From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
Maintenance of Murine Long-Term Repopulating Stem Cells in Ex Vivo
Culture Is Affected by Modulation of Transforming Growth Factor-/3 But
Not Macrophage Inflammatory Protein - 1a Activities
By Toshihiro Soma, Jian Mei Yu, and Cynthia E. Dunbar
Transforming growth factor-j? (TGF-j?) and macrophage inflammatory protein-la (MIP-la) are both well-described inhibitors of committed and multipotential hematopoietic progenitors. The effect of these cytokines on true stem cell
activity in ex vivo culture systems as assayed by murine
long-term repopulating activity (LTRA) has not been examined. We studied the stem cell effects of the addition of these
cytokines to ex vivo cultures containing interleukin-3 (IL-31,
IL-6, and stem cell factor (SCF), using the murine competitive
repopulation assay. We also tested the impact of adding an
anti-TGF-P neutralizing antibody, to ask whether abrogation of autocrine/paracrine TGF-j? may protect or enhance
the survival of LTRA during ex viva culture. TGF-Pl had significant suppressive effects on both short- and long-term
repopulating activities, and anti-TGF-6 antibody had enhancing effects compared with control cultures containing
IL-3, IL-6, and SCF alone. MIP-la had no significant effects
on either short- or long-term repopulating ability. These data
suggest that abrogation of TGF-p during suspension culture
may allow enhanced survival or even expansion of primitive
cells ex vivo, with implicationsfor many applications, including gene therapy.
This is a US government work. There are no restrictions on
its use.
T
formation in v i t r ~ . ” ~
Inhibitory
’~
actions on committed progenitors have also been reported after in vivo administration,
and protection of progenitors from chemotherapy-induced
cytotoxicity has also been demon~trated.’~*~~
Thus, MIP- la
has been classified as one of the main inhibitory hematopoietic cytokines along with TGF-Ps. But more recent reports
studying more primitive LTCIC and/or purified progenitor
populations have found evidence for a stimulatory or protective role for MIP-la during ex vivo culture.18~21~22
Thus far there have been no investigations of the effects
of TGF-0s or MIP-la on true long-term repopulating stem
cell activity. This knowledge could be critical for ex vivo
expansion or gene therapy applications directed at hematopoietic progenitor or stem cells, especially because there is
strong evidence for autocrine and paracrine production of
these cytokines by hematopoietic and microenvironmental
cell^.^.'^ Thus, we designed a series of experiments focusing
on the effects of TGF-P, MIP-la, or anti-TGF-P on ex vivo
culture of murine stem cells, using the murine competitive
repopulation assay system that allows quantitation of true
long-term repopulating a ~ t i v i t y . ’ ~ . ~ ~
RANSFORMING growth factor-ps (TGF-0s) are a
group of pleiotropic cytokines with significant effects
on hematopoiesis.”*Of the three mammalian isoforms, TGFP l is the best characterized. But recent reports have shown
that TGF-P3 is also important in hematopoietic reg~lation.~
These cytokines have a complex array of inhibitory and
stimulatory activities in the hematopoietic and immune syst e m ~ .The
~ . ~type of activity may be dependent on the differentiation stage of the target cell population.’ The recently
reported TGF-P1 knockout mouse model has shown the
prominent role for this cytokine in preventing hyperactive
inflammatory responses and immune system over activity.’
Many investigators have explored the effects of TGF-0s
on hematopoietic progenitor cells. The plating efficiencies
of differentiated, single-lineage committed progenitors such
as colony-forming unit-erythroid (CFU-E), CFU-granulocyte (CFU-G), and CFU-macrophage (CFU-M) are stimulated.’ But more immature multipotential progenitors such
as CFU-granulocyte, erythroid, macrophage, megakaryocyte
(CFU-GEMM), high proliferative potential-colony-forming
cell (HPP-CFC), and CFU-spleen (CFU-S) are inhibited, and
these actions can be modified by specific combinations or
temporal sequences of administration of stimulatory cytokines.”.” Human long-term culture initiating cells (LTCIC)
are also inhibited. ‘ - I 3 Administration of pharmacologic
doses of TGF-PI to mice in vivo decreases platelet and red
blood cell numbers but increases leukocyte c o ~ n t s . The
’~
origin of TGF-Ps in the hematopoietic microenvironment is
also complex. They are produced by stromal elements as
well as by hematopoietic progenit~rs.’~’’~
The balance of
autocrine and paracrine positive and negative regulators
present in the microenvironment, both in vivo or during ex
vivo culture, presumably has a major impact on the homeostasis of hematopoietic cell production.
Macrophage inflammatory protein-la (MIP-la) is one of
the family of chemokines produced by macrophages, and
has also been called “stem cell inhibitor.”16 This cytokine
inhibits immature multipotential CFU-S and CFU-A.I6 Although there is clear evidence that MlP-la inhibits stem cell
factor-induced proliferation of human CFU-GM and burstforming units-erythroid, other investigators have shown both
inhibitory or stimulatory actions on committed progenitors
depending on the other cytokines used to support colony
’
Blood, Vol 87, No 11 (June I ) , 1996: pp 4561-4567
MATERIALS AND METHODS
Mice. C57BW6J, B6.C-H-1ByJ (abbreviated HW80), and
WBB6F, WIW‘ or +/+ mice were purchased from Jackson Labs
(Bar Harbor, ME).
Competitive repopulation assay. This assay was performed as
described previo~sly?~
Briefly, donor C57BL/6J or HW80 mice were
injected intravenously with 5-fluorouracil (5-FU;150 mgkg) 2 days
before bone marrow (BM) harvest. On the day of harvest, BM from
tibias and femurs was expelled into D15 media: Dulbecco’s minimal
From the Hematology Branch, National Heart, Lung and Blood
Institute, Bethesda, MD.
Submitted November 14, 1995; accepted January 23, 1996.
Address reprint requests to Cynthia E. Dunbar, MO,Hematology
Branch, NHLBI, Bldg 10, Room 7C103, National Institutes of
Health, Bethesda, MD 20892.
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 18 U.S.C. section 1734 solely to
indicate this fact.
This is a US government work. There are no restrictiom on its use.
0006-4971/96/871I -0008$0.OO/0
4561
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
SOMA, YU, AND DUNBAR
4562
essential medium with 4 mmol/L glutamine. 50 mg/mL penicillin,
and 50 mg/mL streptomycin, (all from GIBCOIBRL, Gaithersburg,
MD). and 15% fetal calf serum (Hyclone, Logan, UT) dispersed to
a single-cell suspension by passage through a 21-gauge needle. BM
cells were cultured at 5 X 10' nucleated cellslmL in D15 medium
with cytokines on non-tissue culture treated 100-mm dishes (Sarstedt. Princeton, NJ). Murine interleukin-3 (IL-3; Genzyme. Boston,
MA), human IL-6 (Amgen, Thousand Oaks, CA), and rat stem cell
factor (Amgen, Thousand Oaks, CA) were added to every culture at
concentrations of 10 ng/mL. 50 nglmL, and 100 ng/mL, respectively.
Some C57BW6J cultures had purified natural human TGF-Dl at
5 ng/mL or murine MIP-la at 100 ng/mL added (R&D Systems,
Minneapolis. MN). Other C57BL/6J plates had anti-TGF-8 I .2,3
murine monoclonal antibody (MoAb; Genzyme) added at 20 pglmL.
which inhibits up to 2.5 nglmL of TGF-Dl activity. This antibody has
been shown to be neutralizing in both ex vivo and in vivo assays."
Control polyclonal mouse IgG was added to some control plates of
HW80 cells. After 4 days of incubation at 37°C. 5% COz. the BM
cells were collected from the HW80 and C57BL/6J plates, counted,
and pooled for each experimental competitive condition. C57BU6J
cells (either treated with TGF-0, MIP-la, anti-TGF antibody or no
addition) were mixed with cells representing the same initial dosage
(same plate equivalent) of nontreated or isotype control-treated
HW80 cells. Cells were injected via tail vein into nonirradiated W/
W' recipient mice. The total number of cells injected ranged from
0.85 to 1.5 X IOh per WIW' recipient, corresponding to 5 X IOr'
cells initially plated.
Hb electrophoresis. Heparinized peripheral blood (PB) was obtained from the retro-orbital sinus every 2 to 4 weeks. Hemoglobin
(Hb) phenotype was determined by cellulose acetate electrophoresis
of heparinized PB as previously described, using reagents and a
CliniScanII instrument from Helena Laboratories (Helena Laboratories, Beaumont, TX).'" The relative repopulating ability of the
HW80 (control) compared with the C57BL/6J (treated or control)
marrow was determined by the relative intensities of the Hb diffuse
(HW80) versus Hb single (C57BL/6J) bands.
CFII-S12nssciy. The CFU-SI2 assay was performed by the
method of Till and McCulloch.'" WBB6FI -+/+ recipients received
900 rads total body irradiation using a '17Csgamma irradiator. Within
2 to 4 hours, the mice were injected with 100 to 1O.ooO cultured
marrow cells via the tail vein. Twelve days later the spleens were
removed and fixed in 70% ethano1:acetic acid:formalin at 20:1:1,
and macroscopic CFU-SI, colonies were counted.
CFU-C assay. Cultured marrow cells were plated at 10,OOO to
100,000/mL in murine methylcellulose media (Stemcell Technologies Inc. Vancouver, Canada) containing 3 U/mL human erythropoietin (Amgen. Thousand Oaks, CA) and 10%WEHI-3B-conditioned
media. Colonies of greater than 50 cells were counted after incubation at 37°C in 5% CO2 for 14 days.
Southern blot analvsis. At least 4 months after transplantation,
recipient WIW' mice were killed and high-molecular-weight DNA
was extracted from the thymus and BM. BM and thymus DNA was
digested with EcoRl (GIBCOI BRL) and probed with a 61 I-bp Psr I/
BamHI fragment from the second intervening sequence of the murine
murine Hb gene as described.27The two p"ple
alleles (C57BLI
63) have an expected fragment size of 10 kb, whereas the ~ l ' " " Y
alleles (pmnJ"'andp'"'")
have expected fragment sizes of 14 kb and
7 kb. with the probe hybridizing well only to the 7-kb allele.
Statisricd ana/ysi.s. Statistical significance was assessed by the
two-tailed Student's t-test using Statviewll computer software (Abacus Concepts, Sunnyvale. CA).
RESULTS
Design qf competitive repopulation assay. We tested the
effect of altering TGF-,B o r MIP-la levels under ex vivo
HW80
C57BLl6J
d
S
F
U
&FU
4
HarveslBone Marrow
f'.
2 days aher injection
f'.
WRY"
WRY"
0
0
-
0
-
WRY"
WRYY
('57111.
0
0
0
-
0
HWSD
Hb Electrophoresis
0
0
-
0
Fig 1. Experimental design-Competitive repopulation assay.
mice were injected with 5C57BL/6J (Hbp""O'')and HW80 ~Hbpdm"")
FU at 150 mg/kg, and 2 days later BM cells were collected from
femurs and tibias and cultured in D15 media plus IL-3, IL-6, and SCF.
C57B1/6J cells additionally had the experimental cytokines or antibodies as indicated in the figure added to the culture media. After 4
days of culture, the cells were obtained and marrow cells from equivalent number of plates of HW80 BM and C57BL/6J BM were combined and injected into at least four W / W recipient mice for each
cell mixture. Samples of the C57BL/6J marrow were also used for
CFC and CFU-S assays. Every 2 to 3 weeks, mice were bled for Hb
analysis.
culture conditions similar to those used for transduction of
murine BM cells with retroviral vectors. To enrich for cycling progenitor cells, donor marrows were obtained 2 days
after 5-FU treatment.** As shown in Fig I and detailed in
Materials and Methods, marrow cells from the two donor
strains were cultured for 4 days. At the end of the culture
period, HW80 cells grown under standard conditions (IL-3,
IL-6 and SCF, plus isotype control antibody in experiment 4)
were mixed with C57BL/6J cells grown under experimental
conditions (IL-3, IL-6, and SCF, plus either no additional
factors, TGF-01, MIP-la, or anti-TGF-0 antibody). The
two types of donor cells were then mixed as equal plate
equivalents, so that each mixture contained cells derived
from equal numbers of starting cells from the two populations. A total of four separate experiments, each including
all four experimental conditions, were performed.
Cell numbers (XIO") at the end of the culture per 10 X
IO" cells plated were: HW80 control 2.95 +- 0.8, C57BL/6J
control 2.97 +- 1.1, anti-TGF-,B 2.97 +- 0.62, TGF-0 2.75
2 0.75, and MIP-la 2.55 2 0.5. In each experiment, at least
four WIW' recipient mice received cells cultured under each
experimental condition, with cell numbers injected ranging
from 0.85 to I S O x I O6 celIs/mouse.
Effects on committed CFU-C and CFU-S. The numbers
of CFU-C and CFU-S present in ex vivo cultures under the
various experimental conditions were assessed after 4-day
incubations. As shown in Fig 2, the anti-TGF-,B antibodycontaining cultures had CFU-C and CFU-S numbers increased by 135% and 90%, respectively. In contract, TGF-
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
MURINE STEM CELL MAINTENANCE AFFECTED BY TGF-P
4563
Table 1. Short-Term ReconstitutingAbiliw of Treated BM
CFC
f
4W 1
A
Culture Conditions of Hb"'"''" (C57EU6J)Marrow
Experiment No.
1 (Day 56)
2 (Day 46)
3 (Day 53)
4 (Day 61)
MIP-la
TGF-81
46.5 r 0.93
38.2 t 2.6
59.4 t 2.4
ND
21.2 5 2.4'
14.3 t 3.4t
34.2 2 2.0'
49.7 2 17.4
Anti-TGF-/3 Ab
57.0 5
50.8 t
73.3 ?
90.0 ?
3.05$
3 3
3.0t
5.37
Control
46.8 f 0.97
40.4 t 4.7
61.2 t 2.2
54.1 ? 11.0
Data shown are mean % lib"'""'"t SD in the PE of each set of animals.
P < ,001.
t P < .01.
P < .05.
*
Anii-TCiF+ Anlthdy
TW-p
MIP-la
CFU-S
B
-r
Anti-TGF$ Antibody
TGF-E
MIP-la
Fig 2. Effects of culture conditions on committed progenitors. (A)
CFU-C present after 4 days of culture under the experimental conditions comparedwith the number present after 4 days of culture under
control conditions (IL-3, IL-6, and SCF alone). (B) CFU-S present after
4 days of culture under the experimental conditions compared with
the number present after 4 days of culture under control conditions
(IL-3, 11-6, and SCF alone). Results were calculated from two IMIPl a ) or four (TGF-P and anti-TGF-p) independent experiments. * P <
.05;, * * P < .01.
Dl -containing cultures had CFU-C and CFU-S decreased
by 82% and 73%, respectively. MIP-la only marginally
decreased the yield of CFU-S and CFU-C, and this effect
was not statistically significant.
Efects on short- nnd long-term repopulating cells. Every 2 to 4 weeks, Hb phenotype was analyzed by cellulose
acetate electrophoresis and quantitated by densitometry.
Data on short-term repopulation (40 to 60 days posttransplantation) in four separate experiments are shown in Table
1. The percent pIlncle Hb, corresponding to the repopulation
of the erythroid compartment derived from the "treated"
(CS7BW6J) marrow, is shown. At these early time points,
there was a strong inhibitory effect of TGF-PI and an opposite positive effect of anti-TGF-P antibody. MIP-la had no
significant effect. Table 2 shows the effect of the culture
conditions on long-term repopulating ability (defined as
greater than 120 days) in the same four experiments. Effects
were similar to those noted at the shorter-term time points,
but variability from animal to animal increased, especially
in the animals receiving TGF-PI -treated marrow. Again,
MIP-la had no stimulatory or inhibitory effects.
Figure 3A shows reconstitution over time for each of the
four experiments. expressed as the mean Hb""F'' value for
each experimental group of mice (CS7BW6J marrow cultured with the addition of either TGF-PI or anti-TGF-P
antibody competed against HW80 cultured under standard
conditions or with control antibody) divided by the mean
HVIttgk value for each control group of mice (C57BU6J marrow cultured under standard conditions competed against
HW80 marrow cultured under standard conditions or with
control antibody) at each time point. A ratio of 1.0 means
that the experimental group of mice had no difference in
reconstitution compared with the control group of mice that
received marrow of both types incubated under identical
conditions. All four experiments found that anti-TGF-P antibody-containing cultures produced relatively constant enhancement of reconstitution compared with standard culture
conditions over the whole time period, although the absolute
levels varied between experiments, suggesting a uniform
protective activity on both long- and short-term reconstituting abilities. The most potent inhibitory effects of TGF-PI
were seen in the first 2 to 3 months, with some dispersion
of later values. Figure 3B shows the same reconstitution time
course for MIP-la-treated marrow. MIP-la was neither
inhibitory nor stimulatory under our experimental conditions.
Individual animals receiving TGF-P-cultured marrow in
all four experiments began to develop more marked variability in results after day 60 to 90 posttransplantation, leading
to loss of statistical significance although maintenance of an
overall inhibitory effect. The anti-TGF-P antibody and
MIP-la animals did not show this individual variability.
Table 2. LongTerm ReconstitutingA b i l i of Treated BM
Culture Conditions of Hb"""" (C57EU6.J)Marrow
Experiment No.
1
2
3
4
(Day
(Day
(Day
(Day
141)
124)
1231
122)
MIP-le
TGF-01
49.5 t 12.9
54.1 z 5.3
76.1 t 8.2
ND
35.5 t 3.8.
32.6 z 9 . l t
47.3 ? 2.3t
50.0 t 17.5
Anti-TGF-0 Ab
62.9 5
74.8 5
87.0 2
92.9 2
2.lt
7.lt
9.1
4.1.
Control
55.0 z 2.4
50.6 z 4.1
76.1 t 9.8
49.5 t 17.5
Data shown are mean % Hb""'' -r SD in the PE of each set of animals.
P < .01.
t P < .05.
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
4564
SOMA, YU, AND DUNBAR
Reconstitution Time Course
*I
that discriminates between the patterns of P-Hb alleles found
in the two donor strains. These band pattems can be seen in
lanes representing the two donor (HW80 and C57BL/6J) and
the W W WBBFl recipient strains. Animals receiving TGFPI -cultured marrow competed against standard conditions
had dominant bands corresponding to the
Hb phenotype's alleles (pmaJor
and ,Bminor),
concordant with the Hb data
and indicating that culture of marrow in TGF-PI was inhibitory to reconstitution in multiple lineages. Similar confirmatory results were seen in the anti-TGF-P animals, with a
phenotype. Quandominant band corresponding to the
titation of the densities of Southern blot bands and the corresponding Hbsinglebands on cellulose acetate electrophoresis
are also shown. The numbers correlate well, indicating similar effects of either the TGF-Plor the anti-TGF-P cultures
on myeloid, erythroid, T-cell, and B-cell lineages.
DISCUSSION
0.0
30
60
90
120
150
180
Days Post-Transplant
Fig 3. Reconstitution over time in the competitive repopulation
assay. Time course of reconstitutionin TGF-8 and anti-TGF-0 groups
(A) and MIP-la group (8).Data are presented as the ratios of mean
HbP.'"** values of the treated group (C57BL/6J marrow culturedwith
either TGF-/3, anti-TGF-8 antibody or MIP-la competed against
HW80 marrow cultured with 11-3, 11-6, and SCF alone) to the mean
HbP.ins''values of the control group (C57BL16.l marrow culturedwith
IL-3, IL-6, and SCF alone or with isotype control antibody competed
against HW80 marrow culturedwith IL-3, IL-6, and SCF alone) at each
assayed time point after transplantation.Each line represents an individual experiment, and is the mean of at least four mice for each
experiment.
This increased variability in the TGF-P group might be explained by a decreased number of functional long-term repopulating cells present after culture in the presence of TGFP, leading to repopulation by a very small number of C57BL/
65 clones and thus marked changes in Hbsingle levels over
time in individual animals as well as marked differences
between animals in these groups.
DNA analysis of reconstitution patterns of other hematopoietic lineages. Other investigators have verified that erythroid reconstitution is representative of reconstitution in
other nonerythroid hematopoietic lineages in the W W competitive repopulation assay.29 We also validated our Hb reconstitution results by analyzing the relative proportions of
the P-Hbgene alleles present in other hematopoietic tissues
at least 120 days posttransplantation. Figure 4 shows a
Southem blot analysis of DNA derived from BM spleen and
thymus of reconstituted animals. A P-Hb probe was used
In our experiments, we have shown that TGF-01 is inhibitory to long-term repopulating activity (LTRA) in ex vivo
suspension culture, and that addition of anti-TGF-P antibody can improve LTRA compared with standard supematant cultures containing 1L-3, IL-6, and SCF alone. This
expands the known levels of activity of this cytokine to the
most primitive repopulating cells, beyond effects on progenitor cells previously described by other investigators.
A recent study reported that TGF-PI downregulates c-kit
expression on both immortalized progenitor cell lines and
primary BM cells.'"." Because SCF signaling via c-kit upregulates the sensitivities of progenitor cells to various stimulatory cytokines, it is quite probable that TGF-,O participates
in negative regulation of hematopoiesis through downregulation of cytokine receptors. SCF and other stimulatory cytokines probably exert much of their stimulatory or protective
activities via inhibition of apoptosis. A recent study reported
that the viability of purified murine Lin-Scal' hematopoietic
cells was compromised by the inclusion of TGF-P during
culture with SCF or granulocyte colony-stimulating factor.32
These cells were shown to undergo apoptosis more frequently in the presence of TGF-P ,and anti-TGF-P antibody
was shown to be protective. These data correlate well with
our functional data on actual LTRA.
Our results showing a protective or stimulatory effect of
the inclusion of a neutralizing anti-TGF-P antibody during
ex vivo culture suggests that autocrine or paracrine production of TGF-Ps by marrow elements may be an important
negative regulator of hematopoiesis and impediment to successful ex vivo expansion or gene transduction. Antisense
oligonucleotides to TGF-P have been shown to enhance colony formation by purified human progenitor cells, suggesting
a direct autocrine loop involving this cytokine on primitive
progenitor cells.'' Individual sorted primitive murine hematopoietic cells also have been shown to have enhanced viability in cultures containing the neutralizing antibody, further
supporting a significant component of autocrine regulation
of primitive progenitors by TGF-p.32 In addition, marrow
stromal elements also produce TGF-Ps. In long-term Dextertype cultures, concentrations of TGF-0s were as high as 2
to 5 ng/mL I day after media change.'* Our experiments used
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
MURINE STEM CELL MAINTENANCE AFFECTED BY TGF-0
MIP
Ob Hb-single
o
4565
COntro I
TGF
Anti-TGF
9
3
v!
CO
(3
m
CO
*
W
c?
(D
CO
Fig 4. Southern blot analysis of the p-globin locus of reconstituted mice. EcORl digestion of DNA from homozygous C57BL/W (Hb'i"P"/
Hb'i"P1. donors gives one band of 10 kb (band 21, representing the p-single genes. EcoRl digestion of DNA from homozygous HW80 donors
(HbdflUU/HbddfU.. ) donors gives two bands: a 7-kb (band 3) band representing the p-major allele, and a 14-kb (band 1) band representingthe
p-minor gene (weakly hybridized by our probe). W / W recipient heterozygotes show all three bands. Analysis of DNA extracted from the
spleen, thymus, and BM of repopulated mice are shown. C57BL/6J marrow received the treatment indicated before transplantation. Below
each lane the percent of signal representingthe IO-kb band (band 2) by phosphoimager analysis is given, with the percent Hb""g"found on
Hb electrophoresis just before sacrifice.
nonfractionated marrow cclls, thus we cannot distinguish
autocrine from paracrine effects.
Another interesting feature observed in our experiments
were the temporal patterns of the effects of TGF-PI versus
anti-TGF-P antibody. The suppressive effect of TGF-01
appeared to be strongest on short-term reconstituting cells.
and eventually weakened and became more variable. On the
other hand, anti-TGF-8 neutralizing antibody appeared to
be more consistently protective of both short- and long-term
repopulating activities. These differences suggest that shortterm reconstituting cells may be more susceptible to exogenous TGF-01 than long-term repopulating cells because of
a greater reliance on ex vivo stimulation through SCFlc-kit
interactions or other pathways. The presence of large
amounts of TGF-01 may induce the most primitive cells to
in some way modulate their cell-cycle status or responsiveness to cytokines in such a way as to be relatively protected. The presence of the neutralizing antibody may instead
allow both short- and long-term repopulating cells to remain
protected from a much lower endogenous production level.
It is also possible that autocrine or paracrine TGF-P2 or 3
are responsible for inhibition of the most primitive longterm repopulating cells, and that the antibody was effective
due to neutralization of all three isoforms, whereas addition
of TGF-P 1 had its strongest effects only on short-term reconstituting cells.
We have also shown that inclusion of MIP-la did not
have inhibitory or stimulatory actions on LTRA under our
experimental conditions. Other investigators have shown that
MIP-la does not have inhibitory actions on more primitive
human progenitors despite marked inhibition of CFU-GM
formation in the presence of IL-3." In fact, it has been
reported that MIP-la allowed better maintenance of human
LTClC in extended cultures including a stromal layer."
When MIP-la was administered to mice just after treatment
with 5-FU. LTRA was not protected from a second dose of
5-FU administered 7 days later, suggesting that the MIP- la
did not prevent stem cells from going into cell cycle after
the first dose of 5-FU." This lack of an in vivo effect on
stem cell activity, despite activity in the same experiment
on more committed progenitor cells, agrees with our conclusions.
After in vivo administration. both MIP- I a and TGF-PI
have shown an ability to protect against the immediate myelosuppressive effects of subsequently administered chemoIt is hypothesized that this is due to prevention
of cycling of progenitor cells and thus protection from killing
by cell-cycle-specific agents. These in vivo experiments
document a marked contrast between in vivo and in vitro
effects of these cytokines. In vitro, progenitor and stem cells
are exposed to cytokines, both stimulatory and inhibitory,
without the usual microenvironmental components that may
completely alter the response of cell populations to these
influences in vivo. The addition of as yet unknown soluble
factors produced by stromal cells or inclusion of adhesive
molecules such as fibronectin may further improve the survival or proliferation of stem cells during ex vivo culture.".'5
Efficient gene transfer to true repopulating stem cells, especially in large animals or humans, has proven very diffic~lt.'~.''The most commonly used vector systems depend
on the target cell going through S phase to integrate vector
genes into the genome." A major hurdle to effective gene
therapy may be the fact that primitive hematopoietic stem
cells are not commonly in S phase."' Existing ex vivo suspension culture systems using cocktails of hematopoietic cytokines are very efficient at cycling and expanding committed
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
4566
SOMA, YU, AND DUNBAR
progenitor cells, but data suggest, even in the murine model,
that true LTRA decreases during ex vivo culture even in
the presence of cytokine cocktails that are very effective
at expanding committed progenitor^.^^,^^^^ These cytokine
cocktails may be driving stem cells to irreversibly differentiate, decreasing LTRA.4’ Thus, an understanding of the function of inhibitory cytokines in the context of stem cell biology may yield important new approaches to improving
LTRA and stem cell gene transfer. Abrogation of the effects
of TGF-P or other inhibitory cytokines may allow more
successful ex vivo culture or even expansion of LTRA for
gene therapy or stem cell transplantation applications. Primitive cells pushed into cycle may go through apoptosis if
TGF-P is present, and ex vivo culture may stimulate the
autocrine production of TGF-P for some reason.
It will be important in the future to clarify the mechanism
of the effects observed in our studies. The use of purified
progenitor and stem cell populations will allow us to study
apoptosis, growth factor receptor and ligand expression, and
LTRA in response to TGF-P and anti-TGF-P exposure.
These studies may also provide important information about
maintenance and deviation from quiescence of LTRAs,
which is an essential issue in hematologic gene transfer and
ex vivo expansion applications.
RE FER E NCES
1. Roberts A, Spom M: Physiological actions and applications of
transforming growth factor @. Growth Factors 8:1, 1993
2. Nilsen-Hamilton M: Transforming growth factor-@and its actions on cellular growth and differentiation. Curr Topics Develop
Bio 24:95, 1990
3. Jacobsen SEW, Keller JR, Ruscetti FW, Kondaiah P, Roberts
AB, Falk LA: Bidirectional effects of transforming growth factor @
(TGF-@)on colony-stimulating factor-induced human myelopoiesis
in vitro: Differential effects of distinct TGF-P isoforms. Blood
78:2239, 1991
4. Hooper WC: The role transforming growth factor-p in hematopoiesis: A review. Leuk Res 15:179, 1991
5. Wahl SM: TGF-@in inflammation: A cause and a cure. J Clin
Immunol 12:61, 1992
6. Ruscetti FW, Dubois CM, Jacobsen SE, Keller JR: Transforming growth factor @ and interleukin-1: A paradigm for opposing
regulation of haemopoiesis. Baillieres Clin Haematol 5:703, 1992
7. Lahm H, Odartchenko N: Role of transforming growth factor
p in colorectal carcinoma. Growth Factors 9:1, 1993
8. Christ M, McCartney-Francis NL, Kulkami AB, Ward JM,
Mizel DE, Mackall CL, Gress RE, Hines KL, Tian H, Karlsson
S , Wahl SM: Immune dysregulation in TGF-pl-deficient mice. J
Immunol 153:1936, 1994
9. Han ZC, Caen JP: Cytokines acting on committed haematopoietic progenitors. Baillieres Clin Haematol 7:65, 1994
10. Hestdal K, Jacobsen SEW, Ruscetti FW, Longo DL, Boone
TC, Keller JR: Increased granulopoiesis after sequential administration of transforming growth factor-@l and granulocyte-macrophage
colony-stimulating factor. Exp Hematol 21:799, 1993
11. Hayashi S, Gimble JM, Henley A, Ellingsworth LR, Kincade
PW: Differential effects of TGF-@I on lymphohemopoiesis in longterm bone marrow cultures. Blood 74:1711, 1989
12. Eaves CJ, Cashman JD, Ray RI, Dougherty GJ, Otsuka J,
Gaboury LA, Hogge DE, Lansdorp DM, Eaves AC, Humphries R K
Mechanisms that regulate the cell cycle status of very primitive
hematopoietic cells in long-term human marrow cultures. 11. Analy-
sis of positive and negative regulators produced by stromal cells
within the adherent layer. Blood 78:110, 1991
13. Waegell WO, Higley HR, Kincade PW, Dasch JR: Growth
acceleration and stem cell expansion in Dexter-type cultures by neutralization of TGF-P. Exp Hematol 22:1051, 1994
14. Carlino JA, Higley HR, Creson JR, Avis PD, Ogawa Y , Ellingsworth LR: Transforming growth factor @lsystemically modulates
granuloid, erythroid, lymphoid, and thrombocytic cells in mice. Exp
Hematol 20:943, 1992
15. Hatzfeld J, Li M, Brown EL, Sookdeo H, Levesque J, O’Toole
T, Gumey C, Clark SC, Hatzfeld A: Release of early human hematopoietic progenitors from quiescence by antisense transforming
growth factor @loligonucleotides. J Exp Med 174:925, 1991
16. Graham GJ, Wright EG, Hewick R, Wolpe SD, Wilkie NM,
Donaldson D, Lorimore S, Pragnell IB: Identification and characterization of an inhibitor of haemopoietic stem cell proliferation. Nature
344:442, 1990
17. Broxmeyer HE, Sherry B, Cooper S, LU L, Maze R, Beckmann MP, Cerami A, Ralph P: Comparative analysis of the human
macrophage inflammatory protein family of cytokines (chemokines)
on proliferation of human myeloid progenitor cells. J Immunol
150:3448, 1993
18. Keller JR, Bartelmez SH, Sitnicka E, Ruscetti FW,Ortiz M,
Gooya JM, Jacobsen SEW: Distinct and overlapping direct effects
of macrophage inflammatory protein-la and transforming growth
factor p on hematopoietic progenitor/stem cell growth. Blood
84:2175, 1994
19. Dunlop DJ, Wright EG, Lorimore S, Graham GJ, Holyoake
T, Kerr DJ, Wolpe SD, Pragnell IB: Demonstration of stem cell
inhibition and myeloprotective effects of SCYrhMIPla in vivo.
Blood 79:2221, 1992
20. Lord BI, Dexter TM, Clements JM, Hunter MA, Gearing
AJH: Macrophage-inflammatory protein protects multipotent hematopoietic cells from the cytotoxic effects of hydroxyurea in vivo.
Blood 79:2605, 1992
21. Verfaillie CM, Cateanzarro PM, Li WN: Macrophage inflammatory protein la, interleukin 3 and diffusable marrow stromal
factors maintain human hematopoietic stem cells for at least eight
weeks in vitro. J Exp Med 179:643, 1994
22. Verfaillie CM, Miller JS: CD34+/CD33- cells present in
MIP- la +IL3 supplemented ‘ ‘stroma-non-contact’’ cultures contain
more than 20 percent primitive progenitors that can reinitiate longterm cultures. Blood 84:1442, 1994
23. Harrison DE, Jordan CT, B o n g RK, Astle CM: Primitive
hemopoietic stem cells: Direct assay of most productive repopulation
with simple binomial, correlation and covariance calculations. Exp
Hematol 21:206, 1993
24. Bodine DM, Crosier PS, Clark SC: Effects of hematopoietic
growth factors on the survival of primitive stem cells in liquid suspension culture. Blood 78:914, 1991
25. Dasch JR, Pace DR, Waegell W, Inenaga D, Ellingsworth L:
Monoclonal antibodies recognizing transforming growth factor-@.
Bioactivity neutralization and transforming growth factor p-2 affinity purification. J Immunol 142:1536, 1989
26. Till JE, McCulloch EA: A direct measurement of radiation
sensitivity of normal mouse bone marrow cells: Evidence for a preCFU-S cell. Nature 281:381, 1979
27. Dunbar CE, Crosier PS, Nienhuis AW: Introduction of an
activated RAS oncogene into murine bone marrow lymphoid progenitors via retroviral gene transfer results in thymic lymphomas. Oncol
Res 6:39, 1991
28. Bodine DM, McDonagh KT, Seidel NE, Nienhuis AW: Survival and retrovirus infection of murine hematopoietic stem cells in
vitro: Effects of 5-FU and method of infection. Exp Hematol 19:206,
1991
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
MURINE STEM CELL MAINTENANCE AFFECTED BY TGF-8
29. Barker JE, Greer J, Bacon S, Compton ST:Temporal replacement of donor erythrocytes and leukocytes in nonanemic W44jlw44j
and severely anemic W N v mice. Blood 78:1432, 1991
30. Jacobsen SE, Ruscetti FW, Dubois CM, Lee J, Boone TC,
Keller JR: Transforming growth factor-@trans-modulates the expression of colony stimulating factor receptors on murine hematopoietic
progenitor cell lines. Blood 77:1706, 1991
31. Dubois CM, Ruscetti FW, Stankova J, Keller JR: Transforming growth factor-@regulates c-kit message stability and cellsurface protein expression in hematopoietic progenitors. Blood
83:3138, 1994
32. Jacobsen FW, Stokke T, Jacobsen SEW: Transforming
growth factor-@potently inhibits the viability-promoting activity of
stem cell factor and other cytokines and induces apoptosis of primitive murine hematopoietic progenitor cells. Blood 86:2957, 1995
33. Quesniaux VFJ, Graham GJ, Pragnell I, Donaldson D, Wolpe
SD, Iscove N, Fagg B: Use of 5-fluorouracil to analyze the effect
of macrophage inflammatory protein-la on long-term reconstituting
stem cells in vivo. Blood 81:1497, 1993
34. Uchida N, Fleming WH, Alpem EJ, Weissman IL: Heterogeneity of hematopoietic stem cells. Cum Opin Immunol 5:177, 1993
35. Moritz T, Patel VP, Williams DA: Bone marrow extracellular
matrix molecules improve gene transfer into human hematopoietic
cells via retroviral vectors. J Clin Invest 93:1451, 1994
36. Dunbar CE, Cottler-Fox M, O’Shaughnessy JA, Doren S ,
4561
Carter CS, Berenson R, Brown S, Moen RC, Greenblatt J, Stewart
FM, Leitman SF, Wilson W, Cowan KH, Young NS, Nienhuis AW:
Retrovirally-marked CD34-enriched peripheral blood and bone marrow and cells contribute to long-term engraftment after autologous
transplantation. Blood 85:3048, 1995
37. Dunbar CE, Emmons RVB: Gene transfer into hematopoietic
progenitor and stem cells: Progress and problems. Stem Cells 12563,
1994
38. Miller DG, Adam MA, Miller AD: Gene transfer by retrovirus
vectors occurs only in cells that are actively replicating at the time
of infection. Mol Cell Biol 10:4239, 1990
39. Harrison DE, Lemer C P Most primitive hematopoietic stem
cells are stimulated to cycle rapidly after treatment with 5-flUOrOuracil. Blood 78:1237, 1991
40. Knobel KM, McNally MA, Berson AE, Rood D, Chen K,
Kilinski L, Tran K, Okarma TB, Lebkowski JS: Long-term reconstitution of mice after ex vivo expansion of bone marrow cells: Differential activity of cultured bone marrow and enriched stem cell populations. Exp Hematol 22:1227. 1994
41. Williams DA: Ex-vivo expansion of hematopoietic stem and
progenitor cells-Robbing Peter to pay Paul? Blood 81:3169, 1993
42. Peters SO, Kittler ELW, Ramshaw HS, Quesenbeny PJ: Murine marrow cells expanded in culture with IL-3, IL-6, IL-I 1, and
SCF acquire an engraftment defect in normal hosts. Exp Hematol
23:461, 1995
From www.bloodjournal.org by guest on June 17, 2017. For personal use only.
1996 87: 4561-4567
Maintenance of murine long-term repopulating stem cells in ex vivo
culture is affected by modulation of transforming growth factor-beta
but not macrophage inflammatory protein-1 alpha activities
T Soma, JM Yu and CE Dunbar
Updated information and services can be found at:
http://www.bloodjournal.org/content/87/11/4561.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.