IJC

International Journal of Cancer

Leukemia stem cells

Eike C. Buss and Anthony D. Ho

Special Section Paper

Department of Internal Medicine V, Heidelberg University Medical Center, Im Neuenheimer Feld 410, Heidelberg, Germany

Leukemia stem cells (LSCs) might originate from malignant transformation of normal hematopoietic stem cells (HSCs), or
alternatively, of progenitors in which the acquired mutations have re-installed a dysregulated self-renewal program. LSCs are
on top of a hierarchy and generate leukemia cells with more differentiated characteristics. While most leukemia cells are
initially sensitive to chemo- and radiotherapy, LSCs are resistant and are considered to be the basis for disease relapse after
initial response. Albeit important knowledge on LSC biology has been gained from xenogeneic transplantation models
introducing human leukemia cells into immune deficient mouse models, the prospective identification and isolation of human
LSC candidates has remained elusive and their prognostic and therapeutic significance controversial. This review focuses on
the identification, enrichment and characterization of human LSC derived from patients with acute myeloid leukemia (AML).
Experimental data demonstrating the clinical significance of estimating LSC burden and strategies to eliminate LSC will be
summarized. For long-term cure of AML, it is of importance to define LSC candidates and to understand their tumor biology
compared to normal HSCs. Such comparative studies might provide novel markers for the identification of LSC and for the
development of treatment strategies that might be able to eradicate them.

Stem cells in health and in cancer

In analogy to physiological differentiation, a stem cell paradigm has been proposed for malignant development.1 There
is increasing evidence that human cancers may originate
from transformation of stem cells, or alternatively, from progenitors in which the acquired mutations have re-installed a
dysregulated self-renewal program. The rst hints were found
in hematological malignancies where only a small subset of
slowly dividing cells was able to induce transplantable acute
myeloid leukemia.2
Since the rst description by Lapidot et al.,2 many groups
have conrmed the existence of leukemia stem cells (LSCs)
for acute myeloid leukemia (AML). The engraftment of
human LSC preparations into xenogeneic transplantation
models has been regarded as evidence for the presence of leukemia initiating cells for human AML. In such studies, a
dened number of leukemia cells were administered to
immunodecient mice and the presence of LSC in the primary material was retrospectively assessed after engraftment
of human leukemia.3 Albeit relevant knowledge on LSC biology has been gained from these sophisticated experiments,
there has been thus far no reliable prospective method for
the estimation of LSC burden.

Key words: Leukemia, Stem cells, HSC, LSC, ALDH

DOI: 10.1002/ijc.26318
History: Received 13 Apr 2011; Accepted 12 Jul 2011; Online 27 Jul
2011
Correspondence to: Anthony D. Ho, Department of Internal
Medicine V, Heidelberg University Medical Center, Im Neuenheimer
Feld 410, 69120 Heidelberg, Germany,
E-mail: Anthony.Ho@med.uni-heidelberg.de

C 2011 UICC
Int. J. Cancer: 129, 23282336 (2011) V

The abundance of LSC has been associated with clinical

relapse or with refractory disease. This hypothesis has also
remained speculative because published data thus far were
not able to substantiate these claims in a denitive manner.
The prospective identication and isolation of human LSC
candidates have remained elusive and their clinical and prognostic signicance controversial.4,5
Similar to normal HSCs, LSCs are characterized by their
ability to self-renew, by their unlimited repopulating potential, and by the production of a multitude of progeny cells
with more differentiated characteristics, LSCs have also been
reported to express stem cell markers and to survive indenitely upon serial transplantations in animal models.
Whereas common leukemia blasts multiply prolically, the
LSC divide slowly.6 The latter characteristic is especially associated with their resistance to conventional chemotherapy
strategies that target rapidly dividing cells. Like their normal
counterparts, the LSCs are also extremely rare.

Xenogeneic transplantation models

Thus far, engraftment of human leukemia in an in vivo
xeno-transplantation model represents the ultimate proof for
the concept of LSC. The developmental steps for this animal
model have led in the nineties to the severe combined immunodeciency (SCID) model7 and thereafter to the successful
transplantation of selected subpopulations of AML cells into
SCID mice.2 This was followed later by the more efcient
NOD/SCID model.3 The transplanted leukemia cells were
highly similar to the original disease from the respective
patient and hence the stem cell nature of the transplanted
cells was considered proven.
The determination of the number of leukemic stem cells
is usually performed by limiting-dilution experiments. A

2329

Buss and Ho

Year

Assay

Authors

1961

Spleen colony forming cells (CFU-S) as progenitors after transplantation into

irradiated recipient mice

Till and McCulloch11

1965/66

Clonogenic assays in semisolid media for blood progenitor cells (CFU-GM, CFU-B)

Pluznik and Sachs12;

Bradley and Metcalf13

1977

Stroma-based cultivation methods for in vitro identication of LTC-ICs

Dexter et al.14

1989

In vitro stroma-based limiting dilution cobblestone area forming cell (CAFC)

assay for blood progenitor and stem cells

Ploemacher et al.15

1994

First in vivo demonstration of a human AML-initiating cell in immunodecient SCID mice

Lapidot et al.2

1997

Transplantation of AML stem cells in more immunodecient NOD/SCID mice

Bonnet and Dick3

null

mice with extended immunodeciency and

Shultz et al.9

2000

Development of NOD/LtSz-Rag1
higher engraftment rates

2005

Development of NOD/SCID/IL2R?null mice with extended immunodeciency and

higher engraftment rates

Ishikawa et al.8

2010

Development of NOD/SCID/IL2R?null mice plus transgenetic expression of human growth

factors SCF, GM-CSF and IL-3 with higher permissiveness for human myeloid cells

Wunderlich et al.10

dened number of cells in log-scale reductions were seeded

in colony-formation or transplantation studies and the
growth or engraftment is subsequently assessed and extrapolated to estimate the original stem/progenitor cell number.
Early in vivo experiments estimated a frequency of AML-initiating cells of about 0.4 per 105 mononuclear cells.2 This
was further rened by the demonstration of a self-renewal
capacity by serial transplantations. The most recent improvements of this model system include murine models with
increased immunodeciency8,9 and increased permissiveness
to normal as well as leukemia human cells by expression of
transgenes for human growth-factors in these animals.10
Methods derived from these experiments have become blueprints for cancer stem cell research derived from other
tissues.
Albeit the relevance of LSC has been suggested by extensive experiments in animal models, translation of this knowledge into the clinic has faced two major challengescontroversy in the identication of LSC candidates and
subsequently their prospective separation, as well as the denition of their biologic properties as compared to normal
HSC (Table 1).

Surface antigen markers for enrichment of myeloid

LSC
In analogy to identication of human HSC, most investigators have used surface antigen markers to separate and enrich
the LSC subset from the primary leukemia population. The
usual marker prole to distinguish and select the leukemic
stem cell population is based on CD34 as starting point. This
antigen is expressed on most HSC, although there is probably
a small fraction of dormant and primitive HSC without this
antigen. CD34 is also expressed on committed progenitor
cells and then lost in the further hematopoietic differentiation
process. Thus, it is not exclusive for stem cells. Similarly,
CD34 is expressed on a majority of AML stem cells. A comC 2011 UICC
Int. J. Cancer: 129, 23282336 (2011) V

monly used surface antigen marker for myeloid differentiation is CD38, which can also be found on several nonmyeloid
cell types and which is absent from HSC. The signicance of
CD38-expression for enrichment of LSC is controversial. In
most of the initial studies, LSCs were dened by a
CD34CD38
phenotype, but recently Taussig et al. demonstrated that seven of seven tested individual AML samples
could engraft from CD34CD38 cells. However, the cell
dose required for engraftment of CD34CD38 cells in animal models was in the range of 106 cells per animal, whereas
that of CD34CD38
was in a much lower range.16
This marker combination CD34/CD38
was able to
enrich leukemia-initiating cells or LSC, as rst demonstrated
by Lapidot et al. in 1994. In this seminal publication, the
injection of 2  105 CD34/CD38
AML cells into immunodecient SCID mice was sufcient to initiate an AML in the
transplanted recipients.2 Given the recent development of
more permissive immunodecient mouse models and transplantation techniques, considerably lower numbers of LSC
candidates with as few as 2001000 cells per animal have
been reported to result in stable engraftment of AML.17,18
Other authors have also indicated that LSC could be
enriched in the cell fraction that is CD33 or CD9019,20 or
CD34CD90
.21 A different approach is the staining with
broad intracellular dyes. The most prominent is Hoechst
33342, a mitochondrial dye that is also a substrate for ATPbinding cassette (ABC)-type transporters. The Hoechst
33342low population is termed side-population and has stem
cell features both in healthy and malignant cells.22

Pitfalls of using surface antigen markers

Most correlative studies have demonstrated LSC activity in
xenograft studies within the CD34/CD38
fraction. Taussig
et al.,23 however, recently demonstrated that the phenotype
of LSC from NPM-mutated AML is characterized by low
CD34 expression and is different to that reported for

Special Section Paper

Table 1. Development of blood stem cell and leukemia stem cell assays

2330

Leukemia stem cells

Special Section Paper

Table 2. Antigens and functional markers for ow-cytometry based stem cell sorting25,26
Staining target

Description

CD34

Marker antigen for hematopoietic stem and progenitor cells should be positive and also
reports on CD34
stem cells and leukemic stem cells23

CD38

Maturation marker, normal HSC should be negative for this marker

CD33

Myeloid lineage marker

CD90 (Thy-1)

HSC and myeloid LSC are usually CD90

CD117 (c-kit)

The receptor (tyrosine kinase) for stem cell factor regularly positive on HSC and myeloid LSC

CD123 (IL3RA)

Often a strong marker for LSC24

CLL-1

C-type lectin-like molecule-1, a stem cell marker, also for myeloid LSC23

Hoechst 33342

Intracellular dye, stem cells are low22, analysed as a so-called side-population

ALDH

High activity of ALDH in myeloid LSC25,26 and also HSC; see also Figure 1.

unselected AML. In some samples, the LSC were exclusively

CD34
, whereas other samples had both CD34
and CD34
LSC. Their study has suggested that there is greater heterogeneity in the phenotype of LSC than previously thought. There
are indications that LSC might change their phenotype and
LSC might even be found in multiple fractions with different
intensities of CD34 and CD38 expressions.
Hence present evidence indicates that leukemia blasts display extremely diverse molecular and phenotypic features,
which are reected correspondingly in their LSCs. In some
cases, aberrant markers are identied that can be very useful
for sorting strategies and are also clinically relevant for detection of minimal residual diseases. These aberrant surface
antigens are markers of lymphoid cells that are usually not
found on healthy myeloid cells, often T-cell antigens like
CD4 or CD7. The antibody panel used for characterization of
LSC therefore often comprises other markers such as CD90
(Thy-1), CD117 (c-kit) and CD123 (IL3RA). Especially
CD123 is in many cases a strong marker for leukemic stem
cells24 (Table 2).

Heterogeneity of the LSC candidates

Based on the evidence listed in the previous paragraph, LSCs
are probably fairly heterogeneous and using surface markers
alone is not adequate for enrichment of LSC candidates.
Our group has recently demonstrated that the aldehyde dehydrogenase (ALDH)bright/CD34high subset had the highest leukemic long-term culture initiating cell (LTC-IC) frequencies as
compared to other subsets. In a leukemia LTC-IC assay of individually sorted cells, i.e., single cell LTC-IC assay of each of
these aforementioned subsets of ALDHbright cells, the progeny
cell-colonies were assessed for cytogenetic markers characteristic of the original AML. We found that a varying proportion of
colonies showed the original cytogenetic aberrations. Our observation indicated that the LSCs are heterogeneous and that
some of the LSC might not bear the characteristic cytogenetic
marker at all. Another possibility is that under in vitro conditions, survival and growth of normal HSCs are preferentially
favored over that of LSCs, in analogy to engraftment experiments in xeno-transplantation models.27,28

20,21

, there are also reports on CD34CD90

21

LSC

Nevertheless, our results have provided evidence that

varying proportions of residual HSC might be found in LSC
preparations. Future challenge will include the separation of
normal HSC versus LSC from the same individual and preliminary experiments exploiting the expression of typical
aberrant markers on LSC candidates are promising (unpublished results). In conjunction with index sorting and single
cell deposition, the functional and molecular characteristics
of these LSC as well as HSC candidates from the same
patient can be compared in future studies.

Divisional kinetics and ALDH activity of normal and

leukemia stem cells
In a series of experiments, our group has provided evidence
that other parameters such as divisional kinetics and asymmetric divisions might facilitate the identication of the most
primitive HSC. Other authors have shown that, in analogy to
normal HSC, LSC have comparable slow divisional kinetics
and the ability for extensive self-renewal.29,30 We and others
have reported that a slow dividing fraction (PKHbright) of
HSC is superior to a fast dividing fraction (PKHdim) in
reconstituting the NOD/SCID mouse not only with myeloid
cells, but also T cells and B cells.27,31
In preliminary experiments, we attempted to isolate LSC
based on slow divisional kinetics using dilution of PKH
membrane dye or dilution of cytosolic carboxyuorescein
succinimidyl ester (CFSE) dye during divisions as a parameter. With this technology, we were able to isolate a very limited number of leukemia cells that fullled some of the criteria for LSC candidates. However, we were not able to recover
an adequate number for functional studies (Ho AD, unpublished results, 2011).27
Another recent development is the use of the enzyme
ALDH as a marker for primitive HSC.4,25,32 ALDH is a group
of enzymes catalyzing the conversion of a broad range of
aldehydes to the corresponding acid via a NAD-dependent
irreversible reaction. ALDH can be detected by activation of
the dye aldeuor and its high uorescence then signies hematopoietic and leukemic stem cells (Fig. 1). ALDHbright cells
have also been identied in hematopoietic stem cells (HSCs)
C 2011 UICC
Int. J. Cancer: 129, 23282336 (2011) V

2331

Special Section Paper

Buss and Ho

Figure 1. Example for the sorting of an ALDH leukemia stem cell population. FACS gating of a stained sample of blood or bone marrow
cells. (a) Gating on a population of intact leukocytes in an FSC/SSC plot ( size/granularity). (b) Gating on propidium iodide (PI) negative
cell ( living cells). (c) Gating on Aldeuor-positive cells as a marker for ALDH-activity, in uorescence channel FL1. There are about 2.4%
ALDH stem cells within the mononuclear cells of this sample. (d) As a control for specicity by inhibition of ALDH activity. (Ran, Schubert,
Eckstein; Reproduced with kind permission.39)

from umbilical cord blood.3335 High ALDH activity has

been reported to delineate distinct CD34 or CD133 stem
cell subsets that are more primitive than the ALDH-negative
fractions.32,36,37 These normal HSCs have been reported to
demonstrate LTC-IC and NOD/SCID mouse repopulating
activity.
Using ALDH activity and low side-scatter pattern as a parameter, Cheung et al. reported that LSC candidates could be
isolated from patients with AML.25 They showed that these
cells were able to engraft in the NOD/SCID mouse model
and induced human AML growth. In another study reported
by Pearce et al., the ALDHbright and CD34 subpopulation in
AML largely overlapped, but a signicant amount of ALDHbright
cells with LSC characteristics did not express the
CD34CD38
/low phenotype.4
Our group has shown that LSC candidates could be reproducibly enriched by combining the markers ALDH and
CD34.26 Functional studies in vitro have demonstrated that
ALDHbright cells from AML patients divided slowly, were
more adhesive to the stroma, gave rise to LTC-IC and leukemia colonies that showed the same cytogenetic aberrations as
C 2011 UICC
Int. J. Cancer: 129, 23282336 (2011) V

the parent leukemia.27 Moreover, ALDHbright/CD34 cells,

when compared with ALDHbright cells, yielded similar results.
Schubert et al. have demonstrated that in the NOD/SCID
mouse model, repopulating human AML initiating cells were
recovered from ALDHbright as well as from slow dividing
(PKHbright) cells.27 Divisional kinetics was determined by the
membrane dye (PKH) dilution method, as described previously by our group and other authors.31,38 Comparing all
these methods for enrichment of LSC candidates, we found
that isolation using ALDH activity, either alone or in combination with CD34 expression, yielded comparable results
according to functional parameters. The efciency of isolation
of viable ALDHbright cells for functional experiments was,
however, consistently higher than using other methods.

Clinical signicance of identifying LSC candidates

Using ALDH activity and low side-scatter pattern as a parameter, Cheung et al. reported that LSC candidates could be
isolated from 43 cases of AML.25 They then showed that the
presence of LSC was associated with adverse cytogenetic
markers, a strong leukemic engraftment in the NOD/SCID

Special Section Paper

2332

mouse model and probably with a poorer prognosis. Similarly, Pearce et al. showed that engraftment potential after i.v.
transplantation of unmanipulated AML cells (107 to 108) into
irradiated NOD/SCID immunodecient mice correlated signicantly with poor prognosis40; 51% of 59 cases in this
study engrafted successfully and possible associations with
homing capacities, white blood cell, FLT-3 and nucleophosmin mutations, CXCR4-expression were excluded. Nevertheless, engraftment of AML samples with cytogenetic aberrations (ve of ve), as well as a strong relationship between
engraftment in NOD/SCID mice and poor overall survival
(OS) (13 of 25) in these patients indicated the clinical relevance of engraftment data in the NOD/SCID model.
In a series of articles using a combination of surface
markers and side scatter characteristics, the group of Schuurhuis has reported that detection of high frequencies of LSC
in the marrow of patients with AML at diagnosis or in remission is associated with poor survival.5 Van Rhenen et al.41
suggested that aberrant marker patterns are expressed on the
CD34/CD38
cells in patients with AML which permitted
the separation of malignant from the normal stem cell compartments from the same patient. Using a combination of
forward and side scatter behaviors in ow cytometric studies
as well as expression of CD34 and aberrant markers for
detection of LSC, Terwijn et al.14,42 have documented that
detection of residual LSC in remission bone marrow predicts
relapse in patients with AML and represents an independent
prognostic factor in the identication of poor prognostic
patients. Albeit the experimental approaches were different,
the observations by these authors are compatible with the
notion that the frequency of LSC candidates at the time of
diagnosis correlates with poor prognosis.
In a recent study in 101 patients with AML, we have
demonstrated that ALDHbright cells in the marrow as a surrogate marker for LSC candidates at the time of diagnosis of
AML is an independent prognostic factor predicting refractory disease and poor clinical outcome.28 A remarkable nding is the signicant relationship between levels of LSC candidates at the time of diagnosis with the persistence of
leukemia blasts in the marrow after the rst induction chemotherapy (Ran et al., manuscript in preparation).
Univariate and multivariate Cox regression analyses on
relapse free survival (RFS) as well as on OS further conrmed
the relevance of LSC candidates for long-term outcome. In
univariate models, high frequencies of LSC candidates represent signicant prognostic factors for decreased RFS as well
as decreased OS. Similarly, genetic factors were also relevant
prognostic factors for progression free survival and OS. In
the multivariate model (stepwise regression analysis), frequency of ALDHbright cells was the strongest prognostic
marker affecting OS. Cytogenetic prognostic markers showed
no strong effect on OS in the multivariate model. Our observations support the notion that higher levels of LSC candidates are associated with resistant disease and that LSC cannot be eliminated by conventional chemotherapy alone.28

Leukemia stem cells

Interactions between LSC and bone marrow niche

The essential role of the interactions between stroma and tumor cells43 as well as normal HSC and the marrow niche for
maintenance of stem cell properties has been reported extensively. Within this context, adhesion molecules binding HSC
to the cellular determinants play a vital role.44,45 Most of the
evidence has been derived from studies in animal models. In
all the engraftment studies using immunodecient animal
models, the murine bone marrow environment represents a
substitute niche that is suboptimal and might not be extrapolated for engraftment of HSC or LSC in the human marrow
microenvironment. The results from animal models, especially on the interactions between human stem cells and the
marrow niche, should therefore be validated using human
cells. Within this context, the human mesenchymal stromal
cell (MSC) preparations might represent a more appropriate
surrogate. Problems of the effect of the microenvironment
and of the chosen host model have been discussed in detail
by Rosen and Jordan.46
In analogy, LSCs are maintained in a dormant state and
protected by the niche from cytotoxicity of chemotherapeutic
agents.47,48 Using HSC and MSC, derived from human marrow
as a surrogate model for the interaction between stem cells and
their niche, we have shown that direct contact between stem
cells and the microenvironment is essential in regulating asymmetric divisions and promoting stem cell renewal.44,4952
To characterize the interactions between human HSC and
stromal cells Wuchter et al. have systematically analysed the
homotypic cell-cell contact among HSC and MSC.53 Although
among HSC, dened as CD34/CD38
cells, no prominent
junctions of cell-cell contacts were evident, remarkable junctions and junction complexes were found between MSC. The
mesenchymal cells were interconnected by occasional gap
junctions and two morphotypes of adhering junctions, i.e.,
typical puncta adhaerentia and an abundant and elaborate
form of variously sized, invaginating villi-to-vermiform junction complex (complexus phalloides).
Using an immunodecient mouse model, Ishikawa et al.
have demonstrated that CD34CD38
AML LSC home and
engraft to the osteoblast rich endosteal area of the animals.17
Results from our group have demonstrated that co-culture of
leukemia blasts or LSC with human MSC as a surrogate
niche model would increase the resistance of these cells
against chemotherapeutic agents.27 Other authors have
reported the signicant role of the adhesive mechanisms
between LSC and the bone marrow niche, leading to dormancy and drug resistance of the LSC.48 The adhesion molecule CD44 might also play an essential role, as targeting of
CD44 might eradicate human myeloid LSC.54

Special features of lymphoid LSC

The rst phenotype of acute lymphoid leukemia (ALL) stem
cells was described as immature CD34/CD19
cells.55,56 This
was later challenged by other authors who reported ALL-initiating cells in the CD34/CD19 subset.57,58 Lymphoid LSC
C 2011 UICC
Int. J. Cancer: 129, 23282336 (2011) V

candidates derived from patients characterized by t(9;22) or

t(4;11) might be found in the CD34/CD19
subset.57,59 Serial
transplantation studies have suggested that human lymphoid
LSC with more mature phenotypes were able to repopulate and
propagate B-precursor ALL.60,61 Kong et al. reported that leukemia initiating activity could be found both in the CD34/
CD38/CD19 as well as in the CD34/CD38
/CD19 subset.
Thus, lymphoid LSCs, similar to the myeloid counterparts, are
fairly heterogeneous.60
By enumeration of copy number alterations (CNAs) of
the ETV6-RUNX1 fusion product in individually analyzed
LSC derived from childhood ALL, Anderson et al. have identied distinctive genetic signatures of subclones and their frequencies, based on which they have inferred the evolutionary
architecture of the leukemia subclones.62 By monitoring FISH
stainings, the ETV6-RUNX1 (TEL-AML1) fusion at different
time points of the disease and by serial transplantations into
immunodecient NOD/SCID/IL2Rcnull mice, they have demonstrated clonal diversity and evolution in the LSC department and concluded that clonal architecture is subject to
alterations at diagnosis and in relapse.62
Notta et al. have demonstrated the genetic diversity of leukemia initiating cells and they reported that many diagnostic
samples from patients with ALL contain multiple genetically
distinct subclones of LSC.63 This group focused on the evolution of adult Philadelphia positive (BCR-ABL positive) ALLinitiating cells. With copy-number alteration (CNA) proling,
they could demonstrate several subclones at diagnosis that
could be followed through repopulation studies in immunodecient mice. During the repopulation process, the contributions
of subclones changed and smaller ones could outgrow the initial major subclone. Their ndings implicate that there are
probably genetically distinct subclones at the LSC level that
undergo clonal evolution processes during the disease process.

Therapies directed against LSC

The ultimate goal of research in LSC is to induce long-term
cure for patients with AML by treatment strategies that will
enable us to eradicate the LSC. Different treatment principles
have been proposed in this respect.64,65
Molecules targeting proteins or pathways that are essential
and specic for survival and maintenance of LSC could be an
ideal approach.66 Thus far, there is little evidence for agents
that show specic effect against LSC, leaving normal HSC
unharmed. The combined inhibition of two specic pathways
together has been proven to be effective in eradicating LSC in
animal models. The rst pathway is the induction of oxidative
stress combined with the inhibition of nuclear factor jB (NFjB), physiologically transmitting survival signals. Guzman et
al.67 have reported that the proteasome inhibitor MG-132
might successfully inhibit both these processes and have led
to the preferential apoptosis of LSC in vitro and in vivo.
A compound that has attracted attention is parthenolide,
which is extracted from the herb feverfew. Preclinical experiments have demonstrated remarkable activity against LSC. The
C 2011 UICC
Int. J. Cancer: 129, 23282336 (2011) V

mechanism of cytotoxic activity was traced back to the inhibition

of NF-jB and exertion of cellular oxidative stress.68 To dene
the underlying mechanisms, a genetic expression prole was
generated from parthenolide-treated LSC. These results were
then exploited for generation of a search pattern for expression
databases to identify substances with similar molecular effects.
By means of this innovative method, two further agents have
been identied: celastrol and 4-hydroxy-2-nonenal (HNE),
which have been shown to effectively eradicate AML cells69 as
well as their corresponding progenitors and stem cells.
Another approach is inuencing the pertubation of the adhesion between LSC and their bone marrow niche.70,71 Mobilization or priming of dormant LSC should thus release LSC
from their protective microenvironment. First indications that
this strategy might work have been suggested by Lowenberg
et al. In this clinical trial, G-CSF has been administered before
chemotherapy to patients with AML as a priming strategy.72
The interpretation of the results of this study has remained
controversial. Recently the concept was strengthened by new
experimental data by Ishikawa and colleagues.18 The recent development of another molecule, plerixafor, a potent CXCR4
antagonist and modulator, might be promising. Pre-clinical
studies in animal models have shown encouraging results48,73
and a clinical trial in AML is already activated.74
A novel approach to make leukemic cells vulnerable to
chemotherapy is by forcing them into cell cycle by inducing
stress.75 One possible stress mechanism is the use of interferon-a (IFN-a) for chronic myeloid leukemia (CML). Based
on investigations in animal models, Trumpp et al. suggested
that the application of IFN-a before treatment with 5-FU
might lead to an exhaustion of the dormant stem cell pool in
murine model.59,76
A further approach was presented by the group of Pandol et al. who demonstrated the essential role of the PML
tumor suppressor protein already well known from acute
promyelocytic leukemia (APL) for the maintenance of
healthy HSC as well as CML LSC.77 Consequently, an inhibition of PML by the already clinically used arsenic trioxide
(As2O3) together with conventional chemotherapy lead to apoptosis of LSC and increased survival of mice in a transgenic
mouse model mimicking CML. Further targeted approaches
against CML stem cells are discussed.74,78

Conclusions
It is of utmost importance to understand the mechanisms of
interaction between cellular niche and HSCs versus LSCs to provide a basis for the development of more efcient treatment strategies. The application of such principles might induce long-term
cure as they could eradicate the ultimate source of leukemia.

Acknowledgement
The authors wish to thank Dan Ran, Mario Schubert, Volker Eckstein and
the team of the Bone Marrow Laboratory of the Department of Internal
Medicine V, Medical Center of the University of Heidelberg, for providing
the gures for this manuscript.

Special Section Paper

2333

Buss and Ho

2334

Leukemia stem cells

References

Special Section Paper

1.

Reya T, Morrison SJ, Clarke MF,

Weissman IL. Stem cells, cancer, and
cancer stem cells. Nature 2001;414:10511.
2. Lapidot T, Sirard C, Vormoor J, Murdoch
B, Hoang T, Caceres-Cortes J, Minden M,
Paterson B, Caligiuri MA, Dick JE. A cell
initiating human acute myeloid leukaemia
after transplantation into SCID mice.
Nature 1994;367:6458.
3. Bonnet D, Dick JE. Human acute myeloid
leukemia is organized as a hierarchy that
originates from a primitive hematopoietic
cell. Nat Med 1997;3:7307.
4. Pearce DJ, Taussig D, Simpson C, Allen K,
Rohatiner AZ, Lister TA, Bonnet D.
Characterization of cells with a high
aldehyde dehydrogenase activity from cord
blood and acute myeloid leukemia samples.
Stem Cells 2005;23:75260.
5. van Rhenen A, Feller N, Kelder A, Westra
AH, Rombouts E, Zweegman S, van der
Pol MA, Waissz Q, Ossenkoppele GJ,
Schuurhuis GJ. High stem cell frequency in
acute myeloid leukemia at diagnosis
predicts high minimal residual disease and
poor survival. Clin Cancer Res 2005;11:
65207.
6. Ho AD, Wagner W. Bone marrow niche
and leukemia. Ernst Schering Found Symp
Proc Vol. 2006-5, 2007:12539.
7. Shultz LD, Ishikawa F, Greiner DL.
Humanized mice in translational
biomedical research. Nat Rev Immunol
2007;7:11830.
8. Ishikawa F, Yasukawa M, Lyons B, Yoshida
S, Miyamoto T, Yoshimoto G, Watanabe
T, Akashi K, Shultz LD, Harada M.
Development of functional human blood
and immune systems in NOD/SCID/IL2
receptor {gamma} chain(null) mice. Blood
2005;106:156573.
9. Shultz LD, Lang PA, Christianson SW,
Gott B, Lyons B, Umeda S, Leiter E,
Hesselton R, Wagar EJ, Leif JH, Kollet O,
Lapidot T, et al. NOD/LtSz-Rag1null mice:
an immunodecient and radioresistant
model for engraftment of human
hematolymphoid cells. HIV infection, and
adoptive transfer of NOD mouse
diabetogenic T cells. J Immunol 2000;164:
2496507.
10. Wunderlich M, Chou FS, Link KA,
Mizukawa B, Perry RL, Carroll M, Mulloy
JC. AML xenograft efciency is
signicantly improved in NOD/SCIDIL2RG mice constitutively expressing
human SCF GM-CSF and IL-3. Leukemia
2010;24:17858.
11. Till JE, McCulloch EA. A direct
measurement of the radiation sensitivity of
normal mouse bone marrow cells. Radiat
Res 1961;14:21322.

12. Pluznik DH, Sachs L. The cloning of

normal mast cells in tissue culture. J Cell
Physiol 1965;66:31924.
13. Bradley TR, Metcalf D. The growth of
mouse bone marrow cells in vitro. Aust J
Exp Biol Med Sci 1966;44:28799.
14. Dexter TM, Allen TD, Lajtha LG.
Conditions controlling the proliferation of
haemopoietic stem cells in vitro. J Cell
Physiol 1977;91:33544.
15. Ploemacher R, van der Sluijs J, Voerman J,
Brons N. An in vitro limiting-dilution
assay of long-term repopulating
hematopoietic stem cells in the mouse.
Blood 1989;74:275563.
16. Taussig DC, Miraki-Moud F, Anjos-Afonso
F, Pearce DJ, Allen K, Ridler C, Lillington D,
Oakervee H, Cavenagh J, Agrawal SG, Lister
TA, Gribben JG, et al. Anti-CD38 antibodymediated clearance of human repopulating
cells masks the heterogeneity of leukemiainitiating cells. Blood 2008;112:56875.
17. Ishikawa F, Yoshida S, Saito Y, Hijikata A,
Kitamura H, Tanaka S, Nakamura R,
Tanaka T, Tomiyama H, Saito N, Fukata
M, Miyamoto T, et al. Chemotherapyresistant human AML stem cells home to
and engraft within the bone-marrow
endosteal region. Nat Biotechnol 2007;25:
131521.
18. Saito Y, Uchida N, Tanaka S, Suzuki N,
Tomizawa-Murasawa M, Sone A, Najima
Y, Takagi S, Aoki Y, Wake A, Taniguchi S,
Shultz LD, et al. Induction of cell cycle
entry eliminates human leukemia stem
cells in a mouse model of AML. Nat
Biotechnol 2010;28:27580.
19. Buccisano F, Rossi FM, Venditti A, Del
Poeta G, Cox MC, Abbruzzese E, Rupolo
M, Berretta M, Degan M, Russo S,
Tamburini A, Maurillo L, et al. CD90/Thy1 is preferentially expressed on blast cells
of high risk acute myeloid leukaemias. Br J
Haematol 2004;125:20312.
20. Nilsson L, Astrand-Grundstrom I,
Anderson K, Arvidsson I, Hokland P,
Bryder D, Kjeldsen L, Johansson B,
Hellstrom-Lindberg E, Hast R, Jacobsen
SE. Involvement and functional
impairment of the CD34()CD38(-)Thy1() hematopoietic stem cell pool in
myelodysplastic syndromes with trisomy 8.
Blood 2002;100:25967.
21. Blair A, Hogge DE, Ailles LE, Lansdorp
PM, Sutherland HJ. Lack of expression of
Thy-1 (CD90) on acute myeloid leukemia
cells with long-term proliferative ability in
vitro and in vivo. Blood 1997;89:310412.
22. Moshaver B, van Rhenen A, Kelder A, van
der Pol M, Terwijn M, Bachas C, Westra
AH, Ossenkoppele GJ, Zweegman S,
Schuurhuis GJ. Identication of a small

23.

24.

25.

26.

27.

28.

29.

30.

31.

subpopulation of candidate leukemiainitiating cells in the side population of

patients with acute myeloid leukemia. Stem
Cells 2008;26:305967.
Taussig DC, Vargaftig J, Miraki-Moud F,
Griessinger E, Sharrock K, Luke T,
Lillington D, Oakervee H, Cavenagh J,
Agrawal SG, Lister TA, Gribben JG, et al.
Leukemia-initiating cells from some acute
myeloid leukemia patients with mutated
nucleophosmin reside in the CD34(-)
fraction. Blood 2010;115:197684.
Jin L, Lee EM, Ramshaw HS, Buseld SJ,
Peoppl AG, Wilkinson L, Guthridge MA,
Thomas D, Barry EF, Boyd A, Gearing DP,
Vairo G, et al. Monoclonal antibodymediated targeting of CD123 IL-3 receptor
alpha chain eliminates human acute
myeloid leukemic stem cells. Cell Stem Cell
2009;5:3142.
Cheung AM, Wan TS, Leung JC, Chan LY,
Huang H, Kwong YL, Liang R, Leung AY.
Aldehyde dehydrogenase activity in
leukemic blasts denes a subgroup of acute
myeloid leukemia with adverse prognosis
and superior NOD/SCID engrafting
potential. Leukemia 2007;21:142330.
Ran D, Schubert M, Pietsch L, Taubert I,
Wuchter P, Eckstein V, Bruckner T,
Zoeller M, Ho AD. Aldehyde
dehydrogenase activity among primary
leukemia cells is associated with stem cell
features and correlates with adverse clinical
outcomes. Exp Hematol 2009;37:142334.
Schubert M, Herbert N, Taubert I, Ran D,
Singh R, Eckstein V, Vitacolonna M, Ho
AD, Zoller M. Differential survival of AML
subpopulations in NOD/SCID mice. Exp
Hematol 2011;39:25063.
Ran D, Schubert M, Taubert I, Eckstein V,
Bellos F, Jauch A, Bruckner T, Chen H,
Saffrich R, Wuchter P, Ho AD. Frequency
of leukemia stem cell candidates at
diagnosis of acute myeloid leukemia is a
signicant prognostic factor for response.
2011, submitted.
Holyoake T, Jiang X, Eaves C, Eaves A.
Isolation of a highly quiescent
subpopulation of primitive leukemic cells
in chronic myeloid leukemia. Blood 1999;
94:205664.
Holtz MS, Forman SJ, Bhatia R.
Nonproliferating CML CD34 progenitors
are resistant to apoptosis induced by a
wide range of proapoptotic stimuli.
Leukemia 2005;19:103441.
Vitacolonna M, Schubert M, Herbert N,
Taubert I, Singh R, Ho A, Zoller M.
Improved T and B cell recovery by the
transfer of slowly dividing human
hematopoietic stem cells. Leuk Res 2010;34:
62230.

C 2011 UICC
Int. J. Cancer: 129, 23282336 (2011) V

32. Storms RW, Trujillo AP, Springer JB, Shah

L, Colvin OM, Ludeman SM, Smith C.
Isolation of primitive human hematopoietic
progenitors on the basis of aldehyde
dehydrogenase activity. Proc Natl Acad Sci
USA 1999;96:911823.
33. Christ O, Lucke K, Imren S, Leung K,
Hamilton M, Eaves A, Smith C, Eaves C.
Improved purication of hematopoietic
stem cells based on their elevated aldehyde
dehydrogenase activity. Haematologica
2007;92:116572.
34. Muramoto GG, Russell JL, Sa R, Salter
AB, Himburg HA, Daher P, Meadows SK,
Doan P, Storms RW, Chao NJ, McDonnell
DP, Chute JP. Inhibition of aldehyde
dehydrogenase expands hematopoietic stem
cells with radioprotective capacity. Stem
Cells 2010;28:52334.
35. Pierre-Louis O, Clay D, Brunet de la
Grange P, Blazsek I, Desterke C, Guerton
B, Blondeau C, Malfuson JV, Prat M,
Bennaceur-Griscelli A, Lataillade JJ, Le
Bousse-Kerdiles MC. Dual SP/ALDH
functionalities rene the human
hematopoietic Lin-CD34CD38- stem/
progenitor cell compartment. Stem Cells
2009;27:255262.
36. Gentry T, Foster S, Winstead L, Deibert E,
Fiordalisi M, Balber A. Simultaneous
isolation of human BM hematopoietic,
endothelial and mesenchymal progenitor
cells by ow sorting based on aldehyde
dehydrogenase activity: implications for cell
therapy. Cytotherapy 2007;9:25974.
37. Hess DA, Meyerrose TE, Wirthlin L, Craft
TP, Herrbrich PE, Creer MH, Nolta JA.
Functional characterization of highly
puried human hematopoietic repopulating
cells isolated according to aldehyde
dehydrogenase activity. Blood 2004;104:
164855.
38. Young JC, Varma A, DiGiusto D, Backer
MP. Retention of quiescent hematopoietic
cells with high proliferative potential
during ex vivo stem cell culture. Blood
1996;87:54556.
39. Buss EC, Ho AD. Cancer stem cells
nding and hitting the roots of cancer. In:
Emmert-Streib F, Dehmer M, eds. Medical
biostatistics for complex diseases.
Weinheim: Wiley-VCH, 2010.2544.
40. Pearce DJ, Taussig D, Zibara K, Smith LL,
Ridler CM, Preudhomme C, Young BD,
Rohatiner AZ, Lister TA, Bonnet D. AML
engraftment in the NOD/SCID assay
reects the outcome of AML: implications
for our understanding of the heterogeneity
of AML. Blood 2006;107:116673.
41. van Rhenen A, van Dongen GA, Kelder A,
Rombouts EJ, Feller N, Moshaver B,
Stigter-van Walsum M, Zweegman S,
Ossenkoppele GJ, Jan Schuurhuis G. The
novel AML stem cell associated antigen
CLL-1 aids in discrimination between

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

C 2011 UICC
Int. J. Cancer: 129, 23282336 (2011) V

normal and leukemic stem cells. Blood

2007;110:265966.
Terwijn M, Feller N, van Rhenen A, Kelder
A, Westra G, Zweegman S, Ossenkoppele
G, Schuurhuis GJ. Interleukin-2 receptor
alpha-chain (CD25) expression on
leukaemic blasts is predictive for outcome
and level of residual disease in AML. Eur J
Cancer 2009;45:16929.
Mueller MM, Fusenig NE. Friends or foes
bipolar effects of the tumour stroma in
cancer. Nat Rev Cancer 2004;4:83949.
Wagner W, Roderburg C, Wein F,
Diehlmann A, Frankhauser M, Schubert R,
Eckstein V, Ho AD. Molecular and
secretory proles of human mesenchymal
stromal cells and their abilities to maintain
primitive hematopoietic progenitors. Stem
Cells 2007;25:263847.
Walenda T, Bork S, Horn P, Wein F,
Saffrich R, Diehlmann A, Eckstein V, Ho
AD, Wagner W. Co-culture with
mesenchymal stromal cells increases
proliferation and maintenance of
haematopoietic progenitor cells. J Cell Mol
Med 2010;14:33750.
Rosen JM, Jordan CT. The increasing
complexity of the cancer stem cell
paradigm. Science 2009;324:16703.
Rozenveld-Geugien M, Baas IO, van
Gosliga D, Vellenga E, Schuringa JJ.
Expansion of normal and leukemic
human hematopoietic stem/progenitor
cells requires rac-mediated interaction
with stromal cells. Exp Hematol 2007;35:
78292.
Zeng Z, Shi YX, Samudio IJ, Wang RY,
Ling X, Frolova O, Levis M, Rubin JB,
Negrin RR, Estey EH, Konoplev S,
Andreeff M, et al. Targeting the leukemia
microenvironment by CXCR4 inhibition
overcomes resistance to kinase inhibitors
and chemotherapy in AML. Blood 2009;
113:621524.
Wagner W, Saffrich R, Wirkner U,
Eckstein V, Blake J, Ansorge A, Schwager
C, Wein F, Miesala K, Ansorge W, Ho
AD. Hematopoietic progenitor cells and
cellular microenvironment: behavioral and
molecular changes upon interaction. Stem
Cells 2005;23:118091.
Wagner W, Wein F, Roderburg C, Saffrich R,
Faber A, Krause U, Schubert M, Benes V,
Eckstein V, Maul H, Ho AD. Adhesion of
hematopoietic progenitor cells to human
mesenchymal stem cells as a model for cellcell interaction. Exp Hematol 2007;35:
31425.
Walenda T, Bokermann G, Ferreira MV,
Piroth DM, Hieronymus T, Neuss S, Zenke
M, Ho AD, Muller AM, Wagner W.
Synergistic effects of growth factors and
mesenchymal stromal cells for expansion
of hematopoietic stem and progenitor cells.
Exp Hematol, in press.

52. Wein F, Pietsch L, Saffrich R, Wuchter P,

Walenda T, Bork S, Horn P, Diehlmann A,
Eckstein V, Ho AD, Wagner W. Ncadherin is expressed on human
hematopoietic progenitor cells and
mediates interaction with human
mesenchymal stromal cells. Stem Cell Res
2010;4:12939.
53. Wuchter P, Boda-Heggemann J, Straub BK,
Grund C, Kuhn C, Krause U, Seckinger A,
Peitsch WK, Spring H, Ho AD, Franke
WW. Processus and recessus adhaerentes:
giant adherens cell junction systems
connect and attract human mesenchymal
stem cells. Cell Tissue Res 2007;328:
499514.
54. Jin L, Hope KJ, Zhai Q, Smadja-Joffe F,
Dick JE. Targeting of CD44 eradicates
human acute myeloid leukemic stem cells.
Nat Med 2006;12:116774.
55. Cobaleda C, Gutierrez-Cianca N, PerezLosada J, Flores T, Garcia-Sanz R,
Gonzalez M, Sanchez-Garcia I. A primitive
hematopoietic cell is the target for the
leukemic transformation in human
philadelphia-positive acute lymphoblastic
leukemia. Blood 2000;95:100713.
56. Cox CV, Evely RS, Oakhill A, Pamphilon
DH, Goulden NJ, Blair A. Characterization
of acute lymphoblastic leukemia progenitor
cells. Blood 2004;104:291925.
57. Castor A, Nilsson L, Astrand-Grundstrom
I, Buitenhuis M, Ramirez C, Anderson K,
Strombeck B, Garwicz S, Bekassy AN,
Schmiegelow K, Lausen B, Hokland P, et
al. Distinct patterns of hematopoietic stem
cell involvement in acute lymphoblastic
leukemia. Nat Med 2005;11:6307.
58. Hong D, Gupta R, Ancliff P, Atzberger A,
Brown J, Soneji S, Green J, Colman S,
Piacibello W, Buckle V, Tsuzuki S, Greaves
M, et al. Initiating and cancer-propagating
cells in TEL-AML1-associated childhood
leukemia. Science 2008;319:3369.
59. Hotlder M, Rottgers S, Rosemann A,
Schrauder A, Schrappe M, Pieters R,
Jurgens H, Harbott J, Vormoor J.
Leukemic stem cells in childhood high-risk
ALL/t(9;22) and t(4;11) are present in
primitive lymphoid-restricted
CD34CD19- cells. Cancer Res 2005;65:
14429.
60. Kong Y, Yoshida S, Saito Y, Doi T,
Nagatoshi Y, Fukata M, Saito N, Yang SM,
Iwamoto C, Okamura J, Liu KY, Huang
XJ, et al. CD34CD38CD19 as well as
CD34CD38-CD19 cells are leukemiainitiating cells with self-renewal capacity in
human B-precursor ALL. Leukemia 2008;
22:120713.
61. le Viseur C, Hotlder M, Bomken S,
Wilson K, Rottgers S, Schrauder A,
Rosemann A, Irving J, Stam RW, Shultz
LD, Harbott J, Jurgens H, et al. In
childhood acute lymphoblastic leukemia,

Special Section Paper

2335

Buss and Ho

2336

Special Section Paper

62.

63.

64.

65.

66.

67.

blasts at different stages of

immunophenotypic maturation have stem
cell properties. Cancer Cell 2008;14:
4758.
Anderson K, Lutz C, van Delft FW,
Bateman CM, Guo Y, Colman SM,
Kempski H, Moorman AV, Titley I,
Swansbury J, Kearney L, Enver T, et al.
Genetic variegation of clonal architecture
and propagating cells in leukaemia. Nature
2011;469:35661.
Notta F, Mullighan CG, Wang JC, Poeppl
A, Doulatov S, Phillips LA, Ma J, Minden
MD, Downing JR, Dick JE. Evolution of
human BCR-ABL1 lymphoblastic
leukaemia-initiating cells. Nature 2011;469:
3627.
Burnett A, Wetzler M, Lowenberg B.
Therapeutic advances in acute myeloid
leukemia. J Clin Oncol 2011;29:48794.
Smits EL, Lee C, Hardwick N, Brooks S,
Van Tendeloo VF, Orchard K, Guinn BA.
Clinical evaluation of cellular
immunotherapy in acute myeloid
leukaemia. Cancer Immunol Immunother
2011;60:75769.
Roboz GJ, Guzman M. Acute myeloid
leukemia stem cells: seek and destroy.
Expert Rev Hematol 2009;2:66372.
Guzman ML, Swiderski CF, Howard DS,
Grimes BA, Rossi RM, Szilvassy SJ, Jordan
CT. Preferential induction of apoptosis for

Leukemia stem cells

68.

69.

70.

71.

72.

73.

primary human leukemic stem cells. Proc

Natl Acad Sci USA 2002;99:162205.
Guzman ML, Rossi RM, Karnischky L, Li
X, Peterson DR, Howard DS, Jordan CT.
The sesquiterpene lactone parthenolide
induces apoptosis of human acute
myelogenous leukemia stem and progenitor
cells. Blood 2005;105:41639.
Hassane DC, Guzman ML, Corbett C, Li
X, Abboud R, Young F, Liesveld JL, Carroll
M, Jordan CT. Discovery of agents that
eradicate leukemia stem cells using an in
silico screen of public gene expression data.
Blood 2008;111:565462.
Lane SW, Scadden DT, Gilliland DG. The
leukemic stem cell niche: current concepts
and therapeutic opportunities. Blood 2009;
114:11507.
Konopleva MY, Jordan CT. Leukemia stem
cells and microenvironment: biology and
therapeutic targeting. J Clin Oncol 2011;29:
5919.
Lowenberg B, van Putten W, Theobald M,
Gmur J, Verdonck L, Sonneveld P, Fey M,
Schouten H, de Greef G, Ferrant A,
Kovacsovics T, Gratwohl A, et al. Effect of
priming with granulocyte colonystimulating factor on the outcome of
chemotherapy for acute myeloid leukemia.
N Engl J Med 2003;349:74352.
Buss EC, Kalinkovich A, Schajnovitz A,
Kollet O, Dar A, Tesio M, Fruehauf S,

74.

75.

76.

77.

78.

Hotlder M, Ho AD, Shultz LD, Lapidot

T. In vivo mobilization of leukemic human
precursor-B-ALL cells by the CXCR4antagonist AMD3100 is via secretion of
SDF-1 and synergistically by catecholamine
action. ASH Ann Meet Abstr 2008;112:
1920.
Konopleva M, Zhihong Z, Wang R-Y,
Thall PF, McCormick G, Lu H, Chen JJ,
Shpall EJ, Ciurea SO, Kebriaei P, Alousi
AM, Popat U, et al. A phase I/II trial of
plerixafor/G-CSF combined with IV Bu/Flu
conditioning regimen in AML/MDS
patients undergoing allogenic stem cell
transplantation. ASH Ann Meet Abstr 2010;
116:2358.
Trumpp A, Essers M, Wilson A.
Awakening dormant haematopoietic stem
cells. Nat Rev Immunol 2010;10:2019.
Essers MA, Offner S, Blanco-Bose WE,
Waibler Z, Kalinke U, Duchosal MA,
Trumpp A. IFNalpha activates dormant
haematopoietic stem cells in vivo. Nature
2009;458:9048.
Ito K, Bernardi R, Morotti A, Matsuoka S,
Saglio G, Ikeda Y, Rosenblatt J, Avigan DE,
Teruya-Feldstein J, Pandol PP. PML
targeting eradicates quiescent leukaemiainitiating cells. Nature 2008;453:10728.
Pellicano F, Sinclair A, Holyoake T. In
search of CML stem cells deadly weakness.
Curr Hematol Malig Rep 2011;6:827.

C 2011 UICC
Int. J. Cancer: 129, 23282336 (2011) V