Effects of Dietary 17β-Estradiol Exposure on Serum Hormone Concentrations and Testicular Parameters in Male Crl:CD BR Rats

TOXICOLOGICAL SCIENCES ARTICLE NO.

44, 155–168 (1998)

TX982470

Effects of Dietary 17b-Estradiol Exposure on Serum Hormone Concentrations and Testicular Parameters in Male Crl:CD BR Rats1 Jon C. Cook,2 Larry Johnson,* John C. O’Connor, Lisa B. Biegel, Cindy H. Krams, Steven R. Frame, and Mark E. Hurtt DuPont Haskell Laboratory for Toxicology and Industrial Medicine, P.O. Box 50, Newark, Delaware 19714; and *Department of Veterinary Anatomy and Public Health, College of Veterinary Medicine, Texas A & M University, College Station, Texas 77843-4458 Received November 10, 1997; accepted April 2, 1998

Effects of Dietary 17b-Estradiol Exposure on Serum Hormone Concentrations and Testicular Parameters in Male Crl:CD BR Rats. Cook, J. C., Johnson, L., O’Connor, J. C., Biegel, L. B., Krams, C. H., Frame, S. R., and Hurtt, M. E. (1998). Toxicol. Sci. 44, 155–168. A 90-day/one-generation reproduction study was conducted in male and female Crl:CD BR rats using dietary levels of 0, 0.05, 2.5, 10, and 50 ppm 17b-estradiol. The goals of this study were to set dose levels and evaluate several mechanistic endpoints for inclusion in multigeneration reproduction and combined chronic toxicity/oncogenicity studies with 17b-estradiol. In this report we discuss the effects of dietary 17b-estradiol exposure on serum hormonal levels and sperm parameters from P1 and F1 male rats. Sperm parameters were also evaluated in recovery P1 and F1 male rats that were fed control diets for 105 and 103 days, respectively, following 97 and 86 –94 days of estradiol exposure, respectively. Measurement of Sertoli cell number from F1 male rats was performed to test the hypothesis that in utero exposure to estrogens will decrease Sertoli cell number and sperm production. Other findings from this 90-day/ one-generation reproduction study are summarized elsewhere. 17b-Estradiol produced a dose-dependent decrease in body weight in P1 male rats at >2.5 ppm and in the F1 male rats at 2.5 ppm. This decrease in body weight was due to a combination of reduced food consumption and food efficiency. In the recovery P1 males, body weight increased in the affected groups, albiet not to control levels, due to food consumption returning to control levels accompanied by an increase in food efficiency. However, in F1 males there was no corresponding rebound in body weight. In the P1 rats, exposure to 17b-estradiol decreased testis and epididymis weights in the 10 and 50 ppm groups, while no effects were seen in the P1 2.5 ppm group. In contrast, epididymis weights in the F1 and F1 recovery 2.5 ppm groups were statistically decreased; however, there were no histopathological effects observed. The decreases in testis weights in the P1 generation correlated with histopathologic evidence of interstitial cell atrophy and seminiferous tubule degeneration and 1 We gratefully acknowledge partial funding of this project by the Chlorine Chemistry Council, 1300 Wilson Boulevard, Arlington, VA 22209. 2 To whom reprint requests should be addressed. Fax: (302) 451-4827.

reduced sperm production. Correlative changes in the epididymides of P1 rats were characterized by oligospermia or aspermia, the presence of germ cell debris in the lumen of tubules, and atrophy of epididymal tubules. 17b-Estradiol decreased testicular spermatid numbers, epididymal sperm numbers, and sperm motility in the P1 males in the 10 and 50 ppm groups, but not in the 2.5 ppm group. Following a 105-day recovery period in the P1 males, all sperm parameters and reproductive organ weights returned to control values except for the epididymal sperm count. Overall, the decline in testicular spermatid and epididymal sperm numbers in the P1 rats correlated with the reduced organ weights and the observed histopathological changes and appeared primarily related to the decrease in serum testosterone levels. In the F1 rats, no significant decreases were noted in the testicular spermatid number but a slight decrease in epididymal sperm number was seen in the 2.5 ppm group, which showed no evidence of recovery. Using morphometric analysis, no change was seen in the number of Sertoli cell nuclei per testis in F1 males. The pattern of hormonal responses seen in this study was characteristic of an estrogen receptor agonist such as 17b-estradiol: increased serum prolactin and decreased testosterone, luteinizing hormone, and follicle stimulating hormone levels. The data demonstrate that in utero and postnatal dietary administration of 17bestradiol at levels which increased serum estradiol levels to approximately 400% of control and decreased testosterone levels to 33% of control did not reduce the number of Sertoli cell nuclei per testis. © 1998 Society of Toxicology.

Based largely on ecologic data, it has been hypothesized that endocrine active compounds (EACs) are responsible for worldwide increases in cancer incidence (breast and reproductive tract) and developmental abnormalities (male urogenital defects), and decreasing sperm counts in humans (Birnbaum, 1994; Carlse¨n et al., 1992; Colborn and Clement, 1992; Colborn et al., 1993; Davis et al., 1993; Davis and Bradlow, 1995; Kelce et al., 1994; Sharpe and Skakkebaek, 1993; Wolff et al., 1993). Because of the implications to human health and the environment, this hypothesis has

155

1096-6080/98 $25.00 Copyright © 1998 by the Society of Toxicology. All rights of reproduction in any form reserved.

156

COOK ET AL.

come under intense international scrutiny within the scientific and regulatory communities (NAS, 1995; Harrison et al., 1995; Toppari et al., 1995; Kavlock et al., 1996; Ankley et al., 1997). Decreased reproductive success and developmental defects (masculinization/feminization) in wildlife also have been reported and appear to be primarily associated with areas of industrial pollution (Colborn and Clement, 1992). However, there is considerable controversy over whether adverse human health effects can be attributed to exposure to EACs in the environment (Stone, 1994; Safe, 1995; Houghton and Ritter, 1995; Crisp et al., 1997). Much of the concern around EACs has focused on estrogenicity (Harrison et al., 1995; Toppari et al., 1995). This focus is in part due to the adverse responses seen in humans following in utero exposure to diethylstilbestrol (DES), a drug prescribed to pregnant women in the 1950s and 1960s in the mistaken belief that it would prevent miscarriage. In utero DES exposure was subsequently linked to the production of vaginal clear cell adenocarcinoma in daughters (Herbst et al., 1971) and reproductive abnormalities in sons (Gill et al., 1979) of exposed mothers: increased incidence of genital malformations (cryptorchidism, epididymal cysts, hypoplastic testes, microphallus) and reduced sperm counts (79% of control) when compared to the placebo control group. This same cohort was studied again for evidence of altered fertility when these men had reached the ages of 38 to 41 years. There was no evidence that the DES-exposed males had altered fertility or sexual function (Wilcox et al., 1995). However, this assessment did not include measurement of sperm counts. There are no reports that this cohort had an increased incidence of testicular cancer. In 1992, a meta-analysis of 61 studies published between 1938 and 1991 was performed and the authors concluded that human sperm counts had declined by 42% and mean ejaculate volume had decreased by 20% based upon linear regression analysis (Carlse¨n et al., 1992). Other investigators have also reported declining sperm counts over time (Ginsburg et al., 1994; Auger et al., 1995; Irvine et al., 1996). These data supported the hypothesis that a widespread decline in sperm counts is occurring in humans. However, when the data from the meta-analysis were evaluated using additional models, the best statistical fit was not a linear regression but rather a stair-step model (Olsen et al., 1995). The stair-step model suggests that the sperm counts decreased prior to 1970 and that from 1970 through 1990 sperm counts were either constant or slightly increased. In addition, other studies of sperm number, spanning the time period from 1958 to 1996, have reported no decline in human sperm production (MacLeod and Wang, 1979; Suominen and Vierula, 1993; Fisch et al., 1996; Paulsen et al., 1996). Interestingly, Fisch and co-workers (1996) reported regional variations (New York, Minnesota, and California) in sperm count within the United States, where New York had the highest numbers. It is noteworthy that the earlier studies which were the reference point for the apparent ‘‘de-

cline’’ reported in the Carlse¨n meta-analysis were primarily from New York. Hence, the decline in sperm counts reported by Carlse¨n and co-workers (1992) may be a reflection of regional variations in sperm counts (Fisch and Goluboff, 1996). Clearly, while there may be regional variations in sperm counts, the available data do not support the hypothesis that every region of world is experiencing a decline in sperm counts (Crisp et al., 1997). Based largely on the DES data, Sharpe and Skakkebaek (1993) provided a plausible mechanistic explanation whereby in utero exposure to environmental estrogens could result in decreasing sperm counts and increased incidence of testicular cancer and urogenital defects. They proposed that in utero exposure to estrogenic compounds could decrease sperm number via estrogen-induced reduction of FSH levels which in turn would reduce Sertoli cell number. Because daily sperm production is conventionally believed to be directly proportional to Sertoli cell number in humans, reduced Sertoli cell number would result in decreased daily sperm production (Johnson et al., 1984b, 1997). Consistent with this hypothesis, in utero exposure to xenoestrogens has been reported to reduce testicular size and sperm production (Sharpe et al., 1995). Regarding the increased incidence of testicular cancer and urogenital defects in men, Sharpe and Skakkebaek also proposed that in utero exposure to estrogenic compounds could reduce FSH levels which in turn would reduce Sertoli cell production of anti-Mu¨llerian hormone (AMH). Decreased AMH levels would cause Mu¨llerian duct retention and impair transabdominal descent of the testes resulting in testicular nondescent. Testicular nondescent is the major risk factor for testicular cancer in humans (Buetow, 1995; Forman et al., 1994; Giwercman et al., 1988). Reduced AMH levels also increase germ cell number which increases the probability that these germ cells will give rise to carcinoma in situ (CIS), the major testicular cancer type seen in men. The recently passed Food Quality Protection Act of 1996 requires the U.S. EPA to implement screening strategies for EACs within the next 2 years. A key issue in interpreting results from screening tests will be distinguishing between physiologic and adverse responses. However, addressing these issues is complicated by the absence of traditional dietary rodent bioassays with model estrogenic compounds such as 17b-estradiol. Thus, a 90-day/one-generation reproduction study was designed to set dose levels and evaluate mechanistic endpoints for inclusion in multigeneration reproduction and combined chronic toxicity/oncogenicity studies. From pilot studies and available literature, dietary levels of 0, 0.05, 2.5, 10, and 50 ppm 17b-estradiol were chosen for the following study. The traditional 90-day/one-generation endpoints as well as estrous cyclicity and female hormonal data are reported elsewhere (Biegel et al., 1998a,b). The present paper reports the effects of dietary 17b-estradiol exposure on serum hormone levels and sperm parameters in P1 and F1 male rats and Sertoli cell number in F1 male rats. In addition, the effects of

DOES IN UTERO EXPOSURE REDUCE SERTOLI CELL NUMBER?

a 101- to 105-day recovery period on sperm parameters were evaluated in P1 and F1 male rats. MATERIALS AND METHODS Test materials. 17b-Estradiol (98 –100% pure), phosphate-buffered saline (PBS), toluidine blue, and bovine serum albumin (BSA; Fraction V) were purchased from Sigma Chemical Company (St. Louis, MO). Eosin Y was purchased from J. T. Baker (Phillipsburg, NJ). Epon, 2% glutaraldehyde, osmium, and sodium cacodylate buffer were purchased from Electron Microscopy Sciences (Fort Washington, PA). All other materials were obtained from the following manufacturers: Certified Rodent Chow 5002 meal, PMI Feeds, Inc. (St. Louis, MO); luteinizing hormone (LH), prolactin (PRL), and follicle stimulating hormone (FSH) radioimmunoassay (RIA) kits, Amersham Corp. (Arlington Heights, IL); testosterone (T) and estradiol (E2) RIA kits, Diagnostics Products Corp. (Los Angeles, CA). Diet preparation and analysis. A complete description of the diet preparation, concentration/stability verification, and analytical methods can be found in the accompanying paper (Biegel et al., 1998a). Briefly, dietary concentrations of 17b-estradiol were measured by high-performance liquid chromatography (HPLC) with UV detection for the 10 and 50 ppm diets and with fluorescence detection for the 0.05 and 2.5 ppm diets. The initial three diet analyses suggested that 17b-estradiol was not homogeneously distributed in the diet using the criteria of 20% of nominal concentration as within specifications. In subsequent diets, homogeneity was achieved by dissolving 17bestradiol in acetone (0.005% v/w for acetone in diet) before addition to the diet. 17b-Estradiol was added to PMI Feeds, Inc., Certified Rodent Diet 5002 and mixed for 6 min. During the test period, male rats in each group were fed a diet of PMI Feeds, Inc., Certified Rodent Diet 5002 that contained 0, 0.05, 2.5, 10, or 50 ppm 17b-estradiol. Animals and treatment. Three-hundred ninety-five male and 520 female Crl:CD BR rats were purchased from Charles River Laboratories Inc. (Raleigh, NC) at approximately 35 days of age for a 90-day feeding and one-generation reproduction study with 17b-estradiol. The materials, methods, and results for that study are described in detail elsewhere (Biegel et al., 1998a). Briefly, male and female rats were fed diets containing concentrations of 0, 0.05, 2.5, 10, or 50 ppm 17b-estradiol for the entire study duration and were assigned either to the 90-day feeding or to the one-generation reproduction subsets. After 70 days of diet administration, rats were bred (1:1 basis) within their respective treatment groups. Mated-female rats were transferred to polycarbonate pans on gestation day 20 until weaning. On postnatal day 4, litters were culled to 8 (4/sex when possible). At weaning (postnatal day 21), selected F1 rats were either euthanized for a gross pathological examination or selected to continue on study at which time rats were individually housed. The current study represented a subset of males used to evaluate the effects of dietary 17bestradiol on testicular and hormonal parameters. Ten males per group were assigned for the sperm parameter assessment for the P1, P1 recovery, F1, and F1 recovery groups and an additional 15 males/group were evaluated at four time points for hormonal status. The histopathological evaluation of the testis and epididymis was conducted on animals assigned to the subchronic (P1) and one-generation (F1) subsets. All rats were housed individually in stainless-steel, wire-mesh cages during the study, except where previously noted. Rats were provided tap water (United Water Delaware) and PMI Feeds, Inc., Certified Rodent Diet 5002 (with or without 17b-estradiol) ad libitum. Rats were quarantined for 6 days and observed with respect to weight gain and any gross signs of disease or injury. Rats were released from quarantine by the laboratory veterinarian on the bases of body weights and clinical observations. Following quarantine release, rats were assigned to treatment groups by computerized, stratified randomization into male control and treatment groups so that there were no statistically significant differences among group body weight means. Additionally, the animals for the current study were randomly assigned to sperm parameter assessment or hormonal evaluation from their respective dose

157

groups. Animal rooms were maintained at a temperature of 23 6 1°C and a relative humidity of 50 6 10% and were artificially illuminated with fluorescent light on a 12-h light/dark cycle (0600 – 1800 h). In a few instances, the temperature and humidity were outside the stated ranges, but the magnitude and duration were minimal and judged to be of no consequence. Rats were weighed weekly during the study. Individual food consumption was also determined weekly throughout the study. From these determinations, individual body weight gain, daily food consumption, and food efficiency were determined. Cage-side examinations were conducted at least daily throughout the study. At least once each week, rats were individually handled and carefully examined for abnormal behavior and/or appearance. Hormonal measurements. Serum hormone concentrations were evaluated at four time points during the course of the study; these time points occurred at 7, 28, and 90 days of feeding in the P1 generation and on test day 77 in the F1 generation males. At each of these time points blood was collected from the tail vein of the 15 designated male rats/group. Blood was collected in the morning between the hours of 0600 and 0900 h. From previous work, we have determined that the tail vein blood collection procedure increases serum prolactin levels when compared to rats whose blood was collected with an indwelling catheter. Even though serum prolactin levels are elevated, compound-induced increases in prolactin (e.g., haloperidol treatment) are detected using this method of blood collection (unpublished data). Serum was prepared and stored between 265 and 285°C until analyzed for T, E2, PRL, LH, and FSH by commercially available RIA kits. These assays were conducted using the protocols supplied by the manufacturers. The range of the calibrators for each assay are as follows: T (20 – 1600 ng/ml), E2 (1–500 pg/ml), PRL (0.8–50 ng/ml), LH (0.8–50 ng/ml), and FSH (1.6 –100 ng/ml). Sperm parameters. The 10 males/group that were designated for sperm assessment were euthanized by carbon dioxide anesthesia and exsanguination. Testis and epididymis weights have been shown to be body weight independent in dietary restriction studies; hence, these data were reported on an absolute rather than on a relative to body weight ratio basis (Chapin et al., 1993; Cook et al., 1993). The right epididymis was removed and weighed. The right cauda epididymis was excised and placed in 32°C PBS with 10 mg/mL BSA, pricked with a needle to facilitate release of sperm, and placed in an incubator at 32°C for 5 min. After incubation, an aliquot was placed in a sample chamber (50 mM depth, Conception Technology, La Jolla, CA) and transferred to a heated microscope stage (34°C) containing a camera linked to a video cassette recorder. At least 10 fields per sample were videotaped for approximately 10 s per field. The videotapes were later visually analyzed, and the percentage of motile cells among at least 200 cells examined per animal was determined. Motility was manually determined by marking the initial head location of the sperm and advancing the videotape 10 frames. Sperm were classified as motile if there was any progressive movement of the sperm head after the videotape was advanced 10 frames (1 s 5 30 frames). After removal of an aliquot of sample for videotaping, the remainder of the right cauda epididymis was minced and further incubated. An aliquot was stained with eosin, and smears were prepared on microscope slides. Sperm smears were examined to determine the frequency of morphologically abnormal sperm, expressed as the percentage of normal cells among at least 200 cells examined per animal. Sperm morphology was performed as a subjective analysis. Sperm were classified as abnormal if they did not display the characteristic ‘‘hook-shaped’’ head or if the tail was compressed resulting in a ‘‘kink-like’’ appearance. The specific types of abnormal morphology, if any, were not categorized. The left epididymis and testis were flash frozen in liquid nitrogen and stored between 265 and 285°C until analyzed. After thawing, the epididymis was weighed and the cauda epididymis was excised, weighed, and homogenized. After thawing, the testis was weighed, decapsulated, and homogenized. Sperm count per cauda epididymis and spermatid count per testis (homogenizationresistant spermatid heads) were determined as previously described (reviewed in Amann, 1981). Sertoli cell morphometry. The right testis was removed and fixed vascularly via the testicular artery (Johnson et al., 1980, 1984a) and immersed

158

COOK ET AL.

in 2% glutaraldehyde in sodium cacodylate buffer. Five pieces of testicular tissue from each rat were further fixed in 1% osmium in sodium cacodylate buffer and embedded in Epon. Tissues were sectioned or serial sectioned at 0.5 or 20 mm. The 0.5-mm Epon sections were stained with toluidine blue and used for stereologic determination of the volume density (percentage) of Sertoli cell nuclei (Johnson et al., 1984b). The 20-mm Epon sections were observed unstained by Nomarski optics (Johnson et al., 1980, 1984b, 1996a) in order to determine the maximum height and width of a Sertoli cell nucleus totally embedded in the Epon sections (Johnson et al., 1996a,b, 1986). The average height and width (based on measurement of 35–50 nuclei) of the nuclei were used to calculate a rough estimate of the volume of a single nucleus, assuming the nucleus to be a sphere. Since Sertoli cell nuclei are not spherical, a correction factor was used to obtain a corrected final volume for an individual nucleus. This correction factor was calculated at 0.663 6 0.025 for intact rats (Johnson et al., 1996a). The corrected volume of a single nucleus was the product of the rough volume estimate of a single nucleus times 0.663. The number of Sertoli cells per testis was calculated when the product of the volume density of Sertoli cell nuclei, parenchymal volume, and the approximated histological correction factor for section thickness and nuclear diameter (assuming the most closely related spherical model) (Weibel and Paumgarter, 1978) was divided by the corrected volume of a single Sertoli cell nucleus. The approximated histological correction factor was 0.96 for volume density of Sertoli cell nuclei. The relative section thickness (average maximum diameter divided by section thickness of 0.5 mm) was much less than the 0.1-mm cutoff point (at which correction has no significant value) needed to correct for spherical structures (Bolender, 1978). While the correction factor for section thickness and nuclear diameter lowered the estimate of Sertoli cells/testis by 4%, no correction would result in an overestimation of only a few percentage points of the absolute value (Bolender, 1978). Volume density of Sertoli cell nuclei in each rat was based on the number of points over Sertoli cell nuclei divided by total points applied using a point-counting method and a 50-point ocular grid at 10003 magnification (Johnson et al., 1980; Elias et al., 1978). Sections averaging over 10 mm2 each were analyzed by two observers for each testis by scoring a total of 15,000 points. Precision of Sertoli cell nuclei volume density has been estimated at 14% coefficient of variation for our laboratory (Johnson and Nguyen, 1986). In addition, the ratio of the spermatid number/testis to the number of Sertoli cell nuclei/testis was determined. Statistics. Body weights, body weight gains, food consumption, food efficiency, and organ weights were analyzed by a one-way analysis of variance. When the test for differences among test group means (the value of the F test statistic) was significant, pairwise comparisons between test and control groups were made with Dunnett’s test. When the Bartlet’s test for normality was significant ( p , 0.005) for organ weights, pairwise comparisons between test and control groups were made with Dunn’s test. Hormonal measurements, testicular spermatid number, epididymal sperm number, sperm motility, and sperm morphology were statistically analyzed using Jonckheere’s test for trend. If a significant dose–response trend was detected, data from the top dose group was excluded and the test repeated until no significant trend was detected. When only two groups were present Mann-Whitney test was used. The number of Sertoli cell nuclei per gram parenchyma and number of Sertoli cell nuclei per testis were analyzed using Student–Newman–Keuls test. The level of significance was p # 0.05.

RESULTS

In-life measurements. In the subset of P1 rats used for sperm analysis, 17b-estradiol produced a dose-dependent decrease in body weight from test day 7 onward (Fig. 1A). These decreases in body weight were statistically significant in the 2.5 to 50 ppm 17b-estradiol groups. Dose-dependent and sta-

FIG. 1. Body weights of P1 and F1 rats. In the P1 generation, rats were 49 days of age on test day 0. In the subset of P1 rats used for sperm analysis, 17b-estradiol produced a dose-dependent decrease in body weight from test day 7 onward (A). These decreases were statistically significant in the 2.5 to 50 ppm 17b-estradiol groups. On test day 97, P1 rats were placed on control diet for 105 days (A, arrow). Body weights in the P1 recovery rats never returned to control levels, but significant recovery was observed. In the F1 generation, rats were 21 days of age on test day 0. In the F1 rats, body weight was significantly decreased throughout the study in the 2.5 ppm 17b-estradiol group (B). On test day 77, F1 rats were placed on control diet for 103 days (B, arrow). Body weights in the 2.5 ppm F1 recovery rats also failed to return to control levels.

tistically significant decreases in overall body weight gain (test days 0 –97) were observed at the 2.5, 10, and 50 ppm 17bestradiol groups (71, 44, and 16% of control, respectively) (Table 1). Parallel decreases in overall food consumption and food efficiency were seen in the 2.5 to 50 ppm 17b-estradiol groups. Body weight, body weight gain, food consumption, and food efficiency in the 0.05 ppm 17b-estradiol group were similar to that of the control. In the P1 recovery subset, statistically significant increases in overall body weight gain (test days 97–202) were observed in the 10 and 50 ppm 17b-estradiol groups (188 and 252% of control, respectively) (Table 1). The recovery time period was 105 days on control diet. Overall food consumption was similar across the dose groups. In contrast, overall food efficiency was significantly increased in the 10 and 50 ppm 17b-estradiol recovery groups (188 and 250% of control, respectively). How-

159

DOES IN UTERO EXPOSURE REDUCE SERTOLI CELL NUMBER?

TABLE 1 Effect of Dietary Estradiol on Overalla Body Weight Gains, Food Consumption, and Food Efficiency

Concentration (ppm)

Body weight gain (g)

0 0.05 2.5 10 50

348.8 6 17.0 324.3 6 21.0 246.3 6 11.0* 152.3 6 14.0* 55.7 6 9.0*

Food efficiency (g body weight gain/g food consumed)

Food consumption (g/day)

Food consumption (g/day)

Body weight gain (g)

P1 b

P1 recovery

26.9 6 0.7 26.1 6 1.0 22.8 6 0.5* 19.7 6 0.7* 17.8 6 2.1*

0.166 6 0.01 0.157 6 0.01 0.140 6 0.01* 0.093 6 0.01* 0.043 6 0.01*

115.7 6 8.2 112.3 6 4.2 130.8 6 17.0 217.9 6 19.0* 291.3 6 12.0*

F1 0 0.05 2.5

397.9 6 9.7 434.0 6 13.0 317.7 6 14.0*

Food efficiency (g body weight gain/g food consumed)

27.4 6 0.6 27.3 6 0.9 26.0 6 0.9 27.2 6 1.1 27.7 6 0.8

0.040 6 0.00 0.040 6 0.00 0.047 6 0.01 0.075 6 0.00* 0.100 6 0.00*

F1 recovery

23.4 6 0.5 24.3 6 0.5 20.7 6 1.1*

0.243 6 0.00 0.256 6 0.00 0.224 6 0.01*

149.6 6 11.0 163.0 6 12.0 147.8 6 16.0

27.0 6 0.8 29.3 6 0.8 25.4 6 1.3

0.055 6 0.00 0.055 6 0.00 0.056 6 0.00

a

The overall period for P1, P1 recovery, F1, and F1 recovery are test days 0 –97, 97–202, 0 –70, and 77–178, respectively. Mean 6 SE, n 5 10. * p # 0.05 (Dunnett’s test).

b

ever, the body weights in the 2.5 to 50 ppm 17b-estradiol recovery groups never returned to control levels (Fig. 1). Body weight gain, food consumption, and food efficiency in the 0.05 and 2.5 ppm 17b-estradiol recovery groups were similar to the control during the recovery period. Due to the absence of pups in the 10 and 50 ppm 17bestradiol groups, there were only three groups in the F1 generation (Biegel et al., 1998a). In the 2.5 ppm F1 17b-estradiol group, body weight was significantly decreased throughout the study and was decreased to 75% of control on test day 77 (98 days of age) (Fig. 1B). The 2.5 ppm F1 17b-estradiol group

also had significantly decreased overall body weight gain, food consumption, and food efficiency (80, 88, and 92% of control, respectively) (Table 1). In the 0.05 ppm F1 17b-estradiol group, overall body weight gain, food consumption, and food efficiency were similar to control (Table 1). In the F1 recovery subset, body weight gain, food consumption, and food efficiency in the 0.05 and 2.5 ppm 17b-estradiol groups were similar to control (Table 1). However, the body weights in the 2.5 ppm 17b-estradiol recovery group never returned to control levels while the weights in the 0.05 17b-estradiol recovery group were greater than the control (Fig. 1B).

TABLE 2 Effect of Dietary Estradiol on Testicular and Epididymal Weights Concentration (ppm)

Testis (g)

0 0.05 2.5 10 50

1.73 6 0.06 1.64 6 0.05 1.64 6 0.02 1.19 6 0.12** 0.43 6 0.06**

Epididymis (g)

Testis (g)

P1

P1 Recovery 0.618 6 0.014 0.621 6 0.017b 0.610 6 0.019 0.299 6 0.050** 0.113 6 0.008**

a

b

1.72 6 0.03 1.71 6 0.04 1.78 6 0.06 1.71 6 0.04 1.72 6 0.05

F1 0 0.05 2.5

1.44 6 0.06 1.57 6 0.04 1.47 6 0.03

Mean 6 SE, n 5 10 unless stated otherwise. n 5 9. * p # 0.05 (Dunnett’s test). ** p # 0.05 (Dunn’s test). a b

Epididymis (g)

0.659 6 0.015 0.675 6 0.017 0.664 6 0.023 0.652 6 0.015 0.628 6 0.020 F1 Recovery

0.573 6 0.017 0.579 6 0.009 0.500 6 0.009* b

1.77 6 0.07 1.71 6 0.08 1.66 6 0.07b

0.673 6 0.015 0.626 6 0.026 0.577 6 0.023b,*

160

COOK ET AL.

Organ weights and histopathology. In the P1 rats, 17bestradiol produced a statistically significant decrease in absolute testis weight in the 10 and 50 ppm groups (69 and 25% of control, respectively) (Table 2). Similarly, statistically significant decreases in absolute epididymal weight were observed in the 10 and 50 ppm 17b-estradiol groups (48 and 18% of control, respectively) (Table 2). The decreases in testis weights correlated with histopathologic evidence of interstitial cell atrophy and seminiferous tubule degeneration. Microscopically, slight to moderate, diffuse atrophy of interstitial cells was present in both the 10 and 50 ppm 17b-estradiol groups. Atrophied interstitial cells had scant cytoplasm and loss or reduction of normal cytoplasmic eosinophilia. In the 50 ppm 17b-estradiol group, changes in seminiferous tubules were characterized by marked degeneration and loss of germ cells, most notably elongate spermatids. As a result, the adluminal layer of germ cells was typically composed of round spermatids or spermatocytes (Fig. 2). At 10 ppm 17b-estradiol, changes in the seminiferous tubules were generally very slight and were most commonly characterized by degeneration of individual spermatocytes and spermatids and retention of elongated spermatids near the basement membrane tubules in the second half of the cycle (Fig. 3). Correlative changes in the epididymides were characterized by oligospermia or aspermia, the presence of germ cell debris in the lumen of tubules, and atrophy of epididymal tubules. In the P1 recovery subset, testis and epididymis weights returned to control levels after having been on control diet for 105 days. Microscopic evaluation of the testis and epididymis was not performed in the P1 or F1 recovery rats. In the F1 generation, absolute testis weight was similar among the treatment groups. However, absolute epididymis weight was statistically decreased in the 2.5 ppm 17b-estradiol group (87% of control) (Table 2). As in the F1 generation, the F1 recovery testis weights were similar to the control group, while the epididymal weight in the 2.5 ppm 17b-estradiol recovery group was significantly decreased (86% of control). There were no compound-related histopathological effects observed in the F1 male rats (Biegel et al., 1998a), where the standard battery of subchronic study guideline-specified organs were examined. P1 generation serum hormone levels. As expected, dietary administration of 17b-estradiol resulted in a dose-dependent increase in serum E2 levels at all time points (Table 3). On test day 7, serum E2 levels were significantly increased in the 2.5, 10, and 50 ppm 17b-estradiol groups to 433, 1267, and 4135%

of control, respectively; on test day 28, serum E2 levels were 135, 1041, 6985, and 34794% of control, respectively; and, on test day 90, serum E2 levels were 272, 803, and 3551% of control, respectively. The statistically significant increase in serum estradiol levels in the 0.05 ppm 17b-estradiol group (135% of control) on test day 28 was in all likelihood a spurious finding based on the low concurrent control value (2.4-fold lower than the mean of the control values from either test days 7 or 90) and the absence of any concurrent change in any of the other measured hormonal levels. Hence, there were no compound-related hormonal effects seen in the 0.05 ppm 17b-estradiol group. On test day 28, serum E2 levels in the 10 and 50 ppm 17b-estradiol groups appeared to reach peak concentrations when compared to the measured levels on test days 7 and 90. To confirm this apparent peak, a subset of rats in the 10 and 50 ppm 17b-estradiol groups from each time point were reanalyzed. Upon reanalysis, the serum E2 levels from test days 7 and 28 were similar to their original values but the serum E2 levels on test day 90 were more similar to the concentrations measured on test day 28. These results suggest that a plateau rather than a peak in serum E2 levels had occurred on test day 28. Due to insufficient amounts of serum, complete reanalysis of the E2 levels from the 90-day time point could not be performed to confirm these findings. At all measured time points, dietary administration of 17b-estradiol produced statistically significant increases in serum PRL in the 2.5, 10, and 50 ppm 17b-estradiol groups (Table 3). On test day 7, serum PRL levels were increased to 183, 246, and 699% of control in the 2.5, 10, and 50 ppm 17b-estradiol groups, respectively; on test day 28, serum PRL levels were increased 351, 419, and 218% of control, respectively; and, on test day 90, serum PRL levels were increased 559, 450, and 277% of control, respectively. Interestingly, the highest value in serum PRL levels occurred in the 50 ppm 17b-estradiol group on test day 7, in the 10 ppm 17b-estradiol group on test day 28, and in the 2.5 ppm 17b-estradiol group on test day 90. Dietary administration of 17b-estradiol resulted in a time and dose-dependent decrease in serum T levels (Table 3). On test day 7, serum T levels were significantly decreased in the 50 ppm 17b-estradiol group (10% of control). On test day 28, serum T levels were significantly decreased in the 10 and 50 ppm 17b-estradiol groups (12 and 0% of control, respectively). On test day 90, serum T levels were significantly decreased in the 0.05, 2.5, 10, and 50 ppm 17b-estradiol groups (65,77, 8, and 0% of control, respectively).

FIG. 2. Photomicrograph showing testicular degeneration/atrophy. Testes from a P1 control rat (A) and a P1 rat fed 50 ppm 17b-estradiol (B). Seminiferous tubules of the 17b-estradiol-treated rat are small and the adluminal layer is composed of spermatocytes or round spermatids. Note the diffuse atrophy of the interstitium; bar, 100 mm. FIG. 3. Photomicrograph showing spermacyte degeneration and spermatid retention. Testes from a P1 control rat (A) and a rat fed 10 ppm 17b-estradiol (B). Degenerating spermatocytes (arrowheads) are present in a stage VII tubule, and retention of elongate spermatids is present in a stage X tubule of the 17b-estradiol-treated rat. Note atrophy of Leydig cells in the interstitium (IS) of the treated rat; bar, 30 mm.

DOES IN UTERO EXPOSURE REDUCE SERTOLI CELL NUMBER?

161

162

COOK ET AL.

TABLE 3 Effects of Dietary Estradiol on Hormone Levels of P1 Male Rats Concentration (ppm)

Estradiol (pg/ml)

Prolactin (ng/ml)

0 0.05 2.5 10 50

20.3 6 1.8 25.5 6 4.1 87.9 6 17.0* 257.2 6 32.1* 839.4 6 206.8*

Testosterone (ng/ml)

LH (ng/ml)

FSH (ng/ml)

1.53 6 0.08 1.69 6 0.08 1.90 6 0.15 1.51 6 0.10 1.19 6 0.05*

18.2 6 1.0 18.3 6 1.2 21.9 6 1.3 20.0 6 1.1 12.7 6 0.9*

1.60 6 0.10 1.93 6 0.14 1.60 6 0.09 1.23 6 0.11* 1.11 6 0.04*

14.2 6 0.7 16.2 6 0.7 14.9 6 0.9 12.6 6 0.5 8.5 6 0.4*

2.14 6 0.13c 2.40 6 0.10 2.18 6 0.16c 1.67 6 0.07* 1.77 6 0.11*

12.7 6 0.7c 13.3 6 0.4 14.1 6 0.7 12.6 6 0.5 10.6 6 1.2*

Test day 7 a

32.1 6 5.6 38.6 6 9.1 58.8 6 5.3* 79.1 6 5.9* 224.3 6 38.7*

1.19 6 0.20 1.66 6 0.27 1.42 6 0.21 1.03 6 0.22 0.12 6 0.03* Test day 28

0 0.05 2.5 10 50

8.6 6 0.8 11.6 6 1.0** 89.5 6 9.6* 600.7 6 75.2* 2992.3 6 651.0*

51.0 6 5.4 40.6 6 4.5 179.1 6 15.2* 213.8 6 22.6* 111.3 6 11.1*

0 0.05 2.5 10 50

21.7 6 4.1 14.8 6 1.3 59.0 6 7.2c,* 174.2 6 20.3* 770.6 6 117.9d,*

31.0 6 5.6 30.0 6 2.7 173.3 6 41.6* 139.6 6 12.0* 85.8 6 12.3*

1.38 6 0.42 1.44 6 0.32 0.91 6 0.19 0.16 6 0.06* 0.00 6 0.00* Test day 90

b

1.58 6 0.26c 1.03 6 0.10** 1.22 6 0.26* 0.13 6 0.04* 0.00 6 0.00*

c

Mean 6 SE, n 5 15 unless stated otherwise. n 5 12. c n 5 14. d n 5 13. * p # 0.05 (Jonckheere’s test for trend). ** p # 0.05 (Mann-Whitney tests; performed when only two groups were present). a b

Serum LH and FSH levels were significantly decreased by dietary administration of 17b-estradiol at all time points, however the magnitude of response was less than that observed for the previously discussed hormones. On test day 7, serum LH and FSH levels in the 50 ppm 17b-estradiol group were significantly decreased to 78 and 70% of control, respectively (Table 3). On test day 28, serum LH levels in the 10 and 50 ppm 17b-estradiol groups were significantly decreased to 77 and 69% of control, respectively; while only serum FSH levels in the 50 ppm 17b-estradiol group were significantly decreased (60% of control). On test day 90, serum LH levels in the 10 and 50 ppm 17b-estradiol groups were significantly decreased to 78 and 83% of control, respectively, while only serum FSH levels in the 50 ppm 17b-estradiol group were significantly decreased to 83% of control. F1 generation serum hormone levels. Dietary administration of 17b-estradiol produced significantly increased serum E2 and PRL levels (443 and 267% of control, respectively) and significantly decreased T and LH levels (33 and 83% of control, respectively) in the 2.5 ppm 17b-estradiol group (Table 4). Serum FSH levels were unchanged in the 2.5 ppm 17bestradiol group. There were no statistically significant changes in serum hormone levels in the 0.05 ppm 17b-estradiol group.

Sperm parameters. In the P1 rats, 17b-estradiol decreased testicular spermatid and epididymal sperm numbers in the 10 and 50 ppm groups and decreased sperm motility and percentage of normal sperm (not statistically significant) in the 10 ppm group (Table 5). Sperm motility and morphology could not be evaluated in the 50 ppm 17b-estradiol group due to the absence of sperm. In the 10 and 50 ppm 17b-estradiol groups, testicular spermatid number was significantly decreased to 53 and 1% of control, respectively, while epididymal sperm number was significantly decreased to 23 and 0% of control, respectively. In the 10 ppm 17b-estradiol group, the percent motile sperm was significantly decreased to 93% of control. After 105 days of recovery, P1 testicular spermatid number, sperm motility, and morphology had completely recovered in the 10 and 50 ppm 17b-estradiol groups (Table 5). Although epididymal sperm number had nearly returned to control levels, it was still significantly decreased in the 10 and 50 ppm groups (82 and 78% of control, respectively). In contrast to the P1 rats, testicular spermatid number in the F1 rats was significantly increased to 135 and 140% of control in the 0.05 and 2.5 ppm 17b-estradiol groups, respectively (Table 6). This increase appears to be due to the unusually low control value. Epididymal sperm number was significantly

163

DOES IN UTERO EXPOSURE REDUCE SERTOLI CELL NUMBER?

TABLE 4 Effects of Dietary Estradiol on Hormone Levels of F1 Male Rats Concentration (ppm)

Estradiol (pg/ml)

Prolactin (ng/ml)

Testosterone (ng/ml)

LH (ng/ml)

FSH (ng/ml)

3.22 6 0.16 2.93 6 0.11 2.66 6 0.12*

17.3 6 0.7 16.5 6 0.6 17.1 6 0.6

Test day 77a 0 0.05 2.5

4.2 6 0.6b 3.5 6 0.6 18.6 6 1.9c,*

25.5 6 1.4 27.3 6 3.4 68.2 6 8.9*

2.31 6 0.33 1.90 6 0.24 0.77 6 0.18*

a

Age of F1 male rats on test day 77 was 98 days. Mean 6 SE, n 5 15 unless stated otherwise. c n 5 14. * p # 0.05 (Jonckheere’s test for trend). b

decreased in the 2.5 ppm 17b-estradiol group (75% of control). Sperm motility was unaltered. Even though a statistically significant decrease in morphologically normal sperm was observed in the 2.5 ppm 17b-estradiol group, this decrease was judged not to be biologically significant due to the small magnitude of change and since it is within the expected range for control rats within our laboratory. In the F1 recovery subset, testicular spermatid number, sperm motility, and morphology in the 17b-estradiol groups were similar to the control values (none of these endpoints were adversely altered in the F1 rats) (Table 6). However, no apparent improvement in epididymal sperm number had occurred during the 103-day recovery period in the 2.5 ppm 17b-estradiol group (65% of control).

Sertoli cell number. Morphometric analysis of the testis was conducted on F1 rats euthanized on test day 72. There was no difference (p . 0.05) in the number of Sertoli cell nuclei per gram of parenchyma, number of Sertoli cell nuclei/ testis, or mature spermatid number per Sertoli cell nuclei per testis ratio among any of the 17b-estradiol treatments groups when compared to the control (Table 7). Based on our data and other work, a decrease in Sertoli cell number of approximately 17% would be detected as being statistically significant using a sample size of 10. Likewise, there was no dose-related trend for higher concentrations of estrogen to reduce the number of Sertoli cells. The absolute number of Sertoli cells in each testis of Crl:CD BR rats (Table 7) is within the reported range of 22 3 106/testis (Wing and Christensen, 1982) or 54 3 106/

TABLE 5 Effect of Dietary Estradiol on Sperm Parameters in P1 Male Rats Concentration (ppm)

Testicular spermatid number (per testis 3 106)

0 0.05 2.5 10 50

94.6 6 4.0 84.5 6 5.8 92.0 6 5.8 50.6 6 9.5* 0.9 6 0.4*

Epididymal sperm number (per cauda epididymis 3 106)

Sperm mobility (% motile)

Morphology (% normal)

87.2 6 0.9 87.7 6 1.4 82.9 6 2.5 81.0 6 2.9b,* —d

97.9 6 0.4 96.6 6 1.0 97.7 6 0.8 92.5 6 2.5c —d

80.7 6 3.1 81.3 6 3.1 84.0 6 1.6e 83.8 6 2.1 76.5 6 2.7e

98.1 6 0.5 97.0 6 0.6 97.2 6 0.4e 95.8 6 0.9 98.2 6 0.4

P1 a

195.2 6 11.6 188.3 6 6.1 192.4 6 11.5 44.2 6 14.1* 0.1 6 0.1* P1 Recovery

0 0.05 2.5 10 50

76.9 6 6.0 72.9 6 4.6 74.4 6 6.9 82.2 6 4.6 82.0 6 6.3

Mean 6 SE, n 5 10 unless stated otherwise. n 5 5. c n 5 7. d Parameter could not be assessed as a result of zero sperm present. e n 5 9. * p # 0.05 (Jonckheere’s test for trend). a b

216.2 6 13.7 207.3 6 16.2 201.4 6 17.5 176.9 6 13.2* 168.8 6 16.7*

164

COOK ET AL.

TABLE 6 Effect of Dietary Estradiol on Sperm Parameters in F1 Male Rats Concentration (ppm)

Testicular spermatid number (per testis 3 106)

0 0.05 2.5

46.8 6 4.7 63.3 6 4.1* 65.6 6 3.8**

Epididymal sperm number (per cauda epididymis 3 106)

Sperm motility (% motile)

Morphology (% normal)

85.7 6 2.0 83.9 6 1.1 82.7 6 2.4

99.2 6 0.4 98.6 6 0.5 97.0 6 0.7**

82.5 6 3.1 79.8 6 3.8b 79.6 6 2.8b

96.1 6 1.0 96.0 6 1.1 97.8 6 0.4b

F1 144.1 6 5.5 157.9 6 10.6 107.9 6 2.1**

a

F1 recovery 118.6 6 9.2 114.1 6 9.7 109.9 6 9.5b

0 0.05 2.5

229.1 6 9.2 199.8 6 15.8 148.8 6 12.0b,**

Mean 6 SE, n 5 10 unless stated otherwise. n 5 9. * p # 0.05 (Mann-Whitney tests; performed when only two groups were present). ** p # 0.05 (Jonckheere’s test for trend). a

b

testis (Hochcreau-de-Reviers and Courot, 1978) for rats. However, the number of Sertoli cells in the testis of a Crl:CD BR rat is greater than the 26 3 106 found in adult Fischer rats (Johnson et al., 1996) evaluated at the same laboratory. DISCUSSION

Much of the concern around EACs has focused on estrogenicity (Harrison et al., 1995; Toppari et al., 1995) because of the unexpected findings that several commodity chemicals such as bisphenol A (Dodds and Lawson, 1936; Krishnan et al., 1993) and nonylphenol (Soto et al., 1991; Jobling and Sumpter, 1993) are weakly estrogenic and could leach out of finished products. These findings have led some individuals to call for additional toxicology testing for estrogenicity (Colborn et al., 1996). The recently passed Food Quality Protection Act of 1996 requires the U.S. EPA to implement screening strategies for EACs within the next 2 years. Because of the absence of dietary exposure data with 17b-estradiol, a 90-day/one-

TABLE 7 Effect of Dietary Estradiol on Sertoli Cell Number from F1 Male Rats

Concentration (ppm)

Number of Sertoli cell nuclei/g parenchyma (106)

Number of Sertoli cell nuclei/testis (106)

Ratio (spermatid number/testis)/ (number Sertoli cell nuclei/testis)

0 0.05 2.5

30.7 6 2.5a 35.2 6 1.3 31.6 6 1.6

47.5 6 3.4 54.6 6 1.5 45.8 6 2.8

1.1 6 0.2 1.2 6 0.1 1.5 6 0.1

Note. No significant differences detected using Student-Newman-Keuls test. Mean 6 SE, n 5 10.

a

generation reproduction study was designed to set dose levels for a multigeneration reproduction and a combined chronic toxicity/oncogenicity studies. The purpose of these studies will be to provide a benchmark for other estrogenic compounds. Administration of dietary levels of 0.05, 2.5, 10, and 50 ppm 17b-estradiol produced a dose-dependent decrease in body weight in P1 male rats in the 2.5 to 50 ppm groups. This decrease in body weight gain was due to a combination of reduced food consumption and food efficiency. Growth inhibition has been proposed as a sensitive measure of the potency of estrogens (Heywood and Wadsworth, 1980). Similar changes were observed in the F1 male rats in the 2.5 ppm 17b-estradiol group, demonstrating that in utero exposure did not result in a lower no-observable-adverse-effect level (NOAEL) for these endpoints. When a subset of P1 male rats was placed on control diet for 105 days, body weight increased due to food consumption returning to control levels accompanied by an increase in food efficiency. In contrast, the recovery F1 male rats in the 2.5 ppm 17b-estradiol group did not have a corresponding increase in body weight or food efficiency. The body weights in the P1 2.5 to 50 ppm and F1 2.5 ppm 17bestradiol groups never returned to control levels. The inability of body weights to fully recover to control levels has been shown to occur in dietary restricted rats following a 3-week recovery (Grewal et al., 1971). Although their recovery period was much shorter than the current study, the data in the current study demonstrate that the body weight increased mostly during the first 3 weeks of recovery followed by a slower but parallel increase relative to the control for the remainder of the recovery period. Hence, comparison of data from Grewal and co-workers (1971) to this study is reasonable and suggests that the inability of the body weights to return to control levels is not unique to 17b-estradiol.

DOES IN UTERO EXPOSURE REDUCE SERTOLI CELL NUMBER?

In the P1 rats, 17b-estradiol decreased testis and epididymis weights in the 10 and 50 ppm groups, while no adverse effects were seen in the P1 2.5 ppm groups. In contrast, epididymis weights in the F1 and F1 recovery 2.5 ppm groups were significantly decreased: however, there were no histopathological effects observed. The decreases in testis weights at P1 10 and 50 ppm correlated with histopathologic evidence of interstitial (Leydig) cell atrophy and seminiferous tubule degeneration. Correlative changes in the epididymis of P1 rats were characterized by oligospermia or aspermia, the presence of germ cell debris in the lumen of tubules, and atrophy of epididymal tubules (see also Biegel et al., 1998a). Similar changes were reported in the testis and epididymis when 17bestradiol was administered by subcutaneous injection for 12 weeks (Attia and Zayed, 1989). When 17b-estradiol was administered for 15 days by intraperitoneal injection, a minor spermatid retention in the basal regions of the late-stage tubules was observed (O’Connor et al., 1998), which was also present in the current study. In the P1 recovery subset, testis and epididymis weights returned to controls levels after having been on control diet for 105 days, which is consistent with the dramatic recovery of sperm production. These data suggest that following exposure of adult male rats to high levels of 17bestradiol nearly complete recovery can occur once compound administration is ceased. The recovery is certainly not unexpected based on the fact that there was no evidence of effects on the stem cells of the testis in this study. The pattern of hormonal responses (as well as the histopathological changes in the testis) seen in this study are characteristic of an estrogen receptor agonist such as 17bestradiol. Namely, estrogen receptor agonists increase serum PRL and decrease T, LH, and FSH (Knobil and Neill, 1994; DeGroot, 1995). There were no significant hormonal effects in the P1 and F1 generations at 0.05 ppm, even though the F1 generation had a higher mean daily intake of 17b-estradiol than the P1 generation (Biegel et al., 1998a). Clearly, the pattern of hormonal responses between the two generations were not dramatically different at the 2.5 ppm 17b-estradiol level. However, there are a few noteworthy observations in this study. First, on test day 28, serum E2 levels in the 10 and 50 ppm 17b-estradiol groups appeared to reach peak concentrations. To confirm this apparent peak, serum E2 levels were reanalyzed in a subset of rats in the 10 and 50 ppm 17b-estradiol groups from each time point. This reanalysis suggested that a plateau in serum E2 levels had occurred, but due to insufficient amounts of serum, complete reanalysis of the E2 levels from the 90-day time point could not be performed to confirm these findings. Second, the decreases in serum LH and FSH were not as severe as would be predicted by the magnitude of the T decreases. This inconsistency may result from competing feedback loops. Increasing levels of serum E2 would feedback to centrally inhibit gonadotropin release (Knobil and Neill, 1994; DeGroot, 1995) and could also directly inhibit T biosynthesis

165

by the Leydig cell by inhibiting the 17a-hydroxylase/C-17, 20-lyase enzyme (Tsai-Morris et al., 1986; Nishihara et al., 1988). This decrease in T would also feedback centrally to stimulate gonadotropin secretion, hence attenuating the signal received by the elevated serum E2 levels (Knobil and Neill, 1994; DeGroot, 1995). Finally, the peak in serum PRL levels occurred in the 50 ppm 17b-estradiol group on test day 7, in the 10 ppm 17b-estradiol group on test day 28, and in the 2.5 ppm 17b-estradiol group on test day 90. The subsequent decreases in serum PRL at the higher dose groups may be due to reduced secretory capacity of the lactotrophes, possibly due to desensitization, similar to that reported with insulin (Flier, 1995). Dietary administration of 17b-estradiol to male rats caused reductions in testicular spermatid and epididymal sperm numbers and sperm motility. In the P1 males, these effects were observed in the 10 and 50 ppm groups, except in the P1 males where aspermia precluded evaluation of sperm motility and sperm morphology. Following a 105-day recovery period in the P1 males, all sperm parameters returned to control values except for epididymal sperm number, which had nearly recovered. The decline in testicular spermatid and epididymal sperm numbers correlated with the reduced organ weights and the observed histopathological changes. In the F1 rats, no significant decreases were noted in the testicular spermatid number. The control value for the F1 males was considerably lower than expected and resulted in the statistically significant increase observed in the testicular spermatid number in the 0.05 and 2.5 ppm groups. Since T is the primary paracrine hormone of the testis, these changes would appear to be primarily due to the dramatic decrease in serum T levels observed in the P1 10 and 50 ppm groups. However, other endocrine factors could contribute to this decline. A slight decrease in T was observed in the 0.05 and 2.5 ppm groups at the 90-day time point, although no changes were seen in the sperm parameters in that group. The role of PRL on testis function is less clear (Sharpe, 1988). PRL has been shown to help regulate the number of Leydig cell LH receptors where hyperprolactemia increases LH receptor number and alters the sensitivity to LH (reviewed in Sharpe, 1993). Hyperprolactemia in men is associated with impotence and infertility, and in rats severe hyperprolactemia produces testicular atrophy (Sharpe, 1994). In the current study, the P1 2.5 ppm 17b-estradiol group had significantly elevated PRL levels and minimally decreased serum T, yet there were no changes in testicular and epididymal weight, histopathology (Biegel et al., 1998a), or sperm parameters. The data in the current study support the conclusion of Sharpe (1994) that regardless of the effects of PRL on the hypothalalmic–pituitary–testicular axis, all of the changes of spermatogenesis appear to be a consequence of altered T levels. Overall, the testicular spermatid numbers in the present study were lower than expected based on our laboratory data and that reported in the literature (Blazak, 1989; Johnson et al., 1980). However, only the F1 data are below the range reported

166

COOK ET AL.

in the literature for the Fischer-344 (F-344) strain, which has a very similar testicular sperm production rate as the Sprague– Dawley (SD). The sperm production rates per gram testis per day (3106) for the F-344 and SD rats are 17 and 15, respectively. The Wistar and Long-Evans strains have a much higher sperm production rate (22–25) (Blazak, 1989; Gray et al., 1989). The reason for the lowered testicular spermatid numbers in the F1 rats is unclear. In contrast to the testicular spermatid numbers, a significant decrease in epididymal sperm numbers was observed in the F1 males, which correlated with the decreased epididymal weight. In addition, this decrease was present following a 101-day recovery period and showed no recovery. Due solely to the magnitude of the epididymal decrease (25%), it would have been expected that the value would have reached control levels following the recovery period. These data demonstrate that a longer recover period is necessary. The decreases in epididymal weight and sperm number in the F1 males appear compound related and certainly warrant additional investigation. If reproducible, the results would suggest an effect of in utero administration of 17b-estradiol on epididymal function, since the present study observed no histological changes in the epididymis of the F1 males. The morphometric evaluation of the testis to determine the number of Sertoli cells was done to specifically address the hypothesis of Sharpe and Skakkebaek (1993). They proposed that in utero exposure to estrogenic compounds could decrease sperm number via estrogen-induced reduction of FSH levels, which in turn would reduce Sertoli cell number. To test this hypothesis, Sertoli cell number was evaluated using samples sizes at least twofold greater than those traditionally used for morphometric analysis (n 5 10) in order to increase statistical power. The number of Sertoli cell nuclei in the testis of F1 rats was similar to control when expressed on a per gram of parachyma or whole testis basis. The values for the number of Sertoli cell nuclei per testis for the control and test groups were all at the high end of the range reported in the literature (15.5–53) (Russell, 1993). The ratio of spermatid number per testis to the number of Sertoli cell nuclei per testis was also unchanged. This ratio is considered a surrogate measure of Sertoli cell function. However, the ratio was much lower than expected (Wing and Christensen, 1982) due to the decreased numerator (spermatid number/testis) and the increased denominator (Sertoli cell nuclei/testis). The reason for the decreases in spermatid number and the increases in Sertoli cell number is unclear. Despite the ratio values being lower than expected, the values were clearly comparable across all groups indicating consistent data collection and support no significant difference in Sertoli cell number. It is especially noteworthy that this absence of Sertoli cell effects occurred at a dietary level of 17b-estradiol that increased serum E2 levels to approximately 400% of control and reduced T levels to 33% of control. This is the first data that the authors are aware of

that evaluate the Sharpe and Skakkebaek (1993) hypothesis. The current study data indicate that in utero exposure to 17b-estradiol does not cause a decrease in Sertoli cell number even at concentrations which produce definitive evidence of endocrine perturbation. The purpose of this range-finder study was to evaluate the effects of dietary 17b-estradiol exposure on serum hormone levels, sperm parameters and Sertoli cell number from P1 and F1 male rats. Overall, there were no apparent compoundrelated effects in the P1 and F1 males at 0.05 ppm, but definitive effects were observed in both generations at 2.5 ppm and above. Subsequent studies would need to be conducted at concentrations below 10 ppm since no litters were produced at this dietary concentration. It appears prudent to select an upper concentration range that explores effects between 2.5 and 10 ppm. Of the endpoints evaluated, in-life measurements (body weight, food consumption, and food efficiency) and serum hormone levels were the most sensitive and did not demonstrate marked differences between the P1 and F1 males. Finally, the measurement of Sertoli cell nuclei per testis was conducted to test the hypothesis that in utero exposure to estrogens will decrease Sertoli cell number and sperm production (Sharpe and Skakkebaek, 1993). The data from the current study do not support the hypothesis that in utero exposure to 17b-estradiol would cause such a decrease in Sertoli cell number. ACKNOWLEDGMENTS We thank Dr. A. Michael Kaplan (DuPont) and Ann M. Mason (Chlorine Chemistry Council) for their stewardship of this project. We also acknowledge the invaluable advice of Drs. Steven F. Arnold (Tulane-Xavier Center for Bioenvironmental Research), James A. Barter (PPG Industries), Robert E. Chapin (NIEHS), William R. Kelce (U.S. EPA), Roland R. Miller (Dow Chemical Company), and Ellen K. Silbergeld (University of Maryland at Baltimore). The expert technical assistance of Suzanne C. Craven, Bryan Crossley, Mary Ann Jacobs, Stephen Novak, Vivian Thompson, and Merralyn Vaillancourt was also appreciated. Finally, the technical support provided by the reproductive toxicology group at Haskell Laboratory was essential in the conduct of this study and was greatly appreciated.

REFERENCES Amann, R. P. (1981). A critical review of methods for evaluation of spermatogenesis from seminal characteristics. J. Androl. 2, 37–58. Ankley, G. T., Johnson, R. D., Toth, G., Fomar, L. C., Detenbeck, N. E., and Bradbury, S. P. (1997). Development of a research strategy for assessing the ecological risk of endocrine disruptors. Rev. Toxicol. 1, 231–267. Attia, M. A. and Zayed, I. (1989). Thirteen-weeks subcutaneous treatment with high dose of natural sex hormones in rats with special reference to their effect on the pituitary-gonadal axis. I. Oestradiol. Dtsch. tieraztl. Wschr. 96, 438 – 445. Auger, J., Kunstman, J. M., Czyglik, F., and Jouannet, P. (1995). Decline in semen quality among fertile men in Paris during the past 20 years. N. Engl. J. Med. 332, 281–285. Biegel, L. B., Flaws, J. A., Hirshfiel, A. N., O’Connor, J. C., Elliott, G. S., Ladics, G. S., Silbergeld, E. K., Van Pelt, C. S., Hurtt, M. E., Cook, J. C., and Frame, S. R. (1998a). 90-day feeding and one-generation reproduction study in Crl:CD BR rats with 17b-estradiol. Toxicol. Sci. 44, 116 –142.

DOES IN UTERO EXPOSURE REDUCE SERTOLI CELL NUMBER? Biegel, L. B., Cook, J. C., Hurtt, M. E., and O’Connor, J. C. (1998b). Effects of 17b-estradiol on serum hormone concentrations and estrous cycle in female Crl:CD BR rats: Effects on parental and first generation rats. Toxicol. Sci. 44, 143–154. Birnbaum, L. S. (1994). Endocrine effects of prenatal exposure to PCBs, dioxins, and other xenobiotics; implications for policy and future research. Environ. Health Perspect. 102, 676 – 679. Blazak, W. F. (1989). Significance of cellular end points in assessment of male reproductive toxicity. In Toxicology of the Male and Female Reproductive Systems (P. K. Working, Ed.), pp. 157–172. Hemisphere Publishing, New York.

167

D. L. Loriaux, J. C. Marshall, W. D. Odell, Jr., and A. H. Rubenstein, Eds.), pp. 1592–1604. Saunders, Philadelphia. Forman, D., Pike, M. C., Davey, G., Dawson, S., Baker, K., Chilvers, C. E. D., Oliver, R. T. D., Coupland, C., Cartwright, R. A., Elwood, P. C., Birch, J., Tyrell, C., Brett, R., Bush, T., Isbell, V., Cornwell, A., Steer, R., Thistlethwaite, S., Gellman, H., Hughes, J., Llewellyn, M., Andernjones, A., Allen, A., Hilton, E., Lloyd, B., McVeigh, S., Thorne, M., Trowbridge, P., and Reid, S. (1994). Aetiology of testicular cancer: association with congenital abnormalities, age at puberty, infertility, and exercise. Br. Med. J. 308, 1393–1399.

Bolender, R. P. (1978). Correlation of morphometry and sterology with biochemical analysis of cell fractions. Int. Rev. Cytol. 55, 247–289.

Gill, W. B., Schumacher, F. B., Bibbo, M., Straus, F. H., and Schoenberg, H. W. (1979). Association of diethylstilbestrol exposure in utero with cryptorchidism, testicular hypoplasia and semen abnormalities. J. Urol. 122, 36 –39.

Buetow, S. A. (1995). Epidemiology of testicular cancer. Epidemiol. Rev. 17, 433– 449.

Ginsburg, J., Okolo, S., Prelevic, G., and Hardiman, P. (1994). Residence in London area and sperm density. Lancet 343, 230.

Carlse¨n, E., Giwercman, A., Keiding, N., and Skakkebaek, N. E. (1992). Evidence for decreasing quality of semen during past 50 years. Br. Med. J. 305, 609 – 613.

Giwercman, A., Mu¨ller, J., and Skakkebaek, N. E. (1988). Cryptorchidism and testicular neoplasis. Horm. Res. 30, 157–163.

Chapin, R. E., Gulati, D. K., Barnes, L. H., and Teague, J. L. (1993). The effects of feed restriction on reproductive function of Sprague–Dawley rats. Fundam. Appl. Toxicol. 20, 23–29. Colborn, T., and Clement, C. (1992). Chemically Induced Alterations in Sexual and Functional Development: The Wildlife/Human Connection. Princeton Scientific, Princeton, NJ.

Gray, L. E., Jr., Ostby, J., Ferrell, J. Rehnberg, G., Linder, R., Cooper, R. Goldman, J., Slott, V., and Laskey, J. (1989). A dose–response analysis of methoxychlor-induced alterations of reproductive development and function in the rat. Fundam. Appl. Toxicol. 12, 92–108. Grewal, T., Mickelsen, O., and Hafs, H. D. (1971). Androgen secretion and spermatogenesis in rats following semistarvation. PSEBM 138, 723–727.

Colborn, T., vom Saal, F. S., and Sota, A. M. (1993). Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ. Health Perspect. 101, 378 –384.

Harrison, P. T. C., Humfrey, C. D. N., Litchefield, M., Peakall, D., and Shuker, L. K. (1995). IEH Assessment on Environmental Oestgrogens: Consequences to Human Health and Wildife. MRC Institute for Environment and Health, Leicester, UK.

Colborn, T., Dumanoski, D., and Myers, J. P. (1996). Our Stolen Future: Are We Threatening Our Fertility, Intelligence, and Survival? A Scientific Detective Story. Dutton Books, New York.

Herbst, A. L., Ulfelder, H., and Poskanzer, D. C. (1971). Adenocarcinoma of the vagina: Association of maternal stilbestrol therapy with tumor appearance in young women. N. Engl. J. Med. 284, 878 – 881.

Cook, J. C., Mullin, L. S., Frame, S. R., and Biegel, L. B. (1993). Investigation of a mechanism for Leydig cell tumorigenesis by linuron in rats. Toxicol. Appl. Pharmacol. 119, 195–204.

Heywood, R., and Wadsworth, P. F. (1980). The experimental toxicology of estrogens. Pharmacol. Ther. 8, 125–142.

Crisp, T. M., Clegg, E. D., Cooper, R. L., Anderson, D. G., Baetcke, K. P., Hoffmann, J. L., Morrow, M. S., Rodier, D. J., Schaeffer, J. E., Touart, L. W., Zeeman, M. G., Patel, Y. M., and Wood, W. P. (1997). Special Report on Environmental Endocrine Disruption: An Effects Assessment and Analysis, EPA/630/R-96/012, Risk Assessment Forum. U.S. Environmental Protection Agency, Washington, DC. Davis, D. L., Bradlow, H. L., Wolff, M., Woodruff, T., Hoel, G., and AntonCulver, H. (1993). Medical hypothesis: Xenoestrogens as preventable causes of breast cancer. Environ. Health Perspect. 101, 372–377.

Hochereau-de-Reviers, M. T., and Courot, M. (1978). Sertoli cells and development of seminiferous epithelium. Ann. Biol. Anim. Biochem. Biophys. (2B) 18, 573–583. Houghton, D. L., and Ritter, L. (1995). Organochlorine residues and risk of breast cancer. J. Am. Coll. Toxicol. 14, 71– 89. Irvine, S., Cawood, E., Richardson, D., MacDonald, E., and Aitken, J. (1996). Evidence of deteriorating semen quality in the United Kingdom: Birth cohort study in 577 men in Scotland over 11 years. Br. Med. J. 312, 467– 471.

Davis, D. L., and Bradlow, H. L. (1995). Can environmental estrogens cause breast cancer? Sci. Am. 273, 166 –172.

Jobling, S., and Sumpter, J. P. (1993). Detergent components in sewage effluent are weakly oestrogenic to fish: An in vitro study using rainbow trout (Oncorhynchus mykiss) hepatocytes. Aquat. Toxicol. 27, 361–372.

DeGroot, L. J. (1995). Endocrinology (L. J. De Groot, M. Besser, H. G., Burger, J. L. Jameson, D. L. Loriaux, J. C. Marshall, W. D. Odell, Jr., and A. H. Rubenstein, Eds.). Saunders, Philadelphia.

Johnson, L., Petty, C. S., and Neaves, W. B. (1980). A comparative study of daily sperm production and testicular composition in humans and rats. Biol. Reprod. 22, 1233–1243.

Dodds, E. C., and Lawson, W. (1936). Synthetic oestrogenic agents without the phenanthrene nucleus. Nature 137, 996.

Johnson, L., Lebovitz, R. M., and Samson, W. K. (1984a). Germ cell degeneration in normal and microwave-irradiated rats: Potential sperm production rates at different developmental steps in spermatogenesis. Anat. Rec. 209, 501–507.

Elias, H., Pauly, J. E., and Burns, E. R. (1978). Quantitative microscopy (sterology and morphometry). In Histology and Human Microanatomy, pp. 547–576. Wiley, New York. Fisch, H., and Goluboff, E. T. (1996). Geographic variations in sperm counts: A potential cause of bias in studies of semen quality. Fertil. Steril. 65, 1044 –1046. Fisch, H., Goluboff, E. T., Olson, J. H., Feldshuh, J., Broder, S. J., and Barad, D. H. (1996). Semen analyses in 1,283 men from the United States over a 25-year period: no decline in fertility. Fertil. Steril. 65, 1009 –1014. Flier, J. S. (1995). Syndromes of insulin resistance and mutant insulin. In Endocrinology (L. J. DeGroot, M. Besser, H. G., Burger, J. L. Jameson,

Johnson, L., Zane, R. S., Petty, C. S., and Neaves, W. B. (1984b). Quantification of the human Sertoli cell population: its distribution, relation to germ cell numbers, and age-related decline. Biol. Reprod. 31, 785–795. Johnson, L., and Nguyen, H. B. (1986). Annual cycle of the Sertoli cell population in adult stallions. J. Reprod. Fertil. 76, 311–316. Johnson, L., Suggs, L. C., Norton, Y. M., and Zeh, W. C. (1996a). Effect of developmental age or time after transplantation on Sertoli cell number and testicular size in inbred Fischer rats. Biol. Reprod. 54, 948 –959. Johnson, L., Suggs, L. C., Norton, Y. M., Welsh, T. J., and Wilker, C. E.

168

COOK ET AL.

(1996b). Effect of hypophysectomy, sex of host, and/or number of transplanted testes on Sertoli cell number and testicular size of syngeneic testicular grafts in Fischer rats. Biol. Reprod. 54, 960 –969. Johnson, L., Welsh, T. H., Jr., and Wilker, C. E. (1997). Anatomy and physiology of the male reproductive system and potential targets of toxicants. In Comprehensive Toxicology, Volume 10, Reproductive and Endocrine Toxicology (K. Boekelheide, R. E. Chapin, P. B. Hoyer, and C. Harris, Eds.), pp. 5– 61. Elsevier, New York. Kavlock, R. J., Daston, G. P., DeRosa, C., Fenner-Crisp, P., Gray, L. E., Jr., Kaattari, S., Lucier, G., Luster, M., Mac, M. J., Maczka, C., Miller, R., Moore, J., Rolland, R., Scott, G., Sheehan, D. M., Sinks, T., and Tilson, H. A. (1996). Research needs for the assessment of health and environmental effects of endocrine disuptors: a report of the U.S. EPA-sponsored workshop. Environ. Health Perspect. 104(Suppl. 4), 715–740. Kelce, W. R., Monosson, E., Gamscik, M. P., Laws, S. C., and Gray, L. E., Jr. (1994). Environmental hormone disruptors: Evidence that vinclozolin developmental toxicity is mediated by antiandrogenic metabolites. Toxicol. Appl. Pharmacol. 126, 276 –285. Knobil, E., and Neill, J. D. (1994). The Physiology of Reproduction (E. Knobil, J. D. Neill, G. S. Greenwald, C. L. Markert, and D. W. Pfaff, Eds.). Raven Press, New York. Krishnan, A. V., Stathis, P., Permuth, S. F., Tokes, L., and Feldman, D. (1993). Bisphenol A: An estrogenic substance is released from polycarbonate flasks during autoclaving. Endocrinology 132, 2279 –2286. MacLeod, J., and Wang, Y. (1979). Male fertility potential in terms of semen quality: A review of the past, a study of the present. Fertil. Steril. 31, 103–116. National Academy of Sciences (NAS) (1995). Hormone-Related Toxicants in the Environment. National Research Council, Commission on Life Sciences, Co-operative Agreement with EPA, Washington, DC. Nishihara, M., Winters, C. A., Buzko, E., Waterman, M. R., and Dufau, M. L. (1988). Hormonal regulation of rat Leydig cell cytochrome P-45017a mRNA levels and characterization of a partial length rat P-45017a cDNA. Biochem. Biophys. Res. Commun. 154, 151–158. O’Connor, J. C., Frame, S. R., Biegel, L. B., Cook, J. C., and Davis, L. G. (1998). Sensitivity of a tier I screening battery compared to an in utero exposure for detecting the estrogen receptor agonist 17b-estradiol. Toxicol. Sci. 44, 169 –184. Olsen, G. W., Bodner, K. M., Ramlow, J. M., Ross, C. E., and Lipshultz, L. I. (1995). Have sperm counts been reduced 50 percent in 50 years? A statistical model revisited. Fertil. Steril. 63, 887– 893. Paulsen, C. A., Berman, N. G., and Wang, C. (1996). Data from men in greater Seattle area reveals no downward trend in semen quality: Further evidence that deterioration of semen quality is not geographically uniform. Fertil. Steril. 65, 105–1015. Russell, L. D. (1993). Form, dimensions, and cytology of mammalian Sertoli

cells. In The Sertoli Cell (L. D. Russell and M. D. Griswold, Eds.), pp. 1–37. Cache River Press, Clearwater, FL. Safe, S. H. (1995). Environmental and dietary estrogens and human health: is there a problem? Environ. Health Perspect. 103, 346 –351. Sharpe, R. M. (1988). Endocrinology and paracrinology of the testis. In Physiology and Toxicology of Male Reproduction (J. C. Lamb, IV, and P. M. D. Foster, Eds.), pp. 71–102. Academic Press, San Diego. Sharpe, R. M. (1994). Regulation of spermatogenesis. In The Physiology of Reproduction (E. Knobil and J. D. Neill, Eds.), 2nd ed., pp. 1363–1434. Raven Press, New York. Sharpe, R. M., and Skakkebaek, N. E. (1993). Are oestrogens involved in falling sperm counts and disorders of the male reproductive tract? Lancet 341, 1392–1395. Sharpe, R. M., Fisher, J. S., Millar, M. M., Jobling, S., and Sumpter, J. P. (1995). Gestational and lactational exposure of rats to xenoestrogens results in reduced testicular size and sperm production. Environ. Health Perspect. 103, 1136 –1143. Soto, A. M., Justica, H., Wray, J. W., and Sonnenschein, C. (1991). p-Nonylphenol: An estrogenic xenobiotic released from ‘‘modified’’ polystyrene. Environ. Health Perspect. 92, 167–173. Stone, R. (1994). Environmental estrogens stir debate: News and Comment. Science 256, 308 –310. Suominen, J., and Vierula, M. (1993). Semen quality of Finnish men. Br. Med. J. 306, 1579. Tsai-Morris, C. H., Knox, G., Luna, S., and Dufau, M. L. (1986). Acquisition of estradiol-mediated regulatory mechanism of steroidogenesis in cultured fetal rat Leydig cells. J. Biol. Chem. 261, 3471–3474. Toppari, J., Larsen, J. C., Christiansen, P., Giwercman, A., Grandjean, P., Guillette, L. J., Jr., Jegou, B., Jensen, T. K., Jouannet, P., Keiding, N., Leffers, H., McLachlan, J. A., Meyer, O., Muller, J., Rajpert-De Meyts, E., Scheike, T., Sharpe, R., Sumpter, J., and Skakkebaek, N., E. (1995). Male Reproductive Health and Environmental Chemicals with Estrogenic Effects, Miljoprojekt nr. 290. Report of the Ministry of Environment and Energy, Danish Environmental Protection Agency, Copenhagen, Denmark. Weibel, E. R., and Paumgartner, D. (1978). Integrated sterological and biochemical studies on hepatocytic membranes. II. Correction of section thickness effect on volume and surface density estimates. J. Cell Biol. 77, 584 –597. Wilcox, A. J., Baird, D. D., Weinberg, C. R., Hornsby, P. P., and Herbst, A. L. (1995). Fertility in men exposed prenatally to diethylstilbestrol. N. Engl. J. Med. 332, 1411–1416. Wing, T. Y., and Christensen, A. K. (1982). Morphometric studies on rat seminiferous tubules. Am. J. Anat. 165, 13–25. Wolff, M. S., Toniolo, P. G., Lee, 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.