Ethinyl estradiol is a potent synthetic estrogen widely used in pharmaceutical preparations. Its high potency and widespread use led to its selection by the National Toxicology Program for inclusion in studies to examine endocrine disrupting compounds with estrogenic activity, both because of its utility as a positive control to which weaker estrogens can be compared and because of potential human developmental exposures resulting from unintentional continuation of the use of oral contraceptives containing ethinyl estradiol during early pregnancy. Because of these concerns, ethinyl estradiol was selected as one of the compounds to be examined in a protocol utilizing Sprague-Dawley rats designed to evaluate the effects of short-term multigenerational and long-term exposures to doses of estrogenic agents that produce subtle reproductive tract lesions in developmentally exposed Sprague-Dawley rat pups (see Figure 1 of Overview). Results of short-term reproductive dose range-finding and mutigenerational reproductive toxicology studies are reported in this Technical Report, and results of the 2-year study are reported separately (NTP, 2010).
Reproductive Dose Range-Finding StudyA series of short-term studies with ethinyl estradiol was conducted with two goals: to obtain data necessary to establish exposure concentrations to be used in the subsequent multigenerational reproductive toxicology and chronic toxicity studies and to evaluate the effects of ethinyl estradiol on estrogen-sensitive endpoints outside the reproductive tract. Ethinyl estradiol was administered in a soy- and alfalfa-free diet at concentrations of 0, 0.1, 1, 5, 25, 100, or 200 ppb to pregnant Sprague-Dawley dams starting on gestation day 7 (GD 7) and continuing through pregnancy. These dietary exposure concentrations resulted in ingested doses of approximately 0.008, 0.08, 0.39, 1.77, 7.26, or 13.33 µg ethinyl estradiol/kg body weight per day to the dams. Dietary exposure of the dams continued through lactation, during which time ingested doses were approximately 0.03, 0.26, 1.37, 6.53, 29.68, or 51.93 µg/kg per day. Pups from five litters, culled to eight per litter with an equal sex distribution on postnatal day (PND) 2 were maintained on the same dosed feed as their mother after weaning until sacrifice at PND 50. Ingested doses were approximately 0.02, 0.22, 1.14, 5.48, 21.00, or 45.24 µg/kg per day for male pups and 0.02, 0.22, 1.18, 5.60, 22.92, or 45.87 µg/kg per day for female pups.
Daily body weights of pregnant dams showed a negative exposure concentration-related trend with significantly decreased body weights in the 100 and 200 ppb groups relative to the controls on GDs 12 to 21 and 10 to 21, respectively. Daily feed consumption was also decreased in the 100 and 200 ppb groups on multiple days in the early period of treatment (within the period from GDs 8 to 14). Overall body weight gain and feed consumption during pregnancy also showed significant negative trends and were significantly less than controls in the 100 and 200 ppb groups.
Mean live pup birth weight was significantly less than controls in the 100 and 200 ppb groups. Other pregnancy parameters (gestation duration, proportion of vaginal plug-positive dams producing litters) or litter data (total pups per litter, proportion of stillborn pups, sex ratio, anogenital distance) did not show significant exposure concentration-related effects. Preputial separation, a marker of male puberty, was accelerated in the 5 and 25 ppb groups relative to the controls; however, the proportion of male pups showing preputial separation in the 200 ppb group by the time of scheduled sacrifice at PND 50 was less than that in the control group. Vaginal opening, a marker of female puberty, was accelerated in the 25, 100, and 200 ppb groups relative to the control group. The mean body weights of 200 ppb males and females were significantly less than those of controls from PND 42 onward. Total body weight gain and feed consumption after weaning were not significantly altered by treatment for either sex. Organ weights were analyzed by three statistical models, one utilizing the absolute organ weight and the others incorporating a body weight adjustment by using organ-weight-to-body-weight ratio or by using body weight as a covariable in an analysis of covariance. For 200 ppb males, ventral prostate gland (absolute and relative) and testis (all statistical models) weights were decreased relative to controls while the relative pituitary gland weight was increased. Regardless of the statistical model used, the dorsolateral prostate gland weight in the 5 ppb group was increased relative to the control group. In 200 ppb females, absolute and relative ovary weights were decreased while relative liver weight was increased.
Microscopic evaluation indicated exposure-induced changes in multiple organs of both sexes. Relative to the control group, incidences of ductal mammary gland hyperplasia were significantly increased in males exposed to 25 ppb or greater. In the testis, incidences of degeneration of pachytene spermatocytes and depletion of elongated spermatids in the 100 and 200 ppb groups and degeneration of round spermatids in the 200 ppb group were significantly increased compared to the control group. Testicular spermatid head counts were significantly less in the 200 ppb group. Relative to the control group, the seminal vesicle showed increased incidences of depletion of secretory material in the 100 and 200 ppb groups and atrophy in the 200 ppb group. The incidences of mild mineralization of renal tubules were increased in 100 and 200 ppb males. In females, significant disturbance of the estrous cycle occurred in animals in the 100 and 200 ppb groups, with the ovaries of 2 of 15 and 14 of 15 animals, respectively, diagnosed as anestrus. In the 200 ppb group, significantly increased incidences of uterine atrophy and vaginal mucocyte metaplasia and dystrophy occurred.
The severity of reproductive tract effects in 200 ppb male and female pups clearly eliminated this exposure concentration from consideration for the multigenerational reproductive toxicology study, while the effects of 100 ppb on dam body weight and feed consumption and reproductive tract effects in pups were primary reasons for concern for the use of this exposure concentration in the multigenerational reproductive toxicology study. The high exposure concentration for the multigenerational reproductive toxicology study was thus set at 50 ppb. Intermediate exposure concentrations of 2 and 10 ppb were selected to bracket the 5 ppb exposure concentration used in the reproductive dose range-finding study.
Multigenerational Reproductive Toxicology Study
The multigenerational reproductive toxicology study (F0 through F4, with F5 litters terminated at weaning) focused on reproductive endpoints. Animals were exposed from the time that the F0 generation was 6 weeks old through weaning of the F3 generation, and animals of the F0 through F4 generations were necropsied at 20 weeks of age. Exposure concentrations of 0, 2, 10, or 50 ppb resulted in ingested doses of approximately 0, 0.1, 0.7, or 4 µg ethinyl estradiol/kg body weight per day to males and 0, 0.2, 1, or 6 µg/kg per day to females during the time that the rats were directly consuming dosed feed. Animals (140 of each sex) from the NCTR CD (Sprague-Dawley) rat colony were obtained at weaning. Thirty-five animals per sex were assigned to exposure groups by a weight-ranked randomization procedure prior to the start of dietary exposure of the parental (F0) generation at 6 weeks of age. At the time of mating, males were paired with females from the same exposure group and they were housed together until evidence of successful mating was detected or for a maximum of 14 days. Litters were randomly standardized to four males and four females on PND 2, and 25 litters per exposure group and their associated sires and dams were randomly selected to continue on study to produce the next generation (through F5) and then necropsied at termination at 20 weeks (F0 through F4) of age. Similar procedures were used to produce each generation. Dosed feed was removed from the F3 pups at the time of weaning, and this generation and subsequent generations were maintained on control feed for the remainder of the study. The F5 litters were terminated at weaning.
In the postweaning period, exposure to 50 ppb ethinyl estradiol reduced body weights of males and females of generations in which rats were ingesting the compound throughout adulthood (F0 through F2). Significantly decreased body weights were also observed in the 10 ppb F0 female group and the 2 and 10 ppb F2 male groups. The body weight decreases were not consistently linked to decreased feed consumption. While pup birth weights were not significantly affected by exposure in any generation, during the preweaning period, significantly decreased body weight gains were observed in the 50 ppb groups of the F1, F2, and F3 generations.
Measures of fertility (mating, pregnancy, and fertility indices, time to mating, gestation length, litter size, pup birth weight) were not adversely affected by ethinyl estradiol exposure. The sex ratio of the litters was also not altered. Anogenital distance (AGD) of exposed male pups measured on PND 2 and covaried by body weight, was significantly less than that of controls in the F3 generation. In exposed females, AGD covaried by body weight was significantly increased relative to controls in the F2 generation, but decreased in the F3 generation. In all cases, the AGD differences in exposed groups relative to controls were less than 10% and were of questionable biological significance. Females exposed to 50 ppb ethinyl estradiol showed an accelerated time of vaginal opening in the F1, F2, and F3 generations. Body weight at vaginal opening was also decreased in the 50 ppb groups of the F1, F2, and F3 generations and the 10 ppb group of the F1 generation. When examined shortly after vaginal opening, the estrous cycles in all exposed groups of the F1 generation and the 50 ppb group of the F2 generation were significantly longer than those in their respective control groups and were approximately doubled in length in the 50 ppb groups. Compared to the control groups, the 50 ppb groups of the F1 and F2 generations also had significant increases in the percentage of time that they were in estrus and increased percentages of abnormal cycles. When the estrous cycles of older animals were examined after pregnancy and lactation and prior to termination, there were no significant treatment effects. No significant treatment-related effects on male sexual development were noted with the exception of an increased time of preputial separation (an indication of delayed puberty) in the 50 ppb F2 group and increased or decreased time of testicular descent in the 2 ppb groups of the F1 and F4 generations, respectively. Sporadic statistically significant effects on ovarian follicle, epididymal sperm, and testicular spermatid head counts were not convincingly treatment-related as the magnitudes of the effects were generally within the variation seen in control animals and did not show a consistent pattern in the exposed generations. While multiple statistically significant effects on organ weights in both sexes were observed, these appeared for the most part to be secondary to body weight changes and/or were not consistent across exposed generations. In males, but not females, relative pituitary gland weights were significantly greater in the 50 ppb groups of the F0 through F2 generations than in the respective control groups. Relative spleen weights were similarly greater in these males, while relative spleen weights of females were greater in the 2 ppb group of the F1 generation and in all exposed groups of the F2 generation.
Biologically significant treatment-related microscopic lesions appeared to be confined to the male mammary gland and kidney. Relative to the controls, incidences of mammary gland alveolar/ductal hyperplasia were increased in the 50 ppb groups of the F0, F1, F2, and F3 generations, the 2 and 10 ppb groups of the F1 generation, and the 10 ppb group of the F2 generation. The effect of ethinyl estradiol on the occurrence of male mammary gland hyperplasia was more pronounced in the continuously exposed F1 and F2 generations compared to the late adolescent and adult exposure of the F0 generation and the preweaning-only exposure of the F3 generation, indicating that both developmental and adult exposures contributed to the maintenance of this effect into adulthood. Although a slight increase in the incidence of mammary gland alveolar hyperplasia occurred in 50 ppb males in the unexposed F4 generation, the increase was not statistically significant. Significant effects of ethinyl estradiol on the male kidney were limited to the 50 ppb group of the continuously exposed F1 and F2 generations, where incidences of mild mineralization of the renal tubules were increased relative to those in the controls.
Ethinyl estradiol administered at exposure concentrations of 2, 10, or 50 ppb in a low phytoestrogen diet to NCTR CD (Sprague-Dawley) rats showed clear biological activity including potentially adverse effects. Both preweaning and postweaning body weights of males and females were decreased during periods of direct exposure to dosed feed. Ethinyl estradiol accelerated the attainment of puberty of females under continuous exposure conditions (F1 and F2) and of animals where dosing was terminated at weaning (F3). Perturbation of the estrous cycle (prolonged cycles, aberrant cycles, time in estrus) in young females after vaginal opening and prior to mating was observed in the the F1 and F2 generations. In males, statistically significant inductions of male mammary gland hyperplasia (F0 through F3 generations) and mild mineralization of renal tubules (F1 and F2 generations) were observed. The majority of these effects were observed at 50 ppb, but significant effects on body weight reduction and male mammary gland hyperplasia were observed at the lowest exposure concentration (2 ppb). With the possible exception of a 1.5-day delay of preputial separation in the F2 males, effects of ethinyl estradiol did not appear to be magnified across exposed generations.