University of Groningen Biology of acute myeloid leukemia stem cells Mattes, Katharina

(1)

Biology of acute myeloid leukemia stem cells

Mattes, Katharina

DOI:

10.33612/diss.98637951

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date: 2019

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Mattes, K. (2019). Biology of acute myeloid leukemia stem cells: the role of CITED2 and mitochondrial activity. Rijksuniversiteit Groningen. https://doi.org/10.33612/diss.98637951

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

(2)

535611-L-sub01-bw-Mattes 535611-L-sub01-bw-Mattes 535611-L-sub01-bw-Mattes 535611-L-sub01-bw-Mattes Processed on: 23-9-2019 Processed on: 23-9-2019 Processed on: 23-9-2019

Processed on: 23-9-2019 PDF page: 9PDF page: 9PDF page: 9PDF page: 9

1.

(3)

Processed on: 23-9-2019 PDF page: 10PDF page: 10PDF page: 10PDF page: 10 10

1 Hematopoiesis

In human adults, more than 300 billion blood cells are produced every day. Blood cells are quite short lived and therefore blood formation needs to be a continuous and complex coordinated process which ensures that a large amount of different blood cells can be replenished within a short period of time. The main location of blood formation shifts during human development: The first site of hematopoiesis is the yolk sac; in the later embryogenesis the liver is the main organ for blood production and in the adult body the majority of blood cells is generated in the bone marrow. Hematopoiesis results in the production of several types of mature blood cells which can be assigned to four different blood lineages: the erythroid lineage (erythrocytes), the megakaryocytic lineage (platelets), the lymphoid lineage (T cells, B cells, NK-cells) and the myeloid lineage (granulocytes, monocytes). The different types of cells fulfill distinct functions: Erythrocytes are the oxygen-transporters of the cells, platelets are important for blood clotting, lymphocytes are the key-players of the adaptive immune system and granulocytes and monocytes are crucial components of the innate immune response.

All the different blood cells arise from one type of cell: The hematopoietic stem cell (HSC). In blood, HSCs are the only type of cell that possess both self-renewal and multipotency - which means that by asymmetric cell division they can give rise to both another HSC and a daughter cell that is capable of differentiating to any of the distinct blood cell types. Therefore, HSCs are commonly visualized as the top of a hierarchical tree, whereas the

mature blood cells form the bottom. However, it is still not completely understood which roadmaps HSCs take on the route towards differentiating into fully functional mature blood cells. One of the first models for hematopoietic maturation proposed a gradual way of HSC-differentiation via oligopotent progenitor-intermediates that stepwise narrow down their lineage potential along the way. This model was based on the identification of a common

lymphoid progenitor (CLP)1_that _can

give rise to T-, B-, and NK-cells and a

common myeloid progenitor (CMP),2

that is capable of forming all myeloid cells by further differentiating into the granulocyte-macrophage progenitor (GMP) or megakaryocyte-erythrocyte

progenitor (MEP) (Figure 1A).

Remarkably, also very recent studies with modern techniques support this view of a tree-like hematopoietic system, however, mostly with less clear defined intermediates.3,4_{A recent study proposed} a gradual model of hematopoiesis where HSCs give rise to progenitor cells that are viewed as a “continuum of low-primed undifferentiated hematopoietic stem and progenitor cells” (CLOUD-HSPCs)5 (Figure 1B). Additionally, in contrast to these gradual concepts, there are also distinct models that completely question the existence of oligopotent progenitors in adult hematopoiesis and suggest that there are either multipotent HSCs or unipotent progenitors that have acquired already a defined lineage fate soon after emerging from HSCs (Figure 1C).6 Several recent studies demonstrated that at least the megakaryocytic lineage seems to directly evolve from HSCs without passing through a stage of an oligopotent progenitor (Figure 1C).7–10_{Therefore, it is currently under} debate if even HSCs themselves form a

(4)

1

homogenous cell population or if there is rather a pool of distinct HSCs that have already a bias towards a certain blood lineage. In addition, evidence exists that steady-state hematopoiesis in adulthood is not driven by the classical defined HSCs, but rather by a large

number of long-lived progenitors.11,12 According to this model, HSCs have particular important functions in rebuilding the hematopoietic system e.g. after transplantation, however, don’t play a leading part under steady state conditions.

Figure 1. Hematopoiesis Models. Multiple concepts exist which describe potential stages of hematopoietic development.In all models, formation of blood cells is maintained by a multipotent, self-renewing hematopoietic stem cell (HSC). (A) The classical model proposes a gradual differentiation of HSCs into mature blood cells via well-defined oligopotent

progenitors. MPP: multipotent progenitor; CMP: common myeloid progenitor; MEP: megakaryocyte-erythrocyte progenitor; GMP: granulocyte-macrophage progenitor; CLP: common lymphoid progenitor. (B) In the Lineage Continuum Model, HSCs

give rise to less defined intermediates that are viewed as a “continuum of low-primed undifferentiated hematopoietic stem and progenitor cells (CLOUD-HSPCs)”. From these, unipotent progenitors arise. (C) The direct model proposes that unipotent

progenitors acquire a defined lineage fate soon after emerging from HSCs without passing through a stage of an oligopotent progenitor. Recent studies suggest that the megakaryocytic lineage directly evolves from HSCs.

(5)

1

analyzed. Furthermore, in vivo studies allow to distinguish between short-term HSCs whose ability of hematopoietic reconstitution is limited to a first round of transplantation, and long-term HSCs, which can contribute to hematopoiesis in recipients even in serial-transplantation experiments.15

Phenotypic HSC features

Using the above-named methods, multiple cellular surface proteins have been found to mark HSCs and collectively describe the HSC phenotype (Figure 2A). A well-known marker that is uniquely expressed on human HSPCs but not on differentiated cells is CD34,16 and it is frequently used to enrich for HSCs in studies that aim to further characterize them. Functionally, CD34 is a glycoprotein ligand that can bind to vascular selectins and thereby is supposed to have a role in cell migration.17_Further enrichment for HSCs within the CD34+

population was achieved with the surface markers CD38,18,19 _CD90,20_CD49f21_and CD45RA.22 _{HSCs characteristically have} low or no expression of CD38, are positive for CD90 and CD49f and lack expression of CD45RA, resulting in the distinctive

CD34+_CD38-_CD90+_CD49f+_CD45RA

-phenotype. However, already the

CD34+_CD38- _{cell fraction is highly}

enriched for HSCs, and often only these two markers are used to isolate human HSCs for further applications. Notably, phenotypic markers for murine HSCs are distinct from human. Purification strategies for murine HSPCs usually involve a positive selection for the markers c-kit and Sca-1, and a negative selection for mature hematopoietic cell lineage (Lin) markers (usually Gr-1, Mac-1, B220, CD4, CD8, and Ter-119),23

resulting in the characteristic Lin

-Characteristics of

hematopoietic stem cells

Strategies for HSC identification

In total, the human body contains between 20 and 30 trillion blood cells from which the majority is red blood cells. HSCs in contrast are a very rare cell population, however, due to its unique and crucial function for hematopoiesis a lot of research is focused on studying and characterizing these particular cells. Functionally, HSCs are defined by their ability to both self-renew and reconstitute the entire hematopoietic system. In order to test if - and to which extent- a certain hematopoietic cell population has these particular characteristics, several functional in

vitro and in vivo assays have been

developed. A frequently used in vitro assay to rapidly identify hematopoietic progenitors is the Colony Forming Cell (CFC) assay,13 _{which reads out the ability} of cells to form colonies on a semi-solid medium that contains growth factors required for differentiation towards the distinct lineages. Subsequent replating of colonies can further provide preliminary insight into the self-renewal capacity of those cells, however, this information is limited due to the rather short duration of this assay. In order to identify the most primitive HSC population in vitro, the Long-Term Culture Initiating Cell (LTC-IC) assay14_{is often used, in which} cells are first co-cultured on a supportive stromal layer for several weeks prior to reading out their sustained colony-forming ability. However, the current gold standard for identifying HSCs are

in vivo studies, in which hematopoietic

cells are transplanted into irradiated mice and their ability of contributing to hematopoiesis in the recipient animal is

(6)

1

Sca-1+_c-kit+ _{(LSK) phenotype. Further}

enrichment of HSCs from the LSK fraction can be achieved by isolating LSK-CD34-_Flk2-24_{or LSK-150}+_CD48-25 cells. In addition, also LSK-independent strategies to enrich for murine HSCs were described.26–28_{Besides surface} markers that are convenient for the identification and isolation of stem cells,

HSCs also express certain receptors that are crucial for their function and which can be stimulated by their corresponding ligands – a fact that is used when culturing HSCs in vitro. Two receptor/ ligand pairs that have been shown to be critical for HSC function are c-kit/ SCF29,30_{and MPL/TPO}31_{(SCF: Stem Cell} Factor; MPL: Thrombopoietin-Receptor, Figure 2. Characteristics of Hematopoietic stem cells (HSCs). (A) HSCs can be distinguished from other cells

with the help of surface marker that are typically either absent or present on HSCs. Human HSCs have the characteristic CD34+_CD38-_CD90+_CD49f+_CD45RA-_{phenotype, whereof often only CD34 and CD38 are used to enrich them (highlighted}

in bold). Murine HSCs are often described as Lin-_c-kit+_Sca-1+_CD34-_Flk2-_CD150+_CD48- _(Lin-_{: negative selection for the lineage}

markers Gr-1, Mac-1, B220, CD4, CD8, and Ter-119). (B) HSCs reside in bone marrow niches surrounded by other cell types

such as endothelial cells, mesenchymal stromal cells, megakaryocytes, osteoblasts or nerve cells. Cytokines produced by those cells can bind to surface receptors of HSCs and influence their function. SCF: stem cell factor, TPO: thrombopoietin, Ang-1: Angiopoietin 1; TGF-β: transforming growth factor beta; CXCL4/12: chemokine (C-X-C motif) ligand 4/12. (C) Stimuli from the microenvironment can trigger signaling cascades via receptors located on the HSC surface and consequently result in activation of transcription factors that mediated stem cell quiescence or self-renewal and expansion. Figure adapted from Lin et al, Stem cell Investig. 2015.61_{(D) HSC function and maintenance are influenced by a complex network of transcription}

(7)

1

renewal and maintenance, for example by inducing Wnt-signaling,42 Notch-signaling43_{or TGF-β-signaling}44₍_Figure

2C). In addition to niche-associated factors, also other components such as growth factors, inflammatory cytokines and availability of nutrients and oxygen influence signaling in HSCs. Eventually, activation of various signaling cascades by diverse stimuli results in translocation or activation of transcription factors (TF) that further determine the characteristic gene expression profile of HSCs. TFs that are thought to have crucial roles in HSC regulation are β-catenin,42 _c-Myc,45 SMAD,46_STAT3/5,47_ICN,43 _C/EBPα,48 JUNB,49_PU.1,50_RUNX,51 _GATA2,52 HOXB4,53_GFI,54,55 _FOXOs56 _{and HIFs.}57 However, the transcriptional network responsible for HSC regulation is highly complex and also involves additional factors to the ones listed here.58–60 Together, numerous TFs regulate the expression of genes involved in regulation of cell cycle, proliferation, differentiation, apoptosis or metabolism- and thereby key programs for cell fate, -survival, and -function. Notably, HSC function can be strongly influenced by their metabolic state, which is reviewed in more detail in chapter 4.

Acute myeloid leukemia

Leukemia is a form of cancer that arises from immature blood cells that normally would develop into the different mature, functional cells types. However, during leukemia development this process gets disturbed and results in accumulation of dysfunctional blood cells. In acute myeloid leukemia (AML), the development of the myeloid cell lineage is disturbed, and results in infiltration of the bone marrow, blood and other tissues by expansion of poorly TPO: Thrombopoietin). Furthermore,

the tyrosine kinase receptor FMS-like tyrosine kinase 3 (FLT3) was found to be highly expressed on HSPCs32 _{and its} ligand FLT3-L was shown to enhance the proliferation of primitive progenitor cells.33,34_{Therefore, a cytokine-cocktail} composed of SCF, TPO and FLT3-L is frequently used for in vitro expansion of HS(P)Cs.

HSCs reside in stem cell niches and are regulated by a complex transcription factor network

The balance between HSC self-renewal vs. HSC differentiation is strongly influenced by their microenvironment. HSCs can reside in different bone marrow regions, so-called stem cell niches, and depending on their location HSCs are supposed to have a distinct self-renewal capacity.35_{Different cell} types in the niche such as endothelial cells, mesenchymal stromal cells, megakaryocytes, osteoblasts or nerve cells can impact the function of HSCs (Figure 2B).36_{Mechanistically, these} cells can regulate HSCs directly by secretion of cytokines, or indirectly for example by stimulating the cytokine release of other cells. Quiescent HSCs are thought to be located in the so called “perivascular niche” close to perivascular stromal cells and arterioles with low levels of oxygen.37_{Cells in this} niche produce cytokines such as SCF, TPO, Angiopoetin-1 (Ang-1), C-X-C motif chemokine 12 and 4 (CXCL12, CXCL4) and transforming growth factor β (TGF-β), which all have been shown to impact HSC function.38–41 Stimuli from the microenvironment can trigger signaling cascades via receptors located on the HSC surface and consequently impact HSC

(8)

self-535611-L-sub01-bw-Mattes 535611-L-sub01-bw-Mattes 535611-L-sub01-bw-Mattes 535611-L-sub01-bw-Mattes Processed on: 23-9-2019 Processed on: 23-9-2019 Processed on: 23-9-2019

1

cohesin-complex genes (e.g. STAG2), and (9) spliceosome-complex genes62 (Figure 3).

Clonal disease evolution

Cells from AML patients usually have a lower number of gene mutations compared to cells from other cancers, with an average of 13 mutation per

AML case.62_{Of these, an average}

of 5 are driver mutations which are found in genes recurrently mutated in AML. Other mutations are passenger mutations which might have a role in the malignant cell behavior, but have not shown to be directly involved in malignant transformation. Notably, not all the AML cells of a patient contain the same alterations. Distinct subclones usually arise from one founding clone that accumulates additional alterations throughout disease progression, a progress termed “clonal evolution”. Longitudinal sequencing studies discovered that the order of mutation acquisition follows certain patterns: Early mutations in AML evolution are thought to happen in the stem cell population and often effect epigenetic regulators (e.g DNMT3A, ASXL1,

IDH2, and TET2), which gives rise to

so called “pre-leukemic” stem cells that have altered self-renewal properties, but are still capable of multilineage differentiation.63,64_{In contrast, later} mutations commonly occur in genes related to proliferation and signaling activation (e.g. FLT3, KRAS/NRAS,

PTPN11) and lead to frank leukemia

(Figure 4). Leukemic blasts are thought to derive from leukemia stem cells (LSCs) that have self-renewal properties similar to normal- or preleukemic HSCs, but also carry the late genetic alterations that affect cellular proliferation and differentiation.65

or abnormally differentiated cells. In contrast to chronic myeloid leukemia (CML), which is characterized by an abnormal accumulation of more mature myeloid cells, AML is defined by a high percentage (≥ 20%) of immature blasts in the bone marrow.

Mutational spectrum in AML

AML is a very heterogeneous disease, and many different mutations and chromosomal alterations can lead to disease development. Mechanistically, AML-associated mutations can drive disease development via multiple strategies. Gene mutations can lead to hyper-activation of signaling cascades that trigger uncontrolled cell proliferation, confer resistance to apoptosis, or directly affect transcription factors crucial for cell differentiation. Furthermore, mutations can induce aberrant gene expression by changing DNA structure and accessibility or the subcellular location of transcription factors. Thereby, mutations can impact self-renewal and life-span of the stem- and progenitor cell population, or affect important metabolic regulators. According to The Cancer Genome Network, which analyzed the genome of 200 de novo AML patients, gene mutation relevant for AML pathogenesis can be assigned to 9 different categories:

(1) transcription-factor fusions (e.g.

PML-RARA, AML1-ETO), (2) the gene encoding nucleophosmin (NPM1), (3) tumor-suppressor genes (e.g. TP53,

WT1), (4) DNA-methylation-related

genes (e.g. DNTM3A, IDH1, IDH2,

TET2), (5) signaling genes (e.g. FLT3, KIT, KRAS/NAS), (6) chromatin-modifying

genes (e.g. ASXL1, EZH2, MLL-fusion proteins), (7) myeloid transcription-factor genes (RUNX1, C/EBPα), (8)

(9)

1

Figure 3. Categories of genes commonly mutated in acute myeloid leukemia (AML). Mutations in signaling genes such as FLT3 lead to hyperactivation of signaling cascades and enhanced cell proliferation (upper left box). Mutations in myeloid transcription factors or their fusion by chromosomal translocations can impair hematopoietic differentiation (center left box). Cohesin-complex gene mutations such as RAD21 or STAG2 can impair chromosome segregation and induce gene expression changes (center middle box). Mutations in tumor suppressor genes such as TP53 confer resistance to apoptosis through impaired transcriptional activation of apoptosis-related genes. Moreover, mutations alter kinetics of p53 degradation by MDM2 and PTEN (upper middle box). Mutations in epigenetic regulators such as DNMT3A, TET2 or IDH1/2 (which changes 2-hydroxyglutarate (2HG) levels) can change DNA methylation levels and thus transcriptional activity (upper right box). Mutations in other epigenetic modifiers such as EZH2 or ASXL1 can alter chromatin modifications such as methylation of histones H3 and H2A on lysine residues K27, K119 and K79, respectively. Gene fusions such as KMT2A-MLLT3 alter the activity of the methyltransferase DOT1L (center right box). Mutations of genes that are part of the splicing machinery affect RNA processing and thereby the transcriptional output (bottom right box). Mutations in the gene encoding the nucleocytoplasmic shuttling protein NPM1 result in aberrant location of NPM1 itself or NPM1 interacting proteins (bottom left box). Figure adapted from Döhner et al, NEJM 2015.66

(10)

1

SCT at the age ≤65 years.67_{While these} therapy approaches commonly lead to an initial reduction of the leukemic blast in the majority of patients, disease relapse represents the major reason patients are not cured. Especially in patients with an unfavorable prognosis, including presence of complex cytogenetics,

TP53 mutations or a high allelic ratio

of FLT3-ITD.68_{Apparently, certain}

AML cells survive the chemotherapy treatment, expand during remission and contribute to leukemia reoccurrence. Since these cells are rare following the chemotherapy treatment, knowledge about cell biological characteristics of these cells is limited. Various concepts exist that speculate on the nature of these resistant cells: It might be that pre-leukemic HSCs survive therapy and acquire again additional alteration that drive relapse. Alternatively, it could be that minor genetic leukemic sub-clones are less responsive to therapy, persist and expand after treatment.69

Challenges of AML therapy

Therapy success of AML is strongly dependent on the molecular background of the disease, however, in general, AML can be viewed as a disease with a rather poor prognosis. AML is cured in 35-40% of the adult patients younger than 60 years old, and only in 5-15% of the patients older than 60 years of age.66 The general strategy for AML induction therapy usually consists of chemotherapy with the pyrimidine nucleoside analog cytarabine in combination with anthracyclines, applied in the ‘7+3’ regimen. Follow up treatments on this induction therapy are based on the risk stratification for the individual patients. Based on the ELN risk classification, high-risk and intermediate-risk patients under the age of 75 years usually proceed to allogenic stem cell transplantation (SCT). Low-risk AML patients can receive additional cycles of chemotherapy or might be considered for autologous

Figure 4. AML evolution from pre-leukemia to leukemia. In this simplified model, early mutations in AML evolution are thought to occur in normal hematopoietic stem cells (HSCs) and mainly affect epigenetic regulators such as DNMT3A, ASXL1, IDH1/2, or TET2, resulting in enhanced self-renewal and formation of pre-leukemic HSCs.63,78_Pre-leukemic HSCs promote clonal hematopoiesis, but do not necessarily result in leukemic transformation. Acquisition of additional hits (“late mutations”) in signaling genes such as FLT3, KRAS or NRAS lead accumulation of leukemic blasts. Leukemia stem cells (LSCs) are thought to have both self-renewal and proliferation abilities and maintain the blast cell population. In AML, characteristically several genetically distinct subclones are found, which arise from one founding clone that accumulates additional alterations throughout disease progression.

(11)

1

HSCs in animal transplants, LSCs are functionally defined by their ability to initiate leukemia when transplanted into immunodeficient animals.74,79 _In contrast to bulk leukemic blasts, LSCs are able to self-renew, initiate leukemia even in serial transplants and give rise to non-LSC blast cells. However, it has to be noted that not all leukemia subclones engraft in current mouse models, or at least not equally efficient, and thereby this way of identifying LSCs also has its limitations.80

LSCs partly share characteristics with normal HSCs. For instance, multiple studies indicated that the LSC population contains quiescent cells,81–83 which have altered expression of genes that restrain cell cycle progression and prevent differentiation.84_Furthermore,

HSCs and LSCs partly share

immunophenotypic characteristics. In CD34+_{AMLs, which is the majority of}

AMLs that is defined by the presence of >10% CD34 positive blast cells, LSCs - similar to HSCs- seem to predominantly reside in the CD34+_CD38-_fraction.74–76 However, other studies demonstrated that LSCs are definitely not restricted to this fraction, but can be found also in the CD34+_CD38+_{cell population or}

less frequently even in CD34-_cells.85,86 In AMLs with low CD34 expression, which are often AMLs with mutations in the NPM1 gene, LSCs mainly reside in the CD34-_fraction.87_{Another similarity} between HSCs and LSCs is that they share expression of stemness-related genes such as ABCB1, MEIS1, ERG or homeobox genes, as well as activation of signaling pathways such as JAK-STAT or Notch.86 _{This correlation was observed in} a study where AML cells were first sorted in different populations based on CD34/ CD38 expression and subsequently Leukemia-associated mutations persist

in approximately half of the patients during complete remission, and detection of molecular minimal residual disease (except mutations in DNMT3A,

TET2, and ASXL1) was associated with

an increased risk for disease relapse.70 In the majority of AML patients, driver mutations present at relapse overlap with driver mutations detected at diagnosis. However, frequently gain- or loss of gene mutations is observed, indicating that the relapse-clone has evolved from a preexisting clone by accumulation or loss of mutations.71–73_{It is supposed that} relapse originates from a rare leukemia stem cell (LSC) population - a concept that will be explained in more detail in the next section.72 _{LSCs represent,} similar to HSCs, the most immature cell fraction of AML cells and are associated with a CD34+_CD38-_phenotype.74–76_It was recently demonstrated that a high number of CD34+_CD38-_{cells at the time}

of diagnosis correlates with a higher incidence of relapse and poor survival.77 Importantly, these distinct models on disease relapse are likely not mutually exclusive and different concepts could apply in different AML cases and might be dependent on the age of the patient. Research in AML is currently aimed to identify the characteristics of cells that drive disease relapse in order to improve AML therapy.

The leukemia stem cell

theory

The leukemia stem cell theory proposes that AML is a hierarchical organized disease that is maintained by a rare fraction of self-renewing LSCs that form the apex of the hierarchy.65_{Similar to} the repopulation capacity of normal

(12)

1

stress-response pathways or signal

transduction.94 _{The transcriptional}

co-regulator CITED2

(CBP/p300-interacting transactivator with glutamic acid (E) and aspartic acid (D)–rich tail 2) is involved in the regulation of several of those pathways crucial for stem cell homeostasis. CITED2 expression was shown to positively correlated with stem cell activity and maintenance,95,96_and can therefore be viewed as a “stemness gene”.

Structural features of CITED2

The protein CITED2 lacks a classical DNA binding domain, but consists of three highly conserved regions (CR 1-3) and a characteristic serine-glycine rich junction (SRJ) that serve as domains for the interaction with various partners. CITED2 was originally identified as a protein binding to the cysteine-histidine-rich (CH1) domain of the acetyltransferases CBP and p300,97,98 which function as transcriptional activators. Mechanistically, CBP/p300-mediated acetylation of histone proteins can open up chromatin structures at the promoter region of genes, whereas acetylation of transcription factors can change their conformation and activity. Many transcription factors require binding to CBP/p300 for their full activation, and this interaction can be either increased or blocked by the modulator CITED2. By competing with their binding for CBP/p300, CITED2 was shown to function as a negative regulator for HIF-1α,97,98_NF-κB,99_and TP53.100_{In contrast, CITED2 increased} CBP/p300 mediated co-activation of

the transcription factors TFAP2,101

MYC,102_PPARα,103_SMAD2104_and

SMAD3.105_{Besides CBP/p300, CITED2} was also found to interact with the applied to xenotransplant assays to

identify LSC-enriched populations.86 Other gene expression data suggested that LSCs in general have more in common with normal progenitors than with normal HSCs. Notably, in that study the sorting strategy to enrich for LSCs included the CD45RA+ _cell

population, which corresponds to the normal progenitor cell population, and could explain the difference compared to previous studies.75_{Therefore, it is still}

not entirely defined if LSCs directly arise from HSCs, or if LSCs are progenitor cells that acquire stem-cell features, or if maybe both concepts are possible depending on the AML case.

Despite certain similarities, HSCs and LSCs also have distinct features, which might represent targeting-option for therapy. Multiple cell surface markers have been found to be higher expressed in LSCs compared to HSCs,65,88_for instance CD123 (which is part of the IL-3 receptor),89 _TIM3,90_IL1RAP,91 CD9392 _{or GPR56.}93_{Recently, evidence} accumulates that LSCs across distinct AML genotypes share metabolic features such as low mitochondrial activity, low levels of reactive oxygen species and low amounts of ATP. These characteristics partly overlap with HSCs, however, LSCs seem to depend on different pathways for maintenance of their metabolic state. Thus, interference with those pathways represents a promising strategy for LSC-selective targeting. The characteristic metabolic features of LSCs will be described in detail in chapter 4.

The “stemness gene” CITED2

Stem cell maintenance and proliferation is regulated by multiple factors that influence cell cycle, apoptosis,

(13)

1

binds to multiple ETS-binding sites in the CITED2 promoter.96,111

CITED2 functions in the hematopoietic system

CITED2 is a ubiquitously expressed protein and important for various biological processes. Full deletion of Cited2 in mice is embryonically lethal and is associated with cardiac malformations, adrenal agenesis, neural crest defects and exencephaly.101_In the hematopoietic system, Cited2 was shown to be especially important for the functionality of stem cells, since conditional deletion in mice using Mx1-Cre resulted in depletion of HSCs and

multilineage bone marrow failure.112

In contrast, conditional deletion of Cited2 in committed lymphoid and myeloid lineages had no impact on the maintenance of these lineages.112_{In line} with the murine data, overexpression of CITED2 in human CD34+_{stem- and}

progenitor cells enhanced quiescence

and maintenance of CD34+_CD38

-HSCs. In AML, CITED2 expression is found overexpressed in a subgroup of patients compared to normal HSCs, and knockdown of CITED2 suppressed the growth of AML cells in both in vitro and

in vivo studies.96

Mechanistically, CITED2 contributes to increased stem cell homeostasis by regulating crucial pathways such as cell cycle, proliferation and apoptosis (Figure 5). Increased HSC maintenance in human cord blood cells upon CITED2 overexpression was associated with increased expression of the cycline-dependent kinase (CDK) inhibitor

p21,96_{which is known to mediate}

quiescence. In murine cells, Cited2 was shown to repress the proliferation acetyltransferase GCN5,106,107 _thereby

preventing GCN5-mediated acetylation and hence inactivation of peroxisome proliferator-activated receptor-coactivators 1α (PGC-1α).

Regulation of CITED2 expression

The expression of CITED2 can be induced by various cytokines and biological stimuli such as 1α, IL-2, IL-3, IL-4, IL-6, IL-9 and IL-11,

granulocyte/macrophage

colony-stimulating factor (GM-CSF), interferon

gamma (IFN-γ), platelet-derived

growth factor, insulin, serum, and lipopolysaccharide.108 _{In the promoter} region of CITED2 multiple bindings sites for the transcription factor STAT5 have been identified,109,110_{which is known as} an important downstream mediator of cytokine signaling. Furthermore, the promoter region of CITED2 contains a FOXO3A binding element and it was shown that both STAT5 and FOXO3A are required for highest CITED2 expression in erythroid cells.110_{Interestingly, STAT5} and FOXO3A represent quite opposing pathways, since STAT5 signaling is associated with cell proliferation, whereas FOXO3A signaling rather inhibits proliferation or drives apoptosis. Thus, CITED2 acts as a critical modulator at the interface of differing pathways. The promoter region of CITED2 further contains binding elements for the transcription factors HIF-1 and NF-κB, which are in turn negatively regulated by CITED2, and thereby most likely form a negative feedback loop by promoting CITED2 expression.97,109 _{Additionally, the} transcription factor c-MYC stimulates CITED2 expression via direct binding to the CITED2 promoter region.102_CITED2 expression is negatively regulated by the transcription factor PU.1, which directly

(14)

1

p57 and Hes1.105_{ERG1 has a crucial}

role in HSC maintenance by regulating retention of HSCs in the bone marrow environment and preventing their proliferation.113 _{ERG1 is a target gene of} the TGF-β/SMAD3 signaling pathway,114 and CITED2 was shown to directly interact with SMAD3 and co-activate EGR1 expression.105 _{The cyclin dependent} inhibitors p16Ink4a _{and p19}Arf_{, as well as}

the anti-apoptotic protein p53, thereby preventing depletion of HSCs. Additional deletion of the Ink4a/Arf locus or Tp53 was shown to restore HSC functionality in Cited2-deficient mice.112_{Deletion of}

Cited2 in HSCs was further shown to be

accompanied by decreased expression of stemness-related genes such as Egr1,

Figure 5. Pathways of CITED2-mediated increase of stem cell maintenance. CITED2 regulates multiple pathways affecting cell cycle, proliferation and metabolism, and thereby influences HSC maintenance. CITED2 has been shown to promote the expression of the proliferation inhibitors p57 and HES1 in a HIF-1 dependent manner. CITED2 blocks the CBP/300-mediated transactivation of HIF-1 and thereby most likely interferes with HIF-1-mediated repression of p57 and HES1 (upper box). CITED2 interferes with PI3K/AKT-mediated phosphorylation of the transcription factors FOXO1 and FOXO3A. FOXO proteins drive the expression of genes important for glycolysis (upper right box) and mitochondrial regulation (lower right box). In liver cells, CITED2 was shown to interact with the acetyltransferase GCN5 and to inhibit GCN5-mediated acetylation of the mitochondrial regulator PGC1-α, thereby promoting mitochondrial biogenesis and turnover. This might also apply for HSCs (bottom box). CITED2 represses the expression of the cyclin-dependent-kinase (CDK) inhibitors p16INK4 and p19ARF, and prevents activation of the pro-apoptotic protein p53 (lower left box). CITED2 promotes expression of the CDK inhibitor p21 which increases quiescence (center left box). CITED2 co-activates the transcription factor SMAD3 by recruiting the acetyltransferase CBP/p300 and thereby promotes ERG1-mediated HSC maintenance.

(15)

1

kinase (CDK) inhibitor p57 is required for quiescence and maintenance of adult HSCs,115,116_{whereas the Notch signaling} target HES1 was shown to restrict GATA-2 expression in emerging HSCs.117_The expression of both genes is repressed in a HIF-1α-dependent manner, which can be prevented by CITED2 that interferes with HIF-1α activation. Furthermore, CITED2 was shown to be involved in metabolic pathways crucial for HSC functionality. HSCs mainly rely on glycolysis for their energy production and are further characterized by low mitochondrial activity (see chapter 4).

Cited2 deletion in murine HSCs reduced

the expression of glycolysis related genes such as pyruvate dehydrogenase kinase 2 and 4 (Pdk2, Pdk4) and lactate dehydrogenases B and D (Ldhb, Ldhd), and resulted in elevated mitochondrial membrane potential, elongation of mitochondrial shape, and elevated ROS

production.118_{Interestingly, loss of}

CITED2 was accompanied by increased phosphorylation and thus deactivation of the transcription factors FOXO1 and FOXO3A, which are involved in regulation of those metabolic genes.118 Notably, in liver cells CITED2 was also shown to positively regulate mitochondrial metabolism by mediating increased activation of PGC-1α, which promotes mitochondrial biogenesis. CITED2 blocks the acetylation of PGC-1α by GCN5, which increases its transcriptional activity.106_{It may seem} contradictory that CITED2 can both block and stimulate mitochondrial activity, however, increased biogenesis also contributes to increased mitochondria turnover, which promotes mitochondrial viability and low ROS production. This highlights that CITED2-mediated regulation of mitochondrial homeostasis is critical for HSC integrity.

Scope of this thesis

The success of current treatment strategies for AML is limited due to the high rate of disease relapse following intensive chemotherapy and stem cell transplantation. Apparently, certain cell populations survive conventional therapy approaches, grow out after cessation of therapy and are responsible for disease recurrence. Originally it was thought that chemotherapy treatment might introduce additional mutations in leukemia cells that confer drug resistance.119_{However, more recent} studies suggest that therapy resistant cell populations are already present at the moment of diagnosis prior to start of therapy, as revealed by sequencing studies that analyzed patients samples before and after chemotherapy treatment.71,120,121 _Evidence _exists that relapse is initiated by leukemia stem cells (LSCs), which have, in contrast to the bulk of AML cells, self-renewal properties similar to normal hematopoietic stem cells (HSCs), and are not efficiently targeted by standard therapy approaches.76,82,122 _{It is therefore} important to study the characteristics of LSCs and unravel the molecular mechanisms that drive their stemness- and drug-resistant- properties. This might help to identify potential strategies for LSC targeting and to improve current AML therapy protocols.

The first part of this thesis will focus on studying the transcriptional co-regulator CITED2, which was shown to have a crucial role in regulating the maintenance of both HSCs and LSCs. Whereas overexpression of CITED2 in HSCs results in their increased

quiescence and maintenance,96

(16)

1

cells leads to increased HSC cycling and apoptosis.105_{Similar, downregulation of} CITED2 in AML stem- and progenitor cells severely impairs their survival and

expansion,96_{indicating that CITED2}

contributes to leukemia maintenance by regulating the stem cell properties. Hence, CITED2 represents a potential target for AML therapy. Chapter 2 aims to gain insight in the molecular mechanism that drive reduced AML cell survival upon reduction of CITED2 levels. AML cells are transduced with lentiviral constructs for RNAi-mediated knockdown of CITED2 and potential changes in signal transduction pathways and gene expression are studied.

Leukemic transformation is thought to be a multistep process, in which initial mutations occur in hematopoietic stem and progenitor cells (HSPCs) and alter their lifespan and/or maintenance and lead to clonal hematopoiesis.123–125 Additional mutations in such preleukemic HSPCs that affect differentiation- or proliferation pathways might result in the formation of LSCs and consequently

lead to leukemia onset.63_Thus,

combining multiple genetic alterations might have cooperative effects on generating LSCs or influencing their maintenance. While CITED2 is known to increase stem cell maintenance, the transcription factor PU.1 is crucial for proper differentiation of the myeloid lineage.126,127_{AML-associated mutations} frequently result in the activation of signaling pathways that drive CITED2 expression,97,108,110,128–130 _{and also often} result in impaired expression or activation of PU.1.131–137 _{Notably, PU.1}

was found to be a negative regulator of CITED2 expression,96_{indicating that} both factors act in the same signaling network and might have cooperative

effects on regulating cellular properties.

Chapter 3 investigates the potential

effects of combined CITED2 upregulation and PU.1 downregulation on stem cell maintenance or differentiation. In addition, it questions whether combined CITED2/PU.1 deregulation is sufficient for leukemic transformation.

Cited2 deletion in murine HSCs was

previously shown to impact the HSC metabolism. Loss of CITED2 resulted in decreased expression of glycolysis genes, increased mitochondrial activity and higher levels of reactive oxygen species

(ROS).87_{Recent studies highlighted}

that LSCs also have different metabolic properties compared to the leukemic bulk cell population, which will be the focus of the second part of this thesis. LSCs characteristically have low levels of ROS, which apparently result from a combination of low mitochondrial activity and high activity of ROS-removing pathways such as autophagy.138,139 _{The review in chapter}

4 will summarize in detail

ROS-regulating pathways and their relevance for the functionality of both normal- and leukemic stem cells. This chapter will also highlight that the metabolic state of LSCs implicates certain vulnerabilities of these cells. Even though LSCs are characterized by a low mitochondrial activity, they rely on this remaining activity for their survival and are highly dependent on mechanisms that regulate mitochondrial integrity. In particular, the anti-apoptotic protein BCL2 and proteins regulating mitophagy play an important role in LSC maintenance and survival and represent promising targets for AML therapy.139,140

Previous studies demonstrated that LSCs are enriched in the fraction with

(17)

1 References

1. Kondo, M., Weissman, I. L. & Akashi, K. Identification of clonogenic common lym-phoid progenitors in mouse bone marrow. Cell 91, 661–672 (1997).

2. Akashi, K., Traver, D., Miyamoto, T. & Weissman, I. L. A clonogenic common my-eloid progenitor that gives rise to all mymy-eloid lineages. Nature 404, 193–197 (2000).

3. Pei, W. et al. Polylox barcoding reveals haematopoietic stem cell fates realized in vivo. Nature 548, 456–460 (2017).

4. Athanasiadis, E. I. et al. Single-cell RNA-sequencing uncovers transcriptional states and fate decisions in haematopoiesis. Nat. Commun. 8, 2045 (2017).

5. Velten, L. et al. Human haematopoietic stem cell lineage commitment is a contin-uous process. Nat. Cell Biol. 19, 271–281 (2017).

6. Notta, F. et al. Distinct routes of lineage development reshape the human blood

hierarchy across ontogeny. Science 351, aab2116 (2016).

7. Adolfsson, J. et al. Identification of Flt3+ lympho-myeloid stem cells lacking eryth-ro-megakaryocytic potential a revised road map for adult blood lineage commitment. Cell 121, 295–306 (2005).

8. Yamamoto, R. et al. Clonal analysis unveils self-renewing lineage-restricted progenitors generated directly from hematopoietic stem cells. Cell 154, 1112–1126 (2013).

9. Carrelha, J. et al. Hierarchically relat-ed lineage-restrictrelat-ed fates of multipotent haematopoietic stem cells. Nature 554, 106–111 (2018).

10. Rodriguez-Fraticelli, A. E. et al. Clonal analysis of lineage fate in native haemato-poiesis. Nature 553, 212–216 (2018).

11. Sun, J. et al. Clonal dynamics of native haematopoiesis. Nature 514, 322–327 (2014).

12. Busch, K. et al. Fundamental properties of unperturbed haematopoiesis from stem cells in vivo. Nature 518, 542–546 (2015).

13. Pereira, C., Clarke, E. & Damen, J. Hematopoietic colony-forming cell assays. Methods Mol. Biol. 407, 177–208 (2007).

14. Liu, M., Miller, C. L. & Eaves, C. J. Hu-man long-term culture initiating cell assay. Methods Mol. Biol. 946, 241–256 (2013).

15. Purton, L. E. & Scadden, D. T. Limiting factors in murine hematopoietic stem cell assays. Cell Stem Cell 1, 263–270 (2007).

16. Berenson, R. J. et al. Antigen CD34+ marrow cells engraft lethally irradiated ba-boons. J. Clin. Invest. 81, 951–955 (1988).

17. AbuSamra, D. B. et al. Not just a mark-er: CD34 on human hematopoietic stem/ progenitor cells dominates vascular selec-tin binding along with CD44. Blood Adv. 1, 2799–2816 (2017).

the lowest ROS signal when total mononuclear AML cells are stained with a fluorescent ROS dye.138_{However, it}

has not been investigated yet in detail

how ROS levels within the CD34+

AML cell population correlate with cellular features such as morphology, metabolic activity, gene expression and drug responsiveness. Chapter 5 aims to identify characteristics of ROS-low

and ROS-high CD34+_{AML cells and}

to investigate if ROS levels within the

CD34+_{population are indicative for}

the presence of stem cell properties. Furthermore, the responsiveness of

ROS-low and ROS-high CD34+_AML

cells to the BCL2 inhibitor Venetoclax is tested, which is thought to target the LSC population and is currently intensively studied.140

Finally, chapter 6 summarizes and discusses the research described in this thesis.

(18)

1

18. Hao, Q. L., Thiemann, F. T., Petersen, D., Smogorzewska, E. M. & Crooks, G. M. Extended long-term culture reveals a highly quiescent and primitive human hemato-poietic progenitor population. Blood 88, 3306–3313 (1996).

19. Petzer, A. L., Hogge, D. E., Landsdorp, P. M., Reid, D. S. & Eaves, C. J. Self-renew-al of primitive human hematopoietic cells (long-term-culture-initiating cells) in vitro and their expansion in defined medium. Proc. Natl. Acad. Sci. U. S. A. 93, 1470–1474 (1996).

20. Baum, C. M., Weissman, I. L., Tsukamoto, A. S., Buckle, A. M. & Peault, B. Isolation of a candidate human hematopoietic stem-cell population. Proc. Natl. Acad. Sci. U. S. A. 89, 2804–2808 (1992).

21. Notta, F. et al. Isolation of single hu-man hematopoietic stem cells capable of long-term multilineage engraftment. Science 333, 218–221 (2011).

22. Majeti, R., Park, C. Y. & Weissman, I. L. Identification of a hierarchy of multipotent hematopoietic progenitors in human cord blood. Cell Stem Cell 1, 635–645 (2007).

23. Challen, G. A., Boles, N., Lin, K. K.-Y. & Goodell, M. A. Mouse hematopoietic stem cell identification and analysis. Cytometry. A 75, 14–24 (2009).

24. Adolfsson, J. et al. Upregulation of Flt3 expression within the bone marrow Lin(-) Sca1(+)c-kit(+) stem cell compartment is accompanied by loss of self-renewal capaci-ty. Immunity 15, 659–669 (2001).

25. Kiel, M. J., Radice, G. L. & Morrison, S. J. Lack of evidence that hematopoietic stem cells depend on N-cadherin-mediated adhe-sion to osteoblasts for their maintenance. Cell Stem Cell 1, 204–217 (2007).

26. Goodell, M. A., Brose, K., Paradis, G., Conner, A. S. & Mulligan, R. C. Isolation and

functional properties of murine hematopoi-etic stem cells that are replicating in vivo. J. Exp. Med. 183, 1797–1806 (1996).

27. Dykstra, B. et al. High-resolution vid-eo monitoring of hematopoietic stem cells cultured in single-cell arrays identifies new features of self-renewal. Proc. Natl. Acad. Sci. U. S. A. 103, 8185–8190 (2006).

28. Balazs, A. B., Fabian, A. J., Esmon, C. T. & Mulligan, R. C. Endothelial protein C re-ceptor (CD201) explicitly identifies hema-topoietic stem cells in murine bone marrow. Blood 107, 2317–2321 (2006).

29. Nocka, K. et al. Expression of c-kit gene products in known cellular targets of W mu-tations in normal and W mutant mice--evi-dence for an impaired c-kit kinase in mu-tant mice. Genes Dev. 3, 816–826 (1989).

30. Zsebo, K. M. et al. Stem cell factor is encoded at the Sl locus of the mouse and is the ligand for the c-kit tyrosine kinase re-ceptor. Cell 63, 213–224 (1990).

31. Kimura, S., Roberts, A. W., Metcalf, D. & Alexander, W. S. Hematopoietic stem cell de-ficiencies in mice lacking c-Mpl, the receptor for thrombopoietin. Proc. Natl. Acad. Sci. U. S. A. 95, 1195–1200 (1998).

32. Matthews, W., Jordan, C. T., Wiegand, G. W., Pardoll, D. & Lemischka, I. R. A receptor tyrosine kinase specific to hematopoietic stem and progenitor cell-enriched popula-tions. Cell 65, 1143–1152 (1991).

33. Ku, H. et al. Soluble thrombopoietin receptor (Mpl) and granulocyte colony-stim-ulating factor receptor directly stimulate proliferation of primitive hematopoietic progenitors of mice in synergy with steel factor or the ligand for Flt3/Flk2. Blood 88, 4124–4131 (1996).

34. Yonemura, Y., Ku, H., Lyman, S. D. & Ogawa, M. In vitro expansion of hematopoi-etic progenitors and maintenance of stem

cells: comparison between FLT3/FLK-2 li-gand and KIT lili-gand. Blood 89, 1915–1921 (1997).

35. Haylock, D. N. et al. Hemopoietic stem cells with higher hemopoietic potential re-side at the bone marrow endosteum. Stem Cells 25, 1062–1069 (2007).

36. Birbrair, A. & Frenette, P. S. Niche heterogeneity in the bone marrow. Ann. N. Y. Acad. Sci. 1370, 82–96 (2016).

37. Morrison, S. J. & Scadden, D. T. The bone marrow niche for haematopoietic stem cells. Nature 505, 327–334 (2014).

38. Pinho, S. et al. PDGFRalpha and CD51 mark human nestin+ sphere-forming mes-enchymal stem cells capable of hematopoi-etic progenitor cell expansion. J. Exp. Med. 210, 1351–1367 (2013).

39. Bruns, I. et al. Megakaryocytes regu-late hematopoietic stem cell quiescence through CXCL4 secretion. Nat. Med. 20, 1315–1320 (2014).

40. Nakamura-Ishizu, A., Takubo, K., Fu-jioka, M. & Suda, T. Megakaryocytes are essential for HSC quiescence through the production of thrombopoietin. Biochem. Bio-phys. Res. Commun. 454, 353–357 (2014).

41. Zhao, M. et al. Megakaryocytes main-tain homeostatic quiescence and promote post-injury regeneration of hematopoiet-ic stem cells. Nat. Med. 20, 1321–1326 (2014).

42. Luis, T. C., Ichii, M., Brugman, M. H., Kin-cade, P. & Staal, F. J. T. Wnt signaling strength regulates normal hematopoiesis and its de-regulation is involved in leukemia develop-ment. Leukemia 26, 414–421 (2012).

43. Lampreia, F. P., Carmelo, J. G. & An-jos-Afonso, F. Notch Signaling in the Regula-tion of Hematopoietic Stem Cell. Curr. stem cell reports 3, 202–209 (2017).

(19)

1

44. Blank, U. & Karlsson, S. TGF-beta sig-naling in the control of hematopoietic stem cells. Blood 125, 3542–3550 (2015).

45. Laurenti, E. et al. Hematopoietic stem cell function and survival depend on c-Myc and N-Myc activity. Cell Stem Cell 3, 611– 624 (2008).

46. Hinge, A. & Filippi, M.-D. Deconstruct-ing the Complexity of TGFbeta SignalDeconstruct-ing in Hematopoietic Stem Cells: Quiescence and Beyond. Curr. stem cell reports 2, 388–397 (2016).

47. Staerk, J. & Constantinescu, S. N. The JAK-STAT pathway and hematopoietic stem cells from the JAK2 V617F perspective. JAK-STAT 1, 184–190 (2012).

48. Avellino, R. & Delwel, R. Expression and regulation of C/EBPalpha in normal myelo-poiesis and in malignant transformation. Blood 129, 2083–2091 (2017).

49. Passegue, E., Wagner, E. F. & Weissman, I. L. JunB deficiency leads to a myeloprolif-erative disorder arising from hematopoietic stem cells. Cell 119, 431–443 (2004).

50. Iwasaki, H. et al. Distinctive and indis-pensable roles of PU.1 in maintenance of hematopoietic stem cells and their differ-entiation. Blood 106, 1590–1600 (2005).

51. de Bruijn, M. & Dzierzak, E. Runx tran-scription factors in the development and function of the definitive hematopoietic sys-tem. Blood 129, 2061–2069 (2017).

52. Rodrigues, N. P., Tipping, A. J., Wang, Z. & Enver, T. GATA-2 mediated regulation of normal hematopoietic stem/progenitor cell function, myelodysplasia and myeloid leuke-mia. Int. J. Biochem. Cell Biol. 44, 457–460 (2012).

53. Schiedlmeier, B. et al. HOXB4’s road map to stem cell expansion. Proc. Natl. Acad. Sci. U. S. A. 104, 16952–16957 (2007).

54. van der Meer, L. T., Jansen, J. H. & van der Reijden, B. A. Gfi1 and Gfi1b: key reg-ulators of hematopoiesis. Leukemia 24, 1834–1843 (2010).

55. Shooshtarizadeh, P. et al. Gfi1b regu-lates the level of Wnt/beta-catenin signaling in hematopoietic stem cells and megakary-ocytes. Nat. Commun. 10, 1270 (2019).

56. Santo, E. E. & Paik, J. FOXO3a & hae-matopoietic stem cells: Goodbye PI3K, hello SIRT1? Cell cycle (Georgetown, Tex.) 15, 879–880 (2016).

57. Gezer, D., Vukovic, M., Soga, T., Pollard, P. J. & Kranc, K. R. Concise Review: Genetic Dissection of Hypoxia Signaling Pathways in Normal and Leukemic Stem Cells. Stem Cells 32, 1390–1397 (2014).

58. Beck, D. et al. Genome-wide analysis of transcriptional regulators in human HSPCs reveals a densely interconnected network of coding and noncoding genes. Blood 122, e12-22 (2013).

59. Pouzolles, M., Oburoglu, L., Taylor, N. & Zimmermann, V. S. Hematopoietic stem cell lineage specification. Curr. Opin. Hematol. 23, 311–317 (2016).

60. Kent, D. et al. Regulation of hemato-poietic stem cells by the steel factor/KIT signaling pathway. Clin. Cancer Res. 14, 1926–1930 (2008).

61. Lin, S., Zhao, R., Xiao, Y. & Li, P. Mech-anisms determining the fate of hemato-poietic stem cells. Stem cell Investig. 2, 10 (2015).

62. Ley, T. J. et al. Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N. Engl. J. Med. 368, 2059–2074 (2013).

63. Corces-Zimmerman, M. R., Hong, W.-J., Weissman, I. L., Medeiros, B. C. & Majeti, R. Preleukemic mutations in human acute

my-eloid leukemia affect epigenetic regulators and persist in remission. Proc. Natl. Acad. Sci. U. S. A. 111, 2548–2553 (2014).

64. Shlush, L. I. et al. Identification of pre-leukaemic haematopoietic stem cells in acute leukaemia. Nature 506, 328–333 (2014).

65. Thomas, D. & Majeti, R. Biology and relevance of human acute myeloid leukemia stem cells. Blood 129, 1577–1585 (2017).

66. Dohner, H., Weisdorf, D. J. & Bloomfield, C. D. Acute Myeloid Leukemia. N. Engl. J. Med. 373, 1136–1152 (2015).

67. Dohner, H. et al. Diagnosis and man-agement of AML in adults: 2017 ELN rec-ommendations from an international expert panel. Blood 129, 424–447 (2017).

68. Estey, E. H. Acute myeloid leukemia: 2019 update on risk-stratification and management. Am. J. Hematol. 93, 1267– 1291 (2018).

69. Shlush, L. I. & Mitchell, A. AML evolution from preleukemia to leukemia and relapse. Best Pract. Res. Clin. Haematol. 28, 81–89 (2015).

70. Jongen-Lavrencic, M. et al. Molecular Minimal Residual Disease in Acute Myeloid Leukemia. N. Engl. J. Med. 378, 1189–1199 (2018).

71. Ding, L. et al. Clonal evolution in re-lapsed acute myeloid leukaemia revealed by whole-genome sequencing. Nature 481, 506–510 (2012).

72. Shlush, L. I. et al. Tracing the origins of relapse in acute myeloid leukaemia to stem cells. Nature 547, 104–108 (2017).

73. Mauvieux, L., Miguet, L. & Fornecker, L. Molecular Minimal Residual Disease in Acute Myeloid Leukemia. The New England journal of medicine 378, 2442–2443

(20)

1

(2018).

74. Bonnet, D. & Dick, J. E. Human acute myeloid leukemia is organized as a hierar-chy that originates from a primitive hema-topoietic cell. Nat. Med. 3, 730–737 (1997).

75. Goardon, N. et al. Coexistence of LMPP-like and GMP-LMPP-like leukemia stem cells in acute myeloid leukemia. Cancer Cell 19, 138–152 (2011).

76. Ishikawa, F. et al. Chemotherapy-resis-tant human AML stem cells home to and engraft within the bone-marrow endosteal region. Nat. Biotechnol. 25, 1315–1321 (2007).

77. Zeijlemaker, W. et al. CD34(+)CD38(-) leukemic stem cell frequency to predict out-come in acute myeloid leukemia. Leukemia 33, 1102–1112 (2019).

78. Chan, S. M. & Majeti, R. Role of DN-MT3A, TET2, and IDH1/2 mutations in pre-leukemic stem cells in acute myeloid leukemia. Int. J. Hematol. 98, 648–657 (2013).

79. Lapidot, T. et al. A cell initiating human acute myeloid leukaemia after transplanta-tion into SCID mice. Nature 367, 645–648 (1994).

80. Wang, K. et al. Patient-derived xeno-transplants can recapitulate the genetic driver landscape of acute leukemias. Leuke-mia 31, 151–158 (2017).

81. Guan, Y., Gerhard, B. & Hogge, D. E. Detection, isolation, and stimulation of qui-escent primitive leukemic progenitor cells from patients with acute myeloid leukemia (AML). Blood 101, 3142–3149 (2003).

82. Terpstra, W. et al. Fluorouracil selec-tively spares acute myeloid leukemia cells with long-term growth abilities in immu-nodeficient mice and in culture. Blood 88, 1944–1950 (1996).

83. Hope, K. J., Jin, L. & Dick, J. E. Acute myeloid leukemia originates from a hierar-chy of leukemic stem cell classes that differ in self-renewal capacity. Nat. Immunol. 5, 738–743 (2004).

84. Lechman, E. R. et al. miR-126 Regulates Distinct Self-Renewal Outcomes in Normal and Malignant Hematopoietic Stem Cells. Cancer Cell 29, 214–228 (2016).

85. Sarry, J.-E. et al. Human acute myeloge-nous leukemia stem cells are rare and het-erogeneous when assayed in NOD/SCID/ IL2Rgammac-deficient mice. J. Clin. Invest. 121, 384–395 (2011).

86. Eppert, K. et al. Stem cell gene expres-sion programs influence clinical outcome in human leukemia. Nat. Med. 17, 1086– 1093 (2011).

87. Taussig, D. C. et al. Leukemia-initiating cells from some acute myeloid leukemia pa-tients with mutated nucleophosmin reside in the CD34(-) fraction. Blood 115, 1976– 1984 (2010).

88. de Boer, B. et al. Prospective Isolation and Characterization of Genetically and Functionally Distinct AML Subclones. Cancer Cell 34, 674-689.e8 (2018).

89. Jin, L. et al. Monoclonal antibody-me-diated targeting of CD123, IL-3 receptor alpha chain, eliminates human acute my-eloid leukemic stem cells. Cell Stem Cell 5, 31–42 (2009).

90. Kikushige, Y. et al. TIM-3 is a promising target to selectively kill acute myeloid leuke-mia stem cells. Cell Stem Cell 7, 708–717 (2010).

91. Askmyr, M. et al. Selective killing of can-didate AML stem cells by antibody targeting of IL1RAP. Blood 121, 3709–3713 (2013).

92. Iwasaki, M., Liedtke, M., Gentles, A. J. & Cleary, M. L. CD93 Marks a Non-Quiescent

Human Leukemia Stem Cell Population and Is Required for Development of MLL-Rear-ranged Acute Myeloid Leukemia. Cell Stem Cell 17, 412–421 (2015).

93. Pabst, C. et al. GPR56 identifies prima-ry human acute myeloid leukemia cells with high repopulating potential in vivo. Blood 127, 2018–2027 (2016).

94. Terskikh, A. V, Miyamoto, T., Chang, C., Diatchenko, L. & Weissman, I. L. Gene ex-pression analysis of purified hematopoietic stem cells and committed progenitors. Blood 102, 94–101 (2003).

95. Zhong, J. F. et al. Gene expression pro-file of murine long-term reconstituting vs. short-term reconstituting hematopoietic stem cells. Proc. Natl. Acad. Sci. U. S. A. 102, 2448–2453 (2005).

96. Korthuis, P. M. et al. CITED2-mediated human hematopoietic stem cell mainte-nance is critical for acute myeloid leukemia. Leukemia 29, 625–635 (2015).

97. Bhattacharya, S. et al. Functional role of p35srj, a novel p300/CBP binding protein, during transactivation by HIF-1. Genes Dev. 13, 64–75 (1999).

98. Freedman, S. J. et al. Structural basis for recruitment of CBP/p300 by hypoxia-induc-ible factor-1 alpha. Proc. Natl. Acad. Sci. U. S. A. (2002). doi:10.1073/pnas.082117899

99. Lou, X. et al. Negative feedback regula-tion of NF-kappaB acregula-tion by CITED2 in the nucleus. J. Immunol. 186, 539–548 (2011).

100. Wu, Z.-Z., Sun, N.-K. & Chao, C. C.-K. Knockdown of CITED2 using short-hairpin RNA sensitizes cancer cells to cisplatin through stabilization of p53 and enhance-ment of p53-dependent apoptosis. J. Cell. Physiol. 226, 2415–2428 (2011).

101. Bamforth, S. D. et al. Cardiac mal-formations, adrenal agenesis, neural crest