Differential carcinogenic effects of intraperitoneal initiation with 7,12-dimethylbenz(a)anthracene or urethane and topical promotion with 12-O-tetradecanoylphorbol-13-acetate in skin and internal tissues of female SENCAR and BALB/c mice.

Groups of female SENCAR or BALB/c mice were initiated once intraperitoneally with 300 micrograms/mouse of 7,12-dimethylbenz(a)anthracene (DMBA) or 20 mg/mouse of urethane at 7 weeks of age. Beginning one week later, mice received topically applied acetone or 12-O-tetradecanoylphorbol-13-acetate (TPA), once weekly, at 2.5 micrograms/mouse for weeks 1 through 6 and 1.25 micrograms/mouse for weeks 7 through 52. The skin lesions were evaluated clinically. A complete necropsy was performed on all mice at week 52. SENCAR mice exposed to DMBA/TPA and urethane/TPA had more skin tumors than SENCAR mice exposed to DMBA or urethane alone and more than BALB/c mice in any treatment group. Of all skin carcinomas diagnosed histologically in DMBA/TPA-exposed mice, less than one-third had been identified clinically while the mice were alive. Most of the carcinomas arose within papillomas. BALB/c mice developed more vascular and uterine tumors than did SENCAR mice injected with DMBA and more lung and vascular tumors than did SENCAR mice injected with urethane. TPA exposure after treatment with either initiator had no significant effect on internal tumor development in either SENCAR or BALB/c mice.

Introduction SENCAR mice were originally bred for susceptibility to two-stage carcinogenesis in mouse skin (1). It has been demonstrated that these mice are exceptionally more sensitive to skin carcinogenesis by chemicals (2)(3)(4) and by ultraviolet irradiation (5) than most other strains or stocks of mice, and that SENCAR mice have a population of spontaneously initiated cells (3), as well as dark cells (2), which are possible targets of carcinogens. An earlier study of two-stage carcinogenesis in SENCAR and BALB/c mice has shown that SENCAR *Tumor Pathology and Pathogenesis Section, Laboratory of Comparative Carcinogenesis, Division of Cancer Etiology, National Cancer Institute, NIH, Frederick, MD 21701. tMicrobiological Associates, Inc., Bethesda, MD 20816. tLaboratory of Cellular Carcinogenesis and Tumor Promotion, Division of Cancer Etiology, National Cancer Institute, NIH, Bethesda, MD 20892. mice are considerably more sensitive than BALB/c mice (3) to two-stage carcinogenesis. The aim of this current study was to evaluate the differential carcinogenic effects of 7,12-dimethylbenz(a)anthracene (DMBA) and urethane on internal tumor development in female SENCAR and BALB/c mice and to observe the effects of a skin tumor promoter, topically applied 12-0-tetradecanoylphorbol-13-acetate (TPA), on development of both epidermal and nonepidermal neoplasms,

Materials and Methods
Animals Two hundred female SENCAR mice (Oak Ridge National Laboratories, Oak Ridge, TN) and two hundred female BALB/c mice (Charles River Breeding Laboratories, Wilmington, MA) were received and held until they were 7 weeks of age, at which time they were initiated. All animals were housed 5 per 7 in. x 11.5 in. polyearbonate cage on 1/8 in. corn cob bedding, according to strain and treatment. Water and Purina Lab Chow were available ad libitum throughout the experiment. All water bottles and cages were changed and sanitized twice per week, the adequacy of food and water was checked daily, and death checks were performed twice daily. Working solutions of both carcinogens were prepared in yellow light under a hood, contained in amber multidose serum vials, and administered immediately after preparation to the animals. DMBA stocks used for dose preparations were stored at -20°C over Drierite in amber glass vials; urethane stocks used for dose preparations were stored at room temperature in foilwrapped vessels.
Promoter. TPA purchased from CCR Inc. (Eden Prairie, MN) was prepared as a 12.5 ,ug/mL or 6.25 jig/ mL solution in acetone (Baker Chemical, Reagent Grade) depending on the dose to be given. Stock solutions of TPA were kept in tightly stoppered glass vessels wrapped in foil and kept refrigerated between uses. Stocks of TPA were stored in the dark at -20°C over Drierite.   treated at 0 experimental time (7 weeks of age). DMBA (1200 ,ug/mL in DMSO/SSV), urethane (80 mg/mL in saline), or the appropriate vehicle were given under a portable charcoal-filtered lab hood by injecting 0.25 mL/ mouse using 1-mL tuberculin syringes attached to appropriate-sized hypodermic needles.

Preparation of Animals and Methods of Administration
TPA solutions (12.5 R±g/mL or 6.25 pig/mL in acetone) or acetone vehicle were applied topically to appropriate groups at 0.2 mL, once weekly, from the first week after initiation through the 52nd experimental week. All mice were shaved biweekly at least 24 hr before promotion in a standardized area to expose the back skin to topically applied chemical treatments. Solutions were applied from Eppendorf multipipettes delivering 0.2 mL      of solution. Disposable tips were changed between groups receiving TPA or vehicle. The concentration of TPA stocks used from weeks 1 through 6 was 12.5 jig/ mL; 6.25 ,ug/mL was given from weeks 7 through 52. Using these stocks resulted in doses of 2.5 ,ug/mouse or 1.25 ,ug/mouse per week, respectively. The doses were lowered when toxic skin reactions to TPA, which had previously been shown to be progressive, were observed.
Experimental Design Table 1 summarizes the designation, composition, and treatment for each group within this study. Mice were weighed monthly, and papilloma counts were performed biweekly as previously described (3).

Pathology
Necropsies were performed on all animals found dead, on moribund mice euthanized during the study, and on mice killed at termination (52 weeks) unless precluded by autolysis or cannibalization. The following tissues were routinely preserved in 10% formalin: skin, including all malignant and benign tumors; any TPA-induced skin lesions; and backskin, spleen, pancreas, lungs, heart, thymus, trachea, larynx, thyroid, lung tumors, kidneys and adrenals, liver, ovaries and uterus, brain, pituitary, and grossly abnormal lymph nodes and other lesions. Avidin-biotin peroxidase-complex (ABC) immunocytochemistry (6) was performed to localize keratin in selected trypsinized skin tumors, surfactant apoprotein (SAP) and Clara cell antigen (CCA) in lung tumors by using rabbit antibodies to human keratin (DAKO Corp., Santa Barbara, CA), specific antibodies to SAP and CCA provided by Dr. G. Singh (6), and the Vectastain kit (Vector Laboratories, Burlingame, CA).

Body Weight and Survival
The growth of female SENCAR mice in all groups was similar except for mice exposed to DMBA/TPA. The latter mice gained weight at a considerably slower rate than did other SENCAR mice (Fig. 1). BALB/c mice of all groups gained weight at the same rate (Fig.  1). SENCAR mice exposed to DMBA/TPA also had the worst survival rate (Fig. 2), since only 23% of the mice survived to the end of the experiment, whereas 63% of the SENCAR mice receiving both urethane and TPA survived, and 77% to 100% of the SENCAR mice in other groups survived. From 87% to 100% of the BALB/c mice were alive at 52 weeks (Fig. 2).

Tumors in Control Mice
At 52 weeks after treatment with the initiator vehicle (DMSO/SSV or saline) and promoter (TPA) or promoter vehicle (acetone) (59 weeks of age) skin and other tumors were found in SENCAR or BALB/c mice (Tables   2 and 3). An increased incidence of skin tumors was associated with TPA promotion in SENCAR mice after DMSO/SSV vehicle initiation. In contrast, only a few skin neoplasms were associated with TPA promotion following saline vehicle initiation.

DMBA Initiation and TPA Promotion
DMBA/TPA-treated SENCAR mice that died early in the study had many skin tumors as previously reported (3) ( Table 2). The occurrence of skin (Fig. 3), vascular, and uterine tumors was strain-related (Table  2). In the DMBA/TPA-treated SENCAR mice, 17 mice had 37 squamous cell carcinomas of the skin (10 seen grossly), 3 of which metastasized to lymph nodes and 2 to lungs. Several carcinomas were seen to arise within papillomas (Fig. 4). Keratin was found in well and poorly differentiated carcinomas (Fig. 5). The vascular tumors were usually small hemangiomas and were found in 35 of 60 BALB/c mice and in 7 of 60 SENCAR mice exposed to DMBA. In BALB/c mice, 57 vascular tumors were found in 35 mice. These tumors usually involved the serosa of intestines (18 mice), uterus (23), mesentery (5), and other abdominal tissues (11). Uterine polypoid adenomas were found more frequently in BALB/c mice (11/60) than in SENCAR mice (1/60). Few skin tumors were found in BALB/c mice. Other tumors were found in similar incidences in both strains, although mammary and intestinal tumors were found only in SENCAR mice. Control mice had few tumors of the skin, blood vessels, and uterus. Lung tumors (adenomas or adeno- carcinomas) occurred in higher incidence, whether confirmed microscopically (14/30 vs. 10/30) or counted visually (17/30 vs. 11/30), and in greater multiplicity (2.0 vs. 0.5), in BALB/c mice exposed to DMBA/TPA than in BALB/c mice given DMBA alone. SENCAR mice showed no differences between these treatment groups.

Urethane Initiation and TPA Promotion
In the urethane experiments, skin tumors were found primarily in SENCAR mice given urethane and TPA (Fig. 6). Twelve mice had squamous cell carcinomas, confirmed histologically, of which seven had been observed clinically. Metastasis occurred in 5 of the 12 mice with squamous cell carcinoma; 4 had metastases to lymph nodes and 2 to lungs. Vascular tumors were significantly more common in BALB/c mice (16/60) than in SENCAR mice (3/59). Most were hemangiosarcomas of the uterus and peritoneal surfaces. Uterine polypoid adenomas occurred primarily in BALB/c mice (6/60). The occurrence of lung tumors noted by gross inspection was subjected to histologic confirmation ( Table 4). The number of microscopically visible lung tumors could have actually been more numerous since only gross lesions were sectioned. The many gross nodules in salinetreated SENCAR mice were found histologically to be focal, chronic inflammatory lesions probably caused by chronic Sendai infection, although these histological lesions were also seen in treated mice. Histologically, tumors were identified as alveolar, solid, tubular, or papillary adenomas and adenocarcinomas (Table 5). Dis- Selected representative lung tumors were evaluated for the presence of pulmonary antigens. Immunocytochemically, SAP was found in the cytoplasm of many tumor cells of alveolar and solid lung tumors (Figs. 7 and 8) but in few cells of tubular and papillary tumors. Rarely, tumor cell nuclei of papillary tumors contained the antigen. CCA was never found in any lung tumors, not even in papillary tumors (Fig. 9). TPA application alone after using either of the two vehicles caused a few skin papillomas and carcinomas in SENCAR mice by 52 weeks (Tables 2 and 3).

Discussion
SENCAR mice were much more susceptible to twostage induction of skin tumors than were BALB/c mice as previously reported (3) but developed fewer carcinogen-initiated vascular and uterine tumors than did BALB/c mice. The increased susceptibility of BALB/c mice to vascular tumors has been previously reported (7)(8)(9). The decreased susceptibility to these induced tumors in SENCAR mice cannot be readily explained. It was shown that metabolism of some carcinogens in selected tissues does not differ significantly in these mice (1), but their increased susceptibility to development of skin tumors, however, may be related to the dark cell population (2). Strains and stocks of mice generally differ in their susceptibility to carcinogens because of differences in genetics, pharmacokinetics, background tumor incidences, and other factors. The glomerulonephritis and associated immunological lesions in our SENCAR mice alter their responses to initiated cells, preneoplastic lesions, or tumors (10). Skin tumors induced by ultraviolet (UV) light have been found to be associated with immunological dysfunction that may be related to skin carcinogensis by chemicals (11). Since SENCAR mice are also more susceptible to UV skin carcinogenesis than other mouse strains, similar immunologic mechanisms may act in SENCAR mice. Internal tumor development in these mice may not, however, depend so much on immunological changes (12,13).
No significant increase in lung carcinogenesis was seen in either SENCAR mice initiated with urethane and promoted with TPA or in BALB/c mice. It has been reported that phorbol promotes lung tumors when initiated with dimethylnitrosamine (14). We may have expected a greater tumor incidence and multiplicity in urethane/TPA-treated mice, but they died earlier than the SENCAR mice treated only wvith urethane. Also, Sendai virus infection, suggested in these mice by pulmonary inflammation, has been shown to reduce induced lung tumor incidence (15) and may have affected the outcome of our experiment. Selected urethane-induced lung tumors were usually immunoreactive for SAP, a marker for alveolar Type II cells (6). Clara cell antigen was never detected in any lung tumors, even in papillary tumors, which have been suggested to be of Clara cell origin (16).
The increased number of skin carcinomas seen histologically over those found by clinical examination is in agreement with the poor gross/microscopic correlation seen in our other studies. In two other studies with topical DMBA and TPA, we found 24 carcinomas in 16 mice histologically, whereas only 6 were found clinically in six mice. The ratio of three to four carcinomas found histologically to every one identified by clinical examination appears to be consistent in our three studies. It has been previously described (17) that squamous cell carcinomas of the mouse skin arise within papillomas. Thus, some of the early carcinomas arising within papillomas are obviously seen clinically as papillomas. Other malignant lesions, however, can be mistaken for papillomas. It would be prudent to section all skin tumors in studies when carcinoma development is a critical endpoint.