Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Targeting centrosome amplification in aneuploid B-cell precursor acute lymphoblastic leukemia Guo, Meiyun 2020

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2020_november_guo_meiyun.pdf [ 13.59MB ]
Metadata
JSON: 24-1.0390963.json
JSON-LD: 24-1.0390963-ld.json
RDF/XML (Pretty): 24-1.0390963-rdf.xml
RDF/JSON: 24-1.0390963-rdf.json
Turtle: 24-1.0390963-turtle.txt
N-Triples: 24-1.0390963-rdf-ntriples.txt
Original Record: 24-1.0390963-source.json
Full Text
24-1.0390963-fulltext.txt
Citation
24-1.0390963.ris

Full Text

TARGETING CENTROSOME AMPLIFICATION IN ANEUPLOID B-CELL PRECURSOR ACUTE LYMPHOBLASTIC LEUKEMIA  by  Meiyun Guo  B.Sc., Simon Fraser University, 2017  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Experimental Medicine)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) May 2020  © Meiyun Guo, 2020 ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the thesis entitled:    Targeting centrosome amplification in aneuploid B-cell precursor acute lymphoblastic leukemia   submitted by   Meiyun Guo  in partial fulfillment of the requirements for   the degree of    Master of Science     in    Experimental Medicine     Examining Committee:    Christopher Maxwell, Pediatrics  Supervisor   Gregor Reid, Pediatrics  Supervisory Committee Member   Xiaoyan Jiang, Medical Genetics  Additional Examiner   Additional Supervisory Committee Members:    Suzanne Vercauteren, Pathology and Laboratory Medicine  Supervisory Committee Member  iii  Abstract   B-cell precursor acute lymphoblastic leukemia (B-ALL) remains the single largest contributor to relapse in the pediatric leukemia patient population and new treatments are sorely needed to address this clinical challenge. Centrosomes play an important role in cell division, and centrosome abnormalities are a common feature in cancer cells. Mitotic cells with centrosome amplification are likely to form multipolar spindles, which generally lead to cell death. Cancer cells, therefore, must cluster supernumerary centrosomes to form pseudo-bipolar mitotic spindles and maintain cancer cell viability. My study investigates the efficacy of emerging inhibitors of centrosome clustering as new therapies to target pediatric B-ALL cells. As normal cells do not need to use centrosome clustering pathways, these inhibitors have the potential for low toxicity to healthy and growing tissues. However, tumor cells often resist targeted therapies and it is prudent to expect tumor adaption. My study shows that centrosome clustering inhibitors induce genetic and genomic instability in refractory leukemia cells, including micronuclei, which localize the DNA sensor cGAS and increase production of pro-inflammatory signals. Thus, refractory tumor cells may be immunogenic and activate an innate immune response. Overall, these findings identify centrosome clustering inhibitors as potential therapies to kill tumor cells and condition an immunogenic population that may be targeted by immune-based therapies to achieve long-term immune protection.     iv  Lay Summary  Acute lymphoblastic leukemia (ALL) is the most common blood cancer in children. The main treatment for children with ALL is chemotherapy. However, a return of ALL and long-term side effects of chemotherapy remain significant clinical challenges. To minimize the side effects after chemotherapy, personalized therapies target processes needed by leukemia cells that are likely to kill leukemias specifically with less or no effects on normal patient tissues. However, tumor cells often resist targeted therapies. So, it is prudent to expect resistance and plan for tumor adaption. In this thesis, I demonstrate that a class of agents can kill leukemic cells. In response to these drugs, surviving leukemic cells become more unstable, which in turn may make them more visible to the immune system and susceptible to immune-based treatments. Overall, this work identifies a more efficient way to recognize and kill both primary and resistant leukemic cells. v  Preface  Dr. Maxwell formulated the hypothesis for the thesis. Along with Dr. Maxwell, I designed the experiments in the thesis. In addition, I performed most of the experiments and data analysis with the following exceptions:  • For Chapter 3, Dr. Jihong Jiang and I performed the in vitro drug response analysis of B-ALL cell-lines.  • Dr. Gregor Reid’s laboratory (BC Children’s Hospital Research Institute, Canada) provided human cell-lines, cells from the Eµ-ret mouse model, and patient-derived xenograft (PDX) samples.  • Dr. Dario Campana’s laboratory (National University Cancer Institute, Singapore) generated the GFP-hTERT transformed mesenchymal stromal cells.  • For Chapter 5, Dr. Jihong Jiang contributed to the generation of 289 resistant cell-lines. • All studies involving tissues or cell samples derived from human participants were approved by the UBC-affiliated Research Ethics Board (certificate #H18-01197). vi  Table of Contents  Abstract ................................................................................................................................... iii Lay Summary .......................................................................................................................... iv Preface .......................................................................................................................................v Table of Contents .................................................................................................................... vi List of Tables ........................................................................................................................... xi List of Figures ......................................................................................................................... xii List of Abbreviations ..............................................................................................................xiv Acknowledgements ............................................................................................................... xvii Dedication ............................................................................................................................ xviii Chapter 1: Introduction ............................................................................................................1 1.1 Pediatric cancer........................................................................................................... 1 1.1.1 Prognosis and treatments ......................................................................................... 1 1.1.2 Genomic landscape of pediatric cancer .................................................................... 3 1.2 Acute lymphoblastic leukemia .................................................................................... 5 1.2.1 Leukemogenesis...................................................................................................... 5 1.2.2 Specific subtypes in pediatric ALL.......................................................................... 6  1.2.2.1 Hyperdiploid ALL .............................................................................................. 7  1.2.2.2 Hypodiploid ALL ........................................................................................ 8 1.3 Aneuploidy ................................................................................................................. 9 1.3.1 Generation of aneuploidy ...................................................................................... 11 1.3.2 Aneuploidy in cancer ............................................................................................ 12 vii  1.4 Centrosome amplification ......................................................................................... 13 1.4.1 Generation of centrosome amplification ................................................................ 13 1.4.2 Centrosome amplification and aneuploidy in cancer .............................................. 14 1.4.3 Centrosome clusteirng in cancer cells .................................................................... 15 1.5 Immunity and cancer................................................................................................. 17 1.5.1 Immune surveillance and cancer development ....................................................... 17 1.5.2 Resistance mechanisms in cancer .......................................................................... 19 1.6 Genomic instability and the immune system ............................................................. 19 1.6.1 Micronucleation .................................................................................................... 20 1.6.2 Innate immune resposne to dsDNA fragments and micronuclei ............................. 20 1.7 Mouse models for ALL ............................................................................................. 21 1.7.1 Patient-derived xenograft models .......................................................................... 21 1.7.2 Eµ-ret transgenic mouse model ............................................................................. 22 1.8 Rationale, hypothesis and significance of study ......................................................... 24 1.8.1 Summary of rationale ............................................................................................ 24 1.8.2 Hypothesis and aims of study ................................................................................ 25 1.8.3 Significance .......................................................................................................... 25 Chapter 2: Materials and methods ......................................................................................... 26 2.1 Cell culture ............................................................................................................... 26 2.1.1 Maintenance of B-ALL cells ................................................................................. 26 2.1.2 Maintenance of mesenchymal stromal cells (MSCs) .............................................. 26 2.1.3 Generation of 289 cells resistant to centrosome clustering inhibitors ..................... 27 2.2 Primary B-ALL or hematopoietic stem cells co-cultured with hTERT-MSCs ............ 27 viii  2.3 Immunofluorescence and image acquisition .............................................................. 28 2.4 Measurements of cell viability and cytotoxicity......................................................... 28 2.5 High-content imaging ............................................................................................... 29 2.6 Reverse-transcription PCR and quantitative real-time PCR ....................................... 29 2.7 Drugs used in the study ............................................................................................. 30 2.8 Reagents and antibodies used in the study ................................................................. 30 2.9 Statistics ................................................................................................................... 30 Chapter 3: Centrosome clustering is required for the viability of aneuploid B-ALL cells .. 32 3.1 Rationale and hypothesis........................................................................................... 32 3.2 Results ...................................................................................................................... 33 3.2.1 Centrosome amplification is a common feature in aneuploid B-ALL cells ............. 33 3.2.2 Centrosome clustering inhibitors are effective against aneuploid B-ALL cells ....... 36 3.2.3 Centrosome clustering inhibitors induce multipolar spindles in aneuploid B-ALL . 39 3.2.4 Sensitivity of B-ALL cells to centrosome clustering inhibitors correlates with the level of centrosome amplification in the treated population ............................................... 41 3.2.5 Primary B-ALL cells frequenctly exhibit centrosome abnormalities ...................... 44 3.2.6 Primary B-ALL cells cocultured on hTERT-MSC are viable during the drug screening protocol ............................................................................................................ 48 3.3 Key findings ............................................................................................................. 52 3.4 Discussion ................................................................................................................ 52 Chapter 4: Centrosome clustering pathway is targetable in primary aneuploid B-ALL ..... 58 4.1 Rationale and hypothesis........................................................................................... 58 4.2 Results ...................................................................................................................... 59 ix  4.2.1 Ex vivo analysis of primary B-ALL cells co-cultured on hTERT-MSC provides reproducible measurements of centrosome amplification and drug responses .................... 59 4.2.2 Stattic has poor selectivity against primary leukemia cells relative to normal stem cells and hTERT-MSC ..................................................................................................... 63 4.2.3 IC50 doses for centrosome clustering inhibitors show selectivity against primary B-ALL relative to stem cells derived from non-involved bone marrows................................ 67 4.2.4 Selectivity of centrosome clustering inhibitors correlates with the frequency of centrosome amplification .................................................................................................. 73 4.3 Key findings ............................................................................................................. 75 4.4 Discussion ................................................................................................................ 75 Chapter 5: Centrosome clustering inhibitors induce genome instability and the expression of pro-inflammatory signals in refractory aneuploid B-ALL ................................................ 78 5.1 Rationale and hypothesis........................................................................................... 78 5.2 Results ...................................................................................................................... 79 5.2.1 Primary pediatric B-ALL cells exhibit micronuclei following treatment ................ 79 5.2.2 Primary pediatric B-ALL cells show cGAS localized to micronuclei ..................... 82 5.2.3 Generation of 289 cells resistant to centrosome clustering inhibitors ..................... 84 5.2.4 289 resistant cell-lines exhibit higher basal levels of micronuclei .......................... 86 5.2.5 289 resistant cell-lines show cGAS located at micronuclei .................................... 88 5.2.6 289 resistant cell-lines show increased level of g-H2AX foci ................................. 90 5.2.7 289 resistant cell-lines increase the expression of pro-inflammatory signals .......... 92 5.3 Key findings ............................................................................................................. 95 5.4 Discussion ................................................................................................................ 95 x  Chapter 6: Discussion and conclusions ................................................................................. 100 6.1 Summary .................................................................................................................100 6.2 Centrosome clustering inhibitors as inducers of protective anti-ALL immunity ........102 6.3 Combination of centrosome clustering inhibitors and immune checkpoint  blockade ..............................................................................................................................103 6.4 Use of centrosome clustering inhibitors as a conditioning treatment for CAR-T  therapy  ...............................................................................................................................104 6.5 Suggested future studies ...........................................................................................106 Bibliography .......................................................................................................................... 107        xi  List of Tables  Table 2.1 Genes in mouse immune panel .................................................................................. 31 Table 2.2 Primary antibodies dilutions...................................................................................... 31 Table 3.1 Immortal cell-lines used in the study ......................................................................... 33 Table 3.2 Centrosome clutsering inhibitors used in the study .................................................... 36 Table 3.3 Centrosome clutsering inhibitors (expanded) used in the study.................................. 41 Table 3.4 Primary cells ............................................................................................................. 44 Table 4.1 Overview of patient sample characteristics................................................................ 59 Table 4.2 Characteristics of human primary cells derived from B-ALL involved (n=3) or non-involved (n=1) bone marrows  .................................................................................................. 60 Table 4.3 Characterization of 4 patient normal stem cells and 11 patient B-ALL ...................... 67    xii  List of Figures  Figure 1.1 Generation of tetraploidy is associated with chromosomal instability ....................... 10 Figure 1.2 Diagram depicting centrosome clustering pathway .................................................. 15 Figure 3.1 Centrosome phenotype in immortal B-ALL ............................................................. 34 Figure 3.2 Centrosome amplification frequency in B-ALL cell-lines ........................................ 35 Figure 3.3 Efficacy of centrosome clustering inhibitors in mouse and human B-ALL cell-lines  ................................................................................................................................................. 38 Figure 3.4 Formation of multipolar spindles in mitotic 289 cells treated with centrosome clustering inhibitors .................................................................................................................. 40 Figure 3.5 Correlation between CA and IC50 from 9 drugs in pediatric B-ALL cell-lines......... 43 Figure 3.6 Centrosome phenotype in mouse primary B-lymphocytes and aneuploid B-ALL ..... 46 Figure 3.7 The frequency of centrosome amplification in primary B-lymphocytes and B-ALL cells.......................................................................................................................................... 47 Figure 3.8 The development of ex vivo co-culture model using PDX3 ...................................... 49 Figure 3.9 Cell viability and drug response of PDX3 in three protocols .................................... 51 Figure 4.1 The correlation between two experiments (CA and IC50) in human primary cells ... 62 Figure 4.2 Efficacy of centrosome clustering inhibitors against human hTERT-MSC and primary cells.......................................................................................................................................... 64 Figure 4.3 Correlation between the frequency of mitotic centrosome amplification (CA) and the measured IC50 values for drug treatment of human primary stem cells, B-ALL and hTERT-MSC ................................................................................................................................................. 66 xiii  Figure 4.4 Drug responses of human normal stem cells treated with centrosome clustering inhibitors .................................................................................................................................. 69 Figure 4.5 Drug responses of human primary B-ALL cells treated with centrosome clustering inhibitors .................................................................................................................................. 70 Figure 4.6 Efficacy of centrosome clustering inhibitors in human primary cells ........................ 71 Figure 4.7 The correlation between CA and IC50 in human primary stem cells and B-ALL ..... 74 Figure 5.1 The frequency of micronuclei in DMSO and CC inhibitors treated primary cells ..... 79 Figure 5.2 Correlation between CA and MN in human primary stem cells and B-ALL ............. 81 Figure 5.3 cGAS co-localized to MN in primary B-ALL cells (C00652)................................... 83 Figure 5.4 Generation of the refractory B-ALL cells lines (289r) .............................................. 85 Figure 5.5 289r cells refractory to centrosome clustering inhibitors show elevated levels of micronuclei .............................................................................................................................. 87 Figure 5.6 Micronuclei-positive 289r cells frequently show cGAS co-localization.................... 89 Figure 5.7 Increased levels of γ-H2AX foci are observed in 289r cells ..................................... 91 Figure 5.8 Ct values for expression of 48 mouse immune genes between 289r (AZ82) cell lysates and 289r (DMSO) control cell lysates ....................................................................................... 93 Figure 5.9 Representative diagrams for signaling pathways of up- and down-regulated genes .. 99 Figure 6.1 Working models .....................................................................................................101  xiv  List of Abbreviations  ALL   Acute Lymphoblastic Leukemia AML   Acute Myeloid Leukemia APC   Antigen Presenting Cell B-ALL  B Cell Precursor Acute Lymphoblastic Leukemia BCP  B Cell Precursor CA  Centrosome Amplification CAR   Chimeric Antigen Receptor CCND2  Cyclin D2 CD40L  Cluster of Differentiation 40 Ligand CDKN2A  Cyclin-Dependent Kinase Inhibitor 2A CLL   Chronic Lymphocytic Leukemia CML   Chronic Myeloid Leukemia CNS   Central Nervous System CTLA-4  Cytotoxic T Lymphocyte Associated Protein 4 DAMP  Damage-Associated Molecular Pattern DAPI  4’,6-Diamidino-2-Phenylindole DC   Dendritic Cell DMEM Dulbecco’s Modified Eagle Medium  DMSO  Dimethyl Sulfoxide DNA  Deoxyribonucleic Acid  FBS  Fetal Bovine Serum xv  FDA   Food and Drug Administration GFP/luc  Green Fluorescent Protein and Firefly Luciferase GM-CSF  Granulocyte-Macrophage Colony-Stimulating Factor HLA   Human Leukocyte Antigen HNSCC  Head and Neck Squamous Cell Carcinoma IFN   Interferon IFN-γ   Interferon-gamma Ig   Immunoglobulin IKZF3   Ikaros Family Zinc Finger Protein 3 IL-12   Interleukin-12  IL-12R  Interleukin-12 Receptor JAK   Janus Kinase LIC   Leukemia-Initiating Cell mAb   Monoclonal Antibody MAPK  Mitogen-Activated Protein Kinase MHC-I  Major Histocompatibility Complex Class I MHC-II  Major Histocompatibility Complex Class II MM   Multiple Myeloma MRD   Minimal Residual Disease mTOR  Mammalian Target of Rapamycin MyD88  Myeloid Differentiation Primary Response 88 NF-κB  Nuclear Factor Kappa B NHL  Non-Hodgkin lymphoma xvi  NK   Natural Killer PBS  Phosphate Buffered Saline PCM   Pericentriolar Material PD-1   Program Cell Death 1 PD-L1  PD-1 Ligand 1 Ph   Philadelphia PI3K   Phosphoinositide 3-Kinase PLK1  Polo-like kinase 1 qPCR   Quantitative Polymerase Chain Reaction RET   Rearranged During Transfection RFP   Ret Finger Protein RNA  Ribonucleic Acid SD   Standard Deviation SEM   Standard Error of the Mean  siRNA  Small Interfering RNA STAT3  Signal Transducer and Activator of Transcription 3 TCR   T Cell Receptor TGF- β  Transforming Growth Factor-Beta TIL   Tumor-Infiltrating Lymphocytes TLR   Toll-Like Receptor TNF   Tumor Necrosis Factor TNF-α  Tumor Necrosis Factor Alpha TP53   Tumor Protein 53 xvii  Acknowledgements  This project was made possible with the help and support of many wonderful people. At the beginning, I would like to thank my supervisor, Dr. Chris Maxwell for supporting my research and providing diverse opportunities. Without your continuous encouragement and guidance, I would not dig deep within myself and complete the challenges. I appreciate the training and graduate school experience received during my graduate study. I would also like to thank my supervisory committee, Dr. Gregor S. Reid and Dr. Suzanne M. Vercauteren for their insightful advice on my research study.  To members of the Maxwell lab, both past and present, I am grateful for their help and support, especially for the trouble shooting on experimental set-up and teaching on microscopy. Also, I appreciate the help from other BCCHR trainees and the fun talks we had during lunch time.  Lastly, I would like to thank my family and friends for their support and company all these years. To my parents, I am keeping forward because of you. xviii  Dedication  This work is dedicated to my parents for their unwavering support and unconditional love. 1  Chapter 1: Introduction  1.1 Pediatric cancer Childhood cancer is relatively rare; however, it is the most common disease-related cause of death among children. In Canada, there are approximately 850 children diagnosed with cancer every year, and about one in six of these kids will die from the disease 1. The most prevalent subtypes of pediatric cancer include leukemia (32%), brain and central nervous system (19%), and lymphoma (11%).  Cancers in children are different from those in adults. Pediatric cancers differ in pathology, appearance, rate of growth, and response to treatment. Some cancers are exclusive to children, such as neuroblastoma, Wilms tumor, rhabdomyosarcoma and retinoblastoma. In general, tumors in children often grow faster and can quickly metastasize to other parts of the body.   1.1.1 Prognosis and treatments Survival rates for pediatric cancer have continued to improve over the past 50 years. The 5-year survival rates increased from less than 30% in the early 1960s to 80% in the late 1990s 2,3. Over the past decades, survival rates have shown marked improvements for acute lymphoblastic leukemia (ALL), Non-Hodgkin lymphoma (NHL) and brain and central nervous system cancers 4. A greater decline in mortality has been seen in lymphoid cancers (ALL, NHL, and Hodgkin disease) than nonlymphoid cancers with the annual mortality rate declining to 1.9% 4. Moreover, 5-year survival rates are higher in specific diagnostic groups, including retinoblastoma (99%), renal tumours (92%), lymphomas (89%), and germ cell tumours (89%). In contrast, neuroblastoma (70%) and malignant bone tumours (72%) have relatively low survival rates 1. Despite the  2  increased survival rates, there are still many children that die from cancer every year, which highlights the need for improved therapies for pediatric cancers. Although survival rates are encouraging, two-thirds of pediatric cancer survivors are likely to suffer chronic or long-term side effects after cancer treatment. The long-term effects include infertility, weak growth, cardiac damage, psychosocial effects, and the development of second cancers (3 to 12% of survivors). Endocrine and metabolic complications are the most prevalent late-effects in children survivors, accompanied by sensory problems, neurocognitive impairment, cardiopulmonary dysfunction, gastrointestinal disorders, and secondary malignant neoplasms 5. In general, these adverse effects from chemotherapy and radiation may appear several years after cure and some may result in long-term complications 6.  A main reason for mortality is disease relapse 7. It is reported that the leading cause of death of 5-year survivors is disease recurrence 8. For some cancers, such as ALL and Hodgkin lymphoma, the increased 5-year survival rates have been followed by an increase in late relapse rates 9–11. Over the last two decades, relapse rates are 15 - 20% in developed countries 12,13, and the overall survival rates after relapse are 40 - 70% depending on the follow-up time and the type of risk groups 14,15.  Treatment of pediatric cancer depends on the type of cancer and the stage of the disease. General treatments include chemotherapy, surgery, radiation, and stem cell transplantation. Treatment for pediatric solid tumors usually consists of a combination of chemotherapy, radiation, and surgery. Even though solid tumors tend to be localized at initial diagnosis, the spreading of cancer cells to distinct locations is often observed 16. Consequently, intensification of chemotherapy is applied to kill the spreading and presumably more resistant cancer cells. Although  3  the intensification of therapy can improve outcomes, long-term side effects are a major clinical problem that encourages the development of novel and less toxic treatments.  Over the past two decades, technological improvements in the measurement of genetic changes in malignant cells has revealed common mutations, translocations and structural changes17, which promote the proliferation and survival of cancerous cells under cellular stress. Therapies that target these alterations are termed precision medicine. For example, the chromosomal translocation t(9;22) results in the formation of a chimeric protein consisting of the breakpoint cluster region (BCR) gene with the Abelson murine leukemia viral oncogene homolog 1 (BCR-ABL) fusion gene, which activates the tyrosine kinase ABL and promotes proliferation. By applying the tyrosine kinase inhibitor imatinib, the proliferation of leukemic cells can be inhibited and durable remission achieved 18. In addition to targeted therapy, cancer immunotherapy has quickly developed during the last 5 years. Generally, the immune system has the ability to kill most abnormal cells. The past decade has shown that the immune system can be reactivated to recognize and eliminate tumor cells. Monoclonal antibodies, oncolytic virus therapy, cancer vaccines, chimeric antigen receptor T-cell (CAR-T) therapy, and bispecific T-cell engagers are each different forms of immunotherapy.  1.1.2 Genomic landscape of pediatric cancer Cancer is a genetic disease that arises from the accumulation of genetic alterations. Most cancers contain somatic mutations that occur in a given tissue or organ. Somatic mutations are likely induced by environmental factors such as viral infection, UV exposure or chemical reagents, as well as a defective endogenous DNA repair system. Cancers with somatic mutations are frequently observed in older people, consistent with the epidemiology studies, which showed an  4  increased incidence of cancer in the population aged more than 60 years 19. Up to 10% of all cancers arise from inherited genetic defects 20. In contrast to somatic mutations, a germline mutation is harbored in every cell of the body and is heritable. Germline heterozygote mutations in cancer predisposition genes are usually not lethal because of the compensation by a normal second allele. When the second allele has a somatic mutation, the predisposition gene is totally inactivated. Therefore, germline mutations cause a higher risk for cancer development in a carrier. As a result, hereditary cancers are often seen in young people.  Genomic sequencing studies reveal key differences between pediatric and adult cancers. Of particular note, the number of somatic mutations in most pediatric cancers is typically lower than those seen in adult cancers 21,22. However, pediatric cancers have a higher prevalence of germline mutations in cancer predisposition genes 23. It is proposed that approximately 7% of pediatric tumors carry a mutation in a cancer predisposition gene 24. Analysis of ALL samples obtained at diagnosis and relapse from the same patient demonstrate the acquisition of new genetic alterations and loss of initially detected genetic events 25, suggesting clonal evolution in leukemia. The CREBBP gene, a transcriptional coactivator, is frequently mutated in relapsed ALL, which leads to resistance to glucocorticoids 26. Other studies also showed that the acquisition of additional mutations in a single subclone from diagnosis leads to therapy resistance 27. Whole genome sequencing (WGS) analysis of diagnostic and post-therapy medulloblastomas revealed heterogeneity in the dominant clone after therapy, and less than 12% of diagnostic alterations were retained at relapse 28.  WGS analysis has also revealed some novel genetic mutations, which can serve as therapeutic targets in pediatric cancer. For example, chemotherapies combined with ABL1 tyrosine kinase inhibitor (imatinib) were able to improve three-year event-free survival (EFS) in patient with BCR-ABL1 positive ALL 29. Moreover, the  5  identification of JAK2 mutations in high risk ALL results in the introduction of JAK inhibitor (ruxolitinib) for relapsed and refractory malignancy 30. Thus, the identification of molecular pathways, or processes, that are more prominent in malignant cells than in normal cells provides rationale for the development of novel treatment combinations.  1.2 Acute lymphoblastic leukemia Acute lymphoblastic leukemia (ALL) is a cancer of white blood cells that is characterized by the overproduction and accumulation of immature lymphocytes in the bone marrow. It is the most common form of cancer in children and accounts for 30% of childhood cancer 31. The frequency of ALL is higher in males than in females with a sex ratio of 1.4 : 1 32. ALL is classified into two groups, T-ALL and B-ALL, based on immunophenotype and modern treatments have led to overall survival rates approaching 90% 33. However, most of these cases arise from a fetal B cell precursor (BCP) cell of origin 34,35.   1.2.1 Leukemogenesis A common characteristic among cancers, including leukemia, is unregulated cell growth. Leukemogenesis is proposed to be a multistep progression in which normal cells need to develop a number of alterations before they become malignant 36. The alterations may involve structural and functional changes in genes that are responsible for clonal expansion of defective precursor B cells. Though the specific genetic events involved in leukemogenesis are not well understood, it has been suggested that mutations that cause proliferative advantage and impairment of the maturation process are required.  6  Leukemias develop mostly from hematopoietic stem cells that obtain the initial mutation in the bone marrow, leading to a prolonged “pre-leukemic” phase composed of leukemia initiating cells (LIC) 37. Most pediatric leukemias probably originate in utero through chromosome changes 38, leading to immune tolerance to associated antigens. Chromosome number alterations and/or translocations, which cause the formation of fusion genes, can also initiate leukemia in both children and adult 38. In aneuploid ALL, abnormal chromosome number is an early event in leukemogenesis, which is confirmed by the detection of the same chromosome abnormality in both leukemic and LIC populations. In addition, twin and neonatal blood studies demonstrate that chromosome number alterations can initiate leukemogenesis, resulting in subsequent development of hyperdiploid pre-leukemic cells 39. Therefore, alterations in chromosome number are likely an initial step 39–42. Chromosome abnormalities can initiate leukemia and lead to clinically silent pre-leukemic clones 38,39. The rate of generation for silent pre-leukemic clones is 100 times higher than the rate of diagnosis for high hyperdiploid (HeH) ALL 38. Therefore, a subsequent postnatal event is required to complete leukemogenesis 38,39. During disease progression, pre-leukemic cells may acquire secondary mutations, such as copy number alterations and point mutations, and evolve towards malignant leukemia 43. Clonal evolution can further shape the leukemia with subsequent mutations that increase or decrease the adaptive fitness of pre-leukemic clones, leading to positive or negative selection 44.  1.2.2 Specific subtypes in pediatric ALL Subtypes of ALL are often defined by specific genomic characteristics, including aneuploid genomes or structural chromosome alterations. Certain genetic alterations are  7  associated with favorable prognosis, such as a hyperdiploid genome (51-65 chromosomes) and ETV6-RUNX1 fusions. Other genetic alterations are considered to be high-risk disease, such as a hypodiploid genome (<44 chromosomes) or a BCR-ABL1 fusion.  BCR-ABL1-like ALL is a high-risk subtype that has a B-cell precursor phenotype and a similar gene expression profile to BCR-ABL-positive ALL with an IKZF1 alteration 45. This subtype occurs in approximately 10% of pediatric ALL 46. Around half of BCR-ABL1-like ALL has a CRLF2 rearrangement 47 and many CRLF2-rearranged cases also contain JAK mutations 30. Transcriptome analysis and WGS revealed that BCR-ABL1-like ALL lacking CRLF2 dysregulation had alternative genetic alterations which can activate cytokine receptor and kinase signaling 48. Overall, this genotype represents a high risk of relapse, which is independent of age, cytogenetics and minimal residual disease (MRD) levels after remission induction 46.  1.2.2.1 Hyperdiploid ALL High hyperdiploid ALL is the most common subgroup in childhood B-cell precursor acute lymphoblastic leukemia (BCP-ALL), which accounts for 25-30% of pediatric B-ALL cases 49. High hyperdiploid ALL is characterized by nonrandom gain of chromosomes X, 4, 6, 10, 14, 17, 18, and 21 with modal number of 51 – 67 chromosomes 49. Chromosome gains are usually trisomies but tetrasomies are also seen 50. However, high hyperdiploidy (51-67 chromosomes) is rare in adult ALL and T-ALL 49, which highlights the strong association between hyperdiploidy and pediatric B-ALL. Clinically, high hyperdiploid  ALL is mainly associated with children under 5-years old and characterized by a relatively low white blood cell count 51. Although high hyperdiploid  pediatric ALL generally has a favorable prognosis, about 20% of diagnosed patients will suffer a relapse and 10% of diagnosed patients will die from the disease 52.   8  The age and sex of the patient and the presence of specific chromosome trisomies have strong independent effects on disease outcomes 52. For example, the frequency of having bone marrow relapse is higher in boys than girls 53; however, there is no significant difference in overall survival. In general,  boys older than 9-years with high hyperdiploid B-ALL have a higher risk of relapse 52. Early studies suggested that the gain of specific trisomies (+4,+6,+10,+17) is correlated with an improved prognosis 52,54–56, and the gain of chromosome 5 is associated with poor prognosis 57.  1.2.2.2 Hypodiploid ALL Hypodiploid ALL is a rare subtype of BCP-ALL with poor prognosis 58, which accounts for 0.5% of pediatric ALL cases 59 and is defined as having a genome with less than 45 chromosomes. Low hypodiploidy (32-39 chromosomes) is proposed to be a prognostic biomarker for high-risk disease in both pediatric and adult ALL 60. Patients with low hypodiploidy are older 59 and often have inherited genetic alterations of IKZF2, RB1, and TP53 61. Near-haploid (24-31 chromosomes) ALL is associated with patients that are younger 59 and have genetic alterations targeting receptor tyrosine kinase signaling, Ras signaling, as well as alterations of IKZF3 61. Hypodiploid ALL generally has simple karyotypes with few structural aberrations 59. No recurrent structural abnormality and no common fusion gene have been identified 62. In addition, relatively few microdeletions are seen possibly because many chromosomes have already been lost through the generation of hypodiploidy 61.  The frequency of hypodiploidy, as well as other high-risk genetic aberrations, is four times higher in adults compared to children and adolescents 63, indicating the strong correlation between age and prognosis 63. Patients with hypodiploid ALL will have a high risk of relapse and/or death  9  if treated as standard risk 64. A recent study demonstrated that minimal residual disease (MRD) is the most important prognostic indicator for pediatric hypodiploid ALL and MRD-guided therapy can substantially improve the outcome 65. In this study, the 5-year EFS was 85.1% for the 14 patients who achieved negative MRD status at the end of remission induction and 44.4% for the six patients with detectable MRD.   1.3 Aneuploidy Aneuploidy is defined as an abnormal number of chromosomes in a cell. Generally, aneuploid genomes differ from diploid genomes by only one or a small number of chromosomes 66. Tetraploidization is one example of aneuploidy and is proposed to associate with the initiation of chromosomal instability (CIN) via various mechanisms (Fig. 1.1)  10   Figure 1.1 Generation of tetraploidy is associated with chromosomal instability Proper DNA repair and chromosomal segregation during cell division play an important role in the maintenance of genomic stability. (1) DNA and centrosome are duplicated during S phase, and replication errors must be repaired prior to mitosis. Potential mechanisms to generate tetraploidy include: (2) Centrosome amplification, (3) telomerase dysfunction, (4) failure of the spindle assembly checkpoint, and (5) failure of the post mitotic checkpoint, which gives rise to a single tetraploid cell (4N) instead of two diploid cells (2N). Therefore, the generation of a tetraploid cell leads to the propagation of genomic instability to subsequent generations. Adapted from “Genomic instability in human cancer: Molecular insights and opportunities for therapeutic attack and prevention through diet and nutrition” by L.R. Ferguson et al., 2015, Semin Cancer Biol, 35, S5-S24. Copyright (2015) by the authors. Adapted with permission.   11  1.3.1 Generation of aneuploidy Aneuploidy is often caused by chromosome mis-segregation during mitosis, which is guarded against by the mitotic checkpoint (or spindle assembly checkpoint, SAC). This cell cycle control mechanism delays the progression from metaphase to anaphase and allows proper bi-oriented chromosome attachment to spindle microtubules. The SAC is essential in mammalian cells, as complete inactivation of this checkpoint leads to cell death and early embryonic lethality 67–69. In normal circumstances, the SAC will delay mitotic progression in response to unattached kinetochores 70. However, if the expression of SAC proteins is weak or mutated, anaphase may initiate abnormally and result in chromosome mis-segregation and aneuploidy. Altered expression of mitotic checkpoint components was identified in a subset of aneuploid human cancers, including leukemia, breast, and lung cancer 71.  The assembly of multipolar mitotic spindles is another potential mechanism to generate aneuploidy. Extra centrosomes are highly associated with human cancer cells and aneuploidy 72. The centrosome is responsible for the formation of a mitotic spindle pole, and cells that possess more than two centrosomes might form multipolar spindles and subsequent aneuploid daughter cells. Cells that pass through a multipolar spindle will generate an increased frequency of merotelic chromosome attachments, which occur when microtubules from both poles attached to a single kinetochore. Merotelic attachments frequently result in lagging anaphase chromosomes, chromosome mis-segregation and aneuploidy.  Telomere maintenance is also crucial for the cell cycle and the maintenance of genome stability. The failure to maintain telomeres can lead to telomere attrition and activation of DNA damage responses that result in the termination of cell proliferation (telomere crisis). Early studies showed the presence of transient telomere crisis in early stages of cancer, followed by the  12  stabilization of shorter telomeres through telomerase reactivation 73–75. Interestingly, telomere dysfunction can lead to genome-doubling through cytokinesis failure or endoreduplication 76. Tetraploid daughter cells are frequently genomically unstable and tumorigenic 77.   1.3.2 Aneuploidy in cancer Genomic instability has a critical role in cancer development and it is commonly manifested by structural or numerical chromosomal aberrations 78. The integrity of the genome is protected at every stage of the cell cycle. The presence of aneuploid or tetraploid cells indicates the failure of one or many of the repair pathways. Studies in mice show that newly formed tetraploid cells exhibit chromosomal instability (CIN) and can initiate tumor formation 77.  Single-cell analysis of paired primary and metastatic breast cancers show that primary tumor cells with the predominance of a highly aneuploid clone have ongoing CIN, which gives rise to metastasis 79. Aneuploid cells have a random combination of chromosomes, leading to the generation of complex phenotypes 80. Although aneuploidy dampens homeostatic cell proliferation, it allows clones to possess growth advantages when exposed to chemotherapeutic drugs, and it can drive the acquisition of entirely new phenotypes 81,82. Cancer cells are often aneuploid. For example, chromosome number alterations and/or other cytogenetic abnormalities are observed in more than 20% of acute myeloid leukemia (AML) cases 83,84. There is also correlation between specific chromosome alterations and cancer types. For example, monosomy of chromosome 7 is prevalent in myeloid malignancy and indicates a poor prognosis 85. Patients with hyperdiploid ALL have a good prognosis and the gain of specific trisomies (+4,+6,+10,+17,+18) are correlated with an improved prognosis 52,54–56. Conversely, the gain of  13  chromosome 5 is associated with poor prognosis 57. Importantly, aneuploidy is generally preserved from diagnosis to relapse 86, indicating that it likely plays an important role in disease progression.  1.4 Centrosome amplification The centrosome is often the major microtubule-organizing center in the cell and controls many cellular processes via the nucleation and organization of cytoplasmic microtubules (MT) 87. The centrosome duplicates once during a normal cell cycle to give rise to two centrosomes that later function as the spindle poles in a bipolar mitotic spindle 88. The bipolar spindle ensures equal segregation of sister chromatids to each daughter cell 89. In experimental studies, mutations in centrosome-associated kinases dysregulated centrosome duplication and led to aberrant mitosis and centrosome amplification (CA) 90, which is characterized by combinations of increases in centrosome number, size, abnormal structure and function 91.  1.4.1 Generation of centrosome amplification There are two possible mechanisms that can lead to centrosome amplification: cell doubling events (cytokinesis failure or cell–cell fusion) and centriole overduplication 92. Centriole overduplication can be introduced by the overexpression of the centriole regulator, polo-like kinase 4 (PLK4) 93. Inhibition of PLK4 activity with centrinone B reduces the percentage of centrosome amplification in melanoma cell-lines. However, the overexpression of PLK4 does not drive most cases of centriole overduplication in melanoma, indicating the presence of other mechanisms 93. In vitro studies suggest PLK4, SAS6, STIL, and pericentrin each have roles in centriole overduplication 94–97, and increased PLK4 expression is found in medulloblastoma, breast, colorectal, prostate, and ovarian cancers 98–103.   14  Another mechanism to generate centrosome amplification is tetraploidization. Unscheduled tetraploidization arises from either defective cell division or cell-to-cell fusion, which usually bestows the daughter cell with supernumerary centrosomes 104. Mitotic dysfunction, such as endoreplication or cytokinesis failure, generates either mononucleated or binucleated tetraploid cells with duplicated centrosomes. In the presence of mechanical stress, cell-cell fusion may occur 105, which also generates binucleated tetraploid cells with two centrosomes.  1.4.2 Centrosome amplification and aneuploidy in cancer Centrosome amplification is commonly seen in cancer cells. For example, 44.3% of melanomas exhibit centrosome amplification 93. It is proposed that supernumerary centrosomes can lead to multipolar cell division, which cause chromosome mis-segregation during mitosis and lead to mitotic death or aneuploidy. Although cancer cells with supernumerary centrosomes possess centrosome clustering pathways to form pseudo-bipolar spindles, the transient formation of multipolar spindles has been observed 106. Such chromosome instability is a hallmark of cancer and is believed to promote tumorigenesis 107. Indeed, centrosome amplification is found in precursor lesions and, thus, may be an early or even initiating event in carcinogenesis 108,109.  Centrosome amplification is correlated with an increase in aneuploidy and chromosomal instability 93,110 111, which are important for both the initiation and evolution of cancer. Invasive breast tumors indicate a high correlation between aneuploidy and centrosome amplification; that is, centrosome size and number both showed a positive, significant linear correlation with aneuploidy and chromosomal instability 110. Moreover, in multiple myeloma (MM), centrosome abnormalities were found more frequently in bone marrows from MM patients compared to patients with pre-malignant monoclonal gammopathy of undetermined significance or normal  15  plasma cells 112. Overall, many studies have shown centrosome amplification is associated with CIN and aneuploidy indicating a potential role in cancer initiation and progression 111,113  1.4.3 Centrosome clustering in cancer cells Centrosome amplification is associated with multipolar mitotic spindles and consequent multipolar cell division. However, extra centrosomes also assemble into small clusters during prometaphase to form distinct groupings at pseudo-bipolar spindles 114. This is termed centrosome clustering and there are basically two types of forces that contribute to this process 115 (Fig. 1.2). The first type of force is exerted by astral microtubules on centrosomes at the beginning of mitosis to facilitate clustering. Anti-parallel microtubule-bundling complexes generate pulling force in overlapping regions. Second, forces exerted by the cell cortex on astral microtubules will restrict the movement of centrosomes and contribute to aggregation by facilitating the movement of centrosomes towards each other.   Figure 1.2 Diagram depicting centrosome clustering pathway Centrosome clustering pathways in cancer cells are mainly achieved by two forces: pulling (red box) and pushing (blue box) forces. Pulling force is generated by microtubule motor proteins (ie. KIFC1, KIF4). Motor proteins cross-link microtubules anchored at distinct centrosomes to facilitate centrosome aggregation. Moreover, astral microtubules from centrosomes interact with actin filaments at the cell cortex to restrict the movement of centrosomes and stabilize the pulling force by motor proteins.  16  Several motor proteins contribute to the process of centrosome clustering. For example, KIFC1/HSET is a minus-end directed kinesin-14 family member with a motor domain and an N-terminal microtubule-binding domain that can cross-link adjacent microtubules. Because of its minus-end motor activity, and the interaction with the centrosomal protein CEP215/CDK5RAP2, this motor protein can focus microtubule minus ends at spindle poles 116,117. In addition, KIFC1 also participates in the clustering of acentrosomal spindle poles, which are more common in cancer cells than normal cells 118.  Astral microtubules are another important component in the centrosome clustering process. The nucleation of microtubules at the centrosome is a complex process that involves multiple proteins. The initiation step is the recruitment of γ-tubulin ring complexes to the centrosome, which is mediated by Nedd1 119,120. PLK1, a mammalian homolog of Drosophila Polo 121,122, plays an important role in promoting the interaction between MT and γ-tubulin ring complexes 123. PLK1 interacts with and phosphorylates Nedd1 123, promoting MT nucleation at the centrosome 124. Another protein that participates in centrosome clustering is nuclear mitotic apparatus protein 1 (NUMA1). NUMA1 localizes to the nucleus during interphase and accumulates at the spindle poles during mitosis 125,126. NUMA1 binds to microtubule minus ends and dynactin to regulate the clustering of microtubules for bipolar spindle formation 127.  By utilizing centrosome clustering pathways, cancer cells with supernumerary centrosomes can avoid lethal effects from multipolar division 128. Should supernumerary centrosomes fail to cluster and form multipolar spindles, however, apoptosis of tumor cells is often triggered 115. Because normal cells do not require centrosome clustering pathways, the inhibition of these pathways may specifically target tumor cells 129. These studies suggest that centrosome clustering pathways may be potential therapeutic targets against aneuploid pediatric ALL cells.  17  1.5 Immunity and cancer Targeted therapies, including monoclonal antibodies and kinase inhibitors, have improved patient responses in many tumor types, such as metastatic melanoma and non-small cell lung cancer (NSCLC) 130,131. Although significant progress has been made, tumor cells often develop resistance leading to disease progression. While the majority of tumor cells are likely sensitive to the drug, a small amount of cells with genetic alterations may survive and continue to grow under drug pressure, resulting in disease resistance and recurrence 132.   1.5.1 Immune surveillance and cancer development Immune surveillance is a theory that proposes the immune system not only surveys and responds to pathogenic infections but also to cancerous tissue. Successful immune surveillance can recognize and eliminate tumor cells by recognizing tumor-associated antigens. It is thought that genetic instability in tumor cells can generate true tumor-specific neoantigens. As a consequence, the immune system is a potentially powerful therapeutic tool to use against cancers. The diversity of receptors in the adaptive immune response, such as T-cell receptor (TCR) and antibodies, contributes to the capacity for specific targeting of cancer cells. Moreover, the diverse cell killing patterns from both the innate and adaptive immune system offer the potential to kill any abnormal cell that is appropriately recognized. The concept that cancer cells develop in the presence of a functional immune system is termed immunoediting 133, and it is divided into 3 phases: elimination, equilibrium, and escape 134. Generally, tumor cells present tumor antigens via MHC class I molecules to CD8+ effector cells and NK cells. Dendritic cells (DC) can also take up and cross-present tumor antigens to T cells, including natural killer T (NKT) cells. These effector cells contribute to anti-tumor immune  18  response by producing IFN- γ. CD8+ T cells interact with FAS and TRAIL receptors on tumor cells or secrete perforin and granzymes to induce the tumor cell’s apoptosis. Effector T cells express CD28, CD137, GITR, and OX40, which are co-stimulatory molecules to increase their proliferation and survival. Innate immune cells, such as macrophages (M1) and granulocytes, release TNF-α, IL-1, and IL-12 that can mediate anti-immune effects by inhibiting tumor cell proliferation and angiogenesis. Each of the active processes in the elimination phase may only eliminate a portion of tumor cells. In this case, a temporary stage of equilibrium occurs between the immune system and the tumor population. During this period, the population of tumor cells may evolve with the accumulation of mutations that can shut down T cells responses specific for tumor-associated antigens. In T cells, the ultimate quantity and quality of the T cell response is regulated by a balance between co-stimulatory and co-inhibitory signals, which are often termed immune checkpoints 135. Immune checkpoints are responsible for the maintenance of immune homeostasis and self-tolerance. However, tumor cells may co-opt their expression to silence effector immune responses. A major inhibitory cytokine produced by many cell types is TGF-b, which normally arrests the cell cycle at G1 stage 136. In cancer cells, a mutation in the TGF-b pathway promotes resistance to cell cycle inhibition, and results in uncontrolled proliferation. As the equilibrium process continues, the immune system applies selective pressure to control tumor progression. This pressure also selects for tumor cell variants that resist anti-tumor immune responses, leading to the escape phase. In the escape phase, the immune system is less able to restrict tumor cell proliferation, causing clinically apparent disease.   19  1.5.2 Resistance mechanisms in cancer Tumor cells resist therapy through many mechanisms, including secondary alterations such as mutation or amplification. For example, acquired resistance to an mTOR inhibitor requires activating mutations in the MTOR gene 137. The high level of kinase activity that results from these mutations allows resistance to mTOR inhibitors. Likewise, in anti-PD-1 treated melanomas, clinical resistance arises by the mutation of the beta-2-microglobulin (B2M) gene, which can result in improper folding and localization of major histocompatibility complex class I 138, and reduce the presentation of peptide fragments from cytosolic proteins. Thus, immune elimination is dampened for cancer cells with such a mutation. Moreover, the upregulation of anti-apoptotic genes (Bcl2) and downregulation of pre-apoptotic genes (Bax) in cancer cells can increase resistance to chemotherapy 139 while mutations in the TP53 gene also promote drug resistance 140.  Another common resistance mechanism in cancer cells is to enable signaling pathways to “bypass” the drug target. For example, melanoma cells with a BRAF mutation can eventually resist BRAF inhibition by reactivating alternative pathways 137. Cancer cells also become resistant by reducing the activity of drugs 141. For example, while cytarabine has no effect on cancer cells, the phosphorylated form of cytarabine is lethal 142. Mutations in the enzyme that promotes the cytarabine phosphorylation pathway reduces the drug activity and enables drug-resistant cancer cells 143.   1.6 Genomic instability and the immune system Immunotherapy has produced significant responses however many patients still do not have a complete and persistent response to the therapies. The reasons are largely unclear. Therefore, new ways to activate the anti-tumor immune response in non-responding patients are  20  needed. Recent studies indicate that activation of the innate immune system, rather than a focus on the adaptive system, may present a potential new therapy to eliminate cancer cells 144,145.   1.6.1 Micronucleation Nuclear atypia and defects in nuclear architecture are believed to be major events in cancer progression 146. Micronuclei are one type of nuclear architecture defects, which are small extranuclear bodies containing one or a few chromosomes that form because of improper mitotic chromosome segregation. Lagging chromosomes recruit a nuclear envelope during telophase, which causes the separation of the micronucleus and main nucleus. However, micronuclei are not only the result of genomic instability but can also introduce genomic instability inside the cell.  Micronuclei contain acentric chromosomes, chromatid fragments, or lagging chromosomes that possess a distinct nuclear envelope from the main nucleus 147. Acentric chromosomes and chromatid fragments are the consequence of unrepaired or improperly repaired DNA lesions. When a single kinetochore from one chromosome is attached to microtubules from more than one spindle pole (merotelic attachment), this chromosome is simultaneously pulled towards opposite pole, resulting in the formation of lagging chromosome 148. During anaphase, acentric chromosomes, chromatid fragments, and lagging chromosomes result in the formation of a micronucleus. Chromosomes in a micronucleus are subject to massive DNA fragmentation, which can cause a mutational pattern commonly presented in cancer genome called chromothripsis 149,150.  1.6.2 Innate immune response to dsDNA fragments and micronuclei In contrast to the main nucleus, micronuclei often undergo nuclear membrane rupture during interphase, which exposes the inside DNA to the cytoplasm 149,151. Therefore,  21  micronucleation is proposed to be associated with DNA damage and activation of innate immune pathways. Cytoplasmic DNA from micronuclei can be mistaken as foreign DNA, and therefore, initiate elimination through the innate immune response. A cytoplasmic enzyme cGAS (cyclic 2’-3’-GMP-AMP synthase) is designed to detect cytoplasmic DNA 152. Once it binds to cytoplasmic DNA, the enzymatic activity of cGAS is activated, producing cyclic GMP-AMP (cGAMP), the small molecule second messenger 153. Thereafter, cGAMP binds to the central scaffold of the innate immune response called STING, which initiates a cascade of events that modulates the transcription of inflammatory cytokines.  1.7 Mouse models for ALL Over the past century, murine models are widely used in biomedical research. In my study, I used two types of mouse models to analyze B-ALL: (i) patient-derived xenografts in immune-deficient recipient mice; and, (ii) cells generated from transgenic immune-competent mice.  1.7.1 Patient-derived xenograft models Patient-derived xenograft (PDX) models are patient tumor cells or tissues engrafted in immune-deficient mice. PDX models faithfully resemble the original tumors and, as a consequence, are applied in various types of cancer, such as chronic lymphocytic leukemia (CLL) 154, large B cell lymphoma 155, pancreatic cancer 156, colorectal cancer 157, gastric cancer 158, high-grade serous carcinoma 159, and intrahepatic cholangiocarcinoma 160. In general, PDX models accurately reflect patient tumors with regards to genomic and genetic mutations, as well as therapeutic response.  22  PDX models provide valuable in vivo evidence that support basic science studies of cancers, such as the processes of tumor progression and metastasis. In addition, PDX models can serve as a guide for the clinical treatment of cancer cells. PDX models are used in preclinical drug testing to indicate drug safety, efficacy and dosage in many different types of cancers, such as pancreatic cancer 161, non-small cell lung cancer (NSCLC) 162, melanoma 163, breast cancer 164, colon cancer 165, and prostate cancer 166. Unfortunately, most human cancer cell-lines frequently lose the characteristics of human malignancy 167. In contrast, PDX models can better reflect the pathology of an individual patient. Overall, PDX models provide in vivo platforms for researchers to have a better idea about the cellular and molecular mechanisms of therapy resistance of cancers 168 and identify the beneficial chemotherapies for patients with cancers 169.  1.7.2 Eµ-ret transgenic mouse model BCP-ALL is the most common childhood cancer, which results from the over-expansion of pro-B or pre-B cells in the bone marrow. However, the study of BCP-ALL leukemogenesis using human patient samples is limited due to the undetectable preleukemic stage. Therefore, it is unclear whether and how genetic abnormalities occur and develop with the transformation of pro- or pre-B cells. To solve this problem, several approaches using animal models have been developed that allow researchers to study the transformed population that emerges from a pool of B-cell precursor cells. One example of this is the generation of mice with transgenic constructs that transform B lineage cells.  In normal B lineage development, the pro-B cell stage is characterized by the rearrangement of the immunoglobulin heavy chain (IgH) loci. The pre-B cell stage is characterized by the expression of cytoplasmic IgM heavy chain (µ) protein. The B cell stage includes cells that  23  express surface IgM and subsequent IgD proteins 170. In murine B lineage development, the late pro-B cells match the most commonly transformed B lineage cell in human B cell precursor ALL. Late pro-B cells are in the process of completing the rearrangement of the IgH chain loci or VDJ rearrangement. Completion of VDJ rearrangement will result in the production of µ protein. Pre-B cell stages are the most prominent stages; in these stages, cells undergo proliferation 171. However, failure to complete VDJ rearrangement will result in cell death 172. The RET tyrosine kinase is expressed during the pro-B cell stages in murine bone marrow B lineage development 173. As a result of µ expression, RET expression is downregulated at the pre-B cell stage 173. Constitutively expressed RET tyrosine kinase is oncogenic 174. Transgenic mice expressing Eµ-ret develop B lymphoid malignancies; in these mice, a RFP/RET fusion gene is expressed under the transcriptional control of the IgH enhancer 175. The leukemia cells from Eµ-ret mice are defined as B cell precursors because of the expression of CD45R, the presence of clonal IgH chain rearrangements, and the lack of sIgM expression 176. Therefore, Eµ-ret mice not only provide opportunity for the study of leukemogenesis but also for pathology analysis of BCP-ALL.   24  1.8 Rationale, hypothesis and significance of study 1.8.1 Summary of rationale ALL is the most common cancer in children. Current treatments cure approximately 90% of patients. However, about 20% of patients suffer a disease relapse, which results in low survival rates 177. Moreover, patients that survive current treatments are at risk for adverse late-effects due to the potential damage current treatments cause to normal tissues 178. Therefore, new therapies are required that target relapsed ALL with reduced long-term effects in children. Many B-ALL cases are defined by aneuploidy (gain or loss of whole chromosomes), which includes hyperdiploidy (the most common pediatric subtype) and hypodiploidy. Aneuploidy is likely generated by a defective cell division with an abnormal number or size of centrosomes. Cancer cells cluster these extra centrosomes together to generate pseudo-bipolar cell divisions and avoid lethal effects from multipolar divisions. As normal cells do not often have supernumerary centrosomes, centrosome clustering pathways could be specific targets for cancer cell therapy with low toxicity to normal tissues. Tumor cells often resist targeted therapies resulting in disease progression. In response to centrosome clustering inhibitors, the resistant cancer cells may exhibit heightened genome instability. While genome instability is a hallmark of cancer and a proposed enabling characteristic for relapse 179, it might create a population of resistant B-ALL cells that are likely to be targeted by immune cells. For example, micronuclei can activate cGAS-STING signaling and a downstream proinflammatory response. Therefore, cells with micronuclei may become more visible to the immune system.    25  1.8.2 Hypothesis and aims of study Hypothesis:  Centrosome clustering is required for the viability of aneuploid B-ALL cells and inhibition of this process will be lethal. Few B-ALL cells will survive and these refractory cells will be prone to genomic instability, which should augment the production of pro-inflammatory signals and prime for anti-tumor immune responses.  Aims: 1. Determine the frequency of centrosome amplification and the efficacy of centrosome clustering inhibitors in immortal B-ALL cells. 2. Target centrosome clustering in primary B-ALL cells taken directly from patients. 3. Evaluate genomic instability and cGAS-STING pathway in refractory B-ALL cells.  1.8.3 Significance B-ALL remains the single largest contributor to relapse in the pediatric leukemia patient population and new therapies are sorely required to address this clinical need. Targeting disease features that are commonly shared by B-ALL cells has the potential to significantly reduce the incidence of disease progression. While targeted therapies may induce response, resistance or relapse often occur. Based on previous findings in the literature, we proposed that tumor cells may adapt to declustering events potentially through the acquisition of genome instability. Generation of micronuclei links genome instability to the activation of innate immunity, which may enhance the clinical responses to immune-based treatments for pediatric B-ALL patients.  26  Chapter 2: Materials and methods 2.1 Cell Culture  2.1.1 Maintenance of B-ALL cells Human B-ALL cell-lines, including RS4;11, RCH-ACV and 380 cells, were a gift from Dr. Gregor Reid’s laboratory (BC Children’s Hospital Research Institute, Canada). All human B-ALL cell-lines were cultured in RPMI 1640 with L-glutamine (Gibco) supplemented with 10% fetal bovine serum (FBS, Invitrogen), 20 U/ mL penicillin (Invitrogen), 20 µg/ mL streptomycin (Invitrogen), 2mM L-glutamine (Gibco), 20mM HEPES (Gibco) and 91.5 µM 2-mecaptoethanol (2ME, Invitrogen).  A mouse B-ALL cell-line, termed 289 cells, was provided by Dr. Gregor Reid’s laboratory (BC Children’s Hospital Research Institute, Canada). 289 cells were derived from spontaneous primary leukemic cells arising in Eµ-ret mice. These cells were cultured in complete Dulbecco's Modified Eagle Medium (DMEM, Gibco) supplemented with 20% FBS, 2mM L-glutamine, 20 U/ mL penicillin, 20 µg/ mL streptomycin, 20 mM HEPES, 1X NEAA, 91.5µM 2ME and 100pg/mL recombinant murine IL-7 (Peprotech). All cell-lines were grown at 37°C in a 5% (v/v) CO2 incubator.   2.1.2 Maintenance of mesenchymal stromal cells (MSCs) Human mesenchymal stromal cells (MSCs) were a gift from Dr. Dario Campana’s laboratory (National University Cancer Institute, Singapore). These human MSCs were immortalized through the expression of TERT and GFP in primary bone marrow MSCs in Dr. Dario Campana’s laboratory. These cells were maintained in RPMI 1640 medium supplemented  27  with 10% fetal bovine serum (FBS, Invitrogen) and 10-6 mol/L hydrocortisone (Sigma). Cells were grown at 37°C in a 5% (v/v) CO2 incubator.  2.1.3 Generation of 289 cells resistant to centrosome clustering inhibitors 289 cells that resist centrosome clustering inhibitors were generated in the Maxwell lab. 289 cells were passaged 24 hours before seeding in a 12-well plate (Corning). These cells were separately exposed to an IC50 dose of centrosome clustering inhibitor (AZ82 or PJ34) or equivalent DMSO control. The drug, or DMSO, was refreshed every two days for 2 weeks. Then, the drug concentration was increased by 33% (1.33 fold) for an additional 2 weeks. This cycle was repeated six times until a population of 289 cells that were resistant to lethal doses of each drug was generated. These cell-lines were termed 289r (DMSO), 289r (AZ82), or 289r (PJ34).  2.2 Primary B-ALL or hematopoietic stem cells co-cultured with hTERT-MSCs In a 96-well plate (Corning), hTERT-MSCs were seeded at 3000 per well in 200 µL of RPMI-1640 medium containing 10% fetal bovine serum (FBS, Invitrogen) and 1µM hydrocortisone (Sigma) 24 hours prior to seeding with primary B-ALL or stem cells from peripheral blood or bone marrow. RPMI-1640 complete medium was removed before adding 5x104 B-ALL cells, recovered from cryopreserved samples, in 200 µL of AIM-V medium. Both primary cells and hTERT-MSCs were incubated at 37°C in a 5% (v/v) CO2 incubator. The protocol described by Frismantas et al 180 was adapted from their published 384-well format to a 96-well format used in this thesis.   28  2.3 Immunofluorescence and image acquisition Cells were seeded on poly-L-lysine coated coverslips and fixed with ice-cold methanol for 10 minutes at -20°C. Cells were blocked in PBS with 0.2% Triton X-100 and 3% BSA for 1 hour at room temperature. Antibodies were diluted in PBS with 0.2% Triton X-100 and 3% BSA. Primary antibodies were diluted and incubated with coverslips overnight at 4°C. Cells were then washed three times in PBS. The coverslips were incubated with diluted secondary antibodies at room temperature for 1 hour in the dark. Coverslips were washed three times in PBS and mounted with ProLong Gold Antifade Reagent containing DAPI (Invitrogen).  Fixed cells were imaged using the Fluoview software (Olympus) connected to the Olympus Fluoview FV10i confocal microscope (Olympus). Image stacks of 20 optical sections with a spacing of 0.5 µm through the cell volume were taken using a 60X 1.2 NA oil objective. ImageJ v1.46j (National Institute of Health) was used to generate maximum intensity projection of the fluorescent channels.  2.4 Measurements of cell viability and cytotoxicity The cell cytotoxicity assay kit was purchased from Abcam (ab 112118). Cell viability was assessed by treating cells with 1/5 volume of assay solution for 4 hours at 37°C. Absorbance at 570 and 605nm wavelengths was measured with an EnSpireTM multilabel reader (PerkinElmer). The background absorbance of the blank wells was subtracted from the values for the wells containing the cells. The ratio of OD570 to OD605 was used to determine the cell viability in each well.   29  2.5 High-content imaging Cells were seeded in a 96-well plate (Corning) and placed in the chamber of the high content image analysis system (ImageXpress Micro XL). Images were taken through a 40X 0.75 NA dry objective on the ImageXpress Micro XL epifluorescence microscope (Molecular Devices Inc) controlled by the MetaXpress 5.0.2.0 software (Molecular Devices Inc). For the analysis of the proportion of living cells, images were taken once per site using 50-ms exposures, 2x2 binned resolution, with 100% of full lamp intensity for each channel, and 25 optical sections spaced 500 µm apart. For analysis of micronuclei and gamma-H2AX, images were taken once per site using 50-ms exposures, 2x2 binned resolution, with 100% of full lamp intensity for each channel, and 100 optical sections spaced 700 µm apart. Post-acquisition processing of images was performed using MetaXpress offline.  2.6 Reverse-transcription PCR and quantitative real-time PCR RNA was extracted from 289r (DMSO) cells and 289r (AZ82) cells using AllPrep DNA/RNA mini kit (Qiagen). Preparations were measured with a NanoDrop (Thermo-Fisher). Reverse transcription PCR reaction was carried out using the SuperScript IV VILO Master Mix kit. gDNA digestion was performed at 37°C for 2 minutes and the following reverse transcription PCR conditions were: 25°C for 10 minutes, followed by 50°C for 10 minutes, followed by 85°C for 5 minutes. Then, cDNA quantity was determined by the QuantiFluor dsDNA System (Promega). Quantitative PCR (qPCR) reactions were run in a 7000 series machine (Applied Biosystems) using StepOnePlus system. qPCR cycling conditions were: 50°C for 2 min (UNG incubation) and 95°C for 20 seconds (enzyme activation), followed by 40 cycles at 95°C for 1 second (denaturing stage) and at 60°C for 20 seconds (annealing/extension stage). Analysis of  30  qPCR results was done using the Ct values, which were normalized to 18S ribosomal RNA (18S rRNA). Genes are listed in Table 2.1.  2.7 Drugs used in the study AZ82 (Caymanchem), Stattic (Sigma), PJ34 (SelleckChem), AZ9482 (AdipoGen), Olaparib (SelleckChem), Talazoparib (SelleckChem), Doxorubicin (Sigma), Taxol (SelleckChem) and Methotrexate (SelleckChem) were dissolved in DMSO.   2.8 Reagents and antibodies used in the study Cell Trace Violet (Invitrogen), CyQuant (Invitrogen), Poly-L-lysine (Sigma), DAPI (Sigma), and TaqMan Fast Advance Master Mix (ThermoFisher) were used as indicated by the manufacturer. Primary antibodies are listed in Table 2.2. Secondary antibodies were conjugated to Alexa Fluor-488, -594 or -679 (Invitrogen).  2.9 Statistics Data were expressed as mean ± standard deviation (SD) or as mean ± standard error of mean (SEM) as indicated in each figure. Statistical analysis was performed by ANOVA as indicated in each figure. The results were considered significant at P<0.05.     31  Table 2.1 Genes in mouse immune panel  GENE SYMBOL ENDOGENOUS CONTROL GENES 18S, Gapdh, Hprt1, Gusb IMMUNE RESPONSE ASSOCIATED GENES Agtr2, Bax, Bcl2, Bcl2l1, C3, Ccl19, Ccl2, Ccl3, Ccl5, Ccr2, Ccr4, Ccr7, Cd19, Cd28, Cd34, Cd38, Cd3e, Cd4, Cd40, Cd40lg, Cd68, Cd80, Cd86, Cd8a, Csf1, Csf2, Csf3, Ctla4, Cxcl10, Cxcl11, Cxcr3, Cyp1a2, Cyp7a1, Edn1, Fas, Fasl, Fn1, Gzmb, H2-Ea, H2-Eb1, Hmox1, Icos, Ifng, Ikbkb, Il10, Il12a, Il12b, Il13, Il15, Il17a, Il18, Il1a, Il1b, Il2, Il2ra, Il3, Il4, Il5, Il6, Il7, Il9, Lrp2, Lta, Nfkb1, Nfkb2, Nos2, Prf1, Ptgs2, Ptprc, Sele, Selp, Ski, Smad3, Smad7, Socs1, Socs2, Stat1, Stat3, Stat4, Stat6, Tbx21, Tgfb1, Tnf, Tnfrsf18, Vcam1, Vegfa, Ace, Icam1, Lif, Ly96, Nfatc3, Nfatc4  Table 2.2 Primary antibodies dilutions PROTEIN COMPANY HOST DILUTION Centrobin Abcam Mo IF: 1:100 CP110 Proteintech Rb IF: 1:2500 Pericentrin Convance Rb IF: 1:500 β-tubulin Cell Signaling 647 Conjugated IF: 1:600 γ-tubulin Sigma Mo IF: 1:2500 Phospho-Histone H3 (Ser10) Cell Signaling Rb IF: 1:500 γ-H2AX Millipore Mo IF: 1:250 cGAS (D1D3G) Cell Signaling Rb IF: 1:500 cGAS (D3O8O) Cell Signaling Rb IF: 1:500    32  Chapter 3: Centrosome clustering is required for the viability of aneuploid B-ALL cells  3.1 Rationale and hypothesis An initiating mitotic event that seeds aneuploidy in B-ALL cells will induce centrosome amplification, which in turn will serve as a precise mark for leukemia cells and a potential therapeutic target. Recently published work identified centrosome abnormalities in 18 of 26 pediatric B-ALL tumors (including 12 of 13 aneuploid tumors) using a binary present or absent assessment 181. As cancer cells frequently contain supernumerary centrosomes, these extra centrosomes must be clustered together during cell division to form pseudo-bipolar mitotic spindle and avoid lethal effects from multipolar cell division 115. As normal cells do not have supernumerary centrosomes, the centrosome clustering pathway could be a potential specific target for cancer therapy with low off-target effects on normal tissue.  I hypothesize that B-ALL cells require centrosome clustering pathways for viability such that inhibition of these pathways will be lethal to tumor cells.     33  3.2 Results 3.2.1 Centrosome amplification is a common feature in aneuploid B-ALL cells To measure centrosome abnormalities, I examined the centrosome phenotypes using immunofluorescence in three human B-ALL cell-lines and one mouse B-ALL cell-line (Table 3.1).   Table 3.1 Immortal cell-lines used in the study IMMORTAL  CELL- LINES SPECIES KARYOTYPE ORIGIN CYTOGENETICS 289 Mouse 41 - 43 (hyperdiploid) + chr. 9,12,17 Transgenic Eµ-ret+ mice RFP/RET fusion gene RCH-ACV Human 43 - 50 (2N) + chr. 8 Bone marrow cells taken at relapse from an 8-year-old girl t(1;19) (q23;p13.3) 380 Human 43 - 47 (2N) - chr. 14 Peripheral blood taken at relapse from a 15-year-old boy EBNA negative, t(8;14;18) (q24;q32;q21) RS4;11 Human 46 (2N) Bone marrow cells taken at relapse from a 32-year-old women t(4;11) (q21;q23), isochromosome for the long arm of chr.7   Cells were passaged one day prior to the immunofluorescence protocol so they were in logarithmic growth phase at the time of analysis.  Cells were fixed and stained with three different antibodies that each recognize different components of the centrosome. I defined a centrosome by co-staining the centrosome component (γ-tubulin) with an internal centriole (CP110) and the nucleation of microtubules (b-tubulin) (Fig. 3.1). Figure 3.1 indicates examples of a cell with an enlarged centrosome (top panel, cell #1) and a cell with a supernumerary centrosome (bottom panel, cell # 4) as determined in the immortal mouse B-ALL cell-line 289 (hereafter 289 cells). In  34  interphase, cells with either one or two centrosomes and centrosome size smaller than 4 µm2 were considered normal (Fig. 3.12,3). Cells containing more than two centrosomes (Fig. 3.14) or cells containing a centrosome larger than 4 µm2 (Fig. 3.11) were counted as an abnormal centrosome or centrosome amplification.    Figure 3.1 Centrosome phenotype in immortal B-ALL Centrosomes were stained with γ-tubulin (centrosome), CP110 (centrioles) and β-tubulin (microtubules) in methanol-fixed immortal mouse 289 cells. Cells were stained with DAPI to mark the DNA. All the confocal images represent a maximum intensity projection of stacks of optical sections acquired at 0.5 µm intervals through the cell Z-axis. Scale bar=4 µm. Centrosomes in each cell were selected and enlarged in the right panels, and each centrosome is indicated with an arrow. Scale bar=1 µm.   35  To query whether centrosome amplification is a frequent occurrence in B-ALL cell-lines, I measured the frequency of centrosome amplification in three human B-ALL cell-lines and one mouse B-ALL cell-line (Table 3.1). As cells with 2 centrosomes may be normal in G2-phase or abnormal in G1- phase, centrosomes were measured by immunofluorescence in both mitotic and interphase cells. Three different centrosome markers were used to confirm the identity of centrosomes and cells with either supernumerary centrosomes (≥ 3) or a large centrosome (≥4 µm2) were classified as centrosome amplification. I found 40.7% of mitotic 289 cells exhibited centrosome amplification, and lower frequencies were observed in RCH-ACV (32%), 380 (16.7%), and RS;411 (19.2%) cells; up to 15% interphase cells from all immortal B-ALL cells have centrosome amplification (Fig. 3.2). Thus, mitotic B-ALL cell-lines demonstrate centrosome amplification, which may also be a common feature in primary human B-ALL cells.  Figure 3.2 Centrosome amplification frequency in B-ALL cell-lines Centrosome size and number were evaluated in interphase and mitotic cells, including abnormal centrosome number (red) and larger centrosome size (yellow). More than 20 mitotic and 100 interphase cells were counted for each cell-line. Data is representative of two independent experiments.   36  3.2.2 Centrosome clustering inhibitors are effective against aneuploid B-ALL cells To evaluate the sensitivity of B-ALL cells to centrosome clustering inhibitors, four immortal B-ALL cell-lines (Table 3.1) were treated with 6 graded doses each of six centrosome clustering inhibitors (Table 3.2). The six inhibitors were selected due to their ability to induce multipolar mitotic spindles (Wang et al., 2018; Chang et al., 2005; Chondrou et al., 2018; Zasadil et al., 2014), including: the HSET molecular motor inhibitor AZ82, the Stat3 inhibitor Stattic, PARP inhibitors (Olaparib and PJ34), and standard chemotherapies, such as doxorubicin, and paclitaxel.  Table 3.2 Centrosome clustering inhibitors used in the study INHIBITORS TARGET DESCRIPTION AZ82 (21898) KIFC1/HSET KIFC1 cross-links adjacent microtubules and can focus microtubule minus ends at spindle poles. Stattic (S7947) STAT3 Stattic alters the SH2 domain of STAT3, which acts as transcription factor and also regulates the Stathmin-PLK1 cascade to reduce astral microtubules. PJ34 (S7300) PARP Inhibition of the enzyme poly ADP ribose polymerase (PARP) causes multiple double strand breaks, which can induce multipolar spindles in cancer cells. Olaparib (AZD2281) PARP 1/2 Doxorubicin (D1515) topoisomerase II Doxorubicin is a widely used anticancer drug, which intercalates in DNA, inhibits topoisomerase II, generates free radicals and induces multipolar spindles. Paclitaxel (T7402) Tubulin Paclitaxel inhibit microtubules dynamics, arrests cells in mitosis and induces multipolar spindles.     37  All cell-lines were passaged 24 hours prior to the drug screening protocol and seeded in 96 well plates. At day 0, drugs were dissolved in DMSO, and an equivalent amount of DMSO was added to each of three control wells. Then, after 48 hours of drug incubation, a colorimetric assay was performed to measure cell viability or the approximate number of metabolically-active cells.  Cell viability was normalized to that measured in DMSO-treated control wells, and a nonlinear regression model (log (inhibitor) vs response) was applied to determine the cell killing curve, and specifically the inhibitory concentration 50 (IC50) value for each drug. As shown by the dose response curves in Figure 3.3, hyperdiploid 289 cells (red lines), which contain elevated levels of centrosome amplification, were more sensitive to the majority of centrosome clustering inhibitors, including AZ82, Stattic, Olaparib, PJ34, and Doxorubicin.  An interesting exception is that these cells were less sensitive to the microtubule stabilizing drug paclitaxel. Overall, these experiments show that centrosome clustering inhibitors have high efficacy against immortal B-ALL cell-lines with higher frequency of centrosome amplification.   38   Figure 3.3 Efficacy of centrosome clustering inhibitors in mouse and human B-ALL cell-lines Immortal cell-lines were separately exposed to serial dilutions of AZ82, Stattic, Olaparib, PJ34, Doxorubicin or Taxol for 48 hours. Cell viability was determined using a colorimetric assay. Viability in each dose was normalized to that in DMSO control. The kill curve for each centrosome clustering inhibitor was plotted and IC50 is indicated by a dash line. (mean ± SEM, n=3 experiments).    39  3.2.3 Centrosome clustering inhibitors induce multipolar spindles in aneuploid B-ALL  To assess whether leukemic cells treated with these inhibitors exhibit multipolar spindles, 289 cells, with the highest levels of centrosome amplification and aneuploidy, were treated with three centrosome clustering inhibitors (AZ82, Stattic, and PJ34), due to their high efficacy measured in Section 3.2.2. Cells were exposed to 1 µM dose, or equivalent DMSO, which approximates the measured IC50 for each drug. Again, cells were passaged 24 hours before drug exposure and seeded in a 24-well plate. Five hours post inhibitor treatment, cells were fixed in 4% PFA following ice-cold methanol and stained with γ-tubulin (centrosome), CP110 (centriole), and tubulin B (microtubule) for the measurement of centrosome phenotype and spindle morphology. Compared to DMSO-treated 289 cells, declustered supernumerary centrosomes and multipolar spindles were observed at an elevated level in 289 cells treated with centrosome clustering inhibitors (Fig. 3.4a); that is, 37.4% of AZ82-treated, 44.4% of Stattic-treated, and 34.8% of PJ34-treated 289 mitotic cells show multipolar spindles (Fig. 3.4b), indicating that centrosome clustering inhibitors are able to induce multipolar spindles in aneuploid B-ALL with centrosome amplification.  In summary, the results of these experiments reveal that inhibition of centrosome clustering pathways is effective at reducing the viability of aneuploid B-ALL cells and inducing multipolar spindle phenotypes.  40   Figure 3.4 Formation of multipolar spindles in mitotic 289 cells treated with centrosome clustering inhibitors (a) Mitotic 289 cells were identified by phosphorylated histone H3 staining. These cells were also immunostained for γ-tubulin, β-tubulin, and DAPI. All images represent a maximum intensity projection of stacks of optical sections acquired at 0.5 µm intervals through the cell Z-axis. Examples of bipolar and multipolar spindles are shown. The spindle poles in H3-positive cells are highlighted by an arrow. Scale bar = 5 µm.  (b) The frequency of mitotic 289 cells with multipolar spindles is plotted for drug-treated (1µM of AZ82, Stattic or PJ34) and DMSO control-treated mitotic 289 cells after 5 hours. (mean ± SD, n=3, >20 cells per bar, *P<0.05, **P<0.01, One-way ANOVA).  41  3.2.4 Sensitivity of B-ALL cells to centrosome clustering inhibitors correlates with the level of centrosome amplification in the treated population To select the optimal centrosome clustering inhibitors prior to transitioning from immortal B-ALL cell-lines to samples derived from primary human tumor tissues, IC50 values were measured following exposure to two additional PARP inhibitors (AZ9482 and Talazoparib) and one additional clinical drug (Methotrexate) (Table 3.3).  Table 3.3 Centrosome clustering inhibitors (expanded) used in the study INHIBITORS TARGET DESCRIPTION AZ9482 (SYN-3046) PARP Inhibition of the enzyme poly ADP ribose polymerase (PARP) causes multiple double strand breaks, which can induce multipolar spindles in cancer cells. Talazoparib (BMN 673) PARP 1 Methotrexate (S1210) Dihydrofolate reductase (DHFR) Methotrexate suppresses inflammation and prevents cell division in cancer. Methotrexate inhibits DNA synthesis by competitively inhibiting dihydrofolate reductase (DHFR).     42  For the four immortal B-ALL cell-lines, the levels of mitotic centrosome amplification (from Section 3.2.1) and the IC50 values for 9 drugs were plotted as a heatmap with yellow indicating low values and red indicating high values (Fig. 3.5). The correlation between the levels of centrosome amplification (CA) and the IC50 values was measured by Pearson r (vs CA) to generate a P value of each drug, which revealed a strong negative correlation between the levels of centrosome amplification and the IC50 concentration for most drugs; that is, cell-lines with high CA had lower IC50 values. The negative correlation was significant for AZ82 (r= -0.98, P=0.02) and doxorubicin (r= -0.99, P=0.005) and trends were observed for Stattic (r= -0.87, P=0.13) and PJ34 (r= -0.66, P=0.34). Taxol was an exception that showed a positive correlation (r= 0.82, P=0.18) while AZ9482 (r= 0.45, P=0.55) and methotrexate (r= 0.04, P=0.96) did not show a trend. It is important to note the caveat that these correlations are only based on the responses of 4 immortal B-ALL cell-lines, but these findings do indicate that cells with a higher frequency of centrosome amplification are more sensitive to some centrosome clustering inhibitors. These findings, therefore, support further evaluation of five inhibitors (AZ82, Stattic, PJ34, Olaparib, and doxorubicin) against human patient-derived samples.    43    Figure 3.5 Correlation between CA and IC50 from 9 drugs in pediatric B-ALL cell-lines Horizontal scaled heatmaps for mitotic centrosome amplification (CA) and IC50 values for 9 drugs versus four pediatric B-ALL cell-lines (mouse 289; RCH ACV; 380; RS4;11). Red or yellow colors indicate the high or low value of either CA or IC50. IC50 values were generated from Figure 3.3. Pearson r (vs. CA) values and the P value (in parentheses) are provided with an asterisk indicating significant correlation. (*P<0.05, **P<0.01).  44  3.2.5 Primary B-ALL cells frequently exhibit centrosome abnormalities To query whether centrosome amplification is also prevalent in primary B-ALL cells, I analyzed the frequency of centrosome amplification in primary mouse splenic cells, in which the majority of the cells are B lymphocytes, as well as primary leukemic cells derived from transgenic (Eµ-ret) mice and three patient B-ALL samples expanded in immune-deficient mice (PDX1, PDX2, and PDX3) (Table 3.4). It is important to note that the PDX samples were originally generated from human primary minimal residual disease tissues, in which the leukemia burden is <5%, however the PDX samples were composed of the overt leukemia that were expanded in immune-deficient recipient mice.  Table 3.4 Primary cells PRIMARY CELLS SPECIES KARYOTYPE ORIGIN CYTOGENETICS SPLEEN CELLS mouse N/A N/A N/A EΜ-RET mouse + chr. 9, 12, 17 BALB/c mice express the RET fusion protein under control of the Ig heavy chain enhancer RFP/RET fusion PDX 1 human + chr. X, 4, 5, 8, 14, 17 Post-treatment day 29 minimal residual disease (MRD) bone marrow from patient 1 expanded in immune-incompetent mice del CDKN2A, a germline ATM mutation, a PTPN11 mutation PDX 2 human - chr. 2, 3, 4, 7, 12, 13, 15, 16, 17, 20 Post-treatment MRD bone marrow from patient 2 expanded in immune-incompetent mice del CDKN2A, p53 mutation PDX 3 human + chr. X, 14, 18, 21 Relapsed ALL from patient 3 (4-year-old girl) expanded in immune-incompetent mice N/A    45  Similar to the protocol followed in Section 3.2.1, centrosomes were stained with pericentrin (pericentriolar material), centrobin (daughter centriole), and beta-tubulin (microtubules) for the identification of centrosome phenotypes (Fig. 3.6). In B lymphocytes derived from wild-type mice, the interphase cells have either one or two centrosomes with normal size (Fig. 3.6a). In contrast, centrosomes in mouse hyperdiploid B-ALL cells exhibit either sizes larger than 4 µm2 (Fig. 3.6b1) or greater than 2 centrosomes per cell (Fig. 3.6b3). Moreover, fragmentation of pericentriolar material (Fig. 3.6b2), which is commonly seen in cancer cells 183 , was also observed in mouse leukemic cells. The levels of abnormal centrosomes (size and numbers) measured in either interphase and mitotic splenic cells derived from wild-type mice were low. However, close to half of B-ALL mitotic cells contained abnormal centrosome phenotypes whether these cells were derived from Eµ-ret mice (48%) or one of three human PDX mouse models: PDX 1 (41.7%), PDX 2 (49.2%), and PDX 3 (45%) (Fig. 3.7). I also found that up to 25% interphase cells also contained abnormal centrosome (Fig. 3.7) similar to the levels seen in immortal B-ALL cell-lines. In summary, primary B-ALL cells derived from either transgenic mice or patient-derived material possess a higher frequency of centrosome abnormalities as measured in mitotic cells.      46    Figure 3.6 Centrosome phenotype in mouse primary B-lymphocytes and aneuploid B-ALL Centrosomes were stained with pericentrin (PCM), centrobin (daughter centriole), β-tubulin (microtubules), and DAPI in (a) primary B-lymphocytes and (b) primary aneuploid B-ALL cells. All the images are maximum intensity projections of optical sections acquired at 0.5 µm intervals through the cell Z-axis. Scale bar = 4 µm. In each image, the centrosome is highlighted by an arrow in enlarged panel. Scale bar =1 µm.   47    Figure 3.7 The frequency of centrosome amplification in primary B-lymphocytes and B-ALL cells Centrosomes were evaluated in interphase and mitotic cells. Abnormalities include number (red) and size (yellow). For each measurement, more than 20 mitotic (m) and 100 interphase (i) cells were counted.     48  3.2.6 Primary B-ALL cells cocultured on hTERT-MSC are viable during the drug screening protocol Primary B-ALL cells exhibit high levels of spontaneous death ex vivo. To address this problem, I evaluated three protocols for the expansion of primary B-ALL cells derived from PDX3 (Fig. 3.8a). As the co-culture protocol mixes leukemia cells with hTERT-immortalized mesenchymal stromal cells (hTERT-MSCs), it was necessary to use image analysis to directly measure leukemia cell responses. In these experiments, only two centrosome clustering inhibitors (AZ82 and Stattic) were used due to the limited number of B-ALL cells available from PDX 3.  Three culture conditions were tested including: adherent culture alone, adherent culture with CD40 Ig stimulation, and co-culture with hTERT-MSCs. One day prior to seeding with primary B-ALL cells, CD40 Ig or hTERT-MSCs were added to wells in a 96 well plate. A fluorescent cell staining dye (CellTrace, 405/450 nm) was added to track B-ALL cell proliferation and to distinguish B-ALL from hTERT-MSCs. Five hours after cell seeding, six doses of either AZ82 or Stattic were added, as well as separate wells treated with equivalent DMSO. After 48 hours of exposure to drug, B-ALL cells in all three protocols were stained with a live cell staining dye (CyQuant, 480/520 nm), and then the number of live and dead cells were measured by high-content imaging. B-ALL cells were identified by the presence of both Cell Trace and CyQuant whereas hTERT-MSCs, which exhibited larger nuclei, lacked the Cell Trace stain (Fig. 3.8b).  The number of viable cells per well was independently measured through high-content image analysis or manual counting. There was a significant correlation in the results obtained by these two methods for primary B-ALL cells treated with either AZ82 (r2=0.99, P <0.0001) or Stattic (r2=0.98, P <0.0001) (Fig. 3.8c). Consequently, my subsequent measurements utilized automated image analysis.    49   Figure 3.8 The development of ex vivo co-culture model using PDX3 (a) Schematic for the ex vivo primary cell culture procedure, including for: 1. adherent culture alone; 2. CD40 Ig stimulated; or, 3. co-culture with hTERT-MSCs.  (b) Before seeding, primary B-ALL cells were stained with Cell Trace Violet to distinguish them from hTERT-MSCs. After 48 hours of culture on hTERT-MSCs, wells were stained with CyQuant, which is taken up by viable cells. Then, cell viability was measured by high-content imaging. Viable B-ALL cells will be Cell Trace + and CyQuant +, dead B-ALL cells will be Cell Trace + and CyQuant -ve, while viable MSCs will have larger nuclei and be Cell Trace - and CyQuant +. Scale bar=20 µm.  (c) Cell viability measured by high-content image analysis has a significant positive correlation with manual cell counting in AZ82-treated (r2=0.99, P <0.0001) and Stattic-treated (r2=0.98, P <0.0001) primary B-ALL cells.    50  The viability of B-ALL cells derived from PDX 3 was measured using trypan blue dye at thaw, as well as at 24 hours or 48 hours after culture in three conditions. At 24 hours, B-ALL cells had good viability in all three conditions; however, only primary B-ALL cells co-cultured with hTERT-MSCs maintained their viability through 48 hours of tissue culture (Fig. 3.9a).  When I compared the cell killing curves for cells grown on plastic versus cells co-cultured with hTERT-MSCs (including MSCs alone treatment), I found that B-ALL cells grown on plastic died quickly and were much more sensitive to centrosome clustering inhibitors than the cells cultured with hTERT-MSCs (Fig. 3.9b). This potentially indicates the importance of robust viability in ex vivo drug response analyses. In these experiments, I noted that MSCs alone appear quite resistant to centrosome clustering inhibitors as would be expected for cells lacking centrosome amplification. In summary, these experiments revealed that co-culture with hTERT-MSCs improves primary cell survival and enables drug response analysis via high-content imaging.  51   Figure 3.9 Cell viability and drug response of PDX3 in three protocols (a) Analysis of cell viability in three protocols at thaw, and 24 or 48 hours of culture. (mean ± SEM, n=3). (b) After 48 hours of culture in the presence of AZ82 or Stattic, viable cells were counted, and normalized to DMSO. Kill curves were generated in Graphpad and IC50 is indicated by a dash line. (mean ± SEM, n=3).   52  3.3 Key findings 1. Centrosome amplification (both size and number) is a frequent occurrence in aneuploid B-ALL cells, including immortal cell-lines and primary cells derived from transgenic mice or patient-derived xenograft models. 2. Centrosome clustering inhibitors (AZ82, Stattic, Olaparib, PJ34, and Doxorubicin) are effective agents that induce multipolar spindles and death in treated immortal aneuploid B-ALL cells. 3. The level of centrosome amplification in immortal B-ALL cells negatively correlates with the dose of centrosome clustering inhibitors needed to kill those cells. 4. Primary human B-ALL cells co-cultured with hTERT-mesenchymal stromal cells are viable ex vivo for at least 48 hours.  3.4 Discussion Here, I analyzed centrosome phenotypes in B-ALL cell-lines, including mouse aneuploid B-ALL cells and human diploid and near-diploid B-ALL cells, and primary mouse and human B-ALL cells. The results from my study show that the frequency of centrosome abnormalities was lower in interphase cells than in mitotic cells. Moreover, there was no significant difference in the level of centrosome amplification measured at interphase in aneuploid and (near) diploid B-ALL cells. The lower levels of centrosome amplification measured in interphase cells could have a technical component. That is, I defined a cell with 2 centrosomes in interphase as normal however, this cell may be abnormal should it be in G0 or G1 phase. Moreover, my measurement of centrosome size, and the cutoff I used to define an abnormally large centrosome, was qualitative and, thus, may include abnormal sized centrosomes in the population of cells that I deemed to be  53  normal. Finally, my measurements of centrosome size were based on X-Y measurements and did not consider the depth (Z axis) of the centrosome. Thus, I consider the frequency of centrosome amplification measured in mitotic cells to be more reflective of the true levels of abnormalities, which is further supported by the levels of multipolar spindles induced following treatments with centrosome clustering inhibitors.  Across the four B-ALL cell-lines and primary B-ALL cells, a positive correlation between centrosome amplification at mitosis and aneuploidy was observed. I found few abnormal centrosomes in normal splenic cells or hTERT-MSCs but found that 40 - 50% of mitotic hyperdiploid immortal 289 cells contain abnormal centrosomes, and a similar level of abnormalities were seen in primary Eµ-ret hyperdiploid B-ALL cells as well as aneuploid B-ALL cells from PDX1, PDX2 and PDX3. This observation is consistent with a previous study performed in acute myeloid leukemia cells, which showed that the level of numerical and structural centrosome aberrations was higher in cells with chromosome instability 184.  Centrosome amplification is proposed to be associated with the development of a clonal population of potentially preleukemic aneuploid cells 185. Cancer cells divide successfully by clustering supernumerary centrosomes into two pseudo-bipolar spindle poles. Bipolar spindle formation via centrosome clustering is associated with an increased frequency of lagging chromosomes during anaphase 186. Spindle assembly checkpoints (SACs) are responsible for proper chromosome segregation, and heterozygosity for SAC proteins results in increased level of aneuploidy 187–189, as well as tumor formation in mice after long latency 188,189. These studies suggest a role of aneuploidy and/or CIN in tumorigenesis. My study demonstrates a correlation between centrosome amplification and aneuploidy in pediatric B-ALL. Thus, in combination with the findings of earlier studies, my findings suggest centrosome amplification may promote  54  leukemogenesis by introducing chromosome instability and aneuploidy in leukemia initiating cells. However, the role of centrosome amplification in tumorigenesis remains partially understood, and the improvement of centrosome profiling and disease evaluation in pediatric cancers are needed. Supernumerary centrosomes in normal cells can form multipolar spindles, resulting in multipolar cell division or mitotic failure. Multipolar spindles are a hallmarks of many tumor types and can lead to apoptosis 190. Multipolar spindles may also lead to unequal chromosome segregation and may constitute the major generator of aneuploidy and genomic instability in tetraploid cells 191. In tetraploid cancer cells, the high frequency of segregation errors exhibited by newly formed tetraploids is due to the concomitant presence of supernumerary centrosomes and double chromosomes 129. Thus, a putative path exists from supernumerary centrosomes to mitotic failure and tetraploidization to resolve with aneuploidy, such as hyperdiploidy or hypodiploidy.  Centrosome clustering is proposed to be a promising target for cancer therapy. AZ82 acts as an inhibitor of KIFC1 and was discovered using integrated high-throughput synthesis and screening 192. Wu et al., (2013) used AZ82 to target cancer cells with amplified centrosomes and provided evidence that the inhibition of KIFC1 by suppressing its enzymatic activity could be a potential approach for cancer therapy 193. A similar screening approach for chemical inhibitors of centrosome clustering was used to identify Stattic, a STAT3 inhibitor 194. STAT3 is typically responsible for transcription 195. However, the inhibitory effect of Stattic on centrosome clustering is transcription-independent. The authors found that STAT3 regulates the Stathmin-PLK1 cascade and increases gamma-tubulin levels at centrosomes, which promotes centrosome clustering 194. A recent study demonstrated that inhibition of Protein mono-ADP-ribosyltransferase 6 (PARP6) induces multipolar spindles via Chk1 and apoptosis in breast cancer cells 182. Moreover, PARP1/2  55  inhibitors can induce multipolar spindles by regulating nuclear mitotic apparatus protein 1 (NUMA1) directly or indirectly through PARP5a 196. NUMA1 binds to microtubule minus ends and dynactin to regulate the clustering of microtubules for bipolar spindle formation 127. Depletion of PARP results in multipolar spindles, leading to the activation of mitotic checkpoints and metaphase arrest 197. In addition, certain standard-of-care drugs, such as doxorubicin, paclitaxel, and methotrexate, are able to induce multipolar spindles in cancer cells. Doxorubicin has various mechanisms, including DNA intercalation, topoisomerase II inhibition, and generation of free radicals and apoptosis 198–200. In leukemic cells, doxorubicin-treated tumor cells have an increased level of multipolar metaphases in comparison to the control cultures 201. Paclitaxel affects microtubule dynamics 202. In primary breast cancer, paclitaxel- mediated cell death is due to chromosome missegregation through abnormal multipolar mitotic spindles 203. Finally, methotrexate can selectively kill cancer cells with centrosome amplification 204, suggesting that it is a potential centrosome clustering inhibitor. My studies show that centrosome clustering inhibitors (AZ82, Stattic, and PJ34) can induce multipolar spindles and the frequency of centrosome amplification correlates with IC50 values in immortal B-ALL cells. Compared to human diploid B-ALL cells, aneuploid B-ALL cells with higher level of centrosome abnormalities were more sensitive to centrosome clustering inhibitors, especially to AZ82, Stattic, Olaparib, PJ34, and Doxorubicin. To test these inhibitors against primary human B-ALL cells, I required a culture method that overcame the low viability and elevated cell death that is generally observed when primary cells are grown in vitro. Importantly,  56  we show that primary cells co-cultured with hTERT-MSC were able to survive in vitro for a sufficient period for ex vivo analysis of drug responses. Although the majority of centrosome clustering inhibitors were more effective against cells with elevated levels of centrosome amplification, my studies indicate that cells with centrosome amplification may be more resistant to taxol, a mitotic inhibitor. This effect may due to the enhanced microtubule nucleation capacity of cells with supernumerary centrosomes 94. Paclitaxel binds beta-tubulin and stabilizes microtubules, resulting in mitotic arrest and cell death 202. However, clinical studies have found a correlation between the loss of function and/or low expression of BRCA1 and resistance to paclitaxel in different cancers 205–207. The loss of BRCA1 increases microtubule dynamicity, which decreases the sensitivity of microtubules to paclitaxel, preventing the activation of caspase-8 and subsequent cell death 208. Therefore, this study suggests the increased microtubule density caused by centrosome amplification may also reduce the sensitivity to paclitaxel.  Several oncogenic and tumor suppressor proteins are known to localize to the centrosome, deregulation of which may evoke centrosome abnormalities 209. I find the frequency of centrosome amplification is elevated in aneuploid B-ALL cells, including immortal cell-lines and primary cells derived from transgenic mice or patient-derived xenograft models. Therefore, it is reasonable to postulate that centrosome amplification is likely to be widespread in B-ALL cells and quantitation of centrosome amplification could offer valuable information about tumor aggressiveness in patients. However, in order to better distinguish between normal centrosomes and centrosome  57  amplification, and to determine the impacts of the centrosome amplification frequency towards disease aggression, a more precise method for centrosome quantification may be needed.  In conclusion, my work shows that the frequency of centrosome amplification has a negative correlation with drug sensitivity (IC50), especially for B-ALL cells treated with either AZ82 or Doxorubicin. It is possible, therefore, that studying the extent of centrosome amplification, and its correlation with tumor grade, may be very useful to identify patients who might benefit from declustering drugs.   58  Chapter 4: Centrosome clustering pathway is targetable in primary aneuploid B-ALL  4.1 Rationale and hypothesis Immortal cell-lines generally have high proliferation rates in vitro and are amendable to culture and manipulation (e.g. transfection) making them indispensable tools for scientific discovery. However, cell-lines often show different genetic and phenotypic morphology from original patient-derived material 210,211, which may be due to their adaptation to the two-dimensional culture environment. Although primary cells isolated directly from tissues have a limited lifespan and restricted proliferation potential in tissue culture, it is generally accepted that they better represent human cell morphology and physiology 210,211. Therefore, primary human cancer cells are excellent model systems for preclinical studies of drug responses.  In Chapter 3, I found centrosome amplification is prevalent in human or mouse immortal B-ALL cell-lines, and the inhibition of centrosome clustering is an effective way to prevent the proliferation of these immortal B-ALL cells. Moreover, I evaluated three different culture conditions to optimize the ex vivo viability and proliferation of primary human B-ALL cells.  Here, I hypothesize that centrosome amplification is common in human pediatric leukemia cells derived directly from patients (primary samples), and the inhibition of centrosome clustering pathways will be more cytotoxic to leukemic cells with lower toxicity to primary cells derived from normal tissues.      59  4.2 Results 4.2.1 Ex vivo analysis of primary B-ALL cells co-cultured on hTERT-MSC provides reproducible measurements of centrosome amplification and drug responses Primary samples were sourced through the BCCH Biobank, including 45 primary B-ALL samples (5 hypodiploid, 15 hyperdiploid, 28 euploid) and 6 control stem cell samples (Table 4.1).  Leukemia cells were derived from bone marrow samples taken from B-ALL patients while normal primary stem cells were derived from non-involved bone marrow or peripheral blood samples taken from siblings or autologous donors.  Table 4.1 Overview of patient sample characteristics Samples N Age (year) Sex SV (t+, fusion) Risk     Male Female No Yes Av. High V. High Controls Sibling BM 2 12.6 ± 5.0 100% 0%      Autologous Stem cells 4 9.3 ± 6.3 75% 25%      Diagnosis Euploid 15 6.1 ± 4.9 66% 33% 40% 60% 53% 40% 7% Hyperdiploid 11 5.8 ± 4.7 18% 82% 55% 45% 55% 44% 0% Hypodiploid 5 9.2 ± 5.3 60% 40% 60% 40% 20% 20% 60% Relapse Euploid 10 11.1 ± 4.5 *14.6 ± 4.6 66% 33% 60% 70%    Hyperdiploid 4 7.8 ± 5.5 *12.6 ± 6.9 50% 50% 100% 0%    Note: Mean age at diagnosis is provided with * indicating the mean age at relapse, if applicable.     60  A challenge with the analysis of primary samples is the limited amount of material available and the potential for differences between aliquots, a so-called batch effect.  I prioritized the evaluation of responses to the top five inhibitors as determined in Chapter 3, including: AZ82, Stattic, PJ34, Olaparib, and Doxorubicin. However, it was not feasible to measure these five inhibitors against redundant aliquots of every patient sample. Thus, I first chose primary samples with the highest documented number of cells per aliquot, including: three B-ALL samples (each clinically designated high risk) and one normal stem cell sample (Table 4.2). The high number of cells per aliquot in these samples allowed me to test all five inhibitors against a single aliquot. Thus, I performed redundant measurements on two separate aliquots to assess the reproducibility of measurements across aliquots.  Table 4.2 Characteristics of human primary cells derived from B-ALL involved (n=3) or non-involved (n=1) bone marrows CELLS ALIQUOT KARYOTYPE ORIGIN CYTOGENETICS SPECIMEN TYPE (%BLAST) C01052 100M per 1.3ml N/A Autologous peripheral blood stem cells from a 1-year-old boy with relapsed choroid plexus carcinoma N/A Stem cells (N/A) C00023 R3 15M Per 1ml N/A BCP-ALL from 16-year-old boy at 3rd relapse t(1;5)(q21;q33), t(6;9)(p21;p13), rea(9),t(13 ;17)(q32;q21),  t(15 ;16)(q24;q13) t(6;9) Whole bone marrow (90%) C00652 51.3M Per 1ml 54, XY + chr. X, Y, 9, 9, 14, 14, 21, 21 Relapsed BCP-ALL from 17-year-old boy N/A Mononuclear cells  (74%)  C00373 12.8M Per 1ml Euploid Diagnosis B-ALL from an older than 10-year-old boy N/A N/A (90%)   61  I used the protocol followed in Section 3.2.6 to co-culture primary cells with hTERT-MSCs, which were seeded 24 hours prior to seeding with primary cells. A fluorescent cell staining dye (CellTrace, 405/450 nm) was added to the primary cells (B-ALL or control stem cells) prior to seeding on hTERT-MSC feeders to distinguish B-ALL cells and allow the subsequent measurement of viability. Wells were then treated with one of six graded doses each of five selected centrosome clustering inhibitors, as well as separate wells treated with equivalent DMSO. After 48 hours of exposure to drug, B-ALL cells and hTERT-MSC were stained with a live cell staining dye (CyQuant, 480/520 nm), and then the number of live and dead cells per well was automatically measured by high-content imaging. Cell viability was normalized to that measured in wells treated with DMSO control. Primary B-ALL cells and normal stem cells, as well as hTERT-MSC, were fixed in ice-cold methanol and stained with three different centrosome markers for the analysis of centrosome amplification. Again, the centrosome identity was defined by co-staining the centrosome component (γ-tubulin) with an internal centriole (CP110) and the nucleation of microtubules (tubulin B). Two separate aliquots of four primary samples (Table 4.2) were thawed and the frequency of centrosome amplification and IC50 values of selected centrosome clustering inhibitors was measured in these separate aliquots. Figure 4.1 shows the correlation in the results obtained by these two experiments in the analysis of centrosome amplification and IC50 values for AZ82-, PJ34-, and Doxorubicin-treated primary cells. Measurements were significantly correlated across the two distinct aliquots for centrosome amplification frequency (n= 4 values, r2= 0.97, P<0.05) and IC50 values for the three different drugs (n=12 values, r2=0.94, P<0.001) (Fig. 4.1), indicating these measurements are reproducible across biological replicates of patient samples.    62   Figure 4.1 The correlation between two experiments (CA and IC50) in human primary cells (a) Biological replicate measurements of centrosome amplification in distinct aliquots taken from three primary human B-ALL patients and one patient with control non-involved stem cells. Statistically significant correlation was observed for the values measured in replicate aliquots (r2=0.97, P<0.05).  (b) Biological replicate measurements of IC50 concentrations (log10 µM) for AZ82-, PJ34-, and Doxorubicin-treated primary B-ALL cells. Statistically significant correlation was observed for the values measured in replicate aliquots (r2=0.94, P<0.001).   63  4.2.2 Stattic has poor selectivity against primary leukemia cells relative to normal stem cells and hTERT-MSC  In these initial studies, I was also interested in the potential for selective toxicity of individual drugs against primary B-ALL cells relative to control stem cells. Thus, I compared IC50 values obtained against the three primary human B-ALL samples and one control stem cell sample, as well as hTERT-MSC alone. It is important to note that the evaluation of primary cell responses to Stattic were performed on a single aliquot of patient samples as these initial measurements revealed poor selectivity against cancer cells.  Based on the experiment in Section 4.2.1, I applied a nonlinear regression model (log (inhibitor) vs response) to determine the drug kill curve, and specifically the IC50 value for each drug. As shown by the dose response curves in Figure 4.2, primary B-ALL cells were more sensitive than normal stem cells, or hTERT-MSCs alone, to the selected centrosome clustering inhibitors. Responses to Stattic (top, RHS) were an exception to this pattern as the normal stem cell sample (C01052, red line) overlapped with the responses of primary B-ALL samples (black lines).    64    Figure 4.2 Efficacy of centrosome clustering inhibitors against human hTERT-MSC and primary cells (a) TERT-immortalized mesenchymal stromal cells (hTERT-MSCs, yellow line) and primary cells derived from non-involved bone marrow (red line), or primary cells derived from B-ALL patient marrows (black lines), were co-cultured with hTERT-MSC and separately exposed to serial dilutions of AZ82, Stattic, Olaparib, PJ34, or Doxorubicin for 48 hours. Primary B-ALL and normal stem cells were stained with Cell Trace Violet to track the proliferation of B-ALL or B-lymphocytes. CyQuant was used to detect live cells. Cell viability was determined using high-content image analysis. Viability in each dose was normalized to that in cells treated with DMSO control. The kill curve for each centrosome clustering inhibitor was plotted and IC50 is indicated by a dash line and (b) IC50 values are tabulated.   65  To better visualize the selectivity of drug responses, the levels of mitotic centrosome amplification and the IC50 values (from Section 4.2.1) for the 5 drugs were plotted as a heatmap for the three primary B-ALL cells, one normal stem cell, and hTERT-MSC cell-line (Fig. 4.3). In this heatmap dataset, values were hierarchically clustered using average linkage (Euclidean) and scaled horizontally with yellow indicating low values and red indicating high values. The levels of centrosome amplification (CA) and the IC50 dose for the four primary patient cells was correlated by Pearson r (vs CA) to generate a P value of each drug. Negative correlation between the levels of centrosome amplification and the IC50 concentration was observed across the four primary samples and one immortal line with a significant negative correlation observed for AZ82 (r= -0.97, P= 0.03) and Olaparib (r= -0.95, P= 0.049) and trends observed for PJ34 (r= -0.94, P=0.058) and Doxorubicin (r= -0.86, P=0.14). However, the correlation between centrosome amplification frequency and drug sensitivity for Stattic (r= -0.78, P= 0.22) was weak.  As Stattic treatment is known to have broad effect on other pathways, such as immune response 212,213, we excluded Stattic from subsequent analyses of primary samples and proceeded with the further evaluation of four inhibitors (AZ82, PJ34, Olaparib, and doxorubicin) against a larger cohort of human patient samples (both B-ALL and non-involved marrow samples).    66    Figure 4.3 Correlation between the frequency of mitotic centrosome amplification (CA) and the measured IC50 values for drug treatment of human primary stem cells, B-ALL and hTERT-MSC Heatmap display of hierarchical clustered IC50 values from five centrosome clustering inhibitors against hTERT-MSC, primary stem cells and primary B-ALL cells. Mitotic centrosome amplification (CA) frequency in hTERT-MSCs and B-ALL cells, interphase CA frequency in normal stem cells, and IC50 values were scaled horizontally. Red or yellow colors indicate high or low values, respectively. The dendrogram displayed on top was based on hierarchical clustering of the samples (both CA and IC50 data) using the average linkage method. Pearson r (vs. CA) is calculated as is the P value. (the correlation and p values only include four primary patient cells, *P<0.05).   67  4.2.3 IC50 doses for centrosome clustering inhibitors show selectivity against primary B-ALL relative to stem cells derived from non-involved bone marrows To query whether centrosome clustering inhibitors inhibit the proliferation of pediatric B-ALL cells with lower cytotoxic effects on normal cells, I performed drug response analysis in three additional normal stem cell samples (n= 4 total) and ten additional pediatric B-ALL samples (n= 11 total) (Table 4.3).   Table 4.3 Characterization of 4 patient normal stem cells and 11 patient B-ALL  CELLS KARYOTYPE ORIGIN CYTOGENETICS SPECIMEN TYPE (%BLAST) C03696 Normal Autologous peripheral blood stem cells from a 15-year-old girl with small cell carcinoma N/A Stem cells (N/A) C03938 Normal Autologous bone marrow stem cells from a 6-year-old boy with embryonal sarcoma N/A Stem cells (N/A) C02571 Normal Autologous peripheral blood stem cells from a 13-year-old boy with malignant lymphoma N/A Stem cells (N/A) C01052 Normal Autologous peripheral blood stem cells from a 1-year-old boy with relapsed choroid plexus carcinoma N/A Stem cells (N/A) C00652 54, XY + chr. X, Y, 9, 14, 21 Relapsed BCP-ALL from 17-year-old boy N/A Mononuclear cells (74%) C00373 Euploid Diagnosis B-ALL from an older than 10-year-old boy N/A N/A (90%)  68  C00023 R3 N/A BCP-ALL from 16-year-old boy at 3rd relapse t(1;5)(q21;q33), t(6;9)(p21;p13), rea(9),t(13 ;17)(q32;q21), t(15 ;16)(q24;q13) t(6;9) Whole bone marrow (90%) C03632 28, XY  Diagnosis B-ALL from a 7-year-old boy N/A Mononuclear cells (93%) C00023 R2 Euploid BCP-ALL from 13-year-old boy at 2nd relapse abnormal 2p, 3p, 5q, 6p, 14q, 16q, t(8;9) and t(13;17) Whole bone marrow (96%) C00115 Euploid Relapsed BCP-ALL from 10-year-old girl t(16;22), 37% of interphase nuclei have BCR/ABL dual fusion Whole bone marrow (82%) C00189 49-52, XX + chr. X, X, 5, 8, 10, 21, 21 - chr. 20 Relapsed BCP-ALL from 16-year-old girl t(2,8)(p13;p21), del(9)(p21.3), dup (17)(q24.2qter) Whole blood (93%) C01818 Euploid Relapsed BCP-ALL from 6-year-old girl del(9)(p21.3)(CDKN2A-, cen9+) Mononuclear cells (98%) C00269 47, XX + chr. 21c Relapsed BCP-ALL from 11-year-old girl N/A Whole bone marrow (99%) C02100 Euploid Relapsed BCP-ALL from 19-year-old boy t(9;22)(q34;q11.2)(ABL1+,BCR+,ABL1+) Mononuclear cells (54%) C00118 47, XY + chr. 21c Relapsed BCP-ALL from 15-year-old boy N/A Mononuclear cells (88%)   Cell viability, which was scored by high-content imaging and automated analysis, was normalized to that measured in DMSO control wells. I applied a nonlinear regression model (log (inhibitor) vs response) to determine the kill curve, and specifically IC50 value for each drug. Figure 4.4 shows dose responses for each of the four normal control samples (red lines). Dose responses for each of the eleven B-ALL samples (black lines) are plotted with a mean dose response curve for the four normal samples (red line) (Fig. 4.5). A mean dose response curve for the four normal samples (red line) and a mean dose response curve for eleven B-ALL samples  69  (black line) are plotted together to better demonstrate the heightened sensitivity observed for primary B-ALL cells, with IC50 values for each sample shown in the boxes (Fig. 4.6).     Figure 4.4 Drug responses of human normal stem cells treated with centrosome clustering inhibitors Primary cells co-cultured with hTERT-MSC were separately exposed to serial dilutions of the indicated drug for 48 hours. Primary normal stem cells were stained with Cell Trace Violet. CyQuant was used to detect live cells using high-content image analysis. Viability was normalized to that in cells treated with DMSO. The kill curve for each centrosome clustering inhibitor is plotted. The plot indicates the drug responses of stem cells derived from non-involved bone marrow with the 4 samples (red lines). IC50 is indicated by a dash line.    70    Figure 4.5 Drug responses of human primary B-ALL cells treated with centrosome clustering inhibitors 11 primary samples of pediatric B-ALL cells co-cultured with hTERT-MSC were separately exposed to serial dilutions of the indicated drug for 48 hours. Primary B-ALL cells were stained with Cell Trace Violet. CyQuant was used to detect live cells using high-content image analysis. Viability was normalized to that in cells treated with DMSO. The plot indicates the drug responses of 11 primary B-ALL samples with the mean of the four normal samples (red line) plotted as a comparator. IC50 is indicated by a dash line.  71    Figure 4.6 Efficacy of centrosome clustering inhibitors in human primary cells The plot indicates the mean drug responses of 4 normal samples (red line) and 11 primary B-ALL samples (black line). IC50 is indicated by a dash line.  For the HSET inhibitor AZ82, the IC50 values for the four patient samples derived from non-involved bone marrow stem cells (red lines) were similar and ranged from 2.5 to 15 µM with a mean value of 10.4 ± 4.0 µM; conversely, the IC50 values for AZ82 against each of the 13 primary B-ALL cells co-cultured on hTERT-MSC (black lines) ranged from 1.07 to 5 µM or up to 10-fold more sensitive than the mean value for the normal stem cell controls.  For the PARP inhibitor Olaparib, the IC50 values for the four patient samples derived from non-involved bone marrow stem cells (red lines) were similar and ranged from 6 to 10 µM with a mean value of 7.1 ± 1.4 µM; conversely, the IC50 values for Olaparib against primary B-ALL  72  cells co-cultured on hTERT-MSC ranged from 0.1 to 10 µM or up to 70-fold more sensitive than the mean value for the normal stem cell controls.  For the PARP inhibitor PJ34, the IC50 values for the four patient samples derived from non-involved bone marrow stem cells (red lines) were similar and ranged from 12.6 to 44.7 µM with a mean value of 27.2 ± 9.9 µM; conversely, the IC50 values for PJ34 against primary B-ALL cells co-cultured on hTERT-MSC ranged from 0.3 to 28.2 µM or up to 100-fold more sensitive than the mean value for the normal stem cell controls.  For the standard-of-care drug Doxorubicin, the IC50 values for the four patient samples derived from non-involved bone marrow stem cells (red lines) were similar and ranged from 270 to 631 nM with a mean value of 461.8 ± 138.4 nM; conversely, the IC50 values for Doxorubicin against primary B-ALL cells co-cultured on hTERT-MSC ranged from 5 to 141.3 nM or up to 100-fold more sensitive than the mean value for the normal stem cell controls.  While responses to centrosome clustering inhibitors were variable across patient samples, these data indicate consistently much lower concentrations (up to 100-fold) were needed to achieve 50% cytotoxicity against primary pediatric B-ALL samples than that measured against stem cells derived from non-involved bone marrow samples.     73 4.2.4 Selectivity of centrosome clustering inhibitors correlates with the frequency of centrosome amplification To better visualize the correlation, if any, between the level of centrosome amplification in mitotic cells from primary samples and drug sensitivity, these measurements (from Section 4.2.3) were plotted as a heatmap for the 15 primary samples (Fig. 4.5). It is important to note, however, that I measured the frequency of centrosome amplification in interphase cells for stem cells derived from non-involved bone marrows. The plot was hierarchically clustered using average linkage (Euclidean) and scaled horizontally with yellow indicating low values and red indicating high values. The levels of centrosome amplification (CA) in mitotic cells and the IC50 for each drug was correlated by Pearson r (vs CA) to generate a P value. These correlations indicate that centrosome amplification frequency in primary patient samples is negatively correlated with the IC50 values of selected centrosome clustering inhibitors; while this negative correlation is significant for AZ82 (r=-0.80, P<0.001), PJ34 (r=-0.60, P=0.021), and Doxorubicin (r=-0.78, P<0.001), there was also a trend observed for Olaparib (r=-0.47, P=0.075).  For these analyses, I selected 11 B-ALL samples from 45 samples received from the BCCH Biobank primarily based upon the documented number of cells per sample aliquot rather than to draw any inference about the putative association between centrosome amplification and response to CC inhibitors or to severity of disease. However, hierarchical clustering of the data clearing separated the four primary samples derived from non-involved bone marrows, which had low levels of centrosome amplification and were relatively insensitive to centrosome clustering inhibitors. Two other primary B-ALL samples had relatively low levels of centrosome amplification (C00269: 19.1%; and, C02100: 25%); interestingly, these samples were clinically stratified into the average risk group and unknown risk, respectively, and showed relative   74 insensitivity to the HSET inhibitor AZ82. Overall, my studies indicate that primary patient B-ALL cells with a higher frequency of centrosome amplification are more sensitive to centrosome clustering inhibitors. The elucidation of a putative association between disease risk, centrosome amplification and sensitivity to centrosome clustering inhibitors will require the further evaluation of additional primary B-ALL samples.   Figure 4.7 The correlation between CA and IC50 in human primary stem cells and B-ALL Heatmap plot for the frequency of centrosome amplification (CA) and IC50 values for four CC inhibitors against four primary non-involved stem cells and eleven primary B-ALL samples. Mitotic CA in B-ALL, interphase CA in normal stem cells, and IC50 values were scaled horizontally. Red or yellow colors indicate high or low values of CA or IC50. The dendrogram displayed on top was based on hierarchical clustering of the samples using the average linkage method indicating the strong relation of 11 primary B-ALL cells, and 4 normal stem cells. Pearson r (vs. CA) is calculated as is the P value. Corresponding patient information about risk and stage is showed in the bottom rows. Black or white colors indicate high or low risk. (N=normal, Dx=diagnosis, R=relapse, *P<0.05, ***P<0.001).   75 4.3 Key findings 1. The frequency of centrosome amplification and IC50 values show high reproducibility in biological replicates from 3 primary B-ALL samples and 1 normal stem cell sample. 2. Stattic, a Stat3 inhibitor, is poorly selective against 3 primary B-ALL samples relative to 1 normal stem cell sample. 3. Relative to normal stem cells, high risk primary B-ALL cells are more sensitive to centrosome clustering inhibitors (AZ82, Olaparib, PJ34, and Doxorubicin).  4. For primary patient B-ALL samples, sensitivities to AZ82, PJ34, and Doxorubicin significantly correlate with the frequency of centrosome amplification.  4.4 Discussion Centrosome amplification is so frequently detected in human tumors 214–216  that it is proposed as a candidate hallmark of cancer cells. After the optimization of my protocols and the selection of candidate inhibitors using immortal cell-lines and xenograft expanded patient-derived samples, I studied the presence of centrosome amplification in 11 patient B-ALL samples using antibodies to three different centrosomal proteins. The results from my study show that the frequency of centrosome amplification varies from 19% to 50% in primary B-ALL mitotic cells; in contrast, primary stem cells derived from non-involved bone marrows show centrosome amplification in less than 5% of interphase cells. I studied interphase cells rather than mitotic cells in normal stem cells due to the very limited number of phospho-histone H3 positive mitotic cells that I observed in the four primary samples. Consistently, human hematopoietic stem cells are known to only average a replication rate of once every 40 weeks 217. However, it is published that   76 non-malignant cells, including human foreskin fibroblasts and human embryonic stem cell-derived cells, have very low rates (2 - 5%) of multi-centrosomal mitosis 218.  Centrosome abnormalities are frequently present in patients with chromosome or DNA index abnormalities. Abnormal centrosomes can directly lead to the uneven distribution of chromosomes in cancer cells and result in ongoing chromosomal instability and aneuploidy 219,220. In my study, patient B-ALL cells with aneuploidy, chromosome aberrations or translocations (Table 4.3) also had high levels of centrosome amplification (Fig. 4.5). With the exception of C00373, 10 of the 11 pediatric B-ALL samples had genomic or genetic abnormalities, and 90% (9/10) of these samples had high centrosome amplification frequency. Although the cohort number is small, my results suggest that centrosome amplification associates with chromosome abnormalities in pediatric leukemia, which is also described in previous studies 181,221. In Chapter 3, I identified five centrosome clustering inhibitors that induced high cytotoxicity against immortal aneuploid B-ALL cell-lines. However, one of these inhibitors, termed Stattic, showed poor selectivity against primary B-ALL cells. This lack of selectivity might be due to the broad signaling pathways regulated by STAT3, including in the immune system; indeed, Stat3 gene deletion in mice leads to embryonic lethality  212,213. Thus, I restricted my latter experiments to the study of responses to AZ82, PJ34, Olaparib, and Doxorubicin in a larger cohort of primary cells. In solid tumors, the frequency of centrosome amplification appears to be more prevalent in high-risk diseases 183. Here, I show that high-risk B-ALL patient samples had high levels of centrosome amplification (35-50%). I observed two outlier samples (C00269 and C02100), which were not risk stratified as high risk and had lower levels of centrosome amplification (19.1% and 25%, respectively). Interestingly, C00269 also show relative insensitivity to the HSET inhibitor   77 AZ82. This may indicate that a requirement for centrosome clustering is a prerequisite for the killing of leukemia cells by AZ82.  The relapse sample C00118 had high centrosome amplification (50%) but about the same sensitivity as the normal samples to PARP inhibitors Olaparib (10 µM) and PJ34 (28.2 µM). In addition, a hypodiploid B-ALL sample showed relative insensitivity to Olaparib (C03632: 5 µM), but had high frequency of centrosome amplification (C03632: 35%). The relative insensitivity to PARP inhibitors despite a high level of centrosome amplification might be due to the mutation of TP53, which is frequently mutated in hypodiploid and relapsed ALL cells. The p53 tumor suppressor responds to DNA damage and triggers cell cycle arrest, DNA repair or apoptosis 222,223. Indeed, p53 loss in human breast cancer or colorectal cancer cells associates with increased resistance to PARP-mediated cell death 224. Thus, pediatric hypodiploid B-ALL samples may be less sensitive to PARP inhibitors due to the mutation of TP53. While current treatments have led to a significant improvement in outcomes for pediatric patients, the prognosis for adult patients with B-ALL still remains poor. While most adult patients with leukemia respond to chemotherapy, only 30-40% of them will achieve durable remission 225. Therefore, personalized treatments for adult ALL are needed to improve the clinical outcome. High hyperdiploidy is one of the most prevalent cytogenetic subgroups in adult B-ALL 226 and can often coexist with t(9;22) 227; about 30% of adult B-ALL with high hyperdiploidy harbour a Ph 228. While Ph-positive patients with high hyperdiploidy have improved outcome when compared with Ph-positive patients without high hyperdiploidy after induction chemotherapy 229, new treatments are still needed for these aggressive diseases. My findings in pediatric B-ALL patient samples suggest that targeting centrosome clustering pathway may also be an effective treatment for adult hyperdiploid ALL.    78 Chapter 5: Centrosome clustering inhibitors induce genome instability and the expression of pro-inflammatory signals in refractory aneuploid B-ALL   5.1 Rationale and hypothesis In Chapter 3 and 4, I found the inhibition of centrosome clustering was an effective way to prevent the proliferation of pediatric B-ALL cells, both immortal cell-lines and primary B-ALL cells derived directly from patients. However, tumor cells generally respond to targeted therapies but often a population of cells becomes refractory. While genome instability is a hallmark of cancer and a proposed enabling characteristic for relapse 179, it also improves the recognition of cancers by immune system: high mutational burden and activation of innate immune responses. It is proposed that an elevated mutational burden is a strong indicator for the success of immune checkpoint inhibitors 230. Similarly, micronuclei 144,145 and the improper chromosome segregation 231 are linked with cytosolic DNA sensing and activation of innate immunity.  Here, I hypothesize that centrosome clustering inhibitors will generate elevated genomic instability and pro-inflammatory signals in the population of surviving and refractory B-ALL cells.    79 5.2 Results 5.2.1 Primary pediatric B-ALL cells exhibit micronuclei following treatment To test whether centrosome clustering inhibitors induce genomic instability in the refractory primary B-ALL and normal stem cells, I measured the frequency of micronuclei, stained via the DNA staining dye CyQuant, in primary cells that survived treatment with approximately IC50 doses of AZ82, Olaparib, PJ34, and Doxorubicin (Section 4.2.3). Because the size of B-lymphocytes and B-ALL cells are small, I was not able to use automatic detection of micronuclei with the high-content image analysis program. Therefore, I manually measured the frequency of primary B-ALL cells (Cell Trace Violet) with micronuclei (Fig. 5.1a).    Figure 5.1 The frequency of micronuclei in DMSO and CC inhibitors treated primary cells (a) Representative images of micronuclei (CyQuant) phenotype in centrosome clustering inhibitor treated primary B-ALL cells (CellTrace Violet). Arrowhead indicates micronucleus. Scale bar=5µm.  (b) Averaged micronuclei frequency in DMSO, AZ82, Olaparib, PJ34, and Doxorubicin treated 4 normal stem cells (SC) and 11 primary B-ALL cells. Experiment was performed once with duplicated seeding. Each dot represents one sample’s micronuclei frequency measured from total 10 fields of view. (mean ± SEM, *P<0.05, **P<0.01***P<0.001, ANOVA).   80 For each condition, I analyzed 10 fields of view for the frequency of micronuclei. I compared the mean micronuclei frequency for the four normal samples (red dots) with the mean micronuclei frequency for the 11 B-ALL samples (black dots) following treatment with ~IC50 doses of one of four centrosome clustering inhibitors or DMSO-treated controls (Fig. 5.1b). In DMSO-treated primary cells, there was no significant difference in micronuclei frequency between normal stem cells and primary leukemia samples. However, the frequency of micronuclei in primary B-ALL cells treated with the centrosome clustering inhibitors was significantly higher than that observed in drug-treated non-involved bone marrow stem cells. These data indicate that centrosome clustering inhibitors increase the formation of micronuclei in primary B-ALL cells, suggesting the induction of elevated genome instability in the surviving, and potentially drug-insensitive, pediatric leukemic cells. To better visualize the association between the level of mitotic centrosome amplification in the treated population (from Section 3.2.1) and the micronuclei frequency across samples, I plotted these values as a heatmap for the four non-involved bone marrow and eleven primary B-ALL cells. The data was hierarchically clustered using average linkage (Euclidean) and scaled horizontally with yellow or red color indicating low or higher values, respectively (Fig. 5.2). Correlation between the levels of centrosome amplification and micronuclei frequency was measured by Pearson r (vs CA), which revealed a significant positive correlation for all drugs. That is, primary cells with high CA had more micronuclei following exposure to centrosome clustering inhibitors. These findings indicate that populations of cells with a higher frequency of centrosome amplification are more likely to contain micronuclei, an indicator of genome instability, after treatment with centrosome clustering inhibitors.      81   Figure 5.2 Correlation between CA and MN in human primary stem cells and B-ALL Heatmap plot for the frequency of centrosome amplification (CA) and micronuclei (MN) following treatment with four CC inhibitors or DMSO in four primary non-involved stem cells and 11 primary B-ALL samples. Mitotic CA in B-ALL, interphase CA in normal stem cells, and MN frequency are scaled horizontally. Red or yellow colors indicate high or low values, respectively. The dendrogram displayed on top is based on hierarchical clustering using the average linkage method. Pearson r (vs. CA) is calculated and the P value is indicated. Patient sample information about risk and stage is showed in the bottom rows. Black or white colors indicate high or low risk, respectively. (N=normal, Dx=diagnosis, R=relapse, ***P<0.001).    82 5.2.2 Primary pediatric B-ALL cells show cGAS localized to micronuclei Due to an insufficient or porous nuclear envelope, micronuclei are frequently exposed to the cytoplasm and are capable of activating the cytosolic DNA sensing pathway, cGAS-STING. So, I determined whether the cytosolic DNA sensing pathway is activated by micronuclei that follow exposure to centrosome clustering inhibitors in drug-insensitive primary B-ALL cells. To do so, I recovered B-ALL cells following their exposure to approximately IC50 doses of AZ82 (1µM), Olaparib (1µM), PJ34 (1µM), or Doxorubicin (100nM). Unfortunately, immunofluorescence detection of cGAS-positive micronuclei was not possible in 96 well plates after treatment (as was performed for micronuclei detection in Fig. 5.1), as the washing steps needed for the immunofluorescence protocol remove the inhibitor-insensitive cells from the wells. So, I transferred the drug-treated cells to poly-L-lysine coated coverslips prior to the immunofluorescence protocol. For this analysis, I prioritized the evaluation of cGAS-micronuclei co-localization using the primary B-ALL sample with the highest documented number of cells per aliquot (C00652).  C00652 was first exposed to each drug, or DMSO control, and surviving cells were recovered after 48 hours and seeded on poly-L-lysine coated coverslips in a 24-well plate. Approximately 3x105 cells were required for each dose for each coverslip. Cells were then fixed and stained with DAPI and cGAS for immunofluorescence analysis (Fig. 5.3a). Ten fields of view were selected and the micronuclei frequency and cGAS co-localization was measured in each dose. Again, hTERT-MSC were distinguished from B-ALL cells based on nuclear size and cell shape. Drug-treated primary B-ALL cells (C00652) had an increased level of micronuclei compared to DMSO-treated cells. Moreover, a large proportion of cells with micronuclei had cGAS co-localized to the micronuclei, especially in cells treated with centrosome clustering inhibitors (Fig.   83 5.3b). While this putative association between micronucleation and cGAS activation requires the evaluation of additional primary B-ALL samples, this data reveals the potential activation of the cGAS-STING DNA sensing pathway by micronuclei induced in inhibitor-insensitive primary leukemia cells.   Figure 5.3 cGAS co-localized to MN in primary B-ALL cells (C00652) (a) Primary B-ALL cells were fixed and stained with DAPI (nucleus) and cGAS by immunofluorescence. Scale bar=2µm.  (b) Percentage of primary B-ALL cells with micronuclei. Cells were fixed and stained with DAPI (nucleus) and cGAS 48 hours after either treatment with DMSO or IC50 doses from four CC inhibitors. In each bar, the inside gray bar shows the percentage of cell with cGAS-bound MN. (one experiment, >100 cells per bar).   84 5.2.3 Generation of 289 cells resistant to centrosome clustering inhibitors  Because of the limited amount of primary material available, it was not feasible to further evaluate the association between micronuclei and cGAS-STING activation in patient samples. So, I used immortal hyperdiploid (mouse) 289 cells to generate sublines that are selectively resistant to one of two centrosome clustering inhibitors: AZ82 or PJ34. To do so, 289 cells were passaged 24 hours before seeding in a 12-well plate. These cells were separately exposed to an IC50 dose of each drug or equivalent DMSO control. The drug, or DMSO, was refreshed every two days for 2 weeks. Then, the drug concentration was increased by 33% (1.33- fold) for an additional 2 weeks. This cycle was repeated six times until we generated a population of 289 cells that were resistant to lethal doses of each drug (Fig. 5.4a). These sublines were termed 289r (applied drug) cells. The overall growth characteristics of the 289 resistant cells (289r), as measured by doubling time, were similar when grown in the presence of DMSO (Fig. 5.4b). When grown in lethal doses of centrosome clustering inhibitors, however, the two 289r cells each demonstrated a selective growth advantage (Fig. 5.4b). These data indicate that the drug selection process did not significantly alter the selected cell population’s basal growth rate. Moreover, the selected resistant populations may not be utilizing a broadly-acting resistance mechanism(s), such as a generalized drug efflux mechanism, as the sublines were selectively resistant to the applied drug.     85  Figure 5.4 Generation of the refractory B-ALL cells lines (289r) (a) Schematic diagram for the in vitro generation of resistant B-ALL cells. Hyperdiploid 289 cells received either DMSO or increasing concentrations of centrosome clustering inhibitors (AZ82 or PJ34) for approximately 4 months. (b) 289r cell-lines have selective growth advantages to applied drugs. Viable cells were counted at 0, 24, 48 and 72 hours. DMSO or drug containing media was refreshed at 48 hours. (mean ± SEM, n=3).   86 5.2.4 289 resistant cell-lines exhibit higher basal levels of micronuclei Approximately 5% of drug-insensitive primary B-ALL cells contain micronuclei following their treatment with IC50 doses of centrosome clustering inhibitors (Fig. 5.1b). To determine how closely, or not, the 289r B-ALL cells model the micronucleation rate observed in drug-insensitive primary B-ALL cells, I used high-content imaging to measure micronuclei in the 289r sublines. In each experiment, 10 fields of view were selected and micronuclei frequency (%) was measured (Fig. 5.5a). As a positive control to induce micronuclei, parental 289 cells were subjected to graded doses of irradiation. In these control experiments, I observed dose-dependent increases in the frequency of micronuclei in the population of parental 289 cells (Fig. 5.5b). Notably, 289r cells were continuously grown in medium containing the drug or equivalent DMSO. For 289r (AZ82) and 289r (PJ34) sublines, I observed a significantly increased frequency of micronuclei relative to 289r (DMSO) control cells (Fig. 5.5b), and these increased levels approximated those observed in drug-insensitive primary cells (3% – 8%, Fig. 5.1b).     87  Figure 5.5 289r cells refractory to centrosome clustering inhibitors show elevated levels of micronuclei (a) 289 cells were stained with DAPI to identify nuclei and micronuclei (arrowhead). Scale bar=5µm.  (b) Micronuclei frequency in 289r (DMSO), 289r, and parental 289 cells one day after irradiation. In each experiment, 10 fields of view were analyzed. (mean ± SEM, n=3 experiments, >500 cells per bar, *P<0.05, **P<0.01, ANOVA).       88 5.2.5 289 resistant cell-lines show cGAS located at micronuclei Centrosome clustering inhibitors induce micronuclei in refractory primary and immortal B-ALL cells, and cGAS co-localized to micronuclei in drug-treated primary B-ALL cells (Fig. 5.3b). The micronucleation rate observed in 289r cells approximates that observed in the drug-treated primary B-ALL cells. Thus, 289r cells may serve as models that allow me to better understand the downstream events. So, I first examined the localization of cGAS to micronuclei in 289r cells. Again, parental 289 cells were subjected to graded doses of irradiation as a positive control to induce micronuclei. Prior to the immunofluorescence protocol, 289 cells were allowed to attach to poly-L-lysine coated coverslip. In each condition, 10 fields of view were selected and micronuclei frequency (%) as well as the localization of cGAS was measured (Fig. 5.6a). Again, I observed a dose-dependent increase in micronuclei frequency in irradiated parental 289 cells and similar frequencies of micronuclei in 289r cells (Fig. 5.6b). In those cells with micronuclei (5% - 10% of total), cGAS co-localized to micronuclei in roughly half of the cells. These values are similar to the values I observed for drug-insensitive primary B-ALL cells (Fig. 5.3b). Thus, micronuclei are frequently cGAS-positive in drug-treated and refractory B-ALL cells.      89   Figure 5.6 Micronuclei-positive 289r cells frequently show cGAS co-localization (a) 289 cells were fixed and stained with DAPI (nucleus) and cGAS following exposure to drugs, DMSO equivalent or gamma radiation. Arrowheads indicate micronuclei and cGAS localization. Scale bar=2µm.  (b) Percentage of cells with micronuclei in 289 control cells - 289r (DMSO) or sham-treated -, 289r cells, and irradiated 289 parental cells. In each bar, the inside gray bar indicates the percentage of cell with cGAS-positive micronuclei (cGAS-MN). (mean ± SEM, n= 3 experiments, >200 cells per bar, *P<0.05, **P<0.01, ***P<0.001, ANOVA)    90 5.2.6 289 resistant cell-lines show increased levels of g-H2AX foci Following treatment with chemotherapy, cancer cells will frequently possess an increase in markers of DNA damage, such as the presence of γ-H2AX foci. Moreover, micronuclei are known to be particularly unstable and prone to ongoing DNA damage. To query the level of DNA damage that may accompany the increased frequency of micronuclei in 289r cells, cells were fixed in poly-L-lysine coated 24-well plate and stained with γ-H2AX for DNA damage (Fig. 5.7a). Again, parental 289 cells exposed to graded levels of radiation were used as a positive control. Cells were continuously grown in medium containing the drug or DMSO and passaged 24 hours before fixation. In each experiment, 10 fields of view were selected and measured.  Compared to the control-treated cells, the level of γ-H2AX was significantly elevated in 289r cells and in parental 289 cells exposed to graded levels of radiation (Fig. 5.7b). Perhaps unsurprisingly, the levels of γ-H2AX foci were highest in the parental 289 cells that received the greatest amount of radiation damage (3 Gy) or the 289r cells grown in the presence of a PARP inhibitor, which directly impedes the repair of DNA damage. Thus, 289r cells show evidence of elevated genomic instability (micronuclei frequency) and genetic instability (γ-H2AX foci) that may trigger the transcription of pro-inflammatory signals.    91  Figure 5.7 Increased levels of g-H2AX foci are observed in 289r cells. (a) Representative images from high-content image (top panel) and confocal image (bottom panel) of irradiated 289 cells stained with DAPI and γ-H2AX (red). Arrowheads indicate micronuclei. Scale bar=5µm (High-content images), 2µm (confocal images).  (b) The average number of γ-H2AX foci per cell was measured with high-content image analysis in 289r (DMSO), 289r, and parental 289 cells following irradiation. In each experiment, 10 fields of view were analyzed. (mean ± SEM, n=3 experiments, >500 cells per bar, **P<0.01, ***P<0.001, ANOVA).   92 5.2.7 289 resistant cell-lines increase the expression of pro-inflammatory signals Drug-insensitive primary B-ALL cells and 289r cell-lines show evidence of genetic and genomic instability. Micronuclei are proposed to link chromosomal instability to activation of the innate immune system via cGAS-STING pathway 144,145. Thus, I measured the activation of immune responses in 289r cells through a TaqMan Mouse Immune Array to quantitate the expression of 92 mRNA associated with immune activation and 4 mRNA for housekeeping genes, including 18S rRNA. For this analysis, I performed duplicate experiments to compare expression levels in 289r (AZ82) cells with levels in 289r (DMSO) cells. For each experiment, I extracted total mRNA, using an AllPrep DNA/RNA mini kit, and measured RNA quantity and purity by spectrophotometric absorbance using a NanoDrop spectrophotometer. Following qRT-PCR analysis, 48 genes were found to be detected in each of four reactions. For these genes, the normalized cycle threshold (Ct) values were averaged and then compared between control 289r (DMSO) and 289r (AZ82) cells (Fig.5.8). I noted that gene expression was highly correlated in 289r (AZ82) and 289r (DMSO) cell lysates, however, 17 genes were outliers (95% confidence interval). Among these 17 genes, I selected the top 8 genes which have at least 10% higher or lower Ct values (approximately 5-fold change in gene expression) in 289r (AZ82) cell lysates than in control 289r (DMSO) cell lysates. These 8 genes are: STAT4 (-30%), CXCL10 (-12%), FAS (-11%), CCL3 (-10%), CD38 (33%), SOCS1 (11%), TNFRSF18 (15%), and H2-Eb1 (13%). Note: lower Ct values represent elevated expression in the 289r (AZ82) cell lysate.    93  Figure 5.8 Ct values for expression of 48 mouse immune genes between 289r (AZ82) cell lysates and 289r (DMSO) control cell lysates.  Replicate measurements of Ct values in distinct RNA extractions taken from 289r (AZ82) and control. Each dot represents a gene and dotted lines show the associated 95% confidence interval. Red or green color indicate the up- or down-regulated genes in 289r (AZ82) cells. Note a lower normalized Ct value (red) indicates elevated expression in 289r (AZ82) vs 289r (DMSO) cell lysates. (n=2 experiments, mean ± SEM).  Upregulated genes were either pro-inflammatory signals (CXCL10, CCL3), a transcription factor (STAT4) or a death receptor (FAS). Each of these gene products play an important role in the activation of apoptosis and immune responses. For example, STAT4 is involved in the IL-12 signaling pathway and regulates IFN-γ gene transcription 232. Both CXCL10 and CCL3 are cytokines, which are released from cells to regulate immune response. Finally, the FAS death receptor on the cell surface is able to induce programmed cell death. Moreover, among the downregulated genes, SOCS1 is a member of suppressor of cytokine signaling and the   94 overexpression of SOCS genes in cell-lines results in the inhibition of signaling by a wide range of cytokines 233. Overall, the gene expression altered in 289r (AZ82) cells may serve to activate the immune response, leading to further tumor elimination of the refractory B-ALL cells.  Interesting, I noted that CD38 and H2-Eb1 were downregulated in 289r (AZ82) cells. CD38 is a type II transmembrane glycoprotein that locates at cell surface and is a marker of various tumor cells, including leukemia. H2-Ea and H2-Eb1 encode H2-E which is a classical major histocompatibility complex (MHC) class II in mice. H2-Ea and H2-Eb locate at the surface of antigen presenting cells to present antigenic peptides to CD4+ T helper cells. Moreover, TNFRSF18 encodes the glucocorticoid-induced tumor necrosis factor receptor-related protein (GITR) and was also downregulated in 289r (AZ82) cells. GITR plays an important role in self-tolerance and effective T cell activation. These downregulated genes in mouse resistant B-ALL cells might associate with the disease progression, however, future evaluation of B-ALL cells is needed.  Overall, my study demonstrates that B-ALL cells that are refractory to centrosome declustering events show an increased level of genomic (micronuclei) and genetic (DNA damage) instability, which correlates with an increased production of pro-inflammatory signals, with the potential to activate the anti-tumor immune response.      95 5.3 Key findings 1. After 48 hours treatment with the dose nearest IC50, the measurements of nucleus phenotypes in primary cells revealed an increased frequency of micronuclei in surviving and, potentially, refractory primary B-ALL cells. 2. The frequency of induced micronuclei positively correlates with the frequency of centrosome amplification in the treated population of primary B-ALL cells. 3. In B-ALL cells treated with centrosome clustering inhibitors, approximately 50% of cells with micronuclei locate cGAS to the micronucleus. 4. 289 resistant cell-lines, 289r (AZ82) and 289r (PJ34), have growth advantages specific to the applied drug. 5. High-content image analysis reveals evidence of elevated DNA damage, as measured by γ-H2AX, and micronucleation in 289r and irradiated 289 cells. 6. The frequency of micronuclei is elevated in 289r cells to a level seen 48 hours after irradiation, which approximates the levels seen in drug-treated primary B-ALL cells. 7. Relative to the 289r (DMSO) cells, STAT4, CXCL10, CCL3, and FAS are overexpressed in 289r (AZ82) cells, indicating the potential to activate anti-tumor immune responses.  5.4 Discussion Recent studies have found that micronuclei are associated with extensive DNA damage 149 and nuclear envelope rupture 151, and can activate the cGAS-STING pathway for proinflammatory responses 145. Moreover, other studies reveal the elimination of solid tumors in mice via direct activation of cGAS-STING pathway 234. In addition, Coussens et al., (2013) show that proinflammatory signaling has the potential to cure tumors 235. To test the hypothesis that   96 centrosome clustering inhibitors will generate elevated genomic instability and pro-inflammatory signals in the population of surviving and refractory B-ALL cells, I measured the frequency of micronuclei in four normal primary stem cells derived from non-involved bone marrows and eleven primary B-ALL cells derived directly from patients, as well as in 289 cells resistant to centrosome clustering inhibitors. I found that exposure to each of the tested centrosome clustering inhibitors - AZ82 (KIF1 inhibitor), Olaparib and PJ34 (PARP inhibitor), and doxorubicin (standard-of-care drug) - increase the formation of micronuclei in primary B-ALL cells. Similarly, I observed a significantly increased frequency of micronuclei in 289r cells relative to 289r (DMSO) control cells, and the level of γ-H2AX was also significantly elevated in 289r cells. Taken together, these data reveal that refractory B-ALL cells show evidence of elevated genomic instability (micronuclei frequency) and genetic instability (γ-H2AX foci), which may trigger pro-inflammatory NF-kB signaling and its downstream proteins. Generally, micronuclei can lead to p53-mediated cell cycle arrest followed by apoptotic cell death 236,237. However, cancer cells are able to exclude micronuclei 238,239, resulting in the survival of cells after chemotherapy 240. Micronuclei are proposed to reincorporate their DNA into the main nucleus during replication 149,150, which contributes to genomic instability and cancer promotion 241. Moreover, chromosomes in micronuclei are under-replicated with accumulated DNA damage 149,151, and this under-replication of the chromosome in micronucleus results in a copy number asymmetry between daughter cells 150. Therefore, the low level of micronuclei (<10%) in resistant and refractory B-ALL cells may be due to the arrested G2/M phase and reincorporation or under-replication of chromosomes in micronucleus. However, the relatively early timepoint of my analysis (at 48 hours) may also indicate that the low level of micronuclei is reflective of the rate of occurrence of mis-segregated chromosomes.   97 I measured cGAS localization in micronuclei-positive drug-insensitive primary B-ALL cells and 289r cells. I found approximately 50% of resistant B-ALL cells with micronuclei have cGAS localized to micronuclei. An earlier study in mouse embryonic fibroblasts (MEFs) demonstrates that nuclear envelope rupture of micronuclei releases the inside DNA to the cytoplasmic compartment, which leads to cGAS localized to micronuclear chromatin 145. My findings reveal the potential activation of the cGAS-STING DNA sensing pathway by micronuclei induced in inhibitor-insensitive primary leukemia cells. So, I measured the production of pro-inflammatory signals in 289r (AZ82) and 289r (DMSO) cell-lines. TaqMan analysis of mouse immune response indicates significant upregulation of innate immunity regulators, including CXCL10, CCL3, STAT4, and FAS, in 289r (AZ82) cells. In addition, CD38, SOCS1, TNFRSF18, and H2-Eb1 were downregulated in 289r (AZ82) cells relative to 289r (DMSO).  Signal transducer and activator of transcription 4 (STAT4) is a transcription factor which is involved in IL-12 signaling pathway (Fig. 5.9a). IL-12 stimulate anti-tumor responses in various clinical and pre-clinical studies, making it an important candidate for cancer immunotherapy 242. STAT4 is required for the development of Th1 cells from naïve CD4+ T cells 243 and responsible for the production of IFN- γ 232.  CXCL10 and CCL3 are chemotactic cytokines that control the migration of immune cells from the blood to the site of inflammation and are critical for the function of the innate immune system (Fig. 5.9a). It is shown that CXCL10 mediates the recruitment of tumour-suppressive CXCR3+ T cells and natural killer (NK) cells into solid cancers 244,245. Correspondingly, a high intra-tumoral concentration of these chemokines is associated with a higher lymphocytic infiltrate and an improved survival in several malignancies 246–248.    98 FAS is responsible for multiple pro-apoptotic processes 249 and proposed to be involved in the pathogenesis of various malignancies 250 (Fig. 5.9b). FAS expression is often downregulated in human tumor cells to avoid FAS-mediated apoptosis signaling 251,252, especially in metastatic human colorectal cancer 253,254.  Cytokines bind to the receptors on cell surface and deliver information about the changes in external environment. The message from the cell surface is rapidly transported to the nucleus using different signaling cascades, including Janus kinase and signal transducer and activator of the transcription (JAK-STAT) pathway 255. One of its negative regulators that is known to contribute to cytokine inhibition is the protein family of suppressors of cytokine signaling (SOCS) (Fig. 5.9a). The overexpression of SOCS genes in cell-lines results in the inhibition of signaling by a wide range of cytokines 233.  Studies demonstrate that SOCS genes are constitutively expressed at relatively high levels in human breast cancer cells 256,257, which may confer the resistance to proinflammatory cytokines and trophic factors by shutting down STAT1/STAT5 signaling 257.  Therefore, the overexpression in refractory B-ALL cells of STAT4, CXCL10, CCL3, and FAS and the silencing of SOCS1 reveals a potential to activate immune system responses. NF-κB is responsible for the transcription of genes which encodes many pro-inflammatory signals, including CXCL10, CCL3, FAS, and it is involved in cGAS-STING pathway. Thus, the increased expression of pro-inflammatory signals in 289r (AZ82) cells might be regulated by cGAS-STING-NF-kB pathway.      99   Figure 5.9 Representative diagrams for signaling pathways of up- and down-regulated genes (a) Following binding of IL-12 molecules to IL-12Rβ1 and IL-12Rβ2, Jak2 and Tyk2 get transphosphorylated. Phosphorylated IL-12Rβ2 binds to STAT4 which then dimerizes with another STAT4 molecule. These activated STAT4 enter nucleus and initiate the transcription of genes. SOCS1 inhibits transcription by blocking JAK2 phosphorylation. Produced signals are released from cells to modulate immune responses.  (b) The binding of FAS ligand (FASL) to FAS leads to the recruitment of FADD, which binds to inactivated caspase 8 (pro-caspase 8) through DED domain, causing the formation of death-inducing signaling complex (DISC). Then caspase 8 is activated and initiates downstream apoptosis directly by activating caspase 3.     100 Chapter 6: Discussion and conclusions  6.1 Summary Disease relapse and the long-term side effects from current treatments remain significant clinical challenges in pediatric cancer. Hence, the goals of my study were to explore new therapies that target pediatric ALL cells with reduced effects on normal stem cells and potential to induce immune responses that can enhance immunogenicity and maintain remission in pediatric B-ALL. My study demonstrates that both primary and immortal aneuploid pediatric B-ALL cells frequently exhibit centrosome amplification and are sensitive to centrosome clustering inhibitors. These findings may highlight the importance of centrosome amplification in the diagnosis of pediatric B-ALL and reveal a novel specific treatment in pediatric leukemia. In addition, B-ALL cells resistant to centrosome clustering inhibitors have increased genomic and genetic instability, and potentially activate innate immune responses through the cGAS-dsDNA pathway (Fig. 6.1). These findings suggest the ability of centrosome clustering inhibitors to generate long-term protection against pediatric leukemia by inducing immune responses to refractory B-ALL cells. Overall, my study implicates the advantages of using centrosome clustering inhibitors on the treatment of pediatric ALL and suggests that broadening the immune responses may enhance the clinical outcomes for ALL patients.    101  Figure 6.1 Working models (a) During cell division, B-ALL cells cluster extra centrosomes together to generate pseudo-bipolar spindle cell division and generate two stable daughter cells.  (b) Inhibition of centrosome clustering pathways triggers multipolar spindle formation and elicits programmed cell death. But some B-ALL cells survive from the declustering events and gain increased genomic instability.  (c) Micronucleation is often caused by chromosome mis-segregation, (1) which can be induced by radiation and centrosome clustering inhibitors. (2) Chromothripsis generated by micronucleus that cannot undergo resorption indicates the presence of tumor-specific neoantigen. Micronuclei undergo nuclear membrane rupture during interphase, which exposes the inside DNA to the cytoplasm cGAS. Activated cGAS sends signals to scaffold protein STING, (3) which phosphorylates NF-kB and IRF3. Phosphorylated NF-kB and IRF3 enter the nucleus and modulate the transcription of pro-inflammatory signals for innate immune response.   102 6.2 Centrosome clustering inhibitors as inducers of protective anti-ALL immunity Non-self antigens from tumors are proposed to improve tumor immunogenicity and activate anti-tumor immunity 258. TLR9, which is involved in innate immunity, recognizes the non-methylated CG motifs from nucleotide sequences that occur in bacterial and viral DNA, but rarely in mammalian DNA 259,260. Currently, a TLR9 agonist, CpG ODNs (oligonucleotides), has been investigated in various solid and hematologic tumors 261. In these studies, it is demonstrated that TLR9 agonists have anti-tumor activity in patients with refractory tumors 262,263. These observations suggest that the expression of neoantigens by B-ALL cells may lead to the generation of protective immune activity.  While the ability of genomic instability to induce an innate immune response by activating the cGAS-STING pathway has been demonstrated in solid tumors 144,145, it has yet to be revealed in pediatric B-ALL cells treated with targeted therapies. The results presented in Chapter 5 show the link between micronuclei and cGAS localization, and reveal the elevated production of pro-inflammatory cytokines. Moreover, chromosomes in micronuclei are subjected to extensive mutagenesis 150, which may induce the presence of tumor-specific neoantigen in cancer cells. The results in Chapter 5, therefore, suggest the adaptive immune response may also be activated in addition to the innate immune response.  According to these studies, the modulation of the immune environment may enhance innate and adaptive immune responses to pediatric B-ALL. Therefore, neoantigen formation driven by declustering agents, combined with immune modulation by the accompanying cGAS-dsDNA pathway activation or adjuvant TLR9-CpG ODN/antibody treatment, may lead to a broader anti-ALL immune response, resulting in long-term protection from ALL outgrowth.    103 6.3 Combination of centrosome clustering inhibitors and immune checkpoint blockade Cancer immunotherapies, such as CAR-T, bispecific antibody and immune checkpoint inhibitor, enhance cancer-specific immune reactivity and eliminate cancer cells in patients 264–266. Tumor cells often dysregulate immune checkpoints to survive and escape from immune surveillance 267. Cytotoxic T lymphocyte antigen-4 (CTLA-4) and programmed cell death-1 (PD-1) are the two immune checkpoints that have been extensively studied. Blockade of CTLA4 achieves long-term survival in ~20% of patients with metastatic melanoma, and the inhibition of PD-1/PD-L1 results in a response rate of ~30% in several solid tumors 268. Nonetheless, only subsets of patients respond to the blockade. Recent studies demonstrate that somatic nonsynonymous mutations positively correlate with the clinical efficacy of PD-1 blockade pembrolizumab 269,270. The level of mutations or neoantigens, therefore, might be a strong indicator for the success of immunotherapy.  Several features of pediatric B-ALL make it a challenging target for immunotherapy. These include the low mutational burden 27; the occurrence of driver mutations in utero 271; and the lack of costimulatory molecule expression on leukemia cells which contributes to T cell anergy induction 272. Therefore, strategies that alter the endogenous immune environment and enable a broadening of the anti-ALL immune response are needed. Although small cell lung cancer is often resistant to immune checkpoint blockades, PARP inhibition is found to elevate PD-L1 expression and potentiate the anti-tumor effects of the PD-L1 blockade 273. This indicates the necessity of mutations and neoantigens in immunotherapy. Therefore, pediatric leukemia with low mutational loads may not have significant endogenous immune responses when treated with immune checkpoint blockade as a single agent. The combination of centrosome clustering inhibitors and   104 immune checkpoint blockades could potentially induce a strong immune response to prevent cancer proliferation and progression. Sarcomas are a heterogeneous group of cancers that arise from bone or soft tissues. Pediatric sarcomas account for approximately 10% of pediatric cancer 274, and the majority of cases are rhabdomyosarcoma and osteosarcoma. Similar to pediatric B-ALL, relapse and long-term side effects are major clinical problems that encourage the development of novel and less toxic treatments. According to the overall success of immune checkpoint blockades for the treatment of metastatic melanoma and non-small cell lung cancer, the potential of CTLA-4 and PD-1 blockade is being explored in pediatric sarcomas. Anti-CTLA-4 antibodies promote tumor rejection in murine sarcoma models 275, and expression of PD-L1 is positively correlated with tumor-infiltrating lymphocytes (TILs) in osteosarcoma 276. These studies reveal the potential of immune checkpoint blockades in pediatric sarcomas. Because of the low mutational rates and MHC expression in childhood sarcomas 277, immune checkpoint blockades, as single agents, may be less effective against pediatric sarcomas. As centrosome amplification has been detected in sarcomas 278,279, the combination of centrosome clustering inhibitors and immune checkpoint blockades may further improve the prognosis in both diagnosis and relapsed pediatric sarcomas.  6.4 Use of centrosome clustering inhibitors as a conditioning treatment for CAR-T therapy CAR-T cells are T cells that originated from patients and have been genetically engineered to specifically target antigens present in patients. CAR-T cells have achieved complete remission rates approaching 85% in even heavily pre-treated refractory and relapsed pediatric ALL patients 280. The U.S. Food and Drug Administration (FDA) has approved the adoptive transfer of autologous CD19-targeted CAR-T cells for patients with relapsed or heavily pre-treated leukemia,   105 including B-ALL 281. Last year, the first CAR-T therapy was approved in Canada 282. Now, eligible patients are able to access Kymriah (tisagenlecleucel), a CD19-directed modified autologous T cell therapy, from hospitals in Toronto and British Columbia. Currently, the price for Kymriah is $475,000 USD for one-time treatment in the US, and the true cost of CAR-T therapy for patients in Canada is not clear. Moreover, the development of systems for administration, infrastructure, and post-infusion care for successful delivery of CAR-T therapy remains a challenge 283. Despite these challenges and unknowns, CAR-T therapy is approved in Canada, and patients with hard-to-treat leukemias can be treated with this therapy and may have better responses. Nonetheless, disease progression occurs within one year in 50% of cases treated with CAR-T therapy 280. This disease progression may be due to the loss of engineered receptors, which results in antigen positive relapse. Therefore, strategies to improve the outcome for CAR-T therapy are required. The combination of CD19-targeted CAR-T cells and pembrolizumab, a PD-1 inhibitor, leads to better prognosis in patient with refractory diffuse large B-cell lymphoma 284. This indicates that the enhanced immunogenicity through PD-1 blockade may improve the efficacy of CAR-T cells. Moreover, the alteration of tumor antigens in murine B-ALL cells can lead to prolonged survival in mouse models 285. This study indicates that the ability to drive the generation of additional neoantigens in ALL cells may provide a means to diversify immune responses and influence the effectiveness of targeted immunotherapy. Thus, the application of centrosome clustering inhibitors in pediatric B-ALL cells may enhance the T cell response diversification and consolidate the CAR-T-induced remission as a conditioning treatment.     106 6.5 Suggested future studies Chapter 5 reveals that cGAS colocalizes with micronuclei in 289r cells, which suggests that the cytosolic DNA sensing pathway is activated. Similar results were also observed in one primary B-ALL cell sample. It will be important to further analyze the cGAS-micronuclei colocalization in PDXs or primary B-ALL cells treated with centrosome clustering inhibitors. Immunofluorescence identified the localization of cGAS to micronuclei while TaqMan analysis indicates significant upregulation of innate immunity regulators in immortal 289r cells. These results indicate that micronucleation is a common response in cells refractory to centrosome clustering inhibitors and may then augment immunogenicity. However, the significant association between cGAS and pro-inflammatory signals is not yet investigated. It would be interesting to create cGAS knock-out cells to determine the necessity of cGAS for the expression of pro-inflammatory signals in refractory B-ALL cells. Lastly, to evaluate centrosome clustering inhibitors as inducers of protective anti-ALL immunity, adoptive transfer of parental or modified 289 B-ALL cells into syngenetic mice (wt BALB/c or RAG1-/- mice) is needed to measure and manipulate adaptive or innate immune responses. It would be interesting to see whether immunomodulation will enhance the responses to neoantigens generated by declustering agents. As such, the impact of CpG ODN and/or immune checkpoint blockades on leukemia-specific T cell responses and survival in Eµ-ret mice challenged with 289r cells can be evaluated in the future.         107 Bibliography 1. Canadian Cancer Statistics 2008. Cancer 1–108 (2008). 2. Harras, A., Edwards, B.K., Blot, W.J., et al., E. Cancer Rates and Risks. The National Cancer Institute (1996). Available at: https://www.scirp.org/reference/ReferencesPapers.aspx?ReferenceID=2237785. (Accessed: 20th October 2019) 3. Ries LAG, Harkins D, Krapcho M, Mariotto A, Miller BA, Feuer EJ, Clegg L, Eisner MP, Horner MJ, Howlader N, Hayat M, Hankey BF, E. B. (eds). SEER Cancer Statistics Review 1975-2003. National Cancer Institute. Bethesda, MD (2003). Available at: https://seer.cancer.gov/archive/csr/1975_2003/. (Accessed: 20th October 2019) 4. Smith, M. & Hare, M. L. An overview of progress in childhood cancer survival. J. Pediatr. Oncol. Nurs. 21, 160–4 (2004). 5. Sklar, C. A. Overview of the effects of cancer therapies: the nature, scale and breadth of the problem. Acta Paediatr. Suppl. 88, 1–4 (1999). 6. Blatt, J., Copeland, D. & Bleyer, W. Late effects of childhood cancer and its treatment. In: Pizzo PA and Poplack DG (eds) Principles and practices of pediatric oncology. Philadelphia, Lippincott-Raven 1303–29 (1997). Available at: https://shop.lww.com/Principles-and-Practice-of-Pediatric-Oncology/p/9781451194234. (Accessed: 22nd October 2019) 7. Stary, J. et al. Intensive chemotherapy for childhood acute lymphoblastic leukemia: Results of the randomized intercontinental trial ALL IC-BFM 2002. J. Clin. Oncol. 32, 174–184 (2014). 8. Mertens, A. C. et al. Cause-specific late mortality among 5-year survivors of childhood   108 cancer: The childhood cancer survivor study. J. Natl. Cancer Inst. 100, 1368–1379 (2008). 9. Brierley, J. D., Rathmell, A. J., Gospodarowicz, M. K., Sutcliffe, S. B. & Pintillie, M. Late relapse after treatment for clinical stage I and II Hodgkin’s disease. Cancer 79, 1422–7 (1997). 10. Rivera, G. K. et al. Effectiveness of intensified rotational combination chemotherapy for late hematologic relapse of childhood acute lymphoblastic leukemia. Blood 88, 831–7 (1996). 11. Sadowitz PD, Smith SD, Shuster J, Wharam MD, Buchanan GR, R. G. Treatment of late bone marrow relapse in children with acute lymphoblastic leukemia: a Pediatric Oncology Group study. - PubMed - NCBI. Blood 81(3):602–9 (1993). Available at: https://www.ncbi.nlm.nih.gov/pubmed/8427957. (Accessed: 20th October 2019) 12. Escherich, G., Horstmann, M. A., Zimmermann, M., Janka-Schaub, G. E. & COALL study group. Cooperative study group for childhood acute lymphoblastic leukaemia (COALL): long-term results of trials 82,85,89,92 and 97. Leukemia 24, 298–308 (2010). 13. Möricke, A. et al. Long-term results of five consecutive trials in childhood acute lymphoblastic leukemia performed by the ALL-BFM study group from 1981 to 2000. Leukemia 24, 265–284 (2010). 14. Tallen, G. et al. Long-term outcome in children with relapsed acute lymphoblastic leukemia after time-point and site-of-relapse stratification and intensified short-course multidrug chemotherapy: Results of trial ALL-REZ BFM 90. J. Clin. Oncol. 28, 2339–2347 (2010). 15. Eckert, C. et al. Minimal residual disease after induction is the strongest predictor of prognosis in intermediate risk relapsed acute lymphoblastic leukaemia - long-term results   109 of trial ALL-REZ BFM P95/96. Eur. J. Cancer 49, 1346–55 (2013). 16. Devita, V. T., Young, R. C. & Canellos, G. P. Combination versus single agent chemotherapy: A review of the basis for selection of drug treatment of cancer. Cancer 35, 98–110 (1975). 17. Stratton, M. R., Campbell, P. J. & Futreal, P. A. The cancer genome. Nature 458, 719–724 (2009). 18. Holyoake, T. L. & Vetrie, D. The chronic myeloid leukemia stem cell: Stemming the tide of persistence. Blood 129, 1595–1606 (2017). 19. Risques, R. A. & Kennedy, S. R. Aging and the rise of somatic cancer-associated mutations in normal tissues. PLoS Genetics 14, (2018). 20. Wang, Q. Cancer predisposition genes: Molecular mechanisms and clinical impact on personalized cancer care: Examples of Lynch and HBOC syndromes. Acta Pharmacologica Sinica 37, 143–149 (2016). 21. Vogelstein, B. et al. Cancer genome landscapes. Science 340, 1546–1558 (2013). 22. Lawrence, M. S. et al. Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature 499, 214–218 (2013). 23. Alejandro Sweet-Cordero, E. & Biegel, J. A. The genomic landscape of pediatric cancers: Implications for diagnosis and treatment. Science (80-. ). 363, 1170–1175 (2019). 24. Gröbner, S. N. et al. The landscape of genomic alterations across childhood cancers. Nature 555, 321–327 (2018). 25. Mullighan, C. G. et al. Genomic analysis of the clonal origins of relapsed acute lymphoblastic leukemia. Science (80-. ). 322, 1377–1380 (2008). 26. Mullighan, C. G. et al. CREBBP mutations in relapsed acute lymphoblastic leukaemia.   110 Supplementary Information. Nature 471, 235–9 (2011). 27. Ma, X. et al. Rise and fall of subclones from diagnosis to relapse in pediatric B-acute lymphoblastic leukaemia. Nat. Commun. 6, (2015). 28. Morrissy, A. S. et al. Divergent clonal selection dominates medulloblastoma at recurrence. Nature 529, 351–7 (2016). 29. Schultz, K. R. et al. Improved early event-free survival with imatinib in Philadelphia chromosome-positive acute lymphoblastic leukemia: a children’s oncology group study. J. Clin. Oncol. 27, 5175–81 (2009). 30. Harvey, R. C. et al. Rearrangement of CRLF2 is associated with mutation of JAK kinases, alteration of IKZF1, Hispanic/Latino ethnicity, and a poor outcome in pediatric B-progenitor acute lymphoblastic leukemia. Blood 115, 5312–21 (2010). 31. Ward, E., Desantis, C., Robbins, A., Kohler, B. & Jemal, A. Childhood and Adolescent Cancer Statistics , 2014. 64, 83–103 (2014). 32. Kim, S. B. et al. Sex differences in prevalence and risk factors of asymptomatic cholelithiasis in Korean health screening examinee: A retrospective analysis of a multicenter study. Med. (United States) 96, (2017). 33. Smith, M. A. et al. Outcomes for children and adolescents with cancer: Challenges for the twenty-first century. J. Clin. Oncol. 28, 2625–2634 (2010). 34. Gale, K. B. et al. Backtracking leukemia to birth: Identification of clonotypic gene fusion sequences in neonatal blood spots. Proc. Natl. Acad. Sci. 94, 13950–13954 (1997). 35. Taub, J. W. et al. High frequency of leukemic clones in newborn screening blood samples of children with B-precursor acute lymphoblastic leukemia. 99, 2992–2997 (2018). 36. Foulds, L. Neoplastic development. Br. J. Surg. 56, 800–800 (1969).   111 37. Kong, Y. et al. CD34+CD38+CD19+ as well as CD34+CD38-CD19+ cells are leukemia- initiating cells with self-renewal capacity in human B-precursor ALL. 22, 1207–1213 (2008). 38. Greaves, M. F., Wiemels, J. & Laboratories, C. B. ORIGINS OF CHROMOSOME TRANSLOCATIONS IN CHILDHOOD. 3, 1–11 (2003). 39. Maia, A. T. et al. Identification of Preleukemic Precursors of Hyperdiploid Acute Lymphoblastic Leukemia in Cord Blood. 43, 38–43 (2004). 40. Bateman, C. M. et al. Evolutionary trajectories of hyperdiploid ALL in monozygotic twins. 29, 58–65 (2014). 41. Talamo, A., Chalandon, Y., Marazzi, A. & Jotterand, M. Clonal heterogeneity and chromosomal instability at disease presentation in high hyperdiploid acute lymphoblastic leukemia. Cancer Genet. Cytogenet. 203, 209–214 (2010). 42. Alpar, D. et al. Sequential and Hierarchical Chromosomal Changes and Chromosome Instability are Distinct Features of High Hyperdiploid Pediatric Acute Lymphoblastic Leukemia. Pediatr Blood Cancer 61, 2208–2214 (2014). 43. Bateman, C. M. et al. Acquisition of genome-wide copy number alterations in monozygotic twins with acute lymphoblastic leukemia. 115, 3553–3559 (2018). 44. Greaves, M. & Maley, C. C. CLONAL EVOLUTION IN CANCER. Nature 481, 306–313 (2012). 45. Mullighan, C. G. et al. Deletion of IKZF1 and prognosis in acute lymphoblastic leukemia. N. Engl. J. Med. 360, 470–80 (2009). 46. Den Boer, M. L. et al. A subtype of childhood acute lymphoblastic leukaemia with poor treatment outcome: a genome-wide classification study. Lancet. Oncol. 10, 125–34   112 (2009). 47. Harvey, R. C. et al. Identification of novel cluster groups in pediatric high-risk B-precursor acute lymphoblastic leukemia with gene expression profiling: correlation with genome-wide DNA copy number alterations, clinical characteristics, and outcome. Blood 116, 4874–84 (2010). 48. Roberts, K. G. et al. Genetic Alterations Activating Kinase and Cytokine Receptor Signaling in High-Risk Acute Lymphoblastic Leukemia. Cancer Cell 22, 153–166 (2012). 49. Paulsson, K. & Johansson, B. High Hyperdiploid Childhood Acute Lymphoblastic Leukemia. 660, 637–660 (2009). 50. Paulsson, K. et al. Genetic landscape of high hyperdiploid childhood acute lymphoblastic leukemia. 2–7 (2010). doi:10.1073/pnas.1006981107 51. Chromosomal Abnormalities in Acute Lymphoblastic Leukemia : Structural and Numerical Changes in 234 Cases. Cancer Genet. Cytogenet. 4, 101–110 (1981). 52. Moorman, A. V. et al. Outcome heterogeneity in childhood high-hyperdiploid acute lymphoblastic leukemia. Blood 102, 2756–2762 (2003). 53. Chessells, J. M., Richards, S. M., Bailey, C. C., Lilleyman, J. S. & Eden, O. B. Gender and treatment outcome in childhood lymphoblastic leukaemia: report from the MRC UKALL trials. Br. J. Haematol. 89, 364–72 (1995). 54. Schultz, K. R. et al. Risk- and response-based classification of childhood B-precursor acute lymphoblastic leukemia: a combined analysis of prognostic markers from the Pediatric Oncology Group (POG) and Children’s Cancer Group (CCG). Blood 109, 926–935 (2007). 55. Jackson, J. F. et al. Favorable prognosis associated with hyperdiploidy in children with   113 acute lymphocytic leukemia correlates with extra chromosome 6. A Pediatric Oncology Group study. Cancer 66, 1183–9 (1990). 56. Harris MB, Shuster JJ, Carroll A, Look AT, Borowitz MJ, Crist WM, Nitschke R, Pullen J, Steuber CP, L. V. Trisomy of leukemic cell chromosomes 4 and 10 identifies children with B-progenitor cell acute lymphoblastic leukemia with a very low risk of treat... - PubMed - NCBI. Blood (1992). Available at: https://www.ncbi.nlm.nih.gov/pubmed/1596572. (Accessed: 20th October 2019) 57. Heerema, N. A. et al. Prognostic impact of trisomies of chromosomes 10, 17, and 5 among children with acute lymphoblastic leukemia and high hyperdiploidy (> 50 chromosomes). J. Clin. Oncol. 18, 1876–87 (2000). 58. Heerema, B. N. A. et al. Hypodiploidy With Less Than 45 Chromosomes Confers Adverse Risk in Childhood Acute Lymphoblastic Leukemia: A Report From the Children’s Cancer Group. 94, 4036–4046 (2018). 59. Harrison, C. J. et al. Three distinct subgroups of hypodiploidy in acute lymphoblastic leukaemia. 552–559 (2004). doi:10.1111/j.1365-2141.2004.04948.x 60. Moorman, A. V. The clinical relevance of chromosomal and genomic abnormalities in B-cell precursor acute lymphoblastic leukaemia. Blood Rev. 26, 123–35 (2012). 61. Holmfeldt, L. et al. THE GENOMIC LANDSCAPE OF HYPODIPLOID ACUTE LYMPHOBLASTIC LEUKEMIAfile:///Users/meiyung/Downloads/1-s2.0-0092867486903181-main.pdf file:///Users/meiyung/Downloads/1-s2.0-S0304419X08000267-main.pdf file:///Users/meiyung/Downloads/1-s2.0-S0092867400808752-ma. Nat Genet 45, 242–252 (2013). 62. Nachman, J. B. et al. Outcome of treatment in children with hypodiploid acute   114 lymphoblastic leukemia. 110, 1112–1116 (2018). 63. Moorman, A. V. New and emerging prognostic and predictive genetic biomarkers in B-cell precursor acute lymphoblastic leukemia. Haematologica 101, 407–416 (2016). 64. Moorman, A. V. et al. Apopulation-based cytogenetic study of adults with acute lymphoblastic leukemia. Blood 115, 206–214 (2010). 65. Mullighan, C. G. et al. Outcome of children with hypodiploid ALL treated with risk-directed therapy based on MRD levels. Blood 126, 2896–2899 (2015). 66. Griffiths, A. J. et al. An Introduction to Genetic Analysis. (New York: W. H. Freeman, 2000). 67. Michel, L. S. et al. premature anaphase and chromosome instability in mammalian cells. 355–359 (2001). 68. Dobles, M., Liberal, V., Scott, M. L., Benezra, R. & Sorger, P. K. Chromosome Missegregation and Apoptosis in Mice Lacking the Mitotic Checkpoint Protein Mad2. 101, 635–645 (2000). 69. Kalitsis, P., Earle, E., Fowler, K. J. & Choo, K. H. A. Bub3 gene disruption in mice reveals essential mitotic spindle checkpoint function during early embryogenesis. 2277–2282 (2000). doi:10.1101/gad.827500.tubule-depolymerising 70. Rieder, C. L., Cole, R. W., Khodjakov, A. & Sluder, G. The Checkpoint Delaying Anaphase in Response to Chromosome Monoorientation Is Mediated by an Inhibitory Signal Produced by Unattached Kinetochores. Cell Biol. 130, 941–948 (1995). 71. Weaver, B. A. A. & Cleveland, D. W. Does aneuploidy cause cancer ? 658–667 (2006). doi:10.1016/j.ceb.2006.10.002 72. Brinkley, B. R. Managing the centrosome numbers game : from chaos to stability in   115 cancer cell division. 11, 18–21 (2001). 73. Rudolph, K. L., Millard, M., Bosenberg, M. W. & DePinho, R. A. Telomere dysfunction and evolution of intestinal carcinoma in mice and humans. Nat. Genet. 28, 155–9 (2001). 74. Meeker, A. K. et al. Telomere shortening is an early somatic DNA alteration in human prostate tumorigenesis. Cancer Res. 62, 6405–6409 (2002). 75. Chin, K. et al. In situ analyses of genome instability in breast cancer. Nat. Genet. 36, 984–8 (2004). 76. Sansregret, L. & Swanton, C. The role of aneuploidy in cancer evolution. Cold Spring Harbor Perspectives in Medicine 7, a028373 (2017). 77. Davoli, T. & de Lange, T. Telomere-driven tetraploidization occurs in human cells undergoing crisis and promotes transformation of mouse cells. Cancer Cell 21, 765–76 (2012). 78. Roschke, A. V. et al. Karyotypic Complexity of the NCI-60 Drug-Screening Panel. Cancer Res. 63, 8634–8647 (2003). 79. Navin, N. et al. Tumour evolution inferred by single-cell sequencing. Nature 472, 90–95 (2011). 80. Potapova, T. A., Zhu, J. & Li, R. Aneuploidy and chromosomal instability: A vicious cycle driving cellular evolution and cancer genome chaos. Cancer Metastasis Rev. 32, 377–389 (2013). 81. Pavelka, N. et al. Aneuploidy confers quantitative proteome changes and phenotypic variation in budding yeast. Nature 468, 321–5 (2010). 82. Rancati, G. et al. Aneuploidy underlies rapid adaptive evolution of yeast cells deprived of a conserved cytokinesis motor. Cell 135, 879–93 (2008).   116 83. Breems, D. A. et al. Monosomal karyotype in acute myeloid leukemia: A better indicator of poor prognosis than a complex karyotype. J. Clin. Oncol. 26, 4791–4797 (2008). 84. Grimwade, D. et al. Refinement of cytogenetic classification in acute myeloid leukemia: Determination of prognostic significance of rare recurring chromosomal abnormalities among 5876 younger adult patients treated in the United Kingdom Medical Research Council trials. Blood 116, 354–365 (2010). 85. Hasle, H. et al. Myelodysplastic syndrome, juvenile myelomonocytic leukemia, and acute myeloid leukemia associated with complete or partial monosomy 7. European Working Group on MDS in Childhood (EWOG-MDS). Leuk.  Off. J. Leuk. Soc. Am. Leuk. Res. Fund, U.K 376–385 (1999). 86. Parkin, B. et al. Clonal evolution and devolution after chemotherapy in adult acute myelogenous leukemia. Blood 121, 369–377 (2013). 87. Kifschner, M. & Mitchison, T. Beyond Self-Assembly : From Microtubules to Morphogenesis. 45, 329–342 (1986). 88. Kellogg, D. R. Organizing cytoplasmic events. Nature 340, 99 (1989). 89. Khodjakov, A. & Rieder, C. L. Centrosomes Enhance the Fidelity of Cytokinesis in Vertebrates and Are Required for Cell Cycle Progression 7. 153, 237–242 (2001). 90. Murphy, K. L. & Rosen, J. M. Mutant p53 and genomic instability in a transgenic mouse model of breast cancer. (2000). 91. Assoro, A. B. D., Lingle, W. L. & Salisbury, J. L. Centrosome amplification and the development of cancer. 6146–6153 (2002). doi:10.1038/sj.onc.1205772 92. Godinho, S. A. & Pellman, D. Causes and consequences of centrosome abnormalities in cancer. (2014).   117 93. Denu, R. A. et al. Centriole Overduplication is the Predominant Mechanism Leading to Centrosome Ampli fi cation in Melanoma. 517–528 (2018). doi:10.1158/1541-7786.MCR-17-0197 94. Godinho, S. A. et al. Oncogene-like induction of cellular invasion from centrosome amplification. Nature 510, 167–71 (2014). 95. Shinmura, K. et al. SASS6 overexpression is associated with mitotic chromosomal abnormalities and a poor prognosis in patients with colorectal cancer. Oncol. Rep. 34, 727–38 (2015). 96. Tang, C.-J. C. et al. The human microcephaly protein STIL interacts with CPAP and is required for procentriole formation. EMBO J. 30, 4790–804 (2011). 97. Loncarek, J., Hergert, P., Magidson, V. & Khodjakov, A. Control of daughter centriole formation by the pericentriolar material. Nat. Cell Biol. 10, 322–8 (2008). 98. Korzeniewski, N., Hohenfellner, M. & Duensing, S. CAND1 promotes PLK4-mediated centriole overduplication and is frequently disrupted in prostate cancer. Neoplasia (United States) 14, 799–806 (2012). 99. Marina, M. & Saavedra, H. I. Nek2 and Plk4: prognostic markers, drivers of breast tumorigenesis and drug resistance. Front. Biosci. (Landmark Ed. 19, 352–65 (2014). 100. Mason, J. M. et al. Functional characterization of CFI-400945, a Polo-like kinase 4 inhibitor, as a potential anticancer agent. Cancer Cell 26, 163–76 (2014). 101. Sredni, S. T. & Tomita, T. The polo-like kinase 4 gene (PLK4) is overexpressed in pediatric medulloblastoma. Childs. Nerv. Syst. 33, 1031 (2017). 102. Li, Z. et al. Expression of Polo-Like Kinase 4(PLK4) in Breast Cancer and Its Response to Taxane-Based Neoadjuvant Chemotherapy. J. Cancer 7, 1125–32 (2016).   118 103. Macmillan, J. C., Hudson, J. W., Bull, S., Dennis, J. W. & Swallow, C. J. Comparative expression of the mitotic regulators SAK and PLK in colorectal cancer. Ann. Surg. Oncol. 8, 729–740 (2001). 104. Castedo, M. & Kroemer, G. Illicit survival of cancer cells during polyploidization and depolyploidization. 1403–1413 (2011). doi:10.1038/cdd.2010.145 105. Kim, J. H. et al. Mechanical Tension Drives Cell Membrane Fusion. Dev. Cell 32, 561–573 (2015). 106. Ring, D., Hubble, R. & Kirschner, M. Mitosis in a cell with multiple centrioles. J. Cell Biol. 94, 549–56 (1982). 107. Boveri, T. Concerning the Origin of Malignant Tumours by Theodor Boveri . Translated and annotated by Henry Harris. Cell Sci. 121, 1–84 (2008). 108. Li, J. J. et al. centrosome amplification , chromosomal instability , and breast cancer in female ACI rats. 101, 18123–18128 (2004). 109. Basto, R. et al. Centrosome Amplification Can Initiate Tumorigenesis in Flies. Cell 133, 1032–1042 (2008). 110. Lingle, W. L. et al. Centrosome amplification drives chromosomal instability in breast tumor development. (2001). 111. Marthiens, V. et al. Centrosome amplification causes microcephaly. Nat. Cell Biol. 15, 731–740 (2013). 112. Maxwell, C. A., Keats, J. J., Belch, A. R., Pilarski, L. M. & Reiman, T. Receptor for Hyaluronan-Mediated Motility Correlates with Centrosome Abnormalities in Multiple Myeloma and Maintains Mitotic Integrity. 850–861 (2005). 113. Levine, M. S. et al. Centrosome Amplification Is Sufficient to Promote Spontaneous   119 Tumorigenesis in Mammals. Dev. Cell 40, 313-322.e5 (2017). 114. Ring, D., Hubble, R. & Kirschner, M. Mitosis in a Cell with Multiple Centrioles. 0–7 (1982). 115. Marthiens, V., Piel, M. & Basto, R. Never tear us apart – the importance of centrosome clustering. Cell Sci. 125, 3281–3293 (2012). 116. Chavali, P. L. et al. A CEP215-HSET complex links centrosomes with spindle poles and drives centrosome clustering in cancer. Nat. Commun. 7, (2016). 117. She, Z.-Y. & Yang, W.-X. Molecular mechanisms of kinesin-14 motors in spindle assembly and chromosome segregation. J. Cell Sci. 130, 2097–2110 (2017). 118. Kleylein-Sohn, J. et al. Acentrosomal spindle organization renders cancer cells dependent on the kinesin HSET. J. Cell Sci. 125, 5391–5402 (2012). 119. Lüders, J., Patel, U. K. & Stearns, T. GCP-WD is a γ-tubulin targeting factor required for centrosomal and chromatin-mediated microtubule nucleation. Nat. Cell Biol. 8, 137–147 (2006). 120. Wiese, C. & Zheng, Y. Microtubule nucleation: γ-tubulin and beyond. J. Cell Sci. 119, 4143–4153 (2006). 121. Archambault, V. & Glover, D. M. Polo-like kinases: Conservation and divergence in their functions and regulation. Nature Reviews Molecular Cell Biology 10, 265–275 (2009). 122. Dai, W. Polo-like kinases, an introduction. Oncogene 24, 214–216 (2005). 123. Johmura, Y. et al. Regulation of microtubule-based microtubule nucleation by mammalian polo-like kinase 1. Proc. Natl. Acad. Sci. U. S. A. 108, 11446–11451 (2011). 124. Zhang, X. et al. Sequential phosphorylation of Nedd1 by Cdk1 and Plk1 is required for targeting of the γTuRC to the centrosome. J. Cell Sci. 122, 2240–2251 (2009).   120 125. Zeng, C. NuMA: A nuclear protein involved in mitotic centrosome function. Microsc. Res. Tech. 49, 467–477 (2000). 126. Compton, D. A., Szilak, I. & Cleveland, D. W. Primary structure of NuMA, an intranuclear protein that defines a novel pathway for segregation of proteins at mitosis. J. Cell Biol. 116, 1395–1408 (1992). 127. Hueschen, C. L., Kenny, S. J., Xu, K. & Dumont, S. NuMA recruits dynein activity to microtubule minus-ends at mitosis. (2017). doi:10.7554/eLife.29328.001 128. Rhys, A. D. et al. Loss of E-cadherin provides tolerance to centrosome amplification in epithelial cancer cells. 1–15 (2018). 129. Nigg, E. A. CENTROSOME ABERRATIONS : CAUSE OR CONSEQUENCE OF CANCER. 2, (2002). 130. Grimaldi, A. M. et al. MEK Inhibitors in the Treatment of Metastatic Melanoma and Solid Tumors. Am. J. Clin. Dermatol. 18, 745–754 (2017). 131. Li, G., Dai, W. & Shao, F. Effect of ALK-inhibitors in the treatment of non-small cell lung cancer : a systematic review and meta-analysis. 3496–3503 (2017). 132. Ahronian, L. G. & Corcoran, R. B. Strategies for monitoring and combating resistance to combination kinase inhibitors for cancer therapy. 9, 37 (2017). 133. Dunn, G. P., Bruce, A. T., Ikeda, H., Old, L. J. & Schreiber, R. D. Cancer immunoediting: From immunosurveillance to tumor escape. Nature Immunology 3, 991–998 (2002). 134. Dunn, G. P., Old, L. J. & Schreiber, R. D. The Three Es of Cancer Immunoediting. Annu. Rev. Immunol. 22, 329–360 (2004). 135. Chen, L. & Flies, D. B. Molecular mechanisms of T cell co-stimulation and co-inhibition. Nature Reviews Immunology 13, 227–242 (2013).   121 136. Chen, C., Kang, Y., Siegel, P. M. & Massague, J. E2F4 / 5 and p107 as Smad Cofactors Linking the TGF ␤ Receptor to c- myc Repression. 110, 19–32 (2002). 137. Rodrik-Outmezguine, V. S. et al. Overcoming mTOR Resistance Mutations with a New Generation mTOR Inhibitor. Nature 534, 272–276 (2016). 138. Zaretsky, J. M. et al. Mutations Associated with Acquired Resistance to PD-1 Blockade in Melanoma. N Engl J Med 375, 819–829 (2016). 139. Higgins, C. F. ABC Transporters: From microorganisms to man. Annual Review of Cell Biology 8, 67–113 (1992). 140. De Vree, J. M. L. et al. Mutations in the MDR3 gene cause progressive familial intrahepatic cholestasis. Proc. Natl. Acad. Sci. U. S. A. 95, 282–287 (1998). 141. Druker, B. J. et al. Activity of a specific inhibitor of the BCR-ABL tyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblastic leukemia with the Philadelphia chromosome. N. Engl. J. Med. 344, 1038–1042 (2001). 142. Zahreddine, H. & Borden, K. L. B. Mechanisms and insights into drug resistance in cancer. Frontiers in Pharmacology 4 MAR, 28 (2013). 143. Michael, M. & Doherty, M. M. Tumoral drug metabolism: Overview and its implications for cancer therapy. J. Clin. Oncol. 23, 205–229 (2005). 144. Harding, S. M. et al. Mitotic progression following DNA damage enables pattern recognition within micronuclei. Nature 548, 466–470 (2017). 145. MacKenzie, K. J. et al. CGAS surveillance of micronuclei links genome instability to innate immunity. Nature 548, 461–465 (2017). 146. Valderrabano, P. et al. Cancer risk associated with nuclear atypia in cytologically indeterminate thyroid nodules: A systematic review and meta-analysis. Thyroid 28, 210–  122 219 (2018). 147. Fenech, M. et al. Molecular mechanisms of micronucleus, nucleoplasmic bridge and nuclear bud formation in mammalian and human cells. Mutagenesis 26, 125–32 (2011). 148. Thompson, S. L. & Compton, D. A. Chromosomes and cancer cells. Chromosome Research 19, 433–444 (2011). 149. Crasta, K. et al. DNA breaks and chromosome pulverization from errors in mitosis. Nature 482, 53–8 (2012). 150. Zhang, C.-Z. et al. Chromothripsis from DNA damage in micronuclei. Nature 522, 179–84 (2015). 151. Hatch, E. M., Fischer, A. H., Deerinck, T. J. & Hetzer, M. W. Catastrophic nuclear envelope collapse in cancer cell micronuclei. Cell 154, 47–60 (2013). 152. Sun, L., Wu, J., Du, F., Chen, X. & Chen, Z. J. Cyclic GMP-AMP synthase is a cytosolic DNA sensor that activates the type I interferon pathway. Science (80-. ). 339, 786–791 (2013). 153. Cai, X., Chiu, Y. H. & Chen, Z. J. The cGAS-cGAMP-STING pathway of cytosolic DNA sensing and signaling. Molecular Cell 54, 289–296 (2014). 154. Dürig, J. et al. A novel nonobese diabetic/severe combined immunodeficient xenograft model for chronic lymphocytic leukemia reflects important clinical characteristics of the disease. Cancer Res. 67, 8653–61 (2007). 155. Chapuy, B. et al. Diffuse large B-cell lymphoma patient-derived xenograft models capture the molecular and biological heterogeneity of the disease. Blood 127, 2203–13 (2016). 156. Tignanelli, C. J., Herrera Loeza, S. G. & Yeh, J. J. KRAS and PIK3CA mutation frequencies in patient-derived xenograft models of pancreatic and colorectal cancer are   123 reflective of patient tumors and stable across passages. Am. Surg. 80, 873–7 (2014). 157. Seol, H. S. et al. Development and characterization of a colon PDX model that reproduces drug responsiveness and the mutation profiles of its original tumor. Cancer Lett. 345, 56–64 (2014). 158. Zhu, Y. et al. Establishment and characterization of patient-derived tumor xenograft using gastroscopic biopsies in gastric cancer. Sci. Rep. 5, 8542 (2015). 159. Dong, R. et al. Histologic and molecular analysis of patient derived xenografts of high-grade serous ovarian carcinoma. J. Hematol. Oncol. 9, 92 (2016). 160. Cavalloni, G. et al. Establishment of a patient-derived intrahepatic cholangiocarcinoma xenograft model with KRAS mutation. BMC Cancer 16, 90 (2016). 161. Rubio-Viqueira, B. et al. An in vivo platform for translational drug development in pancreatic cancer. Clin. Cancer Res. 12, 4652–61 (2006). 162. Merk, J., Rolff, J., Becker, M., Leschber, G. & Fichtner, I. Patient-derived xenografts of non-small-cell lung cancer: a pre-clinical model to evaluate adjuvant chemotherapy? Eur. J. Cardiothorac. Surg. 36, 454–9 (2009). 163. Fiebig, H.-H. et al. Gene signatures developed from patient tumor explants grown in nude mice to predict tumor response to 11 cytotoxic drugs. Cancer Genomics Proteomics 4, 197–209 (2007). 164. de Plater, L. et al. Establishment and characterisation of a new breast cancer xenograft obtained from a woman carrying a germline BRCA2 mutation. Br. J. Cancer 103, 1192–200 (2010). 165. Jin, K. et al. Assessment of a novel VEGF targeted agent using patient-derived tumor tissue xenograft models of colon carcinoma with lymphatic and hepatic metastases. PLoS   124 One 6, e28384 (2011). 166. Yoshida, T. et al. Antiandrogen bicalutamide promotes tumor growth in a novel androgen-dependent prostate cancer xenograft model derived from a bicalutamide-treated patient. Cancer Res. 65, 9611–6 (2005). 167. Bousquet, G. & Janin, A. Patient-Derived Xenograft: An Adjuvant Technology for the Treatment of Metastatic Disease. Pathobiology 83, 170–6 (2016). 168. Ter Brugge, P. et al. Mechanisms of Therapy Resistance in Patient-Derived Xenograft Models of BRCA1-Deficient Breast Cancer. J. Natl. Cancer Inst. 108, (2016). 169. Dobbin, Z. C. et al. Using heterogeneity of the patient-derived xenograft model to identify the chemoresistant population in ovarian cancer. Oncotarget 5, 8750–8764 (2014). 170. Burrows, P. D. & Cooper, M. D. B cell development and differentiation. Curr. Opin. Immunol. 9, 239–44 (1997). 171. Hardy, R. R., Carmack, C. E., Shinton, S. A., Kemp, J. D. & Hayakawa, K. Resolution and characterization of pro-B and pre-pro-B cell stages in normal mouse bone marrow. J. Exp. Med. 173, 1213–25 (1991). 172. Lu, L. & Osmond, D. G. Apoptosis during B lymphopoiesis in mouse bone marrow. J. Immunol. 158, 5136–45 (1997). 173. Wasserman, R., Li, Y. S. & Hardy, R. R. Down-regulation of terminal deoxynucleotidyl transferase by Ig heavy chain in B lineage cells. J. Immunol. 158, 1133–8 (1997). 174. Goodfellow, P. J. & Wells, S. A. RET gene and its implications for cancer. J. Natl. Cancer Inst. 87, 1515–23 (1995). 175. Iwamoto, T. et al. Preferential development of pre-B lymphomas with drastically down-regulated N-myc in the E mu-ret transgenic mice. Eur. J. Immunol. 21, 1809–14 (1991).   125 176. Wasserman, R., Zeng, X. X. & Hardy, R. R. The evolution of B precursor leukemia in the Eµ-ret mouse. Blood 92, 273–282 (1998). 177. Hunger, S. P. & Mullighan, C. G. Acute Lymphoblastic Leukemia in Children. N Engl J Med 373, 1541–52 (2015). 178. Mody, R. et al. Twenty-five – year follow-up among survivors of childhood acute lymphoblastic leukemia : a report from the Childhood Cancer Survivor Study. 111, 5515–5524 (2018). 179. Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: The next generation. Cell 144, 646–674 (2011). 180. Frismantas, V. et al. Ex vivo drug response profiling detects recurrent sensitivity patterns in drug-resistant acute lymphoblastic leukemia. Blood 129, e26–e37 (2017). 181. Kerketta, L. S., Ghosh, K., Nadkarni, A., Madkaikar, M. & Vundinti, B. R. Centrosome aberration frequency and disease association in B-Acute lymphoblastic leukemia. In Vivo (Brooklyn). 31, 215–220 (2017). 182. Wang, Z. et al. Pharmacological inhibition of PARP6 triggers multipolar spindle formation and elicits therapeutic effects in breast cancer. Cancer Res. 78, 6691–6702 (2018). 183. Denu, R. A. et al. Centrosome amplification induces high grade features and is prognostic of worse outcomes in breast cancer. BMC Cancer 16, (2016). 184. Neben, K., Glesecke, C., Schweizer, S., Ho, A. D. & Krämer, A. Centrosome aberrations in acute myeloid leukemia are correlated with cytogenetic risk profile. Blood 101, 289–291 (2003). 185. Kearns, W. G., Yamaguchi, H., Young, N. S. & Liu, J. M. Centrosome amplification and   126 aneuploidy in bone marrow failure patients. Genes Chromosom. Cancer 40, 329–333 (2004). 186. Ganem, N. J., Godinho, S. A. & Pellman, D. A mechanism linking extra centrosomes to chromosomal instability. Nature 460, 278–82 (2009). 187. Wang, Q. et al. BUBR1 deficiency results in abnormal megakaryopoiesis. Blood 103, 1278–1285 (2004). 188. Iwanaga, Y. et al. Heterozygous deletion of mitotic arrest-deficient protein 1 (MAD1) increases the incidence of tumors in mice. Cancer Res. 67, 160–166 (2007). 189. Michel, L. S. et al. MAD2 haplo-insufficiency causes premature anaphase and chromosome instability in mammalian cells. Nature 409, 355–359 (2001). 190. Holland, A. J. & Cleveland, D. W. Boveri revisited: Chromosomal instability, aneuploidy and tumorigenesis. Nat Rev Mol Cell Biol 10, 478–487 (2009). 191. Gisselsson, D. et al. When the genome plays dice: Circumvention of the spindle assembly checkpoint and near-random chromosome segregation in multipolar cancer cell mitoses. PLoS One 3, (2008). 192. Yang, B. et al. Discovery of potent KIFC1 inhibitors using a method of integrated high-throughput synthesis and screening. J. Med. Chem. 57, 9958–9970 (2014). 193. Wu, J. et al. Discovery and mechanistic study of a small molecule inhibitor for motor protein KIFC1. ACS Chem. Biol. 8, 2201–8 (2013). 194. Morris, E. J. et al. Stat3 regulates centrosome clustering in cancer cells via Stathmin/PLK1. Nat. Commun. 8, (2017). 195. Bromberg, J. F. et al. Stat3 as an oncogene. Cell 98, 295–303 (1999). 196. Boehler, C. et al. Poly(ADP-ribose) polymerase 3 (PARP3), a newcomer in cellular   127 response to DNA damage and mitotic progression. Proc. Natl. Acad. Sci. U. S. A. 108, 2783–2788 (2011). 197. Chang, P., Coughlin, M. & Mitchison, T. J. Tankyrase-1 polymerization of poly(ADP-ribose) is required for spindle structure and function. Nat. Cell Biol. 7, 1133–1139 (2005). 198. Xu, X., Persson, H. L. & Richardson, D. R. Molecular pharmacology of the interaction of anthracyclines with iron. Mol. Pharmacol. 68, 261–71 (2005). 199. Walsh, V. & Goodman, J. From taxol to Taxol: the changing identities and ownership of an anti-cancer drug. Med. Anthropol. 21, 307–36 (2002). 200. Huang, T.-C. & Campbell, T. C. Comparison of weekly versus every 3 weeks paclitaxel in the treatment of advanced solid tumors: a meta-analysis. Cancer Treat. Rev. 38, 613–7 (2012). 201. Chondrou, V. et al. Combined study on clastogenic, aneugenic and apoptotic properties of doxorubicin in human cells in vitro. J. Biol. Res. 25, (2018). 202. Schiff, P. B., Fant, J. & Horwitz, S. B. Promotion of microtubule assembly in vitro by taxol [19]. Nature 277, 665–667 (1979). 203. Zasadil, L. M. et al. Cytotoxicity of paclitaxel in breast cancer is due to chromosome missegregation on multipolar spindles. Sci Transl Med. 2014 6, (2014). 204. De Almeida, B. P., Vieira, A. F., Paredes, J., Bettencourt-Dias, M. & Barbosa-Morais, N. L. Pan-cancer association of a centrosome amplification gene expression signature with genomic alterations and clinical outcome. PLoS Comput. Biol. 15, (2019). 205. Reguart, N. et al. BRCA1: a new genomic marker for non-small-cell lung cancer. Clin. Lung Cancer 9, 331–9 (2008). 206. Byrski, T. et al. Response to neo-adjuvant chemotherapy in women with BRCA1-positive   128 breast cancers. Breast Cancer Res. Treat. 108, 289–96 (2008). 207. Quinn, J. E. et al. BRCA1 mRNA expression levels predict for overall survival in ovarian cancer after chemotherapy. Clin. Cancer Res. 13, 7413–7420 (2007). 208. Sung, M. & Giannakakou, P. BRCA1 regulates microtubule dynamics and taxane-induced apoptotic cell signaling. Oncogene 33, 1418–1428 (2014). 209. Fukasawa, K. Oncogenes and tumour suppressors take on centrosomes. Nature Reviews Cancer 7, 911–924 (2007). 210. Alge, C. S., Hauck, S. M., Priglinger, S. G., Kampik, A. & Ueffing, M. Differential protein profiling of primary versus immortalized human RPE cells identifies expression patterns associated with cytoskeletal remodeling and cell survival. J. Proteome Res. 5, 862–878 (2006). 211. Pan, C., Kumar, C., Bohl, S., Klingmueller, U. & Mann, M. Comparative proteomic phenotyping of cell lines and primary cells to assess preservation of cell type-specific functions. Mol. Cell. Proteomics 8, 443–450 (2009). 212. Hillmer, E. J., Zhang, H., Li, H. S. & Watowich, S. S. STAT3 signaling in immunity. Cytokine and Growth Factor Reviews 31, 1–15 (2016). 213. Takeda, K. et al. Targeted disruption of the mouse Stat3 gene leads to early embryonic lethality. Proc. Natl. Acad. Sci. U. S. A. 94, 3801–3804 (1997). 214. Krämer, A., Neben, K. & Ho, A. D. Centrosome aberrations in hematological malignancies. Cell Biol. Int. 29, 375–83 (2005). 215. Krämer, A. et al. Centrosome aberrations as a possible mechanism for chromosomal instability in non-Hodgkin’s lymphoma. Leukemia 17, 2207–13 (2003). 216. Neben, K., Giesecke, C., Schweizer, S., Ho, A. D. & Krämer, A. Centrosome aberrations   129 in acute myeloid leukemia are correlated with cytogenetic risk profile. Blood 101, 289–91 (2003). 217. Catlin, S. N., Busque, L., Gale, R. E., Guttorp, P. & Abkowitz, J. L. The replication rate of human hematopoietic stem cells in vivo. Blood 117, 4460–4466 (2011). 218. Holubcová, Z. et al. Human embryonic stem cells suffer from centrosomal amplification. Stem Cells 29, 46–56 (2011). 219. Cosenza, M. R. & Krämer, A. Centrosome amplification, chromosomal instability and cancer: mechanistic, clinical and therapeutic issues. Chromosome Res. 24, 105–26 (2016). 220. Krämer, A., Maier, B. & Bartek, J. Centrosome clustering and chromosomal (in)stability: a matter of life and death. Mol. Oncol. 5, 324–35 (2011). 221. Nolte, F. et al. Centrosome aberrations in bone marrow cells from patients with myelodysplastic syndromes correlate with chromosomal instability. Ann. Hematol. 92, 1325–33 (2013). 222. Vousden, K. H. & Prives, C. Blinded by the Light: The Growing Complexity of p53. Cell 137, 413–431 (2009). 223. Vogelstein, B., Lane, D. & Levine, A. J. Surfing the p53 network. Nature 408, 307–310 (2000). 224. Montero, J., Dutta, C., Van Bodegom, D., Weinstock, D. & Letai, A. P53 regulates a non-apoptotic death induced by ROS. Cell Death Differ. 20, 1465–1474 (2013). 225. Jabbour, E., O’Brien, S., Konopleva, M. & Kantarjian, H. New insights into the pathophysiology and therapy of adult acute lymphoblastic leukemia. Cancer 121, 2517–2528 (2015). 226. Moorman, A. V. The clinical relevance of chromosomal and genomic abnormalities in B-  130 cell precursor acute lymphoblastic leukaemia. Blood Rev. 26, 123–135 (2012). 227. Paulsson, K. et al. Distinct patterns of gained chromosomes in high hyperdiploid acute lymphoblastic leukemia with t(1;19)(q23;p13), t(9;22)(q34;q22) or MLL rearrangements. Leukemia 27, 974–977 (2013). 228. Chilton, L. et al. High hyperdiploidy among adolescents and adults with acute lymphoblastic leukaemia (ALL): Cytogenetic features, clinical characteristics and outcome. Leukemia 28, 1511–1518 (2014). 229. Tauro, S., McMullan, D., Griffiths, M., Craddock, C. & Mahendra, P. High-hyperdiploidy in Philadelphia positive adult acute lymphoblastic leukaemia: Case-series and review literature. Bone Marrow Transplant. 31, 763–766 (2003). 230. Zappasodi, R., Merghoub, T. & Wolchok, J. D. Emerging Concepts for Immune Checkpoint Blockade-Based Combination Therapies. Cancer Cell 33, 581–598 (2018). 231. Santaguida, S. et al. Chromosome Mis-segregation Generates Cell-Cycle-Arrested Cells with Complex Karyotypes that Are Eliminated by the Immune System. Dev. Cell 41, 638-651.e5 (2017). 232. Bacon, C. M. et al. Interleukin 12 induces tyrosine phosphorylation and activation of STAT4 in human lymphocytes. Proc. Natl. Acad. Sci. U. S. A. 92, 7307–7311 (1995). 233. Starr, R. et al. A family of cytokine-inducible inhibitors of signalling. Nature 387, 917–921 (1997). 234. Conlon, J. et al. Mouse, but not Human STING, Binds and Signals in Response to the Vascular Disrupting Agent 5,6-Dimethylxanthenone-4-Acetic Acid. J. Immunol. 190, 5216–5225 (2013). 235. Coussens, L. M., Zitvogel, L. & Palucka, A. K. Neutralizing tumor-promoting chronic   131 inflammation: a magic bullet? Science 339, 286–91 (2013). 236. Kisurina-Evgen’eva, O. P., Bryantseva, S. A., Shtil’, A. A. & Onishchenko, G. E. Antitubulin agents can initiate different apoptotic pathways. Biophysics (Oxf). 51, 771–775 (2006). 237. Apraiz, A., Boyano, M. D. & Asumendi, A. Cell-centric view of apoptosis and apoptotic cell death-inducing antitumoral strategies. Cancers 3, 1042–1080 (2011). 238. Fenech, M. et al. Molecular mechanisms of micronucleus, nucleoplasmic bridge and nuclear bud formation in mammalian and human cells. Mutagenesis 26, 125–132 (2011). 239. Luzhna, L., Kathiria, P. & Kovalchuk, O. Micronuclei in genotoxicity assessment: From genetics to epigenetics and beyond. Frontiers in Genetics 4, (2013). 240. Kisurina-Evgenieva, O. P., Sutiagina, O. I. & Onishchenko, G. E. Biogenesis of micronuclei. Biochemistry (Moscow) 81, 453–464 (2016). 241. Kalsbeek, D. & Golsteyn, R. M. G2/M-phase checkpoint adaptation and micronuclei formation as mechanisms that contribute to genomic instability in human cells. International Journal of Molecular Sciences 18, (2017). 242. Del Vecchio, M. et al. Interleukin-12: Biological properties and clinical application. Clinical Cancer Research 13, 4677–4685 (2007). 243. Kaplan, M. H. STAT4: A critical regulator of inflammation in vivo. Immunologic Research 31, 231–241 (2005). 244. Yang, X., Chu, Y., Wang, Y., Zhang, R. & Xiong, S. Targeted in vivo expression of IFN-γ-inducible protein 10 induces specific antitumor activity. J. Leukoc. Biol. 80, 1434–1444 (2006). 245. Wendel, M., Galani, I. E., Suri-Payer, E. & Cerwenka, A. Natural killer cell accumulation   132 in tumors is dependent on IFN-γ and CXCR3 ligands. Cancer Res. 68, 8437–8445 (2008). 246. Denkert, C. et al. Tumor-associated lymphocytes as an independent predictor of response to neoadjuvant chemotherapy in breast cancer. J. Clin. Oncol. 28, 105–13 (2010). 247. Kondo, T. et al. High expression of chemokine gene as a favorable prognostic factor in renal cell carcinoma. J. Urol. 171, 2171–2175 (2004). 248. Zhang, L. et al. Intratumoral T Cells, Recurrence, and Survival in Epithelial Ovarian Cancer. N. Engl. J. Med. 348, 203–213 (2003). 249. Wajant, H. The Fas signaling pathway: More than a paradigm. Science 296, 1635–1636 (2002). 250. Nagata, S. Fas and Fas Ligand: A Death Factor and Its Receptor. Adv. Immunol. 57, 129–134 (1994). 251. Stewart, T. J. & Abrams, S. I. How tumours escape mass destruction. Oncogene 27, 5894–5903 (2008). 252. Hanahan, D. & Weinberg, R. A. The hallmarks of cancer. Cell 100, 57–70 (2000). 253. Sträter, J. et al. Impaired CD95 expression predisposes for recurrence in curatively resected colon carcinoma: Clinical evidence for immunoselection and CD95L mediated control of minimal Residual disease. Gut 54, 661–665 (2005). 254. Möller, P. et al. Expression of APO-1 (CD95), a member of the NGF/TNF receptor superfamily, in normal and neoplastic colon epithelium. Int. J. Cancer 57, 371–377 (1994). 255. Darnell, J. E. STATs and gene regulation. Science (80-. ). 277, 1630–1635 (1997). 256. Raccurt, M. et al. Suppressor of cytokine signalling gene expression is elevated in breast carcinoma. Br. J. Cancer 89, 524–532 (2003).   133 257. Evans, M. K. et al. Expression of SOCS1 and SOCS3 genes is differentially regulated in breast cancer cells in response to proinflammatory cytokine and growth factor signals. Oncogene 26, 1941–1948 (2007). 258. Schumacher, T. N. & Schreiber, R. D. Neoantigens in cancer immunotherapy. Science 348, 69–74 (2015). 259. Kawai, T. & Akira, S. The role of pattern-recognition receptors in innate immunity: Update on toll-like receptors. Nature Immunology 11, 373–384 (2010). 260. Hemmi, H. et al. A Toll-like receptor recognizes bacterial DNA. Nature 408, 740–745 (2000). 261. Krieg, A. M. Toll-like receptor 9 (TLR9) agonists in the treatment of cancer. Oncogene 27, 161–167 (2008). 262. Kim, Y. H. et al. Phase i trial of a Toll-like receptor 9 agonist, PF-3512676 (CPG 7909), in patients with treatment-refractory, cutaneous T-cell lymphoma. J. Am. Acad. Dermatol. 63, 975–983 (2010). 263. Zent, C. S. et al. Phase i clinical trial of CpG oligonucleotide 7909 (PF-03512676) in patients with previously treated chronic lymphocytic leukemia. Leuk. Lymphoma 53, 211–217 (2012). 264. Yu, S. et al. Chimeric antigen receptor T cells: a novel therapy for solid tumors. J. Hematol. Oncol. 10, 78 (2017). 265. Yu, S. et al. Recent advances of BsAb in solid tumors; BsAb formats. J. Hematol. Oncol. 10, 155 (2017). 266. Yi, M. et al. Biomarkers for predicting efficacy of PD-1/PD-L1 inhibitors. Mol. Cancer 17, 129 (2018).   134 267. Pardoll, D. M. The blockade of immune checkpoints in cancer immunotherapy. Nature Reviews Cancer 12, 252–264 (2012). 268. Topalian, S. L., Drake, C. G. & Pardoll, D. M. Immune checkpoint blockade: A common denominator approach to cancer therapy. Cancer Cell 27, 450–461 (2015). 269. Rizvi, N. A. et al. Mutational landscape determines sensitivity to PD-1 blockade in non–small cell lung cancer. Science (80-. ). 348, 124–128 (2015). 270. Le, D. T. et al. PD-1 Blockade in Tumors with Mismatch-Repair Deficiency. N Engl J Med 372, 2509–2520 (2015). 271. Greaves, M. F., Maia, A. T., Wiemels, J. L. & Ford, A. M. Leukemia in twins: Lessons in natural history. Blood 102, 2321–2333 (2003). 272. Cardoso, A. A. et al. Pre-B acute lymphoblastic leukemia cells may induce T-cell anergy to alloantigen. Blood 88, 41–48 (1996). 273. Sen, T. et al. Targeting DNA damage response promotes antitumor immunity through STING-mediated T-cell activation in small cell lung cancer. Cancer Discov. 9, 646–661 (2019). 274. Pizzo, P. & Poplack, D. Principles and Practice of Pediatric Oncology - Philip A. Pizzo, David G. Poplack - Google Books. Philadelphia, PA: Wolters Kluwer/Lippincott Williams & Wilkins Health (2011). Available at: https://books.google.ca/books?hl=en&lr=&id=NM36CQAAQBAJ&oi=fnd&pg=PT81&ots=IN81XWCnhS&sig=kagaZPpX4IEtQ0IHVL7zCUYzgck&redir_esc=y#v=onepage&q&f=false. (Accessed: 22nd February 2020) 275. Kawano, M., Itonaga, I., Iwasaki, T. & Tsumura, H. Enhancement of antitumor immunity by combining anti-cytotoxic T lymphocyte antigen-4 antibodies and cryotreated tumor   135 lysate-pulsed dendritic cells in murine osteosarcoma. Oncol. Rep. 29, 1001–1006 (2013). 276. Shen, J. K. et al. Programmed cell death ligand 1 expression in osteosarcoma. Cancer Immunol. Res. 2, 690–698 (2014). 277. Haworth, K. B. et al. Going back to class I: MHC and immunotherapies for childhood cancer. Pediatric Blood and Cancer 62, 571–576 (2015). 278. Gisselsson, D., Pålsson, E., Yu, C., Mertens, F. & Mandahl, N. Mitotic instability associated with late genomic changes in bone and soft tissue tumours. Cancer Lett. 206, 69–76 (2004). 279. Al-Romaih, K. et al. Chromosomal instability in osteosarcoma and its association with centrosome abnormalities. Cancer Genet. Cytogenet. 144, 91–9 (2003). 280. Maude, S. L. et al. Tisagenlecleucel in Children and Young Adults with B-Cell Lymphoblastic Leukemia. N. Engl. J. Med. 378, 439–448 (2018). 281. Administration, U. S. F. and D. FDA approves tisagenlecleucel for B-cell ALL and tocilizumab for cytokine release syndrome. (2017). Available at: https://www-fda-gov.ezproxy.library.ubc.ca/drugs/resources-information-approved-drugs/fda-approves-tisagenlecleucel-b-cell-all-and-tocilizumab-cytokine-release-syndrome. (Accessed: 5th February 2020) 282. NOVARTIS. Novartis completes certification of initial sites in Ontario for first approved Canadian CAR-T therapy, Kymriah® (tisagenlecleucel)1 | Novartis Canada. (2019). Available at: https://www.novartis.ca/en/news/media-releases/novartis-completes-certification-initial-sites-ontario-first-approved-canadian. (Accessed: 23rd February 2020) 283. Medlior Health Outcomes Research. CAR-T THERAPY : CHALLENGES FOR. (2019). 284. Chong, E. A. et al. PD-1 blockade modulates chimeric antigen receptor (CAR)-modified T   136 cells: Refueling the CAR. Blood 129, 1039–1041 (2017). 285. Jo, S. et al. Generation of a multi-antigen-directed immune response for durable control of acute lymphoblastic leukemia. Leukemia 32, 539–572 (2018).  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0390963/manifest

Comment

Related Items