Download Estradiol-dependent Collagenolytic Enzyme

[CANCER RESEARCH 35, 2039-2048, August 1975]
Estradiol-dependent Collagenolytic Enzyme Activity in Long-Term
Organ Culture of Human Breast Cancer'
Jean-Claude Heuson, Jean-Lambert Pasteels, Nicole Legros, Jeanine Heuson-Stiennon,
and Guy Leclercq
Servicede Médecine
et Laboratoire d'Investigation Clinique de I'Institut Jules Bordet [J.-C. H., N. L., G. L.], and Laboratoire d'Histologie, Faculté
de
Médecine,UniversitéLibre de Bruxelles [J..L. P., J. 11.-SI, Brussels, Belgium
SUMMARY
nent of the hormone
growth.
dependency
of human
breast cancer
An organ culture method suitable for the maintenance of
viable human breast cancer for at least 14 days has been
INTRODUCTION
described. This method was applied to a total of 94 breast
cancer specimens. It allowed good survival of―soft―tumors
The mechanisms
of hormone dependence of human
of various histological types, with loose connective stroma
breast cancer that form the basis of its endocrine treatments
even in hormone-free
medium. In contrast, “scirrhous―
are far from being understood. Their study can be ap
cancers showed poor survival in hormone-free
medium;
proached by in vitro experiments
that may serve the
viable cells were maintained only at the very periphery of
additionally useful goal of selecting patients for hormonal
the explants. Supplementation
of the medium with insulin
therapies. Organ culture is likely to provide an appropriate
(10 zg/ml), ovine prolactin (5 zg/ml), and hydrocortisone
method for that purpose because it preserves the normal
(1 zg/ml) in various combinations
seemed to induce en
relationships
of epithelial tumor cells and surrounding
largement
of viable cancer cells and moderate
loosening of
connective tissue. In contrast to cell or tissue culture
the stroma in some cases. However, it did not improve the
methods, it avoids growth competition between tumor cells
survival
of central
tumor
cords in scirrhous
explants.
and fibroblasts
and allows a clear recognition of the
Further supplementation
of the medium with 17 @-estradiol
histological nature of the cells. The technique should be able
(minimum effective dose, 0.1 to 10 ng/ml), although it did
to sustain living tumor tissue for several weeks, ensuring
not affect soft tumors, markedly improved survival of the
that hormone assays, even if performed early in the culture
cancer cells of scirrhous tumors throughout
the whole
experiment, are not carried out on dying cells (17).
explants, with evidence of collagen digestion around the
Organ culture methods were successfully applied to
neoplastic cells. This was observed in 18 of 20 scirrhous
mammary tumors of laboratory animals (7, 10, 14- 16, 27,
cancers subjected to this treatment. Estradiol need not be
35).
Unfortunately,
long-term
organ
culture
of human
present during the whole culture period; the results at 14
breast cancer appears to be especially difficult, particularly
days were identical in explants treated with estradiol for the
in the case of scirrhous tumors (I, 21, 32, 34, 36). This is
first 7 days only or for the entire period. Addition of
probably the reason why several authors have proposed the
purified collagenase during the first 24 or 48 hr of culture
use of cell or tissue cultures (2, 3, 11, 25) or very short-term
resulted in complete dissolution of the collagen. After such
organ cultures (4, 5, 18, 21, 29, 30, 32).
treatment, culture under the usual conditions resulted in
The present paper describes experiments of organ culture
excellent survival of the explants without improvement from
where explants from human breast cancer, including the
hormone supplementation;
thus, while estradiol was neces
scirrhous type, were maintained for periods up to 14 days. It
sary when collagen was present, it was no longer required
describes unexpected findings on the influence of estrogen
after collagen digestion. It can be concluded that breast
on survival of the scirrhous tumors.
cancer cells in organ culture are only slightly, or not at all,
hormone dependent for survival, provided that they are not
restrained by a dense collagen barrier. The estrogen-induced
MATERIALS AND METHODS
changes allowing survival inside the scirrhous explants
strongly suggest the presence of an estrogen-dependent
Ninety-four breast cancer tissue samples were obtained
collagenolytic enzyme system in the collagen-rich breast
from 88 female patients and I male patient. These samples
cancers. This system could represent an important compo
were from primary breast carcinomas in 75 cases, from both
the primary and an invaded axillary lymph node in 5 cases,
I This
work
was
supported
by
grants
from
the
“
Fonds
Cancé
and from metastatic lesions (nodes, skin) in 9 cases. The
rologique de Ia Caisse Générale
d'Epargne et de Retraite de Belgique―and
patients' ages ranged from 29 to 94 years, with a mean value
performed
in part under contract of the Ministère de Ia Politique
of 60 years. Twenty-three patients were premenopausal,
8
Scientifique within the framework of the Association Euratom, University
were within I year of their last menstrual period, 56 were
of Brussels, University
of Pisa.
postmenopausal,
and I was of undetermined status.
Received January 8, 1975; accepted April 24, 1975.
AUGUST
1975
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1975 American Association for Cancer Research.
2039
J.-C. Heuson el a!.
Immediately after excision, the sterile tissue samples were
maintained in Earle's base at 0° and transferred to the
laboratory
within 15 mm. In the 1st 23 experiments,
explants were prepared for culture by mincing the specimens
with opposed surgical scalpel blades operated on a porcelain
surface. However, histological examinations revealed that
this procedure produced a considerable amount of crushing
when applied to scirrhous carcinomas.
From then on, in
order to avoid this harmful effect, the samples were
prepared on a solidified paraffin surface by means of
opposed razor blades (Schick Ultra Platinium; Eversharp
Nederiand B.V., Amsterdam, The Netherlands) selected for
the sharpness and hardness of their cutting edge. Adhering
fat, connective tissue strands, and necrotic areas were
dissected out and the tumors were cut into 1- to 2-mm
explants. Five explants were placed on a Millipore filter,
pore size 0.45 jzm (Millipore, Malakoff, France), or on a
small block of agar gel (8 x 8 x 4 mm) (Difco Laboratories,
Inc., Detroit, Mich.) and transferred on stainless steel grids
in Falcon disposable Petri dishes (Falcon Plastics, Los
Angeles, Calif.) containing 2.5 ml of liquid culture medium.
The height of the grids was so calculated that the explants
were at the liquid-gas interphase. Agar gel was prepared by
the method of Willcox and Thomas (36). A I % solution in
Earle's base was autoclaved and then poured into sterile
Petri dishes to form a 4-mm thick layer that was solidified
at room temperature.
Agar-gel blocks of appropriate size
were then cut. The cultures were carried out at 37°in a
water-saturated
atmosphere of 95% 02:5% CO2. Medium
199 Earle's base (Flow, Irvine, Scotland, or Gibco Biocult,
Glasgow, Scotland), supplemented with glucose to reach a
final concentration
of 250 mg/ 100 ml, was used as basic
medium. It was changed every 2 days. Hormone or enzyme
supplementation
was carried out as follows.
Insulin. A stock solution of insulin (Sigma Chemical
Co., St. Louis, Mo.) was prepared by dissolving hormone in
0.005 N HCI; it was added to the medium at a final
concentration of 10 sg/ml.
Prolactin. Prolactin (ovine, NIH-P-S-l0,
Bethesda, Md)
was dissolved at a concentration
of 0.33 mg/ml in 0.9%
NaCl, pH 8.4; the final concentration in the culture medium
was 5 @ig/ml.
Hydrocortisone
and 17@3-Estradiol. The stock solutions
(Sigma) were prepared by dissolving the hormones at a
concentration of I mg/mi in absolute ethanol. Hydrocorti
sone was added to the culture medium at a final concentra
tion of 1 zg/ml; estradiol was added at final concentrations
of 0. 1 and 1 ng/ml. Some experiments involved additional
concentrations
from 0.01 ng to 10 sg/ml. The final ethanol
concentration never exceeded 0.1%.
Bacterial Collagenase. A preparation (Calbiochem, San
Diego, Calif.) titrating at 100,000 C units/g was added to
the culture medium at a final concentration of 1 mg/mi.
The cultures were maintained for 4, 6, 7, or 14 days. At
the end of each culture period, all explants from 2 or 3
Millipore discs or agar-gel blocks for each experimental
condition were fixed in Bouin's fluid for histological exami
nation. They were cut serially, and 8 regularly spaced
sections were placed on a slide. They were stained with
hematoxylin and eosin. In selected experiments,
specific
2040
staining of the collagen was carried out by use of Masson
trichrome, Van Gieson staining, and periodic acid-Schiff
reaction (24).
Histological
typing of the tumors was performed on
samples collected at the start of the experiment according to
Stewart (31) and Kuzma (20). The tumors were further
subdivided into 2 categories: (a) “soft―
tumors with scanty
or loose stroma,
and (b) “collagen-rich― (scirrhous)
tumors.
Before culture, a tissue specimen ofeach tumor was saved
for estrogen receptor determinations.
The method used has
been described (23) and is based on the preparation of the
tissue cytosol fraction and measurement
of its specific
binding affinity for 17@3-[3H}estradiol.
RESULTS
Histological
Typing
The histological types of the 94 human breast cancers
studied are listed in Table 1. Thirty-two were soft tumors
and 62 were collagen-rich tumors. The mean age of patients
bearing tumors of these categories was 6 1 and 59 years old,
respectively, with little variation among histological types.
The proportion of postmenopausal
patients in each group
was 58 and 68%, respectively, the remainder being pre
menopausal or rarely menopausal. The groups and various
histological types were thus homogeneous with respect to
age and menopausal status.
Table I
Survival of human breast cancer tissue in organ culture according to
histological type
The cultures were carried out for 4 to 14 days; the medium contained no
hormone or was supplemented with insulin, prolactin, or hydrocortisone,
alone or in combination.
Survival―Histological
eralcSoft
of
casesCen
typingNo.
tralbPeriph
tumorsInfiltratingductalcarcinoma121212(carcinoma
simplex)Medullary
carcinoma502Colloid
carcinoma444Comedocarcinoma444Papillary
(infiltrating)333Paget's
carcinoma
(infiltrating)II1Infiltrating
disease
carcinoma323Collagen-rich
lobular
tumorsInfiltrating
carcinoma(scirrhous)60160ComedocarcinomaI01Paget's
ductal
disease (infiltrating)I01
a Central
or
peripheral
survival,
of the culture period in central
respectively.
b Comparison
6l.55,p
ofcentral
of
viable
soft
versus
cancer
cells
at
the
end
regions of the explants,
collagen-rich
tumors;
Xc2
< 0.001.
CComparison
Xe'
survival;
presence
or in peripheral
3.35,
not
of peripheral
survival; soft versus collagen-rich
tumors;
significant.
CANCER RESEARCH
VOL. 35
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1975 American Association for Cancer Research.
Estradiol-dependent
Effect of Hormones in Culture
Insulin, Prolactin, and Hydrocortisone. Explants cul
tured in hormone-free medium or in medium supplemented
with insulin, prolactin, or hydrocortisone, either alone or in
various combinations, failed to display obvious differences.
It is possible that, in occasional cases, insulin alone or in
combination
with prolactin induced some enlargement of
viable cancer cells and moderate loosening of the stroma.
Actually, survival during culture depended on the type of
tumors, whether soft or collagen rich (Table I).
In the soft tumors, survival was very satisfactory. The
architecture
of the explants was essentially the same as
before culture, and viable cells were present in central as
well as in peripheral areas, provided the diameter did not
exceed I mm (Fig. 1). Only the 5 medullary carcinomas,
which are tumors characterized
by heavy lymphocyte
infiltration, showed poor survival.
In striking contrast, survival of the collagen-rich tumors
was very poor. The only viable cells were located at the
extreme periphery of the explants; in deeper parts, the
tumor cells were completely pyknotic. The appearance of a
typical scirrhous infiltrating ductal carcinoma before and
after 6 days of culture is shown in Fig. 2, A and B. A few
bundles of cancer cells survived only at the very periphery of
the explants. In their depth, cell death usually spared
connective tissue cells and, when present, normal and
dyspiastic mammary tissues.
17@9-Estradiol. Supplementation
of the medium with
17fl-estradiol in combination with insulin or with insulin and
prolactin was tested on 23 breast cancers.
Three of the tumors were soft tumors: I carcinoma
simplex, I comedocarcinoma,
and I papillary carcinoma. In
none did estradiol improve survival of the cultures that was
already excellent under control conditions; neither did
steroids induce stromal changes.
Dramatic changes occurred, however, in most collagen
rich tumors. Twenty such tumors were studied: I comedo
carcinoma and 19 scirrhous infiltrating ductal carcinomas.
In 18 ofthem, culture in the presence ofestradiol resulted in
marked loosening and dissolution of the collagen, as well as
in significant improvement of cancer cell survival in the
depth of the explants. Fig. 2C illustrates this effect of
estradiol after 6 days of culture. Collagen had vanished and
cords of very healthy tumor cells were present throughout
the whole explant, interspersed with connective tissue cells.
Fig. 3A is a higher-magnification
micrograph of the same
explant. Occasional mitoses were seen in the tumor cords.
Complete dissolution of the collagen, as observed in Figs.
2C and 3A, did not occur in all cases. In some experiments,
estradiol-induced
lytic effects on the collagen were limited
to the area surrounding strands ofviable cancer cells. This is
illustrated by the experiment of Fig. 4. In this experiment,
after 7 days of culture in the absence of estradiol, viable
cancer cells located at the periphery of explants were
surrounded by intact collagen, whereas cancer cells located
more centrally were pyknotic (Fig. 4A). Culturing with
added estradiol for 4 or 7 days resulted in a progressive
digestion of the collagen around cords of living cells (Fig. 4,
B and C). Special
AUGUST
staining
of the material
indicated
that the
Collagenolysis
in Breast Cancer
lytic effects described here were indeed exerted on the
collagen itself. Lytic effects on the collagen were not
observed far from epithelial cancer cells or in explants that,
by random occurrence during preparation, did not contain
such cells. The minimum concentration of estrogen effect
ing these changes was usually I ng/ml but ranged from 0.1
to 10 ng/ml. Concentrations
as high as 10 @sg/mldid not
appear distinctly noxious, but the tests were too few to reach
valid conclusions. Failure to respond to estrogen at any
concentration occurred in only 2 cases.
Estradiol need not be present for the whole culture period
to insure cell survival in the depth of scirrhous explants.
This was demonstrated
in 5 experiments where explants
from scirrhous carcinomas were cultured for 7 days in a
medium containing insulin, prolactin, and 17@9-estradiol, I
ng/ml, and then cultured for an additional 7 days in the
same medium without estradiol. At the end of each 7-day
period, examination of the explants revealed an excellent
survival of cancer cells in cords surrounded by a loosened
stroma. Figure 4D represents a culture at the end of the 2nd,
estradiol-free period.
Effect of Collagenase
Explants from 25 tumors were cultured for an initial
period of 24 to 48 hr in the presence of added collagenase.
After washing out collagenase,
they were then further
cultured under the usual conditions.
Hormones,
when
present, were added at the start of culture. Of these 35
tumors, 8 were soft: 4 cases of carcinoma simplex, 2
papillary carcinomas,
1 comedocarcinoma,
and I colloid
carcinoma. The 27 others were collagen-rich tumors: 25
scirrhous infiltrating ductal carcinomas,
1 infiltrating Pa
get's disease, and I comedocarcinoma.
At the end ofthe 1or 2-day period of culture with collagenase, histological
examination disclosed that complete dissolution of collagen
had occurred. Tumor cells and connective tissue cells
appeared intact and settled on the culture support, whether
Millipore filter or agar gel, forming flattened explants of
very high cellular density. When further cultured for periods
ranging from 5 to 13 days in fresh medium with or without
added hormones, the cells maintained essentially the same
pattern. With respect to the soft tumors, cell survival was
not improved over that in explants not initially treated with
collagenase, where survival was already excellent. In sharp
contrast, scirrhous tumors cultured after an initial treat
ment with collagenase were considerably improved over
their controls. In each case the entire explant was composed
of viable tumor cells and connective tissue cells (Figs. 2D
and 3B). Pyknosis was very infrequent. The tumor cells
arranged themselves as epithelial cords, allowing clear
distinction between tumor and connective tissue. Mainte
nance of these explants did not seem grossly influenced by
added hormones. As already described, control explants
without initial culture with collagenase showed only poor
peripheral survival of tumor cells (Fig. 2B).
Tissue Estrogen Receptors
Samples
of tumor tissue were saved before culture
1975
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1975 American Association for Cancer Research.
for
2041
f-C.
Heuson et a!.
estrogen receptors assay in all cases but I . Specific receptors
were found present in 26 of 32 soft tumors (81%) and in 47
of 61 collagen-rich tumors (79%). The proportion of recep
tor-positive tumors and the mean concentration of receptors
were essentially the same in all histological types ofTable I.
In an experiment described above, l7fl-estradiol was found
to improve survival of cultured scirrhous carcinomas in 18
of 20 cases. Seventy-two % of these 18 tumors contained
receptors. Both tumors found unaffected by estradiol in
culture
were receptor
positive.
DISCUSSION
Whereas successful organ cultures of soft, collagen-poor
types of breast cancer were reported by various laboratories
(34, 36), this paper describes the 1st fully successful attempt
at maintaining scirrhous ductal carcinomas in long-term
organ cultures. This success may be ascribed partly to
technical improvements
and partly to the use of proper
hormonal supplementation.
One difficulty encountered in culturing scirrhous breast
cancer lies in the preparation of the explants. Extreme care
is required during the cutting procedure in order to avoid
shearing and crushing of the cells engulfed in the dense
collagen matrix. The best results were obtained in our hands
by the use of crossed razor blades selected for the sharpness
and hardness of their cutting edge and operated on a
solidified paraffin surface. Explants prepared by this proce
dure even from very hard collagen-rich tumors showed
excellent survival abilities under appropriate conditions.
Survival of the soft tumors of whatever histological type
was satisfactory in unsupplemented
synthetic culture me
dium. Cords of cells at the very periphery of scirrhous
explants also survived, but those located in their depth died.
Supplementation
of the medium with various combinations
of insulin, prolactin, and hydrocortisone
seemed to effect
in some cases minor changes that could not be accurately
defined by histological
methods. It would thus appear
that mere survival of breast cancer cells in soft tumors or at
the periphery of the scirrhous explants is only slightly, if at
all, hormone dependent in vitro. It is quite unlikely that
explants would contain enough endogenously
produced
hormones at the time of explantation to assure cell survival
for 14 days. Survival in excess of this duration was not
investigated.
In scirrhous carcinomas, normal or benign hyperplastic
mammary epithelial cells were able to survive in the depth
of the explants, while cancer cells survived only at their
periphery. This may indicate that the cancer cells have more
stringent requirements for nutrients or oxygen than does the
normal or hyperplastic
epithelium and that the dense
collagen is the factor responsible for the shortage of
nutrients or oxygen available to the cancer cells inside these
scirrhous explants.
The experiments reported here seem to suggest a new
effect of estradiol on breast cancer. Other authors (4, 30)
have reported that estrogen could improve the survival of
short-term organ cultures of human breast cancers in a
2042
small proportion of cases. In the present study, estradiol in
combination with insulin or with insulin and prolactin was
found to permit the survival of breast cancer cells in the
depth of scirrhous explants during long-term culture in most
cases. Furthermore,
this estradiol-induced
survival was
accompanied by lysis of the collagen around the cancer cell
cords and, in some cases, total dissolution of the collagen in
the whole explant. One possible explanation for these effects
is that estradiol stimulates the connective tissue components
to greater activity in a putative role of supporting life of the
cancer cells. The importance of stromal-epithelial
interac
tions has been well documented (12), in particular during
embryonic development of the mammary gland (19). An
other explanation is that estradiol would be required at the
tumor cell level to insure survival. In both hypotheses
assuming inherent collagenolytic properties of the tumor
cells, their survival would be the cause of the observed lytic
effects on the surrounding collagen. A 3rd explanation is
that estradiol would stimulate elaboration or activation in
breast cancer cells of collagenolytic enzymes. The partial or
total destruction of the surrounding collagen would then
suppress the shortage of nutrients or oxygen and thereby
allow cell survival. This 3rd explanation
is the most
plausible for 1 major reason: breast cancer cells do not seem
to be estrogen dependent for survival in culture, provided
they are not surrounded by dense collagen. Thus, good
survival
was
obtained
without
added
estradiol
under
the
following circumstances having in common the absence of a
collagen barrier: (a) when the cells were located at the
periphery of the explant; (b) when centrally located cells had
already digested the surrounding collagen during an initial
culture with added estradiol; (c) when the collagen had been
eliminated by added exogenous collagenase during the 1st 1
or 2 days of culture.
From these experiments,
one can perhaps deduce a
tentative hypothesis stating that survival of breast cancer
cells in scirrhous ductal carcinoma depends in organ culture
upon elaboration
or activation by these cells, under the
influence of estradiol, of enzymes dissolving the collagen
matrix, thus allowing diffusion ofthe necessary nutrients. In
all experiments, estradiol was added in combination with
insulin or insulin and prolactin. It cannot therefore be
answered whether estradiol alone would produce the same
effect. Nor can it be said that estradiol is specific in
triggering this enzymatic process. In rare instances, it was
suspected that insulin or insulin plus prolactin induced some
loosening of the stroma adjacent to peripherally located
cords of cancer cells. Yet, this effect was insufficient to
permit survival of centrally located cancer cells. Progester
one and other steroids were not tested in the present
experiments. Additional studies are required, therefore, to
draw a complete picture of the hormonal requirements of
human breast cancer in organ culture.
Since specific tissue estrogen receptors are the supposed
mediators of the metabolic effects of estrogens in their
target tissues, they were assayed in all breast cancer samples
subjected to culture and found present in about 80%. In the
estradiol culture experiments, 5 of the 18 estradiol-respon
sive tumors were devoid of receptors while the 2 unrespon
CANCER RESEARCH
VOL. 35
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1975 American Association for Cancer Research.
Estradiol-dependent
sive ones were receptor positive. This discrepancy might be
only apparent, because, as reported elsewhere (22), negative
receptor assays might be due to methodological
artifacts.
The present data support this view and further suggest that
most if not all breast cancers, at least of the scirrhous type,
have estrogen-sensitive
metabolic reactions.
The results reported here on the action of estradiol on
collagen-rich breast cancer in vitro are in keeping with the
conclusions of Elliot and Turner (8) on the development of
the normal mammary gland. These authors described the
presence of a spreading factor in the mammary gland of
different animal species during parenchymal growth. This
factor had the properties of an enzyme but, unlike other
tissue-spreading
factors, it was not hyaluronidase.
The
possibility was suggested that it could act on substrates like
collagen. The authors' opinion was “thatthe spreading
factor is either elaborated
or activated in the growing
mammary gland cells by hormones (especially estrogens)
which initiate both duct and lobule-alveolar growth of the
mammary glands. The spreading factor may then break
down the connective tissue in some manner allowing the
rapid forward growth of the mammary ducts and lobes.―
The question arises whether this hypothesis may apply to
breast cancer in vivo. In the present culture experiments, the
minimum effective concentrations
of estradiol were quite
low, ranging from 100 to 10,000 pg/ml. The lower concen
trations are not far above physiological levels (15 pg/mI
in the serum of postmenopausal
women). In earlier in
vivo studies,
Emerson
et a!. (9) described
the loosening
of
connective tissue stroma around foci of active cancer in
breast cancer patients subjected to estrogen therapy. The
estrogen stilbestrol was administered
at high doses in
order to try to inhibit tumor growth. In their careful de
scription, the authors observed that “theimmediate cause
of the softening of the tumor was in each case a loosening of
the connective tissue surrounding the tumor cells.― Such
loosening appeared to result from partial dissolution of the
collagen, for there was an overall depletion of collagen. It
mainly occurred in the connective tissue immediately adja
cent to the epithelial elements.
Several authors reported collagenolytic enzyme activity
in various neoplasms, as demonstrated in homogenates of a
tumor (13) or in organ culture experiments (6, 26, 28, 33),
suggesting that membrane-bound
collagenase could confer
upon tumor cells a distinct degree of mobility and invasive
ness. Such studies were conducted on a few cultures of
breast cancer with negative results (6, 26). No hormone
supplementation
was tried in these works and, as stated by
Riley and Peacock (26), “cell
death results in instantaneous
cessation of the enzyme activity, and the failure of many
tissue fragments to produce collagenolytic enzyme could be
failure of tissue culture methods to sustain life for 5 days.―
Our own data, bearing on viable long-term cultures, strongly
suggest that human breast cancer, at least of the collagen
rich types, is able to produce collagenolytic enzymes, but
that this production is estrogen dependent. Moreover, this
estrogen-dependent
coilagenolytic activity could well be an
important
part of the mechanism
underlying hormone
dependency of human breast cancer growth.
AUGUST
Collagenolysis
in Breast Cancer
ACKNOWLEDGMENTS
We are grateful to the surgeons and to the pathologists ofthe Institut J.
Bordet for kindly providing the specimens. We are indebted to S. Verset
and to M. C. Deboel for their skillful technical assistance. Our thanks are
due to Dr. P. G. Condliffe from the National Institutes of Health, who
generously
provided ovine prolactin.
REFERENCES
I. Archer, F. L. Normal and Neoplastic Human Tissue in Organ Culture.
Arch. Pathol., 85: 62-71, 1968.
2. Barker, J. R., and Richmond,
C. Human
Breast Carcinoma
Culture:
The Effects of Hormones. Brit. J. Surg., 58: 732-734, 1971.
3. Burstein, N. A., Kjellberg, R. N., Raker, J. W., and Schmidek, H. H.
Human Carcinoma of the Breast, in vitro: The Effect of Hormones.
Cancer,27: 1112-1116,1971.
4. Chayen, J., Altmann, F. P., Bitensky, L., and Daly, J. R. Response of
Human Breast-Cancer Tissue to Steroid Hormones in vitro. Lancet, I.'
868-870, 1970.
5. Dc Souza, I., Morgan, L., Lewis, U. J., Raggatt, P. R., Salih, H., and
Hobbs, J. R. Growth Hormone Dependence among Human Breast
Cancers. Lancet, 2: 182- 184, 1974.
6. Dresden, M. H., Heilman, S. A., and Schmidt, J. D. Collagenolytic
Enzymes in Human Neoplasms. Cancer Res., 32: 993-996, 1972.
7. Elias, J. J., and Rivera, E. Comparison of the Responses of Normal,
Precancerous, and Neoplastic Mouse Mammary Tissues to Hormones
in Vitro.Cancer Res.,19:505-511,1959.
8. Elliot, J. R., and Turner, C. W. The Mammary Gland Spreading
Factor. Res. Bull. Univ. Missouri CoIl. Agri. 537: 1-52, 1953.
9. Emerson, W. J., Kennedy, B. J., Graham, J. N., and Nathanson, I. T.
Pathology of Primary and Recurrent Carcinoma of the Human Breast
after Administration
of Steroid Hormones. Cancer, 6: 641 -670, 1953.
10. Flaxel, J., and Wellings, S. R. Toxic Effects ofTestosterone
on Organ
Cultures of Mammary Carcinoma Cells of C3H/CrgI mice. Federa
tionProc.,22:331, 1963.
11. Foley, J. F., and Aftonomos, B. T. Growth of Human Breast
Neoplasms in Cell Culture. J. Natl. Cancer Inst.. 34: 217229, 1965.
12. Grobstein, C. Mechanism of Organogenetic
Tissue Interaction.
J.
NatI. Cancer Inst. Monograph, 26: 279-299, 1967.
13. Harris, E. D., Faulkner, C. S., and Wood, S. Collagenase in
Carcinoma
1972.
Cells. Biochem.
Biophys. Res. Commun.,
48: 1247 1253,
14. Heuson, J. C., Coune, A., and Heimann, R. Cell Proliferation Induced
by Insulin in Organ Culture of Rat Mammary Carcinoma. Exptl. Cell
Res., 45: 351 -360, 1967.
15. Heuson, J. C., and Legros, N. Study ofthe Growth-Promoting
Effect
of Insulin in Relation to Carbohydrate
Metabolism in Organ Culture
of Rat Mammary Carcinoma. European J. Cancer, 4: I -7, 1968.
16. Heuson, J. C., and Legros, N. Effect of Insulin on DNA Synthesis and
DNA Polymerase Activity in Organ Culture of Rat Mammary
Carcinoma and the Influence of Insulin Pretreatment
and of Alloxan
Diabetes. Cancer Res., 31: 59-65, 1971.
17. Heuson, J. C., and Tagnon, H. J. Androgen Dependence of Breast
Cancers. Lancet. 2: 203-204, 1973.
18. Hobbs, J. R., Salih, H., Flax, H., and Brander, W. Prolactin
Dependence among Human Breast Cancers. In: J. L. Pasteels and C.
Robyn (eds.), Human Prolactin, pp. 147- 155. Amsterdam: Excerpta
Medica,
1973.
19. Kratochvil, K. Organ Specificity in Mesenchymal Induction Demon
strated in the Embryonic
Development
of the Mammary
Gland of the
Mouse. Develop. Biol., 20: 141- 153, 1969.
20. Kuzma, J. F. The Breast. In: A. D. Andersen
(ed), Pathology.
1975
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1975 American Association for Cancer Research.
Ed. 2,
2043
J@C. Heuson et a!.
pp. 1102-1119. London: Henry Kimpton, 1953.
21. Lagios, M. D. Hormonally Enhanced Proliferation of Human Breast
Cancer in Organ Culture. Oncology, 29: 22- 33, 1974.
22. Leclercq, G., Heuson, J. C., Deboel, M. C., and Mattheiem, W. H.
Oestrogen Receptors in Breast Cancer. A Changing Concept. Brit.
Med. J.,I:185-189,
1975.
23. Leclercq, G., Heuson, J. C., Schoenfeld,
R., Mattheiem,
W. H., and
Tagnon, H. J. Estrogen Receptor in Human Breast Cancer. European
J. Cancer, 9. 665-673, 1973.
24. Pearse, A. G. E. Histochemistry.
Theoretical and Applied. London:
Churchill, Ltd., 1960.
25. Riley, P. A., Latter, T., and Sutton, P. M. Hormone Assays on
Breast-Tumour
Cultures. Lancet, 2.' 818-820, 1973.
26. Riley, W. B., Jr., and Peacock, E. E., Jr. Identification,
Distribution
and Significance of a Collagenolytic Enzyme in Human Tissues. Proc.
221:259-260, 1969.
29. Salih, H., Flax, H.,
Dependence
Brander,
W., and
Breast Cancers.
Hobbs,
Lancet,
J. R. Prolactin
2: 1103- 1105, 1972.
30. Salih, H., Flax, H., and Hobbs, J. R. In Vitro Estrogen Sensitivity of
Breast Cancer Tissue as a Possible Screening Method for Hormonal
Treatment. Lancet, 1: 1198-1202, 1972.
3 1. Stewart, F. W. Tumors of the Breast. In: Atlas of Turnpr Pathology,
Section 9, Fascicle 34, pp. 5- I 14. Washington: The Armed Forces
Institute of Pathology, 1950.
32. Stoll, B. A. Investigation ofOrgan Culture as an Aid to the Hormonal
Management of Breast Cancer. Cancer, 25: 1228—1233, 1970.
33. Taylor, A. C., Levy, B. M., and Simpson, J. W. Collagenolytic
Activity of Sarcoma Tissues in Culture. Nature, 228: 366-367, 1970.
34. Wellings, S. R., and Jentoft, V. L. Organ Cultures of Normal,
Dysplastic,
Soc. Exptl. Biol. Med., 124: 207-210, 1967.
27. Rivera, E. M., Elias, J. J., Bern, H. A., Napalkov,
in Human
Hyperplastic,
and Neoplastic
Human
Mammary
Tissues.
J. Nail. Cancer Inst., 49: 329-338, 1972.
P., and Pitelka, D.
R. Toxic Effects of Steroid Hormones on Organ Cultures of Mouse
Mammary Tumors, with a Comment on the Occurrence of Viral
Inclusion Bodies. J. NatI. Cancer Inst., 31: 671-687, 1963.
28. Robertson, D. M., and Williams, D. C. In Vitro Evidence of Neutral
Collagenase Activity in an Invasive Mammalian Tumour. Nature,
35. Welsch, C. W., and Rivera, E. M. Differential Effects of Estrogen and
Prolactin on DNA Synthesis in Organ Cultures of DMBA-induced
Rat Mammary Carcinoma.
Proc. Soc. Exptl. Biol. Med., 139:
623-626, 1972.
36. Willcox, P. A., and Thomas, G. H. Oestrogen Metabolism in Cultured
Human Breast Tumours. Brit. J. Cancer, 26: 453-460, 1972.
Fig. I . Culture of a soft infiltrating ductal carcinoma. H & E. A , initial condition. The tumor (large cells) invades nontumoral mammary tissue (small
cells). Connective tissue is sparse, with loose collagen fibrils. B, central region of a 5-day organ culture of the same tumor. Culture in chemically defined
Medium 199 supplemented with insulin (10 @g/ml)and hydrocortisone (I @g/ml).As a general rule, tumor cells are well preserved. Collagen liquefaction
is not necessary for the maintenance of viable tumor within the whole explant.
Fig. 2. Culture of a scirrhous infiltrating ductal carcinoma. H & E. A , initial condition. Tumor is characterized by dense collagen stroma. The explants
were prepared without apparent loss of tumor cells. B, peripheral area of 6-day organ culture in Medium 199 + insulin ( 10 @zg/ml)+ prolactin (5 @g/ml).
A cluster of viable tumor cells is present (arrow). Well-preserved fibroblasts and intact collagen bundles can be found even in deeper locations (5). C,
same conditions as for B, but with further supplementation of the medium with 17$-estradiol (0.3 ng/ml). Nearly complete disappearance of collagen
resulting in increased cellular density throughout the whole explant is observed. Under such conditions, viable tumor cords are present in central parts of
the explant (arrows). D, same conditions as for B, except that the initial 48-hr period of culture was carried out in the presence of collagenase (4 mg/mI).
The collagen is totally dissolved. The micrograph shows the whole explant with its very high cellular density, probably explained in part by the
disappearance of the collagen bulk. Most viable cells are epithelial tumor cords (for higher magnification, see Fig. 3B). Collagenase treatment was
sufficient
for the maintenance
of viable tumor cells in the absence of estrogen.
Fig. 3. Higher-magnification micrographs showing the effects of estradiol and collagenase. H & E. A, same conditions as for Fig. 2C. Detail of the
central area of an explant exposed to I7@-estradiol (0.3 ng/ml). A cluster of viable, healthy tumor cells is observed. It is surrounded by connective tissue
with increased cellular density and a very loose network ofthin collagen fibrils. B, same conditions as for Fig. 2D. As a result ofcollagenase treatment, the
collagen is totally dissolved. Viable tumor cells were maintained in the absence of estrogen. They are easily recognized from fibroblasts as a cluster of
densely packed epithelial cells.
Fig. 4. Details of culture of a scirrhous
infiltrating
ductal carcinoma.
H & E. A, 7-day culture in medium containing
insulin (10 @.sg/mI)and ovine
prolactin (5 @tg/ml).Peripheral area of an explant; viable tumor cells are still present. No liquefaction of collagen around them is observed. In deeper
locations, the tumor is pyknotic (arrows). Such deeper locations are to be compared to B, C, and D for survival of tumor in the central area of the explants.
B, 4-day culture in the same medium but further supplemented with I 7@-estradiol (0. I ng/ml). Central area of an explant; survival of tumor is observed,
with dissolution of the collagen in close contact with tumor cells. Survival of this tumor cord should be compared to the necrotic condition in the deeper
locations indicated with arrows on A . C, 7-day culture in the same medium as for B. Central area of an explant; extensive digestion of collagen is visible.
The picture was focused on stroma. Only tumor cells can be clearly recognized in the lower part of the central cluster. They look healthy, with prominent
nucleoli. D, 14-day culture in the same medium as for B and C, except that l7@-estradiol (0. 1 ng/ml) was added only during the initial 7-day period of
culture. Central area of an explant. Once sufficient loosening of stroma is obtained in such a central area, estradiol is no longer necessary for the
maintenance of very healthy tumor cells. The presence of prominent nucleoli, suggesting ribosomal synthesis, is proof of their good condition.
2044
CANCER RESEARCH
VOL. 35
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1975 American Association for Cancer Research.
Estradiol-dependent
Co!!agenolysis
in Breast Cancer
@-,flI
@
,.
,-.‘
.
.@
I—;.
AUGUST
@.
4@
*
1975
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1975 American Association for Cancer Research.
2045
@
;@.
@
._‘._.b;.
@
,‘
;•@
.,@.L
@
..
@
.-
S.
,.
@.
.@
v'),'.,
!‘@
$
..@‘
•,@‘‘.4'r-,..:@@
.
@
I
5,-
‘$1
,. ,.
I
.@
,4
@
...
•:@r.'
I?
@
•,
•
@
@@
@
-@
p
“I,
t
#
I,
q
.,,
.@.: @\@\
,@
@
._p@
‘
@
@
“‘I.'
@
‘@d•@p'.
@
‘
@
@
p@
@.‘
‘ S
‘@!t
@
I
:
‘9&.@',@
‘@.
,
@
C,
;
S
@
@w
@
.
@
:
‘•@@
@.
$ •.
S
•
.-.
2046
,
?@
@r
@.e-4&
,
@,&‘- ‘,.
:@a;@,
,,;@“
‘,‘‘
.@
,
;;:,
•
@
,@
.:@
.-‘—
@,I... @.
..
. f.,@,
@%.,
•
‘@1I@
_
‘my',
:,@‘@Q'
_S_.._
CANCER RESEARCH
VOL. 35
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1975 American Association for Cancer Research.
Estradiol-dependent
AUGUST
Collageno!ysis
in Breast Cancer
1975
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1975 American Association for Cancer Research.
2047
f-C.
Heuson et a!.
@
@.‘
@
.
p
.@
r
.4'
:@
a
4
@
@
10 @.LTh
:@“ - ‘k
2048
,—
jO@L11@—
..
‘4
@.
-
CANCER
RESEARCH
VOL. 35
Downloaded from cancerres.aacrjournals.org on June 18, 2017. © 1975 American Association for Cancer Research.
Estradiol-dependent Collagenolytic Enzyme Activity in
Long-Term Organ Culture of Human Breast Cancer
Jean-Claude Heuson, Jean-Lambert Pasteels, Nicole Legros, et al.
Cancer Res 1975;35:2039-2048.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/35/8/2039
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 18, 2017. © 1975 American Association for Cancer Research.