Actin dynamics regulate subcellular localization of the F-actin-binding protein PALLD in mouse Sertoli cells

in Reproduction

Sertoli cells undergo terminal differentiation at puberty to support all phases of germ cell development, which occurs in the mouse beginning in the second week of life. By ∼18 days postpartum (dpp), nearly all Sertoli cells have ceased proliferation. This terminal differentiation is accompanied by the development of unique and regionally concentrated filamentous actin (F-actin) structures at the basal and apical aspects of the seminiferous epithelium, and this reorganization is likely to involve the action of actin-binding proteins. Palladin (PALLD) is a widely expressed F-actin-binding and bundling protein recently shown to regulate these structures, yet it is predominantly nuclear in Sertoli cells at puberty. We found that PALLD localized within nuclei of primary Sertoli cells grown in serum-free media but relocalized to the cytoplasm upon serum stimulation. We utilized this system with in vivo relevance to Sertoli cell development to investigate mechanisms regulating nuclear localization of this F-actin-binding protein. Our results indicate that PALLD can be shuttled from the nucleus to the cytoplasm, and that this relocalization occurred following depolymerization of the F-actin cytoskeleton in response to cAMP signaling. Nuclear localization was reduced in Hpg-mutant testes, suggesting the involvement of gonadotropin signaling. We found that PALLD nuclear localization was unaffected in testis tissues from LH receptor and androgen receptor-mutant mice. However, PALLD nuclear localization was reduced in the testes of FSH receptor-mutant mice, suggesting that FSH signaling during Sertoli cell maturation regulates this subcellular localization.

Abstract

Sertoli cells undergo terminal differentiation at puberty to support all phases of germ cell development, which occurs in the mouse beginning in the second week of life. By ∼18 days postpartum (dpp), nearly all Sertoli cells have ceased proliferation. This terminal differentiation is accompanied by the development of unique and regionally concentrated filamentous actin (F-actin) structures at the basal and apical aspects of the seminiferous epithelium, and this reorganization is likely to involve the action of actin-binding proteins. Palladin (PALLD) is a widely expressed F-actin-binding and bundling protein recently shown to regulate these structures, yet it is predominantly nuclear in Sertoli cells at puberty. We found that PALLD localized within nuclei of primary Sertoli cells grown in serum-free media but relocalized to the cytoplasm upon serum stimulation. We utilized this system with in vivo relevance to Sertoli cell development to investigate mechanisms regulating nuclear localization of this F-actin-binding protein. Our results indicate that PALLD can be shuttled from the nucleus to the cytoplasm, and that this relocalization occurred following depolymerization of the F-actin cytoskeleton in response to cAMP signaling. Nuclear localization was reduced in Hpg-mutant testes, suggesting the involvement of gonadotropin signaling. We found that PALLD nuclear localization was unaffected in testis tissues from LH receptor and androgen receptor-mutant mice. However, PALLD nuclear localization was reduced in the testes of FSH receptor-mutant mice, suggesting that FSH signaling during Sertoli cell maturation regulates this subcellular localization.

Introduction

Sertoli cells reside within the seminiferous epithelium and play critical roles in germ cell development. These include support of the spermatogonial stem cell niche, maintenance of the blood–testis barrier (BTB), remodeling of spermatid nuclei, reduction of spermatid cytoplasm, and release of spermatids into the tubular lumen. During the second week of life in the mouse, Sertoli cells respond to hormones including follicle-stimulating hormone (FSH), testosterone, and thyroid hormone to mature into an adult appearance (Walker 2003, Walker & Cheng 2005). In particular, they stop dividing, become polarized, and span the epithelium to support all phases of germ cell development. These dramatic changes in the Sertoli cell cytoskeleton accompany terminal differentiation, with regional concentrations of filamentous actin (F-actin) at dynamic structures termed the basal and apical ectoplasmic specializations (ESs) respectively (shown in Supplementary Figure S1, see section on supplementary data given at the end of this article, thoroughly reviewed in Vogl et al. (2008) and Lie et al. (2010)). The basal ES is a homotypic tight junction between adjacent the Sertoli cells forming the BTB that partitions the seminiferous epithelium into two distinct compartments, the basal and adluminal, and it must restructure during stages VIII–IX to allow for the translocation of preleptotene spermatocytes into the adluminal compartment to enter meiosis (Russell 1977). By contrast, the apical ES is a heterotypic adherens junction between Sertoli cells and associated elongating and condensing spermatids, and serves to anchor the spermatid until spermiation (Ross 1976, Romrell & Ross 1979).

After many years of controversy, it is now accepted that the nucleus contains abundant G-actin, although F-actin has not been detected (Gettemans et al. 2005, Pederson 2008, Zheng et al. 2009, Visa & Percipalle 2010, de Lanerolle & Serebryannyy 2011, Oma & Harata 2011). However, a number of F-actin-binding proteins are found in nuclei. Although their exact function(s) remain unclear, proposed roles include regulation of mRNA processing and transport, chromatin structure, and transcription (Ozanne et al. 2000, Loy et al. 2003, Archer et al. 2005, Prante et al. 2008, Zheng et al. 2009). The Palladin (PALLD) gene encodes a highly conserved F-actin-bundling protein that is both widely expressed and essential for development (Parast & Otey 2000, Luo et al. 2005, Otey et al. 2009). We previously showed that PALLD relocates from the cytoplasm to nuclei of Sertoli cells as they undergo terminal differentiation (Niedenberger et al. 2013). This predominant nuclear localization was also indicated in a report published at the same time (see Fig. 1D in Qian et al. (2013)). In that study, the role of PALLD in Sertoli cells was assessed by siRNA-mediated knockdown (KD) in vitro. The most obvious phenotype following PALLD KD in vitro was a loss of organization of the F-actin cytoskeleton, with multiple short fragments scattered throughout the cytosol (Qian et al. 2013). In vivo treatment with adjudin, a specific disrupter of anchoring junctions in the apical ES causing infertility (Cheng et al. 2001, Mok et al. 2011), resulted in loss of PALLD signal at the apical ES by 48 h (Qian et al. 2013).

In this study, we utilized primary Sertoli cells in culture as a model to investigate the mechanisms regulating PALLD nuclear localization that occurs as Sertoli cells mature during puberty. In serum-free culture, PALLD remained nuclear, but the addition of serum resulted in a dramatic relocation to the cytoplasm. We found that PALLD nuclear localization is regulated by increased cAMP levels through Rho signaling and responds to the modulation of the G:F-actin ratios. We also found that localization is modified in vivo by perturbation of gonadotropin signaling. Taken together, these results provide a functional link between hormonal signaling and localization of a critical actin-bundling protein that regulates cytoskeletal dynamics in a wide variety of tissues and cell types. In addition, this provides the first example we are aware of where localization of an F-actin binding protein is determined by changes in polymerization of the actin cytoskeleton.

Materials and methods

Primary Sertoli cell culture

The Institutional Animal Care and Use Committee at East Carolina University approved all procedures involving mice (AUP# A178). CD1 mice were killed at 10 and 18 days postpartum (dpp) and Sertoli cells isolated essentially as previously described (Anway et al. 2003), without hypotonic shock treatment. After 2 days in culture, Sertoli cells represented ≥95% of adherent junctions of cells. The cells were cultured up to 4 days on coverslips in 50:50 DMEM:Ham's F-12 (Cellgro-Mediatech, Manassas, VA, USA) with or without 10% FBS (ATCC, Manassas, VA, USA)-containing 1× penicillin–streptomycin at 33 °C in 95% air with 5% CO2. The primary cell cultures were treated with the following compounds at least twice with the following (final concentration): 10 μM cytochalasin D (A2618, Sigma–Aldrich), 2.5 μM latrunculin B (L5288, Sigma–Aldrich), 100 μM 8-bromo-cAMP (1140, Tocris Bioscience, Bristol, UK), 10 μM forskolin (BML-CN100, Enzo Biochem, New York, NY, USA), 1 μg/ml C3 transferase (CT104, Cytoskeleton, Inc., Denver, CO, USA), 50 nM leptomycin B (L2913, Sigma–Aldrich), and 0.1 U/ml FSH (F2293, Sigma–Aldrich).

Indirect immunofluorescence on coverslips

The cells grown on coverslips were fixed for 1 h in 4% PFA. Blocking and antibody dilutions were carried out in PBS+0.1% Triton X-100-containing 3% BSA, pH 7.3. C-terminal anti-PALLD (1:200, ProteinTech Group, Inc., Chicago, IL, USA) or N-terminal PALLD (1:1000, Rachlin & Otey 2006) was added and incubated overnight at 4 °C. Alexa Fluor-488 anti-rabbit IgG (1:500, Life Technologies) was added with Alexa Fluor-594 phalloidin (1:1000, Life Technologies) for 1 h at room temperature. The coverslips were mounted with Vectastain-containing DAPI (Vector Laboratories, Burlingame, CA, USA) and images acquired using a Nikon E600 fluorescence microscope with an Orca II CCD camera (Hamamatsu Corp., Middlesex, NJ, USA) or a Fluoview FV1000 confocal laser scanning microscope (Olympus America). DAPI staining was used to visually determine the position of nuclei and PALLD localization. If PALLD staining was of greater intensity within the area stained by DAPI than outside, PALLD localization was determined to be nuclear. If PALLD staining was of lesser or similar intensity within the area stained by DAPI than outside, PALLD localization was determined to be cytoplasmic. For each treatment, when PALLD nuclear localization was determined, it was observed in >85% of Sertoli cells.

Indirect immunofluorescence on frozen sections

The frozen sections were prepared as described previously (Niedenberger et al. 2013) from mouse testes. PBS+0.1% Triton X-100 containing 3% BSA of pH 7.3 was used for both blocking and antibody dilution. After blocking for 1 h, N-terminal PALLD (1:1000, Rachlin & Otey 2006) was applied to sections. Alexa Fluor-488 anti-rabbit IgG (1:500, Life Technologies) was added with Alexa Fluor-594 phalloidin (1:1000, Life Technologies) for 1 h at room temperature. The images were acquired using a Fluoview FV1000 confocal laser scanning microscope (Olympus America).

Immunohistochemistry

The testes were fixed in 4% PFA overnight at 4 °C and then processed using standard methods. Immunohistochemistry (IHC) was carried out on 5 μm sections as described previously (Niedenberger et al. 2013). Blocking and antibody dilutions were done in 0.05 M Tris–HCl, 0.15 M NaCl, 0.02% Triton-X-100+1% BSA, pH 7.6. The antibodies were against PALLD C-terminus (1:500, ProteinTech Group, Inc.) and AR (#sc-816, 1:500, Santa Cruz Biotechnology, Inc.). The sections were imaged using a Zeiss Axio Observer A1 (Carl Zeiss Microscopy LLC, Thornwood, NY, USA) and images captured using a Dage XL16C digital camera and Dage Exponent software (Dage-MTI, Michigan City, IN, USA).

RNA isolation and quantitative (q)RT-PCR

Total RNA was isolated from the tissues using Trizol (Life Technologies), according to the manufacturer's instructions. Briefly, 100 ng of RNA from cultured cells were reverse transcribed and amplified in triplicate using the iScript One-Step RT-PCR Kit with SYBR green on an iQ5 real-time PCR detection system (Bio-Rad Laboratories). The primer pairs were designed to amplify across an intron/exon junction to avoid amplification of genomic DNA. Primer sequences are listed in Table 1.

Table 1

Primer sequences.

GeneAccessionForward (5′→3′)Reverse (5′→3′)
PalldJX477684TACAGAGCGTATTTGGTGCCCAGTTCAAAGGCCTCAGCAATGACCAAG
TrfNM_133977TAGACAGAACCGCTGGTTGGAACATTGAGTGGGCCAATACACAGGTCA
FosNM_010234AGAGAAACGGAGAATCCGAAGGGAATTGGCAATCTCAGTCTGCAACGC
VclNM_009502ACAGTGGCAGAGGTAGTGGAAACTTCACCAACATCACACGGTGTTCCT
B2mNM_009735CCGTGATCTTTCTGGTGCTTCGTAGCAGTTCAGTATGTTCG
ActbNM_007393TCCGATGCCCTGAGGCTCTTTTCCTTGCTGATCCACATCTGCTGGAA

Steady-state mRNA levels for commonly used housekeeping genes such as glyceraldehyde 3-phosphate dehydrogenase (Gapdh) and beta actin (Actb) increased in response to serum (Schmittgen & Zakrajsek 2000); therefore Ct values were normalized to beta 2 microglobulin (B2m), which changed the least in response to serum. Relative mRNA levels were calculated with the delta–delta Ct (ddCt) method. Student's t-test was used for all statistical analyses, and significant differences are indicated with an asterisk at P<0.05).

Immunoblot analysis

Fifteen microgram of protein were separated on a 10% SDS–polyacrylamide gel before immunoblot analysis using standard methods as described previously (Niedenberger et al. 2013). The gels were blotted onto PVDF membranes and blocked in 5% nonfat milk in PBS-T. The antibodies used were against C-terminal PALLD (1:1000, ProteinTech Group, Inc.), N-terminal PALLD, (1:1000, Rachlin & Otey 2006), and ACTB (1:1000, A2066, Sigma).

Results

PALLD localization is dynamic in primary Sertoli cells

We first isolated primary Sertoli cells from the testes at 10 dpp (when PALLD is cytoplasmic) and 18 dpp (when PALLD is nuclear) and maintained them in serum-free culture for up to 4 days to determine whether PALLD retained proper in vivo localization (Fig. 1A). We carried out indirect immunofluorescence (IIF) and found that PALLD localization was predominantly cytoplasmic at 10 dpp (Fig. 1B, Supplementary Figure S2A, see section on supplementary data given at the end of this article), but was mostly nuclear in Sertoli cells isolated from 18 dpp mice (Fig. 1C, Supplementary Figure S2B), as was seen in vivo (Niedenberger et al. 2013). It was previously shown that Sertoli cell morphology and cytoskeletal organization were modified in vitro by the addition of serum to culture media (Hutson et al. 1980). We tested whether the addition of serum would affect PALLD localization in 18 dpp Sertoli cells and found that PALLD relocated to the cytoplasm in the presence of 10% serum (Fig. 1D, Supplementary Figure S2C). Conversely, in 10 dpp Sertoli cells, PALLD localization was not significantly affected by the addition of 10% serum. The addition of as little as 1% serum was sufficient to direct PALLD cytoplasmic localization (data not shown). To determine whether removal or addition of serum would change PALLD localization once it was established, we maintained 18 dpp Sertoli cells in serum-free medium for 3 days, then serum-stimulated for 24 h, and found that PALLD relocated from the nucleus to the cytoplasm. We did not, however, observe PALLD nuclear localization in cells that were maintained with serum for 3 days and then serum-starved for 24 h (data not shown). Therefore, in the remainder of this report, we employed primary Sertoli cells isolated from 18 dpp mice to investigate the mechanisms regulating the nuclear and cytoplasmic localization of PALLD and gross changes in the F-actin cytoskeleton were monitored by phalloidin staining (Supplementary Figures S3, S4 and S5, see section on supplementary data given at the end of this article).

Figure 1

Download Figure

Figure 1

PALLD localization in cultured primary Sertoli cells. Treatments with various inhibitors are described in the Materials and Methods section. (A) Timeline for primary Sertoli cell culture experiments. (B, C, and D) IIF to detect PALLD (in green) in primary Sertoli cells from mice aged 10 dpp (B) and 18 dpp (D and C) in the absence (B and C) or presence (D) of serum. Phalloidin-labeled F-actin cytoskeleton (in red). (E and F) Levels of Palld mRNA (E) and protein (F) were slightly elevated in Sertoli cells grown in the presence of serum (P<0.0271). Scale bars=25 μm.

Citation: REPRODUCTION 148, 4; 10.1530/REP-14-0147

The addition of serum elevates Palld mRNA and protein levels

We investigated whether increased expression of PALLD might underlie its dramatic nuclear relocalization. We found that the addition of serum caused a small but statistically significant increase (P=0.027) in mRNA levels for the 140 kDa Palld isoform (Fig. 1E), which is the predominant isoform in Sertoli cells (Niedenberger et al. 2013). This was accompanied by a small increase in the level of PALLD protein (Fig. 1F). The addition of serum to cultured cells has been reported to increase the abundance of most housekeeping mRNAs, including Actb and Gapdh (Schmittgen & Zakrajsek 2000). In accordance with this, we saw increased levels for beta actin (Actb)=+1.8-fold (P=0.004). In contrast, mRNAs encoding markers of differentiated Sertoli cells showed no change (Vcl, P=0.273) or decreased in response to serum, including FBJ osteosarcoma oncogene (Fos)=−3.4-fold (P=0.0003), and transferrin (Trf)=−3.8-fold (P=0.000001) (Chaudhary et al. 1996, Gineitis & Treisman 2001; Fig. 2).

Figure 2

Download Figure

Figure 2

Gene expression changes in primary Sertoli cells in response to serum. Message levels were quantitated by qRT-PCR for the indicated genes (Vcl, vinculin; Actb, beta actin; Fos, c-FBJ osteosarcoma oncogene; Trf, transferrin). Statistical significance is indicated by an asterisk, with associated P values.

Citation: REPRODUCTION 148, 4; 10.1530/REP-14-0147

PALLD localization is dependent upon XPO1

A leucine-rich (LxLxxxL) putative nuclear export signal (NES) resides at the extreme C-terminus of PALLD (la Cour et al. 2004). The NES of target proteins is bound by exportin 1 (XPO1, previously CRM1), which directs their exit from the nucleus (Kutay & Guttinger 2005). We previously found that deletion of C-terminal sequence containing the NES resulted in nuclear localization of GFP-PALLD expression constructs in the TM4 Sertoli cell line (Niedenberger et al. 2013). We tested whether shuttling of endogenous PALLD occurred in primary Sertoli cells by treating serum-stimulated cells with leptomycin B, a specific inhibitor of XPO1. As early as 1 h after treatment with leptomycin B, PALLD became predominantly nuclear (Fig. 3B), and this was enhanced by 7 h (Fig. 3C).

Figure 3

Download Figure

Figure 3

PALLD accumulates in nuclei after inhibition of nuclear export. (A, B, and C) IIF to detect PALLD (in green) in serum-stimulated Sertoli cells grown without (A) or with leptomycin B treatment for 1 h (B) and 7 h (C). Phalloidin-labeled F-actin cytoskeleton (in red). Scale bars=25 μm.

Citation: REPRODUCTION 148, 4; 10.1530/REP-14-0147

Depolymerization of the F-actin cytoskeleton results in PALLD nuclear localization

Addition of serum to cells in culture stimulates polymerization of the F-actin cytoskeleton, including in primary Sertoli cells (Hutson et al. 1980). Accordingly, the phalloidin-labeled Sertoli F-actin cytoskeleton appeared less robust in serum-free culture. We treated primary Sertoli cells with cytochalasin D and latrunculin B, two drugs which cause rapid depolymerization of the F-actin cytoskeleton (Brenner & Korn 1979, Brown & Spudich 1979, Lin & Lin 1979, Flanagan & Lin 1980, MacLean-Fletcher & Pollard 1980, Spector et al. 1983). In Sertoli cells maintained in serum-containing medium, PALLD rapidly relocalized to nuclei following the addition of cytochalasin D (Fig. 4B) or latrunculin B (Fig. 4C), coinciding with breakdown of the F-actin cytoskeleton as assessed by phalloidin staining.

Figure 4

Download Figure

Figure 4

PALLD accumulates in nuclei in response to increased cAMP and after disruption of the F-actin cytoskeleton. (A, B, and C) IIF to detect PALLD (in green) in serum-stimulated Sertoli cells grown without (A) or with cytochalasin D (B), latrunculin (C), 8-bromo-cAMP (D), forskolin (E), or C3 transferase (F). Phalloidin-labeled F-actin cytoskeleton (in red). Scale bar=25 μm.

Citation: REPRODUCTION 148, 4; 10.1530/REP-14-0147

cAMP and Rho signaling regulate PALLD nuclear localization in vitro

The Sertoli cell cytoskeleton undergoes substantial changes during the prepubertal period, with the creation of unique and highly specialized actin-rich basal and apical ES structures (Mruk & Cheng 2010). Addition of cAMP to Sertoli cells in culture profoundly affected cellular morphology and the cytoskeleton (Tung & Fritz 1975, Tung et al. 1975, Hutson 1978, Hutson et al. 1980), and the addition of serum blocked these effects (Tung et al. 1975). Therefore, we tested whether localization of PALLD, as a dominant modulator of the F-actin cytoskeleton, was affected in response to elevated cAMP levels using two approaches. We treated serum-stimulated Sertoli cells with 8-bromo-cAMP, a stable analog resistant to phosphodiesterase degradation, and forskolin, which stimulate adenylyl cyclase activity. PALLD relocates to the nucleus within 24 h in serum-stimulated Sertoli cells in response to both 8-bromo-cAMP (Fig. 4D) and forskolin (Fig. 4E). Serum stimulation drives actin polymerization by activating Rho signaling (Van Aelst & D'Souza-Schorey 1997, Bishop & Hall 2000, Hall 2012), and Rho-induced actin polymerization results in relocalization of MKL1 to the nucleus (Miralles et al. 2003). We therefore tested whether inactivation of Rho signaling directed PALLD nuclear localization. We treated serum-stimulated primary Sertoli cells with the specific Rho inhibitor C3-transferase, and PALLD relocated to the nucleus within 24 h (Fig. 4F).

Gonadotropin signaling regulates PALLD nuclear localization in vivo

Several studies uncovered the roles of hormone signaling in remodeling the Sertoli cell cytoskeleton in vivo (Muffly et al. 1993, 1994, Allan & Handelsman 2005). We tested whether PALLD nuclear localization depended upon gonadotropin signaling in vivo in Hpg-mutant mice, which have a spontaneous mutation in the gonadotropin-releasing hormone (Gnrh1) gene. Male Hpg mice are infertile, with dramatically reduced levels of FSH and luteinizing hormone (LH) and small testes (Cattanach et al. 1977, Mason et al. 1986). In contrast to the robust nuclear PALLD localization that we found in adult WT testes (Fig. 5A, Niedenberger et al. 2013), only a subset of immature-appearing Sertoli cells in Hpg mutants displayed faint nuclear PALLD staining, with many having no detectable PALLD (Fig. 5B and C). As GNRH1 stimulates the anterior pituitary to release LH and FSH, this suggests that impaired signaling through LH or FSH decreased nuclear localization in vivo.

Figure 5

Download Figure

Figure 5

PALLD nuclear localization is disrupted in response to aberrant gonadotropin signaling in vivo. IHC was used to detect PALLD (brown), and sections were counterstained with hematoxylin (blue). PALLD in adult testis sections from WT (A), Hpg mutant (B and C), SCARKO (D and E), or Fshb KO (F) mice. Arrows indicate unstained Sertoli cell nuclei in panels C–D. Scale bars=25 μm.

Citation: REPRODUCTION 148, 4; 10.1530/REP-14-0147

PALLD nuclear localization is not dependent upon AR, but is reduced in Fshb KO testes

We tested the hypothesis that nuclear translocation of AR (downstream of testosterone production by LH) was required for PALLD localization within Sertoli cell nuclei using testes from Sertoli cell Ar knockout (SCARKO) mice (De Gendt et al. 2004, Zhou et al. 2011). As expected, AR was not detectable, although it was present in interstitial Leydig cells (Fig. 5D). In contrast, PALLD was present in all Sertoli cell nuclei in adult SCARKO mice (Fig. 5E), indicating that the loss of testosterone signaling through AR did not affect nuclear localization of PALLD. However, it is possible that translocation of AR depends on PALLD, rather than the other way around as we originally predicted.

cAMP is one of the most well-studied molecules induced by FSH during pubertal Sertoli cell differentiation in vivo (Walker & Cheng 2005), which is the time period during which PALLD localization changes from cytoplasmic to nuclear (Niedenberger et al. 2013). Therefore, we determined whether nuclear localization was affected in the absence of FSH signaling. We carried out IHC with anti-PALLD on sections from Fshb KO mice (Kumar et al. 1997). Nuclear PALLD was reduced, with localization around the periphery of the nucleus (Fig. 5E and F). We tested whether FSH would direct PALLD nuclear localization in primary Sertoli cells from 10 dpp mice (when it is cytoplasmic). We found that the addition of 0.1 U/ml FSH for 24 h did not affect PALLD nuclear localization, suggesting that other mechanisms are preventing PALLD nuclear localization in immature Sertoli cells.

Discussion

In this study, we demonstrate that nuclear localization of the essential F-actin bundling protein PALLD is regulated by changes in the polymerization status of the actin cytoskeleton. Our results suggest that FSH signaling modulates the nuclear localization of the actin binding protein PALLD in Sertoli cells, and this relationship provides a functional link between hormone signaling, intracellular signaling downstream of cAMP, and Sertoli cell cytoskeletal changes during puberty.

It was shown many years ago that addition of serum to primary Sertoli cells resulted in dramatic changes in morphology and cytoskeletal organization (Hutson et al. 1980), and we found that it also caused a dramatic relocalization of PALLD from the nucleus to the cytoplasm. Serum contains a complex but poorly defined mixture of macromolecules, nutrients, hormones, and growth factors. Therefore, instead of trying to identify responsible component(s), we focused on the fact that serum induces polymerization of the F-actin cytoskeleton. We tested the hypothesis that PALLD relocalization was regulated by the polymerization status of the actin cytoskeleton and showed that depolymerization by latrunculin B and cytochalasin D induced rapid PALLD nuclear localization. Therefore, our results support the model that PALLD is in the cytoplasm in cells with abundant F-actin, and in the nucleus with low F-actin levels. Three positively charged amino acid residues in the third immunoglobulin domain of PALLD were found to direct its F-actin binding and bundling capabilities (Dixon et al. 2008, Beck et al. 2013). Although WT recombinant PALLD was detected in the cytoplasm, mutation of these residues (K to A) resulted in dramatic nuclear localization of mutant recombinant PALLD (Beck et al. 2013). Taken together, these data indicate that, when not bound to F-actin, PALLD migrates into the nucleus. This localization paradigm provides a counterexample to the actin polymerization-directed nuclear localization of the G-actin binding-transcription factors MKL1 (previously MAL and MRTF-A) and MKL2 (previously MRTFB). As the actin cytoskeleton polymerizes, there is reduced G-actin to retain MKL1 and MKL2 in the cytoplasm, and therefore they shuttle into the nucleus (Wang et al. 2002, Miralles et al. 2003, Olson & Nordheim 2010), where they activate transcription in association with serum response factor (Wang et al. 2002, Miralles et al. 2003).

These elegant studies involving MKL1 and MKL2 demonstrate that changes in cytoplasmic G:F-actin result in their nuclear localization to ultimately regulate transcription. This provides a mechanism by which the cell can sense and respond to cytoplasmic changes at the level of gene expression. We envision a similar scenario for PALLD, whose localization is regulated in a reciprocal manner. There is precedence for a role of PALLD in transcription, as other studies have shown that PALLD also partially localizes to the nucleus in certain cell lines (Endlich et al. 2009, Goicoechea et al. 2010, Jin et al. 2010), where it can bind to transcriptional regulators that influence the patterns of gene expression.

However, others and we have shown that PALLD is present in both Sertoli cell nuclei and the cytoplasmic F-actin-rich structures that are alternately formed and disassembled at precise points during the cycle of the seminiferous epithelium (Niedenberger et al. 2013, Qian et al. 2013). The change in the intracellular localization of PALLD temporally accompanies the development of the ES structures described earlier, suggesting an important role for PALLD in F-actin cytoskeletal rearrangements during Sertoli cell development. It likely that the appearance of specialized F-actin cytoskeletal structures (BTB at puberty, and apical ES later) shifts cytoplasmic F:G-actin ratios, resulting in the redirection of PALLD into the nucleus. Alternatively, this shuttling mechanism may function to reduce cytoplasmic levels of PALLD, a dominant regulator of the cytoskeleton, as the F-actin cytoskeleton changes during Sertoli cell development. It has been shown that transfection of excess PALLD into cells results in dramatic changes in the cytoskeleton, with the formation of robust cable-like and stellate actin arrays (Rachlin & Otey 2006). Therefore, PALLD may play multiple roles in Sertoli cells by regulating transcription and ES formation, and may also provide a signal to the nucleus as to the status of the F-actin cytoskeleton.

We found that PALLD nuclear localization was impaired in Sertoli cells in Hpg mice, which indicates that it was at least indirectly affected by either LH or FSH signaling. The major role of LH in the testis is to direct the production of testosterone by interstitial Leydig cells. In Sertoli cells, testosterone binding to the androgen receptor (AR) mediates its nuclear translocation (Bremner et al. 1994), which occurs in a similar developmental pattern to PALLD (Niedenberger et al. 2013). As mentioned previously, this nuclear translocation was facilitated through an interaction with another F-actin-binding protein, filamin A (Ozanne et al. 2000, Loy et al. 2003). Therefore, this association seemed to be a likely mechanism for directing PALLD nuclear localization. However, we did not see any difference in SCARKO testes, indicating that PALLD localization does not depend upon AR, although the converse may be true. Our results do, however, suggest a role for FSH signaling in the nuclear localization of PALLD, and there have been a number of reports supporting a link between FSH signaling and cytoskeletal regulation. Cultured primary Sertoli cells from 20 dpp rats undergo significant morphological changes in response to FSH, but not other hormones such as insulin, LH, growth hormone, testosterone, or estradiol (Hutson 1978). FSH signaling is required for qualitatively and quantitatively normal fertility (Allan et al. 2004, Walker & Cheng 2005), and it stimulates adenylate cyclase in Sertoli cells to increase intracellular cAMP levels (Murad et al. 1969, Kuehl et al. 1970, Dorrington et al. 1972, Means & Vaitukaitis 1972, Bhalla & Reichert 1974, Fritz et al. 1975, Tung et al. 1975). The addition of FSH to rats receiving GNRH immunogen (reducing testosterone and FSH) increased actin-containing ES and adherens junctions (Sluka et al. 2006). FSH also induced elongated cytoplasmic extensions resembling the arbor-like cytoplasmic processes of differentiated Sertoli cells in vivo (Tung & Fritz 1975, Tung et al. 1975, Hutson 1978, Hutson & Stocco 1978), and it regulated proper localization of actin, espin, and vimentin (Sluka et al. 2006). We also found that increased cAMP resulted in robust nuclear localization of PALLD, and this effect was likely mediated by Rho activation.

In conclusion, the differentiating Sertoli cell population provides an excellent model system to investigate the functions of nuclear-localized actin-binding proteins such as PALLD in a relevant developmental context. We have used a combination of primary cell culture and tissues from mutant mice to explore the mechanisms regulating nuclear localization of PALLD. Our current model is that FSH-induced organization of the Sertoli cell cytoskeleton resulting in the formation of discrete F-actin-containing structures results in a reduction in the overall F-actin cytoskeleton, which in turn directs the nuclear localization of PALLD. Future experiments will test these complex interactions and investigate the specific role of nuclear PALLD.

Supplementary data

This is linked to the online version of the paper at http://dx.doi.org/10.1530/REP-14-0147.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding

This study was supported by startup funds from the Department of Anatomy and Cell Biology and Division of Research and Graduate Studies at ECU (to C B Geyer).

Acknowledgements

The authors thank Joani Zari-Oswald for technical assistance, and David Tulis and Myles Cabot (East Carolina University) for helpful advice. SCARKO testis sections were provided by Marvin Meistrich (MD Anderson), and Fshb KO testis sections were provided by Rajendra Kumar (UKMC).

References

  • AllanCMHandelsmanDJ2005Transgenic models for exploring gonadotropin biology in the male. Endocrine26235239. (doi:10.1385/ENDO:26:3:235)

  • AllanCMGarciaASpalivieroJZhangFPJimenezMHuhtaniemiIHandelsmanDJ2004Complete Sertoli cell proliferation induced by follicle-stimulating hormone (FSH) independently of luteinizing hormone activity: evidence from genetic models of isolated FSH action. Endocrinology14515871593. (doi:10.1210/en.2003-1164)

  • AnwayMDFolmerJWrightWWZirkinBR2003Isolation of Sertoli cells from adult rat testes: an approach to ex vivo studies of Sertoli cell function. Biology of Reproduction689961002. (doi:10.1095/biolreprod.102.008045)

  • ArcherSKClaudianosCCampbellHD2005Evolution of the gelsolin family of actin-binding proteins as novel transcriptional coactivators. Bioessays27388396. (doi:10.1002/bies.20200)

  • BeckMRDixonRDGoicoecheaSMMurphyGSBrungardtJGBeamMTSrinathPPatelJMohiuddinJOteyCA2013Structure and function of Palladin's actin binding domain. Journal of Molecular Biology42533253337. (doi:10.1016/j.jmb.2013.06.016)

  • BhallaVKReichertLEJr1974Properties of follicle-stimulating hormone–receptor interactions. Specific binding of human follicle-stimulating hormone to rat testes. Journal of Biological Chemistry2494351.

  • BishopALHallA2000Rho GTPases and their effector proteins. Biochemical Journal348241255. (doi:10.1042/0264-6021:3480241)

  • BremnerWJMillarMRSharpeRMSaundersPT1994Immunohistochemical localization of androgen receptors in the rat testis: evidence for stage-dependent expression and regulation by androgens. Endocrinology13512271234. (doi:10.1210/endo.135.3.8070367)

  • BrennerSLKornED1979Substoichiometric concentrations of cytochalasin D inhibit actin polymerization. Additional evidence for an F-actin treadmill. Journal of Biological Chemistry25499829985.

  • BrownSSSpudichJA1979Cytochalasin inhibits the rate of elongation of actin filament fragments. Journal of Cell Biology83657662. (doi:10.1083/jcb.83.3.657)

  • CattanachBMIddonCACharltonHMChiappaSAFinkG1977Gonadotrophin-releasing hormone deficiency in a mutant mouse with hypogonadism. Nature269338340. (doi:10.1038/269338a0)

  • ChaudharyJWhaleyPDCuppASkinnerMK1996Transcriptional regulation of Sertoli cell differentiation by follicle-stimulating hormone at the level of the c-fos and transferrin promoters. Biology of Reproduction54692699. (doi:10.1095/biolreprod54.3.692)

  • ChengCYSilvestriniBGrimaJMoMYZhuLJJohanssonESasoLLeoneMGPalmeryMMrukD2001Two new male contraceptives exert their effects by depleting germ cells prematurely from the testis. Biology of Reproduction65449461. (doi:10.1095/biolreprod65.2.449)

  • la CourTKiemerLMolgaardAGuptaRSkriverKBrunakS2004Analysis and prediction of leucine-rich nuclear export signals. Protein Engineering Design & Selection17527536. (doi:10.1093/protein/gzh062)

  • De GendtKSwinnenJVSaundersPTSchoonjansLDewerchinMDevosATanKAtanassovaNClaessensFLecureuilC2004A Sertoli cell-selective knockout of the androgen receptor causes spermatogenic arrest in meiosis. PNAS10113271332. (doi:10.1073/pnas.0308114100)

  • DixonRDArnemanDKRachlinASSundaresanNRCostelloMJCampbellSLOteyCA2008Palladin is an actin cross-linking protein that uses immunoglobulin-like domains to bind filamentous actin. Journal of Biological Chemistry28362226231. (doi:10.1074/jbc.M707694200)

  • DorringtonJHVernonRGFritzIB1972The effect of gonadotrophins on the 3’,5’-AMP levels of seminiferous tubules. Biochemical and Biophysical Research Communications4615231528. (doi:10.1016/0006-291X(72)90780-2)

  • EndlichNSchordanECohenCDKretzlerMLewkoBWelschTKrizWOteyCAEndlichK2009Palladin is a dynamic actin-associated protein in podocytes. Kidney International75214226. (doi:10.1038/ki.2008.486)

  • FlanaganMDLinS1980Cytochalasins block actin filament elongation by binding to high affinity sites associated with F-actin. Journal of Biological Chemistry255835838.

  • FritzIBLouisBGTungPSGriswoldMRommertsFGDorringtonJH1975Biochemical responses of cultured Sertoli cell-enriched preparations to follicle stimulating hormone and dibutyryl cyclic AMP. Current Topics in Molecular Endocrinology2367382. (doi:10.1007/978-1-4613-4440-7_26)

  • GettemansJVan ImpeKDelanoteVHubertTVandekerckhoveJDe CorteV2005Nuclear actin-binding proteins as modulators of gene transcription. Traffic6847857. (doi:10.1111/j.1600-0854.2005.00326.x)

  • GineitisDTreismanR2001Differential usage of signal transduction pathways defines two types of serum response factor target gene. Journal of Biological Chemistry2762453124539. (doi:10.1074/jbc.M102678200)

  • GoicoecheaSMBednarskiBStackCCowanDWVolmarKThorneLCukiermanERustgiAKBrentnallTHwangRF2010Isoform-specific upregulation of Palladin in human and murine pancreas tumors. PLoS ONE5e10347. (doi:10.1371/journal.pone.0010347)

  • HallA2012Rho family GTPases. Biochemical Society Transactions4013781382. (doi:10.1042/BST20120103)

  • HutsonJC1978The effects of various hormones on the surface morphology of testicular cells in culture. American Journal of Anatomy1515569. (doi:10.1002/aja.1001510106)

  • HutsonJCStoccoDM1978Specificity of hormone-induced responses of testicular cells in culture. Biology of Reproduction19768772. (doi:10.1095/biolreprod19.4.768)

  • HutsonJCGarnerCWStoccoDM1980Effects of serum components on the morphology of Sertoli cells in culture. Anatomical Record197205211. (doi:10.1002/ar.1091970209)

  • JinLGanQZiebaBJGoicoecheaSMOwensGKOteyCASomlyoAV2010The actin associated protein Palladin is important for the early smooth muscle cell differentiation. PLoS ONE5e12823. (doi:10.1371/journal.pone.0012823)

  • KuehlFAJrPatanelliDJTarnoffJHumesJL1970Testicular adenyl cyclase: stimulation by the pituitary gonadotrophins. Biology of Reproduction2154163. (doi:10.1095/biolreprod2.1.154)

  • KumarTRWangYLuNMatzukMM1997Follicle stimulating hormone is required for ovarian follicle maturation but not male fertility. Nature Genetics15201204. (doi:10.1038/ng0297-201)

  • KutayUGuttingerS2005Leucine-rich nuclear-export signals: born to be weak. Trends in Cell Biology15121124. (doi:10.1016/j.tcb.2005.01.005)

  • de LanerollePSerebryannyyL2011Nuclear actin and myosins: life without filaments. Nature Cell Biology1312821288. (doi:10.1038/ncb2364)

  • LiePPMrukDDLeeWMChengCY2010Cytoskeletal dynamics and spermatogenesis. Philosophical Transactions of the Royal Society of London. Series B Biological Sciences36515811592. (doi:10.1098/rstb.2009.0261)

  • LinDCLinS1979Actin polymerization induced by a motility-related high-affinity cytochalasin binding complex from human erythrocyte membrane. PNAS7623452349. (doi:10.1073/pnas.76.5.2345)

  • LoyCJSimKSYongEL2003Filamin-A fragment localizes to the nucleus to regulate androgen receptor and coactivator functions. PNAS10045624567. (doi:10.1073/pnas.0736237100)

  • LuoHLiuXWangFHuangQShenSWangLXuGSunXKongHGuM2005Disruption of Palladin results in neural tube closure defects in mice. Molecular and Cellular Neurosciences29507515. (doi:10.1016/j.mcn.2004.12.002)

  • MacLean-FletcherSPollardTD1980Mechanism of action of cytochalasin B on actin. Cell20329341. (doi:10.1016/0092-8674(80)90619-4)

  • MasonAJHayflickJSZoellerRTYoungWSIIIPhillipsHSNikolicsKSeeburgPH1986A deletion truncating the gonadotropin-releasing hormone gene is responsible for hypogonadism in the hpg mouse. Science23413661371. (doi:10.1126/science.3024317)

  • MeansARVaitukaitisJ1972Peptide hormone "receptors": specific binding of 3 H-FSH to testis. Endocrinology903946. (doi:10.1210/endo-90-1-39)

  • MirallesFPosernGZaromytidouAITreismanR2003Actin dynamics control SRF activity by regulation of its coactivator MAL. Cell113329342. (doi:10.1016/S0092-8674(03)00278-2)

  • MokKWMrukDDLiePPLuiWYChengCY2011Adjudin, a potential male contraceptive, exerts its effects locally in the seminiferous epithelium of mammalian testes. Reproduction141571580. (doi:10.1530/REP-10-0464)

  • MrukDDChengCY2010Tight junctions in the testis: new perspectives. Philosophical Transactions of the Royal Society of London. Series B Biological Sciences36516211635. (doi:10.1098/rstb.2010.0010)

  • MufflyKENazianSJCameronDF1993Junction-related Sertoli cell cytoskeleton in testosterone-treated hypophysectomized rats. Biology of Reproduction4911221132. (doi:10.1095/biolreprod49.5.1122)

  • MufflyKENazianSJCameronDF1994Effects of follicle-stimulating hormone on the junction-related Sertoli cell cytoskeleton and daily sperm production in testosterone-treated hypophysectomized rats. Biology of Reproduction51158166. (doi:10.1095/biolreprod51.1.158)

  • MuradFStrauchBSVaughanM1969The effect of gonadotropins on testicular adenyl cyclase. Biochimica et Biophysica Acta177591598. (doi:10.1016/0304-4165(69)90324-9)

  • NiedenbergerBAChappellVKKayeEPRenegarRHGeyerCB2013Nuclear localization of the actin regulatory protein Palladin in Sertoli cells. Molecular Reproduction and Development80403413. (doi:10.1002/mrd.22174)

  • OlsonENNordheimA2010Linking actin dynamics and gene transcription to drive cellular motile functions. Nature Reviews. Molecular Cell Biology11353365. (doi:10.1038/nrm2890)

  • OmaYHarataM2011Actin-related proteins localized in the nucleus: from discovery to novel roles in nuclear organization. Nucleus23846. (doi:10.4161/nucl.2.1.14510)

  • OteyCADixonRStackCGoicoecheaSM2009Cytoplasmic Ig-domain proteins: cytoskeletal regulators with a role in human disease. Cell Motility and the Cytoskeleton66618634. (doi:10.1002/cm.20385)

  • OzanneDMBradyMECookSGaughanLNealDERobsonCN2000Androgen receptor nuclear translocation is facilitated by the f-actin cross-linking protein filamin. Molecular Endocrinology1416181626. (doi:10.1210/mend.14.10.0541)

  • ParastMMOteyCA2000Characterization of Palladin, a novel protein localized to stress fibers and cell adhesions. Journal of Cell Biology150643656. (doi:10.1083/jcb.150.3.643)

  • PedersonT2008As functional nuclear actin comes into view, is it globular, filamentous, or both?Journal of Cell Biology18010611064. (doi:10.1083/jcb.200709082)

  • PranteBCGarmanKLSimsBNLindseyJS2008Matrix-coated transwell-cultured TM4 Sertoli cell testosterone-regulated gene expression mimics in vivo expression. In Vitro Cellular & Developmental Biology. Animal44434443. (doi:10.1007/s11626-008-9135-8)

  • QianXMrukDDWongEWLiePPChengCY2013Palladin is a regulator of actin filament bundles at the ectoplasmic specialization in adult rat testes. Endocrinology15419071920. (doi:10.1210/en.2012-2269)

  • RachlinASOteyCA2006Identification of Palladin isoforms and characterization of an isoform-specific interaction between Lasp-1 and Palladin. Journal of Cell Science1199951004. (doi:10.1242/jcs.02825)

  • RomrellLJRossMH1979Characterization of Sertoli cell–germ cell junctional specializations in dissociated testicular cells. Anatomical Record1932341. (doi:10.1002/ar.1091930103)

  • RossMH1976The Sertoli cell junctional specialization during spermiogenesis and at spermiation. Anatomical Record18679104. (doi:10.1002/ar.1091860107)

  • RussellL1977Movement of spermatocytes from the basal to the adluminal compartment of the rat testis. American Journal of Anatomy148313328. (doi:10.1002/aja.1001480303)

  • SchmittgenTDZakrajsekBA2000Effect of experimental treatment on housekeeping gene expression: validation by real-time, quantitative RT-PCR. Journal of Biochemical and Biophysical Methods466981. (doi:10.1016/S0165-022X(00)00129-9)

  • SlukaPO'DonnellLBartlesJRStantonPG2006FSH regulates the formation of adherens junctions and ectoplasmic specialisations between rat Sertoli cells in vitro and in vivo. Journal of Endocrinology189381395. (doi:10.1677/joe.1.06634)

  • SotiropoulosAGineitisDCopelandJTreismanR1999Signal-regulated activation of serum response factor is mediated by changes in actin dynamics. Cell98159169. (doi:10.1016/S0092-8674(00)81011-9)

  • SpectorIShochetNRKashmanYGroweissA1983Latrunculins: novel marine toxins that disrupt microfilament organization in cultured cells. Science219493495. (doi:10.1126/science.6681676)

  • TungPSFritzIB1975Follicle stimulating hormone and cyclic AMP directed changes in ultrastructure and properties of cultured Sertoli cell-enriched preparations: comparison with cultured testicular peritubular cells. Current Topics in Molecular Endocrinology2459508. (doi:10.1007/978-1-4613-4440-7_34)

  • TungPSDorringtonJHFritzIB1975Structural changes inducted by follicle-stimulating hormone or dibutyryl cyclic AMP on presumptive Sertoli cells in culture. PNAS7218381842. (doi:10.1073/pnas.72.5.1838)

  • Van AelstLD'Souza-SchoreyC1997Rho GTPases and signaling networks. Genes and Development1122952322. (doi:10.1101/gad.11.18.2295)

  • VisaNPercipalleP2010Nuclear functions of actin. Cold Spring Harbor Perspectives in Biology2a000620. (doi:10.1101/cshperspect.a000620)

  • VoglAWVaidKSGuttmanJA2008The Sertoli cell cytoskeleton. Advances in Experimental Medicine and Biology636186211. (doi:10.1007/978-0-387-09597-4_11)

  • WalkerWH2003Molecular mechanisms controlling Sertoli cell proliferation and differentiation. Endocrinology14437193721. (doi:10.1210/en.2003-0765)

  • WalkerWHChengJ2005FSH and testosterone signaling in Sertoli cells. Reproduction1301528. (doi:10.1530/rep.1.00358)

  • WangDZLiSHockemeyerDSutherlandLWangZSchrattGRichardsonJANordheimAOlsonEN2002Potentiation of serum response factor activity by a family of myocardin-related transcription factors. PNAS991485514860. (doi:10.1073/pnas.222561499)

  • ZhengBHanMBernierMWenJK2009Nuclear actin and actin-binding proteins in the regulation of transcription and gene expression. FEBS Journal27626692685. (doi:10.1111/j.1742-4658.2009.06986.x)

  • ZhouWWangGSmallCLLiuZWengCCYangLGriswoldMDMeistrichML2011Gene expression alterations by conditional knockout of androgen receptor in adult Sertoli cells of Utp14b jsd/jsd (jsd) mice. Biology of Reproduction84400408. (doi:10.1095/biolreprod.110.090530)

If the inline PDF is not rendering correctly, you can download the PDF file here.

Downloadable materials

 

    Society for Reproduction and Fertility

Article Information

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 155 155 30
PDF Downloads 51 51 6

Altmetrics

Related Articles

Figures

  • View in gallery

    PALLD localization in cultured primary Sertoli cells. Treatments with various inhibitors are described in the Materials and Methods section. (A) Timeline for primary Sertoli cell culture experiments. (B, C, and D) IIF to detect PALLD (in green) in primary Sertoli cells from mice aged 10 dpp (B) and 18 dpp (D and C) in the absence (B and C) or presence (D) of serum. Phalloidin-labeled F-actin cytoskeleton (in red). (E and F) Levels of Palld mRNA (E) and protein (F) were slightly elevated in Sertoli cells grown in the presence of serum (P<0.0271). Scale bars=25 μm.

  • View in gallery

    Gene expression changes in primary Sertoli cells in response to serum. Message levels were quantitated by qRT-PCR for the indicated genes (Vcl, vinculin; Actb, beta actin; Fos, c-FBJ osteosarcoma oncogene; Trf, transferrin). Statistical significance is indicated by an asterisk, with associated P values.

  • View in gallery

    PALLD accumulates in nuclei after inhibition of nuclear export. (A, B, and C) IIF to detect PALLD (in green) in serum-stimulated Sertoli cells grown without (A) or with leptomycin B treatment for 1 h (B) and 7 h (C). Phalloidin-labeled F-actin cytoskeleton (in red). Scale bars=25 μm.

  • View in gallery

    PALLD accumulates in nuclei in response to increased cAMP and after disruption of the F-actin cytoskeleton. (A, B, and C) IIF to detect PALLD (in green) in serum-stimulated Sertoli cells grown without (A) or with cytochalasin D (B), latrunculin (C), 8-bromo-cAMP (D), forskolin (E), or C3 transferase (F). Phalloidin-labeled F-actin cytoskeleton (in red). Scale bar=25 μm.

  • View in gallery

    PALLD nuclear localization is disrupted in response to aberrant gonadotropin signaling in vivo. IHC was used to detect PALLD (brown), and sections were counterstained with hematoxylin (blue). PALLD in adult testis sections from WT (A), Hpg mutant (B and C), SCARKO (D and E), or Fshb KO (F) mice. Arrows indicate unstained Sertoli cell nuclei in panels C–D. Scale bars=25 μm.

PubMed

Google Scholar