Please indicate author’s corrections in blue, setting errors in red

146705 BREA ART.NO 967-97 ORD.NO 230332

Address for offprints and correspondence: Maria Grazia Sacco, ITBA-CNR, via Ampere 56, 20131 Milan, Italy; Tel: (0039) 270643384;Fax: (0039) 22663030

Breast Cancer Research and Treatment 47: 171–180, 1998. 1998 Kluwer Academic Publishers. Printed in the Netherlands.

Report

Establishment and characterization of a new mammary adenocarcinomacell line derived from MMTV neu transgenic mice

Maria Grazia Sacco,1 Laura Gribaldo,2 Ottavia Barbieri,3 Gino Turchi,4 Ileana Zucchi,1 Angelo Collotta,2 LucaBagnasco,1 Domenico Barone,5 Cristina Montagna,1 Anna Villa,1 Erminio Marafante,2 and Paolo Vezzoni11Istituto di Tecnologie Biomediche Avanzate, CNR, Milano; 2European Communities, Joint Research Center,ECVAM, Ispra; 3Dip. Oncologia Clinica e Sperimentale, Università di Genova, Genova; 4Laboratorio di GeneticaUniversità, Pisa; 5Istituto di Ricerche Biomediche ‘A. Marxer’ RBM SpA, Colleretto di Giacosa (TO), Italy

Key words: mammary adenocarcinoma cell line, MG 1361, MMTV-neu transgenic mice, breast cancer,hormone responsiveness, apoptosis

Summary

A new murine cell line, named MG1361, was established from mammary adenocarcinomas arising in aMMTV-neu transgenic mouse lineage where breast tumors develop in 100% of females, due to the over-expression of the activated rat neu oncogene in the mammary gland. The MG1361cell line shows an epithelial-like morphology, has a poor plating efficiency, low clonogenic capacity, and a doubling time of 23.8 hours.Karyotype and flow cytometry analysis revealed a hypotetraploid number of chromosomes, whereas cell cycleanalysis showed 31.2% of cells to be in the G1phase, 21.4% in S and 47.4% in G2 + M. This cell line maintains ahigh level of neu expression in vitro. The MG1361 cell line was tumorigenic when inoculated in immunodef-icient (nude) mice and the derived tumors showed the same histological features as the primary tumors fromwhich they were isolated. MG1361cells were positive for specific ER and PgR binding which was competed bytamoxifen, making this cell line useful for the evaluation of endocrine therapy. Moreover, they were sensitiveto etoposide treatment, suggesting that they could be a model for the study of chemotherapy-induced apopto-sis. As the tumors arising in MMTV-neu transgenic mice have many features in common with human mam-mary adenocarcinomas (Sacco et al., Gene Therapy 1995; 2: 493–497), this cell line can be utilized to performbasic studies on the role of the neu oncogene in the maintenance of the transformed phenotype, and to testnovel protocols of therapeutic strategies.

Introduction

Breast cancer is one of the most frequent tumorsoccurring in humans, and although primary tumorscan be surgically removed with some success, over50% of affected women die as a result of the meta-static spread of the disease [1]. Overexpression ofthe Erb-B2 gene (neu oncogene) due to genomic

amplification often occurs in human breast tumorsand is recognized as a factor contributing to a badprognosis in a subset of these neoplasms [2–5]. Inorder to investigate the role of this oncogene in thepathogenesis of mammary carcinomas, a transgenicmouse lineage in which overexpression of the acti-vated rat neu oncogene was directed to the mam-mary gland by the regulatory regions of the mouse

Please indicate author’s corrections in blue, setting errors in red

146705 BREA ART.NO 967-97 ORD.NO 230332

172 MG Sacco et al.

mammary tumor virus long terminal repeat(MMTV-LTR) was produced [6, 7]. As previouslyreported, breast tumors arising in these mice pre-sent several features in common with human mam-mary tumors, including the ability to induce metas-tases to distant organs [7, 8]. For this reason, thistransgenic lineage has been used as a model for aretroviral-mediated gene therapy approach [9]. Al-though a significant reduction of the tumor growthwas obtained in comparison to the control tumors,several factors related to the particular character-istics of the adenocarcinoma cells have been shownto reduce the efficacy of the treatments [9]. As invivo studies are difficult to perform, the availabilityof a cell line representative of this kind of neoplasiaallows the preliminary evaluation of new ther-apeutic protocols. Therefore, we established abreast adenocarcinoma cell line, MG1361, fromthese tumors. This cell line seems to maintain thesame features as the primary tumors (epithelial-likemorphology, high level of neu expression, and hightumorigenicity when inoculated in nude mice), pro-viding the material to investigate the pathwaysleading to neu-dependent transformation in vitro.In addition, this cell line can be useful for studyingthe role of apoptotic cell death in response to che-motherapeutic agents.

Apoptosis is a programmed form of cell deathwhich is now recognised as being important in thepathogenesis of several human diseases, includingneoplasia. The circumvention of programmed celldeath permits the survival of genetically alteredclones that would otherwise have been deleted.This allows the propagation of defects that wouldnormally have been eliminated through p53 de-pendent apoptosis [10]. The ability to modify sensi-tivity to apoptosis has clear implications for thetreatment of malignancies [11]. Potential strategiesfall into three categories: direct triggering of apop-tosis by cytotoxic agents, enhancing susceptibilityto apoptosis to increase the efficacy of other ther-apies, and boosting the resistance of normal cells toapoptosis with survival factors [11, 12]. In particular,tamoxifen and other estrogen antagonists induce anincrease in cell death in human mammary cancercell line MCF7 [13]. The finding that the survival oftumor cells may be hormone-dependent may sug-

gest new strategies for tumor therapy. The cell linedescribed in this paper is positive for ER and PgRand sensitive to tamoxifen. This will make the in-vestigation of different therapeutic approaches us-ing an in vitro system which can be directly correlat-ed with the correspondent in vivo model possible.

Materials and methods

Transgenic mouse lineage and MG1361 cell lineestablishment

The production of MMTV-neu transgenic mice aswell as the histological features of the tumors havebeen previously described [7, 8]. RNA extractionand Northern blot analysis was performed accord-ing to standard methods [14]. Breast tumors from4-month old transgenic mice were removed andrinsed separately 3 times in PBS with 100 U/ml pen-icillin and 100 µg/ml streptomycin (pen/strep).Each tumor was then minced into small pieces(0.5 mm each) with a sterile scalpel and transferredinto trypsin-EDTA (Sigma) solution at 37 °C for 30min. The cell suspension was then centrifuged at1200 rpm for 5 min, and the pellet containing thecells was plated and cultured in William’s E mediumsupplemented with Foetal Calf Serum (FCS) (Sig-ma), antibiotics (pen/strep), non-essential aminoa-cids (GIBCO/BRL), and 2 mM glutamine, until thecells formed a subconfluent monolayer. Prelimina-ry experiments had been performed to define theoptimum in vitro growth conditions by growing thecells in different culture media (Dulbecco’s Mod-ified Eagle’s Medium, DMEM, RPMI-1640 or Wil-liam’s E Medium; Sigma) supplemented with 10 or15% FCS (Sigma). Two cell lines derived from twoindependent tumors were obtained. In particular,clone MG1361 was maintained in culture for 80 pas-sages and fully characterized.

Removal of sex steroids by CD treatment of serum

Charcoal (Norit A, acid-washed; Sigma) waswashed twice with cold sterile water immediatelybefore use. A CD (50 g/l charcoal-5 g/l dextran T70,

Please indicate author’s corrections in blue, setting errors in red

146705 BREA ART.NO 967-97 ORD.NO 230332

Breast adenocarcinoma cell line from transgenic mice 173

Pharmacia-LKB, Uppsala Sweden) suspension wasprepared. A volume similar to that of serum to beprocessed was centrifuged at 100 g for 10 min. Su-pernatant was aspirated, serum was added, and thepellet was resuspended by rolling at 4 cycles/min at37 °C for 1 hour. The suspension was centrifuged at2000 g for 20 min. The supernatant was then filteredthrough a 0.45 µm (pore size) Nalgene filter. Morethan 99% of serum sex steroids was removed by thistreatment. This serum was added to the culture me-dium as previously described, in order to comparethe growth characteristics of the MG1361 cell line inthe presence or absence of steroid hormones.

Cell morphology and in vitro growth characteristics

The morphology of clone MG1361 was observedunder an inverted microscope (Olympus) using cyt-ological staining with 100% May-Grunwald for 5min counterstained with 20% Giemsa for 20 min.

Plating efficiency. Cells were trypsinized, counted,and seeded in 35 mm plates (2500, 1200, 600, 300,150, 75 cells each in 6 plates). Two independent ex-periments were performed with 6 replicates. After 7days, adherent colonies were stained with 20%Giemsa for 20 min and counted. The plating effi-ciency was calculated as the number of colonies/number of cells seeded × 100.

Population doubling time. The growth curve wasperformed by plating 105 cells/well on 6-well plates(Falcon) and counting them after 24, 48, 72, and 96hours of culture. Each experiment was performed 4times.

Soft-agar assay. Anchorage-independent growthwas studied by a clonogenic assay determined in abilayer soft agar system modified by Hamburger etal. [15]. The cells at the 30th passage were plated in0.3% agar at three different concentrations: 103, 104,and 5 × 104 cells for 60 mm plates. Formation ofclones was evaluated by light microscopy after 7, 14,and 21 days. Colonies consisting of more than 50cells were counted as agar-growing colonies.

Karyotype analysis. Cells in logarithmic growth(70% confluent) were exposed to colcemid (10 µg/ml, Gibco, Scotland) for 2 hours at 37 °C, harvestedusing trypsin/EDTA, and centrifuged. The pelletwas treated with hypotonic KCl (0.075 M) for 15min at 37 °C. Cells were fixed in 3 passages of meth-anol:acetic acid solution (3:1) and spread on slides.The chromosome complement of cells was estab-lished after scoring at least 100 metaphases stainedaccording to the conventional Giemsa technique.

Coulter Epics profile. For the flow cytometric analy-sis, cells were trypsinized, washed, suspended inPBS, and fixed in 70% ethanol. Fixed cells weretreated with RNAse (0.1 mg/ml) and stained withpropidium iodide (40 µg/ml) for 30 min at roomtemperature. DNA measurements were carried outby analyzing 106 cells on Coulter Epics.

Analysis of RNA expression

Total RNA was extracted from the control mam-mary gland, transgenic breast tumor, and MG1361cell line with the RNA-easy kit (Promega). Five mi-crograms of each RNA were run on a denaturatinggel, blotted to Hybond N + membrane, and hybrid-ized with a rat neu cDNA probe according to stan-dard procedures, with minor modifications [14].

Analysis of tumorigenic potential in nude mice

The tumorigenicity of clone MG1361was assayed invivo in nude mice. For this purpose, the cells weretrypsinized, resuspended in PBS, and inoculatedsubcutaneously in nude mice (Harlan Nossan) atthe concentration of 1, 3, 10 ×106 in 0.3 ml and left togrow until the tumor mass reached about 0.5 cm3.For histopathological analysis a complete necropsywas performed, essentially as previously described[8]. In particular, tumors were removed and fixedby immersion in 10% formalin for at least 24 hoursat room temperature and processed in paraffin wax.Serial sections (3–5 µm) were stained with hema-toxylin and eosin for conventional evaluations.

Please indicate author’s corrections in blue, setting errors in red

146705 BREA ART.NO 967-97 ORD.NO 230332

174 MG Sacco et al.

Table 1. In vitro growth characteristics of MG1361 cells

Population doubling time (h) 23.8Plating efficiency

(n. of cells) %75 1.5150 1.5300 3.0600 3.01200 4.02500 4.0

Agar clonogenicity(n. of cells) %103 0.5104 3.15 × 104 18.4

Figure 1. Morphology of MG1361 cells observed by an invertedmicroscope (400 × magnification).

Sensitivity to chemotherapy through evaluation ofapoptosis

MG1361 cells were incubated with either the medi-um alone or together with etoposide (VP16) at adosage of 10 µg/ml and 100 µg/ml. After 24 hours ofexposure, the cells which had detached and werefloating in the supernatant were trypsinized andcollected in different samples for staining with An-nexin V (Bender System) and propidium iodide forflow cytometry analysis as previously described[16]. In order to determine the DNA laddering, to-tal DNA was extracted according to standard pro-cedures [14] from supernatant and adherent cellstreated for 24 hours with 0, 10, and 100 µg/ml etopo-side and electrophoresed on a 2% agarose gel.

Estrogen and progesterone receptor determinationand sensitivity to tamoxifen

MG1361 cells were plated in 24 well-plates (200,000cells ± 5%/well) in William’s E medium withoutphenol red and with 15% CD-treated FCS, 2 mMglutamine, pen/strep, and non-essential aminoacidsfor 24 hours. The medium was then replaced withthe same medium without phenol red and serum for90 min at 37 °C.

Radioligands, 10 µl of 3H-estradiol (S.A. 84.1 Ci/mmol, NEN), and 3H-progesterone (S.A. 94.0 Ci/mmol, Amersham) were then added with 10 µl ofculture medium (controls) or of a solution of dipro-pionated estradiol or cold progesterone (10-6 M fi-nal concentration for the determination of the re-spective non-specific binding) or of tamoxifen(SIGMA, 10-7 M final concentration).

The plates were gently shaken for 5 min and thecells incubated for 1 hour as stated above. At theend of the incubation, the medium was removed byaspiration and the cells washed 3 times with PBS atroom temperature. 1 ml of 80% EtOH was thenadded to each well and left overnight at room tem-perature. The radioactivity of each well was thencounted in minivials by diluting the EtOH (1 ml)with 4 ml of a scintillation cocktail (Filger-CountPackard), 5 min count, about 40% counting effi-ciency). The experiment was run twice in triplicate.

The final concentration of 3H estradiol was between0.25 and 12.5 nM. 3H progesterone concentrationranged from 0.23 and 12 nM.

Results

In vitro growth characteristics

Cells maintained in culture are polygonal, the nu-clei show many nucleoli, and the cells maintain anepithelial-like morphology (see Figure 1). Theirgrowth is influenced by the culture medium. Theoptimal growing conditions are obtained by usingWilliam’s E medium supplemented with 15% FCS,non essential amino acids, 100 U/ml penicillin, and

Please indicate author’s corrections in blue, setting errors in red

146705 BREA ART.NO 967-97 ORD.NO 230332

Breast adenocarcinoma cell line from transgenic mice 175

Table 2. Flow cytometric analysis of DNA index, coefficient of variance (CV) and cell cycle of MG1361 cells

Cell line DNA index CV Cells (%) in

G1 S G2 + M

erythrocytes 1.0 3.7 95.0 1.9MG 1361 1.7 4.6 31.2 21.4 47.74

Figure 2. Histogram of the distribution of the number of chromosomes observed in cultured breast tumor cell line (clone MG1361)derived from MMTV-neu transgenic mice. The values are based on the analysis of 100 metaphases and are expressed as the percentage ofmetaphases in each ploidy class.

100 µg/ml streptomycin. The cells grow with thesame characteristics in medium supplemented withcharcoal-dextran treated serum, showing a hor-mone-independent growth.

The MG1361 cell line was plated in 6 well plates(105 cells/plate), left to grow, and counted every 24hours with a Burker chamber. The cells were247,000, 546,000, 776,000, and 1,240,000 after 24, 48,72, and 96 hours of culture respectively. As summa-rized in Table 1, the MG1361 cell line shows a pop-ulation doubling time of 23.8 h. They have a poorplating efficiency (1.5%–4%) as described in Mate-rials and Methods, and a low clonogenic capacity(18.4% with the highest concentration of cells).

Karyotype analysis

Figure 2 shows the distribution of the number ofchromosomes in 100 metaphases. MG1361 does notdemonstrate a diploid karyotype. At the 21st pas-sage the chromosome number distribution in themetaphases was bimodal, showing peaks at 39–40and 80. The range of distribution was 39–174, and69% of metaphases analyzed were hypotetraploid.

FACS analysis

Figure 3 and Table 2 show the DNA distributionprofile of the MG1361 cell line obtained by flow cy-tometry. The DNA index was calculated as the ratio

Please indicate author’s corrections in blue, setting errors in red

146705 BREA ART.NO 967-97 ORD.NO 230332

176 MG Sacco et al.

Table 3. Estrogen and progesterone receptor binding

Bmax* fM KDnM KAnM-1

3H-estradiolexp. n. 1 47.75 0.70 1.43exp. n. 2 52.50 1.60 0.63+ Tamoxifen 10-7 M 52.50 3.25 0.31

3H-progesteroneexp. n. 1 102.0 0.93 1.08exp. n. 2 120.0 0.73 1.37

* specific binding referred to 200,000 ± 5% cells.

Figure 3. Flow cytometric plots. a) MG1361 cells; b) erythrocytecontrol cells. In the abscissa the increasing fluorescence intensityin proportion to the DNA content (channel number) is reported.In the ordinate the number of cells is shown.

Figure 4. Northern blot analysis of neu transgene expression inthe control mammary gland (lane 1), MG1361 cell line (lane 2),and MMTV neu transgenic breast tumor (lane 3).

Figure 5. Histopathological analysis of paraffin embedded sec-tions of tumors derived from the in vivo inoculation of MG1361cells (a) compared with spontaneous arising tumors of MMTV-neu transgenic mice (b).

between the mean channel of G0/G1 peak ofMG1361 sample to that of the diploid control sam-ple (avian erythrocyte nuclei). The S-phase per-centage represents the proliferative activity of thecells and is determined by using an analysis pro-gram. For each population the G1 peak variance isshown (Table 2). MG1361 had its main G1 peak atchannel 170 and the second one at channel 339. Thecontrol channel was 95. We can, therefore, confirmthat MG1361 is a hypotetraploid cell line as previ-

ously determined with the karyotypic analysis. Asshown in Table 2, cell cycle analysis indicated that31.2% of the cells were in G1, 47.4% in G2, and21.4% in the S phase.

Please indicate author’s corrections in blue, setting errors in red

146705 BREA ART.NO 967-97 ORD.NO 230332

Breast adenocarcinoma cell line from transgenic mice 177

Figure 6. Flow cytometric analysis of apoptosis in MG1361cells. The cells were cultured in medium (CTRL) or in medium + etoposide (10and 100 µg/ml) for 24 hours. Supernatant and trypsinized cells were directly analyzed after addition of annexin V (FITC) and propidium.iodide (PI). Annexin positive/P1 negative cells represent the percentage of apoptotic cells. (a) supernatant (SN) CTRL; (b) SN 100 µg/mletoposide; (c) SN 10 µg/ml etoposide; (d) CTRL; (e) 100 µg/ml etoposide; (f) 10 µg/ml etoposide.

Analysis of RNA expression

Figure 3 shows the expression level of the neu onco-gene in the non-transgenic mammary gland, in theMG1361 cell line, and in the transgenic breast tu-mor. Neu expression was absent in the normalmammary gland (lane 1), and high in the primarytumor (lane 3). This high expression, typical of tu-mors arising in MMTV-neu mice, is maintained at asimilar level in the MG1361 cell line (lane 2), con-firming that the neu oncogene expression was notaffected by in vitro culture conditions.

Tumorigenicity in vivo

To investigate whether the malignant potential ismaintained by the MG1361 cell line, cells were in-oculated subcutaneously in immunodeficient miceand grown until the nodules reached the mass ofabout 0.5 cm3. The tumors arose as solid masses thatbecame palpable after 10 days post-inoculation andreached the indicative mass at days 22, 34, and 38for the concentrations of 10, 3, 1 × 106 cells, respec-tively. Due to the short period in which these tu-mors were allowed to grow, no metastases were ob-served in nude mice. The histopathological analysisof these adenocarcinomas shows that the morphol-

Please indicate author’s corrections in blue, setting errors in red

146705 BREA ART.NO 967-97 ORD.NO 230332

178 MG Sacco et al.

Figure 7. Evaluation of estrogen receptors (ER) in MG1361 cellsand dose-related inhibition of radiolabeled 3H-estradiol bindingby 10-7 M tamoxifen.

ogy of the grafted tumors (Figure 4a) is comparableto that of the primary mammary adenocarcinomas(Figure 4b) arising in MMTV-neu transgenic mice.

Sensitivity to chemotherapy through apoptosisevaluation

Figure 5 shows the percentage of apoptotic cells inthe MG1361 cell population grown in physiologicalconditions, and after treatment with etoposide. Af-ter 24 hours of treatment with 10 µg/ml of etopo-side, 51.8% of the cells in the supernatant expressedphosphatidylserine at the cell surface, which is anearly marker for apoptosis, and were negative forpropidium iodide (PI). This represents a slight in-crease with respect to the control sample whichshows 45% of apoptotic cells. With 100 µg/ml of eto-poside, 84% of cells in the supernatant were dead(positive for annexin and also for PI). This indicatesthat the apoptotic cells detach themselves from themonolayer and are found mainly in the supernatant[16]. Cells trypsinized from the monolayer showed anegligable rate of apoptosis in the controls (< 2%).We only obtained 13% of apoptosis in the monolay-er with the highest dose (100 µg/ml) of etoposide.Induction of apoptosis was also determined by ana-lyzing the laddering of the DNA extracted from su-pernatant and adherent cells treated with 0, 10, and100 µg/ml etoposide. The laddering was only pre-

sent in the DNA extracted from the cells of the su-pernatant treated with 100 µg/ml etoposide (datanot shown) confirming the results obtained with theannexin V method.

Estrogen receptors determination and sensitivity totamoxifen

As shown in Figure 7, MG1361 cells are positive forER and PgR and the levels have been determined(Table 3).

Dose-related competition of tamoxifen againstestrogen ER-binding was assessed (Figure 7). Thepresence of tamoxifen 10-7 M reduced the receptorbinding affinity for 3H-estradiol, but not the num-ber of the receptor sites. The difference in specifici-ty in presence of tamoxifen was eliminated by in-creasing the concentrations of 3H-estradiol, bothcurves reaching the Bmax, showing that tamoxifenwas able to compete against binding of radiolabeledestradiol to ER.

Discussion

Mice transgenic for oncogenes are a valuable modelfor studying the multiple-step pathway involved inthe pathogenesis of neoplastic diseases [17–19].They provide spontaneously growing tumors inwhich at least one of the primary determinants isclearly defined. In this regard, we have extensivelycharacterized a neu-dependent breast neoplasmarising in transgenic mice which has several fea-tures in common with its human neoplastic counter-part, and we have exploited it as an appropriatemodel for gene therapy studies [8, 9].

This study describes a new murine cell line de-rived from a mammary adenocarcinoma arising inthese mice. The new cell line MG1361 shows an epi-thelial-like morphology typical of adenocarcinomacells, a hypotetraploid number of chromosomes,and a high percentage of cells in the G2 + M phase(47,4%). The in vitro studies demonstrated a shortdoubling time (23.8 h), poor agar clonogenicity(18.4% with the highest concentration of cells con-sidered), and a very low plating efficiency (1.5–4%).

Please indicate author’s corrections in blue, setting errors in red

146705 BREA ART.NO 967-97 ORD.NO 230332

Breast adenocarcinoma cell line from transgenic mice 179

This suggests that this cell line needs a mutual cell-cell interaction or some growth factor release for itsreplication. The pathway for physiological as wellas VP16-induced apoptosis is maintained, pointingout that presumably there are no mutations in Bcl-2and p53 proteins [20–22]. At least in some modelsystems, initiated cells and pre-malignant cell pop-ulations have been found to exhibit evident cell rep-lication, and also enhanced apoptotic activity, com-pared to the normal tissue [23]. Furthermore, someobservations have shown that tumor cell growthmay still depend on the presence of trophic hor-mones [24]. This is in keeping with our findings of anormal p53 gene in the MMTV-neu primary tu-mors, determined by PCR amplification and se-quencing (Sacco MG, unpublished). When inocu-lated in vivo in immunodeficient mice, these cellsgive rise to tumors with the same histological char-acteristics as the spontaneously arising breast neo-plasias observed in MMTV-neu transgenic mice,which have many features in common with humanmammary adenocarcinomas [7, 8].

This cell line allows us to perform basic studies onthe molecular and cellular components of this tu-mor. The maintenance of the sensitivity to apopto-sis could be exploited to investigate the mechanismof action of chemotherapeutics such as etoposide ordoxorubicin [21, 22, 25, 26] or to test new drugswhich act, at least in part, through induction of thispathway. Chemotherapeutic agents and radiationinduce tumor cell death primarily by causing dam-age which induces the cells to commit suicide [27].Furthermore, breast cancer often undergoes a re-gression when treated with estrogen antagonists.Tamoxifen is a widespread anti-hormonal com-pound used for the treatment of human breast can-cer. Its effects are mediated via estrogen receptors,but the molecular basis of its activity has yet to beclarified. This established cell line, which is positivefor ER and PgR receptors and sensitive to tamoxi-fen, allows us to investigate the cellular and mole-cular events in vitro and in vivo which underline thedevelopment of breast cancer, and to evaluate moremechanistic approaches in therapy. In addition,MG1361 cells can be used to evaluate novel genetherapy approaches. Amplification and overex-pression of the Erb-B2 mutated gene (neu onco-

gene) are a characteristic of the majority of humanbreast adenocarcinomas [3–5]. Since these cellsmaintain a high level of neu expression in vitro,comparable with the level of expression in primarytumors of MMTV-neu transgenic mice, the MG1361cell line could be a suitable tool for the in vitro test-ing of experimental protocols of tumor growth in-hibition targeted to the neu function, such as thosebased on immuno-therapy directed against Erb-B2mutated protein [28] and antisense inhibition of ex-pression [29] that are now beginning to be studied.

The efficacy of therapeutic protocols based onantisense oligonucleotides as well as on viral vec-tors (retroviral or adenoviral) carrying suicidegenes or antisense constructs, can be assayed in thiscell line before their application to the in vivo trans-genic model in which tumors arise spontaneouslyand escape tumor immunosurveillance. This will al-low rapid preliminary screening of different ap-proaches using the in vitro growing cells as a screen-ing strategy.

Acknowledgements

The technical assistance of Dario Strina, Lucia Su-sani, Valeria Fasolo, and Enrica Mira Cato is ac-knowledged. We thank Dr. Silvana Canevari for heruseful suggestions and Ms. Sophie Bevan for hertyping of the manuscript. The financial support ofAIRC to AV is gratefully acknowledged. The workwas also partly supported by a grant from PFACRO of CNR. This is manuscript no. 8 of the Ge-nome 2000/ITBA Project funded by CARIPLO.

References

1. Klijn JGM, Berns PMJJ, Bontenbal M, Alexieva-Figusch J,Foekens JA: Clinical breast cancer: New developments inselection and endocrine treatment of patients. J Steroid Bio-chem Molec Biol 43: 211–221, 1992

2. Porter-Jordan K, Lippman ME: Overview of the biologicmarkers of breast cancer. Hematol Oncol Clin North Am 8:73–100, 1994

3. Maguire HC, Greene MI: Neu (c-erbB-2), a marker in carci-noma of the female breast. Pathobiol 58: 297–303, 1990

Please indicate author’s corrections in blue, setting errors in red

146705 BREA ART.NO 967-97 ORD.NO 230332

180 MG Sacco et al.

4. Hynes NE, Stern DF: The biology of erbB-2neu/HER-2 andits role in cancer. Biochim Biophys Acta 1198: 165–184, 1994

5. Dougall WC, Quian X, Peterson NC, Miller MJ, Samanta A,Greene MI: The neu-oncogene: signal transduction path-ways, transformation mechanisms and evolving therapies.Oncogene 9: 2109–2123, 1994

6. Muller WJ, Pattengale PK, Wallace R, Leder P: Single-stepinduction of mammary adenocarcinoma in transgenic micebearing the activated c-neu oncogene. Cell 54: 105–115, 1988

7. Lucchini F, Sacco MG, Hu N, Villa A, Brown J, Cesano L,Mangiarini L, Rindi G, Kindl S, Sessa F, Vezzoni P, Clerici L:Early and multifocal tumors in breast, salivary, harderianand epidermal tissues developed in MMTV-Neu transgenicmice. Cancer Lett 64: 203–209, 1992

8. Sacco MG, Mangiarini L, Villa A, Macchi P, Barbieri O, Sac-chi MC, Monteggia E, Fasolo V, Vezzoni P, Clerici L: Localregression of breast tumor following intramammary ganci-clovir administration in double transgenic mice expressingneu oncogene and herpes symplex virus thymidine kinase.Gene Ther 2: 493–497, 1995

9. Sacco MG, Benedetti S, Duflot-Dancer A, Mesnil M, Bag-nasco L, Strina D, Fasolo V, Villa A, Macchi P, Faranda S,Vezzoni P, Finocchiaro G: Partial regression, yet uncom-plete eradication of mammary tumors in transgenic mice byretroviral mediated HSV-TK transfer in vivo. Gene Ther, inpress

10. Williams GT: Programmed cell death: apoptosis and onco-genesis. Cell 65: 1097–1098, 1991

11. Hickman JA: Apoptosis induced by anticancer drugs. Can-cer Metas Rev 11: 121–139, 1992

12. Thompson CB: Apoptosis in the pathogenesis and treat-ment of disease. Science 267: 1456–1462, 1995

13. Kienzi H: Antiestrogen induced cell death in culturedMCF7 cells, submitted

14. Sambrook J, Fritsch EF, Maniatis T: Molecular Cloning. ALaboratory Manual. Cold Spring Harbor Laboratory Press,Cold Spring Harbor, NY, 1989

15. Hamburger A, Salmon SE, Kim MB, Trent JM, SoehnlenBJ, Alberts DS, Schmidt HJ: Direct cloning of ovarian carci-noma cells in agar. Cancer Res 38: 3438–3444, 1978

16. Koopman G, Reutelingsperger CPM, Kuijten GAM, Keeh-nen RMJ, Pals ST, van Oers MHJ: Annexin V for flow cyto-metrical detection of phosphatidylserine expression on Bcells undergoing apoptosis. Blood 84: 1415–1420, 1994

17. Hanahan D: Transgenic mice as probes into complex sys-tems. Science 246: 1265–1275, 1989

18. Pattengale PK, Stewart TA, Leder A, Sinn E, Muller W,Tepler I, Schmidt E, Leder P: Animal models of human dis-ease. Pathology and molecular biology of spontaneous neo-plasms occurring in transgenic mice carrying and expressingactivated cellular oncogenes. Am J Pathol 135: 39–61, 1989

19. Cardiff RD, Sinn E, Muller W, Leder P: Transgenic onco-gene mice. Am J Pathol 39: 495–501, 1989

20. Lowe SW, Bodis S, McClatchey A, Remington L, Ruley HE,Fisher DE, Housman DE, Jacks T: P53 status and the effica-cy of cancer therapy in vivo. Science 266: 807–810, 1994

21. Aas T, Borresen AL, Geisler S, Smith-Sorensen B, JohnsenH, Varhaug JE, Akslen LA, Lonning PE: Specific p53 muta-tions are associated with de novo resistance to doxorubicinin breast cancer patients. Nature Med 2: 811–814, 1996

22. Hashimoto H, Chatterjee S, Berger NA: Inhibition of etopo-side (V-16)-induced DNA recombination and mutant fre-quency of Bcl-2 protein overexpression. Cancer Res 55:4029–4035, 1995

23. Wyllie AH, Kerr JFR, Currie AR: Cell death: the signifi-cance of apoptosis. Int Rev Cytol 68: 251–300, 1980

24. Bursch W, Liehr JG, Sirbarsku DA, Putz B, Taper H,Schulte-Hermann R: Control of cell death (apoptosis) bydiethylstilbestrol in an estrogen dependent kidney tumor.Carcinogenesis 12: 855–860, 1991

25. Muss HB, Thor AD, Berry DA, Kute T, Liu ET, Koerner F,Cirrincione CT, Budman DR, Wood WC, Barcos M: c-erbB-2 expression and response to adjuvant therapy inwomen with node-positive early breast cancer. N Engl JMed 330: 1260–1266, 1994

26. Rasbridge SA, Gillet CE, Seymour AM, Patel K, RichardsMA, Rubens RD, Millis RR: The effects of chemotherapyon morphology, cellular proliferation, apoptosis and onco-protein expression in primary breast carcinoma. Br J Cancer70: 335–341, 1994

27. McDonnell TJ, Meyn RE, Robertson LE: Implications ofapoptotic cell death regulation in cancer therapy. SeminCancer Biol 6: 53–60, 1995

28. Deshane J, Siegal GP, Alvarez RD, Wang MH, Feng M,Cabrera G, Liu T, Kay M, Curiel DT: Targeted tumor killingvia an intracellular antibody erbB-2. J Clin Invest 96: 2980–2989, 1995

29. Vaughn JP, Iglehart JD, Demirdji S, Davis P, Babiss LE,Caruthers MH, Marks JR: Antisense DNA downregulationof the ERBB2 oncogene measured by a flow cytometricassay. Proc Natl Acad Sci USA 92: 8338–8342, 1995