[CANCER RESEARCH 54. 1587-1595. March 15. 1994]
`
`Altered Expression of Estrogen-regulated Genes in a Tamoxifen-resistant and
`
`ICI 164,384 and ICI 182,780 Sensitive Human Breast Cancer Cell Line,
`
`MCF-7/TAMR-1‘
`
`Anne E. Lykkesfeldt,’ Mogens W. Madsen, and Per Briand
`Department of Tumor Endocrinology, Division for Cancer Biology, The Danish Cancer Society, Strandboulevarden 49, DK-2100 Copenhagen 0, Denmark
`
`ABSTRACT
`
`A stable, tamoxifen-resistant subline, MCF-7/TAM“-1, of the human
`breast cancer cell line MCF-7 has been established in tissue culture after
`long-tenn treatment with 10" M tamoxifen. The MCF-7/TAM"-1 cell line
`grows equally well in the presence and absence of tamoxifen, whereas the
`steroidal antlestrogens ICI 164,384 and ICI 182,780 exert profound in-
`hibitory activity on cell proliferation, although higher concentrations are
`required to inhibit these cells compared to the parent cells. The MCF-7/
`TAM“-1 cells grown in tissue culture deviate from parent characteristics
`by the complete lack of expression of progesterone receptors even when
`grown with estradiol, by an altered tamoxifen regulation of M, 52,000
`cathepsin D synthesis and secretion, and by lack of tamoxifen stimulation
`of an estradiol down-regulated M, 42,000 protein with presumed growth
`inhibitory function. MCF-7/TAM"-1 cells are estrogen receptor positive.
`The estrogen receptors have wild-type characteristics with respect to (a)
`binding of estradiol, tamoxifen, and ICI 164,384; (b) estrogen and anties-
`trogen regulation of the estradiol-regulated proteins pS2, M, 61,000 a.-
`antitrypsin-like protein, M, 66,000 at,-antichymotrypsin-like protein, and
`corresponding mRNAs; and (c) estrogen and antiestmgen regulation of a
`transiently transfected estrogen responsive reporter gene. We suggest that
`the lack of tamoxifen up-regulation of the M, 42,000 protein synthesis in
`MCF-7/TAM"-l cells may at least partly explain the resistance to tamox-
`ifen treatment. The sensitivity to the growth inhibitory activity of [Cl
`164,384 and lCl 182,780 may be ascribed to the maintenance of the pure
`antagonistic effect of these steroidal antiestrogens on MCF-7/TAM"-1
`cells. Our results indicate that treatment with pure antiestrogens may be
`effective when patients become refractory to tamoxifen therapy.
`
`INTRODUCTION
`
`Treatment with the antiestrogen tamoxifen is first line endocrine
`therapy for most breast cancer patients with estrogen receptor-positive
`primary tumor, but almost all patients with advanced disease will
`eventually develop tamoxifen resistance. Fortunately, many of these
`patients respond to second and third line of hormone therapy (1).
`However, the mechanisms for antiestrogen resistance are not yet clari-
`fied; therefore, it is extremely important for a rational treatment strat-
`egy to be devised to obtain detailed knowledge of antiestrogen resis-
`tance. Recent results from studies on clinical material and with
`
`antiestrogen-resistant human breast cancer cell lines in tissue culture
`or in athymic nude mice have revealed that several different mecha-
`nisms may be underlying the development of resistance (2-4).
`Presence of estrogen receptors is obligatory for response to anties-
`trogen treatment, and loss of expression of ER’ is, therefore, the most
`
`obvious mechanism for development of antiestrogen resistance. The
`incidence of progression to ER negativity has been evaluated by a
`comparison of ER content in primary breast tumors and metastases
`from the same patient. Patients with uniformly ER-positive tumors
`usually have ER-positive metastases (93%), whereas ER heteroge-
`neous tumors in most cases give rise to ER-negative metastases (5). In
`general, tumors classified as ER positive give rise to ER-positive
`metastases in about 70% of the cases (5, 6), and the observation of a
`response rate to endocrine therapy of about 60-70% in patients with
`advanced disease with ER-positive primary tumors could, at least for
`some of the patients, be explained by lack of ERs in the metastatic
`lesions. Relapse following successful antiestrogen therapy may also
`be due to progression of ER-negative tumor cells during treatment;
`however, clinical data show that many tumors that progress after first
`responding to tamoxifen therapy express ERs (7).
`Expression of ERs with abnormal function could be a mechanism
`that leads to antiestrogen resistance in tumor cells classified as ER
`positive, and at present, papers have reported on the presence of ER
`proteins deviating from the wild-type protein (8), expression of vari-
`ant mRNAs generated by alternative RNA splicing, and mutations in
`the ER gene (9-14). In a few cases, the function of such variant ERs
`has been studied, and two interesting abnormal functions have been
`disclosed: (a) one which is considered dominant positive by being
`transcriptionally active without binding hormone (15), and (b) one
`which is dominant negative by inhibiting the action of the normal
`receptor (12, 16). A review of papers describing the ER-positive
`antiestrogen-resistant cell lines established in tissue culture after long
`tenn treatment with antiestrogen reveals that most of the ER functions
`in these cells appear normal; however, each resistant cell line displays
`one or more features of abnonnal ER function (17-22).
`Antiestrogen resistance may not necessarily be associated with loss
`of ERs or abnormal function of ER. Tumors, which during the pro-
`gression to hormone insensitivity up-regulate the synthesis of growth
`stimulatory factors and/or down-regulate the synthesis of growth in-
`hibitory factors, may not, in spite of normal binding of antiestrogen
`and normal function of the ER machinery, exhibit a decrease in growth
`potential (23). ER-negative tumor cells or stromal cells may also
`support growth of ER-positive cells via paracrine mechanisms and
`thereby sustain growth of ER-positive tumor cells in spite of tamox-
`ifen treatment. Finally, the observed tamoxifen resistance in trans-
`planted tumors of MCF-7 cells in athymic nude mice seems to be due
`to a phannacological change in the metabolism of tamoxifen giving
`rise to compounds with estrogenic activity (24). Similar indications of
`changed pharmacology of tamoxifen have also been found in human
`breast tumor biopsies (25).
`In this paper, we describe a tamoxifen-resistant cell line established
`in our laboratory from MCF-7 cells via two series of long term
`treatment with tamoxifen in tissue culture. The antiestrogen resistant
`cell line MCF-7/TAM“-1, originally called the AL-1 cell line (22, 26),
`is estrogen receptor-positive. Most of the ER-mediated functions are
`regulated normally, but with cells grown in tissue culture, we have
`
`Received 7/6/93; accepted 1/14/94.
`The costs of publication of this article were defrayed in pan by the payment of page
`charges. This article must therefore be hereby marked advertisement in accordance with
`18 U.S.C. Section 1734 solely to indicate this fact.
`‘ This work was supported by the Danish Cancer Society, the Astrid Thaysen Foun-
`dation, and the John and Birthe Meyer Foundation.
`2 To whom requests for reprints should be addressed. at Department of Tumor Endo-
`crinology, Division for Cancer Biology. The Danish Cancer Society, Strandboulevarden
`49, DK-2100 Copenhagen 6, Denmark.
`3 The abbreviations used are: ER, estrogen receptor; PgR, progesterone receptor; DME,
`Dulbecco’s minimal essential medium; F12, Ham's nutrient mixture F-12; FCS, fetal calf
`serum; PBS. phosphate-buffered saline; SDS. sodium dodecyl sulfate; SSPE, sodium
`chloride-sodium phosphate-EDTA buffer; SSC, standard sodium citrate; CAT, chloram-
`to untreated controls; GAPDH, glyceraldehyde-3-phosphate-dehydrogenase; vit-tk-CAT,
`phenicol acetyl transferase; lC5o, the concentration that inhibited growth by 50% relative
`vitellogenin-thymidin kinase-chloramphenicol acetyl transferase.
`1587
`
`Downloaded from cancerres.aacrjourna|s.org on June 17, 2016. © 1994 American Association for Cancer Research.
`
`MYLAN PHARMS. INC. EXHIBIT 1036 PAGE 1
`
`

`
`TAMOXIFEN~RESlSTANl‘ HUMAN BREAST CANCER CELL LINE
`
`MATERIALS AND METHODS
`
`observed a complete lack of PgR expression and estrogen inducibility
`and a lack of tamoxifen regulation of secreted proteins with growth
`regulatory function. This stable, tamoxifen-resistant cell line is sen-
`sitive to the steroidal antiestrogens ICI 164,384 and ICI 182,780,
`indicating that these new antiestrogens may be valuable compounds in
`the treatment of breast cancer patients with tumor progression while
`on tamoxifen therapy.
`
`antihormones were added, and medium was renewed every second or third day.
`At day 6, the wells were washed with PBS and labeled for 6 h with [355]-
`methionine (NEG-0O9H; NEN Research Products; specific activity, ~1000
`Ci/mmol) or a mixture of [35S]methionine and [35S]cysteine (NEG-072) in
`semm-free DME-medium with 1:10 the normal concentration of the amino
`acids used for labeling. Treatment of labeled, conditioned medium and SDS-
`polyacrylamide gel electrophoresis analysis on 15%, 15-20% gradient, and
`20% polyacrylamide gels under denaturating conditions were performed as
`described previously (32). Kodak XAR-5 films were used for autoradiography,
`and scanning was performed by a LKB Ultroscan laser densitometer.
`Northern Analysis. Cells were seeded in T-150 flasks, and after 2 days, the
`Cells and Growth Experiments. The MCF-7 cell line was originally ob-
`medium was changed to experimental medium containing either 10'“ M estra-
`tained from The Breast Cancer Task Force Cell Culture Bank, Mason Research
`diol, 10" M tamoxifen, or 10" M ICI 164,384. The medium was renewed on
`Institute (Worcester, MA). The cell line has been gradually adapted to grow in
`days 3 and 5, and on day 6 the cells were harvested by trypsinization, pelleted,
`medium with low serum concentration (27), and at the initiation of these
`and frozen in liquid nitrogen. RNA preparation and Northern analysis were
`experiments, the cells were propagated in DME/F12 medium (121) supple-
`preformed as described previously (33). Briefly, the cell pellets were homog-
`mented with 0.5% FCS, 2 mM glutamine, and 6 ng/ml insulin (Novo-Nordisk,
`enized in 4 M guanidinium thiocyanate, and RNA was isolated by centrifuga-
`Copenhagen, Denmark). The MCF-7/TAM“-1 cell
`line, previously called
`tion over a CsCl cushion. To ensure the quality of RNA, 10 pg RNA from each
`AL-1, was established by two series of treatments with 10" M tamoxifen as
`preparation was denatured by dimethylsulfoxide/glyoxal, electrophoresed on a
`described earlier (22). The MCF-7/TAM“-1 cells were propagated continu-
`1.0% agarose gel in 10 mM phosphate buffer (pH 7.0), and stained with acridine
`ously in the presence of 10" M tamoxifen from passage 372 (onset of the
`orange. Ten-pg RNA samples from both cell lines were electrophoresed as
`second tamoxifen treatment), which is considered as passage 1 of the resistant
`described above and transferred by Northern blotting to the same nylon mem-
`cell line. After the discovery of the weak estrogenic activity of phenol red (28),
`both MCF-7 and the MCF-7/TAM“-1 cells were propagated in phenol red-free
`brane (NY 13N; Schleicher and Schuell, Dassel, Germany). RNA ladder,
`0.24—9.5 kilobases, (BRL, New York) was used as molecular size markers.
`DME/F12 medium supplemented with 1% FCS, 2 mM glutamine, and 6 ng/ml
`insulin. The shift to growth medium without phenol red and with 1% FCS had
`Purified plasmid-derived complementary probes were labeled with [a-”P]-
`neither influence on cell proliferation rate nor on the pattern of secreted
`dATP (Amersham, Buckinghamshire, England) using the Multiprime DNA
`proteins. Both MCF-7 and MCF-7/TAM“-1 cells were passaged every week by
`labeling kit (Amersham) to a specific activity of 1-2 X 109 dpm/pg DNA, and
`trypsinization and were seeded with 5 X 103 cells/cm’. Medium was changed
`Northern blots were hybridized as described before (33). Oligonucleotide
`every second or third day. The cell lines were regularly tested by the Hoechst
`probes were end-labeled with [gamma-32P]ATP (Amersham) to a specific
`staining method (29) and found to be free of Mycoplasma.
`activity of 0.5-2 X 107 dpm/pmol using T4 polynucleotide kinase (BRL), and
`Cultures used for growth experiments were seeded in plastic T25 flasks or
`prehybridization of Northern blots was carried out at 55°C for at least 1 h in
`2 cm’ multiwell dishes (Nunc, Denmark) 2 days before the onset of the
`6X SSPE (1X SSPE, 150 mM NaCl-10 mM NaI~l2P04-1 mM EDTA), 0.1%
`experiments. Estradiol (Collaborative Research), tamoxifen, ICI 164,384, and
`Ficoll, 0.1% polyvinylpyrrolidone, 0.1% bovine serum albumin, and 0.5%
`[Cl 182,780 (lCl Pharmaceuticals) were dissolved in 96% ethanol in stock
`SDS. For hybridization, the nylon membranes were then incubated 12 h at
`solutions 1000-fold higher than the concentration used and added to the ex-
`55°C in 6X SSPE and 0.1% SDS containing 4 X 105 dpm/ml of “P-labeled
`perimental media at the time of change of medium, which was every second or
`oligonucleotide. Unbound probe was removed from the membranes by wash-
`third day during the experiment. Triplicate flasks or quadruplicate wells were
`ing 1.5 h at 55°C with 2X SSC (IX SSC, 0.15 M NaCl-15 mM sodium citrate,
`used for determination of cell numbers, either after trypsinization and counting
`pH 7.0,) and 0.1% SDS followed by a 10-min wash at 55°C with 6X SSC, 0.1%
`in a Biirker-Tiirch chamber or after cell lysis and counting of cell nuclei in an
`SDS, and a 3-min wash at 70°C in 6X SSPE and 0.1% SDS. The blots were
`automatic cell counter (Analys Instrument ABNDA; Ref. 26).
`exposed to Kodak XAR-5 or AGFA Structurix-D7pFW films at —80°C. Scan-
`Tirmorigeniclty. Near confluent cultures were trypsinized, washed once
`ning of autoradiograms was performed by a LKB Ultroscan laser densitometer.
`with PBS, and resuspended in PBS. Approximately 107 cells were inoculated
`All Northern analyses were repeated two times on independent RNA prepa-
`s.c. in female nude mice at the site of the fourth mammary gland. Estrogen was
`rations, and the results were reproducible.
`administered as estrone in the drinking water (0.5 pg/ml estrone; Sigma
`Probes. The probes used were cathepsin D, oligonucleotide 5’-TI‘AACG-
`E9750). The animals were observed for tumor formation weekly for a mini-
`TAGGTGCFGGACITG-TCGCI‘G'I'I‘GTACI'l‘-3’ (34); pS2, EcoRl linear-
`mum of 11 weeks and up to 6 months. Tumors were excised and frozen at
`ized pS2 (35); a1-antitrypsin, 1.4-kilobase EcoRl fragment from phAT85
`—80°C until ER and PgR determination.
`(American Type Culture Collection No. 61595); arantichymotrypsin, 1.0-
`ER and PgR Determinations. Tumor tissue was homogenized in a Mikro-
`kilobase Pstl fragment from phACI‘235 (American Type Culture Collection
`Dismembrator (B Braun) with precooled (-180°C) pulverization chambers.
`no. 61601); and rat GAPDH, 1.2-kilobase Pstl fragment from pRGAPDI-I-13
`Cytosols were prepared according to the European Organization for Research
`(36).
`and Treatment of Cancer guidelines (30). The ERs were extracted from the
`Transient 'lYansfection Studies. Cationic liposome-mediated transfection
`nuclear pellet with an extraction buffer consisting of 10 mM Tris (pH 8.5)-1 mM
`with Lipofectin reagent (BRL) was performed according to the manufacturer’s
`monothioglycerol-10% glycerol-6()0 mM KCI for 1 h at 0°C; the supernatant,
`instructions. About 80% confluent cultures in 60-mm tissue culture dishes
`after centrifugation at 100,000 X g, was used for determination of nuclear
`(Nunc, Denmark) were cotransfected with 1 p.g vit-tk—CAT (37) and 10 pg
`receptors. Tissue culture cells were harvested from near confluent cultures and
`pBGal plasmid (38) using 50 pl Lipofectin reagent. pBR322 (11 p.g) was used
`homogenized in a glass-glass tightly fitting Potter-Elvehjem homogenizer;
`for mock transfections, and 10 pg pRSV~CAT (39) were used as a positive
`cytosol and nuclear extract were prepared as described previously (27). The
`control for transfection. Twenty-four h were used for transfection, followed by
`content of free ER and free PgR in cytosol preparations was detennined by
`48-72 h in experimental medium before harvest of the cells. The cells were
`ligand binding assay (dextran coated-charcoal assay) according to the recom-
`washed twice with PBS and harvested with 1.5 ml 1 mM EDTA in PBS by
`mendations of the European Organization for Research and Treatment of
`scraping with a rubber policeman. Cells were collected by centrifugation at
`Cancer Receptor Group (30). ER contents in the nuclear extracts were deter-
`10,000 X g for 5 min and resuspended in 100 p.l 0.25 M Tris buffer, pH 7.8. The
`mined with the Abbott ER-EIA monoclonal kit (Abbott Laboratories, North
`cells were lysed by freezing and thawing five times, and cell debris was
`Chicago, IL), which is insensitive to KCl concentrations up to 800 mM and
`removed by centrifugation at 10,000 X g for 15 min at 4°C. Equal amounts of
`which measures both free and hormone-bound ERs (31).
`supematants (50 pl) were used for B-galactosidase assay for 4 h (40), and cell
`Secreted Proteins. MCF-7/TAM“-1 cells were propagated for 1 week
`extracts (heated to 65°C for 10 min to remove deacylase activity) adjusted to
`without
`tamoxifen before onset of the experiments. MCF-7 and MCF-7/
`TAM“-1 cells were seeded in 2 cm’ multiwell dishes in control medium with
`equal B-galactosidase activity were analyzed for CAT activity (40), using
`[“‘C]chloramphenicol (59.5 mCi/mmol; NEC-408; NEN Research Products).
`1% FCS at a density of 1.5 X 10‘ cells/well. Two days after seeding, experi-
`mental media containing the indicated concentrations of hormones and/or
`Thin layer chromatography on silica gel 60 F254 (0.2 mm; Merck) was
`1588
`
`Downloaded from cancerres.aacrjourna|s.org on June 17, 2016. © 1994 American Association for Cancer Research.
`
`MYLAN PHARMS. INC. EXHIBIT 1036 PAGE 2
`
`

`
`TAMOXIFEN-RESISTANT HUMAN BREAST CANCER CELL IJNE
`
`performed as described in Sambrook et al. (40). Kodac XAR-5 films were used
`for the autoradiograms.
`
`RESULTS
`
`Establishment of the Tamoxifen-resistant MCF-7/TAM"-1 Cell
`Line and Growth Studies. The tamoxifen-resistant cell
`line was
`established from MCF-7 cells treated in two series with 10‘‘‘ M ta-
`
`moxifen. During the first treatment, nearly all cells died, but a few
`grew up as isolated colonies in medium without tamoxifen. Tamoxifen
`retreatment of cells propagated from one of the surviving colonies
`gave rise to the MCF-7/TAM“-1 cell line (22). Growth curves for the
`MCF-7 and the MCF-7/TAM“-1 cell lines are shown in Fig. 1. The
`doubling time of MCF-7 cells is close to 24 h, and a similar doubling
`time is found for the tamoxifen-resistant cell line in both control
`
`medium (phenol red containing DME/F12 medium supplemented
`with 0.5% FCS, 2 mM glutamine, and 6 ng/ml insulin) and in medium
`with 10" M tamoxifen. Cell proliferation of the parent MCF-7 cell line
`is inhibited by tamoxifen, and after growing in medium with tamox-
`ifen for 7 days, the majority of these cells will not survive a subcul-
`tivation in the presence of tamoxifen, whereas the MCF-7/TAM“-1
`cell line can be subcultured in the presence of tamoxifen with high
`plating efficiency.
`To test the stability of the tamoxifen-resistant phenotype of the
`MCF-7/TAM“-1 cell line, propagation was performed for 15 passages
`without tamoxifen and then a growth experiment similar to the one
`presented in Fig. 1 was performed. No difference in cell proliferation
`rate was observed in medium with and without 10* M tamoxifen (data
`
`100
`
`80
`
`80
`
`140
`
`120
`
`100
`
`B0
`
`60
`
`10"‘
`
`10'"
`
`10"
`
`10"
`
`10”
`
`10‘
`
`lcl 164,384 concentration (M)
`
`number(%ofcontrol)
`Cell
`number(%ofcontrol)
`Cell
`
`10°
`
`10‘
`
`Ceflnumber
`
`Days
`
`Fig. 1. Growth experiment with MCF-7 and MCF-7/TAM“-1 cells. The MCF-7/
`TAM“-l cells were grown for 2 weeks without tamoxifen before onset of the experiment.
`Cells (1.25 X 105) were seeded per 125 flask in control medium (DME/F12 medium
`supplemented with 0.5% FCS, 2 mM glutamine, and 6 ng/ml insulin). '1\vo days after
`seeding (day 0), the medium was changed to experimental medium, which could be either
`control medium or medium with 10" M tamoxifen. Medium was renewed every second
`or third day. On days 3, 5, 6, 7, and 10, triplicate cultures were trypsinized, and cell
`number was determined by counting in a Biirker-Turck chamber. Mean cell number and
`SD are shown by: O, MCF-7 cells in control medium; 0, MCF-7 cells in control medium
`with 10" M tamoxifen; D, MCF-7/TAM“-1 cells in control medium; I, MCF-7fTAM“-1
`cells in control medium with 10" M tamoxifen.
`
`10-"
`
`10‘‘‘’
`
`10"
`
`10‘
`
`10"
`
`ICI 182,780 concentration (M)
`
`Fig. 2. Dose-response curves for effect of ICI 164,384 (A) and ICI 182,780 (B) on cell
`proliferation of MCF-7 cells and MCF-7/1‘/KM"-1 cells. Cells were seeded in multiwell
`dishes with 1.5 X 10‘ cells per well. Two days after seeding, medium was changed to
`experimental medium containing the indicated concentrations of the 1C] compounds.
`Medium was renewed every second or third day. Cell numbers were determined after 6
`days in experimental medium by counting of cell nuclei in an automatic cell counter. The
`mean values (expressed as a percentage of the corresponding control cultures) and SDs of
`four to six experiments are shown.
`
`not shown), and the MCF-7/TAM“-1 cell line grown for 15 passages
`without tamoxifen could be transferred to tamoxifen-containing me-
`dium and passaged continuously in the presence of tamoxifen, indi-
`cating that this subline grown in tissue culture under the above men-
`tioned conditions expresses a stable, tamoxifen-resistant phenotype
`with respect to growth characteristics.
`To elucidate whether resistance to tamoxifen results in cross-resis-
`
`tance to other antiestrogens, we have studied the effect of the new
`steroidal antiestrogens ICI 164,384 and ICI 182,780 on growth of
`MCF-7/TAM“-1 cells. Fig. 2 clearly illustrates that the tamoxifen-
`resistant cell line is sensitive to both steroidal antiestrogens, although
`the sensitivity of the MCF-7/TAM“-1 cells to the steroidal antiestro-
`gens is reduced compared to parent MCF-7 cells. The IC50 for ICI
`164,384 was 9 X 10‘9 M for MCF-7fl‘AMR-1 cells, which is 18-fold
`higher than the IC50 of 5 X 10"° M for ICI 164,384 inhibition of
`parent MCF-7 cells. The IC5o for ICI 182,780 was 7 X 10"° M for
`MCF-7fI‘AMR-1 cells, which is 5-fold higher than the [C50 of 1.4 X
`10"” M for ICI 182,780 inhibition of MCF-7 cells. For both cell lines,
`the growth inhibitory effect of the steroidal antiestrogens in concen-
`trations up to 10'7 M is fully estrogen reversible (results not shown).
`Thmorigenicity in Nude Mice and ER and PgR Content.
`MCF-7 cells will form tumors in nude mice provided estrogen is
`administered, and most tumors are already palpable after 1 week
`(Table 1). The antiestrogen-resistant cell line also requires estrogen
`administration for tumor formation in nude mice, and except for one
`1589
`
`Downloaded from cancerres.aacrjourna|s.org on June 17, 2016. © 1994 American Association for Cancer Research.
`
`MYLAN PHARMS. INC. EXHIBIT 1036 PAGE 3
`
`

`
`TAMOXIFEN-RESISTANT HUMAN BREAST CANCER CELL LINE
`
`tumor that appeared after 1 week, palpable tumors in the other five
`animals inoculated with MCF-7/TAM“-1 cells appeared after 5 to 11
`weeks (Table 1). Tumors originating from either inoculated MCF-7
`or MCF-7/TAM“-1 cells (excised from animals treated with estro-
`gen) contain both free ERs and free PgRs. No difference in PgR
`level is observed between tumors arising from parent or tamoxifen-
`resistant cells. The ER level is significantly lower in tumors from
`animals injected with MCF-7/TAM"-1 cells (Mann-Whitney U test;
`P = 0.002).
`ER and PgR Content in MCF-7 Cells and in MCF-7/TAM"-1
`Cells Grown in Tissue Culture with Different Growth Media.
`Table 2 shows that the tamoxifen-resistant cell line contains free ER
`
`when grown in tissue culture in control medium with 1% FCS (and
`without tamoxifen for 15 days). The average level of free ER is lower
`than in the parent MCF-7 cell line grown under similar conditions, but
`due to the large SD values, the ER levels are not significantly differ-
`ent. The equilibrium dissociation constant for binding of estradiol to
`ER is the same for the parent and the resistant cell lines, about 10“°
`M. There is no measurable content of PgRs in the tamoxifen-resistant
`cell line grown in control medium, whereas the parent cell line con-
`tains a significant level of PgR in control medium, which contains low
`levels of estrogen compounds supplied via the FCS content. In me-
`dium supplemented with optimal estradiol concentration (10"’ M), the
`MCF-7 cells contain very few free ERs and a high level of PgR, 2050
`finol/mg cytosol protein. The MCF-7/I‘AM“—1 cells grown in estra-
`diol-containing medium contain no free ERs and few, if any, free
`PgRs. MCF-7 cells grown in medium with 10‘‘’ M tamoxifen contain
`few free ERs and a significantly reduced level of PgR compared to
`MCF-7 cells grown in control medium. The tamoxifen-resistant cell
`line contains neither free ERs nor free PgRs when grown with tamox-
`ifen. After growth in the presence of 10‘8 M ICI 164,384, no detectable
`level of free ER or free PgR is observed in the MCF-7 cells, whereas
`a low level of free ER and no free PgR are found in the MCF-7/
`TAM“-1 cell line. Both cultures of MCF-7 and MCF-7/TAM“-1 cells
`grown in control medium and in medium with estradiol or antiestro-
`gens contain salt-extractable ERs in the nucleus (results not shown).
`The complete lack of PgR expression and estrogen inducibility
`observed to occur in the MCF-7/TAM“-1 cells grown in tissue culture
`is an interesting alteration, and we have investigated whether this
`characteristic is maintained after long term growth in medium without
`tamoxifen. The result of PgR determinations disclosed a small in-
`crease in estrogen—induced PgR level during the first 2-4 weeks with-
`out tamoxifen, and the level then stayed rather constant during the
`propagation for up to 15 passages without tamoxifen. The average
`estrogen-induced PgR level was 290 frnol/rng cytosol protein, which
`is only 14% of the PgR level in MCF-7 cells grown with estradiol.
`
`Table 2 ER and PgR content in MCF-7 and in MCF-7/TAM“-I cells grown in tissue
`culture in control medium, medium with estradiol, and medium with the antiestrogen:
`tamoxifen or ICI 164,384
`MCF-7/TAM“-1 cells were grown for 9 to 10 days without tamoxifen before onset of
`the experiment. Estradiol and antiestrogens were added to cultures in the early exponential
`growth phase, and cells harvested after 6 days. Mean and SD for ER and PgR content
`determined by dextran-coated charcoal assay on cytosol preparations are shown, and the
`number of experiments are indicated in the parentheses.
`MCF-7
`
`Mci-'-7/rAM"—1
`
`Control medium 1% FCS
`
`Medium with 10’9 M estradiol
`
`Medium with 10" M tamoxifen
`
`Medium with 10'8 M ICI 164,384
`
`" fmol/mg, cytosol protein.
`
`Free ER”
`(II)
`442 : 272
`(8)
`14 I 2
`(4)
`18 1 24
`(3)
`<10
`(3)
`
`Free PgR‘
`(II)
`361 I 83
`(8)
`2050 I 504
`(4)
`70 : 64
`(3)
`<10
`(3)
`
`Free ER"
`(H)
`271 I 138
`(3)
`<10
`(4)
`<10
`(4)
`42 1 52
`(4)
`
`Free PgR‘
`('1)
`<10
`(3)
`17 t 17
`(4)
`<10
`(4)
`<10
`(4)
`
`MCF-7
`
`TAM"-1
`
`5
`3
`
`2 _
`3
`2 0 75 9 3
`
`I _
`u 35 2 3
`
`q
`3
`
`zoo»
`97_
`
`69>
`
`Z.‘ .. B 3 ;; 3
`E 2 3 =
`=3
`i I
`.,‘ 6
`.
`,
`C — Q ‘I 31
`
`
`A ‘mi
`
`46> e
`30> 4 A
`
`_ D .
`
`233
`
`Fig. 3. Autoradiograrn of [35S]methionine-labeled proteins secreted from MCF-7 and
`MCF-7/I'AM"—1 cells analyzed by SDS-polyacrylamide gel electrophoresis. The cells
`were grown as described in Fig. 2. The concentrations of hormones and antihonnones in
`the experimental media were 10“‘ M tamoxifen (TAM), 10’3 M lCl 164,384 (ICI), and 10‘‘’
`M estradiol (E2). [35S]Methionine in serum-free medium was added to the cultures at day
`6, and conditioned media were collected after 6 h. Equal amounts of TCA-precipitable
`cpm were applied on each lane. C, lanes with conditioned medium from control cultures;
`MW, lane with molecular weight markers. Arrows to the right, positions for the estradiol-
`regulated proteins M, 66,000 oz.-antichymotrypsin, M, 61.000 a,-antitrypsin, M, 52,000
`cathepsin D, and M, 42,000 protein.
`
`This result shows that the MCF-7/TAM“-1 cells do not regain parent
`capacity to induce PgR synthesis when grown in tissue culture without
`tamoxifen.
`
`Table 1 Tumor takes in athymic nude mice. Estrogen and
`progesterone receptor content
`Near confluent cultures were harvested by trypsinization and about 107 cells were
`inoculated s.c. at the fourth mammary gland of female athymic nude mice. Some of the
`animals received estrogen supplementation. The animals were observed weekly for about
`6 months. ER and PgR contents were determined by the dextran charcoal-coated assay on
`cytosols prepared from tumors originating from inoculated MCF-7 or MCF-7/1"AMR-1
`cells (mean and SD). The numbers in parentheses are the number of tumors included in
`the analyses.
`
`Tumor takes
`— estrogen
`1171;
`
`P t bl
`Tumor mm :";',:,s°
`+ estrogen
`(wk)
`5/3
`1.6
`
`Cell line
`MC].-.7
`
`R
`°’“
`“°‘”'7”"“ "
`1: fmovmg’ cytosol P,o,cin_
`’’One small nonprogressively growing tumor.
`
`6”
`
`H‘
`
`Fm Egg
`(n)
`315 -_~ 206
`2 (8)
`‘gs?’
`
`Effect of Estradiol and Antiestrogens on the Synthesis of Se-
`creted Proteins in MCF-7 and MCF-7/TAM“-1 Cells. The obser-
`vation of lack of synthesis and estrogen inducibility of the PgR protein
`encouraged a further analysis of hormone regulation of estrogen-
`regulated proteins in MCF-7 cells. Fig. 3 shows [”S]methionine-
`labeled secreted proteins from MCF-7 and MCF-7/PAM“-1 cells
`grown in control medium, medium with 10" M tamoxifen, medium
`with 10’5 M ICI 164,384, and medium with 10‘9 M estradiol. The
`patterns of secreted proteins in control cultures of MCF-7 and MCF-
`7/TAMR-1 are different in that MCF-7 cells secrete substantially more
`M, 52,000 cathepsin D (5-fold; densitometric scanning) than MCF-7/
`TAM“-1 cells. The MCF-7/TAM“-1 cells secrete about twice as much
`.
`.
`.
`.
`.
`.
`Fm PER. M, 61,000 oi,-antitrypsin as MCF-7 cells. A striking difference in.the
`(n)
`effect of tamoxifen on secreted proteins is noticed because tamoxifen
`252 3 195
`has no effect on secreted proteins in MCF-7fI'AM“-1 cells, whereas
`(E)
`tamoxifen stimulates the secretion of the M, 52,000 cathepsin D
`”°(;)“
`protein (3-fold) and the M, 42,000 protein (3-fold) in the parent
`MCF-7 cells. The steroidal antiestrogen ICI 164,384 inhibits M,
`52,000 cathepsin D secretion and stimulates M, 42,000 protein syn-
`1590
`
`Downloaded from cancerres.aacrjourna|s.org on June 17, 2016. © 1994 American Association for Cancer Research.
`
`MYLAN PHARMS. INC. EXHIBIT 1036 PAGE 4
`
`

`
`TAMOXIFEN-RESISTANT HUMAN BREAST CANCER CELL LINE
`
`thesis in both cell lines. Estradiol stimulates secretion of M, 52,000
`cathepsin D and M, 61,000 oz.-antitrypsin and inhibits the secretion of
`the M, 42,000 protein in both cell lines. The M, 66,000 a,-antichy-
`motrypsin protein is secreted at a rather low level in control cultures
`of both cell lines, and no significant estrogen or antiestrogen regula-
`tion was found in these experiments.
`In both cell lines, estradiol significantly stimulates the secretion of
`a [35S]cysteine-labeled protein with M, 7000-8000, presumed to be
`the pS2 protein (41), and cultures of both cell lines grown with either
`tamoxifen or ICI 164,384 do not secrete detectable levels of this
`protein (results not shown).
`Effects of Estradiol and Antiestrogens on the Synthesis of Spe-
`cific mRNAs Encoding Estrogen-regulated Proteins. To investi-
`gate whether the observed changes between the MCF-7/TAM“-1 and
`the parent cell line in secreted estrogen-regulated proteins could be
`reflected also at the mRNA level, we studied the mRNA expression of
`cathepsin D, a,-antitrypsin, at,-antichymotrypsin, and pS2. Densito-
`metric scannings of the Northern blots (adjusted for difference in
`loading by measuring the level of the housekeeping gene GAPDH)
`reveal
`that
`the level of cathepsin D, oz,-antitrypsin, and al-
`antichymotrypsin mRNA in MCF-7/TAM“-1 cells grown in both
`control medium and in medium with estradiol, tamoxifen, or ICI
`164,384 are considerably lower than in the corresponding parent cul-
`tures. Fig. 4A shows that the 2.2—kilobase cathepsin D mRNA ex-
`pression in MCF-7 is significantly stimulated by both estradiol and
`tamoxifen, whereas only estradiol exerts a marked stimulation in
`
`MCF-7/TAM“-1 cells. The level of pS2 mRNA in control cultures of
`MCF-7/PAM“-1 cells is significantly higher than in parent MCF-7
`cells, and estradiol stimulates pS2 mRNA level to the same extent in
`both cell
`lines. Tamoxifen inhibits pS2 mRNA in both cell
`lines
`(Fig. 4B). The oz,-antitrypsin mRNA expression is stimulated by es-
`tradiol and to a smaller extent by tamoxifen in both cell lines (Fig.
`4C), whereas the oz,-antichymotrypsin mRNA synthesis in both
`MCF-7 and MCF-7/TAM“-1 cells is stimulated by estradiol and is
`unrespons