The transcription factor MEF/ELF4 regulates the quiescence of
primitive hematopoietic cells
H. Daniel Lacorazza,1,2,* Takeshi Yamada,2Yan Liu,1Yasuhiko Miyata,1Mariela Sivina,2Juliana Nunes,1
and Stephen D. Nimer1,*
1Molecular Pharmacology and Chemistry Program, Sloan-Kettering Institute, Memorial Sloan-Kettering Cancer Center,
1275 York Avenue, New York, New York 10021
2Department of Pathology, Baylor College of Medicine, 6621 Fannin Street, MC 1-2261, Houston, Texas 77030
*Correspondence: firstname.lastname@example.org (S.D.N.); email@example.com (H.D.L.)
transcription factor known as MEF (or ELF4), which is targeted by the t(X;21)(q26;q22) in acute myelogenous leukemia, reg-
ulates the proliferation of primitive hematopoietic progenitor cells at steady state, controlling their quiescence. Mef null
HSCs display increased residence in G0with reduced 5-bromodeoxyuridine incorporation in vivo and impaired cytokine-
driven proliferation in vitro. Due to their increased HSC quiescence, Mef null mice are relatively resistant to the myelosup-
pressive effects of chemotherapy and radiation. Thus, MEF plays an important role in the decision of stem/primitive progen-
itor cells to divide or remain quiescent by regulating their entry to the cell cycle.
uously generate progeny that maintain a steady flow of periph-
eral blood cells. HSCs can remain quiescent, enter senescence,
die, self-renew, or differentiate into multiple lineages. Adult
HSCs are maintained quiescent in bone marrow niches where
osteoblasts provide self-renewal signals, mainly BMP and
Notch ligands (Calvi et al., 2003; Zhang et al., 2003). However,
the downstream molecules involved in this process are largely
unknown. At steady state, cytokines and chemokines drive
a small percentage of HSCs to self-renew in order to maintain
a constant number of stem cells. Following bone marrow abla-
tion, by cytotoxic agents or radiation, stem cells are recruited
from their quiescent niche to promote the rapid reconstitution
sential to protect the HSC pool from cell cycle-dependent injury
(e.g., from toxins including chemotherapy) and from acquiring
mutations during numerous rounds of replication. At the same
time, the proper expansion and differentiation of HSCs is crucial
for basal hematopoietic cell production and for reconstitution
following myelotoxic injury.
A delicate balance between dormancy and cell cycle entry
must take place to ensure maintenance of stem cell numbers
and adequate production of mature blood lineages. These
events need to be tightly regulated, particularly during the tran-
sition of HSCs from G0to G1(from a quiescent to a cycling and
potentially differentiating state). Most research has focused on
mechanisms of self-renewal, and there is little information on
the regulation of stem cell dormancy and its reentry into the
cell cycle. The ability to expand HSCs would have clear clinical
applications to stem cell transplantation and the correction of
genetic disorders via gene transfer. Conversely, altering HSC
covery after bone marrow transplantation or chemotherapy.
Furthermore, understanding how normal HSCs regulate quies-
cence and entry into the cell cycle will set the stage for defining
these properties in leukemic stem cells.
MEF, also known as ELF4, is a member of the ETS family of
winged helix-turn-helix transcription factors (Lacorazza and
Nimer, 2003; Mao et al., 1999; Miyazaki et al., 1996, 2001). Re-
cent studies have implicated MEF in tumorigenesis. The MEF
gene is located on Xq26, within a region of LOH in advanced
ovarian and breast carcinomas (Choi et al., 1997, 1998; Piao
and Malkhosyan, 2002). While the MEF gene has tumor sup-
pressor gene activity in lung carcinoma cell lines (Seki et al.,
roviral mutagenesis studies using the Em-Myc Pim12/2Pim22/2
S I G N I F I C A N C E
The normal balance between the self-renewal and differentiation of HSCs is perturbed in acute leukemia. Quiescence is an alternative
state for HSCs, and several genes that encode cell cycle regulatory proteins or transcriptional regulators have been shown to regulate
of HSCs by facilitating cell cycle entry. MEF null hematopoietic cells demonstrate resistance to myelotoxic injury, conferring recipient
mice with enhanced hematological recovery following chemotherapy or radiation exposure. Treatments that decrease MEF expres-
sion could potentially improve hematopoietic recovery in cancer patients undergoing myeloablative treatments. However, quiescent
leukemic stem cells could use similar mechanisms to escape from the cytotoxic effects of chemotherapy.
A R T I C L E
CANCER CELL 9, 175–187, MARCH 2006 ª2006 ELSEVIER INC.DOI 10.1016/j.ccr.2006.02.017175
and cdkn2a2/2murine models (Akagi et al., 2004; Suzuki et al.,
2002), and in MSCV-Sox4 virus-induced leukemia as well
(Du et al., 2005). Three groups have reported that MEF expres-
sion is repressed by the AML1-ETO and PML-RARa leukemia-
associated fusion proteins, suggesting a role for dysregulation
of MEF expression in the pathogenesis of acute myeloid leuke-
mias (AML) (Alcalay et al., 2003; Muller-Tidow et al., 2004; Park
et al., 2003). An analysis of MEF levels in AML patient samples
showed that MEF RNA levels are particularly low in FAB M2
and M3 AML cells, which contain AML1-ETO and PML-RARa,
ples (Fukushima et al., 2003). The recent identification of an
ELF4-ERG fusion transcript generated by the t(X;21)(q26;q22)
in a patient with AML definitively endorses the involvement of
MEF (ELF4) in hematological malignancies (Moore et al., 2005).
Given the potential importance of MEF downregulation in hu-
man leukemic cells, we evaluated the behavior of hematopoietic
stem and progenitor cells in Mef-deficient mice. We found that
the bone marrow of Mef2/2mice contains a higher fraction of
HSCs (Lin2Sca-1+c-kit+Flt32CD342cells), with more ‘‘side
population’’ (SP) cells and more cobblestone area-forming cells
(CAFCs). Mef null HSCs are more quiescent than normal at
in vitro. Nonetheless, Mef null HSCs can reconstitute an ablated
host and compete with wild-type HSCs in secondary trans-
plants. This quiescence confers Mef2/2mice, and wild-type
mice transplanted with Mef2/2bone marrow cells, with clinically
important protection from myelotoxic drugs (such as 5-FU and
busulfan) and from radiation. The low levels of MEF expression
that are seen in certain subtypes of AML may play an important
role in AML pathogenesis by altering the growth properties of
leukemic stem cells similarly to that observed in Mef null HSCs.
Mef regulates the proliferation of hematopoietic
Mef RNA is expressed in lymphoid and myeloid cells including
Lin2Sca-1+c-kit+(LSK) cells, a population enriched in HSCs
line). As this transcription factor is also involved in lymphoid cell
development (Lacorazza et al., 2002) and is regulated during the
plays a critical role in the behavior of HSCs in the bone marrow.
We first examined the number of primitive hematopoietic cells
in the bone marrow of Mef2/2compared to wild-type controls.
We used the CAFC assay and the long-term culture-initiating
cell (LTC-IC) assay, since they correlate with the in vivo repopu-
lating potential of primitive hematopoietic progenitors (Ploe-
macher et al., 1991). Mef2/2bone marrow cells contain at least
5-fold more CAFCs compared to wild-type bone marrow (Fig-
ure 1A). The increased frequency of primitive progenitors was
also evident in the LTC-IC assay (Figure 1B). In contrast, meth-
ylcellulose colony-forming unit (CFU) assays, performed to
quantify myeloid progenitor cells, showed no significant differ-
ences in the number or type of colonies formed (Figure 1C and
data not shown), aresultthatisconsistentwith the normalblood
counts observed in Mef2/2mice (Table S1). Taken together,
these data indicate the accumulation of primitive hematopoietic
cells rather than committed progenitors in the absence of Mef.
To further address the frequency of primitive progenitors in
Mef null bone marrow, we performed serial replating studies
using methylcellulose cultures (Figure 1D). Wild-type BMMCs
depletion of HSCs; in contrast, we saw a 8- to 10-fold higher fre-
quency of primitive progenitors using Mef2/2bone marrow cells
creased in Mef2/2bone marrow cells at steady
A: The steady-state level of bone marrow primi-
tive progenitor cells was evaluated using the
cobblestone area-forming cell assay, scoring
colonies at week 5 (****p < 0.0001).
B: LTC-IC was also used to enumerate primitive
progenitor cells, using limiting dilutions of BMMCs
that were cultured for 5 weeks on MS5 stroma
and then cultured on methylcellulose for the
C: Myeloid progenitors were quantified by meth-
ylcellulose culture using BMMCs from Mef+/+or
Mef2/2mice (**p < 0.01).
D: Methylcellulose cultures were serially replated
weekly, for 5 weeks.
E: Mef2/2BMMCs were transduced with retrovi-
ruses containing either the MEF cDNA or the
empty vector; GFP-positive cells were purified
and serially replated on methylcellulose culture
for 3 weeks.
Data are shown as mean 6 standard deviation
(n = 5).
A R T I C L E
CANCER CELL MARCH 2006
ted progenitors indicates their primitiveness and ability to re-
tain stemness in culture. This replating was seen using purified
Mef2/2Lin2Sca-1+c-kit+cells and not Lin2Sca-12c-kit+cells
poietic cells rather than committed progenitors have aberrant
growth properties. To demonstrate that this effect is directly
linked to the absence of Mef, we reintroduced Mef into Mef2/2
BMMCs using retroviruses that express MEF-IRES-GFP
(MIGR1-MEF) or the empty IRES-GFP (MIGR1) and plated puri-
Mef abrogated the ‘‘preservation’’ of early progenitor cells (Fig-
ure 1E), suggesting that restoring Mef expression restores the
normal level of proliferation of Mef null hematopoietic progenitor
cells, leading to their exhaustion in this in vitro assay.
These in vitro assays nicely correlate with the immunopheno-
typic characterization of various hematopoietic subsets within
the bone marrow. Mef2/2bone marrow contains three to four
based on their increased percentage in a normocellular bone
marrow (Figure 2B). We detected more long-term HSC (LT-
HSC) and short-term HSC (ST-HSC) based on Flt3 and CD34
expression (Flt32CD342LSK cells = LT-HSCs and Flt32CD34+
Figure 2. Increased primitive progenitor popula-
tions in the bone marrow of Mef-deficient mice
A: A greater percentage of Lin2Thy1lowSca-1+
c-kit+cells are found in Mef2/2bone marrow
cells than in Mef+/+bone marrow cells (represen-
tative dot plots are shown).
B: There is no difference in the bone marrow cel-
lularity of wild-type and Mef-deficient mice (two
femurs and two tibias were analyzed for six wild-
type and six Mef2/2mice; p = 0.130). Data are
shown as mean 6 standard deviation.
C: Expression of Flt3 and CD34 in LSK cells from
wild-type and Mef2/2mice. Mef null LSK cells
(3-fold) and Flt32CD34+(2.3-fold) LSK cells, which
correspond to LT-HSC and ST-HSC. There is a de-
crease in the transition from ST-HSC to MPP
D: Side population (SP) cells were identified by
Hoechst 33342 staining and the use of blue and
red filters (Goodell et al., 1996). The identification
of SP cells was confirmed by their disappear-
ance in the presence of Verapamil (50 mM).
Two gates (SP-low and SP-intermediate) and
the fold increase (Mef null/Mef wild-type) are
E: SP cells contained within the Lin2Sca-1+c-kit+
population are also shown (Mef null cells show
a 2.9-fold increase over wild-type).
A R T I C L E
CANCER CELL MARCH 2006177
cells than in wild-type cells (Figure 2C), with a concomitant de-
crease in Flt3+CD34+
LSK cells (multipotent progenitors
[MPPs]). Although this suggests a developmental holdup in the
maturation of HSCs to MPPs, stem cell function is not signifi-
cantly affected, as Mef null LSK cells are able to reconstitute
host hematopoiesis in a myeloablative transplant model (data
We used another assay to identify murineHSCs, namely dual-
wavelength flow cytometric detection of cells stained with
Hoechst 33342, which is based on the ability of HSCs to efflux
thisdye(Goodell etal.,1996).The Hoechst-stained SP identifies
HSCs with long-term repopulating activity by a functional fea-
turerather thanbytheexpression ofcellsurfaceproteins.Asex-
pected, Mef-deficient BMMCs contain a 5-fold greater number
of SP cells (Figure 2D), even gating on the highest dye efflux
activity (SPlow) (0.002% versus 0.011% for +/+ and 2/2, re-
spectively). This population, SPlowcells, is known to be nearly
homogenous for quiescent, long-term HSCs (Camargo et al.,
2006). Since SP cells could include some committed progeni-
and observed a consistent increase (at least 2.9-fold) in the
number of SP-LSK cells (Figure 2E). To validate our SP cell de-
tection, we also performed this assay in the presence of verap-
amil and saw disappearance of the SP cells (Figure 2D). Based
on all these assays, we conclude that in the absence of MEF
hematopoietic stem/primitive progenitor cells accumulate in
the bone marrow, primarily LT-HSCs and ST-HSCs.
Bone marrow transplant reveals a functional defect
in the cell cycle
We used a competitive repopulating assay and the distinct con-
genic markers (CD45.1 and CD45.2) to define the long-term re-
constituting ability of Mef2/2BMMCs and to enumerate stem
cell numbers. By injecting BMMCs from Mef2/2(C57BL/6,
CD45.2+) and Mef+/+(B6.SJL, CD45.1+) into B6/SJL.F1 (which
express CD45.1+and CD45.2+) at different input ratios (contain-
ing 0%, 20%, 50%, 80%, or 100% Mef2/2BMMCs) we can
tution. We observed the same magnitude increase in LSK cells
in the bone marrow of B6/SJL.F1 mice transplanted with
100% Mef2/2BMMCsasinthe Mefnullmice(Figure S3),further
cell accumulation. The percentage of LSK cells in the bone mar-
row of these mice increases as the percent input of Mef2/2
BMMCs increases (Figure S3), yet we did not observe a greater
contribution of Mef2/2BMMCs to the steady-state circulating
lymphoid cells (Figure S3) or myeloid cells (Figure 3) 5 months
after hematologic reconstitution. This suggests that even
though Mef2/2BMMCs contain more primitive progenitors
Figure 3. Transplantation studies to assess stem
A: The contribution of each donor (Mef+/+versus
Mef2/2) to the peripheral blood cell populations
5 months after transplantation is shown for the
competitive reconstitution study (far left). The
same mice were then sublethally irradiated
(450 rads), and the source of hematopoietic re-
covery was monitored in the peripheral blood
cells was also evaluated in the bone marrow of
recipient mice 4 months postradiation (9 months
posttransplant). Data are shown as mean 6
standard deviation (n = 4).
B: Primary recipients were transplanted with a
mixture of wild-type and Mef2/2BMMCs (1 3
105BMMCs each). After reconstitution, PBCs
showed a similar contribution of each donor.
Wild-type (CD45.1+) and Mef2/2(CD45.2+) cells
were then purified from the bone marrow of
these mice and then transplanted into second-
ary recipients in a ratio 1:1. After hematologic
reconstitution, there is a predominance of Mef
null-derived cells in peripheral blood.
A R T I C L E
CANCER CELL MARCH 2006
they do not have a repopulating advantage over wild-type pro-
genitors in a competitive situation.
The greater contribution of Mef2/2HSCs following radiation
could be due to either increased survival of Mef2/2cells during
the radiation or to enhanced proliferation in response to stress.
We did not observe a competitive advantage over wild-type
cells in primary competitive repopulation (CR) experiments, de-
spite using different limiting doses (data not shown). We believe
this illustrates a limitation of in vivo CR experiments, namely the
use of this assay to determine the number of functionally abnor-
rified Mef+/+(CD45.1+) and Mef2/2(CD45.2+) cells, obtained
from mice previously transplanted with a mixture of wild-type:
knockout (1:1), and used these cells in a secondary CR assay,
the vast majority of PBCs were derived from the Mef2/2bone
marrow cells, even though the donor mice had approximately
50:50 peripheral blood mixed chimerism at the time of the har-
vest (Figure 3B). Mef2/2bone marrow cells clearly out-compete
wild-type cells that are compromised from the primary trans-
plant. Thus, although Mef2/2BMMCs contain more HSCs with
repopulating capacity, this can only be seen in a secondary
CR, due to their cell cycle deregulation.
chimera mice,animalstransplanted formorethan5monthswith
both Mef+/+and Mef2/2cells were sublethally irradiated. The
vast majority of the peripheral blood cells were at that time de-
rived from the Mef2/2progenitor cells (Figure 3A). Most impor-
tantly, 4 months following the radiation treatment (and 9 months
after the initial transplant) nearly all the bone marrow cells of the
transplanted mice were derived from Mef null stem cells (Fig-
ure 3A). This indicates that the proportion of Mef2/2to wild-
type HSCs in the bone marrow of transplanted mice was higher
than the ratio of BMMCs initially transplanted and points to
a proliferative defect at steady state that is overcome by the
response to stress (i.e., radiation).
Mef regulates the quiescence of primitive
The accumulation of primitive progenitor cells in Mef2/2mice
could be cell intrinsic and due to enhanced self-renewal, quies-
cence, or survival. Alternatively, faulty environmental cues could
of Mef loss. CFSE studies showed normal homing ability; we
also observed nearly normal G-CSF induced PBPC migration
(data not shown). We next performed serial bone marrow trans-
plants and saw no improvement in the survival of mice that re-
ceived bone marrow cells serially transplanted four times
(Figure S4). This implies that enhanced stem cell self-renewal
may not be the primary cause of the stem cell accumulation
that occurs in Mef2/2mice, even though we did find greater
CAFC and CFC activity in mice serially transplanted twice with
purified Mef null LSK cells, indicating preservation of stem cell
properties (Figure S4). The absence of a survival advantage in
progenitors, suggests that Mef2/2HSCs have impaired prolifer-
To assesstheability ofMefnullHSCtoenterthecell cycle,we
measured DNA content in purified LSK cells incubated for 24 hr
in the presence and absence of SCF, IL-3, and IL-6. The percent
of Mef2/2LSK cells in S phase increased by only 24%, whereas
wild-type LSK cells showed at least a 52% increase (Figure 4A).
Furthermore, by monitoring the growth of individual LSK cells,
placed one cell per well into 96-well plates, we showed that
Mef-deficient cells grow more slowly than wild-type cells (ten
representative clones per genotype are shown in Figure 4B),
demonstrating impaired cytokine responsiveness at the single
movement of stem cells into the cell cycle, perhaps the transi-
tion from G0to G1. We used multiparameter flow cytometry (Py-
ronin Y and Hoechst 33342 to monitor RNA and DNA content,
respectively) to distinguish between G0and G1, as quiescent
primitive progenitors are defined as cells with low Pyronin Y
centage of cells in G0within the Lin2Sca-1+cells was clearly
higher in the Mef null cells than the Mef+/+control cells (Fig-
ure 4D). We also measured Ki67 expression on purified Lin2
Pyronin-Ylowcells; we found more Ki67-negative cells in Mef
null LSK Pyronin-Ylowcells compared to wild-type, which was
also evident by measuring Ki67 expression on gated LSK cells
(Figure S5). The increased quiescence observed in Mef null cells
suggests that Mef regulates the entrance of primitive progenitor
cells into the cell cycle at steady state.
To confirm the reduced proliferation of LSK cells using in vivo
assays, we determined the proportion of LSK cells that incorpo-
rate BrdU over a 2 day period. Nuclear BrdU was measured in
LSK cells using a DNA-labeling protocol. Only 20% of Mef null
LSK cells proliferated during 48 hr BrdU exposure while the
BrdU incorporation into wild-type LSK cells was 60% (Fig-
ure 4E). Thus, Mef null LSK cells proliferate less than normal un-
der steady-state conditions.
Mef null BMMCs are protected from cell
The presence of more quiescent HSCs in Mef2/2mice could re-
sult in enhanced recovery from myelosuppressive treatments
that target cycling hematopoietic cells. To measure the kinetics
of hematopoietic recovery, we administered a single dose of
a chemotherapeutic agent (i.e., 5-FU) and serially followed pe-
ripheral blood counts. 5-FU-treated Mef2/2mice had less se-
vere leukopenia than wild-type mice, with more rapid recovery
(Figure 5A). The speedy restoration of myeloerythroid cell line-
ages improved the overall survival, as fewer Mef null mice suc-
cumbed to life-threatening infections due to severe neutropenia
(100% survival compared to 33% in wild-type mice). The milder
leukopenia seen after 5-FU treatment is likely due to the pres-
fined as LSK CD342Flt32and LSK CD34+Flt32cells) in the
bone marrow. LSK CD34+Flt32cells can efficiently protect
myeloablated mice from life-threatening cytopenias (Yang
et al., 2005). To address the possibility that Mef deficiency could
alter the expression of enzymes involved in metabolizing 5-FU,
we also tested busulfan, adifferent cytotoxic agent that also tar-
gets proliferating cells. As shown in Figure S6, WBC recovery
post-busulfan is also faster in Mef-deficient mice, with milder
leukopenia compared to wild-type mice. As irradiated Mef2/2
mice also display a more rapid recovery of WBC counts (Fig-
ure S6), this protection from myelosuppression is likely due to
a combination of enhanced quiescence and an increased num-
ber of HSCs.
Although it was recently reported that bone marrow cells
undergo senescence following myelosuppressive therapies,
A R T I C L E
CANCER CELL MARCH 2006179
evidenced by increased levels of p16 and p19 (Wang et al.,
2006), Mef2/2bone marrow cells isolated at different times
post-5-FU treatment showed a similar pattern of p16 and p19
expression as the wild-type group (Figure S7), implying that
loss of Mef does not affect chemotherapy-induced cell senes-
To correlate the hematopoietic recovery observed in the pe-
ripheral blood with bone marrow morphologic features, we ex-
amined the location of hematopoietic cells within the femurs of
Mef2/2mice following 5-FU treatment. We saw no appreciable
difference in the trabecular bone, the growth plate, or the epiph-
ysis of untreated wild-type and Mef2/2femurs stained with
hematoxylin and eosin (Figure 5B). This suggests that Mef-
deficient mice have no gross alterations in stem cell niches
(Calvi et al., 2003; Zhang et al., 2003). Examining femoral sec-
tions at various time points showed that although the initial
bone marrow findings are similar in Mef2/2and Mef+/+femurs
(Figure 5C), clusters of hematopoietic cells can be observed
as early as day 4 in the osteoblastic zone of the femur of
Mef2/2mice. Six days post-5-FU, the bone marrow of Mef2/2
mice had retained over 50% of its cellularity in contrast to
wild-type mice (10%); the expansion of megakaryocytes and
myeloid cells is also increased in ablated Mef null mice (data
not shown). The early appearance of polyploid megakaryocytes
in the vascular niche, which is usually observed at much later
times (Avecilla et al., 2004), accounts for the more rapid platelet
recovery seen following 5-FU administration to Mef-deficient
mice (data not shown).
Figure 4. Loss of Mef leads to increased quies-
cence of hematopoietic stem cells
A: Defective entry to cell cycle of Mef2/2LSK
cells in response to a 24 hr incubation with SCF,
IL-3, and IL-6. Cell cycle analysis was performed
on nuclei stained with propidium iodide and an-
alyzed using the ModFit software. LSK cells were
purified from four mice per group.
B: Single LSK cells were purified by cell sorting into
96-well plates (1 cell/well), and then cell growth
was evaluated microscopically for each differ-
ent clone. Only ten representative clones for
each group are shown.
C: Multiparameter flow cytometry was used to
determine the percentage of hematopoietic
progenitor cells in the G0phase of the cell cycle.
Total bone marrow cells were stained with anti-
Sca-1, with a lineage cocktail (Lin), and with
Pyronin Y and Hoechst 33342. Lin2Sca-1+cells
in G0are defined as cells with low Pyronin Y con-
tent that contain 2n DNA (G0/G1).
D: The percentage of Lin2Sca-1+cells in G0is
plotted (n = 7; **p < 0.005, two-tailed distribution
of two samples with equal variance).
E: The proliferation of LSK cells was measured by
in vivo BrdU incorporation. Lower proliferation of
Mef null LSK cells was seen (22% versus 60% for
wild-type; n = 3).
A R T I C L E
CANCER CELL MARCH 2006
The rapid recovery from a myeloablative treatment in Mef null
mice seems paradoxically opposed to the increased quies-
cence observed in homeostasis. However, the steady-state
etic progenitors postablation is likely controlled by different
mechanisms. We monitored stem cell expansion post-5-FU
treatment using SP cell detection and found the in vivo prolifer-
ative response of Mef null cells to myeloablation similar to that
seen in Mef+/+mice (Figure 6A). We also observed similar
BrdU incorporation in LSK cells and Lin2Sca-1+(LS) cells
post-5-FU treatment (Figure 6B). Thus, although the absence
of Mef enhances the quiescence of primitive progenitors, its ab-
sence does not prevent cell division in response to myeloabla-
Resistance to chemotherapy is intrinsic to Mef null HSCs
The improved hematopoietic recovery from 5-FU treatment
seen in Mef null mice is cell intrinsic and transplantable. Normal
B6 mice transplanted with Mef2/2bone marrow mononuclear
cells showed the same rapid recovery of peripheral blood
WBC and platelet counts post-5-FU (Figure 7A) as Mef null
mice (Figure 5A). Similarly, increased numbers of hematopoietic
cells are found in the osteoblastic and vascular zones of the fe-
murs of mice receiving Mef null BMMCs on day 3 (Figure 7A),
when bone marrow ablation is usually reached (Figure 5). Re-
markably, wild-type mice transplanted with Mef null BMMCs
also exhibit an expansion of megakaryocyte (Figure 7A), which
Given the importance of stem cell quiescence in regulating
hematopoiesis, we investigated whether the acute lowering of
MEF expression in human hematopoietic cells could lead to
the same perturbations in cell cycle that we observed in Mef
null mice. We reduced MEF expression in human cord blood
CD34+cells by siRNA and observed reduced proliferation (per-
centage in S phase) and increased numbers of quiescent cells
(percentage in G0, defined as Ki67-negative cells within G0/G1
phase of cell cycle) even when MEF levels were only reduced
by 30%–40% (Figure 7B). Thus, decreasing MEF levels in these
human cells enhances quiescence, which could similarly pro-
vide in vivo protection from myelosuppression.
Figure 5. Hematopoietic recovery after treat-
ment with myelotoxic agents
serial peripheral blood count monitoring of mice
(Mef+/+and Mef2/2) injected with a single dose
of 5-FU (200 mg/kg, i.p.). WBC counts are shown
as a percentage of the initial values for each
group of mice (mean 6 standard deviation; n =
3 for each time point). Overall survival is indi-
cated for the mice treated with 5-FU.
B: Gross morphology of femurs from untreated
mice and mice 8 days after 5-FU injection is
shown. Slides were stained with hematoxylin-
eosin. The epiphysis (E), growth plate (GP), and
trabecular bone (TB) regions of metaphysis are
C: Bone marrow morphology was examined fol-
magnification in the diaphysis of the bone (indi-
cated by rectangles in B). Localization of the
osteoblastic niche (ON) and vascular niche
(VN) are indicated. Arrowheads show clusters
of hematopoietic cells in the osteoblastic zone
4 days after 5-FU administration.
A R T I C L E
CANCER CELL MARCH 2006181
The preservation of HSCs throughout life is essential to ensure
an adequate supply of mature blood cells under physiologic
and stress situations. HSC numbers remain constant over
time, with approximately 8% of the HSCs randomly undergoing
cell division each day to generate another HSC and/or to differ-
entiate along specific lineages (Passegue et al., 2003). The fate
of HSCs that commit to enter the cell cycle is self-renewal, dif-
ferentiation, senescence, or apoptosis. The molecular determi-
nants of stem cell maintenance have not been studied in detail,
although the involvement of Hox proteins such as HoxB4
(Antonchuk et al., 2002) and polycomb group proteins such as
Bmi-1 (Lessard and Sauvageau, 2003a, 2003b) in stem cell
self-renewal has been unveiled. The regulation of stem cell dor-
mancy is particularly relevant to bone marrow failure states, and
for patients treated with myelotoxic chemotherapy or radiation.
Furthermore, efficientengraftment ofHSCsduring bonemarrow
transplantation may depend on their cell cycle status, in fact
their quiescence (Passegue et al., 2005).
We have shown that loss of the transcription factor Mef leads
to increased stem cell quiescence, diminished in vitro cytokine-
induced cell proliferation of primitive progenitors, and in vivo re-
sistance to cell cycle-dependent myelotoxicity. These findings
demonstrate that Mef facilitates the entrance of HSCs to the
cell cycle, and bone marrow transplantation studies show that
of Mef expression. The absence of Mef results in an accumula-
tion of primitive hematopoietic cells with repopulating capacity
(both LT-HSCs and ST-HSCs), which is accompanied by en-
hanced steady-state quiescence, no change in cell survival,
and minimal changes in self-renewal (not detectable in serial
bone marrow transplantation [sBMT] assays). This accumula-
croenvironment, as we have seen nearly normal mobilization of
stem cells from the bone marrow of Mef-deficient mice into the
bloodstream in response to G-CSF (data not shown).
We have confirmed the increased proportion of primitive pro-
genitors in Mef2/2mice by a variety of in vitro and in vivo assays
commonly used to identify stem cells, including CAFC, LTC-IC,
immunophenotyping (Lin2Thy1lowSca-1+c-kit+cells, SP-LSK
cells, and CD342Flt32LSK cells), CFU-S12(data not shown),
and serial replating in methylcellulose. Multilineage rescue of
contain HSCs with long-term repopulating capacity when trans-
planted into both primary and secondary recipients. Interest-
ingly, although Mef null primitive progenitors cells do not enter
the cell cycle efficiently at steadystate, their progeny do expand
normally in vivo, following bone marrow transplantation or dur-
ing recovery from chemotherapy. The role of MEF in cell cycle
regulation in HSCs versus committed progenitors shows its
cell stage-specific effects. As a result of MEF loss, LT-HSCs
and ST-HSCs tend to accumulate, due to a lower proliferation
capacity without a major impact on self-renewing divisions.
Lymphoid and myeloid development proceeds normally in the
progenitor populations, leading to nearly normal peripheral
The increased stem cell pool in Mef null mice is likely due to
their increased quiescence. The abundance of LT-HSCs and
ST-HSCs in Mef null mice, as a result of increased dormancy,
enables a faster hematologic recovery postmyelosuppression.
Although the converse could also be true, that the increase in
stem cell numbers triggers the increased quiescence, our trans-
plantation experiments suggest that this is not the case, as con-
sistent biological properties areseenindependent ofthe infused
cell dose. The increased quiescence in the absence of MEF was
Figure 6. Normal expansion of Mef2/2hematopoietic progenitor cells dur-
ing bone marrow ablation
A: SP cells were identified by flow cytometric analysis in BMMCs obtained
from mice 2 and 4 days after 5-FU administration. The number of SP cells
that resisted this treatment is indicated as percentages.
B: In vivo proliferation of LSK and LS (Lin2Sca-1+) cells following 5-FU treat-
ment. Mice (n = 3) were injected with 5-FU 2 days before the BrdU injection
and then left for 2 additional days with BrdU in their water. Statistical analysis
is shown in a bar graph (mean 6 standard deviation).
A R T I C L E
CANCER CELL MARCH 2006
evidenced by the incorporation of BrdU in vivo, by in vivo resis-
tance to cell cycle-dependent cytoxicity, by measurements of
pression within LSK Pyronin-Ylowcells. Furthermore, acute
knockdown of MEF in vitro using siRNA increased quiescence
of human CD34+cells.
Several cell cycle regulators have been shown to play key
roles in hematopoiesis. The absence of the cdk inhibitor p27
(kip1) affects the proliferation of progenitor cells rather than
HSCs (Cheng et al., 2000a). In contrast, mice deficient in the
p21 cdk inhibitor have increased stem cell numbers with higher
proliferative capacity but also lower quiescence than normal
(Cheng et al., 2000b). This leads to stem cell exhaustion follow-
ing serial 5-FU administration or sBMT. A recent study shows
that the cdk inhibitor p18 INK4C acts earlier than p21 in G1,
and its absence increases the self-renewing divisions of HSCs
(Yuan et al., 2004). Mef also regulates the size of the primitive
progenitor cell pool, regulating the release of HSCs from quies-
cence. This suggests that p21 and Mef could regulate stem cell
pool size via opposing effects.
We propose a model where MEF (ELF4) facilitates cell cycle
entry of primitive progenitors from their quiescent state, acting
earlier than p18 and antagonizing p21 (Figure 8). The enforced
expression of GATA-2 has been shown to block the expansion
loinsufficiency results in fewer CD342LSK cells with poor com-
petitive repopulating ability. Increased quiescence is seen in
GATA-2+/2cells in the presence of normal self-renewal. Thus,
MEF and GATA-2 could be the first transcription factors to be
implicated in the maintenance of stem cell quiescence (Ro-
drigues et al., 2005) (Figure 8A). The zinc finger repressor Gfi-1
appears to restrict the proliferation of HSCs, likely in a p21-
dependent fashion as Gfi-1 null HSCs show downregulation of
levels in Mef null BMMCs (data not shown). The contrasting ef-
fects of Mef and Gfi-1 in terms of cell cycle regulation, CR, and
engraftment capacity might be due to their different actions on
pressor (Gfi-1). Although Tie-2 expression has been recently
Figure 7. Enhanced hematopoietic recovery is
due solely to hematopoietic progenitors and
their resistance to 5-FU
A: Wild-type mice transplanted with Mef+/+or
Mef2/2BMMCs were injected with 5-FU, and
the hematopoietic recovery was measured in
peripheral blood (WBC and PLT counts). Data
are shown as mean 6 standard deviation; n = 3
for each time point. The histology of femurs 3
days posttreatment is shown on the right (filled
arrowheads, osteoblastic zone; empty arrow-
heads, vascular niche). Megakaryocytic ex-
pansion in the femur is shown at day 13 post-
treatment (multinuclear megakaryocytes are
indicated with arrowheads). The number of
platelets (PLT) in the peripheral blood at day 13
is also indicated.
B: Effect of deregulated MEF expression in hu-
man hematopoietic cells. Cord blood CD34-
positive cells were transfected with control and
MEF siRNA. Ki67 and DNA content were then an-
alyzed by flow cytometry.
A R T I C L E
CANCER CELL MARCH 2006183
linked to quiescent HSCs (Arai et al., 2004), we have not de-
tected significant alterations of Tie-2 expression in Mef null
bone marrow cells (data not shown). A full molecular delineation
of the Mef2/2HSCs is planned; such characterization will help
define the cell-intrinsic mechanisms that control stem cell main-
Tumorigenesis is a multistep process where cells accumulate
genetic alterations that result in their progressive transforma-
tion. MEF-deficient mice seem to have a normal lifespan, with-
out spontaneous tumor formation. However, MEF2/2HSCs
might be prone to transformation, as a larger stem cell pool
could increase the likelihood that these cells accumulate addi-
tional mutations. Leukemia cells could benefit from inactivation
of MEF (via mutations or via transcriptional repression such as
occurs with AML1-ETO, which can block MEF function [Mao
et al., 1999]). Studies implicating the MEF (also known as
ELF4) gene in malignant transformation include its downregula-
tion by two human leukemia-associated fusion proteins (Alcalay
et al., 2003; Muller-Tidow et al., 2004; Park et al., 2003), its po-
tential role as a tumor suppressor gene in solid tumors (Seki
et al., 2002), and also its upregulation by retroviral insertional
activation in murine cancer models (Akagi et al., 2004; Du
et al.,2005;Suzukietal., 2002).Thesefeatures areinfactsimilar
to those implicating the AML1 transcription factor in hemato-
logic malignancies, namely the functional downregulation of
AML-1 function by AML1-ETO (Frank et al., 1995), haploinsuffi-
ciency studies in a familial AML predisposition syndrome (FPD/
AML) implicating AML1 asatumorsuppressorgene (Song et al.,
1999), and the retroviral insertional activation of Runx1, Runx2,
or Runx3 in murine cancer models (Stewart et al., 1997, 2002;
Wotton et al., 2002). Although lack of AML1 function during de-
velopment is embryonically lethal, the lack of MEF alters stem
cell biology but does not block hematopoietic development.
In other studies, we have shown that deregulated MEF ex-
pression can be tumorigenic in solid tumors; overexpression
of MEF in ovarian cancer cell lines increases proliferation and
aggressiveness, and MEF overexpression transforms NIH3T3
cells leading to tumor growth in nude mice (J.-J. Yao, Y.L.,
H.D.L., R.A. Soslow, J.M. Scandura, S.D.N., and C.V. Hedvat,
unpublished data). The recent discovery of t(X;21)(q26:q22) in
a patient with AML that fuses exon 2 of the MEF gene with
most of the coding region of the ERG gene, which encodes an-
other ETS protein, strongly supports the role of MEF dysregula-
tion in hematological malignancies (Moore et al., 2005). In this
female patient, the t(X;21) generates at least haploinsufficiency
it deregulates ERG expression, which is now transcriptionally
controlled by the MEF promoter.
Advances in our understanding of HSC biology will hasten the
development of new therapeutic approaches to cancer. HSCs
may make an active choice to neither self-renew nor differenti-
ate, but rather to remain quiescent. MEF regulates the quies-
cence of primitive hematopoietic progenitors; therefore, treat-
ments that block or diminish MEF expression could contribute
to improve hematopoietic recovery postmyelosuppression. He-
matopoiesis in Mef2/2mice is far less affected by 5-FU admin-
istration than that in normal mice, leading to minimal cytopenias
andfewerfatal infections.Our studiesusingsiRNAtolowerMEF
levels in human CD34+cells and enhance quiescence suggest
that this could even be clinically relevant. Myelotoxicity induced
by chemotherapy or radiotherapy could be prevented by main-
taining stem cells in a quiescent state during their administration
to cancer patients. However, another implication of our work is
that tumor stem cells are more quiescent than more differenti-
ated tumor cells and could use similar mechanisms to resist
the effects of chemotherapy or radiation (Chevallier et al.,
2004; Nakamura et al., 2002; Snowden et al., 2003).
The generation of Mef-deficient mice was described previously (Lacorazza
et al., 2002). C57BL/6 (CD45.1+), B6.SJL (CD45.2+), and F1 B6/SJL mice
were purchased from Jackson Laboratories. All mice were maintained in
MSKCC Animal Facility and Baylor College of Medicine Animal Facility, ac-
cording to IACUC-approved protocols, and kept in Thorensten units with fil-
tered germ-free air.
Stem and progenitor cell assays
To evaluate the most primitive progenitors, we used the CAFC assay. Briefly,
1.5 3 105bone marrow cells were seeded on MS5 stroma and cultured in
aMEM containing 10% FBS, 10% horse serum, 1 mM hydrocortisone, and
1 mM glutamine. Medium was semireplenished every week, and ‘‘cobble-
stone’’ colonies were scored at week 5 and expressed as number of CAFC
Figure 8. Model of MEF function on the proliferation of hematopoietic stem
A: Diagram depicting the cell cycle of stem cells (left) and progenitor cells
(right) and the cell cycle regulators known to control their proliferation.
B: LT-HSCs were defined in this work as Lin2Sca-1+c-kit+SP cells and as Lin2
Sca-1+c-kit+CD342Flt32cells. ST-HSCs were identified as Lin2Sca-1+c-kit+
CD34+Flt32cells. In the absence of MEF, we found a higher fraction of prim-
itive progenitors (LT-HSC and ST-HSC) in spite of normal differentiation into
lymphoid and myeloid cells. The abundance of ST-HSCs further contributes
to the enhanced hematologic recovery seen post-myeloid suppression, in
addition to the increased primitive progenitor cell quiescence.
A R T I C L E
CANCER CELL MARCH 2006
formed. After 5 weeks of weekly semireplenishment of the media, cells were
trypsinized and plated on methylcellulose media (MethoCult GF M3434,
StemCell Technologies) and cultured for 10 days before the percentage of
negative wells per dilution was scored. Frequencies were calculated using
Poisson statistics (L-cal program from StemCell Technologies).
Thecontentofprimitive myeloidprogenitorswas determined bytheday12
bone marrow cells.Spleensweredissected12 daysposttransplant andfixed
in Bouin’s solution. Colonies per spleen were counted by visual inspection
and expressed as number of CFU-S12colonies per 1 3 105BMMCs.
Clonogenic progenitors were determined in methylcellulose medium
(MethoCult GF M3434, StemCell Technologies) using 2 3 104BMMCs per
well (6-well plate). Colonies were scored after 10 days of incubation and ex-
pressed as number of CFUs per 2 3 104BMMCs.
Bone marrow transplantation
Recipient mice were lethally irradiated with 10 Gy of whole-body irradiation.
Two to three million BMMCs were injected intravenously into the lateral tail
vein of recipient mice previously warmed under a heat lamp. Animals were
daily monitored for signs of toxicity and sacrificed 2 months postinfusion.
sBMT was used to test the self-renewal capacity of the Mef2/2stem cells.
We used Lin2Sca-1+c-kit+cells (HSCs) purified by cell sorting (MoFlow
cell sorter, Cytomation); cells were then injected (5000 LSKs per mouse)
into lethally irradiated mice (B6.SJL). We also performed sBMT using do-
nor-derived BMMCs. B6 (CD45.2) mice were used as donors, and B6.SJL
(CD45.1) mice were used as recipients. For each transplant, donor-derived
cells (CD45.2+) were purified by cell sorting and injected into lethally irradi-
ated B6.SJL mice (approximately 1–2 3 105BMMCs/mouse). In both cases,
hematopoietic reconstitution was monitored in peripheral blood, as de-
In the CR study, we injected different percentages of BMMCs from wild-
type (B6.SJL) and Mef2/2(B6) mice (0%, 20%, 50%, 80%, or 100%) into
lethally irradiated B6.F1 mice (CD45.1/CD45.2). After 5 months of reconstitu-
tion, peripheral blood was obtained by retro-orbital eye bleeding. The RBCs
were lysed, and the PBMCs were stained with anti-CD45.2 FITC and anti-
CD45.1 PE and analyzed by flow cytometry. These mice were then irradiated
with 4.5 Gy of whole-body radiation, and hematopoietic recovery in periph-
eral blood was analyzed at different time points, as described below.
Measurement of cell cycle parameters using Pyronin Y
and Hoechst staining
Bone marrow cells were first stained for the Lin cell surface markers and for
Sca-1. Then, cells were resuspended in a Hank’s balanced salt solution,
20 mM HEPES, 1 g/l glucose, 10% FCS, 1.7 mM Hoechst 33342 and incu-
bated for 1hr at 37ºC. After asinglewash, bone marrow cells were incubated
in the same buffer containing Pyronin Y (1 mg/ml) for an additional hour at
37ºC. Finally, cells were analyzed using a MoFlow cytometer (Cytomation)
and FlowJo program (Cytomation).
Mice received an intraperitoneal injection of 3 mg BrdU (Becton Dickinson)
and admixture of 1 mg/ml of BrdU (Sigma) to drinking water for 2 days.
Bone marrow cells were isolated and lineage negative purified using the
BD-Imag system (Becton Dickinson) and then stained with c-kit PE and
Sca-1 FITC antibodies. Then, the analysis of BrdU incorporation was per-
formed using the APC BrdU flow kit (Becton Dickinson).
Cell cycle analysis
LSK cells were purified by cell sorting pooling four mice per genotype (wild-
type and knockout). Cellswere incubated for24hr in Xvivo-15 supplemented
(or not) with SCF (100 ng/ml), IL-3 (10 ng/ml), and IL-6 (6 ng/ml). Cells were
centrifuged and resuspended in a hypotonic buffer (0.1% sodium citrate,
0.1% Triton X-100) containing 100 mg/ml RNase and 50 mg/ml propidium io-
dide. Samples were analyzed within 1 hr using the FACScanto flow cytome-
ter (Becton Dickinson) and the Modfit LT software (Verity).
Detection of SP cells
We followed the SP detection procedure using Hoechst dye as previously
described (Goodell et al., 1996). Briefly, BMMCs were resuspended at 1 3
106cells/ml (DMEM + 10% FBS) and incubated with Hoechst 33342
(5 mg/ml) for 90 min at 37ºC. Cells were then washed and stained for cell sur-
face markers (lineage, Sca-1, and c-kit). Analysis was performed on a MoFlo
flow cytometer (Cytomation) equipped with a 351 nm laser for UV. Red and
blue fluorescence derived from UV excitation was separated with a 610 di-
chromatic DRSP. Blue Hoechst fluorescence was collected with a 450/20 fil-
ter, and red Hoechst/PI fluorescence was collected via a 675 ESLP filter. We
collected 5 3 106events to have enough cells in the SP-LSK gate.
We used 5-FU (200 mg/kg, i.p.), busulfan (15 mg/kg, s.c.), and 4.5 Gy radia-
tion for myelosuppression, in order to follow hematopoietic recovery in the
peripheral blood. PBCs were monitored once a week in each mouse but
not more than three times overall. Therefore, each experimental group
(wild-type and knockout) was formed by two sets of three to four mice per
set that were bled alternatively. Retro-orbital peripheral blood (<100 ml)
was obtained from anesthetized mice and collected into EDTA-coated cap-
illary tubes. Complete blood counts and differential counts were measured in
the Genetically Engineered Mouse Phenotyping Core (MSCKK/Cornell/
Rockefeller) using an automated blood cell counter.
The Supplemental Data include Supplemental Experimental Procedures,
seven supplemental figures, and two supplemental tables and can be found
with this article online at http://www.cancercell.org/cgi/content/full/9/3/175/
The authors would like to thank the staff of the Flow Cytometry and Mouse
Genotyping Core Facilities in MSKCC. We are also grateful to Dr. Margaret
Goodell for her suggestions on conducting SP analysis; to Drs. John Petrini,
Malcolm Moore, and Vladimir Jankovic for their critical review of our manu-
Threeton for the SP detection at Texas Children’s Hospital. This work was
funded by NIH RO1 DK52208 (S.D.N.), a Howard Temin Award from NCI/
NIH to H.D.L. (KO1-CA099156), the Moran Foundation Award (H.D.L.), the
Curtis Hankamer Basic Research Fund (Junior-Faculty Seed Fund Award,
Baylor College of Medicine; H.D.L.), and The Laurie Strauss Leukemia Foun-
dation Award (H.D.L.).
Received: September 1, 2005
Revised: December 23, 2005
Accepted: February 13, 2006
Published: March 13, 2006
Akagi, K., Suzuki, T., Stephens, R.M., Jenkins, N.A., and Copeland, N.G.
(2004). RTCGD: retroviral tagged cancer gene database. Nucleic Acids
Res. 32, D523–D527.
Alcalay, M., Meani, N., Gelmetti, V., Fantozzi, A., Fagioli, M., Orleth, A., Riga-
nelli, D., Sebastiani, C., Cappelli, E., Casciari, C., et al. (2003). Acute myeloid
leukemia fusion proteins deregulate genes involved in stem cell maintenance
and DNA repair. J. Clin. Invest. 112, 1751–1761.
Antonchuk, J., Sauvageau, G., and Humphries, R.K. (2002). HOXB4-induced
expansion of adult hematopoietic stem cells ex vivo. Cell 109, 39–45.
Arai, F., Hirao, A., Ohmura, M., Sato, H., Matsuoka, S., Takubo, K., Ito, K.,
Koh, G.Y., and Suda, T. (2004). Tie2/angiopoietin-1 signaling regulates
hematopoietic stem cell quiescence in the bone marrow niche. Cell 118,
Avecilla, S.T., Hattori, K., Heissig, B., Tejada, R., Liao, F., Shido, K., Jin, D.K.,
Dias, S., Zhang, F., Hartman, T.E., et al. (2004). Chemokine-mediated
A R T I C L E
CANCER CELL MARCH 2006185
interaction of hematopoietic progenitors with the bone marrow vascular
niche is required for thrombopoiesis. Nat. Med. 10, 64–71.
Calvi, L.M., Adams, G.B., Weibrecht, K.W., Weber, J.M., Olson, D.P., Knight,
M.C., Martin, R.P., Schipani, E., Divieti, P., Bringhurst, F.R., et al. (2003).
Osteoblastic cells regulate the haematopoietic stem cell niche. Nature 425,
Camargo, F.D., Chambers, S.M., Drew, E., McNagny, K.M., and Goodell,
M.A. (2006). Hematopoietic stem cells do not engraft with absolute efficien-
cies. Blood 107, 501–507.
Cheng, T., Rodrigues, N., Dombkowski, D., Stier, S., and Scadden, D.T.
(2000a). Stem cell repopulation efficiency but not pool size is governed by
p27kip1. Nat. Med. 6, 1235–1240.
Cheng, T., Rodrigues, N., Shen, H., Yang, Y., Dombkowski, D., Sykes, M.,
and Scadden, D.T. (2000b). Hematopoietic stem cell quiescence maintained
by p21cip1/waf1. Science 287, 1804–1808.
Chevallier, N., Corcoran, C.M., Lennon, C., Hyjek, E., Chadburn, A., Bard-
well, V.J., Licht, J.D., and Melnick, A. (2004). ETO protein of t(8;21) AML is
a corepressor for Bcl-6 B-cell lymphoma oncoprotein. Blood 103, 1454–
Choi, C., Cho, S., Horikawa, I., Berchuck, A., Wang, N., Cedrone, E., Jhung,
S.W., Lee, J.B., Kerr, J., Chenevix-Trench, G., et al. (1997). Loss of heterozy-
gosity at chromosome segment Xq25-26.1 in advanced human ovarian car-
cinomas. Genes Chromosomes Cancer 20, 234–242.
Choi, C., Kim, M.H., and Juhng, S.W. (1998). Loss of heterozygosity on chro-
mosome XP22.2-p22.13 and Xq26.1-q27.1 in human breast carcinomas.
J. Korean Med. Sci. 13, 311–316.
Du, Y., Spence, S.E., Jenkins, N.A., and Copeland, N.G. (2005). Cooperating
cancer-gene identification through oncogenic-retrovirus-induced insertional
mutagenesis. Blood 106, 2498–2505.
Frank, R., Zhang, J., Uchida, H., Meyers, S., Hiebert, S.W., and Nimer, S.D.
(1995). The AML1/ETO fusion protein blocks transactivation of the GM-CSF
promoter by AML1B. Oncogene 11, 2667–2674.
Fukushima, T., Miyazaki, Y., Tsushima, H., Tsutsumi, C., Taguchi, J., Yosh-
ida, S., Kuriyama, K., Scadden, D., Nimer, S., and Tomonaga, M. (2003).
The level of MEF but not ELF-1 correlates with FAB subtype of acute my-
eloid leukemia and is low in good prognosis cases. Leuk. Res. 27,
Goodell, M.A., Brose, K., Paradis, G., Conner, A.S., and Mulligan, R.C.
(1996).Isolation andfunctional propertiesofmurinehematopoieticstemcells
that are replicating in vivo. J. Exp. Med. 183, 1797–1806.
Hock, H., Hamblen, M.J., Rooke, H.M., Schindler, J.W., Saleque, S., Fuji-
wara, Y., and Orkin, S.H. (2004). Gfi-1 restricts proliferation and preserves
functional integrity of haematopoietic stem cells. Nature 431, 1002–1007.
Lacorazza, H.D., and Nimer, S.D. (2003). The emerging role of the myeloid
Elf-1 like transcription factor in hematopoiesis. Blood Cells Mol. Dis. 31,
Lacorazza, H.D., Miyazaki, Y., Di Cristofano, A., Deblasio, A., Hedvat, C.,
Zhang, J., Cordon-Cardo, C., Mao, S., Pandolfi, P.P., and Nimer, S.D.
(2002). The ETS protein MEF plays a critical role in perforin gene expres-
sion and the development of natural killer and NK-T cells. Immunity 17,
Lessard, J., and Sauvageau, G. (2003a). Bmi-1 determines the proliferative
capacity of normal and leukaemic stem cells. Nature 423, 255–260.
Lessard, J., and Sauvageau, G. (2003b). Polycomb group genes as epige-
netic regulators of normal and leukemic hemopoiesis. Exp. Hematol. 31,
Mao, S., Frank, R.C., Zhang, J., Miyazaki, Y., and Nimer, S.D. (1999). Func-
tional and physical interactions between AML1 proteins and an ETS protein,
MEF: implications for the pathogenesis of t(8;21)-positive leukemias. Mol.
Cell. Biol. 19, 3635–3644.
Miyazaki, Y., Sun, X., Uchida, H., Zhang, J., and Nimer, S. (1996). MEF,
a novel transcription factor with an Elf-1 like DNA binding domain but distinct
transcriptional activating properties. Oncogene 13, 1721–1729.
Miyazaki, Y., Boccuni, P., Mao, S., Zhang, J., Erdjument-Bromage, H.,
Tempst, P., Kiyokawa, H., and Nimer, S.D. (2001). Cyclin A-dependent phos-
phorylation of the ETS-related protein, MEF, restricts its activity to the G1
phase of the cell cycle. J. Biol. Chem. 276, 40528–40536.
Moore,S.D.,Offor,O.,Ferry, J.A.,Amrein, P.C.,Morton,C.C., andDalCin,P.
(2005). ELF4 is fused to ERG in a case of acute myeloid leukemia with
a t(X;21)(q25-26;q22). Leuk. Res., in press. Published online November 21,
Muller-Tidow, C., Steffen, B., Cauvet, T., Tickenbrock, L., Ji, P., Diederichs,
S., Sargin, B., Kohler, G., Stelljes, M., Puccetti, E., et al. (2004). Translocation
matopoietic cells. Mol. Cell. Biol. 24, 2890–2904.
Nakamura, H., Morishita, R., and Kaneda, Y. (2002). Molecular therapy via
transcriptional regulation with double-stranded oligodeoxynucleotides as
decoys. In Vivo 16, 45–48.
Park, D.J., Vuong, P.T., de Vos, S., Douer, D., and Koeffler, H.P. (2003).
Comparative analysis of genes regulated by PML/RARa and PLZF/RARa
in response to retinoic acid using oligonucleotide arrays. Blood 102,
Passegue, E.,Jamieson,C.H., Ailles, L.E.,and Weissman, I.L.(2003). Normal
sition of stem cell characteristics? Proc. Natl. Acad. Sci. USA Suppl. 100,
Passegue,E., Wagers,A.J.,Giuriato, S.,Anderson, W.C.,and Weissman,I.L.
(2005). Global analysis of proliferation and cell cycle gene expression in the
regulation of hematopoietic stem and progenitor cell fates. J. Exp. Med.
Persons, D.A., Allay, J.A., Allay, E.R., Ashmun, R.A., Orlic, D., Jane, S.M.,
Cunningham, J.M., and Nienhuis, A.W. (1999). Enforced expression of the
GATA-2 transcription factor blocks normal hematopoiesis. Blood 93, 488–
Piao, Z., and Malkhosyan, S.R. (2002). Frequent loss Xq25 on the inactive X
chromosome in primary breast carcinomas is associated with tumor grade
and axillary lymph node metastasis. Genes Chromosomes Cancer 33,
Ploemacher, R.E., van der Sluijs, J.P., van Beurden, C.A., Baert, M.R., and
Chan, P.L. (1991). Use of limiting-dilution type long-term marrow cultures
in frequency analysis of marrow-repopulating and spleen colony-forming he-
matopoietic stem cells in the mouse. Blood 78, 2527–2533.
Rodrigues, N.P., Janzen, V., Forkert, R., Dombkowski, D.M., Boyd, A.S.,
Orkin, S.H., Enver, T., Vyas, P., and Scadden, D.T. (2005). Haploinsufficiency
of GATA-2 perturbs adult hematopoietic stem-cell homeostasis. Blood 106,
Seki, Y., Suico, M.A., Uto, A., Hisatsune, A., Shuto, T., Isohama, Y., and Kai,
H. (2002). The ETS transcription factor MEF is a candidate tumor suppressor
gene on the X chromosome. Cancer Res. 62, 6579–6586.
Snowden, A.W., Zhang, L., Urnov, F., Dent, C., Jouvenot, Y., Zhong, X.,
Rebar, E.J., Jamieson, A.C., Zhang, H.S., Tan, S., et al. (2003). Repression
zinc finger transcription factors. Cancer Res. 63, 8968–8976.
Song, W.J., Sullivan, M.G., Legare, R.D., Hutchings, S., Tan, X., Kufrin, D.,
Ratajczak, J., Resende, I.C., Haworth, C., Hock, R., et al. (1999). Haploinsuf-
ficiency of CBFA2 causes familial thrombocytopenia with propensity to de-
velop acute myelogenous leukaemia. Nat. Genet. 23, 166–175.
Stewart, M., Terry, A., Hu, M., O’Hara, M., Blyth, K., Baxter, E., Cameron, E.,
Onions, D.E., and Neil, J.C. (1997). Proviral insertions induce the expression
of bone-specific isoforms of PEBP2aA (CBFA1): evidence for a new myc col-
laborating oncogene. Proc. Natl. Acad. Sci. USA 94, 8646–8651.
Stewart, M., MacKay, N., Cameron, E.R., and Neil, J.C. (2002). The common
retroviral insertion locus Dsi1 maps 30 kilobases upstream of the P1 pro-
moter of the murine Runx3/Cbfa3/Aml2 gene. J. Virol. 76, 4364–4369.
Suzuki, T., Shen, H., Akagi, K., Morse, H.C., Malley, J.D., Naiman, D.Q.,
Jenkins, N.A., and Copeland, N.G. (2002). New genes involved in cancer
identified by retroviral tagging. Nat. Genet. 32, 166–174.
A R T I C L E
CANCER CELL MARCH 2006
Wang, Y., Schulte, B.A., LaRue, A.C., Ogawa, M., and Zhou, D. (2006). Total Download full-text
body irradiation selectively induces murine hematopoietic stem cell senes-
cence. Blood 107, 358–366.
Wotton, S., Stewart, M., Blyth, K., Vaillant, F., Kilbey, A., Neil, J.C., and Ca-
meron, E.R. (2002). Proviral insertion indicates a dominant oncogenic role for
Runx1/AML-1 in T-cell lymphoma. Cancer Res. 62, 7181–7185.
and Jacobsen, S.E. (2005). Identification of Lin2Sca1+kit+CD34+Flt32short-
term hematopoietic stem cells capable of rapidly reconstituting and rescuing
myeloablated transplant recipients. Blood 105, 2717–2723.
Yuan, Y., Shen, H., Franklin, D.S., Scadden, D.T., and Cheng, T. (2004).
In vivo self-renewing divisions of haematopoietic stem cells are increased
in the absence of the early G1-phase inhibitor, p18INK4C. Nat. Cell Biol. 6,
Zeng, H., Yucel, R., Kosan, C., Klein-Hitpass, L., and Moroy, T. (2004). Tran-
scription factor Gfi1 regulates self-renewal and engraftment of hematopoi-
etic stem cells. EMBO J. 23, 4116–4125.
Zhang, J., Niu, C., Ye, L., Huang, H., He, X., Tong, W.G., Ross, J., Haug, J.,
Johnson, T., Feng, J.Q., et al. (2003). Identification of the haematopoietic
stem cell niche and control of the niche size. Nature 425, 836–841.
A R T I C L E
CANCER CELL MARCH 2006 187