The p53 Tumor Suppressor protein Regulates Hematopoietic Stem Cell Fate

Summary

The p53 tumor suppressor protein is a key transcription factor that regulates several signaling pathways involved in the cell’s response to stress. Through stress-induced activation, p53 accumulates and triggers the expression of target genes that protect the genetic integrity of all cells including hematopoietic stem cells (HSCs). These protective mechanisms include cell-cycle arrest, DNA repair, induction of apoptosis, or initiation of senescence. In addition to its function under stress conditions, p53 has important functions during steady-state hematopoiesis, regulating HSC quiescence and self-renewal. In addition, it appears that p53 levels affect HSC competition for the hematopoietic niche, with the less p53 activated HSCs preferentially surviving. The specific genes and precise mechanisms underlying p53’s effects on normal HSCs are slowly being clarified. p53 also plays an important role in leukemia stem cell (LSC) behavior, with p53 loss affecting drug resistance and disease progression. Pharmacologic activation of p53 function could overcome the adverse impact of p53 inactivation in LSCs. Thus, understanding the p53 regulatory mechanisms active in HSCs and LSCs may promote the development of new therapeutic strategies that could eliminate the population of largely quiescent LSCs.

Introduction

p53 was originally isolated as a cellular partner of simian virus 40 (SV-40)-derived tumor antigens (Lane and Crawford, 1979; Linzer and Levine, 1979), and a decade later shown to be an important tumor suppressor (Baker et al., 1989; Donehower et al., 1992). p53 functions as a transcription factor (Bargonetti et al., 1991; el-Deiry et al., 1992; Farmer et al., 1992; Kern et al., 1991), mediating DNA damage responses to a variety of cellular stresses and inducing cell-cycle arrest (Mercer et al., 1990; Scheffner et al., 1990), senescence (Serrano et al., 1997), and apoptosis (Shaw et al., 1992; Yonish-Rouach et al., 1991) in order to maintain genomic instability (Liu et al., 2004). These imply that p53 serves as a guardian of the genome under stress conditions.

In the steady state, p53 activity is strictly restrained through its ubiquitylation and proteasome mediated degradation, involving several E3 ubiquitin ligases, primarily MDM2 (Jones et al., 1995; Montes de Oca Luna et al., 1995; Ringshausen et al., 2006). p53 is stabilized and activated in response to stresses, such as acute genotoxic stress or oncogenic activation (Meek, 2009; Toledo and Wahl, 2006; Vousden and Lane, 2007). However, questions remain as to whether p53 is kept inactive until its expression is induced by cellular stress. Recent studies have revealed an important role for p53 under conditions of apparently normal growth and development, by promoting the survival of slightly damaged cells, through effects on DNA repair (Lassus et al., 1996), or by lowering reactive oxygen species (ROS) levels and reducing DNA damage (Sablina et al., 2005). Moreover, p53 has been shown to play a homeostatic function regulating of hematopoietic stem cell quiescence and self-renewal. We will focus on the role of p53 in HSCs in this review (Figure 1).

Fig 1. Multiple functions of p53 at physiological and stress conditions.

p53 can induce a variety of cellular responses depending on the nature, level and duration of stress signals. When expressed at low level under physiological condition, p53 plays homeostatic roles such as anti-oxidant function and maintenance of stem cell functions. Increasing level of stress augments the level of p53 expression which induces apoptosis, thereby eliminate damaged cells.

Hematopoietic stem cells (HSCs) have the ability to differentiate into all blood cell lineages and to self-renew. Under steady state conditions, most HSCs are quiescent, while a fraction enter the cell cycle, giving rise to progenitors of each lineage, in order to adequately replenish the circulating blood cells each day (Attar and Scadden, 2004). Under stress conditions, HSCs will enter the cell cycle and maintain their cell pool by self-renewal. Cell division will promote the repair of double strand breaks (DSBs) in HSCs via homologous recombination (Passegue et al., 2005). In contrast, quiescent HSCs appear to utilize nonhomologous end joining (NHEJ)-mediated DNA repair, which can be associated with acquisition of genomic rearrangements (Mohrin et al., 2010).

Although p53-null mice show almost normal hematopoiesis (Lotem and Sachs, 1993), many studies have identified roles for p53 in the proliferation, differentiation, apoptosis, and aging of HSCs (Dumble et al., 2007; Kastan et al., 1991; Park et al., 2003; Shounan et al., 1996). Furthermore, recent more detailed analyses of p53-null mice have gradually revealed the important function of p53 in HSCs.

Function of p53 in Steady-state Hematopoiesis

HSC quiescence is critical for preserving a lifelong pool of HSCs that can sustain a highly regenerative hematopoietic system. The functions and dynamics of HSCs are strictly controlled by both HSC-intrinsic and bone marrow microenvironmental mechanisms (Hock et al., 2004; Krosl et al., 2003; Lacorazza et al., 2006; Ling et al., 2004; Wilson and Trumpp, 2006; Zeng et al., 2004) and many proteins and signaling pathways have been implicated in regulating HSC quiescence, including cell-intrinsic transcription factors, such as MEF/ELF4 (Lacorazza et al., 2006), MLL (Jude et al., 2007), GATA-2 (Tipping et al., 2009), Pbx1 (Ficara et al., 2008), FoxO (Miyamoto et al., 2007), and cell cycle regulators, such as p21^cip1/waf1 (Cheng et al., 2000), and Cyclin C (Miyata et al., 2010). Interactions between the bone marrow microenvironment and HSCs, such as Tie2/angiopoietin-1 signaling (Arai et al., 2004), Wnt/Frizzled signaling (Fleming et al., 2008), Thrombopoietin/MPL signaling (Yoshihara et al., 2007), and CXCR4/CXCL12 signaling (Nie et al., 2008) regulate HSC quiescence as do several tumor suppressor genes, including PML, Fbxw7, and PTEN (Ito et al., 2008; Perry and Li, 2008; Thompson et al., 2008; Yilmaz et al., 2006; Zhang et al., 2006).

Our studies of quiescence evolved from characterizing the biological function of an Ets transcription factor, MEF/ELF4, which we initially identified based on its ability to regulate cytokine gene expression in hematopoietic cells (Miyazaki et al., 1996). MEF/ELF4 generally promotes cell growth and functions as an oncogene: MEF/ELF4 can transform NIH3T3 cells and is overexpressed in some human ovarian cancer tumor samples (Yao et al., 2007). We generated MEF/ELF4-null mice and subsequently identified its role in regulating both HSC quiescence and self-renewal (Lacorazza et al., 2006). After discovering that MEF/ELF4 can directly regulate MDM2 expression (Sashida et al., 2009), we showed that MEF/ELF4-null long-term reconstituting HSCs (LT-HSCs) and HSC-enriched Lineage⁻Sca-1⁺ c-kit⁺ (LSK) cells express high levels of p53 (Liu et al., 2009), that could play an important role in HSC physiology.

p53-null mice have a two- to threefold increase in LSK cells and in LT-HSC-enriched SLAM⁺ (CD150⁺ CD48⁻) LSK cells (Akala et al., 2008; Chen et al., 2008; Liu et al., 2009; TeKippe et al., 2003). p53 promotes HSC quiescence and in its absence, HSCs more easily to enter the cell cycle (Liu et al., 2009). However, lack of p53 returned the enhanced stem cell quiescence of MEF/ELF4-null HSCs to normal, indicating that p53 functions to block cell-cycle entry of HSCs and maintain their quiescence status. This does not involve the standard p21-mediated pathway (el-Deiry et al., 1993; Liu et al., 2009).

p53 plays a cytoprotective function within the hematopoietic compartment, mediated by its effects on HSC quiescence, helping to protect HSCs from DNA damage. Furthermore, elevated ROS levels have been shown to limit the life span of HSCs in vivo (Ito et al., 2006) and p53 has been shown to lower ROS levels, thereby reducing DNA damage and the mutational rate (Sablina et al., 2005).

In vivo bone marrow transplantation experiments (Zon, 2008) have shown that p53 negatively regulates HSC self-renewal (Akala et al., 2008; Chen et al., 2008; Liu et al., 2009; TeKippe et al., 2003). p53-null bone marrow cells outcompete wild type bone marrow cells in competitive repopulation assays (Liu et al., 2009; TeKippe et al., 2003), and although multipotent progenitor cells generally lack the ability to self-renew, p16^Ink4a and p19^Arf null multipotent progenitor cells gain the ability to reconstitute long-term hematopoiesis when p53 is absent (Akala et al., 2008). The pathways controlled by these proteins are commonly repressed in the course of oncogenesis, which would allow hematopoietic progenitor cells to gain the ability to self-renew, and become more susceptible to transformation by oncogenic mutations.

Based on these findings, inhibition of p53 activity has been suggested to represent a therapeutic strategy capable of amplifying the HSC pool. Small molecule inhibitors of p53, the pififthrins (PFTs) have been identified, that can suppress genotoxic stress-induced p53-dependent apoptosis, thereby protecting mice from otherwise lethal doses of irradiation or chemotherapy (Komarov et al., 1999). A recent study showed that PFTs can stimulate HSC proliferation in vitro and in vivo; the amplified HSCs are functional in bone marrow transplantation experiments, without promoting tumor development (Marion et al., 2009), which suggests that transient exposure to p53 inhibitors could be used to stimulate HSC self-renewal or survival.

As p53 has been shown to negatively regulate neural stem cell self-renewal (Meletis et al., 2006) and mammary stem cell self-renewal (Cicalese et al., 2009), such strategies may not be limited to the hematopoietic compartment. In addition, disruption of the p53 network enhances the generation of induced pluripotent stem (iPS) cells, which are capable of self-renewal and of giving rise to multiple types of differentiated cells (Hong et al., 2009; Kawamura et al., 2009; Li et al., 2009; Marion et al., 2009; Utikal et al., 2009). Thus, p53 impedes somatic cell reprogramming, yet another p53 function in addition to its role as a regulator of cellular stress responses.

What are the p53 target genes relevant to the steady-state behavior of HSCs? While the cyclin-dependent kinase inhibitor p21 is a major target of p53, p21 has been shown to play a modest role in regulating HSC quiescence, in a mixed mouse strain background (Cheng et al., 2000), but not in a pure C57BL/6 background (van Os et al., 2007). In our study of p53 function, p21 did not appear to be involved in regulating the quiescence of either wild type or MEF/ELF4-null HSCs. By performing transcript profiling of LSK cells isolated from p53-null and p53, MEF/ELF4-double null mice, we identified Gfi-1 and Necdin as two direct p53 target genes that could regulate HSC quiescence (Liu et al., 2009).

Gfi-1 (Growth factor independent-1) is a SNAG-domain-containing zinc-finger transcriptional repressor that promotes the proliferation of T cells and sometimes functions as a cooperating oncogene in lymphoid cells (Gilks et al., 1993; Zhu et al., 2002). Gfi-1 has been shown to restrict HSC proliferation and preserve HSC functional integrity. Gfi-1 null HSCs demonstrate excessive cycling status and impaired self renewal, shown by in vivo competitive repopulation assays (Hock et al., 2004; Zeng et al., 2004).

Necdin is a growth-suppressing protein first identified in post-mitotic neurons (Maruyama et al., 1991). The gene encoding Necdin is one of the several genes which are deleted in individuals with Prader-Willi syndrome (Jay et al., 1997; MacDonald and Wevrick, 1997), a disorder associated with a mildly increased risk of myeloid leukemia (Davies et al., 2003). Like the retinoblastoma protein, Necdin interacts with multiple cell cycle promoting proteins, such as simian virus 40 large T antigen, adenovirus E1A, and transcription factor E2F1 (Hu et al., 2003; Taniura et al., 2005; Taniura et al., 1999). Necdin is highly expressed in LT-HSCs (Forsberg et al., 2005; Liu et al., 2009). Downregulation of Necdin diminished HSC quiescence, whereas its upregulation increased HSC quiescence, identifying its role as a rheostat controlling HSC quiescence (Liu et al., 2009). Necdin and p53 has been shown to inhibit cell growth in an additive manner (Taniura et al., 1999), suggesting the presence of a positive feedback loop that may control quiescent HSCs. Recently our analysis using Necdin-null HSCs shows that Necdin protects HSCs from genotoxic stress (Liu and Asai et al., in submission) (Figure 2).

Fig 2. Role of p53 in hematopoietic stem cells.

(A) Under normal condition, p53 has been shown to regulate hematopoietic stem cell quiescence mediated by Gfi-1 and Necdin. It is likely to maintain self-renewal by yet unknown mechanism. (B) Increasing level of stress triggers ATM activation which will increase p53 activation. By downstream mediators such as p21, Puma and Slug, p53 triggers cell cycle arrest, DNA repair or apoptosis.

Function of p53 in Stress Hematopoiesis

Hematopoiesis is perturbed by genotoxic stresses such as γ-irradiation or chemotherapy and in response to high levels of genotoxic stress, HSCs and progenitor cells undergo apoptosis, leading to severe anemia, bleeding, and infections. Some sensor proteins (Mre11, Rad50, Nbs1, Rad9, Rad1, Rad17, Hus1) bind damaged DNA and relay the DNA damage signal to transducer proteins such as ATM and ATR, which phosphorylate several effector kinases including Chk1 and Chk2. These effector kinases, together with ATM, phosphorylate p53 and other target proteins (Zhou and Elledge, 2000). Once activated, p53 induces growth arrest and DNA repair, or apoptosis depending on many variables (Balint and Vousden, 2001).

p53 is a critical regulator of apoptosis in HSCs. Lack of Bmi-1 in the hematopoietic compartment leads to impaired self-renewal capacity and bone marrow failure due to p19^Arf accumulation, which triggers p53-dependent cell death (Park et al., 2003). Similarly, inactivation of Fbxw7, an SCF-type ubiquitin ligase complex subunit, causes premature loss of HSCs due to active cell cycling and p53-dependent apoptosis (Matsuoka et al., 2008).

While p53 initiates apoptosis within HSCs, inhibition of p53 function leads to resistance to the apoptosis, promoting effects of ionizing irradiation (Komarov et al., 1999; Komarova et al., 2004; Leonova et al., 2010; Westphal et al., 1997) or DNA-damaging chemotherapeutic agents (Lotem and Sachs, 1993). The transcription-independent pro-apoptotic effects of p53 are triggered by the movement of p53 to the mitochondria (Erster et al., 2004), where it can react with Bcl-xL and Bcl-2 and antagonize their mitochondrial membrane stabilizing effects (Mihara et al., 2003). Mitochondrial p53 also increases permeabilization of the outer mitochondrial membrane, leading to the release of cytochrome c (Moll et al., 2005), and it directly promotes the pro-apoptotic functions of Bak (Leu et al., 2004).

The precise p53 role in HSCs under genotoxic stress is not fully understood; using γ-H2AX as an indicator of DNA damage, we demonstrated that p53, MEF/ELF4 double-null HSCs have more foci than MEF-null HSCs 3 hours following ionizing irradiation (Liu et al., 2009), which suggests that p53 facilitates DNA damage repair in HSCs. The transcriptional regulatory functions of p53 control the expression of its pro-apoptotic target genes, such as Puma (Han et al., 2001; Nakano and Vousden, 2001; Yu et al., 2001), Noxa (Oda et al., 2000), Bid (Mandal et al., 2008), and Bax (Miyashita and Reed, 1995) (Table 1). Puma (p53 upregulated mediator of apoptosis) is one of the most potent p53 target genes that induce apoptosis in HSCs under conditions of genotoxic stress. Puma has been reported to be essential for hematopoietic cell death triggered by ionizing radiation and cytokine withdrawal among others (Jeffers et al., 2003; Villunger et al., 2003). The function of Puma in determining the sensitivity of HSCs to high-dose ionizing irradiation has been characterized (Shao et al., 2010; Yu et al., 2010); in the absence of Puma, HSCs are highly resistant to ionizing irradiation in a cell autonomous manner. Puma null HSCs also show enhanced quiescence and more efficient DNA repair than wild type HSCs (Yu et al., 2010), as do Puma-null hematopoietic progenitor cells (Shao et al., 2010). In contrast to Puma, activation of Slug, a transcriptional repressor induced by p53 upon irradiation, protects hematopoietic progenitors from apoptosis by repressing the transcription of Puma (Wu et al., 2005). Thus, promoting Slug function or blocking pro-apoptotic p53 targets, such as Puma, may be potential strategies to protect HSCs from the myelosuppressive effects of intensive radiotherapy or chemotherapy.

Table 1.

List of p53 target genes that mediate p53 activities.

Cellular Effect	Target genes	References
Cell cycle arrest	p21	(el-Deiry et al., 1993)
	14-3-3-sigma	(Chan et al., 1999; Weber et al., 2002)
	Cdc25	(Resnick-Silverman et al., 1998)
DNA repair	DDIT4	(Ellisen et al., 2002)
	Gadd45	(Canman et al., 1995)
Apoptosis	Puma	(Han et al., 2001; Nakano and Vousden, 2001; Yu et al., 2001)
	Noxa	(Oda et al., 2000)
	Slug	(Inoue et al., 2002)
	Bid	(Mandal et al., 2008)
	Bax	(Miyashita and Reed, 1995)
Stem cell quiescence	Gfi1	(Liu et al., 2009)
	Necdin	(Liu et al., 2009)

In response to moderate levels of genotoxic stress, p53 can initiate the repair of damaged HSCs by triggering cell cycle arrest and activating the DNA repair machinery. Among the p53 target genes implicated in these processes are DDB2 (Takimoto et al., 2002), DDIT4 (Ellisen et al., 2002), and Gadd45 (Canman et al., 1995), which are involved in DNA damage repair, and p21 (CDKN1A) (el-Deiry et al., 1993), 14-3-3-sigma (SFN) (Chan et al., 1999; Weber et al., 2002), and Cdc25 (Resnick-Silverman et al., 1998), which are involved in inducing cell cycle arrest (Table 1).

Function of p53 in HSC aging

Aging can be defined as a progressive functional decline associated with an increasing risk of mortality over time (Sharpless and DePinho, 2007). Advancing age is accompanied by a number of pathophysiological changes in the hematopoietic system, possibly reflecting loss of homeostatic control of hematopoietic stem and progenitor cell (HSPC) behavior (Rossi et al., 2008). Accumulation of DNA damage is thought to be a physiological consequence of HSC aging, which may contribute to the diminished capacity of older HSCs to return to their basal state after exposure to acute stress or injury (Rossi et al., 2007)

The relationship between stem cell aging and tumor suppressor gene (TSG) expression has received much attention recently. Inactivation of tumor suppressors contributes to the development of cancer, while TSG activation contributes to HSC aging. Increased expression of p16^INK4a has been proposed to be one of the principal biomarkers of aging (Krishnamurthy et al., 2004). p16^INK4a is elevated in HSCs isolated from older mice, and the HSC repopulating defects and apoptosis of older HSCs are less apparent in older p16^INK4a-null mice (Janzen et al., 2006).

p53 has also been implicated in HSC aging (Dumble et al., 2007; Tyner et al., 2002). Knock-in mice expressing a truncated but active form of p53 (p53^+/m) exhibit an early aging phenotype with fewer proliferating HSCs compared with older wild-type mice (Maier et al., 2004; Tyner et al., 2002). p53^+/− mice, that have slightly reduced p53 levels, show the opposite phenotype. In addition, p53^+/m HSCs have reduced engraftment capacity, compared to wild-type or p53-null HSCs (Dumble et al., 2007). In contrast, mice carrying three functional copies of the p53 gene (super-p53) have a normal lifespan but some evidence of early aging (Garcia-Cao et al., 2006). It is possible that the constitutively high activity of p53 in p53^+/m mice accelerates aging, while the extra copy of p53 in the super-p53 mice is subject to normal regulation, and therefore p53 functional activity is unchanged.

p53-mediated apoptosis may be involved in the physiological regulation of HSC population size and function with aging. Mice carrying a hypermorphic form of the Rad50 DNA repair protein exhibit precipitous bone marrow failure due to enhanced signaling through an ATM-Chk2-p53-dependent apoptotic pathway (Bender et al., 2002; Morales et al., 2005). On the other hand, mice that overexpress BCL-2 within the hematopoietic compartment show an expanded HSC pool and improved HSC repopulating capacity (Domen et al., 2000), similar to the p53-null mice (TeKippe et al., 2003). These studies illustrate the tenuous relationship between tumor suppression and HSC aging (Gatza et al., 2007). Although manipulating the p53 pathway could delay aging, this could occur at the expanse of a marked increased risk of developing cancer.

p53-mediated HSC competition

Cell competition is an important aspect of many homeostatic processes, and two groups recently identified a role for p53 in the competition among HSPCs (Bondar and Medzhitov, 2010; Marusyk et al., 2010). In competitive repopulation experiments, unirradiated HSCs were shown to outcompete HSCs that were treated with a low dose of ionizing irradiation (1 Gy), dependent on p53 (Bondar and Medzhitov, 2010). Similarly, wild type HSCs outcompete MDM2-null HSCs in the absence of external stress, and low- or middle-dose irradiation enhances this competitive advantage, suggesting that the p53 level itself is critical for HSC competition within the stem cell niche. This p53-dependent HSC competition is related to cell proliferation, and not to a higher rate of apoptosis in the outcompeted cells. It is also mediated by the expression of growth arrest and senescence-related genes in the outcompeted cells, such as p16 (Bondar and Medzhitov, 2010; Janzen et al., 2006).

It is problematic that p53-mutated HSCs could potentially dominate the stem cell niche and outcompete normal HSCs. As p53-null HSCs are more proliferative and less quiescent than normal HSCs (Liu et al., 2009), it is possible that mechanisms exist to detect p53-mutated HSCs and limit their inherent advantage (Green, 2010; Marusyk et al., 2010). The presence of p53 may help cells tolerate stress via metabolic processes (Vousden and Ryan, 2009), or overall patterns of gene expression may be optimized during p53 regulated G0 or G1 phases of the cell cycle. Defining how wild-type HSCs sustain the advantage over p53-mutated HSCs under various conditions will help us understand leukemogenesis and how regulating p53 activity could affect disease initiation and progression.

Function of p53 in Leukemia Stem Cells

p53 is the most common gene targeted for inactivation by deletion and/or mutation in human tumors (Harris and Hollstein, 1993). p53 mutations are much less frequent in leukemia than in other solid tumors (Peller and Rotter, 2003), occurring in less than 10% of de novo acute myeloid leukemia (AML) (Fenaux et al., 1992; Greenblatt et al., 1994; Imamura et al., 1994; Slingerland et al., 1991). Patients with p53 mutations are generally resistant to chemotherapy and have relatively short survival (Haferlach et al., 2008; Nahi et al., 2008; Wattel et al., 1994). p53 mutations or deletions are thought to be an independent prognostic factor for survival and they are more common in elder patients with complex karyotypes (Nakano et al., 2000; Stirewalt et al., 2001). Similar observations have been made in patients with myelodysplastic syndrome (MDS) (Padua et al., 1998). However, in patients who develop AML or MDS following exposure to alkylating agents, the incidence of p53 mutation increases to 30%, and is associated with increased resistance to chemotherapy and shorter overall survival (Christiansen et al., 2001). In addition, p53 mutations are found in 25% of blast phase chronic myelogenous leukemia (CML) (Feinstein et al., 1991; Kelman et al., 1989), generally accompanying disease progression (Ahuja et al., 1989). Thus, p53 mutations occur late in the course of these diseases and promote drug resistance.

Leukemia stem cells (LSCs) are thought to be resistant to various types of therapy, because they are in a relatively quiescent state (Komarova and Wodarz, 2007). Leukemia relapse may occur because most therapies eliminate proliferating cells, but not the quiescent cells, such as the LSCs, that can reinitiate the disease after some latency period (Holtz et al., 2007). New therapeutic approaches that can target LSCs will help eradicate acute leukemia and given the effect of p53 on HSC quiescence, understanding how p53 promotes quiescence may lead to therapeutic strategies that could eliminate the largely quiescent LSCs (Liu et al., 2009).

p53 mutations are found in a minority of human leukemias, and if p53 signaling is intact, then disrupting the p53-MDM2 interact could lead to the induction of apoptosis in the leukemic cells (Saha et al., 2010; Shangary and Wang, 2008; Tovar et al., 2006). One of the most promising p53 activating agents is nutlin (and its active form, nutlin-3a or nutlin-3), which is a small-molecule inhibitor of MDM2 (Vassilev et al., 2004). Recent studies demonstrated that nutlin-3, used alone or in combination with other drugs, effectively induces apoptosis in AML (Kojima et al., 2005; Kojima et al., 2008; Zhang et al., 2010). While its precise function on LSCs has not been clarified, gene expression analyses suggest that the p53-regulated genes, Bax, Gadd45, and possibly p21^cip1/waf1, are involved in triggering LSC apoptosis (Guzman et al., 2002). Other approaches to target LSCs include the combination of a proteasome inhibitor (MG-132) with the anthracycline idarubicin, which induces p53-dependent apoptosis of LSCs while leaving normal HSCs viable (Guzman et al., 2002; Tergaonkar et al., 2002) In addition, parthenolide, a naturally occurring small molecule, induces robust apoptosis in LSCs, through similar mechanisms (Guzman et al., 2005; Guzman et al., 2007). Therefore, p53 may play a different role in inducing apoptosis of LSCs, compared to HSCs.

p53 has been shown to be involved in some mouse models of leukemia. Loss of p53 dramatically accelerates AML1/ETO9a-driven, but not MLL/ENL-driven leukemogenesis (Zuber et al., 2009). Loss of p53 also cooperates with overexpression of sPRDM16, the short isoform of PR domain containing 16, in leukemogenesis promoting LSC self renewal (Shing et al., 2007), and it accelerates the development of acute lymphoblastic leukemia (ALL) in Fbxw7 null mice (Matsuoka et al., 2008).

Recently, a mouse model of the human 5q- syndrome was reported (Barlow et al., 2010). These mice have macrocytic anemia and dysplastic bone marrow features, which are also observed in human 5q- MDS; the mice show an increase in p53-positive cells in the bone marrow with elevated apoptosis and defective hematopoietic progenitor development. Loss of p53 rescues the defect in hematopoietic progenitor development, suggesting that p53-dependent mechanisms underlie the pathogenesis of the 5q- syndrome (Barlow et al., 2010).

Summary and Consideration for Future Investigation

The p53 tumor suppressor protein is a key transcription factor that regulates signaling pathways controlling the cellular stress response. Through stress-induced activation, p53 accumulates and mediates the expression of genes that protect the genetic integrity of HSCs. During steady-state hematopoiesis, basal-level p53 activity regulates HSC quiescence and self-renewal. Inhibition of p53 may be clinically applicable to amplify the HSC pool, and recent studies show that the level of p53 is critical for HSC competition within the hematological niche, allowing the least damaged HSCs to survive. The target genes and precise mechanisms underlying basal p53 functional activity will be clarified in the future. On the other hand, p53 loss is related to decreased apoptosis, increased drug resistance, and disease progression of leukemic cells. Activation of p53 function represents a positive strategy to overcome the adverse impact of p53 inactivation in LSCs. Understanding the p53 regulatory mechanisms active in HSCs vs LSCs will shed light on new therapeutic strategies that could eliminate the largely quiescent LSCs.