Reproductive Toxicology Division, MD-72, National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711
Received May 14, 2000; accepted July 24, 2000
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key Words: atrazine; rat; early pregnancy; embryo implantation; strain difference; progesterone; prolactin; hormones.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Progesterone is essential for the initiation of pregnancy in the rat and human (De Feo, 1967; Psychoyos, 1973
). During the first 10 days of pregnancy in the rat, there are two daily surges of prl, a nocturnal surge with a peak between 0300 and 0500 h, and a diurnal surge peaking between 1700 and 1900 h (Terkel, 1988
). It is these surges of prl that are luteotropic, leading to the rescue and maturation of the corpora lutea (CL) and to the subsequent dramatic increase in the secretion of the progesterone that, along with ovarian estrogen, directly permits implantation of the embryo in the uterus (Morishige and Rothchild, 1974
; Smith et al., 1976
). In fact, previous work has shown that the administration of bromocryptine, a dopamine agonist that inhibits the synthesis and release of pituitary prl (Barrow and Tindall, 1983
), to rats during gestation day (GD) 18 can block implantation, presumably due to its ability to prevent prl secretion and thereby block the necessary rise in serum progesterone (Cummings et al., 1991
). In the human it is LH that supports CL function prior to implantation and human chorionic gonadotropin (hCG) that exerts postimplantation support of the CL and progesterone secretion (Zeleznik and Fairchild-Benyo, 1994
).
Effects of ATR on pregnancy in rats have been investigated to a limited extent. According to Infurna et al. (1988), the administration of technical ATR to rats during GD 615 had no effect at the 70 mg/kg dose, whereas 700 mg/kg produced maternal mortality. A report by Peters and Cook (1973) states that 1000 or 2000 mg/kg ATR, administered on GD 3, 6, and 9 resulted in embryonic resorption and embryotoxicity. In work reported by Narotsky et al. (1999), exposure of rats to ATR during GD 610 produced full litter resorptions at 200 mg/kg. None of these studies investigated the effects of relatively low doses of ATR or effects on implantation.
Our study is based on the potential for ATR to disrupt early pregnancy and implantation in the rat via a suppression of prl surges. The hypothesis was that if ATR suppresses prl during early rat pregnancy, then such a suppression of either the diurnal or nocturnal surge of prl by ATR would impair implantation in rats. The testing of this hypothesis has been done indirectly: ATR was administered to rats during early pregnancy and the impact of the chemical on implantation and pregnancy maintenance until day 9 was assessed. The two different times of ATR exposure used correspond to intervals preceding the diurnal and nocturnal prl surges (Terkel, 1988). A previous investigation of strain sensitivity to ATR during the postimplantation interval (Narotsky et al., 1999
) led us to incorporate four different rat strains. The strains employed were 1) Holtzman (HLZ) rats, which are historically used in physiologically based pregnancy studies (Yochim and De Feo, 1963
); 2&3) LE and SD, strains that were previously shown to have a sensitivity to ATR with respect to altered estrous cyclicity (Cooper et al., 1995
); and 4) Fischer 344 (F344), animals that have been used in previous work on effects of ATR on pregnancy (Narotsky et al., 1999
). Thus, these studies examine the possibility that different strains of rats are differentially sensitive to the effects of potential prl suppression by ATR, leading to adverse effects on pregnancy.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Experiment 1.
Rats were dosed by gavage with ATR, following daily weighing, on days 18 of pregnancy at 1400 h. This corresponds to a period prior to the diurnal prl surge. Technical ATR (6-chloro-N-ethyl-N`-(1-methylethyl)-1,3,5-triazine-2,4-diamine) was supplied by Novartis (Greensboro, NC) and was of 97.1% purity. ATR was suspended in 1% methyl cellulose and diluted so that animals received 2.5 ml/kg at each dose level. Groups of 810 rats received 0 (1% methyl cellulose vehicle), 50, 100, or 200 mg ATR/kg/day. On day 9, rats were decapitated at 1400 h and trunk blood collected. Following separation from red cells, serum was frozen for later assay of LH, progesterone, and estradiol. Serum prl was not measured, as the experimental design (including times and method of euthanasia) precluded the determination of prl surge levels or patterns. At necropsy the following parameters were assessed: body weight, number of implantation sites, number of corpora lutea (CL), ovarian weight, uterine weight, and number of resorptions apparent by day 9.
Experiment 2.
Groups of rats were treated and evaluated exactly as in Experiment 1 except that ATR was administered at 0200 h (prior to the nocturnal surge of prl) and rats were killed at 2100 h on day 8 of pregnancy (19 h after the last dose of ATR).
Radioimmunoassays (RIAs).
Serum estradiol was measured by a double antibody RIA kit obtained from Diagnostic Products (Los Angeles, CA), and serum progesterone was measured using a coat-a-count RIA kit supplied by the same company. The determination of LH was performed using materials supplied by the National Institute of Arthritis, Diabetes, Digestive, and Kidney Diseases. The iodination preparation was I-6, the reference preparation was RP-2, and the antisera was S-8. Radiolabeling of the iodination preparations was performed with 125I (New England Nuclear, Boston) using the Chloramine-T method of Greenwood et al. (1963), after which labeled hormone was purified on a BioGel P-60 (Bio-Rad Laboratories, Melville, NY) column. The assay for LH had a sensitivity of 15 pg/tube, and inter-and intra-assay coefficients of variation of 7.7% and 5.9%, respectively.
Statistics.
Parametric data were analyzed by a 3-way ANOVA (dose x strain x time) using the General Linear Models (GLM) procedure of SAS statistical software (SAS Institute, 1985). When the overall ANOVA was significant at p < 0.05, comparisons between groups were made using t-tests (Least Squares Means; SAS). In most cases, only comparisons across dose within strain and time were made. For hormone levels in control groups, comparisons across strain and time were made. Percent data (pre- and postimplantation loss) were analyzed by 3-way ANOVA as above after arcsin square-root transformation. Significant differences between data points were indicated when the p value was less than 0.05.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
|
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In a related set of studies, Narotsky et al. (1999) examined strain differences in the sensitivity of rats to ATR and bromodichloromethane (BDCM), a by-product of drinking water disinfection. In those experiments, rats were dosed on GD 610, a postimplantation, LH-dependent interval, and evaluated following parturition. When the pregnancies of ATR- or BDCM-treated dams were allowed to proceed to term, an all-or-nothing phenomenon of full litter resorptions (FLR) was observed. Whereas BDCM treatment caused FLR in F344 but not SD rats, ATR treatment (200 mg/kg/day) led to comparable incidences of FLR in F344, SD, and LE rats (Narotsky et al., 1999). Based on the current data, it is not known whether further evidence of postimplantation loss and/or FLR might have been observed if the pregnancies in which the dams were treated on days 18 were allowed to proceed to term. However, such a finding would corroborate data reported by Narotsky et al. (1999). The increase in preimplantation loss observed here is likely the result of preimplantation exposure to ATR producing a suppression of prl levels, which in turn would impair CL maturation, elevation of progesterone secretion, and the promotion of implantation. Such a prl suppression, induced by bromocryptine, has been shown to block implantation in a similar model of early pregnancy exposure in the rat (Cummings et al., 1991
).
The finding of no effect of ATR on serum progesterone in three out of four strains of rats is surprising in light of the essential role of prl in CL development and progesterone secretion and also in light of the putative action of ATR to alter serum prl (Cooper et al., 2000). The measurement of baseline prl at necropsy on day 9 (with euthanasia performed rapidly enough to prevent stress-induced prl elevation) was not considered useful, as it is the twice-daily surges of the hormone that are important for the stimulation of progesterone secretion (Smith et al., 1976
). In a study such as that reported here, the introduction of an additional variable, that of using a series of sequential necropsies to determine the rise and fall of each prl surge, for each time of dosing, dose level, and strain data point was considered to be unjustifiable with respect to the number of animals required relative to the amount of information obtained. Although it is impossible to state with certainty the effect of ATR on prl during early pregnancy in this study, the data on progesterone suggest either that ATR, at these dosages, had little or no effect on the surges of prl during early pregnancy, or that ATR-induced changes in prl did not affect CL function sufficiently to interfere with pregnancy.
With the exception of one case (F344/diurnal dosing/all dosages), treatment of every strain at both time intervals yielded at least a trend toward reduced baseline serum LH, and this effect was shown to be significant in six cases. This is not a measure of LH surge suppression, but merely an assessment of changes in baseline levels. In two of the six cases of baseline LH suppression, there was also a significant increase in percent postimplantation loss (HLZ/diurnal dosing/100 and 200 mg/kg). In the other four cases, there is usually at least a trend toward increased postimplantation loss that parallels the decline in LH in each group. The number of resorptions detectable on day 9 does not always reflect the full complement of resorptions that might be seen later in pregnancy or near term. In work reported by Narotsky et al. (1999), for example, FLR was detected in rats at term following treatment with ATR during early pregnancy.
Nocturnal but not diurnal dosing with ATR was effective in increasing preimplantation loss in F344 rats. No time-of-dosing effects were observed in other strains for this end point. Both nocturnal and diurnal dosing (independently) produced postimplantation loss in HLZ rats, but no effects on this end point were evident in other strains. Serum LH showed a mixed pattern with respect to effects of dosing interval. In LE rats, dosing at each interval reduced serum LH. However, in HLZ and F344 rats only diurnal or nocturnal (respectively) dosing was effective. The finding of increased serum estradiol was seen in only one case, following diurnal dosing. The apparent elevation of estradiol levels in this one group of SD rats may be simply the result of variability in the data for that hormone. Overall, the effect of different intervals of dosing depended on the end point measured and the strain of rat employed.
A comparison of serum levels of progesterone, estradiol, and LH on day 9 in control animals across strains yields some interesting observations. In animals killed at 1400 h on day 9 (following diurnal dosing), serum progesterone levels were significantly higher in F344 and HLZ animals than in SD or LE rats. At the same necropsy time, estradiol levels in F344 rats were lower (undetectable) than those of the other three strains, and serum LH showed a pattern where F344 and HLZ rats were lowest, SD rats were intermediate, and LE rats were significantly higher than all others. In animals killed at 2100 h on day 8 (following nocturnal dosing), serum progesterone levels were again high in the F344 strain and low in the SD strain, but in LE rats progesterone was high whereas HLZ rats showed low levels of the hormone. At 2100 hr, estradiol levels were much higher in LE rats than in any other strain, and estradiol levels were above detectable levels in F344 rats at this time. Serum LH patterns were similar to those seen at the earlier time: HLZ and F344 rats showed low levels, whereas SD and LE rats showed significantly higher levels of LH. These differences in hormone levels in control animals are a function not only of strain but also of the time of euthanasia, suggesting another caution for the design of toxicological studies that incorporate the measurement of ovarian or pituitary hormones. It is possible that the differences in control hormone levels across strains may play a role in the differential sensitivity of the various strains to the effects of chemicals. For example, LE rats have much higher levels of estradiol than other strains at 2100 h. A connection may exist between this and the apparent resistance of LE rats to effects of ATR on implantation.
Although previous work has suggested that ATR can suppress prl levels under specific conditions unrelated to pregnancy (Cooper et al., 2000; Simpkins et al., 1998
), it is possible that the chemical may not be as effective in altering prl when administered during early pregnancy. The only strain in which ATR produced significant preimplantation loss was F344, and in those animals no significant effect on progesterone levels was observed. On the other hand, it is possible that ATR indeed reduces prl during early pregnancy, but that the minimal amount of prl necessary to sustain CL function and implantation is so low that no outward adverse effect was observed in conjunction with the hormonal effect.
In summary, exposure of rats to ATR during early pregnancy produced no adverse effects in LE or SD rats. HLZ and F344 rats appear most sensitive to the effects of ATR on early pregnancy. Our hypothesis that ATR should produce a preimplantation loss that might be based on the chemical's effect on prl has been shown to be true only for F344 rats. Postimplantation effects, likely to be mediated by LH, were produced by ATR only in HLZ rats. We conclude that the effects of ATR on early pregnancy may be selective with respect to strain and are significant only in the sensitive strains. Further work to delineate effects of ATR on prl in susceptible strains during early pregnancy is necessary.
![]() |
ACKNOWLEDGMENTS |
---|
![]() |
NOTES |
---|
2 Present address: Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC 27599-7295.
The research presented in this article was funded wholly by the U.S. Environmental Protection Agency. It has been subjected to review by the National Health and Environmental Effects Research Laboratory and approved for publication. Approval does not signify that the contents reflect the views of the agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Cooper, R. L., Parrish, M. B., McElroy, W. K., Rehnberg, G. L., Hein, J. F., Goldman, J. M., Stoker, T. E., and Tyrey, L. (1995). Effect of atrazine on the hormonal control of the ovary. The Toxicologist 15, 294.
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, 257264.[ISI][Medline]
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, 297307.
Cummings, A. M., Perreault, S. D., and Harris, S. T. (1991). Use of bromoergocryptine in the validation of protocols for the assessment of mechanisms of early pregnancy loss in the rat. Fundam. Appl. Toxicol. 17, 563574.[ISI][Medline]
De Feo, V. J. (1967). Decidualization. In Cellular Biology of the Uterus (R. M. Wynn, Ed.), pp. 191290. Appleton-Century-Crofts, Meredith, NY.
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, 114123.[ISI]
Gressel, J., Ammon, H. U., Fogelfors, H., Kay, Q. O. N., and Kees, H. (1984). Discovery and distribution of herbicide-resistant weeds outside of North America. In Herbicide Resistance in Plants (H. M. LeBaron, J. Gessel, Eds.), pp. 3146. John Wiley & Sons, New York.
Infurna, R., Levy, B., Meng, C., Yau, E., Traina, V., Rolofson, G., Stevens, J., and Barnett, J. (1988). Teratological evaluations of atrazine technical, a triazine herbicide, in rats and rabbits. J. Toxicol. Environ. Health 24, 307319.[ISI][Medline]
Morishige, W. K., and Rothchild, I. (1974). Temporal aspects of the regulation of corpus luteum function by luteinizing hormone, prolactin and placental luteotrophin during the first half of pregnancy in the rat. Endocrinology 95, 260274.[ISI][Medline]
Narotsky, M. G., Guidici, D. L., Best, D. S., Bielmeier, S. R., and Cooper, R. L. (1999). Strain comparisons of susceptibility to pregnancy loss in rats. Annual Meeting of the Endocrine Society, p. 492.
Peters, J. W., and Cook, R. M. (1973). Effects of atrazine on reproduction in rats. Bull. Environ. Contam. Toxicol. 9, 301304.[ISI][Medline]
Psychoyos, A. (1973). Hormonal control of ovoimplantation. Vitamins and Hormones 31, 201256.[Medline]
SAS Institute. (1985). SAS User's Guide: Statistics, 5th ed. SAS Institute, Cary, NC.
Simpkins, J. W., Eldridge, C., and Wetzel, L. T. (1998). Role of strain-specific reproductive patterns in the appearance of mammary tumors in atrazine-treated rats. In Triazine Herbicides Risk Assessment (L. G. Ballantine, J. E. McFarland, and D. S. Hackett, Eds.), pp. 399413. American Chemical Society, Washington, DC.
Sinclair, J. L., and Lee, T. R. (1992). Biodegradation of atrazine in subsurface environments. Environmental Research Brief, EPA/600/S-92/001.
Smith, M. S., McLean, B. K., and Neill, J. D. (1976). Prolactin: the initial luteotropic stimulus of pseudopregnancy in the rat. Endocrinology 98, 13701377.[Abstract]
Stevens, J. T., and Sumner, D. D. (1991). Herbicides. In Handbook of Pesticide Toxicology (W. J. Hayes and E. R. Laws, Eds.), Vol. 3, pp. 13171408. Academic Press, NY.
Stevens, J. T., Breckenridge, C. B., Wetzel, L. T., Gillis, J. H., Luempert, L. G., and Eldridge, J. C. (1994). Hypothesis for mammary tumorigenesis in Sprague-Dawley rats exposed to certain triazine herbicides. J. Toxicol. Environ. Health 43, 139153.[ISI][Medline]
Terkel, J. (1988). Neuroendocrine processes in the establishment of pregnancy and pseudopregnancy in rats. Psychoneuroendocrinology 13, 528.[ISI][Medline]
Wetzel, L. T., Leumpert, L. G., 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. Health 43, 169182.[ISI][Medline]
Yochim, J. M., and De Feo, V. J. (1963). Hormonal control of the onset, magnitude, and duration of uterine sensitivity in the rat by steroid hormones of the ovary. Endocrinology 72, 317326.[ISI]
Zeleznik, A. J., and Fairchild-Benyo, D. (1994). Control of follicular development, corpus luteum function, and the recognition of pregnancy in higher primates. In The Physiology of Reproduction, 2nd ed. (E. Knobil and J. D. Neill, Eds.), pp. 751782. Raven Press, Ltd., New York.