Src family non-receptor tyrosine kinases are involved in signaling pathways which mediate cell growth, differentiation, transformation and tissue remodeling in various organs. In an effort to elucidate functional involvement of p60c-Src (c-Src) in spermatogenesis, the postnatal changes in c-src mRNA and c-Src protein together with kinase activity and subcellular localization were examined in mouse testes. c-src mRNA levels in testes increased during the first 2 weeks of postnatal development (PND). Following a decrease at puberty (PND 28), the c-src mRNA levels re-increased at adulthood (PND 50). Src kinase activity of testes was low at PND 7 but sharply increased prepubertally (PND 15) and highest at adulthood. Upon Western blotting, the level of c-Src protein was the highest in prepubertal testes but rather decreased in adult testes at PND 50. In adult testes, ubiquitination of c-Src proteins was apparent compared with immature one at PND 7, suggesting active turnover of c-Src by ubiquitination. In immature testes, c-Src immunoreactivity was largely found in the cytoplasm of the Sertoli cells. By contrast, in pubertal and adult testes intense immunoreactivity was localized at the adluminal and basal cytoplasm of Sertoli cells bearing elongated spermatids and early germ cells, respectively. The immunoreactivity of c-Src in the Leydig cells was increased during pubertal development, suggesting the functional involvement of c-Src in differentiated adult Leydig cells. Throughout postnatal development, some spermatogonia and spermatocytes showed intensive c-Src immunoreactivity compared with other germ cells, suggesting a possible role of c-Src in germ cell death. Taken together, it is suggested that c-Src may participate in the remodeling of the seminiferous epithelia and functional differentiation of Leydig cells during the postnatal development of mouse testes.
Src family non-receptor tyrosine kinases (NRTKs) are involved in various events such as cell survival, proliferation, apoptosis, differentiation, transformation and tissue remodeling in various organs (Sorge et al. 1984; Novotny-Smith & Gallick 1992; Nishio et al. 1995; Tsunoda et al. 1996; Masaki et al. 1998; Puceat et al. 1998; Zhong et al. 2002). The cellular proto-oncogene product p60c-Src, serves as the prototype for a family of eight mammalian kinases including Blk, c-Fgr, Fyn, Hck, Lck, Lyn and c-Yes. A wide range of biologically active ligands result in increased Src kinase activity, which is important for signaling from transmembrane receptors lacking tyrosine kinase activity, or further amplifies tyrosine phosphorylation events following engagement of receptor tyrosine kinases. The Src kinase family is a well-known example of functional complementation. Most cells express more than one member of the Src family, and a loss of an individual Src family member can be compensated for by other Src family members (Lowell et al. 1996). To date, several NRTKs including ferT, c-Src, Fyn and Mak were found in testes (Keshet et al. 1990; Matsushime et al. 1990; Yoshinaga et al. 1991; Jinno et al. 1993; Nishio et al. 1995; Berruti & Borgonovo 1996; Kharbanda et al. 1998; Vincent et al. 1998; Maekawa et al. 2002). Basically, spermatogenesis is characterized by the extensive remodeling of seminiferous epithelia accompanying the proliferation, cell death, postmeiotic differentiation and morphogenetic movement of germ cells across the Sertoli cells toward the lumen of the seminiferous tubule, and detachment of germ cells from the seminiferous epithelia by the process of spermiation (de Kretser & Kerr 1988; Kierszenbaum 2001). Among Src kinases, p60c-Src has been known to function in the rearrangement of cytoskeletons and junctional complexes, and germ cells death in rat testes (Nishio et al. 1995; el-Sabeawy et al. 1998; Wang et al. 2000; Jindo et al. 2001; Lee & Cheng 2005). It was also emphasized that Src kinase signaling involved in the mechanism desensitizing Leydig cells to luteinizing hormone (LH) stimulation and thus regulates phosphodiesterase activity and steroid production in vitro (Taylor et al. 1996, 1997; Taylor 2002). Recently, C-terminal Src kinase (Csk) which phosphorylates c-Src and thus control Src kinase activity and stability is found in mouse testes (Gye et al. 2004; Lee & Cheng 2005), suggesting tight regulation of Src activity by Src-Csk loop in testes. This body of evidences suggests that Src family proteins are involved in control of various aspects of spermatogenesis in cooperation with many regulatory proteins. To date, however, the spatiotemporal change in the expression of c-src and its kinase activity during postnatal development of testes was uncovered in mice.
Materials and Methods
Testes were removed from postnatal day (PND) 7, 15, 28, and 50 ICR male mice under asphyxiation in CO2. After washing in phosphate buffered saline (PBS), testes were decapsulated, frozen in liquid nitrogen, and subjected to RNA and protein extraction (n = 4 each). For immunohistochemical analysis, testes were fixed in Bouin solution overnight and processed for paraffin section.
RNA preparation and reverse transcription polymerase chain reaction (RT-PCR)
Frozen tissues were ground using a ceramic grinder. Total RNA was isolated using TRIzol (Gibco BRL, Gaithersuburg, MD, USA). The RNA samples (4 µg) were reverse transcribed for 50 min at 42°C in a 20 µl reaction with 200 units of SuperScriptII reverse transcriptase and 0.5 µg of oligo(dT)12–18 primer by the standard protocol of the supplier (Life Technologies, Rockville, MD, USA). The primers for mouse c-src were designated 5′-CTGCTGGACTTTCTCAAGGG-3′ (forward) and 5′-GTACAGAGCAGCTTCAGGGG-3′ (reverse) according to mouse c-src cDNA sequence (GenBank accession number NM_009271). For semiquantitative analysis of c-src mRNA expression, the G3PDH transcript was amplified as an internal control. The primers for mouse G3PDH were 5′-AGTGGAGATTGTTGCCATCAACGA-3′ (forward) and 5′-GGGAGTTGCTGTTGAAGTC GCAGGA-3′ (reverse) according to mouse G3PDH cDNA sequence (GenBank accession number M32599). These primer sets gave rise to c-src and G3PDH diagnostic fragments of 273 and 791 bp, respectively. PCR was subsequently performed using Ex Taq polymerase (Takara, Otsu, Japan) an optimized protocol consisting of between 20 and 38 cycles. Each cycle consisted of the following: 95°C, 30 s; 60°C, 30 s; 72°C, 45 s. The increase in RT-PCR products of c-src was linear in the 26–38 cycles of amplification (Fig. 1A,B). Thus, the number of the amplification cycle was fixed at 30 thereafter. As negative control for c-src RT-PCR, RNA without RT reaction was amplified (Fig. 1A). The PCR products were run on 2% agarose gels containing 0.5 µg/mL ethidium bromide and photographed under ultraviolet (UV) light. PCR product for each gene was subcloned into pGEM-TEasy vectors and automatically sequenced.
Western blotting analysis
Decapsulated testes were homogenized in 10 volumes of extraction buffer (20 mm Tris-HCl, pH 7.5, 1% NP40, 0.25% sodium deoxycholate, 150 mm NaCl, 1 mm ethylenediamine tetraacetic acid, 1 mm Na3VO4, 1 mm NaF) containing a protease inhibitor cocktail (Complete, Roche, Mannheim, Germany) on ice. Homogenate was centrifuged at 12 000 g for 1 h at 4°C. The supernatant was collected and subjected to determination of protein concentration using commercial protein assay kit (Bio-Rad, Richmond, CA, USA). Protein samples or c-Src immunoprecipitates were mixed with equal volumes of twofold sample buffer (Laemmli 1970) and resolved on 12% sodium dodecylsulfate (SDS)-polyacrylamide gels. Following electrophoretic transfer to enhanced chemiluminescence (ECL) membrane (Amersham Bioscience, Buckinghamshire, UK), the blot was probed with c-Src antibody (1:1000 dilution in Tris-buffered saline (TBS)) raised against a peptide mapping at the C-terminus of p60c-Src of human origin (sc-18, Santa Cruz Biotechnology, Santa Cruz, CA, USA) or ubiquitin antibody (mAb P4D1, Santa Cruz Biotechnology). The blot was incubated with horseradish peroxidase-labeled antirabbit or antimouse IgG (Amersham Bioscience) diluted 1:1000 in TBS. Coloring reaction was done using an ECL kit (Amersham Bioscience). As an internal control, beta actin expression was analyzed in parallel blots using the beta actin antibody (sc-8432, Santa Cruz Biotechnology).
Immunoprecipitation and kinase assay
Decapsulated testes were homogenized with 20 volumes of radioimmunoprecipitation assay (RIPA) buffer (50 mm Tris, pH 7.5, 150 mm NaCl, 5 mm ethyleneglycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA), 250 µm sodium orthovanadate, 0.1% NP-40, 0.1% sodium deoxycholate, 10 µg/mL leupeptin, 10 µg/mL aprotinin, 10 µg/mL pepstatin) on ice. Homogenates were centrifuged at 12 000 g for 1 h at 4°C. The supernatant was collected and subjected to immunoprecipitation experiments. Briefly, 100 µL of supernatant was mixed with 1 mg of rabbit c-Src antibody (Sc8056, Santa Cruz Biotechnology) and incubated for 60 min at 4°C with agitation. 100 µL of the protein A/G plus agarose solution (Sigma Chemical) was added and incubated for 60 min at 4°C with agitation. At the end of the incubation, protein A/G-agarose-immune complexes were rinsed threefold with assay dilution buffer (ADB, 1 mm sodium orthovanadate, 25 mmβ-glyceralphosphate, 5 mm EGTA, 1 mm dithiothreitol (DTT), 20 mm 3-(N-morpholino) propanesulfonic acid (MOPS), pH 7.2) and subjected to kinase assay or Western blotting. Reaction mixture consisted of 50 µm Src kinase substrate peptide (KVEKIGEGTYGVVYK), 5 µL of kinase inhibitor cocktail (20 µm H7, 2 µm H8, and 20 µm R24571), 5 µL of magnesium-adenine triphosphate (ATP) solution, 5 µL of r-(32P) ATP (10 µCi/µL), and 5 µL of protein extracts. After incubation at 37°C for 10 min, reaction was terminated by placing the tube in ice. After spun centrifugation, 20 µL of supernatant was applied on the p81 phosphocellulose paper (PC) and followed by washing threefold in 0.75% phosphoric acid solution. After drying the PC paper, radioactivity was measured by scintillation counting. As a blank, substrate-free reaction mix was measured. The count per minute (CPM) value of the sample was subtracted by that of the blank. The mean values from three trials were statistically analyzed by Student's t-test.
Paraffin sections (5 µm) of testes were processed for immunohistochemical analysis (n = 3 each). Briefly, after the clearing the paraffin endogenous peroxidases were blocked with 3% H2O2 for 15 min. The sections were incubated for 5 min with a PBS containing 0.1% Tween 20 (PBST) and then incubated with c-Src antibody (sc-18) at a 1:200 dilution in PBST containing 0.1% bovine serum albumin (BSA) for 2 h at room temperature (RT). After rinsing with PBST, the sections were incubated with biotinylated antirabbit IgG (Vector Laboratories, Burlingame, CA, USA) dilution 1:200 in PBST containing 1:20 dilution of normal serum included in the Vectorstain ABC kit (Vector Laboratories) for 1 h at RT and then incubated with avidin-peroxidase complex for 30 min at RT. 3,3-diaminobenzidine and H2O2 were used as the chromogen. As a negative control, non-immune rabbit IgG was used instead of c-Src antibody. Following the counterstaining with Mayer hematoxylin the slides were mounted permanently. Observation and photography were performed with BX50 microscope (Olympus, Tokyo, Japan).
Image analysis and statistics
Band intensity of RT-PCR product and Western blot were analyzed using the Bioprofil (Vilber Lourmat, Marne-la-Vallée, France) and the relative amount of c-src versus internal control was plotted. The statistical significance of the results was determined by Student's t-test.
c-Src protein expression in testes during postnatal development
To determine whether the expression of c-src is developmentally regulated in testes, we examined its expression during the postnatal development of testes by semiquantitative RT-PCR. The amplification curves of RT-PCR product of c-src mRNA (273 bp) showed a linear increase between 26 and 38 cycles of amplification (Fig. 1A,B), suggesting relative abundance of c-src mRNA in the adult testes. Thereafter, 30 cycles of amplification were conducted for semiquantitative RT-PCR analysis of c-src mRNA levels. c-src mRNA levels in the testes increased during the first 2 weeks of development after birth (PND 7 and 15). Following a decrease at the pubertal stage (PND 28), testicular c-src mRNA levels re-increased at the adult stage (PND 50) (Fig. 1C,D). There was no specific amplification product in the negative control experiment using no RT-reaction (right panel in Fig. 1A). In Western blot, c-Src protein levels was the highest at PND 15 and the lowest at PND 28. In adult testes, c-Src protein levels was a little higher than the pubertal one (Fig. 2A,B).
Changes in Src kinase activity during postnatal development of testes
Src kinase activity phosphorylating the Src kinase substrate peptides remained low until PND 7 but sharply increased at PND 15. Following a little decrease at PND 28, Src kinase activity peaked at adulthood (PND 50) (Fig. 3).
Ubiquitination of c-Src in testes
Western blotting of c-Src immunoprecipitate of testes extract using ubiquitin antibody showed ubiquitinated c-Src in adult testes at PND 50. There was no visible signal for ubiquitinated c-Src in the immature testes at PND 7 (Fig. 4).
Immunohistochemical localization of c-Src in mouse testes
Throughout the postnatal development, c-Src immunoreactivity was found in both seminiferous epithelia and Leydig cells (Fig. 5A–D). In PND 7 and 15 testes, moderate c-Src immunoreactivity was found in the Sertoli cells and Leydig cells. There was no visible difference in signal intensity among the seminiferous tubules in immature testes (Fig. 5A,B). In pubertal (PND 28) and adult testes (PND 50), adluminal cytoplasm adjacent to the elongating spermatids and basal cytoplasm of the Sertoli cells showed intense c-Src immunoreactivity compared with earlier stage germ cells (Fig. 5C–H). In adult testes (PND 50), the immunoreactivity of c-Src in the seminiferous epithelia showed remarkable heterogeneity among the spermatogenic stage. Throughout the postnatal development, nuclei of some germ cells showed an intensive immunoreactivity compared with others, and c-Src positive germ cells were more frequently observed in the immature (PND 9 and 15) testes than the adult testes. In adult testes, c-Src-positive germ cells were largely pachytene spermatocytes and round spermatids in stage IX seminiferous epithelia (Fig. 5E). A concentrated area of immunoreactivity was also found at the basal sites of the epithelia (Fig. 5E–H). In Leydig cells, c-Src immunoreactivity in pubertal testes onwards was stronger than that of the immature ones (insets in Fig. 5A–D). In a negative control experiment using non-immune rabbit IgG, no specific signal was found in the seminiferous tubule (Fig. 5I).
There was a large increase in the expression of c-src during postnatal development until 2 weeks after birth. Consistently, the expression of c-src increases during the first few weeks of postnatal development in rat testes (Nishio et al. 1995). In mouse testes, meiotic prophase begins at approximately PND 9 with the appearance of preleptotene spermatocytes. The majority of cells reach early and mid-pachytene stage by PND 15 (Bellve et al. 1977; de Rooij 2001). Meanwhile, Sertoli cells, which express a large amount of c-Src, rapidly proliferate during the first two weeks after birth in mice (Nishio et al. 1995; Baker & O'Shaughnessy 2001). Recently, Tokuchi et al. (1999) reported the expression of c-src transcript in the Sertoli cells but not in the purified pachytene spermatocytes and spermatids. Taken together, this increase in c-src mRNA during neonatal to prepubertal development may largely attribute to proliferation of Sertoli cells. In testes, c-Src participates in negative regulation of LH receptor signaling in Leydig cells (Taylor 2002). In mice, the number of Leydig cells reaches approximately half of their adulthood number during the first 2 weeks of postnatal development (Baker & O'Shaughnessy 2001). Therefore, it cannot be excluded that the increase in c-src mRNA in testes could be partly attributed to the proliferation of Leydig cells.
Following the relatively high levels of c-src mRNA during the first few weeks after birth, testicular c-src expression decreased at puberty (PND 28). In mice, the first postmeiotic germ cells appear between 17 and 22 days after birth, thereby signifying the onset of spermiogenesis (Nebel et al. 1961; Bellve et al. 1977). At this time, Sertoli cells cease proliferation and begin terminal differentiation. Therefore, the decrease in testicular c-src mRNA levels in PND 28 testes could be attributed to the increase of germ cells expressing relatively lesser amounts of c-Src. Although c-Src protein levels in PND 7 testes was not vastly different from adult testes, Src kinase activity was markedly lower than others. Because LH receptor signaling activate c-Src in Leydig cells, this may be partly due to the lack of LH support or the inactive state of immature Leydig cells at this stage in mice (Taylor et al. 1996; O'Shaughnessy et al. 2002). At PND 15, c-Src kinase activity sharply increased, suggesting onset of LH signaling. Unexpectedly, c-Src kinase activity at puberty (PND 28) was a little lower than that of the prepubertal (PND 15) testes. Similarly, Src compensation of spermatogenic defect in fyn-deficient mice is insufficient at puberty (Maekawa et al. 2002). The Src kinase family is a well-known example of functional complementation. Most cells express more than one member of the Src family, and a loss of an individual Src family member can be compensated for by other Src family members (Lowell et al. 1996). This suggests possible development of mechanisms for inhibiting c-Src activity during prepubertal to pubertal stages in testes. The expression of cytoplasmic protein tyrosine phosphatase PTP-RL10/PTPD1/PTP2E, as well as receptor protein tyrosine phosphatase PTP-RL10 associated with the c-Src, is developmentally regulated in rodent testes (Mauro et al. 1994; Tokuchi et al. 1999), suggesting possible downregulation of testicular c-Src kinase activity at puberty by protein phosphatases in testes. To date, however, the reason for downregulation of testicular c-Src kinase activity during pubertal development is still obscure.
In adult testes at PND 50, c-src mRNA levels and Src kinase activity were significantly higher than that of the pubertal testes but c-Src protein levels were not. Basically, Csk is abundant in Sertoli cells as well as germ cells in mouse testes (Gye et al. 2004; Lee & Cheng 2005), and phosphorylation of C-terminal tyrosine residue of c-Src by Csk decreases the kinase activity of c-Src. Conversely, dephosphorylation of c-Src causes an increase in its kinase activity (Cooper & King 1986; Okada & Nakagawa 1989; Zheng et al. 1992). The activated form of c-Src is less stable than the inactive form and prone to ubiquitin-mediated degradation (Harris et al. 1999). In our results, Western blot of testicular c-Src immunoprecipitates revealed apparent increase in ubiquitinated c-Src in adult testes compared with immature one. Therefore, it is strongly suggested that relatively high c-Src kinase activity but decreased amount of c-Src protein in adult testes may attribute to the active turnover of the activated form of c-Src protein. It should be also emphasized that c-Src kinase activity is tightly regulated by ubiquitin-mediated degradation of c-Src in testes.
The developmental change in c-Src expression in testes might be of particular interest concerning the remodeling of the seminiferous tubule during spermatogenesis and spermiogenesis. Basically, spermatogenesis in adult testes is characterized by the extensive remodeling of seminiferous epithelia accompanying the proliferation, cell death, postmeiotic differentiation and morphogenetic movement of germ cells across the Sertoli cells toward the lumen of seminiferous tubule, and detachment of germ cells from seminiferous epithelia by a process of spermiation (de Kretser & Kerr 1988; Kierszenbaum 2001). c-Src plays pivotal roles in the rearrangement of cytoskeletons and junctional complexes, and germ cell death in rat testes (Nishio et al. 1995; Wang et al. 2000; Jindo et al. 2001; Lee & Cheng 2005). Therefore, an increase in c-src mRNA expression and c-Src kinase activity in adult testes possibly reflects a functional requirement of c-Src during the terminal differentiation of seminiferous epithelia. No visible difference was found in the c-Src immunoreactivity among the seminiferous tubules in immature testes (PND 7 and 15). In contrast, in the pubertal and adult testes, subcellular localization of c-Src in Sertoli cells was different among the seminiferous tubules. In adult testes, c-Src signal in the adlumnal cytoplasm of Sertoli cells bearing the elongated spermatids was stronger than that bearing earlier stage germ cells. This suggests that the functional requirement for c-Src during the remodeling of the seminiferous epithelia during spermiogenesis. In rat seminiferous epithelia, c-Src is found in the ectoplasmic specialization, a unique junctional complex formed near the base of the seminiferous epithelium forming the blood–testis barrier and lumen of the seminiferous tubule embracing the acrosome region of the developing spermatids (Wang et al. 2000; Mulholland et al. 2001; Toyama et al. 2003; Lee & Cheng 2005). Concentrated areas of c-Src immunoreactivity were also found at the basal side of the seminiferous epithelia. Basically, c-Src is involved in the cell contact with extracellular matrix via activation of focal adhesion kinase (FAK) which is a direct target of c-Src (Clark & Brugge 1995; Schaller et al. 1999; Chaudhary et al. 2002; Siu et al. 2003). In the basement membrane of seminiferous epithelia, extracellular matrix proteins including laminin, fibronectin, collagen IV and collagen I, contacts with Sertoli cells, spermatogonia and early spermatocytes (Yazama et al. 1997). This suggests a role of c-Src in the adhesion between basement membrane and seminiferous epithelia. At the beginning of meiosis, leptotene spermatocytes detached from basal lamina and move across the Sertoli cells toward the adluminal compartment of seminiferous tubule. This morphogenetic movement of germ cells is largely driven by Sertoli cells and requires remodeling of inter-Sertoli junctional complex as well as cellular structures made between germ cells and the basement membrane (Russell 1977; Ulvik 1983; de Kretser & Kerr 1988). Complex paracrine and enzymatic interactions between the germ cells and soma are crucial for this remodeling process (Monsees et al. 1997; Siu & Cheng 2004). Therefore, c-Src kinase activity as well as the expression should be tightly regulated for consistent remodeling of seminiferous epithelia.
Interestingly, strong c-Src immunoreactivity was found in the nuclei of some germ cells and the incidence of c-Src-positive germ cells was relatively higher in the immature testes. In adult testes, c-Src-positive germ cells were largely pachytene spermatocytes and round spermatids in stage IX seminiferous epithelia. Basically, germ cell apoptosis normally occurs in testes to ensure the production of normal sperm (Print & Loveland 2000; Yan et al. 2000; Kierszenbaum 2001). In mice, the peaks in apoptosis of spermatogenic cells occur at 13 days of gestation and around 10–13 days after birth (Wang et al. 1998). Several stages of spermatocytes are vulnerable to toxicant-induced cell death and c-Src might be involved in physiological as well as toxicant-induced germ cell death in rat testes (Wang et al. 2000; Jindo et al. 2001). Similarly, in 3Y1 rat fibroblasts, c-src overexpression results in the generation of apoptotic signals (Zhong et al. 2002). Therefore, it is tempting to speculate that c-Src may be also involved in germ cell death in mouse testes.
In Leydig cells, c-Src immunoreactivity in the pubertal and adult testes was markedly higher than the immature ones. In mice, fetal Leydig cells arise soon after testicular differentiation, at about 12.5 days postcoitum and after birth. Following the fetal secretion of androgen, Leydig cells revert to a relatively undifferentiated state until they are again activated at puberty. The adult Leydig cells arise after birth, proliferate, terminally differentiate and secrete testosterone which induces male behavior, secondary sexual characteristics and fertility. During prepubertal development until PND 15, adult Leydig cells largely populate mice testes. However, they are not fully differentiated and secrete little steroids, although this could be attributed to the lack of LH support at this stage (O'Shaughnessy et al. 2002). Src regulates production of steroids by modulating the phosphodiesterase activity in LH receptor signaling pathways, crucial for function of adult Leydig cells (O’Connell et al. 1996; Taylor et al. 1996, 1997). Therefore, increased c-Src immunoreactivity in Leydig cells in pubertal testes onward may contribute to functional differentiation of adult Leydig cells and fulfill the role of c-Src as a modulator for LH receptor signaling in adult Leydig cells.
In summary, spatiotemporal change in the expression and kinase activity of c-Src in mouse testes suggests a functional involvement of c-Src in the multiple aspects of the remodeling of the seminiferous epithelia and germ cell death during spermatogenesis, and differentiation of Leydig cells during the postnatal development of mouse testes.