| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Population Council, New York, New York, 10021
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, formerly called AF-2364, 1-(2,4-dichlorobenzyl)-IH-indazole-3-carbohydrazide], the event of AJ disruption was also associated with a transient iNOS (inducible nitric oxide synthase, NOS2) induction. Immunohistochemistry has illustrated that NOS2 was intensely accumulated in Sertoli and germ cells in the epithelium during adjudin-induced germ cell loss, with a concomitant accumulation of intracellular cGMP and an induction of PRKG but not cAMP or protein kinase A (cAMP-dependent protein kinase, PRKA). To identify the NOS-mediated downstream signaling partners, coimmunoprecipitation was used to demonstrate that NOS2 and eNOS (endothelial nitric oxide synthase, NOS3) were structurally associated with the N-cadherin (CDH2)/ß-catenin (CATNB)/actin complex but not the nectin-3 (poliovirus receptor-related 3, PVRL 3)/afadin (myeloid/lymphoid or mixed lineage-leukemia tranlocation to 4 homolog, MLLT4) nor the integrin ß1 (ITB1)-mediated protein complexes, illustrating the spatial vicinity of NOS with selected AJ-protein complexes. Interestingly, CDH2 and CATNB were shown to dissociate from NOS during the adjudin-mediated AJ disruption, implicating the CDH2/CATNB protein complex is the likely downstream target of the NO signaling. Furthermore, PRKG, the downstream signaling protein of NOS, was shown to interact with CATNB in the rat testis. Perhaps the most important of all, pretreatment of testes with KT5823, a specific PRKG inhibitor, can indeed delay the adjudin-induced germ cell loss, further validating NOS/NO regulates Sertoli-germ cell AJ dynamics via the cGMP/PRKG pathway. These results illustrate that the CDH2/CATNB-mediated adhesion function in the testis is regulated, at least in part, via the NOS/cGMP/PRKG/CATNB pathway.
adherens junction, anchoring junction, ß-catenin, N-cadherin, nitric oxide, nitric oxide synthase, spermatogenesis, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the testis, adhesion between Sertoli cells and spermatids is conferred by cell-cell actin-based adherens junctions (AJs), which are mediated via three protein complexes: the cadherin (CDH)/catenin (CATN), the nectin (PVRL)/afadin (MLLT4)/ponsin, and the 
6ß1-integrin/ laminin 
3 (ITGA6B1/LAMC3) complexes. These protein complexes, in turn, interact with other peripheral proteins and adaptors to form three functional protein complexes [for reviews, see 7–9]. Unlike tight junctions (TJs), which are restricted between Sertoli cells at the basal compartment of the seminiferous epithelium [for reviews, see 7, 10], AJs are found at the Sertoli-Sertoli and Sertoli-germ cell interface in the epithelium from the basal to the adluminal compartment. In the seminiferous epithelium, the two best studied cell-cell actin-based AJ types that are specific to the testis are the ectoplasmic specialization (ES) [9, 11–15] and the tubulobulbar complex (TBC) [16]. The ES found between Sertoli cells at the blood-testis barrier (BTB) is known as the basal ES, which, coexisting with TJs, TBC, and desmosome-like junctions, constitute the BTB; whereas ES found between Sertoli cells and round/elongating/elongate spermatids is known as apical ES and is restricted to the adluminal compartment of the seminiferous epithelium [for reviews, see 7–11]. Recent studies have shown that ES dynamics are regulated by kinases, phosphatases, and GTPases [17–21; for reviews, see 7, 22]. However, it is not known if NOS and NO play any role in AJ regulation in the testis, yet both molecules have been the focus of numerous studies involving junction restructuring in other epithelia. For instance, NOS3 was found to colocalize with platelet endothelial cell adhesion molecule 1 (PECAM1), an integral membrane protein, in endothelial cells [23]. Furthermore, NO had been shown to regulate the actin cytoskeleton via the guanylate cyclase (GUCY)/cGMP/protein kinase G (PRKG) pathway in human cervical epithelial cells [24]. Collectively, these studies have illustrated that NOS and NO are crucial regulators of junction-related events. More important, recent studies from our laboratory have shown that NO produced from NOS is a crucial regulator of Sertoli cell TJ dynamics in vitro [25]. Yet it is not entirely known if NOS plays any role in Sertoli-germ cell AJ dynamics in the testis and, if it does, the downstream signaling pathway(s) that is (are) being used remains uncharacterized. Because the regulation of Sertoli-germ cell adhesion function during restructuring of the seminiferous epithelium at spermatogenesis is an area of interest to reproductive biologists, the following studies were conducted using both an in vitro model of Sertoli-germ cell cocultures and an in vivo model based on adjudin [a molecule that induces , formerly called AF-2364, 1-(2,4-dichlorobenzyl)-IH-indazole-3-carbohydrazide]. The results of these studies have clearly illustrated the interesting role of NOS in Sertoli-germ cell AJ restructuring, in particular how it regulates the cell adhesion of the CDH2/CATNB/actin protein complex in the seminiferous epithelium.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sprague-Dawley (outbred) rats were obtained from Charles River Laboratories (Kingston, NY). The use of animals in this report was approved by the Rockefeller University Animal Care and Use Committee with Protocol Numbers 00111 and 03017. All animals were housed at the Rockefeller University Laboratory Animal Research Center with 12L:12D cycle with free access to standard chow and water.
Antibodies
The antibodies used in this study were raised either in rabbits, mice, or goats and derived from the corresponding human (NOS3, CDH2, CATNB, PVRL3, MLLT4, ITGB1, and actin) and mouse (NOS2) proteins. These include NOS2 (M-19; cat: sc-650; lot: J151), NOS3 (C-20; cat: sc-654; lot: K291), CDH2 (H-63; cat: sc-7939; lot: C081), CATNB (H-102; cat: sc-7199; lot: L060), PVRL3 (C-19; cat: sc-14806; lot: K261), MLLT4 (cat: 610732; lot: 2), ITGB1 (M-106; cat: sc-8978; lot: E221), and actin (H-196; cat: sc-7210; lot: C222). Bovine anti-rabbit IgG, bovine anti-mouse IgG, or bovine anti-goat IgG conjugated to horseradish peroxidase (HRP) was used as a secondary antibody for the corresponding primary antibody. Antibodies listed above were obtained from Santa Cruz Biotechnology (Santa Cruz, CA), except for MLLT4, which was obtained from BD Transduction Laboratories (San Diego, CA). Rabbit anti-PRKG (cat: 539729; lot: B49342) and rabbit anti-PRKA (cat: AB1613; lot: 23020786) were purchased from Calbiochem and Chemicon (Temecula, CA), respectively.
Primary Sertoli Cell Cultures
Sertoli cells were isolated from testes of 20-day-old rats and cultured in F12/DMEM as previously described [26]. Isolated Sertoli cells were plated at 0.5 x 106 cells/cm2 on Matrigel-coated dishes. On Day 2 (
36 h after plating), cells were hypotonically treated with 20 mM Tris, pH 7.4, for 2.5 min to lyse the contaminating germ cells [27], such that the resulting Sertoli cells had a purity of >95% [26]. Sertoli-Sertoli cell TJs formed within 2–3 days, creating an intact cell epithelium when characterized as earlier described [28–30] including the use of electron microscopy.
Sertoli-Germ Cell Cocultures
Cocultures were prepared essentially as earlier described [26, 31]. In brief, Sertoli cells isolated from testes of 20-day-old rats as described above were plated on Matrigel-coated 12-well dishes at 0.5 x 106 cells/ cm2 containing 3-ml of F12/DMEM and cultured alone for 5 days, forming an intact cell epithelium [25]. On Day 6, total germ cells, including spermatogonia, spermatocytes, and round/elongating/elongate spermatids, isolated from adult rat testes as described [32], were added onto the Sertoli cell epithelium to initiate AJ assembly. Cocultures were terminated at specified time points at 15, 30 min; 1, 2, 3, 6 h; and 1, 2, 3, and 4 days. Controls included Sertoli cells cultured alone without germ cell addition and terminated at the above specified time points.
Seminiferous Tubule Cultures
Seminiferous tubules were isolated from adult rat (300 g body weight [BW]) testes as previously described [26].
Treatment of Rats with adjudin to Disrupt AJs Between Sertoli and Germ Cells in the Seminiferous Epithelium
Adjudin was synthesized with a purity of greater than 99.98% as earlier described [33]. Adult rats (n = 3 for each time point) weighing between 250 and 300 g received either three doses of adjudin at 40 mg/kg BW, i.p., on Days 0, 7, and 14 (i.e., q 1-wk); or a single dose of adjudin at 50 mg/kg BW by gavage on Day 0, to induce germ cell loss from the seminiferous epithelium by perturbing AJs between Sertoli and germ cells as described [33, 34]. Using both regimens as reported herein, the time course of germ cell loss from the epithelium was virtually indistinguishable. For instance, the tubules (>98% of the tubules examined) were virtually devoid of germ cells by Day 7 (see Fig. 3D) regardless if the rats were treated with adjudin either at 40 mg/kg BW (i.p.) or 50 mg/kg BW (by gavage) on Day 0. However, in rats that received two subsequent doses of adjudin on Days 7 and 14, it took a significantly longer time for these rats to regain fertility (monitored by mating studies), which is consistent with 50 different regimens to assess the efficacy of adjudin during the past 10 yr in our laboratory [35]. The rationale of using these two different treatment regimens was to assess if the time course of germ cell loss from the epithelium and the changes in signaling molecules downstream of the NOS in response to adjudin were similar or if an administration via i.p. versus oral could reduce the efficacy of the drug possibly because of differences in metabolism. The slightly different regimens (i.e., 40 mg/kg BW via i.p. with three doses versus 50 mg/kg BW by gavage with one dose), which were shown herein that could both induce germ cell loss from the epithelium, were also used to assess if they could activate the signaling molecules of the NOS/cGMP/PRKG pathway. Furthermore, in the experiments reported in Figure 3, while the specific time points used for analyses by immunoblotting, reverse transcription-polymerase chain reaction (RT-PCR), and histology/immunohistochemistry were slightly different, they did cover the same time range (between Time 0 and Day 21) during germ cell loss from the epithelium. This is because the testes from these animals were either frozen in liquid nitrogen (for immunoblotting and immunohistochemistry) or homogenized in RNA-STAT-60 for RNA extraction at the time of their termination. It is also important to note that the trends of changes in protein and steady-state mRNA levels are consistent in both regimens.
|
Semiquantitative RT-PCR
Semiquantitative RT-PCR was performed essentially as previously described [25, 26]. PCR was performed using 2 µl RT products, 0.6 µg of each sense and antisense primer of a target gene, coamplified with the rat ribosomal S16 primer pairs, 0.01 µg each (Table 1). Each PCR reaction tube consisted of the above primer pairs, 5 µl of 10x PCR buffer, 3 µl MgCl2 (25 mM), 8 µl deoxy(d)-NTPs (200 µM each of dATP, dCTP, dGTP, and dTTP), 1.25 U 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 56–59°C for 2 min, and extension at 72°C for 3 min, for a total of 28–29 cycles, which was followed by a final extension of 15 min.
|
Polyacrylamide Gel Electrophoresis (PAGE)
Cells and cultures were terminated using lysis buffer (10 mM Tris, 0.15 M NaCl, 2 mM PMSF, 2 mM EDTA, 2 mM N-ethylmaleimide, 1% NP-40 [v/v], 10% glycerol [v/v], pH 7.4 at 22°C). Whole cell and testis lysates were prepared as described [38]. About 100 µg protein (30 µl) for each sample was resolved by SDS-PAGE under reducing conditions [39]. Electroblotting and detection of target proteins were performed as described [25, 26].
Immunofluorescent Microscopy
Fluorescent microscopy was performed using frozen rat testis sections as described [26]. The following antibody pair was used to detect the localization of two different antigens. Rabbit anti-PRKG and mouse anti-ß-catenin (anti-CATNB) (Chemicon; cat: MAB2081; lot: 23080643) were used at a working dilution of 1:300. For secondary antibodies, FITC-conjugated goat anti-rabbit IgG (H+L) (cat: 62-6111; Zymed, South San Francisco, CA) and Cy3-conjugated goat anti-mouse IgG (H+L) (cat: 81-6515; Zymed) were used at a dilution of 1:30 and 1:60, respectively. DAPI (Vector Laboratories, Inc., Burlingame, CA) was used to visualize nuclei.
Immunohistochemistry
Testes were isolated from control rats and rats treated with adjudin at 50 mg/kg BW, by gavage. Immunohistochemistry was performed essentially as earlier described and detailed elsewhere [38]. Rabbit anti-NOS2 antibody and the corresponding biotinylated goat anti-rabbit IgG (Vector) were used at a dilution of 1:300 and 1:1000, respectively. Micrographs were obtained using an Olympus BX40 microscope with an Olympus DP70 12.5 MPa digital camera interfaced to an HP Vectra VL800 Workstation via Firewire.
Coimmunoprecipitation
Coimmunoprecipitation (Co-IP) was performed as previously described [25]. Briefly, lysates of testes, seminiferous tubules, and Sertoli cells were prepared using the lysis buffer as described above. Equal amounts of proteins (400 µg) were pretreated with normal rabbit serum (1:150) for 3 h at room temperature. Thereafter, 20 µl Protein A/G PLUS-agarose (Santa Cruz Biotechnology) was added and incubated at room temperature for an additional 3 h to precipitate serum proteins that would non-specifically interact with IgG. After removal of the agarose beads by centrifugation at 1000 x g for 5 min, supernatants from each sample were incubated overnight with the corresponding antibody (1:150) at room temperature with agitation on a rotator at 24 rpm (Glas-Col, Glass Tech Supplies Inc., Fullerton, CA). Subsequent incubation was done at room temperature with the addition of 20 µl of a Protein A/G PLUS-agarose suspension for 4 h to recover the immunocomplexes. Immunoprecipitates were washed four times with the lysis buffer by resuspension and centrifugation (1000 x g, 5 min each). Samples were denatured in SDS sample buffer and resolved by SDS-PAGE under reducing conditions.
In Vivo Treatment of Adult Rats with KT-5823, a PRKG Inhibitor, Via Intratesticular Injection, to Blockthe Disruptive Effects of adjudin on Sertoli-Germ CellAJ Function
To confirm that Sertoli-germ cell AJ dynamics are indeed regulated by NO via the cGMP/PRKG signaling pathway, we sought to investigate if a blockade of PRKG could block or delay adjudin-induced germ cell loss from the seminiferous epithelium. Assuming the volume of an adult rat testicle was 1.6 ml, 1.6 or 16 nM KT-5823 (C29H25N3O5) (BIOMOL), a PRKG inhibitor [40], was administered via intratesticular injection using 26-gauge needles as described [38, 41]. In brief, KT-5823 was prepared as a stock in dimethylsulfoxide and was diluted in saline to a final volume of 140 µl (with 5% dimethylsulfoxide) and injected into two sites per testis (i.e., 70 µl per site). The dosages of KT-5823 used were selected based on published findings [40]. Thereafter, rats were treated with a single dose of adjudin (50 mg/kg BW, by gavage). Controls included rats without any treatment or treated with KT-5823 alone. Positive controls were rats treated with adjudin alone. On Day 4, animals were killed, with n = 4 rats per treatment group, and hematoxylin-eosin staining was performed using paraffin sections. To quantify the effects of KT-5823 (16 nmol per testis) on adjudin-induced germ cell loss from the epithelium, rats treated with KT-5823 plus adjudin were scored, at least 200 tubules were counted from each rat, and three rats were scored for the number of abnormal tubules (note: one of the four rats was used for pathology examination at the Rockefeller University Laboratory Animal Research Center for a preliminary toxicology study, which is outside the scope of this investigation; as such, only three rats were analyzed; however, changes in testicular weight were reported in Fig. 7G from all four rats). A tubule from rats treated with adjudin, KT-5823 + adjudin, or KT-5823 alone, which contained no elongating or elongate spermatids versus control tubules (for control normal testes, the number of elongating/elongate spermatids in typical frozen cross section of a seminiferous tubule at stages I–VI, VII– VIII, XI–XIV were 152 ± 22, 184 ± 35, 136 ± 22, respectively, and elongating/elongate spermatids were not found in stage IX–X tubules, which represented <10% of all the tubules scored. It was noted that, by Day 4 after adjudin treatment, fewer than 10% of the tubules had any elongating/elongate spermatids in the epithelium and the tubular diameter was reduced by 40%) was scored as a damaged tubule with significant loss in elongating/elongate spermatids from the epithelium. The percentage of seminiferous tubules (ST) with normal elongating/elongate spermatids after different treatments was calculated as follows:
 |
|
cAMP and cGMP Measurements
Intracellular levels of cAMP and cGMP were quantified in lysates of testes essentially as earlier described [25]. In brief, testes (about 0.1 g) were sonicated with an extraction buffer (50 mM Tris, pH 7.5, at 22°C, containing 5 mM EDTA, 2 mM PMSF, 1 mM N-ethylmaleimide, and 2 mM caffeine) and centrifuged at 15 000 x g for 25 min at 4°C to remove cellular debris. The levels of cAMP in the supernatant were quantified using [3H]-cAMP assay kits from Amersham Biosciences (Piscataway, NJ). The detection limit was at 0.05 pmol/assay tube and the 50% displacement was at 7 pmol. The cGMP levels were determined using a [3H]-cGMP assay kits from Amersham with a detection limit at 0.04 pmol/ assay tube and the 50% displacement at 3.5 pmol. The inter- and intraassay coefficients of variation for both assays was at 8–10% and 6–8%, respectively. All samples within an experimental group were processed and assayed simultaneously in the same assay session to eliminate interassay variations. Protein estimation was performed by Coomassie blue dye-binding assay using BSA as a standard [42].
Statistical Analyses
Statistical analyses were performed by either ANOVA using honest significant test (HST) by comparing data between samples at different times versus controls at Time 0 or between pairs of samples at other time points within an experimental group, or Student t-test using the GB-STAT Statistical Analysis Software Package (Version 7.0; Dynamic Microsystems, Inc., Silver Spring, MD). Each in vitro experiment was repeated at least three times using different batches of cells with replicate or triplicate cultures per time point. For in vivo studies, each experiment was repeated at least twice with at least three adult rats per time point.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Earlier electron microscopy studies have shown that functional anchoring junctions, such as desmosome-like junctions and AJ (e.g., ectoplasmic specialization) were established in Sertoli-germ cell cocultures within 24–48 h when they were prepared as described in the Materials and Methods [17, 43]. Interestingly, when germ cells were attaching to the Sertoli cell epithelium between 1 and 6 h before the establishment of functional anchoring junctions, an induction of NOS2 (Fig. 1A) and NOS3 (Fig. 1B) in these cocultures were detected (Fig. 1, C and D), illustrating their likely involvement in AJ assembly. Because the downstream activators of NO include cyclic nucleotides (e.g., cGMP and cAMP), their intracellular levels were also quantified in these cocultures. It was noted that a surge in intracellular cGMP (Fig. 2A) but not cAMP (Fig. 2B) was detected within 1 h when germ cells attached onto the Sertoli cell epithelium, and this elevated level was maintained throughout the entire experimental period during Sertoli-germ cell AJ assembly. Furthermore, the downstream activator of cGMP, namely PRKG (Fig. 2C) but not PRKA (which is the downstream activator of cAMP, only the 55 kDa regulatory subunit II was shown herein) (Fig. 2D), was also induced in these cocultures by about 6 h (Fig. 2, C versus D). Collectively, these results illustrate that NOS was activated within the same time frame as the intracellular cGMP level, which was followed by PRKG.
|
|
Changes in the Expression and Localization of NOS2 and/or NOS3 in the Testis During adjudin-Induced Germ Cell Loss from the Epithelium
Adjudin is known to induce germ cell loss from the seminiferous epithelium in vivo by perturbing Sertoli-germ cell AJ function, in particular those found at the site of apical ES between Sertoli cells and late spermatids [33, 34, 44] without affecting the BTB integrity [43]. Interestingly, the adjudin-mediated AJ disruption was associated with a significant surge in the expression of NOS2 (Fig. 3, A–C), but not NOS3 (Fig. 3, A–C), at the mRNA transcript and protein levels within 4–5 h after treatment. This pattern of changes, shown in Figure 3, A–C, where rats were administered 40 mg/kg BW i.p. (three doses, q 1-wk on Days 0, 7, and 14) is consistent with results using samples derived from the treatment regimen using a single dose of adjudin (50 mg/kg BW by gavage) to induce germ cell loss from the epithelium by monitoring the status of spermatogenesis from Day 0 up to Day 21 (data not shown). The induction of NOS2 was further confirmed by immunohistochemistry using an NOS2-specific antibody (Fig. 3, D and E). Virtually no staining was detected in the seminiferous epithelium when the NOS2 antibody was substituted with normal rabbit serum (Fig. 3, D versus E). In addition, a single band of 131 kDa corresponding to the immunoreactive NOS2 protein product in the testis was detected by immunoblotting, as shown in Figure 3F, which illustrates the specificity of the antibody. In normal (control) adult rat testes, NOS2, which appeared as brownish precipitates, was localized largely to pachytene spermatocytes and Sertoli cells (Figure 3D, control), consistent with results of an earlier report [5]. In addition, some brownish precipitates of immunoreactive NOS2 were also detected surrounding the heads of elongate spermatids at stage VII and VIII tubules (Fig. 3D, normal rat testis, control) (see the boxed area marked with an asterisk in Control, which was magnified and shown in the inset). By 4 h after adjudin treatment, the overall intensity of NOS2 staining increased mildly (data not shown) and the difference was more obvious by 16 h posttreatment (Fig. 3D, see 16 h versus control normal testes) (see arrowheads in 16H-a, which is the magnified view of the inset area in 16H). At that time, NOS2 was also accumulated intensively in Leydig cells (see arrows). Furthermore, germ cells (e.g., round spermatids) were also found in the tubule lumen by 16 h post-adjudin treatment. By 2 days after treatment, intense NOS2 staining of Sertoli cells and germ cells persisted and were detected in the seminiferous epithelium at different stages of the epithelial cycle (Fig. 3D). By 4 days after treatment, the overall staining intensity of NOS2 in the seminiferous epithelium was significantly decreased when most elongating/elongate and round spermatids were depleted; and by 7 days, when nearly all germ cells, except for some spermatocytes and spermatogonia, were depleted from the epithelium, NOS2 staining was greatly reduced and was restricted mostly to the basal compartment of the epithelium (Fig. 3D). Collectively, these data (Fig. 3, A–F) further implicate the possible involvement of NOS2 in AJ dynamics. In this context, it is noted that the set of micrographs shown in Figure 3D were derived from the regimen in which rats received a single dose of AF-2364 at 50 mg/ kg BW (by gavage); however, this trend of germ cell loss and the pattern of localization of NOS2 in the epithelium is virtually identical to the other treatment regimen (i.e., 40 mg/kg BW, q 1-wk by i.p. on Days 0, 7, and 14) reported herein (data not shown). Furthermore, over the course of the past 10 yr, our laboratory had performed more than 50 different regimens to monitor the efficacy of adjudin, and it was noted that at a dose ranging between 37.5 and 50 mg/kg BW (by gavage, i.p., or i.m.), adjudin yielded virtually identical antifertility effects in adult rats when monitored both by mating studies and histological analyses [35]. Figure 3G also depicts the changes in testicular weight (per pair testes) in the two regimens versus normal rats, and both treatment regimens induced similar kinetics of testicular weight loss (per pair testes). For instance, while the loss of testicular weight is not significant between 4 and 16 h (see also Fig. 3D) in both treatment regimens, the per pair testes weight (n = 3 rats) reduced to 1.9 ± 0.1 (40 mg/kg BW, three doses, i.p.) and 1.7 ± 0.1 (50 mg/kg BW, one dose, by gavage) from 3.4 ± 0.3 (normal rats) (Fig. 3G), representing an almost 50% decline in testicular weight because of germ cell loss from the epithelium consistent with the histology data shown in Figure 3D.
Induction in the Levels of cGMP and PRKG but Not cAMP and PRKA During the adjudin-Induced Germ Cell Loss from the Epithelium
The levels of cGMP, cAMP (Fig. 4, A and B), and their corresponding downstream activators (Fig. 4, C and D) were also quantified during adjudin-induced germ cell loss from the epithelium. In brief, the levels of cGMP and PRKG were induced during adjudin-mediated Sertoli-germ cell AJ disruption in vivo (Fig. 4, A and C). On the contrary, there was a steady reduction in the levels of cAMP and PRKA during adjudin-induced germ cell loss (Fig. 4, B and C). The immunoblots shown in Figure 4C for PRKG and PRKA were also reprobed for actin to confirm equal protein loading. Collectively, these data suggest that the NOS/cGMP/PRKG is one of the putative signaling pathways that is associated with the adjudin-mediated germ cell loss from the seminiferous epithelium in the rat testis.
|
Structural Association of NOS2 and NOS3 with Components of the CDH2/CATNB/Actin Protein Complex in the Testis and Their Dissociation from CDH2 and CATNB During adjudin-Mediated Germ Cell Loss from the Epithelium
Because results of the above studies have implicated the possible involvement of NOS in regulating Sertoli-germ cell AJ dynamics, we sought to examine if it interacted with components of the AJ protein complexes, such as CDH2, CATNB, PVRL3, MLLT4, ITGB1, and actin by Co-IP. Figure 5A summarizes results of this study, illustrating the association of NOS2 and NOS3 with CDH2, CATNB, and actin, but not PVRL3, MLLT4, and ITGB1, suggesting that NOS was physically associated with the CDH2/CATNB/ actin protein complex. We next investigated if there were any changes in these protein-protein associations during adjudin-induced germ cell loss from the epithelium because adaptors, such as CATNB, are essential in maintaining the complexity and stability of the CDH/CATN complex in the testis [for a review, see 45]. Interestingly, at the time of germ cell loss from the epithelium, while there was an overall increase in both CDH2 and CATNB in the epithelium (Fig. 5, B and C, panel a), there was a drastic loss in the association of either CDH2 or CATNB with NOS (Fig. 5, B and C, panels b and c).
|
Colocalization of CATNB and PRKG in Normal and adjudin-Treated Rat Testes
Because NOS was shown to dissociate from CATNB and CDH2 during adjudin-mediated germ cell loss in the testis, it was postulated that the downstream signaling proteins of NOS, such as PRKG, would be crucial to induce the dissociation of the CDH/CATN complex, possibly by phosphorylating CATNB, as reported earlier in other epithelia [for reviews, see 46, 47]. As such, it was anticipated that CATNB would colocalize with PRKG in the seminiferous epithelium. Indeed, PRKG was shown to colocalize with CATNB in the basal and adluminal compartment of the seminiferous tubules at the sites of basal and apical ES in control rats (normal testes) (Fig. 6A, 0H). Interestingly, colocalization of CATNB and PRKG was still detected in the seminiferous epithelium during adjudin-mediated Sertoli-germ cell AJ disruption in the testis by 2 days posttreatment (Fig. 6A). It is noted that the tubule by 2 days post-adjudin treatment was typified by the presence of spermatocytes and round spermatids in the tubule lumen (see DAPI staining of germ cell DNA). The antibodies used for this experiment appeared to be specific, as illustrated in the immunoblots shown in Figure 6B showing the electrophoretic mobility of PRKG (76 kDa) and CATNB (92 kDa) using testis lysates under reducing conditions (Fig. 6B). Furthermore, the presence of physical contacts between CATNB and PRKG was further confirmed by immunoprecipitation, in which Co-IP using an anti-PRKG antibody with testis lysates was shown to pull out both PRKG and CATNB and vice versa (Fig. 6C), validating results of colocalization by fluorescent microscopy shown in Figure 6A.
|
KT-5823, a PRKG Inhibitor, Could Block and Delay the adjudin-Induced Germ Cell Loss from the Seminiferous Epithelium
To further delineate if NOS/NO indeed regulates Sertoli-germ cell AJ dynamics via the cGMP/PRKG signaling pathway, rats were pretreated with KT-5823, a known PRKG inhibitor [40], to investigate if it can block the adjudin-induced germ cell loss from the epithelium by Day 4 (Fig. 7, A–F). Indeed, when adult rat testes were pretreated with 16 nmol of KT-5823, it was shown that the number of damaged tubules typified by the absence of elongating/ elongate spermatids in the seminiferous epithelium; or the presence of spermatocytes and round spermatids in the tubule lumen was significantly reduced versus adjudin treatment alone. For instance, in rats treated with adjudin alone on Day 4, virtually no elongating and elongate spermatids were found in the seminiferous epithelium of all tubules examined (Fig. 7, A and B versus C and D, and G, upper panel). Tubule marked with an asterisk shown in Figure 7C represents a tubule with obvious germ cell loss from the epithelium including elongating/elongate spermatids, round spermatids, and some spermatocytes; for those that are not marked, while they could be tubules at stages IX or X (note: in normal rat testes, stages IX and X tubules having no elongating/elongate spermatids in the epithelium represent <10% of the total tubules scored; see Fig. 7, A and B). It is also obvious that, in rats treated with adjudin on Day 4, there was a significant reduction in the tubular diameter by as much as 40% versus control (normal) testes in >95% of the tubules examined (see Fig. 7, A and B versus C). However, in rats where testes were pretreated with KT-5823, while elongating and elongate spermatids were also not found in 80% of the tubules examined (Fig. 7G, upper panel), still the number of tubules having elongating/elongate spermatids was significantly higher in the KT-5823 + adjudin treatment group versus the adjudin treatment group (Fig. 7G, upper panel). Furthermore, the number of spermatocytes and round spermatids in most tubules examined was relatively normal, except for about 10% of the tubules in which germ cell detachment indeed occurred (Fig. 7, E and F versus C and D, and G). Furthermore, the reduction in tubular diameter in rats pretreated with KT-5823 before adjudin treatment was less severe than adjudin alone on Day 4 when compared with control testes, 20% versus 40% (see Fig. 7, A and C versus E). These results are also consistent with changes in testicular weight, as shown in the lower panel of Figure 7G. These results are significant because they illustrate the crucial role of PRKG in regulating the cell adhesion function between Sertoli and germ cells in the epithelium, and a disruption of PRKG using a specific inhibitor (e.g., KT-5823) can indeed perturb the adjudin-mediated germ cell loss from the epithelium.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
NOS and NO are known to participate in diverse physiological and pharmacological functions, such as inflammation, junction permeability, and cell signaling [for reviews, see 1, 48, 49]. Recent studies in vitro from our laboratory have also demonstrated the significance of NOS/ NO in the regulation of Sertoli cell TJ-permeability barrier, which is mediated via the sGC/cGMP/PRKG pathway [25]. Because Sertoli-Sertoli AJs (e.g., basal ES and basal TBC) and TJs are intriguingly interacting with each other in the basal compartment of the seminiferous epithelium, which in turn constitutes the BTB in the mammalian testis, including the rat [9, 15, 50], it is anticipated NOS/NO may also play a role in AJ regulation. Indeed, it was shown that NOS is a crucial regulator of AJ dynamics in the testis. Studies conducted in vitro have clearly illustrated that the assembly of Sertoli-germ cell anchoring junctions (e.g., ES and desmosome-like junctions) was associated with an increase in NOS with a concomitant increase in intracellular cGMP, to be followed by an increase in PRKG but not cAMP nor PRKA levels. Collectively, these data illustrate that NOS/NO regulates Sertoli-germ cell AJ dynamics in vitro, which is likely mediated via the cGMP/PRKG signaling pathway. To further validate this information and to assess its physiological relevance in vivo, a series of in vivo studies were conducted using a recently established model of AJ dynamics in the rat testis using adjudin [9, 43, 50].
Interestingly, while the physiological action of NOS2 and NOS3 are similar because both proteins produce NO as their enzymatic product, their effects on AJ dynamics apparently are quite different. For instance, a drastic induction in NOS2, but not NOS3, expression was detected during adjudin-mediated AJ disruption in the rat testis. In this context, it is of interest to note that the changes in protein levels of these two NOSs are different during maturation of germ cells and rat testes [25]. For instance, the steady-state mRNA level of NOS2 is declining in aging germ cells and testes, whereas their NOS3 levels are increasing during maturation [25]. Collectively, these results suggest that these two NOSs may be using different downstream regulatory pathways to affect cellular function. The primary effect of adjudin, a potential male contraceptive, in the testis is to perturb AJ function between Sertoli and germ cells, in particular between Sertoli cells and spermatids/spermatocytes without killing these cells [33, 34]. Studies using [3H]-adjudin have shown that, following its absorption via the gastrointestinal tract, it peaked in the systemic circulation within 6–8 h and it was virtually cleared from the host body within 24–48 h [35]. In order for it to induce germ cell loss from the seminiferous epithelium so efficiently, adjudin is anticipated to exert its effect rapidly at the Sertoli-germ cell interface, at least in part, via activation of some crucial signaling pathways that regulate AJ integrity. In fact, the trend exhibited by NOS2 during adjudin-mediated AJ disruption is similar to that displayed by a Sertoli-germ cell AJ disruption marker, testin [41], which is also induced during the first 3 days post-adjudin treatment and returns to its basal level 14 days later [33]. This trend of induction of NOS2 mRNA and protein is consistent with results of immunohistochemistry when immunoreactive NOS2 was localized in the epithelium during adjudin-induced germ cell loss. For instance, an intense accumulation of immunoreactive NOS2 in Sertoli and germ cells after adjudin treatment was detected versus control testes. In addition, an intense staining was also associated with Leydig cells at the time of germ cell loss, yet earlier studies have shown that the serum levels of FSH, LH, and testosterone remain relatively unaltered in rats treated with adjudin to induce transient infertility [33, 34], implicating that adjudin treatment may have affected another Leydig cell function besides steroidogenesis. Further experiments must be conducted to explore this possibility.
In this context, it is of interest to note that an increase in the levels of NOS and its downstream cGMP/PRKG signaling molecules was detected during both AJ assembly (in vitro studies) and disassembly (in vivo studies), suggesting an activation of the NOS/cGMP/PRKG can take part in both the events of junction assembly and disassembly. Such a bifunctional role of a signaling pathway or a signaling molecule is not entirely unprecedented. For instance, the signaling of focal adhesion kinase (FAK), an important signaling molecule in focal adhesion complex remodeling at the cell-matrix interface, which is also found in the apical ES [51] in the testis, is indeed crucial to both focal adhesion assembly and disassembly [52–55] via a yet-to-be-defined mechanism that is reminiscent of the observations reported herein. Furthermore, dibutyryl cAMP was also shown to have a biphasic effect on the Sertoli cell TJ-permeability barrier in vitro [25, 56]. For instance, at 4–20 µM or 100– 500 µM, dibutyryl cAMP can either stimulate or inhibit the Sertoli cell TJ-barrier function in vitro [25, 56]. Nonetheless, much work is needed in future studies to delineate if such biphasic effects of the NOS/cGMP/PRKG signaling pathway on Sertoli-germ cell anchoring junction dynamics are the results of differential levels and activities of cGMP and PRKG, respectively.
Proposed Mechanism of the NOS/NO-Mediated Effects on AJ Dynamics in the Testis
PRKG, but not PRKA, is the downstream effector of NOS at the AJ site in the testis
Sertioli-germ cell AJ dynamics are intriguingly regulated (and finely coordinated) by an array of AJ integral membrane proteins, such as CDHs, PVRLs, and ITGA6B1, as well as their associated adaptors (e.g., CATNs, MLLT4s, p130Cas, -actinin, vinculin, paxillin, zyxin, axin, WASP) (for reviews, see [45, 57, 58]) (see Fig. 8). These molecules are, in turn, regulated by the interplay of phosphatases (e.g., myotubularins) and kinases (e.g., cSrc, CSNK2 [casein kinase II, formerly called CK-2], Csk, Fer kinase) that determines the phosphorylation status of the adaptors and/or integral membrane protein, and by small GTP-binding proteins, such as Rho and Rac [for reviews, see 7, 22, 43]. To add onto this list of potential regulators, we have shown in this article that NOS/NO apparently also plays a crucial role in regulating Sertoli-germ cell AJ dynamics in the testis via the downstream cGMP pathway. For instance, an induction in the level of cGMP but not cAMP during Sertoli-germ cell AJ assembly in vitro and disruption in vivo was observed, suggesting that the event of Sertoli-germ cell AJ restructuring involves cGMP and PRKG, the downstream protein kinase of cGMP. This is further supported by the observation that a blockade of PRKG by KT-5823, a specific PRKG inhibitor, can indeed delay the depletion of round spermatids and spermatocytes from the seminiferous epithelium as shown herein. The likely target of this pathway appears to be the CDH/CATN complex as shown in this report. In this context, it is of interest to note that the cGMP/PRKG pathway has been shown to downregulate the dynamics of cytoskeleton, such as actin, in other epithelia, and actin is the known attachment site of the CDH2/CATNB protein complex in the seminiferous epithelium of the rat testis [26, 31]. For instance, NO-mediated activation of GC/cGMP/PRKG pathway can lead to fragmentation of the actin cytoskeleton in human cervical epithelial cells [24]. Besides, PRKG activation can trigger the dissociation of CDH from actin in endothelial cells, leading to loss of cell adhesion function [59]. Besides the cGMP/PRKG pathway, NOS can also regulate AJ component proteins via other downstream pathways, such as via activation of adenylate cyclase (ADCY)/ guanylate cyclase (GUCY) and mitogen-activated protein kinase (MAPK) [for reviews, see 48, 60, 61]. Taken together, these findings suggest that NOS regulates the CDH/ CATN-based AJ structural proteins in the testis, at least in part, via the sGC/cGMP/PRKG pathway in the testis (Fig. 8). Because the apical ES is composed of two additional structural protein complexes, namely the ITG/LAM and the PVRL/MLLT4/ponsin complexes (see Fig. 8), it is anticipated that other signaling pathways are in play, which may converge with the soluble GUCY/cGMP/PRKG/CATNB to modulate spermatid adhesion function in the seminiferous epithelium.
|
The CDH/CATN protein complex is the likely downstream target of NOS
In this study, we have used the technique of Co-IP to investigate whether NOS is associated with any of the three known AJ complexes, namely the CDH/CATN, the nectin/MLLT4, and the ITGAB1/laminin 3 (LAMC3) complexes, in the testis [20, 26, 62–66; for reviews, see 7, 8]. NOS2 and NOS3 were found to associate with CDH2, CATNB, and actin but not PVRL-3, MLLT4, and ITGB1, implicating the specific interaction of NOS with the CDH2/CATNB/actin complex for the AJ regulation (see Fig. 8). This physical interaction may bring NOS to the vicinity of CDHs, CATNs, and actin, facilitating the action of NO on this protein complex and its downstream effector, such as PRKG, on this protein complex. This finding is consistent with an earlier report demonstrating the colocalization of NOS3 with vascular endothelial-cadherin at the plasma membrane in late confluent microvascular endothelial cells [23], implicating the potential interaction between NOS3 and CDH. Furthermore, we have also illustrated that PRKG, the downstream effector of NOS, is colocalized with CATNB in the seminiferous epithelium, suggesting that the testis is using PRKG as one of the regulators of the CDH/CATN complex. Interestingly, NOS was found to dissociate from CATNB and CDH2 during adjudin-mediated Sertoli-germ cell AJ disruption and the eventual germ cell loss from the epithelium. This illustrates that one of the actions of adjudin on the CDH2/CATNB protein complex is to alter the protein-protein interactions of the NOS/CDH2-CATNB protein complex, causing their dissociation, which is likely the result of an increase in PRKG intrinsic activity. Earlier studies have also suggested a link between an alteration of the CATNB localization and the induction of NOS2 expression in azoxymethane-induced rat colon adenocarcinoma cells in vitro [67]. Other studies performed using immortal mouse colonic epithelial cells have also demonstrated that NO can induce the release of CATNB from CDH, facilitating the formation of the nuclear CATNB/LEF-1 complex [68, 69], supporting the notion that NO is a crucial regulator of CATNB function at the cell junction site and its intracellular trafficking. Furthermore, PRKG was shown to be a regulator of the CATNB production in colon tumor cells [70]. Collectively, these results have illustrated the novel functional relationship between NOS/NO and the CDH-based cell adhesion complexes in the seminiferous epithelium. More important, the regulatory function of NOS/NO on AJ dynamics is mediated, at least in part, via the cGMP/PRKG/CATNB signaling pathway.
|
|
FOOTNOTES |
|---|
3 Current address: Department of Surgery, Faculty of Medicine, Queen Mary Hospital, University of Hong Kong, Hong Kong, China
2 Correspondence: C. Yan Cheng, Population Council, Center for Biomedical Research, 1230 York Avenue, New York, NY 10021. FAX: 212-327-8733; Y-Cheng{at}popcbr.rockefeller.edu
Received: 8 February 2005.
First decision: 9 March 2005.
Accepted: 26 April 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
P < 0.01 by comparing between KT-5823 + adjudin versus adjudin alone. A–C and E bar = 100 µm; D and F bar = 30 µm
