Basic Fibroblast Growth Factor Confers Growth Inhibition ...and Dentistry of New Jersey-New Jersey Medical School, Newark, New Jersey 07103 [R. W., H. W.] ABSTRACT The effect of basic

Vol. 3, 135-142, January 1997 Clinical Cancer Research 135

Basic Fibroblast Growth Factor Confers Growth Inhibition and

Mitogen-activated Protein Kinase Activation in Human

Breast Cancer Cells1

Eyal Fenig, Robert Wieder, Shoshana Paglin,

Huisheng Wang, Roger Persaud,

Adriana Haimovitz-Friedman, Zvi Fuks,

and Joachim Yahalom2Department of Radiation Oncology, Memorial Sloan-Kettering CancerCenter, New York, New York 10021 [E. F., S. P., R. P., A. H-F.,z. F., J. Y.], and Department of Medicine, University of Medicineand Dentistry of New Jersey-New Jersey Medical School, Newark,New Jersey 07103 [R. W., H. W.]

ABSTRACT

The effect of basic fibroblast growth factor (bFGF)

on human breast cancer cells was studied in vitro. Expo-sure to bFGF resulted in significant growth inhibition,

decreased DNA synthesis, and accumulation of cells in

G0-G1. The IC50 for growth inhibition in MCF-7 cells was50 pglml, and it was abrogated by neutralizing antibodies

against bFGF. Inhibition of growth by bFGF was pre-

dominant over the growth stimulatory effects of 17�-estradiol, insulin, or epidermal growth factor. Binding

and cross-linking studies of ‘25I-labeled bFGF in intactMCF-7 cells demonstrated 5.2 X iO� saturable bFGF

binding sites per cell, a dissociation constant of 57 pM, and

a Mr 142,000 ‘25I-labeled bFGF cross-linked protein.

Stimulation of MCF-7 cells with bFGF at concentrationswhich effected growth inhibition also resulted in activa-

tion of p42maPk (ERK2) and �44maPk (ERK1) mitogen-

activated protein kinases. These data demonstrate that

whereas bFGF inhibits the growth of several breast can-

cer cell lines, it concomitantly activates ERK1 and ERK2,generally considered to signal mitogenic rather thangrowth inhibitory responses. Whether there is association

between these phenomena remains unknown.

Received 3/5/96; revised 9/10/96; accepted 10/9/96.The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.1 This work was supported in part by a Career Development Award fromthe U. S. Army Breast Cancer Research Program (DAMDI7-94-J-4463;to R. W.) and a grant from the Foundation of the University of Medicineand Dentistry of New Jersey (16-95; to R. W.).2 To whom requests for reprints should be addressed, at Departmentof Radiation Oncology, Memorial Sloan-Kettering Cancer Center,1275 York Avenue, New York, NY 10021. Phone: (212) 639-5999;Fax: (212) 639-7742; E-mail: [email protected] or [email protected].

INTRODUCTIONbFGF,3 also known as FGF-2, is a member of a family of

growth and angiogenic factors ubiquitously expressed in normal

and malignant tissues (1, 2). bFGF controls the growth of

several normal and malignant mesodermal and neuroectoderm-

derived cells (3) and malignant epithelial cells (4-7). Its mito-

genic activity is mediated via binding to specific high-affinity

cell surface receptors and activation of their tyrosine kinase

domains (8-12). The downstream signaling was reported to

involve the proline-directed serine/threonine MAPKs, ERKI

(�44maPk) and ERK2 (�42maPk. Refs. 8-12). Activation of

MAPKs was shown to be required for the mitogenic response of

bFGF in fibroblasts (13, 14).

MCF-7 is a hormone-responsive human breast cancer cell

line (15). Several hormones and growth factors stimulate

MCF-7 growth, including 17�3-estradiol, insulin, and EGF (4, 6).

In serum-free systems, bFGF was reported to produce a growth

stimulatory effect on MCF-7 cells (5, 6). However, in MDA-

MB- 134, a human breast cancer cell line that overexpresses the

FGF receptors 1 and 4, acidic FGF, or bFGF were reported to

inhibit cell growth (16). In the present study, we demonstrate

that bFGF inhibits the growth of several serum-fed MCF-7 cell

sublines as well as other breast cancer cell lines (MDA-MB-453

and T47-D). Exposure of MCF-7 cells to bFGF also activated

ERK1 and ERK2, but whether the latter effect was associated

with bFGF-induced inhibition of MCF-7 cell growth remains

unknown.

MATERIALS AND METHODS

Materials. Human recombinant bFGF (157 amino ac-

ids), antihuman bFGF polyclonal IgG, and normal goat IgG

were purchased from R&D Systems, Inc. (Minneapolis, MN).

Na’ 251 and ECL were purchased from Amersham Corp. (Ar-

lington Heights, IL), and DSS was obtained from Pierce Chem-

ical Co. (Rockford, IL). All reagents used for SDS-PAGE were

purchased from Bio-Rad (Melville, NY). [methy!-3HjThymidine

(5 Ci/mmol) was purchased from New England Nuclear (Bos-

ton, MA). Heat-inactivated FCS was obtained from Hyclone

Laboratories, Inc. (Logan, UT), and all tissue culture media

were purchased from Life Technologies, Inc. (Grand Island,

NY). Insulin was obtained from Eli Lilly (Indianapolis, IN).

Immobilon P membrane was purchased from Millipore Corp.

(Marlborough, MA), anti-MAPK (ERK2) monoclonal antibody

3 The abbreviations used are: bFGF, basic fibroblast growth factor;MAPK, mitogen-activated protein kinase; EGF, epidermal growth fac-tor; DSS, dissacinimidyl suberate; BAEC, bovine aortic endothelial cell;PMSF, phenylmethylsulfonyl fluoride; PKC, protein kinase C; ECL,enhanced chemiluminescence.

Research. on February 20, 2021. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

136 Growth Inhibition of Breast Cancer Cells by bFGF

was purchased at United Biomedical, Inc. (Lake Placid, NY),

antiphosphotyrosine monoclonal antibody was purchased from

Transduction Laboratories (Lexington, KY), protein G-plus/

protein A-agarose was purchased from Oncogene Science

(Uniondale, NY). Fetal human recombinant EGF and all chem-

icals were purchased from Sigma Chemical Co. (St. Louis,

MO). All chemicals were reagent grade.

Cell Culture. MCF-7, T-47D, and MDA-MB-453 cell

lines were obtained from the American Type Culture Collection

(Rockville, MD). Cells were cultured in phenol red-containing

DMEM (high glucose). The medium was supplemented with 2

mM glutamine, 5% heat-inactivated FCS, penicillin (50 units!

ml), streptomycin (50 p�gi’ml), and insulin (10 IU!liter). In some

experiments, when the effect of estrogen deprivation on cellgrowth was studied, the cells were cultured in phenol red-free

DMEM supplemented with 5% dextran-coated charcoal-

stripped heat-inactivated serum (17). Cell cultures were main-

mined at 37#{176}Cin 5% CO2-humidified incubators. Clonal pop-

ulations of BAECs were established from the intima of a bovine

aorta (18). These cells were cultured in DMEM containing 10%

FCS and maintained in 37#{176}Cin 10% C02-humidified incubators.

Cell Growth Studies. bFGF stock solutions were dis-

solved in PBS containing 0.1% human serum albumin, kept

frozen at -20#{176}C, and used only once after thawing. MCF-7

monolayers were dispersed with trypsin-EDTA, counted in a

Coulter Counter (Coulter Electronic, Inc. Hialeah, FL), and

plated in 35-mm dishes, at an initial density of 2 X i0� cells!dish, in 2 ml of standard growth medium. The medium was

replaced 24 h later with growth medium containing bFGF or the

vehicle as described in Fig. 1 . The medium was replaced there-

after with fresh medium of the same composition every 2 days.

Thymidine Incorporation Assay. Cells (1.6 X i0�

cells!well) were incubated in 24-well plates (Falcon, Oxnard,

CA) in standard medium for 24 h. The medium was changed to

the experimental conditions, and the resulting mitogenic activity

was measured by pulsing with [methyl-3H]thymidine. Each well

received 1 p.Cilwell on day 4 for 2 h of incubation. Incorporated

thymidine was determined by scintillation spectroscopy of tri-chloroacetic acid extracts of cells dissolved in 0.1 N NaOH.

Cell Cycle Analysis. The effect of bFGF on cell cycle

distribution of subconfluent MCF-7 cells was determined by

analysis of DNA content using flow cytometry as described

previously (19). In short, monolayers were dispersed with tryp-

sin-EDTA and the cells were fixed in 70% ethanol and washed

with PBS. Cellular RNA was digested with 160 p.g!ml of RNase

A. After washing, DNA was stained with 50 �.ag!mi of pro-

pidium iodide. The content of stained DNA was then determined

by using a FACScan flow cytometer (Becton Dickinson, Moun-

tam View, CA). Cell cycle analysis of the DNA histograms was

carried out with the Multicycle Computer program (Phoenix

Flow Systems, San Diego, CA) developed by P. S. Rabinovitz

(University of Washington, Seattle, WA).

lodination of bFGF. Recombinant bFGF was iodinated

with chloramine T as described previously (20, 21). Biological

activity of ‘�I-labeled FGF was determined by its ability to

stimulate growth of cultured BAECs (3).

High-Affinity Binding of ‘�I-labeled bFGF. Subcon-fluent cultures (10� MCF-7 cells in 16-mm wells) were pre-

cooled to 4#{176}Cand washed once with cold PBS. Binding buffer

[25 mz�t HEPES (pH 7.5) and 0.2% gelatin in DMEMJ was

added to each well followed by addition of ‘25I-labeled FGF to

the desired concentration. Nonspecific binding was determinedin the presence of 100-fold excess of unlabeled bFGF. The cells

were incubated for 2 h at 4#{176}C.At the end of the incubation, cells

were washed twice with PBS supplemented with 0.1% BSA and

once with 2 M NaCl in 20 mz�i HEPES (pH 7.5) to remove the

low-affinity bound bFGF (22). Cells were then lysed in PBS

containing 1% Triton X-100, and cell-associated radioactivity

was determined in a gamma counter (Gamma 5500; Beckman

Instruments, Fullerton, CA).

Cross-Linking of 1�I-labeled bFGF to Its Binding Sites.

Cells were grown to confluence in 10-cm plates and washed

twice with cold PBS. The cells were then incubated for 2 h at

4#{176}Cin binding buffer containing ‘25I-labeled bFGF (5 ng!ml).

After the incubation, cells were washed with cold PBS supple-

mented with 0.1% BSA. Bound ‘25I-labeled bFGF was cross-

linked by adding DSS to a final concentration of 0.15 m� for 20

mm at room temperature. The reaction was terminated by ad-

dition of 1 mM Tris-HC1 (pH 7.5) and 150 m�i glycine for 5 mm

followed by a wash with cold PBS. Cells were scraped from the

plate, centrifuged, and resuspended in 40-60 p.1 of lysis buffer

[10 mr�i Tris-HC1 (pH 7.0), 0.5% NP4O, 0.1 nmi EDTA, and 1

mM PMSF) for 10 mm at 4#{176}C.After centrifugation at 12,000

rpm in an Eppendorf microcentrifuge for 5 mm, the proteins in

the supernatant were resolved on 7.5% SDS-polyacrylamide

gels. The gels were fixed, dried, and exposed to Kodak X-

OMAT autoradiography film at -70#{176}C for detection of 1251..

labeled bFGF cross-linked peptides.

Determination of ERK1 and ERK2 Phosphorylation.

Standard culture media were replaced with serum-free DMEM,

2 mr�i glutamine, and 0.5% BSA fraction V (Calbiochem, La

Jolla, CA) 24 h before stimulation with bFGF. Cells were then

removed from the plates by a 10-15-mn incubation in PBS

containing 1 mM EDTA (pH 8.0) at 37#{176}C.Following centrifu-

gation, cells were incubated at 37#{176}Cin a shaker bath with the

indicated concentrations of bFGF for various time periods. After

10 mm of centrifugation at 2000 rpm at 4#{176}C,the cells were

prepared for Western blotting or for immunoprecipitation as

follows: for Western blotting (23), cells were disrupted by

sonication in 200 p.1 of lysis buffer [20 m�i Tris-HC1 (pH 7.4),

1 mM EGTA, 50 l.LM NaVO4, 50 nmi NaF, 0.01 unit/mI leupep-

tin, and 1 mM PMSF]. The proteins were resolved on 12.5%

SDS-polyacrylamide gels, transferred onto Immobilon P mem-branes, immunobloned with antiphosphotyrosine monoclonal

antibody, and visualized with the ECL system. The membranes

were then stripped with 2% SDS!100 m’vi 2-mercaptoethanol in

62.5 mr�t Tris-HC1 (pH 6.7), reblotted with anti-ERK2 antibody,

and visualized as above. For immunoprecipitation, the cellswere lysed in KLB buffer (1% Triton X-lOO, 0.05% SDS, 10

mM Na2HPO4, and 150 mr�i NaC1) passing through a 26-gauge

needle. The proteins (400 p.g/0.5 ml) were heated at 100#{176}Cfor

10 mm, and after cooling were incubated at 4#{176}Cwith 1 jig of

anti-ERK2 monoclonal antibody overnight. The immune corn-

plexes were incubated with 15 pA of protein G plus!protein

A-agarose for 2 h at 4#{176}C.After washing twice with 1 ml of KLB

without SDS and twice with 50 m�i Tris (pH 7.5), the immune

complexes were resolved on SDS-polyacrylamide gels and im-

rnunoblotted as described above.



10�

-0--

1

-0--U

Fig. 1 Inhibition of growth of MCF-7 cells bybFGF. Cells (2 X l0�/thsh) were seeded into35-mm dishes containing standard (STD) or hor-mone-deprived gmwth media (HDM) with orwithout bFGF. Culture media containing freshbFGF or control (0.1% albumin) were replacedevery other day. Points, means of triplicates; bars,SD.

standard growth condltlon(STD)

STD+bFGF(5Opg/mI)STD+ bFGF(500pg/ml)hormons.deprlvsd medlum(HDU)

H DM+bFGF(500pglmI)hiU

.0

Ez

U0

0 2 4 6 8

Time in Culture(days)

Clinical Cancer Research 137

Determination of ERK1 and ERK2 Enzymatic Activity.

bFGF was added to the tissue culture dishes for 10 mm. At the

end of the incubation, the cells were washed twice with PBS,

scraped into lysis buffer [20 rnr�t Tris (pH 8.0), 40 mr�i Na4P2O7,

50 mM NaF, 5 mM MgCl2, 100 p.M Na3VO4, 10 nmi EDTA, 1%

Triton X-lOO, 0.5% sodium deoxycholate, 0.1% SDS, 20 mg!ml

leupeptin, 20 mg/mi aprotonin, and 3 mM PMSF}, and incubated

on ice at 4#{176}Cfor 10 mm. The lysate was centrifuged at 10,000 X

g for 10 mm and boiled after adding sample buffer to a final

concentration of 62.5 nmi Tris (pH 6.8), 2.3% SDS, 5 m�

EDTA, 10% glycerol, and 100 nmi DTF. ERK1 and ERK2

activity was determined by an in-gel kinase assay as described

previously (24, 25). Briefly, the proteins were resolved on a

10% SDS-polyacrylamide gel containing 0.5 p.g!ml myelin ba-

sic protein. The gel was washed for 1 h in each of the following

solutions: twice in 20% 2-propanol in 50 mi�i Tns-HC1 (pH 8.0),

once in 5 mM 2-mercaptoethanol in 50 nmi Tris-HC1 (pH 8.0),

and twice in 6 M guanidine-HC1. The gel was then washed for

16 h with three changes of 5 rn�i 2-mercaptoethanol and 0.04%

Tween 40 in 50 nmi Tris-HC1 (pH 8.0). This was followed by

incubation with 100 ml of phosphorylation buffer [25 rni�i

HEPES (pH 7.4), 10 rni� MnCl2, and 250 pCi of [-y-32PJATP]

for 3 h. After incubation, the gel was washed successively with

water, 40 rn�i HEPES (pH 7.4), and 1% sodium PP1 in 40 nmi

HEPES (pH 7.4). The gel was vacuum dried and exposed to

autoradiography. Phosphorylated myelin basic proteins associ-

ated with proteins in the lysates which comigrated with immu-

noprecipitated ERK1 and ERK2 were cut, and their associated

radioactivity was measured in a scintillation counter.

Statistical Analysis of Data. High-affinity binding pa-rameters of �‘I-labeled bFGF were determined using Scatchard

analysis (26). Student’s t test was used to analyze the differences

between mean values.

RESULTSEffect of bFGF on Growth and DNA Synthesis in Hu-

man Breast Cancer Cells. Fig. 1 shows that bFGF inhibits

0

��C0�gL�Un;C�.��,.-

300 j200�

IIiI

i001�l�i�

�1-I�I�0I .1

U

�

bFGF�

Control Insulin Estradlol EGF

Fig. 2 Effect ofbFGF on DNA synthesis in the presence of stimulatorygrowth factors. MCF-7 cells (1.6 X 103/well) were seeded into 24-wellplates in hormone-deprived media supplemented with insulin (1 p.g/ml),

l7�3.estraiJiol (10 mM), or EGF(l0 mg/mi). bFGF (500 pg/mi) was addedto some of the wells, and DNA synthesis was assessed on day 4 by(3H]thymidine incorporation as described in “Materials and Methods.”Columns, means of triplicates; bars, SD.

the growth of early passage MCF-7 cells obtained from the

American Type Culture Collection. Exposure to bFGF (500

pg!ml) induced a significant growth inhibition of cells cultured

in either standard medium (P < 0.001) or in hormone-deprived

medium (P < 0.01). Concomitantly, there was a 90% decrease

in DNA synthesis, as demonstrated by the use of the [3H]thy-

midine incorporation assay (Fig. 2). Thymidine uptake was

increased in the presence of l7�3-estradiol, insulin, or EGF, but

costimulation with bFGF abrogated the effects of these factors

on DNA synthesis (Fig. 2).

The inhibition of DNA synthesis by bFGF was dose de-

pendent. The IC50 in MCF-7 cells was 50 pg!ml, and a maximal

effect was observed with concentrations equal to or greater than

250 pg!ml (Fig. 3). The antiproliferative effect ofbFGF was also

dependent on the duration of exposure. An inhibitory effect was



10 50 100 250 500 75010000 1

bFGF Concentration (pg/mI)

A B C D E F


thymidine incorporation in a variety of breast cancer cells. In

Fig. 3 Dose-response curve of the inhibition of DNA synthesis by

bFGF. MCF-7 cells (1 .6 X 10� cells/well) were seeded into 24-wellplates and exposed to increasing concentrations of bFGF. DNA synthe-sis was assessed by thymidine incorporation after 96 h. Columns, meansof quadruplicates; bars, SD.

E0.0

C0

000.00C

dlC�0

E

I-

Fig. 4 Reversal of bFGF-induced DNA synthesis inhibition by neu-tralizing anti-bFGF antibodies. Cells were plated in standard media in24-well plates. After 24 h, the medium was replaced with standardcondition medium (A), bFGF (500 pg/mI, B), anti-bFGF antibodies (25�tg/ml, C), bFGF and anti-bFGF antibodies (D), normal goat IgG (25p.g/ml, E), or bFGF and normal goat IgG (F). Thymidine incorporationwas determined after 4 days of exposure as described in “Materials andMethods.” Columns, means of quadruplicates; bars, SD.

noticed at 8 h after addition of bFGF (500 pg/mi) and was

maximal after 96 h of exposure (data not shown).

The specificity of the bFGF effect was determined using

neutralizing anti-bFGF antibodies (Fig. 4). Furthermore, the

inhibition of growth by bFGF was reversible, indicating a cy-

tostatic rather than a cytotoxic effect. Trypan blue exclusion,

evaluated after 96 h of bFGF exposure, revealed that 95% of the

cells were viable (data not shown). Replacement of bFGF-

containing medium with standard growth medium resulted in

resumption of exponential growth, as demonstrated by an in-

crease in cell number and in DNA synthesis (data not shown).

Table 1 shows the effect of bFGF on proliferation and

Table 1 Effect of bFGF (500 pg/ml) on proliferation and thymidineincorporation of various breast cancer cell lines and BAECs

Thymidine incorporation Cell count on day 7Cell line (% of control) (% of control)

MCF-7 (ATCC) 15 ± 4 7.7 ±4MCF-7R 47±7 56±1MCF-7L 64±10 22±2MCF-7P 64±6 19±6T47D 83±13 25±4

MDA-MB-453 61 ± 7 28 ± 15BAEC 385 ± I I 469 ± 17

addition to MCF-7 sublines from different sources, the estrogen

receptor-negative breast cancer line MDA-MB-453 and the es-

trogen receptor-positive breast cancer cell line T47-D were

significantly inhibited by bFGF. BAECs were used as a positive

control for cells undergoing growth stimulation by bFGF. In

these cells, bFGF induced a 3.8-fold increase in thymidine

uptake.

Effect of bFGF on Cell Cycle Distribution. Exposureof proliferating MCF-7 cells to bFGF (1 ng/ml) for 24 h resulted

in a significant change in cell cycle distribution compared to

controls (Fig. 5). The fraction of cells in G0-G1 increased after

exposure to bFGF from 51 to 82%, whereas the fraction of cells

in the S-phase decreased from 41% to 14% and in G2-M from

9% to 4%. These data indicate that bFGF induces a G1 block in

MCF-7 cells.

Binding Affinity of ‘�I-labeled bFGF to Intact MCF-7Cells. High-affinity binding of ‘25I-labeled bFGF to intactcells at 4#{176}Cdisclosed the existence of saturable binding sites

(Fig. 6A). Scatchard analysis of the binding (Fig. 6A, inset)

revealed 5.2 x i0� high-affinity binding sites/cell, with a dis-

sociation constant of 57 pM. Receptor cross-linking to 125J..

labeled bFGF revealed a Mr 160,000 protein (Fig. 6B). Assum-

ing that the molecular weight of the cross-linked bFGF peptide

is 1 8,000, the corresponding molecular weight of the cross-

linked receptor would be approximately 142,000. This is similar

to data on the molecular weight of the FGF receptor obtained by

several investigators in other systems (8, 12, 27).

Effect of bFGF on ERK1 and ERK2 Activation. Ac-

tivation of ERK1 and ERK2 leads to tyrosine and threonine

phosphorylation in target proteins (28). Activation of ERK1 and

ERK2 by bFGF was evaluated by the tyrosine phosphorylation

of ERK1 and ERK2 and by an in-gel kinase assay. Fig. 7 shows

that exposure of MCF-7 cells to bFGF resulted in ERK1 and

ERK2 phosphorylation. Western blotting analysis of the cell

lysates showed that Mr 42,000 and 44,000 phosphoproteins that

comigrated with ERK2 and ERK1 were tyrosine phosphorylated

by bFGF in a dose-dependent manner (data not shown). Immu-

noprecipitation of MAPKs with anti-ERK2 monoclonal anti-

body and immunoblotting with antiphosphotyrosine monoclonal

antibodies confirmed a dose-dependent phosphorylation of

ERKs by bFGF (Fig. 7). Maximal phosphorylation was ob-

served with 1 ng/ml of bFGF, which increased the activity of

ERK1 and ERK2 by 2.1- and 6.2-fold above control, respec-

tively. Incubation with a higher concentration of bFGF (10

ng/ml) resulted in a similar (2.3-fold) increase in ERK1 activity

and a 9.9-fold increase in ERK-2 activity. Fig. 8 demonstrates



A

G1=51%S = 40%

2c�J

G2/M=9%

0�.

0N.

Th,’..tJ

�L

�

0U,U, B

0�.�.

G1=82%S= 14%

G2/M-4%

Cc’4c�4

C

0....r�2#{212}0 ‘i#{244}o 6#{212}0 8O0 i#{252}b#{252}

Fig. 5 The effect of bFGF on thecell cycle phase distribution ofMCF-7 cells. A, cell cycle distribu-tion of proliferating cells in stand-

ard growth conditions. B, 24 h af-

ter exposure of MCF-7 cells tobFGF (1 ng/ml). DNA content wasmeasured by staining the DNAwith propidium iodide (P1).

A2#{176}

2#{212}0 4#{212}0

DNA Content (P1)

6#{212}0 8#{212}0 iobo

B

200

97

69

46

1000 2000 3000 4000


z

U

125IbFGF pg!weII

Fig. 6 Concentration dependence of ‘25I-labeled bFGF binding to cells and cross-linking of ‘�I-labeled bFGF to cell surface receptors. A, confluentMCF-7 cells were incubated with various concentrations of ‘25I-labeled bFGF as described in “Materials and Methods.” Nonspecific binding wasdetermined in the presence of a 100-fold excess of unlabeled bFGF. The amount of specifically bound ‘25I-labeled bFGF was then determined. Inset,

data plotted according to Scatchard (26). Points, means of triplicate determinations and the SD did not exceed ± 10%. B, confluent cultures of MCF-7cells were incubated with ‘251-labeled bFGF (5 ng/ml) in the absence (Lane a) or presence (Lane b) of a 100-fold excess of unlabeled bFGF at 4#{176}Cfor 2 h. The bFGF receptor complexes were cross-linked by adding DSS as described in “Materials and Methods” and were analyzed usingSDS-PAGE. The resulting autoradiogram (4-day exposure) is shown. Left, molecular weights in kDa.

the effect of duration of exposure of bFGF (10 ng/ml) on ERKs

phosphorylation. Maximal effect was observed after 5 mm of

bFGF exposure and after 4 h it returned to a near baseline level.

ERK1 and ERK2 activities were similar in cells grown in

standard media and in cells preincubated in serum-free media

24 h prior to stimulation with bFGF.

Exposure of MCF-7 cells to insulin (1-100 jig!ml) also

induced ERK phosphorylation, but similar exposure to l7�3-

estradiol (0.1-10 nM) had not affected ERK1 and ERK2 phos-

phorylation (data not shown). Coincubation of bFGF (10 ng/ml)

with either 17�3-estradiol or insulin yielded the same level of

ERK phosphorylation as observed with bFGF alone (data not

shown).

DISCUSSION

This study demonstrates that bFGF is a potent growth

inhibitor of the hormone-responsive human breast cancer cell

line MCF-7. bFGF induced a G1 block, and growth inhibition by

bFGF was predominant over the mitogemc effect of estrogen,

insulin, or EGF. The negative growth regulation conferred by

bFGF was dose and time dependent and was reversible upon



0 0.1 0.3 1 .0 3.0 10 30 100 PMA MInutes Hours

4 4 -0.�

4 2 �

I II

0 .5 2 5 15 30 4

UUU

Cl)

.�

U

8 24

bFGF(nglmi)


Fig. 7 Effect of increasing concentrations of bFGF on MAPK tyrosinephosphorylation. MCF-7 cells were exposed to increasing concentra-tions (0.1-100 ng/ml) of bFGF. Phorbol-2-myristate 13-acetate (PMA;

100 nM) was used as a positive control. After 15 mm of exposure to

bFGF, the cells were lysed and immunoprecipitated by anti-ERK2monoclonal antibody and then electrophoresed, immunoblotted with anantiphosphotyrosine monoclonal antibody, and visualized with an ECL

system. Relative intensity of signals was determined by a scanning

densitometer. Molecular weights are indicated in kDa.

removal of bFGF from the culture medium. The specificity of

the bFGF antiproliferative effect was demonstrated by its neu-

tralization with a specific anti-bFGF antibody.

The marked inhibitory effect of bFGF on breast cancer

cells contrasts the reported mitogenic effect of bFGF on a large

number of normal and transformed cells (1-3). Consistent with

our observations, however, Schweigerer et a!. (29) reported that

bFGF purified from SK-ESI Ewing sarcoma cells inhibited

SK-ES 1 cell growth, although it stimulated proliferation of other

cells. In addition, Zhou and Serrero (30) reported that bFGF

inhibited the proliferation of a highly tumorigenic insulin-inde-

pendent teratoma-derived cell line. Unlike the reversibility of

the growth-negative effect observed in our cells and in 5K-ES I

cells (29), the bFGF inhibitory effect in the teratoma-denved

cell line was irreversible, although the cell remained viable,

suggesting the induction of terminal differentiation.

Several studies reported that bFGF may stimulate or inhibit

the growth of he same cell type, depending on the culture

conditions under which the experiments were carried out. Karey

and Sirbasku (6) reported that under serum-deprived conditions,

bFGF alone stimulated the growth of MCF-7 cells, whereas in

combination with insulin, insulin-like growth factor I, or EGF,

it inhibited cell proliferation relative to the growth in the ab-

sence of bFGF. Stewart et a!. (4) reported that under estrogen-

deprived conditions, bFGF alone had a minimal growth stimu-

lation effect on MCF-7 cells, but increased proliferation was

observed when bFGF was given in combination with estradiol,

beyond the proliferative effect of estradiol alone. When these

cells lines were grown in the presence of estradiol and insulin-

like growth factor I, however, bFGF induced a significant dose-

.� .�

� � � �

�1ip�--”ThP-� -.� � 0 p42 � � � HCl) 61� � � � H

U � . � � � � �> H i-’� !H� JH �

� 4� ‘r� � �.� � � � � � � ]-�I

� � I � � �0 .5m 2m 5m 15m 30m 4h 8h 24h

bFGF Exposure Time

Fig. 8 Effect of duration of bFGF exposure on MAPK phosphoryla-

tion. Cells were exposed to bFGF ( 10 ng/ml) for the indicated timeintervals. Then cells were lysed and immunoprecipitated by anti-ERK2monoclonal antibody, electrophoresed, and immunoblotted with an an-

tiphosphotyrosine monoclonal antibody. Signals were visualized with an

ECL system. Relative intensity of signals was determined by a scanningdensitometer. Molecular weights are indicated in kDa.

dependent growth inhibition. These observations are consistent

with our data that demonstrated a maximal growth inhibitory

effect by bFGF in the presence of estradiol and insulin, and a

lower growth inhibitory effect in hormone-deprived media (Fig.

1). Furthermore, the growth of MCF-7L, MCF-7P, and MCF-7R

sublines was significantly inhibited by bFGF when cultured in

standard media (Table 1). The same cells, however, were re-

ported to show increased thymidine uptake when stimulated by

bFGF under serum and hormone deprivation (5, 7).

Taken together, these data indicate that bFGF is a pleio-

tropic biological activator capable of inducing mutually exclu-

sive cellular functions (i.e. , proliferation and an antiproliferative

effect) under different conditions. The fact that distinctly op-

posing end results are observed in response to bFGF raises the

possibility that the pattern of bFGF signal transmission and the

ultimate outcome may be subject to transmodulatory regulation

through signaling systems concomitantly activated by other

cellular stimulators such as estradiol and insulin. Transmodula-

tory interactions between signaling systems, leading to an out-

come which is different from that expected of a single stimu-

lating agent, have been reported for activation of T-helper cells

by interleukin 1 and the T-cell receptor, the reciprocal regulation

of apoptosis via ceramide and PKC (3 1-33), and the ability of

1,2-diacylglycerol to convert an apoptotic response to ceramide

into a proliferative response in T Jurkat cells (34). Coordinate

signaling resulting from costimulation by bFGF, estradiol, in-

sulin, and possibly other serum factors may explain the differ-

ences in response of breast tumor cells to bFGF under various

culture conditions. This hypothesis would suggest that the out-

come of bFGF stimulation may depend not only on specific

phenotypic features of MCF-7 cells, but also on variations of the




microenvironment in which the bFGF signal is generated. This

concept may have important implications for understanding the

clinical effects of biological response modifiers and other ther-

apeutic agents when used in combinations in the management of

breast cancer.

The observation that MAPK is activated in MCF-7 cells

under conditions in which bFGF signals an antiproliferative

effect is inconsistent with reported observations on the involve-

ment of MAPK in mediating the effects of bFGF (13, 14). The

FGF receptors engage in at least two distinct signal transduction

pathways. Autophosphorylated tyrosine residues of activated

bFGF receptors bind and transactivate the -y isoform of phos-

pholipase C to initiate turnover of membrane phosphoinositides,

generate l,2-diacylglycerol, and activate PKC (35, 36). This

signaling pathway is apparently not essential for regulating the

mitogenic effect of bFGF, since a point mutation at the tyrosine

residue 766 of the receptor was found to inhibit the phospho-

lipase C-y-PKC cascade, but had no effect on the mitogenic

response (37, 38). An alternative pathway was shown to involve

activation of Ras and Rafl, which initiate signaling through the

MAPK cascade to induce transcription of early response genes

and progression through G1-S-phase and G2-M checkpoints of

the cell cycle and cell division (39). Furthermore, in some cells

PKC was found to mediate the mitogenic effect of FGF (40),

and this effect was shown to be mediated via the Rafl-MAPK

cascade (41). Whether MAPK is causatively associated in

MCF-7 growth inhibition by bFGF remains unknown. More

details on the signaling initiated by bFGF in MCF-7 cells will be

required to delineate which elements are associated downstream

with the antiproliferative effect.

Our study demonstrates that bFGF, while acting as an

inhibitory growth factor in human breast cancer cells, also

activates ERKI and ERK2. The data do not, however, provide

a cause-and-effect association between these effects. Our data

also show that the cytostatic effect of bFGF is reversible and can

be modulated by other biological response modifiers that affect

MCF-7 cells. The demonstration that biological agents may

have opposing effects on human breast tumor cells when coex-

posed to other agents may have important implications in the

design of combined agent therapy in the management of breast

cancer.

REFERENCES

I. Burgess, W. H., and Maciagn, T. The heparmn-binding (fibroblast)growth factor family of proteins. Annu. Rev. Biochem., 58: 575-606,1989.

2. Rifkin, D. B., and Moscatelli, D. Recent developments in the cellbiology of basic fibroblast growth factor. J. Cell Biol., 109: 1-6, 1989.

3. Gospodarowicz, D., Ferrara, N., Schweigerer, L., and Neufeld, G.Structural characterization and biological functions of fibroblast growthfactor. Endocr. Rev., 8: 1-20, 1987.

4. Stewart, A. J., Westley, B. R., and May, F. E. B. Modulation of theproliferative response of breast cancer cells to growth factors by oes-

trogen. Br. J. Cancer, 66: 640-648, 1992.

5. Briozzo, P., Badet, J., Capony, F., Pien, I., Montcourrier, P.,Barritault, D., and Rochefort, H. MCF7 mammary cancer cells respondto bFGF and internalize it following its release from extracellularmatrix: a permissive role of cathepsin D. Exp. Cell Res., 194: 252-259,

1991.

6. Karey, K. P., and Sirbasku, D. A. Differential responsiveness ofhuman breast cancer cell lines MCF-7 and T47D to growth factors andl7�3-estradiol. Cancer Res., 48: 4083-4092, 1988.

7. Peyrat, P. J., Hondermark, H., Louchez, M. M., and Biolly, B.Demonstration of basic fibroblast growth factor high and low affinity

binding sites in human breast cancer. Cancer Commun., 3: 323-329,

1991.

8. Neufeld, G., and Gospodarowicz, D. The identification and partialcharacterization of the fibroblast growth factor receptor of baby hamster

kidney cells. J. Biol. Chem., 260: 3860-3868, 1985.

9. Raz, V., Kelman, Z., Avivi, A., Neufeld, G., Givol, D., and Yarden,Y. PCR-based identification of new receptors: molecular cloning of areceptor for fibroblast growth factors. Oncogene, 6: 753-760. 1991.

10. Dionne, G. A., Crumley, G., Bellot, F., Kaplow, J. M., Scartoff, G.,Ruts, M., Burgess, W. H., Jaye, M., and Schlessinger, J. Cloning andexpression of two distinct high-affinity receptors cross-reacting withacidic and basic fibroblast growth factors. EMBO J., 9: 2685-2692,

1990.

1 1. Givol, D. and Yayon, A. Complexity of FGF receptors: geneticbasis for structural diversities and functional specificity. FASEB J., 6:

3362-3369, 1992.

12. Johnson, D. E., and Williams, L. 1. Structural and functionaldiversity in the FGF receptor multigene family. Adv. Cancer Res., 60:1-41, 1993.

13. Pages, G., Lenormand, P., L’Allemain, G., Chambard, J. C.,

Meloche, S., and Pouyssegar, J. Mitogen-activated protein kinasesp42MAPK and p44MAPK are required for fibroblast proliferation. Proc.Natl. Acad. Sci. USA, 90: 8319-8323, 1993.

14. Anderson, N. G., Maller, J. L., Tonks, N. K., and Strugill, 1. W.

Requirement for integration of signals from two distinct phosphoryla-tion pathways for activation of MAPKinase. Nature (Lond.), 343: 651-653, 1990.

15. Lippman, M. E., Bolan, G., and Huff, K. K. The effect of estrogensand antiestrogens on hormone responsive breast cancer in long termculture. Cancer Res., 36: 4595-5601, 1976.

16. McLeskey, S. W., Ding, I., Y. F., Lippman, M. E., and Kern, F. G.MDA-MB- 134 breast carcinoma cells overexpressed fibroblast growth

factor (FGF) receptors and are growth-inhibited by FGF ligands. Cancer

Res., 54: 523-530, 1994.

17. Dabre, P., Yates, J., Curtis, S., and King, R. J. B. Effect of estradiolon human breast cancer cells in culture. Cancer Res., 43: 349-354,

1983.

18. Gospodarowicz, D., Moran, J., Braun, D., and Birdwell, C. Clonalgrowth of bovine aortic endothelial cells: fibroblast growth factor as asurvival agent. Proc. NatI. Acad. Sci. USA, 73: 4120-4124, 1976.

19. Gray, G. W., and Coffino, P. Cell cycle analysis by flow cytometry.Methods Enzymol., 58: 233-248, 1979.

20. McConahey, P. J., and Dixon, F. J. Radioiodination of proteins by

the use of the chloramine-T methods. Methods Enzymol., 70: 210-2 13,

1980.

21 . Vaisman, N., Gospodorowicz, D., and Neufeld, G. Characterization

of the receptor for vascular endothelial growth factor. J. Biol. Chem.,265: 19461-19466, 1990.

22. Moscatelli, D. High and low affinity binding sites for basic fibro-blast growth factor on cultured cells: absence of a role for low affinitybinding in the stimulation of plasminogen activator production by ho-

vine capillary endothelial cells. J. Cell. Physiol., 131: 123-130, 1987.

23. Raines, M. A., Golde; D. W., Daeipour, M., and Nd, A. E. Gran-

ulocyte-macrophage colony-stimulating factor activated microtubule-

associated protein 2 kinases in neutrophils via a tyrosine kinase-depen-

dent pathway. Blood, 79: 3350-3354, 1992.

24. Hutchcroft, J. E., Anostasrio, M., Harrison, M. L., and Geahlen,R. L. Renaturation and assay of protein kinases after electrophoresis in

sodium dodecyl sulfate-polyacrylamide gels. Methods Enzymol., 200:417-423, 1991.




25. Kameshita, I., and Fujisawa, H. A sensitive method for detection ofcalmodulin-dependent protein kinase H activity in sodium dodecyl sul-fate-polyacrylamide gel. Anal. Biochem., 183: 139-143, 1989.

26. Scatchard, G. The attractions of proteins for small molecules andions. Ann. NY Aced. Sci., 51: 660-672, 1949.

27. Neufeld, G., and Gospoclarowicz, D. Basic and acidic fibroblastgrowth factors interact with the same cell surface receptors. J. Biol.Chem., 261: 5631-5637, 1986.

28. Leevers, S. J., and Marshall, C. J. Activation ofextracellular signal-related kinase, ERK2, by p21 ras oncoprotein. EMBO J., 11: 569-574,1992.

29. Schweigerer, L., Neufeld, G., and Gospodarowicz, D. Basic fibro-blast growth factor as a growth inhibitor for cultured human tumor cells.J. Clin. Invest., 80: 1516-1520, 1987.

30. Thou, J., and Serrero, 0. Fibroblast growth factor inhibits prolifer-ation of a highly tumorigenic insulin-independent teratoma-derived cellline. Growth Factor, 9: 123-131, 1993.

31. Haimovitz-Friedman, A., Kan, C. C., Ehleiter, D., Persaud, R. S.,McLoughlin, M., Fuks, Z., and Kolesnick, R. N. Ionizing radiation acts

on cellular membranes to generate ceramide and initiate apoptosis. J.Exp. Med., 180: 525-535, 1994.

32. Jarvis, W. D., Fornari, F. A., Jr., Browning, J. L., Gewirtz, D. A.,Kolesnick, R. N., and Grant, S. Attenuation of ceramide-induced apop-tosis by diglyceride in human myeloid leukemia cells. J. Biol. Chem.,269: 31685-31692, 1994.

33. Obeid, L. M., Linardic, C. M., Karolak, L. A., and Hannun, Y. A.Programmed cell death induced by ceramide. Science (Washington DC),259: 1769-1771, 1993.

34. Kolesnick, R., and Fuks, Z. Ceramide: a signal for apoptosis ormitogenesis? (Review). J. Exp. Med., 181: 1949-1952, 1995.

35. Ullrich, A., and Schiessinger, J. Signal transduction by receptorswith tyrosme kinase activity. Cell, 61: 203, 1990.

36. Nishizuka, Y. Intracellular signaling by hydrolysis of phospholipidsand activation of protein kinase C (Review). Science (Washington DC),258: 607-614, 1992.

37. Mohamandi, M., Dionne, C. A., Li, W., Li, N., Spivak, T.,Honegger, A. M., Jaye, M., and Schlessinger, J. Point mutation in FGFreceptor eliminates phosphatidylinositol hydrolysis without affectingmitogenesis. Nature (Lond.), 358: 681-684, 1992.

38. Peters, K. G., Marie, J. E., Wilson, E., Ives, H. E., Escobedo, J., DelRosario, M., Mirda, D., and Williams, L. T. Point mutation of a FGFreceptor abolished phosphatidylinositol turnoff and Ca�2 flux but not

mitogenesis. Nature (Lond.), 358: 678-681, 1992.

39. Wang, J. K., Gao, 0., and Goldfarb, M. Fibroblast growth factorreceptors have different signaling and mitogenic potentials. Mol. Cell.Biol., 14: 181-188, 1994.

40. Paris, S., and Pouyssegur, J. Mitogenic effects of fibroblast growthfactors in cultured fibroblasts. Interaction with the G-protein-mediatedsignaling pathways (Review). Ann. NY Acad. Sci., 638: 139-148,1991.

41. Huang, J., Mohammadi, M., Rodrigues, G. A., and Schlessinger, J.Reduced activation of RAF-l and MAP kinase by a fibroblast growthfactor receptor mutant deficient in stimulation of phosphatidylinositolhydrolysis. J. Biol. Chem., 270: 5065-5072, 1995.



1997;3:135-142. Clin Cancer Res E Fenig, R Wieder, S Paglin, et al. cancer cells.mitogen-activated protein kinase activation in human breast Basic fibroblast growth factor confers growth inhibition and

Updated version

http://clincancerres.aacrjournals.org/content/3/1/135

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/3/1/135To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Basic Fibroblast Growth Factor Confers Growth Inhibition ...and Dentistry of New Jersey-New Jersey Medical School, Newark, New Jersey 07103 [R. W., H. W.] ABSTRACT The effect of basic