[CANCER RESEARCH 40. 2361-2366, July 1980]0008-5472/80/0040-OOOOS02.00

Epidermal Growth Factor Stimulation of Human Breast Cancer Cells inCulture1

C. Kent Osborne,2 Barbara Hamilton, Grace Titus, and Robert B. Livingston

Department of Medicine, University of Texas Health Science Center ¡C. K O.. B. H.¡.and the Audie Murphy VA Hospital ¡G. T.. R. B. L ]. San Antonio.Texas 78284

ABSTRACT

Epidermal growth factor (EGF), a polypeptide found in humanand animal blood and secretions, has been found to stimulatea variety of tissues in vitro including normal and malignantrodent mammary epithelium and human breast epithelial cellsand fibroadenoma. We have studied the influence of EGF onmalignant human breast tissue with a model system comprisinghuman breast carcinoma cells growing in tissue culture. EGFstimulated growth of MCF-7 cells in serum-free medium. After7 days in culture, a 2-fold increase in cell number and DMAcontent and a 3-fold increase in total protein were observed in

cells incubated with EGF (10 ng/ml). As little as 0.01 ng/ml ofEGF stimulated growth; 10 ng/ml was maximal. EGF effectson growth were noted for cells plated at a high as well assparse (cloning) density. EGF also stimulated the rates ofthymidine, uridine, and leucine incorporation into macromole-cules in a dose- and time-dependent fashion. Stimulation of

uridine and leucine incorporation was evident by 3 hr, whereasEGF stimulation of thymidine incorporation was delayed until12 to 18 hr. EGF increased the proportion of cells active inDMA synthesis by nearly 2-fold. The combination of optimal

concentrations of insulin (also a growth factor for these cells)and EGF did not stimulate growth above that seen with eitherhormone alone, suggesting a common step in their mechanismof action. The EGF effect was not dependent on the presenceof serum and was not enhanced by dexamethasone as reportedfor other types of cells. EGF had no effect on another humanbreast cancer cell line, the MDA-231. These studies suggest

that growth of some human breast cancers may be influencedby EGF.

INTRODUCTION

EGF3 is a single-chain polypeptide isolated first from mouse

submaxillary glands (7) and later from human urine (8). Although mouse EGF and human EGF differ slightly with regardto molecular weight, amino acid composition, and immunolog-

ical properties, they both compete for the same cell membranereceptor and demonstrate similar biological activities (6). Thetissue site of EGF production and the regulation of its secretionare not understood. However, it is known that daily urinaryexcretion of EGF in humans exceeds 50 fig/day, is significantly

' This investigation was supported in part by Grant BC-288 from the American

Cancer Society. Grant CA 24129 awarded by the National Cancer Institute.Department of Health, Education, and Welfare, and by a grant from the RuthEstrin Goldberg Memorial for Cancer Research.

2 To whom requests for reprints should be addressed.3 The abbreviations used are: EGF, epidermal growth factor; TCA, trichloro-

acetic acid; IMEM, Richter's improved Eagle's minimal essential medium ZO; TLI,

thymidine labeling index.Received September 17, 1979; accepted April 14. 1980.

higher in females, and is 50% higher in women who are takingoral contraceptives (9). In addition, EGF has been detected inhuman plasma, human milk, and other secretions in measurableconcentrations (ng/ml) (9, 22). EGF appears to be closelyrelated if not identical to the gastric antisecretory hormoneurogastrone (12).

Although the relevance of EGF to normal human physiologyand disease has not been clarified, data suggest that thispolypeptide may be an important regulator of cell growth. EGFis mitogenic for a panoply of ectoderm-, mesoderm-, andendoderm-derived cells in tissue culture (11 ). However, not all

cells in culture respond to EGF, which suggests some targettissue specificity and supports the notion that it may have aphysiological role in vivo.

Several studies indicate that EGF may regulate growth ofmammary epithelium: (a) EGF promotes growth of normal rodent mammary tissue and rodent breast cancer (25); (o) itenhances growth of normal human mammary epithelial cells inshort-term culture (24); and (c) it stimulates mitosis of cells

derived from benign human breast fibroadenomas (23). Theeffect of EGF on malignant human breast tissue has not beenreported in detail, but one might speculate that certain breastcancers could retain the sensitivity to EGF observed withnonmalignant mammary cells. In the present studies, we haveexamined the effects of EGF on human breast cancer cellsmaintained in long-term tissue culture, a system well suited for

biochemical studies of hormone action since the environmentalmilieu of the cells can be carefully defined. This model systemhas been characterized in detail and is used by several laboratories for studies of steroid and peptide hormones (18). Wefind that EGF stimulates proliferation of a human breast cancercell line, suggesting that it may be an additional hormonalfactor regulating breast cancer growth in certain patients withthis disease.

MATERIALS AND METHODS

Materials. Mouse EGF was purchased from CollaborativeResearch, Inc. (Waltham, Mass.). Porcine insulin (Lot 615-D63-10, 25.4 units/mg) was the gift of Dr. M. Root, LillyResearch Laboratories (Indianapolis, Ind.). Crystalline TCA waspurchased from Baker Chemical Co. (Phillipsburg, N. J.), and[3H]thymidine (47 Ci/mmol), [5-3H]uridine (29 Ci/mmol), andL-[U-'"C]leucine (342 mCi/mmol) were obtained from the Ra-

diochemical Centre, Amersham, England.Cells and Tissue Culture Techniques. Both human breast

cancer cell lines were initially derived from malignant effusionsof women with metastatic breast cancer and have been incontinuous tissue culture for at least 4 years. Characterizationof the human and mammary origin of these cells has beensummarized previously (10, 13, 18). The MCF-7 cell line was

JULY 1980 2361

on March 29, 2019. © 1980 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

C. K. Osborne et al.

provided by Dr. H. D. Soûleat the Michigan Cancer Foundation(21). These cells are maintained in monolayer culture in IMEM(Grand Island Biological Co., Grand Island, N. Y.) supplemented with glutamine (0.6 g/liter), penicillin (62 mg/liter),streptomycin (130 mg/liter), insulin (1.0 nw), and 5% calfserum. The MDA-231 cell line was initially isolated by Dr. R.Cailleau at M. D. Anderson Hospital and Tumor Institute (4) andsupplied to us by Dr. M. Lippman of the National CancerInstitute. These cells are maintained in the same medium asthe MCF-7 cells but with 10% fetal bovine serum and no insulinsupplementation. Both cell lines are grown in a humidifiedincubator in 5% CO? at 37°,and both are free of Mycoplasme

contamination. The cells are subcultured weekly with the useof 0.05% trypsin:0.02% EDTA in 150 mw NaCI to remove cellsfrom the culture dish. In all experiments, cells were harvestedby suspending them with this same solution.

Precursor Incorporation and Growth Studies. Methods forthe measurement of precursor incorporation into macromole-

cules and cell growth have been summarized previously (17,20). Briefly, incorporation studies were performed by platingcells in replicate to a subconfluent density (about 7 x 10s

cells/well) in 4 ml of IMEM supplemented with the appropriateserum in multiwell culture dishes (FB-4-TC; Linbro Chemical

Co., New Haven, Conn.). After the cells had attached to thedish (24 hr), the medium was exchanged for serum-free IMEM.After a further 24 hr, hormones (or hormone-free carrier for

controls) were added directly to the incubation medium, andthe cells were incubated for specified times. During the last 2hr of incubation, cells were pulsed with labeled thymidine (0.5fiCi/ml), uridine (1.0 to 2.0 /¿Ci/ml), or leucine (0.5 /nCi/ml)and then were harvested. An aliquot of the cell suspension wastaken for DNA determination (3), and the remaining cells weredisrupted by sonicating for 3 sec in a Heat-Systems Ultrasonic

sonicator (Plainview, N. Y.) with a microtip at the lowest setting.Precursor incorporation was determined by measuring theradioactivity precipitated in 10% TCA after collection on Milli-pore filters (0.45 firn, type HA). The rates of precursor incorporation were linear under the conditions used.

Cell growth studies were performed by determining the totalDNA (3) or protein (15) content of cells plated in replicate onPetri dishes (Falcon Plastics, Oxnard, Calif.) after incubation inserum-free medium (unless stated otherwise) with or withoutEGF or other factors. Cell number was determined by visualcounts in a hemocytometer. Changes in DNA content closelyparalleled changes in population size.

TLI. The TLI was determined by a modification of previouslypublished methods (14). Cells were plated in replicate andincubated with hormones, as described for the incorporationstudies. At the end of the incubation period, they were pulsedfor 1 hr with [3H]thymidine (5 juCi/ml), washed twice, harvested,

and centrifuged (600 rpm for 2 min). The supernatant containing the harvesting medium was discarded. The cells wereresuspended in chilled (4°) Dulbecco's phosphate-bufferedsaline (pH 7.4, without Ca2+ or Mg2+) at a concentration ofabout 2.5 x 105 cells/ml. Two-tenths ml of the diluted cell

suspension was added to a cytocentrifuge cup and centrifugedfor 5 min at 1000 rpm (Shandon Elliott, Southern Instruments,Inc., Sewickley, Pa.). After a drying time of 4 hr, the cytocen-trifuge-prepared slides were dipped 10 times in each of 2

staining dishes containing 5% cold TCA and then dipped 5times in a dish containing methyl alcohol. After drying, the

slides were dipped in Kodak NTB-2 emulsion, exposed for 24hr in a light-tight box, developed with D19 developer, and fixed

with Kodak general purpose fixer. The slides were dried andthen stained with May-Grunwald-Giemsa. A total of 500 cellswere counted for each of 4 slides representing one experimental group (2000 cells total). The labeling index (labeled/unla-

beled cells) was calculated by counting the fraction of intactcells containing 8 or more grains over the nucleus. Backgroundcounts for the autoradiographs were always less than 5 grains/cell nucleus.

Statistical Analysis. When applicable, the means of experimental groups were compared by using Student's f test.

RESULTS

Effects of EGF on Growth. EGF significantly stimulatesgrowth of the MCF-7 human breast cancer cells (Chart 1 A and

B). Cell proliferation as determined by measuring the total DNAand protein content of dishes plated in replicate was augmented by the addition of EGF (10 ng/ml) to cells growing inserum-free medium. Saturation density of the cells was reachedby 7 days in both control and EGF-treated cultures. At thistime, DNA content was increased nearly 2-fold and total proteinwas increased 3-fold in cultures incubated with the poly-

peptide. The increased DNA content in this experiment paralleled a 2-fold increase in actual cell number after 5 days[control = 7.2 ±0.2 x 10s (S.D.) cells/dish; EGF = 13.3 ±0.9 x 105 cells/dish], indicating that EGF stimulates cell

division and not just synthesis of DNA and protein. The observation that EGF stimulates protein synthesis to a greater extentthan DNA synthesis and mitosis (a hypertrophie effect) wasreproducible in multiple experiments. In a separate but similarexperiment (Chart 1C), the effect of EGF on actual cell numberwas determined. The initial doubling time was reduced fromabout 36 to 24 hr by the hormone. It should be emphasizedthat control cells in serum-free medium alone were not dyingunder these conditions but continued to synthesize macromol-

ecules and to divide. Thus, the EGF effect is not simply to delayor prevent cell death but to increase multiplication of thesecells.

Growth of MCF-7 cells is extremely sensitive to EGF (Chart

2). Stimulation of DNA and protein synthesis was observedwith as little as 0.01 ng of EGF per ml (p < 0.05 and 0.005,respectively). Half-maximal stimulation was noted with concentrations between 0.1 and 1.0 ng/ml, and maximal stimulationwas noted at 10 ng/ml. These EGF concentrations are similarto those required for a mitogenic effect in other tissues (5, 11).Furthermore, these concentrations are within the range ofconcentrations reported for EGF in human secretions andplasma by radioimmunoassay (9, 22). A modest but consistentreduction in the stimulatory effect of EGF was noted at concentrations of 50 ng/ml or greater.

The effect of EGF on growth is evident even when MCF-7cells are plated at a very low ("cloning") density (Table 1). In

this experiment, cells were plated at a density of 500 cells/60-mm Petri dish and incubated for 15 days in serum-free medium

with and without EGF or 5% calf serum. Surprisingly, cellsmaintained in serum-free medium alone slowly divided to reacha saturation density of 11.4 x 10" cells. The addition of EGF

resulted in a 2-fold increase in saturation density. However, theeffect of EGF is only about 10% of that observed with calf

2362 CANCER RESEARCH VOL. 40

on March 29, 2019. © 1980 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Stimulation of Human Breast Cancer by EGF

1000-,

0 I

50 n

0 I

1 5-

357Days

Chart 1. Effects of EGF on growth of MCF-7 cells Cells were plated inreplicate (1 x 10s cells/100-mm Petri dish) as described in "Materials andMethods." After 24 hr in serum-free medium, some dishes received EGF at 10

ng/ml (Day 0). Cells were fed with fresh medium with or without EGF on Days 2.4, 6, 8, and 10. On the days indicated, cells were harvested for determination oftotal protein in A (A. controls; A, EGF) or total DMA in e (•,controls; O. EGF). C.a similar experiment in which the cell number was determined on the daysindicated (•.controls; O. EGF). Results are expressed as the means of triplicates.Bars. S.E.

again, no EGF effect was observed (Table 2). Similar experiments using higher serum concentrations (1%) or serumstripped of steroid hormones and probably many proteins bytreatment with dextran-coated charcoal failed to produce EGFstimulation of the MDA-231 cells (data not shown). Of interest

is our previous observation that these cells are also insensitiveto insulin (20). The factors in serum responsible for growthstimulation of this cell line remain unknown.

Effects of EGF on Precursor Incorporation. We also examined the effects of EGF on the rates of incorporation ofradioactive precursors into macromolecules in MCF-7 cells.

The sensitivity of these cells to physiological concentrations ofEGF was confirmed by the increased incorporation of [3H]-thymidine, [3H]uridine, and [14C]leucine into TCA-precipitable

material (Chart 3). Slight stimulation was observed with 0.10ng of EGF per ml, and maximal stimulation was observed with10 ng/ml, in parallel with the dose-response curve of growth

stimulation (Chart 2). The magnitude of the maximal EGFstimulation of precursor incorporation above control values (30

IOO

Chart 2. EGF dose response in MCF-7 cells Cells were plated as describedin Chart 1 (4 x 105 cells/100-mm Petri dish), and EGF was added at the

indicated concentrations. The cells were fed fresh medium on Day 3 and harvested on Day 5 for DMA and protein determination Results are the means oftriplicates. Bars, S.E.

Table 1Effect of EGF on growth of MCF- 7 cells plated at cloning density

Cells were plated as described in "Materials and Methods in medium con

taining 5% calf serum at a density of 500 cells/60-mm Petri dish. After the cellshad attached to the dish (24 hr). the medium was changed to serum-free mediumalone (control), serum-free medium plus EGF (10 ng/ml). or fresh medium plus5% calf serum. EGF was added to appropriate cultures every other day. Spentmedium was exchanged for fresh medium with appropriate additions every 4days Cells were harvested for cell counts, DMA, and protein determination onDay 15.

ControlEGFCalf serumNo.

of cells (x10")11.4

±0.6a

20.2 ±0.7229.4 ±5.0DNA

(fig)3.8

±0.27.5 ±0.3

85.8 ±1.4Protein

(/ig)55.9

± 4.0149.3 ± 6.4

1091.4 ±60.1

Mean ±S.E. of quadruplicate determinations.

serum, indicating that other factors in serum are also importantfor optimal growth of these cells.

In contrast to the effect on MCF-7 cells, EGF had no stimulatory effect on DNA or protein synthesis in the MDA-231

human breast cancer cell line (Table 2); in fact, a slight decrease in DNA and protein was observed with cells incubatedwith EGF. Because an EGF effect on certain cells in culturehas been reported to require a cofactor present in low concentrations of serum (11, 23), we also examined the effect of EGFon these cells in the presence of calf serum (0.5%). Calf serumitself significantly augmented macromolecular synthesis, but,

Table 2Effect of EGF on growth of the MDA-231 cell line

Cells were plated as described in 'Materials and Methods" and in Chart 1.

On Day 0, cells were divided into 4 groups: control (serum-free medium); EGF(10 ng/ml in serum-free medium); 0.5% calf serum; or 0.5% calf serum plusEGF. Fresh medium was added on Day 3. and cells were harvested for DNA andprotein determination on Day 5.

ControlEGFCalf serumCalf serum + EGFDNA

0,g)76.2±1.6a

69.2 ±2.3133.6 ±1.5128.0 ±1.1Protein

(/ig)1561

±341355 ±34

2587 ±1702406 ±68

Mean ±S.E. of triplicate determinations.

JULY 1980 2363

on March 29, 2019. © 1980 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

C. K. Osborne et al.

to 100% depending on the experiment) was similar to themaximal stimulation induced by insulin in these cells (17).

The time course of the effects of EGF on thymidine, uridine,and leucine incorporation into macromolecules is also similarto that observed with insulin (17). Stimulation of leucine anduridine incorporation occurred shortly after EGF addition (by 3hr) and was maximal from 12 to 18 hr (Chart 4). On the otherhand, EGF stimulation of thymidine incorporation was evidentonly after a lag period of about 12 hr; maximal stimulationoccurred at about 24 hr incubation with EGF. This ordered,sequential stimulation of precursor incorporation by EGF suggests that it is not simply due to altered precursor uptake intothe cell (usually an immediate effect) with subsequent changesin the specific activity of the precursor pool. The fact that EGFalso stimulates the thymidine labeling index (see below) supports that contention and suggests that an effect of EGF maybe to increase the proportion of cells actively synthesizingDMA.

TLI. The TLI provides an estimate of the fraction of cells in apopulation actively engaged in DMA synthesis at a given time(14). EGF significantly augmented the TLI of MCF-7 cells (Table3). Nearly a 2-fold increase in TLI above that seen in cellsmaintained in control medium was observed for cells incubatedwith EGF for 24 hr. This effect on the TLI was identical to that

80 n

O.I IO IOEGF (nq/ml)

Chart 3. Effects of EGF concentration on precursor incorporation. Cells wereplated in replicate as described in "Materials and Methods.' After 24 hr in

serum-free medium, EGF was added at the indicated concentrations for 20 hr.During the last 2 hr. the cells were pulsed with [3H]thymidine (0.5 /iCi/ml),[3H]uridine (2.0 /iCi/ml), or ["C]leucine (0.5 fiCi/ml). The radioactivity precipi

tatale by 10% TCA was determined and expressed as dpm/dish/2 hr and shownhere as a percentage above control. Results are means of triplicates. S.E.'s were

between 2 and 10%.

Chart 4. Time course of the effect of EGF on labeled precursor incorporationinto macromolecules by MCF-7 cells. Cells were plated in replicate as describedin "Materials and Methods." EGF (10 ng/ml) was added at Time 0. [3H]Thymidine(0.5 fiCi/ml). [3H]uridine (2.0 /iCi/ml). or [MC]leucine (0.5 fiCi/ml) were added

for a 2-hr pulse just prior to harvesting at the times shown. Incorporation intoTCA-precipitable material was determined and expressed as dpm per /ig DNAper 2 hr. Results are the means of triplicates or quadruplicates. Bars, S.E.

Table 3Effect of EGF and insulin on the TU of MCF- 7 cells

Cells were plated in replicate as described in Chart 1 and then incubated withEGF (10 ng/ml) and/or insulin (10 nM) for 20 hr. The cells were pulsed with[3H]thymidine (5 fiCi/ml) for 1 hr, and the percentage of cells with labeled nucleiwas determined by autoradiography as described in "Materials and Methods.

ControlEGFInsulinEGF

+ insulinTLIa20

±1b35

±236±233±1

Expressed as percentage of labeled nuclei.6 Mean ±S.E.

seen with insulin (10 nw). Interestingly, no additive effect onTLI was observed in cells incubated with optimal concentrationsof the 2 hormones together. Thus, a major effect of EGF andinsulin in these cells may be to increase the fraction of cells inthe population that synthesize DMA. Additional studies will berequired to more accurately define the effects of these hormones on other parameters such as cell cycle transit times.

Because the effects of EGF and insulin on the TLI were notadditive, we wondered whether cell proliferation was influencedin a similar fashion by these 2 growth factors. When MCF-7

cells were incubated with optimal concentrations of EGF orinsulin for 6 days, the effect on cell growth as measured bytotal DMA or protein content of dishes plated in replicate wasstrikingly similar (Table 4). A 50% increase in DMA content anda 75 to 100% increase in total protein were observed. However,an additive effect on DMA and protein content was not observedwhen the cells were incubated with both hormones. On thecontrary, a slight decrease in DMA was noted for cells exposedto both EGF and insulin together, and only a minimal increase(not additive) in total protein was noted in this group. Thesedata suggest the possibility that the stimulation of MCF-7 cells

by EGF and insulin may involve a common step in the mechanism of action of these hormones. Furthermore, they suggestthat the pathways of regulation of DMA and protein synthesisby these hormones may be different, since changes in proteincontent do not always parallel changes in DNA or cell number.

Influence of Glucocorticoids and Serum on the EGF Response. Several reports have indicated that the presence ofeither serum or the glucocorticoid dexamethasone enhancesthe mitogenic response of human fibroblasts in culture to EGF(1,5, 11, 23). Presumably, serum provides a cofactor necessary for a maximum EGF response, whereas dexamethasonemay increase the binding of EGF to its receptor. In contrast,the response of MCF-7 cells to EGF is not enhanced bydexamethasone or by low concentrations of calf serum (Tables5 and 6, respectively). Dexamethasone (1.0 or 100 nw) for 6days resulted in a minimal increase in total cell protein or DNAper dish (Table 5). However, when dexamethasone was addedwith EGF, a decrease in protein and DNA compared to EGFalone was observed (p < 0.025). This antagonism of the EGFresponse by dexamethasone is similar to antagonism by dexamethasone of the mitogenic response to insulin in MCF-7 (1 7)and ZR 75-1 (19) human breast carcinoma cells.

Similarly, calf serum did not synergistically enhance theeffect of EGF in MCF-7 cells (Table 6). In serum-free mediumEGF caused a 3-fold rise in total protein per dish and a 70%

increase in DNA over untreated cells. Calf serum stimulatedDNA and protein accumulation in a dose-dependent manner

2364 CANCER RESEARCH VOL. 40

on March 29, 2019. © 1980 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

Stimulation of Human Breast Cancer by EGF

Table 4Effect of EGF and insulin on MCF-7 cell growth

Cells were plated as described in Chart 1 (2 x 10s cells/60-mm Retri dish)

and exposed to EGF (10 ng/ml) and/or insulin (1.0 nM)for 6 days with a mediachange on Day 3.

ControlEGFInsulinInsulin + EGFDMA

/dish(¿ig)33.5±2A°

49.6 ±0.750.4 ±2.544.5 ±1.4Protein/dish

(fig)406

±24805 ±27728 ±20931 ±56

Mean ±S.E.

Table 5Effect of dexamethasone on the MCF-7 response to EGF

Cells were plated as described in Chart 1 (5 x 10s cells/100-mm Petri dish).EGF (10 ng/ml), dexamethasone, or both were added to the cells for 6 days.Fresh media with hormones were exchanged for spent media on Day 3.

ControlEGFDexamethasone

(1.0nM)Dexamethasone(1.0nM)+

EGFDexamethasone(100nM)Dexamethasone(100MM)+

EGFDNA/dish

(fig)115±5a216

±4133

±4185±6137

±8164±4Protein/dish

(jig)1709

±1134233±1252321±1763403±1802405

±243518± 280

s Mean ±S.E.

Table 6Effect of serum on the MCF-7 response to EGF

After plating (1 x 105cells/100-mm Petri dish), cells were incubated with the

factors shown for 5 days prior to harvesting for protein and DMAdetermination.Spent media were exchanged for fresh media on Day 3.

ControlEGF (10 ng/ml)Calf serum (0.1%)Calf serum (0.1%) + EGFCalf serum (0.5%)Calf serum (0.5%) + EGFDNA/dish

(/ig)19.6±1.5a

33.6 ±0.530.4 ±0.446.2 ±0.862.7 ±1.656.1 ±1.6Protein/dish

(^g)234

± 43676 ± 30513 ± 17965 ± 23

1082 ±1171326 ±87a

Mean ±S.E.

but did not augment the EGF effect. In fact, in the presence of0.5% calf serum EGF increased protein accumulation 30%above that seen with the calf serum alone, but no additionalstimulation of DMA content was observed. Thus, serum factorsare not required for, nor do they synergistically enhance, themitogenic response to EGF of MCF-7 cells.

DISCUSSION

The importance of EGF for cellular growth regulation in vivohas not been defined. In vitro studies, however, suggest that avariety of tissues may be targets for EGF action. Mammaryepithelium is a potential target for this hormone. It is of interestthat levels of EGF in humans are higher in females and areincreased further by pregnancy or exogenous sex steroidhormone administration (9), conditions which drastically altermammary gland growth and development. Furthermore, cellculture studies suggest that EGF regulates growth of rodentand human mammary epithelial cells (23-25). The present

studies demonstrate that EGF is also a potent growth stimulusfor human breast cancer cells in culture, thus confirming apreliminary report that EGF is a necessary ingredient for growthmaintenance of these cells in a defined serum-free medium (2).Physiological concentrations of EGF enhance MCF-7 cell mul

tiplication and increase the rates of thymidine, uridine, andleucine incorporation into macromolecules in these cells. Amajor effect of this hormone appears to be an increase in thenumber of S-phase cells in the population. Whether this is due

to recruitment of nonproliferating cells into the cell cycle orchanges in cell cycle transit times will require further study.The observation that another breast cancer cell line, the MDA-

231, is not sensitive to EGF suggests that only certain breasttumors retain the biomolecular machinery necessary for anEGF effect and that the stimulatory effect of EGF on the MCF-7 cells is not an artifact of tissue culture conditions.

Unlike effects on certain other types of cells in culture, theeffect of EGF on MCF-7 cells is evident in serum-free medium

and is not exaggerated in the presence of calf serum. Inaddition, dexamethasone does not augment the EGF responsein these cells, but, in fact, slightly antagonizes the response.These disparate results among different cells in culture may bedue to culture conditions, or may reflect genuine differencesamong tissues in their hormonal responsiveness. We have alsoobserved that the mitogenic action of insulin is antagonized byglucocorticoids in human breast cancer cells in tissue culture(17, 19). One possible explanation for these effects is thatdexamethasone could cause a nonspecific alteration of the cellmembrane, thereby affecting both EGF and insulin receptors.Another possibility is that dexamethasone opposes the actionof EGF and insulin at a common site distal to receptor binding.

The effects of EGF on the MCF-7 cell line are strikingly

similar to the effects we have reported previously for insulin(1 7, 19). (a) The stimulation of growth and precursor incorporation into macromolecules is quantitatively equivalent withboth hormones, (b) Both hormones stimulate cell proliferationand the TLI, and the effects are not additive when the cells areexposed to optimal concentrations of both hormones, (c) Thetime course of stimulation of thymidine, uridine, and leucineincorporation by insulin and EGF are nearly identical, (d) Glucocorticoids modulate the effects of both hormones in a similarmanner, (e) The MDA-231 cell line, which is unresponsive to

insulin, also fails to respond to EGF. These observations suggest that EGF and insulin may stimulate a common biochemicalpathway in MCF-7 cells leading to enhanced growth. Sincespecific receptors for each of these polypeptides have beendetected in target cells and since neither hormone competesfor the receptor of the other in MCF-7 cells (20)," it is unlikelythat a "common pathway" involves a common receptor binding

site at the cell membrane. However, after initial binding of thesehormones to receptors dispersed on the cell surface, it hasbeen shown by fluorescent techniques that bound insulin andEGF rapidly migrate on the membrane to form patches whereboth are internalized within the same vesicle by a commonpathway (16). Whether patching and internalization are relatedto the growth effects of these hormones is not known, but it isinteresting to speculate that the similar responses to EGF andinsulin observed in MCF-7 cells are related to this event.

In summary, we have shown that EGF stimulates growth ofhuman breast cancer cells in tissue culture. These results, inconcert with other studies of rodent and human mammaryepithelium, suggest the possibility that EGF has a regulatoryrole in the growth and development of the breast and perhapsof breast cancer.

' C. K. Osborne and B. Hamilton, unpublished data.

JULY 1980 2365

on March 29, 2019. © 1980 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

C. K. Osborne et al.

REFERENCES

1. Baker, J. B., Barsh. G. S., Carney. D. H., and Cunningham, D. D. Dexa-methasone modulates binding and action of epidermal growth factor inserum-free cell culture. Proc. Nati. Acad. Sei. U. S. A., 75. 1882-1886.1978.

2. Barnes, D. W., and Sato, G. Growth of a human breast cancer cell line (MCF-7) in a defined, serum-free medium. In Vitro (Rockville), 75. 130A. 1979.

3. Burton, K. A study of the conditions and mechanism of the diphenylaminereaction for the colorimetrie estimation of deoxyribonucleic acid. Biochem.J.. 62. 315-323, 1956

4. Cailleau, R.. Young, R., Olive, M., and Reeves, W. J., Jr. Breast tumor celllines from pleural effusions. J Nati. Cancer Inst., 53. 661-674, 1974.

5. Carpenter, G., and Cohen, S. Human epidermal growth factor and theproliferation of human fibroblasts. J. Cell. Physiol., 88 227-238, 1975.

6 Carpenter, G., and Cohen, S. Biological and molecular studies of themitogenic effects of human epidermal growth factor. In: J. Papaconstantinouand W. J. Rutter, (eds.), Molecular Control of Proliferation and Differentiation, pp. 13-31. New York: Academic Press, Inc., 1978.

7. Cohen, S. Isolation of a mouse submaxillary gland protein acceleratingincisor eruption and eyelid opening in the new-born animal. J. Biol. Chem.,237. 1555-1562, 1962.

8. Cohen. S.. and Carpenter, G. Human epidermal growth factpr: isolation andchemical and biological properties. Proc. Nati. Acad. Sei. U. S. A. 72:1317-

1321, 1975.9. Dailey. G. E., Kraus, J. W , and Orth. D. N. Homologous radioimmunoassay

for human epidermal growth factor (urogastrone). J. Clin. Endocrinol. Me-tab , 46 929-936, 1978.

10. Engel. L. W., and Young. N. A. Human breast carcinoma cells in continuousculture: a review. Cancer Res., 38. 4327-4339, 1978.

11. Gospodarowicz, D.. Greenburg, H.. Bialecki, H., and Zetter, B. R. Factorsinvolved in the modulation of cell proliferation in vivo and in vitro: the role offibroblast and epidermal growth factors in the proliferative response ofmammalian cells. In Vitro (Rockville), 74. 85-118. 1978.

12. Gregory, H. Isolation and structure of urogastrone and its relationship toepidermal growth factor. Nature (Lond.). 257: 325-327. 1975.

13. Lippman. M. E., Osborne, C. K., Knazek, R , and Young. N. In vitro modelsystems for the study of hormone-dependent breast cancer. N. Engl. J.Med.. 296 154-159. 1977.

14. Livingston, R. B , Ambus, U., George, S. L., Freireich, E. J. and Hart, J. S.In vitro determination of thymidine-3H labeling index in human solid tumors.

Cancer Res., 34: 1376-1380. 1974.15. Lowry, O. H., Rosebrough, N. J., Farr, A. L.. and Randall, R. J. Protein

measurement with the Folin phenol reagent. J. Biol. Chem., 793. 265-275,

1951.16. Maxfield, f. R., Schlessinger, J.. Shechter, Y.. Pastan. I., and Willingham.

M. C. Collection of insulin. EGF and «2-macroglobulin in the same patcheson the surface of cultured fibroblasts and common internalization. Cell. 14:805-810. 1978.

17. Osborne. C. K.. Bolán, G., Monaco. M. E., and Lippman, M E. Hormoneresponsive human breast cancer in long-term tissue culture: effect of insulin.Proc. Nati. Acad. Sei. U. S. A.. 73. 4536-4540, 1976.

18. Osborne, C. K.. and Lippman, M. E. Human breast cancer in tissue culture:the effects of hormones. In: W. L. McGuire (ed.). Breast Cancer. AdvancesIn Research and Treatment, vol. 2, pp. 103-154. New York: Plenum Pub

lishing Corp., 1978.19. Osborne, C. K., Monaco, M. E.. Kahn, C. R.. Huff, K.. Bronzert. D., and

Lippman, M. E. Direct inhibition of growth and antagonism of insulin actionby glucocorticoids in human breast cancer cells in culture. Cancer Res., 39.2422-2428, 1979.

20. Osborne, C. K., Monaco, M. E., Lippman. M E., and Kahn. C R. Correlationamong insulin binding, degradation, and biological activity in human breastcancer cells in long-term tissue culture. Cancer Res.. 38: 94-102. 1978.

21. Soule, H. D., Vazquez, J.. Lang. A , Albert. S.. and Brennan, M. A. A humancell line from a pleural effusion derived from a breast carcinoma. J. Nati.Cancer Inst.. 51: 1409-1417. 1973

22. Starkey, R. H., and Orth, D. N. Radioimmunoassay of human epidermalgrowth factor (urogastrone). J. Clin. Endocrinol. Metab.. 45: 1144-1153,1977.

23. Stoker, M. G. P.. Pigott, D.. and Taylor-Papadimitrious. J. Response toepidermal growth factor of cultured human mammary epithelial cells frombenign tumors. Nature (Lond.). 264. 764-767, 1976.

24. Taylor-Papadimitrious, J.. Shearer. M.. and Stoker. M. G. P Growth requirements of human mammary epithelial cells in culture. Int. J. Cancer, 20. 903-

908, 1977.25. Turkington, R. W. Stimulation of mammary carcinoma cell proliferation by

epithelial growth factor in vitro. Cancer Res.. 29 1457-1458. 1969.

2366 CANCER RESEARCH VOL. 40

on March 29, 2019. © 1980 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

1980;40:2361-2366. Cancer Res   C. Kent Osborne, Barbara Hamilton, Grace Titus, et al.   Cells in CultureEpidermal Growth Factor Stimulation of Human Breast Cancer

  Updated version

  http://cancerres.aacrjournals.org/content/40/7/2361

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

  .pubs@aacr.orgDepartment 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://cancerres.aacrjournals.org/content/40/7/2361To request permission to re-use all or part of this article, use this link

on March 29, 2019. © 1980 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from