Interaction of Human Bone Marrow Fibroblasts With

From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
Interaction of Human Bone Marrow Fibroblasts With Megakaryocytes: Role of
the c-kit Ligand
By Hava Avraham, David T. Scadden, Sherry Chi, Virginia C. Broudy, Krisztina M. Zsebo, and Jerome E. Groopman
Human kit ligand (KL), also known as stem cell factor (SCF),
steel factor, or mast cell growth factor, is a recently identified
hematopoietic growth factor whose receptor is the product
of the c-kit proto-oncogene. Alternative splicing of the premRNA of KL/SCF results in secreted and membrane-bound
forms of the protein. We and others have recently shown that
the c-kit gene product is expressed on human megakaryocytes and that soluble KL/SCF in combination with granulocyte-macrophage colony-stimulating factor, interleukin-3 (IL3). or IL-6 increased megakaryocyte progenitor colony
formation (CFU-MEG) and stimulated mature megakaryocytes. Here we show that adhesion of human megakaryocytes t o bone marrow stromal fibroblasts, which express the
membrane-bound form of KL/SCF (mKL/SCF), is mediated
in part by the interaction between mKL/SCF and the c-kit
protein. This interaction also results in marrow fibroblaststimulated proliferation but not an increase in ploidy of
megakaryocytes; when the t w o cell types were separated by
a transoluble membrane, proliferation did not occur. Adhesion and proliferation of human megakaryocytesto an immortalized murine stromal cell line SI/SI lacking the KL/SCF
gene was impaired, whereas transfection of SI/SI cells with
human mKL/SCF significantly increased both adhesion and
proliferation. Marrow stromal fibroblast mKL/SCF may serve
both as an adhesion structure and as a growth-potentiating
factor for megakaryocytes in the bone marrow.
0 1992by The American Society of Hematology.
T
be anchored to stromal elements within the bone marrow
microenvironment and positioned to respond to cytokines.
HE HUMAN kit ligand (KL), also known as stem cell
factor (SCF), mast cell growth factor, or steel factor
has recently been identified as the product of the S1
The cell surface receptor for KL/SCF is the product of the
c-kit proto-oncogene. Mice with mutations in the SI locus
have abnormalities in hematopoiesis, germ cells, and melanocytes. Two forms of KL/SCF have been described that
arise through differential RNA processing: a membranebound species and a soluble secreted form.3 Soluble KL/
SCF in vitro increased the growth response of progenitor
cells to later-acting cytokines such as interleukin-3 (IL-3),
erythropoietin, and granulocyte colony-stimulating factor
(G-CSF).1-3 It has recently been observed that colonyforming unit-megakaryocyte progenitors (CFU-Meg) and
mature human megakaryocytes proliferated in response to
soluble recombinant KL/SCF.4,5
Cos cells expressing surface murine KL/SCF adhered
with murine mast cells: suggesting that the membranebound molecule may have a different function from the
soluble species. A study of murine mast cell adhesion to
mesenchymal cells derived from Sl/Sl mice found that the
extracellular domain of mKL/SCF was required to mediate
this adhesion.’
The interaction of human megakaryocytes with adhesion
molecules present on bone marrow stromal cells has not
been defined. To explore the potential role of mKL/SCF as
an adhesion structure for megakaryocytes and other c-kit
expressing hematopoietic cells, we studied adhesive interactions of human megakaryocytic cells with bone marrow
stromal fibroblasts.
Our studies showed that adhesion of human megakaryocytes could be mediated in part via their c-kit receptor
binding to membrane-associated KL/SCF expressed by
bone marrow fibroblasts. Further, direct interaction between stromal fibroblasts and megakaryocytes induced
DNA synthesis as measured by thymidine incorporation in
the megakaryocytic cells. The dual function of mKL/SCF as
both an adhesion structure and a regulator of proliferation
is novel among hematopoietic growth factors, and provides
a model whereby hematopoietic cells, such as megakaryocytes, mast cells, and early hematopoietic progenitors may
Blood, Vol80, No 7 (October I),
1992: pp 1679-1684
MATERIALS AND METHODS
Cells. Human bone marrow was obtained by aspiration from
the iliac crest of normal donors who gave informed consent in a
protocol approved by the New England Deaconess Hospital
Institutional Review Board. The marrow was aspirated into preservative-free heparin (Sigma Chemical CO, St Louis, MO) and
separated by centrifugation through Ficoll-Hypaque (Pharmacia,
Piscataway, NJ) at 1,200g at room temperature for 30 minutes.
After two washes with sterile lx phosphate-buffered saline (PBS),
the cells were resuspended in Iscove’s modified Dulbecco’s medium (IMDM) with 20% fetal calf serum (FCS), penicillin/
streptomycin (PIS), and L-glutamine; seeded onto T-75 tissue
culture flasks (Corning, Corning, NY);and incubated at 37°C in
5% COz. After 48 hours, the nonadherent cells were gently
removed, and the adherent cells were refed with fresh media, The
cells were refed with fresh medium every 3 days and trypsinized
and split after 1 week or when confluent. Cells underwent three
cycles of trypsinization and splitting before characterization or use
in experimental protocols. Cultures of marrow stromal fibroblasts
prepared by this method were uniformly positive for vimentin,
negative for cytokeratin, negative for von Willebrand’s factor, and
negative for nonspecific esterase by immunochemical or histochem-
From the Division of HematologyIOncoIogy, Department of Medicine, New England Deaconess Hospital, Harvard Medical School,
Boston, 11.24; the Division of Hematology, University of Washington,
Seattle, WA; and Amgen, Inc, Thousand Oaks, CA.
Submitted February 20,1992; accepted June 8, 1992.
Supported in part by grants from the National Institutes of Health
HL33774, HL42112, HL43510, HL46668, AI-29847, CA5.5520, and
the Department of Defense 17-90-CO106.
Presented in part at the meeting of the American Society of
Hematology, December 1991, Denver, CO.
Address reprint requests to Jerome E. Groopman, MD, New
England Deaconess Hospital, Division of HematologylOncologv, I I O
Francis St, Suite 4A, Boston, 11.24 02215.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with I8 U.S.C.section I734 solely to
indicate this fact.
0 1992 by The American Society of Hematology
0006-4971I92 l8007-0006$3.00/0
1679
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
AVRAHAM ET AL
1680
ical staining. Human marrow megakaryocytes were isolated by a
method employing immunomagneticbeads using antihuman glycoprotein GPllbIllla monoclonal antibody (MoAb) as described
previously! All of the isolated cells were recognizable as megakaryocytes by morphology and/or specific immunofluorescence using
antiplatelet antibodies, GPllbIllla, and Gplb. H9 T cells were
obtained from Dr Robert Gallo. the National Cancer Institute, and
cultured in RPMI 1640 medium with 10% FCS, PIS, and Lglutamine.
The CMK cell line, provided by Dr T. Sat0 and derived from a
child with megakaryoblastic leukemia, has properties of cells of
megakaryocytic lineage, including surface expression of glycoproteins Ib and Ilb/llla, synthesis of platelet factor 4. platelet derived
growth factor, von Willebrand factor. and becomes polyploid on
induction with phorbol esters.".l" No myeloid or lymphoid surface
markers have been found on our cultured CMK cells. The CMK
cell line was cultured in RPMI 1640 + 10% FCS.
An immortalized stromal fibroblast cell line derived From the
fetal hematopoietic microenvironment of a murine homozygous
(SIISI) embryo that lacks the entire coding sequence of the Steel
gene has been derived." This stromal cell line (SI/SI) has been
used for the transfection of a cDNA encoding a 220 amino acid
SCF protein (hSCF2x') lacking the proteolytic cleavage site because of the exclusion of exon 6 by Toksoz et al." Therefare. the
220-amino acid polypeptide remains membrane-bound in the
transfected SIISI cells. Neither the original cell line (SIISI) nor
transfectants containing the 220 cDNA (SI/SI-SCF2x') secreted
detectable amounts of KLISCF. The SI/SI-hSCF22"transfectant
showed a high level of membrane-associated hKLISCF protein as
measured by FlTC goat antimouse staining of cells labeled with a
primary MoAb to human KLISCF."
Cell-cell adhesion assay. To measure cell-cell adhesion, lo(
CMK cells or bone marrow megakaryocytes in 2 mL of Dulbecco's
modified Eagle's medium (DMEM) with 10% FCS were added to
marrow stromal fibroblasts. The cells were allowed to settle for I
hour at 37°C. followed by swirling to redistribute the floating cells
and the cells were resettled for an additional hour. Following
c-kit
1
1
PCR Analysis
Southern Blot
A
B
settling of the cells, nonadherent cells were then washed away in
four changes of medium, centrifuged at 1,500 rpm for 10 minutes
and counted. The adherent megakaryocytes were quantitated by
dislodging them by vigorous pipeting up and down and then
counting with Trypan blue. For inhihition studies, we used the
following specific antibodies: control IgG ascites (at dilution
1:l.OOO); purified SR-I MoAb, which recognizes the human KLI
SCF receptor c-kit (at dilution of 100 ng/mL)I2; purified control
IgG ascites (at dilution of 100 ngImL); MoAbs TSlI22 and
TSI 118. which recognize the LFA-la subunit and LFA-IS subunit.
respectively (Dr T.A. Springer, Center For Blood Research, Harvard Medical School, Boston, MA); and MoAb RRIII, which
recognizes ICAM-I (CD54) (also provided by Dr T.A. Springer).
MoAbs to human GPlb, GPIlb/Illa, and L e u 4 were obtained
from Becton Dickinson (Mountain View, CA). All MoAbs were
used at a dilution of 1:lOO. To assay inhibition of adhesion, cells
were treated (30 minutes, 37°C) with the relevant antibody, as
indicated in the text. After 30 minutes, the cells were washed three
times with PBS and then added to stromal fibroblast monolayers
with incubation for 1 hour at 37°C. Nonadherent cells were
removed and the remaining adherent cells were collected,
trypsinized, and counted as described above.
RNA and DNA analysis. RNA was isolated by the guanidium
isothiocyanate method. First-strand cDNA was synthesized at 37°C
for 1 hour in a final volume of 10 pL with oligo-dT as primers; 4.S
pL RNA in DEPC-dH20, 2.0 pL 5 x buffer (250 mmolIL
Tris-HCI, pH 8.3.375 mmolIL KCI, SO mmolIL dithiothreitol, 15
mmolIL MgC12, and 250 pg/mL actinomycin D), 0.5 pL RNAsin
(40 U/pL, Promega, Madison, WI), 1.0 pL dNTP (dATP, dCTP,
dGTP, dTTP mix, 10 mmolIL each (Pharmacia), 1.0 pL oligo-dT,
1.0 pL Maloney murine leukemia virus (MMLV) reverse transcriptase (2,000 UImL, Boehringer-Mannheim, Chicago, IL).
Polymerase chain reacrion (PCR). Eighty microliters of PCR
mix was added to 10 pL of first-strand cDNA. PCR mix contains:
53.5 pL sterile water, 10 pL 10 x buffer, 16 pLof dNTP mix (each
at 1.25 mmolIL) and 0.5 pL (2.5 U) of the Themus aqitariars
thermostable DNA polymerase (Cetus-Perkin Elmer, Emeryville,
SCF
PCR Analysis Southem Blot
C
D
Fig 1. PCR analpis of c-kit and KL/SCF cDNAs from megakaryocytes, CMK cells, H9 T-cells, and bone marrow fibroblasts. PCR was performed
on cDNA preparedfrom RNA isolatedfrom the indicated source of cells as described. The PCR products were separated on 2% or 1% agarose gels
(c-kit and KL/SCF, respectively) run in Tris-borate EDTA buffer (TBE) containing 0.25% pgImL ethidium bromide. Southern blot hybridizationof
the PCR productswas performedusing specific probesfor e-kit and KLISCF. The probe of KL/SCF was generatedby linearizing pGEM3:hKL/SCF
no. 9 with EcoRI-Hindlll using the 218-bp insert as a probe. The probe of c-kit was generated by linearizing phc-kit 171 with Sstl and using the
1.25-kb insert as a probe5
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
KIT LIGAND MEDIATES MEGAKARYOCME ADHESION
-,
,-
Adhesion
7
SCF TreatmentI-
:
1681
Statistical analysis. The results were expressed as the mean -C
SEM of data obtained from three or more experiments performed
in duplicate or four replicate, as indicated in the text. Statistical
significance was determined using the Student's t-test.
20
RESULTS
c
0 10
Adhesion of megakatyocytes to bone marrow fibroblasts.
CMK cells and isolated human megakaryocytes expressed
the c-kit proto-oncogene (Fig 1A and B). Bone marrow
stromal fibroblasts expressed mRNAs related to both the
A
r Untreated
McA
CMK Cells c
b
C@n'vl Ascttes
Hours
Minutes
E
F
LAscites IgG
GplbGpIlbflIIaLFA-1 h
Leu8 I
C-klt
i+
-I
c.kll + G n l l h / l l I ~
Human Isolated Meg.
Untreated
Fig 2. Adhesion of megakaryocytes to bone marrow fibroblasts.
(A,C) Adherent CMK cells or marrow megakaryocyteswere counted in
cytocentrifugepreparations stained with trypan blue. (B,D) CMK cells
or marrow megakaryocytes were treated with human recombinant
soluble KL/SCF (50 ng/mL) for 24 hours. The cells were washed and
then added to the bone marrow fibroblast monolayers for 1 hour to
permit adhesion. Values are mean f SEM, n = 8. Adhesion was
significantly inhibited by soluble SCF/KL treatment, P c .05. (E)
Photograph of clusters of isolated marrow megakaryocyte adherent
to bone marrow fibroblasts. (F) Photograph of isolated marrow
megakaryocytes pretreatedwith human recombinant soluble KL/SCF
(50 ng/mL) for 24 hours. There is minimal adhesion to the bone
marrow fibroblasts. Bone marrow fibroblast cells are visible as having
fibroblast-like morphology and megakaryocytes as small round refractile cells.
CA).'O Five microliters of each primer was added to give a final
primer concentration of 1 pmol/L, and the mixture was then
subjected to PCR amplification using the Perkin Elmer thermal
cycler set for 40 cycles. The temperatures used for PCR were:
denature 94°C. 1 minute; primer anneal 55°C. 2 minutes; primer
extension 72"C, 3 minutes. Normally, 1-minute ramp times were
used between these temperatures. The specific primers for c-kit
and KLlSCF were obtained from Amgen (Thousand Oaks, CA).
Southern blotting was performed by standard procedures. The
hybridization step was carried out in 50% formamide at 42°C. The
probe of KLlSCF was generated by linearizing pGEM3:hSCF#9
with EcoRI-Hind111using the 218 insert as a probe. The probe of
c-kit was generated by linearizing phc-kit 171 with Sstl and using
the 1.25-kb insert as a probe.
Assay of 17H] thymidine uptake by cultured cells. DNA synthesis
and cell viability were respectively assessed by ['HI thymidine
incorporation and by Trypan blue exclusion (0.4% Trypan blue
stain in 0.85% saline; GlBCO Laboratories, Grand Island, NY).
Cells were pulsed with 0.5 pCi per well of [-'HI
thymidine (New
England Nuclear, Boston, MA) and incubated for an additional 5
hours. Samples were harvested onto glass fiber filters washed with
saline and then counted by liquid scintillation spectrometry.
Quantitation of viable cell number by Trypan blue exclusion
thymidine incorporation.
confirmed proliferation results with [3H]
Control Ascites
LAscites IgG
20
0
200
100
ki
a
2
200
40
60
4%
80
99 %
- 4
1
CMK
Cells
\i
8%
3
a
82 %
1
1
Human
Isolated
Meg.
1
Fluorescence lntensity
Fig 3. (A) Effects of specMc antibodies against c-kit, GPllb/llla,
GPlb, LFA-1, and Leu 8 on adhesion of megakaryocytic cells to bone
marrow fibroblasts. CMK cells or isolated marrow megakaryocytes
were incubated without antibodies, or with MoAbs for GPlb, GPllb/
Illa, LFA-1, Leu 8, or c-kit for 30 minutes at 37°C. The cells were then
washed and added to bone marrow fibroblasts for 1 hour at 37°C.
Values are mean f SEM, n = 6. Adhesion was significantly inhibited,
P c .05, compared with control. (B) Immunofluorescence analysis
using flow cytometry of CMK cells or isolated marrow megakaryocytes stained with specific antibodiesfor e-kit. CMK cells or megakaryocytes were incubated with purified control IgG ascites (100 ng/mL)
or purified SR-1 antibody (100 ng/mL), which recognizes human c-kit.
Cells, 105, were analyzed in each instance and fluorescence intensity
was displayed in relative intensity on a logarithmic scale. Percentage
of positive cells, calculated between channel numbers 50 and 200, are
indicated. An unrelated FITC-labeled conjugate (Swine antirabbit Ig)
stained about 2% of each suspension (curves not shown).
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
AVRAHAM ET AL
1682
soluble and membrane forms of KL/SCF (Fig 1C and D).
Low but detectable amounts of secreted soluble KL/SCF
( 0.3 ng/mL) were measured using a radioreceptor binding assay in the conditioned media of stromal fibroblasts.'
Incubation of CMK megakaryocytic cells with primary
bone marrow stromal fibroblasts resulted in adhesion of the
normally nonadherent CMK cells to the fibroblast monolayer. This interaction was rapid and was observed after 15
minutes (Fig 2A). Addition of soluble recombinant human
KL/SCF at concentrations of 50 ng/mL inhibited subsequent adhesion of CMK cells to the fibroblasts (Fig 2A and
B).
Adhesion also occurred between isolated human marrow
megakaryocytes and bone marrow stromal fibroblasts. These
studies showed adhesion of marrow megakaryocytes within
30 minutes that persisted for more than 24 hours (Fig 2C
and E). Treatment with soluble recombinant KL/SCF (at
50 ng/mL) for 24 hours inhibited adhesion of marrow
megakaryocytes to stromal fibroblasts (Fig 2D and F).
Inhibition of adhesion was also observed on pretreatment of CMK cells or marrow megakaryocytes with the
SR-1 neutralizing MoAb to c-kit,'* but not in control
cultures treated with an irrelevant MoAb of the same
isotype, the control IgG ascites, or a control MoAb to HLA
surface antigen (Fig 3A). Flow cytometry of CMK cells or
marrow megakaryocytes treated with the anti-c-kit SR-1
antibody showed that 99% and 83%, respectively, stained
positively (Fig 3B). Adhesion was also inhibited by pretreatment of stromal fibroblasts with polyclonal anti-KL/SCF
antibodies, but not with control rabbit serum (data not
shown).
-
Isolated Bone Marrow
Megakaryocytes
CMK
1
I
1
I
I
A
To ascertain which other cell surface adhesion molecules
could mediate cell-cell interaction between CMK cells and
the marrow fibroblasts, cross-blocking cell adhesion experiments were performed using inhibitory MoAbs specific for
the LFA-1 complex, GPIIblIIIa, GPIb, and Leu 8 struct u r e ~ . 'MoAbs
~
to LFA-1, GPIb, GPIIb/IIIa, and Leu 8
blocked adhesion of CMK cells. The combination of these
MoAbs with the SR-1 antibody to c-kit antibodies resulted
in a small increase in inhibition of adhesion of CMK cells
(Fig 3A).
To further confirm that the mKL/SCF and c-kit interaction mediated adhesion and that the anti-c-kit antibody was
not simply sterically blocking interactions involving other
adhesion molecules on the megakaryocyte cell surface, two
approaches were used. First, Fab-specific antibodies against
c-kit, GPIIb/IIIa, GPIb, and LFA-1 were prepared and
tested for their effects on adhesion. These Fab antibodies
inhibited adhesion of CMK cells comparable to that seen
with the intact MoAbs (data not shown). The second
approach involved immunofluorescence analysis by flow
cytometry of CMK cells using specific antibodies for c-kit,
LFA-1, Leu 8, GPIIb/IIIa, and GPIb. In these experiments,
CMK cells were first stained with the SR-1 MoAb to c-kit.
The immunofluorescence staining of these other adhesive
molecules on CMK cells treated with SR-1 was confirmed
by flow cytometry (data not shown). Thus, each neutralizing
MoAb was able to interact with its adhesion molecule on
the cell surface even in the presence of SR-1 binding to
c-kit.
Taken together, these observations suggested that c-kit
and mKL/SCF interactions were directly mediating adhe-
1
0
Meg.
Co-cultured
Meg.tBMF
D CMofMeg.
Add to BMF
E CMofBMF
-5
Add to Meg,
4
n.d.
F
9
m
10
Co-cultured
with
membrane
1
24
48
72
1
Time of Culture ( h )
24
48
72
Fig 4. Stimulation of DNA synthesis measured
by [JH]thymidine incorporationin megakaryocytic
cells cocultured with bone marrow fibroblasts.
Isolated human megakaryocytes or CMK cells were
added to a subconfluent bone marrow fibroblast
(BMF) monolayer in 24-well plates (Costar, Cambridge, MA). After the indicatedcoculture time, the
entire coculture was assayed for PHI thymidine
uptake over a 5-hour period. Each assay was performed on duplicate wells. Results are given as
means f SEM and are the pooled results of four
independent experiments. 'Significantly elevated
in the cocultures compared with the bone marrow
fibroblasts, CMK cells, or isolated megakaryocytes
alone (columns A plus 6, P < .(E).*Significantly
elevated compared with the sum of the conditioned
media (CM) results (columns D plus E, P < .05).
3Significantly elevated compared with cocukure
separated by a transoluble membrane (column F,
P < .05).
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
KIT LIGAND MEDIATES MEGAKARYOCYTE ADHESION
sion and additional adhesive molecules, such as GPIIbl
IIIa, LFA-1, and Leu 8, also can participate in the adhesion
process between megakaryocytes and marrow fibroblasts.
DNA synthesis in megakaryocytes. Additional experiments were performed to determine whether adhesion of
bone marrow fibroblasts and CMK cells or isolated marrow
megakaryocytes triggered DNA synthesis in these cells (Fig
4). In a coculture assay, bone marrow fibroblasts were
incubated directly with CMK cells or isolated megakaryocytes. Stimulation of marrow megakaryocyte [3H]thymidine
uptake was observed. Similar stimulation, though of a much
lesser magnitude, was seen with the permanent CMK cell
line, which grows autonomously. CMK cells or human
megakaryocytes alone, bone marrow fibroblasts alone, bone
marrow fibroblasts incubated in CMK or megakaryocyte
conditioned media, or isolated megakaryocytes or CMK
cells cultured in media conditioned by marrow fibroblasts
did not achieve similar levels of [3H]thymidine incorporation. In addition, CMK cells or marrow megakaryocytes
cultured in the presence of marrow fibroblasts but separated by a transoluble membrane filter did not show
augmented [3H] thymidine incorporation, indicating that
direct contact was required for DNA synthesis and proliferation. Quantitation of viable cell number by Trypan blue
exclusion confirmed proliferation results with [3H] thymidine incorporation (data not shown). Attachment of CMK
cells or megakaryocytes to marrow fibroblasts resulted in
cell division not nuclear endoreduplication. In addition,
these experiments were repeated with irradiated bone
marrow fibroblasts. The results were similar to that of
nonirradiated bone marrow fibroblasts (data not shown).
Adhesion of megakaryocytes to Sl/Sl-hSCF220. Another
approach to further show that the membrane form of
fibroblast KLiSCF mediated megakaryocyte adhesion was
pursued. In a coculture assay, CMK or marrow megakaryocytes were incubated directly with murine SI/Sl cells or
SI/SI-hSCPZ0, then floating megakaryocytic cells were
washed away and the numbers of adherent cells determined. Megakaryocytes showed marked adhesion and
spreading within 30 minutes and a maximal degree of
adhesion was reached after 3 hours (54% f 9%;
mean t SEM of five experiments). Only a small number of
megakaryocytes adhered to the nontransfected murine
SliSl cells (21% ? 7%; mean f SEM of five experiments),
even after longer incubation times. The difference in
adhesion of megakaryocytes to SI/SI-hSCFZz0compared
with parent murine SliSl cells was statistically significant
(P < .05). These results indicated that murine SI/SIhSCFZ2O cells, which expressed the membrane form of
human KL/SCF, promoted adhesion of human megakaryocytes, whereas parental murine SliSl cells did not. Increased [3H] thymidine incorporation was observed in
cocultures of adherent CMK cells or marrow megakaryocytes and irradiated S1/SI-hSCFZz0cells, but not with
irradiated parental Sl/Sl cells (Fig 5), showing that direct
interaction of c-kit and mKL/SCF resulted in DNA synthesis in the megakaryocytic cells.
1683
1
Fibroblasts
1 Co-cultured cells1
0
Isolated
Bone
Marrow
Meg.
-15
-
-10
-
-5
5.-
E
F
z
0
Fig 5. Stimulation of DNA synthesis measured by 3[H] thymidine
incorporation in megakaryocytic cells cocultured with SI/SI or SI/SIhKL/SCFzzOfibroblast cells. Isolated human megakaryocytes or CMK
cells (referred to as MEG in figure) were added to a subconfluent
irradiatedfibroblast monolayer (3 Gy), in 24-well plates (Costar).After
72 hours, the entire coculture was assayed for [3H]thymidine uptake.
Each assay was performed on four replicate wells in six different
experiments. Results are given as means f SEM. Megakaryocytes
cocultured with SI /SI-hKL/SCF" gave incorporation significantly
above the SI/SI cocultured system with P c .01. There was no
significant difference in thymidine incorporation in CMK cells alone
compared with CMK cells cocultured with nontransfectedSI/SI cells.
DISCUSSION
Our results show that human megakaryocytes can adhere
to bone marrow stromal fibroblasts, which results in induction of DNA synthesis in the megakaryocytes. This adhesion may be mediated in part by the specific interaction of
the c-kit product on the megakaryocyte cell surface with the
membrane-associated form of KL/SCF on the marrow
fibroblast. This conclusion is based on the observations that
treatment of megakaryocytes with soluble recombinant
KL/SCF inhibited this adhesion, and that inhibition of
adhesion was also observed with the SR-1 antibody to c-kit
or with polyclonal antibodies to IU/SCF. Direct cell-cell
interaction of megakaryocytes with stromal fibrobIasts was
required for DNA synthesis in the megakaryocytes, as
shown by the coculture studies. This was most clearly seen
using primary bone marrow megakaryocytes, which are
likely to be more physiologically relevant compared with
the immortalized CMK cell line.
During the course of our experiments, two other research
groups presented data indicating that the product of the
Steel gene may mediate mast cell adhesion. Flanagan et a16
transfected cos cells with murine KL/SCF and observed
aggregation of murine mast cell lines. Kaneko et a17 also
reported murine mast cell adhesion with the extracellular
domain of IUISCF. Our work addressed the functions of
membrane-associated human KLiSCF in the context of
megakaryocytopoiesis. The bone marrow stromal microen-
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
AVRAHAM ET AL
1684
vironment may serve a number of functions in megakaryocyte growth and maturation, acting as a support framework
for developing progenitor cells as well as a rich source of
growth factors.14J5Expression of alternative forms of KL/
SCF by stromal fibroblasts may be a point of regulation of
megakaryocytes and other c-kit expressing cells, because
such hematopoietic cells may use this surface structure as
a n adhesion molecule. Furthermore, direct interaction of
mKL/SCF with t h e c-kit receptor may result in potentiation
of cell growth, particularly in conjunction with other cytokines such as IL-3, IL-6, and granulocyte-macrophage
CSF.4,5J6
The dual function of mIU/SCF-mediating megakaryocyte adhesion and proliferation-is novel among the
hematopoietic growth factors reported to modulate cells of
this lineage.15 Whether adhesion and proliferation are
regulated by independent functional domains within m a /
SCF or its receptor, c-kit, is subject t o further study.
ACKNOWLEDGMENT
We thank Dr T.A. Springer (Harvard Medical School, Boston,
MA) for providing the LFA-1 MoAbs, Dr D. Williams and Dr
Denise Toksoz for providing us with the SI/SI and SI/SI-hSCF2”
cell lines, Dr Larry Bennett for assaying soluble KL/SCF and
providing anti-KL/SCF antibody, and Patricia Grynkiewicz for
preparation of the manuscript.
REFERENCES
9. Komatsu N, Suda T, Moroi M, Tokuyama N, Sakato Y,
1. Zsebo KM, Williams DA, Geissler EN, Broudy VC, Martin
Okada M, Nishida T, Hirai Y, Sat0 T, Fuse A, Miura Y: Growth
FH, Atkins HL, Hsu RY, Birket NC, Okino KH, Murdock DC,
and differentiation of a human megakaryohlastic cell line, CMK.
Jacobsen FW,Langley KE, Smith KA, Takeishi T, Cattanach BM,
Blood 74:42,1989
Galli SJ, Suggs SV: Stem cell factor is encoded at the SI locus of the
mouse and is the ligand for the c-kif tyrosine kinase receptor. Cell
10. Avraham HA, Vannier E, Chi SY, Dinarello CA, Groopman
63:213, 1990
JE: Cytokine gene expression and synthesis by human megakaryo2. Huang E, Nocka K, Beier DR, Chu T-Y, Buck J, Lahm H-W,
cytic cells. Int J Cell Cloning 1070,1992
Wellner D, Leder P, Besmer P: The hematopoietic growth factor
11. Toksoz D, Birkett N, Smith K, Zsebo K, Williams D A
KL is encoded at the SI locus and is the ligand of the c-kif receptor,
Analysis of activity of 2 distinct forms of the steel gene product
the gene product of the Wlocus. Cell 63:225,1990
(steel factor-SCF) expressed in deficient murine stromal cell lines.
3. Anderson DM, Lyman SD, Baird A, Wignall JM, Eisenman J,
Blood 78:160a, 1991 (abstr)
Rauch C, March CJ, Boswell HS, Gimpel SP, Cosman D, Williams
12. Broudy VC, Lin N, Zsebo KM, Birkett NC, Smith KA,
DE: Molecular cloning of mast cell growth factor, a hematopoietin
Bernstein ID, Papayannopoulou T Isolation and characterization
that is active in both membrane bound and soluble forms. Cell
of a monoclonal antibody that recognizes the human c-kit receptor.
63:235,1990
Blood 79:338,1992
4. Avraham H, Vannier E, Cowley S , Jiang S, Chi S, Dinarello
13. Springer TA Adhesion receptors of the immune system.
CA, Zsebo KM, Groopman JE: Effects of stem cell factor, c-kif
Nature 346:425, 1990
ligand, on human megakaryocytic cells. Blood 79:365,1992
14. Dexter TM, Moore MAS: In vitro duplication and cure of
5. Briddell PA, Bruno E, Cooper RJ, Brandt JE, Hoffman R:
hematopoietic
defects in genetically anemic mice. Nature 269:412,
Effect of c-kit ligand on in vitro human megakaryocytopoiesis.
1977
Blood 78:2854,1991
15. Hoffmann R: Regulation of megakaryocytopoiesis. Blood
6. Flanagan JG, Chan D, Leder P: Transmembrane form of the
74:1196, 1989
kit ligand growth factor can be regulated by alternative splicing and
16. Miyazawa K, Henrie PC, Mantel C, Wood K, Ashman LK,
is deleted in the SI mutant. Cell 64:1025, 1991
Broxmeyer HE: Comparative analysis of signaling pathways be7. Kaneko Y, Takenawa J, Yoshida 0, Fujita K, Sugimoto K,
tween mast cell growth factor (c-kit ligand) and granulocyteNakayama H, Fujita J: Adhesion of mouse mast cells to fibroblasts:
macrophage colony-stimulating factor in a human factor-depenAdverse effectsof steel (SI) mutation. J Cell Physiol 147:224, 1991
dent myeloid cell line involves phosphorylation of Raf-1, GTPase8. Tanaka H, Ishida Y, Kaneko T, Matsumoto N: Isolation of
activating protein and mitogen activated protein kinase. Exp
human megakaryocytes by immunomagnetic beads. Br J Haematol
Hematoll9:1110, 1991
73:18, 1989
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
1992 80: 1679-1684
Interaction of human bone marrow fibroblasts with megakaryocytes:
role of the c-kit ligand
H Avraham, DT Scadden, S Chi, VC Broudy, KM Zsebo and JE Groopman
Updated information and services can be found at:
http://www.bloodjournal.org/content/80/7/1679.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.