|Home | About | Journals | Submit | Contact Us | Français|
The blood-testis barrier (BTB) is an important ultrastructure in the seminiferous tubule of the mammalian testis that segregates the events of spermatogenesis, in particular post-meiotic germ cell development, from the harmful substances in the environment including toxicants and drugs, as well as from the unwanted hormones and biomolecules in the systemic circulation. It is known that the BTB is assembled by ~15–21 days postpartum (dpp) in rats coinciding with the onset of late cell cycle progression, namely the formation of zygotene and pachytene spermatocytes by day 15–18 dpp. This is to prepare for: (1) the differentiation/transformation of pachytene spermatocytes to diplotene and dictyate spermatocytes and (2) meiosis I and II, which take place by 23–26 and 26 dpp, respectively. Recent findings have shown spermatogonia/spermatogonial stem cells (SSC) in the tubules failed to re-initiate spermatogenesis by differentiating spermatogonia beyond type A spermatogonia in the absence of a functional BTB, leading to meiotic arrest. These studies thus illustrate that a functional BTB is crucial to the initiation and/or re-initiation of spermatogenesis. Herein, we sought to examine the precise time window when a functional and intact BTB is established in the developing rat testis during the final stage of cell cycle progression and meiosis. Using the techniques of: (1) dual-labeled immunofluorescence analysis to assess the distribution of integrated proteins at the tight junction (TJ), basal ectoplasmic specialization [basal ES, a testis-specific atypical adherens junction (AJ) type] and gap junction (GJ) at the BTB, (2) functional assay to assess the BTB integrity in vivo, (3) immunoblot analysis to monitor changes in steady-state levels of adhesion proteins at the BTB, and (4) co-immunoprecipitation to assess changes in protein-protein interactions at the BTB, it was shown that a BTB was being assembled by day 15–20 dpp, but a functional BTB was not fully established until day 25 dpp in Sprague-Dawley rats, tightly associated with the onset of meiosis I and II. These findings thus illustrate the significance of the BTB on cell cycle progression and the preparation for meiosis, such as germ cell differentiation beyond type A spermatogonia.
The blood-testis barrier (BTB) is an important ultrastructure in the seminiferous tubules of the mammalian testis. It segregates the seminiferous epithelium into the basal and the adluminal (apical) compartments so that post-meiotic germ cell development takes place in a unique microenvironment. This thus prevents the production of anti-spermatid antibodies, many of which express transiently during spermiogenesis.1–3 Besides conferring the testis its immune-privilege status,2,4 BTB also serves as a “gatekeeper” to restrict paracellular flow and/or transport of biomolecules, water, electrolytes, ions, and hormones across the barrier, as well as maintaining cell polarity in the epithelium.2,5 BTB was reported to be established in rats by age 15–186,7 or 218 day postpartum (ppd), and this time also coincides with the cell cycle progression from zygotene to pachytene spermatocytes by 15–18 dpp, and then diplotene and dictyate spermatocytes by 23–26 dpp, to be followed immediately by secondary spermatocytes and round spermatids to complete both meiosis I and II by 26 dpp.9 These findings thus implicate the likely involvement of a functional BTB and cell cycle progression and meiosis.
We recently reported that the fertility of adult rats treated with adjudin, 1-(2,4-dichlorobenzyl)-1H-indazole-3-carbohydrazide (formerly called AF-2364, a potential male contraceptive), failed to recover using an acute high-dose at 250 mg/kg b.w. vs. the low-dose group received adjudin at 50 mg/kg b.w. where fertility was found to rebound.10 Earlier studies have shown that adjudin disrupts germ cell adhesion in the seminiferous epithelium, most notably elongating/elongated spermatids, to be followed by round spermatids and spermatocytes.11–13 It was noted that in rats from the low-dose treatment group, spermatogonia/spermatogonial stem cells (SSC) were not depleted so that spermatogenesis could re-initiate to repopulate the voided tubules with germ cells of all classes gradually after adjudin was metabolically cleared from the host animals and fertility could rebound.11–15 Interestingly, although the fertility failed to recover in rats from the highdose group, the population of spermatogonia/SSC in the tubules was found to be similar to that of the low-dose treated group and untreated control rats.10 Detailed analysis of these animals revealed that in the low-dose group, the BTB was only transiently disrupted, yet the BTB was irreversibly damaged in the high-dose group, and spermatogonia/SSC were found to fail to differentiate beyond type A spermatogonia in rats from the high-dose group.10 These findings thus illustrate for the first time that a functional BTB is crucial to spermatogonia/SSC differentiation beyond type A spermatogonia, and to prepare for meiosis. This postulate is supported by an earlier study in which neonatal rats treated with diethylstilbestrol, a synthetic nonsteroidal estrogen, that delayed the BTB assembly by ~4-wk was found to delay the initiation of cell cycle progression and meiosis by 4-wk since type A spermatogonia failed to differentiate to type B spermatogonia and preleptotene spermatocytes.8
Since developmental data have shown that (1) late cell cycle progression to diplotene and dictyate spermatocytes does not occur until 23–26 dpp; (2) meiosis I and II do not complete until 26 dpp in rats,9 and (3) in light of the recent findings using the adjudin model regarding the onset of diakinesis and meiosis I/ II that failed to occur in the absence of a functional and intact BTB,10 we thought it pertinent to re-assess the precise time window when a functional BTB is being established during postnatal development in rats. This study should help to reassess the physiological significance of BTB in cell cycle progression and the onset of meiosis during spermatogenesis. This is the subject of the present report.
In order to determine the time when a functional BTB is being established in the testis during development, the steady-state levels of several BTB proteins including (1) TJ proteins: occludin, claudin-11 and ZO-1; (2) basal ES proteins: N-cadherin and β-catenin; (3) GJ proteins: connexin-43; and (4) non-receptor protein tyrosine kinases that are known to regulate BTB: FAK and c-Src were assessed by immunoblotting (Fig. 1) using the corresponding specific antibodies (see Table 1). These proteins were selected for the present study because they are the known components of TJ, basal ES, and GJ, which, in turn, constitute the functional Sertoli cell BTB via their coexistence at the site in the rat testis, and more importantly, these antibodies are well characterized as reported earlier.2,15–18 It was noted that all the TJ and GJ proteins examined as reported herein showed a transient surge in their steadystate level during development and was peaked by ~17–25 dpp. The surge of these proteins was apparently being used as building blocks to assemble the BTB, consistent with earlier reports that BTB was assembled at ~15–18 dpp based on electron microscopy. 6-8 Interestingly, the steady-state levels of two BTB regulatory protein kinases known to be involved in BTB dynamics19,20 also induced at 17–20 dpp, but the steady-state levels of c-Src and FAK remained elevated well into the adulthood versus immature rats at 12–15 dpp (Fig. 1A and B). On the other hand, the protein levels of basal ES proteins N-cadherin and β-catenin were high by day 12–15 dpp, but gradually declined during development (Fig. 1). In short, it is obvious from these findings that the BTB appears to initiate its assembly beginning on ~15–20 dpp (as illustrated by cluadin-11) and persisted through ~25 dpp as illustrated by ZO-1, N-cadherin, occludin, connexin-43, FAK and c-Src (Fig. 1A and B). The levels of occludin, ZO-1, N-cadherin, β-catenin, connexin-43, FAK and c-Src remained steady after 25 dpp because they need to maintain the BTB integrity during the seminiferous epithelial cycle, such as during BTB restructuring at stage VIII of the epithelial cycle to facilitate the transit of preleptotene spermatocytes at the site to enter the adluminal compartment for further development.
We next examined changes in the patterns of localization and co-localization of several BTB-associated proteins in the seminiferous epithelium during postnatal development. This was to assess their distribution in developing rat testes vs. adult rat testes when a functional BTB is known to be present. For instance, besides inducing the constituent proteins during development as illustrated by immunoblot analysis shown in Figure 1, the expressed constituent proteins require proper localization such that a continuous belt-like structure is formed near the basement membrane in the tubules.7,21 Thus, dual-labeled immunofluorescence analysis using fluorescent microscopy was used to study the localization of dig-11 (Fig. 2A and B), basal ES protein N-cadherin (Fig. 3) and GJ protein connexin-43 (Fig. 4) vs. the adaptor protein ZO-1 (Figs. 2–4) which is known to interact with TJ, basal ES, and GJ proteins in the rat testes.22,23 It was noted that the TJ proteins occludin and cluadin-11 formed a weak but diffused beltlike structure at the base of the seminiferous epithelium beginning on 15 dpp, and with obvious co-localization with ZO-1 by 17 dpp (Fig. 2A and B). The “occludin-ZO-1 belt-like structure” and the “claudin-11-ZO-1 belt-like structure” became more and more defined during development (Fig. 2A and B) and by 25 dpp, occludin and ZO-1 were more confined to the BTB site, similar to those found in adult testis by 120 dpp (Fig. 2A). However, claudin-11 was restrictively localized to the BTB site only by 38 dpp (Fig. 2B). On the other hand, unlike occludin, a belt-like structure with a relatively diffused pattern was observed by 20 dpp for the basal ES protein N-cadherin (Fig. 3), and a restrictive distribution that confined to the BTB site was not detected until 30 dpp, and only at this age that N-cadherin was found to co-localize with ZO-1, displaying a pattern similar to rats by 38 and 120 dpp (Fig. 3). The GJ protein connexin-43 began to form a diffused belt-like structure near the basement membrane by 15 dpp, and connexin-43 was found to co-localize with ZO-1 by 17 dpp (Fig. 4). But a well defined belt-like structure of connexin-43 that was restricted to the base of the seminiferous epithelium and co-localized with ZO-1 was not apparent until 25 to 30 dpp (Fig. 4). Thus, the findings based on dual-labeled immunofluorescence analysis to assess the distribution of TJ-, basal ES-, and GJ-proteins illustrate that BTB began to be assembled from 15 dpp, but not all the BTB-associated proteins were restricted to the BTB site until at least by 25 dpp. For instance, occludin and ZO-1 were found to be restrictive to the BTB site by ~20 dpp, but this pattern of restrictive localization at the BTB was not found for cluadin-11, N-cadherin and connexin-43 until ~25 to 30 dpp (Figs. 3 and and44 vs. vs.22).
To further define the relative timing regarding the establishment of a functional BTB in developing rat testes, an in vivo functional assay was used to assess the presence of an effective BTB that blocked the entry of fluorescence tag from entering the apical compartment following its administration via the jugular vein, and that semi-quantitative data can be obtained from such an assay for statistical comparison and analysis (Fig. 5). It was found that by 12 dpp at the time the BTB remained to be assembled, it was almost as “leakly” as rats from the cadmium-treated group which is known to induce BTB disruption21,24,25 (Fig. 5). Even though by 15 dpp when the BTB begins its assembly based on dual-immunofluorescence analysis data shown in Figures 2–4, the BTB remained uneffective to block the movement of fluorescence tag across the barrier into the adluminal compartment (Fig. 5A). Although statistical analysis illustrates that a functional BTB was established by day 15 and 17 dpp (Fig. 5B), but many tubules remained “leaky” to the fluorescence tag (Fig. 5A), consistent with dual-labeled immunofluorescence data shown in Figure 2–4. These findings, nonetheless, are in agreement with earlier reports that a functional BTB was being established by 16–18 dpp, but it was not formed until 21 dpp.6–8,26 However, Figure 5 has convincingly demonstrated that a fully effective, intact and functioning BTB was only established until at least day 25 dpp, since only at this age, most of the tubules displayed the ability to block the transit of fluorescence tag across the BTB, blocking their entry into the adluminal compartment (Fig. 5A and B). The findings depicted in Figure 5 are also in agreement with data obtained by dual-labeled immunofluorescence analysis shown in Figures 2–4, depicting the relative patterns of localization of TJ-, basal ES- and GJ-proteins and their co-localization with ZO-1 at the BTB in developing rat testes vs. adult rats at 120 dpp, demonstrating a functional BTB was not established until 25 dpp, coinciding with the final steps of meiosis I and the completion of II which occurs by 26 dpp.9
Since FAK was recently shown to be an integrated component of the occludin-ZO-1 adhesion protein complex at the BTB, and it is known to confer proper phosphorylation status to occludin at the BTB,20,27 we thus sought to examine if the pattern of protein-protein interactions between FAK and two BTB integral membrane proteins correlated with the establishment of the BTB in the testis during development. Connexin-43 was also selected in this study because recent studies have shown that besides being an integrated component of the GJ at the BTB,23 connexin-43 also regulates reassembly of the TJ-barrier28 which is essential to maintain the immunological barrier during the transit of preleptotene spermatocytes at the BTB to prepare for meiosis I. Using the technique of co-immunoprecipitation, the association between FAK and occludin as well as between FAK and connexin-43 gradually increased during development and peaked by day 20 at the time of BTB assembly (see Figs. 1–5), perhaps being used to maintain the proper phosphorylation status of the TJ and GJ proteins, so that they can be assembled to the TJ-fibrils and the coexisting GJ, respectively, to be used to establish the BTB (Fig. 6).
In this study, we have assessed the time in which a fully functional BTB is assembled in the rat testis. We sought to use a combination of techniques including (1) immunoblotting by assessing the steady-state levels of a number of adhesion and constitutent proteins at the BTB, (2) dual-labeled immunofluorescence analysis of BTB-associated proteins including proteins at the TJ, basal ES, and GJ by fluorescent microscopy, (3) an in vivo BTB integrity assay, and (4) co-immunoprecipitation. Collectively, these results have demonstrated that the BTB begins its assembly by 15–20 dpp, however, a fully functioning BTB is not established until 25 dpp. First, this conclusion is supported by the results from immunoblotting that illustrated the transient increase of TJ proteins necessary for BTB assembly including occludin,29 claudin-1130,31 and ZO-132 was at 15–20 dpp. Moreover, GJ protein connexin-43 which is important for maintaining BTB integrity23,28 was also transiently increased at 15–20 dpp. Second, proper localization of integral membrane proteins and their adaptors can also affect the BTB integrity. Earlier studies have shown that a functional BTB is associated with a continuous belt-like structure formed by TJ proteins that constitute the TJ-fibrils near the basement membrane.7,21 Additionally, several studies have demonstrated that in pathological conditions, such as during tumorigenesis associated with male reproductive dysfunction when the BTB was disrupted and compromised, mis-localization of claudin-1133,34 or ZO-132 was detected by immunohistochemistry. Moreover, a recent study has shown that although occludin and claudin-11 were induced significantly at the BTB after acute doses of adjudin in adult rats, BTB was still disrupted because the significantly induced occludin and claudin-11 were mis-localized,10 illustrating proper localization of proteins at the BTB is crucial to maintain its integrity. Based on dual-labeled immunofluorescence analysis as reported herein, well-defined and co-localized “occludin-ZO-1,” “claudin-11-ZO-1” and “connexin-43-ZO-1” belt-like structures were only apparent by 25 dpp. This thus demonstrates that a fully functioning BTB is established by 25 dpp. Third, based on findings from the in vivo BTB assay, a drastic reduction of fluorescence signals in the adluminal compartment was found to be diminished by 25 dpp vs. pups at 12, 15, 17 and 20 dpp. Thus, this finding, coupled with results of the co-immunoprecipitation experiment further suggests that the BTB is only functionally intact by 25 dpp. However, it is noteworthy that by 20 dpp, some of the tubules displayed a significantly reduced fluorescent signal penetration beyond the BTB vs. other tubules in the same cross-sections, and statistical analysis also supported the notion that the BTB integrity was established by 20 dpp, yet the BTB functionality in the testis as a “whole” was not truly established until 25 dpp when the findings depicted herein were analyzed collectively. Since by the age of 25 dpp, the ability of BTB to restrict the movement of FITC-inulin to the basal compartment is analogous to that of an adult rat testis by 120 dpp, illustrating no fluorescence was detected at the apical compartment in >98% of the tubules scored. This finding, nonetheless, indicates that the final phase of BTB establishment is taking place on 20 dpp, and some tubules are better established vs. others at that age. This observation, however, is in agreement with a previous study indicating that the BTB is not established synchronously along the cords, its establishment is depending on the stage of meiosis of germ cells present in tubules.7
Previous studies have indicated that the BTB is established by 16–21 dpp in rats,6–8,26 which is slightly earlier than the time estimated in this study. This slight discrepancy can be due to the differences in methods that were used in different studies. For example, an earlier study performed by Russell et al. to determine the BTB integrity during postnatal development was based on initial washing of the testis with hypertonic solution prior to perfusion using fixative, to be followed by light microscopic examination; and it was noted that the BTB was formed by 18 dpp.6 However, another study performed by Bergmann et al. using hypertonic fixative coupled with electron microscopy to assess the BTB integrity during development indicated that the BTB was established by 19 dpp.7 In a more recent study by Toyama et al. based on the use of cytochrome c coupled with light and electron microscopy, the BTB was shown to be established by 21 dpp.8 Herein, we have shown unequivocally that that depending on the BTB markers that were selected for duallabeled immunofluorescence microscopy, the timing of BTB establishment could be slightly different. For instance, occludin and ZO-1 was assembled to the BTB by as early as 17 dpp, but for claudin-11, N-cadherin and connexin-43, it was 38, 30 and 38 dpp, respectively. These observations thus illustrate the constituent proteins at the BTB that are being assembled to the site do not occur uniformly during development.
Earlier studies have shown that the BTB is composed of an array of TJ-, basal ES-, and GJ proteins' and in addition to these proteins, desmosomes are also key integrated components at the BTB,2,16 making the BTB a unique blood-tissue barrier among other barriers in the mammalian body, including blood-brain barrier, blood-retina barrier, blood-epididymal barrier, blood-biliary barrier and the gut-epithelial barrier. Since in these latter barriers, they are constituted almost exclusively by the TJ between endothelial and/or epithelial cells instead of coexisting TJ, basal ES, GJ and desmosome, thus a loss of any TJ-proteins can significantly impede the functioning of these blood-tissue barriers. Hence, a fully functional BTB in the testis does not require the presence of all the proteins from TJ, basal ES, GJ and desmosome simultaneously, and changes in the expression and localization of different BTB-associated proteins in the seminiferous epithelium during testicular development as reported herein also support this argument since the steady-state of some proteins are declining in adulthood (e.g., claudin-11), whereas others remain relatively constant following the establishment of a functional BTB (e.g., connexin-43, occludin). However, it is likely that a set of proteins is necessary to confer its functionality. Perhaps this is physiologically necessary and important to maintain the unique features of the BTB since unlike other blood-tissue barriers, the BTB undergoes extensive restructuring during spermatogenesis, such as at stage VIII of the epithelial cycle, to facilitate the transit of preleptotene spermatocytes. Thus, the “old” BTB site above the transiting preleptotene spermatocytes can be “gradually” degenerating while the “new” BTB site behind the spermatocytes can also be “gradually” assembling since a “loss” of either TJ, basal ES, GJ or desmosome is not going to lead to a “dissolution” of the immunological barrier. It is obvious that many of the signaling events that coordinate the proper functioning of these junction types that constitute the BTB remain unknown. Nonetheless, the findings report herein does not contradict to earlier reports regarding the assembly of the BTB that occurs by age 16–21 dpp in rats.6–8,26 For instance, the use of different marker proteins at the BTB indeed supports these earlier observations, such as the localization of occludin at the BTB. However, a fully functional BTB, based on a semi-quantitative in vivo assay, is only formed by 25 dpp in rats as depicted in Figure 7. This timing of BTB assembly also coincides precisely with the occurrence of meiosis I and II which is known to take place by ~26 dpp in rats,9 illustrating the significance of BTB and meiosis. This conclusion is indeed supported by findings in humans in which infertile men with oligospermia were having a disrupted BTB, and spermatogonia and spermatocytes failed to enter meiosis.35
In summary, we have shown that an intact BTB is relatively established by 20 dpp during postnatal development in rats, however, a fully functioning BTB is not assembled until 25 dpp. Figure 7 summarized the key findings in this report.
Sprague-Dawley (outbred) rats were purchased from Charles River Laboratories and housed at the Rockefeller University Comparative Bioscience Center (CBC). The use of Sprague-Dawley rats for studies reported herein was approved by the Rockefeller University Institutional Animal Care and Use Committee (Protocols 06018 and 09016). Antibodies were obtained commercially from different vendors and the working dilutions of these antibodies for various experiments are listed in Table 1.
The BTB integrity in vivo in rats was assessed by an assay established earlier in our laboratory as described.10,36,37 This is a semi-quantitative assay to assess the BTB integrity based on the ability of an intact BTB that blocks the diffusion of a small fluorescence tag, such as FITC-inulin, from the basal to the apical compartment of the seminiferous epithelium when the tag is administered to live animals at the jugular vein. In short, rats were under anesthesia with ketamine HCl (60 mg/kg b.w.) in the presence of an analgesic xylazine (10 mg/kg b.w.) (Sigma-Aldrich) administered i.m. A small incision, about 1-cm, was made in the area over the jugular vein with a surgical scissors to expose the blood vessel and depending on the age of rats, about 0.5–1.5 mg FITC-conjugated inulin (Mr 4.6 kDa) (Sigma-Aldrich) in 100–300 µl PBS, depending on the age and testis weight of the animal, was administered into the jugular vein with a 28-gauge needle, such that the fluorescence tag could reach the testis, attempting to traverse the BTB. Thereafter, the skin was closed with two wound clips (9-mm Autoclip from Clay Adams, Becton Dickinson). Rats were allowed to recover, and about 30–45 min thereafter, rats were euthanized by CO2 asphyxiation in a carbon dioxide rat chamber. Testes were removed immediately and snap-frozen with liquid nitrogen. Rats treated with CdCl2 at 5 mg/kg b.w. i.p. for 3-d were used as positive control since the BTB was reported to be irreversibly damaged by this treatment.21,24,25 The distribution of FITC (green fluorescence) in the seminiferous epithelium among tubules was monitored by fluorescence microscopy using a Nikon Eclipse 90i Fluorescence Microscope and images were obtained using a built-in Nikon DS-Qi1Mc digital camera at 12.5-Megapixel (Mpx) with Nikon NIS Elements Advanced Research Imaging Software (Version 3.2) (Nikon Instruments Inc.) in an HP xw8600 Workstation. Images were exported to TIFF format and analyzed in PhotoShop using Adobe Creative Suite (Version 3.0), such as image overlay. BTB was considered not fully formed when fluorescence signal was no longer confined to the basal compartment but detected in the adluminal compartment since in normal adult rat testes, all the fluorescence was blocked from entering the adluminal compartment. To obtain semi-quantitative data on the BTB integrity, the ratio between the distance traveled by the fluorescence signal from the basement membrane in a seminiferous tubule (DSignal) and the radius of the tubule (DRadius) was obtained for each tubule and each time point is the mean ± SD from a total of 180–240 tubules from 3 rats (i.e., 60–80 tubules from each rat). For instance, in the positive control group in which rats were treated with CdCl2 in which fluorescence was found in the center of a tubule in virtually all tubules, the ratio of DSignal/DRadius was at 1. If the cross-section of a seminiferous tubule was oval-shaped because it was obliquely sectioned, DRadius was obtained by averaging the shortest and the longest distance from the basement membrane.
Lysates of testes were prepared in IP lysis buffer [50 mM Tris, pH 7.4, at 22°C, containing 0.15 M NaCl, 1% Nonidet P-40 (vol/vol), 1 mM EGTA, 2 mM N-ethylmaleimide, 10% glycerol (vol/vol)] supplemented with protease inhibitor mixture (Sigma-Aldrich) and phosphatase inhibitor mixture I and II (Sigma-Aldrich) at a dilution of 1:100 to block protease and phosphatase activities using a Cole Parmer Ultrasonic Processor as described.18,38 Protein concentration was estimated using a BioRad Dc Protein Assay kit in 96-well plates with a BioRad Model 680 Spectrophotometry Reader. Approximately 100 µg protein were used for immunoblotting and ECL images were captured using a FujiFilm LAS-4000 mini chemiluminescence imaging system, with images captured using FujiFilm Multi Gauge software package (Version 3.1), converted to TIFF format and analyzed with Scion image as earlier described.39 All data were normalized against actin to correct for possible differences in protein loading between samples in a given experiment. All samples within an experimental group were processed simultaneously to eliminate inter-experimental variations. Antibodies used for immunoblot analysis were listed in Table 1. Each immunoblotting experiment was repeated at least three times with three different set of testis samples for statistical analysis.
Frozen sections (7 µm) were obtained in a cryostat at −20°C, which were fixed immediately in either Bouin's fixative or 4% paraformaldehyde (wt/vol) for 10 min. Sections were then permeablized with 0.1% Triton X-100 in PBS (10 mM sodium phosphate, 0.15 M NaCl, pH 7.4 at 22°C) (vol/vol) for 4 min. Sections were blocked using 1% BSA (wt/vol) for 30 min, to be followed by an overnight incubation of primary antibodies diluted in blocking solution at room temperature. Sections were then incubated with Alexa Fluor—conjugated secondary antibodies (Invitrogen; red fluorescence, Alexa Fluor 555; green fluorescence, Alexa Fluor 488) at 1:250 diluted with the corresponding blocking solution at room temperature. Sections were mounted with Vectorshield Antifade mounting media with DAPI (4′,6-diamidino-2-phenylindole, for staining cell nucleus) (Invitrogen) for fluorescence microscopy. Fluorescence images were captured using an Olympus DP71 12.5 Mpx digital camera interface to an Olympus BX61 fluorescence microscope using the Olympus MicroSuite Five Imaging software package (Version 1226) to obtain images in TIFF format. Image overlay and analysis were performed using PhotoShop in Adobe Creative Suite Design Premium software package (Version 3.0). All staining experiments were performed 2–3 times with different sets of testes. To reduce inter-experimental variations, testes from all time points within an experimental group were processed simultaneously in a single experimental session by placing 2–3 cross-sections of testes on a single microscopic slide. Negative control included the use of normal mouse or normal rabbit IgG diluted in PBS to substitute the primary antibody.
To assess changes in protein-protein interactions between occludin and FAK as well as between connexin 43 and FAK during testicular development, 500 µg of testis lysates from rats on 12-, 17-, 20-, 25- and 30-dpp were initially incubated with 1.5 µg normal IgG, to be followed by an incubation with 15 µl Protein A/G Plus (Santa Cruz) for 2 h in each sample, thereafter, the supernatant was obtained following a brief centrifugation (1,000 g, 5 min) which was used for subsequent co-immunoprecipitation. This pre-clearing step is important to eliminate unwanted non-specific interactions between the target protein [and its binding partner(s)] with IgG in the corresponding antibody. Lysates after the pre-clearing step were then incubated with 2 µg of primary antibody, either antioccludin or anti-connexin-43 IgG, overnight at room temperature (and for negative control, primary antibody was substituted with 2 µg of normal IgG). Thereafter, each sample was incubated with Protein A/G Plus (15 µl) so that the immunocomplexes containing IgG-target protein/binding partner(s) were bound to the resin and were obtained following centrifugation, which were then extracted in SDS-sample buffer (0.125M Tris, pH 6.8 at 22 C, containing 1% SDS, wt/vol, 1.6% 2-mercaptoethanol, vol/vol, and 20% glycerol, vol/vol) at 100 C for 5 min, to be followed by SDS-PAGE and immunoblot analysis using an anti-FAK antibody. Antibodies used for this coimmunoprecipitation experiment were listed in Table 1 and this experiment was repeated at least three times with three different set of testes to obtain sufficient data for statistical comparison.
Statistical analysis of data derived from immunoblotting, co-immunoprecipitation and BTB integrity assay was performed with GB-STAT software package (Version 7.0; Dynamic Microsystems) using two-way ANOVA followed by Newman-Keul's test. For data obtained from the BTB integrity assay, Student's t-test was used to compare corresponding groups annotated by brackets of different colors, such as 12 dpp vs. 12-, 17-, 20-, 25-, 30-, 38- or 120 dpp group.
This work was supported by grants from the National Institutes of Health (NICHD, U54 HD029990 Project 5 to C.Y.C.; R01 HD056034 to C.Y.C.; R01 HD056034-02-S1 to C.Y.C.). , and CRCG Small Project Grant from the University of Hong Kong (to W.M.L.). K.W.M. was supported by a Postgraduate Research Award from the University of Hong Kong.
No potential conflicts of interest were disclosed.