Androgen Dependent Mammary Gland Virilism in Rats Given the Selective Estrogen Receptor Modulator LY2066948 Hydrochloride

Abstract

A selective estrogen receptor modulator (SERM) is a nonsteroidal compound with tissue specific estrogen receptor (ER) agonist or antagonist activities. In animals, SERMs may produce morphologic changes in hormonally-sensitive tissues like the mammary gland. Mammary glands from female rats given the SERM LY2066948 hydrochloride (LY2066948) for 1 month at ≥ 175 mg/kg had intralobular ducts and alveoli lined by multiple layers of vacuolated, hypertrophied epithelial cells, resembling in part the morphology of the normal male rat mammary gland. We hypothesized that these SERM-mediated changes represented an androgen-dependent virilism of the female rat mammary gland. To test this hypothesis, the androgen receptor antagonist flutamide was co-administered with LY2066948 (175 mg/kg) to female rats for 1 month. Female rats given SERM alone had hyperandrogenemia and the duct and alveolar changes described here. Flutamide cotreatment did not affect serum androgen levels but completely blocked the SERM-mediated mammary gland change. In the mouse, a species that does not have the sex-specific differences in the mammary gland observed in the rat, SERM treatment resulted in hyperandrogenemia but did not alter mammary gland morphology. These studies demonstrate that LY2066948 produces species-specific, androgen-dependent mammary gland virilism in the female rat.

Introduction

Mammary gland development and morphology is dependent on the balance of several trophic hormones including androgens, estrogens, progesterone, growth hormone, corticosterone, and prolactin (Goldman et al., 1976; Russo et al., 1989). Estrogens, growth hormone and prolactin control mammary gland duct development in female nongravid rats and the lobuloalveolar proliferation observed in pregnancy is dependent on the combined effects of estrogen and progesterone (Russo et al., 1989). The role of androgens in rat embryonic mammary gland development is well characterized (Goldman et al., 1976); however, less is known about the importance of androgens in adult rat mammary gland physiology.

The rat mammary gland is unique because of striking sex-dependent differences in its morphologic appearance (see review in Cardy, 1991). Female rat mammary glands have a tubuloalveolar morphology characterized by frequent interlobular and intralobular ducts and fewer alveoli. Ducts and alveoli are lined usually by 1 layer of low cuboidal epithelium with scant to moderate amounts of cytoplasm depending on the stage of the estrous cycle (Masso-Welch et al., 2000) (Figure 1A,1C). In contrast, male rat mammary glands have a lobuloalveolar morphology characterized by lobules of tightly clustered alveoli and fewer ducts. Ducts and alveoli are lined by multiple layers of large, cuboidal or columnar epithelial cells with abundant vacuolated cytoplasm (Figure 1B,1D). Sex differences in the circulating levels of various hormones may be important for the expression of this sex-dependent morphologic phenotype in rats (Cardy, 1991).

Selective Estrogen Receptor Modulators (SERMs) are nonsteroidal chemicals that bind the estrogen receptor (ER) and impart tissue selective antagonist or agonist effects. The mechanism of this selectivity is based on target tissue differences in ER expression and ligand mediated effects on ER conformational state or recruitment of coregulatory proteins (Katzenellenbogen and Katzenellebogen, 2002; Shang and Brown, 2002; Riggs and Hartmann, 2003). SERMs have consistently demonstrated ER antagonist effects in mammary gland-derived cell lines and in rodent mammary glands. These antagonist effects include blockade of cell proliferation (Simard et al., 1997; Geiser et al., 2005) and estrogen receptor-mediated transcription in estrogen sensitive cell lines (Tremblay et al., 1998; Labrie et al., 2001; Shang and Brown, 2002), attenuation of mammary gland lobuloalveolar development in rodent animal models (Gotze et al., 1984; Luo et al., 1997), and protection from spontaneous epithelial proliferation or chemically induced mammary gland tumors in rodents (Luo et al., 1997; Labrie et al., 2001, 2003).

Mammary glands from female rats given ≥ 175 mg/kg/day of the SERM LY2066948 hydrochloride (LY2066948) for one month had the expected tubuloalveolar appearance of a female but also had intralobular ducts and alveoli lined by multiple layers of vacuolated, hypertrophied epithelial cells. The stratified appearance of the ductal and alveolar epithelium and the hypertrophic, vacuolated epithelial cells observed in the SERM-treated female rats had some resemblance to the normal morphology of the mammary gland of control male rats. Intralobular ducts and alveoli in normal males are usually lined by >2 cell layers and have epithelial cells with abundant cytoplasm that is often vacuolated. Based on these observations, we hypothesized that the SERM-mediated mammary gland alteration represented an androgen-dependent virilism of the female rat mammary gland, unique to the rat because of the morphologic differences in the male and female rat mammary gland. To test the virilism hypothesis, we evaluated the hormonal status and mammary gland changes in female rats given LY2066948 alone or co-administered with the androgen receptor antagonist flutamide. We also examined the hormonal and mammary gland effects of LY2066948 in female mice, a species that does not have the sex-dependent morphologic differences observed in the rat mammary gland.

Methods

Compounds and Vehicles

LY2066948 hydrochloride (LY2066948) was stored at room temperature and dosed as an oral suspension. Suspensions for oral dosing were prepared weekly and refrigerated. The vehicle was 1% carboxymethylcellulose sodium, 0.25% polysorbate 80, and 0.05% simethicone (Antifoam 1510-US, Dow Corning, Midland, MI) in purified water. Doses were corrected for 93.2% potency. Flutamide (Sigma-Aldrich, St. Louis, MO) was stored at room temperature and dosed by subcutaneous injection as a solution in corn oil. Flutamide solutions were made daily in corn oil with an assumed potency of 100%.

Animals

Approximately 8-week-old CD-1 mice or Fischer 344/NHsd rats were purchased from Harlan Sprague—Dawley Inc. (Indianapolis, IN), and housed in individual stainless steel cages. Animals were given ad libitum access to laboratory chow (No. 5002, Purina Mills, Inc., St. Louis, MO) and chlorinated potable water. All animals were allowed at least 1 week for acclimation to the animal facilities and diet before being used in any experiment. Animal room thermostats were set to maintain a temperature of 69°F to 75°F. The environmental control system was set to maintain a minimum relative humidity of 30% and a maximum of 70%. The photoperiod was approximately 12 hours light and 12 hours dark, changing at approximately 0600 and 1800 hours. Prior to treatment initiation, animals were randomly assigned to treatment groups by body weight stratification using a partitioning algorithm. The Institutional Animal Care and Use Committee of Eli Lilly and Co. approved all experimental protocols.

Experimental Design

LY2066948 1 month rat GLP toxicity study

The initial rat 1-month toxicology study was done according to Good Laboratory Practices (GLP). While the study was a standard 1-month rat toxicology study, for the purposes of this manuscript, only mammary gland methods and results are reported. Fifteen female Fischer 344 (F344) rats were assigned to control and mid-dose (175 mg/kg/day) groups and 10/group were assigned to low (20 mg/kg/day) and high-dose (1500 mg/kg/day) groups and dosed for 1 month. For Groups 1 and 3, the last 5 animals from the toxicity were assigned to the 1-month reversibility phase of the study.

Doses selected for this study were based on results from 1-and 2-week non-GLP (nGLP) studies in rats in which doses up to 1500 mg LY2066948/kg were administered. There were no mammary gland changes in these preliminary studies and findings were limited to changes in the female reproductive tract that represent expected pharmacology (uterine atrophy) and class effects of SERMs described elsewhere (Long et al., 2001; Geiser et al., 2005).

1-Month flutamide rat mammary gland and hormone study

Treatment groups included Fischer 344 female rats for hormonal and pathology evaluation (8/group) and toxicokinetics (3/group). The LY2066948 suspension for oral dosing was prepared at a concentration of 17.5 mg/mL and refrigerated. The flutamide solution was made at a concentration of 8 mg/mL in corn oil and stored at room temperature. Groups were given one of the following for 1 month: purified water plus corn oil (control group), purified water plus 20 mg/kg flutamide, 175 mg LY2066948/kg plus corn oil (SERM alone), or 175 mg LY2066948/kg plus 20 mg/kg flutamide. Individual doses were adjusted weekly for changes in body weight.

Purified water and LY2066948 were given once daily by oral gavage at a dose volume of 10 mL/kg, and corn oil and flutamide were given twice daily approximately 8 hours apart by subcutaneous injection at a dose volume of 1.25 mL/kg. The total daily dose of flutamide was 20 mg/kg. The flutamide dose was chosen based on published data in rats (Luthy et al., 1987; Gallagher et al., 1996; O’Connor et al., 1998). The 175 mg LY2066948/kg dose was chosen because this dose produced mammary gland alteration in female rats in the GLP study following 1 month of treatment.

1-Month mouse mammary gland and hormone study

Treatment groups included female CD-1 mice for hormonal and pathology evaluation (16/group) and toxicokinetics (6/group). Purified water was administered to control mice. Daily doses of 300 or 1000 mg LY2066948/kg of body weight were administered to treated mice orally by gavage. The LY2066948 suspensions for oral dosing were prepared at concentrations of 32.2 or 107.3 mg LY2066948/mL and refrigerated. The dose volume was 10 mL/kg for all groups. Individual doses were adjusted weekly for changes in body weight.

Doses of LY2066948 were selected based on 4-day tolerability and toxicokinetic studies in mice and were projected to be well tolerated and to produce systemic exposures observed at doses that produced mammary gland alteration in female rats.

In life and hormonal evaluations

For the rat flutamide and mouse studies, all rats or mice were observed daily for mortality and moribundity. On the day of necropsy, terminal body weights were taken and all animals were euthanized by decapitation approximately 2 hours after dosing. Trunk blood was collected from each animal, processed to serum and stored at or below −60°C. Serum samples were assayed for estradiol, progesterone, luteinizing hormone (LH), prolactin (PRL), dihydrotestosterone (DHT), testosterone, and/or androstenedione. Assay methods are summarized in Table 1.

Gross and microscopic pathology

For the GLP study and rat flutamide and mouse studies, a necropsy was conducted on each rat or mouse. The right inguinal mammary gland was collected from each animal and preserved in 10% neutral buffered formalin, trimmed, processed through graded alcohol and clearing agent, infiltrated and embedded in paraffin, sectioned at 4–5 m, and stained with hematoxylin and eosin. Sections of mammary gland were examined microscopically from all animals.

Toxicokinetics

For the rat flutamide and mouse studies, concentrations of LY2066948 were determined by LC/MS/MS at 1, 2, 4, 8, and 24 hours postdose on Day 30. Plasma was obtained from orbital (up to 2 samples/rat) or cardiac (1 sample/mouse) blood collected in tubes containing EDTA from isoflurane anesthetized animals in each treatment group. Rats and mice were euthanized by exsanguination after the last orbital (rat) or the cardiac (mouse) blood sample collection. Maximum plasma concentration (C_max) and area under the plasma concentration time curve (AUC_0–24hr) were calculated for LY2066948 and LY2066948-conjugated metabolites.

Data evaluation and statistical analysis

Qualitative pathology assessments were made using a 5-point grading scale, from least to most severe, of minimal, slight, moderate, marked, and severe. For hormone data, mean, standard error, and sample size for each treatment group were reported at each time point. For each hormone, mean values were analyzed using a 1-factor analysis of variance (ANOVA) (Winer, 1971). The mean of each treatment group was compared to the mean of the control group (Group 1) using Dunnett’s t-test (Dunnett, 1964) performed at the 0.05 significance level. Shapiro–Wilk’s test (Shapiro and Wilk, 1965) for normality was performed at the 0.05 significance level to detect outlying observations, and Levene’s test (Levene, 1960) for homogeneity of variance was performed at the 0.05 significance level to assist in interpretation of treatment effects. If either of the 2 diagnostic tests were significant, then the analysis was performed on the rank average transformed data.

Results

Mammary Gland Ductular and Alveolar Alteration in SERM-Treated Female Rats

In the GLP 1 month toxicity study, mammary glands from female rats given 175 or 1500 mg LY2066948/kg had the normal tubuloalveolar appearance of a female control rat; however, there was an increased density of alveoli, and intralobular ducts and alveoli had multiple layers of vacuolated, hypertrophied epithelial cells. Ductal changes were largely restricted to intralobular ducts rather than interlobular ducts. In some intralobular ducts the stratified epithelium was thrown into folds that extended into the lumen. Most alveolar and several smaller ductal lumina were filled with epithelial cells. Epithelial cells had increased amounts of cytoplasm that was basophilic or eosinophilic and often contained small to medium sized clear vacuoles. Nuclear size and shape varied slightly and mitoses were infrequent. The mammary gland alteration was diagnosed as minimal to slight atypical hyperplasia and observed in approximately 75% of rats given ≥ 175 mg/kg. The alteration was completely reversible in rats given 175 mg/kg after a 1-month drug-free period. The 175 mg/kg group was the only group evaluated for reversibility. Mammary glands from females given 20 mg/kg and from all compound-treated males were normal.

Hyperandrogenemia and Androgen-Dependent Mammary Gland Alteration in SERM-treated Female Rats

Female rats were given LY2066948 (175 mg/kg) alone or in combination with the androgen receptor antagonist flutamide (20 mg/kg) for 1 month. Flutamide co-administration did not alter systemic exposure to LY2066948 or alter the LY2066948-induced changes in the hormones monitored in this study (Table 2). The mammary gland alteration observed in the GLP study occurred in females given SERM alone (Figure 2B). Flutamide coadministration completely blocked the mammary gland alteration and mammary glands had morphologic characteristics comparable to control rats (Figure 2A, 2C). Rats given flutamide alone had normal female mammary gland morphology (Figure 2D).

LY2066948-treated rats had significantly increased levels of circulating androgens (Figure 3A–C) and elevations in circulating levels of both estradiol and LH but not prolactin and progesterone (Figure 3D–G). The hormonal changes were not altered with flutamide cotreatment and rats given flutamide alone had a hormonal profile comparable to vehicle-treated controls (Figure 3A–G).

Hyperandrogenemia without Mammary Gland Virilization in SERM-Treated Female Mice

Female mice given either dose of LY2066948 had mammary gland morphology comparable to female mice given vehicle (Figure 4). Mouse mammary gland morphology was characterized by a single layer of epithelial cells with high nuclear/cytoplasmic ratios lining ducts and alveoli. Plasma exposure (AUC_{0–24 hr}) to LY2066948 was only slightly higher at 1000 mg/kg (3.3 μg·hr/mL) than at 300 mg/kg (2.6 μg·hr/mL), and was approximately 5-fold less at 1000 mg/kg than the exposure observed in the rat flutamide study for female rats given 175 mg/kg (17.3 μg·hr/mL). Despite lower systemic exposure in mice, the hormone changes were similar to that observed in rats given 175 mg LY2066948/kg. A 300 mg LY2066948/kg dose significantly elevated mean circulating LH while 1000 mg LY2066948/kg significantly elevated mean circulating LH, estradiol and testosterone without affecting progesterone levels (Figure 5A–D).

Discussion

In toxicology studies, the morphologic changes observed in rat reproductive tissues often reflect xenobiotic-induced changes in the hormonal milieu. The rat is an especially interesting preclinical model for these hormonal perturbations because of the striking sex-dependent morphologic differences in the mammary gland. Our data show how a change in the hormonal balance of the female rat resulted in distinct changes in mammary gland morphology. Specifically, SERM-treated female rats had hyperandrogenemia and virilism of the female rat mammary gland that was completely blocked by cotreatment with the androgen receptor antagonist flutamide. To our knowledge this is the first time that SERM-mediated virilism of the female male rat mammary gland has been described and demonstrated to be androgen-dependent.

Our work showed that androgen stimulation of the female rat mammary gland in ovary intact rats was required for the LY2066948-mediated virilism. However, our studies were not designed to determine whether hyperandrogenemia by itself could produce virilism of the mammary gland in ovary-intact adult female rats. It is known that in fetal rats with little or no circulating estrogens, androgens are critical determinants in the differentiation of the mammary gland to the male phenotype (Goldman et al., 1976). Also, androgen treatment of ovariectomized rats that do not have circulating ovarian hormones results in virilism of the female mammary gland (Sourla et al., 1998). However there are no descriptions of mammary gland virilism in ovary-intact rats caused by nonaromatizable androgens like dihydrotestosterone. Likewise, LY2066948 and the SERMs tamoxifen and EM-652 are potent ER antagonists in vitro in MCF-7 mammary gland cell lines and in vivo in the mammary gland (Gotze et al., 1984; Labrie et al., 2001; 2003; Geiser et al., 2005), but, to our knowledge, estrogen receptor antagonists have not been reported to produce mammary gland virilism in rats (Center for Drug Evaluation and Research 1977, 1997; Watanabe et al., 1980; Greaves et al., 1993). It is possible that in the LY2066948-treated rats in the present study both ER antagonism and hyperandrogenemia were necessary for mammary gland virilism to occur. Additional studies would be necessary to determine if both ER antagonism and AR stimulation are necessary for the SERM-mediated virilism described here.

In the SERM rat studies described here, changes in the circulating levels of other hormones could modulate the morphologic appearance of the mammary gland. For example, in our experience, xenobiotics that increase levels of prolactin in rats cause lobular hyperplasia of the mammary gland in females and feminization of the mammary gland in males (unpublished observations). Also, prolactin and progesterone are important for the ductal and lobuloalveolar development of the female mammary gland, respectively (Russo et al., 1989). To understand whether changes in circulating levels of prolactin and progesterone contributed to the morphologic changes observed in SERM treated rats, we monitored these hormones in our studies. Prolactin and progesterone levels in SERM-treated rats were comparable to vehicle controls and rats given SERM coadministered with flutamide demonstrating that changes in these hormones was not important in the pathogenesis of the mammary gland virilism.

The high doses used in toxicology studies can result in loss of selectivity for the pharmacologic target and unanticipated activity at other receptors. If LY2066948 was an AR agonist at the high systemic exposures achieved in our toxicology studies, the mammary gland virilism could be a direct effect of the compound. Both in vitro and in vivo data suggested that a direct effect of LY2066948 on AR was unlikely. Testosterone treatment of intact male rats results in Leydig cell atrophy of the testis and increased prostate and seminal vesicle weights (O’Connor et al., 2000) neither of which were observed in male rats given LY2066948 for 1 month in the GLP study. Also in a standard in vitro nuclear hormone screen, LY2066948 had no measurable AR binding (data not shown).

The hyperandrogenemia observed in rats given 175 mg LY2066948/kg for 1 month in the present study was considered secondary to SERM effects on the hypothalamic-pituitary-ovarian (HPO) axis. In contrast to the lack of LH effects at pharmacologic doses (Geiser et al., 2005), female rats given LY2066948 at the dose used in the present studies have elevated levels of LH resulting in hyperstimulation of the ovary. The effects on the HPO axis were expected and occur because of ER antagonism in the hypothalamus and loss of estrogen negative feedback on LH secretion (Long et al., 2001; Crouse et al., 2003). In the rat, estrogen and androgen synthesis occur in the ovary from a common intermediate, the androgen androstenedione (Stryer, 1999). The sustained LH levels stimulate the ovary to synthesize androstenedione that is converted to testosterone and estrogens (Yuan, 1991). This was considered the mechanism for hyperandrogenemia and hyperestrogenemia in the present studies with LY2066948.

The principal changes in female rats with SERM-mediated mammary gland virilism were an increase in the number of layers of epithelial cells lining ducts and alveoli and epithelial cell hypertrophy and vacuolation. This suggests that androgens may control the function, differentiation, proliferation, and/or apoptosis of ductal and alveolar epithelial cells in the adult rat female mammary gland. In the female rat mammary gland, stem cells for ductal and alveolar epithelial cells are present in the terminal end buds (TEB) (Dulbecco et al., 1982) and androgen and estrogen receptors are present on the epithelial cells of ducts and acini in female rats (Pelletier, 2000, unpublished observations). It is well established that estrogens promote terminal end bud development and duct elongation and that progesterone promotes duct enlargement and alveolar formation, but the role of androgens in this process is not defined (Russo et al., 1989). It is possible that androgens interact with TEB stem cells in female rats and modulate stem cell differentiation and duct and alveolar bud formation. Androgens are known to inhibit apoptosis in male secondary sex glands such as the prostate by inducing the expression of apoptosis control proteins (Tenniswood et al., 1992; Chang et al., 2002; Omezzine et al., 2003). Androgens may induce similar proteins in intralobular ductal and alveolar bud epithelial cells in SERM treated female rats resulting in the mammary gland morphology observed in normal males and LY2066948-treated female rats.

Mammary glands from male and female mice do not have the morphologic differences observed in male and female rat mammary gland. Female mice given 1000 mg LY2066948/kg had no histologic changes in the mammary gland. While mice given 1000 mg LY2066948/kg had an approximately 5-fold lower systemic exposure than female rats given a dose that caused mammary gland virilism (175 mg/kg), mice had comparable hormonal alterations. It is unlikely that the lower systemic exposure was the explanation for the lack of LY2066948-mediated mammary gland virilism in mice. Instead, mammary gland virilism probably did not occur because the mammary glands of male and female mice do not have the morphologic differences observed in rats (Cardy, 1991) and the adult mouse mammary gland does not respond to androgens (Kratochwil, 1977). The morphologic changes observed in the virilized mammary gland of LY2066948-treated rats would also not be expected in humans and other preclinical animal species like dogs and nonhuman primates that do not have the sex-specific differences in mammary gland morphology observed in the rat mammary gland (Cardy, 1991).

In the GLP 1-month LY2066948 rat study, the diagnostic terminology used for the female mammary gland alteration was atypical hyperplasia. Atypical hyperplasia is a specific diagnosis that implies a preneoplastic condition in animals and man. As a diagnostic term, atypical hyperplasia of the female rat mammary gland is part of the Standardized System of Nomenclature and Diagnostic Criteria for Toxicologic Pathology (SSNDC) (Mann et al., 1996) and is characterized as focal irregular proliferation of epithelium within ducts or alveoli that is distinguished by cellular atypia. Features of cellular atypia include cellular hypertrophy, increased cytoplasmic eosinophilia, basophilia, and/or vacuolation, and vesicular or hyperchromatic nuclei. The characteristics of hyperplasia and cellular atypia listed in the SSNDC for atypical hyperplasia in female rats are similar to the normal characteristics of ducts and alveoli in male rat mammary glands or, in our case, a virilized female mammary gland. To our knowledge, no standard terminology has been established for mammary gland virilism in female rats. We propose that glandular virilization would be an appropriate morphologic diagnosis for virilism of the female rat mammary gland.

In summary, we describe LY2066948-mediated hyperandrogenemia and androgen-dependent virilism of the female rat mammary gland and demonstrate the importance of the mammary gland as an indicator of hormonal balance in the rat. The mammary gland virilism was characterized by an alteration of the female mammary gland towards the morphology of the normal male that was completely blocked by coadministration with the androgen receptor antagonist flutamide. Because of the rat-specific nature of the mammary gland virilization, these morphologic changes did not occur in LY2066948-treated mice despite hyperandrogenemia. These data demonstrate the species-specific nature of this SERM-mediated mammary gland virilism.