| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Testis |
a Population Council, New York, New York 10021
b Department of Zoology, University of Hong Kong, Hong Kong, China
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-catenin, ß-catenin, and p120ctn by semiquantitative reverse transcription-polymerase chain reaction and immunoblotting. Furthermore, the assembly of AJs between Sertoli and germ cells was associated with a transient induction in the steady-state mRNA and protein levels of cadherins and catenins. These analyses reveal, to our knowledge for the first time, that the testis may indeed be using the cadherin/catenin complex as one of the functional units to regulate AJ dynamics between Sertoli and germ cells in addition to 6ß1 integrin and the nectin/afadin complex. To further confirm the existence of such a complex between Sertoli and germ cells, immunoprecipitation experiments were performed using Sertoli-germ cell lysates during AJ assembly. An anti-N-cadherin antibody can pull out ß-catenin, whereas N-cadherin can also be pulled out using an anti-ß-catenin antibody. To further expand and validate these in vitro biochemical studies, immunofluorescent histochemistry was performed, which colocalized N-cadherin and ß-catenin to the same site of Sertoli-Sertoli and Sertoli-germ cell AJs, possibly ectoplasmic specializations near the basal compartment, at the lower third of the seminiferous epithelium in vivo as well as between Sertoli cells cultured in vitro. Furthermore, studies by cross-linking using dithiobis(succinimidylpropionate) confirmed that the cadherin/catenin complex between Sertoli cells as well as between Sertoli and germ cells indeed structurally linked to actin but not to vimentin (an intermediate filament protein) or to tubulin (a microtubule protein). These results thus unequivocally demonstrate that the cadherin/catenin complex, which can be up-regulated by testosterone, is indeed present between Sertoli and germ cells and is used for the assembly of functional AJs.
Sertoli cells, spermatogenesis, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2-macroglobulin [6, 7]; cytokines, such as transforming growth factor (TGF) ß2 and TGFß3 [5]; and kinases and phosphatases, such as myotubularin [8–10]. For instance, protein tyrosine phosphatase inhibitors, protein tyrosine kinase inhibitors, and recombinant TGFß3 were all shown to perturb the Sertoli cell TJ-permeability barrier in vitro [5, 8, 10]. However, the precise mechanism by which the dynamics of AJ, a cell-cell anchoring (or adhering) junction type utilizing actin filaments as its attachment site, are regulated is entirely unknown. Furthermore, morphological studies have shown that the testis is equipped with a modified type of AJ, called the ectoplasmic specializations (ES), which is found between Sertoli cells as well as between Sertoli cells and spermatids/spermatocytes (for review, see [11]). The ES is constituted largely by 6ß1 integrin [12] and, possibly, by nectin/afadin (for review, see [1]) (see also Fig. 1). Whether the cadherin/catenin complex contributes to the making of ES in the testis is not yet certain. Furthermore, two other recent reports using immunofluorescent microscopy have failed to colocalize the cadherin/catenin complex with actin in the testis, implicating that this complex may be linked to the cytoskeleton network via vimentin-based intermediate filaments [12, 13]. Considered collectively, these results seemingly suggest that the testis is using the classical AJ-integral membrane proteins, such as cadherins, instead of the desmosomal cadherins, such as desmogleins and desmocollins (for review, see [1]), for constituting desmosome-like anchoring junctions between Sertoli cells and germ cells in the seminiferous epithelium.
|
In other mammalian nongonadal epithelia, the functional unit of AJs between neighboring cells is either the cadherin/catenin complex (for reviews, see [1, 14–16) or nectin/afadin/ponsin complex (for reviews, see [1, 17]). The dynamics of AJs in turn are regulated by a number of AJ-signaling molecules, such as c-Src, Csk (for reviews, see [1, 18, 19]), and CK2 [20], via changes in the phosphorylation status of the cadherin/catenin and nectin/afadin complexes (for review, see [1]). Yet, some controversy still exists regarding whether the testis is utilizing the cadherin/catenin complex as a functional unit to regulate AJ dynamics and whether this unit links to the actin network. For instance, an immunohistochemistry study failed to localize N-cadherin between Sertoli cells and late spermatids [21]. Other studies also failed to colocalize ß-catenin with N-cadherin at the adluminal ES [13, 22], and -catenin was not detected at the sites of Sertoli-germ cell contact [23]. Instead, - and ß-catenins were found largely to be associated with Leydig cells and the Sertoli cell junctional complex [12, 23]. Moreover, the presence of E-cadherin in the rat testis as determined by immunohistochemistry is also the subject of controversy, even though E-cadherin is the principal cell adhesion molecule in the actin-based cell-cell AJs in epithelial cells [24, 25]. For instance, E-cadherin was not detected in the rat testis [26], but three independent studies using reverse transcription-polymerase chain reaction (RT-PCR) and immunoblotting techniques have demonstrated its presence in the rat testis [27–29]. Furthermore, N-cadherin antibody was shown to precipitate ß-catenin and p120ctn in seminiferous tubule lysate [30]. However, it is difficult to conclude from this latter study if the cadherin/catenin complex indeed exists between Sertoli and germ cells instead of being restricted only to the Sertoli-Sertoli cell junctional complexes because of the source of tissue extracts that were used for immunoprecipitation. Nonetheless, in vitro binding experiments have repeatedly demonstrated that the binding of germ cells onto Sertoli cells requires cadherin, because an anticadherin antibody can perturb the binding of germ cells onto Sertoli cells, which is a prerequisite for the subsequent AJ assembly [31, 32]. Furthermore, recent studies using immunohistochemistry have demonstrated N-cadherin at the basal inter-Sertoli AJs and around the head of elongate spermatids [13, 33], consistent with its localization at ES, but ß-catenin could not be detected at the site surrounding the heads of elongate spermatids [22]. As such, this latter study is in contrast to another report [33] that both ß-catenin and cadherin colocalize around the heads of the elongate spermatids. Considering these findings collectively, they clearly illustrate a controversial issue: if the cadherin/catenin complex indeed is used by Sertoli and germ cells or regulates AJ dynamics.
In light of these somewhat conflicting reports, it is crucial to settle the issue regarding the presence of the cadherin/catenin complex in AJs between Sertoli and germ cells in vitro using highly purified testicular cells and whether this complex is linked to the actin cytoskeleton or the vimentin-based intermediate filament network. Our findings demonstrate that such a regulatory functional unit based on cadherin/catenin is, indeed, present between Sertoli and germ cells using actin as its attachment site. The apparently conflicting reports in the literature likely may be the result of differences in antibody specificity used by different investigators.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sprague-Dawley rats were purchased from Charles River Laboratories (Kingston, NY). The use of animals in this study was approved by the Rockefeller University Animal Care and Use Committee with protocol numbers 00111, 97117, and 95129-R.
Antibodies
Polyclonal antibodies used in the studies reported herein were raised in rabbits against the corresponding AJ target proteins of human (N-cadherin, E-cadherin, -catenin, ß-catenin, -tubulin, and actin), porcine (vimentin), or mouse (p120ctn) origin and were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). These included antibodies against N-cadherin (H-63; catalog no. sc-7939, lot no. C081), E-cadherin (H-108; catalog no. sc-7870, lot no. K080), -catenin (H-297; catalog no. sc-7894, lot no. E141), ß-catenin (H-102; catalog no. sc-7199, lot no. L060), p120ctn (S-19; catalog no. sc-1101, lot no. A079), actin (H-196; catalog no. sc-7210, lot no. C222), -tubulin (H-300; catalog no. sc-5546, lot no. D042), and vimentin (V9; catalog no. sc-6260, lot no. B252). A separate rat anti-mouse E-cadherin antibody (catalog no. 13-1900, lot no. 11168061) was obtained from Zymed (San Francisco, CA). For N-cadherin, E-cadherin, -catenin, ß-catenin, -tubulin, and actin, the source of antigen used for antibody production was derived from corresponding human recombinant protein; for p120ctn, the source was a peptide fragment derived from mouse p120ctn protein. All antibodies used in the present study cross-reacted with the corresponding rat protein as indicated by the manufacturer. Bovine anti-rabbit immunoglobulin (Ig) G or bovine anti-mouse IgG conjugated to horseradish peroxidase (catalog no. sc-2370, lot no. 1071) used as a secondary antibody was obtained from Santa Cruz Biotechnology.
Sertoli Cell Cultures
Sertoli cells were isolated from 20-day-old rats as previously described [34]. Freshly isolated Sertoli cells were resuspended in serum-free Ham F-12 nutrient mixture and Dulbecco modified Eagle medium (F12/DMEM, 1:1 [v/v]; Sigma, St. Louis, MO) supplemented with 15 mM Hepes, 1.2 gm/L of sodium bicarbonate, 10 µg/ml of bovine insulin, 5 µg/ml of human transferrin, 2.5 ng/ml of epidermal growth factor, 20 mg/L of gentamicin, and 10 µg/ml of bacitracin. These cells were plated at high cell density (0.5 x 106 cells/cm2) on Matrigel (Collaborative Research, Inc., Bedford, MA)-coated, 12-well dishes (effective area, 
3.8 cm2/well) with 3 ml of F12/DMEM per well. Sertoli cells were incubated at 35°C in a humidified atmosphere of 5% (v/v) CO2/95% (v/v) air. Cultures were designated as cultures at time 0 at the time of cell plating. Approximately 36 h after plating, Sertoli cells were hypotonically treated with 20 mM Tris (pH 7.4) for 2.5 min to lyse residual germ cells and then washed twice to remove cellular debris [35]. The resulting Sertoli cells had a purity of 95%. To study the effects of steroids on the expression of AJ target genes, cultures at Day 3 were incubated with desired concentrations of dihydrotestosterone (DHT) or testosterone for 6 h before termination. For coculture experiments, Sertoli cells were cultured for 5 days alone to allow the assembly of TJs and AJs, forming an epithelium [6, 34], before adding germ cells to initiate Sertoli-germ cell AJ assembly.
Germ Cell Isolation and Sertoli-Germ Cell Cocultures
Germ cells were isolated from 20- and 90-day-old rat testes as previously described [6, 36]. Freshly isolated germ cells were used within 3 h. For coculture experiments, germ cells isolated from 90-day-old rats were added onto the Sertoli cell epithelium, which had been cultured in vitro for 5 days alone at 0.5 x 106 cells/cm2 using a Sertoli:germ cell ratio of 1:1 to initiate AJ assembly as previously described [6]. Previous morphological studies have shown that specialized AJs, such as desmosome-like junctions and ES, were formed between Sertoli and germ cells within 24–48 h using this coculture system [37, 38], which had previously been characterized in our laboratory [6]. In selected experiments, the integrity of the Sertoli cell epithelium was assessed both by morphological analysis as previously described [6, 34] and by quantifying transepithelial electrical resistance across the cell epithelium to ensure the presence of a Sertoli cell TJ-permeability barrier [5]. These cocultures were designated as time 0 when freshly isolated germ cells were added to the Sertoli cell epithelium, and cocultures were incubated for up to 4 days. Cells were terminated at specified time points for RNA extraction. In selected experiments, cocultures were terminated by first removing the spent media, and the remaining Sertoli-germ cell cocultures in each well of the 12-well dishes were then lysed by adding 0.8 ml of lysis buffer A (0.125 M Tris, 1% [w/v] SDS, 1.2% [v/v] 2-mercaptoethanol, 1 mM EDTA, and 2 mM PMSF, pH 6.8 at 22°C) to obtain whole-cell lysates for immunoblotting. This buffer was also used for the preparation of lysates from testes and seminiferous tubules.
Assessment of Cell Purity
The purity of Sertoli and germ cell cultures used in studies described in this report was assessed by either one or a combination of the following procedures as detailed elsewhere [6, 7, 9, 36, 39, 40]. First, direct microscopic examination of germ and Sertoli cells was performed following acetone fixation and toluidine blue staining [9, 41]. Second, RT-PCR was used to detect 3ß-hydroxysteroid dehydrogenase (a Leydig cell marker) [42] and fibronectin (a myoid cell marker) [43] in either Sertoli or germ cell preparation to assess Leydig and peritubular myoid cell contamination. Germ cell contamination in Sertoli cell preparations was monitored by amplifying c-Kit receptor (a spermatogonium marker) [44, 45]; Sertoli cell contamination in germ cell preparations was assessed by using testin (a Sertoli cell marker) [46, 47]. These analyses revealed that the germ and Sertoli cell preparations used in the studies presented herein were contaminated with negligible numbers of other cell types.
Isolation of Seminiferous Tubules and Their Culture In Vitro
Seminiferous tubules were isolated from adult rat (300 g body weight) testes as previously described [48]. Briefly, tubules were isolated by enzymatic digestion using collagenase/dispase (0.05% [w/v]) treatment (25 min at 35°C). Interstitial cells were removed by extensive washing in F12/DMEM by sedimentation under gravity. Tubules were preincubated for 4–6 h and then washed twice in a 2-h interval to remove steroids released from tubules [48]. These tubules were then trimmed into 2-mm fragments and incubated for 36–48 h at 35°C in F12/DMEM with insulin (20 µg/ml), transferrin (20 µg/ml), gentamicin (100 µg/ml), and penicillin (100 IU/ml) in a final volume of one testis per 25 ml of F12/DMEM. Thereafter, these tubules were used for cross-linking experiments using dithiobis(succinimidylpropionate) (DSP; Pierce, Rockford, IL) (see below). The final tubule preparations were nonresponsive to hCG (10 ng/ml) treatment when the levels of testosterone in spent media were quantified, suggesting that they had negligible Leydig cell contamination.
Treatment of Rats with 1-(2,4-Dichlorobenzyl)-Indazole-3-Carbohydrazide
Adult rats (250–300 g body weight) were fed with a single dose of 1-(2,4-dichlorobenzyl)-indazole-3-carbohydrazide (AF-2364) at 300 mg/kg body weight, which is known to perturb Sertoli-germ cell AJs by inducing germ cell loss, particularly spermatids from the seminiferous epithelium, except for spermatogonia and primary spermatocytes [49, 50]. The AF-2364 was suspended in 0.25% (w/v, in sterile water) methylcellulose to a concentration of 50–100 mg/ml. Rats receiving vehicle alone (0.25% methylcellulose) were terminated at time 0 as controls. Testes were removed at specified time points with three to five rats per time point for RNA extraction. For protein extraction, testes were homogenized with a Tissumizer (Tekmar, Cincinnati, OH) using a tissue:lysis buffer A ratio of 1:3 (w/v).
Semiquantitative RT-PCR
Semiquantitative RT-PCR was performed as previously described [2, 41]. Briefly, 2 µg of total RNA were reverse transcribed into cDNAs using 1 µg of oligo(dT)15 with a Moloney murine leukemia virus RT kit (Promega Corp., Madison, WI) in a final reaction volume of 25 µl. Thereafter, each PCR reaction was performed by combining 2 µl of RT product, 0.6 µg each of sense and antisense primers of a target gene (coamplified with either rat ribosomal S-16 or rat ß-actin using 0.01 µg each of the sense and antisense primer) (Table 1), 5 µl of 10x PCR buffer, 3 µl of MgCl2 (25 mM), 8 µl of deoxy(d)-NTPs (200 µM each of dATP, dCTP, dGTP, and dTTP), 1.25 U of Taq DNA polymerase, and sterile water to a final reaction volume of 50 µl. The cycling parameters for PCR were as follows: denaturation at 94°C for 1 min, annealing at 46–63°C (see Table 1) for 2 min, and extension at 72°C for 3 min, for a total of 20–29 cycles, which was followed by a final extension of 15 min in a DNA Thermal Cycler (model 2400; Perkin-Elmer, Foster City, CA). To obtain data that permitted subsequent quantitative analysis using autoradiograms for densitometric scanning, hot PCR was performed by using 
-32P-labeled sense primers as previously described [2, 41]. The relative ratio of [32P]sense primer of a target gene to [32P]sense S-16 or ß-actin was the same as that of the unlabeled primers. To eliminate interassay variations, all samples from a single experiment were processed simultaneously for RNA extraction and RT-PCR. A master mix containing all the required components in a PCR reaction was prepared and dispensed to each PCR tube within a given experiment to eliminate pipetting errors. To ensure that production of the target gene and either S-16 or ß-actin were in their linear phase, preliminary experiments were performed using different concentrations of template cDNAs and primer pairs as well as different annealing temperatures. Production of both the target gene and the housekeeping gene, either S-16 or ß-actin, was in the linear phase, as demonstrated in preliminary experiments. Yet, because of the disparity between the endogenous levels of mRNAs encoding the target gene and either S-16 or ß-actin, the production of S-16/ß-actin was close to its saturation phase, whereas the linearity of the target gene, such as E-cadherin, was in its exponential phase. Autoradiography of PCR products was performed using Kodak X-OMAT AR films (Eastman Kodak, Rochester, NY). Autoradiograms were densitometrically scanned and normalized against S-16 or ß-actin.
|
Preparation of Cell Lysates for Analytical PAGE
At specified time points, media were aspirated from 12-well dishes, and the remaining Sertoli and germ cells were terminated by adding 0.8 ml of lysis buffer A to each well. Using an Ultrasonic Homogenizer (4710 Series; Cole-Parmer Instrument Co., Vernon Hills, IL), cells were sonicated twice for 15 sec, with a 30 sec interval during which samples were kept in ice, and vortexed to enhance protein solubilization. Cellular debris was pelleted by centrifugation at 12 000 x g for 15 min at 4°C. The clear supernatant was used as whole-cell lysate. Total protein concentration in the cell lysate was determined using BSA as a standard [51]. Approximately 150 µg of protein (50 µl) from each sample within a given experimental group was resolved by SDS-PAGE under reducing conditions [52, 53]. After electrophoresis, proteins on gels were electroblotted onto nitrocellulose membranes (0.45 µm; Schleicher and Schuell, Inc., Keene, NH) for immunoblot analyses [47] using a chemiluminescent-based enhanced chemiluminescence (ECL) system (Amersham Pharmacia Biotech, Piscataway, NJ). Nonspecific binding sites on the nitrocellulose membrane were blocked using 6% (w/v) nonfat dry milk (Nestle, Solon, OH) in PBS-Tris (10 mM Tris, 10 mM sodium phosphate, and 0.15 M NaCl, pH 7.4 at 22°C, containing 0.1% [v/v] Tween-20). Fluorescence on the nitrocellulose membrane was detected using Kodak BioMax Light Films. To examine a second target protein in the same blot, the initial primary and secondary antibodies were removed by incubating blots in stripping buffer (62.5 mM Tris-HCl, pH 6.7 at 22°C, containing 100 mM 2-mercaptoethanol and 2% [w/v] SDS) at 55°C for 30 min in a reciprocating water bath at 110 rpm. The blot was then blocked in milk and reprobed with different primary and secondary antibodies to detect a second target protein. To ensure equal loading of proteins between samples in a given experiment, blots were reprobed with an antibody against actin, a cytoskeletal protein.
Preparation of Membrane and Cytosolic Fractions to Assess the Relative Distribution of AJ-Associated Proteins During Sertoli-Germ Cell AJ Assembly In Vitro
Sertoli-germ cell cocultures in 12-well dishes were terminated at specified time points by first removing the spent media; the remaining cells in each well were then suspended in 0.5 ml of extraction buffer (10 mM NaH2PO4, 0.15 M NaCl, 2 mM PMSF, 2 mM N-ethylmaleimide, and 2 mM EDTA, pH 7.4 at 22°C). Samples were sonicated as described above, and cell membrane was pelleted at 15 000 x g for 10 min at 4°C. The supernatant was used as the cytosolic fraction. The pelleted membrane fraction was sonicated again in 0.5 ml of SDS sample buffer [52]. Proteins were resolved onto 7.5% (w/v) T SDS-PAGE under reducing conditions. Electroblotting and ECL detection of specific target proteins were performed as described above.
Immunoprecipitation of the Cadherin-Catenin Complexes in Sertoli-Germ Cell Cocultures Using Either an Anti-N-cadherin or Anti-ß-Catenin Antibody
Sertoli-germ cells cocultured in 12-well dishes or seminiferous tubular fragments were terminated at specified time points by first removing the spent media. Thereafter, ice-cold lysis buffer B (0.8 ml/well; 10 mM Tris, pH 7.4 at 22°C, containing 0.15 M NaCl, 2 mM PMSF, 2 mM EDTA, 2 mM N-ethylmaleimide, 1% [v/v] NP-40, 1 mM sodium orthovanadate, 0.1 µM sodium okadate, and 10% [v/v] glycerol) was added to the remaining cells. Cells were sonicated twice for 15 sec with a 30-sec interval. Whole-cell lysates were obtained by centrifugation at 12 000 x g for 15 min at 4°C to remove cellular debris. Equal amounts of proteins (400 µg) in 100 µl from Sertoli-germ cell lysates were incubated with either 2 µg of anti-N-cadherin or anti-ß-catenin antibody for 4 h at 4°C with agitation. Thereafter, 20 µl of a Protein A/G PLUS-agarose (Santa Cruz Biotechnology) suspension was added to each sample for immunoprecipitation by a second incubation of 4 h at 4°C. Immunoprecipitates were washed four times with the above lysis buffer by resuspension and centrifugation (1000 x g, 5 min each). Fifty microliters of SDS sample buffer (0.125 M Tris, pH 6.8 at 22°C, containing 20% [v/v] glycerol, 1% [w/v] SDS, and 1.6% [v/v] 2-mercaptoethanol) were added to each sample tube, which was heated at 100°C for 10 min to extract the bound target proteins. After removing the Protein A/G PLUS-agarose by a brief centrifugation at room temperature, supernatant was resolved by SDS-PAGE under reducing conditions. Whole-cell lysates without incubation with antibodies and the supernatant derived from the above immunoprecipitation step were used as controls.
Immunofluorescent Microscopy for Colocalization of N-Cadherin and ß-Catenin in the Rat Testis In Vivo and Sertoli Cells Cultured In Vitro
Immunofluorescent microscopy was performed essentially as previously described for testin from this laboratory [54], except that double-fluorescent probes, namely fluorescein isothiocyanate (FITC) and Cy3, were used to visualize both N-cadherin and ß-catenin in the same tissue sections or cells cultured in vitro. Briefly, testes removed from adult rats were frozen in liquid nitrogen and embedded in OCT compound (Miles Scientific, Elkhart, IN). Frozen sections of testis (thickness, 8 µm) were cut at -20°C with a cryostat and mounted on poly-L-lysine (Mr, 150 kDa)-coated glass slides. Sections were air-dried at room temperature and fixed in modified Bouin solution (4% [v/v] formaldehyde in picric acid) for 5 min. Sections or cells were then treated with 3% (w/v) H2O2 in methanol to block endogenous peroxidase activity, followed by 10% nonimmune goat serum to minimize nonspecific antibody binding as previously described [55, 56]. Sections were then incubated with a mouse anti-N-cadherin antibody (catalog no. 33-3900, lot no. 11268187; Zymed) and followed by a goat-anti-mouse IgG-Cy3 (catalog no. 81-6515, lot no. 11067429; Zymed). Thereafter, sections were washed in PBS (10 mM sodium phosphate and 0.15 M NaCl, pH 7.4 at 22°C) and incubated with a rabbit anti-ß-catenin antibody (catalog no. 71-2700, lot no. 11067640; Zymed) and followed by a goat anti-rabbit IgG-FITC (catalog no. 62-6111, lot no. 10665523; Zymed). Sections were then mounted in Vectashield (Vector Laboratories, Burlingame, CA), and fluorescent microscopy was performed using an Olympus BX40 microscope equipped with Olympus UPlanF1 fluorescent optics (Melville, NY). For Sertoli cells cultured in vitro at 0.5 x 106 cells/cm2 on Matrigel-coated, poly-L-lysine-coated slides, cells were fixed in Bouin solution on Day 5, and immunofluorescent microscopy was performed as detailed above. All images were digitally acquired and analyzed in Adobe Photoshop (version 7.0) (San Jose, CA) in a Compaq SP700 Workstation (Houston, TX) running under Windows XP (Microsoft, Redmond, WA).
Cross-Linking of Proteins in Sertoli Cell Cultures, Sertoli-Germ Cell Cocultures, and Seminiferous Tubules Using DSP
Cross-linking of the cadherin-catenin complex and actin in Sertoli cell cultures, Sertoli-germ cell cocultures, and seminiferous tubule cultures was performed by using DSP essentially as previously described [57]. A thio-cleavable, membrane-permeable, homobifunctional, and amine-reactive cross-linker, DSP can gain access to cytoplasmic proteins. Here, DSP was dissolved in dimethyl sulfoxide as a stock solution of 25 mM. Sertoli cells were cultured alone at 0.75 x 106 cells/cm2 for 7 days. For Sertoli-germ cell cocultures, Sertoli cells were cultured alone for 6 days, forming a cell epithelium; thereafter, freshly isolated germ cells from adult rat testes were isolated and cocultured with Sertoli cells for 3 days. On Day 7 and Day 3 of the Sertoli cell cultures and Sertoli-germ cell cocultures, respectively, and on Day 2 of the seminiferous tubule cultures, media were removed and replaced with PBS containing 2.5 mM DSP to initiate cross-linking. Reaction was terminated 15–30 min thereafter by adding lysis buffer B (see above) for immunoprecipitation using anti-N-cadherin antibody. Proteins were extracted from the Protein A/G-PLUS agarose with or without 2-mercaptoethanol and resolved by SDS-PAGE for immunoblotting using either anti-N-cadherin, antiactin, anti--tubulin (a constituent protein of microtubule), or antivimentin (a constituent protein of intermediate filament) antibodies. Using DSP, all proteins associated with the cadherin-catenin complex, both extracellular, transmembrane, and intracellular, were cross-linked and immunoprecipitated by the anti-N-cadherin antibody. In selected experiments, immunoprecipitation was performed using an antivimentin antibody (1:200 [v/v] dilution), and the immunocomplexes were visualized by either an antivimentin or anti-N-cadherin antibody after SDS-PAGE.
Statistical Analysis
Statistical analyses were performed using the GB-STAT Statistical Analysis Software Package (version 7.0; Dynamic Microsystems, Inc., Silver Spring, MD).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Morphological analysis The purity of Sertoli and germ cells used for the studies described in the present report is shown in Figure 2, A–C. Figure 2A illustrates that Sertoli cell cultures were contaminated with negligible germ cells. Likewise, Figure 2, B and C, illustrates negligible somatic cell contamination in germ cells isolated from 20- and 90-day-old rats.
|
RT-PCR Analysis Germ cell preparations used in our studies had negligible Sertoli cell contamination, because protein blotting and RT-PCR failed to detect testin, a putative Sertoli cell protein [47] (Fig. 2, D and E) in these cultures. Likewise, c-Kit receptor, a spermatogonium marker [44, 45], was only detected in germ cell cultures and testes and not in Sertoli cell cultures (Fig. 2E), indicating that germ cell contamination in Sertoli cell cultures, if any, was negligible. Both cell culture preparations had negligible Leydig and peritubular myoid cell contamination, because no 3ß-hydroxysteroid dehydrogenase, a Leydig cell marker [42], and fibronectin, a myoid cell marker [43], could be detected. These analyses thus unequivocally demonstrate the purity of the cell cultures used in our studies.
Differential Expression of N-Cadherin, E-Cadherin, -Catenin, ß-Catenin, and p120ctn in Sertoli and Germ Cells
To assess whether Sertoli and germ cells were equipped with the AJ-associated protein components, such as N-cadherin, E-cadherin, -catenin, ß-catenin, and p120ctn, their steady-state mRNA and protein levels were assessed in 20-day-old Sertoli and 20- and 90-day-old germ cells. Sertoli cells were cultured at 0.5 x 106 cells/cm2 on Matrigel-coated dishes for 3 days before termination so that contaminating germ cells were removed on Day 2 by hypotonic treatment. Germ cells were used within 3 h after their isolation, because >90% of the germ cells became nonviable within 18 h in vitro [39]. When total RNA isolated from these samples was analyzed by semiquantitative RT-PCR, it was found that both Sertoli and germ cells expressed N-cadherin, E-cadherin, -catenin, ß-catenin, and p120ctn (Fig. 3). The steady-state mRNA level of N-cadherin (Fig. 3, A and G) and p120ctn (Fig. 3, E and G) in Sertoli cells was 2-fold higher than that of germ cells. The -catenin (Fig. 3, C and G) and ß-catenin (Fig. 3, D and G) mRNA level between Sertoli and germ cells was similar. However, the steady-state mRNA level of E-cadherin in germ cells was 3-fold higher than that of Sertoli cells (Fig. 3, B and G). The presence of cadherins and catenins in Sertoli and germ cells was also confirmed by immunoblottings using corresponding antibodies, with actin as a loading control (Fig. 3, F and G).
|
A possible critique to these results is that Sertoli and germ cells may not necessarily have the same level of ß-actin and/or S-16, which in turn would invalidate our conclusion. To address this concern, immunoblotting was performed using an equivalent amount of total proteins extracted from Sertoli and germ cells and visualized by the corresponding antibodies. Results of this analysis, shown in Figure 3F (right panel), illustrate that the relative abundance of different cadherins and catenins between Sertoli and germ cells was, indeed, similar to the semiquantitative RT-PCR results shown in Figure 3, A–E (see Fig. 3G). The observations also confirm that germ cells are, indeed, equipped with the needed machineries to assemble the cadherin/catenin complex (see Fig. 1).
Transient Induction in Both Steady-State mRNA and Protein Levels of N-Cadherin, E-Cadherin, -Catenin, ß-Catenin, and p120ctn During Sertoli-Germ Cell AJ Assembly In Vitro
To examine the participation of different AJ-associated proteins during Sertoli-germ cell AJ assembly in vitro, Sertoli cells isolated from 20-day-old rat testes were cultured alone at 0.5 x 106 cells/cm2 on Matrigel-coated dishes for 5 days, forming an epithelium with intact TJs and AJs. On Day 5, germ cells freshly isolated from 90-day-old rats were added onto this Sertoli cell epithelium at a Sertoli:germ cell ratio of 1:1 to initiate Sertoli-germ cell AJ assembly. Cultures were terminated at specified time points for semiquantitative RT-PCR (Fig. 4) and immunoblotting (Fig. 5). A transient induction was observed in the mRNA levels of N-cadherin (Fig. 4, A and F), E-cadherin (Fig. 4, B and F), -catenin (Fig. 4, C and F), ß-catenin (Fig. 4, D and F), and p120ctn (Fig. 4, E and F) during AJ assembly between Sertoli and germ cells in vitro. These same changes were detected by immunoblotting when cell lysates from parallel cocultures were used (Fig. 5), confirming the RT-PCR results (Fig. 4). To verify that the induction of gene expression shown in Figure 4, A–E, was, indeed, contributed by Sertoli and germ cells during AJ assembly rather than the result of endogenous changes, Sertoli cells cultured alone, without germ cells, were terminated at selected points for RT-PCR and immunoblottings (Figs. 4G and 5G). The loading control for the coculture immunoblots was shown in Figure 5F. These results thus unequivocally demonstrate a transient induction in cadherin and catenin at both the mRNA and protein levels, suggesting that a cadherin/catenin complex might, indeed, exist between Sertoli and germ cells.
|
|
Relative Localization of AJ-Associated Proteins During Sertoli-Germ Cell AJ Assembly In Vitro
To correlate the cellular distribution and induction of proteins of cadherins and catenins during Sertoli-germ cell AJ assembly in vitro, membrane and cytosolic fractions were isolated from these cells in the Sertoli-germ cell cocultures at specified time points. It was shown that N-cadherin (Fig. 6, A and F) and E-cadherin (Fig. 6, B and F) were associated exclusively with the membranes, whereas approximately one-third to one-fourth of -catenin (Fig. 6, C and F), ß-catenins (Fig. 6, D and F), and p120ctn (Fig. 6, E and F) were found in the cytosol, with the remainder associated with the cell membrane during AJ assembly in vitro. These results are consistent with the fact that cadherins are AJ-membrane integral proteins. Whereas the majority of catenins were associated with the cadherin forming the cadherin/catenin complex, a small pool of free catenins were found in the cytosol (for reviews, see [1, 58, 59]). However, catenins that are loosely associated with the cadherin/catenin complex might become dissociated during the extraction procedures in the immunoprecipitation experiments because of the presence of NP-40, which is an nonionic detergent. Yet, for the membrane-associated cadherins and catenins, a similar transient induction in protein level was also detected (Fig. 6) during AJ assembly in vitro, which is consistent with those results obtained by semiquantitative RT-PCR (Fig. 4) and immunoblotting (Fig. 5). This transient induction in protein level, however, was not detected in the cytosolic catenins, which is consistent with earlier studies showing that the small pool of catenins in the cytosol may not take part in AJ assembly and, instead, regulates AJ dynamics (for reviews, see [1, 58, 59]).
|
N-Cadherin/ß-Catenin/-Catenin Complex Is a Functional AJ Unit as Demonstrated by Immunoprecipitation Using Either an Anti-N-Cadherin or an Anti-ß-Catenin Antibody
The literature contains conflicting reports with regard to the existence of cadherin/catenin complexes in Sertoli-germ cell AJs as determined by immunohistochemistry [23, 40, 43]. To resolve this controversy, a biochemical approach using immunoprecipitation was used. Immunoprecipitation was performed using an anti-N-cadherin antibody to pull out its associated proteins in whole-cell lysates prepared from Sertoli-germ cell cocultures at specified time points. It was found that N-cadherin was associated with -catenin and ß-catenin but not with p120ctn (Fig. 7A). Similar results were obtained when an anti-ß-catenin antibody was used instead (Fig. 7B). Whole-cell lysates without incubation with any antibody from selected time points and supernatant after immunoprecipitation with an anti-ß-catenin antibody were used as controls (Fig. 7C). For instance, the presence of an anti-ß-catenin antibody immunoprecipitated virtually all N-cadherin, ß-catenin, and -catenin from the lysates, but not p120ctn (Fig. 7C). These results thus demonstrate unequivocally that the N-cadherin/ß-catenin/-catenin complex is the functional AJ unit between Sertoli and germ cells in the rat testis.
|
Colocalization of N-Cadherin and ß-Catenin to the Seminiferous Epithelium in the Adult Rat Testes and to the Cell-Cell Contact Sites Between Sertoli Cells In Vitro
To expand and confirm the above in vitro studies, immunofluorescent microscopy was used to colocalize N-cadherin and ß-catenin in the adult rat testis (Fig. 8, A–C). It was shown that N-cadherin indeed colocalized with ß-catenin in the seminiferous epithelium near the basal and the lower one-third of the adluminal compartment (Fig. 8, C vs. A and B). Furthermore, both N-cadherin and ß-catenin were colocalized at the site of cell-cell contacts between Sertoli cells and spermatocytes (Fig. 8C, arrowheads). The arrowheads in Figure 8C indicate the site of germ cell nuclei, which were not stained. We did not attempt to perform a similar immunofluorescent microscopy study on E-cadherin, because two anti-E-cadherin antibodies from two different vendors had shown that these antibodies cross-reacted with several low Mr proteins, making them unsuitable for immunohistochemistry (data not shown). Considering these results collectively, cadherin/catenin is a putative AJ unit in the testis, particularly between Sertoli and germ cells. Furthermore, studies using immunofluorescent microscopy also confirm the colocalization of N-cadherin and ß-catenin to the cell-cell contact sites between Sertoli cells in vitro (Fig. 8, D–F), which is consistent with in vivo results shown in Figure 8, A–C. The results shown in Figure 8, D–F, are also consistent with earlier studies [37, 38] that showed functional AJ structures, such as ES and desmosomes, are formed when testicular cells are cultured in vitro.
|
Cadherin/Catenin Complex Is Linked to the Actin Network, but Not to the Microtubule or Intermediate Filament Network, in Sertoli Cell, Sertoli-Germ Cell Cocultures, and Isolated Seminiferous Tubules
Because two reports in the literature [12, 13] suggest that the cadherin x catenin complex may be an intermediate filament-based anchoring junction based on immunohistochemical colocalization studies, we investigated this issue by using cross-linking and immunoprecipitation techniques. We chose to use Sertoli cell cultures and Sertoli-germ cell cocultures in our initial studies instead of whole tubules or testes, because these latter tissues could be contaminated with AJs found in Leydig cells, myoid cells, and possibly, fibroblasts. Cross-linking experiments using DSP were performed to investigate whether the cadherin/catenin complex links to the actin cytoskeleton network using either Sertoli cell cultures on Day 7 or Sertoli-germ cell cocultures on Day 3. Cell lysates were prepared for both the Sertoli cell cultures and Sertoli-germ cell cocultures after cross-linking with DSP and were subjected to immunoprecipitation using an anti-N-cadherin antibody. Because results using Sertoli cell cultures and Sertoli-germ cell cocultures were identical, only data obtained from Sertoli cell cultures are shown herein. It was noted that DSP, a thio-cleavable and membrane-permeable cross-linker [60], cross-linked the cadherin/catenin and its associated proteins, forming a complex >200 kDa on a 10% (w/v) T SDS-polyacrylamide gel when the immunocomplex was visualized by either an antiactin or anti-N-cadherin antibody (Fig. 9A). However, in the presence of 2-mercaptoethanol, the cross-linked complex was cleaved into individual component proteins, and both actin and N-cadherin were detected using the corresponding antibodies (Fig. 9A). When these blots were stained with either an anti--tubulin or antivimentin antibody, the immunocomplexes pulled down by the anti-N-cadherin antibody did not contain any -tubulin or vimentin (Fig. 9A). These biochemical analyses thus unequivocally demonstrate that the cadherin/catenin complex in the testis is an actin-based AJ unit and is not likely associated with the microtubule (e.g., -tubulin) and intermediate filament (e.g., vimentin) network, which is consistent with several earlier reports that ES is an actin-based AJ type [61–63].
|
To further expand and confirm the above studies, we also used isolated seminiferous tubules initially cross-linked with DSP, and the lysates were then immunoprecipitated with an anti-N-cadherin antibody. When the immunocomplexes were examined under reducing (with 2-mercaptoethanol) and nonreducing (without 2-mercaptoethanol) conditions, it was again shown that the N-cadherin immunoprecipitated complexes associated with actin (Fig. 9B, left) but not with vimentin (Fig. 9B, right), which is consistent with results obtained using Sertoli cells or Sertoli-germ cells cultured in vitro (Fig. 9, B vs. A). The middle panel in Figure 9B is the positive control stained with anti-N-cadherin. To vigorously validate the data shown in Fig. 9, A and B, that the cadherin/catenin complex is not associated with the intermediate filament network, Sertoli cells or tubules cross-linked with DSP were also immunoprecipitated with an antivimentin antibody. When the complexes were visualized by an antivimentin antibody (Fig. 9C, left), vimentin (55-kDa protein) was detected in both testis lysate (control, with neither cross-linking using DSP nor immunoprecipitation using antivimentin) and in Sertoli cells and seminiferous tubules (both cross-linked with DSP and immunoprecipitated with antivimentin) (Fig. 9C). Yet, when the same blot was visualized by an anti-N-cadherin antibody (Fig. 9C, right), N-cadherin was detected only in testicular lysates (positive control) and not in Sertoli cells or tubules subjected to DSP cross-linking and immunoprecipitation using an antivimentin antibody. Considering these results collectively, the cadherin/catenin complex in the testis is an actin-based AJ structure.
Effects on the Expression and Levels of AJ-Associated Proteins in the Rat Testis During AJ Disruption by Treatment of Adult Rats with AF-2364, a Male Contraceptive that Induces Germ Cell Loss from the Seminiferous Epithelium In Vivo
Adult rats weighing between 250 and 300 g were treated with one dose of AF-2364 (300 mg/kg body weight) by gavage [49, 50], and testes were removed at specified time points for RNA and protein extraction. An induction in the mRNA level of N-cadherin (Fig. 10, A and F), E-cadherin (Fig. 10, B and F), -catenin (Fig. 10, C and F), ß-catenin (Fig. 10, D and F), and p120ctn (Fig. 10, E and F) was detected after AF-2364 treatment at the time before any germ cell loss was visible in the testis when examined histologically [49, 50]. This induction in mRNA level was further confirmed by immunoblot analysis (Fig. 11) using lysates from testes with the corresponding antibodies.
|
|
Changes in the Steady-State mRNA Levels of Cadherins and Catenins in Sertoli Cell Cultures after Treatment with Testosterone or DHT
A previous study has shown that testosterone induces the expression of TJ proteins, such as occludin [2], implicating the effect of testosterone on TJ dynamics via its effect on TJ-associated proteins. Also, androgens can facilitate TJ assembly in Sertoli cell cultures in vitro [2, 64]. Another active androgen in the male reproductive system is DHT [65]. Thus, we sought to examine whether testosterone and/or DHT would have any effects on the gene expression of different AJ-associated proteins. Testosterone and DHT at 10-11 to 10-5 M were added to Sertoli cell cultures (0.5 x 106 cells/cm2) on Day 3 (i.e., 24 h after hypotonic treatment) and incubated for an additional 6 h before their termination for RNA extraction. A testosterone-induced increase was observed in the mRNA levels of N-cadherin (Fig. 12, A and F), E-cadherin (Fig. 12, B and F), -catenin (Fig. 12, C and F), and p120ctn (Fig. 12, E and F) but not in the mRNA level of ß-catenin (Fig. 12, D and F). Similar induction in the mRNAs of N-cadherin (Fig. 13, A and B) and E-cadherin (Fig. 13, C and D) were detected in DHT-treated Sertoli cell cultures.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-catenin physically interacting with the catenin-binding domain near the C-terminus of cadherin (see Fig. 1). This complex in turn links to the actin cytoskeleton network via -catenin as the linker protein (for reviews, see [1, 14, 15]). Other AJ-associated signaling molecules, such as Src, CK2, and Csk, regulate the functionality of the cadherin/catenin complex by inducing phosphorylation of the complex, changing its structural conformation, which then triggers changes in the underlying actin cytoskeleton, possibly via activation of small GTPases (for reviews, see [1, 69]). The net result will either be the disassembly or reassembly of AJs. Despite extensive studies investigating the role of the cadherin/catenin complexes in AJ dynamics, the presence of cadherins and their structurally and functionally associated proteins was not known in the testis until the early 1990s [23, 26]. Since then, studies investigating their roles in endocrine and reproductive tissues have proliferated (for review, see [70]). For instance, a recent study has demonstrated the presence of at least 24 cadherins in the rat testis using degenerate PCR cloning technique [27], implicating the importance of cadherins in testicular functions, such as spermatogenesis. Needless to say, extensive junction-restructuring events in the testis involve assembly and disassembly of AJs to permit the movement of developing germ cells across the seminiferous epithelium during spermatogenesis, and an intricate mechanism that regulates these events must exist. Yet, few studies have been performed to examine these events, let alone the underlying regulatory mechanism. Furthermore, the actin-based AJs found between Sertoli and germ cells, particularly spermatids (i.e., ES), are unique to the testis when examined by conventional and freeze-fracture electron microscopy (for reviews, see [11, 71]. These morphological observations seemingly suggest that the testis may be employing a different functional unit to regulate AJ functionality between Sertoli and germ cells. Indeed, transgenic Caenorhabditis elegans, either without or having significant reduction of the cadherin/catenin system, apparently maintain normal AJ functionality and structure [72]. However, E-cadherin mice died at the embryo stage around the time of implantation [73]. Also, N-cadherin-/- mice also died at the early embryo stage [74], with severe cardiac defects [75]. Considered collectively, these data suggest the complexity of AJ regulation and the crucial function of the cad
