[CANCER RESEARCH 51, 4028-4037, August 1. 1991]
Isolation of Two Distinct Epithelial Cell Lines from a Single Feline Mammary
Carcinoma with Different Tumorigenic Potential in Nude Mice and Expressing
Different Levels of Epidermal Growth Factor Receptors1
Jules M. H. M. Minke, Ed Schuuring, Raymond van den Berghe, John A. M. Stolwijk, Johannes Boonstra,
Cees Cornelisse, John Hilkens, and Wim Misdorp2
Departments of Veterinary Pathology fJ. M. H. M. M., J. A. M. S., W. M.J and Molecular Cell Biology [J. B.J, State University Utrecht, Yalelaan l, P.O.B. 80158,
3508 TD, Utrecht; Division of Tumor Biology, Netherlands Cancer institute (Antoni van Leeuwenhoek Huis), Pleamanlaan 121, 1066 CX, Amsterdam fJ. M. H. M. M.,
E. S., R. v. d. B., J. H., W. M.]; and Department of Pathology [C. C.¡,University Leiden, Leiden, The Netherlands
overexpression
ABSTRACT
From a single spontaneous feline mammary carcinoma, two subpopu
lations of epithelial tumor cells have been isolated. The variant cells were
established as cell lines designated K248C and K248P. DNA ploidy
analysis showed that the two cell lines represented cell populations
already present in the original tumor. Chromosome analysis confirmed
the feline origin of K248C and K248P and demonstrated that in addition
to unique marker chromosomes characteristic for each cell line, both cell
lines had several marker chromosomes in common. These data suggest
that the two cell populations arose from a hypothetical single ancestor
which diverged during tumor progression. The K248C and K248P cell
lines differed from one another with respect to their tumorigenicity in
athymic mice and epidermal growth factor (EGF) receptor content. The
K248C cells were highly tumorigenic as indicated by a short latency
period and high take rate. The K248P cells were poorly tumorigenic.
Southern blot analysis revealed that the K248C cells contained an am
plified EGF receptor gene that was accompanied by elevated levels of
EGF receptor RNA and protein. The K248C cells were growth inhibited
in vitro at EGF concentrations that stimulated growth of K248P cells.
The amplification of the EGF receptor gene could be detected only in
DNA derived from K248C cells at high passage numbers and not in DNA
derived from the original tumor and K248C cells at low passage numbers.
These data suggest that amplification of the EGF receptor gene occurred
during establishment of the K248C cell line.
of the EGF3 receptor gene has been demon
strated (6). Amplification of the EGF receptor gene, however,
is infrequently found in human breast cancer (5), suggesting
that other yet unknown molecular mechanisms are responsible
for the observed overexpression. Despite the close similarity of
feline mammary carcinoma to human breast cancer, no infor
mation is available on the nature of the genetic changes involved
in the development and progression of feline mammary
carcinoma.
Studies on feline mammary carcinomas have been impeded
by the lack of suitable in vitro model systems. Therefore, the
availability of several well-characterized cell lines would facili
tate the understanding of the biology and heterogeneity of this
disease. Thus far, the establishment of only one feline mammary
carcinoma cell line has been reported (7).
This study reports the successful isolation and subsequent
characterization of two distinct cell types derived from a single
feline mammary carcinoma. We describe the properties of the
individual cell types differing from one another with respect to
their epithelial characteristics, tumorigenic potential in athymic
mice, karyotype, DNA index, and activation of the EGF recep
tor gene.
MATERIALS AND METHODS
INTRODUCTION
Studies on tumors in domestic animals can contribute to a
better understanding of similar diseases in humans. Several
reports (1-4) have focused on feline mammary carcinoma as a
potential model for human breast cancer. Comparative studies
have shown that feline mammary carcinoma shares important
features with human breast cancer, including similar biological
behavior and histológica! appearance (1, 2). The tumors show
local infiltrative-destructive growth and metastasize to regional
lymph nodes, lungs, and pleura at an early stage of the disease.
Histologically, feline mammary carcinomas are frequently char
acterized by extensive heterogeneity among tumor cells within
one tumor. In a recent study (2), 32% of the feline mammary
carcinomas were classified as compound carcinomas, consisting
of equal parts of different structures.
Analysis of genetic changes in cancer during the last decade
has revealed a group of genes involved in cell growth and
differentiation. These genes, when activated, are believed to
play an important role in the development and progression of
cancer. In human breast cancer, for example, amplification of
the myc and neu oncogenes frequently occurs (5). In addition,
Received 11/19/90; accepted 5/24/91.
The costs of publication of this article were defrayed in part 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.
1E. S. was supported by Dutch Cancer Society Grant NK87-12.
2To whom requests for reprints should be addressed.
Clinical Data on the K248 Donor. The tumor cells were collected at
autopsy from a pulmonary metastasis of a mammary carcinoma in a
14-year-old spayed Siamese cat (K248). which had undergone radical
mastectomy 3 months earlier for infiltrating adenocarcinoma of the
mammary gland. No chemo- or radiotherapy had been given. Autopsy
disclosed a widespread disease, with métastasesto the regional lymph
nodes, presternal lymph node, lungs, pleura, and kidney. The tumor
was classified according to the WHO classification for tumors in
domestic animals (8). The classification was extended by the introduc
tion of a distinct cribriform tumor type. Histologically, the tumor was
classified as a compound adenocarcinoma with areas of papillary,
cribriform, and solid tumor growth.
Establishment of the K248 Cell Lines. After autopsy the pulmonary
metastasis of cat 248 was placed in ice-cold tissue culture medium and
within l h was processed for tissue culture. Part of the tumor was
removed for histopathological diagnosis, and the remainder of the
tumor was debrided of necrosis and minced into 1- to 2-mm3 pieces.
The minced tumor was disaggregated overnight in 0.9% collagenase
(CLS III; Cooper Biochemicals) in DMEM (Gibco, Grand Island, NY)
at room temperature with constant stirring. The resulting cell suspen
sion was cultured in DMEM supplemented with 10% PCS (Gibco), 10
Mg/ml insulin (Sigma, St. Louis, MO), 100 lU/ml penicillin, and 1 mg/
3The abbreviations used are: EGF. epidermal growth factor; DMEM, Dulbecco's modified Eagle's medium; PCS, fetal calf serum; PBS, phosphate-buffered
saline; DF medium, DMEM/Ham F-12 medium; MAb, monoclonal antibody;
SDS, sodium dodecyl sulfate; SSC. standard saline citrate (NaCl 150 mM-Na3
citrate-2H2O-15 mm, pH 7.0); cDNA, complementary DNA; DI, DNA index;
UM, unidentified marker; HSR, homogeneously staining region; FeLV-gp70, M,
70,000 envelope glycoprotein of feline leukemia virus.
4028
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HETEROGENEITY
OF CELLS FROM A FELINE MAMMARY CARCINOMA
ml streptomycin (both from Seromed), and after washing in PBS it was
xenografted s.c. between the scapulae in two 8-12-week-old athymic
male B6Kh nude mice (0.15 ml centrifugad packed cell clumps in PBS).
Confluent monolayers were dissociated with 0.25% trypsin (Difco) and
0.02% EDTA in Ca2+- and Mg2+-free PBS.
In the primary culture of tumor K248 two different cell types were
identified. These variants were isolated by repeated differential trypsinization and two successive rounds of single cell cloning using 6.5-mm
cloning cylinders as described previously (9). All cells were screened
monthly for Mycoplasma contamination by Hoechst staining and were
always found to be negative.
Cloning Efficiency in Soft Agar. Dispersed cells were resuspended in
0.25% agar (Difco) in culture medium at concentrations of IO4 and 5
x IO4cells/ml and layered on top of 0.5% agar in culture medium. For
anchorage-independent growth assay in the presence of EGF cells were
plated identically, except that the medium added the next day and once
every week contained 50, 0.5, or 0.01 ng/ml EGF (culture grade;
Collaborative Research). Four weeks after initiating cultures, colonies
were counted.
Saturation Densities and Doubling Times. Logarithmically growing
cell cultures were trypsinized. and 5 x IO4cells were plated in triplicate
in 12-well plates (Costar) and cultured in DMEM supplemented with
10% FCS. Cells were counted every second day by hemacytometer. The
doubling time and saturation densities were calculated from a plot of
cell numbers against time.
Growth in Serum-free Medium. Trypsinized cells were washed once
in a 1:1 medium mixture of DF (Gibco) with 10% FCS to inactivate
trypsin. Subsequently, the cells were washed three times with DF
medium, seeded in the same medium without FCS, and supplemented
with 10 Mg/ml insulin and 10 ^g/ml transferrin (Sigma). When subcultured, 0.1% (w/v) soybean trypsin inhibitor (Sigma) in DF medium
instead of 10% FCS-supplemented medium was used to inactivate
trypsin.
Effect of EGF on Cell Proliferation. K248C, K248P, and A431 cells
were seeded on 35-mm plates (Falcon) at a density of IO4cells/plates
in DF medium containing 10% FCS. After attachment overnight,
cultures were washed three times with PBS, and the medium was
changed to DF medium containing 10 /jg/ml insulin and 10 Mg/ml
transferrin. After overnight incubation, the growth assay was performed
in the presence of various concentrations of EGF (0.01-50 ng/ml) and
in the absence of EGF. Cells were incubated for 10 days (K248C and
K248P) or 6 days (A431), with medium changes every 48 h. The cells
were then trypsinized and counted by hemocytometer. Data shown are
the average of determinations from triplicate plates. Results were
expressed as the percentage change in cell number for cells grown in
EGF-supplemented media compared with cells grown in control media.
Other Cell Lines and Monoclonal Antibodies. A431, a human squamous carcinoma cell line with an amplified and rearranged EGF recep
tor gene (10), was obtained from the American Type Culture Collection.
FEA, a feline fibroblast line, was the gift of Dr. O. Jarre« (11). Both
cell lines were grown in DMEM plus 10% FCS. The feline lymphoblast
cell lines, F422 and FL74, were the gift of Dr. K. Weijer and were
grown in RPMI 1640 (Gibco) supplemented with 10% FCS. Mouse
MAb 425 was the generous gift of Dr. U. Rodeck (Wistar Institute,
Philadelphia) and was directed against the protein backbone of the
extracellular domain of the human EGF receptor (12). Mouse MAb
139H2 was raised against human episialin (13). Mouse MAb E3C4
reacts with keratin 8 and 18 in both humans and cats.4 Mouse MAb
3.17 was raised against FeLV-gp70 (14).
Electron Microscopy. Cells were grown in culture flasks and fixed in
situ with 2.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2, for 1
h and postfixed with 1% osmium tetroxide in the same buffer for 2 h.
After fixation the cells were scraped off the flasks using a rubber
policeman and centrifuged. The pellet was dehydrated and embedded
in Epon-Araldite. Thin sections were stained with uranyl acetate and
lead hydroxide. The preparations were examined with a Philips CM 10
electron microscope.
4 D. Kami, personal communication.
Karyotype Analysis. Colchicine (l ¿jg/ml)was added to logarithmi
cally growing cell cultures, followed by an incubation for 20 min at
37°C.The cells were harvested with trypsin and incubated for 12 min
at 37°Cin a hypotonie solution containing 0.075 M KC1. After centrifugation, the pellet was fixed in three changes of fresh methanohglacial
acetic acid (3:1, v/v). The fixed cells were placed on 6 M HCl-treated
slides and air dried.
Chromosome spreads were "aged" at 60°Cfor 16 h and subsequently
either conventionally stained in Giemsa for chromosome counts or Gbanded. For each culture a minimum of 100 spreads was evaluated for
modal chromosome number, and at least 7 G-banded karyotypes were
analyzed for the detection of chromosomal aberrations. Karyotypes of
each cell line were arranged according to the recommendations of the
first international conference on standardization of banded karyotypes
of domestic animals (15).
Ploidy Analysis. Multiple samples derived from frozen and 40-^m
paraffin-embedded sections of the original K248 pulmonary metastasis
were used for flow cytometric DNA analysis according to the method
previously described (16). Briefly, frozen sections were thawed at room
temperature, cut into small pieces, treated with detergent-trypsin, and
stained with propidium iodide (Sigma). Nuclear suspensions were ana
lyzed with chicken RBC and normal feline WBC as internal standards.
From deparaffinized sections, cell suspensions were prepared by treat
ment with a solution of pepsin (no. P7000; Sigma) and stained with
4',6-diamidino-2-phenylindole
dihydrochloride (Boehringer, Mann
heim, Germany). Stained samples were measured on an 1CP 22 flow
cytometer (Ortho. Westwood, MA).
Suspensions of cultured cells in 40 HIM citrate buffer, pH 7.6,
containing 250 mM sucrose and 5% (v/v) dimethyl sulfoxide, were
frozen in liquid nitrogen and analyzed as described for nuclei derived
from frozen sections.
Tumorigenicity. Eight- to 12-week-old male B6Kh nu/nu mice were
used as recipients for tumor cell injections. Cell cultures at passage 60
were trypsinized and washed 3 times in PBS, and IO7 cells suspended
in 0.1 ml PBS were injected s.c. between the shoulder blades. For i.v.
injections, mice were anesthetized with ether, and 8 x IO5cells in 0.1
ml PBS were injected in the tail vein with a 25-gauge needle.
All animals were sacrificed by cervical dislocation when the tumors
reached a diameter of 2 cm or after 4 months in the absence of palpable
tumors. Tumors and selected organs (axillary lymph nodes, lungs, liver,
and kidneys) were removed, fixed in phosphate-buffered 10% formalin,
and processed for histopathological examination.
Immunofluorescence. For detection of keratins, cells were grown on
glass coverslips and fixed in cold acetone for 4 min. For FeLV-gp70
analysis, a suspension of IO6cells/ml was incubated on multitest slides
(Flow, Irving, Scotland) for 1 h at 37°C.The reaction of monoclonal
antibodies in the appropriate dilutions with the two cell lines was tested
in an indirect immunofluorescence assay using fluorescein-conjugated
goat anti-mouse immunoglobulin as the second antibody (Nordic Di
agnostic, Tilburg. The Netherlands). Normal mouse serum (diluted
1:100) served as a negative control. The cells were examined under a
fluorescence microscope.
DNA Isolation and Blotting. High-molecular-weight genomic DNA
was isolated by SDS-proteinase K lysis, phenol/chloroform extractions,
and ethanol precipitation as described (17). Fifteen /ig DNA were
digested with restriction enzymes (EcoR\, BamHl, Hindlll, and Xbal)
according to the manufacturer's recommendations (Boehringer), frac
tionated on 0.8% agarose gels in buffer (0.04 M Tris-acétate,pH 8.20.02 M sodium acetate-0.018 M NaCI-0.002 M EDTA), transferred onto
nitrocellulose filters (Schleicher and Schuell) according to the method
of Southern (18), and fixed at 80°Cfor 2 h.
RNA Isolation and Blotting. Total RNA was isolated in 3 M lithium
chloride/6 M urea as described previously (19). Twenty /ig of total RNA
were electrophoresed on 1% agarose gels, transferred with 20 x SSC
to nitrocellulose filters, and fixed by heating at 80°Cfor 4 h.
Hybridization and Probes. Nitrocellulose filters, containing DNA or
RNA, were prehybridized for 2 h in 10% dextran sulfate, 5 x Denhardt's
solution (Ix = 0.02% polyvinylpyrrolidone-0.02% Ficoll-0.02% bovine
serum albumin), and 50 ¿ig/mlsalmon sperm DNA at 65°C.32P-Labeled
4029
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HETEROGENEITY
OF CELLS FROM A FELINE MAMMARY CARCINOMA
probes were added and incubated for 16 h at 65°C.The filters were
washed once with 3 x SSC, 0.1 % SDS, once with 1 x SSC, 0.1 % SDS,
and twice with 0.1 x SSC, 0.1% SDS, for 30 min at 65°C,depending
on the homology of the probes. The filters were dried and exposed to
Kodak XAR-5 films with Dupont Lightning-Plus intensifying screens.
Whole plasmids or purified DNA fragments were labeled by random
priming using [-"P]dCTP (specific activity IO9 cpm/^g DNA; Amersham). The following probes were used in this study: a human HER
64.1 cDNA clone (20); a BamH\ fragment of rat neu (21); a 1.3-kilobase
Bam\\\ genomic feline sis fragment; a 2.3-kilobase BamHl genomic
feline myc fragment (both obtained from Dr. W. van de Ven); a 1.8kilobase Hincll/Pstl env fragment of feline RD 114 (provided by Dr.
P. Roy-Burman); and a 1.3-kilobase Pst\ fragment of rat glyceraldehyde
triphosphate dehydrogenase cDNA (22).
Immunoblotting. Membrane preparations (40 ¿ig)were separated by
SDS-polyacrylamide gel electrophoresis under nonreducing conditions
and blotted onto nitrocellulose according to the method described by
Towbin et al. (23). To prevent nonspecific binding the blot was blocked
with 0.5% gelatin in PBS for 60 min at room temperature. After 3
washes in PBS, the blot was incubated for 60 min at room temperature
with protein A-purified MAb 425 and MAb 139H2 (13) as a negative
control labeled with 125I(0.2 x IO5 cpm/ml). After 5 washes in PBS
the blot was dried and exposed to a Kodak XAR-5 film at -70°C.
Immune Complex Kinase Assay. Trypsinized cells (2 x IO6) were
sive rounds of single cell cloning. Only cells beyond passage 60
were characterized unless otherwise stated.
Characteristics of the K248 Cell Lines. The K248C cells grew
in tightly packed colonies, which detached readily after trypsin
ization. At saturation density, the cells were heterogeneous in
size and focally formed ridges (Fig. \A). Dome formation was
present only after several days of confluency. The K.248P cul
ture consisted of more open colonies. At high cell densities the
K248P cells formed a uniform layer of dark, "droplet" cells.
Abundant dome formation occurred spontaneously shortly after
the cultures reached confluency (Fig. 1B). The number of domes
in the K248P cell cultures was much higher compared to the
number of domes in the K248 cultures. The morphology re
mained unchanged during 2.5 years of continuous culture (over
200 passages).
The main growth characteristics (saturation density, doubling
time, soft agar growth, and serum-free growth) of both cell lines
are depicted in Table 1. Doubling times for the two variants
were not significantly different. K248C reached a higher satu
ration density. Both cell lines were serum dependent for growth
and did not form colonies in soft agar.
Ultrastructural analysis (Table 2) showed that both cell lines
washed 3 times in PBS and solubilized in lysis buffer consisting of 50 contained desmosomes, a typical feature of epithelial cells. The
mM Tris-HCl (pH 7.4), 0.15 M NaCl, 1% Triton X-100, 10% glycerol,
epithelial phenotype of K248C and K248P cells was further
100 units/ml of trasylol, and 1 mM phenylmethylsulfonyl fluoride. Cell
lysates were clarified by centrifugation at 12,000 x g for 15 min at 4°C. confirmed by the detection of keratin 8 and 18 using MAb
E,C4. The presence of FeLV in both cell lines was investigated
The cell lysates were preincubated for 15 min at room temperature in
immunologically and electron microscopically. No FeLV-gp70
the presence or absence of EOF (400 ng/ml) as described earlier (24).
Immunoprecipitationof the EGF receptor was accomplished with MAb
425 (4.65 mg/ml). The cell lysates were incubated for l h at 4°Cwith
1 /¿I
of MAb 425 or with 1 ^1 normal mouse serum/ml cell lysate. The
immune complexes were precipitated for 30 min at 4°Cwith 25 /¿I
donkey anti-mouse immunoglobulin-coated beads (IDS). The immune
precipitates were washed three times in lysis buffer and resuspended in
25 nl kinase buffer (50 mM Tris-HCl. pH 7.5-50 m\i NaCl-0.1 % Triton
X-100-10% glycol-10 HIM/?-nitrophenylphosphate-40 /«M
Na,VO4-10
mM MnClj-1 mM phenylmethylsulfonyl fluoride). The kinase reaction
was initiated by adding 10 /iCi -y-ATP followed by a 10-min incubation
on ice. The reaction was terminated by boiling with 1.5 x SDS sample
buffer (0.1 MTris/HCI, pH 6.8-2% SDS-6 M urea-7% mercaptoethanol10% glycerol-0.15% bromophenol blue). Proteins were analyzed on
7.5% SDS-polyacrylamide gels under reducing conditions. Dried gels
were exposed to Kodak XAR-5 films at -70°C.
Receptor Analysis. The receptors for estrogen, progesterone, and
EGF were determined according to methods previously described (25,
26). Briefly, competition assays were performed on intact cells for the
detection of EGF receptors and on cell lysates for the detection of
estrogen receptor and progesterone receptor, using '"Mabeled ligand
and multiple concentrations of cold ligand. The EGF receptor data
were analyzed according to the method of Scatchard (27); regression
curves were calculated with the Ligand computer program (26). The
estrogen receptor and progesterone receptor binding data were analyzed
according to the method of Scatchard. Values of >10 and >20 fmol/
mg cytosolic protein were taken as positive for estrogen and progester
one receptor, respectively.
RESULTS
Establishment of Cell Lines. The K248 primary culture con
tained at least two morphologically different cell types. One
cell type, called K248C, was isolated by repeated differential
trypsinization from the second cell type adhering strongly to
the culture flask, referred to as K248P. At passage 6, both
cultures entered a "crisis," during which most cells died and
detached from the culture vessel. Some of the K248C and
K248P cells survived the crisis and were cloned for two succes-
Fig. 1. Phase contrast micrographs of K248 established cell lines. A, K248C
cells at saturation density, x 240. B, K248P cells at saturation density with dome.
x 384.
4030
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HETEROGENEITY OF CELLS FROM A FELINE MAMMARY CARCINOMA
Table 1 Growth characteristics of the K248C and K248P cell lines
growth-EGF
agar growth
density"7.5
Cell
DomesK248C
lines
+
K248P
+++Saturation
' Cells/cm2 x 10" .
* Concentration for optimal growth, 0.01 ng/ml.
' Concentration for optimal growth, 10 ng/ml.
time(h)24
6.0Doubling
Serum-free
+EGF
-EGF
+EGF+"
23Soft
Table 2 (Ultra-) structural features of the K248C and K248P cell Unes
virusElectron
Cell
lineK248C
at
saturation den
sityPolygonal
Microvilli
croscopy"+
Desmosomes
Keratins
mi-
of
recep
tor/progesterone
leukemia
receptor (fmol/
virus*—Estrogen
mg
protein)'<
+
+Feline
K248PMorphologyPolygonalPresence
" No virus particles were detected with electron microscopy.
* Determined in an indirect immunofluorescence assay using MAb 3-17, directed against feline leukemia virus gp-70.
c A cut of value of 10 and 20 fmol/mg protein was taken for estrogen receptor and progesterone receptor positivity, respectively.
10/25.1
<10/26.5
could be detected in either cell line in an indirect immunofluo
DNA Ploidy. Table 3 presents the DNA indices of the original
rescence assay using M Ab 3-17. No virus-like particles could
tumor and the cell lines at various passage numbers. The
be detected by electron microscopy. The presence of RD-114
pulmonary metastasis from which the cell lines were derived
provirus, an endogenous virus of felines, was demonstrated by was multiploid and contained three cell populations corre
hybridization studies with a RD-114 envelope-specific probe on sponding to DI of 1.09,1.15, and 1.23. Flow cytometric analysis
DNA derived from both cell lines. No estrogen receptors and of the mixed primary culture (passage 2) showed that two cell
low amounts of progesterone receptor were detected in K248C
populations (DI 1.09 and DI 1.25) were represented in the
and K248P (Table 2).
primary culture. Repeated differential trypsinization separated
Karyotype Analysis. We have analyzed the karyotypes of both
the two cell populations. At passage 4, the K248C cells had a
cell lines at passages 5, 100, 107, and 108. Giemsa banding
DI of 1.27, whereas the K248P culture still contained two cell
patterns confirmed the feline origin of both lines. Representa
populations, one with a DI of 1.09 and a minor cell population
tive karyotypes of K248C and K248P at passage 100 are de
with a DI of 1.24. During crisis a new cell population emerged
picted in Fig. 2.
in the K248C culture with a DI of 1.17; the K248P cells
Chromosome counts showed that the K248C cell line con
remained in the near-diploid range (DI 1.06). Thereafter the
tained a modal chromosome number per cell of 41 (normal cat; ploidy of K248C and K248P remained constant for more than
2 n = 38), with a range from 35 to 46 chromosomes. Consistent
160 passages (2.5 years of culture).
structural chromosomal abnormalities were present. The con
Tumorigenicity. The xenografted K248 pulmonary metastasis
sensus karyotype (Fig. 2A) was interpreted as 35-46,XX,-A3,
produced a tumor 2 weeks after s.c. inoculation in nude mice
-D2,-Fl,-Fl,-F2,t(Bl;?),
and 9 unidentified marker chro
and could be maintained as a serially transplantable tumor line.
mosomes (UM 1-9). Marker chromosome UM 6 contained a
The K248C and K248P cell lines showed variation with respect
HSR.
to their tumorigenic abilities when injected into athymic nude
Chromosome counts revealed that the K248P cell line was mice (Table 4). K248C cells (passage 60) formed tumors at
nearly diploid, with a modal chromosome number of 39 (chro
both high and low cell concentrations in all of the s.c. inoculated
mosomes 33-44). Consistent structural chromosomal abnor
male nude mice with a latency period of 2-3 weeks. The tumors
malities were present. The consensus karyotype (Fig. IB) was reached a diameter of 1-2 cm after 3 months. Histologically,
interpreted as 33-44,XX,-B4,+C2,-Fl,-Fl,-F2,t(A2;?),
the tumors were classified as cribriform mammary carcinoma
t(Bl;?),t(Bl;?),del(qB2),del(qD4),
and 5 unidentified marker
(Fig. 3B) resembling the cribriform growth pattern seen in part
chromosomes (UM 1-5). Metaphase cells obtained from
K248C and K248P at passage 5 had chromosome numbers of of the original tumor (Fig. 3/Õ).In 1 of 13 tumors vascular
invasion was observed. Experimental metastasis was tested by
45 (chromosomes 43-47) and 39 (chromosomes 37-42),
injection of 8 x IO5 K248C cells i.v., resulting in lung coloni
respectively.
zation in 3 of 4 recipient animals within 5 months after inocu
Comparison of the karyotypes of K248C and K248P revealed
lation. K248P cells (passage 60) were poorly tumorigenic at
that both lines shared several marker chromosomes [t(Bl;?);
both high (3 of 14 take) and low (1 of 9 take) cell concentration
UM 1-5] in common that were already present in the K248C
upon s.c. inoculation into male nude mice within an observation
and K248P cells at passage 5 before onset of crisis, suggesting
period
of 4 months. The latency time was variable, ranging
a common ancestor for the two cell lines. In addition, the two
from
4
to 16 weeks. The 4 tumors that arose proliferated very
cell lines contained several unique marker chromosomes [UM
6-9 in K248C; t(A2;?), t(Bl;?), del(qB2), and del(qD4) in slowly. All K248P tumors were classified as adenocarcinomas
(Fig. 3C). Extensive necrosis was characteristic of all tumors.
K248P].
The presence of vascular invasion in 3 of 4 tumors was remark
Of particular interest is the finding of a HSR on chromosome
able. Injection of 8 x IO5 K248P cells i.v. did not result in lung
UM 6 in the K248C cells, since HSRs have been associated
colonization as verified at autopsy 3.5 months later. Tumori
with amplified chromosomal DNA. The HSR on chromosome
UM 6 was already present in the K248C cells at passage 5, thus
genicity experiments with K248C and K248P cells were re
excluding the UM 6 chromosome as the locus for the amplified
peated at passage 112 in female mice; similar results were
EGF receptor sequences.
obtained.
4031
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HETEROGENEITY
OF CELLS FROM A FELINE MAMMARY CARCINOMA
*
H
B
f
II
I
)Ì*
n u u
*r
1
» »
«
f
2
M
-
3
n in
U
D lì I-
Î »t .>
M
234
4
•
I
«9
»
1
2
3
I
1
UM
2
1C
I»
X X
X X
I I ««
1
234
5
UM
I
ÃŽ l
«
12345
6789
A
B
Fig. 2. Giemsa-banded karyotypes of the K248 cell lines. A, representative karyotype of a cell from K248C (passage 100), with 9 unidentified marker chromosomes
(UM), monosomy of chromosome (A3, D2, F2), a translocation of chromosome Bl (*). and no chromosomes (FI). B, representative karyotype of a cell from K248P
(passage 105). with 5 unidentified marker chromosomes (UM), monosomy of chromosome (B4, E2), trisomy of chromosome (C2). and no chromosomes (FI). *.
several structural alterations including t(Bl;?). t(Bl;?), t(A2;?), del(qB2), and del(qD4).
In conclusion, K248C is a highly tumorigenic cell line, while
K248P is poorly tumorigenic.
Oncogene Activation in the K248 Cell Lines: EGF Receptor
Gene. To investigate whether oncogenes, known to be involved
in human breast cancer development, were activated in the
original K248 tumor and K248 cell lines, DNA was isolated
from the original tumor and from the K248C and K248P cells
at low passage (<7 and 22) and high passage (>60) number and
screened for amplification and rearrangements of the myc and
neu oncogenes and the EGF receptor gene.
Hybridization using the 64.1 EGF receptor cDNA probe
showed that the K248C cells at high passage number contained
an amplified (4-6 times) EGF receptor gene (Fig. 4, Lane 5).
The EGF receptor gene in K248P cells at high passage number
was not amplified (Fig. 4, Lane 6). DNA isolated from normal
feline placenta was used as a control for normal EGF receptor
gene copy number (Fig. 4, Lane 4). Hybridization of the same
filters to a rat neu (Fig. 4, Lanes 4-6), a feline sis (Fig. 4, Lanes
4-6), and a feline myc probe (not shown) indicated that equal
amounts of DNA were loaded on the gel. The original K248
tumor (not shown), the K248C cells (Fig. 4, Lane 2), and
K248P cells (Fig. 4, Lane 3) at low passage number (<7 and
22) were all negative for EGF receptor gene amplification
(DNA from passage 22 is shown). DNA isolated from normal
feline liver served as a control for normal EGF receptor gene
copy number (Fig. 4, Lane 1). Hybridization of the same filter
4032
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HETEROGENEITY
OF CELLS FROM A FELINE MAMMARY CARCINOMA
Table 3 Comparison of the DNA indices of the original K248 tumor with the
DNA ploidy of the K248 cell lines at different passage numbers
in the K248 specimen (Fig. 5,4). Northern blot analysis of RNA
derived from A431 cells showed not only overexpression of the
EGF receptor gene but also the presence of a prominent 2.9kilobase transcript (Fig. 5.4) as was described previously (10).
Normal levels of neu and myc oncogene transcripts were
found in the original K.248 tumor and the two K248 cell lines
at low passage number (not shown).
The results of the Southern and Northern blot analyses
cultureK248
Tumor/cell
tumorPrimary
original
1.1.08; 1.232524*
culture"K248C
1.1.271.171.181.09;
passage41286K248P
at
passage42685DNA1.09;
at
1.1.061.11index15:
" Analyzed at passage 2.
* Minor cell population.
um
m?
Table 4 Tumorigenicity ofK248C and K248P in male B6Kh-nu/nu mice
of injec
tioni.V.
S.C.3/49/9*4/4*0/31/9*3/14*Latentperiod"(wk)2-32-3154,8,
Cell
lineK248CK248PInoculum(xlO6)0.85100.8510Site
15HistologyCribriformAdenocarcinomaAdenocarcinoma
" Weeks after s.c. inoculation to development of palpable tumors.
'Differences in tumorigenic potential between K248C (13 of 13 take) and
K248P (4 of 23 take) are statistically significant (P< 0.0001; Fisher exact test).
to a rat neu probe (Fig. 4, Lanes 1-3) showed that equal
amounts of DNA were loaded on the gel.
The EcoRl (Fig. 4, Lanes 5 and 6), Hindlll (Fig. 4, Lanes 5
and 6), BantHl, and Xbal (both not shown) restriction patterns
did not show any evidence of rearrangements of the EGF
receptor gene, sis gene, and neu gene in the K248C and K248P
cells at high passage number. No EGF receptor gene or neu
rearrangements were found in the EcoRl digest of DNA isolated
from the original K248 tumor (not shown) and the K248C (Fig.
4, Lane 2) and K248P (Fig. 4, Lane 3) cells at low passage
number. Hybridization using the 64.1 probe revealed an ampli
fied and rearranged EGF receptor gene in the control A431 cell
line (not shown).
To determine whether amplification of the EGF receptor
gene in the K248C cells could lead to increased expression of
the EGF receptor, total RNA was isolated from the K248
tumor, the K248 cell lines at high (passage 96) and low (pas
sages 22 and 23) passage number, and the A431 cell line; blotted
to nitrocellulose; and analyzed for levels of EGF receptor
transcripts. The filter was hybridized to the 64.1 EGF receptor
cDNA probe (Fig. 5A) and to a rat glyceraldehyde triphosphate
dehydrogenase probe (Fig. 5B) to check for the amount of RNA
loaded on the gel. After correction for the amount of RNA
loaded on the gel, we concluded that the K248C cells at high
passage number contained increased levels of normal-sized
EGF receptor transcripts (Fig. 5,4). The size of normal EGF
receptor transcripts was determined using normal feline salivary
gland RNA (not shown). The EGF receptor transcripts were
present in lower quantities in the original K248 tumor (not
shown), in the K248P cells at high passage number (Fig. 5,4),
and in the K248C and K248P cells at low passage number (Fig.
5/1). In addition, no expression of the EGF receptor gene was
found in the FL74 and F422 feline lymphoblast cell lines (not
shown). No truncated EGF receptor transcripts were detected
_¿£'--W
•
*
Fig. 3. A, original K248 pulmonary metastasis, with differentiated adenocarcinomawith cribriform (C) and solid (S) type of growth. H & E. x 165. B, typical
histológica! appearance of a s.c. growing K248C xenograft, with cribriform type
of growth. H & E, X 130. C, histológica! appearance of a s.c. growing K248P
xenograft showing central necrosis (A7).H & E, x 130.
4033
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HETEROGENEITY
Eco Rl
EcoRI
1
456456
2 3
OF CELLS FROM A FELINE MAMMARY CARCINOMA
425 and tested for autophosphorylation activity. The A431 cells
were used as a control for autophosphorylation activity. Fig. 7
shows that both the K248C and the K248P cells contained a
basal level of autophosphorylation activity. The basal levels of
autophosphorylation activity in K248C and K248P cannot be
compared with each other since the membrane preparations
loaded onto the gel were not corrected for the amount of
receptors present. Both K248C and K248P cells responded to
EGF stimulation by an increased autophosphorylation of the
EGF receptor. These results suggest that in both K248C and
K.248P a functional EGF receptor was present. Only a slight
increase in phosphorylation activity of the EGF receptor was
observed in the A431 cell line after addition of EGF (Fig. 7).
Estimation of EGF Binding Sites. Scatchard plot analysis of
the EGF binding data (Fig. 8) revealed two receptor classes for
each cell type with dissociation constants in the range of 10~8
to 10~'°M. The total amount of EGF receptors in K248C was
Hindlll
EGFR
23.0-
^.9.4-6.6--4.4-
CO
CM
\ 2.0-
CO
O)
CM
CM
O
û.
CO
CO
O
CD
O
-
CO
CO
CM
CM
CO
r-Neu
—6.6-
CM
—9.4
CM
—4.4
SIS
9.4 -. '
—6.6
kb
Fig. 4. Southern blot analysis of DNA from the K248C and K248P cell lines
at high and low passage number, normal feline liver and normal feline placenta.
DNA was digested with EcoRi or //mdlll restriction endonuclease. blotted to
nitrocellulose and hybridized to the 64.1 EGF receptor cDNA probe (EGFR) to
a rat neu(r-Neu) and a feline sis (SIS) probe. Lane I, normal feline liver; Lane 2,
K248C at passage 22; Lane 3, K248P at passage 22; Lane 4, normal feline
placenta; Lane 5, K248C at high passage number (>60); Lane 6, K248P at high
passage number (>60). Molecular markers (in kilobases) are derived from electrophoresis of XDNA digested with //mdlll.
showed that amplification of the EGF receptor gene in K248C
cells at high passage number correlated with overexpression of
the EGF receptor gene, without rearranging the gene.
Further characterization of the EGF receptor on the protein
level was performed with K248C and K248P cells beyond
passage 60.
Analysis of EGF Receptor Levels in Cell Lines. In order to
determine whether increased EGF receptor gene expression
correlated with elevated levels of EGF receptor protein, we
analyzed the amount of EGF receptor in K248C and K248P by
Western immunoblotting using purified 125I-labeled monoclo
nal antibody 425. As a negative control M Ab 139H2 was used.
Equal amounts of membrane proteins, as determined with the
BioRad protein assay, were loaded on the gel. The results (Fig.
6) revealed that K248C contained more M, 170,000 EGF
receptors than did K248P. In conclusion, the difference in EGF
receptor RNA levels between the K248C and K.248P cell lines
was reflected at the protein level.
EGF Receptor Kinase Activity. Upon ligand binding to the
EGF receptor a tyrosine-specific kinase becomes activated
which results in autophosphorylation of the EGF receptor. To
determine whether the EGF receptors of both cell lines were
functionally active, EGF receptors from both K248C and
K248P were immune precipitated with monoclonal antibody
-4.8
-2.0
B
-1.4
Fig. 5. Northern blot analysis of total RNA from K248C and K248P at low
passage (23 and 22, respectively) and high passage (96) number and from A431.
The same nitrocellulose filter was hybridized to the 64.1 EGF receptor cDNA
probe (A) and the glyceraldehyde triphosphate dehydrogenase probe (B). Kb,
kilobases.
4034
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HETEROGENEITY
425
OF CELLS FROM A FELINE MAMMARY CARCINOMA
eration of A431 cells is in accordance with literature data (28).
EGF did not induce growth of K248C and K248P cells in soft
agar.
139H2
J>J^CO-1IOJ^00oIO^00•o3>*CO-1104^k00oN)^00•o
DISCUSSION
Thus far, the establishment of feline mammary carcinoma
cell lines has met with low success rates. Recently, a cell line
was isolated from 1 of 30 mammary carcinomas (7). In another
study, no feline mammary carcinoma cell lines were established
from 14 mammary carcinomas (29). We succeeded in establish
ing 4 cell lines of 135 feline mammary carcinomas; 3 of the 4
kd
200-
NMS
425
11692-
O
A431
K248C
K248P
te$
»e
EGF
200-
68-
11668-
45-
45-
Fig. 6. Western immunoblot analysis of the EGF receptor from A431. and
from K248C and K248P at high passage number. Equal amounts of membrane
protein were separated by SDS-polyacrylamide gel electrophoresis under nonreducing conditions and blotted onto nitrocellulose filters. The EGF receptor was
detected by protein A-purified MAb 425 coupled to '"I. The specificity of the
reaction was determined by incubating the filter with protein A-purified MAb
139H2 (negative control) coupled to '"1. Size markers are in thousands.
5.6 times the number calculated for K248P (Table 5). When a
distinction was made between the number of high- and lowaffinity receptors in both lines it was observed that K248C
contained 5 times more low-affinity receptors and 10 times
more high-affinity receptors than K248P (Table 5).
In Vitro Response of Cell Lines to EGF. The effect of increas
ing concentrations of EGF on K248C, K248P, and A431 cell
proliferation is shown in Fig. 9. Under the serum-free condi
tions used, the proliferative response of K248C cells to the
addition of EGF was biphasic with stimulation of growth at
0.01 and 0.1 ng/ml EGF and inhibition at EGF concentrations
exceeding 1 ng/ml. The K248C cell line was capable of growth
in serum-free medium in the presence of 0.01 ng/ml EGF for
an observation period as long as 3 months.
In contrast, EGF at doses ranging from 0.01 to 50 ng/ml
stimulated the growth of K248P cells under the serum-free
conditions used. K248P cells were capable of growth in serumfree medium in the presence of EGF at concentrations ranging
from 0.01 to 50 ng/ml and have been carried as long as 3
months in this manner. The observed effect of EGF on prolif-
Fig. 7. Analysis of protein kinase activity of the EGF receptor from AA431
cells and from K248C and K248P cells at high passage number. Cell lysates from
2 X IO6 K248C, K248P, and A431 cells were incubated for 15 min at 20°Cin the
presence or absence of EGF (400 ng/ml) and immunoprecipitated with either
MAb 425 or normal mouse serum (NMS). Immunoprecipitates were incubated
for 10 min at 4°Cwith 10 ^Ci [-y-32P]ATP. The reaction was terminated by
addition of sample buffer, and the proteins were analyzed by gel electrophoresis
under reducing conditions and autoradiography. Size markers are in thousands.
0.010
O K248C
•K24BP
0008
0.004
0002
15
20
(mol/TO5 c»«t
25
bound
Fig. 8. Scatchard analysis of EGF binding to K248C and K248P cells. Cells
were incubated at 4°Cfor 3 h with a constant amount of '25I-labeled EGF together
with increasing amounts of unlabeled EGF. Cell-associated radioactivity was
measured and Scatchard analysis was performed as described in "Materials and
Methods." Nonspecific binding was determined using a 1000-fold excess of
unlabeled EGF.
4035
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HETEROGENEITY
OF CELLS FROM A FELINE MAMMARY CARCINOMA
Table 5 EGF receptor content of the K248C and K248P cell lines
receptors0Cell
No. of EGF
lineK248C
affinity2.8
x IO4
0.28 x 10"A",
(nM)0.29
affinity14
x IO4
2.7 x IO4A'j
K248PHigh
0.06Low
'' Each result is the mean of two separate binding assays
10 WO 001 01
1
w wo am ai i
concOTtraUon
gene nor detected truncated EGF receptor transcripts. Ampli
fication of the EGF receptor gene accompanied by elevated
levels of RNA and EGF receptors has been reported for a
number of human breast cancer cell lines (38, 39) and primary
breast carcinomas (40). Neither EGF receptor gene amplifica
tion nor EGF receptor gene overexpression was found in the
original tumor or in the K248C cells before onset of crisis as
detected by Southern and Northern blot analyses. Moreover,
immunohistochemistry using MAb 425 on frozen sections of
the original tumor was completely negative (not shown). Al
though we cannot exclude the possibility that a minor cell
population with an amplified EGF receptor gene was present
in the original tumor, we believe that amplification of the EGF
receptor gene occurred during establishment of the K248C cell
line. Apparently, elevated EGF receptor levels conferred a
proliferative advantage to K248C cells in vitro, leading to their
establishment as a cell line. A similar observation has been
made in human squamous cell carcinomas, which showed a low
incidence of amplification in primary' tumors but a high inci
(n\i)34
31
io wo
al EGF(ngAnl)
Fig. 9. Effect of EGF on the growth of K248C, K248P, and A431 cells. Cells
were plated onto 35-mm plates at a density of IO4 cells/plate in DMEM/F-12
medium containing 10% FCS. After attachment overnight, medium was changed
to DMEM/F-12 with insulin and transferrin and increasing concentrations of
EGF. No EGF was added to control plates. Cell counts of triplicate plates were
made at day 10 for K248C and K248P and at day 6 for A431. Points, percentage
of the control without EGF; bars, SD; *, significant differences from the control
(*. P < 0.01: ",P<
0.005; •**,
P < 0.001: Student's t test).
cell lines were isolated from pulmonary métastases.Here, we
report the properties of two distinct epithelial cell lines derived
from a single multiploid feline mammary carcinoma. Compar
ison of DNA indices of the original tumor and the K248C and
K248P cells suggested that the two cell types were already
present in the original tumor. Chromosome analysis showed
that in addition to unique marker chromosomes characteristic
for each line, both cell lines had several marker chromosomes
in common. The common marker chromosomes were already
present at low passage number, suggesting that the two cell
lines arose from a common ancestor. During tumor progression
these cell populations may have diverged and coexisted in the
animal. Our study suggests that the characteristic heterogeneity
in the histológica! appearance of individual feline mammary
carcinomas might be a reflection of the presence of clonally
related but distinct tumor cell subpopulations. The impact of
this observation on tumor therapy is obvious and has been
discussed elsewhere (30). Tumor heterogeneity and the isolation
of distinct cell types in culture have been described for a few
other tumor systems in different species (9, 31-36).
The two cell lines differed from each other with respect to
the number of EGF receptors expressed on the cell surface and
the tumorigenic potential in athymic mice. Since the EGF
receptor content in the K248P cell line was comparable with
the numbers found in two other feline mammary carcinoma
cell lines, in a feline fibroblast line (data not shown), and in
several human mammary carcinoma cell lines (37), we consider
these values to represent the basal receptor level. We have
shown that in the K248C cell line at high passage number,
overexpression of the EGF receptor gene was accompanied by
an increase in the level of EGF receptor protein and of EGF
binding. The marked increase in EGF receptor RNA levels was
probably the result of EGF receptor gene amplification only.
We neither observed rearrangements near the EGF receptor
dence in cell lines established from these tumors (41). The data
presented in this study emphasize the risk of extrapolating
results obtained in vitro to the in vivo situation.
A more fundamental question is whether abnormal expres
sion of the EGF receptor gene is restricted to feline mammary
carcinoma cells in culture or that overexpression of the EGF
receptor gene is also causally related to the development of
feline mammary carcinomas. To address these issues a system
atic study on EGF receptor alterations in a large series of feline
primary and metastatic mammary carcinomas is now in prog
ress at our institute. Studies on human breast (6, 40), bladder
(42), and squamous cell carcinomas (43) suggest a causal rela
tionship
between
EGF
receptor
overexpression
and
tumorigenesis.
The tyrosine kinase activity of the EGF receptor, functioning
as a second messenger, is essential for many biological re
sponses, including DNA synthesis (44). The EGF receptor
proteins identified by MAb 425 showed functional tyrosine
kinase activity as detected by receptor autophosphorylation.
The EGF receptor in K248C and K248P cells was biologically
competent since both cell lines were EGF dependent for growth
under serum-free conditions. However, the effect of EGF on
the growth of the two cell lines was different. A correlation was
found between the number of EGF receptors and the prolifer
ative response of K248C and K248P cells to the addition of
EGF. These data are consistent with the threshold model pro
posed by Kawamoto et al. (28), indicating an optimum ratio
between EGF and EGF receptors.
The observed relationship between the high number of EGF
receptors in K248C and its tumorigenic potential in nude mice
has also been described for the human A431 (45) and MDA468 (46) cell lines in which the extent of amplification of the
EGF receptor gene in the parental and isolated variant lines
correlated with the rate of tumor growth and length of the
latency period in athymic mice.
The K248 cell lines established in this work could be useful
for extending insight into the biology and heterogeneity of
feline mammary carcinomas. Since these lines can be grown in
cats (47) they may provide useful models for in vitro and in vivo
targeting studies with monoclonal antibodies.
ACKNOWLEDGMENTS
We gratefully acknowledge the gifts of monoclonal antibodies from
Dr. U. Rodeck (425), Dr. K. Weijer (3.17), and Dr. D. Ivanyi (E3C4)
4036
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
HETEROGENEITY
OF CELLS FROM A FELINE MAMMARY CARCINOMA
and probes from Dr. J. Neill (myc), Dr. W. van der Ven (sis), Dr. A.
Ulrich (64.1 ), and Dr. P. Roy-Burman (RD-114). We thank Jos Hoekstra, Nel Dijkshoorn, Els Verhoeven, Nel de Haan, and Ton Ultee for
technical assistance and Ir. G. Hart for statistical analysis of the data.
Our thanks are due to Dr. A. Sonnenberg, Dr. G. Verheijden, Dr. A.
Bosma, and Dr. A. Jongsma for useful discussions and to Dr. W. Mooi
for critical reading of the manuscript.
REFERENCES
1. Weijer, K., Head, K. W., Misdorp, W., and Hampe, J. F. Feline mammary
tumors. I. Morphology and biology: comparisons with human and canine
mammary carcinomas. J. Nati. Cancer Inst., 49: 1697-1704, 1972.
2. Weijer, K., and Hart, A. A. Prognostic factors in feline mammary carcinoma.
J. Nati. Cancer Inst., 70: 709-716, 1983.
3. Stolwijk, J. A. M., Minke. J. M. H. M., and Misdorp, W. Breast cancer,
model no. 202, supplemental update, 1987. In: C. C. Capen, T. C. Jones,
and G. Migaki (eds.). Handbook Animal Models of Human Disease, Fasciniile 16. Registry of Comparative Pathology, Washington, DC: Armed Forces
Institute of Pathology, 1987.
4. Stolwijk, J. A. M., Minke, J. M. H. M., Rutteman, G. R., Hoekstra, J., Prop,
F. J. A., and Misdorp, W. Feline mammary carcinomas as a model for human
breast cancer. II. Comparison of in vivo and in vitro Adriamycin sensitivity.
Anticancer Res., 9: 1045-1048, 1989.
5. van de Vijver, M. J., Peterse, J. L., Mooi, W. J., Lomans, J., Verbruggen,
M., van de Bersselaar, R., Devilee, P., Cornelisse, C., Bos, J. L., Yarnold, J.,
and Nusse, R. Oncogene activations in human breast cancer. Cancer Cells,
in press, 1991.
6. Sainsbury, J. R. C., Sherbet, G. V., Farndon, J. R., and Harris, A. L.
Epidermal growth factor receptors and estrogen receptor in human breast
cancer. Lancet, 1: 364-366, 1985.
7. Norval, M., Maingay, J., and Else, R. W. Characteristics of a feline mammary
carcinoma cell line. Res. Vet. Sci., 39: 157-164, 1985.
8. Hampe, J. F., and Misdorp, W. International histological classification of
tumors in domestic animals: tumor and dysplasias of the mammary gland.
Bull. W. H. O., 50: 111-133, 1974.
9. Sonnenberg, A., Daams, H., Calafat, J., and Hilgers, J. In vitro differentiation
and progression of mouse mammary tumor cell. Cancer Res., 46: 59135922, 1986.
10. Merlino, G. T., Xu, Y-H., Ishii, S., Clark, A. J., Semba, K., Toyoshima, K.,
Yamamoto. T., and Pastan, I. Amplification and enhanced expression of the
epidermal growth factor receptor gene in A431 human carcinoma cells.
Science (Washington DC), 224:417-419, 1984.
11. Jarrett, O., Laird, H. M., and Hay, D. Determinations of the host range of
feline leukemia viruses. J. Gen. Virol., 20: 169-175, 1973.
12. Murthy, U., Basu, A., Rodeck, U., Herlyn, M., Ross, A. H., and Das, M.
Binding of an antagonistic monoclonal antibody to an intact and fragmented
EGF-receptor polypeptide. Arch. Biochem. Biophys.. 2: 549-560, 1987.
13. Hilkens, J., and Buys, F. Biosynthesis of MAM-6, an epithelial sialomucin.
J. Biol. Chem., 263: 4215-4222, 1988.
14. Weijer, K., Uytdehaag, F. G. C. M., Jarrett, O., Lutz, H., and Osterhaus, A.
D. M. E. Post-exposure treatment with monoclonal antibodies in a retrovirus
system: failure to protect cats against feline leukemia virus infection with
virus neutralizing monoclonal antibodies. Int. J. Cancer, 38: 81-87, 1986.
15. Ford, C. E., Pollock, D. L., and Gustavsson, I. Proceedings of the first
internal conference on standardization of banded karyotypes of domestic
animals. Hereditas, 92: 145-162, 1980.
16. Minke, J. M. H. M., Cornelisse, C. J., Stolwijk, J. A. M., Kuipers-Dijkshoorn, N., Rutteman, G. R., and Misdorp, W. Flow cytometric DNA-ploidy
analysis of feline mammary tumors. Cancer Res., 50: 4003-4007, 1990.
17. Maniatis, T., Fritsch, E. F., and Sambrook, J. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory,
1982.
18. Southern, E. M. Detection of specific sequences among DNA fragments
separated by gel electrophoresis. J. Mol. Biol., 98: 503-517, 1975.
19. Schuuring, E., Van Deemter, E., Roelink, H., and Nusse, R. Transient
expression of the proto-oncogene int-l during differentiation of P-19 em
bryonal carcinoma cells. Mol. Cell. Biol., 9: 1357-1361, 1989.
20. Ullrich, A., Coussens, L., Hayflick, J. S., Dull, T. J., Gray, A., Tarn, A. W.,
Lee, J., Yarden, Y., Libermann, T. A., Schlessinger, J., Downward, J., Mayes,
E. L. V., Whittle, N., Waterfield, M. D., and Seeburg, P. H. Human
epidermal growth factor receptor cDNA sequence and aberrant expression
of the amplified gene in A431 epidermoid carcinoma cells. Nature (Lond.),
309: 418-425, 1984.
21. Bargmann, C. I., Hung, M-C., and Weinberg, R. The neu oncogene encodes
an epidermal growth factor receptor-related protein. Nature (Lond.), 319:
226-230, 1986.
22. Fort, P. H., Marty, L., Piechaczyk, M., Sabrouth, S. E. L., Dani, C. H.,
Janteau, Ph., and Blanchard, J. H. Various rat adult tissues express only one
major niRNA species from the glyceraldehyde-3P-dehydrogenase gene family. Nucleic Acids Res., 13: 1417-1431, 1985.
23. Towbin, H., Staehelin, T., and Gordon, J. Electrophoretic transfer of proteins
from polyacrylamide gels to nitrocellulose sheets: procedure and some appli
cations. Proc. Nati. Acad. Sci. USA, 76: 4350-4354, 1979.
24. Defize, L. H. K., Moolenaar, W. H., van der Saag, P. T., and deLaat, S. W.
Dissociation of cellular responses to epidermal growth factor using antireceptor monoclonal antibodies. EMBO J., 5: 1187-1192, 1986.
25. Rutteman, G. R., Willekes-Koolschijn, N., Bevers, M. M., van der Gughten,
A. A., and Misdorp, W. Prolactin binding in benign and malignant mammary
tissue of female dogs. Anticancer Res., 6: 829-835, 1986.
26. Boonstra. J., Mummery, C. L., van der Saag, P. T., and de Laat, S. W. Two
receptor classes for epidermal growth factor on pheochromocytoma cells,
distinguishable by temperature, lectins and tumor promoters. J. Cell. Physiol., 123: 347-352, 1985.
27. Scatchard, G. The attraction of proteins for small molecules and ions. Ann.
NY Acad. Sci., 51: 660-672, 1949.
28. Kawamoto, T., Mendelsohn, J., Le, A., Sato, G. H., Lazar, C. S., and Gill,
G. N. Relation of epidermal growth factor receptor concentration to growth
of human epidermoid carcinoma A431 cells. J. Biol. Chem., 259: 77617766, 1984.
29. Elshire, H. D. Some studies on the feline mammary adenocarcinoma as a
model of human breast cancer. Ph.D. Dissertation. Los Angeles: University
of Southern California, 1984.
30. Kimball, P. M., and Brattain, M. G. Isolation of a cellular subpopulation
from a human colonie carcinoma cell line. Cancer Res., 40: 1574-1579,
1980.
31. Bennett, D. C., Peachey, H. D., and Rudland, P. S. A possible mammary
stem cell line. Cell, /5: 283-298, 1978.
32. Dexter, D. L., Kowalski. H. M., Blazar, B. A., Fligiel, Z., Vogel, R., and
Heppner, G. H. Heterogeneity of tumor cells from a single mouse mammary
tumor. Cancer Res., 38: 3174-3181, 1978.
33. Brattain, M., Fine, W. D., Khaled, M., Thompson, J., and Brattain, D. E.
Heterogeneity of malignant cells from a human colonie carcinoma. Cancer
Res.,«: 1751-1756, 1981.
34. Whitehead, R. H., Bertoncello, I., Webber, L.. and Pedersen, J. S. A new
human breast carcinoma cell line (PMC42) with stem cell characteristics. I.
Morphologic characterization. J. Nati. Cancer Inst., 70: 649-661, 1983.
35. Schuuring. E., van der Leede, B. J., Willems, R., Daams, H., van der Valk,
M., van de Vijver, M., van Leeuwen, F., Sonnenberg, A., and Nusse, R.
Differentiation dependent expression of provirus-activated i'nf-1 oncogene in
clonal cell lines derived from a mouse mammary tumor. Oncogene, 5: 459465, 1990.
36. Peterson, O. W.. van Deurs, B., Nielsen, K. V., Madsen, M. W., Laursen, I.,
Balslev, I., and Briand, P. Differential tumorigenicity of two autologous
human breast carcinoma cell lines, HMT-3909S1 and HMT-3909S8, estab
lished in serum-free medium. Cancer Res., 50: 1257-1270, 1990.
37. Imai, Y., Leung, C. K., Friesen, H. G., and Shiu, P. C. Epidermal growth
factor receptor and effect of epidermal growth factor on growth of human
breast cancer cells in long-term tissue cultures. Cancer Res., 42: 4394-4398,
1982.
38. Filmus, J., Pollak, M. N., Cailleau, R., and Buick, R. N. MDA-468, a human
breast cancer cell line with a high number of epidermal growth factor (EGF)
receptors, has an amplified EGF receptor gene and is growth inhibited by
EGF. Biochem. Biophys. Res. Commun., 128: 898-905, 1985.
39. King, C. R., Kraus, M. H., Williams, L., Merlino, G. T., Pastan, I. H., and
Aaronson, S. A. Human tumor cell lines with EGF receptor gene amplifica
tion in the absence of aberrant sized mRNAs. Nucleic Acids Res., 13: 84778486, 1985.
40. Slamon, D. J., Clark, G. M., Wonh, S. G., Levin, W. J., Ullrich, A., and
McGuire, W. L. Human breast cancer: correlation of relapse and survival
with amplification of the HER-2/neu oncogene. Science (Washington DC),
235: 177-182, 1987.
41. Yamamoto, T., Kamata, N., Kawano, H., Shimizu, S., Kuroki, T., Toyoshima, K., Rikimaru. K., Nomura, N., Pastan, I., Gamou, S., and Shimizu, N.
High incidence of amplification of the epidermal growth factor receptor gene
in human squamous carcinoma cell lines. Cancer Res., 46:414-416, 1986.
42. Neal, D. E., Marsh, C, Bennett, M. K., Abel, P. D., Hall, R. R.. Sainsbury,
J. R. C., and Harris, A. L. Epidermal growth factor receptors in human
bladder cancer: comparison of invasive and superficial tumours. Lancet, /:
366-368, 1985.
43. Kamata, N., Chida, K., Rikimaru, K., Horikoshi, M., Enomoto, S., and
Kuroki, T. Growth-inhibitory effects of epidermal growth factor receptor and
overexpression of its receptors on human squamous cell carcinomas in
culture. Cancer Res., 46: 1648-1653, 1986.
44. Moolenaar, W. H., Bierman, A. J., Tilly, B. C, Verlaan, I., Defize, L. H. K.,
Honegger, A. M.. Ullrich, A., and Schlessinger, J. A point mutation at the
ATP-binding site of the EGF-receptor abolished signal transduction. EMBO
J., 7:707-710, 1988.
45. Santon, J. B., Cronin, M. T., MacLeod, C. L., Mendelsohn, J., Masui, H.,
and Gill, G. N. Effects of epidermal growth factor receptor concentration on
tumorigenicity of A431 cells in nude mice. Cancer Res., 46: 4701-4705,
1986.
46. Filmus, J., Trent, J. M., Pollak, M. N., and Buick, R. N. Epidermal growth
factor receptor gene-amplified MDA-468 breast cancer cell line and its
nonamplified variants. Mol. Cell. Biol., 7: 251-257, 1987.
47. Minke, J. M. H. M., Weijer, K., and Misdorp, W. Allotransplantation of the
K248 feline mammary carcinoma cell line in cats: a model for monoclonal
antibody guided detection and therapy of human breast cancer. Lab. Invest.,
in press, 1991.
4037
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.
Isolation of Two Distinct Epithelial Cell Lines from a Single
Feline Mammary Carcinoma with Different Tumorigenic
Potential in Nude Mice and Expressing Different Levels of
Epidermal Growth Factor Receptors
Jules M. H. M. Minke, Ed Schuuring, Raymond van den Berghe, et al.
Cancer Res 1991;51:4028-4037.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/51/15/4028
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1991 American Association for Cancer Research.