The Pennsylvania State University
The Graduate School
College of Medicine
THE NOVEL MECHANISTIC ROLE OF PIGN IN LEUKEMIA PROGRESSION
A Dissertation in
Biomedical Sciences by
Emmanuel Kwame Teye
2017 Emmanuel Kwame Teye
Submitted in Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
December 2017
The dissertation of Emmanuel Kwame Teye was reviewed and approved* by the following:
Jeffrey J Pu Assistant Professor of Medicine Dissertation Advisor Chair of Committee
Jong K Yun Associate Professor of Pharmacology Director, Translational Therapeutics Option, Biomedical Sciences
Douglas B Stairs Assistant Professor of Pathology and Pharmacology
Gregory S Yochum Associate Professor of Biochemistry and Molecular Biology
Hong-Gang Wang Professor of Pediatrics and Pharmacology
Ralph L Keil Associate Professor of Biochemistry and Molecular Biology Chair, Biomedical Sciences Graduate Program
*Signatures are on file in the Graduate School
iii
ABSTRACT
Genomic instability plays a pivotal role in the leukemia progression of myelodysplastic
syndromes (MDS). However, the precise genetic cause and the underlying mechanisms of MDS
leukemia progression or transformation to acute myeloid leukemia (AML) remain elusive.
Moreover, the current approach to MDS progression risk-stratification using the International
Prognostic Scoring System (IPSS) or WHO Prognostic Scoring System (WPSS) is limited and fails
to address the dynamic nature of the disease. This presents a need for the identification of novel prognostic markers and a better understanding of the mechanisms involved.
In this study, phosphatidylinositol glycan anchor biosynthesis, class N (PIGN), a gene encoding an enzyme participating in the final steps of the glycophosphatidylinositol-anchored protein (GPI-AP) biosynthesis pathway and a cancer chromosomal instability (CIN) suppressor, was highly ranked as a predictor of the risk of MDS leukemia progression. We also observed the progressive loss of PIGN during MDS progression to AML. Moreover, PIGN gene expression
aberrations (i.e. increased gene expression but diminished to no protein production) were observed
in a subset of high-risk MDS and AML patients with myelodysplasia-related changes. PIGN gene
expression aberrations were associated with increased frequency of GPI-AP deficiency in leukemic
cells and correlated with the elevation of genomic instability that was independent of the TP53
regulatory pathway. PIGN gene expression aberrations were attributed to novel partial intron
retentions between exons 14 and 15 resulting in frameshifts and premature termination.
Interestingly, in an MDS leukemia progression model, this mutation was identified in the leukemia
stage (i.e. MDS-L) cells but was absent in the MDS stage (i.e. MDS92) cells.
Transient suppression or ablation of PIGN induced DNA damage response which was
rescued following PIGN restoration. We also observed an increase in the frequency of CIN with
PIGN loss. Moreover, PIGN physically interacted with and/or regulated the spindle assembly
iv checkpoint via MAD1, MAD2, MPS1, and BUBR1. Thus, PIGN is crucial in the regulation of mitotic integrity for the maintenance of chromosomal stability and ultimately prevents leukemic transformation/progression.
This study for the first time identified PIGN as a prognostic marker of MDS transformation and revealed the link between PIGN gene expression aberration, genomic instability, and MDS progression/leukemia transformation. PIGN gene expression aberration is associated with genomic instability and leukemogenesis and could serve as a basis for improved risk-stratification of MDS patients.
v
TABLE OF CONTENTS
List of Figures ...... vii
List of Tables ...... viii
Abbreviations ...... ix
Preface ...... xii
Acknowledgements ...... xiii
LITERATURE REVIEW ...... 1
Myelodysplastic Syndromes (MDS) ...... 2 Current MDS therapies ...... 4 MDS risk-stratification and current challenges ...... 6 Molecular markers of MDS progression ...... 9 Chromosomal instability ...... 15 Role of CIN in MDS/AML progression ...... 20 CIN as a marker of MDS leukemic transformation...... 22 Therapeutic targeting of CIN ...... 24 The spindle assembly checkpoint ...... 25 Cross-talk between DDR and CIN induction ...... 27 SAC/mitotic checkpoint dysregulation in MDS/AML progression ...... 29 Phosphatidylinositol Glycan Class N (PIGN) ...... 30 PIGN and GPI-AP biosynthesis ...... 32 Role of PIGN in CIN and leukemic progression ...... 36
PIGN gene expression aberration is associated with genomic instability and leukemic progression in acute myeloid leukemia with myelodysplasia-related changes ...... 37
Abstract ...... 38 Introduction ...... 39 Materials and Methods ...... 41 Results ...... 49 Discussion ...... 73 Conclusions ...... 78
PIGN spatiotemporally regulates the spindle assembly checkpoint ...... 79
Abstract ...... 80 Introduction ...... 81 Materials and Methods ...... 83 Results ...... 88 Discussion ...... 101 Conclusions ...... 107
vi
SUMMARY & FUTURE DIRECTIONS ...... 108
Translational Application ...... 118 Study Limitations ...... 119 CONCLUSION ...... 122 Appendix A Permissions ...... 123 Appendix B Supplemental Tables ...... 128 Appendix C Primer Sequences...... 129 Appendix D Cell cycle frequency post-PIGN knockout ...... 130 Appendix E RT-qPCR profile post-PIGN knockout ...... 131 Appendix F Co-localization analyses of mutant PIGN ...... 132 REFERENCES...... 133
vii
LIST OF FIGURES
Figure 1-1 Representative Wright-Giemsa stained film of a high-risk MDS patient with refractory anemia with excess blasts...... 3
Figure 1-2 Mutational overlap of frequently mutated genes in MDS...... 14
Figure 1-3 CIN is a form of GIN and could be sub-classified as nCIN and sCIN ...... 17
Figure 1-4 Lagging chromosomes and anaphase bridges are the results of segregation errors in CIN...... 19
Figure 1-5 The SAC signaling pathway...... 26
Figure 1-6 Schematic of the TP53-mediated cross-talk between SAC and DDR...... 28
Figure 1-7 PIGN transfers the first EtNP residue to a mannose residue in the multi-step GPI-AP biosynthesis and remodeling pathway...... 31
Figure 1-8 The intracellular localization pattern of PIGN...... 33
Figure 2-1 PIGN is a predictive biomarker for MDS risk-stratification...... 51
Figure 2-2 PIGN gene expression aberration was due to truncation...... 56
Figure 2-5 PIGN expression aberration was associated with genomic instability in leukemic cells and was TP53- pathway-independent...... 66
Figure 2-6 PIGN gene expression suppression was associated with genomic instability, and reintroduction of PIGN gene expression restored genomic stability in a TP53- pathway-independent manner...... 72
Figure 3-1 PIGN expression is cell cycle-regulated...... 89
Figure 3-2 PIGN loss or suppression results in the disrupted expression of SAC components...... 94
Figure 3-4 PIGN co-localizes with SAC components during SAC activation...... 99
Figure 3-5 Models of the spatiotemporal interaction between PIGN and SAC components during the cell cycle and SAC activation...... 105
Figure 4-1 This simplified model summarizes the relationship between PIGN expression aberration, SAC regulation, and leukemia progression...... 115
viii
LIST OF TABLES
Table 1-1 Current MDS Classification and Prognostic Scoring Systems...... 7
Table 1-2 Mutations indicated as markers of leukemia progression in MDS patients ...... 11
Table 1-3 Germline mutations in PIGN as associated congenital and developmental conditions...... 35
Table 2-1 PIGN gene and protein expression status in MDS or AML-MRC patients...... 53
Table 2-2 Verified common NL and leukemic mutations in the TP53 gene ...... 69
ix
ABBREVIATIONS
MDS Myelodysplastic Syndromes
AML Acute Myeloid Leukemia
PIGN Phosphatidylinositol Glycan Class N
PIGA Phosphatidylinositol Glycan Class A
MAD1 Mitotic Arrest Deficient (MAD) 1
MAD2 Mitotic Arrest Deficient (MAD) 2
AML-MRC AML with Myelodysplasia-Related Changes
GPI-AP Glycophosphatidylinositol-Anchored-Protein
SAC Spindle Assembly Checkpoint
CIN Chromosomal instability
GIN Genomic instability
MIN Microsatellite instability
EtNP Phosphoethanolamine
IPSS International Prognostic Scoring System
MPS Monopolar spindle 1-like 1
BUBR1 Budding Uninhibited by Benzimidazoles 1 Beta
AURKA Aurora kinase A
AURKB Aurora kinase B
BUB1 Budding Uninhibited by Benzimidazoles 1
BUB3 Budding Uninhibited by Benzimidazoles 3
PLK1 Polo-like kinase 1
CDC20 Cell Division Cycle 20
CENPE Centromeric Protein E
APC/C Anaphase-Promoting Complex
x
DPM2 Dolichol Phosphate-Mannose Biosynthesis Regulatory Protein
CDC25A Cell Division Cycle 25A
CENPA Centromeric Protein A
CENPB Centromeric Protein B
CCNB1 Cyclin B1
CCNB2 Cyclin B2
CDK1 Cyclin-Dependent Kinase 1
CFC Colony forming cell
CGH Comparative genomic hybridization
COSMIC Catalog of somatic mutation in cancer
DDR DNA damage repair
FAB French-American-British
FISH Fluorescent in-situ hybridization
GR Glucocorticoid receptor
HSCT Hematopoietic stem cell transplantation
HSPC Hematopoietic stem and progenitor cells
IGV Integrated Genome Viewer
IRB Institutional Review Board
LOH Loss of heterozygosity
MACS Magnetic-activated cell sorting
MC Metaphase cytogenetics
MCC Mitotic checkpoint complex
MPD Myeloproliferative disease
NL Non-leukemic
xi
NMD Nonsense-mediated decay
PTC Premature termination codons
RA Refractory anemia
RAEB Refractory anemia with excess of blasts
RARS Refractory Anemia with Ring Sideroblasts
SIFT Sorting Intolerant From Tolerant
SNP Single-nucleotide polymorphism
SRA Sequence raw archive
WHO World Health Organization
WPSS WHO Prognostic Scoring System
xii
PREFACE
Permission to reprint published article was obtained from the publisher and is shown below:
Oncotarget
“Oncotarget applies the Creative Commons Attribution 3.0 License (CC BY 3.0) to all works we publish (read the human-readable summary or the full license legal code). Under the CC BY, authors retain ownership of the copyright for their article, but authors allow anyone to download, reuse, reprint, modify, distribute, and/or copy articles in Oncotarget, so long as the original authors and source are cited.”
No permission is required from the authors or the publishers.
xiii
ACKNOWLEDGEMENTS
I owe my utmost thanks to Jesus Christ my Lord and Savior for I would not have made it this far without Him. My beautiful wife Simi and my lovely daughter Adura continue to be my biggest fans. They cheered me on especially when things got tough in the lab. I dedicate this dissertation to my parents John and Rebecca Teye. I also dedicate this dissertation to my god- parents Dr. Martin and Mary Catherine Limbird through whose beneficence I had the opportunity to study in the United States. The constant prayers and support of my family and friends in Ghana and the USA especially Vasty, Victoria, Jojo and Clement have been a great source of encouragement to me. I am also grateful to my church family, Deeper Life Bible Church for giving me a family away from home and place of succor in trying times. My pastors Pastor Vincent Bello, Pastor Michael Dada and Pastor Wale Davis were always there to pray with me and encourage me.
Drs. Lehman, Spinelli, Karelia, Tan, Hasanali, Kidacki, Keil, Bronson, Coleman, Nelson, Shantz, Degraff and Stairs made themselves available whenever I called upon them to just talk, to peruse my write-ups or critique my experimental designs. The joint lab meetings established by Dr. Stairs were a great resource for my research. I gained helpful feedback and constructive criticism on my project during lab meetings. I cannot overlook the contributions of my past mentors Drs. Owiredu, Vann and Hafer throughout my career. I am a product of their astounding mentorship. I would like to thank Drs. Yang, Amin, Sharma, Kirschbaum and El-Deiry for giving me the opportunity to rotate in their prestigious laboratories. I am immensely grateful to my dissertation committee, Drs. Wang, Stairs, Yochum and Yun for all their support and advice throughout this project. This project would not have seen any success without their invaluable counsel.
I would like to extend my profound gratitude to the patients and their families for contributing valuable specimens for this project. My appreciation goes out to the wonderful staff of the Special Hematology Lab, Phlebotomy Lab, HSCT bank, Hematology office, the Clinical trials office, Pathology laboratory and administrative staff and the graduate affairs office for all their love and support. The Gittlen, Moldovan, Stairs, Yochum and Sample labs were very generous with their equipment, lab space and reagents. Special thanks go to Ms. Karen Shields and Ms. Xin Ping for all the encouragement and support they gave me especially during times of dejection. Ms. Ping was also a great resource to me for designing and conducting experiments.
Last but not least, my heartfelt gratitude goes to my mentor Dr. Jeffrey J Pu. His overwhelming support and guidance in and out of lab cannot be overstated. I am filled with gratitude for the opportunity he gave me to work in his laboratory. He took the bold risk of admitting me into his lab as his first graduate student when no one else would. This project would not have been possible without his support and guidance.
LITERATURE REVIEW
A portion of the material in this chapter has been published as part of an original research article
in Oncotarget.
A portion of this review has been published as United States Patent and Trademark Office
Application No. 15/434,774 ‘Method and Therapeutic Use of PIGN and Other Genes or Gene
Products That PIGN Interacts With for Prognosis and Treatment of Hematological Neoplasias’.
2 Myelodysplastic Syndromes (MDS)
Myelodysplastic syndromes (MDS) also previously referred to as “preleukemia” are a clinically diverse and biologically heterogeneous spectrum of chronic, clonal hematological malignancies of hematopoietic stem or progenitor cells that are characterized by ineffective myeloid hematopoiesis, limited self-renewal, and uni- or multi-lineage dysplasia (Figure 1-1)1–4. In
adults, MDS represents the most common class of acquired bone marrow failure diseases and may
arise de novo (i.e. 65%) or because of prior chemo- or radiotherapy which constitutes ~10% of
cases (i.e. treatment-related MDS)5,6. In the past, it was problematic to accurately estimate the number of MDS cases due to the lack of data collection by national registries and because MDS was not previously classified as neoplastic. MDS is more prevalent in males and is typically diagnosed in individuals between 65-70 years of age7,8. Aside from advanced age and prior exposure to cytotoxic therapies, family history of hematopoietic cancers, exposure to industrial
solvents and agrochemicals and smoking are common risk factors. The disease affects about
30,000-40,000 individuals per annum within the United States with a 30% propensity of
progression into a more aggressive secondary AML (sAML). Essentially, one out of every 4 MDS
patients eventually progresses to AML with an annual transformation rate of 3%9–11. Moreover,
high-risk MDS is often characterized by losing of chromosomal segments and translocations that disrupt genes involved in proliferation and differentiation12. A myeloid blast count ≥20% in the blood or bone marrow and a complex karyotype mark the leukemic transformation from MDS to
AML13.
3
Figure 1-1 Representative Wright-Giemsa stained film of a high-risk MDS patient with refractory anemia with excess blasts.
This image was originally published in ASH Image Bank by Peter Maslak. The bone marrow aspirate reveals dysplastic erythroid cells with myeloid maturational arrest. 04/25/2004; image number-00002535. Permission provided by ©The American Society of Hematology.
4 By convention, MDS is re-classified as AML when bone marrow blast population reaches
or exceeds 20%. This secondary AML is more aggressive and molecularly diverse and involves
unconstrained proliferation of aberrant myeloid progenitor cells. These abnormal cells possess
clonal chromosomal aberrations, populate the bone marrow and peripheral blood, and contribute to
AML progression by driving clonal evolution14. MDS transformation to AML involves the survival and self-renewal ability of chromosomally unstable leukemic clones15. As a result, the prognosis is grim, and the disease becomes more refractory to treatment post-MDS progression to AML16.
Current MDS therapies
The clinical heterogeneity of MDS attests to the molecularly diverse nature of the disease.
Current diagnosis of MDS entails an assessment of the morphology of peripheral blood and bone
marrow cells, blast counts, cytopenia and cytogenetic analyses17. Clonal cytogenetic aberrations
occur in about 80% of therapy-related MDS whereas it occurs in about 50% de novo MDS
patients18. The molecularly diverse nature of MDS and inadequate assessment complicates
treatment. Moreover, the toxicity associated with current drug therapies makes the treatment of
elderly patients difficult. Thus, there is an urgent need for more effective targeted therapies.
Chemotherapy
Low-risk MDS patients with cytopenia are typically treated with erythropoiesis-
stimulating agents with or without growth factors such as granulocyte-colony stimulating factor19.
Additionally, immunomodulatory agents such as lenalidomide have been shown to be efficacious
in MDS patients with 5q deletions20–26. In an attempt to delay AML progression, high-risk MDS patients are treated with cytarabine, idarubicin, and daunorubicin27. Immunosuppressive drug
5 therapies including anti-thymocyte globulin and cyclosporine have been successfully used in immune-mediated MDS 28. Alternatively, due to ineffective hematopoiesis, patients clinically
present with variable peripheral blood cytopenia which may be treated with low-dose of the DNA
methyltransferase inhibitors azacytidine or decitabine in about 50% of high-risk MDS patients29–
33. Unfortunately, response to treatment is abysmal in high-risk MDS patients, particularly in older patients (i.e. ≥ 60 years old) with a median survival of approximately 6-12 months. As such, there is an urgent need for novel targeted therapies despite limitations in the availability of molecular targets6,34.
Hematopoietic stem cell transplantation
Currently available MDS therapeutic agents are insufficient, and may only prolong life but are not curative29. Therefore, allogeneic hematopoietic stem cell transplantation (HSCT) remains the only potentially curative treatment35. Unfortunately, finding a matching donor is very tasking,
and some patients refuse to undergo the procedure because of concerns about complications and
adverse effects on their quality of life. However, the introduction of cord blood from haploidentical
donors and less intensive conditioning for transplantation particularly in older patients has helped
address some of these concerns 36. Sadly, only 5% of MDS patients undergo HSCT despite the
contingencies above being established to reduce adverse effects in patients 37. Relapse is the leading
cause of HSCT failure in MDS patients. In such cases, palliative care, low-dose treatment with
hypomethylating agents, donor lymphocyte infusions or a second HSCT are recommended38. MDS
patients who remain refractory to treatment or relapse are eventually enlisted for clinical trials.
6 MDS risk-stratification and current challenges
MDS risk-stratification is vital for accurate diagnosis and the identification of patients at risk of progression to AML. Risk-stratification helps to proactively develop and select therapeutic strategies to prevent or delay MDS progression for better patient survival. The most accepted scoring systems are included in Table 1-1 and have been defined by the French-American-British
(FAB) criteria and the World Health Organization (WHO) classification and prognostic scoring system (WPSS) and the Revised International Prognostic Scoring System (IPSS-R) which incorporates blast percentage, chromosomal aberrations (karyotype) and assessment of peripheral blood cytopenias39–41. The IPSS-R system has over the years become the de facto standard prognostic approach.
MDS risk-stratification can be complicated and inconsistent due to the variability of clinical outcomes even between MDS patients within the same disease subtype and the subjective disparities in inter-observer morphological assessment42. Methods such as flow cytometry, gene expression profiling, and genome-wide copy number analyses have been proposed, but these approaches need to be validated and standardized for MDS prognostication in the clinic43,44.
Moreover, it is challenging to conduct longitudinal studies to follow the progression of MDS
patients into AML due to the limitations of the current risk-stratification approach.
7 Table 1-1 Current MDS Classification and Prognostic Scoring Systems.
*FAB 39 MPSS45 IPSS-R46 WPSS47
MDS with single lineage
dysplasia (MDS-SLD)
MDS with multi-lineage
dysplasia (MDS-MLD)
Refractory anemia (RA) Low Very Low-risk (≤1.5)
Low-risk (>1.5 to 3)
Refractory anemia MDS with ring sideroblasts
with ring sideroblasts (RARS) (MDS-RS)
MDS-RS with single lineage
dysplasia (MDS-RS-SLD)
MDS-RS with multi-lineage
dysplasia (MDS-RS-MLD)
MDS with isolated del(5q)
Refractory anemia with excess of Intermediate 1 Intermediate risk (> 3 to MDS with excess blasts
blasts (RAEB) Intermediate 2 4.5) (MDS-EB)
Refractory anemia with excess of High High-risk (>4.5 to 6) MDS-EB-1
blasts in transformation (RAEB- Very high-risk (>6) MDS-EB-2
t)
Chronic myelomonocytic
leukemia (CMML)
MDS, unclassifiable (MDS-U)
*Refer to the individual citations for details
8 The FAB classification is the oldest of the three MDS prognostic scoring methods and
stratifies MDS into five subtypes by bone marrow morphology, bone marrow and peripheral blood blast percentage, the occurrence of ring sideroblasts and circulating monocytes39. Subsequently, the
WPSS approach was developed using the FAB approach as a template but with better
characterization of the number of dysplastic lineages while incorporating cytogenetic aberrations
and was revised only last year47. These risk-stratification and scoring systems although far from
perfect, have provided some level of risk-stratification and facilitated the prediction of MDS risk
of AML progression, optimal treatment evaluation and assessment of overall survival. In the IPSS-
R system, MDS patients are assigned to one of five risk groups (i.e. very low-risk, low-risk,
intermediate risk, high-risk, very high) based on the cytogenetic abnormalities, cytopenias, and
bone marrow blasts. Median survival based on very low-risk to very high-risk MDS ranges from
8.8 years to 0.8 years respectively46.
Karyotypic scoring via the assessment of metaphase cytogenetics (MC) is a major
component of these scoring methods and has become a gold standard in MDS prognostication48–51.
However, this is limited by natural cytogenetic diversity (including rare cytogenetic abnormalities) and heterogeneity while fraught with overlaps39,46,47,52,53. Moreover, the use of metaphase cytogenetic assessments in the clinical laboratory is technically demanding and time-consuming in addition to the difficulties associated with culturing and maintaining MDS mononuclear cells or hematopoietic stem and progenitor cells (HSPCs) in culture.
Recent updates to the WPSS method of MDS stratification and the IPSS scoring system
(IPSS-R) have attempted to ameliorate some of the heterogeneity encountered with the prior versions of classification, the latter being the most widely accepted for MDS risk assessment46,47.
Moreover, the recent efforts to incorporate somatic mutation signatures as part of risk assessments
have been challenged due to the detection of some of these mutations in healthy individuals,
particularly in aged populations54–58. The IPSS-R method seems to be the most reliable because it
9 is intuitive, highly reproducible and was generated from a treatment naïve cohort of de novo MDS
patients but its utility becomes limited in serial studies46. However, this is not the case in the M.D.
Anderson Prognostic scoring system (MPSS) which incorporates adverse prognostic parameters such as old age, thrombocytopenia, anemia, increased bone marrow blasts, leukocytosis, complex cytogenetics and prior transfusions to group MDS patients into four prognostic groups45. The MPSS
system is ideal for serial studies to investigate the same MDS patient over the course of disease progression regardless of prior exposure to therapy59.
Despite the inadequacy and shortfalls of the traditional classification systems, these scoring
systems are currently employed for clinical diagnosis and prognostication. Moreover, these widely
adopted scoring methods fail to acknowledge molecular genetic abnormalities or the highly
dynamic and heterogeneous nature of MDS. Thus, incorporation of prognostic molecular markers
into these existing classification systems will improve MDS risk-stratification, facilitate the
selection of personalized therapies and ultimately yield better patient outcomes.
Molecular markers of MDS progression
Research to identify molecular markers for the prognostication of MDS progression has
become an ever increasingly popular area of study due to the heterogeneity of the disease. In the prognostication of MDS progression, it might later be necessary to integrate recently discovered molecular genetic abnormalities into the current MDS classification systems. Genetic mutations have a high correlation with disease phenotypes, and this makes them ideal for prognostication42,60–
62. These genetic markers may be involved in growth factor signaling, transcriptional regulation,
RNA splicing, apoptosis, epigenetic regulation and genomic instability27,63–65. The goal is to
identify a robust prognostic molecular marker that is distinctly mutated or dysregulated in a
10 subclass of MDS. This would provide deeper insight into the mechanism and pathobiology of MDS
transformation for the improvement of leukemia-free survival in high-risk MDS patients.
Markers associated with transcriptional regulation
Point mutations in the gene encoding the transcription factor, RUNX1 have been detected
in up to 20% of patients with MDS and AML with MDS-related changes (AML-MRC) (Table 1-
2). However, RUNX1 mutations are more prevalent in secondary MDS than in de novo MDS with
a worse prognosis in the former compared to patients with no RUNX1 mutation66,67. Another genetic marker that has been explored is the transcriptional regulator ASXL1. ASXL1 mutations occur in up to 15% of MDS patients and are associated with poor overall survival and poor prognosis independent of other risk factors related to the development and progression of MDS61,68.
11
Table 1-2 Mutations indicated as markers of leukemia progression in MDS patients
Mutation Frequency Clinical significance/Survival
TET2 Up to 26% Improved response to azacitidine therapy69
RUNX1 Up to 20% Poor survival70
ASXL1 Up to 15% Poor survival 71
EZH2 Up to 6% Poor survival 72
DNMT3A Up to 8% Poor survival with risk of transformation to AML73
IDH1 Up to 10% Poor survival 74
12 Markers associated with epigenetic regulation
TET2 mutations have been detected in about 25% of MDS patients and were related to
favorable prognosis in a cohort of MDS patients with nearly 80% survival rate over a period of 5
years. Moreover, leukemia progression-free survival was ~90% compared to 63.7% in patients with
no detectable TET2 mutations. Moreover, response to azacytidine treatment was almost four times
higher in the MDS patients with TET2 mutations versus the wild-type TET2 group75,76. Nonetheless,
further studies are required to determine the prognostic utility of TET2 as a marker for MDS risk-
stratification.
The gene encoding the methyltransferase, EZH2 is mutated or overexpressed in several
myeloid malignancies, more specifically those patients with aggressive disease including high-risk
MDS77,78. EZH2 mutations were detected in 6% of a small cohort of 126 MDS patients78.
Overexpression of EZH2 in MDS may indicate a poor prognosis even in lower-risk MDS cases60.
Mutations in the epigenetic regulator, DNMT3A occur in about 8% of de novo MDS patients and might be an early event in MDS disease progression59,73. MDS patients with DNMT3A mutations have an overall poor survival and greater risk of progression to AML61,62. Acquired isocitrate dehydrogenase 1/2 (IDH1/2) mutations occur in ~9% of MDS cases80. In MDS patients with only
5q deletion, an additional mutation in IDH1/2 has been linked with poor overall survival and
leukemia-free survival74,81. A subsequent study conducted two years later more specifically
reported adverse prognosis for MDS patients with mutant IDH1 but not in those patients bearing the IDH2 mutation74,82.
13 Markers associated with the RNA splicing machinery
RNA splicing is the most predominantly perturbed molecular mechanism in MDS pathogenesis6. Primary mRNA splicing mediators including U2AF1, SRSF2, and SF3B1 are frequently (i.e. ~44%) dysregulated during MDS progression83,84. Other studies have reported
splicing factor mutations in more than 50% of MDS patients6. U2AF1 and SRSF2 mutations are
associated with poor overall survival in MDS patients, unlike SF3B1 which is associated with a favorable prognosis85. SF3B1 is frequently (i.e. >70%) mutated in MDS patients with ring
sideroblasts but has no additional prognostic utility for MDS risk-stratification62,82,85. Despite the prevalence of splicing factor dysregulation in MDS, their involvement in the pathogenesis and progression of the disease remains elusive due to mutational overlaps within MDS subgroups
(Figure 1-2).
14
Figure 1-2 Mutational overlap of frequently mutated genes in MDS.
Catalog of somatic mutation in cancer (COSMIC) data of genes with ≥9% frequency of mutation in MDS patients.
15 Chromosomal instability
David Paul von Hansemann in 1891 was the first, more than a century ago to describe and
link cancer with asymmetric chromosomal divisions now referred to as chromosomal instability
(CIN). CIN is a form of genomic instability (GIN) that involves a high rate of gross imbalances in
chromosome number within a cell and loss of heterozygosity (LOH) resulting in either double
strand breaks or continuous chromosome missegregation during mitosis86–88. Some studies have
attempted to distinguish between structural (sCIN) and numerical (nCIN) chromosomal instability
although both are similarly generated (Figure 1-3). Structural CIN pertains to an increased rate of
chromosomal breaks, deletions, inversions, reciprocal or non-reciprocal translocations and DNA
amplification. Numerical CIN describes changes in chromosome counts associated with relatively
higher rates of whole-chromosome gains or losses compared to normal cells (nCIN)89.
Chromosomally stable cells have a low (i.e. ~1% of cell divisions) missegregation rate while
missegregation rates may be as high as occurring on at least every fifth cell division in CIN-positive
cells90,91. Aberrant cohesion between the sister chromatids of a chromosome, supernumerary centrosomes and dysregulated expression of centromere proteins have also been shown to hinder proper segregation during mitotic division and to result in CIN ultimately92,93. CIN is a dynamic state and has previously been described as a hallmark and an enabling characteristic of solid tumors and hematological malignancies94,95. CIN manifests morphologically as anaphase bridges or
lagging chromosomes (Figure 1-4). Anaphase bridges are primarily continuous strings of chromatin
that stretch between the two anaphase poles of the dividing cells. Lagging chromosomes simply
are lagging of whole chromosomes and sometimes acentric fragments between the dividing sets of
chromosomes during metaphase to anaphase transition96. CIN could be an early event during
tumorigenesis and might precede oncogenic mutations that ultimately arise due to the activation of
16 DNA damage response to prior cytotoxic treatments as well as environmental or occupational toxins that induce oncogenic stress and centrosome or kinetochore aberrations63,95,97–99.
17
Figure 1-3 CIN is a form of GIN and could be sub-classified as nCIN and sCIN
18 Although CIN has been implicated in tumorigenesis and multidrug resistance, its
mechanistic role in cancer progression remains unclear100,101. In contrast to CIN-negative cells,
CIN-positive cell lines possess intrinsic multidrug resistance irrespective of the somatic mutational architecture and proliferation rate of the cells101. An in-depth understanding of the genetic basis and underlying molecular mechanisms of CIN will aid in the development of novel therapies to curb it.
19
Figure 1-4 Lagging chromosomes and anaphase bridges are the results of segregation errors in
CIN.
Representative images of missegregation errors observed in HEK293 cells; Blue (chromosomes),
Red (centromere). White arrows indicate lagging chromosomes and the positions of anaphase bridges.
20 Role of CIN in MDS/AML progression
Chromosomal instability (CIN) is a type of genomic instability in which either whole
chromosomes or parts of chromosomes are duplicated or deleted102. The unbalanced distribution of
DNA to daughter cells upon mitosis results in a failure to maintain euploidy leading to
aneuploidy98,103. CIN plays a pivotal role in driving the leukemic progression of MDS to AML
which occurs in up to 50% of high-risk MDS patients104. This is buttressed by the role that CIN plays in the leukemic progression of Fanconi’s anemia, an inherited bone marrow failure syndrome
associated with predisposition to MDS and transformation to AML105–109.
The conversion of a normal hematopoietic stem cell to MDS (preleukemia) and ultimate progression into AML (leukemia) is a multi-step process involving the clonal accumulation of
somatic mutations (i.e. NRAS, KRAS, FLT3, IDH2, RUNX1, TET2, ASXL1, and TP53),
chromosomal losses, deletions or translocations61,64,110,111. In fact, close to 80% of MDS patients possess at least one somatic mutation27. These somatic mutations act as critical drivers of
leukemogenesis and may induce CIN due to the role some of these genes play in ensuring the proper
chromosome segregation, the correction of kinetochore-microtubule (kMT) attachment and
maintenance of centromere architecture112. However, the underlying mechanism(s) that drives CIN to propel leukemic progression is yet to be fully delineated. HSPCs become genetically unstable in nearly half of MDS patients and undergo clonal evolution14,110,113. However, it is still unknown in
which of the myeloid progenitor cells this event triggering MDS transformation occurs whether in
the pluripotent hematopoietic cell or the secondary hematopoietic stem cell. This clonal evolution
and selection predisposes the affected cell to further mutations, and could ultimately lead to AML
progression. Several important tumor suppressor genes involved in the proliferation and
differentiation of progenitor cells during hematopoiesis may be lost or mutated as a result of these
chromosomal losses, duplications, inversions or translocations65.
21 The molecular basis for the pathogenesis of MDS, from the initial trigger or genetic event
to the development of AML remains elusive although DNA damage repair (DDR) defects and CIN
have been broadly suggested. Moreover, disease progression from MDS to AML is closely linked to CIN. In fact, CIN has been proposed as a better prognostic marker of leukemia transformation compared to the conventional approach of risk assessment114. Thus, CIN could be used as a
prognostic marker for the early identification and treatment of MDS patients with high-risk of
progression to AML114. This opens an opportunity to explore the importance DDR and CIN-
associated genes and their role in the progression of MDS to AML. Genomic instability gene
signature panels including the CIN25, CIN70 signature and another 12-gene signature have been
used to predict cancer recurrence and risk-stratification in MDS patients115,116.
Aside from transcription factors, one of the main signaling pathways dysregulated in the multi-step leukemic progression of MDS is the mitotic checkpoint117. Extensive serial studies to investigate the correlation between CIN and leukemic progression are limited due to the lack of data on well-defined cohorts at the various stages of leukemia progression. Moreover, such a study involving a wide gene panel would be essential to link molecular alterations with pathogenesis and provide insight on the mutational architecture and dynamics involved in MDS progression to AML and could ultimately inform biomarker discovery (for more efficient patient risk assessment) and targeted therapy. Recent studies revealed that certain driver mutations in genes including NRAS,
WT1, FLT3, IDH2, and TP53 were implicated in MDS transformation into secondary AML with very poor therapeutic outcomes111,118–120. It is of utmost importance to conduct functional studies to validate these mutations to understand how to therapeutically target the affected signaling pathways to improve patient outcome ultimately. Although several preclinical MDS mouse models have been developed, most of them have failed to recapitulate human disease during MDS to AML progression121,122. However, a recent study using the NHD13 (i.e. NUP98-HOXD13) MDS mouse model indicated that GPI-anchor protein loss and the level of genomic instability are correlated
22 with leukemia progression123–125. Similarly, elevated CIN in MDS patients is associated with poor
outcomes regardless of cytogenetic subtype114. It is well established that tumor suppressors have
multiple regulatory roles encompassing cell cycle regulation, growth and differentiation and the suppression of genomic instability126. Dysregulation of these tumor suppressors or oncogenic
signaling pathways not only affect cell cycle regulation but also disrupts mitotic division and
ultimately results in CIN. Thus, targeting CIN may in a way rob the cells of the very spindle
assembly checkpoint (SAC) machinery designed to quell it unless the leukemic cells are
specifically targeted without any effects on the non-leukemic cells. Also, further research is
required to identify CIN drivers and suppressors and elucidate the mechanisms by which they
influence CIN, particularly during leukemic progression.
CIN as a marker of MDS leukemic transformation
This portion of the literature review explores the pathological and molecular evidence that
reveal the need for CIN assessment in MDS patients. Moreover, the challenges associated with the
current system of MDS risk-stratification will be highlighted, and the proposal of CIN status as a
better alternative for MDS risk-stratification and prediction of treatment response will be presented.
Copy number variations have been associated with the development of CIN in blood cancers127,128.
However, the presence of CIN in MDS continues to be a reliable prognostic factor. Thus, the role
that CIN plays in driving leukemic transformation CIN status could be exploited for MDS risk-
stratification in the clinical setting. Multiple studies have attempted to use CIN scoring to
prognosticate disease resurgence and response to therapy and in so doing have enhanced our
understanding of the genetic and molecular basis of MDS progression114,129. Unfortunately, these studies employed different techniques for CIN detection and measurement. Some of these studies have indirectly used aneuploidy status as a surrogate for CIN status while others have used more
23 direct measurements of CIN114,129–131. Regardless of which technique is used, there is more room for confounding variables that could mask very small associations between CIN and prognosis. The subjective nature of morphologic assessments in the risk assessment of MDS transformation poses a direct problem with regards to the concurrence of clinical diagnosis and therapy. However, the main impediment to the utilization of CIN assessment for MDS risk-stratification is our limited understanding of the mechanistic basis of this form of genomic instability and the lack of prospectively validated methods for its measurement in the clinical setting. Nonetheless, a recent study indicated that nCIN could better and earlier detect leukemic transformation than the established methods of prognostication and regardless of cytogenetic risk114. Moreover, unbalanced translocations, which occur in about 50% of MDS cases and complex karyotypes (i.e. at least three chromosomal aberrations), are correlated with poor patient outcomes27,132.
The relationship between CIN, prognostication of leukemic progression and drug
resistance is very complicated. Moreover, the paradox that extreme CIN is associated with better patient outcomes countered with the lack of clinically defined thresholds for CIN estimation is another hurdle to the implementation of clinical CIN assessment for prognostication87,95,117.
Regardless, stratification of treatment response based on CIN status would aid in the identification
of therapeutic agents with improved sensitivity and facilitate better clinical risk assessment of MDS
patients. Thus, the urgent need for methods and assays for the measurement of CIN cannot be
overemphasized. Such a method would have to consider the dynamic nature of CIN and involve
the direct measurement of numerical and structural chromosomal defects across multiple clones of
myeloid progenitor cell populations. Another approach would be to determine the frequency of
anaphase segregation errors within these myeloid cell populations similar to what was earlier done
in diffuse B-cell lymphoma cells133. The challenge, however, would be due to the low proliferation index and morphology of CD34+ HSPCs isolated from the blood or bone marrow of these MDS patients. In the interim, fluorescent in-situ hybridization (FISH) which uses fluorescently tagged
24 DNA probes that mark specific chromosomes, has been employed clinically for the detection of
CIN status in patients. However, FISH is expensive and very labor intensive which limits its
development into an automated high throughput system. To overcome the cost and high-
performance hurdle, flow cytometry for the measurement of DNA content or ploidy has been used
for CIN measurement in colorectal and pancreatic cancers134. A more sensitive approach for
determining sCIN and nCIN at the single-cell level is comparative genomic hybridization (CGH)
but this is also expensive, and the development of this method as a high throughput clinical
diagnostic assay will be very challenging95,135,136. Nonetheless, CGH could be used to determine
copy number alterations and as such the chromosomal complexity of MDS patient myeloid
progenitor cells61,137. Recently, genome-wide copy number alterations and LOH as measures of
CIN can be determined with a better resolution by employing high-density, single-nucleotide
polymorphism arrays (SNP-A)138.
Therapeutic targeting of CIN
Even though the sufficiency of the therapeutic targeting of CIN is yet to be established, the
direct targeting of CIN has become an intriguing area of research. As earlier discussed, the diverse
ways in which CIN can be induced suggests the potential for the development of several forms of
CIN-targeted therapeutics. In fact, the Karyotypic complexity of the NCI-60 drug panel has been
characterized as part of the quest to discover CIN-targeted anti-cancer agents95,139,140. Thus, some potential CIN-targeted compounds have been identified and need to be further characterized. The detrimental impact of extreme CIN to cancer progression could be leveraged as a therapeutic strategy. For instance, paclitaxel could be used to selectively kill cancer cells with downregulated levels of the SAC regulators MPS1 and BUBR1141,142. Interestingly, the antimalarial drug, chloroquine was shown to target cells with aneuploidy selectively 143. Another potential target
25 implicated in CIN induction is HSP90, and its inhibition may similarly trigger excessive
chromosomal missegregation and ultimately result in the selective death of those cells with pre-
existing levels of CIN144. More research effort should be channeled towards studying the possibility
of long-term tolerance and disease resurgence post-CIN-targeted therapy.
The spindle assembly checkpoint
Genome stability is maintained by two main control points: DDR and the SAC. The SAC
is involved in the regulation of metaphase-anaphase transition and mitotic exit by ensuring proper microtubule attachment to the kinetochore at the centromere145. The core components of the SAC include MAD1, MAD2, BUBR1, and BUB3. MAD1 monitors spindle tension between the microtubule and kinetochore (k-MT). The SAC is activated when problems are sensed due to improper k-MT attachment or unbalanced tension between the opposing spindle poles during the metaphase-anaphase transition. Thus, anaphase cannot proceed until proper k-MT attachment is achieved. Because of MAD1 activation, MAD2 undergoes conformational changes via its interaction with MAD1 at the kinetochore. This leads to MAD2 binding to BUBR1 and BUB3 to form the mitotic checkpoint complex (MCC). The MCC then binds to and sequesters CDC20, thus preventing CDC20 from interacting with and activating the anaphase-promoting complex/cyclosome (APC/C) (Figure 1-5).
26
Figure 1-5 The SAC signaling pathway.
Model of SAC signaling during activation (Right) and proper mitotic segregation (Left). The SAC
serves as a surveillance mechanism for ensuring proper separation and is triggered by MAD1
activation leading to the formation of the MCC and ultimately APC/C inactivation.
27 In the case of proper k-MT attachment, the E3 ubiquitin ligase, APC/C upon activation
by CDC20 polyubiquitinates and degrades securin which is an inhibitor of the protease, separase.
Separase is then free to cleave cohesin which is responsible for maintaining cohesion between the sister chromatids146,147. Thus, the SAC can inhibit APC/C and anaphase progression until proper k-MT attachment is achieved.
Cross-talk between DDR and CIN induction
Although debatable, a recent study has revealed that CIN may be linked to pre-mitotic as
opposed to mitotic defects, suggesting the role of DNA damage via replication stress93,148.
Nonetheless, the DNA damage checkpoint is responsible for the maintenance of genome stability
during replication stress via DDR but is defective in more than half of cancers and is active for the entirety of the cell cycle except during mitotic cell division149–152. However, the level of cross-talk
between DDR and maintenance of mitotic fidelity via the SAC signaling remains unclear (Figure
1-6). In fact, TP53 has been suggested as a link between these two seemingly different checkpoint signaling pathways due to its role in conferring tolerance to CIN via the induction of DDR post- chromosomal mis-segregation153,154. TP53 does this by increasing the expression of some SAC-
related mediators in response to DNA damage155.
28
Figure 1-6 Schematic of the TP53-mediated cross-talk between SAC and DDR.
Image modified from156. The broken arrow represents an indirect link while the solid arrows indicate direct links.
29 Other proteins involved in the DDR including Chk1 have also been implicated due to their role in centrosome replication, the correction of kMT attachment errors via regulation of SAC components MAD2 and BUBR1157–161. Chk2 loss has also been shown to delay mitotic exit and ultimately induce CIN in a TP53-independent manner162,163. Interestingly, proteins formerly
associated with DNA repair, specifically, non-homologous end joining have been reported to localize at the kinetochores of chromosomes to ensure mitotic fidelity although their mechanistic role at this site remains unclear164. Thus, there is adequate evidence to prove the link between the
DNA damage checkpoint pathway and the SAC in the correction of kMT attachment errors, which ultimately result in CIN. On the other hand, in anaphase, chromosome missegregation may lead to
DNA double strand breaks during cytokinesis when lagging chromosomes become trapped in the cytokinetic furrow. Moreover, these lagging chromosomes may subsequently (i.e. G1 phase) form micronuclei that are incapable of replication165–167. DNA damage and double strand breaks have
been linked to mitotic chromosomal missegregation and subsequently during interphase165,167.
Moreover, aberrant DDR signaling pathways have been reported to lead to the formation of
chromosomal breaks168,169.
SAC/mitotic checkpoint dysregulation in MDS/AML progression
The SAC is frequently dysregulated or weakened in MDS/AML cells with CIN117.
Amplified expression of the SAC-related components, AURKA, AURKB, CDC20, and MAD2L1,
in MDS patients, has been linked to abnormal karyotype, poor prognosis and high-risk of
MDS/AML progression117,170,171. Interestingly, gene expression profiling had earlier been proposed as part of risk assessment in the prognostic evaluation of MDS172. Some of these SAC proteins
including MAD1 are located at the nuclear pore complex during interphase but relocate to the
30 kinetochore during mitotic checkpoint activation, and this may suggest a pre-mitotic
activation173,174.
Phosphatidylinositol Glycan Class N (PIGN)
PIGN was initially identified as a mannose phosphoethanolamine (EtNP) transferase in the
penultimate steps of the GPI-anchored protein (GP-AP) biosynthesis pathway (Figure 1-7)175. The
PIGN gene is located at the 18q21.33 locus and has 22 transcript variants, 2 of which code the full- length protein176. Interestingly, about 11-40% of AML subjects with complex karyotypes may have
18q losses, and this suggests the involvement of a tumor suppressor gene within that region that
plays a critical molecular role in preventing leukemia progression and MDS patients may harbor undetected chromosome deletions that may cause inactivation of tumor suppressor genes177.
31
Figure 1-7 PIGN transfers the first EtNP residue to a mannose residue in the multi-step GPI-AP biosynthesis and remodeling pathway.
Image adapted from Fujita and Kinoshita, 2012178.
32 PIGN and GPI-AP biosynthesis
PIGN is one of twenty-six enzymes involved in the multi-step GPI-AP biosynthesis and remodeling pathway179. GPI-APs are important for cell signaling, immunity (i.e. complement regulation, host-pathogen interaction, T cell development and antigen activation), cell proliferation and cell cycle regulation180–183. These proteins are anchored on glycolipids composed of phosphatidylinositol, mannose, carbohydrate residues and glucosamine184. The initial steps in the
GPI-anchor biosynthesis pathway involves the PIGA, PIGC, PIGH, GPI1, PIGY, PIGP and DPM2-
mediated transfer of N-acetyl glucosamine (GlcNAc) to phosphatidylinositol (PI)185. This is followed by the deacetylation of the GlcNAc-PI product by PIGL186. Subsequently, the inositol
group of the deacetylated product, GlcN-PI undergoes acylation followed by stepwise addition of mannose and EtNP at the endoplasmic reticulum187. PIGN is primarily responsible for the addition
of the first EtNP to the anchor188. The preassembled GPI-anchors are covalently linked to the nascent protein via an amide bond between the carboxyl terminus of the GPI-attachment signal peptide of the protein and an amino group of EtNP on the anchor187. This allows GPI-APs to function as membrane-associated enzymes, receptors, adhesion molecules, differentiation markers, and protease inhibitors189. GPI-APs have been shown to play a vital role in normal embryonic and
congenital neuronal development184. Moreover, GPI-AP loss on hematopoietic stem cells results in complement-mediated hemolysis of red cells associated with the acquired clonal disease, paroxysmal nocturnal hemoglobinuria (PNH)190.
Unlike PIGA, a partial shift in GPI-AP availability with PIGN loss may be indicative of a significant alternative role of PIGN related to genomic instability188. PIGN is primarily located in
the endoplasmic reticulum membrane where it very recently been shown to be involved in
preventing protein aggregation aside from its role in GPI-AP biosynthesis191. PIGN also has some
plasma membrane and cytoplasmic localization192 (Figure 1-8).
33
Figure 1-8 The intracellular localization pattern of PIGN.
K562 (upper panels) and HEK293 (lower panels) cells. PIGN (green) and nucleus (blue).
34 Diverse germline mutations in PIGN (Table 1-3) are associated with inherited GPI-anchor deficiencies resulting in multiple congenital anomalies, developmental defects and a wide range of intellectual disabilities189,193–201. Moreover, PIGN knockout mice develop features linked to neurological and growth defects202. No studies have yet reported PIGN mutations in cancer patients.
However, somatic mutations in PIGN have been reported in multiple solid tumor cell lines as well as in hematopoietic and lymphoblastoid cell lines203. Moreover, overall survival is significantly
reduced with a median survival of 8.2 months in AML patients with dysregulated PIGN gene expression compared to cases without aberrant gene expression204,205.
35 Table 1-3 Germline mutations in PIGN as associated congenital and developmental conditions.
Modified data from the OMIM database.
Disease Mutations Study Reference
Multiple Congenital Anomalies- c.2125G-A (p.R709Q) Maydan et al., 2011
Hypotonia Seizures Syndrome 1
(MCAHS1) c.808T-C (p.S270P) Nakagawa et al., 2016
c.963G-A(p.A322VfsTer24) Ohba et al., 2014
c.755A-T(p.D252V) Khayat et al., 2016
c.2340T-A(p.Y780X); Fleming et al., 2016
c.548_549+6del (p. L183fs)
c.1434+5G-A (p. IVS17DS)
chr18.59821582_59824939del, GRCh37
c.Ex5-7del
Fryns Syndrome c.649A-T (p.K232X) McInerney-Leo et al., 2016
c.1966C-T (p.E656X)
Syndromic congenital diaphragmatic c.1574+1G>A Brady et al., 2014
hernia Exon17 is skipped leading to frameshift
and truncated protein
36 Role of PIGN in CIN and leukemic progression
The loss of GPI-AP has been indicated as a marker of genomic instability and leukemic
transformation206. Also, GPI-AP deficient cells have been identified in a subset of MDS patients
and been linked to genomic instability and MDS transformation to leukemia in an MDS mouse
model123,124. Interestingly, a recent study identified PIGN as a suppressor of CIN, a form of genomic
instability, in a colorectal cancer model and linked replication stress to CIN148. In that study, PIGN
was identified as one of three suppressors of chromosomal instability (CIN) in colon cancer cells.
Furthermore, PIGN gene silencing and aneuploidy were observed during the process of adenoma-
carcinoma transition148. CIN has been linked with risk of leukemic transformation of MDS and is
associated with poor overall survival in patients114. However, the role of PIGN gene expression
aberration in the transformation and progression of hematological malignancies has not yet been
addressed. Moreover, the partial shift in GPI-AP availability with PIGN loss may be indicative of
the minor role of PIGN in GPI-AP biosynthesis and defines an alternative function for PIGN related
to genomic instability175,188. Regardless, the findings above reveal PIGN as a common thread
linking GPI-AP biosynthesis, CIN induction, and leukemic progression. In an MDS mouse model,
the progressive loss of GPI-AP was associated with genomic instability, leukemic progression and decreased survival, thus implicating defective GPI-AP biosynthesis in MDS leukemia
progression207. Thus, the dysregulation of a GPI-AP biosynthesis enzyme, such as PIGN could
potentially be used to prognosticate MDS disease progression.
This review has highlighted the urgent need for improved MDS risk-stratification for better
prognostication and therapeutic design while advocating for the relevance of CIN in MDS risk
assessment. This review also makes a case for the link between the GPI-AP biosynthesis enzyme
PIGN and CIN and how this link could be leveraged in the development of PIGN as a prognostic
marker of MDS transformation.
37
PIGN gene expression aberration is associated with genomic instability and leukemic progression in acute myeloid leukemia with myelodysplasia-related changes
Emmanuel K. Teye, Abigail Sido, Ping Xin, Niklas K. Finnberg, Prashanth Gokare, Yuka I.
Kawasawa, Anna C. Salzberg, Sara Shimko, Michael Bayer, W. Christopher Ehmann, David F.
Claxton, Witold B. Rybka, Joseph J. Drabick, Hong-Gang Wang, Thomas Abraham, Wafik S. El-
Deiry, Robert A. Brodsky, Raymond J. Hohl, Jeffrey J. Pu
This work has been published and is reformatted here with permission from Oncotarget
Copyright © Teye et al. This is an open-access article distributed under the terms of the Creative
Commons Attribution License (CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Oncotarget, 2017, Vol. 8, (No. 18), pp: 29887-29905
Author Contributions
EKT and JJP designed this study and wrote this manuscript; EKT, JJP, AS, PX, NKF, PG, and SS conducted experiments and performed data analyses; YIK and ACS conducted bioinformatics study and analyses; TA contributed to Confocal imaging study; JJP, MB, WCE, DFC, and WBR provided patient care; EKT, NKF, YIK, ACS, HGW, JJD, WSE, RAB, and RJH participated in manuscript formation by providing comments and suggestions.
A portion of this work has been published as United States Patent and Trademark Office
Application No. 15/434,774 ‘Method and Therapeutic Use of PIGN and Other Genes or Gene
Products That PIGN Interacts With for Prognosis and Treatment of Hematological Neoplasias’.
38 Abstract
Previous studies have linked increased frequency of glycosylphosphatidylinositol-anchor
protein (GPI-AP) deficiency with genomic instability and the risk of carcinogenesis. However, the
role of the GPI-AP biosynthesis protein, phosphatidylinositol glycan anchor biosynthesis class N
(PIGN) in hematological malignant disease progression is yet to be investigated. In this study, the
PIGN gene expression profiles of patients with either myelodysplastic syndrome (MDS) or acute myeloid leukemia with myelodysplasia-related changes (AML-MRC) were examined. Moreover, the correlation between PIGN expression aberration, genomic instability, and leukemia status in both patient samples and cell lines were investigated. A Random Forest analysis of the gene expression array data from 55 MDS patients (GSE4619) demonstrated a significant (p = 0.0007) correlation (Pearson r = -0.4068) between GPI-anchor biosynthesis gene expression and genomic instability, in which PIGN was ranked as the third most important after DPM2 and PIGP in predicting the risk of MDS progression. Furthermore, PIGN expression aberrations (i.e. increased gene expression but diminished to no protein production) were associated with increased frequency of GPI-AP deficiency in leukemic cells during leukemic progression. PIGN expression aberrations were attributed to partial intron retentions between exons 14 and 15 resulting in frameshifts and premature termination which were confirmed by examining the RNA-seq data from a group of
AML patients (phs001027.v1.p1). PIGN expression aberration correlated with the elevation of genomic instability marker expression that was independent of the TP53 regulatory pathway. PIGN suppression or elimination caused a similar pattern of genomic instability that was rescued by PIGN restoration. In conclusion, PIGN expression aberration is associated with genomic instability and leukemogenesis in a subgroup of AML-MRC patients and prevents leukemic transformation/progression.
39
Introduction
Myelodysplastic syndromes (MDS) are a group of clonal hematopoietic stem cell diseases with various presentations and therapeutic outcomes27. MDS affects about 13,000 people annually in the United States alone with about a one-third propensity of progression into AML208. MDS is
conventionally classified as AML with myelodysplastic changes (AML-MRC) when blood or bone marrow blast populations reach or exceed 20% with dysplastic morphology in 50% or more cells in more than two myeloid lineages47,209. AML constitutes 25% of blood cancers with about 18,860 new cases and 10,460 deaths annually210. Traditionally, AML is treated with chemotherapy.
Unfortunately, it has been observed that chemotherapy is not an effective treatment for this disease and often results in secondary cancers, high incidence of relapse and refractory conditions211. We
would like to use historical data to study this disease to discover new treatments options for this
disease. AML is more aggressive and molecularly heterogeneous and involves unconstrained
proliferation of aberrant myeloid progenitor cells212. These aberrant myeloid progenitor cells possess genetic aberrations, populate the bone marrow and peripheral blood, and contribute to leukemia progression by driving clonal evolution14. AML may develop de novo or secondary to the treatment of other cancers with chemotherapy or radiotherapy; or may evolve from other bone marrow failure conditions such as MDS, aplastic anemia, or myeloproliferative neoplasms213.
Genomic instability is frequently associated with cancer initiation and progression and is
responsible for the clonal accumulation of genetic abnormalities that drive the transformation of
MDS to AML207,214–216. In fact, the frequency of cytogenetic aberrations at the initial presentation of MDS is less than 50%, but this rate increases with progression due to loss or gain of large chromosomal segments177,212. However, the precise genetic cause of the leukemic transformation
of MDS into AML remains elusive and presents the need for an understanding of the underlying
mechanisms for the identification of prognostic markers.
40 Previous studies have linked genomic instability with increased frequency of GPI-AP
deficiency124,125,129,130. Interestingly, GPI-AP loss has been proposed as a marker to predict the risk of leukemic transformation125,130,131,206. However, the biomarker and the underlying mechanism that
link GPI-AP to genomic instability and leukemic transformation are yet to be elucidated.
Fortunately, a gene participating in GPI-AP biosynthesis called phosphatidylinositol glycan anchor biosynthesis class N (PIGN) was recently identified as a cancer chromosomal instability (CIN)
suppressor in a colon cancer model148. In that study, Burrell et al. demonstrated that PIGN gene silencing by either chromosomal 18q deletion or transfection of short interfering RNAs (siRNAs) resulted in DNA replication stress, structural chromosome abnormalities, and chromosome missegregation. Furthermore, PIGN silencing and aneuploidy were observed during the process of adenoma-carcinoma transition. PIGN encodes a phosphoethanolamine (EtNP) transferase involved in the final steps of GPI-AP biosynthesis175. Interestingly, CIN, a form of genomic instability has been linked with risk of leukemic transformation and was associated with poor overall survival in
MDS patients114. Germline mutations in PIGN have been implicated in GPI-AP deficiency and are associated with multiple congenital anomalies and developmental defects175,188,193. We also
identified common non-synonymous PIGN mutations from whole genome sequencing data in
AML patients (dbGAP accession: phs000159.v8.p4). However, there is at yet no published work that addresses the role of PIGN gene expression aberration in the transformation or progression of hematological malignancies. Thus, this study investigated the relationship between PIGN gene expression aberration, genomic instability, and leukemia transformation or progression. The goal of this study was to demonstrate for the first time, the vital role that PIGN plays in the maintenance of genomic stability and the prevention of leukemia transformation or progression in a subgroup of patients with MDS or AML-MRC. Thus, we hypothesize that PIGN gene is a CIN suppressor and its aberrant expression causes genomic instability, MDS progression, and leukemia transformation.
Thus, we investigated the PIGN gene expression profiles in patients with either MDS or AML with
41 myelodysplasia-related changes and the correlations among PIGN expression aberration, genomic
instability, and leukemia status in both patient samples and cell lines.
Materials and Methods
Bioinformatics analyses and statistical analyses
The GENE-E (http://www.broadinstitute.org/cancer/ software/GENE-E/) matrix visualization and analysis platform were used to generate a heat map of the CIN70 gene expression profiles of the CD34+ cells of 55 MDS patients and 11 healthy controls utilizing data generated on the Affymetrix GeneChip U133 Plus2.0 platform from the study GSE4619116,217. The Random
Forest v4.6-12 R package with default parameters was used in a Random Forest analysis to classify
patients based on MDS risk-stratification: control, RA, RARS, RAEB1, and RAEB2. The importance of each gene (predictor) was calculated using the package’s Mean Decrease in
Accuracy, which measures how much the model fit decreases when a variable is dropped.
In the RNA-seq analysis study, raw RNA-seq files for the dbGAP study phs001027.v1.p1 were downloaded as sequence raw archive (SRA) files and then converted to FASTQ, using the SRA toolkit version 2.5.4. The RNA-seq reads were aligned to the most recent reference genomes (hg38) using TopHat (v2.0.9) by allowing up to 2 mismatches218. The junction files data analyses were
loaded on to Integrated Genome Viewer (IGV) to survey intronic frameshifts in the PIGN gene.
Patient selection
This study was approved by the Institutional Review Board (IRB) of the Pennsylvania State
University College of Medicine. We recruited patients with MDS or AML bearing myelodysplastic
42 features as per clinical criteria or WHO MDS/AML classification47. A translational research protocol: “Using Human Tissue and Cells for Clonal Bone Marrow Disease Translational
Research” was approved by Penn State College of Medicine IRB (Protocol No.40969). Informed consent was obtained from each patient before obtaining peripheral blood and bone marrow
aspirates by the approved IRB protocol. The peripheral venous blood and bone marrow aspiration
for experiment use were drawn into heparin-containing tubes. All patients in this study were
evaluated at the Milton S Hershey Cancer Institute at the Pennsylvania State University College of
Medicine either at the outpatient clinic or inpatient wards between January 2013 and August 2016.
MDS and AML were diagnosed as per WHO criteria. A Hematopathologist evaluated bone marrow
aspiration and biopsy samples. The data collection has generated data for the study by collecting
and analyzing 48 high-risk MDS/AML-MRC patient samples and 12 healthy controls. The sample sizes calculation determined the number of MDS/AML-MRC patients and healthy samples needed for statistical significance based on a t-test. The power analysis calculations were performed using
G*Power 3.1.9.2219.
Isolation of mononuclear and CD34+ cells
Mononuclear cells were isolated from peripheral blood or bone marrow aspirates using
Ficoll-Paque PLUS (GE Healthcare Life Sciences) according to the manufacturer’s protocol and as earlier described124. CD34+ cells were sorted from peripheral blood or bone marrow mononuclear
cells via magnetic-activated cell sorting (MACS) ® Technology (Miltenyi Biotec) in conjunction with the human anti-CD34 MicroBead Kit (Miltenyi Biotec) by the manufacturer’s protocol.
43 Leukemic blast cell sorting
Leukemic and non-leukemic cells were sorted via flow cytometry. Leukemic blast cell sorting was conducted under BSL-2 conditions with a 16-color BD FACSAria SORP high-speed
cell sorter (Becton Dickinson) in the Institutional Flow Cytometry Core Facility. CD34+ cells that
expressed HLA-DR, CD13, CD117, and CD45, were sorted as leukemic cells as previously described with few modifications220. CD13 was included in the panel as a myeloid cell marker.
CD34+ HSPCs that did not express HLA-DR, CD13, CD117, and CD45 were sorted as non- leukemic cells.
Selection of proaerolysin-resistant colony forming cell (CFC)s and GPI-AP deficiency frequency analysis
Proaerolysin-resistance CFC assay is a very efficient and widely accepted method to assess
the extent of GPI-AP deficiency. The collection of proaerolysin-resistant CFCs was conducted as
previously described with some modifications124. Bone marrow CD34+ cells were sorted via
MACS for the growth of CFCs. The bone marrow CD34+ cells were cultured in 1ml HSC-CFU
complete w/o Epo (Miltenyi Biotec) with or without 1 nM of proaerolysin. The selection of
proaerolysin-resistant CFCs was conducted as earlier described124. The frequency of GPI-AP
deficiency was calculated as previously described124,130. GPI-AP proteins were detected on the
FACS Diva version 6.1.1 with the following tagged antibodies CD59-FITC (Life Technologies),
FLAER-AlexaFuor-488 (Cedarlane Labs) and CD45-APC-H7 (BD Biosciences). Flow cytometric results were analyzed using FlowJo software (FlowJo LLC).
44 Gene expression analyses
Total RNA was isolated from 1x106 cells using the RNeasy Mini Kit (Qiagen) according
to the manufacturer’s instructions and quantified using the NanoDrop 1000 (Thermo Scientific).
Total RNA (20 ng/µl) was reverse transcribed using the High Capacity cDNA reverse transcription
kit (Life Technologies) on the Mastercycler® Nexus (Eppendorf). The quantitative real-time
(qPCR) step was conducted on the StepOnePlus™ real-time PCR System (ABI systems) using validated PIGN gene expression assay Hs00202443_m1 (Life Technologies) and 18S rRNA
Hs99999901_s1 (Life Technologies) gene expression as an internal reference control. The PIGN gene expression assay was designed to have the best coverage for the reported transcript variants of PIGN including short transcripts and those that may undergo nonsense-mediated decay. The RT- qPCR assay employed targets the exon boundary between exons 5 and exon 6. H2AX, Baxα, DR5,
SIRT1, SAE2 and p21 primers as well as GAPDH and HPRT internal control primers (Integrated
DNA Technologies) were used with Power SYBR® Green PCR Master Mix (Life Technologies).
Please refer to Appendix C for primer sequences.
The selected gene expression profile in the leukemic phase and non-leukemic phase were conducted via reverse transcription–qPCR (RT-qPCR) as earlier described221. RNA was reverse
transcribed as earlier described and RT-qPCR experiments were performed using PrimeTime®
qPCR Assays (Integrated DNA Technologies). For all experiments, gene expression was calculated
using the ΔΔCt method. The reference gene and the expression of the genes under consideration
were normalized to the corresponding transcripts in mononuclear cells obtained from a non-
leukemic healthy subject.
45 Cell lines and cell culture
KG1, KG1a and K562 cells (ATCC) were cultured in IMDM supplemented with 20% FBS.
MDS92 and MDS-L cells were grown in RPMI-1640 supplemented with 10% FBS, 50ng/ml IL-3
(Miltenyi Biotec) and 50μM 2-mercaptoethanol (Fisher Scientific). HEK293 and HEK293 PIGN
CRISPR/Cas9 Knockout cells were cultured in DMEM supplemented with 10% FBS. Cell lines were passaged on average every 3-4 days. CD34+ mononuclear cells were grown in DMEM/F12 supplemented with 10% FBS, 50μM 2-mercaptoethanol, Glutamax (Life Technologies), MEM non-essential amino acids (Life Technologies) and StemMACS HSC expansion cocktail (Miltenyi
Biotec). All cells were cultured at 37°C under 5% CO2 conditions.
HEK293 PIGN CRISPR/Cas9 stable knockout cells (HEK293 PIGN KO) were a gift from Drs.
Yoshiko Murakami and Taroh Kinoshita (Osaka University, Japan). Drs Sherri Rennoll and
Gregory Yochum (Pennsylvania State University College of Medicine, USA) kindly provided
HEK293 cells. K562 cells and HL60 cells were a gift from Drs. Su-Fern Tan and Thomas P.
Loughran, Jr (University of Virginia Cancer Center, USA) and were cultured in IMDM 20% FBS.
MDS92 and MDS-L cells were a gift from Dr. Kaoru Tohyama, Department of Laboratory
Medicine, Kawasaki Medical School Kurashiki, Okayama, Japan. HEK293 CRISPR KO cells and pMEPURO3HAhPIGN expression plasmid were the gifts from Drs. Taroh Kinoshita and Yoshiko
Murakami, Research Institute for Microbial Diseases, Osaka University.
Plasmids and transfection
HEK293 PIGN KO cells were transfected with pMEPURO3HAhPIGN or pMEPURO3HA
using the Lipofectamine3000 transfection reagent (Invitrogen) per the manufacturer's protocol for
24-72 hours. The pMEPURO3HAhPIGN and pMEPURO3HA expression plasmids were gifts from
46 Drs. Taroh Kinoshita and Yoshiko Murakami, Research Institute for Microbial Diseases, Osaka
University.
PIGN knock down and CRISPR/Cas9 Knockout studies
RNA interference-mediated PIGN knockdown experiments were conducted using the
Nucleofector ™II Device (Amaxa) in conjunction with the Cell line Nucleofector® Kit V reagent
kit (Amaxa) according to the manufacturer’s recommended protocols for the various cell lines. The
cells were transfected with 100 nM siGENOME™ siRNA Human PIGN [D-012463-01
(Dharmacon)] or 100 nM siGENOME™ Control siRNA Non-targeting siRNA #2 [D-001210-02-
05 (Dharmacon)] and incubated for 24-72 hours.
CRISPR/Cas9 experiments were conducted according to a modified LentiCRISPRv2
(Addgene plasmid #49535) protocol222. The gRNA (AAACGGTCATGTAGCTCTGATAGC)
employed targets PIGN at exon 4 and results in a frameshift193. The Lentiviral-transduced CD34+ mononuclear cells were harvested for downstream applications nine days’ post-infection.
Western blot analyses
Cells were lysed in RIPA lysis buffer (Sigma-Aldrich) with protease and phosphatase inhibitor cocktails (Sigma). Protein concentration was determined using the Coomassie protein assay reagent (Thermo Scientific). Protein (15-45 µg) was separated on a NuPAGE™ 4-12% Bis-
Tris Gel (Life Technologies) or 10% NuPAGE Bis-Tris gels and transferred to Immun-Blot® PVDF
Membrane (Bio-Rad). The blots were incubated overnight with a rabbit polyclonal antibody to
PIGN (HPA039922, Atlas Antibodies AB), anti-Histone H2A.X (D17A3, Cell Signaling
Technology) anti-phospho-Histone H2A.X ser139 (S139, Cell Signaling Technology) and HA-tag
47 (C29F4, Cell Signaling Technology) in TBS-T with 5% Non-Fat Dry Milk. The anti-PIGN
antibody (HPA039922, Atlas) that was screened and selected for Western blot analyses binds to a
protein sequence encoded close to the N-terminus between exons 5-7. Thus, the anti-PIGN antibody
binds to a sequence encoded by this region in the two full-length transcript variants and 7 of the shorter transcript variants. The blots were subsequently probed with a horseradish peroxidase- conjugated goat anti-Rabbit IgG Antibody (AP132P, Millipore) or goat anti-mouse IgG antibody
(AP124P, Millipore) followed by image development by ECL Prime (Amersham). Mouse beta- actin (C-4, sc-47778, Santa Cruz Biotechnology) was used as loading control. Relative intensity was determined using the NIH ImageJ software223.
PIGN sequence analysis
RT-PCR was conducted with Invitrogen SuperScript One-step RT-PCR with Platinum Taq kit (Invitrogen) using total RNA as the template. PCR products were sub-cloned into PCR2.1
TOPO vector (Invitrogen) and individual clones were selected on Amp +X-gal blue/white selection plates. The plasmids were isolated using the GeneJET Plasmid Miniprep Kit (Thermo Scientific).
M13 universal primers (Invitrogen) were used for screening the PCR insert size, and sequencing was conducted at MCLAB Molecular Cloning Laboratories, South San Francisco, CA. Sequencing data were analyzed with reference to PIGN NCBI Reference Sequences: NG_033144.1and
NM_012327.5 using the CLC Sequence Viewer Version 7.6 (QIAGEN Aarhus A/S) in conjunction with manual inspection. The impact of aberrations on protein coding was conducted using the
Sequence Manipulation Suite and Polyphen-2 with reference to PIGN protein sequence
NP_036459.1224,225.
48 TP53 sequence analyses
One microgram of DNA isolated from non-leukemic cells and leukemic cells was amplified using primers covering exon 2-11 of the TP53 gene including intron/exon boundaries per instructions in the IARC database (http://p53.iarc.fr)226. Seven PCR reactions per sample (non-
leukemic and leukemic) were spin column purified and sequenced using the 3130XL Genetic
Analyzer (ABI systems) with the same primers in both reverse and forward directions. The obtained
sequence was analyzed using the software FinchTV version 1.4.0 and nucleotide BLAST
(http://blast.ncbi.nlm.nih.gov/BLAST)227. Sequence alterations were aligned to the TP53 GenBank sequence NC_000017.9, and sequence alterations were identified partially by manual inspection.
Sequence alterations were aligned with the coding sequence of the TP53 protein, and the impact on protein was determined using the IARC database (R18, April 2016)
(http://p53.iarc.fr/p53Sequence.aspx) and SIFT (Sorting Intolerant From Tolerant)226,228.
Statistical analyses
Statistical calculations were conducted using GraphPad5 (GraphPad Software Inc.) and
Microsoft Excel 2010. Two-tailed independent Student’s t-tests and one-way ANOVA followed by
Tukey’s post hoc tests were used for group comparisons. For all analyses, p-values <0.05 were
considered statistically significant.
49 Results
PIGN gene expression profile links to genomic stability, especially MDS progression risk-
stratification
Presently, there are scarcely any longitudinal studies that have examined the gene as well
as protein expression changes during the transformation of MDS into AML. For this reason, we
examined the differential expression of PIGN in studies that have considered gene expression status
in MDS/AML patients as well as between various MDS classes including RA, Refractory Anemia
with Ring Sideroblasts (RARS) and Refractory Anemia with Excess Blasts (RAEB). Array data
generated from 55 MDS patients and 11 healthy controls (GSE4619) were initially analyzed217. The
patients were sub-classified as follows: RA (18 patients), RARS (19 patients), RAEB1 and REAB2
(18 patients). Overall, CIN70 genes were expressed in a MDS disease subtype-dependent manner
with a relatively lower expression in high-risk disease subtypes (REAB1 and RAEB2) compared
to the low-risk subtypes (RA and RARS) and normal controls116. This gene expression heat map
showed that the expression of the CIN70 gene panel was associated with MDS risk-stratification
(Figure 2-1A). A Random Forest analysis further demonstrated a significant (p = 0.0007) correlation (Pearson r =-0.4068) between the GPI-anchor biosynthesis gene panel and the CIN70 genomic instability marker panel (Figure 2-1B). Furthermore, the Mean Decrease in Accuracy
identified PIGN as highly relevant (i.e. 3rd ranked) among the GPI-AP biosynthesis genes in
predicting MDS progression risk (Figure 2-1C).
50
51
Figure 2-1 PIGN is a predictive biomarker for MDS risk-stratification.
A. Gene expression heat map showing expression of the CIN70 signature was associated with MDS risk-stratification in CD34+ cells isolated from bone marrow samples of 55 MDS patients and 11 healthy controls (GSE4619)116,217. B. 2D scatter plot showed a significantly (p = 0.0007) negative
correlation (Pearson r =-0.4068) between the GPI-anchor biosynthesis gene panel and the CIN70
signature by plotting the first principal component (PC1) of each per gene panel. C. PIGN was
ranked third among GPI-AP biosynthesis genes in predicting MDS risk-stratification based on a
Random Forest classifier using Mean Decrease in Accuracy as a predictor.
52 PIGN gene expression aberrations occur in a subgroup of patients with MDS or AML-MRC
RT-qPCR was used to determine the PIGN gene expression profiles of CD34+ mononuclear cells harvested from the peripheral blood or bone marrow aspirates of 48 patient samples (Table 2-1) with either high-risk MDS or AML-MRC and 12 healthy volunteers.
53 Table 2-1 PIGN gene and protein expression status in MDS or AML-MRC patients
ID Age/Sex TP53 PIGN Protein a PIGN Gene Fold Karyotype Deletion Expression Expression (Normal/Complex) M1 60/F + - 3.787 Complex
M2 27/F + - 7.653 Complex
M3 87/M - + 1.187 Complex
M4* 59/F - + 3.927 Complex
M5* 59/F - - 2.703 Complex
M6 61/M - - 4.639 Complex
M7 78/M - + 1.636 Complex
M8 64/M - - 0.398 Complex
M9 29/F - + N.D Complex
M10 29/F - + 1.435 Complex
M11 68/M - + 2.002 Complex
M12 61/F - + 3.228 Normal
M13 63/F + + 0.873 Complex
M14 55/F - N.D 2.323 Normal
M15 67/F - - 5.158 complex M16 67/F - - 15.633 complex
M17 73/F - N.D 1.150 Complex
M18 27/M - + 2.513 Complex
M19 48/F - - 7.756 Complex
M20 66/F - N.D 3.045 Complex
M21 72/F + - 6.974 Complex
M22 45/F - + 5.737 Normal
M23 46/M - - 3.857 Complex
M24 59/M - - 10.461 Complex
M25 47/F - N.D 6.246 Complex
M26 27/F + - 5.227 Complex
M27 37/F - N.D 18.311 Complex
M28 56/F + - 9.260 Complex
M29 84/F - N.D 3.068 Complex
M30 61/M - + 3.328 Normal
M31 74/M + N.D 5.849 Complex
54
M32 61/M - + 2.354 Complex
M33 74/M - N.D 16.343 Complex
M34 74/M - N.D 3.031 Complex
M35 65/F - N.D 0.808 Normal
M36 65/F - + 0.959 Normal
M37 81/F + + 1.490 Complex
M38 75/F - + 3.671 Normal
M39 85/M - N.D 3.026 Normal
M40 71/M - + 3.021 Complex
M41 77/M - + 0.962 Complex
M42 62/M - - 7.275 Complex
M43 49/F - - 9.842 Complex
M44 51/F - + 3.676 Normal
M45 51/F - N.D 1.096 Normal
M46 58/M + + 2.825 Complex
M47 68/F + + 1.113 Complex
M48 59/M - N.D 8.061 Complex a Mean fold difference in gene expression in patients compared to PIGN gene expression in normal healthy control PBMCs. +: detected -: not detected N.D: no data available Mono: mononuclear cells *: M4 and M5 from the same patient; M4 at pre-treatment phase and M5 at relapse phase.
55 The results revealed that the majority (~60%) of these patients had a significantly
(p<0.0001) higher expression of the PIGN gene in comparison with the cells from healthy normal
controls (Figure 2-2A). Moreover, 15 of 35 patient samples examined for both PIGN transcription
and translation had an unusual expression pattern (i.e. increased transcriptional activity but
diminished to no protein production) (Table 2-1 and Figure 2-2B). Overall, these data indicated
that a subgroup of patients with high-risk MDS or AML-MRC appeared to have PIGN expression aberration with increased gene expression but diminished protein production.
56
Figure 2-2 PIGN gene expression aberration was due to truncation.
57 A. RT-qPCR data on patient samples showed that a subgroup of MDS/AML-MRC patients had a
significant difference (***p<0.0001) in PIGN gene expression than healthy controls. B. In that
same subpopulation of patients, their PIGN protein expression was lost or suppressed. C. Sequence analyses of CD34+ cells revealed the presence of intron fragment retentions resulting from splice defects between exons 14 and 15 caused frameshifts and premature termination; samples M1, M2 and M4 were from AML patients; samples 1-11 represented the results of RNA-seq junction file data analyses from AML patients in the dbGAP study phs001027.v1.p1. Intron base positions (bp) were based on NCBI reference sequence NG_033144.1. *partial intron retention variant identified in another clone in patient M2.
58 PIGN gene expression aberrations were caused by novel intronic retention mutation between
exons 14 and 15
The reason for this PIGN gene expression aberration was further explored by cloning and
sequencing the PIGN transcripts from 3 patient samples (M1, M2, and M4) and a cell line (MDS-
L) that had significantly high PIGN gene expression but no protein expression. To examine whether
PIGN is progressively lost during leukemic transformation, we employed a cell line model of MDS
transformation to AML. This model involves two cell lines (i.e. MDS92 and MDS-L) derived from
a single patient but with distinct phenotypes that represent the MDS phase and AML stage of
leukemic progression respectively229,230. We examined gene and protein expressions of PIGN in these two cell lines. We observed that an MDS phase cell line (MDS92) and its leukemic phase- derived cell line (MDS-L) have different PIGN mutation and gene expression profiles. The MDS-
L cell line lost its capability to express PIGN protein expression compared to MDS92 cells (Figure
2-4 C). Moreover, our results revealed the retention of aberrant short intronic fragments (i.e. 11bp
to 142bp) between exons 14 and 15 (Figure 2-2 C; 1-4). The predicted product of this mutation is
a truncated protein around ~46 kDa which is less than half of the standard protein size (i.e. ~106 kDa). Interestingly, we identified similar variants of this mutation in 11 AML patients from junction files generated from the RNA-seq data of 19 AML patients (dbGaP Study Accession: phs001027. v1.p1) (Figure 2-2 C; 1-11)218. We have identified a similar mutation in an MDS-
derived cell line (i.e. MDS-L) that represents an AML transformed state but interestingly this
mutation was not detected in the corresponding MDS stage cell line (i.e. MDS92). Further
examination at the resolution of individual bases of these aberrant transcripts revealed that these
intron fragments were similar to those initially identified in the patients with PIGN gene expression
aberrations. The endogenous form of this predicted truncated protein was not detected in the
patients we examined likely due to proteasome degradation of the truncated protein.
59 The novel intronic retention mutations are present in leukemic cells but not in non-leukemic
cells and are associated with a relatively high frequency of GPI-AP deficiency
To understand this unique PIGN expression aberration pattern, we conducted additional
experiments on two AML patients (M1 and M2). Using RT-qPCR, we examined PIGN gene
expression in sorted leukemic cells from these 2 AML patients. Both patients contained TP53 gene
deletion mutations (Table 2-2). PIGN gene expression in the leukemic cells from these two patients was at least 3-7-fold higher than in the non-leukemic (NL) cells, but PIGN protein expression was not detectable in those leukemic cells (Figure 2-3 A and B). We sub-cloned and sequenced PIGN transcripts from the sorted leukemic cells and NL cells. Interestingly, we observed the retention of segments (38 bp and 142 bp) of the intervening intron between exons 14 and 15 in the leukemic cells which resulted in frameshifts and led to the occurrence of premature termination codons
(PTCs) (Figure 2-2 C); but not in the NL cells. To understand the nature of PIGN gene expression aberration, we sequenced the coding region of PIGN gene in both leukemic and NL cells. We
identified the retention of segments (38 bp and 142 bp) of the intervening intron between exons 14
and 15 in leukemic cells, but not NL cells. This intron fragment retention resulted in a frameshift
and led to the occurrence of PTCs. The presence of these PTCs in the truncated forms of PIGN
cause early termination of translation or may trigger nonsense-mediated decay that results in
mRNA destruction before it is translated into protein despite a compensatory upregulation of gene
transcriptional expression231.
Elevated frequency of GPI-AP deficiency has been linked to genomic instability and
leukemic progression206. Our laboratory uses a stringent method of proaerolysin-resistant CFC
assay analysis to determine the frequency of GPI-anchor deficiency in MDS patients. To explore
the genetic stability status of those patients, we conducted proaerolysin-resistant CFC assays on
both sorted leukemic and NL cells from patients M1 and M2. We then calculated the GPI-AP
60 deficiency frequency of the two AML patients as previously described124. The median frequencies
(GPI-AP deficiency frequency) of proaerolysin-resistant leukemic CFC formation for M1 and M2
were 1.20% and 4.71% (ranging from 0.27 to 3.02% and 2.88 to 6.46%) respectively; however, the
median frequencies (GPI-AP deficiency frequency) of proaerolysin-resistant NL CFC formation
were 0.009% and 0.029% (ranging from 0.004% to 0.013% and 0.007% to 0.075%) respectively
(Figure 2-3C). The GPI-AP deficiency frequency in a normal population is approximately
0.002%124. The GPI-AP deficiency frequencies in the leukemic cells were 100 times higher than in the NL cells. Thus, the leukemic cells from both patients bore significantly elevated GPI-AP deficiency frequency, which indicated that those leukemia cells indeed were in a state of genomic instability.
61
Figure 2-3 PIGN expression aberration resulted in an increased frequency of GPI-AP deficiency.
A. RT-qPCR showed that PIGN gene expression in leukemic cells from AML patients M1 and M2 were significantly (***p<0.0001) higher (i.e. 3- to 7-fold) than in normal control cells from a healthy individual (NL). One-way ANOVA Tukey’s posthoc test; error bars represent standard deviation from the mean fold change in gene expression. B. PIGN protein expression was lost in patients M1 and M2. Samples NL, M1 and M2 were from different sections of the same blot. C.
The frequency of GPI-AP loss was much higher in leukemic clones than in the NL clones in the individual AML patients. For detailed calculations of the frequency of GPI-AP deficiency, please review citation124. *Leukemic and NL cells were sorted using the following markers: HLA-DR,
CD13, CD117, and CD45 as described earlier with some modifications220.
62 PIGN gene expression aberrations occur during leukemic transformation and progression
Due to our initial identification of partial intron retentions in the sorted leukemic cells, we
examined PIGN gene and protein expression in relation to disease progression in a refractory AML
patient (Figure 2-4 A and B). That patient had 65% leukemic blasts during the pre-treatment phase
(M4) and 42% leukemic blasts at the relapse phase (M5). We detected an intron fragment retention between exons 14 and 15 in the pre-treatment mononuclear cells of this patient that was like those intron fragment retentions earlier identified in the sorted leukemic cells in M1 and M2 (Figure 2-
2C). However, this intron fragment was not detected in the mononuclear cells collected at the relapse phase. Furthermore, we observed PIGN gene expression aberrations in both phases of disease progression (M4 and M5) in this AML patient, with higher gene expression (~4-fold) in the pre-treatment phase than in the relapse phase (~2.5-fold) compared to normal healthy control cells
(Figure 2-4 A and B), but more suppressed protein expression in the relapse phase.
To examine whether PIGN gene expression aberration occurs during leukemic transformation, we employed a cell line model of MDS transformation to AML. This model involves two cell lines MDS92 and its blastic sub-line MDS-L generated from a single patient but with distinct phenotypes representative of the MDS phase and the AML phase of leukemic progression respectively229. We examined PIGN gene and protein expression in these two cell lines.
PIGN protein expression was relatively higher in MDS92 cell line but was not detected in MDS-L cell line (Figure 2-4C). Moreover, we observed a relatively high PIGN gene expression in MDS92 cells (~5.1-fold) and MDS-L cells (~2.2-fold) compared to healthy NL mononuclear cells (Figure
2-4D).
63
Figure 2-4 PIGN expression aberration was a marker of leukemic transformation and progression.
A. Progressively loss of PIGN in an AML patient (M4 and M5) B. PIGN gene expression was significantly (***p<0.0001) downregulated from pre-treatment (M4) to relapse (M5). C. PIGN protein progressively lost from the MDS phase (MDS92 cells) to the leukemic phase (MDS-L cells) and D. PIGN gene expression was significantly (***p<0.0001) higher in MDS92 cells than in
MDS-L cells. E. PIGN protein expression was more suppressed in the myeloblastic phase (KG1) comparing to its myeloid derivative (KG1a), F. No significant (NS) difference in gene expression observed between the KG1a and KG1cell lines. PIGN gene transcriptions in all the samples were elevated 2- to 5- fold in comparison with PIGN gene expression in CD34+ cells from healthy individuals. Error bars represent standard deviation from the mean fold change in gene expression.
G. Model depicting the loss of PIGN protein with disease progression from a less aggressive disease stage to a more aggressive disease stage.
64 Thus, PIGN gene expression aberration was more apparent in the leukemic phase than in the MDS phase. Interestingly, we detected the same intron fragment retention in the leukemic phase
MDS-L cell line like the one we identified in leukemic cells from M2 and M4 (Figure 2-2C). This mutation was however not detected in the MDS92 cells. Thus, PIGN expression aberration occurs during MDS leukemic transformation and progression and is marked by the presence of partial intron retention mutations between exons 14 and 15, and ultimately the progressive loss of PIGN protein expression. PIGN is progressively lost during leukemic transformation. We also observed a similar PIGN expression aberration pattern in one (KG1) of two leukemia cell lines (KG1 and
KG1a) originated from a single patient, KG1 harboring a myeloblast phenotype but KG1a bearing a stem/ progenitor-like phenotype (Figure 2-4 E and F). However, no intron fragment retention was detected in either cell line while PIGN gene expression was only marginally different between these two cell lines. Overall, the progressive loss of PIGN protein expression in these leukemic cells and cell lines in the different MDS/leukemia progression phases indicated that PIGN loss might mark myeloid leukemia progression from a less aggressive disease state to a more aggressive one (Figure
2-4G). The partial intron retention mutations between exons 14 and 15 only occur in a subgroup of patients, especially those patients who are chemotherapy naïve.
65 PIGN gene expression aberration drove the genomic instability status in leukemic cells and was
TP53 regulatory pathway-independent
We further investigated the role of PIGN gene expression aberration in genomic instability
by comparing the gene expression levels of a group of genomic instability/DNA damage related
biomarkers in peripheral blood mononuclear cells collected from patient M2 at active leukemia
phase and leukemia remission phase. We observed that the biomarkers not regulated by TP53
(H2AX and SAE2) manifested a significant transcriptional activation in the active leukemia phase but not in the remission phase (Figure 2-5 A and B). H2AX, a member of the CIN-70 chromosomal instability marker panel is a genomic instability suppressor gene, and SAE2 encodes a protein involved in double strand DNA break repair232,233. BAXα, a proapoptotic gene, was significantly
downregulated in the active leukemia phase as well (Figure 2-5C). However, the expression of
TP53 target gene p21 and the TP53 deacetylase gene SIRT1 was not significantly different between
leukemia phase and remission phase (Figure 2-5 D and E). The TP53-dependent TRAIL death receptor DR5 was upregulated in the remission phase but was still about 50% below the DR5 gene expression in the normal control (Figure 2-5F). No significant difference in sequence identity (%) was found between NL and leukemic cells derived from patient M2 (Figure 2-5 G and H). The overall mutation rate was also similar between NL and leukemic cells with 11.4/kbp and 12.5/kbp respectively (Figure 2-5 I). NL cells displayed a total of 14 sequence alterations in the coding sequence whereas leukemic cells showed 22 (Figure 2-5 J and K). However, PIGN gene expression aberration was only observed in the leukemic cells. Thus, we proposed that PIGN gene expression aberration may be the driving force of high genomic instability in the leukemic cells.
66
Figure 2-5 PIGN expression aberration was associated with genomic instability in leukemic cells and was TP53- pathway-independent.
A-B. TP53-independent genomic instability/DNA damage markers (H2AX and SAE2) gene expression were significantly (p<0.05) upregulated in the leukemic phase compared to remission period. C. The expression of TP53-targeted apoptosis marker BAXα was downregulated in both leukemic phase and remission phase though it was more significantly (p<0.05) in the PMNC rich with leukemic cells. D. The TP53 target gene involved in cell cycle control (p21) was not significantly (NS) different between active leukemia and remission phase and could point to a
TP53-independent mechanism. E. The TP53 deacetylase and deactivator, SIRT1 was also not
67 significantly (NS) different between leukemia and remission phases of disease progression. F. The expression of the TP53 target, TRAIL death receptor 5 (DR5) was significantly downregulated in the leukemic cell rich, active leukemia phase compared to the remission phase, but DR5 expression was below 50% of the normal control in both leukemia and remission phases. Genomic instability biomarkers not regulated by TP53 (H2AX and SAE2) showed significantly transcriptional activation in mononuclear cells rich with leukemic cells in the active leukemia phase but not in mononuclear cells in the remission phase. Results were analyzed using a one-way ANOVA followed by Tukey’s post hoc tests. P-values <0.05 were considered statistically significant. G and
H. The PBMCs from patient M2 were sorted into leukemic and NL cell populations followed by
Sanger sequencing of approximately 2,300 bp of DNA encoding for intron and exon regions ranging from exons 2-11 of the TP53 gene revealed no significant (NS) difference in the overall sequence identity (%) of sequenced (G) introns and (H) exons between the NL and the leukemic cells with reference to the TP53 gene sequence (NC_000017.9). I. Combined intron and exon mutation frequency normalized to kilo bp (kbp) of the TP53 gene in NL and leukemic cells in patient M2. Qualitative analysis of sequence alterations of TP53 gene coding sequences (exons) of
J. NL cells and K. leukemic cell populations in patient M2 showed no significant difference.
Statistical differences were analyzed using Student’s t-test. P-values ≤0.05 were considered statistically significant. ‘NS’ indicates statistically nonsignificant (p>0.05) differences.
68 TP53 gene mutation analysis on leukemic and NL cells from patient M2 was conducted to
explore the role of TP53 deletion in leukemia progression. TP53 gene deletion was observed in both M2’s leukemic cells and NL cells. No significant difference in TP53 mutation profile between leukemic cells and NL cells. Approximately 2,300 bp of DNA from NL and leukemic cells
spanning exons 2-11 of the TP53 gene was analyzed by Sanger sequencing. However, only three
sequence alterations could be verified as a conserved deletion or missense mutations between the
different cell types (Table 2-2).
69 Table 2-2 Verified common NL and leukemic mutations in the TP53 gene
Positiona Exon Nucleotide change Type AA Change SIFT
11,031 2 G>A Non-synonymous p.S9N neutral
11,470 4 C>G Non-synonymous p.P80R neutral
12,379 5 C>- Frameshift deletion p.N131fs NA
a Position in GenBank NC_000017.9 b http://p53.iarc.fr/TP53GeneVariations.aspx
70 To further test our hypothesis that PIGN gene expression aberration can contribute to
genomic instability regardless of TP53 gene status, we employed the transient knockdown of PIGN
in multiple cell lines with various TP53 gene mutations. We investigated the impact of PIGN gene
expression suppression or silencing on the gene expression of p21 and H2AX in HL60 (TP53
deletion), K652 (TP53 mutation), HEK293 and HEK293 PIGN KO (TP53 wild-type) cell lines and
CD34+ mononuclear cells from a healthy individual. We observed that PIGN suppression in
HEK293 cells resulted in the upregulation of H2AX expression and γH2AX induction. Transient suppression of PIGN expression via RNAi led to an increased expression of γH2AX (Figure 2-6 A and B). Moreover, CRISPR/ Cas9 Knockout of PIGN in HEK293 cells confirmed the link between
PIGN loss and the induction of genomic instability in cells and involved an increased transcription
(~15-fold) of H2AX (Figure 2-6C). However, genomic instability was reduced as shown by γH2AX downregulation with the ectopic overexpression of PIGN (Figure 2-6 E). Similar experiments were conducted on a leukemia patient sample (M4), HL60 and K562 cells with similar observations. We confirmed these findings in K562 and HL60 cell lines (Figure 2-6 F-H). The knockdown of PIGN
K562 TP53 mutant cells does seem to induce γH2AX expression, and although significant, there was only a marginal increase in the expression of the CIN70 gene marker, H2AX (Figure 2-6 H).
However, no γH2AX induction was detected post-RNAi-mediated PIGN knockdown in the K562 cells (data not shown). Thus, to prove PIGN involvement rather than TP53, we also employed non- transformed TP53 wild-type (WT) hematopoietic CD34+ mononuclear cells from a healthy individual (Figure 2-6 I and J). Interestingly, H2AX expression in all the cells examined was not influenced by their TP53 gene mutation status, and PIGN gene expression status did not affect p21 gene expression. These results are consistent with the relationship between TP53 and p21 in theTP53-p21 inter-regulatory axis. Thus, PIGN loss or suppression induced genomic instability in a TP53 pathway-independent manner.
71
72 Figure 2-6 PIGN gene expression suppression was associated with genomic instability, and reintroduction of PIGN gene expression restored genomic stability in a TP53-pathway-independent manner.
A. PIGN suppression (***p=0.0008) in HEK293 (TP53wt) cells resulted in the upregulated gene expression of H2AX (**p=0.0029) but no significant change (NS) in p21 gene expression. B.
PIGN suppression resulted in DDR via a ~50% upregulation of γH2AX transcription and translation in HEK293 cells. C. PIGN deletion (***p< 0.0001) results in a ~15-fold increase
(***p=0.0003) in H2AX gene expression but a marginal increase in TP53-dependent p21 gene expression in HEK293 cells. D. Restoration of PIGN in PIGN null (CRISPR/Cas9 deletion)
HEK293 cells via transfection of PIGN expression plasmid results in a marked upregulation (i.e.
~400-fold) in PIGN gene expression with a ~1.6-fold increase (**p= 0.0056) of H2AX transcription and no significant (NS) change in p21 gene expression. E. Restoration of PIGN expression in PIGN null HEK293 cells ameliorates genomic instability as indicated by γH2AX suppression while increasing the mono-ubiquitination of H2AX which is critical in the initiation of DDR. F and G. PIGN loss (***p= 0.0003) in HL60 cells (TP53 null) results in a significant
(**p=0.0013) upregulation of H2AX in both (F) transcription level and (G) translation level but not (NS) p21 gene transcription. H. PIGN suppression (**p=0.0019) resulted in a marginal increase (*p=0.0387) in H2AX gene expression (NS) but not p21 gene expression (NS) in K562 cells (TP53 inactivation mutation). I. CRISPR/Cas9 ablation (***p<0.0001) of PIGN in normal healthy donor CD34+ mononuclear cells results in a significant (*p=0.0261) upregulation in
H2AX transcriptional activation without a significant (NS) increase in p21 transcriptional activation. J. PIGN loss via CRISPR/Cas9 ablation induces upregulation of γH2AX translation in normal healthy donor CD34+ mononuclear cells.
73 Discussion
Genomic instability is a driving force for cancer initiation and progression. It is an
emerging hallmark of cancer and has been implicated in the leukemic transformation of MDS into
AML94. However, the mechanism(s) by which genomic instability drives leukemic transformation has not been fully explored. Previous studies have indicated that cell lines with genomic instability
(i.e. Fanconi anemia and colon cancer cells with a mutator phenotype) had a marked increase in
the frequency of acquiring GPI-AP deficiency129–131. Moreover, MDS and myeloproliferative disease (MPD) patients bearing high frequency of GPI-AP deficiency have an increased risk for leukemia transformation125. This study sought to examine the role that the recently identified CIN
suppressor PIGN plays in this process. The PIGN gene expression profile of CD34+ mononuclear
cells from bone marrow aspiration of 48 high-risk MDS patients and 6 MDS/AML cell lines were
examined. Our preliminary data indicated a subgroup of MDS/AML patients and cell lines
containing complex cytogenetics appeared to have PIGN gene expression aberration with increased
gene expression but diminished protein production due to mutation caused frameshifting and early
translational termination (Figure 2-2). PIGN gene silencing leads to DNA replication stress and
chromosomal abnormalities in colorectal cancer cells during adenoma-carcinoma transition148.
Bioinformatics analyses were used to screen existing databases, and PIGN was identified as a predictor of MDS progression risk217. A unique gene expression aberration pattern within a subgroup of patients with MDS or AML-MRC was also observed. PIGN protein loss was linked to the presence of partial retentions of the intervening intron between exons 14 and 15 that resulted in frameshifts and early translational termination. The presence of this intron retention mutation was confirmed in multiple AML patients based on RNA-seq data analyses (phs001027.v1.p1) that verified the conserved nature of this mutation. Interestingly, no such natural alternative splice variants of PIGN gene have been reported (Ensembl release 85: ENSG00000197563)176. PIGN has
74 20 transcript variants of which two are known to encode the full-length PIGN protein, 13 are predicted to encode shorter length proteins, 2 undergo nonsense-mediated decay, and 3 are non- protein coding. Of these three non-coding transcripts, only 1 one has a retained intron. However, this intron does not occur in the same region as the one reported in this study. Similar deleterious splice defects have previously been reported for genes including DMD, C9orf72, ZNF9 and GR associated with muscular dystrophy, amyotrophic lateral sclerosis/ frontotemporal dementia, myotonic dystrophy type 2 and small cell lung cancer respectively234–237. It can however not be
ignored that, splicing factor genes have been reported to be frequently mutated in MDS and AML and may have played a role in the occurrence of this partial intron retention238. PIGN gene expression aberration and partial intron retention in a pre-treatment sample (M4) from an AML patient were also observed. Despite the progressive loss of PIGN protein expression in the relapse sample (M5); this intron fragment retention was not detected even though M5 was in the state of
PIGN gene expression aberration. Chemotherapy may have eliminated the clone harboring the
PIGN partial intron retention but the more chemoresistant clones survived and proliferated, which could explain why PIGN protein expression was not observed in M5. Our findings in patient M5 are in line with earlier studies that have reported the occurrence of multiple clones with various sensitivities to chemotherapy239,240.
The data from the patient samples, cell lines, and existing databases are consistent with
previous observations that, high frequency of GPI-AP deficiency is a marker of genomic instability
and might predict a risk of leukemic transformation and progression130,131,206. Previous studies have demonstrated that cell lines with genomic instability (i.e. Fanconi anemia and colon cancer cells with a mutator phenotype) had a marked increase in the frequency of acquiring glycosylphosphatidylinositol (GPI)-anchor protein deficiency129,206. Additionally, MDS patients with high GPI-anchor deficiency frequency bear a significant risk of leukemic transformation
75 within six months125. Those MDS patients and cell lines with a high rate of GPI-anchor deficiency
also presented with PIGN gene expression aberration (increased transcription activity but lost
translation capacity). Thus, in MDS PIGN gene expression aberration is a marker of genomic
instability and may predict a risk of leukemic transformation. Furthermore, this study indicated that
PIGN gene expression aberration might be the key factor linking GPI-AP deficiency with CIN and
leukemogenesis. Previous studies have demonstrated that, unlike PIGA, PIGN gene loss does not
completely terminate GPI-AP biosynthesis175,193. This valuable piece of information may explain
why the leukemic cells from patients M1 and M2 still showed signs of CFC formation reduction in proaerolysin-containing medium although the CFC counts were significantly higher than in the normal control.
The role of PIGN gene expression aberration in genomic instability/leukemic progression
and the role of the TP53 signaling pathway in the regulation of genomic instability during leukemia
progression were further investigated by studying the leukemic cells from patient M2, and several
cell lines. No significant difference in TP53 mutation profile between leukemic cells and NL cells.
In this patient, TP53 gene deletions were observed in both leukemic cells and NL cells and
manifested a similar mutation profile. However, the PIGN gene expression aberrations only occurred in the leukemic cells. A large number of TP53 gene sequence alterations found in both
NL and leukemic cells may be the result of patient M2’s extensive hydroxyurea (HU) exposure or the result of sequencing artifacts because the vast majority of TP53 gene sequence alterations were not conserved between the cell types241–243. Neither of the missense mutations present in codons 9 and 80 was believed to be oncogenic since they did not negatively impact TP53’s ability to execute downstream activation of its target genes considered to be essential for TP53’s ability to function as a tumor suppressor. Nonetheless, the frameshift deletion at codon 131 might be deleterious to the expression of the affected TP53 allele. It is plausible that this mutation represented an initiating
76 oncogenic event that inactivated one TP53 allele and predisposed to the deletion of the second
allele on chromosome 17. A similar phenomenon was reported in the normal epithelium of benign
breast tissue within the same breast cancer patients244. Moreover, Wong et al. had previously
reported the presence of functional TP53 mutations in mononuclear cells isolated from healthy
individuals245. It has also been suggested that TP53 loss may be permissive rather than causative
with regards to genomic instability148,153. Thus, this is likely a reflection of the loss of the CIN
suppressor PIGN, facilitating TP53 gene LOH in the leukemic cells isolated from this patient.
Finally, the role of PIGN gene expression aberration and TP53 gene deficiency in
escalating genomic instability during leukemia progression was investigated. The gene expression
of TP53-independent genomic instability/DNA damage markers (H2AX and SAE2) were
upregulated and the expression of the pro-apoptosis marker BAXα was downregulated specifically
in the leukemic cell-rich mononuclear cells at the active leukemia phase when compared to the
cells from the remission phase. However, the expression of TP53-dependent biomarkers, such as p21 and SIRT1, were not significantly different between the active phase and the remission phase.
Furthermore, the expression of the TRAIL death receptor 5 (DR5) was below 50% of that of the normal control in both active phase and the remission phase. We further observed that suppression or elimination of PIGN gene expression in several cell lines and CD34+ mononuclear cells from healthy individuals induced a similar TP53 independent pattern of genomic instability which could be reversed via PIGN gene expression restoration (Figure 2-6 A-J). The genomic instability makers not regulated by TP53 showed significant transcriptional activation in mononuclear cells rich with leukemic cells in the leukemia phase but not in mononuclear cells in the remission phase. However, the expression of those genomic instability markers regulated by TP53 did not show a significant difference between leukemia phase and remission phase. Those experiments were conducted on the samples collected from patient M2. In this patient, TP53 gene deletion was observed in both
77 leukemic cells and NL cells and manifested with a similar mutation profile. However, the PIGN
gene expression aberration only occurred in the leukemic cells. Thus, this is likely a reflection of
the loss of the CIN suppressor PIGN, facilitating TP53 gene LOH in leukemic cells and
corroborates the fact that TP53 loss alone is insufficient for the promotion of genomic instability
in those cells148,153,246. The findings in this study are also consistent with previous observations in
Li-Fraumeni Syndrome patients and may explain why such patients are prone to develop therapy-
related MDS with complex cytogenetics and poor prognosis247. Thus, PIGN gene expression
aberration may be the cause of high genomic instability in leukemia cells. PIGN gene expression
aberration may be the cause of leukemia progression in a subgroup of MDS and AML-MRC
patients.
To test this hypothesis, we investigated the impacts of PIGN aberrations on the expression
of CIN70 marker, H2AX (TP53-independent genomic instability marker) and p21 gene (TP53
dependent genomic instability marker) in HL60 (TP53 deletion), K562 (TP53 mutant), HEK293
and HEK293 PIGN KO (TP53 WT) cell lines. The results showed that both transient and permanent
suppression of PIGN gene expression led to increased H2AX expression that was not influenced by
TP53 status (Fig. 2-6 A-J). Overall, the data revealed that PIGN gene expression aberration was associated with genomic instability in leukemic cells and was independent of the TP53 regulatory pathway. Thus, high GPI-anchor deficiency frequency in MDS patients may be a marker of genomic instability and may predict a risk of leukemic transformation in MDS.
78 Conclusions
PIGN gene expression aberration was observed in a subset of these high-risk MDS or
AML-MRC patients. Moreover, PIGN gene expression aberration might contribute to genomic
instability and AML progression in a subgroup of MDS and AML-MRC patients. Finally, PIGN
links genomic instability and MDS leukemic transformation and is a marker of leukemic
transformation. This study provides additional evidence for the necessity of re-stratifying
MDS/AML risk estimation.
79
PIGN spatiotemporally regulates the spindle assembly checkpoint
Emmanuel K. Teye, Wenrui Yang, Shasha Lu, Thomas Abraham, Jong K. Yun, Douglas B. Stairs,
Gregory S. Yochum, Hong-Gang Wang, Jeffrey J. Pu
This work has been drafted as a manuscript for submission and has been reformatted here
Author Contributions
EKT and JJP designed this study and wrote this manuscript; EKT, WY and SL conducted experiments and EKT performed data analyses; TA and EKT performed confocal imaging
analyses; TA, JKY, DBS, HGW and GSY participated in manuscript formation by providing
comments and suggestions.
A portion of this work has been published as United States Patent and Trademark Office
Application No. 15/434,774 ‘Method and Therapeutic Use of PIGN and Other Genes or Gene
Products That PIGN Interacts With for Prognosis and Treatment of Hematological Neoplasias’.
80 Abstract
The spindle assembly checkpoint complex (SAC) also referred to as the mitotic or
metaphase checkpoint is responsible for proper chromosomal segregation during the metaphase to anaphase transition in mitosis. A novel function of PIGN as a potential integrator of the SAC was
revealed in this study. PIGN regulated the SAC via direct interaction with MAD1, MAD2 and the
mitotic kinase MPS1. The transient downregulation or elimination of PIGN resulted in impaired mitotic checkpoint activation via the disrupted expression of the SAC-related proteins MAD1,
MAD2, BubR1, and MPS1. Moreover, PIGN acts as a CIN suppressor to protect the SAC by regulating the genes that encode these mitotic checkpoint components. The restoration of PIGN via ectopic overexpression stabilized the expression of MAD1 and MAD2. Also, PIGN stabilizes
MAD1 and MAD2 during SAC activation to ensure proper chromosome segregation and it thereby
suppresses chromosomal instability (CIN). Ultimately, PIGN modulation could be potentially
adopted as a targeted therapeutic approach in cancer treatment, and more specifically in the
prevention of leukemia progression in high-risk MDS patients.
81 Introduction
The phosphoethanolamine (EtNP) transferase, phosphatidylinositol glycan anchor
biosynthesis class N (PIGN) was initially identified as a GPI-AP biosynthesis enzyme but has recently been shown in independent studies to suppress CIN and prevent protein aggregation in the
endoplasmic reticulum148,188,191. CIN commonly occurs in solid and hematological cancers98. CIN has been described as an enabling characteristic of cancer and ultimately results in aneuploidy and cancer progression94. Thus, a better understanding of the mechanisms involved is essential for therapeutic targeting141. In the previous chapter, PIGN gene expression aberrations arising from
partial intron retentions between exons 14 and 15 resulted in frameshifts and premature termination
and were linked to genomic instability and leukemia progression in a subset of MDS/AML patients.
The dynamic signaling interactions between mitotic kinases and SAC proteins/protein complexes
are involved in the induction of CIN and sensing of bi-oriented sister chromatids for chromosome
segregation93. Thus, for PIGN to support the maintenance of chromosomal stability, it should be
involved directly or indirectly in the regulation of mitotic checkpoint signaling. However, it
remains unknown whether PIGN mechanistically interacts with the SAC or other signaling
pathways. Thus, this study sought to elucidate the mechanistic role of PIGN as a CIN suppressor.
We examined the dynamic interplay between PIGN and SAC signaling. This was done by
exploring the potential function of PIGN as an interacting partner and regulator of SAC signaling
which is often dysregulated in leukemia248. The dynamic signaling interactions between the mitotic kinases and SAC proteins and protein complexes are involved in the induction of CIN and sensing of bi-oriented sister chromatids for chromosome segregation. Some of these SAC-related proteins, including TP53, BUBR1, BUB1, MAD2, AURK A, AURK B and PLK1 that are involved in cell cycle control are frequently dysregulated in AML248–251. Regulation of the SAC is vital for proper
segregation of sister chromatids during the transition from metaphase to anaphase in mitosis252,253.
82 This mitotic checkpoint is crucial for the protection of the genome from chromosome copy number alterations and aneuploidy254. The SAC stalls mitotic exit until the proper attachment of mitotic spindles to the chromosomes and bi-orientation of the chromosomes on the spindles is achieved255,256. Essentially, the SAC prevents premature chromosomal segregation and entry into anaphase. When chromosomes enter metaphase, and align, there are quality control checks to ensure proper kinetochore-microtubule (k-MT) attachment and proper tension. This is required to pull apart the sister chromatids to opposite spindle poles equally257.
The kinetochore is a complex signaling structure located at the centromere and mediates
the attachment of chromosomes to the microtubules. In the case, of improper k-MT attachment, the
SAC is activated via MAD1 which then recruits MAD2 and converts it from its inactive open form
to its activated closed conformation (c-MAD2) and can self-dimerize at the kinetochore258–261. CIN-
positive cells have excessively stable k-MT attachments compared to CIN-negative cells which
result in mitotic errors and ultimately chromosome missegregation93. Although not entirely clear, upon recruitment, phosphorylation, and activation of Aurora kinase B at the kinetochore, MPS1 phosphorylates KNL1 which then serves as a scaffold for the recruitment of additional SAC
proteins, including MAD1, MAD2, BUB3, BUB1, BUBR1, and CDC20253. At this point, the SAC effector also referred to as the MCC composed of activated MAD2, BUBR1, and BUB3 sequesters
CDC20 and prevents it from binding to and co-activating the ubiquitin ligase APC/C. The activated
MAD2 can interact with CDC20 to block APC/C activation which is required to polyubiquitinate and degrade cyclin B and securin. Securin holds the two sister chromatids together by preventing the activation of separase, which is responsible for chromosome separation262–264. As a result, SAC
activation stalls mitotic exit until the proper k-MT attachment and tension are achieved265,266. This
current study sought to investigate the regulatory role of PIGN on SAC components and elucidate
the possible interactions between PIGN and the SAC. Here we report that PIGN interacts with
members of the SAC and PIGN loss results in the dysregulation of some members of the SAC
83 including MAD1, MAD2, BUBR1, and MPS1 or the expression of associated genes. We primarily
focused on these proteins because of their direct involvement in the SAC and mitotic cell cycle
signaling.
Materials and Methods
Cell culture
Leukemia and lymphoblastoid cell lines HL60, K562, Jurkat, KCL22, KG1 and KG1a cells
were cultured at 37 C 5% CO2 with RPMI or IMDM supplemented with 20% FBS. HEK293 and
HEK293 PIGN CRISPR/Cas9⁰ Knockout cells were grown in DMEM supplemented with 10% FBS.
Cell lines were passaged on average every 3-4 days. As earlier described, CD34+ mononuclear
cells were isolated and grown in DMEM/F12 supplemented with 10% FBS, 50μM 2- mercaptoethanol, Glutamax (Life Technologies), MEM non-essential amino acids (Life
Technologies) and StemMACS HSC expansion cocktail (Miltenyi Biotec)267. All cells were cultured at 37°C under 5% CO2 conditions. Mononuclear cells were isolated from the blood or bone marrow aspirates of donors using the Ficoll-Paque PLUS reagent (GE Healthcare) as earlier described.
Gene expression analyses
RT-qPCR experiments were conducted as earlier described267. Total RNA was isolated
from the cells using the RNeasy Mini Kit (Qiagen) by the manufacturer’s protocol. Total RNA was reverse transcribed using the High Capacity cDNA reverse transcription kit (Life Technologies) on the Mastercycler® Nexus (Eppendorf). The RT-qPCR step was conducted on the StepOnePlus™
84 real-time PCR System (ABI systems) using PIGN (Hs00202443_m1), MAD1 (Hs00269119_m1)
and MAD2 (Hs01554513_g1) gene expression assay (Life Technologies) and 18S (rRNA
Hs99999901_s1) (Life Technologies) gene expression as an internal reference control. BUBR1 and
MPS1 primers as well as GAPDH internal control primers (Integrated DNA Technologies) (see
Appendix C for primer sequences) were used with Power SYBR® Green PCR Master Mix (Life
Technologies). For all experiments, samples were run in triplicates, and expression data were
normalized to PIGN gene expression in the control group. Gene expression fold changes were
calculated using the ΔΔCt method.
HA-tag immunoprecipitation (IP)/ and co-immunoprecipitation (Co-IP) analyses
PIGN-HA IP/Co-IP experiments were conducted by transient transfection of
CRISPR/Cas9 PIGN knockout HEK293 cells with the SRα promoter-driven expression vector pMEPuro3HAhPIGN or the empty vector without PIGN cDNA cloned193. HEK293 CRISPR KO cells and pMEPURO3HAhPIGN expression plasmids were the gifts from Drs. Taroh Kinoshita and
Yoshiko Murakami, Research Institute for Microbial Diseases, Osaka University. The cells were transfected with 2.5-5 μg of the vector using the Lipofectamine 3000 transfection reagent (Life
Technologies) according to the manufacturer’s protocol. Protein samples were obtained 24-72 hours post-transfection, and 250-500 μg protein was used with the HA-tag IP/Co-IP kit according to the manufacturer's protocol. The eluates and 10% of the input lysates were used for Western blot analyses. Co-IP experiments were conducted with the PierceTM Co-IP kit per the manufacturer’s protocol. Prior to the Co-IP experiments, the cells were treated with 60 ng/µl Taxol (Bristol-Myers
Squibb) and 60-100 ng/µl Nocodazole (Sigma) for 12 hours to activate the SAC. For each experiment 2 mg of the whole-cell lysate was used and 20% of the lysate sample was loaded for
85 use as input control. The MAD1 (GTX109519, Genetex) antibody was employed for the Co-IP experiments.
Western blot analyses
Western blot analysis was conducted as earlier described267. Briefly, total protein extraction
was done using RIPA cell lysis buffer supplemented with phosphatase inhibitor cocktail and
protease inhibitor (Sigma). Whole-cell lysates were subjected to electrophoresis in a NuPAGE™
4-12% or 10% Bis-Tris Gel (Life Technologies) and transferred to an Immun-Blot® PVDF
Membrane (Bio-Rad). The membranes were blocked in 5% milk in TBS-T (i.e. 0.1% v/v Tween
20) for 1 hour at room temperature and subsequently treated with primary antibodies (1:500-
1:1000) prepared in 5% milk/TBS-T overnight at 4ºC. The blots were incubated overnight with anti-PIGN antibodies (HPA039922, Atlas Antibodies) anti-MAD1 antibodies (Clone BB3-8,
MABE867, Millipore), anti-Histone H2A.X (D17A3, Cell Signaling Technology) anti-phospho-
Histone H2A.X ser139 (S139, Cell Signaling Technology), cyclin B1 (4138, Cell Signaling
Technology) and HA-tag(C29F4, Cell Signaling Technology) in TBS-T with 5% Non-Fat Dry
Milk. Mouse beta-actin (C-4, sc-47778, Santa Cruz Biotechnology) was used as loading control.
The membranes were then washed three times with TBS-T and incubated for 1-2 hours with a horseradish peroxidase-conjugated goat anti-rabbit (AP132P) or goat anti-mouse (AP124P) IgG secondary antibody (1:5000-1:10000; Millipore). Afterward, membranes were washed three times with TBS-T and detection was conducted by treatment with ECL Prime Western Blotting detection reagents (Amersham).
86 PIGN knockdown and CRISPR/Cas9 Knockout studies
RNAi-mediated PIGN knockdown experiments were conducted using the Nucleofector™
II Device (Amaxa) in conjunction with the Cell line Nucleofector™ Kit V reagent kit (Amaxa).
CRISPR/Cas9 experiments were performed according to a modified LentiCRISPRv2 (Addgene
plasmid #49535) protocol268. The gRNA (AAACGGTCATGTAGCTCTGATAGC) we employed
targets PIGN at exon 4 and results in a frameshift193. Lentiviral-transduced CD34+ mononuclear
cells were harvested for downstream applications nine days post-infection according to a modified protocol269. PIGN knockdown and CRISPR/Cas9 Knockout studies RNAi-mediated PIGN knockdown experiments were conducted using the Nucleofector™ II Device (Amaxa) in conjunction with the Cell line Nucleofector™ Kit V reagent kit (Amaxa) by the manufacturer’s recommended protocols for the respective cell lines. The cells were transfected with 100 nM siGENOME™ siRNA Human PIGN, (D-012463-01, Dharmacon), ON-TARGETplus Human
MAD1L1(8379) siRNA-SMARTpool (L-006825-00-0005, Dharmacon) or 100 nM siGENOME™
Control siRNA Non-targeting siRNA #2, (D-001210-02-05, Dharmacon) and incubated for 24-72 hours.
Cell cycle analyses and SAC activation
Cell cycle synchronizations were performed as earlier reported270. Serum starvation for 72
hours was used to synchronize cells in G0 as earlier described270. Cells were synchronized at the
Go/G1, S and G2/M phases. Cell cycle synchrony was monitored using propidium iodide stained cells with the FACS Calibur flow cytometer (BD Biosciences). Protein lysates were obtained and used for Western blot analyses as earlier described267. Please refer to Figure 3-1A for schematics of the cell cycle synchronization procedures.
87 Immunofluorescence and confocal microscopy
For missegregation and co-localization analyses, cells were blocked in early S-phase via the double-thymidine treatment and released for 6-8 hours into the mitotic phase. Adherent cells were cultured on chambered slides. Cytospin was used to fix suspension cells onto the slides. Cells were fixed in 4% paraformaldehyde in PBS and permeabilized in -20°C 100% methanol. The slides were blocked with 2.5% normal goat serum diluted in PBS. For all wash steps, the cells were washed three times for 5 minutes each in PBS. The cells were treated at 4°C overnight with primary antibodies: Human anti-centromere (kinetochore) (15-234, Antibodies Incorporated), PIGN
(HPA039922, Atlas Antibodies), MAD1 (Clone BB3-8, MABE867, Millipore), MAD2 (sc-
374131, Santa Cruz), BUBR1 (720297, ThermoFisher) or MPS1 (05-682, Millipore) followed by washing and treatment for 2 hours at room temperature in the dark with secondary antibodies
(1:1000): Alexa Fluor® 647 goat anti-human IgG (H+L) (A21445, Life Technologies), Goat anti- rabbit IgG, Dylight® 488 Conjugated Highly cross-adsorbed (35553, Life Technologies) or Goat anti-mouse IgG, Dylight™ 633 Conjugated (35513, Life Technologies) respectively. Antibodies were diluted in 1% normal goat serum diluted in PBS. The slides were partially dried in the dark and mounted in Vectashield Hard Set™ mounting medium with DAPI (H-1500, Vector
Laboratories). Images were acquired using the Leica SP8 inverted confocal microscope at the
Microscopy Imaging Facility at Penn State College of Medicine. Three-dimensional image stacks were acquired in 0.15-μm steps using a ×40 1.4 N.A oil immersion objective. Deconvolution and analyses of image stacks were performed using the Huygens workstation (Scientific Volume
Imaging B.V.) and the Imaris Microscopy Image Analysis (Bitplane AG) and Volocity 6.3
(PerkinElmer Inc).
88 Statistical analyses
GraphPad Prism 5 software and Microsoft Excel 2010 were used for statistical analyses.
Two-tailed Student’s t-tests or one-way ANOVA followed by Tukey’s post hoc tests were used for comparisons. P-values ≤ 0.05 were considered statistically significant.
Results
PIGN expression is cell cycle regulated
To clarify the mechanistic role of PIGN in the maintenance of chromosomal stability, we first examined the impact on PIGN on the cell cycle signaling and more specifically, SAC pathways. The SAC is primarily responsible for ensuring proper chromosomal segregation during metaphase-anaphase transition255. The mechanistic role of the CIN suppressor, PIGN in maintaining chromosomal stability via SAC regulation was investigated. Thus, to initially determine the relationship between PIGN and the SAC, we performed cell cycle synchronization experiments using myeloid and lymphoblastoid cell lines and examined PIGN expression in different stages of the cell cycle. Cell cycle synchronization experiments were conducted by
blocking cell cycle progression at G0/G1, G1/S and G2/M phases in K562, HL60, KCL22 and
Jurkat cells via serum starvation, double-thymidine and nocodazole treatment respectively (Figure
3-1A). Synchronization of myeloid and lymphoblastoid cell lines at the various cell cycle stages revealed that PIGN expression was cell cycle-regulated (Figure 3-1 B-E).
89
Figure 3-1 PIGN expression is cell cycle-regulated.
A. Schematic of cell cycle synchronization protocols. Cells were treated with double-thymidine (2 mM) to arrest them in early S-phase or with nocodazole (100 ng/µl), a microtubule destabilizer, to block the cells in the G2/M phase. Cells were synchronized in Go/G1 phase by serum starvation for 72 hours. Cells were synchronized in G1/S-phase via double-thymidine block and in G2/M by double-thymidine block and release followed by nocodazole treatment for 18 hours. Cells were then washed with PBS and collected for total protein extraction. Total RNA was isolated to examine the cell cycle-dependent gene expression of PIGN, MAD1, MAD2, and BUBR1. GAPDH and 18S
90 were used as internal reference controls for RT-qPCR. Cell synchronizations were confirmed by propidium iodide staining and flow cytometry analyses. FACS, RT-qPCR and Western blot
analyses for multiple cell lines are shown in B-D. PIGN and MAD1 were similarly expressed in a
cell cycle-dependent manner in multiple cell lines: B. HL60 C. K562 D. KCL22 and E. JURKAT
with suppressed expression from early S-phase to the G2/M phase. Error bars represent standard deviation from the mean fold change in gene expression relative to gene expression at Go/G1.
91 Overall, PIGN and MAD1 were least expressed during the G2/M phase of the cell cycle,
and protein levels of PIGN steadily decreased from early S-phase into G2/M phase in almost all the cell lines that were examined. In some instances, the pattern of PIGN expression was also comparable to the model of the expression of other components of the SAC including MAD2 and
BUBR1 during mitosis. These experiments unveiled a correlation between PIGN and the SAC in multiple cell signaling and phenotypic contexts, myeloid and lymphoblastoid alike. Thus, PIGN expression is cell cycle-regulated like SAC components.
PIGN loss or suppression results in the dysregulation of SAC components
To determine whether PIGN loss or downregulation could impact the SAC RNAi-mediated knockdown and CRISPR/Cas9 ablation of PIGN was employed. CRISPR/Cas9 mediated knockout of PIGN in CD34+ mononuclear cells derived from a healthy individual resulted in the suppression of both MAD1 and MAD2 protein and gene expression (Figure 3-2A). A similar observation was made in HEK293 PIGN CRISPR/Cas9 Knockout cells. PIGN loss in the HEK293 cells resulted in the suppression of MAD1, MAD2 and MPS1 expression but an upregulation in BUBR1 expression.
In PIGN ablated-K562 myeloid leukemia cells, the expression of other SAC components including the MAD1 interacting partner, MAD2 were dysregulated both at the gene and protein expression levels (Figure 3-2 A-B). Except for MPS1, in which an increase in gene expression but a corresponding decrease in protein expression was observed, all the other SAC components gene and protein expression were comparable (Figure 3-2B). The impact of PIGN ablation on the expression of other SAC-related genes was also examined in multiple cell lines. Moreover, PIGN loss was accompanied by an increase in the frequency of missegregation errors in PIGN
CRISPR/Cas9 HEK293 KO cells and increased frequency of CIN-associated missegregation errors
(Figure 3-2 C-E). We observed a higher percentage of mitotic cells with lagging chromosomes in
92 the PIGN KO cells compared to the cells with no PIGN ablation. However, the ectopic
overexpression of PIGN resulted in the modulation of MAD2 (Figure 3-2F).
Next, we sought to determine whether transient downregulation of PIGN could lead to
changes in the expression of those SAC components that were impacted by complete ablation of
PIGN. RNAi-mediated suppression of PIGN caused MAD1 and MAD2 suppression and BUBR1 upregulation in K562 myeloid leukemia cells and was accompanied by an increase in the frequency of missegregation errors (Figure 3-2G and H). Complete loss of PIGN via CRISPR/Cas9 ablation resulted in the suppression of MAD1, MAD2, BUBR1 and MPS1 (Figure 3-2I). To further elucidate the regulatory relationship between PIGN and the SAC, the RNAi-mediated knockdown of a major starting component of the SAC, MAD1 was employed. PIGN downregulation accompanied RNAi-mediated suppression of MAD1 in conjunction with an upregulation in
BUBR1 and MPS1 expression (Figure 3-2J).
93
94 Figure 3-2 PIGN loss or suppression results in the disrupted expression of SAC components.
A-B. Western Blot and RT-qPCR analysis of MAD1 and MAD2 in wild-type normal CD34+ mononuclear cells and following PIGN KO in normal CD34+ mononuclear cells. C-D. Comparison
of PIGN WT HEK293 cells and PIGN null (KO) HEK293 cells: (C) Western blot and (D) RT-
qPCR analyses E. PIGN CRISPR/Cas9 ablation increased the frequency of missegregation errors.
F. Ectopic overexpression of PIGN in HEK293 PIGN KO cells. G. RNAi-mediated PIGN
suppression in K562 cells. H. Missegregation errors following RNAi-mediated suppression of
PIGN in K562 cells. I. Western blot analyses comparing CRISPR/Cas9 KO of PIGN in K562 cells compared to wild-type cells. J. Western blot analyses of RNAi-mediated knockdown of MAD1 in
K52 cells. Error bars represent standard deviation from mean fold change in gene expression relative to gene expression in the wild-type cells (*p < 0.05, ***p<0.0001).
95 However, no detectable change was observed in MAD2 expression post-MAD1
knockdown. Essentially, MAD1 suppression resulted in decreased expression of PIGN. Overall,
similarities in the impact of either MAD1 or PIGN downregulation on the SAC components MAD2,
BUBR1 and MPS1 were observed. Thus, PIGN loss affects the SAC assembly like MAD1. These
results point to a unique reciprocal regulatory relationship between the primary SAC component
MAD1 and PIGN.
PIGN spatiotemporally interacts with SAC components
To further investigate the relationship between PIGN and MAD1, we transfected
CRISPR/Cas9 PIGN KO HEK293 cells with an HA-tagged PIGN and performed HA-tag IP experiments. HA-tag IP experiments were employed to test interacting partners of PIGN. A direct interaction between PIGN and MAD1 with the highest interaction at 48 hours post-transfection was observed (Figure 3-3 A-B). The detection of PIGN interaction with MAD2 was highest at the 48-
hour time point post-transfection with the HA-tagged PIGN at which point PIGN expression was
also the highest (Figure 3-3 A-B). Thus, PIGN may directly interact and form a complex with
MAD1 and MAD2 during SAC activation. Because the SAC is activated during mitotic cell
division, an experiment was conducted to verify whether this PIGN-inclusive SAC was formed
during mitosis, the HA-tag IP experiment was repeated using G2/M synchronized cells (Figure 3-
3 C). The results of this analysis showed that PIGN interacted with MAD1 and MAD2 during the
mitotic phase and revealed PIGN as a novel interacting partner (Figure 3-3 C). Thus, PIGN
physically associated with the SAC complex during the mitotic phase of the cell cycle. However,
we did not observe any interaction of PIGN with BubR1 and MPS1 in either asynchronous or
mitotic cells. Moreover, MAD1 was co-precipitated with PIGN and MAD2 during chemically-
induced SAC activation (Figure 3-3 D). The two drugs share a similar target β-tubulin and halt
96 mitotic division. However, there was an observable difference in the level of PIGN, MAD1 and
MAD2 interactions between the Taxol-treated and the nocodazole-treated cells.
97
Figure 3-3 PIGN physically interacted with SAC components.
A-B. Western blots of HA-tag pull-down assay in PIGN null HEK293 cells with ectopically overexpressed 3HA-tagged PIGN in asynchronous cells. C. HA-tag pull-down assay in G2/M synchronized PIGN null HEK293 cells with ectopically overexpressed 3HA-tagged PIGN. D.
Western blot of Co-IP assay during SAC activation via Taxol (60 ng/µl) or nocodazole (60-100 ng/µl) treatment of K562 cells for 12 hours. E. HA-tag pull-down and cell cycle analyses (FACS) of cells released from early S-phase into mitosis. Inputs represent 5-10% of total protein lysate.
98 Next, we sought to determine, whether these interactions were cell cycle-dependent and not limited to the mitotic stage of the cell cycle. To do this, we synchronized cells in early S-phase
(G1/S-phase) via double-thymidine block and release to examine whether the interactions between
PIGN, MAD1, and MAD2 also occur before mitotic entry (Figure 3-3 E). Also in this experiment,
MPS1 which is involved in the recruitment of MAD1 to the kinetochore during SAC activation was examined. The results showed interactions with MAD1 but not MAD2 at all the time-points
(0-10 hours). However, we identified MPS1 as an interacting partner of PIGN during this 10-hour time course (Figure 3-3 E). It was observed that the interaction between MPS-1 and PIGN following early S-phase release was time-dependent. These experiments provide preliminary evidence of the spatiotemporal interactions between PIGN and SAC components. Moreover, these results may also reveal the sequence in which these interactions occur, with PIGN and MAD1 binding to each other and MAD2 binding to the MAD1 and PIGN complex later. The above data suggest that PIGN maintains chromosomal stability via regulation and interaction with the SAC protein MAD1, MAD2, and MPS1 during the cell cycle. PIGN may be required for SAC protein activation, migration, and interaction.
PIGN co-localizes with SAC components during SAC activation
To visualize and verify the interaction between PIGN and the SAC components, confocal microscopy, and Co-IP experiments were conducted. Co-IP studies were used to verify that the interaction between PIGN, MAD1, and MAD2 occurred endogenously during SAC activation
(Figure 3-3 D). Thus, the interaction between PIGN, MAD1, and MAD2 may be an early event during SAC activation. Co-immunofluorescence analyses revealed a dynamic co-localization of
PIGN, MAD1and MAD2 during various stages of mitosis (Figure 3-4 A-B).
99
Figure 3-4 PIGN co-localizes with SAC components during SAC activation.
100 Co-localization of PIGN (green) with A. MAD1 [(% co-localization) 1. Prometaphase (42.5%) 2.
Metaphase (52.5%) 3. Anaphase (13.6%) 4. Late Anaphase (60.4%) and B. MAD2 [Prophase
(upper panel), Prometaphase (Lower panel)] during SAC activation. HEK293 PIGN CRISPR/Cas9
Knockout cells were transfected with a pMEPuro3HAPIGN plasmid for 48 hours followed by treatment with nocodazole (100 ng/ul) for 12 hours. The cells were fixed with 4% paraformaldehyde and treated with mouse anti-MAD, anti-MAD2 or anti-MPS1 and rabbit anti-
HA, followed by treatment with the respective fluorescently-labeled secondary antibodies.
Chromosomes were stained with DAPI (blue). Laser scanning confocal microscopy was used to visualize the stained cells, and image analyses were conducted using the Volocity 6.3 High- performance 3D imaging software (PerkinElmer). Scale bars, 2-3µm.
101 PIGN and MAD1 showed localization patterns commensurate with what has previously
been reported192. We expected PIGN to translocate to the nucleus and thus the chromosomes during
the transition from G1/S to G2/M phase. However, this implied that PIGN translocation may be
dependent on nuclear membrane breakdown and may not occur until the mitotic phase.
Discussion
Since its discovery as a CIN suppressor, the molecular mechanism by which the loss of the
GPI-AP biosynthesis protein PIGN leads to CIN has remained elusive. PIGN ablation results in
genomic instability by triggering DDR via gamma-H2AX upregulation267. This current study
further investigated the mechanistic role of PIGN in maintaining genomic stability and sought to
elucidate the mechanistic basis of PIGN as a suppressor of CIN. In this study, we showed for the
first time that PIGN interacted with and regulated the SAC, and its expression was cell cycle-
dependent (Figures 3-1; 3-2). The downregulated expression of PIGN and other SAC proteins during the G2/M phase compared to the S-phase points to the tight regulation of the expression of these proteins during the mitotic phase and more so during SAC activation. Moreover, the SAC has been reported to be activated as early as during the S-phase which may explain the relatively higher expressions observed during the S-phase271,272. The reciprocal regulation between PIGN and MAD1
as well as their similar cell cycle-regulated expression indicates that either of these proteins may be a limiting factor in the regulation of the other SAC components (Figure 3-2 G and J). We also
demonstrated that the spatiotemporal interactions between PIGN and the SAC occur from early S-
phase into mitosis (Figure 3-3). Although PIGN was not previously known as a SAC protein, it
may have a functional role in the regulation and activation of the SAC and ultimately as a CIN
suppressor. Interestingly, several proteins including WT1, RED, NUP153, YY1, SMURF2,
102 TRRAP, and TAp73 which previously had no physical interaction or functional association with the SAC have recently been reported to interact with or regulate the SAC273–280.
The dysregulation of these SAC components with PIGN suppression or loss may indicate
the role that PIGN plays in stabilizing these proteins and regulating their availability and thus
migration to disrupted spindle attachment to the kinetochore. PIGN loss abrogates SAC function,
but it is yet to be elucidated the way PIGN regulates or ensures kinetochore localization of MAD1,
MAD2, BUBR1, and MPS1. Our observation of PIGN interacting with MAD1 and MPS1 from early S-phase through G2/M phase indicates that MAD1 and MPS1 interaction with PIGN could be involved their recruitment to the kinetochore. It would be interesting to explore this potential role of PIGN further. Nonetheless, due to the widespread cytoplasmic localization of PIGN, it remains to be determined the exact way in which PIGN can impact the kinetochore localization of these SAC components during SAC activation. However, our finding that MAD1 and MPS1 interact with PIGN from early S-phase into the mitotic phase may point to the role that PIGN may play in the recruitment of these proteins to the kinetochore during SAC activation (Figure 3-3 A-
E). Moreover, MAD2 recruitment may be a later event following MAD1 and MPS1 recruitment because it seems to only interact with PIGN in the mitotic phase and during SAC activation (Figure
3-3 C-E). The recruitment of MPS1 and MAD1 to the kinetochore followed by MAD2 recruitment during SAC activation with MAD1 having an additional function of promoting the mitotic checkpoint has earlier been reported281–283. However, other studies have indicated that during SAC
activation, the recruitment of MAD1 to the kinetochore is facilitated by the Rod-Zw10-Zwilch
complex, the NDC80/HEC1 complex and the centromere-kinetochore complex protein, CENP-
F266,284. Interestingly, the ectopic expression of PIGN restores MAD1 and MAD2 expression and
may also point to a degree of regulatory control between PIGN and the two SAC proteins. Aside
from their role in the SAC, MAD2, and BUBR1 have also been implicated in the timing of mitotic
regulation even when they are cytosolic and not yet recruited to the kinetochore285. So, the fact that
103 PIGN seems to regulate the expression of these proteins points to another potential regulatory
control on mitotic timing. Our preliminary finding that the depletion of PIGN accelerated the
frequency of cell cycle progression corroborated this observation (See Appendix E).
The SAC protein MAD1 is necessary for ensuring proper alignment and segregation of
chromosomes284,286. Moreover, PIGN depletion resulted in the occurrence of lagging chromosomes mitotic defects similar to what was earlier reported in colorectal cancer cells148. Thus, the increased
frequency of missegregation errors arising from MAD1 downregulation following PIGN depletion was expected (Figure 3-2 E-F). It is well known that any slight changes in the expression of SAC components are enough to cause missegregation errors and ultimately CIN. Also, the concept of the recruitment of SAC components to the kinetochore to ensure proper k-MT attachment during metaphase to anaphase transition has been extensively documented255,266. Thus, it would be
interesting to see how PIGN depletion affects SAC and more specifically, MAD1 recruitment to
the kinetochore. This is important because of the dynamic localization of MAD1 between the
spindle attached-kinetochore and the spindle poles for the respective activation and silencing of the
SAC255,261,287.
The observation that PIGN and MAD1 co-precipitate during SAC activation may point to a similar relocalization pattern between PIGN and MAD1 as part of SAC activation and deactivation (Figure 3-3 D). The observed difference in the effects of Taxol and nocodazole could be because nocodazole disrupts mitotic spindle function and as such may induce a stronger SAC response, unlike Taxol which promotes the formation of highly stable microtubules that are resistant to depolymerization. Nocodazole reversibly interferes with microtubule dynamics and polymerization and induces mitotic arrest. The fact that depletion of PIGN was accompanied by
MAD1 depletion and vice versa is an indicator of the complementary role that PIGN and MAD1 could play in stabilizing each other (Figure 3-2 G and J). SAC dysregulation has been implicated in CIN induction. Thus this study focused on examining the role of PIGN in the regulation of the
104 SAC. The genetic ablation of SAC components may have dramatic consequences on cell survival depending on the cell type and the genetic context266. In the same study, we demonstrated that PIGN and MAD1 are interacting partners and thus, identified PIGN as a novel SAC interacting protein.
105
Figure 3-5 Models of the spatiotemporal interaction between PIGN and SAC components during the cell cycle and SAC activation.
106 PIGN interacts with MPS-1 and MAD1 in a pre-SAC complex in early S-phase till late S-phase where they might dissociate and reassociate just before entry into mitosis. During SAC activation in mitosis MPS1 is released and PIGN forms a complex with MAD1 and MAD2.
107 However, it remains to be elucidated the role that PIGN plays in the regulation of the SAC.
Furthermore, we sought to identify at which stage of the cell cycle these interactions occurred and other SAC components that may be involved. As expected, we identified in G2/M synchronized cells that, the interactions between PIGN, MAD1 and MAD2 took place during mitosis (Figure 3-
3 B). The current data does not describe the exact effect of PIGN on the SAC. However, it does show that PIGN may stabilize the SAC components/mediators MAD1, MAD2, MPS1 and BUBR1 for proper SAC function. It is yet unclear how PIGN interacts with the SAC and which protein sequences are essential for these interactions. Nonetheless, the verdict based on our observations and others indicate that proteins that were previously thought to be non-SAC-related such as PIGN could be actively involved in the regulation and activation of SAC during metaphase to anaphase transition and ultimately prevent CIN and leukemia progression (Figure 3-5).
Conclusions
PIGN knockdown, knockout, overexpression studies and co-immunoprecipitation studies were performed to determine the impact of PIGN loss on SAC signaling. Additionally, we tested whether the depletion of PIGN results in aberrant cell cycle signaling and thus defective chromosomal segregation. The absence of PIGN results in defective mitotic checkpoint signaling and causes a faster mitotic exit with an accumulation of missegregation errors. Our observations revealed that PIGN physically interacted with and or regulated the SAC proteins, MAD1, MAD2,
MPS1 and BUBR1 during mitotic cell cycle progression and SAC activation. The co-purification of PIGN with some of these mitotic checkpoint proteins indicated the possibility that PIGN plays a direct role in the regulation of mitotic checkpoint signaling. Thus, PIGN as a CIN suppressor is crucial in the regulation of mitotic integrity via the SAC to maintain genome stability.
SUMMARY & FUTURE DIRECTIONS
Genomic instability is associated with cancer initiation and progression and has been
indicated as a driver of the clonal evolution of MDS to AML with poor overall survival99,110. It is
also responsible for the accumulation of genetic abnormalities that contribute to the transformation
of MDS into AML94,106,114,212. An in-depth understanding of the molecular basis of MDS will further
improve prognostication of disease outcomes in the patients and guide treatment options. Sadly, no
specific molecular biomarkers or diagnostic tests are available to predict the leukemic
transformation from MDS to AML. Moreover, the current methods of risk-stratification based on
standard cytogenetics and karyotype complexity are not sensitive. In fact, up to 50% of MDS/AML patients may present with normal karyotypes104,288. Thus, there is an urgent need to identify
potential molecular biomarkers of genomic instability that could be utilized for the prognostication of leukemia transformation.
PIGN gene expression aberration, self-renewal of leukemic clones and MDS transformation
Recently, PIGN, a protein participating in the final steps of GPI-AP biosynthesis, was
identified as a CIN suppressor in a colon cancer model148. However, the role of PIGN in hematological malignancy initiation and progression had never been addressed. Thus, this research
investigated the mechanistic basis of the involvement of PIGN in CIN and leukemia progression.
Initially, partial retention of the intervening intron between exons 14 and 15 of the PIGN gene was
identified and resulted in frameshifts and premature termination of the encoded protein. Moreover,
this partial intron retention was associated with PIGN gene expression aberration that was linked
109 to genomic instability, MDS progression, and leukemic transformation. PIGN gene expression
aberrations were associated with increased frequency of GPI-AP deficiency in leukemic cells
during leukemic transformation/progression. Thus, PIGN loss could potentially mark the leukemic
transformation of MDS to AML. However, these findings need to be further validated in a
longitudinal study involving a larger cohort of MDS/AML patients.
Genomic instability drives clonal selection and self-renewal thus, assessment of the self-
renewal ability of cells with PIGN gene expression aberration in comparison to those with intact
PIGN is essential. It is important to answer the specific question of how PIGN more directly contributes CIN suppression in leukemic clones289. Leukemic clones with PIGN gene expression aberration should have high self-renewal ability. However, subpopulations of leukemic clones with dysregulated checkpoint signaling may undergo mitotic catastrophe and result in the induction of apoptosis290,291. Thus, the functional impact of PIGN gene expression aberrations on the induction of apoptosis in leukemia cells also needs to be explored.
This study revealed for the first time that, healthy NL cells in an MDS/AML patient harbored TP53 mutations. This might have been because the patient under consideration was not naïve to chemotherapy. Thus, the chemotherapy might have negatively selected for TP53-mutant clones. Moreover, TP53-mutant clones may have been preferentially selected during the leukemic blasts cell sorting experiment because the patients under consideration had about 90% TP53 deletion based on FISH data. However, these possibilities may be unlikely because functional TP53 mutations have been detected in small populations of peripheral blood cells of healthy- chemotherapy naïve elderly individuals and the normal epithelium of benign breast cancer tissue244,245. PIGN linked genomic instability to leukemic transformation independent of TP53 signaling. PIGN involvement in DDR and mitotic regulation in multiple cells was also reported in this study. Interestingly, Burrell et al. and Bunz et al. have independently presented data that
110 suggest that TP53 loss may be permissive rather than causative with regards to genomic instability148,153. The findings of this study corroborate the fact that TP53 loss is insufficient for the promotion of genomic instability and this has been demonstrated in human and animal studies alike148,153,246. PIGN could potentially be employed as a molecular biomarker for risk-stratification
for better treatment and management of high-risk MDS patients.
PIGN involvement in CIN induction and SAC regulation
Despite the ubiquity of CIN in cancer, there is still limited knowledge about the
mechanism(s) involved. Moreover, conflicting opinions exist in the literature with regards to the origins of CIN, whether pre-mitotic or mitotic. One group holds that CIN is the result of pre-mitotic
defects associated with replication stress while the other proposes mitotic origins93,102,148,154,292.
Nonetheless, the findings of this research study demonstrated that both schools of thought apply to
PIGN because depletion of PIGN resulted in segregation errors during mitosis and triggered DDR.
This study also revealed a novel mechanistic role of PIGN in the maintenance of genome stability
via SAC regulation. The SAC primarily ensures bipolar segregation of sister chromatids during the
transition from metaphase to anaphase253. The SAC is responsible for sequestering the adapter protein, CDC20 from activating APC/C293. A cohesin complex holds sister chromatids together
until anaphase where securin is targeted for destruction via polyubiquitination and proteasome destruction by CDC20-activated APC/C266. This is a crucial step in SAC activation, and it would be worth examining in future studies the effect of PIGN loss on the rate of securin degradation which is a marker of APC/C activation. Interestingly, mutations in STAG2 one of the genes that encode a protein in the cohesin complex has been linked to CIN294. The degradation of securin, in
turn, releases separase which cleaves the cohesin complex and releases the chromosomes for
separation295. In fact, overexpression of securin and separase have been linked to CIN induction
111 and cancer progression296,297. Thus, a faulty SAC is the main culprit in the development of CIN due to missegregation errors298,299.
The intersection between GPI-AP biosynthesis defects, CIN and leukemia progression
Increased frequency of glycosylphosphatidylinositol-anchor protein (GPI-AP) deficiency has been linked to genomic instability and carcinogenesis125,129,131. Thus, it was necessary to explore
the pathophysiological impacts of PIGN expression aberrations, due to the role that GPI-APs play
in the maintenance of cellular structure and protection, signal transduction, and enzymatic
biological processes184. In fact, GPI-AP loss has been proposed as a predictor of progression and risk of leukemic transformation, and about 42% of high-risk MDS patients harbor a small population of GPI-anchor-deficient granulocytes125,129,206. PIGN may be the first GPI-AP biosynthesis enzyme to be mechanistically linked to the regulation of the SAC and leukemia progression. The role of PIGN in leukemogenesis had not been reported until now. However previous studies have associated germline mutations in PIGN with mental retardation and multiple developmental defects193,300,301. In fact, proper chromosome segregation during meiotic cell division
is vital to normal embryonic development302,303. As such, it makes sense that germline mutations in
PIGN could be implicated in congenital malformations and mental retardation. Interestingly, a couple of studies have reported concurrent mental retardation or developmental defects and hematological malignancies including MDS304,305. However, both studies did not specifically report
any data related to PIGN expression or mutational profile. Interestingly, one of such studies
reported a patient with Jacobsen syndrome with a 18q deletion which may indicate PIGN loss because PIGN is located at the 18q21.33 locus305. Thus, PIGN mutations could have implications for embryonic development and leukemogenesis. Future studies could focus on elucidating the specific role that GPI-AP biosynthesis plays in SAC regulation and CIN suppression.
112 PIGN involvement in SAC regulation and mitotic exit
Previous studies related to SAC regulation focused on proteins with functional links to cell cycle progression. This study for the first time demonstrated that PIGN physically interacted with
SAC-related proteins and further revealed that PIGN suppression or loss altered the expression of
SAC-related proteins. PIGN ablation was marked by an increased frequency of mitotic
missegregation and altered expression of SAC-related proteins including MAD1, MAD2, BUBR1,
and MPS1. Despite this discovery of PIGN as an interacting partner and potential regulator of the
SAC, the exact manner in which this regulation occurs at multiple levels either transcriptionally or
translationally remains an enigma. Current knowledge is limited with regards to the transcriptional
regulation of SAC-related genes during mitotic progression. However, this opens up the possibility
of PIGN to be studied as a tumor suppressor because its loss significantly altered the expression of some SAC members. Altered expression of SAC-related genes have been reported in the literature and are associated with aneuploidy across multiple human cancers and more specifically in AML despite the fact that mutations in SAC-related genes are not frequent248,306. Regardless, the effect
of PIGN ablation or suppression on the expression of SAC-related gene and protein expression may
point to a regulatory interaction between PIGN and the SAC. This regulation may occur indirectly
via common factors involved in the transcriptional regulation of gene expression or the stability of
these proteins307–309. Optimal expression, timely activation, modulation and silencing of the SAC-
related factors are very critical for the efficiency, and the stepwise time-bound precision of cell
cycle progression255,310. Moreover, the cell cycle-regulated expression and transcriptional regulation of these SAC-related proteins help to maintain balanced levels of these proteins during the cell cycle311. For this reason, altered expression of SAC-related factors as observed in this
research study adversely impacts SAC function promotes CIN and ultimately contributes to leukemia progression.
113 Currently, the SAC or cell cycle-related transcription factors that are regulated by PIGN
are unknown. Moreover, the possibility of the translocation of PIGN to the nucleus for the
regulation the transcription of cell cycle or SAC-related genes is confounded by the fact that the
nuclear membrane breaks down during mitosis312,313. However, the transcription factors, NF-Y,
E2F, c-Myc, and FoxM1 have been linked to the cell cycle-regulated transcription of mitotic genes including CDC25A, CENPA, CENPB, PLK1, BUBR1, AURKA, AURKB, CCNB1, and CCNB2314–
319. Moreover, E2F transcriptionally regulates the genes encoding the core SAC component MAD2 and the APC/C activator CDC20320,321. Interestingly, PLK1, the serine/threonine kinase, phosphorylates its transcription factor and proto-oncoprotein FoxM1 in a positive-feedback loop to
regulate its transcription322. However, other SAC proteins such as BUB3 have been shown to be regulated by forming a complex with histone deacetylases to repress their expression323. These
findings reveal the complexity of the regulatory mechanisms involved in SAC regulation and the
importance of transcriptional regulation in SAC induction.
PIGN involvement downstream of the SAC signaling pathway remains elusive and leaves
several important questions to answer: Is PIGN loss one of many paths that lead to CIN? If so, is
PIGN involved in other associated pathways that cross-talk with the SAC pathway or interact and
regulate other proteins downstream of the SAC signaling pathway? Our preliminary evidence in
multiple cells demonstrated the potential for PIGN to regulate some of the downstream mediators
of the SAC signaling pathway including BUB1, AURKA, AURKB, p31 COMET, CDC20, BUB3,
PLK1, and CENPE (refer to Appendix E). Although differential effects were observed in different
cells with various TP53 mutational backgrounds, the results of our preliminary studies suggested
that PIGN may be involved in the transcriptional regulation of the expression of SAC-related genes.
It is unknown whether PIGN ablation impacts transcription factors involved in the transcriptional
regulation of these SAC-related genes.
114 Tumor suppressors and proto-oncogenes are responsible for the transcriptional regulation
of SAC-related genes321. The loss of tumor suppressors like TP53, Brd4, and BRCA1 have been
associated with the altered expression of SAC-related genes, genomic instability and ultimately
tumor progression324325326. TP53 has been shown in multiple studies to transcriptionally regulate
the genes encoding MAD1, CDC20, cyclin A1 and cyclin B327–330. Moreover, the tumor
suppressors, Rb and VHL transcriptionally regulate the gene expression of the core SAC
component MAD2112,321,331. Thus, PIGN could indirectly influence the transcriptional regulation of these SAC-related genes but, the impact of PIGN on these intermediary transcription factors would
need to be thoroughly investigated. Also, the regulatory role that PIGN plays in the transcription of some of these genes involved in SAC regulation or cell cycle progression remains unclear.
On the other hand, the interaction of PIGN with the SAC-related MAD1 indicated a more direct effect related to the stability of these SAC-related proteins with MAD1 being a limiting factor in SAC activation260,282,286,332. Moreover, during mitosis, the ER membrane absorbs the nuclear membrane and dissolves into the cytosol which could explain the interaction between PIGN and
MAD1 during mitosis despite the two proteins occupying separate cellular compartments174,312,313.
Thus, nuclear proteins that are not normally localized at the ER may typically show up in the ER
or cytosol during mitosis. However, it remains unclear how and where the interactions between
PIGN and MAD1 occurs whether at the kinetochore, cytosol or the ER. Regardless, the mitotic
interaction between MAD1 and PIGN seemed to involve MAD2. Based on the findings of this
research study that the suppression of MAD1 and more so MAD2 follows PIGN ablation, it is likely
that PIGN stabilizes the interaction between MAD1 and MAD2 in the formation of the SAC but
this interpretation needs to be experimentally validated. Nonetheless, this study revealed that PIGN
maintains chromosomal stability, by interacting with and potentially regulating the integrity of the
SAC. Overall, PIGN gene expression aberration during leukemogenesis weakens genome stability
by altering the interaction between PIGN and the SAC (Figure 4-1).
115
Figure 4-1 This simplified model summarizes the relationship between PIGN expression
aberration, SAC regulation, and leukemia progression.
PIGN depletion results in SAC dysregulation and ultimately results in missegregation errors and
aneuploidy which contribute to leukemia progression. PIGN plays a vital role in the regulation of mitotic integrity to maintain chromosomal stability.
116 PIGN interaction with the SAC in space and time
This study uncovered a novel mechanism that links PIGN loss to CIN. PIGN gene
expression aberration was attributed to partial intron retention between exons 14 and 15 that
resulted in premature translational termination and protein truncation. However, this endogenous
truncated protein product could not be detected by Western blot analyses despite the capability of
the antibody to target that region. This may be the result of the truncated protein undergoing
proteasomal degradation, although this needs to be confirmed by the incorporation proteasome
inhibitors into future experiments. Regardless, ectopic overexpression in CRISPR/Cas9 PIGN
knockout cells showed that this mutant form of PIGN had a diffuse localization pattern and had
little to no co-localization with the SAC via MAD1 (Appendix F). It is, however, yet to be
determined whether the loss of the C-terminal region resulting from the partial intron retention is
relevant for the interaction of PIGN with the SAC. HA-tag pull-down assays could be performed
using the overexpressed partial intron-retaining mutant form of the full-length HA-PIGN in
CRISPR/Cas9 PIGN knockout cells. No PIGN-SAC protein interactions will be observed if the C- terminus that is lost in the truncated mutant is crucial for the interaction between PIGN and the
SAC proteins. Alternatively, if interactions are observed, it could mean that the N-terminus of the truncated protein is essential for interaction with the SAC. Moreover, PIGN may serve as a scaffold protein for multiple binding molecules of the SAC.
Effects of PIGN-imposed post-translational modifications on SAC
Post-translational modifications such as the ubiquitination of multiple SAC-related substrates by APC/C is very crucial for normal and proper chromosomal segregation333. Also, the
phosphorylation by the mitotic cyclin-dependent kinase, CDK1 plays a vital role in the activation
117 of some of these SAC proteins334. Another level of complexity in SAC regulation involves
phosphorylation283,335. SAC-related kinases such as AURKA, AURKB, and PLK1 are involved in
phosphorylation-mediated cell cycle regulation and have been marked for targeted cancer
therapeutics334,336. The phosphorylation of BUB1 mediates the recruitment of MAD1 to the kinetochore by MPS1337. However, PIGN is a phosphoethanolamine (EtNP) transferase and
possesses a predicted Type I phosphodiesterase transferase domain338. The EtNP moiety transferred
by PIGN to the first mannose residue during GPI-anchor biosynthesis is involved in phospholipid
turnover175,339,340. Interestingly, synthetic EtNP has been shown to possess anti-tumor effects via apoptosis induction in leukemia cells341. PIGN may directly or indirectly regulate mitotic fidelity
via its interaction with and post-translational modification of SAC proteins. Moreover, PIGN may
impact kinetochore integrity by conferring modifications onto some of these SAC proteins. Thus,
the utility of PIGN as an enzyme in the regulation of the SAC and ultimately kinetochore integrity
needs to be further elucidated.
Could SAC proteins potentially become GPI-anchored because of interaction with PIGN?
Could PIGN enzyme activity as an EtNP transferase be involved in stabilizing SAC proteins or in
some cases mark them for degradation? Proteins associated with CIN or SAC function could be
initially examined for features (i.e., threonine and asparagine residues) related to the likelihood of
the transfer of EtNP by PIGN 338,342. These features might also be associated with GPI-anchor signal sequences or the presence of mannose residues343. Such studies would provide insight into the regulatory role that PIGN plays in conferring of activating or deactivating post-translational modifications to the SAC proteins we considered in this current study. The GPI-anchored mitotic
checkpoint protein, GPC1 which regulates mitotic exit by inhibiting the anaphase-promoting complex (APC/C)is a good candidate344. Preliminary gene expression analyses showed a potential
regulatory relationship between PIGN and GPC1, but this needs to be further validated at the
118 protein level (Appendix E). The possibility for some of these SAC proteins to be GPI-anchored
could be assessed by searching for potential GPI-modification sites using the GPI prediction server
big-PI Predictor or GPI-SOM343,345. Moreover, an in vitro cell-free assay could be used to dissect the post-translational modifications that PIGN may confer on SAC components and to determine how these changes could impact the proteins associated with the SAC. A collision-induced and electron transfer dissociation mass spectrometry in vitro cell-free assay incorporating recombinant
PIGN and synthetic EtNP could be used to assess the ability of the PIGN as an enzyme to modify mitotic spindle assembly proteins post-translationally similar to a study conducted by Cullen et al.338.
Translational Application
Accurate prognostication of MDS transformation to AML is vital to both patients and
physicians alike because the clinical outcomes for MDS patients are variable even in diseases
within the same sub-class42. The MDS patient needs to know his or her expected lifespan and impact
on quality of life while the physician needs to know the therapeutic approach which will be ideal
for the treatment of the patient. In this research project, the mechanistic role of PIGN in leukemia
transformation and progression was unveiled. This study revealed a novel mechanism by which
PIGN regulates MDS transformation and leukemia progression via the regulation of the SAC. Thus,
the utility of PIGN as a molecular biomarker of MDS leukemic transformation or progression is
promising. The data accumulated in this research project provide a basis for the reassessment of
MDS risk-stratification. Ultimately, the findings from this research project will help improve MDS patient risk-stratification and aid in the development of novel therapeutic strategies to prevent MDS transformation and leukemia progression.
119 Study Limitations
PIGN gene expression aberration is not always linked to complex karyotype
PIGN gene expression aberrations contributed to a high frequency of GPI-AP deficiency,
which was observed in leukemic CFCs. However, a complex karyotype was seen in almost all cells
analyzed. It may be contested that, all the cells with a complex karyotype should be GPI-AP deficient, but this is not the case because PIGN gene expression aberrations or loss might not be the sole cause of genomic instability. Heilig et al. had earlier reported that CIN was associated with risk of MDS transformation and disease progression regardless of the cytogenetic risk group of the patients114. Thus, a complex karyotype might not necessarily directly correlate with PIGN gene
expression aberrations.
Limited cohort of treatment naïve MDS/AML patients
The conclusions of this research project, are derived from the study of a relatively small cohort of patients and controls despite the significant statistical power [Power (1-β err prob) =
0.924] of the number of subjects and controls that were recruited for this study. Moreover, the
variation in PIGN gene expression within this cohort was an indicator of the inherent heterogeneity
and the mechanistic diversity associated with the pathogenesis of MDS. Sadly, this might
eventually limit the immediate diagnostic utility of the findings of this study. Moreover, the study
should have incorporated a cohort of MDS/AML patients that did not progress to AML as an additional control group. The Penn State Cancer Institute admits about 70 MDS patients over a period of 2 years, some of whom may not consent to be enrolled in such a study. Thus, patient samples are limited and as such not all observations can be confirmed in all the patients. Some of the more fundamental observations need to be confirmed in a larger cohort of patients from multiple
120 institutions. More patients are being recruited to evaluate PIGN expression aberrations in a larger
cohort using multiple methods including RT-qPCR and RNA-seq to detect partial intron retention.
The possibility of cryptic splice site mutations
Splice site and spliceosome-related gene mutations in the genomic DNA of patients with
the PIGN partial intron retention mutation were initially examined but yielded no positive results.
Moreover, publicly accessible TCGA AML patient data were analyzed to identify mutations that
could be linked to this partial intron retention, but none were identified. Thus, there might be cryptic
splice site mutations that contribute to the occurrence of these PIGN partial intron retention in
MDS/AML patients346. Moreover, the Sanger sequencing approach employed in this study may
have lacked the sensitivity to detect smaller clones of leukemic cells with PIGN partial intron
retention mutations. An alternative approach will be to adopt a gene-specific and or single-cell
RNA-seq platform which from a practical standpoint is not widely available and very expensive.
However, this would better answer the question of whether this acquired mutation was directly
involved in leukemic transformation.
Chemical inhibition of PIGN activity
Chemical inhibition of PIGN activity by YW3548/BE49385A which has been shown to
increase septation index in yeast cells could have been incorporated into this study175.
Unfortunately, tireless attempts to secure this compound proved futile. Instead, CRISPR/Cas9
PIGN Knockouts in CD34+, K562, and HEK293 cells were used to model PIGN loss. There is
cross-talk between multiple pathways, and as such, it was challenging to demonstrate that the loss
of PIGN was sufficient to induce CIN. Other disrupted genes could be equally capable of CIN
121 induction while others may only lead to CIN in the context of other gene mutations. For these
reasons, it remains unclear whether PIGN mutation is a driver or passenger. Nonetheless, the effects
observed in the PIGN CRISPR/Cas9 Knockout in normal CD34+ mononuclear cells were
remarkable due to the very low probability for the presence background mutations.
Exclusion of non-structural CIN-related effects
The lack of consideration for non-structural changes associated with CIN was a major
limitation in this study. Thus, only one dimension of CIN induction via PIGN depletion was
examined in this research project. However, in subsequent studies, array CGH and high sensitivity
SNP analyses could be used to detect alterations in the chromosomal number and LOH respectively.
Regions of structural chromosomal losses or gains based on whole chromosome copy numbers should also be examined. Karyotype complexity should be scored as the sum of all whole- chromosome losses and gains. Any regional loss above one megabase could be excluded to avoid overestimation of structural aberrations.
Investigation of limited number of SAC-related proteins
Proteins with SAC-related functions such as EB1, Survivin, HEC1, MCAK, and INCENP are overexpressed and have been linked to CIN-associated features in multiple cancers347.
Moreover, there are at least 70 genes and proteins that have been associated with CIN, cell cycle
regulation, SAC and leukemia progression115,248,348. Thus, the scope of this study may have been limited. Preliminary data generated in multiple cells as part of this project suggested that PIGN
may regulate other SAC or CIN-related factors at various degrees (refer to Appendix E). Based on
122 these preliminary data, this study could be expanded to include other proteins that have been
associated with the mitotic checkpoint and cell cycle regulation.
CONCLUSION
PIGN gene expression aberration was associated with disease progression in a subgroup of
MDS and AML-MRC patients. Primary patient data and cell line models were employed to demonstrate that, PIGN interacts with the SAC to suppress CIN and may prevent MDS transformation. The findings of this project have therapeutic implications for MDS patient risk- stratification and potentially, CIN-targeted therapy.
123 APPENDICES
Appendix A Permissions
Contents in Chapters 1, 2 and 3 have been previously published as part of a US Patent application
and in Oncotarget.
No permission was required due to the publisher's open-access license:
“Oncotarget applies the Creative Commons Attribution 3.0 License (CC BY 3.0) to all works we publish (read the human-readable summary or the full license legal code). Under the CC BY, authors retain ownership of the copyright for their article, but authors allow anyone to download, reuse, reprint, modify, distribute, and/or copy articles in Oncotarget, so long as the original authors and source are cited.”
124
125
126 Figure 1-1 was reprinted by permission from the American Society of Hematology.
Please find a copy of the permission below.
127 Figure 1-7 was reprinted by permission from Elsevier. Please find a copy of the license below.
License Number 4165700995051 License date Aug 10, 2017 Licensed Content Publisher Elsevier Licensed Content Biochimica et Biophysica Acta (BBA) - Molecular and Cell Publication Biology of Lipids Licensed Content Title GPI-anchor remodeling: Potential functions of GPI-anchors in intracellular trafficking and membrane dynamics Licensed Content Author Morihisa Fujita,Taroh Kinoshita Licensed Content Date Aug 1, 2012 Licensed Content Volume 1821 Licensed Content Issue 8 Licensed Content Pages 9 Start Page 1050 End Page 1058 Type of Use reuse in a thesis/dissertation Portion figures/tables/illustrations Number of 1 figures/tables/illustrations Format electronic Are you the author of this No Elsevier article? Will you be translating? No Original figure numbers Figure 1 Title of your THE NOVEL MECHANISTIC ROLE OF PIGN IN LEUKEMIA thesis/dissertation PROGRESSION Expected completion date Aug 2017 Estimated size (number of 200 pages) Requestor Location EMMANUEL TEYE 7883 Clearfied St
HARRISBURG, PA 17111 United States Attn: EMMANUEL TEYE Publisher Tax ID 98-0397604 Total 0.00 USD
128 Appendix B Supplemental Tables
Table B1 Samples with partial intron retentions between exons 14 and 15
# Sample Description Dx Status 1 Leukemic sorted M2 cells clone #2 AML-MRC 2 MDS-L MDS transformed to AML 3 M3 AML-MRC 4 Leukemic sorted M2 cells clone #1 AML-MRC 5 AML003 Dx AML a 6 AML139Dx AML a 7 AML028 Dx AML a 8 AML059 Dx AML a 9 AML125 Dx AML a 10 AML 117 Dx AML a 11 AML103 Rel relapsed AML a 12 AML128 Dx AML a 13 AML 051 Dx AML a 14 AML107 Rel relapsed AML a 15 AML117 Rel relapsed AML a a. RNA-seq junction file data analyses from dbGAP phs001027.v1.p1
Table B2 A Random Forest algorithm was used to predict the following classifications of MDS disease subgroups using the CIN-70 gene panel as markers. The OOB estimate of error rate was 42.42%. Thus, it makes sense to consider the CIN-70 panel important because the data is good at predicting the various disease classes of MDS (GSE4619).
Normal RA RAEB1 RAEB2 RARS Class Error Normal 8 1 1 0 1 0.2727273 RA 2 10 4 1 1 04444444 RAEB1 1 5 13 0 0 0.3157895 RAEB2 0 4 2 2 1 0.7777778 RARS 0 2 1 1 5 0.44444444
Table B3 A Random Forest algorithm was used to predict the following classifications of MDS disease subgroups using GPI-AP biosynthesis genes as markers. The OOB estimate of error rate was 46.97%. Thus, it makes sense to consider the GPI-AP biosynthesis gene panel as important because the data is good at predicting the various disease classes of MDS (GSE4619).
Normal RA RAEB1 RAEB2 RARS Class Error Normal 5 4 1 0 1 0.5454545 RA 3 11 4 0 0 0.3888889 RAEB1 0 3 16 0 0 0.1578947 RAEB2 0 6 2 0 1 1 RARS 0 4 2 0 3 0.6666667
129 Appendix C Primer Sequences
Table C1 Primers covering the coding sequence of PIGN
Gene Exona Forward Primer 5’-3’ Reverse Primer 5’-3’ (bp) Product Position PIGN 3-8 TTCTTCGCCTCCATCTTTGACA GAACCCCAGTCTGTCATTCCAT 767 414-1180 8-16 ATGGAATGACAGACTGGGGTTC GACAGGCTTGAATCAGCAGAAA 726 1159-1884 16-19 AAGCCATCTCCTGCCTTGTA AGATGTCTGGCTTTCGACCT 413 1811-2223 19-29 TGTAGGTCGAAAGCCAGACATC AGCTGCCATAATCCTTGACCAA 824 2201-3024 a Exon and product positions are based on PIGN mRNA transcript variant 2 NM_012327.5
Table C2 RT-qPCR primer sequences used to compare gene expression in AML patient during leukemia and remission
Gene Forward primer 5’-3’ Reverse primer 5’-3’ SIRT1 GGGGCTGCGGTTCCTACTG GTCCAGTCACTAGAGCTTGCAT H2AX GTCGTGCTTCACCGGTCTAC TCAGCGGTGAGGTACTCCAG DR5 CTCCTTTTCTGCTTGCGCTG TGATGCCTGATTCTTTGTGGAC BAXα CGGGGACGAACTGGACAGTAAC GCGTCCCAAAGTAGGAGAGGA SAE2 AGCTGCCCGAAACCATGTTA TCTGGGTCGGCTTAGGATGA p21 AGAAGAAGTCTGCTGGTCACAGCG CTGGACCTTTCCGGGCCGTG
Table C3 RT-qPCR primer sequences used to examine gene expression of SAC-related genes
Gene Forward primer 5’-3’ Reverse primer 5’-3’ CDC20 GACCACTCCTAGCAAACCTGG GGGCGTCTGGCTGTTTTCA BUB1 AGCCCAGACAGTAACAGACTC GTTGGCAACCTTATGTGTTTCAC BUBR1 GCACCGACAATTCCAAGCTC TGTGCTTCGTTGTGGTACAGA BUB3 GGACCCATGATGCCCCTATC CCCAGCATTACAAGGAGTTCTG CMT2/p31comet AGCATATCATGTATCAACGCCAG CCAGGGCTTGTTGGCATTTC PLK1 CCTGCACCGAAACCGAGTTAT CCGTCATATTCGACTTTGGTTGC AURKA GGAATATGCACCACTTGGAACA TAAGACAGGGCATTTGCCAAT AURKB CAGAAGAGCTGCACATTTGACG CCTTGAGCCCTAAGAGCAGATTT MPS1/TTK GTGGAGCAGTACCACTAGAAATG CCCAAGTGAACCGGAAAATGA CENPE GATTCTGCCATACAAGGCTACAA TGCCCTGGGTATAACTCCCAA ZW10 AGCTGATTGTATGGAAGTTCCCA TCTTTGTGCGATTGTTCAGTGT ROD/KNTC1 GAGAGAAGTGGCAACCTACATC CGCCGATTTTCATCGTTAGCTTT ZWILCH CACAGCATCACAAACTGCGAT CCATTCAGTGACACCAGTCCT HEC1/NDC80 TCAAGGACCCGAGACCACTTA GGGAGCTTGTAGAGATTTCATGG GPC1 TGAAGCTGGTCTACTGTGCTC CCCAGAACTTGTCGGTGATGA CCNA2 TGGAAAGCAAACAGTAAACAGCC GGGCATCTTCACGCTCTATTT CCNB1 AATAAGGCGAAGATCAACATGGC TTTGTTACCAATGTCCCCAAGAG CENPF CTCTCCCGTCAACAGCGTTC GTTGTGCATATTCTTGGCTTGC CDK1 GGATGTGCTTATGCAGGATTCC CATGTACTGACCAGGAGGGATAG ANAPC1 AAGGGCAACGATGATTGCAG TGCACCATCAGAAGACCATAAC
130 Appendix D Cell cycle frequency post-PIGN knockout
Figure D1 Cell cycle frequency (1/day) is higher in PIGN Knockout cells than in PIGN WT
HEK293 cells. Mean cell counts were obtained over a period of 3-days at 12-hour intervals. The
/ frequency of cell cycles was calculated using the formula f = × derived from the ln Nt No 1 tf � ln 2 � t formula Nt = N0 2 where Nt is the number of cells at time t, N0 is the initial number of cells and f
is the frequency of cell cycles per unit time. M. Beals, L. Gross, S. Harrell. 1999. Quantifying cell division. http://www.tiem.utk.edu/~gross/bioed/webmodules/celldivision.html. 7/12/17 12:27AM.
131
Appendix E RT-qPCR profile post-PIGN knockout
Figure E1 RT-qPCR data of the expression of SAC-related genes following PIGN ablation by
CRISPR/Cas9. A. HEK293 B. CD34+ mononuclear cells c. HL60 cells. Error bars represent standard deviation from the mean fold change in gene expression relative to gene expression in the wild-type (WT) cells.
132 Appendix F Co-localization analyses of mutant PIGN
Figure F1 Partial intron retention between exons 14 and 15 in PIGN results in diffuse localization
patttern of PIGN and disrupts the co-localization of PIGN with MAD1.
Intracellular co-localization of PIGN (green) and endogenous MAD1(red) during SAC activation in HEK293 cells. HEK293 PIGN CRISPR/Cas9 Knockout cells were transfected with pMEPuro3HAPIGN (upper panel) or mutant plasmid (Lower panel) cloned by inserting a 38bp partial intron sequence into the WT plasmid via restriction enzyme digestion and religation. The cells were incubated for 48 hours followed by treatment with nocodazole (100 ng/ul) for 12 hours.
Cells transfected with either of the two plasmids were fixed with 4% paraformaldehyde and treated with mouse anti-MAD, anti-MAD2 or anti-MPS1 and rabbit anti-HA, followed by treatment with fluorescently-labeled secondary antibodies. Chromosomes were stained with DAPI (blue). Laser scanning confocal microscopy was used to visualize the stained cells. Scale bars, 2-3 µm.
133 REFERENCES
1. Fenaux P, Ades L. How we treat lower-risk myelodysplastic syndromes. Blood.
2013;121(21):4280–4286.
2. Nowell PC. Cytogenetics of preleukemia. Cancer Genet. Cytogenet. 1982;5(3):265–278.
3. Koeffler HP. Human acute myeloid leukemia lines: models of leukemogenesis. Semin.
Hematol. 1986;23(3):223–236.
4. Janssen JW, Buschle M, Layton M, et al. Clonal analysis of myelodysplastic syndromes:
evidence of multipotent stem cell origin. Blood. 1989;73(1):248–254.
5. Cazzola M, Rossi M, Malcovati L. Biologic and clinical significance of somatic mutations
of SF3B1 in myeloid and lymphoid neoplasms. Blood. 2013;121(2):260–269.
6. Bejar R, Steensma DP. Recent developments in myelodysplastic syndromes. Blood.
2014;124(18):2793–2803.
7. Neukirchen J, Schoonen WM, Strupp C, et al. Incidence and prevalence of
myelodysplastic syndromes: Data from the Dusseldorf MDS-registry. Leuk. Res.
2011;35(12):1591–1596.
8. Rollison DE, Howlader N, Smith MT, et al. Epidemiology of myelodysplastic syndromes
and chronic myeloproliferative disorders in the United States, 2001-2004, using data from
the NAACCR and SEER programs. Blood. 2008;112(1):45–52.
9. Ma X, Does M, Raza A, Mayne ST. Myelodysplastic syndromes: Incidence and survival
in the United States. Cancer. 2007;109(8):1536–1542.
10. Mufti GJ. Pathobiology, classification, and diagnosis of myelodysplastic syndrome. Best
Pract. Res. Clin. Haematol. 2004;17(4 SPEC.ISS.):543–557.
11. Malcovati L, Nimer SD. Myelodysplastic syndromes: diagnosis and staging. Cancer
Control. 2008;15 Suppl(4):4–13.
134 12. Tefferi A, Vardiman JW. Myelodysplastic syndromes. N. Engl. J. Med.
2009;361(19):1872–1885.
13. Vardiman JW. Hematopathological concepts and controversies in the diagnosis and
classification of myelodysplastic syndromes. Hematol. Am Soc Hematol Educ Progr.
2006;199–204.
14. Genovese G, Kähler AK, Handsaker RE, et al. Clonal hematopoiesis and blood-cancer
risk inferred from blood DNA sequence. N. Engl. J. Med. 2014;371(26):2477–87.
15. Bochtler T, Fröhling S, Krämer A. Role of chromosomal aberrations in clonal diversity
and progression of acute myeloid leukemia. Leukemia. 2015;29(6):1243–1252.
16. Takahashi K, Cabrero M, Borthakur G, et al. from MDS to AML predicts for very poor
outcomes. 2015;
17. Mughal TI, Cross NCP, Padron E, et al. An international MDS/MPN working group’s
perspective and recommendations on molecular pathogenesis, diagnosis and clinical
characterization of myelodysplastic/myeloproliferative neoplasms. Haematologica.
2015;100(9):1117–1130.
18. Kawankar N, Vundinti BR. Cytogenetic abnormalities in myelodysplastic syndrome: an
overview. Hematology. 2011;16(3):131–8.
19. Santini V. Treatment of low-risk myelodysplastic syndromes. ASH Educ. Progr. B.
2016;2016(1):462–469.
20. Pellagatti A, Jadersten M, Forsblom A-M, et al. Lenalidomide inhibits the malignant clone
and up-regulates the SPARC gene mapping to the commonly deleted region in 5q-
syndrome patients. Proc. Natl. Acad. Sci. 2007;104(27):11406–11411.
21. Golshayan AR, Jin T, Maciejewski J, et al. Efficacy of growth factors compared to other
therapies for low-risk myelodysplastic syndromes. Br. J. Haematol. 2007;137(2):125–132.
22. Jädersten M, Hellström-Lindberg E. Myelodysplastic syndromes: Biology and treatment.
135 J. Intern. Med. 2009;265(3):307–328.
23. Greenberg PL. Current therapeutic approaches for patients with myelodysplastic
syndromes. Br. J. Haematol. 2010;150(2):131–143.
24. List AF, Baker AF, Green S, Bellamy W. Lenalidomide: Targeted anemia therapy for
myelodysplastic syndromes. Cancer Control. 2006;13(SUPPL.):4–11.
25. Komrokji RS, List AF. Role of lenalidomide in the treatment of myelodysplastic
syndromes. Semin. Oncol. 2011;38(5):648–657.
26. Sekeres MA, Tiu R V., Komrokji R, et al. Phase 2 study of the lenalidomide and
azacitidine combination in patients with higher-risk myelodysplastic syndromes. Blood.
2012;120(25):4945–4951.
27. Visconte V, Tiu R V., Rogers HJ. Pathogenesis of myelodysplastic syndromes: An
overview of molecular and non-molecular aspects of the disease. Blood Res.
2014;49(4):216–227.
28. Dobbelstein C, Ganser A. Immunosuppressive therapy for Myelodysplastic syndromes.
Curr. Pharm. Des. 2012;18(22):3184–3189.
29. Fenaux P, Mufti GJ, Hellstrom-Lindberg E, et al. Efficacy of azacitidine compared with
that of conventional care regimens in the treatment of higher-risk myelodysplastic
syndromes: a randomised, open-label, phase III study. Lancet Oncol. 2009;10(3):223–232.
30. Welch JS, Petti AA, Miller CA, et al. TP53 and Decitabine in Acute Myeloid Leukemia
and Myelodysplastic Syndromes. N. Engl. J. Med. 2016;375(21):2023–2036.
31. M. L. DNA methylation inhibitors in the treatment of leukemias, myelodysplastic
syndromes and hemoglobinopathies: Clinical results and possible mechanisms of action.
Curr. Top. Microbiol. Immunol. 2000;249:135–164.
32. Kantarjian H, Issa JP, Rosenfeld CS, et al. Decitabine improves patient outcomes in
myelodysplastic syndromes: results of a phase III randomized study. Cancer.
136 2006;106(8):1794–1803.
33. Sekeres MA. The euphoria of hypomethylating agents in MDS and AML: Is it justified?
Best Pract. Res. Clin. Haematol. 2013;26(3):275–278.
34. Klepin HD. Myelodysplastic Syndromes and Acute Myeloid Leukemia in the Elderly.
Clin. Geriatr. Med. 2016;32(1):155–173.
35. Passweg JR, Baldomero H, Bregni M, et al. Hematopoietic SCT in Europe: data and
trends in 2011. Bone Marrow Transplant. 2013;48(9):1161–7.
36. Ballen KK, Gluckman E, Broxmeyer HE. Umbilical cord blood transplantation: the first
25 years and beyond. Blood. 2013;122(4):491–498.
37. Sekeres MA, Schoonen WM, Kantarjian H, et al. Characteristics of US patients with
myelodysplastic syndromes: Results of six cross-sectional physician surveys. J. Natl.
Cancer Inst. 2008;100(21):1542–1551.
38. de Witte T, Bowen D, Robin M, et al. Allogeneic hematopoietic stem cell transplantation
for MDS and CMML: recommendations from an international expert panel. Blood.
2017;129(13):1753–1762.
39. Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the classification of the
myelodysplastic syndromes. Br. J. Haematol. 1982;51(2):189–199.
40. Vardiman J, Hyjek E. World Health Organization Classification , Evaluation , and
Genetics of the Myeloproliferative Neoplasm Variants Introduction : WHO classification
of MPN variants. Am. Soc. Hematol. 2008;250–256.
41. Malcovati L, Porta MG Della, Strupp C, et al. Impact of the degree of anemia on the
outcome of patients with myelodysplastic syndrome and its integration into the WHO
classification-based prognostic scoring system (WPSS). Haematologica.
2011;96(10):1433–1440.
42. Bejar R. Clinical and genetic predictors of prognosis in myelodysplastic syndromes.
137 Haematologica. 2014;99(6):956–964.
43. Tiu R V., Gondek LP, O’Keefe CL, et al. Prognostic impact of SNP array karyotyping in
myelodysplastic syndromes and related myeloid malignancies. Blood.
2011;117(17):4552–4560.
44. Thol F, Yun H, Sonntag AK, et al. Prognostic significance of combined MN1, ERG,
BAALC, and EVI1 (MEBE) expression in patients with myelodysplastic syndromes. Ann.
Hematol. 2012;91(8):1221–1233.
45. Kantarjian H, O’Brien S, Ravandi F, et al. Proposal for a new risk model in
myelodysplastic syndrome that accounts for events not considered in the original
International Prognostic Scoring System. Cancer. 2008;113(6):1351–1361.
46. Greenberg PL, Tuechler H, Schanz J, et al. Revised international prognostic scoring
system for myelodysplastic syndromes. Blood. 2012;120(12):2454–2465.
47. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revision to the World Health
Organization (WHO) classification of myeloid neoplasms and acute leukemia. Blood.
2016;127(20):2391–2405.
48. Slovak ML, Dewald GW. International Working Group on MDS cytogenetics: October
2007 meeting report. Leuk. Res. 2008;32(9):1329–1332.
49. Steidl C, Steffens R, Gassmann W, et al. Adequate cytogenetic examination in
myelodysplastic syndromes: Analysis of 529 patients. Leuk. Res. 2005;29(9):987–993.
50. Shaffer LG, Ballif BC, Schultz RA. The use of cytogenetic microarrays in
myelodysplastic syndrome characterization. Methods Mol. Biol. 2013;973:69–85.
51. Haase D, Germing U, Schanz J, et al. New insights into the prognostic impact of the
karyotype in MDS and correlation with subtypes: Evidence from a core dataset of 2124
patients. Blood. 2007;110(13):4385–4395.
52. Greenberg P, Cox C, LeBeau MM, et al. International scoring system for evaluating
138 prognosis in myelodysplastic syndromes. Blood. 1997;89(6):2079–88.
53. Vardiman JW, Harris NL, Brunning RD. The World Health Organization (WHO)
classification of the myeloid neoplasms. Blood. 2002;100:2292–2302.
54. Steensma DP, Bejar R, Jaiswal S, et al. Clonal hematopoiesis of indeterminate potential
and its distinction from myelodysplastic syndromes. Blood. 2015;126(1):9–16.
55. Laurie CC, Laurie CA, Rice K, et al. Detectable clonal mosaicism from birth to old age
and its relationship to cancer. Nat. Genet. 2012;44(6):642–650.
56. Cargo CA, Rowbotham N, Evans PA, et al. Targeted sequencing identi fi es patients with
preclinical MDS at high risk of disease progression. 2015;126(21):2362–2366.
57. Jacobs KB, Yeager M, Zhou W, et al. Detectable clonal mosaicism and its relationship to
aging and cancer. Nat. Genet. 2012;44(6):651–658.
58. Holstege H, Pfeiffer W, Sie D, et al. Somatic mutations found in the healthy blood
compartment of a 115-yr-old woman demonstrate oligoclonal hematopoiesis. Genome
Res. 2014;24(5):733–742.
59. Otrock ZK, Tiu R V., Maciejewski JP, Sekeres MA. The need for additional genetic
markers for myelodysplastic syndrome stratification: what does the future hold for
prognostication? Expert Rev Hematol. 2013;6(1):59–68.
60. Bejar R, Stevenson KE, Caughey BA, et al. Validation of a prognostic model and the
impact of mutations in patients with lower-risk myelodysplastic syndromes. J. Clin.
Oncol. 2012;30(27):3376–3382.
61. Bejar R, Stevenson K, Abdel-Wahab O, et al. Clinical effect of point mutations in
myelodysplastic syndromes. N. Engl. J. Med. 2011;364(26):2496–2506.
62. Papaemmanuil E, Gerstung M, Malcovati L, et al. Clinical and biological implications of
driver mutations in myelodysplastic syndromes. Blood. 2013;122(22):3616–3627.
63. Nolte F, Hofmann WK. Myelodysplastic syndromes: Molecular pathogenesis and genomic
139 changes. Ann. Hematol. 2008;87(10):777–795.
64. Hirai H. Molecular mechanisms of myelodysplastic syndrome. Jpn. J. Clin. Oncol.
2003;33(4):153–160.
65. Look a T. Molecular Pathogenesis of MDS. Hematology Am. Soc. Hematol. Educ.
Program. 2005;156–60.
66. Harada Y, Harada H. Molecular mechanisms that produce secondary MDS/AML by
RUNX1/AML1 point mutations. J. Cell. Biochem. 2011;112(2):425–432.
67. Steensma DP, Gibbons RJ, Mesa RA, Tefferi A, Higgs DR. Somatic point mutations in
RUNX1/CBFA2/AML1 are common in high-risk myelodysplastic syndrome, but not in
myelofibrosis with myeloid metaplasia. Eur. J. Haematol. 2005;74(1):47–53.
68. Thol F, Friesen I, Damm F, et al. Prognostic significance of ASXL1 mutations in patients
with myelodysplastic syndromes. J. Clin. Oncol. 2011;29(18):2499–506.
69. Bejar R, Lord A, Stevenson K, et al. TET2 mutations predict response to hypomethylating
agents in myelodysplastic syndrome patients. Blood. 2014;124(17):2705–2712.
70. Chen CY, Lin LI, Tang JL, et al. RUNX1 gene mutation in primary myelodysplastic
syndrome - The mutation can be detected early at diagnosis or acquired during disease
progression and is associated with poor outcome. Br. J. Haematol. 2007;139(3):405–414.
71. Gelsi-Boyer V, Brecqueville M, Devillier R, et al. Mutations in ASXL1 are associated
with poor prognosis across the spectrum of malignant myeloid diseases. J. Hematol.
Oncol. 2012;5(1):12.
72. Xu F, Li X. The role of histone methyltransferase EZH2 in myelodysplastic syndromes.
Expert Rev. Hematol. 2012;5(2):177–85.
73. Walter MJ, Ding L, Shen D, et al. Recurrent DNMT3A mutations in patients with
myelodysplastic syndromes. Leukemia. 2011;25(7):1153–1158.
74. Thol F, Weissinger EM, Krauter J, et al. IDH1 mutations in patients with myelodysplastic
140 syndromes are associated with an unfavorable prognosis. Haematologica.
2010;95(10):1668–1674.
75. Kosmider O, Gelsi-Boyer V, Cheok M, et al. TET2 mutation is an independent favorable
prognostic factor in myelodysplastic syndromes (MDSs). Blood. 2009;114(15):3285–
3291.
76. Langemeijer S, Kuiper R, Berends M, et al. P052 Acquired mutations in TET2 are
common in myelodysplastic syndromes. Leuk. Res. 2009;33(SUPPL. 1):
77. Xu F, Li X, Wu L, et al. Overexpression of the EZH2, RING1 and BMI1 genes is
common in myelodysplastic syndromes: Relation to adverse epigenetic alteration and poor
prognostic scoring. Ann. Hematol. 2011;90(6):643–653.
78. Nikoloski G, Langemeijer SMC, Kuiper RP, et al. Somatic mutations of the histone
methyltransferase gene EZH2 in myelodysplastic syndromes. Nat. Genet. 2010;42(8):665–
667.
79. Thol F, Winschel C, Lüdeking A, et al. Rare occurrence of DNMT3A mutations in
myelodysplastic syndromes. Haematologica. 2011;96(12):1870–1873.
80. Chotirat S, Thongnoppakhun W, Wanachiwanawin W, Auewarakul CU. Acquired somatic
mutations of isocitrate dehydrogenases 1 and 2 (IDH1 and IDH2) in preleukemic
disorders. Blood Cells, Mol. Dis. 2015;54(3):286–291.
81. Pardanani A, Lasho TL, Finke CM, et al. IDH1 and IDH2 mutation analysis in chronic-
and blast-phase myeloproliferative neoplasms. Leukemia. 2010;24(6):1146–1151.
82. Patnaik MM, Hanson CA, Hodnefield JM, et al. Differential prognostic effect of IDH1
versus IDH2 mutations in myelodysplastic syndromes: a Mayo Clinic Study of 277
patients. Leukemia. 2012;26(1):101–105.
83. Yoshida K, Sanada M, Nagata Y, et al. Whole exome analysis of myelodysplastic
syndromes. Haematologica. 2011;96:429.
141 84. Visconte V, Makishima H, Maciejewski JP, Tiu R V. Emerging roles of the spliceosomal
machinery in myelodysplastic syndromes and other hematological disorders. Leukemia.
2012;26(12):2447–2454.
85. Papaemmanuil E. Somatic SF3B1 mutation in myelodysplasia with ring sideroblasts. N
Engl J Med. 2011;365:1384–1395.
86. Janssen A, Medema RH. Mitosis as an anti-cancer target. Oncogene. 2011;30(25):2799–
2809.
87. Geigl JB, Obenauf AC, Schwarzbraun T, Speicher MR. Defining “chromosomal
instability.” Trends Genet. 2008;24(2):64–69.
88. Vajen B, Thomay K, Schlegelberger B. Induction of chromosomal instability via telomere
dysfunction and epigenetic alterations in myeloid neoplasia. Cancers (Basel).
2013;5(3):857–874.
89. Gollin SM, Skarja SR. Mitosis: Chromosomal Rearrangements. Encycl. Life Sci. 2006;1–
5.
90. Thompson SL, Compton DA. Chromosome missegregation in human cells arises through
specific types of kinetochore-microtubule attachment errors. Proc. Natl. Acad. Sci.
2011;108(44):17974–17978.
91. Lengauer C, Kinzler KW, Vogelstein B. Genetic instabilities in human cancers. Nature.
1998;396(6712):643–649.
92. Barber TD, McManus K, Yuen KWY, et al. Chromatid cohesion defects may underlie
chromosome instability in human colorectal cancers. Proc. Natl. Acad. Sci. U. S. A.
2008;105(9):3443–8.
93. Bakhoum SF, Silkworth WT, Nardi IK, et al. The mitotic origin of chromosomal
instability. Curr. Biol. 2014;24(4):R148–R149.
94. Hanahan D, Weinberg RA. Hallmarks of cancer: The next generation. Cell.
142 2011;144(5):646–674.
95. McGranahan N, Burrell RA, Endesfelder D, Novelli MR, Swanton C. Cancer
chromosomal instability: therapeutic and diagnostic challenges. EMBO Rep.
2012;13(6):528–538.
96. Gisselsson D. Classification of chromosome segregation errors in cancer. Chromosoma.
2008;117(6):511–519.
97. Shih IM, Zhou W, Goodman SN, et al. Evidence that genetic instability occurs at an early
stage of colorectal tumorigenesis. Cancer Res. 2001;61(3):818–822.
98. Nowak MA, Komarova NL, Sengupta A, et al. The role of chromosomal instability in
tumor initiation. Proc. Natl. Acad. Sci. U. S. A. 2002;99(25):16226–31.
99. Saletta F, Matullo G, Manuguerra M, et al. Exposure to the tobacco smoke constituent 4-
aminobiphenyl induces chromosomal instability in human cancer cells. Cancer Res.
2007;67(15):7088–7094.
100. Duesberg P, Stindl R, Hehlmann R. Explaining the high mutation rates of cancer cells to
drug and multidrug resistance by chromosome reassortments that are catalyzed by
aneuploidy. Proc. Natl. Acad. Sci. 2000;97(26):14295–14300.
101. Lee AJX, Endesfelder D, Rowan AJ, et al. Chromosomal instability confers intrinsic
multidrug resistance. Cancer Res. 2011;71(5):1858–1870.
102. Bakhoum SF, Swanton C. Chromosomal Instability, Aneuploidy, and Cancer. Front.
Oncol. 2014;4:.
103. Giam M, Rancati G. Aneuploidy and chromosomal instability in cancer: a jackpot to
chaos. Cell Div. 2015;10(1):3.
104. Olney HJ, Le Beau MM. The cytogenetics of myelodysplastic syndromes. Best Pract. Res.
Clin. Haematol. 2001;14(3):479–495.
105. Alter BP, Caruso JP, Drachtman RA, et al. Fanconi anemia: Myelodysplasia as a predictor
143 of outcome. Cancer Genet. Cytogenet. 2000;117(2):125–131.
106. Mehta PA, Harris RE, Davies SM, et al. Numerical chromosomal changes and risk of
development of myelodysplastic syndrome-acute myeloid leukemia in patients with
Fanconi anemia. Cancer Genet. Cytogenet. 2010;203(2):180–186.
107. Tischkowitz M, Dokal I. Fanconi anaemia and leukaemia - Clinical and molecular aspects.
Br. J. Haematol. 2004;126(2):176–191.
108. Cerabona D, Sun Z, Nalepa G. Leukemia and chromosomal instability in aged Fancc-/-
mice. Exp. Hematol. 2016;44(5):352–357.
109. D. C, Z.A. S, E.S. P, Y. H, Z. S. Impaired spindle assembly checkpoint in vivo promotes
MDS/AML in a novel mouse model of fanconi anemia. Blood. 2015;126(23):1211.
110. Graubert T, Walter MJ. Genetics of Myelodysplastic Syndromes: New Insights.
Hematology. 2011;2011(1):543–549.
111. Sperling AS, Gibson CJ, Ebert BL. The genetics of myelodysplastic syndrome: from
clonal haematopoiesis to secondary leukaemia. Nat. Rev. Cancer. 2016;17(1):5–19.
112. Orr B, Compton DA. A double-edged sword: how oncogenes and tumor suppressor genes
can contribute to chromosomal instability. Front Oncol. 2013;3(June):164.
113. Walter MJ, Shen D, Shao J, et al. Clonal diversity of recurrently mutated genes in
myelodysplastic syndromes. Leukemia. 2013;27(6):1275–1282.
114. Heilig CE, Loffler H, Mahlknecht U, et al. Chromosomal instability correlates with poor
outcome in patients with myelodysplastic syndromes irrespectively of the cytogenetic risk
group. J. Cell. Mol. Med. 2010;14(4):895–902.
115. Mettu RKR, Wan YW, Habermann JK, Ried T, Guo NL. A 12-gene genomic instability
signature predicts clinical outcomes in multiple cancer types. Int. J. Biol. Markers.
2010;25(4):219–228.
116. Carter SL, Eklund AC, Kohane IS, Harris LN, Szallasi Z. A signature of chromosomal
144 instability inferred from gene expression profiles predicts clinical outcome in multiple
human cancers. Nat. Genet. 2006;38(9):1043–8.
117. Genga KR, Filho FDR, Ferreira FV de A, et al. Proteins of the mitotic checkpoint and
spindle are related to chromosomal instability and unfavourable prognosis in patients with
myelodysplastic syndrome. J. Clin. Pathol. 2015;68(5):381–7.
118. Makishima H, Yoshizato T, Yoshida K, et al. Dynamics of clonal evolution in
myelodysplastic syndromes. Nat. Genet. 2016;(December):
119. Martin-Cabrera P, Jeromin S, Peglorová K, et al. Acute myeloid leukemias with ring
sideroblasts show a unique molecular signature straddling secondary acute myeloid
leukemia and de novo acute myeloid leukemia. Haematologica.
2017;haematol.2016.156844.
120. Kennedy JA, Ebert BL. Clinical implications of Genetic mutations in Myelodysplastic
syndrome. J. Clin. Oncol. 2017;35(9):968–974.
121. Zhou T, Kinney MC, Scott LM, Zinkel SS, Rebel VI. Revisiting the case for genetically
engineered mouse models in human myelodysplastic syndrome research. Blood.
2015;126(9):1057–1068.
122. Beurlet S, Chomienne C, Padua RA. Engineering mouse models with myelodysplastic
syndrome human candidate genes; How relevant are they? Haematologica.
2013;98(1):10–22.
123. Byrne M, Bennett RL, Cheng X, May WS. Progressive genomic instability in the Nup98-
HoxD13 model of MDS correlates with loss of the PIG-A gene product. Neoplasia.
2014;16(8):627–33.
124. Pu JJ, Hu R, Mukhina GL, et al. The small population of PIG-A mutant cells in
myelodysplastic syndromes do not arise from multipotent hematopoietic stem cells.
Haematologica. 2012;97(8):1225–1233.
145 125. Pu, J J., Spivak J. L. , Brodsky, R. A., Moliterno A. The origin of GPI-AP deficient cells
in MDS, MPD, and aplastic anemia and its significance in predicting leukemic
transformation. J Clin Oncol. 2013;1–2.
126. Coschi CH, Dick FA. Chromosome instability and deregulated proliferation: An
unavoidable duo. Cell. Mol. Life Sci. 2012;69(12):2009–2024.
127. Walter MJ, Payton JE, Ries RE, et al. Acquired Copy Number Alterations in Adult Acute
Myeloid Leukemia Genomes. Source Proc. Natl. Acad. Sci. United States Am.
2009;106(31):12950–12955.
128. Greenman CD, Bignell G, Butler A, et al. PICNIC: An algorithm to predict absolute
allelic copy number variation with microarray cancer data. Biostatistics. 2010;11(1):164–
175.
129. Araten DJ, Martinez-Climent JA, Perle MA, et al. A quantitative analysis of genomic
instability in lymphoid and plasma cell neoplasms based on the PIG-A gene. Mutat. Res. -
Fundam. Mol. Mech. Mutagen. 2010;686(1–2):1–8.
130. Araten DJ, Golde DW, Zhang RH, et al. A quantitative measurement of the human
somatic mutation rate. Cancer Res. 2005;65(18):8111–8117.
131. Peruzzi B, Araten DJ, Notaro R, Luzzatto L. The use of PIG-A as a sentinel gene for the
study of the somatic mutation rate and of mutagenic agents in vivo. Mutat. Res. - Rev.
Mutat. Res. 2010;705(1):3–10.
132. Solé F, Espinet B, Sanz GF, et al. Incidence, characterization and prognostic significance
of chromosomal abnormalities in 640 patients with primary myelodysplastic syndromes.
Br. J. Haematol. 2000;108(2):346–356.
133. Bakhoum SF, Danilova O V., Kaur P, Levy NB, Compton DA. Chromosomal instability
substantiates poor prognosis in patients with diffuse large B-cell lymphoma. Clin. Cancer
Res. 2011;17(24):7704–7711.
146 134. Merkel DE, Dressler LG, McGuire WL. Flow cytometry, cellular DNA content, and
prognosis in human malignancy. J. Clin. Oncol. 1987;5:1690–1703.
135. Fiegler H, Redon R, Carter NP. Construction and use of spotted large-insert clone DNA
microarrays for the detection of genomic copy number changes. Nat. Protoc.
2007;2(3):577–587.
136. H. F, J.B. G, S. L, et al. High resolution array-CGH analysis of single cells. Nucleic Acids
Res. 2007;35(3):no pagination.
137. Pinkel D, Albertson DG. Array comparative genomic hybridization and its applications in
cancer. Nat. Genet. 2005;37(6s):S11–S17.
138. Stephens PJ, Greenman CD, Fu B, et al. Massive genomic rearrangement acquired in a
single catastrophic event during cancer development. Cell. 2011;144(1):27–40.
139. Roschke A V., Tonon G, Gehlhaus KS, et al. Karyotypic Complexity of the NCI-60 Drug-
Screening Panel. Cancer Res. 2003;63(24):8634–8647.
140. Tanaka K, Hirota T. Chromosomal instability: A common feature and a therapeutic target
of cancer. Biochim. Biophys. Acta - Rev. Cancer. 2016;1866(1):64–75.
141. Janssen A, Kops GJ, Medema RH. Targeting the Mitotic Checkpoint to Kill Tumor Cells.
Horm. Cancer. 2011;2(2):113–116.
142. Kops GJPL, Foltz DR, Cleveland DW. Lethality to human cancer cells through massive
chromosome loss by inhibition of the mitotic checkpoint. Proc. Natl. Acad. Sci.
2004;101(23):8699–8704.
143. Tang YC, Williams BR, Siegel JJ, Amon A. Identification of aneuploidy-selective
antiproliferation compounds. Cell. 2011;144(4):499–512.
144. Solit DB, Rosen N. Hsp90: a novel target for cancer therapy. Curr. Top. Med. Chem.
2006;6(11):1205–14.
145. Lampert F, Westermann S. A blueprint for kinetochores - new insights into the molecular
147 mechanics of cell division. Nat. Rev. Mol. Cell Biol. 2011;12(7):407–412.
146. Jia L, Kim S, Yu H. Tracking spindle checkpoint signals from kinetochores to APC/C.
Trends Biochem. Sci. 2013;38(6):302–311.
147. Lawrence KS, Engebrecht J. The spindle assembly checkpoint: More than just keeping
track of the spindle. Trends cell Mol. Biol. 2015;10:141–150.
148. Burrell RA, McClelland SE, Endesfelder D, et al. Replication stress links structural and
numerical cancer chromosomal instability. Nature. 2013;494(7438):492–496.
149. Hollstein M, Sidransky D, Vogelstein B, Harris CC. P53 Mutations in Human Cancers.
Science (80-. ). 1991;253(5015):49–53.
150. Hoeijmakers JH. Genome maintenance mechanisms for preventing cancer. Nature.
2001;411(6835):366–74.
151. Hoeijmakers JHJ. DNA repair mechanisms. Maturitas. 2001;38(1):17–22.
152. Dereli-Öz A, Versini G, Halazonetis TD. Studies of genomic copy number changes in
human cancers reveal signatures of DNA replication stress. Mol. Oncol. 2011;5(4):308–
314.
153. Bunz F, Fauth C, Speicher MR, et al. Targeted inactivation of p53 in human cells does not
result in aneuploidy. Cancer Res. 2002;62(4):1129–1133.
154. Thompson SL, Bakhoum SF, Compton DA. Mechanisms of Chromosomal Instability.
Curr. Biol. 2010;20(6):.
155. Kurinna S, Stratton SA, Coban Z, et al. p53 regulates a mitotic transcription program and
determines ploidy in normal mouse liver. Hepatology. 2013;57(5):2004–2013.
156. Ohashi A, Ohori M, Iwai K, et al. Aneuploidy generates proteotoxic stress and DNA
damage concurrently with p53-mediated post-mitotic apoptosis in SAC-impaired cells.
Nat. Commun. 2015;6:7668.
157. Zachos G, Black EJ, Walker M, et al. Chk1 Is Required for Spindle Checkpoint Function.
148 Dev. Cell. 2007;12(2):247–260.
158. Chilà R, Celenza C, Lupi M, Damia G, Carrassa L. Chk1-Mad2 interaction: A crosslink
between the DNA damage checkpoint and the mitotic spindle checkpoint. Cell Cycle.
2013;12(7):1083–1090.
159. Burke DJ, Stukenberg PT. Linking Kinetochore-Microtubule Binding to the Spindle
Checkpoint. Dev. Cell. 2008;14(4):474–479.
160. Carrassa L, Sanchez Y, Erba E, Damia G. U2OS cells lacking Chk1 undergo aberrant
mitosis and fail to activate the spindle checkpoint. J. Cell. Mol. Med. 2009;13(8 A):1565–
1576.
161. Peddibhotla S, Rosen JM. Chking and executing cell division to prevent genomic
instability. Cell Cycle. 2009;8(15):2339–2342.
162. Stolz A, Ertych N, Bastians H. Tumor suppressor CHK2: Regulator of DNA damage
response and mediator of chromosomal stability. Clin. Cancer Res. 2011;17(3):401–405.
163. Stolz A, Ertych N, Bastians H. Loss of the tumour-suppressor genes CHK2 and BRCA1
results in chromosomal instability. Biochem. Soc. Trans. 2010;38(6):1704–8.
164. Shang ZF, Huang B, Xu QZ, et al. Inactivation of DNA-dependent protein kinase leads to
spindle disruption and mitotic catastrophe with attenuated checkpoint protein 2
phosphorylation in response to DNA damage. Cancer Res. 2010;70(9):3657–3666.
165. Janssen A, van der Burg M, Szuhai K, Kops GJPL, Medema RH. Chromosome
Segregation Errors as a Cause of DNA Damage and Structural Chromosome Aberrations.
Science (80-. ). 2011;333(6051):1895–1898.
166. Hoffelder D, Luo L, Burke N, et al. Resolution of anaphase bridges in cancer cells.
Chromosoma. 2004;112(8):.
167. Crasta K, Ganem NJ, Dagher R, et al. DNA breaks and chromosome pulverization from
errors in mitosis. Nature. 2012;482(7383):53–58.
149 168. Mills KD, Ferguson DO, Alt FW. The role of DNA breaks in genomic instability and
tumorigenesis. Immunol. Rev. 2003;194:77–95.
169. Bayani J, Selvarajah S, Maire G, et al. Genomic mechanisms and measurement of
structural and numerical instability in cancer cells. Semin. Cancer Biol. 2007;17(1):5–18.
170. Heredia FF, De Sousa JC, Junior HLR, et al. Proteins related to the spindle and checkpoint
mitotic emphasize the different pathogenesis of hypoplastic MDS. Leuk. Res.
2014;38(2):218–224.
171. de Oliveira FM, Lucena-Araujo AR, Favarin M do C, et al. Differential expression of
AURKA and AURKB genes in bone marrow stromal mesenchymal cells of
myelodysplastic syndrome: Correlation with G-banding analysis and FISH. Exp. Hematol.
2013;41(2):198–208.
172. Hofmann WK, de Vos S, Komor M, et al. Characterization of gene expression of CD34+
cells from normal and myelodysplastic bone marrow. Blood. 2002;100(10):3553–3560.
173. Rodriguez-Bravo V, Maciejowski J, Corona J, et al. Nuclear pores protect genome
integrity by assembling a premitotic and mad1-dependent anaphase inhibitor. Cell.
2014;156(5):1017–1031.
174. Buch K, Collin P, Rodriguez-bravo V, et al. Nuclear Pores Protect Genome Integrity by
Assembling a Premitotic and Mad1-Dependent Anaphase Inhibitor. 2014;2:1017–1031.
175. Hong Y, Maeda Y, Watanabe R, et al. Pig-n, a mammalian homologue of yeast Mcd4p, is
involved in transferring phosphoethanolamine to the first mannose of the
glycosylphosphatidylinositol. J. Biol. Chem. 1999;274(49):35099–35106.
176. Aken BL, Ayling S, Barrell D, et al. The Ensembl gene annotation system. Database.
2016;2016:baw093.
177. Mrózek K. Cytogenetic, Molecular Genetic, and Clinical Characteristics of Acute Myeloid
Leukemia With a Complex Karyotype. Semin. Oncol. 2008;35(4):365–377.
150 178. Fujita M, Kinoshita T. GPI-anchor remodeling: Potential functions of GPI-anchors in
intracellular trafficking and membrane dynamics. Biochim. Biophys. Acta - Mol. Cell Biol.
Lipids. 2012;1821(8):1050–1058.
179. Dissecting and manipulating the pathway for glycosylphos- phatidylinositol-anchor
biosynthesis Taroh Kinoshita * and Norimitsu Inoue. Curr. Opin. Chem. Biol. 2000;632–
638.
180. Wang B, David MD, Schrader JW. Absence of caprin-1 results in defects in cellular
proliferation. J. Immunol. 2005;175(7):4274–4282.
181. Marmor MD, Julius M. The function of GPI-anchored proteins in T cell development,
activation and regulation of homeostasis. J. Biol. Regul. Homeost. Agents. 2000;14(2):99–
115.
182. Kinoshita T, Fujita M. Biosynthesis of GPI-anchored proteins: special emphasis on GPI
lipid remodeling. J. Lipid Res. 2016;57(1):6–24.
183. Orlean P, Menon AK. Thematic review series: lipid posttranslational modifications. GPI
anchoring of protein in yeast and mammalian cells, or: how we learned to stop worrying
and love glycophospholipids. J. Lipid Res. 2007;48:993–1011.
184. Kezawa HI. Glycosylphosphatidylinositol ( GPI ) -Anchored Proteins.
2002;25(April):409–417.
185. Hu R, Mukhina GL, Lee SH, et al. Silencing of genes required for
glycosylphosphatidylinositol anchor biosynthesis in Burkitt lymphoma. Exp. Hematol.
2009;37(4):423–434.e2.
186. Watanabe R, Ohishi K, Maeda Y, Nakamura N, Kinoshita T. Mammalian PIG-L and its
yeast homologue Gpi12p are N-acetylglucosaminylphosphatidylinositol de-N-acetylases
essential in glycosylphosphatidylinositol biosynthesis. Biochem. J. 1999;339(1):185–192.
187. Tartakoff AM, Singh N. How to make a glycoinositol phospholipid anchor. Trends
151 Biochem. Sci. 1992;17(11):470–473.
188. Yada T, Sugiura R, Kita A, et al. Its8, a Fission Yeast Homolog of Mcd4 and Pig-n, Is
Involved in GPI Anchor Synthesis and Shares an Essential Function with Calcineurin in
Cytokinesis. J. Biol. Chem. 2001;276(17):13579–13586.
189. Jezela-Stanek A, Ciara E, Piekutowska-Abramczuk D, et al. Congenital disorder of
glycosylphosphatidylinositol (GPI)-anchor biosynthesis - The phenotype of two patients
with novel mutations in the PIGN and PGAP2 genes. Eur. J. Paediatr. Neurol.
2016;20(3):462–473.
190. Brodsky RA. Narrative review: Paroxysmal nocturnal hemoglobinuria: The physiology of
complement-related hemolytic anemia. Ann. Intern. Med. 2008;148(8):587–595.
191. Ihara S, Nakayama S, Murakami Y, et al. PIGN prevents protein aggregation in the
endoplasmic reticulum independently of its function in the GPI synthesis. J. Cell Sci.
2016;
192. Uhlen M. A Human Protein Atlas for Normal and Cancer Tissues Based on Antibody
Proteomics. Mol. Cell. Proteomics. 2005;4(12):1920–1932.
193. Ohba C, Okamoto N, Murakami Y, et al. PIGN mutations cause congenital anomalies,
developmental delay, hypotonia, epilepsy, and progressive cerebellar atrophy.
Neurogenetics. 2014;15(2):85–92.
194. Maydan G, Noyman I, Har-Zahav A, et al. Multiple congenital anomalies-hypotonia-
seizures syndrome is caused by a mutation in PIGN. J. Med. Genet. 2011;48(6):383–389.
195. Chen CP, Lin HM, Leung C, et al. Partial monosomy 9p (9p22.2→pter) and partial
trisomy 18q (18q21.32→qter) in a female infant with anorectal malformations. Genet.
Couns. 2012;23(2):201–206.
196. Khayat M, Tilghman JM, Chervinsky I, et al. A PIGN mutation responsible for multiple
congenital anomalies-hypotonia-seizures syndrome 1 (MCAHS1) in an Israeli-Arab
152 family. Am. J. Med. Genet. Part A. 2016;170(1):176–182.
197. Fleming L, Lemmon M, Beck N, et al. Genotype-phenotype correlation of congenital
anomalies in multiple congenital anomalies hypotonia seizures syndrome
(MCAHS1)/PIGN-related epilepsy. Am. J. Med. Genet. Part A. 2016;170(1):77–86.
198. McInerney-Leo AM, Harris JE, Gattas M, et al. Fryns Syndrome Associated with
Recessive Mutations in PIGN in two Separate Families. Hum. Mutat. 2016;37(7):695–
702.
199. Nakagawa T, Taniguchi-Ikeda M, Murakami Y, et al. A novel PIGN mutation and
prenatal diagnosis of inherited glycosylphosphatidylinositol deficiency. Am. J. Med.
Genet. Part A. 2016;170(1):183–188.
200. Brady PD, Moerman P, De Catte L, et al. Exome sequencing identifies a recessive PIGN
splice site mutation as a cause of syndromic Congenital Diaphragmatic Hernia. Eur. J.
Med. Genet. 2014;57(9):487–493.
201. Couser NL, Masood MM, Strande NT, et al. The phenotype of multiple congenital
anomalies-hypotonia-seizures syndrome 1: Report and review. Am. J. Med. Genet. Part A.
2015;167(9):2176–2181.
202. McKean DM, Niswander L. Defects in GPI biosynthesis perturb Cripto signaling during
forebrain development in two new mouse models of holoprosencephaly. Biol. Open.
2012;1(9):874–83.
203. Forbes SA, Beare D, Gunasekaran P, et al. COSMIC: Exploring the world’s knowledge of
somatic mutations in human cancer. Nucleic Acids Res. 2015;43(D1):D805–D811.
204. Gao J, Aksoy BA, Dogrusoz U, et al. Integrative Analysis of Complex Cancer Genomics
and Clinical Profiles Using the cBioPortal. Sci. Signal. 2013;6(269):pl1-pl1.
205. Cerami E, Gao J, Dogrusoz U, et al. The cBio Cancer Genomics Portal: An open platform
for exploring multidimensional cancer genomics data. Cancer Discov. 2012;2(5):401–404.
153 206. Chen R, Eshleman JR, Brodsky RA, Medof ME. Glycophosphatidylinositol-anchored
protein deficiency as a marker of Mutator phenotypes in cancer. Cancer Res.
2001;61(2):654–658.
207. Byrne M, Bennett RL, Cheng X, May WS. Progressive genomic instability in the Nup98-
HoxD13 model of MDS correlates with loss of the PIG-A gene product. Neoplasia.
2014;16(8):627–33.
208. American Cancer Society. Cancer Facts & Figures 2016. 2016.
209. Vardiman J, Reichard K. Acute myeloid leukemia with myelodysplasia-related changes.
Am. J. Clin. Pathol. 2015;144(1):29–43.
210. Siegel RL, Miller KD, Jemal A. Cancer statistics. CA Cancer J Clin. 2016;66(1):7–30.
211. Badar T, Ravandi F. Relapsed Acute Myeloid Leukemia: Need for Innovative Treatment
Strategies to Improve Outcome. Clin. Lymphoma, Myeloma Leuk. 2015;15(S):S104–S108.
212. Grove CS, Vassiliou GS. Acute myeloid leukaemia: a paradigm for the clonal evolution of
cancer? Dis. Model. Mech. 2014;7(8):941–951.
213. Østgrd LSG, Kjeldsen E, Holm MS, et al. Reasons for treating secondary AML as de novo
AML. Eur. J. Haematol. 2010;85(3):217–226.
214. Sheffer M, Bacolod MD, Zuk O, et al. Association of survival and disease progression
with chromosomal instability: a genomic exploration of colorectal cancer. Proc. Natl.
Acad. Sci. U. S. A. 2009;106(17):7131–6.
215. Tainsky MA, Bischoff FZ, Strong LC. Genomic instability due to germline p53 mutations
drives preneoplastic Progression toward cancer in human cells. Cancer Metastasis Rev.
1995;14(1):43–48.
216. Negrini S, Gorgoulis VG, Halazonetis TD. Genomic instability--an evolving hallmark of
cancer. Nat. Rev. Mol. Cell Biol. 2010;11(3):220–8.
217. Pellagatti A, Cazzola M, Giagounidis AA, et al. Gene expression profiles of CD34+ cells
154 in myelodysplastic syndromes: Involvement of interferon-stimulated genes and correlation
to FAB subtype and karyotype. Blood. 2006;108(1):337–345.
218. Trapnell C, Pachter L SS. TopHat: discovering splice junctions with RNA-Seq.
Bioinformatics. 2009;25(9):1105–1111.
219. Faul F, Erdfelder E, Buchner A, Lang A-G. Statistical power analyses using G*Power 3.1:
Tests for correlation and regression analyses. Behav. Res. Methods. 2009;41(4):1149–
1160.
220. Sandes AF, Kerbauy DMB, Matarraz S, et al. Combined flow cytometric assessment of
CD45, HLA-DR, CD34, and CD117 expression is a useful approach for reliable
quantification of blast cells in myelodysplastic syndromes. Cytom. Part B - Clin. Cytom.
2013;84 B(3):157–166.
221. Hasanali ZS, Saroya BS, Stuart A, et al. Epigenetic therapy overcomes treatment
resistance in T cell prolymphocytic leukemia. Sci. Transl. Med. 2015;7(293):293ra102.
222. Sanjana NE, Shalem O, Zhang F. Improved vectors and genome-wide libraries for
CRISPR screening. Nat. Methods. 2014;11(8):783–784.
223. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image
analysis. Nat. Methods. 2012;9(7):671–675.
224. Stothard P. The sequence manipulation suite: JavaScript programs for analyzing and
formatting protein and DNA sequences. Biotechniques. 2000;28(6):.
225. Adzhubei IA, Schmidt S, Peshkin L, et al. A method and server for predicting damaging
missense mutations. Nat. Methods. 2010;7(4):248–249.
226. Petitjean A, Mathe E, Kato S, et al. Impact of mutant p53 functional properties on TP53
mutation patterns and tumor phenotype: Lessons from recent developments in the IARC
TP53 database. Hum. Mutat. 2007;28(6):622–629.
227. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search
155 tool. J. Mol. Biol. 1990;215(3):403–10.
228. Ng PC, Henikoff S. Predicting deleterious amino acid substitutions. Genome Res.
2001;11(5):863–874.
229. Tohyama K, Tsutani H, Ueda T, Nakamura T, Yoshida Y. Establishment and
characterization of a novel myeloid cell line from the bone marrow of a patient with the
myelodysplastic syndrome. Br J Haematol. 1994;87:235–242.
230. Matsuoka A, Tochigi A, Kishimoto M, et al. Lenalidomide induces cell death in an MDS-
derived cell line with deletion of chromosome 5q by inhibition of cytokinesis. Leukemia.
2010;24(4):748–55.
231. Wong JJL, Au AYM, Ritchie W, Rasko JEJ. Intron retention in mRNA: No longer
nonsense: Known and putative roles of intron retention in normal and disease biology.
BioEssays. 2016;38(1):41–49.
232. Turinetto V, Giachino C. Multiple facets of histone variant H2AX: a DNA double-strand-
break marker with several biological functions. Nucleic Acids Res. 2015;43(5):2489–98.
233. Puddu F, Oelschlaegel T, Guerini I, et al. Synthetic viability genomic screening defines
Sae2 function in DNA repair. EMBO J. 2015;34(11):1509–22.
234. Gonorazky H, Liang M, Cummings B, et al. RNAseq analysis for the diagnosis of
muscular dystrophy. Ann. Clin. Transl. Neurol. 2016;3(1):55–60.
235. Niblock M, Smith BN, Lee Y-B, et al. Retention of hexanucleotide repeat-containing
intron in C9orf72 mRNA: implications for the pathogenesis of ALS/FTD. Acta
Neuropathol. Commun. 2016;4(1):18.
236. Parks LL, Tumey MK, Kovacs WJ. An ACTH-producing small cell lung cancer expresses
an aberrant glucocorticoid receptor (GR) transcript from a normal GR gene. J. Investig.
Med. 1996;44(1):30A.
237. Häsler R, Kerick M, Mah N, et al. Alterations of pre-mRNA splicing in human
156 inflammatory bowel disease. Eur. J. Cell Biol. 2011;90(6–7):603–611.
238. Ogawa S. Splicing factor mutations in myelodysplasia. Int. J. Hematol. 2012;96(4):438–
442.
239. Stiehl T, Baran N, Ho AD, Marciniak-Czochra A. Clonal selection and therapy resistance
in acute leukaemias: mathematical modelling explains different proliferation patterns at
diagnosis and relapse. J. R. Soc. Interface. 2014;11(94):20140079.
240. Cogle CR, Bosse RC, Brewer T, et al. Acute myeloid leukemia in the vascular niche.
Cancer Lett. 2016;380(2):552–560.
241. Golkar SO, Czene S, Gokarakonda A, Haghdoost S. Intracellular deoxyribonucleotide
pool imbalance and DNA damage in cells treated with hydroxyurea, an inhibitor of
ribonucleotide reductase. Mutagenesis. 2013;28(6):653–660.
242. Santos JL, Bosquesi PL, Almeida AE, Chin CM, Varanda EA. Mutagenic and genotoxic
effect of hydroxyurea. Int. J. Biomed. Sci. 2011;7(4):.
243. Flanagan JM, Howard TA, Mortier N, et al. Assessment of genotoxicity associated with
hydroxyurea therapy in children with sickle cell anemia. Mutat. Res. - Genet. Toxicol.
Environ. Mutagen. 2010;698(1–2):38–42.
244. Kandel R, Li S-Q, Ozcelik H, Rohan T. p53 protein accumulation and mutations in normal
and benign breast tissue. Int. J. Cancer. 2000;87(1):73–78.
245. Wong TN, Ramsingh G, Young AL, et al. Role of TP53 mutations in the origin and
evolution of therapy-related acute myeloid leukaemia. TL - 518. Nature. 2015;518 VN-
(7540):552–555.
246. Hermsen R, Toonen P, Kuijk E, et al. Lack of Major Genome Instability in Tumors of p53
Null Rats. PLoS One. 2015;10(3):e0122066.
247. Talwalkar SS, Yin CC, Naeem RC, et al. Myelodysplastic syndromes arising in patients
with germline TP53 mutation and Li-Fraumeni syndrome. Arch. Pathol. Lab. Med.
157 2010;134(7):1010–1015.
248. Schnerch D, Yalcintepe J, Schmidts A, et al. Cell cycle control in acute myeloid leukemia.
Am. J. Cancer Res. 2012;2(5):508–28.
249. Schnerch D, Schmidts A, Follo M, et al. BubR1 is frequently repressed in acute myeloid
leukemia and its re-expression sensitizes cells to antimitotic therapy. Haematologica.
2013;98(12):1886–1895.
250. Krapf G, Kaindl U, Kilbey a, et al. ETV6/RUNX1 abrogates mitotic checkpoint function
and targets its key player MAD2L1. Oncogene. 2010;29(22):3307–3312.
251. Lin S-F, Lin P-M, Yang M-C, et al. Expression of hBUB1 in acute myeloid leukemia.
Leuk. Lymphoma. 2002;43(2):385–91.
252. Musacchio A. Spindle assembly checkpoint: the third decade. Philos. Trans. R. Soc. Lond.
B. Biol. Sci. 2011;366(1584):3595–604.
253. Lara-Gonzalez P, Westhorpe FG, Taylor SS. The spindle assembly checkpoint. Curr. Biol.
2012;22(22):.
254. Kops GJPL, Weaver B a a, Cleveland DW. On the road to cancer: aneuploidy and the
mitotic checkpoint. Nat. Rev. Cancer. 2005;5(10):773–85.
255. Musacchio A, Salmon ED. The spindle-assembly checkpoint in space and time. Nat. Rev.
Mol. Cell Biol. 2007;8(5):379–393.
256. Akera T, Watanabe Y. The spindle assembly checkpoint promotes chromosome bi-
orientation: A novel Mad1 role in chromosome alignment. Cell Cycle. 2016;15(4):493–
497.
257. Thompson SL, Compton DA. Chromosome missegregation in human cells arises through
specific types of kinetochore-microtubule attachment errors. Proc. Natl. Acad. Sci.
2011;108(44):17974–17978.
258. Mapelli M, Massimiliano L, Santaguida S, Musacchio A. The Mad2 Conformational
158 Dimer: Structure and Implications for the Spindle Assembly Checkpoint. Cell.
2007;131(4):730–743.
259. Tipton AR, Tipton M, Yen T, Liu ST. Closed MAD2 (C-MAD2) is selectively
incorporated into the mitotic checkpoint complex (MCC). Cell Cycle. 2011;10(21):3740–
3750.
260. Antoni A De, Pearson CG, Cimini D, et al. The Mad1 / Mad2 Complex as a Template for
Mad2 Activation in the Spindle Assembly Checkpoint University of North Carolina at
Chapel Hill. Curr.Biol. 2005;15(3):214–225.
261. Schuyler SC, Wu Y-F, Kuan VJ-W. The Mad1-Mad2 balancing act - a damaged spindle
checkpoint in chromosome instability and cancer. J. Cell Sci. 2012;125(Pt 18):4197–206.
262. Han JS, Holland AJ, Fachinetti D, et al. Catalytic assembly of the mitotic checkpoint
inhibitor BubR1-Cdc20 by a Mad2-induced functional Switch in Cdc20. Mol. Cell.
2013;51(1):92–104.
263. Lara-Gonzalez P, Scott MIF, Diez M, Sen O, Taylor SS. BubR1 blocks substrate
recruitment to the APC/C in a KEN-box-dependent manner. J. Cell Sci.
2011;124(24):4332–4345.
264. Varetti G, Guida C, Santaguida S, Chiroli E, Musacchio A. Homeostatic control of mitotic
arrest. Mol. Cell. 2011;44(5):710–720.
265. Nezi L, Musacchio A. Sister chromatid tension and the spindle assembly checkpoint.
Curr. Opin. Cell Biol. 2009;21(6):785–795.
266. Musacchio A. The Molecular Biology of Spindle Assembly Checkpoint Signaling
Dynamics. Curr. Biol. 2015;25(20):R1002–R1018.
267. Teye EK, Sido A, Xin P, et al. PIGN gene expression aberration is associated with
genomic instability and leukemic progression in acute myeloid leukemia with
myelodysplastic features. Oncotarget. 2017;5(0):1–18.
159 268. Sanjana NE, Shalem O, Zhang F. Improved vectors and genome-wide libraries for
CRISPR screening. Nat. Methods. 2014;11(8):783–784.
269. Mandal PK, Ferreira LMR, Collins R, et al. Efficient ablation of genes in human
hematopoietic stem and effector cells using CRISPR/Cas9. Cell Stem Cell.
2014;15(5):643–652.
270. Jackman J, O’Connor PM. Methods for synchronizing cells at specific stages of the cell
cycle. Curr. Protoc. Cell Biol. 2001;Chapter 8:Unit 8.3.
271. Magiera MM, Gueydon E, Schwob E. DNA replication and spindle checkpoints cooperate
during S phase to delay mitosis and preserve genome integrity. J. Cell Biol.
2014;204(2):165–175.
272. Lawrence KS, Chau T, Engebrecht J. DNA Damage Response and Spindle Assembly
Checkpoint Function throughout the Cell Cycle to Ensure Genomic Integrity. PLOS
Genet. 2015;11(4):e1005150.
273. Shandilya J, Toska E, Richard DJ, Medler KF, Roberts SGE. WT1 interacts with MAD2
and regulates mitotic checkpoint function. Nat. Commun. 2014;5:4903.
274. Shandilya J, Roberts SGE. A role of WT1 in cell division and genomic stability. Cell
Cycle. 2015;4101(March):37–41.
275. Yeh PC, Yeh CC, Chen YC, Juang YL. RED, a spindle pole-associated protein, is
required for kinetochore localization of MAD1, mitotic progression, and activation of the
spindle assembly checkpoint. J. Biol. Chem. 2012;287(15):11704–11716.
276. Lussi YC, Shumaker DK, Shimi T, Fahrenkrog B. The nucleoporin Nup153 affects
spindle checkpoint activity due to an association with Mad1. Nucleus. 2010;1(1):71–84.
277. Kassardjian A, Rizkallah R, Riman S, et al. The Transcription Factor YY1 Is a Novel
Substrate for Aurora B Kinase at G2/M Transition of the Cell Cycle. PLoS One.
2012;7(11):.
160 278. Osmundson EC, Ray D, Moore FE, et al. The HECT E3 ligase Smurf2 is required for
Mad2-dependent spindle assembly checkpoint. J. Cell Biol. 2008;183(2):267–277.
279. Li H, Cuenin C, Murr R, Wang Z-Q, Herceg Z. HAT cofactor Trrap regulates the mitotic
checkpoint by modulation of Mad1 and Mad2 expression. EMBO J. 2004;23(24):4824–
4834.
280. Tomasini R, Tsuchihara K, Tsuda C, et al. TAp73 regulates the spindle assembly
checkpoint by modulating BubR1 activity. Proc Natl Acad Sci U S A. 2009;106(3):797–
802.
281. Heinrich S, Sewart K, Windecker H, et al. Mad1 contribution to spindle assembly
checkpoint signalling goes beyond presenting Mad2 at kinetochores. EMBO Rep.
2014;15(3):291–298.
282. Kuijt TEF, Omerzu M, Saurin AT, Kops GJPL. Conditional targeting of MAD1 to
kinetochores is sufficient to reactivate the spindle assembly checkpoint in metaphase.
Chromosoma. 2014;123(5):471–480.
283. Tipton AR, Ji W, Sturt-Gillespie B, et al. Monopolar spindle 1 (MPS1) kinase promotes
production of closed MAD2 (C-MAD2) conformer and assembly of the mitotic
checkpoint complex. J. Biol. Chem. 2013;288(49):35149–35158.
284. Ballister ER, Riegman M, Lampson MA. Recruitment of Mad1 to metaphase kinetochores
is sufficient to reactivate the mitotic checkpoint. J. Cell Biol. 2014;204(6):901–908.
285. Meraldi P, Draviam VM, Sorger PK, et al. Timing and checkpoints in the regulation of
mitotic progression. Dev. Cell. 2004;7(1):45–60.
286. Kienitz A, Vogel C, Morales I, Müller R, Bastians H. Partial downregulation of MAD1
causes spindle checkpoint inactivation and aneuploidy, but does not confer resistance
towards taxol. Oncogene. 2005;24(26):4301–4310.
287. London N, Biggins S. Signalling dynamics in the spindle checkpoint response. Nat. Rev.
161 Mol. Cell Biol. 2014;15(11):736–47.
288. de Greef GE, Hagemeijer A. Molecular and cytogenetic abnormalities in acute myeloid
leukaemia and myelodysplastic syndromes. Baillieres Clin Haematol. 1996;9(1):1–18.
289. Humphries RK, Eaves a C, Eaves CJ. Self-renewal of hemopoietic stem cells during
mixed colony formation in vitro. Proc. Natl. Acad. Sci. U. S. A. 1981;78(6):3629–33.
290. Zhivotovsky B, Kroemer G. Apoptosis and genomic instability. Nat. Rev. Mol. Cell Biol.
2004;5(9):752–762.
291. Kerbauy DB, Deeg HJ. Apoptosis and antiapoptotic mechanisms in the progression of
myelodysplastic syndrome. Exp. Hematol. 2007;35(11):1739–1746.
292. Burrell RA, McClelland SE, Bartek J, Swanton C. Response to Bakhoum et al. Curr. Biol.
2014;24(4):.
293. Yu H. Cdc20: A WD40 Activator for a Cell Cycle Degradation Machine. Mol. Cell.
2007;27(1):3–16.
294. Solomon DA, Kim T, Diaz-Martinez LA, et al. Mutational inactivation of STAG2 causes
aneuploidy in human cancer. Science (80-. ). 2011;333(6045):1039–43.
295. Musacchio A, Salmon ED. The spindle-assembly checkpoint in space and time. Nat. Rev.
Mol. Cell Biol. 2007;8(5):379–93.
296. Zhang N, Ge G, Meyer R, et al. Overexpression of Separase induces aneuploidy and
mammary tumorigenesis. Proc. Natl. Acad. Sci. U. S. A. 2008;105(35):13033–8.
297. Tu R, Lu W, Chen J, McCabe CJ, Melmed S. Overexpressed Pituitary Tumor-
Transforming Gene Causes Aneuploidy in Live Human Cells. Endocrinology.
2003;144(11):4991–4998.
298. Torres EM, Williams BR, Amon A. Aneuploidy: Cells losing their balance. Genetics.
2008;179(2):737–746.
299. Holland AJ, Cleveland DW. Losing balance: the origin and impact of aneuploidy in
162 cancer. EMBO Rep. 2012;13(6):501–14.
300. Khayat M, Tilghman JM, Chervinsky I, et al. A PIGN mutation responsible for multiple
congenital anomalies-hypotonia-seizures syndrome 1 (MCAHS1) in an Israeli-Arab
family. Am. J. Med. Genet. Part A. 2016;170(1):176–182.
301. Brady PD, Moerman P, De Catte L, et al. Exome sequencing identifies a recessive PIGN
splice site mutation as a cause of syndromic Congenital Diaphragmatic Hernia. Eur. J.
Med. Genet. 2014;57(9):487–493.
302. Thornhill AR, Handyside AH. The role of aneuploidy screening in human preimplantation
embryos. Bienn. Rev. Infertil. Vol. 1. 2009;277–288.
303. Mastenbroek S, Repping S. Preimplantation genetic screening: back to the future. Hum.
Reprod. 2014;29(9):1846–50.
304. Bader-Meunier B, Mielot F, Tchernia G, et al. Myelodysplastic syndromes in childhood:
report of 49 patients from a French multicentre study. French Society of Paediatric
Haematology and Immunology. Br J Haematol. 1996;92(2):344–350.
305. Hou JW, Teng RJ, Tien HF, Lin KH. Jacobsen distal 11q deletion syndrome with a
myelodysplastic change of hemopoietic cells. Am. J. Med. Genet. 1998;75(4):341–344.
306. Holland AJ, Cleveland DW. Losing balance: the origin and impact of aneuploidy in
cancer. EMBO Rep. 2012;13(6):501–14.
307. Wang R, Burton JL, Solomon MJ. Transcriptional and post-transcriptional regulation of
Cdc20 during the spindle assembly checkpoint in S. cerevisiae. Cell. Signal. 2017;33:41–
48.
308. Mena AL, Lam EWF, Chatterjee S. Sustained spindle-assembly checkpoint response
requires de novo transcription and translation of cyclin B1. PLoS One. 2010;5(9):.
309. Castro D, Ca G De, Pérez de Castro I, de Cárcer G, Malumbres M. A census of mitotic
cancer genes: new insights into tumor cell biology and cancer therapy. Carcinogenesis.
163 2007;28(5):899–912.
310. Whitfield ML, Sherlock G, Saldanha AJ, et al. Identification of Genes Periodically
Expressed in the Human Cell Cycle and Their Expression in Tumors. Mol. Biol. Cell.
2002;13(6):1977–2000.
311. Nath S, Ghatak D, Das P, Roychoudhury S. Transcriptional control of mitosis:
deregulation and cancer. Front. Endocrinol. (Lausanne). 2015;6(May):60.
312. Kutay U, Hetzer MW. Reorganization of the nuclear envelope during open mitosis. Curr.
Opin. Cell Biol. 2008;20(6):669–677.
313. Hetzer MW. The nuclear envelope. Cold Spring Harb. Perspect. Biol. 2010;2(3):.
314. Müller G a, Engeland K. The central role of CDE/CHR promoter elements in the
regulation of cell cycle-dependent gene transcription. FEBS J. 2010;277(4):877–893.
315. Laoukili J, Kooistra MR, Bras A, et al. FoxM1 is required for execution of the mitotic
programme and chromosome stability. Nat. Cell Biol. 2005;7(2):126–136.
316. Ishida S, Huang E, Zuzan H, et al. Role for E2F in Control of Both DNA Replication and
Mitotic Functions as Revealed from DNA Microarray Analysis. Mol. Cell. Biol.
2001;21(14):4684–4699.
317. Den Hollander J, Rimpi S, Doherty JR, et al. Aurora kinases A and B are up-regulated by
Myc and are essential for maintenance of the malignant state. Blood. 2010;116(9):1498–
1505.
318. Caretti G, Salsi V, Vecchi C, Imbriano C, Mantovani R. Dynamic recruitment of NF-Y
and histone acetyltransferases on cell-cycle promoters. J. Biol. Chem.
2003;278(33):30435–30440.
319. Menssen A, Epanchintsev A, Rezaei N, et al. c-MYC delays prometaphase by direct
transactivation of MAD2 and BubR1: Identification of mechanisms underlying c-MYC-
induced DNA damage and chromosomal instability. Cell Cycle. 2007;6(3):339–352.
164 320. Nath S, Chowdhury A, Dey S, et al. Deregulation of Rb-E2F1 axis causes chromosomal
instability by engaging the transactivation function of Cdc20-anaphase-promoting
complex/cyclosome. Mol. Cell. Biol. 2015;35(2):356–69.
321. Hernando E, Nahlé Z, Juan G, et al. Rb inactivation promotes genomic instability by
uncoupling cell cycle progression from mitotic control. Nature. 2004;430(7001):797–802.
322. Fu Z, Malureanu L, Huang J, et al. Plk1-dependent phosphorylation of {FoxM1} regulates
a transcriptional programme required for mitotic progression. Nat. Cell Biol.
2008;10(9):1076–1082.
323. Yoon YM, Baek KH, Jeong SJ, et al. WD repeat-containing mitotic checkpoint proteins
act as transcriptional repressors during interphase. FEBS Lett. 2004;575(1–3):23–29.
324. Kidokoro T, Tanikawa C, Furukawa Y, et al. CDC20, a potential cancer therapeutic target,
is negatively regulated by p53. Oncogene. 2008;27(11):1562–1571.
325. Bae I, Rih JK, Kim HJ, et al. BRCA1 regulates gene expression for orderly mitotic
progression. Cell Cycle. 2005;4(11):1641–1666.
326. You J, Li Q, Wu C, et al. Regulation of aurora B expression by the bromodomain protein
Brd4. Mol. Cell. Biol. 2009;29(18):5094–103.
327. Chun ACS, Jin DY. Transcriptional regulation of mitotic checkpoint gene MAD1 by p53.
J. Biol. Chem. 2003;278(39):37439–37450.
328. Rivera A, Mavila A, Bayless KJ, Davis GE, Maxwell SA. Cyclin A1 is a p53-induced
gene that mediates apoptosis, G2/M arrest, and mitotic catastrophe in renal, ovarian, and
lung carcinoma cells. Cell. Mol. Life Sci. 2006;63(12):1425–1439.
329. Krause K, Wasner M, Reinhard W, et al. The tumour suppressor protein p53 can repress
transcription of cyclin B. Nucleic Acids Res. 2000;28(22):4410–8.
330. Kidokoro T, Tanikawa C, Furukawa Y, et al. CDC20, a potential cancer therapeutic target,
is negatively regulated by p53. Oncogene. 2008;27(11):1562–1571.
165 331. Thoma CR, Toso A, Gutbrodt KL, et al. VHL loss causes spindle misorientation and
chromosome instability. Nat. Cell Biol. 2009;11(8):994–1001.
332. Ryan SD, Britigan EMC, Zasadil LM, et al. Up-regulation of the mitotic checkpoint
component Mad1 causes chromosomal instability and resistance to microtubule poisons.
Proc. Natl. Acad. Sci. U. S. A. 2012;109(33):E2205-14.
333. Stegmeier F, Rape M, Draviam VM, et al. Anaphase initiation is regulated by antagonistic
ubiquitination and deubiquitination activities. Nature. 2007;446(7138):876–881.
334. Malumbres M, Barbacid M. Cell cycle kinases in cancer. Curr. Opin. Genet. Dev.
2007;17(1):60–65.
335. von Schubert C, Cubizolles F, Bracher JM, et al. Plk1 and Mps1 Cooperatively Regulate
the Spindle Assembly Checkpoint in Human Cells. Cell Rep. 2015;12(1):66–78.
336. Lapenna S, Giordano A. Cell cycle kinases as therapeutic targets for cancer. Nat. Rev.
Drug Discov. 2009;8(7):547–566.
337. London N, Biggins S. Mad1 kinetochore recruitment by Mps1-mediated phosphorylation
of Bub1 signals the spindle checkpoint. Genes Dev. 2014;28(2):140–152.
338. Cullen TW, Madsen JA, Ivanov PL, Brodbelt JS, Trent MS. Characterization of unique
modification of flagellar rod protein FlgG by Campylobacter jejuni lipid A
phosphoethanolamine transferase, linking bacterial locomotion and antimicrobial peptide
resistance. J. Biol. Chem. 2012;287(5):3326–3336.
339. Wiedman JM, Fabre AL, Taron BW, Taron CH, Orlean P. In vivo characterization of the
GPI assembly defect in yeast mcd4-174 mutants and bypass of the Mcd4p-dependent step
in mcd4?? cells. FEMS Yeast Res. 2007;7(1):78–83.
340. Artimo P, Jonnalagedda M, Arnold K, et al. ExPASy: SIB bioinformatics resource portal.
Nucleic Acids Res. 2012;40(W1):.
341. Ferreira AK, Santana-Lemos BAA, Rego EM, et al. Synthetic phosphoethanolamine has
166 in vitro and in vivo anti-leukemia effects. Br. J. Cancer. 2013;109(11):2819–2828.
342. Farriol-Mathis N, Garavelli JS, Boeckmann B, et al. Annotation of post-translational
modifications in the Swiss-Prot knowledge base. Proteomics. 2004;4(6):1537–1550.
343. Fankhauser N, Mäser P. Identification of GPI anchor attachment signals by a Kohonen
self-organizing map. Bioinformatics. 2005;21(9):1846–1852.
344. Qiao D, Yang X, Meyer K, Friedl A. Glypican-1 regulates anaphase promoting
complex/cyclosome substrates and cell cycle progression in endothelial cells. Mol. Biol.
Cell. 2008;19(7):2789–2801.
345. Eisenhaber B, Bork P, Eisenhaber F. Prediction of Potential GPI-modification Sites in
Proprotein Sequences. J. Mol. Biol. 1999;292(3):741–758.
346. Kapustin Y, Chan E, Sarkar R, et al. Cryptic splice sites and split genes. Nucleic Acids
Res. 2011;39(14):5837–5844.
347. Cimini D. Merotelic kinetochore orientation, aneuploidy, and cancer. Biochim. Biophys.
Acta - Rev. Cancer. 2008;1786(1):32–40.
348. Carter SL, Eklund AC, Kohane IS, Harris LN, Szallasi Z. A signature of chromosomal
instability inferred from gene expression profiles predicts clinical outcome in multiple
human cancers. Nat. Genet. 2006;38(9):1043–1048.
VITA Emmanuel Kwame Teye Education 2012-2017: Doctor of Philosophy, Biomedical Sciences (Translational Therapeutics) Pennsylvania State University College of Medicine, Hershey, PA. 2009-2011: Master of Science, Cell & Molecular Biology, Ball State University, Muncie, IN. 2010-2011: Graduate Certificate, Biotechnology, Ball State University, Muncie, IN 2004-2008: Bachelor of Science (Hons), Medical Laboratory Technology, University of Science and Technology, Kumasi, Ghana
Work Experience 2012-2017: Graduate Lecturer/Researcher (Cancer Center), Pennsylvania State University College of Medicine, Hershey, PA 2011-2012: Analytical Scientist (In vitro Diagnostics), Novartis Institutes for Biomedical Research, Cambridge, MA 2010-2011: Graduate Research/Teaching Assistant, Ball State University, Muncie, IN
Board Certification 2012-date: International Medical Technologist, American Medical Technologists (AMT)
Awards and Recognitions Science Program for Excellence in Science; American Association for the Advancement of Science (AAAS) Pennsylvania State University College of Medicine Summer Research Program (SURIP) Junior Mentorship Recognition
Patent Contribution US Patent Application No. 15/434,774
Activities Pennsylvania State University Faculty Senate and Select Committee on Global Programs Pennsylvania State College of Medicine Summer Research Program Junior Mentor Pennsylvania State College of Medicine Library Advisory Committee Graduate Students' Association International Committee Chair Christian Medical Society
Publications Teye EK, Sido A, Xin P, Finnberg NK, Gokare P, Kawasawa YI, Salzberg AC, Shimko S, Bayerl M, Ehmann WC, Claxton DF, Rybka WB, Drabik JJ, et al. PIGN gene expression aberration is associated with genomic instability and leukemic progression in acute myeloid leukemia with myelodysplastic features. Oncotarget. 2017; 8:29887-29905. Teye EK, Yang W, Lu S, Pu JJ. PIGN spatiotemporally regulates the spindle assembly checkpoint. 2017 [pending submission] Wang W, Lulla AR, Hernandez-Borrero LJ, Dicker DT, Teye EK, Talekar M K, Dolloff NG, Pu JJ, & El-Deiry WS. Synergistic Effect of Quinacrine with Chemotherapeutics or TRAIL in Hematopoietic Malignant Cells. Blood, 2014; 124(21), 5239. Owiredu WKBA*, Teye EK*, Quaye L. Proficiency testing of total serum cholesterol assay by the ATAC 8000® Random Access Chemistry Autoanalyzer at the Komfo Anokye Teaching Hospital. J Med Biomed Sci. 2013; 2: 22–9. *Co-first author