Rapid mobilization of hematopoietic progenitors by AMD3100 and catecholamines is mediated by CXCR4-dependent SDF-1 release from bone marrow stromal cells.

Abstract

Steady-state egress of hematopoietic progenitor cells can be rapidly amplified by mobilizing agents such as AMD3100, the mechanism, however, is poorly understood. We report that AMD3100 increased the homeostatic release of the chemokine stromal cell derived factor-1 (SDF-1) to the circulation in mice and non-human primates. Neutralizing antibodies against CXCR4 or SDF-1 inhibited both steady state and AMD3100-induced SDF-1 release and reduced egress of murine progenitor cells over mature leukocytes. Intra-bone injection of biotinylated SDF-1 also enhanced release of this chemokine and murine progenitor cell mobilization. AMD3100 directly induced SDF-1 release from CXCR4(+) human bone marrow osteoblasts and endothelial cells and activated uPA in a CXCR4/JNK-dependent manner. Additionally, ROS inhibition reduced AMD3100-induced SDF-1 release, activation of circulating uPA and mobilization of progenitor cells. Norepinephrine treatment, mimicking acute stress, rapidly increased SDF-1 release and progenitor cell mobilization, whereas β2-adrenergic antagonist inhibited both steady state and AMD3100-induced SDF-1 release and progenitor cell mobilization in mice. In conclusion, this study reveals that SDF-1 release from bone marrow stromal cells to the circulation emerges as a pivotal mechanism essential for steady-state egress and rapid mobilization of hematopoietic progenitor cells, but not mature leukocytes.

Full text

Page 1

ORIGINAL ARTICLE
Rapid mobilization of hematopoietic progenitors by AMD3100 and catecholamines
is mediated by CXCR4-dependent SDF-1 release from bone marrow stromal cells
A Dar1,7, A Schajnovitz1,7, K Lapid1,7, A Kalinkovich1, T Itkin1, A Ludin1, W-M Kao2, M Battista2, M Tesio1, O Kollet1,
NN Cohen1, R Margalit1, EC Buss1, F Baleux3, S Oishi4, N Fujii4, A Larochelle5, CE Dunbar5, HE Broxmeyer6, PS Frenette2
and T Lapidot1
1Department of Immunology, Weizmann Institute of Science, Rehovot, Israel;2Department of Medicine, Immunology Institute,
and Black Family Stem Cell Institute, Mount Sinai School of Medicine, New York, NY, USA;3Institute Pasteur, Paris, France;
4Graduate School of Pharmaceutical Sciences, Kyoto University, Sakyo-ku, Kyoto, Japan;5Molecular Hematopoiesis Section,
Hematology Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MA, USA and
6Department of Microbiology and Immunology, Indiana University School of Medicine, Indianapolis, IN, USA
Steady-state egress of hematopoietic progenitor cells can be
rapidly amplified by mobilizing agents such as AMD3100, the
mechanism, however, is poorly understood. We report that
AMD3100 increased the homeostatic release of the chemokine
stromal cell derived factor-1 (SDF-1) to the circulation in mice
and non-human primates. Neutralizing antibodies against
CXCR4 or SDF-1 inhibited both steady state and AMD3100-
induced SDF-1 release and reduced egress of murine progenitor
cells over mature leukocytes. Intra-bone injection of biotiny-
lated SDF-1 also enhanced release of this chemokine and
murine progenitor cell mobilization. AMD3100 directly induced
SDF-1 release from CXCR4þhuman bone marrow osteoblasts
and endothelial cells and activated uPA in a CXCR4/JNK-
dependentmanner. Additionally,
AMD3100-induced SDF-1 release, activation of circulating uPA
and mobilization of progenitor cells. Norepinephrine treatment,
mimicking acute stress, rapidly increased SDF-1 release and
progenitor cell mobilization, whereas b2-adrenergic antagonist
inhibited both steady state and AMD3100-induced SDF-1
release and progenitor cell mobilization in mice. In conclusion,
this study reveals that SDF-1 release from bone marrow stromal
cells to the circulation emerges as a pivotal mechanism
essential for steady-state egress and rapid mobilization of
hematopoietic progenitor cells, but not mature leukocytes.
Leukemia advance online publication, 15 April 2011;
doi:10.1038/leu.2011.62
Keywords: rapid mobilization; AMD3100; catecholamines; uPA;
SDF-1/CXCR4; hematopoietic progenitor cells
ROSinhibition reduced
Introduction
Proliferation and differentiation of primitive hematopoietic stem
cells in the bone marrow (BM) reservoir is followed by leukocyte
release to the circulation. This process is regulated by dynamic
interactions between the nervous and immune systems with the
stromal microenvironment.1,2Although the majority of stem and
progenitor cells reside within the BM, a very small subset of
immature cells are also found in the peripheral blood as part of
steady-state homeostasis.3However, the mechanisms governing
progenitor cell egress to the circulation are currently poorly
defined. The basal low levels of circulating progenitor cells are
dramatically amplified by stress signals such as injury, bleeding
and bacterial or viral infection, presumably contributing to host
defense and repair mechanisms.4Clinical stem cell mobilization
regimens, including repeated daily stimulations with the
cytokine granulocyte colony stimulating factor (G-CSF), mimic
this process, leading to enhanced proliferation, differentiation
and recruitment of stem and progenitor cells to the circulation,
allowing their harvest for stem cell transplantation protocols.5–8
The chemokine stromal cell derived factor-1 (SDF-1, CXCL12) is
a potent chemoattractant for human and murine hematopoietic
stem cells.9SDF-1 and CXCR4 are highly expressed in human
and murine BM endothelium, reticular cells and endosteal
osteoblasts.9–12Enhancement of plasma SDF-1 levels using
adenoviral vectors,13stabilized methionine-SDF-1 (ref. 14) or
injection of sulfated polysaccharides,15,16as well as adminis-
tration of a CXCR4 agonist17correlated with induced progenitor
cell mobilization. Complementing the established role of the
SDF-1/CXCR4 axis in mobilization, the sympathetic nervous
system recently emerged as a novel regulator of stem and
progenitor cell egress from the BM in steady state,18as well as
following G-CSF administration, via norepinephrine (NE)
signaling, suppression of osteoblast function and downregula-
tion of SDF-1 in the bone.19Neurotransmitters together with
myeloid cytokines also directly regulate human progenitor cell
migration and development, as well as in vivo proliferation and
mobilization of murine progenitor cells.20In addition, a role for
the fibrinolytic system in G-CSF mobilization was recently
demonstrated, as G-CSF-induced mobilization resulted in
increased levels of chemotactic soluble plasminogen activator
receptor (uPAR),21whereas addition of plasmin to G-CSF
increased mobilization of both murine and human hematopoie-
tic progenitors.22G-CSF-induced mobilization also involves
Reactive Oxygen Species (ROS) generation in hematopoietic
progenitors, correlating with their enhanced egress and motility,
involving c-Met signaling.23Although G-CSF-induced mobiliza-
tion is a multi-step process that includes enhanced proliferation
and differentiation in the BM, rapid mobilization protocols are
characterized by recruitment of stem and progenitor cells from
the existing BM reservoir to the circulation within a few hours
after a single injection of the mobilizing agent.8,24One such
agent is AMD3100 (also termed plerixafor), which inhibits SDF-1-
mediated migration in vitro by blocking the chemokine binding
to its major receptor CXCR4.25,26AMD3100 has been shown to
rapidly mobilize immature progenitor cells from the BM into the
blood in murine,27non-human primate28and humans.26,29It
has been recently approved for clinical mobilization in
Received 2 February 2010; revised 10 February 2011; accepted
23 February 2011
Correspondence: Professor T Lapidot, Department of Immunology,
Weizmann Institute of Science, Rehovot 76100, Israel.
E-mail: Tsvee.Lapidot@weizmann.ac.il
7These authors contributed equally to this work.
Leukemia (2011), 1–11
& 2011 Macmillan Publishers Limited All rights reserved 0887-6924/11
www.nature.com/leu

Page 2

lymphoma and multiple myeloma patients undergoing auto-
logous transplantation.26
When
AMD3100 synergistically augments mobilization of human,
primate and murine progenitor cells, which have increased in
vitro migration to a gradient of SDF-1 and repopulation of
transplanted non-obese diabetic/severe combined immunodefi-
cient mice.27,30–32However, the mechanisms mediating rapid
mobilization of stem and progenitor cells from the BM are
poorly understood. As SDF-1/CXCR4 interactions are needed for
stem and progenitor cell homing and retention in the BM, it is
thought that disruption of these interactions by CXCR4
antagonists, such as AMD3100 or T140, is the cause of
accelerated egress to the circulation.8,24,27–29Unexpectedly,
we have now discovered that AMD3100 has an additional
active role in the rapid mobilization of CXCR4þimmature
murine progenitor cells. Both AMD3100- and NE-induced
mobilization are dependent on enhanced release of functional
SDF-1 to the murine circulation from activated BM stromal cells.
Moreover, AMD3100 treatment activates the protease uPA in a
CXCR4 and JNK-dependent manner.
combined withG-CSF,
Materials and methods
See Supplementary Information in the Leukemia website (http://
www.nature.com/leu) for detailed description of experimental
protocols.
Cell culture and reagents
Human BM aspirates were obtained from healthy donors for
allogeneic transplantations following informed consent. Human
primary BM endothelial cells (BMEC) were isolated from these
samples as previously described.11The purity of endothelial
cells, at first passage, was 91–94% as assessed by expression of
von Willebrand’s factor (Dako, Glostrup, Denmark), determined
by flow cytometry. The human osteosarcoma cell line MG63
were cultured, as described.10The following reagents were
used: recombinant human SDF-1a (R&D Systems, Minneapolis,
MN, USA); AMD3100 (Sigma-Aldrich, Rehovot, Israel or
Anormed, Langley, BC, CanadaFsee Supplementary Informa-
tion); N-Acetyl-L-Cysteine (NAC), NE and ICI (Sigma-Aldrich)
and JNK inhibitor (Calbiochem, Darmstadt, Germany). Chemi-
cally synthesized SDF-1 and biotinylated SDF-1 3/6 (bSDF-1 3/
6) were kindly provided by Drs Fujii (Kyoto, Japan) and Baleux
(Paris, France), respectively.
Animals and treatment procedures
The Weizmann Institute Animal Care and Use Committee
approved all animal experiments. BALB/c and C57BL/6 mice
(8–10 weeks) were purchased from Harlan, (Rehovot, Israel).
Non-obese diabetic/severe combined immunodeficient mice
were bred and maintained as described.7Rhesus macaques
(Macaca mulatta) were housed and handled in accordance with
guidelines set by the Committee of Care and Use of Laboratory
Animals of the Institute of Laboratory Animal Resources,
National Research Council, and the protocol was approved by
the Animal Care and Use Committee of the National Heart,
Lung and Blood Institute. For functional SDF-1 concentration
analyses and cell mobilization studies, mice received either
subcutaneous injections of AMD3100 (5mg/kg in 200ml
phosphate-buffered saline (PBS)), intraperitoneal injections of
either NE (5mg/kg in 400ml PBS), b2 adrenergic antagonist ICI
(5mg/kg in 400ml PBS) or intra-femoral injections of bSDF-1 3/6
(5mg in 5ml of PBS) to anesthetized mice. Control mice received
matched injections of PBS. Where indicated, a total of 50mg
mouse anti-human CXCR4 (12G5, R&D Systems), rabbit anti-
rat-CXCR4 (Torrey Pines Biolabs, East Orange, NJ, USA), anti-
SDF-1 (K15C, INRA, Paris, France) or matched control IgG Ab
were first injected into the peritoneum (25mg in PBS) followed
by intravenous injection of additional 25mg Ab, 30min later,
with or without subcutaneously injected AMD3100. A double
dose of Ab (2?50mg) was given to the C57BL/6 strain.
Noteworthy, the anti-human CXCR4 Ab (clone 12G5) that was
used in our model is cross-reactive to murine CXCR4 (ref. 11)
and is also capable of efficiently inhibiting migration of murine
BM mononuclear cells towards SDF-1 similarly to the poly-
clonal rabbit anti-rat CXCR4 Ab (Supplementary Figure S1).
SP600125, JNK inhibitor (30mg/kg) in 200ml PPCES vehicle or
PPCES alone (control) was administered into the peritoneum
15min before AMD3100 administration.33See Supplementary
Information for NAC treatment in vivo. Mice were asphyxiated
with CO2 and peripheral blood samples were collected by
cardiac aspiration in heparinized tubes and BM was flushed
with PBS. Where indicated, mice were killed exactly 10–60min
post intra-femoral injection and placed on ice, bones were
immediately removed and flushed in PBS. After centrifugation,
plasma and BM supernatants were collected and used for
detection of SDF-1 protein by ELISA and uPA by zymography
(See Supplementary Information for details). Cell pellets were
used for colony-forming assay or flow cytometry as detailed in
Supplementary Information. For the primate studies, blood
samples for CBC determination and CD34þenumeration were
collected 1h after administration of a single subcutaneous dose
of 1mg/kg AMD3100.
SDF-1 secretion assays
Confluent primary human BMEC and MG-63 cells cultured in
24-well plates, washed twice in pre-warmed PBS and stimulated
by either AMD3100, NE or ICI at the indicated concentrations in
serum-free media supplemented with 100mg/ml cycloheximide
(Sigma-Aldrich) at 371C. After 3h, the supernatants were
collected and analyzed for N-terminally intact SDF-1 content
by ELISA.
Statistical analysis
Data were analyzed statistically by analysis of variance. Single
factor or P-value was calculated using two-tailed Student’s t-test,
Excel 2008. Error bars represent s.e.
Results
CXCR4-dependent release of SDF-1 from BM stromal
cells promotes egress and rapid mobilization of
hematopoietic progenitor cells
We examined whether or not functional SDF-1 (intact
N-terminal) is released to the circulation during steady state,
and evaluated its potential effects on egress of immature and
mature leukocytes. Mice were injected with neutralizing
antibodies (Ab) for SDF-1 or CXCR4. At 1h after administration,
CXCR4 neutralization by Ab markedly decreased the concen-
tration of circulating SDF-1 (Figure 1a). Interestingly, increased
levels of functional SDF-1 were found in the BM (Figure 1a),
suggesting that SDF-1 transport to the circulation is defective,
thus it is accumulated in the BM. Although SDF-1-neutralizing
Ab did not affect egress of leukocytes to the circulation
CXCR4-dependent rapid mobilization of progenitors
A Dar et al
2
Leukemia

Page 3

(Figure 1b), CXCR4-neutralizing Ab increased their numbers
(Figure 1b). In contrast, the egress of hematopoietic progenitor
cells was selectively inhibited by neutralizing monoclonal
anti-SDF-1 Ab, and more dramatically by mouse monoclonal
anti-CXCR4 Ab (Figure 1b). Rabbit polyclonal anti-CXCR4 Ab
was also capable of inhibiting steady-state egress of progenitor
cells (data not shown). Administration of rabbit isotype control
Ab did not affect the numbers of circulating cells in steady state
(data not shown) and administration of mouse isotype control
Ab increased the levels of both mature leukocytes and immature
progenitor cells in the circulation (Figure 1b), most probably due
to activation of the murine complement system in a IgG-
dependent manner as previously described.34To rule out
nonspecific effects, which could be mediated by activation of
the complement system via Ab binding to CXCR4þcells in vivo,
a complement inhibitor was given to the mice. Administration of
anti-CXCR4 Ab to mice pre-treated with a complement system
inhibitor resulted in the same reduction of progenitor cell egress
as observed with mice, which were not pre-treated (Supple-
mentary Figure S2), thus excluding indirect effects of the Ab. To
study whether there is a physiological role for SDF-1 release
from the BM to the circulation, we tested its levels after
AMD3100 treatment, as a model for rapid mobilization in
mice27and non-human primates.28A single injection of
AMD3100 resulted in a rapid increase in SDF-1 concentrations
in the circulation of mice (Figure 1c) and non-human primate
(Figure 1e).Accordingly,decreased
documented in the BM of treated mice within 1h (Figure 1c).
Elevation in circulating SDF-1 concentrations corresponded
with rapid mobilization of both mature and immature murine
(Figure 1d) and non-human primate (Figures 1f and g)
hematopoietic cells. Of note, AMD3100 from two different
sources (i.e.,Sigma andAnormedFcurrently
Cambridge, MA, USA) was compared, demonstrating similar
mobilization rates of different types of colony-forming hemato-
poietic progenitor cells (Supplementary Figures S3a and b). We
have previously demonstrated that BMEC transfer SDF-1
between the blood and the BM via their CXCR4 receptors.11
In accordance and similarly to the effect in steady-state
conditions (Figures 1a and b), co-injection of neutralizing
mouse anti-CXCR4 Ab to mice treated with AMD3100 markedly
decreased the levels of circulating SDF-1 compared with isotype
control Ab-treated mice (Figure 1c). Rabbit polyclonal anti-
CXCR4 Ab was also capable of inhibiting AMD3100-induced
progenitor cell mobilization (data not shown). Although our
murine strain model is BALB/c, we have examined also the
C57Bl/6 strain, demonstrating that a double dose of anti-CXCR4
Ab is required to inhibit AMD3100-induced mobilization of
progenitors (Supplementary Figures S4a and b) as compared
with BALB/c mice. In addition, co-administration of anti-CXCR4
SDF-1levelswere
Genzyme,
Intact SDF-1 levels
(fold change of control)
PBS control
*
*
*
*
Number of WBC
(x106/ml blood)
Number of CD34+ cells
(x105 /ml blood)
Control
*
Intact SDF-1 levels
(fold change of control)
Intact SDF-1 levels
(fold change of control)
Number of WBC
(x106/ml blood)
Number of CFU-C
(per 2x105 MNCs)
*
*
*
*
25
20
15
10
5
0
25
20
15
10
5
0
25
20
15
10
5
0
20
35
40
25
20
15
10
5
0
Baseline
0.08%
1 hour
0.14%
*
*
*
*#
*#
*
*
*
0
2
4
6
8
10
12
0
1
2
3
4
5
0
1
2
3
4
5
6
0
2
4
6
8
10
12
Number of WBC
(x106/ml blood)
Number of CFU-C
(per 2x105 MNCs)
WBC
CFU-C
WBC
CD34+
cells
WBC
CFU-C
0
0.5
1
1.5
2
2.5
3
Plasma
BM Sup.
Plasma
BM Sup.
Anti-CXCR4
Isotype control
PBS
control
Anti-CXCR4
Anti-SDF-1Isotype
control
AMD+Anti-
CXCR4
AMD+Isotype
control
AMD PBS control
AMD
+Anti-
CXCR4
AMD
+Anti-
SDF-1
AMD
+Isotype
control
AMD PBS
control
AMDControlAMD
103
102
101
100
103
102
101
100
103
CD45
102
101
100
103
102
101
100
CD34
Figure 1
progenitor cells. (a) Steady-state homeostasis fold change in the levels of N-terminally intact SDF-1 in murine plasma and BM supernatants upon
administration of PBS, isotype control Ab or anti-CXCR4 Ab. Values of plasma SDF-1 levels: 0.7±0.2, 0.8±0.3 and 0.2±0.2ng/ml, respectively.
(b) Total number of circulating WBC and the frequency of colony forming progenitors (CFU-C) in the presence of Ab against SDF-1, CXCR4 or
isotype control Ab, in steady state. (c, d) AMD3100-induced release of functional SDF-1 in murine plasma and BM sup. (c) and the frequency of
circulating murine WBC and CFU-C (d) 1h after administration. Values of plasma SDF-1 levels: 0.61±0.2, 1.19±0.34, 1.79±0.03 and
1.12±0.35ng/ml, respectively. (e–g) AMD3100-induced release of functional SDF-1 in primate plasma (e) and the frequency of circulating
primate WBC and CD34þprogenitor cells (f), 1h after administration. Representative flow-cytometry analyses at 1h after AMD3100
administration (g). *Po0.05 compared with AMD3100 treated mice (d) or to matched controls (a–c, e and f).#Po0.05 compared with AMD3100-
treated mice (c). n¼3–5 mice or non-human primates per group.
CXCR4-dependent steady state and AMD3100-induced release of SDF-1 mediates preferential egress and rapid mobilization of
CXCR4-dependent rapid mobilization of progenitors
A Dar et al
3
Leukemia

Page 4

Ab together with AMD3100 inhibited the mobilization of
Lin?Sca-1þc-Kitþcells, which are enriched for hematopoietic
stem cells (Supplementary Figure S4c). These results, taken into
account together with the accumulation of SDF-1 in the BM,
imply that blockage of CXCR4 did not interfere with SDF-1
secretion from BM stromal cells, but impaired CXCR4-depen-
dent translocation of SDF-1 from the BM to the circulation. The
inhibition of SDF-1 release was accompanied by a significant
reduction in the number of circulating immature colony forming
cells, but not mature leukocytes (Figure 1d). Additionally, SDF-
1-neutralizing Ab significantly reduced circulating progenitor
cells, compared with AMD3100 alone, with no significant
change in total leukocyte counts (Figure 1d). Noteworthy,
independent results in BALB/c mice co-treated with Anormed-
derived AMD3100 and neutralizing anti-CXCR4 Ab, revealed
similar inhibition of AMD3100-induced progenitor cell mobili-
zation, over mature leukocytes (Supplementary Figures S3c and
d). Short-term effects of AMD3100 administration on SDF-1
release and mobilization were evaluated 10min after a single
injection of AMD3100. Although the concentrations of SDF-1 in
the BM extracellular fluids were significantly increased within
10min, the levels of circulating SDF-1 were similar to those of
PBS-treated mice (Figure 2a). Mobilization of colony forming
cells could not be identified at this early time point (Figure 2b).
Altogether, these results point out that SDF-1 release from the BM
to the circulation is a key step in progenitor cell egress and rapid
mobilization from the BM. Following a second injection of
AMD3100 after an interval of additional hour, the levels of SDF-
1, the number of circulating WBCs and colony forming
progenitors were not further increased in comparison with those
measured 1h after a single injection of AMD3100 (Figures 2c and
d). By 24h after a single injection of AMD3100, the levels of
circulating SDF-1 (Figure 2c) and the number of circulating
progenitors (Figure 2d) did not exceed basal levels compared with
control mice. These observations indicate that a single AMD3100
stimulation can induce rapid and substantial release of SDF-1
from the BM to the circulation, which correlates with rapid
recruitment of BM-residing progenitor cells to the circulation.
AMD3100 directly induces SDF-1 secretion from
CXCR4þBM stromal cells
To further demonstrate the ability of the BM to release SDF-1 to
the circulation and the associated recruitment of CXCR4þ
0
10
20
30
0
10
20
30
3.5
*
*
*
*
*
Intact SDF-1 levels
(fold change of control)
Intact SDF-1 levels
(fold change of control)
PBS
Number of WBC
(x106 /ml blood)
Number of WBC
(x106 /ml blood)
Number of CFU-C
(per 2x105 MNCs)
Number of CFU-C
(per 2x105 MNCs)
Plasma
BM Sup.
WBC
CFU-C
Plasma
BM Sup.
0
0.5
1
1.5
2
2.5
3
*
0
0.5
1
1.5
2
2.5
3
AMD 10 min
PBSAMD 10 min
0
3
6
9
12
15
18
21
0
3
6
9
12
15
18
21
AMD
24 hour
AMD
2x1 hour
AMD
1 hour
PBS
AMD
24 hour
AMD
2x1 hour
AMD
1 hour
PBS
WBC
CFU-C
Figure 2
plasma and BM sup. (a) and the frequency of circulating murine WBC and CFU-C (b), 10min after administration. Values of plasma SDF-1 levels:
1.2±0.17 and 1.3±0.18ng/ml, respectively. (c, d) Fold changes in SDF-1 levels in murine plasma and BM sup. (c) and the frequency of circulating
WBC and CFU-C (d) were determined 1 and 24h after AMD3100 administration or following two injections of AMD3100 every hour (AMD
2?1h). Values of plasma SDF-1 levels: 0.76±0.03, 2.1±0.3, 1.2±0.07 and 0.85±0.08ng/ml, respectively. *Po0.05 compared with matched
controls. n¼4–5 mice per group.
Kinetics of AMD3100-induced functional SDF-1 release and rapid mobilization. (a, b) AMD3100-induced release of SDF-1 in murine
Figure 3
SDF-1 (bSDF-1 3/6) and were either treated with AMD3100 or PBS (s.c). Concentrations of intact intra-femur injected bSDF-1 in BM sup of injected
femurs and in non-injected femurs (a) and in the plasma (b) at the indicated time points.@Values of the injected bone are presented as divided by
10, therefore real values are 10-fold higher. Matched control samples from PBS-injected mice were negative for the presence of bSDF-1. (c) Total
numbers of circulating WBC and the frequency of CFU-C in the PB and BM. n¼4–6 mice per group. *Po0.05 compared with control mice;
$Po0.05 compared with 30min after bSDF-1 injection;#Po0.05 compared with 60min after bSDF-1 injection (without AMD3100 treatment).
(d) Immunofluorescent labeling of the BM of control or AMD3100-treated mice. Immunoreactivity of SDF-1 (red) is observed in PBS-treated murine
BM: upper panelsFosteopontin (green) positive osteoblasts in the bone shaft; lower panelsFvon Willebrand’s factor (vW, green) positive blood
vessels (BV). Blood vessels are marked with arrows or a dashed border. SDF-1 immunoreactivity is not observed in AMD3100-treated murine BM.
Nuclear staining by DAPI (blue). Original magnifications: upper panelsF?40; lower panelsF?200 (PBS), ?630 (AMD3100). (e) Untreated or
AMD3100 (100ng/ml) treated primary human BMEC were immunolabeled for functional SDF-1 (red) and stained for nuclear DNA (purple).
(f) Secretion of SDF-1 from cultured BMEC or the osteoblastic human cell line MG-63 in response to AMD3100, measured by ELISA. n¼4,
*Po0.05 compared with control.
AMD3100 agonistically affects BM stromal cells to release functional SDF-1. (a–c) Mice were injected intra-bone with biotinylated
CXCR4-dependent rapid mobilization of progenitors
A Dar et al
4
Leukemia

Page 5

progenitor cells, biotin-labeled SDF-1 (bSDF-1) was injected
into the murine femur with or without administration of
AMD3100. To exclude a potential passive release of SDF-1-
binding heparan sulfate16to the circulation, we used the bSDF-1
analog, bSDF-1 3/6, which cannot bind heparan sulfate but can
still bind and activate CXCR4.35The concentrations of the active
N-terminally intact form of bSDF-1 3/6 were measured in the
plasma and in the extracellular fluids of the injected femur, as
well as in non-injected bones, 10, 30 and 60min after intra-
femur injection. bSDF-1 3/6 was detected within the BM fluids
control
AMD3100
Osteopontin marks osteoblasts; vW Factor marks endothelilal cells
M = megakaryocye; BV = blood vessel; vW Factor = von Willebrand’s Factor
BM CFU-C
SDF-1
DAPI
injected bone@
non-injected bone
primary human BMEC
MG63
*
*
*
*
*
*
*
*#
*
*#
*
*#
*#
$
#
0
1
2
3
4
0
10
20
30
40
50
0
10
20
30
40
50
ng/ml AMD3100
PBS
control
Intact bSDF-1 3/6 levels
(ng/mg BM proteins)
Intact SDF-1 levels
(ng/ml CM)
Number of circulating WBC
(x106 /ml blood)
Number of CFU-C
(per 2x105 PB MNCs
or 1.5x104 BM cells)
Intact bSDF-1 3/6 levels
(ng/mg BM proteins)
0
0.01
0.02
0.03
0.04
0.05
60min
+AMD
60min30min
10min
60min
+AMD
60min30min
10min
bSDF-1 + AMD
60min
bSDF-1
60min
bSDF-1
30min
bSDF-1
10min
Control
AMD3100
Control
AMD3100
0.1
0
0.2
0.3
0.4
0.5
100 100
WBC
PB CFU-C
CXCR4-dependent rapid mobilization of progenitors
A Dar et al
5
Leukemia

Page 6

of non-injected bones, as early as 10min post administration
(Figure 3a). In addition, the levels of bSDF-1 3/6 in the BM fluids
of the injected femur decreased with time and were significantly
lower 60min post administration (Figure 3a). Co-injection of
AMD3100 (s.c) with bSDF-1 3/6 (intra-bone) resulted in a
significant reduction in bSDF-1 3/6 levels in the injected bone,
as well as in non-injected bones 60min after administration
(Figure 3a). In accordance, the concentrations of circulating
bSDF-1 3/6 gradually increased up to 60min following intra-
femoral injection, further enhanced by AMD3100 co-adminis-
tration (s.c) (Figure 3b). In parallel, we observed a significant
increase in the frequency of hematopoietic progenitors in the
circulation and their decrease in the BM, 60min after intra-bone
injection of bSDF-1 3/6, further enhanced by AMD3100 co-
administration (s.c) (Figure 3c). The changes in progenitor cell
levels correlated with the elevation of circulating bSDF-1 3/6
and reduction of bSDF-1 3/6 concentrations in the BM fluids,
suggesting that SDF-1 actively mediates progenitor trafficking.
Thus, SDF-1 release from the BM to the plasma, which is
enhanced by AMD3100 treatment, could trigger egress and
recruitment of hematopoietic progenitors from the BM reservoir
to the circulation. BM stromal cells, such as endosteal bone
lining osteoblasts, endothelial and reticular cells, functionally
express CXCR4 and are the main source of SDF-1 in human10
and murine BM.11,36We therefore analyzed the effect of
AMD3100 (s.c) injection on the expression levels of SDF-1 in
BM sections of treated mice. In steady state, high expression of
SDF-1 was detected in endosteal osteopontin-positive osteo-
blasts of the bone shaft (diaphysis) and in von Willebrand’s-
positive endothelial cells of large blood vessels (Figure 3d).
SDF-1 expression was notably reduced in both compartments,
1h after a single injection of AMD3100 (Figure 3d). Likewise,
only weak immunoreactivity of intracellular SDF-1 was detected
in cultured primary human BMEC, which also highly express
CXCR4,11treated with 100ng/ml of AMD3100 as compared
with untreated cells, which express high levels of intracellular
SDF-1 (Figure 3e). Incubation of CXCR4þ
CXCR4þhuman osteoblastic cell line MG63 with AMD3100
resulted in increased functional SDF-1 release to the condi-
tioned medium (Figure 3f). These results are compatible with
release of SDF-1 from internal storage compartments of BM
stromal cells in response to AMD3100. In addition, circulating
mononuclear cells released negligible amounts of SDF-1 in
comparison with BM stromal cells, and did not respond to
AMD3100 stimulation (Supplementary Figure S5). Interestingly,
BMEC, or the
AMD3100 at concentrations higher than 500ng/ml failed to
induce SDF-1 secretion from primary human BMEC or MG63
(Supplementary Figure S6). These results demonstrate that SDF-1
is actively secreted from stromal cells rather than released from
the cell surface due to competitive binding of AMD3100.
Collectively, these observations reveal complex activities of this
compound in vivo and suggest that the rapid AMD3100-induced
mobilization of hematopoietic progenitor cells involve diverse
CXCR4 interactions, SDF-1 secretion and downstream events
within the BM compartment affecting CXCR4-expressing hema-
topoietic and stromal cells.
ROS signaling is involved in AMD3100-induced
mobilization of hematopoietic progenitor cells
Increased ROS levels are associated with HGF and G-CSF-
induced mobilization of hematopoietic progenitor cells.23
Concomitant inhibition of ROS by NAC resulted in a significant
reduction in the number of G-CSF-mobilized hematopoietic
progenitors.23We sought to find out whether ROS signaling also
has a role in rapid mobilization to the circulation induced by
AMD3100. ROS inhibition by NAC led to a preferential
inhibition of AMD3100-induced progenitor mobilization, over
mature WBC (Figure 4a). In addition, NAC inhibited AMD3100-
induced SDF-1 release to the circulation, correlating with
inhibition of mobilization (Figure 4b). Of interest, NAC treatment
in vivo does not affect the levels of circulating WBC and progenitors
in steady state,23suggesting generation of ROS in response to
AMD3100. Altogether, the involvement of ROS suggests an
activation of signaling pathways by AMD3100 in a SDF-1-dependent
manner, and further supports the notion of rapid mobilization as an
active process rather than a passive inhibitory one.
Adrenergic signals induce rapid SDF-1 release and
recruitment of progenitor cells
We tested whether NE, previously implicated in G-CSF-induced
mobilization,19,20can induce rapid mobilization by a mechan-
ism similar to AMD3100. We found that a single injection of NE
caused a significant increase in circulating SDF-1 within 1h
(Figure 5a). Concomitantly, we observed an increase in the
frequency of circulating progenitors, with no change in the
number of circulating mature leukocytes (Figure 5b). We tested
whether inhibition of adrenergic signaling could induce the
opposite effect, under steady-state conditions. Injection of ICI, a
WBC
CFU-C
*
*
* #
#
PBSAMDAMD + NAC
PBSAMD AMD + NAC
Intact SDF-1 levels
(fold change of control)
Number of CFU-C
(per 2x105 MNCs)
Number of WBC
(x106 /ml blood)
Figure 4
as described in Figure 1. In addition, mice were treated with the ROS inhibitor NAC or PBS as control. (a) At 1h post injection of AMD3100,
circulating WBC (white bars) and CFU-C (black bars) were measured. (b) Functional SDF-1 levels in the plasma. n¼4–5 mice per group. Values of
plasma SDF-1 levels: 0.71±0.06, 1.72±0.2 and 0.86±0.12ng/ml, respectively. *Po0.05 compared with PBS alone and#Po0.05 compared with
AMD3100 treatment.
ROS inhibition reduces AMD3100-induced mobilization and SDF-1 release. (a, b) Mice were treated with AMD3100 or PBS as control
CXCR4-dependent rapid mobilization of progenitors
A Dar et al
6
Leukemia

Page 7

b2-adrenergic receptor inhibitor, reduced the concentrations of
SDF-1 (Figure 5a), as well as the frequency of progenitors in the
circulation (Figure 5b). Neither NE nor ICI altered the numbers
of circulating leukocytes when administered alone or when
injected with AMD3100 (Figure 5b). In addition, the concentra-
tions of SDF-1 were elevated in the BM of ICI injected mice in
comparison with those measured in the BM of control mice,
coinciding with decreased levels of circulating SDF-1 and
progenitor cells (Figures 5a and b). To evaluate a potential role
for adrenergic receptors in direct regulation of SDF-1 secretion,
the cell surface expression of these receptors on human BMEC
was evaluated. Flow cytometry and reverse-transcription PCR
analyses revealed that cultured BMEC expressed b2-adrenergic
receptors (Figure 5c). Moreover, these cells responded to ICI by
reducing SDF-1 secretion in vitro (Figure 5d), implying that
b2-adrenergic signaling is involved in SDF-1 secretion. As
CXCR4 is also expressed by neurons,37,38it is possible that b2-
adrenergic signaling may exert an effect downstream to CXCR4.
Alternatively, these two pathways may exert an effect indepen-
dently in the regulation of SDF-1 secretion. To distinguish
between these possibilities, we isolated catecholaminergic
neurons from superior cervical ganglia (SCG), which express
CXCR4 by PCR (data not shown), and tested whether AMD3100
or SDF-1 might alter the uptake or the release of radiolabeled NE
in SCG culture explants. As shown in Supplementary Figure S7a,
although the re-uptake inhibitor desipramine markedly reduced
the uptake of radiolabeld NE, pre-treatment of the explants with
AMD3100, with or without SDF-1, did not affect NE uptake by
SCG explants. In addition, NE release from SCG explants was
not enhanced by AMD3100 or SDF-1 (Supplementary Figure
S7b). However, KCl-induced NE release was significantly
enhanced in the presence of SDF-1 (Supplementary Figure
S7b), implying a positive feedback loop between SDF-1 and NE
release during progenitor cell mobilization. Given the similar
effects of AMD3100 and NE administration, we examined
whether these signaling pathways cooperate with each other.
Chemical sympathectomy by treatment of neonate mice with
6-hydroxybutamine (6OHDA) did not affect the number of
progenitors in the BM (data not shown), but it noticeably
reduced the number of progenitors elicited in the circulation
(Figure 5e). However, 6OHDA treatment of adult mice, which
does not cross the blood–brain barrier, did not affect AMD3100-
induced mobilization (Figure 5e), suggesting that this effect is
mediated by the central nervous system.
AMD3100 induces protease activation and progenitor
cell mobilization in a CXCR4/JNK signaling-dependent
manner
The involvement of the plasminogenic system in G-CSF-induced
mobilization was recently reported.21,22SDF-1, CXCR4 and
c-jun N-terminal kinase (JNK) signaling were shown to be
involved in activation of the serine protease urokinase plasmi-
nogen activator (uPA) in a breast cancer invasion model.39
Considering that AMD3100 induces SDF-1 secretion, we
evaluated uPA activity levels in AMD3100-treated mice.
Circulating uPA activity was upregulated 1h post AMD3100
administration (Figures 6a and c). The AMD3100 amplification
effect on uPA activity is CXCR4-dependent as injection of
neutralizing CXCR4 Ab together with AMD3100 inhibited the
elevation effect (Figure 6a). This implies that BM-released SDF-1
mediates uPA secretion to promote hematopoietic progenitor
cell mobilization in a CXCR4-dependent manner. Comparable
to CXCR4 neutralization, injection of a specific JNK inhibitor
before AMD3100 administration reduced the activation of uPA
(Figure 6c), as well as mobilization of WBC and progenitor cells
(Figure 6b). Inhibition of JNK signaling in steady-state conditions
did not affect the constitutive egress of mature WBC and
immature progenitors (Figure 6b) nor did it affect the constitutive
release of SDF-1 to the circulation (Figure 6d). This suggests that
40
40
30
30
20
20
10
10
0
Control
0
7
6
Control
0
1
2
3
4
5
8
60
60
50
50
NE
NE
ICI
ICI
AMD
AMD
NE+AMD
NE+AMD
ICI+AMDICI+AMD
β2 adrenergic receptor
-RT +RT
Control
0.0
0.1
0.2
0.3
0.5
0.4
ICI
Counts
ADRB2
*
*
*
*
#
#
#
#
*
Plasma
BM sup
CFU-C
WBC
*
*
*
*
#
#
Intact SDF-1 levels
(ng/ml CM)
Number of WBC
(x106/ml blood)
Number of CFU-C
(per 2X105 MNCs)
Intact SDF-1 levels
(fold change of control)
Number of CFU-C
(per ml blood)
IgG
control
β2 adrenergic
receptor
*
*
Neonate mice
AMD
0
800
600
400
200
AMD
+6OHDA
AMD AMD
+6OHDA
Adult mice
Figure 5
BM sup. (a) and circulating WBC and progenitor cells (b) in mice treated with NE or the b2 adrenergic antagonist ICI, 1h after administration.
Control mice received injections of PBS. n¼6 mice/group. Values of plasma SDF-1 levels: 1.1±0.17, 1.8±0.5, 0.6±0.06, 2±0.3, 2.8±0.5 and
1.2±0.1ng/ml, respectively. *Po0.05 compared with control mice.#Po0.05 compared with AMD3100-treated mice. (c) RT-PCR analysis (top)
for mRNA expression and flow-cytometry analysis (bottom) for cell surface expression of b2-adrenergic receptor on cultured primary human
BMEC. –RT¼cDNA was prepared without reverse transcriptase as a control. (d) SDF-1 release from primary human BMEC in response to
stimulation with ICI (10ng/ml). n¼3 (e) AMD3100-induced mobilization of progenitors in control and sympathectomised (6OHDA) of either
neonate or adult mice. *Po0.005. n¼10–15 mice per group.
Neurotransmitter stimulation induces functional SDF-1 release and rapid progenitor mobilization. (a, b) SDF-1 levels in the plasma and
CXCR4-dependent rapid mobilization of progenitors
A Dar et al
7
Leukemia

Page 8

JNK functions as a ‘stress kinase’ that is recruited upon
AMD3100 administration to facilitate cell mobilization. In
addition, mice that were pre-treated with JNK inhibitor did not
mobilize progenitors in response to AMD3100 injection, and
also exhibited a mild but significant reduction in SDF-1 release
to the circulation (Figure 6d). These results suggest that JNK
signaling not only regulates protease activation, but is also
involved in SDF-1 release to the circulation. Furthermore, a
significant reduction in AMD3100-induced uPA activation in
the blood circulation was observed following ROS inhibition by
NAC (Figure 6e). Taken together, these data imply that in order
to promote progenitor cell mobilization, AMD3100 activates a
signaling pathway, involving ROS, to induce SDF-1 secretion,
which in turn mediates uPA activation in a CXCR4/JNK-
dependent manner.
Discussion
It is currently believed that disruption of the interactions
between the CXCR4 receptor expressed by hematopoietic
progenitors and SDF-1 expressed by BM stromal cells is
sufficient to detach anchored progenitors from their BM niches,
leading to their rapid mobilization to the peripheral blood.
Nevertheless, although neutralizing anti-CXCR4 or anti-SDF-1
antibodies inhibit G-CSF-induced mobilization,7,23steady-state
progenitor cell egress and AMD3100-induced mobilization
(current study), these antibodies are incapable of inducing
progenitor cell mobilization by themselves. Importantly, several
reports demonstrate that mobilization of hematopoietic and
pro-angiogenic progenitor cells is triggered and maintained by
rapid or sustained elevation in the concentrations of circulating
SDF-1, or by repeated intra-peritoneum injections of this
ligand.13,14,16,40,41On the basis of our findings, we introduce
a novel agonistic effect of AMD3100, resulting in enhanced
SDF-1 release from BM stromal cells to the circulation that
together with its inhibitory effect in vivo (which we think is
transient and short-lived) induces rapid mobilization of hema-
topoietic progenitor cells. CXCR4-dependent SDF-1 transporta-
tion to the circulation actively promotes preferential egress of
hematopoietic progenitors from the BM during steady-state
homeostasis and in the course of rapid mobilization with
AMD3100 or catecholamines. However, we cannot rule out
involvement of other cell types in this rapid SDF-1 release, such
as SDF-1 secretion by activated platelets.42Several reports have
demonstrated a correlation between elevated SDF-1 levels in the
circulation or reduced SDF-1 levels in the BM and the
AMD3100-induced mobilization rate of different populations,
including hemangiogenic cells,43monocytes44and hemato-
poietic progenitors.45Of interest, monocytes, sequestered in the
BM following CCR2 antagonism,44and hematopoietic progeni-
tors, which did not egress to the circulation in wounded diabetic
mice,45were capable of being mobilized in response to the
elevated SDF-1 levels in the circulation following AMD3100
treatment. In addition, repetitive administration of AMD3100
did not enhance the mobilization of hemangiogenic cells43or
hematopoietic progenitors in another study.46These findings
support our results and previous observations that a higher dose
of AMD3100 results in a slightly weaker capacity to mobilize
(Broxmeyer et al.27and Supplementary Figure S2), and that an
Vehicle
control
AMD
+vehicle
AMD
+JNK inhibitor
uPA activity
uPA activity
uPA activity
WBC
CFU-C
*
#
AMD
+vehicle
AMD
+vehicle
AMD
+JNK
inhibitor
AMD
+JNK
inhibitor
*
*
#
#
Vehicle
control
JNK
inhibitor
Vehicle
control
JNK
inhibitor
Number of CFU-C
(per 2x105 MNCs)
Intact SDF-1 levels
(fold change of control)
AMD
+Isotype
control
AMD
+Anti-CXCR4
Number of WBC
(x106/ml blood)
0
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0
5
5
10
10
15
15
20
20
25
25
30
30
AMD
PBS
control
AMD
+NAC
Isotype
control
Figure 6
of uPA activity in the plasma of mice treated with either PBS or AMD3100 with or without neutralizing Ab against CXCR4 and matched isotype
control, 1h post administration. (b) The frequency of circulating WBC and CFU-C 1h after administration of vehicle or AMD3100 in the presence
or absence of JNK inhibitor. *Po0.05 compared with control mice.
(c) Activity of uPA in the plasma after AMD3100 administration in the presence or absence of JNK inhibitor. (d) Fold change in the levels of
circulating functional SDF-1 1h after vehicle or AMD3100 administration in the presence or absence of JNK inhibitor. n¼6 mice/group. Values of
plasma SDF-1 levels: 0.38±0.02, 0.34±0.09, 2±0.04 and 1.6±0.3ng/ml, respectively. *Po0.05. (e) Activity of uPA in the plasma after
AMD3100 administration in the presence or absence of ROS inhibitor NAC.
Rapid mobilization induced by AMD3100 involves JNK and uPA release in a CXCR4-dependent manner. (a) Representative zymography
#Po0.05 compared with AMD3100-treated mice. n¼6 mice/group.
CXCR4-dependent rapid mobilization of progenitors
A Dar et al
8
Leukemia

Page 9

additional injection of AMD3100 within 2h failed to induce
another wave of mobilization along with its inability to further
elevate SDF-1 levels. Interestingly, conditional deletion or
inhibition ofsphingosine-1-phosphate
resulted in impaired AMD3100-induced mobilization of im-
mature B cells47and hematopoietic progenitors.48,49Further-
more, sphingosine-1-phosphate (S1P) treatment was found to
modulate SDF-1 secretion.49Of interest, the plasma levels of
S1P, which serves as a potent chemoattractant for hematopoietic
progenitors as well, are increased following AMD3100 admin-
istration.50,51Thus, S1P, SDF-1 and probably additional factors
may cooperate in the AMD3100-induced preferential mobiliza-
tion of progenitor cells. In another study, murine hematopoietic
progenitor cells that lack fibroblast growth factor receptor 1
(FGFR1) showed impaired SDF-1-induced migration in vitro and
AMD3100-induced mobilization in vivo.52Altogether, the
defects of these different models in responding to AMD3100
suggest an active rather than a passive inhibitory process of
AMD3100-induced rapid mobilization. In this study, we found
that ROS inhibition by NAC treatment preferentially reduced the
numbers of AMD3100-mobilized progenitor cells, correlating
with inhibition of SDF-1 release and activation of uPA in the
blood circulation. As ROS signaling is associated with G-CSF-
induced mobilization and motility and does not affect
steady-sate egress,23we may further assume that AMD3100
administration induces active signaling. Regarding cell motility,
it should be noted that the use of AMD3100 in combination with
G-CSF in vivo was shown to increase the responsiveness of
mobilized human CD34þprogenitors to SDF-1 chemotaxis
in vitro.27These findings suggest that AMD3100 engages
additional players in vivo and therefore its effects include more
than only a direct antagonism of CXCR4, as previously shown
in vitro. This hypothesis surprisingly gains support from another
field of research, showing different effects of AMD3100 in vitro
and in vivo. AMD3100 is able to block the CXCR4 receptor and
to inhibit HIV replication in vitro at nanomolar concentra-
tions.53,54However, it was unable to decrease HIV viral load in
HIV-infected individuals.55One possible difference is the
involvement of BM stromal cells that also functionally express
CXCR4.9In support of our results, it was recently demonstrated
that a layer of BMEC reduces the inhibitory effect of AMD3100
on CXCR4-dependent migration by 10–100 fold, using in vitro
trans-endothelial migration assay, wherein CXCR4þCXCR7þ
tumor cells were allowed to migrate across BMEC towards
SDF-1.56We have now shown that blockage of SDF-1 and
CXCR4 signaling preferentially inhibits steady-state egress and
AMD3100-induced mobilization of progenitors, but not mature
leukocytes. We found that administration of NE is accompanied
by increased levels of circulating SDF-1 and rapid mobilization,
preferentially also of progenitor cells. In agreement with our
results, it has recently been shown that circadian modulations in
the number of circulating progenitors are mediated by rhythmic
synthesis of SDF-1 in the BM, which is regulated by b3-
adrenergic receptors.18In addition, NE regulates G-CSF-induced
mobilization of progenitor cells, affecting both hematopoietic
progenitor motility via adrenergic receptors20and production of
SDF-1 by osteoblasts in the BM.19These findings together with
our results suggest that the rich and tight association of efferent
nerve terminals with BM stromal cells57directly impacts on
hematopoietic progenitor cell retention and rapid mobilization
via modulation of SDF-1 expression and release from BM
stromal cells. G-CSF-induced mobilization of human and
murine progenitor cells also involves SDF-1 secretion and
CXCR4 signaling.7,58,59
Neutralization of CXCR4 strongly
reduces G-CSF-induced mobilization of both human and murine
receptor-1(S1P1)
progenitor cells.7
hematopoietic cells in CXCR4þBM chimeras do not mobilize
in response to G-CSF.60Furthermore, rapid mobilization of
murine progenitor cells using fucoidan, which releases SDF-1
from heparan sulfate complexes, was selectively inhibited with
co-administration of neutralizing Ab against SDF-1.16Likewise,
mobilization by a daily injection of SDF-1, HGF or RANKL for
3–5 days was shown to preferentially recruit murine progenitor
cells over mature leukocytes in a SDF-1/CXCR4-dependent
manner.41Of note, activation of the complement cascade is also
needed for optimal AMD3100-induced mobilization.61Activa-
tion of MMP-9 in response to G-CSF-induced mobilization is
well established,62,63and was also documented following
AMD3100 administration61,64(and our unpublished data),
demonstrating that MMP-9 is also involved in the process of
rapid mobilization. Furthermore, the plasminogen system was
recently found to participate in murine and human mobiliza-
tion.21,22Systemic administration of plasminogen activator
generates circulating plasmin from plasminogen, which in turn
activates MMP-9 in the BM.65Altogether, these and our
results imply that the enhanced activation and release of
MMP-9, in the course of AMD3100-induced mobilization, can
be related to the increased release of SDF-1 and uPA to the
circulation. In another model, interactions between CXCR4-
expressing human breast carcinoma cells and SDF-1-presenting
stromal cells were shown to enhance in vitro malignant cell
invasion via upregulation of both uPA and uPAR expression,
mediated by sustained activation of JNK.39We reveal that
CXCR4-dependentJNKsignaling,
AMD3100-induced SDF-1 secretion from BM stromal cells,
results in activation of uPA,66which participates in the release
of SDF-1 to the circulation and is crucial for progenitor cell
mobilization. Of note, JNK also seems to be involved directly in
leukocyte motility, as its inhibition strongly reduces AMD3100-
induced mobilization. Importantly, neutralizing CXCR4 Ab, as
well as ROS inhibition inhibited AMD3100-induced SDF-1
release and uPA activation. We have previously reported that
BM endothelial and other stromal cells actively participate in
regulation of SDF-1 availability due to their surface expression
of CXCR4,11and thus control the homing of human progenitors
across the murine blood–bone marrow barrier. In this study, we
revealed that inhibition of progenitor cell egress and mobiliza-
tion can occur because of impairment of CXCR4-dependent
translocation of BM SDF-1 across this barrier to the circulation.
In summary, we introduce data to support a new concept that
CXCR4, which is functionally expressed on BM stromal cells, as
well as on hematopoietic progenitor cells, actively participates
in preferential regulation of steady-state egress and rapid
mobilization of progenitors, via modulation of SDF-1 secretion
and release and protease activation. Enhanced CXCR4-depen-
dent release of SDF-1 from BM stromal cells can potentially
induce the replacement of former SDF-1/CXCR4 anchoring
interactions with directional migratory signals, leading to rapid
mobilization of progenitors to the circulation. Our results further
underline a direct and active role of CXCR4 signaling, SDF-1
release, proteolytic enzymes and signals from the nervous
system in regulation of steady-state egress and rapid mobiliza-
tion of hematopoietic progenitor cells as part of host defense and
repair mechanisms with clinical relevance.
In support of our findings, CXCR4?/?
initiated byin vivo
Conflict of interest
The authors declare no conflict of interest.
CXCR4-dependent rapid mobilization of progenitors
A Dar et al
9
Leukemia

Page 10

Acknowledgements
This study was partially supported by the Helen and Martin
Kimmel Institute for Stem Cell Research at the Weizmann
Institute, Israeli Science Foundation grant 544/09, the European
Union (Advance Cell-based Therapies for the Treatment of
Primary Immunodeficiency HEALTH-F5-2010-261387) and the
Legacy Heritage Fund (TL). PSF is an established investigator of the
American Heart Association supported by the National Institutes
of Health. This work was also partially supported by Grant-in-Aid
for Scientific Research from the Ministry of Education, Culture,
Sports, Science, and Technology of Japan (SO and NF). TL holds
The Edith Arnoff Stein Professorial Chair in Stem Cell Research.
We would like to thank Dr Abraham Avigdor for supplying us with
human BM aspirations from healthy donors and to Dr Scott
Cooperfor performingexperiments
AMD3100. Our special thanks to Drs Sara Rankin, Isabelle Petit,
Shoham Shivtiel and Jonathan Canaani for fruitful discussions and
for critically reviewing the manuscript.
with Anormed-derived
References
1 Spiegel A, Kalinkovich A, Shivtiel S, Kollet O, Lapidot T. Stem
cell regulation via dynamic interactions of the nervous and
immune systems with the microenvironment. Cell Stem Cell
2008; 3: 484–492.
2 Lapidot T, Kollet O. The Brain-Bone-Blood Triad: Traffic Lights for
Stem-Cell Homing and Mobilization. Hematology Am Soc
Hematol Educ Program 2010; 1–6.
3 Wright DE, Wagers AJ, Gulati AP, Johnson FL, Weissman IL.
Physiological migration of hematopoietic stem and progenitor
cells. Science 2001; 294: 1933–1936.
4 Kollet O, Dar A, Lapidot T. The multiple roles of osteoclasts in host
defense: bone remodeling and hematopoietic stem cell mobiliza-
tion. Annu Rev Immunol 2007; 25: 51–69.
5 Lapid K, Vagima Y, Kollet O, Lapidot T. Egress and mobilization of
hematopoietic progenitor cells. StemBook, Harvard Stem Cell
Institute 2009; 1.39.1.
6 Lapidot T, Petit I. Current understanding of stem cell mobilization:
the roles of chemokines, proteolytic enzymes, adhesion molecules,
cytokines, and stromal cells. Exp Hematol 2002; 30: 973–981.
7 Petit I, Szyper-Kravitz M, Nagler A, Lahav M, Peled A, Habler L
et al. G-CSF induces stem cell mobilization by decreasing bone
marrow SDF-1 and up-regulating CXCR4. Nat Immunol 2002; 3:
687–694.
8 Papayannopoulou T, Scadden DT. Stem-cell ecology and stem
cells in motion. Blood 2008; 111: 3923–3930.
9 Dar A, Kollet O, Lapidot T. Mutual, reciprocal SDF-1/CXCR4
interactions between hematopoietic and bone marrow stromal
cells regulate human stem cell migration and development in
NOD/SCID chimeric mice. Exp Hematol 2006; 34: 967–975.
10 Ponomaryov T, Peled A, Petit I, Taichman RS, Habler L, Sandbank
J et al. Induction of the chemokine stromal-derived factor-1
following DNA damage improves human stem cell function. J Clin
Invest 2000; 106: 1331–1339.
11 Dar A, Goichberg P, Shinder V, Kalinkovich A, Kollet O, Netzer N
et al. Chemokine receptor CXCR4-dependent internalization and
resecretion of functional chemokine SDF-1 by bone marrow
endothelial and stromal cells. Nat Immunol 2005; 6: 1038–1046.
12 Sugiyama T, Kohara H, Noda M, Nagasawa T. Maintenance of the
hematopoietic stem cell pool by CXCL12-CXCR4 chemokine
signaling in bone marrow stromal cell niches. Immunity 2006; 25:
977–988.
13 Hattori K, Heissig B, Tashiro K, Honjo T, Tateno M, Shieh JH et al.
Plasmaelevation of stromal
mobilization of mature and immature hematopoietic progenitor
and stem cells. Blood 2001; 97: 3354–3360.
14 Shen H, Cheng T, Olszak I, Garcia-Zepeda E, Lu Z, Herrmann S
et al. CXCR-4 desensitization is associated with tissue localization
ofhemopoieticprogenitor
5027–5033.
cell-derivedfactor-1 induces
cells.J Immunol 2001; 166:
15 Frenette PS, Weiss L. Sulfated glycans induce rapid hematopoietic
progenitor cell mobilization: evidence for selectin-dependent and
independent mechanisms. Blood 2000; 96: 2460–2468.
16 Sweeney EA, Lortat-Jacob H, Priestley GV, Nakamoto B, Papayan-
nopoulou T. Sulfated polysaccharides increase plasma levels of
SDF-1 in monkeys and mice: involvement in mobilization of stem/
progenitor cells. Blood 2002; 99: 44–51.
17 Tchernychev B, Ren Y, Sachdev P, Janz JM, Haggis L, O’Shea A
et al. Discovery of a CXCR4 agonist pepducin that mobilizes bone
marrow hematopoietic cells. Proc Natl Acad Sci USA 2010; 107:
22255–22259.
18 Mendez-Ferrer S, Lucas D, Battista M, Frenette PS. Haematopoietic
stem cell release is regulated by circadian oscillations. Nature
2008; 452: 442–447.
19 Katayama Y, Battista M, Kao WM, Hidalgo A, Peired AJ, Thomas
SA et al. Signals from the sympathetic nervous system regulate
hematopoietic stem cell egress from bone marrow. Cell 2006; 124:
407–421.
20 Spiegel A, Shivtiel S, Kalinkovich A, Ludin A, Netzer N,
Goichberg P et al. Catecholaminergic neurotransmitters regulate
migration and repopulation of immature human CD34+ cells
through Wnt signaling. Nat Immunol 2007; 8: 1123–1131.
21 Selleri C, Montuori N, Ricci P, Visconte V, Carriero MV, Sidenius
N et al. Involvement of the urokinase-type plasminogen activator
receptor in hematopoietic stem cell mobilization. Blood 2005;
105: 2198–2205.
22 Tjwa M, Janssens S, Carmeliet P. Plasmin therapy enhances
mobilization of HPCs after G-CSF. Blood 2008; 112: 4048–4050.
23 Tesio M, Golan K, Corso S, Giordano S, Schajnovitz A, Vagima Y
et al. Enhanced c-Met activity promotes G-CSF induced mobiliza-
tion of hematopoietic progenitor cells via ROS signaling. Blood
2011; 117: 419–428.
24 Abraham M, Beider K, Wald H, Weiss ID, Zipori D, Galun E et al.
The CXCR4 antagonist 4F-benzoyl-TN14003 stimulates the
recovery of the bone marrow after transplantation. Leukemia
2009; 23: 1378–1388.
25 Hatse S, Princen K, Bridger G, De Clercq E, Schols D. Chemokine
receptor inhibition by AMD3100 is strictly confined to CXCR4.
FEBS Lett 2002; 527: 255–262.
26 Devine SM, Vij R, Rettig M, Todt L, McGlauchlen K, Fisher N et al.
Rapid mobilization of functional donor hematopoietic cells
without G-CSF using AMD3100, an antagonist of the CXCR4/
SDF-1 interaction. Blood 2008; 112: 990–998.
27 Broxmeyer HE, Orschell CM, Clapp DW, Hangoc G, Cooper S,
Plett PA et al. Rapid mobilization of murine and human
hematopoietic stem and progenitor cells with AMD3100, a CXCR4
antagonist. J Exp Med 2005; 201: 1307–1318.
28 Larochelle A, Krouse A, Metzger M, Orlic D, Donahue RE, Fricker
S et al. AMD3100 mobilizes hematopoietic stem cells with long-
term repopulating capacity in nonhuman primates. Blood 2006;
107: 3772–3778.
29 Liles WC, Broxmeyer HE, Rodger E, Wood B, Hubel K, Cooper S
et al. Mobilization of hematopoietic progenitor cells in healthy
volunteers by AMD3100, a CXCR4 antagonist. Blood 2003; 102:
2728–2730.
30 Hess DA, Bonde J, Craft TP, Wirthlin L, Hohm S, Lahey R et al.
Human progenitor cells rapidly mobilized by AMD3100 repopu-
late NOD/SCID mice with increased frequency in comparison to
cells from the same donor mobilized by granulocyte colony
stimulating factor. Biol Blood Marrow Transplant 2007; 13: 398–411.
31 Gazitt Y, Freytes CO, Akay C, Badel K, Calandra G. Improved
mobilization of peripheral blood CD34+ cells and dendritic cells
by AMD3100 plus granulocyte-colony-stimulating factor in non-
Hodgkin’s lymphoma patients. Stem Cells Dev 2007; 16: 657–666.
32 Cashen A, Lopez S, Gao F, Calandra G, MacFarland R, Badel K
et al. A phase II study of plerixafor (AMD3100) plus G-CSF for
autologous hematopoietic progenitor cell mobilization in patients
with Hodgkin lymphoma. Biol Blood Marrow Transplant 2008; 14:
1253–1261.
33 Bennett BL, Sasaki DT, Murray BW, O’Leary EC, Sakata ST, Xu W
et al. SP600125, an anthrapyrazolone inhibitor of Jun N-terminal
kinase. Proc Natl Acad Sci USA 2001; 98: 13681–13686.
34 Reca R, Cramer D, Yan J, Laughlin MJ, Janowska-Wieczorek A,
Ratajczak J et al. A novel role of complement in mobilization:
immunodeficient mice are poor granulocyte-colony stimulating
CXCR4-dependent rapid mobilization of progenitors
A Dar et al
10
Leukemia

Page 11

factor mobilizers because they lack complement-activating im-
munoglobulins. Stem Cells 2007; 25: 3093–3100.
35 Amara A, Lorthioir O, Valenzuela A, Magerus A, Thelen M,
Montes M et al. Stromal cell-derived factor-1alpha associates with
heparan sulfates through the first beta-strand of the chemokine.
J Biol Chem 1999; 274: 23916–23925.
36 Nagasawa T, Hirota S, Tachibana K, Takakura N, Nishikawa S,
KitamuraY etal.Defects
bone-marrow myelopoiesis in mice lacking the CXC chemokine
PBSF/SDF-1. Nature 1996; 382: 635–638.
37 Lavi E, Strizki JM, Ulrich AM, Zhang W, Fu L, Wang Q et al.
CXCR-4 (Fusin), a co-receptor for the type 1 human immunode-
ficiency virus (HIV-1), is expressed in the human brain in a variety
of cell types, including microglia and neurons. Am J Pathol 1997;
151: 1035–1042.
38 Zou YR, Kottmann AH, Kuroda M, Taniuchi I, Littman DR.
Function of the chemokine receptor CXCR4 in haematopoiesis and
in cerebellar development. Nature 1998; 393: 595–599.
39 Serrati S, Margheri F, Fibbi G, Di Cara G, Minafra L, Pucci-Minafra I
et al. Endothelial cells and normal breast epithelial cells enhance
invasion of breast carcinoma cells by CXCR-4-dependent up-
regulation of urokinase-type plasminogen activator receptor
(uPAR, CD87) expression. J Pathol 2008; 214: 545–554.
40 Jin DK, Shido K, Kopp HG, Petit I, Shmelkov SV, Young LM et al.
Cytokine-mediated deployment of SDF-1 induces revasculariza-
tion through recruitment of CXCR4+ hemangiocytes. Nat Med
2006; 12: 557–567.
41 Kollet O, Dar A, Shivtiel S, Kalinkovich A, Lapid K, Sztainberg Y
et al. Osteoclasts degrade endosteal components and promote
mobilization of hematopoietic progenitor cells. Nat Med 2006; 12:
657–664.
42 Massberg S, Konrad I, Schurzinger K, Lorenz M, Schneider S,
Zohlnhoefer D et al. Platelets secrete stromal cell-derived factor
1alpha and recruit bone marrow-derived progenitor cells to arterial
thrombi in vivo. J Exp Med 2006; 203: 1221–1233.
43 Petit I, Jin D, Rafii S. The SDF-1-CXCR4 signaling pathway: a
molecular hub modulating neo-angiogenesis. Trends Immunol
2007; 28: 299–307.
44 Wang Y, Cui L, Gonsiorek W, Min SH, Anilkumar G, Rosenblum S
et al. CCR2 and CXCR4 regulate peripheral blood monocyte
pharmacodynamics and link to efficacy in experimental auto-
immune encephalomyelitis. J Inflamm (Lond) 2009; 6: 32.
45 Tepper OM, Carr J, Allen Jr RJ, Chang CC, Lin CD, Tanaka R et al.
Decreased circulating progenitor cell number and failed mechan-
isms of stromal cell-derived factor-1alpha mediated bone marrow
mobilization impair diabetic tissue repair. Diabetes 2010; 59:
1974–1983.
46 Bonig H, Chudziak D, Priestley G, Papayannopoulou T. Insights
into the biology of mobilized hematopoietic stem/progenitor cells
through innovative treatment schedules of the CXCR4 antagonist
AMD3100. Exp Hematol 2009; 37: 402–415 e1.
47 Allende ML, Tuymetova G, Lee BG, Bonifacino E, Wu YP, Proia
RL. S1P1 receptor directs the release of immature B cells from
bone marrow into blood. J Exp Med 2010; 207: 1113–1124.
48 Downey N, Thien M, Juares J, Bradstock K, Bendall L. Sphingo-
sine-1-phosphate mediates the egress of hematopoietic stem cells
from the bone marrow into the peripheral blood. Exp Hematol
ISEH Ann Meet 2010; 38: Abstract 122.
49 Golan K, Vagima Y, Ludin A, Itkin T, Kalinkovich A, Cohen-Gur S
et al. The lipid chemokine S1P regulates hematopoietic progenitor
cell egress and mobilization via its major receptor S1P1 and by
SDF-1 inhibition, in a p38/Akt/mTOR dependent manner. ASH
Ann Meeting Abstr 2010; 116: Abstract 553.
50 Ratajczak MZ, Lee H, Wysoczynski M, Wan W, Marlicz W,
Laughlin MJ et al. Novel insight into stem cell mobilization-plasma
sphingosine-1-phosphate is a major chemoattractant that directs
of B-celllymphopoiesis and
the egress of hematopoietic stem progenitor cells from the bone
marrow and its level in peripheral blood increases during
mobilization due to activation of complement cascade/membrane
attack complex. Leukemia 2010; 24: 976–985.
51 Ratajczak MZ, Kim CH, Wojakowski W, Janowska-Wieczorek A,
Kucia M, Ratajczak J. Innate immunity as orchestrator of stem cell
mobilization. Leukemia 2010; 24: 1667–1675.
52 Ross J. FGFR1 Is Required for and Elucidates Multiple Mechanisms
of HSC Mobilization. ASH Ann Meet Abstr 2009; 114: Abstract
2532.
53 Donzella GA, Schols D, Lin SW, Este JA, Nagashima KA, Maddon
PJ et al. AMD3100, a small molecule inhibitor of HIV-1 entry via
the CXCR4 co-receptor. Nat Med 1998; 4: 72–77.
54 Schols D, Struyf S, Van Damme J, Este JA, Henson G, De Clercq E.
Inhibition of T-tropic HIV strains by selective antagonization of the
chemokine receptor CXCR4. J Exp Med 1997; 186: 1383–1388.
55 Hendrix CW, Collier AC, Lederman MM, Schols D, Pollard RB,
Brown S et al. Safety, pharmacokinetics, and antiviral activity of
AMD3100, a selective CXCR4 receptor inhibitor, in HIV-1
infection. J Acquir Immune Defic Syndr 2004; 37: 1253–1262.
56 Zabel BA, Wang Y, Lewen S, Berahovich RD, Penfold ME, Zhang P
et al. Elucidation of CXCR7-mediated signaling events and
inhibition ofCXCR4-mediated
migration by CXCR7 ligands. J Immunol 2009; 183: 3204–3211.
57 Yamazaki K, Allen TD. Ultrastructural morphometric study of
efferent nerve terminals on murine bone marrow stromal cells, and
the recognition of a novel anatomical unit: the ‘neuro-reticular
complex’. Am J Anat 1990; 187: 261–276.
58 Aiuti A, Webb IJ, Bleul C, Springer T, Gutierrez-Ramos JC. The
chemokine SDF-1 is a chemoattractant for human CD34+
hematopoietic progenitor cells and provides a new mechanism
to explain the mobilization of CD34+ progenitors to peripheral
blood. J Exp Med 1997; 185: 111–120.
59 Semerad CL, Christopher MJ, Liu F, Short B, Simmons PJ, Winkler I
etal. G-CSFpotentlyinhibits
CXCL12 mRNA expression in the bone marrow. Blood 2005;
106: 3020–3027.
60 Christopher MJ, Liu F, Hilton MJ, Long F, Link DC. Suppression of
CXCL12 production by bone marrow osteoblasts is a common and
critical pathway for cytokine-induced mobilization. Blood 2009;
114: 1331–1339.
61 Lee HM, Wysoczynski M, Liu R, Shin DM, Kucia M, Botto M et al.
Mobilization studies in complement-deficient mice reveal that
optimal AMD3100 mobilization of hematopoietic stem cells
depends on complement cascade activation by AMD3100-
stimulated granulocytes. Leukemia 2009; 24: 573–582.
62 Heissig B, Hattori K, Dias S, Friedrich M, Ferris B, Hackett NR
et al. Recruitment of stem and progenitor cells from the bone
marrow niche requires MMP-9 mediated release of kit-ligand. Cell
2002; 109: 625–637.
63 Pelus LM, Bian H, King AG, Fukuda S. Neutrophil-derived MMP-9
mediates synergistic mobilization of hematopoietic stem and
progenitor cells by the combination of G-CSF and the chemokines
GRObeta/CXCL2 and GRObetaT/CXCL2delta4. Blood 2004; 103:
110–119.
64 Jujo K, Hamada H, Iwakura A, Thorne T, Sekiguchi H, Clarke T
et al. CXCR4 blockade augments bone marrow progenitor
cell recruitment to the neovasculature and reduces mortality
after myocardial infarction. Proc Natl Acad Sci USA 2010; 107:
11008–11013.
65 Heissig B, Lund LR, Akiyama H, Ohki M, Morita Y, Romer J et al.
The plasminogen fibrinolytic pathway is required for hematopoie-
tic regeneration. Cell Stem Cell 2007; 1: 658–670.
66 Binder BR, Mihaly J, Prager GW. uPAR-uPA-PAI-1 interactions and
signaling: a vascular biologist’s view. Thromb Haemost 2007; 97:
336–342.
tumor celltransendothelial
osteoblastactivity and
Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)
CXCR4-dependent rapid mobilization of progenitors
A Dar et al
11
Leukemia