UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigations into the cellular pathways underlying ETV6-NTRK3-mediated transformation Lannon, Christopher L. 2004

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

Item Metadata


831-ubc_2005-995003.pdf [ 18.17MB ]
JSON: 831-1.0092365.json
JSON-LD: 831-1.0092365-ld.json
RDF/XML (Pretty): 831-1.0092365-rdf.xml
RDF/JSON: 831-1.0092365-rdf.json
Turtle: 831-1.0092365-turtle.txt
N-Triples: 831-1.0092365-rdf-ntriples.txt
Original Record: 831-1.0092365-source.json
Full Text

Full Text

INVESTIGATIONS INTO THE CELLULAR PATHWAYS UNDERLYING ETV6-NTRK3-MEDIATED TRANSFORMATION by  CHRISTOPHER L. LANNON B.Sc.H., Saint Mary's University, 1996 M . S c , Dalhousie University, 1998  Submitted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY in F A C U L T Y OF G R A D U A T E STUDIES (Department of Pathology and Laboratory Medicine)  THE UNIVERSITY OF BRITISH C O L U M B I A  December 2004  © Christopher  L. Lannon 2004  ii  ABSTRACT Receptor tyrosine kinases are integral components of cellular signaling pathways, and are frequently deregulated in malignancies. There is increasing interest in the potential role of the NTRK family of neurotrophin receptors in human neoplasia. These proteins are known to mediate neuronal cell survival and differentiation, but altered NTRK signaling has also been implicated in oncogenesis. The ETV6-NTRK3 (EN) gene fusion results from a t(12; 15)(pl3; q25) translocation, and occurs in human pediatric spindle cell sarcomas and secretory breast carcinoma. EN fusion transcripts encode a chimeric protein tyrosine kinase, formed by the fusion ofthe SAM dimerization domain of ETV6 and the tyrosine kinase domain of NTRK3. The resultant EN fusion protein functions as a constitutively active protein tyrosine kinase with potent transforming activity. Further, EN-mediated transformation is associated with activation of the Ras-MAPK and PI3K-Akt pathways, increased levels of cyclin DI, and constitutive phosphorylation of the insulin receptor substrate 1 (IRS-1) through an interaction with the phosphotyrosine binding (PTB) domain of IRS-1. We have recently identified two C-terminal mutants (A614 and Y615F) that abolish and reduce binding to IRS-1, and subsequent transformation. Neither this region nor full-length EN protein contain any known PTB interaction motifs (i.e., NPXY); however, it does contain a TPIY, which may function as a putative PTB recognition sequence. TPIY mutants appear to abrogate anchorage-independent growth (soft agar assay and spheroid formation). These mutants most likely introduce a conformational change that affects protein interactions N-terminal to this region. However, these mutants appear to still bind IRS-1. Therefore, IRS-1 binding does not guarantee transformation in EN-expressing cells. To elucidate the role of EN in tumourigenesis, we next generated transgenic mice expressing ETV6-NTRK3 under the direction of two ubiquitously expressing promoters (CMV or (3-actin/CMV), as this fusion protein appears to be implicated in a range of tumour types. We developed 13 independent founder mouse strains. Approximately 30% ofthe transgenic mice from two independent strains developed lymphomas after a long latency period. Additionally, a single mouse developed a fibrosarcoma expressing EN; this lesion was histologically identical to clinical cases of EN-expressing congenital fibrosarcoma. The low penetrance of tumour formation coupled with the advanced age of the mice raises several questions as to the cellular environment that may be prerequisite for ETV6-NTRK3 oncogenesis in these mice.  iii  TABLE OF CONTENTS INVESTIGATIONS INTO THE CELLULAR PATHWAYS UNDERLYING ETV6-NTRK3 TRANSFORMATION. ABSTRACT  i ii  T A B L E OF CONTENTS  iii  LIST OF FIGURES  vi  LIST OF T A B L E S  viii  LIST OF ABBREVIATIONS  ix  ACKNOWLEDGEMENTS  xi  C H A P T E R I: I N T R O D U C T I O N  1  1.1 SYNOPSIS A N D R A T I O N A L E F O R T H E THESIS  1  1.2 R E G U L A T I O N O F N O R M A L C E L L G R O W T H & PROLIFERATION  1.2.1 1.2.2 1.2.3 1.2.4 1.2.5 1.2.6  Growth Factor Receptors The Ras-Rafl-Mek-Erk Mitogen Activated Protein Kinase Pathway The PI3K-AKT Pathway Cross-Talk between Ras-MAPK and PI3-K Signal Transduction Pathways Cell Cycle Apoptosis  1.3 C A N C E R  2  2 5 6 8 9 12 13  1.3.1 1.3.2  Mechanisms of Oncogenesis Tumour Suppressor Genes Retinoblastoma Protein (RB) p53 1.3.3 Oncogenes 1.3.4 Chromosomal Rearrangements 1.3.5 Fusion genes 1.3.6 The Laboratory Mouse as a Model System for Cancer 1.4 T H E ETV6-NTRK3 CHIMERIC ONCOPROTEIN 1.5ETV6 1.5.1 The ETV6 Gene as a Target of Chromosomal Translocations 1.6NTRK3 1.6.1 NTRK Expression in Other Human Tumours  14 14 15 16 18 21 25 27 29 32 32 33 35  1.7 EXPRESSION OF EN FUSION TRANSCRIPT IN H U M A N MALIGNANCIES  35  1.8 EN SIGNAL TRANSDUCTION  38  1.8.1 1.8.2 1.8.3  Role of the Insulin-Like Growth Factor 1 Receptor Signaling Axis in EN Transformation 40 Role oftheTGF-(3 Pathway 42 Higher Order Polymer Formation of the EN Oncoprotein 43  1.9 A I M S & OBIECTIVES  44  C H A P T E R II: M A T E R I A L S A N D M E T H O D S  46  IV  2.1  C E L L C U L T U R E - B A S E D TRANSFORMATION STUDIES  2.1.1 2.1.2 2.1.3 2.1.4 2.1.5 2.1.6. 2.1.7 2.2  46  Cell Lines Generation of Full-Length ETV6-NTRK3 and Mutant cDNA Transduction of Genes Using the Retroviral Vector MSCVpac Assessment of Transformation Protein Analysis Immunofluorescence Homology Modeling of Kinase Domain of EN  TRANSGENIC M I C E  2.1.1 2.2.2 2.2.3 2.2.4 2.2.5 2.2.6 2.2.7 2.2.8 2.2.9 2.2.10 2.2.11 CHAPTER  54  Vector Construction and Confirmation of Expression Construct Injection Preparation of Tail DNA for Genotype Analysis Genotype Analysis by PCR Pathology of Transgenic Tissues RNA Isolation and RT-PCR Protein Analysis FACS Analysis Tumour Transplantation Assay Mouse Cross-Breeding EN Targeted ES Cells  III: A H I G H L Y C O N S E R V E D  46 46 48 50 52 52 53  NTRK3  C-TERMINAL  54 56 57 59 59 59 61 62 62 63 63 SEQUENCE  I N T H E ETV6-NTRK3 O N C O P R O T E I N B I N D S T H E P T B D O M A I N O F IRS-1  65  3.1  RESULTING PUBLICATION & CONTRIBUTION OF INDIVIDUAL A U T H O R S  65  3.2  INTRODUCTION  66  3.3  RESULTS  68  3.4  DISCUSSION  85  CHAPTER  I V : T H E C - T E R M I N U S O F ETV6-NTRK3 IS E S S E N T I A L F O R  TRANSFORMING ACTIVITY  91  4.1  INTRODUCTION  92  4.2  RESULTS  93  4.3  DISCUSSION  106  CHAPTER  V: E NTRANSGENIC MICE DEVELOP  LYMPHOMAS AFTER A LONG  LATENCY PERIOD  110  5.1  R E S U L T I N G PUBLICATION & CONTRIBUTION O F INDIVIDUAL A U T H O R S  110  3.2  INTRODUCTION  111  3.3  RESULTS  112  3.4  DISCUSSION  135  V  CHAPTER VI: SUMMARY AND FUTURE DIRECTIONS  141  6.1  GENERAL SUMMARY  141  6.2  E N SIGNAL TRANSDUCTION  142  6.3  M O D E L SYSTEMS F O R INVESTIGATING E N TUMOURIGNESIS  153  6.4  FINAL C O M M E N T S  157  REFERENCES  159  VI  LIST OF FIGURES FIGURE # FIGURE 1.  TITLE OF FIGURE  PAGE  Chromosomal translocations can result in a potentially oncogenic gene rearrangement.  23  FIGURE 2.  Schematic diagram of the ETV6-NTRK3 (EN) fusion protein.  31  FIGURE 3.  Outline of EN signal transduction.  41  FIGURE 4.  Constructs used for transgenesis.  55  FIGURE 5.  ETV6-NTRK3 (EN) fusion binds to the phosphotyrosine (PTB) domain of IRS-1.  70  Generation of ETV6-NTRK3 (EN), EN-A614, and EN-Y615F expressing cells and assessment of transformation.  72  IRS-1 is not constitutively tyrosine phosphorylated in EN-A614 expressing NIH3T3 cells.  74  Assessment of the transformation ability of EN-A614 and ENY615F.  77  FIGURE 6.  FIGURE 7.  FIGURE 8.  FIGURE 9.  Differential Mekl, cyclin D1 and Akt activation in EN- and EN mutant-expressing NIH3T3s.  79  FIGURE 10.  EN-A614 does not associate with p85 or Grb2 via IRS-1.  80  FIGURE 11.  Co-expression of IRS-1 C (PTB/PH domains) disrupts EN/IRS-1 complexes.  82  FIGURE 12.  Overexpression of IRS-1 potentiates EN transformation.  84  FIGURE 13.  FIGURE 15.  The C-terminal 29 amino acids of EN, indicating the position and sequence of EN mutants. NIH 3T3s expressing TPIY ("NPXY") mutants display transformed morphology. EN-P626A reduces tumour growth in SCID mice.  97  FIGURE 16.  A624 mutant does not affect colony growth in soft agar.  99  FIGURE 14.  94 96  FIGURE 17.  Y615F+Y628Q double mutant reduces colony size in soft agar.  100  FIGURE 18.  Y615F+Y628Q mutant, but not the A624 mutant, decrease tumour formation in an immunocompromised mouse injection model.  102  EN C-terminal mutants block anchorage-independent growth (spheroid formation).  103  FIGURE 20.  EN C-terminal mutants interact with the PTB domain of IRS-1.  105  FIGURE 21.  EN constructs for transgenesis are able to induce morphological transformation in NIH3T3s.  114  FIGURE 19.  FIGURE 22.  EN-constructs for pronuclear injection are not able to form colonies in soft agar.  115  FIGURE 23.  EN transcription is detected in various tissues from transgenic mice.  117  FIGURE 24.  Histology of Large Cell Lymphoma from 2015-7 mouse expressing EN protein.  124  FIGURE 25.  Lymphomas in EN transgenic mice are of T and B-cell origin.  127  FIGURE 26.  Histology of fibrosarcoma from EN transgenic Mouse is identical to clinical CFS. EN expression is detectable at both the RNA and protein level in a single fibrosarcoma.  130 131  FIGURE 28.  Expression of ETV6-NTRK3 in targeted murine ES cells.  132  FIGURE 29.  EN expression in ES cells does not induce activation of Akt,  FIGURE 27.  MEK, or cyclin D l .  134  FIGURE 30.  Homology modeling oftheNTRK3 portion of EN.  145  FIGURE 31.  A614 mutant shows different cellular localization than EN.  151  LIST OF TABLES  TABLE #  TITLE OF TABLE  PAGE  TABLE 1.  Proto-oncogenes can be activated to become oncogenes through a variety of mechanisms.  23  TABLE 2.  Recurrent Chromosomal Translocations in Soft-Tissue Sarcoma.  24  TABLE 3.  The ETV6 locus is involved in a range of chromosomal translocations.  28  TABLE 4.  Primer sequences used in site-directed mutagenesis and sequence analysis.  58  TABLE 5.  Summary of fertilized eggs injected and subsequent generation of founder strains.  116  TABLE 6.  Site of EN transcription in 6 / 13 founder strains.  118  TABLE 7.  Incidence of tumours in the 2015-7 strain by 18 months of age.  120  TABLE 8.  Incidence of tumours in the 2029-4 strain by 18 months of age.  120  TABLE 9.  Distribution of tumours in the 2015-7 strain.  122  TABLE 10.  Distribution of tumours in the 2029-4 strain.  123  TABLE 11.  Protein crystal structures used as templates for homology modeling.  143  LIST OF ABBREVIATIONS ABL ALK ALL AML APC ARF ARG ATP BAD Bcl-2 BCR BDNF bp BSA CDK cDNA CFS CHOP CIP CKIs CML CMML CMN CS CTLC DAPI DFSP DMEM DMSO DNA DSRCT  Abelson murine leukemia anaplastic lymphoma kinase acute lymphoid leukemia acute myeloid leukemia adenomatosis polyposis coli alternate reading frame Abelson-related gene adenosine triphosphate bcl-2 antagonist of cell death b-cell CLL/lymphoma 2 breakpoint cluster region brain-derived neurotrophic factor base pair bovine serum albumin cyclin dependent serine / threonine kinases complimentary deoxyribonucleic acid congenital fibrosarcoma C/EBP-homologous protein calf intestinal phosphatase cyclin-dependent kinase inhibitor chronic myeloid leukemia chronic myelomonocytic leukemia congenital mesoblastic nephroma calf serum clathrin heavy chain diamidino-2-phenylindole dihydrochloride hydrate dermatofibrosarcoma protuberans Dulbecco's modified eagle medium dimethylsulfoxide deoxyribonucleic acid desmoplastic small round cell tumour  EDTA EGF EGFR EN Erbb2  ERG ERK ES ETS ETV6 EWS FADD Fas FGF FGFR FKHR Fli FLT GEF Grb2 GSK3 GTPase H&E HA HLH HPV IDC IGF-1R IGF2 IL ILK IRS  ethylene-diaminetetraacetic acid epidermal growth factor epidermal growth factor receptor ETV6-NTRK3 v-erb-b2 erythroblastic leukemia viral oncogene homolog 2 ets-related gene extracellular signal regulated kinase embryonic stem E-26 transforming specific ets variant gene 6 Ewings sarcoma fas-associated death domain fibroblast growth factor fibroblast growth factor receptor forkhead in rhabdomyosarcoma friend leukemia virus integration fms-related tyrosine kinase guanine exchange factor growth factor receptorbound protein 2 glycogen synthase kinase-3 guanosine triphosphatase hematoxylin and eosin hemagglutinin helix-loop-helix human papillomavirus infiltrating ductal carcinoma insulin-like growth factor 1 receptor insulin growth factor 2 interleukin integrin-linked kinase insulin receptor substrate  X  JAK kb kDa LPS MAPK MDS MEK mRNA MSCV NF-KB  NIPA  nM NMR NPM nt NT-3 NTRK3 PBS PCR PDGF PDGFR PDK PI3K PKC PLC PML PMSF PTB PTEN  PTK RAR RB RD  janus family of tyrosine kinases kilo-base kilo-daltons liposarcoma mitogen-activated protein kinase myelodysplastic syndrome map kinase/erk-activating kinase messenger ribonucleic acid murine stem cell virus nuclear factor-kappa B nuclear interacting partner of anaplastic lymphoma kinase nanomolar nuclear magnetic resonance nucleophosmin nucleotide neurotrophin-3 neurotrophic tyrosine kinase receptor type 3 phosphate buffered saline polymerase-chain reaction platelet derived growth factor platelet-derived growth factor receptor phosphoinositide dependent protein kinase phosphoinositol-3' kinase protein kinase c phospholipase c promyelocytic leukemia phenylmethylsulfonyl fluoride phosphotyrosine binding phosphatase and tensin homolog deleted on chromosome ten protein tyrosine kinase retinoic acid receptor retinoblastoma Rag (recombination activating gene)-deficient  RMS RNA RTK SAIN SAM SBC SCID SDS SH2 SH3 SHC SHIP2 SNT SOS SRC STAT TBS TCR TEL TGF TK TLS TNF TPM TRKC TSG VEGF VEGFR  rhabdomyosarcoma ribonucleic acid receptor tyrosine kinase SHC and IRS- NPXYbinding sterile alpha motif secretory breast carcinoma severe combined immune deficiency sodium dodecyl sulfate src-homology 2 src homology 3 src homology and collagen SH2-containing inositol phosphatase 2 SUC1-associated neurotrophin factor target son of sevenless Schmidt-Ruppin A-2) viral oncogene homolog signal transducers and activators of transcription tris-buffered saline T cell antigen receptor translocation, ets, leukemia transforming growth factor tyrosine kinase translocated in liposarcoma tumour necrosis factor tropomyosin tropomyosin receptor kinase c tumour-suppressor gene vascular endothelial growth factor vascular endothelial growth factor receptor  XI  Acknowledgements I would like to thank everyone who helped me with this thesis either directly or indirectly. I am grateful for the support of my supervisor, Dr. Poul Sorensen as well as members of my supervisory committee: Drs. Rick Hegele, Keith Humphries, and Elizabeth Simpson. I am indebted to past and present members of the Sorensen lab for their technical support, guidance, insightful scientific discussions, and sense of humour. I would like to thank Drs. Gareth Jevon, Derek DeSa, and Maureen O'Sullivan for their assistance with the mouse histology and the CMMT Transgenic facility, particularly Ms. Kayla Shayne, for assistance with animal husbandry. I acknowledge financial support from the Canadian Institute for Health Research and the Michael Smith Foundation for Health Research. My family has been incredibly supportive during the past six years and I am grateful for that. Finally this thesis would not have been possible without the love, support, patience, and inspiration of Amber.  1  CHAPTER I INTRODUCTION  1.1  SYNOPSIS AND RATIONALE FOR THESIS The presence of chromosomal translocations in human cancers has always been of  interest to both the diagnostic pathologist and the clinician. Further, many chromosomal translocations that occur in human malignancies lead to the expression of chimeric oncogenes formed from the in-frame fusion of coding sequences from two different genes. These translocations (and their resultant fusion product) are usually very tumour specific and are used to sub-classify many tumours, both diagnostically and prognostically. In many cases, the chimeric fusion gene is a mechanism by which tumours hyper-activate a proto-oncogene, converting it to an oncogene. Accordingly, therapeutic strategies targeting these fusion genes have been proposed (and developed) for many years. The ETV6-NTRK3  fusion gene was first described in congenital fibrosarcoma, and  detection thereof was quickly employed to distinguish this tumour from several other, histologically similar, tumours (1, 2). This fusion is rare in that it has since been identified in several different malignancies of varying cell lineages. Numerous studies have identified the ETV6-NTRK3 fusion as causal in the development of these cancers. Therefore, a further understanding of the biology of this fusion protein may lead to novel therapeutic interventions for a range of malignant diseases. In this thesis, I have studied the cellular biology induced by the ETV6-NTRK3 protein. In particular, these studies have focused on signal transduction events at the carboxy-terminus of  2 this protein. Additionally, I have created a transgenic mouse model of ETV6-NTRK3 expression, with the intention of recapitulating the clinical manifestations of this fusion. In the following sections, a general review of normal and abnormal cell biology will be discussed, with an emphasis of how normal cellular pathways are deregulated in human malignancy, with a particular focus on growth factor receptor activation. This will be followed by a brief discussion of the general mechanism of oncogenesis, including chromosomal translocations and the chimeric fusion genes that often result from them. Finally, our current understanding of ETV6-NTRK3 cell biology will be presented with reference to the current published literature. For additional discussions of normal and dysregulated growth, the reader is directed to the list of references at the end of this thesis.  1.2  REGULATION OF NORMAL C E L L GROWTH & PROLIFERATION Growth factors are involved in controlling cell proliferation and survival, as well as  migration along extracellular matrices and guidance by chemotactic or repulsive cues underlying normal development. The interaction of a growth factor with its receptor by specific binding initiates a cascade of biochemical events that result in a wide range of biological responses. Perturbation of this tightly regulated cascade at any point can contribute to the altered cell growth associated with malignancy. In this section, the cellular processes that control these key events will be discussed.  1.2.1  Growth Factor Receptors During normal embryonic development and in adult life, signal transduction needs to be  precisely coordinated and integrated, and properly regulated differentiation signals are critical  3 for preventing oncogenesis (3). Growth factors mediate their diverse etiologic responses by binding to and activating cell-surface receptors with intrinsic protein kinase activity (4). The initial step of most signal transduction cascades is the binding of extracellular growth factors to transmembrane receptors that, once activated, bind cytoplasmic proteins. These proteins include cytoplasmic kinases that are phosphorylated at tyrosine or serine and threonine residues, resulting in a conformational change and activation of their kinase domain. Consequently, a sequential cascade of phosphorylation and dephosphorylation is initiated, resulting in propagation of the signal to the nucleus (see (5) for review). Among the best-understood growth factor regulated pathways are those mediated by receptor tyrosine kinases (RTKs). Sequencing data from the Human Genome Project has revealed that there are more than 90 protein tyrosine kinases (PTKs) in the human genome, 58 of which encode for RTKs (3, 6). PTKs comprise a large portion of known oncogenes and tumour suppressor genes (7). Recently, an impressive study used high-throughput bioinformatics to identify a large number of known and novel mutations in tyrosine kinase genes in colorectal cancer (8). Therefore, the physiological regulation of RTKs is key to understanding the mechanisms causing their oncogenic activation (3). Signaling by RTKs requires ligand-induced receptor oligomerization; dimerization of RTKs is followed by receptor autophosphorylation, usually by one receptor phosphorylating the other in the dimer (9). In the unphosphorylated state, the receptor possesses a low catalytic activity due to the particular conformation of a specific domain in the kinase region, which interferes with the phosphotransfer event (10). Phosphorylation of the activation loop of the kinase domain removes this inhibition, and the catalytic activity is enhanced and persists for some time independently of the presence of the ligand. Once phosphorylated, RTKs are catalytically active and the phosphorylated residues act as docking sites for cytoplasmic  4  signaling proteins that recognize specific phospho-tyrosine residues (10). In this manner, active signaling transducers complex at the cell surface with growth factor receptors through the recognition of phospho-tyrosines by specific binding domains. Recruitment to phosphorylated tyrosine residues on receptors leads to activation of the signaling molecule through a range of mechanisms: tyrosine phosphorylation (e.g., phospholipases  and STATs); conformational  changes induced by the binding of the Src homology-2 (SH2) and other protein tyrosine binding domains (e.g., Grb2, Src); and translocation to the plasma membrane for stimulation (e.g., Ras) (3,11). Given the extensive activation of RTKs in human cancer, and extensive data supporting their causal role in the development and progression of many human cancers, it is no surprise that both RTKs and their growth-factor ligands have become rational targets for therapeutic intervention. The characterization of both the crystal structure of these RTKs as well as elucidation of their primary and secondary functions of these receptors and their ligands in tumourigenesis (and for that matter, normal homeostasis) has allowed for the development of the first target-specific cancer therapeutics (12). The first of such therapeutics was directed against a R T K with a high level of homology to human epidermal growth factor receptor (EGFR), named ERBB2 (v-erb-b2 erythroblastic leukemia viral oncogene homolog 2; frequently referred to as HER2/neu)  (13). Subsequently, the therapeutic antibody trastuzumab (Herceptin®, Genentech,  Inc.) began clinical trials for use in patients with metastatic breast cancer. There are currently more than 20 RTK-targeted therapies that are approved or in clinical development, including compounds directed against B C R - A B L (Gleevec®, Novartis Pharmaceuticals Inc.), plateletderived growth factor receptor (PDGFR), vascular endothelial growth factor receptor 1 and 2  5  (VEGFR1/2), fms-related tyrosine kinase 3 (FLT3), and tropomyosin receptor kinases (TRKs) (reviewed in (12)) and the IGF1R (14).  1.2.2  The RAS-RAF1-MEK-ERK Mitogen Activated Protein Kinase Pathway The Ras-Raf-Mek-Erk pathway is a central signal transduction pathway, which transmits  signals from multiple cell receptors to transcription factors in the nucleus. This pathway is frequently referred to as the M A P K (mitogen-activated protein kinase) pathway, and can be stimulated by mitogens, cytokines, and growth factors. Ras is a 21 kDa molecular weight protein that is targeted to the cell membrane by farnesylation, and cycles between active GTP-bound and inactive GDP-bound configurations (5). Inactivation of the Ras protein occurs via its intrinsic GTPase activity, which is facilitated by the family of GTPase-activating proteins (GAPs) (5). When Ras is bound to GTP it is active, whereas GDP-bound Ras is inactive (15). Ras activation occurs at the cell membrane, downstream of growth factor receptor activation, and is accomplished by guanine nucleotide exchange factors such as SOS (Son of Sevenless) that exchange free GTP for Ras-bound GDP (5). In the case of R T K signaling, receptor activation leads to direct activation of Ras via the SH2-adapter molecule GRB2, which links the activated receptor to SOS, thereby facilitating activation of Ras. Upon activation, Ras recruits several downstream effector molecules that initiate a number of changes in cellular phenotype. Among Ras effectors is Rafl, a serine/threonine kinase that becomes phosphorylated upon binding to activated Ras (15). Rafl phosphorylates MEK1 and M E K 2 (MAPK-ERK kinases) on serine residues; MEK1 and M E K 2 are dual-specificity kinases that in turn activate ERK1 and ERK2 (extracellular stimulus regulated kinases) by phosphorylating threonine and tyrosine residues (5). Once phosphorylated, ERK2 translocates to  6  the nucleus where it phosphorylates a number of different transcription factors, including c-Myc, Elk-1, ETS-1, C R E B , AP-1, and N F - K B (16). Activation of these transcription factors leads to D N A replication and cell division. Interestingly, activation of the other MAPKs, (for example, SAPK/JNK (stress-activated protein kinase/jun amino-terminal kinase), occurs via structurally related molecules in parallel signaling pathways, but leads to growth arrest rather than proliferation (5).  1.2.3  The P I 3 K - A K T Pathway The ability of trophic factors to promote survival has been attributed, at least in part, to  the phosphatidylinositide 3'-OH kinase (PI3K)-AKT kinase cascade. Survival stimuli generally mediate intracellular signaling through ligation of transmembrane receptors, which either possess intrinsic tyrosine kinase activity (e.g., the insulin-like growth factor one receptor, IGF1R), are indirectly coupled to tyrosine kinases (e.g., integrins), or are coupled to seven transmembrane, G protein-coupled receptors (15). Activation of these receptors results in the recruitment of PI3K members (such as p85) to the inner surface of the plasma membrane, typically through the SH2 domain of p85, or indirectly through intermediate phosphoproteins such as the insulin receptor substrates IRS1 and IRS2 (17). Additionally, PI3K can be activated indirectly via intermediate activation of Ras; thus, like R A F , PI3K is a Ras effector and links proliferative and survival pathways (15, 18). Following recruitment of the PI3K heterodimer (regulatory and catalytic subunits) to the cell membrane, and interaction of p85 with RTKs (for example), p i 10 catalyzes the transfer of phosphate from A T P to membrane-localized phospholipids termed phosphoinositides. The principally generated 3'-phosphorylated phosphoinositides are phosphoinositol 3,4 bisphosphate  7  (PI3,4P)  and phosphoinositol  intermediates  regulating  3,4,5  triphosphate (PI3,4,5P) which function as signaling  downstream  signal  transduction  cascades  (15,  19).  The  3'-  phosphorylated phosphoinositides activate the serine/threonine kinase A K T (also termed PKB) by phosphorylation at serine  308  and threonine  473  by potentially  three  molecules:  phosphoinositide-dependent kinase (PDK)l, PDK2 (15) and integrin-linked kinase (ILK) (20). Further, A K T can also be activated by non-PI3K dependent means such as members of the protein kinase A pathway (21) and the calcium/calmodulin dependent kinase kinase (CAMKK1) (22), which lead to phosphorylation of A K T . Active A K T is then translocated, through an unknown mechanism, to the nucleus where many of its substrates are localized (e.g., FKHR, F O X O , and CREB) (19). The main biological consequences of A K T activation can be loosely characterized into three processes: survival / anti-apoptosis, proliferation (increased cell number) and growth (increased cell size) (19). The anti-apoptotic effects of A K T occur through phosphorylation of a variety of targets, notably BCL2-antagonist of cell death (Bad). Phosphorylated Bad interacts with 14-3-3 proteins, inhibiting the ability of Bad to interact with Bcl-2 and Bcl-X ; Bcl-X then L  L  binds to pro-apoptotic Bax molecules and prevents their induction of apoptosis (15). A K T can also influence cell survival through indirect effects on both N F - K B (nuclear factor of kappa light chain gene enhancer in B-cells) (23) and p53 (24). A K T has been shown to phosphorylate and activate IKB kinase (IKK), a kinase that induces degradation of the N F - K B inhibitor, IKB (23). A K T can also influence the activity of the pro-apoptotic tumour suppressor p53 (discussed in detail below), through phosphorylation of the p53-binding protein mdm2; mdm2 targets p53 for degradation through the E3 ubiquitin ligase activity of mdm2 (19).  8  1.2.4  Cross-Talk between Ras-MAPK and PI3-K Signal Transduction Pathways Interaction between the Ras-MAPK and PI3K/AKT pathways, or crosstalk, is an area of  intense research (25) and multiple lines of evidence support the existence of these interactions, the best characterized of which is outlined below. First, cells treated with pharmacological inhibitors of PI3K also have decreased levels of E R K phosphorylation (26, 27). Second, A K T phosphorylates and negatively regulates RAF1 (28, 29) and, as discussed above, PI3K has been shown to be an effector of Ras (18). Third, a dominant-negative mutant of the p85 subunit of PI3K decreased MEK1 and E R K activities and, correspondingly, proliferation (30). Fourth, it has been demonstrated that while an activated form of M E K can induce cell division in quiescent fibroblasts, PI3K signals are required for this effect (31). Finally, evidence from several experimental systems illustrates that Grb2 recruitment results in association of Gab2 (32, 33), which can associate with SHP2 (coupling to ERK1/2 activation) and the p85 subunit of PI3K (coupling to A K T phosphorylation). Thus, a significant amount of cross-talk exists between the Ras-MAPK proliferation pathway and the PI3K survival pathway; this cross-talk is crucial for the regulation of cell proliferation, death and differentiation. Disruption of these cascades and their mechanisms of cross-talk leads to dysregulated cell growth and is a common feature of many malignancies. Cross-talk permits more finely tuned regulation of homeostasis than would the action of individual independent pathways. For example, it is well established that constitutive expression of RAS can induce apoptosis or cell cycle arrest through mechanisms involving p l 9  A R F  , p21  CIP1  ,  and p53 (reviewed in (34)). Further, concomitant activation of the PI3K-AKT pathway prevents Ras-induced cell cycle arrest or apoptosis and allow proliferation to occur (25, 31, 35). Multiple nodes of interaction in each pathway also allows for multiple potential regulation points as well.  9  However, inappropriate cross talk can cause second messengers to be misinterpreted. Thus modification of a single pathway component by mutation or targeted therapy could have profound effects on cellular signaling. It is important to note that cross-talk of these, and other pathways, may be cell-type and stimulus-specific; ergo, investigators should exercise care when extrapolating from other experimental systems.  1.2.5  Cell Cycle Homeostasis of normal tissues is maintained by an intricately regulated balance between  cell proliferation, growth arrest and differentiation, and cell death. In malignancy, these cell functions become deregulated, leading to an increase in cell number. Since both Ras-MAPK and PI3K7AKT pathways discussed above converge on the cell cycle, regulation of the cell cycle will now be summarized. The cell division cycle can be divided into four phases: two functional phases, S and M phases, and two preparatory phases, GI and G2. S phase is defined as the phase in which D N A synthesis occurs. Fully replicated chromosomes are segregated to two genetically identical daughter cells through a process called mitosis in M phase. The other two phases of the cell cycle, GI and G2, are gaps between mitosis and S phase, and S phase and mitosis, respectively. GI is the primary growth phase; during G2, D N A synthesis is terminated and cell growth continues with accumulation of proteins and organelles to be divided between the two daughter cells during mitosis. The length of the GI phase in highly variable, and can range from several hours to several days, depending on cell type and environmental conditions (36). Cells that persist in GI for extended periods of time enter a distinct (quiescent) state called GO. Cells in GO can re-enter the cycle or remain in GO indefinitely.  10  Two main checkpoint control mechanisms in the cell cycle exist to ensure that cells progress without errors: G l / S (also called the restriction or 'R' point) and G2/M. Upon appropriate stimulation, cells are able to initiate proliferation in both the Gland GO the phases of the cell cycle. In cultured cells, once a cell passes the R point, it is committed to enter S phase, regardless of stimulatory withdrawal. In vivo, however, cells may arrest at different points within G l in response to different inhibitory signals; thus, in reality, there may be several R points in different cell types that restrict cell cycle progression (37). The G2/M checkpoint is much less studied, but is thought to prevent the cell from entering mitosis (M phase) if the genome is damaged, largely through the action of the tumour suppressors A T M and p53 (36). In the following sections, the major regulators of cell cycle progression will be discussed. In general, cell cycle transitions are controlled by the interactions of cyclins and cyclindependent kinases (CDKs) (38). As activation of CDKs is the central event in cell cycle transitions, their activity is quite tightly regulated at several levels (39). The active C D K holoenzyme is composed of a catalytic subunit and the cyclin regulatory subunit. Mammalian cyclin family members include cyclins A to H , which all share a conserved sequence of about 100 amino acids (36). In mammalian cells, the C D K family includes seven members that are conserved in size between 32-40 kDa, and share approximately 40% sequence homology (37). CDKs are expressed at constant levels throughout the cell cycle and, once bound to cyclins, are active serine/threonine kinases (36). Full activation of the cyclin-CDK complex is dependent both on phosphorylation of a conserved threonine in the catalytic cleft by CDK-activating kinase (CAK) and on dephosphorylation of inhibitory threonine sites by phosphatases of the cdc25 family (37).  11  Transition from GI to S phase through the restriction point is mediated by cyclin D- and E-dependent kinases (38). The D-type cyclins ( D l , D2, and D3) bind to CDK4 and CDK6 to create six different  holoenzymes,  which are expressed in tissue-specific patterns  (38).  Hypophosphorylated RB (the retinoblastoma protein) represses the transcription of genes whose products are required for D N A synthesis, largely by binding transcription factors such as the E2Fs (40). Activated cyclin D-CDK4/6 complexes phosphorylate, and thus inactivate R B I , enabling E2Fs to function as transcriptional activators (38). Completion of RB phosphorylation is accomplished by the cyclin E-CDK2 complex which is activated in response to E2F mediated induction of the cyclin E gene (36). This shift in RB phosphorylation from mitogen-dependent cyclin D-CDK4/6 complexes to mitogen-independent cyclin E-CDK2 accounts in part for the loss of dependency on extracellular growth factors at the restriction point (38). Cyclin-dependent kinase activity is tightly regulated by cell-cycle inhibitors, and loss of this regulation is often the cause of human cancer; these inhibitors include the p21 (also known as WAF1/CIP1), p27 (KIP1), p57 (KIP2) proteins (which interact with cyclin-CDK complexes in all phases of the cell cycle), and the INK4 proteins which specifically inhibit cyclin Ddependent  kinases (38).  The CIP/KIP family of polypeptide inhibitors bind complexes  containing cyclins D, E , and A. By binding, these CKIs lead to the inhibition of C D K activity and thus preventing cell cycle progression (36). The INK4 proteins (named for their ability to inhibit CDK4) sequester CDK4/6, preventing binding of CDK4/6 to cyclin D, and thereby indirectly inhibiting cyclin E-CDK2 to ensure cell cycle arrest (38). Not surprisingly, pl6INK4a is a human tumour suppressor because loss of both copies of the gene encoding this protein is known to contribute to tumour formation (36).  12  1.2.6  Apoptosis Programmed cell death plays critical roles in a wide variety of physiologic processes  during fetal development and in adult tissues. Apoptosis is the active mechanism of programmed cell death ((41), reviewed in (42)). Defects in apoptotic cell death contribute to neoplastic diseases by preventing or delaying normal cell turnover, thus promoting cell accumulation. Defects in apoptosis also facilitate tumour progression by rendering cancer cells resistant to death mechanisms relevant to metastasis, hypoxia, growth factor-deprivation, chemotherapy, and irradiation (42). An extensive discussion of apoptosis is not directly relevant to this thesis; however, given its importance to the oncogenic process, a brief summary is presented. Apoptosis is a highly conserved mechanism by which eukaryotic cells commit suicide. Unlike necrosis, apoptosis results from the activation of a genetic program in which cells lose their viability, fragment, and are ingested before losing membrane integrity. The molecular hallmark of apoptosis is the activation of caspases, cysteine proteases that cleave their targets at aspartic acid residues (42). Based on their order of activation, caspases are classified into two families: the initiator caspases and the effector caspases (43). In response to appropriate stimuli, the initiator caspases (also known as 'apical caspases') undergo a complex course of autocatalytic processing and activation, which usually require several auxiliary factors (43). Once activated, an initiator caspase specifically cleaves and hence activates an effector caspase zymogen. For example, the initiator caspase-9 is activated by the assembly of a multimeric complex (the 'apoptosome') involving Apaf-1 and cytochrome c. Once activated, caspase-9 cleaves and activates caspase-3 and caspase-7. Caspases-3 and -7 rapidly degrade a large number of cellular proteins that ultimately kill a cell (44).  13  Apoptosis can be triggered by a wide variety of stimuli: Fas ligand, tumour necrosis factor (TNF), growth factor withdrawal, viral or bacterial infection, oncogenes, irradiation, ceramide, and chemotherapeutic drugs (45). Essentially, there are two major apoptotic pathways in mammalian cells: 1) the death-receptor pathway is triggered by members of the death-receptor superfamily (e.g., CD95 and T N F receptor 1). Binding of the CD95 ligand to the CD95 induces receptor clustering, which recruits multiple procaspase-8 molecules via the adapter molecule F A D D (Fas-associated death domain). Caspase-8 is activated and mediates cleavage on a variety of  substrates critical for cell survival (such as nuclear lamins, D N A polymerases, and  cytoskeletal proteins) (46). 2) The mitochondrial pathway is used extensively in response to extracellular cues and internal insults such as D N A damage (often through p53 induction of Bax, Bcl2-associated X protein). Ultimately, intrinsic and extrinsic pathways unite upon the common final degradation phase, which is responsible for the chromatin condensation and nuclear fragmentation in the apoptotic cell. Many proteins initiate tumour formation or develop therapeutic resistance by manipulating the apoptotic process.  1.3  CANCER The term 'cancer' is used to describe a variety of malignant diseases that result from  uncontrolled cell proliferation. The dividing cells form large masses called neoplasms, or tumours, which can invade neighboring tissues or may metastasize to more distant sites. Although the ancient origins of the term are somewhat uncertain, it most likely derives from the Latin for crab, 'cancer' - presumably because a cancer "adheres to any part that it seizes upon in an obstinate manner like the crab" (47). In contrast, benign proliferations consist of cells that  14  neither invade other tissues nor metastasize. An estimated 145,500 new cases of cancer and 68,300 deaths from cancer will occur in Canada in 2004 (48). Malignant cells have defects in the regulatory circuits that govern normal (controlled) proliferation and homeostasis. Hanahan and Weinberg have proposed that malignant cells exist due to six essential alterations in cellular physiology: self-sufficiency  in growth signals,  insensitivity to growth-inhibitory ('anti-growth') signals, evasion of programmed cell death (apoptosis), limitless replicative potential, sustained angiogenesis, and tissue invasion and metastasis (49). Each of these physiologic changes - perhaps unique in their combination and levels in every tumour sub-type - counteracts homeostatic mechanisms which govern normal cell function. These six capabilities are proposed to be shared in common by most and perhaps all types of human tumours (49).  1.3.1  Mechanisms of Oncogenesis The notion that inherited abnormalities predispose some individuals to cancer was  suggested more than a century ago (reviewed in (50)); advances in molecular biology within the past three decades have enabled us to identify cancer-associated genes. These were found to include both oncogenes (dominant gain-of-function proteins) and tumour-suppressor genes (recessive loss-of-function proteins), and both classes of genes were discovered principally by virtue of their alteration in human cancers (51).  1.3.2  Tumour Suppressor Genes Tumour Suppressor Gene (TSG)s, as their name implies, function to regulate cell growth,  differentiation, or other basic functions in order to prevent unregulated proliferation. Generally  15  speaking, cancers can arise if both copies of a specific T S G are lost or inactivated, although loss of  one  allele  may  be  sufficient  for  malignant  transformation  with  certain  TSGs  (haploinsufficiency) (52). Inactivation of tumour suppressor genes can occur via a variety of mechanisms including loss of the gene (deletion) or mutation (point, missense or nonsense) (53). Tumour-suppressor genes probably evolved to perform specific functions related to development and homeostasis rather than to prevent cancer (53). Interestingly, many tumour suppressors are essential regulators in key signal transduction pathways, but are only involved in specific subtypes of particular tumours (54). Loss of tumour-suppressor function therefore contributes to cancer in some contexts, but leads to pathophysiological states that are distinct from cancer in other contexts (55). Some well-characterized examples of TSGs include APC, RBI, and p53. A brief discussion on the inactivation of TSGs will now be discussed using RBI and p53 as examples. Retinoblastoma Protein (RB) The first T S G to be cloned was the retinoblastoma (RBI) gene, discovered during studies into familial retinoblastoma by Alfred Knudson and others, and resulting in the 'two hit' hypothesis for cancer development (56-58). 'Hits' (mutations) are required for retinoblastoma formation; the first being an inherited (germline) mutation; the second mutation, a somatic one, during the course of development of the retina. The protein product of the RBI gene (pRbl) is responsible for controlling the transition from GI to S in the cell cycle; loss of the key regulator results in uncontrolled cell cycling. Interestingly, patients with germline mutations in one allele of RBI are predisposed to other malignancies including osteosarcomas, soft tissue sarcomas, and melanoma later in life (53). Additionally, RBI also acts as an anti-apoptotic factor through  16  association with p53 (discussed below) (59). Since the discovery of R B I , two other members have been added to the RB gene family: pl07 (60) and RB2/pl30 (61). As with the founding member of this family, both pl07 and RB2 are able to control the cell cycle by negatively modulating the transition between the G l and S phases (described above) (62). The functional activity of RBI is regulated by phosphorylation during the cell cycle (62). Accordingly, RBI appears to be predominantly unphosphorylated or hypophosphorylated in the G l phase of the cell cycle and maximally phosphorylated in G2 (53). Hypophosphorylated RBI represses the transcription of genes whose products are required for D N A synthesis, largely by binding  transcription  factors  such  as  the  E2Fs  (53).  Cyclin  D-CDK4/6  mediated  hyperphosphorylation of RB disrupts the interaction between E2Fs and RB, E2Fs are subsequently released and transcriptional activation ensues (38) (as previously described in section 1.2.5).  p53 The tumour suppressor p53 was originally identified as an oncogene, as overexpression  of p53 appeared to induce cellular transformation (63). However, subsequent studies showed that these p53 proteins under study were in fact missense mutations, and p53 was actually a tumour suppressor gene (53). It is thought to be the most frequently mutated gene in human cancer (64). p53 encodes a nuclear DNA-binding phosphoprotein that normally exists as a tetramer, which binds to specific D N A sequences (59). As suggested by the earliest p53 experiments, the vast majority of p53 mutations in common human cancers are missense mutations, apparently impairing p53's ability to bind specific recognition sequences (64).  A number of other  mechanisms for inactivation of p53 in tumours include deletion of one or both p53 alleles,  17  truncation of the protein by nonsense or splice mutations, functional inactivation through the expression of the E6 gene of human papillomavirus (HPV), and germline p53 mutations (resulting in predisposition to breast cancers, sarcomas, brain tumours, lymphomas and L i Fraumeni syndrome) (53, 65, 66). p53 is a cellular gatekeeper (67); it has numerous cellular functions, including cell cycle regulator ( G i / S and G 2 / M checkpoints), D N A repair, induction of apoptosis, differentiation, sensitivity to chemotherapeutic agents and radiation, and protection from hypoxic and nutrient stress ((59, 65). Although the biochemical mechanisms by which p53 induces apoptosis following insult are not fully understood, it does appear that p53 is involved in both extrinsic (death receptors) and intrinsic (mitochondrial membrane) death pathways (65). p53 activates transcription of a number of genes involved in control of the cell cycle, including the regulator of Cdk activity p21/WAFl/CIPl (68), a growth-arrest DNA-damage inducible gene GADD45 (growth arrest and DNA-damage-inducible 45) (69), the E3 ubiquitin ligase mdm2 (transformed mouse 3T3 cell double minute 2) and the intracellular signal transducer  14-3-3G  (reviewed in (70)). It also stimulates transcription of bcl-2 pro-apoptotic  family members B A X , N O X A , and P U M A (53), L R D D (PIDD, p53-induced protein with death domain) (71) and a number of genes involved in the generation of reactive oxygen species (53, 72). In addition, p53 has been shown to repress the expression of a number of genes and at least some of them, such as cyclin BI and survivin, are negative regulators of apoptosis (70). Finally, p53 is regulated at basal levels through a continuous cycle of ubiquitination and protein degradation through association with mdm2 and PIRH2, a novel ubiquitin ligase that appears to function in parallel with mdm2 (73). Studies targeting ubiquitination as a means to regulate and reactivate p53 in tumours are currently in progress (70). As such, RBI interacts with p53  18  pathways in regulating cell cycle arrest and apoptosis. It has been suggested that disruption of the RB 1 or p53 pathways probably occurs in virtually every human cancer (36).  1.3.3  Oncogenes  Oncogenes, first identified in cancer-causing viruses, are now well established as major contributors to the development of cancer in humans. Oncogenes may be viral in origin or may be derived from normal cellular genes referred to as proto-oncogenes. A recent census of genes mutated and causally implicated in cancer development has identified 291 cancer genes, more than 1% of the human genome (74). Proto-oncogenes are highly conserved in evolution and their products are important in the regulation of cell growth and differentiation in organisms ranging from primitive eukaryotes to humans (75). The expression of cellular proto-oncogenes is tightly regulated in normal cells but, if converted to oncogenes, can induce tumour formation. Oncogenes can be classified into several groups based on the functional and biochemical properties of protein products of their normal counterparts (proto-oncogenes); these include growth factors, growth factor receptors, signal transducers, and transcription factors (Table 1). Conversion of proto-oncogenes into oncogenes can occur by several mechanisms including proviral insertion, gene amplification, point mutation, and chromosomal rearrangement (discussed in more detail below). Activation of oncogenes by proviral insertion is complex and involves recombination between viral and cellular genomes following infection and integration of the virus into the cell. In this manner, viral sequence is integrated adjacent to the cellular proto-oncogene resulting in alterations that convert the normal gene to its oncogenic counterpart (76). The first described example of proviral insertion leading to oncogene activation was the 50- to 100- fold elevation  19  of c-myc transcription observed in bursal lymphomas induced by avian leukosis virus (77). In this situation, viral sequence insertion leads to enhanced and unregulated expression of the cmyc proto-oncogene (leading to uncontrolled cellular proliferation), thus converting it to an oncogene. A second mechanism of oncogene activation observed both in transformed cells and in tumours is gene amplification, the expansion in gene copy number. The process of gene amplification occurs through redundant replication of genomic D N A , potential creating karyotypic abnormalities called double minute chromosomes and homogenous staining regions (78). Amplification leads to the increased expression of genes, often resulting in a selective advantage for cell growth (75). Amplified proto-oncogenes are found in human tumours, often in tumour-specific patterns, and the presence of multiple copies of proto-oncogenes in tumour cells is associated with a poor prognosis (e.g., c-myc in neuroblastoma) (79). Gene amplification of two other proto-oncogenes - c-erbB2 (HER2/neu) in breast cancer (80) and Ras family members in colorectal (81) and esophageal cancers (82) - are also well established. Mutations can also activate proto-oncogenes through structural alterations in their encoded proteins. These alterations, which usually involve critical protein regulatory regions, often lead to the uncontrolled, continuous activity of the mutated protein (75). Various types of mutations are capable of activating proto-oncogenes, such as base substitutions (point mutations), deletions, and insertions (83). It has been estimated that as many as 15-20% of all human tumours may harbor a mutation of a Ras family member, with some tumour subtypes (e.g., pancreatic carcinoma) containing mutations in 87% of tumours analyzed (84). In this instance, such single-base mutations alter the amino acid sequences of the Ras proteins, decreasing  20  T A B L E 1. Proto-oncogenes can be activated to become oncogenes through a variety of mechanisms. Selected oncogenes are presented below (adapted from Pierotti et al, (75)). Oncogene PDGFB FGF4 MET EGFR  Mechanism of Activation Constitutive production Constitutive production Constitutive activation  Protein Function  Neoplasm  B chain PDGF FGF family member Receptor tyrosine kinase Receptor tyrosine kinase  Glioma / fibrosarcoma Kaposi's sarcoma Renal carcinoma  Receptor tyrosine kinase Receptor tyrosine kinase Receptor tyrosine kinase Receptor tyrosine kinase Tyrosine kinase Tyrosine kinase  Leukemia ( C M L / A M L )  PDGFR  Gene Amplification / Increased Protein Gene rearrangement  NTRK  Constitutive activation  ERBB2  Gene amplification  RET  Constitutive activation  SRC ABL  Constitutive activation Constitutive activation  MOS  Constitutive activation  RAF  Constitutive activation  PIM-1  Constitutive activation  U-RAS  Point mutation  Serine/threonine kinase Serine/threonine kinase Serine/threonine kinase G Protein  K-RAS  Point mutation  G Protein  N-RAS DBL OST CRK N-MYC C-MYC  Point mutation D N A rearrangement D N A rearrangement Constitutive activation Gene amplification Gene amplification  G Protein GEF GEF SH2/SH3 adapter Transcription factor Transcription factor  MDM2 CCND1 FOS JUN  Gene amplification Gene amplification Deregulated Activity Deregulated Activity  E3 ubiquitin ligase Cell cycle regulator Transcription factor Transcription factor  Squamous cell carcinoma  Colon, thyroid, & breast cancer, fibrous tumours Breast cancer Neuroblastoma Thyroid cancer Colon carcinoma Leukemia (CML/ALL) Sarcoma Sarcoma T-cell lymphoma Colon, lung, pancreas carcinomas A M L , thyroid cancer, melanoma Melanoma Lymphoma osteosarcoma Lung, breast carcinomas Neuroblastoma; lung Many neoplasms Burkitt's lymphoma Sarcoma Breast cancer Osteosarcoma Sarcoma  G E F = Guanine Exchange Factor; A M L = acute myeloid leukemia; C M L = chronic myelogenous leukemia; FGF = fibroblast growth factor; PDGF = platelet-derived growth factor.  21  intrinsic GTP-ase activity and constitutive activation of Ras and its downstream proliferative pathways (85).  1.3.4  Chromosomal Rearrangements It is generally accepted that chromosomal rearrangements occur more frequently in  malignant cells than in normal ones, presumably due to the significant amount of genomic instability found in many cancers (86). The most common mutation class among the known cancer genes is a chromosomal translocation (74). Chromosome translocations result from deoxyribonucleic acid (DNA) double-strand breakages in two or more chromosomes, followed by reciprocal exchange of the chromosomal segments. Two patterns  of  chromosome  translocation have been observed in human cancers (Figure 1). The first pattern is observed in many solid tumours and can result in gains and losses of large portions of chromosomal material. These complex translocations appear to be random and are not tumour-specific (87). The second pattern of chromosomal translocation (referred to as the simple type) is characterized by distinctive rearrangements of chromosomal segments in disease-specific manners. These tumourspecific translocations are often felt to be causal events in the development of these tumours, and are often found in sarcomas and leukemias and rarely in carcinomas (Table 2). The overall pattern of specificity of translocations has two possible explanations: 1) certain genomic regions (known as breakpoint cluster regions) are located in regions of open chromatin in actively expressed genes and are therefore especially vulnerable to breaks, or 2) breaks and translocations occur randomly throughout the genome, and only those which provide a clonal advantage lead to disease (88). It is probable that a combination of both hypotheses is correct.  22  The exact mechanism(s) by which chromosomal translocations arise is still poorly understood, but double-stranded D N A breaks are considered an important step in this process. These may be induced by both exogenous and endogenous agents (86). Endogenous processes associated with D N A double-strand breakage include intrachromosomal rearrangement as the immunoglobulin or T cell receptor loci, meiotic recombination between homologous chromatids, malfunctions of the D N A repair process and topoisomerase enzymes, and the production of D N A damaging agents such as free radicals (86, 88). Exogenous insults known to induce D N A double-strand breakage include ionizing radiation and chemotherapy (86). There are two different mechanism by which a translocation can give rise to a malignant cell: 1) the transcriptional activation of proto-oncogenes (as evidenced by the t(8;14)(q24;q32) found in 85% of cases of Burkitt's lymphoma, placing the c-myc gene under the regulatory control of the immunoglobulin heavy chain locus (89)); 2) deletion of a tumour suppressor gene or fusion of two unrelated gene in frame, creating a novel fusion gene with oncogenic potential (such as the bcr-abl fusion resulting from a t(9;22)(q34;qll) in chronic myelogenous leukemia (90)). Fusion genes encode chimeric proteins with transforming activity. In general, both fusion partners contribute to the transforming potential of the chimeric oncoprotein (75). For example, the ETV6-PDGFR(3 fusion protein induces transformation through constitutive activation of the tyrosine kinase domain of PDGFR(3, facilitated by the dimerization domain of ETV6.  23  POSSIBLE RESULTS O F C H R O M O S O M A L T R A N S L O C A T I O N :  Breaks in heterochromatin  No observed effect  Deletion of gene 1 or gene 2  Potential disruption of tumour suppressor  Stop codon created in gene 1 or gene 2  Truncated Protein  Out of frame fusion between genes 1 & 2  Non-functional protein  In-frame fusion between gene 1 & gene 2  Functional chimeric fusion protein (e.g., ETV6NTRK3)  F I G U R E 1. Chromosomal translocations can result in rearrangement through a number of different mechanisms.  a potentially  oncogenic  gene  24  T A B L E 2. Recurrent Chromosomal Translocations in Soft-Tissue Sarcoma. Abbreviations: DFSP, dermatofibrosarcoma protuberans; DSRCT, desmoplastic small round-cell tumour. TUMOUR  CYTOGENETIC EVENT  MOLECULAR EVENT,..,  FREQUENCY  DIAGNOSTIC UTILITY? ....  Alveolar Soft Part Sarcoma  t(X;17)(pll;q21)  ASPL-TFE3  >90%  Yes  Angiomatoid Fibrous t(12;16)(ql3;pll) Histiocytoma Clear Cell Sarcoma t(12;22)(ql3;ql2)  fusion FUS-ATF1  fusion  ?  Yes  EWS-ATF1  fusion  >75%  Yes  ?  No  DFSP  t(17;22)(q22;ql3) COL1A1-  DSRCT  t(ll;22)( 13;ql2)  EWS-WT1 fusion  95%  Yes  Endometrial Stromal Tumour  t(7;17)(pl5;q21)  JAZF1-JJAZ1  30%  Yes  Ewing's Sarcoma  t(ll;22)(q24;ql2) t(21;22)(ql2;ql2) t(2;22)(q33;ql2) t(7;22)(p22;ql2) t(17;22)(ql2;ql2) t(12;15)(pl3;q25)  EWS-FLI1  >85% 10-15% <1% <1% <1% >90%  Yes Yes Yes Yes Yes Yes  t(12;15)(pl3;q25)  ETV6-NTRK3  >90%  Yes  Iriflarnmatory Myofibroblastic Tumour Liposarcoma (Myxoid)  t(2;var)(p23;var)  ALK fusion genes  >50%  Yes  t(12;16)(ql3;pll) t(12;22)(ql3;ql2)  TLS-CHOP EWS-CHOP  95% 5%  Yes Yes  Rhabdomyosarcoma (Alveolar)  t(2;13)(q35;ql4)  PAX3-FKHR  >75%  Yes  t(l;13)(q36;ql4) double minutes t(X;18)(pll;qll)  PAX7-FKHR  10-20%  Yes  >90%  Yes  >85%  Yes  PDGFB  Congenital Fibrosarcoma Congenital Mesoblastic Nephroma  P  fusion  fusion fusion  EWS-ERG fusion EWS-FEV fusion EWS-ETV1 EWS-E1AF  fusion fusion  ETV6-NTRK3  fusion fusion  fusion  fusion  Synovial Sarcoma  fusion fusion SYT-SSX1 or SYT-  SSX2 fusion  Extraskeletal Myxoid Chondrosarcoma  t(9;22)(q22;ql2)  EWSR1-TEC  25  Chromosomal translocations are specific to certain biologically distinct subtypes of cancer. Moreover, balanced translocations tend to be found in leukemias as well as pediatric sarcomas; conversely, epithelial carcinomas typically have intra-clonally diverse karyotypes (such as deletions and unbalanced chromosome translocations), usually indicative of excessive chromosome instability (88). However, a recent study by Mitelman et al. has proposed that karyotypes of epithelial tumours are too complex to allow for a detailed characterization; this excessive 'noise' has precluded identification of recurrent genetic lesions, and it is a combination of small sample size and difficulty in obtaining completely characterized karyotypes in these tumours that has biased the investigator against the importance of deregulated and rearranged genes in these tumours (91).  Additionally, tissue-specific differences  in mechanisms of  tumourigenesis may also contribute to the paucity of important structural aberrations in solid tumours (92).  For example, recombination is an essential step in the development  of  hematopoietic progenitor cells in which leukemias arise. This may allow for genes to be activated by recombination-mediated translocations in leukemias, whereas they are activated more often by amplification in solid tumours (92).  1.3.5  Fusion genes Both hematologic malignancies and solid tumours express chimeric proteins from gene  fusions. In solid tumours, one of the genes involved often encodes for a transcription factor. The fusion protein consequently functions as a chimeric transcription factor that is able to recognize the same D N A sequences as the parent transcription factor. However, transcriptional activation would be altered due to contributions from the domains of the partner gene (93). Two examples  26  are the t(ll;22)(q24;ql2) EWS-FLI1  (94, 95) and t(21;22)(q22;ql2) EWS-ERG  (96) fusions  identified in Ewing sarcoma. Whereas EWS is a ubiquitously expressed gene of unknown function, both FLU and ERG are members of the ETS family of transcription factors. The fusion of EWS to either of these two genes juxtaposes the EWS transcriptional-activation domain with the FLU or ERG ETS DNA-binding domain, while the EWS RNA-binding domain is lost (9496). The classic example of the fusion oncogene is the Philadelphia chromosome (Ph ) 1  identified by Rowley in virtually all patients with chronic myelogenous leukemia (97). Ph is the 1  derivative  Chromosome  22  that  results from  a t(9;22)(q34;qll)  rearrangement.  This  translocation fuses the c-ABL proto-oncogene on Chromosome 9 with the breakpoint cluster region (BCR) gene on Chromosome 22. Expression of the B C R - A B L fusion protein results in increased A B L kinase activity and in the malignant transformation of hematopoietic cells (98, 99). Interestingly, the t(9;22) is also found in up to 20% of cases of adult acute lymphoblastic leukemia (ALL) and the majority of cases of pediatric A L L (100). In these cases, the breakpoint in the bcr gene differs from that found in C M L , resulting in a 185 kDa bcr-abl fusion protein (as opposed to the 210 kDa product in CML) (100); both fusion proteins have transforming ability. Interestingly, certain genes such as M L L (mixed-lineage  leukemia) and ETV6 are  promiscuous, and form chimeric fusions with a large number of partners (101). The M L L gene fuses with over 30 diverse partner genes (102). ETV6 is widely expressed in hematopoietic and non-hematopoietic  tissues (103,  104), although ETV6 rearrangements generally occur in  hematopoietic malignancies (Table 3). ETV6 is involved in fusion oncogenes with many 5' and 3' partners; the majority of these breakpoints occur after ETV6 exon 5. Two exceptions to this are E T V 6 - T T L (exon 1 breakpoint) (105) and ETV6-CDX2 (with the E T V 6 breakpoint after  27  exon 2) (106). These proteins would lack the ETV6 helix-loop-helix domain involved in breakpoints after exon 5, and if in fact are tumourigenic, would presumably involve the ETV6 promoter. To date, NTRK3 is the only non-hematopoietic fusion partner identified for ETV6. Gene fusions produce tumour-specific molecules as the chimeric R N A and protein product occur only in cells with the translocation and are (presumably) malignant. Accordingly, these  unique  molecules  are  potential  tumour-specific  therapeutic  targets  that  would  correspondingly reduce treatment side effects (107). The most popular and seemingly easiest approach is targeting of the fusion protein with small molecules. The most well known example of this is the anti-kinase drug STI571 (Gleevec), targeted against the kinase activity of the A B L portion of B C R - A B L (108).  1.3.6  The Laboratory Mouse as a Model System for Cancer In vivo models of cancer development are needed for a comprehensive understanding of  the biology of cancer (120). Our understanding of the mechanisms by which cancer initiates and progresses has been advanced by the development of mouse models of tumourigenesis (121). Analysis of transgenic mice in which oncogenes are overexpressed has been complemented by the creation of mice carrying germline mutations in tumour suppressor genes. The generation of mice predisposed to cancer has allowed scientists to establish direct causal links between the mutation of individual genes and specific cellular changes that lead to cancer. Further, these models provide the investigator with a in vivo system for studying both tumour development, as well as the design of directed therapeutic modalities. Mice with reduced E R B B receptor function show strikingly similar disease patterns to treatment side effects in patients receiving  28  T A B L E 3. The ETV6 locus is involved in a range of chromosomal translocations. More than 40 different fusion partners have been reported; selected ones are presented below. Abbreviations: T K = Tyrosine Kinase, S A M = Pointed Domain, ORF = Open Reading Frame, C M M L = Chronic Myelomonocytic Leukemia, M P D = Myeloproliferative Disease, aCML = atypical Chronic Myelogenous Leukemia, CFS = Congenital Fibrosarcoma, C M N = congenital mesoblastic nephroma, A M L = Acute Myelogenous Leukemia, M D S = Myelodysplastic syndrome.  TRANSLOCATION  FUSION GENE  STRUCTURE  DISEASE  REF  t(9;12)(q34;pl3)  ETV6-ABL  SAM-TK  CMML  (109)  t(5;12)(q33;pl3)  ETV6-PDGFR/3  SAM-TK  CMML  (103)  t(3;12)(q26;pl3)  ETV6MDS1/EVI1  S A M - unknown  MPD  (110)  t(4;12)(qll-ql2;pl3)  BTL - ETV6  B R X like SAM+DNA B D  AML  (111)  t(9;12)(p24;pl3)  ETV6-JAK2  SAM-TK  A L L , aCML  (112)  t(12;15)(pl3;q25)  ETV6 - NTRK3  SAM-TK  A M L , CFS, CMN  (2)  t(12;22)(pl3;q22)  ETV6 - MN1  SAMglutamine repeat  MPD, meningioma  (113)  t(12;13)(pl3;ql2)  ETV6 - CDX2  No ETV6 domain - homeobox  AML  (106)  t(6;12)(q23;pl3)  ETV6-STL  No functional significance  ALL  (114)  t(5;12)(q31;pl3)  ACS2 - ETV6  No functional significance  A M L , MDS  (115)  t(l;12)(q25;pl3)  ETV6-ARG  SAM-TK  AML  (116)  t(12;21)(pl3;q22)  ETV6 - AML1  S A M - runt  ALL  (117)  t(l;12)(p36;pl3)  ETV6-MDS2  unknown  MDS  (118)  t(12;13)(pl3;ql4)  ETV6-TTL  Both transcripts expressed  ALL  (105)  t(9;12)(q22;pl2)  ETV6-SYK  SAM-TK  MDS  (119)  29  ERBB-targeted drugs (122), suggesting that genetically engineered mouse models can serve as valuable tools to predict targeted therapy toxicity and identify mutations that predispose individuals to side effects. The  development  of appropriate mouse models would significantly enhance our  understanding of translocation-bearing sarcomas (123). To date, it has proven difficult to recapitulate the human disease using animal models. The most successful model so far is the FUS-DDIT3 (also known as TLS-CHOP) fusion gene found in myxoid liposarcoma (124). This fusion functions as an aberrant transcription factor involved in adipocyte differentiation and growth arrest (125). In this traditional transgenic (random insertion) model, mice from two independent founder lines developed transplantable tumours of white fat that morphologically resemble human liposarcoma. Interestingly, tumours only develop in fat tissues, despite the expression of the Tls-Chop transgene in most tissues (124). These finding are consistent with the theory that the transforming effect of the fusion gene is restricted to very specific cell lineages, due to content-dependent activity of the fusion gene (123).  1.4  T  H  E  ETV6-NTRK3  C H I M E R I C  O N C O P R O T E I N  "Sarcomas are relatively rare malignant tumours derived from mesenchymal tissues - nonepithelial tissues derived from the embryonic mesodermal layer" (123). However, pediatric sarcomas are responsible for considerably higher morbidity and mortality than those affecting adults. It has become apparent that specific molecular alterations associated with particular tumour subtypes are more important diagnostically and prognostically than earlier classifications based on the site of the tumour (126). Sarcomas can be stratified into two groups, based on cytogenetic and molecular rearrangements. One group consists of genetically complex sarcomas,  30  more typically affecting older patients than the second group. These complex chromosomal rearrangements, without any recurrent reciprocal translocations, present a great challenge to the cancer researcher (123). The second group is characterized by relatively simple karyotypes, often involving disease-specific chromosomal translocations such as the well-studied  PAX3-FKHR  fusion in alveolar rhabdomyosarcoma (127), and the ETV6-NTRK3 gene fusion that is the subject of this thesis. The ETV6-NTRK3 gene fusion was first identified by cloning of the t(12;15)(pl3;q25) translocation in congenital (or infantile) fibrosarcoma (2), a mesenchymal malignancy of very young children (see below). This rearrangement fuses the N-terminal S A M domain of ETV6 to the C-terminal PTK domain of NTRK3 (also known as TrkC), generating a fusion protein that is similar in structure to other ETV6 chimeric PTKs (Figure 2). NTRK3 is the transmembrane surface receptor for neurotrophin-3 and is primarily expressed in the central nervous system where it is involved in growth, development, and cell survival of neuronal cells (reviewed in (128). The ETV6-NTRK3 (EN) fusion protein has potent in vivo and in vitro transforming activity in several cell lineages including fibroblasts (129), hematopoietic cells (130), and breast epithelial cells (131). E N induces phenotypic transformation, soft agar colony formation, and nude mouse tumours when expressed in NIH3T3 as well as Scg6 and Eph4 breast epithelial cells (131). Transformation of each cell type requires an active P T K domain as well as an intact S A M oligomerization domain. Mutations of the A T P binding site or any of the three activation loop tyrosines of the NTRK3 PTK completely abolish transformation activity, as do deletions of the S A M domain (129). These findings are consistent with the notion that E N self-association through its S A M domain leads to P T K activation and stimulation of downstream signaling pathways required for transformation.  A9  ETV6  NH — 2  SAM  DNA binding~|—  COOH  V112  KFG S H C TM  NTRK3  N H  2  ^ '  ligand binding  PTK  |ATP binding  A9<  ETV6-NTRK3  NH  SAM V112  H ATP Binding K380  hr C O O H  Activation Loop Tyrosines  PTK ft Activation Y 6 1 5 Loop Y513.Y517, Y518  PLCy Binding Tyrosine  COOH PLCy Binding Tyrosine  F I G U R E 2. Schematic diagram of the E T V 6 - N T R K 3 (EN) fusion protein. E N contains the S A M dimerization domain of E T V 6 fused to the protein tyrosine kinase (PTK) domain of N T R K 3 . Note that E N does not contain the transmembrane domain or SHC and K F G binding sites of wild-type N T R K 3 .  32  1.5 ETV6 ETV6 (ETS, E-26 transforming specific, variant gene 6) is a member of the ETS family of transcriptional regulators, and is thought to play a role in early hematopoiesis and angiogenesis (104,  132-136). The Etv6  gene is widely expressed throughout embryonic  development and in the adult (132). Murine gene targeting experiments have shown that ETV6~'~ are embryonic lethal, with defective yolk sac angiogenesis and intra-embryonic apoptosis of mesenchymal and neural cells (104). ETV6, formerly abbreviated T E L (translocation ETS leukemia), encodes the sterile alpha motif in exons 3 and 4 and the ETS D N A binding domain in exons 6 to 8 (132) (Figure 2). ETV6 preferentially binds to the sequence T(G/T)(A/C)GGAAGT (137) and functions as a transcriptional repressor via interactions with the mSin3A, N-CoR, and SMRT co-repressors, as well as histone deacetylase (138, 139). E T V 6 represses the matrix metalloproteinase Stromelysin-1  (140)  and has been shown to induce apoptosis through  repression of Bcl-XL (141). ETV6 can inhibit Ras-dependent colony growth in soft agar and hinder proliferation in a range of cell types (140, 142). Further, loss of heterozygosity at chromosome 12pl3 (ETV6) is found in many types of malignancies, including leukemias and tumors of the breast and ovary (103, 143-146). Clearly E T V 6 may play a role as a tumour suppressor.  1.5.1  The ETV6 Gene as a Target of Chromosomal Translocations For  reasons  that remain unclear ETV6  is frequently  targeted by chromosomal  translocations in human malignancies, particularly leukemias, resulting in the expression of oncogenic ETV6 gene fusions (Table 3). Chimeric oncoproteins often contain the S A M (sterile  33  alpha motif; also known as 'pointed'or helix-loop-helix, HLH) oligomerization domain of ETV6 fused to either a D N A binding transcription factor such as A M L 1 (117, 147), or more commonly to a PTK domain such as that of PDGFRp (103), A B L (109, 148), JAK2 (112, 149, 150), A R G (116, 151), or FGFR3 (152)). These chimeric proteins have predominantly been discovered in human leukemias, and expression appears to be sub-type specific (153). The well-characterized ETV6-CBFA2 (TEL-AML1) fusion is often accompanied by deletion of the normal ETV6 allele on Chromosome 12 (154). The fusion of E T V 6 to NTRK3 in congenital fibrosarcoma (CFS) is the first report of this ETS gene being involved in a human solid tumour (2).  1.6 NTRK3 The development and survival of the mammalian nervous system is highly dependent on the existence of soluble neurotrophic factors (155). These neurotrophins recognize two different classes of receptors, the N T R K family of protein tyrosine kinases and the low-affinity p75 death receptor, a member of the tumour necrosis factor (TNF) receptor superfamily. The N T R K family of protein tyrosine kinases includes NTRK1, NTRK2, and NTRK3, which were formerly called T R K A , T R K B , and T R K C , respectively. NTRK1 is the N G F receptor, NTRK2 serves to bind brain-derived neurotrophic factor (BDNF), and NTRK3 is the primary receptor for neurotrophin3 (NT-3) (156,  157). NT-3 can also bind to and activate NTRK1 and NTRK2 at high  concentrations (155). N T R K family members are very similar at the amino acid level; in fact, their protein tyrosine kinase (PTK) domains are over 70% homologous (158). N T R K family members undergo ligand-induced dimerization and autophosphorylation that  subsequently  activate signal transduction cascades (159). The NTRK2 and NTRK3 genes have been shown to  34  engage in alternative splicing that results in non-catalytic receptor isoforms whose functions remain unknown. NTRK3 encodes the transmembrane surface receptor for neurotrophin-3 (NT-3) and mediates many aspects of growth and development in the central nervous system (155, 160, 161). There are two human NTRK3 splice variants: active NTRK3 consists of 825 amino acids but a non-functional variant contains a 14 amino-acid insert occurring after NTRK3 arginine 701 (162). Both variants are equally proficient in binding NT-3 and in autophosphorylation; however, only NTRK3 lacking the insert can mediate signal transduction (163). NTRK3 is activated by cell surface ligand-mediated oligomerization, which facilitates autophosphorylation of NTRK3 cytoplasmic tyrosine residues and subsequent kinase activation (155, 158, 164). Recruitment of N T R K effectors to the plasma membrane initiates signaling events that promote growth, differentiation, and survival. Phosphorylated NTRK3 tyrosine 516 binds the SHC (Src-homology and collagen) adapter protein via the SH2 domain of the latter (163, 165), and is also the site of association with the regulatory (p85) subunit of PI3K (phosphoinositol-3' kinase) (165, 166). Subsequently, SHC activates the Ras-MAPK signaling cascade, while p85 activates the pllO catalytic subunit of PI3K and the A K T survival pathway. Phosphorylated NTRK3 tyrosine-820 is the binding site for phospholipase-Cy (PLCyl) (165, 167). Additionally, the sue 1-associated neurotrophin factor target (SNT) protein has been shown to bind to an NTRK3 juxtamembrane K F G sequence (lysine - phenylalanine - glycine; NTRK3 residues 461-463) and become tyrosine phosphorylated (168, 169). Other molecules are known to associate with activated NTRK3 and potentially be involved in NTRK3 signaling: these include the SH2 domain-containing protein SH2B(3, and the rat homologue for an adapter molecule containing a Pleckstrin homology domain and an SH2 domain (rAPS) (170).  35  1.6.1  NTRK Expression in Other Human Tumours A number of reports highlight a potentially more general role for N T R K receptors in  oncogenesis. NTRK1  (TRKA) sequences were originally isolated from a colon carcinoma biopsy  as part of an oncogene encoding the amino terminal portion of tropomyosin (TPM3) fused to a truncated tyrosine kinase receptor. TPM3-NTRK1  fusions  were subsequently detected in  papillary thyroid carcinomas (171), and altered N T R K signaling has been implicated in other neoplasms (reviewed in (172)) including pancreatic adenocarcinoma (173). Reuther et al. have recently identified an activating NTRK1 mutation in a case of A M L (174), and others have described potential roles for N T R K signaling in prostatic cancer cell survival (175) and invasion (176), as well as in breast cancer cell proliferation (177). Recent reports describe NTRK3 expression in 44 / 51 human soft tissue sarcomas (178), and NTRK3 mutations that potentially activate the P T K in colorectal carcinomas (8). Expression of either NTRK1 or NTRK3 is a marker of favorable prognosis in neuroblastomas (179) and medulloblastomas (180). Therefore N T R K proteins appear to have oncogenic activity in several lineages, and their abnormal expression (i.e., in a non-neuronal tissue) may result in increased N T R K signaling in an environment lacking regulatory mechanisms. Elucidation of pathways activated by E N may provide novel insights into how N T R K signaling contributes to oncogenesis.  1.7 EXPRESSION OF E N FUSION TRANSCRIPT IN HUMAN MALIGNANCIES Congenital  Fibrosarcoma  Congenital fibrosarcoma (CFS) is a pediatric spindle cell tumour of the soft tissues that occurs almost exclusively before two years of age. Although these tumours display histologic features of malignancy and frequently recur, they have a relatively good prognosis and only  36  rarely metastasize (181). CFS is difficult to distinguish histologically  from adult-type  fibrosarcoma, which has a worse prognosis and often metastasizes (181). Moreover, CFS can resemble  several benign but cellular fibroblastic lesions of infancy, including infantile  fibromatosis and myofibromatosis (182). ETV6-NTRK3  transcripts were absent in adult-type  fibrosarcoma, infantile fibromatosis, and myofibromatosis as well as other histologically similar spindle cell neoplasms (1). Expression of the ETV6-NTRK3  gene fusion hence appears to be  specific for CFS among childhood soft tissue tumours and is a useful diagnostic tool for these tumours (126).  Congenital Mesoblastic  Nephroma  Several groups have also identified the E N fusion in another pediatric solid tumour, the cellular variant of congenital mesoblastic nephroma (CMN) (183-185). ETV6-NTRK3  transcripts  are not present in so-called classical C M N , but are observed in the mixed form of this disease (183, 184)). The salient clinical features of cellular C M N , including its excellent prognosis and occurrence in very young children, overlap with those of CFS, and a relationship between CFS and congenital  mesoblastic  nephroma has been  proposed based on morphologic and  ultrastructural similarities (186). The presence of ETV6-NTRK3 that cellular congenital mesoblastic  transcripts support the concept  nephroma is histogenetically  related to CFS (187).  Interestingly, in several studies, virtually all cases of CFS and cellular C M N cases expressing the ETV6-NTRK3  gene fusion also demonstrated an extra copy of Chromosome 11 (183, 184). One  possible explanation is that trisomy 11 provides an additional copy of the insulin-like growth factor 2 gene (IGF2), which is localized to chromosome llpl5.5 and has been shown to bind the  37  IGF-I receptor, provoking an anti-apoptotic signaling cascade (188). The relevance of this finding is discussed in detail below.  Secretory Breast  Carcinoma  Tognon et al. recently reported that expression of the ETV6-NTRK3  gene fusion occurs in  human secretory breast carcinoma (SBC), a rare subtype of infiltrating ductal carcinoma (IDC) (131). SBC was originally described in children but is now known to occur in the adult population as well. SBC is generally considered to have a favorable prognosis (189,  190).  Although recurrences and nodal metastases have been observed in both male and female cases (191), distant metastases are extremely rare (192). Therefore the prognosis for SBC was initially thought to be excellent compared to typical IDC, with an up to 100% five year survival rate (193, 194). However, more recent studies have suggested that the favorable outcome is age related, and that in older patients the prognosis is similar to typical IDC (195). The ETV6-NTRK3  gene  fusion appears to be specific for secretory breast carcinoma, and was not detected in any cases of EDC (131). Further, the E N chimeric protein can transform normal mouse mammary epithelial cells, strongly implicating this oncoprotein as a primary genetic lesion in the development of secretory breast carcinoma (131).  Acute Myelogenous  Leukemia  A variant of the ETV6-NTRK3  gene fusion has been reported in a single case of acute  myelogenous leukemia in an adult patient (out of 100 cases studied) (196). This chimeric transcript consists of exons 1-4 of ETV6 fused to exons 13-18 of NTRK3, thereby differing from the originally described fusion transcript by lacking ETV6 exon 5. This variant, which can  38  presumably still dimerize resulting in constitutive PTK activity, has been shown to transform murine hematopoietic cells and induce similar signaling transduction cascades (130, 197). It is possible that this so-called 'Tel-TrkC )' variant may exist as a second form of (L  ETV6-NTRK3  that  is specific for acute myeloid leukemia. Neither variant was detected in a screen of 13 pediatric A M L and 58 acute lymphoblastic leukemia cases (198). However, full-length E N cDNA is able to confer IL-3 independence to murine hematopoietic cells (Martin and Sorensen, unpublished results) (130), and so the involvement of E N in hematologic malignancies may be worthy of further investigation. E N is the first translocation-associated fusion protein to be identified in mesenchymal, epithelial, and hematopoietic malignancies. This calls into question the prevailing view that fusion genes show strict tumour specificity. In fact, several other chimeras have been reported in distinct tumour subtypes. The chimeric protein TPM3-ALK, been  documented  in  both  anaplastic  large  myofibroblastic tumours (200), while CTLC-ALK  cell  resulting from a t(l;2)(q25;p23), has  lymphoma  (199)  and inflammatory  was demonstrated in large cell lymphoma (201)  and inflammatory myofibroblastic tumour (202). This divergence is not limited to chimeric PTKs, as FUS-ERG,  originally described in acute myeloid leukemia (203), encodes a chimeric  transcription factor and has recently been detected in Ewing tumours (204). Moreover, the ASPL-TFE3 chimeric transcription factor of alveolar soft part sarcoma (205) is also expressed in forms of renal cell carcinoma (206).  1.8  E N SIGNAL TRANSDUCTION The ETV6-NTRK3 oncoprotein is similar to other chimeric PTKs such as B C R - A B L  (207), ETV6-PDGFR(3 (208), ETV6-JAK2 (149), E T V 6 - A B L (109), and N P M - A L K (209) in  39  that it functions as a constitutively active tyrosine kinase. E N is capable of homodimerization (or heterodimerization with endogenous ETV6) via the E T V 6 S A M domain (129). This domain therefore mediates ligand-independent dimerization and subsequent P T K activation (129, 130, 210). E N expression leads to constitutive elevation of cyclin D l mRNA and protein levels, and EN-induced cyclin D l expression correlates with increased cell cycle progression in fibroblasts (210) and breast epithelial cells (Tognon and Sorensen, unpublished results). Moreover, E N expression leads to constitutive activation of two of the major effector pathways of wild-type NTRK3, namely the Ras-MAPK mitogenic pathway and the PI3K pathway leading to activation of the A K T cell survival factor (210). These effects occur even under serum-free conditions. Phenotypic transformation, soft agar colony formation, and tumourigenesis in nude mice by E N are blocked by inhibition of either Mekl  or PI3K (210). However, we failed to detect  interactions between E N and adapter molecules known to link NTRK3 to Ras-MAPK and PI3KA K T pathways such as SHC, GRB2, SH2Bp\ or the PI3K p85 subunit (129), as well as A B L , SRC, and SHJP2 (unpublished data). Wild-type NTRK3 proteins utilize juxtamembrane tyrosine Y485 to interact with several of these adapters, but this residue is not present in the E N oncoprotein due to the position of the fusion point (129). As will be described below, evidence suggests that the adapter molecule linking E N to Ras-MAPK and PI3K-AKT is insulin receptor substrate-1 (IRS-1). The unique ability of E N to activate both Ras-MAPK and PI3K-AKT pathways may be crucial to its oncogenic activity (Figure 3). While each pathway has been shown to individually contribute to oncogenic signaling, there is increasing evidence for a synergistic effect of these two pathways in transformation (31, 35). Stimulation of both pathways simultaneously may allow for appropriate activation of either pathway, without induction of an anti-tumourigenic  40  (apoptotic) response. For instance, it has been shown that continuous Ras activity will induce cell cycle arrest or apoptosis unless the PI3K-AKT survival pathway is activated concomitantly (reviewed in (25)). Further, these pathways may be acting in concert to continuously activate cyclin DI and drive cell cycle progression (210). While our studies indicate that EN-induced constitutive elevation of cyclin as well as increased cell cycle progression is mediated predominantly by the Ras-MAPK pathway, the PI3K-AKT also appears to play a role in these processes (210).  1.8.1  Role of the Insulin-Like Growth Factor 1 Receptor Signaling Axis in E N Transformation  Accumulating literature points to a critical role for the IGF1R axis in cellular transformation (211, 212). Moreover, up-regulation of the IGF1R pathway has been observed in a number of pediatric and adult neoplasms, either through over-expression of the ligands (IGF1 or 2) or activation of the IGF1R itself (211, 213, 214). It has been demonstrated that there is elevated expression of the IGF2 gene in CFS and C M N tumours (215), and that the IGF1R pathway must be intact for E N transformation (215). E N fails to transform mouse embryo fibroblasts derived from mice with a targeted disruption of the IGF1R gene, but re-introduction of IGF1R into these cells restores E N transformation activity (215). E N is therefore similar to other dominantlyacting oncoproteins such as activated Ras, c-Src, SV40 large T antigen, and over-expressed receptor PTKs in failing to transform IGF1R null cells (reviewed in (213)), including the childhood tumour associated oncoproteins EWS-FLI1 and PAX3-FKHR of Ewing tumour and  IGFs  F I G U R E 3. Outline of E N signal transduction. E N undergoes multimerization in order to become phosphorylated and active. This results in constitutive activation of both the Ras-MAPK proliferation and P I 3 K - A K T survival pathway, through an interaction with IRS-1. We hypothesize that IGF1R localizes the signaling complexes at the cell membrane.  42  alveolar rhabdomyosarcoma, respectively (216, 217). Furthermore, IRS-1, the major IGF1R substrate, is constitutively tyrosine phosphorylated in EN-transformed cells, and physically associates with E N (215). E N / IRS-1 complexes bind both Grb2 and the PI3K p85 regulatory subunit. This strongly suggests that IRS-1 is functioning as the adapter molecule linking E N to Ras-MAPK and PI3K-AKT signaling pathways, respectively (215). A large number of theories have been put forth to explain how IGF1R is functioning in oncogenesis, including either enhancement of mitogenesis or suppression of apoptosis (reviewed in (188)). However, no theory fully explains the phenomenon, and the mechanism by which loss of IGF1R blocks transformation surprisingly remains unclear (212). It is currently postulated that IGF1R is complementing E N transformation at least in part by contributing to anchorageindependent growth of transformed cells, although the mechanisms involved remain unclear.  1.8.2  Role of the TGFp Pathway 4  Transforming growth factor-(3 (TGF-(3) is a member of a family of structurally homologous dimeric proteins (TGF-|3l-5) (218, 219). TGF-P elicits a variety of biological activities including growth stimulation and arrest, stimulation of extracellular matrix formation, stimulation of angiogenesis, immunosuppression, induction of apoptosis, and induction of differentiation of several cell lineages (218, 219). TGF-|3 factors initiate signaling by assembling receptor complexes that activate S M A D transcription factors (220). TGF-f3 was originally identified on the basis of its ability to induce anchorage-independent growth (phenotypic transformation) of fibroblasts (221). However, in most epithelial, endothelial, and hematopoietic cells, TGF-(3 is a potent inhibitor of cell proliferation, by stimulating production of the cyclindependent protein kinase inhibitors plS "" , p21 11  018  CIP1  , and p27  KIP1  (222, 223). In cancer cells, it is  43  thought that alterations in the TGF-(3 pathway confer resistance to growth inhibition by TGF-p\ thus allowing uncontrolled proliferation of cells (222). TGF-P is also thought to be involved in local invasion and metastasis (224), angiogenesis (225), host immunosuppression (226), as well as tumour suppression in both mouse and human studies (222). NIH3T3 cells expressing ETV6-NTRK3 show increased levels of TGF-(3l and -p2 mRNA, as compared with vector controls (Tognon and Sorensen, unpublished results). Further, as documented for a number of EWS-containing fusion protein (227, 228), TGF-(3RII appears to be down regulated in EN-expressing cells (229)). Expression of TGF-|3 has been shown to be upregulated in some prostate tumours, despite its growth-inhibitory effect on normal epithelial and carcinoma cells of the prostate. It is hypothesized that these carcinoma cells down-regulate the receptor to avoid the effects of autocrine TGF-(3l (230). These tumours then lose the inhibitory effect of TGF-(3 on proliferation (i.e. gain proliferation). The production and secretion of TGF-|3 by certain cancer cells may suppress the activities of infiltrating immune cells, thereby helping the tumour escape host immune surveillance (226). Accordingly, T G F - P l may play an immunemodulatory role in tumourigenesis. Based on these findings, the Sorensen laboratory is currently investigating the role of the TGF-p family in ETV6-NTRK3 signaling.  1.8.3  Higher Order Polymer Formation of the E N Oncoprotein It was initially hypothesized that, similar to other chimeric PTKs, SAM-mediated  dimerization of E N leads to constitutive activation of the P T K and downstream signaling cascades. However, replacement of the E N S A M domain was with the inducible FK506 binding protein (FKBP) dimerization system, FKBP-NTRK3 chimeras failed to transform NTH3T3 cells even though transient P T K activation was observed (231). It was recently shown that the E T V 6  44  S A M domain has two potential interacting surfaces rather than one, raising the possibility that this domain can mediate protein polymerization as opposed to dimerization (232, 233). In fact, Kim et al. demonstrated that the isolated ETV6 S A M domain forms an insoluble homopolymer in bacterial cells, and self-associates in a head-to-tail fashion to crystallize as an extended helical polymer (232). Mutation of key amino acids in each E N S A M binding interface completely blocked the ability of E N to polymerize, to activate its PTK, and to transform NIH3T3 cells (231). Furthermore, while E N formed large polymeric structures in cells, mutant E N proteins existed only as monomers. When isolated S A M domains were co-expressed in E N transformed cells, a dominant negative effect on E N polymer size and transformation was observed. Although a number of possibilities can be put forth to explain these observations, it is interesting to speculate that polymerization facilitates optimal positioning of P T K domains for crossphosphorylation of E N molecules. For example, perhaps the P T K phosphorylation partners in the E N polymer need to be located one helical turn apart for proper positioning as substrates for each other. Therefore higher order polymerization may be  a critical  requirement for the  transformation activity of E N and potentially other ETV6-PTK fusion proteins.  1.9  AIMS & O B J E C T I V E S The ETV6-NTRK3 fusion has been demonstrated to be involved in a group of relatively  rare diseases. However, studies into the relatively rare tumour retinoblastoma have given considerable insight into our understanding of tumour suppressors. Mechanisms discovered by such rare diseases can be more clearly characterized than mechanisms involved in common diseases. These findings can then be extrapolated to studies of these clearly defined mechanisms in more common conditions. Further, there is considerable evidence to support the role of N T R K  45  receptors in human malignancy. This fusion is only one mechanism for the activation of the kinase domain of the NTRK3 receptor, and therefore this research has direct relevance to any oncogenic activation of NTRK3. Moreover, E N is a model system to elucidate the pathways involved in oncogenic N T R K dysregulation. As presented earlier, oncogenic fusion are typically very cell-type specific. E N is unique among chimeric fusion proteins in that it has been identified in mesenchymal, epithelial, and hematopoietic  malignancies. Studies into EN-induced signaling may reveal  mechanisms  responsible for this lack of specificity. Despite extensive characterization of the signaling pathways induced by E N , the precise mechanism of activation of these pathways (e.g., characterization of binding partners) is not yet known. Further, while extremely difficult to recapitulate for sarcomas, animal models that mimic human disease have given considerable insight into mechanisms of tumourigenesis. Complete understanding of the oncogenic process mediated by E N will be important to the development of novel treatment strategies that target these tumours in vivo. We hypothesize that E N is a potent oncoprotein capable of transforming multiple cell lineages. Further, an interaction between the carboxy-terminus of E N and IRS-1 is essential for this transforming activity. Therefore, the studies presented in this thesis were undertaken with the following aims:  1) To determine the mechanism of IRS-1 binding to E N . 2) To identify additional amino acid residues involved in activation of downstream signaling pathways. 3) To develop an animal model of ETV6-NTRK3 induced tumourigenesis using transgenic technology.  46  CHAPTER II MATERIALS AND METHODS 2.1  C E L L CULTURE-BASED TRANSFORMATION STUDIES  2.1.1  C E L L LINES NIH3T3 cells, derived from murine embryonic fibroblasts, were obtained for long-term  culture from Dr. Robert Kay at Terry Fox Laboratories in Vancouver, Canada. These fibroblasts, as well as all other cell lines described below, were cultured at 37 °C using standard methods (234). The cells were grown in Dulbecco's Modified Eagle Medium (DMEM) containing 9% calf serum (CS), high glucose, sodium pyruvate, and pyridoxine hydrochloride (GibcoBRL), and supplemented with antibiotic-antimycotic (GibcoBRL). Stocks from each cell line were preserved by freezing the cells, in media with 10% D M S O (Fisher), in liquid nitrogen. BOSC 23 packaging cells were used to produce mature ecotropic viruses from the various  ETV6-NTRK3  constructs cloned into the retroviral vector, MSCVpac (described further in Section 2.1.3). BOSC 23 cells were cultured at 37 °C using standard methods (234), in D M E M containing 10% fetal calf serum (FCS). Confluent flasks were split no lower than 1:5 prior to transfection. Transient transfection experiments were performed in FfEK 293T cells, a human embryonic kidney cell line. HEK293T cells were obtained from A T C C and grown in 10% FBS / D M E M .  2.1.2  GENERATION OF F U L L - L E N G T H ETV6-NTRK3  AND MUTANT CDNA  47 Full-Length E N cDNA. The cDNA encoding full-length ETV6-NTRK3 and kinasedead-EN (K380N) was inserted into the retroviral vector MSCVpuro or MSCVneo at the EcoRI site as described (129). Truncated Constructs. The cDNA encoding EN-A614 and EN-A624 were produced via amplification of residues 1-614 of ETV6-NTRK3 using primers containing Hpal I EcoRI restriction sites (5' primer: 5' TGCAGTTAACGTTCCTGATCTCTCTCGCTGTG CTGAATTCCTAGATCTCCTTGATGTTCAACC  3'; 3';  A614 A624  3' 3'  primer:  5'  primer:  5'  C G C G G A T C C C T A G G C C T T C C C C A A A G C A T G G A G ) . The PCR product was digested with Hpal I EcoRI and cloned into MSCVpuro. Full-length clones were screened by D N A sequencing using  ETV6 primers TEL352 and TEL701 (Table 4; (117)), NTRK3 primers Trkl and Trk3, and  M S C V 5' and 3' primers (Table 4). All sequencing was performed on an ABI 3100 genetic analyzer (Applied Biosystems) using the BigDye terminator reaction, and analyzed using DNASTAR™ software. Site-Directed Mutagenesis The QuikChange™ Site-Directed Mutagenesis Kit (Stratagene) was employed to create various tyrosine point mutants at the C-terminus of E N (Table 4). All SDM-PCRs were carried out as follows: 95°C for 30 s, followed by 12 cycles of 95°C for 30 sec, 55°C for 1 min, and 68°C for 10 min 15 sec. D N A sequencing was used to confirm all the point mutations; sequences using standard primer T7 or NTRK3 primer PS2 (Table 4) were obtained and analyzed as  48  described above. Double site-directed mutants (Y615F+Y628Q, T625A / P626A) were made by performing two separate site-directed mutagenesis experiments for each amino acid.  2.1.3  TRANSDUCTION OF GENES USING T H E RETROVIRAL V E C T O R MSCVpac Cloning ETV6-NTRK3  constructs into MSCVpac  The murine stem cell virus (MSCV) vector, MSCVpac, was derived from the murine embryonic stem cell virus and the L N retroviral vectors (235) (236). Upon transfection into a packaging cell line, MSCVpac transiently expresses (or integrates and stably expresses) a transcript containing the extended viral packaging signal, the puromycin N-acetyl transferase (pac) resistance gene, and a gene of interest inserted into its multiple cloning site. The vectors achieve stable, high-level gene expression through a specifically designed 5' long terminal repeat. The entire ORFs of ETV6-NTRK3,  along with the 11 mutated constructs, were subcloned into  MSCVpac at the .EcoRI site. Clones were screened using 5gZII-digestion (NEB), as well as M S C V - F / Trk2 PCR-amplification to ensure the proper orientation of the ETV6-NTRK3  cDNA  (Table 4). Recombinant viruses were produced using BOSC 23 packaging cells. Transfection of BOSC 23 packaging cell line The BOSC 23 cell line packages retroviral R N A genomes into infectious, replicationincompetent retroviral particles. MSCVpac does not contain the gag, pol, and env genes necessary for viral particle formation and replication: these genes are stably integrated into the BOSC 23 genome (237-239). Introduction of MSCVpac into the packaging cell line results in production of high-titer, replication-incompetent infectious virus particles.  49  Calcium-chloride (CaCb) mediated transfection was used to introduce the recombinant vectors into the BOSC 23 cells (240). Packaging cells from a 10-cm dish were used to seed 6well plates. On the day of the transfection, the media from each well was replaced by 1 mL of D M E M containing 9% FCS and 25 p,M of chloroquine diphosphate (Sigma). 2 u.g of D N A , mixed with 200 uL of 250 m M CaCl (Fisher) and 200 uL of 2X Hank's Balanced Salt solution 2  (GibcoBRL), was added to the media. The cells were returned to the incubator for 8 to 10 hours. The supernatant was replaced with fresh D M E M / 9% FCS at 8 - 10 hours and at 30 - 36 hours post- transfection. The supernatant containing mature viruses was collected at 45 - 50 hours. This solution was filtered (22 ^im filter) prior to being used directly for infection or stored at -70°C. Infection of NIH3T3 cells NIH3T3 cells were infected with recombinant retroviruses in order to establish cell lines stably expressing ETV6-NTRK3 and various mutants. NIH3T3 cells were seeded at low density on to 6-well plates. On the day of infection, the medium was replaced with 0.5 mL of D M E M / 10% CS containing 20 [ig/mL hexadimethrine bromide (Sigma). 0.5 mL of viral supernatant was added, and the cells were incubated overnight. The medium was then replaced with 2 mL of fresh D M E M / 10% CS. 36 hours after the start of infection, the cells were placed under antibiotic selection (2 ^ig/mL of puromycin (Sigma) for 4 days), then cultured to sub-confluency and confirmed for protein expression. Co-Immunoprecipitation Experiments. cDNAs encoding E N , EN-A614 and EN-Y615F mutant constructs were C-terminally tagged with V5-His tag using pcDNA3.1/V5-His-TOPO (Invitrogen). Different combinations of  50  H A - and V5-tagged constructs were transiently co-transfected in HEK293T cells using FuGENE 6 transfection reagent (Roche). HA-tagged phospho-tyrosine binding (PTB) domain of IRS-1 (HA-IRS-1C, a generous gift of Mr. Matthew Martin) was transiently transfected in NIH3T3s expressing  EN  or  EN-mutants.  Lysates  were  prepared  as  described  above,  and  immunoprecipitations (a-HA, a-V5) were performed on lysates collected 36 hours after transfection, followed by a-V5 or a-HA western blotting.  2.1.4 A S S E S S M E N T O F T R A N S F O R M A T I O N Retrovirally transduced cell lines were initially assessed for transformation based on morphologic criteria. Transformation was associated with spindling, elaboration of cellular processes, increased nuclear-to-cytoplasmic ratios and refractility, and focus formation with loss of contact inhibition (241). Soft Agar Assay To assess the transformation ability of these transfected cells, the soft agar assay was employed to assess anchorage-independent growth as described (215). Briefly, transfected NIH3T3 were plated in triplicate at a density of approximately 8 X 10 cells per 35 mm dish 3  (Falcon). Each agar plate contains a bottom and top (cell) layer. Bottom layers were made up of 0.4% agar in 9% CS supplemented D M E M (GibcoBRL). Cells were re-suspended in a top layer of 0.2% agar in 9% CS D M E M . Cells were fed every other day by adding two drops of serumcontaining media. After two weeks at 37 °C, the number of single cells and colonies per high power field were counted. Colonies were defined as an aggregation of four or more cells, and results were formulated as a percentage of colonies formed per total number of cells plated.  51  Photomicrographs and pictures of the plates were taken 16 to 19 days after plating, and the plates were observed for up to 28 days for macroscopic colony growth. As the data did not meet the assumption for analysis of variance (normal distribution, random independent samples and equal variance), a non-parametric test was utilized. Statistical significance of differences in the respective groups was evaluated using the Mann-Whitney ranks test; P values < 0.05 were considered to be of statistical significance. Injection into SCID mice Pathogen free male Fox Chase SCID (severe combined immune deficiency) mice (C.B-17), five to six weeks old were obtained from Charles River Laboratories. One or two million NTH3T3 cells infected with control vector, ETV6-NTRK3, or mutant E N , were injected subcutaneously (s.c.) at four sites per animal (three animals / group), for a total of 12 monitored sites for each cell line. Animals were housed in laminar flow racks and microisolator cages under specific pathogen free conditions and received autoclaved food and water. Once palpable tumours were established, tumour volume measurements were taken every two to three days using calipers, until which time tumour growth in the ETV6-NTRK3 group necessitated the termination • of the experiment. Statistical significance of differences in tumour size in the respective groups was evaluated using the MannWhitney ranks test; P values < 0.05 were considered to be of statistical significance. Anchorage-independent Multi-cellular Spheroid Cultures For anchorage independent suspension cultures, ambient environmental conditions and media were identical to adherent monolayer cultures. To establish suspension cultures, confluent monolayers were trypsinized, resuspended as single cells, and replated on tissue culture dishes  52  that had been coated with 1.4% agar. Suspended cells were cultured for four days before spheroid formation was assessed.  2.1.5  PROTEIN ANALYSIS Lysates were prepared from cell lines (as described above), and prepared for either  immunoprecipitation or as direct lysates for western blotting as described above in section 2.1.7. Additionally, the following antibodies were used: Phospho-Mekl/2 (Ser217/221)(IB: 1:1000; Cell Signaling); Phospho-AKT (Ser473, IB:  1:1000; Cell Signaling); 4G10 (IB:  1:10000;  Upstate Biotechnology); TrkC (C-14) (WB: 1:1000; IP: 5ul/IP; Santa Cruz Biotechnology); cyclin Dl/2  (IB:  1:2000; Upstate Biotechnology);  Grb2 (WB: 1:5000, Transduction  Laboratories); G S T - E T V 6 - H L H (IB: 1:5000; IP: 2ixl/IP; a generous gift from Peter Marynen, Flanders Interuniversity Institute for Biotechnology, University of Leuven, Leuven, Belgium); IRS-1 (IP: 2 ug/IP; WB 1:1000, Upstate Biotechnology Inc., Lake Placid, NY); H A antibody (HA. 11) (IB: 1:2000 dilution; IP: 5ul of 1/50 dilution/IP; BabCO); V5 (IB: 1:5000; Invitrogen).  2.1.6.  IMMUNOFLUORESCENCE V5-tagged  constructs  of  ETV6-NTRK3,  A614,  and Y615F were  made  using  pcDNA3.1V5/His-TOPO cloning vector (Invitrogen). As the V5 portion was placed on the 3' end of the various constructs, the stop codon was replaced with a glycine codon to allow for the continued translation of V5. This was accomplished using the QuikChange™ Site-directed mutagenesis kit (Stratagene)  according to the manufacturer's instructions. V5-tagged E N  constructs were cloned into MSCVpuro retroviral plasmid using the Kpnl I Pmel digest and Kpnl cut / CEP treated MSCVpuro. Constructs, including vector control, were retrovirally infected into  53  NIH3T3 cells and selected as described above. Prior to immunofluorescence experiments, equal levels of V5-tagged proteins were confirmed by western analysis. Cells were grown on Fisherbrand (Fisher) glass slides as described above until the confluency reached approximately 50%. Cells were serum starved for a minimum of 24 hours in duplicate, with one group subject to serum stimulation ( D M E M with 9% calf serum) for 30 minutes. The slides were then quickly rinsed with PBS and fixed in 4% paraformaldehyde (pH 7.4) (Sigma) in PBS for 10 minutes at room temperature, followed by permeabilization with 0.5% NP-40 (5 minutes at room temperature). The slides were then washed three times in PBS, blocked with 5% skim milk powder (30 minutes at room temperature) and incubated overnight with a-V5 antibody (Invitrogen, 1:1000). Cells were washed with PBS, incubated with secondary antibody (Alexa 488, 1:1000) before being mounted with VectaShield™ containing DAPI counterstain, and visualized using a Zeiss Axioplan epifluorescent microscope equipped with a C O H U - C C D camera.  2.1.7  H O M O L O G Y MODELING OF KINASE DOMAIN O F E N The entire NTRK3 portion of the E N fusion protein (containing the catalytic kinase  domain and juxtamembrane region) was submitted to SWISS-MODEL Protein Modeling Server (242, 243) on the World Wide Web with the following parameter settings: (i) B L A S T search P value <0.00001, (ii) global degree of sequence identity (SIM) >25%, and (iii) minimal projected model length = 25 amino acid residues. Based on strongest alignment, proteins lir3A (244)), IgagA (245)), lirk (246), lufA (247), and lrqq (248) were selected as the structure templates for the query sequence, and all of them were catalytic domains of the tyrosine kinases.  54  2.2  TRANSGENIC MICE All mice were housed under specific pathogen-free conditions in the transgenic facility at  the Centre for Molecular Medicine and Therapeutics (CMMT) according to protocols approved by the Animal Care Committee at the University of British Columbia. C57BL/6J and C B A mice were from Jackson Laboratory (Bar Harbor, Maine).  2.1.1  VECTOR CONSTRUCTION AND CONFIRMATION OF EXPRESSION As the E N fusion transcript had been demonstrated in mesenchymal and hematopoietic  cells, transgenic mice expressing E N cDNA were generated under the direction of two different ubiquitously-expressing promoters, as described below.  pIRES2-EGFP-EN The full-length E N open reading frame was excised from pBluescript-EN (129) by digesting with EcoRI, and ligated into EcoRI -digested, Calf Intestinal Phosphatase (ClP)-treated pIRES2-EGFP expression vector (Clontech, Palo Alto, CA) (249) to create pIRES2-EN (see Figure 4). Prior to blastocyst injection, pIRES2-EN was linearized by digestion with Nsil to generate a 4848 bp fragment containing the promoter, transgene, and reporter, as well as two fragments of 72 bp and 2331 bp containing vector backbone, which were excluded from injection to improve transgene expression.  55  P-actin Promoter / CMV-IE enhancer  1 ETV6p I R E S 2 - E G F P - E r | NTRK3  on SV40 ori Rabbit P-globin poly-A  poly A  EcoRI (3657)  EGFP  F I G U R E 4. Constructs used for transgenesis. Cloning E T V 6 - N T R K 3 into p C X and pIRES2E G F P vectors. Both constructs were created by ligating ZscoRI-cut transgene into .EcoRI, C I P treated vectors.  pCX-EN Expression vector p C X - E G F P was obtained from Dr. Andras Nagy (Toronto, ON), with permission from the creator (Dr. Jun-Ichi Miyazaki, Osaka University Medical School, Osaka, Japan). This vector contains the chicken beta-actin promoter and cytomegalovirus immediateearly ( C M V - I E ) enhancer ( p C A G G S ) (250); expression from this vector i n transgenic mice had been demonstrated in various tissues, including heart, kidney, brain, thymus, spleen, intestine, testis, lung, muscle adipose tissue, and adrenal glands (250), (251). Due to concerns of toxicity of the G F P protein, it was decided to remove this protein from one of the two transgenic constructs. Accordingly, the E G F P reporter was removed from p C X - E G F P with an E c o R I digest. E c o R I -prepared E N O R F was ligated into CIP-treated vector (Figure 4). Prior to zygote injection, p C X - E N was linearized by digestion with Sail to generate a 6716 bp fragment.  56  Each construct was confirmed by restriction digest, and sequence verified twice. Linearized constructs were electrophoresed on an agarose gel, excised with a sterile scalpel, and purified using a QIAquick gel extraction kit (QIAGEN), according to the manufacturer's protocol. D N A was eluted in (filter-sterilized) 10 mM Tris pH 8.0, 0.1 m M E D T A , and stored at - 70 °C until injection.  Confirmation  of  Expression  To verify expression of transgenic constructs, p C X - E N and pIRES2-EN, as well as empty vectors, were transiently transfected into NIH3T3 cells. Cells were plated in 10 cm dishes such that they were 90-95% confluent on the day of transfection. For each construct, 24 \ig vector D N A was diluted in 1.5 ml of Opti-Mem (Gibco-BRL) without serum. Sixty ul of LIPOFECTAMINE 2000 was diluted in 1.5 ml Opti-Mem and incubated at room temperature for five minutes, before combining with the D N A mixture and incubating 20 minutes. D N A Lipofectamine complexes were added to the cultures and incubated for 24 hours. 48 hours posttransfection; cells were selected in neomycin (G418, Gibco, 900 |ig/ml for seven days). Transfection of fibroblastic cell cultures was assessed by western analysis for ETV6-NTRK3 (described in detail below). To assess the transformation ability of these transfected cells, soft agar assays were performed as described above (Section  2.2.2  CONSTRUCT INJECTION All pronuclear microinjections were done on a F2 hybrid background (C57BL/6J x C B A  cross was designated F l ; F l x F l cross was designated F2). Each construct was injected into  57  approximately 300 eggs, by Mrs. Anita Borowski, technician, Canadian Genetic Diseases Network, Centre for Molecular Medicine and Therapeutics. Injected eggs were transplanted into pseudopregnant females (females that had been mated with a vasectomized male) by oviduct injection using standard procedure.  2.2.3  PREPARATION OF TAIL DNA FOR GENOTYPE ANALYSIS For genotyping, approximately 1 cm tail clips were obtained from anesthetized mice, and  digested in 400 [i\ lysis buffer containing 1.2 mg/ml proteinase K, 50 m M Tris-HCl pH 8.0, 10 mM E D T A , 0.1% SDS, and 100 mM NaCl. Following digestion, an equal volume of 1.0 M TrisHCl Buffer pH 8.0 and phenol / chloroform / isoamyl alcohol (25:24:1) was added. Samples were hand mixed, and centrifuged at 13, 000 rpm for three minutes. The aqueous phase was removed and the phenol / chloroform / isoamyl alcohol step was repeated. An equal volume of chloroform was added to the aqueous phase; samples were mixed, and centrifuged. The D N A was precipitated by adding l/10  th  volume of 5 M Ammonium Acetate and an equal volume of  ice-cold isopropanol (approximately 500 |il). The D N A was pelleted by centrifugation, washed several times with 70% EtOH, air-dried, and resuspended in 100 |Ltl T E buffer (10 m M Tris-Cl, pH8.0, 1 m M EDTA).  58  T A B L E 4. Primer sequences used in site-directed mutagenesis and sequence analysis.  PRIMER  COMMENTS  S E Q U E N C E (5'-3')  TEL352  ETV6 nt 352-373  GGT G A T G T G C T C T A T G A A C T C C  TEL541  ETV6 nt 541-560  CCT C C C A C C A T T G A A C T G TT  TEL541rev  ETV6 nt 541-560  AAC AGT TCA ATG GTG GGA G G  TEL701  £ 7 V 6 n t 701-720  AGA ACA ACC ACC AGG AGT CC  ETV6-F  ETV6 nt 905-924  AGCCCATCAACCTCTCTCAT  PS2  NTRK3 nt 2304-2324  GTA A T G CAC TCA A T G A C C TC  Trkl  NTRK3 nt 2414-2431  TCT C C T T G A TGT T C A A C C  Trk2  NTRK3 nt 1821-1838  CCG CAC ACT CCA T A G A A C  Trk3  NTRK3 nt 1601-1620  C C TCT T A A TGT G C T G C A C A T  NTRK3-R  NTRK3 nt 1073-1092 C T C G G C C A G G A A G A C CTT T C  Tel 971  ETV6 nt 971-995  A C C A C A T C A T G G TCT C T G TCT C C C  Trk 1059  NTRK3 1038 - 1059  C A G T T C T C G CTT C A G C A C G A T G  Delta614-F  E N amino acids 1-614  T G C A G T T A A CGT T C C T G A T C T C T C T C G C T G T G  Delta 614-R  E N amino acids 1-614  CTG A A T TCC T A G A T C T C C TTG A T G TTC A A C C  Y594F  CCC A A A G A G GTG TTC GAT GTC A T G CTG  Y615F  ATC A A G G A G A T C TTC A A A A T C CTC  Delta 624-R  C G C G G A T C C C T A G G C CTT C C C C A A A G C A T G G A G  P626A  GCT TTG G G G A A G GCC A C C G C A A T C T A C CTG G A C ATT  T625A/P626A  GCT T T G G G G A A G G C C G C C G C A A T C T A C C T G G A C A T T  I627E  GGG A A G GCC A C C C C A G A A T A C CTG G A C ATT  Y628Q-for  G G C C A C C C C A A T C C A G C T G G A C A T T CTT G G C  Y628Q-rev  GCC A A G A A T GTC C A G C T G G A T T G G GGT GGC C  M S C V 3'  CCC TTG A A C CTC CTC GTT C G A C C  M S C V 5'  G A G A C G TGC T A C TTC C A T T T G TC  59  2.2.4  GENOTYPE ANALYSIS BY PCR The genotype of E N transgenic mice was identified by PCR of tail genomic D N A by  using Tel541 and Trk2 primers (all E N primer sequences are described in Table 4) using the following conditions: 94 °C for 5 min, followed by 35 cycles of 94 °C for 1 min, 60 °C for 1 min, and 72 °C for 1 min, and a final extension of 72 °C for 10 min. E N positive mice were identified by the presence of a 731-bp product when visualized following standard agarose gel electrophoresis. To confirm integrity of genomic D N A , (^-microglobulin was amplified using the following primers: 5' primer: 5' C A C C G G A G A A T G G G A A G C C G A A 3'; 3' primer: 5' T C C A C A C A G A T G G A G C G T C C A G 3'.  2.2.5  PATHOLOGY OF TRANSGENIC TISSUES When mice were found dead or moribund, necropsies were performed examining all  tissues. Thymus, lymph nodes, bone marrow, spleen and solid organs were analyzed histologically. Specimens were fixed by immersion for 24 hours in 10% phosphate-buffered formalin before embedding in paraffin. Sections of 35 \xm thickness were cut and placed on glass slides. Tissue specimens were dehydrated, dewaxed and stained with Hematoxylin and Eosin (H&E). Some larger lesions were additionally analyzed by F A C S or snap frozen in liquid nitrogen for immunohistochemical and protein analysis.  2.2.6  RNA ISOLATION AND RT-PCR Total R N A was extracted from cell lines and primary tissues using the acid-guanidinium-  phenol / chloroform method of Chomczynski and Sacchi (252). Briefly, cell pellets or approximately ten 5 u,m-thick sections of primary tumour were resuspended and homogenized in  60  1 ml of Trizol (Gibco-BRL, Life Technologies) and RNA then isolated from the aqueous phase. Purity  and integrity  of  R N A was  assessed by  1% agarose  gel  electrophoresis  and  spectrophotometric analysis of the OD A 6o/A 8o ratio. Isolated mRNA was then treated with 2  deoxyribonuclease  I (Invitrogen)  to  2  remove  any contaminating cDNA  before  reverse  transcription using the Biosciences Titanium One Step RT-PCR kit (Clontech) using oligo(dT) primers according to the manufacturer's recommendations. Reverse transcriptase-polymerase chain reaction (RT-PCR) was used to screen for ETV6NTRK3 fusion transcripts using ETV6 and NTRK3 primers (Table 4) (185). PCR conditions were as follows: 94 °C for 1.5 minutes, followed by 35 cycles of 94 °C for 45 seconds, 60 °C for 1 minute and 72 °C for 1 minute and a final extension of 72 °C for 10 minutes. The presence of any contaminating D N A was accounted for by PCR on non-transcribed mRNA samples. Amplified products were visualized by electrophoresis using 2% agarose gels stained with ethidium bromide. All samples were confirmed for the presence of the breakpoint by Southern blot analysis as described above using a D N A oligo (Tel 1033) spanning the breakpoint:  5'-  G G G A G A A T A G C A G A T G T G C A G C A C - 3 ' . The presence of amplifiable R N A in all samples was  confirmed  by  RT-PCR  TGTGATGGTGGGAATGGGTCAG or  hypoxanthine  using  either  13-Actin  (5'  primer:  3'; 3' primer: 5' T T T G A T G T C A C G C A C G A T T T C C  phosphoribosyltransferase  CCTGCTGGATTACATTAAAGCACTG  3';  (HPRT)  (278  279  G T C A A G G G C A T A T C C A A C A A C A A A C 3') (253) primers as a control.  primer: primer:  5' 3') 5' 5'  61  2.2.7  PROTEIN ANALYSIS One ml of Lysis Buffer (1.5 mM M g C l (Fisher), 150 mM NaCl (Fisher), 50 mM Hepes 2  (Sigma), 10 m M NaF (Sigma), 10 m M N a P 0 (Sigma), 2 m M N a V 0  4  ethylene-diamine-tetraacetic  2  4  2  7  3  (Sigma), 2 m M  acid (EDTA) (Fisher), 2 m M N a M o 0 - 2 H 0 (Sigma), 10% 4  Glycerol (Fisher), 0.5% Nonidet P-40 (Fisher), Leupeptin (1:1000 dilution of 2 mg/ml stock made in H 0 ) (Sigma), Aprotinin (1:1000 dilution of lOmg/mL stock made in H 0 ) (Sigma), 2  2  Phenylmethylsulfonyl Fluoride (PMSF) (1:200 dilution of a 100 m M solution made in dimethyl sulfoxide) (Sigma)) was then added to the phosphate buffered saline-rinsed cells or tissues, and incubated for 15 minutes on ice.  Lysates were then cleared at 12000 rpm for 10 minutes, at  which point the supernatant was transferred into a fresh tube. For immunoprecipitation analysis, 1000 /xg of lysate was incubated with gentle agitation for three hours at 4 °C with either NTRK3 antibody (5 pil) (C-14, Santa Cruz Biotechnology) or a-ETV6:HLH (2 JU.1) (generous gift of Dr. P. Marynen) along with 10  of Protein A-Sepharose  (Pharmacia). The tubes were centrifuged at 2500 rpm for 5 minutes and the supernatant discarded. The pellet was washed two to three times in wash buffer (0.1% Nonidet P-40), boiled in Laemmli buffer and electrophoresed on a 7.5, 10 or 15% polyacrylamide gel overnight at 70 100 Volts according to standard methods (254). Transfer of the proteins from the gel to Immobilon-P (Millipore) was accomplished with the Bio-Rad Trans-Blot SD Semi-Dry Transfer cell at 25 volts for 60-90 minutes using Towbin Transfer Buffer (25 m M Tris (Fisher), 192 m M glycine (Fisher), 20% methanol (Fisher). The membranes were blocked with Blocking Buffer ( I X TBS, 1% BSA, 0.05% Tween-20) for one hour at room temperature with gentle agitation. The membrane was then incubated with one of the following antibodies: anti-TrkC (C14) (1 jug/ml, Santa Cruz Biotechnology), RC20-Horse  62  Radish Peroxidase conjugated (1:2500) (Transduction Laboratories), 0 E T V 6 : H L H (1:5000). Proteins were detected using HRP-conjugated secondary antibodies and E C L reagent, according to established protocols (254).  2.2.8  FACS ANALYSIS Freshly isolated tumour tissue was incubated with crude collagenase (0.25%, Invitrogen)  at 37 °C for several hours to disaggregate the tissue. Cell suspensions were passed through a 40 (j,M mesh filter to achieve an approximately single cell suspension. Cells were analyzed immediately or resuspended in 10% dimethylsulphoxide in fetal bovine serum and stored in liquid nitrogen before analysis. Flow cytometry was performed on a FACSCalibur  Flow  cytometry system with CellQuest and Modfit L T analytic software (Becton Dickinson, San Jose, CA). For analysis of cell surface molecules, all samples were labeled with directly conjugated fluorescent antibodies at 4 °C for 20 min. The samples were then washed and resuspended in 1.5% paraformaldehyde and analyzed within 24 hours. The following antibodies, all obtained from B D Pharmingen (San Diego, C A , USA), were used for cell surface labeling: CD3, Thy-1.2, and B220. The relevant labeled isotype control antibodies were included in all experiments.  2.2.9  T U M O U R TRANSPLANTATION ASSAY To determine if the EN-expressing tumours were transplantable, tumours were harvested  from E N transgenic mice, finely minced with crossed scalpels and filtered through 30 ^im nylon mesh to remove clumps. Cells were washed in PBS, resuspended in D M E M (Invitrogen Life Technologies,  Burlington,  ON) without  serum, and injected  (10  6  cells/0.1  ml) both  intraperitoneally and into the lateral tail veins of seven to eight week old male severe combined  63  immunodeficient mice (Fox-Chase, CB-17, bred in-house, Jack Bell Research Centre under the direction of Dr. Martin Gleave). Mice were housed in microisolator cages and were provided with autoclaved chow and acidified water. Animals were monitored daily for signs of distress (hunching, labored breathing, lack of activity, scruffy coat) and were sacrificed at the first signs of distress.  2.2.10 M O U S E C R O S S - B R E E D I N G Transgenic mice were maintained as heterozygotes, and backcrossed to C57BL/6J mice. To determine the relevance of a functional immune system in suppressing tumour formation, E N heterozygotes were crossed to Rag-deficient mice (B6.129S7-Z?ag  fniMonyj^  T  h  e  j  a  c  k  s  o  n  Laboratory, Bar Harbor, Maine) (255). E N mice were also mated with mice that overexpress A K T (PTEN heterozygote mice). Pten " (B6A29-Pten" ) +/  nlRps  mice were created by Dr. Ramon  Parsons (256) and obtained, with permission, from Dr. Frank Jirik (University of Calgary). Pten " mice were maintained as heterozygotes, and genotyped as described (257). +/  2.2.11 E N T A R G E T E D E S C E L L S E N knock-in cells were created by targeting the CJ7 ES cell line using double drug selection (neomycin and thymidine kinase), followed by a PCR / Southern blot screen to identify correctly targeted clones (experiments performed by Dr. Zhe L i , Post-Doctoral fellow, Dr. Stuart Orkin, Harvard Medical School). Targeted clones were maintained on a embryonic fibroblastic feeder cell layer in D M E M supplemented with 20% heat-inactivated fetal bovine serum, 2 m M L-glutamine, 6.25 ml in-house nucleoside mix, non-essential amino acids (GIBCO), 0.1 ^iM 2mercaptoethanol, and 100 units/ml of recombinant leukemia inhibitory factor (Sigma) at 37 °C in  64  an atmosphere of 5% CO2 in air. Nucleoside mix consisted of 80 mg adenosine (Sigma A-4036), 85 mg guanosine (Sigma G-6264), 73 mg uridine (Sigma U-3003), 73 mg cytidine (Sigma C4654), 24 mg thymidine (Sigma T-1895) dissolved in 100 ml distilled water. For western analysis, cells were grown on gelatinized plates (0.1%) for several passages to remove feeder cell contamination.  65  C H A P T E R  III  A HIGHLY CONSERVED NTRK3 C-TERMINAL SEQUENCE IN THE ETV6-NTRK3 ONCOPROTEIN BINDS THE PTB DOMAIN OF IRS-1  3.1 RESULTING PUBLICATION & CONTRIBUTION OF INDIVIDUAL AUTHORS This chapter has been published: Lannon, C. L . , Martin, M . J., Tognon, C. E . , Jin, W., Kim, S. J., Sorensen, P. H.B. (2004). A highly conserved NTRK3 C-terminal sequence in the ETV6-NTRK3 oncoprotein binds the phosphotyrosine binding domain of insulin receptor substrate-1: an essential interaction for transformation. J Biol Chem 279(8): 6225-34. The project described in this chapter was conceived and initiated by the Sorensen Lab, and developed into an intellectual collaboration with the laboratory of Dr. Seong-Jin Kim during the latter stages of experimentation and analyses. I generated the majority of the data, wrote the paper, and saw the manuscript through to publication.  The following  figures  were not generated  by myself, and are duly  acknowledged:  Figure 5. Cloning by Matthew Martin; experimental results generated by Cristina Tognon Figure 11. Experimental Results from Matthew Martin Figure 12. Experimental Results from Matthew Martin  66  3.2  INTRODUCTION The transforming properties of E N require both an active P T K domain and the S A M  oligomerization domain (129). E N expression leads to constitutive elevation of cyclin D l mRNA and protein levels, and EN-induced cyclin D l expression correlates with increased cell cycle progression (210). Moreover, the Sorensen lab found that E N expression in NIH3T3 cells leads to constitutive activation of two of the major effector pathways of wild-type NTRK3, namely the Ras-Rafl-Mekl-Erkl/2 M A P kinase (Ras-MAPK) mitogenic pathway and the phosphatidyl inositol-3-kinase (PI3K) pathway leading to activation of the A K T survival factor (210). Phenotypic transformation and soft agar colony formation by EN-expressing cells are blocked by inhibition of either Mekl or PI3K. However, we failed to detect interactions between E N and adapter molecules known to link NTRK3 to Ras-MAPK and PI3K-AKT pathways such as SHC, GRB2, S H 2 B p \ or the p85 subunit of PI3K (129), as well as A B L , SRC, and SHTP2 (unpublished data). Wild-type N T R K proteins utilize juxtamembrane tyrosine residues (e.g. NTRK3 Y485) to interact with several of these adapters including SHC (163), GRB2 (166), and p85 (165, 166), but this residue is not present in the E N oncoprotein due to the position of the fusion point (129). Recently, my colleagues observed that E N fails to transform mouse embryo fibroblasts derived from mice with a targeted disruption of the insulin-like growth factor 1 receptor (IGF1R)  gene, but that re-introduction of IGF1R  into these cells restores E N  transformation activity (215). This led me to examine in more detail the relationship between IGF1R signaling and E N transformation. The Sorensen laboratory has previously observed a direct physical interaction between E N and the major IGF1R substrate, insulin receptor substrate-1 (IRS-1) (215). IRS-1 is constitutively tyrosine phosphorylated in EN-transformed cells, and that E N / IRS-1 complexes bind both Grb2 and the PI3K p85 regulatory subunit. This  67  strongly suggests that IRS-1 is functioning as the adapter molecule linking E N to Ras-MAPK and PI3K-AKT signaling (215). However, the mechanism by which IRS-1 interacts with E N remains unknown. IRS-1 along with IRS-2-4 are a family of tyrosine phosphorylated scaffold proteins that are substrates for IGF1R and the insulin receptor (IR) (17). Although IRS proteins lack enzymatic activity, they are thought to play key adapter roles in linking IGF1R and IR to downstream pathways. Three different domains in IRS-1 have been identified as potentially contributing  to  IGF1R  and IR binding: the  pleckstrin homology  (PH) domain, the  phosphotyrosine binding (PTB) domain, and the SHC and IRS-1 NPXY-binding (SAIN) domain (258). P H domains bind phospholipids, thereby mediating the interaction of signaling proteins such as IRS-1 with the plasma membrane (259). PTB domains in adapter proteins bind to phosphorylated tyrosines within N P X Y motifs in interacting proteins such as cell surface receptors (260), thus promoting receptor/adapter interactions. The SAIN domain of IRS-1 remains poorly characterized but has been postulated to contain other potential protein-protein interaction motifs (258, 261). I now show that the distal C-terminal sequence of E N , which is highly conserved among NTRK3 proteins across species, interacts specifically with the PTB domain of IRS-1; further, this interaction is essential for E N transformation activity. Moreover, the transformation activity of E N can be inhibited by a dominant-negative IRS-1 construct while IRS-1 over-expression in EN-transformed cells enhances the tumourigenic activity of this oncoprotein.  68  3.3  RESULTS  The PTB domain of IRS-1 mediates its association with EN.  Morrison  et  al.  previously  demonstrated  that  IRS-1  is  constitutively  tyrosine  phosphorylated in EN-transformed cells and that E N associates with IRS-1 in vivo (215). Further, IRS-1 functions as the adapter protein linking E N to the Ras-MAPK and PI3K-AKT cascades that are essential for E N transformation (215). To further characterize the E N / IRS-1 interaction, we created a series of HA-tagged IRS-1 constructs expressing specific portions of the protein (Figure 5A). These constructs were transiently transfected into FTEK293T cells along with V5-tagged E N . Expression of each construct was confirmed by a - H A immunoprecipitation followed by a - H A immunoblotting (Figure 5B). V5-EN expression was confirmed by a-V5 immunoprecipitation followed by a-V5 immunoblotting (Figure 5C). To determine the region of IRS-1 responsible for binding to E N , lysates were immunoprecipitated with either a-V5 or aH A antibodies, followed by immunoblotting with a-HA or a-V5 antibodies. H A tagged -EN, IRS-1 full-length (FL), -IRS1C, TRS-ID and -IRS-1E were all able to pull down V5-EN (Figure 5B) V5-tagged E N was able to immunoprecipitate HA-tagged E N as well as all IRS-1 constructs containing the phosphotyrosine binding (PTB) domain (HA-IRS-1C, HA-IRS-1D, and HA-IRS1E (Figure 5C). Only the fragment consisting of the IRS-1 PH domain alone was incapable of associating with V5-EN, indicating that the PTB domain of IRS-1 is required for its interaction with E N (Figure 5C).  Generation of C-terminal EN mutants.  PTB domains are generally thought to bind phosphorylated tyrosine residues on target interacting proteins within NPXpY motifs, where pY is a phosphorylated tyrosine and X is any  69  FIGURE 5. ETV6-NTRK3 (EN) fusion binds to the phosphotyrosine (PTB) domain of IRS1. (A) Schematic diagram describing the five HA-tagged IRS-1 constructs. (B) Immunoblots of H A immunoprecipitations from HEK293T cells co-transfected with V5-tagged E N and either 1) H A alone 2) HA-tagged E N or 3) one of the five HA-tagged IRS-1 constructs. Blots were probed with anti-HA and anti-V5 to detect immunoprecipitated HA-tagged IRS-1 constructs and coimmunoprecipitated V5-tagged E N protein, respectively. V5-tagged E N was used as a positive control. (C) Immunoblots of V5 immunoprecipitations from HEK293T cells co-transfected with V5-tagged E N and 1) H A alone, 2) HA-tagged E N , or 3) one of the five HA-tagged IRS-1 constructs. Blots were probed with anti-HA and anti-V5 antibodies to detect coimmunoprecipitated HA-tagged IRS-1 constructs and immunoprecipitated V5-EN, respectively.  y  W  —  •—•  t/3  IT)  I  I  HI SHI V H  =  CIISHI V H  > Z  w +  31 SHI V H  a i SHI  t  >  + 41 11 i  VH VH  KHVH  > AIN  CM PQ  u  C O  CD  ECU  CQ  mm  S3 MH  a C2  m  £ in  _  Pi  %  < PC  U  V3  Pi & <  w  GO  ssoo  X  CM CQ  11 a <  ^ u  c ao  u -  JO m >  >  71  amino acid (reviewed in (260)). E N sequences were searched for tyrosine residues, particularly within NPXpY motifs, that might mediate the E N / IRS-1 interaction. In wild-type NTRK3 signaling, the juxtamembrane NPQpY516 sequence binds SNT (169) and SHC (163). This sequence is lost in E N due to the position of the fusion point (2, 129), and previous studies confirmed that E N does not bind She (129). Of the 19 remaining tyrosines in E N (GenBank Accession Number AF041811), none are within classical NPXpY motifs (data not shown). None of the seven tyrosines within the ETV6 portion are likely to mediate the E N / IRS-1 interaction, as replacement of the entire E T V 6 sequence with the inducible F K B P dimerization domain (262) still results in activation of the Ras-MAPK and PI3K-AKT cascades (unpublished data). The other twelve tyrosines reside within the NTRK3 portion of the fusion protein, nine of which are conserved among human N T R K family members (NTRK1-3) (GenBank Accession Numbers NP_002520, NP_006171, and Q16288, respectively). If any of these tyrosines mediate the E N / IRS-1 PTB interaction, then their mutation should block E N transformation. Therefore, tyrosine to phenylalanine (Y to F) substitutions of each tyrosine were made and the resulting E N mutants transfected into NIH3T3 cells and assessed for soft agar colony formation as described (129, 210). As reported previously (129, 215), mutation of the three P T K activation loop tyrosines Y513, Y517, and Y518 completely blocked colony formation while mutation of Y560, Y594, and Y628 (the PLCyl binding site of EN) had no inhibitory effects on colony formation whatsoever (data not shown). Of the remaining tyrosines, only mutation of Y615 reduced soft agar colony formation (see below in Figure 8A). Y615 lies within a sequence of distal Cterminal NTRK3 amino acids that are highly conserved across vertebrate species (Figure 6B). Moreover, there is substantial sequence dissimilarity from either NTRK1 or NTRK2 (Figure 6A), suggesting that this region may be important for NTRK3-specific signaling. We therefore  72  created a deletion mutant, EN-A614, lacking the last 19 amino acids of E N (including Y615). Levels of expression of EN-A614 and the EN-Y615F mutant were similar to that of E N , and both became tyrosine phosphorylated in NIH3T3 cells (Figure 6C). A  B  NTRK Family Members NTRK3  PQQRLNIKEIYKILHALGKATPIYLDILG  NTRK2  .HT.K...N.HTL.QN.A..S.V  NTRK1  .HS..DVHAR.Q..AQ.P.V  V . .  Vertebrate Species human pig mouse rat chicken  PQQRLNIKEIYKILHALGKATPIYLDILG PQQRLNIKEIYKILHALGKATPIYLDILG PQQRLNIKEIYKILHALGKATPIYLDILG PQQRLNIKEIYKILHALGKATPIYLDILG PQQRLNIKEIYKILHALGKATPIYLDILG  Y615  Y628  kDa  72.668.0  IgG IP: C C - E T V 6 ;  RC20  (P-Tyr) Western  F I G U R E 6. Generation of E T V 6 - N T R K 3 (EN), EN-A614, and EN-Y615F expressing cells and assessment of transformation. (A) Sequence alignment of the C-terminal 30 amino acids of the three members of the N T R K family of protein tyrosine kinases: NTRK1, NTRK2, and NTRK3. Residues identical to the NTRK3 sequence are indicated by (.). Numbers delineate amino acid boundaries of each protein segment. This region of NTRK3 is considerably different from other members of the N T R K family, suggesting it may contain unique signaling residues. (B) Sequence alignment of the same C-terminal 30 amino acids of NTRK3 between four different vertebrate species. (*) indicates the position of tyrosine 615 (Y615) of EN. Conservation of this region suggests that it may contain essential residues for unique NTRK3 signaling. (—) indicates a potential region for IRS-1 PTB binding. (C) Retroviral gene transfer was used to produce E N - and E N mutant-expressing NIH3T3 cells. To confirm protein expression, whole cell lysates were subjected to immunoprecipitation with a-ETV6 antibodies  73  followed by western blotting with anti-phosphotyrosine antibodies. Arrows highlight the 73/68 kDa doublet of E N and EN-Y615F and the predicted 70/66 kDa doublet of EN-A614. Note that both EN-A614 and EN-Y615F are also tyrosine phosphorylated.  EN-A614 fails to associate with IRS-1.  To determine whether these E N mutants could associate with IRS-1, NIH3T3 fibroblasts were retrovirally infected with E N , EN-A614, EN-Y615F, or vector (MSCV) alone. Lysates prepared from serum-starved cells were subjected to immunoprecipitation using a-ETV6 antibodies followed by immunoblotting with a-phosphotyrosine antibodies. Consistent with our previous findings (215), E N cells demonstrated a tyrosine phosphorylated band of -180 kDa (see Figure 7A; top panel), which was confirmed to represent IRS-1 by re-probing with a-IRS-1 antibodies (data not shown). Levels of this band were consistently lower in immunoprecipitates from EN-Y615F cells (Figure 7A; top panel), and completely absent (either phosphorylated or non-phosphorylated) in those from EN-A614 or vector (MSCV) control lysates (Figure 7A; top panel). Immunoprecipitation from the same lysates with a-IRS-1 antibodies followed by immunoblotting with a-phosphotyrosine antibodies demonstrated that total IRS-1 tyrosine phosphorylation was correspondingly absent or reduced in EN-A614 or EN-Y615F cells, respectively (Figure 7A; middle panel). To confirm that equal levels of IRS-1 were present in each cell line, lysates were analyzed by western blotting using an a-IRS-1 antibody (Figure 7A; bottom panel). To further demonstrate that EN-A614 does not interact with IRS-1, HEK293T cells were transiently transfected with HA-tagged IRS-1 and either V5-tagged E N , kinase-dead E N (K380N), or EN-A614 constructs. Lysates were immunoprecipitated with a-V5 antibodies,  74  >  U  •J-J  IP: ETV6  §  Z I SO >  W  so  sO  <  IB: P-Tyr IP: IRS-1 IB: P-Tyr Total Cell Lysate IB: IRS-1  + HA-IRS-l  B i  F I G U R E 7. IRS-1 is not constitutively tyrosine phosphorylated in EN-A614 expressing NIH3T3 cells. (A) N I H 3 T 3 cells infected with M S C V control, E N , EN-A614, and E N - Y 6 1 5 F retroviral constructs were grown in 100 mm dishes to 75% confluence and then serum-deprived for 18 hours in 0.5% serum. Whole cell lysates were prepared, immunoprecipitated with an a E T V 6 antibody, separated by S D S - P A G E , and probed with anti-phosphotyrosine antibodies (PTyr). The top panel shows a differentially tyrosine phosphorylated band at approximately 180 kDa, which was confirmed to be IRS-1 by re-probing with a-IRS-1 antibodies (data not shown). To confirm the presence of IRS-1 in each lysate, 50 [ig of whole cell lysate was separated by S D S - P A G E and probed with an a-IRS-1 antibody (bottom panel). (B) H E K 2 9 3 T cells were transiently transfected with H A - I R S - 1 and either V5-tagged E N , E N - A 6 1 4 mutant, or kinasedead E N (K380N), lysed, and immunoprecipitated with a - V 5 antibody. A s expected, both the E N and E N - A 6 1 4 fusion proteins were tyrosine phosphorylated, whereas the kinase-dead mutant ( V 5 - K 3 8 0 N ) was not. E N interacted with and phosphorylated IRS-1, while kinase-dead E N did  75  not. Interestingly, EN-A614 was not able to immunoprecipitate IRS-1 (middle panel). Expression of V5-tagged constructs (data not shown) and equal levels of IRS-1 expression (bottom panel) were confirmed with a-V5 and a-HA immunoblots.  followed by western blot analysis with a-phosphotyrosine antibodies. As shown in Figure 7B, a tyrosine phosphorylated band of -180 kDa was only pulled down in E N lysates, the identity of which was confirmed to be IRS-1 by re-probing with a-HA (Figure 7B) or a-IRS-1 antibodies (data not shown). Equal expression of IRS-1 was confirmed by western blotting of lysates using a-HA antibodies (Figure 7B). These data indicate that the distal 19 C-terminal amino acids of E N are essential for the interaction of E N with IRS-1 and for the increased total IRS-1 tyrosine phosphorylation observed in EN-transformed cells.  EN-A614 and EN-Y615F  are defective in their transformation  activity.  To determine whether EN-A614 and EN-Y615F retain E N transformation activity, NIH3T3 cells stably expressing each construct were assessed morphologically. EN-A614 expressing cells exhibited a non-transformed phenotype that was identical to controls cells (data not shown). EN-Y615F cells showed partial morphologic transformation, with increased spindling, elaboration of cellular processes,  increased nuclear-to-cytoplasmic ratios and  refractility, and evidence of focus formation. To further characterize the transforming ability of these mutants, cells were plated in soft agar to determine their ability to grow under anchorage independent conditions. Both EN-A614 and EN-Y615F expressing cells formed only occasional macroscopic soft agar colonies at rates that were significantly reduced compared to cells expressing similar levels of E N (EN versus EN-A614 cells, p< 0.001; E N versus EN-Y615F cells, p<0.001; Figure 8A). We next investigated the transformation activity of EN-A614 and E N -  76  Y615F mutants in vivo. NIH3T3 cells expressing similar levels of each E N construct (data not shown) were injected subcutaneously into SCID mice and observed for tumour formation over a period of 20 days, at which time the size of tumours in E N positive control mice necessitated the termination of the experiment (Figure 8B). After 20 days, there were no detectable tumours in any of the 12 injection sites in either EN-A614 or the empty vector (MSCV) control. Fibroblasts expressing EN-Y615F were able to form small tumours in SCID mice, averaging 1127 mm at 20 days, compared with 4086 mm for E N (p<0.01). Therefore EN-A614 and to a lesser extent 3  the EN-Y615F mutant are defective in E N transformation, even though these proteins remain tyrosine phosphorylated. Taken together, these data indicate that the distal 19 C-terminal amino acids of E N mediating IRS-1 binding are essential for E N transformation activity.  Mekl/2  and AKT activation  are blocked  in EN-A614  expressing  NIH3T3  cells.  If IRS-1 links E N to serum-independent constitutive activation of the Ras-MAPK and PI3K-AKT pathways and to constitutive over-expression of cyclin D l (210, 215), then mutants lacking the IRS-1 binding region should fail to activate these cascades. To assess this the activation states of Mekl/2 and A K T , as well as cyclin D l levels were assessed in NIH3T3 cells expressing E N , EN-A614, EN-Y615F, and vector alone. Levels of phosphorylated (activated) Mekl/2 and A K T as well as cyclin D l were markedly reduced in EN-A614 expressing cells after serum starvation compared to cells expressing E N (see Figure 9). These patterns were reproducible over numerous independent experiments, and strongly indicate a role for IRS-1 binding in constitutive activation of Ras-MAPK and PI3K-AKT pathways in EN-transformed cells. Interestingly, EN-Y615F was able to activate Mekl/2 and to elevate cyclin D l / 2 to almost the same degree as E N , while A K T activation was consistently moderately lower than in  77  50%  F I G U R E 8. Assessment of the transformation ability of EN-A614 and EN-Y615F. A , anchorageindependent growth was assessed by the ability of cells to form macroscopic colonies (greater than 0.1 mm in size) when plated in soft agar. Histogram comparing colony formation of NIH3T3 cells infected with vector alone (MSCV) or constructs containing E N , EN-A614, or EN-Y615F. Results are expressed as the ratio of colonies formed per number of cells seeded. Each cell line was assessed in triplicate in 35 mm wells, and each experiment was performed at least six times. Statistical analysis was performed using the Mann-Whitney ranks test: P-value comparing E N and EN-A614 (*) cells (p< 0.001); P-value comparing E N and EN-Y615F (**) cells (p<0.001). B, NIH3T3s expressing M S C V (vector control (—)), E N (•), EN-A614 (•), and EN-Y615F (•) were injected subcutaneously into SCTD mice (2 x 10 cells / injection site x 4 sites/mouse). Mice were euthanized after 20 days. Growth curves shown are representative tumour volumes for each of the cell lines tested, as measured using a caliper. Statistical analysis was performed using the Mann-Whitney ranks test: P-value comparing E N and EN-Y615F cells (p<0.01). 6  78  E N cells (Figure 9). This slight discrepancy suggested that EN-Y615F might retain signaling to Ras-MAPK via other pathways. For example, it is possible that EN-Y615F (and EN) also activate Ras-MAPK through the PLCy-PKC pathway by binding P L C y l at E N tyrosine 628 (129). However, cells expressing either EN-Y628Q (an E N mutant which is unable to bind PLCyl (129)) or an EN-Y615F/Y628Q double mutant were both were comparable to E N in Mekl/2 activation and cyclin D l expression (Figure 9). This strongly rules against PLCyl binding and subsequent P K C activation as a major contributor to Ras-MAPK pathway activation in E N and EN-Y615F expressing cells. Consistent with this, non-PLCyl binding EN-Y628Q mutants retain full E N transformation activity (129). Another possibility suggested by the above findings is that E N (and therefore EN-Y615F) can also link to the Ras-MAPK cascade by activation of She, which is well documented to function as an adapter between N T R K receptors and the Ras-MAPK pathway (166). However, as in previous studies (129, 210), we failed to detect interactions of E N or the above mutants with She, nor was there evidence of She tyrosine phosphorylation in cells expressing these chimeras (data not shown). Therefore She is unlikely to play a major role in Ras-MAPK activation by E N . Taken together, our findings demonstrate that the last 19 C-terminal amino acids of E N are essential for its ability to constitutively activate Ras-MAPK and PI3K-AKT pathways and to induce cyclin D l over-expression, but that this is not due solely to the presence of an intact Y615 residue. Therefore the failure of EN-A614 and EN-Y615F to fully transform cells is likely due to defects in IRS-1 mediated activation of these transformation associated pathways.  EN-A614  fails to recruit Grb2 and p85 via  IRS-1.  Upon insulin or IGF stimulation of their respective receptors, IRS-1 becomes tyrosine phosphorylated and recruits SH2-containing proteins such as Grb2 and the PI3K p85 regulatory  79  o CO r|  ^O  —  fa  O  o >  VO  r< vo  o  ir,  >  fa  "r,  oo  !*  vo  !PH  VO  < > P-Mek P-Akt H | = Cyclin D l / 2 total Mek  F I G U R E 9. Differential M e k l , cyclin D l and A K T activation in E N - and E N mutantexpressing NIH3T3s. NIH3T3s infected with vector control (MSCV), E N , EN-A614, and E N Y615F expressing constructs were grown in 35 mm dishes to 75% confluence and then serumdeprived in 0.5% serum D M E M for 18 hours. Cells were then stimulated with (+) or without (-) 9% serum for 1 hour. Whole cell lysates were collected and used for western blot analysis with antibodies to activated A K T (p-AKT), activated Mekl/2 (p-Mek), or cyclin Dl/2. Western blotting using a total Mek antibody was used to demonstrate equal protein loading.  subunit, in turn activating downstream Ras-MAPK, PI3K-AKT and other pathways (263). Previous studies showed that tyrosine phosphorylated IRS-1 within E N / IRS-1 complexes recruits both Grb2 and the p85 subunit (215). Therefore the levels of Grb2 and p85 associated with  IRS-1  in  cells  expressing  EN,  EN-Y615F,  and  EN-A614  were  analyzed.  Immunoprecipitation from lysates of E N - and mutant-EN-expressing cells were performed under  80  serum-starved  conditions  using  IRS-1  antibodies.  Similar  levels  of  IRS-1  were  immunoprecipitated from each cell line (Figure 10, top panel). As expected, in E N expressing cells IRS-1 was constitutively tyrosine phosphorylated and its association with both Grb2 and p85 was readily detected (Figure 10). However, EN-Y615F cell lysates showed reduced IRS-1 tyrosine phosphorylation and diminished recruitment of Grb2 and p85, while in EN-A614 cell lysates there was no apparent association between IRS-1 and Grb2 and p85. These data strongly suggest a direct correlation between E N associated IRS-1 tyrosine phosphorylation and the ability of IRS-1 to recruit Grb2 and p85, and suggest that the capacity of EN-Y615F to activate Ras-MAPK and PI3K-AKT compared to EN-A614 is likely due to the fact that it retains limited ability to bind IRS-1.  B u CO  i §  H W  2  vo <  o  IB: IRS-1 IP: IRS-1 <  180 kDa IB: P-Tyr IB: p85 IB: Grb2  Total cell lysate  Grb2  F I G U R E 10. EN-A614 does not associate with p85 or Grb2 via IRS-1. Whole cell lysates were prepared, immunoprecipitated with an a-IRS-1 antibody, separated by SDS-PAGE, and probed with an anti-phosphotyrosine antibody (P-Tyr), as well as antibodies to p85 and Grb2. E N expression is associated with constitutive IRS-1 phosphorylation, and subsequent  81  recruitment of p85 and Grb2. EN-A614-expressing cells do not show constitutive IRS-1 phosphorylation, and are unable to recruit Grb2 and p85. EN-Y615F expression leads to reduced tyrosine phosphorylation of IRS-1 (compared to EN), and consequently is associated with lower levels of Grb2 and p85. To confirm equal protein loading, lysates were probed with an a-Grb2 antibody.  Co-expression  of IRS-1 C reduces EN-mediated  transformation.  Since E N was shown to specifically interact with the PTB domain of IRS-1, we postulated that overexpression of an IRS-1 fragment containing this domain (IRS-1C) should be able to block the E N / IRS-1 interaction and thus E N transformation in a dominant negative fashion. Retroviral gene transfer was used to co-express E N along with either HA-IRS-1C or empty  vector  in  NIH3T3  fibroblasts.  Levels  of  E N expression  were confirmed by  immunoprecipitation with 0C-NTRK3 antibodies, and expression of HA-IRS-1C was confirmed by a - H A western blotting (data not shown). Lysates from the same cells were then subjected to immunoprecipitation with an a-IRS-1 antibody that recognizes the C-terminus (but does not bind IRS-1C). Significantly less IRS-1 associated with E N in cells co-expressing E N and IRS-1C compared to cells expressing E N alone, despite higher levels of E N in the co-expressing cells (data not shown). All lines were shown to contain equal amounts of endogenous IRS-1 (data not shown). These cells were then assessed for their ability to form colonies in soft agar. Cells coexpressing both E N and the IRS-1C showed a significant reduction in colony formation compared with cells expressing E N alone (p<0.0001; Figure 11). Together, these data indicate that IRS-1C can function as a dominant negative regulator of EN-mediated transformation by reducing the interaction with endogenous IRS-1.  82  IRS-1 overexpression potentiates EN-mediated  transformation.  The above findings underscore the importance of IRS-1 in EN-mediated transformation. This led us to assess the influence of IRS-1 expression levels on E N transformation activity. To study this we retrovirally overexpressed IRS-1 in EN-transformed fibroblasts. We used a different murine fibroblast cell line for these studies, namely murine R+ cells. These cells are derived from insulin-like growth factor receptor 1 (IGF1R) null R- cells ((264, 265)) that have been engineered to re-express IGF1R (215).  MSCV  MSCV / IRS-1C  EN  IRS-1C / EN  F I G U R E 1 1 . Co-expression of I R S - 1 C (PTB/PH domains) disrupts E N / IRS-1 complexes. Anchorage-independent growth of NIH 3T3s expressing E N and/or IRS-1C constructs was assessed by the ability to form macroscopic colonies in soft agar. Each cell line was assessed in triplicate in 35 mm wells, and the experiment was repeated five times. Cells expressing both HA-IRS-1C and E N show a significantly lower rate of colony formation than those expressing E N alone (p<0.0001), as assessed by Mann-Whitney ranks test.  83  We chose R+ cells as we had previously showed that while they are transformed by E N , transformation activity is lower than in NIH3T3 cells (215). IRS-1 infected R+ cells were shown by a-IRS-1 western blot analysis to have significantly higher (at least 3-fold) levels of IRS-1 compared to levels of endogenous IRS-1 in vector alone R+ cells (M. Martin, personal communication). Soft agar colony assays were used to assess transformation activity. R+ and R+/IRS-1 cells showed little or no colony formation as expected (Figure 12). Consistent with previous results (215), R+/EN cells formed macroscopic colonies at a lower rate than typically observed for EN-transformed NIH3T3 cells (compare to Figure 11). However, in the presence of IRS-1 overexpression, EN-mediated anchorage-independent growth was greatly increased (80%) (p<0.0001; Figure 12). Furthermore, the rate of colony formation was greatly accelerated and colonies were larger in R+ cells expressing both E N and IRS-1 constructs compared to E N alone (data not shown), which was reproducible over numerous experiments. These findings not only confirm the crucial role of IRS-1 in E N transformation, but also suggest that actual IRS-1 protein levels might influence E N transformation activity.  84  p< 0.0001  1  0.9-,  MSCV  IRS-1  EN  =  EN/IRS-1  F I G U R E 12. Overexpression of IRS-1 potentiates E N transformation. The R+ fibroblast cell line was retrovirally transfected with expression constructs of E N and/or full-length IRS-1. Anchorage-independent growth was assessed by the ability to form macroscopic colonies in soft agar, as described in materials and methods. Flistogram comparing R+ control (MSCV) cells with those overexpressing EN, IRS-1, or both E N and IRS-1. Results are expressed as the ratio between colonies and number of cells plated. Each cell line was assessed in triplicate in 35 mm wells, and the experiment was repeated five times. Differences were found to be statistically significant using the Mann-Whitney ranks test (p<0.0001).  85  3.4  DISCUSSION  The t(12;15)-associated ETV6-NTRK3 oncoprotein is similar to other chimeric PTKs such as B C R - A B L (266), ETV6-PDGFR(3 (103), ETV6-JAK2 (149), and N P M - A L K (209) in that it functions as a constitutively active tyrosine kinase. Activation of the E N P T K is linked to induction of downstream signaling pathways of wild-type NTRK3 including Ras-MAPK and PI3K-AKT, leading to elevated cyclin DI expression and aberrant cell cycle progression (210). Rather than utilizing known NTRK3 adapters to link to these pathways, E N does so through an interaction with the IGF1R substrate IRS-1, and E N transformation is associated with constitutive IRS-1 tyrosine phosphorylation (215). Here we demonstrate that the E N / IRS-1 interaction occurs via the PTB domain of IRS-1 binding to the C-terminus of E N . Specifically, the last 19 amino acids of E N are essential for this interaction. A mutant E N protein lacking the distal C-terminal 19 amino acids (EN-A614), though still tyrosine phosphorylated when expressed in NIH3T3 cells, does not bind IRS-1 and is non-transforming. In contrast to E N expressing cells which are characterized by strong constitutive IRS-1 tyrosine phosphorylation even when serum-starved, those expressing EN-A614 show a complete block in total and E N associated IRS-1 tyrosine phosphorylation under these conditions. Moreover, Mekl/2 and A K T activation as well as cyclin DI elevation, hallmarks of E N transformation (129, 210), were markedly reduced in EN-A614 expressing cells after serum starvation compared to cells expressing E N . This directly correlated with the ability of IRS-1 to recruit Grb2 and the p85 subunit of PI3K, which link E N to the Ras-MAPK and PI3K-AKT pathway, respectively (210, 215). In EN-expressing cells, IRS-1 bound both Grb2 and p85, while this was not evident in cells expressing the EN-A614 mutant. These observations indicate that the interaction of the C -  86  terminus of E N with IRS-1 is essential for activation of signaling pathways underlying E N mediated oncogenesis. Numerous reports highlight a potentially more general role for N T R K receptors in oncogenesis. NTRK1  (TRKA) sequences were originally isolated from a colon carcinoma biopsy  as part of an oncogene encoding the amino terminal portion of tropomyosin (TPM3) fused to a truncated tyrosine kinase receptor (267). TPM3-NTRK1 fusions were subsequently detected in papillary thyroid carcinomas (171), and altered N T R K signaling has been implicated in other neoplasms (reviewed in (172)) including pancreatic adenocarcinoma (173), acute myeloid leukemia (174), prostatic cancer (175, 176), breast cancer (177), and human soft tissue sarcomas (178). Recently, potentially activating mutations in the NTRK3 P T K domain were reported in colon carcinoma (8). Therefore NTRK molecules appear to contribute to oncogenesis in a number of lineages, and it will be important to determine whether IRS-1 binding to wild-type or activated N T R K molecules has a more general role in the oncogenic activity of these receptors. In keeping with this possibility, IRS-1 was recently shown to associate with the NTRK1 protein (268). Tyrosine phosphorylated IJRS-1 proteins are known to efficiently bind a number of SH2 domain-containing proteins involved in activation of downstream signaling pathways, including PI3-kinase p85, Grb2, SHP-2, Nek and Crk (reviewed in (269)). There is increasing interest in the potential role of IRS-1 in oncogenesis. Over-expression of IRS-1 in NIH3T3 fibroblasts leads to increased activation of the Ras-MAPK cascade and cell transformation (270, 271). IRS-1 overexpression also contributes to the progression of hepatocellular carcinoma, possibly by inhibiting transforming growth factor [31-mediated apoptosis (272). Although the LNCaP prostate cancer cell line does not express IRS-1 or has very low levels of IRS-2, introduction of  87  either protein in combination with IGF1R converts these cells to a more aggressive phenotype (273). A recent study examining endogenous IRS-1 shows that it is constitutively tyrosine phosphorylated in a wide range of human tumour samples, suggesting that IRS-1 activation may be a common phenomenon in tumours (274). Moreover, a relationship between IRS-1 activation and fusion oncoproteins has already been established. TRK-T1 (268) and B C R - A B L (275) have both been shown to bind IRS-1 and to be associated increased IRS-1 tyrosine phosphorylation. Therefore IRS-1 activation may be a more general mechanism for transformation mediated by fusion oncogenes. We found that an approximately three-fold elevation in IRS-1 levels can lead to a dramatic increase in the transformation potential of the E N oncoprotein. We hypothesize that this leads to an increased number of IRS-1 molecules to be phosphorylated by E N , thus providing an increased number of docking sites for maximal activation of the Ras-MAPK and PI3K-AKT cascades. Conversely, a dominant negative IRS-1 construct partially blocked E N transformation in our studies, further suggesting that stoichiometric relationships between E N and IRS-1 are important in E N transformation. In this study we found that the phosphotyrosine binding (PTB) domain of IRS-1 mediates its association with E N . PTB domains were initially described to preferentially bind to proteins with a phosphorylated NPXpY motif (where X is any amino acid and pY is phosphotyrosine) (261, 276-278). The NTRK3 portion of E N does not contain the N P Q Y sequence which may be used by wild-type NTRK3 to bind IRS-1, due to the position of the fusion point (2). No other classical N P X Y motifs are found in the E N protein. However, close inspection of the C-terminal 19 amino acids of E N necessary for the IRS-1 interaction reveals several similarities to those associated with classical N P X Y motifs (Figure 6B). First, amino acids 625-628 within these Cterminal 19 residues of E N represent a TPIY sequence. While this varies from a classical N P X Y  88  sequence, threonine (T) has structural similarities to asparagine (N). Therefore the TPIY sequence may structurally mimic an N P X Y motif. Second, it is known that the PTB-binding N P X Y motifs of several proteins including IGF1R, the IL-4 receptor and N T R K A contain hydrophobic residues immediately N-terminal to their N P X Y motifs (reviewed in (260)). In E N , residues 620-624 N-terminal to the TPIY motif are A L G K A (Figure 6B), all of which are potentially hydrophobic, including lysine (K). This region of E N may represent an NPXY-like motif that binds the IRS-1 PTB domain. An obvious flaw in this model is the fact that when the tyrosine residue within the TPIY motif (Y628 - the PLCyl-binding tyrosine residue of EN) is mutated, resulting EN-Y628Q mutants still retain full transformation activity (129,  215).  However, studies have shown that PTB domains can bind non-phosphorylated N P X Y motifs with high affinity or even independently of the N P X Y consensus sequence, suggesting that there is greater plasticity in PTB-binding motifs than previously appreciated (reviewed in (260)). Cells expressing the EN-Y615F mutant showed reduced soft agar colony formation and tumourigenic activity. EN-Y615F bound less tyrosine phosphorylated IRS-1 than E N , and that total IRS-1 tyrosine phosphorylation was decreased compared to EN-expressing cells. Moreover, this correlated with reduced binding of Grb2 and the PI3K p85 subunit to IRS-1. Although tyrosine residue 615 (Y615) is proximal to the TPIY motif, it lies within the C-terminal 19 amino acids deleted in the non-IRS-1 binding EN-A614 mutant. It is possible that Y615 stabilizes IRS-1 binding, or that the Y615F mutation somehow affects the three-dimensional structure of E N leading to reduced binding of IRS-1 to EN-Y615F. Interestingly, while A K T phosphorylation was moderately decreased compared to EN-expressing cells, the EN-Y615F mutant still induced Ras-MAPK activation and cyclin D l expression at levels that were almost equivalent to E N (Figure 9). The Sorensen laboratory previously showed that both the Ras-  89  M A P K or PI3K-AKT pathways are required for E N transformation, and that it is the former that regulates cyclin D l expression (210). Therefore the observed decrease in transformation activity in EN-Y615F cells could be accounted for by the decrease in PI3K-AKT activation even though cyclin D l elevation is retained. However, the fact that one of the two transformation-associated pathways was attenuated in EN-Y615F cells to a greater degree than the other suggests additional complexities in how IRS-1 links E N to these cascades. These findings do not support a role for IRS-1 independent links to the Ras-MAPK cascade by the PLCy-PKC pathway or through the She adapter protein, although contributions by other pathways cannot be ruled out. The most likely explanation for the observed discrepancy is that the absolute level of IRS-1 binding  differentially  regulates  activation  of  downstream  pathways  mediating  EN  transformation. In this scenario, perhaps sufficient IRS-1 and subsequently Grb2 is associated with EN-Y615F to attain near maximal Ras-MAPK activation, but the amount of p85 subunit binding to the EN-Y615F/IRS-1 complex is below the threshold level required for full A K T activation. This model is corroborated by our finding that increasing IRS-1 levels in NIH3T3 cells can dramatically increase the transformation potential of E N , suggesting that IRS-1 levels may be rate-limiting for E N transformation. If, as our studies suggest, IRS-1 plays a pivotal role in E N transformation (and potentially that of activated NTRK3 in other tumours), then blocking the E N / IRS-1 interaction offers an interesting avenue for potential cancer therapeutics. Over-expression of the dual specificity phosphatase P T E N reduces the overall level of IRS-1 phosphorylation and induces growth arrest in the MCF-7 breast cancer cell line (279). An IRS-1 molecule in which all 18 tyrosines were mutated to phenylalanine, thus negating tyrosine phosphorylation, can act in a dominant negative fashion to reduce anchorage-independent growth of breast cancer cells (274). Expression of an  90  N-terminal portion of IRS-1 blocks the tumourigenic phenotype of human hepatocellular carcinoma (272). A recent report shows that over-expression of the IRS-1 PH/PTB domains blocks binding of IRS-1 to the human JC virus T-antigen and prevents subsequent cellular transformation by this protein (280). In our study a truncated PH/PTB IRS-1 fragment disrupted the E N / IRS-1 interaction and reduced EN-mediated transformation. Therefore agents blocking the interaction between the C-terminus of E N and the IRS-1 adapter may provide a novel approach for treatment of EN-expressing tumours.  91  C H A P T E R IV  THE C-TERMINUS OF ETV6-NTRK3 IS ESSENTIAL FOR TRANSFORMING ACTIVITY  92  4.1  INTRODUCTION The kinase domain is the most conserved domain among tyrosine kinase receptors, and  an intact PTK domain is absolutely required for receptor signaling. Mutations of a single lysine in  the A T P binding site, prevents  A T P binding to the  active  site, blocks receptor  autophosphorylation, and subsequently completely inactivates receptor biologic function (10). Further, the tyrosine kinase domain of the RTKs  is subject to autoinhibition in the  unphosphorylated basal state via steric interactions involving the activation loop. A mutation in the activation loop designed to relieve autoinhibition, asparagine to alanine, substantially increases the ability of the unphosphorylated kinase of the insulin receptor to bind A T P (281).  The carboxy-terminal tail sequences are among the most divergent sequences between known RTKs (282). However, the carboxy-terminal domain of the receptor is thought to play an important role in regulating kinase activity (10, 36). This region typically contains several tyrosine residues, which are phosphorylated by the active kinase. In fact, the receptor itself is quite often the major tyrosine phosphorylated species following ligand binding (10). The presence of specific amino acid residues in this domain is integral to the activation of specific downstream signaling effectors (10). We have demonstrated that an E N protein mutated at both Y615F and Y628Q (ENY615F+Y628Q, Chapter III) still activates (phosphorylates) Mekl/2 and increases levels of cyclin D I . However, previous research in the Sorensen laboratory has suggested that the Y628 may be associated with IRS-1 binding. Morrison et al. showed that in NLH3T3s expressing E N Y628Q, binding to IRS-1 (as measured by ability to co-immunoprecipitate) is decreased compared to E N (215). Interestingly, tyrosine phosphorylation of IRS-1 is not concomitantly  93  decreased in Y628Q-expressing cells. This suggests that there may be multiple tyrosine residues involved in the putative EN-IRS-1 interaction, and that the Y628 residue is one of these sites. The A614 and Y615F mutations identified and discussed in the previous chapter abrogate and diminish IRS-1 binding (and subsequent transformation) through an unknown mechanism. We therefore hypothesized that there may be residues within the region of the protein truncated by the A614 mutation (see Figure 13) that are essential for the transforming ability of E N .  4.2  RESULTS  Can TPIY sequence function  as an NPXY-like  motif?  We have demonstrated that an interaction between E N and and the PTB domain of LRS-1 is essential for transformation (215, 283). It was proposed in Chapter III that the TPIY sequence at position 625-628 of E N resembles the N P X Y motif that binds PTB domains. Threonine (T) has some structural similarities to asparagine (N), and therefore the TPIY sequence may structurally mimic an N P X Y motif. Wai et al. previously demonstrated that mutation of the tyrosine at position 628 (Y628, the PLCyl -binding site) is dispensable for transformation (129). However, it is known that PTB domains can bind non-phosphorylated N P X Y motifs with high affinity (reviewed in (260)). Several studies into PTB binding have demonstrated that alanine mutants of the asparagine (N) and proline (P) residues of the N P X Y motif either greatly decrease or completely abolish binding (261, 277). We therefore made alanine mutants of the residues as well as a glutamic acid (E) mutation of the isoleucine (I) at position 627 (primers used listed in Table 4, Materials and Methods), and retrovirally transfected fibroblasts to express these mutants (see Figure 13 for the position of all mutations created for this Chapter). The PTB domain of IRS-1 favors a small, hydrophobic amino acid (such as alanine) at the pY-1 position for high  94  affinity binding; substitution of alanine with glutamic acid results in a significant reduction in IRS-1 binding (284, 285). Therefore, isoleucine to glutamic acid mutation at position 627 (pY-1), may function as a low affinity mutant. NIH3T3s infected with virus for the single T625A mutant did not survive drug selection, presumably due to low viral titres. However, He et al. showed that a double alanine mutant (both asparagine and proline residues) exhibit the same reduction in PTB binding as either single mutant (277), and therefore there was no attempt to remake this single mutant.  T625 P626  human pig rat chicken  PQQRLNIKE 3YKILHALGKATPIYLDILG PQQRLNIKEIYKILHALGKATPIYLDILG PQQRLNIKEIYKILHALGKATPIYLDILG PQQRLNIKEIYKILHALGKATPIYLDILG  FIGURE 13. The C-terminal 29 amino acids of EN, indicating the position and sequence of EN mutants. We hypothesize the TPIY sequence may function as an N P X Y (Phosphotyrosine Binding, PTB) motif. The A624 construct is truncated at the position of the arrow, and is therefore missing the most distal nine amino acids.  NIH3T3s  over  expressing  all  of  the  TPIY  mutants  exhibited morphological  characteristics of transformation (Figure 14), including refractility, spindling, and a high nuclearto-cytoplasmic ratio, and closely resembled EN-expressing cells. Empty vector (MSCV) and A614 cells are non-transformed, and were used as a negative control (Figure 14). To further  95  assess the role of these mutants in transformation, NIH3T3s expressing these mutants were assessed by both the soft agar assay and injection into immunocompromised mice. Both the P626A and the T625A+P626A markedly reduced soft agar colony formation, while as expected an I627E mutation had no effect (data not shown). As both the P626A and T625A+P6262A mutants behaved similarly in the soft agar assay, only the P626A mutant was assessed for growth in immunocompromised mice. To confirm the findings of Wai et al. that the PLCyl binding site (Y628) was dispensable for transformation (129), the Y628Q mutant was included in this study. Equal levels of each construct (EN, P626A, Y628Q) were assessed prior to injection into four sites per mouse, with one million cells per injection site (Figure 15). This method allows the investigator to use a minimal number of mice (three per cell line), while achieving required statistical power. Cells containing M S C V (empty vector) control were also injected. As previously published, E N expressing cells readily formed tumours, and the Y628Q mutant had no statistically significant effect on growth. The P626A mutant drastically reduced tumour formation in this assay, forming tumours averaging 1437 mm after 23 days, compared with 3  2866 mm in the E N positive control group (p< 0.05) and 2196 mm in the Y628Q group (Figure 15).  vector  A614  I627E  F I G U R E 14. N I H 3T3s expressing TPIY ('NPXY') mutants display transformed morphology. NIH3T3s were retrovirally infected with E N , mutant E N , or empty vector and assessed for morphologic criteria of transformation as described in Materials and Methods. Photomicrographs taken using a lOx objective.  97  PLCy  IP: ct-Tel WB: P-Tyr  F I G U R E 15. EN-P626A reduces tumour growth in SCID mice. Equal levels of phosphorylated protein were confirmed prior to injection into immunocompromised mice (top). Bottom panel shows reduced growth of EN-P626A cells, compared to E N (p<0.05). The difference between Y628Q and E N was not found to be statistically significant.  98  Y628Q is not the second tyrosine site crucial for  transformation  Studies with the PDGFR(3 have shown that the contributions of the PLCyl-binding tyrosine could only been observed after elimination of the substantially stronger PI3K signal transduction cascade (286, 287). These studies have suggested that PLCy activation by the PDGFRP may be negatively regulated (288), and that RTKs may selectively activate PI3K and PLCy signaling pathways, either individually or collectively (289). Previously, it had been shown that less Y628Q mutant co-immunoprecipitated with IRS-1 (compared with EN), suggesting that this site may play a role in IRS-1 binding (215). Since the Y615F mutant has already been shown to be diminish IRS-1 binding, we therefore wished to assess a double mutant, Y615F+Y628Q for both IRS-1 binding and transforming ability, as well as a deletion mutant that eliminates this proposed IRS-1 binding site (EN-A624). This deletion mutant would effectively eliminate this region, and subsequently eliminate IRS-1 binding (and potentially transformation) if the hypothesis is true. To determine whether EN-A624 and EN-Y615F+Y628Q retain E N transformation activity, NIH3T3 cells stably expressing each construct were assessed morphologically. Both EN-A624  and EN-Y615F+Y628Q expressing  cells exhibited  evidence  of morphologic  transformation, with increased spindling, elaboration of cellular processes, increased nuclear-tocytoplasmic ratios and refractility, and evidence of focus formation (data not shown). As expected, an interaction with PLCy with this mutant was not detectable (as the known binding site had been eliminated). Interestingly, deletion of the distal nine amino acids (A624) had no effect on colony formation in two separate sequence-verified preparations of this mutant.  99  I 'a  1.  MSCV  EN  Delta 624  Y615F + Y628Q*  Y628Q  F I G U R E 16. EN-A624 mutant does not affect colony growth in soft agar assay. The A624 mutant (missing the TPIY residues) forms a similar amount of colonies in soft agar as E N . *Y615F+Y628Q double mutant was able to form colonies in soft agar, but these colonies were visibly much smaller than colonies formed by other cell lines (investigated further in Figure 17).  As colonies formed by the Y615F+Y628Q double mutant appeared smaller than those of EN-expressing cells, we next assessed colony size quantitatively. The soft agar was employed as described, and the diameter of the first 60 random colonies was measured with an ocular micrometer. After 15 days on culture, colonies from E N and Y628Q expressing cells measured 0.34 and 0.27 ^m, respectively. Y615F colonies had reached 0.27 u.m, while Y615F+Y628Q double mutant expressing cells only 0.17 \im, suggesting that in combination with the Y615F mutations, Y628Q mutations decreased colony size more than either single mutation alone (Y615F+Y628Q versus either Y615F or Y628Q, p<0.005).  100  0.40  0.30  ^  0.20  0.10  0.00 MSCV  EN  Y628Q  Y615F+Y628Q  Y615F  F I G U R E 17. Y615F+Y628Q double mutant reduces colony size in soft agar. NIH3T3s infected with various constructs were plated in soft agar, and then assessed for colony size by measuring colony diameter from 60 random colonies. Y615F+Y628Q double mutant formed significantly smaller colonies than other mutants and EN-expressing cells. Note, Y615F mutant showed reduction in colony formation as previously discussed; the colonies that did form, however, were of similar size to those found in E N .  We  next investigated transformation activity of EN-A624 and EN-Y615F+Y628Q  mutants in vivo. NIH3T3 cells expressing similar levels of each E N construct (Figure 18) were injected subcutaneously into SCID mice and observed for tumour formation over a period of 22 days, at which time the size of tumours in EN-A624 and E N positive control mice necessitated the termination of the experiment (Figure 18). After 22 days, fibroblasts expressing E N and E N -  101  A624 were able to form large tumours averaging 2745 mm and 2596 mm , respectively (no 3  3  statistical significance on this difference); empty vector (MSCV) control achieved barely palpable small growths averaging 57 mm . Interestingly, fibroblasts expressing Y615F+Y628Q were able to form small tumours in SCID mice, averaging 841 mm (p<0.05 for E N versus 3  Y615F+Y628Q at all time points measured) (Figure 18).  EN C-terminal  mutants block anchorage-independent  growth  Spheroids are multi-cellular structures of intermediate complexity between in vivo tumours and monolayer cultures and, as such, may demonstrate biologic characteristics that are more closely related to those of primary tumours (290, 291). To study growth under nonadherent conditions, sub-confluent, retrovirally-infected monolayer cultures were trypsinized and placed as single cell suspensions in medium overlayed on agar-coated dishes (preventing attachment to plastic), and their growth characteristics monitored morphologically. Within one hour of replating, the single cells began to form loose clumps that continued to grow in density over time (including vector control). However, by 48 hours spheroid formation was beginning to disappear in the non-transformed cell lines, and had completely disappeared by 96 hours, presumably through induction of apoptosis. Conversely, in dishes with transformed cell lines (EN, A624, and I627E in particular), almost no single cells remained and irregular shaped clumps were evolving into tight spheroidal structures (see Figure 19). This ability to form multicellular aggregates was highly dependent on seeding density, with at least 10 cells plated in a 6  10cm dish required for induction of spheroid formation in the transformed cell lines. Nonetheless, this assay is particularly useful for assessing anchorage independent growth rapidly, and results  3500 I  3000  ^2500 |  2000  | 1500 § j— 1 0 0 0  T  I  e  IP: a-ETV6  WB: P-Tyr  J  500 i  —S 0  IT  j  T  J  5 - - ^ = jE  "  •—i  •  r  *  i ^—i  D a y 1 2 D a y 14 D a y 16 D a y 1 8 D a y 2 0 D a y 2 2 D a y 2 4  E N  Delta624 — H - Y 6 1 5 F + Y 6 2 8 Q  —•—MSCV  F I G U R E 18. Y615F+Y628Q mutant, but not the A624 mutant, decrease tumour formation in an immunocompromised mouse injection model. Prior to injection, NIH3T3s expressing the four constructs under study were analyzed by IP / western to confirm equal levels of phosphorylated protein prior to injection (left panel). Two million NIH 3T3s expressing E N and mutant E N at equal levels were injected sub-cutaneously into nude mice. E N and A624 expressing cells formed tumours at approximately the same rate. Interestingly, Y615F+Y628Q-expression resulted in decreased tumour formation compared to E N (p<0.05) (right panel).  F I G U R E 19. E N C-terminal mutants block anchorage-independent growth (spheroid formation). NIH3T3s expressing E N and mutants (at relatively equal levels, data not shown) were plated on plates coated with 1.4% agarose. Photomicrographs were taken at low power (lOx objective) four days after plating, and are representative of three experiments. Cells expressing M S C V and A614 were not able to form colonies. Cells expressing Y615F, Y628Q, Y615F+Y628Q, P626A, and T625A+P626A form small aggregates of cells of varying cell number. Interestingly, Y615F and T625A+P626A mutations create problems with cell adhesion.  ©  104  correlate  well  with  colony  formation  in soft  agar and ability to  form tumours in  immunocompromised mice.  EN C-terminal  mutants bind IRS-1  We hypothesized that the TPIY sequence at the C-terminus of E N acted as a PTB-like manner to bind IRS-1, and the impaired transforming ability of mutants of the residues was a result of altered ability to bind IRS-1. To assess this, NIH3T3s expressing E N and various mutants with the HA-tagged PTB domain of IRS-1 were transiently transfected (HA-IRS-1C, described in Chapter III). Lysates from these cells were immunoprecipitated with a-HA antibodies, and probed for the E N doublet with phosphotyrosine antibodies. HA-tagged IRS-1C was able to immunoprecipitate E N proteins from all samples, except from M S C V or A614 (Figure 20). In addition, this demonstrates that the TPIY region is not responsible for IRS-1 binding, and mutations in this region affecting the transforming ability of E N must be affecting other interactors directly or through altered conformation of the protein. Further this suggests that IRS-1 must be either binding E N N-terminal to this region directly through an, as yet, undiscovered mechanism, or indirectly through an unknown adapter protein.  105  F I G U R E 20. E N C-terminal mutants interact with the IRS-1 P T B (phosphotyrosine binding) domain. NEH 3T3s stably expressing E N and EN-mutants at approximately equal levels (data not shown) were transiently transfected with HA-tagged PTB domain of IRS-1, lysed, and subject to immunoprecipitation with anti-HA antibody. The PTB domain of IRS-1 was shown to interact with E N and all mutants except E N - A614. Western blotting for H A detects equal expression of the PTB domain of IRS-1 in all samples.  106  4.3  DISCUSSION We proposed that the TPIY sequence in the C-terminus of E N functioned as an N P X Y -  like motif to mediate IRS-1 binding. Further, this interaction with IRS-1 was essential to the transforming ability of E N . It would appear that our hypothesis regarding the TPIY site was incorrect. While mutation of the threonine at residue 625 and proline at residue 626 each have a profound effect on transformation, the A624 mutant (which truncated the protein N-terminal to this TPIY sequence) was fully transforming, suggesting that these residues are not crucial for transformation. Further, these C-terminal mutants retained their ability to bind IRS-1. Therefore, while EN-induced transformation appears to be associated with IRS-1 binding, IRS-1 binding does not guarantee transformation. These data do suggest that the C-terminus of E N is crucial for transformation. The abrogated and diminished IRS-1 binding observed in the A614 and Y615F mutants, respectively, is most likely due to binding N-terminal to amino acid 614. These mutants must alter the protein conformation is such a manner than interactions N-terminal to this region are unable to occur. It is likely, therefore, that the interaction with IRS-1 involves residues N-terminal to amino acid 614, as mutation of proposed sequences between residues 616-633 did not greatly affect IRS-1 binding. The double mutant Y615F+Y628Q showed reduced association with IRS-1 (Figure 20), but this is most likely due to the effect of the Y615F mutation alone. No significant difference in levels of associated IRS-1 could be consistently observed between Y615F and Y615F+Y628Q mutants. Previous findings from our laboratory have demonstrated that the Y628Q mutant coimmunoprecipitates poorly with IRS-1 (215). However, this study was not able to demonstrate a difference in IRS-1 association between E N and EN-Y628Q. These discrepancies could be due  107  to different antibodies used in these studies for immunoprecipitation of EN-associated proteins (CX-ETV6 in this study versus a-NTRK3 used by Morrison et al.) (215). C-terminal binding or conformational changes induced by the Y628Q mutation could be interfering with the ability of the a-NTRK3 antibody to bind at the C-terminus of the protein. Alternatively, an interaction with IRS-1 could be very transient (owing perhaps to the correct localization of the E N protein), and the Y628 residue may be involved indirectly. These two possibilities could be addressed using protein crystal structures or binding efficiency experiments. It is interesting to observe that while the P626A and T625A+P626A mutants were still able to induce hallmarks of morphological transformation (increased refractility, increased nuclear to cytoplasmic ratio, spindling), these mutants had decreased growth in soft agar and ability to form tumours in immunocompromised mice (P626A). These mutants must be inducing a conformational change that is preventing transformation, perhaps by disrupting binding of an, as yet, undetermined protein. Studies to address the downstream signaling consequences of these mutations are currently in progress, and will provide valuable insight into the oncogenic pathways induced by E N . Traditional cell culture models of sarcomas and many other human solid tumors involve growth of cells as adherent monolayers on plastic dishes in the presence of serum and other growth factors. However, among the in vitro features that differentiate transformed cell cultures from those of normal cells are their decreased growth factor requirements and their ability to grow in an anchorage-independent environment (241). It is likely these properties correlate with the clinical features of malignant tumors (ability to infiltrate surrounding tissues and to establish distant metastases). Spheroids are multi-cellular structures of intermediate complexity between in vivo tumors and monolayer cultures and, as such, may demonstrate biologic characteristics  108  that are more closely related to those of primary tumors (291). Several studies have indicated that anchorage-independent survival and proliferation of tumour cells are dependent on cell-cell adhesion, which is mimicked when tumour cells form multi-cellular spheroids (reviewed in (291). We therefore cultured cell lines expressing E N and E N mutants on agar-coated plates and found differences in their ability to spontaneously form multi-cellular spheroids. We have utilized the spheroid formation assay to rapidly assess transforming ability of site-directed mutants. While this approach works well with most cell lines examined, there are some notably discrepancies between ability to form spheroids and growth in soft agar (or in vivo growth in an immunocompromised host). For example, EN-Y628Q formed colonies in soft agar and grew in the SCID injection assay, but formed significantly smaller spheroids. Perhaps the smaller sized spheroids are a reflection of the reduction in soft agar colony size (Figure 17) and reduction in tumour volume in the SCID injection study. Further, EN-T625A+P6262A was not able to form colonies in soft agar, but formed loose aggregates of cells in the spheroid assay, suggesting that either different signaling pathways are responsible for growth in either assay or that the spheroid assay imposes less stringent criteria than the soft agar assay. It appears that anchorage-dependent growth is more dependent on the PI3K-Akt survival than the Ras-MAPK proliferation pathway. Various mutants deficient in their ability to stimulate Akt (Y615F, Y615F+Y628Q) exhibited decreased spheroid formation and ability to grow in soft agar. This finding is consistent with other data from our laboratory. R-EN cells (fibroblasts deficient for the IGF1R) cannot activate Akt and are unable to maintain long-term spheroid formation; this can be rescued with overexpression of an activated Akt construct (myristoylated Akt) (Matthew Martin and Poul Sorensen, unpublished results). Further, the PI3K inhibitor LY294002 blocks spheroid formation and induces apoptosis in EN-expressing cells, but has no  109  effect on monolayer cultures at the same concentration. The suppression of anoikis (apoptosis resulting from loss of cell-matrix interactions) by NTRK2 through a PBK-Akt-dependent mechanism has recently been reported (292). Further, it is likely that Akt-mediated signaling is responsible for cohesive spheroid formation, as several mutants deficient in activation of this pathway were only able to form loose aggregations. In summary, this study has identified a number of residues important for the transforming ability of E N . It is likely that C-terminal signaling is required for the oncogenic signal transduction induced by E N . Future studies investigating the downstream effectors altered by these mutants, as well as determination of the crystal structure of the kinase domain of NTRK3, will enhance our understanding of these signaling events.  CHAPTER V  EN TRANSGENIC MICE DEVELOP LYMPHOMAS AFTER A LONG LATENCY PERIOD  5.1 RESULTING PUBLICATION & CONTRIBUTION O F INDIVIDUAL AUTHORS This chapter is currently in preparation for submission to International Journal of Cancer.  I acknowledge the following contributors to this work. Pronuclear microinjections and oviduct injections were performed by Mrs. Anita Borowski, technician, Canadian Genetic Diseases Network, Centre for Molecular Medicine and Therapeutics. Figure 27. Dr. Gregor Reid performed FACS analysis on prepared cell suspensions. E N knock-in ES cells were created in the laboratory of Dr. Stuart Orkin, Harvard Medical School.  Ill  5.1  Introduction The  development  of appropriate mouse models would significantly enhance our  understanding of translocation-bearing sarcomas (123). To date, it has proven difficult to recapitulate the human disease using animal models. The most successful model so far is the FUS-DDIT3  (also known as TLS-CHOP)  fusion gene found in myxoid liposarcoma (124). This  fusion functions as an aberrant transcription factor involved in adipocyte differentiation and growth arrest (125). In this model, transgenic mice from two independent founder lines developed transplantable tumours of white fat that morphologically resemble human liposarcoma. Interestingly, tumours only develop in fat tissues, despite the expression of the  Tls-Chop  transgene in most tissues (124). These finding are consistent with the theory that the transforming effect of the fusion gene is restricted to very specific cell lineages, due to contentdependent activity of the fusion gene (123). Transgenic mice have been created to express the Pax3-Fkhr  fusion gene, which is  associated with a subtype of alveolar rhabdomyosarcoma (126). By targeting the Fkhr gene to the Pax3 locus, the fusion gene was expressed in developing neural-crest and muscle precursors. Although developmental anomalies including cardiac defects were observed, no tumours developed in these animals or in chimeric mice after up to 1.5 years (293). Lack of tumour development in these mice may be due to inadequate expression of PAX3-FKHR, or absence of yet to be defined cooperating mutations in the pathogenesis of this disease (123). To determine the mechanism of transformation of ETV6-NTRK3 (EN) in vivo, we decided to create a transgenic mouse model. As the EN gene fusion has been identified in epithelial, mesenchymal, and endothelial cell lineages, we generated transgenic mice expressing full-length E N cDNA under the direction of the two ubiquitously-expressing promoters.  112  Several groups have attempted various strategies to recreate chromosomal translocation in an in vivo model, with varying efficiencies (294-296). As our initial transgenic model had not developed a robust, early-onset tumour phenotype, we began to generate a second mouse model of EN-expressing tumours using an embryonic stem (ES) cell-based transgenic approach (297) in parallel with the above methodology. This technique uses homologous recombination to allow the fusion of sequences with the endogenous gene in ES cells, creating a fusion gene that utilizes the endogenous promoter to control expression, as would occur following chromosomal translocation. This technique (known as a 'knock-in' fusion) has been well established for the study of leukemia-related fusion genes (120, 298-300). We therefore began to create an in-frame fusion of E N oncogene with exon 5 of mouse ETV6, resembling the rearrangement that occurs clinically following a t(12;15)(pl3;q25) in EN-expressing tumours.  5.2  RESULTS  Confirmation  of Construct  Expression  Both constructs for transgenesis were sequenced verified twice to confirm intact, inframe inserts. To confirm expression of E/V-containing vectors, NIH 3T3s were transiently transfected using Lipofectamine™ reagent with either pCX-EN  or pIRES2-£'A/ or parental vector  control. NIH3T3 cells transfected with EN exhibited evidence of a transformed phenotype compared to empty vector control cells (Figure 21a). These transformed cells possessed a high nucleus-to-cytoplasm ratio, their cell bodies refracted light, and their growth was not inhibited by contact to other fibroblasts. NIH3T3 cells infected with vector alone were indistinguishable from wild-type NIH3T3 cells (data not shown). Western analysis of E N transfected cells revealed the presence of the 73/68 kDa phosphorylated doublet distinguishable as E N , with no  113  detectable protein in the empty vector lysates (Figure 21b). Protein expression is much lower in the transiently transfected constructs than the retrovirally infected control (MSCV-EN), most likely due to differences between the two transfection systems. To fully assess the transforming ability of these constructs, cells transiently transfected with EN-containing and empty vector as well as cells retrovirally-expressing E N were tested for their ability to grow under anchorageindependent conditions, as measured by the soft agar assay. Only cells expressing E N from the retroviral vector were able to grow in soft agar, suggesting that the transient transfection system does not permit sufficiently high expression for this assay (Figure 22). Transiently transfected N Ras and myristoylated A K T were also not able to form colonies in soft agar (data not shown), supporting this hypothesis.  114  pCX-EGFP  pCX-EN •* \- ^ wSJ  go**  1 pIRES-EGFP  7,1  pIRES-EN  IP: a-TrkC F I G U R E 21. E N constructs for transgenesis are able to induce morphological transformation in NIH3T3s. NIH3T3 cells were transiently transfected with either empty or EN-containing vector to verify expression. (A) Cells transfected with empty vector (pCX and pIRES-EGFP) display a flattened, contact-inhibited, non-transformed phenotype, while cells with EN-containing vectors display a transformed phenotype with increased refractility, loss of contact inhibition and focus formation in monolayer culture. (20x objective) (B) Cells from above were lysed and assessed for E N protein by immunoprecipitation and western blotting. The 68/73 kDa E N characteristic doublet was observed in p C X - E N and pIRES-EN only at levels substantially lower than retrovirally infected control cells (MSCV-EN), presumably due to inherent differences in transfection efficiency between transient transfection and retroviral infection.  115  F I G U R E 22. EN-constructs for pronuclear injection are not able to form colonies in soft agar. Typical results of three separate infection experiments are shown for NIH3T3 cells transiently expressing empty vectors (pCX and pIRES2-EGFP), E N (pCX-EN and pIRES-EN), Harvey-Ras, and retrovirally infected positive control (MSCV-EN). Only positive control cells were able to form colonies, presumably due to higher levels of transgene expression associated with the retroviral expression system. Photomicrographs were taken using a lOx objective.  116 Generation of  ETV6-NTRK3  Transgenic Mice  To examine the direct consequences of E N expression in vivo, the cDNA of the human E N chimeric protein was cloned downstream of the C M V or beta-actin / C M V - I E promoters to direct expression to all tissues (see Figure 4, Chapter II) and injected into C57BL/6  x CBA  fertilized eggs (Table 5). Fifteen founders containing the E N transgene were identified by PCR/Southern analysis on genomic DNA, and each of the founders transmitted the E N transgene to the progeny, except founder 2027-3 which died before any mating could be performed. To generate larger cohorts of E N transgenic mice for analyses, each of the founder mice were mated to C57BL/6 mice and a strain established.  T A B L E 5 . Summary of fertilized eggs injected and subsequent generation of founder strains. pCAGGS is a combined chicken P-actin promoter with C M V enhancer (250). r Promoter  Construct  # Fertilized Eggs Injected  Pups Born Alive (Dead)  Number of Founders (Dead)  pCX-EN  pCAGGS  303  34(1)  9  pIRES2-EGFPEN  CMV  303  39 (3)  5(1)  ETV6-NTRK3  Transgene is Expressed in Multiple Tissues  To examine in more detail the expression pattern of E N transgenic mice, both a message and protein screen on various tissues isolated from five to six mice from each founder strain was performed. Total mRNA and protein lysate were prepared from freshly isolated brain, heart, lungs, thymus, liver, stomach, spleen, intestine, pancreas, kidney, bone marrow, skin, and muscle from mice aged three to six months. Of the 14 founder strains, six strains showed variable E N  117  message expression by RT-PCR. Message expression was identified in a number of different tissues: spleen, thymus, intestine, heart, kidney, pancreas, and bone marrow (results of a typical experiment shown in Figure 23). Distribution of transcription varied from line to line, with one mouse exhibiting E N transcription in five organs (line # 2029-4) (heart, lung, kidney, pancreas, and bone marrow). We were not able to document corresponding protein expression in these tissues, as evaluated by western analysis using ETV6 and NTRK3 antibodies (data not shown), presumably due to the low levels of E N protein in these tissues. Based on the extent of E N transcription, and the associated phenotype observed within the strains under study (see Table 6), we limited subsequent studies to two strains: 2015-7 and 2029-4. Both of these strains were derived from fertilized eggs injected with the pfRES2-EN construct.  F I G U R E 23. E N Transcription is detected in various tissues from E N transgenic mice, as shown by R T - P C R . R N A was extracted from various tissues and treated with DNase I to remove contaminating genomic DNA. cDNA was synthesized and amplified using primer sets flanking the fusion breakpoint. PCR products were electrophoresed on a 1.5% agarose gel, transferred to Nylon membrane, and labeled with an internal oligo to the PCR product to confirm (see above). A) Shows RT-PCR results for one mouse from 2029-4 strain; B) RT-PCR results from one mouse from 2015-7 strain. Note the tissue variation of transcription from different strains. Further, there was also transcriptional variation between mice from the same strain (i.e., not all mice from 2029-4 strain displayed transcription in all five tissues (see Table 6). A clinical case of congenital fibrosarcoma (CFS) was used as a positive control.  118  T A B L E 6. Site of transcription in 6 / 14 founder strains, indicating any detectable pathology. Based on widest range of detected expression and prevalence of pathology, we focused on two strains for the remainder of this study: 2015-7 and 2029-4.  MOUSE STRAIN  TISSUES EXPRESSING (RT-PCR)  PATHOLOGY  2010-1  spleen, liver  2010-5 2012-5 2015-7  thymus, heart intestine bone marrow, intestine  2029-4  heart, kidney, lung, pancreas, muscle liver  lymphoid hyperplasia (spleen & Peyers' patches) Not detected Not detected atypical lymphoid infiltrates; leukemia / lymphoma lymphoblastic lymphoma, fibrosarcoma Not detected  2029-5  Incidence of Spontaneous Neoplasms in EN Transgenic Mice To determine whether E N transgene expression could lead to an increase in tumour development, mice of the two transgenic strains under study (as well as littermate controls) were kept under observations for two years for clinical symptoms of disease (ruffled fur, failure to nest, listlessness, hunched back, loss of appetite, or gross evidence of tumour). Numerous mice died in both the 2015-7 and 2029-4 strains, most of them between ten and eighteen months, either due to tumour burden, chronic skin infection endemic to the animal facility or undetermined causes. Tumours were observed in approximately 30% of the sick and dead transgenic mice. Symptomatic mice were killed; spleen, thymus, liver, mesenteric lymph node, and bone marrow were examined. Splenomegaly was always apparent, accompanied by abdominal lymphadenopathy and occasional leukemic infiltrates of the liver and lung. Statistical significance of tumour formation was determined using Fisher's exact test, as it is more suitable for small sample sizes that the Chi-squared test. To date, the incidence of  119  tumours in the 2015-7 strain is 32% in the transgenics versus 0% in the non-transgenics (p=0.005, see Table 7). This data has 100% sensitivity (i.e., proportion of mice that are transgenic and develop tumours, 10 / 10) and a specificity of 56% (e.g., proportion of mice without tumours that are not transgenic, 27 / 48). Due to the 32% incidence of disease in the transgenic group, the E N transgene has more strength as a negative predictor as a positive predictor. Said another way, the proportion of mice without the E N transgene who don't have tumours is 100%, but the proportion of mice with E N who actually have tumours is 32%. The 2029-4 strain has an incidence of tumours of 29% in the transgenics versus 3% in the non-transgenics (p<0.005, see Table 8). This data has 91% sensitivity (i.e., proportion of mice that are transgenic and develop tumours, 10 / 11) and a specificity of 57% (e.g., proportion of mice without tumours that are not transgenic, 33 / 58). Consistent with results found with the 2015-7 strain, due to the 29% incidence of disease in the transgenic group, the E N transgene has more strength as a negative predictor than as a positive predictor. That is, the proportion of mice without the E N transgene who don't have tumours is 97%, but the proportion of mice with E N who actually have tumours is 29%. Another 15% of mice die annually of causes unrelated to neoplasia (skin infection, malocclusion, anal prolapse, etc.), regardless of genotype.  120  T A B L E 7. Incidence of tumours in the 2015-7 strain by 19 months of age: 32% in the transgenics versus 0% in the non-transgenics (p=0.005).  Tumour  No Tumour  Total  Transgenic  10  21  31  Non-Transgenic  0  27  27  T A B L E 8. Incidence of tumours in the 2029-4 strain by 19 months of age: 29% in the transgenics versus 3% in the non-transgenics (p<0.005).  Tumour  No Tumour  Total  Transgenic  10  25  35  Non-Transgenic  1  33  34  Abdominal and Splenic Lymphomas Both 2015-7 and 2029-4 strains develop aggressive lymphomas after a considerable latency period, compared with an absence of tumours in the non-transgenic littermate control group. These lesions usually presented as a massive swelling in the abdomen. Upon necropsy, large masses in the intestinal lymph nodes or the spleen were noted, with involvement of the liver, abdominal lymph nodes, lungs, and bone marrow in advanced stages. Lymphoid organs of littermate control mice stained with Hematoxylin and Eosin were normal. Conversely,  121  approximately 30% transgenic mice had either a large-cell or lymphoblastic lymphoma (Table 9 and Table 10). The normal splenic architecture was replenished (in some cases completely) with lymphomatous infiltrates. RT-PCR analysis showed E N transcription in hematopoietic tissues and organs infiltrated with lymphoma cells. Histological analysis typically revealed aggressive, cellular proliferations with a high mitotic rate (Figure 24). In occasional tumours, detection of E N protein was also possible, at levels comparable to clinical samples rather than cell culture controls (see Figure 24b).  T A B L E 9. Distribution of tumours in the 2015-7 strain. ND=Not Detected by IP / Western analysis.  MOUSE  TRANSGENIC  RT-PCR  PROTEIN  TUMOUR SITE  PATHOLOGY  2015-7-3-7  Y  Y  ND  Blood, Marrow, spleen abdominal  Leukemia / Lymphoma  2015-7-4-1  Y  Y  Y  2015-7-4-3  Y  Y  ND  2015-7-4-4  Y  Y  ND  2015-7-7-5  Y  Y  ND  2015-7-8-5  Y  Y  ND  2015-7-8-1-1-2  Y  2015-7-8-1-1-4  Y  2015-7-7-8  Y  2015-7-7-7  Y  Not performed Not performed Not performed Not performed  Not performed Not performed Not performed Not performed  D A T E OF TUMOUR INCIDENCE 19 months  Large Cell Lymphoma Lymphoma  19 months  Lymphoma  17 months  Lymphoma  14 months  abdominal  Lymphoma  19 months  abdominal  Lymphoma  18 months  liver  Lymphoma  15 months  abdominal  Lymphoma  14 months  Intestine, liver, spleen  Lymphoma  14 months  Intestine, spleen Abdominal, liver Intestine, spleen  15 months  T A B L E 10. Distribution of tumours in the 2029-4 strain. ND=Not Detected by IP / Western analysis.  MOUSE  TRANSGENIC  RT-PCR  PROTEIN  TUMOUR SITE  PATHOLOGY  2029-4-4-8  Y  Y  Y  Shoulder  fibrosarcoma  D A T E OF TUMOUR INCIDENCE 16 months  2029-4-4-9  Y  Y  ND  Abdominal  Lymphoma  19 months  2029-4-2-6  Y  Y  ND  Lymphoma  19 months  2029-4-3-4  Y  Y  ND  Lymphoma  19 months  2029-4-4-9-6-3  Y  Y  ND  Abdominal, thymus Intestine, spleen Spleen  Lymphoma  19 months  2029-4-2-12  Y  Y  ND  Lymphoma  13 months  2029-4-4-9-3-4  Y  18 months  Y  abdominal  Lymphoma  19 months  2029-4-4-9-1-8  Y  abdominal  Lymphoma  19 months  2029-4-4-9-4-6  Y  abdominal  Lymphoma  17 months  2029-4-3-3  N  Not performed Not performed Not performed Not performed ND  Lymphoma  2029-4-4-9-1-4  Not performed Not performed Not performed Not performed N  Intestine, spleen Abdominal  Trunk  Rhabdomyosarc oma  17 months  <9 f  # f ?  X  # <?  N  f  ^  IP: a-ETV6; WB: P-Tyr 1000 |Lig input  F I G U R E 24. Large Cell Lymphoma from 2015-7 mouse expressing E N protein. (A) H & E stained section (lOx objective) of spleen from E N transgenic showing extensive invasion and near complete replenishment of red pulp by lymphoma cells (top). Highpower magnification (40x objective) of lymphoma cells showing large lymphocytes with loose, open chromatin and prominent nucleoli (bottom). (B) Tumour cells shown in (a) along with unaffected kidney and several controls were lysed and immunoprecipitated with an a-ETV6 antibody, separated by SDS-PAGE and probed with an a-phosphotyrosine antibody. The E N doublet was detected only in tumour tissue from the transgenic mouse. For comparison of levels, compare to EN-positive clinical case of pediatric secretory breast carcinoma.  125  Preliminary  Immunophenotyping  Histological analysis of the lesions indicated that these masses were lymphomatosis; specifically large cell and lymphoblastic lymphomas. To characterize these tumours further, we analyzed tumour cells for expression of cell surface markers. Out of six mice analyzed, all showed positive staining for either B or T cell markers, as indicated by positivity for either B220, CD3, or Thy 1.2 staining (Figure 25). B220 (CD45R) is commonly used as a pan B-cell marker, but has also been found on activated subsets of natural killer (NK) cells (301, 302). No obvious differences can be seen between the 2015-7 and 2029-4 strains on this small data set. We cannot rule out that some T-cell staining may be due to inflammatory infiltrates within the tumours. Both types of lymphomas were negative for myeloid marker Mac-1 (data not shown). A single tumour showed evidence of biphenotypic staining, with 30% of the tumour population positive for both B220 and Thy 1.2. This suggests that E N transformation occurred at a very early stage of lymphoid development. The ability of a transgene to induce lymphomas of varying origins has been frequently reported, including a model of N P M - A L K (303, 304).  Fibrosarcoma A single mouse in the 2029-4 strain developed a large, white mass on the right shoulder, weighing 3.5g. There were no obvious metastases, nor any other abnormalities noted. Histological  analysis  revealed spindle cell proliferation with dense cellularity, nuclear  pleomorphism and moderate mitotic activity forming occasional herringbone patterns. Cells showed no obvious evidence of specific differentiation other than having the appearance of possible fibroblastic origin. The tumour was positive immunohistochemically for vimentin, but not for muscle specific actin, desmin, S100, histiocytic markers, or endothelial markers (data not  126  shown),  and was therefore diagnosed fibrosarcoma, virtually indistinguishable from the  congenita] fibrosarcoma from which E N was originally identified (Figure 26). RT-PCR analysis was positive for E N transcription, and E N protein was also detectable by IP / western analysis (Figure 27).  Serial  Transplantation  of ETV6-NTRK3-positive  tumour cells into secondary  recipients  To determine whether tumours generated in E N transgenic mice were transplantable, we employed the immunocompromised mouse injection assay (305, 306). Cryopreserved cells were thawed rapidly and rinsed several times to remove any cell debris. One million E/V-positive tumour cells derived from abdominal masses with lymphomatous features were injected both intravenously and intraperitoneally into two immunocompromised mice. Mice were monitored daily for clinical symptoms of disease, and sacrificed at the first signs of morbidity (17-20 days post-injection). Upon necropsy, animals were found with large masses in the thoracic cavity and lungs, with smaller lesions visible in the peritoneum. These lymphomas observed in the secondary recipient animals were histologically and immunophenotypically identical to the primary tumours (data not shown). Further, transcription of the EN transgene was documented by RT-PCR in all recipient animals (data not shown).  Analysis  of Potentially  Cooperating  Events in EN-Induced  Tumourigenesis  To create an environment that may be more permissive to E N expression and subsequent transformation, E N heterozygous mice were crossed into either Rag (Recombination Activating Gene) -deficient  (immuno-compromised) or Pten  +A  (Phosphatase  and Tensin Homolog)  backgrounds. Rag-deficient null (RD" ) mice are totally deficient in both mature T cells and B /_  F I G U R E 25. Lymphomas in E N transgenic mice are of T and B-cell origin.. Tumour cells confirmed for message expression of EN were disaggregated into a single suspension and stained for surface expression of B (B220) and T ( C D 3 , Thy 1.2) cell markers. Histological analysis showed original masses to contain little normal tissue. (A) Tumour population contains both B and T cells. (B) Tumour cells show evidence of biphenotypic pattern.  128  cells (255, 307); this lack of an immune system may allow E N expression to cause tumours more quickly (308). Crosses with RD " and E N transgenic mice did not result in an 7  acceleration of disease, or a statistically significant number of tumours versus non-transgenic controls; however, the data does suggest a higher incidence of tumours in the transgenic group. For the 2015-7 strain, 14% (3/22 mice) of transgenic mice bred to RD" mice developed a A  tumour phenotype, compared with 0% in the non-transgenic group (0/24 mice) (p = 0.10). While not statistically significant, this data does have 100% sensitivity (i.e., all of the mice that have tumours are transgenic). A determination of sample size required for significance indicates that either of the following would provide statistical significance (p>0.05): 1) analyze one more EN " +/  with the anticipation it contains a tumour (13.6% likelihood, given the results to date); 2) analyze three more negative (EN "") animals and anticipate they do not have a tumour (100% likelihood, /  given the results to date). For the 2029-4 strain, 14% (3/21 mice) of transgenic mice bred to RD " 7  mice developed a tumour phenotype, compared with 0% in the non-transgenic group (0/19 mice) (p = 0.10). Sample size determination for this crossbreeding indicates that either seven nontransgenics at 19 months of age must be tumour-free, or two more E N " would need to develop +/  tumours in order to attain statistical significance (i.e., p<0.05). The  Sorensen laboratory have previously shown that E N expressing cells show  constitutive activation of Akt, and that pharmacologic blockade of this pathway abrogates transformation (210). We therefore hypothesized that constitutive activation of the PI3K-Akt pathway may allow for increased EN-induced transformation. P T E N (also known as M M A C 1 , mutated in multiple advanced cancers) is a protein tyrosine phosphatase that was originally identified as a tumour suppressor in multiple organ systems (256, 309-311). P T E N is essential  129  for embryonic development and PTEN " mice are embryonic lethal (312) (256). Heterozygous 7  mice, PTEN ", are characterized by tumours of the endometrium, liver, prostate, gastrointestinal +/  tract (gut-associated lymphoid tissue), thyroid, and thymus (256). Further, cells from Pten " mice +/  have elevated levels of phosphorylated Akt (257, 313). As elevated levels of phosphorylated Akt is a biochemical hallmark of E N transformation, E N transgenic mice from both 2015-7 and 2029-4 strains were crossed with Pten " mice and monitored for any increases in morbidity or +/  mortality. For undetermined reasons, these crosses were very difficult to obtain. Four different mating pairs were used over a 14 month period, and only eight litters were produced (each parent was a confirmed successful breeder when matched with a B16 control). All litters were small in number (less than five pups each). From a total of 26 pups born, there was only one confirmed double positive (EN ~, PTEN ~). Based on Mendelian genetics, we would have expected at +/  +/  least six double positive mice to have been born. Chi-squared analysis indicates these findings to be significant (p<0.05).  Development of ETV6-NTRK3  Knock-in Mouse  As the murine syntenic region of human 12pl3 (Chromosome 6q) had not been sequenced and submitted to a public database, I therefore sequenced more than 20kb of murine Etv6 and began constructing the targeting vector to create a knock-in mouse. While working on the final cloning step, we were contacted by Stuart Orkin, Harvard Medical School, regarding the possibility of collaborating. Dr. Orkin's had successfully created an E T V 6 - A M L 1 knock-in mouse, and were interested in using this system for an E N knock-in mouse; I provided the  -Y  •  m  Mm  *  ****  i  <-—-  FIGURE 26. Histology of fibrosarcoma from E N transgenic Mouse is identical to clinical CFS. H & E staining of a clinical case  of CFS (A, C) and the fibrosarcoma that developed in a transgenic mouse (B, D) showing identical histological features. Low power view (lOx objective, top) of cellular spindle cell lesions arranged in interdigitating fascicles in both tumours. High power magnification (40x objective, bottom) shows undifferentiating spindle cells with overlapping nuclei and occasional mitotic figures (circled), characteristic of this tumour. i—i  O  131  f 300—  mm*  200 —  mmm  V?  ^ °  #  mm  100 —  ^  ^  WB: RC20 WB: a-ETV6 IP: a -NTRK3  F I G U R E 29. E N expression is detectable at both the R N A and protein level in a single fibrosarcoma. Top panel shows RT-PCR of ETV6-NTRK3 fusion transcripts from total R N A treated with DNase I to remove contaminating cDNA. Amplification of the 110 base pair product using the Tel 971 and Trk 1059 primers. Bottom panel shows the presence of the E N protein after immunoprecipitation of a whole cell lysate prepared from the tumour tissue.  132  NTRK3 cDNA to modify their existing construct. The following experiments were performed on embryonic cells created by Dr. Zhe Li, post-doctoral fellow under Dr. Stuart Orkin.  Expression of Conditional ETV6-NTRK3 after Retroviral Cre Recombinase  Administration  Multiple independent-targeted ES clones carrying the NTRK3 knock-in cassette (NTRK3 cDNA, polyadenylation signal and neomycin resistance gene) were obtained following drug selection, PCR and Southern blot screening process. To confirm conditional expression of the knocked-in allele, ES cells were treated with adenoviral Cre and analyzed by RT-PCR. E N message expression was documented in two independent ES cell lines, TN28.12 and TN28.19, as well to a much lower degree in the targeted ES cells without Cre, indicating some 'leakiness' in the conditional system. Western analysis on these clones documented the presence of the 73/68 kDa E N protein in both TN28.12 and TN28.19, and not in the targeted ES cells without Cre (TN28) (see Figure 28).  Cre:  +  HlCS  QC  SC  QC  Z H  Z H  i-H  rl  z  u Z H  o u a  >  H CO  W B : P-Tyr IP: a-Tel F I G U R E 28. Expression of E T V 6 - N T R K 3 in targeted murine E S cells. E N protein is expressed only after infection of conditional MEFs with retroviral-Cre in two independent clones (TN28.12 and TN28.19). Conversely, targeted ES cells without Cre administration do not express the fusion (TN28). Retrovirally infected NIH3T3s were used as controls.  133  Analysis of Transformation Effector Pathway Signaling in Cre-Infected ES Cells The activation of the well-characterized E N effector pathways, M A P K and PI3K, was assessed by preparing whole cell lysates from serum-starved ES cells, and immunoblotting with phosphorylation state-specific anti-MEK and anti-Akt, and anti-cyclinDl/2 antibodies. ES cells expressing E N (TN28.12 and TN28.19) unexpectedly contained equal or even decreased levels of phosphorylated Akt, phosphorylated M E K , and cyclin D compared to non-expressing ES cells (TN28) (Figure 29) or wild-type ES cells (data not shown). This data is very similar to the results of Tuveson et al. (314), where mutant Ras expressing MEFs induced equal or lower levels of E R K 1/2 and Akt activation than wild-type MEFs. In that study, the Ras-induced signaling cascade (and subsequent morphologic changes) could only be confirmed through use of pharmacological inhibitors of the M A P K and PI3K pathways (314).  Current Status of EN knock-in Mouse  Collaboration  Dr. Orkin's lab has successfully targeted ES cells and has obtained heterozygous mice that are being mated with GatalCre mice (expression in germline and hematopoietic  cells).  Heterozygous E N animals (without excision) are born at normal ratio and look normal, but most of the GatalCre; E N animals seem to die during gestation. To overcome this embryonic lethality, E N heterozygotes are being crossed to a number of different conditional Cre-expressing mice. Further, microarray analysis of EN-expressing and wild-type mouse embryonic fibroblasts (MEFs) is also in progress.  134  Cre:  -  + iH  cc  CC  Z  z  H  P-Akt  + cc CC <N  Z H  c  z H  H  J - 52 kDa T - 9 6 kDa  P-MEK  Cyclin Dl/2 Wt ETV6  "52 kDa  36 kDa 56 kDa  F I G U R E 29. E N expression in ES cells does not induce activation of A K T , M E K , or cyclin  DI. Targeted ES cells as well as retrovirally infected NIH 3T3 controls were serum starved for 24 hours, rinsed thoroughly, lysed and assessed for hallmarks of EN transformation. Creadministered (EN-expressing) ES cells TN28.12 and TN28.19 showed no evidence of activation of AKT, MEK, and cyclin Dl/2. Probing for wild-type ETV6 confirms equal loading in all samples.  135  5.3  DISCUSSION Animal models of human disease allow the investigator to analyze the genetic factors that  influence disease phenotype, as well as design rational intervention strategies. The utility of the laboratory mouse as a model organism has expanded immensely during the past 20 years (315). Moreover, transgenic and knockout technologies are continuously evolving; new mouse models of cancer will undoubtedly continue to benefit our understanding of tumourigenesis. In an attempt to model the tumourigenic process invoked by the ETV6-NTRK3 oncoprotein, transgenic mice overexpressing the E N fusion gene were generated. As we wished to express E N in as many tissues as possible, we selected two ubiquitously-expressing promoters to drive transgene expression. E N transcription was documented in various cell types. Interestingly, transgene expression only led to tumour development when expressed in lymphocytes and fibroblasts,  suggesting  the  cellular  environment  is  crucial  in  mediating  EN-induced  tumourigenesis. Two strains of E N transgenic mice were selected for detailed characterization, based on their variability of tissue expression and preliminary phenotype. Both of these strains displayed remarkable consistency in phenotype, with a considerable latency period before disease induction. This is presumably because non-transformed / normal cells will not tolerate excessively high levels of E N protein expression, and expression above a certain threshold will likely induce apoptosis. A recent transgenic model of N P M - A L K fusion demonstrated a prolonged onset of disease concomitant with a lower-expressing founder line (304). Turner et al. proposed that founders with high expression levels die in utero or very soon after birth (304). Of course, there may be alternate explanations for the relative low levels of E N expression and long latency period before disease induction. The C M V promoter / enhancer that is an  136  essential component of both constructs used for transgenesis has been shown to be frequently methylated, most probably as a mechanism to down-regulate expression of a viral genome (316). Studies have demonstrated the in vivo methylation (and subsequent decrease in transcription) of CMV-driven transgenes (317, 318). Additionally, fusion protein expression (an abnormal protein for that cellular environment) is being driven by an artificial promoter, which may have a multitude of unknown effects. This may include induction of apoptosis, as suggested by work with other fusion proteins such as EWS-Flil (S. Baker, St. Jude Children's Research Hospital, Memphis, T N , personal communication) or others (PHB Sorensen, personal communication). The NTRK oncogene was originally isolated from colon carcinoma as a fusion with the tropomyosin gene (TPM3), resulting in constitutive activation in the tyrosine kinase activity (319). TPM3-NTRK1 fusions were subsequently detected in papillary thyroid carcinomas (171), and altered N T R K signaling has been implicated in other neoplasms (reviewed in (172)) including pancreatic adenocarcinoma (173), and an activating NTRK1 mutation in a case of A M L (174). Thus, oncogenic NTRKs appear to be exclusively observed in human malignancies with nonneuronal origin. In this animal model, overexpression of the kinase domain of NTRK3 in lymphoblasts (non-neuronal) has resulted in lymphomas of B- and T-cell origin. To our knowledge, there are presently no animal models of wild-type NTRK3 overexpression. The development of lymphomas in E N transgenic mice was surprising, although not unexplainable. A recent study has suggested that transformation by an endogenous  K-Ras  oncogene is highly dependent on cellular context. Guerra et al. demonstrated in a knock-in mouse model that although oncogenic Ras (K-Ras expression  ) was expressed in a wide range of cell types, only  in the lungs resulted in tumour formation (multifocal  lung adenomas and  adenocarcinomas) (320). It may be that, for undetermined reasons, a precursor lymphoid cell  137  contains the correct cellular milieu to allow for E N transformation. The lengthy period of time for disease induction is reminiscent of the EN fusion in a single case of the acute myeloid leukemia (196). Moreover, the biology of "liquid" tumours (leukemias and lymphomas) is substantially different that that of solid tumours (321, 322), and there may be substantial differences in the pathways induced or inactivated. While E N is capable of transforming multiple cell lineages, the specific mechanisms may vary between cell types. This may explain the paradox between the long latency period observed in these mice, and the short (or absent) latency period of the congenital fibroblastic tumours. We had initially hypothesized that additional cooperating mutations may be required for EN-induced tumourigenesis.  For instance, it has been demonstrated that ETV6-NTRK3  transformation requires an intact IGF1R (215). We therefore crossed both strains of mice to both immunocompromised (Rag-deficient) mice and mice with constitutively activated Akt (PTEN heterozygotes). Crosses with the Rag-deficient strain did not result in an increase.in disease incidence or decrease the latency period significantly. The long latency period seen in this study may suggest that other cooperating mutations may be required for maintenance or even initiation of  EN-induced  tumourigenesis,  as  suggested by  other  studies  with  single  oncogene  overexpression (323-325). However, the majority of clinical cases of t(12;15) resulting in the E N fusion are congenital or occurring in early childhood, suggesting against the accumulation of additional 'hits.' This data suggest that levels of E N may be more crucial to its regulation and expression than cooperating mutations. Mating experiments between E N  + A  and Pten " appear to suggest that double heterozygotes +/  are lethal, given that 1/26 mice were of the desired genotype (Mendelian ratio expects 6.5 pups to be both E N " and PTEN ". It should be noted that these crosses in general did not yield sizable +/  +/  138  litters, and other unknown factors may be involved or responsible for this. It is known that expression of E N results in constitutive phosphorylation of Akt; Pten " mice also show hyper+/  activation of Akt. It may be possible that too much Akt could be toxic to the cell and induce cell death. However, there is no published literature to support this hypothesis. Another possibility is that embryonic cells in the developing mouse may not tolerate high levels of reactive oxygen species (ROS). NIH3T3 cells transfected with E N have been shown to contain higher levels of ROS than inactive kinase control (Tognon and Sorensen, unpublished results), and it is well documented that fibroblasts are particularly adept at sustaining high levels of ROS without inducing apoptosis (326-328). Embryonic cells in the developing mouse may not be able to tolerate such high levels of ROS and, consequently, induce apoptosis. Finally, while the lipid phosphatase activity of P T E N has garnered much attention due to its association with tumour suppressor function, little is known about the protein phosphatase activities of PTEN. It is possible that in the context of oncogenic stress, P T E N is dephosphorylating (and down-regulating) a key component of the E N transformation cascade. In MCF7 cells, P T E N expression inhibited M A P K phosphorylation and Akt activation by dephosphorylating IRS-1 (279). It may be, therefore, that decreased P T E N levels (as would be seen in heterozygous mice) already contain some activation of Ras-MAPK proliferation cascade which, when crossed to an EN-expressing background, results in cell cycle arrest due to excessive activation of Ras-MAPK (discussed in detail elsewhere in this thesis). The E N fusion oncogene is unique in that it has been identified in fibroblastic (2), hematopoietic (196), and epithelial malignancies (131), and in vitro data has shown this fusion is capable of transforming cells of these lineages (129, 131) (Matthew Martin, unpublished results). Further, using a retroviral bone marrow transplantation assay, Liu et al. demonstrated that a  139  modified E N transcript (containing 14 amino acid kinase insert) causes a long-latency, pre-B-cell lymphoblastic lymphoma (130). It may not, therefore, be surprising that overexpression of E N in a transgenic mouse model induces lymphomas with a considerable latency period. I  Loss of heterozygosity at chromosome 12pl3 (ETV6) is found in many types of malignancies (reviewed in (140)). It has also been reported that several translocations involving 12pl3 also have the second ETV6 allele deleted (139). ETV6 has also been shown to inhibit the invasiveness of Ras-transformed cells in vitro and in vivo (140, 329), suggesting that ETV6 may function as a tumour suppressor. Ablation of a single ETV6 allele with this knock-in approach may be more representative of the biology of E N expressing tumours, through disruption of this tumour suppressor. Such a model has considerable advantages over other methods for recreating chromosomal translocations (such as inducible expression, Cre-lox systems, 'hit-and-run' strategies, multiple genetic manipulations or combinations thereof (294, 315, 330-332). We are currently awaiting the creation of these mice by our collaborators. A potential pitfall of a knock-in strategy is that it does not include the balanced translocation and does not imitate the cell-type-specific, somatic aspects of a tumourigenic translocation. The possibility that both expression products of a translocation can alter the disease phenotype is well illustrated by work on mouse models of the P M L - R A R a (promyelocytic leukemia-retinoic acid receptora) leukemogenic translocation (333). However, a large amount of in vitro and in vivo evidence suggests that the reciprocal translocation product is not required for EN-induced tumourigenesis (129, 210). We have demonstrated that our collaborators have correctly targeted the Etv6 locus with the knock-in (NTRK3) cDNA. Preliminary studies into the transforming ability of these E N expressing ES cells have not been able to demonstrate M A P K activation; in fact, there may be  140  some attenuation of M A P K activity in these cells relative to non-targeted cells. In contrast to the ease of identifying such signaling events following E N expression in NIH3T3 (210, 283), it is difficult to see evidence of Ras effector pathway activation in Cre-administered ES cells. As suggested by Tuveson et al. for studies with knocked-in activated K-Ras, E N effector pathway signaling that accompanies endogenous expression levels of E N may not be reliably assessed by isolated analysis of phospho-Mek or phospho-Akt (314). Further to this point, we have not been able to demonstrate elevation of phospho-Erk in NIH3T3s overexpressing E N , despite elevated levels of phospho-MEK (210, 283). Alternately, there may be inherent differences between analysis of proliferation in ES cells and NIH3T3s such that the proliferative advantage provided by an oncogene may be insignificant to the levels of endogenous levels of proliferation in these embryonic cells. Studies assessing the phosphorylation status of IRS-1, as well as IGF1R status, should be performed to determine the role of this pathway in ES cells expressing E N . In summary, we have shown that ubiquitous E N expression in a murine model results in B and T cell lymphomas. Thus, the E N fusion is unique in that it has potent transforming abilities in four cell lineages (fibroblastic, epithelial, myeloid, and lymphoid cells). However, the precise mechanism in which E N operates in these different tumour types is not clear, and remains to be determined.  141  CHAPTER VI SUMMARY AND FUTURE DIRECTIONS  6.1  GENERAL SUMMARY The EN fusion gene is unique in that it has been identified in tumours derived from  varying cell lineages. E N expression provides an interesting model of tumourigenesis in that it has transforming properties in multiple cell lineages. Originally identified in CFS, and subsequently in C M N , it has since been identified in an adult patient with A M L , and pediatric and adult cases of a ductal breast cancer called SBC. E N is a potent oncoprotein in vitro (129), and previous studies have demonstrated that it constitutively activates both the Ras-MAPK and PBK-Akt pathways (210) via an interaction with IRS-1 (215). However, the exact nature of the E N / IRS-1 interaction  remained unknown. We  now  demonstrate  that  E N specifically  binds  the  phosphotyrosine binding (PTB) domain of IRS-1 via an interaction at the C-terminus of E N . An E N mutant lacking the C-terminal 19 amino acids does not bind IRS-1 and lacks transforming ability. These findings indicate that E N / IRS-1 complex formation through the NTRK3 Cterminus is essential for E N transformation. It was hypothesized that a highly conserved motif (TPIY) in the C-terminal tail of E N may function as a pseudo-consensus site for PTB binding (NPXY); however, this did not prove to be true. To further characterize the transforming ability of E N , we developed transgenic mice expressing E N under the direction of two ubiquitously expressing promoters. Approximately 30% of transgenic mice in two different strains developed lymphomas after a long latency period. Further, a single transgenic mouse developed a fibrosarcoma with identical histology to CFS. In the following two sections, major observations  142  observed in this thesis will be discussed in the context of potential avenues for further experimentation.  6.2  E N SIGNAL TRANSDUCTION These studies of the signaling mechanisms behind EN-induced transformation have shown  that IRS-1 interacts specifically with the C-terminus of E N . Various mutants lacking IRS-1 binding were non-transforming. It would therefore appear that IRS-1 is functioning as an adapter protein that connects E N to the Ras-MAPK and PI3K pathways. The involvement of IRS-1 as an adapter for other proteins has already been proven (17, 271, 334). It is well established that N T R K molecules bind the adapters SHC and the p85 subunit of PI3K via their SH2 domain, allowing for the activation of downstream effectors (165, 335). Studies in the Sorensen laboratory are currently assessing whether the interaction with IRS-1 is a novel mechanism for wild-type N T R K molecules to activate these pathways. Preliminary analysis has suggested this interaction with IRS-1 is unique to E N (Martin and Sorensen, unpublished results), presumably because NTRK3 molecules have binding sites for the adapters SHC and GRB2, which may circumvent the need for this connection found in an abnormal fusion protein. We had hypothesized that the TPIY sequences (residues 625-628) found in the distal part of the C-terminus of E N was a novel PTB interaction motif, due to its similarity to the canonical N P X Y sequence; however, this did not turn out to be true. Further, IRS-1 binding must occur through an interaction N-terminal to the A614 truncation. We hypothesize that the deletion mutant is folding back upon itself in such a manner that it is either blocking the specific binding site or affecting the three-dimensional structure such that an improper conformation is attained.  143  Comparative modeling is useful for relating patterns observed among a small number of structures to the generalized functions of a large protein family (for reviews, see (336, 337)). Recent improvements in comparative modeling make it practical for homology-based assignment of biological function to unknown proteins and structure-based design of novel pharmaceuticals (336 6798)). A number of different structure-prediction servers are available for this purpose. SWISS-MODEL (242, 243) is a fully automated protein structure homology-modeling server, accessible via the ExPASy web server, or from the program Deep View (Swiss Pdb-Viewer), and was utilized to create a model of the NTRK3 portion of E N . The amino acid sequence of the kinase portion of E N was submitted to the SWISSM O D E L Protein Modeling Server (338) with fixed parameter settings given by the Server to screen the structure templates further. The Server gave predictions of 3D structures for five proteins: four of which were the structures of the activated and inactivated catalytic domain of the insulin receptor kinase, and one of which was the structure of the activated M U S K tyrosine kinase (Table 11).  T A B L E 11. Protein crystal structures used as templates for homology modeling. PDB = Protein Data Bank (http://www.expasy.org/swissmod/SWISS-MODEL.html). APS = adapter protein with pleckstrin homology (PH) and src homology 2 (SH2) domains.  PDB C O D E  PROTEIN  llufA  Musk Tyrosine Kinase Insulin Receptor Kinase in Complex with the Sh2 Domain of APS Insulin Receptor Kinase in Complex with an inhibitor Phosphorylated Insulin Receptor Tyrosine Kinase Tyrosine Kinase Domain of the Human Insulin Receptor  irqq IgagA lir3A lirk  HOMOLOGY TO NTRK3 KINASE 50.77% 39.76%  REFERENCE  39.16%  (245)  39.16%  (244)  39.16%  (246)  (247) (248)  144  As seen with other kinases (such as insulin receptor and Src), when E N is inactive the activation loop threads through the active site, preventing binding of A T P and possibly protein substrates (see Figure 30). The inactive form of the E N catalytic domain is modeled after four different structures, and consequently the structure is most likely valid. When phosphorylated, the activation loop 'flips' outwards and the active site is accessible. However, insight into possible mechanisms of interaction and binding partners of this active kinase is highly speculative, as only a single kinase was used for modeling of the active molecule. This model shows the C-terminus and tyrosine 615 as part of a helix and consequently, is probably not a substrate-binding site. The tyrosine side chains of amino acid 615 do project from this helix, and it may be possible that this is available for interaction, but this is purely speculative. Higher resolution crystal structures are required in order to make such inferences. Sequence homology searching using the SWISS-MODEL  similarity tool  (BLASTP)  indicated the highest level of homology of N T R K portion of E N to the M U S K tyrosine receptor kinase, a transmembrane protein expressed exclusively in skeletal muscle (339). R T K catalytic activity can be regulated via several autoregulatory mechanisms, including autophosphorylation of the kinase activation loop (including A T P and substrate binding) and autophosphorylation of key tyrosine residues in the juxtamembrane region. M U S K kinase activity appears to be regulated through both of these mechanisms (247). Due to the position of the E N fusion breakpoint, the regulatory juxtamembrane tyrosine is missing (SHC binding site in wild-type NTRK3, tyrosine 485), suggesting that E N only utilizes the activation loop mode of inhibition. Perhaps the loss of the juxtamembrane inhibition mechanism in the E N fusion is crucial to its ability to be involved in many malignancies.  F I G U R E 30. Homology modeling of the N T R K 3 portion of E N . A ribbon diagram of the crystal structure of the NTRK3 portion of E N , with P-strands shown in cyan and a-helices shown in red. The inactive protein is modeled on the left; active protein (showing an accessible active site) on the right. As the most distal portion of the C-terminus is unique to NTRK3, no homologous sequence existed to serve as a template. Consequently, the most distal ten amino acids of E N are missing from these models. Accordingly, caution should be used when interpreting the structure of this area in particular. Tyrosine 615 (and side chain) is indicated in green at the C terminus of the protein.  146  The structure model presented here for the catalytic domain of E N may be categorized as moderately accurate as the model is based on a sequence identity of 30-50%, suitable for virtual screening and docking of small ligands and defining antibody epitopes, but not prediction of protein partners or docking of macromolecules (340, 341). Models of this quality (as defined by sequence homology  between target  and template  sequences)  allow  for locating and  characterizing active sites. Therefore, the model presented here can serve as a rough guide for the allocation of amino acid residues of importance for further investigations or for the further refinement of the models of the E N catalytic domain. These studies (SPOT peptide array and NMR structure / X-ray crystallization of the kinase portion of EN) are currently in progress. To this end, studies into the three-dimensional structure of E N are currently in progress. Nuclear magnetic resonance (NMR) spectroscopy and subsequently X-ray crystallography will be performed on the kinase domain of E N . Ideally, these experiments will also be performed in conjunction with the PTB domain of IRS-1 and the SH2 domain of P L C y l . Such a strategy will determine the precise relationship between these proteins. While previous research in the Sorensen laboratory has demonstrated that the PLCyl-binding site (Y628) is dispensable for transformation (129), it is possible that signaling through this residue is responsible for the welldifferentiated morphology observed in these tumours. Both the SHC and PLC-y binding sites on NTRK1 are required for maximal activation of M A P K (342). It may be that this site contributes to, but is not essential for, the transformation signal cascade induced by E N . It would therefore be plausible that E N binds two effectors at the C-terminus: PLC-y to induce differentiation and IRS-1 to induce proliferation / survival signaling. The three-dimensional structural approach outlined above assumes that the interaction between E N and IRS-1 is direct. While this is a likely hypothesis, it is quite possible that there is  147  an intermediary step (s) involved. This may explain the relatively low levels of IRS-1 protein found in E N immunoprecipitates (and vice versa). The most efficient method to determine whether this interaction is direct, is an in vitro translation experiment where the IRS-1 protein is translated (from a T7 promoter-containing vector) and mixed with GST-fused E N (or E N mutant). A pull-down for the GST-tagged proteins with glutathione beads, and western analysis for the presence of IRS-1 would determine if this is a direct interaction. Alternately, due to potential problems with translating a large protein such as IRS-1, the reverse experiment could be performed or it may be possible to use only the PTB fragment of IRS-1 (TRS-1C) outlined in Chapter III. Additionally, peptide (SPOT) array analysis of the tyrosine 615 is currently in progress. This technique assesses substrate binding to a series of ordered amino acids adhered to a cellulose membrane, and thereby identifies the peptide or consensus motif required for binding (343, 344). Manipulation of amino acid residues both N - and C-terminal to this tyrosine may provide insight into other residues involved in IRS-1 binding beside Y615. To date, we have only identified two proteins that interact with EN: P L C y l and IRS-1. It is conceivable that many other interactors may exist, and analysis of these interactors as well as their downstream binding partners may provide clues to presently unknown facets of E N tumourigenesis. One approach is mass spectrometric analysis of differential protein bands in silver-stained gels. Preliminary experiments by this author have identified a number of targets for further analysis. Immunoprecipitates from non-transforming EN-A614 and empty vector were associated with higher levels of Heat Shock Protein (Hsp) 70 and (to a lesser extent) Hsp90 (data not shown) than those from wild-type E N . HSPs are chaperone proteins, and are involved in protein folding, transport and complex formation, and are thought to play a role in cellular stress recovery (345). Studies have suggested that HSPs are up-regulated by oncogenes (such as  148  N P M - A L K and BCR-ABL), to permit accumulation of abnormal proteins (346, 347), but low levels may also be associated with aberrant growth control and cell death (348). HSPs could therefore be controlling a negative regulator of E N signaling; the low levels of HSP in E N expressing cells would thereby allow for the accumulation (and subsequent signal transduction) of E N . In addition, E N immunoprecipitates were found to contain highest levels of periplakin, a cytolinker protein that has been shown to act as a scaffold and possible localization signal in AKT-mediated signaling (349). Plakins are expressed in tissues that experience mechanical stress, where they play a vital role in maintaining tissue integrity (350). Immunoprecipitates from partially transforming EN-Y615F containing lower levels of periplakin, EN-A614 contained lower levels again and no periplakin was immunoprecipitated in samples from vector control cells. While a strong association between periplakin and autoimmune disease has been established, efforts to characterize the function of this protein (including a murine knockout model) have been unsuccessful (351). Periplakin has been shown to interact with the pleckstrin homology domain of A K T , and possibly serve as a localization signal in AKT-mediated signaling (349). Perhaps periplakin functions to stabilize the Akt activation signal induced by EN. Given that both HSPs and periplakin have been identified in E N immunoprecipitates, further studies into their role in EN-mediated transformation are warranted. These studies would include co-immunoprecipitation studies with E N expressing cells transfected with either protein, or an interactor screen such as the yeast two-hybrid if various mutants of both proteins were to be assayed simultaneously. This would allow for determination of the domains required for any potential interaction.  149  Two alternate approaches for identifying interacting proteins are the yeast two-hybrid interactor screen and FRET (fluorescence resonance energy transfer) analysis. Previous attempts with E N and a yeast two-hybrid cDNA library screen in the Sorensen laboratory have not been fruitful, presumably due to overwhelming activation of the reporter by the Etv6 transcription factor portion of E N . It may be possible to use this system to determine if the E N / IRS-1 interaction is direct by using the two proteins as 'bait' and 'prey', rather than screening a library. FRET analysis relies on the non-radioactive transfer of energy from an excited donor fluorophore to an acceptor fluorophore and can be used to study protein interactions as well as changes in protein conformation, folding, and stability (352, 353). Imaging and FRET analysis with confocal and fluorescence microscopy allows for real-time monitoring of protein-protein interactions, and is especially informative for assessment of compartmentalization (354, 355). If, as hypothesized, IRS-1 localizes E N to the cell membrane where it induces key signaling cascades, FRET analysis would be a powerful tool to analyze this. Based on two studies with various fusion tyrosine kinases (FTKs), including B C R - A B L , E T V 6 - A B L , ETV6-JAK2, ETV6-PDGFpR, N P M - A L K , FIIP1- PDGFpR and E N , E N is the only F T K that does not activate STAT5 (130, 197). The phosphorylation of STAT (Signal Transducers and Activators of Transcription) family members triggers nuclear translocation, and activation of a number of nuclear targets involved in proliferation, differentiation, and apoptosis. It may be that STAT5 activation of non-EN FTKs limits the oncogenic potential to a leukemogenic process. Further, the ability of E N to transform multiple cell lineages would be due to its ability to activate many signaling pathways, including Ras-MAPK, PI3K-Akt, and TGF-(3 rather than the STAT pathway alone. Various Stat-deficient models exist, and analysis of  150  E N transformation in these models should define the contributions of STATs, if any, in E N transformation. Recently, the adaptor protein GRB2 has been shown to bind the fusion E T V 6 - A B L via a tyrosine residue in the ETV6 portion (356). Mutation of this interaction site blocked ETV6ABL-induced leukemogenesis in vitro and in vivo, similar to studies involving mutation of the GRB2-binding site in B C R - A B L (98, 357). This tyrosine has also been shown to be responsible for GRB2 association in ETV6-JAK2 (358). However, two independent investigators in our laboratory were not able to detect Grb2 in immunoprecipitates of the E N fusion protein. Further, there are obvious differences in the signaling cascades resulting from mutation of this tyrosine in E T V 6 - A B L and ETV6-JAK2. In contrast to the inactive mutant Y314F in E T V 6 - A B L , mutation of this site in ETV6- JAK2 showed impaired activation of Ras but no decrease in E R K / M A P K activation (356, 358). This suggests that GRB2-dependent signals induced by oncogenic tyrosine kinases differ depending on the cell type and the particular kinase involved. Despite the extensive involvement of ETV6 fusion proteins in the development of cancer, very little is known about the localization of these proteins. To further understand the oncogenic mechanism of action of E N within the cell, we performed localization studies with V5-tagged E N and non-transforming A614. These preliminary results suggest that E N is present throughout the cell, while the A614 mutation restricts localization to the cytoplasm only (Figure 31). These results show significant differences in localization between these two phenotypically different mutants, and suggest that protein subcellular localization is important for transformationinducing interactions.  151 DAPI  a-V5  merge  V5-EN  V5-A614  FIGURE 31. EN-A614 mutant shows different cellular localization than EN. NIH3T3s expressing either V 5 - E N or V5-A614 were fixed in paraformaldehyde, incubated with anti-V5 antibody and analyzed by immunofluorescence microscopy. NIH3T3s expressing either V 5 - E N (top panels) or V5-A614 were probed with an a - V 5 antibody. These experiments show expression of V 5 - E N throughout the cell (cytoplasm and nucleus), while there appears to be less V5-A614 expression in the nucleus.  Caution should be exercised, however, when interpreting these preliminary results since they conflict with previous attempts at identifying the subcellular localization of the E N protein, which show localization of E N and kinase-dead mutant to the cytoplasm (Dr. Stevan Knezevich, unpublished results). These previous studies were performed with antibodies to endogenous protein (either wild-type E T V 6 or N T R K 3 ) and may be reacting with endogenous proteins. Recently, a C-terminally tagged E N protein was found to be exclusively cytoplasmic, while SAM  domain mutants (lacking polymerization ability) localized to both the nucleus and  cytoplasm (231). Dimerization of w i l d type E T V 6 and E N has been documented (129), and is presumed to occur in the nucleus given that wild-type E T V 6 functions as a transcription factor.  152  Recent reports have also suggested that wild-type ETV6 delocalizes from the cytoplasm to the nucleus following sumoylation at lysine 99 (359), and it could also be in the cytoplasm that E N heterodimerizes with wild-type ETV6. The present study, however, is not without limitations. Both V5-EN and V5-A614 are C-terminally tagged. This thesis has provided ample evidence to the importance of protein interactions at the C-terminus of E N , and the V5-tag may interfere structurally with these interactors. Future localization studies should include multiple methods of protein detection to avoid confounding artifactual results. The localization of oncogenic fusion proteins is gathering increasing interest. Studies of the N P M - A L K fusion indicate it is cytoplasmic, but it localizes to the nucleus upon dimerization with the predominantly nuclear N P M shuttle protein (360). In the nucleus, N P M - A L K interacts with NIPA (Nuclear Interacting Partner of Anaplastic Lymphoma Kinase), which subsequently prevents the induction of apoptosis (360). N P M - A L K has also been shown to interact with IRS-1 (361), but mutation of the NPXY-binding site in N P M - A L K did not affect transformation, indicating that either IRS-1 is not vital for transformation or that ERS-1 binding is through a nonN P X Y mechanism (as studies in this thesis have suggested). Finally, a single paper has recently demonstrated the presence (and biological role) of IRS-1 in the nucleus (362). Clearly, further study into the subcellular localization and spatial interactors of oncogenic fusion kinases is warranted. Clearly the most important experimental challenge  is to determine the  specific  mechanism by which E N interacts with IRS-1, and the consequences of this interaction. Given that transformation by E N induces constitutive phosphorylation of IRS-1, it is possible that E N is the kinase that phosphorylates IRS-1. We also hypothesize that IRS-1 localizes E N to the cell membrane. Determination of the nature of the interaction between these two proteins is therefore  153  paramount. There are a number of approaches possible, the most direct being a series of truncation constructs as employed in Chapter III. These E N mutants would be initially screened for their ability to interact with the PTB domain of IRS-1, and then with the entire protein. Crystallization studies done in conjunction with the PTB domain of IRS-1 and the SH2 domain of PLCyl will undoubtedly contribute to the nature of this interaction and suggest possibilities for disruption, but will require considerable labour investment. Given that both B C R - A B L and N P M - A L K have been shown to interact with IRS-1, there is considerable potential for application of these results to other fusion tyrosine kinases.  6.3  M O D E L SYSTEMS FOR INVESTIGATING E N TUMOURIGENESIS Random insertion transgenic strains can generally be established more quickly and less  expensively than those in which the transgene is targeted to a specific genetic site. When we first initiated experiments to generate mice transgenic for EN, random insertion strategies were still in favor, and conditionally targeted approaches were deemed too complicated for initial attempts at in vivo modeling. Since then, concerns regarding abnormal temporal and spatial expression, as well as endogenous levels of gene expression, have risen such that the random insertion strategy may be less appropriate for studies of oncogenic fusion genes. As we have observed E N expression in different tumour types, we have generated transgenic mice by placing the cDNA of the E N fusion oncogene under the control of two ubiquitous promoters. After a considerable latency period, 30% of these transgenic mice develop neoplasms. The demonstration that E N expression induces lymphomas in a transgenic model is the first data suggesting that E N can transform lymphoid cells, in addition to already published reports of transformation activity in fibroblastic (129), myeloid (196), and epithelial cells (131).  154  This oncoprotein is therefore very unique in its ability to transform several different cell lineages. Only one other fusion oncoprotein is able to transform multiple cell lineages, TPM3-ALK, with oncogenic activity in both mesenchymal (inflammatory myofibroblastic tumours) and lymphoid (anaplastic lymphoma) cell types (200). There have been no published reports of t(12;15) in any human lymphomas to date but gains of 15q (NTRK3 is positioned at 15q25) have been reported (363-365). Therefore, further investigation into the presence of E N in clinical cases of lymphoma is warranted in light of these results. Additionally, studies into the capability of E N to transform lymphoid cells (such as Whitlock-Witte cultures (366)) should be carried out to investigate the specific pathways induced upon E N expression. An alternate hypothesis is that the induction of lymphomas by E N is a species-specific phenomenon, and many not correlate well with human clinical samples, as demonstrated by studies with the Ras oncoprotein. In humans, Ras is partly responsible for thyroid, lung and colon cancer and causes 90% of pancreatic cancer (367). By contrast, the gene does not cause pancreatic cancer, but rather breast and skin cancer in (random insertion) transgenic mice (368). Hamad et al. showed (in a cell culture system) that Ras-induced tumourigenesis required RalGEFs in human cells, but not in murine cells, suggesting species variation underlies differential transformation requirements (368). These differential in vivo results may also be due to the random insertion model employed, although this is extremely unlikely given that two independent strains have very similar phenotypes and incidence of disease. Recent models of Ras activating mutants expressed at endogenous levels have shown pancreatic neoplasia (369) and metaplasia (in cooperation with a secondary mutation) (320). The detection of E N protein in lymphomas from transgenic mice was extremely difficult. Moreover, it has been noted that the detection of R N A transcripts in clinical samples has proven  155  challenging (370), leading to speculation that that levels of ETV6-NTRK3 fusion transcripts might be below the threshold of detection for clinical (paraffin-based) assays. It is intriguing to note that although the ETV6-NTRK3 chimeric protein has very potent transforming ability when expressed in immortalized murine fibroblasts (129), CFS, C M N and S B C are relatively nonaggressive clinically (126, 187). One possible explanation for this apparent contradiction is that above a certain level, ETV6-NTRK3 expression in human cells may be incompatible with cell survival. It is well-established that high (above endogenous) levels of Ras can induce either cell cycle arrest or apoptosis (reviewed in (34)). Further to this point, we have not been able to identify E N protein by immunohistochemistry in clinical samples and detection of protein in tumours from transgenic mice by Western analysis has been difficult, suggesting low levels of the protein. It is therefore possible that only cells that express low fusion transcript levels survive and go on to form tumours, which may explain the low metastatic ability of EN-expressing tumours. It may be that levels of E N protein in tumours arising in the transgenic mice are below detectable levels. We are currently collaborating with Dr. Stuart Orkin, Harvard Medical School, on a targeted approach for E N tumourigenesis. While the knock-in strategy appears to be the strongest to recapitulate EN-induced tumourigenesis, several limitations do exist with this strategy. A knock-in strategy does not include the balanced translocation and does not imitate the cell-type-specific, somatic aspects of a tumourigenic translocation. The possibility that both expression products of a translocation can alter the disease phenotype is well illustrated by work on mouse models of the P M L - R A R a (promyelocytic leukemia-retinoic acid receptor alpha) leukemogenic translocation. Using conventional transgenesis, it was shown that expression of the non-leukemogenic R A R a - P M L translocation product affects the phenotype of P M L - R A R a  156  leukemias (333). However, a reciprocal gene fusion was not detected in the E N index cases (2) and work to date with EN has indicated that the reciprocal translocation is not associated with the transformed phenotype. Dr. Orkin's lab has successfully created conditionally-expressing E N knock-in mice, and is currently breeding these mice with strains expressing the Cre recombinase with varying spatial and temporal patterns. Preliminary data from these breeding experiments has suggested that hematologic expression of the E N fusion during embryogenesis induces lethality. Excision of E N in the bone marrow through crosses with the Mx-Cre mouse (371) results in a very aggressive myeloproliferative disease similar to the retroviral bone marrow studies of Liu et al. (130). To date, all clinical cases of congenital fibrosarcoma and cellular mesoblastic nephroma have been associated with a trisomy 11 as well as the E N fusion (183). Neither the random insertion model, nor the knock-in model under development, has addressed this. We postulate the extra copy of Chromosome 11 provides additional IGFII, resulting in an up-regulation of the IGF1R pathway. Moreover, E N is unable to transform cells deficient for IGF1R (215). Mouse models of IGFII overexpression do exist (372-374), and cross-breeding with these strains may be required to fully mimic these clinical lesions. Interestingly, preliminary analysis of secretory breast carcinoma has not revealed the presence of trisomy 11, nor was there any mention of it in the case report of E N in A M L (196). This suggests that E N may use alternate mechanisms in these lineages to elicit transformation. Studies such as microarray and proteomics will address the similarities and differences between various E N transformed cells. Another group is also developing a transgenic model of E N expression (375). The laboratory of Dr. Mel Greaves has created two E N constructs under the direction of the S C L /  157  Tal-1 promoter, which restricts expression to the hematopoietic system (376). EN-expression driven by the S C L / Tal-1 promoter resulted in embryonic lethality, with finding very similar to those of the ETV6 knockout mouse (defective yolk sac angiogenesis) (104). Preliminary findings from transgenic mice created from a second construct, encoding a transgene lacking exon 5 of E N ('Tel-TrkC(L)'), develop leukemias and lymphomas of B cell origin. Interestingly, these tumours did not contain detectable levels of E N protein; transgene expression was only detectable by RT-PCR. These findings are remarkably similar to those presented in this thesis, and support the hypothesis that E N expression above a certain threshold is not well tolerated in vivo.  6.4  FINAL C O M M E N T S While the number of patients with an E N expressing tumour is, to date, relatively low  compared to adult cancers such as breast, colon, and prostate, it is important to examine these (and, for that matter, all childhood cancers) from the point of view of life years saved. A survivor of childhood cancer has many more years to contribute to society than, for instance, the average prostate cancer patient. The EN gene fusion is only one mechanism by which NTRK3 can be activated in an abnormal context, and studies into E N tumour biology may have broader reaching implications than select subsets of childhood malignancy. The malignant consequences of abnormal and mutant N T R K signaling are gaining increasing interest (8, 292). Overexpression of NTRK2 (or its ligand BDNF) has been documented in pancreatic and prostate cancers, Wilms' tumour and neuroblastomas, and is frequently associated with aggressive behavior and poor prognosis (377-  158  380). It is likely that activation of N T R K molecules in a non-neural environment contributes to uncontrolled proliferation and / or survival signals without appropriate feedback mechanisms for regulating these signals. Further investigations, both in vivo and in vitro, into the mechanism of NTRK-mediated transformation will contribute to the design of novel treatment modalities for the lesions.  159  R E F E R E N C E S  1.  Bourgeois, J.M., Knezevich, S.R., Mathers, J.A. and Sorensen, P.H.B. (2000) Molecular detection of the ETV6-NTRK3 gene fusion differentiates congenital fibrosarcoma from other childhood spindle cell tumours. Amer J Surg Path, 2 4 , 937-946.  2.  Knezevich, S.R., McFadden, D.E., Tao, W., Lim, J.F. and Sorensen, P.H. (1998) A novel ETV6-NTRK3 gene fusion in congenital fibrosarcoma. Nat Genet, 18,184-7.  3.  Blume-Jensen, P. and Hunter, T. (2001) Oncogenic kinase signalling. Nature, 4 1 1 , 35565.  4.  Aaronson, S.A. (1991) Growth Factors and Cancer. Science, 2 5 4 , 1146-1152.  5.  Zanke, B.W. (1998) Growth factors and intracellular signaling. In Tannock, I.F. and Hill, R.P. (eds.), The Basic Science of Oncology. 3rd ed. McGraw-Hill, Toronto, pp. 106-33.  6.  Robinson, D.R., Wu, Y . M . and Lin, S.F. (2000) The protein tyrosine kinase family of the human genome. Oncogene, 1 9 , 5548-57.  7.  Futreal, P.A., Kasprzyk, A., Birney, E . , Mullikin, J.C., Wooster, R. and Stratton, M.R. (2001) Cancer and genomics. Nature, 4 0 9 , 850-2.  8.  Bardelli, A., Parsons, D.W., Silliman, N., Ptak, J., Szabo, S., Saha, S., Markowitz, S., Willson, J.K., Parmigiani, G., Kinzler, K.W. et al. (2003) Mutational analysis of the tyrosine kinome in colorectal cancers. Science, 3 0 0 , 949.  9.  Hei din, C H . (1995) Dimerization of cell surface receptors in signal transduction. Cell, 8 0 , 213-23.  10.  Fedi, P., Kimmelman, A. and Aaronson, S.A. (2000) Growth Factor Signal Transduction in Cancer. In Bast, R.C. (ed.), Holland-Frei Cancer Medicine. 5th ed. B C Decker Inc., Hamilton.  11.  Marshall, C.J. (1995) Specificity of receptor tyrosine kinase signaling: transient versus sustained extracellular signal-regulated kinase activation. Cell, 8 0 , 179-85.  12.  Gschwind, A., Fischer, O.M. and Ullrich, A. (2004) The discovery of receptor tyrosine kinases: targets for cancer therapy. Nat Rev Cancer, 4 , 361-70.  13.  Coussens, L . , Yang-Feng, T.L., Liao, Y . C . , Chen, E . , Gray, A., McGrath, J., Seeburg, P.H., Libermann, T . A . , Schlessinger, J., Francke, U . et al. (1985) Tyrosine kinase receptor with extensive homology to E G F receptor shares chromosomal location with neu oncogene. Science, 230,1132-9.  14.  Mitsiades, C.S., Mitsiades, N.S., McMullan, C.J., Poulaki, V . , Shringarpure, R., Akiyama, M . , Hideshima, T., Chauhan, D., Joseph, M . , Libermann, T.A. et al. (2004) Inhibition of the insulin-like growth factor receptor-1 tyrosine kinase activity as a therapeutic strategy  160  for multiple myeloma, other hematologic malignancies, and solid tumors. Cancer Cell, 5, 221-30. 15.  Steelman, L.S., Pohnert, S.C., Shelton, J.G., Franklin, R.A., Bertrand, F . E . and McCubrey, J.A. (2004) JAK7STAT, Raf/MEK/ERK, PI3K7Akt and B C R - A B L in cell cycle progression and leukemogenesis. Leukemia, 18, 189-218.  16.  Chang, F., Steelman, L.S., Lee, J.T., Shelton, J.G., Navolanic, P.M., Blalock, W.L., Franklin, R . A . and McCubrey, J.A. (2003) Signal transduction mediated by the Ras/Raf/MEK/ERK pathway from cytokine receptors to transcription factors: potential targeting for therapeutic intervention. Leukemia, 17, 1263-93.  17.  White, M . F . (2002) IRS proteins and the common path to diabetes. Am J Endocrinol  Metab,  Physiol  283,E413-22.  18.  Rodriguez-Viciana, P., Warne, P.H., Dhand, R., Vanhaesebroeck, B., Gout, I., Fry, M.J., Waterfield, M.D. and Downward, J. (1994) Phosphatidylinositol-3-OH kinase as a direct target of Ras. Nature, 370, 527-32.  19.  Vivanco, I. and Sawyers, C.L. (2002) The phosphatidylinositol 3-Kinase A K T pathway in human cancer. Nat Rev Cancer, 2, 489-501.  20.  Troussard, A.A., Mawji, N.M., Ong, C , Mui, A., St -Arnaud, R. and Dedhar, S. (2003) Conditional knock-out of integrin-linked kinase demonstrates an essential role in protein kinase B/Akt activation. J Biol Chem, 278, 22374-8.  21.  Filippa, N., Sable, C.L., Filloux, C , Hemmings, B. and Van Obberghen, E . (1999) Mechanism of protein kinase B activation by cyclic AMP-dependent protein kinase. Mol Cell Biol, 19, 4989-5000.  22.  Yano, S., Tokumitsu, H . and Soderling, T.R. (1998) Calcium promotes cell survival through CaM-K kinase activation of the protein-kinase-B pathway. Nature, 396, 584-7.  23.  Romashkova, J.A. and Makarov, S.S. (1999) NF-kappaB is a target of A K T in antiapoptotic PDGF signalling. Nature, 401, 86-90.  24.  Mayo, L . D . and Donner, D.B. (2001) A phosphatidylinositol 3-kinase/Akt pathway promotes translocation of Mdm2 from the cytoplasm to the nucleus. Proc Natl Acad Sci USA, 98, 11598-603.  25.  Scheid, M.P. and Woodgett, J.R. (2000) Protein kinases: six degrees of separation? Curr Biol, 10, R191-4.  26.  Von Willebrand, M . , Jascur, T., Bonnefoy-Berard, N., Yano, H., Altman, A., Matsuda, Y. and Mustelin, T. (1996) Inhibition of phosphatidylinositol 3-kinase blocks T cell antigen receptor/CD3-induced activation of the mitogen-activated kinase Erk2. Eur J Biochem, 235, 828-35.  161  27.  Ferby, I.M., Waga, I., Hoshino, M . , Kume, K. and Shimizu, T. (1996) Wortmannin inhibits mitogen-activated protein kinase activation by platelet-activating factor through a mechanism independent of p85/pll0-type phosphatidylinositol 3-kinase. J Biol Chem, 271,11684-8.  28.  King, W . G . , Mattaliano, M.D., Chan, T.O., Tsichlis, P.N. and Brugge, J.S. (1997) Phosphatidylinositol 3-kinase is required for integrin-stimulated A K T and Raf1/mitogen-activated protein kinase pathway activation. Mol Cell Biol, 17,4406-18.  29.  Rommel, C , Clarke, B.A., Zimmermann, S., Nunez, L., Rossman, R., Reid, K., Moelling, K., Yancopoulos, G.D. and Glass, D J . (1999) Differentiation stage-specific inhibition of the Raf-MEK-ERK pathway by Akt. Science, 286, 1738-41.  30.  Craddock, B.L., Hobbs, J., Edmead, C.E. and Welham, M.J. (2001) Phosphoinositide 3kinase-dependent regulation of interleukin-3-induced proliferation: involvement of mitogen-activated protein kinases, SHP2 and Gab2. J Biol Chem, 276, 24274-83.  31.  Treinies, I., Paterson, H.F., Hooper, S., Wilson, R. and Marshall, C.J. (1999) Activated M E K stimulates expression of AP-1 components independently of phosphatidylinositol 3-kinase (PI3-kinase) but requires a PI3-kinase signal To stimulate D N A synthesis. Mol Cell Biol, 19, 321-9.  32.  Gu, H . , Maeda, H . , Moon, J.J., Lord, J.D., Yoakim, M . , Nelson, B . H . and Neel, B.G. (2000) New role for She in activation of the phosphatidylinositol 3-kinase/Akt pathway. Mol Cell Biol, 20, 7109-20.  33.  Sattler, M . , Mohi, M . G . , Pride, Y.B., Quinnan, L.R., Malouf, N.A., Podar, K., Gesbert, F., Iwasaki, H., Li, S., Van Etten, R.A. et al. (2002) Critical role for Gab2 in transformation by B C R / A B L . Cancer Cell, 1, 479-92.  34.  Roovers, K. and Assoian, R.K. (2000) Integrating the M A P kinase signal into the G l phase cell cycle machinery. Bioessays, 22, 818-26.  35.  Gille, H . and Downward, J. (1999) Multiple ras effector pathways contribute to G(l) cell cycle progression. J Biol Chem, 274, 22033-40.  36.  Andreef, M . , Goodrich, D.W. and Pardee, A . B . (2003) Cell Proliferation and Differentiation. In Kufe, D.W. (ed.), Holland-Frei Cancer Medicine. 6th ed. B C Decker Inc., Hamilton, pp. 27-39.  37.  Slingerland, J.M. and Tannock, LF. (1998) Cell Proliferation and Cell Death. In Tannock, I.F. and Hill, R.P. (eds.), The Basic Science of Oncology. 3rd ed. McGraw-Hill, Toronto, pp.134-165.  38.  Sherr, C.J. (2000) The Pezcoller lecture: cancer cell cycles revisited. Cancer 3689-95.  Res, 60,  162  39.  Harper, J.W. and Adams, P.D. (2001) Cyclin-dependent kinases. Chem Rev, 1 0 1 , 251126.  40.  Nevins, J.R. (1998) Toward an understanding of the functional complexity of the E2F and retinoblastoma families. Cell Growth Differ, 9, 585-93.  41.  Kerr, J.F., Wyllie, A . H . and Currie, A.R. (1972) Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer, 26, 239-57.  42.  Reed, J.C. (2003) Apoptosis and Cancer. In Holland-Frei Cancer Medicine. 6th ed. B C Decker Inc., Hamilton, pp. 41-52.  43.  Shi, Y . (2002) Mechanisms of caspase activation and inhibition during apoptosis. Mol Cell, 9, 459-70.  44.  Thornberry, N.A. and Lazebnik, Y . (1998) Caspases: enemies within. Science, 2 8 1 , 1312-6.  45.  Wyllie, A . H . (1997) Apoptosis and carcinogenesis. Eur J Cell Biol, 73,189-97.  46.  Earnshaw, W . C . , Martins, L . M . and Kaufmann, S.H. (1999) Mammalian caspases: structure, activation, substrates, and functions during apoptosis. Annu Rev Biochem, 6 8 , 383-424.  47.  Cotran, R.S., Kumar, V . and Robbins, S.L. (eds.) (1994) Robbins Pathologic Basis of Disease. 5th ed. W.B. Saunders & Co., Philadelphia.  48.  • Canada, N.C.I.o. (2004) Canadian Cancer Statistics 2004. NCIC, Canadian Cancer Society, Toronto.  49.  Hanahan, D. and Weinberg, R.A. (2000) The hallmarks of cancer. Cell, 100, 57-70.  50.  Hungerford, D.A. (1978) Some early studies of human chromosomes, 1879—1955. Cytogenet Cell Genet, 2 0 , 1-11.  51.  Classon, M . and Harlow, E . (2002) The retinoblastoma tumour suppressor in development and cancer. Nat Rev Cancer, 2 , 910-7.  52.  Santarosa, M . and Ashworth, A. (2004) Haploinsufficiency for tumour suppressor genes: when you don't need to go all the way. Biochim Biophys Acta, 1654, 105-22.  53.  Park, B.H. and Vogelstein, B. (2003) Tumour-Suppressor Genes. In Kufe, D.W. (ed.), Holland-Frei Cancer Medicine. 6th ed. B C Decker Inc., Hamilton, pp. 87-105.  54.  Kopnin, B.P. (2000) Targets of oncogenes and tumor suppressors: key for understanding basic mechanisms of carcinogenesis. Biochemistry (Mosc), 6 5 , 2-27.  55.  Baker, S.J. and McKinnon, P.J. (2004) Tumour-suppressor function in the nervous system. Nat Rev Cancer, 4, 184-96.  163  56.  Knudson, A . G . (1978) Retinoblastoma: a prototypic hereditary neoplasm. Semin Oncol, 5, 57-65.  57.  Friend, S.H., Bernards, R., Rogelj, S., Weinberg, R.A., Rapaport, J.M., Albert, D . M . and Dryja, T.P. (1986) A human D N A segment with properties of the gene that predisposes to retinoblastoma and osteosarcoma. Nature, 323, 643-6.  58.  Fung, Y . K . , Murphree, A.L., T'Ang, A., Qian, J., Hinrichs, S.H. and Benedict, W.F. (1987) Structural evidence for the authenticity of the human retinoblastoma gene. Science, 236,1657-61.  59.  Pucci, B., Kasten, M . and Giordano, A. (2000) Cell cycle and apoptosis. Neoplasia, 291-9.  60.  Ewen, M.E., Xing, Y . G . , Lawrence, J.B. and Livingston, D . M . (1991) Molecular cloning, chromosomal mapping, and expression of the cDNA for pl07, a retinoblastoma gene product-related protein. Cell, 66, 1155-64.  61.  Mayol, X., Grana, X . , Baldi, A., Sang, N., Hu, Q. and Giordano, A. (1993) Cloning of a new member of the retinoblastoma gene family (pRb2) which binds to the E1A transforming domain. Oncogene, 8, 2561-6.  62.  Mulligan, G. and Jacks, T. (1998) The retinoblastoma gene family: cousins with overlapping interests. Trends Genet, 14, 223-9.  63.  Lane, D.P. and Benchimol, S. (1990) p53: oncogene or anti-oncogene? Genes Dev, 4, 1-8.  64.  Hollstein, M . , Hergenhahn, M . , Yang, Q., Bartsch, H., Wang, Z.Q. and Hainaut, P. (1999) New approaches to understanding p53 gene tumor mutation spectra. Mutat Res, 431, 199209.  65.  Hofseth, L.J., Hussain, S.P. and Harris, C C . (2004) p53: 25 years after its discovery. Trends Pharmacol  2,  Sci, 25, 177-81.  66.  Vogelstein, B. and Kinzler, K.W. (1992) p53 function and dysfunction. Cell, 70, 523-526.  67.  Kinzler, K.W. and Vogelstein, B. (1997) Cancer-susceptibility genes. Gatekeepers and caretakers. Nature, 386, 761, 763.  68.  El-Deiry, W.S., Tokino, T., Velculescu, V . E . , Levy, D.B., Parsons, R., Trent, J.M., Lin, D., Mercer, E., Kinzler, K.W. and Vogelstein, B. (1993) WAF1, a potential mediator of p53 tumor suppression. Cell, 75, 817-825.  69.  Kastan, M.B., Zhan, Q., El-Deiry, W.S., Carrier, F., Jacks, T., Walsh, W.V., Plunkett, B.S., Vogelstein, B. and Fornace, A.J. (1992) A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia. Cell, 71, 587597.  164  70.  Yang, Y . , L i , C C . and Weissman, A . M . (2004) Regulating the p53 system through ubiquitination. Oncogene, 23, 2096-106.  71.  L i n , Y . , M a , W . and Benchimol, S. (2000) Pidd, a new death-domain-containing protein, is induced by p53 and promotes apoptosis. Nat Genet, 26, 122-7.  72.  Y u , J., Zhang, L . , Hwang, P . M . , Kinzler, K . W . and Vogelstein, B . (2001) P U M A induces the rapid apoptosis of colorectal cancer cells. Mol Cell, 7, 673-82.  73.  Leng, R . P . , L i n , Y . , M a , W . , W u , H . , Lemmers, B . , Chung, S., Parant, J . M . , Lozano, G . , Hakem, R. and Benchimol, S. (2003) Pirh2, a p53-induced ubiquitin-protein ligase, promotes p53 degradation. Cell, 112, 779-91.  74.  Futreal, P . A . , C o i n , L . , Marshall, M . , D o w n , T., Hubbard, T., Wooster, R., Rahman, N . and Stratton, M . R . (2004) A census of human cancer genes. Nat Rev Cancer, 4, 177-83.  75.  Pierotti, M . A . , Sozzi, G . and Croce, C M . (2003) Oncogenes. In Holland-Frei Cancer Medicine. 6th ed. B C Decker Inc., Hamilton, pp. 73-85.  76.  Kanter, M . R . , Smith, R . E . and Hayward, W . S . (1988) Rapid induction of B-cell lymphomas: insertional activation of c-myb by avian leukosis virus. J Virol, 62, 1423-32.  77.  Hayward, W . S . , Neel, B . G . and Astrin, S . M . (1981) Activation of a cellular one gene by promoter insertion in A L V - i n d u c e d lymphoid leukosis. Nature, 290, 475-80.  78.  Cowell, J . K . (1982) Double minutes and homogeneously staining regions: gene amplification in mammalian cells. Annu Rev Genet, 16, 21-59.  79.  Hiyama, E . , Yokoyama, T., Ichikawa, T., Ishii, T. and Hiyama, K . (1990) N-myc gene amplification and other prognosis-associated factors in neuroblastoma. / Pediatr Surg, 25, 1095-9.  80.  Jiang, W . , Kanter, M . R . , Dunkel, I., Ramsay, R . G . , Beemon, K . L . and Hayward, W . S . (1997) M i n i m a l truncation of the c-myb gene product in rapid-onset B - c e l l lymphoma. J Virol, 71, 6526-33.  81.  Bos, J.L. (1988) The ras gene family and human carcinogenesis. Mutat Res, 195, 255-71.  82.  Galiana, C , Lozano, J . C , Bancel, B . , Nakazawa, H . and Yamasaki, H . (1995) High frequency of Ki-ras amplification and p53 gene mutations in adenocarcinomas of the human esophagus. Mol Carcinog, 14, 286-93.  83.  Bishop, J . M . (1991) Molecular Themes in Oncogenesis. Cell, 64, 235-248.  84.  Minamoto, T., M a i , M . and Ronai, Z . (2000) K-ras mutation: early detection in molecular diagnosis and risk assessment of colorectal, pancreas, and lung cancers—a review. Cancer  Detect Prev, 24, 1-12.  165  85.  Park, M . (1998) Oncogenes. In Vogelstein, B. and Kinzler, K . W . (eds.), The Genetic Basis of Human Cancer. The McGraw-Hill Companies, Inc.. New York.  86.  Oliveira, A . M . and Fletcher, J.A. (eds.) (2003) Translocation Macmillan Publishers Ltd, Nature Publishing Group.  87.  Lengauer, C , Kinzler, K.W. and Vogelstein, B. (1998) Genetic instabilities in human cancers. Nature, 396, 643-9.  88.  Greaves, M.F. and Wiemels, J. (2003) Origins of chromosome translocations in childhood leukaemia. Nat Rev Cancer, 3, 639-49.  89.  Siebert, R., Matthiesen, P., Harder, S., Zhang, Y . , Borowski, A . , Zuhlke-Jenisch, R., Metzke, S., Joos, S., Weber-Matthiesen, K., Grote, W. et al. (1998) Application of interphase fluorescence in situ Hybridization for the detection of the Burkitt translocation t(8;14)(q24;q32) in B-cell lymphomas. Blood, 91, 984-90.  90.  Shtivelman, E . , Lifshitz, B., Gale, R.P. and Canaani, E . (1985) Fused transcript of abl and bcr genes in chronic myelogenous leukaemia. Nature, 315, 550-554.  91.  Mitelman, F., Johansson, B. and Mertens, F. (2004) Fusion genes and rearranged genes as a linear function of chromosome aberrations in cancer. Nat Genet, 36, 331-4.  92.  Albertson, D . G . , Collins, C , McCormick, F. and Gray, J.W. (2003) Chromosome aberrations in solid tumors. Nat Genet, 34, 369-76.  93.  Sorensen, P.H. and Triche, T.J. (1996) Gene fusions encoding chimaeric transcription factors in solid tumours. Semin Cancer Biol, 7, 3-14.  94.  Delattre, O., Zucman, J., Ploustagel, B., Desmaze, C , Melot, T., Peter, M . , Kovar, H . , Joubert, I., de Jong, P., Rouleau, G. et al. (1992) Gene fusion with an ETS D N A binding domain caused by chromosome translocation in human cancers. Nature, 359, 162-165.  95.  May, W.A., Gishizky, M.L., Lessnick, S.L., Lunsford, L.B., Lewis, B.C., Delattre, O., Zucman, J., Thomas, G. and Denny, C T . (1993) Ewing sarcoma 11;22 translocation produces a chimeric transcription factor that requires the DNA-binding domain encoded  breakpoints  in cancer.  by F L U for transformation. Proc Natl Acad Sci USA, 90, 5752-5756.  96.  Sorensen, P.H.B., Lessnick, S.L., Lopez-Terrada, D., Liu, X.F., Triche, T.J. and Denny, C T . (1994) A second Ewing's sarcoma translocation, t(21;22), fuses the EWS gene to another ETS-family transcription factor, ERG. Nature Genet, 6, 146-151.  97.  Rowley, J.D. (1973) A new consistent chromosomal abnormality in chronic myelogenous leukaemia identified by quinacrine fluorescence and Giemsa staining. Nature, 243, 290293.  166  98.  Puil, L., Liu, J., Gish, G., Mbamalu, G., Bowtell, D., Pelicci, P.G., Arlinghaus, R. and Pawson, T. (1994) Bcr-Abl oncoproteins bind directly to activators of the Ras signalling pathway. Embo J, 13, 764-73.  99.  Cortez, D., Kadlec, L . and Pendergast, A . M . (1995) Structural and signaling requirements for BCR-ABL-mediated transformation and inhibition of apoptosis. Mol Cell Biol, 15,5531-41.  100.  Bartolo, C. and Viswanatha, D.S. (2000) Molecular diagnosis in pediatric acute leukemias. Clin Lab Med, 20, 139-82, x.  101.  Ayton, P.M. and Cleary, M . L . (2001) Molecular mechanisms of leukemogenesis mediated by M L L fusion proteins. Oncogene, 20, 5695-707.  102.  Bloomfield, C D . and Caligiuri, M . A . (2000) Gene Fusions in Leukemia. In DeVita, V.T. (ed.), Principles and Practice of Oncology. 6 ed. J.B. Lippincott, Philadelphia, Vol. 2, pp. 2392-99.  103.  Golub, T.R., Barker, G.F., Lovett, M . and Gilliland, D . G . (1994) Fusion of PDGF receptor beta to a novel ets-like gene, tel, in chronic myelomonocytic leukemia with t(5;12) chromosomal translocation. Cell, 11, 307-16.  104.  Wang, L . C , Kuo, F., Fujiwara, Y., Gilliland, D.G., Golub, T.R. and Orkin, S.H. (1997) Yolk sac angiogenic defect and intra-embryonic apoptosis in mice lacking the Ets-related factor T E L . Embo J, 16, 4374-83.  105.  Qiao, Y., Ogawa, S., Hangaishi, A., Yuji, K., Izutsu, K., Kunisato, A., Imai, Y., Wang, L., Hosoya, N., Nannya, Y. et al. (2003) Identification of a novel fusion gene, T T L , fused to ETV6 in acute lymphoblastic leukemia with t(12;13)(pl3;ql4), and its implication in leukemogenesis. Leukemia, 17,1112-20.  106.  Chase, A., Reiter, A., Burci, L . , Cazzaniga, G., Biondi, A., Pickard, J., Roberts, I.A., Goldman, J.M. and Cross, N.C. (1999) Fusion of ETV6 to the caudal-related homeobox gene CDX2 in acute myeloid leukemia with the t(12;13)(pl3;ql2). Blood, 93, 1025-31.  107.  Rabbitts, T.H. and Stocks, M.R. (2003) Chromosomal translocation products engender new intracellular therapeutic technologies. Nat Med, 9, 383-6.  108.  Druker, B. (2001) Signal transduction inhibition: results from phase I clinical trials in chronic myeloid leukemia. Semin Hematol, 38, 9-14.  109.  Golub, T.R., Goga, A., Barker, G.F., Afar, D . E . , McLaughlin, J., Bohlander, S.K., Rowley, J.D., Witte, O.N. and Gilliland, D . G . (1996) Oligomerization of the A B L tyrosine kinase by the Ets protein T E L in human leukemia. Mol Cell Biol, 16,4107-16.  110.  Raynaud, S.D., Baens, M . , Grosgeorge, J., Rodgers, K., Reid, C D . , Dainton, M . , Dyer, M . , Fuzibet, J.G., Gratecos, N., Taillan, B. et al. (1996) Fluorescence in situ  167  hybridization analysis of t(3; 12)(q26; pl3): a recurring chromosomal abnormality involving the T E L gene (ETV6) in myelodysplastic syndromes. Blood, 88, 682-9. 111.  Cools, J., Bilhou-Nabera, C , Wlodarska, I., Cabrol, C , Talmant, P., Bernard, P., Hagemeijer, A. and Marynen, P. (1999) Fusion of a novel gene, B T L , to ETV6 in acute myeloid leukemias with a t(4;12)(qll-ql2;pl3). Blood, 94, 1820-4.  112.  Peeters, P., Raynaud, S.D., Cools, J., Wlodarska, I., Grosgeorge, J., Philip, P., Monpoux, F., Van Rompaey, L., Baens, M . , Van den Berghe, H . et al. (1997) Fusion of T E L , the ETS-variant gene 6 (ETV6), to the receptor- associated kinase JAK2 as a result of t(9;12) in a lymphoid and t(9;15;12) in a myeloid leukemia. Blood, 90, 2535-40.  113.  Buijs, A., Sherr, S., van Baal, S., van Bezouw, S., van der Plas, D., Geurts van Kessel, A., Riegman, P., Lekanne Deprez, R., Zwarthoff, E . , Hagemeijer, A. et al. (1995) Translocation (12;22) (pl3;qll) in myeloproliferative disorders results in fusion of the ETS-like T E L gene on 12pl3 to the MN1 gene on 22qll [published erratum appears in Oncogene 1995 Aug 17;11(4):809]. Oncogene, 10, 1511-9.  114.  Suto, Y., Sato, Y . , Smith, S.D., Rowley, J.D. and Bohlander, S.K. (1997) A t(6;12)(q23;pl3) results in the fusion of ETV6 to a novel gene, STL, in a B-cell A L L cell line. Genes Chromosomes Cancer, 18, 254-68.  115.  Yagasaki, F., Jinnai, I., Yoshida, S., Yokoyama, Y., Matsuda, A., Kusumoto, S., Kobayashi, H., Terasaki, H., Ohyashiki, K., Asou, N. et al. (1999) Fusion of T E L / E T V 6 to a novel ACS2 in myelodysplastic syndrome and acute myelogenous leukemia with t(5;12)(q31;pl3). Genes Chromosomes Cancer, 26, 192-202.  116.  Iijima, Y., Ito, T., Oikawa, T., Eguchi, M . , Eguchi-Ishimae, M . , Kamada, N., Kishi, K., Asano, S., Sakaki, Y. and Sato, Y . (2000) A new E T V 6 / T E L partner gene, A R G (ABLrelated gene or ABL2), identified in an A M L - M 3 cell line with a t(l;12)(q25;pl3) translocation. Blood, 95, 2126-31.  117.  Golub, T.R., Barker, G.F., Bohlander, S.K., Hiebert, S.W., Ward, D.C., Bray-Ward, P., Morgan, E . , Raimondi, S.C., Rowley, J.D. and Gilliland, D.G. (1995) Fusion of the TEL gene on 12pl3 to the AML1 gene on 21q22 in acute lymphoblastic leukemia. Proc Natl Acad Sci USA, 92, 4917-4921.  118.  Odero, M . D . , Vizmanos, J.L., Roman, J.P., Lahortiga, I., Panizo, C , Calasanz, M.J., Zeleznik-Le, N.J., Rowley, J.D. and Novo, F.J. (2002) A novel gene, MDS2, is fused to E T V 6 / T E L in a t(l;12)(p36.1;pl3) in a patient with myelodysplastic syndrome. Genes  Chromosomes Cancer, 35, 11-9.  119.  Kuno, Y., Abe, A., Emi, N., Iida, M . , Yokozawa, T., Towatari, M . , Tanimoto, M . and Saito, H . (2001) Constitutive kinase activation of the TEL-Syk fusion gene in myelodysplastic syndrome with t(9;12)(q22;pl2). Blood, 97, 1050-5.  120.  Corral, J., Lavenir, I., Impey, H . , Warren, A.J., Forster, A., Larson, T . A . , Bell, S., McKenzie, A.N., King, G. and Rabbitts, T.H. (1996) An M11-AF9 fusion gene made by  168  homologous recombination causes acute leukemia in chimeric mice: a method to create fusion oncogenes. Cell, 85, 853-61. 121.  Macleod, K.F. and Jacks, T. (1999) Insights into cancer from transgenic mouse models. J Pathol, 187, 43-60.  122.  Roberts, R.B., Arteaga, C L . and Threadgill, D.W. (2004) Modeling the cancer patient with genetically engineered mice: prediction of toxicity from molecule-targeted therapies. Cancer Cell, 5, 115-20.  123.  Helman, L.J. and Meltzer, P. (2003) Mechanisms of sarcoma development. Nat Rev Cancer, 3, 685-94.  124.  Perez-Losada, J., Pintado, B., Gutierrez-Adan, A., Flores, T., Banares-Gonzalez, B., del Campo, J . C , Martin-Martin, J.F., Battaner, E . and Sanchez-Garcia, I. (2000) The chimeric FUS/TLS-CHOP fusion protein specifically induces liposarcomas in transgenic mice. Oncogene, 19, 2413-22.  125.  Sanchez-Garcia, I. and Rabbitts, T.H. (1994) Transcriptional activation by T A L I and FUS-CHOP proteins expressed in acute malignancies as a result of chromosomal abnormalities. Proc Natl Acad Sci USA, 91, 7869-73.  126.  Triche, T.J. and Sorensen, P.H.B. (2002) Molecular Pathology of Pediatric Malignancies. In Pizzo, P. A. and Poplack, D.G. (eds.), Principles and Practice of Pediatric Oncology. 4th ed ed. Lippincott Williams & Wilkins, Philadelphia, pp. 161-204.  127.  Sorensen, P.H., Lynch, J . C , Qualman, S.J., Tirabosco, R., Lim, J.F., Maurer, H . M . , Bridge, J.A., Crist, W.M., Triche, T.J. and Barr, F.G. (2002) PAX3-FKHR and PAX7F K H R gene fusions are prognostic indicators in alveolar rhabdomyosarcoma: a report from the children's oncology group. J Clin Oncol, 20, 2672-9.  128.  Kaplan, D.R. and Miller, F.D. (2000) Neurotrophin signal transduction in the nervous system. Curr Opin Neurobiol, 10, 381-91.  129.  Wai, D.H., Knezevich, S.R., Lucas, T., Jansen, B., Kay, R.J. and Sorensen, P.H. (2000) The ETV6-NTRK3 gene fusion encodes a chimeric protein tyrosine kinase that transforms NIH3T3 cells. Oncogene, 19, 906-15.  130.  Liu, Q., Schwaller, J., Kutok, J., Cain, D., Aster, J . C , Williams, I.R. and Gilliland, D.G. (2000) Signal transduction and transforming properties of the T E L - T R K C fusions associated with t(12;15)(pl3;q25) in congenital fibrosarcoma and acute myelogenous leukemia. EMBO J, 19,1827-38.  131.  Tognon, C , Knezevich, S.R., Huntsman, D., Roskelley, C D . , Melnyk, N., Mathers, J.A., Becker, L . , Carneiro, F., MacPherson, N., Horsman, D. et al. (2002) Expression of the ETV6-NTRK3 gene fusion as a primary event in human secretory breast carcinoma. Cancer Cell, 2, 367-76.  169  132.  Baens, M . , Peeters, P., Guo, C , Aerssens, J. and Marynen, P. (1996) Genomic organization of T E L : the human ETS-variant gene 6. Genome Res, 6,404-13.  133.  Edel, M.J. (1998) The ETS-related factor T E L is regulated by angiogenic growth factor V E G F in HUVE-cells. Anticancer Res, 18, 4505-9.  134.  O'Connor, H.E., Butler, T.A., Clark, R., Swanton, S., Harrison, C.J., Seeker-Walker, L . M . and Foroni, L . (1998) Abnormalities of the ETV6 gene occur in the majority of patients with aberrations of the short arm of chromosome 12: a combined PCR and Southern blotting analysis. Leukemia, 12, 1099-106.  135.  Poirel, H . , Oury, C , Carron, C , Duprez, E . , Laabi, Y . , Tsapis, A., Romana, S.P., Mauchauffe, M . , Le Coniat, M . , Berger, R. et al. (1997) The T E L gene products: nuclear phosphoproteins with D N A binding properties. Oncogene, 14, 349-57.  136.  Wang, L . C . , Swat, W., Fujiwara, Y., Davidson, L . , Visvader, J., Kuo, F., Alt, F.W., Gilliland, D.G., Golub, T.R. and Orkin, S.H. (1998) The T E L / E T V 6 gene is required specifically for hematopoiesis in the bone marrow. Genes Dev, 12, 2392-402.  137.  Szymczyna, B.R. and Arrowsmith, C.H. (2000) D N A binding specificity studies of four ETS proteins support an indirect read-out mechanism of protein-DNA recognition. J Biol Chem, 275, 28363-70.  138.  Chakrabarti, S.R. and Nucifora, G. (1999) The leukemia-associated gene T E L encodes a transcription repressor which associates with SMRT and mSin3A. Biochem Biophys Res Commun, 264, 871-7.  139.  Wang, L . and Hiebert, S.W. (2001) T E L contacts multiple co-repressors and specifically associates with histone deacetylase-3. Oncogene, 20, 3716-25.  140.  Fenrick, R., Wang, L., Nip, J., Amann, J.M., Rooney, R.J., Walker-Daniels, J., Crawford, H.C., Hulboy, D.L., Kinch, M.S., Matrisian, L . M . et al. (2000) T E L , a putative tumor suppressor, modulates cell growth and cell morphology of ras-transformed cells while repressing the transcription of stromelysin-1. Mol Cell Biol, 20, 5828-39.  141.  Irvin, B.J., Wood, L.D., Wang, L., Fenrick, R., Sansam, C.G., Packham, G., Kinch, M . , Yang, E . and Hiebert, S.W. (2003) T E L , a putative tumor suppressor, induces apoptosis and represses transcription of Bcl-XL. J Biol Chem, 278, 46378-86.  142.  Van Rompaey, L., Dou, W., Buijs, A. and Grosveld, G. (1999) Tel, a frequent target of leukemic translocations, induces cellular aggregation and influences expression of extracellular matrix components. Neoplasia, 1, 526-36.  143.  Hatta, Y., Takeuchi, S., Yokota, J. and Koeffler, H.P. (1997) Ovarian cancer has frequent loss of heterozygosity at chromosome 12pl2.3-13.1 (region of T E L and K i p l loci) and chromosome 12q23-ter: evidence for two new tumour-suppressor genes. Br J Cancer, 75, 1256-62.  170  144.  Sato, Y., Suto, Y., Pietenpol, J., Golub, T.R., Gilliland, D.G., Davis, E . M . , Le Beau, M . M . , Roberts, J.M., Vogelstein, B., Rowley, J.D. et al. (1995) T E L and KTP1 define the smallest region of deletions on 12pl3 in hematopoietic malignancies. Blood, 86, 1525-33.  145.  Takeuchi, S., Bartram, C.R., Miller, C.W., Reiter, A., Seriu, T., Zimmerann, M . , Schrappe, M . , Mori, N., Slater, J., Miyoshi, I. et al. (1996) Acute lymphoblastic leukemia of childhood: identification of two distinct regions of deletion on the short arm of chromosome 12 in the region of T E L and KIP1. Blood, 87, 3368-74.  146.  Spirin, K.S., Simpson, J.F., Takeuchi, S., Kawamata, N., Miller, C.W. and Koeffler, H.P. (1996) p27/Kipl mutation found in breast cancer. Cancer Res, 56, 2400-4.  147.  Romana, S.P., Mauchauffe, M . , Le Coniat, M . , Chumakov, I., Le Paslier, D., Berger, R. and Bernard, O.A. (1995) The t(12;21) of acute lymphoblastic leukemia results in a telAML1 gene fusion. Blood, 85, 3662-70.  148.  Papadopoulos, P., Ridge, S.A., Boucher, C.A., Stocking, C. and Wiedemann, L . M . (1995) The novel activation of A B L by fusion to an ets-related gene, T E L . Cancer Res, 55, 34-8.  149.  Lacronique, V., Boureux, A., Valle, V . D . , Poirel, H . , Quang, C.T., Mauchauffe, M . , Berthou, C , Lessard, M . , Berger, R., Ghysdael, J. et al. (1997) A T E L - J A K 2 fusion protein with constitutive kinase activity in human leukemia. Science, 278, 1309-12.  150.  Schwaller, J., Frantsve, J., Aster, J., Williams, I.R., Tomasson, M . H . , Ross, T.S., Peeters, P., Van Rompaey, L . , Van Etten, R.A., Ilaria, R., Jr. et al. (1998) Transformation of hematopoietic cell lines to growth-factor independence and induction of a fatal myeloand lymphoproliferative disease in mice by retrovirally transduced TEL/JAK2 fusion genes. Embo J, 17, 5321-33.  151.  Cazzaniga, G., Tosi, S., Aloisi, A., Giudici, G., Daniotti, M . , Pioltelli, P., Kearney, L . and Biondi, A. (1999) The tyrosine kinase abl-related gene A R G is fused to ETV6 in an AML-M4Eo patient with a t(l;12)(q25;pl3): molecular cloning of both reciprocal transcripts. Blood, 94, 4370-3.  152.  Yagasaki, F., Wakao, D., Yokoyama, Y., Uchida, Y., Murohashi, I., Kayano, H . , Taniwaki, M . , Matsuda, A. and Bessho, M . (2001) Fusion of E T V 6 to fibroblast growth factor receptor 3 in peripheral T-cell lymphoma with a t(4;12)(pl6;pl3) chromosomal translocation. Cancer Res, 61, 8371-4.  153.  Rowley, J.D. (1999) The role of chromosome translocations in leukemogenesis. Semin Hematol, 36, 59-72.  154.  Cave, H . , Cacheux, V., Raynaud, S., Brunie, G., Bakkus, M . , Cochaux, P., Preudhomme, C , Lai, J.L., Vilmer, E . and Grandchamp, B. (1997) ETV6 is the target of chromosome 12p deletions in t(12;21) childhood acute lymphocytic leukemia. Leukemia, 11, 1459-64.  155.  Barbacid, M . (1995) Structural and functional properties of the T R K family of neurotrophin receptors. Ann N Y Acad Sci, 766, 442-58.  171  156.  Hallbook, F., Ibanez, C.F. and Persson, H . (1991) Evolutionary studies of the nerve growth factor family reveal a novel member abundantly expressed in Xenopus ovary. Neuron, 6, 845-58.  157.  Bothwell, M . (1995) Functional interactions of neurotrophins and neurotrophin receptors. Annu Rev Neurosci, 18, 223-53.  158.  Barbacid, M . (1994) The Trk family of neurotrophin receptors. / Neurobiol, 25, 1386403.  159.  Barbacid, M . (1995) Neurotrophic factors and their receptors. Curr Opin Cell Biol, 7, 148-55.  160.  Conover, J.C. and Yancopoulos, G.D. (1997) Neurotrophin regulation of the developing nervous system: analyses of knockout mice. Rev Neurosci, 8, 13-27.  161.  Ichaso, N., Rodriguez, R.E., Martin-Zanca, D. and Gonzalez-Sarmiento, R. (1998) Genomic characterization of the human trkC gene. Oncogene, 17, 1871-5.  162.  McGregor, L . M . , Baylin, S.B., Griffin, C.A., Hawkins, A . L . and Nelkin, B.D. (1994) Molecular cloning of the cDNA for human TrkC (NTRK3), chromosomal assignment, and evidence for a splice variant. Genomics, 22, 267-72.  163.  Guiton, M . , Gunn-Moore, F.J., Glass, D.J., Geis, D.R., Yancopoulos, G.D. and Tavare, J.M. (1995) Naturally occurring tyrosine kinase inserts block high affinity binding of phospholipase C gamma and She to TrkC and neurotrophin-3 signaling. J Biol Chem, 270, 20384-90.  164.  Lemmon, M . A . and Schlessinger, J. (1994) Regulation of signal transduction and signal diversity by receptor oligomerization. Trends Biochem Sci, 19,459-63.  165.  Obermeier, A., Lammers, R., Wiesmuller, K . H . , Jung, G., Schlessinger, J. and Ullrich, A. (1993) Identification of Trk binding sites for SHC and phosphatidylinositol 3'- kinase and formation of a multimeric signaling complex. J Biol Chem, 268, 22963-6.  166.  Hallberg, B., Ashcroft, M . , Loeb, D.M., Kaplan, D.R. and Downward, J. (1998) Nerve growth factor induced stimulation of Ras requires Trk interaction with She but does not involve phosphoinositide 3-OH kinase. Oncogene, 17, 691-7.  167.  Obermeier, A., Halfter, H . , Wiesmuller, K . H . , Jung, G., Schlessinger, J. and Ullrich, A. (1993) Tyrosine 785 is a major determinant of Trk—substrate interaction. Embo J, 12, 933-41.  168.  Rabin, S.J., Cleghon, V. and Kaplan, D.R. (1993) SNT, a differentiation-specific target of neurotrophic factor-induced tyrosine kinase activity in neurons and PC12 cells. Mol Cell Biol, 13, 2203-13.  172  169.  Peng, X., Greene, L . A . , Kaplan, D.R. and Stephens, R . M . (1995) Deletion of a conserved juxtamembrane sequence in Trk abolishes NGF- promoted neuritogenesis. Neuron, 15, 395-406.  170.  Qian, X., Riccio, A., Zhang, Y. and Ginty, D.D. (1998) Identification and characterization of novel substrates of Trk receptors in developing neurons. Neuron, 21, 1017-29.  171.  Bongarzone, I., Pierotti, M.A., Monzini, N., Mondellini, P., Manenti, G., Donghi, R., Pilotti, S., Grieco, M . , Santoro, M . , Fusco, A. et al. (1989) High frequency of activation of tyrosine kinase oncogenes in human papillary thyroid carcinoma. Oncogene, 4, 145762.  172.  Pahlman, S. and Hoehner, J.C. (1996) Neurotrophin receptors, tumor progression and tumor maturation. Mol Med Today, 2, 432-8.  173.  Miknyoczki, S.J., Lang, D., Huang, L . , Klein-Szanto, A.J., Dionne, C A . and Ruggeri, B.A. (1999) Neurotrophins and Trk receptors in human pancreatic ductal adenocarcinoma: expression patterns and effects on in vitro invasive behavior. Int J Cancer, 81, 417-27.  174.  Reuther, G.W., Lambert, Q.T., Caligiuri, M . A . and Der, C J . (2000) Identification and characterization of an activating TrkA deletion mutation in acute myeloid leukemia. Mol Cell Biol, 20, 8655-66.  175.  Weeraratna, A . T . , Arnold, J.T., George, D.J., DeMarzo, A. and Isaacs, J.T. (2000) Rational basis for Trk inhibition therapy for prostate cancer. Prostate, 45, 140-8.  176.  Walch, E.T. and Marchetti, D. (1999) Role of neurotrophins and neurotrophins receptors in the in vitro invasion and heparanase production of human prostate cancer cells. Clin Exp Metastasis, 17, 307-14.  177.  Melck, D., De Petrocellis, L . , Orlando, P., Bisogno, T., Laezza, C , Bifulco, M . and Di Marzo, V. (2000) Suppression of nerve growth factor Trk receptors and prolactin receptors by endocannabinoids leads to inhibition of human breast and prostate cancer cell proliferation. Endocrinology, 141, 118-26.  178.  Hisaoka, M . , Sheng, W.Q., Tanaka, A. and Hashimoto, H . (2002) Gene expression of TrkC (NTRK3) in human soft tissue tumours. J Pathol, 197, 661-7.  179.  Brodeur, G . M . (1993) T R K - A Expression in Neuroblastomas: A New Prognostic Marker With Biological and Clinical Significance. J Natl Cancer Inst, 85, 344-345.  180.  Segal, R.A., Goumnerova, L . C , Kwon, Y . K . , Stiles, C D . and Pomeroy, S.L. (1994) Expression of the neurotrophin receptor TrkC is linked to a favorable outcome in medulloblastoma. Proc Natl Acad Sci USA, 91,12867-71.  173  181.  Fisher, C. (1996) Fibromatosis and fibrosarcoma in infancy and childhood. Eur J Cancer, 32A, 2094-100.  182.  Enzinger, F . M . and Weiss, S.W. (1995) Soft Tissue Tumors. C V . Mosby, St. Louis.  183.  Knezevich, S.R., Garnett, M.J., Pysher, T.J., Beckwith, J.B., Grundy, P.E. and Sorensen, P.H. (1998) ETV6-NTRK3 gene fusions and trisomy 11 establish a histogenetic link between mesoblastic nephroma and congenital fibrosarcoma. Cancer Res, 58, 5046-8.  184.  Rubin, B.P., Chen, C.J., Morgan, T.W., Xiao, S., Grier, H.E., Kozakewich, H.P., PerezAtayde, A.R. and Fletcher, J.A. (1998) Congenital mesoblastic nephroma t(12; 15) is associated with ETV6-NTRK3 gene fusion: cytogenetic and molecular relationship to congenital (infantile) fibrosarcoma. Am J Pathol, 153, 1451-8.  185.  Argani, P., Fritsch, M . , Kadkol, S.S., Schuster, A., Beckwith, J.B. and Perlman, E.J. (2000) Detection of the ETV6-NTRK3 chimeric R N A of infantile fibrosarcoma/cellular congenital mesoblastic nephroma in paraffin- embedded tissue: application to challenging pediatric renal stromal tumors. Mod Pathol, 13, 29-36.  186.  O'Malley, D.P., Mierau, G.W., Beckwith, J.B. and Weeks, D.A. (1996) Ultrastructure of cellular congenital mesoblastic nephroma. Ultrastruct Pathol, 20, 417-27.  187.  Argani, P. and Sorensen, P.H.B. (2004) Congenital mesoblastic nephroma. In Eble, J.N., Sauter, G., Epstein, J.I. and Sesterhenn, L A . (eds.), Tumours of the Urinary System and Male Genital Organs. IARC Press, Lyon.  188.  Baserga, R. (2000) The contradictions of the insulin-like growth factor 1 receptor. Oncogene, 19, 5574-81.  189.  Oberman, H.A. (1980) Secretory carcinoma of the breast in adults. Am J Surg Pathol, 4, 465-70.  190.  Page, D . L . , Anderson, T.J. and Sakamoto, G. (1987) Infiltrating carcinoma: major histologic types. In Page, D.L. and Anderson, T.J. (eds.), Diagnostic Histopathology of the Breast. Churchill Livingstone, Edinburgh.  191.  de Bree, E . , Askoxylakis, J., Giannikaki, E., Chroniaris, N., Sanidas, E . and Tsiftsis, D.D. (2002) Secretory carcinoma of the male breast. Ann Surg Oncol, 9, 663-7.  192.  Herz, H . , Cooke, B. and Goldstein, D. (2000) Metastatic secretory breast cancer. Nonresponsiveness to chemotherapy: case report and review of the literature. Ann Oncol, 11, 1343-7.  193.  Tavassoli, F.A. and Norris, H.J. (1980) Secretory carcinoma of the breast. Cancer, 45, 2404-13.  194.  Rosen, P.P. and Cranor, M.L. (1991) Secretory carcinoma of the breast. Arch Pathol Lab Med, 115, 141-4.  174  195.  Maitra, A., Tavassoli, F.A., Albores-Saavedra, J., Behrens, C , Wistuba, II, Bryant, D., Weinberg, A.G., Rogers, B.B., Saboorian, M . H . and Gazdar, A . F . (1999) Molecular abnormalities associated with secretory carcinomas of the breast. Hum Pathol, 30, 143540.  196.  Eguchi, M . , Eguchi-Ishimae, M . , Tojo, A., Morishita, K., Suzuki, K., Sato, Y., Kudoh, S., Tanaka, K., Setoyama, M . , Nagamura, F. et al. (1999) Fusion of E T V 6 to neurotrophin-3 receptor T R K C in acute myeloid leukemia with t(12;15)(pl3;q25). Blood, 93, 1355-63.  197.  Slupianek, A., Hoser, G., Majsterek, I., Bronisz, A., Malecki, M . , Blasiak, J., Fishel, R. and Skorski, T. (2002) Fusion tyrosine kinases induce drug resistance by stimulation of homology-dependent recombination repair, prolongation of G(2)/M phase, and protection from apoptosis. Mol Cell Biol, 22, 4189-201.  198.  Alessandri, A.J., Knezevich, S.R., Mathers, J.A., Schultz, K.R. and Sorensen, P.H. (2001) Absence of t(12;15) associated ETV6-NTRK3 fusion transcripts in pediatric acute leukemias. Med Pediatr Oncol, 37, 415-6.  199.  Lamant, L., Dastugue, N., Pulford, K., Delsol, G. and Mariame, B. (1999) A new fusion gene T P M 3 - A L K in anaplastic large cell lymphoma created by a (I;2)(q25;p23) translocation. Blood, 93, 3088-95.  200.  Lawrence, B., Perez-Atayde, A., Hibbard, M.K., Rubin, B.P., Dal Cin, P., Pinkus, J.L., Pinkus, G.S., Xiao, S., Yi, E.S., Fletcher, C D . et al. (2000) T P M 3 - A L K and T P M 4 - A L K oncogenes in inflammatory myofibroblastic tumors [see comments]. Am J Pathol, 157, 377-84.  201.  Touriol, C , Greenland, C , Lamant, L., Pulford, K., Bernard, F., Rousset, T., Mason, D.Y. and Delsol, G. (2000) Further demonstration of the diversity of chromosomal changes involving 2p23 in ALK-positive lymphoma: 2 cases expressing A L K kinase fused to C L T C L (clathrin chain polypeptide-like). Blood, 95, 3204-7.  202.  Bridge, J.A., Kanamori, M . , Ma, Z., Pickering, D., Hill, D.A., Lydiatt, W., Lui, M.Y., Colleoni, G.W., Antonescu, C.R., Ladanyi, M . et al. (2001) Fusion of the A L K gene to the clathrin heavy chain gene, C L T C , in inflammatory myofibroblastic tumor. Am J Pathol, 159,411-5.  203.  Prasad, D.D.K., Ouchida, M . , Lee, L . , Rao, V . N . and Reddy, E.S.P. (1994) TLS/FUS fusion domain of TLS/FUS-erg chimeric protein resulting from the t(16;21) chromosomal translocation in human myeloid leukemia functions as a transcriptional activation domain. Oncogene, 9, 3717-3729.  204.  Shing, D . C , McMullan, D.J., Roberts, P., Smith, K., Chin, S.F., Nicholson, J., Tillman, R.M., Ramani, P., Cullinane, C. and Coleman, N. (2003) F U S / E R G gene fusions in Ewing's tumors. Cancer Res, 63,4568-76.  205.  Ladanyi, M . , Lui, M.Y., Antonescu, C.R., Krause-Boehm, A., Meindl, A., Argani, P., Healey, J.H., Ueda, T., Yoshikawa, H . , Meloni-Ehrig, A. et al. (2001) The  175  der(17)t(X;17)(pll;q25) of human alveolar soft part sarcoma fuses the transcription factor gene to ASPL, a novel gene at 17q25. Oncogene, 20, 48-57.  TFE3  206.  Argani, P., Antonescu, C.R., Illei, P.B., Lui, M.Y., Timmons, C.F., Newbury, R., Reuter, V . E . , Garvin, A.J., Perez-Atayde, A.R., Fletcher, J.A. et al. (2001) Primary renal neoplasms with the ASPL-TFE3 gene fusion of alveolar soft part sarcoma: a distinctive tumor entity previously included among renal cell carcinomas of children and adolescents. AmJ Pathol, 159, 179-92.  207.  McWhirter, J.R., Galasso, D . L . and Wang, J.Y. (1993) A coiled-coil oligomerization domain of Bcr is essential for the transforming function of Bcr-Abl oncoproteins. Mol Cell Biol, 13, 7587-95.  208.  Carroll, M . , Tomasson, M.H., Barker, G.F., Golub, T.R. and Gilliland, D.G. (1996) The TEL/platelet-derived growth factor beta receptor (PDGF beta R) fusion in chronic myelomonocytic leukemia is a transforming protein that self-associates and activates PDGF beta R kinase-dependent signaling pathways. Proc Natl Acad Sci USA, 93, 14845-50.  209.  Bai, R.Y., Dieter, P., Peschel, C , Morris, S.W. and Duyster, J. (1998) Nucleophosminanaplastic lymphoma kinase of large-cell anaplastic lymphoma is a constitutively active tyrosine kinase that utilizes phospholipase C-gamma to mediate its mitogenicity. Mol Cell Biol, 18, 6951-61.  210.  Tognon, C , Garnett, M . , Kenward, E., Kay, R., Morrison, K. and Sorensen, P.H. (2001) The chimeric protein tyrosine kinase ETV6-NTRK3 requires both Ras- Erkl/2 and PI3kinase-Akt signaling for fibroblast transformation. Cancer Res, 61, 8909-16.  211.  Baserga, R. (1995) The insulin-like growth factor I receptor: a key to tumor growth? Cancer Res, 55, 249-52.  212.  Baserga, R., Peruzzi, F. and Reiss, K. (2003) The IGF-1 receptor in cancer biology. Int J Cancer, 107, 873-7.  213.  Baserga, R. (1999) The IGF-I receptor in cancer research. Exp Cell Res, 253,1-6.  214.  Toretsky, J.A. and Helman, L.J. (1996) Involvement of IGF-II in human cancer. / Endocrinol, 149, 367-72.  215.  Morrison, K.B., Tognon, C.E., Garnett, M.J., Deal, C. and Sorensen, P.H. (2002) ETV6NTRK3 transformation requires insulin-like growth factor 1 receptor signaling and is associated with constitutive IRS-1 tyrosine phosphorylation. Oncogene, 21, 5684-95.  216.  Toretsky, J.A., Kalebic, T., Blakesley, V . , LeRoith, D. and Helman, L.J. (1997) The insulin-like growth factor-I receptor is required for EWS/FLI-1 transformation of fibroblasts. J Biol Chem, 272, 30822-7.  176  217.  Wang, W., Kumar, P., Epstein, J., Helman, L., Moore, J.V. and Kumar, S. (1998) Insulinlike growth factor II and PAX3-FKHR cooperate in the oncogenesis of rhabdomyosarcoma. Cancer Res, 58,4426-33.  218.  Massague, J. (1990) The transforming growth factor-beta family. Annu Rev Cell Biol, 6, 597-641.  219.  Roberts, A . B . , and Sporn, M . B. (1990) Peptide Growth Factors and Their Receptors. Springer-Verlag, Berlin.  220.  Massague, J. (1998) TGF-beta signal transduction. Annu Rev Biochem, 67, 753-91.  221.  Zhu, X., Scharf, E. and Assoian, R.K. (2000) Induction of anchorage-independent growth by transforming growth factor- beta linked to anchorage-independent expression of cyclin Dl.JBiol Chem, 275, 6703-6.  222.  Blobe, G.C., Schiemann, W.P. and Lodish, H.F. (2000) Role of transforming growth factor beta in human disease. ./V Engl J Med, 342, 1350-8.  223.  Datto, M . B . , L i , Y . , Panus, J.F., Howe, D.J., Xiong, Y. and Wang, X . F . (1995) Transforming growth factor beta induces the cyclin-dependent kinase inhibitor p21 through a p5 3-independent mechanism. Proc Natl Acad Sci USA, 92, 5545-9.  224.  Maehara, Y., Kakeji, Y., Kabashima, A., Emi, Y., Watanabe, A., Akazawa, K., Baba, H., Kohnoe, S. and Sugimachi, K. (1999) Role of transforming growth factor-beta 1 in invasion and metastasis in gastric carcinoma. J Clin Oncol, 17, 607-14.  225.  Dickson, M.C., Martin, J.S., Cousins, F . M . , Kulkarni, A . B . , Karlsson, S. and Akhurst, R J . (1995) Defective haematopoiesis and vasculogenesis in transforming growth factorbeta 1 knock out mice. Development, 121, 1845-54.  226.  Norgaard, P., Hougaard, S., Poulsen, H.S. and Spang-Thomsen, M . (1995) Transforming growth factor beta and cancer. Cancer Treat Rev, 21, 367-403.  227.  Hahm, K.B., Cho, K., Lee, C , lm, Y . H . , Chang, J., Choi, S.G., Sorensen, P.H., Thiele, C.J. and Kim, S.J. (1999) Repression of the gene encoding the TGF-beta type II receptor is a major target of the EWS-FLI1 oncoprotein [published erratum appears in Nat Genet 1999 Dec;23(4):481]. Nat Genet, 23, 222-7.  228.  Im, Y . H . , Kim, H.T., Lee, C , Poulin, D., Welford, S., Sorensen, P.H., Denny, C T . and Kim, S.J. (2000) EWS-FLI1, EWS-ERG, and EWS-ETV1 oncoproteins of Ewing tumor family all suppress transcription of transforming growth factor beta type II receptor gene. Cancer Res, 60, 1536-40.  229.  Kim, S.J. (2004) Role of TGFbeta family in EN-induced tumourigenesis. In Lannon, C.L., (ed.).  177  230.  Park, B.J., Park, J.L, Byun, D.S., Park, J.H. and Chi, S.G. (2000) Mitogenic conversion of transforming growth factor-betal effect by oncogenic Ha-Ras-induced activation of the mitogen-activated protein kinase signaling pathway in human prostate cancer. Cancer Res, 60, 3031-8.  231.  Tognon, C.E., Mackereth, C D . , Somasiri, A . M . , Mcintosh, L.P. and Sorensen, P.H.B. (2004) Mutations in the S A M domain of the ETV6-NTRK3 chimeric tyrosine kinase block polymerization and transformation activity. Mol Cell Biol, 24, 4636-50.  232.  Kim, C.A., Phillips, M.L., Kim, W., Gingery, M . , Tran, H.H., Robinson, M.A., Faham, S. and Bowie, J.U. (2001) Polymerization of the S A M domain of T E L in leukemogenesis and transcriptional repression. Embo J, 20, 4173-82.  233.  Tran, H.H., Kim, C.A., Faham, S., Siddall, M . C and Bowie, J.U. (2002) Native interface of the S A M domain polymer of T E L . BMC Struct Biol, 2, 5.  234.  Freshney, R.I. (1994) Culture  of Animal  Cells: A Manual  of Basic  Technique.  Wiley-Liss,  New York. 235.  Grez, M . , Akgun, E., Hilberg, F. and Ostertag, W. (1990) Embryonic stem cell virus, a recombinant murine retrovirus with expression in embryonic stem cells. Proc Natl Acad Sci US A , 87, 9202-6.  236.  Miller, A.D. and Rosman, G.J. (1989) Improved retroviral vectors for gene transfer and expression. Biotechniques, 7, 980-2, 984-6, 989-90.  237.  Mann, R., Mulligan, R.C. and Baltimore, D. (1983) Construction of a retrovirus packaging mutant and its use to produce helper-free defective retrovirus. Cell, 33,153-9.  238.  Morgenstern, J.P. and Land, H . (1990) Advanced mammalian gene transfer: high titre retroviral vectors with multiple drug selection markers and a complementary helper-free packaging cell line. Nucleic Acids Res, 18, 3587-96.  239.  Miller, A . D . and Buttimore, C. (1986) Redesign of retrovirus packaging cell lines to avoid recombination leading to helper virus production. Mol Cell Biol, 6, 2895-902.  240.  Pear, W.S., Nolan, G.P., Scott, M . L . and Baltimore, D. (1993) Production of high-titer helper-free retroviruses by transient transfection. Proc Natl Acad Sci USA, 90, 8392-6.  241.  Ponten, J. (1971) Spontaneous and virus induced transformation in cell culture. Monogr, 8, 1-253.  242.  Schwede, T., Kopp, J., Guex, N. and Peitsch, M . C . (2003) SWISS-MODEL: An automated protein homology-modeling server. Nucleic Acids Res, 31, 3381-5.  243.  Guex, N. and Peitsch, M . C . (1997) SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis, 18, 2714-23.  Virol  178  244.  Hubbard, S.R. (1997) Crystal structure of the activated insulin receptor tyrosine kinase in complex with peptide substrate and ATP analog. Embo J, 16, 5572-81.  245.  Parang, K., Till, J.H., Ablooglu, A J . , Kohanski, R.A., Hubbard, S.R. and Cole, P.A. (2001) Mechanism-based design of a protein kinase inhibitor. Nat Struct Biol, 8, 37-41.  246.  Hubbard, S.R., Wei, L., Ellis, L . and Hendrickson, W.A. (1994) Crystal structure of the tyrosine kinase domain of the human insulin receptor. Nature, 372, 746-54.  247.  Till, J.H., Becerra, M . , Watty, A., Lu, Y., Ma, Y., Neubert, T . A . , Burden, S J . and Hubbard, S.R. (2002) Crystal structure of the MuSK tyrosine kinase: insights into receptor autoregulation. Structure (Camb), 10, 1187-96.  248.  Hu, J., Liu, J., Ghirlando, R., Saltiel, A.R. and Hubbard, S.R. (2003) Structural basis for recruitment of the adaptor protein APS to the activated insulin receptor. Mol Cell, 12, 1379-89.  249.  Bock, M . , Bishop, K.N., Towers, G. and Stoye, J.P. (2000) Use of a transient assay for studying the genetic determinants of F v l restriction. J Virol, 74, 7422-30.  250.  Niwa, H., Yamamura, K. and Miyazaki, J. (1991) Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene, 108, 193-9.  251.  Ikawa, M . , Yamada, S., Nakanishi, T. and Okabe, M . (1999) Green fluorescent protein (GFP) as a vital marker in mammals. Curr Top Dev Biol, 44, 1-20.  252.  Chomczynski, P. and Sacchi, N. (1987) Single-step method of R N A isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162, 156-159.  253.  Melton, D.W., Konecki, D.S., Brennand, J. and Caskey, C T . (1984) Structure, expression, and mutation of the hypoxanthine phosphoribosyltransferase gene. Proc Natl Acad Sci USA, 81, 2147-51.  254.  Sambrook, J., Fritch, E.F. and Maniatis, T. (1989) Molecular cloning: a laboratory manual, second ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.  255.  Mombaerts, P., Iacomini, J., Johnson, R.S., Herrup, K., Tonegawa, S. and Papaioannou, V . E . (1992) RAG-1-deficient mice have no mature B and T lymphocytes. Cell, 68, 86977.  256.  Podsypanina, K., Ellenson, L.H., Nemes, A., Gu, J., Tamura, M . , Yamada, K . M . , Cordon-Cardo, C , Catoretti, G., Fisher, P.E. and Parsons, R. (1999) Mutation of Pten/Mmacl in mice causes neoplasia in multiple organ systems. Proc Natl Acad Sci U S A, 96, 1563-8.  257.  Podsypanina, K., Lee, R.T., Politis, C , Hennessy, I., Crane, A., Puc, J., Neshat, M . , Wang, H . , Yang, L., Gibbons, J. et al. (2001) An inhibitor of mTOR reduces neoplasia  179  and normalizes p70/S6 kinase activity in Pten+/- mice. Proc Natl Acad 10320-5.  Sci USA,  98,  258.  White, M.F. (1997) The insulin signalling system and the IRS proteins. Diabetologia, Suppl 2, S2-17.  40  259.  Yenush, L . , Makati, K.J., Smith-Hall, J., Ishibashi, O., Myers, M . G . , Jr. and White, M.F. (1996) The pleckstrin homology domain is the principal link between the insulin receptor and IRS-1. J Biol Chem, 271, 24300-6.  260.  Yan, K.S., Kuti, M . and Zhou, M . M . (2002) PTB or not PTB - that is the question. Lett, 513, 67-70.  261.  Gustafson, T . A . , He, W., Craparo, A., Schaub, C D . and O'Neill, T.J. (1995) Phosphotyrosine-dependent interaction of SHC and insulin receptor substrate 1 with the NPEY motif of the insulin receptor via a novel non- SH2 domain. Mol Cell Biol, 15, 2500-8.  262.  Spencer, D . M . , Wandless, T.J., Schreiber, S.L. and Crabtree, G.R. (1993) Controlling signal transduction with synthetic ligands [see comments]. Science, 262, 1019-24.  263.  FEBS  Myers, M . G . , Jr. and White, M.F. (1996) Insulin signal transduction and the IRS proteins. Annu Rev Pharmacol  Toxicol,  36, 615-58.  264.  Baker, J., Liu, J.P., Robertson, E.J. and Efstratiadis, A . (1993) Role of insulin-like growth factors in embryonic and postnatal growth. Cell, 75, 73-82.  265.  Liu, J.P., Baker, J., Perkins, A.S., Robertson, E.J. and Efstratiadis, A. (1993) Mice carrying null mutations of the genes encoding insulin-like growth factor I (Igf-1) and type 1 IGF receptor (Igflr). Cell, 75, 59-72.  266.  Naldini, L., Stacchini, A., Cirillo, D . M . , Aglietta, M . , Gavosto, F. and Comoglio, P.M. (1986) Phosphotyrosine antibodies identify the p210c-abl tyrosine kinase and proteins phosphorylated on tyrosine in human chronic myelogenous leukemia cells. Mol Cell Biol, 6,1803-11.  267.  Reinach, F . C and MacLeod, A.R. (1986) Tissue-specific expression of the human tropomyosin gene involved in the generation of the trk oncogene. Nature, 322, 648-50.  268.  Miranda, C , Greco, A., Miele, C , Pierotti, M . A . and Van Obberghen, E . (2001) IRS-1 and IRS-2 are recruited by TrkA receptor and oncogenic TRK-T1. J Cell Physiol, 186, 35-46.  269.  Giovannone, B., Scaldaferri, M . L . , Federici, M . , Porzio, O., Lauro, D., Fusco, A., Sbraccia, P., Borboni, P., Lauro, R. and Sesti, G. (2000) Insulin receptor substrate (IRS) transduction system: distinct and overlapping signaling potential. Diabetes Metab Res Rev, 16, 434-41.  180  270.  Tanaka, S., Ito, T. and Wands, J.R. (1996) Neoplastic transformation induced by insulin receptor substrate-1 overexpression requires an interaction with both Grb2 and Syp signaling molecules. J Biol Chem, 271, 14610-6.  271.  Ito, T., Sasaki, Y . and Wands, J.R. (1996) Overexpression of human insulin receptor substrate 1 induces cellular transformation with activation of mitogen-activated protein kinases. Mol Cell Biol, 16, 943-51.  272.  Tanaka, S. and Wands, J.R. (1996) Insulin receptor substrate 1 overexpression in human hepatocellular carcinoma cells prevents transforming growth factor betal-induced apoptosis. Cancer Res, 56, 3391-4.  273.  Reiss, K., Wang, J.Y., Romano, G., Furnari, F.B., Cavenee, W.K., Morrione, A., Tu, X. and Baserga, R. (2000) IGF-I receptor signaling in a prostatic cancer cell line with a P T E N mutation. Oncogene, 19, 2687-94.  274.  Chang, Q., Li, Y., White, M.F., Fletcher, J.A. and Xiao, S. (2002) Constitutive activation of insulin receptor substrate 1 is a frequent event in human tumors: therapeutic implications. Cancer Res, 62, 6035-8.  275.  Traina, F., Carvalheira, J.B., Saad, M.J., Costa, F.F. and Saad, S.T. (2003) B C R - A B L binds to IRS-1 and IRS-1 phosphorylation is inhibited by imatinib in K562 cells. FEBS Lett, 535, 17-22.  276.  Kavanaugh, W . M . , Turck, C.W. and Williams, L.T. (1995) PTB domain binding to signaling proteins through a sequence motif containing phosphotyrosine. Science, 268, 1177-9.  277.  He, W., O'Neill, T.J. and Gustafson, T.A. (1995) Distinct modes of interaction of SHC and insulin receptor substrate-1 with the insulin receptor N P E Y region via non-SH2 domains. J Biol Chem, 270, 23258-62.  278.  Songyang, Z., Margolis, B., Chaudhuri, M . , Shoelson, S.E. and Cantley, L . C . (1995) The phosphotyrosine interaction domain of SHC recognizes tyrosine-phosphorylated N P X Y motif. JBiol Chem, 270, 14863-6.  279.  Weng, L.P., Smith, W.M., Brown, J.L. and Eng, C. (2001) P T E N inhibits insulinstimulated M E K / M A P K activation and cell growth by blocking IRS-1 phosphorylation and IRS-l/Grb-2/Sos complex formation in a breast cancer model. Hum Mol Genet, 10, 605-16.  280.  Lassak, A., Del Valle, L., Peruzzi, F., Wang, J.Y., Enam, S., Croul, S., Khalili, K. and Reiss, K. (2002) Insulin receptor substrate 1 translocation to the nucleus by the human JC virus T-antigen. JBiol Chem, 277, 17231-8.  281.  Till, J.H., Ablooglu, A.J., Frankel, M . , Bishop, S.M., Kohanski, R.A. and Hubbard, S.R. (2001) Crystallographic and solution studies of an activation loop mutant of the insulin receptor tyrosine kinase: insights into kinase mechanism. J Biol Chem, 276, 10049-55.  181  282.  Yarden, Y . and Ullrich, A. (1988) Growth factor receptor tyrosine kinases. Annu Rev Biochem, 57, 443-78.  283.  Lannon, C.L., Martin, M.J., Tognon, C.E., Jin, W., Kim, S.J. and Sorensen, P.H. (2004) A highly conserved NTRK3 C-terminal sequence in the ETV6-NTRK3 oncoprotein binds the phosphotyrosine binding domain of insulin receptor substrate-1: an essential interaction for transformation. J Biol Chem, 279, 6225-34.  284.  Wolf, G., Trub, T., Ottinger, E . , Groninga, L . , Lynch, A., White, M.F., Miyazaki, M . , Lee, J. and Shoelson, S.E. (1995) PTB domains of IRS-1 and She have distinct but overlapping binding specificities. J Biol Chem, 270, 27407-10.  285.  Farooq, A., Plotnikova, O., Zeng, L . and Zhou, M . M . (1999) Phosphotyrosine binding domains of She and insulin receptor substrate 1 recognize the NPXpY motif in a thermodynamic ally distinct manner. J Biol Chem, 274, 6114-21.  286.  DeMali, K.A., Whiteford, C C , Ulug, E.T. and Kazlauskas, A. (1997) Platelet-derived growth factor-dependent cellular transformation requires either phospholipase Cgamma or phosphatidylinositol 3 kinase. J Biol Chem, 272, 9011-8.  287.  Tallquist, M.D., Klinghoffer, R.A., Heuchel, R., Mueting-Nelsen, P.F., Corrin, P.D., Heldin, C.H., Johnson, R.J. and Soriano, P. (2000) Retention of PDGFR-beta function in mice in the absence of phosphatidylinositol 3'-kinase and phospholipase Cgamma signaling pathways. Genes Dev, 14, 3179-90.  288.  Valius, M . , Secrist, J.P. and Kazlauskas, A. (1995) The GTPase-activating protein of Ras suppresses platelet-derived growth factor beta receptor signaling by silencing phospholipase C-gamma 1. Mol Cell Biol, 15, 3058-71.  289.  Assefa, Z., Valius, M . , Vantus, T., Agostinis, P., Merlevede, W. and Vandenheede, J.R. (1999) JNK/SAPK activation by platelet-derived growth factor in A431 cells requires both the phospholipase C-gamma and the phosphatidylinositol 3-kinase signaling pathways of the receptor. Biochem Biophys Res Commun, 261, 641-5.  290.  Lawlor, E.R., Scheel, C , Irving, J. and Sorensen, P.H. (2002) Anchorage-independent multi-cellular spheroids as an in vitro model of growth signaling in Ewing tumors. Oncogene, 21, 307-18.  291.  Santini, M . T . and Rainaldi, G. (1999) Three-dimensional spheroid model in tumor biology. Pathobiology, 67, 148-57.  292.  Douma, S., Van Laar, T., Zevenhoven, J., Meuwissen, R., Van Garderen, E . and Peeper, D.S. (2004) Suppression of anoikis and induction of metastasis by the neurotrophic receptor TrkB. Nature, 430, 1034-9.  293.  Lagutina, I., Conway, S.J., Sublett, J. and Grosveld, G . C (2002) Pax3-FKHR knock-in mice show developmental aberrations but do not develop tumors. Mol Cell Biol, 22, 7204-16.  182  294.  Collins, E.C., Pannell, R., Simpson, E . M . , Forster, A. and Rabbitts, T.H. (2000) Interchromosomal recombination of Mil and Af9 genes mediated by cre- loxP in mouse development. EMBO Rep, 1, 127-32.  295.  Huettner, C.S., Zhang, P., Van Etten, R.A. and Tenen, D.G. (2000) Reversibility of acute B-cell leukaemia induced by BCR-ABL1. Nat Genet, 24, 57-60.  296.  Buchholz, F., Refaeli, Y., Trumpp, A. and Bishop, J.M. (2000) Inducible chromosomal translocation of AML1 and E T O genes through Cre/loxP-mediated recombination in the mouse. EMBO Rep, 1, 133-9.  297.  Capecchi, M.R. (1989) Altering the genome by homologous recombination. Science, 244, 1288-92.  298.  Castilla, L . H . , Wijmenga, C , Wang, Q., Stacy, T., Speck, N.A., Eckhaus, M . , MarinPadilla, M . , Collins, F.S., Wynshaw-Boris, A. and Liu, P.P. (1996) Failure of embryonic hematopoiesis and lethal hemorrhages in mouse embryos heterozygous for a knocked-in leukemia gene CBFB-MYH11. Cell, 87, 687-96.  299.  Castellanos, A., Pintado, B., Weruaga, E . , Arevalo, R., Lopez, A., Orfao, A. and Sanchez-Garcia, I. (1997) A BCR-ABL(pl90) fusion gene made by homologous recombination causes B- cell acute lymphoblastic leukemias in chimeric mice' with independence of the endogenous bcr product. Blood, 90, 2168-74.  300.  Okuda, T., Cai, Z., Yang, S., Lenny, N., Lyu, C J . , van Deursen, J.M., Harada, F£. and Downing, J.R. (1998) Expression of a knocked-in A M L 1 - E T O leukemia gene inhibits the establishment of normal definitive hematopoiesis and directly generates dysplastic hematopoietic progenitors. Blood, 91, 3134-43.  301.  Rolink, A., ten Boekel, E., Melchers, F., Fearon, D.T., Krop, I. and Andersson, J. (1996) A subpopulation of B220+ cells in murine bone marrow does not express CD 19 and contains natural killer cell progenitors. J Exp Med, 183, 187-94.  302.  Ballas, Z . K . and Rasmussen, W. (1993) Lymphokine-activated killer cells. VII. IL-4 induces an NK1.1+CD8 alpha+beta- TCR-alpha beta B220+ lymphokine-activated killer subset. J Immunol, 150, 17-30.  303.  Marinkovic, D., Marinkovic, T., Mahr, B., Hess, J. and Wirth, T. (2004) Reversible lymphomagenesis in conditionally c - M Y C expressing mice. Int J Cancer, 110, 336-42.  304.  Turner, S.D., Tooze, R., Maclennan, K. and Alexander, D.R. (2003) Vav-promoter regulated oncogenic fusion protein N P M - A L K in transgenic mice causes B-cell lymphomas with hyperactive Jun kinase. Oncogene, 22, 7750-61.  305.  Skoda, R.C., Tsai, S.F., Orkin, S.H. and Leder, P. (1995) Expression of c - M Y C under the control of GATA-1 regulatory sequences causes erythroleukemia in transgenic mice. J Exp Med, 181, 1603-13.  183  306.  Rifkin, I.R., Channavajhala, P.L., Kiefer, H.L., Carmack, A.J., Landesman-Bollag, E . , Beaudette, B.C., Jersky, B., Salant, D.J., Ju, S.T., Marshak-Rothstein, A . et al. (1998) Acceleration of lpr lymphoproliferative and autoimmune disease by transgenic protein kinase CK2 alpha. J Immunol, 161, 5164-70.  307.  Mombaerts, P. (1995) Lymphocyte development and function in T-cell receptor and RAG-1 mutant mice. Int Rev Immunol, 13, 43-63.  308.  JAX, C. (2000) Immunodeficient Model Selection: Choosing a nude, scid, or Ragl strain. No. 2 ed. The Jackson Laboratory.  309.  Li, J., Yen, C , Liaw, D., Podsypanina, K., Bose, S., Wang, S.I., Puc, J., Miliaresis, C , Rodgers, L., McCombie, R. et al. (1997) PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer. Science, 275, 1943-7.  310.  Steck, P.A., Pershouse, M.A., Jasser, S.A., Yung, W.K., Lin, H., Ligon, A . H . , Langford, L . A . , Baumgard, M . L . , Hattier, T., Davis, T. et al. (1997) Identification of a candidate tumour suppressor gene, M M A C 1 , at chromosome 10q23.3 that is mutated in multiple advanced cancers. Nat Genet, 15, 356-62.  311.  Li, D . M . and Sun, H . (1997) TEP1, encoded by a candidate tumor suppressor locus, is a novel protein tyrosine phosphatase regulated by transforming growth factor beta. Cancer Res, 57, 2124-9.  312.  Di Cristofano, A., Pesce, B., Cordon-Cardo, C. and Pandolfi, P.P. (1998) Pten is essential for embryonic development and tumour suppression. Nat Genet, 19, 348-55.  313.  Crackower, M.A., Oudit, G.Y., Kozieradzki, I., Sarao, R., Sun, H., Sasaki, T., Hirsch, E., Suzuki, A., Shioi, T., Irie-Sasaki, J. et al. (2002) Regulation of myocardial contractility and cell size by distinct PI3K-PTEN signaling pathways. Cell, 110, 737-49.  314.  Tuveson, D.A., Shaw, A.T., Willis, N.A., Silver, D.P., Jackson, E.L., Chang, S., Mercer, K . L . , Grochow, R., Hock, H . , Crowley, D. et al. (2004) Endogenous oncogenic K ras(G12D) stimulates proliferation and widespread neoplastic and developmental defects. Cancer  Cell, 5, 375-87.  21, 5504-14.  315.  Jackson-Grusby, L. (2002) Modeling cancer in mice. Oncogene,  316.  Doerfler, W. (1992) D N A methylation: eukaryotic defense against the transcription of foreign genes? Microb Pathog, 12, 1-8.  317.  Brooks, A.R., Harkins, R.N., Wang, P., Qian, H.S., Liu, P. and Rubanyi, G.M. (2004) Transcriptional silencing is associated with extensive methylation of the C M V promoter following adenoviral gene delivery to muscle. J Gene Med, 6, 395-404.  318.  Sutter, D. and Doerfler, W. (1980) Methylation of integrated adenovirus type 12 D N A sequences in transformed cells is inversely correlated with viral gene expression. Proc Natl Acad Sci USA, 77, 253-6.  184  319.  Martin-Zanca, D., Hughes, S.H. and Barbacid, M . (1986) A human oncogene formed by the fusion of truncated tropomyosin and protein tyrosine kinase sequences. Nature, 319, 743-8.  320.  Guerra, C., Mijimolle, N., Dhawahir, A., Dubus, P., Barradas, M . , Serrano, M . , Campuzano, V. and Barbacid, M . (2003) Tumor induction by an endogenous K-ras oncogene is highly dependent on cellular context. Cancer Cell, 4,111-20.  321.  Vogelstein, B. and Kinzler, K.W. (2004) Cancer genes and the pathways they control. Nat Med, 10,789-99.  322.  Nowell, P.C. (2002) Tumor progression: a brief historical perspective. Semin Cancer Biol, 12, 261-6.  323.  Greenwald, R.J., Tumang, J.R., Sinha, A., Currier, N., Cardiff, R.D., Rothstein, T.L., Faller, D . V . and Denis, G.V. (2004) E mu-RD2 transgenic mice develop B-cell lymphoma and leukemia. Blood, 103, 1475-84.  324.  Langdon, W.Y., Harris, A.W., Cory, S. and Adams, J.M. (1986) The c-myc oncogene perturbs B lymphocyte development in E-mu-myc transgenic mice. Cell, 47, 11-8.  325.  Adams, J.M., Harris, A.W., Pinkert, C.A., Corcoran, L . M . , Alexander, W.S., Cory, S., Palmiter, R.D. and Brinster, R.L. (1985) The c-myc oncogene driven by immunoglobulin enhancers induces lymphoid malignancy in transgenic mice. Nature, 318, 533-8.  326.  Lee, A . C , Fenster, B.E., Ito, H., Takeda, K., Bae, N.S., Hirai, T., Yu, Z.X., Ferrans, V.J., Howard, B . H . and Finkel, T. (1999) Ras proteins induce senescence by altering the intracellular levels of reactive oxygen species. J Biol Chem, 274, 7936-40.  327.  Busuttil, R.A., Rubio, M . , Dolle, M . E . , Campisi, J. and Vijg, J. (2003) Oxygen accelerates the accumulation of mutations during the senescence and immortalization of murine cells in culture. Aging Cell, 2, 287-94.  328.  Chen, Q.M., Bartholomew, J . C , Campisi, J., Acosta, M . , Reagan, J.D. and Ames, B.N. (1998) Molecular analysis of H202-induced senescent-like growth arrest in normal human fibroblasts: p53 and Rb control GI arrest but not cell replication. Biochem J, 332 (Pt 1), 43-50.  329.  Rompaey, L . V . , Potter, M . , Adams, C. and Grosveld, G. (2000) Tel induces a GI arrest and suppresses Ras-induced transformation. Oncogene, 19, 5244-50.  330.  Johnson, L., Mercer, K., Greenbaum, D., Bronson, R.T., Crowley, D., Tuveson, D.A. and Jacks, T. (2001) Somatic activation of the K-ras oncogene causes early onset lung cancer in mice. Nature, 410, 1111-6.  331.  Orsulic, S., L i , Y., Soslow, R.A., Vitale-Cross, L . A . , Gutkind, J.S. and Varmus, H . E . (2002) Induction of ovarian cancer by defined multiple genetic changes in a mouse model system. Cancer Cell, 1, 53-62.  185  332.  Aguirre, A.J., Bardeesy, N., Sinha, M . , Lopez, L . , Tuveson, D.A., Horner, J., Redston, M.S. and DePinho, R.A. (2003) Activated Kras and Ink4a/Arf deficiency cooperate to produce metastatic pancreatic ductal adenocarcinoma. Genes Dev, 17, 3112-26.  333.  Pollock, J.L., Westervelt, P., Kurichety, A.K., Pelicci, P.G., Grisolano, J.L. and Ley, T.J. (1999) A bcr-3 isoform of RARalpha-PML potentiates the development of P M L RARalpha-driven acute promyelocytic leukemia. Proc Natl Acad Sci USA, 96, 15103-8.  334.  White, M . F . (1998) The IRS-signalling system: a network of docking proteins that mediate insulin action. Mol Cell Biochem, 182, 3-11.  335.  Kaplan, D.R. and Stephens, R.M. (1994) Neurotrophin signal transduction by the Trk receptor. J Neurobiol, 25, 1404-17.  336.  Al-Lazikani, B., Jung, J., Xiang, Z. and Honig, B. (2001) Protein structure prediction. Curr Opin Chem Biol, 5, 51-6.  337.  Marti-Renom, M.A., Stuart, A.C., Fiser, A., Sanchez, R., Melo, F. and Sali, A. (2000) Comparative protein structure modeling of genes and genomes. Annu Rev Biophys Biomol  Struct, 29, 291-325.  338.  Kopp, J. and Schwede, T. (2004) The SWISS-MODEL Repository of annotated threedimensional protein structure homology models. Nucleic Acids Res, 32 Database issue, D230-4.  339.  Valenzuela, D . M . , Stitt, T.N., DiStefano, P.S., Rojas, E . , Mattsson, K., Compton, D.L., Nunez, L . , Park, J.S., Stark, J.L., Gies, D.R. et al. (1995) Receptor tyrosine kinase specific for the skeletal muscle lineage: expression in embryonic muscle, at the neuromuscular junction, and after injury. Neuron, 15, 573-84.  340.  Kopp, J. and Schwede, T. (2004) Automated protein structure homology modeling: a progress report. Pharmacogenomics, 5, 405-16.  341.  Baker, D. and Sali, A. (2001) Protein structure prediction and structural genomics. Science,  294, 93-6.  342.  Stephens, R.M., Loeb, D . M . , Copeland, T.D., Pawson, T., Greene, L . A . and Kaplan, D.R. (1994) Trk receptors use redundant signal transduction pathways involving SHC and PLC-gamma 1 to mediate NGF responses. Neuron, 12, 691-705.  343.  Li, C , Iosef, C , Jia, C.Y., Han, V . K . and Li, S.S. (2003) Dual functional roles for the X linked lymphoproliferative syndrome gene product SAP/SH2D1A in signaling through the signaling lymphocyte activation molecule (SLAM) family of immune receptors. J Biol Chem, 278, 3852-9.  344.  Rodriguez, M . , Li, S.S., Harper, J.W. and Songyang, Z. (2004) An oriented peptide array library (OPAL) strategy to study protein-protein interactions. J Biol Chem, 279, 8802-7.  186  345.  Haiti, F . U . and Hayer-Hartl, M . (2002) Molecular chaperones in the cytosol: from nascent chain to folded protein. Science, 295, 1852-8.  346.  Bonvini, P., Gastaldi, T., Falini, B. and Rosolen, A. (2002) Nucleophosmin-anaplastic lymphoma kinase (NPM-ALK), a novel Hsp90-client tyrosine kinase: down-regulation of N P M - A L K expression and tyrosine phosphorylation in ALK(+) CD30(+) lymphoma cells by the Hsp90 antagonist 17-allylamino,17-demethoxygeldanamycin. Cancer Res, 62, 1559-66.  347.  Ninane, J., Gosseye, S., Panteon, E . , Claus, D., Rombouts, J.J. and Cornu, G. (1986) Congenital fibrosarcoma. Preoperative chemotherapy and conservative surgery. Cancer, 58, 1400-6.  348.  Nollen, E . A . and Morimoto, R.I. (2002) Chaperoning signaling pathways: molecular chaperones as stress-sensing 'heat shock' proteins. J Cell Sci, 115, 2809-16.  349.  van den Heuvel, A.P., de Vries-Smits, A . M . , van Weeren, P.C., Dijkers, P.F., de Bruyn, K . M . , Riedl, J.A. and Burgering, B . M . (2002) Binding of protein kinase B to the plakin family member periplakin. J Cell Sci, 115, 3957-66.  350.  Kalinin, A . E . , Idler, W.W., Marekov, L.N., McPhie, P., Bowers, B., Steinert, P.M. and Steven, A . C . (2004) Co-assembly of envoplakin and periplakin into oligomers and Ca(2+)-dependent vesicle binding: implications for cornified cell envelope formation in stratified squamous epithelia. J Biol Chem, 279, 22773-80.  351.  Aho, S., L i , K., Ryoo, Y., McGee, C , Ishida-Yamamoto, A., Uitto, J. and Klement, J.F. (2004) Periplakin gene targeting reveals a constituent of the cornified cell envelope dispensable for normal mouse development. Mol Cell Biol, 24, 6410-8.  352.  Philipps, B., Hennecke, J. and Glockshuber, R. (2003) FRET-based in vivo screening for protein folding and increased protein stability. J Mol Biol, 327, 239-49.  353.  Riven, I., Kalmanzon, E . , Segev, L . and Reuveny, E . (2003) Conformational rearrangements associated with the gating of the G protein-coupled potassium channel revealed by FRET microscopy. Neuron, 38, 225-35.  354.  Miyawaki, A., Llopis, J., Heim, R., McCaffery, J.M., Adams, J.A., Ikura, M . and Tsien, R.Y. (1997) Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature, 388, 882-7.  355.  Day, R.N., Periasamy, A. and Schaufele, F. (2001) Fluorescence resonance energy transfer microscopy of localized protein interactions in the living cell nucleus. Methods, 25, 4-18.  356.  Million, R.P., Harakawa, N., Roumiantsev, S., Varticovski, L. and Van Etten, R.A. (2004) A direct binding site for Grb2 contributes to transformation and leukemogenesis by the Tel-Abl (ETV6-AM) tyrosine kinase. Mol Cell Biol, 24, 4685-95.  187  357.  Pendergast, A . M . , Quilliam, L.A., Cripe, L.D., Bassing, C.H., Dai, Z., L i , N., Batzer, A., Rabun, K . M . , Der, C J . , Schlessinger, J. et al. (1993) BCR-ABL-induced oncogenesis is mediated by direct interaction with the SH2 domain of the GRB-2 adaptor protein. Cell, 75,175-85.  358.  Ho, J.M., Nguyen, M.H., Dierov, J.K., Badger, K . M . , Beattie, B.K., Tartaro, P., Haq, R., Zanke, B.W., Carroll, M.P. and Barber, D.L. (2002) T E L - J A K 2 constitutively activates the extracellular signal-regulated kinase (ERK), stress-activated protein/Jun kinase (SAPK/JNK), and p38 signaling pathways. Blood, 100, 1438-48.  359.  Wood, L . D . , Irvin, B.J., Nucifora, G., Luce, K.S. and Hiebert, S.W. (2003) Small ubiquitin-like modifier conjugation regulates nuclear export of T E L , a putative tumor suppressor. Proc Natl Acad Sci USA, 100, 3257-62.  360.  Ouyang, T., Bai, R.Y., Bassermann, F., von Klitzing, C , Klumpen, S., Miething, C , Morris, S.W., Peschel, C. and Duyster, J. (2003) Identification and characterization of a nuclear interacting partner of anaplastic lymphoma kinase (NIPA). J Biol Chem, 278, 30028-36.  361.  Fujimoto, J., Shiota, M . , Iwahara, T., Seki, N., Satoh, H . , Mori, S. and Yamamoto, T. (1996) Characterization of the transforming activity of p80, a hyperphosphorylated protein in a Ki-1 lymphoma cell line with chromosomal translocation t(2;5). Proc Natl Acad Sci USA, 93,4181-6.  362.  Tu, X., Wu, A., Maiorana, A. and Baserga, R. (2003) Subcellular localization of IRS-1 in cell proliferation and differentiation. Horm Metab Res, 35, 734-9.  363.  Flordal Thelander, E . , Walsh, S.H., Thorselius, M . , Laurell, A., Landgren, O., Larsson, C , Rosenquist, R. and Lagercrantz, S. (2004) Mantle cell lymphomas with clonal immunoglobulin V(H)3-21 gene rearrangements exhibit fewer genomic imbalances than mantle cell lymphomas utilizing other immunoglobulin V(H) genes. Mod Pathol.  364.  Vizcarra, E . , Martinez-Climent, J.A., Benet, I., Marugan, I., Terol, M J . , Prosper, F., Marco, J., Sanchez, D., Ferrandez, A., Tormo, M . et al. (2001) Identification of two subgroups of mantle cell leukemia with distinct clinical and biological features. Hematol J, 2, 234-41.  365.  Zunino, A., Viaggi, S., Ottaggio, L . , Fronza, G., Schenone, A., Roncella, S. and Abbondandolo, A. (2000) Chromosomal aberrations evaluated by C G H , FISH and G T G banding in a case of AIDS-related Burkitt's lymphoma. Haematologica, 85, 250-5.  366.  Whitlock, C A . and Witte, O.N. (1982) Long-term culture of B lymphocytes and their precursors from murine bone marrow. Proc Natl Acad Sci USA, 79, 3608-12.  367.  Bafico, A. and Aaronson, S.A. (2003) Growth Factors and Signal Transduction in Cancer. In Holland-Frei Cancer Medicine. 6th ed. B C Decker Inc., Hamilton, pp. 53-71.  188  368.  Hamad, N.M., Elconin, J.H., Karnoub, A.E., Bai, W., Rich, J.N., Abraham, R.T., Der, C J . and Counter, C M . (2002) Distinct requirements for Ras oncogenesis in human versus mouse cells. Genes Dev, 16, 2045-57.  369.  Hingorani, S.R., Petricoin, E.F., Maitra, A., Rajapakse, V . , King, C , Jacobetz, M . A . , Ross, S., Conrads, T.P., Veenstra, T.D., Hitt, B.A. et al. (2003) Preinvasive and invasive ductal pancreatic cancer and its early detection in the mouse. Cancer Cell, 4, 437-50.  370.  Argani, P., Fritsch, M.K., Shuster, A . E . , Perlman, E J . and Coffin, C M . (2001) Reduced sensitivity of paraffin-based RT-PCR assays for ETV6-NTRK3 fusion transcripts in morphologically defined infantile fibrosarcoma. Am J Surg Pathol, 25, 1461-4.  371.  Kuhn, R., Schwenk, F., Aguet, M . and Rajewsky, K. (1995) Inducible gene targeting in mice. Science, 269, 1427-9.  372.  Sun, F.L., Dean, W.L., Kelsey, G., Allen, N.D. and Reik, W. (1997) Transactivation of Igf2 in a mouse model of Beckwith-Wiedemann syndrome. Nature, 389, 809-15.  373.  Eggenschwiler, J., Ludwig, T., Fisher, P., Leighton, P.A., Tilghman, S.M. and Efstratiadis, A. (1997) Mouse mutant embryos overexpressing IGF-II exhibit phenotypic features of the Beckwith-Wiedemann and Simpson-Golabi-Behmel syndromes. Genes Dev, 11,3128-42.  374.  Bol, D.K., Kiguchi, K., Gimenez-Conti, I., Rupp, T. and DiGiovanni, J. (1997) Overexpression of insulin-like growth factor-1 induces hyperplasia, dermal abnormalities, and spontaneous tumor formation in transgenic mice. Oncogene, 14,1725-34.  375.  Greaves, M.F. (2004) Childhood leumekia - backtracking its origins. 36th Congress of the International Society of Pediatric Oncology. Pediatric Blood and Cancer, Oslo, Norway, Vol. 43 (4), pp. 305 - 503.  376.  Bockamp, E.O., McLaughlin, F., Murrell, A . M . , Gottgens, B., Robb, L . , Begley, C . G . and Green, A.R. (1995) Lineage-restricted regulation of the murine SCL/TAL-1 promoter. Blood, 86, 1502-14.  377.  Brodeur, G.M. (2003) Neuroblastoma: biological insights into a clinical enigma. Nat Rev Cancer, 3, 203-16.  378.  Eggert, A., Grotzer, M.A., Ikegaki, N., Zhao, H . , Cnaan, A., Brodeur, G.M. and Evans, A . E . (2001) Expression of the neurotrophin receptor TrkB is associated with unfavorable outcome in Wilms' tumor. J Clin Oncol, 19, 689-96.  379.  Aoyama, M . , Asai, K., Shishikura, T., Kawamoto, T., Miyachi, T., Yokoi, T., Togari, H., Wada, Y., Kato, T. and Nakagawara, A. (2001) Human neuroblastomas with unfavorable biologies express high levels of brain-derived neurotrophic factor mRNA and a variety of its variants. Cancer Lett, 164, 51-60.  189  380.  Nakagawara, A., Azar, C.G., Scavarda, N J . and Brodeur, G . M . (1994) Expression and function of TRK-B and BDNF in human neuroblastomas. Mol Cell Biol, 14,759-767.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items