et al., 2008). Our study on IGF-1R resistance in NSCLC, on the other hand, showed that IGF-1R inhibition could activate EGF-R signaling and promote tumor growth, suggesting another mechanism of IGF-1R/EGF-R signaling crosstalk.

IGF-1R as a target for cancer treatment

Since IGF-1R plays an important role in tumorigenesis and cancer progression, it is generally accepted that IGF-1R could be a potential therapeutic target for cancer therapy. In 1989, Arteaga. et al reported that inhibition of IGF-1R using an antibody

22 showed a slower growth of tumors in a mouse xenograft breast cancer model. This is the first study suggesting that an antibody against IGF-1R could be used to treat cancer.

Since then, there has been more and more evidence indicating that the IGF-1R signaling pathway could be used as a therapeutic target for cancer therapy. The finding that mouse embryos carrying null mutations of the gene encoding Igf-1r display growth deficiency

(45% normal size) (Liu et al., 1993) suggests that IGF-1R is not 100% required for normal embryo growth and that there are other pathways that could contribute to the growth. Moreover, the effect of IGF-1R on cell growth has been evaluated in anchorage independent conditions. The growth of human melanoma cells is strongly inhibited when an antisense RNA against IGF-1R RNA is introduced into nude mice, while there is no effect when the same antisense RNA is introduced into the same cells grown in monolayers (Resnicoff et al., 1994). Since IGF-1R is important for mediating viral and cellular oncogene transformation, it is reasonable to hypothesize that downregulation or inhibition of the IGF-1R signaling pathway may play a role in suppressing tumor growth.

Surprisingly, downregulation of IGF-1R not only inhibited tumor growth, but also caused apoptosis especially in anchorage independent conditions. Therefore, IGF-1R signaling has been used as a well-studied therapeutic target in cancer treatment. Currently there are three different groups of strategies to target IGF-1R signaling: anti-receptor antibodies, anti-ligand neutralizing antibodies and small molecule tyrosine kinase inhibitors. Many studies have used a number of different IGF-1R specific inhibitory antibodies to inhibit tumor xenografts in mice with promising results. So far, many of these antibodies are under phase I and II clinical trials (Gualberto and Pollak, 2009). While the results of

23 phase I and II clinical trials indicated low toxicity of these antibodies (mainly hyperglycemia and increase insulin levels), a recent phase III trial of figitumumab (a human monoclonal antibody against IGF-1R) in NSCLC suggested a lack of efficacy.

This might be due to the increased resistance since we have found that tumor cells became resistant to IGF-1R inhibitory antibody treatments (MK-0646) owing to increased EGF-R activity (Liang et al., 2012). Furthermore, other mechanisms to evade

IGF-1R inhibition have been suggested in different types of cancer. For instance, in rhabdomyosarcoma (an aggressive muscle cancer of children and young adults), IGF-1R inhibition is associated with HER2 overexpression and MAPK reactivation (Abraham et al.). Therefore, the above evidence strongly indicates that the IGF-1R inhibitory antibody treatments should be used in combinational or sequential therapies or certain patients with predictive biomarkers.

THE EGF-R SIGNALING PATHWAY

An overview of the EGF-R pathway

Epidermal growth factor receptor (EGF-R) can be found in many types of cells throughout the body and plays an important role in cell growth and cell survival. EGF-R is a transmembrane glycoprotein, containing an extracellular ligand binding domain, a transmembrane domain, and an intracellular tyrosine kinase domain. EGF-R belongs to the ErbB family of receptor tyrosine kinases (RTKs), which includes EGF-R, ErbB-2

(HER-2), ErbB-3, and ErbB-4 (Walker, 1998). ErbB-2 (HER-2) is highly homologous to

EGF-R. However, there is no known high affinity ligand of ErbB-2, and it is in a

24 constitutively active conformation (Burgess et al., 2003). ErbB-3 and ErbB-4 are structurally related proteins. ErbB-3 does not have tyrosine kinase activity and it can only be activated by heterodimerizing with other ErbB family members. EGF-R can be activated by certain ligands such as epidermal growth factor (EGF). Upon ligand binding to the extracellular domain of EGF-R, the receptor either undergoes homo- or heterodimerization (with other ErbB family proteins) and allows autophosphorylation of the tyrosine residues (Olayioye et al., 2000). Phosphorylated tyrosine residues on the receptors then serve as docking sites for various adaptor proteins and enzymes, which will initiate a number of downstream signaling cascades via recruiting the adaptors growth factor receptor-bound protein 2 (GRB2) and the SH2 domain-containing family of proteins. The main signaling cascades include the Ras-Erk1/2 pathway, the phosphotidylinositol-3 kinase (PI3K)-Akt pathway, the phospholipase Cγ-protein kinase

C pathway and the signal transducer and activator of transcription (STAT) signaling pathway.

Endocytic sorting of EGF-R

Upon ligand binding, subsequent dimerization and autophosphorylation of EGF-R leads to the recruitment of adaptor protein GRB2, which in turn binds to an E3 ubiquitin- protein ligase Cbl (Casitas B-lineage Lymphoma). EGF-R can be either mono- or polyubiquitinated by Cbl (Levkowitz et al., 1998), and then recognized and bound by ubiquitin-binding proteins of the clathrin coat, such as epidermal growth factor receptor substrate 15 (EPS15) (Hawryluk et al., 2006) and hepatocyte growth factor-regulated Tyr

25 kinase substrate (HRS). Upon internalization into clathrin-coated pits, EGF-R will then be transported to the early endosome, and subjected to either recycling back to the plasma membrane or lysosomal degradation. In the early endosome, EGF-R that is designated for lysosomal degradation is still associated with Cbl, and is continuously being ubiquitinated. Ubiquitinated EGF-R can then be recognized by ESCRT complexes

(HRS/STAM and ESCRT-I to III), which mediate receptor translocation into intraluminal vesicles (ILVs) of multivesicular bodies (MVBs) or lysosomes for degradation processes

(Katzmann et al., 2002). However, ubiquitination of EGF-R by Cbl is considered an essential step for EGF-R degradation, but not an absolutely required event for clathirin- mediated endocytosis (Duan et al., 2003; Huang et al., 2007).

Effects of ligands on EGF-R endocytic sorting

Seven EGF-R ligands have been described so far: epidermal growth factor (EGF), transforming growth factor α (TGFα), heparin-binding EGF (HB-EGF), β-cellulin

(BTC), amphiregulin (AR), epiregulin (EPI), and epigen. In regards to ligand mediated

EGF-R endocytosis and degradation, both EGF and TGFα have been studied extensively.

Unlike EGF that binds to EGF-R, which leads to lysosomal degradation of EGF-R, the binding between TGFα and EGF-R causes receptor recycling (Decker, 1990). This is the reason that TGFα is a stronger mitogen than EGF (Waterman et al., 1998). It is believed that the binding between EGF and EGF-R is relatively stable in the acidic environment of endosomes, compared to the binding between TGFα and EGF-R. Once TGFα binds to

EGF-R and targets EGF-R into the endosome, the binding between TGFα and EGF-R

26 quickly dissociates due to the low pH in endosome, leading to EGF-R de-phosphorylation and de-ubiquitination. Deubiquitinated EGF-R will then be recycled back to the plasma membrane (Longva et al., 2002). Interestingly, one study on amphiregulin mediated effects of EGF-R degradation suggests that amphiregulin stimulation results in EGF-R recycling rather than degradation, even in the condition that Cbl is overexpressed and

EGF-R is extensively ubiquitinated (Stern et al., 2008). Furthermore, many studies have indicated that high expression levels of amphiregulin could be used as a marker to select patients for anti-EGF-R therapy. For example, metastatic colorectal cancer patients with high expression levels of amphiregulin and epiregulin are more sensitive to anti-EGF-R antibody cetuximab treatment (Khambata-Ford et al., 2007). In addition, it has been suggested that NSCLC patients with wild-type EGF-R and positive for amphiregulin may obtain a clinical benefit from EGF-R tyrosine kinase inhibitor treatment (Chang et al.,

2011).

CHAPTER 2

RATIONALE AND DISSERTATION OUTLINE

Lung cancer is the leading cause of cancer related deaths in the United State and worldwide. Each year, the number of mortalities due to lung cancer is more than that of colon, breast, and prostate cancers combined. Due to the lack of early detection method,

NSCLC is often diagnosed at advanced or metastatic stages. So far, chemotherapy is still the first line of therapy to treat advanced or metastatic NSCLC (see Table 1.2 for a full list of standard treatments for NSCLC). However, the standard third-generation platinum- based doublet agents, even in combination with targeted therapy agents, only result in a small improvement for patients with advanced NSCLC. Therefore, the identification of new therapeutic targets and discovery of new combinational therapeutic regimens to treat

NSCLC are of extreme importance. In addition to the unsatisfactory results of the current therapeutic regimens in NSCLC, the hypoxic tumor microenvironment presented in all solid tumors, including lung cancer, adds another layer of difficulty to the development of therapeutic agents targeting advanced NSCLC. Notably, it has been shown that the cells present in hypoxic tumor regions are responsible for resistance to chemo/radiotherapies. Moreover, hypoxia has been linked with many other effects in promoting tumor growth and metastasis, such as increased angiogenesis, enhanced metastatic potential, DNA replication and reduced protein synthesis. Therefore, the

27 28 purpose of this dissertation is to identify potential therapeutic targets for the treatment of advanced NSCLC under hypoxic conditions.

We and other groups have shown that Notch signaling plays a crucial role in

NSCLC. Notch signaling is an evolutionarily conserved signaling pathway that has been extensively studied in cell fate regulation and stem cell homeostasis. Notch signaling is frequently dysregulated in human cancers. In mammals, there are four Notch receptors

(Notch-1 to 4) and five canonical Notch ligands, which include three Delta-like and two

Jagged. In T-cell acute lymphoblastic leukemia/lymphoma (T-ALL), Notch-1 is activated through chromosomal translocations (Ellisen et al., 1991) or activating mutations (Weng et al., 2004). Similarly, increased Notch activity has also been observed in one third of

NSCLCs, either due to loss of expression of a negative regulator Numb or gain-of- function mutations in the Notch-1 gene (Westhoff et al., 2009). Additionally, high expression levels of Notch targets Hes-1, HeyL and Hey-1 have been found in a number of NSCLC cell lines (Collins et al., 2004), which further suggests the important role of

Notch activity in NSCLC. Moreover, the importance of different Notch receptor isoforms in maintaining tumor growth and promoting tumorigenesis has also been investigated in

NSCLC. The activity of Notch-3 is linked with increased proliferation of NSCLC cells

(Konishi et al., 2007) as well as enhanced tumorigenic potential in a subpopulation of self-renewing NSCLC stem-like cells (Sullivan et al., 2010). Our previous results showed that Notch-1 activity is activated in hypoxia, which is critical in providing survival signals to NSCLC cells. Furthermore, targeting Notch-1 signaling by either GSI or siRNA against Notch-1 caused significant cell death, specifically under hypoxic

29 conditions (Chen et al., 2007). All of the above studies have elucidated an emerging but critical role of Notch signaling in NSCLC, and have also highlighted the therapeutic potential of targeting Notch signaling in the treatment of NSCLC.

To better understand the detailed mechanism underlying Notch-1 mediated survival of NSCLC cells under hypoxia, and to develop new potential therapeutic agents to treat hypoxic NSCLC, our lab focused on studying the downstream signaling events that are activated by Notch-1 signaling and contribute to cell survival under hypoxia condition. We found that Notch-1 provides necessary survival signals to NSCLC cells by positively regulating insulin-like growth factor-1 receptor (IGF-1R) to activate the Akt-1 pathway. This effect is exacerbated by the fact that Notch-1 represses phosphatase and tensin homolog (PTEN) expression (Eliasz et al., 2010).

IGF-1R is a transmembrane glycoprotein consisting of two extracellular subunits and two intracellular subunits. The two intracellular subunits contain the tyrosine kinase domain that is responsible for the activation of downstream signaling components IRS-

1/PI3K/AKT pathway and the Ras/Raf/MAPK pathways. The activation of these pathways leads to cell proliferation, promotes transformation, survival and migration.

High expression level of IGF-1R has been detected in patients with advanced NSCLC. In

NSCLC, IGF-1R signaling is involved in cell proliferation, tumor invasion and survival.

In addition to our finding that IGF-1R is a target of Notch-1 in NSCLC cells, the crosstalk between IGF-1R and Notch pathways in T-ALL has recently been reported, which also suggests that Notch signaling promotes tumor growth and survival through regulating IGF-1R expression (Medyouf et al., 2011). The same group also reported that

30 the expression of IGF-1R is regulated by Notch in T-ALL through microRNA miR-223

(Gusscott et al., 2012). Interestingly, IGF-1R downstream signaling PI3K/Akt/mTOR has been shown to regulate the levels of HIF-1α in different types of cancer. In prostate cancer and glioblastoma-derived cell lines, Akt activation leads to increased HIF-1α expression, whereas PTEN attenuates hypoxia-mediated HIF-1α expression (Zhong et al., 2000; Zundel et al., 2000). On the other hand, HIF-1α protein translation was elevated upon activation of PI3K/Akt/mTOR pathway in breast cancer (Laughner et al.,

2001). Since the intracellular Notch (NICD) was shown to be stabilized and activated by binding to HIF-1α (Gustafsson et al., 2005), not only IGF-1R signaling can be activated by Notch, the Notch downstream signaling may also be triggered by IGF-1R signaling through regulating the level of HIF-1α, to form a positive feedback loop. Therefore, an inhibitory antibody targeting the IGF-1R or a pan Akt inhibitor may synergistically act in combination with GSI (a γ-secretase inhibitor that blocks Notch signaling) to completely inhibit the IGF-1R and Akt signaling pathways, and contribute to depletion of tumor mass in hypoxic areas, thus providing advantages to the treatment of advanced NSCLC.

The overall hypothesis of this dissertation is that γ-secretase inhibitor (GSI), MRK-

003, or a fully humanized antibody against the human IGF-1R that promotes the receptor degradation (IGF-1R inhibitory antibody), MK-0646, or a pan-Akt inhibitor, MK-2206, alone or in various combinations, could improve the efficacy of chemotherapy for advanced NSCLC. We tested this hypothesis in mouse preclinical models of orthotopic, advanced (metastatic) NSCLC, since the stage IV NSCLC patients who receive the best supportive therapy alone has a median survival of 3.5 months. To parallel this situation

31 we should have started therapy in mice with a life expectancy of about 2.7 days at the time of intervention. We arbitrarily started therapy when mice had a bit more than a week of life expectancy (8 or 9 days).

In Aim 1, we wanted to test whether inhibition of the Notch-1/IGF-1R/Akt axis is a suitable therapeutic tool to target hypoxic NSCLC. Specifically, we tested this hypothesis by using a number of drugs that inhibit Notch-1, IGF-1R, or Akt, as well as some other drugs currently in clinical use either as standard of care or second line therapeutics (cisplatin and erlotinib). We tested these drugs in an orthotopic mouse model of advanced NSCLC. We sought to determine the best drug combination that improves the median survival of mice with advance NSCLC. We used the orthotropic model instead of a subcutaneous xenograft model because this model can better recapitulate the hypoxic environment of tumors grown in the lungs, which are more heterogeneous and contain hypoxic areas in combination with well-oxygenated tumor tissues. In our experience, a subcutaneous xenograft is mostly hypoxic, hence less heterogeneous in terms of tumor microenvironment. Additionally, we also wanted to test whether the hypoxic tumor region is indeed decreased and if cancer metastasis is reduced after drug treatments. Specifically, we measured the expression levels of hypoxic markers (human

VEGF-A and GLUT-1) in tumors excised from GSI treated and vehicle treated mice.

Both VEGF-A and GLUT-1 are known HIF specific target genes (Carroll and Ashcroft,

2006; Schipani et al., 2009), and we also identified GLUT-1 as a good marker of hypoxia in NSCLC. Using the TUNEL assay, we visualized specific cell death in hypoxic tumor areas. Furthermore, we quantified the tumor burdens in the lungs, livers and brains of

32 mice by measuring the ratio of genomic human versus mouse Glyceraldehyde 3- phosphate dehydrogenase (GAPDH) by quantitative PCR (the assay was previously validated using known number of cells).

During the course of treating tumor-bearing mice with IGF-1R inhibitory antibody (MK-0646), we found that at the beginning of the treatment, MK-0646 had a tumoristatic effect on NSCLC cells. However, after the first two weeks, tumor cells resumed growth and ultimately led to the death of the experimental animals. The recurrent malignancy and eventual fatality are commonly seen in anti-cancer therapy in humans (with the exception of some hormone-driven malignancies). Regardless of the therapeutic approach, tumors become resistant to adjuvant radio/chemotherapy and eventually lead to the death of the patients. In Aim 2 of this dissertation, our objective was to uncover the mechanism(s) that allow cancer cells to evade IGF-1R deprivation. A recent study suggested that inhibition of Akt-1 signaling leads to a compensatory activation of several receptor tyrosine kinases (RTKs) (Chandarlapaty et al., 2011). Since our preliminary results showed that treatment with MK-0646 in NSCLC cells caused rapid loss of IGF-1R total protein and dramatic loss of Akt activation, we wanted to investigate if tumor cells evade MK-0646 treatment through activating alternative RTKs. Proteomic RTK array analysis with lung tumor samples from mice treated with IGF-1R inhibitory antibody (MK-0646) showed that the level of phosphorylated EGF-R (pEGF-R) was significantly increased compared to vehicle controls. As activation of the EGF-R signaling pathway leads to cell survival and proliferation, this could be a potential mechanism underlying the tumor survival when

IGF-1R signaling is impaired. In Aim 2, we hypothesized that IGF-1R inhibitory antibody treatment activates the EGF-R signaling pathways, leading to tumor cell survival and proliferation. This hypothesis was tested both in vitro, using NSCLC cell lines cultured in MK-0646 containing media, and in vivo, with lung tumor samples obtained from MK-0646 treated mice. Furthermore, we also wanted to further advance this study by identifying the detailed molecular mechanisms that lead to EGF-R activation. We focused on determining if EGF-R mRNA or protein levels were affected, whether the levels of EGF-R ligands were changed and whether the EGF-R endocytosis pathway was altered by MK-0646 treatment.

Since NSCLC cells treated with MK-0646 had overactivated EGF-R, we asked whether such cells could have become more sensitive to the EGF-R inhibitor (erlotinib).

In Aim 3, we hypothesized that administration of the IGF-1R inhibitory antibody followed by the EGF-R inhibitor (erlotinib) has better efficacy in treating advanced

NSCLC. This hypothesis was again tested both in vitro and in vivo. Briefly, NSCLC cells were cultured in MK-0646 containing media for three weeks to activate EGF-R signaling.

Then the cells were either treated with the EGF-R inhibitor (erlotinib) or transduced with a lentiviral vector expressing shEGF-R to stably knock down EGF-R. Cell viability relative to vehicle treatment was then assessed. We also compared the median survival of mice treated with sequential therapy (MK-0646 for three weeks and then erlotinib) to each single agent and to a to a combination therapy protocol (MK-0646 plus erlotinib from the beginning of treatment).

Inhibiting Notch-1 through a GSI or other pharmacological interventions in non- small cell lung cancer may preferentially target hypoxic areas and the tumor-initiating cells within them. The crosstalk between Notch-1 and IGF-1R signaling suggests a possible positive feedback loop with these signaling pathways. For this reason, the use of

Notch inhibitors in combination with inhibitors of the IGF-1/Akt pathway and/or cytotoxic chemotherapy may result in synergistic anti-tumor effects in NSCLC. Also, elucidating the detailed mechanisms by which tumor cells evade from the IGF-1R depletion could reveal new therapeutic targets for the treatment of this deadly malignancy. In conclusion, data obtained by this project provide knowledge on designing novel therapeutic regimens to target advanced NSCLC, emphasizing the hypoxic tumor regions that are responsible for chemo/radiotherapy resistance and tumor relapse. The related signaling pathways and specific AIMs of this dissertation are summarized in the following flowchart:

Figure 2.1 Dissertation outline. Aim1: Determine whether inhibition of the Notch- 1/IGF-1R/Akt axis is a suitable therapeutic tool to target hypoxic NSCLC. Aim2: Delineate the detailed mechanism by which tumor evasion occurred during IGF-1R inhibitory antibody treatment. Aim3: Study if administration of IGF-1R inhibitory antibody sensitizes NSCLC cells to EGF-R inhibition.

CHAPTER 3

MATERIALS AND METHODS

Cell Culture

NSCLC cells (A549, H1437, H226, H1299 and H1650) were grown in RPMI

1640 supplemented with 5% or 10% fetal bovine serum (FBS). All cell lines were fingerprinted using the GenePrint fluorescent STR system (Promega, Madison, WI). Prior to injection in mice, A549 and H1437 were transduced with a lentivirus expressing luciferase (ViraPowerTM T-Rex Lentiviral expression system, Grand Island, NY; see below for cloning details). Some experiments were performed in cell lines re-derived from mouse lung tumors obtained through mechanical dissociation of tumor nodules after necropsies. After 5 passages, cell lines contained 100% human cells as assessed by Q-

PCR using species-specific primers for GAPDH. For experiments in hypoxia, cells were grown in 1% O2, 5% CO2, 94% N2, maintained in chambers (Stem Cell Technologies,

Vancouver, BC, Canada) filled with a certified, aforementioned mixture of gases (Airgas

North Central, West Chicago, IL, USA) at 37°C. MiniOX1 oxygen meters (Mine Safety

Appliances Company, Pittsburgh, PA, USA) were used to measure oxygen concentration.

Reagents

γ-secretase inhibitor MRK-003, the humanized monoclonal IgG1 antibody against

IGF-1R MK-0646, and the pan-Akt inhibitor MK-2206 were a generous gift from Merck

36 37

& Co. (Whitehouse Station, NJ). Erlotinib (N-3-ethylnylphenyl)-6,7-bis(2- methoxyethoxy)-4 quinazolinamine) was donated by patients of the Loyola University

Chicago Medical Center who had become unresponsive to it. IGF-1 was from Insight

Genomics (Fall Church, VA). Cisplatin and NH4Cl were from Sigma-Aldrich (St. Louis,

MO). MG-132 was from Calbiochem (San Diego, CA). The working concentrations and the duration of exposure used in each experiment are specified in the text and in figure legends.

Mouse procedures

Five-week-old female NOD.CB17-Prkdcscid/J mice (Jackson Laboratories, Bar

Harbor, ME) were injected in the tail vein with 2.5 × 106 cells in 100 µl of sterile saline solution. Mice were housed in a pathogen-free animal facility at Loyola University

Medical Center. All procedures were performed in accordance with the Institutional

Animal Care and Use Committee of the Loyola University Chicago Medical Center.

Animals were treated as indicated in the text and were monitored daily until they reached one of the end points (observed: dyspnea or 20% weight loss). At the time of euthanasia, human cells comprised 93% ± 0.8% of the total lungs of the mice. Tumor burden quantification was performed using bioluminescence imaging (Xenogen Vivo-Vision

IVIS 100 In Vivo Imaging System, Caliper Life Science, Hopkinton, MA) and by qPCR measuring human versus mouse GAPDH after euthanasia. Mice were monitored weekly for bioluminescence. All data analysis was performed with the Living Image 3.2

Software (Caliper) after intraperitoneal injection (i.p.) of 0.25 mg of D-luciferin (Gold

Biotechnology, St Louis, MO) into animals under isoflurane anesthesia. Quantitative

PCR (qPCR) was performed using human and mouse specific GAPDH primers designed after sequence alignments of human and GAPDH genomic regions (Clustalw software) to discriminate the 2 genes (Table 3.1). Primers were carefully validated, and calibration curves to match qPCR results to cell number were developed. Lungs, livers, and brains were dissected and flash frozen for molecular analyses. Genomic DNA from these tissues was purified using DNeasy Blood & Tissue Kit (Qiagen, Valencia, CA) according to the manufacture’s protocol.

Table 3.1 Primers used for human and mouse specific GAPDH

Targeted Gene Primer Sequence (5’ to 3’) Accession # Position h-GAPDH Fw AGGCCCCGGGATGCTAGTG NC_000012 1486-1864 h-GAPDH Rev CACACGCGACTCCACCCATC NC_000012 1993-2012 m-GAPDH Fw AGCACAACTCAAAACTACCTGCA NW_001030907.1 3202-3225 m-GAPDH Rev TGAGGTGTTTTGCTCCCAGT NW_001030907.1 3332-3351

Plasma glucose concentration was measured from 100 µl of blood drawn from the lateral tail vein of mice prior euthanasia using the Glucose Assay Kit (Eton Bioscience

Inc., Cambridge, MA) following the manufacturer’s instructions.

Cell Lysis, Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and Western Blotting

Cells growing in 10 cm diameter dishes were rinsed once with cold PBS and lysed on ice in 300 µl of RIPA buffer (0.1% SDS, 1% NP-40, 200uM PMSF, 0.5% sodium deoxycholate, 0.2 mM sodium vanadate, 50 mM sodium fluoride, protease inhibitor cocktail (Sigma-Aldrich) and phosphatase inhibitor cocktail (Sigma-Aldrich) in PBS).

Cells then were harvested using a cell scraper and transferred to a 1.5 ml tube on ice, followed by sonication and centrifugation at 13,000 rpm for 10 min. Supernatants were collected in new tubes after centrifugation, and Bradford assays were used to determine the protein concentration. Then, 50-100 µg of protein lysates were mixed with a 2x sample buffer (100mM Tris-HCl pH6.8, 4% SDS, 0.2% (wt/vol) bromophenol blue and

20% glycerol), boiled for 5 min at 95°C, and loaded onto 10% SDS-polyacrylamide gel along with a molecular weight marker (PageRuler Plus Prestained Protein Ladder,

Thermo Scientific, Product# 26619). Samples were electrophoresed in running buffer (25 mM Tris, 192 mM glycine and 0.1% SDS) at 120 volts at room temperature (RT) using a

Mini-PROTEAN 3 system (Bio-Rad). Separated proteins were then electrophoretically transferred onto Hybond-C Extra nitrocellulose membrane (Amersham Biosciences) in transfer buffer (25 mM Tris, 192 mM glycine, 0.37% SDS and 20% (vol/vol) methanol) at 300 mA overnight at 4°C. After transfer, membranes were blocked with 5% (wt/vol) non-fat milk made in TBST (Tris buffered saline containing Tween-20; 20 mM Tris pH

7.6, 137 mM NaCl, 0.1% (vol/vol) Tween-20) for 1 hr at RT on a rocking platform shaker. After blocking, the membranes were incubated with the appropriate primary

40 antibody diluted in 5% milk-TBST for 2 hr at RT or overnight at 4°C while rocking.

Following incubation, the membranes were washed 3 times (5 minutes each) with TBST and then incubated for 45 min at RT with the appropriate secondary antibody conjugated with horse radish peroxidase (HRP) made in 5% milk-TBST. Finally, blots were washed

5 times (10 min each) with TBST at RT and developed by incubating with enhanced chemiluminescence (ECL) reagent (SuperSignal West Dura Extended Duration Substrate,

Prod# 34076, Thermo Scientific) and exposed to x-ray films (Denville Scientific Inc.) using an automated film processor. As an additional control for equal loading, nitrocellulose membranes were stained with Ponceau S solution (Sigma-Aldrich).

Chromatin Immunoprecipitation (ChIP)

Cells were grown to ~80-90% confluency in a 150 mm culture dish containing 20 ml of growth media. Then the Magna ChIP A immunoprecipitation Kit (Millipore,

Catalog # 17-610) was used for the ChIP experiments. To crosslink proteins and DNA,

550 µl of 37% formaldehyde was added to 20 ml of cell culture media (1% final concentration), and incubated at room temperature (RT) for 10 min. After crosslinking, the unreacted formaldehyde was quenched using 2 ml of 1.25 M glycine at RT for 5 min.

Then the dishes were placed on ice, washed with 20 ml of PBS twice, and harvested in 2 ml of cold PBS containing protease inhibitor cocktail II by scraping. Cells were then centrifuged at 800 x g at 4°C for 5 min, and the cell pellets were resuspended in 0.5 ml of

Lysis Buffer containing Protease Inhibitor Cocktail II. To completely lyse the cells, cell lysates were incubated further on ice for 15 min, with vortexing every 5 min.

Supernatants were then removed after centrifugation at 800 x g at 4°C for 5 min, and cell pellets were resuspended in 0.5 ml of Nuclear Lysis Buffer containing Protease Inhibitor

Cocktail II. To shear crosslinked DNA, the cell lysates were sonicated on ice for 8 seconds, 10 times (samples were incubated on ice for 50 seconds between each sonication). A 5 µl aliquot of sheared DNA was collected for agarose gel (2%) analysis to determine the efficiency of sonication. Then, 50 µl aliquots of crosslinked chromatin were diluted with Dilution Buffer containing Protease Inhibitor Cocktail II for immunoprecipitation (5 µl of the supernatant was removed as the ‘Input’). Chromatin was immunoprecipitated overnight at 4°C with rotation using 1 µg of anti-Notch1 (c-20)

(Santa Cruz Biotech), anti-MAML1 (Millipore), anti-Acetyl Histone H3 (AcH3)

(Millipore), anti-p300 (H272) (Santa Cruz Biotech) or anti-rabbit IgG (Millipore) together with 20 µl fully suspended protein A magnetic beads. The Protein A magnetic beads were pelleted with the magnetic separator (Magna Grip Rack (8 well), Cat.# 20-

400) and supernatant was removed. Then the Protein A bead-antibody/chromatin complex was washed once with 0.5 ml of each of the following buffers: Low Salt

Immune Complex Wash Buffer (Cat.# 20-154), High Salt Immune Complex Wash Buffer

(Cat.# 20-155), LiCl Immune Complex Wash Buffer (Cat.# 20-156), and TE Buffer

(Cat.# 20-157). Finally, the protein/DNA complex was eluted with ChIP Elution Buffer with Proteinase K at 62°C for 2 hours with shaking then 95°C for 10 min. The beads were separated from the elution using a magnet and transferred to a new tube. Eluted

DNAs were then purified using spin columns provided with the Kit and regular PCR or

42 quantitative PCR (qPCR) was performed to determine the presence of CBF-1 binding sites (see detailed primer sequences in Table 3.2).

PCR and Quantitative PCR (qPCR) for Chromatin Immunoprecipitation Assay

The regular PCR and qPCR for Chromatin Immunoprecipitation experiments were performed using PTC-200 Peltier Thermal Cycler and ABI 7300 thermal cycler

(Applied Biosystems), respectively. For standard PCR experiments, each 25 µl reaction contained 4 µl ChIP DNA, 12.5 µl of 2X PCR master Mix (Thermo Fisher Scientific) and

1 µM of forward and reverse primers. PCR reactions were performed using the following protocol: 1 cycle at 95°C for 20 minutes, 40 cycles at 95°C for 1 minute and 1 cycle at

60°C for 20 minutes, followed by 4°C hold. Taqman primers were used in qPCR for the

ChIP experiment. Each 20 µl Taqman PCR reaction contained 2 µl ChIP DNA, 10 µl of

2X Taqman master mix (Applied Biosystem), 0.9 µM of forward and reverse primers and

0.25 µM of probe. The Taqman PCR reactions were performed as shown below: 1 cycle at 50°C for 2 minutes, 1 cycle at 95°C for 10 minutes, and 40 cycles at 95° for 15 seconds followed by 1 cycle at 60°C for 1 minutes.

Table 3.2 Primers used for Chromatin Immunoprecipitation (ChIP) Analysis

Targeted Gene Primer Sequence (5’ to 3’) Accession # Position Beta-globin-F CCAGCCTTATCCCAACCATA NC_000011.8 328-347

Beta-globin-R TATCATGCCTCTTTGCACCA NC_000011.8 495-514

IGF1R-CBF1- TGTGTGTGTCCTGGATTTGG NT_035325.6 See text Up-F IGF1R-CBF1- AGAAACGCGGAGTCAAAATG NT_035325.6 See text Up-R IGF1R-CBF1- GGTTGCCGAGGGTATGCA NT_035325.6 See text DWN-F IGF1R-CBF1- GTGAAGGCTCAGTCGTGATTTTT NT_035325.6 See text DWN-R IGF1R-CBF1- 6FAM-TGCCGATTAACTTTG-MGBNFQ NT_035325.6 See text DWN-probe

Immunofluorescent Microscopy

We performed immunofluorescent staining experiments on 8 µm thick lung tumor sections or NSCLC cells plated in chamber slides. The tumor samples or cells were fixed in 3.7% paraformaldehyde in PBS for 15 minutes. After fixation, cells were permeabilized with 50 mM NH4Cl in PBS containing 0.1% Triton X-100 for 15 minutes.

Then the slides were washed 3 times with PBS (5 minutes each), and blocked for 1 hour with 5% BSA in PBS containing 0.05% Triton X-100. Cells were incubated with a primary antibody diluted in 1% BSA in PBS containing 0.05% Triton X-100 overnight at

4°C on a rocking platform (See Table 3.2 for detailed information of antibodies). Rabbit

IgG or mouse IgG was used as negative control in the step of the primary antibody staining. After washing with PBS containing 0.1% Triton X-100 5 times (5 minutes each), goat anti-mouse Alexa Fluor 488 (or 568) or goat anti-rabbit Alex Flour 568 (or

488) was used in the step for the secondary antibody staining. Secondary antibodies were also diluted in 1% BSA in PBS containing 0.1% Triton X-100, and the incubation time was 1 hour. Finally, cells were washed 5 times with PBS containing 0.1% Triton X-100

(5 minutes each), rinsed once with PBS, and mounted with a fluorescence mounting medium (Dako, Glostrup, Denmark). The images were acquired with an AX80 microscope (Olympus, Center Valley, PA, USA). Lenses were UPlanApo (Olympus) with a numerical aperture of 10X. The camera was RETIGA 4000R (QImaging, Surrey,

BC, Canada). The software for image acquisition was Photoshop 6.0 (Adobe Systems

Inc., San Jose, CA, USA). Images were processed (overlaid) using Illustrator CS software

(Adobe).

Plasmids and transfections

The plasmid encoding DN-MAML-1 was described earlier (Weng et al., 2003).

Transient transfections DN-MAML-1 or control plasmid were transfected into NSCLC cells by electrophoresis. Cells were washed 3 times with Opti-MEM to remove any remaining serum, the presence of which decreases the efficiency of electroporation. Then the cells were transferred into electroporation cuvettes with 2 µg of plasmid per million cells, and electrophoresis was performed using an electroporator (GenePulserII, Bio-Rad,

Hercules, CA, USA) under the following parameters: 300kV, 975 µF. After electroporation, cells were quickly resuspended in media containing serum and plated into dishes.

Constitutively active Akt1 (NH2-terminal myristoylatable Akt1, Myr-Akt1) and dominant negative Akt1 (K179M mutant Akt1) were cloned into the pUSEamp(+) expression plasmid (pUSE empty vector was used as the negative control) (Upstate,

Millipore, Temecula, MA). Transient transfections were done using Fugene HD

(Promega, Madison, WI) according to the manufacturer’s instructions. In brief, cells were plated 1 million per 10cm dish one day before the transfection. On the day of the transfection, 3 µg of plasmid and 9 µl Fugene HD reagent were incubated together in 150

µl Opti-MEM for 5 minutes, and added to each well of 6-well plates containing attached cells.

Lentiviral systems

The luciferase gene was excised from pGL2 basic vector (Promega) by restriction digestion at the Kpn I and Bam HI sites and ligated into the Kpn I and Eco RI sites of pENTR4 (Invitrogen, Grand Island, NY). The incompatibility between the Bam HI and

Eco RI sites was circumvented using a linker oligonucleotide in the ligation mixture. The luciferase sequence was transferred via Gateway recombination into the lentiviral backbone plasmid pLenti4/TO/V5-DEST (ViraPower T-Rex Lentiviral Expression

System, Invitrogen). To produce viruses containing the luciferase gene, 239FT cells were cotransfected with ViralPower Mix together with pLenti4/TO/V5-DEST containing luciferase using lipofectamine 2000 (Invitrogen). 72 hours after transfection, supernatants were collected and viruses were concentrated by ultracentrifugation at 25,000 x g for 1 hour and 45 minutes at 4°C. Pelleted viruses were then resuspended with RPMI at 4°C overnight. To generate luciferase stable expressing cell lines, cells were infected with packaged lentivirus, and selected with Zeocin (Invitrogen) as recommended (Invitrogen).

Dual-Luciferase Reporter Assay System (Promega) was used to verify the expression of the luciferase gene, as recommended by the manufacturer, and relative light units (RLUs)

46 were measured using a Femtomaster FB15 luminometer (Zylux Corporation, Maryville,

TN).

Plasmids encoded for lentiviral packaging plasmids, pLKO lentiviral vector and pLKO-shEGF-R were gifts from Dr. Takeshi Shimamura. Lentiviruses were packaged in

293T cells. Briefly, 239T cells were plated as 4 x 106 cells per 10 cm tissue culture dishes with 10 ml of DMEM containing 10% FBS one day before the transfection. The following day, the culture media was changed to 6ml of DMEM containing 10% FBS 1 hour before the transfection. The transfection mixture of plasmid DNA and Fugene HD transfection reagent (Invitrogen) was prepared according to the manufacturer’s instructions. A 1:3 ratio of DNA: Fugene HD reagent was used. Viral packaging plasmids pLP1/2 and pLP/VSVG were cotransfected with pLKO-shEGF-R to produce lentiviruses expressing shEGF-R. Cell culture media were changed to 6ml of DMEM containing 20%

FBS per plate and incubated for additional 48 hours to generate viruses. Lentiviruses were harvested by centrifugation of culture media at 1500 rpm for 10 minutes and the supernatants were collected and added to NSCLC cells. A549 and H1437 cells were incubated with lentiviruses expressing either vector control or shEGF-R overnight in the incubator and stably transduced cells were selected in growth media supplemented with puromycin (3 µg/ml).

Receptor tyrosine kinase (RTK) arrays

Human Phospho-RTK (Human Phospho-RTK Array Kit, R&D Systems Inc.,

Minneapolis, MN) arrays were used according to the manufacturer’s instructions. Briefly,

47 tumor tissue was washed with cold PBS and homogenated in NP-40 lysis buffer (1% NP-

40, 20 mM Tris-HCl, pH 8.0, 137 mM NaCl, 10% glycerol, 2 mM EDTA, 1 mM sodium orthovanadate, 10 µg/ml Aprotinin, 10 µg/mL Leupeptin, and 10 µg/mL Pepstatin), and

250 µg of cell homogenate was incubated with preblocked membranes overnight.

Membranes were subsequently washed and incubated with freshly diluted detection antibody (anti-phosphotyrosine-HRP) for 2 hours at room temperature (R.T.) on a rocking platform shaker. Spots were detected using enhanced chemiluminescence (ECL)

(SuperSignal West Dura Extended Duration Substrate, Prod# 34076, Thermo Scientific) and exposed to x-ray films (Denville Scientific Inc.).

RNA isolation and cDNA preparation

Total RNA from cultured cells or lung tissue samples was extracted with the

RNeasy Mini kit (Qiagen, Valencia, CA) according to the manufacture’s protocol. cDNA was synthesized with iScript Reverse Transcription Supermix for RT-qPCR (Bio-Rad

Laboratories, Hercules, CA) using oligo (dT) and random primers. Each reaction contained 1 µg of RNA, 4 µl of 5x iScript reverse transcription supermix and nuclease- free water up to a total of 20 µl. The following incubation protocol was used for cDNA synthesis: 5 minutes at 25°C followed by 30 minutes at 42°C and then 5 minutes at 85°C.

No-RT control reactions were also performed with the same amount of total RNA to ensure that there was no genomic DNA contamination.

Quantitative RT-PCR (qRT-PCR)

Quantitative RT-PCR to analyze gene expression levels using the previously described cDNA was performed with SYBR Green PCR Master Mix (Applied

Biosystems, Foster City, CA, USA). For each sample, a serial dilution of cDNA template was measured in duplicate using 2 µl diluted cDNA, 12.5 µl 2X SYBR Green Master

Mix, 0.2 µM each forward and reverse primers and nuclease-free water up to a total of 25

µl. The standard SYBR Green PCR reaction cycles were run in an ABI 7300 thermal cycler (Applied Biosystems) as follows: 1 cycle at 95°C for 20 minutes and 40 cycles at

95°C for 1 minute and 1 cycle at 60°C for 20 minutes. No-RT reactions were used as negative controls for templates. The detailed primer sequences are described in Table 3.3.

Relative expression levels of all analyzed genes were normalized to human ribosomal protein RPL13A expression levels. Experiments were repeated three times.

Table 3.3 Primers used for quantitative RT-PCR (qRT-PCR) analysis

Targeted Gene Primer Sequence (5’ to 3’) Accession # Position h-EGF-R Fw CTTGCCGCAAAGTGTGTAAC NM_005228.3 1241-1260

h-EGF-R Rev TGTGGATCCAGAGGAGGAGT NM_005228.3 1415-1396

h-ErbB3 Fw TAGCCAGCTGTCCCCATAAC NM_001982 1136-1155

h-ErbB3 Rev TGTTGCTCGAGTCCACAGTC NM_001982 1318-1299 h-amphiregulin TGGTGCTGTCGCTCTTGATA NM_001657.2 242-261 Fw h-amphiregulin CCCTGAAGACATCTCACTTCTT NM_001657.2 390-369 Rev

h-EGF Fw CGTGCCCTGTAGGATTTGTT NM_001178131.1 1468-1487

h-EGF Rev GAGCTGGCTACATCCACCAT NM_001178131.1 1676-1657 h-HB-EGF Fw AGAAGAGGGACCCATGTCTT NM_001945.2 583-602 h-HB-EGF Rev AGGATGGTTGTGTGGTCATAG NM_001945.2 763-743 h-betacellulin GCAGCTACCACCACACAATCA NM_001729.2 507-527 Fw h-betacellulin CCAGAATCTGTCCTCTGTCTCCT NM_001729.2 708-686 Rev h-TGF-α Fw TTAATGACTGCCCAGATTCCCA NM_001099691.2 375-396

h-TGF-α Rev CAACGTACCCAGAATGGCAGA NM_001099691.2 479-459 h-RPL13A Fw CATAGGAAGCTGGGAGCAAG NM_012423 749-768 h-RPL13A Rev ACAAGATAGGGCCCTCCAAT NM_012423 915-896 h-VEGF-A Fw ATGACCCAGTTTTGGGAACA NM_001025366 2383-2402 h-VEGF-A Rev TCCTGAATCTTCCAGGCAGT NM_001025366 2600-2581 h-GLUT-1 Fw GTCGGAGTCAGAGTCGCAGT NM_006516 401-420 h-GLUT-1 Rev GGCATTGATGACTCCAGTGTT NM_006516 630-611

h-Hey-L Fw CATAGAGAAACGGCGTCGAGA NM_014571 201-221

h-Hey-L Rev CGAAAACCAATGCTCCGGAA NM_014571 416-397

h-Hey-1 Fw TCCTGCCTCCTTCTCTTTGA NM_012258 2034-2053

h-Hey-1 Rev TCCTGCCTCCTTCTCTTTGA NM_012258 2140-2121

HK-2 Fw GGCTGTGGATGAGCTTTCACTC NM_000189.4 4107-4128

HK-2 Rev CGATTTCACCAAGCGTGGA NM_000189.4 4221-4203

PDK-2 Fw CTGTGACCACCACATGACG NM_002611.4 2416-2434

PDK-2 Rev CTGTCTGCTCCAGTTGTTGG NM_002611.4 2593-2574

ALDOA Fw CTGCCCCCTCCCACTCTT NM_000034.3 2221-2238

ALDOA Rev GCTGTTTATTTGGCAGTGTGC NM_000034.3 2371-2351

Histological study (Haematoxylin and eosin)

Frozen tissue sections of 8-µm thickness were fixed in 10% formalin for 10 minutes followed by rinsing with PBS and H2O before staining. Slide sections were incubated with Haematoxylin for 5-9 minutes and rinsed with running tap water for 3 minutes. Then the sections were incubated with 1 or 2 dips 0.3% acid alcohol (1 ml of concentrated HCl in 400 ml of 70% ethanol) for differentiation, and rinsed with running tap water for 3-4 minutes to remove over-staining. Sections were then incubated in 0.3% ammonium hydroxide for 30 seconds to 1 minute followed by rinsing with running tap water for 2 minutes. The sections were then incubated in 95% ethanol for 1 minute followed by Eosin for 30 seconds to 1 minute. They were then dipped 10-15 times in

95% ethanol, then dipped 10-30 times in 100% ethanol, and incubated in 100% ethanol for 20-30 seconds. Slide sections were immediately incubated in Xylene for 2 minutes and another 5 minutes with fresh Xylene followed by mounting with Permount Mounting

Medium.

Protein phosphorylation expression and analysis

Western blot analyses were performed as previously described. Cells were lysed in NP-40 lysis buffer (1% NP-40, 20 mM Tris-HCl, pH 8.0, 137 mM NaCl, 10% glycerol, 2 mM EDTA, 1 mM sodium orthovanadate, 10 µg/ml Aprotinin, 10 µg/mL

Leupeptin, and 10 µg/mL Pepstatin). For immunoprecipitation, lysates were incubated overnight at 4°C with a specific anti-EGF-R primary antibody followed by a 1-hour incubation with Protein G-agarose beads (Santa Cruz Biotech, CA). Then the beads were washed 4 times in lysis buffer, and a 2x sample buffer (100mM Tris-HCl pH6.8, 4%

SDS, 0.2% (wt/vol) bromophenol blue and 20% glycerol) was added and boiled for 5 minutes at 95°C to dissociated precipitated proteins from the beads. The immunoprecipitates were resolved by 8% SDS-PAGE, and blotted for phosphorylated

Tyrosine HRP (pTyr-HRP) (R&D Systems) and EGF-R (NeoMarkers).

Cell surface biotinylation

For biotinylation of cell membrane proteins, cells were washed with ice-cold PBS and then incubated for 1 hour at 4°C with Sulfosuccinimidobiotin (EZ-Link Sulfo-NHS-

Biotin; Pierce Biotechnology, Rockford, IL) freshly dissolved in H2O (2 mM final concentration). The reaction was blocked by rinsing the cells 3 times with 100 mM glycine in PBS. Cell lysates were incubated for 1 hour at 4°C with streptavidin-agarose

(Sigma-Aldrich) to pull down biotinylated membrane proteins. The supernatant was stored for further manipulation (first fraction). Streptavidin-biotin complexes (surface protein-enriched fraction) were incubated overnight at 4°C with a primary antibody

52 specific for EGF-R. The suspension was then incubated for 1 hour with protein A conjugated to magnetic beads (Millipore), and immunocomplexes were captured using a dedicated magnet. To estimate the amount of EGF-R not biotinylated (therefore not present on the plasma membrane), the first fraction was immunoprecipitated using an antibody specific for EGF-R.

Cell viability and proliferation

NSCLC cells were plated in 100 µl of RPMI with 5% FBS into 96-well cell culture plates (5,000 cells/well). Cell viability was determined by using the 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide cell MTT assay kit (Roche) according to the manufacturer’s instructions. Briefly, 10 µl of MTT labeling reagent was added to each well and incubated at 37°C for 4 hours in the incubator. Then 100 µl solubilization solution was added and incubated at 37°C overnight. The spectrophotometric absorbance was measured at 575 nm using a fluorescent multiplate reader (Polar Star Omega, Ortemberg, Germany). The results were expressed as mean ±

SD of 3 independent experiments. In parallel, dye exclusions and cell count assays were performed. Proliferation was analyzed by bromodeoxyuridine (Br-dU) incorporation

(FITC BrdU Flow kit, BD Pharmingen, San Diego, CA)/7-aminoactinomycin D (7-AAD,

Sigma-Aldrich) staining. Briefly, 100 µl of BrdU (1mM in DPBS) was added to each 10 cm dish of experimental NSCLC cells containing 10ml of culture media, allowing 4 hours incubation time for the incorporation to occur. One million cells were then harvested in FACs tubes, washed once with Staining Buffer (1 x DPBS + 3% heat

53 inactivated Fetal Bovine Serum + 0.09% sodium azide) followed by fixation and permeabilization with 100 µl of BD Cytofix/Cytoperm Buffer for 15 minutes at room temperature and washed once with 1ml of 1 x BD Perm/Wash Buffer. Then cells were incubated with 100 µl BD Cytoperm Plus Buffer for 10 minutes on ice, washed once with

1 x BD Perm/Wash Buffer, and followed by re-fixation with 100 µl of BD

Cytofix/Cytoperm Buffer for 5 minutes at room temperature and washed with 1 x BD

Perm/Wash Buffer. After fixation and permeablization, the cells were treated with 100 µl of DNase (300 µg/ml) to expose incorporated BrdU for 1 hour at 37C, and washed with 1 x BD Perm/Wash Buffer. Then cells were resuspended in 50 µl of BD Perm/Wash Buffer containing diluted FITC conjugated anti-BrdU antibody (1:50) for 20 minutes at room temperature in the dark, followed by washing the cells with 1 x BD Perm/Wash Buffer.

For total DNA staining for cell cycle analysis, the cells were resuspended with 20 µl of the 7-amino-actinomycin D (7-AAD) solution. Finally, 1 ml of Staining Buffer was added to each tube to resuspend cells for FACS analysis. FACS analysis was performed using a BD FACS Canto II instrument (Becton Dickinson, San Jose, CA) measuring

30,000 events for each sample at a rate no greater than 400 events/sec.

Statistical analysis

We performed statistical analysis by Student’s t test. Values were considered statistically significant at P < 0.05 in 2-tailed tests. Survival curves were analyzed using the Mantel-Cox and Gehan-Breslow-Wilcoxon tests (Prism 5 software).

Table 3.4 List of all antibodies used in this dissertation work

Target Type Company Cat. Number Active Notch-1 Rabbit polyclonal Cell Signaling 2421 (Val1744) Acetyl Histone H3 Rabbit polyclonal Millipore 06-599B

IGF1Rβ(C-20) Rabbit polyclonal Santa Cruz Biotech. 713

GFP Rabbit polyclonal Abcam Ab290

HIF1α (H-206) Rabbit polyclonal Santa Cruz Biotech. 10790

IgG Rabbit polyclonal Santa Cruz Biotech. 2027

IgG Rabbit polyclonal Millipore PP64B

IgG Mouse polyclonal AbCam 18448

MAML1 Rabbit polyclonal Millipore AB5975

Notch-1 (C-20) Rabbit polyclonal Santa Cruz Biotech. 6014

P300 (H-272) Rabbit polyclonal Santa Cruz Biotech. 8981

p-Akt (Ser473) Rabbit polyclonal Cell Signaling 4058 p-Akt (Ser473) Rabbit monoclonal Cell Signaling 4060 (D9E) Akt-1 (G-5) Mouse monoclonal Santa Cruz Biotech. 55523

pERK 1/2 Rabbit polyclonal Cell Signaling 4370

ERK 2 Mouse monoclonal Santa Cruz Biotech. 154

EGF-R Mouse monoclonal NeoMarkers 400-P1ABX

pTyr-HRP Mouse monoclonal R&D Systems 841403

Ubiquitin (PD41) Mouse monoclonal Santa Cruz Biotech. 8017

Caspase-3 Rabbit polyclonal Cell Signaling 9662

GLUT-1 Mouse monoclonal Abcam Ab40084

GAPDH Mouse monoclonal Chemicon MAB374 Rabbit IgG labeling AlexaFluor 568 Zenon Z25306 kit Rabbit IgG labeling AlexaFluor 488 Zenon Z25302 kit AlexaFluor 488 Mouse IgG labeling Zenon Z25002

CHAPTER 4

RESULTS

Previous work from our laboratory indicates that Notch-1 is activated and stabilized by hypoxia inducible factor-1α (HIF-1α) under hypoxic conditions. Activated

Notch-1 can then provide survival signals to NSCLC cells by positively regulating insulin-like growth factor-1 receptor (IGF-1R) to activate the AKT pathway.

Additionally, this effect is exacerbated by the fact that Notch-1 depresses phosphatase and tensin homolog (PTEN) expression. We have also found that inhibition of Notch-1 signaling in NSCLC cells causes cell death specifically under hypoxia. The results are summarized in Figure 4.1.

56 57

Figure 4.1 Schematic of Notch-1 activation in hypoxia, which provides survival signals to NSCLC cells through activating Akt-1. In hypoxic tumor microenvironment, HIF-1α is stabilized. By binding to and stabilizing intracellular Notch-1 (Notch-1IC), HIF-1α enhances the transcriptional activity of Notch-1IC, which in turn increases the expression IGF-1R and decreases PTEN expression. Both of these events lead to hyperactivation of Akt-1, which provides survival signals to NSCLC cells (and possibly other solid tumors) specifically under hypoxic conditions.

Notch-1 directly regulates IGF-1R transcription through its association with the

+1478 DNA region

To further investigate the mechanism by which Notch-1 regulates the expression level of IGF-1R, we performed experiments with NSCLC cells by transfecting known downstream targets of Notch-1: HES-1, HES-5, HEY-1, HEY-L and c-MYC (Palomero et al., 2006)). However, none of these targets seemed to regulate the level of IGF-1R, neither at the mRNA level nor at the protein level. Therefore, we then thought that

Notch-1 might regulate the expression of IGF-1R by directly binding to its promoter. We have analyzed the sequence in the DNA region surrounding the IGF-1R initiation codon

(position +1), and found two regions containing canonical RBP-Jkappa/CBF1 consensus sequence (CGTGGGAA): one at position -612 and a second at position +1478. A schematic overview of these regions is represented in Figure 4.2.

To this point, we performed chromatin immunoprecipitation (ChIP) experiment targeting these two regions using an antibody recognizing Notch-1IC. We found that there is no association of Notch-1IC at region -612 in seven independent experiments.

Instead, we detected that Notch-1IC was present at region +1478 (Figure 4.3). We then performed ChIP experiments on Notch transcriptional coactivators: Mastermind-like-1

(MAML-1) (Nam et al., 2006) and p300 (Fryer et al., 2002). The ChIP results suggested that both MAML-1 and p300 bind to DNA region +1478 (Figure 4.4a). Furthermore, when we transfected NSCLC cells with a plasmid encoding dominant negative MAML-1

(DN-MAML-1), we did not detect the binding of Notch-1, MAML-1 and p300 at DNA region +1478, and the amount of acetylated histone H3 (AcH3) was also significantly reduced over that region (Figure 4.4a).

In the DN-MAML-1 plasmid, the central binding domain of MAML-1 for p300 is replaced by green fluorescent protein (GFP) (Weng et al., 2003). Therefore it would not be able to bind p300, resulting in less acetylated histone H3. These results suggested that a complete coactivator complex containing MAML-1 and p300 was needed to allow stable Notch-1IC association with DNA region +1478. Alternatively, the absence of p300 activity in this site may decrease the overall acetylation of histone H3 (Figure 4.4a), leading to a local rearrangement of the chromatin structure that could render this region less accessible to transcription factors/coactivators. Furthermore, the transfection of

NSCLC cells with DN-MAML-1 reduced IGF-1R expression significantly (Figure 4.4b).

Collectively, the above data suggest that Notch-1 directly regulates IGF-1R transcription through its association with the +1478 DNA region.

Figure 4.2 Schematic representation of the promoter region of the IGF-1R gene. Gray: region upstream of the initiation of transcription; darker pink: 5’ UTR; lighter pink: first fragment of the CDS (+1, initiation of translation); green: 5’ region of the first intron. Note that there are two canonical CBF-1 binding sites: at regions -612 and +1478. (a) At region +1478, but not -612, we detected the association of Notch-1, MAML1 and p300. (b) Upon transfection of a plasmid that encodes dominant negative MAML1 (DN- MAML-1), we no longer detected any association of these proteins at region +1478. Additionally, the amount of acetylated histone H3 was significantly reduced.

Figure 4.3 Notch-1 associates with the +1478 region of the IGF-1R gene. ChIP experiment was performed on A549 cells. DNA fragments were analyzed using end-point PCR with specific primers that amplify DNA region +1478 (See Table 3.1 for detailed primer sequences). Different pre-immune IgGs were used in this experiment: one from Millipore (IgG M) and the other from Santa Cruz Biotechnology (IgG SC). Notch-1 was immunoprecipitated using the antibody C20 (Santa Cruz Biotechnology).

Figure 4.4 Notch-1, MAML-1 and p300 associate with DNA region +1478 of IGF-1R, and transfection of NSCLC cells with plasmid DN-MAML-1 disrupts these associations. (a) A549 cells were transfected with either control plasmid (black columns) or a plasmid that encodes dominant negative MAML-1 (DN-MAML-1) (gray columns). The ChIP experiments were performed using antibodies against acetylated histone H3 (AcH3), Notch-1, MAML-1 and p300 for immunoprecipitation. The amount of associated DNA fragment at region +1478 (62 bp-long products) was quantified with Real-Time PCR method using Taqman primers (see Table 3.1 for detailed primer sequences). The IgG column represents the average of three different pre-immune IgGs. Each column represents the average of four (black columns) or five (gray columns) independent experiments. The IgG column represents nine independent immunoprecipitation experiments. (b) Western Blot analysis of A549 cells transfected with either control plasmid or DN-MAML-1 (the binding domain for p300 is replaced by green fluorescence protein (GFP)). At 24hr after transfection, the protein levels of IGF- 1R were substantially decreased compared to cells transfected with control plasmid. Since DN-MAML-1 seemed to be toxic to NSCLC cells following transfection, all the experiments shown here were conducted 24hr post transfection.

Establishment of mouse orthotopic models with advanced NSCLC

To generate the orthotopic mouse models, we injected 2.5 x 106 NSCLC cells

(A549 and H1437) transduced with lentiviral vector expressing firefly luciferase (for tumor burden imaging purpose) (see Material and Methods for detailed cloning and transduction information) into the lateral tail vein of SCID mice. By intraperitoneal (i.p.) injection of the mice with luciferin, the luciferase substrate, we were able to detect the tumor cells using a Xenogen luminescence system. Most of the injected cells home to the lung, which is the original site of NSCLC development. The presence of tumor cells in mice lungs was readily visible 5 minutes after tail-vein injection. The tumor nodules were clearly detected throughout the lungs of mice 5-weeks after injection (Figure 4.5a).

Figure 4.5a shows the upper third of the left lung of a representative mouse (H&E staining) at the initiation of therapy. There are two large tumor nodules, indicating the advanced tumor progression stage at the time when mice were euthanized. Moreover, we transduced A549 cells with the luciferase gene in a lenti-viral vector, and were therefore able to detect luminescence after giving luciferin substrate to the mice with lung tumors.

With this method, we were able to normalize the mice lung tumor burden before the initiation of therapy. Before we administered treatment, mice were imaged for light emission and only the mice within ±20% of average light emission in thorax and little or no light emission at the site of injection were assigned to treatment groups. Figure 4.5b shows a representative example of the preliminary imaging analysis to assess equal tumor burden prior to therapy initiation.

In order to make sure that the tumor burdens of the experimental mice are equal before treatments, we performed quantitative PCR analysis at this 5-week time point.

Specifically, Quantitative PCR using species-specific primers for the glyceraldehyde 3- phosphate dehydrogenase (GAPDH) gene on total DNA extracted from 12 mice lungs, showed that the mean ratio of human/mouse cells of 0.0495 ± 0.0029 (Table 4.1). This result suggests that there was minimum tumor burden variation at the beginning of therapy.

In addition to A549 and H1437 cells, we also tried to inject NSCLC cell lines

H1299 and H838 into the tail veins of these mice. However, neither cell line generated acceptable tumor burden in the mice lungs. Five weeks after injection, we could not detect any tumor mass in the mice injected with H838, and only detected brain tumors in the H1299 injected mice.

Figure 4.5 Characterization of lung tumors in advanced NSCLC mice model at the time of therapy initiation (5-weeks after tail vein injection). (a) Histopathological (H&E) evaluation of an 8 µm-thick lung section (upper third of the left lung lobe) from a representative mouse five weeks after tail vein injection of A549 cells. (b) Imaging analysis of tumor burden of mice injected with A549 cells prior to therapy. The mice that showed substantial light emission at or around the site (tail) of injection were discarded from study.

Table 4.1 Quantitative PCR analysis of the human/mouse GAPDH ratio in the whole left lungs of 12 representative animals 5-weeks after tail vein injection.

Hypoxia/Notch-1/IGF-1R/Akt-1 axis also takes place in vivo

Previous findings in our lab suggested that Notch-1 is activated in hypoxia, which provided survival signals to NSCLC cells by upregulating IGF-1R expression and downregulating PTEN levels. Both of these events led to hyperactivation of Akt-1 signaling pathway. We wanted to confirm these findings in vivo using a mouse model with human metastatic NSCLC (see Material and methods for details). We performed a series of co-immunofluorescent staining on snap frozen, 8 µm-thick tissue slides containing sections of mice lungs. To stain hypoxic regions in mice lung sections, we used glucose transporter-1 (GLUT-1) as a hypoxia marker (Behrooz and Ismail-Beigi,

1999; Ebert et al., 1995). We confirmed the previous finding that GLUT-1 is a good hypoxia marker, since Western Blot and immunofluorescent staining showed elevated

GLUT-1 levels upon incubating the cells under hypoxic conditions (Figure 4.6).

Strikingly, co-immunofluorescent staining of sections obtained from tumor-bearing lungs of SCID mice injected with A549 cells through the tail vein showed that only hypoxic tumor areas expressed Notch-1IC, maximally expressed IGF-1R, and appeared to be the only tumor areas in which Akt-1 is phosphorylated (Figure 4.7a-d). By contrast, PTEN was never expressed in the tumor hypoxic areas that expressed hypoxic marker GLUT-1

(Figure 4.7e). These in vivo results are consistent with our previous in vitro findings and the hypothesis that Notch-1 is activated under hypoxia, resulting in activation of Akt-1 through upregulation of IGF-1R and downregulation of PTEN expression.

Figure 4.6 GLUT-1 is an effective hypoxia marker in NSCLC cells. (a-d) immunofluorescent staining of A549 cells cultured in normoxia (21% oxygen) (a and c) and hypoxia (1% oxygen) (b and d) conditions. The primary antibodies used during the staining procedure are pre-immune normal rabbit IgG (negative control) (a and b) and rabbit anti-GLUT-1 antibody (c and d). (e) Western blot analysis of GLUT-1 and Actin (loading control) performed on H1299 cells cultured at the indicated oxygen concentrations.

Figure 4.7 Activated Notch-1 (Notch-1IC) is expressed in hypoxic tumor cells, together with IGF-1R and phosphorylated Akt-1 (Ser473) but not with PTEN. Co- immunofluorescence stainings of the indicated proteins on 8 µm-thick slides obtained

68 from tumor-bearing lungs of SCID mice injected with A549 cells through the tail vein. (a) Co-expression of GLUT-1 and HIF-1α. (b) Co-expression of GLUT-1 and Notch-1IC. (c) Expression of IGF-1R is maximal in GLUT-1 expressing (hypoxic) tumor cells. (d) Phosphorylated Akt-1 (S473) is detected in GLUT-1 expressing (hypoxic) tumor cell only. (e) PTEN is expressed in GLUT-1 negative tumor cell only. Note that GLUT-1 and PTEN positive cells appear to be contiguous. This is likely an event due to the lateral, asymmetric nature of Notch signaling. Bar, 20 µm.

AIM1: DETERMINE WHETHER INHIBITION OF THE NOTCH-1/IGF-1R/AKT

AXIS IS A SUITABLE THERAPEUTIC TOOL TO TARGET HYPOXIC NSCLC

Survival analysis of mice treated with GSI, IGF-1R inhibitory antibody or Akt inhibitor, alone or in combination

Since Notch-1 signaling is activated under hypoxia and provides critical prosurvival signals to NSCLC cells by activating IGF-1R/Akt signaling, we questioned whether interfering with the Notch-1/IGF-1R/Akt axis could deplete tumor mass in hypoxic areas in the established orthotopic, very advanced (metastatic) NSCLC mouse model. We performed the in vivo experiment with GSI (MRK-003), an inhibitory antibody targeting the IGF-1R (MK-0646), and a pan AKT inhibitor (MK-2206). MRK-

003 interacts with γ-secretase component presenilin and blocks its catalytic properties.

MK-0646 is a humanized IgG1 monoclonal antibody (mAb) against IGF-1R, which induces receptor internalization and degradation (Broussas et al., 2009). There are many different IGF-1R monoclonal antibodies (including MK-0646) currently in clinical development (phase I to III) of breast, colorectal, pancreatic and lung cancer (Gualberto and Pollak, 2009; Hewish et al., 2009; Weroha and Haluska, 2008). Many Phase I and II

69 studies using different monoclonal antibodies against IGF-1R suggested promising clinical outcomes and minimal toxicity (Haluska et al., 2007; Karp et al., 2009; Tolcher et al., 2009). However, a recent Phase III study using figitumumab (another IGF-1R inhibitory antibody) in combination with paclitaxel and carboplatin in NSCLC patient showed disappointing activity and metabolic toxicity (mainly hyperglycaemia) (Pollak,

2012). Therefore, we planned our experiment using MK-0646 in combination with GSI that targets hypoxic tumor regions, only in a window of time during therapeutic interventions. MK-2206 is a pan Akt inhibitor that interacts with the PH domain and prevents Akt interaction with PIP3.

Notch inhibition results. To inhibit Notch-1 signaling, we used γ-secretase inhibitor (GSI) MRK-003 (Merck & Co., Inc., Whitehouse Station, NJ) in this study. The therapy regimen was 100 mg/kg of MRK-003 in 0.05% methylcellulose administered via oral gavaging three consecutive days a week. The therapy combinations were as follows:

GSI plus cisplatin (3 mg/kg, one weekly i.p. injection for four consecutive weeks), GSI plus the fully humanized monoclonal antibody MK-0646 targeting the human IGF-1R

(10 mg/kg, one weekly injection). There were 16 experimental animals assigned to each treatment group. The results are plotted as survival curves shown in Figure 4.8, left panel, while the statistical analysis of median survivals are shown in Table 4.2. Due to the fact that we have numerous comparisons between treatment groups (in total we performed 40 comparisons), it is difficult to assess the most effective therapeutic regimen because of the required Bonferroni correction. Specifically, to determine the best therapeutic regimen over other groups of treatments, the drug treatment group should have had a

70 corrected P value of 0.0125 or less. Nonetheless, the statistical analysis indicated that all treatments significantly prolonged the median survival of mice compared with mice treated with a vehicle (methylcellulose) control. In addition, we observed that combination therapies of GSI plus cisplatin and GSI plus MK-0646 were statistically better treatment regimens compared to a single agent alone.

Akt pan-inhibition results. The therapeutic regimen of Akt pan-inhibition was

80mg/kg (3 consecutive days per weeks) of pan Akt inhibitor MK-2206 (Merck & Co).

We have tested this dosage in tumor free SCID mice, and this was the maximum tolerated dose. However, in the tumor bearing mice, we found that the MK-2206 treated mice reached their endpoint (20% loss of body weight) within 8 days after initiation of therapy.

This was due to glucose intolerance caused by Akt depletion, since the plasma glucose concentration in mice at end point was 316 ± 23 mg/dl (average of 8 mice ± standard deviation).

IGF-1R inhibiton results. To inhibit IGF-1R signaling, we treated tumor-bearing mice with IGF-1R inhibitory antibody MK-0646 (Merck & Co) with the same regimen described earlier. For the combination therapy, MK-0646 plus erlotinib (100mg/kg, 3 times per week via oral gavaging) were administered together at the beginning of treatment. For the sequential therapy, mice were treated with MK-0646 for three weeks, followed by MK-0646 plus erlotinib treatment. The IGF-1R inhibitory antibody MK-

0646 alone substantially improved the median survival of treated mice compared with that of vehicle control treated mice. Again, due to the multiple comparisons it is difficult to assess the best therapeutic regimen (Figure 4.8 right panel, and statistic analysis of

71 median survivals are shown in Table 4.2). Interestingly, we found that sequential treatment (MK-0646 followed by the addition of erlotinib after 3 weeks) significantly prolonged mice survival compared with the combination therapy (MK-0646 plus erlotinib from therapy initiation) or a single agent alone. The experiment design and rationales are discussed later.

Figure 4.8 Survival curves of mice injected with A549 and H1437 cells treated with indicated agents. Median survivals of A549-injected mice (days): control (C), 9.0; cisplatin, 12.5; MRK-003 (GSI), 15.5; GSI+cisplatin, 19.5; GSI+MK-0646, 26.0; erlotinib, 16.0; MK-0646, 25.50; MK-0646+erlotinib, 15.0; 3 weeks of MK-0646 as single agent then combination of MK-0646 and erlotinib, 36.5. Median survivals of H1437-injected mice (days): control, 8.0; cisplatin, 21.0; MRK-003 (GSI), 30.0; GSI+cisplatin, 38.0; GSI+MK-0646, 42.0; erlotinib, 19.0; MK-0646, 26.0; MK-

0646+erlotinib, 18.5; 3 weeks of MK-0646 as single agent then combination of MK-0646 and erlotinib, 35.0.

Table 4.2 Statistical analysis of the mice survival curves shown in Figure 4.8.

A549

Median Survival C 9.000; cisplatin 12.50 Ratio 0.7200 95% CI of ratio 0.2018 to 1.238 Hazard Ratio Ratio: 2.435 95% CI of ratio 1.096 to 5.407 Log-rank (Mantel Cox) Test Chi square: 4.78; P value: 0.0288 Gehan-Breslow-Wilcoxon Test Chi square: 4.121; P value: 0.0424

Median Survival C 9.000; GSI 15.50 Ratio 0.5806 95% CI of ratio 0.06245 to 1.099 Hazard Ratio Ratio: 3.718 95% CI of ratio 1.642 to 8.420 Log-rank (Mantel Cox) Test Chi square: 9.914; P value: 0.0016 Gehan-Breslow-Wilcoxon Test Chi square: 8.801; P value: 0.0030

Median Survival C 9.000; GSI+cisplatin 19.50 Ratio 0.4615 95% CI of ratio -0.06745 to 0.9905 Hazard Ratio Ratio: 10.51 95% CI of ratio 4.260 to 25.92 Log-rank (Mantel Cox) Test Chi square: 26.08; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 24.79; P value: <0.0001

Median Survival C 9.000; GSI+MK-0646 26.00 Ratio 0.3462 95% CI of ratio -0.1720 to 0.8643 Hazard Ratio Ratio: 21.55 95% CI of ratio 7.947 to 58.45 Log-rank (Mantel Cox) Test Chi square: 36.39; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 29.21; P value: <0.0001

Median Survival cisplatin 12.50; GSI 15.50

Ratio 0.8065 95% CI of ratio 0.3064 to 1.307 Hazard Ratio Ratio: 1.668 95% CI of ratio 0.7594 to 3.663 Log-rank (Mantel Cox) Test Chi square: 1.624; P value: 0.2026 Gehan-Breslow-Wilcoxon Test Chi square: 0.8228; P value: 0.3644

Median Survival cisplatin 12.50; GSI+cisplatin 19.50 Ratio 0.6410 95% CI of ratio 0.1311 to 1.151 Hazard Ratio Ratio: 4.104 95% CI of ratio 1.738 to 9.690 Log-rank (Mantel Cox) Test Chi square: 10.38; P value: 0.0013 Gehan-Breslow-Wilcoxon Test Chi square: 9.620; P value: 0.0019

Median Survival cisplatin 12.50; GSI+MK-0646 26.00 Ratio 0.4808 95% CI of ratio -0.01932 to 0.9809 Hazard Ratio Ratio: 11.47 95% CI of ratio 4.255 to 30.93 Log-rank (Mantel Cox) Test Chi square: 23.25; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 21.68; P value: <0.0001

Median Survival GSI 15.50; GSI+cisplatin 19.50 Ratio 0.7949 95% CI of ratio 0.2849 to 1.305 Hazard Ratio Ratio: 2.604 95% CI of ratio 1.169 to 5.801 Log-rank (Mantel Cox) Test Chi square: 5.485; P value: 0.0192 Gehan-Breslow-Wilcoxon Test Chi square: 5.833; P value: 0.0157

Median Survival GSI 15.50; GSI+MK-0646 26.00 Ratio 0.5962 95% CI of ratio 0.09606 to 1.096 Hazard Ratio Ratio: 3.206 95% CI of ratio 1.399 to 7.349 Log-rank (Mantel Cox) Test Chi square: 7.578; P value: 0.0059 Gehan-Breslow-Wilcoxon Test Chi square: 11.10; P value: 0.0009

Median Survival GSI+cisplatin 19.50; GSI+MK-0646 26.00 Ratio 0.7500 95% CI of ratio 0.2400 to 1.260

Hazard Ratio Ratio: 1.042 95% CI of ratio 0.4958 to 2.190 Log-rank (Mantel Cox) Test Chi square: 0.01183; P value: 0.9134 Gehan-Breslow-Wilcoxon Test Chi square: 2.275; P value: 0.1315

Median Survival C 9.000; erlotinib 16.00 Ratio 0.5625 95% CI of ratio 0.06241 to 1.063 Hazard Ratio Ratio: 4.015 95% CI of ratio 1.684 to 9.571 Log-rank (Mantel Cox) Test Chi square: 9.832; P value: 0.0017 Gehan-Breslow-Wilcoxon Test Chi square: 12.03; P value: 0.0005

Median Survival C 9.000; MK-0646 25.50 Ratio 0.3529 95% CI of ratio -0.1472 to 0.8530 Hazard Ratio Ratio: 28.79 95% CI of ratio 9.435 to 87.86 Log-rank (Mantel Cox) Test Chi square: 34.84; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 28.61; P value: <0.0001

Median Survival C 9.000; MK-0646+erlotinib 15.00 Ratio 0.6000 95% CI of ratio 0.09991 to 1.100 Hazard Ratio Ratio:3.923 95% CI of ratio 1.642 to 9.374 Log-rank (Mantel Cox) Test Chi square: 9.460; P value: 0.0021 Gehan-Breslow-Wilcoxon Test Chi square: 12.42; P value: 0.0004

Median Survival C 9.000; MK-0646 then erlotinib 36.50 Ratio 0.2466 95% CI of ratio -0.2535 to 0.7467 Hazard Ratio Ratio: 34.03 95% CI of ratio 10.92 to 106.0 Log-rank (Mantel Cox) Test Chi square: 37.02; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 29.91; P value: <0.0001

Median Survival erlotinib 16.00; MK-0646 25.50 Ratio 0.6275 95% CI of ratio 0.1274 to 1.128 Hazard Ratio Ratio: 7.872 95% CI of ratio 3.067 to 20.20

Log-rank (Mantel Cox) Test Chi square: 18.41; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 17.46; P value: <0.0001

Median Survival erlotinib 16.00; MK-0646+erlotinib 15.00 Ratio 1.067 95% CI of ratio 0.5666 to 1.567 Hazard Ratio Ratio: 0.7894 95% CI of ratio 0.3747 to 1.663 Log-rank (Mantel Cox) Test Chi square: 0.3868; P value: 0.5340 Gehan-Breslow-Wilcoxon Test Chi square: 0.1714; P value: 0.6788

Median Survival erlotinib 16.00; MK-0646 then erlotinib 36.50 Ratio Ratio 0.4384 95% CI of ratio -0.06174 to 0.9384 Hazard Ratio Ratio: 29.54 95% CI of ratio 9.856 to 88.56 Log-rank (Mantel Cox) Test Chi square: 36.54; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 30.02; P value: <0.0001

Median Survival MK-0646 25.50; MK-0646+erlotonib 15.00 Ratio 1.700 95% CI of ratio 1.200 to 2.200 Hazard Ratio Ratio: 0.07699 95% CI of ratio 0.02814 to 0.2106 Log-rank (Mantel Cox) Test Chi square: 24.93; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 22.53; P value: <0.0001

Median Survival MK-0646 25.50; MK-0646 then erlotinib 36.50 Ratio 0.6986 95% CI of ratio 0.1985 to 1.199 Hazard Ratio Ratio: 2.524 95% CI of ratio 1.090 to 5.842 Log-rank (Mantel Cox) Test Chi square: 4.672; P value: 0.0307 Gehan-Breslow-Wilcoxon Test Chi square: 5.882; P value: 0.0153

Median Survival MK-0646+erlotinib 15.00;MK-0646 then erlotinib 36.50 Ratio 0.4110 95% CI of ratio -0.08913 to 0.9111 Hazard Ratio Ratio: 30.71 95% CI of ratio 10.11 to 93.28 Log-rank (Mantel Cox) Test Chi square: 36.49; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 29.84; P value: <0.0001

H1437

Median Survival C 8.000; cisplatin 21.00 Ratio 0.3810 95% CI of ratio -0.1134 to 0.8753 Hazard Ratio Ratio: 39.06 95% CI of ratio 11.63 to 131.2 Log-rank (Mantel Cox) Test Chi square: 35.17; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 29.23; P value: <0.0001

Median Survival C 8.000; GSI 30.00 Ratio 0.2667 95% CI of ratio -0.2277 to 0.7611 Hazard Ratio Ratio: 33.11 95% CI of ratio 10.06 to 109.0 Log-rank (Mantel Cox) Test Chi square: 33.17; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 28.37; P value: <0.0001

Median Survival C 8.000; GSI+cisplatin 38.00 Ratio 0.5526 95% CI of ratio 0.05254 to 1.053 Hazard Ratio Ratio: 20.12 95% CI of ratio 6.801 to 59.50 Log-rank (Mantel Cox) Test Chi square: 29.43; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 25.73; P value: <0.0001

Median Survival C 8.000; GSI+MK-0646 42.00 Ratio 0.1905 95% CI of ratio -0.3039 to 0.6849 Hazard Ratio Ratio: 39.06 95% CI of ratio 11.63 to 131.2 Log-rank (Mantel Cox) Test Chi square: 35.17; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 35.17; P value: <0.0001

Median Survival cisplatin 21.00; GSI 30.00 Ratio 0.700 95% CI of ratio 0.1999 to 1.200 Hazard Ratio Ratio: 3.020 95% CI of ratio 1.2610 to 7.241 Log-rank (Mantel Cox) Test Chi square: 6.138; P value: 0.0132 Gehan-Breslow-Wilcoxon Test Chi square: 3.325; P value: 0.0682

Median Survival cisplatin 21.00; GSI+cisplatin 38.00 Ratio 0.5526 95% CI of ratio 0.05254 to 1.053 Hazard Ratio Ratio: 20.12 95% CI of ratio 6.801 to 59.50 Log-rank (Mantel Cox) Test Chi square: 29.43; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 25.73; P value: <0.0001

Median Survival cisplatin 21.00; GSI+MK-0646 42.00 Ratio 0.5000 95% CI of ratio -9.128e-005 to 1.000 Hazard Ratio Ratio: 16.03 95% CI of ratio 5.558 to 46.25 Log-rank (Mantel Cox) Test Chi square: 26.35; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 24.04; P value: <0.0001

Median Survival GSI 30.00; GSI+cisplatin 38.00 Ratio 0.7895 95% CI of ratio 0.2894 to 1.290 Hazard Ratio Ratio: 3.089 95% CI of ratio 1.349 to 7.074 Log-rank (Mantel Cox) Test Chi square: 7.121; P value: 0.0076 Gehan-Breslow-Wilcoxon Test Chi square: 7.790; P value: 0.0053

Median Survival GSI 30.00; GSI+MK-0646 42.00 Ratio 0.7143 95% CI of ratio 0.2142 to 1.214 Hazard Ratio Ratio: 4.534 95% CI of ratio 1.882 to 10.93 Log-rank (Mantel Cox) Test Chi square: 11.35; P value: 0.0008 Gehan-Breslow-Wilcoxon Test Chi square: 9.153; P value: 0.0025

Median Survival GSI+cisplatin 38.00; GSI+MK-0646 42.00 Ratio 0.9048 95% CI of ratio 0.4047 to 1.405 Hazard Ratio Ratio: 1.105 95% CI of ratio 0.5179 to 2.357 Log-rank (Mantel Cox) Test Chi square: 0.06641; P value: 0.7966 Gehan-Breslow-Wilcoxon Test Chi square: 0.08055; P value: 0.7765

Median Survival C 8.000; erlotinib 19.00 Ratio 0.4211

95% CI of ratio -0.07904 to 0.9211 Hazard Ratio Ratio: 34.11 95% CI of ratio 10.71 to 108.7 Log-rank (Mantel Cox) Test Chi square: 35.65; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 29.53; P value: <0.0001

Median Survival C 8.000; MK-0646 26.00 Ratio 0.3077 95% CI of ratio -0.1867 to 0.8021 Hazard Ratio Ratio: 30.97 95% CI of ratio 9.752 to 98.34 Log-rank (Mantel Cox) Test Chi square: 33.91; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 27.90; P value: <0.0001

Median Survival C 8.000; MK-0646+erlotinib 18.50 Ratio 0.4324 95% CI of ratio -0.06766 to 0.9325 Hazard Ratio Ratio: 34.11 95% CI of ratio 10.71 to 108.7 Log-rank (Mantel Cox) Test Chi square: 35.65; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 29.53; P value: <0.0001

Median Survival C 8.000; MK-0646 then erlotinib 35.00 Ratio 0.2286 95% CI of ratio -0.2658 to 0.7230 Hazard Ratio Ratio: 30.97 95% CI of ratio 9.752 to 98.34 Log-rank (Mantel Cox) Test Chi square: 33.91; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 27.90; P value: <0.0001

Median Survival erlotinib 19.00; MK-0646 26.00 Ratio 0.7308 95% CI of ratio 0.2364 to 1.225 Hazard Ratio Ratio: 1.391 95% CI of ratio 0.5550 to 3.488 Log-rank (Mantel Cox) Test Chi square: 0.4959; P value: 0.4813 Gehan-Breslow-Wilcoxon Test Chi square: 0.1642; P value: 0.2001

Median Survival erlotinib 19.00; MK-0646+erlotinib 18.50 Ratio 1.027 95% CI of ratio 0.5269 to 1.527 Hazard Ratio Ratio: 0.6593

95% CI of ratio 0.2847 to 1.526 Log-rank (Mantel Cox) Test Chi square: 0.9459; P value: 0.3308 Gehan-Breslow-Wilcoxon Test Chi square: 1.399; P value: 0.2368

Median Survival erlotinib 19.00; MK-0646 then erlotinib 35.00 Ratio 0.5429 95% CI of ratio 0.04846 to 1.037 Hazard Ratio Ratio: 7.671 95% CI of ratio 2.796 to 21.04 Log-rank (Mantel Cox) Test Chi square: 15.66; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 17.38; P value: <0.0001

Median Survival MK-0646 26.00; MK-0646+erlotonib 18.50 Ratio 1.405 95% CI of ratio 0.9110 to 1.900 Hazard Ratio Ratio: 0.6077 95% CI of ratio 0.2677 to 1.380 Log-rank (Mantel Cox) Test Chi square: 1.418; P value: 0.2338 Gehan-Breslow-Wilcoxon Test Chi square: 2.960; P value: 0.0854

Median Survival MK-0646 26.00; MK-0646 then erlotinib 35.00 Ratio 0.7429 95% CI of ratio 0.2540 to 1.232 Hazard Ratio Ratio: 3.953 95% CI of ratio 1.604 to 9.744 Log-rank (Mantel Cox) Test Chi square: 8.920; P value: 0.0028 Gehan-Breslow-Wilcoxon Test Chi square: 8.198; P value: 0.0042

Median Survival MK-0646+erlotinib 18.50;MK-0646 then erlotinib 35.50 Ratio 0.5286 95% CI of ratio 0.03418 to 1.023 Hazard Ratio Ratio: 7.882 95% CI of ratio 2.974 to 20.89 Log-rank (Mantel Cox) Test Chi square: 17.24; P value: <0.0001 Gehan-Breslow-Wilcoxon Test Chi square: 18.47; P value: <0.0001

GSI treatment resulted in the effective killing of tumors in hypoxic areas

Our previous findings suggested that Notch-1 promotes tumor cell survival under hypoxia, and targeting Notch-1 signaling using either siRNA against Notch-1 or GSI specifically kills NSCLC cells under hypoxia in vitro. Here, we wanted to further understand whether GSI treatment indeed targets hypoxic tumor cells in the tumor- bearing mice. To answer this question, we excised the whole inferior right lung lobes of mice after treatment at their endpoints with GSI or vehicle control, and performed RNA isolation and quantitative RT-PCR (qRT-PCR) experiments using human-specific primers. The results showed that the expression levels of Notch targets HEY-L and HEY-

1 were significantly decreased in GSI treated mice compared to controls. Furthermore, the expression levels of two critical hypoxic markers glucose transporter 1 (GLUT-1) and vascular endothelial growth factor A (VEGF-A) were also greatly reduced in GSI treated mice lungs, suggesting a reduction in hypoxic regions upon GSI treatment. We also found decreased expression of IGF-1R, which confirmed our in vitro finding that Notch-1 activates IGF-1R expression in hypoxic conditions (Figure 4.9). To specifically visualize the cell death in hypoxic tumor regions, we performed TUNEL staining analysis on frozen lung sections obtained from GSI and vehicle treated mice lungs. The results from

TUNEL staining showed that GSI treatment promoted apoptosis in hypoxic tumor regions, whereas control mice did not show TUNEL signals in hypoxic tumor areas

(Figure 4.10a-d). Consistently with the TUNEL results, we also detected cleaved

(activated) caspase-3 in GSI treated mice lungs by GSI treatment (Figure 4.10e). All of

81 the above results suggest that GSI treatment caused specific apoptosis of hypoxic tumor cells.

Figure 4.9 GSI treatment caused significant reduction in the expression of Notch target genes, hypoxic markers and IGF-1R. Quantitative RT-PCR analysis of the indicated mRNAs in control and GSI (MRK-003) treated mice lungs. Total RNA was isolated from the whole inferior right lung lobe of mice after euthanasia, reverse transcribed, and measured by qPCR using human-specific primers. The expression levels of indicated mRNAs were normalized with human ribosomal protein RPL13A. Columns represent averages, while bars represent standard deviation. The experiment was performed on 4 control and 4 GSI (MRK-003) treated mice.

Figure 4.10 GSI treatment led to apoptosis of hypoxic NSCLC cells. Immunofluorescent staining of hypoxic marker GLUT-1 (green) and in situ TUNEL assay (red) in a control mouse (b) and in a mouse treated with GSI (MRK-003) (d). (a) and (c) are the same regions of (b) and (d) in bright field, respectively. 8 µm thick section slides of frozen lungs were first immunolabeled with a mouse monoclonal antibody of GLUT-1, followed by incubation with a goat anti-mouse IgG conjugated with Alexa Fluor 488 (green). Subsequently, the slides were subjected to the reaction mixture for TUNEL labeling containing terminal deoxynucleotidyl transferase and TMR red dUTP,

83 washed and mounted following the protocol provided by the TUNEL kit manufacturer. Original magnification: 200 X. Cells in (d) appear larger compared to (b). We propose that this effect could be an artifact due either to cell death or other reasons. (e) Western blot analysis performed on 100 µg of total lung lysate extracted from two control mice (GSI -) and two mice treated with MRK-003 (GSI +).

GSI treatment reduced brain and liver metastasis

It has been shown that Notch signaling promotes tumor migration and invasion in hypoxia. Using species-specific GAPDH primers, we measured the ratio of human versus mouse cells in total DNA extracted from the whole right brain and left liver lobe. Our results agree with the previous findings: brain and liver metastasis were indeed reduced after GSI treatment (Figure 4.11). Note that all mice were sacrificed at their endpoint and still the tumor burden in the brain and liver of GSI treated mice were less than the controls although the GSI treated mice lived much longer than the control mice.

Figure 4.11 GSI treatment caused significant reduction in brain and liver metastasis. Quantitative PCR (qPCR) analysis of the ratio of human versus mouse cells in DNAs extracted from brain and liver of control mice or GSI (MRK-003) treated mice

84 at the endpoint. This ratio was calculated by qPCR using primers specific for human or mouse GAPDH (this qPCR assay has been previously validated using known amount of cells). In the brain experiment, columns represent the average of 6 mice (controls: days at euthanasia 12.67 ± 4.18; GSI treated mice: days at euthanasia 17.83 ± 8.86). In the liver experiment we analyzed 6 control mice (days at euthanasia: 12.6 ± 4.22) and 7 GSI treated mice (days at euthanasia: 18.67 ± 7.84). Bars represent standard deviation.

GSI treatment decreased the expression of glycolytic enzymes

Recently, activation of Notch signaling has been shown to promote glycolysis through activating PI3K/Akt signaling. We measured the expression of glycolytic enzymes by quantitative RT-PCR. We consistently found that GSI treatment decreased the expression of glycolytic enzymes, such as hexokinase-2 (HK-2), pyruvate dehydrogenase kinase-2 (PDK-2), and aldolase-A (ALDOA) (Figure 4.12), suggesting glycolysis was inhibited upon GSI treatment.

Figure 4.12 GSI treatment decreased the expression of glycolytic enzymes. Quantitative RT-PCR of hexokinase-2 (HK-2), pyruvate dehydrogenase kinase-2 (PDK- 2), and aldolase-A (ALDOA) mRNA in control (C) or GSI (MRK-003) treated mice lungs. Columns represent the average of 4 mice. Bars represent standard deviation. P values are indicated.

AIM2: DELINEATE THE DETAILED MECHANISM BY WHICH TUMOR

EVASION OCCURRED DURING IGF-1R INHIBITORY ANTIBODY

TREATMENT

MK-0646 promoted IGF-1R degradation in a proteasome dependent manner

To identify the mechanism by which MK-0646 inhibits the IGF-1R signaling pathway, we performed in vitro experiments on NSCLC cells exposed to MK-0646 for different amounts of time. We found that MK-0646 treatment of control cells caused rapid loss of IGF-1R protein and effective loss of Akt-1 activation by IGF-1 administration to serum-starved cells. The transient ERK activation (measured by phosphorylated ERK1/2) within 1hr of IGF-1 stimulation was also lost in MK-0646 treated cells (Figure 4.13a). More importantly, we found that MK-0646 reduced IGF-1R protein levels within 1hr of exposure to MK-0646. This effect was mediated by the proteasome degradation pathway, because the MK-0646 mediated IGF-1R degradation was blocked in the presence of a proteasome inhibitor (MG-132), but not a lysosome inhibitor (NH4Cl) (Figure 4.13b).

Figure 4.13 MK-0646 promoted IGF-1R degradation in a proteasome dependent manner. (a) Western blot analysis of H1437 cells treated with or without purified IGF-1 (50ng/ml) or MK-0646 (2.5µg/ml). Cells were serum starved for 1 hour, and then exposed to IGF-1 treatment at the indicated time points. Note that IGF-1R was lost in the MK-0646 treated cells, and that IGF-1 activated Akt and ERK signaling were also inhibited. (b) Western blot analysis of H1437 cells treated with MK-0646 plus vehicle, or MK-0646 plus MG-132 (10µM) or NH4Cl (10mM) for the indicated times.

Tumor evasion occurred after IGF-1R inhibitory antibody (MK-0646) treatment

Our preliminary data suggested that MK-0646 had a tumoristatic effect on

NSCLC cells at the beginning of treatment. However, after the first two weeks, tumor cells resumed growth and ultimately led to the death of the experimental animals (Figure

4.14). The recurrent malignancy and eventual fatality are commonly seen in anti-cancer therapy in humans (with the exception of some hormone-driven malignancies).

Regardless of the therapeutic approach, tumors become resistant to adjuvant

87 radio/chemotherapy and eventually lead to the death of the patients. In this aim, we want to uncover the mechanism(s) that allow cancer cells to evade the deprivation of IGF-1R signaling.

Figure 4.14. Bioluminescence quantification of the tumor burden in the lungs of mice treated with IGF1-R inhibitory antibody (MK-0646). The Graphic shows the units of light recorded by an IVIS 100 at the indicated time points. Sixteen mice were injected with 0.25 mg of luciferin and imaged for 5 minutes.

Phosphorylated EGF-R (pEGF-R) and EGF-R total protein levels were increased upon MK-0646 treatment in vivo and in vitro

A recent study suggested that inhibition of Akt-1 signaling led to a compensatory activation of several receptor tyrosine kinases (RTKs), including EGF-R (Chandarlapaty et al., 2011). Since MK-0646 caused rapid loss of IGF-1R total protein and dramatic loss of Akt activation (Figure 4.13a), we wanted to investigate if tumor cells evaded the effect

88 of MK-0646 treatment through activating alternative RTKs. Proteomic RTK array analysis showed that the level of phosphorylated EGF-R (pEGF-R) was increased in the mouse lungs treated with IGF1-R inhibitory antibody (MK-0646) compared with controls

(Figure 4.15). As activation of EGF-R signaling pathways can lead to cell survival and proliferation, this is a potential mechanism underlying the tumor survival when IGF-1R signaling is impaired.

In order to delineate the detailed mechanism(s) by which EGF-R signaling is activated when IGF1-R signaling is impaired, we wanted to determine if we could achieve the same effect by inhibiting IGF1-R signaling in vitro. Briefly, we cultured

NSCLC cells in the tissue culture media containing MK-0646 for three weeks, and detected increased phosphorylated EGF-R (pEGF-R), as well as total protein level of

EGF-R. The similar results were obtained in tumors of mice treated with MK-0646 in vivo and in cells re-derived after necroscopy from lung tumors of mice treated with MK-

0646 (Figure 4.16c). This appeared to be a cell-autonomous phenomenon since A549 and

H1437 cells treated for three weeks with MK-0646 in vitro also showed increased levels of pEGF-R and total EGF-R (Figure 4.16a, c). However, the increased total EGF-R protein was not due to increased mRNA expression (which indeed was reduced) (Figure

4.16b).

Figure 4.15. Phosphorylated EGF-R (pEGF-R) was increased in IGF-1R inhibitory antibody (MK-0646) treated mouse lung compared to control. Representative human phospho-RTK array analysis of total protein lysates (250µg) obtained from control and IGF-1R inhibitory antibody treated mouse lungs at their endpoints. We performed a total of 4 assays and always detected a dramatic increase in EGF-R phosphorylation. In this particular example we also detected that ErbB3 and FGF-R3 increased phosphorylation upon MK-0646 treatment, but the only constant outcome that we identified was EGF-R hyperactivation. (b) Histogram quantification of selected phosphorylated proteins by measuring the mean spot pixel density from the arrays using image software analysis. Data were collected from 4 independent array experiments. White columns represent protein levels in lysates obtained from irrelevant IgG treated mice; black columns represent protein levels in lysates obtained from MK-0646 treated mice. Bars represent standard deviation. Note that only the increase in phosphorylated EGF-R of the MK-0646 treated mouse lung was statistically significant.

Figure 4.16 MK-0646 treatment increased total EGF-R protein and phosphorylated EGF-R (pEGF-R) levels both in vitro and in vivo. (a) Western blot analysis of the indicated proteins in 2 representative protein lysates obtained from the postcaval lung lobe of a control mouse and MK-0646 treated mouse in 3 pooled cell lines re-derive after necroscopy from lung tumors of control mice and mice treated with MK-0646, in A549 and H1437 cells treated either with an irrelevant human IgG or with MK-0646 for 3 weeks. (b) qRT-PCR of the EGF-R mRNA measured from the sources specified in (a). Columns represent the average of 4 independent experiments; bars represent standard deviation. (C) Immunoprecipitation of the EGF-R from the sources specified in (a) followed by immunoblot analysis using the specified antibodies. IgG lanes: immunoprecipitation performed using a pre-immune mouse IgG (negative control).

Polyubiquitinated EGF-R was reduced and EGF-R protein accumulated in the plasma membrane upon MK-0646 treatment

Since NSCLC cells treated with MK-0646 showed significant induction in EGF-R protein and pEGF-R levels in a transcription independent manner, we wanted to understand where the EGF-R protein accumulated in the cells and whether the EGF-R endocytosis/degradation pathway was affected by the treatment. We biotinylated the cell surface protein and performed a streptavidin pull down experiment to separate the membrane bound EGF-R from cytosolic EGF-R. We then determined the levels of phosphorylated EGF-R and total EGF-R protein by immunoprecipitating EGF-R and blotted for phosphorylated tyrosine. From the results of this experiment, we found that pEGF-R was increased in both the plasma membrane and cytosolic portion, while the total EGF-R protein was only accumulated on the cell plasma membrane (Figure 4.17a).

Consistent with this result, our experiment on ubiquitinated EGF-R showed that there was reduced EGF-R polyubiquitination after MK-0646 treatment (Figure 4.17b), suggesting an altered EGF-R degradation pathway. The fact that total EGF-R protein accumulated on the plasma membrane also suggested that the EGF-R recycling process might be affected by long term MK-0646 treatment.

Figure 4.17 MK-0646 treated NSCLC cells showed increased pEGF-R in the plasma membrane and cytosol, increased EGF-R total protein only in the plasma membrane, and reduced EGF-R polyubiquitination. (a) Biotinylation of cell-surface proteins; streptavidin pull-down followed by EGF-R immunoprecipitation (see Materials and Methods for details) and Western blot analysis performed using anti-phosphorylated tyrosine and anti-EGF-R antibodies. IgG lanes: EGF-R immunoprecipitation performed using a pre-immune mouse IgG (negative control) (b) EGF-R immunoprecipitation followed by Western blot analysis in A549 and H1437 cells treated with MK-0646 in vitro for three weeks, using antibodies against ubiquitin and EGF-R. IgG lanes: same as in (a).

MK-0646 treatment increased the expression of EGF-R ligand amphiregulin by activating Akt-1 signaling

We also found that the IGF-1R inhibitory antibody (MK-0646) treatment led to

Akt-1 activation, both in cells re-derived from MK-0646 treated mice lungs and in cells treated with MK-0646 for three weeks in vitro (Figure 4.18a). Since we were able to detect an increased level of pEGF-R in cells treated with MK-0646 in vitro for three weeks, this EGF-R activation effect appeared to be a cell-autonomous phenomenon.

Therefore, we performed qRT-PCR analysis to access the expression levels of all known

EGF-R ligands (including EGF, TGFα, HBEGF, betacellulin, epiregulin, amphiregulin

93 and epigen) in MK-0646 treated cells compared to controls treated with an irrelevant IgG.

We found that only the mRNA of amphiregulin was elevated in tumors obtained from mice lungs, as well as in cell lines re-derived from tumors and in cells treated with MK-

0646 in vitro (Figure 4.18b).

In order to verify whether the increased amphiregulin mRNA was dependent on

Akt-1 activation, A549 and H1437 cells were treated with either IGF-1 (50 ng/ml) or Akt pan-inhibitor (MK-2206) (1 µM). Cells treated with IGF-1 showed significantly increased mRNA levels of amphiregulin, whereas Akt pan-inhibitor treated cells displayed reduced mRNA levels of amphiregulin in comparison to controls (Figure

14.19a). Transfecting cells with plasmid encoding a constitutively active Akt-1, or a dominant-negative Akt-1 (Figure 14.19b) further confirmed that the expression level of amphiregulin is regulated by Akt signaling. Taken together, the results in Figure 4.18 and

Figure 4.19 suggested that MK-0646 treatment increased the expression level of EGF-R ligand amphiregulin by activating Akt-1 signaling. Since it has been shown that amphiregulin targets EGF-R for membrane recycling rather than lysosomal degradation

(Stern et al., 2008), these results also provided a possible explanation for the increased

EGF-R protein on the plasma membrane shown in Figure 4.17.

Figure 4.18 MK-0646 treatment led to Akt activation and increased ampheregulin expression. (a) Western blot analysis of phosphorylated Akt (pAkt) on Ser 473 and total Akt proteins in 3 pooled cell lines re-derived after necroscopy from lung tumors of control mice and mice treated with MK-0646, in A549 cells treated either with an irrelevant human IgG or with MK-0646 for 3 weeks. (b) qRT-PCR experiment of amphiregulin mRNAs in RNA samples extracted from the postcaval lung lobe of a control mouse and a MK-0646 treated mouse, in 3 pooled cell lines re-derived after necroscopy from lung tumors of control mice and mice treated with MK-0646, in A549 and H1437 cells treated either with an irrelevant human IgG or with MK-0646 for 3 weeks.

Figure 4.19 The expression level of amphiregulin is regulated by Akt signaling. (a) NSCLC cell lines A549 and H1437 were treated either with vehicle (C), with 50 ng/ml of IGF-1, or with 1 µM of MK-2206. Left, representative Western blot analysis using antibodies against phosphorylated Akt (pAkt) (Ser473), total Akt and GAPDH; right, qRT-PCR of the amphiregulin mRNA. Columns represent the average of 4 independent experiments; bars represent standard deviation. (b) A549 and H1437 cells were transfected either with a control plasmid (C), with a plasmid expressing constitutively active Akt-1 (Myr-Akt), or with a plasmid expressing dominant negative Akt-1 (DN- Akt). Left, representative Western blot analysis using the same antibodies described in (a); right, qRT-PCR of the amphiregulin mRNA. Columns represent the average of 4 independent experiments; bars represent standard deviation.

AIM3: STUDY IF ADMINISTRATION OF IGF-1R INHIBITORY ANTIBODY

SENSITIZES NSCLC CELLS TO EGF-R INHIBITION

MK-0646 treated NSCLC cells became sensitive to EGF-R inhibiton in vitro

Since NSCLC cells treated with MK-0646 overactivated EGF-R both in vivo and in vitro, we questioned whether such cells could have become more sensitive to the EGF-

R inhibitor erlotinib. To assess this hypothesis, three cell lines (A549, H226, and H1437) were treated with MK-0646 for three weeks in vitro (control cells were exposed to an irrelevant human IgG). Control and MK-0646 treated cells were then exposed to 5 µM erlotinib. We measured Br-dU incorporation (an example is provided in Figure 4.20a), and growth curves were assessed using MTT assays (Figure 4.20b). In all cases, cells pretreated with MK-0646 displayed the highest sensitivity to erlotinib. We further confirmed these results using a lentiviral vector expressing shEGF-R to stably knock down EGF-R in cells pre-treated with MK-0646 (Figure 4.21). The results also indicated that NSCLC cells became sensitive to EGF-R downregulation after MK-0646 treatment.

Figure 4.20 NSCLC cells pre-treated with MK-0646 became sensitive to EGF-R inhibitor (erlotinib). (a) Representative 5-bromo-2'-deoxyuridine (Br-dU) incorporation/7AAD staining assay followed by FACS analysis of A549 cells exposed either to a pre-immune human IgG (C) or to MK-0646 for 3 weeks (MK-0646). From left to right: control cells, control cells exposed to 5 µM erlotinib, MK-0646 treated cells exposed to 5 µM erlotinib for 48 hours. (b) NSCLC cell lines A549, H226 and H1437 were exposed either to a pre-immune IgG (C) or to MK-0646 for 3 weeks (MK-0646). Cells were then treated with vehicle or 5 µM erlotinib and MTT assays were performed at the indicated time points (24hr, 48hr and 72hr). Curves summarize 3 independent experiments. Bars represent standard deviation.

Figure 4.21 NSCLC cells pre-treated with MK-0646 became sensitive to EGF-R downregulation. (a) MTT assay performed with NSCLC cells A549 and H1437 pre- treted with and without MK-0646 stably transduced with either vector or lentivirus expression shEGF-R at indicated timepoints (24 hr, 48 hr and 72 hr). (b) Western blot analysis of A549 and H1437 cells transduced with vector or lentivirus expressing shEGF- R using antibodies against EGF-R and GAPDH.

Administration of MK-0646 followed by EGFR inhibitor (erlotinib) had improved efficacy in treating advanced NSCLC in vivo

If the tumor evasion that occurred during MK-0646 treatment was due to activated EGF-R signaling, then administration of IGF-1R inhibitory antibody followed by an EGFR inhibitor should have better efficacy in treating advanced NSCLC. To test this hypothesis, we performed in vivo experiments by treating experimental animals with

MK-0646 for three weeks (this is when tumors rebounded, see Figure 4.14), followed by erlotinib treatment (sequential therapy) in the same orthotopic mice models we have established. For comparison, we also included a combination therapeutic regimen in which the tumor-bearing mice were treated with MK-0646 together with erlotinib from

99 day one. The results confirmed that mice treated with MK-0646 for three weeks followed by erlotinib treatment significantly improved median survival compared to combination therapy using MK-0646 and erlotinib from the beginning of treatment or using single agent alone (Figure 4.8 right panel and Table 4.2).

CHAPTER 5

DISCUSSION

Tumor hypoxia is commonly seen in many types of cancer. It has been linked with cancer recurrence, radio/chemotherapy resistance, and maintaining the survival and growth of the cancer stem cells population (Keith and Simon, 2007). Current anticancer therapies that combine chemo/radiotherapy with antiangiogenic agents have generally failed and led to tumor recurrence because they are unable to target the hypoxic tumor microenvironment (Milas and Hittelman, 2009). Previous work from our laboratory indicated that under hypoxic conditions, Notch-1 is activated and stabilized by hypoxia inducible factor-1α (HIF-1α). On one hand, activated Notch-1 increases the expression level of IGF-1R and, on the other hand, it decreases PTEN expression. Both of these events lead to Akt-1 activation, which provides survival signals to NSCLC cells (and possibly other solid tumors) specifically under hypoxic condition (Eliasz et al., 2010).

In this dissertation, the detailed mechanism through which Notch-1 regulates the expression of IGF-1R has been studied. First, we tested the effect of Notch-1 downstream targets on IGF-1R regulation, including HES-1, HES-5, HEY-1, HEY-L and c-MYC.

However, none of these genes were able to regulate IGF-1R expression. Therefore, we hypothesized that Notch-1 directly regulates IGF-1R expression through CBF-1.

Chromatin immunoprecipitation experiment suggested that Notch-1 directly associated to the 5’ region of the IGF-1R first intron along with CBF-1 and the coactivators MAML-1

100 101 and p300. Additionally, transfection of the NSCLC cells with a plasmid encoding dominant-negative MAML-1 disrupted this association and dramatically reduced the amounts of acetylated histone H3 in that DNA region. It has been shown that transcriptional regulatory regions can be located in an intron or exon. For example, IGF-1 expression was shown to be regulated by an E box in the exon 1 promoter region

(McLellan et al., 2006). The transcriptional upregulation of IGF-1R by Notch-1 is consistent with previous findings that most tumor suppressors repress IGF-1R gene transcription, while oncogenes enhance the expression level of IGF-1R. For example, wild-type tumor suppressor p53 dramatically suppresses the IGF-1R promoter activity by interacting with the TATA box-binding protein, thus preventing the formation of the initiation complex at the IGF-1R promoter. However, mutant p53 stimulates the expression of IGF-1R and activates IGF-1R signaling (Werner et al., 1996). Similar to p53, expression of tumor suppressor breast cancer type 1 susceptibility protein (BRCA-1) also resulted in significant reduction in endogenous IGF-1R levels and IGF-1R promoter activity (Maor et al., 2000).

We also found that Notch-1 is activated in hypoxic tumor regions, which leads to maximal IGF-1R expression and Akt-1 activation in a mouse model of advanced

NSCLC. This in vivo finding further confirmed our previous in vitro results that Notch-1 is activated under hypoxia to activate IGF-1R and Akt-1 signaling, since it has been shown that mTORC-1 activity is inhibited in the hypoxic tumor microenvironment

(Vadysirisack and Ellisen, 2012), and that autophagy is promoted in these tumor areas

(Noman et al., 2011; Vadysirisack and Ellisen, 2012). The fact that IGF-1R is maximally

102 expressed in hypoxic tumor regions could be due to the loss of a negative feedback loop mediated by mTORC-1 activity on the insulin receptor substrate 1 (IRS-1).

We have gathered evidence suggesting that hypoxia activates Notch-1, which in turn leads to the increased expression of IGF-1R and activated Akt-1 signaling, and eventually survival of the hypoxic tumor cells. Additionally, Notch downstream signaling can also be triggered by IGF-1R signaling through increasing the level of HIF-1α

(Laughner et al., 2001; Zhong et al., 2000; Zundel et al., 2000). Therefore, a positive feedback loop is formed between Notch-1 and IGF-1R signaling pathways. To better target the hypoxic tumor microenvironment and to improve the efficacy of current therapeutic regimens for NSCLC, the major aim in this dissertation is to target the Notch-

1/IGF-1R/Akt-1 pro-survival signaling pathway using specific inhibitors, alone or in combination, to interfere with the survival of hypoxic NSCLC cells. We injected NSCLC cell lines A549 and H1437 through the tail vein to generate the orthotopic mouse model rather than a subcutaneous xenograft mouse model, because the orthotopic model can better recapitulate the heterogeneous tumors of the lungs, which have a mix of hypoxic and well-oxygenated tumor regions. Moreover, tumor masses in the lungs are likely to produce cardiopulmonary syndrome, thus further reproducing conditions similar to what is clinically observed. We limited our experiments to two cell lines, because a number of other NSCLC cell lines we have tested, including H1299, H1650 and H838, did not yield an acceptable tumor take in the mice lungs. Injection of H1299 into mice eventually generated brain tumors and, 5-weeks after injection, H838 and H1650 did not show any tumor formation measured by luminescence from luciferase activity. One possible

103 explanation for the unaccepted tumor intake is that the highly oxygenated environment of the lung interferes with the localization and growth of NSCLC tumor initiating cells, which requires hypoxic conditions (Conley et al., 2012; Gustafsson et al., 2005; Keith and Simon, 2007). We also decided to initiate the therapies when the tumor burden was extensive, because most of the patients diagnosed with NSCLC are at advanced stages

(IIIB and mostly IV).

Mice treated with GSI had a significantly prolonged median survival. Moreover, I have found that GSI treatment specifically depleted hypoxic tumor cells, which confirmed our previous finding that Notch-1 activation provided survival signals to

NSCLC cells specifically in hypoxia (Chen et al., 2007; Eliasz et al., 2010). Most of the

GSI treated experimental animals reached their endpoints due to a significant decrease in body weight. This could be explained by the disrupted intestine integrity due to inactivation of Notch signaling by GSI, since it has been shown that simultaneous inhibition of Notch-1 and Notch-2 in the intestine causes a significant loss in the number of intestinal crypt progenitor cells (Riccio et al., 2008; Wu et al., 2010). However, in humans, parenteral supportive care may reduce this undesired side effect significantly, allowing for a more prolonged survival time. Recently, a number of fully humanized monoclonal therapeutic antibodies that target individual Notch receptors have been developed. They can specifically prevent the cleavage of Notch receptors and thereby, the activation of downstream Notch signaling. However, it may not be beneficial to use these antibodies to target a hypoxic tumor microenvironment. Large molecules, such as

104 antibodies may not be able to reach the hypoxic tumor areas due to their distance from the functional vasculature.

The quantitative PCR performed using human and mouse specific GAPDH primers showed that the brains and livers of the mice had a reduced ratio of human versus mouse cells in GSI treated mice. Notably, the GSI treated mice were sacrificed, and the brain and liver tissues were collected at a later time point since they survived much longer (15.5 and 30 days for A549 and H1437 injected mice, respectively) compared with control mice (9 and 8 days for A549 and H1437 injected mice, respectively). These results suggested that GSI treated mice could have reduced tumor metastasis to the brain and liver. Moreover, previous literature has also suggested that Notch signaling is critical in mediating hypoxia-induced tumor cell migration and invasion (Sahlgren et al., 2008).

Since cisplatin, a current chemotherapeutic drug that kills highly proliferating cells, and

GSI treatment specifically targets hypoxic tumor regions that are quiescent and therefore nonresponsive to chemotherapeutic agents, we hypothesized that combination therapy using these two agents may improve the efficacy of NSCLC treatment. Indeed, we found that combining GSI with cisplatin significantly improved the median survival of the mice compared to mice receiving a single agent alone, which also confirmed the previous finding that GSI can sensitize cancer cells to cisplatin exposure (Song et al., 2008).

During the administration of pan Akt inhibitor MK-2206, we found that this drug caused severe glucose intolerance and very rapid weight loss. The treatment schedule and dosage were based on the result of a preliminary experiment, which suggested that the maximum tolerated dosage in SCID mice was 3 consecutive administrations of 80 mg/kg

105 per week. This finding is in accordance with the phenotype of Akt-2 and Akt-3 double knockout mice, which are highly insulin and glucose intolerant and display a 25% reduction in body mass, and with the lethal phenotype of the Akt triple knockout

(Dummler et al., 2006). However, She et al. (She et al., 2010) have used a similar compound alone (100 mg/kg for 5 consecutive days per week) and in combination with a

MEK inhibitor in nude mice for more than 20 days, obtaining slower tumor growth of subcutaneous xenograft models. Hirai et al., who used MK-2206 at a dose of 120 mg/kg with our administration schedule, both alone and in combination with lapatinib or erlotinib in nude mice for more than 15 days, also attained slower tumor growth of subcutaneous xenograft models (Hirai et al., 2010). The discrepancy between our results and those obtained in the aforementioned studies could be explained by a higher susceptibility of NOD-SCID mice to Akt inhibition compared with nude mice or by differences in glucose consumption between a subcutaneous tumor model compared with multiple tumor masses in the lungs.

IGF-1R inhibitory antibody (MK-0646) treatment significantly extended the survival of mice compared to the control treatment. Combinational therapy using GSI and

MK-0646 significantly increased median survival compared to a single agent alone.

Although the IGF-1R is maximally expressed in hypoxic tumor areas, treatment with this antibody did not appear to target the hypoxic tumor microenvironment specifically, probably due to the fact that an antibody has difficulty penetrating poorly vascularized tumor regions.

106

Three weeks after MK-0646 treatment, tumor burden started to rebound and eventually led to the death of the experimental animals. We identified that EGF-R is the most important player in mediating this treatment evasion process. Therefore, we hypothesized that sequential therapy with MK-0646 followed by erlotinib treatment may improve the efficacy of NSCLC treatment. Indeed, in our orthotopic model, mice treated with MK-0646 for three weeks followed by erlotinib treatment had a dramatically improved median survival compared to a single agent or combination therapy (MK-0646 and erlotinib simultaneously started from day one of treatment). IGF-1R inhibitory antibodies have been implicated in anticancer therapies targeting different types of cancer.

Phase I and II clinical trials with these antibodies were very promising, and suggested little toxicity, mainly of hyperglycemia and increased insulin levels. However, a recent phase III trial of figitumumab, another humanized monoclonal antibody against IGF-1R in NSCLC, did not improve overall survival in combination with carboplatin plus paclitaxel, suggesting a lack of efficacy. This might be due to increased resistance since we have found that tumor cells become resistant to IGF-1R inhibitory antibody treatments (MK-0646) owing to increased EGF-R activity. Therefore, the above evidence strongly indicates that the IGF-1R inhibitory antibody treatments should be used in combinational or sequential therapies to overcome the problem of tumor evasion occurring after IGF-1R deprivation. Here, for the first time, we provide evidence that

MK-0646 could be used in a window of time during therapeutic intervention to sensitize NSCLC cells to erlotinib. This drug is used as a second line therapeutic in

107

NSCLC patients who have failed standard carboplatin plus paclitaxel and it extends median survival for about 3 months.

To further study the detailed mechanism by which EGF-R is activated upon IGF-

1R deprivation, we cultured the NSCLC cells in the culture media containing MK-0646 for three weeks in vitro and found EGF-R activation, similar to the results obtained from mice lung tumors in vivo. This observation highly suggested that the effect mediated by

MK-0646 on EGF-R activation is a cell autonomous effect. We also found that MK-0646 seems to increase the EGF-R expression and activation mainly through protein stabilization. Biotinylation experiments showed that MK-0646 treatment increased EGF-

R expression at the plasma membranes rather than in cytosolic fractions, suggesting rapid recycling of the receptor. Increased recycling and activation of EGF-R could be due to autocrine of amphiregulin. This EGF-R ligand has been shown to cause both fast and slow EGF-R recycling and does not target EGF-R to lysosomal degradation (Stern et al.,

2008). Moreover, amphiregulin stimulation of wild-type EGF-R has been shown to sensitize NSCLC cells to EGF-R inhibitors, which are mostly effective in patients with

EGF-R activating mutations (Yonesaka et al., 2008). Furthermore, immunohistochemistry studies suggested that NSCLC patients with high levels of amphiregulin expression and wild-type EGF-R respond better to EGF-R inhibitors

(Chang et al., 2011). Moreover, we have found that Akt-1 activation positively regulates the amphiregulin mRNA expression levels. Previous studies have placed Akt-1 downstream of amphiregulin/ErbB signaling. It is possible that in NSCLC cells, a

108 positive feedback loop exists between the ligand and one or more of its downstream effectors (Dong et al., 2011; Yotsumoto et al., 2010).

The in vitro experiments in which NSCLC cells were cultured in the presence of

MK-0646 containing culture media showed increased EGF-R protein levels and activation after three weeks of treatment, whereas EGF-R levels remained unchanged after short-term treatment. (We have tested 24 hr, 48 hr and 72 hr, data not shown). One possible explanation for this phenomenon is that MK-0646 treatment may actively select pre-existing cells expressing a higher level of EGF-R. To better understand whether MK-

0646 treatment increases EGF-R protein levels through selecting pre-existing ‘EGF-R high’ (in terms of protein level) cell populations or, directly increases the expression of

EGF-R, we performed FACS analysis of A549 cells for EGF-R surface expression levels, and sorted for ‘EGF-R high’ and ‘EGF-R low’ cells (Figure 5.1a). To understand if

‘EGF-R high’ or ‘EGF-R’ low cells are able to maintain their phenotype after sorting, the cells were cultured for a week in the absence of MK-0646, and reanalyzed for EGF-R surface level by FACS. We found that the distribution of surface EGF-R in both ‘EGF-R high’ and ‘EGF-R low’ cells went back to the same as their parental cells, which were

A549 cells before sorting (Figure 5.1b). This would suggest that ‘EGF-R high’ and

‘EGF-R low’ A549 cells would not keep their phenotype for more than one week.

However, these results could not answer the question of whether MK-0646 treatment actively selects pre-existing ‘EGF-R high’ cells or directly increases the protein level of

EGF-R, because MK-0646 was constantly provided in the culture medium of NSCLC

109 cells when we performed the experiment with MK-0646 and found increased EGF-R protein level and activity.

Figure 5.1 EGF-R surface expression remained the same one-week after sorting. (a) A549 cells were stained with antibody against EGF-R (NeoMarkers) and Alexa Flour

110

488. Cells expressing low surface level of EGF-R (GFP-DIM) and high surface level of EGF-R (GFP-BRIGHT) were FACS sorted. (b) One week after the sorting experiment indicated in (a), the sorted cells and A549 parental cells were surface stained with EGF-R antibody and Alexa Flour 488, or Alexa Flour 488 only (negative control). FACS analysis was performed on the stained cells.

Taken together, we have found that the hypoxic tumor microenvironment could be targeted by administration of GSI. GSI treatment not only reduced tumor growth in hypoxic areas, but also significantly reduced tumor metastasis. Combining GSI and cisplatin had better efficacy in treating mice bearing advanced NSCLC. Therefore, GSI could be used in combination with other chemotherapeutic agents to target both hypoxic and highly proliferating cells. Moreover, we have also identified the mechanism that leads to tumor evasion from IGF-1R inhibition, which is by activating the EGF-R signaling pathway. MK-0646 treatment caused upregulation of amphiregulin, which in turn activates and stabilizes EGF-R. Furthermore, sequential therapy using MK-0646 followed by erlotinib treatment dramatically improved the survival of experimental animals. This study has considerable clinical significance because IGF-1R inhibitory antibodies are under clinical trials and face the problem of lack of efficacy due to resistance. Our study suggests that targeting IGF-1R could be used in a window of time during therapeutic intervention to sensitize NSCLC cells to EGF-R inhibition.

BIBLIOGRAPHY

Abbott, A.M., Bueno, R., Pedrini, M.T., Murray, J.M., and Smith, R.J. (1992). Insulin- like growth factor I receptor gene structure. J Biol Chem 267, 10759-10763.

Abraham, J., Prajapati, S.I., Nishijo, K., Schaffer, B.S., Taniguchi, E., Kilcoyne, A., McCleish, A.T., Nelon, L.D., Giles, F.G., Efstratiadis, A., et al. Evasion mechanisms to Igf1r inhibition in rhabdomyosarcoma. Mol Cancer Ther 10, 697- 707.

Artavanis-Tsakonas, S., Rand, M.D., and Lake, R.J. (1999). Notch signaling: cell fate control and signal integration in development. Science 284, 770-776.

Behrooz, A., and Ismail-Beigi, F. (1999). Stimulation of Glucose Transport by Hypoxia: Signals and Mechanisms. News Physiol Sci 14, 105-110.

Brizel, D.M., Dodge, R.K., Clough, R.W., and Dewhirst, M.W. (1999). Oxygenation of head and neck cancer: changes during radiotherapy and impact on treatment outcome. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology 53, 113-117.

Brizel, D.M., Scully, S.P., Harrelson, J.M., Layfield, L.J., Bean, J.M., Prosnitz, L.R., and Dewhirst, M.W. (1996). Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. Cancer Res 56, 941-943.

Brou, C., Logeat, F., Gupta, N., Bessia, C., LeBail, O., Doedens, J.R., Cumano, A., Roux, P., Black, R.A., and Israel, A. (2000). A novel proteolytic cleavage involved in Notch signaling: the role of the disintegrin-metalloprotease TACE. Mol Cell 5, 207-216.

Broussas, M., Dupont, J., Gonzalez, A., Blaecke, A., Fournier, M., Corvaia, N., and Goetsch, L. (2009). Molecular mechanisms involved in activity of h7C10, a humanized monoclonal antibody, to IGF-1 receptor. Int J Cancer 124, 2281-2293.

111 112

Brown, J.M. (2007). Tumor hypoxia in cancer therapy. Methods in enzymology 435, 297-321.

Bruick, R.K. (2003). Oxygen sensing in the hypoxic response pathway: regulation of the hypoxia-inducible transcription factor. Genes Dev 17, 2614-2623.

Bruick, R.K., and McKnight, S.L. (2001). A conserved family of prolyl-4-hydroxylases that modify HIF. Science 294, 1337-1340.

Burgess, A.W., Cho, H.S., Eigenbrot, C., Ferguson, K.M., Garrett, T.P., Leahy, D.J., Lemmon, M.A., Sliwkowski, M.X., Ward, C.W., and Yokoyama, S. (2003). An open-and-shut case? Recent insights into the activation of EGF/ErbB receptors. Mol Cell 12, 541-552.

Burrow, S., Andrulis, I.L., Pollak, M., and Bell, R.S. (1998). Expression of insulin-like growth factor receptor, IGF-1, and IGF-2 in primary and metastatic osteosarcoma. J Surg Oncol 69, 21-27.

Carroll, V.A., and Ashcroft, M. (2006). Role of hypoxia-inducible factor (HIF)-1alpha versus HIF-2alpha in the regulation of HIF target genes in response to hypoxia, insulin-like growth factor-I, or loss of von Hippel-Lindau function: implications for targeting the HIF pathway. Cancer Res 66, 6264-6270.

Chandarlapaty, S., Sawai, A., Scaltriti, M., Rodrik-Outmezguine, V., Grbovic-Huezo, O., Serra, V., Majumder, P.K., Baselga, J., and Rosen, N. (2011). AKT inhibition relieves feedback suppression of receptor tyrosine kinase expression and activity. Cancer cell 19, 58-71.

Chang, M.H., Ahn, H.K., Lee, J., Jung, C.K., Choi, Y.L., Park, Y.H., Ahn, J.S., Park, K., and Ahn, M.J. (2011). Clinical impact of amphiregulin expression in patients with epidermal growth factor receptor (EGFR) wild-type nonsmall cell lung cancer treated with EGFR-tyrosine kinase inhibitors. Cancer 117, 143-151.

Chen, Y., De Marco, M.A., Graziani, I., Gazdar, A.F., Strack, P.R., Miele, L., and Bocchetta, M. (2007). Oxygen concentration determines the biological effects of NOTCH-1 signaling in adenocarcinoma of the lung. Cancer Res 67, 7954-7959.

113

Collins, B.J., Kleeberger, W., and Ball, D.W. (2004). Notch in lung development and lung cancer. Seminars in cancer biology 14, 357-364.

Comerford, K.M., Wallace, T.J., Karhausen, J., Louis, N.A., Montalto, M.C., and Colgan, S.P. (2002). Hypoxia-inducible factor-1-dependent regulation of the multidrug resistance (MDR1) gene. Cancer Res 62, 3387-3394.

Conley, S.J., Gheordunescu, E., Kakarala, P., Newman, B., Korkaya, H., Heath, A.N., Clouthier, S.G., and Wicha, M.S. (2012). Antiangiogenic agents increase breast cancer stem cells via the generation of tumor hypoxia. Proc Natl Acad Sci U S A 109, 2784-2789.

Coquelle, A., Toledo, F., Stern, S., Bieth, A., and Debatisse, M. (1998). A new role for hypoxia in tumor progression: induction of fragile site triggering genomic rearrangements and formation of complex DMs and HSRs. Mol Cell 2, 259-265.

Covello, K.L., Kehler, J., Yu, H., Gordan, J.D., Arsham, A.M., Hu, C.J., Labosky, P.A., Simon, M.C., and Keith, B. (2006). HIF-2alpha regulates Oct-4: effects of hypoxia on stem cell function, embryonic development, and tumor growth. Genes Dev 20, 557-570.

D'Ambrosio, C., Valentinis, B., Prisco, M., Reiss, K., Rubini, M., and Baserga, R. (1997). Protective effect of the insulin-like growth factor I receptor on apoptosis induced by okadaic acid. Cancer Res 57, 3264-3271.

Decker, S.J. (1990). Epidermal growth factor and transforming growth factor-alpha induce differential processing of the epidermal growth factor receptor. Biochem Biophys Res Commun 166, 615-621.

Delcourt, N., Bockaert, J., and Marin, P. (2007). GPCR-jacking: from a new route in RTK signalling to a new concept in GPCR activation. Trends Pharmacol Sci 28, 602-607.

DeYoung, M.P., Horak, P., Sofer, A., Sgroi, D., and Ellisen, L.W. (2008). Hypoxia regulates TSC1/2-mTOR signaling and tumor suppression through REDD1- mediated 14-3-3 shuttling. Genes Dev 22, 239-251.

114

Dong, A., Gupta, A., Pai, R.K., Tun, M., and Lowe, A.W. (2011). The human adenocarcinoma-associated gene, AGR2, induces expression of amphiregulin through Hippo pathway co-activator YAP1 activation. J Biol Chem 286, 18301- 18310.

Duan, L., Miura, Y., Dimri, M., Majumder, B., Dodge, I.L., Reddi, A.L., Ghosh, A., Fernandes, N., Zhou, P., Mullane-Robinson, K., et al. (2003). Cbl-mediated ubiquitinylation is required for lysosomal sorting of epidermal growth factor receptor but is dispensable for endocytosis. J Biol Chem 278, 28950-28960.

Dummler, B., Tschopp, O., Hynx, D., Yang, Z.Z., Dirnhofer, S., and Hemmings, B.A. (2006). Life with a single isoform of Akt: mice lacking Akt2 and Akt3 are viable but display impaired glucose homeostasis and growth deficiencies. Mol Cell Biol 26, 8042-8051.

Ebert, B.L., Firth, J.D., and Ratcliffe, P.J. (1995). Hypoxia and mitochondrial inhibitors regulate expression of glucose transporter-1 via distinct Cis-acting sequences. J Biol Chem 270, 29083-29089.

Eliasz, S., Liang, S., Chen, Y., De Marco, M.A., Machek, O., Skucha, S., Miele, L., and Bocchetta, M. (2010). Notch-1 stimulates survival of lung adenocarcinoma cells during hypoxia by activating the IGF-1R pathway. Oncogene 29, 2488-2498.

Ellisen, L.W., Bird, J., West, D.C., Soreng, A.L., Reynolds, T.C., Smith, S.D., and Sklar, J. (1991). TAN-1, the human homolog of the Drosophila notch gene, is broken by chromosomal translocations in T lymphoblastic neoplasms. Cell 66, 649-661.

Favoni, R.E., de Cupis, A., Ravera, F., Cantoni, C., Pirani, P., Ardizzoni, A., Noonan, D., and Biassoni, R. (1994). Expression and function of the insulin-like growth factor I system in human non-small-cell lung cancer and normal lung cell lines. Int J Cancer 56, 858-866.

Frankel, S.K., Moats-Staats, B.M., Cool, C.D., Wynes, M.W., Stiles, A.D., and Riches, D.W. (2005). Human insulin-like growth factor-IA expression in transgenic mice promotes adenomatous hyperplasia but not pulmonary fibrosis. Am J Physiol Lung Cell Mol Physiol 288, L805-812.

115

Fryer, C.J., Lamar, E., Turbachova, I., Kintner, C., and Jones, K.A. (2002). Mastermind mediates chromatin-specific transcription and turnover of the Notch enhancer complex. Genes Dev 16, 1397-1411.

Galluzzo, P., and Bocchetta, M. (2011). Notch signaling in lung cancer. Expert review of anticancer therapy 11, 533-540.

Goldstraw, P., Crowley, J., Chansky, K., Giroux, D.J., Groome, P.A., Rami-Porta, R., Postmus, P.E., Rusch, V., Sobin, L., International Association for the Study of Lung Cancer International Staging, C., et al. (2007). The IASLC Lung Cancer Staging Project: proposals for the revision of the TNM stage groupings in the forthcoming (seventh) edition of the TNM Classification of malignant tumours. J Thorac Oncol 2, 706-714.

Gordan, J.D., Bertout, J.A., Hu, C.J., Diehl, J.A., and Simon, M.C. (2007). HIF-2alpha promotes hypoxic cell proliferation by enhancing c-myc transcriptional activity. Cancer cell 11, 335-347.

Gualberto, A., and Pollak, M. (2009). Emerging role of insulin-like growth factor receptor inhibitors in oncology: early clinical trial results and future directions. Oncogene 28, 3009-3021.

Gusscott, S., Kuchenbauer, F., Humphries, R.K., and Weng, A.P. (2012). Notch-mediated repression of miR-223 contributes to IGF1R regulation in T-ALL. Leukemia research 36, 905-911.

Gustafsson, M.V., Zheng, X., Pereira, T., Gradin, K., Jin, S., Lundkvist, J., Ruas, J.L., Poellinger, L., Lendahl, U., and Bondesson, M. (2005). Hypoxia requires notch signaling to maintain the undifferentiated cell state. Dev Cell 9, 617-628.

Haase, V.H. (2010). Hypoxic regulation of erythropoiesis and iron metabolism. American journal of physiology Renal physiology 299, F1-13.

Hakam, A., Yeatman, T.J., Lu, L., Mora, L., Marcet, G., Nicosia, S.V., Karl, R.C., and Coppola, D. (1999). Expression of insulin-like growth factor-1 receptor in human colorectal cancer. Hum Pathol 30, 1128-1133.

116

Haluska, P., Shaw, H.M., Batzel, G.N., Yin, D., Molina, J.R., Molife, L.R., Yap, T.A., Roberts, M.L., Sharma, A., Gualberto, A., et al. (2007). Phase I dose escalation study of the anti insulin-like growth factor-I receptor monoclonal antibody CP- 751,871 in patients with refractory solid tumors. Clinical cancer research : an official journal of the American Association for Cancer Research 13, 5834-5840.

Happerfield, L.C., Miles, D.W., Barnes, D.M., Thomsen, L.L., Smith, P., and Hanby, A. (1997). The localization of the insulin-like growth factor receptor 1 (IGFR-1) in benign and malignant breast tissue. J Pathol 183, 412-417.

Harris, A.L. (2002). Hypoxia--a key regulatory factor in tumour growth. Nat Rev Cancer 2, 38-47.

Hartmann, D., de Strooper, B., Serneels, L., Craessaerts, K., Herreman, A., Annaert, W., Umans, L., Lubke, T., Lena Illert, A., von Figura, K., et al. (2002). The disintegrin/metalloprotease ADAM 10 is essential for Notch signalling but not for alpha-secretase activity in fibroblasts. Human molecular genetics 11, 2615-2624.

Hawryluk, M.J., Keyel, P.A., Mishra, S.K., Watkins, S.C., Heuser, J.E., and Traub, L.M. (2006). Epsin 1 is a polyubiquitin-selective clathrin-associated sorting protein. Traffic 7, 262-281.

Hellawell, G.O., Turner, G.D., Davies, D.R., Poulsom, R., Brewster, S.F., and Macaulay, V.M. (2002). Expression of the type 1 insulin-like growth factor receptor is upregulated in primary prostate cancer and commonly persists in metastatic disease. Cancer Res 62, 2942-2950.

Hewish, M., Chau, I., and Cunningham, D. (2009). Insulin-like growth factor 1 receptor targeted therapeutics: novel compounds and novel treatment strategies for cancer medicine. Recent patents on anti-cancer drug discovery 4, 54-72.

Hirai, H., Sootome, H., Nakatsuru, Y., Miyama, K., Taguchi, S., Tsujioka, K., Ueno, Y., Hatch, H., Majumder, P.K., Pan, B.S., et al. (2010). MK-2206, an allosteric Akt inhibitor, enhances antitumor efficacy by standard chemotherapeutic agents or molecular targeted drugs in vitro and in vivo. Mol Cancer Ther 9, 1956-1967.

117

Hockel, M., Schlenger, K., Aral, B., Mitze, M., Schaffer, U., and Vaupel, P. (1996). Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 56, 4509-4515.

Howe, A.K., Aplin, A.E., and Juliano, R.L. (2002). Anchorage-dependent ERK signaling--mechanisms and consequences. Current opinion in genetics & development 12, 30-35.

Huang, F., Goh, L.K., and Sorkin, A. (2007). EGF receptor ubiquitination is not necessary for its internalization. Proc Natl Acad Sci U S A 104, 16904-16909.

Iso, T., Kedes, L., and Hamamori, Y. (2003). HES and HERP families: multiple effectors of the Notch signaling pathway. J Cell Physiol 194, 237-255.

Ivan, M., Kondo, K., Yang, H., Kim, W., Valiando, J., Ohh, M., Salic, A., Asara, J.M., Lane, W.S., and Kaelin, W.G., Jr. (2001). HIFalpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science 292, 464-468.

Jaques, G., Rotsch, M., Wegmann, C., Worsch, U., Maasberg, M., and Havemann, K. (1988). Production of immunoreactive insulin-like growth factor I and response to exogenous IGF-I in small cell lung cancer cell lines. Exp Cell Res 176, 336-343.

Jeffries, S., Robbins, D.J., and Capobianco, A.J. (2002). Characterization of a high- molecular-weight Notch complex in the nucleus of Notch(ic)-transformed RKE cells and in a human T-cell leukemia cell line. Mol Cell Biol 22, 3927-3941.

Jiang, X., Zhou, J.H., Deng, Z.H., Qu, X.H., Jiang, H.Y., and Liu, Y. (2007). [Expression and significance of Notch1, Jagged1 and VEGF in human non-small cell lung cancer]. Zhong Nan Da Xue Xue Bao Yi Xue Ban 32, 1031-1036.

Kaether, C., Haass, C., and Steiner, H. (2006). Assembly, trafficking and function of gamma-secretase. Neuro-degenerative diseases 3, 275-283.

Kaidi, A., Williams, A.C., and Paraskeva, C. (2007). Interaction between beta-catenin and HIF-1 promotes cellular adaptation to hypoxia. Nature cell biology 9, 210-217.

118

Kaiser, U., Schardt, C., Brandscheidt, D., Wollmer, E., and Havemann, K. (1993). Expression of insulin-like growth factor receptors I and II in normal human lung and in lung cancer. J Cancer Res Clin Oncol 119, 665-668.

Karp, D.D., Paz-Ares, L.G., Novello, S., Haluska, P., Garland, L., Cardenal, F., Blakely, L.J., Eisenberg, P.D., Langer, C.J., Blumenschein, G., Jr., et al. (2009). Phase II study of the anti-insulin-like growth factor type 1 receptor antibody CP-751,871 in combination with paclitaxel and carboplatin in previously untreated, locally advanced, or metastatic non-small-cell lung cancer. J Clin Oncol 27, 2516-2522.

Katzmann, D.J., Odorizzi, G., and Emr, S.D. (2002). Receptor downregulation and multivesicular-body sorting. Nat Rev Mol Cell Biol 3, 893-905.

Keith, B., and Simon, M.C. (2007). Hypoxia-inducible factors, stem cells, and cancer. Cell 129, 465-472.

Khambata-Ford, S., Garrett, C.R., Meropol, N.J., Basik, M., Harbison, C.T., Wu, S., Wong, T.W., Huang, X., Takimoto, C.H., Godwin, A.K., et al. (2007). Expression of epiregulin and amphiregulin and K-ras mutation status predict disease control in metastatic colorectal cancer patients treated with cetuximab. J Clin Oncol 25, 3230- 3237.

Klinakis, A., Szabolcs, M., Politi, K., Kiaris, H., Artavanis-Tsakonas, S., and Efstratiadis, A. (2006). Myc is a Notch1 transcriptional target and a requisite for Notch1- induced mammary tumorigenesis in mice. Proc Natl Acad Sci U S A 103, 9262- 9267.

Knowlden, J.M., Jones, H.E., Barrow, D., Gee, J.M., Nicholson, R.I., and Hutcheson, I.R. (2008). Insulin receptor substrate-1 involvement in epidermal growth factor receptor and insulin-like growth factor receptor signalling: implication for Gefitinib ('Iressa') response and resistance. Breast Cancer Res Treat 111, 79-91.

Konishi, J., Kawaguchi, K.S., Vo, H., Haruki, N., Gonzalez, A., Carbone, D.P., and Dang, T.P. (2007). Gamma-secretase inhibitor prevents Notch3 activation and reduces proliferation in human lung cancers. Cancer Res 67, 8051-8057.

119

Koshiji, M., To, K.K., Hammer, S., Kumamoto, K., Harris, A.L., Modrich, P., and Huang, L.E. (2005). HIF-1alpha induces genetic instability by transcriptionally downregulating MutSalpha expression. Mol Cell 17, 793-803.

Kucejova, B., Pena-Llopis, S., Yamasaki, T., Sivanand, S., Tran, T.A., Alexander, S., Wolff, N.C., Lotan, Y., Xie, X.J., Kabbani, W., et al. (2011). Interplay between pVHL and mTORC1 pathways in clear-cell renal cell carcinoma. Molecular cancer research : MCR 9, 1255-1265.

Kulik, G., Klippel, A., and Weber, M.J. (1997). Antiapoptotic signalling by the insulin- like growth factor I receptor, phosphatidylinositol 3-kinase, and Akt. Mol Cell Biol 17, 1595-1606.

Lai, E.C. (2004). Notch signaling: control of cell communication and cell fate. Development 131, 965-973.

Lando, D., Peet, D.J., Gorman, J.J., Whelan, D.A., Whitelaw, M.L., and Bruick, R.K. (2002). FIH-1 is an asparaginyl hydroxylase enzyme that regulates the transcriptional activity of hypoxia-inducible factor. Genes Dev 16, 1466-1471.

Lanner, F., Lee, K.L., Ortega, G.C., Sohl, M., Li, X., Jin, S., Hansson, E.M., Claesson- Welsh, L., Poellinger, L., Lendahl, U., et al. (2013). Hypoxia-induced Arterial Differentiation Requires Adrenomedullin and Notch Signaling. Stem cells and development.

Laughner, E., Taghavi, P., Chiles, K., Mahon, P.C., and Semenza, G.L. (2001). HER2 (neu) signaling increases the rate of hypoxia-inducible factor 1alpha (HIF-1alpha) synthesis: novel mechanism for HIF-1-mediated vascular endothelial growth factor expression. Mol Cell Biol 21, 3995-4004.

Levkowitz, G., Waterman, H., Zamir, E., Kam, Z., Oved, S., Langdon, W.Y., Beguinot, L., Geiger, B., and Yarden, Y. (1998). c-Cbl/Sli-1 regulates endocytic sorting and ubiquitination of the epidermal growth factor receptor. Genes Dev 12, 3663-3674.

Li, X., Zhang, X., Leathers, R., Makino, A., Huang, C., Parsa, P., Macias, J., Yuan, J.X., Jamieson, S.W., and Thistlethwaite, P.A. (2009). Notch3 signaling promotes the development of pulmonary arterial hypertension. Nat Med 15, 1289-1297.

120

Liang, S., Galluzzo, P., Sobol, A., Skucha, S., Rambo, B., and Bocchetta, M. (2012). Multimodality Approaches to Treat Hypoxic Non-Small Cell Lung Cancer (NSCLC) Microenvironment. Genes Cancer 3, 141-151.

Lin, Q., and Yun, Z. (2010). Impact of the hypoxic tumor microenvironment on the regulation of cancer stem cell characteristics. Cancer biology & therapy 9, 949-956.

Liu, J.P., Baker, J., Perkins, A.S., Robertson, E.J., and Efstratiadis, A. (1993). Mice carrying null mutations of the genes encoding insulin-like growth factor I (Igf-1) and type 1 IGF receptor (Igf1r). Cell 75, 59-72.

Logeat, F., Bessia, C., Brou, C., LeBail, O., Jarriault, S., Seidah, N.G., and Israel, A. (1998). The Notch1 receptor is cleaved constitutively by a furin-like convertase. Proc Natl Acad Sci U S A 95, 8108-8112.

Longva, K.E., Blystad, F.D., Stang, E., Larsen, A.M., Johannessen, L.E., and Madshus, I.H. (2002). Ubiquitination and proteasomal activity is required for transport of the EGF receptor to inner membranes of multivesicular bodies. J Cell Biol 156, 843- 854.

Lu, Y., Zi, X., Zhao, Y., Mascarenhas, D., and Pollak, M. (2001). Insulin-like growth factor-I receptor signaling and resistance to trastuzumab (Herceptin). J Natl Cancer Inst 93, 1852-1857.

Mahon, P.C., Hirota, K., and Semenza, G.L. (2001). FIH-1: a novel protein that interacts with HIF-1alpha and VHL to mediate repression of HIF-1 transcriptional activity. Genes Dev 15, 2675-2686.

Maor, S.B., Abramovitch, S., Erdos, M.R., Brody, L.C., and Werner, H. (2000). BRCA1 suppresses insulin-like growth factor-I receptor promoter activity: potential interaction between BRCA1 and Sp1. Molecular genetics and metabolism 69, 130- 136.

McLellan, A.S., Kealey, T., and Langlands, K. (2006). An E box in the exon 1 promoter regulates insulin-like growth factor-I expression in differentiating muscle cells. American journal of physiology Cell physiology 291, C300-307.

121

Medyouf, H., Gusscott, S., Wang, H., Tseng, J.C., Wai, C., Nemirovsky, O., Trumpp, A., Pflumio, F., Carboni, J., Gottardis, M., et al. (2011). High-level IGF1R expression is required for leukemia-initiating cell activity in T-ALL and is supported by Notch signaling. The Journal of experimental medicine 208, 1809-1822.

Miettinen, P.J., Ebner, R., Lopez, A.R., and Derynck, R. (1994). TGF-beta induced transdifferentiation of mammary epithelial cells to mesenchymal cells: involvement of type I receptors. J Cell Biol 127, 2021-2036.

Mihaylova, V.T., Bindra, R.S., Yuan, J., Campisi, D., Narayanan, L., Jensen, R., Giordano, F., Johnson, R.S., Rockwell, S., and Glazer, P.M. (2003). Decreased expression of the DNA mismatch repair gene Mlh1 under hypoxic stress in mammalian cells. Mol Cell Biol 23, 3265-3273.

Milas, L., and Hittelman, W.N. (2009). Cancer stem cells and tumor response to therapy: current problems and future prospects. Seminars in radiation oncology 19, 96-105.

Minuto, F., Del Monte, P., Barreca, A., Fortini, P., Cariola, G., Catrambone, G., and Giordano, G. (1986). Evidence for an increased somatomedin-C/insulin-like growth factor I content in primary human lung tumors. Cancer Res 46, 985-988.

Miyamoto, Y., Maitra, A., Ghosh, B., Zechner, U., Argani, P., Iacobuzio-Donahue, C.A., Sriuranpong, V., Iso, T., Meszoely, I.M., Wolfe, M.S., et al. (2003). Notch mediates TGF alpha-induced changes in epithelial differentiation during pancreatic tumorigenesis. Cancer cell 3, 565-576.

Moorehead, R.A., Sanchez, O.H., Baldwin, R.M., and Khokha, R. (2003). Transgenic overexpression of IGF-II induces spontaneous lung tumors: a model for human lung adenocarcinoma. Oncogene 22, 853-857.

Morgillo, F., Woo, J.K., Kim, E.S., Hong, W.K., and Lee, H.Y. (2006). Heterodimerization of insulin-like growth factor receptor/epidermal growth factor receptor and induction of survivin expression counteract the antitumor action of erlotinib. Cancer Res 66, 10100-10111.

Morrison, K.B., Tognon, C.E., Garnett, M.J., Deal, C., and Sorensen, P.H. (2002). ETV6- NTRK3 transformation requires insulin-like growth factor 1 receptor signaling and

122

is associated with constitutive IRS-1 tyrosine phosphorylation. Oncogene 21, 5684- 5695.

Nakanishi, Y., Mulshine, J.L., Kasprzyk, P.G., Natale, R.B., Maneckjee, R., Avis, I., Treston, A.M., Gazdar, A.F., Minna, J.D., and Cuttitta, F. (1988). Insulin-like growth factor-I can mediate autocrine proliferation of human small cell lung cancer cell lines in vitro. J Clin Invest 82, 354-359.

Nam, Y., Sliz, P., Song, L., Aster, J.C., and Blacklow, S.C. (2006). Structural basis for cooperativity in recruitment of MAML coactivators to Notch transcription complexes. Cell 124, 973-983.

Nishi, H., Nakada, T., Hokamura, M., Osakabe, Y., Itokazu, O., Huang, L.E., and Isaka, K. (2004). Hypoxia-inducible factor-1 transactivates transforming growth factor- beta3 in trophoblast. Endocrinology 145, 4113-4118.

Noman, M.Z., Janji, B., Kaminska, B., Van Moer, K., Pierson, S., Przanowski, P., Buart, S., Berchem, G., Romero, P., Mami-Chouaib, F., et al. (2011). Blocking hypoxia- induced autophagy in tumors restores cytotoxic T-cell activity and promotes regression. Cancer Res 71, 5976-5986.

Nordsmark, M., and Overgaard, J. (2004). Tumor hypoxia is independent of hemoglobin and prognostic for loco-regional tumor control after primary radiotherapy in advanced head and neck cancer. Acta oncologica 43, 396-403.

Olayioye, M.A., Neve, R.M., Lane, H.A., and Hynes, N.E. (2000). The ErbB signaling network: receptor heterodimerization in development and cancer. EMBO J 19, 3159-3167.

Orphanides, C., Fine, L.G., and Norman, J.T. (1997). Hypoxia stimulates proximal tubular cell matrix production via a TGF-beta1-independent mechanism. Kidney international 52, 637-647.

Oswald, F., Liptay, S., Adler, G., and Schmid, R.M. (1998). NF-kappaB2 is a putative target gene of activated Notch-1 via RBP-Jkappa. Mol Cell Biol 18, 2077-2088.

123

Oswald, F., Tauber, B., Dobner, T., Bourteele, S., Kostezka, U., Adler, G., Liptay, S., and Schmid, R.M. (2001). p300 acts as a transcriptional coactivator for mammalian Notch-1. Mol Cell Biol 21, 7761-7774.

Paez-Ribes, M., Allen, E., Hudock, J., Takeda, T., Okuyama, H., Vinals, F., Inoue, M., Bergers, G., Hanahan, D., and Casanovas, O. (2009). Antiangiogenic therapy elicits malignant progression of tumors to increased local invasion and distant metastasis. Cancer cell 15, 220-231.

Palomero, T., Lim, W.K., Odom, D.T., Sulis, M.L., Real, P.J., Margolin, A., Barnes, K.C., O'Neil, J., Neuberg, D., Weng, A.P., et al. (2006). NOTCH1 directly regulates c-MYC and activates a feed-forward-loop transcriptional network promoting leukemic cell growth. Proc Natl Acad Sci U S A 103, 18261-18266.

Pandini, G., Vigneri, R., Costantino, A., Frasca, F., Ippolito, A., Fujita-Yamaguchi, Y., Siddle, K., Goldfine, I.D., and Belfiore, A. (1999). Insulin and insulin-like growth factor-I (IGF-I) receptor overexpression in breast cancers leads to insulin/IGF-I hybrid receptor overexpression: evidence for a second mechanism of IGF-I signaling. Clinical cancer research : an official journal of the American Association for Cancer Research 5, 1935-1944.

Parks, A.L., Klueg, K.M., Stout, J.R., and Muskavitch, M.A. (2000). Ligand endocytosis drives receptor dissociation and activation in the Notch pathway. Development 127, 1373-1385.

Pavelic, K., Kolak, T., Kapitanovic, S., Radosevic, S., Spaventi, S., Kruslin, B., and Pavelic, J. (2003). Gastric cancer: the role of insulin-like growth factor 2 (IGF 2) and its receptors (IGF 1R and M6-P/IGF 2R). J Pathol 201, 430-438.

Pescador, N., Villar, D., Cifuentes, D., Garcia-Rocha, M., Ortiz-Barahona, A., Vazquez, S., Ordonez, A., Cuevas, Y., Saez-Morales, D., Garcia-Bermejo, M.L., et al. (2010). Hypoxia promotes glycogen accumulation through hypoxia inducible factor (HIF)-mediated induction of glycogen synthase 1. PloS one 5, e9644.

Pollak, M. (2012). The insulin and insulin-like growth factor receptor family in neoplasia: an update. Nat Rev Cancer 12, 159-169.

124

Pollak, M.N., Schernhammer, E.S., and Hankinson, S.E. (2004). Insulin-like growth factors and neoplasia. Nat Rev Cancer 4, 505-518.

Pugh, C.W., and Ratcliffe, P.J. (2003). Regulation of angiogenesis by hypoxia: role of the HIF system. Nat Med 9, 677-684.

Railo, M.J., von Smitten, K., and Pekonen, F. (1994). The prognostic value of insulin-like growth factor-I in breast cancer patients. Results of a follow-up study on 126 patients. Eur J Cancer 30A, 307-311.

Rangarajan, A., Talora, C., Okuyama, R., Nicolas, M., Mammucari, C., Oh, H., Aster, J.C., Krishna, S., Metzger, D., Chambon, P., et al. (2001). Notch signaling is a direct determinant of keratinocyte growth arrest and entry into differentiation. EMBO J 20, 3427-3436.

Renehan, A.G., Zwahlen, M., Minder, C., O'Dwyer, S.T., Shalet, S.M., and Egger, M. (2004). Insulin-like growth factor (IGF)-I, IGF binding protein-3, and cancer risk: systematic review and meta-regression analysis. Lancet 363, 1346-1353.

Resnicoff, M., Coppola, D., Sell, C., Rubin, R., Ferrone, S., and Baserga, R. (1994). Growth inhibition of human melanoma cells in nude mice by antisense strategies to the type 1 insulin-like growth factor receptor. Cancer Res 54, 4848-4850.

Riccio, O., van Gijn, M.E., Bezdek, A.C., Pellegrinet, L., van Es, J.H., Zimber-Strobl, U., Strobl, L.J., Honjo, T., Clevers, H., and Radtke, F. (2008). Loss of intestinal crypt progenitor cells owing to inactivation of both Notch1 and Notch2 is accompanied by derepression of CDK inhibitors p27Kip1 and p57Kip2. EMBO reports 9, 377- 383.

Ries, L.A.G. (2007). Cancer survival among adults : U.S. SEER program, 1988-2001, patient and tumor characteristics (Bethesda, MD: U.S. Department of Health and Human Services, National Institutes of Health, National Cancer Institute).

Ronchini, C., and Capobianco, A.J. (2001). Induction of cyclin D1 transcription and CDK2 activity by Notch(ic): implication for cell cycle disruption in transformation by Notch(ic). Mol Cell Biol 21, 5925-5934.

125

Rotsch, M., Maasberg, M., Erbil, C., Jaques, G., Worsch, U., and Havemann, K. (1992). Characterization of insulin-like growth factor I receptors and growth effects in human lung cancer cell lines. J Cancer Res Clin Oncol 118, 502-508.

Ryan, H.E., Lo, J., and Johnson, R.S. (1998). HIF-1 alpha is required for solid tumor formation and embryonic vascularization. EMBO J 17, 3005-3015.

Sahin, U., Weskamp, G., Kelly, K., Zhou, H.M., Higashiyama, S., Peschon, J., Hartmann, D., Saftig, P., and Blobel, C.P. (2004). Distinct roles for ADAM10 and ADAM17 in ectodomain shedding of six EGFR ligands. J Cell Biol 164, 769-779.

Sahlgren, C., Gustafsson, M.V., Jin, S., Poellinger, L., and Lendahl, U. (2008). Notch signaling mediates hypoxia-induced tumor cell migration and invasion. Proc Natl Acad Sci U S A 105, 6392-6397.

Sandler, A., Gray, R., Perry, M.C., Brahmer, J., Schiller, J.H., Dowlati, A., Lilenbaum, R., and Johnson, D.H. (2006). Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. N Engl J Med 355, 2542-2550.

Schipani, E., Maes, C., Carmeliet, G., and Semenza, G.L. (2009). Regulation of osteogenesis-angiogenesis coupling by HIFs and VEGF. Journal of bone and mineral research : the official journal of the American Society for Bone and Mineral Research 24, 1347-1353.

Schroeter, E.H., Kisslinger, J.A., and Kopan, R. (1998). Notch-1 signalling requires ligand-induced proteolytic release of intracellular domain. Nature 393, 382-386.

Schwarzer, R., Tondera, D., Arnold, W., Giese, K., Klippel, A., and Kaufmann, J. (2005). REDD1 integrates hypoxia-mediated survival signaling downstream of phosphatidylinositol 3-kinase. Oncogene 24, 1138-1149.

Sell, C., Dumenil, G., Deveaud, C., Miura, M., Coppola, D., DeAngelis, T., Rubin, R., Efstratiadis, A., and Baserga, R. (1994). Effect of a null mutation of the insulin-like growth factor I receptor gene on growth and transformation of mouse embryo fibroblasts. Mol Cell Biol 14, 3604-3612.

126

Sell, C., Rubini, M., Rubin, R., Liu, J.P., Efstratiadis, A., and Baserga, R. (1993). Simian virus 40 large tumor antigen is unable to transform mouse embryonic fibroblasts lacking type 1 insulin-like growth factor receptor. Proc Natl Acad Sci U S A 90, 11217-11221.

Semenza, G.L. (1999). Regulation of mammalian O2 homeostasis by hypoxia-inducible factor 1. Annual review of cell and developmental biology 15, 551-578.

She, Q.B., Halilovic, E., Ye, Q., Zhen, W., Shirasawa, S., Sasazuki, T., Solit, D.B., and Rosen, N. (2010). 4E-BP1 is a key effector of the oncogenic activation of the AKT and ERK signaling pathways that integrates their function in tumors. Cancer cell 18, 39-51.

Shigematsu, H., Lin, L., Takahashi, T., Nomura, M., Suzuki, M., Wistuba, II, Fong, K.M., Lee, H., Toyooka, S., Shimizu, N., et al. (2005). Clinical and biological features associated with epidermal growth factor receptor gene mutations in lung cancers. J Natl Cancer Inst 97, 339-346.

Siegel, R., Naishadham, D., and Jemal, A. (2013). Cancer statistics, 2013. CA Cancer J Clin 63, 11-30.

Simon, M.C., and Keith, B. (2008). The role of oxygen availability in embryonic development and stem cell function. Nat Rev Mol Cell Biol 9, 285-296.

Song, L.L., Peng, Y., Yun, J., Rizzo, P., Chaturvedi, V., Weijzen, S., Kast, W.M., Stone, P.J., Santos, L., Loredo, A., et al. (2008). Notch-1 associates with IKKalpha and regulates IKK activity in cervical cancer cells. Oncogene 27, 5833-5844.

Spitz, M.R., Barnett, M.J., Goodman, G.E., Thornquist, M.D., Wu, X., and Pollak, M. (2002). Serum insulin-like growth factor (IGF) and IGF-binding protein levels and risk of lung cancer: a case-control study nested in the beta-Carotene and Retinol Efficacy Trial Cohort. Cancer Epidemiol Biomarkers Prev 11, 1413-1418.

Stern, K.A., Place, T.L., and Lill, N.L. (2008). EGF and amphiregulin differentially regulate Cbl recruitment to endosomes and EGF receptor fate. Biochem J 410, 585- 594.

127

Sullivan, J.P., Spinola, M., Dodge, M., Raso, M.G., Behrens, C., Gao, B., Schuster, K., Shao, C., Larsen, J.E., Sullivan, L.A., et al. (2010). Aldehyde dehydrogenase activity selects for lung adenocarcinoma stem cells dependent on notch signaling. Cancer Res 70, 9937-9948.

Tamura, K., Taniguchi, Y., Minoguchi, S., Sakai, T., Tun, T., Furukawa, T., and Honjo, T. (1995). Physical interaction between a novel domain of the receptor Notch and the transcription factor RBP-J kappa/Su(H). Current biology : CB 5, 1416-1423.

Tao, Y., Pinzi, V., Bourhis, J., and Deutsch, E. (2007). Mechanisms of disease: signaling of the insulin-like growth factor 1 receptor pathway--therapeutic perspectives in cancer. Nature clinical practice Oncology 4, 591-602.

Testa, J.R., and Tsichlis, P.N. (2005). AKT signaling in normal and malignant cells. Oncogene 24, 7391-7393.

Tolcher, A.W., Sarantopoulos, J., Patnaik, A., Papadopoulos, K., Lin, C.C., Rodon, J., Murphy, B., Roth, B., McCaffery, I., Gorski, K.S., et al. (2009). Phase I, pharmacokinetic, and pharmacodynamic study of AMG 479, a fully human monoclonal antibody to insulin-like growth factor receptor 1. J Clin Oncol 27, 5800-5807.

Ullrich, A., Gray, A., Tam, A.W., Yang-Feng, T., Tsubokawa, M., Collins, C., Henzel, W., Le Bon, T., Kathuria, S., Chen, E., et al. (1986). Insulin-like growth factor I receptor primary structure: comparison with insulin receptor suggests structural determinants that define functional specificity. EMBO J 5, 2503-2512.

Vadysirisack, D.D., and Ellisen, L.W. (2012). mTOR activity under hypoxia. Methods in molecular biology 821, 45-58.

Valentinis, B., Morrione, A., Taylor, S.J., and Baserga, R. (1997). Insulin-like growth factor I receptor signaling in transformation by src oncogenes. Mol Cell Biol 17, 3744-3754. van Es, J.H., van Gijn, M.E., Riccio, O., van den Born, M., Vooijs, M., Begthel, H., Cozijnsen, M., Robine, S., Winton, D.J., Radtke, F., et al. (2005). Notch/gamma- secretase inhibition turns proliferative cells in intestinal crypts and adenomas into goblet cells. Nature 435, 959-963.

128

Walker, R.A. (1998). The erbB/HER type 1 tyrosine kinase receptor family. J Pathol 185, 234-235.

Wang, G.L., and Semenza, G.L. (1995). Purification and characterization of hypoxia- inducible factor 1. J Biol Chem 270, 1230-1237.

Wang, R., Zhang, Y.W., Zhang, X., Liu, R., Zhang, X., Hong, S., Xia, K., Xia, J., Zhang, Z., and Xu, H. (2006). Transcriptional regulation of APH-1A and increased gamma-secretase cleavage of APP and Notch by HIF-1 and hypoxia. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 20, 1275-1277.

Ward, C.W., and Garrett, T.P. (2004). Structural relationships between the insulin receptor and epidermal growth factor receptor families and other proteins. Current opinion in drug discovery & development 7, 630-638.

Ward, C.W., Gough, K.H., Rashke, M., Wan, S.S., Tribbick, G., and Wang, J. (1996). Systematic mapping of potential binding sites for Shc and Grb2 SH2 domains on insulin receptor substrate-1 and the receptors for insulin, epidermal growth factor, platelet-derived growth factor, and fibroblast growth factor. J Biol Chem 271, 5603-5609.

Wartenberg, M., Ling, F.C., Muschen, M., Klein, F., Acker, H., Gassmann, M., Petrat, K., Putz, V., Hescheler, J., and Sauer, H. (2003). Regulation of the multidrug resistance transporter P-glycoprotein in multicellular tumor spheroids by hypoxia- inducible factor (HIF-1) and reactive oxygen species. FASEB journal : official publication of the Federation of American Societies for Experimental Biology 17, 503-505.

Waterman, H., Sabanai, I., Geiger, B., and Yarden, Y. (1998). Alternative intracellular routing of ErbB receptors may determine signaling potency. J Biol Chem 273, 13819-13827.

Weng, A.P., Ferrando, A.A., Lee, W., Morris, J.P.t., Silverman, L.B., Sanchez-Irizarry, C., Blacklow, S.C., Look, A.T., and Aster, J.C. (2004). Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science 306, 269-271.

129

Weng, A.P., Nam, Y., Wolfe, M.S., Pear, W.S., Griffin, J.D., Blacklow, S.C., and Aster, J.C. (2003). Growth suppression of pre-T acute lymphoblastic leukemia cells by inhibition of notch signaling. Mol Cell Biol 23, 655-664.

Werner, H., Karnieli, E., Rauscher, F.J., and LeRoith, D. (1996). Wild-type and mutant p53 differentially regulate transcription of the insulin-like growth factor I receptor gene. Proc Natl Acad Sci U S A 93, 8318-8323.

Weroha, S.J., and Haluska, P. (2008). IGF-1 receptor inhibitors in clinical trials--early lessons. Journal of mammary gland biology and neoplasia 13, 471-483.

Westhoff, B., Colaluca, I.N., D'Ario, G., Donzelli, M., Tosoni, D., Volorio, S., Pelosi, G., Spaggiari, L., Mazzarol, G., Viale, G., et al. (2009). Alterations of the Notch pathway in lung cancer. Proc Natl Acad Sci U S A 106, 22293-22298.

Wisnivesky, J.P., Yankelevitz, D., and Henschke, C.I. (2005). Stage of lung cancer in relation to its size: part 2. Evidence. Chest 127, 1136-1139.

Wu, L., Aster, J.C., Blacklow, S.C., Lake, R., Artavanis-Tsakonas, S., and Griffin, J.D. (2000). MAML1, a human homologue of Drosophila mastermind, is a transcriptional co-activator for NOTCH receptors. Nature genetics 26, 484-489.

Wu, Y., Cain-Hom, C., Choy, L., Hagenbeek, T.J., de Leon, G.P., Chen, Y., Finkle, D., Venook, R., Wu, X., Ridgway, J., et al. (2010). Therapeutic antibody targeting of individual Notch receptors. Nature 464, 1052-1057.

Wu, Y., Tewari, M., Cui, S., and Rubin, R. (1996). Activation of the insulin-like growth factor-I receptor inhibits tumor necrosis factor-induced cell death. J Cell Physiol 168, 499-509.

Yee, D. (2006). Targeting insulin-like growth factor pathways. Br J Cancer 94, 465-468.

Yeung, T.M., Gandhi, S.C., and Bodmer, W.F. Hypoxia and lineage specification of cell line-derived colorectal cancer stem cells. Proc Natl Acad Sci U S A 108, 4382- 4387.

130

Yonesaka, K., Zejnullahu, K., Lindeman, N., Homes, A.J., Jackman, D.M., Zhao, F., Rogers, A.M., Johnson, B.E., and Janne, P.A. (2008). Autocrine production of amphiregulin predicts sensitivity to both gefitinib and cetuximab in EGFR wild- type cancers. Clinical cancer research : an official journal of the American Association for Cancer Research 14, 6963-6973.

Yotsumoto, F., Fukami, T., Yagi, H., Funakoshi, A., Yoshizato, T., Kuroki, M., and Miyamoto, S. (2010). Amphiregulin regulates the activation of ERK and Akt through epidermal growth factor receptor and HER3 signals involved in the progression of pancreatic cancer. Cancer science 101, 2351-2360.

Yu, H., Spitz, M.R., Mistry, J., Gu, J., Hong, W.K., and Wu, X. (1999). Plasma levels of insulin-like growth factor-I and lung cancer risk: a case-control analysis. J Natl Cancer Inst 91, 151-156.

Yuan, J., Narayanan, L., Rockwell, S., and Glazer, P.M. (2000). Diminished DNA repair and elevated mutagenesis in mammalian cells exposed to hypoxia and low pH. Cancer Res 60, 4372-4376.

Zavadil, J., Cermak, L., Soto-Nieves, N., and Bottinger, E.P. (2004). Integration of TGF- beta/Smad and Jagged1/Notch signalling in epithelial-to-mesenchymal transition. EMBO J 23, 1155-1165.

Zhao, J., Chen, H., Peschon, J.J., Shi, W., Zhang, Y., Frank, S.J., and Warburton, D. (2001). Pulmonary hypoplasia in mice lacking tumor necrosis factor-alpha converting enzyme indicates an indispensable role for cell surface protein shedding during embryonic lung branching morphogenesis. Developmental biology 232, 204-218.

Zhong, H., Chiles, K., Feldser, D., Laughner, E., Hanrahan, C., Georgescu, M.M., Simons, J.W., and Semenza, G.L. (2000). Modulation of hypoxia-inducible factor 1alpha expression by the epidermal growth factor/phosphatidylinositol 3- kinase/PTEN/AKT/FRAP pathway in human prostate cancer cells: implications for tumor angiogenesis and therapeutics. Cancer Res 60, 1541-1545.

Zundel, W., Schindler, C., Haas-Kogan, D., Koong, A., Kaper, F., Chen, E., Gottschalk, A.R., Ryan, H.E., Johnson, R.S., Jefferson, A.B., et al. (2000). Loss of PTEN facilitates HIF-1-mediated gene expression. Genes Dev 14, 391-396.

VITA

The author, Shuang Liang, was born in Shijiazhuang, Hebei, China on February

18th, 1984 to Jiansheng Liang and Chaoying Liu. She is married to Zhenyu Zhong, and is blessed with an energetic 8-month-old son, Eric Haotian Zhong.

Shuang received a Bachelor of Science in Bioengineering from Hebei University

(China) in June of 2006. Her thesis was on optimizing the expression level of soluble

SARS spike protein mediated by adenovirus in HEK293 cells. She was awarded the

Hebei University Scholarship for Outstanding Students (2002-2005).

After graduating from Hebei University, Shuang joined the department of

Microbiology and Immunology, Rosalind Franklin University of Medicine and Science

(North Chicago, IL) as a graduate student in July 2006.

Shuang joined the Molecular Biology Program of Loyola University Chicago in

August 2007. She is currently completing her doctoral studies under the guidance of Dr.

Maurizio Bocchetta. Her research focuses on targeting the Notch-1/IGF-1R/Akt axis in an orthotopic model of advanced non-small cell lung cancer. She received a fellowship award (2011-2012) for her dissertation proposal from the Oncology Institute of Loyola

University Chicago. After completing her Ph.D., she intends to pursue research in the field of oncology.

131