nogenic hydrocarbons, DMBA, 3-methylcholanthrene and di benz(a,h)anthracene, transform mouse mammary gland in whole organ culture. It was therefore ...
[CANCER RESEARCH 39, 1784-1792, May 1979] 0008-5472/79/0039-0000$02.00
Transformation of Cultured Mouse Mammary Glands by Aromatic Amines and Amidesand Their Derivatives1 Quentin J. Tonelli, R. Philip Custer, and Sam Sorof2 The Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, Pennsylvania 19111
ABSTRACT Mouse mammary glands in whole organ culture, previously demonstrated by others to undergo lobuloalveolar develop ment, functional differentiation, and glandular involution, are transformable by carcinogenic aryl amines and amides and their derivatives. The transformation, which involves an escape from the hormonal controls of these processes, results in the formation of nodule-like alveolar lesions that are morphologi cally similar to presumptive preneoplastic lesions, termed hy perpiastic alveolar nodules, which others have found to arise in mice during viral and chemical mammary tumorigenesis. The transformation system discriminates between carcinogens and noncarcinogens of analogous structures in the fluorenyl and naphthylenic chemical series, and is sensitively responsive to the presence of the amino or amide group or their derivatives. In a series of fluorenyl carcinogens, N-2-fluorenylacetamide displayed very high transforming activity > N-hydroxy-N-2fluorenylacetamide (high activity), > N-acetoxy-N-2-fluoreny lacetamide (moderate activity), and > 2-fluoreneamine and 2nitrofluorene (low activity). In contrast, the noncarcinogen, fluorene, had little if any transforming activity. The two naphthy lenic carcinogens, 1-naphthylammneand 2-naphthylamine, both displayed moderate transforming activity, while the noncarcin ogen, naphthalene,
had no activity. Control glands treated only
with the vehicle (dimethyl sulfoxide) were not transformed. Considerable structural changes at the cellular level were caused by the carcinogens, much less so by the noncarcino gens, and essentially none by the vehicle. It remains to be determined whether the glands transformed by these carcino gens can progress to tumors in vivo. The transformation by the carcinogenic aryl amines and amides and their derivatives (this report) and likewise the malignant transformation by the car cinogenic polycyclic aromatic hydrocarbons (Banerjee and coworkers, Proc. Am. Assoc. Cancer Res., 20: 153, 1979) suggest
that mouse
mammary
gland
in whole
organ
culture
may be transformable by a broad spectrum of chemical carcin ogens. The prospective
linking of this transformation
system to
the considerable body of information presently known con cerning these carcinogens may significantly increase the cur rent understanding of the actions of chemical carcinogens on mammary gland and on epithelial tissues in general.
Mouse mammary gland in whole organ culture undergoes in
part
by
NIH
Grants
CA-21
522
and
CA-05945;
Institutional
Grants CA-09035, CA-06927, and RR-05539 from NIH; and an appropriation from the Commonwealth of Pennsylvania. Presented in part at the 18th Annual Meeting of the American Society for Cell Biology, San Antonio, Texas, November 6, 1978 (35). 2 To
whom
requests
for
reprints
should
be
addressed,
at
The
Institute
for
Cancer Research, Fox Chase Cancer Center, 7701 Burholme Avenue, Fox Chase, Philadelphia, Pa. 191 11. Received September 21, 1978; accepted December 27, 1978.
1784
in vivo. Hormonalmanipulationof this in vitro system brings about full lobuloalveolar development (19), lactogenesis, and alveolar regression (38). In addition, carcinogenic polycyclic aromatic hydrocarbons cause an epithelial cell transformation in the cultured whole mammary gland involving an escape from hormonal regulation of alveolar development. The resultant 4
are
morphologically
similar
to
presumptive
preneo
plastic lesions, termed HAN (5, 20, 21), which arise in mice during viral and chemical mammary tumorigenesis (12, 13, 16, 22, 29). Furthermore,
the carcinogenic
polycyclic
aromatic
hydrocarbon, DMBA, causes a malignant transformation of the epithelial cells of the mammary glands in whole organ culture. Fragments or dissociated cells from such transformed whole mammary glands, after transplantation into mammary gland free (cleared) fat pads of virgin mice, give rise to transplantable hyperactive outgrowths (like those of preneoplastic glands) and lobular adenocarcinoma.5 The mouse mammary gland in whole organ culture thus appears to provide an in vitro model transformation system that is exceptionally well suited to the study of the possible roles of aberrant organ development and differentiation in the chemical carcinogenesis of epithelial tis sues in general,
and of mammary
gland in particular.
Certain aromatic amines and amides cause cancer of the mammary glands and other organs in several rodent species, including mouse (1—3,6, 10, 23—25,26, 28, 34, 36, 37). A considerable body of information has been accumulated on the metabolism and activation of these carcinogens (1, 2, 6, 10, 23, 24, 36, 37). In search of the direct effects of specific
carcinogen-macromolecular interactions on organ develop ment and differentiation and the possible involvement of these effects in carcinogenesis, we have applied the transformation model of the cultured whole mammary gland of mice to the study of the carcinogenic aryl amines and amides and their activated derivatives. MATERIALS AND METHODS Mice. Female BALB/c mice, a strain with only endogenous MuMTV (see Ref. 27 and ‘ ‘Discussion' ‘), were bred in-house. 3 The
INTRODUCTION
I Supported
processes of organ development, functional differentiation, and alveolar involution (regression) which mimic those which occur
abbreviations
used
are:
NLAL,
nodule-like
alveolar
lesions
produced
in
whole mammary organ culture; HAN, hyperpiastic alveolar nodules produced in mammary glands in vivo; DMBA, 7,12-dimethylbenzo(a)anthracene; MuMTV, munne mammary tumor virus; DMSO, dimethyl sulfoxide; FAA, N-2-fluorenyla cetamide; N-OH-FAA, N-hydroxy-N-2-fiuorenylacetamlde; 2-FA, 2-fluorenylam me; 2-NF, 2-nftrofluorene; N-AcO-FAA, N-acetoxy-N-2-fluorenylacetamide; 1NA, 1-naphthylamine;2-NA, 2-naphthylamine. 4 The
NLAL
are
the
nodular
alveolar
lesions
that
are
the
product
of
the
escape
from the hormonal controls of alveolar development in mouse mammary gland in whole organ culture. This change to hormone independence, which is brought about by the chemical carcinogens, constitutes the operational definition of chemical transformation in the entire mammary gland In organ culture. 5 N.
T.
Telang,
A.
B.
Kundu,
A.
P.
lyer,
and
M.
R.
Banerjee,
personal
communication, andProc.Am.Assoc.CancerRes.,20: 153, 1979.
CANCERRESEARCHVOL. 39
Transformation of Cultured Mammary Gland In the normal mammarygland of the 3- to 4-week-old mouse, the epithelial rudiment is restricted to the nipple region of the mammary fat pad (12). At that age, the mice were hormonally primed by s.c. injections of I mg progesterone and 1 @g fi estradiol daily for 9 days (19, 31). The parenchyma then consists mainly of ducts and terminal alveolar buds and extends to the major portion of the mammary fat pad (31 ). On the day following the last injection, the mice were sacrificed by cervical dislocation. Both second thoracic mammary glands were care fully excised and in their entirety were immediately cultured (below). Protocol of Culturing and Transformation. The procedure for the culturing of whole mammary glands of mice has been described by Banerjee et a!. (4). Briefly, both second thoracic mammary glands of each mouse were individually floated on exhaustively water-extracted Dacron rafts in the same tissue culture plastic dish (35 mm; Falcon Division, Becton Dickinson Co., Oxnard, Calif.) containing 2 ml of medium (1 mI/gland). The medium did not contain serum, and it consistedof Way mouth MB752/1 supplemented with L-glutamine (350 @g/ml), potassium penicillin 0 (35 @sg/ml),and the appropriate steroid and polypeptide hormones (below). Cultures were maintained at 37°in a water-saturated atmosphere of 95% oxygen and
.
evidencing an absence or paucity of alveolar buds. Preparation of Histological Sections. The cultured glands were transferred from toluene to xylene, embedded in paraffin, sectioned at 5 @m, and stained with hematoxylin and eosin. Selected sections were also stained with Masson's trichrome and periodic acid-Schiff reagent. The plane of section through most glands was longitudinal and included the common duct below the nipple. Parameters for comparison included quali tative and quantitative changes in ductal and alveololobular epithelium, size and proliferation of ducts and lobules, quantity and character of secretions, and alterations in the interstices. Most groups contained 9 or 10 glands, one 5, another 6, and control glands 19. Each item was scored between 0 and 4+. Chemicals. FAA (191 —1 92°;Fluka Chemical Co., Buchs, Switzerland) and fluorene (1 15—1 16°;Aldrich Chemical Co., Milwaukee, Wis.) migrated as single spots in thin-layer chro matography on activated silica gel (containing fluorescent in dicator; Eastman Kodak Co., Rochester, N. V.) in chloroform: methanol (97:3) and in benzene:ethyl acetate (1 :1). The FAA and fluorene were indicated by their vendors to be > 99 and 98% pure, respectively.
N-OH-FAA (1 50—151 0) was gener
ously donated by Dr. P. D. Lotlikar of the Fels Research Institute. The sample absorbed maximally at 291 and 302 nm, 5% carbon dioxide in plastic compartments within water-jack in agreement with its reported properties (1 1). 2-FA, 2-NF, Neted incubators. AcO-FAA, 1-NA, and 2-NA were obtained from the Chemical Culturing of the second thoracic mammary gland in the above Repository of the National Cancer Institute Carcinogenesis medium containing insulin, prolactin, aldosterone, and hydro Research Program, and were indicated to be of > 99% purity. cortisone, each at 5 @sg/ml (development medium), for the first Commercially obtained DMBA (Eastman) and crystalline naph 9 days resultsin full lobuloalveolardevelopmentand functional thalene (J. T. Baker Chemical Co., Phillipsburg, N. J.) were differentiation (5, 21 , 38). On the third day of culturing, coin further purified by crystallization from absolute ethanol. The cident with the second wave of DNA synthesis (21 ), the medium DMBA and naphthalene(81 0) each migratedas singlespots in was replaced with mediumcontaining0. 1% DMSO (v/v, final) thin-layer chromatography in silica gel (above) in benzene: with or without (control) the test substance. The medium was methanol (19:1 ) and benzene:hexane (15:1 ). All compounds of renewed without DMSO and test substance 24 hr later and the fluorene and naphthalene series were added to cultures in routinely every 2 or 3 days thereafter. Levels of the nonpolar semidarkness from stock solutions of DMSO that were freshly test substances are presented in terms of @tg/ml/gland be prepared, except that of FAA, which was retained up to 1 day cause of the possible concentrating action of the mammary fat in darkness at 1—4°. DMBA was applied to cultures under pad. Following an additional 15 days (ninth to 24th day) in yellow light from stock DMSO solutions that were used fresh or medium containing insulin at 5 @tg/mlwithout other added similarly stored up to 1 month. Solid FAA and DMBA were hormones (regression medium) (4, 5, 38), the nontransformed maintained without light under air at 22°;N-OH-FAA, N-AcO alveoli were involuted (see below). In contrast, mammary FAA, 1-NA, and 2-NA under argon at —20°; and 2-FA, 2-NF, glands that had undergone transformation by chemical carcin fluorene, and naphthalene under argon at 22°. ogens had nodules of alveoli (NLAL) which were not regressed Insulin and hydrocortisone were obtained from Calbiochem by this protocol. (La Jolla, Calif.); prolactin, aldosterone, progesterone, 17$On the 24th day, the glands were fixed with acetic acid: estradiol, and L-glutamine were obtained from Sigma Chemical ethanol (1 :3, v/v), stained with alum carmine (4), and scored Co. (St. Louis, Mo.); Waymouth Medium MB 752/1 (lOx) and for NLAL and cytotoxicity under a dissecting microscope. The potassium penicillin G were obtained from Grand Island Biolog NLAL were revealed after involution of mammary gland alveoli ical Co. (Grand Island, N. V.); and DM80 was obtained from and appeared on a background of ducts and terminal buds. Burdick and Jackson (Muskegon, Mich.). Glands were grouped, coded, and scored for transformation without knowledge of the previous treatment of the glands. RESULTS Each experiment included test and control glands. The results from several matching experiments were combined. Based on Development of Mammary Glands In Whole Organ Culture. results presented in Table 1, test substances that brought In order to demonstrate the state of the morphological devel about transformation in more than about 75% of the glands opment of the mammary glands in vitro prior to their regression, were designated to have very high transforming activity; about representative control and FAA-treated glands were cultured 75 to 50%, high activity; about 50 to 25%, moderate activity; in the development medium for 9 days and then fixed and about 25 to 10%, low activity; and below about 10%, very low stained. After 9 days of development, glands treated with FAA or essentially no activity. Relative cytotoxicity was similarly were distinguishablefrom control glands. The control glands, rated on the basis of the percentage of the regressed glands which had been treated with only DM80 (vehicle) at Days 3 to
MAY 1979
1785
Q. J. Tone!li,
R. P. Custer,
and
S. Sorof
4, exhibitednumerousfullydevelopedlobuloalveolarstructures joined to wide ducts (Fig. 1a). On the other hand, the glands which had been subjected at Days 3 to 4 to 0. 1 @tg FAA per ml per gland were not as fully developed. They contained seem ingly fewer and smaller lobuloalveoli, and these were con nected to relatively thinner ducts (Fig. 1b). Thus, FAA was able to alter the 9-day course of the mammary glandular develop ment which occurred in response to full hormonal supplemen tation in development medium. Thereafter, all glands were routinely maintained in the regression medium from Days 9 through 24 in order to permit the involution of the nontrans formed lobuloalveolar structures and to allow the transformed glands to be revealed as those containing NLAL that persisted in the absence of added hormones other than insulin. Control Mammary Glands. Control glands, which were treated with DMSO without test substance at Days 3 to 4 according to the 24-day protocol, were similar to normal re gressed glands after lactation in vivo. All control glands were not transformed, and only a few glands showed effects of cytotoxicity. A control gland demonstrating the absence of NLAL and of cytotoxicity is presented in Fig. 1c. The compila tions of the control glands in Table 1 represent summations from different experiments relevant to each carcinogen. Transformation by Reference Carcinogen, DMBA. Baner jee et a!. (5, 20, 21) have previously reported that the carci nogenic hydrocarbons, DMBA, 3-methylcholanthrene and di benz(a,h)anthracene, transform mouse mammary gland in whole organ culture. It was therefore desirable to utilize one of these carcinogenic polycyclic aromatic hydrocarbons as a reference transforming agent. In the present study, DMBA exhibited a high transforming activity. The rating is based on the transformation in 64% of the glands and is supported by the presence of 6 NLAL/transformed gland (Table 1). This high transforming activity by DMBA was exhibited at 0.001 @tg/ml/ gland (3.9 pmol/ml/gland), which is 1000-fold lower than the lowest concentration of DMBA (1 .0 pg/mI/gland) previously reported to transform mammary glands of mice in whole organ culture (5, 20, 21). In our experiments, 1.0 @tg/ml/gIandwas very cytotoxic. Even the level of 0.001 @zg/ml/glandin the present study was toxic in 22% of the glands. The greater activity of DMBA in the present investigation, compared to that previously reported, may in part reside in the greater purity and preservation of our DMBA. In any case, the present incidence of transformation and cytotoxicity at the above concentration of DMBA serves as a reference by which to compare the different carcinogens, and to compare our results with those of other studies. Transformation by FAA and Derivatives. FAA was very active in transforming mouse mammary glands. Fig. 1d shows such a transformed gland. The rating of very high transforming activity is based on the finding that up to 83% of the glands were transformed, and it is supported by the up to 7 NLAL/ transformed gland (Table 1). A 1000-fold range of concentra tion of the carcinogen was studied, i.e., from 0.001 to 1.0 sg/ mI/gland, or 4.5 pmol to 4.5 nmol/mI/gland. The concentra tion of FAA that yielded the highest incidence of mammary gland transformation is 1000-fold higher than the above mdi cated active concentration of DMBA, and it is 10-fold higher than the concentration of FAA that produced the greatest number of NLAL per transformed gland. Examination of the
1786
results in Table 1 suggests that below the concentration of 0.1 @zg/ml/gIandthe concentration of carcinogen limited the pro duction of NLAL, whereas above that concentration toxicity was limiting. Thus, no toxicity was observed at the optimum concentration (0.1 @sg/ml/gland)that resulted in the maximum number of NLAL per transformed gland. At higher concentra tions the number of NLAL per transformed gland progressively decreased. Nevertheless, at the higher concentrations, up to 83% of all the glands showed at least 1 NLAL and thus were designated as transformed. Fig. 1e shows a gland evidencing toxicity due to FAA. The mammary gland-transforming activity of FAA decreased with the increasing electrophilicity of its derivatives. Thus, while FAA exhibited very high transforming activity, the moderately activated carcinogen, N-OH-FAA (25), was somewhat less effective with a rating of high transforming activity, and the carcinogenic, strongly electrophilic hydroxamic acetyl ester, N-AcO-FAA (6), was only moderately active. The latter ratings of transforming activity are based on the finding that N-OH FAA transformed up to 62% of the glands and that N-AcO-FAA transformed up to 50% of the glands (Table 1). Mammary glands transformed by these activated carcinogens were mor phologically similar to that of Fig. 1d. Both carcinogens were similarly highly cytotoxic atthe highest concentration examined (1.0 @tg/ml/gland), and both were more toxic than was the parent FAA. The moderately strong carcinogens, 2-FA and 2-NF (26, 28), exhibited low transforming activity. The rating of low transform ing activity results from the finding that both chemicals trans formed only up to 25% of the mammary glands in the 1000fold range of concentrations studied (Table 1). However, even though the activity was low in terms of percentage of number of glands transformed, both carcinogens brought about num bers of NLAL per transformed gland (up to 3) like those of the more active transforming agents, N-OH-FAA and N-AcO-FAA. The results suggest that the main difference between the in
vitro actions of the carcinogensof high and low transforming activities resides in the percentage of the glands that are transformed. Apparently, in glands committed to transforma tion, the carcinogens of high and low transforming activity produced similar numbers of NLAL per transformed gland. Both carcinogens, 2-FA and 2-NF, were moderately toxic in that they brought about a paucity or absence of terminal alveolar buds in up to 28 and 37% of the glands, respectively (Table 1). In considering the basis of the similar magnitudes of the transforming and cytotoxic activities by 2-FA and 2-NF, it seems reasonable to suggest that the similarity resides in their metabolic convergence at N-hydroxy-N-2-fluorenylammne (23, 24). We then determined whether the in vitro system could distin guish between the above fluorenyl carcinogens and a structur ally close fluorenyl noncarcinogen. We accordingly chose to test the known noncarcinogen, fluorene (34). The examination also served to inquire whether the essential feature of the transforming agent resides in the functional substituent group (amino or amide or their derivative), rather than in the fluorene hydrocarbon nucleus. Fluôrenewas found to exhibit very low or essentially no transforming activity. Only between 2 and 11% of the mammary glands were transformed over the 1000fold range of fluorene concentration tested (Table 1). Fig. 1f
CANCERRESEARCHVOL. 39
Transformation of Cultured Mammary Gland shows a typical nontransformed gland following treatment with fluorene. Again, even though so few glands were transformed, those that were transformed had up to 3 NLAL/transformed gland, in agreement with the above stated similar incidences of NLAL per transformed gland brought about by chemicals of high and low transforming activities. On the other hand, with increased concentration of fluorene up to one-third of the glands evidenced cytotoxic effects (Table 1). Three conclu sions therefore seem warranted. The process of transformation per se in the whole mammary organ culture system: (a) can distinguish between strong fluorenyl carcinogens (FAA, N-OH FAA, N-AcO-FAA),
moderate fluorenyl carcinogens
(2-FA, 2-
NF), and a fluorenylnoncarcinogen(fluorene),(b) depends on the presence of the amino or amide groups, or their activated or activatable derivatives, in the fluorenyl carcinogens; and (c) does not depend on the cytotoxic effects of these chemicals. Transformation by Naphthylamines. The carcinogen,2-NA, is considerably more oncogenic in bladder of dogs than is 1NA (1 0). There apparently
has been no description
of either
substance as a significant mammary carcinogen in any species (34). Nevertheless,
both naphthylamines
were equally
and
moderately active in transforming mammary gland over a 1000fold range of concentration (Table 1). The rating of moderate activity was based on the up to approximately 50% incidence of transformation of mammary glands. A gland transformed by 2-NA is shown in Fig. 1g. As with the more active agents,
@
excepting FAA, up to 4 NLAL per transformed gland were produced by either naphthylamine. In addition 1-NA was highly cytotoxic, whereas 2-NA was moderately cytotoxic. We then determined whether the in vitro system could distin guish between the carcinogenic naphthylammnesand the struc turally close noncarcinogen, naphthalene (34). The examina tion likewise served to test the essentiality of the amino group compared to that of the naphthalene hydrocarbon nucleus. Naphthalene exhibited virtually no transforming activity in a 1000-fold range of concentration (Table 1). Fig. 1h shows a nontransformed gland following treatment with naphthalene. Nevertheless, naphthalene was moderately cytotoxic at the higher concentration. Therefore, as with the fluorenyl com pounds, the in vitro mammary transformation system distin guishes between naphthylenic carcinogens and a naphthylenic noncarcinogen, and is dependent on the presence of the amino group. Histopathology. Control glands incubated in the develop ment medium during Days 1 through 9 and treated with DMSO on Days 3 to 4 could not be distinguished from normal glands of pregnant mice. The glands displayed rich lobuloalveolar growth and abundant lactation (Fig. 2, a and b). Partial to complete involution of such control glands in the regression medium for 15 days resulted in a relatively bare ductal skeleton at Day 24 (Fig. 2, c to e). The glands which had been treated with the noncarcinogens, fluorene and naphthalene, had greater residual cellularity at Day 24, and those subjected to the carcinogen had considerably more. Glands exposed to FAA on Days 3 to 4 and maintained during Days 1 to 9 in the development medium showed a rather different picture. Proliferation was confined largely to the ductal system (Fig. 2f), while lobuloalveolar budding and secretory activity were minimal (Fig. 2g). On the other hand, there was no qualitative feature distin guishable between the glands that were treated with the car MAY 1979
cinogenic aromatic amines, amides and their derivatives at Day 3 to 4 and examined after the 24-day protocol. Differences were merely a matter of degree, even with the same agent. Many lobules reflected the cytotoxicity of the agent, in that the original alveolar spaces were marked by casts of inspissated secretion and occasional fragmented cells. In certain large persistent lobules (undoubtedly NLAL), there was concomitant proliferation of ductules and alveoli (Fig. 2, h and ‘), frequently with epithelial dysplasia (Fig. 2n). Squamous metaplasia of ductal and alveolar epithelium was noted in 20% of the glands (Fig. 2, k and n). Major ducts beneath the nipple were com monly the seat of marked hyperplasia (Fig. 2k), and tributaries were often occluded by hyperplastic and dysplastic lining cells (Fig. 2m). Interstices within some of the larger lobules appeared densely hyalmnized(Fig. 2j), and small crescents of hyperplastic myoepithelium were occasionally noted. Periductal connective tissue tended to be thickened, and active fibroblastic prolifer ation was found around some of the larger radicles (Fig. 2!). A mild degree of serous atrophy was common in the supporting fat, and infiltration of lipophages marked areas of ‘ ‘spilled milk― that followed lobuloalveolar disintegration. No actual malignant change was observed, although a pre malignant status comparable to that of HAN in vivo could be sensed. Scoring to evaluate the relative transforming activities of the test substances was not sufficiently fruitful to warrant detailed presentation. Collectively, the glands, treated with the carcinogens scored twice as high as did the control glands, and those given FAA scored highest. It seems therefore that whole organ preparations are probably more reliable objects for evaluating the relative structural alterations associated with transformation in culture than are histological sections. DISCUSSION The present study provides the initial description of the transformation of mammary gland by the carcinogenic aryl amines and amides and their derivatives in any culture system. The transformation occurs in the cultured whole mammary organ, which can undergo in vitro the processes of mammary lobuloalveolar development, functional differentiation, and glandular involution. The chemical transformation involves an escape from the hormonal controls of these processes (see ‘ ‘Introduction'
‘),
and
discriminates
between
carcinogens
and
noncarcinogens of analogous structures in the fluorenyl and naphthylenic chemical series. The ability of chemicals to transform mammary gland is presumed to depend on their ability to generate, or to be activated to, electrophiles that interact with critical macromol ecules in that organ. The noncarcinogens, fluorene and naph thalene, which do not have an activatable or electrophilic substituent, were virtually without transforming activity. In con trast, the carcinogens, which are activated or activatable, were effective transforming agents. The transforming activities of the chemicals were in the following decreasing order: FAA > NOH-FAA > N-AcO-FAA> 2-FA, 2-NF > fluorenein the fluorenyl series, and 2-NA, 1-NA naphthalene in the naphthylenic series. All the carcinogens transformed mouse mammary glands in concentrations of pmol/ml/gland. Several factors probably contributed to the order of relative transforming activ ities. (a) FAA was the most active, presumably in part because 1787
Q. J. Tonelli, R. P. Custer, and S. Sorof
Table1 Transformation of mouse mammary gland by chemical carcinogens in whole organ culture Transformed glands
Chemical Control FAA
Dose(j.tg/ No.of Cytotoxicity5 mi/gland) glands (%) NLAL/glandb 0 44 2 0 0.001 27 15 2,5±i.7c@
Incidence Transforming No.
(%)
0 4
0 15
0.01
26
11
3.4±1.1
9
35
0.1 0.5
27 27
0 14
7.2 ±3.0
8
30
5.6±2.8
19
70
1.0
30
42
4,5 ±2.2
25
83
Control
0
62
2
0
0
0
N-OH-FAA
0.001
26
11
3.3±1.6
6
23
0.01
33
12
0.05
36
25
0.10
32
30
1.0
16
69
3.0 ±2.0 2.5 ±1.3 2.9 ±2.4 3.2±2.1
14 16 20 4
42 44 62 25
0 0.001
64 27
15
0.01
33
0.05
34
15
0.10
33
1.0
Control N-AcO-FAA
2
0
0
4.3±2.4
7
26
3.6±2.7
13
39
3.4±2.1
16
47
24
2.9±2.4
16
48
14
62
2.0±1.1
7
50
0 1
0 1
0 3
3
0
25
2-FA
0.001
32
0 3
0.01
25
20
2.5
2
0.10
28
21
3.3
3
11
1.0
28
28
2.6±0.8
7
25
8
Control
0
18
0
0
0
0
2-NF
0.001 0.01
24 24
12
1
1
4
17
1.5
2
8
0.1 1.0
24 22
37 35
3.0±0.7 3.0
6 3
25 14
0 0.001 0.01
48 48 50
0 10
0 3.0
0 1
0 2
0.1 1.0
36 46
Control
0
33
0
1-NA
0.001 0.01
20 8
45
0.05 0.10 0.50
18 20 21
22
1.0
16
Control
0
30
2-NA
0.001
32
21
1.0
2
4
23
1.7±0.9
4
11
35
2.7
3
7
0
0
0
2.4±0.5
5
25
2.3
3
37
1.8±0.8
5
27
30
3.2±1.4
8
40
52 62
2.3±1.4 3.7±1.6
10 6
48 37
3 20
0 3.7 ±2.4
0 15
0 47
0
0.01
19
31
3.1±1.2
8
42
0.1
17
41
4.2±1.0
4
23
1.0
20
47
3.3±2.4
11
55
Control
0
25
0
0
0
0
Naphthalene
0.001 0.01 0.1
28 23 26
4 9
0 0
0 0
0 0
1.0
27
18 33
1 0
1 0
4 0
0 0 0 0 0 36 Control 22 6.0±3.6 23 64 0.001 36 DMBA a Percentageof regressedglandswithabsenceor paucityofalveolarbuds.
@
b NIAL per transformed gland. c Ratings are based on the maximum percentage
Veryhigh
High
0
Control
Control Fluorene
activity ratingc
of incidences of gland transformation.
Moderate
Low
Low
Very low
Moderate
Moderate
Essentially none
High
Greater than
about75%, very high;about75 to 50%, high;about50 to 25%, moderate;about25 to 10%, low; less thanabout 10%, very low or essentiallynone. d Average ± S.D.
1788
CANCER RESEARCH VOL. 39
Transformation of Cultured Mammary Gland it was activated within the mammary gland itself. The in situ activated metabolites of FAA might therefore be present in high concentrations at critical loci within the gland, and accordingly they might be less consumed in interactions nonproductive in transformation. (b) FAA is among the most chemically stable of the carcinogens tested. In contrast, incubation at 37°in culture medium containing nucleophiles and 95% oxygen atmosphere probably resulted in partial chemical alteration of the amines and their activated derivatives. (c) The balance between cyto toxic and transforming activities contributed to the overall incidences of transformation. Thus, the most active transform ing agent, FAA, was the least cytotoxic of the carcinogens at concentrations up to 0.5 pg/mI/gland, whereas the activated derivatives, N-OH-FAA and N-AcO-FAA, were more toxic. The combination of all these factors probably resulted in the de crease in the transforming activity of the derivatives of FAA (FAA > N-OH-l@AA> N-AcO-FAA) with increase in electrophil icity of its activated derivatives. The finding of in vitro transformation of mouse mammary glands by the aryl amines and amides and their activated derivatives supports the worthiness of the whole mammary organ culture system in studies of many families of carcino gens. The system can discriminate carcinogens from noncar cinogens in the fluorenyl and naphthylenic series (present study), and in the polycyclic aromatic hydrocarbon series (5, 20, 21 ). The system is sensitive to carcinogens
not previously
reported to cause mammary tumors, e.g., 1-NA and 2-NA (10, 34). Further, the present study suggests that whole organ
cultured mouse mammary glands can activate aryl amines and amides to reactive metabolites. Hydrocarbon-induced aryl hy drocarbon hydroxylase has previously been found to be pres ent in mouse mammary glands (8), mammary cell lines (9), and cultured whole mammary organs of mice.6 The findings collec tively suggest that the mouse mammary gland in organ culture may be capable of activating and being transformed by a spectrum of chemical carcinogens. Chemical transformation of mouse mammary gland in whole organ culture may relate to early stages of the chemical car cinogenic process. HAN are considered to be early premalig nant mammary lesions in mice (12, 13, 16, 22, 29) but report edly not in rats (17, 32). HAN have been produced in vivo in mouse mammary glands by 3-methylcholanthrene (16, 22), DMBA, benzo(a)pyrene, urethan, and the carcinogenic aryl amine, 2,7-fluorenediamine (Faulkin, cited in Ref. 16). HAN produced by mouse mammary tumor virus and the carcino genic hydrocarbons have been indicated to be tumorigenic in mice after implantation into gland-free fat pads (12, 13, 22, 29). The in vitro equivalent of mouse mammary HAN seems to
be the NLAL, brought about in whole organ culture previously by the carcinogenic polycyclic aromatic hydrocarbons (5, 20, 21 ), and now by the aryl amines and amides and their activated
derivatives. Furthermore, mouse mammary glands containing
NLALasa resultof transformation byDMBAin wholeorgan
culture are indicated to be tumorigenic after implantation Into mice (see ‘ ‘Introduction' ‘). It is also noteworthy that primary epithelial cell cultures of rat mammary gland have recently been neoplastically transformed by DMBA and N-nitroso methylurea, as evidenced by their ability when transplanted into gland-free mammary fat pads to produce adenocarcinomas a N. T. Telang, A. B. Kundu, and M. R. Banerjee, personal communication.
MAY 1979
(30). It remains to be determined
whether mammary glands
containing NLAL generated in whole organ culture by the carcinogenic aryl amines or amides, or their activated deriva tives can likewise undergo progression to cancer in vivo. Efforts inthisregardare inprogress.
Studies are also underway on the possible roles of carcino gen-virus interactions in the transformation of mouse mammary glands in whole organ culture. In preliminary experiments, both control glands and glands treated with FAA (1 .0 @tg/mI/gland) according to the present protocol have been found at 9 days of organ culturing to contain less than 1 copy of MuMTV RNA per cell, as determined by molecular hybridization with a com plementary DNA probe. In addition, radioimmune assay has failed to detect the presence of the major glycoprotein of MuMTV, a glycoprotein with a molecular weight of 52,000, either in the mammary glands at the ninth day, or in the cul ture medium over the 9-day period. The tentative indication is that the chemical transformation of BALB/c mammary glands by FAA in organ culture appears not to be mediated by TV78 Chemical carcinogens appear to be contributory in the cau sation of most cancers in humans (18). Carcinomas constitute the major type of cancer in humans, while mammary carcino mas are among the most common cancers in women in many countries (7). A considerable need therefore exists for in vitro models of chemical transformation in epithelial tissues, espe cially of mammary gland. Several families of chemical carcin ogens have now been found to bring about in cultured whole mammary glands a transformation of epithelial cells that may relate to the chemical causation of mammary gland neoplasias in particular and carcinomas in general. The aryl amines and amides and their activated derivatives (this report) and the polycyclic aromatic hydrocarbons (5, 20, 21) constitute large groups of carcinogens that carry out this transformation. Fur ther, a considerable body of information already exists regard ing these carcinogens with respect to their metabolic activation to reactive electrophiles, their subsequent interactions with macromolecules in target organs, and their abilities to bring about mutagenesis, transformation in culture, and carcinogen esis in various organs and species (1 , 2, 6, 10, 23, 24, 34, 36, 37). The linking of this large body of information to the trans
formation system of cultured mouse mammary glands will sig nificantly increase the understanding of the actions of chemical carcinogens on mammary gland and epithelial tissues in gen eral. 7 0.
J.
Tonelli,
C.
A.
Long,
A.
B.
Vaidya,
and
S.
Sorof,
unptèlished
findings,
and Proc. Am. Assoc. Cancer Res., 20: 260, 1979. a In another study, the principal carcinogen-protein complexes In mammary organ cytosols of BALB/c mice given[3H)MCA, [3HJenzo(a)pyrene, or[3H]DMBA
werepreviously foundbyconventional molecular sieving tohaveamolecular size
of 80,000 to 90,000 daftons (14). Several other carcinogen-organ systems were preliminarily determined by the same method to have principal carcinogen-protein complexes of this same molecular size (15, 33). However, the addition of blue
dextrandirectlyto thecytosolsof wholemammaryorgan(or blood)asaninternal reference molecular weight marker results in a lowering of the numerical value of the molecular size of the principal mammary complexes to the size of serum albumin. Also, a recent experiment with gland-free (cleared) mammary fat pads showed that these principal complexes in mammary cytosol do not originate from the mammary glandular epithelial cells per as. It seems possible therefore that the mammary cytosolic principal carcinogen-protein complexes, which in part are apparently covalently joined, may derive from or be related to the serum albumin within the mammary glands. The findings demonstrate the difficulty of assigning biochemical properties to the mammary glandular epithellal cells per se (M. S. Dickens, 0. J. Tonelli, and S. Sorof, unpublished findings).
1789
Q. J. Tonelli,
R. P. Custer,
and
S. Sorof
ACKNOWLEDGMENTS We are grateful to Dr. Mihir A. Banerjee, University of Nebraska at Lincoln, for his gracious help preliminary to our use of the whole mammary organ culture system, and for several reports in advance of publication. The valuable donations of cited chemicals by the Chemical Repository of the National Cancer Institute Carcinogenesis Research Program is also acknowledged. We express our grat itude to Dr. Michael S. Dickens for useful discussions. We also commend the valuable laboratory assistance rendered by Reynold A. Panettieri and the pho tographic assistance given by John F. Heinsinger.
REFERENCES
74:4020-4024, 1977.
18. Higginson, J. Present trends in cancer epidemiology. Can. Cancer Conf., 8: 40-75,
1969.
19. Ichinose, R. R., and Nandi, S. Influence of hormones on lobulo-alveolar differentiation of mouse mammary glands in vitro. J. Endocrinol., 35: 331 340, 1966. 20. Kundu, A. B., Telang, N. T., and Banerjee, M. A. The binding of 7,12dlmethylbenz[a]anthracene to mammarygland DNA in organ culture. J. Nail. Cancer Inst., 61: 465—469,1978. 21 . Lin, F. K., Banerjee, M. R., and Crump, L. R. Cell cycle-related hormone
carcinogen interaction during chemical carcinogen induction of nodule-like mammary lesions in organ cuiture. Cancer Res., 36: 1607—1 614, 1976. 22. Medina, D. Preneoplastic lesions in mouse mammary tumorigenesis. Meth
ods Cancer Res., 7: 3—53,1973. 23. Miller, E. C. Some current perspectives on chemical carcinogenesis in 1. Arcos, J. C., and Argus, M. F. Molecular geometry and carcinogenic activity humans and experimental animals: Presidential address. Cancer Res., 38: of aromatic compounds. New perspectives. Adv. Cancer Res., 11: 305— 471, 1968.
1479—1496, 1978.
2. Arcos,J. C., and Argus,M. F. Chemicalinductionof cancer.Structuralbasis 24. Miller, E. C., and Miller, J. A. The metabolism of chemical carcinogens to reactive electrophiles and their possible mechanisms of action in carcino and biological mechanisms, Vol. 2B, pp. 1-339. New York: Academic Press, Inc., 1974. genesis. In: C. E. Searle (ed), Chemical Carcinogens. American Chemical Society Monograph 173, pp. 737-762. Washington, D. C.: American Chem 3. Armstrong, E. C., and Bonser, G. M. The carcinogenic action of 2-acetyl aminofluorene on various strains of mice. J. Pathol. Bacteriol., 59: 19—27, ical Society, 1976. 1947. 25. Miller, E. C., Miller, J. A., and Enomoto, M. The comparative carcinogenici ties of 2-acetylaminofluorene and its N-Hydroxy metabolite in mice, ham 4. Banerjee, M. A., Wood, B. G., Lin, F. K., and Crump, L. R. Organ culture of sters, and guinea pigs. Cancer Res., 11: 2018—2032,1964. whole mammary gland of the mouse. Tissue Culture Assoc. Manual, 2: 457— 462, 1976.
5. Banerjee, M. R., Wood, B. G.. and Washburn, L. L. Brief Communication: Chemical carcinogen-Induced alveolar nodules in organ culture of mouse mammary gland. J. Nail. Cancer Inst., 53: 1387—1 393, 1974. 6. Bartsch, H., Malaveille, C., Such, H. F., Miller, E. C., and Miller, J. A. Comparative electrophilicity, mutagenicity, DNA repair induction activity, and carcinogenicity of some N- and 0-Acyl derivatives of N-hydroxy-2aminofluorene. Cancer Res., 37: 1461 —1 467, 1977. 7. Cancer Around the World, 1972—1 973, World Health Statistics. CA, 28: 28— 29, 1978. 8. Chuang, A. H. L., and Bresnick, E. Aryl hydrocarbon hydroxylase in mouse mammary gland. Cancer Res., 36: 4125—41 29, 1976. 9. Chuang, A. H. L., Howard, E. F., and Bresnick, E. Aryl hydrocarbon hydrox ylase in mouse mammary gland: In vitro study using mammary cell lines. Chem.-Biol. Interact., 17: 9—16,1977. 10. Clayson, D. B., and Garner, R. C. Carcinogenic aromatic amines and related compounds. In: C. E. Searie (ed), Chemical Carcinogens. American Chem ical Society Monograph 173, pp. 366—461 . Washington, D. C.: American Chemical Society, 1976. 11. Cramer, J. W., Miller, J. A., and Miller, E. C. N-Hydroxylation: A new metabolic reaction observed in the rat with the carcinogen 2-acetylaminoflu orene. J. Blol. Chem., 235: 885-888, 1960. 12. DeOme, K. B., Faulkin, L. J. Jr., Bern, H. A., and Blair, P. B. Development of mammary tumors from hyperplastic alveolar nodules transplanted into gland-free mammary fat pads of female C3H mice. Cancer Res., 19: 515— 520, 1959. 13. DeOme, K. B., and Nandi, S. The mammary tumor system in mice. In: W. J. Burdette (ed), Viruses Inducing Cancer, pp. 127—1 39. Salt Lake City, Utah: University of Utah Press, 1966. 14. Dickens, M. S., and Sorof, S. Carcinogen-protein complexes in mammary gland after administration of 3-methylcholanthrene. Biochem. Biophys. Res. Commun., 79: 713—719,1977. 15. Dickens, M. S., and Sorof, S. A common molecular size class of carcinogen protein complexes during chemical carcinogenesis in various systems. Fed. Proc. (abstracts), 37: 1315, 1978. 16. Faulkin, L. J., Jr. Hyperplastic lesions of mouse mammary glands after treatment with 3-methyicholanthrene. J. Nail. Cancer Inst., 36: 289—297, 1966. 17. Haslam, 5. Z., and Bern, H. A. Histopathogenesis of 7,12-dimethylbenz[a]anthracene-induced rat mammary tumors. Proc. NatI. Acad. Sci. U. S. A.,
26. Miller, J. A., Sandin, R. B., Miller, E. C. . and Rusch, H. P. The carcinogenicity
of compounds related to 2-acetylaminofluorene. II. Variations in the bridges and the 2-substituent. Cancer Res., 15: 188—199,1955. 27. Moore, D. H., Long, C. A., Vaidya, A. B., Sheffield, J. B. , Dion, A. S., and
Lasfargues, E. Y. Mammary tumor viruses. Adv. Cancer Res., in press, 1979. 28. Morris, H. P., Dubnik, C. S. , and Johnson, J. M. Studies of the carcinogenic
action in the rat of 2-nitro-, 2-amino-, 2-acetylamino-, and 2-diacetyl-ami nofluorene after ingestion and after painting. J. Nail. Cancer Inst., 10: 1201— 1213,1950. 29. Nandi, S., and McGrath, C. M. Mammary neopiasia in mice. Adv. Cancer Res.,17:353—414, 1973. 30. RIchards, J., and Nandi, S. Neoplastic transformation of rat mammary cells
exposed to 7,12-dimethylbenz[a]anthracene or N-nitrosomethylurea in cell culture. Proc. Nail. Acad. Sci. U. S. A., 75: 3836—3840,1978. 31 . Singh, D. V., DeOme, K. B., Bern, H. A. Strain differences in response of the mouse mammary gland to hormones in vitro. J. Nail. Cancer Inst., 45: 657— 675, 1970. 32. Sinha, D., and Dao, T. L. Hyperpiastic alveolar nodules of rat mammary
gland: Tumor-producing capability in vivo and in vitro. Cancer Left., 2: 153— 160, 1977.
33. Sorof, S., and Dickens, M. S. Evidence for similar principal target proteins of chemical carcinogens in six carcinogen-organ systems. Proc. Am. Assoc. Cancer Res., 19: 37, 1978. 34. John I. Thompson and Co. Survey of compounds which have been tested
for carcinogenic activity, 1951 through 1973, U. S. Public Health Serv. Pubi. No. 149. Washington, 0. C.: U. S. Government Printing Office. 35. Tonelli, 0. J., and Sorof, S. Transformation of mouse mammary glands in whole organ culture by carcinogenic aromatic amines and amides. J. Cell Biol., Abstract, J. Cell Biol. 79: 73a, 1978. 36. Weisburger, E. K., and Weisburger, J. H. Chemistry, carcinogenicity, and
metabolism of 2-fluorenamine and related compounds. Adv. Cancer Res., 5: 331-431, 1958. 37. Weisburger, J. H., and Weisburger, E. K. Biochemical formation and phar macological, toxicological, and pathological properties of hydroxylamines and hydroxamic acids. Pharmacol. Rev., 25: 1—66,1973. 38. Wood, B. G., Washburn, L. L., Mukherjee, A. S., and Banerjee, M. R. Hormonal regulation of lobulo-alveolar growth, functional differentiation and regression of whole mouse mammary gland in organ culture. J. Endocrinol., 65: 1—6,1975.
Fig. 2. Vertical sections through fixed mouse whole mammary glands cultured for 9 or 24 days. The glands were embedded in paraffin, cut at 5 @m, and stained with hemaioxylin and eosin. Arrows point to cells in mitosis. a, DMSO-treated control mammary gland showing normal Iactational activity after maintenance in development medium for Days 1 through 9. x 40 (see Fig. la). b, alveoli from terminal lobule (shown in a) lined by well-preserved secretory epithelium. Lumitia are distended with milk. x 400. c, DM50-treated gland after development and regression in culture, showing on Day 24 striking involution of the secretory units and persistence of the major duct system. x 40 (see Fig. 1c). d, subinvolution of alveoli of mammary gland depicted in c. x 400. e, atrophy of terminal duct from mammary gland shown in c. x 400. f, FAA-treated mammary gland maintained in development medium at Days 1 through 9. There is striking proliferation of the ductal system and an inhibition of alveololobules. x 40 (see Fig. 1b and contrast with Fig. 2a). g, terminal lobule of mammary gland in f, showing ductules and alveolar buds with relatively little secretory activity. x 400 (contrast with b). h, FAA-treated mammarygland after 24 days in culture in the development and regression media. Shown is a relatively dense portion (NLAL) in which cytolysis is minimal. Ductular proliferation is pronounced, as emphasized by hyperchromatic lining epithelium. x 100 (see Fig. 1d). i, FAA-treated mammary gland similar to that in h. The NLAL has a large alveolar component along with considerable ductular proliferation. X 100 (correlate with Fig. 1d). 1,proliferation of alveolar epithelium of the mammary gland in i, showing marked hyalinization of surrounding interstices.
x 400.k,Mammary glandexposed to1-NAaccording tothe24-day protocol. Striking proliferation ofthemajor mammary ductliningwithfociofsquamous metaplasia
are evident. x 400. I, mammary gland treated with FAA in the 24-day protocol. There is a large branch of mammary duct with multilayered columnar epithelial lining. The cells are moderately dysplastic. Fibroblastic proliferation and lymphocytic infiltration are evident in the periductal connective tissue. x 400. m, mammary gland treated with 1-NA according to 24-day protocol. Shown is an intermediary duct with a lobuloalveolar bud. The epithelial lining is markedly hyperplastic and dysplastic, with 5 mitoses (arrowheads) in this single field. x 400. n, mammary gland treated with 2-NA in the 24-day protocol. The alveolar epithelium is hyperplastic and dysplastic and shows squamous metaplasia. x 400 (see Fig. 1g).
1790
CANCERRESEARCHVOL. 39
F
I
1mm
f
1
Fig. 1. Mouse whole mammary glands cultured for 9 or 24 days. Glands were treated with test substance in DMSO on Days 3 to 4. Glands were fixed and stained. Details are provided in the text. All photographs were magnified to the same extent. A 1-mm mark is shown above a. a, DM50-treated control mammary gland, shown after morphological development resulting from maintenance in development medium during Days 1 to 9; b, FAA-treated mammary gland (0.1 @g/ml/gland), demonstrated after morphological development brought about by incubation in development medium during Days 1 to 9; c, DM50-treated control mammary gland, shown after development and regression according to the 24-day protocol. The presence of regressed lobulo-alveolar structures and the absence of NLAL are demonstrated. d, mammary gland treated with FAA (0.1 @g/ml/giand),shown after development and regression according to the 24-day protocol. Arrows, densely stained, nonregressed NLAL. e, mammary gland treated with FAA (1.0 @g/mi/giand),shown after development and regression according to the 24-day protocol. The cytotoxic effects of the carcinogen at this higher concentration are indicated by the lack of terminal alveolar buds along the margin of the gland (arrows). f, nontransformed mammary gland treated with fluorene (1.0 @ag/ml/gland),demonstrated after development and regression according to the 24-day protocol. NLAL are not evident. g, mammary gland treated with 2-NA (0.1 @g/ml/gland),pictured after development and regression according to the 24-day protocol. Arrows, typical NLAL. h, nontransformed mammary gland treated with naphthaiene (1.0 @g/ml/gland),shown after development and regression according to the 24-day protocol. NLAL are not evident.
@
:
Q. J. Tone!li, R. P. Custer, and S. Sorof
‘\@.@L•e;
@
@a@•@•: @,
@
..,
@‘
q • .
@
-
it,. ..:,, - @‘
@
::@:::@..
@
...,I
•1,,,-
,@‘.,
m 2
See legend to Fig. 2 on page 1790, below References.
I 792
CANCER
RESEARCH
VOL.
39
