[CANCER RESEARCH 45, 2695-2699,
June 1985]
Two Distinct Tumor Cell Growth-inhibiting Factors from a Human
Rhabdomyosarcoma
Cell Line
Charlotte M. Fryling,1 Kenneth K. Iwata,2 Patricia A. Johnson, Walter B. Knott, and George J. Todaro3
Laboratory of Viral Carcinogenesis, National Cancer Institute, NIH [C. M. F., K. K. I., P. A. J., G. J. T.], and Program Resources, Inc., Frederick Cancer Research Facility
[W. B. K.], Frederick, Maryland 21701
ABSTRACT
Tumor cell growth-inhibiting factors (TIFs) have been shown
to inhibit the growth of tumor cell lines in culture. TIF-1, the first
TIF to be described, is a low-molecular-weight, acid- and heatstable polypeptide with no antiviral activity. A second class of
TIFs (TIF-2) has now been isolated from the conditioned media
of a human rhabdomyosarcoma cell line and partially purified by
polyacrylamide gel filtration, cation exchange, and reverse-phase
high-pressure liquid chromatography. Partially purified prepara
tions of TIF-2 inhibit the growth of a variety of human tumor cells
in soft agar and monolayer cultures and are mitogenic for normal
human and mouse cells. TIF-2 has no antiviral activity. The
growth-inhibitory effects of TIF-2 are reversible when the affected
cells are no longer exposed to the factor. Although both TIF-1
and TIF-2 are obtained from the same source, they can be
distinguished by their molecular weight, heat lability, elution
pattern from reverse-phase high-pressure liquid chromatogra
phy, and their effect on the growth of mink lung epithelial cells.
The growth of a human tumor cell variant, selected for resistance
to growth inhibition by TIF-1, is inhibited by TIF-2. TIFs may
therefore be a family of related polypeptides which selectively
inhibit the growth of tumor cells.
MATERIALS AND METHODS
Cell Culture. Cell cultures were maintained at 37°C in 75-sq cm
Falcon No. 3024 tissue culture flasks with complete growth medium
composed of Dulbecco's modification of Eagle's medium with 10% FBS
(Grand Island Biological Co., Grand Island, NY). The cell lines A673, a
human rhabdomyosarcoma, A549, a human lung adenocarcinoma, and
A375, a human melanoma line, have been described previously (8). A375
Ag5 is a soft agar clone of A375. A375 Ag5-IR is a TIF-1 -resistant variant
derived by continually treating subconfluent A375 Ag5 cells with TIF-1.
HuF is a normal human foreskin fibroblast strain received from J. Levy
(Cancer Research Institute, University of California at San Francisco).
NIH/3T3 clone 7 was established in this laboratory.
Conditioned Media. A673 cells were grown to confluence on 850 sqcm roller bottles (Corning 25140) in 50 ml of complete growth medium.
The monolayers were rinsed twice with serum-free Dulbecco's modifi
cation of Eagle's medium and incubated in 50 ml of serum-free Waymouth's medium for 8 h. The medium was discarded and replaced with
50 ml of fresh, serum-free Waymouth's medium, and the cells were
INTRODUCTION
Previous studies have shown that the human rhabdomyosar
coma cell line A673 produces TGFs4 (17) and TIFs) (12). TGFs
(17) and sarcoma growth factors (3) confer the transformed
phenotype on certain normal indicator cells and promote the
anchorage-independent growth of these cells. TGFs also en
hance the growth of tumor cells in soft agar (12). The TGFs
produced by this cell line compete for binding to the EGF receptor
and are classified as TGFa (15). TIF-1, which does not bind to
the EGF receptor, is an acid- and heat-stable, low-molecularweight polypeptide (M, 10,000 to 16,000) that is inactivated by
trypsin (12). Preparations of TIF-1 inhibit the growth of human
tumor cells in soft agar and monolayer culture and stimulate the
growth of normal human fibroblasts and a normal human kidney
epithelial cell line. It blocks the stimulatory effect of both TGFa
and EGF in a soft agar assay with human tumor cells.
A second class of TIFs (TIF-2) has been isolated and partially
1To whom requests for reprints should be addressed.
2 Present address: Oncogene Science, Inc., 222 Station Plaza North, Suite 330,
Mineóla, NY 11501.
3 Present address: Oncogen, 3005 First Avenue, Seattle, WA 98121.
4 The abbreviations used are: TGFs, transforming growth factors; TGFa, type a
TGFs; TGF0, type 0 TGFs; TIFs, tumor cell growth-inhibiting factors; EGF, epider
mal growth factor; rp-HPLC, reverse-phase high-pressure liquid chromatography;
FBS, fetal bovine serum; TFA, trifluoroacetic acid; [125l]ldUrd, 5-[125l]iodo-2'-deox-
incubated for 48 h. The conditioned media were then collected, clarified
by centrifugation, concentrated, dialyzed, and recentrifuged. The super
natant fluid was then lyophilized. The procedure has been described in
more detail elsewhere (12).
Bio-Gel P-100 Chromatography of Concentrated Conditioned Me
dia. Concentrates from 227 liters of conditioned media were chromatographed on several Bio-Gel P-100 columns. The lyophilized samples
were resuspended
g for 30 min, and
(100 to 200 mesh;
at 4°C, collecting
in 10 ml of 1 M acetic acid, centrifuged at 175,000 x
applied to a column (5 x 82.5 cm) of Bio-Gel P-100
Bio-Rad), equilibrated, and eluted with 1 M acetic acid
12.4-ml fractions. Protein was determined for every
other fraction by absorbance at 280 nm. Aliquots of fractions were
collected, lyophilized, and utilized as described below.
CM-cellulose Chromatography.
Fractions from several Bio-Gel P100 columns (total protein, 85 mg) containing TGF activity (M, 18,000 to
22,000) were combined, lyophilized, resuspended in 5 ml of 1 M acetic
acid, and dialyzed overnight against 5 HIM ammonium acetate (pH 4.5)
at 4°C.The dialyzed material was centrifuged at 175,000 x g for 30 min
at 22°C, and the supernatant was placed on a column (1.5x3
cm) of
CM-cellulose (Whatman, CM-23). The column was eluted with a 400-ml
linear salt gradient from 5 HIM ammonium acetate (pH 4.5) to 500 mM
ammonium acetate (pH 6.8) with a flow rate of 80 ml/h at 22°C.Fractions
of 10.3 ml were collected, and conductance was determined using a
conductivity meter (Radiometer, Copenhagen, Denmark). Fraction aliquots were sterilized by adding 0.5 ml of 1 M acetic acid and were
lyophilized.
rp-HPLC. Lyophilized samples from the CM-cellulose purification step
yuridine; CM, carboxymethyl.
Received 8/20/84; revised 1/14/85; accepted 2/2e/85.
CANCER
purified from the same rhabdomyosarcoma cell line. TIF-2 shares
many properties in common with TIF-1 but can be distinguished
by its molecular weight, Chromatographie properties, and its
effect on certain target cells, thus indicating that it is a distinct
TIF. The growth of A673 cells, in particular, and of tumor cells,
in general, may be regulated by the TGFs and TIFs which these
tumor cells themselves produce.
RESEARCH
VOL. 45 JUNE 1985
2695
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1985 American Association for Cancer Research.
TUMOR CELL INHIBITORY
FACTORS
FROM A HUMAN CELL LINE
were resuspended in 1 ml of 0.05% TFA and injected on a Waters Ci8¿iBondapak column (0.39 x 30 cm). A Waters 720 System
was used for the gradient system. The column eluate was
with a variable-wavelength UV detector (Waters, Model 481)
nm. Elution was achieved using a 3-step (15 min, 80 min, 15
Controller
monitored
set at 214
min) linear
-80
- 80
gradient of acetonitrile (0 to 25%, 25 to 45%, and 45 to 100%, respec
tively) in 0.05% TFA at a flow rate of 1.0 ml/min.
A fraction containing TIF-2 activity from a rp-HPLC as described above
was lyophilized, resuspended in 1 ml of 0.05% TFA, and rechromatographed by injection on a Waters ds-MBondapak column (0.39 x 30
cm). Elution was achieved using a 3-step (10 min, 80 min, 10 min) linear
gradient of 2-propanol (0 to 29%, 29 to 37%, 37 to 100%, respectively)
60
o
40
in 0.05% TFA at a flow rate of 1.0 ml/min.
Soft Agar Assay. Lyophilized TIF was resuspended in complete
medium containing 104 A375 Ago cells and 0.34% agar (Difco Noble
20
Agar). This suspension was immediately pipeted over a base layer of
0.5% agar in complete medium in 35-mm Falcon 3001 tissue culture
dishes. Cells were incubated at 37°Cin a 5% CO2-95% air atmosphere,
and photomicrographs were made after 8 days.
TGF Assay. TGF activity was determined by the ability to compete in
a 125l-labeled EGF radioreceptor binding assay (17) and to stimulate the
growth of normal rat kidney cells (NRK 49F) in soft agar.
TIF Assay. Test cells were subcultured (100 ^I/well) in 96-well Nunc
167008 tissue culture plates at a density that allowed continuous growth
throughout the time period of the assay. Most cells were seeded at a
density of 3 x 103 cells/well, while A549 required a seeding density of
4.5 x 103 cells/well. Cells were treated with TIF for 96 h and with [125I]
10
20
30
40
Fraction Number
Chartl. CM-cellulose chromatography of fractions containing TGF activity
pooled from several Bio-Gel P-100 columns. Fractions of 10.3 ml were collected
and conductance was determined using a Radiometer (Copenhagen) conductivity
meter. A 2.5->il aliquot and a 100-/il aliquot from every third fraction were tested
for TIF and TGF activities, respectively (see "Materials and Methods"). A549 was
the cell tested in the TIF assay. Shaded area, TIF-2 activity fractions.
IdUrd (Amersham IM.355V) for the last 24 h as described previously (12).
Inhibition-stimulation of growth was expressed as the percentage of
decrease and/or increase of [125l]ldUrd incorporated by cells treated with
aliquots of TIF, relative to [125l]ldUrd incorporated by the untreated cells.
[3H]Thymidine incorporation and [125l]ldUrd incorporation have given
I
80 l
60 ^
£
40 5
almost identical experimental results. TIF fractions which have given the
most inhibition of [125l]ldUrd incorporation have also given the most
20 I
S
inhibition when tested in a soft agar assay.
Assay for Mitogenic Activity. NIH/3T3 clone 7 and HuF cells were
subcultured in complete growth media at a density of 104 cells/well in
96-well Nunc 167008 tissue culture plates. After incubation for 24 h at
37°C,the cells were serum starved for 72 h by replacing the media with
Waymouth's medium containing 0.1% FBS (100 Ml/well). Cells were
treated with TIF and [125l]ldUrd for 24 h and tested for incorporation of
radioactivity as described (12).
RESULTS
Conditioned media from the human rhabdomyosarcoma cell
line A673, partially purified by Bio-Gel P-100 chromatography,
has been shown to contain both TGFs (17) and TIFs (12).
Fractions with TGF activity (M, 18,000 to 22,000) from the P100 column showed little TIF activity as determined by the TIF
assay (12), using the human lung adenocarcinoma cell line A549.
When the fractions containing TGF activity were subjected to rpHPLC, another peak of TIF activity eluted between approximately
38 and 42% acetonitrile, suggesting that TGF was masking the
effect of TIF in the TIF assay as described previously (12). This
TIF has been designated TIF-2. TIF-1 (M, 10,000 to 16,000),
which has been described previously, élûtes
from rp-HPLC be
tween approximately 28 and 33% acetonitrile (12).
In order to further purify TIF-2, those fractions which contained
TGF activity after Bio-Gel P-100 chromatography were combined
and rechromatographed on a CM-cellulose column (Chart 1) as
described in "Materials and Methods." A major peak of TIF-2
activity was separated from the TGF activity. The fractions in the
shaded area, representing this TIF-2 activity, were pooled, lyCANCER
RESEARCH
10
15
20
25
30
35
Fraction Number
Chart 2. rp-HPLC of TIF-2 purified by CM-cellulose chromatography. A pool of
TIF-2 activity shown in the shaded area of Chart 1 was lyophilized, resuspended
in 1 ml of 0.05% TFA, and subjected to rp-HPLC as described in "Materials and
Methods." Fractions of 3 ml were collected, and 180-^1 aliquots were tested for
TIF inhibition of A549 cells (see "Materials and Methods").
ophilized, and rechromatographed on a reverse-phase Ci8-MBondapak column with a 3-step linear acetonitrile gradient (Chart 2).
The TIF-2 activity eluted between approximately 38 and 42%
acetonitrile, while the majority of the protein eluted at a lower
acetonitrile concentration. A single fraction [Fraction 24 (Chart
2)] which eluted between approximately 38 and 39% acetonitrile
and which contained TIF-2 activity was lyophilized and further
purified by rp-HPLC using a 3-step linear 2-propanol gradient
(Chart 3). TIF-2 eluted between approximately 29 and 31 % and
again was purified away from the majority of the protein.
The HPLC-purified TIF-2 was not inactivated by trypsin. It was
stable at 56°Cbut, after heating to 100°Cfor 3 min, the level of
inhibition of cell growth dropped 30% in a TIF assay. No interferon activity was detected when TIF-2 was tested in an antiviral
interferon assay.
Material from each purification step was tested in the TIF
assay using the human lung adenocarcinoma cell line (A549) and
normal human foreskin fibroblasts (HuF) as the test cells. Growth
inhibition of A549 cells and growth stimulation of HuF cells were
observed at each purification step, as can be seen in Table 1.
The growth-inhibitory effects of TIF-2 seem to be reversible,
VOL. 45 JUNE 1985
2696
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1985 American Association for Cancer Research.
TUMOR CELL INHIBITORY
FACTORS
FROM A HUMAN CELL LINE
diameter. The colony sizes of cells treated with TIF-2 were
O 80
significantly reduced (averaging 0.1 to 0.2 mm in diameter), yet
there was no evidence of cytotoxicity. The small colony growth
that was seen occurred within 1-3 days after treatment, and
80
0.38
-60^
there was no further increase in colony size after 3 weeks. The
human lung adenocarcinoma cell line, A549, was also inhibited
in a soft agar assay. TIF-2 thus inhibits anchorage-independent
growth as well as monolayer cell growth.
TIF-2 (partially purified) was also tested for its effect on other
0.25ÃŒo
3- 40g
O
540
CM
0.12*
£20
5
10
Fraction
15
o.
fi
normal and tumor cells. Table 3 shows representative examples
of the effects of TIF-2 preparations on various cell lines. Although
high concentrations of TIF-2 were used, similar results have been
seen with rp-HPLC-purified material. The growth of human fibro
25
20
Number
Chart 3. rp-HPLCof TIF-2 with a linear2-propanolgradient.A fraction containing
TIF-2 activity which eluted from rp-HPLC between approximately 38 and 39%
acetonitnle was lyophilized, resuspended in 1 ml of 0.05% TFA, and subjected to
rp-HPLC with 2-propanol and 0.05% TFA as described in "Materialsand Methods.*
Fraction aliquots were lyophilized and tested for TIF inhibition of A549 cells (see
'Materials and Methods").
Table 1
ComparisonoÃ-CM-celluloseand HPLC-purified TIF-2from A673-conditioned
media
TIF-2 from each purification step was tested in the TIF assay as described in
"Materials and Methods." The results shown are from a representativeexperiment.
tested
Oig/100/J)
ofHuF150
PurificationCM-celluloserp-HPLC
(acetonitrite
gradient)Amount
15
2.6
0.26
% of inhibition of % of stimulation
A549
57 ±9.9a
27 ±30.0
37 ±15.4
63 ±8.0
40 ±7.4
38 ±14.2
80 ±5.6
37 ±6.1
blasts and epithelial embryonic kidney cells was stimulated. In
contrast, human tumor cell growth was inhibited. When high
tissue culture passages of A549 and A673 cells were tested in
the same assay (data not shown), inhibition of cell growth was
greatly reduced, suggesting that the closer the tumor cell line is
to the primary expiant the more sensitive it is to inhibition of
growth by TIF-2. The effect of TIF-2 on the growth of SV40transformed fibroblasts was examined. Whereas the growth of
normal WI-38 cells (CCL 75) was stimulated by TIF-2, the growth
of WI-38 cells transformed by SV40 (CCL 75.1) was inhibited.
The growth of a SV40-transformed mouse cell line was also
inhibited. Normal rat cell growth is stimulated by TIF-2.
The continuous mink lung cell line (CCL 64) is very responsive
to inhibition of growth by TIF-1 (12). As shown in Table 4,
inhibition of the growth of CCL 64 cells was dose dependent
and detected at very low concentrations of TIF-1. TIF-2, how
" Mean±SDobtainedfromtriplicatewellsassayedinoneexperiment.
TableZ
Mitogen assay with TIF-2
TIF-2 was partially purified by Bio-Gel P-100 and CM-cellulosechromatography.
Cells were serum-starvedfor 3 days with medium containing 0.1% FBS, and assay
was performed as described in "Materials and Methods."
ever, had no significant effect on the growth of these cells. Low
concentrations of both TIF-1 and TIF-2 inhibited the growth of
Tables
Effect of TIF-2on normal and transformed cell lines
incor
TIF-2 was partially purified by Bio-GelP-100 and CM-cellulosechromatography.
of con
All cell lines were tested using the TIF assay (see "Materials and Methods") with
lineNIHdone7HuFTreatmentControlTIF-2FBSControlTIF-2FBSAmount
Cell
tested0.1
poration
(cpm)1699
trol100141185194399228100250426340349
TIF-2 concentrations between 13 and 65 jig in a volumeof 100 pi. Each preparation
178a2396
±
was normalized on the basis of comparable inhibition (approximately 50%) of a
Dug0.52
4872814
±
standard, tow, in vitro passage of A549 cells. The HuF, TOD, and HEK cells have
Mg2.6
9853301
±
been described previously (12). The MCF-7, CCL 64, CCL 75 (WI-38), CCL 75.1
ng13.0/<g10%(v/v)0.52
±3136777
(SV40-transformedWI-38),and NRK49F cell lineswere obtainedfrom the American
6673875
±
Type
Culture Collection, Rockville. MD. All other lines were established in this
677153
±
laboratory. A549, A673 and A2058 cells have been described (17).
[I2Sl]ldUrdincorporation
±84383
±131652
„g2.6
196520
+
lineNormal Cell
inhibition000
of
„g13.0^910%(v/v)[125l]ldUrd
stimulation72
of
±120534
humanHuF
±92%
6.7a121
±
tibroblastTOD
foreskin
a Mean ±SD derived from triplicate wells in a representative experiment.
fibroblastWI-38
adult skin
18.611
±
embryonic lung fibroblast
±7.6
kidneyHuman
HEK p3" embryonic
±10.90000052
21
053
since the replacement of TIF-2 with fresh media leads to renewed
growth of A549 cells. Since preparations of TIF-2 stimulated the
growth of normal fibroblasts, it was also tested in a mitogen
assay (see "Materials and Methods") with serum-starved, quies
tumorA549
carcinomaA673
p17 lung
rhabdomyosarcomaMCF-7
p10
cancerA2058
breast
melanomaNonhumanCCL
p13
cent HuF cells and normal mouse NIH 3T3 clone 7 cells (Table
2). Levels of [125l]ldUrd incorporation were as good as or better
than when cells were treated with 10% FBS.
Tumorigenicity in vivo and anchorage-independent growth in
vitro appear to be closely related characteristics of tumor cells
(1, 13). The human melanoma cell line A375 Ago was treated
with TIF-2 in a soft agar assay, and photomicrographs were
made at 8 days. As shown in Fig. 1, control cells typically grow
to colony sizes measuring an average of 0.5 to 1.0 mm in
CANCER RESEARCH
lungNRK-49F
64 normal mink
kidneySV40normal rat
±1.946
11.471
±
8.355
±
±0.55
±18.2024
7.400
±
transformantsSV40-WI-38
(human)SV
6.936
±
3T3 done 5 (mouse)%
±5.2%
" Mean ±SD of triplicate determinations from a representative experiment.
The number following the letter "p" denotes the passage number of the cell in
tissue culture.
VOL. 45 JUNE 1985
2697
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1985 American Association for Cancer Research.
TUMOR CELL INHIBITORY
FACTORS
Table 4
Comparison of inhibitory activity of TIF-1 and TIF-2 on human lung carcinoma
cells and normal mink lung cells
TIF-2 was partially purified by Bio-Gel P-100 and CM-cellulose chromatography.
TIF-1 was partially purified by Bio-Gel P-100 chromatography Test cells were the
human lung adenocarcinoma, A549, and normal mink lung cells, CCL 64.
inhibition0A5490
of
tested¡iff0.000570.00570.0570.570.00320.0320.323.2%
(i<g/100
6423
3.6C37
±
24.045
±
6.378
±
6.059
±
6.075
±
±15.878
0.76
±
3.68
±
TIF-1TIF-2Concentration
±9.433
10.314
±
±17.050
13.29
±
6.565
±
±17.35±
18.2
±5.9CCL
" As in the TIF assay (see "Materials and Methods").
6 Expressed as percentage of decrease of control [125l]ldUrd incorporation as in
the TIF assay.
c Mean ±SD determined from triplicate wells from a representative experiment.
Table 5
Comparison of tumor inhibitors on A375 Ag5 and TIF-1-resistant A375 Ag5-IR
cells
TIF-1 and TIF-2 were partially purified by Bio-Gel P-100 chromatography. TIF-2
was further purified by CM-cellulose chromatography. Partially purified human
leukocyte Interferon (Lot P-321) was obtained from Meloy Laboratories, Inc.,
Springfield, VA.
inhibitionA375
of
SampleTIF-1
TIF-2
InterferonAmount
tested*1.78MJ9.75
„g
100IU%
Ag5-IR2
Ag544
±8.7"
34 + 10.6
55 ±1.5A375
±6.7
25 ±6.8
54 ±6.8
Concentration per 100 n\ as in TIF assay.
0 Mean ±SD.
A549 cells.
A comparison was made of the effects of exposure to TIF-1,
TIF-2, and leukocyte Interferon on the human melanoma, clonal
line A375 Ag5, and a TIF-1-resistant variant of this line, desig
nated A375 Ag5-IR. To derive this variant, A375 Ag5 cells (1 x
10") were seeded on 24-well Nunc 169590 plates, and the
monolayers were treated with TIF-1 for 10 days. The treated
cells became enlarged, rounded, and no longer attached to the
plate. When transferred to a new tissue culture flask, the cells
eventually settled on the surface and grew as a monolayer. After
4-5 passages, they grew as rapidly as did the parent line, A375
Ag5. When both cell types were plated in soft agar, A375 Ag5
grew to colony sizes averaging 0.5 to 1.0 mm in diameter. In
contrast, the A375 Ag5-IR variant showed almost no growth in
soft agar, with only a few small colonies composed of 3 to 5
cells (averaging 0.1 to 0.2 mm in diameter). Both of these lines
were tested in the TIF assay and, as shown in Table 5, the
growth of the parent line is inhibited by TIF-1, TIF-2, and interferon. The variant remained sensitive to growth inhibition by TÕF2 and Interferon, but growth was not inhibited by TIF-1.
DISCUSSION
The human rhabdomyosarcoma cell line, A673, produces
TGFs and 2 classes of TIF designated TIF-1 (12) and TIF-2.
Partially purified preparations of both TIFs inhibit the growth of
human tumor cells in soft agar and monolayer cultures, stimulate
the growth of normal human cells in monolayer cultures, and are
FROM A HUMAN CELL LINE
mitogenic to normal human and mouse fibroblasts. Although high
concentrations of partially purified TIFs were used in these
experiments, the inhibitory and stimulatory activities of both TIF1 and TIF-2 coelute through several purification procedures.
However, until each factor is purified to homogeneity, it will not
be known if the stimulation of normal cells is due to the action
of TIFs or to the presence of other, unidentified, but perhaps
novel growth factors. Tucker ef al. (18) have recently shown that
purified growth inhibitor from BSC-1 cells and TGF/3 from plate
lets have similar biological activities. Both of these molecules
can stimulate or inhibit growth depending on the experimental
conditions (18). The transformed phenotype may be necessary
for the growth-inhibitory activity of the TIFs. This is demonstrated
by the growth inhibition of virally transformed cells the normal
counterpart of which is stimulated to grow (SV40-transformed
WI-38 cells and normal WI-38 cells).
Both TIFs have no detectable antiviral activity. TIF-2 can be
distinguished from TIF-1 by its apparent molecular weight, trypsin insensitivity, heat lability at 100°C,elution pattern from rpHPLC, and inability to inhibit the growth of certain target cells,
the proliferation of which is inhibited by TIF-1. Although the larger
molecular weight of TIF-2 suggests that perhaps it is a precursor
form of TIF-1, the variable effects of these 2 TIFs on mink lung
and human tumor cell lines indicate that TIF-2 and TIF-1 may be
members of 2 distinct classes of factors which inhibit tumor cell
growth.
The continuous mink lung cell line CCL 64 has been shown to
be very responsive to inhibition of growth by TIF-1 (12) and by
the African green monkey kidney growth inhibitor characterized
by Holley ef al. (9-11,18). In contrast, TIF-2 has little or no effect
on the proliferation of this cell line. This difference in sensitivity
may be explained by the presence of specific receptors on the
cell surface with which TIF-1 interacts but which do not bind TIF2.
Holley ef al. (11) has shown that the African green monkey
kidney BSC-1 inhibitor prevents the stimulation of DNA synthesis
by EOF in BSC-1 cells (11). Similarly, TIF-1 and TIF-2 can
antagonize the effects of TGF« in soft agar and monolayer
cultures. Other growth factors which are produced by trans
formed cells [e.g., platelet-derived growth factor (4, 5, 19) and
insulin-like growth factor (2, 7, 14)] may also interact with TIFs
to control tumor cell proliferation. TIFs may thus be another
important class of autocrine growth regulators (16) which are
secreted by tumor cells.
As shown by Downward ef a/. (6) the close similarity between
the EGF receptor and the v-enb-B oncogene product suggests
other possible mechanisms for triggering proliferative responses.
Since TIFs do not bind to the EGF receptor but can counteract
the stimulatory effects of EGF and TGFa (12), perhaps they can
also block the effects of activated growth factor receptors.
The effects of the TIFs on the variant melanoma cell line
demonstrate that tumor cells can become nonresponsive to one
TIF and yet remain sensitive to another. If tumor cells are capable
of producing 2 or more classes of TIFs, it may be possible that
some aggressive primary or metastatic tumors may be lacking
certain TIFs and/or their receptors or may have acquired resist
ance to the TIFs which they are producing. Determination of
levels of TIF production and identification of tumor cell surface
receptors may thus indicate which class of TIF could have
therapeutic potential against particular tumors.
CANCER RESEARCH VOL. 45 JUNE 1985
2698
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1985 American Association for Cancer Research.
TUMOR CELL INHIBITORY
FACTORS
FROM A HUMAN CELL LINE
1423, 1973.
9. Holley,R. W., Armour, R., Baldwin, J. H., and Greenfield,S. Activity of a kidney
epithelial cell growth inhibitor on lung and mammary cells. Cell Biol. Int. Rep.,
7:141-147,1983.
10. Holley, R. W., Armour, R., Baldwin, J. H., and Greenfield, S. Preparation and
properties of a growth inhibitor produced by kidney epithelial cells. Cell Biol.
Int. Rep., 7:525-526,1983.
11. Holley, R. W., Bohlen, P., Fava, R., Baldwin, J. H., Kleeman,G., and Armour,
R. Purification of kidney epithelialcell growth inhibitors. Proc. Nati. Acad. Sci.
USA, 77:5989-5992,1980.
12. Iwata, K. K., Fryling, C. M., Knott, W. B., and Todaro, G. J. Isolation of tumor
cell growth-inhibiting factors from a human rhabdomyosarcomacell line. Can
cer Res., 45: 2689-2694,1985.
13. Kahn, P., and Shin, S. Cellulartumorigenicity in nude mice: test of associations
among loss of cell-surface fibronectin, anchorage independence,and tumorforming ability. J. Cell Biol., 82: 1-16, 1979.
14. Marquardt, H., Todaro, G. J., Henderson, L. E., and Orozlan, S. Purification
and primary structure of a polypeptide with multiplication-stimulatingactivity
from rat liver cell cultures. J. Biol. Chem., 256: 6859-6865,1980.
15. Roberts, A. B., Frolik, C. A., Anzano, M. A., and Sporn. M. B. Transforming
growth factors from neoplastic and nonneoplastic tissues. Fed. Proc. 42'
2621-2626,1983.
16. Sporn, M. B., and Todaro, G. J. Autocrine secretion and malignant transfor
mation of cells. N. Engl. J. Med., 303: 878-880, 1980.
17. Todaro, G. J., Fryling, C. M., and De Larco, J. E. Transforming growth factors
produced by certain humantumor cells: polypeptidesthat interact with epider
mal growth factor receptors. Proc. Nati. Acad. Sci. USA, 77:5258-5262,1980.
18. Tucker, R. F., Shipley, G. D., Moses, H. L., and Holley, R. W. Growth inhibitor
from BSC-1 cells closely related to platelet type tf transforming growth factor.
Science (Wash. DC), 226: 705-707,1984.
19. Waterfield, M. D., Scrace, G. T., Whittle, N., Stroobant, P., Johnsson, A.,
Wasteson, A., Westermark, B., Heldin, C., Huang, J. S., and Deuel, T. F.
Platelet-derivedgrowth factor is structurally related to the putative transform
ing protein p28*sof simian sarcoma virus. Nature (Lond.), 304: 35-39, 1983.
ACKNOWLEDGMENTS
We thank Dr. Raoul E. Benveniste for helpful suggestions and assistance with
the manuscript.
REFERENCES
1. Cifone, M. A., and Fidler, I. J. Correlationof patterns of anchorage-independent
growth with in vivo behavior of cells from a murine fibrosarcoma. Proc. Nati.
Acad. Sci. USA, 77: 1039-1043,1980.
2. De Larco, J. E., and Todaro, G. J. A human fibrosarcoma cell line producing
multiplication stimulating activity (MSA)-relatedpeptides. Nature (Lond.), 272:
356-358, 1978.
3. De Larco, J. E., and Todaro, G. J. Growth factors from murine sarcoma virustransformed cells. Proc. Nati. Acad. Sci. USA, 75. 4001-4005, 1978.
4. Deuel, T. F., Huang, J. S., Huang, S. S., Stroobant, P., and Water-field,M. D.
Expression of a platelet-derived growth factor-like protein in simian sarcoma
virus transformed cells. Science (Wash. DC), 22Õ:1348-1350, 1983.
5. Doolittle, R. F., Hunkapiller, M. W., Hood, L. E., Devare, S. G., Robbins, K. C.,
Aaronson, S. A., and Antoniades, H. N. Simian sarcoma virus one gene, v-s/s,
is derived from the gene (or genes) encoding a platelet-derivedgrowth factor.
Science (Wash. DC), 227: 275-276, 1983.
6. Downward, J., Yarden, Y., Mayes, E., Scrace, G., Totty, N., Stockwell, P.,
Ullrich,A., Schlessinger,J., and Waterfield, M. D. Close similarity of epidermal
growth factor receptor and v-erò-Boncogene protein sequences. Nature
(Lond.), 307:521-527, 1984.
7. Dulak, N. C., and Temin, H. M. Multiplication-stimulatingactivity for chicken
embryo fibroblasts from rat liver cell conditioned medium: a family of small
polypeptides. J. Cell. Physiol., 81:161-170, 1973.
8. Giard, D. J., Aaronson, S. A., Todaro, G. J., Amstein, P., Kersey, J. J., Dosik,
H., and Parks, W. P. In vitro cultivation of human tumors: establishmentof cell
lines derived from a series of solid tumors. J. Nati. Cancer Inst., 5): 1417-
Fig. 1. Effect of TIF-2 on the growth of human melanoma cells (A375 Ag5) in soft agar. A, no treatment; B, TIF-2 (64 /jg/ml) after CM-cellutose chromatography
x84.
CANCER
RESEARCH
VOL. 45 JUNE 1985
2699
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1985 American Association for Cancer Research.
Two Distinct Tumor Cell Growth-inhibiting Factors from a
Human Rhabdomyosarcoma Cell Line
Charlotte M. Fryling, Kenneth K. Iwata, Patricia A. Johnson, et al.
Cancer Res 1985;45:2695-2699.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/45/6/2695
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1985 American Association for Cancer Research.