| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Urology (M/C 955), College of Medicine, University of Illinois, Chicago, Illinois 60612-7310
b Department of Environmental & Molecular Toxicology, North Carolina State University, Raleigh, North Carolina 27695-7633
c Endocrinology Branch, Reproductive Toxicology Division, National Health and Environment Effects Research Laboratory MD72, United States Environmental Protection Agency, Durham, North Carolina 27713
ABSTRACT
Brief exposure of rats to high doses of natural estrogens early in life results in permanent alterations of the prostate gland, which include differentiation defects, altered gene expression, and dysplasia with aging. Whether low-dose treatments can cause similar effects in the developing prostate remains controversial. The current project was designed to determine the dose-response relationship of the prostate gland to estradiol exposure during the developmentally critical neonatal period in the rat. Male Sprague-Dawley (SD) rats were treated on Days 1, 3, and 5 of life by s.c. injections of a 7-log range of doses (0.015 µg/kg to 15.0 mg/kg) of ß-estradiol-3-benzoate (EB) in 25 µl of peanut oil (Arachis) as vehicle. In a separate block, neonatal Fisher 344 (F344) rats received 0.15, 15.0, or 1500.0 µg EB/kg. Rats were killed on Postnatal Day (PND) 35 or 90, and the prostates were microdissected, weighed, and frozen for immunohistochemistry. Preputial separation and hepatic testosterone hydroxlase activities were monitored and measured to determine the onset of puberty. On PND 35, there was an increase in prostate weights of SD rats treated with low doses of EB and a decrease in prostate weights of SD rats treated with high doses. The low-dose effect was entirely abolished by PND 90, and only high-dose suppression of organ sizes was found. The transient nature of the effect in low-dose animals suggests an advancement of puberty as the cause for increased reproductive organ weights on PND 35. F344 rats were more sensitive than SD rats to the suppressive effects of high doses of neonatal EB on PND 90. Despite this heightened responsiveness in the F344 rats, a low-dose estrogenic effect on adult prostate weights was not observed. Thus, in the rat model a sustained effect at low doses of natural estrogens is not present in the prostate glands.
early development, environment, estradiol, prostate, puberty, steroid hormones, toxicology
INTRODUCTION
The estrogen receptor (ER) with its two isoforms, 
and ß, is highly promiscuous in regards to its ligands, binding a great variety of both natural and synthetic estrogenic chemicals [1]. Although binding parameters for these compounds differ depending on their three-dimensional structure, their potential to agonize the ER has nonetheless raised great concerns, especially since an increasing body of evidence suggests that various manmade chemicals can interfere with the endocrine system of animal species and humans via ER activation [2]. Whether environmental contaminants with such disrupting properties pose a present danger to humans and wildlife has been controversial ever since the endocrine disruptor theory was first formulated [3]. Concentrations at which these compounds are detected usually lie below previously reported no-observed-adverse-effect levels (NOAELs), and according to the traditional paradigm of toxicology, exposure should therefore not result in any significant malignant response. However, recent reports have suggested nonmonotonic dose-response relationships for natural and synthetic estrogens, where in utero exposure to low environmentally relevant concentrations produced measurable effects in the adult prostate glands of male mice [4, 5]. Attempts to reproduce the low-dose effect have had inconclusive outcomes. Some attempts failed entirely [6–9], while others confirmed the ability of estrogenic chemicals to affect endpoints other than the prostate after exposure in the low-dose range [10–12]. If future investigations positively confirm the existence of a low-dose effect, current toxicity testing protocols and dose-response assessment methods will have to be reevaluated.
Unlike in humans, rodent prostate development largely occurs postnatally (e.g., growth, branching morphogenesis, epithelial differentiation), thus providing a window of susceptibility to hormonal disruption early in life [13]. Brief exposure of rats to high doses of exogenous natural estrogens during this period was shown to result in permanent alterations of the prostate gland, including marked developmental and differentiation abnormalities, dose-dependent reductions in adult prostate size, altered secretory function during adulthood, permanent alterations in the expression of androgen receptors (ARs) and ERs within prostatic cells and perturberation of the transforming growth factor ß signaling system at multiple cellular levels in the prostate [14–16]. The mechansims of this neonatal imprinting or developmental estrogenization is not fully understood, but upregulation of ER expression within periductal stromal cells along the length of the developing ducts of estrogenized prostates suggests an amplification of estrogenic signals [17]. Neonatal exposure to high concentrations of synthetic estrogenic environmental contaminants can likewise lead to similar disruptions of prostate development, as experiments with diethylstilbestrol (DES) confirmed [18]. However, it is still uncertain whether low-dose exposure to either natural or synthetic estrogens causes similar or inverted responses in the prostate gland (i.e., increased prostate sizes). Likewise, it is unclear whether different time points of exposure result in different responses [19], since fetal exposures target an earlier phase of prostatic development than does neonatal exposure.
Recent studies have demonstrated that genetic predisposition for estrogen responsiveness can play a pivotal role in assessing the effects of fetal or neonatal imprinting. For instance, differential susceptibility to disruption of male development by 17ß-estradiol was observed in C57BL/6J, C17/Jls, and CD-1 mouse strains [20]. Similarly, while the reproductive tract of female Fisher 344 (F344) rats reacted very sensitively to estrogen treatment, Sprague-Dawley (SD) rats were more resistant to these treatments [11, 21, 22]. Although the cause of such strain-specific responses to estrogen exposure is uncertain, both the metabolic clearance rate, ER binding, and immediate early gene transcription have been ruled out as possible reasons in favor of a delayed or intermediate activation of responding genes [22]. Long et al. [22] reported that strain-specific differential induction of DNA synthesis in vaginal epithelium of SD and F344 rats occurred only in response to the synthetic estrogenic compound bisphenol A but not estradiol. Consequently, assessment of the potential of any estrogenic compound to act as a disruptor of reproductive tract development may depend on the endpoints observed and the animal model under study.
The present study was designed to determine the dose response of the developing prostate gland to neonatal estrogen exposure in SD and F344 rats. A companion investigation was designed to analyze the dose response of the male reproductive tract [23]. In particular, our goals were to investigate the response of known estrogenization markers of the prostate to a large dose range of natural estrogen as a model for xenoestrogen-mediated disruption of prostate development and to determine whether neonatal exposure of rats to low-dose estrogens would result in the same effects reported for fetal exposure in mice.
MATERIALS AND METHODS
The critical period for hormonal imprinting of the rodent prostate with high concentrations of natural estrogen is between Days 1 and 5 of life [17]. Therefore, we used the model for neonatal estrogenization established in our laboratory, where newborn pups are administered estrogen on Days 1, 3, and 5 of life. SD and F344 rats were chosen for the experiment because of their reported differential sensitivity to natural and synthetic estrogens. Great interstudy variability due to genetic predisposition for estrogenization may have been responsible for failures to replicate previous results and has stressed the need for comparative analyses of different rodent strains as in vivo models for the investigations of low- and high-dose estrogen effects (National Institute of Environmental Health Sciences and National Toxicology Program Low-Dose Peer Review Panel, unpublished results).
Animals and Housing
All animals were handled in accordance with the principles and procedures of the Guiding Principles for the Care and Use of Animal Research. Timed-pregnant SD rats (Zivic-Miller Laboratories, Pittsburgh, PA) and F344 rats (Harlan Teklad, Madison, WI) were housed individually in a light- and temperature-controlled environment in polypropylene solid-bottom cages with steel covers on corncob bedding (Bed-o'corn, The Andersons, Maumee, OH). Rooms were kept at approximately 21°C and 60% relative humidity, with a 14L:10D schedule. Animals were allowed a special soy-free diet (Sterol Free Rat Diet; Ziegler Bros., Gardners, PA) and access to tap water ad libitum. To avoid additional variability due to varying estrogenicity in the rat diet, animals of all groups and of both experiments (SD and F344) were given food from the same lots.
Pregnant dams were monitored for delivery, with day of parturition being designated Postnatal Day (PND) 0 of the experiment. The total numbers of live and dead pups and of male and female pups were recorded for each litter. In the first experiment (SD rats), each litter was culled randomly on PND 1 to a maximum of 10 neonates, with up to 10 male pups per litter if possible. When fewer than 10 males were available, appropriate numbers of females were retained. Litters with 10 or fewer pups were not altered. Litters were not culled in the second experiment (F344 rats), and all male offspring available were recruited.
Dosing and Treatment
The SD experiment was performed between February and December 1999, with the two experimental blocks (block 1, necropsy on PND 35; block 2, necropsy on PND 90) conducted in succession and with a separate oil control group for each block. Litters of two females were randomly assigned to a treatment group, so that control and treated animals were not monitored consecutively but rather simultaneously. A total of 32 gravid females were used in the study. Newborn SD rats were treated on PNDs 1, 3, and 5 with s.c. injections of a 7-log range of doses (0.015 µg kg BW-1 day-1 to 15 mg kg BW-1 day-1) of ß-estradiol-3-benzoate (EB; Sigma Chemical Co., St. Louis, MO) in 25 µl of peanut oil (Arachis sp.) as vehicle or with s.c. injections of vehicle alone (oil control). The F344 experiment was conducted between April and August 2000. Six timed-pregnant dams were randomly assigned to one of four treatment groups (oil control or 0.15, 15, or 1500 µg EB kg BW-1 day-1) and monitored for birth as described above. A total of 29 gravid females were used in the study. Male neonates were injected s.c. on PNDs 1, 3, and 5 as described for SD pups. In addition, one naive control group consisting of five females was included in the experiment to test for possible sources of variation that may have an effect on the outcome (e.g., handling, potential estrogenicity of the vehicle). Dams of the naive control group and their offspring were not disturbed until the pups were weaned on PND 21. Male pups did not receive a vehicle injection on PNDs 1, 3, and 5 but remained under observation only. Necropsy was performed on PND 90 on all male offspring (i.e., treated, oil control, and naive control animals).
Tissue Removal and Preparation
After animals were killed by cervical dislocation, ventral, lateral, and dorsal prostate lobes were removed from each animal, weighed, and microdissected. Five specimens per treatment group were embedded in OCT mounting medium (Sakura Finetek USA, Torrance, CA) along their longitudinal orientation for future immunohistochemistry (IHC) and subsequently snap frozen in liquid propane pending further analysis. Serum was stored at -80°C for later hormone analyses. All dissections were performed by an operator who was blind to the treatment group of the animals.
Microdissection of the prostate was conducted under an Olympus zoom stereo SZ4045 microscope (Optical Analysis Corporation, Nashua, NH) and followed the protocol described elsewhere [24]. Briefly, the entire urogenital tract was removed and placed immediately into cooled PBS (4°C). Specimens were stripped of adipose tissue, and after sequential removal of the ampucillary glands, vasa deferens, bladder, seminal vesicles, and coagulating glands, the individual prostate lobes were separated. For specimens later subjected to IHC, the ventral lobes were further dissected into smaller ductal arrays and embedded along the longitudinal axis.
Immunohistochemistry
The IHC method was a modification of the method described by Prins et al. [24] using the labeled streptavidin biotin technique (LSAB 2 kit for rat tissues; Dako, Carpintera, CA). Frozen individual lobes were mounted on precooled chucks (-20°C) in a Reichert-Jung cryostat (Leica, Inc., Deerfield, IL), and sections were thaw-mounted onto gelatin-coated glass slides. Whenever possible, individual lobes were sectioned longitudinally to reveal the proximal-distal orientation. Native tissue sections were fixed in 2% paraformaldehyde with 0.2% picric acid, rinsed, blocked against endogenous peroxidase in 3% hydrogen peroxidase, and subsequently incubated with the primary antibody. Specific antibodies, sources, concentrations, and incubation times are presented in Table 1. The primary antibody was reacted with a biotinylated bridging antibody, and after incubation with streptavidin peroxidase detected with 3-amino-9-ethylcarbazole as a chromagen. Specimens were counterstained with Mayer hematoxylin (Electron Microscopy Sciences, Ft. Washington, PA) and mounted and coverslipped in Glycergel Mounting Medium (Dako). Images of immunostained specimens were digitalized with an Axioskop microscope (Carl Zeiss, Inc., Thornwood, NY) and a digital AxioCam camera scanner (Zeiss) and were processed and analyzed using the AxioVision 2.05 system (Zeiss).
|
Normal rabbit IgG (Vector Laboratories, Burlingame, CA) or normal mouse ascites fluid (Sigma Chemical Co., St. Louis, MO) were substituted for primary antibodies as negative controls. For comparative studies, tissues from rats of the same age but different treatments were always processed in parallel to reduce discrepancies related to interassay variability in staining intensity.
Statistics
Statistical analysis of organ weights has drawn increased attention as a possible reason for the differences in results of experiments investigating low-dose estrogen effects. We thus decided to analyze all organ weights both as absolute and as normalized by BW. A Pearson correlation analysis was employed to establish possible positive organ weight-BW correlation, in which case we prioritized the relative weights expressed as mg/100 mg BW. Data were analyzed using a one-way ANOVA to determine significant differences between data sets after establishing homogeneity of groups. Either a post hoc Dunnett multiple comparison test of oil controls and individual treatment groups or a Tukey-Kramer multiple comparison were performed when the overall ANOVA was significant at P < 0.05. If required, data were either log-transformed and compared with a one-way ANOVA or medians were compared by a nonparametric Kruskal-Wallis ANOVA, followed by the Dunn multiple comparison test. Two groups were compared using an unpaired t-test if homogeneity allowed or using the nonparametric Mann-Whitney test if differences between standard deviations were significant. Differences were accepted as significant when at P < 0.05. All values are expressed as means ± SEM unless stated otherwise, with indication of the number (n) of separated determinations corresponding to individual numbers.
RESULTS
Response of SD Rats to Neonatal Estradiol Benzoate Treatment
Effects of treatment on BW and prostate weight Table 2 summarizes the effects of neonatal EB treatment on BW and absolute prostate weights. On PNDs 35 and 90, BWs were significantly different in individual treatment groups, although not in a dose-dependent manner. The overall dose response of prostate sizes on PNDs 35 and 90 was monotonic; absolute weights in all three lobes were decreased at the two highest EB treatment doses (1500.0 and 15000.0 µg/kg BW). In contrast to absolute weights, relative prostate weights (normalized to BW) on PND 35 showed a nonmonotonic dose response (Fig. 1, A–C). They were significantly increased at a low dose (0.15 µg/kg BW) and decreased at the high-dose end of the treatment regimen (1500.0 and 15 000.0 µg/kg BW). Although similar trends were observed for all three lobes, the increase in the low-dose group was only significant for the dorsal prostate lobe (P < 0.01). The observed size reductions at high doses of EB were significant for all three lobes (P < 0.05). A direct comparison of the percentage changes in relative organ weights of all three prostate lobes further revealed differential sensitivity of individual lobes to neonatal estrogenization, which was greatest in the dorsal lobe for both low-dose and high-dose exposure (Fig. 1D). A histologic comparison of sections taken from the midregion of each dorsal prostate lobe from control, low-dose (0.15 µg/kg BW) and high-dose (15 000.0 µg/kg BW) animals showed that the treatment resulted in increased ductal numbers and in an elongation of the proximal aspects of the ducts (Fig. 2, A and B). Unlike on PND 35, relative organ weights on PND 90 displayed a monotonic dose response with high doses of EB (1500.0 or 15 000.0 µg/kg BW) causing permanent and significant weight reductions of the prostate gland, whereas the previously observed stimulatory low-dose effect was no longer present (Fig. 3, A–D).
|
|
|
|
Effects of treatment on prostatic AR and cellular differentiation
Low-dose EB treatment did not affect the presence or localization of any of the differentiation markers investigated immunohistochemically in the ventral prostate lobes on PND 35. However, there was a correlation of presence of differentiation markers and increasing EB concentrations, leading to significant alterations at high EB concentrations (Fig. 4, A–O). Numbers of AR-positive nuclei in luminal epithelial cells appeared unaltered in animals neonatally treated with concentrations of 1.5 µg/kg BW EB or lower, but already at 15.0 µg/kg BW the number of immunostained nuclei was reduced. This trend continued progressively, until at 15 000.0 µg/kg BW virtually no AR-positive epithelial cells were found (Fig. 4, A–C). The distribution of basal and luminal epithelial cells was affected by the EB treatment in a dose-dependent manner and did not follow a nonmonotonic response. In control and low-dose prostates, basal cells in the central and distal regions of the duct epithelium were found intermittently between luminal cells (Fig. 4, D and E). Exposure to increasing concentrations of EB caused a continuous layer of basal cells reaching from the proximal region to the distal tips of the duct (Fig. 4F). Basal cell numbers in the distal regions of the duct increased continuously from 150.0 µg/kg BW onwards, whereas luminal epithelial cell numbers declined progressively (Fig. 4, G–I). In control prostates, - and 
-actin were immunolocalized to a smooth muscle cell layer, which was separated from the basal epithelial membrane by a sheath of fibroblasts along the prostatic ducts (Fig. 4J). While the fibroblast sheath was multilayered around the proximal duct, it progressively thinned out in the distal region, so that the smooth muscle cells came into direct contact with the epithelium. Again, no particular differences were observed at low doses (Fig. 4K). However, with increasing EB concentrations the multilayered fibroblast sheath extended further distally, until at high-dose exposure it reached fully into the distal tips (Fig. 4L). Prostate binding protein (PBP) was highly expressed in epithelial cells of control specimens (Fig. 4M). Although unaltered in low-dose prostates (Fig. 4N), the immunosignal decreased in intensity as EB concentrations increased (Fig. 4O). The same pattern of gradual alteration of differentiation markers was observed on PND 90 as EB doses reached the high-dose range. Immunohistochemical analysis of the dorsal and lateral lobes revealed the same responses for AR, basal and luminal epithelial cells, and smooth muscle cells as observed in the ventral lobe (results not shown).
|
Response of Fisher 344 Rats to Neonatal Estradiol Benzoate Treatment
Effects of treatment on body and organ weights Overall, F344 rats responded markedly differently to the estrogen treatment than did SD rats. Neonatal administration with high doses of EB resulted in a suppression of prostate morphogenesis that did not allow for a distinction of the dorsal and lateral lobe and made their individual microdissection impossible. Thus, at high doses of EB, the entire dorsolateral complex was removed and weighed. Absolute ventral and lateral prostate lobe weights responded in a monotonic fashion and were significantly reduced at the highest EB dose (1500.0 µg/kg BW; P < 0.01; Table 3). However, a comparison of absolute dorsal prostate weights uncorrected for BW revealed a significant increase at the low dose of EB (0.15 µg/kg BW) when compared with controls (Dunnett multiple comparison, P < 0.05) but not when compared with naive animals (Tukey-Kramer multiple comparison, P > 0.05). When absolute dorsolateral complex weights were compared, a significant reduction was found only for animals of the highest dose group (P < 0.01; data not shown). In contrast to SD rats, BW reductions observed in F344 rats as a result of EB treatment appeared to be dose dependent and were only present in the highest dose group (P < 0.001; Table 3). Therefore, relative prostate weights were given precedence in the analysis of dose-related effects on prostatic development and growth. The dose response of relative prostate weights in F344 rats on PND 90 was monotonic, with no effects observable in low-dose animals, whereas high doses significantly reduced prostate sizes (Fig. 5, A–D). At high doses (1500.0 µg/kg BW), relative ventral lobe weights were reduced significantly to 22.34% of the control weights, a response similar to that in SD rats (29.39%). A direct comparison of combined relative dorsal and lateral prostate weights (i.e., dorsolateral complex) and of relative ventral prostate weights between SD and F344 rats at the same exposure dose revealed no significant differences (Mann-Whitney test, P > 0.05) at any dose (Fig. 6, A and B). This finding suggests that the treatment with high-dose estrogen (1500.0 µg/kg BW) affected morphogenesis of prostate glands but not growth as such. In F344 rats, no permanent low-dose effect was observed on any prostate lobe on PND 90.
|
|
|
DISCUSSION
Neonatal exposure of rodents to high concentrations of both natural and synthetic estrogens results in significant and lobe-specific reduction of prostate sizes [14, 17, 25]. Whether exposure to environmentally relevant concentrations below previously reported NOAELs causes any effect is less clear [6, 7]. Vom Saal et al. [5] reported that feeding pregnant female mice with sub-NOAEL concentrations of estradiol or DES resulted in permanently increased prostate sizes in their male offspring, whereas high doses caused significant size reductions. In the present study, we observed a similar inverted U-shaped response curve for prostate sizes during puberty in SD rats neonatally treated with EB, but this effect was not permanent and was absent from adult animals, suggesting that unlike in utero exposure, neonatal treatment produced only a transient effect in the low-dose range (0.15 µg/kg BW). Whether this effect was related to the time point of exposure or was a species-specific response remains unclear.
In the F344 rats, absolute dorsal prostate lobe weights of adult animals treated with EB at 0.15 µg/kg BW increased by 29.25%, which is approximately the same relative margin as the weight increase reported by vom Saal et al. (27% for estradiol) [5]. This increase was significant, but since prostate weight and BW were correlated (Pearson correlation analysis, P < 0.05), we considered this analysis of uncorrected organ weights too liberal. In contrast, changes in relative dorsal prostate lobe weights were not significantly different from controls at the low-dose range, so that from the present data it can only be concluded that neonatal exposure to low doses of EB causes a temporal increase of prostate weights, most prominent in the dorsal lobe that is absent from adult animals.
The mechanism by which low doses of estrogens cause prostate weight increases on PND 35 remains uncertain. Vom Saal et al. [5] observed an overall increase of AR in adult prostates of low-dose animals, which suggested an increased sensitivity of the prostate gland to testosterone. The perinatal rodent prostate is androgen dependent in its growth, branching morphogenesis, and differentiation [26], and neonatal exposure to high doses of EB disrupts these crucial developmental processes by downregulating AR [27, 28]. During this period, androgen-mediated growth of the prostate is believed to be mediated through AR in stromal cells, which produce androgen-dependent paracrine factors that specifically drive prostate growth and differentiation [29]. It is conceivable that amplification of this process through increased levels of AR within stromal cells could permanently augment prostate growth in low-dose estrogen-exposed animals. However, IHC in the present study did not indicate any observable differences in AR presence, localization, or staining intensity in either epithelial or stromal cells of prostates of males exposed to low EB levels. Analysis of IHC data is subjective and requires large changes in signal distribution and intensity for detection of differences, so a more sensitive technique would be needed to detect any subtle differences. Nonetheless, it seems unlikely that the observed increase in prostate weights observed on PND 35 can be explained by a greater sensitivity for testosterone.
Estrogenization with high concentrations of EB impairs prostate development directly by upregulating ER and downregulating AR and indirectly through alteration of the hypothalamic-pituitary-testicular axis [30–32]. The transient character of the increased prostate weights in the present study suggests such an alteration, which by increased gonadotropin levels could lead to precocious puberty and associated larger prostate sizes. This hypothesis is consistent with our observations in the same animals as used in the current study that hepatic activities of 2- and 16-testosterone hydroxylase, two enzymes that increase during puberty in the male rat under growth hormone regulation, were increased in low-dose animals on PND 35 [23]. It therefore appears more conclusive to consider, advanced onset of pubertal processes were probably responsible for the temporary increase in prostate weights observed in low-dose animals.
Developing prostatic ducts display a serial presence of specific cell types and associated steroid receptors along their longitudinal axis that changes during postnatal differentiation and branching in a well-orchestrated fashion. Within the epithelium, two distinct cell types, the basal and luminal cells, can be distinguished by differential IHC with monoclonal antibodies against specific cytokeratins [24]. Although the origin of both cell types is controversial [33], their ratio within the prostate and their localization in the epithelium are reliable differentiation markers that allow assessment of the developmental state of the gland [34]. As the rat prostate epithelium differentiates, basal cells begin to disperse in the central and distal regions of the ducts by PND 10, so that they are localized intermittently rather than as a continuous layer in the proximal regions of the ducts. Between PNDs 10 and 15, luminal cells appear and can be demonstrated immunohistochemically. Thereafter, the cellular proportions change, with increasing luminal cell and decreasing basal cell numbers as the gland matures. The ducts are surrounded by a thin single-cell layer of fibroblasts interspersed between the basement membrane [35], which thins as the ducts elongate and branch [16]. In the distal regions, the fibroblast cell layer disappears completely, leaving the smooth muscle cells in close proximity to the epithelium [34].
Neonatal exposure to high concentrations of EB causes the continuous layer of basal cells normally associated with the proximal regions of the ducts to extend into the distal regions [15]. It also results in a thick periductal layer of fibroblasts that reach far into the distal tips of the developing ducts [16], resulting in a proximalized prostatic phenotype [34]. In the present study, we confirmed earlier observations on the alterations of basal and luminal epithelial cells and thickened fibroblast and smooth muscle cell layers in the high-dose prostate. However, we did not detect any significant changes in presence or localization of these markers in the prostates exposed to low doses of EB. This finding suggests that the early growth spurt observed in low-dose prostates occurs not because of an alteration of the proximal-distal organization of the duct but rather because the low-dose prostate reaches a certain developmental state earlier than does the control prostate, supporting our hypothesis of precocious puberty. If this hypothesis is correct, the two estrogenization types (low versus high dose) are not only distinctly different in phenotype but also in the mechanism leading to their phenotypes.
Recent attempts to replicate the experiments resulting in low-dose effects on the rodent prostate in response to sub-NOAEL levels of natural or synthetic estrogens in mice have resulted in contradictory outcomes, ranging from failures [6, 7] to confirm the observations made by vom Saal et al. [5] to similar observations for the prostate or other endpoints [36–38]. Additional studies on rats that likewise focused on the prostate as an endpoint for a low-dose effect also reported no observable adverse effects [8, 9, 39]. In all these "negative" studies, investigators weighed either the entire prostate gland or the ventral prostate lobe alone. Our results demonstrate that the observed low-dose effect is lobe specific and greatest in the dorsal prostate. A direct comparison of the total prostate weights on PND 35 did not reveal significant differences between low-dose and control groups (Dunn test, P > 0.05); the increase in ventral lobe size was not significant either. Although the time point of administration differed between the cited and the present studies, in light of the clear lobe-specific nature of the observed low-dose effect it appears imperative to analyze lobes individually rather than analyzing the entire prostate gland.
Finally, our results support earlier reports of strain-specific differential responses to estrogen exposure in rodents. For instance, Ben-Jonathan and coworkers showed that prolonged exposure to the estrogenic monomer of polycarbonate plastics bisphenol A caused hyperplasia, hypertrophy, and differentiation in the reproductive tract epithelia of female F344 but not SD rats [21, 22]. Similar genetic disposition for susceptibility to estrogen was reported for mice, where estradiol implants affected testis weight and spermatogenesis differently in three strains of mice [20]. In the present study, combined relative dorsal and lateral prostate weights did not differ significantly between SD and F344 rats, but high doses of EB (1500.0 µg/kg BW) had a greater effect on morphogenesis in F344 rats, where unlike in SD rats the two individual lobes had not formed distinguishably. This finding does not necessarily suggest greater sensitivity to estrogen in one strain but rather indicates endpoint-dependent differential responses between strains. Many studies investigating possible endocrine disruption in rats by natural or synthetic estrogens used SD rats as their in vivo model [8, 9, 39]. For a more accurate assessment of the adverse effects of environmental contaminants with estrogenic properties using the prostate gland as an endpoint, SD and F344 rats should at least be tested in direct comparison.
In summary, the present findings lend support to previously reported nonmonotonic dose-response curves as a result of estrogen exposure and to the hypothesis that sub-NOAEL concentrations of estrogenic chemicals can have a biologic effect. The transient nature of the observed effect indicates that in response to neonatal exposure to low doses of estrogens, male puberty is probably advanced. Differential responses of F344 and SD rats confirm previously reported strain-specific sensitivities to estrogens, stressing the necessity of rethinking the question of the strain of choice for risk assessment studies and investigations of the effects of low doses of estrogen on the developing prostate gland.
|
ACKNOWLEDGMENTS
The authors thank Lynn Birch for her excellent technical assistance.
FOOTNOTES
1 Supported by EPA STAR grant R826299 to G.S.P. 
2 Correspondence: Gail S. Prins, Department of Urology (M/C 955), University of Illinois, 820 South Wood Street, Chicago, IL 60612-7310. FAX: 312 996 1291; gprins{at}uic.edu
Accepted: July 3, 2001.
Received: March 27, 2001.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  L. Li, M. E Andersen, S. Heber, and Q. Zhang Non-monotonic dose-response relationship in steroid hormone receptor-mediated gene expression J. Mol. Endocrinol., May 1, 2007; 38(5): 569 - 585. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-M. Ho, W.-Y. Tang, J. Belmonte de Frausto, and G. S. Prins Developmental Exposure to Estradiol and Bisphenol A Increases Susceptibility to Prostate Carcinogenesis and Epigenetically Regulates Phosphodiesterase Type 4 Variant 4 Cancer Res., June 1, 2006; 66(11): 5624 - 5632. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C-L Hsieh, Z Xie, Z-Y Liu, J E Green, W D Martin, M W Datta, F Yeung, D Pan, and L W K Chung A luciferase transgenic mouse model: visualization of prostate development and its androgen responsiveness in live animals J. Mol. Endocrinol., October 1, 2005; 35(2): 293 - 304. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. O. Goyal, T. D. Braden, C. S. Williams, P. Dalvi, M. Mansour, and J. W. Williams Estrogen-Induced Abnormal Accumulation of Fat Cells in the Rat Penis and Associated Loss of Fertility Depends upon Estrogen Exposure during Critical Period of Penile Development Toxicol. Sci., September 1, 2005; 87(1): 242 - 254. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. O. Goyal, T. D. Braden, C. S. Williams, P. Dalvi, M. M. Mansour, and J. W. Williams Permanent Induction of Morphological Abnormalities in the Penis and Penile Skeletal Muscles in Adult Rats Treated Neonatally With Diethylstilbestrol or Estradiol Valerate: A Dose-Response Study J Androl, January 1, 2005; 26(1): 32 - 43. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Huang, Y. Pu, S. Alam, L. Birch, and G. S. Prins Estrogenic Regulation of Signaling Pathways and Homeobox Genes During Rat Prostate Development J Androl, May 1, 2004; 25(3): 330 - 337. [Full Text] [PDF]
|
|
|
|
|
|
G. P. Daston, J. C. Cook, and R. J. Kavlock Uncertainties for Endocrine Disrupters: Our View on Progress Toxicol. Sci., August 1, 2003; 74(2): 245 - 252. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. G. Ramos, J. Varayoud, L. Kass, H. Rodriguez, L. Costabel, M. Munoz-de-Toro, and E. H. Luque Bisphenol A Induces Both Transient and Permanent Histofunctional Alterations of the Hypothalamic-Pituitary-Gonadal Axis in Prenatally Exposed Male Rats Endocrinology, July 1, 2003; 144(7): 3206 - 3215. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.O. Goyal, A. Robateau, T.D. Braden, C.S. Williams, K.K. Srivastava, and K. Ali Neonatal Estrogen Exposure of Male Rats Alters Reproductive Functions at Adulthood Biol Reprod, June 1, 2003; 68(6): 2081 - 2091. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. S. Prins, W. Y. Chang, Y. Wang, and R. B. van Breemen Retinoic Acid Receptors and Retinoids Are Up-Regulated in the Developing and Adult Rat Prostate by Neonatal Estrogen Exposure Endocrinology, September 1, 2002; 143(9): 3628 - 3640. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Putz, C. B. Schwartz, G. A. LeBlanc, R. L. Cooper, and G. S. Prins Neonatal Low- and High-Dose Exposure to Estradiol Benzoate in the Male Rat: II. Effects on Male Puberty and the Reproductive Tract Biol Reprod, November 1, 2001; 65(5): 1506 - 1517. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
