[CANCER RESEARCH 43, 5200-5204, November 1983] Relative Carcinogenic Activity of Various Synthetic and Natural Estrogens in the Syrian Hamster Kidney1 Jonathan J. Li,2 Sara Antonia Li, John K. Klicka, Jonathan A. Parsons,3 and Luke K. T. Lam Medical Research Laboratories, Veterans Administration Medical Center [J. J. L], and Departments of Urological Surgery [J. J. L, S. A. L, J. K. K.], Anatomy [J. A. P.], and Laboratory Medicine and Pathology [L K. T. L], University of Minnesota Medical School, Minneapolis, Minnesota 55455 ABSTRACT motive effects on cellular differentiation, there is a growing awareness of the carcinogenic potential of both natural and Both synthetic and natural estrogens have been studied for synthetic estrogens (4, 14, 18, 21). At present, there are no their ability to induce renal carcinomas in castrated male ham detailed studies demonstrating the relative carcinogenic activity sters after 9.0 months of treatment. Tumor foci were detected of various estrogens within a given animal tumor system. in frozen serial sections stained histochemically for esterase The hamster renal adenocarcinoma has provided a unique activity. Both diethylstilbestrol (DES) and 17.^-estradiol had equal model to investigate both hormonal and carcinogenic aspects of ability (100%) to induce renal tumors [~20.5 ±3 (S.E.) tumor estrogen-induced tumorigenesis (5). Early studies in our labora foci] in these animals. Hexestrol induced the same incidence and tory strongly support a hormonal role for the induction of these number of renal carcinoma foci as DES or 17/3-estradiol. How tumors by estrogen. The demonstration of a specific high-affinity ever, a-dienestrol and DES 3,4-oxide showed an 86 to 88% estrogen receptor in renal cytosols of castrated hamsters which incidence of renal tumors in hamsters (~10.8 ±3). When equilin is enhanced nearly 3.5-fold after prolonged estrogen administra and d-equilenin, components of therapeutic conjugated estro tion supports this contention (8). Pertinent also is the finding that gens, were tested, only equilin had a 76% incidence of renal estrogen-induced renal tumorigenesis in the hamster can be tumor foci (5.5 ±0.9). The ability of these stilbene and steroidal completely blocked by those antiestrogens which inhibit mark estrogens to compete for renal tumor estrogen receptor gener edly estrogen receptor-binding activity (8). Moreover, estrogen ally correlated well with their ability to cause renal tumorigenesis treatment resulted in at least a 10-fold rise in cytosolic proges in the hamster with one notable exception. Although ethinyl terone receptor concentration in the untransformed kidney, and estradiol competed as well as did DES or 17/3-estradiol for this elevated receptor can be modulated by antiestrogens, an- estrogen receptor, had similar ability to induce renal progester drogens, and some synthetic progestins (11,15). However, our one receptor, and led to similar high serum prolactin levels as recent result concerning the inhibition of estrogen-induced kidney either DES or 17/3-estradiol, it had only weak carcinogenic activity transformation by »-naphthoflavone cannot be readily explained (21%) in the hamster kidney (0.6 ± 0.5 foci). These data by hormonal mechanisms alone (12). In addition, the marked represent the first detailed analysis of the relative carcinogenic suppression of kidney aryl hydrocarbon hydroxylase activity, not activity of different estrogens within a given tumor-inducing evident in the liver, supports a pertinent role for P-450 multisub- system, and based on the carcinogenicity data of hexestrol and strate monooxygenases in effecting the tumorigenic response of a-dienestrol presented herein, they suggest that epoxidation of the hamster to estrogens (14). the olefinic double bond and the p-quinone metabolite of DES In an effort to discern the structural and hormonal require probably are not involved significantly in its carcinogenic activity. ments for estrogen carcinogenicity in the hamster kidney and to Moreover, the poor carcinogenic activity of ethinyl estradiol in elucidate the significance of particular intermediates in a path this system, despite strong estrogenicity. suggests that estronic way, we have undertaken to determine the carcinogenic potential activity alone may not be sufficient to effect renal tumorigenesis of various synthetic and natural estrogens to transform the in the hamster. hamster kidney. These data are correlated with the ability of these estrogens to compete for estrogen receptor binding, their INTRODUCTION ability to induce renal progesterone receptor in the untrans formed kidney, and in some instances their ability to elevate For over a decade, evidence has continued to accumulate serum PRL4 levels as determined by the Nb2 rat lymphoma cell indicating a causal link between estrogens and a variety of human bioassay. cancers involving vaginal, hepatic, breast, endometrial, and cer vical organ sites (1-3,17, 23). Although some of the tumorigenic effects of estrogen exposure are due undoubtedly to their pro- MATERIALS AND METHODS 1This investigation was supported by Grant CA 22008 from the National Cancer Chemicals and Reagents. 17/3-[2,4,6,7-3H]Estradiol (115 Ci/mmol), Institute, NIH. Department of Health and Human Services, and the General Medical [1,2,6,7-3H]progesterone (103 Ci/mmol), and [17a-mef/?y/-3H]R5020 (86 Research Fund of the Veterans Administration. Presented in part at the 24th Ci/mmol) were obtained from New England Nuclear, Boston, Mass. Congress of the European Society of Toxicology, Satellite Symposium on the Radioinert estradiol and progesterone, Chromatographie grade, were Mechanisms of Estrogen Carcinogenicity: Rote of Metabolites, Rome, Italy, March obtained from Calbiochem-Behring, San Diego, Calif., and all other 31,1983. 2To whom requests for reprints should be addressed, at the Medical Research nonlabeled steroids were purchased from Sigma Chemical Co., St. Louis, Laboratories, Veterans Administration Medical Center, 54th Street and 48th Avenue Mo. DES 3,4-oxide was synthesized by a procedure modified from that South, Minneapolis, Minn. 55417. 3 Recipient of Grant AM 26962 of the NIH, Department of Health and Human Services. 4 The abbreviations used are: PRL, prolactin; DES, diethylstilbestrol; FM, Fisch Received May 6, 1983; accepted August 2. 1983. er's medium. 5200 CANCER RESEARCH VOL. 43 Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1983 American Association for Cancer Research. Carcinogenic Activity of Estrogens in Hamster Kidneys of Metzler and McLachlan (22) using m-chloroperbenzoic acid. Trizma and some lower-affinity binding components. Protein concentration of base, Norit A, dextran 80, dithiothreitol, and 2-mercaptoethanol were the tumor cytosols (5 mg/ml) was determined by the method of Lowry supplied by Sigma. Ultrapure sucrose (RNase free) was obtained from et al. (16) using bovine serum albumin as a standard. Radioactivity Schwarz/Mann, Orangeburg, N. Y. Ovine PRL (NIAMDD-o-PRL-14, 31 sample measurements were assessed in a Packard Tri-Carb liquid scin ID/mg) was a gift of the National Hormone and Pituitary Program, tillation spectrometer (Model 3375; Packard Instrument Co., Inc., Down Baltimore Md. FM for leukemic cells of mice, fetal bovine serum, L- ers Grove, III.) with a counting efficiency for tritium of about 43% (10). glutamine, and antibioticiantimycotic (1OOx) were purchased from Grand Bioassay of Hamster PRL. The Mb, node rat lymphoma cells were Island Biological Co., Grand Island, N. Y. Horse serum was supplied by generously provided by Dr. R. L. Nobel and Dr. C. T. Beer, University of Sterile Systems, Logan, Utah. British Columbia, Vancouver, British Columbia, Canada. Cells were main TED (0.01 M Tris-HCI, 0.0015 M EDTA, and 0.001 M dithiothreitol) and tained in suspension culture and were conditioned and distributed for TEDG (TED plus 5% glycerol) buffers, pH 7.4, were prepared at room bioassay as has been described previously in detail (25). Ovine PRL was temperature as indicated previously (8,11,13). diluted with FM to give a final concentration of 0.05 to 5.0 ng/ml, sterilized Animals and Renal Tumor Induction. Adult castrated-male Syrian by filtration (Acrodisc, 0.22 Mm; Gelman Sciences, Inc.), and added to golden hamsters (LAK:LVG, outbred strain) were obtained from Charles 24-well tissue culture plates (Flow Laboratories, Inc., McLean, Va.) in River Lakeview Hamster Colony, Wilmington, Mass. Similar hypophysec- duplicate in aliquots of 100 ¿ilorless. Hamster serum samples, obtained tomized male hamsters were also supplied from the same source. All from the orbit of the eye, were diluted 1:10 with FM containing 10% animals weighed between 85 and 95 g (50 to 55 days old) and were horse serum, filtered as above, and added to wells in aliquots of 50 and acclimated at least 1 to 2 weeks prior to treatment or use. Pure estrogen 100 n\. The plates were incubated at 37° for 72 hr in a humidified pellets, prepared without binder, were implanted in the shoulder region incubator in an atmosphere of 95% air:5% CO2. Determination of cell as described earlier (9). All animals were exposed to different estrogens number was made on a Coulter Counter. During the course of these for 9.0 to 9.2 months for induction of renal carcinomas. To maintain experiments, the dose of ovine PRL which induced a 50% inhibition of constant estrogen levels in these hamsters, hormone pellets were reim- cell ranged between 0.45 and 1.0 ng/ml and averaged 0.56 ±0.15 ng/ planted at 3-month intervals. Ethmyl estradici pellets, however, were ml (8 assays). Within assay coefficient of variation based upon replicate absorbed more rapidly and were therefore reimplanted every 2 months. determination of standards ranged from 1.0 to 8.0%, and between-assay For progesterone receptor induction, hamsters were treated for 3.0 coefficient of variation was 20% (24). Since bioassay of the serum months with various estrogens as reported in detail elsewhere (11,15).