Regulation of Hematopoietic Progenitor Formation in a Shwachman Diamond Syndrome Induced Pluripotent Stem Cell Disease Model

Alice Maria Luca

A thesis submitted in conformity with the requirements for the degree of Masters of Science

Institute of Medical Science University of Toronto

Shwachman-Diamond Syndrome Induced Pluripotent Stem

Cell Disease Model

Alice Maria Luca Master of Science Institute of Medical Sciences University of Toronto 2015

Abstract

Shwachman-Diamond syndrome (SDS) is an inherited bone marrow failure disease, with

90% of SDS patients carrying a mutation in the SBDS gene. Due to limited efficacy or toxicity of current treatments available, and the otherwise reduced life expectancy of SDS patients, novel therapeutic strategies are needed. Since the main morbidity and mortality are related to the blood dyscrasia, studying hematopoiesis will help characterize the hematological phenotype. We hypothesized that the definitive wave of hematopoiesis is markedly impaired. We generated SDS iPSCs that recapitulated the human SDS disease, specifically, the reduced blood cell formation. The SDS iPSCs showed a defect in definitive hematopoiesis, with a marked reduction in the hemogenic endothelium population. We did not observe a defect in primitive hematopoiesis. Our study sheds light on the onset and progression of the SDS hematopoietic phenotype, and provides a platform for the development of novel, potential therapeutic targets to improve patient care.

ii Acknowledgments

My graduate journey has been a wonderful learning experience, and this work would not have been possible without the guidance, support and encouragement I have received from a number of influential people. I would like to take a moment to extend my deepest gratitude to all that were involved in this process.

First of all, I would like to thank my supervisor, Dr.Yigal Dror, without whom this project would not have been possible. His guidance, critical-thinking and support were instrumental to my journey. I am grateful for all the time and attention that he always put into ensuring that I was successful every step of the way. Thank you for always pushing me to think critically, and for your unwavering support, even when I doubted myself.

I would also like to thank Mathura Sabanayagam, for the ribosome profile experiments, and her willingness to spend her time performing them in a 4°C room. A special thank you to Bozana Zlateska for her instrumental role in the generation of the patient-specific iPSCs. Thank you to Hongbing and other lab members for always providing valuable input, and for your continuous encouragement. To Santhosh Dhanraj, thank you for your kind words and friendship.

I am thankful for all the valuable discussion, constructive-criticism and encouragement that Dr.Gordon Keller has provided. I am also very grateful for all the time and patience that Marion Kennedy offered me, as well as her expertise, guidance and support. To

iii Dr.Gordon Keller, Marion Kennedy, and the rest of the Keller lab, thank you for always keeping me laughing, and always making me feel welcomed.

I would also like to thank my committee members, Dr.Michael Glogauer and Walter

Kahr for all of their critical insight to my work.

Last but not least, I would like to thank my personal cheerleaders, my parents, my sister,

Delia, NV and Ryan for their constant words of encouragement every step of the way.

iv Contributions

I would like to thank Dr. Yigal Dror for his instrumental role in designing the research, critically interpreting the data and editing the thesis. Mathura Sabanayagam performed the ribosome profile experiments and analyzed the acquired data. Members of the

Sickkids-University Health Network Flow Cytometry Facility provided their assistance for the cell sorting experiments. Members of The Centre for Applied Genomics (TCAG) at Sickkids Hospital provided their expertise in sequencing the SBDS mutations. The

Centre for Commercialization of Regenerative Medicine (CCRM) generated the Sendai- derived iPSCs, performed reverse transcription quantitative real time polymerase chain reaction and immunofluorescence staining of stem cell pluripotency markers. Gordon

Keller helped in designing the research, as well as critically interpreting the data. Marion

Kennedy helped design the research, critically interpreted the data, and also trained me in induced pluripotent stem cells culturing and hematopoietic development. The rest of the members of the Dror and Keller labs provided their input regarding the design of the project. Dr. Michael Glogauer and Dr. Walter Kahr helped in designing the research. I performed the DNA isolation, polymerase chain reaction amplification, hematopoietic differentiation, clonogenic assay, flow cytometry and microscopy experiments, and analyzed the data. I was also responsible for designing the research and writing the thesis.

This work would not have been possible without funding from the Institute of Medical

Sciences, the Butterfly Guild and Nicola’s Triathlon for Kids.

v Table of Contents

Abstract…………………………………………………………………………………. ii Acknowledgments……………………………………………………………………… iii Contributions…………………………………………………………………………… v Table of Contents………………………………………………………………………. vi List of Tables and Figures……………………………………………………………… ix List of Abbreviations…………………………………………………………………… x

Chapter I General Introduction 1.1 Shwachman Diamond Syndrome 1 1.1.1 Hematological abnormalities 2 1.1.2 Non-hematological abnormalities 6 1.1.3 Treatment 8 1.1.4 Characterization of Shwachman-Bodian Diamond Syndrome gene 9 1.1.5 Current disease models 14 1.2 Hematopoietic development and hematopoietic differentiation models 16 1.2.1 Primitive hematopoiesis 21 1.2.2 Definitive hematopoiesis 22 1.2.3 Hematopoietic differentiation model using control PSCs 28 1.3 Disease-specific hiPSCs 32 1.4 Objective of Study 35 1.4.1 Rationale 35 1.4.2 Hypothesis 39 1.4.3 Specific aims 39

Chapter II Materials and Methods 2.1 Cell culture 2.1.1 Cell lines 40 2.1.1.1 Mouse embryonic fibroblasts 40 2.1.1.2 OP9-DL1 40 2.1.1.3 hiPSCs 41 2.1.2 Feeder depletion 42 2.1.3 Establishing embryoid bodies for hematopoietic differentiation 42 2.1.3.1 Primitive hematopoiesis 43 2.1.3.2 Definitive hematopoiesis 45 2.1.4 Preparation of single-cell suspension from embryoid bodies 48 2.1.5 Preparation of single-cell suspensions from OP9-DL1 co-cultures 48 2.1.6 Clonogenic assay 48 2.1.7 Morphological analysis by Wright-Giemsa staining 49 2.2 DNA isolation 49 2.3 PCR amplification and sequencing 50 2.4 Flow cytometry and cell sorting 50 2.4.1 Antibodies 50 2.4.2. Fluorescence-activated cell sorting and flow cytometry analysis 50

vi 2.5 Sucrose density gradient ultracentrifugation 51 2.6 Statistics 52

Chapter III Results 3.1 SDS and control iPCSs can be generated from SDS patient and control fibroblasts 53 3.1.1 Generation efficiency 53 3.1.2 SDS iPSCs carry the original patient mutations 55 3.1.3 Pluripotency markers 56 3.1.4 Germ layer differentiation 58 3.2 SDS iPSCs are characterized by an abnormal ribosome profile 60 3.2.1 SDS iPSCs show reduced levels of 80S ribosome subunit 60 3.3 SDS iPSCs can form definitive hematopoietic progenitors 62 3.4 SDS iPSCs can form terminally differentiated blood cells characteristic of 64 definitive hematopoiesis 3.4.1 Terminally differentiated blood cells have a morphology characteristic of macrophages, 64 erythrocytes and granulocytes ……------…………………….. 3.5 SDS iPSCs have a reduced capacity to form hematopoietic progenitors characteristic of definitive hematopoiesis 65 3.6 SDS iPSCs have a developmental defect in definitive hematopoiesis 67 3.6.1 Mesoderm induction is not affected in SDS iPSCs 67 3.6.2 Induction of cells with hemogenic endothelium potential is reduced in SDS iPSCs 70 3.6.3 Early hematopoietic progenitor induction is reduced in SDS iPSCs 73 3.6.4 Myeloid cell induction is reduced in SDS iPSCs 76 3.6.5 Granulocyte/Monocyte induction is reduced in SDS iPSCs 78 3.7 SDS iPSCs manifest a delay in the development of primitive 80 hematopoiesis, without a quantitative defect 3.7.1 Mesoderm/Hemangioblast induction is intact in SDS iPSCs 80 3.7.2 Induction of cells with hemogenic endothelium potential is not impaired in SDS iPSCs 82 3.7.3 Early hematopoietic progenitor induction is not affected in SDS iPSCs 85 3.7.4 Mature blood cell induction is not impaired in SDS iPSCs 88 3.7.5 SDS iPSCs showcase a delay in their ability to form hematopoietic progenitors characteristic of primitive hematopoiesis 92

Chapter IV Discussion 4.1 Discussion of results 94 4.1.1 SDS iPSCs can be generated using integrative and non-integrative transgenes 96 4.1.2 Recapitulation of SDS using SDS iPSCs ………………………………... 98 4.1.3 Mesoderm development is intact when SDS iPSCs are induced to undergo definitive hematopoiesis 102 4.1.4 The SDS definitive hematopoietic defect was first noticed at the hemogenic 103 endothelium induction phase 4.1.5 During primitive hematopoiesis, clonogenic potential is delayed, but not impaired 106 4.2 Limitations of study 107 4.3 Significance 109

vii Chapter V Conclusion 112

Chapter VI 114 Future Directions

References 116

viii List of Tables and Figures

Table 1. Human induced pluripotent stem cells generation efficiency 54 Table 2. Human induced pluripotent stem cells germ layer differentiation 59

Figure 1. Human SBDS gene structure 13 Figure 2. Human SBDS protein structure 13 Figure 3. Sites of human hematopoietic development during fetal life and early infancy 20 Figure 4. Current model of human hematopoietic hierarchy 27 Figure 5. Model of primitive and definitive hematopoiesis using pluripotent stem cells 31 Figure 6. Primitive hematopoiesis differentiation scheme for human pluripotent stem cells 44 Figure 7. Definitive hematopoiesis differentiation scheme for human pluripotent stem cells 47 Figure 8. Sequencing of SBDS in SDS induced pluripotent stem cells 55 Figure 9. Human induced pluripotent stem cells pluripotency markers 57 Figure 10. Abnormal ribosome profile in SDS induced pluripotent stem cells 61 Figure 11. Hematopoietic progenitor colony formation 63 Figure 12. Terminally differentiated blood cells 64 Figure 13. Reduced formation of hematopoietic progenitors characteristic of definitive 66 hematopoiesis in SDS induced pluripotent stem cells Figure 14. Mesoderm induction during definitive hematopoiesis: CD34-CD56+ 68 Figure 15. Mesoderm induction during definitive hematopoiesis: CD34-KDR+C-kit- 69 Figure 16. Induction of cells with hemogenic endothelium potential during definitive 71

hematopoiesis: CD34+CD43- Figure 17. Induction of cells with hemogenic endothelium potential during definitive hematopoiesis 72 CD34+CD31+ Figure 18. Early hematopoietic progenitor induction during definitive hematopoiesis: CD34+CD45+ 74

Figure 19. Early hematopoietic progenitor induction during definitive hematopoiesis: CD34+CD43+ 75 Figure 20. Myeloid cell induction during definitive hematopoiesis: CD34-CD45+ 77 Figure 21. Granulocyte/Monocyte induction during definitive hematopoiesis: CD34- 79 CD45+CD11b+ Figure 22. Mesoderm/Hemangioblast induction during primitive hematopoiesis: KDR+C-kit- 81 Figure 23. Induction of cells with hemogenic endothelium potential during primitive hematopoiesis: 83 CD34+CD43- Figure 24. Induction of cells with hemogenic endothelium potential during primitive hematopoiesis: 84 CD34+CD31+ Figure 25. Early hematopoietic progenitor induction during primitive hematopoiesis: CD34+CD43+ 86

Figure 26. Early hematopoietic progenitor induction during primitive hematopoiesis: CD34+CD45+ 87 Figure 27. Mature blood cell induction during primitive hematopoiesis: 90 CD34-CD43+ Figure 28. Mature blood cell induction during primitive hematopoiesis: CD34-CD45+ 91

Figure 29. Delayed formation of hematopoietic progenitors characteristic of primitive 93 hematopoiesis in SDS induced pluripotent stem cells

ix List of Abbreviations

AA Ascorbic acid Act.A Activin A AFP Alpha-fetoprotein AGM Aorta-gonad/mesonephros AHR Aryl hydrocarbon receptor AK2 Adenylate kinase 2 AML Acute myelogenous leukemia APLNR Apelin receptor BCL11A B-cell CLL/lymphoma 11A bFGF Basic fibroblast growth factor BFU-E Burst forming unit – erythroid BM Bone marrow BMI1 BMI1 proto-oncogene BMP-4 Bone morphogenetic protein-4 C/EBPα Ccaat-enhancer-binding proteins CCRM Centre of Commercialization for Regenerative Medicine CD235a Glycophorin A cDNA Complementary DNA CFU-B Colony forming unit - basophil CFU-E Colony forming unit – erythroid

CFU-EO Colony forming unit - eosinophil CFU-G Colony forming unit – granulocyte CFU-GEMM Colony forming unit – granulocyte, erythrocyte, monocyte, megakaryocyte CFU-GM Colony forming unit – granulocyte, monocyte CFU-M Colony forming unit – macrophage CFU-MEG Colony forming unit – megakaryocyte CIMFR Canadian Inherited Marrow Failure Registry CLP Common lymphoid progenitor CMP Common myeloid progenitor

x CSF Colony stimulating factor DAPI 4',6-diamidino-2-phenylindole DBA Diamond-Blackfan Anemia DC Dendritic cell DKC Dyskeratosis Congenita DMEM Dulbecco’s Modified Eagle Medium E Embryonic day EB Embryoid body EHP Early hematopoietic progenitor EHT Endothelial to hematopoietic transition ETP Early thymic progenitor FA Fanconi Anemia FCS Fetal calf serum FLT3L Fms-related tyrosine kinase 3 ligand FYSH Fungal Yhr087wp Shwachman G-CSF Granulocyte-colony stimulating factor GATA1 Globin transcription factor 1 GATA2 Globin transcription factor 2 GFI1 Growth factor independent 1 transcription repressor GM-CSF Granulocyte macrophage colony-stimulating factor GMP Granulocyte-monocyte progenitor GVHD Graft-versus-host disease HAND1 Heart and neural crest derivatives-expressed protein 1 HE Hemogenic endothelium HES-1 Hair and enhancer or split-1 hESC Human embryonic stem cell HgF Fetal hemoglobin hiPSC Human induced pluripotent stem cell HLF Hepatic leukemia factor HOXB4 Homeobox b4 HSC/Ps Hematopoietic stem cells/progenitors

xi HSCT Hematopoietic stem cell transplantation IBMFS Inherited bone marrow failure syndrome ICM Inner cell mass IKZF1 IKAROS family zinc finger 1 IL-11 Interleukin-11 IL-3 Interleukin-3 IL-5 Interleukin-5 IL-6 Interleukin-6 IMDM Iscove’s Modified Dulbecco medium Kb Kilobases KDR Kinase insert domain receptor LEF1 Lymphoid enhancer-binding factor 1 Lin Lineage LTR Long-term-reconstituting M-CSF Macrophage colony-stimulating factor Mb Megabuses MCV Mean corpuscular volume MDS Myelodysplastic syndrome MEFs Mouse embryonic fibroblasts MEP Megakaryocyte-erythroid progenitor MIXL1 Mix paired-like homeobox MLP Multilymphoid progenitor MPP Multipotent progenitor MYC v-myc avian myelocytomatosis viral oncogene homolog NAC N-acetylcysteine NEAA Non-essential amino acids NEUROD1 Neurogenic differentiation 1 NK Natural killer OCT4 Octamer-binding transcription factor 4 OP9-DL1 OP9-Delta-like 1 P-Sp Para-aortic splanchnopleura

xii P53 Tumor protein p53 PAX5 Paired box 5 PDGFRα Platelet-derived growth factor receptor α PSC Pluripotent stem cell PU.1 Spleen focus forming virus (SFFV) proviral integration oncogene RBC Red blood cell RNAi RNA interface ROS Reactive oxygen species RRM RNA recognition motif RT Retroviral RT-PCR Real-time polymerase chain reaction RT-qPCR Reverse-transcription quantitative real time PCR RUNX1 Runt-related transcription factor 1 SBDS Shwachman Bodian Diamond Syndrome gene SBDSP Pseudogene of SBDS SCF Stem cell factor SCL Stem cell leukemia factor SDS Shwachman-Diamond Syndrome SEM Standard errors of the mean SeV Sendai SOX17 Sex determining region-Y box 17 SOX18 Sex determining region-Y box 18 SOX8 Sex determining region-Y box 8 SSEA-4 Stage-specific embryonic antigen-4 STR Short-term-reconstituting T Day Tal-1 T-cell acute lymphocytic leukemia-1 TGF- G Transforming growth factor beta TPO Thrombopoietin VEGF Vascular endothelial growth factor α-MEM α-Minimal Essential Medium

xiii Chapter I

Introduction

1.1 Shwachman Diamond Syndrome

Shwachman Diamond Syndrome (SDS) is an inherited bone marrow (BM) failure syndrome (IBMFS) first described in 1964 by Shwachman and Bodian [1, 2]. It is an autosomal recessive, multisystem disorder, characterized most commonly by a triad of bone marrow failure, exocrine pancreatic insufficiency and skeletal abnormalities [3-5].

Patients may further present with immunological abnormalities, hepatic complications, renal and cardiac disease, insulin-dependent diabetes, growth-hormone deficiency and hypogonadotropic hypogonadism, oral disease and cognitive impairment [3, 6-16]. Much like other IBMFSs, SDS is characterized by a high propensity for development of myelodysplastic syndrome (MDS) and leukemia, specifically acute myelogenous leukemia (AML) – with an estimated risk of 36% at age 30 [17, 18]. The incidence of

SDS was estimated as 1 in 77 000 child births, with no racial or gender bias [7, 19, 20].

Data from the Canadian Inherited Marrow Failure Registry (CIMFR) indicates that SDS is the third most common IBMFS, less prevalent only to Diamond-Blackfan Anemia

(DBA) and Fanconi Anemia (FA), with an estimated median survival rate of about 35 years [21, 22].

The molecular basis of the pleiotropic phenotype characteristic of SDS remains to be elucidated, although recent significant advances by Boocock & colleagues have helped shed some light on the genetic basis of this syndrome. In 2003, Boocock & colleagues identified hypomorphic, disease-associated mutations in the Shwachman Bodian

1 Diamond syndrome gene, SBDS in approximately 90% of SDS patients [23]. Several groups have attributed SBDS with having a role in ribosome biogenesis, cell survival, mitotic spindle assembly, chemotaxis, the formation and activity of hematopoietic stem cells and progenitors (HSC/Ps) in the BM, the regulation of reactive oxygen species

(ROS), and the maintenance of genomic stability [24-36]. Although the identification of the SBDS gene has become an important tool in the diagnosis of SDS, and recent studies have helped elucidate some of its functions, further studies are required to define the pathogenic link between SBDS genotype and the SDS phenotype.

1.1.1 Hematological abnormalities

The bone marrow phenotype in SDS is characterized by a reduction in one or more of the myeloid lineages [3]. Neutropenia, typically defined as a neutrophil count less than 1500 x 106/L, is the most common cytopenia, affecting 88-100% of SDS patients [3, 7, 37].

Approximately two-thirds of SDS patients experience intermittent neutropenia (counts of

<1.5 x 109/L), whereas one-third experience persistent neutropenia [38].

SDS patients experience an increased risk of developing recurrent bacterial, viral and fungal infections, in part due to a deficiency in the number and function of neutrophils, which are normally involved in the host defense against pathogens [3, 4, 7, 39-43]. In particular, cases of otitis media, sinusitis, tonsillitis, pharyngitis, mouth sores, oral cellulitis, abscesses, skin infections, septic arthritis and osteomyelitis have been reported

2 in SDS patients, along with deaths as a consequence of bronchopneumonia, septicemia and respiratory tract infections [3, 6, 7, 13, 38-41, 44-48].

Instrumental studies performed by Aggett & colleagues indicate that neutrophil mobility is significantly impaired in SDS patients [44]. Further studies have confirmed that this phenotype, along with a defect in migration and chemotaxis, are observed in most or all

SDS patients [30, 42, 48-52]. A cytoskeletal defect contributes at least in part to the abnormal chemotaxis of SDS neutrophils, due to the unusual surface distribution and mobility of concanavalin-A receptors [50]. Orelio & colleagues investigated the molecular defect responsible for impaired chemotaxis, and determined that upon chemoattractant stimulation, SBDS and F-actin co-localize in neutrophilic cells.

Furthermore, the authors also demonstrated that F-actin dynamics, particularly polymerization and depolymerization in response to chemoattractants are disturbed in

SDS neutrophils,[35].

Little is known about how SBDS contributes to neutrophil production. Loss-of-function experiments in the murine myeloid 32Dcl3 cell line capable of differentiating to mature neutrophils, have demonstrated that although SBDS is not required for neutrophil maturation, its loss does increase the sensitivity of granulocyte precursor cells to apoptotic stimuli [53]. Recent studies by Tulpule & colleagues, using patient-derived

SDS induced pluripotent stem cells (iPSCs) and SBDS-deficient human embryonic stem cells (hESCs) have shown that autodigestion may be responsible for the hematopoietic

3 phenotype, due to a prominent increase in the granular content, and consequent protease activity, in SBDS-deficient hematopoietic cultures [54].

The second most common cytopenia is anemia, affecting 42-82% of SDS patients. It is defined as a hemoglobin concentration of 2 standard deviations below the mean, adjusted for age and sex. A thorough study analyzing the hematological parameters in 21 SDS patients over a 25-year period reported that the majority of patients (66%) presented with normochromic normocytic anemia at time of diagnosis. Fetal hemoglobin (HgF) levels were elevated for age, while the reticulocyte count was reduced in approximately three- quarters of patients [48]. Recent data from 34 SDS cases registered on the CIMFR indicates that approximately 60% of patients presented with anemia, accompanied by high red blood cell (RBC) mean corpuscular volume (MCV). High HgF was also reported in 72% of SDS patients[37]. Likely a sign of “stressed” erythropoiesis, elevated HgF levels have also been associated with an increase in the apoptotic rate of fetal-type erythropoietic progenitors in MDS [55-58]. A recent study showed that SBDS is critical for normal erythropoiesis, as SBDS-depleted CD34+ HSC/Ps and K562 erythroleukemia cells showcased reduced cell expansion during erythroid differentiation, at least in part due to accelerated apoptosis [34].

Thrombocytopenia, defined as a platelet count of less than 150 x 109/L, has been documented in 24-88% of SDS patients, although it is usually mild [3, 7, 37, 38, 48].

Easy bruising has been reported, as well as fatal bleeding in cases of moderate to severe thrombocytopenia [49, 59-63].

An analysis of SDS patients registered with the CIMFR reported that 24% of patients presented with bilineage cytopenia at time of diagnosis [37]. Pancytopenia has also been reported in 10-65% of SDS patients, neutropenia being the most severe, with mild to moderate anemia and thrombocytopenia. Some pancytopenia patients have also been documented to develop severe aplastic anemia [3, 7, 38, 48].

It has been postulated that the hematopoietic phenotype is a result of a stem cell defect, as

Dror and Freedman reported that SDS patients have significantly reduced levels of BM

CD34+ HSC/Ps, that showcase a reduced ability to generate hematopoietic colonies in vitro [32]. Although the hematopoietic defect primarily affects the myeloid lineage, deficiencies in the number and function of cells of the lymphoid lineage have also been reported. Specifically, low numbers of total circulating B and T-lymphocytes, and natural killer (NK) cells have been documented in SDS patients. B-lymphocytes have also been reported to have an impaired ability to produce specific antibodies, and much like T- lymphocytes, to proliferate in-vitro [6, 13].

There is no clear correlation between BM cellularity and severity of cytopenia, as varying degrees of fat infiltration, as well as hypocellular, normocellular and hypercellular marrows have been observed [3, 7, 32, 48].

5 1.1.2 Non-hematological abnormalities

Exocrine pancreatic insufficiency is another hallmark of SDS [7]. Imaging and pathological studies often reveal a small pancreas, with prominent fatty replacement of the pancreatic acinar cells, but relatively normal ducts and islets [38, 64]. SDS patients commonly present with impaired pancreatic enzyme secretion, as reflected by low serum trypsinogen levels in 91% of patients, low serum amylase levels in 73% of patients, and low serum pancreatic isoamylase levels in 97% of SDS patients [3, 7, 65]. As a result of the exocrine pancreatic insufficiency, patients experience steatorrhea, malabsorption, deficiencies in the levels of the A,D,E,K fat-soluble vitamins, and ultimately, failure to thrive [1, 7, 38]. Despite receiving pancreatic enzyme replacement treatments, SDS patients have a tendency to remain below the 3rd percentile for height [3, 7, 38]. With increasing age, the pancreatic deficiency resolves in up to 50% of patients, and patients become pancreatic-sufficient and able to absorb fat normally. The reason for this improvement remains unclear however [7, 43]. Pancreatic endocrine function seems generally unaffected in the vast majority of SDS patients [66].

Skeletal abnormalities are common in SDS, although their localization and severity varies with age [67]. The most common skeletal abnormality is metaphyseal dysostosis involving the femoral head [3, 7, 67, 68]. Metaphyseal dysostosis is usually asymptomatic, and affects approximately 50% of patients [3, 4, 37, 68]. One third to one half of patients also have rib and/or thoracic cage abnormalities that might lead to thoracic dystrophy and respiratory failure during the newborn period [3, 7, 38, 41, 69].

Cases of progressive spinal deformities, short ribs, digit abnormalities, slipped femoral

6 epiphysis and pathological fractures have also been described [3, 7, 37, 38, 67, 70]. SDS has also been associated with low turnover osteoporosis, characterized by low bone mass and bone turnover, consequently leading to vertebral fragility and fractures [37, 71].

Liver abnormalities including elevated serum liver enzymes and hepatomegaly are seen in 50-75% of patients, although they are rarely severe, and resolve with age [38, 43, 47].

Developmental delay, eczema, dental caries and periodontal disease have also been documented [3, 37]. Heart, kidney, eye, testes, craniofacial structure and nervous system abnormalities have been reported, albeit infrequently [3, 7, 12, 43, 72-74].

SDS is also characterized by an increased propensity for malignant myeloid transformation into MDS and AML, of approximately 19% and 36% by ages 20 and 30, respectively [75]. A literature review in 2005 reported that of 54 SDS patients diagnosed with MDS/AML, 37 developed marrow cytogenetic abnormalities, at a median age of 8 years with a range of 2-42 years [76]. It is unclear whether solid tumors are part of the

SDS phenotype, although four isolated cases of pancreatic adenocarcinoma, breast cancer and dermatofibrosarcoma in SDS patients were recently reported [77-79].

Several IBMFSs, including DBA, dyskeratosis congenital (DKC), SDS and cartilage-hair hypoplasia are characterized by impairments in ribosome biogenesis or function, and further, cancer predisposition. It is yet unclear if and how the ribosomal abnormalities contribute to the cancer predisposition in SDS patients [80]. As reviewed by Ruggero & colleagues however, several studies have indicated that there is an association between

7 ribosome production and neoplastic transformation [81]. Specifically, it has been shown that several tumor suppressors such as tumor protein p53 (p53) and proto-oncogenes such as nucleophosmin, either affect the initiation of protein synthesis or the production of mature ribosomes [81, 82]. Studies in zebrafish show that inactivating mutations in ribosomal protein genes predispose the mutants to tumor development [83, 84].

Additionally, decreased levels of the RPS14 ribosomal subunit protein were found in patients with 5q- myelodysplastic syndrome, further indicating that there is an association between ribosome biogenesis and tumor predisposition [85]. Considering the above- mentioned studies, it is reasonable to associate SBDS mutations with an increased cancer predisposition, as a result of an SBDS-induced defect in ribosome biogenesis. However, leukemic transformation might be related to other functions of the SBDS gene (see below).

1.1.3 Treatment

The only curative option for the hematological complications associated with SDS, including marrow failure and MDS/AML, is hematopoietic stem cell transplantation

(HSCT) [86, 87]. Unfortunately, a review of the literature indicates that HSCT is associated with a significant mortality and morbidity rate, as a result of graft-versus-host disease (GVHD), relapse and transplant-related toxicities. Severe to fatal organ toxicities, including cardiac, neurological, hepatic, pulmonary and immune, due to intensive preparative therapies, are also a significant contributor to the poor outcome of HSCT in

SDS patients [63, 88-90].

8 Although not curative, non-transplant management options are available, and include chronic transfusion programs and cytokine combinations. Granulocyte-colony stimulating factor (G-CSF) is one such example, shown to induce a clinically beneficial response in severely neutropenic SDS patients, by promoting granulopoiesis and reducing the risk of infections [91-93]. Adverse effects are associated with G-CSF treatment however, and include headaches, musculoskeletal symptoms, bone pain and osteopenia [94]. Further, the risk of developing MDS/AML has been reported to increase in SDS and other IMBFS such as congenital neutropenia, following G-CSF treatment [95-97]. Immunosuppressive therapies using corticosteroids and cyclosporine typically do not result in significant response [61, 98]. Although their need may diminish with age, oral pancreatic enzyme supplements and fat-soluble vitamins are used to manage the pancreatic exocrine deficiencies seen in SDS patients [86, 99].

Due to limited efficacy or toxicity of the treatments above, the life expectancy of SDS patients is substantially reduced (about 35 years), and novel therapeutic strategies are critically needed. Since the main morbidity and mortality are related to the blood dyscrasia, studying hematopoiesis may help characterize the hematological phenotype, and develop targeted and effective treatments for the SDS-associated hematological complications.

1.1.4 Characterization of Shwachman-Bodian Diamond Syndrome gene

SDS is associated with mutations in the SBDS gene in 90% of patients [23], and biallelic inactivating mutations are the cause of this autosomal recessive disease.

SBDS is located at the 7q11 centromeric region of chromosome 7, and it encodes for a

250 amino acid protein [19, 23]. It is composed of five exons, encompassing 7.9 kilobases (kb), and its complementary DNA (cDNA) transcript is 1.6 kb (Fig. 1). 5.8 megabases (mb) distal to the SBDS gene is a paralogous duplicon containing a pseudogene called SBDSP. SBDSP shares 97% homology with SBDS, but due to certain deletions and nucleotide changes, its coding potential is impaired. Approximately 75% of

SBDS mutations are due to gene conversions with SBDSP, with 60% of SDS patients carrying two converted alleles, and almost all published SDS patients with biallelic SBDS mutations, carrying at least one.

The most common conversion mutations are 183-184TA>CT in exon 2 and 258+2T>C in the donor splice site of intron 2, leading to an immature stop codon, and an alternative splice site leading to premature truncation by frameshift, respectively (Fig. 1). Compound heterozygotes as well as homozygotes have been identified with respect to 183-

184TA>CT and 258+2T>C mutations, and 258+2T>C mutations, respectively. However, no homozygotes have been identified with respect to 183-184TA>CT, indicating that this severe genotype may lead to early lethality. Aside from conversion mutations, there are also other changes that may occur in the coding region of SBDS, leading to frameshift, nonsense and missense mutations [23]. Most of these mutations lead to a significant reduction in the amount of SBDS protein produced [22].

10 NMR spectroscopy of human SBDS indicates that much like its archaeal ortholog, it is composed of three well-folded domains. Ferreira de Oliveira & colleagues identified a classic RNA binding site at the N-terminal domain – which is also the site of most SDS- associated mutations [23, 100] The N-terminal domain that has been termed fungal,

Yhr087wp, Shwachman (FYSH) by the Warren group [101], is composed of 4 alpha (α) helices and 4 beta (β) strands, connected to the central domain by a flexible linker. The central domain forms a three-helical bundle, which connects to the C-terminal domain by a short flexible linker. The C-terminal domain is composed of a four-stranded anti- parallel β-sheet and two α-helices, folded into an RNA recognition motif (RRM). Further, its surface is characterized by a large acidic path, which indicates that the C-terminal domain may be a site of protein-protein interaction (Fig. 2) [100].

SBDS is highly conserved throughout evolution, expressed in plants, yeast, archaea, and vertebrates. Although SBDS is highly expressed at both the mRNA and protein level throughout human tissues, its biochemical function has yet to be fully elucidated [23,

102-104].

Several studies indicate that SBDS may play a role in chemotaxis, genomic stability and mitotic spindle formation, but the best-characterized functions are within the context of ribosome biogenesis and cell survival [24-31, 33, 100]. Studies from our own lab have shown that SBDS-deficient cells are hypersensitive to Fas ligand, and therefore experience increased apoptosis and reduced growth, through a Fas-mediated pathway

[26-28, 33]. Using SDS patient cells and Sbds-deleted mice, Burwick & colleagues and

11 Finch & colleagues, respectively, demonstrated that ribosome maturation is an SBDS- dependent process. They showed that SBDS helps promote the displacement of eIF6 from the 60s subunit; and that only once this has occurred, can the 60s subunit associate with the 40s subunit to form the mature 80s monomer. SDS patient cells were characterized by lower 80s subunit levels, indicating that much like what has been described in yeast, SBDS plays an important role in ribosome maturation [24, 25, 105,

106].

Ribosomes are integral to the process of protein translation, and a deficit in the amount of available ribosomes may have an impact not only on cell survival, but also on expansion and adaptation to changing environmental conditions [107]. Therefore, a reduction in the ability to efficiently translate cellular proteins may be at least in part responsible for the

SDS phenotype. Additionally, studies have shown that ribosome-related proteins such as

SBDS play integral roles in apoptosis, DNA repair and transcriptional regulation, and that their deregulation is associated with IBMFSs [80, 108-110].

Although the studies outlined above provide evidence that SBDS plays an important role in cell survival and ribosome biogenesis, its specific role in hematopoiesis has yet to be elucidated. The use of patient-derived iPSCs will allow us to further investigate the impact that mutations in SBDS, and a consequent reduction in SBDS production will have on the in-vitro process of early embryogenesis, specifically in the context of mesoderm and hematopoietic differentiation.

Fig 1. SDSD gene structure

Map of the SBDS gene, which contains 5 exons. The locations of the two most common mutations (258+2 T>C, 183-184 TA>CT) are also depicted (Adapted from [23])

Fig 2. Human SBDS protein structure

Ribbon representation of SBDS, containing 3 independent domains: N-terminal domain (residues 1-95) involved in RNA binding, central domain (residues 107-167), and C-terminal domain (residues 173-250), that contains a RNA recognition motif and is involved in protein-protein interaction. (Adapted from [100])

13 1.1.5 Current disease models

The use of patient samples for the study of hematological complications is ideal, although it is often not feasible due to the hypocellular nature of the SDS BM. Using SDS patient- derived samples, Dror & colleagues demonstrated that CD34+ HSC/Ps have a significantly reduced ability to generate hematopoietic colonies in vitro [32]. Further, it was shown that SDS patient-derived hematopoietic progenitors undergo accelerated apoptosis, as a result of hyperactivation of the Fas signaling pathway [27].

Rawls & colleagues generated a model for the study of hematopoiesis in SDS via the lentiviral-mediated RNA interference (RNAi) of SBDS in transplanted murine hematopoietic progenitors. This was the first study to demonstrate that a loss of SBDS was sufficient to cause hematological abnormalities, as SBDS-deficient murine hematopoietic progenitors experienced an increased difficulty to undergo granulocytic differentiation and to produce myeloid progenitors [111].

SBDS-null mouse embryos fail to progress past embryonic day 6.5, indicating that SBDS is developmentally essential [112]. Recently, Tourlakis et al developed a pancreas- specific SBDS knockout mouse model, with a pancreatic phenotype similar to that observed in humans. A BM-specific SBDS knockout mouse model has yet to be explored

[113].

14 A zebra fish model was generated by morpholino-mediated SBDS knockdown at the 1- cell stage embryo, that showed a defect not only in the pancreas but also during granulopoiesis, and could therefore be used to study hematopoiesis [114].

Although animal models provide us with an invaluable resource for the study of the function of the SBDS gene, they have substantial limitations. Animals and humans differ in their biochemical, physiological and anatomical properties [115, 116]. For example, murine models of FA demonstrate defects in their DNA repair ability, much like FA patients. However, unlike FA patients, they do not develop spontaneous BM failure, a hallmark of this disease [117]. Attempts to model the bone marrow phenotype of other

IBMFSs, such as dyskeratosis congenita and Diamond Blackfan anemia in mice also resulted in similar challenges with respect to recapitulating the human diseases [118-

120].

By deriving hESC lines and human induced pluripotent stem cell (hiPSC) lines from patients carrying SDS-causing mutations, Park & colleagues were able to bypass some of the difficulties faced when using animal models to study human genetic diseases [121].

Further, hESC/hiPSC provide us with an unlimited number of patient material, and also allow for the study of early developmental dysfunction in genetic diseases such as SDS.

Tulpule & colleagues validated SDS hESC and retrovirally-derived hiPSCs as an appropriate model for the study of SDS. They found that SBDS-deficient hESCs and hiPSCs manifested impaired hematopoietic development towards myeloid cells, at least in part due to protease-mediated autodigestion [54]. However, analysis of hematopoietic

15 development was limited. Although Tulpule & colleagues used a differentiation protocol that gave rise to hematopoietic progenitors characteristic of definitive hematopoietic cells found within the aorta-gonad/mesonephros (AGM) region, they only assessed for the expression of CD45 in their cultures [54, 122]. They did not investigate the ability of

SBDS-deficient PSC to undergo the step-wise differentiation towards hemogenic endothelium (HE), early hematopoietic progenitors (EHPs), and mature blood cells such as myeloid cells and granulocytes; and did not address whether the hematopoietic defect was also characteristic of the primitive wave. Despite determining that SBDS-deficient

PSCs manifest a disruption in hematopoietic development, Tulpule & colleagues did not investigate the stage at which stage this defect may be initiated – as is the aim of our current study.

1.2 Hematopoietic development and hematopoietic differentiation

models

Hematopoiesis is the lifelong production of blood cells, including the RBCs, megakaryocytes, monocytes/macrophages, mast cells, neutrophils, basophils and eosinophils of the myeloid lineage, and the B-lymphocytes, T-lymphocytes and natural killer cells of the lymphoid lineage [123]. The different stages of hematopoiesis, growth factors and transcriptional regulators have been mainly studied in animal models, including the mouse, zebra fish and bird. [124-126]. However, a growing appreciation for species-to-species differences has driven the generation and use of humanized mouse models, hESC and hiPSCs to study human hematopoiesis and mechanisms of human diseases [123, 125].

Hematopoiesis is initiated during embryonic development in the extra-embryonic mesoderm. Extra-embryonic mesoderm is derived from the primitive endoderm of the hypoblast. The third week of embryonic development marks the initiation of primitive hematopoiesis, with the appearance of blood islands in the mesoderm of the extraembryonic yolk sac (Fig. 3) [127, 128]. The mesodermal germ layer, responsible for blood formation, is characterized by an upregulation of Brachyuri, and an increase in the cell-surface expression of kinase insert domain receptor (KDR) and platelet-derived growth factor receptor α (PDGFRα) [129-134].

Primitive hematopoiesis is characterized by the presence of the hemangioblast, a common precursor to the hematopoietic and endothelial cells of the blood islands [133]. The hemangioblast transitions through a hemogenic endothelium stage, to give rise to a transient population of nucleated erythroblasts expressing embryonic hemoglobins (ζε), macrophages and megakaryocytes [135-138].

Primitive streak formation, and further, gastrulation of the inner cell mass (ICM) allows for the development of intra-embryonic mesoderm, alongside endoderm and ectoderm

[128, 139]. A second wave of hematopoiesis known as definitive hematopoiesis is initiated in the intra-embryonic mesoderm, as early as embryonic day 19, in the human para-aortic splanchnopleura (P-Sp)/ AGM [140, 141].

17 Unlike primitive hematopoiesis, embryonic definitive hematopoiesis is defined by the generation of HSCs from hemogenic endothelium (HE), a specialized population of endothelial cells [124, 142]. Studies using hESCs indicate that HE cells are characterized by cell-surface expression of CD31, CD34, CD117, KDR and CD144 [143, 144]. HE- derived HSCs are multipotent stem cells, able to give rise to all blood cells of both the myeloid and lymphoid lineages, for the entire lifespan of the individual [124].

All fetal and adult sites of blood formation are not intrinsically hematopoietic, and must therefore be colonized by HSCs [123, 145]. At approximately 30 days of gestation, HSCs colonize the fetal liver. Although the liver is the chief organ responsible for blood formation until the sixth month of fetal development, hematopoiesis can also be observed in the spleen and thymus by the third month of gestation [22, 146, 147]. Week 10.5 marks the colonization of the BM with HSCs – deeming it the primary site of blood formation for the rest of the fetal and adult life of the individual (Fig. 3) [148].

Although both fetal and adult HSCs lie at the top of the hematopoietic hierarchy, studies in mice indicate that their properties differ. Fetal liver HSCs cycle at a much higher rate than do adult HSCs of the BM, which are predominantly quiescent [149, 150]. However, both fetal and adult HSCs are multipotent, able to self-renew and maintain the hematopoietic stem cell pool, while also differentiating in a sequential manner, into progenitor cells and then mature blood cells of all hematopoietic lineages [140, 142, 151].

18 When transplanted, long-term-reconstituting HSCs (LTR-HSCs) have the ability to reconstitute the BM for the rest of the lifetime of the individual. During hematopoiesis,

LTR-HSCs differentiate into short-term-reconstituting HSCs (STR-HSCs), and subsequently into multipotent progenitors (MPPs), which although multipotent, have only a transient engraftment capacity, due to reduced self-renewal ability [152, 153].

Classically, MPPs further differentiate into either common myeloid progenitors (CMPs) or common lymphoid progenitors (CLPs), which upon further specification, give rise to all cells of the myeloid and lymphoid lineages, respectively.

The hematopoietic hierarchy is probably more complex than the scheme described above.

Indeed, a recent study points to the presence of a multilymphoid progenitor (MLP) rather than a CLP, that aside from giving rise to all lymphoid lineages, may also hold myeloid potential [125, 154]. CMPs further bifurcate to give rise to granulocyte-monocyte progenitors (GMPs) and megakaryocyte-erythroid progenitors (MEPs). GMPs differentiate into granulocyte and macrophage colony-forming units (CFU-G and CFU-

M, respectively), which proceed to form the mature polymorphonuclear leukocytes

(neutrophils in specific), and monocytes/macrophages, respectively. MEPs give rise to megakaryocyte colony-forming units (CFU-MEG) and erythroid burst-forming units

(BFU-E), which go on to form the erythroid colony-forming units (CFU-E). Further differentiation of the CFU-MEGs and CFU-Es gives rise to platelets, and erythrocytes, respectively [125]. The basophil colony-forming units (CFU-B) which give rise to mature basophils, and the eosinophil colony-forming units (CFU-Eo) which differentiate into the mature eosinophils, are derived from the CMP [22]. On the lymphoid side, MLPs give

19 rise to early thymic progenitors (ETPs) and B/NK cell progenitors, which ultimately produce mature T-cells, B-cells and NK cells. It was recently shown that although MLPs are lymphoid-skewed, they also have myeloid potential, giving rise to monocytes and granulocytes, but not erythrocytes or mekagaryocytes. Similarly, dendritic cells (DCs) may be derived from both a GMP and MLP source [125, 154].

Blood lineage commitment is a complex and stepwise process, with each differentiation step being governed by the expression of specific cell surface markers, activation and inhibition of various regulatory networks, transcription factors, and multiple growth factors [125].

Fig 3. Sites of human hematopoietic development during fetal life and early infancy

Hematopoiesis is initiated in the yolk-sac blood islands during the third week of embryonic development, with the beginning of the primitive wave. Definitive hematopoiesis is also initiated as early as day 19 of development within the AGM. By the 30th day of gestation, HSCs colonize the fetal liver, which becomes the chief organ responsible for blood formation until the 6th month of gestation. By the third month of gestation, hematopoietic activity can also be observed in the spleen. At week 10.5, the HSCs colonize the bone marrow, and the bone marrow becomes the primary site of hematopoiesis for the rest of the lifetime of the individual. (Adapted from [22])

20 1.2.1 Primitive hematopoiesis

Although the earliest stages of human development are fairly inaccessible for experimentation, studies using hESCs and specimens derived from elective pregnancy terminations have helped shed some light on the development of the primitive hematopoietic system [128].

The development of the hemangioblast marks the onset of primitive hematopoietic commitment. Exposure of hESCs to bone-morphogenetic protein – 4 (BMP-4) followed by basic fibroblast growth factor (bFGF) and Activin A (Act.A) promotes the induction of nascent mesoderm, as mirrored by a decrease in octamer-binding transcription factor 4

(OCT4) expression and a transient increase in the expression of brachyuri [155]. Studies by Choi & colleagues indicate that the mesoderm responsible for hemangioblast formation expresses both the apelin receptor (APLNR) and PDGFRα [143]. Keller’s group has shown that Activin/Nodal and Wnt signaling pathways control the specification of mesoderm towards the primitive or definitive hematopoietic lineages, and that CD235a positivity could be used as a distinguishing marker between the two populations [138, 156]. Exposure to vascular endothelial growth factor (VEGF) promotes further commitment of the CD235a+KDR+ mesoderm towards the hemangioblast, as seen by the sequential up-regulation of CD117, CD34 and CD31. The hemangioblast population further gives rise to a population of CD34+CD235a+ cells responsible for the development of erythroid progenitors expressing ε-globin, and a second, CD34+CD43- population. The CD34+CD43- population rapidly transitions through a hemogenic endothelium stage, characterized by the expression of the Runt-related transcription

21 factor 1 (RUNX1), globin transcriptin factor 1 (GATA1), hematopoietic stem cell leukemia factor (SCL) and LOMO2. The primitive CD34+CD43- hemogenic endothelium gives rise to ε and γ -globin expressing erythroblasts, as well as myeloid cells

(macrophages and megakaryocytes) [138]. Much like their yolk-sac-derived counterparts, hESC/hiPSC-derived erythrocytes are CD235a+ (glycophorin A) but CD34-, while monocyte-macrophages are CD14+, CD18+, CD11b+, CD115+,CD34+ and megakaryocytes are CD41a+ [140, 157-159].

1.2.2 Definitive hematopoiesis

Through the process of gastrulation, the PS gives rise to the intra-embryonic mesodermal germ layer, and later the P-sp/AGM, which is responsible for HE development during definitive hematopoiesis.

Studies using hESCs indicate that much like with extra-embryonic mesoderm, PS formation is induced in response to BMP-4, in conjunction with active bFGF and transforming growth factor beta (TGF-β)/Nodal/Activin signaling [160]. Several groups have further shown that non-canonical Wnt signaling is an important modulator in mesoderm induction and its further specification [138, 161, 162]. The development of the

PS and further, the mesoderm, is mirrored by a rapid upregulation of brachyuri and mix paired-like homeobox (MIXL1), along with KDR and PDGFRα cell-surface expression

[131-134]. Studies by Kennedy & colleagues have shown that although Activin/Nodal signaling is required for mesoderm induction, HE development, and subsequent definitive hematopoietic development is not dependent on it [156]. In response to VEGF, mesoderm

22 gives rise to a CD34+CD31+KDR+VE-CAD+CD117+CD45-CD43- population containing the HE, much like what has been observed in the P-sp/AGM region [163].

As HE undergoes an endothelial to hematopoietic transition (EHT), newly formed hematopoietic cells, including HSCs, emerge from the endothelial lining of the aorta as distinct clusters [144]. Sex determining region – Y box 17 (SOX17) plays an integral role in the regulation of HE formation, and its transition towards a hematopoietic fate – as overexpression leads to the expansion of a population with HE characteristics, whereas knockdown inhibits the generation of hematopoietic progenitors [164]. Studies using hESCs have shown that in synergy with bFGF, VEGF plays an integral role during the

EHT of HE [165]. Similarly, RUNX1 has been shown to play an integral role during the

EHT of HE in mice [166]. Canonical Wnt is responsible for hematopoietic differentiation, post-HE formation [162].

The HSCs generated in the Psp/AGM region go on to colonize the fetal liver, and later from there, the BM – making it the primary site of definitive hematopoiesis for the rest of the lifetime of the individual [140]. HSCs are defined as being CD34+CD38-

CD90+CD45RA-CD49f+ as well as negative for all hematopoietic lineage markers (Lin-)

[167-170]. The generation of HSCs is associated with the upregulation of ID genes,

NFIB, Sex determining region – Y box 8 (SOX8) and Sex determining region – Y box 18

(SOX18), among others [152]. Both homeobox B4 (HOXB4) and BMI1 proto-oncogene

(BMI1) are involved in the self-renewal of HSCs, whereas murine studies indicate that both growth factor independent 1 transcription repressor (GFI1) and p53 are responsible

23 for the maintenance of HSC quiescence [171-174]. Hairy and enhancer of split-1 (HES-1) and hepatic leukemia factor (HLF) are involved in the maintenance of repopulating capacity, by enhancing the cell cycle and inhibiting the apoptosis of HSCs, respectively

[175]. Also, antagonism of the aryl hydrocarbon receptor (AHR) promotes the expansion of HSCs [176]. Further, BMP-4 and Wnt are involved in the maintenance of HSC homeostasis, by promoting self-renewal [177, 178]. Although Notch signals play an integral role in the in-vitro expansion of HSCs, recent literature indicates that this is not a requirement in-vivo , in part due to spatial and temporal differences between in-vitro and in-vivo culture conditions (Fig. 4) [179].

The progeny of the LT-HSC is the MPP, and its generation is associated with the upregulation of v-myc avian myelocytomatosis viral oncogene homolog (MYC) and

IKAROS family zinc finger 1 (IKZF1) [152]. Aside from being CD34+CD38- CD45RA-

Lin-, the MPP is further characterized by the loss of CD49f and CD90 expression [152,

180]. Adenylate kinase 2 (AK2) plays an integral role in the survival and metabolism of

MPPs, as mutations in this gene are responsible for defects in the growth and differentiation of both myeloid and lymphoid cells [181]. Downstream of MPPs are the

CD34+CD38+CD45RA-CD135+CD10-CD7-Lin- CMPs and CD34+CD38-

CD45RA+CD10+CD7-Lin- CLPs (Fig. 4).

Aside from experimental studies of hematopoiesis, clinical data from patients with hematological disorders also helps provide some insight into regulatory mechanisms responsible for hematopoietic differentiation in humans. Under the control of spleen

24 focus forming virus (SFFV) proviral integration oncogene (PU.1), CMPs give rise to

CD34+CD38+CD45RA+CD135+CD10-CD7-Lin- GMPs [182]. Clinical data show that mutations in RUNX1 and C/EBPα are associated with MDS and AML; these studies thus indicate that RUNX1 and C/EBPα are key regulators of myeloid development, and that together with GFI1 and lymphoid enhancer-binding factor 1 (LEF1), promote generation of granulocytes from GMPs [183-185]. With the upregulation of IRF8 expression,

RUNX1 and C/EBPα expressing GMPs give rise to monocytes and DC [186, 187]. CMPs may also differentiate into MEPs, which aside from being CD34+CD38+ CD10-CD7-Lin-, are also characterized by the loss of CD45 and CD135 expression. In the presence of

PU.1 and GATA-1, MEPs differentiate further to give rise to megakaryocytes and platelets, or with the upregulation of B-cell CLL/lymphoma 11A (BCL11A) also, to erythrocytes [188, 189].

GATA2 is in part responsible for the differentiation of MPPs to CLPs, which in the presence of PU.1, give rise to B/NK cell progenitors. [190-192]. B/NK cell progenitors further give rise to NK cells, and in the presence of paired box 5 (PAX5) and IKZF1, B cells [193]. T-cells are generated from CLPs, under the control of NOTCH-1 (Fig. 4)

[194].

Aside from transcription factors, the various differentiation steps in hematopoiesis are also dependent on the interaction of hematopoietic cells with growth factors and cytokines. Various colony stimulating factors (CSF), including G-CSF, macrophage colony-stimulating factor (M-CSF), interleukin-5 (IL-5), EPO and thrombopoietin (TPO)

25 act upon relatively late hematopoietic progenitors in a lineage-specific manner to give rise to neutrophils, monocytes, eosinophils, erythrocytes, and megakaryocytes and platelets, respectively [195-198]. Growth factors may also act upon less committed hematopoietic progenitors, therefore promoting the growth of more than one hematopoietic lineage. They include granulocyte macrophage colony-stimulating factor

(GM-CSF) and interleukin-3 (IL-3), which aside from supporting the development of

DCs, also promote the production of granulocyte and monocyte colonies, and multilineage, erythrocyte and megakaryocyte colonies, respectively [199, 200]. In conjunction with other growth factors such as EPO, IL-9 supports erythrocyte growth

[201]. Synergistically with other growth factors, both stem cell factor (SCF) and Fms- related tyrosine kinase 3 ligand (FLT3L) act much earlier on the very early progeny of

HSCs, to promote the growth of multilineage hematopoietic colonies [202, 203]. FLT3L is also in part responsible for the survival of HSCs [204]. The pleiotropic cytokines interleukin-6 (IL-6) and interleukin-11 (IL-11) can also promote the generation of more committed progenitors from HSCs, although IL-6 is more potent in inducing B and T-cell development, whereas IL-11 is more potent in the induction of megakaryocytes (Fig. 4)

[22, 205, 206]. Although hematopoiesis has been studied extensively, much of what is currently known has been derived from mouse studies, and extrapolated to human hematopoiesis. However, because mice and humans are inherently different, there is a need to further investigate hematopoiesis in humans. Further, deeper insight into normal hematopoietic development and the signaling mechanisms responsible for differentiation will help shed some light on the impact that molecular aberrations can have on hematopoiesis. Specifically, a more precise understanding of normal hematopoiesis will

26 help provide a deeper understanding of hematopoietic development in disease states such as SDS.

Fig 4. Current model of human hematopoietic hierarchy

Essential transcription factors and growth factors relevant for each stage of development, starting with the long-term hematopoietic stem cell are shown. Transcription and growth factors important for the survival and self-renewal of HSCs are shown in the dark and light orange boxes, respectively. Transcription factors relevant for multipotent and committed progenitors are shown in the light green boxes. This list is not exhaustive, and details regarding the factors are included in the text. (Adapted from [22, 125]

27 1.2.3 Hematopoietic differentiation model using control PSCs hESC are derived from the inner cell mass of the human blastocyst-stage embryo, and are able to give rise to all three germ layers, and ultimately all cell types of the human body

[207, 208]. iPSCs are pluripotent stem cells derived from adult somatic cells via the forced expression of embryonic stem cell transcription factors, which help drive pluripotency [209]. In the presence of defined factors, hiPSCs can generate the three germ layers, including mesoderm, and further more differentiated cell types, such as those of the hematopoietic lineage [209]. Due to their ability to self-renew indefinitely, and their pluripotent nature, hESCs and hiPSCs provide us with the unprecedented opportunity to study early hematopoietic development in vitro [210].

Multiple groups have demonstrated hematopoietic development with hESCs and hiPSCs in the presence of cytokines and various growth factors, using one of the three aforementioned approaches: forming embryoid bodies (EBs) from pluripotent stem cells

(PSCs), culturing on supportive stromal layers such as OP9s, or as a monolayer on extracellular matrix proteins such as collagen and fibronectin [132, 133, 156, 211-216].

However, for the purpose of this dissertation, only the EB method of differentiation established by Kennedy & colleagues will be further described [133, 156].

The method described below allows for the stepwise differentiation of PSCs towards the

PS/mesoderm, hemangioblast, hemogenic endothelium, EHP, and mature blood cells, including erythrocytes, granulocytes and macrophages (Fig. 5).

28 To begin the process of hematopoietic differentiation, PSC colonies previously maintained on mouse embryonic fibroblasts (MEFs) are dissociated into small cell clumps, and placed in suspension cultures supplemented with BMP-4, to give rise to EBs.

Due to its three-dimensional structure, the EB’s spatial organization resembles that of the human embryo, and is therefore able to give rise to all three germ layers: ectoderm, mesoderm and endoderm [217]. In the presence of BMP-4, the EBs begin to lose expression of pluripotency markers, such as OCT4, stage-specific embryonic antigen-4

(SSEA-4) and T-cell acute lymphocytic leukemia 1 (Tal-1), while upregulating expression of Brachyuri – an indication that the PS and/or mesoderm has been induced.

CD117 and KDR are expressed on PSCs at day 0 of differentiation, but following one day of induction with BMP-4, there is a dramatic drop in the expression of both receptors. Following induction of CD56+ mesoderm with bFGF and Act.A, CD117 and

KDR re-appear on different cell populations between days 3 and 4 of differentiation

[218]. Treatment with Act.A gives rise to both waves of hematopoiesis, although primitive hematopoiesis, which is characterized by generation of the hemangioblast, predominates (Fig. 5).

Hemangioblast specification of the mesoderm is characterized by the emergence of a

KDR+CD117-CD31-CD235a+ population, as evidenced by its ability to generate blast colonies with both endothelial and hematopoietic potential. There is also a KDR-CD117+ population at this stage of differentiation, but it severely lacks hematopoietic potential. In the presence of VEGF, by day 5 of differentiation, a small KDR+CD117+CD34+CD31+ population is present, marking the onset of primitive hematopoiesis [133, 138]. To assay

29 the extent of primitive hematopoiesis, the expression patterns of CD34, CD43, CD41a,

CD45 and CD235a may be monitored. In the presence of various hematopoietic cytokines and growth factors, by day 9 of differentiation, a CD43+ population of CD34+ origin responsible for primitive hematopoiesis is generated, as evidenced by the fact that it is the sole population able to give rise to ε-globin expressing erythroid, myeloid and erythroid/myeloid progenitors. Expression of CD41a and CD235a, but lack of CD45 on the CD43+ population is further evidence that this population contains erythroid and mekagaryocyte progenitors characteristic of primitive hematopoiesis. Although the

CD34+CD43+ population declines over time, the CD34-CD43+ population increases, and a

CD45+ population emerges at day 13 of differentiation (Fig. 5).

Kennedy & colleagues concluded that although the primitive program is dependent on the activation of the Activin/Nodal pathway, this is not the case for definitive hematopoiesis.

Therefore, the inhibition of the Activin/Nodal pathway between day 1-2 of differentiation using the small molecule SB-431542 allows for the enrichment of the CD34+CD43- population with the potential to generate definitive hematopoiesis. Using this approach, a

HE population was identified at day 9 of differentiation, expressing markers indicative of both endothelial and hematopoietic progenitors, including CD34, CD31, VE-Cad, KDR and CD117. In addition, when this population is plated on OP9-DL1 cells, it undergoes an EHT to generate cells of both the myeloid and erythroid lineages, and T-cells when plated on OP9-DL4 cells. As lymphoid potential is a characteristic of definitive hematopoiesis only, the ability to generate T-cells is a clear indication that definitive hematopoiesis is initiated within the CD34+CD43- population at day 9. By day 12 of

30 differentiation, a CD43+ population does form, but unlike primitive hematopoiesis, this

CD43+ population is enriched for myeloid precursors and expresses CD45+ rather than

CD41a or CD235a. In definitive hematopoiesis, CD41a and CD235a are expressed, but at later stages of differentiation, and are confined to the megakaryocyte and erythroid lineages, respectively (Fig. 5) [156].

Fig 5. Model of primitive and definitive hematopoiesis using pluripotent stem cells

The developmental stages leading to mature blood cells are shown, starting with the pluripotent stem cell. The various markers and genes characteristic of the different stages of differentiation are depicted, as well as the signaling pathways responsible for the transition from one cell population to another. Primitive hematopoiesis is seen in a light red box, whereas definitive hematopoiesis is seen in a light green box. (Adapted from [138, 156]).

31 1.3 Disease-specific hiPSCs

Considering that SDS is a rare disease, characterized by a hypocellular BM, it is rather difficult to attain sufficient patient material for the study of its pathogenesis. As per the limitations described previously, animal models frequently do not recapitulate the bone marrow phenotype in IBMFSs. Therefore, patient-derived iPSCs may help circumvent many of the difficulties associated with studying rare, genetic diseases such as SDS

[219]. Considering that iPSCs are derived in the context of one’s genetic identity, several groups have validated their use as in-vitro disease models.

In recent years, iPSCs lines have been generated from patients and used to study not only neurological, metabolic, cardiovascular and primary immunodeficiency diseases, but also hematological conditions, including, but not limited to FA, DBA and DKC [110, 220-

222].

In 2007, it was shown for the first time that human iPSCs can be generated via the retroviral (RT) transduction of the OCT4 (O), KLF4 (K), SOX2 (S) and C-MYC (C) genes

– better known as OKSM [209]. However, this method, along with others using DNA integrating vectors, did not come without its drawbacks. The integrative nature of RT vectors may lead to numerous integrations in the genome that can not only disrupt the endogenous expression of genes, but also ultimately interfere with the differentiation of iPSCs. Secondly, transgene integration may also lead to tumor formation, due to the use of C-MYC, a putative oncogene, as a reprogramming factor [208]. To bypass some of

32 these problems, in recent years, several groups have worked to improve the transduction protocols for generating safe iPSCs.

An alternative approach has been the removal of C-MYC as a reprogramming factor, albeit leading to a significantly reduced reprogramming efficiency [223]. Other groups have used a slightly modified combination of reprogramming factors, including OCT4,

SOX2, NANOG (N) and LIN28 (L) (OSLN) [224]. The number of integration sites has been reduced by incorporating all the reprogramming factors into a single vector, or by using a transposon system [225, 226]. Further improvements have entirely eliminated the risk of integration via the transient expression of reprogramming factors, with the use of episomal, adenoviral, sendai (SeV) or lentiviral vectors, small molecules, or synthetic mRNAs [224, 225, 227-232]. Protein transduction and microRNA (miRNA)-based technologies have also been used to induce iPSCs from human adult cells. One of the disadvantages of using non-integrative adenoviral, SeV or lentiviral vectors is the reduced reprogramming efficiency when compared to the use of RTs. Recent studies however, have shown that the synthetic mRNAs and miRNAs have a high reprogramming efficiency, comparable to that of RT [208, 222].

SeV are particularly promising as a vehicle for reprogramming factors and generating patient-derived iPSCs, due to their non-integrative nature. SeVs are RNA based, and therefore allow the expression of transgenes without going through a DNA phase, or integrating into the host genome [233].

33 Considering that iPSCs partially retain the gene expression profile of the donor cells, their origin will have an effect on not only their reprogramming efficiency, but also on the ability of the generated iPSCs to differentiate [234]. It has been shown that iPSCs can differentiate most efficiently back to their cell type of origin, as compared to other cell fates [235]. Previous studies indicate that most iPSCs have been generated using mesoderm and ectoderm-derived cells, with the highest reprogramming efficiencies belonging to umbilical vein endothelial cells and adipose stem cells, among others [236,

237]. Each cell type has its own specific requirements regarding the type and number of reprogramming factors to be used, as well as its own dynamics of generation [238].

Further, it has been shown that cells from younger people have a tendency to reprogram more efficiently than those of older people [239].

It is important to realize that reprogramming is not a flawless process, and that some cell lines are not completely reprogrammed. It is therefore essential that it be shown that the reprogrammed clones are indeed bona fide iPSCs [116]. The criteria a cell line must meet to be deemed an iPSC line is as follows:

(a) ESC-like morphology, forming tight colonies with smooth borders

(b) Positive staining for alkaline phosphatase (AP)

activity

(d) Presence of endogenous pluripotency markers, including but not limited to OCT4,

TRA-1-60, SOX2, NANOG, TRA-1-81, REX1, SSEA-3 and SSEA4

(e) Demethylation of key pluripotency gene promoters

34 (f) Differentiation into endoderm, mesoderm and ectoderm

(g) Karyotype characteristic of the origin cell, with the correct arm morphology and

number of chromosomes

Once the derived cell lines are confirmed as bona fide patient-derived iPSCs, they must be validated as an appropriate model for the genetic disease in question. Validation is first accomplished by confirming the presence of the disease-specific genetic lesion(s).

Secondly, the disease phenotype must be replicated using the iPSCs model – by differentiating the iPSCs into the cell types that are normally affected in the patient.

Therefore, any new observations made following the validation of the model, such as their behavior in response to drug treatment, will be deemed credible [116, 121].

1.4 Objective of Study

1.4.1 Rationale

SDS is characterized by impaired hematopoiesis, with most patients experiencing neutropenia, and to a lesser extent, anemia and thrombocytopenia. Patients are further characterized by an enhanced predisposition to leukemia, which increases with age.

Hematopoietic abnormalities leading to infections, hemorrhage and leukemia are currently the leading causes of death in both younger and older SDS patients. However, little is currently known about the cellular mechanisms responsible for the disruptive hematopoiesis characteristic of SDS patients. Further, the only curative option for the hematological defects associated with SDS is HSCT – although it is associated with poor outcomes. There is therefore a clinical need for the better characterization of the

35 pathogenesis of the SDS-related hematopoietic defects. Although the SDS BM is hypocellular at least in part due to accelerated apoptosis, this may also be due to an impaired ability of the hematopoietic cells to differentiate and grow. Therefore, determining the onset of the hematological phenotype will help shed light on the process of hematopoiesis and the role of SBDS in hematopoietic development [28]. It will also allow for more targeted studies of this disease, the development of strategies to prevent complications, and the generation of more effective treatments, ultimately improving patient care.

Although little is currently known about the onset of the hematopoietic defects in SDS, previous studies in mice indicate that SBDS is important for early embryogenesis, with the disease phenotype arising early during development. Studies using fetal and adult sources of human tissues support this notion, as they showed that SBDS is widely expressed in the fetal and adult liver, spleen, thymus, lung, heart, kidney, skeletal muscle and brain, and adult lymphocytes, lymph nodes, BM, tonsils and peripheral blood leukocytes, albeit to varying degrees [23]. Much like what is observed in humans, the murine SBDS ortholog is expressed in most adult tissues and at all embryonic stages, beginning with the fertilized egg.

Consistent with the hypothesis that SBDS is an essential gene for early development,

SBDS-/- murine embryos experience lethality prior to embryonic day (E) 6.5. Although the SBDS null embryos did not experience developmental defects pre-implantation, by

E6.5, they displayed severe growth and morphological defects. The SBDS -/- embryos

36 were unable to undergo gastrulation, and rather were completely resorbed by E10.5, with only empty deciduae still present [112].

To date, no human patients homozygous for early truncating mutations, with no residual

SBDS expression, have been identified - indicating that SBDS may also play an important role in early human embryonic development [23]. In addition, exocrine pancreatic and hematological manifestations – both hallmarks of SDS – have been documented as early as during the neonatal period in SDS patients [240-242].

It is unclear however, when and where a deficit in the expression levels of SBDS will have a phenotypic effect on the patient. As previously mentioned, SDS patients are characterized by skeletal and hematopoietic abnormalities, but not cardiac, although all three are derived from the mesodermal germ layer. Similarly, although defects have been observed in some organs originating from the endodermal and ectodermal layers (such as the pancreas and brain, respectively), they have not been observed in all organs originating from those respective germ layers. Given that SBDS is ubiquitously expressed, these observations may indicate an either variable need for the protein in different tissues, or variable threshold of protein level for biological disruption in different tissues. To understand SDS disease pathogenesis in the context of blood development, the onset of the hematopoietic defects must be determined. In accordance with the studies mentioned above, the hematopoietic defects may be initiated during early human development; therefore, an adequate model of SDS must be utilized to study early and late stages of blood development.

We chose the iPSC disease model, as not only does it allow for the study of early embryonic development, but also it helps bypass some of the ethical dilemmas faced with using human SDS embryos. Further, due to their plasticity, iPSCs can be differentiated towards various cell fates, including the hematopoietic lineage, to study disease progression in the context of SDS. Lastly, because they can be derived from adult patient cells, and have the ability to self-renew indefinitely, they provide an unlimited source of material to study SDS.

After derivation of the SDS iPSCs from patient samples, we had to first validate them as an appropriate model for the study of this disease. Specifically, we verified that the iPSCs indeed did carry the same SBDS mutations found in the original patient cells, and that they were indeed morphologically and functionally pluripotent in nature. To further validate our model, we assessed the ribosome profile of SDS iPSCs for aberrations, as

SDS patients are characterized by an aberrant ribosome profile. Lastly, we had to determine if the SDS iPSCs were able to differentiate into the cell types of interest

(hematopoietic); and if they were able to do so to a lesser extent than control iPSCs, as observed in SDS patients.

After confirming that the SDS iPSCs indeed did recapitulate the SDS phenotype, we used a differentiation protocol to study the ability of SDS iPSCs to give rise to various cell populations characteristic of specific stages during hematopoietic development. Given that SDS compound heterozygotes with residual SBDS expression survive gestation, the

38 hematopoietic defect may only become clinically significant later during development.

Therefore, we studied time points in both early (primitive wave) and late (definitive wave) stages of hematopoiesis, to determine if and where there may be a defect in hematopoietic development in SDS iPSCs. Given that SDS iPSCs recapitulate the SDS phenotype, and that hematopoietic abnormalities are the leading cause of death in SDS patients, this warranted the further study of hematopoietic differentiation in SDS iPSCs.

1.4.2 Hypothesis

Although it is well documented that SDS patients have an impaired hematopoietic phenotype, the onset of this blood development abnormality remains to be elucidated. We hypothesized that the hematopoietic manifestations associated with SDS are the result of impaired development of the hematopoietic system, rather than an earlier defect in the stepwise differentiation of pluripotent stem cells towards the mesodermal germ layer.

Since the known SDS phenotype involves definitive hematopoiesis, we hypothesize that the definitive wave of hematopoiesis is markedly impaired.

1.4.3 Specific Aims

1. To develop and validate an in-vitro SDS iPSC disease model that recapitulates the

human hematopoietic SDS disease.

2. To identify the abnormal developmental step(s) leading to reduced blood cell

formation in SDS.

39 Chapter II

Materials and Methods

This study was approved by The Hospital for Sick Children Research Ethics Board.

Samples from SDS patients and controls were obtained after receiving written, informed consent.

2.1 Cell Culture

2.1.1 Cell lines

2.1.1.1 Mouse embryonic fibroblasts

Mouse embryonic fibroblasts (MEFs) were generated in Dr. Gordon Keller’s Laboratory

(McEwan Centre for Regenerative Medicine, Toronto, ON). They were maintained in

Iscove’s Modified Dulbecco medium (IMDM), (Cellgro) supplemented with 1%

Penicillin/Streptomycin (Cellgro) and 20% Fetal Calf Serum (FCS) (Atlas, Fort Collins,

CO).

2.1.1.2 OP9-DL1

OP9 cells that had been transduced to express Delta-like 1 (OP9-DL1) were generously obtained from the lab of Dr. Gordon Keller (McEwan Centre for Regenerative Medicine,

Toronto, ON) and were maintained in α-Minimal Essential Medium (α-MEM) supplemented with 1% L-Glutamine (Cellgro) and 20% human T-cell serum (GemCell).

40 2.1.1.3 hiPSCs

Bone marrow fibroblast cultures were established by plating marrow cells in long-term culture medium (Myelocult, Stem Cell Technologies) and depletion of hematopoietic cells by weekly medium change and three to four passages. Sendai virus-mediated reprogramming and generation of hiPSCs from control individuals, N1-I and N1-K and

SDS patients, P1-A and P1-D patient were performed from bone marrow fibroblasts at the Centre of Commercialization for Regenerative Medicine (CCRM), (Toronto, ON).

Retrovirus-mediated BS-1 and BS-3 control hiPSCs were obtained from the lab of Dr.

Bill Stanford (Ottawa Hospital Research Institute, Ottawa, ON). Retrovirus-mediated,

SDS-iPSC-1 and SDS-iPSC-2 were obtained from the lab of Dr. George Daley (Boston

Children’s Hospital, Boston, MA).

The results obtained from using these cell lines were pooled, categorized and reported as

RT-Control, RT-SDS, SeV-Control and SeV-SDS.

Prior to passaging the hiPSCs, 6-well tissue culture plates were gelatinized using 0.1% gelatin (Sigma, St Louis, MO) and irradiated MEFs were seeded at 70-80% confluency. hiPSCs were maintained on irradiated MEFs in hiPSC medium consisting of Dulbecco’s

Modified Eagle Medium (DMEM): Nutrient Mixture F-12 (DMEM/F12), 1%

Penicillin/Streptomycin, 1% Glutamine, (Cellgro), 20% KnockoutTM Serum

Replacement, 1% Non-essential amino acids (NEAA), (Gibco), 10-4 M β- mercaptoethanol (Sigma, St Louis, MO) and 15ng/mL hbFGF (R&D Systems,

Minneapolis, MN) at 37°C in a humidified 5 % CO2 atmosphere.

Confluent hiPSCs were passaged to new MEFs as clump suspensions, following dissociation of the hiPSC colonies using collagenase B (1 mg/mL; Roche, Indianapolis,

IN) for 5 minutes, and trypsin-EDTA (0.05%) for approximately 2 minutes (Mediatech).

2.1.2 Feeder depletion

Prior to hematopoietic differentiation (day -1), hiPSCs were feeder-depleted by culturing on a thin layer of matrigel (BD Biosciences, Bedford, MA, USA) for 24-48 hours in hiPSC medium.

2.1.3 Establishing embryoid bodies for hematopoietic differentiation

To generate EBs at day 0 (T0), hiPSC colonies on matrigel were dissociated into small clumps of 10-20 cells by treating with collagenase B (1 mg/mL; Roche, Indianapolis, IN) for 20 minutes, followed by trypLETMExpress (1x), (Invitrogen) for approximately 2 minutes. The cells were gently scraped and clumps were washed and distributed into 6- well low cluster plates at a concentration of 3 wells equivalent to 2 wells, (Corning,

Corning, NY) in 2 mL of aggregation medium per well. Aggregation medium (used to help promote the generation and survival of EBs) consisted of StemPro-34 (Invitrogen) supplemented with hBMP-4 (10ng/mL), (R&D Systems, Minneapolis, MN), monothioglycerol (MTG, 4 x 10-4 M), ascorbic acid (50µg/mL), (Sigma), L-glutamine

(10ng/mL), (Cellgro), transferrin (150µg/mL), (Roche) and penicillin/streptomycin

(10ng/mL), (Cellgro). At day 1 of differentiation, 2 mL of aggregation medium with bFGF (5ng/mL),(R&D Systems, Minneapolis, MN) were added to each well of EBs.

42 2.1.3.1 Primitive hematopoiesis

At day 2 of differentiation, media and aggregates were collected into tubes and allowed to settle for 30 minutes at 37°C in a 5% CO2/5% O2/90% NO2 environment. The aggregation medium was aspirated and aggregates were resuspended in Induction

Medium 1 (used to induce mesoderm formation), containing StemPro-34 (Invitrogen), hBMP-4 (10ng/mL), bFGF (5ng/mL), Act.A (0.3ng/mL), (R&D Systems, Minneapolis,

MN), monothioglycerol (MTG, 4 x 10-4 M), ascorbic acid (50µg/mL) (Sigma), L- glutamine (10ng/mL), (Cellgro), transferrin (150µg/mL), (Roche), and penicillin/streptomycin (10ng/mL) (Cellgro). At day 4, media and aggregrates were collected into tubes and washed twice following centrifugation at 57 G for 5 minutes.

Aggregates were resuspended in Induction Medium 2 (used to prime EBs for blood cell formation), consisting of StemPro-34 (Invitrogen), bFGF (5ng/mL), VEGF (10ng/mL), hIL-6 (10ng/mL), hIL-11 (5ng/mL) (R&D Systems, Minneapolis, MN), monothioglycerol (MTG, 4 x 10-4 M), ascorbic acid (50µg/mL), (Sigma), L-glutamine

(10ng/mL), (Cellgro), transferrin (150µg/mL), (Roche), and penicillin/streptomycin

(10ng/mL), (Cellgro) . At day 6, 2 mL of Induction 3 medium (used to further promote blood cell formation) consisting of StemPro-34 (Invitrogen), bFGF (5ng/mL), VEGF

(10ng/mL), hIL-6 (10ng/mL), hIL-11 (5ng/mL), hSCF (50ng/mL) (R&D Systems,

Minneapolis, MN), hEPO (2 units/mL) (Janssen-Ortho Inc.), monothioglycerol (MTG, 4 x 10-4 M), ascorbic acid (50µg/mL), (Sigma), L-glutamine (10ng/mL), (Cellgro), transferrin (150µg/mL), (Roche), and penicillin/streptomycin (10ng/mL), (Cellgro) were added to each well. At day 8 of differentiation, media and aggregates were collected and washed as done at day 4. Aggregates were resuspended in hematopoietic expansion

43 medium (to promote the general expansion of hematopoietic progenitors) containing

StemPro-34 (Invitrogen), hIL-6 (10ng/mL), hIL-11 (5ng/mL), hSCF (100ng/mL), hIGF-1

(25ng/mL), hTPO (50ng/mL), hIL-3 (50ng/mL), hFLT3L (20ng/mL), (R&D Systems,

Minneapolis, MN), hEPO (2 units/mL) (Janssen-Ortho Inc.), monothioglycerol (MTG, 4 x 10-4 M), (Sigma), ascorbic acid (25µg/mL), (Roche), L-glutamine (10ng/mL), (Cellgro) transferrin (150µg/mL), (Roche) and penicillin/streptomycin (10ng/mL). Cultures were maintained at 37°C in a 5% CO2/5% O2/90% NO2 environment for the first 8 days and then transferred to a 5% CO2/air environment for anywhere from 1 to 6 more days (Fig.

6).

Fig 6. Primitive hematopoiesis differentiation scheme for human pluripotent stem cells

The various oxygen conditions, culture modalities and development stages corresponding to each day of differentiation are shown. EBs are generated in the presence of BMP-4 for the first 24 hours, and then cultured together with BMP-4 and bFGF for an additional 24 hours. At day 2, Act. A is added, and the EBs are cultured in its presence, together with BMP-4 and bFGF. 2 days later, hematopoietic cytokines are added to promote the generation of more mature blood cells.

44 2.1.3.2 Definitive hematopoiesis

At day 1.75 of differentiation, media and aggregates were collected into tubes and allowed to settle for 30 minutes at 37°C in a 5% CO2/5% O2/90% NO2 environment. The

Aggregation Medium was aspirated and aggregates were resuspended in Induction

Medium 1 (used to induce mesoderm formation), containing StemPro-34 (Invitrogen), hBMP-4 (10ng/mL), bFGF (5ng/mL), SB-431542 (0.3ng/mL) (Sigma) (which is an inhibitor of the activin receptor-like kinase (ALK) receptors, ALK5, ALK4 and ALK7) , monothioglycerol (MTG, 4 x 10-4 M), ascorbic acid (50µg/mL) (Sigma), L-glutamine

(10ng/mL), (Cellgro), transferrin (150µg/mL), (Roche), and penicillin/streptomycin

(10ng/mL) (Cellgro). At day 4, media and aggregrates were collected into tubes and washed twice following centrifugation at 57 G for 5 minutes. Aggregates were resuspended in Induction Medium 2 (used to prime EBs for blood cell formation), consisting of StemPro-34 (Invitrogen), bFGF (5ng/mL), VEGF (10ng/mL), hIL-6

(10ng/mL), hIL-11 (5ng/mL) (R&D Systems, Minneapolis, MN), monothioglycerol

(MTG, 4 x 10-4 M), ascorbic acid (50µg/mL), (Sigma), L-glutamine (10ng/mL),

(Cellgro), transferrin (150µg/mL), (Roche), and penicillin/streptomycin (10ng/mL),

(Cellgro). At day 6, 2 mL of Induction Medium 3 (used to further promote blood cell formation) consisting of StemPro-34 (Invitrogen), bFGF (5ng/mL), VEGF (10ng/mL), hIL-6 (10ng/mL), hIL-11 (5ng/mL), hSCF (50ng/mL) (R&D Systems, Minneapolis,

MN), hEPO (2 units/mL) (Janssen-Ortho Inc.), monothioglycerol (MTG, 4 x 10-4 M), ascorbic acid (50µg/mL), (Sigma), L-glutamine (10ng/mL), (Cellgro), transferrin

(150µg/mL), (Roche), and penicillin/streptomycin (10ng/mL), (Cellgro) were added to each well. At day 8 of differentiation, media and aggregates were collected and washed

45 as done at day 4. Aggregates were resuspended in hematopoietic expansion medium (to promote the general expansion of hematopoietic progenitors) containing StemPro-34

(Invitrogen), hIL-6 (10ng/mL), hIL-11 (5ng/mL), hSCF (100ng/mL), hIGF-1 (25ng/mL), hTPO (50ng/mL), hIL-3 (50ng/mL), hFLT3L (20ng/mL), (R&D Systems, Minneapolis,

MN), hEPO (2 units/mL) (Janssen-Ortho Inc.), monothioglycerol (MTG, 4 x 10-4 M),

(Sigma), ascorbic acid (25µg/mL), (Roche), L-glutamine (10ng/mL), (Cellgro) transferrin (150µg/mL), (Roche) and penicillin/streptomycin (10ng/mL). Cultures were maintained at 37°C in a 5% CO2/5% O2/90% NO2 environment for the first 8 days and then transferred to a 5% CO2/air environment for anywhere from 1 to 6 more days.

At day 9 of differentiation, EBs were harvested and treated with trypsin-EDTA (0.05%)

(Mediatech) for 5 minutes. Medium with 50% fetal calf serum (FCS)(Cellgro) was added to the trypsinizing EBs, which were then dissociated by passaging through a 20-gauge needle 6 times. The cells were treated for 1 hour with Collagenase Type 1 (0.2%)

(Sigma) at 37°C, and a single cell suspension was generated by passaging through a 20- gauge needle 6 times. Cells were passaged through a blue-cap filter (Corning) to ensure a single-cell suspension, counted using a hemocytometer, centrifuged at 276 G, resuspended in an antibody mix composed of CD34-APC (1:100) and CD43-PE (5:100)

(both from BD Biosciences, San Diego, CA) and left on ice for 30 minutes in the dark.

After a 5 minute centrifugation at 228 G, cells were resuspended in 0.5 mL of OP9-DL1 medium. Cells were sorted with the FACSAriaII (BD) cell sorter for the CD34+/CD43- fraction. After sorting, the CD34+/CD43- fraction was counted using a hemocytometer, and co-cultured in a 24-well plate of irradiated OP9-DL1s at 20 000 cells/well in OP9

46 medium (hTPO (30ng/mL), hSCF (50ng/mL), Flt3l (10ng/mL), hIL11 (5ng/mL), (R&D

Systems, Minneapolis, MN) and hBMP4 (10ng/mL), (Sigma)). At day 1 post-sort, 500

µL of OP9 medium supplemented with hTPO (30ng/mL), hSCF (50ng/mL), Flt3l

(10ng/mL), hIL11 (5ng/mL), (R&D Systems, Minneapolis, MN) and hBMP-4

(10ng/mL), (Sigma) were added. At day 3 post-sort, all medium was aspirated and 1 mL of OP9 medium supplemented with VEGF (5ng/mL), hTPO (30ng/mL), hSCF

(50ng/mL), Flt3l (10ng/mL), hIL11 (5ng/mL), (R&D Systems, Minneapolis, MN) and hBMP-4 (10ng/mL), (Sigma) was added. The media was subsequently changed every two days for the remainder of the culture period (Fig. 7).

Fig 7. Definitive hematopoiesis differentiation scheme for human pluripotent stem cells

The various oxygen conditions, culture modalities and development stages corresponding to each day of differentiation are shown. EBs are generated in the presence of BMP-4 for the first 24 hours, and then cultured together with BMP-4 and bFGF for an additional 24 hours. At day 1.75, SB-431542 is added, and the EBs are cultured in its presence, together with BMP-4 and bFGF. 2 days later, hematopoietic cytokines are added, and at day 9 of differentiation, the CD34+ CD43- population is sorted and further cultured in the presence of various hematopoietic cytokines.

47 2.1.4 Preparation of single-cell suspension from embryoid bodies

If EBs were harvested between day 0 and 8 of differentiation, they were dissociated to a single cell suspension by treating with trypsin-EDTA (0.05%) (Mediatech) for 5 minutes, and passaging through a 20-gauge needle 6 times. If the EBs were harvested post day 8 of differentiation, cells were also treated with Collagenase Type 1 (0.2%) (Sigma) for 1 hour at 37°C, and a single cell suspension was generated by passaging through a 20- gauge needle 6 times. Cells were used for further assays.

2.1.5 Preparation of single-cell suspensions from the OP9-DL1 co-

cultures

On the day of the assay, supernatant was collected and passed through a blue-cap filter

(Corning). Cultures were trypsinized using trypsin-EDTA (0.05%) (Mediatech) for approximately 3 minutes. Wash medium was added, and cells were mixed vigorously and passed through the blue-cap filter. Cells from both, supernatant and adherent cells were combined and centrifuged at 447 G for 5 minutes to be used for further assays.

2.1.6 Clonogenic assay

For evaluation of colony-forming potential, harvested cells were plated at 40 000 cells/35 mm gridded dish (Corning) in IMDM with methylcellulose (55%), hSCF (100 ng/mL), hEPO (4 units/mL), hIL-6 (10ng/mL), hIL-3 (50 ng/mL), hTPO (50 ng/mL), hIL-11 (5 ng/mL), hIGF-1 (50 ng/mL), hVEGF (10 ng/mL), hGM-CSF (1 ng/mL), (R&D Systems,

Minneapolis, MN), L-glutamine (2 mM), (Cellgro), transferrin (150µg/mL), (Roche), plasma-derived serum (15%, PDS; Animal Technologies, Tyler, TX) and protein-free

48 hybridoma medium (5%, PFHM-II; Invitrogen). Hematopoietic colonies were visualized and scored using a Leica DM IRB microscope (Leica Microsystems, Bannockburn, IL) according to size (large if colony > 0.5mm x 0.5mm, small if colony ≤ 0.5mm x 0.5mm, cluster if colony is between 2- 20 cells). Images were captured with a Magnafire digital camera (Optronics, Goleta, CA) using a 20x objective lens.

2.1.7 Morphological analysis by Wright-Giemsa staining

Hematopoietic colonies were picked from the clonogenic assays and spread on glass slides using the cytospin centrifuge for 3 minutes at 500 G. Morphological features were evaluated via Wright-Giemsa staining. Cells were visualized and recorded with a Leica

DMLB microscope (Leica Microsystems, Bannockburn, IL). Images were taken with a

SPOT InsightTM 2 Megapixel Color Mosaic digital camera (Diagnostic Instruments,

Sterling Heights, MI) using a 100x objective.

2.2 DNA isolation

Following a 48-hour feeder depletion period, control and SDS iPSCs were treated with collagenase B (1 mg/mL; Roche, Indianapolis, IN) for 5 minutes, and trypsin-EDTA

(0.05%) for approximately 2 minutes (Mediatech). Cells were harvested and centrifuged for 5 minutes at 228 G at 4°C. Genomic DNA was isolated using the Wizard Genomic

DNA Purification Kit (Promega) according to the manufacturer’s instructions.

49 2.3 PCR amplification and sequencing

Thermo Scientific Maxima Hot Start Taq DNA Polymerase (Thermo Fisher Scientific) was used to amplify the SBDS DNA sequences. The primer pairs used were as follows:

SBDS gene exon 2 forward primer: GGG ATT TGT TGT GTC TTG, reverse primer:

CTT TCC TCC AGA AAA ACA GC. SBDS gene exon 4 forward primer: GCC TTC

ACT TTC TTC ATA GT, reverse primer: GAA AAT ATC TGA CGT TTA CAA CA.

Amplified DNA samples underwent Sanger sequencing at The Center for Applied

Genomics (TCAG, Sickkids Hospital, Toronto, ON). Sequences were viewed and analyzed using Finch TV (Geospiza Inc., Seatlle, WA).

2.4 Flow cytometry and cell sorting

2.4.1 Antibodies

The antibodies used for these studies were as follows: mouse, anti-human CD34-PE-CY7

(clone 4H11, eBiosciences San Diego, CA),KDR-APC (R&D Systems), CD31-FITC

(clone WM59), CD33-FITC (clone HIM 3-4), CD41a-APC (clone HIP8), CD43-PE

(clone 1G10),CD45-APC-eFluor750 (clone 2D1), CD56-APC, CD56-PE, CD117-PE,

CD144-PE, SSEA-4-PE, CD235a-FITC, CD11b-FITC, and mouse, IgG1–PE, IgG1–APC,

IgG1–FITC, IgG1–PE-CY7, IgG1–Pacific Blue (BD Biosciences, San Diego, CA).

2.4.2 Fluorescence-activated cell Sorting and flow cytometry

After harvest from either the EB or OP9-DL1 co-cultures, aliquots of 5 x 104 cells per staining combination were incubated in staining medium containing IMDM (Cellgro),

10% Fetal Calf Serum (FCS) (Atlas, Fort Collins, CO) and the antibodies of interest for

50 30 minutes at 4°C, in the dark. Stained cells were analyzed and/or sorted using an LSRII

(BD Biosciences, San Diego, CA) flow cytometer at various time points. Data analysis was performed using FlowJo software (Treestar, San Carlos, CA). Gating was done with matched isotype control monoclonal antibodies.

2.6 Sucrose density gradient ultracentrifugation

Following a 48-hour feeder depletion period, control and patient iPSCs were treated with collagenase B (1 mg/mL; Roche, Indianapolis, IN) for 5 minutes, and trypsin-EDTA

(0.05%) for approximately 2 minutes (Mediatech). Cells were harvested, centrifuged at

228 G for 5 minutes and resuspended in 10 mL of hiPSC medium. They were then incubated at 37°C for 10 minutes after treating with 100µg/mL cycloheximide (Sigma).

Cells were centrifuged and washed twice in ice cold PBS supplemented with 100µg/mL cycloheximide (Sigma). The cell pellet was then resuspended in ice-cold TMK100 lysis buffer containing 1 M Tris, 100 mMKCl, 5 mM MgCl2, 10 mM HEPES (pH 7.4), 100

µg/mL cycloheximide, 0.5% (v/v) Triton x-100 and 250 Units/mL RNAseOUT, and passaged through a 27.5 gauge needle 3 times. Following centrifugation at 32842 G for 5 minutes, nuclei were removed and total RNA was measured at A260 nm. Lysates were layered on a 5% to 45% (w/v) sucrose gradient in ribosome profile sucrose gradient buffer, containing 100 mMKCl, 5 mM MgCl2 and 10 mM HEPES (pH 7.4). Lysates were centrifuged in a SW41 rotor (Beckman Coulter, Fullteron, CA) for 2.5 hours at 273 865

G. All sucrose gradients were fractionated using the Brandel gradient fractionator system

(Brandel, Gaithersberg, MD), and absorbance was monitored at 254 nm with an ISCO

UA-6 flow cell (Teledyne ISCO, Lincoln, NE).

51 2.7 Statistics

Means and standard errors of the mean (SEM) were used to describe the results concerning ribosomal subunit levels, hematopoietic colony formation and induction of the various populations characteristic of hematopoiesis during differentiation of hiPSCs.

Student’s t-test was used to determine the statistical significance of differences between two means. P value of <0.05 was considered statistically significant.

52 Chapter III

Results

3.1 SDS and control iPCSs can be generated from SDS and control

fibroblasts

3.1.1 Generation efficiency

To reduce the risk of gene disruption and the subsequent consequences of using

integrative viruses for reprogramming, we generated hiPSC lines using non-integrative

SeV viruses through the CCRM (Toronto, Canada). N1-I and N1-K were generated from

the BM fibroblasts of a 41 year old, healthy male, using SeVs containing the embryonic

stem cell transcription factors OKSM. The P1-A and P1-D lines were generated from the

BM fibroblasts of a 17 year old, male SDS patient with the IV2+2T>C and 183-

184TA>CT compound heterozygous SBDS mutations, using SeVs containing OKSM.

The reprogramming efficiency of SDS iPSCs P1-A and P1-D, was 0.002%. In

comparison, the reprogramming efficiency of the control iPSCs was 0.04% (Table 1).

Table 1. Human induced pluripotent stem cell generation efficiency

The molecular defect, donor cell type, age and sex of donor, along with the type of viral vector used to generate the iPSCs are shown. The control BS-1 and BS-3 lines were a generous gift from Dr. Bill Stanford (Ottawa Hospital Research Institute), whereas the patient SDS-iPSC-1 and SDS-iPSC-2 lines were a generous gift from Dr.George Daley (Harvard Stem Cell Institute). Information regarding the generation efficiency of the RV-derived iPSC lines is not available. The N1-I and N1-K control lines were generated with a reprogramming efficiency of 0.04%, whereas the P1-A and P1-D cell lines were generated with a 0.002% reprogramming efficiency.

54 3.1.2 SDS iPSCs carry the original patient mutation

Prior to studying the hematopoietic phenotype in SDS iPSCs, it was imperative to determine if they carry the disease-specific genotype characteristic of the parental somatic cells. DNA sequencing was therefore used to assess for the presence of mutated alleles in the generated SDS iPSCs.

Similar to what was observed in the original patient sample, the SDS-iPSC-1 and SDS- iPSC-2 cell lines were compound heterozygotes for the IV2+2T>C SBDS mutation in the intron 2 splice donor site, and the IVS3-1G>A SBDS point mutation. Similarly, sequencing of the P1-A and P1-D iPSCs lines revealed that they are compound heterozygotes, carrying the IV2+2T>C and 183-184TA>CT SBDS mutations (Fig. 8).

Fig 8. Sequencing of SBDS in SDS induced pluripotent stem cells

Mutated alleles identical to the original donor cells were verified by DNA sequence. SDS-iPSC-1 and SDS- iPSC-2 are both compound heterozygotes with point mutations at IV2+2T>C and IV3-1G>A. P1-A and P1- D are also compound heterozygote with mutations at IV2+2T>C and 183-184TA>CT.

55 3.1.3 Pluripotency markers

Various markers are expressed on the surface of cells at different stages of differentiation.

Specifically, the undifferentiated, pluripotent stem cell state is characterized by the expression of proteins including OCT4, SSEA-4, TRA-1-81, NANOG and TRA-1-60, which are involved in stemness properties of undifferentiated pluripotent stem cells, such as self renewal.

Following reprogramming, only hiPSC colonies morphologically similar to hESCs were selected for expansion and further characterization. The CCRM used immunofluorescence to check for the expression of markers characteristic of the undifferentiated, stem cell state, and to consequently confirm that the generated hiPSCs were indeed pluripotent in nature. The N1-I and N1-K control cells marked by nuclear

4',6-diamidino-2-phenylindole (DAPI) positivity were also both positive for the OCT4,

SSEA-4, TRA-1-81, NANOG and TRA-1-60 markers, indicating that they were indeed in a pluripotent and undifferentiated state (Fig. 9).

The P1-A and P1-D SDS cells positive for DAPI also expressed OCT4, SSEA-4, TRA-1-

81, NANOG and TRA-1-60, indicating that they too were in an undifferentiated, stem cell-like state (Fig. 9). The RT-mediated reprogrammed SDS-iPSC-1 and SDS-iPSC-2 lines have been previously characterized by Park & colleagues. [121].

Fig 9. Human induced pluripotent stem cell pluripotency markers

As shown by immunofluorescence, the N1-I, N1-K, P1-A and P1-D iPSC lines are positive for the presence of pluripotency markers: OCT4, SSEA4, TRA-1-81, Nanog and TRA-1-60. The first row of each panel depicts the single-stain of the marker in focus. DAPI is shown in the middle row of each panel, and it is representative of the entire cell content in that particular field of view. The bottom row of each panel is a merged image of the DAPI and pluripotency marker stainings. The SDS-iPSC-1 and SDS-iPSC-2 lines have been previously characterized by Park & colleagues, and no data is available regarding BS-1 and BS- 3. (Scale bars = 200 µm)

57 3.1.4 Germ layer differentiation

To further confirm their pluripotency and ability to differentiate towards all three germ layers, the SeV-derived hiPSC lines were differentiated as EBs, and assessed for their expression of lineage markers specific to each germ layer by the CCRM. Reverse- transcription quantitative real time PCR (RT-qPCR) was used to determine the expression levels of the alpha-fetoprotein (AFP), heart and neural crest derivatives- expressed protein 1 (HAND1) and neurogenic differentiation 1 (NEUROD1) genes, which are characteristic of the endodermal, mesodermal, and ectodermal germ layers, respectively.

Relative to their undifferentiated states, all 4 hiPSC lines showed upregulated expression of the three different lineage markers. N1-I showed a 91.47, 70.97, and 8.384 fold induction of markers characteristic of endoderm, mesoderm, and ectoderm, respectively.

Comparatively, N1-K showed lower fold inductions, of 2.013 for the endodermal gene,

34.19 for the mesoderm gene, and 0.132 for the ectodermal gene (Table 2). P1-A showed a 157.6, 5084, and 2020 fold induction of markers characteristic of endoderm, mesoderm, and ectoderm, respectively. Similarly, P1-D showcased fold inductions of 364.8, 1269 and 2.724 for the endodermal, mesodermal, and ectodermal, AFP, HAND1, and

NEUROD1 genes, respectively (Table 2).

The variation in fold induction between the different cell lines for the three germ layers may be a result of variation among different hiPSC clones, due to epigenetic changes that may occur during the reprogramming process. Differences in genetic background, as seen

58 between the control and SDS cell lines, may also affect gene-expression patterns and consequently the functional ability of hiPSCs to differentiate towards all three germ layers. Further, the heterogeneity that exists among fibroblasts may also lead to variations in the genetic make-up of the resulting hiPSC colonies, although they may all be clones derived from the same sample – ultimately resulting in hiPSC lines with varying abilities to differentiate towards the three germ layers [243, 244].

Table 2. Human induced pluripotent stem cell germ layer differentiation

Differentiation of the SeV-generated N551-I, N551-K, P357-A and P357-D iPSC lines, followed by RT- qPCR shows upregulated expression of three different lineage markers specific to the three germ layers (AFP for endoderm, HAND1 for mesoderm and NEUROD1 for ectoderm, when compared to their undifferentiated states). The SDS-iPSC-1 and SDS-iPSC-2 lines have been previously characterized by Park& colleagues, and no data is available regarding BS-1 and BS-3. No data available regarding the associated SEM values. n=3.

59 3.2 SDS iPSCs are characterized by an abnormal ribosome profile

3.2.1 Ribosomal profiles of SDS iPSCs show reduced levels of 80S

ribosome subunit

Given that SDS is characterized by an aberrant ribosome profile, sucrose density gradients were used to determine if the SDS iPSCs truly do recapitulate the SDS phenotype. Equal amounts of RNA obtained from both, control and SDS, RT and SeV- derived iPSCs were layered on 5-45% sucrose gradients (Fig. 10A). Following ultracentrifugation, the levels of 40S, 60S, 80S and polysomes were determined by quantifying the area under the curve. Using this approach, we observed a marked reduction in the 80S peak in the undifferentiated, RT and SeV-derived SDS iPSCs when compared to the healthy controls (Fig. 10B-C). Additionally, there was also a significant reduction in the 40S/60S subunit ratio in the SeV SDS iPSCs, with the levels of 40S consistently reduced in all experimental trials (10B, D). However, there were no marked differences in the levels of 60S and polysomes between SeV control and patient iPSCs

(Fig. 10B).

This observation is consistent with what has been previously documented in SDS patient cells, as Burwick & colleagues, have shown that SBDS promotes the association between the 40S and 60S subunits, to form the mature 80S monomer [24]. These results suggest that loss of SBDS may lead to a global defect in ribosome biogenesis, due to the reduction in the generation of the mature 80S subunit. Further, the abnormal ribosome profile in SDS iPSCs supports the notion that these cells recapitulate the SDS phenotype, and may therefore be used for the study of blood development in a diseased state.

Fig 10. Abnormal ribosome profile in SDS induced pluripotent stem cells

A. Layout of the 5-45% sucrose density gradient. Following centrifugation of the sample, low density 40S subunits are found near the top, followed by the 60S subunits, 80S subunits, and lastly, the polysomes, which due to their high density, are at the bottom. B. Equal amounts of RNA isolated from undifferentiated control and SDS RT and SeV-generated iPSCs were layered on top of a 5-45% sucrose gradient, and ultracentrifuged. A representative diagram of the ribosome profiles of the control and SDS RT and SeV is shown below, and the levels of 40S, 60S, 80S subunits and polysomes are indicated. n=2 for RT control and SDS iPSCs, n=3 for SeV control and SDS iPSCs. C. The levels of 80S subunits were determined by quantifying the area under the curve, and reported relative to the control values. As shown, the SeV SDS iPSCs showed a marked reduction in the size of the 80S subunit peak as determined by the T-test (*p<0.05). The error bars represent the SEM. n=3. D. The 40S/60S subunit ratio was determined by quantifying the area under the curve, and reported relative to the control values. As shown, the SeV SDS iPSCs showed a significant reduction (p=0.0004) in the 40S/60S subunit ratio as determined by the T-test (*p<0.05). The error bars represent the SEM. n=3.

61 3.3 SDS iPSCs can form definitive hematopoietic progenitors

To study the hematopoietic defect in SDS, we first aimed to determine if the control and

SDS iPSCs could generate hematopoietic progenitors: granulocyte, erythrocyte, monocyte, megakaryocyte-colony forming units (CFU-GEMMs), granulocyte, monocyte- colony forming units (CFU-GMs) and BFU-Es. The control and SDS iPSCs were assessed for hematopoietic colony-forming ability by using an EB-based differentiation protocol for definitive hematopoiesis. Briefly, the cultures were treated with SB-431542, and after 9 days of differentiation, the EBs were dissociated and the CD34+/CD43- HE population was isolated. CD34+/CD43- cells were plated on OP9-DL1s, to undergo EHT in the presence of VEGF, and 7 days later, the cells were plated into clonogenic assays, to assess for colony-forming potential.

SeV-mediated control and SDS iPSCs both gave rise to CFU-GEMMs, CFU-GMs and

BFU-Es (Fig. 11). The RT-generated control and SDS iPSCs both gave rise to CFU-GMs and BFU-Es, but no CFU-GEMMs (Fig. 11). The absence of CFU-GEMMs in the RT- derived cultures may be as a result of the previously-reported clonal variation in differentiation potential of iPSCs [245]. Given that much like what was observed by

Tulpule & colleagues, the RT-SDS iPSCs did not generate the full range of hematopoietic progenitors, the RT-derived iPSC lines were excluded from further, more in depth analysis. These results show that the control and SDS iPSCs can differentiate towards hematopoietic cells, and can therefore be used to study hematopoietic development in the context of SDS.

Fig 11. Hematopoietic progenitor colony formation

Photographs showing morphology of colonies characteristic of definitive hematopoiesis. Colonies were generated using clonogenic assays, from SB-431542-treated EBs sorted for the CD34+ CD43- population which was co-cultured on OP9-DL1s for 7 days. The RT Control and SDS iPSCs generated CFU-GMs and BFU-Es, but no CFU-GEMMs. The SeV Control and SDS iPSCs generated CFU-GEMMs, CFU-GMs, and BFU-Es. Original magnification x 20.

63 3.4 SDS iPSCs can form terminally differentiated blood cells

characteristic of definitive hematopoiesis

3.4.1 Terminally differentiated blood cells have a morphology

characteristic of macrophages, erythrocytes and granulocytes

To confirm the differentiation functionality of the hematopoietic progenitors generated in methylcellulose, the CFUs were collected and the morphology of the cells was assessed using Giemsa-Wright Staining. The results indicate that both the SeV and RT-generated control and SDS iPSCs have the ability to form cells with a morphology characteristic of granulocytes, macrophages, and erythroid cells (Fig. 12). Therefore, these cells provided us an adequate model for our next in-vitro studies.

Fig 12. Terminally differentiated blood cells

Photographs of Giemsa-stains, showing morphology of the cells isolated from colonies formed in the clonogenic assays. Colonies were generated from SB-431542-treated EBs sorted for the CD34+ CD43- population, which was co-cultured on OP9-DL1s for 7 days. The RT and SeV Control and SDS iPSCs generated granulocytes, macrophages and erythroid cells. Original magnification x 100.

64 3.5 SDS iPSCs have a reduced capacity to form hematopoietic

progenitors characteristic of definitive hematopoiesis

In validating the SDS iPSCs as a model for SDS, there was a need to verify that much like what is seen in patients, SDS iPSCs have a reduced ability to generate hematopoietic cells. The SeV-iPSCs were primed for definitive hematopoiesis as previously described, and CFU-GEMM, CFU-GM and BFU-E colonies were enumerated according to size

(large if colony > 0.5mm x 0.5mm, small if colony ≤ 0.5mm x 0.5mm, cluster if colony is between 2- 20 cells ) after 11 days on methylcellulose. The SDS SeV-iPSCs generated significantly fewer CFU-GEMM, CFU-GM and BFU-E, large and small colonies, relative to the Control SeV–iPSCs. However, both the control and SDS iPSCs generated similar numbers of CFU-GM and BFU-E clusters (Fig. 13). These results mirror the SDS hematopoietic phenotype that has been previously characterized, and further help verify that SDS SeV-iPSCs recapitulate the SDS hematopoietic phenotype.

Fig 13. Reduced formation of hematopoietic progenitors characteristic of definitive hematopoiesis in SDS induced pluripotent stem cells

Colonies generated using clonogenic assays, from SB-431542-treated EBs sorted for the CD34+ CD43- population, which was co-cultured on OP9-DL1s for 7 days were enumerated. Colonies were counted in triplicate for each data point. Hematopoietic colonies were scored according to size (large if colony > 0.5mm x 0.5mm, small if colony ≤ 0.5mm x 0.5mm, cluster if colony is between 2- 20 cells ). SDS SeV iPSC colony numbers are reported relative to the number of colonies generated by the control SeV iPSCs. As shown, SDS SeV iPSCs generate significantly fewer large and small CFU-GEMM, CFU-GM, and BFU-E colonies when compared to the controls, as determined by the T-test (*p<0.05). However, the control and SDS SeV iPSCs generate comparable numbers of CFU-GM and BFU-E clusters, as determined by the T-test (*p>0.05). The error bars represent the SEM. n=4 for N1-I, n=4 for N1-K, n=4 for P1-A, n=4 for P1-D.

66 3.6 SDS iPSCs have a developmental defect in definitive

hematopoiesis

3.6.1 Mesoderm induction is not affected in SDS iPSCs

Given that SDS iPSCs exhibit a reduction in their ability to generate hematopoietic colonies within the definitive wave of hematopoiesis, our next step was to investigate the various developmental stages, and determine where this defect may be initiated. We first looked at the potential of iPSCs to generate mesoderm, a CD34-CD56+ or CD34-KDR+C- kit- population using flow cytometry. Briefly, the iPSCs were preferentially differentiated towards definitive hematopoiesis, and the EBs were dissociated at T4 and T6. The single cell suspension was then assessed for the presence of the CD34, CD56, KDR and C-kit surface markers (Fig. 14A, Fig. 15A).

Both the control and SDS SeV iPSCs generated a comparable percentage of CD34-CD56+ mesoderm cell population at day 4 of differentiation. Likewise, at day 6 of differentiation, the size of the CD34-CD56+ population generated by the control SeV iPSCs was similar to that of the SDS SeV iPSCs (Fig. 14B).

Using an additional set of antigens that mark mesoderm cells at day 4 of differentiation,

CD34-KDR+C-kit- cells, it was again apparent that the control and SDS SeV iPSCs generated comparable mesoderm populations. At day 6 of differentiation, a marked, but similar KDR+C-kit- population was still present in the control and SDS samples (Fig.

15B). Although previous studies have reported that there is no general mesoderm lineage defect in SDS SeV iPSCs, taken together, our results are the first to indicate that

67 mesoderm generation is not affected within the context of specifically, definitive hematopoiesis in SDS SeV iPSCs [54].

Fig 14. Mesoderm induction during definitive hematopoiesis

A. SB-431542 treated EBs were assessed for the generation of mesoderm at days 4 and 6 of differentiation using flow cytometry. The CD56+ CD34- mesoderm population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, the control and SDS Sev iPSCs generate comparable levels of mesoderm at both days 4 (p=0.29) and 6 (p=0.26) of differentiation, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=1 for N1-I at T4, n=3 for N1-K at T4, n=2 for P1-A at T4, n=2 for P1-D at T4, n=2 for N1-I at T6, n=1 for N1-K at T6, n=3 for P1-A at T6, n=0 for P1-D at T6.

Fig 15. Mesoderm induction during definitive hematopoiesis

A. SB-431542 treated EBs were assessed for the generation of mesoderm at days 4 and 6 of differentiation using flow cytometry. (Gating on the CD34- population is not shown). The KDR+ C-kit- CD34- mesoderm population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, the control and SDS Sev iPSCs generate comparable levels of mesoderm at both days 4 (p=0.4) and 6 (p=0.4) of differentiation, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=1 for N1-I at T4, n=4 for N1-K at T4, n=2 for P1-A at T4, n=3 for P1-D at T4, n=1 for N1-I at T6, n=2 for N1-K at T6, n=2 for P1-A at T6, n=1 for P1-D at T6.

69 3.6.2 Induction of cells with hemogenic endothelium potential is

reduced in SDS iPSCs

Mesoderm is the precursor to HE, a specialized population of endothelial cells with hematopoietic and endothelial capabilities. Given that mesoderm induction was not impaired in our differentiation system, our next step was to investigate the generation of population with HE potential, that is CD34+CD43-CD45- or CD34+CD31+.

The iPSCs were differentiated towards definitive hematopoiesis, and the EBs were dissociated at T6, T9, T12 and T15, and assessed for the expression of CD34 and CD43

(Fig. 16A). At day 6 of differentiation, a CD34+CD43-CD45- was already present in both samples, albeit, to a significantly lesser degree in the SDS sample. Likewise, at days 9 and 12, the SDS SeV iPSCs had generated a significantly reduced population with HE potential, when compared to the control SeV iPSCs. By day 15, as the HE-containing population differentiated towards a more mature population, it acquired CD43 positivity, and consequently lost CD34 positivity. The CD34+CD43-CD45- population therefore decreased in size, and a significant difference between the size of the population with HE potential in both samples was no longer apparent at T15 (Fig. 16B).

To further verify that HE induction is in fact impaired in SDS SeV iPSCs, we checked for the presence of a CD34+CD31+ population at day 9 of differentiation (Fig. 17A). As with the CD34+CD43-CD45- population, the CD34+CD31+ HE-containing population size was significantly reduced in the SDS samples, when compared to the controls (Fig. 17B).

Taken together, these novel results indicate that the hematopoietic defect observed in

70 SDS may be initiated as early as during the HE induction stage of definitive hematopoiesis.

Fig 16. Induction of cells with hemogenic endothelium potential during definitive

hematopoiesis

A. SB-431542 treated EBs were assessed for the generation of a HE-containing population at days 6,9,12 and 15 of differentiation using flow cytometry. (Gating on the CD45- population is not shown). The CD34+ CD43- CD45- HE-containing population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS Sev iPSCs generate significantly lower levels of a population with HE potential at days 6 (p=0.045), 9 (p=0.036) and 12 (p=0.049) of differentiation when compared to the controls, as determined by the T-test (*p<0.05). By day 15 of differentiation, the control and SDS SeV iPSCs generate comparable numbers of the CD34+ CD43- CD45- HE population (p=0.45). The error bars represent the SEM. n=2 for N1-I at T6, n=2 for N1-K at T6, n=2 for P1-A at T6, n=1 for P1-D at T6, n=3 for N1-I at T9, n=3 for N1-K at T9, n=3 for P1-A at T9, n=2 for P1-D at T9, n=3 for N1-I at T12, n=3 for N1-K at T12, n=2 for P1-A at T12, n=2 for P1-D at T12, n=3 for N1-I at T15, n=2 for N1-K at T15, n=3 for P1-A at T15, n=1 for P1-D at T15.

Fig 17. Induction of cells with hemogenic endothelium potential during definitive

hematopoiesis

A. SB-431542 treated EBs were assessed for the generation of a HE-containing population at day 9 of differentiation using flow cytometry. (Gating on the CD45- population is not shown). The CD34+ CD31+ population with HE potential is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS Sev iPSCs generate significantly lower levels of a population with HE potential at day 9 (p=0.013), as determined by the T-test (*p<0.05). The error bars represent the SEM. n=3 for N1-I at T9, n=3 for N1-K at T9, n=3 for P1-A at T9, n=2 for P1-D at T9.

72 3.6.3 Early hematopoietic progenitor induction is reduced in SDS

iPSCs

Having determined that the induction of cells with HE potential is impaired in our definitive hematopoiesis differentiation system, we further investigated whether this would have an effect on the generation of EHPs. EHPs derived from pluripotent stem cells express the CD34 antigen and the panhematopoietic CD45 antigen. These cells do not express endothelial antigens such as CD31 and VE-CAM. As well, they are negative for mature lineage specific blood cell markers.

Given that cells with HE potential are seen as early as day 6 of differentiation, we expected to see a more mature population of EHPs starting with day 9. At day 9 of differentiation, the EHP population is small, but increases substantially by day 12. When staining for a CD34+CD45+ population, also characteristic of EHPs, we observed that by day 12 of differentiation, the control iPSCs had generated a significantly larger population when compared to the patient sample (Fig. 18A). Likewise, by day 15, the

EHP population generated by the SDS iPSCs was less than half as large as that generated by the control samples (Fig. 18B).

The Slukvin group showed that EHPs express CD43 antigen in addition to CD34 (Fig.

19A) [246]. By day 12, the control samples had a significantly larger CD34+CD43+ population, as compared to the SDS SeV iPSCs. Although by day 15, the EHP population had decreased in size as the cells lost expression of CD34, the control samples still possessed a significantly greater CD34+CD43+ population (Fig. 19B). These results

73 therefore indicate that definitive hematopoiesis is also affected downstream of the HE stage, as EHP induction is reduced in SDS SeV iPSCs.

Fig 18. Early Hematopoietic Progenitor induction during definitive hematopoiesis

A. SB-431542 treated EBs were assessed for the generation of EHPs at days 9,12 and 15 of differentiation using flow cytometry. The CD34+ CD45+ EHP population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, both SDS and control Sev iPSCs generate very few EHPs at day 9 of differentiation (p=0.41). However, by days 12 (p=0.03) and 15 (p=0.039) of differentiation, the SDS SeV iPSCs generate a significantly reduced EHP population when compared to the controls, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=3 for N1-I at T9, n=3 for N1-K at T9, n=3 for P1-A at T9, n=2 for P1-D at T9, n=3 for N1-I at T12, n=3 for N1-K at T12, n=2 for P1-A at T12, n=2 for P1-D at T12, n=3 for N1-I at T15, n=2 for N1-K at T15, n=3 for P1-A at T15, n=1 for P1-D at T15.

Fig 19. Early Hematopoietic Progenitor induction during definitive hematopoiesis

A. SB-431542 treated EBs were assessed for the generation of EHPs at days 9,12 and 15 of differentiation using flow cytometry. The CD34+ CD43+ EHP population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS and control Sev iPSCs generate comparable levels of EHPs at day 9 of differentiation (p=0.439). However, by days 12 (p=0.047) and 15 (p=0.039) of differentiation, the SDS SeV iPSCs generate a significantly reduced EHP population when compared to the controls, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=3 for N1-I at T9, n=3 for N1-K at T9, n=3 for P1-A at T9, n=2 for P1-D at T9, n=3 for N1-I at T12, n=3 for N1-K at T12, n=3 for P1-A at T12, n=2 for P1-D at T12, n=3 for N1-I at T15, n=2 for N1-K at T15, n=3 for P1-A at T15, n=1 for P1-D at T15.

75 3.6.4 Myeloid cell induction is reduced in SDS iPSCs

As EHPs mature they give rise to more differentiated cells, among which are those of the myeloid lineage. To determine if myeloid cell induction is affected in SDS SeV iPSCs, the cells were stained for CD34 and CD45, and assessed for the presence of a CD34-

CD45+ myeloid-containing population (Fig. 20A). At day 9 of differentiation, the CD34-

CD45+ was quite small, with no significant differences in the sizes of this population between the control and patient samples. However, by day 12, the myeloid-containing population had markedly increased in size in both samples, although the control SeV iPSCs generated a significantly larger population than the SDS SeV iPSCs. By day 15, the CD34-CD45+ had yet again markedly increased, with the control samples generating a significantly larger myeloid population than the SDS samples (Fig. 20B). These results suggest that during definitive hematopoiesis, myeloid cell induction is impaired within

SDS SeV iPSCs – mirroring the myeloid cell deficit commonly observed during the post- natal life of SDS patients.

Fig 20. Myeloid Cell induction during definitive hematopoiesis

A. SB-431542 treated EBs were assessed for the generation of a myeloid cell-containing population at days 9,12 and 15 of differentiation using flow cytometry. The CD34- CD45+ myeloid cell population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS and control Sev iPSCs generated very low levels of myeloid cells at day 9 of differentiation (p=0.22). However, by days 12 (p=0.03) and 15 (p=0.027) of differentiation, the SDS SeV iPSCs generate a significantly reduced CD34- CD45+ myeloid cell population when compared to the controls, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=3 for N1-I at T9, n=3 for N1-K at T9, n=3 for P1-A at T9, n=2 for P1-D at T9, n=3 for N1-I at T12, n=3 for N1-K at T12, n=2 for P1-A at T12, n=2 for P1-D at T12, n=3 for N1-I at T15, n=2 for N1-K at T15, n=3 for P1-A at T15, n=1 for P1-D at T15.

77 3.6.5 Granulocyte/Monocyte induction is reduced in SDS iPSCs

Given that our results showed that myeloid cell induction is impaired in SDS SeV iPSCs, our next step was to investigate whether the granulocyte/monocyte population was specifically affected, as seen in SDS patients. Briefly, the samples were differentiated towards definitive hematopoiesis, and assessed for the presence of a CD34-

CD45+CD11b+ granulocyte/monocyte population at days 12 and 15 (Fig. 21A). At day 12 of differentiation, although the control samples had generated a markedly larger granulocyte/monocyte population as compared to the patient samples, this difference was not significant. However, by day 15, the control SeV iPSCs had a significantly larger granulocyte/monocyte population, that was more than two times greater than that generated by the SDS SeV-iPSCs (Fig 21B). These results indicate that granulocyte/monocyte generation is impaired during definitive hematopoiesis in SDS

SeV iPSCs.

Our results so far suggest for the first time, that definitive hematopoiesis in SDS is characterized by a defect that is initiated as early as during the HE stage, and that it affects all downstream developmental stages, leading to a general reduction in the ability of SDS SeV iPSCs to generate EHPs, myeloid cells, and specifically granulocytes/monocytes.

Fig 21. Granulocyte/Monocyte induction during definitive hematopoiesis

A. SB-431542 treated EBs were assessed for the generation of a granulocyte/monocyte population at days 12 and 15 of differentiation using flow cytometry. (Gating on the CD45+ population is not shown.) The CD34- CD45+ CD11b+ population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS and control Sev iPSCs generate comparable levels of granulocytes/monocytes at day 12 of differentiation (p=0.12). However, by day 15 (p=0.008) of differentiation, the SDS SeV iPSCs generate a significantly reduced CD34- CD45+ CD11b+ granulocyte/monocyte population when compared to the controls, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=3 for N1-I at T12, n=3 for N1-K at T12, n=2 for P1-A at T12, n=2 for P1-D at T12, n=2 for N1-I at T15, n=3 for N1-K at T15, n=3 for P1-A at T15, n=1 for P1-D at T15.

79 3.7 SDS iPSCs manifest a delay in the development of primitive

hematopoiesis, without a quantitative defect

3.7.1 Mesoderm/Hemangioblast induction is intact in SDS iPSCs

Having observed a defect in definitive hematopoiesis in SDS SeV iPSCs, we aimed to determine whether this is mirrored in primitive hematopoiesis as well. The SeV iPSCs were preferentially differentiated towards the primitive wave of hematopoiesis using Act.

A, and the samples were assessed for the presence of a mesodermal population with hemangioblast potential. Briefly, EBs maintained in primitive hematopoiesis culture conditions were dissociated at day 4 of differentiation, and assessed for the presence of a

KDR+c-kit-CD34- population, characteristic of mesoderm at this stage (Fig. 22A). Gating on the CD34- population is not shown for either sample, as all cells were negative for this marker at T4. At day 4 of differentiation, both the control and SDS samples had generated a substantial, but comparable KDR+C-kit-CD34- population (Fig. 22B). These results indicate that much like what is seen with definitive hematopoiesis, there is no defect in the ability of SDS SeV iPSCs to generate mesoderm within the context of primitive hematopoiesis.

Fig 22. Mesoderm/Hemangioblast induction during primitive hematopoiesis

A. Act. A treated EBs were assessed for the generation of a myeloid cell-containing population at day 4 of differentiation using flow cytometry. (Gating on the CD34- population is not shown). The KDR+ C-kit- CD34- mesoderm population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS and control Sev iPSCs generate comparable levels of mesoderm at day 4 of differentiation (p=0.17), as determined by the T-test (*p<0.05). The error bars represent the SEM. n=2 for N1-I at T4, n=3 for N1-K at T4, n=1 for P1-A at T4, n=3 for P1-D at T4.

81 3.7.2 Induction of cells with hemogenic endothelium potential is not

impaired in SDS iPSCs

Having determined that mesoderm generation is not affected in SDS SeV iPSCs, we next investigated the ability of SDS cells to generate a HE-containing population. A study of differentiation kinetics was done by dissociating the EBs at days 9 through 12 of differentiation, and by assessing for the presence of a CD34+CD43- population characteristic of a population with HE potential (Fig. 23A). Upon visual inspection of the flow cytometry plots, we did not observe any striking differences in the size of the HE- containing population between the control and SDS samples at days 9 through 12. More specifically, by day 9 of differentiation, both samples had generated similarly sized

CD34+CD43- populations. Although by day 12, the population with HE potential had subsided, the size of this population was comparable between control and SDS samples

(Fig. 23B).

Given that CD31 is one of the markers expressed on the population with HE potential, we also tested the development of the HE-containing population by assessing the generation of a CD34+CD31+ population at days 9 and 12 of differentiation (Fig. 24A). At day 9, as well as day 12, we noticed no significant differences in the size of the CD34+CD31+ HE- containing population between the control and SDS samples (Fig. 24B). Taken together, these results indicate that the induction of a population with HE potential is not affected within the primitive wave of hematopoiesis in SDS SeV iPSCs.

Fig 23. Induction of cells with hemogenic endothelium potential during primitive

hematopoiesis

A. Act. A treated EBs were assessed for the generation of a HE-containing population at days 9 through 12 of differentiation using flow cytometry. The CD34+ CD43- population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS and control Sev iPSCs generate comparable levels of a HE-containing population at days 9 (p=0.32) and 12 (p=0.092) of differentiation, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=5 for N1-I at T9, n=1 for N1-K at T9, n=4 for P1-A at T9, n=1 for P1-D at T9, n=5 for N1-I at T12, n=1 for N1-K at T12, n=4 for P1-A at T12, n=1 for P1-D at T12, n=3 for N1-I at T15, n=0 for N1-K at T15, n=2 for P1-A at T15, n=0 for P1-D at T15.

Fig 24. Induction of cells with hemogenic endothelium potential during primitive

hematopoiesis

A. Act. A treated EBs were assessed for the generation of a HE-containing population at days 9 and 12 of differentiation using flow cytometry. The CD34+ CD31+ population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS and control Sev iPSCs generate comparable levels of a population with HE- potential at days 9 (p=0.497) and 12 (p=0.206) of differentiation, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=3 for N1-I at T9, n=1 for N1-K at T9, n=2 for P1- A at T9, n=1 for P1-D at T9, n=3 for N1-I at T12, n=0 for N1-K at T12, n=2 for P1-A at T12, n=0 for P1-D at T12.

84 3.7.3 Early hematopoietic progenitor induction is not affected in SDS

iPSCs

Given our observation that the induction of cells with HE potential was not affected within our samples, we next investigated the ability of SDS SeV iPSCs to generate EHPs.

Specifically, the EBs were dissociated at days 9 through 12 of differentiation, and assessed for the presence of a CD34+CD43+ EHP population. Upon visual inspection of the kinetics of differentiation, we did not observe any striking differences between the control and SDS samples - both samples acquired CD43 positivity at a similar rate (Fig.

25A). Specifically, the control and SDS samples generated comparable CD34+CD43+

EHPs populations at both days 9 and 12 of differentiation (Fig. 25B).

Next, we investigated the presence of a CD34+CD45+ EHP population at days 9 through

12 of differentiation. Much like what was seen with the CD34+CD43+, the kinetics of differentiation towards a more mature, CD45+ population in SDS iPSCs were comparable to those of the control SeV iPSCs (Fig. 26A). Specifically, we did not observe any significant differences in the CD34+CD45+ population between the control and SDS samples at both days 9 and 12 of differentiation (Fig. 26B). These results indicate that there is no defect in the ability of SDS SeV iPSCs to generate EHPs during primitive hematopoiesis.

Fig 25. Early Hematopoietic Progenitor induction during primitive hematopoiesis

A. Act. A treated EBs were assessed for the generation of a EHP population at days 9 through 12 of differentiation using flow cytometry. The CD34+ CD43+ HE population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS and control Sev iPSCs generate comparable levels of EHPs at days 9 (p=0.441) and 12 (p=0.27) of differentiation, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=5 for N1-I at T9, n=1 for N1-K at T9, n=3 for P1-A at T9, n=2 for P1-D at T9, n=5 for N1-I at T12, n=1 for N1-K at T12, n=3 for P1-A at T12, n=2 for P1-D at T12.

Fig 26. Early Hematopoietic Progenitor induction during primitive hematopoiesis

A. Act. A treated EBs were assessed for the generation of a EHP population at days 9 through 12 of differentiation using flow cytometry. The CD34+ CD45+ HE population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS and control Sev iPSCs generate comparable levels of EHPs at days 9 (p=0.424) and 12 (p=0.3) of differentiation, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=5 for N1-I at T9, n=1 for N1-K at T9, n=3 for P1-A at T9, n=2 for P1-D at T9, n=5 for N1-I at T12, n=1 for N1-K at T12, n=3 for P1-A at T12, n=2 for P1-D at T12.

87 3.7.4 Mature blood cell induction is not impaired in SDS iPSCs

As EHPs differentiate further towards mature blood cells, they lose CD34 positivity, while retaining CD43 and CD45 positivity. Therefore, to determine if primitive hematopoiesis may be impaired at the mature blood cell stage, we assessed our samples for a CD34-CD43+, or alternatively a CD34-CD45+ population at days 9 through 12 of differentiation. When looking at the differentiation kinetics, we yet again observed a comparable progression towards a CD34-CD43+ from day 9 to 12 between the control and

SDS samples (Fig. 27A). Specifically, we did not observe any significant differences in the size of the CD34-CD43+ mature blood cell population between the samples at both days 9 and 12 (Fig. 27B).

To verify that the SDS iPSCs were not experiencing a defect in their ability to generate mature blood cells, we further investigated the CD34-CD43+ population, as Kennedy & colleagues have previously stated that a CD34-CD43+ population that carries primitive hematopoiesis potential is also positive for CD235a and CD41 [156]. As the CD34-

CD43+ population was greatest at day 12 of differentiation, the CD41/CD235a profile was only investigated at this time point. In both the control and SDS samples we observed a marked, but comparable population positive for CD41, and a separate, marked but comparable population with CD235a positivity (Fig. 27C-D).

We also assessed our samples for a more mature, CD45+ population that has lost expression of CD34, at days 9 through 12 of differentiation. Upon visual inspection, we again observed that both the control and SDS samples progressed towards a CD34- state

88 in a comparable manner from day 9 to 12 (Fig. 28A). Specifically, at both days 9 and 12 of differentiation, there were no significant differences in the sizes of the CD34- CD45+ population between the control and SDS samples (Fig. 28B). Taken together, these results indicate that mature blood cell induction is not impaired in SDS SeV iPSCs during primitive hematopoiesis.

Fig 27. Mature Blood Cell induction during primitive hematopoiesis

A. Act. A treated EBs were assessed for the generation of a mature blood cell population at days 9 through 12 of differentiation using flow cytometry. The CD34- CD43+ mature blood cell population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS and control Sev iPSCs generate comparable levels of mature blood cells at days 9 (p=0.244) and 12 (p=0.318) of differentiation, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=5 for N1-I at T9, n=1 for N1-K at T9, n=3 for P1-A at T9, n=2 for P1- D at T9, n=5 for N1-I at T12, n=1 for N1-K at T12, n=3 for P1-A at T12, n=2 for P1-D at T12. C. The day 12, CD34- CD43+ mature blood cell population was assessed for its expression of CD41 and CD235a. As shown, SDS and control SeV iPSCs both have the ability to generate CD41 single positive, CD235a single positive, as well as CD41/CD235a double positive populations. D. As shown, SDS and control Sev iPSCs generate comparable levels of CD41+ (p=0.0514) and CD235a+ (p=0.448) at day 12 of differentiation, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=5 for N1-I at T12, n=1 for N1-K at T12, n=3 for P1-A at T12, n=2 for P1-D at T12

Fig 28. Mature Blood Cell induction during primitive hematopoiesis

A. Act. A treated EBs were assessed for the generation of a mature blood cell population at days 9 through 12 of differentiation using flow cytometry. The CD34- CD45+ HE population is indicated by a green and red rectangle for the control and SDS SeV iPSCs, respectively. B. As shown, SDS and control Sev iPSCs generate comparable levels of mature blood cells at days 9 (p=0.266) and 12 (p=0.346) of differentiation, as determined by the T-test (*p<0.05). The error bars represent the SEM. n=5 for N1-I at T9, n=1 for N1-K at T9, n=3 for P1-A at T9, n=2 for P1- D at T9, n=5 for N1-I at T12, n=1 for N1-K at T12, n=3 for P1-A at T12, n=2 for P1-D at T12.

91 3.7.5 SDS iPSCs showcase a delay in their ability to form hematopoietic

progenitors characteristic of primitive hematopoiesis

Having determined by flow cytometry that there is no apparent impairment in the induction of the various developmental steps in primitive hematopoiesis, our next step was to verify this finding by using colony assays. Briefly, the samples were preferentially differentiated towards primitive hematopoiesis, the EBs were dissociated at day 9, plated into methylcellulose in the presence of various cytokines, and assessed for their ability to generate hematopoietic progenitors characteristic of primitive hematopoiesis – CFU-GM and BFU-E colonies. Interestingly, when the EBs were dissociated at day 9, the SDS SeV iPSCs gave rise to significantly fewer CFU-GMs as compared to the control SeV iPSCs, although they both gave rise to comparable numbers of BFU-E colonies (Fig. 29A).

Given that we did not observe a marked defect in the kinetics of differentiation of SDS

SeV iPSCs during primitive hematopoiesis, we wished to investigate our results further.

Therefore, we maintained the samples in differentiation media until day 12 rather than day 9, following which we dissociated the EBs and plated them into methylcellulose. We observed that if the SDS SeV-iPSCs are left to differentiate as EBs until day 12, they will generate comparable numbers of CFU-GMs as compared to the control SeV-iPSCs.

Much like what was seen with the day 9 samples, the numbers of BFU-E colonies generated from the day 12 samples were similar between the control and SDS SeV-iPSCs

(Fig. 29B).

92 Our novel results indicate that rather than an impairment in the ability of SDS SeV iPSCs to induce the development of the various stages in primitive hematopoiesis, there is only a delay.

Fig 29. Delayed formation of hematopoietic progenitors characteristic of primitive hematopoiesis in SDS induced pluripotent stem cells

Colonies generated using clonogenic assays from Act. A-treated EBs dissociated at days 9 and 12 of differentiation were enumerated. Colonies were counted in triplicate for each data point. SDS SeV iPSC colony numbers are reported relative to the number of colonies generated by the SeV Control iPSCs.

A. As shown, SeV SDS iPSCs plated at day 9 of differentiation generate significantly fewer CFU- GM colonies when compared to the controls, as determined by the T-test (*p<0.05). However, the control and SDS SeV iPSCs generate comparable numbers of BFU-E colonies. The error bars represent the SEM. n=3 for N1-I, n=3 for N1-K, n=3 for P1-A, n=3 for P1-D. B. With the EBs that are dissociated at day 12 of differentiation, both the SeV Control and SDS iPSCs generate comparable numbers of CFU-GM and BFU-E clusters, as determined by the T- test (*p>0.05). The error bars represent the SEM. n=2 for for N1-I, n=1 for N1-K, n=2 for P1-A, n=1 for P1-D.

93 Chapter IV

Discussion

4.1 Discussion of results

SDS is a rare, autosomal-recessive IBMFS. It is a multisystem disorder characterized by bone marrow failure, pancreatic insufficiency and skeletal abnormalities, as well as predisposition to MDS/AML [3-5, 17, 18]. Given that hematopoietic abnormalities are the leading cause of death among SDS patients, our study was focused on acquiring a better understanding of blood development in SDS.

We aimed to first generate and characterize a human model of SDS using iPSCs. With the CCRM facility, our lab generated SDS iPSCs by reprogramming patient BM fibroblasts using SeVs. The cells demonstrated pluripotency and carried the donor-cell

SBDS mutations. As is seen in SDS patients, SDS iPSCs manifested a defect in ribosome assembly and reduction in hematopoietic progenitor formation. After confirming that the

SDS iPSCs recapitulated the aberrant human SDS ribosome profile and hematopoietic cell formation phenotype, we used them to study the various developmental stages during primitive and definitive hematopoiesis, to determine the hematopoietic stage where the

SDS defect begins, and the various stages that manifested an aberrant hematopoietic phenotype.

We showed that during directed differentiation towards the definitive wave of hematopoiesis, mesoderm induction is not impaired. However, our study showed that there was a defect in the generation of cells with HE potential during definitive

94 hematopoiesis. We also studied the developmental stages downstream of the HE stage

(EHP, myeloid cells and granulocytes), and observed that their formation is also impaired in SDS iPSCs. Next, we asked whether the SDS phenotype is restricted to definitive hematopoiesis, or if it is more global, also affecting the primitive wave of hematopoiesis.

Since thus far, no SDS patients with complete absence of SBDS protein have been identified, and complete knockout in mice results in early demise at the stage of E6.5, it is reasonable to hypothesize that some level of SBDS protein may be required for earlier stages of embryonic development, such as the primitive wave of blood formation. Similar to the definitive wave, SDS iPSCs did not manifest a defect in mesoderm induction during the primitive wave. However, in contrast to definitive hematopoiesis, no deficiency in the generation of a population containing HE cells was observed during primitive hematopoiesis. Further, generation of EHPs and mature blood cells was also intact during the primitive wave of hematopoiesis. As differentiation during the primitive wave was not impaired, we asked whether the primitive hematopoietic precursors generated were functionally intact. We found that the maturation of SDS iPSCs in terms of hematopoietic colony formation potential was delayed, but not impaired.

Taken together, these studies indicate that the SDS hematopoietic phenotype starts at the stage of hematopoietic specification, rather than at an earlier stage in the stepwise development of PSCs towards mesoderm. Specifically, our observations show that definitive hematopoiesis is impaired, with the defect being initiated during the induction of HE, and affecting downstream developmental stages. Our results also showed that primitive hematopoiesis is not impaired, as all developmental stages studied were intact.

95 Rather, we observed that there is a delay in the ability to generate hematopoietic progenitors during the primitive wave.

4.1.1 SDS iPSCs can be generated using integrative and non-integrative

transgenes

Using the stem cell facility at CCRM, we generated control and SDS iPSCs from healthy and SDS patient BM fibroblasts, using Sendai viruses. The reported reprogramming efficiency of iPSCs from skin fibroblasts using SeVs is approximately 1% [244]. In our study, the SeV control and SDS iPSCs were generated with a reprogramming efficiency of 0.04% and 0.002%, respectively. The markedly lower efficiencies observed in our study might be related to the source of cells for reprogramming (BM), as the origin of the parental cell can affect the reprogramming efficiency [234]. The age of the donor may also affect reprogramming efficiency, as cells from younger people have a tendency to reprogram much more efficiently than those of older people [239]. Further, with reprogramming come necessary epigenetic changes, along with inadvertent ones, which can ultimately lead to partial reprogramming, therefore reducing the reprogramming efficiency [243]. Although both control and SDS iPSCs were generated in the presence of the same reprogramming conditions, the control iPSCs had an efficiency that was 20 times greater than that of SDS iPSCs. This may be due to the disease-causing mutations in the SDS samples, which may negatively impact the process of reprogramming – as has been demonstrated with FA patient fibroblasts, which could not be reprogrammed unless the cells were genetically corrected [247]. In FA this was attributed to the DNA-repair defect inherent to FA cells, which makes it difficult to restore the DNA damage that may

96 be induced during reprogramming. In SDS this might be related to the hypersensitivity to

Fas ligand and impaired ribosome assembly characteristic of SDS cells.

The expression of cell-surface markers such as OCT4, SSEA-4, TRA-1-81, NANOG and

TRA-1-60 indicates pluripotency. As such, the SeV control and SDS iPSCs were assessed for their expression of pluripotency markers, OCT4, SSEA-A, TRA-1-81,

NANOG and TRA-1-60, and all were detected. These results are in line with what has been previously reported by Park & colleagues regarding RT SDS iPSC lines, and confirm that the generated iPSC lines are pluripotent in nature [121].

To confirm that the SeV iPSC lines are functionally pluripotent, they were differentiated as EBs and assessed using RT-qPCR for the expression of markers characteristic of all three germ layers. Relative to their undifferentiated states, all 4 hiPSC lines showed upregulated expression of AFP, HAND1 and NEUROD1, which are characteristic of endoderm, mesoderm and ectoderm, respectively. These results are in line with what has been previously reported regarding RT SDS iPSC lines, and confirm that the iPSC lines have multilineage potential [121]. Although all iPSC lines were differentiated in the same culture conditions, we observed skewing of the expression levels of the three germ layer genes in each iPSC line. Although studies have shown that iPSCs have an increased propensity to differentiate towards the original cell type that they were derived from – in this case, cells of mesodermal origin that – we did not observe this in our study [235]. We also noticed differences in the gene expression patterns among all 4 iPSC lines. These deviations from what is expected may be as a result of cloning individual cells during

97 reprogramming, which allows for the preservation of each cell’s unique mutational history, despite being derived from the same source [248]. As mentioned previously, reprogramming can also inadvertently introduce various epigenetic changes that may influence the differentiation capacity of these iPSCs [243].

The finding that RT and SeV control and SDS iPSCs can give rise to hematopoietic progenitors was confirmed by our observation that the RT and SeV iPSC CFUs had generated cells whose morphology was indicative of macrophages, granulocytes and erythrocytes. These results indicate that the SeV SDS iPSCs are an adequate model for the study of hematopoiesis, as they can give rise to terminally differentiated blood cells.

This is the first study to use iPSC lines generated using Sendai viruses to study SDS. Use of integrative viruses such as retroviruses (as used by Tulpule & colleagues) can increase the risk of multiple insertions, and tumorigenicity, and make it cumbersome to silence or activate the viral transgenes. This affects their potential to differentiate and makes their use in the clinic unsafe [54, 249]. However, using non-integrative viruses such as SeV can help eliminate some of these risks.

4.1.2 Recapitulation of SDS using SDS iPSCs

One of the biggest challenges with using iPSCs to model diseases is being able to differentiate them into the cell types of interest. To determine whether SDS iPSCs were an adequate model for SDS, we exposed them to a hematopoietic differentiation protocol skewed towards definitive hematopoiesis. As Kennedy & colleagues have reported, intact

98 PSCs should give rise to CFU-GEMM, CFU-GM and BFU-E colonies in the presence of the aforementioned conditions [156]. While the SeV control and SDS iPSCs both gave rise to CFU-GEMM, CFU-GM and BFU-E colonies, the RT control and SDS iPSCs only gave rise to CFU-GM and BFU-E colonies.

The inability to generate CFU-GEMM colonies by the RT iPSCs might have been as a result of the integrative viruses that were used to generate these cell lines. This might have inadvertently introduced genetic changes, and consequently affected the iPSCs’ differentiation potential.

Alternatively, the differences in genetic composition with respect to the SBDS mutations may have also been responsible for the variations in hematopoietic differentiation potential. Tulpule & colleagues observed a similar phenomenon, albeit within the context of pancreatic development, when using two RT SDS iPSC lines derived from two patients carrying different mutations [54]. The RT and SeV SDS iPSCs both carry the common IV2+2T>C splice-site mutation. The SeV SDS iPSCs carry a second, common, nonsense mutation (183-184TA>CT) that has been documented to give rise to classic features of SDS. The RT SDS iPSCs carry a second splice-site mutation (IVS3-1G>A), which has been reported to have an atypical presentation, with a significantly lower neutrophil count than what is observed in most SDS patients [21]. This indicates that the hematological phenotype of the patient from which the RT SDS iPSCs were derived may have been much more severe than that of the patient from which the SeV iPSCs were derived – and this was reflected in their inability to generate CFU-GEMM colonies [21].

Given the discrepancy in CFU-GEMM colony potential between the RT and SeV iPSC lines, to preserve consistency, only the SeV iPSCs lines were used for further assessment of SDS, as they provided us with a more adequate and appropriate model of definitive hematopoiesis in-vitro.

Given that one of the hallmarks of SDS is impaired blood development, it was important to determine if the SDS iPSCs gave rise to a reduced number of CFUs in comparison to the control iPSCs. Similar to what was reported by Tulpule & colleagues, the SeV SDS iPSCs gave rise to significantly fewer CFUs, with a general reduction across all large and small CFUs identified in culture (CFU-GEMM, CFU-GM and BFU-E colonies) [54].

This finding is consistent with what has been previously reported by Dror & colleagues regarding the significantly reduced ability of SDS CD34+ HSC/Ps to generate hematopoietic colonies in vitro [32]. These results are also in accordance with the SDS hematopoietic phenotype, characterized by neutropenia, anemia and also thrombocytopenia. However, there were no statistically significant differences between the number of CFU-GM and BFU-E clusters (groups of <50 cells) between the SeV control and SDS iPSCs. It is possible that the CFU-GM and BFU-E clusters generated by the SeV iPSCs are a consequence of the accelerated apoptosis that characterizes SDS hematopoietic progenitors [27]. Although SeV SDS iPSCs may hold the potential to give rise to large CFU-GEMM, CFU-GM and BFU-E colonies, follow-up studies should be focused on the activity of the Fas signaling pathway in iPSCs-derived SDS hematopoietic

100 cells, as hyperactivation of this pathway may lead to premature cell death, and prevent colony growth beyond the size of a cluster [27].

Our ribosome profile results indicate that the SDS iPSCs recapitulate the ribosome biogenesis defect that characterizes SBDS-deficient cells. The SDS iPSCs had a marked reduction in the 80S subunit, which is consistent with previous reports from our lab using

SBDS-deficient myeloid cells, as well as reports by Tulpule & colleagues, Burwick & colleagues and Finch & colleagues [24, 34, 54, 106]. These findings indicate that the ribosome profile defect characterizes SDS cells also at the stage of pluripotency, and is consistent with what has been previously reported regarding the role of SBDS in displacing eIF6 from the 60s subunit, only after which can the 60s subunit associate with the 40s subunit to form the mature 80s monomer [24].

Our results also indicate that there is a significant reduction in the 40S/60S subunit ratio in the SDS iPSCs. This result was inconsistent with what has been previously reported in

SDS patient cells and RT iPSCs, where the 40S/60S subunit ratio actually experienced a mild increase in SBDS-deficient cells when compared to controls, due to reduced 60S subunit levels [24, 54]. Ganapathi & colleagues have also reported that there were no changes in the relative ratio of the 40S and 60S subunits, with no reductions in the levels of these subunits in SDS patient lymphoblasts [250]. These inconsistencies may be as a result of the various mutational backgrounds of the SDS cells that were used for the aforementioned studies. The reduction may also point to the possibility of a defect in rRNA processing in SDS cells, which has not been previously reported. These novel

101 findings warrant further investigation, to determine the reason for the aforementioned discrepancies in ribosome profiles.

4.1.3 Mesoderm development is intact when SDS iPSCs are induced to

undergo definitive hematopoiesis

Given the observation that SeV SDS iPSCs give rise to significantly fewer hematopoietic progenitors in methylcellulose cultures when the cells were induced to undergo definitive hematopoiesis, we asked whether this was due to a defect in inducing the mesodermal germ layer, from which blood cells are ultimately derived. Our results indicate that at all time points assessed, the SeV control and SDS iPSCs generated a comparable mesoderm population. The results replicate those previously reported by Tulpule & colleagues, and are in keeping with the known phenotype of SDS patients, where some, but not all mesoderm-derived organs are affected [54]. It is possible that the tissue specificity is as a result of differential requirement for the ribosomal functions of SBDS in various organ systems.

Interestingly, SBDS-null mice experience embryonic lethality, as they are unable to undergo gastrulation, and therefore, do not form any of the three germ layers, including mesoderm [112]. Perhaps the reason why the SDS phenotype is not as severe in humans is because all living SDS patients have residual SBDS protein expression, which is likely sufficient for promoting induction of mesoderm. It is also possible that varying spatial and temporal conditions are not met when SBDS is completely absent, therefore impeding the generation of more committed cell populations [23].

102

Taken together, the comparable ability of SeV SDS iPSCs to generate mesoderm much like the control SeV iPSCs points to a defect in hematopoietic induction downstream of mesoderm formation.

4.1.4 The SDS definitive hematopoietic defect was first noticed at the

hemogenic endothelium induction phase

Considering that mesoderm induction during definitive hematopoiesis was not impaired, we asked whether the onset of the hematopoietic defects was downstream of the mesoderm stage. Our results indicate that the induction of cells with HE potential is impaired during the definitive wave of hematopoiesis in SeV SDS iPSCs. Downstream developmental stages, including the generation of EHPs, and myeloid cells (specifically granulocytes/monocytes) were also impaired, much like what Tulpule & colleagues observed [54]. Although it is unclear whether the reduction in HE is responsible for the reduction in the downstream developmental stages, our novel results indicate for the first time that the onset of the characteristic hematopoietic phenotype may be initiated at the stage of HE development.

There are several plausible explanations for the results described previously. At day 9 of differentiation, cells with HE potential (CD34+CD43-) were isolated from both SeV control and SDS iPSCs. When expanded and assessed for CFU formation, the SeV SDS iPSCs gave rise to significantly fewer CFUs than the SeV control iPSCs. This result indicates that control and SDS HE populations are functionally different. Further studies

103 are needed to better characterize the SeV control and SDS-derived HE populations, and determine whether the SDS-derived HE population has a maturation defect. The more committed progeny derived from the SeV SDS iPSCs, such as the EHPs, also manifested impaired function, and reduced ability to form mature blood cells. As discussed above, hypersensitivity to Fas ligand, that characterizes SDS BM cells, might also be a contributing factor in the defect present at the stage of HE induction, and at least in part responsible for the reduced numbers of HE, EHPs, myeloid cells and granulocytes/monocytes generated by the SeV SDS iPSCs [27, 33].

A study by Sen & colleagues using SBDS-deficient erythroleukemic cells showed that during erythroid differentiation the characteristic ribosomal and global translation abnormalities become much more prominent [34]. Given that during HE induction, cells are actively differentiating, the ribosomal and global translation defects may become more prominent, and interfere with the ability of the cells to meet the increased protein production demands. This added ribosomal stress may lead to enhanced apoptosis, and therefore a reduced number of HE cells. Further studies are needed to test this hypothesis.

It is also possible that cells manifest increased oxidative stress at the HE stage. Given their loss of SBDS, cells may manifest increased ROS and accelerated apoptosis, due to an enhanced response to the Fas ligand [34, 251]. Further studies on ROS levels and their impact on differentiation at the various stages of hematopoietic development will help shed more light on the pathogenesis of SDS.

104

Differential requirements for translational activity might be an alternative explanation to the onset of the hematopoietic phenotype at the HE stage. It is possible that at the HE stage, the requirement for protein synthesis is increased, and the inability of SDS cells to augment it may affect HE generation.

SBDS also has extra-ribosomal functions that may become relevant only at the HE stage and onwards; therefore leading to the impaired developmental stage that was observed in our study. It was previously shown that SBDS co-localizes with the mitotic spindle in myeloid progenitors, suggesting that SBDS may play a role during cell division. Perhaps the myeloid cell reduction observed in our studies is a result of impaired cell division

[252]. Studies by Austin & colleagues have shown that SBDS helps stabilize the mitotic spindle, and prevents genomic instability. These authors showed that no immediate mitotic disruption is observed when SBDS is depleted, since the consequences of spindle abnormalities must accumulate over time for the phenotype to become apparent. These findings may indicate that the SDS phenotype may be as a result of a combination of defects, beginning with the reduced SBDS protein levels. Perhaps this is the reason why the hematopoietic defect is only initiated at the HE stage; because although the reduced

SBDS protein levels are also present at the iPSC and mesoderm stages, additional secondary events must occur for the disease phenotype to arise [253]. Alternatively, it is possible that specific developmental stages, such as HE, are reliant solely on the SBDS pathway to maintain spindle stability, with no compensatory mechanisms present to make up for the the reduced levels of SBDS protein.

105

4.1.5 During primitive hematopoiesis, clonogenic potential is delayed,

but not impaired

Having observed a defect in definitive hematopoiesis, we asked whether this defect was also present during primitive hematopoiesis. Although Tulpule & colleagues also observed that mesoderm induction is not affected; they did not discriminate between definitive and primitive hematopoiesis [54]. Our group is the first to specifically study mesoderm induction during both, definitive and primitive hematopoiesis, and to discover that mesoderm generation is intact during both waves of hematopoiesis. Interestingly, the induction of cells with HE potential and mature blood cells was also not impaired during primitive hematopoiesis. This observation may be explained by the differential requirements for translational activity during primitive hematopoiesis, as compared to definitive hematopoiesis. This indicates that during primitive hematopoiesis, the residual

SBDS expression level is sufficient to promote intact blood development. This might be related to a relatively constant level of protein synthesis that is required during primitive hematopoiesis, which can be sustained by the reduced numbers of functional ribosomes present. Alternatively, there may be a compensatory mechanism during the primitive wave that helps alleviate the stress that may otherwise be experienced by the SDS differentiating cells, while promoting their growth and differentiation. It is also possible that primitive hematopoiesis is not as susceptible to mitotic stressors (such as ROS and genomic instability) as definitive hematopoiesis, and is not limited by the mitotic spindle defect [253].

106 Interestingly however, we observed that when the cells were assessed for colony-forming potential, there was a delay in the ability to form CFU-GM colonies characteristic of primitive hematopoiesis in the SDS cells. BFU-E colony formation was not affected, which is perhaps a reflection of the observation that anemia is not as severe as neutropenia in SDS patients. Given that we did not observe any striking differences in the kinetics of differentiation by flow cytometry, our results indicate that despite being phenotypically similar, the control and SDS hematopoietic progenitors may be functionally different. Specifically, there may be a delay in the ability of the SDS cells to reach functional maturity. Further studies are required to better understand the mechanism governing this delay.

4.2 Limitations of study

Having discussed the results, there are several limitations that may have influenced our study. Although iPSCs are pluripotent, they may not entirely faithfully mimic the epigenetic signature of hESCs. hESCs are the ultimate model of pluripotent stem cells, that best represent embryonic development. However, due to ethical constraints, disease models of hESCs are not available. Although SDS hiPSCs are in many aspects similar to hESCs, genetic changes may have been introduced during the process of reprogramming that may have subsequently impacted our results. Reprogramming can alter differentiation potential as well as the phenotype of the reprogrammed cells [234]. Our planned follow-up studies in which we will be introducing SBDS to an SDS iPSCs line will largely address this limitation, by determining whether the identified defects are specifically due to deficiency of SBDS.

107 A second limitation is the use of retroviruses to reprogram SDS patient fibroblasts, as these viruses are integrative. These vectors may cause inadvertent genetic changes that may affect the properties of iPSCs [224]. We noticed in our study that both the RT control and SDS iPSCs were deficient in their ability to generate CFU-GEMM colonies, perhaps at least in part due to genetic changes introduced during reprogramming. To address this concern, SDS iPSC lines were also generated using non-integrative Sendai viruses.

So far, genotype-phenotype correlation has not been identified in SDS. Nevertheless, it is important to use iPSCs from different patients to eliminate variability that might be related not only to in vitro reprogramming, but also to biological differences between patients. Although our study closely investigated the hematopoietic developmental potential of two SeV SDS iPSC lines, they were both generated from the same patient.

Therefore, it is imperative that follow-up experiments be completed using iPSCs from additional patients.

We used a protocol that was highly designed to enrich differentiation towards definitive hematopoiesis. Although many groups have focused on devising differentiation protocols that separate definitive from primitive hematopoiesis, many caveats still exist. The protocol used in our experiments, which was developed by the Keller group, enriches for the definitive wave of hematopoiesis, with T-cell production as the hallmark. Despite enriching for definitive hematopoiesis, the β-globin - expressing erythroid cells obtained using this protocol also still expressed considerable levels of ε-globin [156]. This may

108 indicate that the definitive wave that we replicated in our experiments might in fact be an intermediate differentiation stage between primitive and definitive hematopoiesis.

Therefore, more defined differentiation protocols are still needed to better replicate both embryonic definitive hematopoiesis, as well as adult-type definitive hematopoiesis.

4.3 Significance

This study demonstrates the value of using patient-specific iPSCs to study human diseases and the potential of this model to answer biological questions that cannot otherwise be answered. Previously we were able to study the hematopoietic system of

IBMFSs mainly by characterizing post HSC stages. However, given that the hematopoietic defects in IBMFS patients may manifest as early as during the neonatal period, being able to study development early on, specifically during embryogenesis, is an incredibly useful tool in learning more about these syndromes. Our validated SDS iPSC disease model provides us with the opportunity to study developmental abnormalities during embryogenesis, prior to, and following the generation of HSCs.

The results of our study shed some light on the sequential steps of hematopoietic development, and helped identify the precise steps that are abnormal. Being able to identify the developmental stage at which the hematopoietic phenotype is initiated in

SDS could also help with the generation of novel potential therapeutic targets, which may help in not only rescuing the phenotype or at least slowing down disease progression, but also in lowering the risk of MDS/AML development. For example, given that the onset was determined to be at the HE stage, and that downstream developmental stages were

109 also impaired, further studies identifying molecules that may promote the growth or survival of functional HE cells and downstream hematopoietic progenitors may have a very valuable clinical application. Being able to generate therapeutics that rescue the hematopoietic phenotype at these various developmental stages may help in preventing

SDS disease progression.

Lastly, on a broader scale, our findings may be relevant for the study and understanding of other inherited bone marrow failure syndromes. Our work provides an example of how iPSCs can provide a reliable model for the study of the pathophysiology of rare diseases associated with genetic mutations, such as other IBMFSs. Whereas previously, studies of

IBMFSs were limited to the adult stages of hematopoietic development, with the advent of iPSCs that faithfully mimic genetic diseases, we can now replicate and study early embryonic development in vitro. Studies of early embryonic development in the context of hematopoiesis can help with the development of novel potential therapeutic targets for

IBMFs. Being able to separate and individually assess developmental stages during primitive and definitive hematopoiesis in SDS, and identify those that are impaired is another example of how blood development can be studied using iPSCs generated from patients with various IBMFS.

By identifying the onset of the blood defects in SDS, we can acquire a more thorough understanding of the function of SBDS during hematopoiesis, but also of normal blood development, and its requirements. Further, by studying the activity of SBDS at the onset of the hematopoietic phenotype, we may be able to identify the biochemical

110 pathways that SBDS is involved in, and identify new genes that may be implicated in bone marrow failure, which may help shed some light on other hematopoietic diseases, and also provide additional therapeutic targets.

111 Chapter V

Conclusion

The aim of this study was to generate an adequate model for the study of hematopoiesis in SDS using patient-specific iPSCs. RT and SeV SDS iPSCs were successfully generated and shown to carry the characteristic SDS phenotype, including the aberrant ribosome profile and reduced blood cell formation. The stepwise differentiation of iPSCs towards the hematopoietic lineage was studied during both the primitive and definitive waves. It was hypothesized that hematopoietic development is markedly impaired during definitive hematopoiesis, with a post-mesoderm generation onset, and this study showed that definitive hematopoiesis was indeed significantly impaired in SDS iPSCs. The study’s findings indicate that the hematopoietic phenotype is initiated at the stage of HE generation, and that all downstream cell populations that were studied were also markedly reduced. In contrast, no such impairment was present during primitive hematopoiesis, with only a delay in the generation of hematopoietic colonies in SDS iPSCs.

This study suggests that a reduced ability to generate HE during definitive hematopoiesis may be responsible for the hematopoietic phenotype in SDS patients, as all stages downstream of HE were also impaired, with an ultimate reduction in the generation of terminally differentiated blood cells, much like what is seen in SDS patients. These findings also indicate that the function of SBDS may become clinically relevant for hematopoietic development only at the HE stage, given that earlier stages of hematopoietic development, specifically mesoderm, are not affected in SDS iPSCs.

112 Given that in our experiments, the SDS iPSCs consistently gave rise to significantly smaller hematopoietic populations during definitive hematopoiesis, further studies are required to determine the pathway responsible for this defect. Also, the result that SDS iPSCs generate significantly fewer hematopoietic colonies than control iPSCs, despite plating equal numbers of cells, indicates that the generated SDS hematopoietic progenitors may be functionally different. Future studies are required to further investigate the functional differences between SDS and control hematopoietic progenitors. Further, these findings suggest that primitive hematopoiesis is not markedly impaired in SDS patients. This may indicate that primitive hematopoiesis may not be as reliant on the function of SBDS as definitive hematopoiesis is, perhaps due to the presence of compensatory mechanisms, or a relatively lower demand for protein translation, when compared to the definitive wave.

Taken together, these results help shed some light on SDS disease progression and its onset, and provide a potential therapeutic target for the treatment of the hematopoietic phenotype in SDS patients. They also have broader relevance to other common genetic diseases that are associated with bone marrow failure, as they indicate that patient- derived PSCs can provide a reliable model for the study of disease pathogenesis. Lastly, this study is an example of how the study of the stepwise differentiation of PSCs can not only provide insight into disease progression, but can also help in identifying novel therapeutic avenues that may be pursued in the treatment of IBMFSs.

113 Chapter VI

Future directions

Given the inherent variations among iPSC lines, our first step will be to replicate our results, and study them in additional SeV SDS iPSC lines generated from other patients with different mutational backgrounds.

Second, to determine whether the defect in HE induction during definitive hematopoiesis is specifically due to SBDS deficiency, we will re-introduce SBDS into the SDS iPSCs, and also knock-down SBDS in control iPSCs to assess impact of these genetic manipulations on generation of cells with HE potential as determined by flow cytometry and clonogenic assays. By re-introducing SBDS, we expect to rescue the hematopoietic phenotype, and therefore see an increase in the HE, EHP and mature blood cell populations in SDS iPSCs during definitive hematopoiesis. By knocking-down SBDS in control iPSCs we expect to recapitulate the SDS hematopoietic phenotype, as observed with our SDS iPSCs.

Third, we will determine whether the reduced blood cell formation during definitive hematopoiesis in SDS iPSCs is due to accelerated cell death, or decreased proliferation.

To do so, we will be using several approaches, including analysis of cells at various stages of differentiation using Ki-67/PI staining via flow cytometry. If cell death is enhanced during HE generation, we expect that re-introduction of SBDS will improve cell survival of SDS cells. To further evaluate the type of cell death (apoptosis or necrosis), we will use the Ki-67/PI flow method mentioned above as well as the

114 PI/annexin V flow cytometry assay. Based on the results, we will investigate whether cell death is mediated by Fas overexpression or accelerated ROS levels, which were both shown to be increased in primary SDS cells and SBDS-knockdown cells in our previous studies [27, 28, 33, 34, 251]. If we observe a marked increase in the levels of Fas and/or

ROS at the onset of the hematopoietic phenotype, we will test whether we can rescue the phenotype by adding antioxidants such as ascorbic acid (AA) and N-acetylcysteine

(NAC) or Fas pathway inhibitors.

Fourth, we will study the gene expression profiles of various genes during different stages of hematopoietic development. We aim to identify gene(s) expression alterations that are responsible for the onset of the hematopoietic defect during HE development.

Follow-up studies will include gain-of-function or loss-of-function experiments of the genes identified, to determine the link between gene expression patterns and phenotype, and to help better understand the underlying mechanism responsible for the hematopoietic phenotype in SDS.

Ultimately, our goal is to uncover novel potential therapeutic targets. Manipulation of these targets will be studied for their effectiveness in either rescuing the hematopoietic phenotype or at least delaying its progression.

115 References

1. Shwachman, H., et al., The Syndrome of Pancreatic Insufficiency and Bone Marrow Dysfunction. The Journal of pediatrics, 1964. 65: p. 645-663.

2. Bodian, M., W. Sheldon, and R. Lightwood, Congenital Hypoplasia of the Exocrine Pancreas. Acta paediatrica, 1964. 53: p. 282-293.

3. Aggett, P., N. Cavanagh, and D. Matthew, Shwachman's syndrome. A review of 21 cases. Archives of disease in …, 1980.

4. Burke, V., et al., Association of pancreatic insufficiency and chronic neutropenia in childhood. Archives of disease in childhood, 1967. 42(222): p. 147-157.

5. Pringle, E., W. Young, and E. Haworth, Syndrome of pancreatic insufficiency, blood dyscrasia and metaphyseal dysplasia. Proceedings of the Royal Society of Medicine, 1968. 61(8): p. 776-778.

6. Dror, Y., et al., Immune function in patients with Shwachman-Diamond syndrome. British journal of haematology, 2001. 114(3): p. 712-717.

7. Ginzberg, H., et al., Shwachman syndrome: phenotypic manifestations of sibling sets and isolated cases in a large patient cohort are similar. The Journal of pediatrics, 1999. 135(1): p. 81-88.

8. Graham, A., et al., Testicular fibrosis and cardiomegaly in Shwachman's syndrome. Archives of pathology & laboratory medicine, 1980. 104(5): p. 242-4.

9. Nezelof, C. and G. LeSec, Multifocal myocardial necrosis and fibrosis in pancreatic diseases of children. Pediatrics, 1979. 63(3): p. 361-368.

10. Nivelon, J., et al., Myocardial fibrosis in Shwachman's syndrome: pathogenic discussion of cardiac complications. Pédiatrie, 1978. 33(5): p. 461-9.

11. Sacrez, R., et al., Hypoplasia of exocrine pancreas. Associated myoendocardial fibrosis in 1 of 2 brothers. Annales de pédiatrie, 1969. 16(1): p. 43-8.

12. Savilahti, E. and J. Rapola, Frequent myocardial lesions in Shwachman's syndrome. Eight fatal cases among 16 Finnish patients. Acta paediatrica Scandinavica, 1984. 73(5): p. 642-651.

13. Kornfeld, S., et al., Shwachman-Diamond syndrome associated with hypogammaglobulinemia and growth hormone deficiency. The Journal of allergy and clinical immunology, 1995. 96(2): p. 247-250.

14. Raj, A., et al., Chromosome 20q deletion and progression to monosomy 7 in a patient with Shwachman-Diamond syndrome without MDS/AML. Journal of pediatric hematology/oncology, 2003. 25(6): p. 508-9.

116 15. Ho, W., et al., Prevalence of oral diseases in Shwachman-Diamond syndrome. Special care in dentistry : official publication of the American Association of Hospital Dentists, the Academy of Dentistry for the Handicapped, and the American Society for Geriatric Dentistry, 2007. 27(2): p. 52-58.

16. Kent, A., G. Murphy, and P. Milla, Psychological characteristics of children with Shwachman syndrome. Archives of disease in childhood, 1990. 65(12): p. 1349- 1352.

17. Woods, W., et al., The occurrence of leukemia in patients with the Shwachman syndrome. The Journal of pediatrics, 1981. 99(3): p. 425-8.

18. Dror, Y., et al., Malignant myeloid transformation with isochromosome 7q in Shwachman-Diamond syndrome. Leukemia, 1998. 12(10): p. 1591-1595.

19. Goobie, S., et al., Shwachman-Diamond syndrome with exocrine pancreatic dysfunction and bone marrow failure maps to the centromeric region of chromosome 7. American journal of human genetics, 2001. 68(4): p. 1048-1054.

20. Alter, B., et al., Malignancies and survival patterns in the National Cancer Institute inherited bone marrow failure syndromes cohort study. British journal of haematology, 2010. 150(2): p. 179-188.

21. Tsangaris, E., et al., Genetic analysis of inherited bone marrow failure syndromes from one prospective, comprehensive and population-based cohort and identification of novel mutations. Journal of medical genetics, 2011. 48(9): p. 618- 628.

22. Orkin, S., et al., Nathan and Oski's Hematology of Infancy and Childhood. 7th ed. 2009, Philadelphia, PA: Elsevier.

23. Boocock, G., et al., Mutations in SBDS are associated with Shwachman-Diamond syndrome. Nature genetics, 2003. 33(1): p. 97-101.

24. Burwick, N., et al., Impaired ribosomal subunit association in Shwachman- Diamond syndrome. Blood, 2012. 120(26): p. 5143-5152.

25. Ganapathi, K., et al., The human Shwachman-Diamond syndrome protein, SBDS, associates with ribosomal RNA. Blood, 2007. 110(5): p. 1458-1465.

26. Rujkijyanont, P., et al., Bone marrow cells from patients with Shwachman- Diamond syndrome abnormally express genes involved in ribosome biogenesis and RNA processing. British journal of haematology, 2009. 145(6): p. 806-815.

27. Dror, Y. and M. Freedman, Shwachman-Diamond syndrome marrow cells show abnormally increased apoptosis mediated through the Fas pathway. Blood, 2001. 97(10): p. 3011-3016.

117 28. Rujkijyanont, P., et al., SBDS-deficient cells undergo accelerated apoptosis through the Fas-pathway. Haematologica, 2008. 93(3): p. 363-371.

29. Austin, K., et al., Mitotic spindle destabilization and genomic instability in Shwachman-Diamond syndrome. The Journal of clinical investigation, 2008. 118(4): p. 1511-1518.

30. Stepanovic, V., et al., The chemotaxis defect of Shwachman-Diamond Syndrome leukocytes. Cell motility and the cytoskeleton, 2004. 57(3): p. 158-174.

31. Wessels, D., et al., The Shwachman-Bodian-Diamond syndrome gene encodes an RNA-binding protein that localizes to the pseudopod of Dictyostelium amoebae during chemotaxis. Journal of cell science, 2006. 119(Pt 2): p. 370-379.

32. Dror, Y. and M. Freedman, Shwachman-Diamond syndrome: An inherited preleukemic bone marrow failure disorder with aberrant hematopoietic progenitors and faulty marrow microenvironment. Blood, 1999. 94(9): p. 3048- 3054.

33. Watanabe, K.-I., et al., SBDS-deficiency results in specific hypersensitivity to Fas stimulation and accumulation of Fas at the plasma membrane. Apoptosis : an international journal on programmed cell death, 2009. 14(1): p. 77-89.

34. Sen, S., et al., The ribosome-related protein, SBDS, is critical for normal erythropoiesis. Blood, 2011. 118(24): p. 6407-6417.

35. Orelio, C., et al., SBDS expression and localization at the mitotic spindle in human myeloid progenitors. PloS one, 2009. 4(9).

36. Vasieva, O., Role of Shwachman-Bodian-Diamond syndrome protein in translation machinery and cell chemotaxis: a comparative genomics approach. Advances and applications in bioinformatics and chemistry : AABC, 2011. 4: p. 43-50.

37. Hashmi, S., et al., Comparative analysis of Shwachman-Diamond syndrome to other inherited bone marrow failure syndromes and genotype-phenotype correlation. Clinical genetics, 2011. 79(5): p. 448-458.

38. Mack, D., et al., Shwachman syndrome: exocrine pancreatic dysfunction and variable phenotypic expression. Gastroenterology, 1996. 111(6): p. 1593-1602.

39. Launiala, K., et al., A syndrome with pancreatic achylia and granulocytopenia. Acta paediatrica Scandinavica, 1967: p. 9.

40. Strevens, M., J. Lilleyman, and R. Williams, Shwachman's syndrome and acute lymphoblastic leukaemia. British medical journal, 1978. 2(6129): p. 18.

118 41. Danks, D., et al., Metaphyseal chondrodysplasia, neutropenia, and pancreatic insufficiency presenting with respiratory distress in the neonatal period. Archives of disease in childhood, 1976. 51(9): p. 697-702.

42. Dokal, I., et al., Adult onset of acute myeloid leukaemia (M6) in patients with Shwachman-Diamond syndrome. British journal of haematology, 1997. 99(1): p. 171-173.

43. Cipolli, M.D.O., Ciro; Delmarco, Antonella; Marchesini, Cristina; Miano, Angelo; Mastella, Gianni, Shwachman's Syndrome: Pathomorphosis and Long- term Outcome. Journal of Pediatric Gastroenterology & Nutrition, 1999. 29(3): p. 265-272.

44. Aggett, P., et al., An inherited defect of neutrophil mobility in Shwachman syndrome. The Journal of pediatrics, 1979. 94(3): p. 391-394.

45. Hudson, E. and T. Aldor, Pancreatic insufficiency and neutropenia with associated immunoglobulin deficit. Archives of internal medicine, 1970. 125(2): p. 314-316.

46. Sacchi, F., et al., Association of neutrophil and complement defects in two twins with Shwachman syndrome. Helvetica paediatrica acta, 1982. 37(2): p. 177-181.

47. Maki, M., et al., Hepatic dysfunction and dysgammaglobulinaemia in Shwachman-Diamond syndrome. Archives of Disease in Childhood, 1978. 53.

48. Smith, O., et al., Haematological abnormalities in Shwachman-Diamond syndrome. British journal of haematology, 1996. 94(2): p. 279-284.

49. Maserati, E., et al., Isochromosome (7)(q10) in Shwachman syndrome without MDS/AML and role of chromosome 7 anomalies in myeloproliferative disorders. Cancer genetics and cytogenetics, 2000. 121(2): p. 167-171.

50. Rothbaum, R., D. Williams, and C. Daugherty, Unusual surface distribution of concanavalin A reflects a cytoskeletal defect in neutrophils in Shwachman's syndrome. Lancet, 1982. 2(8302): p. 800-801.

51. Azzarà, A., et al., In vivo effectiveness of lithium on impaired neutrophil chemotaxis in Shwachman-Diamond syndrome. Acta haematologica, 1991. 85(2): p. 100-102.

52. Ruutu, P., et al., Constant defect in neutrophil locomotion but with age decreasing susceptibility to infection in Shwachman syndrome. Clinical and experimental immunology, 1984. 57(1): p. 249-255.

53. Yamaguchi, M., et al., Shwachman-Diamond syndrome is not necessary for the terminal maturation of neutrophils but is important for maintaining viability of granulocyte precursors. Experimental hematology, 2007. 35(4): p. 579-586.

119 54. Tulpule, A., et al., Pluripotent stem cell models of Shwachman-Diamond syndrome reveal a common mechanism for pancreatic and hematopoietic dysfunction. Cell stem cell, 2013. 12(6): p. 727-736.

55. Choi, J., et al., Significance of fetal hemoglobin-containing erythroblasts (F blasts) and the F blast/F cell ratio in myelodysplastic syndromes. Leukemia, 2002. 16(8): p. 1478-1483.

56. Jane, S. and J. Cunningham, Understanding fetal globin gene expression: a step towards effective HbF reactivation in haemoglobinopathies. British journal of haematology, 1998. 102(2): p. 415-422.

57. Stamatoyannopoulos, G., et al., Hb F production in stressed erythropoiesis: observations and kinetic models. Annals of the New York Academy of Sciences, 1985. 445: p. 188-197.

58. Bard, H., C. Gagnon, and K. Peri, HbF synthesis during stress erythropoiesis as determined by gamma-mRNA/non-alpha-mRNA quantification. Pediatric research, 1999. 45(5 Pt 1): p. 684-686.

59. Nezelof, C. and M. Watchi, [Lipomatous congenital hypoplasia of the exocrine pancreas in children. (2 cases and review of the literature)]. Archives françaises de pédiatrie, 1961. 18: p. 1135-1172.

60. Caselitz, J., et al., Shwachman's syndrome and leukaemia. Virchows Archiv. A, Pathological anatomy and histology, 1979. 385(1): p. 109-116.

61. Woods, W., et al., Aplastic anemia associated with the Shwachman syndrome. In vivo and in vitro observations. The American journal of pediatric hematology/oncology, 1981. 3(4): p. 347-351.

62. Gretillat, F., et al., [Leukemic transformation of Shwachman's syndrome]. Presse médicale (Paris, France : 1983), 1985. 14(1): p. 45.

63. Okcu, F., W. Roberts, and K. Chan, Bone marrow transplantation in Shwachman- Diamond syndrome: report of two cases and review of the literature. Bone marrow transplantation, 1998. 21(8): p. 849-851.

64. Toiviainen-Salo, S., et al., Magnetic resonance imaging findings of the pancreas in patients with Shwachman-Diamond syndrome and mutations in the SBDS gene. The Journal of pediatrics, 2008. 152(3): p. 434-436.

65. Ip, W., et al., Serum pancreatic enzymes define the pancreatic phenotype in patients with Shwachman-Diamond syndrome. The Journal of pediatrics, 2002. 141(2): p. 259-265.

120 66. Myers, K., et al., Endocrine evaluation of children with and without Shwachman- Bodian-Diamond syndrome gene mutations and Shwachman-Diamond syndrome. The Journal of pediatrics, 2013. 162(6): p. 1235.

67. Mäkitie, O., et al., Skeletal phenotype in patients with Shwachman-Diamond syndrome and mutations in SBDS. Clinical genetics, 2004. 65(2): p. 101-112.

68. Taybi, H., A. Mitchell, and G. Friedman, Metaphyseal dysostosis and the associated syndrome of pancreatic insufficiency and blood disorders. Radiology, 1969. 93(3): p. 563-571.

69. Labrune, M., et al., [Shwachman's syndrome with neonatal thoracic manifestations]. Archives françaises de pédiatrie, 1984. 41(8): p. 561-563.

70. Dror, Y., et al., Duplication of distal thumb phalanx in Shwachman-Diamond syndrome. American journal of medical genetics, 1998. 78(1): p. 67-69.

71. Toiviainen-Salo, S., et al., Shwachman-Diamond syndrome is associated with low-turnover osteoporosis. Bone, 2007. 41(6): p. 965-972.

72. Le Gloan, L., N. Blin, and J.-M. Langlard, Atrioventricular septal defect in a case of Shwachman-Diamond syndrome. Cardiology in the young, 2013: p. 1-3.

73. Hauet, Q., et al., Cardiomyopathies and congenital heart diseases in Shwachman- Diamond syndrome: a national survey. International journal of cardiology, 2013. 167(3): p. 1048-1050.

74. Kopel, L., P. Gutierrez, and S. Lage, Dilated cardiomyopathy in a case of Shwachman-Diamond syndrome. Cardiology in the young, 2011. 21(5): p. 588- 590.

75. Donadieu, J., et al., Analysis of risk factors for myelodysplasias, leukemias and death from infection among patients with congenital neutropenia. Experience of the French Severe Chronic Neutropenia Study Group. Haematologica, 2005. 90(1): p. 45-53.

76. Dror, Y., Shwachman-Diamond syndrome. Pediatric blood & cancer, 2005. 45(7): p. 892-901.

77. Singh, S., et al., Breast cancer in a case of Shwachman Diamond syndrome. Pediatric blood & cancer, 2012. 59(5): p. 945-946.

78. Dhanraj, S., et al., Molecular characteristics of a pancreatic adenocarcinoma associated with Shwachman-Diamond syndrome. Pediatric blood & cancer, 2013. 60(5): p. 754-760.

121 79. Sack, J.E., L. Kuchnir, and M.-F.F. Demierre, Dermatofibrosarcoma protuberans arising in the context of Shwachman-Diamond syndrome. Pediatric dermatology, 2010. 28(5): p. 568-569.

80. Ganapathi, K. and A. Shimamura, Ribosomal dysfunction and inherited marrow failure. British journal of haematology, 2008. 141(3): p. 376-387.

81. Ruggero, D. and P. Pandolfi, Does the ribosome translate cancer? Nature reviews. Cancer, 2003. 3(3): p. 179-192.

82. Grisendi, S., et al., Nucleophosmin and cancer. Nature reviews. Cancer, 2006. 6(7): p. 493-505.

83. Amsterdam, A., et al., Many ribosomal protein genes are cancer genes in zebrafish. PLoS biology, 2004. 2(5).

84. Lai, K., et al., Many ribosomal protein mutations are associated with growth impairment and tumor predisposition in zebrafish. Developmental dynamics : an official publication of the American Association of Anatomists, 2009. 238(1): p. 76-85.

85. Ebert, B., et al., Identification of RPS14 as a 5q- syndrome gene by RNA interference screen. Nature, 2008. 451(7176): p. 335-339.

86. Dror, Y. and M. Freedman, Shwachman-diamond syndrome. British journal of haematology, 2002. 118(3): p. 701-713.

87. Smith, O., Shwachman-Diamond syndrome. Seminars in hematology, 2002. 39(2): p. 95-102.

88. Donadieu, J., et al., Hematopoietic stem cell transplantation for Shwachman- Diamond syndrome: experience of the French neutropenia registry. Bone marrow transplantation, 2005. 36(9): p. 787-792.

89. Cesaro, S., et al., Haematopoietic stem cell transplantation for Shwachman- Diamond disease: a study from the European Group for blood and marrow transplantation. British journal of haematology, 2005. 131(2): p. 231-236.

90. Tsai, P., et al., Fatal cyclophosphamide-induced congestive heart failure in a 10- year-old boy with Shwachman-Diamond syndrome and severe bone marrow failure treated with allogeneic bone marrow transplantation. The American journal of pediatric hematology/oncology, 1990. 12(4): p. 472-476.

91. Davies, S., et al., Unrelated donor bone marrow transplantation for children and adolescents with aplastic anaemia or myelodysplasia. British journal of haematology, 1997. 96(4): p. 749-756.

122 92. Grill, J., et al., [Treatment of neutropenia in Shwachman's syndrome with granulocyte growth factor (G-CSF)]. Archives françaises de pédiatrie, 1993. 50(4): p. 331-333.

93. van der Sande, F. and H. Hillen, Correction of neutropenia following treatment with granulocyte colony-stimulating factor results in a decreased frequency of infections in Shwachman's syndrome. The Netherlands journal of medicine, 1996. 48(3): p. 92-95.

94. Dale, D., et al., The Severe Chronic Neutropenia International Registry: 10-Year Follow-up Report. Supportive cancer therapy, 2006. 3(4): p. 220-231.

95. Rosenberg, P., et al., The incidence of leukemia and mortality from sepsis in patients with severe congenital neutropenia receiving long-term G-CSF therapy. Blood, 2006. 107(12): p. 4628-4635.

96. Freedman, M. and B. Alter, Malignant myeloid transformation in congenital forms of neutropenia. The Israel Medical Association journal : IMAJ, 2002. 4(11): p. 1011-1014.

97. Freedman, M., et al., Myelodysplasia syndrome and acute myeloid leukemia in patients with congenital neutropenia receiving G-CSF therapy. Blood, 2000.

98. Barrios, N. and D. Kirkpatrick, Successful cyclosporin A treatment of aplastic anaemia in Shwachman-Diamond syndrome. British journal of haematology, 1990. 74(4): p. 540-541.

99. Burroughs, L., A. Woolfrey, and A. Shimamura, Shwachman-Diamond syndrome: a review of the clinical presentation, molecular pathogenesis, diagnosis, and treatment. Hematology/oncology clinics of North America, 2009. 23(2): p. 233-248.

100. de Oliveira, J., et al., Structure, dynamics, and RNA interaction analysis of the human SBDS protein. Journal of molecular biology, 2010. 396(4): p. 1053-1069.

101. Shammas, C., et al., Structural and mutational analysis of the SBDS protein family. Insight into the leukemia-associated Shwachman-Diamond Syndrome. The Journal of biological chemistry, 2005. 280(19): p. 19221-19229.

102. Austin, K., R. Leary, and A. Shimamura, The Shwachman-Diamond SBDS protein localizes to the nucleolus. Blood, 2005. 106(4): p. 1253-1258.

103. Woloszynek, J., et al., Mutations of the SBDS gene are present in most patients with Shwachman-Diamond syndrome. Blood, 2004. 104(12): p. 3588-3590.

104. Wong, T., et al., SBDS protein expression patterns in the bone marrow. Pediatric blood & cancer, 2010. 55(3): p. 546-549.

123 105. Wong, C., et al., Defective ribosome assembly in Shwachman-Diamond syndrome. Blood, 2011. 118(16): p. 4305-4312.

106. Finch, A.J., et al., Uncoupling of GTP hydrolysis from eIF6 release on the ribosome causes Shwachman-Diamond syndrome. Genes & development, 2011. 25(9): p. 917-929.

107. Nazar, R., Ribosomal RNA processing and ribosome biogenesis in eukaryotes. IUBMB life, 2004. 56(8): p. 457-465.

108. Mukhopadhyay, S., et al., Autophagy and apoptosis: where do they meet? Apoptosis : an international journal on programmed cell death, 2014. 19(4): p. 555-566.

109. Cmejlova, J., et al., Translational efficiency in patients with Diamond-Blackfan anemia. Haematologica, 2006. 91(11): p. 1456-1464.

110. Garçon, L., et al., Ribosomal and hematopoietic defects in induced pluripotent stem cells derived from Diamond Blackfan anemia patients. Blood, 2013. 122(6): p. 912-921.

111. Rawls, A., et al., Lentiviral-mediated RNAi inhibition of Sbds in murine hematopoietic progenitors impairs their hematopoietic potential. Blood, 2007. 110(7): p. 2414-2422.

112. Zhang, S., et al., Loss of the mouse ortholog of the shwachman-diamond syndrome gene (Sbds) results in early embryonic lethality. Molecular and cellular biology, 2006. 26(17): p. 6656-6663.

113. Tourlakis, M., et al., Deficiency of Sbds in the mouse pancreas leads to features of Shwachman-Diamond syndrome, with loss of zymogen granules. Gastroenterology, 2012. 143(2): p. 481-492.

114. Venkatasubramani, N. and A. Mayer, A zebrafish model for the Shwachman- Diamond syndrome (SDS). Pediatric research, 2008. 63(4): p. 348-352.

115. Odom, D., et al., Tissue-specific transcriptional regulation has diverged significantly between human and mouse. Nature genetics, 2007. 39(6): p. 730- 732.

116. Tiscornia, G., E. Vivas, and J. Izpisúa Belmonte, Diseases in a dish: modeling human genetic disorders using induced pluripotent cells. Nature medicine, 2011. 17(12): p. 1570-1576.

117. Ming, C., et al., Inactivation of Fac in mice produces inducible chromosomal instability and reduced fertility reminiscent of Fanconi anaemia. Nature Genetics, 1996. 12.

124 118. McGowan, K.A. and P.J. Mason, Animal models of Diamond Blackfan anemia. Seminars in hematology, 2011. 48(2): p. 106-116.

119. Matsson, H., et al., Targeted disruption of the ribosomal protein S19 gene is lethal prior to implantation. Molecular and cellular biology, 2004. 24(9): p. 4032- 4037.

120. Matsson, H., et al., Erythropoiesis in the Rps19 disrupted mouse: Analysis of erythropoietin response and biochemical markers for Diamond-Blackfan anemia. Blood cells, molecules & diseases, 2005. 36(2): p. 259-264.

121. Park, I.-H., et al., Disease-specific induced pluripotent stem cells. Cell, 2008. 134(5): p. 877-886.

122. Chadwick, K., et al., Cytokines and BMP-4 promote hematopoietic differentiation of human embryonic stem cells. Blood, 2003. 102(3): p. 906-915.

123. Orkin, S. and L. Zon, Hematopoiesis: an evolving paradigm for stem cell biology. Cell, 2008. 132(4): p. 631-644.

124. Dzierzak, E. and N. Speck, Of lineage and legacy: the development of mammalian hematopoietic stem cells. Nature immunology, 2008. 9(2): p. 129-136.

125. Doulatov, S., et al., Hematopoiesis: a human perspective. Cell stem cell, 2012. 10(2): p. 120-136.

126. Tavian, M., et al., Embryonic origin of human hematopoiesis. The International journal of developmental biology, 2010. 54(6-7): p. 1061-1065.

127. Luckett, W., Origin and differentiation of the yolk sac and extraembryonic mesoderm in presomite human and rhesus monkey embryos. The American journal of anatomy, 1978. 152(1): p. 59-97.

128. Lensch, M. and G. Daley, Origins of mammalian hematopoiesis: in vivo paradigms and in vitro models. Current topics in developmental biology, 2004. 60: p. 127-196.

129. Kinder, S., et al., The orderly allocation of mesodermal cells to the extraembryonic structures and the anteroposterior axis during gastrulation of the mouse embryo. Development (Cambridge, England), 1999. 126(21): p. 4691- 4701.

130. Palis, J. and M. Yoder, Yolk-sac hematopoiesis: the first blood cells of mouse and man. Experimental hematology, 2001. 29(8): p. 927-936.

131. Pick, M., et al., Differentiation of human embryonic stem cells in serum-free medium reveals distinct roles for bone morphogenetic protein 4, vascular

125 endothelial growth factor, stem cell factor, and fibroblast growth factor 2 in hematopoiesis. Stem cells (Dayton, Ohio), 2007. 25(9): p. 2206-2214.

132. Ng, E., et al., Forced aggregation of defined numbers of human embryonic stem cells into embryoid bodies fosters robust, reproducible hematopoietic differentiation. Blood, 2005. 106(5): p. 1601-1603.

133. Kennedy, M., et al., Development of the hemangioblast defines the onset of hematopoiesis in human ES cell differentiation cultures. Blood, 2007. 109(7): p. 2679-2687.

134. Davis, R., et al., Targeting a GFP reporter gene to the MIXL1 locus of human embryonic stem cells identifies human primitive streak-like cells and enables isolation of primitive hematopoietic precursors. Blood, 2008. 111(4): p. 1876- 1884.

135. Zambidis, E., et al., Hematopoietic differentiation of human embryonic stem cells progresses through sequential hematoendothelial, primitive, and definitive stages resembling human yolk sac development. Blood, 2005. 106(3): p. 860-870.

136. Palis, J., et al., Primitive erythropoiesis in the mammalian embryo. The International journal of developmental biology, 2010. 54(6-7): p. 1011-1018.

137. Lancrin, C., et al., The haemangioblast generates haematopoietic cells through a haemogenic endothelium stage. Nature, 2009. 457(7231): p. 892-895.

138. Sturgeon, C.M., et al., Wnt signaling controls the specification of definitive and primitive hematopoiesis from human pluripotent stem cells. Nature biotechnology, 2014.

139. Tam, P. and R. Behringer, Mouse gastrulation: the formation of a mammalian body plan. Mechanisms of development, 1997. 68(1-2): p. 3-25.

140. Tavian, M., et al., The human embryo, but not its yolk sac, generates lympho- myeloid stem cells: mapping multipotent hematopoietic cell fate in intraembryonic mesoderm. Immunity, 2001. 15(3): p. 487-495.

141. Peault, B. and M. Tavian, Hematopoietic stem cell emergence in the human embryo and fetus. Annals of the New York Academy of Sciences, 2003. 996: p. 132-140.

142. Ivanovs, A., et al., Highly potent human hematopoietic stem cells first emerge in the intraembryonic aorta-gonad-mesonephros region. The Journal of experimental medicine, 2011. 208(12): p. 2417-2427.

143. Choi, K.-D., et al., Identification of the hemogenic endothelial progenitor and its direct precursor in human pluripotent stem cell differentiation cultures. Cell reports, 2012. 2(3): p. 553-567.

126 144. Rafii, S., et al., Human ESC-derived hemogenic endothelial cells undergo distinct waves of endothelial to hematopoietic transition. Blood, 2013. 121(5): p. 770- 780.

145. Galloway, J. and L. Zon, Ontogeny of hematopoiesis: examining the emergence of hematopoietic cells in the vertebrate embryo. Current topics in developmental biology, 2003. 53: p. 139-158.

146. Migliaccio, G., et al., Human embryonic hemopoiesis. Kinetics of progenitors and precursors underlying the yolk sac----liver transition. The Journal of clinical investigation, 1986. 78(1): p. 51-60.

147. Thomas, D. and J. Yoffey, Human Foetal Haematopoiesis. II. Hepatic Haematopoiesis in the Human Foetus. . British journal of haematology, 1964. 10: p. 193-197.

148. Charbord, P., et al., Early ontogeny of the human marrow from long bones: an immunohistochemical study of hematopoiesis and its microenvironment. Blood, 1996. 87(10): p. 4109-4119.

149. Williams, D., H. Xu, and J. Cancelas, Children are not little adults: just ask their hematopoietic stem cells. The Journal of clinical investigation, 2006. 116(10): p. 2593-2596.

150. Lansdorp, P., Developmental changes in the function of hematopoietic stem cells. Experimental hematology, 1995. 23(3): p. 187-191.

151. Wang, J., M. Doedens, and J. Dick, Primitive human hematopoietic cells are enriched in cord blood compared with adult bone marrow or mobilized peripheral blood as measured by the quantitative in vivo SCID-repopulating cell assay. Blood, 1997. 89(11): p. 3919-3924.

152. Notta, F., et al., Isolation of single human hematopoietic stem cells capable of long-term multilineage engraftment. Science (New York, N.Y.), 2011. 333(6039): p. 218-221.

153. McKenzie, J., et al., Human short-term repopulating stem cells are efficiently detected following intrafemoral transplantation into NOD/SCID recipients depleted of CD122+ cells. Blood, 2005. 106(4): p. 1259-1261.

154. Doulatov, S., et al., Revised map of the human progenitor hierarchy shows the origin of macrophages and dendritic cells in early lymphoid development. Nature immunology, 2010. 11(7): p. 585-593.

155. Cerdan, C., et al., Activin A promotes hematopoietic fated mesoderm development through upregulation of brachyury in human embryonic stem cells. Stem cells and development, 2012. 21(15): p. 2866-2877.

127 156. Kennedy, M., et al., T lymphocyte potential marks the emergence of definitive hematopoietic progenitors in human pluripotent stem cell differentiation cultures. Cell reports, 2012. 2(6): p. 1722-1735.

157. Grigoriadis, A., et al., Directed differentiation of hematopoietic precursors and functional osteoclasts from human ES and iPS cells. Blood, 2010. 115(14): p. 2769-2776.

158. Chang, K.-H., et al., Definitive-like erythroid cells derived from human embryonic stem cells coexpress high levels of embryonic and fetal globins with little or no adult globin. Blood, 2006. 108(5): p. 1515-1523.

159. Tavian, M., M. Hallais, and B. Péault, Emergence of intraembryonic hematopoietic precursors in the pre-liver human embryo. Development (Cambridge, England), 1999. 126(4): p. 793-803.

160. Zhang, P., et al., Short-term BMP-4 treatment initiates mesoderm induction in human embryonic stem cells. Blood, 2008. 111(4): p. 1933-1941.

161. Woll, P., et al., Wnt signaling promotes hematoendothelial cell development from human embryonic stem cells. Blood, 2008. 111(1): p. 122-131.

162. Vijayaragavan, K., et al., Noncanonical Wnt signaling orchestrates early developmental events toward hematopoietic cell fate from human embryonic stem cells. Cell stem cell, 2009. 4(3): p. 248-262.

163. Oberlin, E., et al., Blood-forming potential of vascular endothelium in the human embryo. Development (Cambridge, England), 2002. 129(17): p. 4147-4157.

164. Nakajima-Takagi, Y., et al., Role of SOX17 in hematopoietic development from human embryonic stem cells. Blood, 2013. 121(3): p. 447-458.

165. Wang, C., et al., TGFβ inhibition enhances the generation of hematopoietic progenitors from human ES cell-derived hemogenic endothelial cells using a stepwise strategy. Cell research, 2012. 22(1): p. 194-207.

166. Chen, M., et al., Runx1 is required for the endothelial to haematopoietic cell transition but not thereafter. Nature, 2009. 457(7231): p. 887-891.

167. Civin, C., L. Strauss, and C. Brovall…, Antigenic analysis of hematopoiesis. III. A hematopoietic progenitor cell surface antigen defined by a monoclonal antibody raised against KG-1a cells. The Journal of …, 1984.

168. Baum, C., et al., Isolation of a candidate human hematopoietic stem-cell population. Proceedings of the National Academy of Sciences of the United States of America, 1992. 89(7): p. 2804-2808.

128 169. Bhatia, M., et al., Purification of primitive human hematopoietic cells capable of repopulating immune-deficient mice. Proceedings of the National Academy of Sciences of the United States of America, 1997. 94(10): p. 5320-5325.

170. Lansdorp, P., H. Sutherland, and C. Eaves, Selective expression of CD45 isoforms on functional subpopulations of CD34+ hemopoietic cells from human bone marrow. The Journal of experimental medicine, 1990. 172(1): p. 363-366.

171. Hock, H., et al., Gfi-1 restricts proliferation and preserves functional integrity of haematopoietic stem cells. Nature, 2004. 431(7011): p. 1002-1007.

172. Zeng, H., et al., Transcription factor Gfi1 regulates self-renewal and engraftment of hematopoietic stem cells. The EMBO journal, 2004. 23(20): p. 4116-4125.

173. Rizo, A., et al., Repression of BMI1 in normal and leukemic human CD34(+) cells impairs self-renewal and induces apoptosis. Blood, 2009. 114(8): p. 1498- 1505.

174. Amsellem, S., et al., Ex vivo expansion of human hematopoietic stem cells by direct delivery of the HOXB4 homeoprotein. Nature medicine, 2003. 9(11): p. 1423-1427.

175. Shojaei, F., et al., Hierarchical and ontogenic positions serve to define the molecular basis of human hematopoietic stem cell behavior. Developmental cell, 2005. 8(5): p. 651-663.

176. Boitano, A., et al., Aryl hydrocarbon receptor antagonists promote the expansion of human hematopoietic stem cells. Science (New York, N.Y.), 2010. 329(5997): p. 1345-1348.

177. Huang, J., et al., Maintenance of hematopoietic stem cells through regulation of Wnt and mTOR pathways. Nature medicine, 2012. 18(12): p. 1778-1785.

178. Bhatia, M., et al., Bone morphogenetic proteins regulate the developmental program of human hematopoietic stem cells. The Journal of experimental medicine, 1999. 189(7): p. 1139-1148.

179. Benveniste, P., et al., Notch signals are required for in vitro but not in vivo maintenance of human hematopoietic stem cells and delay the appearance of multipotent progenitors. Blood, 2013.

180. Majeti, R., C. Park, and I. Weissman, Identification of a hierarchy of multipotent hematopoietic progenitors in human cord blood. Cell stem cell, 2007. 1(6): p. 635-645.

181. Lagresle-Peyrou, C., et al., Human adenylate kinase 2 deficiency causes a profound hematopoietic defect associated with sensorineural deafness. Nature genetics, 2009. 41(1): p. 106-111.

129 182. Novershtern, N., et al., Densely interconnected transcriptional circuits control cell states in human hematopoiesis. Cell, 2011. 144(2): p. 296-309.

183. Owen, C., M. Barnett, and J. Fitzgibbon, Familial myelodysplasia and acute myeloid leukaemia--a review. British journal of haematology, 2008. 140(2): p. 123-132.

184. Person, R., et al., Mutations in proto-oncogene GFI1 cause human neutropenia and target ELA2. Nature genetics, 2003. 34(3): p. 308-312.

185. Schwieger, M., et al., A dominant-negative mutant of C/EBP , associated with acute myeloid leukemias, inhibits differentiation of myeloid and erythroid progenitors of man but not mouse. Blood, 2004. 103(7): p. 2744-2752.

186. Hambleton, S., et al., IRF8 mutations and human dendritic-cell immunodeficiency. The New England journal of medicine, 2011. 365(2): p. 127- 138.

187. Becker, A.M., et al., IRF-8 extinguishes neutrophil production and promotes dendritic cell lineage commitment in both myeloid and lymphoid mouse progenitors. Blood, 2012. 119(9): p. 2003-2012.

188. Sankaran, V., et al., Developmental and species-divergent globin switching are driven by BCL11A. Nature, 2009. 460(7259): p. 1093-1097.

189. Chou, S.T., et al., Graded repression of PU.1/Sfpi1 gene transcription by GATA factors regulates hematopoietic cell fate. Blood, 2009. 114(5): p. 983-994.

190. Dickinson, R., et al., Exome sequencing identifies GATA-2 mutation as the cause of dendritic cell, monocyte, B and NK lymphoid deficiency. Blood, 2011. 118(10): p. 2656-2658.

191. Hsu, A., et al., Mutations in GATA2 are associated with the autosomal dominant and sporadic monocytopenia and mycobacterial infection (MonoMAC) syndrome. Blood, 2011. 118(10): p. 2653-2655.

192. Iwasaki, H., et al., The order of expression of transcription factors directs hierarchical specification of hematopoietic lineages. Genes & development, 2006. 20(21): p. 3010-3021.

193. Mullighan, C., et al., Deletion of IKZF1 and prognosis in acute lymphoblastic leukemia. The New England journal of medicine, 2009. 360(5): p. 470-480.

194. Weng, A., et al., Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science (New York, N.Y.), 2004. 306(5694): p. 269-271.

130 195. Dale, D., et al., A randomized controlled phase III trial of recombinant human granulocyte colony-stimulating factor (filgrastim) for treatment of severe chronic neutropenia. Blood, 1993. 81(10): p. 2496-2502.

196. VandePol, C. and M. Garnick, Clinical applications of recombinant macrophage- colony stimulating factor (rhM-CSF). Biotechnology therapeutics, 1991. 2(3-4): p. 231-239.

197. Clutterbuck, E., et al., Recombinant human interleukin 5 is an eosinophil differentiation factor but has no activity in standard human B cell growth factor assays. European journal of immunology, 1987. 17(12): p. 1743-1750.

198. Debili, N., et al., The Mpl-ligand or thrombopoietin or megakaryocyte growth and differentiative factor has both direct proliferative and differentiative activities on human megakaryocyte progenitors. Blood, 1995. 86(7): p. 2516-2525.

199. Leary, A., et al., Recombinant gibbon interleukin 3 supports formation of human multilineage colonies and blast cell colonies in culture: comparison with recombinant human granulocyte-macrophage colony-stimulating factor. Blood, 1987. 70(5): p. 1343-1348.

200. Kelly, K., et al., Development of dendritic cells in culture from human and murine thymic precursor cells. Cellular and molecular biology (Noisy-le-Grand, France), 2001. 47(1): p. 43-54.

201. Sonoda, Y., et al., Human interleukin-9 supports formation of a subpopulation of erythroid bursts that are responsive to interleukin-3. American journal of hematology, 1992. 41(2): p. 84-91.

202. Toksoz, D., et al., Support of human hematopoiesis in long-term bone marrow cultures by murine stromal cells selectively expressing the membrane-bound and secreted forms of the human homolog of the steel gene product, stem cell factor. Proceedings of the National Academy of Sciences of the United States of America, 1992. 89(16): p. 7350-7354.

203. Rappold, I., et al., Functional and phenotypic characterization of cord blood and bone marrow subsets expressing FLT3 (CD135) receptor tyrosine kinase. Blood, 1997. 90(1): p. 111-125.

204. Sitnicka, E., et al., Human CD34+ hematopoietic stem cells capable of multilineage engrafting NOD/SCID mice express flt3: distinct flt3 and c-kit expression and response patterns on mouse and candidate human hematopoietic stem cells. Blood, 2003. 102(3): p. 881-886.

205. Leary, A., et al., Synergism between interleukin-6 and interleukin-3 in supporting proliferation of human hematopoietic stem cells: comparison with interleukin-1 alpha. Blood, 1988. 71(6): p. 1759-1763.

131 206. Burstein, S., et al., Leukemia inhibitory factor and interleukin-11 promote maturation of murine and human megakaryocytes in vitro. Journal of cellular physiology, 1992. 153(2): p. 305-312.

207. Thomson, J., et al., Embryonic stem cell lines derived from human blastocysts. Science (New York, N.Y.), 1998. 282(5391): p. 1145-1147.

208. Chun, Y., K. Byun, and B. Lee, Induced pluripotent stem cells and personalized medicine: current progress and future perspectives. Anatomy & cell biology, 2011. 44(4): p. 245-255.

209. Takahashi, K., et al., Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell, 2007. 131(5): p. 861-872.

210. Lim, W., et al., Hematopoietic cell differentiation from embryonic and induced pluripotent stem cells. Stem cell research & therapy, 2013. 4(3): p. 71.

211. Lengerke, C., et al., Hematopoietic development from human induced pluripotent stem cells. Annals of the New York Academy of Sciences, 2009. 1176: p. 219- 227.

212. Choi, K.-D., et al., Hematopoietic and endothelial differentiation of human induced pluripotent stem cells. Stem cells (Dayton, Ohio), 2009. 27(3): p. 559- 567.

213. Ledran, M., et al., Efficient hematopoietic differentiation of human embryonic stem cells on stromal cells derived from hematopoietic niches. Cell stem cell, 2008. 3(1): p. 85-98.

214. Weisel, K., et al., Stromal cell lines from the aorta-gonado-mesonephros region are potent supporters of murine and human hematopoiesis. Experimental hematology, 2006. 34(11): p. 1505-1516.

215. Salvagiotto, G., et al., A defined, feeder-free, serum-free system to generate in vitro hematopoietic progenitors and differentiated blood cells from hESCs and hiPSCs. PloS one, 2011. 6(3).

216. Niwa, A., et al., A novel serum-free monolayer culture for orderly hematopoietic differentiation of human pluripotent cells via mesodermal progenitors. PloS one, 2011. 6(7).

217. Itskovitz-Eldor, J., et al., Differentiation of human embryonic stem cells into embryoid bodies compromising the three embryonic germ layers. Molecular medicine (Cambridge, Mass.), 2000. 6(2): p. 88-95.

218. Evseenko, D., et al., Mapping the first stages of mesoderm commitment during differentiation of human embryonic stem cells. Proceedings of the National

132 Academy of Sciences of the United States of America, 2010. 107(31): p. 13742- 13747.

219. Kelley, J. and G. Daley, Hematopoietic defects and iPSC disease modeling: Lessons learned. Immunology letters, 2013. 155(1-2): p. 18-20.

220. Batista, L., et al., Telomere shortening and loss of self-renewal in dyskeratosis congenita induced pluripotent stem cells. Nature, 2011. 474(7351): p. 399-402.

221. Müller, L., et al., Induced pluripotent stem cells as a tool for gaining new insights into Fanconi anemia. Cell cycle (Georgetown, Tex.), 2012. 11(16): p. 2985-2990.

222. Robinton, D. and G. Daley, The promise of induced pluripotent stem cells in research and therapy. Nature, 2012. 481(7381): p. 295-305.

223. Nakagawa, M., et al., Generation of induced pluripotent stem cells without Myc from mouse and human fibroblasts. Nature biotechnology, 2008. 26(1): p. 101- 106.

224. Yu, J., et al., Human induced pluripotent stem cells free of vector and transgene sequences. Science (New York, N.Y.), 2009. 324(5928): p. 797-801.

225. Sommer, C., et al., Induced pluripotent stem cell generation using a single lentiviral stem cell cassette. Stem cells (Dayton, Ohio), 2009. 27(3): p. 543-549.

226. Woltjen, K., et al., piggyBac transposition reprograms fibroblasts to induced pluripotent stem cells. Nature, 2009. 458(7239): p. 766-770.

227. Warren, L., et al., Feeder-free derivation of human induced pluripotent stem cells with messenger RNA. Scientific reports, 2012. 2: p. 657.

228. Warren, L., et al., Highly efficient reprogramming to pluripotency and directed differentiation of human cells with synthetic modified mRNA. Cell stem cell, 2010. 7(5): p. 618-630.

229. Si-Tayeb, K., et al., Generation of human induced pluripotent stem cells by simple transient transfection of plasmid DNA encoding reprogramming factors. BMC developmental biology, 2010. 10: p. 81.

230. Zhou, W. and C. Freed, Adenoviral gene delivery can reprogram human fibroblasts to induced pluripotent stem cells. Stem cells (Dayton, Ohio), 2009. 27(11): p. 2667-2674.

231. Miyoshi, N., et al., Reprogramming of mouse and human cells to pluripotency using mature microRNAs. Cell stem cell, 2011. 8(6): p. 633-638.

232. Kim, D., et al., Generation of human induced pluripotent stem cells by direct delivery of reprogramming proteins. Cell stem cell, 2009. 4(6): p. 472-476.

133 233. Fusaki, N., et al., Efficient induction of transgene-free human pluripotent stem cells using a vector based on Sendai virus, an RNA virus that does not integrate into the host genome. Proceedings of the Japan Academy. Series B, Physical and biological sciences, 2009. 85(8): p. 348-362.

234. Marchetto, M., et al., Transcriptional signature and memory retention of human- induced pluripotent stem cells. PloS one, 2009. 4(9).

235. Bar-Nur, O., et al., Epigenetic memory and preferential lineage-specific differentiation in induced pluripotent stem cells derived from human pancreatic islet beta cells. Cell stem cell, 2011. 9(1): p. 17-23.

236. Panopoulos, A., et al., Rapid and highly efficient generation of induced pluripotent stem cells from human umbilical vein endothelial cells. PloS one, 2011. 6(5).

237. Sugii, S., et al., Human and mouse adipose-derived cells support feeder- independent induction of pluripotent stem cells. Proceedings of the National Academy of Sciences of the United States of America, 2010. 107(8): p. 3558- 3563.

238. González, F., S. Boué, and J. Izpisúa Belmonte, Methods for making induced pluripotent stem cells: reprogramming à la carte. Nature reviews. Genetics, 2011. 12(4): p. 231-242.

239. Dimos, J., et al., Induced pluripotent stem cells generated from patients with ALS can be differentiated into motor neurons. Science (New York, N.Y.), 2008. 321(5893): p. 1218-1221.

240. Black, L., et al., Shwachman-Diamond syndrome presenting in a premature infant as pancytopenia. Pediatric blood & cancer, 2008. 51(1): p. 123-124.

241. Kuijpers, T.W., Congenital Aplastic Anemia Caused by Mutations in the SBDS Gene: A Rare Presentation of Shwachman-Diamond Syndrome. Pediatrics, 2004. 114.

242. Mirjana, T.-G., et al., A case of Shwachman-Diamond syndrome in a male neonate. Acta Paediatrica, 2007. 95.

243. Lister, R., et al., Hotspots of aberrant epigenomic reprogramming in human induced pluripotent stem cells. Nature, 2011. 471(7336): p. 68-73.

244. Robinton, D.A. and G.Q. Daley, The promise of induced pluripotent stem cells in research and therapy. Nature, 2012. 481(7381): p. 295-305.

245. Osafune, K., et al., Marked differences in differentiation propensity among human embryonic stem cell lines. Nature biotechnology, 2008. 26(3): p. 313-315.

134 246. Vodyanik, M.A., J.A. Thomson, and I.I. Slukvin, Leukosialin (CD43) defines hematopoietic progenitors in human embryonic stem cell differentiation cultures. Blood, 2006. 108(6): p. 2095-2105.

247. Muller, L.U., et al., Overcoming reprogramming resistance of Fanconi anemia cells. Blood, 2012. 119(23): p. 5449-5457.

248. Young, M.A., et al., Background mutations in parental cells account for most of the genetic heterogeneity of induced pluripotent stem cells. Cell stem cell, 2012. 10(5): p. 570-582.

249. Macarthur, C.C., et al., Generation of human-induced pluripotent stem cells by a nonintegrating RNA Sendai virus vector in feeder-free or xeno-free conditions. Stem cells international, 2011. 2012: p. 564612.

250. Ganapathi, K.A., et al., The human Shwachman-Diamond syndrome protein, SBDS, associates with ribosomal RNA. Blood, 2007. 110(5): p. 1458-1465.

251. Ambekar, C., et al., SBDS-deficiency results in deregulation of reactive oxygen species leading to increased cell death and decreased cell growth. Pediatric blood & cancer, 2010. 55(6): p. 1138-1144.

252. Orelio, C., et al., SBDS expression and localization at the mitotic spindle in human myeloid progenitors. PloS one, 2008. 4(9).

253. Austin, K.M., et al., Mitotic spindle destabilization and genomic instability in Shwachman-Diamond syndrome. The Journal of clinical investigation, 2008. 118(4): p. 1511-1518.

135