Developmental and Hormonal Regulation of the Expression of ...

0013-7227/00/$03.00/0 Endocrinology Copyright © 2000 by The Endocrine Society

Vol. 141, No. 8 Printed in U.S.A.

Developmental and Hormonal Regulation of the Expression of Oligodendrocyte-Specific Protein/Claudin 11 in Mouse Testis* ALI HELLANI, JINGWEI JI, CLAIRE MAUDUIT, CATHERINE DESCHILDRE, ERIC TABONE, AND MOHAMED BENAHMED INSERM, U-407, Communications Cellulaires en Biologie de la Reproduction, Faculte´ de Me´decine Lyon-Sud, F-69921 Oullins, France ABSTRACT The proliferation and differentiation of testicular progenitor stem cells into highly specialized germ cells (spermatozoa) are largely controlled by the hormonally (FSH and testosterone) regulated adjacent supporting Sertoli cells. However, the factors involved in this control remain largely unknown. In the present study, the technique of differential display PCR was used to identify target transcripts to FSH action in cultured murine Sertoli cells. Among these target transcripts, we identified the oligodendrocyte-specific protein (OSP), also known as claudin 11, which had recently been shown to play a key role in the formation of the hematotesticular barrier. Our data show that the testicular expression of OSP is dependent upon male gonad development and systemic and local signaling molecules. Indeed, OSP is expressed early in fetal development in Sertoli cells, immediately after the peak of SRY (sex-determining region, Y gene) expression, but just before that of the anti-Mullerian hormone. Postnatally, OSP expression starts to increase from day 3 to reach a plateau between days 6 and 16 postnatally. In the prepubertal and adult testes, an

S

PERMATOGENESIS IS the process by which spermatogonia (stem cells) develop into highly specialized cells, spermatozoa. Such a process is highly dependent upon Sertoli cells, which were originally defined and are still recognized as nourishing and supporting cells for the highly differentiated germ cells (1– 4). The differentiation of Sertoli cells is believed to be the first key event in the differentiation of the indifferent gonad into the testis (5). The choice of this developmental pathway depends on the presence of a key gene, the sex-determining region on the Y chromosome (SRY) (6 – 8). SRY expression induces the indifferent gonad to begin differentiating as a testis, involving a large variety of biological processes, such as somatic (Sertoli and Leydig) cell differentiation, germ and somatic cell migration and proliferation, testis cord formation, and vascularization (9). The differentiated somatic cells produce two signaling molecules, testosterone and anti-Mullerian hormone (AMH), crucial for male differentiation during fetal development (2, 10, 11). Spermatogenesis in the adult is still dependent upon Sertoli cells, as it is these cells that are under direct endocrine Received February 22, 2000. Address all correspondence and requests for reprints to: Dr. Mohamed Benahmed, INSERM, U-407, Faculte´ de Me´decine Lyon-Sud, BP 12, F-69921 Oullins Cedex, France. E-mail: [email protected]. * This work was supported by INSERM, U-407, and in part by Association pour la Recherche contre le Cancer (ARC 9678 and ARC 7384).

apparent decline in OSP messenger RNA (mRNA) levels was found, probably because of the increasing number of germ cells (which do not express OSP). Among the signaling molecules that control testicular OSP expression, we have identified FSH and tumor necrosis factor-␣ (TNF␣). Indeed, using a model of purified cultured mouse Sertoli cells, we demonstrate that FSH inhibits, in a dose (ED50 ⫽ 4 ng/ml)- and time (maximal effect after 24 h)-dependent manner, the levels of OSP mRNA. Such an inhibitory effect was mimicked by 8-bromo-cAMP, suggesting that FSH may use the cAMP/protein kinase A pathway to inhibit OSP mRNA levels. TNF␣ was also shown to inhibit OSP expression in cultured Sertoli cells. The maximal effect was observed after 48 h of TNF␣ treatment with an ED50 of 4.5 ng/ml. Together, our results indicate that OSP expression 1) starts during fetal life at a critical period, probably under SRY control and during testicular formation; and 2) is regulated by hormones (FSH) and cytokines (TNF␣) in the adult testis, suggesting a critical role for these molecules in the (re)modeling process of the hematotesticular barrier during spermatogenesis. (Endocrinology 141: 3012–3019, 2000)

control (2, 3, 12). Endocrine regulation allows Sertoli cells to develop and reorganize to generate the hematotesticular barrier through their tight junction complexes and to provide nutrients and regulatory factors to the germ cells. This timemodeling process and the production of Sertoli cell factors lead to the constitution of a specific biochemical and cytoarchitectural microenvironment in the adluminal compartment where germ cells will proliferate and differentiate. Among the identified Sertoli cell products are binding transport proteins, proteases and protease inhibitors, energy substrates such as lactate, and local signaling molecules such as growth factors and cytokines (13–18). Although hormones (LH/testosterone and FSH) exert their control on spermatogenesis via these above-mentioned identified components, there are still other hormonally regulated Sertoli cell factors that remain unknown. To identify these factors that are under FSH control, we have used a method based on messenger RNA (mRNA) differential display in a model of purified mouse Sertoli cells cultured in the absence or presence of FSH. In the present study among the different transcripts differentially regulated, oligodendrocyte-specific protein (OSP) mRNA was identified. OSP was initially shown to be produced in oligodendrocytes, which are supporting cells for the highly specialized neurons (19). Recently, OSP has been identified as a new claudin family member (claudin 11) due to its sequence similarity (30%) and its ability to form tight junction strands in transfected fibroblasts (20). The key role

3012

HORMONAL REGULATION OF OSP EXPRESSION IN SERTOLI CELLS

of OSP in central nervous system myelin and Sertoli cell tight junction strands was demonstrated in osp/claudin 11 null mice (21). However, the control of the expression of this potentially crucial factor in the formation of the hematotesticular barrier is yet unknown. In the present study we investigated the possibility that OSP expression might be developmentally and hormonally regulated in the testis. We report here that OSP expression is 1) regulated during testicular development, as it appears just after Sry expression and before that of Amh; and 2) targeted by the actions of FSH and tumor necrosis factor-␣ (TNF␣) in the postnatal testis. This last observation clearly supports the concept that the remodeling process affecting the hematotesticular barrier during spermatogenesis might be under the control of hormones and local signaling factors. Materials and Methods Materials DMEM/Ham’s F-12 medium, deoxy (d)-NTPs, Moloney murine leukemia virus reverse transcriptase, SuperScript reverse transcriptase, deoxyribonuclease I (DNase I), FCS, and TRIzol were obtained from Life Technologies, Inc. (Eragny, France). Collagenase/dispase was obtained from Roche (Mannheim, Germany). Ovine FSH was a gift from A. F. Parlow (National Hormone and Pituitary Program, Torrance, CA). Human recombinant TNF␣ was purchased from TEBU (Le Perray en Yvelines, France). Sigma (Isle d’Abeau, France) was the source for 8bromo-cAMP, transferrin, gentamicin, HEPES, sodium bicarbonate, ␣-tocopherol, trypsin, and DNase I. RNA image kits were purchased from GenHunter Corp. (Nashville, TN). Taq polymerase was obtained from Promega Corp. (Lyon, France). The Amplicycle sequencing kit was purchased from Perkin-Elmer Corp. (Courtaboeuf, France). [␣-33P]dATP (1000 –3000 Ci/mmol) and [␣-33P]dCTP (3000 Ci/mmol) were purchased from Amersham Pharmacia Biotech (Aylesbury, UK). Oligonucleotides were purchased from GENSET (Paris, France). Primary and secondary antibodies and the chemiluminescence detection kit were obtained from Covalab (Lyon, France).

3013

were plated on Falcon petri dishes (Becton Dickinson & Co., Franklin Lakes, NJ; 10-mm diameter, 5 ⫻ 106 cells/dish) and cultured at 32 C in a humidified atmosphere of 5% CO2-95% air.

Isolation of RNA Total RNAs were prepared using TRIzol, a monophasic solution of phenol and guanidine isothiocyanate. This reagent is an improvement over the single-step RNA isolation method developed by Chomczynski and Sacchi (23). The amount of RNA was estimated by spectrophotometry at 260 nm.

mRNA differential display mRNA differential display was performed with the RNAimage kit according to the protocol of Liang and Pardee (24). RNA preparations were submitted to DNase I digestion (1 U DNase I for 1 ␮g RNA, 15 min at room temperature) to remove DNA contamination. Complementary DNA (cDNA) was synthesized from 0.5 ␮g total RNA using SuperScript reverse transcriptase (1 U), one of three 1-base anchored oligo(deoxythymidine) primers (0.2 ␮m), and 20 ␮m dNTPs (1 h, 37 C). The cDNAs generated were used for the PCR display; one of the three reverse primers was used in conjunction with one of the eight forward arbitrary 13-mer primers (AAGCT11V). Briefly, 2 ␮l of 1:4 diluted cDNA were amplified using Taq polymerase (2 U) in a total volume of 20 ␮l containing 10 mm Tris-HCl (pH 9.0), 50 mm KCl, 2.5 mm MgCl2 0.1% Triton X-100, 0.2 ␮m dNTPs, 0.2 ␮m of each primer, and 0.2 ␮l [␣-33P]dATP. After a 94 C, 5-min denaturing step, PCR cycles consisted of 30 sec at 94 C, 2 min at 40 C, and 30 sec at 72 C for 40 cycles, followed by a 72 C, 7-min elongating step. The amplified products were separated on 8 m urea 6% acrylamide gels, then dried and exposed to Kodak film (Eastman Kodak Co., Rochester, NY) for 24 – 48 h. After development, lanes containing amplified products from FSH-treated cells or untreated cells were compared. Bands that showed differential expression were excised from the gel, DNA was extracted by boiling in 100 ␮l sterile H2O and precipitated with ethanol 85% (2 vol) and sodium acetate (0.3 m) in the presence of glycogen (50 ␮g). DNA was reamplified using the original set of primers and the same thermal cycling conditions, except that the deoxynucleoside triphosphate concentration was increased to 250 ␮m. Reaction products were run on a 1% agarose gel and stained with ethidum bromide. Bands were excised from the gel, eluted, and used for sequencing or as a probe for Northern blot analysis.

Animals OF1 strain mice were purchased from IFFA CREDDO (Lyon, France). Virgin female CD-1 mice were individually housed in controlled conditions of lighting (12 h of light, 12 h of darkness), temperature (20 –24 C), and humidity (40 –70%) and were given free access to water and feed. Females were mated on a one to one basis with male animals of the same strain and the same supplier. Day 0 of gestation was when a vaginal plug was observed. At the appropriate times, mice were killed by asphyxiation in CO2. All studies were approved by the INSERM animal care committee.

Isolation and culture of Sertoli cells Sertoli cells were prepared from 16- to 18-day-old mice, as described by Dorrington et al. (22). First, albuginea was removed, then testicular tissues were mechanically dispersed with forceps in DMEM/Ham’s F-12 (1:1) medium (1.2 mg/ml sodium bicarbonate, 15 mm HEPES, and 20 ␮g/ml gentamicin) containing DNase I (0.05 mg/ml). Testicular tissues were further dispersed by collagenase/dispase treatment (0.5 mg/ml, 30 min at 32 C) in the presence of DNase I (0.05 mg/ml) and FCS (2%) in DMEM/Ham’s F-12 medium through mild stirring. At the end of the enzymatic dissociation, testicular cells were submitted three times to gravity sedimentation (3–5 min), and supernatants were removed. The pellets containing the sedimented tubules were further dissociated with a collagenase treatment as described above until small clumps resulted. The cells where then submitted to a gravity sedimentation (10 –15 min). The supernatant was removed, and the sedimented clumps of Sertoli cells were further washed by centrifugation (200 ⫻ g, 10 min). The Sertoli cell pellets were resuspended in DMEM/Ham’s F-12 medium (supplemented with 5 ␮g/ml transferrin and 10 ␮g/ml vitamin E), and cells

Northern blot analysis About 20 ␮g total RNA were loaded on a 1.2% agarose/2.2 m formaldehyde gel. After 5 h of migration in 0.02 m 3(N-morpholino)propane sulfonic acid (MOPS) running buffer, RNAs were transferred to nitrocellulose membrane (Hybond-C Extra, Amersham Pharmacia Biotech) by capillary transfer with 10 ⫻ SSC (standard saline citrate) and fixed at 80 C for 2 h. The probes used for hybridization were a 339-bp OSP cDNA, and a 1.3-kb PstI rat glyceraldehyde-3-phosphate dehydrogenase cDNA (GAPDH). Probes were labeled with 50 ␮Ci [␣-32P]dCTP using a random primed DNA labeling kit (SA, 109 dpm/␮g DNA). The labeled probes were separated from free nucleotides by filtration through a diethylaminoethyl-cellulose column. After 4 h of prehybridization at 42 C, filters were hybridized with the labeled probe (1– 4 106 cpm/ml) overnight at 42 C in 50% formamide, 5 ⫻ SSPE (0.9 m NaCl, 50 mm sodium phosphate, 5 mm EDTA, pH 7.4), 5 ⫻ Denhardt’s solution, 1% SDS, and 100 mg/ml yeast RNA. Then, membranes were washed four times in 2 ⫻ SSC-0.1% SDS (15 min at room temperature), followed by 30 min at 55 C. Filters were exposed to Kodak X-OMAT films for 1–2 days at ⫺70 C. The intensities of the autoradiographic bands were estimated by densitometric scanning using Intelligent Quantifier software (BioImage, Cheshire, UK). The data were expressed as the OSP/GAPDH mRNA ratio.

Semiquantitative RT-PCR cDNAs were obtained from RT of 3 ␮g total RNA using random hexanucleotides as primer (50 ␮m) in the presence of dNTPs (250 ␮m), dithiothreitol (10 ␮m), and Moloney murine leukemia virus (10 U/␮l) for 1 h at 37 C. cDNAs (2 ␮l RT mixture) were amplified by PCR with Taq

3014


polymerase (0.01 U/␮l), dNTP (250 ␮m), [␣-33P]dATP (0.75 ␮Ci), and specific primers (10 ␮m). PCR amplification was performed by first heating the mixture at 92 C for 3 min, followed by X cycles (see Table 1) at 92 C for 30 sec, Tm C for 30 sec, 70 C for 30 sec, then 70 C for 5 min. PCR products were analyzed on an 8% polyacrylamide gel. Dried gels were exposed to Kodak X-OMAT films for 1–2 days at room temperature. PCR analyses were carried out from the logarithmic phase of amplification. PCR products were checked by sequencing. Oligonucleotide primers were designed inside separate exons to avoid any bias due to residual genomic contamination. Moreover, for all primers, no amplification was observed when PCR was performed on DNA preparation.

Sequencing Samples showing significant changes by Northern blot analysis were sequenced directly from the PCR product using the Amplicycle sequencing kit (Perkin-Elmer Corp.) following the manufacturer’s protocol. Differential display-PCR upstream arbitrary primers were used to locate the sequence. DNA sequences were analyzed by the BLAST search program and were screened for homology against the GenBank/EMBL database sequences.

Western blotting analysis and immunohistochemistry Two polyclonal antibodies (Covalab, Lyon, France) raised against synthetic peptides corresponding to the 1) C-amino acid terminal or 2) the N-amino acid terminal sequence of OSP were used for the immunological approaches. Proteins (10 ␮g) from whole adult testis and brain were analyzed on 15% SDS-polyacrylamide gels and electrophoretically transferred to nitrocellulose membranes using 25 mm Tris and 185 mm glycine, pH 8.3, containing 20% methanol. The transfer was performed at a constant voltage of 100 V for 2 h. After transfer, the membrane was incubated in a blocking buffer [Tris-buffered saline (TBS) containing 1% BSA] overnight at 4 C. The membrane was rinsed three times with TBS/0.1% Tween (three times, 10 min each time), then incubated with the primary antibody (1:500 dilution in TBS) for 2 h at room temperature. The membrane was rinsed with TBS/0.1% Tween (three times, 10 min each time) and then incubated with horseradish peroxidase-labeled goat antirabbit IgG (1:1000 dilution in TBS) for 1 h at room temperature. The membrane was thoroughly washed with TBS/Tween (three times, 10 min each time) and then with TBS. Bound antibodies were detected using the chemiluminescence detection kit and Biomax MR films from Eastman Kodak Co. The protein concentration was determined by the Bradford assay. For the immunohistochemistry analysis, paraffin sections (5 ␮m) of Bouin-fixed testis were mounted on glass slides. The sections were deparaffinized and rehydrated. The UltraVision Detection System (Lab Vision Corp., Fremont, CA) was used as recommended by the manufacturer. Briefly, endogenous peroxidases were blocked with 3% H2O2 for 15 min. The sections were incubated for 5 min with a protein-blocking solution to minimize nonspecific binding. The rabbit primary antibody was diluted (1:20) in antibody diluent (DAKO Corp., Trappes, France) and incubated with the sections for 2 h at room temperature. After washing and incubation with the biotinylated secondary antibody, a peroxidase-streptavidin complex was applied. Diaminobenzidene was used as the peroxidase chromogen. Sections were briefly counterstained

Endo • 2000 Vol 141 • No 8

with Harris hematoxylin and mounted in mounting medium. Antibody diluent alone or normal rabbit serum was used as the negative control.

Statistical analysis Experiments were repeated at least three times with independent cell preparations. The statistical significance of the results was determined by Student’s t test when comparing data from three experiments. Data are presented as the mean ⫾ sd.

Results OSP expression in the testis

The mRNA differential display technique was performed using total RNA from untreated or FSH-treated mouse Sertoli cells. Using combinations of primer sets, 14 differentially expressed cDNAs were identified, extracted from the gel, reamplified, and used as probes for Northern blot. Three cDNA fragments showed a differentially expressed pattern. One of these cDNAs, which is presented here, is 253 bp in length. The cDNA reamplification product was sequenced and used to search the GenEMBL database with the BLAST search. This cDNA has shown full homology (1548 –1801 bp) with OSP. Total RNA from a large variety of mouse tissues (lung, heart, brain, kidney, skeletal muscle, ovary, and testis) were hybridized with 32P-labeled OSP probe to identify the specific expression of OSP. The results obtained show specific expression of OSP in the testis and, as expected, in the brain (Fig. 1A). It is interesting to note that OSP was expressed in the testis but not in the ovary, whereas it was expressed in both male and female brains (Fig. 1B). Finally, the data in Fig. 1C show that testicular OSP mRNA is translated into protein. Indeed, by using polyclonal antibodies raised against the N-and C-terminal parts of OSP, we identified through Western blotting analysis a protein with a molecular mass of 21 kDa and showed through an immunohistochemical approach the specific expression of OSP in Sertoli cells, but not in the other cell types (Fig. 2). OSP expression during testicular development

To determine whether OSP is developmentally regulated, we examined its expression in fetal and postnatal testes by semiquantitative RT-PCR and Northern blotting analysis, respectively. In the fetal period, OSP mRNA levels were examined at the time of initiation of testis differentiation between 10 and 15 days postcoitum (dpc). The data in Fig. 3 show that SRY expression was detected at 11 dpc and then was almost absent at 12 dpc. AMH was expressed soon after

TABLE 1. PCR amplification Primer sequences

OSP SRY AMH

␤ACTIN

GATTGGCATCATCGTCACAACG (forward) AGCCAGCAGAATAAGGAGCACC (reverse) GGTTGCAATCATAATTCTTCC (forward) CACTCCTCTGTGACACTTTAG (reverse) GCTGCAGGCCTTTCTGTTTGGC (forward) GCAGCAGTAATAGGGGTTCC (reverse) TTGCTGATCCACATCTGCT (forward) GACAGGATGCAGAAGGAGAT (reverse)

Amplified product size

Tm Temp (C)

No. of cycles

339 bp

55

20

396 bp

55

20

362 bp

55

20

146 bp

61

18


3015

FIG. 1. Specific expression of OSP mRNA in the testis and the brain. A, Twenty micrograms of total RNA extracted from 3-week-old mouse tissues were used to perform Northern blotting analysis with a [32P]OSP (339-bp) probe. B, Twenty micrograms of total RNA extracted from 3-week-old mouse tissues (female and male brain, ovary, and testis) were used to perform Northern blotting analysis with a 32 P-labeled OSP (339-bp) probe. C, Ten micrograms of protein from 3-week-old mouse brain and testis (second and third lanes) were used to perform Western blotting analysis. Membranes were incubated with antibody against the C-terminus of OSP (dilution, 1:500; lane brain and lane testis) or with antibody plus synthetic peptide (lane Ab⫹Ag).

OSP at 13 dpc. Thus, OSP expression appears and is upregulated just before the first sign of testis development; i.e. the formation of the testis cords that is observable at 12 dpc. In the female (XX), no OSP expression was detected during fetal life (data not shown). From day 3 postnatally OSP expression started to increase until a plateau was reached between days 6 and 16 (Fig. 4). From day 16, there was an apparent decline in OSP mRNA. Such a decrease in OSP mRNA levels might be due, at this age, to the increasing number of germ cells that do not express OSP. OSP expression is under the control of FSH and TNF␣ in cultured Sertoli cells

As adult testicular function is under the control of systemic and local signaling molecules and based on our initial observations using the approach of mRNA differential display

FIG. 2. Specific expression of OSP protein in Sertoli cells. Testis sections were incubated with antibody raised against the N-terminal portion of OSP (dilution, 1:20). A, Survey view of two seminiferous tubules showing intratubular positive staining. The anti-OSP antibody specifically stains Sertoli cells. Bar, 100 ␮m. B, Higher magnification of the basal portion of a seminiferous tubule. Germ cell lineage is negative, whereas Sertoli cell cytoplasm is highly positive, underlining the connection between neighboring Sertoli cells. Bar, 20 ␮m).

3016


suggesting that OSP is targeted by FSH action, we tested a possible hormonal and local regulation of OSP. For this purpose, use was made of a model of cultured mouse Sertoli cells treated with several hormones and growth factors potentially present in the microenvironment of the Sertoli cells. Among the different signaling molecules tested that may affect OSP expression, we identified FSH and TNF␣. In Fig. 5, Sertoli cells were exposed to various FSH concentrations (0.5–50 ng/ml) for different times (0 – 48 h), and total RNA was extracted and analyzed through Northern blotting analysis. The data obtained show that FSH decreased OSP mRNA levels. The inhibitory effect of FSH on OSP mRNA levels was dose dependent, as it was detectable at a concentration of 0.5 ng/ml and was maximal at 20 ng/ml (P ⬍ 0.001; Fig. 5A). This inhibitory effect on OSP mRNA was also time dependent, as FSH decreased OSP mRNA levels after 6 h of exposure, with a maximal inhibition observed at 24 h (P ⬍ 0.001) but not at 48 h (Fig. 5B). To determine whether the effect of FSH is mediated by the cAMP/protein kinase A (PKA) pathway, cultured Sertoli cells were exposed to 0.5 mm 8-bromo-cAMP for 1, 6, 12, 24, and 48 h. The data in Fig. 5C show an inhibition of OSP mRNA levels in 8-bromocAMP-treated Sertoli cells close, if not similar (P ⬍ 0.001), to that observed in FSH-treated cells. In Fig. 6, OSP mRNA levels were shown to be inhibited by TNF␣ in a dose- and time-dependent manner. The inhibitory effect of TNF␣ on OSP mRNA levels was detectable at a

FIG. 3. OSP expression in the fetal testis. RNAs from testes of 10 –15 dpc embryos were tested by semiquantitative RT-PCR for the expression of OSP, SRY, and AMH (␤-actin was used as standard).

FIG. 4. Postnatal expression of OSP. A, RNA from 1–55 postnatal mouse testes were analyzed by Northern blotting using a labeled [32P]OSP probe. Upper panel, OSP and 28S transcripts; lower panel, data yielded by scanning autoradiographs, expressed as OSP mRNA/ 28S ratios.

Endo • 2000 Vol 141 • No 8

concentration of 2.5 ng/ml and was maximal at 20 ng/ml (P ⬍ 0.001; Fig. 6A). The TNF␣ inhibitory action was also evidenced after 12 h of exposure and was maximal at 48 h (P ⬍ 0.001; Fig. 6B). Discussion

In the present study we have characterized the developmental expression and the hormonal regulation of the OSP gene in the mouse testis, specifically in Sertoli cells. OSP initially cloned by Bronstein et al. (19) was found to be specific for oligodendrocytes and localized on chromosome 3 (3q26.2–26.3) (25). Initially, using an approach based on mRNA differential display in a model of cultured FSHtreated Sertoli cells, we detected OSP as a hormonally regulated gene in the testis. The testicular expression of OSP was shown in forms of mRNA during fetal and adult testis development in both the whole gonad as well as in purified cultured Sertoli cells under hormonal control. Furthermore, by using antibodies directed against the C-and N-terminal parts of the protein, we have shown that the OSP mRNA is translated into a protein of 21 kDa specifically present in Sertoli cells, but not in the other testicular cells. Concerning OSP expression in the course of testis development, it was found to be present in both the early fetal testis as well as the postnatal testis, but with a different pattern of expression. The testis exhibits two main and distinct activities in the course of its development. The first occurs during fetal life and is mainly related to the process of organogenesis, whereas the second major activity, which occurs throughout adult life, is mainly related to spermatogenesis. During fetal life, the indifferent gonad has the ability to develop into an ovary or a testis depending on several genes, including Sry (sex-determining region 4 chromosome) as well as Sox-9 (an Sry-related gene), WT-1 (Wilm’s tumor gene), and the nuclear receptors DAX-1 and SF-1. Genetic analysis of XX-XY chimeras indicates that SRY functions within the cell lineage that will differentiate into


3017

FIG. 6. Inhibitory effect of TNF␣ on OSP mRNA levels. A, Sertoli cells were incubated for 48 h in the absence or presence of increasing concentrations of TNF␣ (0.5–50 ng/ml). B, Sertoli cells were treated with TNF␣ (20 ng/ml) for 1– 48 h. Total cellular RNAs were then extracted, and Northern blotting analysis was performed using 20 ␮g total RNA/lane. Membranes were successively hybridized with the OSP and GAPDH cDNA. In the upper panel, a representative pattern of Northern blotting is shown; in the lower panel, data yielded by scanning autoradiographs are expressed as OSP/GAPDH mRNA ratios. Values are the mean ⫾ SD. Without TNF␣ (basal) vs. with TNF␣ (20 ng/ml), P ⬍ 0.001 (A).

FIG. 5. Inhibitory effect of FSH on OSP mRNA levels. A, Sertoli cells were incubated for 24 h in the absence or presence of increasing concentrations of FSH (0.5–50 ng/ml). B, Sertoli cells were treated with FSH (20 ng/ml) for 1– 48 h. C, Sertoli cells were incubated with 0.5 mM 8-bromo-cAMP for 1– 48 h. Total cellular RNA was then extracted, and Northern blotting analysis was performed using 20 ␮g total RNA/lane. Membranes were successively hybridized with OSP and GAPDH cDNAs. In the upper panel, a representative pattern of Northern blotting is shown; in the lower panel, data yielded by scanning autoradiographs are expressed as OSP/GAPDH mRNA ratios. Values are the mean ⫾ SD. Without FSH (basal) vs. with FSH (20 ng/ml), P ⬍ 0.001 (A). Without 8-bromo-cAMP (basal) vs. with 8bromo-cAMP, P ⬍ 0.001 (C).

Sertoli cells (9). OSP is expressed in the same lineage and immediately after the peak of SRY, suggesting that SRY could, theoretically, induce OSP expression. As SRY expression is turned off at 12 dpc, whereas OSP continues to be expressed, it is possible that SRY is not required for its main-

tenance. It is also possible that Sox-9 controls OSP, as Sox-9, which is already expressed in undifferentiated gonad, becomes abundant after 10.5 dpc in XY embryos and is absent in XX embryos (9). Furthermore, AMH, which is also a key gene in Sertoli cells, is expressed on 13 dpc, immediately after the expression of OSP. AMH, which belongs to the transforming growth factor-␤ peptide family and is classically known to be responsible for Mullerian duct regression, may also play a role in the development of fetal testicular cells, such as Leydig cells and germ cells (11, 26, 27). Together, these observations indicate that OSP is expressed at a critical period in the cascade of gene expression, between SRY and AMH, during Sertoli cell differentiation. In addition, the absence of OSP expression in the ovary, but not in the female brain, clearly indicates that OSP expression is tissue specific. The promoter of the testicular OSP gene is potentially responsive to the genes involved in the testis (Sertoli cell) differentiation, indicating that in the gonad, but not in the brain, SRY expression is probably required for that of OSP.

3018


After birth, OSP mRNA starts to increase from postnatal day 3 and reaches a maximal level between days 6 and 16 postnatally. At this period, during which OSP mRNA levels are at their highest, a crucial morphogenetic event occurs for the development of the adult testis: the formation of the hematotesticular barrier. Based on this timing of OSP expression in postnatal life, the hypothesis was that OSP may play a key role in the hematotesticular barrier. Such a possibility was demonstrated by two different groups during the course of our present study. Firstly, Morita et al. (20) identified OSP as a new claudin family member due to its sequence similarity (30%) and its ability to form tight junction strands in transfected fibroblasts. Secondly, the absence of tight junction intramembranous strands in central nervous system myelin and between Sertoli cells of osp/claudin 11 null mice (21) reinforced the concept for a key role for OSP in the hematotesticular barrier formation. One of the major contributions of our present work is the first demonstration that such a key factor involved in the formation of the hematotesticular barrier is targeted by gonadotropin and local signaling factor action, specifically by FSH and TNF␣ action. The actions of both molecules were identified using a model of cultured Sertoli cells. The action of FSH was mimicked by 8-bromo-cAMP, indicating that the hormone may use the cAMP/PKA/cAMP response elementbinding protein transducing pathway to reduce OSP mRNA levels. Interestingly, we confirmed such an inhibitory effect of FSH on testicular OSP expression in an in vivo model. Indeed, the daily sc administration of 25 IU FSH (for 2 days) to 32-day-old hypophysectomized rats induced a decline of about 50% in OSP mRNA levels in the testis (our unpublished data). TNF␣ also inhibited OSP expression at a concentration compatible with the Kd of its receptors, particularly the P55 form, which has been reported to be present in Sertoli cells (28, 29). The inhibitory effect of TNF␣ on OSP mRNA levels in primary cultures of Sertoli cells was exerted in the nanomolar concentration range. Such a concentration is consistent with the amounts of TNF␣ reported to be present in the Sertoli cell environment (28). As this testicular TNF␣ originates from the postmeiotic germ cells (28), it may be that in the adult testis, Sertoli cell OSP is also under the control of germ cells in the context of Sertoli cell-germ cell cooperation during spermatogenesis. Such a possibility is reinforced by the fact that higher levels of OSP mRNA were observed in genetically modified mice (white-Spotting, W/Wv) with testes devoid of germ cells (our unpublished data). Finally, for both FSH and TNF␣, it remains to be established whether these factors reduce OSP mRNA levels by acting at a transcriptional level and/or mRNA stabilization levels. The effects of FSH and TNF␣ on OSP expression we report here might be of particular interest concerning the regulatory mechanisms and factors entering the process of the hematotesticular barrier complex dissolution and reformation that occur during germ cell passage into the adluminal compartment. Indeed, as we previously mentioned, the intratesticular factors involved in such a process remain to be identified. We hypothesize that the modulatory actions of FSH and TNF␣ on OSP levels may participate as a key event in the dissolution and reformation of the testicular barrier, allow-

Endo • 2000 Vol 141 • No 8

ing germ cell translocation in the adluminal compartment. The inhibitory effects of the two signaling molecules, FSH and TNF␣, on OSP levels may favor opening of the junctions between Sertoli cells and therefore allow germ cell translocation into the adluminal compartment. Concerning FSH action, its inhibitory effect on OSP testicular expression may appear as a contradictory observation, assuming its well known positive action on the spermatogenesis process. In fact, by inhibiting OSP levels, FSH may favor the remodeling process (opening?) of the hematotesticular barrier and translocation of germ cells into the adluminal compartment. As such, FSH still positively regulates the spermatogenic process. Although we identified here two inhibitory signaling molecules of OSP expression, FSH and TNF␣, it is reasonable to suggest that other signaling molecules also exist that enhance OSP expression and will therefore contribute to a coordinated hematotesticular reformation once the translocated germ cells are present in the adluminal compartment. The identification of such factors stimulating OSP expression is currently under investigation in our laboratory. Finally, whether alterations in OSP expression might be involved in male infertility is as yet unknown. However, such a possibility exists in view of the localization of OSP on chromosome 3 (19). Indeed, this chromosome is of interest in relation to male fertility because 1) several genes, such as DAZH, DAZLA, and TAK1, suspected to be involved in the regulation of spermatogenesis (30, 31) are localized on chromosome 3; and 2) several cases of infertility have been observed in patients with genetic disorders involving this chromosome either in the context of translocations occurring on somatic (autosomal) chromosomes or of translocations between somatic and sex chromosomes (32–34). In summary, by using an approach initially based on DDPCR, we have identified the developmentally and hormonally dependent expression of OSP, a gene expressed in testicular Sertoli cells that has a potential key role in hematotesticular barrier formation. We reported here on 1) its tissue-specific expression, which appears dependent upon SRY in the testis; indeed, OSP is not present before SRY expression, and is absent in the ovary; and 2) its regulation by hormones and cytokines, supporting the concept that the remodeling process affecting the hematotesticular barrier during spermatogenesis is targeted by both the endocrine system and local regulatory factors. Acknowledgments We are indebted to Dr. S. Magre for providing the fetal testicular RNA. We thank Dr. I. Goddard for her help performing Western blotting experiments, and Drs. H. and M. Hellani and Mrs. A. McLeer for critical reading of the manuscript.

References 1. Sertoli E 1865 Dell’ esistenza di particulari cellule ramificate nei canalicoli seminiferi del testicolo umano. Morgagni 7:31– 40 2. Fawcett DW 1973 Observations on the organization of the interstitial tissue of the testis and on the occluding cell function in the seminiferous epithelium. Adv Biosci 10:83–99 3. Sharpe RM 1994 Regulation of spermatogenesis. In: Knobil E, Neill JD, Ewing LL, Greenwald GS, Markert CL, Pfaff DW (eds) The Physiology of Reproduction. Raven Press, New York, pp 1363–1434

HORMONAL REGULATION OF OSP EXPRESSION IN SERTOLI CELLS 4. Griswold MD 1995 Interactions between germ cells and Sertoli cells in the testis. Biol Reprod 52:211–216 5. Jost A, Magre S, Anjelopoulos R 1981 Early stages of testicular differentiatiotion in the rat. Hum Genet 58:59 – 64 6. Koopman P, Gubbay J, Vivian N, Goodfellow P, Lovell-Badge R 1991 Male development of chromosomally female mice transgenic for SRY. Nature 351:117–121 7. Sinclair AH, Berta P, Palmer MS, Hawkins JR, Griffiths BL, Smith MJ, Foster JW, Frischauf AM, Lovell-Badge R, Goodfellow PN 1990 A gene from human sex determinating gene encodes a protein with homology to a conserved DNA binding motif. Nature 346:240 –244 8. Berta P, Hawkins JR, Sinclair AH, Taylor A, Griffiths BL, Goodfellow PN, Fellous M 1990 Genetic evidence equating SRY and the testis determining factor. Nature 348:448 – 450 9. Capel B 1998 Sex in the 90s: SRY and the switch to the male pathway. Annu Rev Physiol 60:497–523 10. Huhtaniemi I, Pelliniemi LJ 1992 Fetal Leydig cells: cellular origin, morphology, life span, and special functional features. Proc Soc Exp Biol Med 201:125–140 11. Josso N, Lamarre I, Picard JY, Berta P, Davies N, Morichon N, Peschanski M, Jeny R 1993 Antimullerian hormone in early human development. Early Hum Dev 33:91–99 12. Griswold MD, Solari A, Tung PS, Fritz IB 1977 Stimulation by folliclestimulating hormone of DNA synthesis and of mitosis in cultured Sertoli cells prepared from testes of immature rats. Biol Reprod 52:211–216 13. Dym M, Fawcett DW 1971 Further observations on the numbers of spermatogonia, spermatocytes, and spermatids connected by intercellular bridges in the mammalian testis. Biol Reprod 4:195–215 14. Russell LD, Peterson RN 1985 Sertoli cell junctions: morphological and functional correlates. Int Rev Cytol 94:177–211 15. Grootegoed JA 1987 Towards a better understanding of Sertoli cell function. Int J Androl 10:727–729 16. Fritz IB, Tung PS, Ailenberg M 1993 Proteases and antiproteases in the seminiferous tubule. In: Russel LD, Griswold MD (eds) The Sertoli Cell. Cache River Press, Clearwater, pp 217–236 17. Benahmed M 1996 Growth factors and cytokines in the testis. In: Comhaire F (ed) Male Infertility. Chapman and Hall, London, pp 55–96 18. Gnessi L, Fabbi A, Spera G 1997 Gonadal peptides as mediators of development and functional control of the testis: an integrated system with hormones and local environment. Endocr Rev 18:541– 601 19. Bronstein JM, Kozak CA, Chen XN, Wu S, Danciger M, Korenberg JR, Farber DB 1996 Chromosomal localization of murine and human oligodendrocytespecific protein genes. Genomics 34:255–257

3019

20. Morita K, Sasaki H, Fujimoto K, Furuse M, Tsukita S 1999 Claudin-11/OSPbased tight junctions of myelin sheaths in brain and Sertoli cells in testis. J Cell Biol 145:579 –588 21. 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 22. Dorrington JH, Roller NF, Fritz IB 1975 Effects of follicle stimulating hormone on cultures of Sertoli cell preparations. Mol Cell Endocrinol 3:57– 62 23. Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156 –159 24. Liang P, Pardee AB 1992 Differential diplay of eukaryotic messenger RNA by means of the polymerase chain reaction. Science 257:967–971 25. Bronstein JM, Popper P, Micevych PE, Farber DB 1996 Isolation and characterization of a novel oligodendrocyte-specific protein. Neurology 47:772–778 26. Vigier B, Watrin F, Magre S, Tran D, Josso N 1987 Purified bovine AMH induces a characteristic freemartin effect in fetal rat prospective ovaries exposed to it in vitro. Development 100:43–55 27. Josso N, Racine C, di Clemente N, Rey R, Xavier F 1998 The role of antiMullerian hormone in gonadal development. Mol Cell Endocrinol 145:3–7 28. 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 29. Mauduit C, Jaspar JM, Poncelet E, Charlet C, Revol A, Franchimont P, Benahmed M 1993 Tumor necrosis factor ␣ antagonizes follicle-stimulating hormone action in cultured Sertoli cells. Endocrinology 133:69 –76 30. Seboun E, Barbaux S, Bourgeon T, Nishi S, Agulnik A, Egashira M, Nikkawa N, Bishop C, Fellous M, McElreavey K, Kasahara M 1997 Gene sequence, localization, and evolutionary conservation of DAZLA, a candidate male sterility gene. Genomic 41:227–235 31. Hirose T, O’Brien DA, Jetten AM 1995 Cloning of the gene encoding the murine orphan receptor TAK1 and cell-type-specific expression in testis. Gene 163:239 –242 32. Faed MJ, Lamont MA, Baxby K 1982 Cytogenetic and histological studies of testicular biopsies from subfertile men with chromosome anormaly. J Med Genet 19:49 –56 33. Matsuda T, Hayashi K, Nonmura M, Yamamoto S, Yoshida O 1989 Azoospermic male with a balanced Y-autosome translocation. Urol Int 44:43– 46 34. Guichaoua MR, Speed RM, Luciani JM, Delafontaine D, Chandley AC 1992 Infertility in human males with autosomal translocations. II. Meiotic studies in three reciprocal rearrangements, one showing tertiary monosomy in a 45-chromosome individual and his father. Cytogenet Cell Genet 60:96 –101