The Effects of Atrazine on Female Wistar Rats: An Evaluation of the Protocol for Assessing Pubertal Development and Thyroid Function

Susan C. Laws*,1, Janet M. Ferrell*, Tammy E. Stoker{dagger}, Judith Schmid{ddagger} and Ralph L. Cooper*

* Endocrinology Branch, {dagger} Gamete and Early Embryo Biology Branch, and {ddagger} Biostatistics and Research Support Staff, Reproductive Toxicology Division, National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711

Received May 15, 2000; accepted September 13, 2000


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The effects of atrazine (ATR), a chlorotriazine herbicide, on the onset of puberty were evaluated in Wistar rats. Female rats were dosed by oral gavage from postnatal day(s) (PND) 22 through PND 41 with 0, 12.5, 25, 50, 100, or 200 mg ATR/kg. Vaginal opening (VO) was significantly delayed 3.4, 4.5, or greater than 6.8 days by 50, 100, and 200 mg/kg, respectively. VO had not occurred in 4 of 15 females in the 200 mg/kg group by the time of necropsies (PND 41). Body weight (bw) at necropsy was reduced in the 200 mg/kg group by 11.6%, but was not different from the control (0) in the 50 and 100 mg/kg groups. To examine the influence of reduced bw on pubertal development, a group of pair-fed controls was included whose daily food intake was dependent upon the amount consumed by their counterpart in the 200 mg/kg group. Although necropsy bw was reduced to the same extent as the ATR females, VO in the pair-fed controls was not significantly delayed. Adrenal, kidney, pituitary, ovary, and uterine weights were reduced by 200 mg/kg ATR. Serum T3, T4, and TSH were unaltered by ATR, which was consistent with no histopathologic/morphologic changes in the thyroid. Estrous cyclicity was monitored in a second group of females from VO to PND 149. The number of females displaying regular 4- or 5-day estrous cycles during the first 15-day interval after VO was lower in the 100 and 200 mg/kg ATR and pair-fed controls. Irregular cycles were characterized by extended periods of diestrus. By the end of the second 15-day interval (PND 57–71), no effects on estrous cyclicity were observed. These data show that ATR can delay the onset of puberty and alter estrous cyclicity in the female Wistar rat ( NOAEL of 25 mg/kg). Reduced food consumption and bw did not account for the delay in VO, because this effect was not observed in the pair-fed controls. In addition, the effect on estrous cyclicity was observed in the 100 mg/kg ATR group where no significant reduction in bw was observed.

Key Words: female reproductive toxicology; puberty; rat; atrazine; vaginal opening; estrous cyclicity.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
To address emerging issues concerning the effects of environmental chemicals on human health and wildlife, the U.S. Environmental Protection Agency (U.S. EPA) was given a mandate by Congress to develop and initiate a screening program to identify chemicals that could disrupt endocrine function (Public Law 104–170 and 104–182, August 1996). Working toward this goal, the U.S. EPA submitted an Endocrine Disruptor Screening Program: Proposed Statement of Policy (Federal Register, 1998) based largely upon recommendations by the Endocrine Disruptor Screening and Testing Advisory Committee (EDSTAC; U.S. EPA, 1998a). The Tier I Screening Battery includes a series of in vivo and in vitro research protocols designed to detect chemicals that alter the estrogen, androgen, or thyroid systems in humans, fish, and wildlife.

An integral part of the Tier I Screening Program is focused upon detecting chemicals that alter male and female pubertal development and thyroid function through steroid-mediated mechanisms of action. The Protocol for the Assessment of Pubertal Development and Thyroid Function in Juvenile Female Rats (Goldman et al., 2000Go; U.S. EPA, 1998b) is currently undergoing testing to evaluate its reliability and reproducibility as such a screen. Marty et al. (1999) evaluated the Female Pubertal Protocol using estrogen and selected pharmaceuticals that are known to alter thyroid function or steroid biosynthesis. Goldman et al. (2000) recently reviewed the physiological and biochemical changes that occur during sexual maturation in the female rat, and also the influence of peripubertal exposures to endocrine disrupting chemicals. While the studies cited in that review used endpoints and dosing paradigms that were chronologically similar to the Female Pubertal Protocol, few investigators have used the protocol to evaluate environmental chemicals. In this study, we report the effects of atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine) on female pubertal development and thyroid function, using the dosing regimen and required endpoints described in the protocol (Goldman et al., 2000Go). Some of the optional endpoints in the protocol were also included in the study, to address specific issues related to the proposed mode of action for atrazine.

Atrazine, a chorotriazine herbicide, is used extensively in the United States. Degradation by-products, as well as the parent compound, have been detected in surface and ground water in areas of major usage (Baker, 1998; Rawn et al., 1998Go). Previous studies have shown that atrazine has adverse effects on the reproductive system in mammals (Stevens et al., 1994Go; Wetzel et al., 1994Go). Eldridge et al. (1998, 1999a, 1999b) reported an earlier onset of mammary tumors in Sprague-Dawley rats following long-term oral exposure. The premature appearance of persistent estrus in these animals suggested that atrazine may have induced early reproductive senescence. Cooper et al. (1996) reported that atrazine (75–300 mg/kg, orally) disrupts estrous cyclicity in adult Long Evans and Sprague-Dawley rats during a 21-day exposure. These authors suggested that the effects on estrous cyclicity were most likely mediated via alterations in the neurotransmitter and hormonal control of gonadal function. Specifically, atrazine has been reported to increase dopamine and reduce norepinephrine concentrations in the hypothalamus (Cooper et al., 1998Go), and to diminish the estrogen-induced surge of luteinizing hormone and prolactin in ovariectomized rats following single or multiple (3 and 21 days) doses of atrazine (Cooper et al., 2000Go). In addition, Stoker et al. (1999) showed that suckling-induced prolactin release in lactating Wistar females is inhibited by atrazine and that male pups nursed by dams treated with 12.5 or 25 mg/kg atrazine twice per day during PND 1–9 exhibited an increased incidence of inflammation of the prostate at PND 120.

In the study reported here, we continue to explore the effects of atrazine on pituitary and ovarian function by evaluating the effect of the chemical on female pubertal development. Since atrazine has been reported to alter the release of prolactin and luteinizing hormone from the pituitary, we hypothesized that this compound would alter the onset of puberty. In addition, these studies evaluate the usefulness of the required and selected optional endpoints included in the female pubertal protocol as a screen for chemicals that alter hypothalamic-pituitary function.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Animals.
Wistar rats (14-day timed pregnant) were obtained from Charles Rivers Laboratories, Raleigh, NC, and were maintained under controlled temperature (20–24°C), humidity (40–50%) and light (14 h light/10 h dark) conditions with Purina Laboratory Rat Chow (5001) and water available ad libitum. Pregnant dams were allowed to deliver their pups naturally; 3 days postpartum (PND 3; PND 0 = the morning of birth) all litters were culled to 10 pups. Females were weaned on PND 21, ranked by body weight (bw) and litter, and placed into treatment groups such that the mean body weight ± SE for all groups were similar. In addition, littermates were equally distributed between the treatment groups. During the treatment and observation periods, animals were housed 1–2 per cage.

Dosing solutions and procedures.
Atrazine (97.1% purity; a gift from Novartis Crop Protection, Inc., Greensboro, NC) was prepared as a suspension in 1.0% methyl cellulose (M-7140, Lot No. 64H0619, Sigma Chemical Co., St. Louis, MO) in distilled water. Dose groups included 0 (vehicle), 12.5, 25, 50, 100, and 200 mg atrazine/kg bw, which were delivered in a volume of 5.0 ml dosing solution/kg bw. All doses were administered by oral gavage.

Experimental design.
The study was conducted in 2 blocks. Eight treatment groups were included in the first block. The females in 4 of the treatment groups received either 0 (vehicle), 50, 100, or 200 mg/kg (n = 7/treatment group) and were killed on PND 41. Females in the other 4 treatment groups received 0, 50, 100, or 200 mg/kg (n = 7/treatment group) and were evaluated for changes in estrous cyclicity from PND 42–149. These same treatment groups were used in a second block (n = 8/treatment group), along with 2 lower dose groups (e.g., 12.5 and 25 mg/kg; n = 15/treatment group; all animals killed on PND 41) to determine the no observed adverse effect level (NOAEL). An additional control group (e.g., pair-fed control; n = 8, killed on PND 41; n = 7, were evaluated for changes in estrous cyclicity from PND 42–149), and this group was also included in the second block to evaluate the effects of restricted food intake and lower body weight on the reproductive endpoints. In this group, the amount of food made available for each pair-fed control was dependent upon the amount of food consumed by its respective counterpart in the 200 mg/kg atrazine group on the previous day. Pair-fed controls also received vehicle.

All animals were dosed daily by oral gavage, beginning on PND 22 and continuing through PND 41 (Fig. 1Go). Body weights were recorded daily and the dose administered each day was adjusted for body weight. For those females killed on PND 41, liver, kidney, adrenal, ovary, uterus, and pituitary weights were recorded. In addition, serum was frozen for triiodothyronine (T3), thyroxin (T4), thyroid stimulating hormone (TSH), and prolactin (PRL). Pituitaries were frozen for PRL assays. Estrous cyclicity was evaluated for the remaining females from PND 42–149.



View larger version (20K):
[in this window]
[in a new window]
 
FIG. 1. Overview of the study protocol. Female rats were weaned on postnatal day (PND) 21, ranked by body weight and litter, and randomly assigned to treatment groups such that the mean body weight ± SE was approximately equal for all groups and litter mates were equally dispersed among the treatment groups. During the treatment period (i.e., PND 22–41), the animals were evaluated daily for vaginal opening (VO). On PND 41, one half of the females were killed (n = 15/treatment group) and serum and tissues collected. Estrous cyclicity was monitored in the remaining females (n = 15/treatment group) from VO through the posttreatment period (PND 42–149).

 
Age at vaginal opening and estrous cyclicity.
Beginning on PND 22, all females in the study were evaluated daily for vaginal opening. The day of complete vaginal opening and body weight on that day were recorded. For those animals where vaginal opening failed to occur prior to necropsy, PND 42 was used as the age at vaginal opening, to determine a mean for each treatment group. Beginning on the day of vaginal opening, daily vaginal smears were collected and evaluated in the 0, 50, 100, and 200 ATR and the pair-fed control groups. Vaginal smears were observed under a low-power light microscope for the presence of leukocytes, nucleated epithelial cells, or cornified epithelial cells, to determine the age of the first complete estrous cycle after vaginal opening and/or any longer-term effects on estrous cyclicity. The vaginal smears were classified as diestrus (presence of leukocytes), proestrus (presence of nucleated epithelial cells), or estrus (presence of cornified epithelial cells) as characterized by Everett (1989). Extended estrus was defined as exhibiting cornified cells with no leukocytes for 3 or more days and extended diestrus as the presence of leukocytes for 4 or more days (Cooper and Goldman, 1999Go).

Histology.
Immediately upon necropsy on PND 41, the uterus, ovaries, and thyroid were placed in formalin for 24 h. The tissues were rinsed and stored in 70% alcohol until embedded in paraffin, sectioned, and stained with hematoxylin and eosin. Tissues from the pair-fed control, 0 (vehicle) and atrazine 200 mg/kg groups were evaluated by Experimental Pathology Laboratory, Inc. (Research Triangle Park, NC) for pathologic abnormalities and potential treatment-related effects.

Radioimmunoassays.
Serum and pituitary prolactin concentrations and serum thyroid stimulating hormone (TSH) were measured by radioimmunoassay using material supplied by the National Hormone and Pituitary Agency. The iodination preparation, reference preparation, and antisera for PRL and TSH were as follows, respectively: I-6, I-9, RP-3, RP-3, S-9, and S-6. The iodination preparation was radiolabeled with 125I (Dupont/New England Nuclear) by a modification of the chloramine-T methods (Greenwood et al., 1963Go). Labeled PRL and TSH were separated from free iodide by gel filtration chromatography, and the radioimmunassays were conducted as described by Goldman et al. (1986).

Totals of T3 and T4 were measured using coat-a-count radioimmunoassay kits obtained from Diagnostic Products Corp. (Los Angeles, CA). The detection limits for T3 and T4 were 0.2 and 10 ng/ml, respectively.

Statistical analyses.
Data were evaluated for differences between the replicate studies (e.g., Blocks 1 and 2) and treatment effects by analysis of variance (ANOVA) using the General Linear Model (GLM) procedure (Statistical Analysis System (SAS), SAS Institute, Inc. Cary, NC). Since no significant differences between Block 1 and 2 were observed, data from the replicates were pooled prior to testing for treatment effects. Body weight and vaginal opening data from those females killed on PND 41 were analyzed separately from data collected for females used for the estrous cyclicity observations, in order to maintain a sample size of 15 as recommended in the Female Pubertal Protocol (Goldman et al., 2000Go). When a significant treatment effect ({alpha} = 0.05) was present, Dunnett's test (control vs. each treatment group) or Bonferroni t-tests (multiple comparisons of treatment groups) was used to compare treatment groups. Bartlett's test (GraphPad InStat, GraphPad Software, San Diego, CA) was used to test for homogeneity of variance, and where heterogeneity of variance was evident, the Welch t-test or Kruskal-Wallis Nonparametric Test with Dunn's Multiple Comparison Test were used. Organ weights were analyzed by analysis of covariance (ANCOVA) using the body weight at necropsy as a covariate. Means and adjusted means relative to necropsy body weight were calculated for liver, kidney, adrenal, and pituitary weights. Adjusted means were compared with the control, using a pairwise t-test with the Bonferroni correction. All data are reported as mean ± SE (n).


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Body Weight
Figure 2Go shows the body weight data from females killed on PND 41. Although the mean body weights ± SE were similar for all treatment groups on the first day of treatment (Fig. 2AGo), animals receiving the highest dose of atrazine (200 mg/kg bw) showed an 11.6% reduction in body weight on the last day of treatment (PND 41), when compared with the vehicle control (Fig. 2BGo). The reduction in body weight on PND 41 was not a result of a major weight loss at any one time during the treatment period, but rather the failure of these animals to gain weight at the same rate as the controls. Weight gains during the treatment period are shown in Figure 2CGo for all treatment groups. Weight gain in the 200 mg/kg atrazine group (81.1 ± 1.9 g [n = 15]) was significantly lower compared with the controls (97.1 ± 2.0 g [n = 15]).



View larger version (36K):
[in this window]
[in a new window]
 
FIG. 2. Changes in body weight during the 20-day exposure period. (A) Body weight (g) on the first day of treatment (PND 22); (B) body weight (g) on the last day of treatment (PND 41); (C) Body weight gain (g) during treatment from PND 22–41. Data are presented as mean ± SE (n = 15). *Significant treatment effect by General Linear Model (GLM) and significantly different from control (0) by Dunnett's Multiple Comparison Test (p < 0.05).

 
Similar changes in body weights were also observed in those females used for estrous cyclicity observations. Again, the mean body weight on PND 41 and body weight gain from PND 22 to 41 were significantly lower in the atrazine 200 mg/kg group. In this set of females, the body weights at PND 22 and PND 41 were 47.6 ± 1.5 g (n = 15) and 141.6 ± 3.0 g (n = 15), respectively, for the vehicle control; and 46.6 ± 1.3 g (n = 15) and 125.7 ± 2.8 g (n = 15) for the atrazine 200 mg/kg group.

Age and Body Weight at Vaginal Opening
The age of vaginal opening was significantly delayed following exposure to 50, 100, or 200 mg/kg atrazine, but was unaltered following exposure to 12.5 or 25 mg/kg (Fig. 3AGo). While vaginal opening occurred in the controls by 32.5 ± 0.51 days (n = 15), the age at vaginal opening was delayed by 3.4, 4.5, or greater than 6.8 days for females in the 50, 100, or 200 mg/kg treatment groups, respectively. In addition, vaginal opening did not occur within the treatment period in 4 of 15 females in the 200 mg/kg group. Body weights at the age of vaginal opening are shown in Figure 3BGo and were significantly increased in the 50, 100, and 200 mg/kg treatment groups as compared with the control (103 ± 2.9 g [n = 15]).



View larger version (44K):
[in this window]
[in a new window]
 
FIG. 3. (A) Age (days) at the time of vaginal opening. Data are presented as mean ± SE (n = 15). Ratio within each box is the number of females that reached vaginal opening by PND 41 as compared with the total number of females in each treatment group. (B) Body weight (g) at the time of vaginal opening. Data are presented as mean ± SE (n = 15). *Significant treatment effect by General Linear Model (GLM) and significantly different from control (0) by Dunnett's Multiple Comparison Test (p < 0.05).

 
Age at vaginal opening was also delayed in the group of females evaluated for changes in estrous cyclicity. In this group of females, vaginal opening occurred in the vehicle control by 32.8 ± 0.5 days (n = 15). The ages of vaginal opening in the 50, 100, and 200 mg/kg groups were 34.3 ± 0.5 days (n = 15), 36.3 ± 0.5 days (n = 15), and 40.8 ± 0.5 days (n = 15), respectively. Eight of 15 females in the highest atrazine group again failed to reach vaginal opening by PND 41. However, once treatment was terminated on PND 41, vaginal opening occurred within 3–4 days in these females. Body weights on the day of vaginal opening were also significantly increased by atrazine treatment in this group of females: 103 ± 2.8 g (n = 15) for the control as compared with 110 ± 2.9 g (n = 15), 114 ± 3.0 g (n = 15), and 124.2 ± 3.0 g (n = 15) for the atrazine 50, 100, and 200 mg/kg groups, respectively.

Body Weight Gain and Pubertal Development in the Pair-Fed Group
Figure 4Go shows the mean daily food consumption by females in the atrazine 200 mg/kg group (n = 15) as compared with controls (n = 10) that were allowed to eat ad libitum from PND 22 to 41. Food intake was significantly reduced in the atrazine group during the treatment period. To evaluate the effects of decreased food consumption and lower body weight gain during the treatment period on pubertal development, a pair-fed control group was included in Block 2 of the study. Food intake for the group`s females was dependent upon the amount of food consumed each day by their respective mates in the atrazine 200 mg/kg group. Body weights on PND 33 (e.g., the age at vaginal opening for the ad libitum control) and PND 41, as well as, the body weight gain during the treatment period were similar in the pair-fed control and atrazine 200 mg/kg groups (Table 1Go). However, these parameters were significantly reduced in the pair-fed control and atrazine 200 mg/kg groups when compared with the vehicle controls fed ad libitum. Lower body weights on PND 33 and body weight gain during the treatment period did not significantly delay the age or body weight at vaginal opening in the pair-fed control group (34.4 ± 0.5 days [n = 15]) as compared with the ad libitum controls (32.8 ± 0.4 days [n = 15]). In contrast, both age and body weight at vaginal opening were significantly increased in the atrazine 200 mg/kg group as compared with the pair-fed control group.



View larger version (14K):
[in this window]
[in a new window]
 
FIG. 4. Mean daily food consumption in control and atrazine 200 mg/kg females. Daily ad libitum food consumption (g) was monitored from PND 22 to 41 for the 0 and 200 mg/kg atrazine females. Data are presented as mean ± SE (n = 10 for Control (0); n = 15 for 200 mg/kg atrazine). Food intake was significantly reduced in the atrazine group. Each pair-fed control female in the study received the same amount of food as her counterpart in the 200 mg/kg atrazine group.

 

View this table:
[in this window]
[in a new window]
 
TABLE 1 Comparison of Body Weight Gain and Pubertal Development in 0- and 200 mg/kg Atrazine and Pair-Fed Control Females
 
Tissue Weights
Tissue weights at necropsy are shown in Tables 2 and 3GoGo for the atrazine dose response. Data from Blocks 1 and 2 were combined as no significant block difference was detected for any of the parameters. Since there was a reduction in necropsy body weight in the highest atrazine treatment group, organ weights were analyzed with and without the body weight at necropsy as a covariate. Results from both analyses for liver, kidney, adrenal, and pituitary weights are shown in Table 2Go. These weights were significantly reduced in the 200 mg/kg atrazine group when evaluated by ANOVA and the Bonferroni t-test for multiple comparison. Using ANCOVA, the adjusted means for these organ weights reflected a significant increase in liver weight relative to necropsy body weight in the atrazine 200 mg/kg group, as well as a reduction in kidney weight at 200 mg/kg. The adjusted mean for pituitary weight was also significantly lower in the atrazine 12.5, 100, and 200 mg/kg groups as compared with the control.


View this table:
[in this window]
[in a new window]
 
TABLE 2 Body Weight and Organ Weights at Necropsy on PND 41
 

View this table:
[in this window]
[in a new window]
 
TABLE 3 Reproductive Organ Weights and Hormone Concentrations at Necropsy on PND 41
 
Weights of the reproductive organs and serum hormone concentrations are reported in Table 3Go. Since the endocrine status of the females at the time of necropsy affects the weights of the reproductive organs and prolactin concentrations, the number of females that were in diestrus, proestrus/estrus, or not cycling at the time was documented for each treatment group. The number of females not cycling at the time of necropsy was substantially higher in the 200 mg/kg atrazine group, and ovarian and uterine (+ fluid) weights were significantly reduced in this group. Uterine weights without the fluid were significantly decreased in both the 100 and 200 mg/kg groups.

Tissue weights in the 0 and 200 mg/kg atrazine groups and the pair-fed control groups are shown in Table 4Go. To keep sample size equal with the pair-fed control (n = 8), only data for the 0 (control) and 200 mg/kg atrazine groups from Block 2 were used for this comparison. While kidney and pituitary weights were significantly reduced in the pair-fed control females as compared with the 0 controls, these weights remained significantly higher than those of the 200 mg/kg atrazine animals. Ovarian and uterine weights were not significantly reduced in the pair-fed females as compared with the controls. However, ovarian and uterine weights (with and without fluid) in the 200 mg/kg atrazine group were significantly lower than those in the pair-fed control group.


View this table:
[in this window]
[in a new window]
 
TABLE 4 Comparison of Organ Weights and Hormone Data in 0 and 200 mg/kg Atrazine and Pair-Fed Controls on PND 41
 
Hormone Assays
Pituitary and serum prolactin concentrations are shown in Tables 3 and 4GoGo. Prolactin levels normally vary during the estrous cycle of the female rat and the data reflect these fluctuations. For example, the mean serum prolactin concentrations in the combined control and atrazine 12.5 mg/kg groups ranged from 4.4 ± 0.9 (n = 17) ng/ml on diestrus, to 80.3 ± 8.2 (n = 4) on proestrus and 60.3 ± 21.3 (n = 5) on estrus. Pituitary prolactin concentrations for diestrus, proestrus, and estrus ranged from 955 ± 79 ng/mg pituitary (n = 17), 658 ± 117 (n = 4), and 963 ± 275 (n = 5), respectively. Thus, when the data were analyzed by ANOVA, there was a significant effect on serum and pituitary prolactin based upon the day of the estrous cycle on which the females were killed, but no significant atrazine treatment effect. Since the number of females within a treatment group was not adequate for analyzing the data based upon endocrine status, the means reported in Tables 3 and 4GoGo contain all females in the treatment group.

Hormones associated with thyroid function were not significantly altered with atrazine treatment. Serum levels of T3, T4, and TSH were not significantly different from the control (Tables 3 and 4GoGo).

Histology
The thyroids from the 0 and 200 mg/kg atrazine groups and the pair-fed controls were evaluated for morphologic changes such as altered follicular epithelial height, the relative abundance and tinctorial characteristics of colloid, the extent of the thyroid vascular supply, and the density, size, and shape of the thyroid follicles. No changes were observed in any of the treatment groups, nor were there any indications of a treatment effect on thyroid hormone secretion.

No pathologic changes of any kind were observed in uterine or ovarian tissues from the control or pair-fed control groups. A decrease in corpus luteum development was observed in the 200 mg/kg atrazine group. This decrease was generally associated with underdevelopment of the reproductive tract on PND 41 in comparison with the controls. In addition, although the average number of atretic follicles (e.g., contained more than 1–2 cells' worth of karyorrhectic nuclear debris within the follicular antrum or lining cells) per ovary was 1.6-fold higher in the 200 mg/kg group, this increase was not statistically significant (p = 0.07). Uterine hypoplasia (e.g., characterized by one or more of the following, decreased uterine horn diameter, decreased myometrial development, decreased or absent endometrial glands, and/or immature endometrial stroma), was observed in 60% of the 200 mg/kg females and was restricted to those females that demonstrated a marked decrease in corpus luteum development.

Estrous Cyclicity
Estrous cyclicity was monitored in the 0, 50, 100, and 200 mg/kg and pair-fed control groups from the day of vaginal opening through PND 149. The age of the first complete 4–5 day estrous cycle after vaginal opening was significantly delayed in the 100 and 200 mg/kg atrazine groups as compared with the control (Table 5Go). A slight delay was also noted during this period for the pair-fed group, but the delay was not significantly different from the control.


View this table:
[in this window]
[in a new window]
 
TABLE 5 Summary of Estrous Cyclicity Data from Vaginal Opening (VO) through PND 149
 
Estrous cyclicity was initially evaluated during the 15-day period following vaginal opening. The number of regular 4- or 5-day cycles during the first 15 days after vaginal opening was reduced for the pair-fed controls, as well as for the 100 and 200 mg/kg atrazine groups. All control (0) females displayed between 2 and 3 regular 4- or 5-day cycles during the 15-day observation period as compared with 46.6% of the 100 or 200 mg/kg atrazine groups and 71.4% of the pair-fed control females. Most of the abnormal cycles observed in the 100and 200 mg/kg atrazine groups during the first 15 days after vaginal opening were characterized by an increase in the number of days of diestrus. A similar response was also observed in the pair-fed control group. This effect was temporal since the differences in estrous cyclicity were no longer evident by PND 57–71 in the pair-fed control or atrazine groups.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The stated purpose of the female pubertal protocol was to identify chemicals that alter receptor function or the synthesis of estrogen, androgen, and thyroid hormones, or that disrupt the hypothalamic-pituitary control of ovarian function. Results from the study reported here clearly demonstrate that atrazine delays puberty in Wistar female rats with a LOAEL (lowest observable adverse effect level) of 50 mg/kg. This dose is greater than that found to induce the premature development of mammary gland tumors (400 ppm or approximately 22.5 mg/kg/day) in chronic feeding studies (Eldridge et al., 1994bGo; Wetzel et al., 1994Go), or the dose (12.5 mg/kg bw, orally, twice per day) found to inhibit suckling-induced prolactin in lactating females during only 4 days of treatment and alter reproductive function in the male offspring at PND 120 (Stoker et al., 1999Go). Also, 50 mg/kg for 3 days has been reported to delay the timing of the estrogen-induced LH and prolactin surges in adult, ovariectomized females (Cooper et al., 2000Go). Thus, the alterations in female pubertal development appear to occur at concentrations that are similar to those doses reported to affect reproductive tissues in other studies.

Data from this study provide a means to evaluate the usefulness of both the required and some of the optional endpoints included in the Female Pubertal Protocol. Of the required endpoints, the age and weight at vaginal opening were the first and perhaps the best indicators of atrazine's effect on pubertal development. Changes in body weight during treatment were also informative since the reduction in body weight in the highest-atrazine treatment group needed to be taken into account when evaluating the data. In this study, the issue of treatment-associated changes in body weight was addressed by using necropsy body weight as a covariate when analyzing tissue weights and by including a pair-fed control group in the second block of the study. The ability to evaluate hypothalamic-ovarian function by monitoring vaginal cytology during the treatment period (PND 22–41) was somewhat limited in this study. This is because there is an increased tendency for irregular cycles in all females at the onset of puberty and also an inability of many of the females in the highest-atrazine-treatment group to reach vaginal opening. By extending the observation period for monitoring estrous cyclicity in the study reported here, a more thorough assessment of cyclicity was conducted. The protocol also allowed for the detection of treatment-related changes in kidney, pituitary, adrenal, ovarian, and uterine weights. However, due to the changes associated with different stages of the estrous cycle, the usefulness of uterine weight and serum prolactin was minimal. Although serum T3, T4, and TSH levels and thyroid histology were unaffected by atrazine, these endpoints are valid and add strength to the screening aspect of the protocol.

The effect of endocrine status at necropsy is a factor that should be considered while evaluating tissue weights and hormone concentrations when using this protocol. Changes in hormone concentrations vary with circadian rhythm as well as during the estrous cycle (Esber et al., 1976Go; Goldman et al., 2000Go). Uterine and ovarian weights can vary depending upon the endocrine milieu and endocrine state (e.g., cycling, pseudopregnant, or anestrous) in postpubertal females. Goldman et al. (2000) have discussed these technical issues at length and suggest that these data be blocked according to the particular day of the estrous cycle that each animal was killed. Here, we included the number of animals within each treatment group that displayed vaginal smears indicative of diestrus, proestrus, or estrus. However, the sample size of 15–16 animals used in the study was not sufficient for analyzing the data by endocrine status. The observation of lower uterine and ovarian weights in the 200 mg/kg atrazine group at PND 41 in this study was most likely due to the immaturity of the endocrine system. In addition, since the variation in the serum and pituitary prolactin concentrations was greatly inflated by the fact that the females were killed on varying days of the estrous cycle, no significant treatment effect could be observed for this measure.

A primary concern with the Female Pubertal Protocol, as well as for any in vivo test of toxicity, is that systemic toxicity may play a key role in any observed effect. For the most part, this may be associated with a decrease in body weight during treatment. For this reason, we included a group of control females that were pair-fed to match the food consumption and reduction in body weight observed in the females dosed with the highest concentration of atrazine. This was especially important when evaluating the results from the Female Pubertal Protocol, since it is known that lower body weight during the peripubertal period, in both humans and laboratory animals, will delay the onset of puberty and reduce fertility in the adult (Goldman et al., 2000Go). While it is generally recognized that there is a threshold for body weight and percent body fat that must be achieved prior to the onset of puberty (Bronson, 1987Go; Holeham and Merry, 1985Go; Wilen and Naftolin, 1978Go), the age of vaginal opening in the females observed in this study was not totally dependent upon body weight. For example, some of the animals with lower body weights reached vaginal opening before others with greater body weights. However, it is interesting to note that the average body weight of the pair-fed control females at vaginal opening was only 5% lower than the control, and the age of vaginal opening was not significantly different from the control. Conversely, the average body weight at vaginal opening for the 200 mg/kg atrazine females was approximately 10 and 14% higher than the control or pair-fed groups, respectively.

Another important observation was that estrous cyclicity was altered in the pair-fed control during the first 15-day interval following vaginal opening. Again, it is well documented that reduced body weight/food consumption and stress can disrupt estrous cyclicity (Bates et al., 1982Go; Warren et al., 1999Go). Although the irregularities of the estrous cycles during this observation period are consistent with data reported by Cooper et al. (1996) and Eldridge et al. (1994a), where single or multiple doses of atrazine disrupt estrous cyclicity, the abnormal cycles observed in the 200 mg/kg group in this study cannot be totally dissociated from a possible correlation with reduced body weight and/or stress. However, the fact that cyclicity was also irregular in the100 mg/kg atrazine females with no reduction in body weight, lends support to the hypothesis that atrazine alone can disrupt estrous cyclicity.

The observation period for estrous cyclicity was extended in this study to evaluate posttreatment effects of atrazine on pituitary and ovarian function. No treatment differences were detected in the number of regularly cycling animals from PND 57–149. However, since Eldridge et al (1999) have observed early reproductive senescence and increased incidences of mammary tumors in rats, following long-term exposure to atrazine, it would have been interesting to have monitored the pubertal females through reproductive aging. Such a precedence for effects becoming apparent later in life has been set by the increase in prostate inflammation in adult male rats following lactational exposure to atrazine (Stoker et al., 1999Go).

The identification of the particular mode of action is not a stated purpose of either the male or female pubertal protocols currently under consideration for the Tier-1 Screening Program (U.S. EPA, 1998b). Nevertheless, by evaluating the changes observed in the present study in light of previously published data, some insight into the primary mode of action for atrazine's ability to delay puberty may be gained. Current evidence suggests that atrazine may work through one or several of the endocrine-disrupting mechanisms targeted by the protocol. For example, although this herbicide does not bind to the estrogen receptor directly nor possess any estrogenic activity in vivo (Conner et al., 1996; Eldridge et al., 1994bGo; Tennant et al., 1994aGo), it has been reported to inhibit estrogen-induced 3H-thymidine incorporation and uterine growth in vivo (Tennant et al., 1994bGo). This observation is consistent with the pubertal delay in the present study. However, atrazine can suppress the pulsatile release of LH in long-term, ovariectomized females, an effect that occurs in the absence of endogenous or exogenous estrogen (Cooper et al., 1996Go). The results of in vitro studies evaluating the effect of atrazine on the estrogen receptor are also mixed. Conner et al. (1996) reported no estrogenic or anti-estrogenic activity of atrazine in several cell lines. In contrast, Tran et al. (1996) reported that atrazine, simazine, and cyanazine all inhibited an estrogen-induced response in yeast cells. However, these data could not be replicated by Graumann et al. (1999) who reported no inhibitive effect of atrazines on 17ß-estradiol-mediated transactivation in yeast. Thus, these observations lead us to conclude that it is unlikely that atrazine alters reproductive development via an estrogen receptor-dependent change.

Importantly, there is a growing line of evidence that atrazine may increase the activity of aromatase, a key enzyme for the conversion of androgens to estrogens. Crain et al. (1997) have observed elevated aromatase activity in the gonadal-adrenal mesonephros in male alligator hatchlings exposed in ovo to atrazine. Sanderson et al. (2000) have recently reported that in vitro exposure to atrazine, simazine, and propazine induces aromatase activity (CYP19) and increases levels of CYP19 mRNA in H295R human adrenocortical carcinoma cells. Indeed, Stoker et al. (2000) report a reduction in intratesticular testosterone and increases in serum estrone and estradiol, following pubertal exposure to 200 mg/kg of atrazine. It is not known whether the changes in steroids in the pubertal males were due to a delay in the onset of pubertal development or an atrazine-induced change in aromatase activity. In the study reported here, estrogen and estrone were not measured because of the variability in hormone concentrations associated with the females being killed on different days of the estrous cycle. However, there was no evidence that atrazine elevated serum estrogen concentrations in the females in this study (e.g., no incidences of advanced vaginal opening, persistent cornified vaginal epithelial cells, or uterotrophic activity were observed). Thus, while an effect on steroidogenesis was not immediately apparent in this study, the issue of a possible change in aromatase activity in those tissues key to neuroendocrine development during puberty should be explored.

The results of this study also support the hypothesis that atrazine delays the onset of puberty by altering hypothalamic-pituitary activity. Atrazine reduces serum LH and prolactin secretion (Cooper et al., 1996Go, 2000Go; Simpkins et al., 1998Go). Both of these hormones are important for normal pubertal development. Although serum and pituitary prolactin were measured at necropsy for the females in this study, no significant differences were observed due to the variability associated with prolactin concentrations on different days of the estrous cycle. Reports that atrazine can reduce hypothalamic norepinephrine concentrations (Cooper et al., 1998Go) and that intravenous injections of GnRH restore the estrogen-induced secretion of LH in ovariectomized, atrazine-treated female rats (Cooper et al., 2000Go) provide additional evidence for a CNS–pituitary mode of action. It is well known that neurotransmitters and their regulation of pituitary hormone synthesis and secretion are critical for the onset of puberty and the maintenance of reproductive capability in the adult female (Goldman et al., 2000Go). Thus, because the onset of puberty is a transitional period during which there are changes in the signaling within the hypothalamic-pituitary-ovarian axis, the hypothesis that atrazine mediates its effects through the CNS and hormonal control of gonadal function warrants further investigation.

In summary, we report that oral exposure to atrazine from PND 22 through 41 delays vaginal opening in Wistar rats in a dose-dependent manner. The LOAEL for a delay in vaginal opening was 50 mg/kg. As compared with pair-fed control group, the reduced body weight in the 200 mg/kg atrazine group was not solely responsible for the delay in vaginal opening. Fewer regular 4- or 5-day estrous cycles were observed during the first 15-day period following vaginal opening in the 100- and 200 mg/kg atrazine groups as compared with the control, but no differences were detected in any treatment group between PND 57 and 149. While this study was not intended to identify a specific mechanism of action, these data are consistent with a possible effect on the CNS and hormonal control of gonadal function that warrant further investigation. In addition, this study documents the effectiveness of the Female Pubertal Protocol as a method for screening for endocrine-disrupting chemicals, which may not mediate their effect through the estrogen, androgen, or thyroid systems.


    ACKNOWLEDGMENTS
 
We gratefully acknowledge Judy McEachern, Rodney Daye, Debbie Crawford, Alvin Moore, Bette Terrill, Femi David-Yerumo and Henry Deas for their technical support and assistance with animal care; Dorothy Guidici and Keith McElroy for their technical assistance with the RIAs and necropsies; and Drs. Audrey Cummings, Parikshit Das, and Susan Makris for their reviews and helpful comments on earlier drafts of the manuscript.


    NOTES
 
This manuscript has been reviewed in accordance with the policy of the National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.

Portions of these data were presented at the 2000 Annual Meeting of the Society of Toxicology, Philadelphia, PA.

1 To whom correspondence should be addressed at MD-72, NHEERL, U.S. EPA, Research Triangle Park, NC 27711. Fax: 919 541-5138. E-mail: laws.susan{at}epa.gov. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Baker, D. B. (1988). Herbicides in drinking water: A challenge for risk communication. In Triazine Herbicides Risk Assessment (L. G. Ballantine, J. E. McFarland, and D. S. Hackett, Eds.), pp. 303–321. ACS Symposium Series. American Chemical Society, Washington, D.C.

Bates, G.W., Bates, S. R., and Whitworth, N. S. (1982). Reproductive failure in women who practice weight control. Fertil. Steril. 37, 373–378.[ISI][Medline]

Bronson, F. H. (1987). Puberty in female rats: Relative effect of exercise and food restriction. Am. J. Physiol. 252, R140–R144.[Abstract/Free Full Text]

Cooper, R. L., and Goldman, J. M. (1999). Vaginal cytology. In An Evaluation and Interpretation of Reproductive Endpoints for Human Health Risk Assessment. (G. Daston and C. Kimmel, Eds.), pp. 42–56. ILSI Press, Washington, DC.

Cooper, R. L., Stoker, T. E., Goldman, J. M., Parrish, M. B., and Tyrey, L. (1996). Effect of atrazine on ovarian function in the rat. Reprod. Toxicol. 10, 257–264.[ISI][Medline]

Cooper, R. L., Stoker, T. E., McElroy, W. K., and Hein, J. F. (1998). Atrazine (ATR) disrupts hypothalamic catecholamines and pituitary function. Toxicologist 42, 160.

Cooper, R. L., Stoker, T. E., Tyrey, L., Goldman, J. M., and McElroy, W. K. (2000). Atrazine disrupts the hypothalamic control of pituitary-ovarian function. Toxicol. Sci. 53, 297–307.[Abstract/Free Full Text]

Connor, K., Howell, J., Chen, I., Liu, H., Berhane, K., Sciarretta, C., Safe, S.O., and Zacharewski, T. (1996). Failure of chloro-S-traizine-derived compounds to induce estrogen receptor-mediated responses in vivo and in vitro. Fundam. Appl. Toxicol. 30, 93–101.[ISI][Medline]

Crain, D. A., Guillette, L., Jr., Rooney, A. A., and Pickford, D. B. (1997). Alterations in steroidogenesis in alligators (Alligator mississippiensis) exposed naturally and experimentally to environmental contaminants. Environ. Health Perspect. 105, 528–533.[ISI][Medline]

Eldridge, J. C., Fleenor-Heyser, D. G., Extrom, P. C., Wetzel, L. T., Breckenridge, C. B., Gillis, J. H., Luempert, L. G., III, and Stevens, J. T. (1994a). Short-term effects of chlorotriazines on estrus in female Sprague-Dawley and Fischer 344 rats. J. Toxicol. Environ. Health 43, 155–167.[ISI][Medline]

Eldridge, J. C., McConnell, R. F., Wetzel, L. T., and Tisdel, M. O. (1998). Appearance of mammary tumors in atrazine-treated female rats: Probable mode of action involving strain-related control of ovulation and estrous cycling. In Triazine Herbicides: Risk Assessment. (L.G. Ballantine, J.E. McFarland, and D.S. Hackett, Eds.), pp. 158–176. ACS Symposium Series. American Chemical Society, Washington, DC.

Eldridge J. C., Tennant, M. K., Wetzel, L. T., Breckenridge, C. B., and Steven, J. T. (1994b). Factors affecting mammary tumor incidence in chlorotriazine-treated female rats: Hormonal properties, dosage, and animal strain. Environ. Health Perspect. 102, 29–36.

Eldridge, J. C., Wetzel, L. T., Stevens, J. T., and Simpkins, J. W. (1999a). The mammary tumor response in triazine-treated female rats: A threshold-mediated interaction with strain and species-specific reproductive senescence. Steroids 64, 672–678.[ISI][Medline]

Eldridge, J. C., Wetzel, L. T., and Tyrey, L. (1999b). Estrous cycle patterns of Sprague-Dawley rats during acute and chronic atrazine administration. Reprod. Toxicol. 13, 491–499.[ISI][Medline]

Esber, H. J., Kuo, E., and Bogden, A. E. (1976). Circadian pattern of prolactin and growth hormone levels in female rats of two strains. Lab. Anim. Sci. 26, 186–189.[Medline]

Everett, J. W. (1989). Neurobiology of Reproduction in the Female Rat. Springer-Verlag, New York.

Goldman, J. M., Laws, S. C., Balchak, S. K., Cooper, R. L., and Kavlock, R. J. (2000). Endocrine-disrupting chemicals: Prepubertal exposures and effects on sexual maturation and thyroid activity in the female rat. A focus on the EDSTAC recommendations. Crit. Rev. Toxicol. 30, 135–196.

Goldman, J. M., Cooper, R. L., Rehnberg, G. L., Hein, J. F., McElroy, W. K., and Gray. L. E., Jr. (1986). Effects of low subchronic doses of methoxychlor on the rat hypothalamic-pituitary reproductive axis. Toxicol. Appl. Pharmacol. 86, 474–483.[ISI][Medline]

Graumann K, Breithofer A., and Jungbauer A. (1999). Monitoring of estrogen mimics by a recombinant yeast assay: Synergy between natural and synthetic compounds? Sci. Total Environ. 225, 69–79.[ISI][Medline]

Greenwood, F. C., Hunter, W. M., and Glover, J. S. (1963). The preparation of 131I-labeled human growth hormone of high specific activity. Biochem. J. 89, 114–123.[ISI]

Holeham, A. M., and Merry, B. J. (1985). The control of puberty in the dietary restricted female rat. Mech. Ageing Dev. 32, 179–191.[ISI][Medline]

Marty, M. S., Crissman, J. W., and Carney, E. W. (1999). Evaluation of the EDSTAC female pubertal assay in CD Rats using 17ß-estradiol, steroid biosynthesis inhibitors, and a thyroid inhibitor. Toxicol. Sci., 52, 269–277.[Abstract]

Rawn, D. F., Halldorson, T. H., and Muir, DC. (1998). Atmospheric transport and deposition: An additional input pathway for atrazine to surface waters. In Triazine Herbicides Risk Assessment. (L.G. Ballantine, J.E. McFarland, and D.S. Hackett, Eds.), pp. 158–176. ACS Symposium Series. American Chemical Society, Washington, DC.

Sanderson, J. T., Seinen, W., Giesy, J. P., and van den Berg, M. (2000). 2-Chloro-s-triazine herbicides induce aromatase (CYP19) activity in H295R human adrenocortical carcinoma cells: A novel mechanism for estrogenicity? Toxicol. Sci. 54, 121–127.[Abstract/Free Full Text]

Simpkins, J. W., Eldridge, J. C., and Wetzel, L. T. (1998). Role of strain-specific reproductive patterns in the appearance of mammary tumors in atrazine-treated rats. In Triazine Herbidcides Risk Assessment. (L.G. Ballantine, J.E. McFarland, and D.S. Hackett, Eds.), pp. 399–413. ACS Symposium Series. American Chemical Society, Washington, DC.

Stevens, J. T., Breckenridge, C. B., Wetzel, L. T., Gillis, J. H., Luempert, L. G., III, and Eldridge, J. C. (1994). Hypothesis for mammary tumorigenesis in Sprague-Dawley rats exposed to certain triazine herbicides. J. Toxicol. Environ. Health 43, 139–153.[ISI][Medline]

Stoker, T. E., Laws, S. C., Guidici, D., and Cooper, R. L. (2000). The effects of atrazine on puberty in male Wistar rats: An evaluation in the protocol for the assessment of pubertal development and thyroid function. Toxicol. Sci. 58, 50–59.[Abstract/Free Full Text]

Stoker, T. E., Robinette, C. L., and Cooper, R. L. (1999). Maternal exposure to atrazine during lactation suppresses suckling-induced prolactin release and results in prostatitis in the adult offspring. Toxicol. Sci. 52, 68–79.[Abstract]

Tennant, M. K., Hill, D. S., Eldridge, J. C., Wetzel, L. T., Breckenridge, C. B., and Steven, J. T. (1994a). Anti-estrogenic properties of chloro-s-triazines in rat uterus. J. Toxicol. Environ. Health 43, 183–196.[ISI][Medline]

Tennant, M. K., Hill, D. S., Eldridge, J. C., Wetzel, L. T., Breckenridge, C. B., and Steven, J. T. (1994b). Chloro-s-triazine antagonism of estrogen action: Limited interaction with estrogen receptor binding. J. Toxicol. Environ. Health 43, 197–211.[ISI][Medline]

Tran, D. Q., Kow, K. Y., McLachlan, J. A., and Arnold, S. F. (1996). The inhibition of estrogen receptor-mediated responses by chloro-s-triazine-derived compounds is dependent on estradiol concentration in yeast. Biochem. Biophys. Res. Commun. 227, 140–146.[ISI][Medline]

U.S. EPA (U.S. Environmental Protection Agency). (1998a). EDSTAC Final Report. Available at: http://www.epa.gov/scipoly/oscpendo/history/finalrpt.ht. Accessed on October 15, 2000.

U.S. Environmental Protection Agency (1998). Endocrine Disruptor Screening Program: Proposed Statement of Policy, 63(248). Federal Register:71541–71568. Available at: http://www.epa.gov/fedrgstr/EPA-TRI/1998/December/Day-28/tri34298.htm. Accessed on October 15, 2000.

U.S. EPA (U.S. Environmental Protection Agency). (1998b). EDSTAC Final Report, Chapter 5. Appendix L. "Protocols for Tier-I Screening Assays." Available at: http://www.epa.gov/scipoly/oscpendo/history/app-lv14.pdf. Accessed on October 15, 2000.

Warren, M. P., Voussoughian, F., Geer, E. B., Hyle, E. P. Adberg, C. L., and Ramos, R. H. (1999). Functional hypothalamic amenorrhea: Hypoleptinemia and disordered eating. J. Clin. Endocrinol. Metab. 84, 873–877.[Abstract/Free Full Text]

Wetzel, L. T., Luempert, L. G., III, Breckenridge, C. B., Tisdel, M. O., Stevens, J. T., Thakur, A. K., Extrom, P. J., and Eldridge, J. C. (1994). Chronic effects of atrazine on estrus and mammary tumor formation in female Sprague-Dawley and Fischer 344 rats. J. Toxicol. Environ. Health43, 169–182.[ISI][Medline]

Wilen, R., and Naftolin, F. (1978). Pubertal food intake and body length, weight, and composition in the feed-restricted female rat: Comparison with well fed animals. Pediatr. Res. 12, 263–267.[ISI][Medline]