Leukemia (2000) 14, 882–888
 2000 Macmillan Publishers Ltd All rights reserved 0887-6924/00 $15.00
www.nature.com/leu
Effects of the chemokine stromal cell-derived factor-1 on the migration and
localization of precursor-B acute lymphoblastic leukemia cells within bone marrow
stromal layers
KF Bradstock1, V Makrynikola1, A Bianchi1, W Shen1, J Hewson2 and DJ Gottlieb1
Departments of Haematology 1Westmead Hospital and 2Concord Hospital, NSW, Australia
Acute lymphoblastic leukemia (ALL) blasts undergo migration
into layers of bone marrow fibroblasts (BMF) in vitro, utilizing
the ␤1 integrins VLA-4 and VL-5 as adhesion molecules. However, it has been unclear as to whether this is a selective process mediated by specific chemoattractant molecules, or simply a reflection of the highly motile nature of early B cell
precursors. We further characterized this process using a
transwell culture system, in which the two chambers were separated by an 8 ␮m diameter microporous membrane, through
which leukemic cells could move. When a BMF layer was grown
on the upper surface of the membrane there was an 84.1%
reduction in transmigration of the human pre-B ALL cell line
NALM-6 into the lower chamber, compared to control membrane with no BMF layer. Localization of leukemic cells under
the BMF layer was confirmed ultrastructurally, suggesting the
possibility that the migration of leukemic cells was directed by
a chemotactic agent secreted by BMF. The involvement of the
chemokine stromal cell-derived factor-1 (SDF-1) in this process
was next investigated. BMF were shown to express m-RNA for
SDF-1. Addition of SDF-1 at 100 ng/ml into the lower chamber
increased transmigration of NALM-6 across the membrane by
2.2-fold, and also induced a 1.4- to 6.1-fold increase in movement of NALM-6 through a BMF layer into the lower chamber.
The receptor for SDF-1, CXCR4, was demonstrated by flow
cytometry on all 10 cases of precursor-B ALL analyzed, as well
as on NALM-6, KM-3 and REH lines. An inhibitory antibody to
CXCR4 was able to block the migration of NALM-6 cells into
BMF monolayers grown on plastic by 51%, and in nine cases
of ALL by 8–40%, as well as partially inhibit transmigration of
leukemic cells through BMF layers along an SDF-1 concentration gradient. These results confirm that precursor-B ALL
cells selectively localize within bone marrow stroma in vitro,
and that this process is partially due to the stromal chemokine
SDF-1 binding to its receptor CXCR4 on leukemic cells. SDF-1
may be important in influencing the localization of precursor-B
ALL cells in marrow microenvironmental inches which regulate
their survival and proliferation. Leukemia (2000) 14, 882–888.
Keywords: acute lymphoblastic leukemia; adhesion; migration;
stromal cell-derived factor-1; chemotaxis; chemokine
Introduction
The microenvironment of the bone marrow plays a crucial
role in regulating the survival, proliferation, and differentiation
of both normal hemopoietic cells and acute leukemia cells.1–5
Localization of hemopoietic progenitors and leukemic cells
involves interaction of leukocyte adhesion molecules with
counterligands present on bone marrow stromal cells and
extracellular matrix.4–7 Adhesion of normal CD34-positive
bone marrow cells and acute leukemia cells to bone marrow
stroma in vitro has been shown to involve the ␤1 integrins
VLA-4 (CD49d) and VLA-5 (CD49e) as well as the ␤2 integrin
LFA-1 (CD11a), while other adhesion molecules may also be
implicated.7–9 In addition, specific homing molecules10 and
Correspondence: KF Bradstock, Haematology Department, Westmead
Hospital, Westmead, NSW, 2145, Australia; Fax: 612 9689 2331
Received 19 July 1999; accepted 30 November 1999
concentration gradients of chemotactic factors11 or cytokines
may be involved in localization of primitive normal and leukemic hemopoietic cells to the marrow microenvironment.
Adhesion of human lymphoid leukemia cells to the surface
of bone marrow stromal layers in vitro is rapidly followed by
tunnelling into and migration beneath the stromal layers.12–14
This process has been demonstrated to be dependent on the
presence of VLA-4 and VLA-5, and is largely specific for the
precursor-B form of acute lymphoblastic leukemia (ALL) and
related cell lines.12–14 In contrast, acute myeloid leukemia
(AML) cells do not migrate into marrow stromal layers in vitro,
although ultrastructural studies have shown evidence of leukemic cell motility.14,15 This raises the question as to whether
the migration of precursor-B ALL cells into marrow stroma
simply reflects an inherently greater level of motility than
AML, or is selective and related to the presence of a chemotactic factor(s) secreted by stromal cells and concentrated in
their local microenvironment. One potential chemoattractant
for precursor-B ALL cells is the chemokine stromal cellderived factor-1 (SDF-1), a recently described member of the
CXC family of chemokines which includes interleukin-8.16,17
In addition to acting as a chemoattractant for CD34-positive
hematopoietic cells, monocytes and T lymphocytes,18,19 SDF1 is a potent chemokine for B cell progenitors. SDF-1 knockout mice have severe defects in B lymphopoiesis, suggesting
that this chemokine plays a vital role in normal early B cell
development through trafficking of progenitors to appropriate
bone marrow microenvironmental niches.20 SDF-1 binds to
the cellular receptor CXCR4, which is expressed on normal
CD34-positive bone marrow cells.21 However, the expression
of CXCR4 on precursor-B ALL, and the effects of SDF-1 on
their migratory behavior have not been reported to date.
In order to investigate this further, we used a transwell culture system, containing a membrane with 8 micron pores
through which cells are able to move. By interposing a bone
marrow stromal layer on one side of the membrane, we were
able to analyze the degree to which ALL cells are selectively
retained within bone marrow stroma, and examine the effects
of SDF-1 on their migration.
Materials and methods
Leukemic cells
Leukemic cell lines used in this study included the human
pre-B ALL lines NALM-6, REH, and KM-3, which were maintained in RPMI-1640 medium (Gibco, Grand Island, NY, USA)
supplemented with 10% fetal calf serum (FCS; ICN Biomedicals, Costa Mesa, CA, USA), L-glutamine, and antibiotics.
Leukemic blast cells were obtained from untreated patients
with acute leukemia, with informed consent and institutional
ethics committee approval. Heparinized peripheral blood or
bone marrow aspirate samples were diluted with sterile
Stromal cell-derived factor-1 effects on pre-B ALL
KF Bradstock et al
medium and centrifuged on Ficoll–Paque (Pharmacia,
Uppsala, Sweden). Washed mononuclear cells were then
cryopreserved in 10% dimethylsulphoxide, 20% FCS, and
medium, prior to use.
Bone marrow fibroblasts
Bone marrow fibroblasts (BMF) were cultured from normal
bone marrow mononuculear cells, as previously described.8,9
BMF were cultured as adherent layers in McCoy’s medium
(ICN), plus 10% FCS, in 75-cm2 tissue culture flasks (Falcon;
Becton Dickinson Labware, Lincoln Park, NJ, USA) before use
in experiments.
added to the lower chambers. 2 × 105 Na2CrO3-labeled leukemic cells (in 200 ␮l medium) were then added to the upper
compartment, and wells incubated for 5 h (37°C) in 5% CO2
in air. In some experiments, SDF-1 (100 ng/ml) was added to
the lower compartment to form a chemoattractant gradient, as
previously described,19 while in other assays leukemic cells
were pre-incubated for 30 min with 4B4 (5 ␮g/ml) or CXCR4
(50 ␮g/ml) antibodies, before being added unwashed to the
upper compartment. The contents of the lower compartments
were collected, and cells migrating through the membrane
from the upper chamber were quantitated in a gamma counter, and expressed as a percentage of total input cells in the
upper chamber.
883
Electron microscopy
Reagents
Immunofluoresent staining of cells and analysis was carried
out as previously described.9 Briefly, 1–2 × 105 leukocytes
were incubated with titered amounts of CXCR4 monoclonal
antibody, or isotype control, washed, reacted with sheep antimouse IgG conjugated with FITC (Silenus, Melbourne,
Australia), and analyzed in a FACScan flow cytometer
(Becton Dickinson).
Ultrastructural studies of migrating leukemic cells were carried out as previously described.14,15 Briefly, leukemic cells
were allowed to adhere to and migrate into BMF layers grown
on microporous transwell membranes. The membranes were
examined by light microscopy to assess areas of leukemic cell
migration. Relevant foci were then carefully cut from the insert
and divided into eight to 10 segments. The segments were
stacked and embedded in 1% agar gel. The blocks were then
post-fixed in osmium tetroxide 1%, stained en bloc with uranyl acetate 0.5%, dehydrated in serial concentrations of ethanol, and infiltrated through serial concentrations of Spurr resin
according to standard techniques. Each block was then oriented in a Beem capsule, embedded in Spurr resin, thick sections (50 ␮m) cut with a glass knife and viewed under a light
microscope. After trimming the block, thin (90 nm) sections
were cut on a Reichert ultracut E ultramicrotome, grid stained
with Reynolds lead citrate, and viewed with a Philips CM10
transmission
electron
microscope
(Eindhoven,
The
Netherlands).
Migration of leukemic cells into BMF layers
Detection of m-RNA for SDF-1
The ability of pre-B ALL cells to migrate into bone marrow
stromal layers grown on plastic surfaces was assessed using
adherent bone marrow fibroblasts. 2 × 104 BMF were seeded
into 96-well flat-bottom tissue culture plates (Costar, Cambridge, MA, USA), and used 48–72 h later when confluent, as
previously described.14 Leukemic cells were labeled with
51
Na2CrO3, washed, and added at 2 × 105/well to confluent
BMF layers. Plates were incubated at 37°C for 5 h. The BMF
layer was then washed to remove cells from the surface, then
solubilized using Triton X-100, and counted in a gamma
counter. The proportion of migrated cells was then compared
to the total input number, as previously described.14
Total RNA was isolated from human bone marrow fibroblasts
using Total RNA isolation Reagent (Advanced Biotechnologies, Surrey, UK). Complementary DNA was then made by MMLV reverse transcriptase (Promega, Madison, WI, USA), and
PCR was performed for SDF-1␣ and ␤ forms, as well as to the
common region of both forms using specific primers homologous to SDF-1 mRNA sequences, as follows: common region
SDF-1 82-101/319-338; SDF-1 alpha 82-101/777-796: SDF-1
beta 82-101/861-880, according to previously published
results.22 PCR products were electrophoresed on 1% agarose
gels, and visualized with ethidium bromide.
Antibodies used in this work consisted of 4B4 (CD29; Coulter
Immunology, Hileah, FL, USA) and CXCR4 (Clone 44716.111;
R&D Systems, Wiesbaden, Germany). WMD 7, reactive with
a canine leukocyte antigen but not with human cells, was
used as an IgG2b isotype control. Recombinant human SDF1 alpha and polyclonal rabbit anti-SDF-1 were obtained from
R&D Systems.
Flow cytometry
Statistics
Transmigration assay
Student’s t-test was used to compare paired experiments.
A transwell culture system was used to characterize the specificity of cell transmigration, and to evaluate the effects of
chemokines and antibodies. BMF were seeded at 2 × 104 per
well on to 8 micron microporous polycarbonate membranes
(Costar), cultured for 48–72 h, and checked by light
microscopy to ensure complete confluency. Coated membranes were then transferred into 24-well tissue culture plates
(Costar) to separate the culture system into upper and lower
compartments. 600 ␮l of RPMI medium plus 10% FCS was
Results
Effects of bone marrow fibroblasts on pre-B ALL
migration
Using the transwell culture system, 27.2 ± 7.8% of NALM-6
cells were found to undergo transmigration into the lower
Leukemia
Stromal cell-derived factor-1 effects on pre-B ALL
KF Bradstock et al
884
Table 1
Effects of bone marrow fibroblast layer and SDF-1 on
transmigration of NALM-6
Expt No.
Membrane
alone
Membrane
+ BMF
Membrane Membrane +
+ BMF + BMF + SDFSDF-1
1 + AntiCXCR4
1
2
3
Mean ± s.d.
24.7 ± 1.0a
21.0 ± 3.4
36.0 ± 6.0
27.2 ± 7.8
1.5 ± 0.6*b
3.0 ± 0.1*
9.8 ± 0.7*
4.8 ± 4.4*
9.1 ± 2.1*c
9.4 ± 0.2*
13.4 ± 0.8*
10.6 ± 2.4*
3.9 ± 0.9*d
8.5 ± 0.4
10.4 ± 1.2
7.6 ± 3.3
BMF, bone marrow fibroblasts; Anti-CXCR4, inhibitory antibody to
CXCR4 receptor.
a
Values represent mean % transmigration ± s.d. of cells across the
membrane into the lower chamber after 5 h. Experiments 1 and 2
carried out in duplicate, experiment 3 in triplicate. SDF-1 used at
100 ng/ml and anti-CXCR4 at 50 ␮g/ml.
b
*P ⭐ 0.05; comparison of BMF vs membrane alone.
c
Comparison of SDF-1 vs BMF.
d
Comparison of anti-CXCR4 vs SDF-1.
compartment over 5 h (column 1, Table 1). The presence of
a confluent BMF layer on the upper surface of the transwell
membrane resulted in an 84.1 ± 10.6% (P ⭐ 0.03) reduction
in NALM-6 transmigration (column 2, Table 1). Transmission
electron microscopy studies on sections of these membranes
showed that this reduction in transmigration through the
membrane was due to retention of the leukemic cells underneath the BMF layer (Figure 1), in agreement with previous
studies in our laboratory.14,15
Figure 2
PCR of BMF mRNA for SDF-1. Lane 1, molecular weight
markers; lane 2, SDF-1 common region; lane 3, SDF-1 alpha; lane 4,
SDF-1 beta.
Expression of m-RNA for SDF-1 by bone marrow
fibroblasts
PCR for SDF-1 mRNA molecules on total RNA isolated from
BMF revealed bands of characteristic molecular sizes for the
␣ and ␤ forms of SDF-1 (Figure 2, lanes 3 and 4) and the common region for both forms (lane 2, Figure 2). These results
indicate that BMF express mRNA for SDF-1.
Expression and function of CXCR4 chemokine
receptor on pre-B ALL cells
In order to determine whether the chemokine SDF-1 had an
influence on the migration behavior of pre-B ALL cells in the
above experiments, we first analyzed cells for expression of
the receptor for SDF-1, namely CXCR4. As is evident from
Table 2, the majority of cells from three pre-B ALL lines and
Table 2
Analysis of CXCR4 expression on precursor-B ALL cells
by flow cytometry
Cell type
Figure 1
Ultrastructure of NALM-6 cell (n) which has migrated
into a bone marrow fibroblast layer (b) established on microporous
membrane (m). Magnification ×9450.
Leukemia
% Positive
Fluorescence intensity
Cell lines
NALM-6
KM-3
REH
99
99
99
567
115
50
Precursor-B ALL cases
1
2
3
4
5
6
7
8
9
10
94
52
80
98
93
62
74
98
91
98
220
39
30
181
34
10
12
41
40
69
Percentage of cells positive for the CXCR4 receptor by indirect
immunofluorescence and flow cytometry, relative to negative
control. Fluorescence intensity based on mean peak channel of
fluorescence on log scale.
Stromal cell-derived factor-1 effects on pre-B ALL
KF Bradstock et al
10 cases of precursor-B ALL showed moderate to strong
positivity for CXCR4 by flow cytometry (Figure 3).
Next, we examined the effects of an inhibitory antibody to
CXCR4 on pre-B cell migration into confluent layers of BMF
in plastic tissue culture wells. As previously demonstrated,14
approximately 30% of NALM-6 cells localize beneath confluent BMF monolayers within 5 h under these conditions.
This migration could be largely abolished by pre-incubation
of the NALM-6 cells with antibody 4B4 (Table 3), which binds
to human ␤1 integrin and blocks initial adhesion of leukemic
cells to the surface of BMF layers.12–14 When NALM-6 cells
were preincubated with an inhibitory antibody to CXCR4, a
significant reduction in migration of cells into BMF layers
(range 28 to 86%, mean 53.0 ± 17.5%; P ⭐ 0.001) was
observed in seven separate experiments using different
sources of BMF, although this reduction was not as great as
that seen with anti-␤1 integrin antibody. Addition of CXCR4
antibody to anti-␤1 integrin had no apparent additional inhibitory effect compared to anti-␤1 integrin alone (Table 3). A
polyclonal antibody to SDF-1 was also able to significantly
inhibit migration of NALM-6 into BMF in one of two
experiments (Table 3).
Similarly, partial inhibition of migration of leukemic cells
from nine cases of precursor-B ALL into BMF layers was seen
using preincubation with CXCR4 antibody (Table 4). Inhibition ranged from 14.5 to 40.9% in seven cases, and reached
significance in five of these cases, while two cases showed
⬍10% inhibition.
885
Chemotactic effects of SDF-1 on pre-B ALL
In order to further clarify the role of SDF-1 as a putative
chemoattractant for pre-B ALL cells, the migration of NALM6 cells through a microporous membrane into a lower
chamber with or without the addition of SDF-1 was examined
(Figure 4). SDF-1 at either 100 or 200 ng/ml induced a significantly greater degree of migration of NALM-6 over 5 h
than medium alone. Interestingly, preincubation of NALM-6
with CXCR4 antibody only partially inhibited this chemotactic
effect (Figure 4).
Finally, the effects of SDF-1 and the blockade of its receptor
on the transmigration of pre-B ALL cells through BMF layers
were analyzed (columns 3 and 4, Table 1 and Table 5). In
three separate experiments, SDF-1 at 100 ng/ml in the lower
chamber was capable of inducing a significantly greater
degree of migration of NALM-6 through the BMF layer, representing a 1.4- to 6.1-fold increase in migration in these three
experiments (Table 1). This effect could be partially blocked
by preincubation with CXCR4 antibody. A similar picture was
observed when six cases of precursor-B ALL were examined
(Table 5). In three of six cases, SDF-1 induced a significant
increase in transmigration of patient blast cells across a BMF
layer, an effect that was fully or partially inhibitable by incubation of leukemic cells with CXCR4 antibody.
Discussion
Figure 3
Flow cytometry analysis of CXCR4 expression on pre-B
ALL cells. (a) NALM-6; (b) precursor-B ALL case 10; (c) precursor-B
ALL case 4. Solid tracing indicates negative isotype control.
This study has revealed two main new findings regarding the
biology of pre-B ALL. Firstly, the double chamber system used
to quantitate leukemic cell motility confirmed that ALL cells
are selectively retained within a layer of bone marrow stromal
cells. Previous studies have shown that pre-B ALL cells are
highly motile in vitro, and migrate rapidly beneath layers of
bone marrow stromal cells adherent to plastic.12–14 This process is dependent on the adhesion molecules VLA-4 and VLA5, which allow attachment of pre-B ALL cells to their ligands
VCAM-1 and fibronectin within stroma.12–14 Our results
clearly indicate that the migration of pre-B ALL cells beneath
layers of bone marrow fibroblasts is a selective process, since
the microporous membrane system allowed cells that had
Leukemia
Stromal cell-derived factor-1 effects on pre-B ALL
KF Bradstock et al
886
Table 3
Inhibition of NALM-6 migration into BMF layera
Expt No.
1
2
3
4
5
6
7
Mean ± s.d.
Control
Anti-CXCR4
Anti-␤1 integrin
Anti-CXCR4 + anti-␤1
integrin
Polyclonal anti-SDF-1
31.7 ± 5.2b
19.1 ± 1.1
23.5 ± 2.0
28.8 ± 3.1
30.5 ± 0.6
27.3 ± 3.1
14.7 ± 2.7
25.1 ± 6.2
13.5 ± 3.9**
9.0 ± 1.9**
3.4 ± 0.8*
13.9 ± 2.2**
20.0 ± 1.3**
20.0 ± 0.2
5.6 ± 0.8**
12.2 ± 6.5**
2.1 ± 0.6**
NT
NT
3.3 ± 0.7**
5.2 ± 1.0**
NT
NT
3.5 ± 1.5**
NT
NT
NT
3.2 ± 0.4**
5.2 ± 1.7**
NT
NT
4.2 ± 1.4**
NT
NT
NT
NT
NT
19.9 ± 0.1*
10.1 ± 1.6
15 ± 6.8
a
Inhibition of migration of NALM-6 into BMF monolayer on plastic over 5 h.
Experiments 1 to 3 were done in triplicate wells, and experiments 4 to 7 in quadruplicate wells. Values represent mean % migration into
BMF layers ± s.d. NALM-6 cells were pre-incubated with CXCR4, 4B4, or isotype control antibodies for 30 min before addition to BMF.
Polyclonal rabbit anti-SDF-1 (20 ␮g/ml) was added to BMF for 30 min prior to the addition of NALM-6 cells.
*P ⭐ 0.05, comparison with isotype control.
**P ⭐ 0.005, comparison with isotype control.
NT, not tested.
b
Table 4
Inhibition of precursor-B ALL migration into BMF layer
using CXCR4 antibodya
Case No.b
1
2
3
4
5
6
7
8
9
Percent migration
Control
Anti-CXCR4
20.6 ± 0.7
6.1 ± 0.3
3.7 ± 0.1
8.7 ± 0.7
46.5 ± 4.5
11.0 ± 1.5
18.9 ± 2.6
17.2 ± 0.5
26.0 ± 1.4
17.5 ± 1.0
4.9 ± 0.7
3.4 ± 0.1
6.1 ± 2.1
34 ± 4.0
6.5 ± 2.6
14.0 ± 1.2
14.7 ± 0.3
24.0 ± 1.1
% Inhibition
15.0*
19.6
8.0
29.0
26.9*
40.9*
25.9*
14.5*
7.7
a
Inhibition of precursor-B ALL migration into BMF monolayer on
plastic over 5 h, using anti-CXCR4 antibody.
Case numbers identical to those in Table 2.
*P ⭐ 0.05 for comparison of anti-CXCR4 with isotype control.
b
penetrated underneath the BMF layer to continue to move into
the lower chamber if their migration was completely nonselective. The fact that the BMF layer on the microporous
membrane dramatically reduced recovery of leukemic cells in
the lower chamber, together with the ultrastructural demonstration of pre-B ALL cells localized beneath the BMF layer,
confirms that the leukemic cells have selectively ‘homed’ to
this location.
The second major finding is that SDF-1 is a chemokine for
pre-B ALL cells, and may be an explanation, at least in part,
for their localization under bone marrow stromal layers in
vitro. We have shown that BMF produce m-RNA for SDF-1.
Our findings demonstrate that pre-B ALL cells express the
cellular receptor for SDF-1, namely CXCR4,16,17 and are
stimulated to migrate across a microporous membrane along
a concentration gradient of SDF-1. Migration of pre-B ALL
cells into BMF layers could be partially inhibited by preincubation with a blocking antibody to the CXCR4 receptor, or by
an inhibitory antibody to SDF-1. Furthermore, SDF-1 could
induce a proportion of NALM-6 and precursor-B ALL cells to
leave a BMF layer and move across a microporous membrane.
Taken together, these data implicate SDF-1 as a factor
regulating the motility of pre-B ALL cells in the marrow
microenvironment.
Leukemia
Figure 4
Effects of SDF-1 on transmigration of NALM-6 across a
microporous membrane. Transmigration of cells into lower chamber
over 5 h; effects of SDF-1 in lower chamber at 100 and 200 ng/ml,
respectively; SDF-1 at 100 ng/ml in lower chamber, NALM-6 pretreated with anti-CXCR4 antibody. Results indicate percentage of cells
transmigrating +/− s.d., with experiments carried out in triplicate
wells. *P ⭐ 0.05, comparison of SDF-1 with membrane; **P ⭐ 0.05,
comparison of SDF(100)+Ab with SDF(100).
SDF-1 has already been shown to be a potent chemoattractant for human bone marrow progenitor cells, and can induce
both chemotaxis and transendothelial migration of normal
CD34+ cells and colony-forming cells.18,19,21 Both normal
CD34+ cells and leukemic blast cells from cases of acute
myeloid leukemia (AML) have been demonstrated to express
CXCR4, the cellular receptor for SDF-1.21 Since it has been
demonstrated that SDF-1 is produced in high concentrations
by bone marrow stromal cells, it has been speculated that
SDF-1 is an important regulator of stem cell homing and may
be involved in clinically important processes such as stem cell
mobilization.19,21 Although the marrow microenvironment
Stromal cell-derived factor-1 effects on pre-B ALL
KF Bradstock et al
Table 5
cellsa
Case No.b
1
2
4
5
6
7
Effects of SDF-1 on transmigration of precursor-B ALL
Membrane +
BMF
15.9 ± 1.3c
12.9 ± 0.7
22.2 ± 0.9
49.2 ± 0.2
8.1 ± 0.3
44.7 ± 0.3
Membrane +
Membrane +
BMF + SDF-1 BMF + SDF-1 +
Anti-CXCR4
19.2 ± 1.2
21.3 ± 1.3**
59.0 ± 5.0**
63.0 ± 8.3
9.0 ± 0.4
52.6 ± 2.5*
17.4 ± 2.4
15.4 ± 0.8**d
31.7 ± 2.7*
49.7 ± 0.8
8.7 ± 0.1
47.2 ± 2.2
a
Effects of SDF-1 and anti-CXCR4 on transmigration of precursorB ALL cells across transwell membrane coated with BMF, over 5 h.
b
Case numbers identical to those in Table 2.
c
Values represent mean % transmigration of cells across membrane after 5 h ± s.d. carried out in triplicate wells (except for cases
4 and 6, carried out in duplicate wells).
*P ⭐ 0.05; **P ⭐ 0.005, in comparison to membrane + BMF.
d
Comparison to membrane + BMF + SDF-1.
has been shown to be critical in determining the fate of precursor-B ALL cells,4,5 no factors with definite chemotactic
activity for human pre-B cells have been identified until
recently. Despite the finding that B lymphopoiesis is grossly
disrupted in mice lacking SDF-1,20 there have been no reports
of SDF-1 activity or CXCR4 receptor expression on normal or
malignant human pre-B cells. However, recently Scheidweiler
and colleagues23 have reported that normal precursor B cells
express CXCR4, and that SDF-1 was chemotactic for these
cells. Our own findings extend these observations by demonstrating the presence of CXCR4 on the majority of cases of
precursor-B ALL, as well as defining a functional role of SDF1 in leukemic cell migration. Our data do not discriminate
between potential chemokinetic (non-specific motility) and
chemotactic (specific migration) effects of SDF-1, and more
formal checkerboard studies would be required to establish
this (eg Ref. 19). However, SDF-1 has been documented to
have specific chemotactic activity on other human blood cell
types,19 and it is likely that the effects on precursor-B ALL
migration observed in our study were also due to chemotaxis.
Interestingly, we were unable to completely inhibit SDF-1
induced migration, either into a BMF layer, or out of a BMF
layer into a SDF-1 gradient, using an inhibitory antibody to
CXCR4, suggesting either that extremely high concentrations
of SDF-1 exist within stromal layers in vitro, that other chemokines for pre-B ALL exist, or that SDF-1 binds to other receptors than CXCR4. It is relevant to note that Moehle and
coworkers21 also found that transendothelial migration of AML
cells in response to a stromal cell conditioned medium containing SDF-1 was only partially blockable by an antibody to
CXCR4, also suggesting that other chemokines or chemokine
receptors may be involved in the motility of acute leukemia
cells.
In conclusion, we have demonstrated that there is preferential localization of motile pre-B ALL cells within bone marrow
stromal layers in vitro. While this requires the concordant
expression of VLA-4 and VLA-5 to allow adhesion of the leukemic cells with stroma, SDF-1 appears to have significant
chemotactic activity, and its secretion by stromal cells may be
crucial in regulating the movement of ALL blasts within the
bone marrow microenvironment. This may have considerable
ramifications for influencing the subsequent proliferation and
survival of ALL cells in vivo.
Acknowledgements
887
This work was supported by research grants from the National
Health and Medical Research Council of Australia (RG
920127) and the New South Wales State Cancer Council.
References
1 Bentley SA. A close-range cell–cell interaction required for stem
cell maintenance in continuous culture. Exp Hematol 1981; 9:
308–312.
2 Torok-Storb B. Cellular interactions. Blood 1988; 72: 373–385.
3 Dorshkind K. Regulation of hemopoiesis by bone marrow stromal
cells and their products. Ann Rev Immunol 1990; 8: 111–137.
4 Bradstock KF, Gottlieb DJ. Interaction of acute leukemia cells with
the bone marrow microenvironment: implications for control of
minimal residual disease. Leuk Lymphoma 1995; 18: 1–16.
5 Patrick CW, Smith TW, McIntire LV, Juneja HS. Cellular interactions among marrow stromal and normal/neoplastic pre-B and Blymphoblastic cells. Leuk Lymphoma 1996; 22: 205–219.
6 Simmons PJ, Zannettino A, Gronthos S, Leavesley D. Potential
adhesion mechanisms for localization of haemopoietic progenitors
to bone marrow stroma. Leuk Lymphoma 1994; 12: 353–363.
7 Liesveld JL, Dispersio JF, Abboud CN. Integrins and adhesive
receptors in normal and leukemic CD34+ progenitor cells: potential regulatory checkpoints for cellular traffic. Leuk Lymphoma
1994; 14: 19–28.
8 Bendall LJ, Kortlepel K, Gottlieb DJ. Human acute myeloid leukemia cells bind to bone marrow stroma via a combination of beta1 and beta-2 integrin mechanisms. Blood 1993; 82: 3125–3132.
9 Bradstock K, Makrynikola V, Bianchi A, Byth K. Analysis of the
mechanism of adhesion of precursor-B acute lymphoblastic leukemia cells to bone marrow fibroblasts. Blood 1993; 82: 3437–3444.
10 Tavassoli M, Hardy CL. Molecular basis of homing of intravenously transplanted stem cells to the marrow. Blood 1990; 76:
1059–1070.
11 Cherry B, Yasumizu R, Toki J, Asou H, Nishino T, Komatsu Y,
Ishehara S. Production of hematopoietic stem cell chemotactic
factor by bone marrow stromal cells. Blood 1994; 82: 3125–3132.
12 Miyake K, Hasunuma Y, Yagita H, Kimoto M. Requirement for
VLA-4 and VLA-5 integrins in lymphoma cells binding to and
migration beneath stromal cells in culture. J Cell Biol 1992; 119:
653–662.
13 Tang J, Scott G, Ryan DH. Subpopulations of bone marrow fibroblasts support VLA-4 mediated migration of B cell precursors.
Blood 1993; 82: 3415–3423.
14 Makrynikola V, Bianchi A, Bradstock K, Gottlieb D, Hewson J.
Migration of acute lymphoblastic leukemia cells into human bone
marrow stroma. Leukemia 1994; 8: 1734–1743.
15 Hewson J, Bianchi A, Bradstock K, Makrynikola V, Gottlieb D.
Ultrastructural changes during adhesion and migration of pre-B
lymphoid leukaemia cells within bone marrow stroma. Br J
Haematol 1996; 92: 77–87.
16 Bleul CC, Fuhlbrigge RC, Casasnovas JM, Aiuti A, Springer TA. A
highly efficacious lymphocyte chemoattractant, stromal cellderived factor 1 (SDF-1). J Exp Med 1996; 184: 1101–1109.
17 Nagasawa T, Nakajima T, Tachibana K, Iizasa H, Bleul CC, Yoshie
O, Matsushima K, Yoshida N, Springer TA, Kishimoto T. Molecular cloning and characterization of a murine pre-B-cell growthstimulating/stromal cell derived factor 1 receptor, a murine homolog of the human immunodeficiency virus entry coreceptor fusin.
Proc Natl Acad Sci USA 1996; 93: 14726–14729.
18 Aiuti A, Webb IJ, Bleul CC, Springer TA, 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.
19 Kim CH, Broxmeyer HE. In vitro behaviour of hematopoietic progenitor cells under the influence of chemoattractants: stromal cellderived factor-1, steel factor, and the bone marrow microenvironment. Blood 1998; 91: 100–110.
20 Nagasawa T, Hirota S, Tachibana K, Takakura N, Nishikawa S,
Kitamura Y, Yoshida N, Kikutani H, Kishimoto T. Defects of B cell
Leukemia
Stromal cell-derived factor-1 effects on pre-B ALL
KF Bradstock et al
888
Leukemia
lymphopoiesis and bone marrow myelopoiesis in mice lacking the
CXC chemokine PBSF/SDF-1. Nature 1996; 382: 635–638.
21 Moehle R, Bautz F, Rafii S, Moore MAS, Brugger W, Kanz L. The
chemokine receptor CXCR4 is expressed on CD34+ hematopoietic
progenitors and leukemic cells and mediates transendothelial
migration induced by stromal cell derived factor-1. Blood 1998;
91: 4523–4530.
22 Shirozu M, Nakana T, Inazawa J, Tashivo K, Tada H, Shinohava
T, Honjo T. Structure and chromosomal localization of the human
stomal cell-derived factor 1 (SDF1) gene. Genomics 1995; 28:
495–500.
23 Sheidweiler K, Ritterman I, Tang J, Fedyk E, Springer T, Ryan D.
SDF-1 and its receptor CXCR4 are required for migration of human
B cell precursors under stroma and their proliferation in culture.
Blood 1998; 92 (Suppl. 1): 24a, Abstr. 88.