carcinogen-induced mammary tumorigenesis in female rat offspring. LEENAHILAKIVI-CLARKE12, ELIZABETH CHO1, IGHOVIE ONOJAFE1,. MARGARITA R A ...
ONCOLOGY REPORTS 6: 1089-1095, 1999
Maternal exposure to genistein during pregnancy increases carcinogen-induced mammary tumorigenesis in female rat offspring LEENAHILAKIVI-CLARKE 12 , ELIZABETH CHO1, IGHOVIE ONOJAFE1, MARGARITA R A YG AD A 1 2 and ROBERT CLARKE1'3 Lombardi Cancer Center, Department of Psychiatry and Department of Physiology and Biophysics, Georgetown University, 3970 Reservoir Rd, NW, Washington, DC 20007, USA Received May 5, 1999; Accepted June 6, 1999
Abstract. A high estrogenic environment in utero may increase subsequent breast cancer risk. It was therefore determined whether a maternal exposure during pregnancy to the phyto estrogen genistein or zearalenone, both of which exhibit estrogenic activities in vitro and in vivo, alters breast cancer risk among female offspring. Pregnant rat dams were treated daily with subcutaneous injections of 20, 100 or 300 |4g genistein, 20 (ig zearalenone, or vehicle between days 15 and 20 of gestation. The offspring were given 7,12-dimethylbenz(a)anthracene (DMBA) at the age of 2 months to induce mammary tumors. The results indicate that in utero exposure to genistein, but not to zearalenone, dose-dependently increased the incidence of DMBA-induced mammary tumors, when compared with the controls. Tumor growth characteristics were not altered. Prior to the carcinogen administration, the number of estrogen receptor (ER) binding sites, determined using a ligand binding assay, were significantly elevated in the mammary glands of genistein offspring. In contrast, the mammary protein kinase C (PKC) activity was significantly reduced in the genistein offspring. Our results suggest that a maternal exposure to subcutaneous administration of genistein can increase mammary tumori genesis in the offspring, mimicking the effects of in utero estrogenic exposures. Further, increased ER protein levels and reduced PKC activity in the mammary gland may be involved in increasing susceptibility to carcinogen-induced mammary tumorigenesis in rats exposed to genistein in utero.
Correspondence to: Dr Leena Hilakivi-Clarke, Research Bldg., Lombardi Cancer Center, Room W405, Georgetown University, 3970 Reservoir Rd, NW, Washington, DC 20007-2197, USA Abbreviations: DES, diethylstilbestrol; DMBA, 7,12-dimethylbenz(a)antracene; E2, estradiol; ER, estrogen receptor; PKC, protein kinase C Key words: genistein, zearalenone, in utero exposure, estrogen receptor, protein kinase C, mammary tumorigenesis
Introduction Phytoestrogens are plant-derived compounds that can exhibit estrogenic activities (1). There are several groups of plant estrogens, including isoflavones, e.g., the genistein present in soy-based foods, and mycotoxins derived from fungal molds, e.g., the resorcyclic acid lactone zearalenone present as a contaminant in stored grains. Genistein is suggested to reduce breast cancer risk (2), and there is some animal (3) and human (4,5) evidence to support the claim. However, a recent work by Hsieh et al (6) indicates that dietary genistein significantly stimulates the growth of human breast cancer cell MCF-7 tumors in ovariectomized athymic mice. Zearalenone exposure also promotes spontaneous mammary tumorigenesis in rodents (7). Timing of estrogen exposure may play a critical role in affecting mammary tumorigenesis. High in utero estrogenicity, in particular, may increase subsequent breast cancer risk. Breast cancer risk is elevated in women who had a high birth-weight (8,9) or who are dizygotic twins (10), both of these being associated with a high in utero estrogenic environment (11). Maternal exposure to the synthetic estrogen diethylstilbestrol (DES) also may increase breast cancer risk in daughters (12). In contrast, daughters whose mothers suffered from pre-eclampsia/eclampsia during pregnancy (characterized by low circulating estrogen levels) exhibit a significantly lower breast cancer risk (13). Animal studies provide more direct evidence and show that in utero exposure to physiological levels of estradiol (14), or an exposure either in utero or during early postnatal life (days 1-5) to DES (15,16), increases the incidence of carcinogen-induced or spontaneous mammary tumors. Given the estrogenicity of genistein (17-19) and zearalenone (1,20), and the associations of increased fetal estrogenicity with increased breast cancer risk, we examined whether in utero exposure to these phytoestrogens might affect breast cancer risk. Estrogenic effects of phytoestrogens on the mammary gland are likely to be mediated through their interactions with estrogen receptor (ER), a nuclear transcription factor that regulates several other genes, including protein kinase C (PKC). In particular, high PKC activity might down-regulate ER in human breast cancer
HILAKIVI-CLARKE et at: In utero EXPOSURE TO GENISTEIN AND BREAST CANCER RISK
cells (21,22), although the exact relationship between these two genes is not clear. We investigated whether a fetal exposure to genistein or zearalenone via subcutaneous injections causes long-lasting changes in the ER and/or PKC in the m a m m a r y g l a n d . T h e s u b c u t a n e o u s r o u t e of administration of genistein and zearalenone was used for comparability with previously published data indicating an inhibitory effect on carcinogen-induced mammary tumorigenesis by prepubertal exposure to genistein (23). Materials and methods Animals. Female Sprague-Dawley rats, purchased from Charles Rivers Breeding Laboratories, were obtained at day 10 of gestation. The animals were housed singly, in standard rat plexiglass cages, at a constant temperature (20-22°C) and humidity (60-65%), under a 12-h light-dark cycle (lights on 06.00 h). Two days after the offspring were born, the males were sacrificed and the females cross-fostered. Ten to twelve female pups were housed with each lactating dam. The female offspring were weaned on postnatal day 22, and thereafter housed in groups of 3-5 animals. Our experimental protocol was reviewed and approved by the Georgetown University Animal Care and Use Committee. Phytoestrogen exposure. In experiment 1, the animals were treated daily with either 20 |J,g genistein, 20 \ig zearalenone (both from Sigma Chemical Co., St. Louis, MO), or vehicle (2% DMSO in peanut oil), administered as subcutaneous injections, in a volume of 0.05 ml, between days 15 and 20 of pregnancy. Phytoestrogens were first dissolved in 2% DMSO, and then mixed with peanut oil. This period was chosen for injections because ER can be detected in the fetal mammary glands on day 13 of gestation (24,25), and because the fetal mammary gland is sensitive to estrogenic manipulations occurring between days 15 and 20 (14,26,27). In experiment 2, pregnant rats were injected daily with vehicle, 100 ^.g or 300 \xg genistein from gestation day 15 to day 20. The choice of the phytoestrogen doses approximate the level of human exposure. The doses of genistein and zearalenone represent 0.1 mg/kg in experiment 1, and 0.5-1.5 mg/kg in experiment 2 ( g e n i s t e i n o n l y ) . T h e level of g e n i s t e i n e x p o s u r e is approximately 5 mg/person (-0.1 mg/kg) in Asia (28), and approximately 20 times less (at most -0.005 fJ,g/kg) in the Western Europe. The exposure levels are higher in the USA than in the Europe, due to wide use of soy as a protein source of for example bread in the USA. The human exposure for zearalenone is 1-5 mg (0.02-0.1 mg/kg) (29). Estrogen receptor (ER). To determine whether a maternal exposure to genistein or zearalenone altered ER protein levels in the d e v e l o p i n g m a m m a r y g l a n d s of female offspring, the number of ER binding sites in the offspring's 4th mammary glands, were measured. The rats (n=5 per group and age), which were sacrificed at the ages of 21 and 35 days, had been exposed to 20 jug of genistein or zearalenone in utero through their pregnant mother. Rats sacrificed by day 45, were exposed to vehicle or 300 ju,g genistein in utero. Samples were frozen at -70°C until assay. ER protein levels in the mammary glands were detected using a ligand binding
assay, as described by Nelson et al (30). Tissues were homogenized, and cytosolic and nuclear fractions prepared. This process removed fat from the tissue samples. [ 3 H]estradiol (Amersham) was used as the radiolabeled ligand, and a 200-fold excess of diethylstilbestrol as the cold competitor. Binding data were analyzed by the Ligand software. This assay does not discriminate between ERa and ERß, but effectively measures total ER. Protein kinase C (PKC). Total PKC activity was measured using a kit from Amersham Life Sciences (Arlington Heights, IL) and following the manufacturer's instructions. The 9th mammary glands of 21 and 45-day-old offspring of mothers e x p o s e d to e i t h e r v e h i c l e , 20 jag g e n i s t e i n , or 20 jig zearalenone on days 15 through 20 of pregnancy (n=5 per group), were used. Samples were homogenized in buffer (50 nM Tris/HCl, 0 . 3 % (w/v) -mercaptoethanol 5 mM E D T A , 10 m M E G T A , and 50 |ag/ml p h e n y l m e t h y l sulphonylfluoride, pH 7.5). Blanks were used to correct for non-specific effects of [- 12 P]-ATP and/or its radiolytic decomposition products binding to the paper. The positive controls were extracts of the MDA-MB-231 human breast cancer cells, which have high PKC activity. Inducing and monitoring mammary tumorigenesis. Mammary tumors were induced by administration of 10 mg (approximately 40 mg/kg body weight) 7,12-dimethylbenz(a)anthracene (DMBA) (Sigma, St. Louis, MO). This is a suboptimal dose that produces sufficient tumorigenesis to allow both reductions and increases in the endpoints of tumorigenicity. Most of the tumors induced by 10 mg DMBA are adenocarcinomas (14,31). The carcinogen was dissolved in peanut oil and administered by oral gavage in a volume of 1 ml. Animals were 45 days old in experiment 1 (n=24 per group), and 50 days old in experiment 2 (n=18 in the vehicle group, n=22 in the 100 jug genistein group, and n=27 in the 300 jag genistein group) at the time of DMBA administration. The animals were checked for mammary tumors once per week. The end-points for data analysis were i) latency to tumor appearance, ii) the number of animals with tumors (tumor incidence), iii) the number of tumors per animal (tumor multiplicity) and iv) tumor proliferation. A tumor was designated as proliferating if it increased regularly in size. Tumor sizes were measured by recording the tumor diameters with a caliper and determining the length of the longest axis and the width perpendicular to the longest axis. The animals were sacrificed when detectable tumor burden approximated 10% of total body weight. The surviving animals and animals that did not appear to develop mammary tumors were sacrificed at either 18 (experiment 1) or 20 weeks (experiment 2) after carcinogen administration. We followed the second study for an additional two weeks because of the lower tumor burden. This would not be expected to substantially affect the outcome, and we saw no evidence that the additional observation time affected either the data or its interpretation. Results Effects of phytoestrogens on pregnancy. Maternal exposure to genistein or zearalenone did not affect pregnancy (Table I).
ONCOLOGY REPORTS 6: 1089-1095, 1999
Table I. Effects of genistein and zearalenone on pregnancyrelated parameters.11 Maternal treatments Vehicle
No. of pregnant animals
No. of litters
% successful pregnancies
Weight gain during pregnancy (g)
Length of pregnancy (days)
No. of pups/litter
Pup weight on day 2 (g) % females per litter
, "Pregnant rat dams were exposed to daily subcutaneous injections of vehicle, 20 ug genistein or 20 u,g zearalenone between days 15 and 20 of gestation; b The values are means ± SEM.
Figure 1. The concentration of estrogen receptor (ER) protein, measured using a ligand binding assay technique, in the 4th mammary gland of 21and 35-day-old female offspring of mothers exposed to 20 |ig genistein, 20 p.g zearalenone or vehicle, or 45-day-old female offspring of mothers exposed to 300 u.g genistein or vehicle during pregnancy. Means ± SEM obtained from 5 mammary glands per group are shown. Significantly different from the vehicle group: 'p