Division of Reproductive Biology, Department of Biochemistry and Molecular Biology, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD 21205 (e-mail: email@example.com).
ABSTRACT: Successful spermatogenesis requires that germ cells remain in physical contact with Sertoli cells until spermiation. Previous studies have shown that the Bcl2-modifying factor (BMF) is a proapoptotic protein found in many epithelial cells which, when phosphorylated by the active form of mitogen-activated protein kinase 8 (p-MAPK8), initiates apoptosis in response to loss of adhesion of the cells to their basal lamina. Based on this, we hypothesized that p-MAPK8 and BMF may play important roles in the apoptotic death of testicular germ cells in response to their detachment from Sertoli cells. Immunohistochemical analysis of the normal rat testis revealed p-MAPK8 expression in spermatocytes and elongated spermatids but not in round spermatids. This localization was opposite to that of BMF, which is expressed in round spermatids but not in spermatocytes or elongated spermatids. When freshly isolated germ cells were cultured in the absence of Sertoli cells, a condition in which there was widespread germ cell apoptosis, an increase in p-MAPK8 relative to overall MAPK8 protein, was seen by Western blot analysis. Additionally, immunocytochemical analysis showed an increase in immunoreactive p-MAPK8 in round spermatids and spermatocytes in association with BMF expression. From these correlative data, we propose that the activation of MAPK8 and redistribution of BMF may be integrally involved in the mechanism by which specific germ cells undergo programmed cell death in response to their detachment from Sertoli cells.
The Bcl2-modifying factor (BMF), a proapoptotic BH3-only member of the Bcl2 family of apoptosis-related proteins, has been shown to be an apoptotic “trigger” protein that initiates programmed cell death in some epithelial cells in response to their loss of adhesion to basal lamina (Puthalakath et al, 2001). In healthy cells, BMF is sequestered to the actin cytoskeleton via a myosin-binding domain (Puthalakath et al, 2001). Previous studies have indicated that when such cells are subjected to conditions that induce apoptosis, mitogen-activated protein kinase 8 (MAPK8; also known as JNK1) phosphorylates BMF on specific serine residues located within and directly adjacent to the BMF actin/myosin-binding domain (Lei and Davis, 2003; Kanda and Miura, 2004), and there is loss of actin-BMF sequestration (Lei and Davis, 2003). BMF, freed from actin, proceeds to the mitochondria, physically interacts with the death-opposing Bcl2 protein, and thus triggers the initiation of apoptosis (Puthalakath et al, 2001).
We demonstrated previously that BMF is expressed in germ cells of the rat testis, specifically in the subacrosomal space of postmeiotic spermatids of steps 4 to 16 (Show et al, 2004). Experimental conditions in which normal testosterone concentrations within the testes are reduced substantially have been shown to result in the apoptotic death of meiotic spermatocytes and in the premature sloughing and subsequent degeneration of postmeiotic round spermatids (Russell and Clermont, 1977; O'Donnell et al, 1994, 1996). Under such conditions, BMF is expressed in spermatocytes and more mature spermatids and is redistributed throughout round spermatids rather than confined to its normal subacrosomal localization (Show et al, 2004).
We hypothesized that in response to the detachment of germ cells from their associated Sertoli cells, MAPK8 would become activated by phosphorylation and that its activation would be associated with the expression/redistribution of BMF in the detached germ cells and thus with the death of the detached cells. We demonstrate that phosphorylated (ie, active) MAPK8 (p-MAPK8) is detected in spermatocytes and elongated spermatids but not in round spermatids. The increased apoptosis of germ cells that occurs when they are cultured in the absence of Sertoli cells was found to be associated with increased p-MAPK8 expression and with the expression and/or redistribution of BMF within the p-MAPK8-expressing apoptotic germ cells. The results suggest that the activation of MAPK8 and the expression/redistribution of BMF may be involved in the mechanism by which specific germ cells undergo programmed cell death in response to loss of their attachment to Sertoli cells.
Materials and Methods
Male Sprague-Dawley rats of 8 to 12 weeks were purchased from Charles River (Kingston, Massachusetts). Rats were housed in a vivarium under a 14:10 light:dark cycle and were provided water and rat chow ad libitum. All protocols were approved by the Johns Hopkins University Animal Care and Use Committee.
Rats were anesthetized and whole-body perfused with phosphate-buffered saline (PBS) to clear the testes of blood and then with neutral buffered formalin for 1 hour at a rate of 7 mL/min. The testes were removed and immersed in neutral buffered formalin overnight at 4°C. The tissue was dehydrated in ice-cold (4°C) 70%, 90%, and 99% ethanol for 1 hour each and then in absolute ethanol for 1 hour at room temperature. Tissue was infiltrated with 50% polyester wax/50% ethanol for 2 hours at 42°C followed by a 90% polyester wax/10% ethanol mixture for 1 hour at 42°C. The tissue was then transferred into 90% wax/10% ethanol in plastic embedding dishes and chilled on ice for 30 minutes or until the wax solidified. Sections (5 μm) were cut and mounted on HiPure subbed glass slides. The slides were dewaxed by immersion into 100%, 90%, and 70% ethanol baths for 10 minutes each. Some slides were blocked in normal goat serum diluted in PBS (1:60; Vector Laboratories Inc, Burlingame, California) and then incubated for 1 hour at room temperature with a rabbit polyclonal antibody raised against the N-terminus of the human BMF protein (1:200; Abcam Ltd, Cambridge, United Kingdom). Bound primary antibodies were detected with a fluorescein isothiocyanate (FITC)–conjugated goat anti-rabbit secondary antibody. Other slides were blocked in PBS-diluted normal horse serum and then incubated with a mouse monoclonal antibody raised against p-MAPK8 (1:100; Santa Cruz Biotechnologies, Santa Cruz, California). In this case, bound primary antibodies were detected with a FITC-conjugated horse anti-mouse secondary antibody (1:100; Vector Laboratories). The p-MAPK8 antibody has been used to localize p-MAPK8 by immunohistochemistry in several previous studies (Shiraishi et al, 2002; Kins et al, 2003; Lysiak et al, 2003; Ishii et al, 2004). Nuclei were stained with 4,6 diamidoino-2-phenylindole (DAPI; Vector Laboratories) in Vectashield anti-fade mounting medium. Antibody specificity was assessed by incubating slides with mouse immunoglobulin G (IgG) followed by secondary antibody application. BMF primary antibody specificity was demonstrated by performing the above immunostaining procedure preceded by the preabsorption of equal concentrations of the BMF primary antibody with the BMF N-terminal peptide used for antibody production (Show et al, 2004). Images were obtained with a Nikon Eclipse 800 microscope equipped with a Nikon planfluor ×40 objective using a Princeton Instruments 5-Mhz cooled CCd camera with custom CRI color filter and IPLab digital image analysis software for Macintosh.
Seminiferous Tubule Microdissection
Seminiferous tubule segments were isolated from rat testes by transillumination-assisted microdissection, as previously described (Parvinen, 1982). For protein analyses, approximately 15 to 20 cm of seminiferous tubules from stages I to V, VII, VIII, and IX to XIV of the spermatogenic cycle were dissected from the testes of control rats. Testes from 3 different rats were included as independent samples for protein isolation.
Germ Cell Isolation and Culture
Each germ cell preparation included both testes of an adult rat. Testes were decapsulated and incubated with 0.5 mg/mL collagenase (Sigma-Aldrich, St Louis, Missouri) in F12 Dulbecco modified Eagle medium (DMEM; Invitrogen, Carlsbad, California) supplemented with 1 mM sodium pyruvate (Sigma-Aldrich) and 13 mM lactate (Sigma-Aldrich) in a shaking water bath for 12 minutes at 34°C. The free seminiferous tubules were washed 3 times with F12 DMEM supplemented with lactate and pyruvate and allowed to settle between each wash. The tubules were digested with 0.5 mg/mL trypsin in supplemented F12 DMEM for 8 minutes in a shaking water bath at 34°C. Following digestion by trypsin, the remaining seminiferous tubule fragments were mechanically dispersed by pipetting until the cellular suspension became homogeneous. The crude germ cell suspension was then filtered through nylon mesh to remove any undigested seminiferous tubule fragments. The filtrate was centrifuged for 5 minutes. The pellet was washed 3 times with supplemented F12 DMEM and centrifuged as above. Finally, the pellet was resuspended and again filtered through nylon mesh. The isolated germ cells were assessed for purity by morphologic analysis and counted with a hemocytometer. The cells averaged approximately 90% purity, with Sertoli and myoid cells the principal contaminants. Following determination of purity, some germ cells were immediately snap-frozen for later protein analysis, and others were labeled with annexin V to determine apoptotic initiation or were cultured in F12 DMEM supplemented with lactate and pyruvate for 4, 8, or 12 hours at 34°Cina5% CO2 atmosphere so as to initiate apoptosis due to loss of Sertoli cell-germ cell adhesion.
Annexin V Labeling of Apoptotic Germ Cells
Freshly isolated germ cells, as well as germ cells that had been cultured in liquid media without Sertoli cells for 4, 8, or 12 hours, were assessed for the initiation of apoptosis by the annexin V-FITC apoptosis detection kit (Pharmingen, San Diego, California), according to the manufacturer's specifications. Briefly, 1 × 106 cells were added to 0.5 mL of Ca2+-binding buffer followed by the addition of 5 mL of annexin VFITC. The cells were incubated for 10 minutes at room temperature, during which they were protected from light. An aliquot of the annexin V—labeled cells was then transferred to a microscope slide. Germ cells labeled with annexin V-FITC were considered to be apoptotic. An apoptotic index was calculated by averaging the total number of annexin V- FITC—positive germ cells vs the total number of germ cells counted using phase-contrast microscopy. Images were obtained with a Nikon Eclipse 800 microscope, as above, equipped with a Princeton Instruments 5-Mhz cooled CCd camera.
Western Blot Analyses
Protein isolated from germ cells or from stage-specific seminiferous tubule segments were homogenized in radioimmunoprecipitation assay buffer (1% Triton X-100, 15 mM HEPES-NaOH [pH 7.5], 0.15 mM NaCl, 1% sodium deoxycholate, 0.1% SDS, 1 mM sodium orthovanadate, 10 mM EDTA, and 0.5% protease inhibitor cocktail [Sigma-Aldrich]), and stored at −80°C until analyzed. Protein concentrations were determined by the bicinchoninic acid method (Pierce, Rockford, Illinois), according to the manufacturer's specifications. Protein samples were added to an equal volume of 2× loading buffer (100 mM Tris [pH 6.8], 4% sodium dodecyl sulfate [SDS], 0.2% bromophenol blue, 20% glycerol). Samples were reduced with 0.1% β-mercaptoethanol, boiled for 2 minutes, and separated by 12% SDS-polyacrylamide gel electrophoresis. Protein was transferred to Protran nitrocellulose membranes (Schleicher & Schuell, Keene, New Hampshire) with a Trans-Blot SD semi-dry electrophoretic transfer cell (Bio-Rad, Hercules, California), according to the manufacturer's specifications.
Membranes were blocked for 1 hour with 5% nonfat dry milk in PBS (blocking solution) at room temperature, followed by anti—p-MAPK8 (1:300) overnight in blocking solution at 4°C. The membranes then were washed 3 times with PBS + 0.1% Tween-20 for 5 minutes and incubated for 30 minutes at room temperature with secondary anti-mouse horseradish peroxidase (HRP)–linked IgG (1:3000; Amersham Pharmacia, Piscataway, New Jersey) in PBS. Signals were detected using the SuperSignal WestPico chemiluminescent kit (Pierce), according to manufacturer's specifications. Protein membranes were stripped using Restore Western blot stripping solution (Pierce), according to the manufacturer's specifications. Membranes were then blocked for 1 hour and probed with anti-MAPK8/MAPK9 (JNK1/JNK2; 1:1000; Sigma-Aldrich) and anti–β-actin (1:1000; Sigma-Aldrich) for 1 hour in blocking solution, followed by washing and subsequent incubation for 30 minutes at room temperature with anti-rabbit HRP-linked IgG (1:3000; Amersham Pharmacia) for anti-MAPK8/MAPK9 or anti-mouse HRP-linked IgG (1:3000) for anti-β-actin. The relative ratios of p-MAPK8 to total MAPK8 normalized to the amount of β-actin in the samples were quantified by determining signal density using Scion Image 4.0.2 (Scion Corp, Frederick, Maryland).
Freshly isolated germ cells or germ cells cultured for 8 hours without Sertoli cells were dried to microscope slides and fixed with neutral buffered formalin immediately following determination of cell purity. Slides were blocked in PBS-diluted normal horse serum (1:60) and then incubated with anti-mouse p-MAPK8 (1:50) and anti-rabbit BMF (1:200) overnight at 4°C. Slides were washed 3 times in PBS for 5 minutes each. Signal intensity for p-MAPK8 was enhanced by incubating the slides with an anti-mouse biotin secondary antibody (Vector Laboratories) in PBS for 1 hour at room temperature. Bound primary antibodies were detected with a FITC-conjugated anti-rabbit IgM secondary antibody (1:100) and Texas Red—conjugated avidin (1:100; Vector Laboratories). Nuclei were stained with Vectashield anti-fade mounting medium containing DAPI. Images were obtained as above using a Nikon Eclipse 800 microscope and Princeton Instruments 5-Mhz cooled CCd camera.
Data are expressed as means ± SEM. Group mean differences were determined by 1-way analysis of variance (ANOVA). If group differences were revealed by ANOVA (<.05), differences between individual groups were determined using Scheffe's least-significant-difference test. Means were considered significantly different at P < .05.
Immunolocalization of Active Form of MAPK8 in Rat Seminiferous Epithelium
Figures 1 and 2 illustrate the distribution of BMF and p-MAPK8, respectively, in sections of stage VI and VIII rat seminiferous tubules, respectively. BMF protein (green stain; Figure 1) localized to the subacrosomal space of the round spermatids (arrows) and was not seen in either spermatocytes (arrowheads) or elongated spermatids (*). In contrast to BMF, p-MAPK8 (green stain; Figure 2) localized to spermatocytes (arrowheads) and elongated spermatids (*) but was not seen in round spermatids (arrows) at stage VIII. Indeed, at all stages, spermatocytes and elongating spermatids stained for p-MAPK8, whereas round spermatids did not (not shown). This was in striking contrast to the staining pattern seen for BMF, for which expression was seen only in spermatids (Show et al, 2004).
Both MAPK8 (46 kDa) and MAPK9 (50 kDa) were identified in isolated tubules by Western blot analysis, but only MAPK8 was phosphorylated (Figure 3A). The ratio of p-MAPK8 to total MAPK8 was significantly greater in stages VII and VIII and IX to XIV than in stages I to V (Figure 3B), consistent with the results of immunohistochemical analysis (not shown).
Apoptosis and MAPK8 Activation in Cultured Germ Cells
Germ cells were isolated via enzymatic digestion, mechanical dispersion, and filtration to a purity of approximately 90%. The isolated germ cells then were cultured at 34°C, in the absence of Sertoli cell support, in media supplemented with lactate and pyruvate. Apoptosis of freshly isolated or cultured germ cells was measured by annexin V-FITC staining, which is an early morphologic marker for cells undergoing programmed cell death (van Engeland et al, 1996). In germ cells cultured for 4 and 8 hours, increases of 2.5- and 3.1-fold, respectively, were seen in the percentages of annexin V—positive cells relative to total cells (Figure 4). Cells cultured for 12 hours were extremely fragile and also exhibited a high percentage of annexin V—positive cells.
MAPK8 activation was examined by Western blot analysis of freshly isolated (control) or cultured (4 or 8 hours) germ cells (Figure 5). The ratio of p-MAPK8 to total MAPK8 normalized to total protein (β-actin) increased significantly in isolated germ cells cultured for 4 (3.2-fold) or 8 (2.2-fold) hours. Immunocytochemical analysis of isolated germ cells was performed to examine the germ cell type(s) in which the increases in MAPK8 activation occurred. As shown in Figures 1 and 2, the round spermatids from control testes exhibited subacrosomal BMF staining and little or no p-MAPK8 staining; spermatocytes showed little or no BMF and little p-MAPK8 staining. However, after 8 hours of culture, BMF staining was redistributed throughout the round spermatids, and this redistribution was associated with an increase in p-MAPK8 staining (Figure 6). Increases in the intensity of p-MAPK8 and BMF staining also were apparent in spermatocytes cultured for 8 hours (not shown).
We showed herein that BMF and p-MAPK8 are both expressed in germ cells of the normal rat testis; BMF protein localized to the subacrosomal space of the round spermatids and was not seen in either spermatocytes or elongated spermatids, whereas p-MAPK8 localized to spermatocytes and elongated spermatids and was not seen in round spermatids. As in the mouse, MAPK8 also localized to arterioles and venules (Lysiak et al, 2003). However, in contrast to the rat, p-MAPK8 was not seen in mouse germ cells (Lysiak et al, 2003). One possible explanation for the different results is the species difference between our and the Lysiak et al (2003) study (rat vs mouse). Another is our use of polyester wax for tissue embedment, which has the advantages of increased antigenicity because embedment is accomplished at relatively low temperature (Oke and Suarez-Quian, 1993) and of solvents such as xylene not being required (Suarez-Quian et al, 1992; Oke and Suarez-Quian, 1993).
When rat germ cells were cultured in the absence of Sertoli cells, BMF staining became distributed throughout the round spermatids rather than restricted to the subacrosomal space; associated with this, there was increased p-MAPK8 staining in these cells. Increases in the intensity of p-MAPK8 and BMF staining also occurred in spermatocytes in response to cultured cells in the absence of Sertoli cells. The mechanisms that explain these increases remain unknown. Culturing the cells resulted in increased levels of apoptosis, as well as in increased levels of p-MAPK8 and redistribution of BMF. It was shown previously that the increased germ cell apoptosis (deoxynucleotidyl transferase—mediated dUTP nick end labeling—positive cells) seen after vasectomy also was correlated with increases in p-MAPK8 levels in these cells (Shiraishi et al, 2001, 2002). From these results, it is tempting to speculate that the separation of germ cells from Sertoli cells might lead to the activation of MAPK8 and that this, in turn, might lead to the phosphorylation of BMF and to apoptosis of germ cells. It also is tempting to speculate that a similar sequence of events might account for germ cell apoptosis when germ cells slough from Sertoli cells in response to hypogonadism, testicular torsion, vasectomy, or other testicular insults. However, it is not clear that this temporal sequence occurs in response to any of these conditions. For example, it is possible that one or more of these conditions leads to biochemical changes that result in sloughing rather than the other way around. Clearly, cause-effect relationships are yet to be established.
Moreover, the sequence of events may be considerably more complex than described above. Thus, the MAPK superfamily is subdivided into 2 groups: the extracellular regulated kinases (Erk), which are activated by mitogens; and the stress-activated protein kinases/JNK, which are activated in response to various stimuli associated with the induction of apoptosis (Strniskova et al, 2002). Mixed-lineage kinase 2 (MLK2) is a MAPK kinase kinase that, when activated, can lead to the activation of MAPK8 (Dorow et al, 1995; Hirai et al, 1997). Phelan et al (1999) demonstrated that MLK2 is present in germ cells throughout all stages of spermatogenesis. It is possible that in response to culturing germ cells in the absence of Sertoli cells, the observed increase in MAPK8 phosphorylation seen in apoptotic germ cells might be due to activation of MLK2. Further studies will be needed to elucidate the role of MLK2 and MAPK8 in the initiation of germ cell apoptosis as well as to define the events that occur upstream of MAPK8 activation. Downstream events also are complex. BMF is a member of the “BH3-only” BCL2 family of proteins (Puthalakath and Strasser, 2002), each member of which is capable of detecting a specific apoptotic stimulus and transmitting this stimulus to downstream effectors to initiate programmed cell death. For example, in the case of some healthy cells, BMF is sequestered to myosin V motors by association with dynein light chain 2. Following certain damage signals, including loss of cell attachment, BMF translocates and binds prosurvival Bcl-2 proteins and thus causes cell death (Puthalakath and Strasser, 2002).
Clearly, BMF has an important, if not exclusive, role in ensuring that germ cells that lose attachment to the Sertoli cells will be eliminated by apoptosis and thus is involved in the quality control mechanism that ensure the completion of proper spermatogenesis. Further studies will be needed to determine the mechanism for increase in MAPK8 activation in relationship to the loss of adhesion, phosphorylation of BMF in relationship to the activation of MAPK8, and increased apoptosis in relationship to these events.
This work was supported by National Institutes of Health grants HD44258 and U54 HD55740.