From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
Transforming Growth Factor PI Directly and Reversibly Inhibits the Initial
Cell Divisions of Long-Term Repopulating Hematopoietic Stem Cells
By Ewa Sitnicka, Francis W. Ruscetti, Gregory V. Priestley, Norman S. Wolf, and Stephen H. Bartelmez
Hematopoiesis appears t o be regulated, in part, by a balance
between extracellular positive and negative growth signals.
Transforming growth factor beta-I (TGF-p,) has been shown
t o be a negative regulator of primitive hematopoietic cells.
This study examined the direct effect of TGF-p, on the proliferation and differentiation of long-term repopulating hematopoietic stem cells (LTR-HSCI in vitro. We previously reported a cell fractionation approach that includes the
selection of low Hoescht 33342/low Rhodamine 123 (low
Ho/Rh) cell fractions that are highly enriched for long-term
repopulating cells (LTR-HSC) and also clone t o a very high
efficiency inthe presence of stem cell factor (SCF) + interleukin-3 (IL-3) 11-6 90% t o 100% of individually cultured low
Ho/Rh cells formed high proliferative potential clones. This
high cloning efficiency of an LTR-HSC enriched cell population enabled proliferation inhibition studies t o be more easily interpreted. In this report, we show that the continuous
presence of TGF-p, directly inhibits the cell division of essen-
tially all low Ho/Rh cells (in a dose-dependent manner) during their 0 t o 5th cell division in vitro. Therefore, it follows
that TGF-p, must directly inhibit the proliferation of LTRHSC contained within these l o w Ho/Rh cells. The time required for some low Ho/Rh cells t o undergo their first cell
division in vitro was also prolonged in the presence of TGFpl. Furthermore, when low Ho/Rh cells were exposed t o
TGF-P, for varying lengths of time before neutralization of
the TGF-p, by monoclonal antibody, the ability t o form macroclones was markedly decreased after -4 days of TGF-p,
exposure. In addition, 1 t o 10 ng/mL of TGF-pl resulted in a
maintenance of high proliferative potential-colony-forming
cell (HPP-CFC) during 8 days of culture compared with loss
of HPP-CFC in cultures with no added TGF-p,. In conclusion,
this study shows that TGF-P, directly inhibits the initial
stages of proliferation of LTR-HSC and appears t o slow the
differentiation of daughter cells of low Ho/Rh cells.
0 1996 b y The American Society of Hematology.
M
been found in the media of long-term marrow cultures.”
The addition of exogenous TGF-PI to long-term cultures
essentially arrested hematopoiesis.’6 Conversely, the addition of TGF-PI neutralizing antibodies into long-term cultures increased the number of burst-forming unit-erythroid
(BFU-E) and granulocyte-macrophage colony-forming cell
(GM-CFC) in S-pha~e.’~.’’
In vivo administration of TGFP I , either systemically or locoregionally, also resulted in a
suppression of hematopoie~is.’~~’~
Furthermore, depending
on the target cell, TGF-PI can enhance or inhibit proliferation. Early myeloid and erythroid progenitors are markedly
inhibited in the presence of TGF-0,
whereas more differentiated murine myeloid progenitors forming granulocyte,
macrophage, granulocyte-macrophage colonies, and erythroid progenitors forming CFU-e are not inhibited by TGFPI.20.22-24 These and other studies suggest, but do not directly
demonstrate, that the proliferation of hematopoietic stem
cells may be directly inhibited by TGF-PI.”.” This study
uses an LTR-HSC selection procedure that yields cell fractions that are both highly enriched for LTR-HSC and that
also clone at very high efficiency (90% to 100%) in the
presence of specific growth factors.’” Among other findings,
we show that TGF-PI directly inhibits the proliferation of
the LTR-HSC contained in these cell fractions.
+
ANY RECENT studies have reported the isolation
and characterization of hematopoietic stem cells with
long-term repopulating ability.’.’I The ability to culture single cells from cell fractions highly enriched for long-term
repopulating hematopoietic stem cells (LTR-HSC) has led
to the identification of specific extracellular signals that are
direct mediators of this small, but vital cell compartment.
The maintenance of the stem cell pool may be accomplished by mechanisms such as asymmetrical or symmetrical
cell division of a stem cell that results in the production of
one or two identical daughter stem cells. Transforming
growth factor beta-1 (TGF-0,) has been shown to be an
important regulator of hematopoiesis and, therefore, may be
involved as a mediator of stem cell differentiation. TGFPI has been shown to regulate cell differentiation during
embryogenesis, angiogenesis, and myogenesis.” It has been
shown by in situ hybridization and immunochemical staining
that TGF-PI is produced in organs of hematopoiesis including bone marrow and fetal
TGF-PI mRNA has been
detected in marrow stromal cells and TGF-P, protein has
From the Department of Pathology, Medical School, University
of Washington, Seattle, WA; and Laboratory of Leukocyte Biology.
Biological Response Modifiers Program, National Cancer InstituteFrederick Cancer Research and Development Center, Frederick,
MD.
Submitted October 3, 1995; accepted February 22, 1996.
Supported in part by Grants No. DK 48708-02 awarded by the
National Institute of Diabetes and Digestive and Kidney Diseases,
and AG 01 751 awarded by the National Institute of Aging, National
Institutes of Health, Department of Health and Human Services.
Address reprint requests to Stephen H. Bartelmez, PhD, Seattle
Biomedical Research Institute, Four Nickerson St, Seattle. WA
98109.
The publication costs ofthis 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.
0 1996 by The American Society of Hematology.
0OO6-4971/96/8801-0006$3.00/0
82
,’”’
MATERIALS AND METHODS
Animals. In all experiments male and female F, hybrids from
the cross C57B1/6 X DBN2 were used. Three- to six-month-old
mice came from our own National Institute of Aging (NIAj-derived
breeding colony, which is maintained under strict specific pathogenfree conditions and is routinely tested and confirmed to be free of
all known mouse pathogens.”
Bone marrow ceZls. Mice were killed, femurs and tibias were
removed aseptically and repetitively flushed of marrow with phosphate-buffered saline (PBS) with 10% CPD (citrate phosphate dextrose solution), 1% fetal bovine serum (FBS) (Biocel, Rancho
Dominguez, CA), 50 U/mL DNAse. This solution was used as collection medium and for cell resuspension through the lineage depletion steps. The cell suspension was drawn three times through
successively smaller bore needles, and finally passed through a 100Blood, Vol 88,No 1 (July l ) , 1996:pp 82-88
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
a3
TGF-0, INHIBITS HEMATOPOIETIC STEM CELLS
mesh stainless steel screen to obtain a single cell suspension. The
low-density bone marrow cells were isolated on a gradient by layering 5-mL aliquots of lo7 cells/mL over 3.0 mL Nycodenz (1.085
g/mL) (Nycomed, Oslo, Norway). Cells were centrifuged 20 minutes
at 400g collected from the interface and washed twice with collection medium at 4°C.
Lineage committed cell depletion and vital dye staining. During
antibody incubation and magnetic bead depletion of lineage committed precursors, cell suspensions were held at 4°C. Rat antimouse
monoclonal antibodies were used in a titrated mixture. Anti-7/4 (neutrophils and activated macrophages) was a gift from Professor Saimon Gordon (Oxford, UK), anti-B220 (B and pre-B lymphocytes),
and anti-YW25.12.7 (erythroid) were a gift from Dr Ivan Bertoncello
(Peter MacCallum Cancer Institute, Melbourne, Australia). AntiL3T4 (T-helper) and anti-Lyt-2 (T cells) were purchased from Becton Dickinson (Mountain View, CA). Ter 119 was kindly provided
by Dr Tatsuo Kina (CDRI, Kyoto University, Kyoto, Japan). A total
of lo8 cells/mL was incubated with an equal volume of the antibody
mixture for 20 minutes at 4"C, washed over a I-mL fetal bovine
serum (FBS) gradient, then rosetted twice with magnetic beads
coated with sheep antirat IgG (Dynabeads M-450, Dynal, Inc, Great
Neck, NY) at a bead:cell ratio of 5:l in the following manner: lo8
cells/mL (4 mL maximum) were transferred to a 50-mL centrifuge
tube (Coming, Coming, NY), and an equal volume of 5 X the number
of magnetic beads was added dropwise to the cell suspension and
gently mixed with a pipette. The suspension was immediately centrifuged at 20g for 3 minutes, then resuspended, transferred to a 5-mL
polypropylene tube and placed into the Dynal magnet. Cells not
bound with the magnetic beads remained in suspension and were
carefully removed (this fraction was designated lineage-negative
cells). Cells were counted and resuspended in 10 p m o m Hoechst
33342 (Ho) (Sigma Chemical, St Louis, MO) in Hanks' balanced
salt solution (HBSS) with 20 mmolR. HEPES, 1 g/L glucose, 10%
FBS, adjusted to pH 7.2 with NaHC03, and incubated at 37°C for
1 hour. Forty minutes into the Ho incubation, Rhodamine 123 (Rh123) (Sigma) at 0.1 mg/mL final concentration was added to the cell
suspension. After dye incubation, the cells were chilled to 4°C
washed twice with PBS + 1% FBS,resuspended in PBS + 1% FBS
+ propidium iodide (2 pg/mL for detection of dead cells and residual
viable neutrophilic promyelocytes/metamyelocytes), and sorted.
Fluorescence-activated cell sorter selection of LTR-HSC enriched
cellfractions. Cell sorting of the lineage depleted preparation was
performed on an Ortho Cytofluorograf model 50W2151 (Ortho Diagnostic System, Westwood, MA) equipped with two spatially separated argon-ion lasers (model 90-6, Coherent Corp, Palo Alto, CA
and model 164-05 Spectra Physics, Mountain View, CA). Monochromatic light at 351 to 364 nm was used for Ho excitation and
fluorescence emission detected at 420 nm using a long pass filter.
Rh-123 excitation was at 488 nm and fluorescence emission was
detected at 515 to 535 nm. After exclusion of propidium iodide (PI)
positive cells, previously published settings" were used to successively select a 10% blast cell window based on high forward, low
right angle scatter, the lowest 3% of the Ho fluorescence histogram,
and the lowest 15% of the Rh-123 fluorescence histogram. Sorted
cells (henceforth designated low Ho/Rh cells) were directly and
individually deposited into 96-well plates containing the culture medium.
Culture conditions. Single sorted cells were cultured in 96-well
U-bottomed plates (Coming) in Iscove's modified Dulbecco's medium (IMDM) (GIBCO-BRL, Grand Island, NY) supplemented with
cytokines and with 12.5% FBS, 12.5% horse serum (HS) (GIBCO),
m o m 2-mercaptoethanol (2-ME) (Sigma),
mom
2 X
hydrocortisone (HC) (Sigma), and antibiotics (penicillidstreptomycin) (GIBCO). In all experiments, cells were cultured for 2 to 3
weeks. The use of U-bottom plates facilitated the settling of single
cells to the bottom center, which allowed direct observation. Clones
ranging from 2 to 64 cells could be directly enumerated. The number
of cells per well were directly counted using a phase contrast inverted
microscope at 200 x magnification. Where indicated, single cell
survival was determined by trypan blue exclusion coupled with an
ability to highly refract light. Cell morphology was estimated on
slides stained by the Giemsa-Wright method.
HPP-CFC and GM-CFC assays. A double layer nutrient agar
culture consisting of a 1-mL underlayer of 0.5% agar plus hematopoietic growth factors: rrSCF (50 ng/mL), rmIL-3 (115 ng/mL),
rhIL-1 (20 ng/mL), rhCSF-1 (500 U/mL), and 0.5 mL overlay of
0.3% agar plus target cells in 35-nun Petri dishes was used as previously described." Cultures were incubated at 37°C in a 5% 02,
5% COz, 90% Nz gas mixture. HPP-CFC and GM-CFC were enumerated at 14 days using a dissecting microscope at 10 X magnification or an inverted microscope at 25 X magnification, respectively.
HPP colonies were identified as macroscopic colonies (>50,000
cells/colony) with diameters > 1.O mm with a dense center.
Sources of cytokines. Purified, recombinant growth factors were
kindly and generously provided as follows: rat SCF from Dr Krisztina M. Zsebo, Amgen Inc, Thousand Oaks, CA; mouse IL-3 from Dr
Andrew Hapel, Australian National University (Canberra, Australia)
(the activity of IL-3 was 5 U per 1 ng); human IL-6 (rhuIL-6) from
Dr Douglas Williams, Immunex Corp, Seattle, WA; recombinant
bovine TGF-Bl was a gift of Tony Purchio (Bristol Myers, Squibb,
Seattle, WA); specific monoclonal antibodies anti-TGF-PI
ID.1 1.1627and 2C~l-1.12'~were previously described. The concentrations of cytokines used in experiments are described in the Results
section.
RESULTS
TGF-P, directly inhibits cell division of essentially all low
Ho/Rh cells cultured as single cells. Low Homh cells were
directly sorted into 96-well plates containing saturating concentrations of SCF, IL-3 and IL-6, and varying concentrations of TGF-B, . The cloning efficiency (defined as the percentage of visually verified single cells that divided at least
once) and the clone size were determined after 14 days of
culture. In the absence of TGF-PI, 96% 2 5% (mean t
standard deviation [SD]) of cells divided at least once (Table
I). When TGF-PI was added at a concentration of 1, 10, or
50 ng/mL, 14% to 20% of low Homh cells were unable to
undergo even the first cell division in the presence of TGFPI (Table 1). However, even though the remainder of cells
underwent cell division in the presence of TGF-PI, they
were unable to complete no more than 4 to 5 cell divisions
(Tables 1 and 2). This inhibitory effect was strictly dependent on the presence of TGF-PI. At 10 and 50 n g h L of
TGF-PI -95% of dividing clones were at 2 to 32 cell stage
at day 7 of the culture compared with 10% at the 2 to 32
cell stage in the control (Table 1). Moreover, in cultures
with no TGF-PI added, 75% of clones went through more
than seven cell divisions (>128 celldclone) by day 7,
whereas, in cultures with TGF-PI, less than 10% of proliferating clones reached that clone size (Table 1).
The clone size analysis at day 14 of the culture showed
that most of the cells that were growth arrested earlier (at
day 7) by TGF-P, did not escape this inhibitory effect of
TGF-P, . When cells were grown without TGF-PI, the average maximum clone size was > 100,000 cells per clone (Table 2). The mean clone sizes at 14 days of the culture with
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
84
SlTNlCKA ET AL
Table 1. Effect of TGF-p, on the Cloning Efficiency and Kinetics of Clonal Profieration of IndividuallyCultured Low Ho/Rh Cells
TGF-8,
Concentration
Cloning
Efficiency
% Clones With
2-32 Cells at Day 7
% Clones With
32-128 Cells at Day 7
0
1 ng/mL
10 ng/mL
50 ng/mL
96% i 5%
86% t 12%*
80% i l l % t
80
10 t 6
94 5 l l t
95 f 17t
92
15 t 6
Ot
Ot
0
% Clones With
128 Cells at Day 7
75 r 7
6 -t 5 t
5 i- 4t
8
Ho/Rh cells were directly sorted to the wells containing medium containing the following growth factors: 50 ng/mL of SCF, 20 ng/mL of IL-6,
and 115 ng/mL of IL-3. Single cells were verified microscopically. The number of single cells that divided at least once divided by the total
single cells plated ( X l O O ) = cloning efficiency. Data represent the mean t SD of 16 experiments with 0 and 1 ng/mL TGF-p, (45 to 60 clones
per experiment), at 10 ng/mL mean i SD of 5 experiments and at 50 ng/mL of a single experiment. In the absence of growth factors, the
addition of TGF-0, at concentrations of 1, 10, or 50 ng/mL did not induce cell division, and cell death ensued within 48 hours.
*P<.Ol.
t P < ,001.
TGF-PI were 8 ? 11, 10 2 9 and 12 2 9 cells (at concentrations of TGF-PI: 1, 10, 50 ng/mL, respectively) (Table 2).
In all experiments, some cells were able to 'escape' the
inhibitory effect of TGF-PI, especially at lower concentrations (Table 2 ) . However, at TGF-PI concentrations >10
ng/mL, essentially all low Ho/Rh cells were directly arrested
as single cells or after the first few cell divisions.
Clone formation from single low Ho/Rh cells is maximally
inhibited by TGF-P, only when added between the 1 to 8
cell stage. When TGF-PI was added to cultures of single
low Homh cells at 1 to 3 days, clonal proliferation was
inhibited in a similar manner as observed with TGF-PI addition at day 0 (Tables 3 and 4). During this culture period,
clones varied from 1 to 4 cells. However, when TGF-PI was
added at day 4 of the culture, the inhibitory effect of TGFPI was lessened, and when TGF-PI was added at day 7, the
inhibitory effect was only partial (Tables 3 and 4). Further-
Table 2. Effect of TGF-pl on the Clone Size of Individually
Cultured Low HolRh Cells
TGF-0,
Concentration
% Clones With
> 100,000
Cells at Day 14
Ave. Max.
Clone Size
(#cells)*
at Day 14
96 t 4
16 t 13*
5 t 3*
2
>100,000
8 t 11*
10 2 9*
12 i 9*
% Clones
That Escaped
lnhibitiont
more, an analysis of the clone size at the time of TGF-PI
addition indicated that daughter cells produced after the -8
cell stage had an increased probability of 'escaping' TGFPI-mediated inhibition. For example, as the average clone
size increased from 1.14 rt 0.35 at day 2 to 7.23 .+- 3.59 at
day 4 to 93.4 ? 43.8 at day 7 the proportion of clones that
escaped TGF-PI inhibition increased from 0% to 2% to 63%,
respectively (Table 3).
TGF-P, delays the time to thejrst cell division of single
low Ho/Rh cells. As described above, 80% to 86% of low
HoRh cells underwent one or more cell divisions in the
presence of a wide range of TGF-PI concentrations (I to 50
ng/mL). The time required to undergo the first cell division
is another major parameter that could affect stem cell usage
in vivo, and so we determined the time to first cell division
of low Ho/Rh cells in the presence or absence of TGF-PI
(see Table 5 and Fig 1). In the absence of TGF-PI, 95% of
low Ho/Rh cells underwent their first cell division between
day 2 and 4 of culture and only 5% of dividing clones
completed their first division between day 5 and 10. In conTable 3. Effect of TGF-p, on Daughter Cells of Individually Cultured
Low Ho/Rh Cells: Effect on Macroclone Formation
~
0
1 ng/mL
10 ng/mL
50 ng/mL
NA
16 ? 13*
5 ? 3*
2
Ho/Rh cells were directly sorted to the wells and cultured as indicated in Table 1. In the absence of growth factors, the addition of
TGF-p1 at concentrations of 1 , 10, or 50 ng/mL did not induce cell
division, and cell death ensued within 48 hours. Data represents the
mean t SD of 16 experiments with 0 and 1 ng/mL TGF-0, (45 to 60
clones per experiment), at 10 ng/mL mean t S D of 5 experiments,
and at 50 ng/mL of a single experiment.
Abbreviations: Ave. Max., average maximum; NA, not applicable.
* Mean clone size at 14 days t SD.The clones that "escaped TGFp1inhibition" were excluded in this analysis. The number of clones
analyzed for 0, 1, IO, and 50 ng/mL TGF-0, was 60, 89, 51, and 39,
respectively.
t "Clones that escape inhibition" not only proliferated in the presence of TGF-PI, but usually attained macroscopic clone size (r90%
of escaping clones).
P < ,001.
*
TGF-0,
Concentration
Time of TGF-8,
Addition (days)
0
1 ng/mL
10 ng/mL
1 ng/mL
10 ng/mL
1 ng/mL
10 ng/mL
1 ng/mL
10 ng/mL
1 ng/mL
10 ng/mL
1 ng/mL
10 ng/mL
0
0
0
1
1
2
2
3
3
4
4
7
7
No. of Cells/Well
When TGF-0,
Added
% Clones With
> 100,000
Cells at 14 Days
1
1
1
100
1
2
1
1.14 t 0.35
0
1.12 i 0.3
2.98 t 1.32
2.46 2 1.12
7.23 ? 3.59
7.06 t 4.28
93.4 t 43.8
97.7 t 41.04
0
0
2
0
0
0
2
2
63
54
~~
Single cells were verified microscopically. All cultures contained 50
ng/mL of SCF, 115 ng/mL of IL-3, 20 ng/mL of IL-6 ? TGF-01 ( 1 and
10 ng/mL) at the days indicated. Data are from a single experiment;
however, two additional experiments yielded similar results (data not
shown). Mean t SD, n = 50 to 60 clones per time point.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
85
T W O 1 INHIBITS HEMATOPOIETIC STEM CELLS
Table 4. Effect of TGF-p, on Daughter Cells of Individually Cultured
Low HolRh Cells: Effect on the Kinetics of Clonal Proliferation
Time of
TGF-0,
96 Clones
% Clones
With 32-128
Cells at Day
7
% Clones
With >128
Cells at
Day 7
39
22
18
26
25
32
17
49
39
61
37
32
30
51
0
0
2
2
0
3
1
0
11
12
50
57
TGF-8,
Concentration
Addition
(days)
With 2-32
Cells at
Day 7
0
1 ng/mL
10 ng/mL
1 ng/mL
10 ng/mL
1 ng/mL
10 ng/mL
1 ng/mL
10 ng/mL
1 ng/mL
10 ng/mL
1 ng/mL
10 ng/mL
0
0
0
1
1
2
2
3
3
4
4
7
7
10
78
82
72
73
68
80
49
69
28
44
18
13
~
~
~~
,,i
50
NS
c
t!
81
0
~~
~
96 2 5
86 2 12
~~
the First Cell
Division at
Days 2-4
the First Cell
Division at
Days 5-10
9 5 2 11
73 2 12NS
524
27 2 16"
_
_
~
Low Ho/Rh cells were individually deposited from the cell sorter in
96-well-U-bottomed plates containing culture medium (IMDM) with
12.5% FBS, 12.5% HS, HC, 2-ME, antibiotics (see Materials and Methods) and growth factors as follows: 115 ng/mL of IL-3/mL, 20 ng of
IL-G/mL, 50 ng of SCF/mL and indicated concentration of TGF-8,. Data
represent mean values 2 SD from 3 to 12 independent experiments,
and in each 20 to 50 clones were analyzed per group. Cells were
cultured for 2 weeks, and the number of cells in each well was counted
under an inverted microscope every day. The difference in growth
conditions was analyzed using Student's t-test.
Abbreviation: NS, not significant.
* P < ,001.
5-7
4
8-10
Days in Culture
The effect of TGF-PI on the differentiation of low Ho/
Table 6. Neutralization of the lnhibkory Effect of TGF-p, by
Specific Anti-TGF-p, Monoclonal Antibody: Effect of the Duration
of TGF-p, Exposure on Individually Cultured Low Ho/Rh Cells
TGF01
Added
Type of
Ab
Added
Time Ab
Addition
(days)
-
-
-
aTGF-01
Control
0
0
+
+
+
+
Table 5. Effect of TGF-p, on the Time of the First Cell Division of
Individually Cultured Low Ho/Rh Cells
% Clones With
3
in TGF-PI for 1 to 8 days, which became progressively less
able to form clones containing > 100,000cells after the TGFPI was neutralized (Table 6).
periods of time, then tested for their ability to subsequently
proliferate by neutralizing the TGF-P1. The proliferative potential (assessed as clone size) of single low Homh cells
exposed to TGF-PItended to decrease over time in replicate
experiments. In the absence of TGF-PI, 94% of clones attained > 100,000 cells, compared with single cells cultured
% Clones With
2
Fig 1. Effect of TGF-p, on the time of the first cell division of
low Ho/Rh cells in single cell cultures. Low Ho/Rh sorted cells were
cultured in the presence of 115 nglmL 11-3, 20 ng/mL IL-6, and 50
ng/mL SCF only (0).(B) Represent clones cultured with the same
cytokines plus TGF-p, at 1 ng/mL. Means f SD are shown. "Represents P < .01; * * P < .001; NS, not significant.
The exposure of single low Ho/Rh cells to TGF-0, for
longer than 4 days altered their ability to proliferate in the
presence of growth factors. Single low HOD& cells were
cultured with SCF, IL-3,IL-6 plus TGF-PI for increasing
SCF, IL-3, IL-6
+1 ng/mL TGF-pl
T
10
trast, the addition of 1 ng/mL, of TGF-PI delayed the first
cell division of a substantial proportion of cells; 73% of Ho/
Rh cells divided between day 2 and 4, whereas in the presence of TGF-P1, 27% cells divided between day 5 and 10.
Growth Factors
2olk
c
Single cells were verified microscopically. All cultures contained 50
ng/mL of SCF, 115 ng/mL of IL-3, 20 ng/mL of IL-6 2 TGF-p, (1 and
10 ng/mL) at the days indicated. Data are from a single experiment,
however, two additional experiments yielded similar results (data not
shown).
Cloning
Efficiency
NS
+
+
~
+
+
+
-
aTGF-01
Control
LYTGF-~~
0
0
1
2
3
4
5
8
Cloning
Efficiency
%Clones
With >lOO,M)O
Cells
83
75
68
60
84
58
64
78
53
70
71
55
93
92
80
0
94
0
93
81
80
64
47
25
Low Ho/Rh cells were individually deposited from the cell sorter in
96-well-U-bottomed plates containing culture medium (IMDM) with
12.5% FBS, 12.5% HS, HC, 2-ME, antibiotics (see Materials and Methods) and growth factors as follows: 115 ng/mL of IL-3/mL, 20 ng of
IL-G/mL, 50 ng of SCF/mL, and 1 ng/mL of TGF-0,. Anti-TGF-p, antibodies were added at indicated days of the culture at the final concentration 20 mg/mL. Data shows values from single representative experiments, in which 20 to 50 clones were analyzed per group. Cells
were cultured for 2 weeks, and the number of cells in each well was
counted under an inverted microscope every day.
Abbreviation: Ab, antibody.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
86
SlTNlCKA ET AL
Rh cells in culture: Maintenance of HPP-CFC. As shown
below, TGF-PI inhibited proliferation of low HoRh cells
(Table 7). The same inhibitory effect was observed when
low HoRh cells were grown in bulk cultures (70 cells per
culture) with TGF-P1. In the presence of IL-3, IL-6, and
SCF without TGF-PI, low Ho/Rh cells proliferated from 70
f 6 cells to 300,000 cells within 8 days. However, this rapid
proliferation was combined with differentiation: at day 8 the
number of cells forming HPP-CFC declined to 0 compared
with 78% at day 0. In contrast, when 1 ng/mL of TGF-PI
was present, 90 f 22 HPP-CFC were recovered after 8 days
of the culture compared with 55 f 11 HPP-CFC present at
day 0. Thus it appears that TGF-PI inhibits differentiation,
as well as proliferation of low Ho/Rh cells.
The effect of TGF-PI on the de novo generation of GMCFC from Ho/Rh cells. In the same experiment described
above, we quantitated the total cell production and total number of GM-CFC generated after 8 days in cultures seeded
with 70 f 6 low Homh cells exposed to SCF, IL-3, and IL6 plus TGF-PI varying from 0.04 to 10 ng/mL (Table 8).
GM-CFC were not detectable in freshly sorted low Ho/Rh
cell fractions, however significant numbers were generated
after 8 days in all cultures containing growth factors. Even
though a wide range of TGF-PI concentrations (10 to 0.4
ng/mL) severely inhibited total and GM-CFC cell production, the proportion of cells that were GM-CFC remained
stable at -10%. As the concentration of TGF-PI was further
decreased, total GM-CFC production significantly increased
above cultures in which no TGF-PI was added. At the same
time, the relative proportion of GM-CFC compared with
other (mostly more differentiated cell types) decreased to
-4%, as the concentration of TGF-PI was further decreased
to 0%.
DISCUSSION
Previous studies have clearly shown that TGF-PI directly
inhibits the proliferation of a substantial proportion of the
cells in fractions highly enriched for mouse primitive hema-
Table 7. Effect of TGF-p, o n t h e Differentiation of L o w HolRh Cells
in Culture: Maintenance of HPP-CFC
Total HPP Forming
Cells per Well at 8
Days in Culture
Total No. of Cells
per Well at 8
Days in Culture
0
0
0.004 ng/mL
0.04 ng/mL
0.2 ng/mL
1.0 ng/mL
10.0 ng/mL
0
0
0
300,000
300,000
225,000
100,000
44,000
4,500
Amount of
TGF-0, Added
at Day 0
90 i 22
28 2 5
Seventy Ho/Rh cells were cultured per well in the presence of 50
ng/mL SCF, 115 ng/mL IL-3, and 20 ng/mL IL-6 for 8 days, after which,
cells were counted and assayed for HPP-CFC. The values represent
data from a representative experiment. At day 0, each culture of 70
c 6 l o w Ho/Rh cells contained 55 ? 11 HPP-CFC. At day 8 of the
culture, cells were assayed for HPP-CFC at 5,000 cells plated per dish
i n three replicates. No inhibition of HPP development was observed
even in the presence of greaterthan 170 GM-CFC per dish. Data shown
are calculated per total number of cells.
Table 8. Effect of TGF-p, o n t h e Differentiation of Low Ho/Rh Cells
in Culture: Effect on De Novo GMCFC Production
TGF-0,
Added at
Day 0
0
0.04 ng
0.2 ng
0.4 ng
1.0 ng
10.0 ng
Total GM-CFC per
Well at 8 Days in
Culture
Total No. of
CellsNVell a t
8 Days in
Culture
% Cells
GM-CFC
12,000 -e 500
20,000 % 800*
17,000 % 2,000t
13,000 c 300 NS
5,600 2 140*
514 c 60*
300,000
300,000
225,000
100,000
44,000
4,500
4 2 0.2
7 f 3NS
8 c 0.9t
13 2 0.3*
13 t 0.3t
11 t 1.3’
Seventy Ho/Rh cells were cultured per well i n the presence of 50
ng/mL SCF, 115 ng/mL IL-3, and 20 ng/mL IL-6 for 8 days, after which,
cells were counted and assayed for GM-CFC. The values represent
data from a representative experiment. GM-CFC were not detectable
at day 0 of culture. At day 8 of the culture, cells were assayed for GMCFC at 5,000 cells plated per dish in three replicates. Data shown are
calculated per total number of cells.
Abbreviation: NS, not significant.
P < ,001.
t P < .01.
topoietic cells. However, the direct inhibition of LTR-HSC
by TGF-PI had not been shown.1s,17,”,22,25
In the present
study, we first show that TGF-PI will directly inhibit the
proliferation of essentially all individually cultured low Ho/
Rh cells. Given that this same cell fraction is highly enriched
for LTR-HSC,” it follows that the LTR-HSC selected by
this purification method are directly inhibited by TGF-PI.
This conclusion is further strengthened by the results of
our more recent characterization of low Ho/Rh cells that
demonstrates that as few as five low Homh cells would
generate long term (>1 year) chimeras that varied from 1%
to 32% of donor-derived lymphohematopoietic cells (Wolf
et al, unpublished results, August 1995).
A closer look at the clonal cell stage at which the cells
became proliferation arrested by TGF-PI shows a heterogeneity of the response. For example, at 10 ng/mL TGF-PI,
low HoRh cells became arrested at -1 to 16 cell stage (10
2 9 cells). Several mechanisms for this inhibition by TGFPI have been suggested. Direct inhibition could occur by a
sustained downregulation of cell surface of cytokine receptors on bone marrow cells. TGF-PI has been shown to downregulate the expression of receptors for IL-1,1L-3, GM-CSF,
G-CSF, and SCF.29-32
Many studies have shown that TGFPI blocks cell cycle transit at the GUS check point. Although
the signal transduction pathways that mediate the inhibitory
effects of TGF-PI remained to be defined, TGF-PI can maintain the retinomablastoma (RB)and related gene products
in the hypophosphorylated state and block cells in G 1.33,34
Furthermore, TGF-PI reduces the expression of CdK4 and
cyclin E and subsequently reduces the activation of both
cyclin D-CdK4 and cyclin E-CdK2 complexes necessary for
G1 tran~it.~’.’~
In addition, several inhibitors of the action of
cyclin-Cdk complexes have been cloned. p21 (Wafl, Cipl,
Sdil), a universal Cdk inhibitor is transcriptionally activated
by ~ 5 3 . ”p27, structurally related to p21, is upregulated in
TGF-PI -treated epithelial cell lines and lymphocytes, binds
to and inactivates both cyclin D-CdK4 and cyclin E-CdK2
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
87
TGF-8, INHIBITS HEMATOPOIETIC STEM CELLS
complexes.35A third potential effector of cell cycle arrest, ~ 1 5 , ~ ' 2. Spangrude GJ, Scollay R A simplified method for enrichment
of mouse hematopoietic stem cells. Exp Hematol 18:920, 1990
is transcriptionally activated by TGF-PI in a human keratino3. Spangrude GJ, Smith L, Uchida N, &uta K, Heimfeld S, Fridecyte cell lines. Thus, growth arrest may involve several pathman J, Weissman IL: Mouse hematopoietic stem cells. Blood
ways that can be modulated by TGF-P', the specific activation
78:1395, 1991
of which depends on the type of cell, stage of activation, and
4. Lemischka IR: The hematopoietic stem cell and its clonal progdifferentiation. Not relevant to this study, but relevant in vivo,
eny: Mechanisms regulating the hierarchy of primitive hematopoietic
indirect inhibition could be mediated by decreased cytokine
cells. Cancer Surv 15:3, 1992
production from stromal cells. TGF-PI has been shown to de5 . Vos 0,Ploemacher RE:Developments in moderm hematology.
crease the synthesis of LIP," tumor necrosis factor (TIW)Boll Soc Ita1 Biol Sper 67:435, 1991
( Y , ~
SCF?' and interferon (IFN)-Y."~This heterogeneity in in6. Uchida N, Aguila HL, Fleming WH, Jerabek L, Wiessman IL:
Rapid and sustained hematopoietic recovery in lethally irradiated
hibitory mechanisms probably explains the heterogeneity in
mice transplanted with purified Thy- 1.110 Lin-Scal + hematopoietic
response seen in the low Ho/Rh cell population.
stem cells. Blood 83:3758, 1994
In contrast to the inhibition of proliferation of freshly
7. Neben S, Redfeam WJ, Parra M, Brecher G, Pallavicini MG:
sorted low Ho/Rh cells by continuous exposure to TGF-P, ,
Short- and long-term repopulation of lethally irradiated mice by bone
daughter cells generated in the presence of growth factors
marrow stem cells enriched on the basis of light scatter and Hoechst
alone rapidly became refractory to TGF-PI inhibition. This
33342 fluorescence. Exp Hematol 19:9, 1991
study shows that clones up to -8 cell stage (-3 cell division)
8. Bartelmez SH, Andrews RG, Berstein ID: Uncovering the hetwere essentially all growth arrested by delayed addition of
erogeneity of hematopoietic repopulating cells. Exp Hematol 19:861,
TGF-PI. However, the majority of clones at the 100 cell
1991
stage (-6 to 7 cell divisions) contained daughter cells that
9. Jones RI,Wagner JE, Celano P,Zicha MS, Sharkis SJ: Separation of pluripotent hematopoietic stem cells from spleen colony
proliferated in the presence of TGF-PI.
forming cells. Nature 347:188, 1990
Other studies have shown a differential sensitivity to TGF10. Li CL, Johnson GR: Rhodamine 123 reveals heterogeneity
PI depending on the maturation stage of the target
within
murine Lin-, Sca-1 + hematopoietic stem cells. J Exp Med
Thus, the sensitivity of hematopoietic cells to inhibition by
175:1443, 1992
TGF-PI might be used as an indicator of the primitiveness
11. Wolf NS, Kone A, Priestley GV, Bartelmez SH: In vivo
(stemness) of fractionated hematopoietic cell populations. It
and in vitro characterization of long-term repopulating primitive
has been reported before2' that TGF-PI promoted increased
hematopoietic cells isolated by sequential Hoechst 33342-rhodamine
GM-CSF-induced GM colony formation and enhanced neu123 FACS selection. Exp Hematol 21:614, 1993
trophilic differentiation. In this report, the effect of TGF12. Rizzino A: Transforming growth factor p: Multiple effects
PI on generation of GM-CFC from Ho/Rh cell population on cell differentiation and extracellular matrices. Dev Biol 130:411,
enriched for long-term repopulating cells was determined.
1988
13. Ellingworth LR, Brennan JE, Fok K, Rosen DM, Bentz H,
The addition of low doses of TGF-PI resulted in both a
Piez KA, Seyedin SM: Antibodies to the N-terminal portion of cartimaintenance of HPP-CFC and an increase in the total numlage-inducing factor A and transforming growth factor p. Immunober of GM-CFC generated. This indicates that TGF-PI inhibchemical localization and association with differentiating cells. J
ited not only proliferation of primitive hematopoietic stem
Biol Chem 261:12362, 1986
cells, but also slowed the differentiation of daughter cells.
14. Nemunaitis J, Tompkins CK, Andrews DF, Singer JW: TransOther studies have shown that TGF-PI modulates cell differforming growth factor p expression in human marrow stromal cells.
entiation during embryogenesis, angiogenesis, and myogenEur J Haematol 46:140, 1991
esis," however, effects of TGF-PI on hematopoietic differ15. Cashman JD, Eaves AC, Raines EW, Ross R, Eaves CJ:
entiation have not previously been reported.
Mechanisms that regulate the cell cycle status of very primitive
The effect of prolonged TGF-0, exposure on low Ho/Rh
hematopoietic cells in long-term human marrow cultures. I. Stimulacells was also measured by neutralization of the TGF-PI and
tory role of a variety of mesenchymal cell activators and inhibitory
role of TGF-0, . Blood 75:96, 1990
observing the ability of the cell to proliferate in response to
16. Hayash SI, Gimble JM, Henley A, Ellingsworth LR, Kincade
growth factors. The results indicate that after 4 days the
PW: Differential effects of TGF-PI on lymphopoiesis in long-term
previously arrested clones tended to either not proliferate at
bone marrow cultures. Blood 74:1711, 1989
all or form only microscopic clones. This apparent growth
17. Eaves JE, Cashman JD, Kay RJ, Dougherty GJ, Otsuka T,
nonresponsiveness may be mediated by a persistent downGaboury LA, Hogge DE, Lansdorp PM, Eaves AC, Humphries RK:
regulation of cytokine receptors by TGF-p132-35
and/or inMechanisms that regulate the cell cycle status of the very primitive
volve other mechanisms discussed above that may prevent
hematopoietic cells in long-term human marrow cultures. 11. Analycells from progressing through the cell ~ y ~ l e . ~ ~ * ~ sis
~ of
- ~positive and negative regulators produced by stromal cells
Thus, it appears that TGF-PI, a naturally occurring and
within the adherent layer. Blood 78:110, 1991
ubiquitous cytokine in the hematopoietic microenviron18. Migdalska A, Molineux G, Demuynck H, Evans GS, Ruscetti
ment,I3'l4plays a major role in restraining primitive hematoFW,Dexter TM: Growth inhibitory effects of transforming growth
poietic stem cells from entering active cycling and in modufactor-pl in vivo. Growth Factors 4:239, 1991
lating the rate of proliferation and differentiation among their
19. Goey H, Keller JR, Back T, Long0 DL, Ruscetti FW, Wiltrout
RH: Inhibition of early murine hematopoietic progenitor cell prolifimmediate generations of daughter cells.
-
REFERENCES
1. Spangrude GJ, Heimfeld S, Weissman IL:Purificationand characterization of mouse hematopoietic stem cells. Science 241:58, 1988
eration after in vivo locoregional administration of transforming
growth factor-pl. J Immunol 143:877, 1989
20. Ruscetti FW,Jacobsen SE, Birchenall-Roberts M, Broxmeyer
HE, Engelmann GL, Dubois C, Keller JR: Role of transforming
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
88
growth factor-PI in regulation of hematopoiesis. Ann NY Acad Sci
628:31, 1991
21. McNiece IK, Bertoncello I, Keller JR, Ruscetti FW, Hartley
CA, Zsebo KM: Transforming growth factor P inhibits the action
of stem cell factor on mouse and human hematopoietic progenitors.
Int J Cell Cloning 10:80, 1992
22. Keller JR, Bartelmez SH, Sitnicka E, Ruscetti FW,Oriz M,
Gooya JM, Jacobsen SEW: Distinct and overlapping effects of macrophage inflammatory protein- 1 alpha (MIP- l a ) and transforming
growth factor-beta (TGF-P) on hematopoietic progenitorktem cell
growth. Blood 84:2175, 1994
23. Ottmann OG, Pelus LM: Differential proliferative effects of
transforming growth factor-p on human hematopoietic progenitor
cells. J Immunol 140:2661, 1988
24. Keller JR, Jacobsen SE, Sill KT, Ellingsworth LR, Ruscetti
FW: Stimulation of granulopoiesis by transforming growth factor
P: Synergy with granulocyte/macrophage-colony-stimulatingfactor.
Proc Natl Acad Sci USA 88:7190, 1991
25. Keller JR, McNiece IK, Sill KT, Ellingsworth LR, Quesenberry PJ, Sing GK, Ruscetti FW: Transforming growth factor P
directly regulates primitive hematopoietic cell proliferation. Blood
75:596, 1990
26. McAlister I, Wolf NS, Pietrzyk ME, Rabinovitch PS, Priestley
GV, Jaeger B: Transplantation of hematopoietic stem cells obtained
by a combined dye method fractionation of murine bone marrow.
Blood 75:1249, 1990
27. Dasch JR, Pace DR, Waegell W, Inenaga D, Ellingsworth
LR: Monoclonal antibodies recognizing transforming growth factorB: Bioactivity, neutralization and TGF-P2 affinity purification. J Immunol 142:1536, 1989
28. Lucas C, Mora-Worms M, Fendly B, Bald L, Figari I, Palladin0 M: The autocrine production of transforming growth factor PI
during lymphocyte activation: A study with a monoclonal antibody
based Elisa. J Immunol 145:1415, 1990
29. Dubois CM, Ruscetti FW, Palaszynski EW, Falk LA, Oppenheim JJ, Keller JR: Transforming growth factor P is a potent inhibitor
of interleukin 1 (IL- 1) receptor expression: Proposed mechanism of
inhibition of IL-1 action. J Exp Med 172:737, 1990
30. Jacobsen SEW, Ruscetti FW, Dubois CM, Lee J, Boone TC,
Keller JR: Transforming growth factor-p trans-modulate 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-p regulates c-kit message stability and cellsurface expression in hematopoietic progenitors. Blood 83:3 138,
1994
SlTNlCKA ET AL
32. de Vos S, Brach MA, Asano Y, Wolf-Dieter L, Bettelheim
P, Gruss H-J, Herrmann F: Transforming growth factor-PI interfers
with the proliferation-inducing activity of stem cell factor in myelogenous leukemia blasts through functional down-regulation of the ckit proto-oncogene product. Cancer Res 53:3638, 1993
33. Pietenpol JA, Stein RW, Moran E, Yaciuk P, Schlegel R.
Lyons RM, Pittelkow MR, Munger K, Howley PM, Moses HL:
TGF-P, inhibition of c-myc transcription and growth in keratinocytes
is abrogated by viral transforming proteins with pRB binding domain. Cell 62:777, 1990
34. Laiho M, DeCaprio JA, Ludlow JW, Livingston DM, Massague J: Growth inhibition by TGF-P linked to suppression of retinoblastoma protein phosphorylation. Cell 62: 175, 1990
35. Polyak K, Kato J, Solomon MJ, Sherr CJ, Massague J , Roberts JM, Koff A: p27Kip1, A cyclin-Cdk inhibitor, links transforming
factor 0 and contact inhibition to cell cycle arrest. Genes Dev 8:9,
1994
36. Koff A, Ohtsuki M, Polyak K, Roberts JM, Massague J:
Negative regulation of GI in mammalian cells: Inhibition of cyclin
E-dependent kinase by TGF-PI . Science 260536, 1993
37. Ewen ME, Sluss HK, Whitehouse LL, Livingston DM: TGFPI inhibition of Cdk4 synthesis is linked to cell cycle arrest. Cell
74:1009. 1993
38. Datto MB, Li Y, Panus JF, Howe DJ, Xiong Y, Wang XF:
Transforming growth factor induces the cyclin-dependent kinase
inhibitor p21 through a p53-independent mechanism. Proc Natl Acad
Sci USA 92:5545, 1995
is a potential effector of TGF39. Hannon GJ, Beach D: p15INKB4
P,-induced cell cycle arrest. Nature 371:257, 1994
40. Chantry D, Turner M, Abney E, Feldmann M: Modulation
of cytokme production by transforming growth factor 0. J Immunol
142:4295, 1989
41. Heinrich MC, Dooley DC, Koeble WW: Transforming growth
factor PI inhibits expression of the gene products for steel factor
and its receptor (c-kit). Blood 85: 1769, 1995
42. Espevik T, Figari IS, Shalby MR, Lackides GA, Lewis GD,
Shepard HM, Palladino MA Jr: Inhibition of cytokine production
by cyclosporin A and transforming growth factor p. J Exp Med
166571, 1987
43. Hatzfeld J, Li M-L, Brown EL, Sookedo H, Levesque J-P,
O’Toole T, Gumey C, Clark SC, Hatzfeld A: Release of early human
hematopoietic progenitors from quiescence by antisence transforming factor PI or Rb oligonucleotides. J Exp Med 174:925, 1991
44. Howe PH, Draetta G, Leof EB: Transforming growth factor
01 inhibition of p34cedc2 phosphorylation and histone Hl kinase
activity is associated with Gl/S-phase growth arrest. Mol Cell Biol
11:1185. 1991
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1996 88: 82-88
Transforming growth factor beta 1 directly and reversibly inhibits the
initial cell divisions of long-term repopulating hematopoietic stem
cells
E Sitnicka, FW Ruscetti, GV Priestley, NS Wolf and SH Bartelmez
Updated information and services can be found at:
http://www.bloodjournal.org/content/88/1/82.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.