Anin VivoBattery for Identifying Endocrine Modulators That Are Estrogenic or Dopamine Regulators

FUNDAMENTAL AND APPLIED TOXICOLOGY ARTICLE NO.

33, 182– 195 (1996)

0156

An in Vivo Battery for Identifying Endocrine Modulators That Are Estrogenic or Dopamine Regulators1 JOHN C. O’CONNOR, JON C. COOK,2 SUZANNE C. CRAVEN, CAROLYN S. VAN PELT,

AND JOHN

D. OBOURN3

Haskell Laboratory for Toxicology and Industrial Medicine, E.I. du Pont de Nemours & Company, P.O. Box 50, Elkton Road, Newark, Delaware 19714 Received April 23, 1996; accepted June 25, 1996

An in Vivo Battery for Identifying Endocrine Modulators That Are Estrogenic or Dopamine Regulators. O’CONNOR, J. C., COOK, J. C., CRAVEN, S. C., VAN PELT, C. S., AND OBOURN, J. D. Fundam. Appl. Toxicol. 33, 182– 195. We have combined several endpoints into a single 5-day in vivo screening procedure to identify estrogenic compounds and dopaminergic modulators, both of which play important roles in enhancing mammary tumorigenesis in rodents. The endpoints evaluated as markers of estrogenicity included increases in uterine fluid and vaginal cornification incidence, serum prolactin levels, uterine weight, uterine epithelial cell height, uterine stromal cell proliferation, and uterine progesterone receptor (PR) number and decreases in uterine estrogen receptor (ER) number. The endpoints evaluated for changes in dopamine regulation included increases in prolactin and decreases in growth hormone levels. The estrogen agonists estradiol (E2) and estriol (E3), the mixed estrogen agonist/ antagonist tamoxifen (TAM), the full antiestrogen ICI-182,780 (ICI), and the dopamine modulators haloperidol (HAL) and reserpine (RES) were tested using a three-time/day (8-hr intervals) intraperitoneal dosing regimen in sexually mature ovariectomized female Crl:CD BR rats. All compounds were evaluated over a range of concentrations. This in vivo battery was used to evaluate the effects of different classes of endocrine modulators on the selected endpoints. For example, the estrogen receptor agonists E2 and E3 display a unique profile based on changes in the uterotrophic endpoints (estrus conversion, uterine fluid imbibition, increases in uterine weight, and uterine endometrial cell proliferation) where full and partial agonists can be distinguished by the magnitude of these responses. Both the estrogen receptor antagonist ICI and the dopamine modulators HAL and RES lack these uterotrophic responses. Dopamine modulators can be distinguished from estrogen receptor agonists by the profile of increased prolactin levels with no uterotrophic changes. Estrogen receptor antagonists can be distinguished from agonists by comparing their effects on ER, PR, and uterotrophic responses. For instance, the full estrogen receptor antagonist ICI decreased ER (to almost 0) and PR levels, but has no uterotrophic effects, while TAM decreases ER (to almost 0) and increases PR with uterotrophic ef1

This work was presented in part at the 1996 Society of Toxicology meeting in Anaheim, CA (Abstract No. 342). 2 To whom reprint requests should be addressed. 3 Supported by a DuPont Postdoctoral Fellowship. 0272-0590/96 $18.00 Copyright q 1996 by the Society of Toxicology. All rights of reproduction in any form reserved.

AID FAAT 2213

/

6k0f$$$201

fects. The most useful endpoints for distinguishing estrogen agonists and dopamine modulators were uterine fluid imbibition, uterine weight, uterine stromal cell proliferation, and serum prolactin levels. In order to distinguish an estrogen agonist from an antagonist, other endpoints, such as receptor levels, are necessary. The advantage of an in vivo screen is that it utilizes a metabolically and physiologically defined system which is especially important with highly integrative systems such as the endocrine system. This battery can be used as a screening tool to identify potential endocrine modulators as well as to identify mode of action following adverse findings in toxicology studies. Last, additional endpoints may be added to identify other classes of endocrine modulators. q 1996 Society of Toxicology

Environmental scientists have been effective in heightening awareness of persistent, bioaccumulative chemicals and their impact on endocrine function. Recent reports have suggested that background levels of agricultural products, industrial chemicals, and other environmental pollutants may contribute to increasing cancer incidence (mammary gland, testis, prostate), decreasing sperm counts, and developmental abnormalities (male urogenital defects) in humans (reviewed in Colborn et al., 1993; Safe, 1995). Decreased reproductive success and developmental defects (masculinization/feminization) in wildlife also have been reported and have been shown to be primarily related to areas of heavy industrial pollution (reviewed in Colborn et al., 1993; Safe, 1995). The incidence of breast cancer has steadily increased in the United States over the past 40 years (reviewed in Henderson et al., 1993). Approximately 10% of women in the United States will develop breast cancer in their lifetime (Dickson et al., 1992; reviewed in Henderson et al., 1993). The hypothesis that environmental estrogens contribute to increased incidence of breast cancer gained considerable attention with the report of Wolff and co-workers (Wolff et al., 1993). They reported that serum levels of DDE were elevated in breast cancer patients. DDE is a major metabolite of DDT, which is an organochlorine pesticide with estrogenic and antiandrogenic activity (Welch et al., 1969). Although more recent data suggest that there is not a correlation between organochlorine exposure and increased risk of hu-

182

09-20-96 11:51:43

ftoxa

AP: Fund Tox

AN IN VIVO BATTERY FOR ENDOCRINE ACTIVITY

man breast cancer (reviewed in Key and Reeves, 1994; Krieger et al., 1994; Adami et al., 1995; Ahlborg et al., 1995), the role of environmental estrogen exposure and breast cancer continues to be intensely studied and debated. The endocrine system has a profound influence on mammary gland tumorigenesis in both humans and rodents (reviewed in Henderson et al., 1993; Welch et al., 1969; Welsch, 1986). For instance, chronic treatment of rodents with estrogens, such as 17b-estradiol and estrone, increases the frequency and decreases the latency period of spontaneous mammary tumors, while early ovariectomy or ovariectomy–adrenalectomy blocks the development of these tumors (Durbin et al., 1966; reviewed in Welsch, 1983; Welsch et al., 1977). In addition, compounds that have no estrogenic activity, but decrease dopaminergic activity of the hypothalamus resulting in an increase in prolactin, have also been associated with an increase in frequency and decrease in latency of mammary tumors in rodents. Examples of these compounds are haloperidol, methyldopa, reserpine, sulpiride, pimozide, and fluphenazine (reviewed in Greaves, 1990). Therefore, the two primary hormones influencing mammary gland tumorigenesis in the rat are estrogens and prolactin (Meites, 1972a,b; Welsch, 1985), which act in synergy (reviewed in Greaves, 1990). In contrast to the rodent, human breast cancer incidence does not appear to be increased by exposure to compounds which elevate prolactin levels, but may be influenced by estrogens (reviewed in Alison et al., 1994; Henderson et al., 1993; Neumann, 1991; FDA, 1977; Schyve, 1978; Kleinberg, 1987). However, while a causative effect of human exposure to estrogen and increased breast cancer incidence has not been clearly defined, a link between estrogen exposure and increased incidence of endometrial cancer has been well documented (Henderson et al., 1993). In addition, exposure to estrogens has also been associated with adverse reproductive and developmental effects (Schardein, 1993). Several in vitro assays, such as competitive receptor binding or the ‘‘E screen’’ test, have been proposed to identify estrogenic agents (reviewed in Soto et al., 1992; Villalobos et al., 1995). The drawbacks of these in vitro approaches are that they focus solely on estrogenicity and ignore a host of other possible mechanisms (while modulation issues of the endocrine system are much more complex). These systems do not have the ability to detect compounds which require metabolic activation or induce estrogen-like responses through secondary mechanisms. Mechanisms for mammary tumorigenesis may include biologically active metabolites, ligand-independent activation of the estrogen receptor through phosphorylation (Ignar-Trowbridge et al., 1992), increases in serum prolactin (Meites, 1972a,b), and other mechanisms that have not yet been identified. To address these concerns, an in vivo battery for endocrine activity has been developed to identify estrogenic/antiestrogenic

AID FAAT 2213

/

6k0f$$$201

09-20-96 11:51:43

183

compounds and dopaminergic modulators. These endocrine endpoints are important for addressing environmental estrogen issues as well as mechanism of action identification for mammary gland tumorigens and risk assessment. For instance, compounds that induce mammary tumors in the rat via increased serum prolactin do not appear to pose a risk to humans while estrogenic compounds do (reviewed in Alison et al., 1994; Henderson et al., 1993; Neumann, 1991; FDA, 1977; Schyve, 1978; Kleinberg, 1987). The uterine weight test was selected as the starting point for addressing endocrine activity because it has been widely used in assessing compounds for estrogenicity and because additional endpoints could be added to potentially enhance detection of other types of endocrine activity (reviewed in Clark and Mani, 1994). Seven-week-old (sexually mature) female rats were selected in order to allow sufficient blood collection for hormonal determinations and uterine tissue for receptor characterization. Ovariectomized rats were chosen to minimize the effects of physiological estrogens and to eliminate confounding of hormonal measurements by cycling rats. A 5-day test duration was selected to optimize detection of weak endocrine activity (reviewed in Clark and Mani, 1994). By using a series of six positive controls we evaluated which of the various selected endpoints were selective, sensitive, and dose-related. The positives included a full estrogen receptor agonist 17b-estradiol (E2) (Clark and Markaverich, 1983), a partial estrogen receptor agonist estriol (E3) (Clark and Markaverich, 1983), a mixed estrogen receptor agonist/antagonist tamoxifen (TAM) (Kangas, 1992), a full estrogen receptor antagonist ICI-182,780 (ICI) (Wakeling et al., 1991), a D2 receptor antagonist haloperidol (HAL) (Goodman, 1990), and a dopamine depletor reserpine (RES) (which acts to lower brain dopamine levels by blocking reuptake of dopamine) (Goodman, 1990). Several markers were evaluated for their ability to detect compounds which were estrogenic, antiestrogenic, or altered the dopamine pathway. The estrogenic/antiestrogenic markers included vaginal cytology (estrus conversion) (reviewed in Edgren, 1994), uterine weight (reviewed in Clark and Mani, 1994), uterine fluid imbibition (Astwood, 1938), uterine endometrial stromal cell proliferation (Kronenberg and Clark, 1985; Martin and Claringbold, 1958), uterine luminal epithelial cell height (Branham et al., 1988a,b, 1993a,b), uterine progesterone (Bhattacharyya et al., 1994), and estrogen receptor number (Bhattacharyya et al., 1994). Measurement of prolactin (Macleod and Lehmeyer, 1974) and growth hormone (Jansson et al., 1985) levels were included as markers for dopamine activity. MATERIALS AND METHODS Test materials. The following materials were obtained from the Sigma Chemical Co. (St. Louis, MO): 5-bromo-2*-deoxyuridine (BrdU), neutralized charcoal, diethylstilbestrol, dextran (No. D-4626), dithiothreitol, EDTA

ftoxa

AP: Fund Tox

184

O’CONNOR ET AL.

(disodium salt), E2, E3, glycerol, HAL, Hepes, leupeptin, RES, sodium bicarbonate, sodium fluoride, sodium molybdate, and TAM. ICI was kindly donated by Dr. A. E. Wakeling (Zeneca Pharmaceuticals, Macclesfield, UK). All other materials were obtained from the following manufacturers: Purina Rat Chow No. 5002, Ralston Purina (St. Louis, MO); [ 3H]17bestradiol, [3 H]promegestone (R5020), and unlabeled promegestone (R5020), New England Nuclear (Boston, MA); osmotic minipumps Model 2ML1, Alza Corp. (Palo Alto, CA); ethanol, Electron Microscopy Science (Fort Washington, PA); pefabloc, Boehringer-Mannheim (Indianapolis, IN); methylcellulose, Fisher Scientific (Springfield, NJ); radioimmunoassays (RIA), Amersham Corp. (Arlington Heights, IL). Test species. Ovariectomized female Crl:CD BR rats, approximately 42 days old, were acquired from Charles River Laboratories, Inc. (Raleigh, NC). Rats were ovariectomized on the day of shipment (41 days old). Upon arrival, rats were housed in stainless steel, wire-mesh cages suspended above cage boards, fed irradiated Purina Certified Rodent Chow No. 5002 checkers, and provided with tap water ad libitum. Animal rooms were maintained on a 12-hr light/dark cycle (fluorescent light) and at a temperature of 23 { 27C and a relatively humidity of 50 { 10%. After a quarantine period of 1 week, rats that displayed adequate weight gain and freedom from clinical signs were divided by computerized, stratified randomization into five groups of 20 rats so that there were no statistically significant differences among group body weight means. Within each group, 14 animals were designated for biochemical/hormonal examination and 6 rats were designated for cell proliferation/morphometric evaluation. Study design. All rats were weighed daily and cage-site examinations were done to detect moribund or dead rats. At each weighing, rats were individually handled and examined for abnormal behavior or appearance. Test compounds were administered by intraperitoneal injection three times daily at approximately 8-hr intervals (8:00 AM, 4:00 PM, 11:00 PM) in order to facilitate detection of short-acting estrogens (Clark and Markaverich, 1983). All experiments were performed with multiple concentrations of test compound suspended in 0.25% methylcellulose vehicle. The following compounds were used in the in vivo battery: E2 (1.0, 2.5, 7.5, and 50 mg/ kg/day), E3 (1.0, 2.5, 7.5, and 50 mg/kg/day), TAM (0.05, 0.25, 0.5, and 1.5 mg/kg/day), HAL (0.01, 0.1, and 1.0 mg/kg/day), RES (0.2, 1.0, and 2.0 mg/kg/day), and ICI (0.05, 0.25, 0.5, and 1.5 mg/kg/day). In addition, a diet restriction experiment was performed to determine which endpoints were body weight dependent in order to evaluate potential confounding from compound-induced decreases in body weight or body weight gain. Rats were given a range of irradiated Purina Certified Rodent Chow No. 5002 meal in order to mimic weight loss seen in previous experiments with the positive controls. The levels of diet restriction were ad libitum control, 17 g chow/day, 14 g chow/day, 11 g chow/day, and 8 g chow/day. Levels of diet restriction were based on calculated food consumption data from previous experiments. Pathological evaluations. On the morning of Test Day /5, rats were anesthetized using CO2 and euthanized by exsanguination. All rats were euthanized between 6:00 and 9:00 AM. The presence of fluid in the uterine horns was recorded. Uteri from the biochemical/hormonal subset were dissected, weighed, and processed to uterine cytosol for measurement of uterine progesterone and estrogen receptor content. Uteri from the cell proliferation subset were dissected, weighed, and fixed in Bouin’s solution. Processed tissues were embedded in paraffin, sectioned at 5 mm, and fixed to slides for staining. Uterine stumps from all rats were saved in 10% neutralbuffered Formalin and processed by the pathologist to confirm the absence of ovarian tissue. Cell proliferation. Six rats from each group were designated for cell proliferation analyses. Rats were anesthetized using isoflurane and implanted (subcutaneously) with Alzet osmotic pumps loaded with 20 mg/ml BrdU dissolved in 0.5 N sodium bicarbonate buffer on Test Day 01. On Test Day /5 the rats were euthanized and the uteri were trimmed, affixed to dental wax, fixed for 2 hr in Bouin’s solution, routinely processed, paraffin embedded, and stained for immunohistochemical analysis of BrdU

AID FAAT 2213

/

6k0f$$$201

09-20-96 11:51:43

TABLE 1 Final Body Weight

Compound

Dosage (mg/kg/day)

Final body weight (g)

Final body weight (% of control)

17b-Estradiol

0 1 2.5 7.5 50

231.3 (2.81) 220.9 (1.87)* 218.1 (2.83)* 209.8 (2.91)* 202.3 (2.24)*

100.0 95.5 94.3 90.7 87.5

Estriol

0 1 2.5 7.5 50

209.9 (3.63) 208.5 (4.09) 205.3 (3.86) 204.2 (3.69) 189.2 (2.94)*

100.0 99.3 97.8 97.3 90.1

Tamoxifen

0 50 250 500 1500

203.4 (3.98) 195.7 (3.23) 185.4 (2.27)* 181.9 (2.19)* 179.4 (1.76)*

100.0 96.2 91.2 89.4 88.2

ICI-182,780

0 50 250 500 1500

204.6 (2.94) 204.3 (2.67) 201.6 (3.47) 205.0 (3.63) 200.7 (2.86)

100.0 99.9 98.5 100.2 98.1

Haloperidol

0 10 100 1000

198.8 (2.97) 197.4 (2.49) 196.0 (2.89) 163.7 (3.80)*

100.0 99.3 98.6 82.3

Reserpine

0 200 1000 2000

189.3 (3.82) 185.6 (3.34) 145.9 (3.77)* 129.9 (3.16)*

100.0 98.0 77.1 68.6

Note. Mean (standard error). * p õ 0.05 compared to ad libitum control.

incorporation into DNA. Immunohistochemical staining was done by the avidin –biotin –peroxidase complex method (Vector Labs, Burlingame, CA) using a monoclonal antibody to BrdU (Becton– Dickinson No. 7580) and DAB substrate (Sigma Fast Tabs). Labeling indices for uterine endometrial stromal cells were determined. One thousand cells were counted. Morphometry. The same six rats used for cell proliferation were used for uterine morphometry. After fixation for 2 hr in Bouin’s solution, the uteri (affixed to dental wax) were washed clear of all Bouin’s solution. Starting at the proximal end (oviduct end) of the uterine horn, four to five sections at 2- to 3-mm intervals were cut perpendicular to the long axis of the horn. All cut sections were embedded proximal end down to avoid double counting any cell, and a consistent pattern of trimming and embedding was used to identify individual sections proximal-to-distal toward the body of the uterus. Epithelial cell height was measured on H&E-stained 5mm sections using an image analyzer with the Olympus Cue-2 Series Software (Galai, Israel). For each animal, uterine epithelial height was taken from the average measurement of a 350-mm basement membrane from each of three uteri cross sections (thus, an approximately 1-mm length of uterine epithelium was measured per animal). Routinely, the second, third, and fourth sections of each uterus were measured on each animal.

ftoxa

AP: Fund Tox

185

AN IN VIVO BATTERY FOR ENDOCRINE ACTIVITY Estrous cycling. Rats assigned to the biochemical/hormonal subset were evaluated for vaginal cytology. Vaginal washes were collected once daily by multiple pipetting of 75 ml of 0.9% sterile saline into the vagina. Slides were air dried and stained by the Wright – Geimsa method and evaluated for conversion out of diestrus (Davis, 1993). Hormonal measurements. Blood was collected from all animals at the time of euthanization. Serum was prepared and stored between 060 and 0807C until analyzed for serum hormone concentrations. Serum prolactin and growth hormone levels were measured by commercially available RIAs (Amersham Corp). Uterine estrogen and progesterone receptor measurements. Uteri from rats assigned to the biochemical/hormonal subset were pooled 2/tube and homogenized (0.1 tissue/1 ml buffer) with a Brinkmann polytron. The homogenization buffer contained 10% glycerol, 25 mM Hepes, 1.5 mM EDTA, 1 mM dithiothreitol, 20 mM sodium molybdate, 50 mM sodium fluoride, 0.2 mM pefabloc, and 0.25 mM leupeptin at pH 7.4. The homogenate was then centrifuged at 68,000g for 90 min at 47C. The resulting supernatant was removed and stored between 060 and 0807C until analyzed for estrogen and progesterone receptor concentration. Receptor concentrations were measured using a single saturating concentration of radiolabeled ligand, 2 nM for estrogen receptor analysis and 12 nM for progesterone receptor analysis. The selection of a saturating radiolabeled ligand concentration was based on a 7-point Scatchard analysis (Scatchard, 1949). Nonspecific binding was determined using a 250-fold excess of nonlabeled competitor, DES for the estrogen receptor and promegestone for the progesterone receptor. Cytosol samples were diluted to 1.0 mg/ml for receptor analysis.

Statistical analyses. Body weights, organ weights, cell proliferation indices, uterine morphometry measurements, serum hormone levels, and progesterone and estrogen receptor levels were analyzed by a one-way analysis of variance (ANOVA). When the corresponding F test for differences among test group means was significant, pairwise comparisons between test and control groups were made with Dunnett’s test. Bartlett’s test for homogeneity of variances was performed and, when significant ( p õ 0.005), was followed by nonparametric procedures. Except for Bartlett’s test, all other significance was judged at p õ 0.05. ID50 and ED50 values were determined using a four-parameter logistic function (Origin 3.5, MicroCal Software, Inc., Northhampton, MA). ED50 values represent the dose at which 50% of the maximal response occurred, while ID50 values represent the dose at which 50% inhibition of the response occurred.

RESULTS

Final body weights. Mean final body weights were decreased in a dose-dependent manner for all test compounds except ICI, and were statistically decreased in rats treated with E2, E3, TAM, HAL, and RES (Table 1). Due to the magnitude of body weight decrements with the test compounds, the potential existed for body-weight-induced changes in the measured endpoints, which would confound interpretation. In order to evaluate which endpoints were body weight dependent, a dietary restriction experiment was performed.

TABLE 2 Effect of Dietary Restriction on Endpoints Treatment groups (g of chow/day) ad libituma

Endpoint

17 g/day

14 g/day

11 g/day

8 g/day

Final body weight (grams)

202.5 (2.54)

191.2* (1.47)

179.9* (1.84)

168.5* (1.34)

160.8* (1.52)

Final body weight (% of control)

100

94.4

88.8

83.2

79.4

Uterine fluid (incidence)

0/14

0/14

0/14

0/14

0/14

Absolute uterine weight (g)

0.096 (0.0027)

0.097 (0.0048)

0.095 (0.0021)

0.102 (0.0029)

0.099 (0.0027)

Estrus conversion (incidence)

0/14

0/14

0/14

0/14

0/14

Prolactin (ng/ml)

2.78 (0.707)

4.54 (0.738)

2.65 (1.086)

3.33 (0.734)

3.28 (0.492)

Growth hormone (ng/ml)

145.31 (67.153)

174.97 (62.299)

132.97 (58.012)

90.58 (20.925)

55.24 (14.414)

Progesterone receptor (fmol/mg protein)

344.3 (49.34)

249.1* (23.00)

192.2* (9.21)

140.4* (10.32)

124.7* (19.62)

Estrogen receptor (fmol/mg protein)

248.1 (26.37)

250.5 (18.40)

265.8 (18.53)

253.3 (14.55)

241.4 (17.42)

Uterine stromal cell proliferation (% of labeled cells)

1.95 (0.429)

1.47 (0.253)

2.22 (0.490)

1.97 (0.285)

1.47 (0.408)

Uterine epithelial cell height (mm)

15.54 (1.065)

15.82 (0.601)

15.66 (1.450)

15.93 (1.484)

16.48 (2.088)

Note. Mean (standard error). * p õ 0.05 compared to ad libitum control. a Ad libitum rats consumed 18.9 g of chow/day.

AID FAAT 2213

/

6k0f$$$201

09-20-96 11:51:43

ftoxa

AP: Fund Tox

186

O’CONNOR ET AL.

Dietary restriction experiment. Daily food consumption for the dietary restriction experiment was assigned in order to produce decreases in body weight that spanned the range of decreases seen with the test compounds. Due to animal welfare concerns, a dietary restriction group was not selected to match the decrements seen in the RES-treated animals. The maximal decrease in final body weight was targeted at 80% of ad libitum controls. Of the nine endpoints, serum growth hormone levels and uterine progesterone receptor number were the only two endpoints judged to be body weight dependent (Table 2). The statistically significant decrease in uterine progesterone receptor content was attributed to increased serum progesterone levels secondary to increased adrenocorticotrophic hormone (ACTH) levels in the diet-restricted animals (data not shown) (DeGroot, 1995). Serum growth hormone levels were also decreased in a treatment-related manner, although statistically significant changes were not observed. Uterine fluid and estrus conversion. Uterine fluid was present only in the uteri from animals treated with the estrogen agonists E2 and E3 (Table 3). E2-induced fluid accumulation was dose-dependent, while E3-induced fluid accumulation was only seen at the highest dose. Estrus conversion was achieved in a dose-dependent manner in animals treated with the estrogen agonists E2 and E3 and the mixed estrogen agonist/antagonist TAM. Neither the full antiestrogen ICI nor the dopamine modulators HAL and RES produced fluid accumulation in the uterus or estrus conversion. Uterine weight, cell proliferation, and height. Absolute uterine weight was statistically increased in rats treated with the estrogen agonists E2 and E3 and the mixed estrogen agonist/antagonist TAM (Figs. 1A and 3). The relative potency of the compounds can be compared by the ED50 values (Table 4) and efficacy by the magnitude of uterine weight increase when compared to the control. The full estrogen agonist E2 (361% of control) produced a greater increase in absolute uterine weight than either the partial estrogen agonist E3 (234% of control) or the mixed estrogen agonist/ antagonist TAM (216% of control). However, the ED50 values were roughly similar for E2 (1.27 mg/kg) and E3 (3.62 mg/kg) but approximately 20-fold less than TAM (48.3 mg/ kg) (Table 4). Neither the full antiestrogen ICI nor the dopamine modulators HAL and RES increased uterine weight. Uterine stromal cell proliferation (Figs. 1B and 3) was statistically increased in rats treated with the estrogen agonists E2 and E3 and the mixed estrogen agonist/antagonist TAM. The partial estrogen agonist E3 (2852% of control) produced a greater increase in stromal cell proliferation than either the full estrogen agonist E2 (1595% of control) or the mixed estrogen agonist/antagonist TAM (1191% of control). However, the ED50 values were roughly similar for E2 (1.56 mg/kg) and E3 (2.61 mg/kg) and were also approximately 20-fold less than TAM (45.1 mg/kg) (Table 4). Neither the

AID FAAT 2213

/

6k0f$$$201

09-20-96 11:51:43

TABLE 3 Incidence of Uterine Fluid and Estrus Conversion

Compound

Dosage (mg/kg/day)

Uterine fluid

Estrus conversion

17b-Estradiol

0 1 2.5 7.5 50

0/14 5/14 12/14 13/14 14/14

(0) a (36) (86) (93) (100)

0/14 2/14 13/14 14/14 14/14

(0) (14) (93) (100) (100)

Estriol

0 1 2.5 7.5 50

0/14 0/14 0/14 0/14 5/14

(0) (0) (0) (0) (36)

0/14 0/14 0/14 9/14 12/14

(0) (0) (0) (64) (86)

Tamoxifen

0 50 250 500 1500

0/14 0/14 0/13 0/14 0/14

(0) (0) (0) b (0) (0)

0/14 0/14 0/14 11/14 13/14

(0) (0) (0) (79) (93)

ICI-182,780

0 50 250 500 1500

0/14 0/14 0/14 0/14 0/13

(0) (0) (0) (0) (0)

0/14 0/14 0/14 0/14 0/13

(0) (0) (0) (0) (0)

Haloperidol

0 10 100 1000

0/14 0/14 0/14 0/14

(0) (0) (0) (0)

0/14 0/14 0/14 0/14

(0) (0) (0) (0)

Reserpine

0 200 1000 2000

0/14 0/14 0/14 0/13

(0) (0) (0) (0)

0/14 0/14 0/14 0/13

(0) (0) (0) (0)

a b

Incidence (percentage). Animal excluded due to the presence of ovarian tissue.

full antiestrogen ICI nor the dopamine modulators HAL and RES increased uterine stromal cell proliferation. Like the endpoints of uterine weight and cell proliferation, uterine epithelial cell height (Figs. 1C and 4) was also statistically increased in rats treated with the estrogen agonists E2 and E3 and the mixed estrogen agonist/antagonist TAM, but unlike the other endpoints, the dopamine antagonists HAL and RES also increased epithelial cell height. The mixed agonist/antagonist TAM (259% of control) produced a greater increase in epithelial cell height than either the full estrogen agonist E2 (218% of control) or the partial estrogen agonist E3 (203% of control). In contrast, the dopamine modulators HAL (115% of control) and RES (136% of control) produced smaller increases in epithelial cell height than the estrogenic test compounds. The range of the ED50 values for epithelial cell height exhibited the greatest differences when compared to the other uterine endpoints: E2 (1.20 mg/ kg), E3 (5.60 mg/kg), TAM (52.4 mg/kg), HAL (253 mg/kg),

ftoxa

AP: Fund Tox

AN IN VIVO BATTERY FOR ENDOCRINE ACTIVITY

187

FIG. 1. Uterine measurements of endocrine modulation. Effects of positive controls on absolute uterine weight (A), uterine stromal cell proliferation (B), uterine epithelial cell height (C), and uterine cytosolic estrogen receptor concentration (D). Uteri were removed at the time of necropsy, weighed, and processed to slides for morphometry and cell proliferation evaluation. Uterine cytosol was prepared for receptor evaluation. Results are expressed as percentage of control. Control values for each endpoint are shown in parentheses. Filled symbols denote statistical significance except for ICI, which is denoted by # (p õ 0.05, Dunnett’s test).

and RES (847 mg/kg) (Table 4). The full antiestrogen ICI did not affect uterine epithelial cell height. Uterine receptor content. Cytosolic ER concentrations were decreased in a dose-dependent manner for all test compounds and were statistically decreased in rats treated with E2, E3, TAM, ICI, and RES (Fig. 1D). For the estrogenic compounds, TAM and ICI (õ1% of control) produced the greatest decreases in estrogen receptor concentration followed by E2 (9% of control) and E3 (60% of control). Suprisingly, the dopamine modulators HAL (57% of control) and RES (36% of control) also decreased the estrogen receptor concentration, although the magnitude of response was similar to the partial estrogen agonist E3. The estrogenic compounds had similar ID50 values (range, 5.10 to 14.1 mg/

AID FAAT 2213

/

6k0f$$$201

09-20-96 11:51:43

kg), while the dopamine modulators HAL (53.9 mg/kg) and RES (435 mg/kg) were much higher (Table 4). Progesterone receptor content was statistically increased in rats treated with E2, E3, and TAM and statistically decreased in rats treated with HAL, RES, and ICI (data not shown). In the dietary restriction study, uterine progesterone receptor content was shown to decrease with decreasing body weight (Table 2). The decreases in progesterone receptor content in rats treated with HAL and RES were attributed to the decreases in body weight based on dietary restriction data (Table 2). However, the decreases seen in the ICItreated rats (36% of control) could not be attributed to body weight changes and were judged to be compound related. The increases in progesterone receptor content in the E2-

ftoxa

AP: Fund Tox

188

O’CONNOR ET AL.

TABLE 4 Summary of ED(ID)50 Values ED(ID)50 (mg/kg/day)

Compound

Absolute uterine weight

Uterine stromal cell proliferation

Uterine epithelial cell height

Serum prolactin

Uterine estrogen receptor a

17b-Estradiol Estriol Tamoxifen ICI-182,780 Haloperidol Reserpine

1.27 3.62 48.3 NA NA NA

1.56 2.61 45.1 NA NA NA

1.20 5.60 52.4 NA 253 847

11.1 NA 129 NA 170 1000

5.10 14.1 9.10 2.91 53.9 435

a

These values are ID50s. NA, not applicable. Value could not be calculated due to absence of response.

and E3-treated rats were not dose-dependent and were maximal at 1.0 mg/kg (304 versus 155% of control for E2 and E3, respectively). In contrast, the increases in progesterone receptor content in the TAM-treated rats were dose-dependent and were maximal at 1500 mg/kg (185% of control). Serum hormone concentrations. Serum prolactin levels were statistically increased in rats treated with the estrogen agonist E2, the mixed estrogen agonist/antagonist TAM, and the dopamine modulators HAL and RES (Fig. 2). The full estrogen agonist E2 (1895% of control) and the D2 receptor antagonist HAL (1601% of control) produced the largest increases, the dopamine depletor RES (502% of control) produced an intermediate response, and the mixed estrogen

agonist/antagonist TAM (290% of control) produced the smallest increase in prolactin levels. Regarding the ED50 values, the most potent was E2 (11.1 mg/kg), intermediate potency values were obtained with TAM (129 mg/kg) and HAL (170 mg/kg), and the least potent was RES (1000 mg/ kg) (Table 4). The partial estrogen agonist E3 and antiestrogen ICI did not increase serum prolactin levels. Serum growth hormone levels were statistically decreased in a dose-dependent manner in rats treated with HAL, RES, and TAM but were unaffected by the other test compounds (data not shown). In the dietary restriction study, serum growth hormone levels were shown to decrease with decreasing body weight (Table 2). The decreases in growth hormone levels in rats treated with RES were attributed to the decreases in body weight based on dietary restriction data (Table 2). However, the decreases seen in the HAL- and TAMtreated rats were judged to be compound related and were maximal at the highest dosage (29% versus 23% of control for HAL and TAM, respectively) (data not shown). DISCUSSION

FIG. 2. Serum prolactin levels. Prolactin levels were measured by radioimmunoassay from serum prepared at necropsy. Results are expressed as percentage of control. Control values are shown in parentheses. Filled symbols denote statistical significance except for ICI, which is denoted by # (p õ 0.05, Dunnett’s test).

AID FAAT 2213

/

6k0f$$$201

09-20-96 11:51:43

An in vivo battery for selected endocrine activities has been developed and partially validated to identify estrogenic/ antiestrogenic compounds and dopaminergic modulators. These endocrine activities are important for addressing environmental estrogen issues as well as identification of mammary tumorigens (reviewed in Colborn et al., 1993; Safe, 1995; Welsch, 1985). The two primary mechanisms for induction of mammary tumors in the CD rat are direct estrogenic stimulation and sustained elevations in serum prolactin through alterations in the dopaminergic pathway (Meites, 1972a,b; Welsch, 1985). To respond to growing concerns regarding the potential for chemicals to alter endocrine homeostasis, reliable and economical screening procedures are needed. We have developed a short term in vivo battery that

ftoxa

AP: Fund Tox

AN IN VIVO BATTERY FOR ENDOCRINE ACTIVITY

189

FIG. 3. Uterine size and cell proliferation. Effects of positive controls on uterine size and uterine stromal cell proliferation shown in cross section. Uterus was removed at the time of necropsy, weighed, and processed to slides for morphometry. Uterine stromal cells (S) are designated by the doubleheaded arrow lying between the uterine luminal epithelium (E) and the mesometrium (M) (see E2). Dose concentrations were E2, 50 mg/kg/day; E3, 50 mg/kg/day; TAM, 250 mg/kg/day; ICI, 1500 mg/kg/day; HAL, 1000 mg/kg/day; RES, 2000 mg/kg/day. Original magnification: 401.

AID FAAT 2213

/

6k0f$$2213

09-20-96 11:51:43

ftoxa

AP: Fund Tox

190

O’CONNOR ET AL.

can be used as a screening tool to identify compounds which possess estrogenic activity or alter the dopamine/prolactin pathway. The overall pattern of the response can not only identify the activity of a compound, but also yield insights on the mode of action. The uterine weight test has been widely used for evaluating compounds for estrogenic activity. One of the limitations of this assay is the minimal responses seen with weak estrogen agonists, which makes definitive conclusions regarding estrogenicity difficult. Other uterotrophic endpoints such as uterine fluid imbibition (Astwood, 1938), estrus conversion (reviewed in Edgren, 1994), and uterine stromal cell proliferation (Kronenberg and Clark, 1985; Martin and Claringbold, 1958) are also valuable tools in assessing compounds that are estrogenic. More recently, additional endpoints have been developed for evaluating estrogenicity, such as uterine luminal epithelial cell height (Branham et al., 1988a, 1993a), uterine progesterone receptor content (Bhattacharyya et al., 1994), and uterine estrogen receptor content (Bhattacharyya et al., 1994). We evaluated these endpoints for their sensitivity (ability to detect weak-acting compounds), specificity (response is limited to one class (activity) of compounds), and dose dependency. In addition, prolactin and growth hormone were evaluated for their ability to identify compounds that modulate the dopamine pathway. While this in vivo battery focuses on uterotrophic responses in the rat, many endocrine modulators may have species- and tissue site-dependent effects (Jordan, 1995; Clute, 1995; Kangas, 1992). Unlike other uterotrophic responses, vaginal cornification (estrus conversion) can be used as a marker for estrogenicity without euthanizing test animals. Allen and Doisy characterized changes in lavageable cell types over the course of the estrous cycle and found that estrus is induced when circulating estrogen levels are at their peak (Allen and Doisy, 1923). Induction of vaginal cornification in ovariectomized rats is a very specific response of estrogen treatment. For instance, treatment with the estrogenic compounds E2, E3, and TAM induced vaginal cornification, while the full antiestrogen ICI and the dopamine modulators HAL and RES did not (Table 3). Regarding sensitivity, all three estrogenic compounds induced cornification, while characterizing the dose-response is more difficult. One of the earliest measurable uterotrophic responses to estrogenic stimulation is the accumulation of water in the uterus, commonly referred to as uterine fluid imbibition (Astwood, 1938; Katzenellenbogen and Gorski, 1975). Astwood and co-workers characterized the biphasic response for uterine weight growth by estrogenic compounds. Uterine fluid imbibition peaks approximately 6 hr after estrogen treatment and occurs from increased vascular permeability mediated by eosinophils (Astwood, 1938; Lee et al., 1989). This is followed by a decrease in uterine weight due to attenuation of fluid accumulation.

AID FAAT 2213

/

6k0f$$$201

09-20-96 11:51:43

Uterine weight peaks again, approximately 30 hr after estrogen exposure, due to estrogen-mediated stimulation of transcription and cell division (Kaye et al., 1972; Stormshak et al., 1976). Based on our data, fluid imbibition is a very specific endpoint for estrogenicity, but it is not sensitive. For instance, the dose-response for fluid imbibition mirrored estrus conversion in E2-treated rats, but was only detected in the high dosage group for E3treated rats and not detected with the mixed agonists TAM (Table 3). Estrogens increase uterine weight through increased cell division (Kaye et al., 1972; Stormshak et al., 1976). The magnitude of the response is proportional to the dose of estrogen (reviewed in Clark and Mani, 1994). However, it is often difficult to evaluate the potencies of different estrogens due to differences in absorption and estrogenic potential (reviewed in Clark and Mani, 1994). Full estrogen agonists such as E2 are capable of increasing uterine weight after a single dose while partial estrogen agonists such as E3 do not (reviewed in Clark and Mani, 1994). Multiple dosing can be used to compensate for differences in pharmacokinetics and potency. Under our experimental conditions, one is able to distinguish between full and weak estrogen agonists by the magnitude of the uterine weight increase. E2 produced the greatest increase in uterine weight (361% of control) followed by the partial estrogen agonist E3 (234% of control) and the mixed estrogen agonist/antagonist TAM (216% of control) (Figs. 1A and 3). While uterine weight is a sensitive marker for detecting weak estrogen agonists, it can be a nonspecific response. Compounds such as progestagens, progesterone, and testosterone at high levels are capable of increasing uterine weight (Jones and Edgren, 1973). Within 12 – 24 hr of E2 treatment, there is a measurable increase in cell proliferation in the uterus (Kaye et al., 1972; Stormshak et al., 1976). Of the endpoints we evaluated, uterine stromal cell proliferation was the most sensitive response to estrogen stimulation based on the magnitude of the dose – response curves (Fig. 1B). Cell proliferation was increased in the E2-, E3-, and TAM-treated rats but not with the other compounds tested (Fig. 1B). The lack of cell proliferation seen with the antiestrogen ICI and the dopamine modulators HAL and RES confirms the specificity of the cell proliferation response. Interestingly, the weak estrogen agonist E3 produced a greater increase in cell proliferation (2852% of control) when compared to control than the full estrogen agonist E2 (1595% of control). This discrepancy is believed to be artifactual due to the unusually low cell proliferation seen in the control group in the E3 experiment (Fig. 1B legend). In support of this conclusion, the maximum increase in the labeling index is higher in the E2-treated rats (29.8%) than in the E3-treated rats (20.8%) (Fig. 1B). In the absence of estrogen, the uterine epithelium consists of two to three layers of cuboidal cells. Within 12– 24 hr

ftoxa

AP: Fund Tox

AN IN VIVO BATTERY FOR ENDOCRINE ACTIVITY

of E2 treatment the cells change from cuboidal-shaped to cylindrical-shaped vacuolated cells (Martin and Claringbold, 1958). Within 24 hr the thickness of the epithelium can increase from two to three layers of cells to six to eight layers of cells. These morphologic changes can be quantitated by measuring uterine epithelial cell height (Branham et al., 1988a,b, 1993a,b). The mixed estrogen agonist/antagonist TAM produced the greatest increase in uterine epithelial cell height (259% of control) followed by the full estrogen agonist E2 (218% of control) and the partial estrogen agonist E3 (203% of control) when compared to their concurrent controls (Figs. 1C and 4). The greater response observed with TAM compared to E2 may be due to the higher doses administered with TAM and/or the higher relative binding affinity for the estrogen receptor for TAM (Obourn et al., 1994). Interestingly, the dopamine modulators HAL (115% of control) and RES (136% of control) also caused an increase in uterine epithelial cell height. It is possible that the HAL- and RES-induced increase in uterine epithelial cell height is due to the increase in serum prolactin levels (Cohen et al., 1993). Hence, uterine epithelial cell height is a sensitive marker, but is not specific to estrogenic compounds. The actions of steroid hormones are mediated through specific steroid receptors such as the progesterone receptor (PR), estrogen receptor (ER), and the androgen receptor (AR) (reviewed in Beato et al., 1995). In the uterus, estrogen treatment decreases cytosolic ER levels (reviewed in Manni et al., 1981; Shupnik et al., 1989) and increases cytosolic PR levels (reviewed in Manni et al., 1981). Kraus and coworkers found that within 48 hr of E2 treatment, uterine PR levels increased sixfold and uterine ER levels dropped to 15% of control (Kraus and Katzenellenbogen, 1993). Our experiments showed a similar pattern for the estrogen receptor agonists E2 and E3 (PR levels of 304 and 155% of controls, respectively; ER levels of 9 and 60% of controls, respectively) and the mixed estrogen receptor agonist/antagonist TAM (PR level of 185% of control; ER level of õ1% of control) (Fig. 1D; PR data not shown). In the E2-, E3-, and TAM-treated rats, a dose-dependent decrease in ER levels was seen, however, the increases in PR were not dosedependent and peaked at much lower concentrations of test compound. Because PR levels are body weight dependent (Table 2), the usefulness of PR as a marker is limited to compounds which increase PR or do not induce body weight effects. For instance, the HAL- and RES-induced decreases in PR level are attributed solely to decrements in body weight (Table 2). The ability of HAL and RES to decrease ER levels is unknown. However, the full estrogen receptor

191

antagonist ICI decreased in a dose-dependent manner both uterine PR levels (36% of control) and ER levels (õ1% of control), and these effects were compound related since there were no body weight effects. Because PR levels are maintained by estrogen stimulus, the ability of ICI to decrease PR levels is likely due to its ability to act as a full estrogen receptor antagonist (Wakeling et al., 1991). Hence, ER and PR levels are useful for markers of estrogenicity (i.e., sensitive), but can also help identify full estrogen receptor antagonists. Clearly, these two endpoints require cautious interpretation due to the lack of specificity and the body weight dependency of PR levels. The two primary hormones influencing mammary gland tumorigenesis in the rat are estrogens and prolactin (Meites, 1972a,b; Welsch, 1985). Compounds which decrease dopamine levels in the brain (such as RES), or dopamine receptor antagonists (such as HAL), increase serum prolactin and decrease growth hormone (Jansson et al., 1985; Macleod and Lehmeyer, 1974). We evaluated serum prolactin and growth hormone for their ability to distinguish between estrogen agonists and dopamine modulators. In our experiments, the full estrogen receptor agonist E2 and the mixed estrogen agonist/antagonist TAM produced dose-dependent increases in serum prolactin with maximal responses of 1895 and 290% of control, respectively (Fig. 2). However, the partial estrogen agonist E3 failed to increase serum prolactin levels. The dopamine modulators HAL (1601% of control) and RES (502% of control) also produced dose-dependent increases in prolactin. While prolactin is a useful endpoint for identifying dopamine modulators, its lack of specificity and sensitivity require it to be used in conjuction with uterotrophic endpoints. Growth hormone levels were evaluated as an additional marker for DA modulation (data not shown). This endpoint was judged not to be a reliable marker because of its variability and because of confounding effects of body weight loss (Table 2). Even if a concurrent pair-fed control group was used, the variability of this endpoint severely limits its usefulness. Other endpoints such as brain dopamine levels could be used in place of growth hormone levels and perhaps provide a more specific and sensitive endpoint for examining dopamine modulators. In conclusion, this in vivo battery appears useful for identifying different classes of endocrine modulators (Table 5). For example, the estrogen receptor agonists E2 and E3 display a unique profile based on changes in the uterotrophic endpoints (estrus conversion, uterine fluid imbibition, increases in uterine weight, and uterine endometrial cell prolif-

FIG. 4. Uterine epithelial cell height. Effects of positive controls on uterine epithelial cell height. Uterus was removed at the time of necropsy, weighed, and processed to slides for morphometry. The thickness of the uterine epithelium is identified by the double-headed arrow. Dose concentrations were E2, 50 mg/kg/day; E3, 50 mg/kg/day; TAM, 250 mg/kg/day; ICI, 1500 mg/kg/day; HAL, 1000 mg/kg/day; RES, 2000 mg/kg/day. Original magnification: 4001.

AID FAAT 2213

/

6k0f$$$201

09-20-96 11:51:43

ftoxa

AP: Fund Tox

192

AID FAAT 2213

O’CONNOR ET AL.

/

6k0f$$2213

09-20-96 11:51:43

ftoxa

AP: Fund Tox

193

AN IN VIVO BATTERY FOR ENDOCRINE ACTIVITY

FIG. 4—Continued

eration) where full and partial agonists can be distinguished by the magnitude of these responses. The estrogen receptor antagonist ICI and the dopamine modulators HAL and RES all lack these uterotrophic responses. Dopamine modulators can be distinguished from estrogen receptor agonists by the profile of increased prolactin levels with no uterotrophic changes. Estrogen receptor antagonists can be distinguished from agonists by comparing their effects on ER, PR, and

uterotrophic responses. For instance, the full estrogen receptor antagonist ICI decreased ER (to almost zero) and PR levels, but has no uterotrophic effects, while TAM decreases ER (to almost zero) and increases PR with uterotrophic effects. The most useful endpoints for distinguishing estrogen agonists and dopamine modulators were uterine fluid imbibition, uterine weight, uterine stromal cell proliferation, and serum prolactin levels. In order to distinguish an estrogen

TABLE 5 Effects of Different Compounds on Selected Endocrine Endpoints Compound classification

Endpoint

Full estrogen agonist (E2)

Partial estrogen agonist (E3)

Mixed estrogen agonist (TAM)

Estrogen antagonist (ICI)

Dopamine modulator (HAL/RES)

Estrus conversion Uterine fluid imbibition Uterine weight Endometrial cell proliferation Uterine epithelial cell height Progesterone receptor content Estrogen receptor content Prolactin Growth hormone

/// /// /// /// /// /// — /// NC

// / // /// /// // 0 NC NC

// NC // /// /// // — / 0

NC NC NC NC NC — — NC NC

NC NC NC NC / ? 0 /// ?

Note. /, Compound increased endpoint response; 0, compound decreased endpoint response; NC, no change; ?, endpoint could not be evaluated due to confounding body weight effects.

AID FAAT 2213

/

6k0f$$$201

09-20-96 11:51:43

ftoxa

AP: Fund Tox

194

O’CONNOR ET AL.

agonist from an antagonist, other endpoints, such as receptor levels, are necessary. In addition, endpoints such as brain dopamine levels may enhance the ability of the screen to identify compounds that modulate serum prolactin levels through changes in dopamine levels. In summary, we have combined multiple endpoints into a single 5-day in vivo test to identify estrogenic compounds and dopaminergic modulators. This battery can be used as a screening tool to identify potential endocrine modulators. It can also be used as a tool to identify mode of action following adverse findings in developmental (e.g., changes in developmental landmarks), reproduction (e.g., decreases in fertility, sperm counts), and oncogenicity (e.g., increased incidence of mammary gland tumors) studies. In contrast to in vitro screening systems that focus solely on estrogenicity, the advantages of a system like this are numerous. The redundancy of the endpoints enhances the ability to detect weakly estrogenic compounds as well as reducing the probability of false negative/positive responses. Recently, we have used this in vivo approach to identify a commercially available placebo carrier as weakly estrogenic (manuscript submitted for publication). Finally, other endpoints may be added to identify other classes of endocrine modulators. Once validated, compounds can be screened quickly and at a relatively low cost. ACKNOWLEDGMENTS Michele Applegate and Susan Nicastro are gratefully acknowledged for their technical support.

REFERENCES Adami, H.-O., Lipworth, L., Titus-Ernstoff, L., Hsieh, C.-C., Hanberg, A., Ahlborg, U., Baron, J., and Trichopoulos, D. (1995). Organochlorine compounds and estrogen-related cancers in women. Cancer Causes Control 6, 551– 566. Ahlborg, U. G., Lipworth, L., Titus-Ernstoff, L., Hsieh, C.-C., Hanberg, A., Baron, J., Trichopoulos, D., and Adami, H.-O. (1995). Organochlorine compounds in relation to breast cancer, endometrial cancer, and endometriosis: An assessment of the biological and epidemiological evidence. Crit. Rev. Toxicol. 25, 463– 531. Alison, R. H., Capen, C. C., and Prentice, D. E. (1994). Neoplastic lesions of questionable significance to humans. Toxicol. Pathol. 22, 179– 186. Allen, E., and Doisy, E. A. (1923). Preliminary report on its localization, extraction and partial purification, and action in test animals. J. Am. Med. Assoc. 8, 819– 821. Astwood, E. B. (1938). A six-hour assay for the quantitative determination of estrogen. Endocrinology 23, 25– 31. Beato, M., Herelich, P., and Schultz, G. (1995). Steroid receptors: Many actors in search of a plot. Cell 83, 851– 857. Bhattacharyya, N., Ramsammy, R., Eatman, E., Hollis, V. W., and Anderson, W. A. (1994). Protooncogene, growth factor, growth factor receptor, and estrogen and progesterone receptor gene expression in the immature rat uterus after treatment with estrogen and tamoxifen. J. Submicrosc. Cytol. Pathol. 26, 147– 162. Branham, W. S., Leamons, M. L., and Sheehan, D. M. (1988a). Estrogen-

AID FAAT 2213

/

6k0f$$$201

09-20-96 11:51:43

and antiestrogen-induced ornithine decarboxylase activity and uterine growth in the rat. J. Steroid Biochem. 29, 153 – 159. Branham, W. S., Lyn-Cook, B. D., Andrews, A., and Sheehan, D. M. (1993a). Growth of separated and recombined neonatal rat uterine luminal epithelium and stroma on extracellular matrix: Effects of in vivo tamoxifen exposure. In Vitro Cell Dev. Biol. 29A, 408 –414. Branham, W. S., Zehr, D. R., Chen, J. J., and Sheehan, D. M. (1988b). Postnatal uterine development in the rat: Estrogen and antiestrogen effects on luminal epithelium. Teratology 38, 29 –36. Branham, W. S., Zehr, D. R., and Sheehan, D. M. (1993b). Differential sensitivity of rat uterine growth and epithelium hypertrophy to estrogens and antiestrogens. Proc. Soc. Exp. Biol. Med. 203, 297 – 303. Clark, J. H., and Mani, S. K. (1994). Actions of ovarian steroid hormones. In The Physiology of Reproduction (E. Knobil and J. D. Neill, Eds.), pp. 1011– 1059. Raven Press, New York. Clark, J. H., and Markaverich, B. M. (1983). The agonistic and antagonistic effects of short acting estrogens: A review. Pharmacol. Ther. 21, 429 – 453. Clute, E. (1995). Tamoxifen—Breast cancer preventive or human carcinogen? Toxicol. Ind. Health 11, 503– 510. Cohen, H., Pageaux, J.-F., Melinand, C., Fayard, J.-M., and Laugier, C. (1993). Normal rat uterine stromal cells in continuous culture: Characterization and progestin regulation of growth. Eur. J. Cell. Biol. 61, 116– 125. Colborn, T., Saal, F. S. V., and Soto, A. M. (1993). Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ. Health Perspect. 101, 378– 384. Davis, B. (Ed.) (1993). Female Reproductive Toxicology. Academic Press, San Diego. DeGroot, L. J. (Ed.) (1995). Endocrinology, pp. 341– 367. Saunders, Philadelphia. Dickson, R. B., Johnson, M. D., Bano, M., Shi, E., Kurebayashi, J., Ziff, B., Martinez-Lacaci, I., Amundadottir, L. T., and Lippman, M. E. (1992). Growth factors in breast cancer: Mitogenesis to transformation. J. Steroid Biochem. Mol. Biol. 43, 69– 78. Durbin, P. W., Williams, M. H., Jeung, N., and Arnold, J. S. (1966). Development of spontaneous mammary tumors over the life-span of the female Charles River (Sprague– Dawley) rat: The influence of ovariectomy, thyroidectomy, and adrenalectomy – avariectomy. Cancer Res. 26 (Part 1), 400 – 411. Edgren, R. A. (1994). Issues in animal pharmacology. In Pharmacology of the Contraceptive Steroids. (J. W. Goldzieher, Ed.), pp. 81– 97. Raven Press, New York. FDA Advisory Committee Report on Antipsychotic Drugs (August 12, 1977). Goodman Gilman, A. (1990). Goodman and Gilman’s The Pharmacological Basis of Therapeutics. Pergamon, New York. Greaves, P. (1990). Factors affecting development and growth of mammary neoplasms. In Histopathology of Preclinical Toxicity Studies: Interpretation and Relevance in Drug Safety Evaluation. pp. 64– 69. Elsevier, Amsterdam. Henderson, B. E., Ross, R. K., and Pike, M. C. (1993). Hormonal chemoprevention of cancer in women. Science 259, 633– 638. Ignar-Trowbridge, D. M., Nelson, K. R., Bidwell, M. C., Curtis, S. W., Washburn, T. F., McLachlan, J. A., and Korach, K. (1992). Coupling of dual signalling pathways: Epidermal growth factor action involves the estrogen receptor. Proc. Natl. Acad. Sci. USA 89, 4658– 4662. Jansson, J.-O., Staffan, E., and Isaksson, O. (1985). Sexual dimorphism in the control of growth hormone secretion. Endocrine Rev. 6, 128– 150.

ftoxa

AP: Fund Tox

AN IN VIVO BATTERY FOR ENDOCRINE ACTIVITY Jones, R. C., and Edgren, R. A. (1973). The effects of various steroids on the vaginal histology in the rat. Fertil. Steril. 24, 284– 291. Jordan, C. V. (1995). An overview of considerations for the testing of tamoxifen as a preventive for breast cancer. Ann. NY Acad. Sci. 768, 141 – 147. Kangas, Lauri. (1992). Agonistic and antagonistic effects of antiestrogens in different organs. Acta. Oncol. 31, 143 –146. Katzenellenbogen, B. S., and Gorski, J. (1975). The Biochemical Actions of Hormones. Academic Press, New York. Kaye, A. M., Sheratzky, D., and Lindner, H. R. (1972). Kinetics of DNA synthesis in immature rat uterus: Age dependence and estradiol stimulation. Biochim. Biophys. Acta 261, 475– 486. Key, T., and Reeves, G. (1994). Organochlorines in the environment and breast cancer. Br. Med. J. 308, 1520 –1521. Kraus, W. L., and Katzenellenbogen, B. S. (1993). Regulation of progesterone receptor gene expression and growth in the rat uterus: Modulation of estrogen actions by progesterone and sex steroid hormone antagonists. Endocrinology 132, 2371– 2379. Krieger, N., Wolff, M. S., Hiatt, R. A., Rivera, M., Vogelman, J., and Orentriech, N. (1994). Breast cancer and serum organochlorines: A prospective study among white, black, and Asian women. J. Natl. Cancer Inst. 86, 589– 599. Kronenberg, M. S., and Clark, J. H. (1985). Changes in keratin expression during the estrogen-mediated differentiation of rat vaginal epithelium. Endocrinology 117, 1480– 1489. Lee, Y. H., Howe, R. S., Sha, S.-J., Teuscher, C., Sheehan, D. M., and Lyttle, R. (1989). Estrogen regulation of an eosinophil chemotactic factor in the immature rat uterus. Endocrinology 125, 3022– 3028. Macleod, R. M., and Lehmeyer, J. E. (1974). Studies on the mechanism of the dopamine-mediated inhibition of prolactin secretion. Endocrinology 94, 1077– 1085. Manni, A., Baker, R., Arafah, B. M., and Pearson, O. H. (1981). Uterine oestrogen and progesterone receptors in the ovariectomized rat. J. Endocrinol. 91, 281– 287. Martin, L., and Claringbold, P. J. (1958). A highly sensitive assay for oestrogens. Nature 181, 620– 621. Meites, J. (1972a). Regulation of prolactin and estrogen to mammary cancer in the rat in response to estrogen. Arch. Pathol. 33, 334 –356. Meites, J. (1972b). Relation of prolactin and estrogen to mammary tumorigenesis in the rat. J. Natl. Cancer Inst. 48, 1217 –1224. Neumann, F. (1991). Early indicators for carcinogenesis in hormone-sensitive organs. Mutat. Res. 248, 341– 356. Newman, C. B., Cosby, H., Friesen, H. G., Feldman, M., Cooper, P., De Crescito, V., Pilon, M., and Kleinberg, D. L. (1987). Evidence for a nonprolactin, non-growth-hormone mammary mitogen in the human pituitary gland. Proc. Natl. Acad. Sci. USA 84, 8110 – 8114.

AID FAAT 2213

/

6k0f$$$201

09-20-96 11:51:43

195

Obourn, J. D., Koszewski, N. J., Notides, A. C. (1994). Hormone- and DNA-binding mechanisms of the recombinant human estrogen receptor. Biochemistry 32, 6229– 6236. Safe, S. H. (1995). Environmental and dietary estrogens and human health: Is there a problem? Environ. Health Perspect. 103, 346– 351. Scatchard, G. (1949). The attractions of proteins for small molecules and ions. Ann. NY Acad. Sci. 51, 660– 672. Schardein, J. L. (1993). Hormones and Hormonal Antagonists, In Chemically Induced Birth Defects. pp. 271– 339. Dekker, New York. Schyve, P. M., Smithline, F., and Meltzer, H. Y. (1978). Neurolepticinduced prolactin level elevation and breast cancer. Arch. Gen. Psychiatry 35, 1291 –1301. Shupnik, M. A., Gordon, M. S., and Chin, W. (1989). Tissue-specific regulation of rat estrogen receptor mRNAs. Mol. Endocrinol. 3, 660– 665. Soto, A. M., Lin, T. M., Justicia, H., Silvia, R. M., and Sonnenschein, C. (1992) An ‘in culture’ bioassay to assess the estrogenicity of xenobiotics (E-screen), In Advances in Modern Environmental Toxicology. Chemically-induced Alterations in Sexual and Functional Development: The Wildlife/Human Connection. Vol. 21. pp. 295– 309. Princeton Sci. Pub., Princeton. Stormshak, F., Leake, R., Wertz, N., and Gorski, J. (1976). Stimulatory and inhibitory effects of estrogen on uterine DNA synthesis. Endocrinology 99, 1501 – 1511. Villalobos, M., Olea, N., Brotons, J. A., Loea-Serrano, M. F., Almodovar, J. M. R., and Pedraza, V. (1995). The E-screen assay: A comparison of different MCF-7 cell stocks. Environ. Health Perspect. 103, 844– 850. Wakeling, A. E., Dukes, M., and Bowler J. (1991). A potent specific pure antiestrogen with clinical potential. Cancer Res. 51, 3867– 3873. Welch, R. M., Levin, W., and Conney, A. H. (1969). Estrogenic action of DDT and its analogs. Toxicol. Appl. Pharmacol. 14, 358 – 367. Welsch, C. W. (1983). Hormones and murine mammary tumorigenesis: An historical perspective. In Hormonal Regulation of Experimental Mammary Tumors (B. S. Leung, Ed.). pp. 1 – 29. Eden Press, Montreal. Welsch, C. W. (1985). Host factors affecting the growth of carcinogeninduced rat mammary carcinomas: A review and tribute to Charles Brenton Huggins. Cancer Res. 45, 3415 –3443. Welsch, C. W. (1986) Interrelationship between dietary fat and endocrine processes in mammary gland tumorigenesis. In Dietary Fat and Cancer (A. Rogers, D. Birt, E. Mettlin, and C. Ip, Eds.). pp. 623 –654. A. R. Liss, New York. Welsch, C. W., Adams, C., Lambrecht, L. K., and Brooks, C. K. (1977). 17b-estradiol and envoid mammary tumorigenesis in C3H/HeJ female mice: Counteraction by concurrent 2-bromo-a-ergocriptine. Br. J. Can. 35, 322– 328. Wolff, M. S., Toniolo, P. G., Leel, E. W., Rivera, M., and Dubin, N. (1993). Blood levels of organochlorine residues and risk of breast cancer. J. Natl. Cancer Inst. 85, 648– 652.

ftoxa

AP: Fund Tox