Defective interleukin six expression and responsiveness in human mammary cells transformed by an adeno 5/SV40 hybrid virus.

British Journal of Cancer (1996) 73, 1356-1361
rt (B) 1996 Stockton Press All rights reserved 0007-0920/96 $12.00

Defective interleukin six expression and responsiveness in human mammary
cells transformed by an adeno 5/SV40 hybrid virus

F Basolo1, L Fiore', S Calvol, V Falcone2, PG Conaldi2, G Fontaninil, AM Caligol, G Merlo3,

Y Gluzman4 and A Toniolo2

'Institute of Pathological Anatomy, University of Pisa, Italy; 2Institute of Medicine and Public Health, University of Pavia, Varese,
Italy; 3Friedrich Miescher-Institut, Basle, Switzerland; 4Molecular Biology Section, Lederle Laboratories, Pearl River, NY, USA.

Summary Mammary epithelial cells (MECs) were isolated and cultured from mammary glands of healthy
women undergoing reduction mammoplasty. Normal MECs were infected with the transforming hybrid virus
adeno-5/SV40. Two transformed epithelial cell lines, Ml and M2, were obtained, characterised phenotypically
and studied for the production of and the response to cytokines and growth regulators. In both cell lines,
expression of the SV40 large T antigen was associated with loss of interleukin 6 (IL-6) production and
responsiveness as well as with down-regulation of IL-8 and transforming growth factor (TGF)-a production.
Both Ml and M2 cell lines were capable of forming colonies in semisolid media, but upon injection into severe
combined immunodeficient (SCID) mice only M2 cells were tumorigenic. DNA synthesis in Ml cells was
partially inhibited by serum or TNF-ac and weakly stimulated by hydrocortisone (HC) and IL-8. In contrast,
M2 cells were totally unresponsive to a variety of growth regulators. Both lines overexpressed the p53 protein
at levels about 20-fold higher than those observed in primary MEC cultures, but no mutations of the p53 gene
could be detected. The data confirm the view that the expression in human mammary cells of different
oncogenes - including the SV40 T antigen - is frequently associated with alterations of cytokine production
and responsiveness.

Keywords: mammary cells; SV40; interleukin 6; cytokines

This study was initiated to investigate in vitro the early
phenotypic alterations associated with the transforming
activity of simian virus 40 (SV40) in human mammary
epithelial cells (MECs) obtained from healthy donors. Several
human breast cancer cell lines have already been obtained in
different laboratories (O'Hire, 1991), but only a few human
MEC lines have been derived from normal breast tissue (Soule
et al., 1990; Paine et al., 1992). Owing to the extremely low
frequency of spontaneous immortalisation, MEC lines have
been established either by chemical treatment, by transfection
with different oncogenes or by infection with oncogenic viruses
(Calaf and Russo, 1993; Ciardiello et al., 1993; Ochieng et al.,
1991; Berthon et al., 1992). We used a replication-defective
recombinant adeno-5/SV40 virus (Ad5/SV40) that carries the
adenovirus type 5 capsid, the early genomic adenovirus region
lacking the origin of replication, and the SV40 large T antigen
(SV40 T Ag; Van Doren and Gluzman, 1984). Primary MEC
cultures obtained from healthy donors undergoing reduction
mammoplasty were infected with this transforming virus. The
SV40 T Ag, which forms stable complexes with tumour-
suppressor gene products p53 and Rb (Ludlow, 1993), appears
to produce alterations of cytokine pathways similar to those
observed in the neoplastic transformation of MECs by other
oncogenes (Basolo et al., 1993a,b). Of the two Ad5/SV40-
immortalised MEC lines that were characterized, only one was
tumorigenic when injected into immunodeficient mice. How-
ever, transformation of MECs by SV40 T Ag was consistently
associated with changes in interleukin 6 (IL-6) expression and
responsiveness.

Materials and methods

Primary cultures of mammary epithelial cells and SV40
infection

Normal breast tissue was obtained from healthy women
undergoing reductive mammoplasty. Primary MECs were

Correspondence: A Toniolo, University of Pavia, Institute of
Medicine and Public Health, Viale Borri, 57-21100 Varese, Italy

Received 26 July 1995; revised 11 December 1995; accepted 20
December 1995

prepared by mechanic and collagenase dissociation and
cultured in Dulbecco's modified Eagle medium (DMEM)/
F12 low-calcium medium (0.04 mM Ca2+) supplemented with
insulin (5 jug ml-'), hydrocortisone (HC; 500 ng ml-'), hu-
man recombinant epidermal growth factor (EGF;
10 ng ml-'), cholera toxin (100 ng ml-') and 5% horse
serum, according to Soule's method (Soule et al., 1990).
Tissue culture reagents were from Sigma, (St. Louis, MO,
USA). Pure MEC cultures were infected between 40 and 55
days post plating with a replication-incompetent ori- hybrid
Ad5/SV40 virus (R4 strain; Van Doren and Gluzman, 1984)
and grown in complete medium with standard calcium
concentration (1.05 mM Ca2+; high-calcium medium,
HCM). After 20- 25 days, colonies of transformed cells
were picked up with the help of glass cylinders and expanded.
MECs that were capable of growing over at least 15 passages
and that were capable of forming colonies in semisolid media
were considered transformed and further characterised. Two
MEC lines originated from different donors (designated Ml
and M2) were analysed in detail. Serum-free medium (SFM)
was used to evaluate the response of MECs to growth
regulators. SFM consisted of DMEM/F12 with 1.05 mM
Ca2 , 0.3% dialysed fetal bovine serum (FBS), 0.2% bovine
serum albumin (BSA), human-transferrin (5 ig ml-1), so-
dium-selenite (5 ng ml- ') and insulin (10 jug ml- '). Primary
MEC cultures, the spontaneously immortalised MCF-1OA
normal breast cell line (Soule et al., 1990) and the breast
carcinoma-derived T47D cell line were used as controls.

Immunocytochemistry and immunoradiometric assay

For immunocytochemical staining, cells were grown in
chamber slides (Nunc, Naperville, IL, USA) and fixed in
1:1 (v/v) acetone -methanol at -20?C. The following
antibodies were used: cytokeratin (K8, K14, K18 and K19;
Unipath, Basingstoke, UK), p53 protein (MAbs 1801, 240,
DOl, Oncogene Science, Manhasset, NY, USA) and SV40 T
Ag (MAb 419, Oncogene Science). Cultures were stained with
0.1 1 Mg of antibody in 100 M1 of phosphate-buffered saline
(PBS)-BSA. Reactivity was revealed by immunoperoxidase
staining with an avidin-biotin complex kit (ABC kit; Vector,
Burlinghame, CA, USA).

The expression of membrane antigens was measured
with an immunoradiometric assay in live cells using MAbs
to surface antigens as reported previously (Basolo et al.,
1992). The following MAbs were used: non-polymorphic
HLA-A,B,C and HLA-DR (Technogenetics, Trezzano S/N,
Italy), human milk fat globule Ag-I and -2 (HMFG-1, -2;
Oxoid Garbagnate Milanese, Italy); epithelial membrane
antigen (EMA; BioGenex, San Ramon, CA, USA);
transferrin receptor (CD7 1, Becton Dickinson, Milan,
Italy); EGF receptor (EGF-R, extracellular domain;
Oncogene Science).

Cell response to growth regulators

DNA synthesis was measured as incorporation of 5-
[1251000]-2'-Deoxyuridine ([1251]UDR Amersham, Milan,
Italy) by cells grown in SFM containing insulin
(10 Mg ml-1). Cultures in 24-well plates were pulsed for
3 h with [125I]UDR (2.5 x 105 c.p.m.) 1, 3 and 5 days post
plating and processed as reported previously (Basolo et al.,
1994). The results are expressed as net c.p.m. per 50 000
cells. The influence of different growth regulators on DNA
synthesis was studied: rh-IL-6 and rh-IL-8 (UBI, Lake
Placid, NY, USA), rh-TNF-a and rh-EGF (Sigma), rh-
TGF-f,l (R&D Systems, Minneapolis, MN, USA), HC
(Sigma), horse serum (Gibco, Grand Island, NY, USA).

Measurement of cytokine levels

Conditioned media from both primary cultures and Ad5/
SV40-infected cell lines were used to measure the release of
IL-1-ax and -/3, IL-6, IL-8 and TGF-a with immunoenzyme
assays [dosage kits were from: Genzyme, Boston, MA, USA
(IL-1-cx; sensitivity 5 pg ml-1; IL-I-fl, sensitivity 1 pg ml-1),
R&D Systems, Minneapolis, MN, USA (IL-6, sensitivity
5 pg ml-'; IL-8, sensitivity 30 pg ml-'), Oncogene Science
(TGF-a, sensitivity 25 pg ml-1)]. Samples were taken 3, 6 and
9 days after plating (reported data refer to day 6).

Analysis of cytokine transcripts

Cytokine-specific mRNAs were detected in total RNA
extracted from 106 cells by the guanidinium thiocyanate
method. After treatment with RNAase-free DNAase for 1 h
at 37?C, cDNA was obtained by using M-MLV reverse
transcriptase in conjunction with random hexamer primers
(Clontech, Palo Alto, CA, USA). cDNA was then amplified
by the polymerase chain reaction (PCR) using Taq
polymerase and cytokine-specific primer pairs (Cytokine
MAPPing Amplimers, Clontech). Amplification was carried
out for the following human transcripts: IL-6 (amplified
products 628 bp), IL-8 (amplified product 289 bp), TGF-a
(amplified product 297 bp) and ,B-actin (amplified product
1126 bp). ,B-Actin was used as a control for mRNA
detectability. Thirty amplification cycles were performed in
a Hybaid thermal reactor (Hybaid, Teddington, UK).
Amplification products were separated by electrophoresis on
2.5% agarose gels and visualised under UV light by staining
with 0.5 Mg ml -' ethidium bromide. (DX174/HaeIII digest was
used as size marker.

Detection of p53 protein by Western blotting and
immunoenzyme assay

Cell lysates were prepared with the following buffer: 150 mM
sodium chloride, 50 mM Tris (pH 8.0), 5 mM EDTA, 1%

Nonidet P-40, 1 mM phenylmethylsulphonyl fluoride,
20 Mg ml-1 aprotinin, and 25 Mg ml-' leupeptin (30 min at
4?C; chemicals from Sigma). Protein extracts were resolved
by sodium dodecyl sulphate (SDS)-polyacrylamide gel
electrophoresis (8% acrylamide, 0.5% bis-acrylamide) and
transferred to nitrocellulose paper by electroblotting (Bio-
Rad, Milan, Italy). Duplicate membranes were incubated for
1 h in a blotting buffer containing 5% Carnation dried milk,

Cytokines in mammary cells transformed by Ad5/SV40 virus
F Basolo et al

1357
followed by 15 h at 4?C in Tris-borate buffer containing
1 Mg ml-' of anti-p53 MAbs (MAbs 1801, DOl, and 240).
MAbs 1801 and DOI react preferentially with the C-terminal
portion of the wild-type protein, but also with mutant forms,
whereas MAb 240 recognises mainly the tertiary structure of
several mutant forms of p53. The reaction was revealed by
immunoperoxidase staining (Vector).

Levels of wild-type and mutant forms of the p53
protein were measured in culture supernatants and in cell
extracts prepared as above. Two ELISAs (Oncogene
Science) were used: pantropic p53 (MAb 1801; sensitivity
10 pg ml-') and mutant-selective p53 (MAb 240; sensitivity
250 pg ml-1). A polyclonal rabbit antibody to p53 and
peroxidase-conjugated goat anti-rabbit IgG allowed antigen
determination.

Analysis of p53 mutations

A human p53 amplimer panel (Clontech) allowed the PCR
amplification of exons 5, 6, 7, 8 and 9. Template DNA
(100 ng) was subjected to 34 amplification cycles in 10 ,l of
reaction mixture containing: 1 x PCR buffer (50 mm potas-
sium chloride, 10 mM Tris, 1.5 mm magnesium chloride),
200 4uM each dNTPs, 3-10 pmol of each primer, 2.5 U of
Taq polymerase and 0.5 ,ul of [32P]dCTP (3000 Ci mmol-1;
Amersham, Milan, Italy). Aliquots of 1 l, of each PCR
product were diluted with 4 pl of SSCP-loading buffer (95%
formamide, 20 mM EDTA, 0.05% bromophenol blue, 0.05%
xylene cyanole), denatured 5 min at 95?C, loaded onto 6%
polyacrylamide-bis-acrylamide gel (49:1), 5% glycerol,
0.5 x TBE and run for 3 h at 15 w. Bands were detected by
autoradiography on Kodak X-AR 5 film.

The entire exon 7 of the transformed Ml and M2 lines
and of a normal primary MEC culture was amplified by
PCR. Specific oligonucleotide primers were designed from the
flanking intronic sequences: upstream (1775) 5'-
CGCGCACTGGCCTCATCTT-3'; downstream (2253) 5'-
TCAGCGGCAAGCAGAGGCTG-3'. The BamHI and
Hindlll restriction sites were added to, respectively, the
upstream and the downstream primer to facilitate subcloning.
Thirty amplification cycles were performed and PCR
products were ethanol precipitated, resuspended in 10 mM
Tris pH 7.5, digested with BamHI and HindlIl enzymes,
purified by 1.5% agarose electrophoresis, eluted and ligated
into BamHI/HindIII-digested pBluescript SKII (Stratagene,
La Jolla, CA, USA). Recombinant clones were obtained and
checked by restriction analysis. Sequencing was performed on
5, g of purified plasmid DNA using the T3 primer
(Boehringer, Mannheim, Germany) with the Sequenase 2.0
system (US Biochemicals, Cleveland, OH, USA).

Anchorage-independent growth and Matrigel colonisation assay
For colony-formation in soft agar and methylcellulose,
2 x 104 cells were suspended in 0.5 ml of HCM supplemented
with either 0.3% Agar Noble (Difco, Detroit, MI, USA) or
0.8% carboxymethylcellulose (Sigma) and layered on a 0.8%
Agar Noble base (0.25 ml in 2 cm2 wells). Twenty-one days
post plating, colonies > 60 ,um in diameter were photo-
graphed and counted.

The Matrigel colonisation assay was performed as
described previously (Albini, 1989). Matrigel (0.5 ml at
10 mg ml-'; Collaborative Research, Bedford, MA, USA)
was polymerised at 370C for 1 h in wells of a 24-well plate.

Cells suspended in HCM were seeded at 5 x 104 onto the top
of the gel. After 1, 2 and 3 weeks incubation, cultures were
observed for morphology and photographed.

Tumorigenicity assay

Cells cultured in HCM were detached by trypsinisation and
washed twice in PBS- 1% BSA. Approximately 1 million and
10 million cells suspended in 0.2 ml of Matrigel were injected
subcutaneously into the left and right flanks of female severe

Cytokines in mammary cells transformed by Ad5/SV40 virus

F Basolo et al

combined immunodeficient (SCID) mice (IFFA-Credo,
Milan, Italy). The animals were maintained in laminar flow
cabinets and given sterile food and acidic water ad libitum.
The sites of injection were inspected weekly.

Results

Phenotypic characterisation of AdS/SV40-infected breast cell
lines

After Ad5/SV40 infection, MECs were maintained in HCM.
Uninfected cultures began to senesce after 2-4 passages. In
contrast, 20-30 days after infection, colonies appeared in
cultures exposed to Ad5/SV40. Colonies were expanded and
passaged serially. At the present time, both lines have been in
culture for over 2 years (i.e. over 300 population doublings).
On passages 10- 15 the cells appeared to stop growing, but
3 -5 weeks later focal growth resumed. Cells of the Ml and
M2 lines were larger than uninfected cells and tended to grow
to higher densities. Transformed lines formed domes similar
to those produced by the spontaneously immortalised MCF-
IOA MEC line (Soule et al., 1990). To see whether SV40
immortalisation could influence the expression of MEC
antigens, cells were reacted with selected MAbs to both
surface (immunoradiometric assay) and intracellular antigens
(immunostaining; data not shown). Transformed cells were
essentially indistinguishable from normal MECs (Basolo et
al., 1992). They expressed HLA class I, but not class II,
molecules, were positive for epithelial markers such as EMA
and HMFG-l/-2 as well as for EGF and transferrin
receptors. Both lines expressed cytokeratins 8, 14 and 18 in
the absence of cytokeratin 19, a finding in agreement with
previous results obtained in cultured MECs (Taylor-
Papadimitriou et al., 1989; Paine et al., 1992; Raux et al.,
1992). Ml and M2 lines showed strong nuclear reactivity for
the SV40 T Ag (over 95% positive cells).

Production of cytokines

Conditioned media from primary MEC cultures and from
transformed lines (MI, M2, and T47D) were tested for the
presence of IL--L/fl, IL-6, IL-8 and TGF-a (Table I). As
previously reported (Basolo et al., 1993a), no culture released
IL-1-ac or -,B, whereas primary MEC cultures produced high
levels of IL-6, IL-8, and TGF-oc. The Ml, M2 and T47D cell
lines were unable to release IL-6 and secreted reduced levels
of both IL-8 and TGF-oe. Analysis of mRNA transcripts
confirmed that IL-6 expression was abolished in transformed
cells (Figure 1).

TGF- a

IL-6 IL-8

1353

1078
872
603

310
281

1 2 3 4 1 2 3 4 1 2 3 4

Figure 1 PCR amplification of total cellular RNA obtained from
the SV40-transformed MI and M2 cell lines and from a control
culture of normal mammary epithelial cells. Primers specific for
TGF-a, IL-6 and IL-8 were used. Ethidium bromide-stained
agarose gel. Lane 1, Ml cells; lane 2, M2 cells; lane 3, primary
culture no. 25; lane 4, positive control. Molecular size markers are
shown on the left.

Table I Levels of cytokines released in the medium by normal mammary epithelial
cells, SV40-transformed MI and M2 cell lines and tumour-derived T47D cell line

(pg ml )a

Cytokine Primary culturesb Ml M2 T47D
IL-l-oc - - - -
IL-1,B - - - -
IL-6 2280+ 320 - - -

IL-8 2150+240 45+28 50+42 64+40
TGF-oc 158+35 56+ 18 68+32 47+28

aCytokine concentrations were measured by enzyme-linked immunoassay (mean + s.d.
of 3 -5 cultures2. Conditioned media were collected 6 days after plating. -, below the
detection limit. Values accumulated from cultures of three different donors.

Table II Incorporation of [1251]UDR in the MCF-lOA control cell line in SV40-

transformed MI and M2 cell lines stimulated with different growth regulatorsa
Additives MCF-JOA Ml M2

None 991 +82 2937+ 31 8459+743
HC (100ngml- 1) 2699 + 134* 4892+415* 8909+ 1141
EGF (lOngml-') 5946+ 187* 3270+245 9494+ 1512
Horse serum (5%) 4668 + 415* 786 + 36* 9957+2333
IL-6 (lOngml- 1) 1825+124* 3160+96 8972+648
IL-8 (lOngml- ) 1110+56 4415+50* 9637+937
TNF-a (IOngml- 1) 571 +52* 2060+63* 8661 +863
TGF-fl (5 ng ml 2) 42+38* 2909+223 10 512+ 1304

aCultures in serum-free medium containing insulin (10 pg ml 1) were stimulated with
additives for 3 days and pulsed with ['251]UDR for 3 h. Values are reported as c.p.m./
50000 cells; mean + s.d. of 3- 5 cultures. Asterisks refer to values that are significantly
different from those obtained in control cultures given no additives (P<0.001).

Proliferative cell responses to growth regulators

Primary MECs are dependent on insulin, HC, EGF and
serum supplementation for optimal differentiation and
anchorage-dependent growth (Soule et al., 1990; Basolo et
al., 1992). To evaluate the effects of the SV40 T Ag on the
response to growth factors and cytokines, the rate of DNA
synthesis was examined in cultures grown with SFM. Table II

shows that, in the absence of additives, ['251]UDR incorpora-

tion was significantly higher in Ml and M2 cell lines than in
the MCF-IOA line used as control. Tests were performed on
days 1, 3 and 5 post plating. Data in Table II refer to day 3,
but equivalent results were obtained on day 5. HC stimulated
DNA synthesis in both MCF-IOA and Ml cells, but not in
M2 cells. Responsiveness to EGF was lost in SV40-
immortalised cells. Horse serum stimulated ['25I]UDR
incorporation in MCF-IOA cells, but depressed it in Ml

cells, while not influencing M2 cells. IL-6 had a stimulatory
effect on MCF-1OA cells, while SV40-transformed lines were
non-responsive. IL-8 and TNF-a had little effect on either cell
line. TGF-j1 inhibited only MCF-IOA cells.

Anchorage-independent growth and differentiation in Matrigel
Ad5/SV40-infected cells were tested for the ability to form

colonies in semisolid media. When 2 x 104 cells were plated,

the Ml and M2 lines were capable of forming large colonies
both in soft agar [48+8 and 35+6 (mean+s.d.) respectively]
and in methylcellulose (71 + 12 and 58+9 respectively).
Uninfected primary cells failed to grow under these
conditions, while the tumour-derived T47D MEC line used
as positive control formed over 100 colonies in both media.

Since it is known that normal MECs obtained from milk
are able to form lobulo/alveolar-type structures in Matrigel
(Bartek et al., 1991; Petersen et al., 1992), Ml and M2 cell

1 2 1 2 1 2

94 kDa
53 kDa

A B C

Figure 2 Western blotting of cell extracts of Ml and M2 cell
lines. Staining of Ml cells (lane 1) and M2 cells (lane 2) with:
MAb 419 to SV40 T large antigen (A); MAb 1801 to wild-type
forms of p53 protein (B); MAb 240 to mutant forms of p53 (C).

Cytokines in mammary cells transformed by Ad5/SV40 virus

F Basolo et al M

1359
lines were tested for their ability to differentiate in Matrigel.
Three weeks after plating, the spontaneously immortalised
MCF-IOA cell line formed only small colonies in Matrigel
(i.e. <50 gim in diameter; Figure 2a). In contrast, Ml cells
were able to differentiate into duct-like structures, whereas
M2 cells formed large colonies (Figure 2b and 2c).

p53 protein expression

It is known that the p53 protein is frequently mutated in
breast tumours and that it is stabilised by the SV40 T Ag. To
investigate the possible role of p53 alterations in producing
the transformed phenotype of AdS/SV40-infected cells, we
studied the expression of p53 in MI and M2 using three
different MAbs. MAbs 1801 and DOI (reacting with both
wild-type and mutant forms of p53) stained over 80% of
nuclei of Ml and M2 cells, but failed to stain normal cells
(primary cultures and MCF-1OA cells). MAb 240 (reacting
mainly with mutant forms of p53) stained only Ml and M2
cells, but not normal cells (data not shown).

The above data were confirmed by Western blot assay
under denaturing conditions (Figure 2). MAb 1801 produced
a 53 kDa band in both lines, while MAb 240 produced a
53 kDa band in M2, but not in Ml cells. A 94 kDa band
corresponding to the SV40 T Ag was found in extracts of
both Ad5/SV40-infected lines by staining with MAb 419.

Enzyme-linked immunoassays capable of detecting either
the wild-type p53 plus its mutant forms or the mutant forms
alone were used to measure the levels of p53 produced by Ml
and M2 cells (Table III). Primary cultures of normal MECs
and the T47D mammary carcinoma cell line carrying a
mutant form of p53 were used as controls. Measurable levels
of wild-type p53 were detected both in cell extracts and in the
supernatant of all test cultures, while 'mutant' forms of p53
were detected only in cultures of transformed lines.

Taken together, the above data indicated that changes in
p53 expression had occurred in SV40-transformed lines.

Analysis of p53 mutations

To understand if the altered forms of p53 found in AdS/
SV40-infected cells were due to sequence mutations or to
conformational changes, hotspot p53 exons 5-9 were
analysed by PCR-single-strand conformation polymorphism
(SSCP). As expected, no abnormal migration patterns of
bands relative to exons 5, 6, 8 and 9 were detected in the Ml
and M2 cells as compared with cultures of normal MECs.
Owing to minor alterations seen in the migration pattern of
the exon 7 amplification product of M1 and M2 cells (data
not shown), this exon was sequenced. Analysis of several
individual clones of M l, M2 and normal primary MEC
cultures failed to show point mutations (data not shown). We
concluded that no p53 mutations had occurred in the SV40-
transformed cell lines.

Table III Levels of wild-type and 'mutant' forms of p53 in conditioned media and cell lysates
of mammary epithelial cells as measured by ELISA under non-denaturing conditions

(pg mI 7)a

MAb 1801 (wild-type p53)b MAb 240 (mutant p53)b

Culture Conditioned medium Cell lysate Conditioned medium Cell lysate
Primaryc 30+ 3 840+72 - -

M1-SV40 161+12 23 500+ 1980 1120+95 19000+1266
M2-SV40 30+2 18 100+987 - 44000+3514
T47D 170+30 18 000+856 1670+143 13 000+1137

aResults are expressed as pg ml 1 (mean + s.d. of 3 - 5 cultures); -, < 250 pg ml - 1. Cultures
were prepared in T-25 flasks containing 5 ml of complete medium. Five days after seeding,
supernatants were collected and clarified by centrifugation. Monolayers were washed extensively
and lysates were prepared with Nonidet-P40. An aliquot of 1 ml of cell lysate was obtained from
2.5 x 106 cells. bMAb-.1801 reacts mainly with wild-type p53, MAb 240 reacts mainly with mutant
forms of p53. cValues accumulated from cultures of three different donors.

a

b

Cytokines in mammary cells transformed by Ad5/SV40 virus

F Basolo et al

1360

Tumorigenicity assay

Ml and M2 cells were injected subcutaneously into SCID
female mice to evaluate their tumorigenic potential. Thirty
days after inoculation, tumours developed in only three out
of ten mice injected with 107 M2 cells. No tumours were
detected in Ml cell-injected mice at 60 days post inoculation.
Tumours derived from the M2 cell line were classified as
adenocarcinomas, were positive for the SV40 T Ag and
showed p53 overexpression (data not shown).

Discussion

This study shows that MEC derived from healthy donors can
be infected by a defective adenovirus-5 that carries the major
SV40 oncogene. The virus can transform mammary cells that
may also be tumorigenic. Transformation of epithelial cells
by the SV40 T Ag has been tentatively ascribed to the co-
operative activity of virus and random genetic changes.
However, in human bronchial, oesophageal and hepatic
epithelial cells immortalisation by SV40 has been reported
in the absence of p53 mutations (Lehman et al., 1993).

Since p53 mutation is one of the most frequent alterations
in breast cancer (Eriksson et al., 1994), we analysed whether
the tumorigenicity of M2 cells was associated with p53
mutations. Molecular analysis of hotspot p53 exons failed to
detect mutations in SV40-transformed lines, suggesting that
SV40 infection can be directly responsible for the develop-
ment of the tumorigenic phenotype in mammary cells
(Berthon et al., 1992; Yilmaz et al., 1993).

In spite of the absence of detectable mutations,
transformed lines overexpressed wild-type and/or 'mutant'
forms of p53 at levels about 20-fold higher than normal
MECs. The following observations may help explain the
data: (a) the p53 protein has a longer half-life in MEC than
in human fibroblasts (3 h vs 30 min; Delmolino et al., 1993);
(b) p53 is complexed by the SV40 T Ag and accumulates in
infected cells (Bartek et al., 1993; Chang et al., 1993). The
latter mechanism may account for the high levels of 'mutant'
p53 found by immunological methods. In fact, complexing of
p53 with SV40 T Ag may produce conformational changes of
the target that could make it reactive with the antibody
'specific' for mutant forms. ELISA showed that tumorigenic
and non-tumorigenic cells produced equivalent amounts of
'mutant' p53, but the tumorigenic M2 cell line did not release
the protein extracellularly. In agreement with that, staining of
M2 cells with antibodies to either SV40 T Ag or p53
produced patterns more intense than those seen in Ml cells.

Normal MECs have the capacity of producing several
different cytokines. Consistent with what was observed in the
case of MECs transformed by various oncogenes (c-Ha-ras,
c-erb-B2, int-2; Basolo et al., 1993b), the secretion of some
cytokines (IL-8, TGF-a) was reduced, while that of IL-6 was
abolished. Moreover, both SV40-transformed MEC lines
were unable to respond to both IL-6 and TGF-/.
Tumorigenic M2 cells were also totally unresponsive to a
variety of growth regulators (HC, EGF, horse serum, IL-8,
TNF-a). Non-tumorigenic Ml cells had an intermediate
phenotype and formed duct-like structures when grown in
Matrigel, i.e. retained some degree of differentiation (Shearer
et al., 1992; Petersen et al., 1992). In contrast, M2 cells
produced mainly large colonies and induced tumours in
SCID mice. Unresponsiveness to TGF-,B - an important
regulator of normal and neoplastic cells (Derynk, 1994;
Emerman and Eaves, 1994) - has been observed in other
experimental models (Reiss et al., 1993; Basolo et al., 1994;
Herman and Katzenellenbogen, 1994).

c

Figure 3 Growth in Matrigel of normal and SV40-transformed
mammary epithelial cell lines 21 days after plating. Small colonies
of normal MCF-IOA cells (a); colonies and duct-like structures of
Ml cells (b); large colonies of M2 cells (c). Dark-field microscopy;
original magnification 33 x.

Taken together, the data confirm the hypothesis that
alterations of IL-6 and TGF-3 pathways represent an early
and common event in mammary carcinogenesis (Basolo et al.,
1993a, b and 1994).

Our results appear to confirm that the expression of the
SV40 T Ag gene in breast epithelial cells may directly cause

Cy tok- ai manuny cds trumssormed by Ad5/SV40 viru
F Basolo et ai

1361

the appearance of a malignant phenotype (Berthon et al.,
1992). The phenotypic differences between non-tumorigenic
Ml and tumorigenic M2 cells are, however, unclear and not
due to detectable p53 mutations. How transformation by
SV40 is linked to the above events and, particularly, to
alterations of IL-6 production and responsiveness remains to
be elucidated.

References

ALBINI A. AUKERMAN SL. OGLE RC. NOONAN DM. FRIDMAN R.

MARTIN GR AND FIDLER IJ. (1989). The in vitro invasiveness
and interaction with laminin of K-1735 melanoma cells. Evidence
for different laminin-binding affinities in high and low metastatic
variants. Clin. Exp. Metastasis, 7, 437-451.

BARTEK J, BARTKOVA J, KYPRIANOU N, LALANI EN, STASKOVA

Z, SHEARER M, CHANG S AND TAYLOR-PAPADIMITRIOU J.
(1991). Efficient immortalization of luminal epithelial cells from
human mammary gland by introduction of simian virus 40 large
tumor antigen with a recombinant retrovirus. Proc. Natl Acad.
Sci. USA, 88, 3520-3524.

BARTEK J, VOJTESEK B AND LANE DP. (1993). Diversity of human

p53 mutants revealed by complex formation to SV40 T antigen.
Eur. J. Cancer, 29A, 101 - 107.

BASOLO F, SERRA C, CIARDIELLO F, FIORE L, RUSSO J, CAMPANI

D, DOLEI A. SQUARTINI F AND TONIOLO A. (1992). Regulation
of surface-differentiation molecules by epidermal growth factor,
transforming growth factor alpha, and hydrocortisone in human
mammary epithelial cells transformed by an activated c-Ha-ras
proto-oncogene. Int. J. Cancer, 51, 634- 640.

BASOLO F, CONALDI PG, FIORE L, CALVO S AND TONIOLO A.

(1993a). Normal breast epithelial cells produce interleukin 6 and 8
together with tumor-necrosis factor: defective IL6 expression in
mammary carcinoma. Int. J. Cancer, 55, 1 - 5.

BASOLO F, CALVO S, FIORE L, CONALDI PG, FALCONE V AND

TONIOLO A. (1993b). Growth-stimulating activity of interleukin 6
(IL6) on human mammary epithelial cells transfected with the int-
2 gene. Cancer Res., 53, 2957-2960.

BASOLO F, FIORE L, CIARDIELLO F, CALVO S, FONTANINI G,

CONALDI PG AND TONIOLO A. (1994). Response of normal and
oncogene-transformed human mammary epithelial cells to
transforming growth factor f (TGF-fl): lack of growth-inibitory
effect on cells expressing the simian virus 40 large-T antigen. Int.
J. Cancer, 56, 736- 742.

BERTHON P, GOUBIN G, DUTRILLAUX B, DEGEORGE A, FAILLE A,

GESPACH C AND CALVO F. (1992). Single-step transformation of
human breast epithelial cells by SV40 large T oncogene. Int. J.
Cancer, 52, 92-97.

CALAF G AND RUSSO J. (1993). Transformation of human breast

epithelial cells by chemical carcinogens. Carcinogenesis, 14, 483-
492.

CHANG F, SYRJANEN S, TERVAHAUTA A AND SYRJANEN K.

(1993). Tumorigenesis associated with the p53 tumour suppressor
gene. Br. J. Cancer, 68, 653-661.

CIARDIELLO F, GOTTARDIS M, BASOLO F, PEPE S, NORMANNO N,

DICKSON RB, BIANCO AR AND SALOMON DS. (1993). Additive
effects of c-erB-2, c-Ha-ras, and transforming growth factor-2
genes on in vitro transformation of human mammary epithelial
cells. Mol. Carcinogenesis, 6, 43 - 52.

DELMOLINO L, BAND H AND BAND V. (1993). Expression and

stability of p53 protein in normal human mammary epithelial
cells. Carcinogenesis, 14, 827-832.

DERYNK R. (1994). Transforming growth factor-beta. In The

Cytokine Handbook, Thomson A (ed.), pp. 319-342. Academic
Press: London.

EMERMAN JT AND EAVES CJ. (1994). Lack of effect of

hematopoietic growth factors on human breast epithelial cell
growth in serum-free primary culture. Bone-Marrow Transplant,
13, 285-291.

ERIKSSON ET, SHIMMELPENNING H, ASPENBLAD U, ZETTER-

BERG A AND AUER GU. (1994). Immunohistochemical excpres-
sion of the mutant p53 protein and nuclear DNA content during
the transition from benign to malignant breast disease. Hum.
Pathol., 25, 1228 -1233.

This work was supported by grants from AIRC (Milan). ISS-AIDS
(Rome, No. 9206-25) and CNR, PF-FATMA (Rome), Italy.

HERMAN ME AND KATZENELLENBOGEN BS. (1994). Alterations

in transforming growth factor-alpha and -beta production and
cell responsiveness during progression of MCF-7 human breast
cancer cells to estrogen-autonomous growth. Cancer Res.. 54,
5867- 5874.

LEHMAN TA, MODALI R, BOUKAMP P. STANEK J. BENNET WP.

WELSH JA, METCALF RA, STAMPFER MR. FUSENIG N. ROGAN
EM AND HARRIS CC. (1993). p53 mutations in human
immortalized epithelial cell lines. Carcinogenesis, 14, 833 - 839.

LUDLOW JW. (1993). Interaction between SV40 large-tumor antigen

and the growth suppressor proteins pRB and p53. FASEB J., 7,
866-871.

OCHIENG J. BASOLO F. ALBINI A. MELCHIOR] A. WATANABE H.

ELLIOTT J, RAZ A. PARODI S AND RUSSO J. (1991). Increased
invasive, chemotactic and locomotive abilities of c-Ha-ras-
transformed human breast epithelial cells. Invasion Metast., 11,
38-47.

O'HIRE MJ. (1991). Breast cancer. In Human Cancer in Primary

Culture, Masters JRW (ed.). pp. 271-286. Kluwer Academic
Publishers: Amsterdam.

PAINE TM, SOULE HD, PAULEY RJ AND DAWSON PJ. (1992).

Characterization of epithelial phenotypes in mortal and immortal
human breast cells. Int. J. Cancer, 50, 463-473.

PETERSEN OW, RONNOV-JESSEN L HOWLETT AR AND BISSEL MJ.

(1992). Interaction with basement membrane serves to rapidly
distinguish growth and differentiation pattern of normal and
malignant human breast epithelial cells. Proc. Nat! Acad. Sci.
USA, 89, 9064 - 9068.

RAUX H, PLANCHON P, CHRAIBI-HAJJI F, MAGNIEN V, SALLE V

AND BROUTY-BOYE D. (1992). Immunophenotype of SV40-T
gene transfected epithelial cell lines derived from human benign
breast tumors. In Vitro Cell. Dev. Biol., 28, 468-470.

REISS M, VELLUCCI VF AND ZHOU Z-L. (1993). Mutant p53 tumor

suppressor gene causes resistance to transforming growth factor
PIl in murine keratinocytes. Cancer Res., 53, 899-904.

SHEARER M, BARTKOVA J, BARTEK J, BERDICHEVSKY F, BARNES

D, MILLIS R AND TAYLOR-PAPADIMITRIOU J. (1992). Studies of
clonal cell lines developed from primary breast cancer indicate
that ability to undergo morphogenesis in vitro is lost early in
malignancy. Int. J. Cancer, 51, 602-612.

SOULE H, MOLONEY T, WOLMAN SR, PETERSON WD, BRENZ R.

MAC GRATH CM, RUSSO J, PAULEY RJ, JONES RF AND BROOKS
SC. (1990). Isolation and characterization of a spontaneously
immortalized human breast epithelial cell line MCF-IOA. Cancer
Res., 50, 6075 - 6086.

TAYLOR-PAPADIMITRIOU J, STAMPFER M, BARTEK J, LEWIS A,

BOSHELL M, LANE EB AND LEIGH IM. (1989). Keratin
expression in human mammary epithalial cells cultured from
normal and malignant tissue: relation to in vivo phenotypes and
influence of medium. J. Cell Sci., 94, 403-413.

VAN DOREN K AND GLUZMAN Y. (1984). Efficient transformation

of human fibroblasts by adenovirus-simian virus 40 recombi-
nants. Mol. Cell. Biol., 4, 1653-1656.

YILMAZ A, GAIDE C, SORDAT B, BOBERNYI Z, LAHM H, IMAN A.

SCHREYER M AND ODARTCHENKO N. (1993). Malignant
progression of SV40-immortalised human milk epithelial cells.
Br. J. Cancer. 68, 868-873.