Corrected Thesis

Notch Signalling in CD34+ Cells in Chronic Myeloid Leukaemia A thesis submitted to the University of Manchester for the degree of PhD in the Faculty of Life Sciences

2008

Abdullah H. Al-Jedai

Contents Page CONTENTS…………………………………………………………………………2 LIST OF FIGURES………………………………………………………………….7 LIST OF ABBREVIATIONS……………………………………………………….10 ABSTRACT…………………………………………………………………………14 DECLARATION……………………………………………………………….......15 COPYRIGHT STATEMENT……………………………………………………….15 ACKNOWLEDGEMENTS…………………………………………………………16

Chapter one: Introduction………………………………………..17 1.1.2.1.1 1.1.2.1.3

1.1.1 Haemopoietic stem cells (HSCs)………………………………..18 1.1.2 Regulation of haemopoiesis……………………………………...22 1.1.2.1 The stem cell niche………………………………………………………....23 Cell-ECM interaction………………………………………………..23 1.1.2.1.2 Soluble factors in the niche ………………………………………….24 Cell-cell interactions……………………………………….. ……….27 1.1.2.2 Genetic control of haemopoiesis ………………………………………….28

1.2 Notch signalling pathway ………………………………………….31 1.2.1 Notch receptors ………………………………………………………………31 1.2.2 Notch ligands…………………………………………………………………32 1.2.3 Molecular mechanisms of Notch signalling …………………………………35 1.2.4 Modulators of Notch signalling………………………………………………38 Notch signalling in haemopoiesis……………………………………………39 1.2.5 1.2.5.1 Notch and haemopoietic stem cell (HSC) fate decisions………………..40 1.2.5.2 Notch signalling in myeloid development………………………………..42 1.2.5.3 Notch signalling in lymphoid cell development………………………44 1.2.6 Notch signalling and cancer…………………………………………………47 1.2.6.1 Notch signalling in leukaemia ………………………………………..48

1.3 Chronic myeloid leukaemia (CML)………………………………52 1.3.1 Molecular phenotype of BCR-ABL……………………………………..53 1.3.2 BCR-ABL oncogenic activities…………………………………………55 1.3.2.1 Altered adhesion………………………………………………….55 1.3.2.2 Inhibition of apoptosis……………………………………………56 1.3.2.3 Proliferative signals………………………………………………56 1.3.2.4 Role of CrKl in BCR-ABL signalling……………………………57 1.3.3 Leukaemic stem cells (LSC) in CML………………………………….57 1.3.4 Imatinib mesylate………………………………………………………60. 1.3.5 Experimental models of CML…………………………………………61 1.3.5.1 Cell lines……………………………………………………………61 1.3.5.2 Animal models……………………………………………………..61 2

1.4 Possible role for Notch in CML……………………………………62 1.5 Research aims and objectives………………………………………64

Chapter 2 Material and methods…………………………65 2.1 Cell Biology techniques…………………………………65 2.1.1 Cell lines……………………………………………………………………..65 2.1.1.1 K562 cell line…………………………………………………………..65 2.1.1.2 NALM-1 cell line………………………………………………………..65 2.1.1.3 ALL-SIL cell line………………………………………………………65 2.1.1.4 JURKAT cell line ……………………………………………………...66 2.1.1.5 Passage of cell lines……………………………………………………66 2.1.1.6 Viable Cell Count………………………………………………………66 2.1.1.7 Cryopreservation of Cell Lines…………………………………………66 2.1.2 Primary CML samples……………………………………………………….66 2.1.2.1 Thawing of cryopreserved CML cells…………………………………..67 2.1.2.2 Short term liquid culture of primary CML CD34+ cells………………..67 2.1.3 Retroviral transfection of K562 cells with Notch1ΔE……………………68 2.2 Flow cytometric techniques……………………………..68 2.2.1 Isolation of mononuclear cells (MNC)……………………………………68 2.2.2 Isolation of haemopoietic progenitor cell populations……………………68 2.2.3 Staining procedures for flow cytometric analysis………………………...69 2.2.3.1 FACS analysis of extra-cellular Notch1 on primary CML cells……69 2.2.3.2 FACS analysis of extra-cellular Notch1 on K562 cells…………….70 2.2.3.3 FACS analysis of intra-cellular Notch1 on K562 cells……………..70 2.2.3.4 The P-crkl assay…………………………………………………….71

2.3 Molecular biology techniques…………………………..73 2.3.1 RNA extraction…………………………………………………………..73 2.3.2 Construction of cDNA from low cell numbers by Poly-A PCR…………73 2.3.3 Construction of cDNA by from high cell numbers………………………78 2.3.4 Gene specific PCR……………………………………………………….78 2.3.4.1 Primers……………………………………………………………..78 2.3.4.2 Optimisation of Primer Sets………………………………………..79 2.3.4.3 PCR reaction ………………………………………………………79 2.3.4.4 Detection of PCR products…………………………………………79

2.3.5 Real time PCR…………………………………………81

3

2.3.5.1 Overview OF Real Time PCR……………………………………………..81 2.3.5.2 Real time PCR protocols…………………………………………………..82 2.3.5.2.1 Real time PCR using TaqMan®probes………………………………82 2.3.5.2.2 Real time PCR using SYBR® Green………………………………..82. 2.3.5.3 Data analysis…………………………………………………………..83 2.3.5.4 Validation of the 2 –ΔΔCT method………………………………………..83 2.3.6

Protein Analysis………………………………86

2.3.6.1. Protein extraction and determination of concentration…………………...86 2.3.6.2 SDS-PAGE and Western Blott……………………………………………87

2.4 Statistics ……………………………………… ………………………88

Chapter 3 Investigating Notch signalling in chronic myeloid leukaemia……………………………90 3.1 Introduction ……………………………………………………90 3.2 Results…………………………………………………………..91 3.2.1. Gene expression analysis………………………………………………….91 3.2.1.1 Expression pattern of Notch genes in CML……………………………91 3.2.1.2 Expression pattern of Notch target genes……………………………....93 3.2.2 Flow cytometric analysis of Notch1 in CML………………………….. ..99

3.3. Discussion……………………………………………………...108 3.3.1 Expression pattern of Notch genes in CM…………………………………108 3.3.2 Expression patterns of Notch target genes in CML……………………….110 3.3.3 Expression of Notch1 protein in CML……………………………………111

Chapter 4 Investigation of BCR-ABL and Notch cross-talk in cell line models 4.1 Introduction…………………………………………...115 4.2 Results…………………………………………………118 4.2.1 Validation of the P-crkl intracellular FACS assay in K562 cells…………118 4.2.2 The effect of cell passage number on the expression of P-crkl in K562 cells……………………………………………………….120 4

4.2.3 Assessment of P-crkl expression in leukaemic cell lines………………….124 4.2.4 Assessment of imatinib mesylate efficacy in K562 cells using the P-crkl assay………………………………………………………..126 4.2.5 Characterisation of Notch signalling in K562 cells…………………………130 4.2.6 Constitutive expression of Notch1 ΔE in K562 cells…………………….....133 4.2.7 The effect of Valproic acid on BCR-ABL and Notch signalling in K562 cells ………………………………………………………………….135 4.2.8 The effect of GSI in K562 cells……………………………………………..140 4.2.9 Cross-talk between Notch and BCR-ABL in K562 cells…………………….142 4.2.9.1 The effect of imatinib induced BCR-ABL inhibition on Notch signalling in K562 cells…………………………………………..142 4.2.9.2 The effect of Notch inhibition by GSI on BCR-ABL in K562 cells........142 4.2.10 ALL-SIL cell line as a model for ABL-Notch cross-talk…………………....146

4.3 Discussion………………………………………………….149 4.3.1 The FACS based P-crkl assay as a surrogate assay for ABL kinase activity…149 4.3.2 P-crkl expression in other leukaemic cell lines ………………………………151 4.3.3 Inhibition of p-crkl by imantinib mesylate in K562 cells……………………152 4.3.4 Notch signalling in K562 cells………………………………………………152 4.3.5 Cross-talk between Notch and BCR-ABL in K562 cells……………………154 4.3.6 Cross-talk between Notch and BCR-ABL in the ALL-SIL cell line model system…………………………………………………………156

Chapter 5 Cross-talk between Notch and BCR-ABL in primary CD34+ CML cells 5.1 Introduction………………………………………………………158 5.2: Results……………………………………………………………160 5.2.1 P-crkl phosphorylation can be detected in primary CD34+ CML cells by intracellular flow cytometry assay……………………………160 5.2.2 Imatinib mesylate (IM) inhibits BCR-ABL activity in chronic phase CML CD34+ cells……………………………………………162 5.2.3 Effect of matinib in CD34+ CML cells upregulates Hes1 Notch target gene expression…………………………………………162 5.2.4 Investigating the effect of Notch inhibition on BCR-ABL activity in CD34+ CML cells………………………………………………168. 5.2.4.1 GSI induced inhibition of Notch signalling in CD34+ CML cells……168 5.2.4.2 Non GSI responding CD34+ CML cells express high mRNA levels of Hes1 ………………………………………………………171 5.2.4.3 Gamma secretase inhibitor (GSI) increases the kinase 5

activity of BCR-ABL in CD34+ CML cells…………………………………171 5.2.4.4 Gamma secretase inhibitor (GSI) decreased the kinase activity of BCR-ABL in CD34+ CML cells from one CML patient…………172

5.3: Discussion…………………………………………………………179 5.3.1 BCR-ABL activity can be monitored in primary CD34+ CML cells by flow cytometry……………………………………………………..179 5.3.2 Imatinib mesylate inhibits BCR-ABL activity and up-regulates Notch activity in CD34+ chronic phase CML cells…………………………181 5.3.3 Notch inhibition enhances BCR-ABL kinase activity in CD34+ chronic CML cells……………………………………………….185

Chapter 6: Final discussion……………………188 References ……………………………………………………………196 Appendex ……………………………………………………………215

6

Final word count: 52367

List of figures Chapter 1 Fig.1.1 The hierarchy of haemopoiesis………………………………………..21 Fig. 1.2 Regulation of haemopoiesis…………………………………………..30 Fig.1.3 Structure of human Notch receptors …………………………………..33 Fig. 1.4 Structure of Notch ligands…………………………………………….34 Fig. 1.5 The CSL-dependent Notch signalling pathway………………………. 37 Fig. 1.6 Notch signalling during T and B cell development…………………..50 Fig. 1.7 The t(9;22)(q34;q11) reciprocal translocation…………………………54 Fig. 1.8.Signal transduction pathways associated with P210 BCR-ABL in CML…………………………………………………………………………….59

Chapter 2 Fig. 2.1. Outline of poly-A PCR technique……………………………………..77 Fig. 2.2. Real Time PCR…………………………………………………………85

Chapter 3 Fig. 3.1. Notch expression of receptor genes in CD34+ populations isolated from normal bone marrow (NBM) and CML samples……………………………94 Fig. 3.2. Real time PCR analysis of Notch1(N1) expression on CD34+ cell subsets from NBM and CML patients……………………………………………95 Fig. 3.3. Real time PCR analysis of Notch2 expression on CD34+ cell subsets from NBM and CML patients……………………………………………………96 Fig. 3.4. Expression of Notch target genes in CD34+ populations isolated from NBM and CML samples…………………………………………...97 Fig. 3.5. Real time PCR analysis of Hes1 expression on CD34+ cell subsets from NBM and CML patients……………………………………………………98 Fig. 3.6. Notch expression on CD34+ myeloid progenitors in CML…………..101 Fig. 3.7. Notch expression on CD34+ lymphoid progenitors in CML…………103 Fig. 3.8. CD34 gating strategy and the Notch expression in CD34+ CD38cell subset in CML………………………………………………………………104 Fig. 3.9. The problem of EA1 non-specific binding within the CD34+ Thy+ cell subset…………………………………………………………105 Fig. 3.10. The expression of Notch1 in the CD34+ Thy+ cell subset………………………………………………………………………..106

Chapter 4 Fig 4.1. Validation of P-crkl intracellular flow cytometry assay in K562 cells…121 Fig 4.2. Comparison of four commercial anti rabbit secondary antibodies

7

used in the P-crkl assay…………………………………………………………..122 Fig. 4.3. The effect of cell passage number on the expression of P-crkl in K562 cell line…………………………………………………………………123 Fig 4.4. Assessment of P-crkl expression in four leukaemic cell lines…………125 Fig. 4.5-a. Assessment of imatinib mesylate (IM) efficacy in K562 cells using a flow based P-crkl assay.............................................................................127 Fig. 4.5-b. Dose dependant effect of imatinib mesylate (IM) on the expression of P-crkl in k562 cells post 48h………………………………………128 Fig. 4.6. Effect of concentration of imatinib mesylates (IM) on P-crkl protein levels………………………………………………………………129 Fig. 4.7 Expression of Notch1 and Hes1 genes in K562 cell line………………...131 Fig. 4.8. FACS analysis of Notch1 expression in K562 cells…………………….132 Fig. 4.9. Constitutive expression of N1ΔE in K562 cells…………………………134 Fig. 4.10 Hes1 expression in K562 cells post valproic acid (VPA) treatment……136 Fig. 4.11. Effect of Valproic acid (VPA) on BCR-ABL activity in K562 cells…..138 Fig. 4.12. Effect of Valproic acid (VPA) on erythroid differentiation in K562 cells……………………………………………………………………..139 Fig. 4.13. Inhibition of Notch signalling by a gamma seretase inhibitor (GSI) in K562 cells………………………………………………………………141 Fig. 4.14. Expression of Hes1 in K562 cells post 48h treatment of imatinib mesylate (IM)…………………………………………………………144 Fig. 4.15. The effect of Notch inhibition on BCR-ABL activity in K562 cells……………………………………………………………………145 Fig. 4.16. Evaluation of the ALL-SIL cell line as a model for ABL-Notch cross-talk………………………………………………………….147 Fig. 4.17. Expression of Hes1 in ALL-SIL cells post 48h treatment of imatinib mesylate (IM)……………………………………………148

Chapter 5 Fig. 5.1. Application of P-CrKl assay to primary chronic myeloid leukaemia (CML) samples……………………………………………161 Fig. 5.2. Inhibition of BCR-ABL activity by imatinib mesylate (IM) in CD34+ cells isolated from CML patients……………………164 Fig 5.3. Evidence of resistance to imatinib mesylate (IM) in CD34+ from two CML patients……………………………………………..165 Fig. 5.4. Hes1 gene expression post imatinib mesylate (IM) treatment in CD34+ cells isolated from imatinib sensitive CML patients……………….166 Fig. 5.5. Hes1 gene expression post imatinib mesylate (IM) treatment in CD34+ cells isolated from IM resistant CML patients…………..167 Fig. 5.6. Hes1 gene expression after gamma secretase inhibitor (GSI) treatment in CD34+ cells isolated from CML patients 2, 4, and 5……………169 Fig. 5.7. Hes1 gene expression after gamma secretase inhibitor (GSI) treatment in CD34+ cells isolated from pateint 1 and 6…………………170 Fig. 5.8. Hes1 gene expression in CD34+ CML cells………………………….173 Fig. 5.9. Assessment of P-crkl in CD34+ CML cells following inhibition of Notch by gamma secretase inhibitor (GSI)………………………174 Fig. 5.10. Assessment of P-crkl in gamma secretase inhibitor (GSI) non responsive CD34+ CML cells………………………………………175 Fig. 5.11. P-crkl in CD34+ CML cells treated with gammas secretase 8

inhibitor (GSI)…………………………………………………………………176 Fig. 5.12. GSI treatment induced both Notch and BCR-ABL inhibition in CD34+ cells from one CML sample………………………………….177 Fig. 6.1. Proposed model for Notch and BCR-ABL cross-talk in CML……………194 Fig.6.2. The cooperative model of activated Notch and BCR-ABL signalling in chronic phase CML…………………………………………………...195 Appendix1 …………………………………………………………………………215

9

Abbreviations

ABL (c-ABL) V-abl Abelson murine leukemia viral oncogene homolog 1 ADAM

A Disintegrin And Metalloprotease

ALCL

Anaplastic Large Cell Lymphoma

ALDH

Aldehyde dehydrogenase

AML

Acute Myeloid Leukaemia

Ank

Ankyrin repeats

APP

Amyloid precursor protein

B-ALL

B-cell acute Lymphoblastic Leukaemia

BAM

Bag-of-marbles

B-CLL

B-cell Chronic Lymphocytic Leukaemia

BCR

Breakpoint cluster region protein

BHLH

Basic-Helix-Loop-Helix

BM

Bone Marrow

BMP

Bone Morphogenic Protein

BMT

Bone Marrow Transplant

CADASIL

Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy

CBF1

C promoter Binding Factor 1

CD

Cluster of Differentiation

CDKs

Cell cycle Dependent Kinases

CLP

Common lymphoid progenitors

CML

Chronic Myeloid Leukaemia

cDNA

complementary Deoxyribose Nucleic Acid

Crkl

Chicken tumor virus CT10 regulator of kinase-like protein

CSL

(CBF1, Suppressor of Hairless, Lag-1) family of transcription factors

CXCR-4

Chemokine (C-X-C motif) Receptor 4

DMEM

Dulbecco's Modified Eagle's Medium

DMSO

Dimethylsulphoxide

dNTP

Deoxynucleotyde Triphosphate

10

Dtx

Deltex

EBD

EGF-motif binding domain

EBNA2

Epstein—Barr virus (EBV) nuclear antigen 2

ECM

Extracellular matrix

ECN

Extracellular Notch domain

EGF

Epidermal growth factor

ELR

EGF-like repeats

FACS

Fluorescence-Activated Cell Sorting

FBS

Fetal Bovine Serum

FITC

Fluorescein isothiocyanate

FLT3

FMS-Like Tyrosine kinase 3

Fz

Frizzled

GAGs

Glycosaminoglycans

GAPDH

Glyceraldehydes-3-Phosphate Dehydrogenase

G-CSF

Granulocyte Colony Stimulating Factor

GM-CSF

Granulocyte Macrophage Colony Stimulating Factor

GMP

Granulocyte Macrophage Progenitors

GPA

Glycophorin A

GRB2

Growth factor receptor-bound protein 2

GSCs

Germ Stem Cells

GSK3-β

Glycogen synthase kinase-3beta

HBSS

Hank’s Balanced Salt Solution

HD

Hodgkin disease

HD domain

Heterodimerization domain

HDAC

Histone deacetylase

Hes1

Hairy and Enhancer of Split

HGF

Haemopoietic Grwoth Factor

Hox

Homebox

HRS

Hodgkin and Reed-Sternberg

HSCs

Haemopoietic stem cells

ICSBP

Interferon Consensus Sequence Binding Protein

IFN-γ

Interferon gamma

IL-3

Interleukin-3

IL-4

Interleukin-4 11

IL6

Interleukin-6

IM

Imatinib Mesylate

JAK2

Janus kinase 2

Lin

Lineage specific antigens

LNR

Lin 12/ Notch repeats

LSC

Leukaemic Stem Cell

MAPK

Mitogen-Activated Protein Kinase

M-CSF

Macrophage Colony Stimulating Factor

MDS

Myelodysplastic Syndrome

MM

Multiple Myeloma

MNC

Mononuclear Cells

MTor

Mammalian target of rapamycin

NCR

Notch CytokineResponse domain

NCS

Newborn Calf Serum

NLS

Nuclear Localization Signal sequences

NOD/SCID

Non-Obese Diabetic/ Sever Combind Immuno Deficient

PB

Peripheral Blood

PcG

Polycomb Group

PCR

Polymerase Chain Reaction

P-crkl

Phosphorylated crkl

PE

Phycoerythrin

PEST

Proline-glutamate-Serine-Threonine-rich

(Ph)+

Philadelphia chromosome positive

PI3K

phosphatidylinositol 3-kinase

PI

Propidium iodide

PolyA PCR

Poly Adenylated polymerase Chain Reaction

PPR

PTH/PTHrP Receptors

PT⍺

Pre-T cell receptor ⍺

Ptc

Patched

PTEN

phosphatase and tensin homologue

RBP-Jκ

Recombination signal sequence Binding Protein-J kappa

RT

Room temperature

RT-PCR

Reverse transcription polymerase chain reaction

SCF

Stem cell factor 12

SCLC

Small cell lung cancer

SDF-1

Stromal Derived Factor 1

SFEM

Serum Free Expansion Media

Shh

Sonic hedgehog

Su(H)

Suppressor of hairless

STAT

Signal Transducer and Activator of Transcription

TACE

Tumour necrosis factor-Alpha Converting Enzyme

TAD

Transcription Activation Domain

T-ALL

T-cell Acute Lymphoblastic Leukaemia

TAN-1

Translocation-Associated Notch homolog-1

TCR

T Cell Receptor

TGF β

Transforming Growth Factor β

TNF-α

Tumour Necrosis Factor-alpha

TPO

Thrombopoietin

VLA-4

Very Late Antigen-4

VPA

Valproic acid

VWF

Von Willebrand’s Factor

Wnt

Wingless-type MMTV integration site family

13

(Abstract) Notch signalling is critical for haemopoietic stem cell self-renewal and survival. Chronic Myeloid Leukaemia (CML) is a stem cell disease characterised by the presence of the Philadelphia (Ph) chromosome, and subsequent expression of the BCR-ABL oncogene. The well established role for Notch signalling in human T-cell acute lymphoblastic leukaemia (T-ALL) and the reported interaction between Notch and ABL in different developmental contexts in Drosophila raise the possibility that Notch signalling may be dysregulated in CML. Therefore, this project aims to investigate whether Notch signalling is altered in CML and to study possible crosstalk between Notch signalling pathway and BCR-ABL in CML. The gene expression patterns of all four human Notch genes and the Notch target gene HES1 were studied in CD34+ stem and progenitor cells isolated from CML patients. Poly-A PCR followed by real time PCR analysis was used to quantitate gene expression levels in comparison with levels in equivalent populations isolated from normal bone marrow (NBM). The expression of Notch1 receptor protein levels expressed on the cell surface was also investigated by flow cytometry. Results showed an up-regulation of Notch1 and Notch2 genes and the target gene Hes1 on the most primitive CD34+ Thy+ subset of CML CD34+ cells as compared with NBM. In addition, Notch1 receptor protein was expressed in distinct lymphoid and myeloid progenitors within the CD34+ population of CML cells. These results suggest that Notch signalling may be highly activated in CML primitive progenitors. To investigate the possible crosstalk between Notch and ABL in vitro human cell line model systems were assessed as possible models to study the interactions between Notch and ABL signalling and the FACS based P-crkl assay was optimised as a rapid method to assess ABL activity. The data showed that K562 and ALL-SIL cell lines are sufficient model systems to investigate the cross-talk between the Notch and ABL signalling pathways. The imatinib induced inhibition of ABL activity in K562 and ALL-SIL cells resulted in significant up-regulation of Notch activity as assessed by Hes1 expression. Similarly, GSI inhibition of Notch signalling in K562 cells resulted in hyperactivation of ABL kinase activity as assessed by P-crkl levels. The antagonistic relationship between Notch and ABL signalling observed in cell lines were further confirmed in CD34+ cells from chronic CML patients. Treatment of CD34+ CML cells with imatinib led to significant up-regulation of Notch activity whereas inhibition of Notch signalling with GSI in CD34+ CML cells resulted in increased ABL activity. It can be concluded therefore, that Notch signalling may be dysregulated in the chronic phase of CML. In addition, the data presented in this project demonstrate for the first time the cross-talk between Notch signalling and ABL signalling in cell line model systems as well as in primary CD34+ CML cells. Future work is required to address the possible mechanisms that underlie the findings observed here and to investigate the biological consequences of the interplay between Notch and ABL signalling in CML.

14

Declaration and Copyright Statement

Declaration No portion of the work referred to in the thesis has been submitted in support of an application for another degree or qualification of this or any other university or other institute of learning.

Copyright Statement Copyright in text of this thesis rests with the author. Copies (by any process) either in full, or of extracts, may be made only in accordance with instructions given by the author and lodged in the John Rylands University Library of Manchester. Details may be obtained from the Librarian. This page must form part of any such copies made. Further copies (by any process) of copies made in accordance with such instructions may not be made without the permission (in writing) of the author. The ownership of any intellectual property rights which may be described in this thesis is vested in The University of Manchester, subject to any prior agreement to the contrary, and may not be made available for use by third parties without the written permission of the University, which will prescribe the terms and conditions of any such agreement. Further information on the conditions under which disclosures and exploitation may take place is available from the Vice-President and Dean of the Faculty of Life Sciences.

15

Acknowledgments I would like to gratefully acknowledge the enthusiastic supervision of Dr. Anne-Marie Buckle for her warm encouragement, support and thoughtful guidance throughout my PhD project. Special thanks are due to Dr. Nick Chadwick for his continuous help in experimental design and excellent advice and help in molecular biology techniques. I am grateful to Dr. Virginia Portillo for her constant support and help with flow cytometry, and Susan Slack, for her help with cell culture. I also wish to express my Appreciation to my colleagues Sarah Hoyle, for her help with Westerns and real time PCR, and Dr. Carl Fennessy for his useful IT support and his help in the lab.

I am forever indebted to my mother and my wife for their understanding, endless patience, eternal optimism, and encouragement when it was most required. Finally, this project would not have been possible without the financial support of the government of Saudi Arabia who provided me the Ph.D. scholarship.

16

Chapter 1: Introduction The study of the molecular and cellular mechanisms that control blood cell production in the bone marrow is one of the rapidly developing fields in haematology. An understanding of bone marrow niche elements and the regulatory signalling pathways which dictate developmental fate decisions of haemopoietic cells in health and disease is vital for both medicine and developmental biology. For instance, the discovery and cloning of soluble haemopoietic growth factors (HGF), which are critical for blood cells survival, proliferation, and differentiation, was a major breakthrough in medicine and transplantation settings for the treatment of cancer. The importance of cell-cell interaction in the bone marrow (BM) niche has been emphasised by the discovery of a new role for sets of signalling molecules such as Notch and Wnt proteins which are emerging as critical regulators of normal haemopoiesis. Notch signalling is an evolutionary conserved mechanism that controls cell fate decisions in various body sites both in vertebrates and invertebrate (Ohishi et al. 2003). Much of our knowledge about Notch is gained from Drosophila developmental biology. The phenotype of Notch was first described in Drosophila by Morgan in 1916 in a mutant fly with ‘notches’ in its wings and the gene causing this phenotype was found to be required for the wing outgrowth (Simpson, 1998; and Lai, 2004). However, the first example of the involvement of Notch in malignant transformation in humans was described in a subset of T-cell acute lymphoblastic leukaemia (TALL) carrying the t (7; 9) (q34; q34.3) translocation in which a constitutive expression of Notch was involved in the leukaemogenesis process (Ellisen et al. 1991). It has been shown since then that Notch signalling is critical for haemopoietic stem cell self-renewal and survival (Varnum-Finney et al. 2000; Stier et al. 2002) and in cell fates specification during lymphopoiesis (Pear et al. 2003; Radtke et al. 2004). Dysregulation of Notch activity has been shown to be associated with malignant transformation in various organs including the haemopoietic compartment. The contribution of Notch signalling to haematologic malignancies is well established in

17

T-ALL leukaemia (Bellavia et al. 2002; and Weng et al. 2004) and has been suggested in other malignancies such as B-CLL (Hubmann et al. 2002), Hodgkins lymphoma, and anaplastic large cell lymphoma (Jundt et al. 2002). The involvement of Notch signalling in myeloid leukaemias has been suggested recently (Chiaramonte et al. 2005). Intriguingly, myeloid leukaemias such as AML and CML are stem cell diseases in which leukaemic stem cells (LSC) retain the unique potentials of normal haemopoietic stem cells (HSCs), such as self-renewal capacity, to initiate and maintain leukaemia (Jordan and Guzman, 2004). The concept of cancer stem cells and the critical and well established role of Notch signalling in the HSCs self-renewal make Notch signalling an attractive pathway to study in myeloid leukaemias. Recently, Armstrong et al. (2008) demonstrated the major role of the Notch pathway activation in human T-ALL development and in the long term growth and maintenance of leukaemia-initiating cells. This chapter aims to review the main concepts of haemopoiesis and Notch signalling pathway in the literatures, and to analyse how blood cells are affected by Notch signalling during normal development and in the malignant transformation process. This chapter aims also, based on recent findings, to establish a hypothesis of altered Notch signalling in chronic myeloid leukemia (CML). An altered signalling activity such as this would reveal part of the mysterious molecular mechanisms which drive leukaemic transformation in CML and may provide a molecular target for therapeutic strategies in CML.

1.1.1 Haemopoietic stem cells (HSCs) Haemopoiesis is the process of blood production. In the adult this takes place in the BM, and is maintained throughout life by stem cells. Haemopoiesis can be described as hierarchical with the rare HSCs at the top of the hierarchy giving rise first to progenitors and then to precursors with single lineage commitment and ending in differentiated mature cells of different lineages (Fig. 1.1). HSCs are stem cells that reside mainly in the bone marrow and to a lesser extent in other tissues such as peripheral blood and placenta. HSCs enjoy the unique properties of stem cells. First, they are capable of producing the major haempoietic cell types as needed. Secondly, 18

HSCs self-renew to generate daughter stem cell to maintain enough HSCs pool in the body to sustain the requirement of high numbers of specialised blood cells during physiologically stressful conditions. Thirdly, HSCs have an extreme proliferation potential which enable them to meet the high demands of haemopoiesis throughout the normal adult life span (Szilvassy, 2003). Although most HSCs remain quiescent and do not enter the cell cycle, HSCs undergo cell divisions to self-renew and differentiate, or either, through asymmetric cell division, to differentiate and generate more mature and specialised blood cells (Rao and Mattson, 2000). A balance between the numbers of stem cells, committed progenitors, and differentiated haemopoietic cells in the bone marrow is maintained throughout an individual’s life. The fate of HSCs in the bone marrow is highly regulated at the molecular level through complex sets of internal and external signals that will be discussed below. The knowledge of these regulatory elements is essential for the improvement of the current clinical use of HSCs for the treatment of patients with haematological disordesr. Human haemopoietic stem cells are extremely rare and difficult to identify. However, several phenotypic and functional characteristics in vitro and in vivo have been used to identify HSCs. Immunophenotypically, HSCs are characterized by the expression of several antigens such as CD34, Thy-1 (CDw90), CD117 and the absence of coexpression of HLA-DR or CD38 without expression of lineage specific antigens (Lin). The more differentiated CD34+ haemopoietic progenitors are CD38+, Thy-1 negative, and might express one of lineage specific markers such as CD19 for B lymphoid progenitors, CD7 for T lymphoid progenitors, CD33 for myeloid progenitors, CD71 or glycophorin-A for erythroid precursors, and CD41 or CD61 for megakaryocytic progenitors (Steidl et al. 2003). Interestingly, some reports have described a small subpopulation of HSC that is CD34- CD38- Lin- and there is evidence that these cells can give rise to CD34+ HSCs (Zanjani et al. 1998, Bathia et al. 1998). Different approaches, using phenotypic and/or functional markers, have been attempted to isolate HSCs. The classical HSC marker CD34 has been used routinely to identify and isolate HSCs by flow cytometry. Recently, CD133 (the human 19

homologue of prominin 5-transmembrane glycoproteins) has been proposed to be a prominent HSC marker. (Bonde et al. 2004). Purification strategies, which are based on both, conserved stem cell function as well as on phenotype, have been suggested to be more representative of stem cells than the classical phenotypic approach. For example, the use of metabolic markers such as rhodamine and Hoechst 33342 dye efflux, and the enzyme ALDH (Aldehyde dehydrogenase) yielded repopulating cells of high stem cell activity in vivo (Bonde et al. 2004). The 'gold standard' method of identifying HSCs is based on their capacity to repopulate the entire haemopoietic system in lethally irradiated recipients following transplantation. Three models have been used for this purpose which are the nonobese diabetic-severe combined immunodeficient mouse model (NOD/SCID), the beta2 microglobulin-deficient (B2m null) NOD/SCID (β2m null NOD/SCID) and the sheep foetus model which, unlike the mouse models, does not require myeloablation before transplantation (Bonnet et al. 2003; Kollet et al. 2000). Repopulating assays using the mouse model have revealed that there appear to be two kinds of HSCs; long-term (LT-HSC) and short- term (ST-HSC) repopulating stem cells. If bone marrow cells from the transplanted mouse can, in turn, be transplanted to second lethally irradiated mouse and restore its haemopoietic system over four months, they are considered to be long-term stem cells that retain their self-renewal capacity. On the other hand, short-term stem cells can immediately regenerate all the different types of blood cells in the transplanted mouse, but lack the ability to renew themselves if transplanted to another lethally irradiated recipient (Coulombel, 2004). In mouse the LT-HSC self-renew for more than four months whereas the ST-HSC has the ability to self-renew for six to eight weeks only. The ST-HSC then advance to the multipotent progenitor (MPP) cells that can self-renew for less than two weeks (Shizuru et al. 2005). In the last few years, several reports have demonstrated the non-lineage restricted potentials of HSCs and their capacity to transdifferentiate into a variety of non haemopoietic cell types such as neural cells, hepatocytes, cardiomyocytes, and endothelial cells, a process termed as plasticity. However, other studies have challenged the concept of stem cell plasticity and suggest that cell fusion, rather than transdifferentiation may explain the acquisition of non-lineage phenotypes and 20

functions (Steidl et al. 2003). Despite the debate in the scientific literature on the molecular mechanisms responsible for these observations, the accumulating data suggests that new therapeutic potentials of haemopoietic stem cells can be used in areas such as heart and brain degenerative diseases.

Fig.1.1 The hierarchy of haemopoiesis. Haemopoiesis can be described as hierarchical process with the rare HSCs at the top of the hierarchy giving rise first to committed progenitors and then to precursors with single lineage commitment and ending in terminally differentiated mature cells of various lineages. Haemopoiesis is dependent on long-term self renewing HSCs (LTHSC) which gives rise to short-term self renewing HSCs (ST-HSC) which, in turn, differentiate to produce multipotent progenitor cells (MPP). MPPs differentiate into Common Lymphoid Progenitors (CLP) or Common Myeloid Progenitors (CMP). The CLP differentiate into cells of the lymphoid (T cells, B cells and natural killer (NK) cells whereas the CMP further subdivide into Megakaryocyte/Erythroid progenitor (MEP) and Granulocyte–Macrophage progenitor (GMP) which give rise to functional mature myeloid cells. Both the CMP and the CLP can be induced to differentiate to dendritic cells. (Modified from Larsson and Karlsson, 2005)

21

1.1.2 Regulation of haemopoiesis The process of haemopoiesis involves complex interactions between the intrinsic genetic processes of haemopoietic cells and the diverse extrinsic regulatory and signalling molecules in their niche. These interactions determine whether HSCs, progenitors, and differentiated cells remain quiescent, proliferate, differentiate, selfrenew, or undergo apoptosis (Fig. 1.2). In general, the intrinsic and extrinsic regulatory mechanisms of haemopoiesis work together to maintain a balance between all these cellular processes to fulfil the requirement of blood cell production in normal steady states, as well as in the event of stress such as bleeding or infection. For instance, under normal conditions, the majority of HSCs and many progenitors are adherent to the niche and are not cycling, whereas many of the more mature blood progenitors are proliferating to generate mature functioning blood cells. Apoptosis balances the rate of proliferating progenitors in the absence of stress. However, during stress or injury, the stored pools of cells in the BM are released to the site of injury. Concurrently, diverse regulatory molecules in the niche signal the quiescent HSCs and progenitors to proliferate and differentiate while decreasing the rate of apoptosis (Smith, 2003). When the stress ceases, the kinetics of haemopoiesis return to base line levels and the anti-apoptotic and proliferative processes wind down. A variety of environmental and genetic regulatory mechanisms involved in haemopoiesis will be discussed here. Programmed cell death (apoptosis) is involved in the regulation of haemopoiesis at different levels of haemopoietic cells development. At the HSCs level, apoptosis regulates the size of the HSC pool by regulating HSCs production and elimination in response to steady or stressful physiological demands (Kondo et al. 2003). HSCs express primarily the anti-apoptotic protein BCLxL of the BCL2 family members that protects HSCs from apoptosis and enhance their survival. Survival of HSCs and haemopoietic precursors is mediated by the availability of certain haemopoietic cytokines in the niche and deprivation from cytokines induces apoptosis. Cytokines show target cell selectivity in preventing apoptosis. For example, stem cell factor selectively promotes survival of primitive hematopoietic cells, IL-3 inhibit apoptosis in more committed progenitors, whereas Flt ligand is selective for progenitors 22

committed to the myeloid lineage (Wickremasinghe and Hoffbrand, 1999). Other cytokines such as IFN-γ and TNF-α modulate haemopoietic cells survival in the bone marrow by inducing apoptosis via increasing expression of FAS on the surface of haemopoietic progenitors (Maciejewski et al. 1995).

1.1.2.1 The stem cell niche The stem cell microenvironment or niche is a term that describes the diverse combination of differentiated cells which surround stem cells and secrete a rich extracellular matrix and substrates that modulate stem cell self renewal and regulate stem cell survival and functions (Calvi et al. 2003; Fuchs et al. 2004). The ability of stem cells to reside within niches is an evolutionally conserved phenomenon, which has been shown to be vital for stem cell survival and functions. For instance, studies on Drosophila germ stem cells (GSC’s) have shown that direct physical interactions between stem cells and their surrounding cells in the niche are crucial for maintaining stem cell survival and self- renewal. In Drosophila ovaries, terminal filament and cap cells line the basal lamina (BL) and constitute a niche for GSC’s where GSC’s are in physical contact with cap cells. When female GSC’s divide, the daughter cells, which are in contact with cap cells, remain as stem cells whereas cells that lose cap contact lose their stemness and differentiate and initiate oogenesis (Fuchs et al. 2004). In mice, adult haemopoietic stem cells (HSCs) reside in the bone marrow niche where they traverse along the inner surface of the bone which is lined by osteoblasts. The BM niche comprises osteoblasts, extracellular matrix (ECM), and marrow stromal cells which are heterogeneous themselves comprising fibroblasts, reticular cells, macrophages, adipocytes, and endothelial cells. The bone marrow niche is defined by cell-cell interactions, cell-ECM interactions, and exposure to diverse soluble factors including cytokines and various signalling molecules (Frisch et al. 2008).

1.1.2.1.1 Cell-ECM interactions Extracellular matrix (ECM) is composed of three major classes of molecules: structural proteins such as collagen and elastin, specialized proteins such as fibronectin and laminin, and proteoglycans which consist of a protein core to which is attached long chains of glycosaminoglycans (GAGs). The general effect of adhesion

23

of HSCs and progenitors to the marrow ECM is suppression of proliferation and prevention of apoptosis (Arai and Suda, 2007). Several adhesion molecules on the HSCs membrane such as integrins, immunoglobulin-like molecules, cadherins, selectins, and mucins mediate adhesion of HSCs to the basal lamina of extra cellular matrix. To be able to reside in their niche, HSCs express high levels of the integrins α4β1 (also termed VLA-4) and

α5β1,

which bind to fibronectin on ECM to promote adhesion to the bone marrow stroma. Integrins binding to fibronectin inhibit differentiation and promote HSCs survival and quiescence through the inhibition of cell cycle dependent kinases (CDKs) such as P27 (Cheng et al. 2000). Moreover, it has been shown that loss or alteration of integrin expression leads to the departure of HSCs from their niche either through differentiation or apoptosis (Nervi et al. 2006). Similarly, c-kit, which belongs to the immunoglobulin super-family, has been found to be highly expressed in HSCs in normal steady state and is down-regulated in mobilised cells (Kondo et al. 2003). Ckit signalling is known to promote survival and proliferation through its binding to stem cell factor and has been shown to be involved in the JAK2 signalling pathway which triggers potent self-renewal effect on HSCs (Zhao et al. 2002). Of particular interest, is the unique ability of a niche to retain its stem cells, a process referred to as homing. This feature has been demonstrated in HSCs transplantation studies in which the homing molecule VLA-4 was found to be vital for successful engraftment (Craddock et al. 1997).

1.1.2.1.2 Soluble factors in the niche In vitro studies have shown that various cytokines and growth factors in the bone marrow niche are secreted by cells adjacent to HSCs which, depending on their concentration or specific combination, support survival, or proliferation and differentiation of haemopoietic progenitors and stem cells. Cytokines

24

Cytokines are low-molecular- weight regulatory proteins or glycoproteins secreted by stromal cells, white blood cells, and other cells in response to different stimuli. (Goldsby et al. 2003). Once secreted, they bind to specific receptors on the membrane of target cells and trigger signal transduction pathways, which ultimately cause gene activation and alter gene expression in the target cells. The haemopoietic growth factors (HGFs) represent a group of cytokines with well-defined effects on the haemopoietic system. The main biological function of HGF, is to act as means for short-range intercellular communication that influences self-renewal, survival, proliferation, and differentiation of haemopoietic cells at different stages of haemopoietic maturation (Fetscher & Mertelsmann, 2002). Growth factors that are secreted by stromal cells have been shown to enhance the proliferation of early haematopoietic stem and progenitor cells are FLT3, SCF, erythropoietin, IL6 and thrombopoietin (TPO) (Krause, 2002). Chemokines These are molecules in the bone marrow niche that regulate blood cell trafficking and homing. β1-Integrins such as VLA-4 and VLA-5, which are expressed on CD34+ cells, play a dominant role in adhesive interactions to mechanically tie HSCs in the niche. Moreover, β1-Integrins regulate HSCs proliferation and survival through different mechanisms, such as the inhibition of cell cycle dependent kinases (CDK’s) such as P27 and the activation of the RAS/ MAPK signal transduction pathway, which result in an increased expression of c-myc which is known to shorten the G1 phase of the cell cycle (Steidl et al. 2003). The β2-integrin LFA-1 plays a similar role in adhesion and trafficking of CD34+ haemopoietic cells and progenitor cells. L-selectins mediate the initial contact of leukocytes with endothelium and might also be involved also in homing of HSCs (Krause, 2002). The adhesion molecule CD44, which binds to hyaluronic acid and fibronectin, has been shown to be highly expressed in CD34 + cells in the bone marrow in steady state conditions as compared to CD34+ cells in peripheral blood (Lataillade et al. 2005). This finding supports the notion of the importance of CD44 in haemopoiesis and stem cell trafficking. Indeed, CD44 monoclonal antibodies against CD44 inhibit adhesion to bone marrow stroma and stopp haemopoiesis in murine long-term bone marrow cultures (Christ et al. 2001).

25

An important interaction between haemopoietic cells and their niche is mediated by the α-chemokine stromal-derived factor 1 (SDF-1) and its receptor CXCR-4. SDF-1, produced by bone marrow stromal cells, binds to CXCR-4 receptor on CD34+ cells and plays a critical role in HSCS and haemopoietic cells migration. As haemopoietic cells in the bone marrow differentiate, they express low levels of the CXCR-4 receptor in preparation to leave the bone marrow niche (Steidl et al. 2003). Recently, it has been reported that CD44 and its hyaluronic acid ligand cooperated with SDF-1 in the trafficking of CD34+ cell to the BM, demonstrating a cross-talk between CD44 and CXCR4 (Avigdor et al. 2004). Another key family of the growth factors in the HSCs niche is the transforming growth factor β (TGF β) family, which is known to play a role in maintaining HSC in a quiescent stem cell state. It has been shown that osteoblasts receive the TGF β signal and respond by increasing their numbers and thus, indirectly promote HSCs stemness by causing more stem cells to adhere to osteoblasts (Zhang et al. 2003). This indirect mode of action of TGF β on mammalian HSCs is in contrast to the direct mode of this niche factor on the GSC’s in Drosophila where DPP (member of the TGF β family) is secreted by cap cells and bind directly to receptors on GSC’s to support their self -renewal by suppressing the differentiation factor BAM (Fuchs et al. 2004). Wnt proteins represent a growing family of secreted signalling molecules in the bone marrow niche that have been shown to be critical for HSCs self renewal and proliferation (Ryea et al, 2003). Wnt proteins bind to FZ-family receptors on HSCs and trigger the Wnt-catenin signalling pathway which, ultimately leads to accumulation of β -catenin in the cytoplasm before it translocates to the nucleus and facilitate transcription of target genes. Purified mouse HSCs that have been transduced with the active form of β catenin showed high proliferation and maintained the immature phenotype of HSCs in long-term cultures (Staal and Clevers, 2005). The Hedgehog (Hh) pathway is another cascade that plays a crucial role in HSC proliferation. Sonic hedgehog (Shh) is one of three trans-membrane proteins that

26

comprise the Hh family in humans and that mediates signalling through cell-to-cell contact between adjacent cells expressing the Patched receptor (Ptch). Alternatively, Hh ligands can be found as soluble molecules in the niche, where they can stimulate cells in the niche that express the Ptc receptor .Shh, Ptc, and Smo (Smoothened, another Hh receptor) are expressed in primitive human CD34+ CD38- Lin- cells as well as in stromal cells which imply that both HSCs and marrow stromal cells can transduce the Hh signals (Bhardwaj et al. 2001; Szilvassy, 2003). It has been shown that the interaction between HSCs and Hh ligands is critical for HSCs survival and expansion ex vivo. This effect is mediated via regulation of the bone morphogenic protein (BMP)/ TGFβ superfamily (Kondo et al. 2003). 1.1.2.1.3. Cell-cell interactions HSCs communicate with osteoblasts, stromal cells, committed haemopoietic progenitor cells and other HSCs via receptor/ligand interactions in their niche. It has been demonstrated that physical contact between HSCs and osteoblasts is critical for stem cells to retain their unique properties of self-renewal and quiescence. When mice are genetically altered to increase osteoblast numbers, the numbers of HSCs increased significantly and their ability to remain quiescent without further differentiation was dependent on their ability to adhere physically to the osteoblasts through N-cadherinmediated adhesion junctions (Zhang et al. 2003). The molecular glue that anchor stem cells to osteoblasts in the BM niche is known as adherens junctions, which are formed by two important molecules, cadherin and catenins (Fuchs et al. 2004). The contribution of osteoblasts to HSCs niche in mammals was elegantly studied by Calvi et al (2003). They found that in mice genetically altered to produce activated PTH/PTHrP receptors (PPR) specifically on osteoblasts, PPR signalling resulted in increase of osteoblast numbers and over expression of the Notch ligand Jagged 1 and subsequent promotion of HSCs self renewal through Notch activation (Calvi et al. 2003). In line with this, Visnjic and others employed a genetic strategy, to selectively, and reversibly, eliminate osteoblasts from bone and found that osteoblast ablation led to dramatic loss of bone marrow cellularity and a reduced number of early haemopoietic progenitors (Visnjic et al. 2004). Another important cell-cell interaction in the bone marrow niche is the interaction between Notch receptors on HSCs and

27

Notch ligands expressed by stromal cells. Notch signalling is known to be critical for HSCs self- renewal and survival and will be discussed later in detail. If HSCs receive the appropriate signal to differentiate in a steady state, or in response to stressful demands, they lose contact with neighbouring osteoblasts and cell matrix and differentiate to specific cell lineages, before they head towards the central bone marrow cavity and traverse into the circulation.

1.1.2.2 Genetic control of haemopoiesis The sequential differentiation decisions in haemopoiesis from HSCs to intermediate progenitors and fully differentiated cells are highly regulated within the cell. Using knock-out animal models and gene expression profiling, it has been found that distinct expression patterns of genes and transcription factors are needed in different developmental stage of haemopoiesis. For instance, Steidl et al. (2003) have found a higher expression of genes for cell cycle progression in BM-CD34+cells as compared to PB-CD34+ cells. In the same study, they have identified the genes responsible for the transition from quiescence to active cycling CD34+. Furukawa and co-workers have studied the expression patterns of cell cycle genes in different differentiation stages in haemopoiesis. They have demonstrated a universal up-regulation of cdc2, cdk4, cyclin A, cyclin B, and p21, and down-regulation of p16 during differentiation of haemopoietic cells (Furukawa et al. 2000). In addition, there are gene expression patterns that are specific for certain haemopoietic lineages, which mean that cell cycle control genes are modulated during haemopoiesis to control the differentiation and proliferation of haemopoietic cells. Transcription factors also play a vital role in the differentiation of HSCs and progenitor cells. Bmi-1, which is a member of the polycomb group (PcG) family of genes, has been shown to be highly expressed in HSCs and declines during haemopoietic development. Competitive repopulation studies have demonstrated that Bmi-1 is crucial for HSCs self-renewal (Stein et al. 2004). The homebox (Hox) genes exhibit distinct pattern of expression during haemopoietic differentiation (Stein et al. 2004). HoxB4, for example, is abundantly expressed in immature haemopoietic cells

28

including HSCs, but declines with lineage differentiation, which underlines a possible central role of this transcription factor in early haemopoiesis. Another member of the Hox transcription factor family, HoxA10, is critical for the regulation of myeloid differentiation. Similarly, PU1 and GATA-1 transcription factors have been shown to initiate myeloid differentiation (Steidl et al. 2003). Similarly, the Pax5 transcription factor has been shown to be vital for the development of B cell progenitors (Nutt et al. 2001). Other transcription factors have also been found to be indispensable for other haemopoietic lineages proliferation and differentiation. These findings support the notion that distinct expression patterns of transcription factors steer the balance between self-renewal and commitment to differentiation of haemopoietic stem cells. Two models have been proposed for the genetic dictation of haemopoietic cells fate decisions during haemopoiesis. The stochastic model argues that the developmental fate of haemopoietic cells is predetermined by intrinsic genetic processes to occur in certain sequence and timing and the external environmental signals then act to modulate these genetic effects (Smith, 2003). The instructional model, however, suggests that the external environmental signals may play a primary role in directing cells toward various developmental fates by inducing the appropriate genetic change.

29

Fig. 1.2 Regulation of haemopoiesis. Extrinsic and intrinsic (genetic) mechanisms control HSC quiescence, self-renewal and differentiation. Extrinsic mechanisms involve the interaction between HSC and the microenvironment. Physical association between HSC and osteoblasts or other cells in the niche trigger diverse signal transduction pathways that initiate expression of downstream target genes. Intrinsic (genetic) mechanisms can also dictate the haemopoietic cells fate decisions during haemopoiesis. (Modified from Rizo et al. 2006) 30

1.2 Notch signalling pathway Notch signalling is an evolutionary conserved mechanism that controls cell fate decisions in various sites in the body, in both vertebrates and invertebrates (Ohishi et al. 2003). Notch signalling involves binding between a Notch receptor on one cell and a ligand on the neighbouring cell, which triggers the cleavage of the intra-cellular domain of Notch from its membrane-bound tether and a subsequent translocation to the nucleus, where it activates transcription of the CSL family of transcription factors (CBF1 or RBP-Jκ in vertebrates, Su(H)) in drosophila, or LAG-1 in Caenorhabditis elegans). This activation leads to increased transcription of certain target genes, such as the Hairy and Enhancer of Split (HES)-1 gene (Mumm and Kopan, 2000). The phenotype of the Notch gene of Drosophila was discovered by Morgan in 1916 in a mutant fly with ‘notches’ in its wings, and the gene was found to be required for the wing outgrowth (Simpson, 1998; and Lai, 2004). However, it was not until 1983 that the Drosophila Notch gene was cloned (Kidd et al. 1983; Artavanis-Tsakonas et al. 1983). Notch has since been found to be a key player in a wide range of developmental processes throughout different organisms, ranging from the fruit fly to the human. Some 13 years after the cloning of Drosophila’s Notch gene, four mammalian Notch genes were cloned, known as Notch 1-4. Notch 1 was the first Notch protein to be identified in Human and was initially named TAN-1 – for Translocation-Associated Notch homologue-1 – (Das et al. 2004). It was cloned as a gene involved in the t(7;9)(q34;q34.3) chromosomal translocation found in a subset of human T-cell leukaemia (Gridley, 2004).

1.2.1 Notch receptors Mammalian Notch genes encode four Notch receptors, Notch 1-4. These are large (300 KDa) single pass trans-membrane proteins that are cleaved within the transGolgi network during biosynthesis by a Furin-like convertase. This cleavage occurs at the site known as S1 and yields a heterodimer cell surface receptor (Maillard et al. 2003; Baron, 2003). 31

Notch receptors consist of two domains, which remain non-covalently bound together by a calcium-dependent interaction. The extra-cellular domain (ECN) consists of 2936 tandem epidermal growth factor (EGF-like) repeats that bind Notch ligands, and three Lin 12/ Notch repeats which are crucial for maintaining Notch in a resting conformation before ligand binding (Nam et al. 2002). The The Notch transmembrane domain (NTM) consists of a transmembrane region and the intracellular Notch domain (ICN). The intracellular Notch domain (ICN) contains a RAM domain, a cdc 10/ ankyrin-like repeats flanked by two nuclear localisation signal sequences (NLS), and a c-terminal proline-glutamate-serine-threonine-rich (PEST) domain, which is important for regulating protein stability (Fig.1.3). In the Drosophila Notch and human Notch 1 and 2, the ICN also has a transcription activation domain (TAD), which is absent in Notch 3-4. Moreover, the ICN has a Notch cytokine response domain (NCR), which may be involved in cytokine signalling. Notch 4 has a shorter intracellular domain that lacks one of the NLS and the whole NCR domain The RAM domain and ANK repeats are binding sites for the downstream transcription factor CBF-1/RBPJ, which is the human homologue of Drosophila Su(H) (Ohishi et al. 2003). Although the RAM domain is the primary binding site for the transcription factor CBF-1/RBPJ, the ANK repeats facilitate this binding, and more importantly, they are the binding sites of many important proteins that modulate Notch signalling, such as Deltex and Mastermind (Kojika and Griffin, 2001; Fleming, 1998).

Notch ligands 1.2.2 Two Notch ligands (Delata and Serrate) have been identified in Drosophila and five ligands have been identified in mammals (Jagged1, Jagged2, Delta-like1, Delta-like3, and Delta-like4) (Maillard et al. 2003). Notch ligands are trans-membrane proteins, which are composed of an extra-cellular domain, transmembrane domain and a relatively short cytoplasmic tail (Fig 1.4). The extra-cellular domain consists of Nterminal domain (NT) of 100-165 amino acids and a unique amino-terminal DSL domain (named for Delta, Serrate from Drosophila, and Lag-2 from C. elegans). Both the NT domain and the DSL domain constitute the EGF-motif binding domain (EBD), which has been found to be indispensable for proper interaction with Notch expressing cells (Fleming, 1998). It is also the EBD region of the ligand protein that regulates ligand recognition and specificity by modulators such as a fringe gene

32

product. Next to the EBD domain and before the transmembrane domain (TM) lie the EGF-like repeats (EGFR) and an additional cysteine-rich region, which is found only in Serrate and Jagged related groups. Although it has been shown that the EGFR domain is not essential for Notch signalling, missense mutation studies suggest that it may be important for stabilising ligand / receptor interaction (Tax et al. 1994; Fleming, 1998). The function of the cysteine-rich region is not clear, but it may be important in ligand specificity since this domain is not found in delta-like ligands. Moreover, it has been demonstrated that the TM domain and the cytoplasmic portion of Notch ligands are crucial for Notch activation (Fleming, 1998). In Drosophila, it has been found that endocytosis of ligands after binding Notch receptor is essential for the activation of Notch signalling (Kojika and Griffin, 2001).

Fig.1.3 Structure of human Notch receptor. The extracellular Notch protien (ECN) is composed of epidermal growth factor like repeates (EGFR), Lin12 Notch repeats

EGF like repeats

(LNR) and the heterodimerization (HD) domain while the intracellular Notch protien (ICN) is composed of a RAM domain, a series of cdc10/ Ankyrin repeats (ANK) flanked by two nuclear localization signal sequences (NLS) (not shown here), a Notch cytokine response domain (NCR) which is absent in Notch 4, a transcriptional activation domain (TAD) which is absent in Notch 3 and 4, and a c-terminal PEST region (P). The Notch transmembrane domain (NTM) consists of a transmembrane region and ICN. The sites of the proteolytic cleavages S1,S2, and S3 are indicated. 33

Extracellu

Fig. 1.4 Structure of Notch ligands. Notch ligands are composed of an extracellular domain, transmembrane domain (TM) and a relatively short cytoplasmic tail. The extracellular domain consists of N-terminal domain (NT) and a unique amino-terminal DSL domain. Both, the NT domain and the DSL domain constitute the EGF-motif

NT DSL

EGF-like

binding domain (EBD), which has been found to be indispensable for proper interaction

with Notch receptors and modulators. Next to the EBD domain and before the transmembrane domain (TM) lie the EGF-like repeats (EGFR). An additional cysteinerich region, which is found only in the Serrate and Jagged ligand family (Modified from Guidos, 2002).

34

1.2.3 Molecular mechanisms of Notch signalling Physical interaction between specific EGF repeats in the DSL domain of a ligand and the EGF repeats of the ECN receptor protein triggers two successive cleavages, resulting in the release of ICN and its subsequent translocation into the nucleus. The first cleavage, mediated by an ADAM metaloprotease (TACE in vertebrates, and possibly Kuzbanian/SUP-17 in invertebrates), occurs external to the transmembrane domain (S2 cleavage site), and releases the majority of the extra-cellular Notch domain. The second cleavage occurs within the trans-membrane domain (S3 site) (Fig. 1.3) and is mediated by γ -secretase, a multi-protein complex with secretase activity whose components include presenilin, nicastrin, Aph1 and Pen2 (Fortini, 2002, Baron, 2003; Lai, 2004). This cleavage releases soluble ICN into the cytoplasm, which then translocates to the nucleus.. Once in the nucleus, the ICN binds directly through its RAM and ANK domains to the CSL transcription factor (CBF1 in vertebrates, Su(H)) in drosophila, or LAG-1 in Caenorhabditis elegans), and converts it from a transcriptional repressor into a transcriptional activator (Fig. 1.5). It has been shown that the binding of ICN to CSL displaces co-repressor complexes from CSL (such as histone deacetylase (HDAC)) and recruits, through ANK and TAD domains of ICN, different transcriptional coactivators such as mastermind in Drosophila (MAML1 in mammals) to convert CSL into transcriptional activator. In this way transcription of Notch downstream target genes (e.g. HES) which eventually control specific developmental decisions in different cellular contexts (Mumm and Kopan, 2000; Lai, 2004; and Wu et al. 2002). The most widely characterised mammalian Notch target gene is Hes1, although Hes5 is also known to be involved in Notch signalling (Kojika and Griffin, 2001). The HES genes in mammals encode basic-helix-loop-helix (bHLH) transcription factors, in which the basic domain is needed for DNA binding and the HLH domain mediates interactions with other bHLH proteins. Once activated by Notch, CSL binds the regulatory sequences of the HES gene in the nucleus (GTGGGAA), and upregulates expression of its encoded bHLH proteins (Artavanis-Tsakonas et al. 1999). bHLH transcription factors contain peculiar WRPW sequences at the carboxyl

35

terminus, which recruit transcriptional repressors such as Groucho in Drosophila or its mammalian homologue TLE. The overall effect is the repression of the transcription of downstream target genes, necessary for cell fate decisions in processes such as myogenesis, somitogenesis, sex determination, vasculogenesis, lymphocytes development and neurogenesis (Davis and Turner, 2001, Iso et al. 2003). Examples of bHLH transcription factors include MASH1 (HASH1 in human), which is important in neurogenesis. Over-expressed HES1 can repress MASH1 transcription, and thus act as a negative regulator of neurogensis, by directly repressing a pro-neural gene, MASH1. Similarly, HES1 acts

as

the

effector

of

Notch

signalling

to

repress

myoD

transcription in myoblasts, and thereby restrict muscle formation (Davis and Turner, 2001). CD4 is another candidate target gene for HES1, where over-expression of HES1 leads to the down-regulation of the endogenous CD4 expression in CD4+ CD8- TH cells (Kim and Siu, 1998). The HES family is not the only known effector of Notch in mammals as a new bHLH family has been isolated and named as HERP (reviewed in Iso et al. 2003) and a wide range of additional Notch target genes such as C-myc, cycline D1, and deltex have been recently identified (Aster et al. 2008). Despite the linear picture of the Notch signalling pathway described above, several lines of evidence support the existence of CSL-independent Notch signalling pathways in Drosophila and vertebrates (reviewed by Martinez Arias et al. 2002). However, the exact mechanisms of these alternative pathways await further explanation.

36

Sending cell

Fig. 1.5 The CSL-dependent Notch signalling pathway. Notch receptor is present on the cell surface as heterodimer. Upon binding with a ligand on adjacent cell, two proteolytic cleavages at sites (S2) and (S3) occur which liberate the intra cellular domain of Notch (ICN) in the cytoplasm. ICN translocates to the nucleus and binds to the transcription factor CSL, which displaces co-repressors (CoR) and recruits co-activators (CoA) including MAML, leading to transcriptional activation of downstream target genes including Hes1,Deltex, and C-Myc. (Modified from Aster et al. 2008)

37

1.2.4 Modulators of Notch signalling Notch signalling is finely regulated through several modulators that act at the extracellular, cytoplasmic or nuclear levels. The EGF repeats on the extra-cellular domain of Notch receptor undergo O-fucosylation and O-glucosylation which modulate various activities of Notch such as Notch-ligand interaction and intracellular trafficking of Notch (Acar et al. 2008). Examples of the extra-cellular modulators of Notch signalling which influence receptor-ligand interaction may include fringe and Ofut1. Fringe is a glycosyltransferase protein that controls the specificity of Notchligand binding. In Drosophila, fringe physically interacts with the extra-cellular domain of Notch and modifies O-linked fucose on specific Notch EGF-repeats; including EGF repeat 12 (EGF12), to restrict Notch activation to the delta ligand (Panin and Irvine, 1998; Schweisguth, 2004). Three vertebrate homologues of fringe have been identified, (L fringe, M fringe, and R fringe – that control the specificity of receptor-ligand binding in different cellular contexts (Kojika and Griffin, 2001). The O-fucosylation of Notch is catalysed by a GDP-fucose protein O-fucosyltransferase encoded by the Ofut1 gene. The regulation of the Ofut1 gene expression is vital for normal Notch signalling, as over-expression of Ofut1 or loss of its function disturbs ligand-Notch interaction and blocks Notch signalling (Schweisguth, 2004). Interestingly, it has also been shown that ubiquitination of Delta ligand in Drosophila up-regulates Delta signalling activity and promotes Notch activity through as yet unrevealed mechanisms (Schweisguth, 2004). Recently, Acar et al (2008) isolated a gene named rumi in Drosophila which encodes the O-glucosyltransferase protein Rumi. The authors showed that Rumi is essential for Notch-ligand binding and proposed that lack of O-glucosylation of Notch in rumi mutants results in a defect in Notch folding and signalling. Several cytoplasmic modulators of Notch signalling have been described. Numb negatively regulates Notch, probably through a direct protein-to-protein interaction that requires the phosphotyrosine-binding (PTB) domain of Numb and the RAM domain of Notch. Numb is a unique protein, which is known to be asymmetrically 38

segregated to only one of the two daughter cells in sensory organ development. A cell acquiring the Numb protein adopts a fate different to its sister cell. Therefore, Notch continues to function positively in the daughter cell that does not inherit Numb. It has been demonstrated that Numb may influence cell fate by negatively regulating the Notch signalling pathway (Guo et al. 1996). Numb inhibits Notch signalling by preventing the intra-cellular domain of Notch from translocating to the nucleus. Recently, it was shown that this was achieved in mammals through promoting the ubiquitination of Notch1, leading to the degradation of the intracellular domain following receptor activation (McGill and McGlade, 2003). Another Notch regulator is Deltex. Deltex was originally identified in Drosophila as cytoplasmic positive regulator of Notch signalling, through direct interaction with the ANK repeats of ICN (Mastuno et al. 1995). However, recent investigation of Deltex action in mammalian cells suggests that enforced expression of Deltex inhibits Notch signalling, probably through its competition with ICN for transcriptional co-activators (Izon et al. 2002). Moreover, it has been found that a significant fraction of DTX1 (a mammalian homologue of Drosophila Deltex) interact physically in the nucleus with the transcriptional co-activator p300, leading to transcription repression of the MASH1 target gene, and the subsequent inhibition of neuronal differentiation (Yamatomao et al. 2001). This finding suggests a possible Notch-independent role of Deltex in transcription regulation. Collectively, it seems that the role of Deltex on Notch signalling is cell context dependent. Additional nuclear modulators of Notch have been identified in Drosophila include SEL-10 and Suppressor of Deltex (SU(Dx) ), both of which are inhibitors of Notch signalling. Suppressor of Deltex is an E3 ubiquitin ligase, that binds to target proteins and adds ubiquitin. It has been suggested that SU(Dx) may interact with ICN and induce ubiquitation, and thus lead to its degradation (Kojika and Griffen, 2001).

1.2.5 Notch signalling in haemopoiesis Notch receptors (N1-4) have been identified in haemopoietic progenitors. Notch-1 has been shown to be expressed in a wide range of haemopoietic cells at different levels 39

of maturation including CD34+ lin- precursors and CD34+lin+ precursors, as well as lymphoid, myeloid, and erythroid precursors. Notch1 hase been also detected in peripheral blood T and B lymphocytes, monocytes and neutrophils (Milner and Bigas, 1999). The expression patterns of Notch-1 and 2 in different haemopoietic lineages are distinct, ranging from low levels in CD34+ precursors to high levels in monocytes (Ohishi et al. 2000; Walker et al. 2001; Singh et al. 2000, and Jonsson et al. 2001). Moreover, the expression patterns of Notch 1-4 genes in various maturation stages of T and B lymphocyte development have been studied recently (detailed in Saito et al. 2003). Notch ligands have also been found in haemopoietic tissue including foetal liver, BM and thymus, and in populations of haemopoietic cells. Jagged-1, Delta-1 and Delta -4 have been detected in bone marrow stromal cells, whereas Jagged-1 is expressed in haematopoietic cells such as macrophages, megakaryocytes and mast cells (Ohishi et al. 2003). These findings, and the evolutionally conserved role of Notch signalling in cell fate decisions, suggest a role of Notch signalling in haemopoiesis.

1.2.5.1. Notch and haemopoietic stem cell (HSC) fate decisions Several lines of evidence have shown that Notch signalling is at the centre of regulating stem cell fate choices, in terms of self-renewal and/or differentiation. In a recent study, inhibition of Notch signalling in mice caused accelerated differentiation of HSCs in vitro and depletion of HSCs in vivo. Interestingly, Notch signalling activity has been demonstrated to be high in HSC and progenitor cells in the bone marrow niche (Duncan et al. 2005). In line with this finding, it has been shown that constitutive Notch-1 signalling in haemopoietic stem cells and progenitors allows the establishment of immortalised cell lines, which retain the capacity to generate either lymphoid or myeloid cells both in vitro and in vivo (Varnum-Finney et al. 2000). Similar findings have been demonstrated in RAG-1 deficient mouse stem cells, where over-expression of Notch-1 promoted stem cell self-renewal over differentiation (Stier et al. 2002). Similarly, constitutively active Notch-4 promotes HSCs self-renewal, while inhibiting differentiation and altering lymphoid development (Vercauteren and Sutherland, 2004; Ye et al. 2004). 40

Studies on Notch ligands have supported the notion of the crucial role of Notch signalling in influencing HSCs fate decisions and suggested the potential use of Notch ligands for ex vivo expansion of HSCs. When cultured with human Jagged-1, human HSCs showed increased survival and expansion potential in vivo (Karanu et al. 2000). A similar effect was reported when mouse Jagged-2 promoted the survival of murine primitive haematopoietic precursors without exogenous cytokines (Tsai et al. 2000). Moreover, the soluble form of human Delta-like-1 suppressed the acquisition of differentiation markers by murine haemopoietic progenitor cells (Lin ) cultured in vitro with cytokines, and promoted the self-renewal of the primitive haemopoietic precursor cells (Han et al. 2000). Taken together, studies on Notch receptors and ligands show that Notch signalling is critical for HSCs self-renewal and survival. The exact mechanisms and pathways by which Notch regulates the developmental fates of HSCs are still a mystery. However, such effects are most likely to be mediated through cross-talk between Notch and various complex signalling pathways, cell cycle modulators and secreted factors in the bone marrow niche. It has been proposed that cytokines may play an important role in the effects of Notch on HSCs (Kojika and Griffin, 2001). Cytokines may modify the Notch- induced self-renewal of HSCs through their interaction with a specific region on Notch 1-3, termed the Notch Cytokine Response region (NCR) (Bigas et al. 1998). Different modulators of Notch may also modify the action of Notch on HSCs fate decisions, in response to various physiological demands. Interestingly, it has been shown that Wnt signalling contributes to the differential expression of known Notch targets in HSCs, and that Wnt and Notch may work together to promote self-renewal of HSCs (Duncan et al. 2005). One critical target molecule of the Notch-1-induced self-renewal of HSCs is the transcription factor cmyc. It has been demonstrated that the expression of c-myc is enhanced during Notch1-induced self-renewal of murine HSCs, and that Notch-1 activates the c-myc promoter directly (Satoh et al. 2004). It is apperant, therefore, that Notch-1 inhibits differentiation and promotes self-renewal of HSCs by up-regulating c-myc,

41

which is known to shorten the G1 phase of cell cycle and induce G1/S transition (Stein et al, 2004).

1.2.5.2. Notch signalling in myeloid development The effect of Notch on the differentiation and proliferation of myeloid cells is still controversial. Various in vitro studies using cell lines support the notion that Notch may inhibit the differentiation of immature myeloid progenitors. For instance, the constitutively activated ICN of mouse Notch-1 inhibited granulocytic differentiation of the myeloid progenitor cell line 32D (Milner et al. 1996). In another study by the same group, the inhibitory effect of Notch-1 and -2 on differentiation of myeloid cells has been shown to be cytokines dependent and that this is controlled through the NCR region of Notch. Notch-1 has been demonstrated to inhibit granulocytic differentiation of 32D myeloid progenitor cells in response to G-CSF, and Notch-2 in response to GM-CSF (Bigas et al. 1998). Interestingly, the expression of constitutively active Notch-4 also inhibited differentiation of human myeloid leukaemia (HL-60) cells, and caused their accumulation in the G0/G1 phases of the cell cycle (Ye et al. 2004). Moreover, in vitro co-culture experiments have shown that the soluble forms of human Notch ligands Jagged-1 and Delta-1 inhibit the differentiation of myeloid progenitors in 32D cells and in mice (Li et al. 1998; Han et al. 2000). Contradictory to the notion that Notch signalling inhibits myeloid differentiation, in vitro and in vivo studies suggest that Notch promotes rather than inhibits, myeloid differentiation. For example, it has been shown that the induction of murine Notch-1 activity in 32D myeloid progenitor cells promotes differentiation (Schroeder and Just, 2000). Schroeder and Just argued that the discrepancy between their results, and those of Milner, was due to the lack of a RAM domain in the construct used by the latter. Tohda et al also found that the immobilised Notch ligand, Delta-1, induced the differentiation of AML cells in two AML cell lines (Tohda et al. 2003). Furthermore, activation of Notch-1 in the FDCP-mix myeloid cell line resulted in the generation of differentiated myeloid cells with loss of self-renewal capacity (Schroeder et al. 2003). The latter study demonstrated the role of the PU.1.transcription factor as a target gene for Notch in its regulatory effect on myeloid cells.

42

Theses in vivo studies lend further support to the concept that Notch signalling in myeloid differentiation is of promoting rather than inhibitory nature. Myeloid differentiation was not shown to be repressed in transduced haemopoietic progenitors that express activated Notch (Ohishi et al. 2003). Several issues should be addressed in order to critically analyse the conflicting outcomes of the in vitro studies of Notch signalling in myeloid development. Firstly, variations in the ICN constructs used in different studies may lead to different outcomes, such as those observed in studies that used 32D cells. Secondly, the inherent differences in Notch receptors and ligand expression and function in isolated cells, or in cell lines used, may have a major impact on the different outcomes of the above studies. Thirdly, soluble ligands of Notch, which were used in studies that suggested an inhibition of differentiation in myeloid cells, may not accurately represent actually the functioning membrane-bound ligands presented by stromal cells. In fact, these forms may act as dominant negative forms as shown in Drosophila (Sun and Artavanis-Tsakonas, 1997). In addition, manipulation of Notch expression, in terms of whether a constitutive or inducible expression is attempted, may influence the differentiation capacity of myeloid cells. It has been shown that some retroviral transfection attempts of producing constitutively activated Notch yielded clones or mutants that lacked differentiation instructive potential (Schroeder et al. 2003). It has been argued that it is likely that such mutants may have been used in some studies that suggest a Notchinduced block of myeloid differentiation. Finally, the nature of the outcomes of Notch signalling in myeloid development may depend on the cellular context, such as presence or absence of cytokines and other signalling molecules in the bone marrow niche, and therefore it is possible that Notch mediates different cellular outcomes in different cellular contexts. Notch signalling has been investigated in monocytes and it has been found that immobilised Delta-1 induced apoptosis in monocytes in response to M-CSF, and inhibited monocytes from differentiation into macrophages in response to GM-CSF (Ohishi et al. 2003). In the same study, Delta-1 promoted the differentiation of 43

monocytes into dendritic cells in response to GM-CSF and IL-4. Notch signalling has also been found to regulate myeloid differentiation along erythroid and megakaryocytic lineages. It has been reported that Notch-1 inhibits the differentiation of erythroid/megakaryocytic cells by inhibiting GATA-1 activity in the erythroid/megakaryocytic cell line K562 (Ishiko et al. 2005). However, only erythroid differentiation has been shown to be inhibited in K562 cells by activated Notch (Lam et al. 2004). Further studies with careful experimental designs, which take into consideration all of the possible sources of discrepancies mentioned above, are needed in order to reach a final model for the role of Notch in myeloid cell development.

1.2.5.3 Notch signalling in lymphoid cell development Lymphoid development is a highly regulated process in which functional lymphocytes are produced from common lymphoid progenitors (CLP). Development of functional mature lymphocytes from CLP is a finely regulated, stepwise process, that depends on the expression of different transcription factors. Studies in loss and gain of function, suggest that Notch signalling is indispensable for developmental decisions of lymphoid cells at different stages of maturation (Fig. 1.6.). Radtke et al provided the first evidence that Notch signalling regulates B Vs T lineage specification (Radtke et al. 1999). Radtke et al used transgenic mice expressing a conditional Notch1 knockout allele to demonstrate that loss of Notch-1 caused a block in T-cell development, and promoted B-cell development that derive from thymic precursors. The block in T cell development has been found to occur at or before the earliest

intrathymic

precursor

stage

(defined

as

lineage

negative

CD44+CD25−CD117+) (Wilson et al. 2001). Gain of function studies lent further support to these findings, in which enforced expression of constitutively active Notch1, (ICN1) in murine HSC, led to ectopic T cell development in the bone marrow that was thymus independent, while inhibiting B-cell development at the earliest stages (Pui et al. 1999). Collectively, these studies suggest that at the level of CLPs, Notch-1 signalling has to be kept inactive in the BM compartment to allow B cell development, and to inhibit ectopic thymic-independent

44

T cell development. At the same time Notch1signalling is necessary and sufficient for T cell fate specification once a CLP enters the thymus (Pear et al. 2003). Since Notch receptors and ligands are expressed in normal bone marrow, certain regulatory mechanisms must exist to modulate Notch signalling and allow normal B cell development in the bone marrow (Radtke et al. 2004). For instance, it has been shown that the B lineage commitment factor, Pax5, inhibits transcription of Notch1, providing a possible mechanism that allows B cell development in the BM, despite expression of Notch-1. Other possible mechanisms that antagonise Notch signalling and allow B cell development in the bone marrow may include Notch inhibitory modulators such as Fringe (Koch et al. 2001) and Deltex1 (Izon et al. 2002), as demonstrated by enforced expression studies. The Notch signalling effects on T cell development have been shown to be mediated by the CSL transcription factor since an inducible deletion of CSL produced a phenotype which is similar to that shown in Notch-1 conditional knockout mice (Han et al. 2002). However, the molecular mechanisms by which Notch influences lymphoid commitment remain largely unknown. One possible mechanism is that Notch blocks B cell commitment through the inhibition of the E47 function, which is the gene product of E2A. E2A is an important transcription factor during early stages of B cell development (Pui et al. 1999; Kojika and Griffen, 2001). Furthermore, it has been suggested that Notch-1 may promote T-cell development by upregulating expression of T-cell specific genes, such as pre T-cell receptor ⍺ (pT⍺), which encodes a critical component of pre-TCR (Reizis and Leder, 2002). The involvement of Notch signalling in the T cell developmental decision of adopting either ⍺β or γδ T cell lineage is still controversial. The first model of a Notch-1 mediated effect at the ⍺β versus γδ maturation choice, proposed that Notch1 signalling promotes ⍺β T cell development at the expense of γδ T cell development. This notion stems from a study in which BM precursors with only one functional Notch1 allele (Notch1+/‫ )ــ‬give rise to relatively more γδ than ⍺β T cells, compared to wild-type precursors in chimeric mice, reconstituted with a mixture of Notch1+/+ and Notch1+/‫ ــ‬BM-derived cells (Washburn et al. 1997). The other model of Notch in ⍺β

45

versus γδ lineage commitment argues that Notch-1 signalling may promote ⍺β T cell development but it does not influence γδ T cell development. This was evidenced by a study in which inactivation of Notch1 gene in the thymus, before pre-TCR expression, severely impaired ⍺β but not γδ T cell development (Wolfer et al. 2002). The role of Notch signalling in the CD4/CD8 fate choice remains largely unresolved. Initially, it has been proposed that Notch1 promotes the development of CD8 + T cells at the expense of CD4+ cells (Robey et al. 1996). Another group who used transgenic mice expressing slightly longer form the ICN reported maturation of both CD4+ and CD8 + T cells (Deftos et al. 2000). A more recent study showed that both transgenic mice used by the two groups display a decrease in mature CD4+ T-cells and an increase in mature CD8+ T cells, suggesting that Notch1 signalling does indeed influence the CD8 versus CD4 lineage choice (Fowlkes and Robey, 2002). To make things more complicated, loss of functions experiments in which the Notch1 gene was inactivated in mice, did not show any developmental skewing toward CD4+ T cells, suggesting that Notch1 is dispensable for the CD4/CD8 lineage decision (Wolfer et al. 2001). Whether the CD4/CD8 lineage choice is regulated in normal lymphopoeisis by other Notch receptors in a redundant fashion, remains to be confirmed in order to validate the loss of function experimental findings. Collectively, there is no consensus as yet on the Notch-1 instructive role in the CD4/CD8 decision. Notch-3 signalling has been postulated to be involved in different peripheral T cell functions such as the regulation and expansion of CD4+ CD25+ regulatory T cells and the promotion of Th1 differentiation from CD4+ T cells in response to antigen stimulation (Radtke et al. 2004). As for the possible functions of Notch signalling in B cell development, it is well documented, as explained above, that lack of Notch-1 signalling promotes B cell development in the bone marrow at early stages of B cell lymphopoiesis. The other possible role of Notch in B cell development is the stimulation of differentiation of marginal zone B cells (MZB) in the spleen, as demonstrated in an RBP-J knockout study (Tanigaki et al. 2002). Since Notch-2 is the most highly expressed Notch receptor in B cells, it has been speculated that Notch-2 is a likely candidate for the Notch-induced marginal zone B cell effect (Pear and 46

Radtke, 2003). This was confirmed by a study in which conditional inactivation of Notch-2 in the BM resulted in the loss of marginal zone B cells without affecting T cell development (Saito et al. 2003). Several questions regarding the role of Notch signalling in lymphoid development are still to be answered. For example, what are the downstream target genes that mediate the effects of different Notch receptors and what is the role of different Notch modulators in haemopoiesis. Of importance also is which ligand specifically triggers different Notch functions in vivo and whether Notch receptors operate in the haempoietic compartment in a redundant fashion. Finally, the interactions between Notch signalling and other signalling pathways, such as NF-κ B and the ras/MAPK pathway, during T cell development are not well established and might be a focus for future investigations (Allman et al. 2002).

1.2.6 Notch signalling and cancer Although many aspects of the involvement of Notch in developmental biology have been revealed during the last decade, much less is known about the involvement of Notch signalling in human diseases, and particularly in the process of malignant transformation. Notch-3 alterations have been associated with non-malignant human diseases such as ‘cerebral

autosomal

dominant

arteriopathy

with

subcortical

infarcts

and

leukoencephalopathy’ (CADASIL) syndrome, a neurodegenerative disease (Joutel et al. 2000). However, the first example of the involvement of Notch in malignant transformation in humans was described in a subset of T-cell acute lymphoblastic leukaemia (T-ALL), carrying the t(7;9) (q34;q34.3) translocation in which a constitutive expression of ICN was involved in the leukaemogenesis process (Ellisen et al. 1991). Direct proof of oncogenic potential of activated Notch-1 was obtained in bone marrow transplant assay (BMT), in which retroviral expression of activated ICN1 in HSCs induced T-ALL in mice (Pear et al. 1996). Similarly, overexpression of ICN domain of Notch-3 induces T-cell leukaemias (Bellavia et al. 2002). Jundt et al (2002) have reported high expression of Notch1 in

47

Hodgkin and anaplastic large cell lymphoma. Notch oncogenic activity in nonhaematological malignancies has also been reported for example, in breast cancer in mice, in which N4 was involved (Callahan and Raafat, 2001). Other reports have demonstrated the involvement of Notch1 signalling in breast cancer in human (Weijzen et al. 2002) and in human cervical cancer (Talora et al. 2002). Interestingly, Notch deficiency, rather than activation, can also contribute to cancer development. For example, it has been shown that Notch1 loss of function resulted in basal-cell-like carcinomas, or squamous cell carcinomas, in mice (Nicolas et al. 2003). In most cellular contexts of Notch-induced tumourigenesis, altered Notch acts as an oncoprotein that exhibits oncogenic functions such as those discussed in Notchmediated T-ALL leukaemias. However, various lines of evidence suggest that Notch may also act as a tumour suppressor, or may exhibit both oncogenic and tumour suppressive potentials depending on the cellular context (Radtke and Raj, 2003).

1.2.6.1 Notch signalling in leukaemia The longest established role of Notch signalling in leukaemia is that of Notch1 and TALL characterised by a t(7;9) (q34;q34.3) chromosomal translocation. As the gene at the chromosome 7 locus, that is fused to the TCR β promoter/enhancer, is very similar to Drosophila Notch, it was named TAN1 for ‘translocation-associated Notch homologue’, and subsequently became known as human Notch1 (Radtke and Raj, 2003). TAN-1 is a truncated Notch1 molecule that encodes a dysregulated, constitutively active intracellular domain (ICN-1) (Ellisen et al. 1991). Although the complete in vivo molecular mechanism by which ICN1 transforms haemopoietic progenitor cells is not well established, it has been found that ICN1-mediated oncogenic function in t(7;9) T-ALL is dependent on a second T-cell-specific signal that is mediated by the pre-TCR (Allman et al. 2001). The leukaemogenesis potential of ICN1 was further investigated in bone marrow transplant (BMT) reconstitution models, in which retrovirally transduced HSCs were transferred to lethally irradiated mice, and constitutive expression of the human ICN1 led exclusively to CD8+CD24+ (immature single positive, ISP) or CD4+CD8+ double positive (DP) T cell leukaemia/lymphomas, with simultaneous inhibition (Zweidler-McKay and Pear, 2004).

48

of B-cell development

Despite the ability of activated Notch1 to induce T-cell leukaemia in mice, less than 1% of human T-ALLs exhibit the t(7;9) translocation. However, activating mutations in NOTCH1 independent of t(7;9) have been identified in the heterodimerization (HD) and PEST domains of Notch1 in most Notch-dependent T-ALL cell lines, and in approximately 55% of primary T-ALLs (Weng et al. 2004). Activating mutations in Notch1 have since been reported in mouse models of T-ALL (O'Neil et al. 2006) The importance of Notch signalling in T-ALL has been further elucidated by the finding that Notch3 is expressed in almost all T-ALL cases in humans (Bellavia et al. 2002). In this study, Notch3 was consistently expressed in a sample of 30 human T cell acute leukaemias, and the expression was dramatically reduced or absent in patients in clinical remission. Furthermore, Notch-3 expression in those patients was associated with the expression of its target gene, HES1, and of the gene encoding pTα. The combined expression of the genes encoding Notch3, pTα and HES1 in human T-ALL suggests that a signalling defect at a specific stage in Tcell development, the pre-TCR checkpoint, is responsible for T-cell leukaemogenesis (Screpanti et al. 2003). It has been suggested that the altered Notch-3 signalling disrupts the normal interaction between pre-TCR signalling and NF-κB signalling in T-cell development. This in turns is thought to lead to the disruption of differentiation of early thymocytes and results in the development of T cell leukaemia (Bellavia et al. 2003). The precise mechanisms by which Notch contributes to T-cell leukaemias are not fully understood. However, it is postulated that several signal-transduction pathways might co-operate in Notch-induced leukaemogenesis.

For instance, pre-TCR

signalling has been shown to be essential for Notch-1 and Notch-3 induced leukaemogenesis (Allman et al. 2001; Bellavia et al. 2002). However, whether pre-Tα is a direct Notch target is controversial (Zweidler-McKay and Pear, 2004). It has been proposed that pre-TCR signalling may promote Notch-induced leukaemia through inhibition of E2A, a gene that is critical in T and B cell development and which acts as a tumour suppressor (Bellavia et al. 2003). Another pathway that may co-operate with Notch signalling in T-ALL leukaemogenesis is the NFκB pathway. This notion was supported by the findings that Notch3-IC transgenic mice exhibited constitutive 49

NFκB activity (Bellavia et al. 2000), and that truncated Notch-1 expression resulted in up-regulation of NFκB2 in T cells (Oswald et al. 1998). Collectively, Notch signalling may contribute to T-ALL leukaemogenesis through its co-operation with pre-TCR signalling to inhibit the E2A tumour suppression gene and by activation of NFκB, which regulates apoptosis and proliferation of T-cells.

Fig. 1.6 Notch signalling during T and B cell development. Notch signalling has been shown to promote T cell over B cell commitment and favour the commitment towards the ⍺β lineage. Signalling through the Notch1 receptor has been shown to be involved in regulating V-DJ rearrangement of the TCRβ locus. Finally Notch signalling has been proposed to influence lineage decisions when DP (CD4+CD8+) thymocytes must choose between the CD4+ (CD4+CD8−) and the CD8+ (CD4−CD8+) cell fates. Notch signalling should be kept inactive in the B cell progentitors in the bone marrow to allow B cell development (Adopted from Pear and Radtke, 2003).

The involvement of Notch signalling in B-cell malignancies has been suggested in many B-cell neoplasms such as B-CLL (Hubmann et al. 2002; Duechler et al. 2005), Hodgkin’s disease and anaplastic large cell lymphoma (Jundt et al. 2002), and in multiple myeloma (Jundt et al. 2004). The Hubmann group found that Notch-2 is 50

overexpressed in B-CLL cases, and may be involved in the regulation and overexpression of CD23, a hallmark of B-cell chronic lymphocytic leukaemia (BCLL) cells which is linked to the failure of apoptosis in B-CLL cells (Hubmann et al. 2002). This data was further confirmed in a recent study by the same group in which the induction of apoptosis by proteasome inhibitors in B-CLL cells was associated with down-regulation of Notch-2 and CD23 expression (Duechler et al. 2005). Moreover, Jundt et al (2002), using cell lines and primary cells, have demonstrated that Notch-1 is highly expressed in B- and T-cell derived tumours of Hodgkin’s (HD) and anaplastic large cell lymphoma (ALCL). In this study, mRNA expression of the Notch1 ligand, Jagged1, was highly expressed in neighbouring cells of Hodgkin’s and Reed-Sternberg cells in vivo, which suggest that Jagged1-induced Notch-1 signalling might contribute to the pathobiology of HD. This notion was further supported by the overexpression of Hes-1, target gene of Notch signalling, following in vitro culture of HRS and ALCL cells in the presence of Jagged1 (Jundt et al. 2002). Notch signalling has also been proposed to be involved in the pathogenesis of multiple myeloma (MM). Jundt et al have demonstrated that Notch receptors and their ligand, Jagged1, are highly expressed in cultured and primary MM cells and that Notch signalling promotes proliferation of myeloma cells (Jundt et al. 2004). Similarly, Notch1-4 have been found to be expressed by myeloma cells in different MM cell lines (Nefedova et al. 2004). Upon ligand activation, only Notch-1 signalling in myeloma cells, protected cells from drug-induced apoptosis, by inhibiting their entry to the cell cycle. Whether Notch signalling contributes to myelomagenesis by inducing proliferation of MM cells (Jundt et al. 2004) or by inhibiting growth of MM cells and inducing anti-apoptotic mechanisms (Nefedova et al. 2004), seems to be dependent on the cellular context and availability or absence of toxic agents. The role of Notch signalling in myeloid leukaemias is not yet known. However, high expression of Jagged1 has been reported recently in 20 primary AML samples (Chiaramonte et al. 2004). Jagged1 expression in AML samples was significantly higher than its expression in the T-ALL or B-ALL patients. In light of the finding that only low levels of Notch1 and its target genes were detected in this study, a possible autonomous role of Jagged1 in supporting AML growth has been proposed. This is in line with the finding that expression of acute myeloid leukaemia (AML) PML/RAR 51

and AML1/ETO fusion proteins results in activation of Jagged1/Notch signalling in blasts derived from AML patients (Alcalay et al. 2003). This has been suggested to confer self-renewal properties to leukaemic blasts in AML. Interestingly, the gene expression profiling of stem cells in myelodysplastic syndrome (MDS), AML, and CML has revealed a selective expression of the gene encoding Delta-like Notch ligand in MDS patients (Miyazato et al. 2001).

1.3 Chronic myeloid leukaemia (CML) CML results from the malignant transformation of a haemopoietic stem cell. This myeloproliferative disease, which accounts for 10-20% of chronic leukaemias, is characterised by a t(9,22) reciprocal chromosomal translocation, generating the Philadelphia (Ph) chromosome in more than 90% of CML patients. The disease can be divided clinically into three phases, a chronic phase, an accelerated phase, and a terminal blastic phase. Clinically, chronic phase disease is characterised by splenomegaly and high white cell count of mainly myeloid lineage cells with normal differentiation. If untreated, the disease progresses gradually to the accelerated phase until it becomes more refractory to treatment. Progression to the blastic phase, which is an acute leukaemia, then occurs in a few months where the more differentiated marrow cells are displaced by 30% or more immature blasts of either myeloid or lymphoid origin (Pallister, 1998). During the chronic phase, which lasts about 3-5 years, the only chromosomal abnormality present on leukaemic stem cells and myeloid progenitors is the t(9;22) (q34;q11). However, the progression into the accelerated and blastic phases is accompanied by the acquisition of additional genetic abnormalities in most cases, which may involve the loss of tumour suppressors or the activation of many protooncogenes in the bone marrow microenvironment (Ren, 2005). The t(9;22) chromosomal translocation characteristic of CML results in a fused 8.5 kb BCR-ABL gene and an abnormal fusion protein, p210 BCR/ABL. The fusion of BCR sequences to ABL during the t(9,22) translocation, increases the tyrosine activity of ABL and brings new domains, highly critical for oncogenic activities of BCR-ABL, such as the growth factor receptor-bound protein 2 (GRB2) SH2 binding site (Ren, 52

2005). In contrast to the native c-ABL which shuttles between the nucleus and the cytoplasm, the p210 BCR-ABL is exclusively located in the cytoplasm (Marley and Gordon, 2005). It has been shown that cytoplasmic localisation of BCR-ABL is vital for avoiding apoptosis (Melo, 2001). The new fusion protein has five-fold higher tyrosine kinase activity than the normal cABL protein, an activity that has been shown to be essential for its transforming potential (Clarkson et al. 1997). In addition, the new location of the constitutively active ABL kinase in BCR-ABL oncoprotein may provide access to novel substrates and interactions. It appears that the BCR-ABL fusion protein binds various substrates in the cytoplasm to activate various signalling pathways in CML stem cells and primitive progenitors and co-operates with cytokines to induce self-renewal, proliferation and survival of leukaemic stem cells (Clarkson et al. 2003).

1.3.1 Molecular phenotype of BCR-ABL The BCR-ABL fusion protein can vary in size from 190 to 230 kD, depending on the breakpoint in the BCR gene. Splicing at the M, m, and µ breakpoints in BCR produces three BCR-ABL variants which are P190 BCR-ABL (e1a2 junction), p210 BCR-ABL (b2a2 or b2a3 junctions), and p230 BCR-ABL (e19a2 junction) (Fig. 1.7). Most patients with chronic-phase CML express a 210-kD protein which is also found in about 20% of Philadelphia positive acute lymphoblastic leukaemia (ALL) patients.

A

Very few CML patients express the 230-kD protein which is associated with a very mild form of CML, denominated Ph-positive neutrophilic CML (Ph+ N-CML). The P190 BCR-ABL protein is associated with most Philadelphia positive (ALL) patients (Kantarjian et al. 2006).

53

Fig. 1.7 The t(9;22)(q34;q11) reciprocal translocation. (A) The t(9;22) translocation results in the formation of a shortened chromosome 22 (the Philadelphia chromosome) carrying the BCR–ABL fusion gene. In addition, the translocation results in a longer chromosome 9 that carries the ABL–BCR fusion gene. The fusion of BCR sequences to ABL during the t(9,22) translocation, increases the tyrosine activity of ABL and leads to constitutive tyrosine kinase activity in the BCR-ABL protein but not in the ABL-BCR protein. (B) Locations of the breakpoints in the ABL and BCR genes. Exons are shown as boxes, and breakpoints are indicated by arrows. Splicing at the m, M, or µ breakpoints in BCR produces three distinc proteins. These three BCR-ABL variants are named P190 (e1a2 junction), P210 (b2a2 or b3a2 junction), and P230 (e19a2 junction). Most CML patients express the P210 BCR-ABL. (Taken from Smith et al (2003) and Inokuchi (2006)).

1.3.2 BCR-ABL oncogenic activities BCR-ABL induce malignant transformation in Ph+ cells via three major mechanisms: altering adhesion to stromal cells and extra-cellular matrix, inhibiting apoptosis, and activating signalling pathways with mitogenic potentials such as RAS and MAP kinase pathways (Deininger et al. 2000). Depending on the cellular context, BCR-

54

ABL can bind adaptor proteins and phosphorylate substrate molecules and signalling proteins which have different physiological functions in the cytoplasm to exhibit its oncogenic activities (Fig 1.8). These substrates of BCR-ABL can be grouped into adapter molecules including (such as GRB2 and CRKL) , proteins with catalytic functions (such as the nonreceptor tyrosine kinase Fes or the phosphatase Syp), and proteins associated with organisation of the cytoskeleton and cell membrane (such as paxillin and talin) (Ren, 2005; and Deininger et al. 2000).

1.3.2.1 Altered adhesion Normal BM progenitors adhere to stroma through a variety of cell surface adhesion receptors, including the α4β1 and α5β1integrin receptors which bind to cell adhesion molecules on the stroma. Interaction between integrins and the cytoskeleton plays a critical role in modulating integrin function both by affecting receptor conformation and ligand binding affinity (Schwartz et al. 1995). Gordon et al. (1989) showed that CML progenitors fail to adhere to BM stroma. Salesse and Verfaillie (2002) have shown the presence of abnormal association between the α4β1 and α5β1integrin receptors and the cytoskeleton proteins which impairs the normal adhesion function of β1integrins. The authors demonstrated in a human CML model that the p210 BCRABL is directly responsible for the defect in adhesion in CML progenitors. BCR-ABL localisation in the cytoplasm results in increased binding to actin and phosphorylation of a number of neighboring cytoskeletal proteins including FAK and paxillin which may alter normal integrin signalling and contribute to abnormal adhesion receptor function (Shet et al. 2002). Because CML cells exhibit reduced adhesion to fibronectin and bone marrow stroma cells they escape the integrinmediated proliferation control, and enjoy high proliferation potential (Salesse & Verfaillie, 2002). In addition, some populations of CML progenitors but not normal progenitors express α2β1 and α6β1 integrin receptors that interact with basement membrane components, lamin and collagens (Lundell et al. 1997). These abnormal findings in the function and expression of certain cell surface adhesion molecules, may explain the premature release of massively expanded myeloid progenitors in the blood of CML patients.

55

1.3.2.2 Inhibition of apoptosis Inhibition of apoptosis is a feature of CML progenitors. This has been demonstrated in haematopoietic cell lines and murine bone marrow cells (Kabarowski & Witte, 2000). BCR-ABL may inhibit apoptosis by inhibiting the activation of caspases by blocking the release of cytochrome C from the mitochondria (Amarante-Mendes et al. 1998). In addition, BCR-ABL has been shown to up-regulate the anti apoptotic proteins Bcl-2 and BclxL (Deininger et al. 2000). Moreover, BCR-ABL may inhibit apoptosis through the phosphorylation of the pro-apoptotic protein Bad (Neshat et al. 2000) and the down-regulation of ICSBP tumour suppressor protein (Hao et al. 2000) (Fig. 1.8).

1.3.2.3 Proliferative signals In addition to altering adhesion and inhibiting apoptosis, BCR-ABL activates various signalling pathways that contribute to proliferation of CML cells. This may confer proliferative capacity, independent of cytokines requirements for growth and survival in CML cells. It has been shown that BCR-ABL, through specific functional domains, interacts with signalling proteins which in turn activate downstream signalling pathways including the RAS, MAPK, JAK-Stat, PI3 kinase, and Myc pathways. For example, activation of RAS has been shown to occur through autophosphorylation of the tyrosine 177 domain of BCR-ABL which provides a docking site for the adapter molecule GRB-2, which then binds to SOS protein which in turn activates RAS signalling pathway (Fig. 1.8) (Deininger et al. 2000). Stat1 and Stat5 transcription factors, components of Jak-Stat pathway, have been shown to be constitutively phosphorylated in many BCR-ABL positive cell lines and primary CML cells (Chai et al, 1997). In addition, BCR-ABL has been shown to bind and phosphorylate the GAB2 protein, which then recruits and activates phosphatidylinositol 3-kinase (PI3K) signalling pathway (Sattler et al, 2002).

56

In addition, BCR-ABL, promotes proliferation and survival by inducing expression of cytokines such as Interleukin-3 (IL3), G-CSF AND GM-CSF, and by downregulating transcription factors that inhibit proliferation and survival such as ICSBP and JUNB (Ren, 2005).

1.3.2.4 Role of CrKl in BCR-ABL signalling The adaptor protein Crkl is the most prominent tyrosine-phosphorylated proteins in CML and appears to play a key role in mediating the oncogenic activities of the BCRABL oncoprotein (Oda et al. 1994; Singer et al. 2006). BCR-ABL binds and phosphorylates CrKl directly through the SH3 domain. The phosphorylated CrKl (Pcrkl) interacts with specific target proteins and mediate the formation of signal transduction pathways. For example, the BCR–ABL-dependent activation of the PI3K pathway has been shown to be mediated by BCR–ABL interaction with Crkl (Sattler et al. 1996a). CrkL has also been found to be the linking protein between Bcr-Abl and Stat signaling (Rhodes et al. 2000). In addition, Crkl, can also activate the Ras signalling pathway in fibroblasts as well as in haemopoietic cells (Deininger et al. 2000; and Arai et al. 2002). CrKL is also involved in the regulation of cellular motility of CML cells and in integrin-mediated cell adhesion by association with other cytoskeleton proteins such as paxillin, the focal adhesion kinase Fak (Sattler et al. 1996b). Interstingly, it has been shown that that tyrosine-phosphorylation of Crkl is a direct consequence of BCR-ABL expression and that phosphorylation of Crkl could be used as a diagnostic indicator for BCR-ABL activity in Ph+ leukaemia (ten Hoeve et al. 1994).

1.3.3 Leukaemic stem cells (LSC) in CML Various studies have shown that CML is a clonal disease of haemopoietic stem cell origin. The Ph chromosome and the BCR-ABL transcript have been detected in all haematopoietic lineages, except natural killer cells (Tahakashi et al, 1998). The leukaemic stem cells (LSC) in CML are very primitive cells that are Ph+, BCR-ABL+ and can be identified in vitro by long-term culture-initiating cells (LTC-IC) assay. It has been shown that about 80% of Ph+ LTC-IC cell in CML are positive for

57

CD34 and Thy-1 cell surface markers (Petzer et al. 1996). Moreover, the LSC in CML were found to have the phenotype of CD34+ CD38-, a phenotype similar to normal HSCs (Petzer and Gunsilius, 2003). Therefore, it is acceptable that in CML patients in which the circulating LTC-IC population is mainly Ph+, the CD34+CD38- or CD34+ Thy-1 + populations are highly enriched with leukaemic stem cells (LSC). In addition to being CD34+ CD38- Thy-1 +, LSCs in CML are highly enriched in the CD34+ HLA-DR+ population (Verfaillie et al, 1992). This is in line with the finding that CD34+ HLA-DR− cells in CML are polyclonal (Delfroge et al, 1999). There is evidence that normal haemopoietic stem cells (HSCs) are relatively well preserved in newly diagnosed CML patients, but tend to rapidly decline with time (Frassoni et al, 1999). However, the leukaemic stem cells (LSCs) seem to be more predominant than normal HSCs in CD34+ CD38- / CD34+ Thy-1 + cell populations in chronic phase of CML. Maguer-Satta et al (1996) demonstrated the presence of BCR-ABL mRNA in about 80% of the CD34+ CD38- cells in patients with chronic phase CML. In line with this, Grand et al (1997) found that the majority of CD34+ CD38- cells were Ph+ and express BCR-ABL transcript. Similar findings were reported in five (out of nine) CML patients in presentation (Holyoake et al. 2001). Recently, FISH analysis of 10 CML chronic phase patients showed that the majority of the primitive CD34+ CD38- cells, both before and following IM exposure, were BCR-ABL positive (Copland et al, 2006). Contradictory to previous

reports,

others

found

that

normal HSCs

frequently

outnumber the LSCs in the chronic phase of CML (Dube et al, 1984). Interestingly, it has been shown that CML stem cells can be traced at least to haemangioblast like cells, earlier than the pluripotent HSCs (Fang et al, 2005).

58

SOS

G GRB2 P

BCR RAS

Figure 1.8.Signal transduction pathways associated with P210 BCR-ABL in CML. Tyrosine phosphorylation of BCR–ABL substrates results in the consequent activation of multiple signal transduction pathways. Ras activation in is mediated by BCR–ABL interaction with the adaptor molecules Grb2 or the adaptor protein Crkl. Activation of signalling pathwas downstream of the RAS pathway results in promoting proliferation and transformation of BCR-ABL+ cells. Proliferative signals are also gained by the BCR-ABL induced expression of IL3 and GM-CSF cytokines. BCR-ABL has also been shown to bind and phosphorylate the GAB2 protein, which then recruits and activates the (PI3K) signalling pathway which interacts with downstream pathways including AKT and NFkB to inhibit apoptosis. Protection from apoptosis can also occur through the activation of STAT signalling pathway via interaction with Crkl adaptor or following direct phosphorylation of STAT proteins by BCR-ABL. Direct or Crkl mediated binding of BCR-ABL to actin and phosphorylation of a number of neighboring cytoskeletal proteins including FAK and paxillin alter normal adhesion of BCR-ABL+ cells to the stroma.

MAPK JNK

RAF

59

ST

ERK

1.3.4 Imatinib mesylate Imatinib mesyalyte (also known as STI-571 or Gleevec) was discovered in 1996 as a small molecule that specifically inhibits few kinases including BCR-ABL, c-Kit, and platelet growth factor receptor (Druker et al. 1996). Imatinib mesylate is an ATP competitive inhibitor and therefore it selectively inhibits BCR-ABL tyrosine activity by occupying the ATP-binding site in the kinase domain of ABL, thereby maintaining the protein in inactive conformation (Nagar, 2007). Because the kinase domain is similar in wild type c-ABL and BCR-ABL, imatinib may be expected to inhibit cABL as well. However, the inhibition of normal c-ABL function was only reported in cardiomyocytes (Kerkela et al. 2006). In 2001 the United States Food and Drug Administration (FDA) approved the drug for the treatment of Philadelphia chromosome positive chronic phase CML (400 mg/d) and blastic phase CML (600 mg/d) after failure of interferon-α therapy. Clinical trials showed that imatinib was highly effective in newly diagnosed chronic phase CML patients, in which the drug induced greater than 90% haematologic response and greater than 80% cytogenetic response. However, CML patients in accelerated and blastic phases showed less sensitivity to imatinib (Druker et al. 2006). Although most patients in chronic phase achieved haematological and cytogentic remission, minimal residual disease could be detected in most patients by sensitive real time PCR (Jorgensen and Holyoake, 2007). Bhatia et al (2003) showed the persistence of about 20% leukaemic stem cells that were CD34+ BCR-ABL+ as well as LTC-ICs in patients who achieved complete cytogenic response. Copland et al (2006) showed that the more primitive CD34+ CD38- cells in chronic phase CML patients are resistant to imatinib, in vitro. Taken together, these findings show that although the majority of CML cells in chronic phase respond very well to imatinib, the rare primitive leukaemic stem cells that maintain the disease remain insensitive to the drug.

60

1.3.5 Experimental models of CML 1.3.5.1 Cell lines Fibrolast lines and haemopoietic cell lines have been extensively used to study the biology of CML. Although expression of BCR-ABL has been shown to transform mouse fibroblast cell lines and reproduced many of the properties of CML cells, the biological effects are diverse, depending on the type of fibroblasts used. In addition, the interaction between certain BCR-ABL domains in transformed fibroblasts and other signalling proteins does not represent a good model for CML disease. This is simply because certain BCR-ABL domains and signalling proteins are functionally vital in fibroblast transformation, but not in haemopoietic cells (Deininger et al. 2000). Therefore, BCR-ABL+ haemopoietic cell lines such as K562 and BV173 may provide a CML model that overcomes the limitations of fibroblast cell lines. However, one limitation of haemopoietic CML cell lines is that they are derived from blast crisis and thus, are not ideal models for the chronic phase of CML. The importance of BCR-ABL+ cell lines remains that they contributed to our understanding of the basic biology of CML, and that the BCR-ABL tyrosine kinase activity can be turned off with imatinib mesylate (STI571 or Gleevec) to study activity of other signalling pathways in CML. 1.3.5.2 Animal models Animal models represent better physiological systems than cell lines for the study of CML molecular biology because they provide the opportunity to study the oncogenic activities of BCR-ABL+ haemopoietic cells within their normal niche. The most commonly used animal models are mice with haemopoietic cells that express BCRABL through various approaches such as transgenic, knock-in or retroviral transduction techniques. Engraftment of BCR-ABL transformed cell lines in syngeneic mice produced a form of acute leukaemia and does not provide a suitable model for chronic phase CML. Similar results were obtained in transgenic mouse models before the use of the Tec

61

promoter (specific for haemopoietic cells) which was able to target the expression in the appropriate cells (Honda et al. 1998). An interesting transgenic mouse model of P210 BCR-ABL, under control of tetracycline, recently provided evidence that BCRABL is required for both initiation and maintenance of leukaemia (Huettner et al. 2000). Transplantation of non-obese diabetes, severe combined immunodeficiency (NODSCID) mice with large inoculum of chronic phase human BCR-ABL+ cells has been shown to provide an excellent model to study certain aspects of CML biology (Deininger et al. 2000). Transduction of murine bone marrow cells with BCR-ABL retroviruses has also been used as a CML model since 1990. A major improvement to this system was the use of murine stem-cell retroviral vector to express the BCR-ABL oncogene in haemopoietic cells which produced a myeloproliferative disease (MPD) that was similar to the chronic phase of human CML with high efficiency (Ren, 2005). The problems with some animal models are that they either produced other haemopoietic neoplasms in mice or failed to yield a similar disease at frequency sufficient to utilise it in the study of CML pathogenesis (Petzer and Gunsilius, 2003). Another limitation in CML animal models is the difficulty of ruling out disease modification by host factors (Deininger et al. 2000).

1.4 Possible role for Notch in CML CML is a stem cell disease and the differentiated cells in CML constitute the bulk of leukaemic cell mass whereas the leukaemic stem cells responsible for the disease maintenance are, like normal HSCs, very rare. It has been shown that imatinib mesylate (also known as STI571, or gleevec) is highly toxic to the more differentiated CML progenitors but not to the leukaemic stem cells which remain viable in a quiescent state, even in the presence of growth factors and gleevec (Graham et al. 2002). Therefore, it is possible that CML stem cells’ survival and self-renewal capacities are related to the same signalling pathways that regulate these potentials in normal HSCs such as Notch and Wnt signalling pathways.

62

Jamieson et al (2004) have shed new light on the interaction between leukaemic stem cells in CML and Wnt signalling pathways, which is important for normal HSC selfrenewal, as discussed above (Ryea et al, 2003). Granulocyte-macrophage progenitors (GMP) from patients with CML in blast crisis have been shown to have elevated levels of β -catenin, the major effector of Wnt signalling pathway, which resulted in enhanced self-renewal capacities of GMP and thus acquisition of stem cell phenotype (Jamieson et al. 2004). Notch signalling integrates with Wnt signalling pathway to confer self-renewal capacities to HSCs, since intact Notch signalling was required for Wnt-mediated maintenance of undifferentiated HSCs (Duncan et al. 2005). Moreover, fusion oncoproteins PML/RAR and AML1/ETO in AML have been associated with activation of Notch signalling which may confer self-renewal properties to leukaemic stem cells in AML (Alcalay et al. 2003). Transcriptional targets of Notch signalling, such as c-myc, which is essential in Notch-mediated self-renewal potential of HSC (Satoh et al. 2004) and for Notch1 oncogenic role in T-ALL (Girard et al. 1996), also play a critical role in the malignant transformation mediated by ABL in CML (Afar et al. 1994). Furthermore, signalling pathways such as RAS, which is involved in the transformation process in CML and activated by BCR-ABL fusion protein (Ren, 2005), has been shown to activate Notch signalling (Weijzen et al. 2002). Taken together, Notch may also be involved in the self-renewal potentials of leukaemic stem cells in CML. Studies on axons development in Drosophila support the hypothesis of possible cooperation between ABL protein kinase and Notch signalling. It has been found that Notch interacts genetically with ABL as Notch, and ABL mutations synergise to cause synthetic lethality in Drosophila axons (Ginger, 1998). Interestingly, Ginger has demonstrated that Disabled (Dab) interacts physically to the RAM region of the intracellular domain of Notch in vitro. Given the fact that Dab interacts genetically and physically with ABL kinase, it appears that Dab acts as an adaptor in the cytoplasm between Notch and ABL in response to a signal from Notch ligands (Ginger, 1998). In another study, Ginger and colleagues have found that Delta ligand

63

and Notch provides a guidance signal to the developing axon by regulating the ABL kinase signalling pathway (Crowner et al. 2003). The fact that Disabled protein has been shown to interact directly with Notch1 in CML CD34+ cells in humans (Ostrowska et al. unpublished) justifies the hypothesis of possible interaction between ABL fusion protein in leukaemic stem cells in CML and Notch receptor via the Disabled adaptor protein. Although ABLl-Notch interaction in Drosophila has been shown to be only CSL-independent (Crowner et al. 2003), it is possible that Notch-ABL interaction, if proven, might be either CSL-dependent or CSL-independent in human. This is simply because ABL protein in Drosophila shows no nuclear localisation unlike the mammalian ABL which can translocate to the nucleus (Saglio and Cilloni, 2004). All the previous findings and the notion that Notch

co-operates

with

several

signal-transduction

pathways

to

induce

leukaemogenesis make it possible for Notch to be integrated with the BCR-ABL fusion protein in leukaemogenesis of CML.

1.5 Research aims and objectives The overall aims of this project are to determine whether there is a role for altered Notch signalling in chronic myeloid leukaemia (CML). Currently, the role of Notch signalling in CML is not yet established. However, several clues raise the possibility that Notch might be involved in CML as detailed in section 1.4. In a nutshell, CML is a clonal disease, which originates from transformed haemopoietic stem cells and Notch is essential in the self-renewal of these haemopoietic stem cells. In addition, the hallmark of CML leukaemogenesis is the presence of BCR-ABL fusion protein, and the ABL protein has been shown to co-operate with Notch in Drosophila. The hypothesis of this project therefore is that Notch signalling might be altered in CML and that Notch may interact with ABL protein expressed in CML cells. To test this hypothesis, the expression of Notch receptors in CML samples and normal control HSCs will be determined using monoclonal antibodies and flow cytometry. The expression of Notch1-4 at the message level will be investigated using PCR

64

technique. The expression of Notch target genes Hes1 and Herp1-2 will be measured using PCR to determine the activity of Notch signalling in CML. The possible crosstalk between Notch and BCR-ABL will be investigated in cell line models as well as in primary CML CD34+ cells.

Chapter 2: MATERIAL AND METHODS 2.1 Cell Biology techniques 2.1.1 Cell lines 2.1.1.1 K562 cell line K562 cells are human chronic myeloid leukaemia suspension cells in myeloid blast crisis which carry the Philadelphia chromosome with a BCR-ABL b3-a2 fusion gene. K562 cells were maintained in RPMI 1640 (Sigma) supplemented with 10% (v/v) fetal bovine serum (FBS - Sigma), 2 mM L-glutamine (Invetrogen) and 0.1 mg/ml

penicillin and streptomycin (Invetrogen) at 37 °C with in 5% CO2. The K562 cells were not used beyond passage 20 before returning to a stock of low passage number stored in liquid nitrogen. 2.1.1.2 NALM-1 cell line NALM-1 cells are human chronic myeloid leukaemia suspension cells in lymphoid blast crisis which carry the Philadelphia chromosome with a BCR-ABL b3-a2 fusion gene. NALM-1 cells are difficult to culture and grow very slowly in the culture medium so it might be of benefit to start culture in 24-well plates. NALM-1 cells were maintained in RPMI 1640 (Sigma) supplemented with 10% (v/v) fetal bovine serum (FBS - Sigma), 2 mM L-glutamine and 0.1 mg/ml penicillin and streptomycin at 37 °C with in 5% CO2. 2.1.1.3 ALL-SIL cell line ALL-SIL cells are human T-ALL (T cell acute lymphoblastic leukemia) suspension cells that carry the NUP214-ABL1 fusion gene. They are also found in the literature as

65

SIL-ALL. They are slow growing cells so it may be of advantage to first culture the cells in a 24-well-plate with 20% FBS. ALL-SIL cells were maintained in RPMI 1640 (Sigma) supplemented with 10% (v/v) fetal bovine serum (FBS - Sigma), 2 mM Lglutamine and 0.1 mg/ml penicillin and streptomycin at 37 °C with in 5% CO2.

2.1.1.4 JURKAT cell line JURKAT cells are human T-ALL suspension cells which are negative for the BCRABL fusion gene. They grow singly or in clumps in suspension. JURKAT cells were maintained in RPMI 1640 (Sigma) supplemented with 10% (v/v) fetal bovine serum (FBS - Sigma), 2 mM L-glutamine and 0.1 mg/ml penicillin and streptomycin at 37 °C

with in 5% CO2. 2.1.1.5 Passage of cell lines K562 and JURKAT cell lines were sub-cultured every 3-4 days and transferred to fresh media to maintain long phase growth. Due to differences in doubling time the splitting ratio was 1:9 for K562 cells and 1:3 for the JURKAT cells. The NALM-1 and ALL-SIL cells were sub-cultured every week by splitting the cells at 1:2 ratio with fresh media. 2.1.1.6 Viable Cell Count Viable cell numbers were determined by using the trypan blue exclusion method. A cell suspension was diluted 1:1 with a 0.4% solution of trypan blue (Sigma). Viable cells were counted using a haemocytometer. 2.1.1.7 Cryopreservation of Cell Lines 1x107 cells ml-1 were slowly re-suspended in FBS (Sigma) with 10% (v/v) Dimethyl sulphoxide (DMSO) (Sigma) and added to cryogenic vials in 1 ml aliquots (Corning). These were then frozen at -20 °C for 1 hour, kept at -80 °C overnight before being cryopreserved in liquid nitrogen for long-term storage.

66

2.1.2 Primary CML samples Fresh or frozen peripheral blood samples from non-treated patients with chronic myeloid leukaemia (CML) in chronic phase were used in this project. Cord blood samples were used as normal controls. .

2.1.2.1 Thawing of cryopreserved CML cells A special thawing solution referred to as ‘DAMP’ solution was used for thawing of cryopreserved CD34+ CML cells from liquid Nitrogen. DAMP thawing solution was prepared in total volume of 500 ml by using the following recipe:

DNase I (2 vials at ~2500 U/vial (1mL), StemCell Technologies) Magnesium chloride (400X, 1.0 M stock)

2 mL 1.25 mL

Trisodium citrate (0.155M, Sigma)

53 mL

Bovine Serum Albumin (20%, Sigma)

25 mL

Dulbecco’s PBS (magnesium/calcium free)

to 500 mL

Frozen CD34+ CML cells were thawed quickly by immersing in a 37 °C water bath before being opened under sterile conditions and transferred into a sterile 10 ml Falcon tube. 10 ml of warm thawing solution were added dropwise to the cells over 10 minutes and centrifuged at 389g for 5 minutes. The washing step was repeated twice with DAMP thawing solution before the cells were filtered by a cell strainer (BD) and counted. 1 x 10 6 / ml CD34+ cells were cultured overnight in a 24 well culture plate in serum free expansion medium (SFEM) supplemented with a five growth factor cocktail (see 2.1.2.2) at 37 °C in 5 % CO2. This initial 24h culture helps the cells to revive and expand before they are being subjected to any treatment or used in future experiments.

2.1.2.2 Short term liquid culture of primary CML CD34+ cells CD34+ cells were cultured in serum free expansion medium (SFEM) (StemCell Technologies) supplemented with 1% glutamine (100 mM) and 1% penicillin– 67

streptomycin (100 mM). SFEM was further supplemented with a five growth factor cocktail comprising 100 ng/ml Flt3- ligand, 100 ng/ml stem cell factor, 20 ng/ml each of interleukin (IL)-3, IL-6 and granulocyte colony stimulating factor (GCSF) (all from R&D systems). To prepare a ready to use cytokine cocktail, cytokines were combined together and diluted in phosphate buffered saline (PBS) containing 0.1% BSA/PBS to create a 100x working stock solution which was stored at 4 °C for up to 3 weeks.

2.1.3 Retroviral transfection of K562 cells with Notch1ΔE 200 µl of 50 µg/ml Retronictin (TaKaRa) was added to each well of a 24 well plate before being placed at 4 °C overnight. The following day, the Retronectin solution was removed before 1 ml of 1% (w/v) bovine serum albumin (BSA- Sigma) in 1x PBS to each well for I hour at RT to reduce non-specific binding. Next, the supernatant was removed, prior to 1 x 105 K562 cells in log phase growth were mixed with 1 ml of retroviral supernatant which contain either Notch1ΔE or the empty vector pmX (Chadwick et al. 2008). Cells were then placed in duplicate onto the Retronectin coated wells and centrifuged for 45 minutes at 1000 xg at 20 °C before being incubated at 37 °C in 5% CO2 overnight and left for 48 hours. The cells were then harvested and the GFP positive cells were FACS sorted and cultured in the K562 cells media [2.1.1.1].

2.2 Flow cytometric techniques 2.2.1 Isolation of mononuclear cells (MNC) Isolation of mononuclear cells from blood samples was done using ficoll-paque (Amersham Pharmacia Biotech) density gradient separation under sterile conditions. Samples were diluted 1:1 with sterile Hanks Balanced Salt Solution (HBSS) supplemented with 5% Newborn Calf Serum (NCS) (Invitrogen). 20 ml of the diluted blood was then carefully layered onto 10 ml Ficoll in a 50 ml falcon tube and centrifuged at 1500 rpm (389 g) for 30 minutes at room temperature (RT). Next, the mononuclear cells were harvested from the interface layer and transferred into a new tube and washed twice with 50 ml HBSS / 5% NCS by centrifugation at 1500 rpm (389 g) at RT for 7 minutes and cell count and viability were done between washes.

68

The pellet was then re-suspended in known volume of HBSS / 5% NCS for FACS sorting, or processed for liquid nitrogen freezing.

2.2.2 Isolation of haemopoietic progenitor cell populations Haemopoietic progenitor cell populations from normal cord blood and CML samples were isolated by positive selection for CD34 expressing cells using StemSep™ kit (StemCell Technologies) according to the manufacturers’ instructions. In summary, the isolated MNC were re-suspended in HBSS / 5% NCS and incubated with 100 µl selection cocktail per ml of cells on ice for 10 minutes. 60 µl magnetic colloid /ml cells were then added to the cells and incubated on ice for 10 minutes. Cells were then washed with 3 ml HBSS/5% NCS and resuspended in 2 ml HBSS/5% NCS. Next, a MidiMax column (Miltenyi) was washed with 2ml HBSS/5% NCS and cells were run through column in 1ml aliquots, the column was then washed with 2 ml HBSS/ 5% NCS. The magnet was then removed and the bound cells were eluted from the column with a plunger in 2 ml HBSS/ 5% NCS. The eluted CD34 cells were then pooled and viability assessed before cells were pelleted and then re-suspended in 100 µl containing 1:20 CD34-APC, 1:20 Thy-PE and 1:20 lin-FITC cocktail. After incubation for 20 minutes at 4 °C in the dark, cells were washed with 2 ml HBSS/5% serum and re-suspended in 1 ml HBSS/5% serum for sorting. Cells were sorted into a 24 well plate using a FACS Vantage (Becton Dickinson) flow cytometer. Sorted cells were then transferred into RNAse free eppendorf tubes.

2.2.3 Staining procedures for flow cytometric analysis 2.2.3.1 FACS analysis of extra-cellular Notch1 on primary CML cells In order to study the expression patterns of Notch1 on CML cells, the mononuclear cells from frozen samples were stained with EA1 monoclonal antibody, which recognises the extra-cellular domain of Notch1, as well as with a set of antibodies that identify different myeloid and lymphoid haemopoietic progenitors. First, cells were washed twice in HBBS/5% NCS and filtered with nylon mesh filter before 10 5 cells were transferred to a FACS tube and pelleted at 389 g (1500 rpm) for 5 minutes at 4°C. After removing the supernatant, the cells were re-suspended in 50 µl EA1 or IgG primary antibodies at the optimum dilution (see table 2.1) and incubated in the dark at 4°C for 20 minutes. The cells were then washed in 2 ml HBBS/5% NCS and

69

centrifuged at 1500 rpm (389 g) for 5 minutes at 4°C. The supernatant was discarded and the cells were re-suspended in 50 µl secondary antibody and incubated in the dark for 20 minutes at 4°C. Cells were then washed as before, and pelleted at 1500 rpm (389 g) for 5 minutes before the supernatant was removed and the conjugated antibody at appropriate dilution was added. After another 20 minutes, incubation period, the cells were washed in 2 ml HBBS/5% NCS, pelleted and re-suspended in 300 µl diluted propidium iodide (PI) for analysis. CD34 antibody was added in all tubes in order to limit the study of Notch1 and other surface molecules expression to CD34+ population in normal blood and CML samples. IgG1 hybridoma supernatant was used as isotype control for all surface markers studied. Flow cytometric analysis was performed on a FACS Vantage (Becton Dickinson) flow cytomter, and CellQuest® software was used for data analysis. 2.2.3.2 FACS analysis of extra-cellular Notch1 on K562 cells In order to study the expression of the extra-cellular Notch1 (ECN1) on the cell surface of K562 cells, 1 x 106 K562 cells were directly stained with EA1 monoclonal antibody according to the same staining protocol described in 2.1.3.1. The EA1 primary antibody was used at 1:100 dilution (stock conc. is 2 mg/ml), the IgG1 hybridoma supernatant at equivalent concentration to the primary antibody was used as an isotype control. Because K562 cells are negative for CD34 surface antigen the analysis gate used here included all live K562 cells. 2.2.3.3 FACS analysis of intra-cellular Notch1 on K562 cells This method was used to analyse the expression of the intra-cellular Notch1 (ICN1) in K562 cells. At least 1 x 105 K562 cells were suspended in 100 µl fixing reagent (Caltag) and incubated at RT for 15 minutes. The cells were then washed with 2 ml HBSS /5% NCS and centrifuged for 5 minutes at 1500 rpm (389 g). The supernatant was removed and and the cell pellet was resuspended with 25 µl permibilzing reagent

(Caltag). The b-TAN20 primary antibody which recognises the ICN1 domain was added directly to the cells at 1:5 dilution (stock conc. is 40 µg/ ml) and the cells were mixed and incubated at RT in the dark for 40 minutes. The cells were washed twice in 2 ml HBBS/5% FBS and centrifuged at 1500 rpm (389 g) for 5 minutes at 4°C. The secondary antibody (anti-rat IgG2 FITC) was then added at 1:50 dilution to the cells which were incubated at RT in the dark for 30 minutes. The cells were then washed 70

with 2 ml of HBBS/5% FBS and spun at 389 g for 5 minutes. Finally, the cells were resuspended in 300 µl of HBSS ready for FACS analysis. An appropriate isotype control (IgG2 rat antibody) was used at a concentration equivalent to the primary antibody.

2.2.3.4 The P-crkl assay Crkl is a prominent substrate of the BCR-ABL oncoprotein in CML and binds to both BCR-ABL and c-Abl. Crkl is prominently and constitutively tyrosine phosphorylated in CML cells and is not phosphorylated in normal haemopoietic cells (Oda et al. 1994). The levels of phosphorylated crkl (P-crkl) were measured by intra-cellular FACS technique and the P-crkl expression was utilised as a marker for ABL kinase activity in this project. Cells from various cell lines or from primary CD34+ CML cells were harvested from culture media and washed once in 3 ml HBBS/5% FBS. At least 1 x 105 cells were resuspended in 100 µL fixing reagent (Caltag Laboratories) and incubated at RT for 15 minutes. The cells were then washed once with 3 ml HBBS/ 5% FBS and centrifuged at 389 g for 5 minutes. The supernatant was removed and the cells were resuspended with 25 µl permeabilizing reagent (Caltag Laboratories) and 2.5 µl of P-crkl primary antibody (New England Biolabs) was added directly to this buffer. The cells were then vortexed and incubated at RT for 40 minutes before being washed twice with 3 ml HBBS/ 5% FBS. After resuspending the cells in 100 µl buffer the secondary antibody was added directly at appropriate dilution and the cells were mixed and incubated at RT for 30 minutes in the dark. Next, the cells were washed twice and analysed by flow cytometry. The P-crkl results were reported as the mean fluorescence intensity (MFI). The isotype control used in the P-crkl assay was rabbit IgG at a concentration equivalent to that of the primary antibody. Positive and negative controls for the P-crkl assay were K562 and JURKAT cells respectively. The secondary antibody used for measuring the P-crkl levels in cell lines was the pre-diluted PE F(ab')2 Donkey anti-Rabbit IgG (BD biosciences) whereas the FITC monoclonal mouse anti-rabbit antibody (Sigma) was used at 1:20 dilution to assess P-crkl expression in CD34+ CML cells.

71

Monoclonal Ab

Fluorochrome conjugate

Target/ lineage Dilution specificity

Supplier

EA1

FITC or PE

Ubiquitous

1:100

In house

b-TAN20

FITC

Ubiquitous

1:5

DSHB

P-crkl

FITC or PE

ABL+ HCs

1:40

Cell signaling

CD90 (Thy-1)

PE

Primitive HCs

1:20

pharmingen

CD34

APC

Primitive HCs

1:20

BD

CD38

FITC

1:50

BD

CD14

FITC

T, B and CD34+ committed cells Myeloid cells

1:50

BD

CD15

FITC

Myeloid cells

1:25

BD

CD16

FITC

Myeloid cells

1:50

BD

CD33

PE

Myeloid cells

1:50

BD

CD7

FITC

T cells

1:25

BD

CD3

FITC

T cells

1:25

BD

CD19

FITC

B cells

1:20

BD

CD45

PE

Pan HCs

1:50

BD

Erythroid cels

1:50

BD

Glycophorin-A FITC

72

IgG1

FITC

Control

1:20

BD

IgG1

PE

Control

1:20

BD

Table 2.1 Monoclonal antibodies and the dilutions used in flow cytometric staining experiments. (HCs: Haemopoietic cells, BD: Becton Dickonson, DSHB: The Developmental Studies Hybridoma Bank at the university of IOWA).

2.3 Molecular biology techniques 2.3.1 RNA extraction

Using RNAse and DNAse free filter tips, sorted cells were transferred to DNAse and RNAse free eppendorf tubes in a laminar flow cabinet and centrifuged for 3 minutes at 3000 rpm (1840g) at 4 °C. The supernatant was then removed and the pellet was resuspended in 200 µl RNAzol B (Biogenesis) before being vortexed and kept on ice for 5 minutes. 20 µl of chloroform (Sigma) was then added, and the solution was vortexed and spun at 13 000 rpm (12470 g) at 4 °C for 15 minutes. Next, the upper aqueous layer was aspirated into a new eppendorf tube. An equal volume of Isopropanol (Sigma) was added, mixed, and then incubated at -70 °C for two hours (up to 1 week). The samples were thawed and pelleted at 13 000 rpm at 4 °C for 30 minutes, before the supernatant was removed and the pellet washed twice with 70% ethanol (molecular biology grade Sigma) for 10 minutes at 4 °C. The supernatant was then removed and the pellet was air dried for 1-2 hours before re-suspended in 10 µl sigma water.

2.3.2 Construction of cDNA from low cell numbers by PolyA PCR This protocol was typically used for the isolation of RNA from less than 1x10 5 cells. Poly-A PCR is a powerful technique for the investigation of gene expression in rare cell populations such as haemopoietic stem cells. Poly-A PCR results in the unbiased amplification of cDNA representing all poly adenylated RNA in a sample as small as a single cell (Brady and Iscove, 1993). Bias for short length cDNA sequences in a sample is avoided by limiting the length of the initial cDNA produced to an average length of 350 bp regardless of the size of the original RNA template. Principles of the

73

poly-A PCR technique are summarised in figure 2.1. Table 2.2 shows a list of the solutions used in the ploy-A procedure.

cDNA reaction A first strand buffer was freshly prepared by mixing 192 µl lysis buffer with 4 µl Rnase inhibitor (Fermentas) and 4 µl primer mix freshly diluted 1:4 with water (Sigma). 10 µl of this buffer was then transferred to fresh tube and 0.5-1 µl RNA (up to 100 ng) was added to it. The mixture was then heated for 1 minute at 65 ºC before it was allowed to cool for 3 minutes at 18 ºC. After cooling on ice, 0.5 µl AMV Reverse Transcriptase (Roche) was added to each sample, and incubated at 37 ºC for 15 minutes and then heat inactivated at 65 ºC for 10 minutes and placed on ice.

cDNA tailing reaction An equal volume of fresh 2X tailing buffer, including 0.5 µl Terminal Transferase (Roche), was added to the samples, before being incubated for 15 minutes at 37 ºC and heat inactivated by incubating at 65 ºC and then placed on ice.

Poly-A PCR A PCR mix was prepared using the following ratios: 40 µl MMM : 2.7 µl Not1 dT oligonucleotide (MWG Biotech): 1.5 µl Taq Polymerase (Roche). The final concentrations of all PCR solution components were as follow: Tris-HCl pH 8.3 KCl MgCl2 dNTPS Triton X-100 BSA Not140 oligo B/M Taq Polymerase

74

23.5 mM 117.4 mM 8.2 mM 2.2 mM 0.23% 47 µ g/ml 8.33 µ M 84.84 units/ml

10 µl of the PCR mix was added to 5 µl tailed cDNA and amplified using the following cycle profile: 25 cycles consisting of 1 minute at 94 ºC, 2 minutes at 42 ºC, and 6 minutes at 72 ºC linked to another 25 cycles consisting of 1 minute at 94 ºC, 1 minute at 42 ºC, and 2 minutes at 72 ºC.

Reamplification of poly-A cDNA The globally amplified cDNA, from the previous step, was diluted 1:100 using sigma water. Next, a 25 µl reaction was prepared by mixing the following: 1µl of the 1:100 diluted global cDNA, 19 µl sigma water, 2.5µl of 10x Taq Buffer (+MgCl 2), 2.5 µl of 2.5 mM dNTPS (Roche), 0.25 µl of 150µM NotI Oligo-dT (MWG Biotech), and 0.25 µl 5U/µl Taq Polymerase (Roche). Samples were then run on a PCR program with the following conditions: 25 cycles consisting of 1 minute at 94 ºC, 1 minute at 42 ºC, and 2 minutes at 72 ºC. To determine the efficiency of the PCR reaction, 1µl of the PCR product was run on 1.5% w/v agarose gel.

75

Solution

Contents

Supplier

cDNA/Lysis Buffer

-1 ml 5X First Strand Buffer. - 10 µ l BSA (Mol Biol Grade 20 mg/ml). - 250 µ l 10% NP-40.

Gibco/BRL - Roche.

- 3.55 ml Water.

- Sigma. - Sigma.

-800 µ l

TaKaRa 2.5 mM

- TaKaRa

- 24 µ l

- 2.5 mM. - 5.8 µM

dT24 (200uM)

-Gibco/BRL

-200 mM K cacodylate pH 7.2

Primer mix

dNTPs.

2X Tailing Buffer

5X Tailing Buffer

-1 ml 5X Tailing Buffer.

Final concentration

NA.

, 4 mM CoCl2, 0.4 mM DTT

- 25µl 100 mM dATP. - 1.475 ml water

- Roche. - Sigma.

- 0.5 Mpotassium cacodylate PH 7.2. - 10 mM CoCl2. - 1 mM DTT.

-Gibco/BRL

76

- 1 mM

MMM Buffer

-1000 µl 10X Taq Buffer.

- Roche.

- 25.94 mM Tris-HCl pH 8.3. - 129.7 mM KCl

- 20 µl 1M MgCl2. - 375 µl 25 mM dNTPs - 10 µl 20mg/ml BSA. - 100 µl 10% Triton X-100. - 2.35 ml H2O

- Roche. - Roche. - Roche. - Sigma. - Sigma.

- 9.07 mM - 2.43 mM - 51.88 µ g/ml - 0.26%

Table 2.2. List of solutions used in poly-A PCR reaction.

mRNA

(1 77

Figure 2.1. Outline of poly-A PCR technique. Preparation of cDNA for poly-A PCR starts by the addition of reverse transcriptase and Oilgo(dT) primer that anneals to the poly-A tail presen at the 3' end of the Mrna (1). Next, an oligo (dA) is added to the 3' of the first strand cDNA using terminal m-RNA transferase to produce tailed cDNA (2). PCR amplification of the dA/dT- bracketed cDNA is then performed using a modified oligo (dT) primer and Taq polymerase to synthesise the global cDNA pool (3). Finally, the poly-A cDNA can then be reamplified using the protocol described in 2.3.2 (4). Bias for short length cDNA sequences in a sample is avoided by limiting the length of the initial cDNA produced to an average length of 350 bp regardless of the size of the original RNA template. Modified from Brady and Iscove (1993).

2.3.3 Construction of cDNA by from high cell numbers High Capacity cDNA Reverse Transcription Archive Kit (Applied Biosystems) was used for the cDNA production from cell numbers in excess of 1x105 cells. The High Capacity cDNA Kit offers superior reverse transcription capacity, efficiency, and linearity over other commercial kits and has the performance necessary for accurate quantitation of RNA targets. A 20 µl volume of the PCR mix from table 2.3 was mixed with 5 µl of RNA. The reaction mixture was amplified by PCR using the following cycling parameters: 10 minutes at 25 ºC, followed by 2 hours at 37 ºC. All procedures were done using DNase/ RNase free filtered tips and all reagents were kept on ice during the preparation of the PCR reaction mix. Table 2.3. Amplification reaction mixture from the High Capacity Kit Component Volume cDNA (µl) Kit

10X RT Buffer 25X dNTPs 10X Random Hexamers Reverse transcriptase (20 U ml-1) Sigma H2O

2.3.4 Gene specific PCR 2.3.4.1 Primers 78

2.5 1.25 2.5 1.25 12.5

For each gene studied, PCR primer pairs were designed to be directed towards the mRNA sequence present within 280-300 bases of the poly-A tail. Table 2.4 shows the sequences of the set of primers used in the gene expression studies. Most of the primers

were

designed

with

the

help

of

the

http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi.

Primer3 All

program

primers

at were

resuspended in H2O (Sigma) to a stock concentration of 100 µM, and used at a working concentration of 10 µM after further dilution with H2O (Sigma).

2.3.4.2 Optimisation of Primer Sets To optimize the annealing temperatures for each set of primers gradient PCR was performed using an Eppendorf Mastercycler gradient PCR thermocycler. The gradient PCR thermocycler is capable of producing a gradient across the block for the annealing temperature. Gradient PCR was performed in 10 µl reactions consisting of 5 µl PCR ReddyMix (ABgene), 0.5 µl of forward and reverse primers, 3 µl water (Sigma) and 1 µl of human genomic DNA diluted 1:500 (Promega). The following cycle parameters for gradient PCR were used: 5 minutes at 95 ºC linked to 30 cycles consisting of 1 minute at 94 ºC, 1 minute at gradient annealing temperature between 5-65 ºC, and 1 minute at 72 ºC. This was followed by final cycle for 5 minute at 72 ºC. 2.3.4.3 PCR reaction In a PCR hood, using dedicated pipettes and filter tips, primers were diluted down to a working dilution of 10µM and a master mix was prepared at 4ºC using the following recipe in 10µl Reaction: Reagent PCR ReddyMix (ABgene) Primer1 Primer2 Water (Sigma)

Conc.

Vol (µl )

10 µM 10 µM

5.0 0.3 0.3 3.4

Final Conc. 300 nM 300 nM

Next, 9 µl of the mastermix was aliquoted into PCR tubes or wells (if using a PCR plate) and 1 µl of cDNA was added. The reaction was then amplified using the 79

following cycle parameters: 5 minutes at 95 ºC linked to 28-30 cycles consisting of 1 minute at 94 ºC, 1 minute at specific primer pair annealing temperature (table 2.4), and 1 minute at 72 ºC. Water (Sigma) was used as negative control.

2.3.4.4 Detection of PCR products Agarose gel electrophoresis was used to resolve and visualise DNA bands. 1.5 % agarose was prepared by adding 1.5 g high gel agarose (Sigma) to 100 ml of 0.5 xTBE. The agarose was dissolved by heating in a microwave. Once dissolved and cooled 5 µl of 1:20000 diluted Vistra Green (Amersham Pharmacia) was added as the staining agent. The gel was allowed to set before it was placed in an electrophoresis tank filled with 0.5 X TBE buffer. Next, 5 µl of the PCR product was loaded into the wells and run for 30-45 minutes at 120 volts. PCR products not amplified with ReddyMix were first diluted 1:6 with Orange G loading buffer. The size of the products was determined by loading the GeneRuler 100bp ladder (Fermentas) along with the samples. Finally, the resulting gel was observed on a Typhon 8600.

Gene name

Notch1

Notch2

Notch3

Notch4

Hes1

Herp1

Primer

Primer Sequence 5’ – 3’

hN1F

GTGAGGGACGTCAGACTTGG

hN1R

AACATCTTGGGACGCATCTG

hN2F

AAAGCATCTGTCAAATAGGAAAC

hN2R

TAAGGAATGTTACAAACCAATCA

hN3F

CAAGCTGGATTCTGTGTACCTAGT

hN3R

CCCCAGCAAGGCTATGGAACA

hN4F

ATATTTATTGGGCACCTACTAATG

hN4R

ATAGCAATAGCAGTGGCTAGAAG

Hes1F

GTATTAAGTGACTGACCATG

Hes1R

TCAAACATCTTTGGCATCAC

Herp1F

TCATTTCTCTACTGTGTGGAG

Herp1R

GTGGTATGTAAAGACTCTTGC

Herp2F

CTAATTTTCCTGGGACTGCC

80

Length Size T (bp) (bp) ºC 20 166 58 20 ºC 23 205 58 23 ºC 24 202 56 21 ºC 23 166 58 23 ºC 20 140 54 20 ºC 21 155 60 21 ºC 20

Herp2 GAPD H BCR-ABL

Herp2R

TCAAACCCAGTTCAGTGGAG

GAPDHF

CCAGCAAGAGCACAAGAGGAAGAG 24

GAPDHR AGCACAGGGATACTTTATTAGATG BCR-ABL F BCR-ABL R

20

24

TCCACTCAGCCACTGGATTTAA

22

TGAGGCTCAAAGTCAGATGCTACT

24

216

60 ºC

180

56 ºC

Table 2.4 Oligonucleotide primer sequences and annealing temperatures (T).

2.3.5 Real time PCR 2.3.5.1 Overview OF Real Time PCR Real time PCR is the ability to monitor the progress of PCR in real time. Real time PCR technique uses the fluorogenic 5´ nuclease chemistry (TaqMan®) or SYBR® Green I dye chemistry to allow for the quantification of gene expression in the early phase of PCR reaction. This is in contrast to the conventional PCR method which uses Agarose gels for detection of PCR amplification at the end-point of the PCR reaction. This difference in principle of detection makes real time PCR far more sensitive approach to use in gene expression studies than the traditional PCR method. Gene quantification by real time PCR can be performed by absolute or relative quantification. The absolute gene quantification is used to measure the input copy number of a target gene by using a standard curve. The relative gene quantification is used to analyze changes in gene expression in a given sample relative to another reference control sample. Real time PCR experiments were performed using either TaqMan® probes or SYBR® Green. TaqMan® probes are specific sequences of DNA that recognise and bind the target DNA between the primers. Each probe is labelled with a reporter, fluorescein5-carboxamide (FAM) at the 5’ site, and a quencher, carboxytetramethylrhodamine (TAMRA) at its 3’ site. While the FAM and TAMRA molecules are attached to the same probe, FAM fluorescence is quenched by TAMRA. During the real time PCR reaction the 5’ exonuclease activity of Taq polymerase releases the reporter (FAM) from the probe so its fluorescence in no longer quenched by TAMRA and as a result FAM emits a fluorescence signal as the real time PCR reaction progresses. Fluorescence from FAM is measured after each PCR cycle and correlate to the 81

60 ºC

amount of the PCR products formed during the PCR reaction. SYBR® Green is a minor-groove DNA binding dye that fluoresces upon binding to double stranded DNA. As the DNA amplification proceeds in the real time PCR reaction SYBR® Green dye binds to each new copy of double-stranded DNA and the result is an increase in fluorescence intensity proportional to the amount of PCR product produced. Because the SYBR Green binds to any double-stranded DNA, it can also bind to nonspecific double-stranded DNA sequences including primer dimmers. Therefore, it was necessary to run a dissociation curve for each amplification to ensure no non-specific amplification has occurred (Fig. 2.2).

2.3.5.2 Real time PCR protocols 2.3.5.2.1 Real time PCR using TaqMan®probes This method was used to measure Notch1, Notch2, and Hes1 genes on cDNA samples from CML patients as well as normal bone marrow samples. The primers used in real time PCR are listed in table 2.5. cDNA samples were diluted 1:100 with H2O (Sigma). Each reaction was made up to a total volume of 25 μl, with 10 μl of diluted cDNA, 12.5 μl of 2x Power SYBR® Green master mix and 0.5 μl of each 10 μM primer and 0.05 μl of 100 μM probe. This mixture was added to a 96 well plate (Bioplastics) and sealed with StarSeal 96 well plate sealant (STARLAB) and the plate was then centrifuged to ensure reagents were at the bottom of the wells. The ABI 7300 PCR

machine (Applied Biosystems) was used for data collection and analysis. The analysis software was set up to ignore SYBR® Green and only look for amplification involving the TaqMan® probe. The cycling parameters used were 95 °C for 10 minutes, followed by 40 cycles of 95 °C for 15 seconds and 60 °C for 1 minute. To minimize contamination, filtered tips were used and plates and reagents were kept on ice at all times. 2.3.5.2.2 Real time PCR using SYBR® Green This method was used to measure Hes1 gene expression on cell lines and on primary CML samples following treatment with IM, GSI, and VPA drugs. cDNA samples were diluted 1:50 with H2O (Sigma). Each reaction was made up to a total volume of

15 μl, with 5 μl of diluted cDNA, 7.5 μl of 2x Power SYBR® Green master mix and 0.3 μl of each 10 μM primer and 1.9 μl of sigma H2O. The reaction mixture was added to 96 well plate and analysed by the ABI 7300 PCR machine as described 82

above [2.3.5.2.1]. Following the PCR amplification the PCR plate were re-run to obtain a dissociation curve to ensure no non-specific amplification has occurred (Fig. 2.2)

2.3.5.3 Data analysis The real time PCR data in this project were analysed using the relative gene quantification method to study changes in gene expression. The relative change in gene expression was determined using the 2 –ΔΔCT method (Livak and Schmittgen, 2001) which is also known as the Comparative CT method. The CT value is the cycle number at which the level of fluorescence exceeds the level of background fluorescence and passes the fixed threshold. The CT value is indicative of the relative amount of the target gene in a sample because samples with a low starting amount of template require more PCR cycles to produce a fluorescence signal above the background level, and therefore have a high CT value whereas samples with a high amount of template require fewer cycles to reach the fixed threshold and therefore have a low CT. In the 2 –ΔΔCT method, the CT (Cycle threshold) values of a sample are compared to those of a biological calibrator sample such as a non-treated sample or RNA from normal tissue. The CT values of both the calibrator and the sample of interest are normalised to an appropriate endogenous housekeeping gene such as GAPDH to ensure that similar levels of total cDNA found in each sample.

Calculation of relative gene expression using the the 2 –ΔΔCT method The internal control gene (housekeeping gene) used in this project was GAPDH whereas the biological calibrator was either normal control sample from healthy donors or untreated sample depending on the experimental design. The CT values provided from real-time PCR instrumentation were imported into a spreadsheet. To analyse the gene expression of Notch genes in CML samples the gene expression in CML and NBM samples was normalised to the GAPDH house keeping gene and represented as DCt values. For each sample the mean DCt value was calculated. Comparison of gene expression between NBM and CML samples was derived from subtraction of NBM DCt values from CML DCt values to give a DDCt value, and relative gene expression was calculated as 2 –ΔΔCT. Due to the inherent variations in gene expression between different normal bone marrow (NBM) samples it is not surprising that the DDCt values for the biological calibrator (NBM) will not be zero

83

and therefore the relative gene expression values (2 –ΔΔCT) from NBM samples will not be equal to one. Using the 2 –ΔΔCT method in experiments where the change in gene expression was studied in a sample following a drug treatment, the data are presented as the fold change in gene expression normalized to an endogenous reference gene (GAPDH) and relative to the untreated control. Comparison of gene expression between treated and untreated cells was derived from subtraction of untreated cells DCt values from treated cells DCt values to give a DDCt value, and relative gene expression was calculated as 2 –ΔΔCT. For the untreated control sample, DDCt equals zero and 20 equals one, so that the fold change in gene expression relative to the untreated control equals one, by definition. For the treated samples, evaluation of 2 –ΔΔCT indicates the fold change in gene expression relative to the untreated control.

2.3.5.4 Validation of the 2 –ΔΔCT method For calculation to be valid, the amplification efficiencies of the target and the housekeeping genes must be approximately equal. This can be established by looking at how ΔCT varies with template dilution. If the plot of cDNA dilution versus ΔCT is close to zero, it implies that the efficiencies of the target and housekeeping genes are very similar. This was calculated for Hes1 and GAPDH primers prior to their use with SYBR® Green PCR and the obtained value was 0.021 indicating the 2 –ΔΔCT method can be used to generate relative quantitative data using the Hes1 and GAPDH primers. Target

Primer

Primer Sequence 5’ – 3’

hN1F

TCCCCCGGCTCTACGG

Notch

hN1R

ACACAGTAAAAATCAACATCTTGGGAC 60

1

hN1TP

CCGCGTGGTGCCATCCCC

hN2F

AGCCATAGCTGGTGACAAACAG

Notch

hN2R

CAACTACTTCGCATTTCCATTGG

2

hN2TP

AGGCACCTTGTCCCTGAGCAACC

Hes1TF

GCCACCCCTCCTCCTAAACTC

Hes1TR

TCAAAGAGAAGGAGGCAAGGAAA

Hes1

84

Annealing Temp °C

60

60

Hes1TP

CAACCCACCTCTCTTCCCTCCGGA Table 2.5. Primers used for real time PCR.

( Fig. 2.2. Real Time PCR. (A) An example of real time PCR reaction. The X-axis represents the PCR cycle number and the Y-axis represents the magnitude of fluorescence signal (ΔRn). The green line represents the fixed threshold. The CT value is the cycle number at which the level of fluorescence exceeds the level of background fluorescence and passes the fixed threshold. The intersection of the green line with the amplification curves on their exponential phase gives a CT value, which can be used to calculate the relative amounts of cDNA present in different samples. (B) An example of the dissociation curve performed after a SYBR® Green PCR. Non specific amplification85of primer dimers is not detected here as they dissociate at around 70 °C. The dissociation curve observed here is indicative of specific amplification of the desired PCR product.

2.3.6

Protein Analysis

2.3.6.1. Protein extraction and determination of concentration K562 cells were centrifuged at 5000 rpm (1840g) for 2 minutes before being transferred to an eppendorf tube and resuspended in 100 µl RIPA buffer (Table 2.6) containing freshly added phosphatase inhibitor and protease inhibitors. The phosphatase inhibitor used in protein extraction was 1 mM Na3 VO4 (Sigma) whereas the protease inhibitors were 1 mM Phenylmethanesulfonyl fluoride (Sigma) and 5 µg/ ml Leupeptin. After leaving the cells with RIPA buffer for 5 minutes on ice the eppendorf tube was spun at 7900 g for 10 minutes at 4 °C. The supernatant was then transferred to a fresh eppendorf tube and stored at -20 °C. Table 2.6: RIPA buffer ingredients and concentrations. Reagent

Concentration

Supplier

Tris-Chloride (PH 7.4)

20 mM

Sigma

Sodium Chloride

150 mM

Sigma

EDTA

5 mM

Sigma

Nonidet P40 (NP-40)

1% (w/v)

Sigma

The concentration of protein was determined by the comparison of absorbance to standards of known concentration. The standards were made up by preparing different dilutions of BSA (Sigma) in water to have final concentrations of 1.25, 2.5, 5, 10, 15, and 20 µg/ ml respectively (Table 2.7). Table 2.7. Standards for protein concentration determination. Standard concentration (µg/ ml)

1.25

2.5

5

10

15

20

(( (µg/ml)of 2mg/ ml BSA (µl) Volume

0.61

1.25

2.5

5

7.5

10

2 86

Volume of water (µl)

9.38

8.75

7.5

5

2.5

0

8 The standards were then made to 1000 μl containing 200 μl of 1:5 diluted BIORAD Protein Assay reagent (BIO-RAD), respective volume of BSA (table 2.6), 1μl RIPA buffer and H2O. From this mixture, 300 μl was added to a 96 well plate in triplicate. For each sample, 1 μl of sample was added to 799 μl of H2O and 200 μl of 1:5 diluted BIORAD Protein Assay reagent before 300 μl were added in triplicate to the wells of a 96 well plate. The absorbance values of each of the samples were measured on an ELx800 plate reader (BIOTEK) and compared to a blank control containing only H2O and BIO-RAD Protein Assay. The KC junior software (BioHit) was used to create a standard curve from which the equation of the graph was used to calculate the protein concentrations in the samples.

2.3.6.2 SDS-PAGE and Western Blott Sodium dodecyl sulphate poly acrylamide gel electrophoresis (SDS-PAGE) method was used to separate the proteins in the cell lysates. Protein separation Glass plates were washed with ethanol and assembled. The running gel solution (Table 2.8) was made with TEMED added last and then the solution was poured into the glass plates. After approximately 15 minutes the stacking gel (Table 2.8) was prepared and added with a comb being inserted to form wells. The protein samples where diluted with sigma water in order to have the same protein concentration in all samples. Protein samples were then boiled in 2x sample buffer and H2O in a volume of 15 μl for 5 minutes, briefly centrifuged. Once the gel had set, the plates were moved to the tank, which was filled with 1x running buffer (Table 2.8) containing 1% SDS (v/v). Protein samples were loaded onto the gel along with 3 μl Precision Blue marker (BIO-RAD). Electrophoresis was carried out at 150 v until the marker reached the bottom of the gel. Protein Transfer and antibody staining Following the separation of protein by gel electrophoresis, the proteins were transferred to a nitrocellulose membrane (Sigma) by electrophoresis. The gel was

87

removed from the tank and soaked in transfer buffer (20% 1x running buffer and 80% methanol). Two Hybond-N pads (Amersham Pharmacia) and a nitrocellulose membrane were soaked in the transfer buffer before a sandwich was made up consisting of a pad, the membrane, the gel and another pad. After removing air bubbles the sandwich was placed in a Transblot Semi Dry Transfer Cell (BIO-RAD) and run at 100 mA for 115 minutes. Next, the membrane was removed and placed in 20 ml Blocking buffer (1x PBS containing 5 % Marvel (Tesco) and 1% (v/v) TWEEN® 20 (Sigma)) for 1 hr at RT on shaking platform to prevent non-specific binding to the membrane. The membrane was then transferred to a heat-sealable bag with 2 ml of antibody solution (the primary antibody diluted in 1x PBS containing 1% Marvel and 1% (v/v) TWEEN® 20) and left on a shaking platform for 2 hrs at RT or at 4 °C overnight. The membrane was then washed once with water and three times with PBST (1x PBS containing 1% (v/v) TWEEN® 20) for 15 minutes at RT on shaking platform. The secondary antibody was added as described above for the primary antibody. After incubation for one hour at RT the membrane was washed for 4 times at 10 minutes intervals at RT on shaking platform. After removing the excess solution the membrane was placed onto saran wrap and mixed with Chemiluminescence Substrate Kit (Pierce) according to the manufacturer’s instructions, before being placed in an autoradiography cassette with a piece of Kodak Biomax film and developed in a Fuji film FPM800A automated developer.

2.4 Statistics Comparison between two different biological samples Since the CML and NBM samples are biological samples in which normal distribution of the samples is unlikely and because the two groups were not matched pairs the Mann-Whitney test was the appropriate test to compare the differences in gene expression between NBM and CML samples. Comparison between the means of one sample matched pairs To compare between the means of one sample before and after a drug treatment a paired T-test was carried out. The GraphPad Prism statistical package (GraphPad Software Inc., USA) was used to run statistical tests in this project.

88

Materials

Running Gel

Stacking gel

2x Sample Buffer

Volume (μl)

Acrylamide

Sigma

1500

1.88 M Tris pH 8.8

Melford

1200

0.5% SDS

Sigma

1200

Distilled H2O

Self

2100

Tetramethylethylenediamine (TEMED) 10% (w/v) ammonium persulphate (APS) in H2O Acrylamide

Sigma

5

Sigma

30

Sigma

330

6.8 M Tris pH 6.6

Melford

400

0.5% (v/v) SDS

Sigma

400

H2O

Self

870

TEMED

Sigma

2

10% (w/v) APS in H2O

Sigma

10

1 M DTT

Sigma

1000

10% SDS (w/v)

Sigma

2000

1 M Tris pH 6.8

Melford

800

1% Bromophenol Blue (w/v)

Sigma

100

Glycerol

Amersham

1500

Distilled H2O

Self

4600

Material 10x Running Buffer

Supplier

Tris (30.2 g)

Supplier Melford

89

Preparation PH adjusted to 8.3 and made up to 1 l. For SDS-PAGE, 5

Glycine (144 g)

BDH

ml of 10% SDS was added to 495 ml of 1x buffer.

Table 2.8. SDS-PAGE and western Blott reagents.

Chapter 3 Investigating Notch signalling in chronic myeloid leukaemia 3.1 Introduction The role of Notch signalling in normal haemopoiesis and in malignant transformation is well established. It was originally found that the Notch1 receptor gene is expressed in human CD34+ hematopoietic precursors, including the more primitive CD34+ Lincell subset (Milner et al. 1994). Subsequently it was shown that Notch1 is also expressed in lymphoid, myeloid, and erythroid precursor populations, as well as in more mature progentors (Milner and Bigas, 1999) suggesting that Notch functions in multiple lineages and at various stages of haemopoiesis. The expression of the Notch ligand Jagged was also reported in human bone marrow stroma and Jagged1-Notch signalling was shown to promote the cell survival of human CD34+ cells (Walker et al. 1999). Interestingly, Notch1 and Notch4 transcript expression were found to be expressed at significantly higher levels in the more primitive human CD34+ CD38- populations as compared with the more mature CD34+ CD38+ progenitors (Vercauteren and Sutherland, 2004). The authors also found that constitutive activation of Notch1 or Notch4 in human CD34+ lin- cells results in the maintenance of stem cells as shown by the increase in long-term culture initiating cells in vitro.

90

The link between Notch signalling and leukaemia has been established in T-ALL in which the t(7;9) breakpoint translocations involving the Notch1 gene results in expression of constitutively activated intracellular Notch1 protein which has been shown to induce T-ALL in a mouse transplantation model (Pear et al. 1996). The role of dysregulated Notch signaling in T-ALL has been further emphasized by the finding that more than 50% of human T-ALLs have activating mutations that involve the extracellular heterodimerization domain and/or the C-terminal PEST domain of Notch1 (Weng et al. 2004). The role of Notch signalling in chronic myeloid leukaemia is not well characterised. Notch signalling activity has been reported to be downregulated in the blastic phase of CML as compared to the chronic phase of the disease (Sengupta et al. 2007). However, this study was limited in that Notch signalling was only investigated in the total CD34+ cells rather than in the leukaemic stem cells in the chronic phase of CML. It also did not evaluate Notch signalling activity in the chronic phase of CML as compared to normal haematopoeitc stem cells.

The aim of this chapter is to investigate Notch signalling in the CD34+ cells in the chronic phase of CML by measuring the gene expression levels of Notch receptors and their target genes by conventional and real time PCR. The expression of Notch1 will also be investigated at the protein level using flow cytometry.

3.2 Results 3.2.1. Gene expression analysis Poly-A cDNA samples derived from both normal bone marrow samples and CML samples were used to study the expression patterns of Notch receptors and Notch target genes by conventional semiquntitative PCR. The results of these experiments were further quantified by real time PCR studies. In all gene expression studies, GAPDH was used as a housekeeping gene. 91

3.2.1.1 Expression pattern of Notch genes in CML The expression of Notch1, Notch2, Notch3, and Notch4 receptors was studied in four CML patient samples along with four normal bone marrow samples which had previously been prepared in the lab by Dr. S. Ainsworth using the PolyA PCR technique. Cells in each sample had been fractionated into CD34+ Thy+, CD34+ Thyand total CD34+ subsets to enable the study of gene expression in haemopoietic progenitors at different maturation levels and sorted cells were of 95% purity. Figure 3.1 shows the PCR profiles of both normal and CML samples. The housekeeping gene glyceraldehydes-3-phosphate dehydrogenase (GAPDH) was used to assess the quality of cDNA samples and to check the uniformity of DNA content among different samples. Notch1 was expressed in all normal samples and in all three haemopoietic CD34+, Thy+, and Thy- subpopulations. One exception was the CD34+ population in NBM4 which was surprising since both the Thy+ and Thy- subsets expressed Notch1. This was also the case for the CML samples with no clear evidence of differences in the expression between the CD34+, Thy+, and Thy- subpopulations were seen (Figure 3.1). Notch2 was weakly expressed in normal and CML samples and in all three CD34+ subpopulations (Figure 3.1). The initial PCR analysis was done at 28 cycles and showed very weak bands in normal cDNA samples so the number of amplification cycles was increased to 30 cycles in order to visualise clearer bands. The annealing temperature for Notch2 primers was also adjusted after determining the optimum annealing temperature to be 58 °C by gradient PCR. Under these conditions weak bands could be seen in all samples. The study of Notch3 on normal and CML samples did not reveal any message using the PCR conditions and the set of primers described here (results not shown). This was repeated and the activity of Notch 3 primers was confirmed on human genomic DNA, as a positive control, where PCR showed clear bands under the same experimental conditions.

92

Notch4 was not constantly expressed and was seen in 2/4 normal CD34+ and in 2/4 CML CD34+ samples (Figure 3.1). In order to determine whether there were any quantitative differences between the Notch expression seen in normal BM and CML CD34+ populations real time PCR was performed. Data from Real time PCR experiments showed no significant increase in Notch1 expression in CD34+ CML samples as compared to normal bone samples. Similarly there was no difference in Notch1 expression in the CD34+ Thysubpopulation. However, a 3-4 fold increase was seen in CD34+ Thy+ cell subset. A Mann-whitney statistical test showed that this upregulation was only significant in the most primitive CD34+ Thy+ cell subset (Figure 3.2). Notch2 was shown to be overexpressed in all CML samples and in all different subpopulations investigated here by real time PCR (n=4) (Figure 3.3). There was more than a 100 fold increase in Notch2 expression in the CD34+ Thy+, CD34+ Thy- , and in the total CD34+ cell subsets as compared with NBM samples. However, the increased level of expression was only significant in the CD34+ Thy+ and in the total CD34+ cell subsets (P value= 0.02 for both cell subsets). Although real time PCR showed an upregulation of Notch2 in the CD34+ Thy- cell subset, this was not statistically significant by the Mann-Whitney test (P value= 0.057).

3.2.1.2 Expression pattern of Notch target genes The expression of Notch target genes Hes1, Herp1, and Herp2 was studied in CML in order to assess the activity of Notch signalling in CML as compared to normal CD34+ cells. Results showed that neither Herp1 nor Herp2 were expressed in either normal marrow samples or CML patient samples (Figure 3.4). Interestingly, Hes1 appeared to be expressed in some normal samples and some CML samples with no precise pattern of activity discernable. This suggests that Notch signalling was activated in these samples. There was no obvious pattern of Hes1 expression at the haemopoietic differentiation levels in the normal or CML samples studied here.

It is worth

mentioning that the initial number of PCR cycles for HES1 was 30 cycles. This had to

93

be increased to 32 cycles to clearly show the difference in gene expression between normal samples and CML patient samples (Figure 3.4). Quantitative real time PCR analysis demonstrated that that Hes1 was overexpressed in CML. A greater than 100 fold change increase in the Hes1 expression was observed in all CML samples and in all the CD34+ cell subsets studied (n=4) (Figure 3.5). The Mann-Whitney statistical test showed that the overexpression of Hes1 in the CD34+ Thy+ and total CD34+ cell subsets was highly significant (P= 0.007) and significant (P= 0.0.2) respectively. On the other hand, the upregulation of Hes1 in the CD34+ Thy- cell subset was just outside the biological level of significance (P= 0.057).

GAPDH

NBM9

Thy+ Figure 3.1. Notch expression of receptor genes in CD34+ populations isolated from normal bone marrow (NBM) and CML samples. PCR reactions are shown for four normal bone marrow samples on the left panel and for four CML samples on the right panel. For each sample the expression within each of the CD34+, Thy-, Thy+ subpopulation is shown. The gene expression profiles for Notch1, Notch2, and Notch4 are shown. Notch3 is not shown as no gene expression was seen in any of the NBM and CML samples. The housekeeping gene GAPDH was used as a control to assess the quality of cDNA in each sample. The number of PCR amplification 94 cycles was 28. Lower left panel shows human genomic DNA (HGDNA) which was used as a positive control for each set of oligonucleotides.

Thy -

Figure 3.2. Real time PCR analysis of Notch1(N1) expression on CD34+ cell subsets from NBM and CML patients. Gene expression was normalised to the GAPDH house keeping gene and represented as DCt values. For each sample the mean DCt value was calculated. Comparison of gene expression between NBM and CML samples was derived from subtraction of NBM DCt values from CML DCt values to give a DDCt value, and relative gene expression was calculated as 2-DDCt. Significant upregulation was observed in the CD34+ Thy+ cell subset (P≤ 0.05). Statistical significance was calculated using Mann-Whitney test. (* = P ≤0.05). 95

*

Figure 3.3. Real time PCR analysis of Notch2 expression on CD34+ cell subsets from NBM and CML patients. Gene expression was normalised to the GAPDH house keeping gene and represented as DCt values. For each sample the mean DCt values was calculated. Comparison of gene expression between NBM and CML samples was derived from subtraction of NBM DCt values from CML DCt values to give a DDCt value, and relative gene expression was calculated as 2-DDCt. Results showed an upregulation of Notch2 in CD34+ CML cells. This upregulation was significant in both the CD34+ Thy+ and in the total CD34+ cells (P≤ 0.05). The expression of Notch2 in the CD34+ Thy- was very close to significance (P= 0.057). Statistical significance was calculated using Mann-Whitney test. (* = P ≤0.05).

100000 96

GAPDH Thy+

NBM9

Figure 3.4. Expression of Notch target genes in CD34+ populations isolated from NBM and CML samples. PCR reactions for HES1, HERP1, and HERP2 are shown for four different normal bone marrow samples (NBM) (left panel) and four different CML samples (right panel). The number of PCR amplification cycles was 28 except for HES1 where the PCR reaction was carried out at 32 cycles. All samples were run in duplicates and the house keeping gene GAPDH was used to assess the quality of each cDNA sample. The lower left panel shows human genomic DNA (HGDNA) which was used as a positive control for each set of97 oligonucleotides.

Thy -

Figure 3.5. Real time PCR analysis of Hes1 expression on CD34+ cell subsets from NBM and CML patients.Gene expression was normalised to the GAPDH house keeping gene and represented as DCt values. For each sample the mean DCt values was calculated. Comparison of gene expression between NBM and CML samples was derived from subtraction of NBM DCt values from CML DCt values to give a DDCt value, and relative gene expression was calculated as 2-DDCt. Results showed an upregulation of Hes1 in CD34+ CML cells. This upregulation was significant in both the CD34+ Thy+ cell subset (P≤ 0.01) and in the total CD34+ cells (P≤ 0.05). The expression of Hes1 in the CD34+ Thy- cell subset was very close to significance (P value = 0.057). Mann-Whitney statistical test was used here to compare the difference in Hes1 expression between CML and NBM samples. (* = P ≤0.05, ** = P ≤0.01).

1000000

98

3.2.2 Flow cytometric analysis of Notch1 in CML The gene expression studies in section 3.2.1 showed that Notch1 and Notch2 were upregulated at the message level in the CD34+ Thy+ population in CML patients in chronic phase. This finding along with the observation of elevated levels of Hes1 in CD34+ populations including CD34+ Thy+ susbset suggested that Notch signalling may be overactivated in CML patients. This observation raised the possibility that Notch1 may be also over-expressed at the protein level. To address this question, mononuclear cells from three CML patients in chronic phase were stained with EA1, a monoclonal antibody produced in the Buckle lab that recognises the extracellular domain of Notch1 (Dr. V. Portillo, personal communication, Appendix 1). Initial FACS staining experiments demonstrated the presence of Notch1 on the cell surface of live CD34+ population from CML patients. When FACS profiles were compared to those in normal cord blood samples, broadly similar staining profiles were seen. The analysis of expression of Notch1 in CML was then expanded to cover a wide range of primitive stem cell populations, myeloid, and lymphoid committed cells using lineage specific surface markers expressed within the CD34+ compartment. These markers were selected to demonstrate the expression of myeloid progenitors (CD33, CD14, CD15, CD16), B-cell lymphoid progenitors (CD19), T-cell lymphoid progenitors (CD3, CD7), and the more primitive Thy-1 (CD90) haemopoietic stem cell marker. Another primitive cell discriminator used was CD38, as stem cells are enriched in the CD38- subset of the CD34+ compartment.

99

Using multi-parameter flow cytometry, triple colour stainig was used with a CD34 monoclonal antibody, a lineage/ primitive cell marker, and the EA1 antibody included in all tubes. All antibodies were of the IgG1 isotype. To correct for background fluorescence in each FACS tube, a mouse IgG1 hybridoma supernatant, which was tested to be negative for Notch1 protein (Dr. V. Portillo, personal communication), was used as an isotype control for the EA1 antibody. Since cryopreserved CML samples were used in the analysis, thawed cells were filtered by nylon mesh filter to remove large clumps of dead cells. Dead cells were also excluded from the analysis by staining cells with propidium iodide (PI) and acquiring only live cells. FACS analysis showed that the percentage of CD34+ population varied between 7 % and 30 % out of the total mononuclear cells in the CML samples examined here. From this CD34+ population, 35 % (± 1.8) of cells expressed Notch1 (n=4) (Fig. 3.6). Notch1 was expressed on the surface of subpopulations of (CD34+ CD14+, and CD34+ CD33+) myeloid progenitors. FACS staining did not detect neither CD34+ CD15+ nor CD34+ CD16+ myeloid progenitors (Figure 3.6). There was no or little expression of B-lymphoid (CD34+ CD19+) and of the Tlymphoid (CD34+ CD3+) progenitors but CD34+ CD7+ progenitors could be found. Interestingly 24.5 % ± 9.6 of the CD34+ CD7+ progenitors expressed Notch1 (n=4) (Fig. 3.7). Finally the presence of Notch1 in the very primitive haemopoietic CD34+ Thy+ and CD34+ CD38- progenitors was assessed (Figure 3.8 & Figure 3.9). The initial staining of EA1 in CD34+ Thy+ cells showed a diagonal pattern of staining which seems non-specific and could not be corrected using the software compensation tool during acquisition of cells before analysing the expression pattern. Therefore, the staining was repeated with an extra step of blocking with purified mouse IgG to prevent non-specific binding in the reaction which was successful in improving the pattern of staining and produced well compensated FACS plots (Figure 3.9). The data obtained from three CML samples showed that Notch 1 was expressed in 21.6 % of ± 2.3 of the CD34+ Thy+ population (Figure 3.10).

100

Table 3.1 indicates percentages of Notch1 expression among different CD34+ cells subsets in CML samples as compared with expression on cord blood (Cord blood data provided by Dr. V. Portillo).

A IgG1

CD14

C

101

Figure 3.6. Notch expression on CD34+ myeloid progenitors in CML. Mononuclear cells from CML samples were stained with CD34, specific myeloid lineage markers and the anti extra cellular Notch (antibody EA1). Staining for CD34 gated cells is shown. Panels A, C, E, and G show costaining with lineage marker and isotype control. Panels B, D, F, and H show costaining with lineage marker and the EA1 antibody. Panel I shows costaining with CD34 and isotype control and panel K shows costaining with CD34 and EA1.IgG1 hybridoma supernatant was used as an isotype control. Data shown is from one experiment representative of three different patient samples.

102

A IgG1

103

Figure 3.7. Notch expression on CD34+ lymphoid progenitors in CML. Mononuclear cells from CML samples were stained with CD34, specific lymphoid lineage marker and the anti extra cellular Notch antibody EA1. Staining for CD34 gated cells is shown. Panels A, and C show costaining with lymphoid lineage marker and isotype control. Panels B, D, and F show costaining with the lineage marker and the EA1 antibody. IgG1 hybridoma supernatant was used as an isotype control except with CD19 in which mouse IgG1 FITC and PE (panel E) was used to set the quadrant markers for background fluorescence signal. Data shown is from one experiment representative of three different patient samples.

(A )

SSC Figure 3.8. CD34 gating strategy and the Notch expression in CD34+ CD38- cell subset in CML. Mononuclear cells from CML samples were costained with CD34 and anti Notch1 antibody (EA1).and the stem cell markers CD34, and CD38. The first plot (A) shows the CD34 gating strategy used in all FACS plots in this study. The expression of Notch1 in the total CD34+ population in CML is shown in the right hand plot on (A) as compared to the isotype control IgG1 in the middle plot. Panel B shows the Notch1 expression in the primitive CD34+ CD38- cell subset which is enriched for stem cells. The percentage of Notch1 in the CD34+ CD38- is 15.3 ± 2.1 104 (n=3).

A IgG1

Figure 3.9. The problem of EA1 non-specific binding within the CD34+ Thy+ cell subset. The expression of Notch1 in the CD34+ Thy+ cell

B

subset showed a nonspecific diagonal pattern of staining (panel A) before it was corrected by including extra blocking step in the reaction (panel B). After staining with EA1 antibody, cells were incubated with purified mouse IgG1 for 10 minutes before the addition of CD34 antibody in the final step of the reaction. Because both EA1 and CD34 were mouse antibodies, the blocking step seems to be necessary to 105 prevent nonspecific binding of CD34 antibody. This approach was repeated twice in two different CML samples.

106

Figure 3.10. The expression of Notch1 in the CD34+ Thy+ cell subset. Mononuclear cells from CML samples (N=3) were costained with EA1 and the stem cell marker thy-1. The upper panel shows the gating strategy where only cells positive for both CD34 and Thy-1 antibodies where used in the analysis of EA1 expression. The lower panel shows that Notch1 is expressed in the primitive CD34+ thy+ population in CML. The percentage of EA1 in CD34+ Thy+ cells is 21.6 ± 2.3 (n=3). IgG1 hybridoma supernatant was used as an isotype control.

Cell subset

Mean of expression of EA1 in CML (±SEM)

Mean of expression of EA1 in CB (±SEM)

Total CD34+

35 ±1.8

21 ±10

CD34+ Thy-1+

21.6 ± 2.3

12 % ± 3

CD34+ CD14+

4.5 ±2.1

ND

CD34+ CD15+

9 ±6.6

ND

CD34+ CD16+

0.5 ±0.5

ND

CD34+ CD33+

33.5 ± 19.9

26 ±5

Table 3.1. The average expression of EA1 in different cell lineages in CML and cord 33 ±7 CD34+ 24.5 ±9.6 blood (CB).CD7+ FACS analysis of Notch1 in different myeloid, lymphoid, and more primitive lineages in CD34+ CD3+by costaining mononuclear 3.6 ±1.4 cells with both EA1 antibody ND and a lineage CML was done specific cell surface marker. Results shown here are representative of the total CD34+ 20 ±7 CD34+ CD19+ ±1.6 cells in each sample. The mean of1.6 expression refers to percentage of each cell population in the left column that was positive for EA1. The means of expression were measured CD34+ 15.3 from fourCD38different CML samples (n=4). The data from CD34+ 15 CML cells were compared to data obtained from crod blood samples (Right column). The EA1 expression was investigated in the CD34+ CD38- cell subset in two CML samples and two CB samples (n=2). 107

3.3. Discussion 3.3.1 Expression pattern of Notch genes in CML A study of Notch receptor genes N1-4 was carried out in four CML patients in chronic phase in order to investigate the status of the Notch signalling pathway in CML. Since CML is a disease that originates in the haemopoietic stem cell compartment, cDNA from CD34+ cells were used in order to have data representative of the progentitor compartment. The study of Notch signalling in CML was restricted to the chronic phase of the disease where the only known genetic abnormality is the BCR-ABL fusion gene. Since cells acquire several and complex genetic changes beside the BCRABL fusion gene as they transform to blast crisis phase, it becomes more difficult to interpret interactions between Notch signalling components and the BCR-ABL. The expression pattern of the Notch receptor genes has not been studied in CML before. The conventional PCR data showed that there was no preferential expression of the gene in the three fractionated cell subsets studied. It appeared that the expression of Notch1 during the chronic phase of CML can be traced at least to the very primitive CD34+ Thy+ haemopoietic cells and the expression continues in the less primitive CD34+ Thy- cell compartment. The highly sensitive real time PCR approach confirmed the previous findings. Interestingly, the quantitative real time PCR data showed an up-regulation of Notch1 in CML samples. This up-regulation was significant in the more primitive CD34+ Thy+ cell subset in the bone marrow. 108

Apart from gene array expression profiling studies, expression of Notch receptor genes has not been fully characterised before in CML. Bruchova et al (2002) used an array technology to study the gene expression profile of hundreds of genes in the chronic phase of CML and showed that Notch1 is down-regulated in CML mononuclear cells. However, their findings cannot be compared with the results of this study because of differences in sampling and techniques between the two studies. The current study looked at Notch1 expression only in the CD34+ cell subsets in CML including the mostly enriched stem cell CD34+ Thy+ cell subset whereas Bruchova group sample was rather heterogeneous and included all mononuclear cells. The finding that Notch1 is significantly up-regulated in the most primitive CD34+ Thy+ in the chronic phase of CML is interesting. This raises the possibility that Notch signalling is involved in the survival or self-renewal of leukaemic stem cells in CML. This possibility is supported by a recent study which demonstrated that leukaemic stem cells in CML are dependent for their self-renewal on the Wnt pathway, a pathway that like the Notch pathway is important for normal haemopoietic stem cell survival and self-renewal. Zhao et al (2007) showed, using a conditional β-catenin

-/-

mouse, that in vivo chronic myeloid leukaemia progression is dependant on β-catenin and that loss of β-catenin significantly reduces BCR-ABL- induced CML development. As for Notch2, normal bone marrow samples (NBM) showed very weak expression in all samples when the PCR reaction was performed at 28 cycles. Interestingly, at 30 cycles, the expression of Notch2 appeared to be expressed in NBM and CML samples with no obvious favoured expression in one of the three cell subsets studied here. Although Notch2 was difficult to detect by conventional PCR, real time PCR data showed clearly an up-regulation of Notch2 by more than 100 fold in all of the three CD34+ cell subsets tested here. Although there was no obvious favoured expression in any one of the three cell subsets, the over-expression of Notch2 was significant in the CD34+ Thy+ and in the total CD34+ cell subset.

109

The transcripts of Notch3 could not be detected in normal bone marrow and CML samples. Primer specificities and PCR reaction conditions for Notch3 were validated by applying the same conditions on human genomic DNA and Jurkat cells (a human T cell lymphoblast-like cell line). The results showed clear bands for Notch3 in both genomic DNA and Jurkat cells which confirms the finding that Notch3 is absent in CD34+ cells in both normal marrow and in CML patients in chronic phase (Data not shown). The conventional PCR data demonstrated that Notch4 is sporadically expressed in both normal bone marrow and CML samples. Judging from the semi-quantitative PCR experiments, it looks that there was no obvious difference in the level of Notch4 expression between normal marrow samples and those of CML samples. However, it appeared that Notch4 may be preferentially expressed in CD34+ Thy+ cells in most CML samples (3 out of 4). It is not clear whether this observation may be of clinical significance or not. The presence of Notch1 and Notch2 genes on normal CD34+ haemopoietic cells is consistent with previous reports (Milner and Bigas, 1999; Vercauteren and Sutherland, 2004). Notch3 expression could not be detected in normal CD34+ cells under the conditions described here which is consistent with the results of Singh et al. (2000). In contrast, Vercauteren and Sutherland (2004) reported a low expression of Notch3 transcripts in normal CD34+ cells in bone marrow from normal healthy individuals. This discrepancy between this study results and those of Vercauteren and Sutherland could be explained by differences in the sensitivity of primers used in the two experiments. In addition, it could be that the Notch3 transcripts may be under the level of detection in normal CD34+ cells in the samples studied here. The PCR results presented here demonstrated the presence of Notch4 in some normal CD34+ haemopoietic cell samples. This expression is in line with the findings of others (Vercauteren and Sutherland, 2004). The current data that shows the presence of Notch4 in normal CD34+ haemopoietic cells is interesting because Notch4 has previously been thought to be an endothelial specific gene (Uyttendaele et al. 1996).

110

3.3.2 Expression patterns of Notch target genes in CML The detection of the intracellular domain of Notch receptors (ICN) in the nucleus, as a landmark of active Notch, has been proven to be very difficult. Since Notch signalling directly activates Hes1 transcription, the use of Hse1 expression level has been widely used as an alternative method to detect the activity of Notch signalling in haemopoietic system (Pear and Radtke, 2003). A possible implication of the up-regulation of Notch1 and Notch2 in CML is that Notch signalling may be activated. To test this, the expression of the Notch target genes Hes1, Herp1, and Herp2 was studied in the same CML samples. Although the message of both Herp1 and Herp2 could not be detected, it appeared that the Hes1 target gene was expressed in most CML samples. The semi-quantitative PCR experiments performed in this study were confirmed by real time PCR and the data suggests that Notch signalling is active in CML patients in chronic phase. The real time PCR data also showed that Hes1 is up-regulated in the CD34+ Thy+, CD34+ Thy-, and in the total CD34+ cell subsets as compared with NBM. This up-regulation was significant in the CD34+ Thy+ cell subset and total CD34+ cell subsets. Activation of Notch signalling in CML reported here can be attributed to in vivo stimulation of Notch signalling by Notch ligands expressed on the cell surface of stromal cells or on haemopoietic progenitors. It is also possible that mutations in the Notch receptor genes may induce activations of Notch signalling independent of ligand binding. However, it remains to be investigated if Notch is mutated in chronic phase CML. Effects of Notch activation may include effects on stem cell self-renewal or differentiation. A possible consequence of this activation may include the promotion of myeloid differentiation in the chronic phase of CML. This possible effect of Notch signalling in myeloid differentiation has been described before on myeloid cell lines (Schroeder and Just, 2000; Schroeder et al. 2003; Tohda et al. 2003). Although others have reported conditions where Notch signalling inhibit differentiation (Bigas et al. 1998), activation of Notch signalling on the CD34+ Thy+ subset is a strong candidate

111

for possibly leukaemic stem cell expansion as normal stem cells are shown to expand when Notch signalling is stimulated (Stier et al. 2002).

3.3.3 Expression of Notch1 protein in CML Over-expression of the Notch1 receptor in CML at the gene level warranted the investigation of Notch1 receptor protein expression in CML samples. EA1 is a novel monoclonal antibody which was generated in the lab by other members of the research team and its specificity for human Notch1 was confirmed by ELISA (Dr. V. Porttilo, personal communication). Unlike other antibodies available for human Notch1 which only detects the intracellular domain of Notch1 (ICN1), EA1 is the first available antibody that recognises the extracellular domain of Notch1 (ECN1). This characteristic of EA1 avoids the need for permeabilizing cells before staining them which is required by other anti human Notch1 antibodies. Therefore, EA1 can be used to stain live intact blood cells in multiparametric flow cytometric approach. The antibody also specifically detects Notch1 protein on the surface of the cell is available for ligand binding and subsequent signal transduction. The use of flow cytometry in the investigation of Notch1 protein allowed the characterisation of the expression levels of human Notch1 receptor on different haemopoietic progenitors within the CD34+ population in CML. Study of the expression patterns of Notch1 in CD34+ cells from cord blood samples was performed previously by Dr. Porttilo and these data were used here as normal control for the expression of Notch1. The flow cytometric staining results in CML samples demonstrated the expression of Notch1 in 35 % of gated CD34+ cells compared with only 21 % of CD34+ cells in cord blood. The presence of low expression of Notch1 protein has been reported before in normal CD34+ cells in normal bone marrow (Ohishi et al. 2000) and in embryonic liver (Dando et al. 20005). Notch1 was expressed in lymphoid and myeloid haemopoietic progenitors within the CD34+ cells in CML samples. Notch1 was also detected in the very primitive CD34+ Thy+ and CD34+ CD38- populations in CML samples. The percentage of CD34+

112

Thy+ cells expressing Notch1 in CML samples was 21.6 ± 2.3 (n=3) compared with 12 % ± 3 in cord blood samples. The difference in Notch1 expression between the CD34+ Thy+ cell subset in CML and that in cord blood was not significant by the Mann-Whitney statistical test. Because of this and because of the low number of CML samples analysed here no conclusions could be drawn from this variation. The expression of Notch1 was also confirmed in another stem cell enriched subset in CML which is the CD34+ CD38- cell subset. The mean percentage of cells expressing Notch1 in this primitive compartment was 15.3 % (n=2) compared with 15 % in CD34+ cells in cord blood suggesting that the expression was similar between normal and CML cells. There were very low or undetectable numbers of CD34+ CD3+ and CD34+ CD19+ cells in the CML samples tested here which did not allow the study of Notch1 in these two populations. This low level of expression was in correlation with the immunophenotype of CD34+ cells in CML in chronic phase (Normann et al. 2003). However, it is documented that these populations express Notch1 in cord blood samples (Dr. V. Porttilo, personal communication). FACS analysis showed the presence of Notch1 in CD34+ myeloid progenitors including CD34+ CD14+ cells and CD34+ CD33+ cells. It is not clear why only about 20% of CD34+ CD33+ myeloid cells express Notch1 in contrast to the CD34+ CD14+ cells in which most of them co-express Notch1. CD16+ cells could not be detected whereas the total number of granulocytes (CD15+) in CML samples was unexpectedly very low (3-5 %). It is not clear whether this is normal in cryopreserved CML samples or whether this is an artificial event due to possible prolonged cryopreservation or thawing and handling of CML specimens. The data also showed the co-expression of the lymphoid marker CD7 on CD34+ CML in approximately 40% of gated CD34+ cells, a finding that has been shown to be of prognostic importance in CML patients in chronic phase (Normann et al. 2003). 24.5 % (±9.6) of the CD34+ CD7+ cells expressed Notch1 on their surface. CD7 is not only expressed in mature T- and natural killer (NK) cells, but it is regarded as an early haemopietic marker (Normann et al. 2003). It has been suggested that CD34+ 113

CD7+ cells may include very primitive stem/progenitor cells capable of differentiating into lymphoid and myeloid lineages (Chabbanon et al. 1992). Recently, it has been suggested that CD34+ CD7+ cells may be involved in maintenance and clonal progression of Ph-positive cells in CML patients in chronic phase (Kosugi et al. 2005). It can be speculated, therefore, that the expression of Notch1 in the primitive CD34+ CD7+ population is of clinical significance. No clear conclusions can be drawn from the study of Notch1 protein in CML samples in terms of the presence of elevated Notch1 expression in CML. Even if Notch1 upregulation in CML at the m-RNA level did not clearly translate into increased protein levels this does not contradict with the finding that Notch signalling is activated in CML as assessed by Hes1 up-regulation. The activation of Notch signalling does not necessarily occur via ligand dependant mechanisms. For instance, Mutations in Notch1 in T-ALL render Notch1 susceptible to ligand-indpendant cleavage at S2 site and subsequently be constitutively active regardless the fact it is not overexpressed on the cell surface (Aster et al. 2008). Mizuno et al. (2008) have reported that Notch1 is a common retroviral integration site in which retrovirus integration formed a constitutively active form of Notch and accelerated leukaemia development in a CML mouse model. Considering the fact Notch1 mutations have not been examined yet in CML, such mutations may explain the hyperactivity of Notch in CML reported in this study. Moreover, fusion oncoproteins PML/RAR and AML1/ETO in AML have been associated with activation of Notch signalling which may confer self-renewal properties to leukaemic stem cells in AML (Alcalay et al. 2003). It is tempting therefore to propose that BCR-ABL fusion protein may crosstalk with Notch signalling to confer survival signals in CML. Nonetheless, this is the first time the presence of Notch1 protein is reported in the CD34+ population in blood samples of CML patients in chronic phase. Moreover, the expression of Notch1 at the protein level in the CD34+ Thy+ and CD34+ CD38- cell subsets is interesting because these populations are enriched for leukaemic stem cells in CML.

114

Chapter 4: Investigation of BCR-ABL and Notch cross-talk in cell line models 4.1Introduction The data from chapter three showed that Notch signalling, as assessed by levels of Hes1 expression, is upregulated in CD34+ cells, including the more primitive CD34+ THY-1 + cell subset in chronic phase CML patients as compared to CD34+ cells from normal donors. Activation of the Notch pathway has been shown to confer cell survival properties on normal haemopoietic stem cells (Stier et al. 2002; and Carlesso et al. 1999) and this raises the possibility that leukaemic stem cells (LSCs) in chronic phase CML may benefit from activated Notch signalling conferring survival signals. It has been found that inhibition of BCR-ABL alone by imatinib mesylate (IM) does not result in loss of self renewal capacity or cell death within the primitive CD34+ CD38- cell compartment in chronic phase CML in vitro (Copland et al. 2006). This has led some researchers to propose a model which suggests that BCR-ABL requires cooperating genetic events at the stem and/or progenitor level to establish a Ph+ leukaemia (Burchert et al. 2007). Recently, Mizuno et al. (2008) have demonstrated that overexpression or enhanced kinase activity of BCR/ABL and altered expression

115

of Notch1 synergises to induce acute leukemia in a transgenic model for CML. This finding and previous genetic interaction evidence of ABL and Notch synergism in Drosophila development (Giniger, 1998; and Crowner et al. 2003) may justify the need to investigate the hypothesis of possible cross-talk between BCR-ABL and Notch in CML. A range of model systems including cell lines derived from leukaemia patients and animals engineered to express BCR-ABL have been used in the past to study the molecular interactions between BCR-ABL and other signalling pathways (Ren, 2005). In this project leukaemic cell lines will be investigated as possible candidate models to explore the proposed interaction of BCR-ABL and Notch signalling in human CML cells. A good candidate model for the study of BCR-ABL and Notch cross-talk should fulfill the basic criteria of having an intact BCR-ABL and Notch components, and to have a demonstrable response to Notch and BCR-ABL pathways modulators. CML cell lines present a suitable model system in which to investigate the molecular mechanisms underlying signalling pathways involved in the pathogenesis of CML and to identify potential therapeutic targets. In fact, much of our understanding of the basic biology of CML comes from studies that have used BCR-ABL+ cell lines. One draw back is that all the available CML cell lines are derived from the blastic phase of the disease and, thus, contain genetic lesions in addition to BCR-ABL. The K562 line has been used in the cloning and analysis of the t(9;22)(q34;q11) breakpoints which involve the genes ABL and BCR (Drexler et al. 1999). In addition, many of the signalling pathways downstream of BCR-ABL and the proteins that interact with BCR-ABL were identified in BCR-ABL+ cell lines (Ren, 2005). An important advantage of the CML cell lines is that most of them remain dependent on BCR-ABL tyrosine kinase activity for their proliferation and survival, as shown by their susceptibility to the effects of BCR-ABL inhibitors (Deininger et al. 2000). This is a very useful criterion of the BCR-ABL+ cell lines as in vitro model systems in which the BCR-ABL tyrosine kinase activity can be turned off with imatinib to study activity of other signalling pathways.

BCR-ABL activity assay

116

The measurement of BCR-ABL activity in CML is critical for the assessment of the disease progression and response to BCR-ABL targeted therapy. The nuclear adaptor protein Crkl has been reported as the major and constitutive tyrosine-phosphorylated protein in chronic phase CML patients and in CML cell lines (Oda et al. 1994). Another study showed that the level of crkl phosphorylation correlated well with the level of BCR-ABL expression and that BCR-ABL tyrosine kinase activity can be determined by measuring the phosphorylation of its down stream substrate crkl (Hoeve et al. 1994). Barnes et al. (2005) showed that the levels of Phosphorylated crkl (P-crkl) correlate very well with BCR-ABL levels in different stages of CML and that P-crkl expression can be used as an indicator of disease progression. Due to its specificity and stability, the expression of P-crkl has been accepted as a reliable method to assess BCR-ABL status and have proved to be a vital practical method for evaluating the effect of imatinib treatment on BCR-ABL kinase activity in CML cell lines and in primary CML cells (Gorre et al. 2001; and Singer et al. 2006). In all of the studies cited so far Western blotting was used as a method to assess Pcrkl expression in BCR-ABL expressing cell lines and primary CML cells. However, this approach is time consuming and requires large number of cells which may not be achievable in the more primitive cell populations in primary CML cells. The recent availability of phosphorylation-specific antibody that detects only phosphorylated crkl in BCR-ABL + cells has made it possible to use flow cytometry instead of Western blotting to measure P-crkl levels in CML cells (Wetzel et al. 2005). However, this approach has only been used by two groups to date. The P-crkl flow cytometry based assay therefore will be validated in this chapter with a view to its use as a tool for the assessment of BCR-ABL activity in the context of BCR-ABL and Notch cross-talk. Upregulation of the Notch target gene Hes1 has been shown to occur as a result of Notch activation and association with the transcription factor RBP-Jk (Jarriault et al.. 1995). Since then the expression of Hes1 has been widely used and accepted as an indicator of Notch activity (Kageyama et al. 2000). Hes1expression therefore will be used in this project to assess Notch signalling in cell lines as well as in primary cells.

Aims of this chapter:

117

The aims of this chapter were: 1- To establish the optimal staining conditions and appropriate controls for the FACS based P-crkl assay before it is being utilised in this project as a marker for ABL kinase activity and as in vitro sensitivity assay for imatinib mesylate. 2- To validate a human cell line based in vitro model system for the study of the cross-talk of ABL and Notch signalling. 3- To investigate the cross-talk between ABL and Notch signalling in cell line model systems using inhibitors of both signalling pathways.

.

4.2 Results 4.2.1 Validation of the P-crkl intracellular FACS assay in K562 cells The ability to detect phosphotyrosine proteins in BCR-ABL+ cell lines by flow cytometry was first published by Desplat et al. (2004). The authors used a Phosphotyrosine (p-tyr) monoclonal antibody to analyse by FACS the intracellular level of the tyrosine hyperphosphorylation of cellular proteins induced by BCR-ABL in BCR-ABL+ cell lines. This approach was further improved by the availability of a monoclonal antibody which specifically recognises phosphorylated crkl, a nuclear adaptor protein and a substrate of BCR-ABL that is constitutively phosphorylated in CML. P-crkl expression levels have been accepted as a surrogate assay to assess the constitutive BCR-ABL tyrosine kinase activity in BCR-ABL+ cells. At the time of this project two groups had attempted, with different protocols, to analyse P-crkl expression in BCR-ABL+ cells by flow cytometry (Wetzel et al. 2005; and Hamilton et al. 2006). To use the FACS based P-crkl assay as a method for the assessment of BCR-ABL activity in this project, it was essential to validate the assay and optimise its parameters and controls.

118

The K562 cell line is a CML-derived leukemia cell line, established in 1970 from the pleural effusion of a 53-year-old woman with CML in myeloid blast crisis (Drexler et al. 1999). K562 cells were used as a positive control and as a model to validate the various staining steps and conditions of the intracellular FACS staining of P-crkl. This is because K562 cells are BCR-ABL+ and have been shown to have high levels of phosphorylated crkl (Hoeve et al. 1994). To control for non specific binding a rabbit IgG polyclonal antibody at equivalent concentration of the P-crkl antibody was used as an isotype control in each experiment. To stain for P-crkl in K562 cells the cells were re-suspended in 100 µl of fixative (Fix and Perm kit, Caltag, UK) for 15 minutes and washed next with 3 ml of HBSS (5% FBS) before being incubated for 40 minutes with 25 µl permeabilising reagent and the primary anti-P-crkl antibody (New England Biolabs (UK) Ltd, Hitchin, UK).Cells were then washed twice and incubated for 40 minutes with a PE conjugated anti rabbit secondary antibody (BD, UK). The wash step was repeated twice and the cells were analysed by flow cytometry. Parallel intracellular staining experiments were also performed in K562 cells to investigate the effect of critical staining conditions such as fixation, secondary antibody background signal, and optimal concentration for primary antibody. The data showed that fixation and permeabilisation of K562 cells alone without antibody staining showed higher background fluorescence signal than that of unfixed cells (Fig. 4.1). K562 cells stained with the P-crkl antibody and the PE conjugated anti rabbit secondary antibody (BD) showed a distinct fluorescence signal that was clearly distinguishable from the isotype control signal. Interestingly, the staining of K562 cells with only the secondary antibody showed a low background signal as compared with the isotype control. In another experiment the P-crkl antibody was titrated to determine the lowest concentration at which the antibody can be used. This showed that the P-crkl antibody can be still used at 1:40 dilution without obvious loss of fluorescence signal intensity as compared to 1:10 dilution recommended by the manufacturer (Fig 4.1). Selection of the best secondary antibody can improve immunostaining and reduce false positive or negative staining. Therefore, the specificity and fluorescence intensity of four commercial anti rabbit secondary antibodies in the P-crkl assay was investigated. K562 cells were fixed and permeabilised as described above. Cells were then stained with P-crkl primary antibody diluted 1:40 and then stained with one of the following secondary antibodies: PE F(ab')2 Donkey anti-Rabbit IgG (BD), PE 119

polyclonal donkey anti-rabbit antibody (Abcam), FITC monoclonal mouse anti-rabbit antibody (Sigma), FITC polyclonal goat anti-rabbit antibody (Caltag) (Fig. 4.2). The FACS analysis showed that the PE conjugated secondary antibody from BD was superior to the Abcam antibody in terms of specificity and signal intensity. The Abcam PE secondary antibody showed little P-crkl expression in K562 cells and this expression was less specific as demonstrated by the overlap between the fluorescence peak of the P-crkl stained cells and that of the isotype control. K562 cells stained with P-crkl and the Sigma FITC secondary antibody showed a staining pattern similar to that of the BD PE secondary antibody in terms of specificity. However, the fluorescence signal was lower than that observed with the BD PE secondary antibody. In contrast, K562 cells stained with P-crkl and the Caltag FITC secondary antibody showed very strong fluorescence signal but with low specificity as compared with the Sigma FITC secondary antibody (Fig 4.2). The staining patterns of the previous secondary antibodies showed clearly that only the BD PE and Sigma FITC secondary antibodies are suitable to be used in the FACS based P-crkl assay with the advantage of the former in terms of fluorescence signal intensity. It can bee seen from previous data that the optimum staining conditions were obtained with the primary P-crkl antibody at 1:40 dilution (2.25 µg/ ml) and with the use of the BD PE secondary antibody. As for fixation and peremeabilisation methods, several kits were evaluated including Caltag fix and perm kit, BD Cytofix/ Cytoperm kit (cat. No 554722). In addition, different fixing times and temperatures and other permeabilisation methods were attempted such as cold 90% methanol for 30 minutes on ice. There was no obvious difference between all these fixation and permeabilisation methods in terms of P-crkl expression in K562 cells (data not shown). Therefore, an optimised staining protocol was used in all P-crkl assay experiments performed in this chapter which involves the use of the Caltag fix & perm kit followed by probing with primary P-crkl antibody at 1:40 dilution and then staining with BD PE secondary antibody.

4.2.2 The effect of cell passage number on the expression of P-crkl in K562 cells During the time of performing different optimisation experiments of the P-crkl assay the K562 cells were propagated for weeks in culture. It was noticed that the P-crkl expression was gradually lost on K562 cells over this time period. To confirm this

120

observation and to further investigate the effect of passage number on P-crkl expression in K562 cells a new batch of K562 cells was taken out of liquid nitrogen and monitored every two weeks for the expression of P-crkl. Data showed that the Pcrkl levels were stable for about 8 weeks in culture which is equivalent to 16 passages (Fig. 4.3). Around week 10 (passage 20) K562 cells showed reduced phosphorylation of P-crkl which was demonstrated by FACS analysis in the form of two populations of K562 cells with only one population maintaining the expression of P-crkl. K562 cells passaged for more than 12 weeks (> 24 Passages) lost the expression of P-crkl (Fig. 4.3). These findings were confirmed in three separate experiments at different time points before the completion of this project and led to the acceptance of only minimally passaged K562 cells as a positive control for the P-crkl assay and as a BCR-ABL+ in vitro model.

A

B 121

Fig 4.1. Validation of P-crkl intracellular flow cytometry assay in K562 cells. Critical staining conditions in the P-crkl intracellular staining were validated in K562 cells. (A) FACS analysis of the effect of fixation on background staining for P-crkl in K562 cells. Background staining on unfixed-unstained cells is shown in blue, and on fixed-unstained cells in green. P-crkl expression as determined after fixation is shown in red. (B) Analysis of effect of secondary antibody staining in P-crkl assay. Background staining on unfixed-unstained cells is shown in blue, and on fixed cells stained only with PE anti rabbit secondary antibody in ligh blue. Fixed cells stained with rabbit IgG (isotype control) and the secondary antibody are shown in green and P-crkl expression is shown in red. (C) Titration of the primary Pcrkl antibody. Cells were fixed and stained with different dilutions of the primary P-crkl antibody (red, green, and light blue histograms). Unstainedunfixed cells and isotype control (in concentration equivalent to the primary antibody dilution) are shown in blue and purple respectively.

A

B B D P E

122

Fig 4.2. Comparison of four commercial anti rabbit secondary antibodies used in the P-crkl assay. K562 cells were fixed, permeabilised and stained with P-crkl primary antibody diluted 1:40 and then stained with: (A) PE polyclonal donkey anti-rabbit antibody (BD) (B) PE polyclonal donkey anti-rabbit antibody (Abcam) (C) FITC monoclonal mouse anti-rabbit antibody (Sigma) (D) FITC polyclonal goat anti-rabbit antibody (Caltag). In each panel the P-crkl staining patterns of the secondary antibodies is shown in pink. Staining signals of unstained cells and cells stained with the isotype control are shown in blue and green respectively.

A

B Fig. 4.3. The effect of cell passage number on the expression of Pcrkl in K562 cell line. K562 cells were taken out from liquid nitrogen and maintained in culture for 12 weeks. Cells were passaged every 3-4 days and P-crkl expression was assessed by flow cytometry every two weeks. (A) FACS staining of P-crkl in K562 cells kept between 2-8 weeks in culture (passage 4-16). (B) FACS staining of P-crkl in K562 cells in week 10 (passage 20). (C) FACS staining of P-crkl in K562 cells passaged for more than 12 weeks (> 24 Passages). In each panel unstained k562 cells are shown in blue, isotype control in green, and k562 cells stained with PE conjugated P-crkl antibody in red. 123

P

4.2.3 Assessment of P-crkl expression in leukaemic cell lines Following the establishment of the optimal staining conditions and the appropriate controls for the P-crkl flow cytometry assay in K562 cells the P-crkl expression was investigated in three other leukaemic cell lines: the BCR-ABL+ NALM-1, the ABL activated SIL-ALL, and the BCR-ABL negative JURKAT cell lines. The NALM-1 cells are BCR-ABL+ cells which were established from the blastic phase of a CML patient. Phenotypically the NALM-1 cells express the lymphoid markers CD19, CD20 and the plasma cell marker CD138 (Wetzel et al. 2005). This CML cell line may offer an additional in vitro model for the study of BCR-ABL and Notch cross-talk if both BCR-ABL and Notch activity were confirmed. Therefore, BCR-ABL activity was assessed in NALM-1 cells by the P-crkl assay. However, crkl phosphorylation could not be detected in NALM-1 cells when these cells were intracellularly stained with P-crkl antibody and the BD PE conjugated secondary antibody (Fig. 4.4). The P-crkl assay was repeated three times with the same finding. The SIL-ALL cells (also known as ALL-SIL) were established from a T cell Acute Lymphoblastic Leukaemia (T-ALL) patient. The SIL-ALL cells express the novel NUP214-ABL1 fusion gene with hyper ABL kinase activity. In addition, these cells have Notch1 activating mutations which result in constitutive active Notch signalling (Quinta´s-Cardama et al. 2008). The reported activation of both ABL and Notch signalling pathways in SIL-ALL cell line makes it a possible in vitro model for the study of the cross-talk of ABL and Notch. Therefore, the P-crkl expression was assessed in the SIL-ALL cells. SIL-ALL cells were intracellularly stained with P-crkl antibody as decribed above before being analysed by flow cytometry. Results showed the expression of P-crkl in SIL-ALL cells in three separate experiments (Fig.4.4). This expression was lower than the P-crkl expression in K562 cells. 124

Finally it was important to confirm the specificity of the P-crkl assay in a BCR-ABL negative cell line. The Jurkat cells were assessed for P-crkl expression by the P-crkl assay and results showed the absence of phosphorylated crkl in theses cells (Fig. 4.4) This finding further confirms the specificity of P-crkl assay in detecting crkl phosphorylation in BCR-ABL+ or ABL+ cells.

A

K562

Fig 4.4. Assessment of P-crkl expression in four leukaemic cell lines. The expression of P-crkl protein as assessed by flow cytometry is shown for the BCR-ABL positive cell line K562 (A), nalm-1 (B), the ABL positive cell line ALL-SIL (C) and the BCR-ABL negative cell line Jurkat (D). In all experiment the BD PE conjugated secondary antibody was used with an isotype control (in green) and P-crkl antibody (in red). The filled histogram in dark blue represent the background signal of unstained cells.

C

SIL-ALL

125

4.2.4 Assessment of imatinib mesylate efficacy in K562 cells using the P-crkl assay Imatinib mesylate (IM) has been developed as a potent inhibitor of the ABL protein tyrosine kinases (Holtz et al. 2002). In the past, the inhibitory effect of IM on BCRABL kinase activity has been demonstrated as a reduction of P-crkl expression detected by western blotting (Chu et al. 2004). Expression of P-crkl following IM treatment has also been investigated recently by flow cytometry in K562 cells (Hamilton et al. 2006). In order to validate IM as a BCR-ABL inhibitor in the context of ABL and Notch cross-talk the effect of IM was investigated in K562 cells by the FACS based P-crkl assay using the same P-crkl assay parameters which have been established earlier in this chapter. K562 cells were cultured in decreasing concentrations of IM (10, 5, 1, 0.5, and 0.1 µM) for 48 h before being assayed for the expression of P-Crkl by FACS as described above. Treated cells showed dose dependent inhibition of P-crkl expression as compared to untreated cells (Fig 4.5-a). The reduction of crkl phosphorylation post IM can be seen as a shift to the left of the fluorescence histogram to the fluorescence channel of the isotype control. The reduction of P-crkl post IM exposure was also shown as a reduction of the mean fluorescence intensity (MFI). The MFI was calculated for treated cells and for the no drug control cells as relative MFI to the isotype control in each condition in order to accurately determine the specific fluorescence signal in the reaction. The MFI of IM treated cells was then compared with the MFI of the no drug control cells (Fig. 4.5-b). This result was confirmed in another two experiments with similar outcomes. To confirm the specificity of the P-crkl antibody in the FACS-based P-crkl assay, the effect of IM on P-crkL was also assessed by Western blot on K562 cells. K562 cells were cultured in the presence or absence of IM for 48 h. Western blot was performed as described in materials and methods with the same specific anti-P-crkl primary antibody (1:1000) (New England Biolabs) and an anti-rabbit IgG, horseradish

126

peroxidase-linked secondary antibody (1:2000) (New England Biolabs). The membranes were then stripped in 1 X Stripping Solution (Thermo scientific) for 15 min at RT and re-probed with anti actin antibody (1:1000) (New England Biolabs) in 5% BSA/phosphate buffered saline with BSA, to confirm equal sample loading. Results were similar to those obtained with the FACS based P-crkl assay and a dose dependent reduction of P-crkl expression was observed post 48h treatment with IM (Fig. 4.6).

A-1 1 0µM

A-3

1 µM Fig. 4.5-a. Assessment of imatinib mesylate (IM) efficacy in K562 cells using a flow based P-crkl assay. K562 cells cultured in decreasing concentrations of IM (10, 5, 1, 0.5, and 0.1µM) for 48 h were assayed for the expression of P-Crkl by FACS (A1-5) P-crkl expression in cells treated with imatinib mesylate is shown in green and in untreated cells in red. The histograms in purple represent the isotype control in each experiment. Data shown is from one experiment representative of three separate experiments (n=3). 127

ean flourescence intensity

110 100 90 80 70

Fig. 4.5-b. Dose dependant effect of imatinib mesylate (IM) on the expression of Pcrkl in k562 cells post 48h. P-crkl expression 60 of (IM) treated and untreated K562 cells represented as mean flourescence intensity (MFI). MFI presented here was measured by subtracting the MFI of treated or untreated cells from the MFI of the isotype control in each condition. Dose of IM is plotted in50 X axis and MFI of IM treated cells relative to MFI of no drug control cells in Y axis. Data shown is from one experiment representative of three separate experiments (n=3). 40

30 20 128

10

Fig. 4.6. Effect of concentration of imatinib mesylates (IM) on P-crkl protein levels. K562 and Jurkat cells cultured in decreasing concentrations of IM (10, 5, 1, 0.5, and 0.1µM) for 48 h were harvested for analysis of P-Crkl protein levels by western blot (upper panel). The blot was reprobed with an anti-pan-actin antibody to compare sample loading (lower panel).

P-crkl

Actin 129

1 µM IM

5 µM IM

10 µM IM

K562

4.2.5 Characterisation of Notch signalling in K562 cells The well preserved BCR-ABL kinase activity in K562 cells offers an opportunity to investigate possible interaction between BCR-ABL and other signalling pathways. This cell line may present a good in vitro model system to study the possible crosstalk between BCR-ABL and Notch if the Notch signalling components were proven to be intact. Therefore, Notch signalling was investigated in K562 cells at the mRNA and protein levels in order to evaluate the activity of Notch in these blast phase CML cells. The semi-quantitative RT-PCR was performed in cDNA prepared from K562 cells and this showed clearly the expression of Notch1 (Fig. 4.7). However, the Notch target gene Hes1 could not be detected even after increasing the number of PCR amplification cycles to 32 cycles. CEM cells were used as a positive control since they represent a cell line with markedly enhanced Notch-1 levels (Weng et al. 2004). Next, Notch1 receptor protein expression was investigated by flow cytometry. FACS analysis showed that the extra cellular domain of Notch1 (ECN1) was partially expressed as detected by the EA1 antibody which specifically detects the ECN1 (Fig. 4.8). To investigate the intracellular domain of Notch1 (ICN1) K562 cells were fixed and permeabilised before being stained with the b-tan20 antibody. FACS ananlysis showed that the ICN1 was highly expressed in K562 cells (Fig. 4.8).

130

Fig. 4.7 Expression of Notch1 and Hes1 genes in K562 cell line. cDNA was prepared from K562 cells, and from a cell line known to have active Notch signalling (CEM). Transcript levels were measured by RT-PCR. RTPCR products were resolved by agarose gel electrophoresis and visualised by vistra green. Duplicate RT-PCR data is shown for the expression of Notch1 and Hes1 in K562 cells (upper two panels). The lower panel shows BCR-ABL expression in K562 cells. Data shown is from one experiment representative of three independent experiments (n=3). The number of PCR amplification cycles was 32 for all three genes.

Notch1

< 1% ± 0.05

Hes1

131

BCR-ABL

Fig. 4.8. FACS analysis of Notch1 expression in K562 cells. K562 cells were stained with EA1 antibody which detectes the extracellular domain of Notch1 (ECN1) and the protein expression was analysed by FACS (upper right panel). K562 cells subjected to fixation and permeabilisation, then stained with b-tan 20 antibody which recognise the intracellular domain of Notch1 (ICN1) are shown in the lower right panel. Appropriate isotype controls were used in each staining (upper and lower left panels). Data shown is from one experiment representative of three separate experiments (n=3). The mean percentage of cells positive for each staining with the associated standard error of the mean is shown in each panel

4.2.6 Constitutive expression of Notch1 ΔE in K562 cells Since Notch activation in K562 cells was not evident as assessed by the expression of Hes1 by conventional PCR, a gain of function approach was needed to further investigate the cross-talk of Notch and BCR-ABL in K562 cells. Therefore, it was decided to establish a K562 cells that are stably transfected with Notch1ΔE plasmid. This gain of function approach was used before to constitutively activate Notch

132

signalling in cell lines (Chadwick et al. 2008). Unlike the intra cellular Notch1 domain (ICN1), Notch1ΔE construct have a membrane tether region upstream of the start of the ICN region and is constitutively activated by gamma secretase, which enables the use of GSI to inhibit Notch activity. The inhibitable Notch1ΔE used here was previously cloned into transfection vectors and tested by Dr Nicholas Chadwick (Faculty of Life Sciences, University of Manchester). In order to establish a stable source of K562 cells with hyperactive Notch activity, K562 cells were retroviral transfected with either the Notch1ΔE or the empty vector pMX and maintained in culture for weeks in order to use these cells in future Notch and BCR-ABL cross-talk studies. However, the number of K562 cells transfected with Notch1ΔE showed a steady decrease in culture when monitored every 48-72h by GFP expression by flow cytometry. To see whether the constitutive expression of Notch affected the survival of K562 cells, GSI was added at 10 µM to both the K562 cells that were transfected with Notch1ΔE and to the K562 cells that were transduced with the empty pMX vector. Cell survival in the culture was then monitored for both conditions every week by counting the live cells using GFP expression and FACS analysis. Results demonstrated that the GSI rescue the Notch1ΔE transfected cells from being lost in culture as compared to K562 cells transfected with the empty vector (Fig. 4.9).

PMX

68 %

133

Fig. 4.9. Constitutive expression of N1ΔE in K562 cells. K562 cells were transfected with either N1ΔE or the Pmx empty vector and kept in culture for three weeks in the absence (left panel) or presence (right panel) of gamma secretase inhibitor (GSI). Cell survival in the culture was monitored every week by GFP expression and shown as the percentage of gated live cells in each condition.

4.2.7 The effect of Valproic acid on BCR-ABL and Notch signalling in K562 cells The last approach failed to produce cells with hyperactive Notch signalling that can be maintained in culture for long periods or can be frozen for future experiments. Therefore, the search continued for another approach to activate Notch in K562 cells. Until now no small-molecule activators of Notch-1 signaling in haemopoietic cells have been described. However, it has been shown recently that Valproic acid (VPA) 134

treatment of human gastrointestinal and pulmonary carcinoid tumor cell lines resulted in Notch-1 signaling activation which was associated with increase in the expression of both full-length Notch-1 and the active Notch-1 intracellular domain (NICD) (Greenblatt et al. 2008). In addition, VPA treatment activates Notch1 signaling in Small cell lung cancer (SCLC) cells and inhibits proliferation in SCLC cells (Platta et al. 2008). Similar Notch activation effect was described in neuroblastoma cell lines in which VPA treatment led to activation of Notch signalling as shown by increased levels of intracellular Notch-1 and Hes-1 protein expression (Stockhausen et al. 2005). In all of the previous tumors Notch activity was observed only at baseline levels or as transient up-regulation of Hes1 and treatment with VPA, which is a well-established histone deacetylase (HDAC) inhibitor, resulted in activation of Notch signalling. Since Notch signalling was shown to be reduced in the blastic phase of CML as compared to the chronic phase of the disease (Sengupta et al. 2007) and that K562 cells were established from the blastic phase of CML It was hypothesized that VPA may activate Notch signalling in K562 cells. To investigate whether treatment of K562 cells with VPA can activate Notch signalling or not, K562 cells were cultured in the presence or absence of 4 mM VPA for 72h and the gene expression of Hes1 was measured by real time PCR. Results showed a significant down-regulation of Hes1, an effect similar to that of GSI (Fig. 4.10). This finding was confirmed in three different experiments.

on

1.2

135

Fig. 4.10 Hes1 expression in K562 cells post valproic acid (VPA) tratment. K562 cells were treated with 4mM VPA for 72h and the gene expression of Hes1 was measured by real time PCR. Gene expression was normalised to the GAPDH house keeping gene and represented as DCt values. Comparison of gene expression between treated (red bar) and untreated cells (blue bar) was derived from subtraction of untreated K562 cells DCt values from treated K562 cells DCt values to give a DDCt value, and relative gene expression was calculated as 2-DDCt. The result shown here is from one experiment representative of three different experiments (n=3). Statistical significance was calculated using student t-test. (* = P ≤0.05).

Next the effect of VPA induced inhibition of Notch was examined on the BCR-ABL activity in K562 cells. To investigate this K562 cells were cultured with or without 4 mM VPA for 72h and the P-crkl assay was performed to assess the BCR-ABL activity following VPA treatment. FACS analysis showed an increase in P-crkl expression as compared with P-crkl levels in untreated cells (Fig. 4.11). This result was reproduced in three separate experiments. To find if VPA effect on Notch signalling was associated with any effect on the differentiation of the erythroleukaemic K562 cells the erythroid differentiation was evaluated on K562 cells following exposure to VPA. K562 cells were cultured with or without 4 mM VPA for 72h and the cells were then incubated with FITC conjugated glycophorin A (GPA) as a marker of erythroid differentiation for 30 minutes at RT. Cells where then washed with 3 ml of HBSS (5% FBS) before being analysed by flow cytometry. Appropriate isotype control was included in the experiment to control for non specific binding. Results showed that treatment of K562 cells with VPA for 72h

136

markedly decreased the expression of GPA (Fig 4.12). This effect was confirmed in three independent experiments

137

Fig. 4.11. Effect of Valproic acid (VPA) on BCR-ABL activity in K562 cells. K562 cells were treated with 4mM VPA for 72h and the activity of BCR-ABL was assessed by FACS analysis of P-crkl expression (green). P-crkl expression of untreated cells and background flourescence of isotype control are shown in red and blue respectively. The data shown is from one experiment representative of three independent experiments (n=3).

A

Fig. 4.12. Effect of Valproic acid (VPA) on erythroid diffrentiation in K562 cells. K562 cells were treated with 4mM VPA for 72h and the expression of glycophorin-A (GPA) was analysed by FACS. An appropriate isotype control (first plot in A) was utilised to only include positive cells for GPA among untreated cells (middle plot in A) and treated cells (last plot in A) . Fluorescence intensity of GPA from the same conditions in (A) is 138 represented in histogram format in (B). Data is from one experiment representative of three independent experiments (n=3).

Isotype control

B

Glyco

4.2.8 The effect of GSI in K562 cells As it had proven difficult to set up a model of K562 cells with experimentally activated Notch1 the activation of the pathway was re-examined in unmanipulated cells, this time using a more sensitive real time PCR assay. In these experiments real time PCR was applied to cDNA samples from K562 cells before and after treatment with the Notch inhibitor GSI. Interestingly the results using this approach showed that the expression of Hes1 was detectable in unmanipulated K562 cells and furthermore this expression could be down regulated by exposure to GSI (Fig 4.13). This effect was reproducible in three separate experiments. This data suggests that Notch signalling is active in K562 cells and that these cells may therefore be a suitable model for investigating cross talk with BCR-ABL.

139

ve gene expression

1.2 1

Fig. 4.13. Inhibition of Notch signalling by a gamma seretase inhibitor (GSI) in K562 cells. Real time PCR of the Notch target gene Hes1 is shown. cDNA was prepared from cells treated with vehicle control (DMSO) or 10 µM GSI for 24h. 0.8 Hes1 expression on K562 cells is shown in real time in the upper plot. Hes1 expression was normalised to the GAPDH house keeping gene and represented as DCt values. Comparison of gene expression between treated (red bar) and untreated 0.6 from subtraction of untreated K562 cells DCt values from cells (blue bar) was derived treated K562 cells DCt values to give a DDCt value, and relative gene expression (y axis) was calculated as 2-DDCt (Lower 140plot). Data shown here is from one experiment representative of three separate experiments (n=3). Statistical significance 0.4 was calculated using student t-test. (** = P ≤0.01).

4.2.9 Cross-talk between Notch and BCR-ABL in K562 cells Results from this chapter suggest that the BCR-ABL+ K562 cells may offer an in vitro model system to investigate the cross-talk between BCR-ABL and Notch. Beside the constitutive activity of BCR-ABL in K562 cells which can be inhibited by imatinib the K562 cells express the Notch target genes Hes1 at levels that can be inhibited by the Notch inhibitor GSI. Therefore the possible interaction between BCR-ABL and Notch in CML can be investigated by using inhibitors of either pathway before looking at changes in downstream target gene or protein expression. In order to avoid possible loss of BCR-ABL activity in culture only K562 cells with less than 12 passages were used in all K562 experiments.

4.2.9.1 The effect of imatinib induced BCR-ABL inhibition on Notch signalling in K562 cells K562 cells were cultured in the presence or absence of 10 µM imatinib for 48h. Following the confirmation of P-crkl inhibition in K562 cells by flow cytometry the RNA was extracted and the cDNA was prepared using the High Capacity cDNA Archive Kit (Applied Biosystems). Results showed that Hes1 was upregulated in

K562 cells after 48h treatment with 10 µM imatinib (Fig 4.14). This up-regulation was significant in three different experiments (P≤ 0.01).

4.2.9.2 The effect of Notch inhibition by GSI on BCR-ABL in K562 cells Gamma secretase inhibitor (GSI) has been widely used as a useful tool to study Notch signalling. GSI has been shown to inhibit Notch signalling in normal CD34+ cells and in T-ALL cell lines (Chadwick et al. 2007; Kogoshi et al. 2207). To investigate the effect on BCR-ABL activity following Notch inhibition K562 cells were cultured with DMSO as a vehicle control or with 10 µM GSI for 24h. The dose and time point used here were tested before in the lab and shown to induce Notch inhibition in leukaemic cell lines (Dr. N. Chadwick, personal communication). The cells where then harvested and the intracellular P-crkl assay was performed as described in 2.2.3.4. An aliquot of the same treated and untreated samples where used for RNA extraction and cDNA preparation. Hes1 expression was assessed by real

141

time PCR to confirm the inhibition of Notch activity by GSI in K562 cells (Fig. 4.13). Real time PCR results showed down-regulation of transcriptional target gene Hes1 in the GSI treated K562 cells. The FACS analysis of the same cells showed a dramatic increase in P-crkl expression in the GSI treated cells as compared to no drug control cells (Fig 4.15). These results were reproduced in three separate experiments.

142

1.8

Relative gene expression

1.6 1.4

Fig. 4.14. Expression of Hes1 in K562 cells post 48h treatment of imatinib mesylate (IM). Real time PCR of the Notch target gene Hes1 in K562 cells after 48h treatment with 1.2(IM). Gene expression was normalised to the GAPDH house 10 µM imatinib mesylate keeping gene and represented as DCt values. Comparison of gene expression between treated (red bar) and untreated 1 cells (blue bar) was derived from subtraction of untreated K562 cells DCt values from treated SIL cells DCt values to give a DDCt value, and relative gene expression was calculated as 2-DDCt. The result shown here is from one experiment representative of three different experiments (n=3). Statistical significance was calculated 0.8 using student t-test. (** = P≤ 0.01).

0.6 0.4 0.2 0

143

•Isotype control •Untreated

Fig. 4.15. The effect of Notch inhibition on BCR-ABL activity in K562 cells. K562 cells were cultured for 24h in the presence or absence of gamma secretase inhibitor (GSI) and BCR-ABL activity was assessed by P-crkl assay. P-crkl expression for cells treated with 10 µM GSI for 24h is shown in red. P-crkl expression of untreated cells and background flourescence of isotype control are shown in green and blue respectively. The data shown is from one experiment representative of three independent experiments (n=3).

•+ GSI

P- crkl PE

144

4.2.10 ALL-SIL cell line as a model for ABL-Notch cross-talk The reported activity of ABL and Notch signalling in ALL-SIL cell line makes it a further possible in vitro model for the study of the cross-talk of ABL and Notch. The FACS based P-ckrl assay showed the expression of P-crkl as a marker for the ABL kinase activity in ALL-SIL cells. To see if the ABL activity can be switched off by the ABL inhibitor imatinib mesylate (IM) and to ask whether the P-crkl assay can be utilised as an IM sensitivity assay the effect of IM was investigated on ALL-SIL cells by P-crkl assay. ALL-SIL cells were cultured in the presence or absence of 10 µM imatinib mesylate (IM) for 48h. The cells were then stained with P-crkl primary antibody and PE secondary antibody (BD) as described above. K562 cells were used in this experiment as a positive control for the P-crkl assay and for the efficacy of IM. Results showed that ABL kinase is active in ALL-SIL cells and this activity is evident by the phosphorylation of crkl in the absence of IM (Fig 4.16). Treatment of ALL-SIL cells with IM resulted in clear reduction of P-crkl expression to levels equivalent to those of the isotype control. This experiment was repeated three times with similar results. The finding that ABL is active in ALL-SIL cells and that this activity can be switched off by IM made it possible to investigate the effect of ABL inhibition on Notch signalling in ALL-SIL cells. To assess Notch activity in ALL-SIL cells following ABL inhibition with IM the expression of the Notch target gene Hes1 was investigated by real time PCR. ALL-SIL cells were cultured in the presence or absence of 10 µM imatinib for 48h and then one aliquot of the cells from each condition was harvested to perform the FACS based P-crkl assay and the other aliquot was used for RNA extraction. Following the confirmation of P-crkl inhibition in ALL-SIL cells by flow cytometry the RNA from same experiment was reverse transcribed and cDNA was prepared using the High Capacity cDNA Archive Kit (Applied Biosystems). Results showed that Hes1 was upregulated in ALL-SIL cells after 48h treatment with 10 µM imatinib (Fig 4.17). This up-regulation was significant in three different experiments (P≤ 0.05).

145

A

B

Fig. 4.16. Evaluation of the ALL-SIL cell line as a model for ABLNotch cross-talk. FACS analysis of P-crkl levels in the ALL-SIL cell line. ALL-SIL cells stained with P-crkl primary antibody and PE secondary antibody (BD) (in red) and isotype control (in blue) (A). Expression of P-Crkl in ALL-SIL cells after incubating the cells for 48h with 10 µM imatinib mesylate (IM) is shown in green and isotype control in blue (B) . P-crkl expression of IM treated cells is shown in green as compared to untreated cells in red (C). The data shown is from one experiment representative of three independent experiments (n=3).

C 146

Relative gene expression

18 16

Fig. 4.17. Expression of Hes1 in ALL-SIL cells post 48h treatment of imatinib mesylate (IM). Real time PCR of the Notch target gene Hes1 in ALL-SIL cells after 48h treatment with 10 µM imatinib mesylate (IM). Gene expression was normalised to the GAPDH house keeping gene and represented as DCt values. Comparison of gene expression between treated (red bar) and untreated cells (blue bar) was derived from subtraction of untreated SIL cells DCt values from treated SIL cells DCt values to give a DDCt value, and relative gene expression was calculated as 2-DDCt. The result shown here is from one experiment representative of three different experiments (n=3). Statistical significance was calculated using student t-test. (* = P ≤0.05).

14 12 10 8 6 4 2 0

147

4.3 Discussion 4.3.1 The FACS based P-crkl assay as a surrogate assay for ABL kinase activity Data from this chapter showed that the P-crkl assay is a fast and reliable method to assess the ABL kinase activity and response to imatinib (IM) in cell line models. Certain technical aspects of the assay, however, may affect the interpretation of the data and need to be addressed. For example it was evident that the choice of the appropriate isotype control was critical for obtaining the right levels of P-crkl expression and that the use of secondary antibody alone as a substitute of the rabbit IgG isotype control may yield false increase in the P-crkl expression in the tested cells. The well established method of analyzing levels of phosphorylated proteins by flow cytometry is by subtracting the mean fluorescence intensity (MFI) of the phospho antibody–labeled sample from the MFI of the corresponding isotypic control (Desplat et al. 2004). Therefore, using secondary antibody as a negative control would result in a lower MFI as compared to the MFI of the IgG isotype control and this incorrectly would results in higher estimation of the P-crkl content in the P-crkl labeled sample. The P-crkl validation experiments performed here suggest that the BD PE F(ab')2 anti rabbit secondary antibody is superior to other secondary antibodies tested in K562 cells in terms of fluorescence intensity and specificity. This was the only antibody among the others tested here which was recommended by the manufacturer for intracellular flow cytometric staining. The Fc-mediated non-specific binding of this antibody to Fc receptor-bearing cells was reduced by removing the whole IgG and Fc fragments which resulted in more specific binding to the P-crkl primary antibody. The Sigma FITC secondary antibody showed a staining pattern similar to that of the BD PE secondary antibody in terms of specificity. This specificity may be explained by the fact that this antibody is a monoclonal antibody to rabbit IgG which is devoid of binding to other species. Although the BD PE secondary antibody was brighter than the Sigma FITC secondary antibody in terms of fluorescence intensity, they both represent a good choice to use as secondary antibodies in the P-crkl assay. This is particularly important when investigating the P-crkl levels in certain cell subsets

148

where surface staining of other cell surface markers is also required in the P-crkl assay. This data is in agreement with Hamilton et al. (2006) who showed the positive expression of P-crkl in K562 cells when they intracellularly stained K562 cells with the P-crkl primary antibody and the Sigma FITC secondary antibody. However, the Pcrkl levels in K562 cells reported by Hamilton and co-workers were higher than the isotype control used in their study by about two fluorescence channels. We only found one fluorescence channel difference between the isotype control and the P-crkl labeled cells even when the BD PE secondary antibody was used in the assay. Our results are similar to those of the manufacturer of the P-crkl antibody in terms of the relative fluorescence difference between the isotype control and the P-crkl labeled K562 cells. K562 cells proved to be a good positive control that can be used in the FACS based Pcrkl assay. However, careful monitoring of the BCR-ABL kinase activity by the Pcrkl assay over a long time period demonstrated gradual reduction of P-crkl content in K562 cells. The data showed that K562 cells should not be used as a control in the Pcrkl assay if the cells were cultured for more than 20 passages. The passage number effect on the protein expression of cell lines has been reported before. For example, the expression and activity of the multidrug resistance protein (MDR1) in Caco-2 cell line has been demonstrated to be higher in lower passages and then decline at higher passage numbers (Siissalo et al. 2007). In addition, alkaline phosphatase activity was reduced signficantly in high-passage Caco-2 cells compared to low-passage cells (Yu et al. 1997). In another study, it has been shown that low and high passage RAW 264.7 cells can be transfected equally but protein expression is significantly reduced in the high-passage cells (Jacobsen1 and Hughes, 2007). The reduced expression of phosphorylated crkl in high passage K562 cells can be explained by two possible mechanisms. Firstly, it is possible that K562 cells may undergo differentiation after certain passage numbers and this may reduce the kinase activity of BCR-ABL. It has been reported that myeloid differentiation is associated with down-regulation of BCR-ABL tyrosine activity (Oda et al. 1994). The second mechanism which may explain the reduction of phosphorylated crkl may be the presence of elevated levels of tyrosine phosphatases in differentiated K562 cells 149

which may mediate a dephosphorylation reaction. In support of this is the finding that differentiation of K562 cells was associated with the expression of protein tyrosine phosphatase SHP-1 which results in dephosphorylation of a specific set of tyrosyl phosphoproteins down stream of BCR-ABL (Bruecher-Encke et al. 2001).

4.3.2 P-crkl expression in other leukaemic cell lines The data in this chapter also showed that P-crkl is hardly detectable in a second BCRABL positive cells line - NALM-1. This result is in agreement with Wetzel et al. (2005) who showed by flow cytometry that the P-crkl expression is very low in the NALM-1 cell line as compared to K562 cells. It is unknown whether the low expression of P-crkl in the NALM-1 cells was due to a weak kinase activity of BCRABL or low abundance of the crkl protein in the NALM-1 cells. However, this finding may suggest that the NALM-1 cell line is not a suitable model to investigate the cross-talk of Notch and BCR-ABL. The P-crkl assay of the BCR-ABL negative Jurkat cell line did not show positive expression of P-crkl which further confirms the specificity of the P-crkl assay and that the Jurkat cells can be utilised as a negative control in the P-crkl assay. The finding that P-crkl is clearly expressed in the ALL-SIL cells confirms the intact activity of the ABL kinase in the ALL-SIL cells and shows that crkl protein is also a substrate down stream of the NUP214-ABL1 fusion protien. This is the first time that the P-crkl expression is demonstrated in ALL-SIL cell line by the FACS based P-crkl assay. The level of P-crkl expression in ALL-SIL cells was relatively low compared to that demonstrated in K562 cells. This finding is in agreement with the recent finding that the NUP214-ABL1 fusion protein has a lower in vitro tyrosine kinase activity than the kinase activity observed with BCR-ABL (De Keersmaecker et al. 2008b). Since the Notch activity is well documented in ALL-SIL cells (Graux et al. 2004; and Keersmaecker et al. 2008), the finding that the ABL fusion protien exhibits a constitutive tyrosine kinase activity that can be assessed by the P-crkl assay may make the ALL-SIL cell line a possible experimental model to study the cross-talk between Notch and ABL signalling pathways.

150

4.3.3 Inhibition of p-crkl by imantinib mesylate in K562 cells. After establishment of the P-crkl assay as a rapid and sensitive method to validate the BCR-ABL activity in K562 cells, the effect of IM on BCR-ABL activity could be evaluated. The efficacy of imatinib mesylate (IM) as a BCR-ABL inhibitor was confirmed in K562 cells by the FACS based P-crkl assay with doses of 5 µM and 10 µM achieving more than 90% reduction of P-crkl expression. The sensitivity and specificity of this assay has been shown in this chapter to correlate very well with the Western blotting technique. These findings are in agreement with those reported by Hamilton et al. (2006).

4.3.4 Notch signalling in K562 cells In order to establish whether K562 was a suitable model of the activated Notch profile seen in primary CD34+ CML cells, the expression of the Notch-1 receptor and the Notch target, Hes-1 was investigated in K562 cells. The study of Notch signalling in K562 cells showed that the intracellular domain of Notch1 (ICN1) is highly expressed in K562 cells. However, the extra cellular domain of Notch1 (ECN1) could not be detected by FACS analysis using the EA1 monoclonal antibody. Gene expression profiling using the conventional PCR technique showed the presence of Notch1 at the mRNA level in K562 cells. The failure to detect the Notch1 expression on the cell surface of K562 cells is not surprising as the ECN1 could not be detected in our lab in other cell lines including CEM and TFI cell lines which exhibit active Notch signalling (Dr. N.Chadwick, personal communication). It is unlikely that the specificity of the EA1 antibody was responsible for the failure to detect ECN1 in K562 cells because we were able to detect ECN1 in primary CML cells as shown in chapter 3. In addition, ECN1 was detected before in our lab in the HEK293 cells transfected with full-length Notch1 (Dr. V. Portillo, personal communication). It therefore appears that Notch-1 is expressed at very low levels on the surface of the K562 cells, possibly because only low levels are normally expressed in this cell type, or perhaps as a result of the type of mutation seen in T-ALL where the extracellular domain is not stably expressed. Another possible explanation for the inability to detect ECN1 in K562 cells is that Notch1 may be modulated at the cell surface by glycosylation. As described in chapter one glycosylation is a process in which a glycosyltransferase protein modifies the EGF repeats on the ECN1 to modulate

151

various activities of Notch such as Notch-ligand interaction, Notch folding and intracellular trafficking of Notch (Acar et al. 2008). Modification of the EGF repeats by glycosylation regulates the specificity of Notch binding to different ligands and it is tempting therefore to speculate that glycosylation of the extra cellular domain of Notch1 in K562 cells may mask the EA1 binding to its epitope on ECN1. In support of this notion, glycosylation modification of the AT1 receptor (Angiotensin II receptor subtype I) in COS-7 cells caused a dramatic decrease in cell surface expression of AT1 receptor (Lanctot et al. 2005). The activity of Notch in K562 cells was initially assessed by Hes1 expression by conventional RT-PCR. This approach showed no Hes1 expression at the message level which is similar to the result reported by Yin et al. (2008) who also used the conventional RT-PCR method to measure the gene expression of Hes1 in K562 cells. However, Hes1 expression was clearly detected by using the more sensitive real time PCR method. The discrepancy between Hes1 gene expression data obtained by the RT-PCR and that obtained by real time PCR can be attributed to the fact that the latter is far more sensitive than the former due to the use of more sensitive fluorescent dyes in the PCR reaction and the ability to detect the target gene at the exponential stage of amplification. The expression of Hes1 in K562 cells was further confirmed by using a gamma secretase inhibitor which induced down-regulation of Hes1 expression as demonstrated by real time PCR. Taken together, the previous findings show that K562 cell line fulfills the basic criteria as a good candidate model for the study of BCRABL and Notch cross-talk in terms of having intact and inhabitable activity of both signalling pathways. As it had initially proven difficult to detect Hes-1 by conventional PCR, attempts were also made to devise a K562 model with activated Notch signalling. Ectopic expression of constitutively activated Notch-1 ∆ E in K562 cells induced a dramatic decrease in K562 cell numbers via apoptosis and/ or inhibition of proliferation as evidenced by loss of transfected cells from the culture. This effect was confirmed by the ability of GSI to rescue the transfected K562 cells. The data presented here is in agreement with Yin et al. (2008) who showed that over-

152

expression of the constitutively active Notch1 repressed the growth of the K562 cells in vitro. The attempt to activate Notch signalling by VPA in K562 cells resulted in unexpected outcomes. VPA exhibited an inhibitory action on Notch signalling in K562 as demonstrated by the down-regulation of the Notch target gene Hes1 and the VPA induced inhibition of Notch signalling resulted in increased phosphorylation of crkl. It is evident therefore that VPA works as Notch inhibitor in K562 cells and produces similar effect to that of GSI on BCR-ABL activity. This effect of VPA on BCR-ABL in K562 cells lends further support to the antagonistic interaction between the Notch signalling pathway and BCR-ABL in the blastic phase of CML which was seen with GSI treatment of K562 cells. The action of VPA on Notch signalling demonstrated here in K562 cells is in contrast to previous published reports which proposed that VPA is an activator of Notch signalling in various cancer cell lines (Stockhausen et al. 2005; Greenblatt et al. 2008; and Platta et al. 2008). However, it can be argued that none of these studies were performed in haemopoietic cells or leukaemic cell lines and therefore the action of VPA on Notch signalling may be cell context dependant. The data showed that inhibition of Notch in K562 cells by VPA was associated by repression of erythroid differentiation as assessed by the expression of glycophorin-A. This is in agreement with a recent study in which ectopic Notch activation in FDCP-mix cells accelerated differentiation along the erythroid lineage (Henning et al. 2007). This effect of VPA on erythroid differentiation may be mediated by Wnt signalling as VPA has been reported to activate Wnt signalling (Wiltse, 2005). Since Wnt activation has been shown to block erythroid differentiation in mice (Kirstetter et al. 2006), it is possible that VPA may inhibit erythroid differentiation in K562 cells by activating Wnt signalling.

4.3.5 Cross-talk between Notch and BCR-ABL in K562 cells Inhibition of BCR-ABL by IM in K562 cells resulted in significant up-regulation of Notch activity as assessed by Hes1 gene expression. This is the first demonstration of an interaction between Notch and BCR-ABL in CML cells, although the biological consequences of Notch activation in K562 cells following IM treatment remain to be fully investigated. It can be seen from the effect of ectopic expression of Notch in K562 cells attempted in this chapter that Notch activation may inhibit proliferation or

153

induce apoptosis in K562 cells. However, it should be noted that K562 cells, unlike the chronic phase CML primary samples in chapter 3, are from the blastic phase of CML. In agreement with this, Robert-Moreno et al. (2007) have shown that activation of Notch positively regulates apoptosis in the MEL erythroleukemia cells and in primary erythroid cells in mice. Notch mediated apoptosis in IM treated K562 cells may explain the profound sensitivity of K562 cells to IM as assessed by reduction of P-crkl expression. It is possible that activation of Notch signalling following IM in K562 cells may result in increased levels of Notch mediated apoptosis. The exact mechanism by which IM up-regulates Notch activity in K562 cells remain to be elucidated. GSK3β is a serine/threonine kinase and is a component of the Wnt signaling pathway. It has been shown in cell line models that GSK3β positively modulate Notch signalling by protecting the intracellular domain of Notch1 (ICN1) from proteasome degradation (Foltz et al. 2002). It has also been reported that GSK3β is inhibited by the protein serine-threonine kinases Akt which is a down stream substrate of the BCR-ABL oncoprotien in CML (Cantley, 2002). It is possible therefore that IM may activate GSK3β by inhibiting BCR-ABL and that the activated GSK3β may subsequently stabilises the ICN1 and thereby up-regulates the Notch target gene Hes1. The inhibition of Notch signalling in K562 cells by GSI resulted in a marked increase in P-crkl expression. This effect implies that Notch signalling may negatively regulate BCR-ABL in the K562 cell line and that down-regulation of Notch may directly or indirectly enhance the activity of BCR-ABL. K562 cells are in the blsatic phase of CML and the association between Notch downregulation and progression to the blastic phase has been proposed recently (Sengupta et al. 2008). In addition, it is well documented that BCR-ABL expression and crkl phosphorylation are higher in progenitor cells of CML patients in blast crisis than those of chronic phase patients (Barnes et al. 2005). It can be seen therefore that the effect of GSI induced Notch inhibition on BCR-ABL activity in K562 cells may mirror the status of Notch and BCR-ABL signalling pathways in the blastic phase primary CML cells. The precise mechanism by which Notch modulate BCR-ABL in the blastic phase remain to be identified. One model that can be postulated for Notch and BCR-ABL 154

crosstalk in the blastic phase of CML is that the blastic phase CML cells remain dependant on BCR-ABL activity for their proliferation and that down-regulation of Notch signalling maintains the oncogenic activity of BCR-ABL. These observations highlight the importance of considering the cell context when investigating these signalling pathways, and that although K562 and other leukaemic cells are useful tools as models for signalling, in order to examine the signalling in the context of chronic phase CML, primary tissue samples need to be used.

4.3.6 Cross-talk between Notch and BCR-ABL in the ALL-SIL cell line model system However it is clear from the work on T-ALL-SIL cells that the observed interactions between the pathways are not confined to blast crisis CML cells. The ALL-SIL cells proved to be a good model system to further validate the cross-talk between Notch and BCR-ABL. It can been seen that this in vitro model system fulfills the basic criteria to investigate the interaction between Notch and BCR-ABL signalling in terms of the well documented active and inhabitable Notch signalling (Graux et al. 2004; and Keersmaecker et al. 2008) and in terms of the presence of a constitutively activated tyrosine kinase which is sensitive to the ABL kinase inhibitor IM. The findings of P-crkl expression and inhibition of the P-crkl levels by IM in the ALL-SIL cells as assessed here by the FACS based P-crkl assay are in agreement with the data reported by Graux et al. (2004) and those by Quinta´s-Cardama et al. (2008) who used western blotting to demonstrate P-crkl expression and the effect of IM on crkl phoshphorylation in the ALL-SIL cells. The previous results indicate that NUP214ABL1 is a constitutively activated tyrosine kinase that may activate similar pathways as BCR-ABL. The inhibition of the NUP214-ABL1 kinase activity by IM resulted in significant upregulation of the Notch target gene Hes1. This effect recapitulated the IM induced effect on K562 cells in which the inhibition of ABL kinase activity led to activation of Notch signalling. Therefore the finding that the aberrant ABL activity may antagonize Notch signalling in leukaemic cells as found in K562 cells can be extended to a second cell line, ALL-SIL cells. The precise mechanism by which IM induces activation of Notch in K562 and ALL-SIL cells remain to be investigated. However, it

155

is possible that this may be mediated by the GSK3β kinase which has been proposed in the previous section.

156

Chapter 5: Cross-talk between Notch and BCRABL in primary CD34+ CML cells 5.1 Introduction The rare leukaemic stem cells (LSCs) in chronic phase CML remain a challenge to the currently available BCR-ABL targeted treatment options (Elrick et al. 2005). BCRABL + CD34+ CD38- cells which are enriched with LSCs do not respond to imatinib mesylate (IM) in vitro (Graham et al. 2002). In vivo, LSCs are most likely to be responsible for the minimal residual disease seen in most patients who were treated with IM and achieved complete cytogenic response (Jorgensen and Holyoake, 2007). Current knowledge suggests that resistance to IM in chronic phase CML patients may be due to either ABL tyrosine kinase mutations and/ or persistence of disease due to BCR-ABL independent mechanisms (Deininger and Holyoake, 2005). In line with this notion, LSCs in CML have been shown to be resistant to Dasatinib, a novel SRC/ BCR-ABL inhibitor, which is reported to inhibit the majority of kinase mutations in IM-resistant CML (Copland et al. 2006). These findings raise the possibility that LSCs in CML, beside their dependence on BCR-ABL, may depend on other survival pathways that were underestimated by current CML molecular targeted therapy. Since the LSCs in CML have the same phenotypic and functional characteristics of normal haemopoeitic stem cells (HSCs), it is very likely that they also share the same self-renewal and survival pathways such as the Notch, Wnt, and Hedgehog signalling pathways. Recently, it has been reported that Wnt and Hedgehog signalling pathways are essential for the survival of LSCs in CML. Zhao et al (2007) showed in a conditional β-Catenin knockout mice that β-catenin deletion causes a profound reduction in the ability of mice to develop BCR-ABL induced CML which demonstrates that Wnt signalling is required for self-renewal of LSCs in CML. More recently, Hu et al (2008) showed in a CML mouse model the existence of a LSC survival pathway that was not inhibited by imatinib even though IM inhibited BCR-ABL phosphorylation and provided evidence of the essential role of Wnt signalling for survival and self-renewal of CML LSCs. However, another

157

report challenges the view that Wnt signalling is a BCR-ABL independent survival pathway for the LSCs in CML. It has been shown that BCR-ABL physically interacts with β-catenin and that BCR-ABL levels control the degree of β-catenin stabilisation (Coluccia et al. 2007). In the same study the authors confirmed in primary cells and in a cell line model that imatinib mesylate (IM) inhibited β-catenin expression in blastic phase of CML. Interestingly, Dierks et al (2008) showed, in human and mice, that Hedgehog signalling is activated in LSCs in CML through up-regulation of the Hedgeohg's transmembrane receptor Smo and that Smo is essential for the expansion of the LSC pool in mice. To determine if Hedgehog signalling was dependent on BCR-ABL kinase activity, the authors used imatinib to inhibit ABL in murine BCR-ABL+ LSCs and found that the transcripts levels of the Hedgohog target gene Gli1 was decreased by only 20%. Data from chapter three shows clearly that Notch signalling is hyperactive in the most primitive LSCs as well as in CD34+ progenitor cells in the chronic phase of CML patients. Knowing that Notch signalling is essential for the survival and self-renewal of normal HSCs and possibly LSCs in CML, it was intriguing therefore to ask if there is cross-talk between Notch and BCR-ABL signalling pathways. This is particularly important to elucidate possible active survival pathways in CML LSCs that are BCRABL independent and are not targeted yet by ABL kinase inhibitors alone. The results shown in chapter four indicates that there may be cross-talk between ABL and Notch signalling in ABL+ leukaemic cell lines. This chapter aims first to apply the p-crkl FACS based assay to monitor the BCRABL activity in primary CML cells. It also aims to investigate the nature of Notch and BCR-ABL interaction in CD34+ chronic phase CML cells by two approaches. Firstly, by inhibition of BCR-ABL kinase activity by the kinase inhibitor imatinib and determining the effect of that on Notch signalling by real time PCR. Secondly, by inhibiting Notch signalling by a gamma secretase inhibitor (GSI) and looking at the BCR-ABL kinase activity by the p-crkl assay.

158

5.2: Results 5.2.1 Crkl phosphorylation can be detected in primary CD34+ CML cells by intracellular flow cytometry assay The P-crkl FACS based assay has been used as a surrogate marker to monitor the BCR-ABL kinase activity in CML and other malignancies and as a parameter for the efficacy of imatinib and other tyrosine kinase inhibitors (Copland et al. 2006). The Pcrkl assay was validated in K562 cells in chapter four and the staining conditions were assessed before application of the assay to primary CD34+ CML cells. In initial experiments CD34+ CML cells from frozen CML samples were fixed and stained with anti-P-crkl primary antibody and PE secondary antibody as established for K562 cells in chapter four. However, no positive P-crkl staining was seen using this method in primary cells. This finding led to a re-evaluation of the staining method for use with primary cells. Firstly, the experiment was performed on cells thawed and stained on the same day, whereas others have cultured primary in cytokines before assessing P-crkl expression (Dr. Sophia Hatziieremia, University of Glasgow, personal communication). CD34+ cells where therefore cultured overnight in serum free medium supplemented with cytokine cocktail comprising 100 ng/ mL Flt3ligand, 100 ng/ mL stem cell factor, and 20 ng/ mL each of interleukin (IL)–3, IL-6 and granulocyte-colony stimulating factor (G-CSF) for 24 hours before assay. However, under these conditions CD34+ cells still did not show P-crkl expression when stained with anti-P-crkl primary antibody and PE secondary antibody. Therefore modifications of the staining conditions including fixation and permeabilisation procedures, primary antibody concentration, and type of secondary antibody used in the assay were reassessed. It was found that P-crkl could not be detected when the PE secondary antibody (BD) was used but was detected when Sigma FITC secondary antibody was used (Figure. 5.1). This observation was only evident with the primary CD34+ CML cells as parallel experiments on K562 cells showed positive P-crkl expression with both PE as well as FITC anti-rabbit secondary antibodies, which is consistent with experiments results reported in chapter four. Therefore, a protocol using FITC anti rabbit secondary antibody was adopted for the remaining experiments to detect P-crkl in primary CD34+ cells.

159

A K562

PE

Fig. 5.1. Application of P-CrKl assay to primary chronic myeloid leukaemia (CML) samples. Mononuclear cells from frozen aliquots of primary CML cells were cultured for 24h in cytokines cocktail before being fixed and stained with P-crkl primary antibody and either PE (A) or FITC (B) conjugated anti-rabbit secondary antibodies. The P-crkl staining patterns in CML samples are shown in the right hand side plots in A and B and K562 cells which were run as a positive control are shown in the left plots. Cells stained with P-crkl PE are shown in red, whereas unstained cells and isotype control are shown in blue and green respectively (panel A). The P-crkl FITC stained cells are depicted in green and isotype control in red (panel B). CML data shown is from one primary CML sample representative of three separate samples from CML patients.

B

K562

FITC

160

5.2.2 Imatinib mesylate (IM) inhibits BCR-ABL activity in chronic phase CML CD34+ cells Inhibition of BCR-ABL activity was the first approach taken to investigate cross-talk between Notch and BCR-ABL. It has been shown in chapter four that imatinib mesylate (IM) can inhibit BCR-ABL kinase activity in the CML cell line K562. Since this chapter aims to study the cross-talk between Notch and BCR-ABL in primary CD34+ CML cells, the efficacy of IM on primary CD34+ CML was tested by using the FACS based P-crkl assay. CML samples from leukapheresis products from patients with chronic phase CML (n=5) were highly enriched for CD34+ and cultured for 24h in serum free medium (SFM) which was supplemented with the five growth factor cocktail described in section 5.2.1. CML cells were then treated with Imatinib mesylayte (10 uM) for 72h and the inhibitory effect of IM on CD34+ CML cells was assessed by the P-crkl flow cytometric assay. At the time of assessment of P-crkl content most of the samples were found to be > 90% CD34+. In two samples in which the CD34+ percentage was 70% the CD34 gating strategy at time of analysis was used to ensure that only CD34+ cells assessed for P-crkl expression. To control for the sensitivity of the P-crkl assay and the efficacy of IM, the assay was performed on untreated and IM treated K562 cells in parallel with the P-crkl assay on primary CML cells. Figure 5.2 and 5.3 show the inhibitory effect of IM on BCR-ABL in CML samples. Expression of P-crkl was clearly reduced on CD34+ from three CML samples as compared to untreated samples. However, there was some expression of Pcrkl in two CML samples after 72h of IM treatment (Figure 5.3) suggesting a level of resistance to IM in these two patients.

5.2.3 Effect of Imatinib in CD34+ CML cells upregulates Hes1 Notch target gene expression Next, the effect of IM induced BCR-ABL inhibition on CD34+ CML cells on Notch signalling was investigated. All CML samples were enriched for CD34+ cells using magnetic CD34 selection. After culture with or without imatinib (as described in section 5.2.2) the percentage of CD34 cells in culture was assessed by FACS and RNA was extracted directly from cultured cells if they were > 90% CD34+ or FACS sorted if they were < 90% CD34+. The Notch target gene Hes1 was used as an

161

indicator of Notch activation and Hes1 transcript levels were investigated by real time PCR in IM treated CD34+ CML cells. Figure 5.4 shows Hes1 gene expression following 72h IM treatment of CD34+ cells isolated from imatinib sensitive CML patients (CML1, 3, and 6). There was a 4 fold increase (n=3 ± 1.1) in Hes1 gene expression following treatment of CD34+ CML cells with IM. This increase was statistically significant in all three individual CML samples that showed up-regulation of Hes1. To ask if Hes1 up-regulation was only found in IM sensitive CD34+ CML cells the expression of Hes1 was also investigated in CD34+ from the two CML samples that showed resistance to IM (CML2 and 4). In both CML samples Hes1 expression in IM treated CD34+ cells was minimally reduced or similar to untreated cells (Figure 5.5) with no significant difference in gene expression observed in either samples. It can be concluded therefore that only CD34+ CML cells that were IM sensitive showed activation of Notch, as assessed by induction of Hes1 gene expression, when BCRABL is inhibited.

162

A

C M L

-u n t r e a t e d

B

+ L M C I M

3

6

1

P- c r k l

( F I T C )

Fig. 5.2. Inhibition of BCR-ABL activity by imatinib mesylate (IM) in CD34+ cells isolated from CML patients. Primary CD34+ cells were isolated from three CML patients (CML3,CML6, and CML1) and cultured overnight with growth factors alone before being kept in the absence (A) or presence (B) of 10 µM IM for 72h. CD34+ cells were then harvested and the P-crkl assay was performed by FACS to assess the activity of BCR-ABL in treated and untreated cells. In each case at least 70% of cells analysed for P-crkl expression were CD34+. Where possible, P-crkl staining of untreated K562 cells (C) and IM treated K562 cells (D was performed at the same time as positive controls for the P-crkl assay and imatinib mesylate efficacy. The P-crkl FITC stained cells are shown in green and isotype control in red in all plots.

163

A

C

-Mu n L t r e Ba t e d C

M

4 Fig 5.3. Evidence of resistance to imatinib mesylate (IM) in CD34+ from two CML patients. CD34+ cells from two CML patients (CML4 and CML2) were isolated and cultured for 24h before being treated with 10 µM IM for 72h. CD34+ cells were then harvested and the P-crkl assay was performed by FACS on untreated (A) and IM treated cells (B) to assess the response to IM. In each case at least 70% of cells analysed for P-crkl expression were CD34+. In one case P-crkl staining of untreated K562 cells (C) and IM treated K562 cells (D) was performed at the same time as positive controls for the P-crkl assay and imatinib mesylate efficacy. The P-crkl FITC stained cells are shown in green and isotype control in red in all plots.

2

164

L

Relative gene expression

CML 3.5 3 2.5 2 1.5 1 0.5 0

Relative gene expression

CML 8 7 6

Fig. 5.4. Hes1 gene expression post imatinib mesylate (IM) treatment in 5 CD34+ cells isolated from imatinib sensitive CML pateints. CD34+ cells isolated from the same CML patients shown in fig. 5.2 were cultured in the 4 or absence (blue bar) of 10 µM IM for 72h. Live CD34+ presence (light bar) cells were then sorted and the gene expression profile of the Notch target 3 gene Hes1 was investigated by real time PCR. Relative gene expression was calculated using the DDCt method. Statistical significance was calculated using student t-test. 2 (* = P ≤0.05, ** = P ≤0.01, *** = P ≤0.001). 1 0 165

Relative gene expression

CML 1.4 1.2 1 0.8 0.6 0.4 0.2

expression

Fig. 5.5. Hes1 gene 0 expression post imatinib mesylate (IM) treatment in CD34+ cells isolated from IM resistant CML pateints. CD34+ cells isolated from the same CML patients shown in fig. 5.3 were cultured in the presence (light bar) or absence (blue bar) of 10 µM IM for 72h. Live CD34+ cells were then sorted and the gene expression profile of the Notch target gene Hes1 was investigated by real time PCR. Relative gene expression was calculated using the DDCt method. Student t-test in both CML samples showed no significant difference in Hes1 expression between treated and untreated cells.

1.2 1 0.8

166

CML4

5.2.4 Investigating the effect of Notch inhibition on BCR-ABL activity in CD34+ CML cells Results from chapter three showed that Notch signalling is hyperactive in CD34+ cells including the most primitive stem cell enriched CD34+ Thy-1+ cell subset in chronic phase CML patients. The results from chapter 4 showed that GSI induced Notch inhibition led to increase of ABL kinase activity in K562 and ALL-SIL cell lines. To further investigate the nature of cross-talk between Notch and BCR-ABL in CML, the gamma secretase inhibitor GSI-IX was used to induce Notch inhibition on CD34+ CML cells before assessing the effect of Notch inhibition on BCR-ABL activity on those cells. 5.2.4.1 GSI induced inhibition of Notch signalling in CD34+ CML cells Gamma secretase inhibitors (GSI) have been widely used as a useful tool to study Notch signalling. It was therefore important to determine whether GSI could induce inhibition of Notch signalling in CD34+ CML cells before looking at BCR-ABL activity status following GSI treatment. Therefore, Hes1 expression post GSI treatment was investigated as an assay for the efficacy of GSI on Notch activity in CD34+ CML cells. CD34+ CML cells from five patients were cultured as described above with a five growth factor cocktail (as described in section 5.2.1) for 72h in the presence or absence of 10 µM GSI before carrying out real time PCR for the Notch target gene Hes1. Figure 5.6 shows that CD34+ cells in three CML samples (CML2, 4, and 5) responded very well to the inhibitory action of GSI as evident by down-regulation of Hes1. This down-regulation was significant in two samples and not significant in one sample (P= 0.09). CD34+ cells from two CML samples (CML1 and 6) showed no evidence of a response to GSI treatment as assessed by a decrease in the expression of Hes1 mRNA (figure 5.7). To confirm these findings the real time PCR experiments were repeated with more concentrated cDNA samples but this again showed no significant difference of Hes1 gene expression between untreated and GSI treated CD34+ CML cells.

167

gene expression

Fig. 5.6. Hes1 gene expression after gamma secretase inhibitor (GSI) treatment in CD34+ cells isolated from CML patients 2, 4, and 5. CD34+ cells were isolated from CML patients and cultured in the presence (red bar) or absence (blue bar) of 10 µM GSI for 72h. Live CD34+ cells were then sorted and the gene expression of the Notch target gene Hes1 was investigated by real time PCR. Relative gene expression was calculated using the DDCt method. Data shown is from three CML samples (n=3). Statistical significance was calculated using student t-test. (* = P ≤0.05).

168

1

1

0

0

169

n

Fig. 5.7. Hes1 gene expression after gamma secretase inhibitor (GSI) treatment in CD34+ cells isolated from pateint 1 and 6. CD34+ cells were isolated from CML patients and cultured in the presence (red bar) or absence (blue bar) of 10 µM GSI for 72h. Live CD34+ cells were then sorted and the gene expression of the Notch target gene Hes1 was investigated by real time PCR. Relative gene expression was calculated using the DDCt method. Student t-test in both CML samples showed no significant difference in Hes1 expression between treated and untreated cells.

5.2.4.2 Non GSI responding CD34+ CML cells express high mRNA levels of Hes1 The finding that GSI treatment failed to downregulate Hes1 in CD34+ cells from two CML samples was interesting given the finding in chapter three of high levels of Hes1 expression in CD34+ cells from CML patients as compared with NBM. It was not possible to perform GSI inhibition experiments on the CML samples which were used to demonstrate upregulation of Hes1 in the CD34+ populations (chapter three). The CML cells tested in section 5.2.4.1 were from frozen material from CML patients whose Notch signalling activity was not confirmed. It was possible that the observed failure to respond to GSI treatment in the two CML samples described here may be due to low Hes1 transcript levels to start with. Therefore, Hes1 gene expression was measured in the CD34+ cells in all CML samples used in this chapter by real time PCR and compared to CD34+ cell from normal bone marrow. It was found that Hes1 expression was upregulated in CD34+ cells in all six CML samples as compared to normal CD34+ cells from normal bone marrow. This upregulation was statistically significant (P ≤0.01) and confirmed the findings in chapter three. This data exclude the possibility that failure of CD34+ cells to respond to GSI in samples 2 and 4 was due to low levels of Hes1 mRNA in the starting material. Taken together these observations also suggest that the high levels of Hes1 in some of these samples are the result of gamma secretase independent Notch signalling. 5.2.4.3 Gamma secretase inhibitor (GSI) increases the kinase activity of BCRABL in CD34+ CML cells To further explore the cross-talk between Notch and BCR-ABL, the effect of the Notch inhibitor GSI on BCR-ABL activity was investigated in CD34+ CML cells. CD34+ CML cells from five patients were cultured as described before with a five growth factors cocktail (as described in section 5.1) for 72h in the presence or absence of 10 µM GSI before measuring BCR-ABL activity by the FACS based P-crkl assay. At the time of P-crkl assay the percentage of CD34+ cells in the culture was measured and counted to be between 70-90% in all samples. CD34+ gating strategy was applied at the time of the P-crkl assay to samples which were found to have < 90% CD34+ cells in order to have at least 90% CD34+ cells in all CML samples analysed for their

170

P-crkl expression. The BCR-ABL positive K562 cells were used as a positive control for the P-crkl assay in each case. To calculate the change in P-crkl expression the mean fluorescence intensity (MFI) of GSI treated or untreated CD34+ cells was first determined by subtracting the MFI of P-crkl stained cells from the MFI of isotype control in each condition. The MFI of GSI treated cells was then compared to the MFI of untreated cells and the change in P-crkl expression was reported as percentage. Interestingly, FACS data showed that GSI treatment increased the P-crkl expression in CD34+ cells between 18-42 % as compared to total P-crkl in untreated CD34+ cells. Figure 5.9 shows increase in the P-crkl expression in CD34+ CML cells from GSI responsive CML samples (CML2, 4, and 5) that showed downregulation of Hes1 mRNA post GSI treatment. This data suggests that the increase in P-crkl expression on these samples is most likely Notch dependant. However, it appears that the other two CML samples which did not show Notch inhibition by PCR post GSI treatment (CML 1 and 6) also exhibited an 18- 40% increase in crkl phosphorylation (figure 5.10). The increase in P-crkl in CD34+ CML cells (n=5) was statistically significant as shown in figure 5.11 (P< 0.01). 5.2.4.4 Gamma secretase inhibitor (GSI) decreased the kinase activity of BCRABL in CD34+ CML cells from one CML patient In contrast to the results shown earlier for five CML patients, GSI treatment of CD34+ cells from one CML patient (CML3) showed inhibition of BCR-ABL activity as can be seen from the reduction of P-crkl expression by 30% as compared to nodrug control (figure 5.12). CD34+ cells from this sample responded very well to the GSI induced inhibition and showed downregulation of Hes1 post 72h GSI treatment. Therefore, the effect of GSI treatment on BCR-ABL activity on CD34+ cells from this sample is most likely a Notch mediated effect. Interestingly, CD34+ cells from the same patient were sensitive to imatinib treatment since after 72h incubation with 10 µM IM the P-crkl expression was markedly reduced as compared to untreated cells (figure 5.12). Imatinib treatment caused significant upregulation of Hes1 in CD34+ cells (P< 0.001). This is the only CML sample which showed BCR-ABL inhibition following the inhibition of Notch signalling by GSI.

171

The findings of Notch and BCR-ABL cross-talk from all six CML patients tested here

Log fold change in gene expression

are summarised in table 5.12.

1000000

** 100000

** **

10000 1000

Fig. 5.8. Hes1 gene expression in CD34+ CML cells. The gene expression profiles of the Notch target gene Hes1 was investigated by real time PCR. Data is shown from CD34+ cells isolated from six CML patients in chronic phase and CD34+ control cells from three normal bone marrow (NBM) samples. Relative gene expression was calculated using the DDCt method. The log fold change in Hes1 gene expression in each CML sample is plotted against the mean of Hes1 expresion in the three NBM samples. Statistical significance was calculated using the Mann-Whitney test. (** = P ≤0.01).

100 10 1

172

CML1

CML2

CM

A

C

-uM n

2

Fig. 5.9. Assessment of P-crkl in CD34+ CML cells following inhibition of Notch by gamma secretase inhibitor (GSI). Primary CD34+ cells were isolated from three CML patients (CML2, CML4, and CML5) and cultured overnight with growth factors alone before being kept in the absence (A) or presence (B) of 10 µM GSI for 72h. CD34+ cells were then harvested and the P-crkl assay was performed by FACS to assess the activity of BCR-ABL in treated and untreated cells. At least 90% of cells analysed for P-crkl expression were CD34+. In each case P-crkl staining of untreated K562 cells (C) was performed at the same time as a positive control for the P-crkl assay. The increase in Pcrkl expression in GSI treated versus untreated cells is shown as a percentage. The P-crkl FITC stained cells are shown in green and isotype control in red in all plots. The GSI induced inhibition of Notch in all CML samples shown here was confirmed by real time PCR (see Fig.5.6).

4

173

A

CML- untreated

B

CML + GSI

C

Untreated K562

40%

1

18%

6

P-crkl (FITC) Fig. 5.10. Assessment of P-crkl in gamma secretase inhibitor (GSI) non responsive CD34+ CML cells. Primary CD34+ cells were isolated from two CML patients and cultured overnight with growth factors alone before being kept in the absence (A) or presence (B) of 10 µM GSI for 72h. CD34+ cells were then harvested and the P-crkl assay was performed by FACS to assess the activity of BCR-ABL in treated and untreated cells. At least 70% of cells analysed for P-crkl expression were CD34+. In each case P-crkl staining of untreated K562 cells (C) was performed at the same time as a positive control for the P-crkl assay. The increase in P-crkl expression in GSI treated versus untreated cells is shown as a percentage. The P-crkl FITC stained cells are shown in green and isotype control in red in all plots. In the two CML samples shown here Notch activity was not inhibited by GSI as revealed by real time PCR (see fig. 5.7).

174

45

Mean % P -crkl

40

**

35 30 25 20 15 10 5

175

Fig. 5.11. P-crkl in CD34+ CML cells treated with gammas secretase inhibitor (GSI). CD34+ cells from five CML patients in chronic phase were cultured in the absence (blue bar) or presence (red bar) of 10 µM GSI for 72h. The change in BCR-ABL activity was assessed by the FACS based P-crkl assay. P-crkl expression was measured by mean fluoresnece intensity (MFI) units in each condition. MFI of P-crkl in GSI treated CD34+ cells was compared to no-drug control in each sample and the percentage of increase in P-crkl was calculated. Data shown here represent the mean of five CML samples. Statistical significance was calculated using student t-test (** = P ≤0.01).

176

A

Untreated

B

e gene expression

P-crkl (FITC) 1.2 1 0.8 177

0.6 0.4

Figure 5.12. GSI treatment induced both Notch and BCR-ABL inhibition in CD34+ cells from one CML sample. CD34+ cells from CML 3 patient were cultured overnight with cytokines and then treated with GSI or IM for 72h before assessing P-crkl expression (A). The increase in P-crkl expression in GSI treated versus untreated cells is shown as a percentage. The P-crkl FITC stained cells are shown in green and isotype control in red in all plots. Hes1 gene expression was investigated by real time PCR on CD34+ cells treated for 72h with GSI (B) or with IM (C). Relative gene expression was calculated using the DDCt method. Statistical significance was calculated using student t-test. (** = P ≤0.01, *** = P ≤0.001).

Response to IM (P-crkl assay)

Effect of IM on Notch activity

Response to GSI )Hes1 expression(

Effect of GSI on ABL activity

Sensitive

Hes1 up-regulation No response

P-crkl overexpression

Resistant

No effect

P-crkl overexpression

Sensitive

Hes1 up-regulation Hes1 downregulation

P-crkl reduction

Resistant

No effect

Hes1 downregulation

P-crkl overexpression

Sensitive

No effect

Hes1 downregulation

P-crkl overexpression

Sensitive

Hes1 up-regulation No response

Hes1 downregulation

P-crkl overexpression

Table 5.1. Summary of Notch- BCR-ABL cross-talk data following treatment of CD34+ CML cells from six CML patients with GSI and IM.

178

5.3: Discussion Cross-talk between BCR-ABL and Notch has been investigated in cell lines in chapter four. Although most of the interactions between BCR-ABL and other signalling pathways have been described in blastic phase cell lines, this may not represent the behaviour of this oncoprotien in chronic phase disease in vivo (Marley and Gordon, 2005). Therefore, it was important to use patient derived material to investigate the possible cross-talk between BCR-ABL and Notch signalling in chronic phase CML.

5.3.1 BCR-ABL activity can be monitored in primary CD34+ CML cells by flow cytometry The activity of BCR-ABL has been shown to be responsible for initiating and maintaining the leukaemic clone in the chronic phase of CML (Quintás-Cardama and Cortes, 2008). Western blotting and immunoprecipitation have proven technically challenging tools to monitor BCR-ABL interactions with other signalling molecules and/or substrates in primary CML cells (Marley and Gordon, 2005). The development of the FACS based P-crkl assay to monitor BCR-ABL activity provides a more reliable method to monitor BCR-ABL activity and its response to drugs in CML patients. This intracellular flow cytometric assay which detects phophorylated crkl by using an anti-phospho crkl (P-crkl) antibody, only requires small cell numbers from patient samples and, unlike western blotting, can be performed in few hours (Hamilton et al. 2006).

179

The P-crkl assay was validated in K562 cells in chapter four but has to be validated in primary CML samples before it can be used as a marker for BCR-ABL activity in patient samples. Data from this chapter shows that the choice of secondary antibody is a critical step in P-crkl staining of primary CD34+ CML cells. P-crkl expression could not be detected in CD34+ cells from chronic CML patients when the PE (Becton Dickinson) conjugated anti rabbit antibody was used. This is in contrast to the blastic phase CML cell line K562 which showed very bright P-crkl expression with PE conjugated secondary antibody (see chapter four). This discrepancy is most likely due to inherent differences between primary chronic phase CD34+ CML cells and the blastic phase K562 cells. The CD34+ CML cells are relatively small cells (Jørgensen and Holyoake, 2007). It is possible that uptake of PE fluorochrome, a 240-kDa protein, by the CD34+ CML cells is difficult to achieve due to the large molecular weight of PE molecule as compared to FITC fluorochrome which has a molecular weight of only 389 daltons. A thorough literature search showed that only few studies have attempted the intracellular flow cytometric P-crkl assay to monitor BCR-ABL in CML patients. In most of these studies, the FITC conjugated anti rabbit secondary antibody was used against P-crkl primary antibody in CD34+ CML cells (Jiang et al. 2007b; Hamilton et al. 2006; and Copland et al. 2006). In contrast, Jilani et al (2008) have used a PE conjugated secondary antibody (Santa Cruz Biotechnology) to detect crkl phosphorylation in imatinib-treated versus imatinib-naïve CML patient peripheral blood cells. Although the authors showed a positive PE fluorescence signal in the naïve CML cells, they used mononuclear cells from CML patients rather than limiting their study to the more primitive CD34+ cells. In addition to studying cell population other than the CD34+ cells studied here, Jilani and co-workers used a different permeabilisation reagent and a secondary antibody from different supplier which makes comparison between their findings and data reported here more difficult. Nonetheless, within the conditions described here, the use of the PE conjugated secondary antibody (BD) to detect crkl phosphorylation in CD34+ CML cells did not show positive P-crkl signal.

180

Data on P-crkl FACS staining showed higher levels of P-crkl expression on K562 cells as compared to primary CD34+ CML cells when cells stained with P-crkl antibody and the FITC conjugated secondary antibody. It is well documented that levels of crkl phosphorylation correlates well with the levels of BCR-ABL expression (Hoeve et al. 1994). It has been shown that BCR-ABL expression and activity as well as crkl phosphorylation is higher in progenitor cells of patients in blast crisis than in those of chronic phase patients (Barnes et al. 2005). Therefore it is perhaps not surprising to see higher P-crkl expression in the blastic phase CML cell line K562 as compared to CD34+ chronic CML cells.

5.3.2 Imatinib mesylate inhibits BCR-ABL activity and up-regulates Notch activity in CD34+ chronic phase CML cells One of the approaches used in this chapter to study BCR-ABL and Notch cross-talk in chronic phase CML was to investigate the effect of BCR-ABL inhibition on Notch activity. Therefore, the BCR-ABL inhibitor imatinib mesylate (IM) was utilised in this chapter as a tool to inhibit BCR-ABL activity in CD34+ cells. The results showed that BCR-ABL activity was inhibited by IM in CD34+ cells of 4/6 patients. This effect was shown as marked reduction of crkl phosphorylation post 72h of 10 µM IM treatment (n=4). This is in agreement with Chu et al (2004) who showed that imatinib exposure in doses between 1-5 µM resulted in inhibition of crkl phosphorylation in CML CD34+ cells in a dose-dependent manner and as early as after two hours of IM treatment. The authors used western blot analysis to examine the IM effect on crkl phosphorylation, a method which was found to correlate very well with the FACS based P-crkl assay used in this project (see chapter four). Copland and colleagues (2006) showed that the majority of CD34+ CML cells were sensitive to 5 µM IM at 16 hours and showed clear reduction of P-crkl. However, they also reported that the surviving CD34+ cells showed minimal reduction in P-crkl at 72 hours. The authors interpreted this as enrichment of IM resistant population post 72h of IM treatment. Data presented in this chapter showed also two CML samples that

181

did not respond to IM at 72h and showed minimal reduction of crkl phosphorylation (Figure 5.3) which may reflect the time point used. In fact, the 72h time point was chosen in our study in order to have enhanced IM mediated inhibition of BCR-ABL activity. Holtz et al (2002) reported that IM induced more significant suppression of CML CFCs after a 96-hour exposure as compared with 24-hour exposure.

Resistance of CML stem cells to imatinib is likely to be multifactorial and the underlying mechanisms may depend on stage of disease, genetic instability within the malignant clone, and duration of treatment (Copland et al. 2006). For instance, it has been shown that the level of BCR-ABL expression determines the resistance to imatinib and that elevated expression of BCR-ABL in CD34+ progenitor cells from CML patients in blast crisis make them much less sensitive to imatinib as compared to chronic CD34+ CML cells (Barnes et al. 2005). In chronic phase CML and in newly diagnosed patients at least two mechanisms for imatinib resistance are postulated: mutations in the tyrosine kinase domain which may be present in some patients even before IM treatment (Roche-Lestienne et al. 2002; and Jiang et al. 2007a) and/or disease persistence, resulting from inherent insensitivity of CML stem cells to IM due to BCR-ABL independent survival signals (Deininger and Holyoake, 2005). The more primitive CD34+ CD38- cell subset which constitutes about 5% of the total CD34+ CML cells and the quiescent CML stem cells which are about 1% of CD34+ cells have been shown to be resistant to imatinib (Copland et al. 2006; Copland et al. 2008). It was found that the levels of BCR-ABL and P-crkl are higher in those more primitive CD34+ populations as compared to the total CD34+ cells. In addition, no BCR-ABL mutations where detected in IM resistant CD34+ CD38- cells which may suggest that their IM resistance may be due to BCR-ABL independent mechanisms (Cpland et al. 2008). It is possible therefore that the IM resistance shown here in total CD34+ cells from two CML patients is due to either BCR-ABL mutations on the CD34+ cells or due to the presence of higher percentages of the most primitive CD34+ CD38- cells in theses CML samples as compared with the other IM sensitive CML samples studied in this chapter. Data presented here and by others shows that IM has an immediate effect on CD34+ CML cells on most chronic phase patients. CD34+ cells that were confirmed by the 182

FACS based P-crkl assay to be IM sensitive were used to study the effect of BCRABL inhibition on Notch signalling activity. Real time PCR data showed that IM induced inhibition of BCR-ABL in CD34+ cells resulted in significant up-regulation of Hes1, the Notch target gene, in three CML patients. This finding is interesting as it shows for the first time that BCR-ABL signalling pathway may interact with the Notch signalling pathway in CD34+ chronic CML cells. This effect was observed only in CML samples which were IM sensitive (CML1, CML3, and CML6) and a similar effect was not obsereved in IM resistant samples. It should be noted also that IM does not target Notch directly and does not influence gamma secretase activity in vitro (Eisele et al. 2007). Therefore it can be concluded that up-regulation of Hes1 post imatinib treatment was a BCR-ABL mediated effect. The concentration of growth factors (GF) used in culturing CD34+ CML cells is critical to the interpretation of drug responses and biological activities of this cell population. For example, Chu et al (2004) showed that imatinib led to unexpected activation of MAPK pathway in CD34+ CML cells and demonstrated by comparing high and low concentrations GF conditions that the increased MAPK activity was growth factor dependent effect. In addition, it may be speculated that CD34+ cells may proceed toward terminal differentiation under high-concentration GF conditions. The concentration of growth factors used in this study is regarded by some groups as a "high concentration" growth factor cocktail (Jiang et al. 2007b; Copland et al. 2008). This was used for the short term cultures used in this study in order to stimulate cell division and achieve BCR-ABL inhibition in CD34+ CML cells as these cells showed much more increased resistance to imatinib when cultured in low growth factors concentrations (Jiang et al. 2007b). The intra cellular domain of Notch receptors have been shown to have a cytokine response region (NCR) which could modulate the activity of Notch in response to different cytokines (Bigas et al. 1998). If Notch activation reported here was in response to cytokines in the culture then it should have occurred in both IM treated and untreated CD34+ CML cells. However, Notch activation was only observed in IM treated CD34+ cells. 183

Interestingly, Copland et al (2008) showed that imatinib in the presence of high– concentration growth factors led to increased numbers of CML stem/ progenitor cells via its anti-proliferative effects. Moreover, careful precautions were taken to ensure that IM induced effect reported here was limited to the CD34+ cell population. For example, PCR experiments were performed on cDNA from either CD34+ sorted cells or cells that were enriched by magnetic selection and confirmed to have > 95% CD34+ cells at the end of IM culture. In addition, the CD34 expression of cultured cells was monitored every day by FACS and results showed an enrichment of CD34+ cells at the end of 72h culture (data not shown). Therefore, it is unlikely that the imatinib effect on Notch activation was due to growth factors induced differentiation of CML cells. Imatinib did not have significant effects on crkl phosphorylation in normal CD34+ cells (Chu et al. 2004; and Hamilton et al. 2006). This suggests that enhanced Notch signalling post imatinib treatment may be specific to CD34+ CML cells. This effect of imatinib on Notch activity was previously shown in the CML cell line K562 as well as in the ABL+ cell line SIL-ALL (chapter four). The mechanism by which imatinib induces activation of Notch in CD34+ CML cells is remain to be investigated. Dishevelled is a an essential cytoplasmic component and key player in the Wnt signalling pathway in which its phosphorylation leads to stabilisation of β-catenin and activation of Wnt signalling (Katoh and Katoh, 2007). In contrast, there is evidence that Dishevelled binds physically to Notch and serves to down-regulate Notch signalling in Drosophila (Panin and Irvineseminars, 1998). Since imatinib has been shown to inhibit Wnt signalling in CML cells (Coluccia et al. 2007), it is possible that imatinib up-regulates Notch signalling by inhibiting Dishevelled and thus abolish the inhibitory effect of Dishevelled on Notch signalling. The activation of Notch signalling following imatinib treatment is interesting when compared with the effects of this BCR-ABL inhibitor on other cell survival pathways such as Wnt and Hedgehog signalling on CD34+ CML cells. As mentioned earlier, IM inhibited Wnt signalling and decreased the expression of Hedgehog target genes by 20%. In contrast, imatinib led to Notch activation on CD34+ CML cells. This is

184

may be confusing as Notch signalling has been shown to be active on the sasme CML samples tested here even before exposure to imatinib. One possible explanation is that IM induced Notch activation may represent a compensatory response to inhibition of BCR/ABL tyrosine activity in CD34+ CML cells. In fact, Notch activation post IM treatment may explain the enrichment of CD34+ cells reported here and by others at the end of CD34+ culture. In support of this hypothesis is the finding that Notch1 activation inhibits differentiation of hematopoietic stem cells both in vitro and in vivo and results in enhanced stem cells numbers (Stier et al. 2002). Moreover, activation of Notch4 in normal human marrow or cord blood cells resulted in enhanced stem cell activity and reduced differentiation (Vercauteren and Sutherland, 2004).

5.3.3 Notch inhibition enhances BCR-ABL kinase activity in CD34+ chronic CML cells Gamma secretase is a protease that is composed of a high molecular weight multicomponent complex of transmembrane proteins. This enzyme act beside Noch receptor on other substrates like the amyloid precursor protein (APP) resulting in the production of β-amyloid protein involved in Alzheimer’s disease pathology. Gamma secreatse processes also other substrates like ErbB4, E-cadherin, and CD44 (Tian et al. 2003). The mechanisms by which gamma secretase reacts with these different substrates remains unknown. Gamma secretase inhibitor (GSI) treatment of CD34+ CML cells resulted in Hes1 down-regulation in most CML samples studied here. However, GSI treatment failed to show a decrease in Hes1 mRNA in CD34+ cells from two CML patients. This was unexpected as GSI has been shown to inhibit Notch signalling in normal CD34+ cells and in T-ALL cell lines (Chadwick et al. 2007; Kogoshi et al. 2207). However, failure to inhibit Notch activity by GSI treatment was reported in cancer cells by two groups. Kogoshi et al (2007) showed that GSI did not decrease Hes1 mRNA in two leukaemic cell lines including one myeloid cell line. Zhang et al (2008) studied the role of Notch in osteosarcoma and confirmed the activation of Notch pathway genes and target genes including Hes1 in osteosarcoma cell lines. However, GSI did not down-regulate Hes1 mRNA in the osteosarcoma cell line COL. These findings and

185

the data reported here can be explained by two possible mechanisms. Firstly, it is likely that Hes1 expression in cells that fail to show Hes1 response to GSI is Notch receptor independent and therefore cannot be down-regulated by GSI. For example, it has been shown in Raji B-lymphoma cells that EBNA2 protein activates the transcription factor RBP-J and activates Notch signaling while bypassing the Notch protein (He et al. 2008). The other possibility is that GSI non responding cells may have an as-yet uncharacterised activating mutation in the Notch pathway. Such mutations may result in truncated forms of Notch receptors that do not require ligand binding or gamma secretase activity for nuclear translocation and active signalling. An example is the t(7,9) translocation found in 1% of T-ALL cases and results in the formation of the truncated active Notch1 (TAN1) which is constitutively active and not inhibited by GSI (Grabher et al. 2006). Interestingly, CD34+ cells from those two CML patients who did not respond to GSI showed significant up-regulation of Hes1 when BCR-ABL activity was inhibited by imatinib. It is possible therefore that BCR-ABL may act as a Notch repressor and its inhibition activates down stream proteins which activates the transcription factor RBP-J directly and results in Notch activation while bypassing the Notch receptor. This hypothesis may be supported by the finding presented in chapter three that Notch proteins are not over-expressed in CML. Since Notch signalling was postulated here as a possible candidate for BCR-ABL independent resistance to imatinib, it was anticipated that GSI treatment would result in reduction of BCR-ABL activity. However, it appears that GSI treatment significantly enhanced crkl phosphorylation in CD34+ CML cells. This effect on BCR-ABL activity was observed in most CML samples treated with GSI (n=5) regardless of the effectiveness of GSI in inhibiting the Notch pathway activity. It can be seen therefore that GSI cannot be used in drug combinations to overcome imatinib resistance in CD34+ cells in CML. Keersmaecker et al (2008a) investigated the effect of combining gamma secretase inhibitor with imatinib in the ALL-SIL cell line. Since these T-ALL cells had Notch1 activating mutations as well as the ABL fusion protein the authors attempted to combine inhibitors of Notch and ABL to see if this combination could offer a 186

therapeutic advantage over using GSI alone to inhibit cell growth. However, it was found that the inhibitory effect of imatinib on cell proliferation was antagonised by GSI when the two drugs were added at the same time. Although the authors could not see an increase in ABL phosphorylation by western blotting following GSI treatment, the data presented here shows, through an unidentified mechanism yet, a significant increase in P-crkl phosphorylation in CD34+ CML cells post GSI treatment. This may suggest that Notch antagonises ABL in CML and inhibition of Notch may increase the activity of BCR-ABL in the contest of CD34+ cells in CML. Similar interaction between Notch and ABL was described in Drosophila. Genetic interaction studies on the mechanisms that regulate the ISNb motor nerve development in Drosophila have shown that Notch and ABL, via unknown mechanism, act antagonistically and that their gain- and loss-of-function phenotypes equally suppress one another in ISNb (Crowner et al. 2003). The mechanism by which Notch antagonises ABL in CML is not yet clear and further research is required to elucidate other signals or pathways that control this interaction. Gamma secreatse inhibitors may also inhibit other targets like the amyloid precursor protein (APP), ErbB4, E-cadherin, and CD44 (Tian et al. 2003). However, a link between the inhibition of any of these substrates and increased BCR-ABL activity in CML seems unlikely.

187

Chapter 6: Final discussion The Notch signalling pathway has been suggested to play a vital role in HSC survival and self renewal (Duncan et al. 2005). Aberrant NOTCH1 expression has been identified as a causative factor in the development of a subset of T-cell acute lymphoblastic leukemia (T-ALL) (Ellisen et al. 1991) and activating mutations in Notch1 have been identified in more than 50% of human T-ALL (Weng et al. 2004). However, the role of Notch signalling in other leukaemias is not well established. The data from this project demonstrates for the first time that expression of Notch1 and Notch2 at the mRNA level is up-regulated in CD34+ cells from chronic phase CML patients as compared with CD34+ cells from normal donors. Furthermore Notch signalling, as assessed by Hes1 expression, is also up-regulated in this cell subset in chronic phase CML patients, indicating a hyperactivation of the Notch signalling pathway. The mechanisms underlying the activation of Notch signalling demonstrated here in chronic phase CML patients remain to be elucidated. However, Notch1 activating mutations have been identified in T-ALL cells and have been shown to contribute to leukaemogenesis by facilitating ligand-independent pathway activation or by increasing the half-life of active Notch1 intracellular domain (ICN1) (Weng et al. 2004). In T-ALL cells inhibition of aberreant Notch signalling by GSIs leads to decreased proliferation (Weng et al. 2004) and increased sensitivity to apoptosis 188

(Chadwick et al. submitted), suggesting that Notch activation contributes to the transformation of the cells through these effects. Therefore, one possibility is that activating Notch mutations also occur in CML and that these are responsible for the increased signalling seen in this study. Consequently it would be intriguing to examine in future studies whether activating Notch1 mutations are found in CML. A second possibility is that Notch activity is high in CML CD34+ cells as a result of changes in ligand concentration in the leukaemic microenvironment. It does not appear likely that there would be alterations in Notch ligand expression on nonleukaemic cells in the microenvironment, but there remains the possibility that an up regulation of ligand on neighboring leukaemic cells is occurring. A detailed analysis of ligand expression on CML and normal cells would indicate whether this was the case. A third possibility is that Notch signalling is up regulated as a result of cross talk with the BCR-ABL signalling pathway and this possibility was investigated in the present study. The interaction between Notch and ABL has been observed before in Drosophila where Notch and abl mutations interact synergistically to produce synthetic lethality and defects in axon extension (Giniger, 1998). The author has also shown in another study that Notch and its ligand Delta function in the ISNb motor nerve patterning in Drosophila mainly to antagonise ABL (Crowner et al. 2003). This shows clearly that the cross-talk between Notch and ABL may be synergistic or antagonistic depending on the developmental context. Most recently Mizuno et al. (2008) have demonstrated that over-expression or enhanced kinase activity of BCRABL and altered expression of Notch1 synergises to induce acute leukemia in a transgenic model for CML. To test the hypothesis that Notch and BCR-ABL may interact in CML, leukaemic cell line models were characterised for the expression of intact and inhibitable Notch and ABL activity. After establishing the optimised parameters for the FACS based P-crkl assay in this project as a surrogate marker of ABL kinase activity and by utilizing the BCR-ABL inhibitor imatinib mesylate (IM) it was confirmed that K562 and ALL-SIL cell lines have a constitutive active ABL kinase activity that can be blocked by IM enabling a study of the effects of BCR-ABL on the Notch signalling pathway. On the other hand, Notch activity was evident in K562 cells as assessed by the expression of Hes1 by real time PCR and this activity 189

was responsive to the inhibition induced by GSI allowing the study of the effects of Notch signalling on BCR-ABL activity. Active Notch signalling in the ALL-SIL cells is well documented by others (Quinta´s-Cardama et al.. 2008; and De Keersmaecker et al.. 2008). It can be concluded therefore that both K562 and ALL-SIL cell lines may offer themselves as suitable in vitro models to investigate the cross-talk between Notch and ABL signalling pathways. By using inhibitors of both Notch and ABL signalling, it was found for the first time that the Notch and ABL pathways antagonise each other in K562 and ALL-SIL leukaemic cell lines. These data were then confirmed in primary CD34+ cells isolated from chronic phase CML patients. The treatment of CD34+ CML cells with IM resulted in significant upregulation of the Notch target gene Hes1 in three out of the four CML samples that responded to IM treatment. Likewise FACS data showed that GSI treatment significantly increased the P-crkl expression in CML CD34+ cells. Taken together this data implies that Notch and BCR-ABL antagonise each other in primary tissue from chronic phase CML patients. The exact mechanisms underlying the GSI induced activation of BCR-ABL signalling and the IM induced activation of Notch signalling are unknown. Giniger at al. (1998) suggested two possible mechanisms to explain the cross talk observed between Notch and ABL in Drosophila. He proposed that Notch may directly bind to ABL (or possibly bind via an adapter protein) and through this direct physical association the two pathways may regulate each other. Alternatively the two molecules may not physically interact but may interact at the level of the downstream components of the pathway. The proposed interaction in this project between Notch and BCR-ABL in chronic phase CML warrants further investigation including coimmunoprecipitation studies to explore the molecular level at which this interaction occurs and whether this interaction is direct or requires other cellular mediators or common signalling pathways. The possibility of cross talk between downstream components is interesting because it has been demonstrated recently in T-ALL cells that Notch activation positively regulates the phosphatidylinositol 3-kinase (PI3K)/AKT pathway (Palomero et al.

190

2007). Activation of the PI3K/AKT downstream of Notch signalling was found to be mediated via inhibition of PTEN (phosphatase and tensin homologue) by the Notch target gene Hes1. Hes1 was up-regulated in the ALL-SIL T-ALL cell line, a cell line that has constitutive Notch and ABL kinase activities, following the exposure to IM as demonstrated in chapter 4. It is possible, therefore, that Hes1 up-regulation following IM exposure in CML cells may also activate the PI3K/AKT pathway and confer anti apoptotic signals to CML cells regardless of the BCR-ABL repressed activity (Fig. 6.1). More work is needed to investigate whether the activation of the PI3K pathway by Notch signalling reported in T-ALL also occurs in CML cells. Although IM has been shown to inhibit BCR-ABL activity in CD34+ chronic phase CML cells, only a mild increase in apoptosis was demonstrated in these cells (Chu et al. 2004). Moreover, it has been shown that IM treatment activated the PI3K/ Akt/ mammalian target of rapamycin (mTor)- anti apoptotic pathway in chronic phase CML patients as well as in BCR-ABL+ Lama cells (Burchert et al. 2005). This was unexpected because BCR-ABL is upstream of the PI3K/AKT pathway and blocking the BCR-ABL activity by IM was anticipated to repress the anti-apoptotic activity of the PI3K signalling and induce apoptosis. The authors proposed that the IM-induced compensatory PI3K-Akt/mTor activation may represent a novel mechanism for the persistence of BCR/ABL-positive cells in IM treated CML patients. In fact, the IM induced activation of Notch signalling reported in this project may also help to explain why blocking BCR-ABL activity by IM is not enough to switch off the PI3K/AKT/mTor anti apoptotic activity. It is also possible that the antagonistic effects between Notch and BCR-ABL signalling seen in this study are a reflection of the involvement of additional pathways active in the CML cells. In the experiments presented here modulation of the Notch and BCR-ABL pathways have been investigated. However, haematopoietic progenitor cells have been reported to secrete Wnt (Duncan et al. 2005) and the possibility that this pathway may therefore be active in these experiments cannot be ruled out. IM has been shown to inhibit Wnt signalling in CML cells (Coluccia et al. 2007) and in the murine myeloid progenitor cell line 32Dcl3 (Tickenbrock et al.. 2008) in a way that may involve inhibition of Dishevelled and activation of GSK3β, both of which are key players in the canonical Wnt signalling pathway. Dishevelled has been reported to

191

bind to Notch and down-regulate Notch signalling in Drosophila (Panin and Irvineseminars, 1998). In contrast, it has been shown in cell line models that GSK3β positively modulates Notch signalling by protecting the intracellular domain of Notch1 (ICN1) from proteasome degradation (Foltz et al. 2002). Taken together, IM may activate Notch signalling by modulating the Wnt components Dishevelled and GSK3β (Fig. 6.1). It is also possible that IM may activate Notch signalling by blocking the inhibitory action of BCR-ABL on its downstream substrate GSK3β. Although the in vitro inhibitor based loss of function approaches described here suggest that Notch and BCR-ABL antagonise each other, the co-existence of activated Notch and BCR-ABL in vivo in chronic phase CML suggest a cooperative interaction between the two signalling pathways. One possibility is that both BCR-ABL and Notch signalling are equally critical for CML cell survival and resistance to apoptosis and that in vitro inhibition of one of the two signalling pathways may trigger a compensatory activation of the other pathway to compensate for the loss of total survival signals (Fig. 6.2). This hypothetical model would require regulatory molecules in the cytoplasm to sense the reduction of the survival signals from one pathway and respond by increasing the activity of the other pathway to compensate for the reduction in total anti-apoptotic signals in BCR-ABL+ cells. These regulatory molecules themselves may be part of a feedback loop of Notch and BRC-ABL signalling targets. This model may also explain the activation of Notch signalling in chronic phase CD34+ CML cells before any manipulation of these pathways. The outcome of Notch signalling is known to be highly cell context specific. For example it may lead to resistance to apoptosis in some cell contexts (Sade et al. 2004), and lead to sensitivity to apoptosis in other cell types (Zweidler-McKay et al. 2005). In this study BCR-ABL and Notch cross-talk has been investigated in the context of CD34+ cells from chronic phase CML patients and it was confirmed in chapter three that Notch signalling was hyperactive in these cells. Notch signalling was also upregulated in the more primitive CD34+Thy+ compartment. Most available BCR-ABL inhibitors have been shown to be effective against CD34+ CML cells but not against the more primitive CD34+ CD38- cell subset in chronic phase CML (Copland et al.. 2006). It would therefore be important in future work to investigate

192

the effect of Notch inhibition on BCR-ABL activity in the context of the CD34+ CD38-/Thy+ cell subset by including an extra CD34 and CD38/Thy surface staining step in the P-crkl assay, enabling thereby the measurement of P-crkl content in the FACS gated CD34+ CD38-/Thy+ cell subset. Similarly, the effect of BCR-ABL inhibition on Notch activity within the CD34+ CD38-/Thy+ cell subset could be attempted in the future by using a BCR-ABL inhibitor that is effective against the primitive CD34+ CD38-/Thy+ cells such as Dasatinib (Copland et al. 2006). Whether survival and self renewal of CML stem cells is critically dependant on intact Notch signalling remains an open question. This question could be addressed by assessing the proliferation and apoptosis status of the cells following the treatments with inhibitors outlined in this study. This could also be investigated in vivo in a mouse model by generating a mouse in which Notch signalling is inactivated by either a dominant-negative version of the RBPJ protein (DNRBPJ) or a dominant negative version of the co-activator MAML (DNMAML) (Maillard et al. 2008). Stem cell enriched cells from control or Notch inactive mice could be then transfected with control retroviruses or viruses carrying the p210 BCR-ABL and transplanted into lethally irradiated mice to test the requirement of Notch signalling in induction and maintenance of CML in vivo. Clearly many more functional studies are required to investigate the biological and cellular events that result from the activation of Notch signalling in chronic phase CML. If Notch signalling was proven to be critical for the survival or proliferation of chronic phase CML cells, as observed in T-ALL, a more detailed model for the role of Notch in CML progression could be established. One possibility is that in the context of the chronic phase of CML Notch is activated and confers proliferation and cell survival of CD34+ CML cells. Progression to the blastic phase of CML is then associated with down-regulation of Notch (Sengupta et al. 2007). This fits very well with the finding in this project that activation of Notch in the blastic phase CML K562 cells resulted in inhibition of proliferation and by the recent findings that this may induce apoptosis in K562 cells (Yin et al. 2008). Therefore, it can be postulated that the outcome of Notch signalling in CML will depend on the cellular context.

193

Activation Inhibition Fig. 6.1. Proposed model for Notch and BCR-ABL cross-talk in CML. Both BCR-ABL and Notch signalling activate the PI3K-AKT-mTOR signalling pathway which confers survival and proliferation signals to CML cells. Blocking BCR-ABL kinase activity by IM is

Hes1

not sufficient to induce apoptosis in CML cells because they may switch their addiction for survial signals to the PI3K signalling activated by Notch. In this model, IM up-regulates

Notch by modulating Wnt pathway components GSK3β and or Dishevelled. IM induced activation of GSK3β or IM induced inhibition of Dishevelled stabilises ICN in the cytoplasm 194 signalling by up-regulation of Hes1 which which in turn activates the PI3K-AKT-mTOR abolish the inhibitory effect of PTEN on PI3K pathway.

A P

Fig.6.2. The cooperative model of activated Notch and BCR-ABL signalling in chronic phase CML. Both Notch and BCR-ABL are activated in chronic phase CML where the two

BCR-ABL

NOTC

signalling pathways may activate survival signalling pathways to inhibit apoptosis in CD34+ CML cells. In vitro inhibition of Notch signalling by GSI results in compensatory activation of BCR-ABL activity to keep the same level of survival signals required for CML cell

survival (A). Exposure to IM in vitro leads to compensatory activation of Notch signalling to maintain the same level of survival signals needed by CML cells to inhibit apoptosis (B). The net effect is maintenance of balanced levels of survival signals that protect CD34+ CML cells from apoptosis in the chronic phase of CML. (GSI: gamma secretase inhibitor, IM: imatinib mesylate).

195

References 1

2 3

4

5 6

7

Acar, M., Jafar-Nejad, H., Takeuchi, H., Rajan, A., Ibrani, D., Rana, N. A., Pan, H., Haltiwanger, R. S. and Bellen, H. J. (2008) Rumi is a CAP10 domain glycosyltransferase that modifies Notch and is required for Notch signaling. Cell 132, 247-258 Afar, D. E., Goga, A., McLaughlin, J., Witte, O. N. and Sawyers, C. L. (1994) Differential complementation of Bcr-Abl point mutants with cMyc. Science 264, 424-426 Alcalay, M., Meani, N., Gelmetti, V., Fantozzi, A., Fagioli, M., Orleth, A., Riganelli, D., Sebastiani, C., Cappelli, E., Casciari, C., Sciurpi, M. T., Mariano, A. R., Minardi, S. P., Luzi, L., Muller, H., Di Fiore, P. P., Frosina, G. and Pelicci, P. G. (2003) Acute myeloid leukemia fusion proteins deregulate genes involved in stem cell maintenance and DNA repair. J Clin Invest 112, 1751-1761 Allman, D., Karnell, F. G., Punt, J. A., Bakkour, S., Xu, L., Myung, P., Koretzky, G. A., Pui, J. C., Aster, J. C. and Pear, W. S. (2001) Separation of Notch1 promoted lineage commitment and expansion/transformation in developing T cells. J Exp Med 194, 99-106 Allman, D., Punt, J. A., Izon, D. J., Aster, J. C. and Pear, W. S. (2002) An invitation to T and more: notch signaling in lymphopoiesis. Cell 109 Suppl, S1-11 Amarante-Mendes, G. P., Naekyung Kim, C., Liu, L., Huang, Y., Perkins, C. L., Green, D. R. and Bhalla, K. (1998) Bcr-Abl exerts its antiapoptotic effect against diverse apoptotic stimuli through blockage of mitochondrial release of cytochrome C and activation of caspase-3. Blood 91, 1700-1705 Arai, A., Kanda, E. and Miura, O. (2002) Rac is activated by erythropoietin or interleukin-3 and is involved in activation of the Erk signaling pathway. Oncogene 21, 2641-2651

196

8 9

10 11 12 13

14

15 16

17

18 19

20 21

Arai, F. and Suda, T. (2007) Maintenance of quiescent hematopoietic stem cells in the osteoblastic niche. Ann N Y Acad Sci 1106, 41-53 Armstrong, F., Brunet de la Grange, P., Gerby, B., Rouyez, M. C., Calvo, J., Fontenay, M., Boissel, N., Dombret, H., Baruchel, A., Landman-Parker, J., Romeo, P. H., Ballerini, P. and Pflumio, F. (2008) NOTCH is a key regulator of human T-cell acute leukaemia initiating cell activity. Blood Artavanis-Tsakonas, S., Muskavitch, M. A. and Yedvobnick, B. (1983) Molecular cloning of Notch, a locus affecting neurogenesis in Drosophila melanogaster. Proc Natl Acad Sci U S A 80, 1977-1981 Artavanis-Tsakonas, S., Rand, M. D. and Lake, R. J. (1999) Notch signaling: cell fate control and signal integration in development. Science 284, 770-776 Aster, J. C., Pear, W. S. and Blacklow, S. C. (2008) Notch signaling in leukemia. Annu Rev Pathol 3, 587-613 Avigdor, A., Goichberg, P., Shivtiel, S., Dar, A., Peled, A., Samira, S., Kollet, O., Hershkoviz, R., Alon, R., Hardan, I., Ben-Hur, H., Naor, D., Nagler, A. and Lapidot, T. (2004) CD44 and hyaluronic acid cooperate with SDF-1 in the trafficking of human CD34+ stem/progenitor cells to bone marrow. Blood 103, 2981-2989 Barnes, D. J., Palaiologou, D., Panousopoulou, E., Schultheis, B., Yong, A. S., Wong, A., Pattacini, L., Goldman, J. M. and Melo, J. V. (2005) Bcr-Abl expression levels determine the rate of development of resistance to imatinib mesylate in chronic myeloid leukemia. Cancer Res 65, 8912-8919 Baron, M. (2003) An overview of the Notch signalling pathway. Semin Cell Dev Biol 14, 113-119 Bellavia, D., Campese, A. F., Alesse, E., Vacca, A., Felli, M. P., Balestri, A., Stoppacciaro, A., Tiveron, C., Tatangelo, L., Giovarelli, M., Gaetano, C., Ruco, L., Hoffman, E. S., Hayday, A. C., Lendahl, U., Frati, L., Gulino, A. and Screpanti, I. (2000) Constitutive activation of NF-kappaB and T-cell leukemia/lymphoma in Notch3 transgenic mice. Embo J 19, 3337-3348 Bellavia, D., Campese, A. F., Checquolo, S., Balestri, A., Biondi, A., Cazzaniga, G., Lendahl, U., Fehling, H. J., Hayday, A. C., Frati, L., von Boehmer, H., Gulino, A. and Screpanti, I. (2002) Combined expression of pTalpha and Notch3 in T cell leukemia identifies the requirement of preTCR for leukemogenesis. Proc Natl Acad Sci U S A 99, 3788-3793 Bellavia, D., Campese, A. F., Vacca, A., Gulino, A. and Screpanti, I. (2003) Notch3, another Notch in T cell development. Semin Immunol 15, 107-112 Bhardwaj, G., Murdoch, B., Wu, D., Baker, D. P., Williams, K. P., Chadwick, K., Ling, L. E., Karanu, F. N. and Bhatia, M. (2001) Sonic hedgehog induces the proliferation of primitive human hematopoietic cells via BMP regulation. Nat Immunol 2, 172-180 Bhatia, M., Bonnet, D., Murdoch, B., Gan, O. I. and Dick, J. E. (1998) A newly discovered class of human hematopoietic cells with SCIDrepopulating activity. Nat Med 4, 1038-1045 Bhatia, R., Holtz, M., Niu, N., Gray, R., Snyder, D. S., Sawyers, C. L., Arber, D. A., Slovak, M. L. and Forman, S. J. (2003) Persistence of

197

22 23 24 25 26 27 28 29

30 31

32 33 34 35 36 37

malignant hematopoietic progenitors in chronic myelogenous leukemia patients in complete cytogenetic remission following imatinib mesylate treatment. Blood 101, 4701-4707 Bigas, A., Martin, D. I. and Milner, L. A. (1998) Notch1 and Notch2 inhibit myeloid differentiation in response to different cytokines. Mol Cell Biol 18, 2324-2333 Bonde, J., Hess, D. A. and Nolta, J. A. (2004) Recent advances in hematopoietic stem cell biology. Curr Opin Hematol 11, 392-398 Bonnet, D. (2003) Biology of human bone marrow stem cells. Clin Exp Med 3, 140-149 Brady, G. and Iscove, N. N. (1993) Construction of cDNA libraries from single cells. Methods Enzymol 225, 611-623 Bruchova, H., Borovanova, T., Klamova, H. and Brdicka, R. (2002) Gene expression profiling in chronic myeloid leukemia patients treated with hydroxyurea. Leuk Lymphoma 43, 1289-1295 Bruecher-Encke, B., Griffin, J. D., Neel, B. G. and Lorenz, U. (2001) Role of the tyrosine phosphatase SHP-1 in K562 cell differentiation. Leukemia 15, 1424-1432 Burchert, A. (2007) Roots of imatinib resistance: a question of selfrenewal? Drug Resist Updat 10, 152-161 Burchert, A., Wang, Y., Cai, D., von Bubnoff, N., Paschka, P., MullerBrusselbach, S., Ottmann, O. G., Duyster, J., Hochhaus, A. and Neubauer, A. (2005) Compensatory PI3-kinase/Akt/mTor activation regulates imatinib resistance development. Leukemia 19, 1774-1782 Callahan, R. and Raafat, A. (2001) Notch signaling in mammary gland tumorigenesis. J Mammary Gland Biol Neoplasia 6, 23-36 Calvi, L. M., Adams, G. B., Weibrecht, K. W., Weber, J. M., Olson, D. P., Knight, M. C., Martin, R. P., Schipani, E., Divieti, P., Bringhurst, F. R., Milner, L. A., Kronenberg, H. M. and Scadden, D. T. (2003) Osteoblastic cells regulate the haematopoietic stem cell niche. Nature 425, 841-846 Cantley, L. C. (2002) The phosphoinositide 3-kinase pathway. Science 296, 1655-1657 Carlesso, N., Aster, J. C., Sklar, J. and Scadden, D. T. (1999) Notch1induced delay of human hematopoietic progenitor cell differentiation is associated with altered cell cycle kinetics. Blood 93, 838-848 Chabannon, C., Wood, P. and Torok-Storb, B. (1992) Expression of CD7 on normal human myeloid progenitors. J Immunol 149, 2110-2113 Chadwick, N., Fennessy, C., Nostro, M. C., Baron, M., Brady, G. and Buckle, A. M. (2008) Notch induces cell cycle arrest and apoptosis in human erythroleukaemic TF-1 cells. Blood Cells Mol Dis 41, 270-277 Chadwick, N., Nostro, M. C., Baron, M., Mottram, R., Brady, G. and Buckle, A. M. (2007) Notch signaling induces apoptosis in primary human CD34+ hematopoietic progenitor cells. Stem Cells 25, 203-210 Chai, S. K., Nichols, G. L. and Rothman, P. (1997) Constitutive activation of JAKs and STATs in BCR-Abl-expressing cell lines and peripheral blood cells derived from leukemic patients. J Immunol 159, 4720-4728

198

38 39 40 41 42 43 44 45

46

47

48 49

50 51

Cheng, T., Rodrigues, N., Dombkowski, D., Stier, S. and Scadden, D. T. (2000) Stem cell repopulation efficiency but not pool size is governed by p27(kip1). Nat Med 6, 1235-1240 Chiaramonte, R., Basile, A., Tassi, E., Calzavara, E., Cecchinato, V., Rossi, V., Biondi, A. and Comi, P. (2005) A wide role for NOTCH1 signaling in acute leukemia. Cancer Lett 219, 113-120 Christ, O., Gunthert, U., Haas, R. and Zoller, M. (2001) Importance of CD44v7 isoforms for homing and seeding of hematopoietic progenitor cells. J Leukoc Biol 69, 343-352 Chu, S., Holtz, M., Gupta, M. and Bhatia, R. (2004) BCR/ABL kinase inhibition by imatinib mesylate enhances MAP kinase activity in chronic myelogenous leukemia CD34+ cells. Blood 103, 3167-3174 Clarkson, B., Strife, A., Wisniewski, D., Lambek, C. L. and Liu, C. (2003) Chronic myelogenous leukemia as a paradigm of early cancer and possible curative strategies. Leukemia 17, 1211-1262 Clarkson, B. D., Strife, A., Wisniewski, D., Lambek, C. and Carpino, N. (1997) New understanding of the pathogenesis of CML: a prototype of early neoplasia. Leukemia 11, 1404-1428 Collins, B. J., Kleeberger, W. and Ball, D. W. (2004) Notch in lung development and lung cancer. Semin Cancer Biol 14, 357-364 Coluccia, A. M., Vacca, A., Dunach, M., Mologni, L., Redaelli, S., Bustos, V. H., Benati, D., Pinna, L. A. and Gambacorti-Passerini, C. (2007) Bcr-Abl stabilizes beta-catenin in chronic myeloid leukemia through its tyrosine phosphorylation. Embo J 26, 1456-1466 Copland, M., Hamilton, A., Elrick, L. J., Baird, J. W., Allan, E. K., Jordanides, N., Barow, M., Mountford, J. C. and Holyoake, T. L. (2006) Dasatinib (BMS-354825) targets an earlier progenitor population than imatinib in primary CML but does not eliminate the quiescent fraction. Blood 107, 4532-4539 Copland, M., Pellicano, F., Richmond, L., Allan, E. K., Hamilton, A., Lee, F. Y., Weinmann, R. and Holyoake, T. L. (2008) BMS-214662 potently induces apoptosis of chronic myeloid leukemia stem and progenitor cells and synergizes with tyrosine kinase inhibitors. Blood 111, 2843-2853 Coulombel, L. (2004) Identification of hematopoietic stem/progenitor cells: strength and drawbacks of functional assays. Oncogene 23, 7210-7222 Craddock, C. F., Nakamoto, B., Andrews, R. G., Priestley, G. V. and Papayannopoulou, T. (1997) Antibodies to VLA4 integrin mobilize longterm repopulating cells and augment cytokine-induced mobilization in primates and mice. Blood 90, 4779-4788 Crowner, D., Le Gall, M., Gates, M. A. and Giniger, E. (2003) Notch steers Drosophila ISNb motor axons by regulating the Abl signaling pathway. Curr Biol 13, 967-972 Dando, J. S., Tavian, M., Catelain, C., Poirault, S., Bennaceur-Griscelli, A., Sainteny, F., Vainchenker, W., Peault, B. and Lauret, E. (2005) Notch/Delta4 interaction in human embryonic liver CD34+ CD38- cells: positive influence on BFU-E production and LTC-IC potential maintenance. Stem Cells 23, 550-560

199

52

53 54

55

56 57 58 59

60

61

62 63

64

Das, I., Craig, C., Funahashi, Y., Jung, K. M., Kim, T. W., Byers, R., Weng, A. P., Kutok, J. L., Aster, J. C. and Kitajewski, J. (2004) Notch oncoproteins depend on gamma-secretase/presenilin activity for processing and function. J Biol Chem 279, 30771-30780 Davis, R. L. and Turner, D. L. (2001) Vertebrate hairy and Enhancer of split related proteins: transcriptional repressors regulating cellular differentiation and embryonic patterning. Oncogene 20, 8342-8357 De Keersmaecker, K., Lahortiga, I., Mentens, N., Folens, C., Van Neste, L., Bekaert, S., Vandenberghe, P., Odero, M. D., Marynen, P. and Cools, J. (2008a) In vitro validation of gamma-secretase inhibitors alone or in combination with other anti-cancer drugs for the treatment of T-cell acute lymphoblastic leukemia. Haematologica 93, 533-542 De Keersmaecker, K., Versele, M., Cools, J., Superti-Furga, G. and Hantschel, O. (2008b) Intrinsic differences between the catalytic properties of the oncogenic NUP214-ABL1 and BCR-ABL1 fusion protein kinases. Leukemia Deftos, M. L. and Bevan, M. J. (2000) Notch signaling in T cell development. Curr Opin Immunol 12, 166-172 Deininger, M. W., Goldman, J. M. and Melo, J. V. (2000) The molecular biology of chronic myeloid leukemia. Blood 96, 3343-3356 Deininger, M. W. and Holyoake, T. L. (2005) Can we afford to let sleeping dogs lie? Blood 105, 1840-1841 Delforge, M., Boogaerts, M. A., McGlave, P. B. and Verfaillie, C. M. (1999) BCR/ABL- CD34(+)HLA-DR- progenitor cells in early chronic phase, but not in more advanced phases, of chronic myelogenous leukemia are polyclonal. Blood 93, 284-292 Desplat, V., Lagarde, V., Belloc, F., Chollet, C., Leguay, T., Pasquet, J. M., Praloran, V. and Mahon, F. X. (2004) Rapid detection of phosphotyrosine proteins by flow cytometric analysis in Bcr-Ablpositive cells. Cytometry A 62, 35-45 Dierks, C., Beigi, R., Guo, G. R., Zirlik, K., Stegert, M. R., Manley, P., Trussell, C., Schmitt-Graeff, A., Landwerlin, K., Veelken, H. and Warmuth, M. (2008) Expansion of Bcr-Abl-positive leukemic stem cells is dependent on Hedgehog pathway activation. Cancer Cell 14, 238249 Drexler, H. G., MacLeod, R. A. and Uphoff, C. C. (1999) Leukemia cell lines: in vitro models for the study of Philadelphia chromosome-positive leukemia. Leuk Res 23, 207-215 Druker, B. J., Guilhot, F., O'Brien, S. G., Gathmann, I., Kantarjian, H., Gattermann, N., Deininger, M. W., Silver, R. T., Goldman, J. M., Stone, R. M., Cervantes, F., Hochhaus, A., Powell, B. L., Gabrilove, J. L., Rousselot, P., Reiffers, J., Cornelissen, J. J., Hughes, T., Agis, H., Fischer, T., Verhoef, G., Shepherd, J., Saglio, G., Gratwohl, A., Nielsen, J. L., Radich, J. P., Simonsson, B., Taylor, K., Baccarani, M., So, C., Letvak, L. and Larson, R. A. (2006) Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med 355, 2408-2417 Druker, B. J., Tamura, S., Buchdunger, E., Ohno, S., Segal, G. M., Fanning, S., Zimmermann, J. and Lydon, N. B. (1996) Effects of a

200

65

66

67

68 69

70

71 72

73

74 75 76 77 78

selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med 2, 561-566 Dube, I. D., Arlin, Z. A., Kalousek, D. K., Eaves, C. J. and Eaves, A. C. (1984) Nonclonal hemopoietic progenitor cells detected in long-term marrow cultures from a Turner syndrome mosaic with chronic myeloid leukemia. Blood 64, 1284-1287 Duechler, M., Shehata, M., Schwarzmeier, J. D., Hoelbl, A., Hilgarth, M. and Hubmann, R. (2005) Induction of apoptosis by proteasome inhibitors in B-CLL cells is associated with downregulation of CD23 and inactivation of Notch2. Leukemia 19, 260-267 Duncan, A. W., Rattis, F. M., DiMascio, L. N., Congdon, K. L., Pazianos, G., Zhao, C., Yoon, K., Cook, J. M., Willert, K., Gaiano, N. and Reya, T. (2005) Integration of Notch and Wnt signaling in hematopoietic stem cell maintenance. Nat Immunol 6, 314-322 Effros, R. B. and Globerson, A. (2002) Hematopoietic cells and replicative senescence. Exp Gerontol 37, 191-196 Eisele, Y. S., Baumann, M., Klebl, B., Nordhammer, C., Jucker, M. and Kilger, E. (2007) Gleevec increases levels of the amyloid precursor protein intracellular domain and of the amyloid-beta degrading enzyme neprilysin. Mol Biol Cell 18, 3591-3600 Ellisen, L. W., Bird, J., West, D. C., Soreng, A. L., Reynolds, T. C., Smith, S. D. and Sklar, J. (1991) TAN-1, the human homolog of the Drosophila notch gene, is broken by chromosomal translocations in T lymphoblastic neoplasms. Cell 66, 649-661 Elrick, L. J., Jorgensen, H. G., Mountford, J. C. and Holyoake, T. L. (2005) Punish the parent not the progeny. Blood 105, 1862-1866 Fang, B., Zheng, C., Liao, L., Han, Q., Sun, Z., Jiang, X. and Zhao, R. C. (2005) Identification of human chronic myelogenous leukemia progenitor cells with hemangioblastic characteristics. Blood 105, 27332740 Fetscher, S., Mertelsmann, R. (2002.) Haematopoietic geowth factors: therapeutic uses. In Nature Encyclopedia of Life Sciences., Nature publishing group. Fleming, R. J. (1998) Structural conservation of Notch receptors and ligands. Semin Cell Dev Biol 9, 599-607 Foltz, D. R., Santiago, M. C., Berechid, B. E. and Nye, J. S. (2002) Glycogen synthase kinase-3beta modulates notch signaling and stability. Curr Biol 12, 1006-1011 Fortini, M. E. (2002) Gamma-secretase-mediated proteolysis in cellsurface-receptor signalling. Nat Rev Mol Cell Biol 3, 673-684 Fowlkes, B. J. and Robey, E. A. (2002) A reassessment of the effect of activated Notch1 on CD4 and CD8 T cell development. J Immunol 169, 1817-1821 Frassoni, F., Podesta, M., Piaggio, G., Rosti, V., Pitto, A., Benvenuto, F., Figari, O., Vassallo, F., Carella, A. M., Zikos, P., Bergamaschi, G., Fugazza, G., Sessarego, M. and Cazzola, M. (1999) Normal primitive haemopoietic progenitors are more frequent than their leukaemic counterpart in newly diagnosed patients with chronic myeloid leukaemia but rapidly decline with time. Br J Haematol 104, 538-545

201

79 80 81

82 83

84 85 86

87 88

89

90

91

92 93

Frisch, B. J., Porter, R. L. and Calvi, L. M. (2008) Hematopoietic niche and bone meet. Curr Opin Support Palliat Care 2, 211-217 Fuchs, E., Tumbar, T. and Guasch, G. (2004) Socializing with the neighbors: stem cells and their niche. Cell 116, 769-778 Furukawa, Y., Kikuchi, J., Nakamura, M., Iwase, S., Yamada, H. and Matsuda, M. (2000) Lineage-specific regulation of cell cycle control gene expression during haematopoietic cell differentiation. Br J Haematol 110, 663-673 Giniger, E. (1998) A role for Abl in Notch signaling. Neuron 20, 667-681 Girard, L., Hanna, Z., Beaulieu, N., Hoemann, C. D., Simard, C., Kozak, C. A. and Jolicoeur, P. (1996) Frequent provirus insertional mutagenesis of Notch1 in thymomas of MMTVD/myc transgenic mice suggests a collaboration of c-myc and Notch1 for oncogenesis. Genes Dev 10, 1930-1944 Goldsby, R. A. and Goldsby, R. A. K. i. (2003) Immunology. W.H. Freeman, New York ; Houndmills Gordon, M. Y., Dowding, C. R., Riley, G. P., Goldman, J. M. and Greaves, M. F. (1989) Adhesive defects in chronic myeloid leukemia. Curr Top Microbiol Immunol 149, 151-155 Gorre, M. E., Mohammed, M., Ellwood, K., Hsu, N., Paquette, R., Rao, P. N. and Sawyers, C. L. (2001) Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science 293, 876-880 Grabher, C., von Boehmer, H. and Look, A. T. (2006) Notch 1 activation in the molecular pathogenesis of T-cell acute lymphoblastic leukaemia. Nat Rev Cancer 6, 347-359 Graham, S. M., Jorgensen, H. G., Allan, E., Pearson, C., Alcorn, M. J., Richmond, L. and Holyoake, T. L. (2002) Primitive, quiescent, Philadelphia-positive stem cells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro. Blood 99, 319-325 Grand, F. H., Marley, S. B., Chase, A., Titley, I., Healy, L., Spencer, A., Reiter, A., Goldman, J. M. and Gordon, M. Y. (1997) BCR/ABLnegative progenitors are enriched in the adherent fraction of CD34+ cells circulating in the blood of chronic phase chronic myeloid leukemia patients. Leukemia 11, 1486-1492 Graux, C., Cools, J., Melotte, C., Quentmeier, H., Ferrando, A., Levine, R., Vermeesch, J. R., Stul, M., Dutta, B., Boeckx, N., Bosly, A., Heimann, P., Uyttebroeck, A., Mentens, N., Somers, R., MacLeod, R. A., Drexler, H. G., Look, A. T., Gilliland, D. G., Michaux, L., Vandenberghe, P., Wlodarska, I., Marynen, P. and Hagemeijer, A. (2004) Fusion of NUP214 to ABL1 on amplified episomes in T-cell acute lymphoblastic leukemia. Nat Genet 36, 1084-1089 Greenblatt, D. Y., Vaccaro, A. M., Jaskula-Sztul, R., Ning, L., Haymart, M., Kunnimalaiyaan, M. and Chen, H. (2007) Valproic acid activates notch-1 signaling and regulates the neuroendocrine phenotype in carcinoid cancer cells. Oncologist 12, 942-951 Gridley, T. (2004) Kick it up a Notch: NOTCH1 activation in T-ALL. Cancer Cell 6, 431-432 Guidos, C. J. (2002) Notch signaling in lymphocyte development. Semin Immunol 14, 395-404

202

94 95

96

97 98

99

100

101

102

103

104 105

106

Guo, M., Jan, L. Y. and Jan, Y. N. (1996) Control of daughter cell fates during asymmetric division: interaction of Numb and Notch. Neuron 17, 27-41 Hamilton, A., Elrick, L., Myssina, S., Copland, M., Jorgensen, H., Melo, J. V. and Holyoake, T. (2006) BCR-ABL activity and its response to drugs can be determined in CD34+ CML stem cells by CrkL phosphorylation status using flow cytometry. Leukemia 20, 1035-1039 Han, H., Tanigaki, K., Yamamoto, N., Kuroda, K., Yoshimoto, M., Nakahata, T., Ikuta, K. and Honjo, T. (2002) Inducible gene knockout of transcription factor recombination signal binding protein-J reveals its essential role in T versus B lineage decision. Int Immunol 14, 637-645 Han, W., Ye, Q. and Moore, M. A. S. (2000) A soluble form of human Delta-like-1 inhibits differentiation of hematopoietic progenitor cells. Blood 95, 1616-1625 Hao, S. X. and Ren, R. (2000) Expression of interferon consensus sequence binding protein (ICSBP) is downregulated in Bcr-Abl-induced murine chronic myelogenous leukemia-like disease, and forced coexpression of ICSBP inhibits Bcr-Abl-induced myeloproliferative disorder. Mol Cell Biol 20, 1149-1161 He, F., Wang, L., Hu, X. B., Yin, D. D., Zhang, P., Li, G. H., Wang, Y. C., Huang, S. Y., Liang, Y. M. and Han, H. (2008) Notch and BCR signaling synergistically promote the proliferation of Raji B-lymphoma cells. Leuk Res Henning, K., Schroeder, T., Schwanbeck, R., Rieber, N., Bresnick, E. H. and Just, U. (2007) mNotch1 signaling and erythropoietin cooperate in erythroid differentiation of multipotent progenitor cells and upregulate beta-globin. Exp Hematol 35, 1321-1332 Holtz, M. S., Slovak, M. L., Zhang, F., Sawyers, C. L., Forman, S. J. and Bhatia, R. (2002) Imatinib mesylate (STI571) inhibits growth of primitive malignant progenitors in chronic myelogenous leukemia through reversal of abnormally increased proliferation. Blood 99, 37923800 Holyoake, T. L., Jiang, X., Jorgensen, H. G., Graham, S., Alcorn, M. J., Laird, C., Eaves, A. C. and Eaves, C. J. (2001) Primitive quiescent leukemic cells from patients with chronic myeloid leukemia spontaneously initiate factor-independent growth in vitro in association with up-regulation of expression of interleukin-3. Blood 97, 720-728 Honda, H., Oda, H., Suzuki, T., Takahashi, T., Witte, O. N., Ozawa, K., Ishikawa, T., Yazaki, Y. and Hirai, H. (1998) Development of acute lymphoblastic leukemia and myeloproliferative disorder in transgenic mice expressing p210bcr/abl: a novel transgenic model for human Ph1positive leukemias. Blood 91, 2067-2075 Hows, J. (2005) Adult stem cell therapy beyond haemopoietic stem cell transplantation? An update. Transpl Immunol 14, 221-223 Hu, Y., Chen, Y., Douglas, L. and Li, S. (2008) beta-Catenin is essential for survival of leukemic stem cells insensitive to kinase inhibition in mice with BCR-ABL-induced chronic myeloid leukemia. Leukemia Hubmann, R., Schwarzmeier, J. D., Shehata, M., Hilgarth, M., Duechler, M., Dettke, M. and Berger, R. (2002) Notch2 is involved in

203

107 108 109

110 111

112 113

114 115 116

117

118 119 120

the overexpression of CD23 in B-cell chronic lymphocytic leukemia. Blood 99, 3742-3747 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 Inokuchi, K. (2006) Chronic myelogenous leukemia: from molecular biology to clinical aspects and novel targeted therapies. J Nippon Med Sch 73, 178-192 Ishiko, E., Matsumura, I., Ezoe, S., Gale, K., Ishiko, J., Satoh, Y., Tanaka, H., Shibayama, H., Mizuki, M., Era, T., Enver, T. and Kanakura, Y. (2005) Notch Signals Inhibit the Development of Erythroid/Megakaryocytic Cells by Suppressing GATA-1 Activity through the Induction of HES1. J Biol Chem 280, 4929-4939 Iso, T., Kedes, L. and Hamamori, Y. (2003) HES and HERP families: multiple effectors of the Notch signaling pathway. J Cell Physiol 194, 237-255 Izon, D. J., Aster, J. C., He, Y., Weng, A., Karnell, F. G., Patriub, V., Xu, L., Bakkour, S., Rodriguez, C., Allman, D. and Pear, W. S. (2002) Deltex1 redirects lymphoid progenitors to the B cell lineage by antagonizing Notch1. Immunity 16, 231-243 Jacobsen, L. and Hughes, P. (2007) Effects of Passage Number on Cell Line Transfection. Biochemica. 3, 32 Jamieson, C. H., Ailles, L. E., Dylla, S. J., Muijtjens, M., Jones, C., Zehnder, J. L., Gotlib, J., Li, K., Manz, M. G., Keating, A., Sawyers, C. L. and Weissman, I. L. (2004) Granulocyte-macrophage progenitors as candidate leukemic stem cells in blast-crisis CML. N Engl J Med 351, 657-667 Jarriault, S., Brou, C., Logeat, F., Schroeter, E. H., Kopan, R. and Israel, A. (1995) Signalling downstream of activated mammalian Notch. Nature 377, 355-358 Jiang, X., Saw, K. M., Eaves, A. and Eaves, C. (2007a) Instability of BCR-ABL gene in primary and cultured chronic myeloid leukemia stem cells. J Natl Cancer Inst 99, 680-693 Jiang, X., Zhao, Y., Smith, C., Gasparetto, M., Turhan, A., Eaves, A. and Eaves, C. (2007b) Chronic myeloid leukemia stem cells possess multiple unique features of resistance to BCR-ABL targeted therapies. Leukemia 21, 926-935 Jonsson, J. I., Xiang, Z., Pettersson, M., Lardelli, M. and Nilsson, G. (2001) Distinct and regulated expression of Notch receptors in hematopoietic lineages and during myeloid differentiation. Eur J Immunol 31, 3240-3247 Jordan, C. T. and Guzman, M. L. (2004) Mechanisms controlling pathogenesis and survival of leukemic stem cells. Oncogene 23, 71787187 Jorgensen, H. G. and Holyoake, T. L. (2007) Characterization of cancer stem cells in chronic myeloid leukaemia. Biochem Soc Trans 35, 1347-1351 Joutel, A., Dodick, D. D., Parisi, J. E., Cecillon, M., Tournier-Lasserve, E. and Bousser, M. G. (2000) De novo mutation in the Notch3 gene causing CADASIL. Ann Neurol 47, 388-391

204

121

122

123 124 125 126

127 128

129 130 131

132 133 134

135

136

Jundt, F., Anagnostopoulos, I., Forster, R., Mathas, S., Stein, H. and Dorken, B. (2002) Activated Notch1 signaling promotes tumor cell proliferation and survival in Hodgkin and anaplastic large cell lymphoma. Blood 99, 3398-3403 Jundt, F., Probsting, K. S., Anagnostopoulos, I., Muehlinghaus, G., Chatterjee, M., Mathas, S., Bargou, R. C., Manz, R., Stein, H. and Dorken, B. (2004) Jagged1-induced Notch signaling drives proliferation of multiple myeloma cells. Blood 103, 3511-3515 Kabarowski, J. H. and Witte, O. N. (2000) Consequences of BCR-ABL expression within the hematopoietic stem cell in chronic myeloid leukemia. Stem Cells 18, 399-408 Kageyama, R., Ohtsuka, T. and Tomita, K. (2000) The bHLH gene Hes1 regulates differentiation of multiple cell types. Mol Cells 10, 1-7 Kantarjian, H. M., Talpaz, M., Giles, F., O'Brien, S. and Cortes, J. (2006) New insights into the pathophysiology of chronic myeloid leukemia and imatinib resistance. Ann Intern Med 145, 913-923 Karanu, F. N., Murdoch, B., Gallacher, L., Wu, D. M., Koremoto, M., Sakano, S. and Bhatia, M. (2000) The Notch Ligand Jagged-1 Represents a Novel Growth Factor of Human Hematopoietic Stem Cells. J. Exp. Med. 192, 1365-1372 Katoh, M. and Katoh, M. (2007) WNT signaling pathway and stem cell signaling network. Clin Cancer Res 13, 4042-4045 Kerkela, R., Grazette, L., Yacobi, R., Iliescu, C., Patten, R., Beahm, C., Walters, B., Shevtsov, S., Pesant, S., Clubb, F. J., Rosenzweig, A., Salomon, R. N., Van Etten, R. A., Alroy, J., Durand, J. B. and Force, T. (2006) Cardiotoxicity of the cancer therapeutic agent imatinib mesylate. Nat Med 12, 908-916 Kidd, S., Lockett, T. J. and Young, M. W. (1983) The Notch locus of Drosophila melanogaster. Cell 34, 421-433 Kim, H. K. and Siu, G. (1998) The notch pathway intermediate HES-1 silences CD4 gene expression. Mol Cell Biol 18, 7166-7175 Koch, U., Lacombe, T. A., Holland, D., Bowman, J. L., Cohen, B. L., Egan, S. E. and Guidos, C. J. (2001) Subversion of the T/B lineage decision in the thymus by lunatic fringe-mediated inhibition of Notch-1. Immunity 15, 225-236 Kogoshi, H., Sato, T., Koyama, T., Nara, N. and Tohda, S. (2007) Gamma-secretase inhibitors suppress the growth of leukemia and lymphoma cells. Oncol Rep 18, 77-80 Kojika, S. and Griffin, J. D. (2001) Notch receptors and hematopoiesis. Exp Hematol 29, 1041-1052 Kollet, O., Peled, A., Byk, T., Ben-Hur, H., Greiner, D., Shultz, L. and Lapidot, T. (2000) beta2 microglobulin-deficient (B2m(null)) NOD/SCID mice are excellent recipients for studying human stem cell function. Blood 95, 3102-3105 Kondo, M., Wagers, A. J., Manz, M. G., Prohaska, S. S., Scherer, D. C., Beilhack, G. F., Shizuru, J. A. and Weissman, I. L. (2003) Biology of hematopoietic stem cells and progenitors: implications for clinical application. Annu Rev Immunol 21, 759-806 Kosugi, N., Ebihara, Y., Nakahata, T., Saisho, H., Asano, S. and Tojo, A. (2005) CD34+CD7+ leukemic progenitor cells may be involved in

205

137 138 139

140

141 142

143

144 145

146

147

148 149

150

maintenance and clonal evolution of chronic myeloid leukemia. Clin Cancer Res 11, 505-511 Krause, D. S. (2002) Regulation of hematopoietic stem cell fate. Oncogene 21, 3262-3269 Lai, E. C. (2004) Notch signaling: control of cell communication and cell fate. Development 131, 965-973 Lam, L. T., Ronchini, C., Norton, J., Capobianco, A. J. and Bresnick, E. H. (2000) Suppression of erythroid but not megakaryocytic differentiation of human K562 erythroleukemic cells by notch-1. J Biol Chem 275, 19676-19684 Lanctot, P. M., Leclerc, P. C., Clement, M., Auger-Messier, M., Escher, E., Leduc, R. and Guillemette, G. (2005) Importance of N-glycosylation positioning for cell-surface expression, targeting, affinity and quality control of the human AT1 receptor. Biochem J 390, 367-376 Larsson, J. and Karlsson, S. (2005) The role of Smad signaling in hematopoiesis. Oncogene 24, 5676-5692 Lataillade, J. J., Clay, D., David, C., Boutin, L., Guerton, B., Drouet, M., Herodin, F. and Le Bousse-Kerdiles, M. C. (2005) Phenotypic and functional characteristics of CD34+ cells are related to their anatomical environment: is their versatility a prerequisite for their bio-availability? J Leukoc Biol Li, L., Milner, L. A., Deng, Y., Iwata, M., Banta, A., Graf, L., Marcovina, S., Friedman, C., Trask, B. J., Hood, L. and Torok-Storb, B. (1998) The human homolog of rat Jagged1 expressed by marrow stroma inhibits differentiation of 32D cells through interaction with Notch1. Immunity 8, 43-55 Livak, K. J. and Schmittgen, T. D. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402-408 Lundell, B. I., McCarthy, J. B., Kovach, N. L. and Verfaillie, C. M. (1997) Activation of beta1 integrins on CML progenitors reveals cooperation between beta1 integrins and CD44 in the regulation of adhesion and proliferation. Leukemia 11, 822-829 Maciejewski, J., Selleri, C., Anderson, S. and Young, N. S. (1995) Fas antigen expression on CD34+ human marrow cells is induced by interferon gamma and tumor necrosis factor alpha and potentiates cytokine-mediated hematopoietic suppression in vitro. Blood 85, 31833190 Maguer-Satta, V., Petzer, A. L., Eaves, A. C. and Eaves, C. J. (1996) BCR-ABL expression in different subpopulations of functionally characterized Ph+ CD34+ cells from patients with chronic myeloid leukemia. Blood 88, 1796-1804 Maillard, I., Adler, S. H. and Pear, W. S. (2003) Notch and the immune system. Immunity 19, 781-791 Maillard, I., Koch, U., Dumortier, A., Shestova, O., Xu, L., Sai, H., Pross, S. E., Aster, J. C., Bhandoola, A., Radtke, F. and Pear, W. S. (2008) Canonical notch signaling is dispensable for the maintenance of adult hematopoietic stem cells. Cell Stem Cell 2, 356-366 Mancuso, P., Burlini, A., Pruneri, G., Goldhirsch, A., Martinelli, G. and Bertolini, F. (2001) Resting and activated endothelial cells are

206

151 152 153

154

155 156 157 158 159 160

161

162 163 164 165

increased in the peripheral blood of cancer patients. Blood 97, 36583661 Marley, S. B. and Gordon, M. Y. (2005) Chronic myeloid leukaemia: stem cell derived but progenitor cell driven. Clin Sci (Lond) 109, 13-25 Martinez Arias, A., Zecchini, V. and Brennan, K. (2002) CSLindependent Notch signalling: a checkpoint in cell fate decisions during development? Curr Opin Genet Dev 12, 524-533 Matsuno, K., Diederich, R. J., Go, M. J., Blaumueller, C. M. and Artavanis-Tsakonas, S. (1995) Deltex acts as a positive regulator of Notch signaling through interactions with the Notch ankyrin repeats. Development 121, 2633-2644 Mayerhofer, M., Valent, P., Sperr, W. R., Griffin, J. D. and Sillaber, C. (2002) BCR/ABL induces expression of vascular endothelial growth factor and its transcriptional activator, hypoxia inducible factor-1alpha, through a pathway involving phosphoinositide 3-kinase and the mammalian target of rapamycin. Blood 100, 3767-3775 McGill, M. A. and McGlade, C. J. (2003) Mammalian numb proteins promote Notch1 receptor ubiquitination and degradation of the Notch1 intracellular domain. J Biol Chem 278, 23196-23203 Melo, J. (2001) Inviting leukemic cells to waltz with the devil. Nat Med 7, 156-157 Milner, L. A. and Bigas, A. (1999) Notch as a mediator of cell fate determination in hematopoiesis: evidence and speculation. Blood 93, 2431-2448 Milner, L. A., Bigas, A., Kopan, R., Brashem-Stein, C., Bernstein, I. D. and Martin, D. I. (1996) Inhibition of granulocytic differentiation by mNotch1. Proc Natl Acad Sci U S A 93, 13014-13019 Milner, L. A., Kopan, R., Martin, D. I. and Bernstein, I. D. (1994) A human homologue of the Drosophila developmental gene, Notch, is expressed in CD34+ hematopoietic precursors. Blood 83, 2057-2062 Miyazato, A., Ueno, S., Ohmine, K., Ueda, M., Yoshida, K., Yamashita, Y., Kaneko, T., Mori, M., Kirito, K., Toshima, M., Nakamura, Y., Saito, K., Kano, Y., Furusawa, S., Ozawa, K. and Mano, H. (2001) Identification of myelodysplastic syndrome-specific genes by DNA microarray analysis with purified hematopoietic stem cell fraction. Blood 98, 422-427 Mizuno, T., Yamasaki, N., Miyazaki, K., Tazaki, T., Koller, R., Oda, H., Honda, Z. I., Ochi, M., Wolff, L. and Honda, H. (2008) Overexpression/enhanced kinase activity of BCR/ABL and altered expression of Notch1 induced acute leukemia in p210BCR/ABL transgenic mice. Oncogene 27, 3465-3474 Mumm, J. S. and Kopan, R. (2000) Notch signaling: from the outside in. Dev Biol 228, 151-165 Nagar, B. (2007) c-Abl tyrosine kinase and inhibition by the cancer drug imatinib (Gleevec/STI-571). J Nutr 137, 1518S-1523S; discussion 1548S Nam, Y., Aster, J. C. and Blacklow, S. C. (2002) Notch signaling as a therapeutic target. Curr Opin Chem Biol 6, 501-509 Nefedova, Y., Cheng, P., Alsina, M., Dalton, W. S. and Gabrilovich, D. I. (2004) Involvement of Notch-1 signaling in bone marrow stroma-

207

166 167

168 169 170 171

172 173 174

175 176 177 178 179

180

mediated de novo drug resistance of myeloma and other malignant lymphoid cell lines. Blood 103, 3503-3510 Nervi, B., Link, D. C. and DiPersio, J. F. (2006) Cytokines and hematopoietic stem cell mobilization. J Cell Biochem 99, 690-705 Neshat, M. S., Raitano, A. B., Wang, H. G., Reed, J. C. and Sawyers, C. L. (2000) The survival function of the Bcr-Abl oncogene is mediated by Bad-dependent and -independent pathways: roles for phosphatidylinositol 3-kinase and Raf. Mol Cell Biol 20, 1179-1186 Nicolas, M., Wolfer, A., Raj, K., Kummer, J. A., Mill, P., van Noort, M., Hui, C. C., Clevers, H., Dotto, G. P. and Radtke, F. (2003) Notch1 functions as a tumor suppressor in mouse skin. Nat Genet 33, 416-421 Normann, A. P., Egeland, T., Madshus, I. H., Heim, S. and Tjonnfjord, G. E. (2003) CD7 expression by CD34+ cells in CML patients, of prognostic significance? Eur J Haematol 71, 266-275 Nutt, S. L., Eberhard, D., Horcher, M., Rolink, A. G. and Busslinger, M. (2001) Pax5 determines the identity of B cells from the beginning to the end of B-lymphopoiesis. Int Rev Immunol 20, 65-82 Oda, T., Heaney, C., Hagopian, J. R., Okuda, K., Griffin, J. D. and Druker, B. J. (1994) Crkl is the major tyrosine-phosphorylated protein in neutrophils from patients with chronic myelogenous leukemia. J Biol Chem 269, 22925-22928 Oertel, J., Oertel, B., Schleicher, J. and Huhn, D. (1996) Immunotyping of blasts in human bone marrow. Ann Hematol 72, 125-129 Ohishi, K., Katayama, N., Shiku, H., Varnum-Finney, B. and Bernstein, I. D. (2003) Notch signalling in hematopoiesis. Semin Cell Dev Biol 14, 143-150 Ohishi, K., Varnum-Finney, B., Flowers, D., Anasetti, C., Myerson, D. and Bernstein, I. D. (2000) Monocytes express high amounts of Notch and undergo cytokine specific apoptosis following interaction with the Notch ligand, Delta-1. Blood 95, 2847-2854 O'Neil, J., Calvo, J., McKenna, K., Krishnamoorthy, V., Aster, J. C., Bassing, C. H., Alt, F. W., Kelliher, M. and Look, A. T. (2006) Activating Notch1 mutations in mouse models of T-ALL. Blood 107, 781-785 Ostrowska, B., Sacha, T., Szostek, M., Zawada, M., Skotnicki, A., Laidler, P. (2003) The involvement of Abl and Bcr-Abl proteins in Notch signalling pathway in haematopoietic normal and CML CD34+ cells. Oswald, F., Liptay, S., Adler, G. and Schmid, R. M. (1998) NF-kappaB2 is a putative target gene of activated Notch-1 via RBP-Jkappa. Mol Cell Biol 18, 2077-2088 Pallister, C. (1998) Blood : physiology and pathophysiology. Butterworth-Heinemann, Oxford Palomero, T., Sulis, M. L., Cortina, M., Real, P. J., Barnes, K., Ciofani, M., Caparros, E., Buteau, J., Brown, K., Perkins, S. L., Bhagat, G., Agarwal, A. M., Basso, G., Castillo, M., Nagase, S., Cordon-Cardo, C., Parsons, R., Zuniga-Pflucker, J. C., Dominguez, M. and Ferrando, A. A. (2007) Mutational loss of PTEN induces resistance to NOTCH1 inhibition in T-cell leukemia. Nat Med 13, 1203-1210 Panin, V. M. and Irvine, K. D. (1998) Modulators of Notch signaling. Semin Cell Dev Biol 9, 609-617

208

181

Pear, W. S., Aster, J. C., Scott, M. L., Hasserjian, R. P., Soffer, B., Sklar, J. and Baltimore, D. (1996) Exclusive development of T cell neoplasms in mice transplanted with bone marrow expressing activated Notch alleles. J Exp Med 183, 2283-2291 182 Pear, W. S. and Radtke, F. (2003) Notch signaling in lymphopoiesis. Semin Immunol 15, 69-79 183 Petzer, A. L., Eaves, C. J., Lansdorp, P. M., Ponchio, L., Barnett, M. J. and Eaves, A. C. (1996) Characterization of primitive subpopulations of normal and leukemic cells present in the blood of patients with newly diagnosed as well as established chronic myeloid leukemia. Blood 88, 2162-2171 184 Petzer, A. L. and Gunsilius, E. (2003) Hematopoietic stem cells in chronic myeloid leukemia. Arch Med Res 34, 496-506 185 Platta, C. S., Greenblatt, D. Y., Kunnimalaiyaan, M. and Chen, H. (2008) Valproic acid induces Notch1 signaling in small cell lung cancer cells. J Surg Res 148, 31-37 186 Pui, J. C., Allman, D., Xu, L., DeRocco, S., Karnell, F. G., Bakkour, S., Lee, J. Y., Kadesch, T., Hardy, R. R., Aster, J. C. and Pear, W. S. (1999) Notch1 expression in early lymphopoiesis influences B versus T lineage determination. Immunity 11, 299-308 187 Quintas-Cardama, A. and Cortes, J. (2008) Molecular biology of BCRABL1-positive chronic myeloid leukemia. Blood 188 Quintas-Cardama, A., Tong, W., Manshouri, T., Vega, F., Lennon, P. A., Cools, J., Gilliland, D. G., Lee, F., Cortes, J., Kantarjian, H. and Garcia-Manero, G. (2008) Activity of tyrosine kinase inhibitors against human NUP214-ABL1-positive T cell malignancies. Leukemia 22, 1117-1124 189 Radtke, F. and Raj, K. (2003) THE ROLE OF NOTCH IN TUMORIGENESIS: ONCOGENE OR TUMOUR SUPPRESSOR? Nature Reviews Cancer Nat Rev Cancer 3, 756-767 190 Radtke, F., Wilson, A. and MacDonald, H. R. (2004) Notch signaling in T- and B-cell development. Curr Opin Immunol 16, 174-179 191 Radtke, F., Wilson, A., Stark, G., Bauer, M., van Meerwijk, J., MacDonald, H. R. and Aguet, M. (1999) Deficient T cell fate specification in mice with an induced inactivation of Notch1. Immunity 10, 547-558 192 Rao, M. S. and Mattson, M. P. (2001) Stem cells and aging: expanding the possibilities. Mech Ageing Dev 122, 713-734 193 Reizis, B. and Leder, P. (2002) Direct induction of T lymphocytespecific gene expression by the mammalian Notch signaling pathway. Genes Dev 16, 295-300 194 Ren, R. (2005) Mechanisms of BCR-ABL in the pathogenesis of chronic myelogenous leukaemia. Nat Rev Cancer 5, 172-183 195 Reya, T., Duncan, A. W., Ailles, L., Domen, J., Scherer, D. C., Willert, K., Hintz, L., Nusse, R. and Weissman, I. L. (2003) A role for Wnt signalling in self-renewal of haematopoietic stem cells. Nature 423, 409-414

209

196 197

198

199

200

201 202 203

204

205

206

207

208

Rhodes, J., York, R. D., Tara, D., Tajinda, K. and Druker, B. J. (2000) CrkL functions as a nuclear adaptor and transcriptional activator in BcrAbl-expressing cells. Exp Hematol 28, 305-310 Rizo, A., Vellenga, E., de Haan, G. and Schuringa, J. J. (2006) Signaling pathways in self-renewing hematopoietic and leukemic stem cells: do all stem cells need a niche? Hum Mol Genet 15 Spec No 2, R210-219 Robert-Moreno, A., Espinosa, L., Sanchez, M. J., de la Pompa, J. L. and Bigas, A. (2007) The notch pathway positively regulates programmed cell death during erythroid differentiation. Leukemia 21, 1496-1503 Robey, E., Chang, D., Itano, A., Cado, D., Alexander, H., Lans, D., Weinmaster, G. and Salmon, P. (1996) An activated form of Notch influences the choice between CD4 and CD8 T cell lineages. Cell 87, 483-492 Roche-Lestienne, C., Soenen-Cornu, V., Grardel-Duflos, N., Lai, J. L., Philippe, N., Facon, T., Fenaux, P. and Preudhomme, C. (2002) Several types of mutations of the Abl gene can be found in chronic myeloid leukemia patients resistant to STI571, and they can pre-exist to the onset of treatment. Blood 100, 1014-1018 Sade, H., Krishna, S. and Sarin, A. (2004) The anti-apoptotic effect of Notch-1 requires p56lck-dependent, Akt/PKB-mediated signaling in T cells. J Biol Chem 279, 2937-2944 Saglio, G. and Cilloni, D. (2004) Abl: the prototype of oncogenic fusion proteins. Cell Mol Life Sci 61, 2897-2911 Saito, T., Chiba, S., Ichikawa, M., Kunisato, A., Asai, T., Shimizu, K., Yamaguchi, T., Yamamoto, G., Seo, S., Kumano, K., NakagamiYamaguchi, E., Hamada, Y., Aizawa, S. and Hirai, H. (2003) Notch2 is preferentially expressed in mature B cells and indispensable for marginal zone B lineage development. Immunity 18, 675-685 Salesse, S. and Verfaillie, C. M. (2002) Mechanisms underlying abnormal trafficking and expansion of malignant progenitors in CML: BCR/ABL-induced defects in integrin function in CML. Oncogene 21, 8605-8611 Satoh, Y., Matsumura, I., Tanaka, H., Ezoe, S., Sugahara, H., Mizuki, M., Shibayama, H., Ishiko, E., Ishiko, J., Nakajima, K. and Kanakura, Y. (2004) Roles for c-Myc in self-renewal of hematopoietic stem cells. J Biol Chem 279, 24986-24993 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., Gu, H., Griffin, J. D. and Neel, B. G. (2002) Critical role for Gab2 in transformation by BCR/ABL. Cancer Cell 1, 479-492 Sattler, M., Salgia, R., Okuda, K., Uemura, N., Durstin, M. A., Pisick, E., Xu, G., Li, J. L., Prasad, K. V. and Griffin, J. D. (1996) The protooncogene product p120CBL and the adaptor proteins CRKL and cCRK link c-ABL, p190BCR/ABL and p210BCR/ABL to the phosphatidylinositol-3' kinase pathway. Oncogene 12, 839-846 Schroeder, T. and Just, U. (2000) Notch signalling via RBP-J promotes myeloid differentiation. Embo J 19, 2558-2568

210

209

Schroeder, T., Kohlhof, H., Rieber, N. and Just, U. (2003) Notch signaling induces multilineage myeloid differentiation and up-regulates PU.1 expression. J Immunol 170, 5538-5548 210 Schwartz, M. A., Schaller, M. D. and Ginsberg, M. H. (1995) Integrins: emerging paradigms of signal transduction. Annu Rev Cell Dev Biol 11, 549-599 211 Schweisguth, F. (2004) Notch signaling activity. Curr Biol 14, R129-138 212 Screpanti, I., Bellavia, D., Campese, A. F., Frati, L. and Gulino, A. (2003) Notch, a unifying target in T-cell acute lymphoblastic leukemia? Trends Mol Med 9, 30-35 213 Sengupta, A., Banerjee, D., Chandra, S., Banerji, S. K., Ghosh, R., Roy, R. and Banerjee, S. (2007) Deregulation and cross talk among Sonic hedgehog, Wnt, Hox and Notch signaling in chronic myeloid leukemia progression. Leukemia 21, 949-955 214 Shet, A. S., Jahagirdar, B. N. and Verfaillie, C. M. (2002) Chronic myelogenous leukemia: mechanisms underlying disease progression. Leukemia 16, 1402-1411 215 Shizuru, J. A., Negrin, R. S. and Weissman, I. L. (2005) Hematopoietic stem and progenitor cells: clinical and preclinical regeneration of the hematolymphoid system. Annu Rev Med 56, 509-538 216 Siissalo, S., Laitinen, L., Koljonen, M., Vellonen, K. S., Kortejarvi, H., Urtti, A., Hirvonen, J. and Kaukonen, A. M. (2007) Effect of cell differentiation and passage number on the expression of efflux proteins in wild type and vinblastine-induced Caco-2 cell lines. Eur J Pharm Biopharm 67, 548-554 217 Simpson, P. (1998) Introduction: Notch signalling and choice of cell fates in development. Semin Cell Dev Biol 9, 581-582 218 Singer, C. F., Hudelist, G., Lamm, W., Mueller, R., Handl, C., Kubista, E. and Czerwenka, K. (2006) Active (p)CrkL is overexpressed in human malignancies: potential role as a surrogate parameter for therapeutic tyrosine kinase inhibition. Oncol Rep 15, 353-359 219 Singh, N., Phillips, R. A., Iscove, N. N. and Egan, S. E. (2000) Expression of notch receptors, notch ligands, and fringe genes in hematopoiesis. Exp Hematol 28, 527-534 220 Smith, C. (2003) Hematopoietic stem cells and hematopoiesis. Cancer Control 10, 9-16 221 Smith, D. L., Burthem, J. and Whetton, A. D. (2003) Molecular pathogenesis of chronic myeloid leukaemia. Expert Rev Mol Med 5, 127 222 Staal, F. J. T. and Clevers, H. C. (2005) WNT SIGNALLING AND HAEMATOPOIESIS: A WNT-WNT SITUATION. Nature Reviews Immunology Nat Rev Immunol 5, 21-30 223 Steidl, U., Kronenwett, R., Martin, S. and Haas, R. (2003) Molecular biology of hematopoietic stem cells. Vitam Horm 66, 1-28 224 Stein, M. I., Zhu, J. and Emerson, S. G. (2004) Molecular pathways regulating the self-renewal of hematopoietic stem cells. Exp Hematol 32, 1129-1136 225 Stier, S., Cheng, T., Dombkowski, D., Carlesso, N. and Scadden, D. T. (2002) Notch1 activation increases hematopoietic stem cell self-

211

226

227 228 229

230

231

232 233 234

235

236

237

238

renewal in vivo and favors lymphoid over myeloid lineage outcome. Blood 99, 2369-2378 Stockhausen, M. T., Sjolund, J., Manetopoulos, C. and Axelson, H. (2005) Effects of the histone deacetylase inhibitor valproic acid on Notch signalling in human neuroblastoma cells. Br J Cancer 92, 751759 Sun, X. and Artavanis-Tsakonas, S. (1997) Secreted forms of DELTA and SERRATE define antagonists of Notch signaling in Drosophila. Development 124, 3439-3448 Szilvassy, S. J. (2003) The biology of hematopoietic stem cells. Arch Med Res 34, 446-460 Takahashi, N., Miura, I., Saitoh, K. and Miura, A. B. (1998) Lineage involvement of stem cells bearing the philadelphia chromosome in chronic myeloid leukemia in the chronic phase as shown by a combination of fluorescence-activated cell sorting and fluorescence in situ hybridization. Blood 92, 4758-4763 Talora, C., Sgroi, D. C., Crum, C. P. and Dotto, G. P. (2002) Specific down-modulation of Notch1 signaling in cervical cancer cells is required for sustained HPV-E6/E7 expression and late steps of malignant transformation. Genes Dev 16, 2252-2263 Tanigaki, K., Han, H., Yamamoto, N., Tashiro, K., Ikegawa, M., Kuroda, K., Suzuki, A., Nakano, T. and Honjo, T. (2002) Notch-RBP-J signaling is involved in cell fate determination of marginal zone B cells. Nat Immunol 3, 443-450 Tax, F. E., Yeargers, J. J. and Thomas, J. H. (1994) Sequence of C. elegans lag-2 reveals a cell-signalling domain shared with Delta and Serrate of Drosophila. Nature 368, 150-154 ten Hoeve, J., Arlinghaus, R. B., Guo, J. Q., Heisterkamp, N. and Groffen, J. (1994) Tyrosine phosphorylation of CRKL in Philadelphia+ leukemia. Blood 84, 1731-1736 Tian, G., Ghanekar, S. V., Aharony, D., Shenvi, A. B., Jacobs, R. T., Liu, X. and Greenberg, B. D. (2003) The mechanism of gammasecretase: multiple inhibitor binding sites for transition state analogs and small molecule inhibitors. J Biol Chem 278, 28968-28975 Tickenbrock, L., Hehn, S., Sargin, B., Evers, G., Ng, P. R., Choudhary, C., Berdel, W. E., Muller-Tidow, C. and Serve, H. (2008) Activation of Wnt signaling in cKit-ITD mediated transformation and imatinib sensitivity in acute myeloid leukemia. Int J Hematol 88, 174-180 Tohda, S., Murata-Ohsawa, M., Sakano, S. and Nara, N. (2003) Notch ligands, Delta-1 and Delta-4 suppress the self-renewal capacity and long-term growth of two myeloblastic leukemia cell lines. Int J Oncol 22, 1073-1079 Tsai, S., Fero, J. and Bartelmez, S. (2000) Mouse Jagged2 is differentially expressed in hematopoietic progenitors and endothelial cells and promotes the survival and proliferation of hematopoietic progenitors by direct cell-to-cell contact. Blood 96, 950-957 Uyttendaele, H., Marazzi, G., Wu, G., Yan, Q., Sassoon, D. and Kitajewski, J. (1996) Notch4/int-3, a mammary proto-oncogene, is an endothelial cell-specific mammalian Notch gene. Development 122, 2251-2259

212

239

240

241

242 243

244

245 246

247 248 249

250

251 252

Varnum-Finney, B., Xu, L., Brashem-Stein, C., Nourigat, C., Flowers, D., Bakkour, S., Pear, W. S. and Bernstein, I. D. (2000) Pluripotent, cytokine-dependent, hematopoietic stem cells are immortalized by constitutive Notch1 signaling. Nat Med 6, 1278-1281 Vercauteren, S. M. and Sutherland, H. J. (2004) Constitutively active Notch4 promotes early human hematopoietic progenitor cell maintenance while inhibiting differentiation and causes lymphoid abnormalities in vivo. Blood 104, 2315-2322 Verfaillie, C. M., Miller, W. J., Boylan, K. and McGlave, P. B. (1992) Selection of benign primitive hematopoietic progenitors in chronic myelogenous leukemia on the basis of HLA-DR antigen expression. Blood 79, 1003-1010 Visnjic, D., Kalajzic, Z., Rowe, D. W., Katavic, V., Lorenzo, J. and Aguila, H. L. (2004) Hematopoiesis is severely altered in mice with an induced osteoblast deficiency. Blood 103, 3258-3264 Walker, L., Carlson, A., Tan-Pertel, H. T., Weinmaster, G. and Gasson, J. (2001) The notch receptor and its ligands are selectively expressed during hematopoietic development in the mouse. Stem Cells 19, 543552 Walker, L., Lynch, M., Silverman, S., Fraser, J., Boulter, J., Weinmaster, G. and Gasson, J. C. (1999) The Notch/Jagged pathway inhibits proliferation of human hematopoietic progenitors in vitro. Stem Cells 17, 162-171 Washburn, T., Schweighoffer, E., Gridley, T., Chang, D., Fowlkes, B. J., Cado, D. and Robey, E. (1997) Notch activity influences the alphabeta versus gammadelta T cell lineage decision. Cell 88, 833-843 Weijzen, S., Rizzo, P., Braid, M., Vaishnav, R., Jonkheer, S. M., Zlobin, A., Osborne, B. A., Gottipati, S., Aster, J. C., Hahn, W. C., Rudolf, M., Siziopikou, K., Kast, W. M. and Miele, L. (2002) Activation of Notch-1 signaling maintains the neoplastic phenotype in human Rastransformed cells. Nat Med 8, 979-986 Weijzen, S., Zlobin, A., Braid, M., Miele, L. and Kast, W. M. (2003) HPV16 E6 and E7 oncoproteins regulate Notch-1 expression and cooperate to induce transformation. J Cell Physiol 194, 356-362 Weng, A. P. and Aster, J. C. (2004) Multiple niches for Notch in cancer: context is everything. Curr Opin Genet Dev 14, 48-54 Weng, A. P., Ferrando, A. A., Lee, W., Morris, J. P. t., Silverman, L. B., Sanchez-Irizarry, C., Blacklow, S. C., Look, A. T. and Aster, J. C. (2004) Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science 306, 269-271 Wetzel, R., Goss, V. L., Norris, B., Popova, L., Melnick, M. and Smith, B. L. (2005) Evaluation of CML model cell lines and imatinib mesylate response: determinants of signaling profiles. J Immunol Methods 305, 59-66 Wickremasinghe, R. G. and Hoffbrand, A. V. (1999) Biochemical and Genetic Control of Apoptosis: Relevance to Normal Hematopoiesis and Hematological Malignancies. Blood 93, 3587-3600 Wilson, A., MacDonald, H. R. and Radtke, F. (2001) Notch 1-deficient common lymphoid precursors adopt a B cell fate in the thymus. J Exp Med 194, 1003-1012

213

253

254

255 256

257

258 259 260

261

262 263 264

265

Wolfer, A., Bakker, T., Wilson, A., Nicolas, M., Ioannidis, V., Littman, D. R., Lee, P. P., Wilson, C. B., Held, W., MacDonald, H. R. and Radtke, F. (2001) Inactivation of Notch 1 in immature thymocytes does not perturb CD4 or CD8T cell development. Nat Immunol 2, 235-241 Wolfer, A., Wilson, A., Nemir, M., MacDonald, H. R. and Radtke, F. (2002) Inactivation of Notch1 impairs VDJbeta rearrangement and allows pre-TCR-independent survival of early alpha beta Lineage Thymocytes. Immunity 16, 869-879 Wu, L., Sun, T., Kobayashi, K., Gao, P. and Griffin, J. D. (2002) Identification of a family of mastermind-like transcriptional coactivators for mammalian notch receptors. Mol Cell Biol 22, 7688-7700 Yamamoto, N., Yamamoto, S., Inagaki, F., Kawaichi, M., Fukamizu, A., Kishi, N., Matsuno, K., Nakamura, K., Weinmaster, G., Okano, H. and Nakafuku, M. (2001) Role of Deltex-1 as a transcriptional regulator downstream of the Notch receptor. J Biol Chem 276, 45031-45040 Ye, Q., Shieh, J. H., Morrone, G. and Moore, M. A. (2004) Expression of constitutively active Notch4 (Int-3) modulates myeloid proliferation and differentiation and promotes expansion of hematopoietic progenitors. Leukemia 18, 777-787 Yin, D. D., Fan, F. Y., Hu, X. B., Hou, L. H., Zhang, X. P., Liu, L., Liang, Y. M. and Han, H. (2008) Notch signaling inhibits the growth of the human chronic myeloid leukemia cell line K562. Leuk Res Yu, H., Cook, T. J. and Sinko, P. J. (1997) Evidence for diminished functional expression of intestinal transporters in Caco-2 cell monolayers at high passages. Pharm Res 14, 757-762 Zanjani, E. D., Almeida-Porada, G., Livingston, A. G., Flake, A. W. and Ogawa, M. (1998) Human bone marrow CD34- cells engraft in vivo and undergo multilineage expression that includes giving rise to CD34+ cells. Exp Hematol 26, 353-360 Zhang, J., Niu, C., Ye, L., Huang, H., He, X., Tong, W. G., Ross, J., Haug, J., Johnson, T., Feng, J. Q., Harris, S., Wiedemann, L. M., Mishina, Y. and Li, L. (2003) Identification of the haematopoietic stem cell niche and control of the niche size. Nature 425, 836-841 Zhang, P., Yang, Y., Zweidler-McKay, P. A. and Hughes, D. P. (2008) Critical role of notch signaling in osteosarcoma invasion and metastasis. Clin Cancer Res 14, 2962-2969 Zhao, C., Blum, J., Chen, A., Kwon, H. Y., Jung, S. H., Cook, J. M., Lagoo, A. and Reya, T. (2007) Loss of beta-catenin impairs the renewal of normal and CML stem cells in vivo. Cancer Cell 12, 528-541 Zhao, S., Zoller, K., Masuko, M., Rojnuckarin, P., Yang, X. O., Parganas, E., Kaushansky, K., Ihle, J. N., Papayannopoulou, T., Willerford, D. M., Clackson, T. and Blau, C. A. (2002) JAK2, complemented by a second signal from c-kit or flt-3, triggers extensive self-renewal of primary multipotential hemopoietic cells. Embo J 21, 2159-2167 Zweidler-McKay, P. A., He, Y., Xu, L., Rodriguez, C. G., Karnell, F. G., Carpenter, A. C., Aster, J. C., Allman, D. and Pear, W. S. (2005) Notch signaling is a potent inducer of growth arrest and apoptosis in a wide range of B-cell malignancies. Blood 106, 3898-3906

214

266

Zweidler-McKay, P. A. and Pear, W. S. (2004) Notch and T cell malignancy. Semin Cancer Biol 14, 329-340

Appendex1

AAppendex1. Immunoreactivity of ECN1 vs ICN1 antibodies . A-Total HEK293 cell lysates were separated with a 8% SDS-PAGE, transferred to nitrocellulose and probed with: (i) EA1 antibody which detects extracellular Notch1 (ECN1 ) and (ii) b-tan20 which detects intracellular Notch1 (ICN1, Iowa). Lanes (1) untransfected, (2&3) full-length hN1 transfected and (4) ICN transfected cells. (1&2) 15µ g (3&4) 40µ g of cell lysates. The arrows indicate the full-length form (~300 kDa), the ECN1 (~180 kDa) and the ICN1(~120 kDa).

i

i

B- HEK293 cells were transfected with full-length N1 and stained with ECN1 and ICN1 antibodies.

kDa

215

1 2 3 4

1

216