Endogenous Estrogens Inhibit Mouse Fetal Leydig Cell Development via Estrogen Receptor
http://www.100md.com
内分泌学杂志 2005年第5期
Unité de Gamétogenèse et Génotoxicité (G.D., C.L., C.D., C.R., R. H.), Institut National de la Santé et de la Recherche Médicale Unité 566, Commissariat à l’Energie Atomique, Université Paris 7, Denis Diderot, 92265 Fontenay-aux-Roses, France; and Department of Physiology, Institute of Biomedicine (P.P.), University of Turku, FIN-20520 Turku, Finland
Address all correspondence and requests for reprints to: Dr. Christine Levacher, Institut National de la Santé et de la Recherche Médicale Unité 566/Commissariat à l’Energie Atomique/Université Paris 7, Denis Diderot, DSV/DRR/SEGG/LDFG, Batiment 5A, RDC, Route du Panorama, 92265 Fontenay aux Roses, France. E-mail: christine.levacher@cea.fr.
Abstract
It is now accepted that estrogens play a role in male fertility and that exposure to exogenous estrogens during fetal/neonatal life can lead to reproductive disorders in the male. However, the estrogen receptor (ER)-mediated processes involved in the regulation of male reproduction during fetal and neonatal development are still largely unclear. We previously reported that ER? deficiency affects gametogenesis in mice but changes neither the number nor the differentiated functions of fetal Leydig cells. We show here that ER-deficient mice (ER–/–) display higher levels of testicular testosterone secretion than wild-type mice from fetal d 13.5 onwards. This results from higher levels of steroidogenic activity per fetal Leydig cell, as indicated by the hypertrophy of these cells and the higher levels of mRNA for StAR, P450c17 and P450scc in the testis, for a similar number of Leydig cells. Because LH is not produced on fetal d 13.5 and because no change in plasma LH concentration was observed in 2-d-old ER-deficient mice, LH is probably not involved in the effects of estrogens on testicular steroidogenesis in fetal and early neonatal Leydig cells. Furthermore, inactivation of ER? did not change the effect of ER inactivation on steroidogenesis. Lastly, in an organ culture system, 1 μM diethylstilbestrol decreased the testosterone secretion of wild-type fetal and neonatal testes but not of ER–/– testes. Thus, this study shows that endogenous estrogens physiologically inhibit steroidogenesis via ER by acting directly on the testis early in fetal and neonatal development.
Introduction
TWO SUCCESSIVE POPULATIONS of Leydig cells arise during normal testicular development (reviewed in Refs. 1 and 2). In the mouse, the fetal population starts to appear about 12.5 d postconception (dpc) and is essential for the masculinization of the fetus (3). The second, adult population begins to differentiate 4 d after birth (4). The fetal Leydig cells differ from the adult population in morphology, physiology, and regulation (2, 5). They are not desensitized by LH and do not require LH for differentiation (2, 6, 7). Furthermore, mutation of the androgen receptor does not affect the development of fetal Leydig cells, whereas it does impair the development of adult Leydig cells (8).
The role played by estrogens in the development and functions of fetal Leydig cells has recently come to the forefront because it has been claimed that the occurrence of alterations in male reproductive function is linked to exposure to environmental pollutants. Indeed, increases in the frequency of male reproductive disorders have been observed in humans and wildlife in many countries over the last 50 yr (reviewed in Refs. 9 and 10). A decrease in sperm count and increases in the incidence of testicular cancer, cryptorchidism, and hypospadia have been reported (9). It is widely thought that all these disorders are caused by an increase in the concentration of xenobiotics, and of xenoestrogens in particular, in the environment and in food (11, 12). These disorders, which are now collectively considered as testicular dysgenesis syndrome (13), may result from the impairment of testicular programing during fetal and neonatal life. This is particularly evident for hypospadia and cryptorchidism because the masculinization of external genitalia depends on the production of testosterone by fetal Leydig cells (3), and the descent of the testis is induced by the secretion of insulin-like factor-3 (INSL3) and testosterone by fetal Leydig cells (14, 15).
It has been demonstrated that high doses of estrogens alter fetal Leydig cell function. Dufau’s group (16) reported that estrogens inhibit testosterone production in cultured dispersed rat fetal Leydig cells. We recently showed, in an organotypic culture model, that estradiol and diethylstilbestrol (DES) decrease the number and differentiated functions of Leydig cells in rat testes explanted at 14.5 dpc (17). Furthermore, the exposure of laboratory animals to high doses of exogenous estrogenic compounds during fetal or neonatal life leads to an increase in the frequency of hypospadias and cryptorchidism (9, 13, 18). In this estrogen-treated pregnant rodent model, estradiol has been shown to affect the differentiation of fetal Leydig cells (19, 20). Lastly, the male offspring of women treated with DES during pregnancy have a higher incidence of cryptorchidism and hypospadias (21, 22, 23).
However, most of the deleterious effects of estrogens observed in the experiments and clinical cases described above were obtained with pharmacological doses of estrogens. There is currently no evidence that low doses equivalent to the level of human exposure to environmental estrogens have any effect. A few studies have investigated the effects of exposure to low doses of estrogens. Neonatal exposure to low doses of DES or genistein, a phytoestrogen, has no long-term adverse effect on testis size or fertility (24, 25), in contrast to what has been observed for high doses. The physiological effects of endogenous estrogens during fetal and neonatal life are unknown, and the hypothesis that endogenous estrogens are involved in regulating fetal steroidogenesis has yet to be proven. Therefore, it is unclear whether small changes in endogenous concentrations due to the presence of environmental xenoestrogens actually have an effect on the development of fetal Leydig cells.
We investigated the role of endogenous estrogens in the development of fetal Leydig cells, using mice with inactivated estrogen receptors (ERs) or ? (26). We previously investigated the effect of ER inactivation on fetal and neonatal testicular gametogenesis. We found that the number of gonocytes per testis is increased if ER? is inactivated but unaffected if ER is inactivated (27). In contrast, we found that ER? inactivation affected neither the number of fetal Leydig cells nor basal and LH-stimulated testosterone production (27). Therefore, in this study, we investigated the ER-mediated action of estrogens on steroidogenesis by studying Leydig cell development and functions in the ER–/– mouse lineage during fetal and neonatal life.
Materials and Methods
Animals
Mice were housed under controlled photoperiod conditions (lights on 0800–2000 h) and were supplied with tap water ad libitum and standard commercial feed (R03, Safe, Epinay-sur-Orge, France) in which the protein source was soy and yeast based. Mice heterozygous for ER (ER+/–) and ER? (ER?+/–) were produced by Dupont et al. (26) and generously provided by Prof. P. Chambon (Institut de Génétique et Biologie Moléculaire et Cellulaire, Illkirch, France). Exon 3 of these genes, encoding the first zinc finger of the DNA binding domain, was targeted for the disruption. These mice have been backcrossed at least 10 times with C57BL/6 mice to establish a C57BL/6 genetic background.
We generated mice homozygous for ER (ER–/–) by caging heterozygous males with heterozygous females for the night. The day after such overnight mating was counted as 0.5 dpc. Natural birth occurred on fetal d 19.5, which was counted as 0 d postpartum (dpp). For one experiment (see Fig. 3), ER double heterozygous mice (ER+/–/ER?+/–) obtained from breeding ER+/– females with ER?–/– males (26) were inbred to generate mutant mice homozygous for ER and ER? disruption designated as ER?KO.
FIG. 3. Ex vivo LH-stimulated testosterone secretion by neonatal testes of mice lacking the ER and ER? genes (ER?KO). Testes from 3-d-old neonates with various ER status (ER+/+, ER+/– and ER–/–) and various ER? status (ER?+/+, ER?+/–, and ER?–/–) were collected and cut into four pieces. One testis was incubated for 2 h in PBS supplemented with 100 ng/ml oLH (+LH). The amount of testosterone secreted into the medium was determined by RIA. Values are means ± SEM and the number of animal is in parentheses under each column. *, P < 0.05 vs. ER+/+ in each class of ER? inactivation in ANOVA (Tukey Kramer’s test).
Pregnant mice were anesthetized on gestational d 13.5 by the ip injection of 4 mg/100 g sodium pentobarbital (Sanofi, Libourne, France), and the fetuses were rapidly removed from the uterus. Fetuses were dissected under a binocular microscope, their sex was determined on the basis of gonad morphology, and the testes were collected. Male neonates were killed by decapitation on postnatal d 2 or 3, and their testes were immediately removed. All the animals were genotyped by PCR of biopsy DNA as previously described (26). All animal studies were conducted in accordance with the Guide for Care and Use of Laboratory Animals (National Institutes of Health Guide).
Chemicals and solutions
The culture medium was Ham’s F12/DMEM [1:1 (vol/vol); Life Technologies, Inc. (Grand Island, NY) supplemented with 80 μg/ml gentamicin (Gentalline Schering-Plough, Levallois-Perret, France). Ovine LH (oLH; NIH.LH S19; 1.01 IU/mg) was donated by Dr. Parlow (National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD). DES was purchased from Sigma (St. Louis, MO). A stock solution (1 mM) was made up in ethanol and diluted in culture medium (1 μM for use).
Ex vivo incubation
Immediately after their removal, testes from 2- or 3-d-old mice were cut into four pieces and incubated in 500 μl PBS or PBS supplemented with 100 ng/ml oLH for 2 h at 37 C with shaking. For each animal, one testis was incubated with PBS, and the other was incubated with PBS supplemented with oLH. After incubations, all media were kept at –20 C until the assay.
Organ culture
Testes were cultured on Millipore (Bedford, MA) filters (pore size 0.45 μm) as previously described (28). Briefly, intact 13.5-dpc fetal testes were placed on 10-mm-diameter Millipore filters. Testes from 2-d-old neonates were cut into six pieces, and all the pieces from the same testis were placed on a 25-mm Millipore filter. The filters were floated on 0.4 (13.5 dpc) or 1.5 (2 dpp) ml of culture medium in tissue culture dishes and incubated at 37 C in an humidified atmosphere containing 95% air/5% CO2 for 72 h. The medium was changed every 24 h. We added 100 ng/ml oLH to all the media for the last 3 h of culture (72–75 h). For cellular analysis, the whole explant was fixed for 2 h in Bouin’s fluid. All the media were kept at –20 C until the assay. The effect of ER inactivation was measured by comparing wild-type testes with ER+/– and ER–/– testes. The response to DES was measured by comparing one testis cultured in control medium with the other testis from the same animal cultured in medium supplemented with 1 μM DES.
Morphometric characteristics of fetal Leydig cells
Cell counting.
The method was previously described for rat fetal testis (29). The testes from mice killed on fetal d 13.5 or on postnatal d 2 and testes explanted on d 13.5 of gestation after organ culture were fixed in Bouin’s fluid for 2 h, embedded in paraffin, and cut into 5-μm sections. The Leydig cells were identified by immunocytochemical detection of 3?-hydroxysteroid dehydrogenase (3?HSD), using an antibody provided by Dr. G. Defaye (Grenoble, France). This enzyme is known for not being regulated by numerous factors and particularly by estrogens in fetal Leydig cells (20). Immunostaining was performed with the Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA). Leydig cells were counted on one section in 10 for fetal stage and on one section in 20 for neonates. The Abercrombie formula (30) was used to correct for double counting resulting from the appearance of a single cell in two successive sections. For this, mean nuclear diameter was determined for each testis from at least 100 random determinations with a computerized video micrometer (Histolab, Microvision Instruments, Evry, France). All counts were done by an investigator blind to the treatment.
Cell size.
Total Leydig cell areas were measured on one section in 20 with a computerized video densitometer (Histolab). Each area was divided by the corresponding number of Leydig cell nuclei counted on each section, to give the mean Leydig cell area per section. The mean of these values corresponds to the mean Leydig cell area per testis.
RNA extraction and expression analysis by RT-PCR
Real-time PCR was used to study expression of steroidogenic genes (StAR, P450scc, and P450c17) in testes from mice killed on postnatal d 2, using the TaqMan PCR method (31). Total RNA was extracted from one whole testis with the RNeasy kit (QIAGEN, Courtaboeuf, France), and residual genomic DNA was eliminated by deoxyribonuclease treatment (DNAse set, QIAGEN). RNAs were quantified by measuring absorbance at 260 nm, and 1 μg of total testicular RNA was reverse-transcribed as previously described (32). The primers and probes used were assays on demand designed by Applied Biosystems (Courtaboeuf, France) (sequences not provided, P450c17, Mm00484040-m1; P450scc, Mm00490735-m1; and StAR, Mm00441558-m1). Real-time PCR was carried out in a final volume of 25 μl/well in 96-well plates. PCR reagents were purchased from Applied Biosystems. Each PCR well contained 20 ng cDNA, reaction buffer, each primer, and probe, as provided by the manufacturer. Reactions were carried out and fluorescence was detected on an ABI Prism 7000 apparatus (Applied Biosystems, Foster City, CA). Each sample was run in duplicate, and a control PCR was also carried out with RNA for each sample. Negative controls were run for every primer/probe combination. The reaction efficiency, determined by running different concentrations of cDNA (1, 5, 10, and 20 nM) of the same sample in each plate, was around 90%. The measured amount of each cDNA was normalized using an internal standard, ?-actin, from the same sample and was compared between the different genotypes.
Hormone assays
Testosterone.
The testosterone secreted into the medium was determined in duplicate by RIA, as previously described (33).
LH and FSH.
Blood was collected from 2-dpp neonates after decapitation. Plasma was recovered by centrifugation at 700x g for 10 min at 4 C and stored at –20 C until the assay. Plasma samples from two to four neonates were pooled. LH and FSH were determined by immunofluorometric assays, as previously described (34, 35).
Statistical analysis
The results are presented as means ± SEM. The statistical significance of the difference between the mean values for two different genotypes was evaluated using Student’s unpaired t test. The statistical significance of the difference between the mean values for the treated and untreated testes from the same fetus was evaluated with Student’s paired t test. One-way ANOVA was used for the comparison of data from more than two groups.
Results
Testosterone secretion in fetuses and neonates with inactivation of the ER gene
In our floating filter culture system, daily testosterone production remained stable during the 3 d of culture for both ages tested and was strongly increased by the addition of 100 ng/ml oLH for the last 3 h (Fig. 1). In this system, testosterone production was much higher in ER–/– animals of both ages than in the respective wild-type controls in basal conditions, throughout culture and after LH stimulation (Fig. 1). ER+/– animals behaved like wild-type animals on postnatal d 2 but produced amounts of testosterone intermediate between ER–/– and wild-type on fetal d 13.5 (Fig. 1). Because ER+/– testes have only half the amount of ER protein found in wild-type animals, this suggests that endogenous estrogens inhibit testosterone production in a dose-dependent manner in early fetal testis development.
FIG. 1. Effect of ER gene inactivation on in vitro testicular fetal and neonatal testosterone secretion. Testes were collected on d 13.5 of gestation and on postnatal d 2 from homozygous (ER–/–), heterozygous (ER+/–), and wild-type (ER+/+) litter mates and were cultured on floating Millipore filters for 75 h. The medium was changed every 24 h, and the testosterone content of the medium was measured by RIA. We added 100 ng/ml of oLH to the medium for the period from 72–75 h. Values are means ± SEM of 16–29 fetuses and 17–24 neonates. For each age, different letters indicate a significant difference (P < 0.05) between categories in an ANOVA test (Tukey Kramer’s test).
Ex vivo basal testosterone production levels have been shown to be strictly correlated with intratesticular testosterone content (36). This parameter can therefore be used to evaluate the in vivo steroidogenic activity of the testes. Furthermore, ex vivo acute testosterone response to LH gives the in vivo steroidogenic capacity of the testes with the contralateral testis (1). Both ex vivo basal and LH-stimulated testosterone secretion levels were increased for homozygous but not for heterozygous ER gene inactivation at 2 d of age compared with wild-type animals (Fig. 2).
FIG. 2. Effect of ER gene inactivation on ex vivo testosterone secretion by neonatal testes. Testes from homozygous (ER–/–), heterozygous (ER+/–), and wild-type (ER+/+) 2-d-old neonates were collected and cut into four pieces. One testis was incubated for 2 h in PBS (basal), and the other was incubated in PBS supplemented with 100 ng/ml oLH (+LH). The amount of testosterone secreted into the medium was determined by RIA. Values are means ± SEM of eight to 14 animals. *, P < 0.05 vs. wild-type in ANOVA (Tukey Kramer’s test).
Lastly, to check out the absence of ER? involvement, we measured ex vivo LH-stimulated testosterone secretion in ER?KO 3-d-old neonates (Fig. 3). Inside each of the three classes of ER? inactivation (ER?+/+, ER?+/–, and ER?–/–), heterozygous inactivation of ER resulted in a slight nonsignificant increase of testosterone production, whereas homozygous inactivation of this receptor significantly increased this production. Also, inside each of the three classes of ER inactivation (ER+/+, ER+/–, and ER–/–), testosterone production was the same whatever the state of ER? inactivation. This shows that ER? does not compensate even partially the effect of ER inactivation on testosterone secretion.
Morphometric analysis of the Leydig cells
The number of Leydig cells identified by 3?HSD immunostaining increased by 5-fold from fetal d 13.5 to postnatal d 2, and no difference was detected between ER–/– animals and their respective wild-type litter mates (Table 1). The Leydig cells considerably enlarged from fetal d 13.5 to postnatal d 2 in wild-type and ER–/– mice (Table 1; Fig. 4). These cells were significantly larger in ER–/– animals than in wild-type litter mates at both ages (Table 1). These changes in the size of fetal Leydig cells are due to changes in the cytoplasmic volume of the cells because the diameter of the nucleus was similar in the various genotypes and was similar in fetuses and in neonates (Table 1).
TABLE 1. Number and size of fetal Leydig cells in testes of 13.5-d-old fetuses and 2-d-old neonates with ER gene inactivation
FIG. 4. Identification of fetal Leydig cells. Testicular sections from wild-type (A and B) and ER–/– (C and D) mice fetuses on d 13.5 of gestation (A–C) and from 2-d-old neonates (B–D) were treated for the immunohistochemical detection of 3?HSD and counterstained with hematoxylin. Note the enlargement of the Leydig cells during development in wild type and ER–/–.
Expression of StAR and steroidogenic enzymes in ER–/– neonates
The differentiated function of the Leydig cells was evaluated in 2-d-old neonates by determining mRNA expression of StAR and steroidogenic enzymes, P450scc and P450c17 (Fig. 5). The expression of all studied mRNA in ER–/– testis were approximately double those in wild-type controls.
FIG. 5. Effect of ER gene inactivation on testicular expression of StAR, P450scc, and P450c17 mRNA in neonatal testes. Testicular RNA was extracted from homozygous (ER–/–) and wild-type (ER+/+) 2-d-old neonates and reverse-transcribed. Real-time PCR was used to measure cDNA levels. The levels of mRNA are expressed as a percentage of the wild-type mean ± SEM of six to eight animals. *, P < 0.05; **, P < 0.01 vs. wild-type in Student’s unpaired t test.
Plasma LH and FSH concentrations in neonates with inactivation of the ER gene
Plasma LH and FSH concentrations, determined at 2 d after birth, did not differ between ER–/– and wild-type litter mates (Table 2). Thus, the higher level of androgen biosynthesis observed in ER–/– neonates was not associated with a change in gonadotropin level.
TABLE 2. Plasma LH and FSH concentrations in 2-d-old neonates with ER gene inactivation
Effect of DES on in vitro testicular testosterone secretion
In organ culture, the presence of 1 μM DES in the culture medium for 3 d decreased both basal and LH-stimulated testosterone secretion in ER+/+ and ER+/– fetal testes explanted on d 13.5 of gestation and LH-stimulated testosterone secretion only in ER+/+ and ER+/– testes from 2-d-old neonates (Fig. 6). DES had no effect if testes from ER–/– fetuses and neonates were used. The inhibitory effect of DES was observed after as little as 24 h of culture with testes from ER+/+ and ER+/– fetuses (data not shown), whereas it was observed only on LH-stimulated testosterone secretion with testes from 2-d-old neonates. These results suggest that the fetal testis is more sensitive to DES than the neonatal testis.
FIG. 6. Effect of ER gene inactivation on in vitro testicular response to DES. Testes from homozygous (ER–/–), heterozygous (ER+/–), and wild-type (ER+/+) fetuses on d 13.5 of gestation, and 2-d-old neonates were cultured on floating Millipore filters for 75 h. One testis from each animal was cultured in control medium and the other in medium containing 1 μM DES. The media were changed every 24 h. All the media were supplemented with 100 ng/ml oLH from 72–75 h. The figure shows the amount of testosterone secreted into the medium from 48–72 (LH–) and from 72–75 (LH+) h. Values are means ± SEM of eight animals, and the number of animal is in parentheses under each column. **, P < 0.01; ***, P < 0.001 vs. control medium in Student’s paired t test.
After 3 d of culture, the number of Leydig cells per testis for wild-type testes explanted on d 13.5 of gestation was not significantly different in control and DES-treated testes (2996 ± 486 and 2831 ± 649 cell/testis, respectively; n = 5), showing that the decrease in testosterone production induced by DES was not due to a decrease in the number of Leydig cells.
Discussion
This study shows for the first time that endogenous estrogens physiologically inhibit fetal Leydig cell development during fetal and neonatal life because the differentiated function of this cell type in ER–/– fetuses is greater than that in wild-type fetuses on d 13.5 of gestation and 2-d-old neonates. Our study also demonstrated that estrogens exert this effect via ER. Indeed, the homologous recombination used here to disrupt ER (Cre-LoxP strategy) resulted in the complete elimination of ER protein, and no truncated RNA has been described (26), ruling out the possibility of an active variant of ER. Moreover, DES, which is known to be a potent estrogen and to bind ERs (37), inhibited testosterone production by wild-type testes in our organ culture model but had no effect on ER-deficient testes. In contrast, ER? is not involved because we previously showed that ER? gene inactivation affects neither testosterone production nor the number of fetal Leydig cells in 2-d-old neonates (27). Interestingly, the study on ER?KO neonates showed that inactivation of ER results in the same increase of testosterone production, whatever the expression of ER?. This shows that ER? does not compensate, even partially, the effect of ER inactivation. Lastly, our results suggest that the fetal testis is more sensitive to estrogens than the neonatal testis because in vitro, DES was effective on wild-type testes in basal condition as soon as 24 h of culture (data not shown) for testes explanted on d 13.5 of gestation, whereas it has no effect with testes from 2-d-old neonates; and the testosterone production of ER+/– fetuses is intermediate between ER–/– and wild type, whereas neonatal ER+/– testis behaves like the wild type.
The inhibition of steroidogenesis by DES observed in vitro here demonstrates that estrogens act directly on the fetal and neonatal testis. This, together with the observed biological effect of ER inactivation and lack of effect of ER? inactivation on testicular steroidogenesis (present results and 27), is consistent with the observation that ER is present in mouse fetal Leydig cells, whereas ER? is not detected (38). Estrogens may therefore have a direct effect on fetal Leydig cells.
We also show here for the first time that fetal Leydig cell function is altered in ER-deficient fetuses as early as fetal d 13.5, shortly after the initiation of steroidogenesis (39). At this age, the hypothalamopituitary system cannot be involved because LH is not detected before fetal d 16 in the mouse (6). Therefore, although the increase in testosterone production in the adult ER-deficient mouse can be attributed both to a direct effect of estrogens on the testis and to stimulation by high levels of LH (40, 41), our results show that, in the fetus, the ER-mediated inhibition of testicular steroidogenesis by estrogen results exclusively from a direct effect. In line with this finding, the ER-deficient 2-d-old neonates displayed an increase in testosterone secretion with no change in circulating LH level. This is consistent with the decrease in testosterone production on fetal d 19.5 with no change in pituitary LH concentration observed in rats exposed to DES during gestation (42). In addition, the lack of change in LH levels in ER-deficient 2-d-old neonates suggests that endogenous estrogens do not regulate LH secretion at this stage. ER is known to be expressed in the pituitary gland from fetal d 17.5 in the rat (43), but the ontogenesis of this receptor is unknown in the mouse hypothalamopituitary system.
Our study provides new insight into the mechanism of action of estrogens in fetal Leydig cell development. We observed no change in the number of Leydig cells in ER-deficient mice. Moreover, treatment for 3 d with 1 μM DES had no effect on the number of fetal Leydig cells in testes explanted on fetal d 13.5. In contrast, in vivo studies have shown that the treatment of pregnant mice with estrogens induces foci of Leydig cell hyperplasia in fetuses on d 16 and 18 of gestation (19, 44), suggesting a possible role of estrogens in regulating the differentiation of new fetal Leydig cells. In a previous study, treatment for 3 d with 4 μM DES in an organ culture system slightly decreased the number of Leydig cells in rat testes explanted on fetal d 14.5 dpc (17). The reasons for these discrepancies with our present results are unclear but may be due to differences in experimental approaches (in vitro vs. in vivo), doses or models (rat vs. mouse). Nevertheless, the results presented here, together with our previous findings that the number of fetal Leydig cell is not affected by ER? gene inactivation (27), suggest that estrogens at endogenous concentrations are not involved in the differentiation of mesenchymal cells to generate fetal Leydig cells in the mouse.
We observed a clear negative effect of endogenous estrogens on the activity and differentiated functions of each fetal Leydig cell analyzed at the morphometric, functional, and molecular levels. Inactivation of the ER gene leads to increases in Leydig cell cytoplasmic volume, basal and LH-stimulated testosterone production, and mRNA levels for StAR, P450c17, and P450scc. These increases in mRNA levels are consistent with other reports showing that treatment with chemicals with estrogenic activity on fetal d 11.5 and 15.5 reduces the amount of P450c17 mRNA, protein, and activity in fetuses on d 17.5 of gestation (20). The molecular mechanism by which estrogens affect the transcription of the StAR, P450c17, and P450scc genes is largely unknown. Estrogens may act directly on the promoters of these genes via a classical mode of action because these sequences contain estrogen response elements (EREs) (Dragon ERE finder version 2) (45). It is also possible that estrogens act indirectly, by interacting with other non-ERE sites (46) or by regulating other transcription factors known to regulate the expression of steroidogenic enzyme genes. For example, estrogen treatment has been shown to decrease levels of mRNA for SF-1 (47), a transcription factor known to regulate the expression of P450 enzyme genes (48).
Little is known about the amount and origin of endogenous estrogens in fetal and neonatal mouse testes. We previously reported an intratesticular estradiol concentration of 4 nM in testes from 2-d-old neonates (27). P450arom, the enzyme that catalyzes the conversion of testosterone to estradiol, is detected from fetal d 17 in the mouse fetal testis (39), suggesting that total estrogen content changes throughout the developmental period considered here. On fetal d 13.5, the testis is not able to produce estradiol by itself, and the mother is probably an essential source of this hormone because the mouse placenta is not involved in the de novo synthesis of steroids during the second half of pregnancy (49). However, estrogens bind to extracellular carrier proteins and are conjugated and metabolized to generate inactive forms (50). For example, in mice, the -fetoprotein produced throughout development in the visceral endoderm of the extra-embryonic yolk sac and in the fetal liver and gut (51) can bind the estrogens synthesized by the mother, thereby increasing estrogen concentration in the fetus. It is therefore difficult to evaluate the actual concentration of estrogens acting on fetal Leydig cells in vivo. Nevertheless, our results show that this steroid is certainly present in the fetal testis and is physiologically efficient in the regulation of steroidogenesis.
In conclusion, this study demonstrates for the first time an ER-mediated in vivo inhibitory effect of endogenous estrogens on the development of testicular steroidogenesis during fetal and neonatal development in the mouse. Because androgen production is one of the key mechanisms underlying male testicular differentiation, these findings provide important insight into the development of the male reproductive tract. They also support the hypothesis that fetal and neonatal exposure to environmental xenoestrogens could impair the masculinization of the male urogenital system and male fertility in adulthood.
Acknowledgments
We thank Prof. P. Chambon and A. Krust (Institut de Génétique et Biologie Moléculaire et Cellulaire, Illkirch, France) for providing transgenic mice and Dr. G. Defaye (Institut National de la Santé et de la Recherche Médicale, Grenoble, France) for providing the anti-3?HSD antibody. We also thank C. Joubert, P. Flament, and V. Neuville (Commissariat à l’Energie Atomique, Fontenay aux Roses, France) for caring for the animals.
References
Saez JM 1994 Leydig cells: endocrine, paracrine and autocrine regulation. Endocr Rev 15:574–626
Habert R, Lejeune H, Saez JM 2001 Origin, differentiation and regulation of fetal and adult Leydig cells. Mol Cell Endocrinol 179:47–74
Jost A 1970 Hormonal factors in the sex differentiation of the mammalian foetus. Philos Trans R Soc Lond B Biol Sci 259:119–130
Vergouwen RP, Jacobs SG, Huiskamp R, Davids JA, de Rooij DG 1991 Proliferative activity of gonocytes, Sertoli cells and interstitial cells during testicular development in mice. J Reprod Fertil 93:233–243
Pakarinen P, Vihko K, Voutilainen R, Huhtaniemi I 1990 Differential response of Luteinizing Hormone receptor and steroidogenic enzyme gene expression to human chorionic gonadotropin stimulation in the neonatal and adult rat testis. Endocrinology 127:2469–2474
O’Shaughnessy P, Baker U, Sohnius U, Haavisto A-M, Charlton H, Huhtaniemi I 1998 Fetal development of Leydig cell activity in the mouse is independent of pituitary gonadotroph function. Endocrinology 139:1141–1146
Migrenne S, Pairault C, Racine C, Livera G, Géloso A, Habert R 2001 LH-dependent activity and LH-independent differentiation of rat fetal Leydig cells. Mol Cell Endocrinol 172:193–202
O’Shaughnessy PJ, Johnston H, Willerton L, Baker PJ 2002 Failure of normal adult Leydig cell development in androgen-receptor-deficient mice. J Cell Sci 115:3491–3496
Sharpe RM 2003 The "oestrogen hypothesis": where do we stand now? Int J Androl 26:2–15
Sharpe RM, Irvine DS 2004 How strong is the evidence of a link between environmental chemicals and adverse effects on human reproductive health? BMJ 328:447–451
Toppari J, Larsen J, Christiansen P, Giwereman A, Grandjean P, Guillette L, Jegou B, Jensen T, Jouannet P, Keiding N, Leffers H, McLachlan J, Meyar O, Muller J, Rajpert de Meyrs E, Scheike T, Sharpe R, Sumpter J, Skakkebaek N 1996 Male reproductive health and environmental xenoestrogens. Environ Health Perspect 104:741–803
Danzo B 1998 The effects of environmental hormones on reproduction. Cell Mol Life Sci 54:1249–1264
Skakkebaek NE, Rajpert-De Meyts E, Main KM 2001 Testicular dysgenesis syndrome: an increasingly common developmental disorder with environmental aspects. Hum Reprod 16:972–978
Kubota Y, Temelcos C, Bathgate RA, Smith KJ, Scott D, Zhao C, Hutson JM 2002 The role of insulin 3, testosterone, Mullerian inhibiting substance and relaxin in rat gubernacular growth. Mol Hum Reprod 8:900–905
Ivell R, Hartung S 2003 The molecular basis of cryptorchidism. Mol Hum Reprod 9:175–181
Tsai-Morris CH, Knox G, Luna S, Dufau ML 1986 Acquisition of estradiol-mediated regulatory mechanism of steroidogenesis in cultured fetal rat Leydig cells. J Biol Chem 261:3471–3474
Lassurguere J, Livera G, Habert R, Jegou B 2003 Time- and dose-related effects of estradiol and diethylstilbestrol on the morphology and function of the fetal rat testis in culture. Toxicol Sci 73:160–169
Grocock CA, Charlton HM, Pike MC 1988 Role of the fetal pituitary in cryptorchidism induced by exogenous maternal oestrogen during pregnancy in mice. J Reprod Fertil 83:295–300
Perez-Martinez C, Garcia-Iglesias M, Ferreras-Estrada M, Bravo-Moral A, Espinosa-Alvarez J, Escudero-Diez A 1996 Effects of in-utero exposure to zeranol or diethylstilbestrol on morphological development of the fetal testis in mice. J Comp Path 114:407–418
Majdic G, Sharpe RM, O’Shaughnessy PJ, Saunders PT 1996 Expression of cytochrome P450 17-hydroxylase/C17–20 lyase in the fetal rat testis is reduced by maternal exposure to exogenous estrogens. Endocrinology 137:1063–1070
Gill WB, Schumacher GF, Bibbo M, Straus 2nd FH, Schoenberg HW 1979 Association of diethylstilbestrol exposure in utero with cryptorchidism, testicular hypoplasia and semen abnormalities. J Urol 122:36–39
Newbold R 1995 Cellular and molecular effects of developmental exposure to diethylstilbestrol: implications for other environmental estrogens. Environ Health Perspect 103(Suppl 7):83–87
Klip H, Verloop J, van Gool JD, Koster ME, Burger CW, van Leeuwen FE 2002 Hypospadias in sons of women exposed to diethylstilbestrol in utero: a cohort study. Lancet 359:1102–1107
Atanassova N, McKinnell C, Turner K, Walker M, Fisher J, Morley M, Millar M, Groome N, Sharpe R 2000 Comparative effects of neonatal exposure of male rats to potent and weak (environmental) estrogens on spermatogenesis at puberty and the relationship to adult testis size and fertility: evidence for stimulatory effects of low estrogen levels. Endocrinology 141:3898–3908
Fielden MR, Samy SM, Chou KC, Zacharewski TR 2003 Effect of human dietary exposure levels of genistein during gestation and lactation on long-term reproductive development and sperm quality in mice. Food Chem Toxicol 41:447–454
Dupont S, Krust A, Gansmuller A, Dierich A, Chambon P, Mark M 2000 Effect of single and compound knockouts of estrogen receptors (ER) and ? (ER?) on mouse reproductive phenotypes. Development 127:4277–4291
Delbes G, Levacher C, Pairault C, Racine C, Duquenne C, Krust A, Habert R 2004 Estrogen receptor {?}-mediated inhibition of male germ cell line development in mice by endogenous estrogens during perinatal life. Endocrinology 145:3395–3403
Habert R, Devif I, Gangnerau MN, Lecerf L 1991 Ontogenesis of the in vitro response of rat testis to gonadotropin-releasing hormone. Mol Cell Endocrinol 82:199–206
Livera G, Rouiller-Fabre V, Durand P, Habert R 2000 Multiple effects of retinoids on the development of Sertoli, germ and Leydig cells of fetal and neonatal rat testis in culture. Biol Reprod 62:1303–1314
Abercrombie M 1946 Estimation of nuclear population from microtome sections. Anat Rec 94:238–248
Bustin SA 2000 Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. J Mol Endocrinol 25:169–193
Migrenne S, Racine C, Guillou F, Habert R 2003 Pituitary hormones inhibit the function and differentiation of fetal Sertoli cells. Endocrinology 144:2617–2622
Habert R, Picon R 1984 Testosterone, dihydrotestosterone and estradiol 17? levels in maternal and fetal plasma and in fetal testes in the rat. J Steroid Biochem 21:193–198
Zhang FP, Poutanen M, Wilbertz J, Huhtaniemi I 2001 Normal prenatal but arrested postnatal sexual development of luteinizing hormone receptor knockout (LuRKO) mice. Mol Endocrinol 15:172–183
Rulli SB, Zitta K, Calandra RS, Campo S 2003 Effect of dihydrotestosterone on pituitary follicle-stimulating hormone isoforms in adult male rats treated with a gonadotropin-releasing hormone antagonist. Neuroendocrinology 78:280–286
Habert R, Picon R 1982 Control of testicular steroidogenesis in fetal rat: effect of decapitation on testosterone and plasma luteinizing hormone-like activity. Acta Endocrinol 99:466–473
Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA 1997 Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors and ?. Endocrinology 138:863–870
Jefferson W, Couse J, Banks E, Korach K, Newbold R 2000 Expression of estrogen receptor ? is developmentally regulated in reproductive tissues of male and female mice. Biol Reprod 62:310–317
Greco TL, Payne AH 1994 Ontogeny of expression of the genes for steroidogenic enzymes P450 side-chain cleavage, 3 ?-hydroxysteroid dehydrogenase, P450 17 -hydroxylase/C17–20 lyase, and P450 aromatase in fetal mouse gonads. Endocrinology 135:262–268
Eddy EM, Washburn TF, Bunch DO, Goulding EH, Gladen BC, Lubahn BD, Korach KS 1996 Targeted disruption of the estrogen receptor gene in male mice causes alteration of spermatogenesis and infertility. Endocrinology 137:4796–4805
Akingbemi BT, Ge R, Rosenfeld CS, Newton LG, Hardy DO, Catterall JF, Lubahn DB, Korach KS, Hardy MP 2003 Estrogen receptor- gene deficiency enhances androgen biosynthesis in the mouse Leydig cell. Endocrinology 144:84–93
Haavisto T, Nurmela K, Pohjanvirta R, Huuskonen H, El-Gehani F, Paranko J 2001 Prenatal testosterone and luteinizing hormone levels in male rats exposed during pregnancy to 2,3,7,8-tetrachlorodibenzo-p-dioxin and diethylstilbestrol. Mol Cell Endocrinol 178:169–179
Nishihara E, Nagayama Y, Inoue S, Hiroi H, Muramatsu M, Yamashita S, Koji T 2000 Ontogenetic changes in the expression of estrogen receptor and ? in rat pituitary gland detected by immunohistochemistry. Endocrinology 141:615–620
Yasuda Y, Kihara T, Tanimura T 1985 Effect of ethinyl estradiol on the differentiation of mouse fetal testis. Teratology 32:113–118
Bajic VB, Tan SL, Chong A, Tang S, Strom A, Gustafsson JA, Lin CY, Liu ET 2003 Dragon ERE Finder version 2: a tool for accurate detection and analysis of estrogen response elements in vertebrate genomes. Nucleic Acids Res 31:3605–3607
Kushner PJ, Agard D, Feng WJ, Lopez G, Schiau A, Uht R, Webb P, Greene G 2000 Oestrogen receptor function at classical and alternative response elements. Novartis Found Symp 230:20–26; discussion 27–40
Saunders P, Majdic G, Parte P, Millar M, Fisher J, Turner K, Sharpe R 1997 Fetal and perinatal influence of xenoestrogens on testis gene expression. In: Holstein IA, ed. The fate of the male germ cells. New York: Plenum Press; 99–110
Hatano O, Takayama K, Imai T, Waterman MR, Takakusu A, Omura T, Morohashi K 1994 Sex-dependent expression of a transcription factor, Ad4BP, regulating steroidogenic P-450 genes in the gonads during prenatal and postnatal rat development. Development 120:2787–2797
Arensburg J, Payne AH, Orly J 1999 Expression of steroidogenic genes in maternal and extraembryonic cells during early pregnancy in mice. Endocrinology 140:5220–5232
McLachlan JA, Newbold RR 1987 Estrogens and development. Environ Health Perspect 75:25–27
Tilghman SM, Belayew A 1982 Transcriptional control of the murine albumin/-fetoprotein locus during development. Proc Natl Acad Sci USA 79:5254–5257(Géraldine Delbès, Christi)
Address all correspondence and requests for reprints to: Dr. Christine Levacher, Institut National de la Santé et de la Recherche Médicale Unité 566/Commissariat à l’Energie Atomique/Université Paris 7, Denis Diderot, DSV/DRR/SEGG/LDFG, Batiment 5A, RDC, Route du Panorama, 92265 Fontenay aux Roses, France. E-mail: christine.levacher@cea.fr.
Abstract
It is now accepted that estrogens play a role in male fertility and that exposure to exogenous estrogens during fetal/neonatal life can lead to reproductive disorders in the male. However, the estrogen receptor (ER)-mediated processes involved in the regulation of male reproduction during fetal and neonatal development are still largely unclear. We previously reported that ER? deficiency affects gametogenesis in mice but changes neither the number nor the differentiated functions of fetal Leydig cells. We show here that ER-deficient mice (ER–/–) display higher levels of testicular testosterone secretion than wild-type mice from fetal d 13.5 onwards. This results from higher levels of steroidogenic activity per fetal Leydig cell, as indicated by the hypertrophy of these cells and the higher levels of mRNA for StAR, P450c17 and P450scc in the testis, for a similar number of Leydig cells. Because LH is not produced on fetal d 13.5 and because no change in plasma LH concentration was observed in 2-d-old ER-deficient mice, LH is probably not involved in the effects of estrogens on testicular steroidogenesis in fetal and early neonatal Leydig cells. Furthermore, inactivation of ER? did not change the effect of ER inactivation on steroidogenesis. Lastly, in an organ culture system, 1 μM diethylstilbestrol decreased the testosterone secretion of wild-type fetal and neonatal testes but not of ER–/– testes. Thus, this study shows that endogenous estrogens physiologically inhibit steroidogenesis via ER by acting directly on the testis early in fetal and neonatal development.
Introduction
TWO SUCCESSIVE POPULATIONS of Leydig cells arise during normal testicular development (reviewed in Refs. 1 and 2). In the mouse, the fetal population starts to appear about 12.5 d postconception (dpc) and is essential for the masculinization of the fetus (3). The second, adult population begins to differentiate 4 d after birth (4). The fetal Leydig cells differ from the adult population in morphology, physiology, and regulation (2, 5). They are not desensitized by LH and do not require LH for differentiation (2, 6, 7). Furthermore, mutation of the androgen receptor does not affect the development of fetal Leydig cells, whereas it does impair the development of adult Leydig cells (8).
The role played by estrogens in the development and functions of fetal Leydig cells has recently come to the forefront because it has been claimed that the occurrence of alterations in male reproductive function is linked to exposure to environmental pollutants. Indeed, increases in the frequency of male reproductive disorders have been observed in humans and wildlife in many countries over the last 50 yr (reviewed in Refs. 9 and 10). A decrease in sperm count and increases in the incidence of testicular cancer, cryptorchidism, and hypospadia have been reported (9). It is widely thought that all these disorders are caused by an increase in the concentration of xenobiotics, and of xenoestrogens in particular, in the environment and in food (11, 12). These disorders, which are now collectively considered as testicular dysgenesis syndrome (13), may result from the impairment of testicular programing during fetal and neonatal life. This is particularly evident for hypospadia and cryptorchidism because the masculinization of external genitalia depends on the production of testosterone by fetal Leydig cells (3), and the descent of the testis is induced by the secretion of insulin-like factor-3 (INSL3) and testosterone by fetal Leydig cells (14, 15).
It has been demonstrated that high doses of estrogens alter fetal Leydig cell function. Dufau’s group (16) reported that estrogens inhibit testosterone production in cultured dispersed rat fetal Leydig cells. We recently showed, in an organotypic culture model, that estradiol and diethylstilbestrol (DES) decrease the number and differentiated functions of Leydig cells in rat testes explanted at 14.5 dpc (17). Furthermore, the exposure of laboratory animals to high doses of exogenous estrogenic compounds during fetal or neonatal life leads to an increase in the frequency of hypospadias and cryptorchidism (9, 13, 18). In this estrogen-treated pregnant rodent model, estradiol has been shown to affect the differentiation of fetal Leydig cells (19, 20). Lastly, the male offspring of women treated with DES during pregnancy have a higher incidence of cryptorchidism and hypospadias (21, 22, 23).
However, most of the deleterious effects of estrogens observed in the experiments and clinical cases described above were obtained with pharmacological doses of estrogens. There is currently no evidence that low doses equivalent to the level of human exposure to environmental estrogens have any effect. A few studies have investigated the effects of exposure to low doses of estrogens. Neonatal exposure to low doses of DES or genistein, a phytoestrogen, has no long-term adverse effect on testis size or fertility (24, 25), in contrast to what has been observed for high doses. The physiological effects of endogenous estrogens during fetal and neonatal life are unknown, and the hypothesis that endogenous estrogens are involved in regulating fetal steroidogenesis has yet to be proven. Therefore, it is unclear whether small changes in endogenous concentrations due to the presence of environmental xenoestrogens actually have an effect on the development of fetal Leydig cells.
We investigated the role of endogenous estrogens in the development of fetal Leydig cells, using mice with inactivated estrogen receptors (ERs) or ? (26). We previously investigated the effect of ER inactivation on fetal and neonatal testicular gametogenesis. We found that the number of gonocytes per testis is increased if ER? is inactivated but unaffected if ER is inactivated (27). In contrast, we found that ER? inactivation affected neither the number of fetal Leydig cells nor basal and LH-stimulated testosterone production (27). Therefore, in this study, we investigated the ER-mediated action of estrogens on steroidogenesis by studying Leydig cell development and functions in the ER–/– mouse lineage during fetal and neonatal life.
Materials and Methods
Animals
Mice were housed under controlled photoperiod conditions (lights on 0800–2000 h) and were supplied with tap water ad libitum and standard commercial feed (R03, Safe, Epinay-sur-Orge, France) in which the protein source was soy and yeast based. Mice heterozygous for ER (ER+/–) and ER? (ER?+/–) were produced by Dupont et al. (26) and generously provided by Prof. P. Chambon (Institut de Génétique et Biologie Moléculaire et Cellulaire, Illkirch, France). Exon 3 of these genes, encoding the first zinc finger of the DNA binding domain, was targeted for the disruption. These mice have been backcrossed at least 10 times with C57BL/6 mice to establish a C57BL/6 genetic background.
We generated mice homozygous for ER (ER–/–) by caging heterozygous males with heterozygous females for the night. The day after such overnight mating was counted as 0.5 dpc. Natural birth occurred on fetal d 19.5, which was counted as 0 d postpartum (dpp). For one experiment (see Fig. 3), ER double heterozygous mice (ER+/–/ER?+/–) obtained from breeding ER+/– females with ER?–/– males (26) were inbred to generate mutant mice homozygous for ER and ER? disruption designated as ER?KO.
FIG. 3. Ex vivo LH-stimulated testosterone secretion by neonatal testes of mice lacking the ER and ER? genes (ER?KO). Testes from 3-d-old neonates with various ER status (ER+/+, ER+/– and ER–/–) and various ER? status (ER?+/+, ER?+/–, and ER?–/–) were collected and cut into four pieces. One testis was incubated for 2 h in PBS supplemented with 100 ng/ml oLH (+LH). The amount of testosterone secreted into the medium was determined by RIA. Values are means ± SEM and the number of animal is in parentheses under each column. *, P < 0.05 vs. ER+/+ in each class of ER? inactivation in ANOVA (Tukey Kramer’s test).
Pregnant mice were anesthetized on gestational d 13.5 by the ip injection of 4 mg/100 g sodium pentobarbital (Sanofi, Libourne, France), and the fetuses were rapidly removed from the uterus. Fetuses were dissected under a binocular microscope, their sex was determined on the basis of gonad morphology, and the testes were collected. Male neonates were killed by decapitation on postnatal d 2 or 3, and their testes were immediately removed. All the animals were genotyped by PCR of biopsy DNA as previously described (26). All animal studies were conducted in accordance with the Guide for Care and Use of Laboratory Animals (National Institutes of Health Guide).
Chemicals and solutions
The culture medium was Ham’s F12/DMEM [1:1 (vol/vol); Life Technologies, Inc. (Grand Island, NY) supplemented with 80 μg/ml gentamicin (Gentalline Schering-Plough, Levallois-Perret, France). Ovine LH (oLH; NIH.LH S19; 1.01 IU/mg) was donated by Dr. Parlow (National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD). DES was purchased from Sigma (St. Louis, MO). A stock solution (1 mM) was made up in ethanol and diluted in culture medium (1 μM for use).
Ex vivo incubation
Immediately after their removal, testes from 2- or 3-d-old mice were cut into four pieces and incubated in 500 μl PBS or PBS supplemented with 100 ng/ml oLH for 2 h at 37 C with shaking. For each animal, one testis was incubated with PBS, and the other was incubated with PBS supplemented with oLH. After incubations, all media were kept at –20 C until the assay.
Organ culture
Testes were cultured on Millipore (Bedford, MA) filters (pore size 0.45 μm) as previously described (28). Briefly, intact 13.5-dpc fetal testes were placed on 10-mm-diameter Millipore filters. Testes from 2-d-old neonates were cut into six pieces, and all the pieces from the same testis were placed on a 25-mm Millipore filter. The filters were floated on 0.4 (13.5 dpc) or 1.5 (2 dpp) ml of culture medium in tissue culture dishes and incubated at 37 C in an humidified atmosphere containing 95% air/5% CO2 for 72 h. The medium was changed every 24 h. We added 100 ng/ml oLH to all the media for the last 3 h of culture (72–75 h). For cellular analysis, the whole explant was fixed for 2 h in Bouin’s fluid. All the media were kept at –20 C until the assay. The effect of ER inactivation was measured by comparing wild-type testes with ER+/– and ER–/– testes. The response to DES was measured by comparing one testis cultured in control medium with the other testis from the same animal cultured in medium supplemented with 1 μM DES.
Morphometric characteristics of fetal Leydig cells
Cell counting.
The method was previously described for rat fetal testis (29). The testes from mice killed on fetal d 13.5 or on postnatal d 2 and testes explanted on d 13.5 of gestation after organ culture were fixed in Bouin’s fluid for 2 h, embedded in paraffin, and cut into 5-μm sections. The Leydig cells were identified by immunocytochemical detection of 3?-hydroxysteroid dehydrogenase (3?HSD), using an antibody provided by Dr. G. Defaye (Grenoble, France). This enzyme is known for not being regulated by numerous factors and particularly by estrogens in fetal Leydig cells (20). Immunostaining was performed with the Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA). Leydig cells were counted on one section in 10 for fetal stage and on one section in 20 for neonates. The Abercrombie formula (30) was used to correct for double counting resulting from the appearance of a single cell in two successive sections. For this, mean nuclear diameter was determined for each testis from at least 100 random determinations with a computerized video micrometer (Histolab, Microvision Instruments, Evry, France). All counts were done by an investigator blind to the treatment.
Cell size.
Total Leydig cell areas were measured on one section in 20 with a computerized video densitometer (Histolab). Each area was divided by the corresponding number of Leydig cell nuclei counted on each section, to give the mean Leydig cell area per section. The mean of these values corresponds to the mean Leydig cell area per testis.
RNA extraction and expression analysis by RT-PCR
Real-time PCR was used to study expression of steroidogenic genes (StAR, P450scc, and P450c17) in testes from mice killed on postnatal d 2, using the TaqMan PCR method (31). Total RNA was extracted from one whole testis with the RNeasy kit (QIAGEN, Courtaboeuf, France), and residual genomic DNA was eliminated by deoxyribonuclease treatment (DNAse set, QIAGEN). RNAs were quantified by measuring absorbance at 260 nm, and 1 μg of total testicular RNA was reverse-transcribed as previously described (32). The primers and probes used were assays on demand designed by Applied Biosystems (Courtaboeuf, France) (sequences not provided, P450c17, Mm00484040-m1; P450scc, Mm00490735-m1; and StAR, Mm00441558-m1). Real-time PCR was carried out in a final volume of 25 μl/well in 96-well plates. PCR reagents were purchased from Applied Biosystems. Each PCR well contained 20 ng cDNA, reaction buffer, each primer, and probe, as provided by the manufacturer. Reactions were carried out and fluorescence was detected on an ABI Prism 7000 apparatus (Applied Biosystems, Foster City, CA). Each sample was run in duplicate, and a control PCR was also carried out with RNA for each sample. Negative controls were run for every primer/probe combination. The reaction efficiency, determined by running different concentrations of cDNA (1, 5, 10, and 20 nM) of the same sample in each plate, was around 90%. The measured amount of each cDNA was normalized using an internal standard, ?-actin, from the same sample and was compared between the different genotypes.
Hormone assays
Testosterone.
The testosterone secreted into the medium was determined in duplicate by RIA, as previously described (33).
LH and FSH.
Blood was collected from 2-dpp neonates after decapitation. Plasma was recovered by centrifugation at 700x g for 10 min at 4 C and stored at –20 C until the assay. Plasma samples from two to four neonates were pooled. LH and FSH were determined by immunofluorometric assays, as previously described (34, 35).
Statistical analysis
The results are presented as means ± SEM. The statistical significance of the difference between the mean values for two different genotypes was evaluated using Student’s unpaired t test. The statistical significance of the difference between the mean values for the treated and untreated testes from the same fetus was evaluated with Student’s paired t test. One-way ANOVA was used for the comparison of data from more than two groups.
Results
Testosterone secretion in fetuses and neonates with inactivation of the ER gene
In our floating filter culture system, daily testosterone production remained stable during the 3 d of culture for both ages tested and was strongly increased by the addition of 100 ng/ml oLH for the last 3 h (Fig. 1). In this system, testosterone production was much higher in ER–/– animals of both ages than in the respective wild-type controls in basal conditions, throughout culture and after LH stimulation (Fig. 1). ER+/– animals behaved like wild-type animals on postnatal d 2 but produced amounts of testosterone intermediate between ER–/– and wild-type on fetal d 13.5 (Fig. 1). Because ER+/– testes have only half the amount of ER protein found in wild-type animals, this suggests that endogenous estrogens inhibit testosterone production in a dose-dependent manner in early fetal testis development.
FIG. 1. Effect of ER gene inactivation on in vitro testicular fetal and neonatal testosterone secretion. Testes were collected on d 13.5 of gestation and on postnatal d 2 from homozygous (ER–/–), heterozygous (ER+/–), and wild-type (ER+/+) litter mates and were cultured on floating Millipore filters for 75 h. The medium was changed every 24 h, and the testosterone content of the medium was measured by RIA. We added 100 ng/ml of oLH to the medium for the period from 72–75 h. Values are means ± SEM of 16–29 fetuses and 17–24 neonates. For each age, different letters indicate a significant difference (P < 0.05) between categories in an ANOVA test (Tukey Kramer’s test).
Ex vivo basal testosterone production levels have been shown to be strictly correlated with intratesticular testosterone content (36). This parameter can therefore be used to evaluate the in vivo steroidogenic activity of the testes. Furthermore, ex vivo acute testosterone response to LH gives the in vivo steroidogenic capacity of the testes with the contralateral testis (1). Both ex vivo basal and LH-stimulated testosterone secretion levels were increased for homozygous but not for heterozygous ER gene inactivation at 2 d of age compared with wild-type animals (Fig. 2).
FIG. 2. Effect of ER gene inactivation on ex vivo testosterone secretion by neonatal testes. Testes from homozygous (ER–/–), heterozygous (ER+/–), and wild-type (ER+/+) 2-d-old neonates were collected and cut into four pieces. One testis was incubated for 2 h in PBS (basal), and the other was incubated in PBS supplemented with 100 ng/ml oLH (+LH). The amount of testosterone secreted into the medium was determined by RIA. Values are means ± SEM of eight to 14 animals. *, P < 0.05 vs. wild-type in ANOVA (Tukey Kramer’s test).
Lastly, to check out the absence of ER? involvement, we measured ex vivo LH-stimulated testosterone secretion in ER?KO 3-d-old neonates (Fig. 3). Inside each of the three classes of ER? inactivation (ER?+/+, ER?+/–, and ER?–/–), heterozygous inactivation of ER resulted in a slight nonsignificant increase of testosterone production, whereas homozygous inactivation of this receptor significantly increased this production. Also, inside each of the three classes of ER inactivation (ER+/+, ER+/–, and ER–/–), testosterone production was the same whatever the state of ER? inactivation. This shows that ER? does not compensate even partially the effect of ER inactivation on testosterone secretion.
Morphometric analysis of the Leydig cells
The number of Leydig cells identified by 3?HSD immunostaining increased by 5-fold from fetal d 13.5 to postnatal d 2, and no difference was detected between ER–/– animals and their respective wild-type litter mates (Table 1). The Leydig cells considerably enlarged from fetal d 13.5 to postnatal d 2 in wild-type and ER–/– mice (Table 1; Fig. 4). These cells were significantly larger in ER–/– animals than in wild-type litter mates at both ages (Table 1). These changes in the size of fetal Leydig cells are due to changes in the cytoplasmic volume of the cells because the diameter of the nucleus was similar in the various genotypes and was similar in fetuses and in neonates (Table 1).
TABLE 1. Number and size of fetal Leydig cells in testes of 13.5-d-old fetuses and 2-d-old neonates with ER gene inactivation
FIG. 4. Identification of fetal Leydig cells. Testicular sections from wild-type (A and B) and ER–/– (C and D) mice fetuses on d 13.5 of gestation (A–C) and from 2-d-old neonates (B–D) were treated for the immunohistochemical detection of 3?HSD and counterstained with hematoxylin. Note the enlargement of the Leydig cells during development in wild type and ER–/–.
Expression of StAR and steroidogenic enzymes in ER–/– neonates
The differentiated function of the Leydig cells was evaluated in 2-d-old neonates by determining mRNA expression of StAR and steroidogenic enzymes, P450scc and P450c17 (Fig. 5). The expression of all studied mRNA in ER–/– testis were approximately double those in wild-type controls.
FIG. 5. Effect of ER gene inactivation on testicular expression of StAR, P450scc, and P450c17 mRNA in neonatal testes. Testicular RNA was extracted from homozygous (ER–/–) and wild-type (ER+/+) 2-d-old neonates and reverse-transcribed. Real-time PCR was used to measure cDNA levels. The levels of mRNA are expressed as a percentage of the wild-type mean ± SEM of six to eight animals. *, P < 0.05; **, P < 0.01 vs. wild-type in Student’s unpaired t test.
Plasma LH and FSH concentrations in neonates with inactivation of the ER gene
Plasma LH and FSH concentrations, determined at 2 d after birth, did not differ between ER–/– and wild-type litter mates (Table 2). Thus, the higher level of androgen biosynthesis observed in ER–/– neonates was not associated with a change in gonadotropin level.
TABLE 2. Plasma LH and FSH concentrations in 2-d-old neonates with ER gene inactivation
Effect of DES on in vitro testicular testosterone secretion
In organ culture, the presence of 1 μM DES in the culture medium for 3 d decreased both basal and LH-stimulated testosterone secretion in ER+/+ and ER+/– fetal testes explanted on d 13.5 of gestation and LH-stimulated testosterone secretion only in ER+/+ and ER+/– testes from 2-d-old neonates (Fig. 6). DES had no effect if testes from ER–/– fetuses and neonates were used. The inhibitory effect of DES was observed after as little as 24 h of culture with testes from ER+/+ and ER+/– fetuses (data not shown), whereas it was observed only on LH-stimulated testosterone secretion with testes from 2-d-old neonates. These results suggest that the fetal testis is more sensitive to DES than the neonatal testis.
FIG. 6. Effect of ER gene inactivation on in vitro testicular response to DES. Testes from homozygous (ER–/–), heterozygous (ER+/–), and wild-type (ER+/+) fetuses on d 13.5 of gestation, and 2-d-old neonates were cultured on floating Millipore filters for 75 h. One testis from each animal was cultured in control medium and the other in medium containing 1 μM DES. The media were changed every 24 h. All the media were supplemented with 100 ng/ml oLH from 72–75 h. The figure shows the amount of testosterone secreted into the medium from 48–72 (LH–) and from 72–75 (LH+) h. Values are means ± SEM of eight animals, and the number of animal is in parentheses under each column. **, P < 0.01; ***, P < 0.001 vs. control medium in Student’s paired t test.
After 3 d of culture, the number of Leydig cells per testis for wild-type testes explanted on d 13.5 of gestation was not significantly different in control and DES-treated testes (2996 ± 486 and 2831 ± 649 cell/testis, respectively; n = 5), showing that the decrease in testosterone production induced by DES was not due to a decrease in the number of Leydig cells.
Discussion
This study shows for the first time that endogenous estrogens physiologically inhibit fetal Leydig cell development during fetal and neonatal life because the differentiated function of this cell type in ER–/– fetuses is greater than that in wild-type fetuses on d 13.5 of gestation and 2-d-old neonates. Our study also demonstrated that estrogens exert this effect via ER. Indeed, the homologous recombination used here to disrupt ER (Cre-LoxP strategy) resulted in the complete elimination of ER protein, and no truncated RNA has been described (26), ruling out the possibility of an active variant of ER. Moreover, DES, which is known to be a potent estrogen and to bind ERs (37), inhibited testosterone production by wild-type testes in our organ culture model but had no effect on ER-deficient testes. In contrast, ER? is not involved because we previously showed that ER? gene inactivation affects neither testosterone production nor the number of fetal Leydig cells in 2-d-old neonates (27). Interestingly, the study on ER?KO neonates showed that inactivation of ER results in the same increase of testosterone production, whatever the expression of ER?. This shows that ER? does not compensate, even partially, the effect of ER inactivation. Lastly, our results suggest that the fetal testis is more sensitive to estrogens than the neonatal testis because in vitro, DES was effective on wild-type testes in basal condition as soon as 24 h of culture (data not shown) for testes explanted on d 13.5 of gestation, whereas it has no effect with testes from 2-d-old neonates; and the testosterone production of ER+/– fetuses is intermediate between ER–/– and wild type, whereas neonatal ER+/– testis behaves like the wild type.
The inhibition of steroidogenesis by DES observed in vitro here demonstrates that estrogens act directly on the fetal and neonatal testis. This, together with the observed biological effect of ER inactivation and lack of effect of ER? inactivation on testicular steroidogenesis (present results and 27), is consistent with the observation that ER is present in mouse fetal Leydig cells, whereas ER? is not detected (38). Estrogens may therefore have a direct effect on fetal Leydig cells.
We also show here for the first time that fetal Leydig cell function is altered in ER-deficient fetuses as early as fetal d 13.5, shortly after the initiation of steroidogenesis (39). At this age, the hypothalamopituitary system cannot be involved because LH is not detected before fetal d 16 in the mouse (6). Therefore, although the increase in testosterone production in the adult ER-deficient mouse can be attributed both to a direct effect of estrogens on the testis and to stimulation by high levels of LH (40, 41), our results show that, in the fetus, the ER-mediated inhibition of testicular steroidogenesis by estrogen results exclusively from a direct effect. In line with this finding, the ER-deficient 2-d-old neonates displayed an increase in testosterone secretion with no change in circulating LH level. This is consistent with the decrease in testosterone production on fetal d 19.5 with no change in pituitary LH concentration observed in rats exposed to DES during gestation (42). In addition, the lack of change in LH levels in ER-deficient 2-d-old neonates suggests that endogenous estrogens do not regulate LH secretion at this stage. ER is known to be expressed in the pituitary gland from fetal d 17.5 in the rat (43), but the ontogenesis of this receptor is unknown in the mouse hypothalamopituitary system.
Our study provides new insight into the mechanism of action of estrogens in fetal Leydig cell development. We observed no change in the number of Leydig cells in ER-deficient mice. Moreover, treatment for 3 d with 1 μM DES had no effect on the number of fetal Leydig cells in testes explanted on fetal d 13.5. In contrast, in vivo studies have shown that the treatment of pregnant mice with estrogens induces foci of Leydig cell hyperplasia in fetuses on d 16 and 18 of gestation (19, 44), suggesting a possible role of estrogens in regulating the differentiation of new fetal Leydig cells. In a previous study, treatment for 3 d with 4 μM DES in an organ culture system slightly decreased the number of Leydig cells in rat testes explanted on fetal d 14.5 dpc (17). The reasons for these discrepancies with our present results are unclear but may be due to differences in experimental approaches (in vitro vs. in vivo), doses or models (rat vs. mouse). Nevertheless, the results presented here, together with our previous findings that the number of fetal Leydig cell is not affected by ER? gene inactivation (27), suggest that estrogens at endogenous concentrations are not involved in the differentiation of mesenchymal cells to generate fetal Leydig cells in the mouse.
We observed a clear negative effect of endogenous estrogens on the activity and differentiated functions of each fetal Leydig cell analyzed at the morphometric, functional, and molecular levels. Inactivation of the ER gene leads to increases in Leydig cell cytoplasmic volume, basal and LH-stimulated testosterone production, and mRNA levels for StAR, P450c17, and P450scc. These increases in mRNA levels are consistent with other reports showing that treatment with chemicals with estrogenic activity on fetal d 11.5 and 15.5 reduces the amount of P450c17 mRNA, protein, and activity in fetuses on d 17.5 of gestation (20). The molecular mechanism by which estrogens affect the transcription of the StAR, P450c17, and P450scc genes is largely unknown. Estrogens may act directly on the promoters of these genes via a classical mode of action because these sequences contain estrogen response elements (EREs) (Dragon ERE finder version 2) (45). It is also possible that estrogens act indirectly, by interacting with other non-ERE sites (46) or by regulating other transcription factors known to regulate the expression of steroidogenic enzyme genes. For example, estrogen treatment has been shown to decrease levels of mRNA for SF-1 (47), a transcription factor known to regulate the expression of P450 enzyme genes (48).
Little is known about the amount and origin of endogenous estrogens in fetal and neonatal mouse testes. We previously reported an intratesticular estradiol concentration of 4 nM in testes from 2-d-old neonates (27). P450arom, the enzyme that catalyzes the conversion of testosterone to estradiol, is detected from fetal d 17 in the mouse fetal testis (39), suggesting that total estrogen content changes throughout the developmental period considered here. On fetal d 13.5, the testis is not able to produce estradiol by itself, and the mother is probably an essential source of this hormone because the mouse placenta is not involved in the de novo synthesis of steroids during the second half of pregnancy (49). However, estrogens bind to extracellular carrier proteins and are conjugated and metabolized to generate inactive forms (50). For example, in mice, the -fetoprotein produced throughout development in the visceral endoderm of the extra-embryonic yolk sac and in the fetal liver and gut (51) can bind the estrogens synthesized by the mother, thereby increasing estrogen concentration in the fetus. It is therefore difficult to evaluate the actual concentration of estrogens acting on fetal Leydig cells in vivo. Nevertheless, our results show that this steroid is certainly present in the fetal testis and is physiologically efficient in the regulation of steroidogenesis.
In conclusion, this study demonstrates for the first time an ER-mediated in vivo inhibitory effect of endogenous estrogens on the development of testicular steroidogenesis during fetal and neonatal development in the mouse. Because androgen production is one of the key mechanisms underlying male testicular differentiation, these findings provide important insight into the development of the male reproductive tract. They also support the hypothesis that fetal and neonatal exposure to environmental xenoestrogens could impair the masculinization of the male urogenital system and male fertility in adulthood.
Acknowledgments
We thank Prof. P. Chambon and A. Krust (Institut de Génétique et Biologie Moléculaire et Cellulaire, Illkirch, France) for providing transgenic mice and Dr. G. Defaye (Institut National de la Santé et de la Recherche Médicale, Grenoble, France) for providing the anti-3?HSD antibody. We also thank C. Joubert, P. Flament, and V. Neuville (Commissariat à l’Energie Atomique, Fontenay aux Roses, France) for caring for the animals.
References
Saez JM 1994 Leydig cells: endocrine, paracrine and autocrine regulation. Endocr Rev 15:574–626
Habert R, Lejeune H, Saez JM 2001 Origin, differentiation and regulation of fetal and adult Leydig cells. Mol Cell Endocrinol 179:47–74
Jost A 1970 Hormonal factors in the sex differentiation of the mammalian foetus. Philos Trans R Soc Lond B Biol Sci 259:119–130
Vergouwen RP, Jacobs SG, Huiskamp R, Davids JA, de Rooij DG 1991 Proliferative activity of gonocytes, Sertoli cells and interstitial cells during testicular development in mice. J Reprod Fertil 93:233–243
Pakarinen P, Vihko K, Voutilainen R, Huhtaniemi I 1990 Differential response of Luteinizing Hormone receptor and steroidogenic enzyme gene expression to human chorionic gonadotropin stimulation in the neonatal and adult rat testis. Endocrinology 127:2469–2474
O’Shaughnessy P, Baker U, Sohnius U, Haavisto A-M, Charlton H, Huhtaniemi I 1998 Fetal development of Leydig cell activity in the mouse is independent of pituitary gonadotroph function. Endocrinology 139:1141–1146
Migrenne S, Pairault C, Racine C, Livera G, Géloso A, Habert R 2001 LH-dependent activity and LH-independent differentiation of rat fetal Leydig cells. Mol Cell Endocrinol 172:193–202
O’Shaughnessy PJ, Johnston H, Willerton L, Baker PJ 2002 Failure of normal adult Leydig cell development in androgen-receptor-deficient mice. J Cell Sci 115:3491–3496
Sharpe RM 2003 The "oestrogen hypothesis": where do we stand now? Int J Androl 26:2–15
Sharpe RM, Irvine DS 2004 How strong is the evidence of a link between environmental chemicals and adverse effects on human reproductive health? BMJ 328:447–451
Toppari J, Larsen J, Christiansen P, Giwereman A, Grandjean P, Guillette L, Jegou B, Jensen T, Jouannet P, Keiding N, Leffers H, McLachlan J, Meyar O, Muller J, Rajpert de Meyrs E, Scheike T, Sharpe R, Sumpter J, Skakkebaek N 1996 Male reproductive health and environmental xenoestrogens. Environ Health Perspect 104:741–803
Danzo B 1998 The effects of environmental hormones on reproduction. Cell Mol Life Sci 54:1249–1264
Skakkebaek NE, Rajpert-De Meyts E, Main KM 2001 Testicular dysgenesis syndrome: an increasingly common developmental disorder with environmental aspects. Hum Reprod 16:972–978
Kubota Y, Temelcos C, Bathgate RA, Smith KJ, Scott D, Zhao C, Hutson JM 2002 The role of insulin 3, testosterone, Mullerian inhibiting substance and relaxin in rat gubernacular growth. Mol Hum Reprod 8:900–905
Ivell R, Hartung S 2003 The molecular basis of cryptorchidism. Mol Hum Reprod 9:175–181
Tsai-Morris CH, Knox G, Luna S, Dufau ML 1986 Acquisition of estradiol-mediated regulatory mechanism of steroidogenesis in cultured fetal rat Leydig cells. J Biol Chem 261:3471–3474
Lassurguere J, Livera G, Habert R, Jegou B 2003 Time- and dose-related effects of estradiol and diethylstilbestrol on the morphology and function of the fetal rat testis in culture. Toxicol Sci 73:160–169
Grocock CA, Charlton HM, Pike MC 1988 Role of the fetal pituitary in cryptorchidism induced by exogenous maternal oestrogen during pregnancy in mice. J Reprod Fertil 83:295–300
Perez-Martinez C, Garcia-Iglesias M, Ferreras-Estrada M, Bravo-Moral A, Espinosa-Alvarez J, Escudero-Diez A 1996 Effects of in-utero exposure to zeranol or diethylstilbestrol on morphological development of the fetal testis in mice. J Comp Path 114:407–418
Majdic G, Sharpe RM, O’Shaughnessy PJ, Saunders PT 1996 Expression of cytochrome P450 17-hydroxylase/C17–20 lyase in the fetal rat testis is reduced by maternal exposure to exogenous estrogens. Endocrinology 137:1063–1070
Gill WB, Schumacher GF, Bibbo M, Straus 2nd FH, Schoenberg HW 1979 Association of diethylstilbestrol exposure in utero with cryptorchidism, testicular hypoplasia and semen abnormalities. J Urol 122:36–39
Newbold R 1995 Cellular and molecular effects of developmental exposure to diethylstilbestrol: implications for other environmental estrogens. Environ Health Perspect 103(Suppl 7):83–87
Klip H, Verloop J, van Gool JD, Koster ME, Burger CW, van Leeuwen FE 2002 Hypospadias in sons of women exposed to diethylstilbestrol in utero: a cohort study. Lancet 359:1102–1107
Atanassova N, McKinnell C, Turner K, Walker M, Fisher J, Morley M, Millar M, Groome N, Sharpe R 2000 Comparative effects of neonatal exposure of male rats to potent and weak (environmental) estrogens on spermatogenesis at puberty and the relationship to adult testis size and fertility: evidence for stimulatory effects of low estrogen levels. Endocrinology 141:3898–3908
Fielden MR, Samy SM, Chou KC, Zacharewski TR 2003 Effect of human dietary exposure levels of genistein during gestation and lactation on long-term reproductive development and sperm quality in mice. Food Chem Toxicol 41:447–454
Dupont S, Krust A, Gansmuller A, Dierich A, Chambon P, Mark M 2000 Effect of single and compound knockouts of estrogen receptors (ER) and ? (ER?) on mouse reproductive phenotypes. Development 127:4277–4291
Delbes G, Levacher C, Pairault C, Racine C, Duquenne C, Krust A, Habert R 2004 Estrogen receptor {?}-mediated inhibition of male germ cell line development in mice by endogenous estrogens during perinatal life. Endocrinology 145:3395–3403
Habert R, Devif I, Gangnerau MN, Lecerf L 1991 Ontogenesis of the in vitro response of rat testis to gonadotropin-releasing hormone. Mol Cell Endocrinol 82:199–206
Livera G, Rouiller-Fabre V, Durand P, Habert R 2000 Multiple effects of retinoids on the development of Sertoli, germ and Leydig cells of fetal and neonatal rat testis in culture. Biol Reprod 62:1303–1314
Abercrombie M 1946 Estimation of nuclear population from microtome sections. Anat Rec 94:238–248
Bustin SA 2000 Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. J Mol Endocrinol 25:169–193
Migrenne S, Racine C, Guillou F, Habert R 2003 Pituitary hormones inhibit the function and differentiation of fetal Sertoli cells. Endocrinology 144:2617–2622
Habert R, Picon R 1984 Testosterone, dihydrotestosterone and estradiol 17? levels in maternal and fetal plasma and in fetal testes in the rat. J Steroid Biochem 21:193–198
Zhang FP, Poutanen M, Wilbertz J, Huhtaniemi I 2001 Normal prenatal but arrested postnatal sexual development of luteinizing hormone receptor knockout (LuRKO) mice. Mol Endocrinol 15:172–183
Rulli SB, Zitta K, Calandra RS, Campo S 2003 Effect of dihydrotestosterone on pituitary follicle-stimulating hormone isoforms in adult male rats treated with a gonadotropin-releasing hormone antagonist. Neuroendocrinology 78:280–286
Habert R, Picon R 1982 Control of testicular steroidogenesis in fetal rat: effect of decapitation on testosterone and plasma luteinizing hormone-like activity. Acta Endocrinol 99:466–473
Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA 1997 Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors and ?. Endocrinology 138:863–870
Jefferson W, Couse J, Banks E, Korach K, Newbold R 2000 Expression of estrogen receptor ? is developmentally regulated in reproductive tissues of male and female mice. Biol Reprod 62:310–317
Greco TL, Payne AH 1994 Ontogeny of expression of the genes for steroidogenic enzymes P450 side-chain cleavage, 3 ?-hydroxysteroid dehydrogenase, P450 17 -hydroxylase/C17–20 lyase, and P450 aromatase in fetal mouse gonads. Endocrinology 135:262–268
Eddy EM, Washburn TF, Bunch DO, Goulding EH, Gladen BC, Lubahn BD, Korach KS 1996 Targeted disruption of the estrogen receptor gene in male mice causes alteration of spermatogenesis and infertility. Endocrinology 137:4796–4805
Akingbemi BT, Ge R, Rosenfeld CS, Newton LG, Hardy DO, Catterall JF, Lubahn DB, Korach KS, Hardy MP 2003 Estrogen receptor- gene deficiency enhances androgen biosynthesis in the mouse Leydig cell. Endocrinology 144:84–93
Haavisto T, Nurmela K, Pohjanvirta R, Huuskonen H, El-Gehani F, Paranko J 2001 Prenatal testosterone and luteinizing hormone levels in male rats exposed during pregnancy to 2,3,7,8-tetrachlorodibenzo-p-dioxin and diethylstilbestrol. Mol Cell Endocrinol 178:169–179
Nishihara E, Nagayama Y, Inoue S, Hiroi H, Muramatsu M, Yamashita S, Koji T 2000 Ontogenetic changes in the expression of estrogen receptor and ? in rat pituitary gland detected by immunohistochemistry. Endocrinology 141:615–620
Yasuda Y, Kihara T, Tanimura T 1985 Effect of ethinyl estradiol on the differentiation of mouse fetal testis. Teratology 32:113–118
Bajic VB, Tan SL, Chong A, Tang S, Strom A, Gustafsson JA, Lin CY, Liu ET 2003 Dragon ERE Finder version 2: a tool for accurate detection and analysis of estrogen response elements in vertebrate genomes. Nucleic Acids Res 31:3605–3607
Kushner PJ, Agard D, Feng WJ, Lopez G, Schiau A, Uht R, Webb P, Greene G 2000 Oestrogen receptor function at classical and alternative response elements. Novartis Found Symp 230:20–26; discussion 27–40
Saunders P, Majdic G, Parte P, Millar M, Fisher J, Turner K, Sharpe R 1997 Fetal and perinatal influence of xenoestrogens on testis gene expression. In: Holstein IA, ed. The fate of the male germ cells. New York: Plenum Press; 99–110
Hatano O, Takayama K, Imai T, Waterman MR, Takakusu A, Omura T, Morohashi K 1994 Sex-dependent expression of a transcription factor, Ad4BP, regulating steroidogenic P-450 genes in the gonads during prenatal and postnatal rat development. Development 120:2787–2797
Arensburg J, Payne AH, Orly J 1999 Expression of steroidogenic genes in maternal and extraembryonic cells during early pregnancy in mice. Endocrinology 140:5220–5232
McLachlan JA, Newbold RR 1987 Estrogens and development. Environ Health Perspect 75:25–27
Tilghman SM, Belayew A 1982 Transcriptional control of the murine albumin/-fetoprotein locus during development. Proc Natl Acad Sci USA 79:5254–5257(Géraldine Delbès, Christi)