Basic fibroblast growth factor mediates its effects on committed

From www.bloodjournal.org by guest on October 21, 2014. For personal use only.
1995 86: 2123-2129
Basic fibroblast growth factor mediates its effects on committed
myeloid progenitors by direct action and has no effect on
hematopoietic stem cells
AC Berardi, A Wang, J Abraham and DT Scadden
Updated information and services can be found at:
http://www.bloodjournal.org/content/86/6/2123.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.
From www.bloodjournal.org by guest on October 21, 2014. For personal use only.
Basic Fibroblast Growth Factor Mediates Its Effects on Committed Myeloid
Progenitors by Direct Action and Has No Effect on Hematopoietic Stem Cells
By Anna C. Berardi, Anlai Wang, Judith Abraham, and David T. Scadden
Basic fibroblast growth factor or fibroblast growth factor-2
(FGF) has been shown t o affect myeloid cell proliferation
and hypothesizedt o stimulate primitive hematopoietic Cells.
We sought t o evaluate the effect of FGF on hematopoietic
stem cells and t o determine if FGF mediated its effects on
progenitor cells directly or through the induction of other
cytokines. To address the direct effects of FGF, we investigated whether FGF induced production of interleukin-l/3 (ILIB), tumor necrosisfactor a,IL-6, granulocyte colony-stimulating factor, or granulocyte-macrophage
colony-stimulating
factor by two types of accessory cells, bone marrow (BM)
fibroblasts and macrophages. Wefurther evaluated whether
antibodies t o FGF-induced cytokines affected colony formation. To determine if FGF was capableof
stimulating
multipotent progenitors, we assessed the outputof different
colony typea after stimulation of BM mononuclearcells
(BMMC) or CD34+ 6°C
and compared the effects of FGF
with the stem cell active cytokine,kit ligand (KL). In addition,
a subset ofCD34+ BMMC with characteristics of hematopoietic stem cells was isolated by functional selection andtheir
response t o FGF was evaluated using proliferation, colonyforming, and single-cell polymerasechain reaction (PCR)
assays. We determined that FGF had a stimulatory effect on
the production of a single cytokine, 11-6, but that theeffects
of FGF on colony formation were not attributable t o that
induction. FGF was more restricted in its in vitro effects on
BM progenitors than KL was, having no effect on erythroid
colony formation. FGF did notstimulate stem cells andFGF
receptors were not detected on stem cells as evaluated by
single-cell reverse transcription PCR. In contrast, FGF receptor gene expressionwas detected in myeloid progenitor
populations. These data support a directly mediated effect
for FGF that appears to be restricted t o lineage-committed
myeloid progenitor cells. FGF does not appear t o modulate
the human hematopoietic stem cell.
0 1995 b y The American Society of Hematology.
B
richment strategy developed in our laboratory,I2determining
the expression of FGF receptors by single-cell reverse transcription-polymerase chain reaction (RT-PCR) as wellas
measuring the biologic response to FGF and other cytokines
using proliferation assays.
ASIC FIBROBLAST GROWTH factor (FGF) affects
the proliferation and function of a wide range of primary tissues, including blood-forming elements.’” The effects of FGF on hematopoiesis have been previously described and include enhancement of myeloid cell numbers
as reflected by increases in granulocyte and granulocytemacrophage colonies (colony-forming unit-granulocyte
[CFU-G] and colony-forming unit-granulocyte-macrophage
[CFU-GM]), increases in burst-forming units-erythroid
(BFU-E), and enhanced megakaryocyte ~roliferation.~.~
FGF
has also been noted to alter the activity of primitive human
hematopoietic progenitors” augmenting the effects of other
cytokines on granulocytehonocyte, erythroid, or mixed lineage colony formation and to increase spleen colony-forming
units (CFU-S) from mice.7 However, the studies to date indicate that there is little effect on hematopoiesis by FGF alone;
rather, this cytokine enhances the activity of other growth
factors on precursor cells. A similar potentiating function
has been noted for the kit ligand (JSL),which acts synergistically with FGF in enhancing colony formation.“
The goals of this study were to further define the mechanisms of FGF’s effects on hematopoiesis, specifically addressing three questions: ( 1 ) does FGF mediate its activity
directly or through the altered production of other cytokines
by accessory cells, ( 2 ) what is the effect of FGF relative
to KL on hematopoietic cells, and (3) is FGF capable of
stimulating the most primitive blood progenitors or only
those committed to the myeloid lineage of differentiation?
To address the first question, we assessed the production of
granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin-l@ (IL-l@), tumor necrosis factor a (TNFa),and
IL-6 by purified populations of human primary bone marrow
(BM) stromal fibroblasts and peripheral blood (PB) monocyte/macrophages in response to FGF. To address the second
question, the relative activity of FGF and KL on BM mononuclear cells (BMMC) was measured in standard colonyforming assays. The issue of activity of FGF on primitive
progenitors was evaluated using a functional stem cell enBlood, Vol 86, No 6 (September 15). 1995: pp 2123-2129
MATERIALS AND METHODS
Cells
BM stromal cell culture. Human BM was obtained by aspiration
from the iliac crest of normal donors who provided voluntary written
informed consent to a Deaconess Hospital Institution Review Board
approved protocol. The marrow was aspirated into preservative-free
heparin (Sigma, St Louis, MO) and separated by centrifugation
through Ficoll-Hypaque (Pharmacia, Piscataway, NJ) at 400g at
room temperature for 30 minutes. After two washes with sterile 1
X Iscove’smodified Dulbecco’s media (IMDM; GlBCO, Grand
Island, NY) with 20%fetal calf serum (FCS; Hyclone, Logan, UT),
penicillin/streptomycin (P/S), and L-glutamine, the cells were seeded
onto T-75 tissue culture flasks (Coming, Coming, 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 medium.
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 to exclude macrophage contamination, as previously reported.”
Macrophage isolation. PB mononuclear cells (PBMC) were iso-
From the Division of Hematology/Oncology, New England Deaconess Hospital, Harvard Medical School, Boston, MA: and Scios
Nova, Inc,Mountain View, CA.
Submitted February 7, 1995; accepted May 8, 1995.
Address reprint requests to David T. Scadden, MD, Division of
Hematology/Oncology, NewEngland Deaconess Hospital, 1 Deaconess Rd. Boston, MA 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 I 8 U.S.C. section 1734 solely to
indicare this fact.
0 I995 by The American Society of Hematology.
0006-4971/95/8606-0035$3.00/0
2123
From www.bloodjournal.org by guest on October 21, 2014. For personal use only.
2124
lated from heparinized blood samples using Ficoll-Hypaque (Pharmacia) at 400g at room temperature for 30 minutes. After two washes
withRPM1 with 10% FCS, P/S,and L-glutamine, the cells were
incubated in a tissue culture dish (Falcon Plastics, Cockeysville,
MD) at37°C in 5% CO, for monocyte (MO) separation. After 4
hours, the nonadherent cells were gently removed and the adherent
cells were refed with fresh medium containing 10 U/mL recombinant
human macrophage colony-stimulating factor (M-CSF; gift of Genetics Institute, Cambridge, MA). The cultures were refed with fresh
medium plus M-CSF every 3 days. In previous studies, greater than
95% of the resultant cells stained positively with a monoclonal antiM02 (CD14) antibody by flow cytometric analy~is.'~
For cytokine
stimulation assays, macrophage or fibroblast cultures were washed
and refed with serum-free, endotoxin-free medium supplemented
with cytokines or lipopolysaccharide (LPS; Sigma) as noted.
Isolation of CD34' enriched BM. BM was obtained as described
above. The cells were layered on Ficoll-Paque (1.077 g/mL; Pharmacia) and centrifuged at 850g at 22°C for 20 minutes. The lowdensity BMMC were collected from the interface, washed twice with
phosphate-buffered saline (PBS), resuspended in IMDM containing
15% FCS, and plated in 75-cm2 flasks (1 to 2 X lo7cells in 10 mL).
After an overnight incubation at 37°C and 5% CO2 in a humidified
cell culture incubator, the nonadherent cells were recovered. To
select for the CD34' phenotype, these nonadherent BMMC were
incubated with an anti-CD34 monoclonal antibody (Amac, Inc,
Westbrook, ME) at4°C for 30 minutes, washed, andmixed with
magnetic microspheres coated with antimouse IgG (Dynal Inc, Great
Neck, NY) at a 0.5 to 1 bead per cell ratio. Rosetted cells were
recovered using magnetic separation and then washed three times
with IMDM.
CD34' enriched BMMC (2
CD34' G, phasecellenrichment.
x 10' cells/mL) were incubated at 37°C in 5% CO, for 7 days with
2.0 mg/mL 5-fluorouracil (5-FU; Solo Pack, Elk Grove Village, L)
in IMDM medium supplemented with10% FCS, KL (stem cell
factor [SCF]; 100 ng/mL; R & D Systems, Minneapolis, MN), and
L-3 (100 ng/mL; Genzyme). The cell population that survived 5FU treatment was used in this study after three washes in IMDM.
Cell Proliferation Assays
Cells obtained by the above methods were washed twice with
PBS, counted, and distributed into 96-well plates in IMDM containing 10% FCS supplemented with cytokines as indicated: FGF
(10 ng/mL; Scios Nova, Mountain View, CA), leukemia inhibitory
factor (LIF; 10 ng/mL), IL-lp (400 pg/mL), IL-6 (50 ng/mL), IL11 (10 ng/mL), SCF (10 ng/mL), or PIXY (10 ng/mL; R & D
Systems). The cells were incubated at 37°C in 5% CO, with twice
weekly refeeding using growth medium supplemented with cytokines. After 3 weeks, the cells were counted and viability was determined by Trypan blue exclusion and transferred into methylcellulose-containing medium for colony-forming assays.
Methylcellulose Colony Formation Assays
Adherence-depleted low-density BMMC (5 X 104/mL),CD34'enriched BMMC (4 X 103/mL),or cytokine-stimulated G, cells were
suspended in 10 X 35 mm gridded culture dishes with 0.9% methylcellulose (Stem Cell Technology, Inc, Vancouver, British Columbia,
Canada), 30% FCS, lo-' m o m 2-mercaptoethanol, 2 mmoVL Lglutamine, 2 U/mL recombinant human erythropoietin, and 10 ng/
mL GM-CSF. KL, FGF, or specified antibodies were added for
specific experiments as indicated. The cells were plated in triplicate,
incubated in a 5% CO, environment at 37°C for 14 days, and scored
by phase microscopic morphology for multilineage colonies (colonyforming unit-granulocyte, erythrocyte, megakaryocyte, macrophage
[CFU-GEMM]), BFU-E, or granulocyte-macrophage colonies
BERARDI ET AL
Table 1. Cytokine Production
IL-6
Macrophages
Control
FGF (1 ng/mL)
FGF (10ng/mL)
FGF (100ng/mL)
LPS (20ng/rnL)
Stromal cells
Control
FGF (1 ng/mL)
FGF (IO ng/mL)
FGF (100ng/mL)
LPS (20ng/mL)
IL-lp
86
0
0
92
95
0
110
0
1141.48445
127
146
185
376
436
TNFa
GM-CSF
31
23
25
28
70
0
0
0
856
0
0
0
0
0
0
0
0
0
0
0
G-CSF
0
0
0
0
71
0
0
0
0
-
Values are in picograms per milliliter. Effect of FGF or controls on
cytokine production by the accessory cell types, monocytes/macrophages. or BM stromal fibroblasts as measured byELISA. Values
below the limits of sensitivity of the assays are indicated as 0. Lower
limits of sensitivity for the respective ELlSAs are as follows: IL-6, 0.35
pg/mL; IL-IP, 0.3 pg/mL;TNFa, 4.8 pg/mL;GM-CSF, 1.5 pg/mL; GCSF, 7.3 pg/mL.
(CFU-GM). The range of control colony production for different
donor marrows was as follows: for CFU-GM, 28 to 68 colonies/
assay from BMMC and 67 to 182 colonies/assay from CD34' cells;
for BFU-E, 18 to 25 burstdassay from BMMC and 28 to 52 bursts/
assay from CD34+ cells.
Single-cell PCR
Using a Becton Dickinson Cell Deposition Unit instrument (Becton Dickinson, Fullerton, CA), single cells were plated in a 96-well
plate directly into 4 pL lysis buffer [50 mmoVL Tris-HCI, pH 8.3;
75 mmoVL KCI; 3 mmoVL MgCI,; 2 pmoVL of each deoxyribonucleotide triphosphate (Pharmacia); 100 ng/mL (dT)24; 100 U/mL
Inhibit Ace (5'-3' Inc, Boulder, CO), 2,000 U/mL RNAguard (Pharmacia), and 0.5% NP-401 as previously described by Brady et al.'5
The samples were heated to 65°C for 1 minute, cooled to 22°C for
3 minutes, and put on ice. The resultant lysate was then subjected
to reverse transcription and PCR.
One hundred units of Moloney (GIBCO-BRL) and 2 U of avian
reverse transcriptase (Pmmega, Madison, WI) were added and the
samples were incubated at 37°C for 15 minutes. The reverse transcriptases were thereafter inactivated at 65°C for 10 minutes. Cellfree and reverse transcriptase-free samples were usedas negative
controls. Poly(A) tailing of the single-cell cDNA was performed in
200 mmoVL potassium cacodylate, 4 mmoVL CoCI,, 0.4 mmoVL
dithiothreitol (DTT), and 10 U of terminal transferase (Bcehringer
Mannheim, Indianapolis, IN). After the addition of 200 pmoVL
dATP, the samples were incubated at 37°C for 30 minutes. After
heat-inactivation of the enzyme, the cDNA was either amplified
immediately or stored at -80°C until use.
The tailed cDNA was added to a PCR buffer containing l 0 mmoll
L Tris-HCI, pH 8.3, 50 mmoVL KCl, 2.5 mmoVL MgCl,, 1 mmoll
L each of dNTP 0.05% Triton X-100, 5 pmoVL (dTL4 X primer,
and 5 U of Taq polymerase (Perkin-Elmer Cetus, Newton Centre,
MA). The sequence of the PCR primer (dT)% X was ATG TCG
TCC AGG CCG CTC TGG ACA AAA TAT GAA TTC dT(24).
Amplification was performed for 25 cycles (1 minute at94"C, 2
minutes at 42"C, and 6 minutes at 72°C plus 10 seconds of extension
per cycle). Thereafter, an additional 5 U ofTaq polymerase was
added, followed by another 25 cycles of amplification. After electrophoresis and transfer to a nylon membrane, the resultant blots were
From www.bloodjournal.org by guest on October 21, 2014. For personal use only.
EFFECTOFBFGF
ON STEM CELLS
2125
3
'I
X
I
1.8
rWn
4
Control
FGF 1 .Ong/ml
KL long/ml
FGF O.lng/ml
FGF IOnglml
1 CD34+
jY
C
1.6C................................................
Control
I
......
FGF 1 .ong/ml
KL 1 Ong/mlI
FGF O.lng/ml
FGF lOr@ml
................
......................
...............................
.........................................
F:!:, . :I I
c,,;;;
,
j;.
.......................................................
'i ii -1 ":
.-..l; ,i.
..
control
FGF O.lng/ml
FGF 1Ongltnl
FGF 1 .*/m1
FGF O.lng/ml
FGF lOnglml
I
Fig 1. Effect of FGF or KL on colony formation by low-density BMMC (CD349 in the presence of GM-CSF and Epo. Mean fold change in
CFU-GM or BFU-E from pooled experiments is shown; error bars indicate standard deviation of data points obtained in duplicate or triplicate
of experiments repeated in duplicate or triplicate.
hybridizedwith"P-radiolabeledprobesfrom
cytokine receptor
cDNAs, including extreme 3' sequences.
Oligonucleotide probesfromthecDNA
sequence ofhuman
FGFR-I (ACACGCCCTCCCCAGACTCCACCGTCAGCTGTAA),
the human FGFR-2 (AGGCAGCACAGCAGACTAGTTAATCTATTGCTTG), humanFGFR-3 (TCGACC7TGAGCAGCCCTCCCTGCTGCTGGTGCA), or humanFGFR-4(GCCTGCCGAAAACAGGAGCAAATGGCGmATA) were also radiolabeledand
used as probes."-*'
cDNA probes were labeled with [a-3*P1-dCTP (DuPontNEN Research Products, Boston, MA) using random primed DNA labeling
(Stratagene, La Jolla, CA). Oligonucleotide probes were end-labeled
using [y-7'P]-ATP (DuPont NEN Research Products) and T, polynucleotide kinase (New England Biolabs, Beverly, MA). After prehybridization, the membranes were hybridized (SO% formamide, 6 x
SSPE, SX Denhardt's solution, 0.2% sodium dodecyl sulfate [SDS],
and SO pg/mL denaturedssDNA for cDNAprobes, or 20%formamide, 6X SSPE, 2X Denhardt's solution, and 1 0 0 yg/mL ssDNA
for oligonucleotide probes) at 42°C for 16 to 18 hours. The memwith medium
stringency
( I X for IS
branes
were
washed
minutes with 5X SSC, 0.58 SDS atroom temperature, 2 x for IS
minutes with 1 X SSC, 0.5% SDS at 37°C and 2 minutes with 0 . 2 ~
SSC, 0.5% SDS at 40°C) and exposed to x-ray film at -80°C for 4
hoursto I week.
RESULTS
Cytokine Stimulation Assays
Monocyte/macrophage or BM stromal fibroblast preparations were generated and exposed to endotoxin-free, serumfree medium containing FGF at concentrations of 0, I , IO,
or 1 0 0 ng/mL or LPS at 20 ng/mL for 24 hours before
harvesting conditioned media for cytokine analysis. Although LPS induced generally robust cytokine responses in
macrophage cultures, only IL-6 was increased above baseline after FGF stimulation (Table 1). Stromal fibroblasts produced minimal amounts of these cytokines in the supernatant
(as has been reported by others),2s.2hexcept IL-6, which increased approximately threefold in the setting of FGF stimulation. These data are consistent with the conclusion that
FGF does not induce a broad range of cytokine output by
accessory cells. Rather, FGF effects appear to be relatively
From www.bloodjournal.org by guest on October 21, 2014. For personal use only.
BERARDI ET AL
2126
m
conbol
d
FGFO.lnghnl
FGFO.lnghnl
FGF 1
.tMghnl
FGF 1
mghnl
d n
FGF-1
/
g h FGF
C l
FGF+mS-I1B .03ug
FGF + anti-FGF 1:500
FGF+~-IL6.14ug
FGF + anti-FGF 1:100
FGF+cont Ab -03ug
FGF+contAb1:500
FGF+cont Ab .14ug
FGF+ ccmt Ab 1:lOO
50
2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 1 0 0
60
70
80
90 100
110
120
1
K)
CFU-GM
CFU-GM
Control
Control
FGFO.lnglinl
FGFO-lnghnl
FGF 1
mglinl
.tMghnl
FGF 1
FGFlclhgmj
FGFlOnghd
FGF+anlS-ILB .03ug
FGF + anti-FGF 1:500
FGF+&-IL6.14ug
FGF + &FGF 1:100
-.
FGF+ant Ab .03ug
FGF+contAb1:500
FGF+cont Ab .Mug
FGF+ ccmt Ab 1:lOO
o
lo 120 140 160
180
200 220 240 260
CFU-GM
CFU-GM
Fig 2. Effect of FGF on colony formation is inhibited by anti-FGF antibody, but not by anti-lL-6 antibody. Antibodies at the indicated
quantities per milliliter or dilutions were coincubated with BMMC or CD34' cells and 10 ng/mL FGF. The mean and standard deviation are
indicated. * P < .05 when compared with control. * * P < .05 when compared with 10 ng/mL FGF using the Student's t-test analysis. The IC,
of the anti-IL-6 antibody is 0.07 pg/mL.
restricted and, among the cytokines tested. limited to increases in IL-6.
Relative Eflect of FGF Versus K L on Colon>l Formation
b y Hematopoietic Progenitor Cells
Low-density, adherence-depleted BMMC were plated in
standard methylcellulose colony assays in the presence of
FGF alone, GM-CSF, and erythropoietin (Epo) or combinations of FGF and GM-CSF + Epo. FGF induced no colony
formation when used by itself, but enhanced the CFU-GM
by 35% f 7% ( P < .OS) and 78% f 16% ( P < .OS) when
used at 1 .and 10 ng/mL, respectively, in conjunction with
GM-CSF and Epo (Fig 1). These increases were similar in
magnitude to what was observed with IO ng/mL KL (64%
t 21 %). However, unlike KL, there was no enhancement
in the production of BFU-E withFGF. Colony size was
approximately equivalent when evaluating CFU-GM or
CFU-mix in the presence of FGF or KL, whereas BFU-E
size was increased by KL alone.
Similar effects were noted when using CD34' cells. The
CFU-GM increased 35% t 10% ( P < .OS) and 54% t 13%
( P < .OS) when FGF at 1 and IO ng/mL, respectively, was
added to cultures containing GM-CSF and Epo (Fig I). Neither BFU-E nor CFU-mix was affected by FGF. BFU-E were
dramatically increased (3.2-fold) by KL treatment.
To assess whether FGF mediated its effect on colony formation via induction of IL-6, we usedneutralizing antibodies
to IL-6 (R & D Systems) and to FGF (kind gift of Dr Michael
Klagsbrun, Children's Hospital, Harvard Medical School,
Boston, MA) in methylcellulose assays (Fig 2 ) . The enhancement of formation of CFU-GM was unaffected by the presence of antibody to IL-6 but was reduced to baseline by antiFGF antibody. These results were consistent when either
BMMC or the CD34' population was evaluated.
effect of FGF on Hematopoietic Stem Cells
Quiescent BMMC with high long-term culture-initiating
cell (LTC-IC) ability were isolated and analyzed for both a
proliferative and differentiative (as reflected by the acquisition of methylcellulose colony-forming capacity) response to
FGF alone or in combination with other cytokines. Cytokine
combinations were compared with a stromal feeder layer
From www.bloodjournal.org by guest on October 21, 2014. For personal use only.
EFFECTOFBFGF
2127
ON STEM CELLS
tive effects of FGF in any of the suspension cultures tested.
No recombinant cytokine alone or cytokine combination
tested was capable of recapitulating the effect of BM stroma
on the cells.
Table 2. G,, Cell Proliferation Assay
Control
FGF
FGF + LIF
FGF + LIF + IL-l
FGF + LIF IL-l + PIXY
FGF LIF + I L - l + PIXY +
FGF + LIF + IL-l + PIXY +
FGF + LIF + IL-l + PIXY +
FGF + LIF + IL-1 + PIXY +
FGF + IL-l PIXY + IL-7
Stromal cocultivation
+
Expression
Cells
+
+
Single cells isolated using the Becton Dickinson Cell Deposition Unit (which has a plating accuracy of 0.1%) were
used to generate cDNA, as previously described.”.’” The
cells used were quiescent, functionally selected CD34’ cells
(G,cells) or cells selected by the immunophenotype shown
(Fig 3) derived from either human BM or PB. Five cells of
each representative type were used to minimize the
likelihood of a single contaminating cell providing misleading
information. The resultant cDNAs were ethidium bromidestained and photographed and, after Southern transfer, blotted for the presence of FGF receptor or other cytokine receptor mRNA transcripts. Oligonucleotides derived from the 3’
untranslated region of the FGFrI Vg), FGFr2 (hek).FGFr3,
or FGFr4 cDNA sequences were used as radiolabeled probes.
These FGF receptor probes were compared with stage specific receptors such as the Epo (Epor) or thrombopoietin (cmpl) receptors or more ubiquitously expressed receptors
such as the gp130 subunit of the IL-6, IL-I I , LIF, and ciliary
neurotropic factor (CTNF) receptors.
The receptor known to be specifically expressed at a later
stage, Epor,wasrestricted
in its detection of messageto
IL-7
IL-7 + IL-6
IL-7 + IL-6 + KL
IL-7 + IL-6 + KL + IL-l1
+ IL-6 + KL + I L - l 1
Quiescent stem cells are unresponsive to FGF alone or in combination with other cytokines. The indicated cytokines at concentrations
indicated in Materials and Methods or stromal control wells were
incubated with functionally selected CD34+ cells and scored qualitatively for cell proliferation at 3 weeks based on phase microscopic
assessment. Cells were subsequently plated in methylcellulose and
scored for colony formation after 2 weeks.
that is capable of inducing cobblestone formation and CFC
detectable by week 3.’’ Because of the difficulty in accurately quantitating cell number in the cocultivation system,
we qualitatively scored cell or colony numbers using the
stroma cocultivation assay as an arbitrary “+ + +” standard. Table 2 indicates the lack of proliferative or differentia-
Go
FGFrl
FGFr2
FGFr3
Fig 3. Quiescent stem cells do not express FGF
receptors in contrast t o myeloid committed progenitor cells. Single-cell RT-PCR products were agarose
gel electrophoresed and stained with ethidium bromide (lower panel of each set) before transfer t o a
nylon filter, hybridization
t o the indicated’*P-labeled
probe, and autoradiography (upper panel of each
set). Each blot shown is unique and has not been
previously probed. Cells used in the analysis were
derived in at least two independent selection experiments.
FGFr4
sf FGF Receptors on Hematopoietic Stem
CD34’33-
CD34+33+
CD19+
CD3+
CDllb’
From www.bloodjournal.org by guest on October 21, 2014. For personal use only.
2128
BERARDI ET AL
cells of erythroid commitment, as previously reported (data
not shown).12 Incontrast, gp130, thought to be expressed on
a wide range of hematopoietic cells, including stem cells,
was detectable (data not shown) in cells representing stem
cells, primitive progenitors (CD34+, CD33-), primitive myeloid progenitors (CD34+, CD33+), mature monocyte/macrophages (CD1 lb+), mature T cells (CD3+), and mature B
cells (CD19+). Having confirmed the relative specificity of
the detection process, we probed with the oligonucleotide
probes specific for each of the four FGFr (Fig 3). Message
was detected in mature CDllb' cells and more immature
CD34+, CD33+ and most CD34+, CD33- progenitor cells.
Although G, cells are CD34+, CD33-, they represent a small
subset of those cells because they are approximately 0.1%
of CD34+ BMMC, whereas CD33- cells represent approximately 10% of CD34+ cells. The disparate expression data
are therefore not inconsistent and support the conclusion that
FGF does not affect quiescent, multipotent hematopoietic
precursor cells with in vitro characteristics of stem cells. In
contrast, kit message was detected in a range of primitive
cells including the quiescent, LTC-IC-rich G, fraction, as
previously reported (data not shown)."
tabolite treatment. We have exploited this phenomenon to
drive those CD34' cells thatrespond to theearly acting
growth factors IL-3 and KL to cell death. Theremaining
viable cells are immunophenotypically CD34', C D 3 3 ~ ~ .
CD38-, HLA-DR-, and c-kit' with a high capacity for LTCIC (76% 2 S%) and the ability to form both lymphoid and
myeloid progeny." These features support, although they do
not conclusively prove, the stem cell nature of the resultant
cell population.
We used the cells isolated by this procedure to evaluate
functionally and molecularly the role of FGF on stem cells.
When these cells are cocultivated with stroma they acquire
the ability to generate CFC after an incubation of approximately 3 weeks. In liquid culture, FGF was unable to induce
either proliferation or the CFC-producing phenotype in these
cells. This lack of effect was observed whether FGF was
used alone or in combination with other early acting hematopoietic growth factors. Further, the receptor cDNAs for the
four known FGF receptors could not bedetected in the qualitative PCR-based system. We would therefore conclude that,
although FGF can modulate cells already committed to specific stages of myeloid differentiation, it has no apparent role
in regulating the biology of the hematopoietic stem cell.
DISCUSSION
FGF is a cytokine with pleiotropic effects including effects
on multiple cell types comprising hematopoietic tissue. It is
known to induce proliferative effects on BM stromal cells
as well as altering the growth, adhesion, and cytokine production properties of blood elements. Whereas FGF has been
noted to affect myeloid colony production, it has not been
determined whether FGF generates these effects directly or
via alterations in accessory cells. Neither has it been defined
whether FGF exerts effects on the most primitive hematopoietic stem cells.
The studies reported here lead to the conclusion that FGF
induces an alteration in production of a single cytokine, IL6, by macrophages and BM fibroblasts. Other cytokines were
unaffected and inhibition of IL-6 by neutralizing antibody
had no impact on the hematopoietic effects seen with FGF.
Coupled with the data demonstrating message for FGF receptors in myeloid precursors, these data suggest that FGF
directly mediates its activity on myeloid progenitor cells.
The data comparing KL with FGF suggest that, although
the effects on CFU-GM are similar, there is a dissimilarity
in effects on the erythroid lineage. FGF, which is known to
also have activity on megakaryocytic cells, would appear to
be more lineage restricted in its effects than is KL. Unlike
KL, which augments proliferation of all descendants of the
myeloid lineage, FGF does not alter erythroid colony number
or size. These data imply that FGF may notaffect hematopoietic stem cells, as has been previously suggested.'."'
Assessing the role of FGF on primitive hematopoietic cells
has been problematic due to the difficulty of isolating these
cells. The stem cell isolation procedure we developed is
based on the observation that cells acquire responsiveness
to growth factors inan orderly fashion as theyproceed
through the process of hematopoietic differentiation. This
growth factor responsiveness is manifest as a proliferative
response that enhances the sensitivity of the cells to antime-
REFERENCES
1. Gospodarowicz D, Ferrara N, Schweigerer L, Neufeld G:
Structural characterization and biological functions of fibroblast
growth factor. Endocr Rev 8:95, 1987
2. Klagsbrun M: The fibroblast growth factor family: Structural
and biological properties. Prog Growth Factor Res I :207, l989
3. Mason IJ: The ins and outs of fibroblast growth factors. Cell
78547, 1994
4. Oliver LJ, Rifkin DB, Gabrilove J, Hannocks MJ, Wilson EL:
Long-term culture of human bone marrow stromal cells in the presence of basic fibroblast growth factor. Growth Factors 3:231, 1990
5. Wilson EL, Rifkin DE, Kelly F, Hannocks MJ, Gabrilove J :
Basic fibroblast growth factor stimulates myelopoiesis in long-term
human bone marrow cultures. Blood 77:954, 1991
6. Gabrilove J, Wong G, Bollenbacher E, White K,Kojima S,
Wilson EL:Basicfibroblast
growth factor counteracts the suppressive effect of transforming growth factor-p1 on human myeloid
progenitor cells. Blood 81:909, 1993
7. Gallicchio VS, Hughes NK, Hulette BC, DellaPuca R, Noblitt
L: Basic fibroblast growth factor (b-FGF) induces early (CFU-S)
and late-stage hematopoietic progenitor cell colony formation (CFUgm, CFU-meg, and BFU-e) by synergizing with GM-CSF, MegCSF and erythropoietin, and is a radioprotective agent in vitro. Int
J Cell Cloning 9:220, 1991
8. Bruno E, Cooper RJ,WilsonEL, Gabrilove JL, Hoffman R :
Basic fibroblast growth factor promotes the proliferation of human
megakaryocyte progenitor cells. Blood 82:430, 1993
9. Avraham H, Banu N, Scadden DT,Abraham J, Groopman
J E Modulation of megakaryocytopoiesis by human basic fibroblast
growth factor. Blood 83:2126, 1994
IO. Gabbianelli M, Sargiacomo M, Pelosi E, Testa U, Isacchi G,
Peschle C: "Pure"human hematopoietic progenitors: Permissive
action of basic fibroblast growth factor. Science 249:1561, 1990
I I . Gabrilove JL, White K, Rahrnan Z, Wilson EL: Stem cell
factor andbasic fibroblast growth factor are synergistic in augmenting committed myeloid progenitor cell growth. Blood 83:907,
1994
12. Berardi AC,Wang A, Levine JD, Lopez P. Scadden DT:
From www.bloodjournal.org by guest on October 21, 2014. For personal use only.
EFFECT OF eFGF
ON STEM CELLS
Functional isolation and characterization of human hematopoietic
stem cells. Science 267: 104, 1995
13. Scadden DT, Zeira M, Woon A, WangZ, Schieve L, Ikeuchi
K, Lim B, Groopman JE: Human immunodeficiency virus infection
of human bone marrow stromal fibroblasts. Blood 76317, 1990
14. Molina JM, Schindler R, Femani R, Sakaguchi M, Vannier
Production of cytokines by periphE, Dinarello CA, Groopman E.
eral blood monocytedmrophages infected with human immunodeficiency virus type 1 ( W - l ) . J Infect Dis 161:888, 1990
15. Brady G, Mary B, Iscove NN: Representative in vitro cDNA
amplificationfrom individual hematopoietic cells andcolonies.
Methods Mol Cell Biol 2: 17, 1990
16. Dionne CA, Crumley G, Bellot F, Kaplow J M , Searfoss G,
Ruta M, Burgess W H , Jaye M, Schlessinger J: Cloning and expression of two distinct high-affinity receptorscross-reacting with acidic
and basic fibroblast growth factors. EMBO J 9:2685, 1990
17.RutaM,HowkR,RiccaG,Drohan
W, ZabelshanskyM,
Laureys G, Barton DE,Francke U, Schlessinger J, Givol D A novel
protein tyrosine kinase gene whose expression is modulated during
endothelial cell differentiation. Oncogene 3:9, 1988
18. Houssaint E, Blanquet PR, Champion-AmaudP, Gesnel MC,
Tomglia A, Courtois Y, Breathnach R: Related fibroblast growth
factor receptor genes exist in the human genome. Proc Natl Acad
Sci USA 87:8180, 1990
19. Keegan K, Johnson DE, Williams LT, HaymanMJ: Isolation
ofan additional member of the fibroblast growth factor receptor
family, FGFR-3. Proc Natl Acad Sci USA 88:1095, 1991
2129
20. Partanen J, Mtikela TP, Eerola E, Korhonen J, Hirvonen H,
Claesson-Welsh L, Alitalo K FGFR-4,anovel acidic fibroblast
growth factor receptor with a distinct expression pattern. EMBO J
101347, 1991
21. Wennstrom S, Sandstrom C, Claesson-WelshL: cDNA cloning and expressionof a humanFGF receptor which binds acidic and
basic FGF. Growth Factors 4197, 1991
22. Hou J, Kan M, McKeehan K, McBrideG,Adams P, McKeehan WL: Fibroblast growth factor receptors from liver vary in three
structural domains. Science 251:665, 1991
23. Werner S, Duan D-S, De Vries C, Peters KG, Johnson DE,
Williams LT: Differential splicing in the extracellular region of
fibroblast growth factor receptor 1 generates receptor variants
with different ligand-binding specificities. Mol Cell Biol 12:82,
1992
24. Wang SY,Su CY, Hsu ML, Chen LY, Tzeng CH, Ho C K
Effect of lipopolysaccharide on the productionof colony-stimulating
factors by the stromal cells in long-term bone marrow culture. Exp
Hematol19:122, 1991
25. Guba SC, Sartor CI, Gottschalk LR, Ye-Hu J, Mulligan T,
Emerson SG: Bone marrow stromal fibroblasts secrete interleukin6 and granulocyte-macrophage colony-stimulating factor in the absence of inflammatorystimulation:Demonstration by serum-free
bioassay,enzyme-linkedimmunosorbentassay,andreverse
transcriptase polymerase chain reaction. Blood 80:1190, 1992
26. Huang S, Terstappen L W " : Lymphoid and myeloiddifferentiation of single human CD34+,HLA-DR+, CD38- hematopoietic
stem cells. Blood 83:1515, 1994