[CANCER RESEARCH 41, 4644-4652, November 1981]0008-5472/81 70041-OOOOS02.00

Failure of Estradici Immunofluorescence in MCF-7 Breast Cancer Cellsto Detect Estrogen Receptors1

Wayne 0. Mercer,2 Dean P. Edwards, Gary C. Chamness, and William L. McGuire3

Department of Medicine. Division of Oncology. University of Texas Health Science Center at San Antonio, San Antonio. Texas 78284

ABSTRACT

An indirect immunofluorescence assay was used to detectestradiol in MCF-7 breast cancer cells to determine if theestradiol-specific fluorescence observed represented estrogenreceptor-bound estradiol. Appropriate controls were used to

demonstrate the ¡mmunological specificity of our assay procedures. Initial studies of estradiol binding in MCF-7 cells wereperformed at 20° for 1 hr with different concentrations of

estradiol. Cytoplasmic and nuclear staining were observedfollowing treatment with 10 nrvi estradiol, but not with lowerconcentrations which were nevertheless still sufficient to saturate estrogen receptor. The staining intensity increased withhigher estradiol concentration, which is consistent with estradiol binding to lower-affinity binding sites. In order to further

determine if estradiol binding by estrogen receptor was beingdetected, we pretreated MCF-7 cells with 5 nw diethylstilbestrolat 37° for 1 hr to translocate all estrogen receptor to the

nucleus and then administered estradiol at varying concentrations for 4 hr at 4°.The estradiol was still primarily detected in

the cytoplasm, although virtually all of the estrogen receptorwas found to be present in the nucleus by standard [3H]estra-

diol binding assays. Additional immunochemical studies usingsucrose gradient analysis to detect antibody-estradiol-receptor

complexes clearly established that these complexes could notbe detected.

The present results suggest that, although immunocyto-chemical assays can specifically detect estradiol in MCF-7cells, the estradiol is bound to lower-affinity binding sites ratherthan to estrogen receptor. Saturation analyses of intact viableMCF-7 cells performed at 37°for 30 min using [3H]estradiol at

concentrations ranging from 0.1 to 93 nw revealed an additional lower-affinity estradiol-binding site besides the receptor,

perhaps analogous to the Type II sites reported in the rat uterusand human breast cancers.

INTRODUCTION

Only about one-third of patients with metastatic breast can

cer will benefit from endocrine treatment or ablative surgery.The original suggestion by Jensen ef al. (18) that estrogenreceptor assays could be used to more accurately predict theestrogen dependence of breast tumors has been amply supported by clinical studies (26). More recent studies have alsoestablished that the effective use of estrogen receptor assays

' Supported in part by National Cancer Institute Grants CB 23862 and CA

22343; NICHO P30 HD 10202; and NICHD Training Grant HD 07139.2 Present address: Kallestad Laboratories, P. O. Box 18029, Austin, Texas

78760.3 To whom reprint requests should be addressed, at Department of Medicine,

Division of Oncology, University of Texas Health Science Center at San Antonio.7703 Floyd Curl Drive, San Antonio, Texas 78284.

Received October 9, 1980; accepted August 6. 1981.

can be enhanced through the measurement of progesteronereceptor, a product of estrogen stimulation (16, 35). But it isclear that the distribution of these receptors in frequentlyheterogeneous tumor cell populations could seriously affectour conclusions. Furthermore, the ability to detect receptors ina relatively small sample of tumor cells such as a pleuraleffusion or a needle biopsy would be helpful. A histologicalmethod for revealing estrogen and progesterone receptorswould therefore be highly desirable.

Such histological assays could be approached through theuse of antibodies to the receptors, or indirectly through the useof antibodies to the steroid hormones to detect the receptor-bound steroids, or by using a labeled ligand which will bind tothe receptor yet can be detected microscopically. Although anantibody to the estrogen receptor has been prepared (14), itsuse in immunohistochemical procedures has not been reported. However, several laboratories have reported the use ofimmunofluorescence (27-34, 37, 39) and immunoperoxidasetechniques using antibody to estradiol (13, 20, 21, 28-31 ), aswell as cytochemical techniques using fluorescein-estradiolconjugates (22, 23, 38, 40), to detect the binding of steroidhormones in human or animal estrogen-dependent tissues andtumors. The results of these studies seemed to suggest thatimmunocytochemical assays could be used to detect estradiolbinding which might correlate with the presence of estrogenreceptors, and therefore, be useful clinically.

In preliminary studies, we observed that (a) a correlationexisted between the percentage of estradiol-binding cells detected by immunocytochemical assays and the amount of estrogen receptor present in human breast cancers, (b) estradiolbinding was restricted to estrogen target tissues and sometumors derived from target tissues in humans and rats, and (c)the percentage of estradiol-binding cells correlated with theamount of tumor regression in mammary tumor-bearing rats(22-29). However, when we further examined the specificity

of the immunocytochemical assays by using human breastcancer cell lines as model systems, we found that high concentrations of estradiol were required to observe estradiol-specific staining. In addition, when we attempted to use DES"

as a competitive inhibitor of estradiol binding, we were not ableto detect reproducible inhibition of staining under our routineassay conditions (30).

We concluded from these studies that immunocytochemicalassays can be used to detect estradiol in tumor cells, but thatthe observed binding was probably due to multiple classes ofsteroid-binding sites, not only to receptor. The purpose of the

4 The abbreviations used are: DES, diethylstilbestrol; BSA, bovine serum

albumin; PBS, phosphate-buffered saline (0.01 M sodium phosphate:0.15 MNaCI, pH 7.4); KRH glucose, Krebs-Ringer-Henseleit glucose buffer (pH 7.3)containing (g/liter) NaCI (8.0), KCI (0.20), Na2HPO.. 7H2O (1.73), KH2PO4 (0.20),CaCI2 (0.10), MgCI2 (0.048), and glucose (1.0); PTG, phosphate buffer (5 mwsodium phosphate, pH 7.4:10 mM thioglycerol:10% glycerol).

4644 CANCER RESEARCH VOL. 41

Research. on November 12, 2020. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Estradici Immunofluorescence in Breast Cancer

current study was to further examine the specificity of immu-

nocytochemical assays to determine if the detected estradiolbinding might be due, at least in part, to interaction with theestrogen receptor.

MATERIALS AND METHODS

Antiserum Preparation. Estradiol antiserum was prepared in female,albino New Zealand rabbits using 17/8-estradiol-6-O-carboxymethylox-ime-BSA (Steraloids, Inc., Wilton, N. H.) as the immunogen (8).

Adsorption of Anti-Estradiol Antiserum. BSA (fraction V; MilesLaboratories, Elkhart, Ind.) was coupled to glutaraldehyde-activatedpolyacrylamide beads (Biogel P-300, -400 mesh; Biorad Laboratories,

Richmond, Calif.) using the method of Ternynck and Avrameas (43).The anti-estradiol antiserum was diluted 1:5 with PBS and adsorbed 3

times with an equal volume of the immunadsorbent beads. Each adsorption was performed for 1 hr at room temperature.

A portion of the BSA-adsorbed anti-estradiol antiserum was thenadsorbed with crystalline estradiol or 17/?-estradiol-6-O-carboxy-

methyloxime for 1 hr at room temperature. Excess undissolved steroidwas removed by centrifugation at high speed for 2 min in a BeckmanMicrofuge B centrifuge. The adsorbed antisera were tested by radioim-

munoassay to confirm that the adsorptions were complete.Methods for Evaluation of Antiserum. The anti-estradiol antiserum

was tested for specificity using a radioimmunoassay we developed forroutine use (30). In brief, approximately 10,000 cpm of [2,4,6,7-3H]-

estradiol (102 Ci/mmol; New England Nuclear, Boston, Mass.) wereincubated with anti-estradiol antiserum diluted 1:6,000 (final dilution)

in radioimmunoassay buffer (0.01 M sodium phosphate:0.15 M NaCI:0.1% gelatin, pH 6.8) for 18 hr at 4°in the presence of estradiol or

other steroid hormones used at concentrations ranging from 0.1 MMto1 ¿IM.Following the incubation, excess steroid was removed by adsorption with dextran-coated charcoal [0.25% Norit A (Sigma ChemicalCo., St. Louis, Mo.):0.025% Dextran T-70 (Pharmacia, Uppsala, Sweden) in radioimmunoassay buffer] at 4° for 30 min. The charcoal

suspension was separated by centrifugation at 2000 x g for 15 min.The supernatant was poured into scintillation vials, 8 ml of aqueouscounting scintillant (Amersham/Searle Corp., Arlington Heights, III.)were added, and the samples were counted in a scintillation counter.Results are plotted (1 ) in Chart 1.

The specificity of the anti-estradiol antiserum was also checked by

incubating the antiserum at 1:6,000 final dilution with 10,000 cpm ofvarious 3H-steroid hormones to determine the percentage of binding of

those substances (results not shown). The assay incubation conditionswere the same as those described above. The 3H-steroid hormones

used were purchased from New England Nuclear and included[2,4,6,7-3H]estrone, 85 Ci/mmol; [1,2,6,7-3H]progesterone, 114 Ci/mmol; [1,2-3H]dihydrotestosterone, 40 Ci/mmol; and [1,2,6,7-3H]cor-

tisol, 80.6 Ci/mmol.Cell Lines and Culture Techniques. The MCF-7 breast cancer cell

line was originally provided by H. D. Soule, Michigan Cancer Foundation, Detroit, Mich. The routine culture conditions were as describedpreviously (15, 17). For experimental studies, the routine serum supplement was removed and replaced by 5% calf serum stripped ofendogenous steroids by a 30-min incubation at 45° with a dextran-

coated charcoal pellet (0.25% Norit A and 0.0025% dextran in 0.01M Tris-HCI, pH 8.0, at 4°, 1 ml/ml serum). Immediately prior to

incubations with hormones, the culture media were removed and replaced by KRH glucose solution. The hormones were added at 1000x concentrated solutions in absolute ethanol.

Cell Harvest. Cells were removed from the surface by a 10-minincubation at 37°with 1 mw EDTA in Ca2*-Mg2+-free Hanks' balancedsalt solution, washed once with Hanks' balanced salt solution at 4°and

once in PTG or KRH glucose solution. The phosphate buffer was usedif estrogen receptor assays were to be performed while the glucosesolution was used if immunocytochemical assays were to be performed.

Steroid Concentration (M)

Chart 1. Specificity of estradiol binding by rabbit anti-estradiol antiserum. Theantiserum was diluted to 1:6000 (final dilution) and incubated with 0.1 pmol[3H]estradiol in the presence of varying concentrations of steroid inhibitors. £,,estrone; E2, estradiol; £.,,estriol; DES; DHT, dihydrotestosterone; CORT, hydro-cortisone; and PRO. progesterone.

Immunofluorescence Studies. The immunocytochemical assayswere performed using frozen sections of cell pellets prepared from thecultured MCF-7 cells. The cells were incubated with the appropriate

concentration of estradiol or other hormone in KRH glucose solution,harvested with EDTA, and washed. The final cell pellet was resus-

pended in 0.2 ml KRH glucose solution and centrifuged for 1 min at25% of maximum speed in a Beckman Microfuge B centrifuge. Thesupernatant was discarded, and the cell pellet was resuspended in 0.1ml 2% ovalbumin prepared in 0.9% NaCI solution. The cell suspensionwas frozen with liqud nitrogen and stored at -70° until use. Frozen

sections (5 /¿m)were prepared in a cryostat microtome, placed onclean glass slides, and stored at -70° until used. The specimens were

briefly rinsed with PBS before fixation with 1% paraformaldehyde (10min; 20°).The sections were washed for 15 min in 3 changes of PBS

and incubated with a 1:20 dilution of rabbit anti-estradiol or control

reagents, as appropriate, in a humid chamber at room temperature for40 min. After another 15-min wash in PBS, the sections were incubatedwith a 1:20 dilution of fluorescein-conjugated goat antiserum to rabbit

immunoglobulins (Cappel Laboratories, Cochranville, Pa). All dilutionsof antiserum and conjugates were made in PBS.

The specimens were given a final rinse in PBS for 15 min and werecounterstained with Eriochrome black (1:100, 30 sec, Difco Laboratories, Detroit, Mich.) and mounted with a coverslip using PBS-glycerol

as a mounting medium (9, 42). The specimens were then examinedusing a Leitz Dialux 20 microscope with epiillumination, xenon lightsource, and a type K filter cube.

The results of the immunocytochemical evaluation were expressedas intensity of staining evaluated on a scale of negative through 3 + .Trace reactions were considered as negative.

Homogenization and Fractionation. The cells were suspended in a1:1 (v/v) ratio in PTG after harvesting and washing, and were homogenized by hand in a glass Duali homogenizer with a tight-fitting Teflon

pestle. Subcellular fractions were prepared by differential centrifugation. Homogenates were centrifuged at 800 x g for 10 min to yield acrude nuclear pellet and cytoplasm (supernatant) fraction. The nuclearpellet was washed one time by resuspending with appropriate bufferand recentrifugation at 800 x g. The supernatant wash was added tothe crude cytoplasm fraction, and this was centrifuged at 105,000 xg for 60 min to yield the cytosol supernatant and microsomal pellet.

Proteins were extracted from nuclear pellets by suspending thepellets for 1 hr at 4°in 1.0 ml of extraction buffer (10 mw Tris, pH 8.5:

1.5 mw EDTA:10% glycerol:0.6 M KCI). Extracted proteins were thenseparated from other nuclear materials by centrifugation at 105,000x g for 30 min.

Estrogen Binding to Cells. Whole-cell binding assays to detectestrogen receptors and other estradiol-binding sites were performedby incubating 1 million MCF-7 cells with varying concentrations of[3H]estradiol in a final volume of 1 ml KRH glucose solution for 30 min

NOVEMBER 1981 4645

Research. on November 12, 2020. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

W. D. Mercer et al.

at 37°. Duplicate assays were performed using 100-fold excess DESat each concentration of [3H]estradiol to estimate nonspecific binding.

Following the incubation, the cells were separated by centrifugationfor 5 min at 300 x g and washed 3 times with 2 ml cold KRH glucose.Radioactivity in the cell pellets was extracted with 2 ml ethanol. Theethanol was then mixed with 5 ml toluene-based scintillation fluid (4.0

g PRO, 0.05 g POPOP, and 1 liter toluene) in a Beckman LS 233 liquidscintillation counter with a counting efficiency of 50% for tritium.

Estrogen Receptor Assays. Unfilled cytosol and unfilled nuclearestrogen receptors were measured by a protamine sulfate assay asdescribed previously (45). Briefly, cytosol samples were diluted withPTG to adjust the protein concentration to approximately 1.5 mg/ml;nuclear extracts were diluted 1:8. Aliquots of the cytosol (200 /il) andnuclear extracts (500 ¡i\)were then precipitated with 0.1% protaminesulfate (250 /il). The protamine precipitates were incubated at 4°for16 hr with 5 nM [3H]estradiol for cytosol and 10 nw [3H]estradiol for

nuclear samples. Nonspecific binding was determined by parallel incubations with [3H]estradiol plus a 100-fold excess of unlabeled DES.

Following incubation, the free estradiol in solution was aspirated, andthe protamine precipitates were washed 3 times with PTG. Radioactivityin the protamine pellets was extracted and counted in 5 ml of scintillation fluid.

Total receptors (unfilled plus estrogen-filled receptor sites) were

measured by an estradiol exchange assay (44) in protamine pellets ofcytosol and nuclear extract by incubation with [3H]estradiol or [3H]-estradiol plus DES at 30°for 5 hr. The difference between the valuesfor 30°binding (total receptor) and 4°binding (unfilled receptor) yields

the value for filled estrogen receptor sites.Saturation analyses of cytosol and nuclear estrogen-binding sites

were performed by incubating protamine-precipitated cytosol or nuclear extracts with various concentrations up to about 100 nM [3H]-estradiol at 4°for 16 hr. DES at 100-fold excess was used in duplicateassays at each concentration of [3H]estradiol to estimate the amount of

nonspecific binding. After incubation, the protamine pellets werewashed 3 times with PTG, extracted with 5 ml toluene-based scintilla

tion fluid, and counted. Data were analyzed using the method ofScatchard (41).

Sucrose Density Gradient Centrifugation. MCF-7 cytosol was pre

pared in PTG containing 0.4 M KCI and adjusted to approximately 5mg protein per ml solution. The cytosol (200 |il) was incubated with200 fi\ [3H]estradiol (1 nM in radioimmunoassay buffer) in the presence

or absence of 100-fold excess DES, anti-estradiol antiserum (100 /¿Iofa 1:2000 initial dilution prepared in radioimmunoassay buffer), or anti-

estradiol antiserum which had been preadsorbed with estradiol. Theincbation was performed at 4° for 4 hr with continuous low-speed

shaking. Two hundred /il of each sample were layered onto 5% to 20%linear sucrose gradients prepared in the PTG-KCI buffer. ['"CJBSA was

used as an internal 4.6S marker. The gradients were centrifuged in aBeckman SW 60 rotor at 53,000 rpm (average, 246,000 x g) for 16.3hr. Fractions (200 /il) were collected and counted in aqueous countingscintillation fluid.

Protein and DMA Assays. Protein concentrations were measuredusing the method of Lowry er al. (25) with BSA standards. The amountof DNA was determined using the method of Burton (2).

RESULTS

All of the immunocytochemical studies presented here usedrabbit anti-estradiol antiserum which had been adsorbed withBSA, since the use of unadsorbed antiserum may result in apositive staining pattern due to the presence of cross-reacting

cellular antigens and/or BSA from the tissue culture medium(Table 1). In order to prove immunocytochemical specificity forestradiol, a portion of the primary antiserum was further adsorbed with either estradiol or estradiol-6-O-carboxymethylox-

ime. Staining was greatly reduced or abolished (Table 1). As anegative control, PBS was used in place of the primary anti-

serum to detect any nonspecific binding of the secondaryantiserum and/or autofluorescence.

For most routine studies, MCF-7 cells were exposed to

estradiol at room temperature for 1 hr. Under these conditions,estrogen-specific immunofluorescent staining was observed in

both cytoplasm and nuclei of cells treated with 10 nw estradiolor higher doses (Table 2). In general, the staining intensityincreased with higher estradiol concentration in both the cytoplasm and nucleus.

When MCF-7 cells were treated with estradiol at 4°,cyto-

plasmic (or plasma membrane) staining was predominant. Nuclear staining was also observed at 4°but this was associatedwith cytoplasmic staining (Fig. 1; Table 3). Warming to 37°after an initial incubation with estradiol at 4°caused a relative

increase in the nuclear fluorescence (Fig. 2; Table 3).These results suggested temperature-dependent transloca

tion of estradiol to nuclei, as expected for receptor. Becauseof the extremely high concentration of estradiol required, however, we examined estradiol immunofluorescence in MCF-7cells pretreated with 5 nw DES at 37° to induce complete

nuclear translocation of the estrogen receptor. DES was chosen as the ligand because it binds to the estrogen receptor butis not detected by the anti-estradiol antiserum at this concen

tration. Table 4 presents the receptor distribution measured incytoplasm and crude nuclei before and after DES treatment.We have shown elsewhere that the receptor associated withcrude nuclei before estrogen treatment is removed to thesoluble fraction by purification of the nuclei, while after treatment, all receptor is completely localized in purified as well as

Table 1Immuno/ogical specificity of estradiol binding in MCF- 7 cells

MCF-7 cells were grown to confluency in T-25 tissue culture flasks. Theroutine growth medium was replaced by KRH glucose solution, and the cellswere incubated with 10 nM estradiol for 1 hr at room temperature. The resultsare expressed as the average intensity of staining observed during 3 separateexperiments.

AdsorptionNone

BSAEstradiolBSA and estradiolBSA and estradiol 6-O-carboxymethylox-

ime%

of original anties-tradiol titer

after adsorption100

86.72.82.2

<0.1Results

of immunofluorescence testsusing MCF-7cellsCyto

plasm3+

1+1+Nucleus3+

2 +1+1+

Table 2

Effect of estradiol concentration on immunofluorescence staining observed inMCF-7 cells

MCF-7 cells were grown to confluency in T-25 tissue culture flasks. Theroutine growth medium was replaced by KRH glucose solution before treatmentwith estradiol for 1 hr at room temperature. The results are expressed as theaverage intensity of estradiol-specific staining observed during 3 separate experiments.

Concentration of estradioltreatmentNone1

nMestradiol10nMestradiol1

00 nMestradiol1JIMestradiolCytoplasm-—1

+2+3

+Nucleus-—1

+1+2

+

4646 CANCER RESEARCH VOL. 41

Research. on November 12, 2020. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Estradiol Immunofluorescence in Breast Cancer

crude nuclei (5). The results in Table 4 confirm that virtually allcytoplasmic receptor was translocated to nuclei by the DEStreatment. Nevertheless, after we then exposed the DBS-

treated cells to estradiol, the most intense fluorescent stainingwas still located in the cytoplasm (Figs. 3 and 4; summarizedin Table 5). The results of this experiment demonstrate clearlythat the estradiol-specific fluorescent staining observed wasnot due to receptor-bound estradiol.

We also wished to know whether estradiol attached to receptor could be detected by antibody at all. Using high-saltsucrose gradient sedimentation to detect the formation of corn-

Table3Effects of temperature on immunofluorescence staining for estradiol in MCF- 7

cellsMCF-7 cells were grown to confluency in T-25 tissue culture flasks. The

growth medium was replaced by KRH glucose solution before treatment withestradiol at the concentrations indicated for 4 hr at 4°.Parallel sets of cells weretreated with estradiol for 3.5 hr at 4°.followed by warming to 37°for 30 min in

the continued presence of hormone. The results are expressed as the averageintensity of staining observed during 3 separate assays of 2 experiments.

HormonetreatmentNone1

nMestradiol10 nMestradiol1

00 nMestradiolNone1

nMestradiol10nMestradiol1

00 nM estradiolTemperature4°4444.

37°4,374,374,37Cytoplasm-—1

+2+_—1

+1+Nucleus-—1

+1+_—2+2

+

Table 4Effect of DES on intracellular distribution of estrogen receptors in MCF- 7 cells

MCF-7 cells were treated with 5 nM DES for 1 h at 37°and estradiol treatmentwas performed for 4 hr at 4°as described for Table 5. Results are expressed inpmol of [3H]estradiol binding per mg cytosol protein.

CytoplasmicERaTreatmentNone/none

DES/noneDES/1 nM E2DES/10nM E,DES/1 00 nME2Free0.364

0.0200.0190.0020.016Occupied0.082

0.0830.0940.037Nuclear

ERFree0.177

0.0270.0190.0070.013Occupied0.632

0.6270.5810.639

ER, estrogen receptors; E2, estradiol.

Table 5

Effect of DES pretreatment on the immunofluorescence staining by antiestradiolserum in MCF- 7 cells

MCF-7 cells were grown to confluency in roller bottles The routine growthmedium was replaced by KRH glucose solution before the cells were pretreatedwith DES for 1 hr at 37°. The medium was poured off. and the cells were

harvested with EDTA and washed twice with Hanks balanced salt solution Thecell pellet was resuspended in 50 ml KRH glucose solution, and 10 ml of the cellsuspension was placed into sterile centrifuge tubes. The cell suspensions werethen exposed to estradiol or glucose solution, as appropriate, for 4 hr at 4°.

Following this incubation, the cells were collected by centrifugador! and washedtwice. An aliquot of the cell pellet was prepared for immunocytochemical testing.The remainder of the cell pellet was used in assays for cytosol and nuclearestrogen receptor. The results are expressed as the average intensity of stainingobserved during 3 separate assays.

Hormone treatment

DESNone5nM5

DM5DM5nMNoneNoneNoneEstradiolNoneNone1

nM10nw100nM1

nM10nM100

nMCytoplasm-——1

+2+—1

+2+Nucleus——_1

+1+—1

+1+

¿- 200 -

Chart 2. Effect of anti-estradiol antibody on 0.4 M KCI sucrose gradientsedimentation of MCF-7 cytosol-receptor-estradiol complex. MCF-7 cytosol (5mg/ml) was incubated with 0.4 nM [3H]estradiol (final concentration) for 4 hr at4°in the presence of 1:10,000 (final dilution) anti-estradiol antibody (•),anti-estradiol antibody plus 100-fold excess DES (•),or anti-estradiol antibody whichhad been preadsorbed with estradiol (A). [14C]BSA (arrow) was used as a 4.6S

internal marker; fraction 20 was the top of the gradient.

plexes, we found that, while there was competition betweenthe 4S estrogen receptor and 7S anti-estradiol antibody for thelimited amount of [3H]estradiol, there was no evidence of a

double complex consisting of antibody-estradiol-estrogen re

ceptor (Chart 2).In order to determine if the fluorescent staining observed

might represent a class(es) of estradiol-binding sites distinctfrom the estrogen receptor, we performed a whole-cell bindingassay using a wide range of [3H]estradiol concentrations (Chart

3A). The Scatchard analysis (Chart 3B) reveals the presenceof at least 2 classes of binding sites, with one class exhibitingthe high-affinity properties of the estrogen receptor and theother a much lower affinity. The lower-affinity sites are detectedmainly at 10 nw or higher concentrations of [3H]estradiol. It

should be noted that this is the lower limit for immunofluorescence detection of estradiol in MCF-7 cells. In addition, the

saturation analysis (Chart 3A) reveals that the nonspecificbinding at the estradiol concentrations at which maximumfluorescence staining was observed represents nearly one-halfof the total [3H]estradiol binding by the cell. These lower-affinity

sites in MCF-7 cells in some ways resemble the type II binding

sites described in rat uterine cells (4, 7), which are discussedfurther below.

DISCUSSION

Because of the great usefulness of estrogen receptor determinations in selecting therapy for breast cancer patients, therehas been considerable interest in the immunocytochemicallocalization of estradiol, assuming that the estradiol detectedis bound to the estrogen receptor. However, rigorous criteriamust be applied in control experiments to demonstrate boththat the immune reaction is specific for estradiol and that theestradiol detected is indeed receptor bound.

As we reported previously (30), it is absolutely essential toperform the immunocytochemical assays using anti-estradiol

antiserum which has been adsorbed with BSA and tested forcompleteness of adsorption. The use of unadsorbed antiserumcan result in false-positive results due to the presence of BSA

(in tissue culture cells) or human serum albumin which ispresent in tissue sections in interstitial spaces, on cell membranes, and/or intracellularly, and which contributes to the

NOVEMBER 1981 4647

Research. on November 12, 2020. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

W. D. Mercer et al.

40 60[3H] Estradiol Concentration (nM)

80

Chart 3. A. binding of [3H]estradiol by MCF-7 cells. MCF-7 cells (2X10'cells/ml) were incubated in KRH glucose butler with 0.1 to 93 nw [3H]estradiolfor 30 min at 37°.Parallel incubations using 100-fold excess DES were performedat each concentration of [3H]estradiol to estimate nonspecific binding. The cell

pellets were washed 3 times, extracted with ethanol, and counted in a liquidscintillation counter. The competible binding (•)was calculated as the difference

173 pmol/mg DNAKd 56 x IO-'°M

1.0 2.0[3H] Estradiol Bound (pmol/ml)

between total binding (O) and nonspecific binding in the presence of DES (A). B,Scatchard analysis (41 ) of [3H)estradiol binding by MCF-7 cells. The values have

been corrected for nonspecific binding. The number (N) of binding sites anddissociation constant (Kd) for the high-affinity component were determined froma linear extrapolation of the straight portion of the curve.

immunocytochemical staining observed. Most reports fromother laboratories with one possible exception (21) neglect tomention whether or not the anti-estradiol antibody was ad

sorbed with BSA. This adsorption is especially important in thelight of our unpublished observations that direct staining ofhuman breast cancers with fluorescein-labeled antiserum to

human serum albumin appears to reveal cellular heterogeneitysimilar to that reported with existing immunocytochemical assays used to detect estradiol.

To prove specificity of the antiserum, normal serum of thesame species is sometimes substituted to detect any nonspecific binding of immunoglobulins in the primary or secondaryantisera. This does not fully demonstrate immune specificity,however, since the normal serum source animal was not exposed to potential antigens such as BSA or impurities in theimmunogen which might give rise to antibodies cross-reacting

with cellular or plasma antigens. The only acceptable control,therefore, is to adsorb the test antiserum itself with the antigenbeing tested for, in this case, estradiol (or estradiol-6-carbox-

ymethyloxime). A marked reduction or total elimination of fluorescence staining in specimens as we find in this studyindicates specificity for estradiol.

Even though some of the immunocytochemical studies reported previously included controls which did demonstrateimmune specificity, there were still a number of points ofuncertainty concerning the validity of these methods for detecting estrogen receptor. These included the availability ofreceptor-bound estradiol to antibody, the concentration of estradiol necessary to visualize the response, and the ability of

other known receptor ligands to compete for the visualizedestradiol binding. We therefore initiated the present investigation in order to (a) seek evidence of antibody-estradiol-receptor

immune complexes, if any, and (6) determine if the intracellularlocalization of visualized estradiol following experimental treatment was consistent with the detection of estrogen receptor.These experiments were performed using the MCF-7 breastcancer cell line, a well-described model system for in vitro

studies of steroid hormone action (6, 15,17, 24), and includedthe use of appropriate assay controls to demonstrate immunespecificity. The results confirm that estradiol-specific fluores

cence staining can be observed in breast cancer cells, but thatthe immunocytochemical assays are not detecting estradiolbound to the estrogen receptor.

First, it is clear that, if antibodies are to recognize receptor-

associated estradiol, either the antibodies must be able to bindestradiol even in the receptor binding site, or else the receptormust first release the bound hormone. Despite several attemptsusing various incubation times and temperatures, differentbuffers, and high- and low-salt-containing gradients, we consistently failed to detect any evidence of antibody-estradiol-

receptor immune complexes in sucrose gradients. This is essentially the same observation reported previously by Castañedaand Liao (3), who also did not find any evidence of theseimmune complexes using sucrose gradient analysis. Theseexperiments were performed on cytoplasmic receptors, leavingopen a possibility that nuclear receptors would behave differently; this seems unlikely, however, considering their immu-

nological (14) and kinetic similarity.

4648 CANCER RESEARCH VOL. 41

Research. on November 12, 2020. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Estradiol Immunofluorescence in Breast Cancer

In addition, the results of several studies which were basedon radioimmunoassay methods to detect competition betweencytosol estrogen receptor and antibody for binding of estradicialso suggest that a complex of antibody-estradiol-receptor wasnot formed (3, 10-12). These experiments involved the use ofinsolubilized antibodies to estradiol, which were incubated witha mixture of [3H]estradiol and cytosol from breast tumors or

other estrogen target tissues. Under these conditions, althoughthe antibody would bind free [3H]estradiol and [3H]estradiol

bound to nonspecific sites, it could not successfully competefor [3H]estradiol bound to the estrogen receptor, which presum

ably reflects a lack of immune complex formation. Similarcompetition experiments have been performed in our laboratory, and we also failed to observe any evidence of reproduciblecompetition suggestive of immune complex formation (data notshown).

An alternative explanation for antibody recognition of receptor-associated estradiol is that the receptor may release theestradiol or at least loosen its hold during fixation (19). If thiswere true, however, it seems unlikely that the freed hormonewould remain precisely at the original site or even in the originalcell to await the antibody. Freed hormone would presumablywash away during washing steps if not fixed to cellular structures. It therefore seems most likely that the intracellular estradiol detected by these methods is in fact bound to structuresother than the estrogen receptor; we will discuss this possibilityfurther below.

At least 3 types of estrogen-binding components in cells can

be distinguished. The estrogen receptor itself, or type I binder,has the highest affinity, with a dissociation constant (Kd) forestradiol on the order of 1CT10 M. This means that half of the

receptors carry hormone at that estradiol concentration, andthat nearly all are filled by the time the estradiol concentrationreaches 1 nM. Above 1 nM, the receptors are saturated, so thatno more estradiol can bind to them no matter how much isadded.

Other types of sites exist, however, which have lower affinitybut higher capacity for estradiol binding. Such sites have beendescribed in the rat uterus by Clark ef al. (4, 7), who termedthem "type II" estrogen-binding sites. Panko e? al. (36) have

found large amounts of type II binding in many human breastcancers. Below 1 nw estradiol, few of these sites are occupied,but at higher concentrations, type II binding becomes substantial with most binding sites occupied following exposure to 100nM estradiol.

There is also another order of estrogen binding within cells,comprised of soluble molecules like albumin and probably alsocertain molecules associated with cell membranes and othercellular structures. Although no one of these has a high affinityfor estradiol, their total binding capacity is very great, andbinding to these sites at very high estradiol concentrations maybecome substantial. These sites we refer to as type III (ornonspecific) estrogen-binding sites.

Finally, it is conceivable that essentially free estradiol couldbe detected within cells, Whether the observed ¡ntracellularfluorescence includes free estradiol was not directly established in our experiments, but we have observed previouslythat MDA-231 breast cancer cells and many tumor cells ob

tained from patients with breast cancer failed to exhibit detectable estradiol-specific fluorescence following treatment of thecells with up to 1 /IM estradiol (30). Therefore, the free estradiol

KTIO'3 IO'

[Estradici]

KT3

Chart 4. Schematic of estradiol binding to different classes of binding sites inbreast cancer cells. Receptor, type II, and type III sites are occupied in turn atprogressively higher estradiol concentrations; the eventual saturation of receptorand type II sites is shown by the dotted lines.

which would be expected to have diffused into these cells wasnot detected, presumably having been removed from the frozensections or fixed cells during washing procedures. When fluorescent staining is present, as with MCF-7, it therefore prob

ably represents estradiol which is bound nonspecifically tocellular structures or specifically to classes of steroid-binding

sites.The binding of estradiol to estrogen receptor, and subse

quently to type II and type III sites as the concentration increases, is diagrammed in Chart 4. It is obvious that, if too higha concentration of estradiol is used to incubate specimens forimmunocytochemical assays, the probability increases thatthose assays will detect estradiol-binding sites other than the

estrogen receptor. In fact, we have found that, with 1 nMestradiol, which should be sufficient to nearly saturate theestrogen receptors, we cannot detect estradiol-specific immu-

nofluorescence at all. It seems likely, therefore, that mostimmunocytochemical methods, some of which use as high as1 mw estradiol, are detecting primarily type II or even type IIInonspecific binding rather than the receptor itself.

In the rat uterus, the only tissue in which type II binding hasbeen well studied, type II is present in cell nuclei and increasesgreatly after estrogen treatment (8, 15). Unlike the estrogenreceptor, however, the type II sites present in the cytoplasm(which may be totally distinct from those in the nuclei) are notdepleted by estrogen treatment. Nor would the location of typeIII sites be expected to be affected by estrogens. In this report,we attempted to make use of this distinction to discover whichbinding was detected by the immunofluorescence procedure.We treated MCF-7 human breast cancer cells with 5 nM DES

and proved by radioligand binding assay that all of the trueestrogen receptor had in fact been translocated to the nuclei.Nevertheless, the cytoplasmic immunofluorescence was notreduced, which is consistent with the detection of estrogenbound to some sort of type II or even type III component ratherthan to estrogen receptor. Kurzon and Sternberger (21) recently reported similar results with their immunocytochemicalassay in rat uterus.

The results of our saturation analyses of MCF-7 cells with[3H]estradiol, which were performed to detect other classes of

binding sites, show that type II sites may be present in MCF-7

cells. Whether the type II sites that we detected represent baselevels of this class of binding sites or include specific inductionin nuclei following estradiol treatment was not established inthis study. Because saturating conditions were not demon-

NOVEMBER 1981 4649

Research. on November 12, 2020. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

W. D. Mercer et al.

strated for this class of binding site, we were not able toestimate the actual number of binding sites or the dissociationconstant. However, from the amount of total binding observedfollowing incubation with 10 nM [3H]estradiol, and by subtract

ing the amount of type I binding (estrogen receptor), we estimated that nearly half of the total estradiol binding was tolower-affinity but competible estradiol-binding sites. Whether

this level of binding can explain the immunofluorescence results for MCF-7 is not yet known.

One necessary criterion for immunofluorescent detection ofreceptor-associated estradiol is elimination of staining with an

excess of DES or antiestrogens, which would fill the receptorsites but cannot be detected by the antibody. Other investigators have reported such apparent competition, and this hasbeen considered evidence for the receptor specificity of theirprocedures. However, the concentration of DES reportedlyused has varied from 1 ¡IMto 1 rrtM, which is in vast excess ofthe concentration necessary to saturate estrogen receptors.Since type II estradiol binding is also inhibited by DES (4), andsufficiently high concentrations of DES might even reducesome estradiol binding to type III sites, this "competition" does

not invalidate our conclusion that immunocytochemical assaysare probably detecting types II or III sites rather than theestrogen receptor. In addition, we have observed that, at DESconcentrations as high as 1 UM, there is a marked decrease inthe binding of antibody to estradiol (Chart 1). If high tissueconcentrations of DES were present, the appearance of competitive inhibition of estradiol binding to cellular macromole-cules might therefore represent, at least in part, inhibition ofantibody activity. In addition, such high concentrations of DESand other steroid hormones approach the limit of solubility inaqueous solutions, and one cannot rule out a decrease inestradiol solubility in the presence of high DES concentrations.Inhibition by appropriate competitors therefore argues for estrogen receptor binding only if the estradiol concentration itselfis appropriate (around 1 nM) and the competition excess is nottoo great (preferably 10 nM but no more than 100 nw for DES).

We have concluded from our study that, although immunocytochemical assays can be used to detect estradiol in breastcancer cells, the hormone specific fluorescence staining observed is not due to direct binding of antibody-estradiol-recep-tor, but rather must be due to the detection of estradiol boundto lower-affinity binding sites.

REFERENCES

1. Abraham. G. E. Solid-phase radioimmunoassay of estradiol-17/8. J. Clin.Endocrino!. Metab.. 29: 866-870. 1969.

2. 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.

3. Castañeda, E., and Liao, S. The use of anti-steroid antibodies in the characterization of steroid receptors. J. Biol. Chem., 250 883-888, 1975.

4. Clark, J. H., Hardin, J. W., Upchurch, S., and Eriksson, H. Heterogeneity ofestrogen binding sites in the cytosol of the rat uterus. J. Biol. Chem., 253.7630-7634. 1978.

5. Edwards, D. P., Martin, P. M., Horwitz, K. B., Chamness. G. C., and McGuire.W. L. Subcellular compartmentalization of estrogen receptors in humanbreast cancer cells. Exp. Cell Res.. 727. 197-213, 1980.

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

7. Eriksson, H., Upchurch, S.. Hardin, J. W., Peck, E. J., Jr., and Clark, J. H.Heterogeneity of estrogen receptors in the cytosol and nuclear fractions ofthe rat uterus. Biochem. Biophys. Res. Commun., 87: 1-7, 1978.

8. Exley, D., Johnson. M. W.. and Dean. P. D. G. Antisera highly specific for17-/9-estradiol. Steroids, 18: 605-620. 1971.

9. Fey, H. Eriochrome black, a means of reduction of nonspecificity in immunofluorescence. Pathol. Microbiol., 38. 271-277, 1972.

10. Fishman, J., and Fishman, J. H. Competitive binding assay for estradiolreceptor using immobilized antibody. J. Clin. Endocrinol. Metab., 39. 603-606, 1974.

11. Fishman, J., Fishman, J. H., Nisselbaum. J. S.. Menendez-Botet, C..Schwartz. M. K., Martucci, C., and Hellman, L. Measurement of the estradiolreceptor in human breast tissue by the immobilized antibody method. J. Clin.Endocrinol. Metab.. 40: 724-727. 1975.

12. Floridi, A.. Citro, G., Sega, E.. and Natali. P. G. Quantitation of estradiolreceptors employing a sensitive radioimmunoassay. Immunol. Commun., 4:603-615, 1975.

13. Ghosh. L.. Ghosh, B. C., and Gupta, T. K. D. Immunocytological localizationof estrogen in human mammary carcinoma cells by horseradish-anti-horse-radish peroxidase complex. J. Surg. Oncol., 10: 221-224, 1978.

14. Greene, G. L., Closs, L. E.. Fleming, H.. DeSombre, E. R., and Jensen, E. V.Antibodies to estrogen receptor: immunochemical similarity of estrophilinfrom various mammalian species. Proc. Nati. Acad. Sei. U. S. A., 74: 3681-3685, 1977.

15. Horwitz. K. B., Costlow, M. E., and McGuire, W. L. MCF-7: a human breastcancer cell line with estrogen, androgen, progesterone and glucocorticoidreceptors. Steroids. 26. 785-795, 1975.

16. Horwitz, K. B., McGuire, W. L., Pearson, O. H.. and Segaloff. A. Predictingresponse to endocrine therapy in human breast cancer: a hypothesis.Science (Wash. D. C.), 789: 726-727, 1975.

17. Horwitz. K. B., Zava, D. T., Thilagar, A. K., Jensen. E. M.. and McGuire, W.L. Steroid receptor analyses of nine human breast cancer cell lines. CancerRes., 38. 2434-2437. 1978.

18. Jensen. E. V., DeSombre. E. R.. and Jungblut, P. W Estrogen receptors inhormone-responsive tissues and tumors. In: R. W. Wissler. T. L. Dao, and S.Wood. Jr. (eds.). Endogenous Factors Influencing Host-Tumor Balance, pp.15-30. Chicago: University of Chicago Press, 1967.

19. Jungblut, P. W. In: Discussion following: Nenci. I.. Piffanelli, A.. Beccati, M.D., and Lanza, G. In vivo and in vitro immunofluorescence approach to thephysiopathology of estradiol kinetics in target cells. J. Steroid Biochem., 7:883-890, 1976.

20. Kopp, F., Martin, P. M., Rolland, P. H., and Bertrand, M-F. A preliminaryreport on the use of immunoperoxidases to study binding of estrogens in ratuteri. J. Steroid Biochem., 11: 1081-1090, 1979.

21. Kurzon, R. M., and Sternberger, L. A. Estrogen receptor immunocytochem-istry. J. Histochem. Cytochem., 26. 803-808, 1978.

22. Lee, S. H. Cytochemical study of estrogen receptor in human mammarycancer. Am. J. Clin. Pathol.. 70. 197-203, 1978.

23. Lee, S. H. Cancer cell estrogen receptor of human mammary carcinoma.Cancer (Phila.), 44: 1-12, 1979.

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

25. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. Proteinmeasurement with the Folin phenol reagent. J. Biol. Chem.. 793. 265-275.1951.

26. McGuire, W. L.. Carbone, P. P., Sears, M. E.. and Escher, G. C. Estrogenreceptors in human breast cancer: an overview. In: W. L. McGuire, P. P.Carbone, and E. P. Vollmer (eds.). Estrogen Receptors in Human BreastCancer, pp. 1-8. New York: Raven Press. 1975.

27. Mercer. W. D.. Carlson. C. A., Wahl. T. M., and Teague. P. O. Identificationof estrogen receptors in mammary cancer cells by immunofluorescence.Am. J. Clin. Pathol.. 70: 330, 1978.

28. Mercer, W., Wahl, T., Carlson, C., and Teague, P. Identification of estrogenreceptors in breast cancer cells by immunological techniques. Fed. Proc..37: 1312, 1978.

29. Mercer, W., Wahl. T.. Carlson, C., and Teague. P. Identification of estrogenand progesterone receptors in breast cancer cells by immunological techniques. Fed. Proc., 38. 913, 1979.

30. Mercer, W. D.. Wahl, T. M., Carlson, C. A., Wahl. D. A., Lippman, M. E.,Lezotte, D., and Teague. P. O. The use of immunocytochemical techniquesfor the detection of steroid hormones in breast cancer cells. Cancer (Phila.),46. 2859-2868, 1980.

31. Nenci, I. Receptor and centriole pathways of steroid action in normal andneoplastic cells. Cancer Res., 38. 4204-4211, 1978.

32. Nenci, I. Estrogen receptor immunocytochemistry. J. Histochem. Cytochem..27: 1053-1055, 1979.

33. Nenci, I., Beccati, M. D., Piffanelli, A., and Lanza, G. Detection and dynamiclocalization of estradiol-receptor complexes in intact target cells by immunofluorescence technique. J. Steroid Biochem., 7. 505-510, 1976.

34. Nenci, I., Piffanelli, A., Beccati, M. D.. and Lanza, G. In vivo and in vitroimmunofluorescent approach to the physiopathology of estradiol kinetics intarget cells. J. Steroid Biochem.. 7. 883-890, 1976.

35. Osborne, C. K., and McGuire, W. L. Current use of steroid hormone receptorassays in the treatment of breast cancer. Surg. Clin. N. Am.. 58: 777-778,1978.

36. Panko, W. B.. Watson, C. S., and Clark, J. H. The presence of a second,specific estrogen binding site in human breast cancer. J. Steroid Biochem.,

4650 CANCER RESEARCH VOL. 41

Research. on November 12, 2020. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

Estradici Immunofluorescence ¡nBreast Cancer

in press, 1981. carcinoma and hyperplasia by fluorescence microscopy. Res. Commun.37. Pertschuk, L. P. Detection of estrogen binding in human mammary carci- Chem. Pathol. Pharmacol.. 22. 427-430, 1978.

noma by immunofluorescence: a new technic utilizing the binding hormone 41. Scatchard, G. The attractions of proteins for small molecules and ions. Ann.in a polymerized state. Res. Commun. Chem. Pathol. Pharmacol., 14: 771- N. Y. Acad. Sci., 5). 660-672, 1949.774, 1976. 42. Schenk, E. A., and Churukian, C. J. Immunofluorescence counterstains. J.

38. Pertschuk, L. P., Gaetjens. E., Carter, A. C., Brigati, D. J., Kim, D. S.. and Histochem. Cytochem., 22. 962-966, 1974.Tobin, E. H. Histochemistry of steroid receptors in breast cancer: an over- 43. Ternynck, T., and Avrameas, S. Polymerization and immobilization of pro-view. Ann. Clin. Lab. Sci.. 9. 219-224, 1979. teins using ethylchloroformate and glutaraldehyde. Scand. J. Immunol.

39. Pertschuk, L. P.. Tobin, E. H., Brigati, D. J., Kim, D. S., Bloom, N. D., Suppl., 3. 29-35, 1976.Gaetjens, E.. Berman, P. J., Carter, A. C., and Degenshein, G. A. Immuno- 44. Zava, D. T., Harrington, N. Y., and McGuire, W. L. Nuclear estradici receptorfluorescent detection of estrogen receptors in breast cancer. Cancer (Phila.). in the adult rat uterus: a new exchange assay. Biochemistry, /5. 4292-41: 907-911, 1978. 4297, 1976.

40. Pertschuk, L. P., Zava, D. T., Gaetjens, E., Macchia, R. J., Brigati, D. J., and 45. Zava, D. T., and McGuire. W. L. Estrogen receptor. Unoccupied sites inKim, D. S. Detection of androgen and estrogen receptors in human prostatic nuclei of a breast tumor cell line. J. Biol. Chem., 252. 3703-3708, 1977.

Fig. 1. Immunofluorescence photomicrograph of a frozen section of MCF-7 cells. The cells were incubated at 4°for 4 hr in the presence of 10 nM estradiol asdescribed in Table 3. The antiserum for this experiment and all of the following photomicrographs was antiestradiol antiserum which had been preadsorbed with BSAx 325.

Fig. 2. Immunofluorescence photomicrograph of a frozen section of MCF-7 cells incubated with estradiol at 37°. The cells were first incubated with 10 nMestradiol at 4°for 4 hr followed by warming the cells in the presence of hormone to 37°for 30 min as described in Table 3. x 325.

Fig. 3. Immunofluorescence photomicrograph of a frozen section of MCF-7 cells incubated with DES and estradiol. The cells were first incubated with 5 nw DESat 37°for 1 hr, washed and cooled to 4°,and then incubated with 10 nM estradiol at 4°for 4 hr as described in Table 4. x 325.

Fig. 4. Immunofluorescence photomicrograph of a frozen section of MCF-7 cells incubated only with DES. The cells were incubated with 5 nw DES at 37°for 1hr, washed and cooled to 4°.and then incubated with KRH glucose buffer at 4°for 4 hr as described in Table 4. x 325.

NOVEMBER 1981 4651

Research. on November 12, 2020. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

W. D. Mercer et al.

4652 CANCER RESEARCH VOL. 41

Research. on November 12, 2020. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from

1981;41:4644-4652. Cancer Res   Wayne D. Mercer, Dean P. Edwards, Gary C. Chamness, et al.   Cancer Cells to Detect Estrogen ReceptorsFailure of Estradiol Immunofluorescence in MCF-7 Breast

  Updated version

  http://cancerres.aacrjournals.org/content/41/11_Part_1/4644

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/41/11_Part_1/4644To request permission to re-use all or part of this article, use this link

Research. on November 12, 2020. © 1981 American Association for Cancercancerres.aacrjournals.org Downloaded from