Androgens and Postmeiotic Germ Cells Regulate Claudin-11 Expression in Rat Sertoli Cells
http://www.100md.com
内分泌学杂志 2005年第3期
Institut National de la Santé et de la Recherche Médicale, Unité 407, Oullins F-69921; Faculté de Médecine Lyon-Sud (A.F., M.M., A.B., A.H., S.C., M.B.), Lyon F-69921, France; BayerCorpScience (R.B.), Sophia-Antipolis F-06903; and Galderma (F.C.), Sophia-Antipolis F-06560, France
Address all correspondence and requests for reprints to: Dr. Mohamed Benahmed, Institut National de la Santé et de la Recherche Médicale, Unité 407, Faculté de Médecine Lyon-Sud, BP 12, 69921 Oullins Cedex, France. E-mail: benahmed@grisn.univ-lyon1.fr.
Abstract
In the present study we investigated whether fetal exposure to flutamide affected messenger and protein levels of claudin-11, a key Sertoli cell factor in the establishment of the hemotesticular barrier, at the time of two key events of postnatal testis development: 1) before puberty (postnatal d 14) during the establishment of the hemotesticular barrier, and 2) at the adult age (postnatal d 90) at the time of full spermatogenesis. The data obtained show that claudin-11 expression was inhibited in prepubertal rat testes exposed in utero to 2 and 10 mg/kg·d flutamide. However, in adult testes, the inhibition was observed only with 2, and not with 10, mg/kg·d of the antiandrogen. It is shown here that these differences between prepubertal and adult testes could be related to dual and opposed regulation of claudin-11 expression resulting from positive control by androgens and an inhibitory effect of postmeiotic germ cells. Indeed, testosterone is shown to stimulate claudin-11 expression in cultured Sertoli cells in a dose- and time-dependent manner (maximum effect with 0.06 μM after 72 h of treatment). In contrast, postmeiotic germ cells potentially exert a negative effect on claudin-11 expression, because adult rat testes depleted in spermatids (after local irradiation) displayed increased claudin-11 expression, whereas in a model of cocultured Sertoli and germ cells, spermatids, but not spermatocytes, inhibited claudin-11 expression. The apparent absence of claudin-11 expression changes in adult rat testes exposed to 10 mg/kg·d flutamide therefore could result from the antagonistic effects of 1) the inhibitory action of the antiandrogen and 2) the stimulatory effect of the apoptotic germ cells on claudin-11 expression. Together, due to the key role of claudin-11 in the hemotesticular barrier, the present findings suggest that such regulatory mechanisms may potentially affect this barrier (re)modeling during spermatogenesis.
Introduction
SEVERAL RESEARCHERS HAVE shown that in experimental rodent models an endocrine (androgen) disruption during gestational life alters androgen-dependent reproductive development and function (1, 2, 3, 4 ; for reviews, see Refs.5 and 6). One of the most frequently used experimental models of antiandrogens is the potent nonsteroidal antiandrogen flutamide, gestational exposure to which causes, in particular, hypospadias, undescended testes, and impaired spermatogenesis in male offspring (7, 8, 9, 10, 11). Although the molecular and cellular mechanisms underlying hypospermatogenesis in the adult rat exposed in utero to flutamide remain poorly understood, we previously reported that such an alteration could be related to a long-term apoptotic cell death process in meiotic and postmeiotic germ cells, with a significant increase in the expression and activation of effector caspase-3 and -6 (12), and an alteration in the balance between pro- and antiapoptotic Bcl-2 related molecules in favor of proapoptotic proteins (13). Although the death process that follows gestational exposure to flutamide does not affect adult somatic cells, their expression of the androgen receptor (AR) makes them a direct target of testosterone (and thus of flutamide) (14, 15), and these findings suggest that in seminiferous tubules, Sertoli cells (and peritubular myoid cells?) may produce androgen-regulated factors involved in the survival of germ cells.
Claudin-11, also known as oligodendrocyte-specific protein (OSP), belongs to the claudin family of intrinsic transmembrane proteins (16). First described in the myelin sheaths of the central nervous system (17), studies have shown that it is specifically expressed in oligodendrocytes in the brain and Sertoli cells in the testis, where it is responsible for the formation of specific parallel tight junction strands (16, 18, 19). In the mouse testis, claudin-11 has been shown to be hormonally regulated (FSH and testosterone) (20, 21) and is a key factor in establishment of the hemotesticular barrier. Indeed, testicular claudin-11 expression is at its maximum between postnatal d 6 and 16, during the formation of the hemotesticular barrier (20), and mice invalidated for this gene are infertile, with the loss of claudin-11 expression causing a disruption of this barrier, leading to an arrest in spermatogenesis (19).
In the present study we investigated whether fetal exposure to flutamide affected claudin-11 messenger and protein levels in rat Sertoli cells at the time of two key events in postnatal testis development: 1) before puberty (postnatal d 14) during the establishment of the hemotesticular barrier, and 2) in adulthood (postnatal d 90) at the time of full spermatogenesis when all germ cell types are present. Claudin-11 mRNA and protein levels were also studied in the brain of adult male rats exposed in utero to the antiandrogen.
Materials and Methods
Materials
Flutamide was purchased from Sigma-Aldrich Corp. (Meylan, France), dissolved in an aqueous solution of methylcellulose 400 (Fluka, Mulhouse, France) at 0.5% (wt/vol), and stored at 5 C (±3 C) for a maximum of 1 wk. Sigma-Aldrich Corp. was also the supplier for BSA, hexanucleotide primers, insulin, transferrin, gentamicin, HEPES, sodium bicarbonate, trypsin, and deoxyribonuclease I. Collagenase-dispase was obtained from Roche (Mannheim, Germany). DMEM-Ham’s F-12 medium (DMEM/F12), deoxy-NTPs (dNTPs), Moloney murine leukemia virus reverse transcriptase kit, and TRIzol were obtained from Invitrogen Life Technologies, Inc. (Eragny, France). Taq polymerase was obtained from Promega Corp. (Lyon, France). Amersham Biosciences (Little Chalfont, UK) was the supplier of [-33P]dATP. Oligonucleotides were purchased from Invitrogen Life Technologies, Inc. (Groningen, The Netherlands).
Animals
Virgin female Sprague Dawley rats (Charles River Laboratories, Saint Aubin les Elbeuf, France) were individually housed under controlled conditions of lighting (12 h of light, 12 h of darkness), temperature (22 ± 2 C), humidity (55 ± 15%), and ventilation (15 air changes/h) and were given free access to water and feed (certified rodent pellet diet A04C, UAR, Villemoisson-sur-Orge, France). Females were mated on a one to one basis with males of the same strain, purchased from the same supplier. Day 0 of gestation (GD0) was the day a vaginal sperm plug was observed. Before mating and during gestation, dams were housed in suspended stainless steel wire mesh cages. Shortly before parturition and during lactation, dams were housed in Makrolon cages (Charles River) with soft wood bedding.
Pregnant rats were administered daily the vehicle (methylcellulose) alone or flutamide by gavage from GD10 to the day before delivery (GD21 or GD22). Animals were administered flutamide at doses of 0, 0.4, 2, and 10 mg/kg body weight·d (adjusted daily for body weight). Dams were weighed daily from GD10 to the day of delivery. At birth, each pup was sexed, weighed, and identified. The male offspring received no flutamide treatment and were raised until study termination (postnatal d 14 or 90), when rats were euthanized by CO2 asphyxiation. The position of each testis was carefully noted. Moreover, each testis was weighed before being fixed or frozen. Only bilateral descended testes were studied in the present report. All studies using animals were conducted in accordance with current regulation and standards approved by Institut National de la Santé et de la Recherche Médicale (French Institute for Health and Medical Research) animal care committee.
Irradiation
Adult Sprague Dawley (90 d old) rats purchased from Iffa-Credo (L’Arbresle, France) were housed in controlled conditions of lighting (12 h of light, 12 h of darkness), temperature (22 ± 2 C), humidity (55 ± 15%), and ventilation (15 air changes/h) and were given free access to water and feed (certified rodent pellet diet, AO4C, UAR). Before the experimental procedure, an acclimation period of 1 wk was allowed. All rats were anesthetized by ip injection of sodium phenobarbital (45 mg/kg; Sanofi Santé Animale, Libourne, France) 20 min before irradiation. The scrotum of each rat was irradiated by x-ray at a dose of 9 Gy. A 120-kv x-ray beam with 2-mm Al filtration was generated by a Stabilipan (Siemens AG, Erlangen, Germany) orthovoltage machine. A 4-cm diameter lead collimator was used to selectively irradiate both testes while protecting the other organs at risk in the rats. The distance between source and skin was 40 cm. The dose was prescribed and specified at 2.5 mm under the skin of the scrotum, at the center of the testes. Control animals received anesthesia and sham irradiation. Irradiated and sham-irradiated animals were euthanized by CO2 inhalation 10, 26, or 45 d after the experiments. One group of control animals was killed 10 d after sham irradiation, and the other group was killed 45 d after sham irradiation. Each control and treatment group consisted of six animals. For each rat, the left testis was removed and fixed for 24 h in Bouin’s fixative, and the right testis was removed for RNA and protein extractions.
Isolation and culture of Sertoli cells
Primary Sertoli cells were isolated from 20-d-old Sprague Dawley rats, as described by Dorrington et al. (22). Briefly, decapsulated testes were submitted to collagenase dissociation (0.5 mg/ml, 5 min at 30 C) in DMEM/F12 (1:1, vol/vol) medium (1.2 mg/ml sodium bicarbonate, 15 mM HEPES, and 20 μg/ml gentamicin) containing deoxyribonuclease I (0.05 mg/ml). At the end of enzymatic dissociation, testicular cells were washed three times by submitting them to gravity sedimentation (3–5 min), and supernatants were removed. The pellets containing the sedimented tubules were also dissociated with a collagenase-dispase treatment, as described above, until small clumps resulted. Cells were then submitted to gravity sedimentation (10–15 min), supernatants were removed, and the sedimented clumps of Sertoli cells were also washed by centrifugation (200 x g, 10 min). Sertoli cell pellets were resuspended in DMEM/F12 medium (supplemented with 5 μg/ml transferrin, 2 μg/ml insulin, and 10 μg/ml vitamin E), and cells were plated on Falcon petri dishes (BD Biosciences, Franklin Lakes, NJ; 100-mm diameter, 7 x 106 cells/dish) and cultured at 32 C in a humidified atmosphere of 5% CO2-95% air. This procedure led to a Sertoli cell population free from Leydig cells or mature germ cells, but which contained approximately 15–20% peritubular myoid cells (data not shown). For the coculture studies, 2 x 106 Sertoli cells were seeded in 6-cm diameter petri dishes. Testicular germ cells (spermatocytes and spermatids) were isolated from adult rats (90 d old) as described by Boussouar et al. (23). Three groups of cultured cells were generated: Sertoli cells cultured alone without germ cells, Sertoli cells cultured with spermatocytes (2 x 106 cells), and Sertoli cells cultured with spermatids (2 x 106 cells). The cells were collected after 72 h of culture. Experiments were repeated at least three times with independent cell preparations.
Total RNA extraction
Total RNAs were extracted using TRIzol, a monophasic solution of phenol and guanidine isothiocyanate, following an improvement of the single-step RNA isolation method developed by Chomczynski and Sacchi (24). The final amount of RNA was estimated by spectrophotometry at 260 nm.
Coamplification RT-PCR with an endogenous control
Coamplification RT-PCR (25) was performed to determine the mRNA levels of claudin-11, AR, and FSH receptor (FSH-R). Briefly, approximately 2 μg total RNA were reverse-transcribed into cDNAs (cDNAs) for 1 h at 37 C using Moloney murine leukemia virus reverse transcriptase (10 U/μl) in 1x first strand buffer, random hexanucleotides as primers (5 μM), dNTPs (250 μM), and dithiothreitol (10 μM). PCRs were then performed on 2 μl RT product, using the appropriate sense- and antisense-specific primers, Taq polymerase (0.5 U), 1x PCR buffer, 2.5 mM MgCl2, dNTPs, and 0.75 μCi [-33P]dATP. Claudin-11 (0.055 or 1 μM) was coamplified with GATA-6 (1 μM) or with the ?-actin (0.1 μM) primer pair, GATA-6 (1 μM) was coamplified with ?-actin (0.04 μM), FSH-R (1 μM) was coamplified with GATA-6 (1 μM), and AR (0.5 μM) was coamplified with hypoxanthine phosphoribosyltransferase (HPRT; 0.5 μM). Coamplification with ?-actin or HPRT was performed to check that equal amounts of cDNAs were amplified in each reaction tube, and coamplification with the GATA-6 gene was performed to compare claudin-11 and FSH-R mRNA levels with those of another Sertoli cell-specific gene, so as to detect eventual changes in the proportion of this cell type in the testes from flutamide-treated or irradiated animals. PCR mixes were submitted to an initial denaturing step at 95 C, followed by X cycles consisting of 30-sec denaturation at 95 C, 30-sec annealing at melting temperature (Tm), and 30-sec extension at 72 C, and the reaction ended with a final extension step at 72 C: claudin-11: X, 19 cycles; Tm, 55 C; GATA-6: X, 19 cycles; Tm, 55 C; AR: X, 25 cycles; Tm, 55 C; and FSH-R: X, 24 cycles; Tm, 65 C. Primers were as follows: for claudin-11: upstream primer, 5'-GATTGGCATCATCGTCACAACG-3'; downstream primer, 5'-AGCCAGCAGAATAAGGAGCAAC-3' (339 bp); for GATA-6: upstream primer, 5'-GTGCCAACCCTGAGAACAGT-3'; downstream primer, 5'-TGGACTGCTGGACAAAATCA-3' (198 bp); for ?-actin: upstream primer, 5'-TTGCTGATCCACATCTGCTG-3'; downstream primer, 5'-GACAGGATGCAGAAGGAGAT-3' (146 bp); for AR: upstream primer, 5'-ATTGTCCATCGTGTCGTCTCC G-3'; downstream primer, 5'-GAGTTGACATTAGTGAAGGACC-3' (447 bp); for HPRT: upstream primer, 5'-CCTGCTGGATTACATTAA AGC-3'; downstream primer, 5'-GTCAAGGGCATATCCAACAAC-3' (354 bp); and for FSH-R: upstream primer, 5'-CTCATCAAGCGACACCAAGA-3'; downstream primer, 5'-ACCTTGAGGGAGGCAGAAAT-3' (108 bp). PCR products were resolved on 8% polyacrylamide gels, which were exposed to a Storage Phosphor Screen (Packard, Meriden, CT), and the signals were analyzed using Cyclone OptiQuant software (Packard). Results from at least three separate experiments were used for statistical analysis.
PCR analyses were carried out from the logarithmic phase of amplification, and different cycle numbers were tested for each primer pair to determine the minimum number of cycles necessary to detect the PCR product. Primers were designed inside separate exons to avoid any bias caused by residual genomic contamination. For all primers, no amplification was observed when PCR was performed on RNA preparations. Finally, PCR-amplified products were checked by direct sequencing using an automated sequencer (ABI PRISM 310, Applied Biosystems, Foster City, CA).
Western blotting analysis
Proteins (40 μg) from whole adult testis or brain were separated on 10% (AR) or 12% (claudin-11) sodium dodecyl sulfate-polyacrylamide gels and electrophoretically transferred to nitrocellulose membranes using 25 mM Tris and 185 mM glycine for claudin-11 or 15 mM Tris and 120 mM glycine for AR, pH 8.3, containing 20% methanol. Transfer was performed at a constant voltage of 100 V for 1 h. After transfer, membranes were blocked in Tris-buffered saline containing 0.05% Tween (TBS-T) containing 3% BSA (claudin-11) or 5% milk (AR), and incubated with the primary antibody [1:2000 for anticlaudin-11 (Covalab, Lyon, France) or 1:200 for anti-AR (sc-816, Santa Cruz Biotechnology, Inc., Santa Cruz, CA)] in TBS-T overnight at 4 C. Membranes were then incubated with horseradish peroxidase-labeled goat antirabbit (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA; 1:2000 in TBS-T) for 1 h at room temperature, and bound antibodies were subsequently detected using a chemiluminescence detection kit (Covalab) and Biomax MR films (Eastman Kodak Co., Rochester, NY). Protein concentrations were determined by the Bradford assay. Reprobing the blot with a rab-bit immunoglobulin G antiactin (20-33, Sigma-Aldrich Corp., L’Isle d’Abeau, France; concentration, 1:500 in TBS-T) checked protein loading.
Terminal deoxynucleotidyltransferase-mediated deoxyuridine triphosphate nick end labeling (TUNEL)
Testes were immediately fixed for 24 h in Bouin’s fluid, stepwise dehydrated in graded ethanol baths, and embedded in paraffin. Paraffin sections of Bouin-fixed testes were sectioned at 5 μm. The sections were mounted on positively charged glass slides (SuperFrost Plus, Menzel-Glaser, Braunschweig, Germany), deparaffinized, hydrated, treated for 20 min at 93–98 C in citric buffer (0.01 M, pH 6), rinsed in osmosed water (twice, 5 min each time), and washed (twice, 5 min each time) in TBS, pH 7.8. TUNEL reaction was performed as previously described (26), and slides were counterstained for 2 min with Mayer’s hematoxylin. In each rat testis, at least 100 random seminiferous tubules were counted. The results are expressed as the number of TUNEL-positive cells per 100 seminiferous tubules.
Hormone assay
Blood samples, taken from 14- or 90-d-old rats selected in each control or treated group across litters, were taken by retroorbital puncture after isoflurane anesthesia. Plasma samples were kept frozen until testosterone or FSH was analyzed using a specific RIA. Specific kits were supplied by Amersham Biosciences (Orsay, France) for FSH and by Im-munotech (Marseille, France) for testosterone. Packard provided the RIASTAR program (Canberra Co., Meriden, CT).
Data analysis
Data are expressed as the mean ± SEM. At least four different male offspring (n = 4–11/condition) from different litters were used. For statistical analysis, one-way ANOVA was performed to determine differences among all groups (P < 0.05), and the Bonferroni/Dunn posttest was performed to determine the significance of the differences between pairs of groups. P < 0.05 was considered significant. The statistical tests were performed on StatView software (version 5.0, SAS Institute, Inc., Cary, NC) on a Macintosh computer (Apple Computer, Cupertino, CA).
Results
Effects of exposure to flutamide during fetal life on claudin-11 mRNA levels in prepubertal and adult rat testes
In these experiments we analyzed the expression of claudin-11 in prepubertal (14 d old) and adult (90 d old) rat testes exposed in utero to 0.4, 2, and 10 mg/kg·d flutamide. Data presented in Fig. 1A show that claudin-11 mRNA levels were significantly reduced in testes from 14-d-old rats treated in utero with 2 (35% decrease; P < 0.0001) and 10 (50% decrease; P = 0.0001) mg/kg·d flutamide. In parallel, a decrease in claudin-11 protein was observed in a significant (30% decrease; P = 0.008) manner at 10 mg/kg·d flutamide (Fig. 1B).
FIG. 1. Effects of in utero exposure to flutamide on claudin-11 mRNA and protein levels in rat testis. Claudin-11 mRNA levels were determined by coamplification RT-PCR in prepubertal (A) and adult (C) rat testes exposed in utero to vehicle alone (0) or 0.4, 2, or 10 mg/kg·d flutamide. Claudin-11 protein levels were measured by Western blotting in prepubertal (B) and adult (D) rat testes exposed in utero to vehicle alone or to the three doses of the antiandrogen. The data show representative autoradiograms (upper panel), and the histograms (lower panel) are expressed as the mean ± SEM determined from at least six (coamplification RT-PCR) or four (Western blotting) animals from different litters. Results are represented as a percentage of the ratio (claudin-11/GATA-6 mRNA or claudin-11/?-actin protein) detected in the control (vehicle treated) animals.
Figure 1C shows that in the adult (90 d old) testes treated in utero with 2 mg/kg·d flutamide, claudin-11 mRNA levels were significantly reduced by 25% (P < 0.0001), whereas at the dose of 10 mg/kg·d flutamide, such a decrease in claudin-11 mRNA levels was no longer observed in adult testes compared with prepubertal (14 d old) testes.
Data obtained by Western blotting analysis showed similar results for protein levels (Fig. 1D). Indeed, in adult rat testes exposed in utero to 2 mg/kg·d flutamide, a 50% decrease in claudin-11 protein levels was found (P < 0.005), and again, in the 10 mg/kg·d-treated rats, this decrease in testicular claudin-11 levels was not observed.
To determine the specificity of claudin-11 expression changes in Sertoli cells, we evaluated in parallel AR mRNA and protein levels, FSH-R mRNA levels, and the expression of the gene coding for GATA-6, a transcription factor member of the GATA-binding protein family that is specifically expressed in Sertoli cells (27). AR mRNA (data not shown) and protein levels in prepubertal (Fig. 2A) and adult (Fig. 2B) rat testes were not affected after in utero exposure to the various doses of flutamide. Similarly, FSH-R mRNA levels in prepubertal (Fig. 2C) and adult (Fig. 2D) rat testes were not modified by fetal treatment with the antiandrogen, as were GATA-6 mRNA levels in prepubertal (data not shown) or adult (Fig. 2E) rat testes.
FIG. 2. Effects of in utero exposure to flutamide on AR protein levels, FSH-R, or GATA-6 mRNA levels in rat testis. AR protein levels were determined in prepubertal (A) and adult (B) rat testes exposed in utero to vehicle alone (0) or 0.4, 2, or 10 mg/kg·d flutamide. The levels of mRNA were determined for FSH-R in prepubertal (C) and adult (D) testes and for GATA-6 in adult (E) rat testes exposed in utero to vehicle alone (0) or to 0.4, 2, or 10 mg/kg·d flutamide. The data show representative autoradiograms (upper panel), and the histograms (lower panel) are expressed as the mean ± SEM determined from at least six (coamplification RT-PCR) or four (Western blotting) animals from different litters. Results are represented as a percentage of the ratio (AR/?-actin protein, FSH-R/GATA-6 mRNA, and GATA-6/?-actin mRNA) detected in control (vehicle-treated) animals.
Because Sertoli cell activity is predominantly under androgen and FSH control, the changes in claudin-11 expression could be related to modifications in plasma testosterone and/or FSH levels. However, the data in Table 1 show that plasma testosterone levels were not significantly affected in either prepubertal (14 d old) or adult (90 d old) rats exposed in utero to flutamide. Similarly, FSH levels in adult rats exposed in utero to flutamide were not altered (Table 2). Taken together, the data in Fig. 2 and Tables 1 and 2 indicate that the alterations in claudin-11 expression in prepubertal and adult testes exposed in utero to flutamide were not related to altered androgen and/or FSH plasma levels and/or changes in AR/FSH-R expression.
TABLE 1. Plasma testosterone levels in prepubertal and adult animals exposed in utero to flutamide
TABLE 2. Plasma FSH levels in adult animals exposed in utero to flutamide
Because the inhibition of claudin-11 expression observed at a dose of 10 mg/kg·d in prepubertal testis was not evident in adult testis, it was hypothesized that testicular claudin-11 expression could be under two (opposite) regulations: 1) a positive regulation exerted by androgens, which could explain the inhibitory effect of flutamide specifically observed at 2 mg/kg·d; and 2) an additive negative effect exerted by postmeiotic germ cells, present in the adult, but not in the prepubertal, testis. In the following experiments, we therefore tested the regulatory effects of testosterone and germ cells on claudin-11 expression in the testis.
Claudin-11 expression is under positive control of testosterone in testis
To study the regulation by androgens of claudin-11 expression, messenger and protein levels were evaluated in a model of cultured rat Sertoli cells. Sertoli cells were treated with various concentrations of testosterone (7 nM to 3.5 μM; 2–1000 ng/ml) for various periods (4–96 h). The data in Fig. 3A show that claudin-11 mRNA levels were up-regulated in a testosterone dose-dependent manner. The maximal effect was obtained with 583 nM (170 ng/ml) testosterone (P < 0.001). Treatment with 875 nM (250 ng/ml) showed that the regulation of claudin-11 by testosterone was time dependent (Fig. 3B), with an increase in claudin-11 mRNA levels after 48 h of testosterone treatment and a maximal effect after 72 h (P < 0.0001). In control experiments we ensured that basal claudin-11 mRNA levels were not affected throughout the culture period (from 4–96 h, Fig. 3B). Similar results were obtained at the protein level (Fig. 3C); claudin-11 protein levels were increased by more than 2-fold after 72 h of treatment with 875 nM (250 ng/ml) testosterone (P < 0.001).
FIG. 3. Claudin-11 mRNA and protein levels in cultured Sertoli cells treated with testosterone. Cultured Sertoli cells were treated with increasing (0–1750 nM) doses of testosterone for 72 h (A) or without () or with () 875 nM (250 ng/ml) testosterone for different treatment times (4–96 h; B) before determining claudin-11 mRNA levels. Sertoli cells were cultured in the absence (0) or the presence of 875 nM testosterone for 72 h before evaluating claudin-11 protein levels (C). The data show representative autoradiograms (upper panel) and histograms (lower panel) and are expressed as the mean ± SEM from at least three different experiments. Results are represented as a percentage of the ratio of claudin-11/?-actin mRNA detected in the controls (untreated cells).
Claudin-11 expression is under negative control of postmeiotic germ cells in testis
To answer the question of whether claudin-11 expression could be regulated by germ cells, we evaluated claudin-11 expression in two models: 1) a model of locally irradiated (9 Gy) adult (90 d old) rat testis, and 2) a model of Sertoli cell-germ cell cocultures. The first model enabled sequential removal of premeiotic and postmeiotic germ cells (28, 29), thus allowing study of the possible regulation of claudin-11 expression in Sertoli cells by the different types of germ cells. Consistent with the literature (29), in control experiments, testes from rats killed 10 d after irradiation were practically devoid of spermatogonia; 26 d after irradiation, very few premeiotic germ cells were observed; and in testes from rats killed 45 d after irradiation, postmeiotic germ cells were very low in number, but spermatogonia were observed (data not shown).
Figure 4A shows an increase in testicular GATA-6 mRNA expression on d 26 (1.3-fold; P < 0.0001) and d 45 (1.5-fold; P < 0.0001) after irradiation. This increase in GATA-6 expression was observed in the testes depleted in the two most represented germ cell types (pachytene spermatocytes and spermatids). This would suggest that the increase in GATA-6 expression is probably related to the subsequent relative concentration of Sertoli cells. For this reason, we expressed claudin-11 mRNA levels as claudin-11/GATA-6 ratios in Fig. 4B. This figure shows that claudin-11/GATA-6 mRNA levels were similar to control levels in testes that were practically devoid of premeiotic germ cells (10 and 26 d after local irradiation), whereas claudin-11/GATA-6 mRNA levels were increased by 2.5-fold (P < 0.0001) in testes lacking postmeiotic germ cells (45 d after irradiation). For claudin-11 protein levels, Fig. 4C shows that a 2-fold increase (P = 0.0031) was found in the testes from rats killed 45 d after irradiation, compared with the sham-irradiated testes. Furthermore, in the second model of Sertoli cell-germ cell cocultures, a significant decrease (40%; P < 0.007) in claudin-11 mRNA levels was observed when Sertoli cells were cocultured with spermatids. In contrast, no changes in claudin-11 levels were evident in Sertoli cell-spermatocyte cocultures (Fig. 4D). Taken together, these data clearly suggest that postmeiotic germ cells exert an inhibitory effect on claudin-11 expression in Sertoli cells.
FIG. 4. Claudin-11 expression in adult rat testes after sequential removal of the different subtypes of germ cells and in Sertoli cell-germ cell (spermatocytes or spermatids) coculture. GATA-6 mRNA (A) and claudin-11 mRNA (B) and protein (C) levels were determined by coamplification RT-PCR and Western blotting in locally irradiated (9 Gy) adult rat testes, allowing specific and sequential removal of pre- and postmeiotic germ cells. Animals were killed 10, 26, or 45 d after local irradiation. Claudin-11 mRNA levels (D) were determined in Sertoli cell primary cultures or in Sertoli cells cocultured with either spermatocytes or spermatids. The data show representative autoradiograms (upper panel), and histograms (lower panel) are expressed as the mean ± SEM determined from at least six (coamplification RT-PCR) or four (Western blotting) animals from different litters. Results are represented as a percentage of the ratio (GATA-6/?-actin mRNA, claudin-11/GATA-6 mRNA, claudin-11/?-actin protein) detected in the control (vehicle-treated) animals.
Effects of exposure to flutamide during fetal life on claudin-11 mRNA and protein levels in adult rat brain
In contrast to that in testis, claudin-11 expression was not altered in adult rat brain after in utero exposure to the antiandrogen for all doses studied. Indeed, mRNA and protein levels in flutamide- treated animals were similar to those in control animals (data not shown).
Discussion
OSP/claudin-11 was first described in the myelin sheaths of the central nervous system by Bronstein et al. (17), where it represents 7% of total myelin proteins, making it a major component of central nervous system myelin (18). OSP cDNA was first identified as being expressed in oligodendrocytes by subtraction cloning and has been localized on human chromosome 3 (3q26.2-q26.3) (17). OSP was later also found to be specifically expressed in Sertoli cells of the testis, but not in the ovary (16, 20). Morita et al. (16) found that OSP shared sequence homology with the claudin superfamily of tight junction proteins, and like other claudins, this protein was found to induce the formation of tight junction strands when transfected in fibroblasts; it was therefore renamed claudin-11 (for a review on claudins, see Ref.30). The study by Gow et al. (19) of OSP/claudin-11-null mice confirmed the essential roles of this claudin in the establishment of specific tight junction strands in the brain and testis by showing both neurological and reproductive defects (hypospermatogenesis) in these mice.
In the present study we primarily investigated whether gestational exposure to flutamide affected claudin-11 mRNA and protein levels in rat Sertoli cells at the time of two key events in postnatal testis development: 1) before puberty (postnatal d 14), during establishment of the hemotesticular barrier, which starts on postnatal d 15–18 in the rat (31); and 2) in adulthood (postnatal d 90), at the time of full spermatogenesis when all germ cell subtypes are present. To study gene expression changes in this model, the doses of the antiandrogen were chosen to avoid or minimize important germ cell loss that may confound interpretation of the effect of flutamide on testicular gene expression (7). Indeed, alterations in testicular cellularity (i.e. the ratio of somatic cells to germ cells) may confound interpretation of the effects of the antiandrogen on gene expression in the various testicular cell types (for review, see Ref.32), specifically here on claudin-11 expression in the Sertoli cell population.
To ensure that the variations in claudin-11 levels were specific and linked to exposure to the antiandrogen, we studied the expression of another gene specifically expressed in Sertoli cells, the GATA-6 gene that codes for a member of the GATA-binding protein family and has been reported not to be hormonally regulated in the testis (27). Our results show that GATA-6 expression was not altered in 0.4, 2, and 10-mg/kg·d flutamide-treated testes in either prepubertal (14 d old) or adult (90 d old) animals, confirming the absence of hormonal dependence of this gene.
The data reported here show that claudin-11 mRNA levels in prepubertal (14 d old) and adult (90 d old) rat testes showed regulation similarities, because neither were affected by exposure to 0.4 mg/kg·d flutamide, but both declined with a dose of 2 mg/kg·d of the antiandrogen. However, at the dose of 10 mg/kg·d, a major difference in claudin-11 mRNA modulation was found depending on the age of the testis. Indeed, the inhibition of claudin-11 expression observed at the dose of 10 mg/kg·d in prepubertal testis was no longer evident in adult testis. We hypothesized that the effect of flutamide at 10 mg/kg·d in adult rat testes could be due either to a U-curve response to the antiandrogen, with an absence of response to flutamide at the dose of 10 mg/kg·d, or to an additive event occurring in the adult testis (the presence of postmeiotic germ cells) that could enhance claudin-11 levels and thus mask the inhibitory effect of the antiandrogen on testicular claudin-11 expression that was detected in prepubertal testes exposed to 10 mg/kg·d flutamide. Based on these observations, it was therefore hypothesized that testicular claudin-11 expression could be under two (opposite) regulations: 1) a positive regulation exerted by androgens, which could explain the inhibitory effect of flutamide specifically observed at 2 mg/kg·d; and 2) an additive negative effect exerted by the meiotic and/or postmeiotic germ cells present in adult, but not prepubertal, testis.
Using a model of cultured Sertoli cells, it was shown that, as in the mouse (21), claudin-11 expression is positively regulated by androgens in the rat testis, because claudin-11 mRNA and protein levels were up-regulated in a time- and dose-dependant manner in cultured rat Sertoli cells treated with various doses of testosterone at different times. The maximal stimulatory effect on mRNA levels was obtained with 583 nM (170 ng/ml) testosterone after 72 h of treatment. Although high, this concentration of testosterone is within the range of those reported for interstitial fluid collected from adult untreated rats (31, 33). It remains to be determined, however, whether testosterone affects claudin-11 gene transcriptional activity, mRNA stability, or both or claudin-11 protein stability and/or turnover and whether the stimulatory effect on claudin-11 is direct, because there is a relatively long period of latency between the time of treatment and the stimulatory effect observed on claudin-11 expression. This period of latency could also be due to participation of peritubular myoid cells in the androgen-dependent control of claudin-11 expression, because the Sertoli cell cultures used contained peritubular myoid cells.
In this study we also report that claudin-11 expression could be negatively regulated by factors produced by postmeiotic germs cells. Indeed, using a model of locally irradiated (9 Gy) rat testis that allows sequential removal of the different populations of germ cells, we found an increase in claudin-11 expression at both mRNA and protein levels in testes that are devoid of postmeiotic germ cells, but not in testes lacking spermatogonia or spermatocytes. These results were also supported by the use of an in vitro model of Sertoli cells cocultured with germ cells (spermatocytes or spermatids), in which claudin-11 mRNA levels were inhibited exclusively in the presence of spermatids. These observations clearly support an inhibitory effect exerted by postmeiotic germ cells on claudin-11 expression in Sertoli cells. The factors mediating the inhibitory effect of germ cells on claudin-11 expression are not known at present, but good candidates could be cytokines such as TNF and TGF?3. Indeed, these factors are produced by germ cells (34, 35, 36) and have receptors localized on Sertoli cells (35, 37, 38), and previous results published by our laboratory (20) and by Lui et al. (39, 40, 41) have shown that they are potent inhibitors of claudin-11 expression.
This dual and opposed regulation by androgens and postmeiotic germ cells of the expression of claudin-11 could therefore explain the apparent lack of effect of fetal treatment with the antiandrogen in testes from adult rats treated with 10 mg/kg·d, at which dose maximal (postmeiotic) germ cell death was observed (Table 3) (12, 13). Indeed, at this dose, both regulatory events could coexist, with a flutamide-induced down-regulation of claudin-11 expression that could be opposed to the loss of postmeiotic germ cells up-regulating claudin-11 expression.
TABLE 3. Apoptotic germ cell number in adult animals exposed in utero to flutamide
Because it was shown that claudin-11 was positively regulated by androgens, both AR expression and plasma testosterone levels were studied in postnatal animals exposed to the antiandrogen. No alterations in AR mRNA and protein levels were found in either prepubertal (14 d old) or adult (90 d old) rat testes exposed in utero to the three doses of flutamide tested. Hormone assay showed that serum testosterone levels were not significantly different from control levels in rats exposed in utero to flutamide either at a prepubertal age (14 d old), which has not been previously reported, or during adulthood (90 d old), and this is consistent with the literature (10, 11, 42). Taken together, these results suggest that specifically in rat testes exposed in utero to flutamide, alterations in the expression of claudin-11 by the antiandrogen are probably exerted at a post-AR level(s). This post-AR lesion(s) could be related to some changes in epigenetic mechanisms; for example in the methylation pattern of the androgen-dependent target gene promoters, as suggested by McLachlan for c-fos and lactotransferrin gene promoters in diethylstilbestrol-treated animals (43). Moreover, because during fetal exposure to the antiandrogen flutamide, AR expression in seminiferous tubules occurs mainly in peritubular myoid cells (44) (our unpublished data), it is possible that the participation of these cells is crucial to alterations in claudin-11 expression in postnatal Sertoli cells. Moreover, because claudin-11 is negatively regulated by FSH (20), we evaluated both plasma FSH levels and FSH-R expression, which were not found to be affected. Finally, the present findings of the positive control exerted by testosterone on claudin-11 expression suggest that testosterone could have an important role in modulation of the structure of the hemotesticular barrier and could be one of the factors controlling the opening and closing of this barrier during germ cell translocation into the adluminal compartment during the course of spermatogenesis. Such a positive regulatory action of the androgen is consistent with the report by Gye (21). The events contributing to the opening and closing of the hemotesticular barrier still remain unclear, but in view of our results and those of others, it could be hypothesized that both the endocrine system and the germ cells themselves, via factors they synthesize, could control the process of hemotesticular barrier complex dissolution and reformation that occurs during germ cell passage into the adluminal compartment. Indeed, the inhibitory effects of FSH, by decreasing claudin-11 expression (20), may favor the opening of junctions between Sertoli cells, allowing germ cell translocation into the adluminal compartment. The corollary of such a hypothesis is the existence of a stimulating factor, which could enhance claudin-11 levels, leading to closing of the hemotesticular barrier at the end of germ cell translocation. Based on our present findings and those of others, it could be suggested that testosterone, by stimulating claudin-11 [either directly (45) or indirectly via proteins interacting with claudin-11, such as integrin ?1, OAP-1 (46), or occludin (45)], could allow the reassembly of the barrier after germ cells have translocated. In contrast, postmeiotic germ cells, by regulating claudin-11 expression, may control translocation of premeiotic germ cells into the adluminal compartment to undergo meiosis, perhaps via (local) factors originating from germ cells, such as TNF (20) and TGF?3 (39, 40, 41), which (negatively) control claudin-11 in Sertoli cells.
Together, fetal androgen disruption appears to alter, in a complex manner, claudin-11 expression in postnatal rat testis. Indeed, although in the prepubertal testis, i.e. at the crucial moment of establishment of the hemotesticular barrier, a clear inhibitory effect is observed, in the adult testis, the apoptotic process that affects postmeiotic germ cells is likely to increase claudin-11 expression, as supported by the locally irradiated testis and coculture models we used here. These findings suggest that in utero exposure to flutamide may induce hypospermatogenesis through alterations of claudin-11 expression, potentially resulting in hemotesticular barrier (re)modeling impairment, although additional experiments are required to confirm this.
Finally, because claudin-11 is also expressed in the brain, we show in this study that claudin-11 expression is altered in the testis, but not in the brain, of male rats treated in utero with the antiandrogen flutamide, suggesting that transcription factors and coregulator elements in testis and brain might be different, thus leading to tissue specificity.
In summary, the present study shows the variations in the expression of claudin-11 in testes of prepubertal and adult rats exposed during gestation to the antiandrogen flutamide. Our results indicate that claudin-11 expression is altered in postnatal testes with fetal androgen disruption, and that in Sertoli cells, this gene and its product are under dual and opposite regulation, i.e. a stimulatory action exerted by androgens and an inhibitory effect originating from postmeiotic germ cells.
References
Kelce WR, Gray LE, Wilson EM 1998 Antiandrogens as environmental endocrine disruptors. Reprod Fertil Dev 10:105–111
Mylchreest E, Cattley RC, Foster PM 1998 Male reproductive tract malformations in rats following gestational and lactational exposure to di(n-butyl) phthalate: an antiandrogenic mechanism? Toxicol Sci 43:47–60
You L, Casanova M, Archibeque-Engle S, Sar M, Fan LQ, Heck HA 1998 Impaired male sexual development in perinatal Sprague-Dawley and Long-Evans hooded rats exposed in utero and lactationally to p,p'-DDE. Toxicol Sci 45:162–173
Gray Jr LE, Wolf C, Lambright C, Mann P, Price M, Cooper RL, Ostby J 1999 Administration of potentially antiandrogenic pesticides (procymidone, linuron, iprodione, chlozolinate, p,p'-DDE, and ketoconazole) and toxic substances (dibutyl- and diethylhexyl phthalate, PCB 169, and ethane dimethane sulphonate) during sexual differentiation produces diverse profiles of reproductive malformations in the male rat. Toxicol Ind Health 15:94–118
Gray LE, Ostby J, Furr J, Wolf CJ, Lambright C, Parks L, Veeramachaneni DN, Wilson V, Price M, Hotchkiss A, Orlando E, Guillette L 2001 Effects of environmental antiandrogens on reproductive development in experimental animals. Hum Reprod Update 7:248–264
Toppari J 2002 Environmental endocrine disrupters and disorders of sexual differentiation. Semin Reprod Med 20:305–312
McIntyre BS, Barlow NJ, Foster PM 2001 Androgen-mediated development in male rat offspring exposed to flutamide in utero: permanence and correlation of early postnatal changes in anogenital distance and nipple retention with malformations in androgen-dependent tissues. Toxicol Sci 62:236–249
Imperato-McGinley J, Sanchez RS, Spencer JR, Yee B, Vaughan ED 1992 Comparison of the effects of the 5-reductase inhibitor finasteride and the antiandrogen flutamide on prostate and genital differentiation: dose-response studies. Endocrinology 131:1149–1156
van der Schoot P 1992 Disturbed testicular descent in the rat after prenatal exposure to the antiandrogen flutamide. J Reprod Fertil 96:483–496
Kassim NM, McDonald SW, Reid O, Bennett NK, Gilmore DP, Payne AP 1997 The effects of pre- and postnatal exposure to the nonsteroidal antiandrogen flutamide on testis descent and morphology in the albino Swiss rat. J Anat 190:577–588
Miyata K, Yabushita S, Sukata T, Sano M, Yoshino H, Nakanishi T, Okuno Y, Matsuo M 2002 Effects of perinatal exposure to flutamide on sex hormones and androgen-dependent organs in F1 male rats. J Toxicol Sci 27:19–33
Omezzine A, Chater S, Mauduit C, Florin A, Tabone E, Chuzel F, Bars R, Benahmed M 2003 Long-term apoptotic cell death process with increased expression and activation of caspase-3 and -6 in adult rat germ cells exposed in utero to flutamide. Endocrinology 144:648–661
Bozec A, Chuzel F, Chater S, Paulin C, Bars R, Benahmed M, Mauduit C 2004 Mitochondrial-dependent pathway is chronically affected in testicular germ cell death in adult rat exposed in utero to anti-androgens. J Endocrinol 183:79–90
Bremner WJ, Millar MR, Sharpe RM, Saunders PT 1994 Immunohistochemical localization of androgen receptors in the rat testis: evidence for stage-dependent expression and regulation by androgens. Endocrinology 135:1227–1234
De Gendt K, Swinnen JV, Saunders PT, Schoonjans L, Dewerchin M, Devos A, Tan K, Atanassova N, Claessens F, Lecureuil C, Heyns W, Carmeliet P, Guillou F, Sharpe RM, Verhoeven G 2004 A Sertoli cell-selective knockout of the androgen receptor causes spermatogenic arrest in meiosis. Proc Natl Acad Sci USA 101:1327–1332
Morita K, Sasaki H, Fujimoto K, Furuse M, Tsukita S 1999 Claudin-11/OSP-based tight junctions of myelin sheaths in brain and Sertoli cells in testis. J Cell Biol 145:579–588
Bronstein JM, Popper P, Micevych PE, Farber DB 1996 Isolation and characterization of a novel oligodendrocyte-specific protein. Neurology 47:772–778
Bronstein JM, Micevych PE, Chen K 1997 Oligodendrocyte-specific protein (OSP) is a major component of CNS myelin. J Neurosci Res 50:713–720
Gow A, Southwood CM, Li JS, Pariali M, Riordan GP, Brodie SE, Danias J, Bronstein JM, Kachar B, Lazzarini RA 1999 CNS myelin and Sertoli cell tight junction strands are absent in Osp/claudin-11 null mice. Cell 99:649–659
Hellani A, Ji J, Mauduit C, Deschildre C, Tabone E, Benahmed M 2000 Developmental and hormonal regulation of the expression of oligodendrocyte-specific protein/claudin 11 in mouse testis. Endocrinology 141:3012–3019
Gye MC 2003 Changes in the expression of claudins and transepithelial electrical resistance of mouse Sertoli cells by Leydig cell coculture. Int J Androl 26:271–278
Dorrington JH, Roller NF, Fritz IB 1975 Effects of follicle-stimulating hormone on cultures of Sertoli cell preparations. Mol Cell Endocrinol 3:57–70
Boussouar F, Mauduit C, Tabone E, Pellerin L, Magistretti PJ, Benahmed M 2003 Developmental and hormonal regulation of the monocarboxylate transporter 2 (MCT2) expression in the mouse germ cells. Biol Reprod 69:1069–1078
Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159
Foley KP, Leonard MW, Engel JD 1993 Quantitation of RNA using the polymerase chain reaction. Trends Genet 9:380–385
Mauduit C, Siah A, Foch M, Chapet O, Clippe S, Gerard JP, Benahmed M 2001 Differential expression of growth factors in irradiated mouse testes. Int J Radiat Oncol Biol Phys 50:203–212
Ketola I, Rahman N, Toppari J, Bielinska M, Porter-Tinge SB, Tapanainen JS, Huhtaniemi IT, Wilson DB, Heikinheimo M 1999 Expression and regulation of transcription factors GATA-4 and GATA-6 in developing mouse testis. Endocrinology 140:1470–1480
Hopkinson CR, Dulisch B, Gauss G, Hilscher W, Hirschhauser C 1978 The effect of local testicular irradiation on testicular histology and plasma hormone levels in the male rat. Acta Endocrinol (Copenh) 87:413–423
Kangasniemi M, Kaipia A, Toppari J, Mali P, Huhtaniemi I, Parvinen M 1990 Cellular regulation of basal and FSH-stimulated cyclic AMP production in irradiated rat testes. Anat Rec 227:32–36
Heiskala M, Peterson PA, Yang Y 2001 The roles of claudin superfamily proteins in paracellular transport. Traffic 2:93–98
Sharpe RM 1994 Regulation of spermatogenesis. In: Knobil E, Neill JD, Ewing LL, Greenwald GS, Markert CL, Pfaff DW, eds. The physiology of reproduction. New York: Raven Press; 1363–1434
Ivell R, Spiess A-N 2002 Analysing differential gene expression in the testis. In: Rommerts FF, Teerds KJ, eds. Testicular tangrams. Berlin: Springer; 99–120
Comhaire FH, Vermeulen A 1976 Testosterone concentration in the fluids of seminiferous tubules, the interstitium and the rete testis of the rat. J Endocrinol 70:229–235
De SK, Chen HL, Pace JL, Hunt JS, Terranova PF, Enders GC 1993 Expression of tumor necrosis factor- in mouse spermatogenic cells. Endocrinology 133:389–396
Caussanel V, Tabone E, Hendrick JC, Dacheux F, Benahmed M 1997 Cellular distribution of transforming growth factor ?s 1, 2, and 3 and their types I and II receptors during postnatal development and spermatogenesis in the boar testis. Biol Reprod 56:357–367
Olaso R, Pairault C, Saez JM, Habert R 1999 Transforming growth factor ?3 in the fetal and neonatal rat testis: immunolocalization and effect on fetal Leydig cell function. Histochem Cell Biol 112:247–254
Mauduit C, Besset V, Caussanel V, Benahmed M 1996 Tumor necrosis factor receptor p55 is under hormonal (follicle-stimulating hormone) control in testicular Sertoli cells. Biochem Biophys Res Commun 224:631–637
Olaso R, Pairault C, Habert R 1998 Expression of type I and II receptors for transforming growth factor ? in the adult rat testis. Histochem Cell Biol 110:613–618
Lui WY, Lee WM, Cheng CY 2001 Transforming growth factor-?3 perturbs the inter-Sertoli tight junction permeability barrier in vitro possibly mediated via its effects on occludin, zonula occludens-1, and claudin-11. Endocrinology 142:1865–1877
Lui WY, Wong CH, Mruk DD, Cheng CY 2003 TGF-?3 regulates the blood-testis barrier dynamics via the p38 mitogen activated protein (MAP) kinase pathway: an in vivo study. Endocrinology 144:1139–1142
Lui WY, Lee WM, Cheng CY 2003 Transforming growth factor ?3 regulates the dynamics of Sertoli cell tight junctions via the p38 mitogen-activated protein kinase pathway. Biol Reprod 68:1597–1612
Goddard I, Florin A, Mauduit C, Tabone E, Contard P, Bars R, Chuzel F, Benahmed M 2003 Alteration of lactate production and transport in the adult rat testis exposed in utero to flutamide. Mol Cell Endocrinol 206:137–146
McLachlan JA 2001 Environmental signaling: what embryos and evolution teach us about endocrine disrupting chemicals. Endocr Rev 22:319–341
You L, Sar M 1998 Androgen receptor expression in the testes and epididymides of prenatal and postnatal Sprague-Dawley rats. Endocrine 9:253–261
Chung NP, Cheng CY 2001 Is cadmium chloride-induced inter-Sertoli tight junction permeability barrier disruption a suitable in vitro model to study the events of junction disassembly during spermatogenesis in the rat testis? Endocrinology 142:1878–1888
Tiwari-Woodruff SK, Buznikov AG, Vu TQ, Micevych PE, Chen K, Kornblum HI, Bronstein JM 2001 OSP/claudin-11 forms a complex with a novel member of the tetraspanin super family and ?1 integrin and regulates proliferation and migration of oligodendrocytes. J Cell Biol 153:295–305(Anne Florin, Magali Maire)
Address all correspondence and requests for reprints to: Dr. Mohamed Benahmed, Institut National de la Santé et de la Recherche Médicale, Unité 407, Faculté de Médecine Lyon-Sud, BP 12, 69921 Oullins Cedex, France. E-mail: benahmed@grisn.univ-lyon1.fr.
Abstract
In the present study we investigated whether fetal exposure to flutamide affected messenger and protein levels of claudin-11, a key Sertoli cell factor in the establishment of the hemotesticular barrier, at the time of two key events of postnatal testis development: 1) before puberty (postnatal d 14) during the establishment of the hemotesticular barrier, and 2) at the adult age (postnatal d 90) at the time of full spermatogenesis. The data obtained show that claudin-11 expression was inhibited in prepubertal rat testes exposed in utero to 2 and 10 mg/kg·d flutamide. However, in adult testes, the inhibition was observed only with 2, and not with 10, mg/kg·d of the antiandrogen. It is shown here that these differences between prepubertal and adult testes could be related to dual and opposed regulation of claudin-11 expression resulting from positive control by androgens and an inhibitory effect of postmeiotic germ cells. Indeed, testosterone is shown to stimulate claudin-11 expression in cultured Sertoli cells in a dose- and time-dependent manner (maximum effect with 0.06 μM after 72 h of treatment). In contrast, postmeiotic germ cells potentially exert a negative effect on claudin-11 expression, because adult rat testes depleted in spermatids (after local irradiation) displayed increased claudin-11 expression, whereas in a model of cocultured Sertoli and germ cells, spermatids, but not spermatocytes, inhibited claudin-11 expression. The apparent absence of claudin-11 expression changes in adult rat testes exposed to 10 mg/kg·d flutamide therefore could result from the antagonistic effects of 1) the inhibitory action of the antiandrogen and 2) the stimulatory effect of the apoptotic germ cells on claudin-11 expression. Together, due to the key role of claudin-11 in the hemotesticular barrier, the present findings suggest that such regulatory mechanisms may potentially affect this barrier (re)modeling during spermatogenesis.
Introduction
SEVERAL RESEARCHERS HAVE shown that in experimental rodent models an endocrine (androgen) disruption during gestational life alters androgen-dependent reproductive development and function (1, 2, 3, 4 ; for reviews, see Refs.5 and 6). One of the most frequently used experimental models of antiandrogens is the potent nonsteroidal antiandrogen flutamide, gestational exposure to which causes, in particular, hypospadias, undescended testes, and impaired spermatogenesis in male offspring (7, 8, 9, 10, 11). Although the molecular and cellular mechanisms underlying hypospermatogenesis in the adult rat exposed in utero to flutamide remain poorly understood, we previously reported that such an alteration could be related to a long-term apoptotic cell death process in meiotic and postmeiotic germ cells, with a significant increase in the expression and activation of effector caspase-3 and -6 (12), and an alteration in the balance between pro- and antiapoptotic Bcl-2 related molecules in favor of proapoptotic proteins (13). Although the death process that follows gestational exposure to flutamide does not affect adult somatic cells, their expression of the androgen receptor (AR) makes them a direct target of testosterone (and thus of flutamide) (14, 15), and these findings suggest that in seminiferous tubules, Sertoli cells (and peritubular myoid cells?) may produce androgen-regulated factors involved in the survival of germ cells.
Claudin-11, also known as oligodendrocyte-specific protein (OSP), belongs to the claudin family of intrinsic transmembrane proteins (16). First described in the myelin sheaths of the central nervous system (17), studies have shown that it is specifically expressed in oligodendrocytes in the brain and Sertoli cells in the testis, where it is responsible for the formation of specific parallel tight junction strands (16, 18, 19). In the mouse testis, claudin-11 has been shown to be hormonally regulated (FSH and testosterone) (20, 21) and is a key factor in establishment of the hemotesticular barrier. Indeed, testicular claudin-11 expression is at its maximum between postnatal d 6 and 16, during the formation of the hemotesticular barrier (20), and mice invalidated for this gene are infertile, with the loss of claudin-11 expression causing a disruption of this barrier, leading to an arrest in spermatogenesis (19).
In the present study we investigated whether fetal exposure to flutamide affected claudin-11 messenger and protein levels in rat Sertoli cells at the time of two key events in postnatal testis development: 1) before puberty (postnatal d 14) during the establishment of the hemotesticular barrier, and 2) in adulthood (postnatal d 90) at the time of full spermatogenesis when all germ cell types are present. Claudin-11 mRNA and protein levels were also studied in the brain of adult male rats exposed in utero to the antiandrogen.
Materials and Methods
Materials
Flutamide was purchased from Sigma-Aldrich Corp. (Meylan, France), dissolved in an aqueous solution of methylcellulose 400 (Fluka, Mulhouse, France) at 0.5% (wt/vol), and stored at 5 C (±3 C) for a maximum of 1 wk. Sigma-Aldrich Corp. was also the supplier for BSA, hexanucleotide primers, insulin, transferrin, gentamicin, HEPES, sodium bicarbonate, trypsin, and deoxyribonuclease I. Collagenase-dispase was obtained from Roche (Mannheim, Germany). DMEM-Ham’s F-12 medium (DMEM/F12), deoxy-NTPs (dNTPs), Moloney murine leukemia virus reverse transcriptase kit, and TRIzol were obtained from Invitrogen Life Technologies, Inc. (Eragny, France). Taq polymerase was obtained from Promega Corp. (Lyon, France). Amersham Biosciences (Little Chalfont, UK) was the supplier of [-33P]dATP. Oligonucleotides were purchased from Invitrogen Life Technologies, Inc. (Groningen, The Netherlands).
Animals
Virgin female Sprague Dawley rats (Charles River Laboratories, Saint Aubin les Elbeuf, France) were individually housed under controlled conditions of lighting (12 h of light, 12 h of darkness), temperature (22 ± 2 C), humidity (55 ± 15%), and ventilation (15 air changes/h) and were given free access to water and feed (certified rodent pellet diet A04C, UAR, Villemoisson-sur-Orge, France). Females were mated on a one to one basis with males of the same strain, purchased from the same supplier. Day 0 of gestation (GD0) was the day a vaginal sperm plug was observed. Before mating and during gestation, dams were housed in suspended stainless steel wire mesh cages. Shortly before parturition and during lactation, dams were housed in Makrolon cages (Charles River) with soft wood bedding.
Pregnant rats were administered daily the vehicle (methylcellulose) alone or flutamide by gavage from GD10 to the day before delivery (GD21 or GD22). Animals were administered flutamide at doses of 0, 0.4, 2, and 10 mg/kg body weight·d (adjusted daily for body weight). Dams were weighed daily from GD10 to the day of delivery. At birth, each pup was sexed, weighed, and identified. The male offspring received no flutamide treatment and were raised until study termination (postnatal d 14 or 90), when rats were euthanized by CO2 asphyxiation. The position of each testis was carefully noted. Moreover, each testis was weighed before being fixed or frozen. Only bilateral descended testes were studied in the present report. All studies using animals were conducted in accordance with current regulation and standards approved by Institut National de la Santé et de la Recherche Médicale (French Institute for Health and Medical Research) animal care committee.
Irradiation
Adult Sprague Dawley (90 d old) rats purchased from Iffa-Credo (L’Arbresle, France) were housed in controlled conditions of lighting (12 h of light, 12 h of darkness), temperature (22 ± 2 C), humidity (55 ± 15%), and ventilation (15 air changes/h) and were given free access to water and feed (certified rodent pellet diet, AO4C, UAR). Before the experimental procedure, an acclimation period of 1 wk was allowed. All rats were anesthetized by ip injection of sodium phenobarbital (45 mg/kg; Sanofi Santé Animale, Libourne, France) 20 min before irradiation. The scrotum of each rat was irradiated by x-ray at a dose of 9 Gy. A 120-kv x-ray beam with 2-mm Al filtration was generated by a Stabilipan (Siemens AG, Erlangen, Germany) orthovoltage machine. A 4-cm diameter lead collimator was used to selectively irradiate both testes while protecting the other organs at risk in the rats. The distance between source and skin was 40 cm. The dose was prescribed and specified at 2.5 mm under the skin of the scrotum, at the center of the testes. Control animals received anesthesia and sham irradiation. Irradiated and sham-irradiated animals were euthanized by CO2 inhalation 10, 26, or 45 d after the experiments. One group of control animals was killed 10 d after sham irradiation, and the other group was killed 45 d after sham irradiation. Each control and treatment group consisted of six animals. For each rat, the left testis was removed and fixed for 24 h in Bouin’s fixative, and the right testis was removed for RNA and protein extractions.
Isolation and culture of Sertoli cells
Primary Sertoli cells were isolated from 20-d-old Sprague Dawley rats, as described by Dorrington et al. (22). Briefly, decapsulated testes were submitted to collagenase dissociation (0.5 mg/ml, 5 min at 30 C) in DMEM/F12 (1:1, vol/vol) medium (1.2 mg/ml sodium bicarbonate, 15 mM HEPES, and 20 μg/ml gentamicin) containing deoxyribonuclease I (0.05 mg/ml). At the end of enzymatic dissociation, testicular cells were washed three times by submitting them to gravity sedimentation (3–5 min), and supernatants were removed. The pellets containing the sedimented tubules were also dissociated with a collagenase-dispase treatment, as described above, until small clumps resulted. Cells were then submitted to gravity sedimentation (10–15 min), supernatants were removed, and the sedimented clumps of Sertoli cells were also washed by centrifugation (200 x g, 10 min). Sertoli cell pellets were resuspended in DMEM/F12 medium (supplemented with 5 μg/ml transferrin, 2 μg/ml insulin, and 10 μg/ml vitamin E), and cells were plated on Falcon petri dishes (BD Biosciences, Franklin Lakes, NJ; 100-mm diameter, 7 x 106 cells/dish) and cultured at 32 C in a humidified atmosphere of 5% CO2-95% air. This procedure led to a Sertoli cell population free from Leydig cells or mature germ cells, but which contained approximately 15–20% peritubular myoid cells (data not shown). For the coculture studies, 2 x 106 Sertoli cells were seeded in 6-cm diameter petri dishes. Testicular germ cells (spermatocytes and spermatids) were isolated from adult rats (90 d old) as described by Boussouar et al. (23). Three groups of cultured cells were generated: Sertoli cells cultured alone without germ cells, Sertoli cells cultured with spermatocytes (2 x 106 cells), and Sertoli cells cultured with spermatids (2 x 106 cells). The cells were collected after 72 h of culture. Experiments were repeated at least three times with independent cell preparations.
Total RNA extraction
Total RNAs were extracted using TRIzol, a monophasic solution of phenol and guanidine isothiocyanate, following an improvement of the single-step RNA isolation method developed by Chomczynski and Sacchi (24). The final amount of RNA was estimated by spectrophotometry at 260 nm.
Coamplification RT-PCR with an endogenous control
Coamplification RT-PCR (25) was performed to determine the mRNA levels of claudin-11, AR, and FSH receptor (FSH-R). Briefly, approximately 2 μg total RNA were reverse-transcribed into cDNAs (cDNAs) for 1 h at 37 C using Moloney murine leukemia virus reverse transcriptase (10 U/μl) in 1x first strand buffer, random hexanucleotides as primers (5 μM), dNTPs (250 μM), and dithiothreitol (10 μM). PCRs were then performed on 2 μl RT product, using the appropriate sense- and antisense-specific primers, Taq polymerase (0.5 U), 1x PCR buffer, 2.5 mM MgCl2, dNTPs, and 0.75 μCi [-33P]dATP. Claudin-11 (0.055 or 1 μM) was coamplified with GATA-6 (1 μM) or with the ?-actin (0.1 μM) primer pair, GATA-6 (1 μM) was coamplified with ?-actin (0.04 μM), FSH-R (1 μM) was coamplified with GATA-6 (1 μM), and AR (0.5 μM) was coamplified with hypoxanthine phosphoribosyltransferase (HPRT; 0.5 μM). Coamplification with ?-actin or HPRT was performed to check that equal amounts of cDNAs were amplified in each reaction tube, and coamplification with the GATA-6 gene was performed to compare claudin-11 and FSH-R mRNA levels with those of another Sertoli cell-specific gene, so as to detect eventual changes in the proportion of this cell type in the testes from flutamide-treated or irradiated animals. PCR mixes were submitted to an initial denaturing step at 95 C, followed by X cycles consisting of 30-sec denaturation at 95 C, 30-sec annealing at melting temperature (Tm), and 30-sec extension at 72 C, and the reaction ended with a final extension step at 72 C: claudin-11: X, 19 cycles; Tm, 55 C; GATA-6: X, 19 cycles; Tm, 55 C; AR: X, 25 cycles; Tm, 55 C; and FSH-R: X, 24 cycles; Tm, 65 C. Primers were as follows: for claudin-11: upstream primer, 5'-GATTGGCATCATCGTCACAACG-3'; downstream primer, 5'-AGCCAGCAGAATAAGGAGCAAC-3' (339 bp); for GATA-6: upstream primer, 5'-GTGCCAACCCTGAGAACAGT-3'; downstream primer, 5'-TGGACTGCTGGACAAAATCA-3' (198 bp); for ?-actin: upstream primer, 5'-TTGCTGATCCACATCTGCTG-3'; downstream primer, 5'-GACAGGATGCAGAAGGAGAT-3' (146 bp); for AR: upstream primer, 5'-ATTGTCCATCGTGTCGTCTCC G-3'; downstream primer, 5'-GAGTTGACATTAGTGAAGGACC-3' (447 bp); for HPRT: upstream primer, 5'-CCTGCTGGATTACATTAA AGC-3'; downstream primer, 5'-GTCAAGGGCATATCCAACAAC-3' (354 bp); and for FSH-R: upstream primer, 5'-CTCATCAAGCGACACCAAGA-3'; downstream primer, 5'-ACCTTGAGGGAGGCAGAAAT-3' (108 bp). PCR products were resolved on 8% polyacrylamide gels, which were exposed to a Storage Phosphor Screen (Packard, Meriden, CT), and the signals were analyzed using Cyclone OptiQuant software (Packard). Results from at least three separate experiments were used for statistical analysis.
PCR analyses were carried out from the logarithmic phase of amplification, and different cycle numbers were tested for each primer pair to determine the minimum number of cycles necessary to detect the PCR product. Primers were designed inside separate exons to avoid any bias caused by residual genomic contamination. For all primers, no amplification was observed when PCR was performed on RNA preparations. Finally, PCR-amplified products were checked by direct sequencing using an automated sequencer (ABI PRISM 310, Applied Biosystems, Foster City, CA).
Western blotting analysis
Proteins (40 μg) from whole adult testis or brain were separated on 10% (AR) or 12% (claudin-11) sodium dodecyl sulfate-polyacrylamide gels and electrophoretically transferred to nitrocellulose membranes using 25 mM Tris and 185 mM glycine for claudin-11 or 15 mM Tris and 120 mM glycine for AR, pH 8.3, containing 20% methanol. Transfer was performed at a constant voltage of 100 V for 1 h. After transfer, membranes were blocked in Tris-buffered saline containing 0.05% Tween (TBS-T) containing 3% BSA (claudin-11) or 5% milk (AR), and incubated with the primary antibody [1:2000 for anticlaudin-11 (Covalab, Lyon, France) or 1:200 for anti-AR (sc-816, Santa Cruz Biotechnology, Inc., Santa Cruz, CA)] in TBS-T overnight at 4 C. Membranes were then incubated with horseradish peroxidase-labeled goat antirabbit (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA; 1:2000 in TBS-T) for 1 h at room temperature, and bound antibodies were subsequently detected using a chemiluminescence detection kit (Covalab) and Biomax MR films (Eastman Kodak Co., Rochester, NY). Protein concentrations were determined by the Bradford assay. Reprobing the blot with a rab-bit immunoglobulin G antiactin (20-33, Sigma-Aldrich Corp., L’Isle d’Abeau, France; concentration, 1:500 in TBS-T) checked protein loading.
Terminal deoxynucleotidyltransferase-mediated deoxyuridine triphosphate nick end labeling (TUNEL)
Testes were immediately fixed for 24 h in Bouin’s fluid, stepwise dehydrated in graded ethanol baths, and embedded in paraffin. Paraffin sections of Bouin-fixed testes were sectioned at 5 μm. The sections were mounted on positively charged glass slides (SuperFrost Plus, Menzel-Glaser, Braunschweig, Germany), deparaffinized, hydrated, treated for 20 min at 93–98 C in citric buffer (0.01 M, pH 6), rinsed in osmosed water (twice, 5 min each time), and washed (twice, 5 min each time) in TBS, pH 7.8. TUNEL reaction was performed as previously described (26), and slides were counterstained for 2 min with Mayer’s hematoxylin. In each rat testis, at least 100 random seminiferous tubules were counted. The results are expressed as the number of TUNEL-positive cells per 100 seminiferous tubules.
Hormone assay
Blood samples, taken from 14- or 90-d-old rats selected in each control or treated group across litters, were taken by retroorbital puncture after isoflurane anesthesia. Plasma samples were kept frozen until testosterone or FSH was analyzed using a specific RIA. Specific kits were supplied by Amersham Biosciences (Orsay, France) for FSH and by Im-munotech (Marseille, France) for testosterone. Packard provided the RIASTAR program (Canberra Co., Meriden, CT).
Data analysis
Data are expressed as the mean ± SEM. At least four different male offspring (n = 4–11/condition) from different litters were used. For statistical analysis, one-way ANOVA was performed to determine differences among all groups (P < 0.05), and the Bonferroni/Dunn posttest was performed to determine the significance of the differences between pairs of groups. P < 0.05 was considered significant. The statistical tests were performed on StatView software (version 5.0, SAS Institute, Inc., Cary, NC) on a Macintosh computer (Apple Computer, Cupertino, CA).
Results
Effects of exposure to flutamide during fetal life on claudin-11 mRNA levels in prepubertal and adult rat testes
In these experiments we analyzed the expression of claudin-11 in prepubertal (14 d old) and adult (90 d old) rat testes exposed in utero to 0.4, 2, and 10 mg/kg·d flutamide. Data presented in Fig. 1A show that claudin-11 mRNA levels were significantly reduced in testes from 14-d-old rats treated in utero with 2 (35% decrease; P < 0.0001) and 10 (50% decrease; P = 0.0001) mg/kg·d flutamide. In parallel, a decrease in claudin-11 protein was observed in a significant (30% decrease; P = 0.008) manner at 10 mg/kg·d flutamide (Fig. 1B).
FIG. 1. Effects of in utero exposure to flutamide on claudin-11 mRNA and protein levels in rat testis. Claudin-11 mRNA levels were determined by coamplification RT-PCR in prepubertal (A) and adult (C) rat testes exposed in utero to vehicle alone (0) or 0.4, 2, or 10 mg/kg·d flutamide. Claudin-11 protein levels were measured by Western blotting in prepubertal (B) and adult (D) rat testes exposed in utero to vehicle alone or to the three doses of the antiandrogen. The data show representative autoradiograms (upper panel), and the histograms (lower panel) are expressed as the mean ± SEM determined from at least six (coamplification RT-PCR) or four (Western blotting) animals from different litters. Results are represented as a percentage of the ratio (claudin-11/GATA-6 mRNA or claudin-11/?-actin protein) detected in the control (vehicle treated) animals.
Figure 1C shows that in the adult (90 d old) testes treated in utero with 2 mg/kg·d flutamide, claudin-11 mRNA levels were significantly reduced by 25% (P < 0.0001), whereas at the dose of 10 mg/kg·d flutamide, such a decrease in claudin-11 mRNA levels was no longer observed in adult testes compared with prepubertal (14 d old) testes.
Data obtained by Western blotting analysis showed similar results for protein levels (Fig. 1D). Indeed, in adult rat testes exposed in utero to 2 mg/kg·d flutamide, a 50% decrease in claudin-11 protein levels was found (P < 0.005), and again, in the 10 mg/kg·d-treated rats, this decrease in testicular claudin-11 levels was not observed.
To determine the specificity of claudin-11 expression changes in Sertoli cells, we evaluated in parallel AR mRNA and protein levels, FSH-R mRNA levels, and the expression of the gene coding for GATA-6, a transcription factor member of the GATA-binding protein family that is specifically expressed in Sertoli cells (27). AR mRNA (data not shown) and protein levels in prepubertal (Fig. 2A) and adult (Fig. 2B) rat testes were not affected after in utero exposure to the various doses of flutamide. Similarly, FSH-R mRNA levels in prepubertal (Fig. 2C) and adult (Fig. 2D) rat testes were not modified by fetal treatment with the antiandrogen, as were GATA-6 mRNA levels in prepubertal (data not shown) or adult (Fig. 2E) rat testes.
FIG. 2. Effects of in utero exposure to flutamide on AR protein levels, FSH-R, or GATA-6 mRNA levels in rat testis. AR protein levels were determined in prepubertal (A) and adult (B) rat testes exposed in utero to vehicle alone (0) or 0.4, 2, or 10 mg/kg·d flutamide. The levels of mRNA were determined for FSH-R in prepubertal (C) and adult (D) testes and for GATA-6 in adult (E) rat testes exposed in utero to vehicle alone (0) or to 0.4, 2, or 10 mg/kg·d flutamide. The data show representative autoradiograms (upper panel), and the histograms (lower panel) are expressed as the mean ± SEM determined from at least six (coamplification RT-PCR) or four (Western blotting) animals from different litters. Results are represented as a percentage of the ratio (AR/?-actin protein, FSH-R/GATA-6 mRNA, and GATA-6/?-actin mRNA) detected in control (vehicle-treated) animals.
Because Sertoli cell activity is predominantly under androgen and FSH control, the changes in claudin-11 expression could be related to modifications in plasma testosterone and/or FSH levels. However, the data in Table 1 show that plasma testosterone levels were not significantly affected in either prepubertal (14 d old) or adult (90 d old) rats exposed in utero to flutamide. Similarly, FSH levels in adult rats exposed in utero to flutamide were not altered (Table 2). Taken together, the data in Fig. 2 and Tables 1 and 2 indicate that the alterations in claudin-11 expression in prepubertal and adult testes exposed in utero to flutamide were not related to altered androgen and/or FSH plasma levels and/or changes in AR/FSH-R expression.
TABLE 1. Plasma testosterone levels in prepubertal and adult animals exposed in utero to flutamide
TABLE 2. Plasma FSH levels in adult animals exposed in utero to flutamide
Because the inhibition of claudin-11 expression observed at a dose of 10 mg/kg·d in prepubertal testis was not evident in adult testis, it was hypothesized that testicular claudin-11 expression could be under two (opposite) regulations: 1) a positive regulation exerted by androgens, which could explain the inhibitory effect of flutamide specifically observed at 2 mg/kg·d; and 2) an additive negative effect exerted by postmeiotic germ cells, present in the adult, but not in the prepubertal, testis. In the following experiments, we therefore tested the regulatory effects of testosterone and germ cells on claudin-11 expression in the testis.
Claudin-11 expression is under positive control of testosterone in testis
To study the regulation by androgens of claudin-11 expression, messenger and protein levels were evaluated in a model of cultured rat Sertoli cells. Sertoli cells were treated with various concentrations of testosterone (7 nM to 3.5 μM; 2–1000 ng/ml) for various periods (4–96 h). The data in Fig. 3A show that claudin-11 mRNA levels were up-regulated in a testosterone dose-dependent manner. The maximal effect was obtained with 583 nM (170 ng/ml) testosterone (P < 0.001). Treatment with 875 nM (250 ng/ml) showed that the regulation of claudin-11 by testosterone was time dependent (Fig. 3B), with an increase in claudin-11 mRNA levels after 48 h of testosterone treatment and a maximal effect after 72 h (P < 0.0001). In control experiments we ensured that basal claudin-11 mRNA levels were not affected throughout the culture period (from 4–96 h, Fig. 3B). Similar results were obtained at the protein level (Fig. 3C); claudin-11 protein levels were increased by more than 2-fold after 72 h of treatment with 875 nM (250 ng/ml) testosterone (P < 0.001).
FIG. 3. Claudin-11 mRNA and protein levels in cultured Sertoli cells treated with testosterone. Cultured Sertoli cells were treated with increasing (0–1750 nM) doses of testosterone for 72 h (A) or without () or with () 875 nM (250 ng/ml) testosterone for different treatment times (4–96 h; B) before determining claudin-11 mRNA levels. Sertoli cells were cultured in the absence (0) or the presence of 875 nM testosterone for 72 h before evaluating claudin-11 protein levels (C). The data show representative autoradiograms (upper panel) and histograms (lower panel) and are expressed as the mean ± SEM from at least three different experiments. Results are represented as a percentage of the ratio of claudin-11/?-actin mRNA detected in the controls (untreated cells).
Claudin-11 expression is under negative control of postmeiotic germ cells in testis
To answer the question of whether claudin-11 expression could be regulated by germ cells, we evaluated claudin-11 expression in two models: 1) a model of locally irradiated (9 Gy) adult (90 d old) rat testis, and 2) a model of Sertoli cell-germ cell cocultures. The first model enabled sequential removal of premeiotic and postmeiotic germ cells (28, 29), thus allowing study of the possible regulation of claudin-11 expression in Sertoli cells by the different types of germ cells. Consistent with the literature (29), in control experiments, testes from rats killed 10 d after irradiation were practically devoid of spermatogonia; 26 d after irradiation, very few premeiotic germ cells were observed; and in testes from rats killed 45 d after irradiation, postmeiotic germ cells were very low in number, but spermatogonia were observed (data not shown).
Figure 4A shows an increase in testicular GATA-6 mRNA expression on d 26 (1.3-fold; P < 0.0001) and d 45 (1.5-fold; P < 0.0001) after irradiation. This increase in GATA-6 expression was observed in the testes depleted in the two most represented germ cell types (pachytene spermatocytes and spermatids). This would suggest that the increase in GATA-6 expression is probably related to the subsequent relative concentration of Sertoli cells. For this reason, we expressed claudin-11 mRNA levels as claudin-11/GATA-6 ratios in Fig. 4B. This figure shows that claudin-11/GATA-6 mRNA levels were similar to control levels in testes that were practically devoid of premeiotic germ cells (10 and 26 d after local irradiation), whereas claudin-11/GATA-6 mRNA levels were increased by 2.5-fold (P < 0.0001) in testes lacking postmeiotic germ cells (45 d after irradiation). For claudin-11 protein levels, Fig. 4C shows that a 2-fold increase (P = 0.0031) was found in the testes from rats killed 45 d after irradiation, compared with the sham-irradiated testes. Furthermore, in the second model of Sertoli cell-germ cell cocultures, a significant decrease (40%; P < 0.007) in claudin-11 mRNA levels was observed when Sertoli cells were cocultured with spermatids. In contrast, no changes in claudin-11 levels were evident in Sertoli cell-spermatocyte cocultures (Fig. 4D). Taken together, these data clearly suggest that postmeiotic germ cells exert an inhibitory effect on claudin-11 expression in Sertoli cells.
FIG. 4. Claudin-11 expression in adult rat testes after sequential removal of the different subtypes of germ cells and in Sertoli cell-germ cell (spermatocytes or spermatids) coculture. GATA-6 mRNA (A) and claudin-11 mRNA (B) and protein (C) levels were determined by coamplification RT-PCR and Western blotting in locally irradiated (9 Gy) adult rat testes, allowing specific and sequential removal of pre- and postmeiotic germ cells. Animals were killed 10, 26, or 45 d after local irradiation. Claudin-11 mRNA levels (D) were determined in Sertoli cell primary cultures or in Sertoli cells cocultured with either spermatocytes or spermatids. The data show representative autoradiograms (upper panel), and histograms (lower panel) are expressed as the mean ± SEM determined from at least six (coamplification RT-PCR) or four (Western blotting) animals from different litters. Results are represented as a percentage of the ratio (GATA-6/?-actin mRNA, claudin-11/GATA-6 mRNA, claudin-11/?-actin protein) detected in the control (vehicle-treated) animals.
Effects of exposure to flutamide during fetal life on claudin-11 mRNA and protein levels in adult rat brain
In contrast to that in testis, claudin-11 expression was not altered in adult rat brain after in utero exposure to the antiandrogen for all doses studied. Indeed, mRNA and protein levels in flutamide- treated animals were similar to those in control animals (data not shown).
Discussion
OSP/claudin-11 was first described in the myelin sheaths of the central nervous system by Bronstein et al. (17), where it represents 7% of total myelin proteins, making it a major component of central nervous system myelin (18). OSP cDNA was first identified as being expressed in oligodendrocytes by subtraction cloning and has been localized on human chromosome 3 (3q26.2-q26.3) (17). OSP was later also found to be specifically expressed in Sertoli cells of the testis, but not in the ovary (16, 20). Morita et al. (16) found that OSP shared sequence homology with the claudin superfamily of tight junction proteins, and like other claudins, this protein was found to induce the formation of tight junction strands when transfected in fibroblasts; it was therefore renamed claudin-11 (for a review on claudins, see Ref.30). The study by Gow et al. (19) of OSP/claudin-11-null mice confirmed the essential roles of this claudin in the establishment of specific tight junction strands in the brain and testis by showing both neurological and reproductive defects (hypospermatogenesis) in these mice.
In the present study we primarily investigated whether gestational exposure to flutamide affected claudin-11 mRNA and protein levels in rat Sertoli cells at the time of two key events in postnatal testis development: 1) before puberty (postnatal d 14), during establishment of the hemotesticular barrier, which starts on postnatal d 15–18 in the rat (31); and 2) in adulthood (postnatal d 90), at the time of full spermatogenesis when all germ cell subtypes are present. To study gene expression changes in this model, the doses of the antiandrogen were chosen to avoid or minimize important germ cell loss that may confound interpretation of the effect of flutamide on testicular gene expression (7). Indeed, alterations in testicular cellularity (i.e. the ratio of somatic cells to germ cells) may confound interpretation of the effects of the antiandrogen on gene expression in the various testicular cell types (for review, see Ref.32), specifically here on claudin-11 expression in the Sertoli cell population.
To ensure that the variations in claudin-11 levels were specific and linked to exposure to the antiandrogen, we studied the expression of another gene specifically expressed in Sertoli cells, the GATA-6 gene that codes for a member of the GATA-binding protein family and has been reported not to be hormonally regulated in the testis (27). Our results show that GATA-6 expression was not altered in 0.4, 2, and 10-mg/kg·d flutamide-treated testes in either prepubertal (14 d old) or adult (90 d old) animals, confirming the absence of hormonal dependence of this gene.
The data reported here show that claudin-11 mRNA levels in prepubertal (14 d old) and adult (90 d old) rat testes showed regulation similarities, because neither were affected by exposure to 0.4 mg/kg·d flutamide, but both declined with a dose of 2 mg/kg·d of the antiandrogen. However, at the dose of 10 mg/kg·d, a major difference in claudin-11 mRNA modulation was found depending on the age of the testis. Indeed, the inhibition of claudin-11 expression observed at the dose of 10 mg/kg·d in prepubertal testis was no longer evident in adult testis. We hypothesized that the effect of flutamide at 10 mg/kg·d in adult rat testes could be due either to a U-curve response to the antiandrogen, with an absence of response to flutamide at the dose of 10 mg/kg·d, or to an additive event occurring in the adult testis (the presence of postmeiotic germ cells) that could enhance claudin-11 levels and thus mask the inhibitory effect of the antiandrogen on testicular claudin-11 expression that was detected in prepubertal testes exposed to 10 mg/kg·d flutamide. Based on these observations, it was therefore hypothesized that testicular claudin-11 expression could be under two (opposite) regulations: 1) a positive regulation exerted by androgens, which could explain the inhibitory effect of flutamide specifically observed at 2 mg/kg·d; and 2) an additive negative effect exerted by the meiotic and/or postmeiotic germ cells present in adult, but not prepubertal, testis.
Using a model of cultured Sertoli cells, it was shown that, as in the mouse (21), claudin-11 expression is positively regulated by androgens in the rat testis, because claudin-11 mRNA and protein levels were up-regulated in a time- and dose-dependant manner in cultured rat Sertoli cells treated with various doses of testosterone at different times. The maximal stimulatory effect on mRNA levels was obtained with 583 nM (170 ng/ml) testosterone after 72 h of treatment. Although high, this concentration of testosterone is within the range of those reported for interstitial fluid collected from adult untreated rats (31, 33). It remains to be determined, however, whether testosterone affects claudin-11 gene transcriptional activity, mRNA stability, or both or claudin-11 protein stability and/or turnover and whether the stimulatory effect on claudin-11 is direct, because there is a relatively long period of latency between the time of treatment and the stimulatory effect observed on claudin-11 expression. This period of latency could also be due to participation of peritubular myoid cells in the androgen-dependent control of claudin-11 expression, because the Sertoli cell cultures used contained peritubular myoid cells.
In this study we also report that claudin-11 expression could be negatively regulated by factors produced by postmeiotic germs cells. Indeed, using a model of locally irradiated (9 Gy) rat testis that allows sequential removal of the different populations of germ cells, we found an increase in claudin-11 expression at both mRNA and protein levels in testes that are devoid of postmeiotic germ cells, but not in testes lacking spermatogonia or spermatocytes. These results were also supported by the use of an in vitro model of Sertoli cells cocultured with germ cells (spermatocytes or spermatids), in which claudin-11 mRNA levels were inhibited exclusively in the presence of spermatids. These observations clearly support an inhibitory effect exerted by postmeiotic germ cells on claudin-11 expression in Sertoli cells. The factors mediating the inhibitory effect of germ cells on claudin-11 expression are not known at present, but good candidates could be cytokines such as TNF and TGF?3. Indeed, these factors are produced by germ cells (34, 35, 36) and have receptors localized on Sertoli cells (35, 37, 38), and previous results published by our laboratory (20) and by Lui et al. (39, 40, 41) have shown that they are potent inhibitors of claudin-11 expression.
This dual and opposed regulation by androgens and postmeiotic germ cells of the expression of claudin-11 could therefore explain the apparent lack of effect of fetal treatment with the antiandrogen in testes from adult rats treated with 10 mg/kg·d, at which dose maximal (postmeiotic) germ cell death was observed (Table 3) (12, 13). Indeed, at this dose, both regulatory events could coexist, with a flutamide-induced down-regulation of claudin-11 expression that could be opposed to the loss of postmeiotic germ cells up-regulating claudin-11 expression.
TABLE 3. Apoptotic germ cell number in adult animals exposed in utero to flutamide
Because it was shown that claudin-11 was positively regulated by androgens, both AR expression and plasma testosterone levels were studied in postnatal animals exposed to the antiandrogen. No alterations in AR mRNA and protein levels were found in either prepubertal (14 d old) or adult (90 d old) rat testes exposed in utero to the three doses of flutamide tested. Hormone assay showed that serum testosterone levels were not significantly different from control levels in rats exposed in utero to flutamide either at a prepubertal age (14 d old), which has not been previously reported, or during adulthood (90 d old), and this is consistent with the literature (10, 11, 42). Taken together, these results suggest that specifically in rat testes exposed in utero to flutamide, alterations in the expression of claudin-11 by the antiandrogen are probably exerted at a post-AR level(s). This post-AR lesion(s) could be related to some changes in epigenetic mechanisms; for example in the methylation pattern of the androgen-dependent target gene promoters, as suggested by McLachlan for c-fos and lactotransferrin gene promoters in diethylstilbestrol-treated animals (43). Moreover, because during fetal exposure to the antiandrogen flutamide, AR expression in seminiferous tubules occurs mainly in peritubular myoid cells (44) (our unpublished data), it is possible that the participation of these cells is crucial to alterations in claudin-11 expression in postnatal Sertoli cells. Moreover, because claudin-11 is negatively regulated by FSH (20), we evaluated both plasma FSH levels and FSH-R expression, which were not found to be affected. Finally, the present findings of the positive control exerted by testosterone on claudin-11 expression suggest that testosterone could have an important role in modulation of the structure of the hemotesticular barrier and could be one of the factors controlling the opening and closing of this barrier during germ cell translocation into the adluminal compartment during the course of spermatogenesis. Such a positive regulatory action of the androgen is consistent with the report by Gye (21). The events contributing to the opening and closing of the hemotesticular barrier still remain unclear, but in view of our results and those of others, it could be hypothesized that both the endocrine system and the germ cells themselves, via factors they synthesize, could control the process of hemotesticular barrier complex dissolution and reformation that occurs during germ cell passage into the adluminal compartment. Indeed, the inhibitory effects of FSH, by decreasing claudin-11 expression (20), may favor the opening of junctions between Sertoli cells, allowing germ cell translocation into the adluminal compartment. The corollary of such a hypothesis is the existence of a stimulating factor, which could enhance claudin-11 levels, leading to closing of the hemotesticular barrier at the end of germ cell translocation. Based on our present findings and those of others, it could be suggested that testosterone, by stimulating claudin-11 [either directly (45) or indirectly via proteins interacting with claudin-11, such as integrin ?1, OAP-1 (46), or occludin (45)], could allow the reassembly of the barrier after germ cells have translocated. In contrast, postmeiotic germ cells, by regulating claudin-11 expression, may control translocation of premeiotic germ cells into the adluminal compartment to undergo meiosis, perhaps via (local) factors originating from germ cells, such as TNF (20) and TGF?3 (39, 40, 41), which (negatively) control claudin-11 in Sertoli cells.
Together, fetal androgen disruption appears to alter, in a complex manner, claudin-11 expression in postnatal rat testis. Indeed, although in the prepubertal testis, i.e. at the crucial moment of establishment of the hemotesticular barrier, a clear inhibitory effect is observed, in the adult testis, the apoptotic process that affects postmeiotic germ cells is likely to increase claudin-11 expression, as supported by the locally irradiated testis and coculture models we used here. These findings suggest that in utero exposure to flutamide may induce hypospermatogenesis through alterations of claudin-11 expression, potentially resulting in hemotesticular barrier (re)modeling impairment, although additional experiments are required to confirm this.
Finally, because claudin-11 is also expressed in the brain, we show in this study that claudin-11 expression is altered in the testis, but not in the brain, of male rats treated in utero with the antiandrogen flutamide, suggesting that transcription factors and coregulator elements in testis and brain might be different, thus leading to tissue specificity.
In summary, the present study shows the variations in the expression of claudin-11 in testes of prepubertal and adult rats exposed during gestation to the antiandrogen flutamide. Our results indicate that claudin-11 expression is altered in postnatal testes with fetal androgen disruption, and that in Sertoli cells, this gene and its product are under dual and opposite regulation, i.e. a stimulatory action exerted by androgens and an inhibitory effect originating from postmeiotic germ cells.
References
Kelce WR, Gray LE, Wilson EM 1998 Antiandrogens as environmental endocrine disruptors. Reprod Fertil Dev 10:105–111
Mylchreest E, Cattley RC, Foster PM 1998 Male reproductive tract malformations in rats following gestational and lactational exposure to di(n-butyl) phthalate: an antiandrogenic mechanism? Toxicol Sci 43:47–60
You L, Casanova M, Archibeque-Engle S, Sar M, Fan LQ, Heck HA 1998 Impaired male sexual development in perinatal Sprague-Dawley and Long-Evans hooded rats exposed in utero and lactationally to p,p'-DDE. Toxicol Sci 45:162–173
Gray Jr LE, Wolf C, Lambright C, Mann P, Price M, Cooper RL, Ostby J 1999 Administration of potentially antiandrogenic pesticides (procymidone, linuron, iprodione, chlozolinate, p,p'-DDE, and ketoconazole) and toxic substances (dibutyl- and diethylhexyl phthalate, PCB 169, and ethane dimethane sulphonate) during sexual differentiation produces diverse profiles of reproductive malformations in the male rat. Toxicol Ind Health 15:94–118
Gray LE, Ostby J, Furr J, Wolf CJ, Lambright C, Parks L, Veeramachaneni DN, Wilson V, Price M, Hotchkiss A, Orlando E, Guillette L 2001 Effects of environmental antiandrogens on reproductive development in experimental animals. Hum Reprod Update 7:248–264
Toppari J 2002 Environmental endocrine disrupters and disorders of sexual differentiation. Semin Reprod Med 20:305–312
McIntyre BS, Barlow NJ, Foster PM 2001 Androgen-mediated development in male rat offspring exposed to flutamide in utero: permanence and correlation of early postnatal changes in anogenital distance and nipple retention with malformations in androgen-dependent tissues. Toxicol Sci 62:236–249
Imperato-McGinley J, Sanchez RS, Spencer JR, Yee B, Vaughan ED 1992 Comparison of the effects of the 5-reductase inhibitor finasteride and the antiandrogen flutamide on prostate and genital differentiation: dose-response studies. Endocrinology 131:1149–1156
van der Schoot P 1992 Disturbed testicular descent in the rat after prenatal exposure to the antiandrogen flutamide. J Reprod Fertil 96:483–496
Kassim NM, McDonald SW, Reid O, Bennett NK, Gilmore DP, Payne AP 1997 The effects of pre- and postnatal exposure to the nonsteroidal antiandrogen flutamide on testis descent and morphology in the albino Swiss rat. J Anat 190:577–588
Miyata K, Yabushita S, Sukata T, Sano M, Yoshino H, Nakanishi T, Okuno Y, Matsuo M 2002 Effects of perinatal exposure to flutamide on sex hormones and androgen-dependent organs in F1 male rats. J Toxicol Sci 27:19–33
Omezzine A, Chater S, Mauduit C, Florin A, Tabone E, Chuzel F, Bars R, Benahmed M 2003 Long-term apoptotic cell death process with increased expression and activation of caspase-3 and -6 in adult rat germ cells exposed in utero to flutamide. Endocrinology 144:648–661
Bozec A, Chuzel F, Chater S, Paulin C, Bars R, Benahmed M, Mauduit C 2004 Mitochondrial-dependent pathway is chronically affected in testicular germ cell death in adult rat exposed in utero to anti-androgens. J Endocrinol 183:79–90
Bremner WJ, Millar MR, Sharpe RM, Saunders PT 1994 Immunohistochemical localization of androgen receptors in the rat testis: evidence for stage-dependent expression and regulation by androgens. Endocrinology 135:1227–1234
De Gendt K, Swinnen JV, Saunders PT, Schoonjans L, Dewerchin M, Devos A, Tan K, Atanassova N, Claessens F, Lecureuil C, Heyns W, Carmeliet P, Guillou F, Sharpe RM, Verhoeven G 2004 A Sertoli cell-selective knockout of the androgen receptor causes spermatogenic arrest in meiosis. Proc Natl Acad Sci USA 101:1327–1332
Morita K, Sasaki H, Fujimoto K, Furuse M, Tsukita S 1999 Claudin-11/OSP-based tight junctions of myelin sheaths in brain and Sertoli cells in testis. J Cell Biol 145:579–588
Bronstein JM, Popper P, Micevych PE, Farber DB 1996 Isolation and characterization of a novel oligodendrocyte-specific protein. Neurology 47:772–778
Bronstein JM, Micevych PE, Chen K 1997 Oligodendrocyte-specific protein (OSP) is a major component of CNS myelin. J Neurosci Res 50:713–720
Gow A, Southwood CM, Li JS, Pariali M, Riordan GP, Brodie SE, Danias J, Bronstein JM, Kachar B, Lazzarini RA 1999 CNS myelin and Sertoli cell tight junction strands are absent in Osp/claudin-11 null mice. Cell 99:649–659
Hellani A, Ji J, Mauduit C, Deschildre C, Tabone E, Benahmed M 2000 Developmental and hormonal regulation of the expression of oligodendrocyte-specific protein/claudin 11 in mouse testis. Endocrinology 141:3012–3019
Gye MC 2003 Changes in the expression of claudins and transepithelial electrical resistance of mouse Sertoli cells by Leydig cell coculture. Int J Androl 26:271–278
Dorrington JH, Roller NF, Fritz IB 1975 Effects of follicle-stimulating hormone on cultures of Sertoli cell preparations. Mol Cell Endocrinol 3:57–70
Boussouar F, Mauduit C, Tabone E, Pellerin L, Magistretti PJ, Benahmed M 2003 Developmental and hormonal regulation of the monocarboxylate transporter 2 (MCT2) expression in the mouse germ cells. Biol Reprod 69:1069–1078
Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159
Foley KP, Leonard MW, Engel JD 1993 Quantitation of RNA using the polymerase chain reaction. Trends Genet 9:380–385
Mauduit C, Siah A, Foch M, Chapet O, Clippe S, Gerard JP, Benahmed M 2001 Differential expression of growth factors in irradiated mouse testes. Int J Radiat Oncol Biol Phys 50:203–212
Ketola I, Rahman N, Toppari J, Bielinska M, Porter-Tinge SB, Tapanainen JS, Huhtaniemi IT, Wilson DB, Heikinheimo M 1999 Expression and regulation of transcription factors GATA-4 and GATA-6 in developing mouse testis. Endocrinology 140:1470–1480
Hopkinson CR, Dulisch B, Gauss G, Hilscher W, Hirschhauser C 1978 The effect of local testicular irradiation on testicular histology and plasma hormone levels in the male rat. Acta Endocrinol (Copenh) 87:413–423
Kangasniemi M, Kaipia A, Toppari J, Mali P, Huhtaniemi I, Parvinen M 1990 Cellular regulation of basal and FSH-stimulated cyclic AMP production in irradiated rat testes. Anat Rec 227:32–36
Heiskala M, Peterson PA, Yang Y 2001 The roles of claudin superfamily proteins in paracellular transport. Traffic 2:93–98
Sharpe RM 1994 Regulation of spermatogenesis. In: Knobil E, Neill JD, Ewing LL, Greenwald GS, Markert CL, Pfaff DW, eds. The physiology of reproduction. New York: Raven Press; 1363–1434
Ivell R, Spiess A-N 2002 Analysing differential gene expression in the testis. In: Rommerts FF, Teerds KJ, eds. Testicular tangrams. Berlin: Springer; 99–120
Comhaire FH, Vermeulen A 1976 Testosterone concentration in the fluids of seminiferous tubules, the interstitium and the rete testis of the rat. J Endocrinol 70:229–235
De SK, Chen HL, Pace JL, Hunt JS, Terranova PF, Enders GC 1993 Expression of tumor necrosis factor- in mouse spermatogenic cells. Endocrinology 133:389–396
Caussanel V, Tabone E, Hendrick JC, Dacheux F, Benahmed M 1997 Cellular distribution of transforming growth factor ?s 1, 2, and 3 and their types I and II receptors during postnatal development and spermatogenesis in the boar testis. Biol Reprod 56:357–367
Olaso R, Pairault C, Saez JM, Habert R 1999 Transforming growth factor ?3 in the fetal and neonatal rat testis: immunolocalization and effect on fetal Leydig cell function. Histochem Cell Biol 112:247–254
Mauduit C, Besset V, Caussanel V, Benahmed M 1996 Tumor necrosis factor receptor p55 is under hormonal (follicle-stimulating hormone) control in testicular Sertoli cells. Biochem Biophys Res Commun 224:631–637
Olaso R, Pairault C, Habert R 1998 Expression of type I and II receptors for transforming growth factor ? in the adult rat testis. Histochem Cell Biol 110:613–618
Lui WY, Lee WM, Cheng CY 2001 Transforming growth factor-?3 perturbs the inter-Sertoli tight junction permeability barrier in vitro possibly mediated via its effects on occludin, zonula occludens-1, and claudin-11. Endocrinology 142:1865–1877
Lui WY, Wong CH, Mruk DD, Cheng CY 2003 TGF-?3 regulates the blood-testis barrier dynamics via the p38 mitogen activated protein (MAP) kinase pathway: an in vivo study. Endocrinology 144:1139–1142
Lui WY, Lee WM, Cheng CY 2003 Transforming growth factor ?3 regulates the dynamics of Sertoli cell tight junctions via the p38 mitogen-activated protein kinase pathway. Biol Reprod 68:1597–1612
Goddard I, Florin A, Mauduit C, Tabone E, Contard P, Bars R, Chuzel F, Benahmed M 2003 Alteration of lactate production and transport in the adult rat testis exposed in utero to flutamide. Mol Cell Endocrinol 206:137–146
McLachlan JA 2001 Environmental signaling: what embryos and evolution teach us about endocrine disrupting chemicals. Endocr Rev 22:319–341
You L, Sar M 1998 Androgen receptor expression in the testes and epididymides of prenatal and postnatal Sprague-Dawley rats. Endocrine 9:253–261
Chung NP, Cheng CY 2001 Is cadmium chloride-induced inter-Sertoli tight junction permeability barrier disruption a suitable in vitro model to study the events of junction disassembly during spermatogenesis in the rat testis? Endocrinology 142:1878–1888
Tiwari-Woodruff SK, Buznikov AG, Vu TQ, Micevych PE, Chen K, Kornblum HI, Bronstein JM 2001 OSP/claudin-11 forms a complex with a novel member of the tetraspanin super family and ?1 integrin and regulates proliferation and migration of oligodendrocytes. J Cell Biol 153:295–305(Anne Florin, Magali Maire)