From www.bloodjournal.org by guest on July 12, 2017. For personal use only.
Synergistic Actions of Stem Cell Factor and Other Burst-Promoting
Activities on Proliferation of CD34' Highly Purified Blood Progenitors
Expressing HLA-DR or Different Levels of c-kit Protein
By Yoshiaki Sonoda, Hideaki Sakabe, Yoshikazu Ohmisono, Shigeatsu Tanimukai, Shouhei Yokota,
Shuichi Nakagawa, Steven C. Clark, and Tatsuo Abe
used as the target, SCF showed a much strongerBPA. Also,
We studied the synergistic effects ofstem cell factor (SCF)
a distinct additive effect between SCF and IL-3 or GM-CSF
(BPAs) such asinterleuand other burst-promoting activities
kin-3 (lL-3), granulocyte-macrophage colony-stimulating fac-on erythrocyte-containingmixed colony formation was observed. On the other hand, when CD34+c-kitioW cellswere
tor (GM-CSF), or IL-9 on proliferation of human peripheral
used as the target, SCF,IL-3, andGM-CSFcouldexpress
blood-derivedhighly purified progenitors. SCF,IL-3, GMBPA. In contrast, IL-9 alonefailed to support erythroid-burst
CSF, and IL-9showed significant BPA when CD34+HLA-DRf
formation. BecauseCD34+c-kithfghcells weakly expressed
cells were used as the target population. IL-3 exerted the
CD34 antigen, these cells appearedto be more mature promost potent BPA, andGM-CSFsupportedapproximately
genitors than CD34+c-kit'" cells. These results suggest
that
40% to 70% of the erythroid burst-forming units that are
IL-9 acts onmore mature progenitors than those of SCF, ILresponsive to IL-3. SCF and IL-9 showed much weaker BPA
3, or GM-CSF andthat the primary target of SCF is multipothan that of IL-3 orGM-CSF. Combinations of IL-3
with other
BPAs did not show synergistic actions supporting erythroid- tential progenitors at the very early stage of development.
0 1994 by The American Societyof Hematology.
burst formation. However, GM-CSF showed a significant
additive effect with IL-9 or SCF. When CD34+c-kithigh cells
were
0
UR UNDERSTANDING OF the mechanisms of control of erythropoiesis isnot fully established. Two
types of erythroid progenitors have been described, the erythroid burst-forming unit (BFU-E) and the colony-forming
unit-erythroid (CFU-E).' Early erythroid progenitors known
as BFU-E require the presence of burst-promoting activity
(BPA) for their proliferation, in addition to erythropoietin
(Epo)? In earlier reports:A we observed that interleukin-3
(IL-3) and granulocyte/macrophage colony-stimulating factor (GM-CSF) have significant BPA, and IL-3 shows a more
potent BPA than GM-CSF. In addition, we reported that
human IL-9, which was originally identified as a growth
factor for the human megakaryoblastic cell line M07E: selectively supports erythroid-burst formation! We also suggested that IL-9 possibly supports the formation of a subpopulation of BFU-E that are responsive to IL-3.6 Our results
were consistent with those of a number of other report^.^"^
Recently, the ligand for c-kit proto-oncogene product was
identified as stem cell factor (SCF), which is also known as
steel factor, mast cell growth factor, or kit ligand.I5-I7Both
membrane-bound and soluble forms of human SCF have
been identified.16 Many studies indicated that the c-kit molecule is expressed on primitive progenitor cells and plays
an essential role in murine andhuman hematopoiesis.'*-24
Furthermore, several investigators have reported that the soluble form of SCF acts as a BPA in human culture systems,20,25-28
To clarify the interactions of BPAs that regulate proliferation and differentiation of early erythroid progenitor cells,
the effects of four known BPAs including IL-3, GM-CSF,
IL-9, and SCF were compared using fluorescence-activated
cell sorter (FACS)-sorted highly purified progenitors derived
from peripheral blood (PB) cells. IL-3, GM-CSF, IL-9, and
SCF exerted significant BPA on erythroid-burst formation
by PB-derived CD34+HLA-DR+cells. The effects of these
four BPAs using PB-derived highly purified CD34+ cell populations that expressed different levels of c-kit protein were
then studied in serum-free culture. When a CD34+c-kithiSh
population was used as the target, these four BPAs showed
significant BPA, and a significant additive or synergistic
Blood, Vol 84, No 12 (December 15). 1994: pp 4099-4106
action between SCF and IL-3, GM-CSF, or IL-9 was observed. In contrast, IL-9 alone failed to support erythroidburst formation by CD34+c-kitl0"cells. These results lead
us to propose that IL-9 supports more mature early erythroid
progenitors as compared with that of those cells supported
by IL-3, GM-CSF, or SCF and that SCF supports primitive
multipotential progenitor cells.
MATERIALSANDMETHODS
Recombinant factors and neutralizing antisera. Purified bacterially derived recombinant human IL-3, GM-CSF and granulocyteCSF (G-CSF) were generous gifts from Grin Brewery Company
(Tokyo, Japan). Purified Chinese hamster ovary (CH0)-cell-derived
recombinant human Epo was also supplied by Kirin Brewery. Purified CHO-cell-derived recombinant human IL-9 was supplied by
Genetics Institute (Cambridge, MA) and had a specific activity of
6.48 X IO6 U/mg protein. Purified Eschericheria coli-derived recombinant human SCF was kindly provided by Amgen (Thousand Oaks,
CA).'h
Rabbit polyclonal antihuman IL-3 (alL-3) and antihuman GMCSF (aGM-CSF) sera were kindly provided by Kirin Brewery. A
1:1,000 dilution of each antisera neutralized 10 ng/mL of a relevant
From the Department of Hygiene, the Department of Pediatrics,
the Second Department of Surgery, the Third Department of Medicine, and the Department of Urology, Kyoto Prefectural University
of Medicine, Kyoto, Japan; and the Genetics Institute, Inc, Cambridge, MA.
Submitted March 15, 1994; accepted August 18, 1994.
Supported in part b.y a Grant-in-Aidfor ScientijcResearch B(No.
05454334) fromthe Ministry of Education, Science and Culture of
Japan and the Japunese Foundation for Multidisciplinary Treatment
of Cancer for 1992.
Address reprint requests to Yoshiaki Sonoda, MD, Department of
Hygiene, Kyoto Prefectural University of Medicine, KawaramachiHirokoji, Kamigyoku, Kyoto 602, Japan.
The publication costsof this article were defrayedin part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section I734 solely to
indicate this fact.
Q 1994 by The American Society of Hematology.
0006-4971/94/8412-0025$3.00/0
4099
From www.bloodjournal.org by guest on July 12, 2017. For personal use only.
4100
factor in titration experiments (data not shown). Rabbit polyclonal
neutralizing antihuman E - 9 (aIL-9) antiserum was kindly provided
by Dr Edward M. Alderman (Genetics Institute, Cambridge, MA).
A 1 :200 dilution of this antisera neutralized 5 U/mL of IL-9 protein
in the M07E proliferation assay.
Cell preparation. After informed consent was obtained, PB mononuclear cells were collected from patients with testicular tumors
by leukapheresis using Fenwall CS-3000 (Fenwall Laboratories, Inc.
Deerfield, IL). In these patients, hematopoietic progenitors were mobilized into PB by a combination of chemotherapy and G-CSF, as
we reported elsewhere.2yCells in the collection bag were centrifuged
at l5Og for 15 minutes, and buffy-coat cells were transferred to the
second bag. These cells were diluted with CaZ+-and Mg*+-free
phosphate-buffered saline (PBS-) and layered on a 40%/60%Percoll
gradient in 15mL polystyrene tubes. After centrifugation at 4001:
for 30 minutes, interphase cells were obtained with Pasteur pipette.
A part of the cells were used for the present studies, and the remaining cells were subsequently frozen using a rate-controlled programmed freezer for autotransplantation as reported elsewhere."'
Cells separated by Percoll gradient contained on average 6.4% ?
3.6% CD34+ cells according to FACS analysis (n = 9). Forthe
present studies, cells were washed twice with a-medium containing
5% fetal calf serum (FCS; Flow Laboratories, Inc), and nonadherent
cells were recovered by overnight adherence to plastic dishes. The
mononuclear nonadherent cell fractions were further enriched by
negative selection using soybean agglutinin (SBA) Micro-CELLector flasks (Applied Immune Science Inc, Menlo Park, CA) following
the instructions of the manufacturer. For some of the experiments,
previously frozen mononuclear cells were also used after rapid thawing as
Pur$cation of hematopoietic progenitor cells by FACS. Highly
purified progenitor cells were separated from PB-derived SBA- cells
using a FACStar Plus (Becton Dickinson Immunocytometry Systems, San Jose, CA) equipped with an argon laser tuned at 488 nm,
as reported elsewhere." First, PB-derived SBA- cells were washed
twice with PBS- containing 2% FCS (staining medium) and were
passed through a stainless steel mesh. Cells were pelleted before
staining with monoclonal antibodies (MoAbs). MoAbs used are as
follows: fluorescein isothiocyanate (FITC)-conjugated HPCA-2
(mouse IgG, CD34 MoAb; Becton Dickinson); phycoerythrin (PE)conjugated HLA-DR (mouse IgC,,; Becton Dickinson); purified antihuman c-kit MoAb (mouse IgM; Immunotech S.A., Marseille
Cedex, France); and goat antimouse IgMwithPE
(Immunotech
S.A.). For double staining, cells were incubated with 20 pL of
HPCA-2(FITC)/10h cells and 20 pL of HLA-DR(PE)/lOh cells for
30 minutes on ice and then were washed twice with staining medium.
Cells were also stained with 20 pL ofpurified antihuman c-kit
MoAb/5 x IO' cells for 30 minutes at room temperature. After
washing with staining medium, cells were incubated with 20 pL of
goat antimouse IgM with PE/5 X 10' cells for 30 minutes at room
temperature. These cells were washed twice with staining medium
and then stained with HPCA-2(FITC) as described. All stained cells
were kept on ice for cell sorting. For negative controls, unstained
cells and cells stained only with second antibody or with isotype
control IgG with FITC or PE were included.
Stained cells were sorted using a FACStar Plus single-laser flow
cytometry system. Sorting gates were established for both forward
scattering and side scattering. A dual-parameter dotogram displaying
FITC(CD34) and PE (HLA-DR or c-kit) fluorescence was then generated from gated events. Using these gated dotograms, CD34+HLADR' cells were sorted for clonal cultures. The sorting windows for
CD34' cells expressing different levels of c-kit protein were also
established as shown in Fig I . CD34+ cells were subdivided into
three fractions including c-kith'ghh,
c-kit'"", and c-kit- populations as
reported.32Data acquisition was performed using the FACStar Plus
SONODA ET AL
n
w
e
W
CD34 (FITC)
Fig 1. Expression of c-kit on PB-derivedCD34'cells is shown.
First, sortinggates were established forboth forward scattering and
side scattering. A dual-parameter dotogram displaying FITC CD341
and PE (c-kit1 fluorescence was then generated from gated events.
CD34' cells were subdivided into three fractions such as CD34'ckith'gh,CD34+c-kit",and
CD34+c-kit- cells.Each population contained 8.01%. 4.62%. and 10.93% of total gated cells, respectively.
C D 3 4 * ~ - k i tcells
~ ' ~ ~expressed low levels of CD34 antigen compared
with that of CD34+c-kit'OWor CD34+c-kit- cells.
Research Software. A minimum of 20,000 events was analyzed on
each sample. We isolated the CD34'HLA-DR+, the CD34+c-kithlKh,
and the CD34'c-kit'"" populations, respectively. The phenotypic purity of sorted cells determined by postsorting flow cytometric analysis exceeded 90%. After sorting, the recovered cells were washed
twice with &-medium and cultured as described below.
Clonal cell culture. Cultures were performed in 35-mm Lux
suspension culture dishes (no. 171099; Nunc Inc, Naperville, IL) as
we reported e l s e ~ h e r e . ~One
. ~ ~milliliter
"~~
of culture contained 200
to 500 sorted cells, I .2% of 1,500 centipoise methylcellulose
(Shinetsu Chemical, Tokyo, Japan), 30% FCS, 1% deionized fraction
V bovine serum albumin (Sigma Chemical CO, StLouis, MO), 5 X
10" m o m 2-mercaptoethanol (Sigma), and 1 or more recombinant
human CSFs. Final concentrations of each CSF were as follows: IL3, 10 ng/mL; GM-CSF, 10 ng/mL; C-CSF, 20 ng/mL; Epo, 2 U/
mL; and IL-9, 100 U/mL. These concentrations supported maximal
total colony formation in preliminary titration experiments (data not
shown). In some experiments, neutralizing concentrations of aIL-3.
aGM-CSF, and aIL-9 antisera were added to cultures. Cultures were
preincubated with these antisera for I hour atroom temperature
before beginning cultures.
Serum-free cultures were performed as reported
Briefly, 200 to 500 sorted cells were plated in 35-mm Lux suspension
culture dishes (Nunc) containing attenuated a - m e d i ~ m 1.2%
, ~ ~ methylcellulose I % globulin-free bovine serum albumin (Sigma) that had
been crystallized and deionized, 5 X IO-' molL 2-mercaptoethanol.
300 ,ug/mL fully iron-saturated human transferrin (95% pure:
Sigma), 5 X
molL sodium selenite (Sigma), I O pg/mL lecithin
(Sigma), 6 pg/mL cholesterol (Sigma), 1 pg/mL bovine pancreas
insulin (Sigma), and various growth factors.
Dishes were incubated at 37°C in a fully humidified atmosphere
flushed with a combination of 5% COz, 5% O,, and 90% NZ.On day
14 of incubation, all colonies were scored on an inverted microscope
From www.bloodjournal.org by guest on July 12, 2017. For personal use only.
SYNERGISTIC ACTIONS OF SCF AND OTHER BPAs
90
4101
1
1. Approximately 50% of cells in this population formed
colonies in the presence of a combination of IL-3, GM-CSF,
T
0
0.1
1
10
SCF (ng/ml>
+
Epo (2 U/ml)
IL-3
20 100
(10 ng/ml>
*
P < 0.01
Fig 2. Dose-responsa effects of SCFon erythroid-burstformation
by 500 PB-derived CD34+HlA-DR+cells in the presence of Epo. Over
10 ng/mL SCF significantly supported erythroid-burst formation ( P
< .01).
according
their
totypical
morphologic
appearance
reas
ported.3.4.6.%34
Colony types identified in situ were granulocyte/macrophage (GM), erythroid-burst (B), eosinophil (Eo), and erythrocytecontaining mixed (E-Mix) colonies.
Srarisrical analysis. The significance of differences in means
was determined using the two-tailed Student's t-test or Mann-Whitney rank-sum test.
G-CSF, and Epo. IL-3 showed the mostpotent BPA, whereas
GM-CSF supported approximately 70% of BFU-E that were
responsive to IL-3. The BPA of SCF was much weaker than
that of IL-3 and was almost identical with that of IL-9. IL3 and GM-CSF, but not IL-9, supported significant E-Mix
colony formation. On the other hand, SCF supported a small
number of E-Mix colony formation.
Effects of aIL-3, aGM-CSF, and aIL-9 antisera on erythroid-burst formation supported by SCF plus Epo. As
shown in Fig 2 and Table 1, SCF showed a significant BPA
on erythroid-burst formation by highly purified progenitor
cells at low cell density; however, 30% FCS was usedin
these cultures. Unknown factors or known BPAs such as IL3, GM-CSF, or IL-9 may have been present in our culture
system and may have affected the BPA of SCF. To rule out
such a possibility, the effects of neutralizing antisera against
these BPAs were investigated. The results are presented in
Table 2. A combination of neutralizing antisera including
aIL-3, aGM-CSF, and aIL-9 abrogated the BPA effect of the
relevant cytokine. However, SCF again showed a significant
BPA in the presence of a combination of aIL-3, aGM-CSF,
and aIL-9 antisera. These results showed that SCF has an
apparent BPA that is independent of IL-3, GM-CSF, and IL9 but do not rule out the possibility that unknown cytokines
might contribute to this activity.
Effects of single BPA or a combination of more than two
BPAs on erythroid-burst or E-Mix colony formation using
PB-derived CD34+HLA-DR+ cells. To clarify the interactions of four BPAs, the effects of single BPA or a combination of more than two BPAs on erythroid-burst or E-Mix
colony formation were studied. We cultured 500 PB-derived
CD34+HLA-DR+ cells and compared the colony-forming
ability of four BPAs. As shown in Fig 3, IL-3 again elicited
the most potent BPA, andGM-CSF supported approximately
70% of BFU-E that were responsive to IL-3. The plating
efficiency of this culture was approximately 50%. The BPA
of IL-9 or SCF was much weaker than that of IL-3 or GMCSF. IL-9 induced a slightly higher number of erythroid
bursts than that supported by SCF. Combinations of IL-3
RESULTS
Effects of various concentrations of SCF on erythroidburst formation. First, we examined the dose-response effects of SCF on erythroid-burst formation by 500 PB-derived
CD34+HLA-DR+ cells that represent a highly enriched population of progenitor^.^^ SCF alone did not support significant colony formation (data not shown). In the presence of
Epo, concentrations of SCF in excess of 10 ng/mL induced
significant erythroid-burst formation as shown in Fig 2.
However, the BPA of SCF was much weaker than that of
IL-3. Based on these data, we chose 20 ng/mL SCF as the
standard concentration in the following experiments.
Comparison of BPA of SCF with other known BPAs. The
BPA of SCF was then compared with other known BPAs
including IL-3, GM-CSF, and L 9 using 300 to 500 PBderived CD34+HLA-DR+ cells as the target population. The
results of one representative experiment are shown in Table
Table 1. Comparison of BPA of SCF With Other Known BPAs
Colony Types
Factors
EPO
SCF
+ Epo
IL-3+ EPO
+
GM-CSF EPO
IL-9+ EPO
CSFs'
None
B
Eo
E-Mix
121
0
0
421
14 2 5
142 1
321
62 2 12
321
14 2 2t
56 2 llt
42 5 It
1 4 2 3$
41 t 1 1
0
0
1252
10+5
0
120
721
321
0
1023
0
0
7f1
0
GM
Total
151
1922
91512
6924
1723
141219
321
Data represent the mean 2 SD of triplicate cultures containing 300
PB-derived CD34' HLA-DR+cells/dish. Abbreviations for colonytypes
are defined in Materials and Methods.
'CSFs contained IL-3,GM-CSF, G-CSF, and Epo.
t P < ,001,percentage of significant change from control.
P < .01,percentage of significant change from control.
*
From www.bloodjournal.org by guest on July 12, 2017. For personal use only.
SONODA ET AL
4102
Table 2. Effects of alL-3, aGM-CSF, and alL-9 Antisera on Ervthroid-Burst Formation SUDDOrted bv SCF Plus EDO
Colony Types
Factors
GM
EPO
SCF + Epo
SCF + Epo + antibodies*
SCF + Epo + control serumt
IL-3 + EPO
IL-3 + Epo antibodies
GM-CSF + EPO
GM-CSF Epo + antibodies
IL-9 + Epo
IL-9 + Epo + antibodies
3 2 1
6+2
5 2 1
7 2 3
9 5 2
3 2 1
13 2 3
5 t 2
4 2 1
21-2
+
+
B
2 t l*
9 2 l*
9 % l§
91-1
72 t 1
3 2 1
44 t 6
3 2 2
14 2 1
I t 1
Eo
E-Mix
0
0
I t 1
2 5 1
1+1
14 i 2
0
6 5 2
0
0
0
0
0
0
15 t 3
0
11 2 2
lrl
0
0
Total
5t2
162 1
16 t 2
17 2 2
110 t 3
523
73 2 3
9 2 3
18%1
4 2 2
Data represent the mean t SD of triplicate cultures containing 500 PB-derived CD34+ HLA-DR' cells/dish. Abbreviations for colony types are
defined in Materials and Methods.
Antibodies contained alL-3 (1:1,000), aGM-CSF 11:1,000). and alL-9 (1:50) antisera.
t (1:lOO) dilution of normal rabbit serum.
P < 0.01.
§ Statistically not significant.
*
with GM-CSF, SCF, or IL-9 didnotshow synergistic actions. In contrast, when GM-CSF was combined with SCF
or IL-9, the number of BFLJ-E increased significantly ( P
< .05 and P < .01). The number of E-Mix also increased,
although this difference was not statistically significant. Interestingly, a combination of SCF and IL-9 exerted a distinct
additive effect on erythroid-burst formation. SCF only supported a small number of E-Mix colony formation; however,
a combination of SCF and IL-9 supported a significant number of E-Mix colonies. A combination of three BPAs such
as GM-CSF, IL-9, and SCF supported almost the same number of erythroid bursts as that supported by IL-3 alone. Finally, a combination of all four BPAs induced the maximal
number of erythroid-burst formations.
Combinations of the different BPAs also affected the sizes
E-Mix
BFU-E
10
0
50
100
of the erythroid burst (Fig 4). The cell number of each burst
was counted using a counting chamber, as we reported preEPO
v i o ~ s l y When
. ~ more than twoBPAswere combined, the
lL-3+Ep0
sizes of erythroid bursts significantly increased ( P CC .05 by
GM-CSF+Epo
Mann-Whitney rank-sum test), with the exception that the
IL-S+Epo
combination of IL-3 with GM-CSF had no effect. A combiSCF+Epo
nation of four BPAs exerted the strongest stimulatory effect
on proliferation of cells within the erythroid bursts; half of
(IL-3+Epo)+GM-CSF
them were macroscopic colonies containing more than 1 X
(IL-3+Epo)+lL-9
IO4 cells per colony.
(IL-3+Epo)+SCF
Effects of single BPA or a combination of more than two
BPAs on erythroid-burst or E-Mix colony formation by PB(GM-CSF+Epo)+IL-S
derived CD34+c-kip8"or CD34+c-kit'"" cells in serumfree
(GM-CSF+Epo) +SCF
culture. In the previous experiments, 30% FCS was used
in
our cultures. FCS most likely contains known or unknown
(IL-S+Epo) +SCF
substances that are capable of modulating the effects of
(GM-CSF+lL-S+SCF)
BPAs such as those of SCF, IL-3, GM-CSF, and 1L-9. To
+Epo
reduce this possibility, PB-derived highly purified progenitors were cultured at low cell density in our serum-free culture ~ystern.4"~
Because Gunji et a133showed thatCD34'
cells could be subdivided into three subfractions (c-kithighh,
c. .. .
kit'"", and c-kit- cells) and that CD34'c-kit'"" cells are an
100 150
10
0
50
independent population, C D 3 4 + ~ - k i tcells
~ ' ~ ~and CD34'cE-Mix
BFU-E (CFU-GM)
kit'"" cells were separately sorted as shown inFig 1, and
Fig3.Effectsof combinationsoffour BPAs on erythroid-burstor
these cells were cultured in the absence of FCS. The results
E-Mix colony formation by 500 PB-derivedCD34+HLA-DR+cells.Open
are shown in 17igs 5 a d 6.
and closed bars represent the numbers of BFU-E and E-Mlx, respeccells as the target, ILWhen we used 200 CD34+c-kithIgh
tively. A shaded bar on the bottom represents the number of GM
3 again showed the most potent BPA, whereas G"CSF
colonies. NS, statistically not significant; *, P < .05; **, P < .02; *I*,
P < .01.
supported approximately 60% of the BFU-E that were re-
From www.bloodjournal.org by guest on July 12, 2017. For personal use only.
SYNERGISTICACTIONS
4103
OF SCF AND OTHER BPAs
sponsive to IL-3 (Fig 5). On the other hand, SCF supported
a significantly higher number of erythroid bursts than did
GM-CSF ( P < .02) and induced a significant number of EMix colony formations. The BPA of SCF was apparently
much stronger than that when PB-derived CD34+HLA-DR+
cells were used as the target. IL-9 showed much weaker
BPA than the other cytokines and did not support E-Mix
colony formation. However, IL-9 significantly increased the
number of E-Mix colonies in the presence of GM-CSF ( P
< .05). As shown in Fig 3, IL-3 did not show a synergistic
action with GM-CSF, IL-9, or SCF on erythroid-burst formation. In contrast, IL-3 plus SCF produced a significantly
higher number of E-Mix colonies compared with that for IL3 alone ( P < .05). In addition, GM-CSF showed a significant
subadditive or additive effect with SCF on erythroid-burst
( P < .01) and E-Mix colony ( P < .05) formation. SCF also
exerted a subadditive effect with IL-9 in support of erythroidburst formation ( P < .01). A combination of GM-CSF, IL9, and SCF supported almost the same number of erythroid
bursts as that supported by IL-3 alone. The number of EMix colonies supported by a combination of four BPAs was
three times higher than that of IL-3 alone ( P < .Ol), and
the plating efficiency of this culture was roughly 80%.
Finally, we compared the effects of the four BPAs on
erythroid-burst and E-Mix colony formations using 500
CD34+c-kitloWcells as the target population (Fig 6). The
plating efficiency of this culture was approximately 50% in
the presence of a combination of IL-3, GM-CSF, G-CSF,
E-Mk
BFU-E
Epo
IL-3+Epo
GM-CSF+Epo
IL-g+Epo
SCF+Epo
(IL-3+EW)+GM-CSF
(IL-3+Epo)+lL-9
(IL-3+Epo)+SCF
(GM-CSF+Epo)+lL-9
(GM-CSF+Epo) +SCF
(IL-g+Epo) +%F
(GM-CSF+IL-~+SCF)
+Epo
(IL-3+0M-CSF
+IL-g+SCF)+Epo
IL-3+GM-CSF
+GCSF+Epo
5b
€-Mix
t
0
a
lh
l&
BFU-E (CN-GM)
Fig 5. Effects of combinations of four BPAs on erythroid-burst or
E-Mix colony formation by 200 PB-derived CD34*c-kithbh cells in serum-free cutture. (See legend to Fig 3.) NS, statistically not significant; *, P < .05; **, P < .01.
l x lo6
* .r -i
l
I
1 x 104,
t
'
I
... .. .. . .
1.
.
....."N...
N
.
.
I
.
.
.IC-
.:.
"
I
I
Fig 4. Deibution of sizes of erythroid bursts supported by single
BPA or a combinationof more thantwo BP&. The sires of erythroid
two BPAs were cornbursts significantly increased when more than
bined ( P < .05 by Mann-Whitney rank-sumtest).
and Epo. This population contained roughly 50% CFU-GM,
40% BFU-E, and 10% E-Mix colonies. In this experiment,
IL-3, GM-CSF, and SCF again showed significant BPA, and
each BPA supported E-Mix colony formation. IL-3 showed
the most potent BPA. SCF supported almost the same number of erythroid bursts as that supported by GM-CSF. The
number of erythroid bursts supported by GM-CSF significantly increased in the presence of SCF ( P < .05). In contrast, IL-9 alone did not show a significant BPA, and the
synergistic action between IL-9 and SCF disappeared. The
number of E-Mix colonies supported by IL-3 or GM-CSF
markedly increased in the presence of SCF ( P < .02). A
combination of four BPAs supported twice the number of
E-Mix colonies induced by IL-3 plus Epo ( P < .05). These
results obtained in serum-free culture strongly suggest that
these four BPAs act on erythroid-burst or E-Mix colony
formation without needing unknown serum factors.
Comparison of sizes of erythroid bursts generated by PBderived CD34+c-kithigh or
CD34+c-kit"" cells in serum-free
cultures. Next, we compared the sizes of erythroid bursts
generated by PB-derived CD34+c-kithighor CD34+c-kit'""
cells in serum-free cultures. The results are shown in Fig 7.
The size of erythroid burst generated by CD34'c-kit'" cells
was significantly larger than thatgenerated by C D 3 4 + ~ - k i t ~ ' ~ ~
cells in the presence of IL-3 plus Epo ( P < .01 by MannWhitney rank-sum test). In contrast, there was no significant
difference in the size of erythroid burst between these two
From www.bloodjournal.org by guest on July 12, 2017. For personal use only.
4104
SONODA ET AL
populations that expressed different levels of c-kit protein
in the presence of SCF plus Epo.
CD34'c-kit
DISCUSSION
I
It has been well-documented that both BPA and Epo are
required for the proliferation of BFU-E.'.' Until now, at least
four cytokines, IL-3, GM-CSF, IL-9, and SCF, have been
identified as BpAs,7.4,6-'4.2S~28
However, the interactions between SCF and other BPAs on erythroid-burst or E-Mix
colony formation by cells originating from different sources
such as bone marrow (BM), PB, or cord blood (CB) are
notyet fully understood. Recently, KuboandNakahata"
reported that BM-, PB-, and CB-derived erythroid progenitors respond differently to different BPAs such as SCF, IL3, or GM-CSF. They concluded that SCF exerted the most
potent BPA on BM-derived BFU-E, and IL-3 was the most
potent stimulator of PB- and CB-derived BFU-E. Their results suggest that the c-kit/SCF system plays an important
role in erythropoiesis in the BM.
In this study, we first compared the effects of the four
BPAs, IL-3, GM-CSF, IL-9, and SCF, on erythroid-burst and
E-Mixcolony formation using PB-derived highlypurified
progenitor cells astarget populations. These progenitors
were obtained from patients with testicular cancer by leukapheresis after chemotherapy and G-CSF administration;
therefore, they might not behave like normalprogenitor cells.
However, all of them could repopulate the BM after a combi-
E-Mix
50
high
BFU-E
0
50
100
b
IL-3+Epo
1 x10
I
CD34'c-kit
NS
low
I
1 x104
P c 0.01
.. ..
. -...
.......
......
t
. ..i
... G
....
.
.......
...... ....
........ .......
..........
e.
v)
S
3
1 x10 -Luy
1 x10
0.
"
1 x103
1 x102
Fig 7. Distributions of sizes of erythroid bursts generated by PBderived CD34+c-kithigh
or CD34+c-kit1" cells in the presence of IL-3
plus Epo or SCF plus Epo in serum-free culture. The size of erythroid
burst generated by CD34+c-kit" cells was significantly larger than
that generated by CD34'c-kithigh cellsin the presence ofIL-3 plus Epo
(P < .01 by Mann-Whitney rank-sum test). In the presenceofSCF
plus Epo, the size of erythroid burst generated by CD34+c-kitlow
cells
was slightly larger than that generated by CD34'c-kithigh cells. However, there was no significant difference between these two progenitor cell populations.
GM-CSF+Epo
IL-Q+Epo
SCF+Epo
nation of high-dose chemotherapy and PB stem cell transplantation using these progenitors, as wereported else-
(IL-3+EpO)+GM-CSF
(lL-3+EpO)+lL-Q
(IL-3+Epo)+SCF
here.'^
NS
**
(IL-Q+Epo) +SCF
(GM-CSF+IL-Q+SCF)
+m0
(IL-3+GM-CSF
+IL-Q+SCF)+Epo
IL-S+GM-CSF
+G-CSF+Epo
50
E-Mix
0
BFU-E (CFU-GM)
Fig 6. Effects of combinations of four BPAs on erythroid-burst or
E-Mix colony formation by 500 PB-derived CD34+c-kit" cells in serum-free culture. (See legend to Fig 3.)NS, statistically not significant; *, P < .05; **, P < .02.
As reported previou~ly,'~"~~"
IL-3 and GM-CSF have
BPA, and IL-3 shows more potent activity than GM-CSF.
In addition, 1L-9 and SCF exerted significant BPA when PBderived CD34+HLA-DR' cells or CD34'c-kithlghcells were
used as targets. These results were consistent with a number
of other studies,7-1~.~~-?X.3~
Although IL-9 could support erythroid-burst formation by PB-derived CD34+c-kithIgh
cells,
IL-9 alone failed to support erythroid-burst formation by
CD34'c-kit'"" cells.
As shown inFig 1, PB-derived CD34+c-kithIgh
cells expressed lower levels of the CD34 antigen on their surfaces.
Therefore, our data suggest that these cells seem to be more
mature progenitors than CD34'c-kit'"" cells that expressed
higher levels of the CD34 antigen. Recently, Katayama et
al" proposed a model in which c-kit protein is expressed at
low levels on primitive multipotential progenitors andat
high levels on more differentiated progenitors in a murine
culture system. Very recently, Gunji et aI3' clearly showed
that the number of long-term culture-initiating cells" was
From www.bloodjournal.org by guest on July 12, 2017. For personal use only.
SYNERGISTICACTIONS OF SCF AND OTHER BPAs
larger inhuman BM-derived CD34+c-kit1"" cells than in
CD34+c-kithigh
cells, and that the latter population derives
from the former after 4 weeks of coculture on stromal cells.
These results from BM strongly suggest that CD34+c-kit1""
cells are more primitive than CD34+c-kithigh
cells. The comparable PB-derived CD34+c-kit1""cells also seem to represent a more immature progenitor population than the
CD34+c-kithtph
population. As we demonstrated, the size of
erythroid burst generated by PB-derived CD34+c-kitl0"cells
was significantly larger than thatgenerated by CD34+c-kithIgh
cells in the presence of IL-3 plus Epo. These results are
consistent with each other and suggest that IL-9 acts on more
mature erythroid progenitors than those responsive to IL-3,
GM-CSF, and SCF. In an earlier report? we suggested that
the primary targets of IL-3 and GM-CSF are multipotential
progenitors at early and intermediate stages of development.
As described, IL-3 and GM-CSF showed apparent synergistic actions with SCF on E-Mix colony formation, indicating that SCF supports multipotential progenitors at very early
stages of development.
Catlett et a14" reported that exposure of BM-derived
CD34' cells to c-kit antisense oligonucleotides significantly
inhibited colony formation supported by IL-3 buthadno
effect on colony formation induced by GM-CSF. In addition,
Ratajczak et a14' reported that c-kit antisense oligonucleotides selectively inhibit erythroid-burst formation induced by
IL-3 plus Epo. In contrast, neither granulocyte/macrophage
nor megakaryocyte colony formation was inhibited by exposure to antisense oligomer. These results suggest that the ckit/SCF system predominantly functions in erythropoiesis
especially in association with IL-3.
To initiate a mitogenic signal, growth factors bind to specific receptors and activate intracellular kinases. The resulting chain of events transmits the growth signal from the
cell surface to the nucleus, resulting in proliferation and
differentiation of the progenitor cells. Our data show that
IL-3 does not synergize with SCF in erythroid-burst formation. Incontrast, GM-CSF did cooperate with SCF to support
erythroid-burst formation. Therefore, IL-3 may share a common signal transduction pathway with SCF in erythropoiesis,
a possibility suggested by the antisense oligonucleotides experiments discussed a b ~ v e . ' ~On
, ~ ' the other hand, the signals
activated by binding of GM-CSF to its specific receptor
may transmit through a different pathway, resulting in a
significant additive action.
The interactions of four BPAs, SCF, IL-3, GM-CSF, and
IL-9, on erythroid-burst or E-Mix colony formation are
rather ~ o r n p l e x . ~However,
' ~ ~ ~ ~ IL-3
~ ~ ~consistently
~~
supported the maximal number of PB-derived BFU-E, and there
was no synergistic action between IL-3 and GM-CSF, IL-9,
or SCF, suggesting that GM-CSF, IL-9, and SCF each support a subpopulation of PB-derived BFU-E that isresponsive
to IL-3. Our studies clearly showed that GM-CSF can cooperate with SCF or IL-9 and that IL-9 is also able to cooperate
with SCF to support erythroid-burst formation. In addition,
a combination of GM-CSF, IL-9, and SCF supported almost
the same number of BFU-E that were supported by IL-3
alone. Based on these data, we propose that GM-CSF, IL9, or SCF supports a partially overlapping but mostly distinct
4105
population of PB-derived BFU-E. SCF exerted a stronger
BPA on PB-derived CD34+c-kithigh
cells, suggesting that progenitors expressing different levels of cell-surface antigens
including c-kit protein respond differently to various BPAs.
In other words, responsive patterns of progenitors to various
BPAs possibly depend on their stage of development. It
has been reported that BM-derived erythroid progenitors are
more enriched in CD34+c-kit1"" cells.33However, our data
indicate that PB-derived erythroid progenitors are highly enriched in CD34+c-kithlghcells. Therefore, BM-derived erythroid progenitors might be different populations of progenitors from PB-derived progenitors. Further studies are
necessary to establish precisely how the c-kit/SCF system
plays a role in human erythropoiesis.
ACKNOWLEDGEMENT
The authors are grateful toAmgen (Thousand Oaks, CA) and
Kirin Brewery CO(Tokyo, Japan) for providing various growth factors used in this study.
REFERENCES
1. Heath DS, Axelrad AA, McLeod DL, Shreeve MM: Separation
of the erythropoietin-responsive progenitors BFU-E and CFU-E in
mouse bone marrow by unit gravity sedimentation. Blood 47:777,
1976
2. Iscove NN, Guibert LJ: Erythropoietin-independence of early
erythropoiesis and a two-regulator model of proliferative control in
the hemopoietic system, in Murphy MJ (ed): In vitro aspects of
erythropoiesis. New York, NY, Springer-Verlag. 1978, p 3
3. Sonoda Y, Yang Y-C, Wong GG, Clark SC, Ogawa M: Erythroid burst-promoting activity of purified recombinant human GMCSF and interleukin-3: Studies with anti-GM-CSF and anti-IL-3
sera and studies in serum-free cultures. Blood 72:1381, 1988
4. Sonoda Y, Yang Y-C, Wong GG, Clark SC, Ogawa M: Analysis in serum-free culture of the targets of recombinant human hemopoietic growth factors: Interleukin 3 and granulocytelmacrophagecolony-stimulating factor are specific for early developmental stages.
Proc Natl Acad Sci USA 85:4360, 1988
5. Yang Y-C, Ricciardi S , Ciarletta A, Calvetti J, Kelleher K,
Clark SC: Expression cloning of a cDNA encoding a novel human
hematopoietic growth factor: Human homologue of murine T-cell
growth factor P40. Blood 74,1880, 1989
6 . Sonoda Y, Maekawa T, Kuzuyama Y, Clark SC, Abe T: Human interleukin-9 supports formation of a subpopulation of erythroid
bursts that are responsive to interleukin-3. Am J Hematol 4134,
1992
7. Sieff CA, Emerson SG, Donahue RE, Nathan DG, Wang EA,
Wong GG, Clark SC: Human recombinant granulocyte-macrophage
colony-stimulating factor: A multilineage hematopoietin. Science
230:1171, 1985
8. Metcalf D, Begley CG, Johnson GR, Nicola NA, Vadas MA,
Lopez AF, Williamson DJ, Wong GG, Clark SC, Wang EA: Biologic
properties in vitro ofa recombinant human granulocyte-macrophage
colony-stimulating factor. Blood 67:37, 1986
9. Migliaccio AR, Bruno M, Migliaccio G : Evidence for direct
action of human biosynthetic (recombinant) GM-CSF on erythroid
progenitors in serum-free culture. Blood 70: 1867, 1987
10. Emerson SG, Yang Y-C, Clark SC, Long MW: Human recombinant granulocyte-macrophage colony stimulating factor and
interleukin-3 have overlapping but distinct hematopoietic activities.
J Clin Invest 82:1282, 1988
1 1. Migliaccio G, Migliaacio AR, Adamson JW: In vitro differentiation of human granulocytelmacrophage and erythroid progenitors:
From www.bloodjournal.org by guest on July 12, 2017. For personal use only.
4106
Comparative analysis of the influence of recombinant human erythropoietin, G-CSF, GM-CSF, and IL-3 in serum-supplemented and
serum-deprived cultures. Blood 72:248, 1988
12. Donahue RE, Yang Y-C, Clark SC: Human P40 T-cell growth
factor (interleukin-9) supports erythroid colony formation. Blood
75:2271, 1990
13. Holbrook ST, Ohls RK, Schibler KR, Yang Y-C, Christensen
RD: Effects of interleukin-9 on clonogenic maturation and cell-cycle
status of fetal and adult hematopoietic progenitors. Blood 77:2129,
1991
14. Lu L, Leemhuis T, Srour EF, Yang Y-C: Human interleukin
(1L)-9 specifically stimulates proliferation of CD34+++DR+CD33erythroid progenitors in normal human bone marrow in the absence
of serum. Exp Hematol 20:418, 1992
15. Zseho KM, Wypych J, McNiece IK, Lu HS, Smith KA, Karkare SB, Sachdev RK, Yuschenkoff VN, Birkett NC, Williams LR,
Satyagal VN, Tung W, Bosselman RA, Mendiaz EA, Langley K E :
Identification, purification, and biological characterization of hematopoietic stem cell factor from buffalo rat liver-conditioned medium.
Cell 63:195, 1990
16.Martin FH, Suggs SV, LangleyKE,Lu HS, Ting J, Okino
KH, Morris CF. McNiece IK, Jacobsen FW, Mendiaz EA, Birkett
NC, Smith KA, Johnson MJ, Parker VP, Flores JC, Patel AC, Fisher
EF, Erjavec HO, Herrera CJ, Wypych J, Sachdev RK, Pope JA,
Leslie I, Wen D, Lin C-H, Cupples RL, Zsebo KM: Primary structure
and functional expression of rat and human stem cell factor DNAs.
Cell 63:203, 1990
17. Williams DE, Eisenman J, Baird A, Rauch C, NessKV,
March CJ, Park LS, Martin U, Mochizuki DY, Boswell HS, Burgess
GS, Cosman D, Lyman SD: Identification of a ligand for the c-kit
proto-oncogene. Cell 63: 167, 1990
18. Ogawa M, Matsuzaki Y, Nishikawa S, Hayashi S, Kunisada
T, Sudo T, Kina T, Nakauchi H, Nishikawa S: Expression and function of c-kit in hemopoietic progenitor cells. J ExpMed 174:63,
l99 I
19. Okada S, Nakauchi H, Nagayoshi K, Nishikawa S, Nishikawa
S, Miura, Suda T: Enrichment and characterization of murine hematopoietic stem cells that express c-kit molecule. Blood 78:1706, 1991
20. Carow CE, Hangoc G, Cooper SH, Williams DE, Broxmeyer
HE: Mastcell growth factor (c-kit ligand) supports the growth of
human multipotential progenitor cells with a high replating potential.
Blood 78:2216, 1991
2 I . Strobl H, Takimoto M, Majdic 0, Hocker P, Knapp W: Antigenic analysis of human hemopoietic progenitor cells expressing the
growth factor receptor c-kit. Br J Haematol 82:287, 1992
22. Yamaguchi Y, Gunji Y, Nakamura M, Hayakawa K, Maeda
M, Osawa H, Nagayoshi K, Kasahara T, Suda T: Expression of ckit mRNA and protein during the differentiation of human hematopoietic progenitor cells. Exp Hematol 21: 1233, 1993
23. Orlic D, Fisher R, Nishikawa S, Nienhuis AW, Bodine DM:
Purification and characterization of heterogeneous pluripotent hematopoietic stem cell population expressing high levels of c-kit receptor. Blood 82:762, 1993
24. Reisbach G, Bartke I, Kempkes B, Kostka G, Ellwart J, Birner
A, Thalmeier K, Mailhammer R, Bomkamm GW, Ullrich A, Dormer
P: Characterization of hemopoietic cell populations from human
cord blood expressing c-kit. Exp Hematol 21:74, 1993
25. Dai CH, Krantz SB, Zsebo KM: Human burst-forming unitserythroid need direct interaction with stem cell factor for further
development. Blood 78:2493, 1991
26. Toksoz D, Zsebo KM, Smith KA, Hu S, Brankow D, Suggs
SV. Martin FH, Williams DA: Support of human hematopoiesis in
SONODA ET
AL
long-term bone marrow cultures by murine stromal cells selectively
expressing the membrane-bound and secreted forms of the human
homolog of the steel gene product, stem cell factor. Proc Natl Acad
Sci USA 89:7350, 1992
27. Migliaccio C, Migliaccio AR,Druzin ML, Giardind P-J V,
Zsebo KM, Adamson JW: Effects of recombinant human stem cell
factor (SCF) on the growth of human progenitor cells in vitro. J
CellPhysiol 148503, 1991
28. Peschle C, Gabbianelli M, Testa U, Pelosi E, Barberi T, Fossati C, Valtieri M, Leone L: c-kit ligand reactivates fetal hemoglobin
synthesis in serum-free culture of stringently purified normal adult
burst-forming unit-erythroid. Blood 81:328, 1993
29. Kuzuyama Y. Sonoda Y, Sakabe H, Nakagawa H,Fujii H,
Abe T: Successful collection of peripheral blood stem cells mobilized by a combination of G-CSF with high-dose Ara-C or highdose etoposide. Jpn J Clin Hematol 34: 1525, 1993
30. Takdue Y, Watanabe T, Kawano Y, Koyama T, Abe T, Suzue
T, Satoh J, Shimokawa T, Ninomiya T, Kosaka M, Shimizu E, Ogura
T, Kuroda Y: Isolation and storage of peripheral blood hematopoietic
stem cells for autotransplantation into children with cancer. Blood
74: 1245, 1989
3 1, Lansdorp PM, Dragowska W: Long-term erythropoiesis from
constant number of CD34' cells in serum-free cultures initiated with
J Exp
highly purified progenitor cells fromhumanbonemarrow.
Med 175:1501, 1992
32. Sonoda Y, Kuzuyama Y, Tanaka S, Yokota S, Maekawa T,
Clark SC, Abe T: Human interleulun-4 inhibits proliferation of megakaryocyte progenitor cells in culture. Blood 81:624, 1993
33. Gunji Y, Nakamura M, Osawa H, Nagayoshi K,Nakauchi
H, Miura Y, Yanagisawa M, Suda T: Human primitive hematopoietic
in kit"'"' cells than
cells.
progenitor cells are
more
enriched
Blood 82:3283, 1993
34. Sonoda Y, Ogawa M: Serum-free culture of human hemopoietic progenitors in attenuated culture media. Am J Hematol 28:227,
1988
35. LuL,WalkerD, Broxmeyer HE, Hoffman R, Hu W, Walker
E: Characterization of adult human marrow hematopoietic progenitors
highly enriched by two-color cell sorting with My10 and major histocompatiblity class monoclonal antibodies. J Immunol 139: 1823, 1987
36. Kubo T, Nakahata T: Different responses of human marrow
and circulating erythroid progenitors to stem cell factor, interleukin3 and granulocyte/macrophage colony-stimulating factor. Int J Hematol 58:153, 1993
37. Nakagawa S, Sugimoto K, Mikami K, Watanabe H, Sonoda
Y, Kuzuyama Y, Abe T, Fujii H: Successful collection of peripheral
blood stem cells mobilized by high-dose etoposide for patients with
chemotherapy-resistant andor poor prognostic testicular cancer. Jpn
J Urology 85:571, 1994
38. Katayama N, Shih J-P, Nishikawa S, Kina T, Clark SC,
Ogawa M: Stage-specific expression of c-kit protein by murine hematopoietic progenitors. Blood 82:2353, 1993
39. Sutherland HJ, Eaves CJ, Eaves AC, Dragowska W, Lansdorp
PM: Characterization and partial purification of human marrow cells
capable of initiating long-term hematopoiesis in vitro.
Blood
74:1563, 1989
40. Catlett JP, Leftwich JA, Westin EH, Grant S, Huff TF: c-kit
expression by CD34+ bone marrow progenitors and inhibition of
response to recombinant human interleukin-3 following exposure to
c-kit antisense oligonucleotides. Blood 78:3 186, 1991
41. Ratajczak MZ, Luger SM, Deriel K, Abrahm J, Calabretta B,
Gewirtz AM: Role of the kit protooncogene in normal and malignant
human hematopoiesis. h o c Natl Acad Sci USA 89:1710, 1992
From www.bloodjournal.org by guest on July 12, 2017. For personal use only.
1994 84: 4099-4106
Synergistic actions of stem cell factor and other burst-promoting
activities on proliferation of CD34+ highly purified blood progenitors
expressing HLA-DR or different levels of c-kit protein
Y Sonoda, H Sakabe, Y Ohmisono, S Tanimukai, S Yokota, S Nakagawa, SC Clark and T Abe
Updated information and services can be found at:
http://www.bloodjournal.org/content/84/12/4099.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.