Suppression of 7,12-Dimethylbenz(a) Anthracene -Induced Breast Carcinoma by Coumarin in the Rat
G. FEUER, J. A. KELLEN, and K. KOVACS
Departments of Clinical Biochemistry, University of Toronto and Sunnybrook Medical Centre and Department of Pathology, St. Michael’s Hospital, University of Toronto
Abstract. The effect of oral administration of coumarin on the induc- tion of breast tumors brought about by DMBA has been studied in female Wistar rats. Coumarin given before DMBA caused a signifi- cant dose-related suppression of the incidence of adenocarcinomas, although no histologic difference was found in the tumor between untreated and coumarin-treated rats. The growth rate, size and multiplicity were also reduced and tumor occurrence was delayed. Parallel with the suppression, hepatic drug metabolizing activity was decreased and serum prolactin level increased. Coumarin given after DMBA elicited no effect on the carcinogenic potency of DMBA. In contrast to the action on mammary tumorigenesis, coumarin provided no protection against hemopoietic and adrenocortical necrosis brought about by DMBA.
Key Words. Dimethylbenz(a)anthracene - Carcinogenicity - Adrenal necrosis - Hepatic metabolism - Coumarin treatment - Prolactin level
Introduction
The biochemical basis for DMBA1 action has been sought since the mammary adenocarcinoma induced by this com- pound has been used as an excellent model system in the study of cancer [1]. These studies have been conducted along three main lines: The relationship between various DMBA metab- olites and carcinogenic potency has been explored [2, 3, 4, 5], the hormone dependence of the cancer has been studied [6, 7, 8], and its biochemical characteristics investigated [9, 10]. In spite of these efforts, the mechanism of DMBA- induced mammary tumors still has not been established.
Previous results from our laboratory indicated that pretreat- ment of rats with coumarin significantly reduced the incidence of breast carcinomas induced by DMBA injections [11]. The protection against this compound was associated with a decrease in the ability of the hepatocyte to metabolize the drug [12]. Since coumarin elicited no significant hepatic im- pairment, the suppression of carcinogenesis in animal experi- ments seemed to provide a scope for further investigations into the mechanism of action of the carcinogen. In the present studies, attempts have been made to clarify whether or not the protection provided by coumarin in mammary tumor induc- tion can be correlated with blood prolactin levels and whether or not coumarin inhibits hemopoietic and adrenocortical in- jury brought about by DMBA.
Materials and Methods
Animals and treatments. Wistar female rats obtained from High Oak Ranch (Richmond Hill, Ont.) were used through- out these studies. The animals were kept on a standard diet with water or coumarin solution ad libitum. They were divided into four groups of 32 rats each. Group 1: DMBA-treated. DMBA (courtesy of Upjohn Company, Kalamazoo, Mich.) was injected via the tail vein in doses of 2 mg/rat on days 50, 53 and 56 of life [7]. Group 2: Coumarin + DMBA-treated. Coumarin was administered dissolved in drinking water (7.6 mg/100 ml tap water) from day 44 to 61 inclusively. The water consumption of each rat was measured and the cou- marin intake calculated. This group received an average 0.06-0.07 mM coumarin/kg body weight/day. DMBA was given as in group 1. Group 3: Coumarin + DMBA-treated. Coumarin was dissolved in arachis oil and given by oral intu- bation, 1 mM/kg body weight/5 ml oil from day 44 to 61; DMBA given as in group 1. Group 4: DMBA + coumarin- treated. DMBA was administered as in group 1; coumarin, 1 mM/kg given from day 56 to 73 inclusive.
Estrous cycles during the pretreatment period were followed by daily vaginal smears. From each group, on day 50, 6 rats were sacrificed in order to measure hepatic drug metabolizing activity and the serum prolactin level. On day 57, from each group 10 rats were killed for liver, femoral bone marrow and adrenal histologic examinations and enzyme assays. The remaining 16 rats in each group were inspected twice weekly for the appearance of palpable tumors. Diameter of each tumor was measured by calipers and tumor size was expressed as the geometrical mean. The growth rate was assessed on the basis of weekly changes. At various intervals some tumors were removed for histologic tests. Some organs from rats died or sacrificed before the termination of the experi- ments were also examined. At 198 days of life the experiments were terminated and at autopsy gross examination was con- ducted on all organs. All tumors removed before or at the termination of the study were kept and a search for small tumors was also carried out in all animals.
Enzyme activity. Coumarin 3-hydroxylase, representing the drug metabolizing activity of the hepatic endoplasmic reti- culum, was measured [13].
Serum prolactin determination. This was done by a modified radioimmunoassay [14]. Rat prolactin assay reagents were
1 Abbreviations: DMBA, 7,12-dimethylbenz(a)anthracene.
| Parameter | Group | ||||
|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | ||
| Number of animals alive at end of study | 16/16 | 13/16* | 16/16 | 12/16* | |
| Body weight, g | 291.3 ± 103.3 ** | 282.9 ± 134.1 | 291.6 ± 80.6 | 300.8 ± 38.3 | |
| Number of animals with tumors | 16/16 | 11/13 | 12/16 | 11/12 | |
| in % | 100 | 84.6 | 75.0 | 91.6 | |
| Number of adenocarcinomas | 0 | 0 | 2 | 4 | 1 |
| 1 | 1 | 4 | 10 | 1 | |
| 2 | 4 | 2 | 1 | 6 | |
| 3 | 4 | 3 | 1 | 1 | |
| 4 | 4 | 1 | 0 | 0 | |
| 5 or more | 3 | 1 | 0 | 3 | |
| Number of fibroadenomas | 2 | 4 | 2 | 1 | |
| Multiplicity | 3.9 | 2.5 | 1.2 | 3.5 | |
| Age at tumor | 1 day | 139 | 156 | 156 | 154 |
| 2 days | 157 | 174 | 175 | 167 | |
| 3 days | 178 | 178 | 198 | 169 | |
| 4 days | 180 | 191 | - | 169 | |
| 5 days or more | 198 | 194 | - | 190 | |
| Average size of tumors, mm | 7.8 ± 29.7 | 11.9 ± 25.7 | 3.6 ± 3.9 *** | 6.9 ± 30.7 | |
| Number of tumors, | |||||
| regressing | 5 | 3 | 3 | 3 | |
| slow growing | 21 | 14 | 6 | 19 | |
| fast growing | 30 | 19 | 6 | 16 | |
| total | 56 | 36 | 15 | 38 | |
* Death of some animals was unrelated to the treatment.
** Measurements are expressed as means ± S.E.
*** Significantly different from group 1, P < 0.05.
provided by the National Pituitary Agency under a program supported by the National Institute of Arthritis and Metabolic Disease. The serum was kept at -20 ℃ until assayed. Assays were carried out according to instructions provided with the reagents.
Histology. At autopsy, specimens from tumors and from the liver, femoral bone marrow and adrenal gland were fixed in buffered formalin and embedded in paraffin. Sections of 4-6 µ thickness were stained with the hematoxylin-phloxine- saffron technique.
The significance of the differences between each group for each parameter was determined by Student’s t-test.
Results
Incidence of breast tumors. The natural history of mammary tumors induced by DMBA is depicted in table I. In group 1, given DMBA alone, the first tumor appeared between the 8th-9th week, followed fairly closely by the second, the third and fourth, reaching a multiplicity of 3.9 tumors per animal of variable size. Group 2 showed only slight differences. Two rats had no tumor, the progress of carcinoma production was somewhat slower than in group 1 and the multiplicity was
reduced to an average of 2,5 per animal. No difference was found either in the diameter or in the growth rate of tumors, when compared with the coumarin untreated group. In group 3, coumarin pretreatment at high dose level brought about a significant reduction in the incidence of tumors. The number of animals bearing tumors was significantly reduced and most of them had only one tumor detectable at the post-mortem examination. There was a significant delay in the first appear- ance of tumors to between 14th-15th week, and a reduction in diameter. The number of progressing tumors was halved when compared with group 2. In group 4, administration of coumarin after DMBA did not reduce tumor formation. The time course, progress, multiplicity and size were similar to those of group 1.
Histology. In the acute experiment (rats killed at day 57) administration of DMBA caused massive hemorrhagic necro- sis in the zona fasciculata and reticularis of the adrenal cortex identical with that described previously [15]. No protection was noted in rats pretreated with coumarin. The extent and histologic features of adrenocortical necrosis did not differ between the untreated and coumarin-treated groups.
In the DMBA-treated rats the bone marrow was edematous and hemorrhagic. The number of hemopoietic cells was strikingly reduced. Coumarin pretreatment did not offer
protection against bone marrow injury. No major hepatic changes were detected in any of the groups.
In the chronic experiment (rats killed at day 198) no histologic differences were revealed among the tumors of the untreated and coumarin-treated rats. The tumors corresponded to adenocarcinomas with varying degrees of differentiation (fi- gure 1). Gland-like structures were apparent. Some of them were dilated and occasionally transformed into cysts, fre- quently containing acidophil secretion and cellular debris in their cavities. The glands were lined by single or multiple layers of columnar or flat epithelial cells. Papillary growths often projected into the glandular lumina. Some areas of the tumors consisted of solid cell nests. Several tumors were anaplastic with numerous mitotic figures whereas others ex- hibited only a slight or moderate pleomorphism resembling benign adenomas. The amount of connective tissue stroma varied considerably from case to case. In some cases it was scanty; in others, however, it occupied a major part of the neoplasm. Hemorrhages and foci of necrosis were not infre- quent. The fibrous stroma was infiltrated in varying degrees by mononuclear cells and by a few polymorphonuclear leuco- cytes. Metastases were not disclosed microscopically but invasion of the neighbouring muscle tissue was a common finding.
Besides the adenocarcinomas a few typical fibroadenomas were also found (figure 2). In one rat a typical stem cell leukemia was revealed. This rat became pale and weak and was killed after 180 days of life. At autopsy marked hepato- splenomegaly was noted with enlargement of the lymph nodes and petechiae of the serous membranes. Histologic examina- tion showed massive infiltration of various organs with primi- tive stem cells (figures 3 and 4). The liver tissue was most severely affected. In some areas it was so extensively replaced by diffuse proliferation of atypical leukoblasts that it was almost impossible to recognize the original hepatic paren- chyma.
Drug metabolizing activity. This showed a gradual decrease depending on the amount of coumarin administered (table II).
Serum prolactin level. Increasing amounts of coumarin caused a rise in serum prolactin level (table II).
Discussion
These experiments have shown that the induction of mam- mary carcinoma in the female rat by DMBA was retarded by
pretreatment with coumarin. When DMBA metabolism was inhibited, tumor incidence, multiplicity, size and growth rate were all significantly reduced. The protective action of coumarin showed a dose-response relationship (table I). These findings agree with our earlier results [11, 12]. In four different studies extended over 3 years, using two different strains (Sprague-Dawley and Wistar albino rats) and a wide range of coumarin levels, a suppression of cancer was ob- served without any apparent toxic side effects. No marked difference was found in the histologic structure of mammary tumors between untreated and coumarin-treated rats, it seems, therefore, that while coumarin reduced the incidence and yield of breast tumors and retarded their formation, it failed to influence the degree of anaplasia once the neoplasm had developed. In contrast, however, coumarin provided no pro- tection from adrenal necrosis brought about by DMBA.
Parallel with the reduction of tumor incidence, coumarin brought about an inhibition of hepatic drug metabolism and an increase of serum prolactin (table II). The inhibitory action on DMBA-metabolism by coumarin was concordant with our previous report [11] and is similar to the protection provided by pregnancy against DMBA tumorigenesis [12]. It was an unexpected finding that the administration of coumarin increased serum prolactin. Certain other pharma- cologic agents - reserpine, phenothiazines and tricyclic anti- depressants - could block the synthesis, release or effect of prolactin - inhibiting factor in man or animals [16, 17] leading to an increased prolactin secretion. Some hormone treat- ments had a similar effect in rat [18] and man [19]. In contrast, other drugs such as the ergot alkaloids, iproniazid and L-dopa elicited a decreased secretion [16, 20, 21]. Estrogens reduced prolactin-inhibiting factors and also caused a direct release of prolactin from the anterior pituitary in rodents [22]. In man reports were inconclusive [23, 24]. Furthermore, it has been reported that hormones which inhibited the induction of breast tumors in rats could stimulate the growth of the normal mammary gland. The same hormones under different circum- stances greatly enhanced tumor growth and accelerated in- duction [25, 26, 27]. The effect of various drugs and other conditions on DMBA-induced mammary tumor was similarly inconsistent [28]. It seemed therefore difficult to elucidate a causal relationship between the action of coumarin on pro- lactin release and DMBA tumor reduction.
| Parameter | Group | ||
|---|---|---|---|
| 1 | 2 | 3 | |
| Coumarin 3-hydroxylase umol/h/mg protein | 1.35±0.23* | 0.92±0.15 | 0.55± 0.08 ** |
| Prolactin ng/ml | 23.2 ±3.7 | 39.2 ±6.2 | 58.7 ±11.5 ** |
* Results represents 6 rats in each group and are given as means ± S.E.
** Significantly different from group 1, P<0.05.
The metabolic aspects of induction and growth of mammary tumors in rat by DMBA have been extensively studied [2, 3, 4, 5]. These investigations suggested and then disputed the involvement of DMBA hydroxylation; however, the mecha- nism of action has still not been clarified. The present experi- ments showed that DMBA tumorigenesis was related to the hepatic metabolism of the carcinogen. Reduction of its metab- olism brought about a decreased tumor incidence. Since the inhibition of drug metabolism by coumarin was not complete, it is possible that the carcinogenic action of DMBA was associated with one or more of its minor metabolites. Inbition of DMBA hydroxylation by coumarin probably reduced the amount of these minor metabolites to a noncarcinogenic level. This assumption is also in agreement with our previous find- ings [11, 12] and appears to add important information in understanding the mechanism of carcinogenesis in this special case.
Acknowledgments
The authors wish to acknowledge the assistance given by Dr. G. M. BROWN and Mrs. E. JOHANSSON and MRC GRANT MT 4749 in measuring the serum prolactin content. The valuable help rendered by Mrs. R. FARKAS, Mr. A. MIRA- KIAN and Mr. L. MUNIAK is greatly appreciated.
References
1 HUGGINS, C .; BRIZIARELLI, G., and SUTTON, H. Jr .: Rapid in- duction of mammary carcinoma in the rat and the influence of hormones on the tumors. J. Exp. Med. 109: 25-42 (1959).
2 BOYLAND, E. and SIMS, P .: Induction of adrenal damage and cancer with metabolites of 7,12-dimethylbenz(a)anthracenę. Nature 207: 816-817 (1965).
3 FLAKS, A .; HAMILTON, J. M .; CLAYSON, D. B., and SIMS, P .: Evaluation of the carcinogenic activity of the hydroxymethyl derivatives of 7,12-dimethylbenz(a)anthracene by an in vitro - in vivo technique. Int. J. Cancer 10: 548-551 (1972).
4 FLESHER, J. W. and SYDNOR, K. L .: Carcinogenicity of deriva- tives of 7,12-dimethylbanz(a)anthracene. Cancer Res. 31: 1951-1954 (1971).
5 WHEATLEY, D. N. and INGLIS, M. S .: Mammary tumors induced in Sprague-Dawley female rats by 7,12-dimethylbenz(a)anthra- cene and its hydroxymethyl derivatives. Brit. J. Cancer 22: 122-127 (1968).
6 DANIEL, P. M. and PRICHARD, M. M. L .: The response of ex- perimentally induced mammary tumors in rats to hypophy- sectomy and to pituitary stalk section. Brit. J. Cancer 17: 446- 453 (1963).
7 HUGGINS, C .; MOON, R., and MORII, S .: Extinction of experi- mental mammary cancer by estradiol-17ß and progesterone. Proc. Nat. Acad. Sci. Wash. 48: 379-386 (1962).
8 JABARA, A. G .; TOYNE, P. H., and HARCOURT, A. G .: Effects of time and duration of progesterone administration on mam- mary tumors induced by 7,12-dimethylbenz(a)anthracene in Sprague-Dawley rats. Brit. J. Cancer 27: 63-71 (1973).
9 HILF, R .; GOLDENBERG, H .; MICHEL, I .; CARRINGTON, M .; BELL, C .; GRUENSTEIN, M .; MERANZE, D. R., and SHIMKIN, M. B .: Biochemical characteristics of mammary glands and mammary tumors of rats induced by 3-methylcholanthrene and 7,12-dimethylbenz(a)anthracene. Cancer Res. 29: 977-988 (1969).
10 HILF, R .: Biochemical studies of experimental mammary tumors as related to human breast cancer; in BUSCH, Methods in Cancer Research 7, 55-113 (1973).
11 FEUER, G. and KELLEN, J. A .: Inhibition and enhancement of mammary tumorigenesis by 7,12-dimethylbenz(a)anthracene in the female Sprague-Dawley rat. Int. J. Clin. Pharmacol. 9: 62-69 (1974).
12 FEUER, G. and KELLEN, J. A .: Link between carcinogenicity and hepatic metabolism of 7,12-dimethylbenz(a)anthracene. Oncology 30: 499-508 (1974).
13 FEUER, G .: 3-Hydroxylation of coumarin or 3-methylcoumarin by rat liver microsomes and its induction by 4-methylcoumarin. Chem. Biol. Interact. 2: 203-216 (1970).
14 NISWENDER, G. D .; CHEN, C. L .; MIDGLEY, A. R. Jr .; MEITES, F., and ELLIS, S .: Radioimmunoassay for rat prolactin. Proc. Soc. Exp. Biol. Med. 130: 793-797 (1969).
15 HUGGINS, C. and MORII, S .: Selective adrenal necrosis and apoplexy induced by 7,12-dimethylbenz(a)anthracene. J. Expt. Med. 114: 741-759 (1961).
16 HORROBIN, D. F .: Prolactin 1974 (Medical and Technical Publishing Co. Ltd., St. Leonardgate, Lancaster, Great Britain 1974).
17 SHERMAN, L. and KOLODNY, H. D .: The effects of drugs on human hypophysiotrophic functions; in STOLL, Mammary Cancer and Neuroendocrine Therapy, pp. 369-380 (Butter- worth, London 1974).
18 LU, K. H. and MEITES, J .: Effects of serotonin precursors and melanotonin on serum prolactin release in rats. Endocrinology 93: 152-155 (1973).
19 BOWERS, C. Y .; FRIESEN, H. G., and HWANG, P .: Prolactin and thyrotropin release in man by synthetic pyroglutamyl-histidyl- prolinamide. Biochem. Biophys. Res. Commun. 45: 1033-1041 (1971).
20 FLOSS, G. H .; CASSADY, F. M., and ROBBERS, J. E .: Influence of ergot alkaloids on pituitary prolactin and prolactin-dependent processes. J. Pharm. Sci. 62: 699-715 (1973).
21 NAGASAWA, H. and MEITES, J .: Suppression of ergocornine and iproniazid of carcinogen-induced mammary tumors in rats: Effect on serum and pituitary prolactin levels. Proc. Soc. Exp. Biol. Med. 135: 469-472 (1970).
22 MEITES, J .: Hypothalamic control of prolactin secretion; in WOLSTENHOLME and KNIGHT Lactogenic Hormones, pp. 325 to 338 (Churchill-Livingstone, Edinburgh 1972).
23 L’HERMITE, M .; DELVOYE, P., and NOKIN, J .: Human prolactin secretion as studied by radioimmunoassay: Some aspects of its regulation; in BOYNS and GRIFFITHS Prolactin and Carcino- genesis, pp. 81-97 (Alpha Omega Alpha Publishing Company, Cardiff, Wales 1972).
24 WILSON, R. G .; BUCHAN, R., and ROBERTS, M. M .: Prolactin and breast cancer. Proc. R. Soc. Med. 66: 865-866 (1973).
25 CLEMENS, J. A .; WELSCH, C. W., and MEITES, J .: Effects of hypothalamic lesions on incidence and growth of mammary tumors in carcinogen-treated rats. Proc. Soc. Exp. Biol. N. Y. 127: 969-972 (1968).
26 HUGGINS, C. and YANG, N .: Induction and extinction of mam- mary cancer. Science 137: 257-262 (1962).
27 KALIBER, M. S .; GRUENSTEIN, M .; MERANZE, D. R., and SHIMKIN, M. B .: Influence of mammary cancer in Sprague- Dawley rats receiving 7,12-DMBA. Cancer Res. 29: 999-1001 (1969).
28 HUGGINS, C .: Pulse-doses of carcinogens; in CUMLEY et al., Car- cinogenesis: A Broad Critique, pp. 725-730 (Williams and Wilkins, Baltimore 1967).
Request reprints from: Dr. G. FEUER, Department of Clinical Bio- chemistry, Banting Institute, Room 521, University of Toronto, Toronto, Ont. M5G 1L5 (Canada)