Muscle glycogen synthase isoform is responsible for testicular glycogen synthesis: Glycogen overproduction induces apoptosis in male germ cells

Animals

All OF1 mice were obtained from the Biosafety Laboratory Facility for Animal Experimentation (Veterinary Science Faculty, Universidad Austral de Chile, Valdivia, Chile). All experiments were performed in accordance with the standards given by the Manual for Biosafety and Bioethics (CONICYT, Chilean Government) and according to guidelines established by the Animal Protection Committee of the Universidad Austral de Chile. Animals were maintained in a 12:12 light:dark cycle environment and supplied with commercial food pellets and water ad libitum.

Transgenic Animals

Targeting vector construction and knock-in strategy was designed and performed by genOway (Lyon, France) and previously described in Duran et al. [2012]. In brief, the superactive MGS (saMGS) was introduced into the Hprt locus by homologous recombination in 129/Ola (E14). Embryonic stem cell (ES) clones were then injected into C57BL/6 blastocysts and implanted into OF1 pseudopregnant females. Highly chimaeric males were selected for breeding with wild-type C57BL/6J females. The saMGS was performed by site-specific mutation of serine residues to alanine. All procedures were approved by the Barcelona Science Park's Animal Experimentation Committee and were carried out in accordance with the European Community Council Directive and National Institutes of Health guidelines for the care and use of laboratory animals.

Cell Culture and Transfections

Sertoli cell line 42GPA9 widely validated as isolated Sertoli cells by the group that isolated this cell line, Georges Pontis and Dominique Segretain at the Université Paris 5, Paris, France [Bourdon et al., 1998] and spermatogonium-like cell line GC-1 were cultured in DMEM/F12 medium (Thermo Scientific HyClone) supplemented with 10% FBS medium (Thermo Scientific HyClone) [Bourdon et al., 1998; Angulo et al., 2008]. Primary Sertoli cells were cultured as previously described [Angulo et al., 2008]. Transfections were performed with Lipofectamine™ 2000 reagent (Invitrogen) using 1 µg of plasmidial DNA and following the manufacturer's instructions. The plasmidial DNA that contained the mouse PTG fused to GFP (pEGFP-C1 vector, Clontech) was a gift from Dr. Guinovart and has been described previously [Vilchez et al., 2007]. All cells were incubated at 37°C with 5% CO₂.

Electrophoresis and Immunoblotting

Tissue and cell-culture plates were processed for protein extract preparation as has been previously reported [Vilchez et al., 2007]. Proteins were resolved by 10% SDS–PAGE, transferred onto a nitrocellulose membrane (Bio-Rad) and probed with the following antibodies: rabbit antibody to human MGS (Cell Signaling) that recognizes MGS independent of its phosphorylation state; rabbit antibody to rabbit LGS (a gift from Dr. Joan J. Guinovart); rabbit antibody to MGS with phosphorylation of Ser640 (Cell Signaling); mouse antibody to beta-actin (Santa Cruz Biotechnology); mouse antibody to ubiquitin (Cell Signaling).

MGS-Polyubiquitination Assay

Cells seeded on 60 mm plates were cultivated in DMEM/F12 medium for 12 and 24 h in the presence of cycloheximide (1 µg/ml), proteosomal inhibitor MG-132 (10 µM) or DMSO (0.1% w/w). They were then processed for protein extract preparation and separated into two fractions. One fraction was used for isolation of the total polyubiquitinated proteins with the Ubiquitinated protein enrichment kit (Calbiochem–Merck) and the other fraction was used to analyse the total protein fraction. Both protein fractions were probed by Western blot method as described previously.

Immunofluorescence and Histochemical Assays

Cells seeded on poly-L-lysine-coated coverslips were fixed for 30 min in PBS containing 4% (w/v) paraformaldehyde. After fixation, cells were permeabilised for 20 min with PBS containing 0.3% (v/v) Triton X-100. Blocking and incubation with the primary and secondary antibodies were carried out in PBS containing 3% (w/v) BSA and 0.15% (v/v) Triton X-100. Coverslips were washed in PBS and 0.3% (v/v) Triton X-100, and mounted onto glass slides using fluorescence mounting medium from Dako. We used primary antibodies to MGS3 and LGS3 (from Dr. Guinovart's laboratory), antibody to cleaved caspase3 (Cell Signaling) and a monoclonal antibody against glycogen (a gift from O. Baba, Tokyo Medical and Dental University). In some cases, nuclei were stained with Hoechst 33342 or TO-PRO®-3 (Molecular Probes). Fluorescence images were obtained with an Olympus FV-1000 confocal microscope (Universidad Austral de Chile). The light source was an argon/krypton laser (75 mW), and optical sections (0.1 mm) were obtained. For immunohistochemistry, tissue sections were fixed for 48 hr in PBS containing 4% (w/v) paraformaldehyde or BOUIN, paraffin-embedded and treated as described previously. The fibrosis studies performed in testis from wild-type and transgenic mice were carried out with Masson's trichrome and Periodic Acid-Schiff (PAS) stains and with haematoxylin-eosin analysis.

Apoptosis Assays

Tissue sections (0.4 µM) were deparrafinised in a graded alcohol series, washed in sterile PBS and processed for TUNEL or active caspase 3 staining. TUNEL assays were carried out using the Strept-Fluo-tag in situ Apoptosis Detection Kit (R&D Systems), by following the manufacturer's instructions. Active caspase 3-positive cells were visualized by immunocytochemistry using the cleaved caspase3 antibody (Cell Signaling). The TUNEL- and active caspase 3-positive cells were photographed with a Nikon Eclipse E-600 microscope using a 40× objective. The percentage of positive cells was estimated in 8–14 fields from each of three coverslips (two independent experiments) for each treatment condition (100 seminiferous tubules, knock-in or wild-type, respectively). Total number of cells was demonstrated after staining nuclei with Hoechst 33342 or TO-PRO3®.

RNA Purification and Retrotranscription

Total RNA was isolated from mouse tissue using Trizol® reagent (Invitrogen) according to the manufacturer's instructions. RNAs were purified for a second time using RNAeasy (Qiagen), and 5 µg total RNA was retrotranscripted with SuperScript™ III First-Strand Synthesis Supermix (Invitrogen). A series of specific primers were designed to specifically amplify a fragment of the coding sequence of mouse MGS, LGS, malin (Emp2b), laforin (Emp2a) and beta-actin for conventional RT-PCR analyses [Vilchez et al., 2007].

Quantitative RT-PCR

We followed the standard RT-PCR protocol of the ABI Prism® 7700 Sequence Detection System, together with the ready-made TaqMan primer and probe sets (Applied Biosystems; #Mm00472712_m1, #Hs99999901_s1 and #Mm03302249_g1). Each sample was analysed in triplicate wells with 100 ng of first-strand cDNA in a total reaction volume of 20 µl. The temperature profile consisted of 40 cycles, of 15 s, at 95°C and of 1 min at 60°C. Data was analysed with the comparative 2^-ΔΔCt method using ribosomal 18S as an endogenous control [Vilchez et al., 2007].

Glycogen Analysis

Total homogenates, supernatants and sediments were boiled at 100°C in 30% KOH and glycogen content was determined by an amyloglucosidase-based assay as previously reported [Vilchez et al., 2007; Duran et al., 2012].

Determination of MGS Activity

Tissue sections were processed as described previously [Vilchez et al., 2007]. Protein concentration was measured with the Bradford Method. MGS activity was measured in homogenates in the presence of 4.4 mM UDP-glucose, and in the presence or absence of 6.6 mM G6P as described previously. The activity measured in the absence of G6P represents the active form of MGS (I or a form), whereas the activity measured in the presence of G6P represents total MGS activity.

GST-Pull Down

E. coli BL21 was transformed with a vector contained the malin-ring domain construct [Gentry et al., 2005]. One hundred microgram total proteins bacterial extract contained the malin-ring overexpressed as protein fusion to GST was immobilized to gluthation-sepharosa beads (GE Healthy) following the manufacturing instructions. The malin-ring immobilized was used as bait to demonstrate its interaction with MGS in testicular cells. 500 µg total proteins extract from GC-1 or 42GPA9 cell lines were incubated with beads overnight at 4°C and continuous stirred. None bound fraction was washed using phosphate buffered saline and 0.1% w/w Tween-20, and the proteins specifically bounded were eluted using Laemmli buffer [Laemmli, 1970]. Equal volumes from each fraction were fractionated by 10% SDS–PAGE and immunodetected using commercial antibody, previously described.

Statistical Analysis

Student's t-test was applied or non-parametric statistical analyses were performed using Prism version 3.0 for Microsoft software.

Muscle Glycogen Synthase Is Expressed in Testis, Is Active and Is Responsible for the Glycogen Store

It is almost 60 years ago that glycogen content and glycogen fluctuations were detected in the testis of different mammals [Arzac, 1950, 1953; Ewing et al., 1966; Fabbrini et al., 1969; Fouquet and Guha, 1969; Fawcett and Dym, 1976; Datta et al., 1988]. However, which isoform of GS was expressed in this tissue was unknown. Using immunofluorescence, RT-PCR and immunoblotting techniques, we found only the muscle isoform of glycogen synthase (MGS) and not the liver isoform in whole testis extracts (Fig. 1, images a–c). MGS was localized in seminiferous epithelium from rat adult testis, as well as at different stages of male germ cell differentiation, from the basal to the adluminal compartment of tubules (Fig. 1, images b and c). The inset shows a magnified portion of germ epithelium that is immunoreactive to MGS (Fig. 1, image c). The MGS-positive signal was very granular, implying many sites of glycogen synthesis, as has been reported in other cell types and tissues [Graham et al., 2010; Roach et al., 2012]. By conventional RT-PCR (Fig. 1B) we only detected the expression of MGS and by using Western blot analysis (Fig. 1C), we confirmed the presence of this enzyme to reveal a main protein band that migrated with an apparent molecular mass of approximately 85 kDa in adult tissue. In addition, we evaluated by RT-PCR (Fig. 1D) and Western blot (Fig. 1E) the presence of MGS in isolated seminiferous epithelial cells. Sertoli cells showed higher levels of MGS than germ cells pool, suggesting important differences between somatic cells and male germ cells respect to glycogen metabolism. In contrast to previously published results by Ballester et al. [2000], we detected the muscle isoform of GS but not the liver isoform in mammalian spermatozoa. In addition, we confirmed the presence of MGS in spermatozoa from human, stallion, boar, dog, rat, and mouse using immunofluorescence (data not shown). We also examined GS activity during testis development (Fig. 2A) and the glycogen store (Fig. 2B) in mice. We detected higher levels of total activity, measured in the presence of glucose-6P, which represents the total amount of enzyme in the first days after birth, specifically at day 5. GS activity then decreases progressively with testis development. We discovered essentially the same result for glycogen levels, an initial increase and then a major decrease after day 5. By Western blot analysis, we confirmed the MGS presence in the same samples used for enzymatic assays (Fig. 2C). At short exposure times, the immunoreactive signal was positive only for the first days after birth; however, at longer exposure times MGS was detected in all ages.

Sertoli Cells as the Main Source of Intratubular Glycogen

Many authors have reported that Sertoli cells are able to synthesise glycogen, which is the main source of glycogen in the seminiferous tubules and considered as a cell maturation signal [Arzac, 1950; Ewing et al., 1966; Fabbrini et al., 1969; Fouquet and Guha, 1969]. By immunofluorescence and confocal microscopy, we analyzed the subcellular localization of MGS and glycogen in 42GPA9 Sertoli cell line (Fig. 3A). MGS signal was granular while glycogen presented a homogeneous distribution within the whole cytoplasm (images a and b). The merged fluorescence pattern, showed partial colocalization of MGS and glycogen, suggesting active glycogen synthesis sites (image c). MGS was also detected in Sertoli primary culture (Fig. 1A, images d and e), confirming the validation of cell line 42GPA9 as previously demonstrated by transferrin, sulfated glycoprotein-2, vitamin C and glucose transporters (SVCTs and GLUTs), and these cells also retained FSH receptors as suggested by their specific responsiveness to the gonadotropin and the detection of FSH receptor mRNAs [Bourdon et al., 1998; Angulo et al., 2008, 2011]. MGS amount present in Sertoli and male germ cell lines was estimated by an enzymatic assay (Fig. 3B). Total GS activity was measured in the presence of G6P showing that Sertoli cell line 42GPA9 (black bar) has nearly 30% more activity than spermatogonium cell line GC-1 (gray bar), suggesting that Sertoli cells have higher levels of the enzyme.

We estimated glycogen stores by biochemical assays in cell line GC-1, representing 10% of the glycogen store in Sertoli cells (42GPA9 cell line and primary culture; Fig. 3C). The glycogen content was very different between germ cells and Sertoli cells, which confirms the contribution of Sertoli cells to the intratubular glycogen storage. MGS detection in each cell line was confirmed by RT-PCR (Fig. 3D) and immunoblotting (Fig. 3E); Sertoli cells (cell line 42GPA9) showed higher levels of MGS than germ cells (cell line GC-1), confirming our results in isolated cells (germ cells pool and Sertoli cells primary culture).

Glycogen Synthesis in Testis Could Be Regulated by the Malin–Laforin Complex

The malin–laforin complex has been described in brain, specifically in neurons, where it regulates MGS availability and activity [Vilchez et al., 2007; Solaz-Fuster et al., 2008; Worby et al., 2008; Roach et al., 2012]. To determine whether this complex might be present in mammalian testis, we evaluated malin and laforin expression in testis from adult mice with real time and convetional RT-PCR and immunodetection of malin and laforin proteins (Fig. 4). The Real-time RT-PCR results show a differential expression of malin and laforin (Fig. 4A, B), suggesting that laforin is present during the first days of testis development. On the contrary, malin transcript is detected after 10 days post-birth, at the same time in which GS and glycogen store decreased. By conventional RT-PCR, we detected the respective transcripts in whole testis, Sertoli cells (cell line 42GPA9) and isolated germ cells with the exception of spermatids, suggesting that these cells do not express the malin gene at that particular differentiation stage (Fig. 4C). With immunoblotting, laforin was detected in testis (Fig. 4D), confirming the expression of this protein as suggested by other authors [Roach et al., 2012].

The presence of laforin and the putative presence of malin in testicular tissue could be an additional mechanism to regulate glycogen synthesis in the seminiferous tubules. To support this hypothesis, we performed GST-pull down using the malin-ring domain as bait and detected the interaction between malin and endogenous MGS in germ cell (cell line GC-1), Sertoli cell (cell line 42GPA9) and whole testis (Fig. 5A) [Gentry et al., 2005]. In all cases, MGS in protein extracts from different cell types were able to interact and bind to the catalytic domain of malin (as E3 ubiquitin ligase), suggesting that MGS in testicular tissue is substrate to malin and confirming that ubiquitination is one mechanism to modulate the activity and availability of GS for testicular glycogen synthesis (Fig. 5A). This regulatory mechanism may involve the malin–laforin complex, as for neurological tissues and could be a common mechanism to other cell types. As it has been previously found that Sertoli cells could be the main source of intratubular glycogen in adult testis and from our own experimental evidence (Fig. 3), we choose the Sertoli cell line 42GPA9 to demonstrate that the malin–laforin complex is able to polyubiquitinate MGS and indirectly regulate glycogen synthesis in testis. After treatment with cycloheximide and MG-132 (a proteosomal inhibitor), we detected accumulation of a polyubiquitinated form of MGS by Western blot analysis (Fig. 5B). The polyubiquitinated-MGS present in the total protein extract has a molecular weight that varies by 20–40 kDa over its normal migration, showing an evident slip in its electrophoretic profile when the MGS was immunoprecipitated and immunodetected using anti-ubiquitin antibody (data not shown). Despite this, the polyUb-MGS is mainly found in a 110 kDa form. To confirm MGS polyubiquitination, we isolated the total polyubiquitinated-protein fraction (polyUb fraction) and investigated the presence of MGS with immunodetection. Results showed a specific immunoreactivity for this protein both in cells treated with MG-132 alone or with MG-132 plus cycloheximide. The molecular weight of MGS in the polyUb fraction is the same as that detected in the total protein extract, suggesting that ubiquitination of MGS and eventually proteosomal degradation in Sertoli cells might be the mechanism by which glycogen synthesis in testis is regulated. Other studies have investigated the malin–laforin complex using several other experimental methods, such as transfection and overexpression or silencing of genes in different cell models. In this work, we present for the first time evidence of an endogenous malin–laforin complex and its activity, specifically in Sertoli cells under normal conditions.

Overproduction of Glycogen Triggers Germ Cell Apoptosis and Seminiferous Tubule Degeneration

To induce male germ cells glycogen overproduction, we transfected spermatogonia-like cell line GC-1 with PTG (protein targeting to glycogen), which targets PP1 to glycogen synthesis sites, inducing MGS activation [Ingebritsen et al., 1983; Graham et al., 2010; Roach et al., 2012]. Overexpression of PTG in GC-1 cells should promote glycogen accumulation by reducing levels of phosphoSer640-MGS (or the less active form), and thus synthesising more glycogen. In addition, PTG overexpression could induce simultaneously an inactivation of glycogen phosphorylase activity promoting more glycogen storage. The PTG effect was measured by immunodetection of phosphoSer640-MGS (pMGS) and cleaved caspase 3 (Fig. 6A), with a decrease in pMGS levels detected. However, cleaved caspase3 and glycogen stores increase (Fig. 6B, C), which suggests a negative effect on cell viability induced by an imbalance on glycogen synthesis.

To better understand the role of glycogen during testis development and spermatogenesis, we investigated the effects of the loss in glycogen homeostasis during testis development (supplementary data). Glycogen overproduction was induced in nestin-positive cells by expression of the superactive form of MGS (saMGS) [Duran et al., 2012], enzyme which is able to synthesise higher amounts of glycogen, unbalancing the normal synthesis rate (elongation-branch and store balance). Histological analysis from Kin testes suggested an infantile phenotype of transgenic testis, characterized as an involution or cell cycle arrest like in many pathological processes of male reproductive tissues (cryptorchidism or varicocele). PAS staining in saMGS Kin testes revealed the presence of strongly PAS-positive structures in the seminiferous tubules (data not shown), confirming the hyperactivity of saMGS and glycogen overproduction [Duran et al., 2012]. The PAS staining showed many apparently dense structures, which suggested the formation of polyglucosan bodies in the seminiferous epithelium. These results were confirmed by immunodetection of MGS and polyglucosan. Immunolocalization of MGS was evaluated in saMGS knock-in (Kin) animals and was found to be specifically over-expressed in nestin-positive cells (supplementary data). Strong glycogen labelling seen in the germinal epithelium from saMGS Kin testis, but not in the control littermate. Glycogen has been previously associated with cell death in neurons and other cell types by an unknown mechanism [Vilchez et al., 2007; Vallés-Ortega et al., 2011; Duran et al., 2012]. To verify germ cell apoptosis, we detected formation of cleaved caspase 3 and we readily found specific labeling for caspase 3 in spermatocytes and spermatogonia Using TUNEL assay, we estimated an increase in the apoptotic rate of approximately eightfold with respect to control animals (supplementary data).