@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Medicine, Faculty of"@en, "Biochemistry and Molecular Biology, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Ho, Maria Ming Chee"@en ; dcterms:issued "2011-01-27T18:38:21Z"@en, "2007"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Multidrug Resistance (MDR), resistance to multiple chemotherapeutic drugs, is a major problem in the treatment of acute myeloid leukemia (AML) . Overexpression of members of the ATPBinding- Cassette (ABC) transporter superfamily has been associated with clinical MDR and failure of conventional chemotherapy. The work in this thesis was the first in investigating expression of ABC transporters and functional effects of their modulation in AML subpopulations along the leukemic stem cell hierarchy: CD34+CD38- (primitive and disease maintaining), CD34+CD38+ (differentiating progenitors), and CD34- (depleted of progenitors). An initial profiling of mRNA expression of the 47 human ABC transporters in total de novo blasts by RT Real-Time PCR showed no consistent differences between patients who subsequently achieved complete remission following conventional remission induction chemotherapy (responders) and patients who remained refractory (non-responders). Subsequent profiling of isolated subpopulations, however, revealed elevated expression of MDR1 and/or BCRP1, two main drug-resistance ABC transporters, in the primitive CD34+CD38- fraction of 7/10 non-responders compared to 0/7 responders. To test their functional activity ex vivo, daunorubicin sensitivity with or without ABC modulators was determined in AML subpopulations by the apoptotic assay. I found high ABC-dependent drug resistance, correlated to high MDR1IBCRP1 expression level and reversible by ABC inhibition, in the CD34+CD38- fraction of non-responders compared to responders. This suggests an active functional role of ABC transporters in the primitive, disease-maintaining fraction. Taken as a whole, my studies suggest a prognostic significance of ABC transporters in the primitive CD34+CD38- leukemic subpopulation, and support a modified approach in investigating the value of ABC modulating agents in AML. It may be possible to pre-screen and identify patients for whom ABC transporters is a major factor for MDR before initial treatment, who are most likely to benefit from the combination of conventional chemotherapy and ABC inhibitors. This will be invaluable especially to patients with a normal karyotype (50% of patients), since cytogenetic aberrations currently remain the most useful prognostic marker for AML."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/30890?expand=metadata"@en ; skos:note "ABC TRANSPORTERS AS PREDICTIVE FACTORS FOR CHEMOTHERAPEUTIC RESPONSE IN ACUTE MYELOID LEUKEMIA by Maria Ming Chee Ho B.Sc., The University of British Columbia, 2001 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Biochemistry and Molecular Biology) The University of British Columbia A p r i l 2007 Maria M i n g Chee Ho, 2007 A B S T R A C T Multidrug Resistance ( M D R ) , resistance to multiple chemotherapeutic drugs, is a major problem in the treatment of acute myeloid leukemia ( A M L ) . Overexpression of members of the A T P -Binding-Cassette ( A B C ) transporter superfamily has been associated with clinical M D R and failure of conventional chemotherapy. The work in this thesis was the first in investigating expression of A B C transporters and functional effects of their modulation in A M L subpopulations along the leukemic stem cell hierarchy: CD34+CD38- (primitive and disease maintaining), CD34+CD38+ (differentiating progenitors), and CD34- (depleted of progenitors). A n initial profiling of m R N A expression of the 47 human A B C transporters in total de novo blasts by R T Real-Time P C R showed no consistent differences between patients who subsequently achieved complete remission following conventional remission induction chemotherapy (responders) and patients who remained refractory (non-responders). Subsequent profiling of isolated subpopulations, however, revealed elevated expression of MDR1 and/or BCRP1, two main drug-resistance A B C transporters, in the primitive CD34+CD38- fraction of 7/10 non-responders compared to 0/7 responders. To test their functional activity ex vivo, daunorubicin sensitivity with or without A B C modulators was determined in A M L subpopulations by the apoptotic assay. I found high ABC-dependent drug resistance, correlated to high MDR1IBCRP1 expression level and reversible by A B C inhibition, in the CD34+CD38-fraction of non-responders compared to responders. This suggests an active functional role of A B C transporters in the primitive, disease-maintaining fraction. Taken as a whole, my studies suggest a prognostic significance of A B C transporters in the primitive CD34+CD38- leukemic subpopulation, and support a modified approach in investigating the value of A B C modulating agents in A M L . It may be possible to pre-screen and identify patients for whom A B C transporters is a major factor for M D R before initial treatment, who are most likely to benefit from the combination of conventional chemotherapy and A B C inhibitors. This w i l l be invaluable especially to patients with a normal karyotype (50% of patients), since cytogenetic aberrations currently remain the most useful prognostic marker for A M L . TABLE OF CONTENTS Abstract.... ii Table of contents •• .....iii List of tables v List of figures •••••• • v i List of abbreviations ....viii Acknowledgements x Dedication xi I Introduction... 1 1.1 - Challenges in cancer therapy ...1 1.2 - Acute Myeloid Leukemia: an overview 3 1.3 - Current prognostic factors for predicting chemotherapeutic response in A M L 5 1.4 - ABC transporters: an overview 7 1.5 - ABC transporters and multidrug resistance in cancer 9 1.6 - ABC transporters in A M L . 12 1.7 - The leukemic stem cell model and its implications in drug response 13 1.8 - ABC transporters in normal and leukemic stem cells 15 1.9 - Thesis objectives 18 II mRNA expression profiling of the ABC transporter superfamily in unfractionated AML patient samples. ..35 2.1 - Introduction 35 2.2 - Materials and Methods 37 2.2.1 - Patient samples, cell lines and culture 37 2.2.2 - RT-Real Time P C R assay. 38 2.2.2.1. Overview 38 2.2.2.2. RNA isolation, DNase treatment and Reverse Transcription 39 2.2.2.3. Primer design and optimization 39 2.2.2.4. Real-Time PCR 39 2.2.2.5. Generation of standard curve 40 2.2.2.6. Data analysis 40 2.2.2.7. Statistical analysis 40 2.3-Results 41 2.3.1 - Profiling of A B C transporters in the drug-sensitive leukemic cell line C E M and its vinblastine-selected, drug-resistant subline CVT.0 41 2.3.2 - Lack of consistent differences was observed in A B C transporter expression between responsive and non-responsive patients ....42 2.4 - Discussion 44 III Expression profiling of drug resistance-related transporters in FACS-sorted AML subpopulations 53 3.1 - Introduction 53 3.2 - Materials and Methods 55 in 3.2.1 - Flow cytometric sorting of A M L subpopulations ..55 3.2.2 - R N A isolation, DNase treatment and RT-Real T ime-PCR 55 3.3-Results 57 3.3.1 - Profiling of selected drug resistance-related transporters in FACS-sorted subpopulations of A M L patient samples 57 3.3.2 - Higher expression oiMDRl and BCRP1 in the CD34+CD38-cells from non-responders ..57 3.4 - Discussion ......59 IV Ex vivo drug sensitivity of primitive and mature subpopulations of Acute Myeloid Leukemia and effects of ABC transporter modulation.68 4.1 - Introduction 68 4.2 - Materials and Methods... 70 4.2.1 - Exposure of A M L cells to drugs 70 4.2.2 - Annexin V-Propidium Iodide assay .70 4.2.3 - M . T S assay..... 71 4.2.4 - Analysis 71 4.3-Results..... 73 4.3.1 - Comparison of the Annexin V - P I apoptotic assay to the M T S proliferation assay on C E M and C V 1 . 0 cell lines 73 4.3.2 - Adaptation of the apoptotic assay to A M L patient cells 74 4.3.3 - Subpopulation size and patient material availability as a source of limitation 75 4.3.4 - Higher daunorubicin resistance and larger effect of A B C modulation in the CD34+CD38- fraction of non-responders 75 4.4 - Discussion 78 V Conclusion and future prospects 93 5.1 - Overall discussion and conclusion 93 5.2 - Future Prospects.... 98 Bibliography 105 Appendix: Drug sensitivity curves of CR and NR patients.. 115 IV LIST OF TABLES Table 1.1: FAB classification of A M L . 19 Table 1.2: WHO classification of A M L 19 Table 1.3: Frequencies of common recurring cytogenetic aberrations in adult A M L 20 Table 1.4: Prognostic significance of frequent chromosomal abnormalities in A M L 21 Table 1.5: List of human ABC transporters, location and physiological function 22 Table 1.6: Substrate specificity of PGP, MRP1 and BCRP1 23 Table 2.1: Patient characteristics 46 Table 4.1: % of CD34+CD38-, CD34+CD38+ & CD34- fractions in A M L patients. 81 v LIST OF FIGURES Figure 1.1: Morphological differences between different A M L subtypes.......... 24 Figure 1.2: A M L translocation (8;21) (left) compared to normal chromosomes 8 and 21 (right) 25 Figure 1.3: A M L translocation (15;17) detected in APL ........25 Figure 1.4: Overall survival of A M L patients with favorable (A) or adverse (B) cytogenetic abnormalities compared to the group with normal karyotype.... 26 Figure 1.5: Schematic diagram of a typical ABC transporter 27 Figure 1.6: X-ray crystallography structure of MsbA 28 Figure 1.7: Proposed model for lipid A transport by MsbA 29 Figure 1.8: PGP as a classic drug pump in cancer cells 30 Figure 1.9: Prognostic significance of PGP in A M L 31 Figure 1.10: Normal human hematopoiesis 32 Figure 1.11: A M L forms a stem cell hierarchy 33 Figure 1.12: Targeting leukemic stem cells as a curative therapy.... 34 Figure 2.1: Flow-chart of RT-Real Time-PCR 47 Figure 2.2: Typical dissociation curve and amplification plot of a Real Time PCR product 48 Figure 2.3: Standard curves for selected genes 49 Figure 2.4: Profiling of the ABC transporter superfamily in C E M and CV1.0 50 Figure 2.5: Profiling of the ABC transporter superfamily in patients CR#7 and NR#9 51 Figure 2.6: mRNA levels of selected ABCs in unfractionated A M L patient samples 52 Figure 3.1: FACS analysis of A M L patient sample NR#9 62 Figure 3.2: mRNA expression levels of MDR1 in FACS-sorted A M L subpopulations 63 Figure 3.3: mRNA expression levels oiMRPl in FACS-sorted A M L subpopulations 64 v i Figure 3.4: mRNA expression levels of BCRP1 in FACS-sorted A M L subpopulations 65 Figure 3.5: Comparison of expression oiMDRl, BCRP1 and MRP1 between the CR and NR groups in the CD34+CD38- fraction 66 Figure 3.6: CD34+CD38- fraction size in CR and NR A M L patients .....67 Figure 4.1: Apoptosis as detected by the Annexin V-PI assay 82 Figure 4.2: Schematic representation of conversion of MTS to formazan 83 Figure 4.3: Effects of PSC-833 and verapamil on daunorubicin sensitivity in C E M and CV1.0 as measured by the Annexin V-PI assay 84 Figure 4.4: Effects of PSC-833 and verapamil on daunorubicin sensitivity in C E M and CV1.0 as measured by the MTS assay 85 Figure 4.5: Toxicity assay of PSC-833 and verapamil on A M L patient cells 86 Figure 4.6: Effect of Pgp inhibition on daunorubicin sensitivity of A M L patient #18 87 Figure 4.7: Effect of Pgp inhibition on daunorubicin sensitivity of A M L patient #25 .88 Figure 4.8: Drug sensitivity and effects of ABC modulation on CR patient #1 subpopulations 89 Figure 4.9: Drug sensitivity and effects of ABC modulation on NR patient #9 subpopulations 90 Figure 4.10: Daunorubicin sensitivity of different A M L subpopulations in CR and NR patients . 91 Figure 4.11: Effects of ABC modulation on drug sensitivity in different A M L subpopulations ..92 Figure 5.1: Models of tumor drug resistance 101 Figure 5.2: Predicting response and overcoming MDR 102 Figure 5.3. Principles of array comparative genomic hybridization.. 103 Figure 5.4. C G H karyogram pf patient NR#3 104 v i i LIST OF ABBREVIATIONS A B C ATP-binding cassette A B C R ATP-binding cassette transporter-retina A E amplification efficiency A M L acute myeloid leukemia A P L acute promyelocyte leukemia Ara-C arabinoside B A A L C Brain and Acute Leukemia Cytoplasmic B A C bacterial artificial chromosome BCRP1 Breast cancer resistance protein 1 bp base pair(s) B S E P Bile-salt export pump also known as SPGP C B F Core-binding Factor C D Cluster of Differentiation molecule C E B P A C C A A T enhancer binding protein C F T R Cystic Fibrosis Transmembrane conductance Regulator C G H comparative genomic hybridization C R complete remission C S C cancer stem cells Ct threshold cycle C y cyanine D M S O dimethylsulfoxide D N A deoxy ribonucleic acid F A B French-American-British F A C S fluorescence-activated cell sorting F B S fetal bovine serum F C S fetal calf serum F I S H fluorescence in situ hybridization Flt3 FMS- l ike tyrosine 3 G A P D H Glyseraldehyde-3-phosphate dehydrogenase G - C S F granulocytic-colony stimulating factor G S H glutathione H S C hematopoietic stem cell I M D M Iscove's modified Dulbecco's medium ITD internal tandem duplication L S C leukemic stem cell M D R multidrug resistance MDR1 Multidrug Resistance 1 M E M minimal essential medium M L L M i x e d Lineage Leukemia MRP1 Multidrug Resistance protein 1 M T S tetrazolium salt N B D nucleotide-binding domain N O D - S C I D non-obese diabetic severe-combined immunodeficiency N R non-responsive P B peripheral blood P C R polymerase chain reaction PgP Permeability-glycoprotein PI propidium iodide P M L promyelocytic leukemia PS phosphatidylserine R A R a Retinoic A c i d Receptor a R N A ribonucleic acid R T reverse transcription Shh Sonic Hedgehog S K Y spectral karyotyping SL- IC severe-combined immunodeficiency leukemia initiating cell SP side population SPGP Sister Permeability-glycoprotein also known as B S E P V L B vinblastine ix ACKNOWLEDGEMENTS I thank my supervisor, Dr. Victor L ing , for giving me the opportunity to work in his laboratory. The guidance, challenge, resources and autonomy he provides are all pivotal in shaping me into an independent thinker. To Dr. Donna Hogge, my collaborator: her support and expert advice are critical to the successful completion of this project. Special thanks to Dr. Jaclyn Hung for constant encouragement and fruitful intellectual discussions. To my fellow lab members and friends, it was a pleasure and privilege working with them. I shall remember this special period of time in my life. I thank my committee members, Dr. Connie Eaves, Dr. Donna Hogge, and Dr. Ross MacGil l ivray, for their help and suggestions on my thesis work. I also thank Gitte Gerhard and Leman Yalcintepe for their technical assistance. Finally, I thank my parents, my sister and brother for their unfailing love and support throughout the years. x To my father, who spurred me on the pursuit of knowledge. To my Father from Above: without Him I am nothing. I Introduction 1.1 - Challenges in cancer therapy Cancer has become the current leading cause of death for people under age 85 in North o America . It is a genetic disease, arising from the transformation of a normal cell, with a derangement in normal regulation of cell proliferation, differentiation and death. One major difficulty in the treatment of cancer stems from its heterogeneity: individual cancer patients typically show different combinations of genetic and/or epigenetic aberrations in specific cellular pathways. This is thought to explain, at least in part, why there is no form of therapy that is equally successful in all patients presented with the same type of cancer. Recognition of this heterogeneity has resulted in increasing support for the concept of \"personalized therapy\" or \"individual therapy\" - tailoring the treatment according to individual patient condition, in particular, therapeutically targeting specific genetic abnormalities for the patient in question - as emphasized in the 2006 annual report from American Society of Clinical Oncology 9 . There are three broadly used types of treatment for cancer: surgery, radiation therapy and chemotherapy. While the former two are very effective on local tumors, chemotherapy remains the main form of systemic treatment for inoperable, metastasized or more advanced cancer. Adverse side effects often ensue, however, due to drug toxicity on normal cells in the body, particularly those that must divide to maintain organ integrity. It would therefore be invaluable i f an individual patient's response to a given regimen could be predicted before deciding on the best form of therapy available. Recent years has seen the development of hypothesis-driven, mechanism-based drug discovery in support of personalized therapy - designing drugs specifically targeting the molecular pathology underlying individual cancers (reviewed by 1 Collins and Workman, 2006 1 0). This type of molecular-targeted therapy is best combined with characterization and pre-screening of patients with the relevant genetic alteration. One prominent example is the use of Gleevac (imatinib, STI571), a specific inhibitor of B C R - A B L tyrosine kinase, in chronic myeloid leukemia ( C M L ) \" ' 1 2 . B C R - A B L is the product of the aberrant Philadelphia chromosome and is present in virtually all cases of the disease. This prototypical specific targeting of a molecular product in cancer achieves high success in newly diagnosed C M L patients, although cases of resistance undermines its effectiveness in more advanced cases (reviewed by Kantarjian, 2006 1 3). In Acute Myelo id Leukemia ( A M L ) , the internal tandem duplication (ITD) of the FLT3 gene is the most common mutation and hence an attractive target for therapy. Flt3 is a tyrosine kinase receptor important in regulation of proliferation, differentiation and apoptosis of hematopoietic progenitors 1 4. Several studies reported that karyotypically normal patients with F L T 3 activation have a poorer prognosis 1 5\" 1 7, and a number of agents against the FLT3 mutation were being tested on patients bearing the mutation 1 8 ' 1 9 . This thesis sought to evaluate the possible prognostic value of A B C transporter expression in anticipating the chemotherapeutic response of the malignant population in patients with A M L . This type of cancer offers a number of unique features relevant to such a study. Technically, it is possible to select patients with over 90% leukemic cell counts in their blood system, so that results are not skewed by a substantial or undefined normal cell contamination, a common problem in solid tumors. Since chemotherapy is the main form of treatment for leukemia, almost all patients w i l l undergo a standard initial drug regimen, thereby allowing direct comparisons with treatment outcome. Furthermore, critical cancer biology models are well established for A M L , notably the origin and perpetuation of the leukemia by a rare subset of \"leukemic stem cells\" of known phenotype 2 ' 2 0 (Discussed in Section 1.7). This forms the basis of 2 my main hypothesis that it is the rare leukemic stem cells, not the mature or differentiated cells comprising the majority of leukemia, which dictate clinical outcome. This is supported by results of this study, which highlights the value of determining A B C transporter levels in a leukemic stem cell-enriched population to predict treatment response. I then went on to illustrate the effects of A B C inhibitors on this small cell fraction, proposing their use as drug-sensitizing agents in combination with chemotherapy to improve initial treatment outcome for otherwise non-responding patients. This is especially crucial for A M L , in which its fast progression augments the importance of choosing the best treatment at diagnosis. 1.2 - Acute Myeloid Leukemia: an overview Leukemias make up ~ 2 % of adult cancers 2 1 but comprise a heterogeneous group of diseases. The lymphoid leukemias affect the lymphoid lineages (notably B - and T-cells) and the myeloid leukemias affect primarily the myeloid lineages including the granulocytes and monocytes/macrophages, red blood cells, and megakaryocytes. Leukemia can be classified as either acute or chronic. A M L is a malignancy of the myeloid elements, the hallmark being a block in normal differentiation, leading to the massive accumulation of immature leukemic \"blast cells\". This usually results in rapid and severe disruption of normal bone marrow function, which can take the clinical form of anemia (decrease in hemoglobin), fever and infection (decrease in white cells), and bleeding and bruising (decrease in platelets). Accumulation of leukemic cells in other tissues is also common, such as the lymph nodes, spleen and skin. A M L requires urgent diagnosis and treatment. If left untreated, the disease results in death within weeks or days. In contrast, the chronic leukemias are characterized by unregulated proliferation 3 and overexpansion of a range of differentiated cells, are slow-growing and usually progress over 2 1 ' a period of several years . The incidence of A M L increases sharply with age, from less than 1/100,000 under age 35 to 15/100,000 over 75. Recent Canadian statistics report an incidence of 816 new cases in 2001 and a mortality rate of 690 deaths in 2003 8 . While the etiology of A M L remains largely unknown, a number of risk factors have been identified, including radiation and/or chemical exposure, tobacco use, prior cancer-related chemotherapy or radiation therapy, old age, genetic syndromes, and a history of prior blood disorders. Certain recurring cytogenetic abnormalities have been found to be closely associated with A M L and form a major area of investigation (see Section 1.3). Diagnosis is based on morphological data which further classify A M L into eight subgroups (MO-7) under the French-American-British ( F A B ) scheme established in 1976 (Table 1.1). A n example of distinct morphological and histochemical differences between A M L subtypes is shown in Figure 1.1. More recently the World Health Organization (WHO) has proposed a new classification which includes immunophenotyping, cytogenetics and clinical features to allow a more prognostically relevant description of A M L (Table 1.2). Given that A M L is a rapidly progressing disease with a fatal outcome i f not adequately controlled, initial treatment is often targeted at eradicating leukemic blasts and re-establishing normal bone marrow function. This is usually achieved through high-dose chemotherapy with general supportive measures such as blood cell transfusions, antibiotics administration, and leukapheresis to temporarily clear patient blood of blasts. The standard remission-induction chemotherapeutic regimen for A M L is a combination of cytosine arabinoside (Ara-C) and an anthracycline such as daunorubicin. Clinical M D R (Multidrug Resistance), cross resistance arising in cancer cells to a wide range of chemically unrelated drugs, is commonly observed and 4 presents a major problem in A M L therapy. While - 7 0 % of patients achieve remission with initial therapy, approximately 75% of these w i l l relapse within 2 years of diagnosis in spite of additional consolidation chemotherapy 2 1. In addition, 20-30% of patients are unresponsive even to initial chemotherapy. Overall, long-term remissions are obtained in only 25% of patients 2 1. A small number of patients under 60 years of age and who have a suitable histocompatible donor are eligible for curative treatment supported by an allogeneic bone marrow or mobilized peripheral blood transplant. However, chemotherapy necessarily remains the main form of post-remission treatment. Given that chemotherapy is a highly invasive treatment with both long term and short term side effects, it would be valuable to be able to predict which patients are more likely to benefit from current chemotherapy treatments. In addition, the poor prognostic rate raises the need for more effective curative therapies. 1.3 - Current prognostic factors for predicting chemotherapeutic response in AML Older A M L patients (over 65 years) have a poor prognosis compared to young patients due to their lower drug tolerance and higher toxicity during high-dose chemotherapy 2 1. They are also considered clinically ineligible for allogeneic transplant therapy. Other than age, the current most important prognostic variable for patients with A M L is the detection of cytogenetic 9 9 9 4 abnormalities in a diagnostic bone marrow sample ~ . Almost 200 different recurring acquired cytogenetic aberrations have been identified in - 5 0 % of A M L patients 2 4. The most common ones are listed in Table 1.3. These abnormalities, in the forms of translocations, inversions, deletions monosomies and trisomies, play an important role in determining the biological basis of A M L . Intense molecular studies of specific genes at the sites of aberration revealed that they are usually 5 involved in normal blood cell development and homeostasis. Most inversions and translocations in A M L result in gene fusion products that can dysregulate proliferation, differentiation or apoptosis of blood cell precursors . One of the best known translocations (8;21) (Figure 1.2) is frequently associated with A M L subtype M 2 . It places the gene AML1 on 21q22 beside the gene E T O (a transcription factor) on 8q22. A M L 1 is the a subunit of the heterodimeric transcription factor Core Binding Factor (CBF) critical for normal hematopoiesis . The fusion protein retains the ability to bind to A M L 1 consensus regions and acts as a competitive inhibitor for the normal AML1 product, resulting in defective myeloid differentiation. Another frequent abnormality, inversion (16), is molecularly related to t(8;21) because it disrupts the gene encoding the P subunit of C B F . A M L subtype M 3 , also known as acute promyelocytic leukemia ( A P L ) , is characterized by translocation (15; 17) (Figure 1.3). This fuses the retinoic acid receptor a gene (RARa) on 17q 12-21 with the promyelocytic leukemia gene ( P M L ) on 15q22, which has a dominant effect on normal R A R a , antagonizing its differentiation function . Large cooperative studies have documented a significant relationship between detection of non-random chromosomal abnormalities and disease outcomes, including complete remission, disease-free survival and overall survival ' \" .. The prognostic significance of common abnormalities is summarized in Table 1.4. The presence of certain aberrations including inversion (16) and translocations (8;21) and (15; 17) is associated with significantly better overall survival compared to normal cytogenetics6 (Figure 1.4A). While use of the novel targeted therapeutic drugs all-trans retinoic acid ( A T R A ) 3 1 ' 3 2 and arsenic trioxide 3 3 \" 3 6 have drastically improved outcome for patients with t(15;17), the molecular basis of higher sensitivity to chemotherapy for t(8;21) or inv(16) A M L remains to be elucidated. On the other hand, other chromosomal changes such as -5, -7 and del(5) are frequently related to poor prognosis 6 compared to patients whose blasts are cytogenetically normal 6 (Figure 1.4B), although the molecular mechanisms responsible remain unclear. Despite the prognostic value of the many cytogenetic abnormalities identified almost 50% of patients present with an apparent normal karyotype. A patient's karyotype is routinely determined by traditional chromosome-banding and less frequently by newer methodologies such as spectral karyotyping ( S K Y ) or fluorescence in situ hybridization (FISH). Studies have confirmed the validity and consistency of these karyotyping techniques in detecting translocations and large-scale gains or losses of genomic D N A 3 7 \" 4 0 . It w i l l be useful, as suggested by large-scale mieroarray gene expression studies 4 1 ' 4 2, to try to subcategorize patients with a normal karyotype into groups with varying prognosis. Indeed, ongoing studies have suggested a few genes that may act as possible prognostic markers for A M L patients with a normal karyotype 4 3, including FMS- l ike tyrosine 3 (FLT3) (see Section 1.1), mixed lineage leukemia (MLL)44, C C A A T enhancer binding protein (CEBPA)45, nucleophosmin (NPM1)46 and brain and acute leukemia cytoplasmic {BAALC)41. Studies on more biomarkers should facilitate predicting treatment outcomes within the cytogenetically normal group. The objective of this thesis was to investigate the prognostic value of expression of the multidrug resistance-related ATP-binding cassette transporters ( A B C transporters). 1.4 - ABC transporters: an Overview ATP-binding cassette ( A B C ) transporters represent the largest transmembrane protein superfamily in eukaryotes and prokaryotes and are important factors in drug resistance. They are ATP-dependent protein transporters that actively pump a wide range of substrates across biological membranes. To date, 48 A B C transporters have been identified in humans 1 ' 4 8 ' 4 9. Figure 1.5 shows 7 the structural organization of the A B C transporter protein and gene. A B C proteins share a highly conserved ATP-binding cassette ( A B C ) , also known as nucleotide-binding domain (NBD) , which consists of the characteristic motifs Walker A (G-X2-G-X-G-K-S/T-T/S-X4-hydrophobic) and Walker B(R-X-hydrophobic2-X 2-P/T/S/A-X-hydrophobic4-D-E-A/P/C-T-S/T/A-AG-hydrophobic-D ) 5 0 . The A B C gene also contains motif C, or the \"signature m o t i f bearing the sequence (hydrophobic-S-X-G-Q-R/K-Q-R-hydrophobic-X-hydrophobic-A) organized between motifs A and B 5 1 . A fully functional unit of A B C transporters consists of two similar halves, each containing a transmembrane domain and an ATP-binding domain. Members of the superfamily can either be \"full transporters\" with both halves present or \"half transporters\" requiring dimerization for function1. The crystal structure of M s b A , an A B C transporter homolog in Escherichia coli, has been described by Chang et al5. This provides insights into the molecular structure and transport mechanism of A B C transporters. As seen in Figure 1.6, M s b A consists of two similar halves embedded in the lipid bilayer, organized much like a clamp hinging on extracellular connecting loops. The twelve transmembrane domains collectively serve as the substrate binding site. Based on this structure, the same group also proposed a general model for lipid transport by A B C transporters (Figure 1.7). In this \"flippase\" model, the transmembrane chamber first interacts with and intercepts the substrate, causing conformational changes that result in A T P hydrolysis by the N B D . This in turn causes a conformational shift that brings the two N B D s together. The resulting change in the chamber now produces an energetically unfavorable environment for the hydrophobic substrate and it \"flips\" from the inner to the outer membrane layer. After the flip, the N B D s separate leading to the expulsion of the substrate to the extracellular environment. Recently, Dawson and Locher proposed an alternate \"access and release\" model based on their crystal structure on the bacterial 8 A B C transporter Savl866, that postulates conformational changes to reflect the hydrolysis state of A T P 5 2 . The human A B C transporters are further categorized into subfamilies A to G based on sequence homology and structural similarity. Table 1.5 lists the current subfamily members, nomenclature and function. Human A B C s perform a variety of physiological functions by facilitating unidirectional shuttling of compounds within the cell as part of a metabolic process or outside the cell to other organs1'4 9. Unlike other types of transport proteins, one unique characteristic of the members of the A B C superfamily are their wide substrate specificity. The first 5 3 and best characterized A B C transporter, MDR1 (encoding the P-glycoprotein, PGP), is a promiscuous transporter of hydrophobic substrates. Physiologically, PGP is important in removing toxic metabolites from cells, especially in the brain 5 4 ' 5 5 . Other A B C s perform functions ranging from liver bile salt excretion (SPGP) to vitamin A transport in photoreceptors ( A B C R ) to cholesterol transport ( A B C A subfamily). A number of A B C transporters have been linked to genetic disorders, for example C F T R (mutations of which causes cystic fibrosis) and SPGP (mutations of which result in progressive familial intrahepatic cholestasis 2 5 6 ) . The promiscuity of A B C transporters, however, becomes a major clinical problem when cancer occurs: some A B C s are able to transport multiple chemotherapeutic drugs out of cancer cells, resulting in Multidrug Resistance. 1.5 - ABC transporters and multidrug resistance in cancer Overexpression of members of the ATP-Binding-Cassette ( A B C ) transporter family is found to be a main factor for M D R in cancer. There are three main A B C transporters involved in multidrug resistance: the classical FGP/MDR1, the multidrug resistance associated protein (MRP1), and the breast cancer resistance protein (BCRP1, ABCG2). MDR1, the prototype A B C 9 transporter, was first cloned from a drug-selected cell line displaying multidrug resistance . Knockout studies showed that mdrla/mdrlb(-l-) (homologs of human MDR1) mice had a high accumulation of drug levels in many tissues, especially the brain, confirming that P G P confers a general detoxification function against xenotoxins 5 7. The observation that not all M D R cell lines overexpress MDR1 led to the discovery of another multidrug resistance transporter, multidrug resistance protein 1 (MRP1). Cloned from a small-cell lung cancer cell line, MRP I transports drugs that are conjugated to glutathione (GSH) via the G S H reductase pathway 5 8. Knockout studies showed that MRP I is important in inflammation as well as for detoxification in the brain 5 9 . The third multidrug resistance transporter, BCRP1, was identified in cell lines selected for mitoxantrone resistance 6 0 ' 6 1. Unlike P G P and M R P 1 which are full A B C transporters, B C R P 1 is a half A B C and is thought to function as a homodimer. Knockout studies have suggested an in vivo role for this transporter in tissue detoxification and heme metabolite transport under hypoxic conditions 6 2 ' 6 3 . Hence all three MDR-related A B C transporters are physiologically important in protecting normal cells from a broad array of xenobiotics. This versatile defense mechanism, however, can be utilized by cancer cells for protection against a wide variety of chemotherapeutic drugs. Early clinical studies found high expression of P G P in many types of cancer, such as the leukemias, colon, kidney, adrenocortical and hepatocellular cancers 6 4 ' 6 5 . Amplification of MDR1 is frequently reported in many drug-selected M D R cell lines, although this is not a commonly observed mechanism for overexpression in clinical cases 6 6. MRP1 is overexpressed in leukemia, esophageal cancer and non-small cell lung cancers . Expression ofBCRPl has been detected in the leukemias , gastric cancer, hepatocellular carcinoma, endometrial cancer, colon cancer, melanoma and lung cancer 6 9 ' 7 0 . Figure 1.8 illustrates the mechanism of drug resistance via these transporters: the incoming 10 chemotherapeutic drug is intercepted by the transporter in the plasma membrane of the cancer cell and expelled, thereby failing to reach the intracellular site of action, resulting in drug resistance and failure of therapy. Other A B C transporters have been implicated in drug resistance. For instance, A B C A 2 71 can confer resistance to estramustine, a nitrogen mustard derivative of oestradiol . SPGP, an 79 A B C that shares high homology to P G P , is reported to confer resistance to paclitaxel . Members of the A B C C subfamily, commonly known as M R P 2 to M R P 9 , bear similarities to M R P 1 and have the potential to confer drug resistance. Given that many members o f the A B C superfamily have not been well characterized, they may also have the capacity to efflux substrates of clinical interest that has not yet been identified. Table 1.6 lists the common chemotherapeutic drugs that are known substrates of these transporters. P G P and B C R P 1 preferentially extrude large hydrophobic, positively charged molecu les 6 0 ' 7 3 7 5 , while M R P 1 can extrude both hydrophobic uncharged molecules and water-soluble anionic compounds 5 8 ' 7 6\" 7 9 . A l l three display capacity to transport daunorubicin 8 0, the common drug used for A M L , although daunorubicin efflux by M R P 1 is strictly dependent on G S H levels 8 1 . Many agents have been investigated in an effort to reverse PGP-mediated clinical M D R . These are generally competitive inhibitors that are substrates for the A B C transporters. Verapamil is a commonly used inhibitor that can modulate activity of a number of A B C transporters (most effective against PGP) . More specific inhibitors have been developed for each transporter, for example PSC-833 for P G P 8 2 \" 8 4 . Clinical trials using A B C inhibitors yielded mixed results in A M L , with complications including high bone marrow and neurological toxicity, confounding interpretations on the usefulness of these agents. Nonetheless, successful therapy has been reported for other types of cancer. In highly drug-resistant retinoblastoma for 11 example, combination of P G P inhibitor cyclosporin A with chemotherapy resulted in marked increase in relapse-free rate, demonstrating the value of A B C modulation in improving clinical outcome 8 5. 1.6 - ABC transporters in AML Expression of MDR1 in leukemic cells likely contributes to chemotherapy resistance in A M L patients. Treatment failure can be observed either as intrinsic chemotherapy resistance at diagnosis or at relapse. Overexpression of P G P is the most extensively studied mechanism of M D R in A M L . The presence of P G P detected by antibody staining as measured by flow cytometry has been demonstrated in 20% to 75% of de novo A M L patients according to different on O Q studies \" . Expression of P G P in patient peripheral blood or bone marrow is associated with lower remission rates 8 7 ' 8 8, shorter overall survival and lower disease-free survival (Figure 1.9) 4 ' 7 ' 9 0 ' 9 1. In addition, a number of studies have also reported an increase in P G P during relapsed disease 9 2\" 9 4. It is suggested that PGP-positive cells might escape initial chemotherapy and remerge at relapse, or alternately, MDR1 expression may be induced during chemotherapy. Ex vivo studies have shown that P G P expression decreases intracellular accumulation of daunorubicin in leukemic cells, and is reversed by P G P inhibitors ' ' . A n additional anti-apoptotic role has also been attributed to P G P in that it appears to protect leukemic cells from caspase-dependent programmed cell death 9 7\" 9 9. Several investigations on the M R P 1 transporter suggested a role in A M L drug resistance 1 0 0\" 1 0 2, but larger studies failed to identify a consistent relationship between M R P 1 and prognosis ' . Studies on the expression of B C R P 1 in A M L have also yielded mixed results. While some showed that up-regulation of BCRP1 is associated with a poor prognosis 1 0 4 ' 1 0 5 and is 12 common during relapse , others reported a lack of consistent overexpression in A M L patients 1 0 7\" 1 0 9. Several studies have reported the combination of different drug-resistant transporters, or their co-expression with cell survival factors, important in drug resistance 8 9 ' 1 0 0 ' 1 0 9 . Almost all past studies, however, have been limited to the three MDR-related transporters. Another important feature oiMDRl and BCRP1 are their changes in expression during the early differentiation of very primitive hematopoietic cells. A s described in the following sections, this feature has important clinical implications in the treatment of A M L . 1.7 - The leukemic stem cell model and its implications in drug response The concept of \"cancer stem cells\" has emerged as an important theme for cancer research in the past few years. Normal stem cells are defined by their dual capacity to regenerate themselves through self-renewal mechanism and to produce mature cells through differentiation 1 1 0. The \"cancer stem cell model\" proposes that a tumor is similarly sustained by a biologically distinct subpopulation of \"cancer stem cells\" (CSC), with the same ability to perpetuate the production of progeny with limited proliferative ability. Three observations support the existence of CSCs in human cancers. First, only a small fraction of a tumor has the capacity to regenerate a new tumor, operationally demonstrable upon transplantation into immunodeficient mice. Second, these tumor-initiating cancer cells can be identified and prospectively isolated by a distinct phenotype, usually based on flow cytometric or immuno-magnetic detection of differentially expressed surface antigens. Third, secondary tumors regenerated by these cells contain an array of mixed tumorigenic and non-tumorigenic cells, recapitulating the heterogeneity of the original tumor 1 1 1 . Hence irrespective of the origin of C S C , 13 this population bears the hallmarks of stem cells - self-renewal and \"differentiation\" into a functional hierarchy of \"primitive\" (tumorigenic) and \"mature\" (non-tumorigenic) cells. The C S C model was first and best developed in leukemia 2 ' 1 1 2 . In 1979, Minden and colleagues first described colony forming cells, termed \"blast progenitor cells\", from A M L patient peripheral b lood 1 1 3 , suggesting that a subpopulation of leukemic cells was clonogenic. It was, however, John Dick and colleagues who provided direct evidence for the leukemic stem cell (LSC) model . They described a primitive leukemic cell they termed SL- IC (SCID leukemia initiating cell) that can initiate A M L in mice. The SL-ICs from most patients tested are CD34++/CD38-, a surface phenotype similar to that of primitive normal cells that can regenerate normal hematopoiesis in immunodeficient mice (Figure 1.10). In addition to its tumorigenic potential, the SL- IC also shows the capacity to differentiate into the non-dividing leukemic cells similar to those which constitute the majority population in the malignant clone in the A M L patient from which it was isolated. The authors conclude that it is possible to isolate a small fraction enriched with L S C activity (0.2-100 stem cells in 10 6 blast cells), and the common surface properties and hierarchical organization support the hypothesis that the L S C originates by transformation of an initially normal hematopoietic stem cell (Figure 1.11). Nevertheless, it has also been shown that leukemic stem cells can be generated from more mature progenitors that reacquire stem cell properties 1 1 4\" 1 1 6. CSCs appear to share many common properties with normal stem cells 3. For example, overlap in their regulation by the Wnt, Notch and Sonic hedgehog (Shh) pathways, are associated both with oncogenesis and normal stem cell renewal 1 1 7 . A s well , C S C can share normal stem cell properties that allow a long life-span, such as protection against cytotoxins via expression of A B C transporters (see section 1.8 below). 14 Validation and adoption of the leukemic stem cell model calls for a paradigm shift in the treatment of the disease. Figure 1.12 gives a schematic representation of new treatment strategies that are likely more effective. Because the leukemia is sustained by the rare L S C s , this small population must be included as a target for effective therapies rather than just the majority of blast cells that have very limited proliferative ability. Existing drug therapies, however, are commonly targeted against the bulk leukemic population or their immediate precursors. Although a dramatic initial response can often be achieved, i f the L S C s are not also effectively eliminated they can eventually regenerate the disease. Hence to achieve more durable responses or even cure, there is a need for novel treatment methods more specifically directed against this primitive subpopulation. 1.8 - ABC transporters in normal and leukemic stem cells In recent years, MDR-related A B C transporters, in particular MDR1 and BCRP1, have been associated with stem cells of the hematopoietic system 1 1 8 . Early investigation by Chaudhary and Roninson first showed elevated expression oiMDRl in primitive (CD34+) normal hematopoietic ce l l s 1 1 9 . A s discussed above, CD34, a cell surface phosphoprotein, is a marker for both normal hematopoietic stem cells (HSC) and L S C s . Engraftment studies in humans , baboons 1 2 1 and m i c e 1 2 2 demonstrated the CD34+ population to be highly enriched with stem cell repopulation activity. It has also been suggested that CD34 expression may be in part regulated by the activation state of stem cel l s 1 2 3 . Further studies demonstrated a marked decline in MDR1 expression during differentiation 1 1 9 ' 1 2 4 . L ike MDR1, expression ofBCRPl is high in normal human H S C , drops dramatically in more committed progenitors, and remains low in mature hematopoietic ce l l s 1 2 5 . In addition, BCRP1 is reported to be a molecular determinant of the side 15 population (SP) in hematopoietic cells, a small distinct cell fraction with enriched H S C repopulating activity in adult mouse bone marrow 1 2 6 or human fetal l i ve r 1 2 7 but not vice versa. The SP, defined by its ability to efflux the fluorescent dye Hoechst 33342, was first identified in 128 126 129 131 murine bone marrow cells and later also found in many different human tissues ' \" , in spite of its failure to be detected on normal adult human H S C . One important property of the SP is its ability to exclude a number of drugs, reflective of the transport activity of A B C transporters. Interestingly, Uchida and colleagues reported that the dye (Rhodamine 123 and 132 Hoechst 33342) efflux ability of P G P and B c r p l are unstable in murine H S C s . This fluctuation appears to parallel changes in other H S C markers such as CD34 and CD38 and relate to the activation state of HSCs , suggesting a common control mechanism operated by a cell cycle checkpoint. One likely function of A B C transporters in stem cells is to protect this population from toxic substances over their long life-span. Another possible function is that these transporters can efflux regulatory molecules that can alter stem cell fate. For example, studies conducted by our lab revealed that P G P is a functional bile salt transporter in SPGP knockout mice, compensating for the lack of SPGP. The recently elucidated connection between A B C transporters and normal hematopoietic stem cells sheds a new light on the drug resistance property of CSCs . It has been suggested that CSCs can employ the same protective mechanisms operating in normal stem cells to defend themselves from chemotherapeutic drugs. Indeed, an increasing number of studies are associating A B C transporters to CSCs . For example, P G P expression has been correlated to CD34 positivity in A M L 1 3 3 , 1 3 4 . A s well , the SP has been identified in neuroblastoma, breast 135 139 cancer, ovarian cancer, glioblastoma and gastrointestinal cancer cell lines \" . These studies demonstrated that this small subpopulation has both enriched tumorigenicity and high drug 16 extrusion capacity. W u l f and colleagues found the SP to be detectable in most A M L patients, displays significantly increased drug efflux ability, and is able to regenerate the disease in N O D -SCID mice 1 4 0 . Studies by Feuring-Buske and Hogge confirmed the prevalence of SP in A M L , although SP+CD34+CD38- cells appeared to represent normal rather than leukemic primitive cells in A M L patients 1 4 1. The emerging paradigm on cancer biology posits functional heterogeneity within a cancer and the existence of a small distinct group of C S C reminiscent of normal stem cells. It remains to be investigated whether higher expression and activity of MDR-related A B C transporters contributes to higher drug tolerance in L S C , the presumptive subpopulation responsible for perpetuation of A M L . 17 1.9 - Thesis objectives The overall objective of this thesis was to investigate A B C transporter expression as a possible predictive factor for initial drug response in A M L patients. M y original hypothesis was that upregulation of A B C transporters is responsible for the lack of response to initial chemotherapy in A M L patients. In that context, I investigated the expression level of A B C transporters in total A M L blast cells and in different subpopulations along the leukemic hierarchy - CD34 + CD38\" (most primitive), C D 3 4 + C D 3 8 + , and CD34\" (most mature). Hence the three goals of this thesis were: 1. To profile m R N A expression patterns of the A B C transporter superfamily in bulk A M L patient materials, and compare profiles between patients that responded or failed to respond to initial chemotherapy. 2. To examine expression patterns of key MDR-related A B C transporters in primitive and mature A M L subpopulations of responders and non-responders. 3. To determine and compare the ex vivo drug sensitivity of CD34+CD38-, CD34+CD38+, and CD34- A M L subpopulations of responders and non-responders. The first and second goals are addressed in Chapters 2 and 3 where the RT-Real-Time-P C R technique was utilized to profile m R N A levels of A B C transporters in unsorted and sorted populations of A M L patient samples. In Chapter 4,1 investigated the ex vivo drug sensitivity of sorted leukemic subpopulations using the fluorescence-based Annexin V - P I assay for apoptosis. 18 Table 1.1. FAB classification of AML. FAB Morphology MO Minimal ly differentiated M l Myeloblas ts leukemia without maturation M 2 Myeloblasts leukemia with maturation M 3 Hypergranular promyelocytic leukemia M 4 Myelomonocytic blasts M 4 E o Variant, increase in marrow eosinophils M 5 Monocytic leukemia M 6 Erythroleukemia M 7 Megakaryoblastic leukemia Table 1.2. WHO classification of AML. A M L with recurrent cytogenetic translocations A M L with t(8;21)(q22;q22) A M L l / C B F a l p h a / E T O Acute promyelocytic leukemia: A M L with t( 15; 17)(q22;ql2) and variants P M L / R A R a l p h a A M L with abnormal bone marrow eosinophils inv(16)(pl3;q22) vagy t(16;16)(pl3;q22) C B F b e t a / M Y H l A M L with 1 lq23 M L L abnormalities A M L with multilineage dysplasia With prior M D S Without prior M D S A M L with myelodysplastic syndrome, therapy related Alkylat ing agent related Epipodophyllotoxin related Other types A M L not otherwise categorized A M L minimally differentiated A M L without maturation A M L with maturation Acute myelomonocytic leukemia Acute monocytic leukemia Acute erythroid leukemia Acute megakaryocyte leukemia Acute basophilic leukemia Acute panmyelosis with myelofibrosis Myelo id sarcoma Acute Leukemias of ambiguous lineage Table 1.3. Frequencies of common recurring cytogenetic aberrations in adult AML. Cytogenetic abnormality Cooperative Group Study (No. of patients) Adults total (n - 4257) No. (%) CALGB (n = 1311) No.(%) MRC (n= 2337) No. (%) SWOG/ECOG (n = 609) No. (%) None (normal karyotype) 582 (44) 1096 (47) 244 (40) 1922 (45) +8 123 (9) 211 (9) 53(9) 387 (9) -7/7q- 95(7) 209 (9) 52 (9) 356 (8) t(15;17)(q22;q21) 88 (7) 210(9) 27 (4) 325 (8) -5/5q- 86 (7) 183 (8) 36(6) 305 (7) t(8;21)(q22;q22) 81(6) 104 (4) 50(8) 235 (6) inv(16) 96 (7) 53 (2) 53 (9) 202 (5) - Y 58(4) N A 20 (3) 78 (4) t / inv(l lq23) 54 (4) 45 (2) 42 (7) 141 (3) +21 28 (2) 51(2) N A 79 (2) abn(17p) 30 (3) N A . 12(2) 42(2) del(9q) 33 (3) 37(2) 17(3) 87 (2) inv(3) 12(1) 61(3) 12(2) 85 (2) Complex with >3 abn 135 (10) N A 71 (12) 206(11) Complex with >5 abn 99 (8) 222(9) 53 (9) 374 (9) Research Counc i l 6 ' 2 9 ; S W O G / E C O G , Southwest Oncology Group/Eastern Cooperation Oncology Group ; abn, abnormality; N A , not available. Modif ied from Mrozek et al, Blood Reviews, 2004 2 4 . Table 1.4. Prognostic significance of frequent chromosomal abnormalities in AML. Good Standard Poor inv(16) Normal -5, del(5q) t(8:21) +8 -7 t(15;17) +21 Abnormal 3q +22 Complex del(7q) del(9q) Abnormal 1 lq23 A l l other structural abnormalities Modified from Grimwade et al, Blood, 1998 . 21 Table 1.5. List of human ABC transporters, location and physiological function. Gene (subfamily) Common name Location Expression Function A B C A I ABC1 9q31.1 Ubiquitous Cholesterol efflux A B C A 2 ABC2 9q34 Brain A B C A 3 ABC3, A B C C 16pl3.3 Lung A B C A 4 A B C R Ip22.1-p21 Photoreceptors N-retinylidene-PE efflux A B C A 5 17q24 Muscle, heart, testes A B C A 6 17q24 Liver A B C A 7 19pl3.3 Spleen, thymus A B C A 8 17q24 Ovary A B C A 9 17q24 Heart A B C A 10 17q24 Muscle, heart A B C A 12 2q34 Stomach A B C A 13 7 p l l - q l l Low in all tissues ABCB1 M D R 1/PGP 7p21 Adrenal, kidney, brain Multidrug resistance ABCB2 TAP1 6p21 A l l cells Peptide transport ABCB3 TAP2 6p21 A l l cells Peptide transport ABCB4 PGY3, MDR3 7q21.1 Liver PC transport ABCB5 7pl4 Ubiquitous A B C B 6 2q36 Mitochondria Iron transport ABCB7 Xql2-ql3 Mitochondria Fe/S cluster transport ABCB8 7q36 Mitochondria A B C B 9 TAPL 12q24 Brain, testis, spinal cord Peptide transport ABCB10 lq42 Mitochondria ABCB11 SPGP, BSEP 2q24 Liver Bile salt transport ABCC1 MRP1 16pl3.1 Lung, testes, P B M C Drug resistance ABCC2 MRP2 10q24 Liver Organic anion efflux ABCC3 MRP3 17q21.3 Lung, intestines, liver ABCC4 MRP4 13q32 Prostate Nucleoside transport ABCC5 MRP5 3q27 Ubiquitous Nucleoside transport ABCC6 MRP6 16pl3.1 Kidney, liver ABCC7 CFTR 7q31.2 Exocrine tissue Chloride ion channel ABCC8 SUR1 l lpl5.1 Pancreas Sulfonylurea receptor ABCC9 SUR2 12pl2.1 Heart, muscle ABCC10 MRP7 6p21 Low in all tissues ABCC11 MRP8 16qll-ql2 Low in all tissues ABCC12 MRP9 16qll-ql2 Low in all tissues ABCD1 A L D P Xq28 Peroxisomes V L C F A transport regulation ABCD2 A L D R 12qll-ql2 Peroxisomes ABCD3 PMP70 Ip22-p21 Peroxisomes ABCD4 PMP69 14q24.3 Peroxisomes ABCE1 OABP, RNASELI 4q31 Ovary, testes, spleen Oligoadenylate binding ABCF1 ABC50 6p2133 Ubiquitous ABCF2 7q36 Ubiquitous ABCF3 3q25 Ubiquitous A B C G I WHITE 1 21q22.3 Ubiquitous Cholesterol transport ABCG2 BCRP1 ,ABCP 4q22 Placenta, intestines Toxin efflux, drug resistance ABCG4 WH1TE2 llq23 Liver ABCG5 WH1TE3 2p21 Liver, intestines Sterol transport ABCG8 WHITE4 2p21 Liver, intestines Sterol transport Modified from Dean et al, Genome Research, 2001 . 22 Table 1.6. Substrate specificity of PGP, MRP1 and BCRP1. Gene Protein Non-chemotherapy substrates Chemotherapy substrates ABCB1 P G P Neutral and cationic organic compounds, many commonly used drugs Doxorubicin, daunorubicin, vincristine, vinblastine, actinomycin-D, paclitaxel, docetaxel, etoposide, teniposide, bisantrene, STI-571 8 6 ABCC1 M R P 1 Glutathione and other conjugates, organic anions, leukotriene C4 Doxorubicin, daunorubicin, epirubicin, etoposide, CO H(L TQ vincristine, methotrexate ' ABCG2 B C R P 1 , A B C P Prazosin Doxorubicin, daunorubicin, mitoxantrone, topotecan, S N -2g60,73-75 23 Figure 1.1. Morphologica l differences between different A M L subtypes. A M L subtypes M l (A & B) and M 6 (C & D) are shown as examples. A , M l myeloblasts with eccentric nuclei. B , Sudan Black stain of M l blasts. C , M 6 erythroblasts. D , coarse P A S stain of M 6 blasts. Courtesy of Haematological Malignancy Diagnostic Service http://www.hmds.org.uk. 24 8 Figure 1.2. A M L translocation (8;21) (left) compared to normal chromosomes 8 and 21 (right). Courtesy of Guide for Detection of M R D in A M L www.meds.com/leukemia/guide. ' l b : mm detfl 7) dft * \"^1 ail ' ^ •d«tl5) \\ • PAC 833D9 9 PAC 933118 Figure 1.3. A M L translocation (15;17) detected in A P L . Two probes (red: chromosome 15, green: chromosome 17) are utilized to visualize the translocation (yellow arrows). Courtesy of Atlas of Genetics and Cytogenetics in Oncology and Hematology. 25 25 -I 2 3 Years from entry 100 normal 60 years old or 1.5 g/m I V q l 2 h x 6 days for patients < 60 years old). Patients who entered a complete morphological remission with this therapy constitute the complete remission group (CR; responders) in this study. Patients who failed to achieve remission with this initial chemotherapy plus one additional cycle of treatment (either a second course of the same regimen 37 or an alternate regimen usually containing cyclophosphamide and etoposide ) constitute the non-responding group (NR; non-responders). Complete remission (CR) was defined as less than 5% blasts in a normocellular bone marrow with a neutrophil count >1.0 x 10 9 /L, and an unsupported hemoglobin of > 100 gm/L and platelet count >100 x 10 9 /L. C R patients received consolidation therapy consisting of either two cycles of additional chemotherapy the same as induction treatment or allogeneic transplantation (for patients <50-60 years of age with a suitable sibling donor and intermediate risk or a suitable sibling or unrelated donor and high risk cytogenetics as defined by the M R C ( U K ) criteria 6). 2.2.2 - RT-Real Time PCR assay 2.2.2.1. Overview. A reverse transcription (RT) -Real Time polymerase chain reaction (PCR) assay was developed and utilized to detect the relative expression levels of A B C transporters in A M L patient samples. This methodology was preferred over expression array chips for its high sensitivity. This proved crucial since many A B C transporters were expressed at low levels. A flow-chart of the assay was shown in Figure 2.1. Total R N A was first extracted from frozen patient samples and DNase was used to eliminate contaminating genomic D N A . R N A was then reverse-transcribed into c D N A and an aliquot of the same R T reaction was used for individual Real Time P C R reactions for each gene. The resulting Ct value (see below) from each P C R was used to calculate the relative expression of each gene of interest. The amount of P C R product was detected and measured by the fluorescence intensity caused by the binding of the fluorescent dye S Y B R Green to D N A . The threshold cycle, Ct, was defined as the cycle when fluorescence first becomes detectable above the threshold value. The Ct is inversely proportional to the amount of starting R N A transcript and is used to calculate the 38 relative expression level of the gene. To correct for differences in amplification efficiency (AE) between primer pairs, a standard curve was constructed for each gene. A housekeeping gene (GAPDH) was amplified from the same sample as an endogenous control to account for variability in concentration and quality of total R N A , and in the R T reaction efficiency. 2.2.2.2. - RNA isolation, DNase treatment and Reverse Transcription. Total R N A was isolated using Trizol for A M L P B samples (Invitrogen) and quantified by measuring its absorbance at 260/280 nm on a U-2000 spectrophotometer (Hitachi, Tokyo, Japan). Total R N A was treated with DNase I (Invitrogen) following manufacturer's instructions and subsequently reverse-transcribed using random hexamers and the Superscript II R T enzyme (Invitrogen) at a concentration of 1 u.g total R N A per 20 ul reaction. 2.2.2.3. Primer design and optimization. Primers for Real-Time P C R for all 47 A B C transporter genes were designed using PrimerExpress software, Version 2.0 (Applied Biosystems, Streetsville, Ontario, Canada). A l l primers were designed to yield a unique gene-specific product that does not overlap with consensus A B C walker sequences. Parameters used for design included 100 bp amplicon size, 19-22 bp primer size, 40-60 % G C content and 80-90 °C melting temperature. The validity of the primers was tested by conventional P C R and products were analyzed by agarose gel electrophoresis to ensure they gave a single product of the correct size. 2.2.2.4. Real-Time PCR. Real-Time P C R was performed with S Y B R Green Real-Time Core Reagents (Applied Biosystems) according to manufacturer's instructions on the A B I Prism 7900 Sequence Detection System (Applied Biosystems). Each 15 ul P C R reaction contained 1.5 ul diluted c D N A (24 ng starting total R N A ) . Thermal cycling conditions were 50°C for 2 min and 95°C for 5 min, followed by 40 cycles of 15 sec at 95°C, 30 sec at 58°C and 30 sec at 72°C. A n 39 additional cycle - 15 sec at 95°C, 15 sec at 60°C and 15 sec at 95°C - was performed at the end of the reaction to generate the dissociation curve of the amplicon to ensure a single, specific product with the corresponding melting temperature was produced (Figure 2.2). A negative R T control without the R T enzyme was included for each sample total R N A to ensure no reminiscent genomic D N A was amplified in the P C R reaction, and a negative P C R water control without c D N A was included per P C R reaction plate to check for reagent contamination. 2.2.2.5. Generation of standard curves. To determine the A E , a standard curve was constructed for each gene on a 2x c D N A dilution series equivalent from 30 ng to 0.47 ng of starting total R N A (Figure 2.3). The A E was calculated from the formula: 1 0 I / M - 1 , where M = the slope of the standard curve. 2.2.2.6. Data analysis. The Sequence Detector Software SDS 2.0 (Applied Biosystems) was used for data analysis. The threshold cycle value (Ct), defined as the cycle at which a statistically significant increase in S Y B R fluorescence (normalized to a passive reference Rox) is first detected, is automatically calculated and reported by SDS 2.0 for each reaction. Ct is inversely proportional to the log of c D N A . To determine the fold expression of a gene relative to the housekeeping gene, we used the formula: (1 + A E ) \" d C t where A E is the amplification efficiency of the specific gene and dCt = (Ct of the gene) - (Ct of the housekeeping gene). 2.2.2.7. Statistical analysis. Two statistical tests, the student's t test and the permutation test, were performed to assess expression differences between the C R and N R patient groups. A two tail-distribution, homoscedastic (assume two sample groups with equal variance) t test was utilized to compare the means of the patient groups. The permutation test is a randomization test which requires no assumptions about statistical distributions (random-sampling, equal variance). Statistical significance was set at p<0.05 for the t test and Z>2 for the permutation test. 40 2.3 - Results 2.3.1 - Profiling of ABC transporters in the drug-sensitive leukemic cell line C E M and its vinblastine-selected, drug-resistant subline CV1.0 I first profiled all 47 human A B C transporters in the lymphoblastic leukemic cell lines C C R F - C E M and C E M / V L B (although 48 human A B C s were predicted, sequences were available for only 47). These were chosen for comparison since the C C R F - C E M parental line is drug-sensitive while the vinblastine-selected C E M / V L B is multidrug resistant (at least 500-fold more resistant to vinblastine and 150-fold more resistant to doxorubicin than C C R F - C E M ) 1 4 3 ' 1 4 4 . A s described in the Materials and Methods Section, Ct values generated from the amplification plots were used to calculate the level of expression of each test gene relative to that of GAPDH (set at 106). Bustin and colleagues reported the transcript copy number per cell of GAPDH to be in the order of 2 x 10 3 to 5 x 10 3 in P B 1 4 6 , 1 4 7 . In addition, I found Ct values above 35 to be generally unreliable because it approaches the machine detection limit (40 = undetectable). Under the conditions used, this translates to approximately 5 x 1 0 -fold fewer transcripts than GAPDH. I therefore set the tentative \"biologically relevant\" expression level at 1/10 of GAPDH (discussed in Section 2.4), which I estimated to correspond to ~2-5 transcripts per cell. Similar to previous reports, MDR1 m R N A was upregulated by at least 2 x 10 3-fold in C E M / V L B cells due to gene amplification as compared to C E M (Figure 2.4) 1 4 8 . The drug-sensitive parental C C R F - C E M cells expressed very low level of MDR1 (almost 10-fold lower than the ~2 copy per cell reference line). There is no significant difference in MRP1 expression and BCRP1 levels are below the level of detection. A number of genes in the ABCA subfamily, notably ABCA5, 6, 9, JO, appear to be also upregulated in the drug-selected cell line. A s these are all clustered on 41 17q24, this chromosomal segment is likely to be amplified independently of MDR1 located on 7q21 during drug selection. Amplification of this cluster of genes in drug resistance warrants further investigation as the function of these A B C transporters are not well known. 2.3.2 - Lack of consistent differences was observed in ABC transporter expression between responsive and non-responsive patients Table 2.1 lists the characteristics of 31 A M L patients selected for expression profiling. Samples were chosen to represent patients differing in clinical response to induction chemotherapy with the combination of cytosine arabinoside and daunorubicin. O f the 31 samples, 18 were from patients who achieved C R after initial therapy and 13 were from the N R group (patients who did not achieve remission). 25 of the 31 patients had only normal karyotypes seen in the diagnostic bone marrow sample. In an initial set of experiments, amplification profiles were generated for all 47 A B C transporters from P B samples of 12 of the C R patients, and 6 N R patients. Figure 2.5 shows representative expression profiles (CR, Patient #7 and N R , Patient #9). Transcripts for over 40 A B C transporters were detectable in these A M L samples. These included the MDR-related transporters MDR1, MRP1 and BCRP1, as well as A B C transporters that have been previously reported to be restricted to other tissues, such as MDR3 (ABCB4) that transports phosphatidylcholine in the liver, SPGP(ABCB11) that exports bile salts in the liver, CFTR (ABCC7) the chloride ion channel in the lung, WHITES (ABCG5) and WHITE4 (ABCG8) which transport sterols in the liver and the intestine. However, expression levels of A B C transporters were generally low (at least 10-fold lower than GAPDH), with many below the reference line (2 copies per cell). This may represent differential expression among subpopulations of cells. A s 42 well , there was a significant variation (10 to 100-fold) among these 18 A M L samples in the m R N A levels of each A B C transporter detected. MDR1 expression in patients generally fell between that of C C R F - C E M and C E M / V L B . To evaluate the difference in expression between the C R and N R patients for every gene, the data set was tested independently using both a t test and a permutation test. Both statistical tests indicated no significant difference (p>0.05 for the t test and Z<2 for the permutation test) in any A B C transporter between the C R and N R groups, implying that m R N A levels of A B C transporters in the bulk population prior to treatment is not predictive of drug response. We also compared expression levels between A M L samples from the N R group and a C R subgroup of 6 patients (#24, #30 to #34) who achieved long-term remission for over 3 years - C R ( L ) . But, again, no statistically significant difference (p>0.05 for the t test and Z<2 for the permutation test) was found for any of the genes examined. Expression of a selected subset of 9 A B C transporters including the MDi?-related A B C transporters MDR], MRP1 and BCRP1, as well as ABCA2, ABCA3, ABCB9, SPGP, MRP4, and WHITE 1was evaluated in an expanded population of 31 A M L samples which included 18 C R and 13 N R patients. These genes were selected because they were either MDR-related or closely related to the MDR-transporter genes. Expression profiles of these genes are shown in Figure 2.6. Again, for each A B C transporter, both groups covered a wide expression range that overlapped with each other. Statistical analysis also confirmed that there was no consistent difference in m R N A expression between the two groups for these A B C transporters. A s apparent in the overlapping expression ranges, some C R patients actually show the highest expression of MDR-related transporters among all patients, while some N R patients have very low expression of these transporters. 43 2.4 - Discussion In this section, I used the sensitive RT-Real Time PCR technique to quantify and compare expression of the A B C transporter superfamily in various leukemic cell populations of defined drug responsiveness. Testing on the parental C E M cell line and its drug-selected sub-line validated the ability of this approach to detect differences in A B C expression levels. Consistent with prior reports, dramatic upregulation of the MDR] gene was observed in the drug-resistant CV1.0 cell line. Interestingly, a number of other A B C transporters were also elevated (the ABCA subfamily), most of which are poorly characterized. Whether or not these are inducible by drugs or related to drug resistance are not known and warrant further investigation. In this initial study I constructed the mRNA expression profiles of the A B C transporter superfamily in patient blast cells of 90% purity. Because A B C transporters are a major contributor to M D R in other cancers, I asked whether this would also be the case in A M L . However, I observed no consistent, statistically significant difference between the CR and N R patients' cells in all A B C transporters examined. Hence my results suggest that expression levels of A B C transporters in the total blast population are not a useful predictor of response to initial chemotherapy. There are several interpretations for this apparent discordance with older studies relating A B C expression to a poor prognosis) 4 ' 7 ' 9 0 ' 9 1 ' 1 0 0\" 1 0 2 ' 1 0 4 ' 1 0 5 . The first may be purely technical since differences in laboratory techniques and experimental conditions are known to produce variable results that confounding comparisons. In particular, old studies commonly measured the protein levels of A B C transporters, setting somewhat arbitrary levels of \"positivity\". Levels of ABCs, however, are likely a continuous variable. Thus my data likely give a more accurate depiction of 44 the expression profiles. On the other hand, m R N A expression levels may not translate to corresponding protein levels and drug efflux activity of the A B C transporters. Third, A B C transporters may simply not contribute significantly towards resistance to initial chemotherapy via high intrinsic levels, as suggested by the very low levels of expression detected. Instead, A B C transporters may play a role in AML- induced resistance via rapid and dramatic upregulation of expression under the stress of drug exposure. In support of this, acute induction of MDR1 expression after exposure to doxorubicin 1 4 9 and carcinogens 1 5 0 has been reported. Another plausible explanation, in line with the existing paradigm of cancer heterogeneity, is that the difference in expression lies not in the heterogeneous bulk sample, but in subpopulations of cells. This is supported by the very low expression levels (below reference line - less than one copy per cell) observed in a significant number of transporters in many patients. Possibly, a small subpopulation expressed a much higher, biologically-relevant level that is \"diluted\" in the average profile for the total population. This could also explain why some C R patients express relatively high levels of known MDR-related transporters and some N R patients express relatively low levels. Perhaps it is expression in the \"relevant\" cell fraction that accounts for differences in drug response. In order to obtain a more comprehensive picture on the subject, an examination of sorted subpopulations could be useful. 45 Table 2.1. Patient characteristics. Age at Patient no. Gender diagnosis F A B Cytogenetics Treatment Response 1 M 36 M4eo Inv(16)(pl3;q22),+22 C R 2 M 39 M 5 B Normal C R 4 F 17 M 5 A +8,del(5)(q31;q33) N R 7 M 54 M 5 A Normal C R 8 M 36 M 2 Inv(3)(q21;q26) N R 9 M 18 M 5 A Normal N R 10 F 63 M 4 Normal N R 11 M 68 M l +11 N R 12 F 25 M l Normal N R 13 M 69 M l Normal N R 14 F 68 M 4 Normal N R 15 F 46 M 2 Normal N R 16 F 25 M 2 Normal N R 17 F 49 M 5 Normal N R 18 M 57 M 2 Normal N R 19 F 28 M 4 Normal N R 20 M 51 M l Normal C R 21 M 41 M 5 B Normal C R 22 M 46 M 4 Normal C R 23 M 36 M 2 Normal C R 24 M 69 M 6 Normal C R ( L ) 25 F 39 M l Normal C R 26 M 69 M 3 Normal C R 27 F 66 M 4 Normal C R 28 M . 27 M 4 Normal C R 29 F 47 M 4 Normal C R 30 M 22 M 4 Normal C R ( L ) 31 M 25 M 2 +4, t(8;21)(q22;q22) C R ( L ) 32 M 64 M 2 Normal C R ( L ) 33 F 62 M 2 Normal C R ( L ) 34 M 50 M4eo inv(16)(pl3;q22), +22 C R ( L ) C R indicates complete remission achieved after induction chemotherapy; C R ( L ) , continuous remission >3years; N R , no response to induction therapy. Total R N A >cDNA DNase Reverse transcription Individual Realtime P C R 11111 Uniplex Real-time PCR: Triplicate for each gene Run PCR reactions in ABI Prism Sequence Detection System Software analysis: Obtain Ct value and dissociation curve for each reaction Calculate relative expression from Ct: (Gene A Ct) - (Housekeeping gene Ct) = dCt Expression fold = (1 + A E ) _ d C t Figure 2.1. Flow-chart of R T - R e a l T i m e - P C R . Total R N A isolated from patient peripheral blood was first DNased and reverse-transcribed. The c D N A was subsequently used for uniplex Real Time P C R for each gene. P C R reactions were prepared and aliquoted into a 3 84-well plate and run in the A B I Prism detection system, generating a Ct value per reaction. Expression of gene A was normalized to the housekeeping gene (GAPDH) by subtracting the housekeeping Ct value from Ct of gene A . The resulting dCt value and the amplification efficiency ( A E , see below) were used to calculate the relative expression, expressed as a fold difference to the housekeeping gene. 47 . D i s s o r c i n t i o r t P l o t 3 .OOO E -/ / 1 • 1 J 1 \\ ••/ J j f i • : -J 1-\\-J / \\ : V '•;>* \\ /' V „- 1 j V , / •vs. \\ s7 4-(• S O . O T o n m p e r a t u r e : i.1 . O O O E+1 -I . O O O E - 1 -\" ' 1 ^ p D O ; E - 2 -1 :ooo E - 3 -'•20* C y c l e Figure 2.2. Typical dissociation curve and amplification plot of a Real Time PCR product. A dissociation curve (top) was generated for the amplified product of the housekeeping gene GapDHin an A M L sample at the end of a Real Time P C R cycle. The melting temperature (Tm), 83 °C, corresponds to that of the expected GapDH amplicon from the primer design. The threshold cycle (Ct) of the product is 24 as observed in the amplification plot (bottom), with the red horizontal line indicating the threshold intensity. 48 Standard curves for selected genes Figure 2.3. Standard curves for selected genes. Ct values of real time PCR reactions were plotted against the amount of starting total RNA. For details, see above. 49 1000000 g 100000 a. a I Q Q. a O c o (0 0) 0) 1. a. X -*-* w a> LY. 10000 1000' 100 10-1 Figure 2.4. Profiling of the ABC transporter superfamily in C E M and CV1.0. m R N A expression levels normalized to GapDH (set at 106) of 47 known A B C transporters in cell lines C E M (black bar) and CV1 .0 (white bar). Solid horizontal line indicates the biologically-relevant reference line. 50 100000 10000-1000-100-10-1 HI 111, § ? g § g 55 Figure 2.5. Profiling of the ABC transporter superfamily in patients CR#7 and NR#9. m R N A expression levels normalized to GapDH (set at 106) of 47 known A B C transporters in cell lines CR#7 (black bar) and NR#9 (white bar). Solid horizontal line indicates the biologically-relevant reference line. 51 1000000-1 II Q 100000 Q . re o D . 10000 o c o (A (A 0 > _re 0) re u (A o> o _l 1000 100 CR NR f g . . . . . . - • f : i 1 I f • ABCA2 ABCA3 MDR1 ABCB9 BSEP MRP1 MRP4 WHITE 1 BCRP Figure 2.6. m R N A levels of selected A B C s in unfractionated A M L patient samples. A B C transporters were profiled in cells from 18 C R patients, and 13 N R patients. For each column (gene), each point represents a single patient from 3 groups (left: remission, right: refractory). Dotted lines indicate expression levels in cell lines (Blue: C E M , Red: CV1.0) 52 Ill Expression profiling of drug resistance-related transporters in FACS-sorted AML subpopulations 3.1 - Introduction Results from Chapter 2 raised the question of whether A B C transporters are differentially expressed in different subpopulations of leukemic cells. Based on the L S C model, high A B C expression in the fraction enriched in leukemia-initiating CD34 + CD38\" cells, would be expected to be the most critical to explain initial A M L treatment failure. A s discussed in Chapters 1.6 and 1.7, this fraction is chiefly responsible for initiation and maintenance of the whole leukemic population, and its frequency has been implicated to be an independent prognostic marker in A M L . Many studies have been conducted to examine the prognostic value of CD34 expression in A M L (typically categorizing patients into CD34-positive and CD34-negative for comparison of clinical outcome), albeit with mixed results (reviewed by Kanda et al in 2000 1 5 1 ) . Although older studies frequently report an association between CD34 \"positivity\" and lower remission rates 1 5 2\" 1 5 7, more recent studies found no such correlation 1 5 8\" 1 6 1 . CD34+CD38- leukemic cells display self-renewal and differentiating properties that are reminiscent of the function of normal H S C . It is, therefore, also likely that these cells would show higher drug tolerance, another characteristic of normal H S C . Recent studies by de Grouw et a l 1 6 2 and Peeters et a l 1 6 3 have demonstrated preferential expression of A B C transporters in both normal and leukemic CD34+CD38- cells. Since CD34+CD38- cells that express MDR-related A B C s w i l l have a survival advantage under cytotoxic stress, the presence of these transporters in this fraction of cells may predict chemotherapeutic failure. This may in part explain the mixed results on the prognostic value of CD34 expression 1 5 1 - it may be A B C 53 \"positivity\" of the CD34 positive leukemic cells, rather than CD34 \"positivity\" among total A M L blasts, that is prognostically significant. In this section, I profiled the key MDR-related A B C transporters - MDR1, MRP1, and BCRP1 - in different A M L subpopulations. M y working hypothesis was that intrinsic A B C expression in the \"relevant\" subpopulation, specifically the CD34+CD38- fraction, might be more predictive of initial treatment response than the bulk cells, and might possibly be an even more useful prognostic factor than the number of these primitive leukemic cells present. 54 3.2 - Materials and Methods 3.2.1 - Flow cytometric sorting of A M L subpopulations Frozen A M L patient cells were thawed at 37 °C and counted in 1:1 trypan blue for viability. A small aliquot was reserved as the unsorted population and the remaining was centrifuged for 9 min at 1000 rpm in 10 ml Iscove's medium (Invitrogen) + 20 % F B S (Invitrogen). Cells were resuspended in H F N (Hank's Balanced Salt Solution + 2% F B S + 0.05% NalSy + 5% human serum at a concentration of 7 x 10 4 cells per ul . For each 2 x 10 7 cells, 12 ul of each of the following antibodies were added: CD3-FITC, CD19-FITC, CD38-PE, CD34-Cy5 ( A P C - A ) . After incubation in the dark for 30 min on ice, 1 ml H F N was added to each tube and centrifuged for 7 min 1000 rpm. Cells were resuspended in 1 ml H F N with 2ug/ml PI, re-centrifuged and resuspended in 1.5 ml H F N . FACS-sort ing was performed on a F A C S A r i a flow cytometer (Becton Dickinson). PI negative cells were first gated as the viable fraction. CD34+CD38-, CD34+CD38+, and CD34- cells were gated within the viable C D 3 - , CD19-fraction. The following controls were included: IGG1-FITC only, I G G 1 - P E only (non-specific staining), C D 3 - F I T C only, CD19-FITC only, CD38-PE only, CD34-Cy5 only (for compensation). A sample F A C S sort is shown in Figure 3.1. 3.2.2 - RNA isolation, DNase treatment and RT-Real Time-PCR Sorted A M L subpopulation cells were centrifuged at 1200 rpm for 5 min. Lysis buffer was added to the cell pellet immediately after centrifugation and stored at -80 °C until R N A isolation. Total R N A was isolated using the RNeasy Micro spin column kit (Invitrogen) according to manufacturer's instructions. N o quantification was performed due to limitation of 55 material, and all total R N A isolated was directly treated with DNase I (Invitrogen), reverse-transcribed and subsequently used for Real Time P C R using a similar protocol as in Chapter 2.2.2.4 scaled down for smaller amounts of R N A . A two-tailed, homoscedastic student's t test was utilized to evaluate statistical difference between the N R and C R groups. Statistical significance was set at p < 0.05. 56 3.3 - Results 3.3.1 - Profiling of selected drug resistance-related transporters in FACS-sorted subpopulations of A M L patient samples Three fractions from each of 7 C R and 10 N R patients were sorted by F A C S : CD34+CD38- (most primitive), CD34+CD38+ (differentiating progenitors), CD34- (depleted of progenitors). CD3-CD19- cells were first gated to exclude contaminating normal T and B lymphocytes. The three major MDR-related A B C transporters were profiled for each subpopulation: MDR] (Figure 3.2), MRP1 (Figure 3.3), BCRP1 (Figure 3.4). Variation in levels of each gene among subpopulations of the same patient is apparent, indicating heterogeneity within the cancer. Expression levels of MDR1 and MRP1 were frequently detected above the reference line. Detectable BCRP1 levels were less frequent and fall below the reference. Significant expression for MDR1 and BCRP1 appeared to be restricted to the CD34- fraction for C R patients, while N R patients showed high levels also in the primitive subpopulations (see below). MRP1 expression, on the other hand, was observed across all fractions. 3.3.2 - Higher expression of MDR1 and BCRP1 in the CD34+CD38- cells from non-responders Figure 3.5 shows a scatter plot of expression levels in the CD34+CD38- fraction. For MRP1, there was significant overlap between the C R and N R groups and no significant difference was observed. MDR1 and BCRP1, however, showed interesting patterns in the primitive fraction. A l l 7 C R patients expressed uniformly low levels of MDR1 and BCRP1 in the CD34+CD38- cells. In contrast, 5/10 N R patients (NR6, N R 8 , NR10 , NR14 , NR15 , 50%) show significantly higher expression of MDR] (all of which are above the median), and 5/10 N R patients (NR6, N R 8 , N R 9 , N R 1 5 , NR18 , 50%) have high expression ofBCRPl (most of which are significantly 57 above the median). Differences for both genes reached statistical significance (p<0.05). In combination, 7/10 N R patients (70%) show high expression of MDR1 and/or BCRP1. Interestingly, MDR1/BCRP1 expression profiles of the N R group do not resemble a continuum. Rather, there seem to be a distinct segregation within the group, with patients' cells expressing high MDR1/BCRP1 being well separated from the rest of the N R patients clustered with the C R group (especially for BCRP1). O f note, MDR1/BCRP1 levels in CD34+CD38- cells are very low, especially BCRP1 (>103-fold lower than the housekeeping gene GAPDH). To test i f the proportion of CD34+CD38- cells among A M L blasts could predict treatment response, the CD34+CD38- fraction size was also compared between N R and C R patients (Figure 3.6). Contrary to old studies reporting the prognostic significance of CD34 expression 1 5 2\" 1 5 7 , however, comparison of the two groups revealed no statistically significant difference (means of % CD34+CD38-for C R and N R = 2.9 and 12 respectively, p>0,05). 58 3.4 - Discussion In this chapter, I profiled the MDR-related transporters in FACS-sorted subpopulations of C R and N R patients. Results demonstrate that, consistent with the concept of cancer heterogeneity, A B C transporters were differentially expressed across subpopulations along the leukemic hierarchy. This argues against the concept that expression levels are homogeneous across all cells in the leukemic clone. Indeed, profiles on subpopulations proved far more revealing on the potential underlying mechanisms of A M L M D R . N o difference in MRP1 expression between the two patient groups was observed, showing that MRP1 is not a useful predictive factor of initial treatment outcome and likely not an important contributor to drug resistance in A M L . On the other hand, expression of MDR1 and/or BCRP1 was elevated in CD34+CD38- cells of 7/10 N R patients compared to uniformly low levels in the C R group. Hence intrinsic high expression of either transporter in this primitive fraction enriched for leukemia-initiating cells appears predictive of poor response. Furthermore, MDR1/BCRP1 m R N A levels in CD34+CD38- cells were more predictive of response than the size of the CD34+CD38- fraction itself, which showed no significant difference between the C R and N R groups. Both MDR1 and BCRP1 can efflux daunorubicin, an important chemotherapeutic drug used for most A M L patients. According to the L S C model, CD34+CD38- cells are responsible for maintenance of the whole leukemic population, thus high expression of these transporters w i l l give them a critical survival advantage under drug exposure, allowing them to regenerate a leukemic population resulting in refractory disease. Although high levels of A B C transporters were also observed in the CD34- fraction of some responders, this mature subpopulation is 59 incapable of propagating the disease on its own and thus w i l l not contribute to treatment outcome. Comparing differences between the C R and N R groups, both MDR1 and BCRP1 reached statistical significance. More notably however, it is observed that within the N R group, expression levels are either significantly elevated (above the mean) or very low like the C R group, making these \"outliers\" clearly identifiable. Hence it may be possible to distinguish the potential ABC-dependent (high expression) responders within the N R group for more specifically targeted therapy (see Chapters 4 and 5). Higher relative expression of A B C transporters in the primitive leukemic fraction is not an entirely surprising finding. In fact, based on the current existing paradigm that L S C originate from transformed H S C , the leukemic CD34+CD38- cells would be expected to be found to express the highest A B C levels among the three subpopulations examined, as previously reported for their normal counterparts 1 1 9 ' 1 2 5. Indeed, profiling of a limited number of CD34+ normal P B samples (with H S C mobilization by G - C S F injection) showed MDR1 transcript levels to be even higher than those of the high-expressing N R patients (data not shown). Hence these patients appear to be retaining existing normal stem cell protection mechanisms, while the low-expression patients may have actually \"lost\" this defense system during leukemogenesis. Even more intriguing is the observation that some CD34- fractions displayed high A B C expression. Since normal CD34- cells do not express high levels of these transporters 1 1 9 ' 1 2 4 ' 1 2 5 , their leukemic counterpart may have abnormally turned on the expression of these genes. Whether this is a random product of aberrant epigenetics in A M L or part of a more systematic mechanism remains to be elucidated. Results of this study suggest the utility of evaluating the leukemic CD34+CD38-subpopulation for both MDR1/BCRP1 expression in predicting a N R outcome. Despite the 60 apparent difference between the CR and N R groups; however, the low MDR1/BCRP1 expression levels detected in N R patients raises the possibility of their biologically relevance. In particular, BCRP1 levels were so low that they fell below the biological reference line. Further studies are required to discern what expression level translates to functionally significant activity. 61 July 06-July24 - NR9 |I*MM| I II IIMj * I I 1 III! I I I till -86 0 10 2 10 3 10 4 1 FL1 FITC-A July 06-Julv24 - NR9 -139 0 10 PE-A Tube; Jufy24 - NR9 Population O Ail Events P1 #Events %Pafent %Total L-~Wi P2 P4 PS m i p s 61.200 44.374 41.521 38.745 6,350 19,824 7,073 72.5 93.6 93,3 16.4 51.2 18.3 100.0 72.5 67.8 63.3 10.4 32.4 11.6 Figure 3.1. F A C S analysis of A M L patient sample NR#9. Left: Staining of cells with C D 3 -F I T C and CD19-FITC to gate the C D 3 - C D 1 9 - (P3) fraction. Right: Staining and sorting of CD34+CD38- (P4), CD34+CD38+ (P5), and CD34- (P6) cells from the P3 gate. Bottom: Size of each gate is shown as a percentage of the parental gate or percentage of total population. 62 20000 17500 TT i Q a ro 0 -2 12500 C o ' » V) 15000 10000-\\ Q. X E a> > 7500 ^ £ 5000 — a> or 2500-I MICD34+38-ES5CD34+38+ ^ CD34-Complete Remission r d Non-responsive *«ict £ ^ * * f ^ ^ / • Figure 3.2. mRNA expression levels of MDR1 in FACS-sorted A M L subpopulations. CD34+CD38- (red), CD34+CD38+ (green) and CD34- (blue) cell fractions of AML patient samples were sorted by FACS analysis for Real Time-PCR. Expression levels were expressed relative to GAPDH (set at 106). Solid horizontal line indicates the biologically-relevant reference line. 90000-, _ 80000-| JL D Q. « CD o c o 70000 60000-\\ S 50000 4> Q. X £ a> 40000 30000 20000 a> 10000-1 III M C D 3 4 + 3 8 -tSS3CD34+38+ ^ CD34-Complete Remission J Non-responsive cf 0# J £ ^_ ^ c$ # / / # * Figure 3.3. mRNA expression levels of MRP1 in FACS-sorted AML subpopulations. CD34+CD38- (red), CD34+CD38+ (green) and CD34- (blue) cell fractions of AML patient samples were sorted by FACS analysis for Real Time-PCR. Expression levels were expressed relative to GAPDH'(set at 106). Solid horizontal line indicates the biologically-relevant reference line. 800 700-JL 600-Q Q. <0 o o *-> 500-c o 'vt M a> i _ 400-Q. X < Z 300-a: E a> > 200-100 • • C D 3 4 + 3 8 -ES3CD34+38+ ^ CD34-Complete Remissi on jaJ J . * B . ts Non-responsive # c# ^ ^ ^ #N ^ / ^ ^ Figure 3.4. mRNA expression levels of BCRP1 in FACS-sorted A M L subpopulations. CD34+CD38- (red), CD34+CD38+ (green) and CD34- (blue) cell fractions of A M L patient samples were sorted by F A C S analysis for Real Time-PCR. Expression levels were expressed relative to GAPDH (set at 106). Expression levels fall significantly below the biologically-relevant reference line at 1000 (not shown). 17500 15000 \"to T — JL i ' £ 12500 (0 O J 10000 ' i _ a. * 7500 < or E 5 5000-ro a: 2500-N R 1 0 A NR14 A N R 6 , 8 * A NR15 250n 225H 200 175 150 125 100 75 50 25 B N R 6 NR9,18 NR15 A N R 8 A . , 70000 '60000 50000 40000 30000' 20000H IOOOOH A A complete remission non-responsive complete remission non-responsive complete remission non-responsive Figure 3.5. Comparison of expression oiMDRl, BCRP1 and MRP1 between the CR and NR groups in the CD34+CD38-fraction. Expression levels of C R patients (square) and N R patients (triangle) were plotted for MDR1 (A), BCRP1 (B) and MRP1 (C). Red and blue horizontal lines represent median levels of C R and N R , respectively. Patients with high MDR1/BCRP1 levels are identified beside their symbols. Expression levels relative to GAPDH (set at 106). Asterisk indicates statistical significance. 70-60H c m CD o. 50-c c o o 40-oo 30-m a o + 8 2 0: 10 complete remission non-responsive Figure 3.6. CD34+CD38- fraction size in CR and NR A M L patients. Percentages of the FACS-sorted CD34+CD38- subpopulation is plotted for C R (square) and N R (triangle) patients. Horizontal lines represent the median of each group. 67 IV Ex vivo drug sensitivity of primitive and mature subpopulations of Acute Myeloid Leukemia and effects of ABC transporter modulation 4.1 - Introduction Expression profiling as described in the previous chapter allowed rapid examination of a high number of genes in patient samples. However, high gene expression does not necessarily correlate to high protein expression or functional activity. To evaluate the biological relevance of expression data in drug response, a functional assay is necessary. In this chapter I investigated the ex vivo drug sensitivity of A M L patient cells to determine the functional relevance of MDR1/BCRP1 gene expression, particularly in the primitive CD34+CD38- subpopulation. I first adapted and validated two functional assays, the Annexin V - P I apoptotic assay and the M T S proliferation assay, on the C E M and C V 1 . 0 leukemic cell lines (see section 2.3.1), These two assays were used to test the sensitivity of C E M and CV1.0 to exposure of daunorubicin, a major chemotherapeutic drug used for A M L treatment and a known substrate for both P G P and B C R P 1 . The C E M cell line with low A B C transporter expression should display higher sensitivity to the drug than the C V 1 . 0 cell line with MDR1 amplification. Addition of A B C modulators PSC-833 (highly specific inhibitor of PGP) and verapamil (inhibitor of A B C transporters including P G P and B C R P 1 ) is expected to result in re-sensitization of C V 1 . 0 to daunorubicin. The objectives of this chapter were two-fold. The first was to examine possible differences in drug sensitivity among FACS-sorted subpopulations from A M L patients. A s discussed earlier, the L S C fraction exhibits properties reminiscent of the normal H S C . Hence I 68 hypothesized that the primitive CD34+CD38- fraction would be associated with drug resistance, mirroring protection of H S C from cytotoxins. Specifically, I hypothesized that the CD34+CD38-fraction of N R patients would exhibit the highest tolerance to daunorubicin. M y second objective was to investigate the functional significance of A B C transporters in A M L subpopulations. Based on my observations of higher A B C gene expression in CD34+CD38- cells from A M L N R as compared to C R patients (Chapter 3), I hypothesized that ex vivo drug resistance would be associated with high expression levels of MDR1 and/or BCRP1, and that addition of A B C inhibitors would have the largest drug re-sensitization effects on the primitive fraction of N R patients, while the low expresser C R fractions would remain unaffected. 69 4.2 - Materials and Methods 4.2.1 - Exposure of A M L cells to drugs Unsorted or FACS-sorted A M L patient cells were centrifuged and re-suspended in Iscove's medium (Invitrogen) + 10% F B S at a concentration of 250 cells per ul . To each well , 100 ul cells, 50 ul daunorubicin at varying dilutions (0.001 to 0.5 jig/ml) and 50 ul PSC-833 (1 u M or 3 uM) or verapamil (5 ug/ml or 20 ug/ml) were added. Cells were incubated at 37 °C for 24 hours before subjected to apoptosis or proliferation assay (see below). 4.2.2 - Annexin V-Propidium Iodide assay The Annexin V - P I assay utilizes two cell-death markers Annexin V (conjugated to F ITC fluorochrome) and propidium iodide (PI) to stain for cells which are undergoing or have undergone apoptosis, respectively. In the early stages of apoptosis, membrane rearrangement causes the translocation of phosphatidylserine (PS) from the inner to the outer leaflet of the plasma membrane. Thus the binding of PS by Annexin V - F I T C allows detection of cells in the early stages of apoptosis. This is coupled with the vital dye PI that stains for later-stage apoptotic cells with a loss of membrane integrity. Cells that are both Annexin V - F I T C and PI negative are defined as viable. Figure 4.1 shows the detection of viable cells as double negatives under exposure of a low and high concentration of daunorubicin. Apoptosis was measured using the Annexin V - F I T C + PI detection kit I ( B D Biosciences) with a modified manufacturer's protocol. After incubation with daunorubicin for 24 hours, A M L patient cells were centrifuged for 5 min at 1200 rpm. Medium (170 ul) was removed from the top of each well with caution before addition of 100 ul of l x Binding Buffer 70 and 20 ul of 1/8 diluted Annexin-V and PI mix. IGG1-FITC Antibody (BD Biosciences) was added to the mock-treatment well as negative control. Additional staining controls were carried out with addition of Annexin V - F I T C only and PI only. Cells were incubated in the dark at room temperature for 15 min. Reactions were terminated by addition of 100 ul l x Binding Buffer. Fluorescence was measured using a FACSCal ibu r flow cytometer (High Throughput Sampler, B D ) and analyzed by Flow Jo v.2.0 software. 4.2.3 - M T S assay The M T S assay is a colorimetric, proliferation assay that utilizes the soluble chemical tetrazolium salt (MTS) . M T S is reduced to formazan by metabolically active cells (Figure 4.2). Production of formazan can be detected by development of a brown colour and is proportional to the number of viable cells. Proliferation was measured using the CellTiter 96 Aqueous Non-Radioactive Cel l proliferation Assay (Promega). After 24-hour incubation with daunorubicin, 20 ul of Aqueous Solution 1 was added to cells and incubated for 2 hours at 37 °C. Color development was quantified as absorbance at 490 nm by the M R X Microplate Reader (Dynex Technology). 4.2.4 - Analysis For the M T S assay, absorbance was plotted against daunorubicin concentration to generate a drug sensitivity curve for each concentration of A B C inhibition (no inhibition, l u M PSC-833, 3 u M PSC-833, 5 ug/ml verapamil, 20 ug/ml verapamil). For the Annexin V - P I assay, % viability was defined as % of cells in the Annexin V - F I T C and PI double-negative quadrant (in reference to the non-specific IGG1 control). Using this value, % k i l l was calculated 71 for each daunorubicin concentration with respect to the viability control (no daunorubicin) using the formula (% viable control-% viable sample)/% viable control x 100%. The drug sensitivity curve was then generated by plotting % k i l l against daunorubicin concentration for each concentration of A B C inhibition. IC50 for each curve was obtained as the daunorubicin concentration at 50% k i l l . Fold change in IC50 by A B C modulation was calculated using the formula IC50 unmodulated/ IC50 highest inhibition dose. A two-tailed, heteroscedastic student's t test was utilized to evaluate statistical difference between the N R and C R groups. Statistical significance was set at p <0.05. 72 4.3 - Results 4.3.1 - Comparison of the Annexin V-PI apoptotic assay to the MTS proliferation assay on C E M and CV1.0 cell lines. The Annexin V - P I assay and the M T S assay were performed to measure and compare drug tolerance of C E M and CV1.0 cells. Daunorubicin sensitivity was first determined by the apoptosis assay. Cells were stained with Annexin V - F I T C and PI after exposure to different concentrations of daunorubicin with or without PSC-833 or verapamil. A s expected, Figure 4.3 shows that C V 1 . 0 cells without A B C modulation is much more resistant to daunorubicin (only 20% k i l l at highest dose of 0.5 ug/ml) than parental C E M cells ( I C 5 0 = 0.025 ug/mi). Neither PSC-833 nor verapamil had a significant effect on C E M while both had a dose-dependent re-sensitizing effect on CV1 .0 cells, which is consistent with MDR1 amplification being a main cause of the drug-resistance mechanism in this cell line. Results from the M T S assay on a similar drug study were represented in Figure 4.4. Again, CV1 .0 cells were more resistant, as indicated by their higher proliferation activity (proportional to higher absorbance) after daunorubicin exposure, and A B C modulation had a dose-dependent effect in decreasing their proliferation/survival. Similar results from the two assays gave results consistent with expectations showing that both assays provided valid measures of ex vivo drug responses of leukemic cells. Because of its single-cell analysis, the apoptotic assay was more sensitive and reliable compared to the colorimetric M T S assay, and was thus used for subsequent studies on A M L patient cells. 73 4.3.2. Adaptation of the apoptotic assay to A M L patient cells. Due to the limitation of sorted cells, I next tested assay conditions on unsorted A M L patient cells to adapt the apoptotic assay further to clinical material. To determine the maximum dose of A B C modulators PSC-833 (dissolved in D M S O ) and verapamil that can be used, bulk cells from A M L patient #18 were subjected to different concentrations of the A B C modulators in culture and % k i l l was determined using the Annexin V - P I apoptotic assay. Because the concentrated PSC-833 stock solution was dissolved in D M S O , the matching volume equivalent of D M S O was also tested for each PSC-833 concentration. A s seen in Figure 4.5, the maximum dose of PSC-833 and verapamil without significant toxicity on A M L primary cells was 3 u M and 20 ug/ml, respectively. D M S O had no independent toxic effect in the assay conditions used. Next, a range of daunorubicin concentrations with or without addition of the modulator PSC-833 was tested on two unsorted patient samples. A s expected, A B C modulation has no significant effect on patient #18 (Figure 4.6), whose expression of MDR1 was among the lowest of all patients (Figure 4.6, small panel). On the other hand, Figure 4.7 shows that patient #25, who had the highest MDR1 expression (Figure 4.7, small panel), displayed a more resistant profile (% k i l l plateaued at about 60%) than patient #18 (% k i l l reached about 100% at 0.1 ug/ml). Moreover, PSC-833 had a dose-dependent effect, increasing the sensitivity of patient #25 cells to the P G P substrate daunorubicin. This validated the application of the apoptotic assay on clinical samples and demonstrated A B C inhibition as a useful measure of A B C transporter activity. 74 4.3.3 - Subpopulation size and patient material availability as a source of limitation. For study coherence and to allow direct comparison between expression data and functional results, I performed the ex vivo drug sensitivity assay on the same patient samples used in Chapters 2 and 3. Unfortunately, there were limitations in the selection of patient samples for functional studies. The main limitation was the size of subpopulations obtainable. Taking into account the length of sort time and number of patient cells per frozen aliquot, a size of at least 2% of each subpopulation was typically needed from a single patient. Table 4.1 lists the fraction sizes of the same patients that have been profiled for A B C expression in their subpopulations, with an asterisk marking those with high expression of MDR1 and/or BCRP1 in the CD34+CD38- cells (Chapter 2). Another limitation was the availability of patient samples (for example, no more cells were available for patient CR#30). Taken together, two C R patients (CR#1 and #34) and six N R patients (NR#8, #9, #14, #15, #16, #19) matched the criteria of availability and ample fraction size. Four of the six N R patients (NR #8, #9, #14, #15) had high MDR1/BCRP1 expression in CD34+CD38- cells. These eight patients were used for the functional assay. 4.3.4. Higher daunorubicin resistance and larger effect of ABC modulation in the CD34+CD38- fraction of non-responders. CD34+CD38-, CD34+CD38+, and CD34- fractions were sorted by F A C S from the eight selected A M L patients. These as well as unsorted cells were subjected to 24-hour daunorubicin exposure in the presence or absence of PSC-833 or verapamil followed by the apoptotic assay. 75 A s illustrated by examples of drug sensitivity curves from the C R and N R groups in Figures 4.8 and 4.9 (other profiles were included in the Appendix), striking differences were observed between the two. For C R patient #1, unsorted (total), CD34+CD38-, CD34+CD38+ and CD34-fractions showed high sensitivity to daunorubicin, reaching over 85% k i l l at 0.1 (a.g/ml (IC50 = 0.03 to 0.04 ug/ml) with or without A B C modulation, and addition of either PSC-833 or verapamil only had a slight effect of increasing its sensitivity. In contrast, N R patient #9 exhibited significant differences among subpopulations. A s shown in Figure 4.9, unmodulated CD34+CD38- and CD34+CD38+ (35% - 40% k i l l at 0.1 ug/ml, I C 5 0 > 0.1 ug/ml) of NR#9 were significantly more resistant than CD34- and unsorted (55% and 70% at 0.1 ug/ml respectively). A B C modulation had a dose-dependent re-sensitizing effect on all subpopulations, with a magnitude following the order CD34+CD38- > CD34+CD38+ > CD34- , suggesting that A B C transporter activity was a significant contributor to drug resistance in the non-responder, especially in the primitive subpopulation. Figure 4.10 shows the IC50 of daunorubicin for each subpopulation of C R and N R patients. Although only two C R patients were tested, they gave similar results. Both showed very high drug sensitivity (IC50 O . 0 4 ug/ml) in the primitive CD34+CD38- fraction, while the mature CD34- fractions were slightly more resistant to daunorubicin. On the other hand, the N R patients showed a range of resistance in terms of higher IC50 values across all fractions. A s a group, N R patients were much more resistant than the C R patients, particularly in the CD34+CD38- fraction in which there was no overlap between the two groups and difference reached statistical significance (p <0.05). Drug sensitivity corresponded with A B C expression, in that the four patients ( N R #8, #9, #14, #15) with high MDR1/BCRP1 expression displayed the highest resistance in this primitive fraction. 76 Figure 4.11 shows the effect of A B C modulation on drug sensitivity for each subpopulation. Both C R patients showed little change in the presence of A B C inhibitors, especially in the CD34+CD38- fraction, and only a minor increase in sensitivity in the CD34- fraction. Contrastingly, A B C inhibition markedly decreased IC50 in the primitive CD34+CD38- and CD34+CD38+ fractions of N R patients. There is a statistically significant difference between the C R and N R groups in CD34+CD38- (p <0.05, student's t test) but not in CD34+CD38+, CD34-and total fractions. A s with drug resistance, observed effects of A B C modulation is consistent with expression data, with the MDRl/BCKPl-expressing N R patients being most affected by A B C inhibition, demonstrating the M D R activity of these transporters. 77 4.4 - Discussion In this section, I followed up on the expression results from Chapter 3 with ex vivo functional studies to test the drug sensitivity of sorted patient subpopulations and the activity of A B C transporters in these fractions. A s with levels of A B C expression, differences in functional characteristics were apparent among subpopulations along the leukemic hierarchy. Sensitivity curves of unsorted cells, as well as the effects of A B C modulation on these, were generally similar to that of the predominant fraction of CD34- cells rather than the small primitive CD34+CD38- fraction, again showing that studies on total population may not be representative of the tumorigenic cells of importance for prognosis. Although the population size is small, results of this study revealed that ABC-dependent daunorubicin resistance in the CD34+CD38-fraction was common among N R patients who are mostly karyotypically normal (not observed in the C R group) and might thus be predictive of poor clinical response to initial therapy. This difference between the C R and N R patients was not seen in either the mature CD34- fraction or the bulk population. For C R patients, the CD34- fraction displayed slightly higher resistance than the primitive cells which can be overcome by A B C modulation, showing some level of A B C transporter activity in this fraction. Nevertheless, this did not correlate with clinical outcome (CR patients), consistent with the hypothesis that the features of the non-tumorigenic fraction may not be prognostic. Modulation of A B C activity increased daunorubicin sensitivity in a dose-dependent fashion in the CD34+CD38- cells of most N R patients studied. Such an effect was most dramatic for NR#8, #9 and #15, patients with high MDR1/BCRP1 expression (3/6 tested). A B C transporter activity was consistent with the expression data in that high MDR1/BCRP1 78 expression corresponded with high drug resistance, reversible by ABC-specif ic inhibition. This can have important clinical implications. Drug studies, especially on subpopulations, is time-consuming and laborious, hence may not be feasible for routine testing in a clinical laboratory, especially for an acute disease such as A M L that requires rapid intervention. On the other hand, because expression studies are faster, require less clinical material, and may be adapted to high through-put protocols, it is potentially more valuable for identification of the N R cases where MDR1/BCRP1 is a significant contributor to drug resistance. Interestingly, although CD34- cells of N R patients exhibited a range of drug resistance like that of CD34+CD38- cells, effects of A B C modulation on CD34- is much smaller (except NR16) . Therefore, while MDR1/BCRP1 expression may significantly contribute to drug resistance in primitive leukemic cells, it is not the dominant resistance mechanism in mature cells. Based on the existing paradigm that the L S C originate from H S C , the L S C wi l l likely retain many of the normal stem cell properties, including its protective mechanism against cytotoxins. A s evident from my studies, L S C from a number of patients did appear to have high A B C expression like HSCs , which became advantageous under chemotherapy resulting in treatment failure. What additional mechanisms drive drug resistance in the CD34- fraction remains to be elucidated. The L S C model is in congruence with older principles of chemotherapy, in particular the classification of cell cycle-specific and non-cell cycle-specific drugs. Cycle-specific drugs, such as Ara-C, have a dramatic but exclusive effect on proliferating cells in the S-phase of the cell cycle, while non-cycle-specific drugs such as daunorubicin are active throughout the cell cycle, ki l l ing both proliferating cells and quiescent (Go) cells. The latter, however, tends to associate with much higher tissue toxicity compared to cycle-specific drugs. In light of the L S C model, an 79 updated interpretation of this is that cycle-specific drugs are effective in ki l l ing the more differentiated, actively proliferating cancer cells that usually comprise of the majority of the tumor. On the other hand, non-cycle-specific drugs are needed to eradicate the rare C S C that may be in Go. Hence in A M L , although Ara-C is useful in tumor de-bulking and alleviating symptoms, it is daunorubicin (the substrate for MDR1/BCRP1) that targets the L S C fraction and is necessary for long-term remission. Following this premise, it makes sense how patients with high MDR1/BCRP1 in CD34+CD38- should fail to achieve complete remission in my study, since daunorubicin is ineffective against the disease-maintaining compartment. 80 Table 4.1. % of CD34+CD38-, CD34+CD38+ & CD34- fractions in A M L patients. Patient % CD34+38- % CD34+38+ % CD34-CR1 12.3 41.2 27.9 CR20 0.1 63.5 11.8 CR21 0.1 0.8 78 CR25 0.1 10.4 60.6 CR30 3.3 39.9 38 CR31 0.2 75.3 1.2 CR34 4.2 21.6 36.5 N R 4 0.6 79.3 *NR6 0.05 2.7 89.3 *NR8 69.8 13.8 12.1 *NR9 16.4 51.2 18.3 *NR10 5.2 1.1 80.7 *NR14 6.1 15.2 70.7 *NR15 4.8 63.9 17.3 N R 1 6 11.9 18.6 i i i i i i i i *NR18 0.1 69.8 11.7 NR19 7.1 48.7 36.4 * - N R patients with high expression of M A K 7 / B C R P 1 in CD34+CD38-. Highlighted - patient samples suitable A N D available for functional studies. Ungated F L 3 \" H : P l Ungated FL3-H: PI 280706CR1^1 A01 280706CR1 -1A06 Event Count: 7000 Event Count: 7000 Figure 4.1. Apoptosis as detected by the Annex in V - P I assay. A M L patient cells were exposed to 0.001 ug/ml (left) and 0.5 ug/ml (right) daunorubicin for 24 hours before stained by Annexin V and PI and analyzed. The blue gate indicates the quadrant containing viable cells. Cells were significantly more viable when exposed to low drug concentrations (80% viable) than high drug concentrations (1.8% viable). 82 Substrate dehydrogenase Product NAD KADH m - ETR-reduced ETR J O V * X » S 0^ %fX*X' ^Ss^sm^__^y^ CX^ ^^ C^^ Figure 4.2. Schematic representation of conversion of M T S to formazan. Production of N A D H or N A D P H by metabolically-active cells is the main cause for conversion of M T S to formazan. Electron transfer occurs from N A D H to an electron transfer reagent (ETR) such as P M S , which subsequently causes the production of brown formazan that can be detected by a colorimetric method. Reproduced from www.ebiotrade.com. 83 — • — C E M DNR only — * — C E M P S C I u M CEM PSC 3uM — * — CEM Verapamil 5ug/ml — • — CEM Verapamil 20ug/ml CV1.0 DNR only CV1.0 PSC 1uM CV1.0 P S C 3 u M ------- CV1.0 Verapamil 5ug/ml ------- CV1.0 Verapamil 20 ug/ml Figure 4.3. Effects of PSC-833 and verapamil on daunorubicin sensitivity in C E M and CV1.0 as measured by the Annexin V-PI assay. C E M and C V 1 . 0 cells were exposed to daunorubicin at concentrations ranging from 0.001 to 0.5 ug/ml for 24 hours, with or without PSC-833 (1 u M , 3 uM) or verapamil (5 ug/ml, 20 ug/ml). Annexin V and PI were subsequently added to the cells and incubated at room temperature for 15 min in the dark. F A C S analysis was performed to determine the viability (defined as both Annexin V and PI negative). 0.001 0.01 0.1 1 DNR ug/ml 84 — • — C E M DNR only — - A — - C E M P S C 1uM • — C E M P S C 3uM — * — C E M Verapamil 5 ug/ml — • — C E M Verapamil 20 ug/ml CV1.0 DNR only CV1.0 P S C 1uM CV1.0 P S C 3uM CV1.0 Verapamil 5 ug/ml CV1.0 Verapamil 20ug/ml 0.001 0.01 0.1 1 DNR ug/ml Figure 4.4. Effects of PSC-833 and verapamil on daunorubicin sensitivity in C E M and CV1.0 as measured by the MTS assay. C E M and C V 1 . 0 cells were exposed to daunorubicin at concentrations ranging from 0.001 to 0.5 ug/ml for 24 hours, with or without PSC-833 (1 u M , 3 uM) or verapamil (5 ug/ml, 20 ug/ml). M T S was subsequently added to the cells and incubated for 2 hours at 37 °C. Color development was measured as absorbance at 490 nm. 85 PSC-833 in medium 0.01 0.1 1 10 PSC-833 uM 100 90 80 70 60 1 50 40 30 20 10 0 0.001 0.01 0.1 1 10 100 1000 Verapamil pg/ml Figure 4.5. Toxicity assay of PSC-833 and verapamil on A M L patient cells. Primary A M L patient cells were exposed to PSC-833 at concentrations ranging from 0.03 u M to 8 u M and the D M S O volume equivalent (A) or verapamil at concentrations ranging from 0.002 to 200 ug/ml (B) for 24 hours. Viabil i ty and % k i l l were determined by the Annexin V - P I assay. 86 DNR ug/ml Figure 4.6. Effect of P G P inhibition on daunorubicin sensitivity of A M L patient #18. Unsorted cells from patient #18, was subjected to daunorubicin exposure with or without P S C -833 for 24 hours before the Annexin V - P I assay. Small panel: MDR1 expression in total population of #18 (arrow) in comparison to other patients. 87 DNR ug/ml Figure 4.7. Effect of PGP inhibition on daunorubicin sensitivity of A M L patient #25. Unsorted cells from patient #25, was subjected to daunorubicin exposure with or without P S C -833 for 24 hours before the Annexin V - P I assay. Small panel: MDR1 expression in total population of #25 (arrow) in comparison to other patients. 88 A10 /ft/ /•'ft/ /, ft / f,' ft/ t' ft/ ft/- 1 - D N R only -m— P S C 0 . 3 u M - • — P S C 1 uM <>- - P S C 3uM — V e r a p a m i l 5 ug/ml -A- - Verapamil 20 ug/ml D N R u g / m l B D N R u g / m l - DNR only - P S C 0.3uM - P S C 1 uM - o - P S C 3uM — w Verapamil 5 ug/ml - -A- - Verapamil 20 ug/ml D 0.001 D N R u g / m l D N R u g / m l OO Figure 4.8. Drug sensitivity and effects of ABC modulation on CR patient #1 subpopulations. FACS-sorted CD34+38- (A) , CD34+38+ (B) and CD34- (C) and unsorted cells (D) were exposed to daunorubicin at a range of concentrations +/- A B C inhibitors verapamil and PSC-833 for 24 hours. Ce l l viability was measured by the A V - P I assay and expressed as % k i l l . Figure 4.9. Drug sensitivity and effects of ABC modulation on NR patient #9 subpopulations. FACS-sorted CD34+38- (A) , CD34+38+ (B) and CD34- (C) and unsorted cells (D) were exposed to daunorubicin at a range of concentrations +/- A B C inhibitors verapamil and PSC-833 for 24 hours. Ce l l viability was measured by the A V - P I assay and expressed as % k i l l . 0.275 n 0.250 0.225 D> 0 200 c o S 0.175 RS 3 T3 O 0.150 E o § 0.125-1 3 O JC 0.100 I J 0.075 0.050 0.025'J 0.000 N R 9 A N R 8 * NR15 N R 1 4 A A A A A A CD34+CD38-* CD34+CD38+ CD34- Unsorted Figure 4.10. Daunorubicin sensitivity of different A M L subpopulations in CR and NR patients. IC 5o of daunorubicin on A M L cells without A B C modulation was plotted for unsorted cells and each subpopulation. Square, C R patients. Triangle, N R patients. Patients with high MDR1/BCRP1 expression are identified beside their symbols. Horizontal line represents median of each group. Statistical significance in difference between C R and N R indicated by an asterisk *. 91 10T CO C TO JC! O o c: II c o If 3 O E u CD < JC o c: CO o 2 : • o li-en 1 4-o c . 3-2n N R 1 4 A N R 9 * N R 8 NR15 A A A A - W - A A A \"tTA-CD34+CD38-* C034+CD38+ CD34- Unsorted Figure 4.11. Effects of ABC modulation on drug sensitivity in different A M L subpopulations. Fold-change in daunorubicin IC50 by the highest dose of A B C inhibitor (IC50 unmodulated / IC50 inhibited) was plotted for unsorted cells and each subpopulation. Square, C R patients. Triangle, N R patients. Patients with high MDR1/BCRP1 expression are identified beside their symbols. Horizontal line represents median of each group. Statistical significance in difference between C R and N R indicated by an asterisk *. 92 V Conclusion and future prospects 5.1 - Overall discussion and conclusion Drug resistance has been a major obstacle in cancer treatment. Some forms of resistance, which result from alteration of a specific drug target or loss of the surface receptor for a given drug, are specific to a small number of related drugs and can be overcome by combination therapy 1 6 4. The emergence of M D R , however, which involves resistance to multiple unrelated drugs, poses a far more difficult problem for which combination therapy is not a solution. To effectively reverse M D R , an understanding of the underlying mechanisms is necessary. This thesis sought to address the problem of M D R in A M L by investigating the role of A B C transporters, a classic contributor to M D R in cell lines. In Chapter 2,1 applied the RT-Real Time P C R assay to profile and compare the expression levels of the 47 known human A B C transporters between the A M L responders and non-responders to initial chemotherapy. This is the first systematic study on the prognostic significance of the full A B C transporter superfamily in A M L . I first asked whether some of these transporters, especially the MDR-related transporters, might be present in high levels in the bulk leukemic cells from N R patients. However, I found no consistent difference in the expression of any A B C gene between the bulk samples of the two patient groups. Based on the L S C model, I then hypothesized that expression differences might be hidden within the most primitive subpopulations. This was addressed in Chapter 3, where subpopulations along the leukemic functional hierarchy were isolated and profiled for expression of MDR-related A B C transporters. High MDR1 and/or BCRP1 expression in the primitive CD34+CD38- fraction was found to be consistently associated with N R outcomes. Neither the more committed CD34+CD38+ 93 progenitors nor the mature CD34- cells provided such an association. This is the first indication of a possible prognostic value of A B C transporter expression in the CD34+CD38- cells, the small subpopulation that is thought to be responsible for maintenance of the leukemia in patients and for relapses. In Chapter 4,1 further investigated the ex vivo drug sensitivity of patient subpopulations and functional relevance of A B C transporters. Using the Annexin V - P I apoptotic assay, I confirmed that ABC-dependent resistance, corresponding to high MDR1IBCRP1 expression and reversibility by ABC-specif ic inhibitors, is common among non-responsive patients, particularly those with a normal karyotype. This suggested that the expressed P G P / B C R P 1 are actively extruding drugs from the L S C and thus may make a significant contributor to intrinsic drug resistance in vivo. M y study demonstrated that the properties of primitive subpopulations may facilitate better understanding of how a cancer operates than examination of the properties of the bulk cells. In particular, the drug response of the primitive CD34+CD38- A M L subpopulation seems a more accurate predictor of treatment outcome than the bulk leukemic population. This is in line with the L S C model and calls for a continued research focus on this small fraction of cells. O f note, the very low m R N A levels observed (especially for BCRP1 which was consistently below the biological reference line) even in the relative \"high\" expressers hints towards a further subset within the CD34+CD38- fraction that expresses much higher levels. Hence although CD34+CD38- marks a subpopulation enriched in tumorigenicity that is important for prognostic and therapeutic purposes, it is likely a heterogeneous group of cells in itself. Figure 5.1 shows the different proposed models of drug resistance in cancer3. The conventional model of cancer drug resistance (Figure 5.1 A ) conceives that a number of cells 94 acquired mutations that confer drug resistance. These cells outgrow the others to form a new resistant tumor population following chemotherapy. In the C S C model (Figure 5.IB), the C S C (CD34+CD38- in A M L ) is inherently drug-resistant. A t least some of these survive chemotherapy to regenerate a tumor similar to the original disease. A variation of the C S C model, the acquired-resistance C S C model (Figure 5.1C), posits that additional mutations in the surviving C S C generate a drug-resistant tumor. In the last intrinsic resistance model (Figure 5.ID), both the C S C and its descendants are intrinsically drug-resistant. Therapies are ineffective, resulting in uncontrolled tumor growth. The results of my study are consistent with the C S C model and the acquired-resistance model in that CD34+CD38- cells with intrinsic high MDR1/BCRP1 expression and activity can survive chemotherapy. These can drive tumor regeneration that is rapid enough to result in persistent disease, hence failure to achieve complete remission (no detectable leukemic cells). In this respect, non-responsiveness may be viewed as a very fast relapse within the chemotherapeutic regimen. During or following initial therapy, the primitive fraction may acquire further mutations that confer M D R to its descendants as well (acquired resistance model). A B C transporter expression may also be induced in the general population upon drug exposure, producing a drug resistant disease often seen in A M L relapse. The presence of drug resistance mechanisms such as A B C transporters in the L S C is a more accurate predictor of response than the de novo size of this fraction. Thus not all L S C are resistant enough to withstand the initial high-dose chemotherapy given to A M L patients. Those patients without high MDR1IBCRP1 expression in CD34+CD38- cells responded well to initial therapy, while those that apparently retained this normal H S C defense system did not respond. Clearly, A M L is a heterogeneous disease and the L S C that originates it possesses varying properties in each patient. 95 It is interesting that non-responders with high A B C expression (MDR1IBCRP1) appear well separated from the other non-responders. Practically, this may represent a convenient means to distinguish the ABC-dependent (high expression) from the ABC-independent (low expression) N R patients. Conceptually, this segregation also demonstrates the value in identifying outliers within a pre-defined group, an approach that has recently been successfully applied by Tomlins and colleagues to discover the T M P R S S 2 - E T S fusion in prostate cancer 1 6 5 . Cancer as a dynamic disease can display heterogeneity within both itself (as exemplified by expression differences observed among subpopulations) and among its assigned \"type\" and \"group\", such as responders and non-responders in this study. In fact, it is important to bear in mind that these \"subcategories\" merely serve as working definitions and hence is subjected to redefinition or further division as more information became known. In this case, patients within the N R group appear to fall into the ABC-dependent and ABC-independent subcategories that may require different strategies in treatment (Section 5.2). Overall, my work fulfilled all three objectives: to profile expression levels of the A B C transporter superfamily in total A M L patient samples, to compare expression of MDR-related A B C transporters in subpopulations along the leukemic hierarchy, and to investigate the drug resistance of subpopulations and functional activity of A B C transporters. Based on my studies, the following conclusions are reached. First, expression of A B C transporters in de novo unfractionated patient samples is not predictive of response. Second, high intrinsic levels of MDR1 and/or BCRP1 in the primitive CD34+CD38- fraction are associated with poor response to initial chemotherapy. Third, high ex vivo functional activity correlates with high expression levels of MDR11BCRP1 in CD34+CD38- cells. Fourth, ABC-dependent drug resistance in CD34+CD38- is common among non-responders, especially those with normal cytogenetics, and 96 is reversible via ABC-inhibi t ion. Taken as a whole, my studies suggest a prognostic significance of A B C transporters in the primitive CD34+CD38- leukemic subpopulation, and support a modified approach in investigating the value of A B C modulating agents in A M L , as discussed in the next section. 97 5.2 - Future Prospects Previous clinical trials of P G P inhibition, such as second-generation inhibitor PSC-833, yielded largely negative results in A M L , with high toxicity and lack of significant improvement in outcome 8 3 ' 8 4 . M y findings suggest a functional role of A B C transporters in the primitive, disease-maintaining fraction of some N R A M L patients. Based on this initial study, disappointing results from clinical A B C inhibition can be viewed in a new light. MDR1IVG? expression and activity was low in all C R samples and half of the N R samples, to whom P G P inhibition is unlikely to have a significant effect. Because past studies did not distinguish these patients from the small subset of ABC-expressing non-responders, clinical benefits of A B C modulation was probably diluted and underestimated. M y work raises the importance and feasibility of pre-screening patients for targeted therapy in A M L . While A B C activity may not be the major mechanism of M D R in all non-responders, it may be possible to identify those where A B C transporter expression is a major factor for drug resistance and apply appropriate therapeutic intervention. Figure 5.2 outlines a proposed scheme for predicting treatment response and overcoming M D R . Before a patient is subjected to chemotherapy, levels of MDR1/BCRP1 in the CD34+CD38- fraction can be quickly determined. Patients with high expression of either A B C transporter are predicted to be A B C -dependent non-responders (to chemotherapy alone). These are likely to significantly benefit from the combination of conventional chemotherapy and A B C inhibitors, outweighing the high toxicity effects. The low-expressers are predicted to be either responsive to chemotherapy alone or ABC-independent non-responders and may be spared of the toxicity of A B C inhibition. A s 98 more MDR-related factors (factor X ) are characterized, it may be possible to incorporate additional testing and targeted-therapeutic options to the treatment scheme. A s discussed in the introduction, cytogenetic aberrations are the major prognostic factor in A M L . But for patients who are karyotypically normal (half of all patients), only a few molecular biomarkers have been investigated (discussed in Section 1.3). Most of the patients (25/31) profiled for A B C transporter expression in my study belonged to this category. In search for other prognostic markers especially for this cytogenetically normal group, I have identified high expression and activity of MDR1/BCRP1 in CD34+CD38- cells as one significant predictive factor of poor response. Although this was a common M D R mechanism among the patients tested, however, about half of the non-responders appear relatively independent of A B C activity. Evidently, A B C transporters are not the only determinants for treatment outcome. Technical advancements in recent years have opened up new possibilities to study the molecular genetics of a disease. One such technique, comparative genomic hybridization (CGH) , can be used to detect micro amplifications and deletions in D N A . Recent construction by Lam and colleagues of a high-resolution array with over 30,000 B A C clones that cover the entire human genome allows analysis of copy-number changes in the global genome, revealing down to gene-size alterations (Figure 5.3) 1 6 6 . While half of the patients are cytogenetically normal, there may be micro-scale alterations in their cancer genome not detected by karyotyping. It is possible that these micro-amplifications and deletions recur at the loci of novel genes that account for the lack of response to chemotherapy. A s a further extension of my thesis, I would seek to explore genomic changes occurring in A M L samples using the novel array-CGH technique. I hypothesize that micro genomic differences can be detected between drug sensitive and resistant patients. The objective of this 99 study would be two-fold: first, to determine whether high expression of MDR1 and BCRP1 in N R cells is due to micro-amplifications; second, to see i f additional amplified markers can be identified in a genomic signature predictive of clinical response. Eight C R patients and nine N R patients (17 in total) from previous studies were selected for arrayCGH, 12 of which have normal cytogenetics. Microalterations not reported by cytogenetics have been found in each sample. A s shown in Figure 5.4, micro-gains and losses were observed in the different chromosomes of patient NR#3 (A). For example, there was a fragment loss on the p arm of chromosome 9 (B) not detected by cytogenetics, and even smaller but clear micro-amplifications and micro-deletions on chromosome 5q. Some of these gains and losses were recurrent in the samples tested, which may indicate genes critical to leukemogenesis. Furthermore, some differences are seen between the C R and N R samples. Further work on these may provide insight into finding possible prognostic patterns in the de novo leukemic genome to predict treatment response, and identifying novel genes that contribute to the pathogenesis and/or drug response of A M L . 100 a MDR eels I) Tumou- stem cell Figure 5.1. Models of tumor drug resistance. A , in the conventional model of cancer drug resistance, rare cells with genetic alterations that confer M D R form a resistance clone (yellow). These cells survive chemotherapy and proliferate, giving rise to relapsed disease with offspring of the resistant clone. B , in the C S C model, the C S C (red) that has protective mechanisms survive chemotherapy while the committed cells (blue) are killed. The C S C repopulates a functional tumor hierarchy. C, in the acquired resistance C S C model, C S C surviving chemotherapy accumulates mutations (yellow) conferring a resistant phenotype in also the committed descendants (purple). D , in the intrinsic resistance model, both the stem cells and the committed cells are inherently resistant. Therapy has no effect, resulting in tumor expansion. Reproduced from Dean et al, Nature Reviews/Cancer, 2005 3 . 101 ^ % D e n 0 V 0 A M L RT-PCR L o w M D R 1 & B C R P 1 in CD34+CD38-High M D R 1 / B C R P 1 in CD34+CD38-CR or ABC-independent Test other Jf factors X ABC-dependent NR Chemotherapy + 4m A B C inhibition \" * Eradication of A M L X-dependent NR Chemo-therapy Chemotherapy + X inhibition Eradication of A M L Figure 5.2. Predict ing response and overcoming M D R . A t diagnosis, A M L patients can be tested for MDR] and BCRP1 expression in the CD34+CD38- cells. Those with high expression are likely ABC-dependent non-responders (NR) to chemotherapy alone and w i l l benefit from the combination of chemotherapy and A B C inhibitors. Patients with low A B C expression are either responders who wi l l achieve complete remission with chemotherapy alone (CR) or A B C -independent non-responders who has alternate drug resistance mechanisms. As more knowledge becomes available, patients may be tested for other MDR-related factors (X) and appropriate therapeutic intervention may be applied for eradication of the disease. 102 \"T\"iJ ITS out* F^ t &fN&f& 1*^0(61 Figure 5.3. Principles of array comparative genomic hybridization. A , normal and tumor D N A samples are isolated and used to create fluorescently labeled probes, commonly with cyanine 3 (Cy3, green) and cyanine 5 (Cy5, red) dyes. The probes are pooled and competitively cohybridized to a glass slide spotted with a known array of mapped genomic clones. The arrays are analyzed with a microarray scanner, producing an image that is used to assess the log2 ratios of the Cy5 to Cy3 intensities for each clone. B , A log2 ratio profile is assembled to determine relative copy number changes between the cancer and the normal samples. Each dot on the graph represents a clone. Values to the left of \"0\" indicate a loss of a genomic region, while values to the right indicate a gain or amplification. Values at \"0\" indicate no change. Reproduced from Davies et al, Chromosome Research, 2005. 103 Figure 5.4. C G H karyogram pf patient NR#3. D N A from patient NR#3 was isolated and subjected to arrayCGH. The raw data was analyzed using the SeeGH program to generate a karygram for every chromosome. Each dot is a B A C clone representing a small section of D N A . Values to the left of the centre line (\"0\") indicate a loss, and values to the right indicate a gain. A , microalterations on the 23 chromosome pairs. B , Fragment loss on chromosome 9p. C, micro-amplifications and micro-deletions on chromosome 5q. 104 Bibliography 1. Dean M , Rzhetsky A , Allikmets R. The human ATP-binding cassette ( A B C ) transporter superfamily. Genome Res. 2001;11:1156-1166. 2. Huntly B J , Gil l i land D G . Leukaemia stem cells and the evolution of cancer-stem-cell research. Nat Rev Cancer. 2005;5:311-321. 3. Dean M , Fojo T, Bates S. Tumour stem cells and drug resistance. Nat Rev Cancer. 2005;5:275-284. 4. De l Poeta G , Venditti A , Stasi R, et al. P-glycoprotein and terminal transferase expression identify prognostic subsets within cytogenetic risk classes in acute myeloid leukemia. Leuk Res. 1999;23:451-465. 5. Chang G , Roth C B . Structure of M s b A from E . coli : a homolog of the multidrug resistance A T P binding cassette ( A B C ) transporters. Science. 2001 ;293:1793-1800. 6. Grimwade D , Walker H , Oliver F, et al. The importance of diagnostic cytogenetics on outcome in A M L : analysis of 1,612 patients entered into the M R C A M L 10 trial. The Medical Research Council Adult and Children's Leukaemia Working Parties. Blood. 1998;92:2322-2333. 7. Wuchter C, Leonid K , Ruppert V , et al. Cl inical significance of P-glycoprotein expression and function for response to induction chemotherapy, relapse rate and overall survival in acute leukemia. Haematologica. 2000;85:711-721. 8. Canadian Cancer Statistics. 2006. 9. Ozols R F , Herbst R S , Colson Y L , et al. Clinical cancer advances 2006: major research advances in cancer treatment, prevention, and screening—a report from the American Society of Clinical Oncology. J C l i n Oncol. 2007;25:146-162. 10. Collins I, Workman P. New approaches to molecular cancer therapeutics. Nat Chem B i o l . 2006;2:689-700. 11. Druker B J , Sawyers C L , Kantarjian H , et al. Activi ty of a specific inhibitor of the B C R -A B L tyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblastic leukemia with the Philadelphia chromosome. N Engl J Med . 2001;344:1038-1042. 12. Druker B J , Talpaz M , Resta D J , et al. Efficacy and safety of a specific inhibitor of the B C R - A B L tyrosine kinase in chronic myeloid leukemia. N Engl J Med . 2001;344:1031-1037. 13. Kantarjian H M , Talpaz M , Giles F, O'Brien S, Cortes J. New insights into the pathophysiology of chronic myeloid leukemia and imatinib resistance. A n n Intern Med . 2006;145:913-923. 14. Stirewalt D L , Radich JP. The role of F L T 3 in haematopoietic malignancies. Nat Rev Cancer. 2003;3:650-665. 15. Whitman SP, Archer K J , Feng L , et al. Absence of the wild-type allele predicts poor prognosis in adult de novo acute myeloid leukemia with normal cytogenetics and the internal tandem duplication of F L T 3 : a cancer and leukemia group B study. Cancer Res. 2001 ;61:7233-7239. 16. Frohling S, Schlenk R F , Breitruck J, et al. Prognostic significance of activating F L T 3 mutations in younger adults (16 to 60 years) with acute myeloid leukemia and normal cytogenetics: a study of the A M L Study Group U l m . Blood. 2002;100:4372-4380. 17. Kainz B , Heintel D , Marculescu R, et al. Variable prognostic value of F L T 3 internal tandem duplications in patients with de novo A M L and a normal karyotype, t(15;17), t(8;21) or inv(16). Hematol J. 2002;3:283-289. 105 18. Smith B D , Levis M , Beran M , et al. Single-agent CEP-701, a novel F L T 3 inhibitor, shows biologic and clinical activity in patients with relapsed or refractory acute myeloid leukemia. Blood. 2004;103:3669-3676. 19. Stone R M , DeAngelo D J , Kl imek V , et al. Patients with acute myeloid leukemia and an activating mutation in F L T 3 respond to a small-molecule F L T 3 tyrosine kinase inhibitor, PKC412 . Blood. 2005;105:54-60. 20. Bonnet D , Dick JE. Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med. 1997;3:730-737. 21. Mitus A , Rosenthal D . The Adult Leukemias. Clinical Oncology. 1999. 22. Schoch C, Haferlach T. Cytogenetics in acute myeloid leukemia. Curr Oncol Rep. 2002;4:390-397. 23. Mrozek K , Heinonen K , Bloomfield C D . Prognostic value of cytogenetic findings in adults with acute myeloid leukemia. Int J Hematol. 2000;72:261-271. 24. Mrozek K , Heerema N A , Bloomfield C D . Cytogenetics in acute leukemia. Blood Rev. 2004;18:115-136. 25. DeVita J V , Hellman S, Rosenberg S. Cancer Principles and practice of oncology. 200 l;2nd edition. 26. Miyoshi H , Shimizu K , K o z u T, Maseki N , Kaneko Y , Ohki M . t(8;21) breakpoints on chromosome 21 in acute myeloid leukemia are clustered within a limited region of a single gene, A M L 1 . Proc Natl Acad Sci U S A . 1991;88:10431-10434. 27. Warrell RP , Jr., de The H , Wang Z Y , Degos L . Acute promyelocytic leukemia. N Engl J Med. 1993;329:177-189. 28. Byrd JC, Mrozek K , Dodge R K , et al. Pretreatment cytogenetic abnormalities are predictive of induction success, cumulative incidence of relapse, and overall survival in adult patients with de novo acute myeloid leukemia: results from Cancer and Leukemia Group B ( C A L G B 8461). Blood. 2002;100:4325-4336. 29. Grimwade D , Walker H , Harrison G , et al. The predictive value of hierarchical cytogenetic classification in older adults with acute myeloid leukemia ( A M L ) : analysis of 1065 patients entered into the United Kingdom Medical Research Council A M L 1 1 trial. Blood. 2001;98:1312-1320. 30. Slovak M L , Kopecky K J , Cassileth P A , et al. Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study. Blood. 2000;96:4075-4083. 31. Fenaux P, Chevret S, Guerci A , et al. Long-term follow-up confirms the benefit of al l -trans retinoic acid in acute promyelocytic leukemia. European A P L group. Leukemia. 2000;14:1371-1377. 32. Tallman M S , Andersen JW, Schiffer C A , et al. All-trans retinoic acid in acute promyelocytic leukemia: long-term outcome and prognostic factor analysis from the North American Intergroup protocol. Blood. 2002; 100:4298-4302. 33. Shen Z X , Chen G Q , N i J H , et al. Use of arsenic trioxide (As203) in the treatment of acute promyelocytic leukemia ( A P L ) : II. Cl inical efficacy and pharmacokinetics in relapsed patients. Blood. 1997;89:3354-3360. 34. Soignet S L , Frankel SR, Douer D , et al. United States multicenter study of arsenic trioxide in relapsed acute promyelocytic leukemia. J C l in Oncol. 2001;19:3852-3860. 106 35. Raffoux E , Rousselot P, Poupon J, et al. Combined treatment with arsenic trioxide and all-trans-retinoic acid in patients with relapsed acute promyelocytic leukemia. J C l i n Oncol. 2003;21:2326-2334. 36. Shen Z X , Shi Z Z , Fang J, et al. All-trans retinoic acid/As203 combination yields a high quality remission and survival in newly diagnosed acute promyelocytic leukemia. Proc Natl Acad Sci U S A . 2004;101:5328-5335. 37. Garson O M , Hagemeijer A , Sakurai M , et al. Cytogenetic studies of 103 patients with acute myelogenous leukemia in relapse. Cancer Genet Cytogenet. 1989;40:187-202. 38. Frohling S, Skelin S, Liebisch C, et al. Comparison of cytogenetic and molecular cytogenetic detection of chromosome abnormalities in 240 consecutive adult patients with acute myeloid leukemia. J C l i n Oncol. 2002;20:2480-2485. 39. Trost D , Hildebrandt B , Mul ler N , Germing U , Royer-Pokora B . Hidden chromosomal aberrations are rare in primary myelodysplastic syndromes with evolution to acute myeloid leukaemia and normal cytogenetics. Leuk Res. 2004;28:171-177. 40. Casas S, Aventin A , Fuentes F, et al. Genetic diagnosis by comparative genomic hybridization in adult de novo acute myelocytic leukemia. Cancer Genet Cytogenet. 2004;153:16-25. 41.. Bullinger L , Dohner K , Bair E , et al. Use of gene-expression profiling to identify prognostic subclasses in adult acute myeloid leukemia. N Engl J Med. 2004;350:1605-1616. 42. V a l k PJ, Verhaak R G , Beijen M A , et al. Prognostically useful gene-expression profiles in acute myeloid leukemia. N Engl J Med. 2004;350:1617-1628. 43. Marcucci G , Mrozek K , Bloomfield C D . Molecular heterogeneity and prognostic biomarkers in adults with acute myeloid leukemia and normal cytogenetics. Curr Opin Hematol. 2005;12:68-75. 44. Hess JL . M L L : a histone methyltransferase disrupted in leukemia. Trends M o l Med . 2004;10:500-507. 45. Frohling S, Schlenk R F , Stolze I, et al. C E B P A mutations in younger adults with acute myeloid leukemia and normal cytogenetics: prognostic relevance and analysis of cooperating mutations. J C l i n Oncol. 2004;22:624-633. 46. Thiede C, Koch S, Creutzig E , et al. Prevalence and prognostic impact of N P M 1 mutations in 1485 adult patients with acute myeloid leukemia ( A M L ) . Blood. 2006;107:4011-4020. 47. Baldus C D , Tanner S M , Ruppert A S , et al. B A A L C expression predicts clinical outcome of de novo acute myeloid leukemia patients with normal cytogenetics: a Cancer and Leukemia Group B Study. Blood. 2003;102:1613-1618. 48. Childs S, L i n g V . The M D R superfamily of genes and its biological implications. Important A d v Oncol. 1994:21-36. 49. Higgins C F . A B C transporters: from microorganisms to man. Annu Rev Cel l B i o l . 1992;8:67-113. 50. Walker JE, Saraste M , Runswick M J , Gay N J . Distantly related sequences in the alpha-and beta-subunits of A T P synthase, myosin, kinases and other ATP-requiring enzymes and a common nucleotide binding fold. Embo J. 1982;1:945-951. 51. Hyde SC, Emsley P, Hartshorn M J , et al. Structural model of ATP-binding proteins associated with cystic fibrosis, multidrug resistance and bacterial transport. Nature. 1990;346:362-365. 107 52. Dawson RJ , Locher K P . Structure of a bacterial multidrug A B C transporter. Nature. 2006;443:180-185. 53. Juliano R L , L ing V . A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta. 1976;455:152-162. 54. Rao V V , Dahlheimer J L , Bardgett M E , et al. Choroid plexus epithelial expression of M D R 1 P glycoprotein and multidrug resistance-associated protein contribute to the blood-cerebrospinal-fluid drug-permeability barrier. Proc Natl Acad Sci U S A . 1999;96:3900-3905. 55. Borst P, Evers R, K o o l M , Wijnholds J. The multidrug resistance protein family. Biochim Biophys Acta. 1999;1461:347-357. 56. Gottesman M M , Ambudkar S V . Overview: A B C transporters and human disease. J Bioenerg Biomembr. 2001;33:453-458. 57. Schinkel A H , Smit JJ, van Tellingen O, et al. Disruption of the mouse mdr la P-glycoprotein gene leads to a deficiency in the blood-brain barrier and to increased sensitivity to drugs. Cel l . 1994;77:491-502. 58. Borst P, Evers R, K o o l M , Wijnholds J. A family of drug transporters: the multidrug resistance-associated proteins. J Natl Cancer Inst. 2000;92:1295-1302. 59. Wijnholds J, deLange E C , Scheffer G L , et al. Multidrug resistance protein 1 protects the choroid plexus epithelium and contributes to the blood-cerebrospinal fluid barrier. J C l i n Invest. 2000;105:279-285. 60. Doyle L A , Yang W , Abruzzo L V , et al. A multidrug resistance transporter from human M C F - 7 breast cancer cells. Proc Natl Acad Sci U S A . 1998;95:15665-15670. 61. Miyake K , Mick ley L , Litman T, et al. Molecular cloning of c D N A s which are highly overexpressed in mitoxantrone-resistant cells: demonstration of homology to A B C transport genes. Cancer Res. 1999;59:8-13. 62. Jonker JW, Buitelaar M , Wagenaar E , et al. The breast cancer resistance protein protects against a major chlorophyll-derived dietary phototoxin and protoporphyria. Proc Natl Acad Sci U S A . 2002;99:15649-15654. 63. Krishnamurthy P, Ross D D , Nakanishi T, et al. The stem cell marker B c r p / A B C G 2 enhances hypoxic cell survival through interactions with heme. J B i o l Chem. 2004;279:24218-24225. 64. Fojo A T , Ueda K , Slamon D J , Poplack D G , Gottesman M M , Pastan I. Expression of a multidrug-resistance gene in human tumors and tissues. Proc Natl Acad Sci U S A . 1987;84:265-269. 65. Goldstein L J , Galski H , Fojo A , et al. Expression of a multidrug resistance gene in human cancers. J Natl Cancer Inst. 1989;81:116-124. 66. Roninson IB. From amplification to function: the case of the M D R 1 gene. Mutat Res. 1992;276:151-161. 67. Gottesman M M , Fojo T, Bates SE. Multidrug resistance in cancer: role of A T P -dependent transporters. Nat Rev Cancer. 2002;2:48-58. 68. Ross D D , Karp JE, Chen TT, Doyle L A . Expression of breast cancer resistance protein in blast cells from patients with acute leukemia. Blood. 2000;96:365-368. 69. Diestra JE, Scheffer G L , Catala I, et al. Frequent expression of the multi-drug resistance-associated protein B C R P / M X R / A B C P / A B C G 2 in human tumours detected by the B X P - 2 1 monoclonal antibody in paraffin-embedded material. J Pathol. 2002;198:213-219. 70. Kawabata S, Oka M , Soda H , et al. Expression and functional analyses of breast cancer resistance protein in lung cancer. C l i n Cancer Res. 2003;9:3052-3057. 108 71. Mack JT, Beljanski V , Tew K D , Townsend D M . The ATP-binding cassette transporter A B C A 2 as a mediator of intracellular trafficking. Biomed Pharmacother. 2006;60:587-592. 72. Childs S, Y e h R L , Hui D , L ing V . Taxol resistance mediated by transfection of the liver-specific sister gene of P-glycoprotein. Cancer Res. 1998;58:4160-4167. 73. Allikmets R, Schriml L M , Hutchinson A , Romano-Spica V , Dean M . A human placenta-specific ATP-binding cassette gene ( A B C P ) on chromosome 4q22 that is involved in multidrug resistance. Cancer Res. 1998;58:5337-5339. 74. Komatani H , Kotani H , Hara Y , et al. Identification of breast cancer resistant protein/mitoxantrone resistance/placenta-specific, ATP-binding cassette transporter as a transporter of NB-506 and J-107088, topoisomerase I inhibitors with an indolocarbazole structure. Cancer Res. 2001;61:2827-2832. 75. Maliepaard M , Scheffer G L , Faneyte IF, et al. Subcellular localization and distribution of the breast cancer resistance protein transporter in normal human tissues. Cancer Res. 2001;61:3458-3464. 76. Cole SP, Bhardwaj G , Gerlach J H , et al. Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line. Science. 1992;258:1650-1654. 77. Loe D W , Deeley R G , Cole SP. Characterization of vincristine transport by the M(r) 190,000 multidrug resistance protein (MRP) : evidence for cotransport with reduced glutathione. Cancer Res. 1998;58:5130-5136. 78. Jedlitschky G , Leier I, Buchholz U , Barnouin K , Kurz G , Keppler D . Transport of glutathione, glucuronate, and sulfate conjugates by the M R P gene-encoded conjugate export pump. Cancer Res. 1996;56:988-994. 79. Mul ler M , Meijer C , Zaman GJ , et al. Overexpression of the gene encoding the multidrug resistance-associated protein results in increased ATP-dependent glutathione S-conjugate transport. Proc Natl Acad Sci U S A . 1994;91:13033-13037. 80. Marbeuf-Gueye C, Broxterman H J , Dubru F , Priebe W , Gamier-Suillerot A . Kinetics of anthracycline efflux from multidrug resistance protein-expressing cancer cells compared with P-glycoprotein-expressing cancer cells. M o l Pharmacol. 1998;53:141-147. 81. Renes J, de Vries E G , Nienhuis E F , Jansen P L , Mul ler M . A T P - and glutathione-dependent transport of chemotherapeutic drugs by the multidrug resistance protein M R P 1 . B r J Pharmacol. 1999;126:681-688. 82. Tidefelt U , Liliemark J, Gruber A , et al. P-Glycoprotein inhibitor valspodar (PSC 833) increases the intracellular concentrations of daunorubicin in vivo in patients with P-glycoprotein-positive acute myeloid leukemia. J C l i n Oncol. 2000;18:1837-1844. 83. Greenberg P L , Lee SJ, Advani R, et al. Mitoxantrone, etoposide, and cytarabine with or without valspodar in patients with relapsed or refractory acute myeloid leukemia and high-risk myelodysplastic syndrome: a phase III trial (E2995). J C l in Oncol. 2004;22:1078-1086. 84. Baer M R , George S L , Dodge R K , et al. Phase 3 study of the multidrug resistance modulator PSC-833 in previously untreated patients 60 years of age and older with acute myeloid leukemia: Cancer and Leukemia Group B Study 9720. Blood. 2002;100:1224-1232. 85. Chan H S , DeBoer G , Thiessen JJ, et al. Combining cyclosporin with chemotherapy controls intraocular retinoblastoma without requiring radiation. C l i n Cancer Res. 1996;2:1499-1508. 86. Ambudkar S V , Dey S, Hrycyna C A , Ramachandra M , Pastan I, Gottesman M M . Biochemical, cellular, and pharmacological aspects of the multidrug transporter. Annu Rev Pharmacol Toxicol . 1999;39:361-398. 109 87. Marie JP, Zittoun R, Sikic BI . Multidrug resistance (mdrl) gene expression in adult acute leukemias: correlations with treatment outcome and in vitro drug sensitivity. Blood. 1991;78:586-592. 88. Pirker R, Wallner J, Geissler K , et al. M D R 1 gene expression and treatment outcome in acute myeloid leukemia. J Natl Cancer Inst. 1991;83:708-712. 89. Leith C P , Kopecky K J , Chen I M , et al. Frequency and clinical significance of the expression of the multidrug resistance proteins MDRl/P-glycoprote in , M R P 1 , and L R P in acute myeloid leukemia: a Southwest Oncology Group Study. Blood. 1999;94:1086-1099. 90. Zochbauer S, Gsur A , Brunner R, Kyrle P A , Lechner K , Pirker R. P-glycoprotein expression as unfavorable prognostic factor in acute myeloid leukemia. Leukemia. 1994;8:974-977. 91. Borg A G , Burgess R, Green L M , Scheper RJ , Y i n J A . Overexpression of lung-resistance protein and increased P-glycoprotein function in acute myeloid leukaemia cells predict a poor response to chemotherapy and reduced patient survival. B r J Haematol. 1998;103:1083-1091. 92. Lyttelton M P , Hart S, Ganeshaguru K , Hoffbrand A V , Mehta A B . Quantitation of multidrug resistant M D R 1 transcript in acute myeloid leukaemia by non-isotopic quantitative cDNA-polymerase chain reaction. B r J Haematol. 1994;86:540-546. 93. Wood P, Burgess R, MacGregor A , Y i n JA. P-glycoprotein expression on acute myeloid leukaemia blast cells at diagnosis predicts response to chemotherapy and survival. B r J Haematol. 1994;87:509-514. 94. Zhou D C , Zittoun R, Marie JP. Expression of multidrug resistance-associated protein (MRP) and multidrug resistance (MDR1) genes in acute myeloid leukemia. Leukemia. 1995;9:1661-1666. 95. Mich ie l i M , Damiani D , Ermacora A , et al. P-glycoprotein, lung resistance-related protein and multidrug resistance associated protein in de novo acute non-lymphocytic leukaemias: biological and clinical implications. B r J Haematol. 1999;104:328-335. 96. Broxterman H J , Sonneveld P, Pieters R, et al. Do P-glycoprotein and major vault protein ( M V P / L R P ) expression correlate with in vitro daunorubicin resistance in acute myeloid leukemia? Leukemia. 1999;13:258-265. 97 r Smyth M J , Krasovskis E , Sutton V R , Johnstone R W . The drug efflux protein, P-glycoprotein, additionally protects drug-resistant tumor cells from multiple forms of caspase-dependent apoptosis. Proc Natl Acad Sci U S A . 1998;95:7024-7029. 98. Johnstone R W , Cretney E , Smyth M J . P-glycoprotein protects leukemia cells against caspase-dependent, but not caspase-independent, cell death. Blood. 1999;93:1075-1085. 99. Robinson L J , Roberts W K , L ing TT , Lamming D , Sternberg SS, Roepe P D . Human M D R 1 protein overexpression delays the apoptotic cascade in Chinese hamster ovary fibroblasts. Biochemistry. 1997;36:11169-11178. 100. van der K o l k D M , de Vries E G , van Putten W J , et al. P-glycoprotein and multidrug resistance protein activities in relation to treatment outcome in acute myeloid leukemia. C l i n Cancer Res. 2000;6:3205-3214. 101. Legrand O, Simonin G , Beauchamp-Nicoud A , Zittoun R, Marie JP. Simultaneous activity of M R P 1 and Pgp is correlated with in vitro resistance to daunorubicin and with in vivo resistance in adult acute myeloid leukemia. Blood. 1999;94:1046-1056. 102. Legrand O, Simonin G , Perrot J Y , Zittoun R, Marie JP. Pgp and M R P activities using ca lce in -AM are prognostic factors in adult acute myeloid leukemia patients. Blood. 1998;91:4480-4488. 110 103. Filipits M , Suchomel R W , Zochbauer S, Brunner R, Lechner K , Pirker R. Multidrug resistance-associated protein in acute myeloid leukemia: N o impact on treatment outcome. C l i n Cancer Res. 1997;3:1419-1425. 104. Damiani D , Tiribell i M , Calistri E , et al. The prognostic value of P-glycoprotein ( A B C B ) and breast cancer resistance protein ( A B C G 2 ) in adults with de novo acute myeloid leukemia with normal karyotype. Haematologica. 2006;91:825-828. 105. Steinbach D , Sell W , Voigt A , Hermann J, Zint l F , Sauerbrey A . B C R P gene expression is associated with a poor response to remission induction therapy in childhood acute myeloid leukemia. Leukemia. 2002;16:1443-1447. 106. van den Heuvel-Eibrink M M , Wiemer E A , Prins A , et al. Increased expression of the breast cancer resistance protein (BCRP) in relapsed or refractory acute myeloid leukemia ( A M L ) . Leukemia. 2002;16:833-839. 107. Abbott B L , Colapietro A M , Barnes Y , Marini F, Andreeff M , Sorrentino B P . L o w levels of A B C G 2 expression in adult A M L blast samples. Blood. 2002;100:4594-4601. 108. van der K o l k D M , Vellenga E , Scheffer G L , et al. Expression and activity of breast cancer resistance protein (BCRP) in de novo and relapsed acute myeloid leukemia. Blood. 2002;99:3763-3770. 109. van der Pol M A , Broxterman H J , Pater J M , et al. Function of the A B C transporters, P-glycoprotein, multidrug resistance protein and breast cancer resistance protein, in minimal residual disease in acute myeloid leukemia. Haematologica. 2003;88:134-147. 110. Reya T, Morrison SJ, Clarke M F , Weissman IL. Stem cells, cancer, and cancer stem cells. Nature. 2001;414:105-111. 111. Dalerba P, Cho R W , Clarke M F . Cancer Stem Cells: Models and Concepts. Annu Rev Med. 2006. 112. Huntly B J , Gil l i land D G . Cancer biology: summing up cancer stem cells. Nature. 2005;435:1169-1170. 113. Buick R N , Minden M D , McCul loch E A . Self-renewal in culture of proliferative blast progenitor cells in acute myeloblastic leukemia. Blood. 1979;54:95-104. 114. Turhan A G , Lemoine F M , Debert C, et al. Highly purified primitive hematopoietic stem cells are P M L - R A R A negative and generate nonclonal progenitors in acute promyelocytic leukemia. Blood. 1995;85:2154-2161. 115. Castor A , Nilsson L , Astrand-Grundstrom I, et al. Distinct patterns of hematopoietic stem cell involvement in acute lymphoblastic leukemia. Nat Med. 2005; 11:630-637. 116. Brown D , Kogan S, Lagasse E , et al. A P M L R A R a l p h a transgene initiates murine acute promyelocytic leukemia. Proc Natl Acad Sci U S A . 1997;94:2551-2556. 117. Taipale J, Beachy P A . The Hedgehog and Wnt signalling pathways in cancer. Nature. 2001;411:349-354. 118. Bunting K D . A B C transporters as phenotypic markers and functional regulators of stem cells. Stem Cells. 2002;20:11-20. 119. Chaudhary P M , Roninson IB. Expression and activity of P-glycoprotein, a multidrug efflux pump, in human hematopoietic stem cells. Cel l . 1991;66:85-94. 120. Berenson R J , Bensinger WI , H i l l RS , et al. Engraftment after infusion of CD34+ marrow cells in patients with breast cancer or neuroblastoma. Blood. 1991;77:1717-1722. 121. Berenson R J , Andrews R G , Bensinger WI , et al. Antigen CD34+ marrow cells engraft lethally irradiated baboons. J C l i n Invest. 1988;81:951-955. I l l 122. Krause D S , Ito T, Fackler M J , et al. Characterization of murine CD34, a marker for hematopoietic progenitor and stem cells. Blood. 1994;84:691-701. 123. Sato T, Laver J H , Ogawa M . Reversible expression of CD34 by murine hematopoietic stem cells. Blood. 1999;94:2548-2554. 124. Zinovyeva M V , Zijlmans J M , Fibbe W E , Visser JW, Belyavsky A V . Analysis of gene expression in subpopulations of murine hematopoietic stem and progenitor cells. Exp Hematol. 2000;28:318-334. 125. Scharenberg C W , Harkey M A , Torok-Storb B . The A B C G 2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors. Blood. 2002;99:507-512. 126. Zhou S, Schuetz JD, Bunting K D , et al. The A B C transporter B c r p l / A B C G 2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med . 2001;7:1028-1034. 127. Uchida N , Fujisaki T, Eaves A C , Eaves CJ . Transplantable hematopoietic stem cells in human fetal liver have a CD34(+) side population (SP)phenotype. J C l in Invest. 2001; 108:1071-1077. 128. Goodell M A , Brose K , Paradis G , Conner A S , Mull igan R C . Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J Exp Med . 1996;183:1797-1806. 129. A l v i A J , Clayton H , Joshi C, et al. Functional and molecular characterisation of mammary side population cells. Breast Cancer Res. 2003;5:R1-8. 130. Asakura A , Rudnicki M A . Side population cells from diverse adult tissues are capable of in vitro hematopoietic differentiation. Exp Hematol. 2002;30:1339-1345. 131. Lechner A , Leech C A , Abraham E J , Nolan A L , Habener JF. Nestin-positive progenitor cells derived from adult human pancreatic islets of Langerhans contain side population (SP) cells defined by expression of the A B C G 2 (BCRP1) ATP-binding cassette transporter. Biochem Biophys Res Commun. 2002;293:670-674. 132. Uchida N , Dykstra B , Lyons K , Leung F, Kristiansen M , Eaves C. A B C transporter activities of murine hematopoietic stem cells vary according to their developmental and -activation status. Blood. 2004;103:4487-4495. 133. te Boekhorst P A , de Leeuw K , Schoester M , et al. Predominance of functional multidrug resistance (MDR-1) phenotype in CD34+ acute myeloid leukemia cells. Blood. 1993;82:3157-3162. 134. Drenou B , Fardel O, Amiot L , Fauchet R. Detection of P glycoprotein activity on normal and leukemic CD34+cells . Leuk Res. 1993;17:1031-1035. 135. Haraguchi N , Utsunomiya T, Inoue H , et al. Characterization of a side population of cancer cells from human gastrointestinal system. Stem Cells. 2006;24:506-513. 136. Hirschmann-Jax C, Foster A E , W u l f G G , et al. A distinct \"side population\" of cells with high drug efflux capacity in human tumor cells. Proc Natl Acad Sci U S A . 2004; 101:14228-14233. 137. Kondo T, Setoguchi T, Taga T. Persistence of a small subpopulation of cancer stem-like cells in the C6 glioma cell line. Proc Natl Acad Sci U S A . 2004;101:781-786. 138. Smalley M J , Clarke R B . The mammary gland \"side population\": a putative stem/progenitor cell marker? J Mammary Gland B i o l Neoplasia. 2005;10:37-47. 112 139. Szotek PP, Pieretti-Vanmarcke R, Masiakos PT, et al. Ovarian cancer side population defines cells with stem cell-like characteristics and Mullerian Inhibiting Substance responsiveness. Proc Natl Acad Sci U S A . 2006; 103:11154-11159. 140. W u l f G G , Wang R Y , Kuehnle I, et al. A leukemic stem cell with intrinsic drug efflux capacity in acute myeloid leukemia. Blood. 2001 ;98:1166-1173. 141. Feuring-Buske M , Hogge D E . Hoechst 33342 efflux identifies a subpopulation of cytogenetically normal CD34(+)CD38(-) progenitor cells from patients with acute myeloid leukemia. Blood. 2001;97:3882-3889. 142. Beck W T , Grogan T M , Wil lman C L , et al. Methods to detect P-glycoprotein-associated multidrug resistance in patients' tumors: consensus recommendations. Cancer Res. 1996;56:3010-3020. 143. Bradley G , Naik M , L ing V . P-glycoprotein expression in multidrug-resistant human ovarian carcinoma cell lines. Cancer Res. 1989;49:2790-2796. 144. Efferth T, Davey M , Olbrich A , Rucker G , Gebhart E , Davey R. Activi ty of drugs from traditional Chinese medicine toward sensitive and M D R 1 - or MRPl-overexpressing multidrug-resistant human C C R F - C E M leukemia cells. Blood Cells M o l Dis. 2002;28:160-168. 145. Johny A , Song K W , Nantel S H , et al. Early stem cell transplantation for refractory acute leukemia after salvage therapy with high-dose etoposide and cyclophosphamide. B i o l Blood Marrow Transplant. 2006;12:480-489. 146. Bustin S A , Gyselman V G , Will iams N S , Dorudi S. Detection of cytokeratins 19/20 and guanylyl cyclase C in peripheral blood of colorectal cancer patients. B r J Cancer. 1999;79:1813-1820. 147. Bustin S A , Siddiqi S, Ahmed S, Hands R, Dorudi S. Quantification of cytokeratin 20, carcinoembryonic antigen and guanylyl cyclase C m R N A levels in lymph nodes may not predict treatment failure in colorectal cancer patients. Int J Cancer. 2004;108:412-417. 148. H u X F , Slater A , Wa l l D M , et al. Rapid up-regulation of mdr l expression by anthracyclines in a classical multidrug-resistant cell line. B r J Cancer. 1995;71:931-936. 149. Abolhoda A , Wilson A E , Ross H , Danenberg P V , Burt M , Scotto K W . Rapid activation of M D R 1 gene expression in human metastatic sarcoma after in vivo exposure to doxorubicin. C l in Cancer Res. 1999;5:3352-3356. 150. Thorgeirsson SS, Huber B E , Sorrell S, Fojo A , Pastan I, Gottesman M M . Expression of the multidrug-resistant gene in hepatocarcinogenesis and regenerating rat liver. Science. 1987;236:1120-1122. 151. Kanda Y , Hamaki T, Yamamoto R, et al. The clinical significance of CD34 expression in response to therapy of patients with acute myeloid leukemia: an overview of 2483 patients from 22 studies. Cancer. 2000;88:2529-2533. 152. Campos L , Guyotat D , Archimbaud E , et al. Surface marker expression in adult acute myeloid leukaemia: correlations with initial characteristics, morphology and response to therapy. B r J Haematol. 1989;72:161-166. 153. Borowitz M J , Gockerman JP, Moore JO, et al. Clinicopathologic and cytogenic features of CD34 ( M y 10)-positive acute nonlymphocytic leukemia. A m J C l in Pathol. 1989;91:265-270. 154. Geller R B , Zahurak M , Hurwitz C A , et al. Prognostic importance of immunophenotyping in adults with acute myelocytic leukaemia: the significance of the stem-cell glycoprotein CD34 (MylO) . B r J Haematol. 1990;76:340-347. 113 155. Campos L , Guyotat D , Archimbaud E , et al. Clinical significance of multidrug resistance P-glycoprotein expression on acute nonlymphoblastic leukemia cells at diagnosis. Blood. 1992;79:473-476. 156. Solary E , Casasnovas R O , Campos L , et al. Surface markers in adult acute myeloblastic leukemia: correlation of C D 19+, CD34+ and C D 14+/DR—phenotypes with shorter survival. Groupe d'Etude Immunologique des Leucemies (GEIL) . Leukemia. 1992;6:393-399. 157. Lee EJ , Yang J, Leavitt R D , et al. The significance of CD34 and TdT determinations in patients with untreated de novo acute myeloid leukemia. Leukemia. 1992;6:1203-1209. 158. Lamy T, Goasguen JE, Mordelet E , et al. P-glycoprotein (P-170) and CD34 expression in adult acute myeloid leukemia ( A M L ) . Leukemia. 1994;8:1879-1883. 159. Bradstock K , Matthews J, Benson E , Page F , Bishop J. Prognostic value of immunophenotyping in acute myeloid leukemia. Australian Leukaemia Study Group. Blood. 1994;84:1220-1225. 160. Fruchart C , Lenormand B , Bastard C, et al. Correlation between CD34 expression and chromosomal abnormalities but not clinical outcome in acute myeloid leukemia. A m J Hematol. 1996;53:175-180. 161. Arslan O, Akan H , Beksac M , et al. Lack of prognostic value of CD34 in adult A M L . Leuk Lymphoma. 1996;23:185-186. 162. de Grouw E P , Raaijmakers M H , Boezeman JB, et al. Preferential expression of a high number of A T P binding cassette transporters in both normal and leukemic CD34+CD38- cells. Leukemia. 2006;20:750-754. 163. Peeters SD, van der K o l k D M , de Haan G , et al. Selective expression of cholesterol metabolism genes in normal CD34+CD38- cells with a heterogeneous expression pattern in A M L cells. Exp Hematol. 2006;34:622-630. 164. Gottesman M M . Mechanisms of cancer drug resistance. Annu Rev Med . 2002;53:615-627. 165. Tomlins S A , Rhodes D R , Perner S, et al. Recurrent fusion of T M P R S S 2 and E T S transcription factor genes in prostate cancer. Science. 2005;310:644-648. 166. Ishkanian A S , Mal lo f f C A , Watson S K , et al. A tiling resolution D N A microarray with complete coverage of the human genome. Nat Genet. 2004;36:299-303. 114 Appendix: Drug sensitivity curves of CR and NR patients. CR34 Effects of PSC-833 and verapamil on DNR sensitivity in AML patient CR#34 CD34+CD38-DNR only P S C 0.3 uM P S C 1 uM P S C 3 uM Verapamil.5 ug/ml • Verapamil 20 ug/ml DNR ug/ml Effects of PSC-833 and verapamil on DNR sensitivity in AML patient CR#34 CD34-100 90 80 70 60 50 40 30 20 10 0 0 / // — • — D N R only -m— P S C 0.3 uM P S C 1 uM * P S C 3 uM X Verapamil 5 ug/ml —•—Verapami l 20 ug/ml •—~~ // •— 'i^r> ^ ^ ^ ^ .001 •0.01 0.1 DNR ug/ml Effects of PSC-833 and verapamil on DNR sensitivity in AML patient CR#34 CD34+CD38+ DNR ug/ml Effects of PSC-833 and verapamil on DNR sensitivity in AML patient CR#34 Unsorted 110 100 90 80 70 60 50 40 30 20 10 0 0. ^ ^ ^ ^ I .H» — • — D N R only - » - P S C 0 . 3 u M P S C 1 uM >• P S C 3 uM - '^^j —*—Verapami l 5 ug/ml • Verapamil 20 ug/ml .001 0.01 0.1 DNR ug/ml N R 8 110 100 90 80 70 60 50 40 30 20 10 0 0 Effect of PSC-833 and Verapamil on DNR sensitivity of AML NR8 CD34+38-- • — D N R only - •>- P S C 0.3uM P S C 1uM P S C 3uM — * - Verapamil 5ug/ml - • — Verapamil 20ug/ml • /r/-\"~\"~~ • ir / /// / /// / /// / //* 1 / / y • /* /•/ i / 001 0.01 0.1 DNR (jg/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AMLNR8CD34-120 110 100 90 80 70 60 50 40 30 20 10 0 - * - DNR only . psc 0.3uM P S C 1uM *>• P S C 3uM —«— Verapamil 5ug/ml —•— Verapamil 20ug/ml —: f yy ^^^^ y?y ' • IE- ' ' 0.001 0.01 0.1 DNR pg/ml 110 100 90 80 70 60 50 40 30 20 10 0 0. Effect of PSC-833 and Verapamil on DNR sensitivity of AMLNR8CD34+38+ - • — D N R only - * - P S C 0.3uM P S C 1uM P S C 3uM - » — Verapamil 5ug/ml - • — Verapamil 20ug/ml /?yy-~—\\ /ir / j / / / / / / / V / // / /* yy^y *sy^$y 001 0.01 0.1 DNR pg/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AML NR8 Unsorted 120 110 100 90 80 70 60 50 40 • 30 20 10 0 0.001 - *— DNR only - * - P S C 0 . 3 u M P S C 1uM \" * P S C 3uM Verapamil 5ug/ml \" - •—Verapami l 20ug/ml . ' ** &s' / II/ / //' ' ///' / .^yy^cy - = = ^ ^ ^ ^ 0.01 0.1 DNR wg/m) N R 1 4 Effect of PSC-833 and Verapamil on DNR sensitivity of -AMLNR14CD34+38-110 -j : 100 0.001' 0.01 0.1 1 DNR ug/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AML NR14 CD34-S? 50 // // «// ''/if —•— DNR only ~-«-PSC 0.3uM P S C 1uM P S C 3uM -*—Verapamil 5ug/ml -•—Verapamil 20ug/ml /// //// /'// • / / / 0.001 0.01 0.1 1 DNR ug/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AMLNR14CD34+38+ 110 i 0.001 • 0.01 0.1 1 DNR pg/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AML NR14 Unsorted 110 r : 0.001 0.01 0.1 1 DNR pg/ml N R 1 5 110 100 90 80 70 60 50 40 30 20 10 0 0 Effect of PSC-833 and Verapamil on DNR sensitivity of AMLNR15CD34+38-« - DNR only »-PSC0.3uM . PSC 1uM y- PSC 3uM •—Verapamil 5ug/ml *— Verapamil 20ug/m: 001 0.01 0.1 DNR pg/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AML NR15 CD34-DNR pg/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AMLNR15CD34+38+ 110 100 90 80 70 60 50 40 30 20 10 0 0. -*— DNR only - * - PSC 0.3uM P S C I u M - PSC 3uM - -*— Verapamil 5ug/ml . —•— Verapamil 20ug/ml / / / // / / / / / / ' / / ) ' / y y / \" y y 001 0.01 0.1 DNR pg/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AML NR15 Unsorted 120 110 100 90 80 H 70 60 50 40 30 20 10 0 - DNR only - PSC 0.3uM PSC 1uM PSC 3uM -Verapamil 5ug/ml -Verapamil 20ug/ml 0.001 0.01 0.1 DNR pg/ml NR16 Effect of PSC-833 and Verapamil on DNR sensitivity of AML NR16CD34+38-DNR ug/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AML NR16 CD34-DNR ug/ml MD 100 90 80 70 60 50 40 30 20 10 Effect of PSC-833 and Verapamil on DNR sensitivity of AMLNR16CD34+38+ 0 0.001 0.01 /S'y/ . : - • — DNR only . - » - P S C 0 . 3 u M P S C I u M * PSC 3uM - * — Verapamil 5ug/ml —•— Verapamil 20ug/ml \" \"~ ^ y y y 0.1 DNR ug/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AMLNR16Unsorted 100 90 80 70 60 50 40 30 20 10 0 0. 001 /si - * - DNR only PSC 0.3uM i PSC 1uM x- PSC 3uM —«h- Verapamil 5ug/ml - • — Verapamil 20ug/ml / y , --^ y 0.01 0.1 DNR pg/ml NR19 Effect of PSC-833 and Verapamil on DNR sensitivity of AML patient NR#19 CD34+38-DNR pg/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AML patient NR#19CD34-DNR pg/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AML patient NR#19 CD34+38+ DNR pg/ml Effect of PSC-833 and Verapamil on DNR sensitivity of AML patient NR#19 Unsorted "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0100341"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Biochemistry and Molecular Biology"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "ABC transporters as predictive factors for chemotherapeutic response in acute myeloid leukemia"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/30890"@en .