UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Hyperactivation of ERK1/2 by DUSP6 inhibition leads to lethality in lung adenocarcinoma Oh, Min Hee 2019

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2019_may_oh_minhee.pdf [ 4.29MB ]
Metadata
JSON: 24-1.0376231.json
JSON-LD: 24-1.0376231-ld.json
RDF/XML (Pretty): 24-1.0376231-rdf.xml
RDF/JSON: 24-1.0376231-rdf.json
Turtle: 24-1.0376231-turtle.txt
N-Triples: 24-1.0376231-rdf-ntriples.txt
Original Record: 24-1.0376231-source.json
Full Text
24-1.0376231-fulltext.txt
Citation
24-1.0376231.ris

Full Text

 Hyperactivation of ERK1/2 by DUSP6 inhibition leads to lethality in lung adenocarcinoma  by  Min Hee Oh  B.Sc., University of British Columbia, 2015  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Interdisciplinary Oncology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  January 2019  ©Min Hee Oh, 2019    ii   The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, a thesis/dissertation entitled:  Hyperactivation of ERK1/2 by DUSP6 inhibition leads to lethality in lung adenocarcinoma  submitted by Min Hee Oh  in partial fulfillment of the requirements for the degree of Master of Science in Interdisciplinary Oncology  Examining Committee: William Lockwood Supervisor Peter Stirling Supervisory Committee Member Wan Lam Supervisory Committee Member Cathie Garnis Additional Examiner    Additional Supervisory Committee Members: Aly Karsan Supervisory Committee Member  Supervisory Committee Member   iii  Abstract   Mutations in the Epidermal Growth Factor Receptor (EGFR) and Kirsten Rat Sarcoma (KRAS) genes occur in a mutually exclusive manner in ~15% and ~30% of all lung adenocarcinomas (LACs), respectively. Using Doxycycline (Dox)-regulated gene expression vectors, we have previously demonstrated that the forced co-expression of EGFR and KRAS mutants in LAC cells induces lethality through the hyperactivation of the RAS-mitogen-activated protein kinase (MAPK) pathway. A subsequent phosphoproteomic assay using Tet-O-KRASG12V-PC9 cells, which carry an endogenous EGFR mutation and was engineered to express KRASG12V upon Dox treatment, revealed that phosphorylation of extracellular signal-regulated kinases (ERKs) increased acutely and dramatically compared to the Tet-O-GFP-PC9 control. This suggested that early activation of ERK1/2 is a crucial event in mediating the observed lethality. Additionally, genetic and pharmacological inhibition of ERK1/2 rescued multiple co-expression LAC cells, confirming that ERK is the main mediator of this phenomena. Here, I aim to investigate whether KRAS- or EGFR-driven LAC cells exploit any existing negative regulatory mechanisms of the ERK to maintain its levels below its upper signalling threshold.  Because MAPK signalling is typically regulated by phosphatases, our group performed an analysis of the MAPK phosphatase expression data comparing two LAC TCGA tumor subsets – tumors with (n=107) and without (n=123) either EGFR or KRAS mutation. This analysis revealed that Dual-specificity phosphatase 6 (DUSP6) is the only phosphatase that is up-regulated in tumors with a mutation in either two genes in comparison to their wildtype counterparts, suggesting that these tumors may be dependent on a robust DUSP6 activity to moderate the P-ERK1/2 levels and prevent ERK hyperactivation. Furthermore, when DUSP6 was inhibited in mutant KRAS or mutant EGFR bearing LAC cells using DUSP6 small-interfering RNAs (siRNAs) or a DUSP6 iv  inhibitor called (E)-2-benzylidene-3-(cyclohexylamino)-2,3-dihydro-1H-inden-1-one (BCI), we observed that only the mutant bearing LAC cells were more sensitive to DUSP6 inhibition than the KRAS and EGFR wildtype cells. Such findings suggest a potential therapeutic scenario in EGFR or KRAS mutant LACs can be targeting through inhibiting DUSP6, a key negative feedback regulator that prevents the hyperactivation of ERK.   v  Lay Summary  Lung adenocarcinoma (LAC) is a common subtype of lung cancer (LC) with limited therapeutic options. Many LACs have a mutation in the EGFR(15%) or KRAS(30%) genes but never both. Such mutations drive tumor development by continuously activating cellular processes that promote growth and proliferation. Currently, drugs that inhibit EGFR have been developed but eligible patients inevitably develop resistance to them. To address these ongoing challenges, we have previously identified that LACs driven by these mutations can also be killed by further amplifying their activity. My findings presented in this thesis show that these LACs are dependent on a protein called DUSP6, which ensures that signalling activities do not exceed tolerable levels. Consequently, blocking DUSP6 in LAC is shown here to be toxic for cancer cells that have a mutation in KRAS or EGFR. This suggests that DUSP6 can become a new therapeutic target in LACs with EGFR or KRAS mutations.    vi  Preface    The experimental design was developed with guidance and assistance from Dr. William W. Lockwood. Analysis of the TCGA RNA-SEQ data was performed by Dr. William Lockwood. Western blots were performed by Bryant Harbourne. All other experiments performed in this study were performed by myself. The writing of this thesis was completed by me with guidance and editing by Dr. William Lockwood.    vii  Table of Contents  ABSTRACT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii LAY SUMMARY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v PREFACE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii LIST OF TABLES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x LIST OF FIGURES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  xi LIST OF SYMBOLS AND ABBREVIATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvi DEDICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii CHAPTER 1: INTRODUCTION ………………………………………………………………1 1.1 Background on cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 1.2 Cell signalling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 1.2.1 Cell signalling, cell stress and extracellular signals . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.2.2 Phosphorylation as a mechanism of cell signalling: kinases and phosphatases . . . . 5 1.3 Lung cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 1.3.1 Lung cancer and aetiologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 1.3.2 Classifications of lung cancer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 1.3.3 Mutation spectrums of lung cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 1.3.4 Current treatment strategies of lung cancer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13 1.3.5 Current challenges in advanced NSCLC therapeutics. . . . . . . . . . . . . . . . . . . . . . . 15 1.3.6 Novel therapeutic approach to LC: Combination therapy and synthetic lethality . . 17 1.4 Oncogenic signalling in lung cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 viii  1.4.1 PI3K-AKT and RAS-MAPK pathway activity in mutant EGFR- and mutant KRAS-driven cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 1.5 Thesis theme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1.5.1 Background: Does functional redundancy underlie the mutual exclusivity between KRAS and EGFR mutations in NSCLC? . . . . . . . . . . . . . .  . . . . . . . . . . . . . . . . . . 23 1.5.1.1 Tet-ON system to establish tetracycline-regulated co-expression status in NSCLC cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1.5.2 Co-expression of EGFR and KRAS induces lethality in LAC cells . . . . . . . . . . . . 26 1.5.3 ERK1/2 as the main mediator of Tet-O-RAS mediated toxicity in mKRAS/mEGFR LAC cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 1.5.4 Therapeutic application: EGFR- and KRAS-driven LACs are reliant on negative regulators of the RAS-ERK axis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 1.5.4.1 Negative regulators of ERK in the RAS-ERK axis . . . . . . . . . . . . . . . . . . . . . . 33 1.5.5 DUSP6 as the main negative regulator in EGFR- and KRAS-driven LACs . . . . . . 36 1.5.6 Rationale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 CHAPTER 2: MATERIALS AND METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 2  2.1 Cell lines and culture conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 2.2 Plasmid and generation of stable cell lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 2.3 Western blot analysis of protein levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 2.4 siRNA transfections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 2.5 Viability measurement assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 2.6 BCI dose-response curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 3    ix  CHAPTER 3: RESULT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  43  3.1 Transient knockdown of DUSP6 increases P-ERK and reduces viability in EGFR- or KRAS-driven LAC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.1.1 Knockdown of DUSP6 increases apoptosis in mKRAS/mEGFR LAC cells . . . . . 47 3.1.2 Confirmation of the specificity of the DUSP6 siRNA pool . . . . . . . . . . . . . . . . . . 49 3.2 Identification of DUSP6 as a druggable target using BCI – a DUSP1, 6 small molecule inhibitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.2.1 BCI induced hyper-ERK toxicity is specific to DUSP6 inhibition, not DUSP1 . . . 55 3.3 Toxicity from DUSP6 knockdown is partially rescued through suppression of ERK . . .57 3.4 Inhibition of ERK rescues cells from BCI induced lethality . . . . . . . . . . . . . . . . . . . . . 59 3.5 Sensitization of a BCI insensitive line, HCC95, through prolonged EGF treatment . . . 61 CHAPTER 4: DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 4  4.1 Therapeutic application of hyper-RAS/ERK lethality . . . . . . . . . . . . . . . . . . . . . . . . . . 64 4.2 DUSP6 inhibition in mutant EGFR-/KRAS-driven NSCLC generates toxicity . . . . . . 65 4.2.1 RNA-interference experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 4.2.2 BCI experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.3 DUSP6 inhibition lethality is mediated through the hyperactivation of ERK . . . . . . . . 68 4.4 P-ERK dynamics in BCI-sensitive vs. insensitive cells . . . . . . . . . . . . . . . . . . . . . . . . . 69 CHAPTER 5: CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70  CHAPTER 6: SUPPLEMENTAL FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73  REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74  x  List of Tables   Table 1.1  List  of  dual-specificity  MAPK  phosphatases  and  their  classifications  and  localizations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35  xi  List of Figures   Figure 1.1 Therapeutic targeting of the hallmarks of cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Figure 1.2 Mutation spectrums of NSCLC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Figure 1.3 Organization of the EGFR-RAF-MEK-ERK signalling axis . . . . . . . . . . . . . . . . . . . . 21 Figure 1.4 Three-tiered signalling hierarchy of the MAPK (ERK, JNK, p38) pathways . . . . . . . 22 Figure 1.5 Illustration of the Tet-ON system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Figure 1.6 Growth assay of Tet-inducible co-expression LAC cells . . . . . . . . . . . . . . . . . . . . . . 27 Figure 1.7 Pharmacological inhibition  of  ERK1/2  activation  rescues  co-expression  lethality in            PC9s, H358s and H1975s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Figure 1.8 Phospho-kinase antibody array in PC9EGFR-DEL-Tet-O-KRASG12V after 24-hours of dox       treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Figure 1.9 Hypothetical model of ERK levels in KRAS/EGFR mutant versus wildtype cells . . . 32  Figure 1.10 Heat map of TCGA RNA-seq data of mKRAS/mEGFR and KRAS/EGFR WT LAC         tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Figure 3.1 Time-course viability assay following double knockdown of DUSP6 in PC9 . . . . . . 44 Figure 3.2 End-point (Day  5) viability  assay  following  a  double-knockdown of DUSP6 in four       NSCLC cell lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Figure 3.3 Western blot analysis of P-ERK levels 24 hours post-transfection. . . . . . . . . . . . . . . 46 Figure 3.4 Western blot analysis of cleaved PARP at end-point (Day 5) of the double-transfection      assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Figure 3.5 Testing of the individual siRNAs of the siDUSP6 pool in PC9s . . . . . . . . . . . . . . . . . 51 xii  Figure 3.6 BCI sensitivity curves in 11 NSCLC cells after 72 hours of treatment . . . . . . . . . . . . 53 Figure 3.7 Assessment of dose-dependent P-ERK changes after 6 hours of BCI treatment . . . . . 54 Figure 3.8 Double-transfection of siDUSP1 and siDUSP6 in H1975 . . . . . . . . . . . . . . . . . . . . . . 56 Figure 3.9 siDUSP6 rescue assays in four different ERK KD PC9 cell lines . . . . . . . . . . . . . . . . 58 Figure 3.10 Combination drug treatment of BCI and VX-11E in H358s . . . . . . . . . . . . . . . . . . . 60 Figure 3.11 HCC95 sensitization to BCI after 10 days of EGF (100ng/mL) treatment . . . . . . . . 62 Figure 3.12 BCI dose-response curves in HCC95s with and without EGF (100ng/mL) pre-        treatment for 10 days . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Fig S1 Low-density NAC-BCI rescue assay in PC9 and H358 . . . . . . . . . . . . . . . . . . . . . . . . . . 73   xiii  List of Symbols and Abbreviations    AKT protein kinase B ALK anaplastic lymphoma kinase ATP adenosine triphosphate BAD BCL-2-associated death promoter BCI (E)-2-benzylidene-3-(cyclohexylamino)-2,3-dihydro-1H-inden-1-one BRAF v-Raf murine sarcoma viral oncogene homolog B BRCA breast cancer susceptibility gene Dox doxycycline DUSP dual-specificity phosphatase  EGFR epidermal growth factor receptor EMT epithelial-to-mesenchymal transition  ERK extracellular signal-regulated kinase ETS E26 transformation-specific factors FDA Food and Drug Administration of U.S. FGFR1 fibroblast growth factor receptor 1 GSH glutathione GSK3β glycogen synthase kinase 3 beta GTP guanosine triphosphate HER2 human epidermal growth factor receptor 2 IL-1 interleukin-1 JNK c-Jun N-terminal kinase KRAS kirsten rat sarcoma LAC lung adenocarcinoma LC lung cancer LCC large-cell lung cancer LDCT low-dose computed tomography MAPK mitogen-activated protein kinase MAPKK mitogen-activated protein kinase kinase MAPKKK mitogen-activated protein kinase kinase kinase xiv  mEGFR mutant EGFR MEK mitogen-activated protein kinase  MET hepatocyte growth factor receptors MKP MAPK phosphatase mKRAS mutant KRAS mTOR mammalian target of rapamycin NAC N-acetyl-cysteine NF-қβ nuclear factor kappa-light-chain-enhancer of activated B cells NNK nitrosamine 4-(methylnitrosamino)-1-1(3-pyridyl)-1-butanone Non-T non-targeting NOS not otherwise specified NSCLC non-small cell lung cancer P- phospho- PAH polycyclic aromatic hydrocarbons PAPSS1 3’-phosphoadenosine 5’-phosphosulfate synthase 1 PARP poly (ADP-ribose) polymerase PD-1 programmed cell death protein 1 PI3K phosphoinositide 3-kinase PIK3CA phosphatidylionositol-4,5-biphosphate 3-kinase catalytic subunit alpha PTEN phosphatase and tensin homolog RAF rapidly accelerated fibrosarcoma proteins RB1 retinoblastoma protein 1 ROS reactive oxygen species R-SH thiol groups RTK receptor tyrosine kinase rtTA reverse TET-controlled transactivator SCLC small cell lung cancer SCR scramble Ser serine shRNA short-hairpin RNAs siRNA small-interfering RNA SOS1 son of sevenless 1 xv  SPRED sprout-related proteins SPRY sproutys SQCC squamous-cell carcinoma STAT3 signal transducer and activator of transcription 3 TET tetracyclin Thr threonine  TKI tyrosine kinase inhibitor TME tumor microenvironment  TNM tumor, lymph nodes, metastasis TP53 tumor protein 53 TRE tetracycline regulated promoter Tyr tyrosine  UTR untranslated region WT wildtype xCT Xc- cysteine transporter   xvi  Acknowledgments    First and foremost, I would like to thank Dr. William Lockwood for his guidance and support through the completion of this thesis. Not only did he teach me a tremendous amount about cancer, but he has also taught me how to be a better team player and to build meaningful connections at work by setting an example for the lab. Because of this experience, I have no doubt that I have evolved not only as a student but as a person.  Also, I would like to thank each member of the lab (past and present) for making the long hours and the constant frustration at the unknown more bearable. In particular, I would like to thank Bryant Harbourne for his patience and teaching me every skill I know in the lab. I would also like to thank Daniel Lu and Jack Calder for all the brainstorming sessions during which they shared their insights and contemplated about my projects as if they were their own.  I would also like to extend my sincerest gratitude to the Varmus group. In particular, I would like to thank Dr. Arun Unni, who is one of the main contributors of this project and have discovered the fundamental ideas of this thesis. Asides from these groups, I would like to extend my sincerest gratitude to all the members of my supervisory committee, Drs. Wan Lam and Aly Karsan, for their time and insightful feedback throughout this project.  I also wish to acknowledge the BC Cancer Foundation, Canadian Institutes of Health Research and the Terry Fox Research Institute for funding this project. Without their generous support, this thesis would not have been completed.  Lastly, I am extremely thankful of my family and Charlee for their unconditional support through this journey.    xvii  Dedication        This thesis is dedicated to my parents, Il-Keun Oh and Young-Im Ju, and my brother, Doo-Hwan Oh. 1  Chapter 1: Introduction  1.1 Background on cancer Cancer is a group of genetic diseases that arise when cells in the body grow uncontrollably [1]. Malignant tumor or neoplasm is an abnormal mass of tissues that can invade its surrounding tissues and relocate to different parts of the body through a process called metastasis. These masses affect the body by disturbing the function and architecture of the invaded tissues and such disruptions can lead to a serious illness or death. However, not all neoplasms are malignant or “harmful” as some masses grow in a controlled and confined manner without ever disturbing the surrounding cells. These neoplasms also do not metastasize to a different site in the body from its primary site of growth. Theses neoplasms are described as “benign” cancers or tumors. Cancers are known to have numerous hereditary and external aetiologies, such as exposures to different carcinogens [1]. Carcinogens are compounds that are known to cause cancer in human cells. Upon exposure, carcinogens can damage cellular DNAs and introduce permanent mutations. Some of these mutations are oncogenic and can initiate a cell’s transformation from normal to a malignant state. However, the development of cancer is a complex process that requires numerous oncogenic mutations that equip the cells with multiple hallmark capabilities of cancer [1]. On top of oncogenic mutations, the tumor microenvironment (TME) is another major variable in cancer development [1,2,3]. The TME is comprised of a heterogeneous mixture of non-cancerous cells, like immune cells, extracellular factors, and stromal and tumor vasculatures. Each of these components individually equip the TME with either a tumor-inducing or suppressive ability [1,2,3].  So far, there are ten hallmark traits of cancer that have been described, which include traits like uncontrolled proliferation and evasion of apoptosis (Figure 1.1). Despite the apparent 2  simplicity of cancer when their behaviors are captured and presented in just ten hallmark traits, cancer is extremely complex. How these ten traits interplay with one another to give rise to the hundreds of different types of tumors is not yet fully understood and thus making cancer unpredictable and difficult to treat [1]. As a matter of fact, even the individual cases that belong to a same class can often have its own unique biology and behavior and the number of types of cancer is continuously growing as research progresses.    3    Figure 1.1: Hallmarks of cancer. (© 2011 Elsevier Inc, by permission).  The figure depicts the ten hallmark traits of cancer and an example of a therapeutic approach that targets each trait. In addition to these ten hallmarks, tumor microenvironment, which is depicted by the illustration inside of the circle of hallmarks, plays a significant role in tumor development through its interactions with the tumor. Figure 6 from © Hanahan, D., & Weinberg, R. (2011). Hallmarks of cancer: the next generation. Cell, 144(5), 646-674. Page 668. By permission from publisher [2].  4  1.2 Cell signalling   1.2.1 Cell signalling, cell stress and extracellular signals Cells constantly adapt to the changes in their external environment, such as cell stress or a flux of an extracellular signalling molecule. When a cell encounters a stressful environment, such as radiation, the extent and the mode of stress (i.e. oxidative or irradiation) are the primary determinants of the type of the cellular response that is elicited [4]. For instance, reactive oxygen species (ROS) exert oxidative stress when their highly-reactive unpaired electrons react with random thiol groups (R-SH) in the cell to interfere with their role in stabilizing many intracellular structures. Depending on the degree of these random ROS attacks, cells can respond by either trying to reduce the rate of oxidations through internal compensatory mechanisms or triggering apoptosis (cell death) [5, 6].  Unlike cell stress, extracellular signals interact with the cell in a specific manner that starts when a extracellular signal molecule interacts with a corresponding extracellular receptor on the cell (i.e. Epidermal Growth Factor (EGF) binds to an EGF-Receptor (EGFR)). This interaction induces a conformational change in the receptor from their default inactive state to a catalytically active state [1]. This is considered as a signalling event as the EGF has effectively signaled the receptor to change its conformation. Activated receptor subsequently performs post-translational modifications on other intracellular signalling molecules and changes their conformation and activity levels. These signalling events occur in a cascade-like manner through multiple tiers of protein pairs and each molecule with the same signalling partner each time. This relay of events continues until the end of a signaling cascade, at which the last signalling molecule alter the transcription profile within the cell. The transcriptional changes ultimately affects different cellular processes in a manner that corresponds to the initial message (i.e. EGF signals for increased 5  proliferation). There are multiple receptor and extracellular signalling molecule pairs that elicit different cellular functions. Such mechanisms are often exploited in experiments to induce specific cellular responses through treating cells with an extracellular signalling molecule. Additionally, all of these pathways are tightly regulated through multiple positive or negative feedback mechanisms that act as safeguards that try to maintain a normal signalling level.   1.2.2 Phosphorylation as a mechanism of cell signalling: kinases and phosphatases While intracellular signalling events include several different forms of post-translational modifications, such as acetylation and methylation, phosphorylation is the most common reaction utilized for signalling [7]. Phosphorylation is an enzymatic reaction in which the γ-phosphoryl group from adenosine triphosphate (ATP) is transferred onto an amino acid of the target protein, which is usually a Serine (Ser), Threonine (Thr) or Tyrosine (Tyr) residue. A class of enzyme called kinase catalyzes these reactions by binding to ATP and the target protein at its active site. When the reaction is complete, and the phosphate group is transferred onto the target protein, the negative charge of the transferred phosphate group reconfigures the protein either to its active or its inactive conformation [7].  Kinases can be divided into two main categories based on their substrate-specificity: Ser/Thr-specific or Tyr-specific kinases. Some kinases, such as the mitogen-activated protein kinases (MEKs), can phosphorylate all three types of residues and are thus referred to as dual-specificity kinases [7]. Moreover, kinases, including the membrane-bound receptor kinases and the cytoplasmic kinases, like MEKs and ERKs, not only phosphorylate but are phosphorylated themselves by other kinases. Kinases contain a protein segment called the activation loop, which is typically located next to the catalytic portion of the enzyme. In its default inactive state, the activation loop is packed 6  away from the cytosol in a hydrophobic pocket. Phosphorylation of the activation loop and the subsequent introduction of charged moieties destabilizes the hydrophobic complex and exposes the catalytic portion of the kinase [7,8]. For instance, the dual phosphorylation of Thr202/Tyr204 residues of the activation loop of ERK1 by MEK1 activates ERK1. However, kinases typically have many more phosphorylation sites than affect their catalytic activity state other than the activation loop. For example, the autophosphorylation site of ERK1 at Tyr207 residue is known to decrease the kinase activity of ERK1 [9].  Kinase activities can also be regulated through a process called dephosphorylation. Dephosphorylation is the reverse process of phosphorylation and is catalyzed by a group of enzymes called phosphatases. Less is known about the structure and function of phosphatases compared to kinases [7]. Phosphatases, like kinases, can be divided into either Ser/Thr-specific phosphatase or Tyr-specific phosphatase. While most phosphatases can be grouped into one or the other, there are dual-specificity phosphatases, also known as DUSPs, that can dephosphorylate Ser, Thr and Tyr.   7  1.3 Lung cancer Lung cancer is a disease in which the cells of the respiratory tract grow uncontrollably to form malignant tumors in the airway. It is the most commonly diagnosed type of cancer and a leading cause of cancer deaths in both men and women, accounting for up to 1.69 million deaths worldwide in 2015 [10]. Despite the continued improvements in LC diagnostics and therapeutics, the average 5-year survival rate has remained low at 18% or less, which is significantly lower than other commonly diagnosed cancer types, such as prostate (99%), breast (89%) and colon (65%) [11]. The major contributing factor for this poor prognosis is that the majority of LC cases are diagnosed at an advanced/metastatic stage for which there is no curative therapy [12,13]. These statistics highlight the need to improve the detection rate for early-stage LC for which there are more effective treatment options available, and the therapeutic efficacy of the treatment options for late-stage LC.   1.3.1 Lung cancer and aetiologies Cigarette smoking is the primary cause of LC as demonstrated through over 200 years of epidemiological, population and biological studies [14,15]. One of the main epidemiological evidences that supports the causal link between smoking and LC is that the countries with some of the highest cigarette consumption rates, like Hungary, tend to have higher LC incidence rates. Additionally, an epidemiological LC trend analysis conducted in United Kingdom has shown that the decline of cigarette consumption by men ages 35-59 in the 1950s was followed by a similar reduction in the LC mortality rate in the same cohort, with an approximate 30-year disease latency period [15]. These statistics, along with other in vitro and in vivo experiments, have established tobacco as the primary cause of LC. Related to the risks associated with smoking, non-smokers that are exposed to secondhand smoke are known to have a higher risk of developing LC. 8  Secondhand smoke is comprised of the mainstream smoke, which is exhaled by the smoker, and the sidestream smoke that is generated at the lighted end of the cigarette [16]. The molecular mechanism as to how cigarette smoking causes LC is only partially understood as the tobacco smoke contains more than 7000 chemicals and only 69 components are presently identified as carcinogens by the Food and Drug Administration of U.S. (FDA) [17]. Amongst the thousands of compounds, nicotine is the main player in initiating and sustaining a smoking addiction [18]. While nicotine plays an essential role in habit-forming, extensive research of its chemical and biological action has revealed that the nicotine itself is not carcinogenic [18]. It is the accompanying compounds of the smoke that play a major role in tumor development. In particular, the polycyclic aromatic hydrocarbon (PAH) and the tobacco-specific nitrosamine 4-(methylnitrosamino)-1-1(3-pyridyl)-1-butanone (NNK) have been linked to a critical role in LC development [18]. Their metabolites covalently attach to DNA and form DNA adducts [18,19,20]. When a cell fails to repair such abnormal alterations, the DNA becomes permanently altered and the mutations are passed on its daughter cells as well. Repeated exposures to these carcinogens increase the number of mutations, along with the likelihood that the mutations will occur in  a tumor-suppressive or tumor-promoting gene. Asides from cigarette smoking, there are many other internal and external aetiologies of LC. For example, genetic factors are known to affect one’s susceptibility to LC. Although rare, individuals with Li-Fraumeni syndrome have an inherited germ-line mutation in a tumor suppressor gene, Tumor Protein 53 (TP53) [1,21]. TP53 and its protein product, p53, act as potent tumor suppressors as they can inhibit the improper propagation or replication of cells with abnormal DNA or metabolic activities by arresting their cell cycles or triggering apoptosis. Suppression of p53 predisposes Li-Fraumeni patients to developing various types of cancers at a young age, which includes LC [1]. 9  1.3.2 Classifications of lung cancer  The two main types of LC are small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) and they account for 15% and 85% of all LC cases, respectively [21]. These two types have distinct biology and disease progression and are treated with different treatment regimes. Both types can be further sub-classified according to their radiological, histological and molecular features, which was most recently outlined by the 2015 World Health Organization classification scheme of lung tumors [22]. SCLC is an aggressive neuroendocrine tumor that tends to metastasize early and it is associated with the poorest prognosis amongst all LC types. SCLC has a distinct appearance and can be diagnosed using a hematoxylin and eosin-stained sections alone [23]. Some of the defining features of SCLC cells include having a small size with scant cytoplasm and a finely granular chromatin [24]. Interestingly, SCLC occurs almost exclusively in cigarette smokers and is frequently found centralized to the respiratory tracts of the lungs [25]. NSCLC has three major histological subtypes: LAC, squamous-cell carcinoma (SQCC) and large-cell lung cancer (LCC). LAC is the most common subtype, accounting for up to 40% of all LC, followed by SQCC (25-30%) and LCC (5-10%) [26]. While all LC types are predominantly found in smoker or ex-smoker patients, this risk association is considered to be relatively lower in LAC as it is the most common subtype found in never-smoker patients [27]. Nevertheless, the rate of LAC occurrences is still significantly higher in smokers than in the never-smoker population [28].  LAC cells appear glandular in cell type like their putative origin cells, type II pneumocytes or Club cells [29]. LAC can further be categorized into five distinct histological groups: lepidic, acinar, papillary, solid, and micropapillary [22,30]. Identification of any one of the five histological patterns can lead to the histological diagnosis of the lung tumor tissue as LAC [22,30]. 10  Additionally, another common trait of LAC is that the adenocarcinoma tumors are typically found distal to the airways, unlike SCLC [31]. SQCC cells, the second most commonly diagnosed NSCLC, appear squamous, or “flattened”, and high in keratin. SQCC is associated with smoking in a dose-dependent manner and is found centralized to the airways [32]. Finally, LCC is composed of non-small carcinoma cells that lack any characteristics belonging to either the adeno or the squamous group [22]. Aside from the four major histological subtypes discussed above – SCLC,  LAC, SQCC and LCC – the remaining portion of lung tumors without an identifiable feature form their own class, referred to as “Not Otherwise Specified” or NOS [22]. Tumor grading systems also differ between SCLC and NSCLC. SCLC can be staged using the binary “limited” or “extensive” annotation and the TNM staging system [33]. In the former staging system, “limited” stage describes SCLCs that are found in only on one side of the chest, while “extensive” SCLC tumors are found in both sides, with metastasis in the surrounding lymph nodes [33]. The TNM staging system is a tumor staging schematic that grades the severity of the disease based on the size of the tumor (T), involvement of the lymph nodes (N) and whether the tumor has metastasized to other sites or organs (M). Some argue about the redundancy of using the TNM staging system to grade SCLCS as most cases present as the most advanced stage in the TNM stages [33]. On the other hand, staging of NSCLC is heavily dependent on the TNM descriptors of cancer and it can be divided 6 different stages: IA, IB, IIA, IIB, IIIA, IIIB, IV. Each staging is associated with a drastically different prognosis, and a corresponding therapeutic recommendations [13,22].  1.3.3 Mutation spectrums of lung cancer Recent large-scale genomic sequencing projects have revealed that each of the major histological subtypes of LC also have a different spectrum of oncogenic mutations. In LAC, the 11  most common driver mutations are EGFR (15%) and KRAS (30%), both of which affect the RAS-MAPK pathways [34,35]. Other commonly occurring mutations include anaplastic lymphoma kinase (ALK) and hepatocyte growth factor receptors (MET) mutations, as shown on Figure 1.2 [35]. In SQCC, fibroblast growth factor receptor 1 (FGFR1) amplification and mutations in the phosphoinositide 3-kinase (PI3K)-protein kinase B (AKT) pathway are frequently observed, while the RAS-MAPK pathway is rarely affected [34,35]. As a matter of fact, mutations in the PI3K-AKT pathway affect up to 60% of all SQCC cases: Phosphatidylionositol-4,5-Biphosphate 3-Kinase Catalytic Subunit Alpha (PIK3CA) amplification (35%), Phosphatase And Tensin Homolog (PTEN) mutations (10%), and PIK3CA mutation (15%). Similarly, in SCLC, mutations in EGFR and RAS are rare. The most prevalent alteration, which is present in most cases, is the co-occurring bi-allelic inactivation of the tumor suppressing TP53 and Retinoblastoma Protein 1 (RB1) genes [36]. Together, these observations suggest that each subtype of LC differ from one another on both a histological and a genomic level.    12    Figure 1.2: Mutation spectrums of NSCLC. (© 2015 Pioneer Bioscience Publishing Company, by permission).  The top pie chart illustrates the distribution of the different NSCLC histological subtypes.  The bottom two pie charts reflect the frequencies of known driver mutations found in SQCCs (left) and LACs (right). Figure 1 from © Chan, B.A., & Hughes, B.G.M. (2015). Targeted therapy for non-small cell lung cancer: current standards and the promise of the future. Transl Lung Cancer Res, 4(1), 36-54. Page 37. By permission from publisher [35].    13  1.3.4 Current treatment strategies of lung cancer Encouraging cessation of cigarette consumption is one of the main priorities for health authorities in trying to lower LC incidences, as tobacco smoking accounts for up to >80% of all LC cases in the US [17]. Also, to improve the incidence rates of LC, researchers are currently investigating the possibility of employing low-dose computed tomography (LDCT) for public LC screening. Because most patients are asymptomatic in the earlier stages of LC, majority of the LC patients are diagnosed at late-stages for which there is no curative therapy currently available. LDCT trial program, such as in B.C, is trying to improve the early detection rate by annually screening high-risk cohorts. Eligible high-risk patients in the B.C. trial program are defined as follows: (1) Individuals who have smoked more than >30 pack years (10,950 packs total) and (2) ages 55-79. The underlying rationale for trying to improve the early detection rate for LC is that patients can potentially benefit from a higher 5-year overall survival rate of 50% if detect at Stage IA than if diagnosed at Stage IV, which has a rate of 2% whilst being the most commonly diagnosed stage [13]. Preliminary results from such trial programs currently show signs of improving early detection rates [37]. Prior to its adoption as a population-wide screening program, however, there is remaining research being conducted on the economic feasibility and the potential side effects of the program, as high-risk individuals receive an annual low-dose CT radiation [37].  Treatment strategies for LC vary widely depending on its subtype and how advanced the tumor is. For the early stages of NSCLC (IA to IIB), surgical resection with chemotherapy is often recommended. Currently, complete resection of a localized tumor is the only available curative treatment option for LC patients. For the later stages (IIIA to IV) at which the cancer has metastasized, surgical resection is not an available option. Advanced NSCLC cases are instead considered for combination therapy involving generic chemotherapy, radiation, targeted therapies or, in select cases, partial resection of the tumor [13]. 14  In particular, targeted therapies, such as EGFR tyrosine kinase inhibitors (TKIs), are often considered as the first-line treatment in select cases of advanced NSCLC, as they have shown to be more effective than the standard radio-chemotherapy. Targeted therapy was developed based on a principle called “Oncogene Addiction,” which was first coined by Weinstein [39]. This principle, which is true for many cancers, states that although cancer cells may have multiple tumor-promoting mutations, they are typically dependent on the activities of a single dominant oncogene and its pathway. When the activity of such oncogene is sufficiently inhibited through the use of TKIs, the cells not only lose their survival advantage but die as a consequence due to their dependency on its signalling activity [39]. This principle is well-illustrated in the case of EGFR-TKIs and EGFR-addicted/driven NSCLC cancers. When tumors with oncogenic mutation in EGFR are treated with EGFR-TKIs, such as Erlotinib or Gefitinib, approximately 81% of all cases maintain a disease-progression free status for at least three months [40]. These statistics are superior to any other modes of treatment, although targeted therapies are not available for all drivers of NSCLC such as in KRAS-driven cancers. Additionally, a recent large-scale phase III clinical trial testing the efficacy of pembrolizumab, an immune checkpoint inhibitor targeting programmed cell death protein 1 (PD-1), has demonstrated that pembrolizumab can improve the overall survival and the progression-free survival rate in late-stage NSCLC patients compared to standard chemotherapy treatments, which are typically composed of pemetrexed and platinum-based drugs [38].  SCLC, which is characterized by a high proliferation rate, is treated with a combination of chemotherapy and radiotherapy as first-line treatment. Compared to NSCLC, SCLC is more responsive to chemo-radiotherapy because both modalities works more effectively against faster replicating cells [41]. However, SCLC rapidly develops resistance to these treatments exhibiting rapid disease progression [42]. 15  1.3.5 Current challenges in advanced NSCLC therapeutics As previously mentioned, the treatment options for advanced NSCLC are limited. EGFR TKI is often considered for first-line therapy for advanced NSCLC cases as they elicit a high initial response rate in patients. However, despite its initial potency, EGFR TKI has had a minimal impact on the overall survival rate for NSCLC patients because cancers inevitably develop resistance to these drugs [34,43,44]. The list of resistance mechanisms found in lung tumors is numerous and thus requires an extensive review to discuss all mechanisms known to date [34,43,44]. However, they can be grouped into three main groups: target modification (~60%), accessory pathway activation (~20%), and histologic transformation (4%) [44]. Target modification describes a class of resistance in which the target protein of TKI, such as EGFR, has a mutation in the receptor that interferes with its interaction with the drug. Consequently, resistant tumors still maintain EGFR signalling even in the presence of EGFR inhibitors due to the reduced ability for drugs to bind to their targets. Tumors can harbor such mutations either prior treatment (primary resistance) or develop them after treatment (secondary resistance) [44]. Second class of resistance mechanism – accessory pathway activation – involves either re-activating the inhibited pathway through activating a downstream node or activating an alternative oncogenic signalling pathway. For instance, few common resistance mechanisms found in EGFR TKI-treated NSCLCs are amplification of MET, human epidermal growth factor receptor 2 (HER2) upregulation and activation of downstream molecules of EGFR, such as v-Raf murine sarcoma viral oncogene homolog B (BRAF) and KRAS activation [44].  Lastly, tumors can develop resistance through histologic transformation. TKI-treated NSCLC cells can undergo either ab epithelial-to-mesenchymal transition (EMT) or transform into SCLC. How these transformations are initiated or impart resistance to tumors is still unclear and this mode of resistance is the least understood amongst the three aforementioned groups [44]. 16  To combat such resistance mechanisms, researchers have developed multiple generations of EGFR TKIs in response. NSCLC tumors treated with the first generation of EGFR-TKIs often acquire a point mutation at the amino acid position 790, which changes a single threonine amino acid to methionine (EGFR-T790M).  In order to overcome this target alteration, second and third generations of TKIs that circumvent the T790M mutation were developed. However, NSCLCs have shown the ability to adapt to all three generations of TKIs, with the fourth generation currently being tested. Such results indicate that inhibiting oncogenic signalling at a single node is not robust enough to overcome the plasticity of cancer.  Due to an increased awareness of how well cancer cells adapt to inhibition-based therapies, many researchers are currently searching for new therapeutic strategies beyond the current single inhibitor-based approach.  In addition to the ongoing challenges of TKIs, KRAS, which is the most common driver in NSCLCs, has been described as an “undruggable” target due to the many failed attempts in trying to develop a RAS-specific agent [45,46]. The constraints of developing a RAS agent can be attributed to the unique biological and chemical properties of RAS. RAS is difficult to target as it lacks multiple deep binding pockets that can provide points of contacts for drugs. Resultantly, RAS-inhibitors cannot achieve sufficient binding affinity to outcompete its other binding partner guanosine triphosphate (GTPs) [45,46]. Despite these challenges, researchers have successfully developed KRAS-G12C targeting agents, ARS-853 and ARS-1620. Their clinical efficacy and potential resistance mechanisms is an area of ongoing research [47, 48].  1.3.6 Novel therapeutic approach to LC: Combination therapy and synthetic lethality One strategy that is currently being investigated for advanced NSCLC is combination therapy. As the name suggests, combination therapy involves treating tumors with more than one targeted agent that either act on the same target (i.e. 1st and 3rd generation of EGFR TKIs), different 17  nodes of the same pathway (i.e. EGFR TKIs with an inhibitor of a downstream signalling molecule) or in different pathways (i.e. EGFR TKIs with MET inhibitors) [34,44,49]. The main rationale for testing this strategy is that certain drug combinations can synergize and kill cancer cells more robustly than a single inhibitor. However, this method is complicated by the issues of balancing varying pharmacokinetics of different drugs and the potential synergistic side effects and toxic effects on the patients. Another potential therapeutic avenue that is currently being investigated is synthetic lethality. Synthetic lethality, first coined by Dobzhansky, describes a biological relationship where co-occurrence of two conditions (i.e. mutations or drug-induced inhibition) induces lethality while the presence of individual condition alone is tolerable [50]. Exploiting synthetic lethality in a cancer therapy setting is best exemplified by Olaparib, an FDA-approved poly (ADP-ribose) polymerase (PARP) inhibitor that is used to treat select cases of Breast cancer susceptibility genes (BRCA)-deficient ovarian and breast cancers [51]. BRCA1/2 is a tumor suppressor that mediates a high-fidelity DNA repair method called homologous recombination, which is one of the two main DNA repair systems that cells rely on. While BRCA1/2 or PARP loss alone can be tolerated by the cell, inhibiting PARP, a key mediator of an alternative DNA repair pathway, in BRCA-deficiency induces lethality [51]. This seminal finding prompted further investigation in other synthetic lethal relationships present in other cancer settings. Even in LC, many synthetic lethal interactions have been uncovered [50]. For instance, it has been recently shown through in vivo models that LAC cells that express low levels of 3’-phosphoadenosine 5’-phosphosulfate synthase 1 (PAPSS1) are more sensitive to Cisplatin, which is a platinum-based standard chemotherapy used to treat advanced LC [52,53].   18  1.4 Oncogenic signalling in lung cancer Oncogenic mutations can exploit normal cellular pathways by constitutively activating key signalling proteins. Mutations activates key signalling proteins through several different mechanisms. Few mechanisms of pathway activation include stabilizing the active state of the signalling molecule, altering protein conformation to a kinetically favourable state or simply increasing the number of signalling molecules available through gene amplification. For instance, the most common oncogenic mutation of EGFR found in NSCLCs, the L858R (exon 21) point mutant,  replaces the residue Lys-858, which is located on the activation loop, with arginine, which exposes the activation loop from its hydrophobic pocket and thereby permanently stabilizing the active form of the receptor. On the other hand, the in-frame deletion of exon 19 has been hypothesized to activate EGFR by re-arranging the conformation of its ATP binding pocket in a way that increases its kinase activity [43,54,55,56]. Also, EGFR is amplified in many tumors, which increases the number of receptors available on the cell surface. Increased number of receptors increases the overall EGFR signalling activity, not only through the simple increase in the number of signalling receptors but also by increasing the degree of EGFR activation through receptor homodimerization, which is more likely to occur when there are more receptors that are in close proximity with each other [57].   KRAS-driven NSCLCs harbor a point mutation at position 12 (93.3%), 13(6.6%) or the rare 61. Point mutations at each of these positions can have many amino acid variations; guanine at position 12 can be altered to alanine (10.0%), cytosine (43.4%), aspartic acid (10.0%), phenylalanine (3.3%), arginine (1.7%), serine (1.7%), or valine (23.3%) [48]. All of these mutations lead to the constitutive activation of the RAS protein by interfering with its innate ability to hydrolyze GTP and return to its inactive state [58].    19  1.4.1 PI3K-AKT and RAS-ERK pathway activity in mutant EGFR- and mutant KRAS-driven cancer Mutant EGFR and mutant KRAS, which sits downstream from EGFR, drive LC development by activating same sets of signalling pathways, such as the RAS-PI3K-AKT and the RAS-MAPK axis. The PI3K-AKT pathway is involved in regulating many cellular processes. Few notable downstream effectors include BCL-2-associated death promoter (BAD), glycogen synthase kinase 3 beta (GSK3β) and mammalian target of rapamycin (mTOR), which affect apoptosis, cell-cycle and cell-survival pathways, respectively [1]. Of equal importance, the RAS-MAPK axis is a potent mitogenic pathway that is also frequently upregulated in many types of cancers. The RAS-MAPK pathway is organized into three signalling tiers, with the MAPKKK tier (MAPKinase kinase kinase) at the top, MAPKK (MAPKinase kinase) and MAPKs at the bottom (Figure 1.3) [59]. Activated Ras, which is tethered to the cell membrane, can activate three different branches of MAPKKK-MAPKK-MAPK axes, which are the ERK, c-Jun N-terminal kinase (JNK) and the p38 branch (Figure 1.4). ERK pathway is known to respond specifically to growth factors, whereas JNK and p38 pathways primarily respond to RAS activation through cell stress, such as heat, osmotic and interleukin-1 (IL-1) [60]. In the ERK branch, RAS activates Rapidly Accelerated Fibrosarcoma protein (RAFs), which is a Ser/Thr-specific kinase. Out of the three known isoforms of RAFs, – A-RAF, BRAF and C-RAF – it has been reported that BRAF has the highest kinase activity [61]. Additionally, BRAF is most frequently mutated in cancers, with rates up to 1-5% and 50% in NSCLC and melanoma, respectively [34, 62]. Subsequently, activated RAFs phosphorylate MEK1/2 (MAPKKs), which are dual-specificity kinases. Although rare, in about 1% of all NSCLC cases, MEK1 mutants act as the driver mutation [34]. The dual-specific kinase nature of MEK1/2 is critical in its subsequent activation of ERK because ERK uniquely requires the phosphorylation of both a threonine and a tyrosine residue to be activated [61]. ERK1 and 2, which are the final 20  kinases of RAS-ERK axis (MAPKs), are splice variants of the same gene. Together, they have many downstream targets in both the cytoplasmic and the nuclear compartment of the cell. Upon phosphorylation, ERK1/2 can translocate to the nucleus and phosphorylate many transcription factors, such as MYC, E26 transformation-specific factors (ETS), and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-қβ) to name a few [1, 61]. The exact mechanism of how each ERK substrate contributes to the orchestrated ERK response is unclear but a few notable effects of ERK1/2 are increased protein synthesis and cell proliferation.   JNK and P38 pathways are also organized by the identical MAPKKK-MAPKK-MAPK structure as depicted in Figure 1.4 [60]. ERK, JNK and P38 each have an independent set of substrates that they activate to elicit different cellular responses, although some targets are shared amongst each other [63]. Aberrant activities in the PI3K-AKT and RAS-MAPK pathways can enhance a cell’s transformative and proliferative capabilities and are frequently exploited by cancer cells to enhance growth and orchestrate the complex process of metastasis.      21    Figure 1.3: Organization of the EGFR-RAF-MEK-ERK signalling axis.  The EGFR-RAF-MEK-ERK signalling axis is organized into a three-tiered cytoplasmic cascade. MAPK is the last kinase of the cascade that is responsible for phosphorylating the effector proteins that regulate cell functions.  The MAPKKK-MAPKK-MAPK organization is present in all signalling pathways involving MAPKs other than ERK [59].    22    Figure 1.4: Three-tiered signalling hierarchy of the MAPK (ERK, JNK, p38) pathways. The three main branches of RAS pathway target three different MAPKs – ERK, JNK and p38 – which all have been implicated to play a role in cancer development [64]. All three branches adhere to the MAPKKK-MAPKK-MAPK organization. The figure represents a simplified and segmented view of the interactions. There is a greater number of interacting partners and crosstalk between pathways than represented in this figure. *activated by proteins other than RAS.         23  1.5 Thesis theme  1.5.1 Background: Does functional redundancy underlie the mutual exclusivity between KRAS and EGFR mutations in NSCLC?  Despite their prevalence in NSCLC, mutations in EGFR and KRAS are found in a mutually exclusive manner in over 600 LAC cases that are listed on TCGA [65]. The most parsimonious explanation for this mutual exclusivity is that there is no selective pressure for altering two genes that overlap in the signalling spectrum, as their co-activation will have a redundant effect on the cell. However, our group has previously challenged this notion by demonstrating that the forced co-expression of mutant KRAS (mKRAS) and mutant EGFR (mEGFR) has a surprising deleterious effect, suggesting that a synthetic lethal relationship between EGFR and KRAS can, in part, explain the underlying cause of the mutual exclusivity of these two oncogenes.   1.5.1.1 Tet-ON system to establish tetracycline-regulated co-expression status in NSCLC cells  The effects of the synthetic lethal interaction of mKRAS and mEGFR within the context of LAC was tested using a Tetracyclin (Tet)-ON system. Our group established two different Doxycycline (Dox) inducible co-expression LAC cell line models by relying on the Tet-ON technology (Figure 1.5). In order to engineer these Dox-induced co-expression cells, Drs. Unni and Lockwood first created DNA vectors that express their gene of interest – GFP, KRASG12V or EGFRL858R – from a Tet regulated promoter (TRE), and a transcription factor called the reverse Tet-controlled transactivator (rtTA), which is regulated by a ubiquitin promoter that allows expression in most eukaryotic cells [65,66,67]. When transduced in a cell, rtTA requires the Dox to bind to TRE and promote the transcription of its downstream gene (Figure 1.5). Effectively, a 24  cell is induced with a vector containing such elements, it expresses the gene of interest only when Dox is present.    Drs. Uni and Lockwood established three Tet-inducible vectors expressing KRASG12V, EGFRL858R or GFP. The Tet-O-EGFRL858R and Tet-O-GFP control vectors were separately transduced into LAC cells that had a KRAS mutation (H358). Tet-O-KRASG12V and Tet-O-GFP were transduced into LAC cells that had an EGFR mutation (PC9). This process established two Tet-regulated co-expression, PC9EGFR-DEL-Tet-O-KRASG12V and H358KRASG12C-Tet-O-EGFRL858R, along with their respective Tet-O-GFP controls. In experiments, the less toxic Dox antibiotics, which is a part of the family of Tet, were used to minimize the effects of toxicity [68].               25   Figure 1.5: Illustration of the Tet-ON system. Representation of how each component of the Tet-On system interact with each other to allow Dox-mediated regulation of gene expression. In steps: (1) Dox binds to the rtTA transcription factor. (2) rtTA alone cannot bind to the TRE gene promoter, and the genes are not transcribed. (3) Dox-bound rtTA can bind to TRE and allow for temporal regulation of the gene of interest.    26  1.5.2 Co-expression of KRAS and EGFR induces lethality in LAC cells  Using the aforementioned Tet-inducible in vitro models, our group demonstrated that the Dox-induced co-expression of mutant KRAS and EGFR is toxic to LAC cells (Figure 1.6). Co-expression also led to an increase in apoptotic and autophagy marker as well as the phospho-levels of the four MAPKs downstream of RAS, namely AKT, ERK1/2, JNK1/2/3 and p38α,β,γ,δ. This seemingly paradoxical phenomena of having decreased cell viability yet increased signalling activities for cell proliferation and growth suggested that it may be possible to hyperactivate EGFR and RAS signalling to an intolerable level in LAC cells. Consistent with this hypothesis, other groups have observed similar phenomena, such as oncogene-induced senescence (OIS) or hyperactivation of RAS triggering cell death has been reported by multiple groups [69, 70, 71]. Also, it has been reported that the co-expression of BRAFV600E with either KRASG12D or NRASQ61R, both of which are mutually exclusive with mutation in BRAF, has a deleterious effect on cancer cells by triggering senescence [72, 73].  Collectively, these results prompt a further investigation into the following areas: first, a detailed mechanism of how the co-expression leads to lethality, and second, how this information can be exploited to develop a novel therapeutic approach for treating KRAS- or EGFR-driven advanced NSCLC.   27    Figure 1.6: Growth assay of Tet-inducible co-expression LAC cells. PC9EGFR-DEL-Tet-O-KRASG12V and H358KRASG12C-Tet-O-EGFRL858R were seeded in 6-well plates along with their respective Tet-O-GFP controls and with or without 100ng/mL of Dox. The left graph shows cell viability measurements of Dox-treated cells using Alamar Blue, a cell viability dye, and after normalizing the values to no Dox controls. The right graph shows the cell number count at endpoint (Day 7). Unni, A.M., et al. Evidence that synthetic lethality underlies the mutual exclusivity of oncogenic KRAS and EGFR mutations in lung adenocarcinoma. Elife, 2015. 4:e06907 [65].   28  1.5.3 ERK1/2 as the main mediator of Tet-O-RAS mediated toxicity in mKRAS/mEGFR LAC cells   Initially, our group questioned whether the lethality shown in Figure 1.6 is specific to the specific types of the EGFR and KRAS mutants used, namely the EGFRL858R and DEL19 and KRASG12V and G12C, or if it is purely a consequence of increasing RAS signalling. To test this, we established two more Tet-O-KRASG12V cell lines in a different EGFR mutant line, H1975-EGFRL858R/T790M, and the previously described KRAS mutant line, H358-KRASG12C, which would express two types of KRAS mutants (G12C and G12V). Dox-induced KRASG12V expression in all three lines generated a similar degree of toxicities, which suggested that the lethality is due to the increase in RAS signalling (Figure 1.7).  To further elucidate the detailed mechanism of hyper-RAS lethality, a phospho-kinase antibody array was performed to assess the degree of kinase activations after Dox treatments. The assay revealed that the P-ERK1/2 and P-AKT1/2/3 levels were significantly elevated after 24 hours of Dox in PC9EGFR-DEL-Tet-O-KRASG12V, implicating ERK and AKT as the significant mediators of the lethality (Figure 1.8). To test whether these identified kinases do play a critical role in lethality, our group performed rescue assays in the aforementioned Tet-O-KRASG12V lines using Trametinib (MEK1/2 inhibitor), SCH772984 (ERK1/2 inhibitor) and Buparlisib (PI3K inhibitor). If ERK and AKT play a critical role in mediating the lethality, attenuating ERK or PI3K activities should moderate the toxicity upon Dox treatment. When we performed the assays, Trametinib and SCH772984 were able to elicit either full or partial rescues in the RAS-induced lines, while Buparlisib was unable to rescue the cells. In addition, we established knockdown lines using short-hairpin RNAs (shRNAs) targeting either ERK1 or ERK2 and observed that the suppression of total ERK1/2 levels can also moderate the effects of hyper-RAS toxicity. Together, these evidences 29  strongly implicate P-ERK1/2, but not AKT, as the main instigator of the Tet-O-KRAS toxicity in LAC cells. Collectively, these findings demonstrate that RAS-signalling follow a well-known principle called the “Goldilocks principle”, which states that a certain attribute of a given sample can be either “too much” or “too little” [74]. Within the context of oncogenic signalling, the toxic consequences of “too little” signalling activity of the oncogenic pathway have been repeatedly demonstrated through the examples of targeted therapies. Here, our findings illustrate the counterpart of that principle, the “too much”; excessive oncogenic signalling activities can also reduce the fitness of the cells. This also suggests that cancer cells with elevated oncogenic signalling may be closer to the upper threshold – the “too much” – than normal cells (Figure 1.9).          30   Figure 1.7: Pharmacological inhibition of ERK1/2 activation rescues co-expression lethality in PC9s, H358s and H1975s.  (A) Tet-inducible PC9[EGFRE746_A752del], H358[KRASG12C] and H1975[EGFRL858R/T790M] that either expresses KRASG12V or GFP were treated with 100ng/mL of Dox and varying doses of Trametinib, a MEK1/2 inhibitor. Alamar Blue values were measured after 7 days of treatment and normalized to cells that did not receive Dox or Trametinib. The error bars represent the standard deviations determined from three replicates. Trametinib either partially or fully rescues cells from hyper-RAS induced toxicity. P-ERK (p42 and p44) and total ERK levels were determined after 24 hours of Trametinib treatments (bottom panels).         Performed by: Dr. Arun Unni     31    Figure 1.8: Phospho-kinase antibody array in PC9EGFR-DEL-Tet-O-KRASG12V after 24 hours of Dox treatment. Tet-O-KRAS PC9s were treated with 100ng/mL of Dox for 24 hours along with untreated control. Lysates containing 200ug of protein were harvested and incubated on membranes from the human phospho-kinase array kit (ARY003B, R&D systems). The signals from the 24-hour treated sample was normalized to the non-treated control.  Performed by: Dr. Arun Unni    32                           Figure 1.9: Hypothetical model of ERK levels in KRAS/EGFR mutant versus wildtype cells. This figure illustrates our hypothetical model of ERK levels in cells according to their mutational status. Through various inhibitors of the MAPK pathway, ERK activity can be suppressed below its lower signalling threshold to induce cell death. Similarly, our evidences suggest that MAPK signalling can be stimulated beyond its hypothetical upper signalling threshold. Mutant KRAS/EGFR LAC cells have higher basal ERK levels and thus are closer to the upper signalling threshold.    33  1.5.4 Therapeutic application: EGFR- and KRAS-driven LACs are reliant on the negative regulators of the RAS-ERK axis  Next, we questioned how mKRAS and mEGFR LACs that are addicted to RAS signalling are able to maintain a high ERK level without exceeding the upper threshold. We hypothesized that these tumor cells may be utilizing the negative regulators of ERK pathway to modulate ERK activity in order to prevent hyperactivation of ERK past a tolerable threshold.   1.5.4.1 Negative regulators of ERK in the RAS-ERK axis  There are two main modes of negatively regulating ERK: phosphorylation-mediated downregulation of ERK and phosphatases of ERK. In the first mode, ERK phosphorylates some of its upstream partners to suppress their activities, such as EGFR, son of sevenless 1 (SOS1), RAF-1, and MEK1. For instance, ERK phosphorylation of the Thr 669 residue of EGFR is known to attenuate the degree of tyrosine phosphorylation of the receptor [75, 76]. Resultantly, less EGFR activation would lead to less phosphorylation of ERK. Collectively, these self-regulatory activities by ERK via phosphorylation at multiple points exert an acute and immediate down-regulation of ERK. On the other hand, ERK phosphatases produce a less acute response compared to the first mode of ERK regulation. This is due to the fact that there is a temporal delay between the initial increase in P-ERK levels and the eventual increase in the phosphatase activity as the phosphatases have to be transcribed and translated first. Therefore, the second mode of ERK regulation is thought to be responsible in moderating the steady-state or the basal levels of ERK output [77]. As previously mentioned, less is known about the functions and the interactions of phosphatases compared to kinases. Amongst the phosphatases, MAPK phosphatases (MKPs), which is a subclass of DUSPs, have been relatively well-described in their roles in how they regulate MAPKs.  There are ten MKPs that can be categorized into three main subclasses [77, 78] (Table 1.1). First class of MKPs is composed of MAPK-specific DUSPs that localize to the 34  nucleus. The second and third classes both localize to the cytoplasm but are distinguished by their substrates: ERK-specific DUSPs and JNK/p38-speicifc DUSPs [78].  One species of MKPs that is particularly well-described is DUSP6. DUSP6, which is transcriptionally regulated by ERK, is a potent repressor of P-ERK1/2. DUSP6 has the capacity to normalize P-ERK1/2 levels within 2 hours of EGF stimulation (100ng/mL) in NSCLC cells that have been transduced to overexpress DUSP6 [79]. Another group of well-described transcriptionally-regulated phosphatases of the MAPK pathway are the Sproutys 1-4 (SPRYs) [77]. They were originally identified as a transcriptionally-inducible inhibitor of receptor tyrosine kinase (RTK) signalling in the developing Drosophila melanogaster. Transcription of SPRYs is transcriptionally controlled by ERK1/2 and SPRYs have been reported to negatively regulate ERK1/2 by dephosphorylating several upstream activators of ERK1/2 [77]. Lastly, Sprouty-related proteins (SPREDs) 1-3 constitute a group of phosphatases that contain a Sprouty-domain, which is a conserved, cysteine-rich domain that are also found in SPRY proteins. In particular, SPRED1 and SPRED2 have been reported to negatively regulate the downstream mediators of RAS to prevent MAPK activation, although their mechanisms are not fully understood [77, 80, 81].        35  Table 1.1 List of dual-specificity MAPK phosphatases and their classifications and localizations    Class Localization Phosphatase Known substrates (ERK/P38/JNK) 1  Nucleus DUSP1/MKP-1 ERK/P38/JNK   DUSP2 ERK/P38/JNK   DUSP4/MKP-2 ERK/P38/JNK   DUSP5 ERK 2 Cytoplasm DUSP6/MKP-3 ERK   DUSP7/MKP-X ERK   DUSP9/MKP-4 ERK 3 Cytoplasm DUSP8 P38/JNK   DUSP10/MKP-5 P38/JNK   DUSP16/MKP-7 P38/JNK    36  1.5.5 DUSP6 as the main negative regulator in EGFR- and KRAS-driven LACs    To test whether EGFR- and KRAS-driven LACs are reliant on any negative regulators, our lab surveyed the gene expression levels of different transcriptionally-regulated phosphatases of the MAPKs, DUSPs, SPRYs, and SPREDs, as these phosphatases have been implicated to influence the steady-state output of ERK. We analyzed the TCGA RNA-seq data of LAC tumors with (n=108) and without (n=123) an KRAS or EGFR mutation and compared the levels of the phosphatases that have been discussed in section 1.5.4.1. We observed that DUSP6 was the only upregulated phosphatase in mKRAS/mEGFR LAC tumors compared to the KRAS/EGFR wildtype (WT) tumors (Figure 1.9). This strongly suggests that mKRAS/mEGFR LAC cells may be dependent on the robust activity of DUSP6 to modulate ERK levels within tolerable limits.   37   Figure 1.10: Heat map of TCGA RNA-seq data of mKRAS/mEGFR and KRAS/EGFR WT LAC tumors. Upon gene expression analysis of transcriptionally regulated negative feedback regulators of the MAPK pathway, DUSP6 is the only significantly upregulated gene in the mKRAS/EGFR tumors relative to the tumors without such mutations. This analysis was Bonferoni corrected with p<0.01 and we performed two-tailed t-test with Welch’s correction.  Performed by Dr. William Lockwood.    -3.00 3.000.00KRAS/EGFRMUTWTKRASMUTWTEGFRMUTWTKRAS/EGFRKRASEGFRDUSP6SPRED2SPRY4SPRY1SPRED1SPRY2DUSP4DUSP5SPRED3DUSP2DUSP1SPRY3DUSP3id-3.00 3.000.00KRAS/EGFRMUTWTKRASMUTWTEGFRMUTWTKRAS/EGFRKRASEGFRDUSP6SPRED2SPRY4SPRY1SPRED1SPRY2DUSP4DUSP5SPRED3DUSP2DUSP1SPRY3DUSP3id-3.00 3.000.00KRAS/EGFRMUTWTKRASMUTWTEGFRMUTWTKRAS/EGFRKRASEGFRSPRED24SPRY1SPRED1SPRY2DUSP45ED3DUSP2DUSP1SPRY3DUSP3id-3.00 3.000.00KRAS/EGFRMUTWTKRASMUTWTEGFRMUTWTKRAS/EGFRKRASEGFRDUSP6SPRED2SPRY4SPRY1SPRED1SPRY2DUSP4DUSP5SPRED3DUSP2DUSP1SPRY3DUSP3idRelative Expression -5 0 5 10 15DUSP3SPRY3DUSP1DUSP2SPDUSP5DUSP4SPRY2SPRED1SPRY1SPRY4SPDUSP6-log2(p-value)mKRAS/EGFR LAC tumors (n= 107) KRAS/EGFR WT LAC tumors (n= 132) 38  1.5.6 Rationale Although we have seen that hyperactivation of ERK induces toxicity in mKRAS/mEGFR LAC cells, we questioned how our findings can be translated for therapeutic purposes. Based on our preliminary findings (Figure 1.9), I hypothesize that DUSP6, which is over-expressed in LAC tumors bearing mKRAS or mEGFR, is a pharmacologically viable target that can sufficiently increase P-ERK levels to induce toxicity upon its inhibition.  To investigate this hypothesis, the aims are as follows:  Objective 1: Establish DUSP6 as a viable therapeutic target in mKRAS or mEGFR LAC tumors by demonstrating that DUSP6 inhibition induces lethality in mKRAS or mEGFR LAC cells but not in KRAS/EGFR WT cells Aim 1: Demonstrate that both the knockdown of DUSP6 and inhibition of its catalytic activity through the use of a DUSP6 small molecule inhibitor generates toxicity only in NSCLC cells with a mutation in KRAS or EGFR.  Objective 2: Validate that the mechanism of lethality induced by DUSP6 inhibition is mediated through the hyperactivation of ERK, as also seen in our Tet-O-KRAS LAC cell line models  Aim 2: Rescue mKRAS or mEGFR LAC cells from DUSP6 inhibition associated toxicity by RNA-interference or pharmacological inhibition of ERK1/2.     39  Chapter 2: Materials and Methods 2   2.1 Cell lines and culture conditions All cell lines used here, PC9 (PC-9), H358 (NCI-H358), H1975 (NCI-H1975), H1648 (NCI-H1648), A549 and HCC95 cells were initially obtained from either American Type Tissue Culture (ATCC) or were gifted from Dr. Adi Gazdar. These cells were grown in RPMI-1640 medium (Thermo Fisher) supplemented with 10% FBS (Sigma), 1% Glutamax (Thermo Fisher) and Pen/Strep (Thermo Fisher). Cells were cultured at 37°; air; 95%; CO2, 5%.  2.2 Plasmid and generation of stable cell lines Knockdown lines were established using pLKO.1-based lentiviral vector that contains shRNA targeting ERK1 or ERK2. A shRNA scramble sequence was used as control. These shRNA constructs were kindly provided by J. Blenis, Weill Cornell Medicine. Lentivirus was generated using 293T cells (ATCC), psPAX2 #12260 (Addgene, Cambridge, MA) and pMD2.G (Addgene plasmid#12259). After transduction, stable cells were selected with puromycin.   2.3 Western blot analysis of protein levels Cells were washed using cold PBS (Thermo Fisher) and lysed using RIPA buffer (Thermo Fisher) containing Halt protease and phosphatase inhibitor cocktail (Thermo Fisher) on ice. The samples were frozen, sonicated and cleared by centrifugation at 15,000 X g for 15 min. The protein concentrations were determined using a Pierce BCA protein assay kit (Thermo Fisher). Samples were denatured by boiling in 4x Laemmli sample buffer (Bio Rad) with 1:10 2-Mercaptoethanol (Thermo Fisher) and 20-25ug of samples were ran using Novex 4-12% Bis Tris Gels (NuPage) at 200V for 50min in MOPS buffer. The proteins were transferred to Immobilon-P PVDF membranes (Millipore) and blocked in TBS-T (0.1% Tween-20)/5% BSA (Sigma).  40   Primary antibodies were suspended at manufacturer’s recommended concentrations in 5% BSA in TBS-T. Blocked membranes were incubated with primary antibodies overnight at 4°C then incubated with appropriate HRP-conjugated secondary antisera (Santa Cruz Biotechnology). The membranes were incubated with ECL (Thermo Fisher) for five minutes and detected. Following antibodies were purchased and used for the experiments: P-p-38 (4511), p38 (8690), P-p44/p42 (ERK1/2) (9101), p44/p42 (ERK1/2) (4695), P-SAPK/JNK (4668), SAPK/JNK (9252), P-EGFR (3777, 2234), EGFR (2232), KRAS (8955), PARP (9542), cleaved-PARP (5625), and β-Actin (3700, 4970). Additionally, antibodies against DUSP1 (ab1351, abcam) and DUSP6 (ab76310, abcam and SC-377070, SC-137426, Santa Cruz) were used.  2.4 siRNA transfections For the double knockdown, time-course experiments, cells were seeded in 6-well plates in following numbers per well: PC9 (50,000), H1975 (75,000), HCC95 (75,000), A549 (75,000). Cells were adhered overnight then transfected with ON-TARGETplus siRNA pools (Dharmacon) against the following targets: EGFR (L-003114-00-0010), KIF11 (L-003317-00-0010), KRAS (L-005069-00-0010), DUSP6 (L-003964-00-0010), and non-targeting control (D-001810-10-20). In addition, to test for DUSP6 pool specificity for its target, individual siRNAs of the pool were purchased and tested (J-003964-06-0005, J-003964-07-0005, J-003964-08-0005 and J-003964-09-0005). All the transfections were performed under identical conditions for consistency. The siRNAs were constituted at 20uM using 1xsiRNA buffer (Dharmacon) and 10uL of the 20uM siRNA was added in 190uL of OptiMEM (Life Technologies) and 5uL of Dharmafect (Dharmacon) was added in 196uL of OptiMEM and each mixture was incubated for 5 minutes in room temperature. After 5 minutes, the siRNA and Dharmafect suspensions were mixed and 41  incubated for 20 minutes prior to transfection. Total of 400uL of the mixture was added to each well containing 1.6mL of RPMI media to make 2mL of the final transfection media. The cells were incubated in this final transfection media for 24 hours before replacing it with 2mL of fresh RPMI media. For the double knockdown experiments, transfections were conducted on Day 0 and again on Day 3, with media change on Day 1 and Day 4. On Day 5, the number of viable cells were measured using AlamarBlue. For end-point, double-knockdown experiments displayed in Figure 3.2, one-tailed, one-sample t-test using the one-sample t-test function on Prism7 (GraphPad) was conducted to determine the statistical significance of the results to the following powers: *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.  2.5 Viability measurement assays For the 6-well experiments, including the siRNA and the drug treatment experiments, the siRNA or the drug media was removed from the wells. The media was replenished with RPMI media containing Alamar Blue (Thermo Fisher) so that each well contain exactly consistent media to Alamar Blue ratio (1 to 10). After incubating the Alamar blue at 37 degrees for the same amount of time across each replicate, quadruplicates of 100uL of Alamar media was sampled from each well and measured using a Cytation 3 Multi Modal Reader with Gen5 software (BioTek).   2.6 BCI dose-response curve  Dose-response curves for BCI was established by seeding quadruplicates into a 96-well plate and the cells were incubated for 72 hours in media with and without BCI, with DMSO concentration of 0.1%. Alamar Blue was pipetted into each well at endpoint maintaining the 1:10 ratio to the media volume in each well.  For the HCC95 +/- EGF experiments, the cells were cultured in +/- EGF Recombinant human protein solution (Life Technologies) 100ng/mL for 10 days in 10cm plates and were passaged 42  every 48 to 72 hours. Then, the cells were seeded in quadruplicates in 96-well plates and were adhered overnight, with the +/- EGF maintained. After 24hours, the cells were treated with 17 different concentrations of BCI, ranging from 0 to 8uM, with 0.5uM increment doses at 0.1% DMSO concentration. Additionally, 100uM of Etoposide (0.1% DMSO) was added as a positive control for cell death. After 72hours of treatment, Alamar Blue was added at the manufacturer’s recommended Alamar Blue to Media ratio (1:10) and the viabilities were measured. Graphpad prism software was used to create the dose response curves.    43  Chapter 3: Result 3   3.1 Transient knockdown of DUSP6 increases P-ERK and reduces viability in KRAS- or EGFR-driven LAC  To test whether suppressing DUSP6 will sufficiently increase P-ERK to lethal levels in mKRAS/mEGFR NSCLC cells, expression of DUSP6 was transiently suppressed by transfecting a siRNA pool targeting DUSP6 in PC9-EGFRE746_A750DEL, H1975-EGFRL858R/T790M, A549-KRASG12S and HCC95s, which is WT for both mutations, as a control line. The experiments were conducted as outlined in the methods section 2.4. As shown in Figures 3.1 and 3.2, the inhibition of both DUSP6 and the driver oncogene reduced cell viability to similar levels in mKRAS/mEGFR cell lines. On the other hand, HCC95s were resistant to the effects of DUSP6 knockdown as the cell viabilities did not decrease upon DUSP6 knockdown. One unexpected observation from the western blot analysis of P-ERK at Day 5, which is 48 hours after the second transfection, showed that the P-ERK levels were lower in the DUSP6 knockdown lines than the levels found in Non-T (Figure 3.1). On the other hand, when we assessed the P-ERK levels at an earlier time point, Day 1, the depletion of DUSP6 resulted in increased levels of P-ERK (P-p44/p42) relative to the Non-T control of each cell line (Figure 3.3). This result partly unexpected as it was initially hypothesized that the P-ERK levels should be higher in DUSP6 knockdown cells compared to the Non-T cells at both Day 1 and Day 5 timepoints (Figure 3.1). Additionally, another interesting observation from the endpoint blots was that the knockdown of DUSP6 led to a decreased expression of the cell’s oncogene and vice versa.  44    Figure 3.1: Time-course viability assay following double knockdown of DUSP6 in PC9.  Double knockdown of DUSP6 was conducted in PC9 per protocol described in sections 2.4 and 2.5. The siRNAs that were used in these experiments are siDUSP6 pool, pooled siRNAs targeting the driver genes of each cell line (siEGFR), and a Non-T control as a transfection control. The siRNAs transfected on day 0 and again on day 3, and the viability measurements were taken on Day 1, 3 and 5 using Alamar Blue. Day 3 and Day 5 values were normalized to Day 1 Non-T values to represent the relative changes in the number of viable cells in each condition, over time. The graph bars are averages of the normalized biological triplicates and the error bars represent their SEM. Western blot analysis was performed at endpoint (right panel) to determine the levels of EGFR, P-p44/42 (P-ERK1/2), p44/42 (ERK1/2) and DUSP6. β-Actin was blotted as a loading control.   1 3 50246810Time (Days)Number of Viable Cells(Relative to Control at Day 1)siDUSP6siEGFRsiNon-TargetPC9[EGFRE746_A750del]β-ActinEGFR P-p44/42 [T202/Y204]p44/42 DUSP6Day 545    Figure 3.2: End-point (Day 5) viability assay following a double-knockdown of DUSP6 in four NSCLC cell lines. Double knockdown of DUSP6 was conducted in four NSCLC line: PC9-EGFRE746_A750DEL, H1975-EGFRL858R/T790M, A549-KRASG12S and HCC95. The siRNAs that were used in these experiments are siDUSP6, siRNAs targeting the driver genes of each cell line (siEGFR or siKRAS), siKIF11 in HCC95, and a Non-T pool. Experiment was conducted in the same manner as in Figure 3.1 except the viability was measured only on Day 5. The graph bars represent the average values across three biological replicates after scaling to Non-T, with +/- SEM of the averages. Western blots were performed at endpoint (Day 5) to assess for knockdown efficiencies of each transfection condition. The asterisks represent the level of statistical significance as determined after performing one-tailed, one-sample t-test comparing each siRNA condition to the Non-T control: *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001     siDUSP6siKRAS NT0.00.51.0Relative Number of Viabile Cells** **siDUSP6siKIF11 NT0.00.51.0Relative Number of Viabile Cells****siDUSP6siEGFR NT0.00.51.0Relative Number of Viabile Cells****siDUSP6siEGFR NT0.00.51.0Relative Number of Viabile Cells******β-ActinEGFR DUSP65 Daysβ-ActinEGFR DUSP65 Daysβ-ActinRASDUSP65 Daysβ-ActinEGFRDUSP65 DaysKIF11PC9[EGFRE746_A750del] H1975[EGFRL858R/T790M]A549[KRASG12S] HCC95[KRASWTEGFRWT]46    Figure 3.3: Western blot analysis of P-ERK levels 24 hours post-transfection. Three separate western blots were performed using lysates collected at 24 hours after transfection to detect the acute changes in the EGFR, P-p44/42, and p44-42 levels. Densitometry was performed on the three blots, normalized to the P-ERK values found in Non-T to determine the relative P-ERK values.     PC9[EGFRE746_A750del]β-ActinEGFR P-p44/42 [T202/Y204]p44/42 24 hrsβ-ActinEGFR P-p44/42 [T202/Y204]p44/42 24 hrsH1975[EGFRL858R/T790M]β-ActinKRASP-p44/42 [T202/Y204]p44/42 24 hrsA549[KRASG12S]β-ActinP-p44/42 [T202/Y204]p44/42 24 hrsHCC95[KRASWTEGFRWT]siDUSP6siKIF11 NT0.00.51.01.52.02.5RelativeP-ERKsiDUSP6siKRAS NT0.00.51.01.52.02.5Relative P-ERKsiDUSP6siEGFR NT0.00.51.01.52.02.5Relative P-ERKsiDUSP6siEGFR NT0.00.51.01.52.02.5Relative P-ERK47  3.1.1 Knockdown of DUSP6 increases apoptosis in mKRAS/mEGFR LAC cells Although the siRNA-mediated interference of DUSP6 RNA expression led to a reduction in the overall cell viability compared to Non-T, I questioned whether the knockdown of DUSP6 only decreases viability or if it also induces cell death and a reduction of cell numbers. Therefore, a western blot analysis was performed to measure the cleaved PARP levels, a marker for apoptosis, in H1975s (EGFR mutant) and HCC95 (KRAS/EGFR WT) after the double DUSP6 knockdown (Figure 3.4). By Day 5, knockdown of DUSP6 resulted in increased cleaved-PARP levels in both HCC95s and H1975s compared to their Non-T controls. This suggests that DUSP6 knockdown triggers apoptosis in mKRAS/mEGFR LAC cells.     48                   Figure 3.4: Western blot analysis of cleaved PARP at end-point (Day 5) of the double transfection assays. Western blot analysis of cleaved PARP, a marker for apoptosis, in H1975s and HCC95s after double knockdown. The experiment was conducted in the same manner as in Figures 3.1 and 3.2 and the lysates were collected on Day 5.     β Actincleaved	PARP	(Asp214)HCC95EGFRN/TH1975DUSP6EGFREGFRDUSP6KIF11N/TsiRNA:49  3.1.2 Confirmation of the specificity of the DUSP6 siRNA pool A pool of four individual siRNAs was used for the DUSP6 knockdown experiments and the specificity of the siRNA pool needed to be confirmed to ensure that the siDUSP6 lethality was not due to off-target effects.  In order to demonstrate the specificity of the siDUSP6 pool and its effects, a pLEX vector that overexpresses WT DUSP6 was stably transduced into PC9 cells. In the DUSP6 overexpressing PC9s, I attempted to perform rescue experiments using 2 independent siDUSP6 that each target a different region in the 3’ untranslated region (UTR) of the DUSP6 mRNA. Theoretically, 3’UTR targeting siRNA would only interfere with the endogenous transcripts of DUSP6 as vector-generated transcripts do not contain a 3’UTR region. Thus, transfecting the 3’UTR targeting siDUSP6 should decrease the viability of WT PC9s but not pLEX-DUSP6-PC9s as these cells have vector-transcribed DUSP6 that can rescue the cells. Unfortunately, none of the commercially available 3’UTR siDUSP6 that were tested specifically and sufficiently suppressed DUSP6 expression. Therefore, assessing the specificity of the DUSP6 siRNA pool was limited to testing the specificity of the individual siRNA of the pool separately and correlating their knockdown efficiency with the degree of reduction in cell viability. Amongst the four individual siRNAs tested, only one siRNA (siDUSP6 #8) elicited a dramatic reduction in cell viability (Figure 3.5A). As a matter of fact, siDUSP6#8 was able to decrease cell viability more drastically than the siDUSP6 pool. This observation is consistent with the western analysis of DUSP6 knockdown efficiency as the siDUSP6#8 depleted DUSP6 more efficiently than the pool (Figure 3.5B). The three other individual siRNAs that were tested suppressed DUSP6 less efficiently than the pool and siDUSP6#8. Interestingly, a modest suppression of DUSP6 by these three individual siRNAs (siDUSP6#6, siDUSP6#7 and siDUSP6#9) led to an increase in cell viability compared to the Non-T control. This suggested that 50  only a modest or partial knockdown of DUSP6 can increase P-ERK enough to confer an increase in proliferation without increasing its levels past a tolerable threshold to start generating toxicity.   51  A       B              Figure 3.5: Testing of the individual siRNAs of the siDUSP6 pool in PC9s. Double transfection 5-Day viability assay was conducted using 4 different individual siRNAs that were contained in the siDUSP6 pool to infer their specificity. The siDUSP6 Pool, siEGFR and Non-T were also included as controls. The values represent the averages of the biological triplicates that were normalized to Non-T, with +/-SEM. (A) AlamarBlue cell viability measurements on Day 5 (B) Western blot analysis of DUSP6 levels at endpoint to determine the knockdown efficiency of each siRNA condition.    siDUSP6 #6siDUSP6 #7siDUSP6 #8siDUSP6 #9siDUSP6 PoolsiEGFRNon-Target0.00.51.01.52.0Relative Number of Viabile Cells52  3.2 Identification of DUSP6 as a druggable target using BCI – a DUSP1,6 small molecule inhibitor I next questioned whether the inhibition of the DUSP6 enzymatic activity through using a DUSP6 inhibitor, BCI, can also induce hyper-ERK lethality. BCI was discovered through an in vivo zebrafish chemical screen as a drug that can bind to DUSP1 and DUSP6 at their allosteric sites and inhibit their catalytic activities [82, 83].  We treated 11 of in-house NSCLC cell lines – 8 with and 3 without known mutations in KRAS or EGFR – with varying doses of BCI for 72 hours (Figure 3.6). The 72-hour dose response experiments showed that NSCLC cell lines harboring a mutation in either KRAS or EGFR are more sensitive to BCI than the lines without. The sensitive lines exhibited IC50s ranging from 1-3uM, with only 10% of the cells surviving at doses at or higher than 3.2uM. One line, H1437, which is WT for both mutations, exhibited intermediate sensitivity with an IC50 > 4uM, while the two remaining WT lines were relatively resistant to BCI with IC50s ≥ 5uM. Concurrently, a western blot analysis of P-ERK levels in BCI-treated cells was performed to test whether the P-ERK levels correlate with sensitivity to BCI (i.e. greater P-ERK increases in cells that show greater sensitivity to BCI) (Figure 3.7A and B). Out of the four mKRAS/mEGFR lines that were tested, H358, PC9 and H1975 exhibited pronounced increases in P-ERK levels compared to the insensitive lines (Figure 3.7). The insensitive lines, HCC95s and H1648s, exhibited only minor fluctuations of P-ERK with varying BCI doses. The remaining sensitive line, A549s, exhibited modest increases of P-ERK and only at high doses of BCI. This is partially consistent with its BCI sensitivity, as A549 exhibited higher IC50 amongst the mKRAS/mEGFR cell lines. The cause for the modest sensitivity of the A549s is not fully understood and this suggests that there may be other secondary factors, other than the KRAS and EGFR mutation status, that can affect a cell’s sensitivity to DUSP6 inhibition. 53    Figure 3.6: BCI sensitivity curves in 11 NSCLC cells after 72 hours of treatment. A panel of 11 NSCLC cell lines were treated with BCI ranging from 0uM to 5uM with 0.1% DMSO concentration. The cells were treated for 72 hours and the values are representative of the average AlamarBlue values from triplicate experiments, +/- SEM. The cell lines showed three distinct trends: sensitive (red), intermediate (green) or insensitive (black). All the mKRAS and mEGFR cell lines from the panels were sensitive to BCI while the three KRAS/EGFR WT lines were in the intermediate or insensitive group.      0 2 4 60.00.51.0Log2 BCI Concentration [µM]Viability (relative to DMSO) A549 (KRASG12S)H460 (KRASQ61H)H23 (KRASG12C)H358 (KRASG12C)H2122 (KRASG12C)H1650 (EGFRE746_A750del)H2009 (KRASG12A)H2030 (KRASG12C)PC9 (EGFRE746_A750del)H1975 (EGFRL858R/T790M)H1437 (EGFR/KRASWT)H1648 (EGFR/KRASWT)HCC95 (EGFR/KRASWT)SensitiveInsensitiveIntermediateRelative Number of Viable CellsBCI Concentration (µM)54  A      B      Figure 3.7: Assessment of dose-dependent P-ERK changes after 6 hours of BCI treatment. (A) Western blot assessment of P-ERK and ERK levels after treating four BCI-sensitive lines (H358, PC9, A549, H1975; red) and 2 insensitive lines (HCC95, H1648; black) with varying doses BCI doses (0, 1, 3, 5, 8, 10uM) for 6 hours. The 0uM is the vehicle control with 0.1% DMSO. (B) Densitometry values obtained from panel 3.7A were plotted. The values are normalized to β-Actin and 0uM (0.1% DMSO) to show the relative P-ERK with increasing BCI dosages at a static time point of 6 hours. Red lines are BCI-sensitive cell lines while black lines are BCI-insensitive lines as determined by the previous dose-response curve assay shown in Figure 3.6.   0 1 3 5 8 100.250.5124816Relative P-ERK Levlels (Log2) H358[KRASG12C]PC9[EGFRE746_A750del]H1975[EGFRL858R/T790M]A549[KRASG12S]HCC95[KRASWTEGFRWT]H1648[KRASWTEGFRWT]BCI Concentration (uM)BCI (µM):p44/42P-p44/42 [T202/Y204]β Actin1 3 5 8 100H358[KRASG12C]H1975[EGFRL858R/T790M]A549[KRASG12S] HCC95[KRASWTEGFRWT]PC9[EGFRE746_A750del]1 3 5 8 100BCI (µM):p44/42P-p44/42 [T202/Y204]β Actin1 3 5 8 100 1 3 5 8 1001 3 5 8 100H1648[KRASWTEGFRWT]1 3 5 8 10055  3.2.1 BCI induced hyper-ERK toxicity is specific to DUSP6 inhibition, not DUSP1 As aforementioned, BCI is a drug that can interact with both DUSP1 and DUSP6. DUSP1 is a phosphatase that negatively regulates MAPKs, including ERKs, p38s and JNKs [84, 85].  Therefore, to test whether BCI lethality is mediated in part through the inhibition of DUSP1, a 5-Day double transfection assay using siDUSP1 and siDUSP6 was conducted using the H1975 cell line, which has an EGFR-L858R/T790M mutation (Figure 3.8A and B), and PC9s and H358s (data not shown). Knockdown of DUSP1 did not decrease cell viability in all three liens, suggesting that the BCI-mediated lethality can only be attributed to the inhibition of DUSP6 and not DUSP1.               56              A                                                B                 Figure 3.8: Double-transfection of siDUSP1 and siDUSP6 in H1975. Double transfection assay was performed in H1975s using siDUSP1, siDUSP6, siEGFR and Non-T. (A) Confirmation of the knockdown efficiencies through western blot analysis of lysates collected at Day5. (B) The cell viability at Day5 was measured by staining the cells with AlamarBlue. Each value reflects the averages of biological triplicates that were normalized to the Non-T, with +/- SEM.   siDUSP1siDUSP6siEGFRNon-Target0.00.51.0Relative Number of Viabile CellsDUSP6DUSP1β-ActinsiDUSP1siDUSP6siEGFRNon	TargetH1975 H1975 57  3.3 Toxicity from DUSP6 knockdown is partially rescued through suppression of ERK If siDUSP6 toxicity is truly mediated through the hyperactivation of ERK, simultaneous suppression of ERK and DUSP should balance their effects and mitigate the siDUSP6 toxicity. To test this, I established two ERK1 knockdown lines (shERK1 #4 and shERK1 #5), two ERK2 knockdown lines (shERK2 #6 and #7) and a Scramble (SCR) control in PC9s (Figure 3.9A). I chose to knockdown ERK1 and ERK2 separately based on a preliminary experiment using a siERK1/2 pool, which transiently and simultaneously knocks down both ERK1 and 2. The experiment showed that depleting ERK1 and ERK2 simultaneously causes an acute toxicity in PC9 cells, while individually knocking down ERK1 and ERK2 is better tolerated. Based on this result, hairpins targeting ERK1 and ERK2 separately were used to create stable lines. Using these ERK1 or ERK2 knockdown PC9 cells, I performed a 5-Day double transfection assay using siDUSP6, siEGFR and Non-T Control to see if suppression of ERK can moderate the siDUSP6-induced lethality in PC9s (Figure 3.9B and C). In two of the four knockdown lines (ERK1-#5 and ERK2-#7), depleting ERK levels partially mitigated the toxic effects of siDUSP6. Although the degree shERK-mediated rescues were modest in both cell lines, the results strongly suggested that the lethality induced by knocking down DUSP6 is mediated through ERK activity in LAC cells.   58  A       B        C        Figure 3.9: siDUSP6 rescue assays in four different ERK KD PC9 cell lines. (A) Western blot demonstrating the knockdown efficiency the hairpins targeting ERK1 (#4, #5) or ERK2 (#6, #7) along with the shSCR control. The top band represents ERK2 (p44) and the bottom band represents ERK1 (p42). (B) 5-Day double knockdown assay using siDUSP6 Pool, siEGFR and Non-T. The AlamarBlue values measured on Day 5 in biological triplicates were averaged and normalized to the AlamarBlue values in the SCR control, with +/- SEM. Only shERK1-5 and shERK2-7 showed a diminishing effect relative to the Non-T control. (C) Knockdown efficiency of siDUSP6 and siEGFR confirmed in western blot analysis of Day 5 lysates.  siDUSP6siEGFRNon-Target0.00.51.01.5Relative Number of Viabile CellsshERK1-5shERK2-7shRNA-ScramEGFRDUSP6β-actinsiRNA:PC9 shERK1	#5 shERK2	#7 shSCRDUSP6EGFRN/TDUSP6EGFRN/TDUSP6EGFRN/T59  3.4 Inhibition of ERK rescues cells from BCI-induced lethality Similar to the shERK-siDUSP6 rescue experiments, I attempted to rescue H358 from BCI by simultaneously treating these cells using an ERK inhibitor called VX-11E. VX-11E allows ERKs to be phosphorylated but interferes with their catalytic abilities by acting as an ATP-competitive inhibitor of ERK. This drug combination assay was performed for 72 hours. At higher doses of VX-11E, the relative cell viabilities remained higher with increasing concentrations of BCI (Figure 3.10A). A concurrent western blot analysis revealed that P-ERK and P-RSK, which is a downstream target of ERK and thus indicative of ERK activity, were elevated after BCI treatment alone. Upon combination treatment with VX-11E, P-ERK levels were still elevated in H358s, while the P-RSK levels were significantly diminished as anticipated (Figure 3.10B).  The diminished P-RSK levels indicates that VX-11E is rescuing H358s from BCI through suppressing the kinase activity of ERK.     60  A                    B               Figure 3.10: Combination drug treatment of BCI and VX-11E in H358s. (A) Relative cell viability after 72hours of VX-11E and BCI treatment. Relative viability reflects the average values of biological triplicates after being normalized to the BCI=0uM (0.1% DMSO) at their respective Vx-11e dosage. The error bars represent +/-SEM of the average values. (B) Western blot analysis of H358s after combination treatment of BCI and VX-11E for 6 hours to assess their effects on ERK and RSK activation.   RSK	(1/2/3)p-p90RSK	[Ser380]p-p44/42[T202/Y204]β Actinp44/42H358BCI	5uMVX11e	2uM- + - +- - + +0 1 2 3 50.00.51.0BCI (uM)Relative ViabilityVX-11e - 0nMVX-11e - 100nMVX-11e - 500nMVX-11e - 1000nM61  3.5 Sensitization of a BCI-resistant line, HCC95, through prolonged EGF treatment Next, I hypothesized that cells that are continuously treated with a high dose of EGF should a higher basal level of P-ERK and thereby be closer to the upper signalling threshold of ERK. In other words, a constitutive and prolonged stimulation of EGFR through EGF treatments in KRAS and EGFR WT cell should increase the P-ERK levels similar to the levels found in mKRAS/mEGFR LC cells and thus be more sensitive to DUSP6 inhibition induced toxicity. I tested this hypothesis by treating HCC95s, an BCI-insensitive line, with a high dose of EGF for 10 days prior to performing a 72-hour BCI dose response with the pre-treated cells (Figure 3.1B and 3.3A). In cells that were pre-treated with EGF, the basal levels of P-EGFR, P-ERK and DUSP6 were elevated compared to the cells without the pre-treatment (Figure 3.11A). Additionally, a western blot analysis of P-ERK revealed that in EGF-treated cells, P-ERK levels dramatically increased with increasing doses of BCI. This fluctuation in P-ERK levels is significantly different than in non-pretreated HCC95s that maintained a constant P-ERK levels even at high doses of BCI, like 8uM (Figure 3.3B, 3.11A,B). Most importantly, EGF pre-treated cells exhibited increased sensitivity to BCI and had a lower IC50 to BCI (Figure 3.12). Asides from these changes, no phenotypic changes were exhibited after 10-days of EGF treatment.   62  A                          B                Figure 3.11: HCC95 sensitization to BCI after 10 days of EGF (100ng/mL) treatment.  (A) HCC95s were cultured in high dose of EGF (100ng/mL) for 10 days prior to the experiment. HCC95s, with and without the 10-day pre-treatment, were then treated for 6 hours with 5 different doses of BCI (0, 1, 3, 5, 8μM) at 0.1% DMSO concentration. Lysates were collected after 6 hours and blotted for P-EGFR, EGFR, P-ERK, ERK, DUSP6 and β-Actin as loading control. (B) Relative P-ERK levels reflect the densitometry values of the P-ERK1/2 bands shown on panel A after being normalized to β-Actin values of their respective lanes.+EGF (100 ng/mL)-EGF HCC95[KRASWTEGFRWT]BCI (µM):p44/42P-p44/42 [T202/Y204]β Actin0 1N/A 3 5 8P-EGFR [Y1068]DUSP6EGFR 0 1N/A 3 5 8HCC95HCC95 + EGF02468Relative pERK 0uM1uM3uM5uM8uM[BCI] HCC95 HCC95+EGF Relative P-ERK Level 63                         Figure 3.12: BCI dose-response curves in HCC95s with and without EGF (100ng/mL) pre-treatment for 10 days. (A) BCI-insensitive HCC95s were treated with EGF for 10 days before being treated with varying BCI doses with 0.5uM increments for 72 hours. The 0uM represents the vehicle control (0.1% DMSO) and the EGF was maintained in the wells containing the pre-treated HCC95s. The plotted values represent the averages of biological triplicates with +/-SEM. The red curve represents the pre-treated line that exhibits increased sensitivity to BCI than the black curve (non-pretreated line).   0 1 2 30.00.10.20.30.40.50.60.70.80.91.01.1Log2 BCI (uM)Relative Number of Viabile CellsHCC95 +EGFHCC95 64  Chapter 4: Discussion 4   4.1 Therapeutic application of hyper-RAS/ERK lethality Our experiments examined the effects of DUSP6 inhibition in NSCLC and the results presented here are support the overarching hypothesis that cancer cells experience toxicity when the oncogenic pathway that they are dependent on is overstimulated. Similar observations of hyperactivation of oncogenic signalling leading to lethality have also been made by other groups [69, 70, 71, 72, 73]. Collectively, these findings corroborate to demonstrate that the Goldilocks principle also applies to the context of cancer. When cancer cells are treated with inhibitors that target their oncogenic pathways, cancer cells die from “too little” signalling activity. When cancer cells are hyperactivated through the inhibition of its key negative regulator or an upregulation of a positive regulator, “too much” signalling activity also leads to cell death or decrease in viability.  Furthermore, the results from the BCI experiments suggest that an effective agonistic therapy can potentially be developed to target advanced mKRAS or mEGFR NSCLCs by developing a clinically viable DUSP6 inhibitors.   65  4.2 DUSP6 inhibition in mutant KRAS-/EGFR-driven NSCLC generates toxicity The experiments using BCI or siRNAs targeting DUSP6, as shown in Figures 3.1, 3.2 and 3.6, demonstrated that inhibiting DUSP6 can induce lethality in LAC cells possessing mutations in either KRAS or EGFR.   4.2.1 RNA-interference experiments One interesting observation from the knockdown experiments was that depleting DUSP6 mRNAs led to the decreased expression of the oncogene in the cells. On the other hand, suppressing the driver oncogene of the cell led to a decreased expression of DUSP6. The latter phenomenon can partly be explained by the feedback relationship between ERK and DUSP6: a decrease in KRAS or EGFR lead to a decrease in P-ERK, which in turn decreases the amount of DUSP6 transcribed [86]. However, this feedback loop between DUSP6 and ERK does not help to explain why DUSP6 inhibition leads to a decrease in KRAS or EGFR levels in NSCLC cells. One possible explanation is that the cells are trying to normalize increased P-ERK levels through decreasing the number of KRAS or EGFR proteins [77,86].  Furthermore, P-ERK levels in mKRAS/mEGFR LAC cells with DUSP6 knockdown were lower than in Non-T cells at day 5 (Fig 3.1). This was unexpected as I initially hypothesized that the P-ERK levels would be higher in the DUSP6 knockdown cells than in the control cells.  This suggests that while the initial burst of P-ERK upon DUSP6 knockdown is required to hyperactivate ERK and induce lethality, these levels are transient and do not have to be maintained for cells to die from hyper-ERK lethality. To confirm that the effects of siDUSP6 and BCI are mediated through DUSP6 and are not due to an artifactual cause, several experiments were performed to test the specificity of the siRNA pool and BCI. First, each siRNA that composes the siDUSP6 pool was tested against the pool 66  Between the 4 individual siRNAs and the pool itself, the 5-Day double transfection showed that the degree of DUSP6 suppression correlates with the degree of reduction in cell viability. This result strongly supports that the observed toxicity is caused by the suppression of DUSP6 (Figure 3.5). In addition, this experiment also revealed an important aspect of how DUSP6 knockdown mediates hyper-ERK lethality; DUSP6 has to reach a critically low point for its absence to sufficiently increase P-ERK to toxic levels. Otherwise, moderate levels of DUSP6 depletion, as shown through siDUSP6 #6, #7 and #9, elicited only a modest increase in P-ERK that led to an increase in cell viability compared to the Non-T.    4.2.2 BCI experiments BCI, a small molecule inhibitor of DUSP1 and DUSP6, was used to determine whether the inhibition of DUSP6 kinase activity, rather than its depletion, can also generate toxicity.  BCI caused acute toxicity in NSCLC cells with a mutation in either KRAS or EGFR but not in NSCLC cells without either mutations. This helped to demonstrate that DUSP6 is a pharmacologically viable target and that NSCLC patients with an oncogenic mutation in KRAS or EGFR can potentially benefit from a DUSP6 inhibitor. However, since BCI has been reported to inhibit both DUSP1 and DUSP6, experiments had to be performed to test whether the BCI toxicity is mediated through the inhibition of DUSP1, DUSP6 or both (Figure 3.8). To address this question, a double-transfection assay comparing siDUSP1 and siDUSP6 was performed, which revealed that the depletion of DUSP1 does not affect the viability of mKRAS/mEGFR LAC cells, suggesting that the effect of BCI is specifically mediated through the inhibition of DUSP6. However, based on this experiment alone, it is not possible to determine whether simultaneously knocking down DUSP1 and DUSP6 can synergize, although unlikely based on the siDUSP1 results. To test this, I performed an experiment by co-67  transfecting two independent siRNAs, namely siDUSP1 and siDUSP6. However, such protocol has not been tested over time as the single transfection protocol and thus is susceptible to many unknown technical limitations. As expected, my experiment did not achieve a sufficient knockdown of either proteins. A follow-up experiment should be conducted to test the possibility of DUSP1 and DUSP6 synergy. One potential experiment that can overcome this technical limitation is to create a stable knockdown or knockout of DUSP6 in a cell line and perform a siDUSP1 double transfection assay and vice versa. If the simultaneous suppression of DUSP1 and DUSP6 do synergize, this can also add to the explanation of why BCI-induced death is more acute (i.e. 72 hours vs. 5 days) and robust than the RNA-interference mediated death. In the BCI-dose response experiments, cell lines exhibited three different BCI sensitivities, which correlated with their mutation status of KRAS and EGFR (Figure 3.6). Not only do these results strongly support my hypothesis but the clear and wide separation of sensitivity between the two groups of cells (mKRAS/mEGFR and WT) further implicates DUSP6 as an attractive therapeutic target. In order to better assess whether the therapeutic window between the mKRAS/mEGFR cells and wildtype cells are sufficiently wide enough to exploit in a clinical setting, response to BCI should be measured using in vivo models in subsequent studies. Tet-ON system can be used to generate transgenic mice that can express mutant KRAS or EGFR in Type II epithelial cells upon Dox administration. The rate of growth and the final tumor sizes can be compared in groups with or without BCI treatment. This experiment can also expose the potential complications of using BCI and any inherent limitations of employing an agonistic therapy, which has never been explored before.      68  4.3 DUSP6 inhibition lethality is mediated through the hyperactivation of ERK   As previously discussed, I wanted to confirm that DUSP6-associated lethality is truly mediated through the hyperactivation of ERK and whether other nodes of the RAS pathway also contribute to the lethality. In order to test this this, a combination knockdown and drug rescue assays were performed using siDUSP6-shERK combo and the BCI-VX-11E combo. Although, experiments supported my initial hypothesis by rescuing cells from DUSP6 inhibition through ERK suppression, the degree of the rescue in both experiments was less dramatic than anticipated (section 3.3). In the shERK rescue experiments, only two hairpins out of the four hairpins (shERK1-5 and shERK2-7) rescued the cells from siDUSP6. A possible explanation as to why only shERK2-7 rescued and not shERK2-6 is that the knockdown efficiency of shERK2-7 was significantly greater than shERK2-6 (Figure 3.9A). The same explanation cannot help to explain why only one of the two ERK1 KD lines rescued as the knockdown efficiencies of the two hairpins were similar (Figure 3.9A). Further investigation on how minute differences in ERK1 levels or targeting different regions of the ERK1 mRNA affects cells can help explain the observed discrepancy. Also, such issues can potentially be overcome by using knockdown/knockout systems that elicit higher knockdown/knockout efficiencies. Since shRNAs typically generate modest levels of knockdown compared to other systems, a different system like a Tet-inducible CRISPR targeting ERK1 and ERK2 can be expected to overcome the technical limitations pertaining to shRNA systems and enhance the degree of rescue.     69  4.4 P-ERK dynamics in BCI-sensitive vs. insensitive cells  One interesting finding was that cells without a pre-existing mutation in KRAS or EGFR have not only remained insensitive to DUSP6 inhibition but their P-ERK levels also remained consistent even at high doses of BCI (Figures 3.7B, 3.11A and B). This was in contrast with the anticipation that the P-ERK levels should increase upon BCI treatment regardless of their mutation status as the DUSP6 is still getting inhibited. This unexpected result cannot be explained without further research on P-ERK dynamics in cells upon DUSP6 inhibition and how different intracellular environments can affect these levels.   Furthermore, I succeeded in sensitizing HCC95 cells, which are known to express high levels of WT EGFR, by treating them with a high dose of EGF for 10 days. The EGF treatment was intended to mimic the constant state of P-EGFR and P-ERK activation that can be found in EGFR-mutant cells. After 10 days of EGF treatment, the HCC95s were sensitized to BCI (Figure 3.12). In addition, in the pre-treated cells, the P-ERK levels increased in a dose-dependent manner while the non-pretreated cells maintained a static P-ERK level across all doses of BCI (0, 1, 3, 5, 8uM) (Figure 3.11A and B).  This experiment demonstrates that, at least in HCC95, elevating the basal P-ERK levels through extracellular receptor stimulation is sufficient for cells to develop a dependency on DUSP6.          70  Chapter 5: Conclusion  Together, these results highlight the potential benefits of developing an ‘agonistic’ therapeutic approach to target tumors, as opposed to solely relying on inhibition-based approach. As described in this thesis, DUSP6 is an attractive therapeutic target for mKRAS/mEGFR NSCLC tumors as this approach can potentially circumvent the current challenges of TKI resistance and developing an effective treatment for mKRAS cancers. However, several remaining questions that must be addressed before attempting to develop an inhibitor for DUSP6. For instance, a detailed mechanism of how this lethality is mediated after the point of ERK hyperactivation is still unclear. Based on the previous co-expression model studies and the results presented here, there seem to be multiple mechanisms of death, including apoptosis, autophagocytosis and methuosis. Other groups have also reported that hyperactive oncogenic signalling results in senescence [71, 72, 73]. These findings collectively raise the possibility that the level of ERK hyperactivation can play a crucial role in committing the cell to a specific mode of toxicity. Answering such questions can reveal ways to further potentiate DUSP6 treatment. For instance, during our BCI rescue experiments we have discovered that N-acetyl-cysteine (NAC), which is an antioxidant that scavenges ROS, can fully rescue all the cell lines from the effects of BCI (supplemental Figure 1) [87]. Unsurprisingly, ERK and p38 have been implicated in ROS generation and ROS in turn activates the MAPK pathways, forming a positive feedback loop [88, 89, 90, 91]. Although the steps in between the initial burst of ERK activation and ROS mediating lethality in unclear, it can be reasonably deduced that drugs that increase ROS levels in cells can synergize with the effects of BCI. A prime example of such drug is sulfasalazine, which is an anti-inflammatory drug that is typically used to treat different inflammatory conditions, such as rheumatoid arthritis and Crohn’s disease [92, 93]. Interestingly, the application of Sulfasalazine has been investigated in different cancers because of its unexpected ability to inhibit Xc- cystine transporter (xCT) [92, 93]. xCT is 71  a membrane bound antiporter that simultaneously exchanges an extracellular cysteine for an intracellular glutamate. This import of cystine is critical for glutathione (GSH) production, which is an intracellular antioxidant that can scavenge ROS [94]. Also, it has been found that in oral cancer cells, smoking increases xCT levels and treating these cells with sulfasalazine can decrease their proliferation rates [94]. In light of these findings, the potential synergistic relationship between BCI and Sulfasalazine should be explored in future studies.  It is important to note that agonistic therapeutic approach does not have to be limited to inhibiting the key negative regulators of oncogenic pathways. There have recent studies that test a new therapeutic mode called “Drug Holiday”. As the name suggests, drug holiday involves intentionally giving TKI holidays for patients. The rationale behind this treatment regime is overtime, cancer cells tend to grow dependent on targeted inhibitors due to constant exposure. Early in vitro results that exploit the BRAF and MEK inhibitor dependency in BRAFV600E mutant melanomas showed that withdrawing these inhibitors from the tumor cells result in parthanatos-related cell death, coupled with an acute burst of P-ERK [95]. Learning from this approach, a similar relationship can be at play for EGFR-TKI resistant NSCLC tumors. As a matter of fact, it has been recently reported that a selection of NSCLC tumors treated with Osimertinib, a third-generation EGFR TKI, acquires KRAS mutation as a resistance mechanism. Withdrawing Osimertinib in this subset of patients will reactivate EGFR signalling and recreate the hyper-RAS/ERK toxicity that was observed in the mKRAS/mEGFR co-expression models. Furthermore, there are other driver mutations in the MAPK pathways, such as BRAF or MEK1, that can also be sensitive to DUSP6 inhibitors, albeit these mutations occur less frequently than in KRAS or EGFR. Similarly, I anticipate that there are other targetable negative regulators in other oncogenic pathways that can trigger a similar mechanism of hyperactivation-associated toxicity. An optimal candidate pathway is the PI3K/AKT pathway, which is frequently altered in many types of cancer. 72  Using the experimental approach described in this thesis, a different negative regulator of the PI3K/AKT pathway can be discovered and tested for its capacity to serve as a target for an agonistic therapy. My thesis and the results presented here is intended to help broaden our understanding of LACs and to stimulate further research in developing an agonistic therapy. 73  Chapter 6: Supplemental Figures  Fig. S1 A                B                    Fig S1: Low-density NAC-BCI rescue assay in PC9 and H358. Cells seeded at low-densities (A. PC9: 200; B. H358: 300) treated with escalating doses of BCI with or without NAC (5mM). AlamarBlue values measured after 14 days of treatment. Each value on the graph represent the averages of the biological triplicates that were normalized to DMSO +/- NAC to reflect the relative viability changes within each plate. The error bars represent the SEM of the triplicates.    74  References   1. Weinberg, R. A. (2014). The biology of cancer (Second edition ed.). New York, NY: Garland Science. 2. Hanahan, D., & Weinberg, R. A. (2011). Hallmarks of cancer: The next generation. Cell, 144(5), 646-674. 10.1016/j.cell.2011.02.013 3. Balkwill, F. R., Capasso, M., & Hagemann, T. (2012). The tumor microenvironment at a glance. Journal of Cell Science, 125(Pt 23), 5591-5596. 10.1242/jcs.116392 4. Fulda, S., Gorman, A. M., Hori, O., & Samali, A. (2010). Cellular stress responses: Cell survival and cell death. International Journal of Cell Biology, 2010, e214074. 10.1155/2010/214074 Retrieved from https://www.hindawi.com/journals/ijcb/2010/214074/ 5. Liou, G., & Storz, P. (2010). Reactive oxygen species in cancer. Free Radical Research, 44(5), 479-496. 10.3109/10715761003667554 6. Schumacker, P. T. (2015). Reactive oxygen species in cancer: A dance with the devil. Cancer Cell, 27(2), 156-157. 10.1016/j.ccell.2015.01.007 7. Ubersax, J. A., & Ferrell, J. E. (2007). Mechanisms of specificity in protein phosphorylation. Nature Reviews. Molecular Cell Biology, 8(7), 530-541. 10.1038/nrm2203 8. Scheeff, E. D., Eswaran, J., Bunkoczi, G., Knapp, S., & Manning, G. (2009). Structure of the pseudokinase VRK3 reveals a degraded catalytic site, a highly conserved kinase fold, and a putative regulatory binding site. Structure (London, England: 1993), 17(1), 128-138. 10.1016/j.str.2008.10.018 9. Lai, S., & Pelech, S. (2016). Regulatory roles of conserved phosphorylation sites in the activation T-loop of the MAP kinase ERK1.Molecular Biology of the Cell, 27(6), 1040-1050. 10.1091/mbc.E15-07-0527 10. Didkowska, J., Wojciechowska, U., Mańczuk, M., & Łobaszewski, J. (2016). Lung cancer epidemiology: Contemporary and future challenges worldwide. Annals of Translational Medicine, 4(8), 150. 10.21037/atm.2016.03.11 11. Miller, K. D., Siegel, R. L., Lin, C. C., Mariotto, A. B., Kramer, J. L., Rowland, J. H., . . . Jemal, A. (2016). Cancer treatment and survivorship statistics, 2016. CA: A Cancer Journal for Clinicians, 66(4), 271-289. 10.3322/caac.21349 12. Dela Cruz, C. S., Tanoue, L. T., & Matthay, R. A. (2011). Lung cancer: Epidemiology, etiology, and prevention. Clinics in Chest Medicine, 32(4), 605-644. 10.1016/j.ccm.2011.09.001 13. Heist, R. S., & Engelman, J. A. (2012). SnapShot: Non-small cell lung cancer. Cancer Cell, 21(3), 448.e2. 10.1016/j.ccr.2012.03.007 14. Cheng, T. D., Cramb, S. M., Baade, P. D., Youlden, D. R., Nwogu, C., & Reid, M. E. (2016). The international epidemiology of lung cancer: Latest trends, disparities, and tumor characteristics. Journal of Thoracic Oncology: Official Publication of the International Association for the Study of Lung Cancer, 11(10), 1653-1671. 10.1016/j.jtho.2016.05.021 15. Peto, R., Darby, S., Deo, H., Silcocks, P., Whitley, E., & Doll, R. (2000). Smoking, smoking cessation, and lung cancer in the UK since 1950: Combination of national statistics with two case-control studies. BMJ (Clinical Research Ed.), 321(7257), 323-329. 16. Hoh, E., Hunt, R. N., Quintana, P. J. E., Zakarian, J. M., Chatfield, D. A., Wittry, B. C., . . . Matt, G. E. (2012). Environmental tobacco smoke as a source of polycyclic aromatic hydrocarbons in settled household dust. Environmental Science & Technology, 46(7), 4174-4183. 10.1021/es300267g 75  17. The health consequences of Smoking—50 years of progress: A report of the surgeon general (2014). . Atlanta (GA): Centers for Disease Control and Prevention (US). Retrieved from http://www.ncbi.nlm.nih.gov/books/NBK179276/ 18. Hecht, S. S. (2012). Lung carcinogenesis by tobacco smoke. International Journal of Cancer, 131(12), 2724-2732. 10.1002/ijc.27816  19. Xue, J., Yang, S., & Seng, S. (2014). Mechanisms of cancer induction by tobacco-specific NNK and NNN. Cancers, 6(2), 1138-1156.  20. Moorthy, B., Chu, C., & Carlin, D. J. (2015). Polycyclic aromatic hydrocarbons: From metabolism to lung cancer. Toxicological Sciences: An Official Journal of the Society of Toxicology, 145(1), 5-15. 10.1093/toxsci/kfv040 21. Herbst, R. S., Heymach, J. V., & Lippman, S. M. (2008). Lung cancer. The New England Journal of Medicine, 359(13), 1367-1380. 10.1056/NEJMra0802714 22. Travis, W. D., Brambilla, E., Nicholson, A. G., Yatabe, Y., Austin, J. H. M., Beasley, M. B., . . . Wistuba, I. (2015). The 2015 world health organization classification of lung tumors: Impact of genetic, clinical and radiologic advances since the 2004 classification. Journal of Thoracic Oncology: Official Publication of the International Association for the Study of Lung Cancer, 10(9), 1243-1260. 10.1097/JTO.0000000000000630 23. Travis, W. D. (2012). Update on small cell carcinoma and its differentiation from squamous cell carcinoma and other non-small cell carcinomas. Modern Pathology: An Official Journal of the United States and Canadian Academy of Pathology, Inc, 25 Suppl 1, 18. 10.1038/modpathol.2011.150  24. Hiroshima, K., & Mino-Kneudson, M. (2017). Update on large cell neuroendocrine carcinoma. Translational Lung Cancer Research, 6(5), 530-539. 10.21037/tlcr.2017.06.12 25. Miyauchi, E., Motoi, N., Ono, H., Ninomiya, H., Ohyanagi, F., Nishio, M., . . . Ishikawa, Y. (2015). Distinct characteristics of small cell lung cancer correlate with central or peripheral origin: Subtyping based on location and expression of transcription factor TTF-1.Medicine, 94(51), e2324. 10.1097/MD.0000000000002324  26. Zappa, C., & Mousa, S. A. (2016). Non-small cell lung cancer: Current treatment and future advances. Translational Lung Cancer Research, 5(3), 288-300. 10.21037/tlcr.2016.06.07  27. Samet, J.M., Avila-Tang, E., Boffetta, P., Hannan, L.M., Olivo-Marston, S., Thun, M.J., & Rudin, C.M. (2009). Lung cancer in never smokers: Clinical epidemiology and environmental risk factors. Clinical Cancer Research, 15(18), 5626-5645. 10.1158/1078-0432.CCR-09-0376 28. Yang, P., Cerhan, J. R., Vierkant, R. A., Olson, J. E., Vachon, C. M., Limburg, P. J., . . . Sellers, T. A. (2002). Adenocarcinoma of the lung is strongly associated with cigarette smoking: Further evidence from a prospective study of women. American Journal of Epidemiology, 156(12), 1114-1122.  29. Sullivan, J. P., Minna, J. D., & Shay, J. W. (2010). Evidence for self-renewing lung cancer stem cells and their implications in tumor initiation, progression, and targeted therapy. Cancer Metastasis Reviews, 29(1), 61-72. 10.1007/s10555-010-9216-5  30. Solis, L. M., Behrens, C., Raso, M. G., Lin, H. Y., Kadara, H., Yuan, P., . . . Moran, C. A. (2012). Histologic patterns and molecular characteristics of lung adenocarcinoma associated with clinical outcome. Cancer, 118(11), 2889-2899. 10.1002/cncr.26584  31. Chen, Z., Fillmore, C. M., Hammerman, P. S., Kim, C. F., & Wong, K. (2014). Non-small-cell lung cancers: A heterogeneous set of diseases. Nature Reviews. Cancer, 14(8), 535-546. 10.1038/nrc3775  32. Derman, B. A., Mileham, K. F., Bonomi, P. D., Batus, M., & Fidler, M. J. (2015). Treatment of advanced squamous cell carcinoma of the lung: A review. Translational Lung Cancer Research, 4(5), 524.  76  33. Kalemkerian, G. P. (2012). Staging and imaging of small cell lung cancer. Cancer Imaging : The Official Publication of the International Cancer Imaging Society, 11(1), 253-258. 10.1102/1470-7330.2011.0036 Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/22245990  34. Lin, J. J., & Shaw, A. T. (2016). Resisting resistance: Targeted therapies in lung cancer. Trends in Cancer, 2(7), 350-364. 10.1016/j.trecan.2016.05.010  35. Chan, B. A., & Hughes, B. G. M. (2015). Targeted therapy for non-small cell lung cancer: Current standards and the promise of the future. Translational Lung Cancer Research, 4(1), 36-54. 10.3978/j.issn.2218-6751.2014.05.01 36. George, J., Lim, J. S., Jang, S. J., Cun, Y., Ozretić, L., Kong, G., . . . Thomas, R. K. (2015). Comprehensive genomic profiles of small cell lung cancer. Nature, 524(7563), 47-53. 10.1038/nature14664 37. Cressman, S., Peacock, S. J., Tammemägi, M. C., Evans, W. K., Leighl, N. B., Goffin, J. R., . . . Lam, S. (2017). The cost-effectiveness of high-risk lung cancer screening and drivers of program efficiency. Journal of Thoracic Oncology: Official Publication of the International Association for the Study of Lung Cancer, 12(8), 1210-1222. 10.1016/j.jtho.2017.04.021 38. Gandhi, L., Rodríguez-Abreu, D., Gadgeel, S., Esteban, E., Felip, E., De Angelis, F., . . . Garassino, M. C. (2018). Pembrolizumab plus chemotherapy in metastatic Non–Small-cell lung cancer. New England Journal of Medicine, 378(22), 2078-2092. 10.1056/NEJMoa1801005  39. Weinstein, I. B., & Joe, A. (2008). Oncogene addiction. Cancer Research, 68(9), 3080; discussion 3080. 10.1158/0008-5472.CAN-07-3293 40. De Grève, J., Van Meerbeeck, J., Vansteenkiste, J. F., Decoster, L., Meert, A., Vuylsteke, P., . . . Teugels, E. (2016). Prospective evaluation of first-line erlotinib in advanced non-small cell lung cancer (NSCLC) carrying an activating EGFR mutation: A multicenter academic phase II study in caucasian patients (FIELT). PloS One, 11(3), e0147599. 10.1371/journal.pone.0147599 41. Bagnyukova, T. V., Serebriiskii, I. G., Zhou, Y., Hopper-Borge, E. A., Golemis, E. A., & Astsaturov, I. (2010). Chemotherapy and signalling. Cancer Biology & Therapy, 10(9), 839-853. 10.4161/cbt.10.9.13738 Retrieved from   42. Minami, T., Kijima, T., Kohmo, S., Arase, H., Otani, Y., Nagatomo, I., . . . Kumanogoh, A. (2013). Overcoming chemoresistance of small-cell lung cancer through stepwise HER2-targeted antibody-dependent cell-mediated cytotoxicity and VEGF-targeted antiangiogenesis. Scientific Reports, 3, 2669. 10.1038/srep02669  43. Gazdar, A. F. (2010). Epidermal growth factor receptor inhibition in lung cancer: The evolving role of individualized therapy. Cancer Metastasis Reviews, 29(1), 37-48. 10.1007/s10555-010-9201-z  44. Chong, C. R., & Jänne, P. A. (2013). The quest to overcome resistance to EGFR-targeted therapies in cancer. Nature Medicine, 19(11), 1389-1400. 10.1038/nm.3388  45. McCormick, F. (2015). KRAS as a therapeutic target. Clinical Cancer Research: An Official Journal of the American Association for Cancer Research, 21(8), 1797-1801. 10.1158/1078-0432.CCR-14-2662 Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/25878360  46. Ledford, H. (2015). Cancer: The ras renaissance. Nature, 520(7547), 278-280. 10.1038/520278a  47. Janes, M. R., Zhang, J., Li, L., Hansen, R., Peters, U., Guo, X., . . . Liu, Y. (2018). Targeting KRAS mutant cancers with a covalent G12C-specific inhibitor. Cell, 172(3), 589.e17. 10.1016/j.cell.2018.01.006  77  48. Marabese, M., Ganzinelli, M., Garassino, M. C., Shepherd, F. A., Piva, S., Caiola, E., . . . Rulli, E. (2015). KRAS mutations affect prognosis of non-small-cell lung cancer patients treated with first-line platinum containing chemotherapy. Oncotarget, 6(32), 34014-34022. 10.18632/oncotarget.5607  49. Yang, P., Cerhan, J. R., Vierkant, R. A., Olson, J. E., Vachon, C. M., Limburg, P. J., . . . Sellers, T. A. (2002). Adenocarcinoma of the lung is strongly associated with cigarette smoking: Further evidence from a prospective study of women. American Journal of Epidemiology, 156(12), 1114-1122.  50. Leung, A. W. Y., de Silva, T., Bally, M. B., & Lockwood, W. W. (2016). Synthetic lethality in lung cancer and translation to clinical therapies. Molecular Cancer, 15(1), 61. 10.1186/s12943-016-0546-y  51. Farmer, H., McCabe, N., Lord, C. J., Tutt, A. N. J., Johnson, D. A., Richardson, T. B., . . . Ashworth, A. (2005). Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature, 434(7035), 917-921. 10.1038/nature03445  52. Leung, A. W. Y., Dragowska, W. H., Ricaurte, D., Kwok, B., Mathew, V., Roosendaal, J., . . . Bally, M. B. (2015). 3′-phosphoadenosine 5′-phosphosulfate synthase 1 (PAPSS1) knockdown sensitizes non-small cell lung cancer cells to DNA damaging agents. Oncotarget, 6(19), 17161-17177.  53. Leung, A. W. Y., Veinotte, C. J., Melong, N., Oh, M. H., Chen, K., Enfield, K. S. S., . . . Lockwood, W. W. (2017). In vivo validation of PAPSS1 (3′-phosphoadenosine 5′-phosphosulfate synthase 1) as a cisplatin-sensitizing therapeutic target. Clinical Cancer Research, 23(21), 6555-6566.  54. Purba, E., Saita, E., & Maruyama, I. (2017). Activation of the EGF receptor by ligand binding and oncogenic mutations: The “Rotation model”. Cells, 6(2), 13. 10.3390/cells6020013  55. Kumar, A., Petri, E. T., Halmos, B., & Boggon, T. J. (2008). Structure and clinical relevance of the epidermal growth factor receptor in human cancer. Journal of Clinical Oncology: Official Journal of the American Society of Clinical Oncology, 26(10), 1742-1751. 10.1200/JCO.2007.12.1178  56. Gazdar, A. F., Shigematsu, H., Herz, J., & Minna, J. D. (2004). Mutations and addiction to EGFR: The achilles 'heal' of lung cancers? Trends in Molecular Medicine, 10(10), 481-486. 10.1016/j.molmed.2004.08.008  57. Dawson, J.P., Berger, M.B., Lin, C., Schlessinger, J., Lemmon, M.A., & Ferguson, K.M. (2005). Epidermal growth factor receptor dimerization and activation require ligand-induced conformational changes in the dimer interface. Molecular and Cellular Biology, 25(17), 7734-7742. 10.1128/MCB.25.17.7734-7742.2005  58. Hobbs, G. A., Der, C. J., & Rossman, K. L. (2016). RAS isoforms and mutations in cancer at a glance. Journal of Cell Science, 129(7), 1287-1292. 10.1242/jcs.182873  59. Cargnello, M., & Roux, P. P. (2011). Activation and function of the MAPKs and their substrates, the MAPK-activated protein kinases. Microbiology and Molecular Biology Reviews: MMBR, 75(1), 50-83. 10.1128/MMBR.00031-10  60. Reuter, C. W., Morgan, M. A., & Bergmann, L. (2000). Targeting the ras signalling pathway: A rational, mechanism-based treatment for hematologic malignancies? Blood, 96(5), 1655-1669.  61. Wimmer, R., & Baccarini, M. (2010). Partner exchange: Protein-protein interactions in the raf pathway. Trends in Biochemical Sciences, 35(12), 660-668. 10.1016/j.tibs.2010.06.001  62. Ascierto, P. A., Kirkwood, J. M., Grob, J., Simeone, E., Grimaldi, A. M., Maio, M., . . . Mozzillo, N. (2012). The role of BRAF V600 mutation in melanoma. Journal of Translational Medicine, 10, 85. 10.1186/1479-5876-10-85  78  63. Turkson, J., Bowman, T., Adnane, J., Zhang, Y., Djeu, J.Y., Sekharam, M., . . . Jove, R. (1999). Requirement for ras/Rac1-mediated p38 and c-jun N-terminal kinase signalling in Stat3 transcriptional activity induced by the src oncoprotein. Molecular and Cellular Biology, 19(11), 7519-7528. 10.1128/MCB.19.11.7519  64. Cicenas, J., Zalyte, E., Rimkus, A., Dapkus, D., Noreika, R., & Urbonavicius, S. (2018). JNK, p38, ERK, and SGK1 inhibitors in cancer. Cancers, 10(1), 1. 10.3390/cancers10010001   65. Unni, A. M., Lockwood, W. W., Zejnullahu, K., Lee-Lin, S., & Varmus, H. (2015). Evidence that synthetic lethality underlies the mutual exclusivity of oncogenic KRAS and EGFR mutations in lung adenocarcinoma. eLife, 4, e06907. 10.7554/eLife.06907  66. Gossen, M., Freundlieb, S., Bender, G., Müller, G., Hillen, W., & Bujard, H. (1995). Transcriptional activation by tetracyclines in mammalian cells. Science (New York, N.Y.), 268(5218), 1766-1769.  67. Bockamp, E., Maringer, M., Spangenberg, C., Fees, S., Fraser, S., Eshkind, L., . . . Zabel, B. (2002). Of mice and models: Improved animal models for biomedical research. Physiological Genomics, 11(3), 115-132. 10.1152/physiolgenomics.00067.2002 68. Saikali, Z., & Singh, G. (2003). Doxycycline and other tetracyclines in the treatment of bone metastasis. Anti-Cancer Drugs, 14(10), 773-778. 10.1097/01.cad.0000099002.92896.cf 69.  Overmeyer, J. H., Kaul, A., Johnson, E. E., & Maltese, W. A. (2008). Active ras triggers death in glioblastoma cells through hyperstimulation of macropinocytosis. Molecular Cancer Research: MCR, 6(6), 965-977. 10.1158/1541-7786.MCR-07-2036 70. Chi, S., Kitanaka, C., Noguchi, K., Mochizuki, T., Nagashima, Y., Shirouzu, M., . . . Kuchino, Y. (1999). Oncogenic ras triggers cell suicide through the activation of a caspase-independent cell death program in human cancer cells. Oncogene, 18(13), 2281-2290. 10.1038/sj.onc.1202538 Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/10327074 71. Collado, M., & Serrano, M. (2010). Senescence in tumours: Evidence from mice and humans. Nature Reviews. Cancer, 10(1), 51-57. 10.1038/nrc2772 72. Cisowski, J., Sayin, V. I., Liu, M., Karlsson, C., & Bergo, M. O. (2016). Oncogene-induced senescence underlies the mutual exclusive nature of oncogenic KRAS and BRAF. Oncogene, 35(10), 1328-1333. 10.1038/onc.2015.186 73. Cisowski, J., & Bergo, M. O. (2017). What makes oncogenes mutually exclusive? Small GTPases, 8(3), 187-192. 10.1080/21541248.2016.1212689 Retrieved from http://www.tandfonline.com/doi/abs/10.1080/21541248.2016.1212689 74. Amin, A. D., Rajan, S. S., Groysman, M. J., Pongtornpipat, P., & Schatz, J. H. (2015). Oncogene overdose: Too much of a bad thing for oncogene-addicted cancer cells. Biomarkers in Cancer, 7(Suppl 2), 25-32. 10.4137/BIC.S29326 75. Northwood, I. C., Gonzalez, F. A., Wartmann, M., Raden, D. L., & Davis, R. J. (1991). Isolation and characterization of two growth factor-stimulated protein kinases that phosphorylate the epidermal growth factor receptor at threonine 669. The Journal of Biological Chemistry, 266(23), 15266-15276. 76. Sato, K., Shin, M., Sakimura, A., Zhou, Y., Tanaka, T., Kawanishi, M., . . . Sakurai, H. (2013). Inverse correlation between thr-669 and constitutive tyrosine phosphorylation in the asymmetric epidermal growth factor receptor dimer conformation. Cancer Science, 104(10), 1315-1322. 10.1111/cas.12225 79  77. Lake, D., Corrêa, S. A. L., & Müller, J. (2016). Negative feedback regulation of the ERK1/2 MAPK pathway. Cellular and Molecular Life Sciences: CMLS, 73(23), 4397-4413. 10.1007/s00018-016-2297-8 78. Caunt, C. J., & Keyse, S. M. (2013). Dual‐specificity MAP kinase phosphatases (MKPs). FEBS Journal, 280(2), 489-504. 10.1111/j.1742-4658.2012.08716.x Retrieved from https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1742-4658.2012.08716.x 79. Zhenfeng Zhang, Susumu Kobayashi, Alain C Borczuk, Rom S Leidner, Thomas Laframboise, Alan D Levine, & Balazs Halmos. (2010). Dual specificity phosphatase 6 (DUSP6) is an ETS-regulated negative feedback mediator of oncogenic ERK signalling in lung cancer cells. Carcinogenesis, 31(4), 577-586. 10.1093/carcin/bgq020 Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/20097731 80. Wakioka, T., Sasaki, A., Kato, R., Shouda, T., Matsumoto, A., Miyoshi, K., . . . Yoshimura, A. (2001). Spred is a sprouty-related suppressor of ras signalling. Nature, 412(6847), 647-651. 10.1038/35088082 81. King, J. A. J., Straffon, A. F. L., D'Abaco, G. M., Poon, C. L. C., I, S. T. T., Smith, C. M., . . . Hovens, C. M. (2005). Distinct requirements for the sprouty domain for functional activity of spred proteins. The Biochemical Journal, 388(Pt 2), 445-454. 10.1042/BJ20041284 82. Molina, G., Vogt, A., Bakan, A., Dai, W., Queiroz de Oliveira, P., Znosko, W., . . . Tsang, M. (2009). Zebrafish chemical screening reveals an inhibitor of Dusp6 that expands cardiac cell lineages. Nature Chemical Biology, 5(9), 680-687. 10.1038/nchembio.190 83. Korotchenko, V. N., Saydmohammed, M., Vollmer, L. L., Bakan, A., Sheetz, K., Debiec, K. T., . . . Tsang, M. (2014). In vivo Structure–Activity relationship studies support allosteric targeting of a dual specificity phosphatase. ChemBioChem, 15(10), 1436-1445. 10.1002/cbic.201402000 Retrieved from https://onlinelibrary.wiley.com/doi/abs/10.1002/cbic.201402000 84. Lang, R., Hammer, M., & Mages, J. (2006). DUSP meet immunology: Dual specificity MAPK phosphatases in control of the inflammatory response. Journal of Immunology (Baltimore, Md.: 1950), 177(11), 7497-7504. 85. Robitaille, A. C., Caron, E., Zucchini, N., Mukawera, E., Adam, D., Mariani, M. K., . . . Grandvaux, N. (2017). DUSP1 regulates apoptosis and cell migration, but not the JIP1-protected cytokine response, during respiratory syncytial virus and sendai virus infection.Scientific Reports, 7(1), 1-15. 10.1038/s41598-017-17689-0 Retrieved from https://search.proquest.com/docview/1983418300 86. Nissan, M. H., Rosen, N., & Solit, D. B. (2013). ERK pathway inhibitors: How low should we go? Cancer Discovery, 3(7), 719-721. 10.1158/2159-8290.CD-13-0245 87. Hong, A., Moriceau, G., Sun, L., Lomeli, S., Piva, M., Damoiseaux, R., . . . Lo, R. S. (2018). Exploiting drug addiction mechanisms to select against MAPKi-resistant melanoma. Cancer Discovery, 8(1), 74-93. 10.1158/2159-8290.CD-17-0682 80  88. Sun, S. (2010). N-acetylcysteine, reactive oxygen species and beyond. Cancer Biology & Therapy, 9(2), 109-110. 10.4161/cbt.9.2.10583 Retrieved from http://www.tandfonline.com/doi/abs/10.4161/cbt.9.2.10583 89. Li, Q., Fu, G., Zheng, J., He, J., Niu, X., Chen, Q., . . . Jiang, B. (2013). NADPH oxidase subunit p22(phox)-mediated reactive oxygen species contribute to angiogenesis and tumor growth through AKT and ERK1/2 signalling pathways in prostate cancer.Biochimica Et Biophysica Acta, 1833(12), 3375-3385. 10.1016/j.bbamcr.2013.09.018 90. Kassan, M., Choi, S., Galán, M., Lee, Y., Trebak, M., & Matrougui, K. (2014). Enhanced p22phox expression impairs vascular function through p38 and ERK1/2 MAP kinase-dependent mechanisms in type 2 diabetic mice. American Journal of Physiology. Heart and Circulatory Physiology, 306(7), 972. 10.1152/ajpheart.00872.2013 91. Son, Y., Cheong, Y., Kim, N., Chung, H., Kang, D. G., & Pae, H. (2011). Mitogen-activated protein kinases and reactive oxygen species: How can ROS activate MAPK pathways? Journal of Signal Transduction, 2011, 792639. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/21637379 92. Sano, M., Fukuda, K., Sato, T., Kawaguchi, H., Suematsu, M., Matsuda, S., . . . Ogawa, S. (2001). ERK and p38 MAPK, but not NF-kappaB, are critically involved in reactive oxygen species-mediated induction of IL-6 by angiotensin II in cardiac fibroblasts.Circulation Research, 89(8), 661-669. 93. Gout, P.W., Buckley, A.R., Simms, C.R., & Bruchovsky, N. (2001). Sulfasalazine, a potent suppressor of lymphoma growth by inhibition of the xc- cystine transporter: A new action for an old drug. Leukemia, 15(10), 1633-1640. 10.1038/sj.leu.2402238 Retrieved from https://search.proquest.com/docview/220553038 94. Plosker, G. L., & Croom, K. F. (2005). Sulfasalazine: A review of its use in the management of rheumatoid arthritis. Drugs, 65(13), 1825-1849. 95. Ji, X., Qian, J., Rahman, S. M. J., Siska, P. J., Zou, Y., Harris, B. K., . . . Massion, P. P. (2018a). xCT (SLC7A11)-mediated metabolic reprogramming promotes non-small cell lung cancer progression. Oncogene, , 1. 10.1038/s41388-018-0307-z Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/29789716  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0376231/manifest

Comment

Related Items