|Home | About | Journals | Submit | Contact Us | Français|
Spermatogenesis is the process by which spermatogonial stem cells divide and differentiate into sperm. The role of growth factor receptors in regulating self-renewal and differentiation of spermatogonial stem cells remains largely unclear. This study was designed to examine Gfra1 receptor expression in immature and adult mouse testes and determine the effects of Gfra1 knockdown on the proliferation and differentiation of type A spermatogonia. We demonstrated that GFRA1 was expressed in a subpopulation of spermatogonia in immature and adult mice. Neither Gfra1 mRNA nor GFRA1 protein was detected in pachytene spermatocytes and round spermatids. GFRA1 and POU5F1 (also known as OCT4), a marker for spermatogonial stem cells, were co-expressed in a subpopulation of type A spermatogonia from 6-day-old mice. In addition, the spermatogonia expressing GFRA1 exhibited a potential for proliferation and the ability to form colonies in culture, which is a characteristic of stem cells. RNA interference assays showed that Gfra1 small interfering RNAs (siRNAs) knocked down the expression of Gfra1 mRNA and GFRA1 protein in type A spermatogonia. Notably, the reduction of Gfra1 expression by Gfra1 siRNAs induced a phenotypic differentiation, as evidenced by the elevated expression of KIT, as well as the decreased expression of POU5F1 and proliferating cell nuclear antigen (PCNA). Furthermore, Gfra1 silencing resulted in a decrease in RET phosphorylation. Taken together, these data indicate that Gfra1 is expressed dominantly in mouse spermatogonial stem cells and that Gfra1 knockdown leads to their differentiation via the inactivation of RET tyrosine kinase, suggesting an essential role for Gfra1 in spermatogonial stem cell regulation.
Spermatogenesis is a complex process that involves the division and differentiation of a subpopulation of type A spermatogonia called spermatogonial stem cells into sperm. Spermatogonial stem cells are unique, since they are the only stem cells in the body that undergo self-renewal throughout life and transmit genetic information to subsequent generations [1, 2]. Within the seminiferous tubules, spermatogonial stem cells can be directed into one of three fates: self-renewal to give rise to new stem cells; differentiation into more mature spermatogonia; or cell death via apoptosis . A better understanding of the molecular mechanisms controlling the fates of spermatogonial stem cells is of paramount importance for the regulation of spermatogenesis. To date, the pathways induced by the inactivation of growth factor receptors and the underlying type A spermatogonial proliferation and/or differentiation remain largely unknown.
Glial cell line-derived neurotrophic factor receptor alpha 1 (GFRA1) is a co-receptor of RET for the growth factor GDNF (glial cell line-derived neurotrophic factor). GDNF plays an important role in the development of peripheral and central neurons [4–6] and the morphogenesis of kidney . GDNF is also able to regulate the proliferation and differentiation of undifferentiated spermatogonia and spermatogonial stem cells [8–11]. It has also been reported that GDNF alone induces a reduction of spermatogonial stem cells (SSCs) derived from C57 × ROSA mice over a 10-wk culture period, whereas GDNF with the addition of soluble GFRA1 potentiates the expansion of the cells duringthis time period, indicating that GFRA1 plays a role in the GDNF-induced replication of SSCs . Physiologic responses to GDNF require the presence of glycosyl phosphadidylinositol (GPI)-linked protein GFRA1, which binds GDNF specifically and mediates the activation of RET protein tyrosine kinase to induce intracellular signal cascades . It is suggested that both GFRA1 and RET receptors are important for spermatogonial development . GFRA receptors also appear to offer a gain control mechanism that enhances the signal-to-noise ratio of the response to GDNF . Thus, we hypothesized that the knockout or knockdown of Gfra1 expression may initiate spermatogonial differentiation in vitro.
The gene knockout approach is useful for clarifying the physiologic role of molecules in vivo; however, it is impossible to study the possible role of Gfra1 in regulating type A spermatogonial proliferation and differentiation, since the knockout of Gfra1 resulted in the death of newborn mice within 24 h due to uremia . To do a loss-of-function study of Gfra1 in vitro, we used RNA interference (RNAi) to knock down the expression of Gfra1 in type A spermatogonia derived from 6-day-old mice. Small interfering RNAs (siRNAs) are capable of recruiting a multienzyme complex called the RNA-induced silencing complex that identifies and cleaves the complementary target mRNA and eventually results in the degradation of endogenous mRNA and the suppression of protein synthesis . In the past few years, the use of siRNAs or short hairpin RNAs (shRNAs), has proven to be a powerful tool with which to knock down the expression of specific genes in mammalian cells [17, 18] and in Caenorhabditis elegans .
Six-day-old mice were employed in the current study because at this developmental age, type A spermatogonia and Sertoli cells are the only two cell types present in the seminiferous epithelium of the testis , thus allowing us to examine specifically the role of Gfra1 on the proliferation and/or differentiation of type A spermatogonia. It has been demonstrated that type A spermatogonia can be cultured alone in vitro for a short period of time, but they are able to survive for 25 days when cocultured with Sertoli cell monolayers without the addition of growth factors . Sertoli cells were thus used in this study as feeder cells to sustain the survival of type A spermatogonia. Within the testes, GDNF is secreted by Sertoli cells in vivo [10, 22], and thus we expected that Sertoli cells can also secrete GDNF in culture.
Gfra1 has been shown to be expressed in undifferentiated spermatogonia in mice [10–12]; however, Gfra1 mRNA and GFRA1 protein expression were recently observed in pachytene spermatocytes and round spermatids in rats . Therefore, one objective of this study was to ascertain whether Gfra1 mRNA and GFRA1 protein are also expressed in differentiated germ cells, including pachytene spermatocytes and round spermatids in mice. We initially examined Gfra1 mRNA and GFRA1 protein expression in the testis of immature and adult mice using RT-PCR, Western blots, and immunostaining. We also measured GDNF production by Sertoli cells in culture using an ELISA. Another objective of this study was to knock down Gfra1 expression in type A spermatogonia using two Gfra1 siRNAs targeting different sites of Gfra1 mRNA in order to examine the potential roles of Gfra1 in regulating type A spermatogonial proliferation and/or differentiation. To assess the degree of gene silencing, we combined semiquantitative RT-PCR, real-time quantitative RT-PCR, Western blots, and immunocytochemistry. To probe whether differentiation occurred after Gfra1 knockdown, we evaluated the expression of specific markers for spermatogonial stem cells, differentiating spermatogonia, spermatocytes, and spermatids. We further examined whether Gfra1 silencing in type A spermatogonia resulted in expression changes of RET phosphorylation.
BALB/c adult male mice (60 days old) and mothers with 6-day-old male pups were obtained from Charles River Laboratories Inc. (Wilmington, MA). All animal care procedures were carried out pursuant to the National Research Council's Guide for the Care and Use of Laboratory Animals, and the experimental protocols employed in this study were approved by the Animal Care and Use Committee of Georgetown University.
The seminiferous tubules were isolated from the testes of immature (6-day-old) and adult (60-day-old) mice using mechanical dissociation and collagenase digestion, and cell suspension containing germ cells and Sertoli cells was obtained by further enzymatic digestion using collagenase, trypsin, and hyaluronidase [20, 24]. Type A spermatogonia from immature mice, as well as pachytene spermatocytes with a purity of 90% and round spermatids with a purity of 95% from adult mice were isolated using the STAPUT method with a 2%–4% BSA gradient [20, 24]. After STAPUT, type A spermatogonia from immature mice were further purified by differential plating to remove the potential contamination by Sertoli and myoid cells, and the purity of the type A was about 95%, as evaluated by immunocytochemical staining using an antibody to GCNA1 (germ cell nuclear antigen 1), a gift from Dr. George C. Enders (Department of Anatomy and Cell Biology, University of Kansas Medical Center, Kansas City, KS).
In order to obtain a germ cell mixture minus the elongating spermatids from adult mice, we prepared a cell suspension containing all the germ cells and Sertoli cells from 60-day-old mice using mechanical dissociation and two-step enzymatic digestion [20, 24]. Elongating spermatids were removed by a 40-μm mesh filter, and the other germ cells were further separated from Sertoli cells by differential plating.
The coculture procedure for type A spermatogonia and Sertoli cells we employed in this study was based on a protocol that our lab previously used , with minor modifications. Specifically, Sertoli cells were obtained from the testes of 6-day-old mice using a two-step enzymatic digestion [20, 24] and differential plating, and they were trypsinized and seeded at an appropriate concentration of cells on 12-, 24-, or 96-well plates with Dulbecco modified Eagle medium (DMEM)/F12 supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, Utah), 100 U/ml penicillin, and 100 mg/ml streptomycin to allow for the attachment and growth of Sertoli cells. Three hours later, freshly isolated type A spermatogonia from 6-day-old mice were seeded onto the Sertoli cells; they were co-cultured for 4 h in the presence of DMEM/F12 and 10% FBS to allow for the attachment of the spermatogonia to the Sertoli cell feeder layer. Before Gfra1 siRNA transfection, the culture medium was changed with fresh DMEM/F12 without FBS, and then siRNA experiments were carried out.
Total RNA was extracted from the seminiferous tubules of immature mice and the various cells using Trizol reagent (Invitrogen, Carlsbad, CA). Total RNA (1 μg) was reverse transcribed into the first-strand cDNA in 20 μl of reaction primed by oligo(dT)12–18 primer using the Superscript II reverse transcriptase (Invitrogen), and 2 μl first-strand cDNA was used as template for the PCR reactions using Taq DNA polymerase (New England Biolabs, Ipswich, MA). The PCR reaction started at 94°C for 2 min and was performed as follows: denaturation at 94°C for 30 sec, annealing at a temperature (Tm) as indicated in Table 1 for 45 sec, and elongation at 72°C for 45 sec. After 35 cycles, the samples were incubated for an additional 7 min at 72°C. The forward and reverse primers of chosen genes were designed and are listed in Table 1. In some experiments, the housekeeping gene Gapdh (glyceraldehyde-3-phosphate dehydrogenase) served as a loading control for total RNA, and PCR products were separated by electrophoresis on 1.2% agarose gels. The gels were exposed to a Transilluminator (Fisher Scientific, Pittsburgh, PA), and pictures were taken with a Photo-Documentation Camera (Fisher Scientific). The data were scanned with an Epson Perfection 3200 PHOTO (Epson America Inc., Long Beach, CA), and densitometric analyses were processed with Adobe Photoshop 7.0 (Adobe Systems, Mountain View, CA).
Freshly isolated seminiferous tubules and type A spermatogonia from 6-day-old mice, pachytene spermatocytes and round spermatids from adult mice, and type A spermatogonia cocultured with Sertoli cells and treated with or without Gfra1 siRNAs were lysed with 1 ml ice-cold RIPA lysis buffer supplemented with a mixture of PMSF, sodium orthovanadate, and protease inhibitor cocktail solution (Santa Cruz Biotechnology Inc., Santa Cruz, CA). After 30 min of lysis on ice, cell lysates were cleared by centrifugation at 12 000 × g, and the concentration of protein was measured by the Bio-Rad Bradford assay (Bio-Rad Laboratories, Hercules, CA), with BSA (Bio-Rad Laboratories) as the standard.
For Western blots, 50 μg total protein from each sample was used for 4%– 12% SDS-PAGE, and MultiMark Multi-Colored Standard (Invitrogen) was used as the molecular mass standard for the proteins. Samples were resolved in the XCell SureLock Novex Mini-Cell apparatus (Invitrogen) and transferred to nitrocellulose membranes for 2 h at room temperature. The membranes were washed with TBS containing 0.1% Tween (TBS-T) and were blocked with 5% nonfat dry milk in TBS-T for 1 h at room temperature. After extensive washes with TBS-T, the membranes were incubated with the chosen antibody, including anti-GFRA1 (Imgenex Corp., San Diego, CA), anti-PCNA (EMD Biosciences Inc., San Diego, CA), anti-POU5F1 (C-10) (Santa Cruz Biotechnology), or anti-KIT (C-19) (Santa Cruz Biotechnology), at an appropriate dilution in TBS-T for 1 h. After three washes with TBS-T, the membranes were incubated with horseradish peroxidase-conjugated anti-rabbit immunoglobulin G (IgG; Imgenex), anti-goat IgG, or anti-mouse IgG (Santa Cruz Biotechnology) at a 1:2000 to 1:5000 dilution for 1 h at room temperature. The proteins were detected using the Western Blotting Luminol Reagent (Santa Cruz Biotechnology) and were exposed to autoradiography film. In some experiments, the membranes were reused for detection of ACTB (also known as beta-actin) using an antibody to ACTB (Imgenex). The films were scanned with Epson Perfection 3200 PHOTO, and densitometric analyses were processed with Adobe Photoshop 7.0.
Type A spermatogonia from 6-day-old mice were cocultured with Sertoli cells and then treated with or without Gfra1 siRNAs. Forty-eight hours later, 100 ng/ml GDNF was added to the culture medium for 10 min, and the cells were lysed in RIPA lysis buffer. Western blots were carried out according to the methods mentioned above using an antibody to phospho-RET (Tyr 1062; Santa Cruz Biotechnology). The membranes were reused to detect ACTB using an antibody to ACTB. The films were scanned with an Epson Perfection 3200 PHOTO, and densitometric analyses were processed with Adobe Photoshop 7.0.
To prepare sections, testes from 6-day-old and 60-day-old mice were fixed in 4% paraformaldehyde for 1.5 h and washed with phosphate-buffered saline (PBS). The testes were dehydrated through a series of graded alcohols and were embedded in paraffin at 60°C overnight. For immunohistochemistry, 5-μm testis sections were dewaxed in xylene and rehydrated through a series of graded alcohols. Antigen retrieval was performed using the antigen retrieval citra plus solution (BioGenex Laboratories Inc., San Ramon, CA), and the endogenous peroxidase activity was quenched with 3% hydrogen peroxide. After permeabilizeation with Triton X-100 and blocking with 10% normal serum as well as avidin and biotin solutions (Vector Laboratories Inc., Burlingame, CA), the testis sections were incubated with primary antibody, including GCNA1, GATA4 (C-20) (Santa Cruz Biotechnology), or GFRA1 (N-18) (Santa Cruz Biotechnology) at 4°C overnight. After washes in PBS, the testis sections were incubated with the biotinylated second antibody and streptavidin-peroxidase and then detected by DAB chromogen (Zymed Laboratories Inc., South San Francisco, CA). The cell nuclei of the testes were counterstained with hematoxylin. Replacement of the primary antibody with PBS was used as a negative control.
Freshly isolated type A spermatogonia from immature mice, pachytene spermatocytes and round spermatids from adult mice, and a germ cell mixture minus the elongating spermatids from adult mice were cytospun onto slides. Type A spermatogonia were incubated with an antibody to GFRA1 (N-18) and an antibody to POU5F1 (C-10) at 4°C overnight after blocking with 10% normal serum. After three washes in PBS, the cells were incubated with fluorescein (FITC)-conjugated donkey anti-goat IgG and rhodamine-labeled IgG (Jackson ImmunoResearch Laboratories) for 1 h at room temperature. DAPI (4′,6′-diamidino-2-phenylindole) was used to stain the cell nuclei, and the cells were observed for epifluorescence using an Olympus Fluoview 500 Laser Scanning Microscope (Olympus, Melville, NY).
Double staining of the adult mouse germ cell mixture, pachytene spermatocytes, and round spermatids was carried out according to the procedure mentioned previously. Primary antibodies used here included anti-GFRA1 (N-18) and anti-germ cell nuclear antigen 1 (GCNA1). The secondary antibodies were Texas red dye-conjugated goat anti-rat IgG (Jackson ImmunoResearch Laboratories) and FITC-conjugated donkey anti-goat IgG.
Immunofluorescent staining of cell colonies obtained from immature mouse type A spermatogonia in culture on Sertoli cell feeder layers for 24 h also was carried out according to the method described above using an antibody to GFRA1 (N-18).
Forty-eight hours after transfection with Gfra1 siRNAs, an immunofluorescent analysis was performed on type A spermatogonia and Sertoli cells to evaluate expression changes of GFRA1 protein using an antibody to GFRA1 (N-18). DAPI was used to stain the nuclei of both type A spermatogonia and Sertoli cells.
Sertoli cells (1.6 × 105) were seeded in 12-well culture plates with DMEM/F12 supplemented with 1% antibiotics, and type A spermatogonia (0.4 × 105) were added to the culture at 3 h after the seeding of Sertoli cells. DMEM/F12 medium was changed at 72 h after plating. GDNF production from the culture medium was measured at 0, 12, 24, 48, 72, 96, 120, and 144 h, respectively, using the GDNF Emax ImmunoAssay System Kit (Promega, Madison, WI) pursuant to the manufacturer's instructions. The ELISA reaction was completed in 96-well plates (Costar Corp., Cambridge, MA), and the optical densities were recorded by the Bio-Rad Ultra Microplate reader at 450-nm wavelength. The ELISA experiments were repeated in triplicate for each sample, and GDNF production was described as mean ± SEM.
The 19-nucleotide siRNA sequences targeting mouse Gfra1 (nucleotides 1570–1588 and 2609–2627, respectively, GenBank entry BC04251) were designed using BLOCK-iT RNAi Designer (Invitrogen) and synthesized by Invitrogen. The sequences for targeting Gfra1 mRNA and Gfra1 siRNAs are listed in Table 2. The Stealth RNAi negative control obtained from Invitrogen was used as a control for monitoring nonsequence-specific effects. A total of 1.6 × 105 Sertoli cells were seeded in 12-well culture plates with DMEM/F12 supplemented with 10% FBS and 1% antibiotics, and 0.4 × 105 type A spermatogonia were added to the culture at 3 h after the seeding of Sertoli cells. The cells were classified into four groups (each group with triplicates): group 1 with Stealth RNAi negative control transfection (the control siRNA group), group 2 without siRNA transfection (the no-siRNA group), group 3 with Gfra1 siRNA-1 transfection, and group 4 with Gfra1 siRNA-2 transfection. Transfection with the BLOCK-iT fluorescent oligo (Invitrogen) was performed to assess the transfection efficiency of siRNAs. The Stealth RNAi negative control, Gfra1 siRNAs, and the BLOCK-iT fluorescent oligo were transfected into type A spermatogonia and Sertoli cells using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instruction, with certain modifications. Briefly, 50 pmol Stealth RNAi negative control, Gfra1 siRNA-1, Gfra1 siRNA-2, and the BLOCK-iT fluorescent oligo were diluted in 100 μl Opti-MEM I Reduced Serum Medium (Invitrogen) and then mixed gently. Lipofectamine 2000 (2 μl) was diluted in 100 μl Opti-MEM I Reduced Serum Medium and incubated for 15 min at room temperature. After a 15-min incubation, the diluted Stealth RNAi negative control, Gfra1 siRNAs, and the BLOCK-iT fluorescent oligo were combined with the diluted Lipofectamine 2000 and incubated for 15 min at room temperature to allow the oligomer-Lipofectamine 2000 complex formation. The culture medium was changed with 1 ml fresh DMEM/F12 without antibiotics before transfection, and the oligomer-Lipofectamine 2000 complexes were added to each well and mixed gently. After transfection, the cells were incubated at 34°C in a CO2 incubator, and fresh DMEM/F12 without antibiotics was added at 24 h following transfection. The cells were cultured and harvested at an appropriate time for assessing the gene silencing of Gfra1 and the phenotypic changes by combining semiquantitative RT-PCR, real-time quantitative RT-PCR, Western blots, and immunofluorescent analyses. Twenty-four hours after transfection, fluorescent uptake was observed under an immunofluorescence microscope.
Total RNA was isolated from the cells at 48 h after transfection with Gfra1 siRNAs or the negative control siRNA using the Trizol reagent (Invitrogen) and was treated with DNase to remove potential genomic DNA contamination using a DNA-free Kit (Ambion Inc., Austin, TX). Total RNA (1 μg) was reverse transcribed into the first-strand cDNA in a reaction primed by random hexamers according to the protocol of the manufacturer (Invitrogen). Primers for Gfra1 and Gapdh used for real-time quantitative PCR are listed in Table 1. Gapdh standard curves were generated by regression analysis of RT-PCRs performed on log10-diluted cDNA using the iCycler iQ real-time detection system (Bio-Rad Laboratories). Real-time quantitative PCR was conducted in a 25-μl reaction volume containing 0.5 μl cDNA template with the following conditions: 50°C for 2 min to prevent carryover contamination, 95°C for 10 min to activate hot-start polymerase, and 36 cycles of 95°C for 15 sec, 58°C for 30 sec, and 72°C for 30 sec. The iCycler software calculated the threshold cycle (CT), and the expression of Gfra1 mRNA relative to Gapdh mRNA in each sample was then calculated by ΔCT (ΔCT = CT[Gfra1] − CT[Gapdh]). The ratio of Gfra1 mRNA to Gapdh mRNA in the no-siRNA group was designated as the control (1.0), and it was used to calculate the level of Gfra1 mRNA relative to Gapdh mRNA in each sample by the comparative CT method (2−ΔΔCT) (ΔΔCT = [CT (Gfra1) − CT (Gapdh)]sample − [CT (Gfra1) − CT (Gapdh)]control). Real-time PCR was repeated three times for each sample, and mean values were used for comparison with the SEM. Statistically significant differences (P < 0.05) among Gfra1 mRNA expression levels obtained from Gfra1 siRNAs treatment groups and the negative control siRNA or no-siRNA group were determined by an ANOVA and Tukey post-hoc analysis.
Sertoli cells (1.6 × 104) were seeded in 96-well microtiter plates with 200 μl DMEM/F12 containing 10% FBS and 1% antibiotics, and type A spermatogonia (0.4 × 104) were added to the culture at 3 h after the seeding of Sertoli cells. The cells were transfected with or without 15 pmol Gfra1 siRNAs according to the procedure mentioned above. The medium was changed with fresh DMEM/F12 supplemented with 10% FBS at 12 h after transfection and then changed daily. The number of cell clusters in each culture well was counted on the fourth day. To evaluate the effect of Gfra1 siRNAs on the number of cells per cluster, a time course experiment of RNA interference was performed pursuant to the above procedure, and cell cultures were maintained for 5 days and analyzed for cluster size every day after siRNA transfection. The experiments were repeated in triplicate, and the data were presented as mean ± SEM. The statistically significant differences (P < 0.05) among the number and size of the clusters obtained from Gfra1 siRNA treatment groups and the negative control siRNA or no-siRNA group were determined by an ANOVA and Tukey post-hoc analysis.
We examined the expression of Gfra1 mRNA and GFRA1 protein in the freshly isolated seminiferous tubules and type A spermatogonia from immature mice, as well as in pachytene spermatocytes and round spermatids of adult mice, using RT-PCR and Western blots, respectively. RT-PCR analysis showed that Gfra1 mRNA was expressed in the seminiferous tubules and type A spermatogonia of immature mice (Fig. 1A). In contrast, Gfra1 mRNA was not detected in pachytene spermatocytes and round spermatids of adult mice (Fig. 1A). Furthermore, Western blots using an antibody to GFRA1 produced a 55-kDa band in the seminiferous tubules and type A spermatogonia of immature mice (Fig. 1B). Conversely, GFRA1 protein was undetected in pachytene spermatocytes and round spermatids of adult mice (Fig. 1B).
We further characterized the cellular localization of GFRA1 in testis sections of immature and adult mice; in the freshly isolated type A spermatogonia from immature mice; in the freshly isolated germ cells, pachytene spermatocytes, and round spermatids from adult mice; and in the colonies of type A spermatogonia cultured on a Sertoli cell feeder layer. GCNA1, a marker for germ cells , was used to stain type A spermatogonia in immature mouse testes (Fig. 2A), and GATA4, a marker for Sertoli cells, was used to stain the Sertoli cells (Fig. 2B). In the testes of 6-day-old mice, GFRA1 was localized to some but not all of the type A spermatogonia, whereas Sertoli cells were negative for this protein (Fig. 2C). The absence of GFRA1 in Sertoli cells in the immature mouse testes as shown by immunohistochemistry was confirmed by our findings that neither Gfra1 mRNA nor GFRA1 protein was detected in Sertoli cells in culture for 48 h as indicated by RT-PCR and Western blots, respectively (data not shown). In the testis sections from adult mice, GFRA1 was also localized to a subset of spermatogonia but not to differentiating germ cells, such as pachytene spermatocytes, round spermatids, and elongating spermatids (Fig. 2E).
Double staining showed that GFRA1 was co-expressed with POU5F1 in a subpopulation of type A spermatogonia (Fig. 3A). About 50% of the type A spermatogonia obtained from immature mice were positive for GFRA1 and POU5F1. The in vitro culture assay further revealed that germ cell colonies generated from type A spermatogonia cultivated on a Sertoli cell feeder layer contained the cells positive for GFRA1 (Fig. 3B). In addition, a subset of germ cells from adult mice were positive for GFRA1 (Fig. 3C), whereas pachytene spermatocytes (Fig. 3D) and round spermatids (Fig. 3E) were negative for GFRA1 but were positive for GCNA1.
To determine whether neonatal Sertoli cells in culture can secrete GDNF, we carried out an ELISA to measure the amount of GDNF released into the culture medium at various time points. As shown in Figure 4, the GDNF concentration increased during the first 3 days of culture, with a maximum of 258 pg/ml. After the culture medium was changed at 72 h, the GDNF concentration remained at a high level (more than 150 pg/ml) during the last 3 days of culture, indicating that Sertoli cells in culture are able to produce GDNF.
To knock down Gfra1 expression in type A spermatogonia, we used two siRNAs targeting different regions of Gfra1 mRNA. The BLOCK-iT fluorescent oligo, which has a proven correlation of transfection efficiency with siRNA molecules, was used to monitor the transfection efficiency of Gfra1 siRNAs. As shown in Figure 5, A and B, the transfection efficiency of Gfra1 siRNAs in type A spermatogonia was approximately 80%, as indicated by the uptake of the BLOCK-iT fluorescent oligo. Semiquantitative RT-PCR analysis showed that the expression of Gfra1 mRNA was reduced markedly in type A spermatogonia with Gfra1 siRNA treatment compared with the control siRNA or the no-siRNA group (Fig. 5C). Densitometric analyses displayed that Gfra1 mRNA expression was decreased by 60% and 70% in type A spermatogonia treated with Gfra1 siRNA-1 and Gfra1 siRNA-2, respectively (Fig. 5C). Real-time quantitative RT-PCR further confirmed that Gfra1 mRNA expression was reduced by 64% and 72% in type A spermatogonia treated with Gfra1 siRNA-1 and Gfra1 siRNA-2 (Fig. 5E). Moreover, Western blots revealed that the expression of GFRA1 protein was significantly decreased in type A spermatogonia with Gfra1 siRNA treatment (Fig. 5D). Densitometric analyses showed that GFRA1 protein was decreased by 70% and 80% in type A spermatogonia treated with Gfra1 siRNA-1 and Gfra1 siRNA-2 (Fig. 5D). This was further confirmed by immunofluorescent analysis showing that GFRA1 protein was reduced in type A spermatogonia when they were treated with Gfra1 siRNAs (Fig. 6).
We then evaluated the proliferation ability of type A spermatogonia after Gfra1 siRNA treatment using a proliferation assay. As shown in Figure 7, both the number and size of the cell clusters in the Gfra1 siRNA transfection groups were decreased statistically when compared with negative control siRNA and no-siRNA groups.
Next, we used specific markers to test whether Gfra1 knockdown in type A spermatogonia resulted in a decrease of proliferation and their differentiation into more mature spermatogonia and early spermatocytes. These markers included PCNA, POU5F1, and KIT. Semiquantitative RT-PCR analysis showed that the expression of both Pcna and Pou5f1 mRNA was decreased in type A spermatogonia with Gfra1 siRNA treatment, whereas the expression level of Kit mRNA was elevated (Fig. 8A). Western blot analysis confirmed that the expression of PCNA and POU5F1 proteins was reduced in type A spermatogonia after Gfra1 siRNA treatment. In contrast, the expression of KIT protein was increased in type A spermatogonia with the addition of Gfra1 siRNAs (Fig. 8B). Densitometric analyses revealed that Pcna mRNA and PCNA protein were decreased by 70% and 80% in type A spermatogonia treated with Gfra1 siRNA-1 and Gfra1 siRNA-2, and Pou5f1 mRNA and POU5F1 protein were reduced by more than 50% and 60% in the cell population with Gfra1 siRNA treatment. In contrast, a 3-fold increase of Kit mRNA and KIT protein was observed in type A spermatogonia treated with Gfra1 siRNA-1 and Gfra1 siRNA-2 compared with the control (Fig. 8, A and B).
We also tested whether type A spermatogonia with Gfra1 siRNA treatment differentiated into late spermatocytes and spermatids using various markers, such as Zfp42 (Rex-1), Mtl5 (tesmin), Prm2 (protamine 2), and Tnp1 (transition protein 1). We did not detect the expression of any of these products in type A spermatogonia treated with or without Gfra1 siRNA (Fig. 8, C and D), indicating that the differentiated cells did not display the phenotypes of spermatocytes and spermatids.
We further explored whether Gfra1 silencing of type A spermatogonia resulted in expression changes of RET phosphorylation. Western blot assay showed that the phosphorylation of RET (two glycosylated proteins of 170 kDa and 150 kDa) was significantly downregulated in type A spermatogonia treated with Gfra1 siRNAs compared with the negative siRNA group or the no-siRNA group (Fig. 8E). Densitometric analyses revealed that the phosphorylation of RET was decreased by 70% and 80% in type A spermatogonia treated with Gfra1 siRNA-1 and Gfra1 siRNA-2 (Fig. 8E).
There are divergent reports regarding the expression of GFRA1 receptor in the mammalian testes. It has been suggested that Gfra1 mRNA and GFRA1 protein are present in spermatogonia, pachytene spermatocytes, and round spermatids in rats  and in Sertoli cells and Leydig cells in humans . We characterized the expression of GFRA1 in the testis of immature and adult mice and explored its function in regulating proliferation and/or differentiation of type A spermatogonia in vitro. Using whole mounts of mouse seminiferous tubules, we previously demonstrated GFRA1 expression in undifferentiated spermatogonia . We now report that Gfra1 mRNA and GFRA1 protein are not expressed in mouse pachytene spermatocytes or spermatids, as indicated by RT-PCR, Western blots, and immunostaining. This differential expression pattern of Gfra1 mRNA and GFRA1 protein between mouse, rat, and human testes remains to be elucidated. The mechanisms that mediate the onset of spermatogenesis, especially the pathways induced by inactivation of growth factor receptors, are poorly understood. The rationale for our choice of Gfra1 as an interesting receptor to further examine its role in mediating type A spermatogonial proliferation and differentiation was on the basis of our observations that GFRA1 was expressed in a subpopulation of type A spermatogonia of immature mice and as a subset of spermatogonia of adult mice, but not in more differentiated germ cells, including pachytene spermatocytes, round spermatids, and elongating spermatids.
Although there are continuing debates on the identity of the true stem cells within the rodent testis, type A single (As) spermatogonia are generally considered to be the spermatogonial stem cells [27, 28]. The spermatogonial stem cell can divide into either two new As cells or type Apaired spermatogonia (Apr) that then produce aligned spermatogonia (Aal). Aal spermatogonia, in turn, give rise to several generations of spermatogonia, including type A1–A4, intermediate (In), and type B spermatogonia. Spermatogonial stem cells (As) and proliferative spermatogonia (Apr and Aal) are morphologically undifferentiated spermatogonia, whereas A1–A4, In, and type B spermatogonia are collectively considered to be differentiating cells [29, 30]. In another model for the mouse spermatogonial stem cell compartment, it is suggested that As are the actual stem cells with the ability of self-renewal, whereas Apr and Aal are potential stem cells that do not self-renew in normal spermatogenesis . In response to the loss of actual stem cells or an emptied stem cell niche, the potential stem cells would switch their mode from transit amplification to self-renewal and result in the genesis of new actual stem cells . It is worth noting that GFRA1 was co-expressed with POU5F1, a marker for spermatogonial stem cells [11, 32–34], in a subpopulation of type A spermatogonia, reflecting that GFRA1-positive cells within the seminiferous tubules were most likely the spermatogonial stem cells. This viewpoint is partially supported by our previous finding showing that GFRA1 was expressed in As and some Apr  cells. The co-expression of GFRA1 and POU5F1 in undifferentiated type A spermatogonia, together with the unique expression patterns of GFRA1 in the testis of immature and adult mice, suggests that GFRA1 plays a pivotal role in maintaining spermatogonial stem cells in an undifferentiated state. Furthermore, germ cell colonies generated from type A spermatogonia were positive for GFRA1, indicating that GFRA1-positive colony cells have a potential to proliferate in vitro, which is a characteristic of stem cells.
Since GDNF is produced by Sertoli cells within the testes [10, 22], we also anticipated to observe GDNF production by Sertoli cells in culture. In this study we have demonstrated that Sertoli cells in culture can produce GDNF for at least 6 days, as shown by an ELISA, which is consistent with the recent finding that GDNF is detected in conditioned medium derived from Sertoli cells by Western blot analysis . Thus, we did the siRNA assay in a coculture system containing type A spermatogonia and Sertoli cells with no addition of exogenous GDNF.
We have for the first time used RNA interference (RNAi) to do a loss-of-function study of Gfra1 in type A spermatogonia of 6-day-old mice. The major obstacle to achieving RNAi with double-strand RNAs longer than 30 nucleotides in mammalian cells was that it resulted in nonspecific RNA degradation and the shutdown of host cell protein synthesis. This obstacle has been overcome by using in vitro-synthesized siRNA with less than 23 nucleotides to trigger gene suppression without causing interferon response [18, 36]. In the current study, we used 19-nucleotide Gfra1 siRNAs and improved confidence of the RNAi data for sequence specific off-target effects by individually transfecting two Gfra1 siRNAs targeting different sequences of Gfra1 mRNA. The negative control siRNA purchased from Invitrogen provides a good way to measure the effect of the experimental RNAi duplex versus background. This negative control is not homologous to anything in the vertebrate transcriptome and has been shown not to induce a stress response. The Lipofectamine 2000, which was originally developed for the delivery of plasmid DNA, also works effectively with siRNAs . We found that siRNAs against Gfra1 could be efficiently transfected into type A spermatogonia using Lipofectamine 2000, as shown by a high uptake of the BLOCK-iT fluorescent oligo, whose fluorescent signal correlates with the delivery of active siRNAs.
Our results showed that Gfra1 siRNAs could knock down the expression of Gfra1 mRNA and GFRA1 protein in type A spermatogonia derived from immature mice, as shown by RT-PCR and Western blots, respectively. More importantly, the reduction of Gfra1 expression caused by Gfra1 siRNAs induced a decrease of proliferation of spermatogonial stem cells and their phenotypic differentiation, as evidenced by a proliferation assay, as well as an elevated expression of KIT and the decreased expression of PCNA and POU5F1. PCNA has been used extensively in the identification of proliferating spermatogonia and early spermatocytes in the testes of rodents and primates , and POU5F1 is regarded as a marker for spermatogonial stem cells [32–34]. On the contrary, KIT has been used as a hallmark for differentiating spermatogonia, and type A1–A4 spermatogonia divisions are considered to be KIT dependent [39–43]. Our results suggest that after the knockdown of Gfra1 in undifferentiated type A spermatogonia, the cells differentiated into more mature germ cells, probably including the KIT expressing type A1–A4 spermatogonia.
During spermatogenesis in mice, primary spermatocytes start to appear at approximately 10 days postpartum [20, 44]. We next determined whether differentiating spermatogonia further differentiated into late spermatocytes and spermatids using spermatocyte-specific markers, such as Zfp42 and Mtl5, as well as spermatid-specific markers, including Prm2  and Tnp1 . It has been reported that Zfp42 expression is limited to spermatocytes, as shown by in situ hybridization and Northern blot analyses of stage-specific germ cell preparations , and Mtl5 is specifically expressed in spermatocytes of adult mouse testis . In this study, the mRNA expression of Zfp42, Mtl5, Prm2, and Tnp1 was undetected in the differentiating cells, indicating that the spermatogonial stem cells did not differentiate into late spermatocytes or spermatids. Considered together, our data suggest that the downregulation of Gfra1 and subsequent partial block of GDNF/GFRA1 signaling pathway may cause the differentiation from As to A1–A4, but not to spermatocytes or spermatids.
GFRA1 is a necessary component of the GFRA1/RET complex, and RET alone is unable to bind GDNF unless it is co-expressed with the GFRA1 receptor . The association of GDNF with RET tyrosine kinase is mediated by GFRA1, and GDNF cannot induce RET autophosphorylation in the cells that lack GFRA1 expression . Our results using Western blot analysis clearly showed that the knockdown of Gfra1 by Gfra1 siRNAs resulted in a decrease of RET phosphorylation, confirming that GFRA1 is required for the activation of RET tyrosine kinase by GDNF. The partial inactivation of RET tyrosine kinase by Gfra1 silencing may sequentially block intracellular signaling and allow for the differentiation of the spermatogonial stem cells.
In summary, we have demonstrated that GFRA1 is expressed dominantly in mouse spermatogonial stem cells and that Gfra1 silencing leads to a switch from proliferation and self-renewal to differentiation of spermatogonial stem cells into A1–A4 spermatogonia through the inactivation of RET tyrosine kinase. This study thus offers a novel insight into understanding the initial stages of mouse spermatogenesis and also sheds further light on how to maintain spermatogonial stem cells in an undifferentiated state.
We thank Dr. Robert J. Lechleider, National Cancer Institute, National Institutes of Health, for his helpful suggestions regarding this study. We are also grateful to Dr. George C. Enders, Department of Anatomy and Cell Biology, University of Kansas Medical Center, for providing an antibody to GCNA1.
1Supported by National Institutes of Health grants HD33728 to M.D. and HD044543 to M.C.H.