The Role of Histone Demethylases JMJD1C in the Pathopyhsiology Of

Aus dem Department Innere Medizin Klinik für Innere Medizin I: Hämatologie, Onkologie und Stammzellentransplantation des Universitätsklinikums Freiburg im Breisgau

The Role of Histone Demathylase JMJD1C in the Pathophysiology of Myeloproliferative Neoplasms

Inaugural-Dissertation

zur Erlangung des Medizinischen Doktorgrades der Medizinischen Fakultät der Albert-Ludwigs-Universität Freiburg im Breisgau

Vorgelegt 2019 von Elias Johannes Louis Potomac Schoenwandt geboren in Gießen Dekan Prof. Dr. Norbert Südkamp 1. Gutachterin Prof. Dr. H. L. Pahl 2. Gutachterin Prof. Dr. Dr. K. Domschke Jahr der Promotion 2020

Contents

1 List of AbbreviationsV

2 Summary XVIII

3 Introduction1 3.1 Hematopoiesis...... 1 3.2 Myeloproliferative Neoplasms...... 9 3.2.1 Current Understanding of Etiopathogenesis...... 11 3.2.2 Polycythemia vera...... 20 3.2.3 Essential Thrombocythemia...... 23 3.2.4 Primary Myelofibrosis...... 27 3.3 Epigenetics...... 31 3.3.1 Histone Demethylases...... 33 3.3.2 Role of Epigenetics in MPN...... 36 3.4 Aim...... 42

4 Materials and Methods 43 4.1 Cloning...... 43 4.1.1 Restriction digest...... 43 4.1.2 Removal of single stranded overhangs...... 45 4.1.3 Vector dephosphorylation...... 45 4.1.4 Ligation...... 45 4.2 Design of shRNA...... 47 4.3 Bacterial transformation...... 49 4.4 Agarose gel electrophoresis...... 51 4.5 Gel purification and extraction...... 53 4.6 Nucleic acid quantification...... 53 4.7 Plasmid DNA preparation...... 53 4.7.1 Mini preparation of plasmid DNA...... 54 4.7.2 Maxi preparation...... 54 4.8 RNA extraction...... 55 4.9 cDNA synthesis...... 55 4.10 Polymerase Chain Reaction...... 56 4.10.1 Colony PCR...... 57 4.10.2 Quantitative real-time PCR...... 58

III Contents

4.11 Protein Techniques...... 61 4.11.1 Protein Extraction...... 61 4.11.2 Protein Quantification...... 63 4.11.3 SDS-PAGE...... 63 4.11.4 Coomassie Staining...... 65 4.11.5 Western blot...... 66 4.11.6 Antibodies...... 66 4.11.7 Reprobing of blots...... 67 4.12 Cell Culture...... 69 4.12.1 Transient Transfection...... 70 4.12.2 Isolation and Handling of CD34+ cells...... 72 4.12.3 Freezing and Thawing of Cells...... 76 4.13 Virological Methods...... 76 4.13.1 pLeGo-iG-hU6...... 77 4.13.2 Production of Viral Particles...... 78 4.13.3 Titration of Viral Yield...... 79 4.14 Fluorescence Activated Cell Sorting...... 80

5 Results 81 5.1 Cloning of JMJD1C ...... 81 5.1.1 Human JMJD1C ...... 81 5.1.2 Murine JMJD1C ...... 82 5.1.3 Introduction of myc-tag...... 84 5.2 Characterization of western blot antibodies...... 87 5.3 Characterization of final shRNAs...... 88 5.4 Verification of knockdown efficiency...... 90 5.4.1 Verification by qRT-PCR...... 90 5.4.2 Verification by western blot...... 92 5.5 Lentiviral transduction...... 94 5.5.1 Virus production...... 95 5.6 JMJD1C knock down in healthy donor CD34+ cells...... 98

6 Discussion 100 6.1 JMJD1C’s Demethylase activity...... 100 6.2 JMJD1C’s Role in Stemness and Differentiation...... 104 6.3 Outlook...... 113

IV 1 List of Abbreviations

1 List of Abbreviations

µ micro

5-FU 5-Fluorouracil

5-LOX 5-Lipoxygenase

12/15-LOX 12/15-Lipoxygenase

A Adenine

A Absorbance

A Alanine

A Ampère aa Amino Acid aaDIPSS age-adjusted DIPSS

ABL1 Abelson murine leukemia viral oncogene homolog 1

ACD Asymetric Cell Division

ADSC Adipose derived stem cells

AEV Avian erythroblastosis virus

AF9 Myeloid/lymphoid or mixed-lineage leukemia; translocated to 3(MLLT3)

AGM Aorta-gonad-mesonephros

AKML Acute megakaryoblastic leukemia

ARCH Age-related Clonal Hematopoiesis

AKT Proteinkinase B alloSCT Allogenic stem cell transplantation

APS Ammonium persulfate

ASA Acetylsalicylic acid

V 1 List of Abbreviations

ASXL1 Additional Sex Combs like 1

ATM Ataxia Telangiectasia Mutated serine/threonine kinase

ATR Ataxia Telangiectasia and Rad3-related protein avWS Acquired von Willebrand syndrome

B2M β2 microglobulin

BCR Breakpoint Cluster Region

BES Bis(2-hydroxyethyl)-2-aminoethanesulfonic acid

BFU-E Burst Forming Unit - Erythroid

BM Bone marrow bp Base pairs

BRCA1 Breast Cancer 1

BSA Bovine Serum Albumin

C Carbon

C Celsius

C Cysteine

C Cytosine c centi

Ca Calcium

CALR Calreticulin

CD Cluster of Differentiation

CCD Charge-coupled Device

CDS Coding DNA Sequece cDNA complementary DNA

CFBβ Core-binding factor subunit beta

VI 1 List of Abbreviations

CFU Colony Forming Unit

CFU-C Colony Forming Unit - Cytokine (Precursor of Granulocytes and Macrophages

CFU-E Colony Forming Unit - Erythroid

CFU-G Colony Forming Unit - Granulocytic

CFU-GEMM Colony formimg unit - Granulocyte, erythrocyte, monocyte, megakaryocyte

ChIP Chromatin Immunoprecipitation

CIP Calf Intestine Phosphatase

Cl Chlorine

CLP Common Lymphoid Progenitor

CML Chronic Myelogenous Leukemia

CMP Common Myeloid Progenitor

CMP-WHO Clinical Molecular and Pathological Criteria – World Health Organization

CoIP Coimmunoprecipitation

CoREST RE1-Silencing Transcription factor corepressor 1

CXCL4 C-X-C chemokine ligand 4

CXCR4 C-X-C chemokine receptor type 4

C- Carboxy c- Cellular protooncogene

C/EBPα CCAAT-enhancer-binding proteinα c-myb MYB proto-oncogene

D Aspartate d deci

VII 1 List of Abbreviations

Da Dalton dd double destilled

DAMP Danger Associated Molecular Pattern

DGCR8 DiGeorge syndrome chromosomal region 8

DIPSS Dynamic International Prognostic Scoring System

DMEM Dulbecco’s Minimally Enhanced Medium

DMSO Dimethyl sulfoxide

DNA Deoxyribonucleic acid dNTP Deoxynucleoside triphosphate

DRED Direct Repeat Erythroid-Definitive

DSB Double Strand Breakage

DTT Dithiothreitol d-HSC Dormant hematopoietic stem cell

E Glutamate

E2A Transcription Factor 3 (TCF3)

ECL Electrochemiluminescence

EDTA Ethylenediaminetetraacetic acid

EEC Endogenous Erythroid Colonies eGFP Enhanced Green Fluorescent Protein

EPO Erythropoietin

EPOR Erythropoietin receptor

ER Endoplasmic reticulum

ESC Embryonic Stem Cell

EMH Extramedullary Hematopoiesis

VIII 1 List of Abbreviations

ET Essential Thrombocythemia

EZH2 Enhancer of zeste homolog 2

F Phenylalanine

FACS Fluorescence Activated Cell Sorting

FAM Fluorescein amidite

FCS Fetal Calf Serum

Fe Iron

FGF5 Fibroblast Growth Factor 5

Flt3 FMS-like kinase 3

FRET Förster resonance energy transfer

G Glycine

G Guanine g gram g Earth’s gravitational acceleration; m/sec2

GATA1 GATA-binding factor 1

GBX2 Gastrulation Brain Homeobox 2

GCSFR Granulocyte-colony stimulating factor receptor

GFP Green Fluorescent Protein

GIPSS Genetically Inspired Prognostic Scoring System

GLI1 Glioma-associated oncogene

GMP Granulocyte-macrophage progenitor

GRN Genetic Regulatory Network gRNA guide RNA

G-CSF Granulocyte-colony stimulating factor

IX 1 List of Abbreviations

H Histone

H Histidine

H Hydrogen h hour

HA Hemaglutinin

Hb Hemoglobin

Hct Hematocrit

HDM Histone Demethylase

HMR High Molecular Risk

HEB Transcription Factor 12 (TCF12)

HEL Human Erythroleukemia

HETE 15-(S)-Hydroxyeicosatetraenoic Acid

HIF Hypoxia Inducible Factor

HODE 13-(S)-Hydroxyoctadecadienoic Acid

HOXA9 Homeobox protein Hox-A9

HP1α Heterochromatin Protein 1α

HRP Horse Raddish Peroxidase

HSA Human Serum Albumin

HSC Hematopoietic Stem Cell

HU Hydroxyurea

I Isoleucine

IDH Isocitrate dehydrogenase

IgG Immunoglobulin G

IL-1β Interleukin 1β

X 1 List of Abbreviations

IL-6 Interleukin 6

IL-7R Interleukin 7 receptor

IL-8 Interleukin 8

IL-33 Interleukin 33

IPSET-thrombosis International Prognostic Score of Thrombosis in World Health Organization – Essential Thrombocythemia

IPSS International Prognostic Scoring System

IRES Internal Ribosome Entry Site

IMDM Iscove’s Modified Dulbecco’s Medium

JAK1 Janus Kinase 1

JAK2 Janus Kinase 2

JmjC Jumonji C domain

JMJD1C Jumonji domain containing 1C

K Lysine

K Potassium kb Kilobase; 1000 bp

L Leucine l liter

LB Lysogeny Broth

LCR Locus Control Region

Ldb1 LIM domain-binding protein 1

LDLR Low Density Lipoprotein Receptor

LepR Leptin Receptor

Lin Lineage

XI 1 List of Abbreviations

Lmo2 LIM domain only 2

LMPP Lymphoidprimed multipotent progenitor

LOH Loss of heterozygosity

LSD1 Lysine-specific histone demethylase 1A

LSK Lin−Sca1+Kit−

LT-HSC Long Term Hematopoietic Stem Cell

LTRA Long Term Repopulation Ability

Lyl1 Lymphoblastic leukemia associated hematopoiesis regulator 1

휆 average replication rate of a hematopoietic cell

M Molar m meter m milli

MAPK Mitogen-activated protein kinase

MCS Multiple Cloning Site

MDC1 Mediator of DNA damage checkpoint protein 1

MDS Myelodysplastic syndrome

MegE Megakaryocytic-erythroid

MEP Megakaryocytic-erythroid progenitor cell

Mg Magnesium

MGB Minor Groove Binder min minute

MIPSS Mutation-enhanced International Prognostic Scoring System miRISC microRNA-Induced Silencing Complex miRNA microRNA

XII 1 List of Abbreviations

MLL1 Myeloid/Lymphoid or Mixed-Lineage Leukemia 1

MNC Mononuclear Cells

MOI Multiplicity of Infection

MPL Myeloproliferative Leukemia Protein

MPN Myeloproliferative Neoplasm

MPN-BP Blast Phase Myeloproliferative Neoplasm

MPP Multipotent Progenitor Cell mPV Masked Polycythemia Vera

MPLV Myeloproliferative Leukemia Virus

MYBBP1A Myb-binding protein 1A

MyRP Myeloid-restricted Repopulating Progenitors

N Asparagine n nano

Na Sodium

NF-κB Nuclear Factor - Kappa B

NFE2 Nuclear Factor Erythroid 2

NG2 Neuron-Glial Antigen 2

NR2C Nuclear receptor subfamily 2, group C

NB Nuclear Bodies

NuRD Nucleosome Remodeling Deacetylase

N- Amino-

O Oxygen

OAC-1 Oct4-activating Compound 1

Oct4 Octamer binding Transcription Factor 4

XIII 1 List of Abbreviations

ORF Origin of Replication

P Proline p300 Adenovirus Early Region 1A-associated Protein p300

PADI4 Peptidyl Arginine Deiminase, Type IV

PAGE Polyacrylamide Gel Electrophoresis

PAMP Pathogen Associated Molecular Patterns

PBS Phosphate Buffered Saline

PcG Polycomb group

PCR Polymerase Chain Reaction

PEI Polyethylenimine

PGC Primordial Germ Cells

PGC-1 Peroxisome Proliferator-activated Receptor Gamma Coactivator 1 pH Pondus Hydrogenii; -lg a(H+)

Ph Philadelphia Chromosome

PI3K Phosphatidylinositol-4,5-bisphosphate 3-kinase

PKB Protein Kinase B

PMF Primary Myelofibrosis

PRMT5 Protein Arginine N-methyltransferase 5

PRV-1 Polycythemia Rubra Vera 1

PTGS Post-transcriptional Gene Silencing

PV Polycythemia Vera

PVDF Polyvinylidene fluoride qRT quantitative Real Time

Q Glutamine

XIV 1 List of Abbreviations

R Arginine

RAP80 BRCA1-A complex subunit RAP80

ROS Reactive Oxygen Species

RIPA Radioimmunoprecipitation assay buffer

RNA Ribonucleic Acid

RNF8 Ring finger protein 8

RNF168 Ring finger protein 168

RNAi RNA interference

RPMI Roswell Park Memorial Institute rpm Revolutions per minute

RT Room Temperature

RUNX1 Runt-related transcription factor 1

S Serine

Sca Stem Cell Antigen-1

SDF-1 Stromal Derived Factor-1; CXCL12

SDS Sodium Dodecyl Sulfate sec second

SFFV Spleen Focus Forming Virus

SFRP Secreted frizzled-related protein

SH2B1 Src homology 2-B adapter protein 1

SH2B3 Src homology 2-B adapter protein 3 shRNA short hairpin RNA siRNA short interferring RNA

SILAC Stable isotope labeling by amino acids in cell culture

XV 1 List of Abbreviations

SIN-LTR Self-inactivating long terminal repeats

SLAM Signaling lymphocytic activation molecule

SNP Single Nucleotide Polymorphism

SOC Super Optimal broth with Catabolite repression

SOX2 Sex determining region Y -box 2

SRSF2 Serine/arginine-rich splicing factor 2

SSEA-1 Stage-specific Embryonic Antigen 1

ST-HSC Short Term Hematopoietic Stem Cell

STAT Signal Transducer and Activator of Transcription

SVT Splanchnic Vein Thrombosis

T Thymine

T Threonine t Translocation

T3 Triiodothyronine

TAE Tris-Acetate-EDTA buffer

TBS Tris Buffered Saline

TEMED Tetramethylethylenediamine

TERT Telomerase Reverse transcriptase

TET2 Ten-eleven-translocation 2

TGF-β Transforming Growth Factor-β

TLR Toll-like receptor

TNF훼 Tumor Necrosis Factor 훼

TP53 Tumor Protein 53

TPOR Thrombopoietin receptor

XVI 1 List of Abbreviations

TR Thyroid Hormone receptor

Tris 2-Amino-2-(hydroxymethyl)propane-1,3-diol

TrxG Trithorax group

U Unit

UTR Untranslated Region

UV Ultraviolet

V Valine

V Volt

VSEL Very Small Embryonic like Stem Cell

VTE Venous Thrombembolism vWF von Willebrand Factor v- Viral Oncogene

W Tryptophane

WHO World Health Organization wnt Wingless/Integrated wPRE Woodchuck Hepatitis Virus Posttranscriptional Regulatory Element

Y Tyrosine

XVII 2 Summary

2 Summary

Myeloproliferative Neoplasms are defined as clonal hyperplasia in the myeloid compartment of hematopoiesis. Despite the identification of driver mutations, the diseases’ etiopathogenesis remains incompletely understood. The transcription factor NFE2 is overexpressed in MPN and transgenic models could prove its contribution to the disease phenotype.

The histone demethylase JMJD1C is one of NFE2’s target genes which is equally overexpressed in PV patients. Thus, the hypothesis that NFE2 ushers in the MPN phenotype through the expression of JMJD1C serves as the starting point for this work. Herein, the role of JMJD1C in hematopoiesis was dissected by virtue of gene silencing experiments. The ground to these experiments was established by cloning mammalian expression vectors of JMJD1C, identifying western blot antibodies detecting JMJD1C and designing shRNA sequences that were finally validated with the aforementioned tools.

Hematopoietic stem cells of healthy donors were lentivirally transduced with either control sequences or shRNA targeting JMJD1C. Silencing of JMJD1C led to a shift towards granulocytic differentiation at the expense of erythroid differentiation. Inlight of recently published data about JMJD1C’s role in leukemia, these results underscore JMJD1C’s role in stem cell biology and myeloid differentiation. A contribution to the MPN phenotype is, thus, possible and should be further investigated against the backdrop of its dubious enzymatic activity and plethora of identified interaction partners.

XVIII 2 Summary

Zusammenfassung

Myeloproliferative Neoplasien sind als klonale Hyperplasie des myeloiden Kompartiments definiert. Die Ätiopathogenese dieser Erkrankungsgruppe ist trotz der Entdeckung von driver-Mutationen unvollständig verstanden. Der Transkriptionsfaktor NFE2 ist bei MPN überexprimiert und sein Beitrag zum Phänotyp der Erkrankung konnte mittels transgener Modelle nachgewiesen werden.

Die Histondemethylase JMJD1C, welche ein Zielgen NFE2s darstellt, wird ebenfalls in PV- Patienten überexprimiert. Daher dient die Hypothese, dass NFE2 den MPN-Phänotyp durch die Expression von JMJD1C in Gang setzt, als Ausgangspunkt dieser Arbeit. Hierbei wurde JMJD1Cs Rolle in der Hämatopoese durch gene silencing-Experimente untersucht. Zur Etablierung dieser Versuche gehörte die Klonierung von JMJD1C- Säugetier-Expressionsvektoren, die Identifikation von Antikörpern, die JMJD1C im Western Blot nachweisen konnten, sowie die Konzeption von shRNA-Sequenzen, die mit den erstgenannten Techniken validiert wurden.

Hämatopoetische Stammzellen gesunder Spender wurden mittels Lentiviren entweder mit Kontrollsequenzen oder mit shRNA gegen JMJD1C transduziert. Die Herabreg- ulation von JMJD1C verschob die Differenzierung zu Ungunsten der erythroiden Dif- ferenzierung hin zur granulozytären Differenzierung. Vor dem Hintergrund der letzten Publikationen zu JMJD1Cs Rolle bei Leukämie unterstreichen diese Ergebnisse JMJD1Cs Einfluss auf Stammzellbiologie und myeloide Differenzierung. Sein Beitrag zumMPN- Phänotyp ist damit möglich und sollte vor allem vor dem Hintergrund der ungeklärten enzymatischen Eigenschaften und der Fülle an Interaktionspartnern untersucht werden.

XIX

3 Introduction

3.1 Hematopoiesis

Hematopoiesis describes the constant production of the canonical corpuscular components of the blood, taking place at a rate of aproximately 1×1012 cells a day or 1.5×106 cells per second. [283][50] In mammals, primitive hematopoietic cells first arise in the embryonic yolk sac and consecutively move to the liver. The definitive hematopoietic stem cells, however, originate from the hemogenic endothelium of the primitive aorta in a process called endothelial-to-hematopoietic transition. These cells populate the bone marrow of the axial skeleton where lifelong physiological hematopoiesis takes place. [380][86] Within the bone marrow, stem cells reside in a perivascular niche which is composed of endothelial cells, smooth muscle cells, pericytes and stromal cells. [337][207][187][4] Besides the unphysiological occurence of extramedually hematopoiesis in disease, the lungs as a site of physiological extrameduallary hematopoiesis have recently gained interest. [219][41]

Although the concept of a hematopoietic stem cell (HSC) was already envisioned at the turn of the last century by Maximow, Dantschakoff, Neumann and Pappenheim [309], its existence was verified only about half a century later by the results of seminal transplantation studies [31]. Since then, hematopoiesis has been understood as a hierarchical cascade of differentation steps that all have their origin in a primitive stem cellthat possesses two crucial features:

a) multipotency which allows it to give rise to all cells of the blood as well

b) the ability to self-renewal

Together, these characteristics endow the HSC with the potential to reconstitute the entire hematopoietic system. Conventionally, long term repopulation ability (LTRA) [96] defines a cell’s potential to give rise to progeny for more than 12 weeksafter transplantation. [72] The governing concept of HSC biology sees these two features - multipotency and self-renewal capacity - vanishing in step as HSCs differentiate into effector cells. [339]

How LT-HSCs replenish themselves remains elusive. Based on observations in D. melanogaster as well as human neuroblasts, it was conjectured that LT-HSCs might undergo asymmetric cell division (ACD), which refers to a polarized mitosis in which the

1 3 Introduction mother cell keeps stem cell features while the daughter cell loses these features and begins to differentiate. Conclusive results that either support or rule out ACD in LT-HSCs are missing until this day. [126][154]

Recently, a theory has been proposed in which distinct perivascular niches exist that are associated with different stem cell fates: on the one hand, theNG2+ peri-arteriolar niche which harbors dormant stem cells, and on the other hand the LepR+ peri-sinusoidal niche which allows for stem cell proliferation. [210] However, the concept has been challenged by conflicting evidence. [4]

Due to the dynamics of the field, even fundamental concepts are challenged bynew evidence. Therefore, it seems more appropriate to summarize current discussions around a set of basic questions of physiological hematopoiesis than to attempt the presentation of an objective truth.

Which cell is the pivotal for unperturbed hematopoiesis? Since the 1980s it has been postulated that the cycling of hematopoietic cells is positively correlated with their degree of differentiation [155]. While rodent LT-HSCs have a cycling half life of just 19 days [45], results from similar BrdU-incorporation studies with baboons show that primate HSCs cycle likewise but at a much lower rate than their rodent counterparts. To answer the question in human, Catlin et al. employed X chromosome inactivation and computer modeling to calculate 휆, the average replication rate of a hematopoietic stem cell: With about one replication per 40 weeks, this is even lower. [61]

Cycling, however, can be increased by challenging the system with 5-FU or G-CSF. [242] A common interpretation sees this relative quiescence as a mechanism to safeguard the HSCs’ genomic stability by avoiding the cellular stress associated with a fast paced cell cycle, while allowing for rapid increase in cell production when necessary. [231] [294][419][37][50] Meanwhile current estimates of stem cell abundance postulate about 1-4 HSCs per 107 mononuclear cells, or about 3-5 stem cells per kilogram bodyweight. [55]

Against the backdrop of both this scarce number of HSCs as well as the limited turn over of these cell, the question arises to which degree unperturbed, homeostatic hematopoiesis is independent of HSC feed in or at least for how long before cellular insufficiencies arise. Recent data from studies avoiding transplantation reveal that ST-HSCs and even multipotent progenitor cell (MPPs) have a rather high self-renewal capacity that endows them with an average lifetime of 330 and 70 days, respectively. Given the very limited

2 3 Introduction

flow from the LT-HSC to the ST-HSC compartment found by the authors, it wasdeducted that ST-HSCs and MPPs are the main source of hematopoietic output. The authors also noted that the oligoclonality observed after bone marrow transplantation could be consequence of a cell production that relies solely on LT-HSCs [52]. This goes in line with Catlin et al. simulations of bone marrow failure that demonstrate that a lack of progenitor

cells and not an insufficient 푅0, the amount of transplanted HSCs, is responsible for graft failure. They argue that HSCs persist but that the speed at which they repopulate the progenitor compartment is insufficient. [61]

The notion that physiological polyclonality relies on progenitors and not LT-HSCs is reinforced by a study using transposon barcoding. It showed that murine granulocytes originated from roughly 800 clones at any given time point; a number rather close to the roughly 1250 clones Catlin et al. calculated to be necessary for human hematopoiesis [61]. Since this abundance was detected over the course of a year the authors concluded that steady-state hematopoiesis is based on successive recruitment of thousands of progenitor cells constantly switching between quiesence and cellular production. [377]

Taken together these results transmit a concept of unperturbed hematopoiesis relying foremost on progenitors and confines the LT-HSCs to an auxiliary role in which the mostly dormant HSCs (d-HSCs) are activated by bone marrow injury or G-CSF and return to dormancy as soon as homeostasis is reattained. [432]

Is hematopoiesis dichotomous? Conventionally, the cells deriving from hematopoiesis are segregated into two domains: While the myeloid lineage is considered to contain erythroid and megakaryocytic cells along side all types of granulocytes, the lymphoid lineage spans T-cells, NK-cells as well as B-cells. (see classical model in Fig.1). This division is based upon histo(patho)logical observations, and proved to be reliable in characterizing hematological malignancies in clinical practice. [183]

First experimental evidence supporting this model was delivered in the form of semi- solid colony assays (CFU-C) that identified the CFU-GEMM (colony forming unit- granulocytes, erythroid cells, megakaryocytes, and macrophages) as the precursor of all non- lymphoid cells. About thirty years later, CFU-GEMM was characterized as the common myeloid progenitor (CMP).[9] Three years before the CMP was finally characterized, the exisistence of a common lymphoid progenitor (CLP) had been postulated. [203] The CMP and CLP represent the fundamental pillars of the classical, dichotomous model. Since

3 3 Introduction

both cell types were identified in the laboratory of Irving L. Weissman, it is conventionally referred to as Weissman model. It prevailed in spite of a series of different models that had been proposed since the 1970s. [3][284][47]

Figure 1. Comparison of three current models of hematopoiesis. Note that surface markers mirror the murine hematopoietic system. Adapted from [280]

Lately, the Weissman model was subjected to critical appraisal as the IL-7R+ CLP population was found to harbor myeloid potential; a feature incompatible with the model’s strict dichotomy. Adolfsson et al. identified expression of FMS-like tyrosine kinase 3 (Flt3) as a bifurcation since cells losing Flt3 expression due to ACD differentiate towards a megakaryo-erytroid (MegE) fate while sustained Flt3 enables cells to develop into both myeloid and lymphoid cells. Because the Flt3+ population was observed to be primed towards lymphoid differentiation due to common coexpression of IL-7R, the cells were termed lymphoid-primed multipotent progenitors (LMPP).[6][131] Taken together, this revised model upholds the concepts of a critical bifurcation at the level of the MPP but sets it between the MegE potential and all other lineages.

A series of recent studies have revealed distinct, lineage-restricted subgroups within the HSC compartment. [252] Dykstra et al. where indeed able to identify two LT-HSC phenotypes that give rise to different profiles of progeny: while cells ofthe β-cluster give rise to both myeloid and lymphoid WBCs, the WBC progeny of α-cluster HSCs was almost exclusively myeloid. The β-cluster can transition through the γ and δ-cluster which increasingly liken the ST-HSC phenotype concomitantly with restricting to lymphoid

4 3 Introduction

progeny. [95] This is echoed by the results of Yamamoto et al. which imply that the MegE lineage cleaves off the other cell fates already within the HSC compartment byvirtue of ACD, suggesting that the CD150+ HSC represents a myeloid-restricted repopulating progenitor (MyRP) with extensive self-renewal capacities. [438] In addition, CD150- positivity was found to correlate with the expression of von Willebrand-factor (vWF). It was observed that platelet primed vWF+ HSCs could give rise to vWF− lymphoid primed HSCs but not the other way around which would place the myeloid biased HSCs at the apex of the hematopoietic hierarchy. [332]

Thus, the myeloid bypass model which is based upon these discoveries goes even further than the revised model as it abandons the dichotomy of the classical system in breaking the primordial, unified LT-HSC into lineage biased subgroups. [438][20] As a consequence it challenges the proper concept of the hematopoietic cell hitherto in which lineage commitment and LTRA were said to be mutually exclusive. Taken together, both the revised model as well as the myeloid bypass model propose new perspectives on hematopoiesis in which the MegE lineage is singled out from the classical myeloid lineage and placed in closer proximity to the most primitive HSC; with the latter fact being of importance for coming parts of this work.

Is differentiation a deterministic or stochastic process? The above presented conun- drum of differentiation is founded upon the more fundamental question whether cellular fate is a deterministic process or stochastic in nature. [192] The idea of randomness in hematopoiesis was first formulated by Till and McCulloch who analyzed the outcome of CFU-S differentiation. [31][359] They postulated that single cells iterate a number of branching points in the differentiation process and that the outcome of each branching point is random. This results in a spectrum of different phenotypes and behaviors that follow a Gaussian distribution. In this model, regulation of hematopoiesis takes place solely at the level of the entire population by skewing the distribution, not the single cell. [426]

This concept lost support as cytokine receptors were identified and their roles in hematopoietic regulation were understood. Instead of large, rather unpredictably behaving populations, subgroups of cells could be identified with predetermined behavior. [355] The idea of stochasticity is revived by taking the complexity of cellular signaling and transcription networks into account. Rather than conceptualizing the role of cytokines as instructive, meaning that they actively drive the differentiation process, the permissive concept em-

5 3 Introduction

phasizes that differentiation happens based upon the cell’s intrinsic transcription program with cytokines providing cues for growth and survival. [313]

Figure 2. Toggle switch of PU.1 and GATA1. Adapted from [17]

A hallmark of the permissive concept is the existence of so called toggle switches in the differentiation program. Toggle switches are composed of two (or more) antagonizing transcription programs that have the ability to inhibit each other while reinforcing themselves through autoregulatory feed forward mechanisms. In the undifferentiated progenitor, all programs lie in an unstable equilibrium that randomly tilts towards one program, thereby initiating the differentiation towards a given kind of cell. A salient example of a toggle switch is the antagonism of GATA1 and the PU.1 (Fig.2) in the determination of the myeloid fate. [165][17]

Loose and Patient attempted to summarize and integrate all known mechanism in a genetic regulatory network (GRN) that governs hematopoiesis. (Fig.3) They highlight that initially several different lineage programs can be active simultaneously in astate of multilineage priming and that minute changes in transcription factor concentrations or cytokine signalling could lead to the pursuit of one distinct fate. [232] Lastly, the discussion remains whether the attempt to grasp the regulation of hematopoiesis in a highly complex, yet finite, GRN is an example of molecular determinism. This theory debunks stochasticity by reducing randomness to a chaotic, but discernable number of minute differences (e.g. cell size, position in the colony etc.) that influence the cellular outcome. [361]

6 3 Introduction

Figure 3. Genetic regulatory network of the erythroid differentiation. Adapted from [379]

7 3 Introduction

As a closing remark, it should be recalled that the majority of the research that led to these concepts is based on studies in mice. Despite the striking similarities to the human hematopoietic system, there is a series of differences. The most obvious difference is that human leukocytes are predominantly granulocytes while murine leukocytes are mostly lymphocytes. [90] The HSCs of the two species differ in various re- gards:

First, the immunophenotypes are very unlike. While both LSK (lin−Sca+kit+)[365][273] as well as SLAM (CD150+CD48−)[187] proved successful to isolate the HSC compartment in mice, they are not of help for the characterization of the human counterparts. In addition, the murine HSC lack CD34 expression while CD34 positivity has been proven as the most reliable marker for HSC isolation in humans. Similarly, murine HSCs are negative for Flt3 while human HSCs are positive. [381][292]

Second, their transcriptional programs differ as illustrated by the overexpression of the transcription factor HOXB4 which leads to a 1000-fold expansion of HSCs in mice but only to a 2-4 fold expansion in human cells. [90] Third, the species’ HSCs differ in their response to DNA damage as human HSCs are more likely to initiate apoptosis than their murine counterparts. One hypothesis states that this difference may have arisen under the evolution of the comparatively long human life span which requires a more stringent maintenance of genomic stability. [37]

Lastly, it should be noted that the mouse strains used for the experiments are inbred and the choice of a specific genetic background may yield results with high internal validity but lack the external validity needed to apply them to the genetically diverse human population. [90][292]

8 3 Introduction

3.2 Myeloproliferative Neoplasms

Though the different diseases where described earlier, it was not until 1951 that Dameshek first hypothesized that what were then called Chronic Granulocytic Leukemia, Idiopathic Myeloid Metaplasia of the Spleen as well as Polycythemia and Megakaryotypic Leukemia could be grouped together: These diseases present with similar clinical courses, transitions between them are possible, and most importantly they share the common feature of myeloid bone marrow hypercellularity due to an inappropriate growth stimulus. Therefore he coined the term "myeloproliferative disorder". [83]

The discovery of the Philadelphia Chromosome (Ph) by Nowell et al. in 1960 demarks a milestone in the history of myeloproliferative disease. [281] Almost all cases of Chronic Myelogenous Leukemia (CML) bear the fusion chromosome which was identified as t(9;22) in 1973. [322] Later the resulting Bcr-Abl fusion protein [353] was identified as the pathogenetic agent in CML [82]. The comprehensive understanding about the disease’s pathophysiology distinguishes the Ph+ MPN from their Ph- counterparts in which an exact pathogenetic mechanism remains elusive in spite of the progress that has been made. Therefore CML is no further addressed in this work and the term "MPN" is used with regard to Ph- MPN if not otherwise denoted.

The disease nomenclature has changed in the recent years: In 2007, the current terminology for primary and secondary myelofibrosis as well as chronic and blast phase was established. [260]. In 2008, "myeloproliferative disorders" were relabeled as "myeloproliferative neoplasms" (MPN) with the introduction of the 3rd edition of the World Health Organization’s (WHO) classification. This reflects the appraisal of the identification of driver mutations in tyrosine kinases which will be elaborated in 3.2.1. This led to the conclusion that the inappropriate growth stimulus Dameshek postulated is indeed an oncogene. [411] With the 2016 revision of the fourth edition of the WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues the category of MPN comprises the entities listed in Tab.1.

Besides the fact that mastocytosis is no longer grouped within MPN due to its widely varying clinical course [25], the introduction of prefibrotic Primary Myelofibrosis as a distinct subentity constitutes the only appreciable change in the definition of the category.

9 3 Introduction

Table 1. List of entities classified as Myeloproliferative Neoplasms

∙ Chronic Myeloid Leukemia, Ph+

∙ Chronic Neutrophilic Leukemia

∙ Polycythemia Vera

∙ Primary Myelofibrosis – PMF, prefibrotic/ early stage – PMF, overt fibrotic stage

∙ Essential Thrombocythemia

∙ Chronic Eosinophilic Leukemia, not otherwise specified

∙ MPN, unclassifiable

Figure 4. Abridged representation of the classification proposed in the WHO- CMP. Adapted from [264]

10 3 Introduction

The 2016 WHO Clinical Molecular and Pathological Criteria (WHO-CMP) for Classifi- cation and Staging of Myeloproliferative Neoplasms was proposed by Michiels et al. It subclassifies MPN phenotypes according to molecular and histopathological markers into stages that span a spectrum ranging between the classical entities. Therefore, it is a keen effort to unify as well as operationalize the complex variations of MPN. Yet the resulting system itself is too intricate to outline it further. [264]

3.2.1 Current Understanding of Etiopathogenesis

Predisposing factors As in virtually all malignancies, the incidence of MPN increases exponentially with age. [257][375] This is due to the rising likelihood of acquiring one of the driver mutations that will be addressed in 3.2.1. Yet there is evidence that not all carriers of a driver mutation develop MPN. It is assumed that many of them develop age- related clonal hematopoiesis (ARCH).[151] Therefore, it is stipulated that the emergence of the MPN phenotype depends on the genetic context in which the driver mutation occurs. Germline mutations play an important role in shaping this landscape, with a series of such mutations discovered throughout the past years:

The 46/1 haplotype, which is a SNP at JAK2 rs10974944 with GG/CC genotype, increases the risk of developing JAK2푉 617퐹 -positive MPN three to four times. 80% of JAK2푉 617퐹 mutations arise in the context of a 46/1 haplotype and 75% of these mutation are located cis of the predisposition allele. Especially the latter finding has fostered the hypothesis that the SNP renders the JAK2 locus prone to mutation. [152][170][285] [189]

Shen et al. could prove that the rs3733609_C/T germline variant in exon 9 of TET2 is likewise positively associated with JAK2푉 617퐹 MPN. The polymorphism leads to decreased TET2 expression by attenuating binding of C/EBPα to the locus. The SNP is not only associated with elevated hemoglobin concentration, leukocyte and thrombocyte counts but also a higher incidence of hepatosplenomegaly as well as thromobembolic events. [342]

The rs2736100_C SNP in the second intron of TERT is associated with all MPN subtypes and driver mutations. TERT codes for the reverse transcriptase in the telomerase complex which led to the hypothesis that this allele may contribute to a global genome instability. The impact of TERT polymorphism was deemed even higher with a population attributable risk of about 50% [282][403]

11 3 Introduction

Lastly, germline duplications of ATG2B and GSKIP, germline mutations in RBBP6 and SH2B3 as well as SNPs in SH2B3 have been identified in familial MPN cases and seem to predispose the development of MPN. [329][148][224][326] Germline mutations contribute to understanding the vital role of heredity of MPN: a first-degree relative suffering from either PV or ET leads to a statistically relevant increase in theriskto develop a myeloid malagnancies; most likely the same entity. Furthermore, MPN is more likely to develop at a younger age if the patient has first-degree relatives with a history of myeloid malignancy. [375]

The role of gender in MPN is most noticeably mirrored by the uneven distribution among the entities: PV is the most common MPN entity among men, while ET is the most common one among female patients. [125] In the case of ET, male sex is a strong independent risk factor that surpasses the role of thrombosis history in overall survival. Whether this is an immediate consequence of the disease’s biology, or a reflection of the conditions relatively benign course letting shine through regular sex differences in ageing remains debatable. [389] Female cases of myelofibrosis are most often secondary toET, whereas male myelofibrosis cases are most often of primary genesis. [125] This observation goes in line with the results of Stein et. al who demonstrated that female JAK2푉 617퐹 - positive MPN patients harbor lower allele burdens than their male counterparts as well as a lower increase in allele burden over time; the relationship between allele burden and MPN entity will be adressed in more detail in 3.2.1. Nevertheless, females are about 4.5 times more likely to progress from ET to PV, while they bear 0.23 times the risk of men to transform from ET to secondary myelofibrosis. [369] Spivak et al. demonstrated that the expression of about twice as many genes was deregulated in male patients compared to female patients. When analyzing the impact on molecular pathways, the authors showed that despite the smaller number of deregulated genes about three times as many pathways were activated in female patients. [366]

Ethnicity’s etiological role in MPN is rather incompletely understood. [195] Data from Korea show that Caucasian populations have a higher incidence of MPN than African American or East Asian populations and that PV is the most common MPN in Caucasians and Hispanics while ET constitutes the most common MPN among African Americans and East Asians. [56] On the one hand, Caucasian ethnicity has been shown to be an independent protective factor for cardiovascular thrombosis and hemorrhagic events in MPN. [184] On the other hand, it constitutes an independent risk factor for transformation from both PV and ET to myelofibrosis. [184] The impact of ethnicity on PMF has been analyzed in Chinese population with contradicting results. [437][128]

12 3 Introduction

Lastly, an Israeli cohort study could demonstrate that obesity in adolescent age is a positive predictor for development of MPN later in life. [220] It would be of interest to analyze whether there is a connection to genetic aberration in SH2B1 as it not only associated with early-onset obesity [401][428][38] at the same time as it known to interact with JAK2 in the context of growth hormone signaling. [324]

Driver mutations Although the exact pathogenesis remains unclear, mutations in JAK2, CALR and MPL are referred to as driver mutations paying tribute to the facts that, first, mutant clones can be found in more than 90% of MPN patients[279] and, second, these mutations have been shown to induce MPN phenotypes in animal models. [92]

Gene expression analysis from patient material shows that all three mutations are linked to an activation of the JAK-STAT pathway, pointing out the pivotal role this pathway seems to take in the pathogenesis of MPN. [310] Hence, it seems even more striking that in spite of this common mechanism coexistence of two driver mutations is found in < 5% of patients. [279][176]

Patients not featuring any of the three major known driver mutations - often termed triple negative - face a double whammy as a comprehension of pathophysiology is not only a relevant prerequisite for the development of a suitable therapy but they are also the ones most in need of it: their survival prognosis is significantly poorer than for those with a known driver mutation. [390]

JAK2 Janus kinase 2 (JAK2) is a member of the Janus kinase (JAK) family of non- receptor protein tyrosine kinases. Upon ligand-mediated dimerization of type I cytokine receptors like EpoR, TpoR and GCSFR, JAK2 dimerizes and transphosphorylates. [234] This leads to conformational changes that allow for phosphorylation of signaling proteins. The substrate of JAK2 that has been shown to be most relevant for the MPN phenotype is signal transducer and activator of transcription (STAT) 5 which dimerizes upon phosphorylation by JAK2 and translocates to the nucleus in order to start various transcription programs. [439] All JAK2 mutations identified in MPN patients have a cytokine-independent activation of phosphorylation in common, which explains the above mentioned hyperactivation of the JAK-STAT pathway.

The most common missense mutation of amino acid 617 from valine to phenylalanine (JAK2푉 617퐹 ) accounts for 95% of PV cases and 50-60% of ET and PMF cases. [390][196]

13 3 Introduction

X-ray crystallography could show that the exchanged side chain exerts excessive π-bonding which rigidifies the backbone of JAK2’s inhibitory JH2 domain. This rigidification is discussed as the reason for the constitutive phosphorylation activity of JAK2푉 617퐹 .[357] In the majority of the JAK2푉 617퐹 -negative PV patients a mutation in exon 12 of JAK2 can be identified. Most of these mutations are situated in the linker region between the SH2 and JH2 domain. [338] Mechanistic studies of the linker imply its function in transducing activating conformational changes from the receptor to the phosphorylase domain. [449]

The capability of JAK2 mutations to induce all three MPN entities is thought to rely on a gene dosage effect. This effect was reproduced in a mouse model comparing heterozygous and homozygous mutants [8][402] as well as observed in MPN patients [402]; both point out a correlation of low mutant expression to an ET phenotype, while higher expression levels show a gradient which ranges from PV over PV with high propensity to secondary myelofibrosis to a phenotype of primary myelofibrosis. One mechanism that contributes to such differing gene dosages is loss of heterozygosity (LOH) of the p-arm of chromosome 9 (9p). Mitotic recombination with consecutive uniparental disomy is considered to be the underlying cause of it. LOH 9p can be identified in 30% of MPN patients andis more common in PV and PMF. [205]

CALR Calreticulin (CALR) is a Ca2+-binding chaperone protein situated in the endo- plasmatic reticulum (ER) [262] which was found to be mutated in 70-84% of JAK2-negative ET and PMF cases. [279] Both PMF as well as ET patients with CALR mutation have a better survival rate compared to patients with JAK2 or MPL mutations. Furthermore, CALR mutated ET patients have a lower incidence of thrombosis than patients with JAK2푉 617퐹 .[196] Mutant CALR seems to promote megakaryocytic differentiation as it drives megakaryocytes to express higher levels of markers such as Mpl, CD41 and NFE2. [143]

Though a multitude of CALR mutations have been detected in ET/PMF patients, they share an array of features: All mutations are located in exon 9 and although different in nature, they all lead to the same one base pair shift to an alternative reading frame. As exon 9 is the exon directly 5’ to the stop codon, the frame shift affects solely the protein’s C-terminus. Two differences between the wild-type and the mutant C-terminus are most striking:

14 3 Introduction

First, the KDEL-motif responsible for CALR’s retention in the ER is lost. Second, the mutant sequence contains substantially fewer negatively charged amino acids. [196] Yet there are also noteworthy differences across the various mutations: 53% of CALR mutations are a 52-bp deletion flanked by 7 base pairs of identical sequence (p.L367fs*46, c.1092_1143del; denoted as type 1 CALR mutation) while about 32% are a 5-bp insertion (p.K385fs*47, c.1154_1155insTTGTC; denoted as type 2 mutation)[279][196] This distinction is not only of genetical but also clinical relevance as a) the survival advantage of CALR mutated over JAK2 or MPL mutated PMF patients is confined to cases of type 1 mutation [393] and b) type 2 mutated ET cases feature significantly higher platelet counts than their type 1 counterparts. [397]

Based upon the observation that the type 1 mutation gives rise to protein that contains less α-helices than wild-type CALR while the type 2 mutation does the opposite, all other known mutations were grouped into the so called Mayo Classification of "type 1-like" or "type 2-like" depending on the mutant protein’s expected secondary structure. Most importantly, the molecular grouping correlated with the respective survival characteristics of type 1 and type 2 mutations. [215]

As the JAK-STAT axis has been established as the canonical pathogenetic mechanism of MPN, current research aims at identifying the link between intracellular signaling and CALR. Studies of subcellular localisation could show that mutant CALR is mislocated in the nucleus as well as pre-Golgi, Golgi apparatus and even the extracellular space. [65][123] This is presumed to be a consequence of the loss of both the KDEL sequence as well as Ca2+-binding motifs. A blockage of the secretory pathway has been demonstrated to cause an endoplasmic accumulation of mutant CALR at the same time as JAK-STAT signaling is decreased. [143] This leads to the hypothesis that mutant CALR exerts its effect via excretion, thus, in an endocrine fashion. Two endocrine models havebeen proposed:

In the indirect model, CALR causes inflammatory signaling that leads to MPNas elaborated in 3.2.1. This model relies on two findings: First, incubation of monocytes with medium from mutant-CALR transfected HeLa cells caused elevated production of cytokines. [123] Second, PMF patients’ levels of circulating mutant CALR correlated with levels of circulating pro-inflammatory Interleukin-6 (IL-6) which in turn correlated with disease parameters like bone marrow fibrosis, spleen size and hemoglobin concentration. [362] The majority of current literature, however, focuses on the finding that CALR mutations rely on the presence of the thrombopoietin receptor (TPOR) and even JAK2

15 3 Introduction

in order to give rise to the MPN phenotype. [16][65][101] In this direct model, CALR functions as a ligand to the TPOR which is backed by the observation that TPOR’s extracellular domains are necessary for CALR induced JAK-STAT signaling. [65] It is not elucidated to which degree the two proteins aggregate already in the Golgi apparatus, implying an autocrine mechanism, or whether CALR is secreted and acts paracrine. Nevertheless, it could be demonstrated that the positive charge of the mutant C-terminus is necessary for the interaction [101] as it interferes with other protein domains allowing CALR’s N-terminal domain to bind to TPOR. [16]

MPL c-mpl was identified as the cellular homologue of the murine myeloproliferative leukemia virus (MPLV) oncogene v-mpl even before it was characterized as the gene encoding TPOR and its ligand TPO was found. [261] Since then c-mpl mutations have been identified in a range of conditions involving megakaryopoiesis:

While loss-of-function mutations are the main cause of congenital amegakaryocytic thrombocytopenia [21], activating mutations like MPLT487A were found in acute megakaryoblastic leukemia (AKML) [243]. Mpl Baltimore (K39N) in African-Americans [270] and MPLP106L in Arab populations [99] are both linked to hereditary thrombocytosis. Activating mutations found in ET and PMF are almost exclusively located in exon 10 [117], which codes for TPOR’s transmembrane domain [303]. The mutations W515L, W515K and W515A occur most frequently with a rate of about 5-8% in PMF and 1-3% in ET patients. [117][291][407][139] They render the receptor constitutively active by tilting the transmembrane helices’ angle. [88]

The dominant-positive S505N allele was found in a Japanese family affected by familial essential thrombocythemia [89] but can also be detected in cases of sporadic ET. [117] This mutation activates the receptor through the same mechanism as the small molecule eltrombopag by inducing α-Helix formation in a domain of the human TPOR that is usually non-helical. [223]

Although a few cases were identified in which c-mpl mutations occurred concomitantly with JAK2 mutations, it is most commonly found in JAK2-wild type patients. [291] Mutations in c-mpl are associated with female sex, older age, higher platelet count, and lower hemoglobin level. [407][33][139] The MPL-mutant PMF patients’ survival does not significantly differ from the survival of CALR-mutant PMF patients. However, MPL- mutant ET patients have an increased propensity to transform to secondary MF, with the underlying mechanism being incompletely understood. [390]

16 3 Introduction

Inflammation There are several reasons to consider inflammatory mechanisms in the pathogenesis of MPN: First, classical inflammation induces physiological leukocytosis which likens the increased output of mature myeloid cells found in MPN. [103] Second, the role of inflammation in the carcinogenesis of solid neoplasia is generally accepted. [246][144]

From the vast field of inflammation three assorted topics shall be presented in their connex- ion with MPN: Toll-like receptors (TLR), cytokines and eicosanoids. TLRs are innate immunity receptors that recognize cognate danger- or pathogen-associated molecular patterns (DAMPs/PAMPs) which trigger transcription of inflammatory cytokines via Nuclear Factor Kappa B (NF-κB). Though TLRs are canonically thought to be expressed by macrophages and neutrophils, Zhao et al. verified the expression of TLR on ST-HSCs and MPPs and proved that upon TLR stimulation these cells are capable of secreting a broad range of cytokines in quantities surpassing the production by myeloid effector cells. [448] Hence, HSCs can initiate and maintain an inflammatory milieu in the bone marrow niche. Furthermore, constant TLR activation has been shown to alter HCS self-renewal capacity in mice with subsequent myeloid bias. [104]

Kleppe et al. found that cytokine production in MPN is not restricted to the malignant clone. While malignant cells produce primarily IL-6, non-malignant cells produce a set of different cytokines. Tumor Necrosis Factor α (TNF-α) is among a third group which is produced in both malignant and non-malignant cells, [198] which is especially interesting as TNF-α does not only correlate with JAK2푉 617퐹 allele burden but MPN clones have been proven to be insensitive to TNF-α induced suppression of colony formation [115]. This provides a unique feed-forward loop for the establishment of monoclonality. Stromal cells attenuate a MPN phenotype via the production of Interleukin 33 (IL-33) which selectively promotes the proliferation of JAK2 mutant cells. [241] In a similar manner the pro-inflammatory cytokine Lipocalin-2 seems to act on PMF: Lipocalin-2 levels are elevated in patient plasma and it induces increased ROS production in the cells of healthy controls but not those of PMF patients. [233]

Clinically, the prognostic value of both cytokines as well as acute phase proteins has been demonstrated for PV [406][238], ET [306][238] and PMF [395][28] in which they could predict both survival as well thromboembolism. Yet there is evidence that inflammation is not a mere byproduct of myeloproliferative disease, but relevant for disease maintenance and even disease initiation:

17 3 Introduction

On the one hand, observational studies point out that prior autoimmune conditions - irrespective of the affected organ system - constitute a risk factor for the development of MPN. [208] On the other hand, there are preclinical data: The transcription factor NFE2 which is overexpressed in MPN [130] is upregulated by Interleukin 1β (IL-1β)[73]. Higher levels of NFE2 cause cell proliferation [173] as well as increased expression of Interleukin 8 (IL8) [424], connecting inflammatory signaling with the MPN genetic landscape.

As will be elaborated in 3.3.2, is the epigenetically acting enzyme TET2 is commonly mutated in both ARCH and MPN. Zhang et al. could prove that IL-6 mediated inflammation in MPN can be at least partially attributed to the loss of TET2.[447] Lastly, Fuster et al. reported that mice reconstituted with TET2-deficient cells displayed increased atherosclerosis which was attributed to increased IL-1β production from tissue macrophages. [118] In synopsis, these findings relate the phenomenon of "hematopoietic aging" which entails a myeloid bias [376][272][190] with theories on inflammation as a trigger for age-related disease, referred to as "inflammaging." [268]

Targeting inflammation in MPN therapy seems logical in the light of such results, andis indeed already part of established therapies: Interferon α (IFN-α) is an immunomodu- latory cytokine that has proven to be successful in PV, ET [188] and PMF [358][305]. Though the exact mechanism of action remains incompletely understood, experimental evidence traced IFN-α effect on allele burden reduction to the recruitment of HSCs tocell cycling which in the long run shall lead to the depletion of MPN stem cells. [149][275] Knockout experiments demonstrated that the production of cytokines in MPN relies mainly on STAT3 signaling and that the phenotype of STAT3 knockout could be emulated by pharmacological JAK inhibition. [115] Ruxolitinib, a JAK1/2 inhibitor, is already used in the treatment of PMF. Although launched as an inhibitor targeting the JAK2 mutation, clinical trials showed that response was independent of the patients’ underlying driver mutation. As Ruxolitinib has been observed to dampen production of a series of interleukins, it has been proposed that the clinical effect may rely more on the general reduction of inflammatory signaling [307][413]

Analyzing the change in the JAK2푉 617퐹 allele burden of the RESPONSE study cohort, Vannucchi et al. could show that those who had received Ruxolitinib developed greater reduction in allele burden than those who had received best available treatment, and that patients with greater reduction allele burden also showed greater reduction in spleen size. Hematological parameters, though, do not correlate with allele burden

18 3 Introduction

reduction. [408] Recent five year analyses of the original COMFORT-I and COMFORT-II study populations could show that spleen size reduction achieved with Ruxolitinib could be maintained over long periods and that the Ruxolitinib groups still show superior survival rates which was interpreted as a sign that Ruxolitinib positively influences that natural course of the disease. [414][147][299]

Unlike cytokines, eicosanoids are not yet therapeutically targeted in MPN though there is evidence about their implication in their pathogenesis. The activity of 12/15-lipoxygenase (12/15-LOX) seems to prevent MPN through 13-(S)-Hydroxyoctadecadienoic acid (HODE) and 15-(S)-Hydroxyeicosatetraenoic acid (HETE) which cause a release of reactive oxygen species (ROS) and influence wnt/β-catenin-singaling via Phosphoinositide 3-kinase (PI3K) and Interferon regulatory factor 8 (IRF8/ICSBP). HSCs of mice deficient in 12/15-LOX (called Alox15 ) displayed a shift from self-renewal towards proliferation and differentiation. Alox15 animals develop a MPN phenotype. [265][193]

A more recent study showed that 5-Lipoxygenase (5-LOX) exerts the contrary effect as 5-LOX is overexpressed in JAK2-mutant MPN and both genetic elimination as well as pharmacological inhibition with Zileuton were able to reverse MPN in JAK2푉 617퐹 - mutant mice. Inhibition of 5-LOX was shown to dampen both Proteinkinase B (PKB or AKT) as well as wnt/β-catenin signaling, with the latter being rather contradictory as in Alox15 mice decreased wnt/β-catenin signaling seemed to be related with the MPN phenotype. [68]

Lastly, Lau et al. found the prostacyclin receptor to phosphorylate STAT3 via G-

Proteins α and α16. Pharmacological inhibition of the receptor as well as STAT3 self impaired the differentiation of Human Erythroleukemia (HEL) cells in vitro. In view of STAT3’s core role in promoting inflammation in MPN, it seems meriting to further dissect the effect of prostacyclin blockage in established animal models ofMPN. [216]

19 3 Introduction

3.2.2 Polycythemia vera

Although the disease’s hallmark feature of plethora - the surplus of blood - had been described by medical scholars as early as Hippocrates [76] or Maimonides [318], it was not until 1892 that Henri Vaquez concisely described the disease. [409] Since it was William Osler’s 1903 paper [289] that cast general interest to the topic, the disease was called Osler-Vaquez disease before the eponym was abandoned in favor of the name Polycythaemia vera which is used to this date. [372] The epithet vera can be understood as "primary", underscoring the difference from the much more common differential diagnosis of secondary polycythemia.

The three most important pathomechanisms leading to secondary polycythemia are a) hemoconcentration due to isotonic or hypertonic dehydration, b) hypoxemia leading to physiological increase in EPO production such as with high altitude, decreased pulmonary gas exchange [98] or hyperaffinitive hemoglobinopathies (Hb Tak, Hb Sherwood) [13] and c) the increase of other endocrine stimuli such as hyperthyroidism [85], acromegaly [300], hyperandrogenism due to hormonal therapy [97] or paraneoplastic EPO production [276].

As a consequence of hyperglobuly, polycythemia patients may present with a vast array of symptoms of microcirculatory disorders, such as transient ischemic attacks, tinnitus or scotoma. The classic presentation of PV, though, is secondary erythromelalgia which refers to painfully burning distal extremities that are warm, red and congested. Erythromelalgia may be accompanied of acrocyanosis and/or Raynaud’s phenomenon, and can progress to gangrene of toes or fingertips. [263][197]

A clinical phenomenon that is not exclusive to but most commonly found in PV is aquagenic pruritus, where water of any temperature causes burning and itching sen- sations on the skin. [371] Even though the underlying pathophysiology of pruritus in MPN remains poorly understood, the results of Ishii et al. imply that MPN entails pruritogenic mast cell dysfunctionality. [164] Splanchnic vein thrombosis (SVT) occurs in 5-10% of PV patients. Conversely, MPNs represent the most common pro- thrombotic condition leading to SVT and 33% of all SVT patients test positive for JAK2V617. [157]

A recent quantitative review of European data published since 2000 estimates the incidence of PV between 0.4 and 2.08 per 105 person years. [274] A Norwegian study found a twofold increase in the incidence of PV over the last ten years, a finding that was mostly attributed

20 3 Introduction

to the progress in diagnostics. [312] Although it is difficult to approximate the prevalence of PV in the population due to incompatible calculations and standardizations [274], most studies assess PV as the most common of the three MPN entities and report male predominance. [257] In Europe, the mean age at diagnosis was found to range from 65 [316] to 74 [168] years.

According to a recent retrospective analysis the 5-year cumulative incidence of progression to blast phase disease (MPN-BP) in PV is 0.72% [352]; compared to 2.3% at 10 years, 5.5% at 15 years after diagnosis [390].Leading risk factors for transformation to secondary AML are advanced age, reticulin fibrosis, leukocytosis, splenomegaly, abnormal karyotype, TP53 or RUNX1 mutations as well as prior therapies with pipobroman, radiophosphorus and busulfan. [62] Upon leukemic transformation, the median survival was two months. [352]

The cumulative incidence of transformation from PV to secondary myelofibrosis ranges between 5-14%; [62] with advanced age, reticulin fibrosis, leukocytosis, splenomegaly and JAK2푉 617퐹 allele burden as main risk factors. [62] The median survival after transformation to post-PV myelofibrosis was found to be 8.1 years. [298] Leukocytosis of ≥ 1.5 × 104/µl has been identified as a risk factor for both venous thrombosis, leukemic transformation and worse overall survival. [122] PV itself lowers the patients’ life ex- pectancy with a relative survival rate of 0.64 10 years and 0.32 20 years after diagnosis. [160]

With the revision of the WHO criteria the diagnosis of PV has seen some changes which mainly constitute a reaction to the recent reappraisal of masked PV. Masked PV (mPV), earlier also referred to as "inapparent PV", represents a condition in which JAK2 mutational status as well as bone marrow morphology fulfill the PV diagnosis, but red cell mass surrogate parameters remain below the levels set out in the 2008 WHO criteria. Therefore, the threshold hemoglobin concentration has been lowered and hematocrit has been introduced as a surrogate parameter for red cell mass in the current edition.Even though the decision to lower the threshold remains controversial, prompting more hematological diagnostics in healthy individuals [331], the measure seems reasonable as mPV patients were found to suffer from a higher mortality compared to cases of overt PV. Whether mPV should be considered a distinct subentity remains subject to debate:

The one side argues that mPV is merely a prodromal or attenuated form of PV because a) increased mortality can be attributed to a higher rate of thromboembolic events

21 3 Introduction

due to the lack of treatment [237] and b) that, although mPV patients may clinically present with features resembling ET, mPV bone marrow histology is indistinguishable from overt PV [212]. The other side points out that mPV patients harbor an increased propensity for transformation to both overt myelofibrosis and leukemia compared to overt PV patients [24], which would merit a separate classification to demarcate not only a perhaps deviant pathophysiology, but most importantly a different clinical management.

In line with the increased effort to detect mPV, evaluation of the bone marrow histology has been upgraded from a minor to a major criterion of diagnosis. Even though it may seem anachronistic in the age of molecular diagnostics, recent studies underpinned the value of bone marrow histology as mPV features panmyelosis - an age-adjusted bone marrow hypercellularity involving all hematopoietic lineages that is pathognomonic for PV. [26] Lastly, the formation of endogenous erythroid colonies (EEC) is no longer a minor criterion as the method was deemed as too inaccessible, costly and unreliable for clinial practice. [25]

Table 2. 2016 WHO diagnostic criteria for PV Major Criteria 1. Hemoglobin (Hb): men >16.5 g/dL, women >16 g/dL or hematocrit (Hct): men >49%, women >48% or increased red cell mass 2. BM biopsy showing hypercellularity for age with trilineage growth (panmyelosis) including prominent erythroid, granulocytic and megakaryocytic proliferation with pleomorphic, mature megakaryocytes (differences in size) 3. Presence of JAK2 mutation Minor Criteria 1. Subnormal serum erythropoietin level Diagnosis requires either all three major criteria or the two first major criteria and the minor criterion

Until today there is no treatment available that has been shown to affect the course of the disease. Hence, the goal in PV treatment remains the prevention of thrombosis. Hematocrit levels should be kept below 45% using phlebotomy and all patients should receive 40-100 mg acetylsalicylic acid per day. Further treatment is risk stratified. (Tab. 3)

All high-risk patients should receive cytoreductive treatment with hydroxyurea (HU) being the first-line treatment. Besides its role as a second-line treatment in patients intolerant or refractory to hydroxyurea, IFN-α should be considered as an alternative

22 3 Introduction

Table 3. Risk stratification for Thrombosis in PV Low Risk < 60 years old and no history of thrombosis High Risk > 60 year old or history of thrombosis

first-line treatment especially in younger patients as it is neither genotoxic nor teratogenic and has been proven to reduce JAK2푉 617퐹 allele burden as mentioned in 3.2.1. The results of the latest RESPONSE II follow-up underpin the value of Ruxolitinib in patients refractory to HU without splenomegaly. [135] Busulfan is considered a last resort for mainly elderly patients. [120]

3.2.3 Essential Thrombocythemia

The description of Essential Thrombocythemia (ET) by Epstein and Goedel in 1934, make it the latest of the MPN entities to be formally characterized. The disease’s hallmark is thrombocytosis arising from a bone marrow with megakaryocytic hyperplasia and hyper- trophy. The monolinear increase of megakaryocytes is an important discriminator between ET and other MPNs. [387] Analogous to PV, essential should be read as "primary" and must be differentiated from the more prevalent reactive thrombocytosis which is secondary to a) decreased storage of thrombocytes in the spleen, b) chronic infections/inflammations and c) various forms of anemia. [427]

Microcirculatory disorders represent the most common symptom of ET. [263] Though thromboembolic events are the most common complication, both PV and ET can cause hemorrhage through acquired von Willebrand syndrome (avWS) in patients with ex- cessively high platelet counts. In the case of ET, young age, higher hemoglobin as well as JAKV617F were identified as additional risk factors for the development of avWS. [321]

Furthermore, ET’s 9-13% prevalence of SVT makes it the highest among the MPNs. Portal vein thrombosis is three times more common among MPN patients than the often described Budd-Chiari syndrome. Risk factors for MPN patients to develop SVT are younger age, female sex but most importantly JAK2푉 617퐹 . The fact that a) the prevalence (71-100%) of JAK2푉 617퐹 in MPN patients with SVT is significantly higher compared with the prevalence among MPN patients suffering other types of VTE and that b) JAK2푉 617퐹 confers a 54 times odds ratio for the risk of developing SVT, imply

23 3 Introduction

that the mutations effect cannot be solely mediated through its impact on theMPN phenotype. [157] One hypothesis is that the splanchnic venous endothelium may be altered as liver and spleen endothelial cells of SVT patients were found to harbor the JAK2푉 617퐹 mutation. [364][320]

The incidence of ET in Europe was found to range from 0.38 to 1.7 per 105. The average age at diagnosis ranged from 64.3 to 73 years. [274] Compared to the general population, ET patients have a relative survival rate of 0.64 10 years after the diagnosis and 0.44 20 years after the diagnosis. [160] The annual rate of progression to MPN-BP lies at 0.37% with equally short survival as for transformation from any other MPN entity. [274] Important risk factors for progression to blast phase disease are advanced age, anemia, extreme thrombocytosis, thrombosis, leukocytosis, reticulin fibrosis and TP53 or RUNX1 mutations.

The development of secondary myelofibrosis is linked to advanced age, anemia, leukocytosis, reticulin fibrosis, absence of JAK2푉 617퐹 , prior therapy with anagrelide and mutations of ASXL1. The cumulative incidence of post-ET myelofibrosis ranges from 0.8-4.9% at 10 years and 4-11% at 15 years after diagnosis. [62] The median survival of post-ET secondary myelofibrosis is not significantly different from those who transformed from PV. [298]

The WHO definition of ET has changed twofold: First, with the discovery oftheCALR mutation, a driver mutation can now be identified in the vast majority of patients. Therefore, clonality analysis has been downgraded to a minor criterion in spite of tis correlation with increased propensity for thrombosis [346] or disease transformation [249]. The same has happened with the disproof of reactive thrombocytosis; the relevance of both criteria is now refrained to cases of triple negative ET.

Second, the criterion of bone marrow morphology has remained largely unchanged yet for the first time it specifies the amount of allowed reticulin fibers. This can be seenasa reaction to the call for more stringent distinction between pre-firbrotic PMF and ET, as the prePMF patients suffer from shorter overall survival as well as increased propensity for leukemic transformation compared to ET patients. [400][325] It is noteworthy that transcription factor NFE2 - which will be addressed in 3.3.2 - has been shown to be differentially localized within erythropoietic cells and can help distinguishing between prePMF and ET. [18] An important tool for the clinical management of ET is the International

24 3 Introduction

Table 4. 2016 WHO diagnostic criteria for ET Major Criteria 1. Platelet count ≥ 450 × 109 /l 2. BM biopsy showing proliferation mainly of the megakaryocyte lineage with increased numbers of enlarged, mature megakaryocytes with hyperlobulated nuclei. No significant left-shift of neutrophil granulopoiesis or erythropoiesis and very rarely minor (grade 1) increase in reticulin fibers 3. Not meeting WHO criteria for BCR-ABL1+ CML, PV, PMF, MDS or other myeloid neoplasm 4. Presence of JAK2, CALR or MPL mutation Minor Criteria 1. Presence of a clonal marker (e.g. abnormal karyotype) or absence of evidence for reactive thrombocytosis ET diagnosis requires meeting all four major criteria or first three major criteria and one minor criterion

Prognostic Score of thrombosis in World Health Organization – essential thrombocythemia (IPSET). In the original version from 2012, age, thrombosis history, JAK2푉 617퐹 and the presence of cardiovascular risk factors (defined as diabetes, hypertension and active tobacco use) were assigned point values. The sum of the score categorized patients in either the low, intermediate or high risk group. [23]

Even though an analysis of 1150 patients could show that CALR mutations do not effect the score’s validity [110], IPSET was revised in 2015. The aim was to account for the interdependence of risk factors with the result that the score is no longer calculated but each risk category represents a fixed set of risk factors, with age and previous thromboembolic events gaining in importance. At the same time, the "very low" risk category was introduced to account for the exceptionally low rate of thrombosis in young JAK2-wild type patients without previous thrombosis. [27]

In 2016, a different group of authors could validate the revised IPSET analysing 585 patients, yet they pointed out that studies with different cutoff ages and different definitions of cardiovascular risk would be needed. Most importantly, the authors state clearly that prospective studies are needed to evaluate the success of risk-adapted therapies. [142]

25 3 Introduction

Table 5. Risk stratification for Thrombosis in ET and treatment adapted from[396] and [119] Very Low Risk ∙ no history of throm- ∙ no cariovascular risk bosis factors: observation only ∙≤ 60 years old ∙ cardiovascular risk ∙ JAK2/MPL unmu- factor: acetylsalicylic tated acid (ASA) once daily

Low Risk ∙ no history of throm- ∙ no cardiovascular risk bosis factors: ASA once or twice daily ∙≤ 60 years old ∙ cardiovascular risk ∙ JAK2/MPL mutated factors: ASA twice daily

Intermediate risk ∙ no history of throm- ∙ no cardiovascular risk bosis factors: ASA twice daily OR Hydrox- ∙ > 60 years old yurea + ASA once ∙ JAK2/MPL unmu- daily tated ∙ cardiovascular risk factors: Hydroxyurea + ASA once daily

High Risk ∙ history of thrombosis ∙ arterial thromboem- OR bolism: Hydroxyurea + ASA twice daily ∙ > 60 years old + JAK2/MPL mutated ∙ venous thromboembolism: Hydroxyurea + systemic antico- agulation (+ ASA once daily with JAK2/MPL mutated OR cariovascular risk factors) 26 3 Introduction

Platelet aggregation inhibition is contraindicated in patients with prior history of gas- troduodenal ulcers as well as platelet counts ≥ 106/µl. The latter group is at risk of developing avWS, requiring a Ristocetin cofactor test before setting in any platelet inhibition. In patients that are refractory or intolerant to HU treatment, Anagrelide as well as IFN-α are second line treatments. Unlike in Germany, Anagrelide is a first-line option for the treatment of high risk patients in both Austria and Switzerland. [119] In spite of a study from 2005 demonstrating Anagrlide’s inferiority to HU [145], Anagrlide just like IFN-α may be preferred in younger patients for their lack of genotoxicity and secondary malignancies. [119]

3.2.4 Primary Myelofibrosis

The first description of myelofibrosis was made by Gustav Heuck in 1879 who noted similarities to CML but differentiated the conditions by virtue of extramedullary hematopoiesis and osteosclerosis. In 1907, Herbert Assmann published a similar case of what he called Osteosclerotic Anemia, leading to the disease’s eponym Heuck-Assmann syndrome.[387]

In the early (often called prefibrotic) phase of primary myelofibrosis (PMF) the bone marrow is hyperplastic, yielding a blood count dominated by granulocytosis with left shift and/or thrombocytosis. In this phase, the disease clinically resembles CML or ET and hematopathology is the key in order to establish the right diagnosis as prePMF features first of all granulocytosis while megakaryocytes maintain regular dimensions. [400] As discussed above, hematopathology can be misleading though, and considerable efforts are being made to improve differentiation in this critical phase.

The later phase of the disease (sometimes referred to as classical) is characterized by bone marrow fibrosis. The exact pathophysiology leading to the fibrosis remains incompletely understood: One theory states that the malignant clone initiates a process mediated by Transforming Growth Factor-β (TGF-β) which involves osteoblasts as well as bone marrow fibroblasts and extracellular matrix. [221] A more recent theory focuses on the Gli1+ stromal cells that differentiate to pro-fibrotic myofibroblast under the influence in CXCL4 secreted by the MPN clone. [340]

The stress exercised by fibrosis and inflammation mobilizes HSCs that home toal- ternative niches - preferably in spleen and liver - eventually causing extramedullary hematopoiesis (EMH). [162] EMH, in turn, is considered to be the pathophysiological

27 3 Introduction

Table 6. 2016 WHO diagnostic criteria for PMF Major Criteria prePMF overt PMF 1. Megakaryocytic proliferation and 1. Megakaryocytic proliferation and atypia, without reticulin fibrosis > grade atypia accompanied by either reticulin 1, accompanied by increased age-adjusted and/or collagen fibrosis (grade 2 or 3) bone marrow cellularity, granulocytic proliferation and often decreased erythropoiesis 2. Not meeting WHO criteria for BCR-ABL+ CML, PV, ET, MDS or other myeloid neoplasm 3. Presence of JAK2, CALR or MPL mutation or in the absence, the presence of another clonal marker or absence of evidence for reactive bone marrow fibrosis Minor Criteria 1. Presence of one or more of the following confirmed in two consecutive determinations: Anemia not attributed to a comorbid condition Leukocytosis ≥ 119/l Palpable splenomegaly LDH level above upper limit of the institutional reference range only overt PMF: leukoerythroblastosis Both prePMF as well as overt PMF diagnosis require all three major criteria and at least one minor criterion. substrate of hepatosplenomegaly; though mild forms of organomegaly may also be observed in other MPN entities in which cases the underlying mechanism is not EMH, but rather congestion. [228] Under these conditions hematopoiesis is ineffective, with cells manifesting both quantitative as well as qualitative defects; myelophthisic anemia, thrombopenia, leukoerythroblastosis and poikilocytosis with dacrocytes can be observed. [399]

PMF is the MPN entity with the highest propensity of leukemic transformation with a risk of about 20% 10 years [390] and 40% 15 years [25] after diagnosis. Due to the much higher transformation rate in myelofibrosis, it has been conjectured that most cases of PV or ET first transition through post-PV/ET-MF before progressing to MPN- BP. [259]

Finally, the combination of ineffective hematopoiesis and inflammatory signaling yields fatigue, cachexia, fever, myalgia, osteoalgia, bleeding and susceptibility for infection. [413][239] PMF patients, thus, not only bear the least favorable prognosis of all MPN entities with a relative survival ratio at 10 years of 0.21 and 0.06 at 20 years after

28 3 Introduction

diagnosis [274], but also feature the most negatively affected quality of life. [146] In Europe, 0.3 per 105 inhabitants develop PMF each year, at an average age ranging from 69 to 76. [274]

Since PMF bears the risk for poor clinical outcome, a series of prognostic classifications have been developed in order to stratify patients and offer them risk adapted treatment. The first scale of major importance was the International Prognostic Scoring System (IPSS) which gave for each of the following criteria one point:

∙ age > 65 years

∙ constitutional symptoms

∙ hemoglobin < 10 g/dl

∙ leukocytes > 2.5 ×1010/l

∙ circulating blasts ≥1%

Based upon the total score patients were discriminated into four distinct risk groups: High with ≥ 3 points, Intermediate 1 with 2 points, intermediate 2 with 1 point and low if the patient does not fulfill any of the criteria. The IPSS score is assessed only once based onthe patient’s values at the time of diagnosis with prospective median survival ranging from 95 months for low risk to 27 months for high risk patients. [63]

In order to assess the patients survival dynamically over the course of the disease, the DIPSS was developed. The criteria and the four-tier classification are the same as for IPSS but the presence of anemia yields two points instead of one. Since patients younger than 65 with a median survival prognosis of less than five years can benefit from allogenic stem cell transplantation (alloSCT) [22], an age-adjusted version (aaDIPSS) was introduced that valued only the leukocytosis criterion with one point while the three other criteria each yield two points. [297]

DIPSS was further enhanced with more criteria that were found to correlate with survival in PMF; for that matter the four DIPSS risk groups themselves were assigned with zero to three points, respectively and the criteria "unfavorable karyotype", platelets < 105/µl and transfusion dependence were each assigned one point. The resulting DIPSSplus risk categories were again low (0 points), intermediate-1 (1-2 points), intermediate-2 (3-4 points) and high (5-6 points). The DIPSSplus score has been shown to correlate with the risk for leukemic transformation: Patients in the low group had a 6% risk after 5 years

29 3 Introduction

and 12% risk after 10 years, whereas patients in the high group had risks of 18% and 31% at respective time points. [121]

The Mutation-enhanced International Prognostic Scoring System 70 (MIPSS70) was published in 2018 and represents the latest attempt to augment the IPSS with criteria based upon established mutations as well as the latest revision of the WHO criteria. The score is validated for a patient collective younger than 70 years with the aim to improve prognostic precision among those who can undergo curative alloSCT. MIPSS70 assigns one point for:

∙ hemoglobin < 10 g/dl

∙ circulating blasts ≥ 2%

∙ fibrosis ≥ grade 2

∙ constitutional symptoms

∙ absence of CALR type 1 – like mutation

∙ presence of one high molecular risk (HMR) mutation1

A score of two was assigned to:

∙ leukocytes > 2.5 ×104/µl

∙ platelets < 105/µl

∙ two or more HMR mutations

The grading system was changed to three tiers: low risk equalling zero or one point, intermediate risk ranging from two to four and high risk from 5 to twelve points. The MIPSS70plus was developed in parallel and factors in the karyotype, just like the DIPSS- plus, while omitting transfusion dependance. [138] The latest update (MIPSS70+ 2.0) distinguishes karyotype risk even further, featuring five risk categories. [391]

As genetic markers gain more and more importance in diagnosis and prognosis of malignancy, Tefferi et al. proposed the genetically inspired prognostic scoring system which shall rely solely on molecular genetics. [398] Despite the obvious advantages, critics point out the lack of universal availability as well as the restricted possibility of dynamic assessment. [42]

1ASXL1, EZH2, SRSF2, IDH1/2

30 3 Introduction

Currently, the clinical management of MIPSS70+ 2.0 very low and low patients is observational. Antithrombotic medications are not standard even though the risk for thrombosis is comparable to that of ET patients [368]; it seems, though, that this risk is restrained to patients with mutated JAK2. [111]

Patients in MIPSS70+ 2.0 risk categories high and very high should be primarily evaluated for alloSCT. Those assigned to the intermediate risk group are best treated in clinical trials. All other forms of therapy are considered palliative: anemia can be managed with thalido- mide/lenalidomide + prednisone, androgens or danazol.

First line therapy for splenomegaly is 500 mg hydroxyurea. Second line treatment is Ruxolitinib, which may be administered independently of splenomegaly to alleviate constitutional symptoms. Drug-refractory splenomegaly is an indication for splenectomy, while radiotherapy is restricted to bone pain and non-hepatosplenic extramedullary hematopoiesis. [388]

3.3 Epigenetics

The term epigenetics was coined by Waddington in 1942 as he was trying to describe the - back then - elusive link between the static genotype and the phenotype changing during development. [415] Seven decades later, the term has evolved to describe any modification in gene expression that can be transmitted to a daughter cell but doesnot involve changes of the DNA sequence. Besides the extensive study in biological processes such as development, epigenetics has been received as a new avenue for understanding the emergence of disease. As epigenetic changes are responsive to environmental factors, epigenetics is seen as a keystone connecting the oftentimes disputed roles of genetics and environment in pathogenesis. [425]

The foundation in understanding epigenetic mechanisms was the discovery of chromatin as the eukaryotic genome packaging composed of DNA and protein and the insight that the degree of chromatin condensation correlates with transcriptional activity. [12] Today, it is understood that both components of chromatin are subject to modification, yet to different extent: Methylation of position 5 of cytosine in CpG dinucleotides is the only known DNA modification and it always represses transcription. [200][356] On the contrary, histone proteins have been shown to undergo a plethora of covalent modifications. Both DNA methylation as well as histone modification are intertwined and dependent on each other in order to orchestrate the epigenome. [167][317] The nucleosome represents the most

31 3 Introduction

Figure 5. Overview of covalent histone modifications and their known positions. Boxed part of the aa sequences denote the histone proteins’ globular domain. Adapted from [158]

32 3 Introduction

basic unit of DNA packaging: the globular domains of two of each, H2A, H2B, H3 and H4, form an octamer which is wrapped by 146 bp of DNA. The assembly and disassembly of the nucleosome structure by ATPases is referred to as nucleosome remodeling and regarded as a proper epigenetic mechanism. [32]

The histone N-terminus - referred to as "tail" - lacks a secondary structure and protrudes out of the nucleosome. It is here most covalent modifications are attached; mostly at the side chains of lysine or arginine. [14][344] As the effect of such modifications on the transcriptional activity are a function of both the attached molecule as well as the position on the histone tail, hypotheses that explain these effects simply through electrostatic interaction with the DNA have been supplemented by the histone code hypothesis. [373]

This model relies on proteins that appear in three different roles: enzymes "write" the code by covalently attaching specific residues at distinct positions on the histone, while "reader" proteins can bind to them specifically and exert their respective effect on chromatin structure and transcriptional levels. Lastly, these effects can be counteracted by "eraser" enzymes that cleave off residues. [185][446]

3.3.1 Histone Demethylases

Histone methylation occurs mainly on the tails of H3 and H4 and includes both arginine and lysine residues. Arginine residues can occur in mono-, and di-methyl form. Often they appear in close proximity to other modifications which has been interpreted as potential cross-talk. The lysine residue can be methylated in three steps yielding mono-, di- or tri-methyl lysine. [7] Unlike other modifications, methylation was for a long time thought to lack an "eraser", making it irreversible and static in nature. This view has changed since the discovery of histone demethylases (HDMs). [14]

There are three reaction mechanism involved in histone demethylation: PADI4 deiminates arginine residues to citrulline. It is debated, though, if PADI4 truly classifies as a HDM for two reasons: 1) PADI4 catalyzes the deimination of both methylated as well as unmethylated side chains. 2) Citrulline cannot be methylated again and serves itself as a distinct histone mark. [14][344]The second mechanism is flavine-dependent amine oxidation. Members of the KDM1 group (KDM referring to lysine demethylase) rely on this mechanism for the demethylation of di- or monomethylated H3K4 and H3K9. [81]

33 3 Introduction

The third mechanism is represented in the largest known group of histone demethylases: the Jumonji C (JmjC) domain-bearing histone demethylases. The JmjC domain was first described by Takeuchi et al. who named thegene Jumonji after a cross- like malformation of the developing neural plate upon its knock out (jumonji is the Japanese name of the cross-like character that signifies "ten" in various East Asian languages). [384]

Today, the JmjC domain-containing protein superfamily contains more than 10,000 members ranging from members that are active as redox enzymes to catalytically inactive members. [141] Structurally, the JmjC domain is described as an almost bar- rel shaped double-stranded β helix. The fact that the conserved HXD/EXnH motif coordinates both iron II (Fe(II)) and 2-oxoglutarate (also known as α-ketoglutarate) places JmjC domain containing proteins in the 2-oxoglutarate oxygenase class of cupin metalloenzymes. [254][93] Due to the domain’s reaction mechanism (Fig.6),

Figure 6. Reaction mechanism of Jmjc oxygenases. Adapted from [430]

JmjC domain containing HDMs are independent of a protonated -amine, enabling them to demethylate all three methylation stage in a stepwise manner. [248] Based on sequence similarity and precise domain architecture, these demethylases have been classified into six subgroups. (Fig.7) The subgroup KDM3 (also referred to as JHDM2 or JMJD1) consists of four proteins: KDM3A (TSGA/JHDM2A/JMJD1A), KDM3B(5qCNA/JHDM2B/JMJD1B), KDM3C(TRIP8/JHDM2C/JMJD1C) as well

34 3 Introduction

Figure 7. Phylogenetic tree of Jmjc domain-containing proteins. Proteins grouped by similarity of the catalytic site. Subgroups represented by colored lines. Adapted from [430]

as hairless. They share a zinc finger-like domain in the middle of their sequences asa common feature. [201]

JMJD1C In humans, the gene is located in 10q21.3 while it is located at 10 B5.1 in mice. In both species, it is composed of 30 exons. For the murine gene there are two annotated isoforms: isoform 1 represents the full length protein with 2530 aa, weighing about 282 kDa. Isoform 2 differs in the 5’ UTR and coding sequence compared to variant 1 which results in a shorter N-terminus and the lack of a single internal aa compared to isoform 1. The murine isoform 2 weighs about 261 kDa.

The human orthologs are more complex: Isoform a is the longest one with 2540 aa, resulting in a molecular weight of 284 kDa. Isoform c (also referred to as s-JMJD1C [434]) is encoded by RNA variants 3 an 4 that have a different 5’ UTR leading to translation initiation at a downstream in-frame start codon; the resulting protein has therefore a shorter N-terminus (2358 aa). The same phenomenon is observed for isoforms d (2252 aa) and f (2321 aa), while isoform e is translated from RNA variant 7 which lacks an in-frame exon in the 5’ coding region (2502aa). [286]

Data regarding the expression profile of different tissues are partially concordant: While the HPA RNA-seq shows the highest expression of human JMJD1C in the bone marrow [105] and the Illumina bodyMap2 trancriptome for white blood cells, another RNA

35 3 Introduction

sequencing study found the human cerebellum to be the most highly expressing tissue. [91] In mice, the ENCODE transcriptome identifies the central nervous system as the location of highest expression. [442]

The zinc finger-domain typical for members of the KDM3 subgroup is a TRI8H1 domain in case of JMJD1C. [181] It is closely followed by a TRI8H2 domain which is responsible for the interaction with the Thyroid hormone receptor (TR) β.[218] The putative nuclear receptor interaction site was bioinformatically narrowed to a region containing a LXXLL motif [181] which is a highly conserved sequence found in a large number of nuclear receptor coactivators. [333] The functionality of the interaction was established by luciferase assays and showed that it is not restricted to the TR: Cells transfected with androgen receptor, glucocorticoid receptor or thyroid hormone receptor beta showed a marked increase in luminescence when contransfected with JMJD1C; an effect that could be further enhanced by exposure to respective hormone analogs. [434] The question whether JMJD1C belongs to the enzymatically active or the inactive part of the JmjC domain-containing proteins is subject to debate which is elaborated in 6.1.

3.3.2 Role of Epigenetics in MPN

In recent years, research has demonstrated that besides the fundamental role of genomic instability [308], epigenetic alterations are not only involved in both cancer initiation as well as cancer progression [171], but oftentimes precede genetic alterations. [177] This has been demonstrated for MPNs where mutations in epigenetic modifiers can precede the JAKV617F mutation. [206]

Lundberg et al. demonstrated in a cohort of 197 MPN patients that about 36% of the patients had mutations in addition to the known driver mutations and that most of these mutations affected epigenetic regulators. The relevance of these aberations is reflected by the overall worse survival of these patients compared to the ones only carrying a driver mutation (see MIPPS scale in 3.2.4). [236]

The following section presents the most relevant mutated epigenetic regulators as well as the consequentially deregulated genes classified according to the categories proposed by Mascarenhas et al. [250]

36 3 Introduction

Category I - Gene Alterations leading to Epigenetic Deregulation of Ph-negative MPNs

TET2 Mutations of the TET (Ten Eleven translocation) protein family were first described in AML featuring t(10;11)(q22;q23). [287] Similar to JmjC domain-containing enzymes, TET proteins utilize Fe(II) and 2-oxoglutarate in order to hydroxylize methylcytosine to hydroxymethylcytosine. [383]

In addition to the germline variants described in 3.2.1, sporadic mutations are reported at an overall prevalence of about 13% in all MPNs; making it the third most common somatic mutation in MPN and the most frequently comutated gene along JAK2 in MPN. [236]. It is significantly more often mutated in푉 JAK2 617퐹 -positive than -negative cases (17% vs. 7%) as well as significantly more often mutated in patients aged 60 and older (23% vs 4%). [394]

Although the mutations’ actual effect on global DNA methylation is unclear [345], murine loss-of-function models demonstrate a disrupted differentiation with concordant expansion of the LSK stem cell compartment. [271]. In the context of JAK2푉 617퐹 -MPN, it has been demonstrated that comutated HSCs are characterized by an increased self- renewal capacity, resulting in competitive advantage and a more severe phenotype. [66] [175]

The phenotype of singular TET2 mutation is mirrored in humans: Mutations in TET2 are also recurrently found in ARCH [53] as mentioned in 3.2.1. ARCH, which is also reffered to as Clonal hematopoiesis of Indeterminate Potential, has a prevalence ranging from 9.5% to 14,5% in individuals older than 70 years and up to 16,5% for those 80 years and older [150]. It is suspected to create a fertile ground for JAK2푉 617퐹 MPN as it predisposes monoclonal hematopoiesis which is a hallmark of MPN. [175] More importantly, a TET2 mutation that precedes JAK2푉 617퐹 affects the transcriptional pattern in such a way that the clone can expand but the proliferation and terminal differentiation of progenitors is reduced, which was found to lead to less thromboembolic complications. [288] However, this phenotype seems to contradict the effects observed for TET2 germline mutations that were outlined in 3.2.1.

IDH1/2 Isocitrate dehyrogenase (IDH) yields 2-oxoglutarate by oxidizing isocitrate, which makes it a pivotal enzyme in the Krebs cycle. The enzyme exists in to isoforms that differ in their subcellular localization: IDH1 is located in the cytosol and peroxisome,

37 3 Introduction while IDH2 is located mitochondrially. IDH mutations had been described in AML before their discovery in MPN. [247] The association with AML is intriguing as IDH1/2 mutations occur at a mutational frequency of 21% in blast-phase MPN and 25% in post-PV/post-PMF AML [392], while Chotirat et al. identified a prevalence of only about 5% in chronic phase MPN. [70]

IDH mutations are neomorphic gain-of-function mutations as the enzyme can convert 2-oxoglutarate into 2-hydroxyglutarate. [84] As described above, TET2 relies on 2- oxoglutarate and in the context of IDH mutation TET2 catalysis is indeed reduced due to lack of 2-oxoglutarate. [436] Furthermore, IDH-mutant murine bone marrow shows a pheno- copy of TET2’s block of differentiation and expansion of the stem cell compartment. [109] Yet it is not only TET2 that is dependent on 2-oxoglutarate as outlined in 3.3.1: Chowd- hury et al. could show that 2-hydroxyglutarate inhibits HDMs of the JmjC family, [71] meaning that the the phenotype of IDH mutant disease may at least partially be conse- quential to the blockade of JmjC domain-containing HDMs.

After promising results from the first phase I/II trials of IDH inhibitor Enasidenib showing induction of myeloblast differentiation in AML [370], studies of primary patient cells and mouse models of JAK2/IDH double-mutant disease with Ruxolitinib/Enasidenib combination therapy show cooperative effects [255] that should prompt trials in double- mutant MPN patients.

EZH2 Enhancer of zeste homolog 2 (EZH2) is the catalytically active component of the histone methyltransferase PRC2 [251] which trimethylates H3K27. [57][360] About 13% of PMF patients harbor EZH2 mutations; with all known mutations thought to cause a loss of function. [102] Murine models of this loss-of-function led to defects in the B-lymphoid compartment while both stem cells as well as the myelopoiesis remained unperturbed. [374]

Contrary to the loss-of-function, an overexpression of EZH2 is documented in various carcinoma types as well as MPN; most often PMF. [107] The overexpression is, at least partially, attributable to the loss of microRNAs such as miR-101 [410] and suggests a role in disease progression via silencing of tumor suppressor genes. [440] Therefore, it is presumable that the EZH2 inhibitors on trial for B-cell lymphoma will eventually be tested in PMF. [140]

38 3 Introduction

ASXL1 About 8% of MPN patients [58] harbor loss-of-function mutations in this gene which is predicted to have dual activator/repressor functions depending on the cellular context. [113] ASXL1 owes its ambivalent roll to the fact that it serves as an enhancer protein to both the transcriptionally activating trithorax group (trxG) as well as the repressing Polycomb group (PcG) histone modifiers. [2] ASXL1’s important role in building these larger protein complexes is illustrated by the lack of ASXL1 resulting in decreased EZH2 recruitment, loss of transcriptionally repressive H3K27 trimethylation and increased HOXA expression. [1]

ASXL1 knockout mice are characterized by a defect in frequency of differentiation to both myeloid and lymphoid cells, yet they feature neither effects on hematopoietic stem cells nor the developement of myelodysplastic or leukemic phenotype. Hence, ASXL1 mutations alone are not sufficient to induce malignant transformation. [114]

Furthermore, ASXL1 mutations are found in the CD34+ cell population, supporting the principle of a primitive hematopoietic stem as the origin of the MPN clone and suggesting that the acquisition of ASXL1 mutations can occur early in the course of MPN pathogenesis. [58][236]

JAK2 JAK2’s canonical role in the JAK/STAT pathway and the hypothesis of MPN pathogenesis based upon it have been outlined in 3.2.1. Dawson et al. took the finding that 35% of the JAK2 regulated genes did not contain a STAT5 binding site as a point of departure to characterize an alternative pathway by which JAK2푉 617퐹 might affect hematopoiesis:

They observed that besides its common cytosolic localization, JAK2 can occur in the nucleus [87] in spite of its lack of any known nuclear localization signal. [136] Furthermore, they found JAK2 to phosphorylate H3Y41 which releases transcriptional repression by HP1α.[87]

In addition, Liu et al. demonstrated the phosphorylation of arginine methyltransferase PRMT5 by JAK2푉 617퐹 . The ensuing reduction of histone arginine methylation affects the gene expression pattern and the behavior of hematopoietic progenitor cells. [229]

39 3 Introduction

JMJD1C The transcription factor Nuclear factor erythroid 2 (NFE2) is implicated in the differentiation and maturation of the erythroid lineage by regulating the expression of key erythroid genes. [450][311] Besides its role as a transcription factor, NFE2 exerts epigenetic regulation as it induces H3 acetylation [169] as well as H3K4 methylation [186] of the β globin locus control region (LCR). NFE2 is a target gene of RUNX1 [421] and - just like RUNX1 - it is upregulated in MPN irrespective of the underlying driver mutation [130] as well as other polycythemia entities featuring increased signaling by Hypoxia inducible factor (HIF) [178].

In vitro models of NFE2 overexpression showed delays in erythroid maturation in step with an EPO-independent increase of erythrocyte production. [277][40] These findings are complemented by the observation of a MPN phenotype in transgenic mice overexpressing human NFE2 in their bone marrow. [182] Mutant NFE2 - which is observed in about 2% of MPN cases - was demonstrated to enhance the function of wild-type NFE2. [173] Chromatin Immunoprecipitation (ChIP) demonstrated that NFE2 binds to the JMJD1C locus and the binding’s functional relevance was underpinned by Luciferase assays and knock down studies.

In vivo, JMJD1C is overexpressed both in the bone marrow of transgenic mice expressing human NFE2 as well as MPN patient peripheral blood cells. The finding that the NFE2 promoter of PV patients show increased binding of JMJD1C as well as a reduced histone demethylation, prompted the hypothesis of an epigenetic feedback loop that maintains the increased NFE2 expression mentioned above. [302]

Category II - Individual Genes affected by Epigenetic Modification in MPN

PRV-1 CD117 is a glycosylphophatidylinositol-anchored protein expressed on the surface of human granulocytes whose expression increases upon sepsis, pregnancy and G-CSF stimulation. [166] In the case of PV, the mRNA levels of polycythemia rubra vera-1 (PRV-1) - the gene encoding CD117 - are increased, yet protein levels remain comparable to those of healthy controls. [199] The expression of CD117 is inversely related to the methylation of the gene’s C30 promoter site, which in turn was found to be inversely related to the JAK2푉 617퐹 allele burden. [166]

40 3 Introduction

SFRP1/2 The Wnt/β-Catenin pathway is essential for HSC maintenance and proliferation [429][161] and its overexpression has been demonstrated in JAK2푉 617퐹 positive and negative MPN. [235][124] Secreted frizzled-related proteins operate as negative regulators of the Wnt/β-Catenin pathway and have been shown to be inactivated in different hematologicaol malignancies. [172][230] Bennemann et al. detected the hypermethylation of SFRP-1 in 4%, SFRP-2 in 25% and SFRP-3 in 2% of MPN cases. Discordantly with the above mentioned evidence, SFRP-2 hypermethylation was significantly correlated with a JAK2푉 617퐹 genotype. [35]

CXCR4 CXC chemokine receptor 4 (CXCR4) is expressed on CD34+ HSCs in humans and functions as a receptor for stromal derived factor-1 (SDF-1; CXCL12). Together, CXCR4/CXCL12 are responsible for orchestrating the homing of CD34+ cells to the bone marrow niche. In PMF cells, the promoter of the CXCR4 gene has an increased propensity for DNA hypermethylation [39], which is thought to lead to abnormal mobilization of HSCs in the peripheral blood and finally resulting in EMH .[319]

41 3 Introduction

3.4 Aim

As outlined above, NFE2 has been demonstrated to contribute to the pathophysiology of myeloproliferative disease. Yet the cellular mechanisms through which it mediates these effects have remained elusive. As the importance of epigenetics and epigenetic modifiers in myeloid disease has been unraveled, NFE2’s regulation of the histone demethylase JMJD1C and JMJD1C’s overexpression in PV patients represent a new avenue in elucidating the pathogenesis of the NFE2 phenotype. As a consequence, the question lying at heart of this work is: Does the silencing of JMJD1C influence hematopoiesis?

Following approaches were chosen in order to address this question experimentally:

1. construction and validation of shRNAs reliably silencing human as well as murine JMJD1C which necessitates

a) the cloning of mammalian expression vectors carrying human or murine JMJD1C

b) the identification of western blot antibodies reliably detecting human and/or murine JMJD1C

2. lentiviral transduction of hematopoietic stem cells with the subsequent analysis of the resulting phenotype

42 4 Materials and Methods

4 Materials and Methods

4.1 Cloning

4.1.1 Restriction digest

Restriction endonucleases are bacterial enzymes that evolved to cut unmethylated DNA introduced into the cells and restrict its propagation, hence their name. Three features render restriction enzymes handy for targeted manipulation of DNA:

First, their activity can be maintained in vitro using adequate buffers, independent of the bacterial species they are derived from. Second, restriction enzymes recognize palindromic sequences, which makes it possible to anticipate where in a given DNA molecule restriction enzymes will cut. Third, restriction enzymes hydrolyze one phosphodiester bond in each strand of a restriction site. The hydrolyzed bonds can be directly opposite from each other, creating a so called blunt end. In other cases, the bonds can be symmetrically arranged around the center of the recognized palindromic sequence creating single stranded overhangs, called sticky ends. Restriction enzymes are used for two main purposes in cloning:

First, they allow a) the linearization of plasmids, which means the plasmid’s circular structure is severed at one point to allow the insertion of DNA and b) the excision of DNA fragments which can be inserted into DNA molecules, like linearized plasmids. For this purpose 2 - 5 µg of plasmid DNA were supplemented with 10× buffer needed for the respective enzyme(s). The solution was incubated with 10 - 50 U of required enzyme(s) for at least 90 min at the temperature recommended by the manufacturer.

Second, digestion of DNA molecules with restriction enzymes creates a characteristic set of fragments. The analysis of the fragments’ length reveals whether DNA was gained or lost in experiments as foreseen. In a 1,5 ml reaction tube, a sample of interest containing 100 - 500 ng DNA was used to dilute the enzyme’s appropriate buffer from× 10 to 1× concentration. After addition of 5 U of the desired enzyme, the reaction was incubated at the enzyme‘s respective temperature for at least one hour. After every digestion, samples were subjected to agarose gel electrophoresis.

43 4 Materials and Methods

Table 7. List of plasmids used for cloning Name Description Source pBC-SK+- Cloning vector con- Kasuza DNA Re- mKIAA1380 taining the murine search Institute, JMJD1C cDNA Chiba, Japan mKIAA1380 pCMV-Kan/Neo- Mammalian ex- OriGene, USA mJMJD2C pression vector containing a murine JMJD2C cDNA pCR-XL-TOPO- Cloning vector con- OriGene, USA BC143722 taining the human JMJD1C cDNA BC143722 pLeGo-iG-hU6 Lentiviral vector used Dr. S. Schwemmers for shRNA transduction

Table 8. Restriction exzymes by New England Biolabs, USA Enzyme Restriction Optimal Re- Optimal Cat. No. site 5’-3’ action Buffer Temperature [∘C] ApaI GGGCC’C CutSmart 25 R0114 AsiSI GCGCAT’CGC CutSmart 37 R0630 BssHII G’CGCGC CutSmart 50 R0199 EcoRI G’AATTC NEBuffer 2.1 37 R0101 EcoRV GAT’ATC NEBuffer 3.1 37 R0195 HindIII A’AGCTT NEBuffer 2.1 37 R0104 HpaI GTT’AAC CutSmart 37 R0105 MluI A’CGCGT NEBuffer 3.1 37 R0198 NcoI C’CATGG NEBuffer 3.1 37 R0193 NotI GC’GGCCGC NEBuffer 3.1 37 R0189 PmlI CAG’GTC CutSmart 37 R0532 StuI AGG’CCT CutSmart 37 R0187 SwaI ATTT’AAAT NEBuffer 3.1 25 R0604 XhoI C’TCGAG CutSmart 37 R0146

44 4 Materials and Methods

4.1.2 Removal of single stranded overhangs

In case a restriction digest yields overhangs that are hindering the following steps in the cloning procedure, one can use the so called Klenow fragment, which is the larger fragment resulting from a digest of E. Coli DNA polymerase I with subtilisin protease. [267] The Klenow fragment retains the 5’-3’-polymerase activity as well as a weak 3’-5’-exonuclease activity, but lacks the 5’-3’-exonuclease activity. Hence, it can be used to insert comple- menting bases, bringing the two strands to an equal length.

For that purpose, DNA was be dissolved in 1× NEBuffer (New England Biolabs, USA) or T4 Ligase buffer (New England Biolabs, USA) and supplemented with33 µM of each dNTP. After adding one unit of Klenow fragment (New England Biolabs, USA) per µg of DNA, the solution was incubated at 25∘C. After 15 min the reaction was stopped by heating the solution to 75∘C for 20 min.

4.1.3 Vector dephosphorylation

Digesting a vector with either one single enzyme or two enzymes that leave the same overhang creates a situation where the two loose ends of the vector can easily reconnect, thus recircularizing the plasmid without taking up the desired insert. In order to prevent this, the 5’-phosphate groups that were exposed by hydrolysis of the phosphodiester-bond are removed using calf intestine phosphatase (CIP). In this state, the two ends of the vector can no longer be ligated. Since the insert is not subjected to dephosphorylation, its two ends can still be ligated to the vector’s ends.

Dephosphorylation reactions were carried out by incubating the desired amount of DNA in CutSmart buffer (NEB, USA) with 1 µl CIP (NEB, USA) for 1 h at 37∘C. After the reaction the product was purified using gel electrophoresis.

4.1.4 Ligation

Ligation refers to the process when the two ends of DNA are in close proximity and then covalently linked to each other as the enzyme ligase catalyzes the formation of a phosphodiester bond. When cloning, one can use the reaction to either insert a desired piece of DNA into a vector or to simply recircularize a vector after ridding it of a desired

45 4 Materials and Methods

piece of DNA. Due to thermodynamics, however, the first scenario is a lot less likely than the latter:

In a reaction mix with linearized vector and DNA pieces to be inserted the majority of the vector will simply ligate to itself since the likelihood is higher that one end of a vector will encounter the other end than one of the freely diffusing DNA pieces.

There are different strategies to counter this pitfall: One is to employ a cloning strategy that comprises compatible cohesive ends on the vector and the insert. Base pairing favors the interaction between vector and insert, allowing for successful insertion. Whenever blunt end cloning is inevitable, vector dephosphorylation as described above is a viable option. However, the best way to increase the probability of encounter between vector and insert is to add a vast excess of insert. For insert of the length of common cDNAs, a ratio of 1:3 to 1:5 of vector to insert is recommended since even higher amounts of insert obstruct the reaction, thus lowering the yield. The ratio should be adjusted, though, to account for the length of the insert, which is particularly important in the case of minute fragments such as shRNAs. The cloning of shRNAs described herein was carried out with vector-insert-ratios of 1:10. Using the following formula the needed amount of insert can be calculated:

푚푎푠푠(푣푒푐푡표푟)[푛푔] × 푙푒푛푔푡ℎ(푓푟푎푔푚푒푛푡)[푏푝] 푚푎푠푠(푓푟푎푔푚푒푛푡)[푛푔] = 5 × 푙푒푛푔푡ℎ(푣푒푐푡표푟)[푏푝]

50ng of vector were used in all ligation reactions, which is why the formula can be simplified to:

250푛푔 × 푙푒푛푔푡ℎ(푓푟푎푔푚푒푛푡)[푏푝] 푚푎푠푠(푓푟푎푔푚푒푛푡) = 푙푒푛푔푡ℎ(푣푒푐푡표푟)[푏푝]

Ligation reactions were carried out by combining the respective amounts of vector and insert DNA together with 1 µl ligation buffer (New England Biolabs, USA) and1 µl T4 DNA ligase (New England Biolabs, USA). The mix was brought to a total volume

of 10 µl by adding the difference in ddH2O. To easily estimate the efficiency of the ligation after bacterial transformation (see 4.3), one can prepare two further reaction mixes:

46 4 Materials and Methods

First, a religation control which contains all the components of the ligation reaction except for the insert DNA. Comparing the amount of colonies on the plate with bacteria which were transformed using the ligation reaction to the amount of colonies on the plate with bacteria transformed with this religation control allows to estimate how many colonies actually harbour the insert: The fewer colonies grow on the plate of the religation control, the higher is the chance that the transformed bacteria carry the insert.

The second control is the digestion control. This reaction contains everything except for the enzyme and the insert which means that colony formation after transformation with this reaction’s product is indicative of inefficient digestion prior to the ligation reaction. If the digestion was effective, the reaction mix is void of circular plasmid DNA and bacteria cannot be transformed. Thus, no growth should occur on a selective plate. In case growth on this plate occurs, one must make sure that a method of detection is used that allows to discriminate between the original vector and the vector carrying the insert (e.g. PCR with one primer binding within the insert and one within the vector backbone).

Finally, all reactions were incubated at 14∘C overnight. Afterwards, reaction products were stored at 4∘C until used for transformation the same day.

4.2 Design of shRNA

The cellular mechanism of RNA interference (RNAi) relies on small anti-sense guide RNAs (gRNA) that direct protein complexes to target RNAs. There are different sources of gRNAs each with a different biogenesis; the most important being canonical microRNAs (miRNAs). They are transcribed from the genome by RNA Polymerase II as complex, hairpin-structured pri-miRNAs. [328][19] These are cropped by the microprocessor complex of Drosha and DGCR8 resulting in roughly 70 bp pre-miRNA that are still double-stranded and ready for nuclear export. [445] Once in the cytoplasm, pre-miRNA are trimmed to a length of 21-30 bp by Dicer [36] and loaded onto Argonaute [443] to form the microRNA-induced silencing complex (miRISC). Lastly, the passenger strand is shed, leaving the guide strand ready to bind the target sequence via complementary base pairing. [258]

RNAi is conserved from yeast to humans and is implicated in different RNA regulatory functions such as viral innate immunity and post-transcriptional gene silencing (PTGS).

47 4 Materials and Methods

[163] About 60% of the human genome are targeted by miRNAs [116][204], highlighting the mechanism’s importance for gene regulation. Silencing can be mediated either by mRNA cleavage, deadenylation or translational repression of the bound target mRNA. [129][304][435]

Besides its physiological role, RNAi can be utilized as a gene technological tool to knock down the expression of a gene of interest. There are different strategies geared towards different experimental setups: For transient gene silencing in cell culture, cells can be transfected with short interfering RNAs (siRNA).[100] For long lasting silencing, vectors that stably express short hairpin RNA (shRNA) can be integrated into the genome.

ShRNAs are small stem-loop structures that resemble pre-miRNAs, which means that they skip parts of the physiological miRNA processing. Unlike endogenous miRNA that are transcribed by RNA polymerase II, shRNAs are dependent on RNA polymerase III due to their small size and lack of regulatory sequences. [49][290][256] These two factors can lead to insufficient supply of mature targeting complexes or cellular toxicity dueto oversaturation of the processing machinery. [253][43][60][137] To this end, shRNAs are increasingly integrated in endogenous pri-miRNA structures that shall ensure constant and stable silencing. [74]

In order to study the effects of JMJD1C expression, shRNAs targeting either human or murine JMJD1C transcripts were designed using theBLOCK-iT RNAi Designer. (Lifetechnologies, USA) 2

Table 9. Settings used for BLOCK-iT RNAi Designer Targeted Region ORF; not 5’- or 3’-UTR Species either human or murine minimal GC content 35% maximal GC content 55% Vector pENTR/hU6 Orientation sense-loop-antisense

Following the algorithm outlined in [108], all known transcripts variants were fed into the algorithm and those sequences with the highest rank for covering all transcript variants were chosen and manually counter-checked to match a maximum of the following criteria:

2http://rnaidesigner.lifetechnologies.com/rnaiexpress/insert.do

48 4 Materials and Methods

Table 10. Criteria for potent shRNA according to Li et al. [225] Starting with G or GG Strong preference for AU at positions 3–7 Relatively GC-rich at positions 14–16 Strong preference for AU at positions 17-19 Strong preference for AU at positions 3, 6, 13 and 18 Preference for AU at position 9 Preference for GC at position 11

shRNA sequences, along with their reverse complements, were pasted into the hairpin cassette which is described in 4.13.1. Oligonucleotides representing either strand of the cassette were ordered from Eurofins, Germany and annealed using the following protocol:

Table 11. 10× Annealing Buffer Reagent Manufacturer Cat.No. 20mM Tris pH 7.5 Roth, Germany 4855.3 10mM MgCl2 Sigma-Aldrich, Germany M2670 50mM NaCl VWR, Germany 27810.295 1mM DTT Roth, Germany 6908.2

20 µl of each the sense as well as the anti-sense oligonucleotide along with 10 µl of 10 ×

annealing buffer as well as 30 µl of ddH2O were combined in a Eppendorf tube, heated to 95∘C and left to slowly cool to RT.

4.3 Bacterial transformation

Reagents, solutions and kits Manufacturer Cat.No. Agar Fluka, Germany 5040 Ampicillin Sigma-Adrich, Germany A 2804 Chloramphenicol Sigma-Adrich, Germany C0378-5G Kanamycin Sigma-Adrich, Germany K4000 Magnesium chloride Sigma-Aldrich, Germany M2670 Potassium chloride Sigma-Aldrich, Germany P9541 Sodium chloride VWR, Germany 27810.295 Tryptone Sigma-Aldrich, Germany T7293 Yeast Extract Roth, Germany 2363.3

49 4 Materials and Methods

One Shot Top10 Chemically Competent E.coli (# C4040-10, Invitrogen, Germany) were used for cloning and amplification of mammalian expression vectors. Genotype: F- mcrA ∆(mrr-hsdRMS-mcrBC) 휑80lacZ∆M15 ∆ lacX74 recA1 araD139 ∆(araleu)7697 galU galK rpsL (StrR) endA1 nupG

One Shot Stbl3 Chemically Competent E.coli (# C7373-03, Invitrogen, Germany) were used for cloning and amplification of lentiviral expression vectors. Genotype: F-mcrB mrrhsdS20(rB-, mB-) recA13 supE44 ara-14 galK2 lacY1 proA2 rpsL20(StrR) xyl-5 휆-leumtl-1

Transformation describes the process of endowing bacteria with genetic material. In case this genetic material is naturally foreign to this species, the resulting organisms are referred to as transgenic.

Since bacteria do not take up DNA readily, their murein wall needs to be permeabilized. Af- ter such a treatment, bacteria are deemed as chemically competent. Owing to DNA’s polar chemical properties, a strong stimulus is needed for it to pass through the cell membrane. Such a stimulus can either be an electric or a heat shock.

In both cases, it is hypothesized that within the cell membrane minute pores transiently open, allowing small DNA molecules to enter the cell before closing again. Nevertheless, the likelihood of a bacterium to absorb a DNA molecule remains rather low. To distinguish between the few bacteria that have successfully taken up the desired genetic material and the majority of untransformed bacteria, vectors carry antibiotic resistance genes, which shall only allow transformed bacteria to grow in or on selection media containing the respective compound.

Table 12. Composition of Lysogeny Broth (LB) Component Amount [g] Yeast Extract 5 Tryptone 10 NaCl 10 add ddH2O to 1l and autoclave

for LB Agar, additional 15 g Agar, add ddH2O to 1l autoclave and add antibiotic to the cooled medium before pouring into 10 cm petri dishes

As a first step, competent E. Coli were thawed on ice. After adding either 5 ng of plasmid DNA or 5 µl of ligation or control reaction to a 50 µl aliquot of bacteria, the suspension

50 4 Materials and Methods

Table 13. Composition of SOC medium Component Concentration Tryptone 2% Yeast Extract 0.5% NaCl 10mM KCl 2.5 mM autoclave at this point MgCl2 10mM Glucose 20mM was incubated for 30 min on ice. Meanwhlie, an aliquot of SOC medium was preheated to 37∘C using a Thermomixer (Eppendorf, Germany). Using a water bath heated to 42∘C, the Top10 cells were shocked for 30 sec, while Stabl3 cells were shocked for 45 sec. After the shock, the cells were immediately chilled on ice for 2 min before adding 250 µl of warm SOC medium. The suspension was then incubated in a Thermomixer at 350 rpm for 1 h. Depending on the expected transformation efficiency, 50-150 µl of bacterial suspension were spread on a prewarmed LB agar plate containing the appropriate antibiotic using either glass beads or a Drigalski spatula. After air-drying the plates for 10 min, they were incubated overnight at 37∘C.

4.4 Agarose gel electrophoresis

Reagents, solutions and kits Manufacturer Cat.No. 2log DNA ladder NEB, Germany N3200 Acetic acid, glacial Roth, Germany 3738.4 Agarose PeqLab, Germany 351020 Bromphenol Blue Roth, Germany A512.1 Ethidium bromide Roth, Germany 2218.1 Ethylenediaminetetraacetic acid (EDTA) Serva, Germany 11280 Glycerol Fluka, Germany 49780 Tris Roth, Germany 4855.3 Xylene Cyanol FF Serva, Germany 38505

For separation of DNA fragments by size agarose gel electrophoresis was used. The total volume of the agarose solution was chosen depending on the size of the needed Easy Cast gel casting chamber (Owl Separation Systems, USA). Agarose was dissolved by heating the mixture in the microwave until it boiled. The evaporated water was replaced with

ddH2O to maintain the solution’s desired ionic strength.

51 4 Materials and Methods

Table 14. 50× Tris-Acetate EDTA (TAE) buffer Component Amount] Tris 242 g Glacial acetic acid 57.1 ml 0.5 M EDTA 100 ml add ddH2O to 1l

Table 15. 10× Loading Dye Component Amount [% w/v] Bromphenol Blue 0.42 Xylene Cyanol FF 0.42 Glycerol 50

Table 16. Agarose concentration as a function of DNA fragment length Length of the fragment of in- %(w/v) [Agarose in 1× TAE terest [kb] buffer] < 1 2 1-5 1 > 5 0,5

Once the agarose solution had cooled down for handling, 1 µl of 1% ethidium bromide solution was added to every 100 ml of volume. The solution was poured into the chamber and left at room temperature for solidification. The respective amounts× of10 loading dye were diluted in the samples. The chamber with the gel was immersed in 1× TAE and the prepared samples loaded into the wells. For size estimation a 2log-ladder (New England BioLabs, USA) was used. After loading, voltage was applied using the following formula: 푉 표푙푡푎푔푒[푉 ] = 5 × electrode distance[푐푚]

Since the wells are placed facing the cathode, the electric field forced the negatively charged DNA through the agarose gel towards the anode. As the speed at which the fragments migrate through the agarose resin logarithmically depends on their length, fragments of different length separate during migration. Once the fragments were sufficiently separated for the desired purpose, the resulting bands could be visualized: the added ethidium bromide fluoresces intensively under UV light as it intercalates nucleic acid strands. Results were documented using an Intas Gel Imager (Intas, Germany).

52 4 Materials and Methods

4.5 Gel purification and extraction

In case of preparative gels, the desired DNA fragment needed to be extracted from the agarose gel to proceed with cloning. Using a 365 nm UV gel table, the corresponding band was sharply, yet swiftly excised in order to reduce excessive irradiation damage. The piece of gel was placed in a 1.5 ml reaction tube (Greiner, Germany) and weighed using a fine scale. Depending on the weight, the buffers provided in the QIAGEN Gel Extraction kit were added, following the manufacturer’s protocol.

4.6 Nucleic acid quantification

Quality and quantity of DNA or RNA were determined using a NanoDrop spectrophotometer (Thermo Fisher Scientific, USA). About 1.5 µl of a sample were illuminated with UV

light of 230, 260 and 280 nm wavelength and the sample’s respective absorbances (퐴230,

퐴260 and 퐴280) recorded. Using the 퐴260, nucleic acid concentrations were calculated with following formulas:

푛푔 푛푔 DNA concentration[ ] = 퐴 × 50 × dilution factor 휇푙 260 휇푙 푛푔 푛푔 RNA concentration[ ] = 퐴 × 40 × dilution factor 휇푙 260 휇푙

The ratios of 퐴260/퐴230 and 퐴260/퐴280 were used as indicators for sample purity. DNA was considered pure when 퐴260/퐴230 ≈ 1.8. RNA was considered pure when 퐴260/퐴230 ≈

2.0. For both DNA and RNA holds 1.8 ≤ 퐴260/퐴280 ≤ 2.0.

4.7 Plasmid DNA preparation

There is a plethora of protocols available in order to purify plasmid DNA from E. Coli cultures. Though they vary greatly in terms of speed, purity and yield, the basic principle remains the same: after separating the bacteria from their culture medium, detergents are used to release intracellular contents and denature protein. Acids are used to precipitate genomic DNA together with proteins and phospholipid membranes, which allows for easy separation from the plasmid containing supernatant.

53 4 Materials and Methods

Reagents, solutions and kits Manufacturer Cat.No. 14 ml round bottom snap cap tube Beckton-Dickinson, USA 352057 Aqua ad injectabilia Fresenius Kabi, Germany 13101221 Cryo.S round bottom tubes Greiner Bio-One, Germany 122263 Dimethylsulfoxide (DMSO) Sigma-Aldrich, Germany D2438 Ethanol Sigma-Aldrich, Germany 32205 Guanidinium thioisocyante Roth, Germany 0017.2 Isobutanol Roth, Germany 6772.2 Plasmid Maxi Kit QIAGEN, Germany 12165

4.7.1 Mini preparation of plasmid DNA

In order to isolate small amounts of plasmid DNA for restriction digestion or PCR, 5 ml of LB Broth containing the plasmid’s respective antibiotic were inoculated with material from a colony which was grown on a selective LB agar plate. The broth was kept at 37∘C for 16-18 h under constant 200 rpm agitation. Bacterial growth was estimated by naked eye. Cultures deemed positive for growth were subsequently used. 1.5 ml suspension from each culture were transferred to reaction tubes (Greiner Bio-One, Germany) and centrifuged at 14 500 g in a table top centrifuge (Heraeus, Germany) for 30 sec. Upon removal of the supernatant, the pellet was resuspended in 50 µl Buffer P1. After addition of 100 µl of Buffer P2 containing RNase, the tubes were inverted five times and incubated for 5 min at RT. Upon addition of75 µl of Buffer P3, samples were immediately vortexed and subsequently centrifuged for 10 min at 13 000 rpm and 4∘C. The supernatant was transferred to new 1.5 ml reaction tubes and 300 µl of 6 M guanidinium thioisocynate added. After addtition of 400 µl isobutanol samples were thoroughly mixed and again centrifuged for 10 min at 13 000 rpm and 4∘C. The supernatant was carefully aspirated and the plasmid DNA pellet washed with 300 µl of 70% ethanol, vortexed and again centrifuged for 10 min at 13 000 rpm and 4∘C. The supernatant was again removed. The remaining pellet was air dried for 10 minutes and

resuspended in 20-25 µl ddH2O.

4.7.2 Maxi preparation

Whenever larger amounts of plasmid DNA were needed for purposes such as transfection or lentivirus production, 400 ml of LB Broth were supplemented with the respective antibiotic and inoculated either with 400 휇l of a mini culture or a tipful of bacterial DMSO stock. The culture was incubated for 15 hours at 37∘C under constant 200 rpm

54 4 Materials and Methods

agitation. The following day, the bacterial suspension was centrifuged for 15 min at 6000 g. The bacterial pellet was further processed according to the instructions of the kit’s manufacturer. The resulting DNA pellet was dissolved in 300 to 500 휇l of aqua ad injectabilia.

DMSO stock Whenever a culture of transformed bacteria needed to be saved for later purposes, the bacteria were thoroughly resuspended and 930 휇l of medium transferred to a cryo tube. 70 휇l of DMSO were added and mixed. Tubes were frozen and stored at -80∘C.

4.8 RNA extraction

Item Manufacturer Cat. No. 훽-Mercaptoethanol Sigma-Aldrich, Germany M7522 Aqua ad injectabilia Fresenius Kabi, Germany 13101221 RNeasy Mini Kit Qiagen, Germany 74104

In order to harvest RNA from cultivated cells, these were detached from culture vessels and a volume containing 2 ×106 cells was spun at 180 g for 5 min. After removal of the supernatant, the pellet was lysed in 350 µl RLT buffer supplemented with 훽- Mercaptoethanol. RNA was isolated using a RNeasy Mini Kit (QIAGEN, Germany) according to the manufacturer’s instructions. Once the RNA was eluted using 30 µl

ddH2O, it was quantified using a Nanodrop Spectrophotometer (ThermoFisher Scientific, USA) and subsequently stored at −80∘C.

4.9 cDNA synthesis

Reagents, solutions Manufacturer Cat.No. and kits Aqua ad injectabilia Fresenius Kabi, Germany 13101221 One Step TaqMan○R Re- Applied Biosystems, USA N808-0234 verse Transcription

RNA can give rise to DNA in a process called reverse transcription. The molecule it yields is referred to as complementary DNA (cDNA). In the experiments described herein

55 4 Materials and Methods

5 µl 10 x RT Buffer 11 µl 25 mM MgCl2 10 µl 2.5 M dNTPs 2.5 µl 50 µM Random Hexamer Primer 20 U RNAse Inhibitor 6.25 U Multiscribe Reverse Transcriptase the "One Step TaqMan○R Reverse Transcription" kit was used. A master mix containing the following components per sample was prepared:

Using the results of the spectrophotometric quantification, RNA extracts were diluted using RNase-free water so that 19.75 µl solution contain 400 ng RNA. Master mix and RNA solution were combined in PCR tubes. The reaction was carried out in a

Table 17. Reverse transcription PCR program Duration [min] Temperature [∘C] 10 25 30 48 5 90 ∞ 4

Mastercycler Pro S (Eppendorf, Germany) using the program outlined in Tab. 17.The reaction product was either directly used for quantitative PCR or stored at −20∘C for subsequent usage.

4.10 Polymerase Chain Reaction

DNA replication is a pivotal process of all living cells and catalyzed by various members of the family of DNA polymerases. In analogy to restriction enzymes used for cloning, polymerases can catalyze DNA replication in vitro given the appropriate substrates and conditions.

However, there is a fundamental restriction as polymerases need single stranded DNA strands as template for the generation of a complementary daughter strand. In vivo the enzyme helicase severs the hydrogen bonds and provides polymerases with the needed access. In vitro helicase is substituted by thermal energy which thermodynamically breaks the hydrogen bonds between the two strands. Since complete disentanglement of the strands necessitates temperatures of about 95∘C, the enzymes of species living at

56 4 Materials and Methods

ambient temperatures would denature. Therefore, polymerases of extremophile bacteria such as T. aquaticus are used which can withstand such elevated temperatures without losing their catalytic abilities. Their optimal temperature for catalysis, however, lies at about 72∘C. The temperature at which primers - the small DNA oligonucleotides essential for DNA polymerases to start - can bind to a single DNA strand, is even lower with 55 − 65∘C.

PCR machines are designed to sequentially heat the machine to 95∘C for DNA denaturation, then to the primer annealing temperature and then to the enzyme’s optimal catalytic temperature. One such sequence is referred to as cycle. The amount of replicated DNA (A) grows exponentially with the number of cycles (n).

푛 퐴푛 = 퐴0 × 2

PCR’s advantage of enormous sensitivity that allows to amplify experimentally significant amounts of DNA from minute traces of template, can also be a disadvantage when the reaction is contaminated with DNA that allows for primer binding, thus, yielding false products with potentially deleterious effects.

4.10.1 Colony PCR

Table 18. Materials needed for colony PCR Reagents, solu- Manufacturer Cat.No. tions and kits Aqua ad injectabilia Fresenius Kabi, Ger- 13101221 many 2 mM dNTPs ThermoFisher Scien- R0241 tific, USA FIREPol Polymerase Solis BioDyne, Esto- 01-01-0100 nia MgCl2 Solis BioDyne, Esto- supplied with FIRE- nia Pol 10× Polymerase Solis BioDyne, Esto- supplied with FIRE- Buffer nia Pol

PCR represents a swift method for screening for the presence of desired DNA in colonies that arose from bacterial transformation. To do so, a small portion of each colony was

transferred to PCR tubes each containing 20 µl ddH2O water and vigorously resuspended.

57 4 Materials and Methods

Meanwhile a PCR master mix, containing the respective primers listed in Tab. 19, was prepared as following:

2 µl 10× Firepol Buffer 2 µl 2 mM dNTPs 1.2 µl 25 mM MgCl2 1 µl 10 mM forward primer 1 µl 10 mM reverse primer 5 U FIREPol Polymerase 8.8 µl aqua ad injectabilia volumes were multiplied by number of colonies to be analyzed

In a set of fresh PCR tubes, 3 µl of each bacterial colony suspension were combined with 17 µl of PCR master mix. PCR reactions were performed in a ThermoCycler(Eppendorf, Germany) and comprised 30 - 35 cycles (Tab. 20). Once PCR was completed, PCR machines cooled products to 8∘C until they were analyzed using agarose gel electrophoresis. Colonies corresponding to lanes with the desired PCR fragment were used to grow mini cultures. When in doubt, PCR fragments were sequenced. All sequencing reactions were conventional Sanger sequencing reactions carried out by GATC Biotech, Germany using primers synthesized by Eurofins, Germany. Sequence data delivered by GATC was analyzed using Lasergene (DNASTAR, USA).

4.10.2 Quantitative real-time PCR

Quantitative real-time PCR (qRT-PCR) is a suitable technique for the relative quantification of RNA abundance in a sample, thus allowing for conclusions about gene expression. Though there exist several implementations, they all rely on cDNA generated by reverse transcription and primer sets that span boundaries of neighboring exons within the cDNA of interest. The qRT-PCR system used for the experiments herein was based on the TaqMan○R technology, which operates with an oligonucleotide probe that specifically binds to a region of the cDNA of interest that lies between the above-mentioned primer set.

Both ends of the oligonucleotide probe are chemically modified: while the 5’ end is coupled to a fluorophore, the 3’ end is coupled to a quencher. The oligonucleotide is holding the two molecules in such a proximity that Förster Resonance Energy Transfer (FRET) can occur: Upon excitation of the fluorophore at its respective wavelength, the energy is

58 4 Materials and Methods

Table 19. List of primers used for colony PCR Name of primer Sequence Annealing temperature [∘C] 09 IP Fp 5’-ATATATCTTG 54.0 TGGAAAGGACG-3’ pG Britta 5’-TGGCCCCAACG 59.3 TTAGCTATTTTCAT- 3’ pCMV6-mKIAA for- 5’-TGACGTCAA 51.4 ward TGGGAGTTT-3’ pCMV6-mKIAA reverse 5’-TCTCACTCTGT 52.8 ATCCATTTAAT-3’ pCMV6-mKIAA pair 2 5’-GCGTGGATAGC 61.8 forward GGTTTGACTC-3’ pCMV6-mKIAA pair 2 5’-ATGTCGTCCT 60.3 reverse GGTATGGCTGAA-3’ pCMV6-BC143722 for- 5’-GGCGTGGATAG 64.0 ward CGGTTTGACTC-3’ pCMV6-BC143722 re- 5’-CCTCATGAAGC 64.0 verse TGCGGGTTGTC-3’ Shakaya B cPCR for- 5’-TCTGACTGTTTT 63.9 ward CTTCTACTAATG CACTCC-3’ Shakaya C cPCR for- 5’- 65.0 ward GGGTGAACTCACGT CAGAATTCAAGAG- 3’ Shakaya D cPCR for- 5’-CGAGAGTGCAGA 64.4 ward CTTATCCGGAGTA- 3’ SFFV shRNA-PCR re- 5’-AAATGGCGTTAC 61.0 verse TGCAGCTAGCTT-3’ All primers used in the colony PCRs were procured from Eurofins, Germany

Table 20. PCR program used for colony PCR Step Temperature [∘C] Duration [min] Initial denaturation 94 3-10 Denaturation 94 0.5 Annealing see Table 191 Elongation 72 depending on the fragment size 1 min/kb Terminal Elongation 72 3-10

59 4 Materials and Methods

Table 21. TaqMan○R Assays used for qRT-PCR Gene Oligonucleotides or Cat. No. forward primer: 5’ CTT TCT GGT GCT TGT CTC ACT GAC 3’ murine β2-microglobuline reverse primer: 5’ GGT GGC GTG AGA TAT GAA CTT AAA C 3’ probe: 5’FAM–ATC CAG AAA ACC CC–MGB3’ human β2-microglobuline Hs00187842_m1 murine JMJD1C Mm01150329_m1 human JMJD1C Hs00405469_m1

directly transferred to the quencher which emits light of a longer wavelength than the fluorophore does. Hence, no signal can be detected for the emission wavelength ofthe fluorophore. As PCR progresses, two things happen that are crucial for○ theTaqMan R approach:

First, Taq polymerases possess 5’-3’-exonuclease activity which means that they digest the oligonucleotide probes. This results in a spatial separation of the fluorophore from its quencher. FRET can no longer occur and the fluorophore begins to emit light that can be picked up. Second, with each PCR cycle the amount of cDNA of interest doubles. This means for the following cycle that twice as many oligonucleotides can bind to their complementary sequence and in turn twice as many fluorophores will be released by 5’-3’-exonuclease activity. Following the exponential nature of PCR, the intensity of emitted light increases exponentially from one cycle to the next. Yet, the curve assumes a sigmoid course due to the limited amount of labled oligonucleotides: The intensity of emission plateaus at a level where virtually all oligonucleotides have been digested.A master mix containing the following components per sample was prepared:

12.5 µl 2x TaqMan Universal PCR Master Mix 1.25 µl 20x TaqMan Gene Expression Assay 8.75 µl RNase-free water

After pipetting the master mix into the wells of Thermo-Fast 96 PCR Detection Plate (Thermo Scientific, UK), the cDNA obtained from reverse transcription was diluted 1:4 with RNase free water and 2.5 µl of diluted cDNA added to the respective wells. The PCR plate was sealed off using Optical adhesive film (Micro Amp, Applied Biosystems,

60 4 Materials and Methods

USA) and centrifuged for 1 min at 180 g. Each sample was analyzed in duplicate in an ABI Prism 7000 (Applied Biosystems, USA).

Table 22. PCR program used for qRT-PCR Step Temperature [∘C] Duration Initial denaturation 95 10 min Denaturation 94 15 sec Annealing and Elongation 60 1 min Denaturation, Annealing and Elongation were repeated in 40 cycles

The data was then analyzed using ABI PRISM software (Applied Biosystems, USA).

One of the easiest ways to evaluate qRT-PCR is the so called ∆∆퐶푡 method which allows for relative quantification: First, a threshold value of fluorescence (t) is set sothatall samples are analyzed during the phase of exponential growth. Next, the amount of

cycles (C푡) the cDNA of interest needed to reach t is normalized to the C푡 of a reference

gene’s cDNA yielding the so called ∆퐶푡. A proper reference gene should be consistently expressed in the given cell type under all conditions; for the experiments described herein

β2 microglobulin (B2M) was used as the reference gene. Once normalized, the ∆퐶푡 values of different cDNAs or different experimental conditions can be subtracted from eachother which in turn yields the ∆∆퐶푡 value. The actual difference in expression was calculated back by 2−∆∆퐶푡

It should be pointed that taking 2 as the base is one of the inherent error sources of this technique as it assumes the theoretically perfect efficiency of the replication reaction that cannot be safeguarded even under the exercise of utmost diligence.

4.11 Protein Techniques

4.11.1 Protein Extraction

For western blot analysis, 107 cultured cells were harvested. After the removal of the culture medium, cells were washed using chilled PBS. Adherent cells were then detached from their culture dish using a rubber cell scraper (Roth, Germany). After 5 min 180 g centrifugation PBS was removed and cells lysed in 200 µl chilled RIPA buffer and transferred to 1.5 ml reaction tubes (Greiner, Germany).

61 4 Materials and Methods

Reagents, solutions Manufacturer Cat.No. and kits Acetic Acid, glacial Roth, Germany 3738.4 Ammonium persulfate Fluka, Germany 09913 (APS) Bove Serum Albumine NEB, Germany B9001.S (BSA) Bio-Rad Protein Assay Dye Bio-Rad, Germany 5000006 Reagent Concentrate Bromphenol blue Roth, Germany A512.1 Dithiothreitol (DTT) Sigma-Aldrich, Germany 43817 Ethanole absolute Sigma-Aldrich, Germany 32205 Gel blotting paper Schleicher-Schüll, Germany 10426694 Glycine Roth, Germany 3908.2 HiMark Pre-stained molec- ThermoFisher, USA LC5699 ular weight marker HiRange Spectra molecular ThermoFisher, USA 26625 weight marker Isopropanol Roth, Germany 6752.3 Methanol Sigma-Aldrich, Germany 322415 PageRuler Prestained Pro- Thermo Scientific, Ger- 26616 tein Ladder, 10 to 180 kDa many Phosphate buffered saline Biochrom AG, Germany L182-10 (PBS) Ponceau S Sigma-Aldrich, Germany 81460-25G Rotiphorese (30 % Acry- Roth, Germany 3029.1 lamid, 0.8 % Bisacrylamid) Immobilion PVDF mem- Merck, Germany IPVH00010 brane (PVDF) Skim Milk Powder Fluka, Germany 70166 Sodium Chloride VWR, Germany 27810295 Sodium Deoxycholate Sigma-Aldrich, Germany D-6750 Sodium Dodecyl Sulfate Roth, Germany 4360.2 (SDS) Tetramethylethylenediamine Roth, Germany 2367.1 (TEMED) Trisaminomethan (Tris) Roth, Germany 9090.2 Trichloroacetic Acid Fluka, USA 91230 TritonX 100 Roth, Germany 3051.2 Trans-Blot Turbo Midi BioRad, Germany 1704157 PVDF Transfer Pack Tween-20 Roth, Germany 9127.1 Western-Lightning Plus- Perkin Elmer, USA NEL103001EA ECL β-Mercaptoethanol Sigma-Aldrich, Germany M7522 62 4 Materials and Methods

Table 23. Composition of RIPA Buffer Component Concentration Sodium Chloride 150 mM Tris pH 8.0 50 mM TritonX 100 1% Sodium Deoxycholate 0.5% SDS 0.1%

In order to precipitate genomic DNA, samples were agitated in a ThermoMixer precooled to 4∘C (Eppendorf, Germany) for 10 min. Finally, the samples were subjected to 10 min of 14500 g centrifugation. The supernatant was transferred to a new, labeled 1.5 ml reaction tube, frozen and stored in a -20∘C freezer.

4.11.2 Protein Quantification

Protein was quantified using the Bradford assay. Measurements were always performed in triplicate in clear polystyrol flat-bottom 96-well plates with lid (Greiner, Germany). Using water, 5× Bradford reagent was diluted to 1× concentration to obtain the needed final volume. 200 µl 1× Bradford reagent was added to every well to be measured. 2 µl of a sample were diluted with 8 µl of MilliQ water (Siemens, Germany). As a control, RIPA lysis buffer was diluted in the same manner. As a standard for protein content, 5 µl of 10 mg/ml BSA (NEB, USA) were diluted in 45 µl of MilliQ water. Adding respective amounts of the solution, standardized amounts of 0, 0.5, 1, 2, 3, 4 and 5 µg protein were generated. Absorbance at 595 nm was measured with an Infinite M200 microplate reader (Tecan, Switzerland). Based upon the standards’ absorbance values, regression curves were generated which allowed to back calculate samples’ protein concentrations.

4.11.3 SDS-PAGE

Polyacrylamide Gel Electrophoresis (PAGE) is a tool that allows for separation of proteins according to the length of their primary structure by virtue of a polyacrylamide resin. In order to polymerize acrylamide into a polyacrylamide resin, a radical reaction is started with APS and TEMED. It is the initial acrylamide concentration that determines the length of polymer chains, which in turn determine the speed with which proteins migrate

63 4 Materials and Methods

through the gel. Bisacrylamide influences the degree of cross linkage between the polymer chains.

Meanwhile SDS serves a dual purpose: First, it is a detergent that disrupts higher order protein structures3 so that solely the primary structure influences how fast a protein, can penetrate the polyacrylamide. Second, due to its sulfate group, SDS is negatively charged and masks the proteins’ native charge. This guarantees that when subjected to an electric field, all proteins migrate from the cathode towards the anode. In orderto ensure that most of the protein is concentrated at a certain height once the separation starts, an additional layer of gel with a lower concentration, referred to as „stacking gel“, is cast above the separating gel.

Table 24. Composition of SDS-Polyacrylamide gels Component 8% Resolving Gel 5% Stacking Gel ddH2O [ml] 4.76 3.4 1 M Tris pH 6.8 [휇l] - 630 1.5 M Tris pH 8.8 [ml] 2.5 - Rotiphorese [ml] 2.67 0.83 10% SDS [휇l] 100 50 10% APS [휇l] 100 50 TEMED [휇l] 3 5

The necessary gels were prepared using a BioRad mini gel casting chamber with spacers of 1 mm thickness. Once all reagents for the separation gel (Tab. 24) were added to a 50 ml conical tube (Corning, USA), the mixture was thoroughly mixed and swiftly pipetted between the glass slabs. About 500 µl isopropanol were pipetted on top to purge remaining bubbles and ensure that the surface was horizontal. Once the left over mixture had solidified in the conical tube, the isopropanol was decanted from the gelcasting chamber. The procedure was repeated using the stacking gel mixture and a comb instead of isopropanol. Samples were prepared by adding 5× SDS loading buffer and subsequently

Table 25.5 × Loading Buffer Component Concentration [% w/v] Tris 3% Glycine 14.4% SDS 1%

3note that disulfide bonds are covalent, meaning that they need to be severed using reducing agents

64 4 Materials and Methods

boiled at 95∘C for 5 min on a thermomixer (Eppendorf, Germany). After rapid cooling on ice and quick centrifugation, samples were loaded onto the gel. Depending on the size of the protein in question, the appropriate stained protein size marker was applied. Electrophoresis was carried out by placing the gel in a Mini-PROTEAN Tetra Cell

Table 26. 10× SDS Running Buffer Component Concentration [% w/v] Tris 3% Glycine 14,4% SDS 1%

(BioRad, Germany) and filling each side of the gel× with1 SDS Running buffer. After about 2 h at 100 V, the bromophenol blue front ran out of the gel and electrophoresis was terminated.

4.11.4 Coomassie Staining

Table 27. Composition of Solutions for Coomassie Staining Solution Composition Fixation Solution 50% Methanol and 10% Acetic Acid Staining Solution 0.1% Coomassie R-250 or G-250, 50% Ethanol, 10% Acetic Acid Destaining Solution 1% Acetic Acid

Once proteins have been resolved using SDS-PAGE, they can be visualized by exposing the gel to the dye Coomassie blue which binds to basic amino acid residues. Staining can be useful in different situations: In case recombinant or purified proteins are analyzed for quality control reasons and mere differentiation by size is sufficient. In the case ofthe experiments described herein, the staining was used after blotting to check if large size proteins were properly transferred onto the PVDF membrane.

After disassembling the blotting apparatus, the voided gel was fully immersed in appropriate amount of fixation solution and moderately agitated for 10 min. The fixation solution was decanted and replaced with the same amount of staining solution in which the gel remained overnight at RT under constant agitation. The vessel was sealed during incubation to prevent evaporation of acetic acid. At last, the staining solution was collected for further usage and replaced with an equal amount of destaining solution.

65 4 Materials and Methods

Since the amount of dye in the gel and in the solution is constantly approaching an equilibrium, it needs to be replaced regularly until bands become visible. The gel was conserved by placing it onto an equally sized Whatman○R paper and applying a vacuum heater.

4.11.5 Western blot

Unlike with DNA, mere separation by size most rarely suffices when analyzing proteins. For further analysis, protein is transferred from the polyacrylamide gel onto a hydrophobic membrane (nitrocellulose or PVDF) using an electric field perpendicular to the gel. The gel and the membrane are sandwiched in a stack of buffer soaked tissue papers.

There are different approaches to ensure the stack is exposed to a homogeneous electric field: The stack is either entirely immersed in transfer buffer or the soaked stack ispressed between two large graphite electrodes. As the proteins are still deterged with SDS, they maintain their direction of migration in the electric field. It is therefore vital to place the membrane on the side of the gel that is facing the anode. Furthermore, all bubbles have to be purged to avoid disturbances in the protein migration.

For most of the experiments a TransBlot turbo blotting apparatus (Biorad, Germany) along with the manufacturer‘s transfer packs was used in the preprogrammed „Hi Molec- ular Weight“ mode (25 V, 1 A, 10 min).

4.11.6 Antibodies

Once the protein is transferred to the membrane, it can be probed using antibodies binding to peptide sequences within the denatured protein of interest. In most cases the antibody that recognizes the protein, called primary antibody, is an IgG and does not carry a reporter activity for visualization. Therefore, the membrane is again probed, this time with a so called secondary antibody which binds to any IgG of the host species of the primary antibody (e.g. an antibody from a rabbit that was immunized with murine IgG).

Nowadays, different approaches exist to finally visualize the immunocomplex of protein- primary antibody-secondary antibody: The conventional method - which was solely employed for these experiments - makes use of horseradish peroxidase (HRP) as a

66 4 Materials and Methods

Table 28. Solutions for Antibody Detection Ponceau Stain 2% Ponceau S, 3% Trichloroacetic Acid Blocking buffer 1× PBS or TBS + 5% skim milk powder + Tween-20 TBS-T 1× TBS + Tween-20 Tween-20 concentrations were adjusted to meet the recommendations of antibody manufacturer’s

reporter. When the appropriate substrate solutions are spread on the membrane, HRP catalyzes an electrochemiluminescent reaction. The resulting light can be detected by charge-coupled devices (CCD) or film.

After disassembling the blotting apparatus, the PVDF membranes carrying the proteins were checked for successful protein transfer using Ponceau stain. After washing away the dye, membranes were placed in 50 ml conical tubes. In order to increase the signal to noise ratio, unspecific binding sites were saturated with milk proteins by adding10 ml blocking solution and incubating for 1 h at RT on a Stuart SRT9D roller mixer (Bibby Scientific, UK). The blocking solution was then discarded and replaced with blocking buffer carrying primary antibody, diluted according to the manufacturer’s specifications.

4.11.7 Reprobing of blots

In order to reprobe a western blot membrane for different proteins, the bound antibodies were removed by incubating the membrane in 50 ml of stripping buffer (Tab.30) for 30 min in a water bath (Julabo, Germany) at 50∘C. The membrane was then washed three times using PBS + 0.1% Tween-20. Finally, the membrane was blocked as outlined above before it could be probed with the antibodies detecting the protein of interest.

67 4 Materials and Methods

Table 29. List of Antibodies used for Protein Detection Antibody Manufacturer Cat.No. Dilution Blocking So- lution Primary Antibodies anti-Actin Sigma-Aldrich, A5441 1:5000 TBS-T 0.1% USA (+5% milk) anti-JAK2 Cell Signaling, 3230 1:1000 TBS-T 0.1% USA (+5% BSA) anti-hJMJD1C Abcam, UK AB130922 1:2000 TBS-T 0.1% (+5% BSA) anti-hJMJD2C Abcam, UK AB85454 1:1000 PBS-T 0.1% (+5% milk) Mouse anti- Tachibana Lab, Clone 2A 1:20 TBS-T 0.1% mJMJD1C Japan (+5% milk) Tachibana Lab, Clone 13A 1:20 TBS-T 0.1% Japan (+5% milk) Rabbit anti- Merck, Ger- 09-817 1:1000 TBS-T 0.1% mJMJD1C many (+5% milk) Abcam, UK AB56538 1:500 TBS-T 0.1% (+5% milk) Abcam, UK AB106457 1:2000 TBS-T 0.1% (+5% milk) Secondary Antibodies anti-Mouse Cell Signaling 7076S 1:2000 TBS-T 0.1% IgG HRP- (+5% milk) linked Donkey anti- GE, USA NA934V 1:10000 TBS-T 0.05% Rabbit HRP- (+5% milk) linked Sheep anti- GE, USA NA931V 1:10000 TBS-T 0.1% Mouse IgG (+5% milk) HRP-linked

Table 30. Stripping Buffer Component Amount Tris-HCl (pH 6.8) 62,5 mM β-Mercaptoethanol 100 mM SDS 2%

68 4 Materials and Methods

4.12 Cell Culture

Reagents, solu- Manufacturer Cat.No. tions and kits DMEM (Dulbecco’s Gibco, Germany 41966-029 Minimally Enhanced Medium) Fetal Calf Serum Gibco, Germany 10270-106 (FCS) RPMI (Roswell Park Invitrogen, Germany 21875-034 Memorial Institute) IMDM (Iscove’s Gibco, Germany 21980-032 Modified Dulbecco’s Medium) L-Glutamine Gibco, Germany 25030 Penicillin/ Strepto- Lonza, Switzerland DE17-602E mycin Phosphate buffered Biochrom AG, Ger- L182-10 saline (PBS) many

For maintenance cell culture, cells were kept in 12.5, 25, 75 or 175 cm2 cell culture flasks (Greiner, Germany) and incubated in a Function Line Incubator (Hereaus, Germany) at ∘ 37 C, 5% CO2 in normal atmosphere, saturated humidity.

For experiments, cells were seeded into petri dishes (Greiner, Germany). Cells were handled in a Hera biosafety cabinet (Heraeus, Germany). For counting, cells were stained by adding 45 µl PBS, 45 µl Trypan Blue and 10 µl cell suspension. 10 µl of the mixture were applied between a Neubauer chamber (Paul Marienfeld GmbH, Lauda-Königshofen) and a coverslip. The average of the four grids was calculated and plugged into the following formula:

Cell density[푐푒푙푙푠/푚푙] = average cell count per grid × 105/푚푙

Adherent Cells All adherent cells used throughout the experiments were cultured in DMEM supplemented with 10% FCS, 1% L-glutamine and 1% penicillin-streptomycin. For counting and splitting of cells, culture medium was aspirated, carefully rinsed with PBS and 1-2 ml of TrypsinLE spread over the cells. After 2 min incubation at 37∘C, trypsin was inactivated using the tenfold amount of culture medium.

69 4 Materials and Methods

∙ HEK 293T 293T cells are derived from human embryonic kidney and were mainly used for transfection experiments. [132] Roughly every other day, cells were passaged and split at a ratio of 1:10.

∙ HeLa These immortalised epithelial cells are derived from the cervix carcinoma of Henrietta Lacks [335]. They were split twice a week.

∙ 3T3 These immortalised murine fibroblasts were used for testing shRNA constructs targeting the murine form of JMJD1C. Cells were split three times a week.

∙ γ2a This cell line is derived from human fibrosarcoma cell line 2C4 and fails to express JAK2. [202] Cells were split twice a week.

Suspension Cells

∙ UKE1 The cells were grown in IMDM supplemented with 10% FCS, 10% horse serum, 1% penicillin-streptomycin, 1% L-glutamine and 1 µM hydrocortisone. Cells were split two to three times a week to keep them at a density of about 0.5×106/ml culture medium.

∙ SET2 Cells were grown in RPMI supplemented with 20% FCS,1% penicillin- streptomycin, 1% L-glutamine. Cells were kept at a maximum density of 1.5×106/ml culture medium.

∙ HEL This cell line was grown in DMEM containing 10% FCS, 1% Penicillin- Streptomycin and 1% L-glutamine.They were split about every other day.

4.12.1 Transient Transfection

In contrast to Stable Transfection where genetic material is introduced into cells and selection methods are used to isolate cells that have integrated the material into their genome, transient transfection aims at delivering DNA or RNA to cultured cells with the aim of an immediate effect of the introduced material. Since metazoan cells donot readily take up naked nucleic acids, transfection relies on packing nucleic acid in a manner that facilitates uptake by the cells. Of the many available solutions, the two following techniques were employed for these experiments:

70 4 Materials and Methods

Calcium chloride precipitation Transfection mixtures were prepared by combining - in the following order - water, the needed volume of plasmid solution and 2.5 M calcium chloride to add to a total amount of 0.5 ml. An equal amount of 2× BES was pipetted to a FACS tube (Greiner, Germany) and kept under constant agitation while the mixture containing the plasmid was pipetted in drop by drop. Precipitation was allowed for 15 to 30 min at RT before the mixture was added to a petridish containing cells that had been seeded the day before.

Table 31. Materials needed for Calcium chloride precipitation Reagents, solutions Manufacturer Cat.No. and kits Bis(2-hydroxyethyl)-2- Sigma-Aldrich, Germany B2891-25G aminoethanesulfonic acid (BES) Calcium chloride Sigma-Aldrich, Germany 10043-52-4 Disodium phosphate Roth, Germany 4984.1 / 500g Sodium chloride VWR-Prolab, Germany 27810.295

Table 32.2 × BES Buffer Component Concentration [mM] BES 50 NaCl 280 Na2HPO4 1 Titration to pH 6.96

Polyethyleneimine Polyethyleneimine (PEI) is a branched, macromolecular polymer which is cationic due to its abundunce of protonable amine groups. Its electrostatic charge allows the molecule to complex DNA as well as interact with negatively charged cell surface structures. PEI-DNA-complexes are taken up via endocytosis. It is hypothesized that the protonation of PEI amine groups results in an influx of protons that disrupts the endosome, releasing PEI-DNA-complexes into the cytosol. [44][10]

PEI transfection solution was concocted by adding 200 mg of PEI to 175 ml of ddH2O. Using HCl, the pH was adjusted to less than 2 and the solution was stirred until PEI had fully dissolved. After adjusting the solution’s volume to 200 ml, it was filter sterilized, aliquoted and stored at −20∘C. For transfection, the needed amount of plasmid DNA was dissolved in 200 - 1000 µl DMEM without any supplements. PEI

71 4 Materials and Methods

Table 33. Materials needed for Polyethyleneimine transfection Reagents, solutions and kits Manufacturer Cat.No. DMEM (Dulbecco’s Modified Eagles Medium) Gibco, Germany 41966-029 Millex-GV Filter Unit Merck, Germany SLGV033RS Polyethylenimine (PEI), linear, ≈ 25kDa Polysciences, USA 23966

transfection solution was added in such an amount that the ratio of DNA mass to PEI mass equaled 1 : 3. After vortexing, complex formation was allowed by 15 min incubation at RT. The mixture was then sprinkled drop by drop into the petridish containing the cells.

4.12.2 Isolation and Handling of CD34+ cells

Isolation Buffy coats of blood donations processed on the same day were first diluted with 0.9% sodium chloride solution at a ratio of 1:1. Erythrocytes were allowed to sediment for 20 min. The supernatant was transferred to 50 ml conical tubes and centrifuged at 1600 rpm for 10 min in a Hereaus Megafuge 1.0 (Hereaus, Germany). The pellet, which is mostly composed of leukocytes contaminated with erythrocytes, was resuspended in 35 ml 0,9% sodium chloride and slowly laid over with 15 ml of Ficoll-Paque○R . The 50 ml conical tube containing the Ficoll was gently tilted while applying the leukocyte suspension to reduce the chances of disturbing the two phases. The tubes were centrifuged at 1800 rpm for 45 min in a Megafuge 1.0. The centrifuge was set to omit the breaking phase which could otherwise disturb the resulting phases.

Table 34. Materials needed for the isolation of CD34+ cells Reagents, solutions Manufacturer Cat.No. and kits anti-CD34-PE Pharmingen, USA 34375X CliniMACS Miltenyi Biotec, Germany 700-25 Direct CD34 Miltenyi Biotec, Germany 130-046-703 Progenitor Cell Isolation Kit, human MACS Ficoll-Paque○R Amersham Biosciences, USA 17-1440-03 Human serum albumin CSL Behring, Germany (HSA) 20% Sodium chloride 0.9% Fresenius Kabi, Germany 1312813

72 4 Materials and Methods

The mononuclear cells (MNC) which are located right above the phase boundary were aspirated using a glass pipette, transferred to fresh 50 ml conical tubes containing 50 ml PBS and centrifuged at 1200 rpm in a Megafuge 1.0. After decanting the supernatant, the MNC pellet was again resuspended in PBS, centrifuged and the PBS decanted. The washed pellet was resuspended in CliniMACS containing 0.5% HSA in such a fashion that the suspension totalled to 10 ml.

After estimating the cell density using a Nebauer counting chamber, the suspension was centrifuged at 1200 rpm for 10 min in a Eppendorf table top centrifuge. The pellet was resuspended in such an amount of CliniMACS + 0.5% HSA that - based on the counting results from before the centrifugation - the resulting cell density should be 1 ×108/300 µl. The suspension was transferred to a 15 ml conical tube and 100 µl per 108 cells of the included blocking agent as well as 100 µl per 108 cells of the anti-CD34-labelled magnetic beads were added to the suspension. The mixture was then incubated for 30 min at 4∘C under constant agitation.

After incubation, the mixture was topped up with CliniMACS + 0.5% HSA to reach a total volume of 15 ml and subjected to 1000 rpm for 10 min in an Eppendorf table top centrifuge. Meanwhile, the supplied column was placed in the adequate magnet and equilibrated with 3-5 ml CliniMACS + 0.5% HSA. Once centrifugation was completed, the pellet was resuspended in 14 ml CliniMACS + 0.5% HSA and then applied onto the column through a supplied nylon filter. The column was washed three times with each3 ml CliniMACS + 0.5% HSA.

All flow through and wash fluids were collected in a 15 ml conical tube. After thelast wash, the nylon filter was removed, the column was taken out of the magnet and placed on top of a 15 ml conical tube. Cells were eluted in 5 ml CliniMACS + 0.5% HSA under careful pressure of a plunger. 10 µl of the eluate were used for counting in a Neubauer chamber. The purity of the CD34+ cells was ascertained by FACS analysis: 1 x 104 cells in 100 µl CliniMACS + 0.5% HSA were transferred to a FACS tube and 2 µl anti-CD34-PE (Pharmingen, USA) added. After 20 min incubation at 4∘C in darkness, the cells were washed with 3 ml ClininMACS + 0.5% HSA, centrifuged at 1200 rpm for 5 min. Once the supernatant was decanted, the cells were resuspended in 500 µl PBS and analyzed using a Fortessa LSR II cytometer (Beckton Dickinson, USA).

73 4 Materials and Methods

Lentiviral Transduction and Methylcelluslose Assay In a first step, methylcellulose media were prepared by taking a batch of H4330 and filling it up to 100 ml using IMDM. After inverting and waiting for air bubbles to dissipate, 2.5 ml aliquots were made and stored at −20∘C.

Table 35. Materials needed for handling CD34+ cells Reagents, solu- Manufacturer Cat.No. tions and kits 6-well plate Greiner, Germany M8562 24-well plate Greiner, Germany M8812 Cellstar TC plate Greiner, Germany 657160 Iscove’s Muodifies StemCellTechnologies, 36150 Dubecco’s Medium Canada (IMDM) Iscove’s Muodifies StemCellTechnologies, 07700 Dubecco’s Medium Canada (IMDM) + 2% FCS MethoCultTM H4330 StemCellTechnologies, 04330 (with EPO) Canada StemSpan○R SFEM StemCellTechnologies, 09600 Canada recombinant human Peprotech, USA 300-18 Thrombopoietin recombinant human PeproTech, USA 300-07 Stem Cell Factor recombinant human PeproTech, USA 200-06 Interleukin 6 recombinant human Peprotech, USA 300-19 Flt3 Ligand

Table 36. Supplementation of StemSpan○R for cultivation of CD34+cells Cytokine Concentration [ng/ml] TPO 20 IL-6 20 Kit Ligand 100 Flt3 Ligand 100

74 4 Materials and Methods

For lentiviral transduction, 1 × 105 CD34+cells for each the knock down as well as the empty vector control group were cultured each in 500 µl of supplemented medium in 24 well plates. Upon the day the culture started (day 0) lentiviral particles carrying piGhU6-Daphne and piGhU6-empty, respectively, were added to the cells at a multiplicity of infection (MOI) of 10. The medium was refreshed daily by taking out 300 µl of supernatant and centrifuging it for 5 min at 5000 rpm in a table top centrifuge in order to sediment any suspended cells.

After centrifugation, the supernatant was carefully removed using a pipette and replaced with 300 µl fresh, supplemented medium. Virus was again added on day 1 but none on day 2. On day 3, the cells were gently resuspended and the suspension transferred to FACS tubes. The transfer of the cells was confirmed by light microscopy of the emptied wells. The cells were sorted using a FACS Aria (Beckton Dickinson, USA) with a 100 µm nozzle. Proper gating for GFP positivity was established using a small set of CD34+cells of the same batch that had not been exposed to lentivirus.

Cells deemed positive were sorted into 5 ml polystyrene round bottom tube (Corning, USA) containing 1.25 ml of methylcellulose medium. Since there had been problems with unforeseen overgrowth in prior experiments, the cells were fractioned among several tubes as outlined in Fig.8.

Figure 8. Experimental Setup for CD34+ Methylcellulose Assay. Numbers represent amount of cells.

The cell suspensions were homogenized by inverting the tubes for 5 min. After waiting another 5 min for any air bubble to dissipate, the content of each tube was split between two wells of a six well plate (Cellstar, Greiner Bio-One, Germany), yielding about 550 µl per well. Four of the six wells of a plate were used for experiments. The center two were

75 4 Materials and Methods

filled with PBS in order to provide for enough humidity to prevent the methylcellulose from drying out. The plates were incubated at the aforementioned conditions. After 14 days, erythroid differentiation was assessed by light microscopy.

4.12.3 Freezing and Thawing of Cells

Item Manufacturer Cat. No. Cryo.S round bottom tubes Greiner Bio-One, Germany 122263 Dimethylsulfoxide Sigma-Aldrich, Germany D2438 Fetal Calf Serum (FCS) Gibco, Germany 10270-106 Isopropanol Roth, Germany 6752.3

Depending on the cell type, 1 ×106 to 1 ×107 cells were centrifuged at 1000 rpm for 5 min. After removal of the supernatant, the pellet was resuspended in 1 ml FCS containing 10% DMSO. The cell suspension was transferred to a cryotube (Greiner, Germany) and placed in a chilled freezing unit (Nalgene, USA) containing absolute isopropanol. The freezing unit containing the cryotubes was stored at −80∘C for at least 24 h to allow for slow cooling. Once frozen, cryotubes were transferred to liquid nitrogen tanks. Cells were thawed by warming cryotubes by hand. Once fluid, DMSO was immediately rinsed off by adding the cell suspension to 10 ml of DMEM. After pelleting the cells by10 min 1000 rpm centrifugation they were seeded in their respective medium at appropriate density.

4.13 Virological Methods

Viruses can transfer genetic information between different genomes in a process called transduction. In order to exploit this randomly occurring natural phenomenon for the targeted transfer of genetic material, a variety of viral vectors for different model species and purposes have been devised.

For mammalian cells the family of Retroviridae is widely used: once inside a cell, the virus’ enzyme reverse transcriptase transcribes the RNA genome. The produced cDNA is then stably integrated into the host cell’s DNA by virtue of the viral enzyme Integrase. Lentivirus, a genus of Retroviridae, is able to do so in the absence of a cell cycle which allows for transduction of resting cells. Creating an artificial lentivirus carrying a gene of interest requires cloning the desired DNA within the integration sequences of a DNA

76 4 Materials and Methods

plasmid representing the lentiviral genome that can be subsequently transfected into cells that start manufacturing virions.

As a security measure aiming to prevent the uncontrolled replication and spreading of artificial, transgene-carrying viruses, the lentiviral genome is partioned into multiple plasmids of which only the transgene-carrying plasmid contains the psi-sequence necessary to be packaged into the virion. [423] The other vectors used for virus production were pMD2-vsvg, encoding the envelope of vesicular stomatitis virus G which enables the virions to bind mammalian low density lipoprotein receptor (LDLR) [112] and pCMV- dR8.74 which encodes reverse transcriptase and integrase along with viral structural proteins. [423]

4.13.1 pLeGo-iG-hU6

The lentiviral vector used in the experiments described herein was LeGo, which is a derivative of a third generation lentiviral vector. The casette carrying the transgene is flanked by two self-inactivating long terminal repeats (SIN-LTR) which serveas cis-elements for integration. They are referred to as "self-inactivating" as they loose their transcriptional activity after integration which prevents the erroneous activation of an oncogene in case the provirus might integrate in the vicinity of a protooncogene. [269]

Figure 9. Illustration of the pLeGo-iG casette. Figure adapted from [423]

Transcription of the transgene is driven by the promoter of spleen focus-forming virus (SFFV) which is especially effective in the context of hematopoietic cells. Using a multiple cloning site (MCS), the transgene can be inserted downstream of the promoter. The transgene’s open reading frame (ORF) is immediately followed by an encephalomyocarditis virus internal ribosome entry site (IRES) which allows for the translation of the adjacent reporter gene; in the case of LeGo-iG that is enhanced Green Fluorescent Protein (eGFP).

77 4 Materials and Methods

The reporter gene is followed by a Woodchuck Hepatitis Virus Posttranscriptional Regu- latory Element (wPRE) which boosts expression of the transgene. [453] Upstream of the SFFV promoter, LeGo contains a human U6 promoter followed by HpaI and XhoI sites that allow for the insertion of a short-hairpin RNA cassette. [315] The human U6 promoter recruits RNA polymerase III which transcribes small RNA molecules. For experiments described herein, the shorthand "piGhU6" stands for pLeGo-iG-hU6.

4.13.2 Production of Viral Particles

Component Manufacturer Cat.No. Bovine Serum Albu- Sigma-Aldrich, Ger- A-1595 mine (BSA) many IMDM (Iscove’s Gibco, Germany 21980-032 Modified Dulbecco’s Medium) Polyallomer Centrifu- Beckman Coulter, 326823 gation tubes Germany Stericup○R GV PVDF Merck, Germany SCVPU02RE filter unit

All procedures were carried out according to virological safety guidelines. Virus was solely handled in the above mentioned Hera biosafety cabinet without the use of suction pumps in order to avoid aerosolization of virus. Whenever handling potentially infectious material, surfaces were treated with 2% Incidin (EcoLab, USA).

24 hours prior to transfection, 1×107 293T cells that had been split 48 h hours before the transfection were seeded in 15 cm culture dishes (Greiner, Germany) with 18 ml supplemented DMEM. Transfection mixtures were prepared by dripping 84 µl of 1 mg/ml PEI under vortexing into 2 ml of DMEM without supplements containing 16 µg lentiviral vector, 8 µg pCMV-dR8.74 and 4 µg pMD2-vsvg. After an incubation period of 15 min at RT, cells were transfected by dripping the mixture into the culture medium of respective petri dishes.

14 h after the transfection, the medium was replaced with 16 ml fresh, supplemented DMEM. 38 h after the transfection, the virus containing medium was harvested and cooled at 4∘C. Again 16 ml of supplemented DMEM were gently applied to the cells. The procedure was repeated at 62 h as well as 86 h after the transfection.

78 4 Materials and Methods

Finally, small patches of cells from the dishes were washed off using PBS and prepared for FACS analysis (see 4.14) of GFP expression as a control of transfection efficiency. Meanwhile, collected viral supernatant was pooled and cleared of cellular debris by passing it through a 20 µm Stericup filter sterilization device (Merck Millipore, Ger- many).

To concentrate the viral load, 32 ml of cleared supernatant were filled into one Polyallomer centrifuge tube (Beckman Coulter, USA) and six tubes placed into the swing buckets of a SW28 rotor (Beckman Coulter, USA) for centrifugation in a Beckman Optima LE-80 K Ultracentrifuge (Beckman Coulter, USA). Hence, the viral supernatant of four 15 cm culture dishes was processed in one run. The centrifuge was evacuated and cooled to to 4∘C before it was run at 19 500 rpm for 2.5 h.

Once the run was completed, the supernatant was carefully decanted. Virions were resuspended from the resulting pellets by adding 100 µl IMDM supplemented with 2 % BSA to each centrifuge tube, followed by 20 min 700 rpm agitation. The viral suspension was transferred to 1.5 ml reaction tubes (Greiner, Germany), centrifuged at 5000 rpm for 1 min in a Biofuge pico (Heraeus, Germany), flash frozen using liquid nitrogen and placed under storage at −80∘C.

4.13.3 Titration of Viral Yield

In order to estimate the yield of a batch of virus production, 5 × 104 293T cells were seeded in each well of a 24-well plate. (Cellstar, Greiner Bio-One, Germany) Each well was supplemented with 500 µl of appropriate medium (4.12).

5 µl frozen virus concentrate were serially diluted by a factor of 10 in culture medium, so that cells were infected with 5, 0.5, 0.05 or 0.005 µl concentrated viral suspension. The cells were infected four to six hours after seeding. 24 hours after infection, the wells were topped up with additional 500 µl fresh culture medium.

72 hours after infection, the culture medium was removed, and the cells flushed out of the wells using PBS. After centrifugation and resuspension in 300 µl PBS, the suspension was transferred to a 5 ml polystyrene round bottom tube (Corning, USA). Using flow cytometry the amount of GFP+ cells was determined and used as a surrogate parameter for infection.

79 4 Materials and Methods

The amount of viral particles in the concentrate was backcalculated using the following formula in which n represents the total number of cells and V the volume of viral suspension:

+ µ푙 푉 푖푟푖표푛푠 푛 × 퐺퐹 푃 [%] × 1000 = 푚푙 푚푙 100% × 푉 [µ푙]

4.14 Fluorescence Activated Cell Sorting

The acronym FACS stands for Fluorescence Activated Cell Sorting which can be described as a complex fusion of microfluidics and laser optics. In brief, a FACS machine takes ina cell suspension and routes it through its tubing system in such a fashion that single cells can pass through an optic array composed of lasers, mirrors and CCD sensors. While scattering in the direction of the laser beam is termed forward scatter and indicates the size of a passing cell, the light scattered perpendicularly is referred to as sideward scatter and provides informartion about the granularity of the cell.

Moreover, sampled cells may contain fluorescent molecules, termed fluorophores. Cells my do so as they either by expressing fluorescent proteins or because they were exposed to fluorophore-coupled antibodies that bind proteins on or within the cell. The lasers within the FACS machine emit light of discrete wavelengths. In case of a cell that passes through the chamber and carries fluorophores, the fluorophores will be excited bythe light and as a consequence of the so called Stokes shift emit light of a larger wavelength. The emitted light is recorded in the CCD element and can be discriminated from the laser light due to the shift in wavelength. Comparing the data of the sample to a reference population that does not carry the fluorophore, allows for characterization and analysis of the cells.

Once analyzed, the cells’ fate depends on the experimental setup. In case only the measured data is needed, the procedure is referred to as flow cytometry and the cells are discarded afterwards. If the cells are needed for further experiments, they can be sorted according to the recorded characteristics via electrostatically charging the droplets that contain them and deflecting these using angular plate electrodes.

80 5 Results

5 Results

5.1 Cloning of JMJD1C

Due to the lack of validation of both the antibodies as well as the custom-designed shRNAs targeting JMJD1C, mammalian expression vectors allowing for transient expression of JMJD1C were constructed. They served as a reliable benchmark for antibody validation and in turn for knock down quantification.

A pCMV6-Kan/Neo carrying murine Jmjd2c cDNA had already been used in previous projects. [302] The cDNA had been inserted between the sites RsrII and NotI. In order to insert JMJD1C cDNA, Jmjd2c cDNA was excised by a restriction digest comprised both KpnI and NotI since their respective sites lie directly adjacent to each end of the insert.

Following gel purification, the ends of the vector backbone were blunted since KpnIand NotI feature incompatible overhangs. After allowing for religation of the blunted vector, the successful excision was controlled by EcoRI digestion. Since EcoRI lies outside of the RsrII and NotI sites used to clone Jmjd2c into pCMV6-Kan/Neo but within the KpnI and NotI sites used to excise the fragment, vectors still carrying the insert were linearized by EcoRI digestion. Meanwhile, recircularized vectors devoid of Jmjd2c remained supercoiled in gel electrophoresis.

5.1.1 Human JMJD1C

The bacterial vector pCR-XL-TOPO carrying a complete coding cDNA of human JMJD1C (NCBI accession number: BC143722) was procured from Origene, Rockville, USA. In order to clone the cDNA into the newly generated pCMV6-Kan/Neo mammalian expression vector, pCR-XL-TOPO-BC143722 was digested using MluI and ApaI while pCMV6-Kan/Neo was digested with ApaI and BssHII. Since the overhangs resulting from MluI and BssHII digestion feature compatible cohesive ends, the fragment carrying the cDNA could be inserted into the expression vector without further processing.

Unlike most cloning strategies where the cDNA is a fraction of the size of the complete plasmid, the BC143722 cDNA is about two to three times larger than the vector backbones

81 5 Results

Figure 10. Excision of Jmjd2c cDNA from pCMV6-Kan/Neo The lane marked with * shows undigested pCMV6-Kan/Neo-Jmjd2c. The adjacent lane marked with # shows the same vector linearized with EcoRI. The lanes to the right of the 2lg ladder show shorter, supercoiled vector devoid of both Jmjd2c cDNA and the adjacent EcoRI restriction site. which should incorporate it. Hence, the backbone to insert-ratio described in 4.1.4 needed to be inverted on favor of more vector.

Transformed bacteria were selected on plates containing Kanamycin and colonies screened using colony PCR. Mini cultures were grown from cultures deemed positive by PCR screening. Plasmid DNA from these cultures was sequenced and affirmed that the cloning of BC143722 into pCMV6-Kan/Neo had worked.

5.1.2 Murine JMJD1C

Since the effects of a Jmjd1c knockdown should possibly be investigated in a mouse model, murine Jmjd1c cDNA was cloned for the same reasons as for the human counterpart. In the process, following challenges needed to be overcome:

First, it was difficult to find a cDNA of a length that could possibly span the gene’sentire coding DNA sequence (CDS). After extensive research on NCBI, mKIAA1380 (NCBI accession number: AK173162) from the ROUGE database of the Kasuza DNA Research Institute’s Mouse cDNA Project was found. The clone was delivered in the pBC-SK(+) plasmid and successfully transformed clones were selected using chloramphenicol. Colonies of these clones were used to grow mini cultures. Plasmid DNA harvested from those mini

82 5 Results

cultures was test digested using EcoRI to check for pBC-SK(+)-mKIAA1380 expected band pattern.

Second, the needed cloning strategy was complex: In the original vector, the mKIAA1380 cDNA is flanked by two BssHII sites, while pCMV6-Kan/Neo features a single BssHII site in its MCS. Thus, the cDNA carrying fragment can insert itself with the start codon either facing in the direction of the CMV promoter or in the wrong way. This does not only lower the probability of a correct insertion by the factor of two but makes it necessary to distinguish the directionality of insertion in positive clones. Therefore, a multi-step approach was used to isolate relevant clones:

First, concomitant HindIII and StuI digestion was employed to screen mini culture plasmid DNA for the presence of the insert. Fig.11 illustrates how empty pCMV6-Kan/Neo is severed into two pieces of 2 kb and 3 kb (middle lane), while the same digestion yields fragments of 5.8 kb, 3 kb, 1.7 kb, 1.5 kb and 0.3 kb once the vector carries mKIAA1380 in the right orientation (lane #).

Figure 11. HindIII/StuI digest of pCMV6-Kan/Neo-mKIAA1380.# and * denote lanes that feature a banding pattern suggestive of mKIAA insertion. 2 lg ladder. Fragment length in kb.

In a second step, clones deemed positive after the first screen were confirmed by NcoI digestion; plasmid DNA carrying mKIAA1380 was expected to show a band at 12 kb and 1.5 kb (Fig.12, lane NcoI#). Furthermore, directionality was checked using XhoI digestion. For plasmids carrying correctly oriented mKIAA1380 a banding pattern of 6.4 kb and 6.9 kb was expected, while the false orientation yields fragments of the size 1.8 kb and 11 kb. (Fig.12, lane XhoI#; band at 2.6 kb due to contamination). Following this approach, a single positive clone was identified whose plasmid DNA was sequenced to assure the correct insertion.

83 5 Results

Figure 12. XhoI/NcoI control digest of pCMV6-Kan/Neo-mKIAA1380.# plasmid carries insert in correct orientation; * plasmid carries an inverted insert. 2 lg ladder.

5.1.3 Introduction of myc-tag

Once JMJD1C cDNAs were cloned into mammalian expression vectors, a myc-tag was cloned into the N-terminal end of the CDS in order to circumnavigate the situation of poor JMJD1C-specific antibodies: Using a myc-tagged protein, one can not only benchmark the detection of a protein of interest by a new antibody to an antibody of known affinity, but it may also facilitate future experiments like co-immunoprecipitation (CoIP). PCR was employed to introduce a myc-tag into a N-terminal fragment of the JMJD1C CDS. Forward primers spanning the AsiSI restriction site lying between the promoter and the first ATG were designed. The AsiSI site is followed by the start codon, the codons of the myc-tag and then the first couple of codons of the JMJD1C CDS.

Using the pCMV6-Kan/Neo vectors carrying human or murine JMJD1C as template, PCRs were carried out as described in 4.10. The oligonucleotides’ appropriate annealing temperature of 60∘C was established empirically using gradient PCR. Fol- lowing gel purification, the PCR fragments were phosphorylated and subcloned into pBC-SK+ that had been prior linearized using EcoRV and subsequently dephosphory- lated.

Transformed bacterial colonies were screened using the restriction enzymes whose sequences are part of the inserted PCR product. Positive clones were propagated and used for maxi cultures. Maxi culture plasmid DNA was purified and digested in order to obtain myc-tagged fragments suitable for ligation.

84 5 Results

Table 37. myc-tag Forward Primers cDNA vector AsiSI vector myc-tag CDS BC143722 5’-GGCCG GCGATCGC GG ATGGCATC CAAGTC AATGCA CCTTAT GAAGCT TCA-3’ GATCTC AGAGGA GGACCTG mKIAA1380 5’-GGCCG GCGATCGC GG ATGGCATC CAAGTC AATGCA CCTTAT GAAGCT TCC-3’ GATCTC AGAGGA GGACCT G

Table 38. myc-tag Reverse Primers cDNA vector restriction site CDS BC143722 5’-TTGGG CACGTG TATAATGGCT GTGAACAGCG TTGACGTTTT GATTGGCAC-3’ mKIAA1380 5’-AAAGT ATTTAAAT TTTCCTGAGC CGCCCTCTCA GAATGAGTGT CGTTCTTGAT AT-3’ restriction site BC143722: PmlI mKIAA1380: SwaI

85 5 Results

In the meantime, the pCMV6 expression vectors carrying the JMJD1C cDNAs were digested with the respective enzymes. The N-terminal cDNA was separated using gel purification and the backbone with the C-terminal JMJD1C fragment was purified for ligation. The myc-tagged N-terminal fragment and pCMV6 carrying the C-terminal part of JMJD1C were religated. Transformed bacterial clones were screened by digesting purified mini culture DNA with XbaI.

Table 39. Expected banding patterns for XbaI digestion of myc-tag JMJD1C expression vectors Name of vector Fragment size [bp] 9639 pCMV6-Kan/Neo-BC143722 religated 2382 110 5068 pCMV6-Kan/Neo-BC143722 with myc-tag 3172 2915 1719 9639 pCMV6-Kan/Neo-mKIAA1380 religated 2382 110 5057 pCMV6-Kan/Neo-mKIAA1380 with myc-tag 4912 2382 110

Figure 13. XbaI digest of pCMV6-Kan/Neo-myc-JMJD1C mini culture DNA

Comparing the results of the digestion with the expected results, the clones BC #1 and mKIAA #2 were deemed positive and propagated. A subsequent sequencing, however

86 5 Results

revealed that only the human JMJD1C cDNA BC143722 had been successfully tagged with myc. The sequence of the murine cDNA had remained unchanged.

5.2 Characterization of western blot antibodies

Since the antibody AB31215 (Abcam, UK) that Jan Peeken had used for his studies of JMJD1C [302] was no longer available, both commercially available antibodies as well as antibodies other laboratories had generated for their published experiments were tested.

Fig. 14 gives an overview of the recognition of murine JMJD1C by a series of tested antibodies. All depicted blots bear the same batch of lysates from HEK293T cells that were either transfected with pCMV6-Kan/Neo-mKIAA1380 (left lane), pCMV6- Kan/Neo-empty (middle lane) or had remained untransfected (right lane). Of each lysate, 50 µg of whole protein were spread on each lane and images were captured at 10 min exposure time. The full length JMJD1C protein is expected at a height of 280 kDa.

Figure 14. Candidate antibodies against mJMJD1C

87 5 Results

Figure 15. Merck 09-817. Rabbit polyclonal antibody. 1:1000 in TBS + 0.1% T +5 % skim milk

Fig.15 shows the detection of 3T3 cell lysate. Cells had either remained untransfected (right lane), been transfected with pCMV6-Kan/Neo-mKIAA1380 without a myc-tag (middle lane and left lane). Compared to the commercially available AB56538 and AB106457 presented in Fig.14, Merck’s antibody 09-817 performs better in specifically detecting full-length murine JMJD1C as it shows a stronger signal at the expected height of 280 kDa compared with the signal strength of unspecific bands.

Though the antibodies from the Tachibana laboratory seem at least equally specific in the detection of murine JMJD1C, the problem lies in the continuous supply as the laboratory was amidst the process of licensing the hybridoma clones to biotechnological companies. This meant that for legal reasons the laboratory was no longer able to freely supply others with the larger amounts of antibody, at the same time as the licensing company had not yet established the production. For all of the above reasons, Merck’s 09-817 emerged as the antibody used for the experiments described in 5.4.

5.3 Characterization of final shRNAs

From the design procedure outlined in 4.2 resulted sequences described in Tab.40 and Tab.41. More sequences were taken from Sigma-Aldrich, USA (Tab.42) and Shakya et al. (Tab.43). In order to ensure that JMJD1C protein expression is sufficiently suppressed, NCBI was employed to verify that the generated sequences were complementary to all known transcript variants.

88 5 Results

Table 40. Daphne Species human Sequence 5′ to 3′ GCACTTATTGGGTCAGAAA Binding site transcript variant 1 2 159 nt Binding site trancript variant 2 2 159 nt Binding site transcript variant 3 2 491 nt

Table 41. Custom Species murine Sequence 5′ to 3′ GGATCCAAATGTTAGTGAT Binding site transcript variant 1 1 611 nt Binding site trancript variant 2 1 843 nt

Table 42. Sigma 62 Species murine Sequence 5′ to 3′ AGTAACTACTTCACTACTT Binding site transcript variant 1 3 253 nt Binding site trancript variant 2 3 271 nt

Table 43. shRNAs taken from Shakya et al. [341] shRNA A shRNA B shRNA C shRNA D Species murine murine murine murine Binding 3249 3540 5163 5939 site transcript variant 1 Binding 3481 3772 5398 6174 site trancript variant 2

89 5 Results

The sequences were cloned into piGhU6 as outlined in 4.2 and 4.13.1. The insertion was verified by restriction digestion with HpaI and EcoRI which is expected to yield alarge fragment of 7483 bp and a smaller fragment spanning 670 bp with or 606 bp without the shRNA cassette.

Fig.16 illustrates the successful cloning of the designed shRNAs into the lentiviral vector piGhU6 as the lower band shifted from around 600 bp for the empty vector to nearly 700 bp for the cloned vectors.

Figure 16. HpaI/EcoRI digest of piGhU6 carrying shRNAs

5.4 Verification of knockdown efficiency

5.4.1 Verification by qRT-PCR

In order to evaluate the knockdown efficacy of the shRNA "Sigma 62" which targets murine JMJD1C, 3T3 cells were transduced with either piGhU6-empty or piGhU6- sigma62.

Fig.17 shows the results of qRT-PCR evaluation of murine JMJD1C knockdown. After a phase of unspecific signals, the fluorescence signals increase exponentially from cycle20 and onward.

90 5 Results

Figure 17. qRT-PCR for JMJD1C from RNA of 3T3 cells transduced with shRNA "Sigma 62" x-axis: number of PCR cycles; y-axis: detected fluorescence in arbitrary units, logarithmic scale

91 5 Results

Although this development may follow the course of the experiment outlined in 4.10.2, the results ought to be interpreted with caution due to the degree of variance between

the replicas’ 퐶푡 values: A deviation of by one cycle length implies a twofold difference in the abundance of the cDNA of interest.

Furthermore, the fluorescence curves are expected to converge to a single plateau value. In this experiment, however, the plateau values diverge by an order of magnitude. In spite of these shortcomings, the data allowed for the calculation of "Sigma 62" knockdown efficacy:

Table 44. Avarage 퐶푡 value The numbers designate the average 퐶푡 over all replica Gene 3T3 piGhU6-empty 3T3 piGhU6-sigma62 B2M 28.29 21.09 JMJD1C 33.32 27.62

Using the ∆∆퐶푡 method outlined in 4.10.2, the fold change in JMJD1C expression due to "Sigma 62" can be calculated as following:

2(27.62−21.09)−(33.32−28.29) ≈ 0.35355

This result concurs with the 62% knock down efficiency the manufacturer described for this shRNA.

5.4.2 Verification by western blot

Surprisingly, the shRNAs that were taken from Shakya et al. [341] showed no effect in western blot analysis. In Figure 5c of the original publication, all shRNAs showed a nearly complete abrogation of JMJD1C expression in the western blot analysis of murine embryonic fibroblasts and murine embryonic stem cells. Fig. 18, however, shows no marked difference in signal intensity between bands of JMJD1C’s molecular weight from 3T3 cells that remained either untransfected (lane ø), transfected with scrambled shRNA (lane scr) or those that were transfected with shRNA "A" (lane A) or "D" (lane D) published by Shakya et al. This implies that, unlike postulated by Shakya et al., these shRNA would lack a substantial effect on the protein levels of JMJD1C.

92 5 Results

Figure 18. Effect of Shakya et al. shRNAs A and D on the levels ofmJMJD1C 50 µg total protein per lane. anti-JMJD1C: Merck 09-817, 5 min exposure time. n=1. HiMark Pre-stained molecular weight marker; molecular weight in kDa.

Fig. 19 illustrates the successful expression of human JMJD1C from the cDNA BC143722 in 293T: The lack of any signal at 280 kDa in the lane carrying protein of untransfected (-/-) 293T cells or cells transfected with both empty vectors (emp/emp) shows that this cell line does not express JMJD1C natively. Lysate of 293T cells that were transfected with the expression vector carrying BC143722 and the lentiviral vector lacking any shRNA (BC/emp) yields a robust signal at JMJD1C’s expected molecular height.

Furthermore it shows how the self-designed shRNA "Daphne" is capable of abrogating this expression as the lysate of 293T cells simultaneously exposed to both BC143722 and piGhU6 carrying "Daphne" show a band at the same height as "BC/emp" lysate, yet with a drastically lower signal intensity, implying reduction in JMJD1C level due to the introduction of "Daphne".

Contrarily to the results of Jan Peeken, there was no endogenous expression of JMJD1C detectable in HEL cells [302]. There is, however, a strong signal visible at 180 kDa in all lysates. Due to the much smaller size, it is unlikely to be a known JMJD1C isoform. It is possible that due to homology within the KDM3 family, that Merck’s antibody 09-817 also detects KDM3B which is about 191 kDa in size. It is noteworthy that KDM3B is situated on chromosome 5 which is deranged in the HEL cell line [240].

93 5 Results

Finally, it is important to note that this lower band seems to be weakened in cells that were transfected with Daphne. Under the assumption that the band is indeed KMD3B, it could indicate a possible off-target effect within the KDM3 family.

Figure 19. Effect of the shRNA "Daphne" on the protein levels of hJMJD1C "BC" denotes BC143722. "sh" denotes shRNA. Open circle indicates unspecific band. 50 µg total protein per lane. anti-JMJD1C: Merck 09-817, 15 min exposure time; anti-β-actin: Sigma A5441, 5 min exposure time. n=3, one representative blot shown. HiRange Spectra molecular weight marker; molecular weight in kDa.

5.5 Lentiviral transduction

In order to control the efficacy of lentiviral gene transfer using piGhU6, γ2a cells were transfected with piGhU6-mJAK2푉 617퐹 . For transfection with either piGhU6-empty or piGhU6-mJAK2푉 617퐹 , 2.5 × 106 cells were plated in 10 cm culture dishes. 48 h after seeding, the cells were transduced with respective viral supernatant. 72 h after the infection, cells were harvested. FACS analysis showed marked fluorescence in both piGhU6-empty γ2a as well as piGhU6-mJAK2푉 617퐹 γ2a compared to untransduced γ2a cells. Cell lysates were used for western blot analysis. Fig. 20 shows how γ2a cells lack expression of JAK2 as mentioned in 4.12 as there is no signal at the expected protein height in either lysate from uninfected cells (left lane) or cells infected with

94 5 Results

Figure 20. Lentiviral transduction of 훾2a cells Closed arrow marks expected height based on murine JAK2 molecular weight. 50 µg total protein per lane. anti- mJAK2: Cell Signaling 3230, 1 min 30 sec exposure time; anti-β-actin: Sigma A5441, 2 min 30 sec exposure time. n=1. PageRuler Prestained Protein Ladder; molecular weight in kDa. virus carrying piGhU6 without any insert (middle lane). The right lane of the depicted blot illustrates the successful transduction of γ2a with mJAK2푉 617퐹 as infection with lentivirus carrying piGhU6-mJAK2푉 617퐹 brings about a strong signal at the height of JAK2’s expected molecular weight, proving that piGhU6 is a viable vehicle for gene transfer.

Since the mJAK2푉 617퐹 CDS is considerably longer than the shRNA cassettes it is valid to assume that transduction of shRNA for knock down should be at least as efficient as the transfer in this experiment. Yet it is important to consider the cell type to transduced as human CD34+ have been shown to have a limited ability in the initiation of reverse transcription compared to cell lines like HeLa. [404]

5.5.1 Virus production

Since the transfection of the virus producing cells with the desired vector represents one of the limiting steps in the complex production chain of lentivirus, an optimization of the transfection method is vital in order to attain relevant viral titers. In an effort to boost

95 5 Results

transfection efficiency, the established calcium precipitation method was tested against the polyethyleneiminie dendrimers. In order to evaluate the difference in gene delivery, 293T cells were transfected with pCMV6-Kan/Neo-BC143722 and protein expression evaluated by western blot analysis. The more intense signal in the lane with lysate from

Figure 21. Western blot comparing the expression of JMJD1C. AB130922. Ex- posure time 5 min. Molecular weight indicated in kDa.

the PEI transfected cells goes to show that PEI is a more efficient transfection reagent

than CaPO4 precipitate. Therefore, PEI was preferred over CaPO4 precipitation for the following rounds of virus production. Titration of lentiviral supernatant was performed as outlined in 4.13.3 in order to quantify the yield of virus production. Fig. 22 illustrates the dose-effect correlation between the amount of virions the cells were exposed toand the detectable fluorescence:

Untransduced control cells were used to account for autofluorescence (top left panel), thus sharply defining GFP positivity. Cells infected to undiluted viral supernatant (top middle panel) showed a marked increase in GFP expression as the majority of cells fluoresced intensively, with the remaining untransduced cells represented by the smaller, indistinct peak to the left. Already at a dilution by an order of magnitude (top right panel), the distribution was reversed with a large, sharp peak of untransduced cells and a distinctly separate small peak of fluorescing infected cells.

A comparison of dilution factors and the respective percentage of GFP-positive cells shows that the trend is not proportional, posing the question which dilution is optimal for back calculating the viral load. Previous work demonstrated that within the range of 5-15% estimates of the viral titer seem most reliable. [174]

96 5 Results

Figure 22. Flow cytometric analysis of Lentiviral Titration. Histograms represent fluorescence of 293T cells after exposure to serial dilutions of piGhU6-Daphne viral supernatant; numbers of bars denote percentage

97 5 Results

Therefore, the percentage of GFP+ cells of the hundredfold dilution was plugged into the formula from 4.13.3 yielding the following approximation:

4 휇푙 5 × 10 × 4.2% × 1000 푉 푖푟푖표푛푠 푚푙 = 4.2 × 107 100% × 0.05휇푙 푚푙

This result was used as the starting point for the calculation of volume of viral supernatant used in later experiments involving piGhU6-Daphne.

5.6 JMJD1C knock down in healthy donor CD34+ cells

The effect of JMJD1C silencing on the erythroid differentiation potential of healthy human HCSs was studied ex vivo via lentiviral transduction of healthy donor CD34+ cells with the self-designed shRNA "Daphne". The experiment was carried out as outlined in 4.12.2. Colonies were considered separate whenever the distance between two adjacent colonies was at least the same as the diameter of the larger colony.

Figure 23. Effect of JMJD1C knock down on the erythroid differentiation of healthy CD34+ cells. CFU-E = erythroid colony forming unit. BFU-E = erythroid blast forming unit. CFU-G = granulocytic colony forming unit. n =1

Fig. 23 illustrates the effect of the knockdown of JMJD1C on the differentiation of healthy CD34+ cell: Cells that had been infected with piGhU6 without a shRNA (empty) gave rise to about equal amounts of BFU-E as well as CFU-G colonies. Meanwhile, CD34+

98 5 Results

that had been transduced with shRNA "Daphne" (see 5.3) showed a 30% reduction in BFU-E progeny and a complete abolishment of CFU-E progeny at the same time as the amount of CFU-G colonies had more than doubled. These results imply that silencing of JMJD1C in healthy CD34+ shifts differentiation away from the erythroid differentiation towards a granulocytic differentiation.

99 6 Discussion

6 Discussion

The results presented above imply that the loss of JMJD1C influences both differentiation as well as proliferation of myeloid cells. Meanwhile JMJD1C has been acknowledged for its role in hematological disorders. Therefore, this section aims to discuss the question of JMJD1C’s enzymatic function as a point of departure for sketching cellular mechanisms that integrate the results of this study with both molecular groundwork as well as clinical outlooks.

6.1 JMJD1C’s Demethylase activity

By sequence similarity, JMJD1C had been grouped into the histone demethylase family KDM3 [5][46], but - unlike the other members of the group - the protein’s demethylase activity remains subject to debate.

For the murine protein, Kim et al. could demonstrate that a truncated version of the protein is capable of both H3K9 demethylation in cell-free biochemical assays as well as a global H3K9 demethylation in transfected cells [191]. Shakya et al., who investigated JMJD1C’s role in retention of embryonic pluripotency, showed a temporal correlation between ectopic JMJD1C expression and decrease of H3K9 methylation. [341] Kuroki et al, however, could not confirm JMJD1C’s influence on global H3K9 methylation upon their studies of JMJD1C’s influence on murine spermatogenesis. [211] Zhu et al. analyzed the bone marrow of mice suffering from MLL-AF9 leukemia by western blot and were not able to detect any difference in H3K9 methylation between f/f and -/- animals. [451]

Results for the human protein seem equally inconclusive: Wang et al., who investigated JMJD1C’s role in human embryonic stem cells (ESC) differentiation, could demonstrate that a knock down of JMJD1C increases H3K9 methylation in a region of the miR-302 promoter that is bound by Oct4. [418] Similarly, Jan Peeken showed that MNCs of PV patients contain less mono- and dimethylated H3K9 in comparison to healthy controls which has been argued to be an effect of the alleged overexpression of JMJD1C in these patients. [302] Sroczynska et al. reported no difference in H3K9 methylation after knocking down JMJD1C in HEK293T as well as SEM cells, even though they observed the most pronounced effect of JMJD1C depletion among all tested leukemia cells in SEM cells.[367] Concordantly, Chen et al. were not able to observe global H3K9 methylation changes upon

100 6 Discussion

the knock down of JMJD1C. In addition, their results contradict Kim et al. insofar as the human recombinant protein they used in cell-free assays exhibited no demethylation of any H3K9me2 substrate and only a weak demethylation of H3K9me1 polynucleosomes. [67] Lastly, a comprehensive investigation of the enzymatic properties of the human KDM3 family, could show that KDM3A and KDM3B, unlike JMJD1C, show robust demethylase activity both in cell-based as well as cell-free biochemical analysis. Furthermore, it was possible to abolish KDM3A/B activity by inserting domains or single amino acids from JMJD1C, yet it was not possible to render JMJD1C functional by introducing features of KDM3A/B. [46] In an effort to explain these findings, various authors proposed different hypotheses regarding JMJD1C’s mechanism of action to explain their findings which shall be summarized and briefly presented.

JMJD1C acts through mere scaffolding This hypothesis is an immediate reaction to the data of Brauchle et al. which showed that JMJD1C has no activity whatsoever. It is mostly based on the assumption that throughout the evolution of the KDM3/JHDM2- family, JMJD1C lost its enzymatic function through either its N-terminal part - which is disproportionately large compared to KDM3A/B - or a mutation in the C-terminal JmjC domain. [46]

Since this and many other studies have identified a plethora of interaction partners - many of which seem to be associated with chromatin - it seems logical that JMJD1C, may exert its function through establishing protein networks and complexes. The pitfall of a hypothesis that reduces JMJD1C to a sterical function is that is lies in a methodological blind spot. Brauchle et al. themselves state that JMJD1C could be an enzyme under conditions that were either not part of the experimental setup or transcend the scope of current methods, rendering it almost impossible to definitely rule out JMJD1C’s enzymatic activity.

JMJD1C demethylates proteins other than histones Watanabe et al. showed that JMJD1C is implicated in the response to DNA double strand breakage (DSB). They showed that JMJD1C is stabilized at DSB sites via binding of RNF8 as well as interaction with the N-terminal SQ-motif of RNF168. JMJD1C demethylates MDC1 at K45, thereby facilitating consecutive polyubiquitination of MDC1 by RNF8 and RNF168. Finally, demethylated MDC1 recruits RAP80-BRCA1 to the damage site. This alleged role in maintaining genome integrity let the authors postulate JMJD1C as a tumor suppressor gene which they saw backed by decreased JMJD1C expression in breast cancer tissue. [422]

101 6 Discussion

These findings prompt the question whether JMJD1C could act as both a potential oncogene, by promoting self-renewal and preventing differentiation, as well as a tumor suppressor through promoting DNA repair.

Sáez et al. presented a P163L mutation of JMJD1C in a patient suffering from Rett disease, an autism spectrum condition associated with severe neurological defects. This mutant protein was unable to demethyalate MDC1. A hematological phenotype, however, was not reported by the authors [327]. This may be explained by Zhu et al.’s observation that JMJD1C seems dispensable for steady-state hematopoiesis. [451]

Two recent findings suggest that this non-histone effect of JMJD1C could be relevant in the context of hematopoiesis: Watanabe et al. were drawn to hypothesize that JMJD1C might interact with other ubiquitin ligases as dual depletion of RNF8 and RNF168 did not abrogate JMJD1C mediated MDC1 demethylation. [422] Interestingly, Manesia et al. have recently identified ubiquitin ligase UHRF1 as essential for the equivalent of fetal liver hematopoiesis in D. rerio.[244] UHRF1 bears a dual function by controlling epigenetic modification as well as initiating DNA damage repair via ubiquitination and interaction with MDC1. [354] Data from the BLUEPRINT database which harbors hematopoietic gene expression profiles suggests that UHRF1 is more abundant in erythroblasts thanin hematopoietic stem cells or progenitor cells. [179]

Furthermore, Brauchle et al. identified ATR as an interaction partner of JMJD1C by quantitative mass spectrometry. ATR is - like ATM - one of the major protein kinases in the DDR, but unlike ATM not involved in DSB but rather replication associated DNA damage. [227] Vanessa Zimmermann showed a marked interaction between NFE2 and MDC1 using SILAC-based quantitative immunoprecipitation of HA-tagged wild-type NFE2. [452]

Taken together, these results hint towards the important connection in hematopoietic stem cells between maintenance of genomic stability on the one hand and stemness on the other hand. [37] Experiments with human HSCs that were transplanted into mice showed that prior irradiation of the HSCs biased the resulting hematopoiesis towards the myeloid lineage in a dose-dependent manner and mimicked the phenotype of aged human hematopoiesis. [417]

Thus, an investigation of JMJD1C’s role at this intersection between genomic stress and myeloid differentiation may elucidate not only the physiology of stressed oraged

102 6 Discussion

hematopoiesis but also response mechanisms of myeloid diseases treated with genotoxic agents.

JMJD1C demethylates histones under special conditions A different hypothesis is to presume that JMJD1C might exhibit either a low substrate affinity, comparable to other histone demethylases like LSD1 [67], or a decreased DNA binding capability as its zinc finger domain differs from KDM3A/B. [46]

In either scenario JMJD1C would depend on its DNA-binding interaction partners to exert its function, which in turn renders JMJD1C’s effect dependent on the abundance of these partner in the respective type of cell. This contextualization of JMJD1C’s activity offers an explanation for the conflicting results from different experimental setups:

In cell free assays, Chen et al. were able to demonstrate that upon addition of a nuclear extract fraction lacking a histone demethalyase activity of its own, JMJD1C’s weak demethylation activity towards H3K9me1 in polynucleosomes became markedly increased. [67] Sroczynska et al. noted that - unlike in leukemia cells - knock down of JMJD1C had no effect on the proliferation of the human osteosarcoma cell line U2OS.[367] This effect could be explained by U2OS cells lacking the promoting factors that Chenet al. postulated to be present in certain nuclear extracts.

Due to their canonical ability of DNA binding, transcription factors represent candidate proteins, as they could attract JMJD1C and position it to unfold its chromatin modifying properties. Indeed, JMJD1C binds to the transcription factor complex built up around the oncogenic fusion protein RUNX1-RUNX1T1. [67] Besides the fusion protein, the complex abbreviated AETFC, is composed of CFBβ, HEB or E2A, Lyl1, Lmo2 and Ldb1. [378]

JMJD1C’s interaction with Lyl1 and HEB seems to be especially relevant as CoIP showed direct interaction of JMJD1C with both proteins. Knock down of either of the two proteins decreases JMJD1C recruitment to the DNA in several AML cell lines; an observation that could be confirmed by overlapping ChIP peaks for JMJD1C and Lyl1. [67]

This interaction is likely to be relevant outside the context of AML as transcription factor complexes containing mostly the same proteins as the AETFC have been described for erythroid progenitors (see Fig.3 for an overview of erythroid transcription factors).

103 6 Discussion

[433][416] CoIP experiments also showed that JMJD1C interacts with transcription factor HOXA9 [451] which plays an important role in healthy and diseased myeloid differentiation. [194][278] Interestingly, HOXA9 has already been shown to interact with TRIP6, a LIM-domain protein like Lmo2, that interacts with the thyroid hormone receptor as JMJD1C which is also known as TRIP8. [218][323]

The existence of a protein complex entailing HOXA9, TRIP6, JMJD1C and possibly the thyroid hormone receptor can be speculated as one takes into account the effects of the thyroid hormone receptor on erythropoiesis which will be discussed in the following section.

6.2 JMJD1C’s Role in Stemness and Differentiation

Recent publications dissecting the role of JMJD1C in AML have shown that it is necessary for leukemogenesis both in the context of RUNX1-RUNX1T1 [67] as well as MLL-AF9 and HOXA9. [451]

At the same time, their data suggest that - in contrast to the diseased state - JMJD1C is largely dispensible for physiological hematopoiesis. Though Jmjd1c -/- animals showed decreased bone marrow cellularity and impaired competitiveness in repopulation, blood counts did not differ from vav-cre controls. The authors argue that the loss ofJMJD1C leads to an increase in both proliferation as well as apoptosis in HSCs and an impaired differentiation ability.

Looking at the cell numbers in the different subsets of BM-MNCs, it is striking tosee that the effect of JMJD1C depletion inversely correlates with the expression of Jmjd1c evaluated by qRT-PCR: LT-HSCs and MEP, which express the highest levels of JMJD1C, do not seem to be affected in their abundance, whereas CMPs, GMPs as well as CLPs, which share a commonly low expression of JMJD1C, are significantly decreased in in the marrow of -/- animals. [451]

These results prompt a model wherein a low expression level JMJD1C is important for the maintenance of a myeloid progenitor phenotype, while higher expressions are associated with stemness as well as the MegE lineage. The model is apt to explain and unify the following findings:

104 6 Discussion

1) Leukemogenesis can be explained by unphysiologically high levels of JMJD1C in myeloid progenitors endowing the cells with characteristics (e.g. proliferation, stemness) of highly expressing stem cells. Interestingly, the cells are not shifted to the MegE lineage, though it also seems to be associated with high JMJD1C expression.

2) The observation that loss of Jmjd1c diminishes leukemic stem cells would constitute the reverse of the above mentioned principle.

3) Knockdown of JMJD1C presented in section 5.6 can be conceptualized as shift from high to low JMJD1C expression; in line with this reasoning, a shift from the typically highly expressing erythroid to the granulocytic lineage with lower JMJD1C levels was observed.

4) The similar expression pattern and behavior upon knockout put MEPs and HSCs in proximity, which echos with the recently proposed models of hematopoiesis outlined in 3.1

Under the presumption of such a model, it seems no longer contradictory that JMJD1C is on the one hand involved in stemness and pluripotency - which will be elaborated further below - and on the other hand is a target gene of NFE2 [302] which has been established as a key component of the MegE differentiation [351][217][334][350]; it rather underpins the notion of a tie between the MegE lineage and the stem cell phenotype.

Yet caution should be exercised when comparing the results presented in this work with those of Zhu et al.: First, the experimental approaches - FACS analysis of bone marrow from knockout mice versus colony assay of healthy CD34+ cells with lentiviral knockdown - differ significantly. Second, and more importantly, the murine and human hematopoiesis are alike yet differ in crucial parameters as was described in 3.1.

That species difference in hematopoiesis should be taken into account is not only illustrated by the complete ablation of erythro- and megakaryopoiesis in Jmjd1c -/- D. rerio [127] but the fact that two highly powered meta-analyses of genome wide association studies found JMJD1C to be influential on mean platelet value as well as platelet countin humans; with both findings going against the notion that JMJD1C is devoid of a roll in physiological hematopoiesis. [127][363]

105 6 Discussion

Finally, one pressing question remains: What are the implications for the classical MPN entities? Following the above mentioned model, it is to speculate about the influence of JMJD1C overexpression in MPN as the lineage most affected by the condition is characterized by a constitutively high JMJD1C level. Additional JMJD1C might allow for increased proliferation without interfering with MegE maturation; an important distinction from AML. Yet, JMJD1C might as well be subject of a ceiling effect where additional JMJD1C has a large effect in the context of low expression (e.g. leukemogenesis in myeloid progenitors), while being idle in the context of MegE’s relatively saturated cellular mechanisms. If indeed relevant to MPN initiation and maintenance, the investigation of JMJD1C’s precise mechanism of action would become of utmost interest with regard to its potential "drugability". The following paragraphs will elaborate different cellular axes through which JMJD1C may influence stemness and differentiation in hematopoiesis, beginning with nuclear receptors as well documented interaction partners.

Thyroid Hormone Receptor Clinical observations correlate hypothyroidism with decreased erythropoiesis and hyperthyroidism with increased erythopoiesis. [85] Even though experimental evidence showed that - similar to JMJD1C - knock out of thyroid hormon receptor (TR) 훼 as well as β did not affect steady state hematopoiesis, thyroxin levels ineu- thyroid subjects positively correlate with erythrocyte indices. [336]

In spite of the incomplete understanding underlying the mechanism of this correlation, the direct influence of the thyroid hormone receptor on hematopoiesis is illustrated through the avian erythroblastosis virus (AEV): its oncogene v-ErbA represents a mutated version of the receptor that resembles the protein’s ligand-free conformation. v-ErbA leads to increased proliferation as well as a complete block of terminal differentiation in infected erythroblasts.

It was demonstrated that erythroblasts overexpressing TRα displayed a similar phenotype when cultured in absence of thyroid hormone Triiodothyronine (T3); an effect that could

be abolished upon addition of T3 to the culture. [30] There is evidence that the link between the TR and erythroid differentiation also exists in humans as Shiraishi etal.

could show that T3 at concentrations of 1-10 nM increased hemin-induced differentiation of human erythroleukemia cell line K562. [348]

Based upon these results, it was postulated that the thyroid hormone receptor functions as a ligand-operated switch in erythroid differentiation which regulates the transition

106 6 Discussion

from self-renewal to terminal differentiation. [30] To this end, it seems noteworthy that a significant association between thyroid disease and acute myeloid leukemia was foundin a recent case control study. [295]

Given that JMJD1C was first appreciated for its interaction with the thyroid hormone recptor [218], its newly discovered role in leukemogenesis [67] should prompt an investigation of JMJD1C’s possible role as a link between thyroid disease and myeloid (pre-)malignancies.

Androgen Receptor Although the influence of androgens on erythropoiesis was already observed 50 years ago [133][431], the mechanism underlying this effect remains elusive. While some results suggest that androgens might exercise their effect on hematopoiesis through means other than the canonical androgen receptor [51], many others prove the contrary both in vivo [54] and in vitro [75].

The hematopoietic effects of androgens are clinically relevant insofar as antiandrogenic treatment in men suffering from prostatic cancer reduces the risk of venous thromboembolism [213], while a decrease in hemoglobin content, red blood cell count as well as mean corpuscular volume has been documented for women receiving anti-androgens. Meanwhile, in women affected by congenital adrenal hyperplasia, serum androgen levels were positively correlated with hemoglobin content and hematocrit. [180] There are even singular reports of polycythemia caused by congenital adrenal hyperplasia. [11][412]. These clinical effects are used for the alleviation of cytopenia in PMF. The androgen analog danozol, for example, was found to have a 30% response rate in anemic PMF patients. [64]

Evidence that JMJD1C may be involved in the effects of androgens is provided by Wolfet al. who found it to act as a coactivator of the androgen receptor. [434] Since this report is of rather limited scope, Wolf et al.’s results should serve as a start for further investigation of the mechanism underlying this connection as well as its implication in the hematopoietic system given the above mentioned observations.

NR2C1/NR2C2 orphan receptors and c-myb/p300 It was demonstrated by quantitative mass spectrometry that JMJD1C C-terminally interacts with MYBBP1A. [46] MYBBP1A belongs to the p160 family of nuclear receptor coactivators and like JMJD1C possess a LXXLL motif. MYBBP1A is implicated in the erythroid differentiation through the inhibition of two proteins it binds to:

107 6 Discussion

First, MYBBP1A binds to the leucin zipper of c-myb’s negative regulatory domain, thereby reducing c-myb transactivation by 100-fold [386]. Assuming that MYBBP1A binds both JMJD1C and c-myb in vivo, one could explain the link between JMJD1C and the c-myb target gene signature described by Sroczynska et al. [367] while at the same time explaining the weak CoIP signal forJMJD1C and c-myb [67] through the indirect nature of the interaction.

In definitive erythropoiesis, c-myb exerts an antidifferential effect which is necessary for the proliferation of erythroblasts. [134] Histone modifying protein p300 can increase c-myb transactivation by up to threefold through its myb-interacting KIX domain. [94] Accord- ingly, animals carrying mutations in the interaction domains of either c-myb or p300 suffer from macrocytic and decreased red blood cell counts while having excess megakaryocytes and increased as well as enlarged platelets. [330]

Beyond its relevance in physiological hematopoiesis, the c-myb/p300 interaction is also required for myeloid leukemogenesis. Cells and animals carrying mutations in the interaction domain fail to develop leukemia upon viral transduction with RUNX1- RUNX1T1 or MLL-AF9. [301] The triterpenoid Celastrol has been identified as a specific disruptor of the interaction and data from mice as well as patient material experiments show decreased proliferation and prolonged survival, making the c-myb:p300 interaction a promising target for AML therapy. [405]. Yet there are chances that Celastrol might also prove to be relevant in the case of MPN as the role of c-myb [48] and p300 [15] in PV has been established.

Looping back to the concept of JMJD1C providing a scaffold for protein complexes, itis possible that in addition to the indirect interaction through MYBBP1A, JMJD1C might directly interact with the c-myb:p300-complex. Such an interaction seems possible not only due to the stereochemistry of these large proteins, but also in light of new evidence that reveals the role of adapter proteins; they bind to the complex and are needed for proper transactivation in hematopoietic cells. [34] Thus, a role in the c-myb:p300 complex should be considered when analyzing the underlying molecular mechanism of hematopoietic JMJD1C knock out phenotypes.

MYBBP1A’s other relevant interaction partner is PGC-1α. PGC-1α influences erythropoiesis through binding NR2C1/NR2C2 4 orphan receptor dimers and potentiating their transactivation. The NR2C1/NR2C2 orphan receptors regulate globin expression [385] through binding to direct repeat elements with subsequent formation of the direct repeat

4also referred to as TR2/TR4

108 6 Discussion

Figure 24. Illustration of a possible network governing erythroid differentiation. Blue arrows represent positive signaling or interaction. Black arrows represent inhibition.

109 6 Discussion erythroid definitive (DRED) complex. DNMT1 and LSD1 constitute the core of the DRED complex which attracts various other chromatin modification complexes such as the NuRD complex or the CoREST complex. [78] Furthermore, NR2C2 seems to heterodimerize with some of the above mentioned nuclear receptors, enabling NR2C2 to modulate their influence on erythroid differentiation. [343][153]

NR2C1/NR2C2-deficient animals display a lack of GATA1, arresting their erythropoiesis in a CD71ℎ푖 Ter119ℎ푖 state. [80] Tanabe et al. could demonstrate both in humans and mice that NR2C1/NR2C2 repress GATA1 only in the later stages of definitive erythropoiesis by binding to the GATA1 hematopoietic enhancer (G1HE). This suggests that TR2/4 might play an important role in mediating the transition to the last, GATA1 independent phase of terminal erythroid differentiation. [385]

In line with the observation that PGC-1α potentiates NR2C1/2 transactivation, mice carrying a compound heterozygous deletion of PGC-1α and PGC-1β develop similar phenotypes displaying decreased globin gene expression at all stages of development as well as erythrocyte maturation deficiencies. [79]. MYBBP1A binds to the negative regulatory domain of transcriptional coactivator PGC-1α, decreasing its transactivating actity. Upon phosphorylation of PGC-1α via the p38 MAPK pathway, MYBBP1A can no longer bind to PGC-1α.[106]

This means that JMJD1C would finally act as an inhibitor of GATA1, and that p38 MAPK signaling leads to the release of this inhibition. Terminal differentiation is ushered in via EPO stimulation which allows cytosolic GATA1 to translocate into the nucleus and abrogate c-myb expression by binding to its promoter region. [134][29] Hence, MYBBP1A’s repression of c-myb synergizes with GATA1 in switching the progenitor’s differentiation program.

As part of the differentiation program, GATA1 increases the expression of NFE2, which a) has been shown to directly interact with MYBBP1A [452] and b) upregulates the expression of MYBBP1A’s interaction partner JMJD1C. Thus, under the premise that the interaction with MYBBP1A is functional, JMJD1C would be part of a regulatory network of erythroid maturation that upon p38 MAPK as well as EPO signaling, mediates the transition to the GATA1 dependent phase of differentiation.

110 6 Discussion

Oct4, Nanog and TET2 Katoh et al. first described that JMJD1C’s promoter features an Oct4 binding site, yet their postulation that this binding site must be functional as Oct4 levels in ECSs correlate with higher JMJD1C expression has never been mechanistically proven. [181] Using ChIP and CoIP, Shakaya et al. demonstrate that Oct4 recruits JMJD1C to Oct4 binding sites where it supposedly demethylates H3K9, thereby promoting Oct4-driven gene expression. One of the genes whose promoter is bound by both Oct4 and JMJD1C is NANOG.[341]

Interestingly, Costa et al. found JMJD1C to be an interaction partner of Nanog which they describe as a key factor for induction of pluripotency. In addition, they found methylcytosine hydroxylases TET1 and TET2 to interact equally strong with Nanog as with JMJD1C, proposing a similar mechanism as Shakya et al.: Nanog recruits TET1 and TET2 to target gene promoters to epigenetically favor expression, thereby contributing to a stem-like cellular program. [77] Especially with regard to the frequency of TET2 mutations in MPN, it seems meritable to investigate whether the loss of TET2 might influence its interaction partner Nanog and in turn JMJD1C.

Paralleling nanog, Wang et al. demonstrated that miR-302 - a target gene of Oct4 involved in maintaining a pluripotency pattern - is upregulated by JMJD1C in human ESCs. [418] miR-302 has several effects: First, it has been shown that miR-302 represses Cyclin D1 translation in human ESCs which leads to a shortened G1 phase and a more "stem cell-like" cell cycle. [59] This finding is mirrored by the cyclin D1 gene set enrichment that Zhu et al. observed in the HSCs of JMJD1C -/- mice. [451] Second, adipose tissue-derived stem cells (ADSCs) transfected with miR-302 showed an upregulation of pluripotency related genes Oct4, Sox2 and Nanog [382] which represents a feed-back loop reinforcing the Oct4 pluripotency program.

Tying back into the section above, orphan nuclear receptor NR2C1 binds DR elements in the Oct4 promoter. It thereby controls Oct4 expression and does so depending on its current SUMOylation state: UnSUMOylated NR2C1 attracts coactivator Pcaf which consequently increases Oct4 expression. UnSUMOylated NR2C1 becomes recruited to PML nuclear bodies (NB) where it is SUMOylated by Ubc9 and Pias and is released from the PML NB. SUMOylated NR2C1, in turn, attracts corepressor Rip140 which progressively displaces Pcaf and leads to a decrease in Oct4 expression. [293] Furthermore, it should be noted that RNF8 and RNF168 - both of which interact with JMJD1C as explained in 6.1 - have been shown to regulate PML NB. [349] Taken together these observations not only imply a second autoregulatory loop for Oct4 expression, but

111 6 Discussion

also a representation of the above mentioned connection between DDR and stem cell characteristics.

Even though it has been recognized that Oct4 is largely irrelevant for somatic stem cells including HSCs under physiological conditions [222] more recent research sheds new light on the subject: Experimentally, it could be demonstrated that unphysiological Oct4 expression in CD34+ cells stimulated by the compound OAC-1 augments ex vivo HSC expansion as well as in vivo repopulation ability. The study’s authors explain the phenomenon through increased HOXB4 expression. [159] It should be noted, though, that the authors overlooked the possibility of an implication of Nanog and TET1 which are likewise upregulated by OAC-1. [226] Clinically, Oct4 has been found to be overexpressed in AML and is associated with poorer prognosis. [441] In the wake of JMJD1C’s role in MPN and the connection between JMJD1C and Oct4, an investigation of Oct4 expression in MPN seems almost imperative as such data are missing to date.

Lastly, the Oct4 may be the answer to the puzzling fact that JMJD1C is equally abundant in testicular tissue [191][211] and intracranial germ cell turmors [420] as it is in HSCs: About 0.02% of murine bone marrow mononuclear cells are very small embryonic-like stem cells (BM-VSEL) that are able to differentiate into tissue of all three germ layers. These Sca+ lin−CD45− cells were found to not only express markers of pluripotency such as SSEA-1, Oct4 and Nanog [209] but in addition GBX2, FGF5 as well as Nodal.

Therefore, it was proposed that they arise from the epiblast during development and are thereby closely related to primordial germ cells (PGC) which migrate through the primitive ridge to populate the aorta-gonad-mesonephros (AGM) region. Since the first appearance of definitive HSCs in the AGM region coincides with thePGCs’ migration, the authors hypothesize that the three cell populations could share a common origin. [347]

This theory is backed by a) BM-VSELs’ retention of hematopoietic potential [347] and b) the recent discovery that HSCs express both follicle stimulating hormone FSH as well as luteinizing hormone LH receptor on their surface. [266] The functionality of the receptors has been demonstrated in a human study in which the number of both VSELs as well as HSCs in the peripheral blood of women increased upon FSH stimulation. [444] Since JMJD1C is a target gene of Oct4 and expressed in both hematopoietic as well as germ cells it provides an interesting starting point to further dwell upon this otherwise elusive connection.

112 6 Discussion

6.3 Outlook

The experiments described herein demonstrate JMJD1C’s relevance for the erythroid differentiation of healthy donor hematopoietic stem cells. In light of its roleinthe pathogenesis of AML, this should spur further investigation of JMDJ1C’s function in both the healthy and the diseased myeloid compartment. In trying to mechanistically characterize JMJD1C, the crucial question of JMJD1C’s enzymatic capability towers over future research for a twofold reason:

On the one hand, by directing basic science analysis of JMJD1C in relevant directions. Under the presumption that a final answer to this question can be achieved, it would be decisive in selecting between either epigenetic studies built around histone modification or proteomic studies aimed at grasping the outlined interactions and networks. To this end, quantitative immunoprecipitation approaches such as SILAC could be employed to rank the most promising proteins to study.

On the other hand, JMJD1C’s enzymatic function is deeply intertwined with its potential clinical relevance: The example of Celastrol goes to show that catalytic activity does not constitute a conditio sine qua non for drugability, yet as the pharmacological inhibition of dysregulated enzymes has taken center stage among modern targeted therapies it would speed up translation to the bedside.

113 List of Figures

List of Figures

1 Comparison of three current models of hematopoiesis. Note that surface markers mirror the murine hematopoietic system. Adapted from [280] 4 2 Toggle switch of PU.1 and GATA1. Adapted from [17]...... 6 3 Genetic regulatory network of the erythroid differentiation. Adapted from [379]...... 7 4 Abridged representation of the classification proposed in the WHO-CMP. Adapted from [264]...... 10 5 Overview of covalent histone modifications and their known positions. Boxed part of the aa sequences denote the histone proteins’ globular domain. Adapted from [158]...... 32 6 Reaction mechanism of Jmjc oxygenases. Adapted from [430].... 34 7 Phylogenetic tree of Jmjc domain-containing proteins. Proteins grouped by similarity of the catalytic site. Subgroups represented by colored lines. Adapted from [430]...... 35 8 Experimental Setup for CD34+ Methylcellulose Assay. Numbers represent amount of cells...... 75 9 Illustration of the pLeGo-iG casette. Figure adapted from [423].. 77 10 Excision of Jmjd2c cDNA from pCMV6-Kan/Neo The lane marked with * shows undigested pCMV6-Kan/Neo-Jmjd2c. The adjacent lane marked with # shows the same vector linearized with EcoRI. The lanes to the right of the 2lg ladder show shorter, supercoiled vector devoid of both Jmjd2c cDNA and the adjacent EcoRI restriction site...... 82 11 HindIII/StuI digest of pCMV6-Kan/Neo-mKIAA1380.# and * denote lanes that feature a banding pattern suggestive of mKIAA insertion. 2 lg ladder. Fragment length in kb...... 83 12 XhoI/NcoI control digest of pCMV6-Kan/Neo-mKIAA1380.# plasmid carries insert in correct orientation; * plasmid carries an inverted insert. 2 lg ladder...... 84 13 XbaI digest of pCMV6-Kan/Neo-myc-JMJD1C mini culture DNA.... 86 14 Candidate antibodies against mJMJD1C ...... 87 15 Merck 09-817. Rabbit polyclonal antibody. 1:1000 in TBS + 0.1% T +5 % skim milk...... 88 16 HpaI/EcoRI digest of piGhU6 carrying shRNAs ...... 90

114 List of Figures

17 qRT-PCR for JMJD1C from RNA of 3T3 cells transduced with shRNA "Sigma 62" x-axis: number of PCR cycles; y-axis: detected fluorescence in arbitrary units, logarithmic scale...... 91 18 Effect of Shakya et al. shRNAs A and D on the levels ofmJMJD1C 50 µg total protein per lane. anti-JMJD1C: Merck 09-817, 5 min exposure time. n=1. HiMark Pre-stained molecular weight marker; molecular weight in kDa...... 93 19 Effect of the shRNA "Daphne" on the protein levels of hJMJD1C "BC" denotes BC143722. "sh" denotes shRNA. Open circle indicates unspecific band. 50 µg total protein per lane. anti-JMJD1C: Merck 09-817, 15 min exposure time; anti-β-actin: Sigma A5441, 5 min exposure time. n=3, one representative blot shown. HiRange Spectra molecular weight marker; molecular weight in kDa...... 94 20 Lentiviral transduction of 훾2a cells Closed arrow marks expected height based on murine JAK2 molecular weight. 50 µg total protein per lane. anti-mJAK2: Cell Signaling 3230, 1 min 30 sec exposure time; anti-β-actin: Sigma A5441, 2 min 30 sec exposure time. n=1. PageRuler Prestained Protein Ladder; molecular weight in kDa...... 95 21 Western blot comparing the expression of JMJD1C. AB130922. Exposure time 5 min. Molecular weight indicated in kDa...... 96 22 Flow cytometric analysis of Lentiviral Titration. Histograms represent fluorescence of 293T cells after exposure to serial dilutions ofpiGhU6- Daphne viral supernatant; numbers of bars denote percentage...... 97 23 Effect of JMJD1C knock down on the erythroid differentiation of healthy CD34+ cells. CFU-E = erythroid colony forming unit. BFU- E = erythroid blast forming unit. CFU-G = granulocytic colony forming unit. n =1...... 98 24 Illustration of a possible network governing erythroid differentiation. Blue arrows represent positive signaling or interaction. Black arrows represent inhibition...... 109

115 List of Tables

List of Tables

1 List of entities classified as Myeloproliferative Neoplasms...... 10 2 2016 WHO diagnostic criteria for PV...... 22 3 Risk stratification for Thrombosis in PV...... 23 4 2016 WHO diagnostic criteria for ET...... 25 5 Risk stratification for Thrombosis in ET and treatment adapted from[396] and [119]...... 26 6 2016 WHO diagnostic criteria for PMF...... 28 7 List of plasmids used for cloning...... 44 8 Restriction exzymes by New England Biolabs, USA...... 44 9 Settings used for BLOCK-iT RNAi Designer ...... 48 10 Criteria for potent shRNA according to Li et al. [225]...... 49 11 10× Annealing Buffer...... 49 12 Composition of Lysogeny Broth (LB)...... 50 13 Composition of SOC medium...... 51 14 50× Tris-Acetate EDTA (TAE) buffer...... 52 15 10× Loading Dye...... 52 16 Agarose concentration as a function of DNA fragment length...... 52 17 Reverse transcription PCR program...... 56 18 Materials needed for colony PCR...... 57 19 List of primers used for colony PCR...... 59 20 PCR program used for colony PCR...... 59 21 TaqMan○R Assays used for qRT-PCR...... 60 22 PCR program used for qRT-PCR...... 61 23 Composition of RIPA Buffer...... 63 24 Composition of SDS-Polyacrylamide gels...... 64 25 5× Loading Buffer...... 64 26 10× SDS Running Buffer...... 65 27 Composition of Solutions for Coomassie Staining...... 65 28 Solutions for Antibody Detection...... 67 29 List of Antibodies used for Protein Detection...... 68 30 Stripping Buffer...... 68 31 Materials needed for Calcium chloride precipitation...... 71 32 2× BES Buffer...... 71 33 Materials needed for Polyethyleneimine transfection...... 72

116 List of Tables

34 Materials needed for the isolation of CD34+ cells...... 72 35 Materials needed for handling CD34+ cells...... 74 36 Supplementation of StemSpan○R for cultivation of CD34+cells...... 74 37 myc-tag Forward Primers...... 85 38 myc-tag Reverse Primers...... 85 39 Expected banding patterns for XbaI digestion of myc-tag JMJD1C expression vectors...... 86 40 Daphne...... 89 41 Custom...... 89 42 Sigma 62...... 89 43 shRNAs taken from Shakya et al. [341]...... 89

44 Avarage 퐶푡 value The numbers designate the average 퐶푡 over all replica . 92

117 References

References

[1] O. Abdel-Wahab, M. Adli, L. M. LaFave, J. Gao, T. Hricik, A. H. Shih, S. Pandey, J. P. Patel, Y. R. Chung, R. Koche, F. Perna, X. Zhao, J. E. Taylor, C. Y. Park, M. Carroll, A. Melnick, S. D. Nimer, J. D. Jaffe, I. Aifantis, B. E. Bernstein, and R. L. Levine. ASXL1 mutations promote myeloid transformation through loss of PRC2-mediated gene repression. Cancer Cell, 22(2):180–193, Aug 2012.

[2] O. Abdel-Wahab, A. Pardanani, J. Patel, M. Wadleigh, T. Lasho, A. Heguy, M. Beran, D. G. Gilliland, R. L. Levine, and A. Tefferi. Concomitant analysis of EZH2 and ASXL1 mutations in myelofibrosis, chronic myelomonocytic leukemia and blast-phase myeloproliferative neoplasms. Leukemia, 25(7):1200–1202, Jul 2011.

[3] S. Abramson, R. G. Miller, and R. A. Phillips. The identification in adult bone marrow of pluripotent and restricted stem cells of the myeloid and lymphoid systems. J. Exp. Med., 145(6):1567–1579, Jun 1977.

[4] M. Acar, K. S. Kocherlakota, M. M. Murphy, J. G. Peyer, H. Oguro, C. N. Inra, C. Jaiyeola, Z. Zhao, K. Luby-Phelps, and S. J. Morrison. Deep imaging of bone marrow shows non-dividing stem cells are mainly perisinusoidal. Nature, 526(7571):126–130, Oct 2015.

[5] S. L. Accari and P. R. Fisher. Emerging Roles of JmjC Domain-Containing Proteins. Int Rev Cell Mol Biol, 319:165–220, 2015.

[6] Adolfsson, J. and Mansson, R. and Buza-Vidas, N. and Hultquist, A. and Liuba, K. and Jensen, C. T. and Bryder, D. and Yang, L. and Borge, O.-J. and Thoren, L.A.M. and Anderson, K. and Sitnicka, E. and Sasaki, Y. and Sigvardsson, M. and Jacobsen, S. E. W. Identification of Flt3+ Lympho-Myeloid Stem Cells Lacking Erythro-Megakaryocytic Potential: A Revised Road Map for Adult Blood Lineage Commitment. Cell, 121:295–306, April 2005.

[7] K. Agger, J. Christensen, P. A. Cloos, and K. Helin. The emerging functions of histone demethylases. Curr. Opin. Genet. Dev., 18(2):159–168, Apr 2008.

[8] H. Akada, D. Yan, H. Zou, S. Fiering, R. E. Hutchison, and M. G. Mohi. Condi- tional expression of heterozygous or homozygous Jak2V617F from its endogenous

118 References

promoter induces a polycythemia vera-like disease. Blood, 115(17):3589–3597, Apr 2010.

[9] K. Akashi, D. Traver, T. Miyamoto, and I. L. Weissman. A clonogenic common myeloid progenitor that gives rise to all myeloid lineages. Nature, 404(6774):193–197, Mar 2000.

[10] A. Akinc, M. Thomas, A. M. Klibanov, and R. Langer. Exploring polyethylenimine- mediated DNA transfection and the proton sponge hypothesis. J Gene Med, 7(5):657–663, May 2005.

[11] M. M. Albareda, J. Rodriguez-Espinosa, A. Remacha, N. Prat, and S. M. Webb. Polycythemia in a patient with 21-hydroxylase deficiency. Haematologica, 85(E- letters):E08, 2000.

[12] C. D. Allis and T. Jenuwein. The molecular hallmarks of epigenetic control. Nat. Rev. Genet., 17(8):487–500, 08 2016.

[13] H. S. Amran, M. A. Aziz, E. George, N. Mahmud, T. Y. Lee, and S. Md Noor. Secondary polycythaemia in a Malay girl with homozygous Hb Tak. Malays J Pathol, 39(3):321–326, Dec 2017.

[14] R. Anand and R. Marmorstein. Structure and mechanism of lysine-specific demethylase enzymes. J. Biol. Chem., 282(49):35425–35429, Dec 2007.

[15] C. L. Andersen, H. Hasselbalch, and K. Grønbæk. Lack of somatic mutations in the catalytic domains of CREBBP and EP300 genes implies a role for histone deacetylase inhibition in myeloproliferative neoplasms. Leuk. Res., 36(4):485–487, Apr 2012.

[16] M. Araki, Y. Yang, N. Masubuchi, Y. Hironaka, H. Takei, S. Morishita, Y. Mizukami, S. Kan, S. Shirane, Y. Edahiro, Y. Sunami, A. Ohsaka, and N. Komatsu. Ac- tivation of the thrombopoietin receptor by mutant calreticulin in CALR-mutant myeloproliferative neoplasms. Blood, 127(10):1307–1316, Mar 2016.

[17] Y. Arinobu, S. Mizuno, Y. Chong, H. Shigematsu, T. Iino, H. Iwasaki, T. Graf, R. Mayfield, S. Chan, P. Kastner, and K. Akashi. Reciprocal activation ofGATA-1 and PU.1 marks initial specification of hematopoietic stem cells into myeloerythroid and myelolymphoid lineages. Cell Stem Cell, 1(4):416–427, Oct 2007.

119 References

[18] K. Aumann, A. V. Frey, A. M. May, D. Hauschke, C. Kreutz, J. P. Marx, J. Timmer, M. Werner, and H. L. Pahl. Subcellular mislocalization of the transcription factor NF-E2 in erythroid cells discriminates prefibrotic primary myelofibrosis from essential thrombocythemia. Blood, 122(1):93–99, Jul 2013.

[19] V. C. Auyeung, I. Ulitsky, S. E. McGeary, and D. P. Bartel. Beyond secondary structure: primary-sequence determinants license pri-miRNA hairpins for processing. Cell, 152(4):844–858, Feb 2013.

[20] S. Babovic and C. J. Eaves. Hierarchical organization of fetal and adult hematopoietic stem cells. Exp. Cell Res., 329(2):185–191, Dec 2014.

[21] M. Ballmaier and M. Germeshausen. Advances in the understanding of congenital amegakaryocytic thrombocytopenia. Br. J. Haematol., 146(1):3–16, Jun 2009.

[22] T. Barbui, G. Barosi, G. Birgegard, F. Cervantes, G. Finazzi, M. Griesshammer, C. Harrison, H. C. Hasselbalch, R. Hehlmann, R. Hoffman, J. J. Kiladjian, N. Kroger, R. Mesa, M. F. McMullin, A. Pardanani, F. Passamonti, A. M. Vannucchi, A. Reiter, R. T. Silver, S. Verstovsek, and A. Tefferi. Philadelphia- negative classical myeloproliferative neoplasms: critical concepts and management recommendations from European LeukemiaNet. J. Clin. Oncol., 29(6):761–770, Feb 2011.

[23] T. Barbui, G. Finazzi, A. Carobbio, J. Thiele, F. Passamonti, E. Rumi, M. Rug- geri, F. Rodeghiero, M. L. Randi, I. Bertozzi, H. Gisslinger, V. Buxhofer-Ausch, V. De Stefano, S. Betti, A. Rambaldi, A. M. Vannucchi, and A. Tefferi. De- velopment and validation of an International Prognostic Score of thrombosis in World Health Organization-essential thrombocythemia (IPSET-thrombosis). Blood, 120(26):5128–5133, Dec 2012.

[24] T. Barbui, J. Thiele, H. Gisslinger, G. Finazzi, A. Carobbio, E. Rumi, M. Luigia Randi, I. Betozzi, A. M. Vannucchi, L. Pieri, V. Carrai, B. Gisslinger, L. Mullauer, M. Ruggeri, A. Rambaldi, and A. Tefferi. Masked polycythemia vera (mPV): results of an international study. Am. J. Hematol., 89(1):52–54, Jan 2014.

[25] T. Barbui, J. Thiele, H. Gisslinger, G. Finazzi, A. M. Vannucchi, and A. Tefferi. The 2016 revision of WHO classification of myeloproliferative neoplasms: Clinical and molecular advances. Blood Rev., 30(6):453–459, Nov 2016.

120 References

[26] T. Barbui, J. Thiele, A. M. Vannucchi, and A. Tefferi. Myeloproliferative neoplasms: Morphology and clinical practice. Am. J. Hematol., 91(4):430–433, Jun 2016.

[27] T. Barbui, A. M. Vannucchi, V. Buxhofer-Ausch, V. De Stefano, S. Betti, A. Ram- baldi, E. Rumi, M. Ruggeri, F. Rodeghiero, M. L. Randi, I. Bertozzi, H. Gisslinger, G. Finazzi, A. Carobbio, J. Thiele, F. Passamonti, C. Falcone, and A. Tefferi. Practice-relevant revision of IPSET-thrombosis based on 1019 patients with WHO- defined essential thrombocythemia. Blood Cancer J, 5:e369, Nov 2015.

[28] G. Barosi, M. Massa, R. Campanelli, G. Fois, P. Catarsi, G. Viarengo, L. Villani, V. Poletto, T. Bosoni, U. Magrini, R. P. Gale, and V. Rosti. Primary myelofibrosis: Older age and high JAK2V617F allele burden are associated with elevated plasma high-sensitivity C-reactive protein levels and a phenotype of progressive disease. Leuk. Res., 60:18–23, 09 2017.

[29] P. Bart?nek, J. Kralova, G. Blendinger, M. Dvorak, and M. Zenke. GATA-1 and c-myb crosstalk during red blood cell differentiation through GATA-1 binding sites in the c-myb promoter. Oncogene, 22(13):1927–1935, Apr 2003.

[30] A. Bauer, W. Mikulits, G. Lagger, G. Stengl, G. Brosch, and H. Beug. The thyroid hormone receptor functions as a ligand-operated developmental switch between proliferation and differentiation of erythroid progenitors. EMBO J., 17(15):4291–4303, Aug 1998.

[31] A. J. BECKER, E. A. McCULLOCH, and J. E. TILL. Cytological demonstration of the clonal nature of spleen colonies derived from transplanted mouse marrow cells. Nature, 197:452–454, Feb 1963.

[32] P. B. Becker and J. L. Workman. Nucleosome remodeling and epigenetics. Cold Spring Harb Perspect Biol, 5(9), Sep 2013.

[33] P. A. Beer, P. J. Campbell, L. M. Scott, A. J. Bench, W. N. Erber, D. Bareford, B. S. Wilkins, J. T. Reilly, H. C. Hasselbalch, R. Bowman, K. Wheatley, G. Buck, C. N. Harrison, and A. R. Green. MPL mutations in myeloproliferative disorders: analysis of the PT-1 cohort. Blood, 112(1):141–149, Jul 2008.

[34] M. Bengtsen, L. S?rensen, L. Aabel, M. Ledsaak, V. Matre, and O. S. Gabrielsen. The adaptor protein ARA55 and the nuclear kinase HIPK1 assist c-Myb in recruiting p300 to chromatin. Biochim. Biophys. Acta, May 2017.

121 References

[35] K. Bennemann, O. Galm, S. Wilop, C. Schubert, T. H. Brummendorf, and E. Jost. Epigenetic dysregulation of secreted frizzled-related proteins in myeloproliferative neoplasms complements the JAK2V617F-mutation. Clin Epigenetics, 4(1):12, Aug 2012.

[36] E. Bernstein, A. A. Caudy, S. M. Hammond, and G. J. Hannon. Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature, 409(6818):363–366, Jan 2001.

[37] S. Biechonski, M. Yassin, and M. Milyavsky. DNA-damage response in hematopoietic stem cells: an evolutionary trade-off between blood regeneration and leukemia suppression. Carcinogenesis, Mar 2017.

[38] E. G. Bochukova, N. Huang, J. Keogh, E. Henning, C. Purmann, K. Blaszczyk, S. Saeed, J. Hamilton-Shield, J. Clayton-Smith, S. O’Rahilly, M. E. Hurles, and I. S. Farooqi. Large, rare chromosomal deletions associated with severe early-onset obesity. Nature, 463(7281):666–670, Feb 2010.

[39] C. Bogani, V. Ponziani, P. Guglielmelli, C. Desterke, V. Rosti, A. Bosi, M. C. Le Bousse-Kerdiles, G. Barosi, and A. M. Vannucchi. Hypermethylation of CXCR4 promoter in CD34+ cells from patients with primary myelofibrosis. Stem Cells, 26(8):1920–1930, Aug 2008.

[40] R. Bogeska and H. L. Pahl. Elevated nuclear factor erythroid-2 levels promote epo-independent erythroid maturation and recapitulate the hematopoietic stem cell and common myeloid progenitor expansion observed in polycythemia vera patients. Stem Cells Transl Med, 2(2):112–117, Feb 2013.

[41] I. Borges, I. Sena, P. Azevedo, J. Andreotti, V. Almeida, A. Paiva, G. Santos, D. Guerra, P. Prazeres, L. L. Mesquita, L. S. B. Silva, C. Leonel, A. Mintz, and A. Birbrair. Lung as a Niche for Hematopoietic Progenitors. Stem Cell Rev, 13(5):567–574, 10 2017.

[42] P. Bose and S. Verstovsek. The evolution and clinical relevance of prognostic classification systems in myelofibrosis. Cancer, 122(5):681–692, Mar 2016.

[43] R. L. Boudreau, I. Martins, and B. L. Davidson. Artificial microRNAs as siRNA shuttles: improved safety as compared to shRNAs in vitro and in vivo. Mol. Ther., 17(1):169–175, Jan 2009.

122 References

[44] O. Boussif, F. Lezoualc’h, M. A. Zanta, M. D. Mergny, D. Scherman, B. Demeneix, and J. P. Behr. A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proc. Natl. Acad. Sci. U.S.A., 92(16):7297–7301, Aug 1995.

[45] G. B. Bradford, B. Williams, R. Rossi, and I. Bertoncello. Quiescence, cycling, and turnover in the primitive hematopoietic stem cell compartment. Exp. Hematol., 25(5):445–453, May 1997.

[46] M. Brauchle, Z. Yao, R. Arora, S. Thigale, I. Clay, B. Inverardi, J. Fletcher, P. Taslimi, M. G. Acker, B. Gerrits, J. Voshol, A. Bauer, D. Schubeler, T. Bouwmeester, and H. Ruffner. Protein complex interactor analysis and differential activity of KDM3 subfamily members towards H3K9 methylation. PLoS ONE, 8(4):e60549, 2013.

[47] G. Brown, C. M. Bunce, and G. R. Guy. Sequential determination of lineage potentials during haemopoiesis. Br. J. Cancer, 52(5):681–686, Nov 1985.

[48] H. Bruchova, D. Yoon, A. M. Agarwal, S. Swierczek, and J. T. Prchal. Erythro- poiesis in polycythemia vera is hyper-proliferative and has accelerated maturation. Blood Cells Mol. Dis., 43(1):81–87, 2009.

[49] T. R. Brummelkamp, R. Bernards, and R. Agami. A system for stable expression of short interfering RNAs in mammalian cells. Science, 296(5567):550–553, Apr 2002.

[50] D. Bryder, D. J. Rossi, and I. L. Weissman. Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am. J. Pathol., 169(2):338–346, Aug 2006.

[51] LESLIE P. BULLOCK and EMMANUEL C. BESA. The role of the androgen receptor in erythropoiesis*. Endocrinology, 109(6):1983, 1981.

[52] K. Busch, K. Klapproth, M. Barile, M. Flossdorf, T. Holland-Letz, S. M. Schlenner, M. Reth, T. Hofer, and H. R. Rodewald. Fundamental properties of unperturbed haematopoiesis from stem cells in vivo. Nature, 518(7540):542–546, Feb 2015.

[53] L. Busque, J. P. Patel, M. E. Figueroa, A. Vasanthakumar, S. Provost, Z. Hamilou, L. Mollica, J. Li, A. Viale, A. Heguy, M. Hassimi, N. Socci, P. K. Bhatt, M. Gonen, C. E. Mason, A. Melnick, L. A. Godley, C. W. Brennan, O. Abdel-Wahab, and

123 References

R. L. Levine. Recurrent somatic TET2 mutations in normal elderly individuals with clonal hematopoiesis. Nat. Genet., 44(11):1179–1181, Nov 2012.

[54] J. W. Byron. Analysis of receptor mechanisms involved in the hemopoietic effects of androgens: use of the Tfm mutant. Exp. Hematol., 5(6):429–435, Nov 1977.

[55] L. V. Bystrykh, E. Verovskaya, E. Zwart, M. Broekhuis, and G. de Haan. Counting stem cells: methodological constraints. Nat. Methods, 9(6):567–574, May 2012.

[56] Ja Min Byun, Young Jin Kim, Taemi Youk, John Jeongseok Yang, Jongha Yoo, and Tae Sung Park. Real world epidemiology of myeloproliferative neoplasms: a population based study in korea 2004–2013. Annals of Hematology, 96(3):373–381, Mar 2017.

[57] R. Cao and Y. Zhang. The functions of E(Z)/EZH2-mediated methylation of lysine 27 in histone H3. Curr. Opin. Genet. Dev., 14(2):155–164, Apr 2004.

[58] N. Carbuccia, A. Murati, V. Trouplin, M. Brecqueville, J. Adelaide, J. Rey, W. Vainchenker, O. A. Bernard, M. Chaffanet, N. Vey, D. Birnbaum, and M.J. Mozziconacci. Mutations of ASXL1 gene in myeloproliferative neoplasms. Leukemia, 23(11):2183–2186, Nov 2009.

[59] D. A. Card, P. B. Hebbar, L. Li, K. W. Trotter, Y. Komatsu, Y. Mishina, and T. K. Archer. Oct4/Sox2-regulated miR-302 targets cyclin D1 in human embryonic stem cells. Mol. Cell. Biol., 28(20):6426–6438, Oct 2008.

[60] D. Castanotto, K. Sakurai, R. Lingeman, H. Li, L. Shively, L. Aagaard, H. Soifer, A. Gatignol, A. Riggs, and J. J. Rossi. Combinatorial delivery of small interfering RNAs reduces RNAi efficacy by selective incorporation into RISC. Nucleic Acids Res., 35(15):5154–5164, 2007.

[61] S. N. Catlin, L. Busque, R. E. Gale, P. Guttorp, and J. L. Abkowitz. The replication rate of human hematopoietic stem cells in vivo. Blood, 117(17):4460– 4466, Apr 2011.

[62] S. Cerquozzi and A. Tefferi. Blast transformation and fibrotic progression in polycythemia vera and essential thrombocythemia: a literature review of incidence and risk factors. Blood Cancer J, 5:e366, Nov 2015.

124 References

[63] F. Cervantes, B. Dupriez, A. Pereira, F. Passamonti, J. T. Reilly, E. Morra, A. M. Vannucchi, R. A. Mesa, J. L. Demory, G. Barosi, E. Rumi, and A. Tefferi. New prognostic scoring system for primary myelofibrosis based on a study of the International Working Group for Myelofibrosis Research and Treatment. Blood, 113(13):2895–2901, Mar 2009.

[64] F. Cervantes, I. M. Isola, A. Alvarez-Larran, J. C. Hernandez-Boluda, J. G. Correa, and A. Pereira. Danazol therapy for the anemia of myelofibrosis: assessment of efficacy with current criteria of response and long-term results. Ann. Hematol., 94(11):1791–1796, Nov 2015.

[65] I. Chachoua, C. Pecquet, M. El-Khoury, H. Nivarthi, R. I. Albu, C. Marty, V. Gryshkova, J. P. Defour, G. Vertenoeil, A. Ngo, A. Koay, H. Raslova, P. J. Courtoy, M. L. Choong, I. Plo, W. Vainchenker, R. Kralovics, and S. N. Con- stantinescu. Thrombopoietin receptor activation by myeloproliferative neoplasm associated calreticulin mutants. Blood, 127(10):1325–1335, Mar 2016.

[66] E. Chen, R. K. Schneider, L. J. Breyfogle, E. A. Rosen, L. Poveromo, S. Elf, A. Ko, K. Brumme, R. Levine, B. L. Ebert, and A. Mullally. Distinct effects of concomitant Jak2V617F expression and Tet2 loss in mice promote disease progression in myeloproliferative neoplasms. Blood, 125(2):327–335, Jan 2015.

[67] M. Chen, N. Zhu, X. Liu, B. Laurent, Z. Tang, R. Eng, Y. Shi, S. A. Armstrong, and R. G. Roeder. JMJD1C is required for the survival of acute myeloid leukemia by functioning as a coactivator for key transcription factors. Genes Dev., 29(20):2123– 2139, Oct 2015.

[68] Y. Chen, Y. Shan, M. Lu, N. DeSouza, Z. Guo, R. Hoffman, A. Liang, and S. Li. Alox5 Blockade Eradicates JAK2V617F-Induced Polycythemia Vera in Mice. Cancer Res., 77(1):164–174, 01 2017.

[69] S. H. Cheshier, S. J. Morrison, X. Liao, and I. L. Weissman. In vivo proliferation and cell cycle kinetics of long-term self-renewing hematopoietic stem cells. Proc. Natl. Acad. Sci. U.S.A., 96(6):3120–3125, Mar 1999.

[70] S. Chotirat, W. Thongnoppakhun, W. Wanachiwanawin, and C. U. Auewarakul. Acquired somatic mutations of isocitrate dehydrogenases 1 and 2 (IDH1 and IDH2) in preleukemic disorders. Blood Cells Mol. Dis., 54(3):286–291, Mar 2015.

125 References

[71] R. Chowdhury, K. K. Yeoh, Y. M. Tian, L. Hillringhaus, E. A. Bagg, N. R. Rose, I. K. Leung, X. S. Li, E. C. Woon, M. Yang, M. A. McDonough, O. N. King, I. J. Clifton, R. J. Klose, T. D. Claridge, P. J. Ratcliffe, C. J. Schofield, and A. Kawamura. The oncometabolite 2-hydroxyglutarate inhibits histone lysine demethylases. EMBO Rep., 12(5):463–469, May 2011.

[72] J. L. Christensen and I. L. Weissman. Flk-2 is a marker in hematopoietic stem cell differentiation: a simple method to isolate long-term stem cells. Proc. Natl. Acad. Sci. U.S.A., 98(25):14541–14546, Dec 2001.

[73] C. K. Chuen, K. Li, M. Yang, T. F. Fok, C. K. Li, C. M. Chui, and P. M. Yuen. Interleukin-1beta up-regulates the expression of thrombopoietin and transcription factors c-Jun, c-Fos, GATA-1, and NF-E2 in megakaryocytic cells. J. Lab. Clin. Med., 143(2):75–88, Feb 2004.

[74] K. H. Chung, C. C. Hart, S. Al-Bassam, A. Avery, J. Taylor, P. D. Patel, A. B. Vojtek, and D. L. Turner. Polycistronic RNA polymerase II expression vectors for RNA interference based on BIC/miR-155. Nucleic Acids Res., 34(7):e53, Apr 2006.

[75] M. Claustres and C. Sultan. Androgen and erythropoiesis: evidence for an androgen receptor in erythroblasts from human bone marrow cultures. Horm. Res., 29(1):17–22, 1988.

[76] Thomas Coar et al. The aphorisms of Hippocrates: with a translation into latin, and english. Classics of Medicine Library, 1982.

[77] Y. Costa, J. Ding, T. W. Theunissen, F. Faiola, T. A. Hore, P. V. Shliaha, M. Fidalgo, A. Saunders, M. Lawrence, S. Dietmann, S. Das, D. N. Levasseur, Z. Li, M. Xu, W. Reik, J. C. Silva, and J. Wang. NANOG-dependent function of TET1 and TET2 in establishment of pluripotency. Nature, 495(7441):370–374, Mar 2013.

[78] S. Cui, K. E. Kolodziej, N. Obara, A. Amaral-Psarris, J. Demmers, L. Shi, J. D. Engel, F. Grosveld, J. Strouboulis, and O. Tanabe. Nuclear receptors TR2 and TR4 recruit multiple epigenetic transcriptional corepressors that associate specifically with the embryonic β-type globin promoters in differentiated adult erythroid cells. Mol. Cell. Biol., 31(16):3298–3311, Aug 2011.

[79] S. Cui, O. Tanabe, K. C. Lim, H. E. Xu, X. E. Zhou, J. D. Lin, L. Shi, L. Schmidt, A. Campbell, R. Shimizu, M. Yamamoto, and J. D. Engel. PGC-1 coactivator

126 References

activity is required for murine erythropoiesis. Mol. Cell. Biol., 34(11):1956–1965, Jun 2014.

[80] S. Cui, O. Tanabe, M. Sierant, L. Shi, A. Campbell, K. C. Lim, and J. D. Engel. Compound loss of function of nuclear receptors Tr2 and Tr4 leads to induction of murine embryonic β-type globin genes. Blood, 125(9):1477–1487, Feb 2015.

[81] J. C. Culhane and P. A. Cole. LSD1 and the chemistry of histone demethylation. Curr Opin Chem Biol, 11(5):561–568, Oct 2007.

[82] G. Q. Daley, R. A. Van Etten, and D. Baltimore. Induction of chronic myelogenous leukemia in mice by the P210bcr/abl gene of the Philadelphia chromosome. Science, 247(4944):824–830, Feb 1990.

[83] W. DAMESHEK. Some speculations on the myeloproliferative syndromes. Blood, 6(4):372–375, Apr 1951.

[84] L. Dang, D. W. White, S. Gross, B. D. Bennett, M. A. Bittinger, E. M. Driggers, V. R. Fantin, H. G. Jang, S. Jin, M. C. Keenan, K. M. Marks, R. M. Prins, P. S. Ward, K. E. Yen, L. M. Liau, J. D. Rabinowitz, L. C. Cantley, C. B. Thompson, M. G. Vander Heiden, and S. M. Su. Cancer-associated IDH1 mutations produce 2-hydroxyglutarate. Nature, 462(7274):739–744, Dec 2009.

[85] K. C. Das, M. Mukherjee, T. K. Sarkar, R. J. Dash, and G. K. Rastogi. Ery- thropoiesis and erythropoietin in hypo- and hyperthyroidism. J. Clin. Endocrinol. Metab., 40(2):211–220, Feb 1975.

[86] F. L. Datz and A. Taylor. The clinical use of radionuclide bone marrow imaging. Semin Nucl Med, 15(3):239–259, Jul 1985.

[87] M. A. Dawson, A. J. Bannister, B. Gottgens, S. D. Foster, T. Bartke, A. R. Green, and T. Kouzarides. JAK2 phosphorylates histone H3Y41 and excludes HP1alpha from chromatin. Nature, 461(7265):819–822, Oct 2009.

[88] J. P. Defour, M. Itaya, V. Gryshkova, I. C. Brett, C. Pecquet, T. Sato, S. O. Smith, and S. N. Constantinescu. Tryptophan at the transmembrane-cytosolic junction modulates thrombopoietin receptor dimerization and activation. Proc. Natl. Acad. Sci. U.S.A., 110(7):2540–2545, Feb 2013.

127 References

[89] J. Ding, H. Komatsu, A. Wakita, M. Kato-Uranishi, M. Ito, A. Satoh, K. Tsuboi, M. Nitta, H. Miyazaki, S. Iida, and R. Ueda. Familial essential thrombocythemia associated with a dominant-positive activating mutation of the c-MPL gene, which encodes for the receptor for thrombopoietin. Blood, 103(11):4198–4200, Jun 2004.

[90] S. Doulatov, F. Notta, E. Laurenti, and J. E. Dick. Hematopoiesis: a human perspective. Cell Stem Cell, 10(2):120–136, Feb 2012.

[91] M. O. Duff, S. Olson, X. Wei, S. C. Garrett, A. Osman, M. Bolisetty, A.Plocik, S. E. Celniker, and B. R. Graveley. Genome-wide identification of zero nucleotide recursive splicing in Drosophila. Nature, 521(7552):376–379, May 2015.

[92] A. Dunbar, A. Nazir, and R. Levine. Overview of Transgenic Mouse Models of Myeloproliferative Neoplasms (MPNs). Curr Protoc Pharmacol, 77:1–14, Jun 2017.

[93] J. M. Dunwell, A. Culham, C. E. Carter, C. R. Sosa-Aguirre, and P. W. Goode- nough. Evolution of functional diversity in the cupin superfamily. Trends Biochem. Sci., 26(12):740–746, Dec 2001.

[94] R. Dutta, B. Tiu, and K. M. Sakamoto. CBP/p300 acetyltransferase activity in hematologic malignancies. Mol. Genet. Metab., 119(1-2):37–43, Sep 2016.

[95] B. Dykstra, D. Kent, M. Bowie, L. McCaffrey, M. Hamilton, K. Lyons, S. J.Lee, R. Brinkman, and C. Eaves. Long-term propagation of distinct hematopoietic differentiation programs in vivo. Cell Stem Cell, 1(2):218–229, Aug 2007.

[96] C. J. Eaves. Hematopoietic stem cells: concepts, definitions, and the new reality. Blood, 125(17):2605–2613, Apr 2015.

[97] E. G. T. Ederveen, F. P. A. M. van Hunsel, M. J. Wondergem, and E. P. van Puijenbroek. Severe Secondary Polycythemia in a Female-to-Male Transgender Patient While Using Lifelong Hormonal Therapy: A Patient’s Perspective. Drug Saf Case Rep, 5(1):6, Feb 2018.

[98] M. Ekstrom and T. Ringbaek. Which patients with moderate hypoxemia benefit from long-term oxygen therapy? Ways forward. Int J Chron Obstruct Pulmon Dis, 13:231–235, 2018.

[99] e. l. H. A. El-Harith, C. Roesl, M. Ballmaier, M. Germeshausen, H. Frye-Boukhriss, N. von Neuhoff, C. Becker, G. Nurnberg, P. Nurnberg, M. A. Ahmed, J. Hubener, J. Schmidtke, K. Welte, and M. Stuhrmann. Familial thrombocytosis caused by

128 References

the novel germ-line mutation p.Pro106Leu in the MPL gene. Br. J. Haematol., 144(2):185–194, Jan 2009.

[100] S. M. Elbashir, J. Harborth, W. Lendeckel, A. Yalcin, K. Weber, and T. Tuschl. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature, 411(6836):494–498, May 2001.

[101] S. Elf, N. S. Abdelfattah, E. Chen, J. Perales-Paton, E. A. Rosen, A. Ko, F. Peisker, N. Florescu, S. Giannini, O. Wolach, E. A. Morgan, Z. Tothova, J. A. Losman, R. K. Schneider, F. Al-Shahrour, and A. Mullally. Mutant Calreticulin Requires Both Its Mutant C-terminus and the Thrombopoietin Receptor for Oncogenic Transformation. Cancer Discov, 6(4):368–381, Apr 2016.

[102] T. Ernst, A. J. Chase, J. Score, C. E. Hidalgo-Curtis, C. Bryant, A. V. Jones, K. Waghorn, K. Zoi, F. M. Ross, A. Reiter, A. Hochhaus, H. G. Drexler, A. Dun- combe, F. Cervantes, D. Oscier, J. Boultwood, F. H. Grand, and N. C. Cross. Inactivating mutations of the histone methyltransferase gene EZH2 in myeloid disorders. Nat. Genet., 42(8):722–726, Aug 2010.

[103] R. Espin-Palazon, B. Weijts, V. Mulero, and D. Traver. Proinflammatory Signals as Fuel for the Fire of Hematopoietic Stem Cell Emergence. Trends Cell Biol., 28(1):58–66, Jan 2018.

[104] B. L. Esplin, T. Shimazu, R. S. Welner, K. P. Garrett, L. Nie, Q. Zhang, M. B. Humphrey, Q. Yang, L. A. Borghesi, and P. W. Kincade. Chronic exposure to a TLR ligand injures hematopoietic stem cells. J. Immunol., 186(9):5367–5375, May 2011.

[105] L. Fagerberg, B. M. Hallstrom, P. Oksvold, C. Kampf, D. Djureinovic, J. Odeberg, M. Habuka, S. Tahmasebpoor, A. Danielsson, K. Edlund, A. Asplund, E. Sjostedt, E. Lundberg, C. A. Szigyarto, M. Skogs, J. O. Takanen, H. Berling, H. Tegel, J. Mulder, P. Nilsson, J. M. Schwenk, C. Lindskog, F. Danielsson, A. Mardinoglu, A. Sivertsson, K. von Feilitzen, M. Forsberg, M. Zwahlen, I. Olsson, S. Navani, M. Huss, J. Nielsen, F. Ponten, and M. Uhlen. Analysis of the human tissue-specific expression by genome-wide integration of transcriptomics and antibody-based proteomics. Mol. Cell Proteomics, 13(2):397–406, Feb 2014.

[106] M. Fan, J. Rhee, J. St-Pierre, C. Handschin, P. Puigserver, J. Lin, S. Jaeger, H. Erdjument-Bromage, P. Tempst, and B. M. Spiegelman. Suppression of

129 References

mitochondrial respiration through recruitment of p160 myb binding protein to PGC-1alpha: modulation by p38 MAPK. Genes Dev., 18(3):278–289, Feb 2004.

[107] N. Fan, Y. Tang, Z. Wu, M. Guan, B. Chen, X. Xu, W. Ma, X. Xu, and X. Zhang. Elevated expression of the EZH2 gene in CALR-mutated patients with primary myelofibrosis. Ann. Hematol., 97(7):1193–1208, Jul 2018.

[108] C. Fellmann and S. W. Lowe. Stable RNA interference rules for silencing. Nat. Cell Biol., 16(1):10–18, Jan 2014.

[109] M. E. Figueroa, O. Abdel-Wahab, C. Lu, P. S. Ward, J. Patel, A. Shih, Y. Li, N. Bhagwat, A. Vasanthakumar, H. F. Fernandez, M. S. Tallman, Z. Sun, K. Wolniak, J. K. Peeters, W. Liu, S. E. Choe, V. R. Fantin, E. Paietta, B. Lowenberg, J. D. Licht, L. A. Godley, R. Delwel, P. J. Valk, C. B. Thompson, R. L. Levine, and A. Melnick. Leukemic IDH1 and IDH2 mutations result in a hypermethylation phenotype, disrupt TET2 function, and impair hematopoietic differentiation. Cancer Cell, 18(6):553–567, Dec 2010.

[110] G. Finazzi, A. Carobbio, P. Guglielmelli, C. Cavalloni, S. Salmoiraghi, A. M. Vannucchi, M. Cazzola, F. Passamonti, A. Rambaldi, and T. Barbui. Calreticulin mutation does not modify the IPSET score for predicting the risk of thrombosis among 1150 patients with essential thrombocythemia. Blood, 124(16):2611–2612, Oct 2014.

[111] M. C. Finazzi, A. Carobbio, F. Cervantes, I. M. Isola, A. M. Vannucchi, P. Guglielmelli, A. Rambaldi, G. Finazzi, G. Barosi, and T. Barbui. CALR mutation, MPL mutation and triple negativity identify patients with the lowest vascular risk in primary myelofibrosis. Leukemia, 29(5):1209–1210, May 2015.

[112] D. Finkelshtein, A. Werman, D. Novick, S. Barak, and M. Rubinstein. LDL receptor and its family members serve as the cellular receptors for vesicular stomatitis virus. Proc. Natl. Acad. Sci. U.S.A., 110(18):7306–7311, Apr 2013.

[113] C. L. Fisher, N. Pineault, C. Brookes, C. D. Helgason, H. Ohta, C. Bodner, J. L. Hess, R. K. Humphries, and H. W. Brock. Loss-of-function Additional sex combs like 1 mutations disrupt hematopoiesis but do not cause severe myelodysplasia or leukemia. Blood, 115(1):38–46, Jan 2010.

[114] C. L. Fisher, N. Pineault, C. Brookes, C. D. Helgason, H. Ohta, C. Bodner, J. L. Hess, R. K. Humphries, and H. W. Brock. Loss-of-function Additional sex combs

130 References

like 1 mutations disrupt hematopoiesis but do not cause severe myelodysplasia or leukemia. Blood, 115(1):38–46, Jan 2010.

[115] A. G. Fleischman, K. J. Aichberger, S. B. Luty, T. G. Bumm, C. L. Petersen, S. Doratotaj, K. B. Vasudevan, D. H. LaTocha, F. Yang, R. D. Press, M. M. Loriaux, H. L. Pahl, R. T. Silver, A. Agarwal, T. O’Hare, B. J. Druker, G. C. Bagby, and M. W. Deininger. TNFα facilitates clonal expansion of JAK2V617F positive cells in myeloproliferative neoplasms. Blood, 118(24):6392–6398, Dec 2011.

[116] R. C. Friedman, K. K. Farh, C. B. Burge, and D. P. Bartel. Most mammalian mRNAs are conserved targets of microRNAs. Genome Res., 19(1):92–105, Jan 2009.

[117] L. V. Furtado, H. C. Weigelin, K. S. Elenitoba-Johnson, and B. L. Betz. Detection of MPL mutations by a novel allele-specific PCR-based strategy. J Mol Diagn, 15(6):810–818, Nov 2013.

[118] J. J. Fuster, S. MacLauchlan, M. A. Zuriaga, M. N. Polackal, A. C. Ostriker, R. Chakraborty, C. L. Wu, S. Sano, S. Muralidharan, C. Rius, J. Vuong, S. Jacob, V. Muralidhar, A. A. Robertson, M. A. Cooper, V. Andres, K. K. Hirschi, K. A. Martin, and K. Walsh. Clonal hematopoiesis associated with TET2 deficiency accelerates atherosclerosis development in mice. Science, 355(6327):842–847, Feb 2017.

[119] Deutsche Gesellschaft für Hämatologie und medizinische Onkolo- gie. Onkopedia: Essentielle (oder primäre) thrombozythämie (et). https://www.onkopedia.com/de/onkopedia/guidelines/essentielle-oder-primaere- thrombozythaemie-et/@@view/html/index.html, 2014. last accessed: Sep 29, 2018.

[120] Deutsche Gesellschaft für Hämatologie und medizinische Onkologie. Onkopedia: Polycythaemia vera (pv. https://www.onkopedia.com/de/onkopedia/guidelines/polycythaemia-vera- pv/@@view/html/index.html, 2018. last accessed: Sep 29, 2018.

[121] N. Gangat, D. Caramazza, R. Vaidya, G. George, K. Begna, S. Schwager, D. Van Dyke, C. Hanson, W. Wu, A. Pardanani, F. Cervantes, F. Passamonti, and A. Tefferi. DIPSS plus: a refined Dynamic International Prognostic Scoring System

131 References

for primary myelofibrosis that incorporates prognostic information from karyotype, platelet count, and transfusion status. J. Clin. Oncol., 29(4):392–397, Feb 2011.

[122] N. Gangat, J. Strand, C. Y. Li, W. Wu, A. Pardanani, and A. Tefferi. Leucocytosis in polycythaemia vera predicts both inferior survival and leukaemic transformation. Br. J. Haematol., 138(3):354–358, Aug 2007.

[123] M. R. Garbati, C. A. Welgan, S. H. Landefeld, L. F. Newell, A. Agarwal, J. B. Dunlap, T. K. Chourasia, H. Lee, J. Elferich, E. Traer, R. Rattray, M. J. Cascio, R. D. Press, G. C. Bagby, J. W. Tyner, B. J. Druker, and K. H. Dao. Mutant calreticulin-expressing cells induce monocyte hyperreactivity through a paracrine mechanism. Am. J. Hematol., 91(2):211–219, Feb 2016.

[124] A. Geduk, E. B. Atesoglu, P. Tarkun, O. Mehtap, A. Hacihanefioglu, E. T. Demirsoy, and C. Baydemir. The Role of β-Catenin in Bcr/Abl Negative Myelopro- liferative Neoplasms: An Immunohistochemical Study. Clin Lymphoma Myeloma Leuk, 15(12):785–789, Dec 2015.

[125] H. L. Geyer, H. Kosiorek, A. C. Dueck, R. Scherber, S. Slot, S. Zweegman, P. A. Te Boekhorst, Z. Senyak, H. C. Schouten, F. Sackmann, A. K. Fuentes, D. Hernandez-Maraver, H. L. Pahl, M. Griesshammer, F. Stegelmann, K. Dohner, T. Lehmann, K. Bonatz, A. Reiter, F. Boyer, G. Etienne, J. C. Ianotto, D. Ranta, L. Roy, J. Y. Cahn, C. N. Harrison, D. Radia, P. Muxi, N. Maldonado, C. Besses, F. Cervantes, P. L. Johansson, T. Barbui, G. Barosi, A. M. Vannucchi, C. Paoli, F. Passamonti, B. Andreasson, M. L. Ferrari, A. Rambaldi, J. Samuelsson, K. Can- non, G. Birgegard, Z. Xiao, Z. Xu, Y. Zhang, X. Sun, J. Xu, J. J. Kiladjian, P. Zhang, R. P. Gale, and R. A. Mesa. Associations between gender, disease features and symptom burden in patients with myeloproliferative neoplasms: an analysis by the MPN QOL International Working Group. Haematologica, 102(1):85– 93, 01 2017.

[126] B. Giebel, T. Zhang, J. Beckmann, J. Spanholtz, P. Wernet, A. D. Ho, and M. Punzel. Primitive human hematopoietic cells give rise to differentially specified daughter cells upon their initial cell division. Blood, 107(5):2146–2152, Mar 2006.

[127] C. Gieger, A. Radhakrishnan, A. Cvejic, W. Tang, E. Porcu, G. Pistis, J. Serbanovic- Canic, U. Elling, A. H. Goodall, Y. Labrune, L. M. Lopez, R. Magi, S. Meacham, Y. Okada, N. Pirastu, R. Sorice, A. Teumer, K. Voss, W. Zhang, R. Ramirez- Solis, J. C. Bis, D. Ellinghaus, M. Gogele, J. J. Hottenga, C. Langenberg,

132 References

P. Kovacs, P. F. O’Reilly, S. Y. Shin, T. Esko, J. Hartiala, S. Kanoni, F. Murgia, A. Parsa, J. Stephens, P. van der Harst, C. Ellen van der Schoot, H. Allayee, A. Attwood, B. Balkau, F. Bastardot, S. Basu, S. E. Baumeister, G. Biino, L. Bomba, A. Bonnefond, F. Cambien, J. C. Chambers, F. Cucca, P. D’Adamo, G. Davies, R. A. de Boer, E. J. de Geus, A. Doring, P. Elliott, J. Erdmann, D. M. Evans, M. Falchi, W. Feng, A. R. Folsom, I. H. Frazer, Q. D. Gibson, N. L. Glazer, C. Hammond, A. L. Hartikainen, S. R. Heckbert, C. Hengstenberg, M. Hersch, T. Illig, R. J. Loos, J. Jolley, K. T. Khaw, B. Kuhnel, M. C. Kyrtsonis, V. Lagou, H. Lloyd-Jones, T. Lumley, M. Mangino, A. Maschio, I. Mateo Leach, B. McKnight, Y. Memari, B. D. Mitchell, G. W. Montgomery, Y. Nakamura, M. Nauck, G. Navis, U. Nothlings, I. M. Nolte, D. J. Porteous, A. Pouta, P. P. Pramstaller, J. Pullat, S. M. Ring, J. I. Rotter, D. Ruggiero, A. Ruokonen, C. Sala, N. J. Samani, J. Sambrook, D. Schlessinger, S. Schreiber, H. Schunkert, J. Scott, N. L. Smith, H. Snieder, J. M. Starr, M. Stumvoll, A. Takahashi, W. H. Tang, K. Taylor, A. Tenesa, S. Lay Thein, A. Tonjes, M. Uda, S. Ulivi, D. J. van Veldhuisen, P. M. Visscher, U. Volker, H. E. Wichmann, K. L. Wiggins, G. Willemsen, T. P. Yang, J. Hua Zhao, P. Zitting, J. R. Bradley, G. V. Dedoussis, P. Gasparini, S. L. Hazen, A. Metspalu, M. Pirastu, A. R. Shuldiner, L. Joost van Pelt, J. J. Zwaginga, D. I. Boomsma, I. J. Deary, A. Franke, P. Froguel, S. K. Ganesh, M. R. Jarvelin, N. G. Martin, C. Meisinger, B. M. Psaty, T. D. Spector, N. J. Wareham, J. W. Akkerman, M. Ciullo, P. Deloukas, A. Greinacher, S. Jupe, N. Kamatani, J. Khadake, J. S. Kooner, J. Penninger, I. Prokopenko, D. Stemple, D. Toniolo, L. Wernisch, S. Sanna, A. A. Hicks, A. Rendon, M. A. Ferreira, W. H. Ouwehand, and N. Soranzo. New gene functions in megakaryopoiesis and platelet formation. Nature, 480(7376):201–208, Nov 2011.

[128] H. Gill, A. Y. Leung, C. C. Chan, J. S. Lau, C. Chan, S. F. Yip, H. Liu, B. Kho, V. Mak, H. K. Lee, S. Y. Lin, C. K. Lau, and Y. L. Kwong. Clinicopathologic features and prognostic indicators in Chinese patients with myelofibrosis. Hematology, 21(1):10–18, Jan 2016.

[129] A. J. Giraldez, Y. Mishima, J. Rihel, R. J. Grocock, S. Van Dongen, K. Inoue, A. J. Enright, and A. F. Schier. Zebrafish MiR-430 promotes deadenylation and clearance of maternal mRNAs. Science, 312(5770):75–79, Apr 2006.

[130] P. S. Goerttler, C. Kreutz, J. Donauer, D. Faller, T. Maiwald, E. Marz, B. Rum- berger, T. Sparna, A. Schmitt-Graff, J. Wilpert, J. Timmer, G. Walz, andH.L.

133 References

Pahl. Gene expression profiling in polycythaemia vera: overexpression of transcription factor NF-E2. Br. J. Haematol., 129(1):138–150, Apr 2005.

[131] A. Gorgens, A. K. Ludwig, M. Mollmann, A. Krawczyk, J. Durig, H. Hanenberg, P. A. Horn, and B. Giebel. Multipotent hematopoietic progenitors divide asym- metrically to create progenitors of the lymphomyeloid and erythromyeloid lineages. Stem Cell Reports, 3(6):1058–1072, Dec 2014.

[132] F. L. Graham, J. Smiley, W. C. Russell, and R. Nairn. Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J. Gen. Virol., 36(1):59–74, Jul 1977.

[133] S. Granick and A. Kappas. Steroid induction of porphyrin synthesis in liver cell culture. I. Structural basis and possible physiological role in the control of heme formation. J. Biol. Chem., 242(20):4587–4593, Oct 1967.

[134] K. T. Greig, S. Carotta, and S. L. Nutt. Critical roles for c-Myb in hematopoietic progenitor cells. Semin. Immunol., 20(4):247–256, Aug 2008.

[135] M. Griesshammer, G. Saydam, F. Palandri, G. Benevolo, M. Egyed, J. Callum, T. Devos, S. Sivgin, P. Guglielmelli, C. Bensasson, M. Khan, J. P. Ronco, and F. Passamonti. Ruxolitinib for the treatment of inadequately controlled polycythemia vera without splenomegaly: 80-week follow-up from the RESPONSE-2 trial. Ann. Hematol., 97(9):1591–1600, Sep 2018.

[136] D. S. Griffiths, J. Li, M. A. Dawson, M. W. Trotter, Y. H. Cheng, A. M.Smith, W. Mansfield, P. Liu, T. Kouzarides, J. Nichols, A. J. Bannister, A. R. Green,and B. Gottgens. LIF-independent JAK signalling to chromatin in embryonic stem cells uncovered from an adult stem cell disease. Nat. Cell Biol., 13(1):13–21, Jan 2011.

[137] D. Grimm, K. L. Streetz, C. L. Jopling, T. A. Storm, K. Pandey, C. R. Davis, P. Marion, F. Salazar, and M. A. Kay. Fatality in mice due to oversaturation of cellular microRNA/short hairpin RNA pathways. Nature, 441(7092):537–541, May 2006.

[138] P. Guglielmelli, T. L. Lasho, G. Rotunno, M. Mudireddy, C. Mannarelli, M. Ni- colosi, A. Pacilli, A. Pardanani, E. Rumi, V. Rosti, C. A. Hanson, F. Mannelli, R. P. Ketterling, N. Gangat, A. Rambaldi, F. Passamonti, G. Barosi, T. Barbui,

134 References

M. Cazzola, A. M. Vannucchi, and A. Tefferi. MIPSS70: Mutation-Enhanced Inter- national Prognostic Score System for Transplantation-Age Patients With Primary Myelofibrosis. J. Clin. Oncol., 36(4):310–318, Feb 2018.

[139] P. Guglielmelli, A. Pancrazzi, G. Bergamaschi, V. Rosti, L. Villani, E. Antonioli, A. Bosi, G. Barosi, and A. M. Vannucchi. Anaemia characterises patients with myelofibrosis harbouring Mpl mutation. Br. J. Haematol., 137(3):244–247, May 2007.

[140] N. Gulati, W. Beguelin, and L. Giulino-Roth. Enhancer of zeste homolog 2 (EZH2) inhibitors. Leuk. Lymphoma, 59(7):1574–1585, Jul 2018.

[141] P. Hahn, J. Bose, S. Edler, and A. Lengeling. Genomic structure and expression of Jmjd6 and evolutionary analysis in the context of related JmjC domain containing proteins. BMC Genomics, 9:293, Jun 2008.

[142] M. Haider, N. Gangat, T. Lasho, A. K. Abou Hussein, Y. C. Elala, C. Hanson, and A. Tefferi. Validation of the revised International Prognostic Score of Thrombosis for Essential Thrombocythemia (IPSET-thrombosis) in 585 Mayo Clinic patients. Am. J. Hematol., 91(4):390–394, Jun 2016.

[143] L. Han, C. Schubert, J. Kohler, M. Schemionek, S. Isfort, T. H. Brummendorf, S. Koschmieder, and N. Chatain. Calreticulin-mutant proteins induce megakaryocytic signaling to transform hematopoietic cells and undergo accelerated degrada- tion and Golgi-mediated secretion. J Hematol Oncol, 9(1):45, 05 2016.

[144] D. Hanahan and R. A. Weinberg. Hallmarks of cancer: the next generation. Cell, 144(5):646–674, Mar 2011.

[145] C. N. Harrison, P. J. Campbell, G. Buck, K. Wheatley, C. L. East, D. Bareford, B. S. Wilkins, J. D. van der Walt, J. T. Reilly, A. P. Grigg, P. Revell, B. E. Woodcock, and A. R. Green. Hydroxyurea compared with anagrelide in high-risk essential thrombocythemia. N. Engl. J. Med., 353(1):33–45, Jul 2005.

[146] C. N. Harrison, S. Koschmieder, L. Foltz, P. Guglielmelli, T. Flindt, M. Koehler, J. Mathias, N. Komatsu, R. N. Boothroyd, A. Spierer, J. Perez Ronco, G. Taylor- Stokes, J. Waller, and R. A. Mesa. The impact of myeloproliferative neoplasms (MPNs) on patient quality of life and productivity: results from the international MPN Landmark survey. Ann. Hematol., 96(10):1653–1665, Oct 2017.

135 References

[147] C. N. Harrison, A. M. Vannucchi, J. J. Kiladjian, H. K. Al-Ali, H. Gisslinger, L. Knoops, F. Cervantes, M. M. Jones, K. Sun, M. McQuitty, V. Stalbovskaya, P. Gopalakrishna, and T. Barbui. Long-term findings from COMFORT-II, a phase 3 study of ruxolitinib vs best available therapy for myelofibrosis. Leukemia, 31(3):775, Mar 2017.

[148] A. S. Harutyunyan, R. Giambruno, C. Krendl, A. Stukalov, T. Klampfl, T. Berg, D. Chen, J. D. Milosevic Feenstra, R. Jager, B. Gisslinger, H. Gisslinger, E. Rumi, F. Passamonti, D. Pietra, A. C. Muller, K. Parapatics, F. P. Breitwieser, R. Her- rmann, J. Colinge, K. L. Bennett, G. Superti-Furga, M. Cazzola, E. Hammond, and R. Kralovics. Germline RBBP6 mutations in familial myeloproliferative neoplasms. Blood, 127(3):362–365, Jan 2016.

[149] S. Hasan, C. Lacout, C. Marty, M. Cuingnet, E. Solary, W. Vainchenker, and J. L. Villeval. JAK2V617F expression in mice amplifies early hematopoietic cells and gives them a competitive advantage that is hampered by IFNα. Blood, 122(8):1464–1477, Aug 2013.

[150] M. Heuser, F. Thol, and A. Ganser. Clonal Hematopoiesis of Indeterminate Potential. Dtsch Arztebl Int, 113(18):317–322, May 2016.

[151] D. A. Hinds, K. E. Barnholt, R. A. Mesa, A. K. Kiefer, C. B. Do, N. Eriksson, J. L. Mountain, U. Francke, J. Y. Tung, H. M. Nguyen, H. Zhang, L. Gojenola, J. L. Zehnder, and J. Gotlib. Germ line variants predispose to both JAK2 V617F clonal hematopoiesis and myeloproliferative neoplasms. Blood, 128(8):1121–1128, 08 2016.

[152] D. A. Hinds, K. E. Barnholt, R. A. Mesa, A. K. Kiefer, C. B. Do, N. Eriksson, J. L. Mountain, U. Francke, J. Y. Tung, H. M. Nguyen, H. Zhang, L. Gojenola, J. L. Zehnder, and J. Gotlib. Germ line variants predispose to both JAK2 V617F clonal hematopoiesis and myeloproliferative neoplasms. Blood, 128(8):1121–1128, 08 2016.

[153] T. Hirose, R. Apfel, M. Pfahl, and A. M. Jetten. The orphan receptor TAK1 acts as a repressor of RAR-, RXR- and T3R-mediated signaling pathways. Biochem. Biophys. Res. Commun., 211(1):83–91, Jun 1995.

[154] A. D. Ho. Kinetics and symmetry of divisions of hematopoietic stem cells. Exp. Hematol., 33(1):1–8, Jan 2005.

136 References

[155] G. S. Hodgson and T. R. Bradley. In vivo kinetic status of hematopoietic stem and progenitor cells as inferred from labeling with bromodeoxyuridine. Exp. Hematol., 12(9):683–687, Oct 1984.

[156] L. Horton, R. J. Coburn, J. M. England, and R. L. Himsworth. The haematology of hypothyroidism. Q. J. Med., 45(177):101–123, Jan 1976.

[157] J. How, A. Zhou, and S. T. Oh. Splanchnic vein thrombosis in myeloproliferative neoplasms: pathophysiology and molecular mechanisms of disease. Ther Adv Hematol, 8(3):107–118, Mar 2017.

[158] H. Huang, B. R. Sabari, B. A. Garcia, C. D. Allis, and Y. Zhao. SnapShot: histone modifications. Cell, 159(2):458–458, Oct 2014.

[159] X. Huang, M. R. Lee, S. Cooper, G. Hangoc, K. S. Hong, H. M. Chung, and H. E. Broxmeyer. Activation of OCT4 enhances ex vivo expansion of human cord blood hematopoietic stem and progenitor cells by regulating HOXB4 expression. Leukemia, 30(1):144–153, Jan 2016.

[160] M. Hultcrantz, S. Y. Kristinsson, T. M. Andersson, O. Landgren, S. Eloranta, A. R. Derolf, P. W. Dickman, and M. Bjorkholm. Patterns of survival among patients with myeloproliferative neoplasms diagnosed in Sweden from 1973 to 2008: a population-based study. J. Clin. Oncol., 30(24):2995–3001, Aug 2012.

[161] M. Ichii, M. B. Frank, R. V. Iozzo, and P. W. Kincade. The canonical Wnt pathway shapes niches supportive of hematopoietic stem/progenitor cells. Blood, 119(7):1683–1692, Feb 2012.

[162] C. N. Inra, B. O. Zhou, M. Acar, M. M. Murphy, J. Richardson, Z. Zhao, and S. J. Morrison. A perisinusoidal niche for extramedullary haematopoiesis in the spleen. Nature, 527(7579):466–471, Nov 2015.

[163] J. J. Ipsaro and L. Joshua-Tor. From guide to target: molecular insights into eukaryotic RNA-interference machinery. Nat. Struct. Mol. Biol., 22(1):20–28, Jan 2015.

[164] T. Ishii, J. Wang, W. Zhang, J. Mascarenhas, R. Hoffman, Y. Dai, N. Wisch, and M. Xu. Pivotal role of mast cells in pruritogenesis in patients with myeloproliferative disorders. Blood, 113(23):5942–5950, Jun 2009.

137 References

[165] H. Iwasaki and K. Akashi. Myeloid lineage commitment from the hematopoietic stem cell. Immunity, 26(6):726–740, Jun 2007.

[166] J. Jelinek, J. Li, Z. Mnjoyan, J. P. Issa, J. T. Prchal, and V. Afshar-Kharghan. Epigenetic control of PRV-1 expression on neutrophils. Exp. Hematol., 35(11):1677– 1683, Nov 2007.

[167] B. Jin, Y. Li, and K. D. Robertson. DNA methylation: superior or subordinate in the epigenetic hierarchy? Genes Cancer, 2(6):607–617, Jun 2011.

[168] P. Johansson, J. Kutti, B. Andreasson, S. Safai-Kutti, L. Vilen, H. Wedel, and B. Ridell. Trends in the incidence of chronic Philadelphia chromosome negative (Ph-) myeloproliferative disorders in the city of Göteborg, Sweden, during 1983-99. J. Intern. Med., 256(2):161–165, Aug 2004.

[169] K. D. Johnson, H. M. Christensen, B. Zhao, and E. H. Bresnick. Distinct mechanisms control RNA polymerase II recruitment to a tissue-specific locus control region and a downstream promoter. Mol. Cell, 8(2):465–471, Aug 2001.

[170] A. V. Jones, A. Chase, R. T. Silver, D. Oscier, K. Zoi, Y. L. Wang, H. Cario, H. L. Pahl, A. Collins, A. Reiter, F. Grand, and N. C. Cross. JAK2 haplotype is a major risk factor for the development of myeloproliferative neoplasms. Nat. Genet., 41(4):446–449, Apr 2009.

[171] P. A. Jones and P. W. Laird. Cancer epigenetics comes of age. Nat. Genet., 21(2):163–167, Feb 1999.

[172] E. Jost, J. Schmid, S. Wilop, C. Schubert, H. Suzuki, J. G. Herman, R. Osieka, and O. Galm. Epigenetic inactivation of secreted Frizzled-related proteins in acute myeloid leukaemia. Br. J. Haematol., 142(5):745–753, Sep 2008.

[173] J. S. Jutzi, R. Bogeska, G. Nikoloski, C. A. Schmid, T. S. Seeger, F. Stegelmann, S. Schwemmers, A. Grunder, J. C. Peeken, M. Gothwal, J. Wehrle, K. Aumann, K. Hamdi, C. Dierks, W. Kamar Wang, K. Dohner, J. H. Jansen, and H. L. Pahl. MPN patients harbor recurrent truncating mutations in transcription factor NF-E2. J. Exp. Med., 210(5):1003–1019, May 2013.

[174] Jonas Jutzi. Mutationen im Transkriptionsfaktor NF-E2 und deren Rolle in der Entwicklung myeloproliferativer Neoplasien. MD thesis, Albert-Ludwigs-Universität, Freiburg, Germany, 2012.

138 References

[175] T. Kameda, K. Shide, T. Yamaji, A. Kamiunten, M. Sekine, Y. Taniguchi, T. Hidaka, Y. Kubuki, H. Shimoda, K. Marutsuka, G. Sashida, K. Aoyama, M. Yoshimitsu, T. Harada, H. Abe, T. Miike, H. Iwakiri, Y. Tahara, M. Sueta, S. Yamamoto, S. Hasuike, K. Nagata, A. Iwama, A. Kitanaka, and K. Shimoda. Loss of TET2 has dual roles in murine myeloproliferative neoplasms: disease sustainer and disease accelerator. Blood, 125(2):304–315, Jan 2015.

[176] M. G. Kang, H. W. Choi, J. H. Lee, Y. J. Choi, H. J. Choi, J. H. Shin, S. P. Suh, M. Szardenings, H. R. Kim, and M. G. Shin. Coexistence of JAK2 and CALR mutations and their clinical implications in patients with essential thrombocythemia. Oncotarget, 7(35):57036–57049, 08 2016.

[177] R. Kanwal, K. Gupta, and S. Gupta. Cancer epigenetics: an introduction. Methods Mol. Biol., 1238:3–25, 2015.

[178] K. Kapralova, L. Lanikova, F. Lorenzo, J. Song, M. Horvathova, V. Divoky, and J. T. Prchal. RUNX1 and NF-E2 upregulation is not specific for MPNs, but is seen in polycythemic disorders with augmented HIF signaling. Blood, 123(3):391–394, Jan 2014.

[179] M. Kapushesky, I. Emam, E. Holloway, P. Kurnosov, A. Zorin, J. Malone, G. Rustici, E. Williams, H. Parkinson, and A. Brazma. Gene expression atlas at the European bioinformatics institute. Nucleic Acids Res., 38(Database issue):D690–698, Jan 2010.

[180] N. Karunasena, T. S. Han, A. Mallappa, M. Elman, D. P. Merke, R. J. Ross, and E. Daniel. Androgens correlate with increased erythropoiesis in women with congenital adrenal hyperplasia. Clin. Endocrinol. (Oxf), 86(1):19–25, Jan 2017.

[181] M. Katoh and M. Katoh. Comparative integromics on JMJD1C gene encoding histone demethylase: conserved POU5F1 binding site elucidating mechanism of JMJD1C expression in undifferentiated ES cells and diffuse-type gastric cancer. Int. J. Oncol., 31(1):219–223, Jul 2007.

[182] K. B. Kaufmann, A. Grunder, T. Hadlich, J. Wehrle, M. Gothwal, R. Bogeska, T. S. Seeger, S. Kayser, K. B. Pham, J. S. Jutzi, L. Ganzenmuller, D. Steinemann, B. Schlegelberger, J. M. Wagner, M. Jung, B. Will, U. Steidl, K. Aumann, M. Werner, T. Gunther, R. Schule, A. Rambaldi, and H. L. Pahl. A novel

139 References

murine model of myeloproliferative disorders generated by overexpression of the transcription factor NF-E2. J. Exp. Med., 209(1):35–50, Jan 2012.

[183] H. Kawamoto and Y. Katsura. A new paradigm for hematopoietic cell lineages: revision of the classical concept of the myeloid-lymphoid dichotomy. Trends Immunol., 30(5):193–200, May 2009.

[184] I. Khan, A. Shergill, S. L. Saraf, Y. F. Chen, P. R. Patel, J. G. Quigley, D. Peace, V. R. Gordeuk, R. Hoffman, and D. Rondelli. Outcome Disparities in Caucasian and Non-Caucasian Patients With Myeloproliferative Neoplasms. Clin Lymphoma Myeloma Leuk, 16(6):350–357, Jun 2016.

[185] S. Khorasanizadeh. Recognition of methylated histones: new twists and variations. Curr. Opin. Struct. Biol., 21(6):744–749, Dec 2011.

[186] C. M. Kiekhaefer, J. A. Grass, K. D. Johnson, M. E. Boyer, and E. H. Bresnick. Hematopoietic-specific activators establish an overlapping pattern of histone acetylation and methylation within a mammalian chromatin domain. Proc. Natl. Acad. Sci. U.S.A., 99(22):14309–14314, Oct 2002.

[187] M. J. Kiel, O. H. Yilmaz, T. Iwashita, O. H. Yilmaz, C. Terhorst, and S. J. Morrison. SLAM family receptors distinguish hematopoietic stem and progenitor cells and reveal endothelial niches for stem cells. Cell, 121(7):1109–1121, Jul 2005.

[188] J. J. Kiladjian, R. A. Mesa, and R. Hoffman. The renaissance of interferon therapy for the treatment of myeloid malignancies. Blood, 117(18):4706–4715, May 2011.

[189] O. Kilpivaara, S. Mukherjee, A. M. Schram, M. Wadleigh, A. Mullally, B. L. Ebert, A. Bass, S. Marubayashi, A. Heguy, G. Garcia-Manero, H. Kantarjian, K. Offit, R. M. Stone, D. G. Gilliland, R. J. Klein, and R. L. Levine. A germline JAK2 SNP is associated with predisposition to the development of JAK2(V617F)-positive myeloproliferative neoplasms. Nat. Genet., 41(4):455–459, Apr 2009.

[190] M. Kim, H. B. Moon, and G. J. Spangrude. Major age-related changes of mouse hematopoietic stem/progenitor cells. Ann. N. Y. Acad. Sci., 996:195–208, May 2003.

[191] S. M. Kim, J. Y. Kim, N. W. Choe, I. H. Cho, J. R. Kim, D. W. Kim, J. E. Seol, S. E. Lee, H. Kook, K. I. Nam, H. Kook, Y. Y. Bhak, and S. B. Seo. Regulation

140 References

of mouse steroidogenesis by WHISTLE and JMJD1C through histone methylation balance. Nucleic Acids Res., 38(19):6389–6403, Oct 2010.

[192] M. Kimmel. Stochasticity and determinism in models of hematopoiesis. Adv. Exp. Med. Biol., 844:119–152, 2014.

[193] M. Kinder, C. Wei, S. G. Shelat, M. Kundu, L. Zhao, I. A. Blair, and E. Pure. Hematopoietic stem cell function requires 12/15-lipoxygenase-dependent fatty acid metabolism. Blood, 115(24):5012–5022, Jun 2010.

[194] P. D. Kingsley, E. Greenfest-Allen, J. M. Frame, T. P. Bushnell, J. Malik, K. E. McGrath, C. J. Stoeckert, and J. Palis. Ontogeny of erythroid gene expression. Blood, 121(6):e5–e13, Feb 2013.

[195] K. Kirtane and S. J. Lee. Racial and ethnic disparities in hematologic malignancies. Blood, 130(15):1699–1705, 10 2017.

[196] T. Klampfl, H. Gisslinger, A. S. Harutyunyan, H. Nivarthi, E. Rumi, J.D. Milosevic, N. C. Them, T. Berg, B. Gisslinger, D. Pietra, D. Chen, G. I. Vladimer, K. Bagienski, C. Milanesi, I. C. Casetti, E. Sant’Antonio, V. Ferretti, C. Elena, F. Schischlik, C. Cleary, M. Six, M. Schalling, A. Schonegger, C. Bock, L. Malcovati, C. Pascutto, G. Superti-Furga, M. Cazzola, and R. Kralovics. Somatic mutations of calreticulin in myeloproliferative neoplasms. N. Engl. J. Med., 369(25):2379–2390, Dec 2013.

[197] P. F. Klein-Weigel, T. S. Volz, and J. G. Richter. Erythromelalgia. VASA, 47(2):91–97, Feb 2018.

[198] M. Kleppe, M. Kwak, P. Koppikar, M. Riester, M. Keller, L. Bastian, T. Hri- cik, N. Bhagwat, A. S. McKenney, E. Papalexi, O. Abdel-Wahab, R. Rampal, S. Marubayashi, J. J. Chen, V. Romanet, J. S. Fridman, J. Bromberg, J. Teruya- Feldstein, M. Murakami, T. Radimerski, F. Michor, R. Fan, and R. L. Levine. JAK-STAT pathway activation in malignant and nonmalignant cells contributes to MPN pathogenesis and therapeutic response. Cancer Discov, 5(3):316–331, Mar 2015.

[199] S. Klippel, E. Strunck, C. E. Busse, D. Behringer, and H. L. Pahl. Biochemical characterization of PRV-1, a novel hematopoietic cell surface receptor, which is overexpressed in polycythemia rubra vera. Blood, 100(7):2441–2448, Oct 2002.

141 References

[200] R. J. Klose and A. P. Bird. Genomic DNA methylation: the mark and its mediators. Trends Biochem. Sci., 31(2):89–97, Feb 2006.

[201] R. J. Klose, E. M. Kallin, and Y. Zhang. JmjC-domain-containing proteins and histone demethylation. Nat. Rev. Genet., 7(9):715–727, 09 2006.

[202] F. Kohlhuber, N. C. Rogers, D. Watling, J. Feng, D. Guschin, J. Briscoe, B. A. Witthuhn, S. V. Kotenko, S. Pestka, G. R. Stark, J. N. Ihle, and I. M. Kerr. A JAK1/JAK2 chimera can sustain alpha and gamma interferon responses. Mol. Cell. Biol., 17(2):695–706, Feb 1997.

[203] M. Kondo, I. L. Weissman, and K. Akashi. Identification of clonogenic common lymphoid progenitors in mouse bone marrow. Cell, 91(5):661–672, Nov 1997.

[204] A. Kozomara and S. Griffiths-Jones. miRBase: annotating high confidence microR- NAs using deep sequencing data. Nucleic Acids Res., 42(Database issue):68–73, Jan 2014.

[205] R. Kralovics, Y. Guan, and J. T. Prchal. Acquired uniparental disomy of chromosome 9p is a frequent stem cell defect in polycythemia vera. Exp. Hematol., 30(3):229–236, Mar 2002.

[206] R. Kralovics, S. S. Teo, S. Li, A. Theocharides, A. S. Buser, A. Tichelli, and R. C. Skoda. Acquisition of the V617F mutation of JAK2 is a late genetic event in a subset of patients with myeloproliferative disorders. Blood, 108(4):1377–1380, Aug 2006.

[207] A. Krinner and I. Roeder. Quantification and modeling of stem cell-niche interaction. Adv. Exp. Med. Biol., 844:11–36, 2014.

[208] S. Y. Kristinsson, O. Landgren, J. Samuelsson, M. Bjorkholm, and L. R. Goldin. Autoimmunity and the risk of myeloproliferative neoplasms. Haematologica, 95(7):1216–1220, Jul 2010.

[209] M. Kucia, R. Reca, F. R. Campbell, E. Zuba-Surma, M. Majka, J. Ratajczak, and M. Z. Ratajczak. A population of very small embryonic-like (VSEL) CXCR4(+)SSEA-1(+)Oct-4+ stem cells identified in adult bone marrow. Leukemia, 20(5):857–869, May 2006.

142 References

[210] Y. Kunisaki, I. Bruns, C. Scheiermann, J. Ahmed, S. Pinho, D. Zhang, T. Mizoguchi, Q. Wei, D. Lucas, K. Ito, J. C. Mar, A. Bergman, and P. S. Frenette. Arteriolar niches maintain haematopoietic stem cell quiescence. Nature, 502(7473):637–643, Oct 2013.

[211] S. Kuroki, M. Akiyoshi, M. Tokura, H. Miyachi, Y. Nakai, H. Kimura, Y. Shinkai, and M. Tachibana. JMJD1C, a JmjC domain-containing protein, is required for long-term maintenance of male germ cells in mice. Biol. Reprod., 89(4):93, Oct 2013.

[212] H. M. Kvasnicka, A. Orazi, J. Thiele, G. Barosi, C. E. Bueso-Ramos, A. M. Vannucchi, R. P. Hasserjian, J. J. Kiladjian, U. Gianelli, R. Silver, T. I. Mughal, and T. Barbui. European LeukemiaNet study on the reproducibility of bone marrow features in masked polycythemia vera and differentiation from essential thrombocythemia. Am. J. Hematol., 92(10):1062–1067, Oct 2017.

[213] V. Kyriazi and E. Theodoulou. Assessing the risk and prognosis of thrombotic complications in cancer patients. Arch. Pathol. Lab. Med., 137(9):1286–1295, Sep 2013.

[214] T. Lamy, A. Devillers, M. Bernard, A. Moisan, I. Grulois, B. Drenou, L. Amiot, R. Fauchet, and P. Y. Le Prise. Inapparent polycythemia vera: an unrecognized diagnosis. Am. J. Med., 102(1):14–20, Jan 1997.

[215] T. L. Lasho, C. M. Finke, A. Tischer, A. Pardanani, and A. Tefferi. Mayo CALR mutation type classification guide using alpha helix propensity. Am. J. Hematol., 93(5):E128–E129, May 2018.

[216] A. H. Lau, H. K. Lai, B. H. Yeung, S. L. Leung, S. Y. Tsang, Y. H. Wong, and H. Wise. Prostacyclin receptor-dependent inhibition of human erythroleukemia cell differentiation is STAT3-dependent. Prostaglandins Leukot. Essent. Fatty Acids, 86(3):119–126, Mar 2012.

[217] P. Lecine, J. L. Villeval, P. Vyas, B. Swencki, Y. Xu, and R. A. Shivdasani. Mice lacking transcription factor NF-E2 provide in vivo validation of the proplatelet model of thrombocytopoiesis and show a platelet production defect that is intrinsic to megakaryocytes. Blood, 92(5):1608–1616, Sep 1998.

[218] J. W. Lee, H. S. Choi, J. Gyuris, R. Brent, and D. D. Moore. Two classes of proteins dependent on either the presence or absence of thyroid hormone for

143 References

interaction with the thyroid hormone receptor. Mol. Endocrinol., 9(2):243–254, Feb 1995.

[219] E. Lefrancais, G. Ortiz-Munoz, A. Caudrillier, B. Mallavia, F. Liu, D. M. Sayah, E. E. Thornton, M. B. Headley, T. David, S. R. Coughlin, M. F. Krummel, A. D. Leavitt, E. Passegue, and M. R. Looney. The lung is a site of platelet biogenesis and a reservoir for haematopoietic progenitors. Nature, 544(7648):105–109, 04 2017.

[220] A. Leiba, A. Duek, A. Afek, E. Derazne, and M. Leiba. Obesity and related risk of myeloproliferative neoplasms among israeli adolescents. Obesity (Silver Spring), 25(7):1187–1190, 07 2017.

[221] O. Leiva, S. K. Ng, S. Chitalia, A. Balduini, S. Matsuura, and K. Ravid. The role of the extracellular matrix in primary myelofibrosis. Blood Cancer J, 7(2):e525, 02 2017.

[222] C. J. Lengner, F. D. Camargo, K. Hochedlinger, G. G. Welstead, S. Zaidi, S. Gokhale, H. R. Scholer, A. Tomilin, and R. Jaenisch. Oct4 expression is not required for mouse somatic stem cell self-renewal. Cell Stem Cell, 1(4):403–415, Oct 2007.

[223] E. Leroy, J. P. Defour, T. Sato, S. Dass, V. Gryshkova, M. M. Shwe, J. Staerk, S. N. Constantinescu, and S. O. Smith. His499 Regulates Dimerization and Prevents Oncogenic Activation by Asparagine Mutations of the Human Thrombopoietin Receptor. J. Biol. Chem., 291(6):2974–2987, Feb 2016.

[224] E. Lesteven, M. Picque, C. Conejero Tonetti, S. Giraudier, N. Varin-Blank, L. Ve- lazquez, J. J. Kiladjian, B. Cassinat, and F. Baran-Marszak. Association of a single-nucleotide polymorphism in the SH2B3 gene with JAK2V617F-positive myeloproliferative neoplasms. Blood, 123(5):794–796, Jan 2014.

[225] L. Li, X. Lin, A. Khvorova, S. W. Fesik, and Y. Shen. Defining the optimal parameters for hairpin-based knockdown constructs. RNA, 13(10):1765–1774, Oct 2007.

[226] W. Li, E. Tian, Z. X. Chen, G. Sun, P. Ye, S. Yang, D. Lu, J. Xie, T. V. Ho, W. M. Tsark, C. Wang, D. A. Horne, A. D. Riggs, M. L. Yip, and Y. Shi. Identification of Oct4-activating compounds that enhance reprogramming efficiency. Proc. Natl. Acad. Sci. U.S.A., 109(51):20853–20858, Dec 2012.

144 References

[227] Z. Li, A. H. Pearlman, and P. Hsieh. DNA mismatch repair and the DNA damage response. DNA Repair (Amst.), 38:94–101, Feb 2016.

[228] B. K. Lim. Clinics in diagnostic imaging (146). Polycythaemia vera (PV). Singapore Med J, 54(5):289–291, May 2013.

[229] F. Liu, X. Zhao, F. Perna, L. Wang, P. Koppikar, O. Abdel-Wahab, M. W. Harr, R. L. Levine, H. Xu, A. Tefferi, A. Deblasio, M. Hatlen, S. Menendez, and S. D. Nimer. JAK2V617F-mediated phosphorylation of PRMT5 downregulates its methyltransferase activity and promotes myeloproliferation. Cancer Cell, 19(2):283–294, Feb 2011.

[230] T. H. Liu, A. Raval, S. S. Chen, J. J. Matkovic, J. C. Byrd, and C. Plass. CpG island methylation and expression of the secreted frizzled-related protein gene family in chronic lymphocytic leukemia. Cancer Res., 66(2):653–658, Jan 2006.

[231] N. A. Lobo, Y. Shimono, D. Qian, and M. F. Clarke. The biology of cancer stem cells. Annu. Rev. Cell Dev. Biol., 23:675–699, 2007.

[232] M. Loose and R. Patient. Global genetic regulatory networks controlling hematopoietic cell fates. Curr. Opin. Hematol., 13(4):229–236, Jul 2006.

[233] M. Lu, L. Xia, Y. C. Liu, T. Hochman, L. Bizzari, D. Aruch, J. Lew, R. Weinberg, J. D. Goldberg, and R. Hoffman. Lipocalin produced by myelofibrosis cells affects the fate of both hematopoietic and marrow microenvironmental cells. Blood, 126(8):972–982, Aug 2015.

[234] X. Lu, R. Levine, W. Tong, G. Wernig, Y. Pikman, S. Zarnegar, D. G. Gilliland, and H. Lodish. Expression of a homodimeric type I cytokine receptor is required for JAK2V617F-mediated transformation. Proc. Natl. Acad. Sci. U.S.A., 102(52):18962–18967, Dec 2005.

[235] M. Lucijanic, A. Livun, C. Tomasovic-Loncaric, T. Stoos-Veic, V. Pejsa, O. Jaksic, Z. Prka, and R. Kusec. Canonical Wnt/β-Catenin Signaling Pathway Is Dysreg- ulated in Patients With Primary and Secondary Myelofibrosis. Clin Lymphoma Myeloma Leuk, 16(9):523–526, 09 2016.

[236] P. Lundberg, A. Karow, R. Nienhold, R. Looser, H. Hao-Shen, I. Nissen, S. Girs- berger, T. Lehmann, J. Passweg, M. Stern, C. Beisel, R. Kralovics, and R. C. Skoda.

145 References

Clonal evolution and clinical correlates of somatic mutations in myeloproliferative neoplasms. Blood, 123(14):2220–2228, Apr 2014.

[237] F. Lussana, A. Carobbio, M. L. Randi, C. Elena, E. Rumi, G. Finazzi, I. Bertozzi, L. Pieri, M. Ruggeri, F. Palandri, N. Polverelli, E. Elli, A. Tieghi, A. Iurlo, M. Ruella, M. Cazzola, A. Rambaldi, A. M. Vannucchi, and T. Barbui. A lower intensity of treatment may underlie the increased risk of thrombosis in young patients with masked polycythaemia vera. Br. J. Haematol., 167(4):541–546, Nov 2014.

[238] F. Lussana, A. Carobbio, S. Salmoiraghi, P. Guglielmelli, A. M. Vannucchi, B. Bottazzi, R. Leone, A. Mantovani, T. Barbui, and A. Rambaldi. Driver mutations (JAK2V617F, MPLW515L/K or CALR), pentraxin-3 and C-reactive protein in essential thrombocythemia and polycythemia vera. J Hematol Oncol, 10(1):54, Feb 2017.

[239] F. Lussana and A. Rambaldi. Inflammation and myeloproliferative neoplasms. J. Autoimmun., 85:58–63, Dec 2017.

[240] R. N. Mackinnon, M. Wall, A. Zordan, S. Nutalapati, B. Mercer, J. Peverall, and L. J. Campbell. Genome organization and the role of centromeres in evolution of the erythroleukaemia cell line HEL. Evol Med Public Health, 2013(1):225–240, Jan 2013.

[241] L. F. Mager, C. Riether, C. M. Schurch, Y. Banz, M. H. Wasmer, R. Stuber, A. P. Theocharides, X. Li, Y. Xia, H. Saito, S. Nakae, G. M. Baerlocher, M. G. Manz, K. D. McCoy, A. J. Macpherson, A. F. Ochsenbein, B. Beutler, and P. Krebs. IL-33 signaling contributes to the pathogenesis of myeloproliferative neoplasms. J. Clin. Invest., 125(7):2579–2591, Jul 2015.

[242] N. Mahmud, S. M. Devine, K. P. Weller, S. Parmar, C. Sturgeon, M. C. Nelson, T. Hewett, and R. Hoffman. The relative quiescence of hematopoietic stem cellsin nonhuman primates. Blood, 97(10):3061–3068, May 2001.

[243] S. Malinge, C. Ragu, V. Della-Valle, D. Pisani, S. N. Constantinescu, C. Perez, J. L. Villeval, D. Reinhardt, J. Landman-Parker, L. Michaux, N. Dastugue, A. Baruchel, W. Vainchenker, J. P. Bourquin, V. Penard-Lacronique, and O. A. Bernard. Activating mutations in human acute megakaryoblastic leukemia. Blood, 112(10):4220–4226, Nov 2008.

146 References

[244] J. K. Manesia, M. Franch, D. Tabas-Madrid, R. Nogales-Cadenas, T. Vanwelden, E. Van Den Bosch, Z. Xu, A. Pascual-Montano, S. Khurana, and C. M. Verfaillie. Distinct Molecular Signature of Murine Fetal Liver and Adult Hematopoietic Stem Cells Identify Novel Regulators of Hematopoietic Stem Cell Function. Stem Cells Dev., Feb 2017.

[245] Robert Mansson, Anne Hultquist, Sidinh Luc, Liping Yang, Kristina Anderson, Shabnam Kharazi, Suleiman Al-Hashmi, Karina Liuba, Lina Thorén, Jorgen Adolfs- son, Natalija Buza-Vidas, Hong Qian, Shamit Soneji, Tariq Enver, Mikael Sigvards- son, and Sten Eirik W. Jacobsen. Molecular evidence for hierarchical transcriptional lineage priming in fetal and adult stem cells and multipotent progenitors. Immunity, 26:407–419, April 2007.

[246] A. Mantovani, P. Allavena, A. Sica, and F. Balkwill. Cancer-related inflammation. Nature, 454(7203):436–444, Jul 2008.

[247] G. Marcucci, K. Maharry, Y. Z. Wu, M. D. Radmacher, K. Mrozek, D. Margeson, K. B. Holland, S. P. Whitman, H. Becker, S. Schwind, K. H. Metzeler, B. L. Powell, T. H. Carter, J. E. Kolitz, M. Wetzler, A. J. Carroll, M. R. Baer, M. A. Caligiuri, R. A. Larson, and C. D. Bloomfield. IDH1 and IDH2 gene mutations identify novel molecular subsets within de novo cytogenetically normal acute myeloid leukemia: a Cancer and Leukemia Group B study. J. Clin. Oncol., 28(14):2348–2355, May 2010.

[248] R. Marmorstein and R. C. Trievel. Histone modifying enzymes: structures, mechanisms, and specificities. Biochim. Biophys. Acta, 1789(1):58–68, Jan 2009.

[249] L. Martinez-Aviles, A. Alvarez-Larran, C. Besses, G. Navarro, E. Torres, R. Lon- garon, A. Angona, C. Pedro, L. Florensa, S. Serrano, and B. Bellosillo. Clinical significance of clonality assessment in JAK2V617F-negative essential thrombocythemia. Ann. Hematol., 91(10):1555–1562, Oct 2012.

[250] J. Mascarenhas, N. Roper, P. Chaurasia, and R. Hoffman. Epigenetic abnormalities in myeloproliferative neoplasms: a target for novel therapeutic strategies. Clin Epigenetics, 2(2):197–212, Aug 2011.

[251] J. Mascarenhas, N. Roper, P. Chaurasia, and R. Hoffman. Epigenetic abnormalities in myeloproliferative neoplasms: a target for novel therapeutic strategies. Clin Epigenetics, 2(2):197–212, Aug 2011.

147 References

[252] F. Mazurier, O. I. Gan, J. L. McKenzie, M. Doedens, and J. E. Dick. Lentivector- mediated clonal tracking reveals intrinsic heterogeneity in the human hematopoietic stem cell compartment and culture-induced stem cell impairment. Blood, 103(2):545– 552, Jan 2004.

[253] J. L. McBride, R. L. Boudreau, S. Q. Harper, P. D. Staber, A. M. Monteys, I. Martins, B. L. Gilmore, H. Burstein, R. W. Peluso, B. Polisky, B. J. Carter, and B. L. Davidson. Artificial miRNAs mitigate shRNA-mediated toxicity inthe brain: implications for the therapeutic development of RNAi. Proc. Natl. Acad. Sci. U.S.A., 105(15):5868–5873, Apr 2008.

[254] M. A. McDonough, C. Loenarz, R. Chowdhury, I. J. Clifton, and C. J. Schofield. Structural studies on human 2-oxoglutarate dependent oxygenases. Curr. Opin. Struct. Biol., 20(6):659–672, Dec 2010.

[255] A. S. McKenney, A. N. Lau, A. V. H. Somasundara, B. Spitzer, A. M. Intlekofer, J. Ahn, K. Shank, F. T. Rapaport, M. A. Patel, E. Papalexi, A. H. Shih, A. Chiu, E. Freinkman, E. A. Akbay, M. Steadman, R. Nagaraja, K. Yen, J. Teruya- Feldstein, K. K. Wong, R. Rampal, M. G. Vander Heiden, C. B. Thompson, and R. L. Levine. JAK2/IDH-mutant-driven myeloproliferative neoplasm is sensitive to combined targeted inhibition. J. Clin. Invest., 128(2):789–804, 02 2018.

[256] M. T. McManus, C. P. Petersen, B. B. Haines, J. Chen, and P. A. Sharp. Gene silencing using micro-RNA designed hairpins. RNA, 8(6):842–850, Jun 2002.

[257] R. J. McNally, D. Rowland, E. Roman, and R. A. Cartwright. Age and sex distributions of hematological malignancies in the U.K. Hematol Oncol, 15(4):173– 189, Nov 1997.

[258] H. A. Meijer, E. M. Smith, and M. Bushell. Regulation of miRNA strand selection: follow the leader? Biochem. Soc. Trans., 42(4):1135–1140, Aug 2014.

[259] R. A. Mesa and R. Tibes. MPN blast phase: clinical challenge and assessing response. Leuk. Res., 36(12):1496–1497, Dec 2012.

[260] R. A. Mesa, S. Verstovsek, F. Cervantes, G. Barosi, J. T. Reilly, B. Dupriez, R. Levine, M. C. Le Bousse-Kerdiles, M. Wadleigh, P. J. Campbell, R. T. Silver, A. M. Vannucchi, H. J. Deeg, H. Gisslinger, D. Thomas, O. Odenike, L. A. Solberg, J. Gotlib, E. Hexner, S. D. Nimer, H. Kantarjian, A. Orazi, J. W. Vardiman, J. Thiele, and A. Tefferi. Primary myelofibrosis (PMF), post polycythemia vera

148 References

myelofibrosis (post-PV MF), post essential thrombocythemia myelofibrosis (post-ET MF), blast phase PMF (PMF-BP): Consensus on terminology by the international working group for myelofibrosis research and treatment (IWG-MRT). Leuk. Res., 31(6):737–740, Jun 2007.

[261] N. Methia, N. Debili, M. Titeux, W. Vainchenker, and F. Wendling. From the v-mpl oncogene to thrombopoietin. C. R. Acad. Sci. III, Sci. Vie, 318(4):479–482, Apr 1995.

[262] M. Michalak, J. Groenendyk, E. Szabo, L. I. Gold, and M. Opas. Calreticulin, a multi-process calcium-buffering chaperone of the endoplasmic reticulum. Biochem. J., 417(3):651–666, Feb 2009.

[263] J. J. Michiels, Z. Berneman, D. Van Bockstaele, M. van der Planken, H. De Raeve, and W. Schroyens. Clinical and laboratory features, pathobiology of platelet- mediated thrombosis and bleeding complications, and the molecular etiology of essential thrombocythemia and polycythemia vera: therapeutic implications. Semin. Thromb. Hemost., 32(3):174–207, Apr 2006.

[264] J. J. Michiels, M. Tevet, A. Trifa, E. Niculescu-Mizil, A. Lupu, A. M. Vladareanu, H. Bumbea, A. Ilea, C. Dobrea, D. Georgescu, O. Patrinoiu, M. Popescu, M. Murat, C. Dragan, F. Mihai, S. Zurac, S. Angelescu, A. Iova, A. Popa, R. Gogulescu, and V. Popov. 2016 WHO Clinical Molecular and Pathological Criteria for Classification and Staging of Myeloproliferative Neoplasms (MPN) Caused byMPN Driver Mutations in the JAK2, MPL and CALR Genes in the Context of New 2016 WHO Classification: Prognostic and Therapeutic Implications. Maedica (Buchar), 11(1):5–25, Mar 2016.

[265] M. K. Middleton, A. M. Zukas, T. Rubinstein, M. Jacob, P. Zhu, L. Zhao, I. Blair, and E. Pure. Identification of 12/15-lipoxygenase as a suppressor of myeloproliferative disease. J. Exp. Med., 203(11):2529–2540, Oct 2006.

[266] K. Mierzejewska, S. Borkowska, E. Suszynska, M. Suszynska, A. Poniewierska- Baran, M. Maj, D. Pedziwiatr, M. Adamiak, A. Abdel-Latif, S. S. Kakar, J. Rata- jczak, M. Kucia, and M. Z. Ratajczak. Hematopoietic stem/progenitor cells express several functional sex hormone receptors-novel evidence for a potential developmental link between hematopoiesis and primordial germ cells. Stem Cells Dev., 24(8):927–937, Apr 2015.

149 References

[267] L. K. Miller and R. D. Wells. Nucleoside diphosphokinase activity associated with DNA polymerases. Proc. Natl. Acad. Sci. U.S.A., 68(9):2298–2302, Sep 1971.

[268] K. Minton. Inflammation: Inflammasome-related ageing. Nat. Rev. Immunol., 17(2):77, 02 2017.

[269] H. Miyoshi, U. Blomer, M. Takahashi, F. H. Gage, and I. M. Verma. Development of a self-inactivating lentivirus vector. J. Virol., 72(10):8150–8157, Oct 1998.

[270] A. R. Moliterno, D. M. Williams, L. I. Gutierrez-Alamillo, R. Salvatori, R. G. Ingersoll, and J. L. Spivak. Mpl Baltimore: a thrombopoietin receptor polymorphism associated with thrombocytosis. Proc. Natl. Acad. Sci. U.S.A., 101(31):11444–11447, Aug 2004.

[271] K. Moran-Crusio, L. Reavie, A. Shih, O. Abdel-Wahab, D. Ndiaye-Lobry, C. Lobry, M. E. Figueroa, A. Vasanthakumar, J. Patel, X. Zhao, F. Perna, S. Pandey, J. Madzo, C. Song, Q. Dai, C. He, S. Ibrahim, M. Beran, J. Zavadil, S. D. Nimer, A. Melnick, L. A. Godley, I. Aifantis, and R. L. Levine. Tet2 loss leads to increased hematopoietic stem cell self-renewal and myeloid transformation. Cancer Cell, 20(1):11–24, Jul 2011.

[272] S. J. Morrison, A. M. Wandycz, K. Akashi, A. Globerson, and I. L. Weissman. The aging of hematopoietic stem cells. Nat. Med., 2(9):1011–1016, Sep 1996.

[273] S. J. Morrison and I. L. Weissman. The long-term repopulating subset of hematopoietic stem cells is deterministic and isolatable by phenotype. Immunity, 1(8):661–673, Nov 1994.

[274] O. Moulard, J. Mehta, J. Fryzek, R. Olivares, U. Iqbal, and R. A. Mesa. Epidemi- ology of myelofibrosis, essential thrombocythemia, and polycythemia vera in the European Union. Eur. J. Haematol., 92(4):289–297, Apr 2014.

[275] A. Mullally, C. Bruedigam, L. Poveromo, F. H. Heidel, A. Purdon, T. Vu, R. Austin, D. Heckl, L. J. Breyfogle, C. P. Kuhn, D. Kalaitzidis, S. A. Armstrong, D. A. Williams, G. R. Hill, B. L. Ebert, and S. W. Lane. Depletion of Jak2V617F myeloproliferative neoplasm-propagating stem cells by interferon-α in a murine model of polycythemia vera. Blood, 121(18):3692–3702, May 2013.

150 References

[276] H. Muta, A. Funakoshi, T. Baba, N. Uike, H. Wakasugi, M. Kozuru, and A. Jimi. Gene expression of erythropoietin in hepatocellular carcinoma. Intern. Med., 33(7):427–431, Jul 1994.

[277] M. Mutschler, A. S. Magin, M. Buerge, R. Roelz, D. H. Schanne, B. Will, I. H. Pilz, A. R. Migliaccio, and H. L. Pahl. NF-E2 overexpression delays erythroid maturation and increases erythrocyte production. Br. J. Haematol., 146(2):203–217, Jul 2009.

[278] T. Nakamura, D. A. Largaespada, M. P. Lee, L. A. Johnson, K. Ohyashiki, K. Toyama, S. J. Chen, C. L. Willman, I. M. Chen, A. P. Feinberg, N. A. Jenkins, N. G. Copeland, and J. D. Shaughnessy. Fusion of the nucleoporin gene NUP98 to HOXA9 by the chromosome translocation t(7;11)(p15;p15) in human myeloid leukaemia. Nat. Genet., 12(2):154–158, Feb 1996.

[279] J. Nangalia, C. E. Massie, E. J. Baxter, F. L. Nice, G. Gundem, D. C. Wedge, E. Avezov, J. Li, K. Kollmann, D. G. Kent, A. Aziz, A. L. Godfrey, J. Hinton, I. Martincorena, P. Van Loo, A. V. Jones, P. Guglielmelli, P. Tarpey, H. P. Harding, J. D. Fitzpatrick, C. T. Goudie, C. A. Ortmann, S. J. Loughran, K. Raine, D. R. Jones, A. P. Butler, J. W. Teague, S. O’Meara, S. McLaren, M. Bianchi, Y. Silber, D. Dimitropoulou, D. Bloxham, L. Mudie, M. Maddison, B. Robinson, C. Keohane, C. Maclean, K. Hill, K. Orchard, S. Tauro, M. Q. Du, M. Greaves, D. Bowen, B. J. P. Huntly, C. N. Harrison, N. C. P. Cross, D. Ron, A. M. Vannucchi, E. Papaemmanuil, P. J. Campbell, and A. R. Green. Somatic CALR mutations in myeloproliferative neoplasms with nonmutated JAK2. N. Engl. J. Med., 369(25):2391–2405, Dec 2013.

[280] R. A. Nimmo, G. E. May, and T. Enver. Primed and ready: understanding lineage commitment through single cell analysis. Trends Cell Biol., 25(8):459–467, Aug 2015.

[281] P. C. NOWELL and D. A. HUNGERFORD. Chromosome studies on normal and leukemic human leukocytes. J. Natl. Cancer Inst., 25:85–109, Jul 1960.

[282] A. Oddsson, S. Y. Kristinsson, H. Helgason, D. F. Gudbjartsson, G. Masson, A. Sigurdsson, A. Jonasdottir, A. Jonasdottir, H. Steingrimsdottir, B. Vidars- son, S. Reykdal, G. I. Eyjolfsson, I. Olafsson, P. T. Onundarson, G. Runarsson,

151 References

O. Sigurdardottir, A. Kong, T. Rafnar, P. Sulem, U. Thorsteinsdottir, and K. Ste- fansson. The germline sequence variant rs2736100_C in TERT associates with myeloproliferative neoplasms. Leukemia, 28(6):1371–1374, Jun 2014.

[283] M. Ogawa. Differentiation and proliferation of hematopoietic stem cells. Blood, 81(11):2844–2853, Jun 1993.

[284] M. Ogawa, P. N. Porter, and T. Nakahata. Renewal and commitment to differentiation of hemopoietic stem cells (an interpretive review). Blood, 61(5):823–829, May 1983.

[285] D. Olcaydu, A. Harutyunyan, R. Jager, T. Berg, B. Gisslinger, I. Pabinger, H. Gisslinger, and R. Kralovics. A common JAK2 haplotype confers susceptibility to myeloproliferative neoplasms. Nat. Genet., 41(4):450–454, Apr 2009.

[286] N. A. O’Leary, M. W. Wright, J. R. Brister, S. Ciufo, D. Haddad, R. McVeigh, B. Rajput, B. Robbertse, B. Smith-White, D. Ako-Adjei, A. Astashyn, A. Badretdin, Y. Bao, O. Blinkova, V. Brover, V. Chetvernin, J. Choi, E. Cox, O. Ermolaeva, C. M. Farrell, T. Goldfarb, T. Gupta, D. Haft, E. Hatcher, W. Hlavina, V. S. Joardar, V. K. Kodali, W. Li, D. Maglott, P. Masterson, K. M. McGarvey, M. R. Murphy, K. O’Neill, S. Pujar, S. H. Rangwala, D. Rausch, L. D. Riddick, C. Schoch, A. Shkeda, S. S. Storz, H. Sun, F. Thibaud-Nissen, I. Tolstoy, R. E. Tully, A. R. Vatsan, C. Wallin, D. Webb, W. Wu, M. J. Landrum, A. Kimchi, T. Tatusova, M. DiCuccio, P. Kitts, T. D. Murphy, and K. D. Pruitt. Reference sequence (RefSeq) database at NCBI: current status, taxonomic expansion, and functional annotation. Nucleic Acids Res., 44(D1):D733–745, Jan 2016.

[287] R. Ono, T. Taki, T. Taketani, M. Taniwaki, H. Kobayashi, and Y. Hayashi. LCX, leukemia-associated protein with a CXXC domain, is fused to MLL in acute myeloid leukemia with trilineage dysplasia having t(10;11)(q22;q23). Cancer Res., 62(14):4075–4080, Jul 2002.

[288] C. A. Ortmann, D. G. Kent, J. Nangalia, Y. Silber, D. C. Wedge, J. Grinfeld, E. J. Baxter, C. E. Massie, E. Papaemmanuil, S. Menon, A. L. Godfrey, D. Dim- itropoulou, P. Guglielmelli, B. Bellosillo, C. Besses, K. Dohner, C. N. Harrison, G. S. Vassiliou, A. Vannucchi, P. J. Campbell, and A. R. Green. Effect of mutation order on myeloproliferative neoplasms. N. Engl. J. Med., 372(7):601–612, Feb 2015.

152 References

[289] W. Osler. Chronic cyanosis, with polycythaemia and enlarged spleen: a new clinical entity. 1903. Am. J. Med. Sci., 335(6):411–417, Jun 2008.

[290] P. J. Paddison, A. A. Caudy, E. Bernstein, G. J. Hannon, and D. S. Conklin. Short hairpin RNAs (shRNAs) induce sequence-specific silencing in mammalian cells. Genes Dev., 16(8):948–958, Apr 2002.

[291] A. D. Pardanani, R. L. Levine, T. Lasho, Y. Pikman, R. A. Mesa, M. Wadleigh, D. P. Steensma, M. A. Elliott, A. P. Wolanskyj, W. J. Hogan, R. F. McClure, M. R. Litzow, D. G. Gilliland, and A. Tefferi. MPL515 mutations in myeloproliferative and other myeloid disorders: a study of 1182 patients. Blood, 108(10):3472–3476, Nov 2006.

[292] C. Parekh and G. M. Crooks. Critical differences in hematopoiesis and lymphoid development between humans and mice. J. Clin. Immunol., 33(4):711–715, May 2013.

[293] S. W. Park, X. Hu, P. Gupta, Y. P. Lin, S. G. Ha, and L. N. Wei. SUMOylation of Tr2 orphan receptor involves Pml and fine-tunes Oct4 expression in stem cells. Nat. Struct. Mol. Biol., 14(1):68–75, Jan 2007.

[294] Y. Park and S. L. Gerson. DNA repair defects in stem cell function and aging. Annu. Rev. Med., 56:495–508, 2005.

[295] Stefano Parodi, Irene Santi, Enza Marani, Claudia Casella, Antonella Puppo, Simona Sola, Vincenzo Fontana, and Emanuele Stagnaro. Chronic diseases, medical history and familial cancer, and risk of leukemia and non-hodgkin’s lymphoma in an adult population: a case–control study. Cancer Causes & Control, 26(7):993–1002, 2015.

[296] J. H. Parr, M. Seed, I. Godsland, and V. Wynn. The effects of reverse sequential anti-androgen therapy (cyproterone acetate and ethinyl estradiol) on hematological parameters. J. Endocrinol. Invest., 10(3):237–239, Jun 1987.

[297] F. Passamonti, F. Cervantes, A. M. Vannucchi, E. Morra, E. Rumi, A. Pereira, P. Guglielmelli, E. Pungolino, M. Caramella, M. Maffioli, C. Pascutto, M. Lazzarino, M. Cazzola, and A. Tefferi. A dynamic prognostic model to predict survival in primary myelofibrosis: a study by the IWG-MRT (International Working Group for Myeloproliferative Neoplasms Research and Treatment). Blood, 115(9):1703–1708, Mar 2010.

153 References

[298] F. Passamonti, T. Giorgino, B. Mora, P. Guglielmelli, E. Rumi, M. Maffioli, A. Rambaldi, M. Caramella, R. Komrokji, J. Gotlib, J. J. Kiladjian, F. Cervantes, T. Devos, F. Palandri, V. De Stefano, M. Ruggeri, R. T. Silver, G. Benevolo, F. Albano, D. Caramazza, M. Merli, D. Pietra, R. Casalone, G. Rotunno, T. Barbui, M. Cazzola, and A. M. Vannucchi. A clinical-molecular prognostic model to predict survival in patients with post polycythemia vera and post essential thrombocythemia myelofibrosis. Leukemia, 31(12):2726–2731, Dec 2017.

[299] F. Passamonti, M. Maffioli, F. Cervantes, A. M. Vannucchi, E. Morra, T. Barbui, D. Caramazza, L. Pieri, E. Rumi, H. Gisslinger, L. Knoops, J. J. Kiladjian, B. Mora, N. Hollaender, C. Pascutto, C. Harrison, and M. Cazzola. Impact of ruxolitinib on the natural history of primary myelofibrosis: a comparison of the DIPSS and the COMFORT-2 cohorts. Blood, 123(12):1833–1835, Mar 2014.

[300] S. Patra, S. N. Biswas, J. Datta, and P. P. Chakraborty. Hypersomatotropism induced secondary polycythaemia leading to spontaneous pituitary apoplexy resulting in cure of acromegaly and remission of polycythaemia: ’The virtuous circle’. BMJ Case Rep, 2017, Dec 2017.

[301] D. R. Pattabiraman, C. McGirr, K. Shakhbazov, V. Barbier, K. Krishnan, P. Mukhopadhyay, P. Hawthorne, A. Trezise, J. Ding, S. M. Grimmond, P. Pap- athanasiou, W. S. Alexander, A. C. Perkins, J. P. Levesque, I. G. Winkler, and T. J. Gonda. Interaction of c-Myb with p300 is required for the induction of acute myeloid leukemia (AML) by human AML oncogenes. Blood, 123(17):2682–2690, Apr 2014.

[302] Jan Peeken. Characterization of JMJD1C and JMJD2c as novel NF-E2 target genes and their contribution to MPN pathophysiology. MD thesis, Albert-Ludwigs- Universität, Freiburg, Germany, 2014.

[303] Y. Pikman, B. H. Lee, T. Mercher, E. McDowell, B. L. Ebert, M. Gozo, A. Cuker, G. Wernig, S. Moore, I. Galinsky, D. J. DeAngelo, J. J. Clark, S. J. Lee, T. R. Golub, M. Wadleigh, D. G. Gilliland, and R. L. Levine. MPLW515L is a novel somatic activating mutation in myelofibrosis with myeloid metaplasia. PLoS Med., 3(7):e270, Jul 2006.

[304] R. S. Pillai, S. N. Bhattacharyya, C. G. Artus, T. Zoller, N. Cougot, E. Basyuk, E. Bertrand, and W. Filipowicz. Inhibition of translational initiation by Let-7 MicroRNA in human cells. Science, 309(5740):1573–1576, Sep 2005.

154 References

[305] M. Pizzi, R. T. Silver, A. Barel, and A. Orazi. Recombinant interferon-α in myelofibrosis reduces bone marrow fibrosis, improves its morphology andis associated with clinical response. Mod. Pathol., 28(10):1315–1323, Oct 2015.

[306] E. Pourcelot, C. Trocme, J. Mondet, S. Bailly, B. Toussaint, and P. Mossuz. Cytokine profiles in polycythemia vera and essential thrombocythemia patients: clinical implications. Exp. Hematol., 42(5):360–368, May 2014.

[307] A. Quintas-Cardama, K. Vaddi, P. Liu, T. Manshouri, J. Li, P. A. Scherle, E. Caulder, X. Wen, Y. Li, P. Waeltz, M. Rupar, T. Burn, Y. Lo, J. Kelley, M. Covington, S. Shepard, J. D. Rodgers, P. Haley, H. Kantarjian, J. S. Fridman, and S. Verstovsek. Preclinical characterization of the selective JAK1/2 inhibitor INCB018424: therapeutic implications for the treatment of myeloproliferative neoplasms. Blood, 115(15):3109–3117, Apr 2010.

[308] M. Qureshi and C. Harrison. Molecular classification of myeloproliferative neoplasms-pros and cons. Curr Hematol Malig Rep, 8(4):342–350, Dec 2013.

[309] M. Ramalho-Santos and H. Willenbring. On the origin of the term "stem cell". Cell Stem Cell, 1(1):35–38, Jun 2007.

[310] R. Rampal, F. Al-Shahrour, O. Abdel-Wahab, J. P. Patel, J. P. Brunel, C. H. Mermel, A. J. Bass, J. Pretz, J. Ahn, T. Hricik, O. Kilpivaara, M. Wadleigh, L. Busque, D. G. Gilliland, T. R. Golub, B. L. Ebert, and R. L. Levine. Integrated genomic analysis illustrates the central role of JAK-STAT pathway activation in myeloproliferative neoplasm pathogenesis. Blood, 123(22):e123–133, May 2014.

[311] L. Rheinemann, T. S. Seeger, J. Wehrle, and H. L. Pahl. NFE2 regulates transcription of multiple enzymes in the heme biosynthesis pathway. Haematologica, 99(10):e208–210, Oct 2014.

[312] C. Roaldsnes, R. Holst, H. Frederiksen, and W. Ghanima. Myeloproliferative neoplasms: trends in incidence, prevalence and survival in Norway. Eur. J. Haematol., 98(1):85–93, Jan 2017.

[313] L. Robb. Cytokine receptors and hematopoietic differentiation. Oncogene, 26(47):6715–6723, Oct 2007.

155 References

[314] A. Roch, S. Giger, M. Girotra, V. Campos, N. Vannini, O. Naveiras, S. Gobaa, and M. P. Lutolf. Single-cell analyses identify bioengineered niches for enhanced maintenance of hematopoietic stem cells. Nat Commun, 8(1):221, 08 2017.

[315] Roland Roelz. Silencing NF-E2 - Silencing Polyzythämia vera? MD thesis, Albert-Ludwigs-Universität, Freiburg, Germany, 2010.

[316] M. Rohrbacher, U. Berger, A. Hochhaus, G. Metzgeroth, K. Adam, T. Lahaye, S. Saussele, M. C. Muller, J. Hasford, H. Heimpel, and R. Hehlmann. Clinical trials underestimate the age of chronic myeloid leukemia (CML) patients. Incidence and median age of Ph/BCR-ABL-positive CML and other chronic myeloproliferative disorders in a representative area in Germany. Leukemia, 23(3):602–604, Mar 2009.

[317] N. R. Rose and R. J. Klose. Understanding the relationship between DNA methylation and histone lysine methylation. Biochim. Biophys. Acta, 1839(12):1362–1372, Dec 2014.

[318] Fred Rosner. Ophthalmology in medical aphorisms of moses maimonides. New York state journal of medicine, 74(4):699–703, 1974.

[319] V. Rosti, M. Massa, A. M. Vannucchi, G. Bergamaschi, R. Campanelli, A. Pecci, G. Viarengo, V. Meli, M. Marchetti, P. Guglielmelli, E. Bruno, M. Xu, R. Hoffman, and G. Barosi. The expression of CXCR4 is down-regulated on the CD34+ cells of patients with myelofibrosis with myeloid metaplasia. Blood Cells Mol. Dis., 38(3):280–286, 2007.

[320] V. Rosti, L. Villani, R. Riboni, V. Poletto, E. Bonetti, L. Tozzi, G. Bergamaschi, P. Catarsi, E. Dallera, F. Novara, M. Massa, R. Campanelli, G. Fois, B. Peruzzi, M. Lucioni, P. Guglielmelli, A. Pancrazzi, G. Fiandrino, O. Zuffardi, U. Magrini, M. Paulli, A. M. Vannucchi, and G. Barosi. Spleen endothelial cells from patients with myelofibrosis harbor the JAK2V617F mutation. Blood, 121(2):360–368, Jan 2013.

[321] A. Rottenstreich, G. Kleinstern, S. Krichevsky, D. Varon, D. Lavie, and Y. Kalish. Factors related to the development of acquired von Willebrand syndrome in patients with essential thrombocythemia and polycythemia vera. Eur. J. Intern. Med., 41:49–54, Jun 2017.

156 References

[322] J. D. Rowley. Letter: A new consistent chromosomal abnormality in chronic myelogenous leukaemia identified by quinacrine fluorescence and Giemsa staining. Nature, 243(5405):290–293, Jun 1973.

[323] J. F. Rual, K. Venkatesan, T. Hao, T. Hirozane-Kishikawa, A. Dricot, N. Li, G. F. Berriz, F. D. Gibbons, M. Dreze, N. Ayivi-Guedehoussou, N. Klitgord, C. Simon, M. Boxem, S. Milstein, J. Rosenberg, D. S. Goldberg, L. V. Zhang, S. L. Wong, G. Franklin, S. Li, J. S. Albala, J. Lim, C. Fraughton, E. Llamosas, S. Cevik, C. Bex, P. Lamesch, R. S. Sikorski, J. Vandenhaute, H. Y. Zoghbi, A. Smolyar, S. Bosak, R. Sequerra, L. Doucette-Stamm, M. E. Cusick, D. E. Hill, F. P. Roth, and M. Vidal. Towards a proteome-scale map of the human protein-protein interaction network. Nature, 437(7062):1173–1178, Oct 2005.

[324] L. Rui, L. S. Mathews, K. Hotta, T. A. Gustafson, and C. Carter-Su. Identification of SH2-Bbeta as a substrate of the tyrosine kinase JAK2 involved in growth hormone signaling. Mol. Cell. Biol., 17(11):6633–6644, Nov 1997.

[325] E. Rumi, E. Boveri, M. Bellini, D. Pietra, V. V. Ferretti, E. Sant’Antonio, C. Cavalloni, I. C. Casetti, E. Roncoroni, M. Ciboddo, P. Benvenuti, B. Landini, E. Fugazza, D. Troletti, C. Astori, and M. Cazzola. Clinical course and outcome of essential thrombocythemia and prefibrotic myelofibrosis according to the revised WHO 2016 diagnostic criteria. Oncotarget, 8(60):101735–101744, Nov 2017.

[326] E. Rumi, A. S. Harutyunyan, D. Pietra, J. D. Feenstra, C. Cavalloni, E. Roncoroni, I. Casetti, M. Bellini, C. Milanesi, M. C. Renna, M. Gotti, C. Astori, R. Kralovics, and M. Cazzola. LNK mutations in familial myeloproliferative neoplasms. Blood, 128(1):144–145, 07 2016.

[327] M. A. Saez, J. Fernandez-Rodriguez, C. Moutinho, J. V. Sanchez-Mut, A. Gomez, E. Vidal, P. Petazzi, K. Szczesna, P. Lopez-Serra, M. Lucariello, P. Lorden, R. Delgado-Morales, O. J. de la Caridad, D. Huertas, J. L. Gelpi, M. Orozco, A. Lopez-Doriga, M. Mila, L. A. Perez-Jurado, M. Pineda, J. Armstrong, C. Lazaro, and M. Esteller. Mutations in JMJD1C are involved in Rett syndrome and intel- lectual disability. Genet. Med., 18(4):378–385, Apr 2016.

[328] H. K. Saini, S. Griffiths-Jones, and A. J. Enright. Genomic analysis ofhuman microRNA transcripts. Proc. Natl. Acad. Sci. U.S.A., 104(45):17719–17724, Nov 2007.

157 References

[329] J. Saliba, C. Saint-Martin, A. Di Stefano, G. Lenglet, C. Marty, B. Keren, F. Pasquier, V. D. Valle, L. Secardin, G. Leroy, E. Mahfoudhi, S. Grosjean, N. Droin, M. Diop, P. Dessen, S. Charrier, A. Palazzo, J. Merlevede, J. C. Meniane, C. Delaunay-Darivon, P. Fuseau, F. Isnard, N. Casadevall, E. Solary, N. Debili, O. A. Bernard, H. Raslova, A. Najman, W. Vainchenker, C. Bellanne-Chantelot, and I. Plo. Germline duplication of ATG2B and GSKIP predisposes to familial myeloid malignancies. Nat. Genet., 47(10):1131–1140, Oct 2015.

[330] M. L. Sandberg, S. E. Sutton, M. T. Pletcher, T. Wiltshire, L. M. Tarantino, J. B. Hogenesch, and M. P. Cooke. c-Myb and p300 regulate hematopoietic stem cell proliferation and differentiation. Dev. Cell, 8(2):153–166, Feb 2005.

[331] A. F. Sandes, M. V. Goncalves, and M. L. Chauffaille. Frequency of polycythemia in individuals with normal complete blood cell counts according to the new 2016 WHO classification of myeloid neoplasms. Int J Lab Hematol, 39(5):528–531, Oct 2017.

[332] A. Sanjuan-Pla, I. C. Macaulay, C. T. Jensen, P. S. Woll, T. C. Luis, A. Mead, S. Moore, C. Carella, S. Matsuoka, T. Bouriez Jones, O. Chowdhury, L. Stenson, M. Lutteropp, J. C. Green, R. Facchini, H. Boukarabila, A. Grover, A. Gambardella, S. Thongjuea, J. Carrelha, P. Tarrant, D. Atkinson, S. A. Clark, C. Nerlov, and S. E. Jacobsen. Platelet-biased stem cells reside at the apex of the haematopoietic stem-cell hierarchy. Nature, 502(7470):232–236, Oct 2013.

[333] R. S. Savkur, J. S. Thomas, K. S. Bramlett, Y. Gao, L. F. Michael, and T. P. Burris. Ligand-dependent coactivation of the human bile acid receptor FXR by the peroxisome proliferator-activated receptor gamma coactivator-1alpha. J. Pharmacol. Exp. Ther., 312(1):170–178, Jan 2005.

[334] M. S. Sayer, P. A. Tilbrook, A. Spadaccini, E. Ingley, M. K. Sarna, J. H. Williams, N. C. Andrews, and S. P. Klinken. Ectopic expression of transcription factor NF-E2 alters the phenotype of erythroid and monoblastoid cells. J. Biol. Chem., 275(33):25292–25298, Aug 2000.

[335] W. F. SCHERER, J. T. SYVERTON, and G. O. GEY. Studies on the propagation in vitro of poliomyelitis viruses. IV. Viral multiplication in a stable strain of human malignant epithelial cells (strain HeLa) derived from an epidermoid carcinoma of the cervix. J. Exp. Med., 97(5):695–710, May 1953.

158 References

[336] R. K. Schindhelm, E. ten Boekel, N. E. Heima, N. M. van Schoor, and S. Simsek. Thyroid hormones and erythrocyte indices in a cohort of euthyroid older subjects. Eur. J. Intern. Med., 24(3):241–244, Apr 2013.

[337] R. Schofield. The relationship between the spleen colony-forming cell andthe haemopoietic stem cell. Blood Cells, 4(1-2):7–25, 1978.

[338] L. M. Scott, W. Tong, R. L. Levine, M. A. Scott, P. A. Beer, M. R. Stratton, P. A. Futreal, W. N. Erber, M. F. McMullin, C. N. Harrison, A. J. Warren, D. G. Gilliland, H. F. Lodish, and A. R. Green. JAK2 exon 12 mutations in polycythemia vera and idiopathic erythrocytosis. N. Engl. J. Med., 356(5):459–468, Feb 2007.

[339] J. Seita and I. L. Weissman. Hematopoietic stem cell: self-renewal versus differentiation. Wiley Interdiscip Rev Syst Biol Med, 2(6):640–653, 2010.

[340] I. F. G. Sena, P. H. D. M. Prazeres, G. S. P. Santos, I. T. Borges, P. O. Azevedo, J. P. Andreotti, V. M. Almeida, A. E. Paiva, D. A. P. Guerra, L. Lousado, L. Souto, A. Mintz, and A. Birbrair. Identity of Gli1+ cells in the bone marrow. Exp. Hematol., 54:12–16, 10 2017.

[341] A. Shakya, C. Callister, A. Goren, N. Yosef, N. Garg, V. Khoddami, D. Nix, A. Regev, and D. Tantin. Pluripotency transcription factor Oct4 mediates stepwise nucleosome demethylation and depletion. Mol. Cell. Biol., 35(6):1014–1025, Mar 2015.

[342] X. H. Shen, N. N. Sun, Y. F. Yin, S. F. Liu, X. L. Liu, H. L. Peng, C. W. Dai, Y. X. Xu, M. Y. Deng, Y. Y. Luo, W. L. Zheng, and G. S. Zhang. A TET2 rs3733609 C/T genotype is associated with predisposition to the myeloproliferative neoplasms harboring JAK2(V617F) and confers a proliferative potential on erythroid lineages. Oncotarget, 7(8):9550–9560, Feb 2016.

[343] L. Shi, M. C. Sierant, K. Gurdziel, F. Zhu, S. Cui, K. E. Kolodziej, J. Strouboulis, Y. Guan, O. Tanabe, K. C. Lim, and J. D. Engel. Biased, non-equivalent gene- proximal and -distal binding motifs of orphan nuclear receptor TR4 in primary human erythroid cells. PLoS Genet., 10(5):e1004339, May 2014.

[344] Y. Shi, F. Lan, C. Matson, P. Mulligan, J. R. Whetstine, P. A. Cole, R. A. Casero, and Y. Shi. Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell, 119(7):941–953, Dec 2004.

159 References

[345] A. H. Shih, O. Abdel-Wahab, J. P. Patel, and R. L. Levine. The role of mutations in epigenetic regulators in myeloid malignancies. Nat. Rev. Cancer, 12(9):599–612, Sep 2012.

[346] L. Y. Shih, T. L. Lin, C. L. Lai, P. Dunn, J. H. Wu, P. N. Wang, M. C. Kuo, and L. C. Lee. Predictive values of X-chromosome inactivation patterns and clinicohematologic parameters for vascular complications in female patients with essential thrombocythemia. Blood, 100(5):1596–1601, Sep 2002.

[347] D. M. Shin, R. Liu, I. Klich, W. Wu, J. Ratajczak, M. Kucia, and M. Z. Ratajczak. Molecular signature of adult bone marrow-purified very small embryonic-like stem cells supports their developmental epiblast/germ line origin. Leukemia, 24(8):1450– 1461, Aug 2010.

[348] M. Shiraishi, Y. Yamamoto, N. Hirooka, Y. Obuchi, S. Tachibana, M. Makishima, and Y. Tanaka. A high concentration of triiodothyronine attenuates the stimulatory effect on hemin-induced erythroid differentiation of human erythroleukemia K562 cells. Endocr. J., 62(5):431–440, 2015.

[349] K. Shire, A. I. Wong, M. H. Tatham, O. F. Anderson, D. Ripsman, S. Gulstene, J. Moffat, R. T. Hay, and L. Frappier. Identification of RNF168 as a PML nuclear body regulator. J. Cell. Sci., 129(3):580–591, Feb 2016.

[350] R. A. Shivdasani and S. H. Orkin. Erythropoiesis and globin gene expression in mice lacking the transcription factor NF-E2. Proc. Natl. Acad. Sci. U.S.A., 92(19):8690–8694, Sep 1995.

[351] R. A. Shivdasani, M. F. Rosenblatt, D. Zucker-Franklin, C. W. Jackson, P. Hunt, C. J. Saris, and S. H. Orkin. Transcription factor NF-E2 is required for platelet formation independent of the actions of thrombopoietin/MGDF in megakaryocyte development. Cell, 81(5):695–704, Jun 1995.

[352] R. Shrestha, S. Giri, J. O. Armitage, and V. R. Bhatt. Leukemic transformation in patients with myeloproliferative neoplasms: a population-based retrospective study. Future Oncol, 13(14):1239–1246, Jun 2017.

[353] E. Shtivelman, B. Lifshitz, R. P. Gale, and E. Canaani. Fused transcript of abl and bcr genes in chronic myelogenous leukaemia. Nature, 315(6020):550–554, 1985.

160 References

[354] Harsimran Sidhu and Neena Capalash. Uhrf1: The key regulator of epigenetics and molecular target for cancer therapeutics. Tumor Biology, 39(2):1010428317692205, 2017.

[355] H. B. Sieburg, R. H. Cho, B. Dykstra, N. Uchida, C. J. Eaves, and C. E. Muller- Sieburg. The hematopoietic stem compartment consists of a limited number of discrete stem cell subsets. Blood, 107(6):2311–2316, Mar 2006.

[356] Z. Siegfried, S. Eden, M. Mendelsohn, X. Feng, B. Z. Tsuberi, and H. Cedar. DNA methylation represses transcription in vivo. Nat. Genet., 22(2):203–206, Jun 1999.

[357] O. Silvennoinen and S. R. Hubbard. Molecular insights into regulation of JAK2 in myeloproliferative neoplasms. Blood, 125(22):3388–3392, May 2015.

[358] R. T. Silver, K. Vandris, and J. J. Goldman. Recombinant interferon-α may retard progression of early primary myelofibrosis: a preliminary report. Blood, 117(24):6669–6672, Jun 2011.

[359] L. SIMINOVITCH, E. A. MCCULLOCH, and J. E. TILL. THE DISTRIBUTION OF COLONY-FORMING CELLS AMONG SPLEEN COLONIES. J Cell Comp Physiol, 62:327–336, Dec 1963.

[360] J. A. Simon and C. A. Lange. Roles of the EZH2 histone methyltransferase in cancer epigenetics. Mutat. Res., 647(1-2):21–29, Dec 2008.

[361] B. Snijder and L. Pelkmans. Origins of regulated cell-to-cell variability. Nat. Rev. Mol. Cell Biol., 12(2):119–125, 02 2011.

[362] D. Sollazzo, D. Forte, N. Polverelli, M. Perricone, M. Romano, S. Luatti, N. Vianelli, M. Cavo, F. Palandri, and L. Catani. Circulating Calreticulin Is Increased in Myelofibrosis: Correlation with Interleukin-6 Plasma Levels, Bone Marrow Fibrosis, and Splenomegaly. Mediators Inflamm., 2016:5860657, 2016.

[363] N. Soranzo, T. D. Spector, M. Mangino, B. Kuhnel, A. Rendon, A. Teumer, C. Willenborg, B. Wright, L. Chen, M. Li, P. Salo, B. F. Voight, P. Burns, R. A. Laskowski, Y. Xue, S. Menzel, D. Altshuler, J. R. Bradley, S. Bumpstead, M. S. Burnett, J. Devaney, A. Doring, R. Elosua, S. E. Epstein, W. Erber, M. Falchi, S. F. Garner, M. J. Ghori, A. H. Goodall, R. Gwilliam, H. H. Hakonarson, A. S. Hall, N. Hammond, C. Hengstenberg, T. Illig, I. R. Konig, C. W. Knouff, R. McPherson, O. Melander, V. Mooser, M. Nauck, M. S. Nieminen, C. J. O’Donnell, L. Peltonen,

161 References

S. C. Potter, H. Prokisch, D. J. Rader, C. M. Rice, R. Roberts, V. Salomaa, J. Sambrook, S. Schreiber, H. Schunkert, S. M. Schwartz, J. Serbanovic-Canic, J. Sinisalo, D. S. Siscovick, K. Stark, I. Surakka, J. Stephens, J. R. Thompson, U. Volker, H. Volzke, N. A. Watkins, G. A. Wells, H. E. Wichmann, D. A. Van Heel, C. Tyler-Smith, S. L. Thein, S. Kathiresan, M. Perola, M. P. Reilly, A. F. Stewart, J. Erdmann, N. J. Samani, C. Meisinger, A. Greinacher, P. Deloukas, W. H. Ouwehand, and C. Gieger. A genome-wide meta-analysis identifies 22 loci associated with eight hematological parameters in the HaemGen consortium. Nat. Genet., 41(11):1182–1190, Nov 2009.

[364] S. Sozer, M. I. Fiel, T. Schiano, M. Xu, J. Mascarenhas, and R. Hoffman. The presence of JAK2V617F mutation in the liver endothelial cells of patients with Budd-Chiari syndrome. Blood, 113(21):5246–5249, May 2009.

[365] G. J. Spangrude, S. Heimfeld, and I. L. Weissman. Purification and characterization of mouse hematopoietic stem cells. Science, 241(4861):58–62, Jul 1988.

[366] J. L. Spivak, M. Considine, D. M. Williams, C. C. Talbot, O. Rogers, A. R. Moliterno, C. Jie, and M. F. Ochs. Two clinical phenotypes in polycythemia vera. N. Engl. J. Med., 371(9):808–817, Aug 2014.

[367] P. Sroczynska, V. A. Cruickshank, J. P. Bukowski, S. Miyagi, F. O. Bagger, J. Walfridsson, M. B. Schuster, B. Porse, and K. Helin. shRNA screening identifies JMJD1C as being required for leukemia maintenance. Blood, 123(12):1870–1882, Mar 2014.

[368] B. L. Stein, J. Gotlib, M. Arcasoy, M. H. Nguyen, N. Shah, A. Moliterno, C. Jamieson, D. A. Pollyea, B. Scott, M. Wadleigh, R. Levine, R. Komrokji, R. Klisovic, K. Gundabolu, P. Kropf, M. Wetzler, S. T. Oh, R. Ribeiro, R. Paschal, S. Mohan, N. Podoltsev, J. Prchal, M. Talpaz, D. Snyder, S. Verstovsek, and R. A. Mesa. Historical views, conventional approaches, and evolving management strategies for myeloproliferative neoplasms. J Natl Compr Canc Netw, 13(4):424– 434, Apr 2015.

[369] B. L. Stein, D. M. Williams, N. Y. Wang, O. Rogers, M. A. Isaacs, N. Pemmaraju, J. L. Spivak, and A. R. Moliterno. Sex differences in the JAK2 V617F allele burden in chronic myeloproliferative disorders. Haematologica, 95(7):1090–1097, Jul 2010.

162 References

[370] E. M. Stein, C. D. DiNardo, D. A. Pollyea, A. T. Fathi, G. J. Roboz, J. K. Altman, R. M. Stone, D. J. DeAngelo, R. L. Levine, I. W. Flinn, H. M. Kantarjian, R. Collins, M. R. Patel, A. E. Frankel, A. Stein, M. A. Sekeres, R. T. Swords, B. C. Medeiros, C. Willekens, P. Vyas, A. Tosolini, Q. Xu, R. D. Knight, K. E. Yen, S. Agresta, S. de Botton, and M. S. Tallman. Enasidenib in mutant IDH2 relapsed or refractory acute myeloid leukemia. Blood, 130(6):722–731, 08 2017.

[371] H. K. Steinman and M. W. Greaves. Aquagenic pruritus. J. Am. Acad. Dermatol., 13(1):91–96, Jul 1985.

[372] Marvin J Stone. Polycythaemia vera: Osler—vaquez disease. Journal of Medical Biography, 9(2):99–103, 2001. PMID: 11304636.

[373] B. D. Strahl and C. D. Allis. The language of covalent histone modifications. Nature, 403(6765):41–45, Jan 2000.

[374] I. H. Su, A. Basavaraj, A. N. Krutchinsky, O. Hobert, A. Ullrich, B. T. Chait, and A. Tarakhovsky. Ezh2 controls B cell development through histone H3 methylation and Igh rearrangement. Nat. Immunol., 4(2):124–131, Feb 2003.

[375] A. Sud, S. Chattopadhyay, H. Thomsen, K. Sundquist, J. Sundquist, R. S. Houlston, and K. Hemminki. Familial risks of primary myeloid leukemia, myelodysplasia and myeloproliferative neoplasms. Blood, Jul 2018.

[376] K. Sudo, H. Ema, Y. Morita, and H. Nakauchi. Age-associated characteristics of murine hematopoietic stem cells. J. Exp. Med., 192(9):1273–1280, Nov 2000.

[377] J. Sun, A. Ramos, B. Chapman, J. B. Johnnidis, L. Le, Y. J. Ho, A. Klein, O. Hofmann, and F. D. Camargo. Clonal dynamics of native haematopoiesis. Nature, 514(7522):322–327, Oct 2014.

[378] X. J. Sun, Z. Wang, L. Wang, Y. Jiang, N. Kost, T. D. Soong, W. Y. Chen, Z. Tang, T. Nakadai, O. Elemento, W. Fischle, A. Melnick, D. J. Patel, S. D. Nimer, and R. G. Roeder. A stable transcription factor complex nucleated by oligomeric AML1-ETO controls leukaemogenesis. Nature, 500(7460):93–97, Aug 2013.

[379] G. Swiers, R. Patient, and M. Loose. Genetic regulatory networks programming hematopoietic stem cells and erythroid lineage specification. Dev. Biol., 294(2):525– 540, Jun 2006.

163 References

[380] G. Swiers, C. Rode, E. Azzoni, and M. F. de Bruijn. A short history of hemogenic endothelium. Blood Cells Mol. Dis., 51(4):206–212, Dec 2013.

[381] S. M. Sykes and D. T. Scadden. Modeling human hematopoietic stem cell biology in the mouse. Semin. Hematol., 50(2):92–100, Apr 2013.

[382] M. F. Taha, A. Javeri, S. Rohban, and S. J. Mowla. Upregulation of pluripotency markers in adipose tissue-derived stem cells by miR-302 and leukemia inhibitory factor. Biomed Res Int, 2014:941486, 2014.

[383] M. Tahiliani, K. P. Koh, Y. Shen, W. A. Pastor, H. Bandukwala, Y. Brudno, S. Agarwal, L. M. Iyer, D. R. Liu, L. Aravind, and A. Rao. Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science, 324(5929):930–935, May 2009.

[384] T. Takeuchi, Y. Yamazaki, Y. Katoh-Fukui, R. Tsuchiya, S. Kondo, J. Motoyama, and T. Higashinakagawa. Gene trap capture of a novel mouse gene, jumonji, required for neural tube formation. Genes Dev., 9(10):1211–1222, May 1995.

[385] O. Tanabe, Y. Shen, Q. Liu, A. D. Campbell, T. Kuroha, M. Yamamoto, and J. D. Engel. The TR2 and TR4 orphan nuclear receptors repress Gata1 transcription. Genes Dev., 21(21):2832–2844, Nov 2007.

[386] F. J. Tavner, R. Simpson, S. Tashiro, D. Favier, N. A. Jenkins, D. J. Gilbert, N. G. Copeland, E. M. Macmillan, J. Lutwyche, R. A. Keough, S. Ishii, and T. J. Gonda. Molecular cloning reveals that the p160 Myb-binding protein is a novel, predominantly nucleolar protein which may play a role in transactivation by Myb. Mol. Cell. Biol., 18(2):989–1002, Feb 1998.

[387] A. Tefferi. The history of myeloproliferative disorders: before and after Dameshek. Leukemia, 22(1):3–13, Jan 2008.

[388] A. Tefferi. Primary myelofibrosis: 2019 update on diagnosis, risk-stratification and management. Am. J. Hematol., Jul 2018.

[389] A. Tefferi, S. Betti, D. Barraco, M. Mudireddy, S. Shah, C. A. Hanson, R.P.Ket- terling, A. Pardanani, N. Gangat, G. Coltro, P. Guglielmelli, and A. M. Vannucchi. Gender and survival in essential thrombocythemia: A two-center study of 1,494 patients. Am. J. Hematol., 92(11):1193–1197, Nov 2017.

164 References

[390] A. Tefferi, P. Guglielmelli, D. R. Larson, C. Finke, E. A. Wassie, L. Pieri, N. Gangat, R. Fjerza, A. A. Belachew, T. L. Lasho, R. P. Ketterling, C. A. Hanson, A. Rambaldi, G. Finazzi, J. Thiele, T. Barbui, A. Pardanani, and A. M. Vannucchi. Long-term survival and blast transformation in molecularly annotated essential thrombocythemia, polycythemia vera, and myelofibrosis. Blood, 124(16):2507–2513, Oct 2014.

[391] A. Tefferi, P. Guglielmelli, T. L. Lasho, N. Gangat, R. P. Ketterling, A. Pardanani, and A. M. Vannucchi. MIPSS70+ Version 2.0: Mutation and Karyotype-Enhanced International Prognostic Scoring System for Primary Myelofibrosis. J. Clin. Oncol., 36(17):1769–1770, Jun 2018.

[392] A. Tefferi, T. L. Lasho, O. Abdel-Wahab, P. Guglielmelli, J. Patel, D. Caramazza, L. Pieri, C. M. Finke, O. Kilpivaara, M. Wadleigh, M. Mai, R. F. McClure, D. G. Gilliland, R. L. Levine, A. Pardanani, and A. M. Vannucchi. IDH1 and IDH2 mutation studies in 1473 patients with chronic-, fibrotic- or blast-phase essential thrombocythemia, polycythemia vera or myelofibrosis. Leukemia, 24(7):1302–1309, Jul 2010.

[393] A. Tefferi, T. L. Lasho, C. Finke, A. A. Belachew, E. A. Wassie, R. P. Ketterling, C. A. Hanson, and A. Pardanani. Type 1 vs type 2 calreticulin mutations in primary myelofibrosis: differences in phenotype and prognostic impact. Leukemia, 28(7):1568–1570, Jul 2014.

[394] A. Tefferi, A. Pardanani, K. H. Lim, O. Abdel-Wahab, T. L. Lasho, J.Patel, N. Gangat, C. M. Finke, S. Schwager, A. Mullally, C. Y. Li, C. A. Hanson, R. Mesa, O. Bernard, F. Delhommeau, W. Vainchenker, D. G. Gilliland, and R. L. Levine. TET2 mutations and their clinical correlates in polycythemia vera, essential thrombocythemia and myelofibrosis. Leukemia, 23(5):905–911, May 2009.

[395] A. Tefferi, R. Vaidya, D. Caramazza, C. Finke, T. Lasho, and A. Pardanani. Circulating interleukin (IL)-8, IL-2R, IL-12, and IL-15 levels are independently prognostic in primary myelofibrosis: a comprehensive cytokine profiling study. J. Clin. Oncol., 29(10):1356–1363, Apr 2011.

[396] A. Tefferi, A. M. Vannucchi, and T. Barbui. Essential thrombocythemia treatment algorithm 2018. Blood Cancer J, 8(1):2, 01 2018.

165 References

[397] A. Tefferi, E. A. Wassie, P. Guglielmelli, N. Gangat, A. A. Belachew, T.L. Lasho, C. Finke, R. P. Ketterling, C. A. Hanson, A. Pardanani, A. P. Wolanskyj, M. Maffioli, R. Casalone, A. Pacilli, A. M. Vannucchi, and F. Passamonti. Type1 versus Type 2 calreticulin mutations in essential thrombocythemia: a collaborative study of 1027 patients. Am. J. Hematol., 89(8):E121–124, Aug 2014.

[398] Tefferi Ayalew, Guglielmelli Paola, Nicolosi Maura, Mannelli Francesco, Mudireddy Mythri, Bartalucci Niccolo, Finke Christy M., Lasho Terra L., Hanson Curtis A., Ketterling Rhett P., Begna Kebede H., Naseema Gangat, Pardanani Animesh, and Vannucchi Alessandro M. GIPSS: genetically inspired prognostic scoring system for primary myelofibrosis. Leukemia, 2018.

[399] J. Thiele and H. M. Kvasnicka. Hematopathologic findings in chronic idiopathic myelofibrosis. Semin. Oncol., 32(4):380–394, Aug 2005.

[400] J. Thiele, H. M. Kvasnicka, L. Mullauer, V. Buxhofer-Ausch, B. Gisslinger, and H. Gisslinger. Essential thrombocythemia versus early primary myelofibrosis: a multicenter study to validate the WHO classification. Blood, 117(21):5710–5718, May 2011.

[401] G. Thorleifsson, G. B. Walters, D. F. Gudbjartsson, V. Steinthorsdottir, P. Sulem, A. Helgadottir, U. Styrkarsdottir, S. Gretarsdottir, S. Thorlacius, I. Jonsdottir, T. Jonsdottir, E. J. Olafsdottir, G. H. Olafsdottir, T. Jonsson, F. Jonsson, K. Borch- Johnsen, T. Hansen, G. Andersen, T. Jorgensen, T. Lauritzen, K. K. Aben, A. L. Verbeek, N. Roeleveld, E. Kampman, L. R. Yanek, L. C. Becker, L. Tryggvadottir, T. Rafnar, D. M. Becker, J. Gulcher, L. A. Kiemeney, O. Pedersen, A. Kong, U. Thorsteinsdottir, and K. Stefansson. Genome-wide association yields new sequence variants at seven loci that associate with measures of obesity. Nat. Genet., 41(1):18–24, Jan 2009.

[402] R. Tiedt, H. Hao-Shen, M. A. Sobas, R. Looser, S. Dirnhofer, J. Schwaller, and R. C. Skoda. Ratio of mutant JAK2-V617F to wild-type Jak2 determines the MPD phenotypes in transgenic mice. Blood, 111(8):3931–3940, Apr 2008.

[403] A. P. Trifa, C. Banescu, M. Tevet, A. Bojan, D. Dima, L. Urian, T. Torok-Vistai, V. M. Popov, M. Zdrenghea, L. Petrov, A. Vasilache, M. Murat, D. Georgescu, M. Popescu, O. Patrinoiu, M. Balea, R. Costache, E. Coles, C. Saguna, N. Berbec, A. M. Vladareanu, R. G. Mihaila, H. Bumbea, A. Cucuianu, and R. A. Popp. TERT rs2736100 A>C SNP and JAK2 46/1 haplotype significantly contribute to

166 References

the occurrence of JAK2 V617F and CALR mutated myeloproliferative neoplasms - a multicentric study on 529 patients. Br. J. Haematol., 174(2):218–226, 07 2016.

[404] N. Uchida, R. Green, J. Ballantine, L. P. Skala, M. M. Hsieh, and J. F. Tisdale. Kinetics of lentiviral vector transduction in human CD34(+) cells. Exp. Hematol., 44(2):106–115, Feb 2016.

[405] S. Uttarkar, E. Dasse, A. Coulibaly, S. Steinmann, A. Jakobs, C. Schomburg, A. Trentmann, J. Jose, P. Schlenke, W. E. Berdel, T. J. Schmidt, C. Muller-Tidow, J. Frampton, and K. H. Klempnauer. Targeting acute myeloid leukemia with a small molecule inhibitor of the Myb/p300 interaction. Blood, 127(9):1173–1182, Mar 2016.

[406] R. Vaidya, N. Gangat, T. Jimma, C. M. Finke, T. L. Lasho, A. Pardanani, and A. Tefferi. Plasma cytokines in polycythemia vera: phenotypic correlates, prognostic relevance, and comparison with myelofibrosis. Am. J. Hematol., 87(11):1003–1005, Nov 2012.

[407] A. M. Vannucchi, E. Antonioli, P. Guglielmelli, A. Pancrazzi, V. Guerini, G. Barosi, M. Ruggeri, G. Specchia, F. Lo-Coco, F. Delaini, L. Villani, S. Finotto, E. Am- matuna, R. Alterini, V. Carrai, G. Capaccioli, S. Di Lollo, V. Liso, A. Rambaldi, A. Bosi, and T. Barbui. Characteristics and clinical correlates of MPL 515W>L/K mutation in essential thrombocythemia. Blood, 112(3):844–847, Aug 2008.

[408] A. M. Vannucchi, S. Verstovsek, P. Guglielmelli, M. Griesshammer, T. C. Burn, A. Naim, D. Paranagama, M. Marker, B. Gadbaw, and J. J. Kiladjian. Ruxolitinib reduces JAK2 p.V617F allele burden in patients with polycythemia vera enrolled in the RESPONSE study. Ann. Hematol., 96(7):1113–1120, Jul 2017.

[409] Henri Vaquez et al. Sur une forme spéciale de cyanose s’ accompagnant d’hyperglobulie excessive et persistante. CR Soc Biol (Paris), 44:384–388, 1892.

[410] S. Varambally, Q. Cao, R. S. Mani, S. Shankar, X. Wang, B. Ateeq, B. Laxman, X. Cao, X. Jing, K. Ramnarayanan, J. C. Brenner, J. Yu, J. H. Kim, B. Han, P. Tan, C. Kumar-Sinha, R. J. Lonigro, N. Palanisamy, C. A. Maher, and A. M. Chinnaiyan. Genomic loss of microRNA-101 leads to overexpression of histone methyltransferase EZH2 in cancer. Science, 322(5908):1695–1699, Dec 2008.

[411] J. W. Vardiman, J. Thiele, D. A. Arber, R. D. Brunning, M. J. Borowitz, A. Porwit, N. L. Harris, M. M. Le Beau, E. Hellstrom-Lindberg, A. Tefferi, and

167 References

C. D. Bloomfield. The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: rationale and important changes. Blood, 114(5):937–951, Jul 2009.

[412] S. Verma, D. Lewis, G. Warne, and M. Grossmann. An X-traordinary stroke. Lancet, 377(9773):1288, Apr 2011.

[413] S. Verstovsek, H. Kantarjian, R. A. Mesa, A. D. Pardanani, J. Cortes-Franco, D. A. Thomas, Z. Estrov, J. S. Fridman, E. C. Bradley, S. Erickson-Viitanen, K. Vaddi, R. Levy, and A. Tefferi. Safety and efficacy of INCB018424, aJAK1and JAK2 inhibitor, in myelofibrosis. N. Engl. J. Med., 363(12):1117–1127, Sep 2010.

[414] S. Verstovsek, R. A. Mesa, J. Gotlib, V. Gupta, J. F. DiPersio, J. V. Catalano, M. W. Deininger, C. B. Miller, R. T. Silver, M. Talpaz, E. F. Winton, J. H. Harvey, M. O. Arcasoy, E. O. Hexner, R. M. Lyons, R. Paquette, A. Raza, M. Jones, D. Kornacki, K. Sun, and H. Kantarjian. Long-term treatment with ruxolitinib for patients with myelofibrosis: 5-year update from the randomized, double-blind, placebo-controlled, phase 3 COMFORT-I trial. J Hematol Oncol, 10(1):55, Feb 2017.

[415] C. H. Waddington and C. H. Waddington. The epigenotype. 1942. Int J Epidemiol, 41(1):10–13, Feb 2012.

[416] I. A. Wadman, H. Osada, G. G. Grutz, A. D. Agulnick, H. Westphal, A. Forster, and T. H. Rabbitts. The LIM-only protein Lmo2 is a bridging molecule assembling an erythroid, DNA-binding complex which includes the TAL1, E47, GATA-1 and Ldb1/NLI proteins. EMBO J., 16(11):3145–3157, Jun 1997.

[417] C. Wang, M. Oshima, G. Sashida, T. Tomioka, N. Hasegawa, M. Mochizuki- Kashio, Y. Nakajima-Takagi, Y. Kusunoki, S. Kyoizumi, K. Imai, K. Nakachi, and A. Iwama. Non-Lethal Ionizing Radiation Promotes Aging-Like Phenotypic Changes of Human Hematopoietic Stem and Progenitor Cells in Humanized Mice. PLoS ONE, 10(7):e0132041, 2015.

[418] J. Wang, J. W. Park, H. Drissi, X. Wang, and R. H. Xu. Epigenetic regulation of miR-302 by JMJD1C inhibits neural differentiation of human embryonic stem cells. J. Biol. Chem., 289(4):2384–2395, Jan 2014.

[419] J. C. Wang and J. E. Dick. Cancer stem cells: lessons from leukemia. Trends Cell Biol., 15(9):494–501, Sep 2005.

168 References

[420] L. Wang, S. Yamaguchi, M. D. Burstein, K. Terashima, K. Chang, H. K. Ng, H. Nakamura, Z. He, H. Doddapaneni, L. Lewis, M. Wang, T. Suzuki, R. Nishikawa, A. Natsume, S. Terasaka, R. Dauser, W. Whitehead, A. Adekunle, J. Sun, Y. Qiao, G. Marth, D. M. Muzny, R. A. Gibbs, S. M. Leal, D. A. Wheeler, and C. C. Lau. Novel somatic and germline mutations in intracranial germ cell tumours. Nature, 511(7508):241–245, Jul 2014.

[421] W. Wang, S. Schwemmers, E. O. Hexner, and H. L. Pahl. AML1 is overexpressed in patients with myeloproliferative neoplasms and mediates JAK2V617F-independent overexpression of NF-E2. Blood, 116(2):254–266, Jul 2010.

[422] S. Watanabe, K. Watanabe, V. Akimov, J. Bartkova, B. Blagoev, J. Lukas, and J. Bartek. JMJD1C demethylates MDC1 to regulate the RNF8 and BRCA1- mediated chromatin response to DNA breaks. Nat. Struct. Mol. Biol., 20(12):1425– 1433, Dec 2013.

[423] K. Weber, U. Bartsch, C. Stocking, and B. Fehse. A multicolor panel of novel lentiviral "gene ontology" (LeGO) vectors for functional gene analysis. Mol. Ther., 16(4):698–706, Apr 2008.

[424] J. Wehrle, T. S. Seeger, S. Schwemmers, D. Pfeifer, A. Bulashevska, and H. L. Pahl. Transcription factor nuclear factor erythroid-2 mediates expression of the cytokine interleukin 8, a known predictor of inferior outcome in patients with myeloproliferative neoplasms. Haematologica, 98(7):1073–1080, Jul 2013.

[425] B. Weinhold. Epigenetics: the science of change. Environ. Health Perspect., 114(3):A160–167, Mar 2006.

[426] Z. L. Whichard, C. A. Sarkar, M. Kimmel, and S. J. Corey. Hematopoiesis and its disorders: a systems biology approach. Blood, 115(12):2339–2347, Mar 2010.

[427] K. Wille, P. Sadjadian, and M. Griesshammer. [Thrombocytosis and thrombocytopenia - background and clinical relevance]. Dtsch. Med. Wochenschr., 142(23):1732–1743, Nov 2017.

[428] C. J. Willer, E. K. Speliotes, R. J. Loos, S. Li, C. M. Lindgren, I. M. Heid, S. I. Berndt, A. L. Elliott, A. U. Jackson, C. Lamina, G. Lettre, N. Lim, H. N. Lyon, S. A. McCarroll, K. Papadakis, L. Qi, J. C. Randall, R. M. Roccasecca, S. Sanna, P. Scheet, M. N. Weedon, E. Wheeler, J. H. Zhao, L. C. Jacobs, I. Prokopenko, N. Soranzo, T. Tanaka, N. J. Timpson, P. Almgren, A. Bennett,

169 References

R. N. Bergman, S. A. Bingham, L. L. Bonnycastle, M. Brown, N. P. Burtt, P. Chines, L. Coin, F. S. Collins, J. M. Connell, C. Cooper, G. D. Smith, E. M. Dennison, P. Deodhar, P. Elliott, M. R. Erdos, K. Estrada, D. M. Evans, L. Gianniny, C. Gieger, C. J. Gillson, C. Guiducci, R. Hackett, D. Hadley, A. S. Hall, A. S. Havulinna, J. Hebebrand, A. Hofman, B. Isomaa, K. B. Jacobs, T. Johnson, P. Jousilahti, Z. Jovanovic, K. T. Khaw, P. Kraft, M. Kuokkanen, J. Kuusisto, J. Laitinen, E. G. Lakatta, J. Luan, R. N. Luben, M. Mangino, W. L. McArdle, T. Meitinger, A. Mulas, P. B. Munroe, N. Narisu, A. R. Ness, K. Northstone, S. O’Rahilly, C. Purmann, M. G. Rees, M. Ridderstrale, S. M. Ring, F. Rivadeneira, A. Ruokonen, M. S. Sandhu, J. Saramies, L. J. Scott, A. Scuteri, K. Silander, M. A. Sims, K. Song, J. Stephens, S. Stevens, H. M. Stringham, Y. C. Tung, T. T. Valle, C. M. Van Duijn, K. S. Vimaleswaran, P. Vollenweider, G. Waeber, C. Wallace, R. M. Watanabe, D. M. Waterworth, N. Watkins, J. C. Witteman, E. Zeggini, G. Zhai, M. C. Zillikens, D. Altshuler, M. J. Caulfield, S. J. Chanock, I. S. Farooqi, L. Ferrucci, J. M. Guralnik, A. T. Hattersley, F. B. Hu, M. R. Jarvelin, M. Laakso, V. Mooser, K. K. Ong, W. H. Ouwehand, V. Salomaa, N. J. Samani, T. D. Spector, T. Tuomi, J. Tuomilehto, M. Uda, A. G. Uitterlinden, N. J. Wareham, P. Deloukas, T. M. Frayling, L. C. Groop, R. B. Hayes, D. J. Hunter, K. L. Mohlke, L. Peltonen, D. Schlessinger, D. P. Strachan, H. E. Wichmann, M. I. McCarthy, M. Boehnke, I. Barroso, G. R. Abecasis, and J. N. Hirschhorn. Six new loci associated with body mass index highlight a neuronal influence on body weight regulation. Nat. Genet., 41(1):25–34, Jan 2009.

[429] K. Willert, J. D. Brown, E. Danenberg, A. W. Duncan, I. L. Weissman, T. Reya, J. R. Yates, and R. Nusse. Wnt proteins are lipid-modified and can act as stem cell growth factors. Nature, 423(6938):448–452, May 2003.

[430] S. T. Williams, L. J. Walport, R. J. Hopkinson, S. K. Madden, R. Chowdhury, C. J. Schofield, and A. Kawamura. Studies on the catalytic domains of multiple JmjC oxygenases using peptide substrates. Epigenetics, 9(12):1596–1603, Dec 2014.

[431] H. G. Williams-Ashman and A. H. Reddi. Actions of vertebrate sex hormones. Annu. Rev. Physiol., 33:31–82, 1971.

[432] A. Wilson, E. Laurenti, G. Oser, R. C. van der Wath, W. Blanco-Bose, M. Jaworski, S. Offner, C. F. Dunant, L. Eshkind, E. Bockamp, P. Lio, H. R. Macdonald, and A. Trumpp. Hematopoietic stem cells reversibly switch from dormancy to self- renewal during homeostasis and repair. Cell, 135(6):1118–1129, Dec 2008.

170 References

[433] N. K. Wilson, S. D. Foster, X. Wang, K. Knezevic, J. Schutte, P. Kaimakis, P. M. Chilarska, S. Kinston, W. H. Ouwehand, E. Dzierzak, J. E. Pimanda, M. F. de Bruijn, and B. Gottgens. Combinatorial transcriptional control in blood stem/progenitor cells: genome-wide analysis of ten major transcriptional regulators. Cell Stem Cell, 7(4):532–544, Oct 2010.

[434] S. S. Wolf, V. K. Patchev, and M. Obendorf. A novel variant of the putative demethylase gene, s-JMJD1C, is a coactivator of the AR. Arch. Biochem. Biophys., 460(1):56–66, Apr 2007.

[435] L. Wu, J. Fan, and J. G. Belasco. MicroRNAs direct rapid deadenylation of mRNA. Proc. Natl. Acad. Sci. U.S.A., 103(11):4034–4039, Mar 2006.

[436] W. Xu, H. Yang, Y. Liu, Y. Yang, P. Wang, S. H. Kim, S. Ito, C. Yang, P. Wang, M. T. Xiao, L. X. Liu, W. Q. Jiang, J. Liu, J. Y. Zhang, B. Wang, S. Frye, Y. Zhang, Y. H. Xu, Q. Y. Lei, K. L. Guan, S. M. Zhao, and Y. Xiong. Oncometabolite 2-hydroxyglutarate is a competitive inhibitor of α-ketoglutarate- dependent dioxygenases. Cancer Cell, 19(1):17–30, Jan 2011.

[437] Z. Xu, R. P. Gale, Y. Zhang, T. Qin, H. Chen, P. Zhang, T. Zhang, L. Liu, S. Qu, and Z. Xiao. Unique features of primary myelofibrosis in Chinese. Blood, 119(11):2469–2473, Mar 2012.

[438] R. Yamamoto, Y. Morita, J. Ooehara, S. Hamanaka, M. Onodera, K. L. Rudolph, H. Ema, and H. Nakauchi. Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells. Cell, 154(5):1112– 1126, Aug 2013.

[439] D. Yan, R. E. Hutchison, and G. Mohi. Critical requirement for Stat5 in a mouse model of polycythemia vera. Blood, 119(15):3539–3549, Apr 2012.

[440] Y. Yang, H. Akada, D. Nath, R. E. Hutchison, and G. Mohi. Loss of Ezh2 cooperates with Jak2V617F in the development of myelofibrosis in a mouse model of myeloproliferative neoplasm. Blood, 127(26):3410–3423, 06 2016.

[441] J. Y. Yin, Q. Tang, L. L. Zhai, L. Y. Zhou, J. Qian, J. Lin, X. M. Wen, J. D. Zhou, Y. Y. Zhang, X. W. Zhu, and Z. Q. Deng. High expression of OCT4 is frequent and may cause undesirable treatment outcomes in patients with acute myeloid leukemia. Tumour Biol., 36(12):9711–9716, Dec 2015.

171 References

[442] F. Yue, Y. Cheng, A. Breschi, J. Vierstra, W. Wu, T. Ryba, R. Sandstrom, Z. Ma, C. Davis, B. D. Pope, Y. Shen, D. D. Pervouchine, S. Djebali, R. E. Thurman, R. Kaul, E. Rynes, A. Kirilusha, G. K. Marinov, B. A. Williams, D. Trout, H. Amrhein, K. Fisher-Aylor, I. Antoshechkin, G. DeSalvo, L. H. See, M. Fastuca, J. Drenkow, C. Zaleski, A. Dobin, P. Prieto, J. Lagarde, G. Bussotti, A. Tanzer, O. Denas, K. Li, M. A. Bender, M. Zhang, R. Byron, M. T. Groudine, D. McCleary, L. Pham, Z. Ye, S. Kuan, L. Edsall, Y. C. Wu, M. D. Rasmussen, M. S. Bansal, M. Kellis, C. A. Keller, C. S. Morrissey, T. Mishra, D. Jain, N. Dogan, R. S. Harris, P. Cayting, T. Kawli, A. P. Boyle, G. Euskirchen, A. Kundaje, S. Lin, Y. Lin, C. Jansen, V. S. Malladi, M. S. Cline, D. T. Erickson, V. M. Kirkup, K. Learned, C. A. Sloan, K. R. Rosenbloom, B. Lacerda de Sousa, K. Beal, M. Pignatelli, P. Flicek, J. Lian, T. Kahveci, D. Lee, W. J. Kent, M. Ramalho Santos, J. Herrero, C. Notredame, A. Johnson, S. Vong, K. Lee, D. Bates, F. Neri, M. Diegel, T. Canfield, P. J. Sabo, M. S. Wilken, T. A. Reh, E. Giste, A. Shafer, T. Kutyavin, E. Haugen, D. Dunn, A. P. Reynolds, S. Neph, R. Humbert, R. S. Hansen, M. De Bruijn, L. Selleri, A. Rudensky, S. Josefowicz, R. Samstein, E. E. Eichler, S. H. Orkin, D. Levasseur, T. Papayannopoulou, K. H. Chang, A. Skoultchi, S. Gosh, C. Disteche, P. Treuting, Y. Wang, M. J. Weiss, G. A. Blobel, X. Cao, S. Zhong, T. Wang, P. J. Good, R. F. Lowdon, L. B. Adams, X. Q. Zhou, M. J. Pazin, E. A. Feingold, B. Wold, J. Taylor, A. Mortazavi, S. M. Weissman, J. A. Stamatoyannopoulos, M. P. Snyder, R. Guigo, T. R. Gingeras, D. M. Gilbert, R. C. Hardison, M. A. Beer, and B. Ren. A comparative encyclopedia of DNA elements in the mouse genome. Nature, 515(7527):355–364, Nov 2014.

[443] P. D. Zamore, T. Tuschl, P. A. Sharp, and D. P. Bartel. RNAi: double-stranded RNA directs the ATP-dependent cleavage of mRNA at 21 to 23 nucleotide intervals. Cell, 101(1):25–33, Mar 2000.

[444] M. Zbucka-Kretowska, A. Eljaszewicz, D. Lipinska, K. Grubczak, M. Rusak, G. Mru- gacz, M. Dabrowska, M. Z. Ratajczak, and M. Moniuszko. Effective Mobilization of Very Small Embryonic-Like Stem Cells and Hematopoietic Stem/Progenitor Cells but Not Endothelial Progenitor Cells by Follicle-Stimulating Hormone Therapy. Stem Cells Int, 2016:8530207, 2016.

[445] Y. Zeng. Principles of micro-RNA production and maturation. Oncogene, 25(46):6156–6162, Oct 2006.

172 References

[446] G. Zhang and S. Pradhan. Mammalian epigenetic mechanisms. IUBMB Life, 66(4):240–256, Apr 2014.

[447] Q. Zhang, K. Zhao, Q. Shen, Y. Han, Y. Gu, X. Li, D. Zhao, Y. Liu, C. Wang, X. Zhang, X. Su, J. Liu, W. Ge, R. L. Levine, N. Li, and X. Cao. Tet2 is required to resolve inflammation by recruiting Hdac2 to specifically repress IL-6. Nature, 525(7569):389–393, Sep 2015.

[448] J. L. Zhao, C. Ma, R. M. O’Connell, A. Mehta, R. DiLoreto, J. R. Heath, and D. Baltimore. Conversion of danger signals into cytokine signals by hematopoietic stem and progenitor cells for regulation of stress-induced hematopoiesis. Cell Stem Cell, 14(4):445–459, Apr 2014.

[449] L. Zhao, H. Dong, C. C. Zhang, L. Kinch, M. Osawa, M. Iacovino, N. V. Grishin, M. Kyba, and L. J. Huang. A JAK2 interdomain linker relays Epo receptor engagement signals to kinase activation. J. Biol. Chem., 284(39):26988–26998, Sep 2009.

[450] Z. Zhou, X. Li, C. Deng, P. A. Ney, S. Huang, and J. Bungert. USF and NF-E2 cooperate to regulate the recruitment and activity of RNA polymerase II in the beta-globin gene locus. J. Biol. Chem., 285(21):15894–15905, May 2010.

[451] N. Zhu, M. Chen, R. Eng, J. DeJong, A. U. Sinha, N. F. Rahnamay, R. Koche, F. Al-Shahrour, J. C. Minehart, C. W. Chen, A. J. Deshpande, H. Xu, S. H. Chu, B. L. Ebert, R. G. Roeder, and S. A. Armstrong. MLL-AF9- and HOXA9- mediated acute myeloid leukemia stem cell self-renewal requires JMJD1C. J. Clin. Invest., 126(3):997–1011, Mar 2016.

[452] Vanessa Zimmermann. Identification of novel nf-e2 interaction partners using silac-based quantitative immunoprecipitation. Molecular medicine, Albert-Ludwigs- Universität, Freiburg, Germany, 2016.

[453] R. Zufferey, J. E. Donello, D. Trono, and T. J. Hope. Woodchuck hepatitis virus posttranscriptional regulatory element enhances expression of transgenes delivered by retroviral vectors. J. Virol., 73(4):2886–2892, Apr 1999.

173 Publication

Peeken, J. C., Jutzi, J. S., Wehrle, J., Koellerer, C., Staehle, H. F., Becker, H., Schoenwandt, E., Seeger, T. S., Schanne, D. H., Gothwal, M., Ott, C. J., Gründer, A., Pahl, H. L. (2018). Epigenetic regulation of NFE2 overexpression in myeloproliferative neoplasms. Blood, 131(18), 2065-2073

174 Curriculum vitae

According to the faculty’s rules and regulations this page was mandated to be included in the issues submitted to the examination committee. As a means of personal data integrity the author opted to void it in the final version for publication.

175 Affidavit

I declare on oath that I completed this work on my own and that information which has been directly or indirectly taken from other sources has been noted as such. Neither this, nor a similar work, has been published or presented to an examination committee.

Freiburg, August 19th 2019

Elias Schoenwandt

176 Contribution

Prof. Heike L. Pahl provided the overarching idea to the MD thesis project described above. The experimental design and scientific guidance were provided by Prof. Heike L. Pahl and Dr. Jonas J. Jutzi jointly. The experiments and the analysis of the data produced was solely conducted by Elias J.L.P. Schoenwandt. The shRNAs described above were intellectually conceived and physically generated by Elias J.L.P. Schoenwandt and represent his contribution to the publication Epigenetic regulation of NFE2 overexpression in myeloproliferative neoplasms. The experiments involving the shRNAs took place after Elias J.L.P. Schoenwandt had left the laboratory and were executed by Dr. Jonas J. Jutzi, representing his contribution to the aforementioned publication.

177 Acknowledgments

First and foremost I would like to extend my gratitude to Professor Heike L. Pahl and Jonas S. Jutzi, MD, PhD who offered me the opportunity to endeavor on this scientific journey and guided me on the way. Your support was invaluable.

I want to thank the MOTIVATE program of the Medical Faculty of the University of Freiburg that through the generous aid of the Else Kröner-Fresenius-Stiftung sup- ported me both pecuniarily as well as ideationally. It was a great growth opportunity that I hope many more generations of future physician-scientists may profit from.

Furthermore, I would like to thank all the members of the lab with whom I had the pleasure to work with. No matter how crammed the space was, you all contributed to a friendly atmosphere that made me want to the lab every morning. Yet none of you ever compromised on your rigorous work ethic, which spurred a positive and uplifting dynamism.

I want to thank the Kazusa DNA Research Institute for providing the pBC-SK+- mKIAA1380 vector, the laboratory of Makoto Tachibana and the laboratory of Jiri Bartek for providing anti-JMJD1C antibodies which they had used for their experiments. It is my firm belief that the ultimate goals of science will best be achieved in the spiritof cooperation and trust.

My sincere gratitude goes to my friends in Freiburg, the rest of Germany and the world. Not only did you counsel me on this manuscript, but your kind words, your warm embraces and cheerful laughter have helped me stay lighthearted and put a smile on my face. Emil, jag är otroligt glad över hur du har berikat mitt liv; jag tackar dig för din förståelse, din medkänsla och din närvaro.

Lastly, I want to thank my wonderful family, again both near and far, but of course formost my two loving parents. They paved my life’s path and through their continuous and unconditional support I am at the place where I am now, confidently looking into the future.

178