| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Testis |
and ß in Neonatal Rat Testis: Identification of Gonocytes as Targets of Estrogen Exposure1
a Division of Hormone Research, Department of Cell Biology, Georgetown University School of Medicine, Washington, District of Columbia 20057
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ß mRNAs, determined by semiquantitative reverse transcription polymerase chain reaction, though diethylstilbestrol had a biphasic effect on both mRNAs. In situ hybridization analysis showed that PDGFR mRNA increased mostly in the interstitium, while PDGFRß mRNA increased both in the interstitium and seminiferous cords. Immunohistochemical studies of PDGFR and ß proteins revealed that both receptors were present in testis before and after birth and that they were upregulated upon treatment with estrogens in 3-day-old rats, with PDGFRß increasing dramatically in gonocytes. PDGFR and ß mRNAs and proteins were also found in purified gonocytes. Our previous finding that PDGF and 17ß-estradiol induce gonocyte proliferation in vitro, together with the present finding that in vivo exposure to estrogens upregulates PDGF receptors in testis, suggest that PDGF pathway is a target of estrogens in testis. In addition, these data identify PDGFRß in gonocytes as a major target of gestational estrogen exposure, suggesting that estrogen may have a physiological interaction with PDGF during gonocyte development. These results, however, do not exclude the possibility that the effects of the compounds examined in this study might be due to estrogen receptor-independent action(s).
estradiol, growth factors, spermatogenesis, testis, toxicology
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(ER) knockout male mice are infertile has crystallized the emerging concept that estrogens are not only involved in the development of the female reproductive tract but that they are also critical for the development of male reproductive organs [9, 10]. Indeed, studies have shown that estrogen exposure disturbs the differentiation and/or proliferation of Leydig [10, 11] and Sertoli cells [12], directly or via perturbation of the hypothalamo-pituitary axis [13]. Some reports have also raised the possibility that testicular gonocytes might be directly affected by estrogens [14–16]. Indeed, we have previously shown that estradiol stimulates the proliferation of neonatal gonocytes in vitro [14]. Gonocytes were further reported to express the ERß [17, 18].
Estrogens exert their effects by binding on the ERs, which in turn trigger the transcription of estrogen-regulated genes [19, 20]. Several xenoestrogens, epitomized by diethylstilbestrol (DES) [21], and phytoestrogens, including plant isoflavones such as genistein [22], have been characterized for their ability to compete with 17ß-estradiol in receptor binding assays and to induce transcriptional activity of the two main ERs, and ß [23, 24]. Besides these classical genomic effects, estrogens have been shown to have nontranscriptional effects on various signal transduction pathways via either the classical intracellular ER [25] or a membrane-bound ER [26]. ER has also been implicated in estrogen-independent types of cross-talk with growth factors signaling pathways such as that of epidermal growth factor [27, 28]. In leiomyoma cells, 17ß-estradiol was shown to stimulate platelet-derived growth factor (PDGF) secretion in an autocrine fashion, resulting in cell proliferation [29], while in mouse uterus and vagina, DES was shown to induce PDGF and PGDF receptor (PDGFR) overexpression [30]. In previous studies, we found that PDGF as well as 17ß-estradiol were able to induce gonocyte proliferation in vitro and that their effects were not additive, suggesting that the two pathways shared a common element [14].
The importance of the PDGF pathway during development is emphasized by the fact that both PDGFR and ß knockout mice do not survive past birth. Indeed, PDGFR knockout mice die in utero and present various defects in crest-derived tissues [31], while PDGFRß knockout mice die right before birth and present abnormal capillary development affecting diverse tissues [32]. An earlier study suggested that the PDGF pathway is involved in Sertoli and myoid cell functions during testis development [33]. Our finding that neonatal testicular gonocytes express PDGFR and proliferate in response to PDGF [14] further suggested that PDGF signaling may also directly regulate germ cell development. Considering our own work and the above-mentioned literature, it appears that estrogens may also play a role in testis development, further suggesting the possibility of a cross-talk between the two growth-promoting pathways in the developing testis.
In the present study, we examined the possibility of a relationship between the estrogen and PDGF signaling pathways in testis by investigating the effects of prenatal exposure to four estrogenic compounds on the expression of PDGFR in neonatal rat testis. These experiments demonstrated that prenatal exposure to xeno-/phytoestrogens alters the expression of PDGFR and ß mRNAs and proteins, further suggesting that estrogen may have a physiological interaction with PDGF during testis development. These data, however, do not exclude the possibility that the effects of these compounds on PDGFR expression and testicular function may be distinct from their estrogenic actions.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bisphenol A (4,4'-isopropylidenediphenol; BPA) was purchased from ICN Biomedicals (Aurora, OH). Diethylstilbestrol, genistein, 17ß-estradiol, and hematoxylin were purchased from Sigma (St. Louis, MO). Coumestrol was purchased from Fluka Chemika (Sigma). Rabbit polyclonal antibodies against the Cterminal domains of PDGFR and PDGFRß were purchased from Santa Cruz Biotechnology (Santa Cruz, CA), while rabbit polyclonal antibodies against the kinase domain of PDGFRß was purchased from BD Biosciences (BD Pharmingen, San Diego, CA). The mAb anti-phosphotyrosine residue (PY20) was from BD Biosciences (BD Transduction Labs, San Diego, CA). Rabbit and mouse IgGs and reagents for immunohistochemistry (Histostain-Plus Kit) were from Zymed Laboratories (South San Francisco, CA). Crystal-mount was from Biomeda Corp. (Foster City, CA). Restriction enzymes were from Stratagene (La Jolla, CA) and New England Biolabs (Beverly, MA). Kits for plasmid isolation and PCR product extraction were from Qiagen (Valencia, CA). Cell culture supplies were purchased from Invitrogen (Grand Island, NY). Electrophoresis reagents and materials were supplied by BioRad (Hercules, CA). All other chemicals used were of analytical grade and were obtained from various commercial sources.
Animals and Treatments
Sprague Dawley female rats or newborn male pups were purchased from Charles River Laboratories (Wilmington, MA). Pregnant rats were treated daily by gavage or s.c. injection (for DES only) with vehicle (corn oil with dimethyl sulfoxide added in the same amount as in the drugs), BPA, DES, genistein, or coumestrol from Gestational Day 14 (14 dpc) to birth (Day 0). The animals were weighed daily and the treatments adjusted accordingly. The dose ranges were, for DES, 0.01–2 µg (0.037–7.5 nmol) kg-1 day-1; for BPA, 0.1–200 mg (0.44–877 µmol) kg-1 day-1; for genistein, 0.1–10 mg (0.37–37 µmol) kg-1 day-1, and for coumestrol, 1–100 mg (3.7–370 µmol) kg-1 day-1. Some animals were also treated by s.c. injection of 17ß-estradiol at 100 µg (0.37 µmol)/kg-1 day-1. Male offspring, from 21 dpc to Neonatal Day 3 (3 dpp), were killed by CO2 inhalation and testes were collected and either fixed in 3.5% buffered formaldehyde or snap-frozen in liquid nitrogen. The animals were handled according to a protocol that was reviewed and approved by the Georgetown University Animal Committee.
Isolation of Gonocytes
In order to obtain pure gonocyte preparations for mRNA analysis, we modified the method of cell purification previously published [14] by adding a step of individual cell sorting after the BSA gradient using a micromanipulator system (TransferMan NK micromanipulator from Eppendorf Scientific; Brinkmann Instruments, Westbury, NY) equipped with a cell-harvesting system coupled to a 15-µm glass transfer micropipette to selectively collect gonocytes from 3-day-old rats. The final pool of pure gonocytes (3000–6000 cells per sorting) was centrifuged and pellets were frozen at -70°C for RNA extraction or protein analysis. The mixed cultures of Sertoli and myoid cells obtained during the process of cell purification were also collected and frozen to serve as comparison for protein analysis.
Protein Analysis
Aliquots of cell preparations were solubilized in Laemmli buffer, their protein content was determined using Coomassie staining quantification, and the proteins were separated by SDS-page electrophoresis on 4–20% polyacrylamide gels. Immunodetection of PDGFRs was carried out by immunoblot analysis using antibodies raised against PDGFR or PDGFRß (kinase or Ct domains; 1:1000 in Tris HCl 20 mM, pH 7.0, NaCl 500 mM, 0.05 % Tween-20) and a chemiluminescence kit (ECL western blotting detection kit; Amersham Pharmacia Biotech, Piscataway, NJ) as previously described [14].
RNA Isolation and Semiquantitative Reverse Transcription Polymerase Chain Reaction Analysis
Total RNA was extracted from frozen testes previously pulverized on dry ice or isolated cells using RNAzol B reagent (TEL-TEST, Friendswood, TX), following the recommendations of the manufacturer, and stored in diethylpyrocarbonate-treated water at -20°C. Reverse transcription polymerase chain reactions (RT-PCR) were carried out using an RNA PCR kit from Perkin Elmer (Branchburg, NJ), according to the manufacturer's protocol, on 10 µg total RNA samples. PCR reactions contained 2 µl of cDNAs, 1 U Taq DNA polymerase, 20 pg of each primer, 5 nmol of each dNTP, 2 µl 10x PCR buffer, 1 µl MgCl2 (25 mM), 12 µl sterile double-distilled water in 20-µl final volume. G3PDH (0.45 kb 5' and 3' G3PDH amplimers, 20 pmol each; Clontech, Palo Alto, CA) was used as house-keeping gene for data normalization. The sets of primers used for PDGFR and PDGFRß (extracellular domain) mRNA quantification (Table 1) generated fragments of 999 and 990 base pairs (bp), respectively. Preliminary experiments were run to determine the optimal number of cycles required to perform the PCR reactions in exponential phase. The cycling parameters for the paired PDGFR/G3PDH and PDGFRß/G3PDH were as described in Table 2. PCR products were separated on 1% agarose gels in 1x TAE buffer and visualized by ethidium bromide staining. Image analysis of the bands obtained was performed using the OptiQuant Acquisition & Analysis program from Packard Biosciences (Meridien, CT). For each sample, the ratio of the densities of the target gene and G3PDH bands was calculated. The ratios of the control samples were used as 100% and given an arbitrary value of 1. The results represent the average ± SEM of values obtained in three independent experiments performed on rats born from different mothers in order to account for the variations in maternal response to the treatments. Each individual RNA sample was isolated from two pooled testis of pups of the same litter to provide sufficient amounts of RNA and to account for variations between animal responses.
|
|
To examine the expression of PDGFR and PDGFRß mRNAs in purified gonocytes and because of the limited number of purified gonocytes, mRNAs were first amplified using the RiboAmp RNA amplification kit from Arcturus Engineering (Mountain View, CA) that uses oligo dT primers. The purity of the gonocyte preparation was also evaluated by measuring the mRNA expression in gonocytes of three somatic cell markers, including P450scc, androgen binding protein, and smooth muscle actin, expressed by Leydig, Sertoli, and myoid cells, respectively. RT-PCR reactions were carried out using primers targeting the 3' end of the different mRNAs examined (Table 1). The detection of the mRNA of P450scc, androgen binding protein, and smooth muscle actin was performed using identical cycling conditions (Table 2). PDGFR mRNA was detected using the same set of primers and cycle temperatures as that used for whole testis RNA analysis (Tables 1 and 2), except that the 5 initial cycles were followed by 30 additional cycles instead of 17. In the case of PDGFRß, the only rat sequence reported in the GenBank was a partial 5' end mRNA sequence that could not be used to amplify 3' end fragments. Thus, we first determined the sequence of the full-length rat mRNA for PDGFRß using an adult rat testis cDNA library (Marathon-Ready cDNA kit, Advantage from Clontech) as template. PCR product sequences were determined using an ABI PRISM dye terminator cycle sequencing ready reaction kit (PE Biosystems, Foster City, CA) and an Applied Biosystems sequencer (Applied Biosystems, Foster City, CA) at the Lombardi Cancer Center Sequencing Core Facility (Georgetown University). A set of primers corresponding to the 3' end of the rat mRNA (Cter set; Table 1) was further used in RT-PCR reactions using cycling conditions similar to those used for PDGFR with the exception of the annealing temperatures, which were 60°C for the first 5 cycles and 58°C for the 30 remaining cycles.
In Situ Hybridization Experiments
Probes of approximately 1000-bp size were synthesized to fit the criteria recommended by Molecular Histology Labs (Gaithersburg, MD) using the rat testis cDNA library mentioned above as template and primers for PDGFR and ß (extracellular domain) genes (Table 1). PCR reaction mixtures contained 10 nmol of each dNTP, 50 pmol of each primer, 1 µl of Taq polymerase (50x, advantage 2 polymerase from the kit), 5 µl 10x PCR buffer, 5 µl cDNA, and 36 µl double-distilled water in a final volume of 50 µl. For PDGFR, the cycling conditions were the same as those described in Table 2 except that the initial 5 cycles were followed by 30 additional cycles. For PDGFRß, the conditions were similar to those presented in Table 2 with the exception that the initial 5 cycles were done at 60°C, followed by 30 cycles at 56°C. PCR products were separated on 1% agarose gels, extracted from the gels, and cloned into a T/A type vector using the TOPO-TA cloning kit (Invitrogen, Carlsbad, CA). The orientation of the plasmids was verified by restriction enzyme digestion and the identity of the probes produced was verified by sequencing. DNA sequences were analyzed by using Entrez and BLAST program against GenBank Database. Once the proper identity of a probe was confirmed, it was produced in large quantity and further purified by conventional organic extraction. Prior to labeling, both sense and antisense products (1 µg each) were linearized by restriction enzyme digestion (five units restriction enzyme per sample) for 2 h at 37°C. Linearized templates were then labeled using the following reaction mixture: 1 µg of T7 or SP6 primer, 60 units of RNA polymerase, 200 µCi 35S-CTP, 2.5 mM of ATP, 2.5 mM GTP, 2.5 mM UTP, 0.1 mM CTP, 100 mM dithiothreitol (DTT), 5x transcription buffer (Promega, Madison, WI), and 20 units of Rnasin ribonuclease inhibitor to protect the newly synthesized RNA transcripts. After a 90-min incubation at 37°C, the DNA templates were removed by further incubation with 1 unit of DNase 1 at 37°C for 15 min. After alkaline hydrolysis, the 35S-labeled cRNA probes were purified by phenol-chloroform extractions and passage through G-50 spun columns and then resuspended in TE buffer containing DTT.
In situ hybridization reactions were performed on paraffin sections of 3-day-old rat testis samples as described by Li and collaborators [34]. Briefly, the sections were rehydrated, acetylated, prehybridized, and heat denatured prior to hybridization in a cocktail (2x SSC, 1x Denhardt, 50 mM phosphate buffer, 50 mM DTT, 500 µg/µl of salmon sperm DNA, 250 µg/µl of tRNA, 5 µg/ml poly(dA), 100 µg/ml poly(A), 0.05 pmol/ml of randomer, 57% dextran/formamide) containing the 35S-labeled cRNA-specific probes or nonsense probes (0.08 µl of probe/µl of cocktail). After an overnight incubation at 45°C in a sealed humidified chamber, the slides were incubated in 76% formamide-0.25x SSC-1.2 mM DTT-0.5 mM EDTA (two times for 30 min at 37°C), 0.25x SSC (10 min at 37°C), and RNase (RNase A, 25 µg/ml; RNase T1, 5 U/ml) in 0.5 M NaCl-1.2 mM DTT-0.01 M Tris-HCl (pH 7.4) for 40 min at 37°C, dehydrated through graded 0.3 M ammonium acetate-ethanol, air dried, dipped in NTB-3 (Eastman Kodak, Rochester, NY) photographic emulsion (45°C), and incubated at 4°C for 5 days in a light-tight box. Slides were developed at 15°C in D-19 developer (Eastman Kodak) for 4 min, fixed for 6 min, counterstained with hematoxylin and eosin, coverslipped with permount, and evaluated under bright- and dark-field microscopy. For each treatment, the hybridization experiments were performed in duplicate on testis sections of rats issued from two different litters. We further quantified the intensity of the signals present in three distinct areas of the cords, respectively, the interstitium (including Leydig cells, blood vessels, fibroblasts, and macrophages), the periphery of the cords (including myoid and Sertoli cells), and the center of the cords (comprising gonocytes and parts of Sertoli cell cytosol) by performing densitometric analysis of these areas in five different zones of each section to reflect possible heterogeneity of the response through the tissue using the OptiQuant Acquisition & Analysis program. The results were expressed in arbitrary units of signal intensity and represent specific signals obtained by subtracting the signals of nonsense probes from those obtained with the antisense probes for each sample. The in situ hybridization illustrations show a representative experiment for each condition.
Immunohistochemistry
Testes from 21-day-old fetuses and 3-day-old rats were fixed in 3.5% buffered formaldehyde and embedded in paraffin. Briefly, 5-µm sections were dewaxed, rehydrated through a graded series of alcohol, washed with PBS and water, permeabilized by heating the slides (immersed in 0.01 M citrate buffer, pH 6.0) in a microwave for 5 min at high power (800 W), as previously described [14]. The sections were then incubated at 4°C overnight with the primary antibodies diluted (1:40–1:200) in PBS containing 10% calf serum, then sequentially with the reagents provided in the Histostain-Plus Kit from Zymed Laboratories (South San Francisco, CA), including biotin-coupled secondary antibodies, streptavidin-coupled horse radish peroxidase solution, and a substrate mixture for the peroxidase (0.03% H2O2 + 0.2 mg/ml of 3-amino-9-ethylcarbazole in 0.05 M Na acetate, pH 5.0). As negative controls, nonspecific IgG of the same species as the first antibodies were used instead of first antibodies in the treatment of several slides. At the end of the reaction, the sections were counterstained with hematoxylin, further coated with Crystal-mount, and examined under bright-field microscopy with an BX40 Olympus microscope (Olympus America, Melville, NY) coupled to a Spot RT color CCD camera (Diagnostic Instruments, Sterling Heights, MI). The results obtained for animals born from 4–6 different mothers were compared and representative results are shown.
Statistical Analysis
Statistical analysis was performed by unpaired t-test with Welch correction using the Instat (v.3.0) package from GraphPad (San Diego, CA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ß mRNAs
Semiquantitative RT-PCR analysis of the PDGFR mRNA was carried out using G3PDH mRNA as a standard for normalization, as described in the Materials and Methods section. As shown in Figure 1, the four estrogenic compounds examined induced a dose-dependent overexpression, from 1.5- to 5-fold increase, of the PDGFR mRNA in whole testis. The biggest increase was obtained following treatment with genistein (Fig. 1C; 4.8-fold increase at 0.1 mg kg-1 day-1), while BPA and coumestrol induced changes of similar amplitude (Fig. 1A, 2.9-fold increase at 200 mg kg-1 day-1; Fig. 1D, 2.6-fold increase for 100 mg kg-1 day-1). In the case of DES, the effect was biphasic, as there was an initial increase (maximum twofold increase at 0.1 µg kg-1 day-1) followed by a return to the control levels for the highest dose (Fig. 1B). A similar semiquantitative RT-PCR analysis of PDGFRß mRNA in neonatal testis revealed that it was also increased in a dose-dependent fashion in pups from treated mothers (Fig. 2). In this case, the two phytoestrogens examined, genistein and coumestrol, appeared to induce a stronger response (Fig. 2, C and D), reaching values of a 3.5-fold increase (0.1 mg kg-1 day-1 genistein) to a 6.2-fold increase (100 mg kg-1 day-1 coumestrol), than the xenoestrogens, BPA, and DES (Fig. 2, A and B). While doses of 1–200 mg kg-1 day-1 BPA had clear and significant effects (P < 0.05 to P < 0.001), reaching a maximum of a 2.6-fold increase for the highest dose, DES induced only a small and nonsignificant increase (maximum 1.45-fold at 0.1 µg kg-1 day-1), followed by a significant inhibition (68%) at the highest dose (2 µg kg-1 day-1). In general, the range of doses giving significant increases for DES, BPA, and genistein indicated an order of potency similar to that predicted from the results of in vitro studies [24]. For instance, DES increased PDGFR expression at 0.01 µg kg-1 day-1, while 0.1 mg kg-1 day-1 genistein and 1 mg kg-1 day-1 BPA were needed to induce an increase of PDGFR expression. However, a dose of 5 mg kg-1 day-1 of coumestrol was required to increase by more than 50% the expression of both mRNAs.
|
|
To compare the effects of these estrogenic compounds with that of the endogenous estrogen 17ß-estradiol, some female rats were treated with s.c. injections of 100 µg kg-1 day-1 17ß-estradiol. 17ß-Estradiol elicited a small but significant (1.45 ± 0.14-fold of control; P < 0.05) increase of PDGFR mRNA but no significant change (1.20 ± 0.27-fold of control) in PDGFRß mRNA.
Comparison of the Levels and Cellular Localization of PDGFR and ß mRNAs by In Situ Hybridization
The previous experiments revealed changes in the total testicular levels of the PDGFR mRNAs but did not indicate in which testicular compartments they were occurring. To answer this question, in situ hybridization experiments were carried out (Fig. 3), and the intensity of the signals obtained in several randomly picked areas of the interstitium and seminiferous cords were quantified (Fig. 4). As shown in Figures 3 and 4, both PDGFR and ß mRNAs were mostly localized in the interstitial compartment in control samples. BPA (200 mg kg-1 day-1) and genistein (10 mg kg-1 day-1) induced the overexpression of PDGFR mRNA in the interstitial compartment and no significant change in the seminiferous cords. Coumestrol (50 mg kg-1 day-1) induced a slight (30%) decrease in the interstitium and no change in the cords of PDGFR mRNA, while DES (1 µg kg-1 day-1) induced a strong (80%) decrease in the interstitium and no change in the cords.
|
|
BPA, genistein, and coumestrol treatments had more generalized effects on PDGFRß mRNA expression, inducing significant increases in all areas of the testis (Figs. 3 and 4). Indeed, these compounds induced a 3- to 7-fold increase of PDGFRß mRNA in the interstitial compartment, a 3- to 12-fold increase in the center of the seminiferous cords, and a 2- to 3-fold increase in the periphery of the cords. At the concentration examined, DES had no effect on PDGFRß mRNA expression in the interstitium but slightly increased the mRNA in the seminiferous cords. The nonspecific signals obtained in reactions performed with nonsense or sense probes were generally similar for most samples and are shown in Figure 3.
Immunolocalization of PDGFR and ß Proteins in Neonatal Testis Following Estrogen Prenatal Exposure
Because the expression of a protein does not always follow the same pattern of expression as its mRNA, we examined the expression of the PDGFR and ß proteins following estrogen prenatal exposure by immunohistochemistry (Fig. 5). In sections from control testis, PDGFR was strongly expressed in the peritubular myoid cells, while its expression varied in gonocytes from a robust to a weak signal. Prenatal exposure to 200 mg kg-1 day-1 BPA and 50 mg kg-1 day-1 coumestrol induced a slight increase of PDGFR expression in gonocytes, while treatment with 10 mg kg-1 day-1 genistein induced an increase in Sertoli cells but not in gonocytes. Sertoli cell immunoreactivity was also increased by coumestrol and in some, but not all, sections from BPA-treated rats. Samples exposed to DES (1 µg kg-1 day-1) retained immunoreactivity for the receptor in myoid cells and in gonocytes. Concerning PDGFRß, it was expressed at low levels in gonocytes and interstitial cells of control samples (Fig. 5A). However, prenatal exposure to 200 mg kg-1 day-1 BPA, 10 mg kg-1 day-1 genistein, and 50 mg kg-1 day-1 coumestrol induced a strong expression of PDGFRß in gonocytes. BPA and genistein also induced a slight increase of PDGFRß immunoreactivity in Sertoli cells. As shown in Figure 5C, PDGFRß was expressed at high levels in gonocytes from BPA-treated samples, in contrast with control samples (Fig. 5A), where only low levels of PDGFRß could be detected. In the testis from DES-exposed rats, minimal levels of PDGFRß were detectable, although it was clearly expressed by a small fraction of gonocytes. Figure 5B illustrates the fact that the background stainings obtained with a rabbit IgG were comparable in all samples independently of the treatments.
|
Immunolocalization of PDGFR and ß Proteins in Prenatal Testis Following Estrogen Gestational Exposure
In order to determine if PDGFR and ß were expressed in testis prior to the neonatal phase of gonocyte proliferation, immunolocalization of the receptors was performed in testis from 21-day-old fetuses, corresponding to a period of quiescence for gonocytes. As shown in Figure 6, A and C (top panel), PDGFR was present in prenatal gonocytes and in Sertoli cells in control samples, while lower levels of the receptor were present in interstitial cells. Interestingly, the expression of PDGFR in gonocytes was not altered by BPA, genistein, or coumestrol treatments. However, the expression of PDGFR decreased in Sertoli and interstitial cells upon DES treatment, while the signal appeared also less robust in gonocytes. Similarly, PDGFRß was noticeably expressed in gonocytes (Fig. 6, A and C, bottom panel) and at low levels in Sertoli cells of control samples. At the doses examined, all estrogens appeared to strengthen the immunoreactive signal present in gonocytes and Sertoli cells. Here again, reactions performed with preimmune rabbit IgGs gave background levels of signal independently of the treatments (Fig. 6B).
|
Effects of Prenatal Exposure to Genistein on the Levels of Phosphotyrosine-Containing Proteins in Neonatal Testis
Besides being an estrogenic compound, genistein is also known to inhibit a wide range of tyrosine kinases in vitro. Thus, we examined whether prenatal exposure to genistein resulted in any changes in the profile of tyrosine-phosphorylated proteins present in testis, as compared with control samples, using an anti-phosphotyrosine antibody. These experiments showed that, in control samples, tyrosine-phosphorylated proteins were strongly expressed in the lateral surfaces of adjacent Sertoli cells, at the periphery of gonocytes, and as a more diffuse signal in Sertoli cell cytoplasm (Fig. 7A). Genistein treatment did not alter the levels of tyrosine-phosphorylated proteins present in most sections (Fig. 7B), although in some samples, there was a slight reduction in the signal between Sertoli cells (data not shown).
|
Expression of PDGFR and ß mRNAs and Proteins in Purified Gonocytes from Testis
To characterize the forms of PDGFR and ß expressed in neonatal gonocytes, pure populations of isolated gonocytes from control rats were prepared. This was achieved by adding a step of single-cell harvesting to the purification procedure that we previously reported [14], using a micromanipulation device that allowed us to collect individual gonocytes from an enriched cell suspension. The appearance of the gonocyte suspension is illustrated in Figure 8A. The purity of the gonocyte preparation was confirmed by analyzing the expression of mRNA species normally found in somatic cells, including P450scc, androgen binding protein, and smooth muscle actin, as markers of Leydig, Sertoli, and myoid cells, respectively. While the PCR products of somatic cell markers were clearly seen in RNA extracts from whole testis, they were absent in purified gonocytes (Fig. 8B), confirming that the gonocyte preparations were devoid of other cell types. Isolated gonocytes were found to express both PDGFR and ß mRNAs (Fig. 8C). The size of the PDGFR mRNA fragments amplified by RT-PCR was identical to what was expected from the rat PDGFR mRNA published sequence, and its sequence was found to be 100% homologous to that reported before [35]. Because the 3' end of the cDNA sequence of PDGFRß was not known, we initially determined its full-length sequence by screening a rat testis library. The resulting full-length PDGFRß cDNA sequence was deposited in the Genbank (GenBank accession no. AY090783) and was found to be 93% homologous to the published mouse sequence [36]. The RT-PCR fragment of PDGFRß mRNA amplified in neonatal gonocytes had a sequence identical to the corresponding sequence in adult rat testis. In addition, immunoblot analysis of protein extracts from purified gonocytes showed that gonocytes express the conventional form of PDGFR and two forms of PDGFRß (Fig. 8D). The high molecular weight form of PDGFRß, detected with an antibody recognizing the kinase domain of the receptor, appeared identical to the form present in Sertoli/myoid cell extracts and corresponding to the classical PDGFRß. However, an antibody raised against the C-terminal of PDGFRß failed to react with this band, suggesting differences in its primary structure compared with the classical form of the receptor. The second form of PDGFRß present in gonocytes had a smaller size (
76 kDa) and was recognized by an antibody raised against the C-terminal domain of PDGFRß but not by the antibody generated against the kinase domain. In light of these data, we performed additional histochemical analysis of paraffin sections of whole testes with the antibody raised against the kinase domain of PDGFRß. Unfortunately, this antibody did not react under the conditions used (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our results showed that in utero exposure to BPA, genistein, and coumestrol had qualitatively similar effects in neonatal testis, increasing the expression of the PDGF receptors and ß in a dose-dependent manner, whereas DES induced an increase in mRNA expression at low doses but was inhibitory at high doses. Interestingly, DES was previously reported to have a biphasic effect on testicular maturation when administered to neonatal rats [16]. Considering that DES binds both ER and ß with high affinity [1, 23, 24] and that both receptors are present in various types of pre- and postnatal testicular cells, it is possible that DES acts simultaneously on ER and ß and several target cells. The doses-responses observed here suggested a ranking of potencies for DES, genistein, and BPA in vivo that was analogous to the ranking deduced from in vitro studies [23, 24]. However, coumestrol appeared to be less potent in vivo than genistein, in contrast with the results of in vitro studies [23, 24, 37, 38]. This discrepancy may reflect differences in the absorption and metabolism of the compounds in vivo or in their target molecules. Indeed, in vitro assays cannot mirror accurately the in vivo effects of compounds, which depend on the route of administration and bioavailability [39–41], on biotransformation [42, 43], and on the time of exposure [44]. Although BPA and genistein were used at doses exceeding their estimated environmentally relevant doses, we obtained significant effects with low doses (0.1 mg kg-1 day-1 genistein; 1 mg kg-1 day-1 BPA) corresponding to levels that were close (2- to 4-fold higher) to those encountered in our environment or diet [42, 45].
Our finding that BPA and genistein induced similar responses at both the mRNA and protein levels suggests that the two compounds have the same targets in fetal testis. In contrast, neonatal exposure to these compounds appeared to provoke opposite effects and thus to target different molecules [16]. Indeed, estrogenic compounds have been shown to interact with various components of the ER regulatory complexes and to possibly act as natural selective ER modulators [46, 47]. Moreover, the strong responses obtained following gestational exposure to BPA suggest that rats may be more susceptible to this compound during gestation than after birth, possibly due to developmental differences, such as those seen in gonocytes between pre-and postnatal days [14, 48]. In view of the complexity and multiplicity of the molecules involved in ER regulation, both at the level of the cytosolic and nuclear transcriptional complexes, it is not surprising that various estrogenic compounds would differentially interact with some of these regulatory molecules in function of the time of exposure.
Besides their ER-dependent effects, estrogens have been shown to exert properties nonrelated to ER, such as their ability to act as antioxidant molecules [49]. In the case of genistein, it is commonly used as a tyrosine kinase inhibitor in vitro [50]. The present study did not show consistent differences in the total levels of tyrosine-phosphorylated proteins in testis between control and genistein-treated rats. However, the possibility that the phosphorylation of specific proteins was inhibited by genistein treatment cannot be excluded because the method used here detected only the global phosphotyrosine immunolocalization in the tissue and not the levels of tyrosine-phosphorylation of individual proteins. A closer examination of the phosphorylation profile of isolated testicular proteins or the measurement of specific tyrosine kinase activities would be required to answer to this question.
In situ hybridization analysis confirmed the RT-PCR data obtained on whole testis for BPA and genistein but only the increase in PDGFRß for coumestrol. However, it should be noted that the dose of 50 mg kg-1 day-1 coumestrol used in these experiments was found to be suboptimal in dose-response studies. Concerning DES, the samples examined by in situ hybridization showed a decrease of PDGFR mRNA that was not found in the pooled RT-PCR results of many samples. Nonetheless, these experiments clearly demonstrated that the alterations in PDGFRs expression were not restricted to a single cell type but rather occurred in all testicular compartments and on different cell types. However, the in vivo paradigm used here did not allow differentiating between primary and secondary targets of estrogens. Although it is generally believed that the primary targets of estrogen in testis are the somatic cells [11, 51], the fact that ERß is expressed in gonocytes [17, 18, 52] and that estradiol induces proliferation in isolated gonocytes [14] suggests that estrogens might also exert a direct effect on germ cells. The recent generation of ERß knockout mice in which male offspring were found to be fertile [53], in contrast with the ER knockout male mice, which are infertile [54], would suggest that the ER rather than ß is regulating testicular development. Thus, the role of the ERß in Sertoli cells and gonocytes remains to be determined. Recent studies have revealed the existence of new forms of ER that can bind several estrogenic compounds, heterodimerize with ER and ß, and induce transcription [37]. The existence of such alternative forms of ER in testis remains to be determined because they could potentially support functions normally attributed to the classical ER and ß, which would explain the maintenance of reproductive functions that were observed in ERß knockout mice.
The present study revealed that PDGFR and ß were present in rat gonocytes in the days preceding and following birth. Moreover, PDGFRß was recently found in human fetal testicular Leydig cells and some gonocytes [55]. At this point, we do not have an indication for the role of PDGF in fetal gonocytes, while our previous studies showed that neonatal gonocytes proliferate in response to PDGF [14]. While low levels of PDGFR and ß mRNAs and proteins were found in neonatal gonocytes from control sections, gestational exposure to estrogens induced a large increase of PDGFR and ß mRNAs in testicular interstitium as well as in the center of the seminiferous cords in the case of PDGFRß. In contrast, immunodetection of the proteins revealed a profound increase of PDGFRß but not in gonocytes and only limited increases in somatic cells. At the present time, we do not have an explanation for this discrepancy, which may be related to posttranscriptional events limiting the rate of translation or affecting receptor turnover.
RT-PCR and immunoblot analysis of purified neonatal gonocytes further confirmed the expression of PDGFRs in germ cells. Interestingly, we found that an antibody raised against the kinase domain of PDGFRß reacted with a form of PDGFR in somatic cells and in gonocytes that had the same molecular weight as the conventional receptor, while an antibody raised against the C-terminal domain of PDGFRß gave different immunoreactive bands for somatic and germ cells. Indeed, this antibody recognized the classical form of PDGFR in somatic cells but not in gonocytes, where it detected instead the presence of a lower molecular weight form of PDGFRß. Considering that the large form of PDGFRß did not react with this antibody and that the lower form did not react with an antibody against the kinase domain, in contrast with the band in somatic cells that was recognized by both antibodies, one might conclude that neonatal gonocytes express two forms of PDGFRß, high- and low-molecular weight forms, that both differ from the classical form in specific domains of their sequence. Because only one of the antibodies worked for the immunohistochemistry of the paraffin-embedded testes, we do not know at present if both forms are affected simultaneously by estrogen treatment. Interestingly, tumoral testicular germ cells have been shown to express mRNAs encoding for variant forms of PDGFR, and the question was raised whether nonpathological germ cells at early developmental stages might express such smaller size PDGFRs [56]. The characterization of the new isoforms of PDGFRß expressed by neonatal germ cells in normal conditions and following estrogen treatment are in progress in our laboratory.
Prenatal and neonatal estradiol treatments have previously been reported to induce changes in testis that were dose-dependent [57, 58] but also strain-dependent [58]. In the present study, we found that prenatal administration of estradiol induced only a small but significant increase in PDGFR mRNA but not in PDGFRß. This may reflect the ability of pregnant rats to regulate their estradiol homeostasis and to maintain the free hormone at physiological levels via interaction with plasmatic steroid binding proteins, in contrast with other estrogens that bind poorly to the carrier proteins [39]. However, this may also indicate that the exogenous estrogens interact with additional molecules that are not regulated by the natural hormone. Although these compounds have all been characterized for their interaction with ERs [23, 24], some, such as coumestrol, have been shown to exert antiestrogenic effects in specific tissues [59] while others have been found to interact with ER-associated molecules [46, 47] or, as in the case of genistein, with transduction pathways unrelated to ER [50]. In this context, alternative effects have also been documented for several natural steroid hormones, which did not involve binding to the classical steroid receptors [60]. Moreover, both DES and genistein were shown to modulate intracellular calcium levels in mammalian cells by yet unknown mechanisms [61]. Thus, it cannot be excluded that the responses observed here might be due to nonestrogenic effects of the compounds.
Our results showing that most prenatal gonocytes express PDGFRs differed noticeably from the situation in 3-day-old pups, where only a portion of gonocytes expressed these receptors. Considering that neonatal gonocyte proliferation occurs during a period of 2 days, affecting only a small portion of gonocytes at any given time [14, 48], and that more mature germ cells no longer express PDGFRs [14, 55], this suggests that PDGFRs expression is physiologically downregulated in gonocytes that have completed their division and have progressed to the next stage of maturation. In this context, naturally circulating estrogen during pregnancy would participate in the regulation of PDGFRs expression in the late fetal period. An increase in estradiol and PDGF production by neonatal Sertoli cells, as those previously described [62, 63], would then result in the concerted activation of the PDGF pathway after birth. However, prenatal exposure to excessive levels of estrogens could disturb these cellular events, not only increasing the expression of both receptors but also antagonizing a physiological process of downregulation of PDGFRs and resulting in the alteration of the paracrine interaction between Sertoli and germ cells in neonatal testis. Further experiments will be needed to determine the consequences of the alterations of PDGFRs on gonocyte functions, including the study of their proliferation and their subsequent maturation.
In conclusion, this study showed that gestational exposure to estrogenic compounds induced changes in the expression of the PDGFR and ß in the testis of neonatal rats, suggesting that PDGF functions in testicular development may be physiologically regulated by estrogen. To our knowledge, this is the first report of an interaction between estrogen and PDGF receptors in testis. Interestingly, an earlier study had described the occurrence of a similar interaction in the uterus of immature mice exposed to DES, suggesting the existence of an autocrine/paracrine loop between estrogens and PDGF [30]. This study also suggests that the effects of estrogenic compounds on testicular development might not be all mediated by the same molecule but rather that they interact with different proteins and testicular cell types, although gonocytes appear to be a common target of the compounds. Considering that testicular gonocytes are the precursors of spermatogonia, it is conceivable that perturbations of their development, either direct or indirect, may result in impaired spermatogenesis and infertility in the adult. However, more studies will be needed to determine if there is a relationship between the estrogen-induced overexpression of the PDGF receptors and the decrease in fertility and formation of testicular tumors that have been associated with estrogen exposure.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Martine Culty, Department of Cell Biology, Georgetown University School of Medicine, 3900 Reservoir Road NW, Washington, DC 20057. FAX: 202 687 7855; e-mail: cultym{at}georgetown.edu
3 These authors equally contributed to this work
Received: 21 July 2002.
First decision: 15 August 2002.
Accepted: 19 September 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and -ß mRNAs in the male reproductive system of the rat as revealed by in situ hybridization. J Mol Endocrinol 2001 26:165-174[Abstract]- and ß-subunits in the mouse uterus and vagina: potential mediators of estrogen action. Endocrinology 1995 136:2325-2340[Abstract]
