Serum Regulation of the Estrogen Responsiveness of the Human Breast Cancer Cell Line MCF-71

Home , Nitromifene

[CANCER RESEARCH 43. 1244-1250. March 1983] 0008-5472/83/0043-OOOOS02.00 Serum Regulation of the Estrogen Responsiveness of the Human Breast Cancer Cell Line MCF-71

Martin J. Page, John K. Field, Nie P. Everett, and Chris D. Green2

Department of Biochemistry. The University of Liverpool, P. O. Box 147. Liverpool L69 3BX. United Kingdom

ABSTRACT England, and tamoxifen was a gift from S. R. Slater, I.C.I. Pharmaceu ticals Division, Alderloy Edge, England. Initially after receiving MCF-7 cells, we were able to confirm Maintenance of Stock Cultures. MCF-7 human breast cancer cells their estrogen responsiveness. We observed significant in were kindly provided by Dr. M. Rich, Michigan Cancer Foundation, creases in thymidine incorporation, in thymidine kinase activity, Detroit, Mich., and Dr. M. Lippman, National Institute of Health, Be- and in cell numbers in response to 10~8 M estradiol. Subse thesda, Md. The cells were grown as monolayer cultures in glass bottles at 37Â° in an atmosphere of 5% CO? in air. Growth medium quently, however, the cells failed to show a response to estra consisted of Dulbecco's minimal essential medium (Gibco, Paisley, diol. A growth response to estradiol could be restored by Scotland), supplemented with 4 mw L-glutamine (BDH, Liverpool, Eng increasing the serum concentration in the medium. Cells grown land) penicillin:streptomycin (250 units/ml; Gibco), and 10% (v/v) in 15% serum (calf or human) responded to estradiol with NBCS3 (Gibco or Flow, Irvine, Ayrshire, Scotland). Cells were harvested increased rates of growth and thymidine incorporation and by rinsing the monolayer with PBS, followed by a 15-min incubation increased activities of thymidine kinase and DNA polymerase. with PBS containing EDTA (0.1%). Cultures were tested for Myco- We suggest that there is present in serum a "factor" which plasma contamination using the Hoechst fluorescent staining method can influence the expression of a growth response to estradiol. (3). Preparation of Steroid-depleted Serum. Serum was stirred over night at 4Â° with 0.25% activated charcoal (Sigma) and 0.0025% INTRODUCTION dextran (Pharmacia, Hounslow, England). The serum was then centri- MCF-7 is a human breast cancer cell line derived from the fuged at 10,000 x g to remove the charcoal and passed through a 0.22-nm Millipore filter. Removal of steroids was monitored by the pleural effusion of a patient with metastatic breast adenocar- addition of a trace of [3H]estradiol at the beginning of the procedure. cinoma (21). These cells have been characterized extensively Greater than 97% of the tritiated steroid was removed. and have been shown to be of human and breast origin. [3H]Thymidine Incorporation Experiments. Cells were uniformly Because growth and macromolecular synthesis of MCF-7 cells plated in 35-mm plastic dishes using medium containing 5% serum. have been reported to be stimulated by physiological concen After the cells were left overnight to allow attachments, the medium trations of estradiol (13, 14), they have been proposed as an was aspirated and replaced by medium containing 0.5% steroid-de in vitro model system for the study of estrogen-responsive pleted serum (unless otherwise stated) and left for 24 hr. The medium was then aspirated and replaced by fresh medium containing hormone breast cancer (15). additions in ethanol. The cells were labeled with [3H]thymidine for the However, reports from other laboratories and our own ex last 2 hr of the next 48-hr incubation period. The dishes were then perience with these cells have shown that the degree of re thoroughly washed with ice-cold PBS, and the cells were harvested sponsiveness of MCF-7 cells to estradiol is rather variable. and spun at 800 x g for 5 min. To the cell pellets was added 1.0 ml Early reports from Lippman ef al. showed that estradiol more distilled water, and the cell suspension was sonicated (MSE Sonicator). than doubled rates of macromolecular synthesis (13) and could To 600 ut of this homogenate was added an equal volume of ice-cold double cell number yield (15). However, more recently, the 10% trichloroacetic acid, and the mixture was left at 4Â°for 30 min. same group has observed an abrupt decrease in the response The resulting precipitate was collected on a GF/C filter (Whatman, of their MCF-7 cells to estradiol (22). Similarly, other workers Maidstone, Kent, England) and washed with ice-cold 5% trichloroacetic have recently shown that estradiol failed to stimulate growth acid (10 ml). The filters were then placed in scintillation vials and dried (5, 8) and produced only a minimal stimulation of [3H]thymidine at 100Â° for 15 min, and the radioactivity was determined using Triton:toluene scintillation fluid (700 ml toluene, 300 ml Triton X-100, incorporation (10). Other recent reports would suggest that estradiol stimulation of MCF-7 cell growth can still be demon 5 g PPO, and 0.3 g POPOP). The remaining 400 Â¡i\ofeach homogenate were used for protein determinations (16). strated (1, 17). Cell Number Determinations. Cells were harvested in PBS:EDTA, In this paper, we describe the characteristics of MCF-7 cells pelleted at 800 x g (5 min), resuspended in culture medium, and which had become apparently unresponsive to estradiol, and counted using a hemocytometer. we also report new conditions under which estrogen respon Thymidine Kinase Assay. Cells were cultured in a manner identical siveness can be regained. to that described for [3H]thymidine incorporation except that the step involving the addition of [3H]thymidine was omitted. Cells were har vested with a rubber policeman in 200 fi\ of 0.05 M Tris-HCI:1 mM MATERIALS AND METHODS dithiothreitol:0.25 M sucrose buffer, pH 7.8, and sonicated. The ho Materials. [3H]Thymidine (48 Ci/mmol) and 17/J-[3H]estradiol (108 mogenate was spun at 20,000 x g for 10 min, and the supernatant Ci/mmol) were obtained from the Radiochemical Centre, Amersham, was transferred to a Sarstedt tube. Blue dextran (10 /il of a 10% England. 17/3-Estradiol was obtained from Sigma Chemical Co., Poole, solution) was added, and the mixture was loaded onto a Sephadex G- 25 column (2 ml) equilibrated with the Tris:dithiothreitol:sucrose buffer ' Supported by grants from the Medical Research Council and the North West at 4Â°. The column was eluted with Tris:dithiothreitol:sucrose buffer, Cancer Research Fund. 2 To whom requests for reprints should be addressed. 3The abbreviations used are NBCS, newborn calf serum; PBS, phosphate- Received May 26, 1982; accepted November 23, 1982. buffered saline.

1244 CANCER RESEARCH VOL. 43

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1983 American Association for Cancer Research. Estrogen-responsive MCF-7 Cells and the blue fraction (excluded volume containing all molecules with Table 1 molecular weight greater than 5000) was collected and assayed for Effects of estradiol and tamoxifen on [3H]thymidine incorporation in ' 'unresponsive MCF- 7 celts thymidine kinase activity as described by Green and Martin (7). The final assay mix (100 fil) contained 4.4 /Â¿molTris-HCI (pH 7 to 8), 0.44 Cells were plated with 5% NBCS medium (2 ml) in 35-mm-diameter culture dishes. After 24 hr, the medium was replaced with 2 ml of fresh medium /Â¿molMgCI2,0.089 /unol NaF, 17.8 /Â¿molunlabeled thymidine, 1.0 /Â¿mol containing 0.5% steroid-depleted serum. After a further 24-hr, the medium was [3H]thymidine, and 50 /Â¿Iofenzyme extract. replaced with 2 ml of fresh medium containing 0.5% steroid-depleted serum together with estradiol (10 8 M) and/or tamoxifen (10~6 M). The cells were DMA Polymerase Assay. Cells were cultured and harvested as for the thymidine kinase assay. DMA polymerase activity was measured as incubated in these media for 48 hr, and 0.5 iiCi of pHJthymidine was added for the final 2 hr. Thymidine incorporation rates were determined as described in incorporation of [3H]dTTP into trichloroacetic acid-insoluble product "Materials and Methods." using the basic method of McLennan and Keir (12) except that acti vated calf thymus DNA was used as the template primer. Assay mixture (75 til) contained 10 mM potassium phosphate (pH 7.4), 4 mM MgSO4, x 10~Vmg pro- 1 mM mercaptoethanol, 15 iig activated DMA, 50 /Â¿MdATP, 50 /Â¿M TreatmentControl tein/hr%17.3 dCTP, 50 /IM dGTP, 0.1 iiCi [3H]dTTP, and enzyme extract. After 1 hr Â±1.6a 120Â° at 37Â°,the reaction was stopped by adding 50 /Â¿Iof50% trichloroacetic Estradiol 20.7 Â±1.8 acid and cooling the mixture on ice. The acid-insoluble product was Tamoxifen 6.7 Â±0.7 39 Tamoxifen + estradiolpHJThymidmeincorporationdpm16.6 Â±1.2100 98C collected on GF/C filters. Mean Â±S.E. of triplicate determinations. Estradici Receptor Assays. MCF-7 cytosol and nuclear extracts " Stimulation by estradiol not significant by ( test ( p = 0.05). were prepared and assayed using the single-dose protamine sulfate c Stimulation by estradiol significant by f test (p = 0.01 ). method of Zava and McGuire (28), except that the assay for total nuclear receptors was carried out at 30Â°for 3 hr. Table 2 Karyotyping. Logarithmically growing cells were treated with col- Cytoplasmic and nuclear estrogen receptor contents of control and estradiol- chicine at a final concentration of 0.0004% for 4 to 5 hr and then treated unresponsive MCF-7 cells swollen with hypotonie solution (0.075 M KCI) and fixed with metha- MCF-7 cells were grown for 5 days in glass roller bottles containing 5% nol:acetic acid (3:1, v/v). Drops of the fixed cell suspensions were steroid-depleted NBCS with or without estradiol (10s M). The cells were then harvested, and cytosol and nuclear extracts were prepared and assayed, in allowed to dry on clean slides which were stained by the trypsiniGiemsa triplicate, for estrogen receptors as described in "Materials and Methods." Each banding method (27). The resulting preparations were examined with result represents mean total estradiol binding minus mean nonspecific estradiol a Zeiss photomicroscope III and photographed with llford Pan F emul binding. sion. [The karyotyping technique used was similar to that of Evans ef cytoplas nuclear recep al. (6) and the photographic techniques of Davidson (4).] mic receptors (pmol es- tors (pmol estradiol/ CellsControlEstradiol-treatedUnoccupiedtradiol/mgDNA)0.73 mgDNA)0.57 Â±0.06a RESULTS Â±0.05 None detectedTotal 0.37 Â±0.02 Loss of Response to Estradici. During the first 6 months Mean Â±S.E. of difference. after receiving cells from the Michigan Cancer Foundation, we were able to confirm the estrogen responsiveness of the MCF- estrogen response in these cells (8). Also, despite the lack of 7 cell line such that a 48-hr exposure to 1CT8 M estradiol, in a response to estradiol alone, an estradiol-reversible response medium containing 0.5% NBCS, produced a 3-fold increase in to the antiestrogen tamoxifen was still observed (Table 1). [3H]thymidine incorporation and a 2- to 3-fold increase in Strobl and Lippman (23) rationalized the loss of response to thymidine kinase activity (19). These changes in thymidine estradiol in terms of long-term retention of estradiol derived incorporation were also reflected in an increase in cell number from the serum used to supplement the growth medium. Thus, yield which became detectable 5 days after treatment with cells maintained in medium containing steroid-depleted serum estradiol. However, during the next 9 months, the behavior of might be expected to regain estrogen responsiveness. How the cells became markedly different. Estradiol (1CT8 M) now ever, we found that, even after 3 months (about 50 cell dou failed to produce a detectable increase in cell number above blings) in steroid-depleted medium, the cells still did not re control, stimulation of [3H]thymidine incorporation was reduced spond to estradiol (10~6 to 10~'Â°M, data not presented). to less than 30% above control (e.g., Table 1), and thymidine It also appeared unlikely that the cells were synthesizing kinase activity was no longer stimulated (data not presented). sufficient endogenous estradiol for maximal stimulation, be As reported by others (23), stock cells which had been stored cause inclusion in the culture medium of an inhibitor [4-hy- droxy-4-androstene-3,17-dione (10~8 to 10~5 M)] of estradiol frozen since the time of full response also now failed to respond to estradiol. synthesis (2) failed to produce an estradiol-reversible inhibition Characteristics of the "Unresponsive" Cells. The loss of of growth (results not shown). response to estradiol was not due to a loss of estradiol receptor It has also been suggested that the response of these cells because cells grown in the absence of estradiol possessed to estradiol may vary with cell density in culture (8). To test normal complements of cytoplasmic and nuclear receptors this, MCF-7 cells were plated at 2 different cell densities, one (Table 2). Furthermore, when grown in the presence of estra 5 times the other, and then tested for their response to estradiol diol, cells showed a complete absence of cytoplasmic recep and antiestrogens. The results in Table 3 show that indeed the tors (Table 2), suggesting that receptor translocation was active response of the cells, as determined by thymidine incorporation in these cells. Not only were receptors translocated into the rates, is dependent on cell density. Cells plated at low density nucleus but they were also "processed" as shown by lower were not responsive to estradiol but were inhibited by anties total cellular receptor content of estradiol-treated cells (Table trogens. Cells plated at high density, on the other hand, did 2) and by the decline in total specifically bound estradiol in show a minimal stimulation by estradiol but were no longer inhibited by 10~6 M antiestrogens. The reason for this variation whole cells after estradiol administration (Chart 1). Such "processing" has been suggested to be an integral part of the in responsiveness is not clear since the control rates of thymi-

MARCH 1983 1245

We were concerned that the loss of 10 chromosomes may indicate that we were no longer culturing an MCF-7 cell line. Nelson-Rees ef al. (18) reported 5 marker chromosomes in MCF-7 but indicated that there may have been others. We have prepared karyotypes by Q and G banding. The karyotype shown in Fig 1 is an example of a trypsin:Giemsa-stained 6 preparation which contains 80 chromosomes, 10 of which are marker chromosomes. Marker chromosomes 1, 2, and 4 are morphologically similar to those found by Nelson-Rees ef al. processed (18) which would indicate that we are still working with an MCF-7 cell line. RnE Table 3 Effect of cell density on the response of thymidine incorporation rate to estradiol and antiestrogens H Cells from the same stock culture were plated out at 2 different densities, one â€¢3 (confluent) 5 times the other (subconfluent). The experiment was then conducted as described in Table 1. I incorpora tion rate (dpm x 10~"/mgprotein/hr)7.1 mediumA. Addition to Confluent cells Control Â±0.4a (100)" Estradiol (10 8 M) 9.6 Â±0.6 (135)Â° Tamoxifen (10 6 M) 6.9 Â±0.5 (97)" NitromifenedO 6 M) 6.8 Â±0.7 (97)d Estradiol (10 " M) + tamoxifen (10~6 M) 9.6 Â±0.6 (135)c Estradiol (10 ' M) + nitromifene (10~6 M)Thymidine 9.8 Â±0.9 (138)Â°

B.cellsControlEstradiol Subconfluent Â±0.4(100)6.3 (10"SM)Tamoxifen Â±0.3(88)e3.9 M)NitromifenedO(10~6 Â±0.2(54)'3.4 6M)Estradiol Â±0.3(47)'7.1 M)Estradiol(10 '" M) + tamoxifen (10~6 Â±0.5(99)e7.4 HOURS (10 " M) + nitromifene (10~6 M)7.2 Â±0.6 (104)e Chart 1. "Processing' of estradiol receptor (flnÂ£) in "unresponsive' MCF-7 Ã¤Mean Â±S.E. of triplicate determinations. cells. Cells were cultured for the indicated times with either 5 nM | 'HJestradiol or Numbers in parentheses, percentage. 5 nM [ 'Hjestradiol plus 500 nM unlabeled estradiol. The cell monolayers were c Stimulation by estradiol significant by ( test. then washed thoroughly with ice-cold PBS, and the cells were harvested. After Inhibition by antiestrogen not significant by f test (p = 0.05). e Stimulation by estradiol not significant by f test (p = 0.05). a further washing with PBS, the cell pellet was extracted twice with ethanol, and the extract was assayed for radioactivity. All points have been corrected for Inhibition by antiestrogen significant by f test. nonspecific binding. The decline in total receptors from the initial optimal value to the value at 5 hr represents those receptors that are said to have undergone intranuclear processing (8) Arrow, the mean chromosome number. dine Incorporation (expressed as dpm/mg protein) are identical at the 2 cell densities. Nevertheless, we were unable to regain a full estrogen response by testing either confluent or subconfluent cells. The growth and macromolecular synthesis of these cells was 20 â€” originally stimulatable by pharmacological levels of androgen (8). It was of interest, therefore, to determine whether the cells had lost their response to androgen as well as to estradiol. We I6 found that, whereas at the time when they were capable of being stimulated by estradiol they were also responsive to testosterone, they were now refractory to both hormones. The growth-stimulatory effects of androgens are reported to occur via the estrogen receptor (29, 30). O We examined the karyotype of our unresponsive cells to see 4 whether their change could be due to a change in chromosome U number. A stem line of 88 chromosomes in the 39th passage C was reported by Soule et al. (21) with a range of about 80 to i 95 chromosomes. Nelson-Rees ef al. (18) reported a modal number of 87 chromosomes with a range of 79 to 89. We have n found a mean chromosome number of 78 (Chart 2) with a 65 70 75 80 90 range of 69 to 85 chromosomes. From this evidence, the MCF- CHROMOSOMES PER CELL 7 cell line appears to have lost 10 chromosomes from the Chart 2. Variation in chromosome content of MCF-7 cells in ninth year of modal number since 1975. culture. Arrow, the mean chromosome number.

1246 CANCER RESEARCH VOL. 43

To confirm that there had not been a genetic change in our cells, we grew from frozen stocks cells of the same passage number as those that had shown a full response to estradiol. When tested, these cells also failed to respond to estradiol with an increase in thymidine incorporation rate. Evidently, the loss of hormone response was not confined to our continuously cultured cells but was also shown by cells frozen at a time when a full response was shown. This confirmed our conclusion that the loss of response is a phenotypic change in response to some environmental stimulus. Serum Regulation of Estrogen Response. During an inves tigation of the serum requirements of the "unresponsive" MCF- 7 cells, it was found that the serum content of the culture medium had a profound effect on the estrogen responsiveness of the cells. While the addition of 10~8 M estradiol to cells growing in medium containing 0.5% NBCS had no effect on cellular growth rate, a significant increase in cell numbers was seen in response to estradiol for cells cultured in medium containing 15% NBCS. This growth response to 10~8 M estradiol was reflected in altered [3H]thymidine incorporation rates (Table 4). Once again, while no response was seen in medium containing 0.5% NBCS, estradiol almost doubled the rate of thymidine incorporation in cells grown in 15% NBCS, even if the serum had not been depleted of steroids. Use of 15% NBCS treated with dextran- coated charcoal did not affect the magnitude of this estradiol response (Table 4). Because the control rate of [3H]thymidine 2468 incorporation was depressed by 15% serum, it could not be DAYS IN CULTURE concluded from these data alone that estradiol was stimulating Chart 3. Stimulation of growth by estradiol in the presence of 15% serum. MCF-7 cells were plated in 25-sq cm Falcon flasks with medium containing 5% growth rather than rescuing cells from an inhibition of growth. NBCS. The next day, the medium was replaced with fresh medium containing However, the results (Table 4) obtained from human serum either 0.5% NBCS (â€¢)plus or minus 10 e M estradiol (curves identical) or 15% NBCS plus (â€¢)or minus (O) 10 8 M estradiol. Cells were harvested at the times (outdated serum obtained from the Mersey Regional Health indicated and counted in a hemocytometer. Results are the mean of triplicate Authority Blood Transfusion Centre) show that a depression of the control rate of [3H]thymidine incorporation was not a pre determinations; oars, S.E. requisite for an estradiol response. It seems most reasonable medium could be synthesizing DNA at a lower rate than did to assume that serum from the 2 sources promotes estradiol cells in low-serum-containing medium. Thus, the apparent de responsiveness by the same mechanism. Similarly, Chart 3 pression of thymidine incorporation by 15% NBCS is almost shows not only that cells in 15% NBCS responded to estradiol certainly an artifact, possibly due to dilution of radiolabeled with increased growth but also that the growth rate of control thymidine with unlabeled thymidine from the serum. Chart 4 cells of 15% NBCS was higher than that of cells in 0.5% serum. shows the results from an experiment in which the ability of It is not possible, therefore, that cells in high-serum-containing 10~8 M estradiol to stimulate [3H]thymidine incorporation by MCF-7 cells was tested at a series of serum concentrations. In 3 such experiments, the optimum concentration of serum for Table 4 Effects of serum on the stimulation of l'HÂ¡thymidine incorporation by estradiol the response to estradiol was in the range of 12 to 25% by Cells were plated with 5% NBCS in 35-mm-diameter culture dishes. After 24 volume. No significant stimulation was seen at serum concen hr, the medium was replaced with fresh experimental medium containing, where trations of 5% or below. necessary, estradiol (1CT8 M). The cells were incubated in these media for 48 hr, Estradiol has been shown to affect DNA precursor uptake and the thymidine incorporation rates were determined during the final 2 hr. and intracellular pool sizes in MCF-7 cells (1 ); this presumably in corporation(dpmx could be the reason for the more modest increase in cell 10~Vmgpro-tein/hr)9.5 numbers (50% by 8 days) compared to thymidine incorpora Treatment0.5% tion. However, Aitken and Lippman (1), after careful measure Â±0.7a8.9 NBCS0.5% ment of precursor pool sizes, were able to conclude that estradiol15%NBCS + Â±0.74.4 NBCS1 Â±0.48.3 estradiol does indeed stimulate true rates of DNA synthesis, estradiol15% NBCS + Â±0.63.4 which would agree with our demonstration of an estradiol NBCS15% steroid-depleted Â±0.26.9 stimulation of growth rate (Chart 3). Autoradiography of MCF- estradiol0.5%5% steroid-depleted NBCS + Â±0.212.4 7 cells labeled with [3H]thymidine has shown that the effect of serum0.5%human Â±0.815.6 estradiol1human serum + Â±2.213.5 estradiol is to shorten the cell cycle time of all the cells rather serum15%5% human Â±1.432.9 than to stimulate a previously noncycling fraction of cells to human serum + estradiol[3H]Thymidine Â±2.3%10094100189B100203ft126Â°100244b divide (25). Mean Â±S.E. of triplicate determinations. 0 Stimulation by estradiol significant by i test (p = 0.01). The response observed in 15% serum was specific for estro c Stimulation by estradiol not significant by f test ( p = 0.05). gen inasmuch as no stimulation of [3H]thymidine incorporation

MARCH 1983 1247

300 significant variation was seen with different batches of NBCS. Our current gel filtration experiments indicate that a serum factor with a molecular weight of about 300 to 1000 is involved 250 in the "serum response."

O ff K 200 Z DISCUSSION O o MCF-7 cells as they were first described appeared to present 1 50 an excellent in vitro model for the study of cellular growth regulation by estrogen. In addition, because of the origin of the 100 cells, it was believed that results obtained with them might be relevant to the treatment of certain types of breast tumors. More recently, however, from our own results, personal com 10 IS 20 munications from other researchers, and a few publications (5, 9, 10, 23), it became clear that the responsiveness of MCF-7 % CAUF SERUM cells to estradiol had become drastically reduced. Chart 4. Estradiol stimulation of [3H]thymidine incorporation by MCF-7 cells Karyotypic analysis showed that the "unresponsive" cells at different calf serum concentrations. MCF-7 cells were plated in 60-mm- were still MCF-7 cells rather than a contaminating (e.g., HeLa) diameter culture dishes with medium containing 5% NBCS. The next day, the medium was replaced with fresh medium containing the indicated concentration cell type. Also, despite the lack of a growth response, the of NBCS with or without estradici (10s M). The cells were incubated in these estrogen receptor system of the cells still appeared to be media for 48 hr, and the thymidine incorporation rates were determined during the final 2 hr. All results were expressed as dpm pHJthymidine incorporated per present and functioning. It had already been suggested (23, mg protein per hr before being presented as percentages of control rates. Results 24) that MCF-7 cells might not respond to added estradiol if are the means of triplicate determinations. The S.E.s of the triplicates used in the they were prestimulated by estrogen derived from the serum. percentage calculations were generally less than 10%. However, the continued lack of response after prolonged Table 5 growth of the cells in medium depleted of steroid hormones or Specificity of estrogen response in 15% NBCS in the presence of an inhibitor of estrogen synthesis showed that the "unresponsive" cells were not being prestimulated by Experimental conditions were the same as for Table 4 except that the cells were grown in multiwell plates (Flow Laboratories, Ltd.; 24- x 2-sq cm wells) either exogenous or endogenous estrogen. This conclusion is instead of individual culture dishes. incorporationTreatmentControl [3H]Thymidine also supported by the fact that unoccupied estrogen receptors were present in the cytoplasm of cells growing in the normal x10"Vmgprotein/hr3.44 experimental medium. The translocation and processing of receptors following ad Â±0.10a ministration of estradiol, together with the continued response Progesterone (10 8 M) 3.58 Â±0.25 10487 Dihydrotestosterone (10~e M) 2.99 Â±0.23 to antiestrogen, suggested that the receptor machinery of the 21 76 Estradiol (10 8 M)dpm 7.46 Â±0.78%100 cells was still functioning. Thus, it was thought to be possible that the "unresponsive" cells were still capable of responding Mean Â±S.E. of 4 replicates. ' Stimulation by estradici significant by i test ( p = 0.01 ). to estradiol but that the primary actions of the estradiolireceptor complex were not being translated into a growth response. In was seen with either progesterone or dihydrotestosterone (Ta this case, the lack of response of cryopreserved early-passage ble 5). Other growth-related responses which are not shown in (responsive) cells would indicate that the culture conditions low serum (5) but which are expressed in the presence of 15% were no longer conducive to the expression of a growth re serum are stimulations of both thymidine kinase activity [con sponse to estradiol. trol, 2.12 Â±0.02 (S.E.) estradiol, 3.75 Â±0.16 pmol dTMP per A reexamination of the requirements of MCF-7 cells for mg protein per min] and DNA polymerase a activity (control, serum, the only undefined culture variable, showed that con 21.7 Â± 1.9; with estradiol, 38.5 Â±3.1 dpm X 10~4 per mg ditions favorable to the expression of a growth response to protein per hr). Although the restoration of estradiol respon estradiol could be created by increasing the concentration of siveness by 15% serum was also shown by MCF-7 cells ob serum in the experimental medium to 15% (v/v). Thus, the "unresponsive" cells were indeed still capable of showing a tained from Dr. Marc Lippman, recent batches of cells obtained from Dr. Marvin Rich failed to respond to estradiol, even in the full specific response to estradici. presence of 15% serum. Other workers have also reported a Whether the original change in behavior of MCF-7 cultures difference in estradiol responsiveness between MCF-7 cells was due to a change in composition of serum or to a change in obtained from these 2 sources (26). Thus, it appears that sensitivity of the cells to serum factors is as yet unknown. variant sublines of MCF-7 may have developed in some labo However, it is clear that both NBCS and human serum contain ratories. a factor, or factors, which can influence the expression of a The original change in the behavior of MCF-7 cultures may growth response to estradiol. We are currently investigating have been due to a change in composition of serum or to a both the identity of the serum factor(s) responsible for this change in sensitivity of the cells to serum factors. In this activity and the behavior of the receptor system under respon connection, it is interesting to note that the human serum sive and "unresponsive" conditions. Preliminary gel filtration batches from individual donors showed considerable variation experiments indicate that the "factor" has an apparent molec in their ability to support an estrogen response, although all ular weight of 300 to 1000. were able to support MCF-7 cell growth. Lesser although still A small pituitary factor (M, < 1000) present in fetal calf

1248 CANCER RESEARCH VOL. 43

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1983 American Association for Cancer Research. Estrogen-responsive MCF-7 Cells serum has been described that specifically stimulates the Effects of estradiol on estrogen receptors and nuclear receptor processing. J. Biol. Chem., 253. 8185-8191, 1978. growth of a line of hormone-dependent rat mammary tumor 10. Jozan, S., Moure, C., Gillois, M., and Bayard, F. Effects of estrone on cell cells and that synergizes with other growth-stimulatory hor proliferation of human breast cancer (MCF-7) in long term tissue culture. J. Steroid Biochem., 10: 341-342, 1979. mones (12). This mammary growth factor has subsequently 11. Kano-Sueoka, T., Cohen, D. M., Yamaizumi, Z.. Nishimura. S.. Mori, M., and been identified as phosphoethanolamine (11). Sirbasku (20) Fujiki, H. Phosphoethanolamine as a growth factor of a mammary carcinoma has presented evidence for specific tumor growth factors the cell line of rat. Proc. Nati. Acad. Sei. U. S. A., 76. 5741-5744, 1979. 12. Kano-Sueoka. T., Errick, J. E., and Cohen, D. M. Effects of hormones and production of which is under estradiol control and that are a novel mammary growth factor on a rat-mammary carcinoma in culture. In: responsible for the growth of hormone-responsive tumors in G. H. Sato and R. Ross (eds.). Hormones and Cell Culture, pp. 499-512. whole animals. The relationship of these tumor growth factors Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory, 1979. 13. Lippman, M. E., and BolÃ¡n, G. Oestrogen-responsive human breast cancer to our serum factor is the subject of present investigations. in long term tissue culture. Nature (Lond.), 256. 592-593, 1975. Sudden changes in responsiveness to estradiol are not re 14. Lippman, M., BolÃ¡n, G., and Huff, K. The effects of estrogens and antiestro- stricted to cultured cells, inasmuch as many estrogen receptor- gens on hormone-responsive human breast cancer in long-term tissue culture. Cancer Res., 36. 4595-4601, 1976. positive human breast tumors which initially respond to endo 15. Lippman, M. E., BolÃ¡n, G., Monaco, M. E., Pinkus, L., and Engel, L. Model crine therapy (i.e., are estrogen responsive) eventually pro Systems for the study of estrogen action in tissue culture. J. Steroid Bio chem., 7. 1045-1051, 1976. gress to an apparently unresponsive state (22). Such tumors 16. Lowry, O. H., Rosebrough, N. J., Farr A. L.. and Randall, R. J. Protein usually retain estrogen receptors and may therefore resemble measurement with the Folin phenol reagent. J. Biol. Chem., )93. 265-275, the "unresponsive" MCF-7 cells. Our results suggest that 1951. 17. Nawata, H., Bronzert, D., and Lippman. M. E. Isolation and characterisation changes in the responsiveness of some breast tumors to estro of a tamoxifen-resistant cell line derived from MCF-7 human breast cancer gen may be the result of mechanisms other than the selection cells. J. Biol. Chem., 256: 5016-5021, 1981. of a nonresponsive variant and that MCF-7 cells are still a 18. Nelson-Rees. W. A., Flandermeyer, R. R., and Hawthorne, P. K. Distinctive banded chromosomes of human tumour cell lines. Int. J. Cancer. 74: 74- useful model system for studying estrogen-responsive breast 82, 1975. cancer. 19. Page, M. J., Field, J. K.. and Green, C. D. The response to hormones and antihormones of human breast tumour cells grown in continuous culture. Biochem. Soc. Trans., 6. 1318-1319, 1978. REFERENCES 20. Sirbasku, D. A. Estrogen-induction of growth factors specific for hormone- responsive mammary, pituitary and kidney tumor cells. Proc. Nati. Acad. Sei. U. S. A. 75. 3786-3790, 1978. 1. Aitken, S. C., and Lippman, M. E. In: L. Jimenez de Asau (ed.). Control 21. SoÃ»le,H. D.. Vasquez, J.. Long, A., Albert, S., and Brennan, M. J. Human Mechanisms in Animal Cells, pp. 133-155. New York: Raven Press, 1980. cell line from a breast carcinoma. J. Nati. Cancer Inst., 5Ã•. 1409-1415. 2. Brodie, A. M. H., Marsh, D. A., Wu, J. T., and Brodie, H. J. Aromatase 1973. inhibitors and their use in controlling oestrogen-dependent processes. J. 22. Stoll. B. A. Selective mechanisms in hormonal therapy. In: B. A. Stoll (ed.), Steroid Biochem. 11: 107-112, 1979. Secondary Spread in Breast Cancer, pp. 169-191. London: Heinemann, 3. Chen, T. R. In situ detection of mycoplasma contamination in cell cultures 1977. by fluorescent Hoechst 33258 stain. Exp. Cell Res., 104: 255-262, 1977. 23. Strobl. S. J.. and Lippman, M. E. Studies of steroid hormone effects on 4. Davidson, N. R. Photographic techniques for recording chromosome band human breast-cancer cells in long-term tissue culture. In: W. L. McGuire ing patterns. J. Med. Genet.. 10: 122-126, 1973. (ed.). Hormones, Receptors and Breast Cancer, pp. 85-106. New York: 5. Edwards. D. P., Murthy, S. R., and McGuire, W. L. Effects of estrogen and Raven Press. 1978. antiestrogen on DNA polymerase in human breast cancer. Cancer Res., 40: 24. Strobl, J. S., and Lippman, M. E. Prolonged retention of estradiol by human 1722-1726, 1980. breast cancer cells in tissue culture. Cancer Res., 39. 3319-3327, 1979. 6. Evans, H. J., Bucktown, K. E.. and Sumner, A. T. Cytological mapping of 25. Weichselbaum, R. R., Hellman. S., Piro, A. J., Nove, J. J., and Little, J. B. human chromosomes. Results obtained with quinacrine fluorescence and Cancer Res.. 38. 2339-2342, 1978. the acetic-saline-giemsa techniques. Chromosoma (Beri.), 35. 310-325, 26. Westley, B., and Rochefort, H. A secreted glycoprotein induced by estrogen 1971. in human breast cancer cell lines. Cell, 20. 353-362, 1980. 7. Green, C. D., and Martin, D. W. Characterisation of a feedback-resistant 27. Worton. R. G., and Duff, C. Karyotyping. Methods Enzymol.. 58. 322-343. phosphoribosylpyrophosphate synthetase from cultured, mutagenised hep- 1979. atoma cells that overproduce purines. Proc. Nati. Acad. Sei U. S. A.. 70. 28. Zava, D. T., and McGuire, W. L. Estrogen receptor. Unoccupied sites in 3698-3702, 1973. nucleic of a breast tumor cell line. J. Biol. Chem., 252. 3703-3708, 1977. 8. Horwitz, K. B., Koseki, Y., and McGuire, W. L. Estrogen control of proges 29. Zava, D. T., and McGuire, W. L. Androgen action through estrogen receptor terone receptor in human breast cancer: role of estradiol and anti-estrogen. in a human breast cancer cell line. Endocrinology, ! 03. 624-631, 1978. Endocrinology, 103: 1742-1751, 1978. 30. Zava, D. T., and McGuire, W. L. Human breast cancer: androgen action 9. Horwitz. K. B., and McGuire, W. L. Nuclear mechanisms of estrogen action. mediated by estrogen receptor. Science (Wash. D. C.) 199: 787-788, 1978.

MARCH 1983 1249

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1983 American Association for Cancer Research. M. J. Page et al. Il 1 III ))( ili J L 11%UH0l mi un limi )ij l 6 _ 7 _ 8 _ 9 _ 10 _ 11_ 12 _ X

â€¢il Â«III il H II IIÂ» 13 _ 14_ 15 _ | D â€¢ff â€¢â€¢â€¢â€¢ ,,** | 19 20 ! 21 22 UnG imiÂ» MINUTES mj ma m3 m4 m5 mg m7 mg mfl m1Q MARKERS Fig. 1. Karyotype of MCF-7 trypsin:Giemsa-stained preparation of metaphase spread. The chromosomes have been ordered into Groups A to G, minutes, and MCF-7 marker chromosomes.

1250 CANCER RESEARCH VOL. 43

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1983 American Association for Cancer Research. Serum Regulation of the Estrogen Responsiveness of the Human Breast Cancer Cell Line MCF-7

Martin J. Page, John K. Field, Nic P. Everett, et al.

Cancer Res 1983;43:1244-1250.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/43/3/1244

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/43/3/1244. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.