| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Testis |
Department of Biochemistry and Cellular and Molecular Biology, University of Tennessee, Knoxville, Tennessee 37996-0840
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
kinases, meiosis, spermatogenesis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reversible protein phosphorylation is universally involved in regulation of both mitotic and meiotic cell divisions [1]. Indeed, during meiosis, protein phosphorylation accompanies synaptonemal complex disassembly, chromosome condensation, and nuclear envelope breakdown [2– 5], and there is a dramatic increase in levels of phosphorylated proteins as spermatocytes exit prophase and enter metaphase (Cobb and Handel, unpublished results). These observations imply that reversible protein phosphorylation plays an important role in the regulation of meiotic divisions during spermatogenesis. However, although phosphorylation reactions are presumed essential, the roster of specific kinases required for the successful completion of the G2/MI transition during male meiosis is not yet known.
Experimental analysis of the G2/MI transition in spermatocytes has been difficult, in part because of the low frequency of testicular spermatocytes progressing through this transition but also because of the lack of unique size characteristics that would facilitate their purification. This difficulty was partially overcome by the discovery that okadaic acid (OA) treatment of pachytene spermatocytes in vitro overrides the normal checkpoints that delay entry into metaphase I and induces a precocious MI-like state, characterized by condensation of bivalent metaphase chromosomes and synaptonemal complex breakdown [2]. OA is thought to act indirectly through activation of maturation promoting factor (MPF). MPF is a complex of two subunits, the catalytic p34cdc2 (CDK1) subunit and the CYCLIN B1 regulatory subunit. Activation of MPF occurs through a two-step process involving the phosphorylation of Thr161 by the CDK-activating kinase and the dephosphorylation of Thr14 and Tyr15 by a CDC25 phosphatase [6, 7].
MPF is suspected to be required for the meiotic G2/MI transition of spermatocytes in vitro since pachytene spermatocytes pretreated with butyrolactone I, a potent and specific inhibitor of the cyclin-dependent kinases, did not condense bivalent chromosomes in response to OA treatment [4]. These observations, however, are also consistent with a role for CDK2, which has recently been shown to be required for the meiotic division of spermatocytes [8, 9]. The role of CDC25C in MPF activation has recently come into question. Mice homozygously null for the Cdc25c gene have no abnormal phenotype and are fertile [10], suggesting that CDC25C is not required or that there are redundant phosphatases that substitute in its absence. The requirement for CDC25C in the spermatocyte's OA-induced G2/MI transition is investigated here.
Evidence suggests that MPF may be necessary but is not sufficient for the meiotic G2/MI transition in spermatocytes. Phosphorylation of histone H3, known to be a hallmark of chromosome condensation, was unaffected after butyrolactone I treatment, suggesting both that phosphorylation of histone H3 is not sufficient for condensation of chromosomes and that CDK1 and CDK2 are not involved in the regulation of histone H3 phosphorylation [4]. If MPF or another butyrolactone I-inhibited CDK is not required for histone H3 phosphorylation, at least one other kinase must play a key role in the spermatocyte's G2/MI transition.
The extracellular signal-regulated protein kinases (ERKs), more commonly known as the mitogen-activated protein kinases (MAPKs), are implicated in regulating cell division processes. ERK1 and ERK2, the two main isoforms of the MAP kinases, are serine/threonine kinases that are activated in response to dual phosphorylation on threonine and tyrosine residues by the MAPK kinases, MEK1/ 2. Activation of MEK1/2 has been shown to occur by two separate pathways. In mitotically dividing cells, MEK1/2 is activated by an extracellular signaling pathway containing RAS and the MAPK kinase kinase, RAF. However, in oocytes, MOS, the product of the Mos "proto-oncogene," is required for MAPK activation during meiosis [11, 12]. Germ cells in male mice lacking the MOS protein progress normally through spermatogenesis without any obvious impairment in fertility, revealing different requirements for regulation of the division phase in male and female meiosis [13–15]. Nonetheless, recent evidence suggests a role for MAPKs in chromatin condensation during the first meiotic division of male germ cells [16, 17].
The present study was designed to investigate the dynamics of kinases potentially involved in regulating the meiotic G2/MI transition of mouse spermatocytes, with particular emphasis on requirements for the activation of the ERK1/2 MAPKs during this transition. The results provide further insight to the mechanism(s) by which OA stimulates a precocious G2/MI transition. Use of genetic models precludes a role for either CDC25C or MOS in the OA-induced G2/MI transition and activation of MPF and the ERK1/2 MAPKs. Data on both expression and inhibition of MAPK pathway molecules reveal complex and possibly overlapping signal transduction pathways regulating the naturally occurring G2/MI transition and the OA-induced transition in spermatocytes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ICR mice were obtained from Harlan (Indianapolis, IN); those carrying the Mostm1Ev (hereafter designated Mos–) and Cdc25ctm1Hpw (hereafter designated Cdc25c–) targeted null alleles were generously provided by Drs. John Eppig (The Jackson Laboratory, Bar Harbor, ME) and Helen Piwnica-Worms (Washington University, St. Louis, MO), respectively. Mice were maintained under standard conditions in accordance with the National Institutes of Health and U.S. Department of Agriculture standards.
Mixed germ cell preparations were obtained by enzymatic digestion of adult testes. The testes were detunicated and digested in 0.5 mg/ml collagenase in Krebs-Ringer bicarbonate (KRB) buffer at 32°C for 20 min, followed by a subsequent digestion in 0.5 mg/ml trypsin at 32°C for 13 min. After digestion, the germ cells were filtered through an 80-µm Nitex filter (Sefar America, Depew, NY) and washed three times before being processed for immunofluorescence as described in the following.
Enriched fractions of germ cells were isolated from mice of varying ages as previously described [18, 19]. Spermatogonial cell populations were collected from 8-day-old mice, leptotene/zygotene spermatocyte and juvenile pachytene spermatocyte populations (approximately 80% pure with most contaminating cells being Sertoli cells) were collected from 17-day-old mice, while pachytene spermatocytes, round spermatids, and residual bodies were collected from mice at least 8 wk of age.
Cell Culture and Treatment
Enriched fractions of purified spermatocytes were washed and resuspended in HEPES-buffered MEM
culture medium (Sigma, St. Louis, MO) [20]. Cells were plated at a concentration of 2.5 x 106 cells/ml and cultured at 32°C in a 5% CO2-humidified atmosphere.
To investigate requirements for MAPK activation, spermatocytes were cultured for 2–12 h in the presence of the MEK inhibitor U0126 (Calbiochem, San Diego, CA), dissolved at 5 mM in dimethylsulfoxide (DMSO), and added to cultures at 10 or 40 µM. As a control, spermatocytes were cultured for 12 h in the presence of the inactive analog of U0126, U0124 (Calbiochem) dissolved at 10 mM in DMSO and added to cultures at 100 µM. Treatment with butyrolactone I was performed as described previously [4]; control cultures were given equivalent volumes of solvent alone. Spermatocytes were treated with either 0.5 or 5 µM OA (Calbiochem), dissolved at 244 µM in 100% ethanol, for 5 h. After treatment, cells were harvested and processed for immunoblot analysis, H1 kinase assays, in-gel kinase assays, or cytological analyses as described in the following sections.
Preparation of Cell Lysates
Purified germ cell fractions were lysed as previously described [3]; the supernatant was frozen and used in subsequent immunoblot and kinase assays. Total protein concentration was determined by the Bradford method (BioRad, Hercules, CA).
Immunoblot Analysis
Total proteins (20 µg) from each sample were separated on 10% SDS-PAGE gels and transferred to nitrocellulose using a BioRad semidry transfer apparatus (BioRad). The blots were blocked in either 5% nonfat milk or 2% BSA dissolved in Tris-buffered saline/0.1% Tween (TBST) for 1 h at room temperature. The blots were probed with anti-ERK1 (Santa Cruz Biotechnology, Santa Cruz, CA), anti-P-MAPK (Promega, Madison, WI), anti-MEK1 (UpState Biotechnology, Charlottesville, VA), anti-MEK2 (Cell Signaling, Beverly, MA), or anti-P-MEK1/2 (Cell Signaling) overnight at 4°C. After washing in TBST, the blots were incubated with horseradish peroxidase-conjugated secondary antibody (Pierce, Rockford, IL) for 1 h at room temperature, then incubated with the ECL chemiluminescent substrate (Amersham, Piscataway, NJ) and exposed to film according to manufacturer's instructions.
In-Gel Kinase Assay
Soluble cell lysates were collected and quantified as described previously for immunoblot analysis. Total protein, 10–15 µg, was diluted in 2x SDS-PAGE sample buffer and separated on 10% SDS-PAGE gels containing 0.1 mg/ml full-length myelin basic protein (MBP). After electrophoresis, gels were washed in 20% 2-propanol, 50 mM Tris-HCl, pH 8.0, for 1 h at room temperature, then equilibrated in 50 mM Tris-HCl, pH 8.0, 5 mM 2-mercaptoethanol. Proteins were denatured for 1 h in 6 M guanidine HCl, 50 mM Tris-HCl, pH 8.0, 5 mM 2-mercaptoethanol, and refolded overnight in 50 mM Tris-HCl, pH 8.0, 0.04% Tween-20, 5 mM 2-mercaptoethanol with gentle shaking at 4°C. Following renaturation, gels were preincubated in reaction buffer (50 mM HEPES, pH 8.0, 2 mM dithiothreitol, 10 mM MgCl2). Phosphorylation of MBP was carried out at 25°C for 1 h in reaction buffer supplemented with 0.5 mM EGTA, 2 µM cAMP-dependent protein kinase inhibitor, 40 µM ATP, and 100 µCi 
ATP. The reaction was stopped by immersing the gel in 5% trichloroacetic acid (TCA). The gel was then washed in 5% TCA containing 10 mM sodium pyrophosphate, dried, exposed to film, and quantitated by scanning densitometry using the UN-SCAN-IT software program (Silk Scientific Inc., Orem, UT).
H1 Kinase Assay
The assay for CDK1/CYCLIN-B1 (MPF) activity was performed as described [3]. Assays were performed with lysates made from frozen cells using exogenously added histone H1 as a substrate. The assay products were resolved on 10% SDS-PAGE gels. Gels were dried and exposed to film. Quantitation was done using the UN-SCAN-IT software program (Silk Scientific).
Cytological Methods
Spermatocytes isolated from germ cell preparations were embedded in fibrin clots as previously described [21]. Primary antibodies detecting SYCP3 [21], ERK1 (Santa Cruz Biotechnology), ERK2 (Santa Cruz Biotechnology), and phospho-MAPK (New England Biolabs) were diluted in PBS/10% goat serum and incubated overnight in a humidified chamber. Rhodamine and fluorescein isothiocyanate conjugated secondary antibodies (Pierce), diluted 1:500, were added and incubated for 2–3 h in the dark. Coverslips were mounted with ProLong-Antifade (Molecular Probes, Eugene, OR) containing 4',6-diamidino-2-phenylindole (DAPI) (0.1 µg/ ml) to visualize the DNA. Control slides were incubated with either preimmune sera or with secondary antibodies only in order to determine antibody specificity. Immunolocalization was observed with an Olympus epifluorescence microscope (Olympus, Tokyo, Japan), and images were captured to Adobe PhotoShop (Adobe Systems, San Jose, CA) using a Hamamatsu color 3CCD camera (Hamamatsu, Shizuoka, Japan).
Chromosome condensation was assessed by a modification of the Evans procedure [22] as previously described [2]. Air-dried cell preparations were stained with Giemsa (Sigma) and scored for chromatin condensation and the presence of MI chromosomes.
All experiments involving indirect immunofluorescence localization or Western blot analysis were repeated at least two times using different sets of samples. Experiments involving treatment with chemical inhibitors and analysis of kinase activity were repeated three times on independently treated samples.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Expression of ERK1 and/or ERK2 throughout spermatogenesis was investigated by Western blot analysis using an ERK1 antibody, which recognizes both the 44-kDa ERK1 protein and the 42-kDa ERK2 protein with similar specificity. As shown in Figure 1A, the 44-kDa ERK1 is expressed at virtually identical levels in all cells examined, from the mitotic spermatogonia to the haploid round spermatids. A similar level of expression was observed for ERK2, except in residual bodies, where the level decreased slightly. The antibody against ERK2, which demonstrates a higher specificity for ERK2 over ERK1, produced similar results (data not shown). Thus, these proteins are apparently not regulated at the level of expression during spermatogenesis. An antibody detecting the phospho-MAPKs was used to investigate the activation status of ERK1/2 throughout spermatogenesis (Fig. 1B). While phosphorylation (activation) of ERK1/2 is detected in early germ cell stages, it is not found in adult mid- to late-pachytene spermatocytes, a pattern similar to previous reports [16, 23].
|
The Western blot analyses were performed using extracts from highly enriched spermatogenic cell preparations [18, 19]. The populations of spermatocytes were obtained from prepubertal male mice (two fractions, enriched for leptotene/zygotene and juvenile spermatocytes respectively) or adult males (one fraction, enriched for mid- to late-pachytene spermatocytes). Each population undoubtedly contains some earlier or later spermatocytes; for example, a significant proportion of the juvenile spermatocytes from 17-day males express histone H1t protein, a marker of mid- to late pachynema [3]. Thus, to more precisely define cell specificity of MAPK expression, indirect immunofluorescence microscopy was used to monitor the localization of the kinases throughout meiotic prophase. Localization patterns in spermatocytes of ERK1 and ERK2 were similar, and both kinases were found throughout the chromatin of spermatocytes (Fig. 1C). Upon entry into the first meiotic division, there was a change from the very general and nonspecific pattern of prophase to a dramatic relocalization of ERK1 and ERK2 to the meiotic spindle at MI (Fig. 1C), confirmed by colocalization with tubulin (data not shown). Interestingly, ERK1 and ERK2 are not present in the chromosomes at MI (Fig. 1C), confirmed by DAPI staining (data not shown). In contrast to this localization pattern for ERK1 and ERK2 in pachytene spermatocytes, the phosphorylated (active) forms of the MAPKs were not observed until diplonema (Fig. 1C). At diplonema, the phosphorylated forms of the MAPKs were detected throughout the entire chromatin area, and as the cells entered the division phase, the intensity of the staining increased. These phosphorylated forms were not detected in the meiotic spindle at MI, although it is possible that the intensity and generality of antibody localization at MI masks localization to the meiotic spindle. Similar to the localization of ERK1 and ERK2, no colocalization of phosphorylated forms with DAPI was detected at MI, suggesting that neither the active nor the inactive forms of ERK1 and ERK2 associate directly with the chromosomes.
Expression of MEK1, MEK2, and the Phosphorylated Forms of MEK1/2 Throughout Spermatogenesis
Since ERK1 and ERK2 localization suggested they could play a role in regulating the onset of the meiotic division phase, the expression levels of their activating kinases, MEK1 and MEK2, was examined by Western blotting (Fig. 2). Analysis of enriched populations of germ cells revealed expression of the 45-kDa MEK1 only in mitotic spermatogonia and the early to midmeiotic prophase spermatocytes retrieved from juvenile males, with little if any expression in the fraction containing predominantly mid- to late-pachytene spermatocytes enriched from adult males (Fig. 2A). In contrast to the limited expression of MEK1 during spermatogenesis, MEK2 was expressed in mitotic spermatogonia, in all stages of meiotic prophase spermatocytes, and in postmeiotic cells (Fig. 2B).
|
The phosphorylated forms of MEK1/2 (Fig. 2C) exhibited an expression pattern identical to that observed for the phosphorylated forms of the MAPKs (Fig. 1B). MEK1/2 is phosphorylated primarily in mitotic spermatogonia and leptotene/zygotene spermatocytes. A lesser degree of phosphorylation was detected in juvenile pachytene spermatocytes, and in the fraction containing primarily mid- to late-pachytene spermatocytes from adult mice, the phosphorylated forms of MEK1/2 were no longer detected.
Taken together, these results demonstrate phosphorylation of both MEK1/2 and ERK1/2 at the same developmental period, suggesting that MEK1/2 could govern phosphorylation and activation of ERK1/2 in early stages of meiosis but that this is less likely in mid- to late stages of meiotic prophase.
Activation of the MAPKs in Response to OA Treatment
The expression and localization patterns of ERK1/2 and MEK2 suggested that they could play a role in the spermatocyte's G2/MI transition, a possibility also assessed by others [16, 17]. Study of this transition was facilitated by treatment of pachytene spermatocytes with the phosphatase inhibitor OA to induce a precocious and rapid MI chromosome condensation [2], with the caveat that OA may abrogate normal regulatory steps.
As shown in Figure 1B, pachytene spermatocytes have almost undetectable levels of phosphorylated ERK1/2, suggesting that the MAPKs do not play a regulatory role in vivo at this stage of meiosis. To determine if ERK1/2 are important for the OA-induced meiotic division phase, pachytene spermatocytes were cultured in the presence of 5 µM OA, and the phosphorylation status of ERK1/2 was monitored by Western blotting. As shown in Figure 3A, cells cultured with solvent alone showed no increase in phosphorylation of ERK1/2, while those cultured in the presence of OA showed an increase in phosphorylation in a time-dependent manner. This demonstrates that the phosphorylation of ERK1/2 is a direct result of OA treatment and not of cell culture and suggests that the ERK1/2 MAPKs could play a role in the OA-induced meiotic G2/ MI transition. To directly demonstrate enzymatic activity of ERK1/2, an in-gel kinase assay was performed with the MAPK substrate, MBP, copolymerized with polyacrylamide. Time-dependent OA-induced activation of the MAPKs was observed by an increase in intensity of the phosphorylated substrate (Fig. 3B, showing the results of two independent experiments). In contrast to the preferential activation of ERK1 observed previously by others [16], no enzymatic activation of ERK1 was detected by the in-gel kinase assay, although phosphorylation of ERK1 was detected on immunoblots after OA treatment (Fig. 3A).
|
Previous studies have shown that brief pretreatment with the MEK inhibitor U0126 blocks ERK1/2 activation [24]. When pachytene spermatocytes were pretreated with 10– 40 µM U0126 for 30 min prior to OA treatment, there was no inhibition of MAPK activation detected by the in-gel kinase assay (data not shown). To ensure that OA was not overriding inhibitory effects of U0126, the concentration of OA was reduced from 5 to 0.5 µM, but this strategy still did not result in U0126 inhibition of either MAPK activation or chromosome condensation in OA-treated spermatocytes (data not shown). When exposure to U0126 was increased to overnight, a treatment scheme used in previous studies [16, 17], there was no decrease in OA-induced MAPK activation detected using the in-gel kinase assay (Fig. 4A). When air-dried chromosome preparations were scored for the presence of MI chromosome configurations, 25.9% of spermatocytes treated with 10 µM U0126 followed by OA exhibited condensed MI-like chromosomes, and a similar frequency (26.3%) of spermatocytes pretreated with 40 µM U0126 also exhibited condensed chromosomes. However, these frequencies of spermatocytes with condensed chromosomes after pretreatment with U0126 are similar to those from spermatocytes treated with OA alone (28.6%). Thus, long exposure to U0126 did not inhibit either OA-induced chromosome condensation or MAPK activation, implying either that the target MEK is not accessible or that spermatocytes activate the ERK1/2 MAPKs by a MEK-independent pathway.
|
The failure of U0126 treatment to inhibit OA-induced chromosome condensation and MAPK activation by pachytene spermatocytes, coupled with low expression of MEK1 in pachytene spermatocytes, led to a test of the role of MEK1/2 in ERK1/2 activation during the earliest stages of spermatogenesis, when the MEKs are abundantly expressed and phosphorylated. Isolated leptotene/zygotene and juvenile pachytene spermatocytes were cultured with either 10 or 40 µM U0126, and phosphorylation/activation of ERK1 and ERK2 after treatment was assessed by Western blot analysis using a phospho-specific MAPK antibody. In leptotene/zygotene spermatocytes, U0126 treatment inhibited endogenous levels of MAPK activity (Fig. 4B). While some phosphorylation was still detected after treatment with 10 µM U0126, phosphorylation was completely inhibited after treatment with 40 µM U0126 for as little as 2 h. U0126 treatment of juvenile pachytene spermatocytes also resulted in inhibition of MAPK activation, even at the lower concentration of 10 µM (Fig. 4C). Treatment of cultured germ cells with the inactive analog of U0126, U0124, had no effect on MAPK activity. Taken together these results suggest that U0126 is an effective inhibitor of ERK1/2 activation in early meiotic prophase spermatocytes, that spermatocytes are permeable to U0126, and that MEK1/2 is a relevant activator of ERK1/2 kinases during early meiotic prophase but not later.
Treatment with the CDK1/CDK2 Inhibitor, Butyrolactone I, Prevents MAPK Activation
Recent evidence suggests that MPF activation can be a prerequisite for MAPK activation [25]. To explore the regulatory relationships between MPF and the MAPKs during spermatogenesis, pachytene spermatocytes were cultured in the presence of the CDK inhibitor, butyrolactone I, previously shown to inhibit OA-induced MPF activation and chromosome condensation [4]. Treatment with butyrolactone I caused a dose-dependent inhibition of chromatin condensation (data not shown) and a decrease in H1 kinase activity (Fig. 5A). Additionally, a butyrolactone I dose-dependent inhibition of MAPK activation was observed (Fig. 5B). The relative decrease in MAPK activity mirrored the decrease in MPF activity, suggesting that spermatocytes may possess a novel MPF-dependent regulatory pathway leading to the activation of the MAPKs.
|
MOS Is Not Required for MAPK Activation During Spermatogenesis
Normal spermatogenesis in Mos–/Mos– (knockout) mice suggests that MOS is not required for spermatogenesis. Mos–/Mos– MI spermatocytes phosphorylate MAPK at levels similar to their heterozygous littermate control spermatocytes (Fig. 6A). Isolated Mos–/Mos– pachytene spermatocytes cultured in the presence of 5 µM OA also condensed bivalent MI chromosomes (Fig. 6B) and activated ERK2 at levels similar to spermatocytes from the heterozygous littermate controls (Fig. 6C). Taken together, these observations demonstrate that MOS is not required for phosphorylation of ERK1/2 in spermatocytes in vivo or in vitro after OA treatment, or for the meiotic G2/MI transition in vivo or in vitro, or for chromosome condensation in spermatocytes.
|
CDC25C Is Not Required for OA-Induced Activation of MPF and MAPKs
Generation of fertile Cdc25c–/Cdc25c– male mice demonstrated that the CDC25C phosphatase is not required for the meiotic G2/MI transition in vivo [10]. To determine if CDC25C is required for OA-induced activation of MPF and MAPKs, Cdc25c–/Cdc25c– spermatocytes were treated with OA. In response to OA, they condensed bivalent MI chromosomes at a frequency equivalent to spermatocytes from the heterozygous littermate controls (data not shown) and activated both MPF (Fig. 7A) and the ERK1/2 MAPKs (Fig. 7B). Immunofluorescence confirmed the presence of activated MAPKs in vivo (data not shown). These observations demonstrate that spermatocytes have a CDC25C-independent pathway for activation of MPF and the ERK1/ 2 MAPKs.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MAPKs Are Expressed in Premeiotic and Meiotic Germ Cells and Are Activated at the Meiotic G2/MI Transition In Vivo and In Vitro
Immunolocalization and Western blot expression data demonstrated that the MAPKs, ERK1/2, are expressed in the testis in mitotically dividing spermatogonia and in meiotic spermatocytes. Even more important, the phosphorylated, or active forms, of the MAPKs were detected only in diplotene and metaphase I spermatocytes, placing these active MAPKs at the right place and time for contributing to the spermatocyte's G2/MI transition. Localization of the unphosphorylated ERKs was general and nonspecific in meiotic prophase spermatocytes, but at MI a dramatic relocalization to the meiotic spindle occurred. This is consistent with studies showing that the MAPKs are essential for spindle assembly and maintenance in both mitotic cells [26, 27] and meiotic division-phase oocytes [12, 28–30]. Our studies did not show localization of either the active or the inactive forms of the MAPKs to meiotic MI chromosomes, in contrast to a previously published report [17]. A reason for this apparent discrepancy may lie in the types of preparations used for localization; our method retains soluble proteins of the cell. Our results are also consistent with immunolocalization of inactive and active forms of the MAPKs in meiotically dividing oocytes [29, 31, 32].
In addition to their role in the naturally occurring G2/ MI transition, the MAPKs are activated during entry into the division phase induced by OA in mammalian oocytes, concurrent with resumption of meiosis and germinal vesicle breakdown in incompetent, prophase-arrested oocytes [33– 38]. Similarly, activation of the MAPKs occurs in pachytene spermatocytes treated with OA to induce a precocious G2/MI transition (Fig. 3), as reported previously [16, 17]. Although a previously published report suggested that OA-induced MAPK activation in spermatocytes preferentially involves ERK1 rather than ERK2 [16], our analysis suggests that it is ERK2 (the 42-kDa band detected by a phospho-specific MAPK antibody) that is preferentially activated (Fig. 3A). A previous study, using an approach similar to ours, also reported that the majority of MAPK activity observed in bovine oocytes is due to activation of ERK2 [39], but other studies have observed equivalent levels of ERK1 and ERK2 activity in oocytes [40].
Activation of the ERK1/2 MAPKs in Pachytene Spermatocytes Occurs Through a MEK-Independent Mechanism
Male mice lacking the MAPK kinase kinase, MOS, are not sterile, suggesting that MOS is not required for the meiotic divisions in vivo [13, 14]. This was not surprising since Mos transcripts had not been detected in spermatocytes but rather were found primarily in haploid round spermatids [41–43]. Since our analysis revealed spermatocytes deficient for Mos do activate the ERK1/2 MAPKs in spermatocytes (Fig. 6A), the pathway for MAPK activation in vivo must be independent of MOS. Additionally, since spermatocytes lacking MOS condense bivalent MI chromosomes and activate the MAPKs in response to OA treatment (Fig. 6, B and C), OA-induced activation of the MAPKs in vitro must also be independent of MOS.
Since the downstream target of MOS is the ERK-activating kinases MEK1/2, these observations called into question the role of MEK1/2 in spermatocytes. The expression of the MEKs and their sensitivity to the MEK1/2-specific inhibitor, U0126, was examined to test their role in MAPK activation. Analysis of protein expression revealed that MEK1 was expressed predominantly in mitotic spermatogonia and in leptotene/zygotene spermatocytes (Fig. 2A) but not in pachytene spermatocytes. In contrast to MEK1, MEK2 was expressed throughout spermatogenesis in all cell types examined (Fig. 2B). Therefore, if the MEKs activate the MAPKs in pachytene spermatocytes, it must be MEK2 that does so. However, recently Mek2-null mice were produced; these mice are fertile [44], suggesting either that MEK2 does not activate the ERK1/2 MAPKs in pachytene spermatocytes or that its function is replaced by another kinase. The expression profile of phosphorylated/activated MEK1/2 was similar to that of activated ERK1/2, that is, present in mitotic spermatogonia and early meiotic prophase spermatocytes but absent in pachytene spermatocytes (Fig. 2C). This suggests that MEK1/2 might regulate ERK1/2 activity in spermatogonia and early meiotic prophase but not later.
Analysis of the effect of the MEK1/2-specific inhibitor U0126 was consistent with observations on expression. Treatment of populations of leptotene/zygotene spermatocytes and juvenile pachytene spermatocytes with U0126 for only 2 h was sufficient to completely block endogenous levels of MAPK activity (Fig. 4, B and C), implying that MEKs activate the ERK1/2 MAPKs in these cells (and also demonstrating that spermatocytes are permeable to U0126). However, treatment of pachytene spermatocytes with U0126 did not inhibit OA-induced activation of the MAPKs (Fig. 4A) or chromosome condensation. This result is consistent with the expression analysis of MEK2 but initially was unexpected because it was in contrast to previously published studies [16, 17]. Several different experimental approaches were taken to resolve this discrepancy, including changes in incubation time, culture medium composition, and serum concentration. In all cases, there was no demonstrable U0126 inhibition of OA-induced MAPK activation or chromosome condensation.
Thus, expression analysis puts MEK2 at the right time and place for activating the MAPKs. However, the ability of spermatocytes lacking MOS to activate the ERK1/2 kinases, the absence of any apparent phenotype in spermatocytes lacking MEK2, and the lack of U0126 inhibition of ERK1/2 activation under physiological conditions all suggest that activation of the ERK1/2 MAPKs may occur through a MEK-independent mechanism in pachytene spermatocytes.
Interestingly, treatment of spermatocytes in vitro with butyrolactone I, which inhibits OA-induced bivalent chromosome condensation and MPF activation [4], also inhibited OA-induced activation of the ERK1/2 kinases (Fig. 5B). This observation provides further support for a MEK-independent mechanism of MAPK activation in pachytene spermatocytes but does not yet define a precise pathway. Butyrolactone I is a selective inhibitor of both CDK1 (the catalytic component of MPF) and CDK2 and has virtually no effect on MAPKs (or protein kinases A and C or casein kinase I or II), and thus our experiments suggest that butyrolactone inhibition of OA-induced MAPK activation is an indirect rather than a direct effect. This indirect effect may be mediated by MPF. However, since CDK2 activity was not assessed, it cannot be excluded that CDK2 could mediate OA-induced MAPK activation. This is an intriguing idea since CDK2 is known to be required for completion of meiosis in mouse spermatocytes in vivo [8, 9], although it is not known if CDK2-deficient spermatocytes can respond to OA with chromosome condensation and activation of MPF and the ERK1/2 kinases.
CDC25C Is Not Required for MPF or MAPK Activation in the OA-Induced Transition
At midpachynema, the time that spermatocytes acquire competence for OA-induced chromosome condensation, CDC25C levels increase [3]. Members of the Cdc25 family are dual specificity protein phosphatases that control the activation of the cyclin-dependent protein kinases (CDKs), including the CDK1 subunit of MPF [6, 7]. It had been suggested that OA treatment may inhibit the type 2A protein phosphatases (PP2A) that act on CDC25C, resulting in activation of CDC25C and MPF [3]. However, the generation of a Cdc25c-null mouse model with no apparent phenotype [10] called this idea into question and provided a means to test the requirement for CDC25C in OA-induced chromosome condensation and activation of MPF and the ERK1/2 MAPKs. In fact, Cdc25c–/Cdc25c– spermatocytes responded to OA treatment with condensation of bivalent MI chromosomes and activation of MPF and the MAPKs (Fig. 7), demonstrating that CDC25C is not required for OA-induced MAPK activation and/or that other molecules, perhaps either CDC25A or CDC25B, can compensate for its loss.
Evidence from oocytes homozygous for the targeted allele Cdc25btm1Pjd (designated Cdc25b–) suggests that the CDC25 story is complex and that there is sexual dimorphism in the germ-cell expression of and requirement for the different isoforms. Cdc25b–/Cdc25b– oocytes arrest at prophase and fail to undergo the G2/MI transition, possibly because of their low levels of MPF activity [45]. On the other hand, Cdc25b–/Cdc25b– male mice are fertile and have no apparent phenotype, revealing a difference in the regulatory molecules required by oocytes and spermatocytes for completion of the G2/MI transition. A possible explanation for this sexual dimorphism may be that spermatocytes use another CDC25 variant; indeed, expression analysis indicates that Cdc25b transcripts are predominantly in the somatic cells of the testis [46], while Cdc25a transcripts are expressed in diplotene spermatocytes [47]. Although CDC25A is in the right place and at the right time for a role in G2/MI activation of MPF and the ERK1/ 2 kinases, mutational analysis will be required to determine its role.
A Role for the ERK1/2 MAPKs in the Spermatocyte's G2/MI Transition?
Taken together, the observations of this study show that the ERK1/2 kinases are activated in spermatocytes at the G2/MI transition both in vivo and in vitro and that neither MOS, MEK1/2, nor CDC25C are required for their activation, suggesting that there is a novel pathway for activation of these kinases both in vivo and in vitro in response to OA. This work has focused on candidate upstream regulators of the ERK1/2 MAPKs using cellular and genetic approaches. While the results show activation of the MAPKs both in vivo and in vitro at the G2/MI transition, genetic analyses also reveal that the expected activators, MOS and MEK, do not play a role. First, the ERK1/2 kinases are activated in spermatocytes that lack MOS. Second, it was demonstrated that only MEK2 is present in spermatocytes, and it has been shown that MEK2 is not required for completion of meiosis in the testis [44]. Additionally, it was consistently found that the MEK inhibitor U0126 did not inhibit OA-induced activation of the ERK1/ 2 MAPKs in vitro. These observations were difficult to reconcile with a previous report documenting inhibition of MAPK activation by U0126 [17]. Differences in the two studies could lie in the relative purities of cell preparations or in the composition of the culture medium coupled with the concentration of DMSO used. We found that both MEK1 and MEK2 are present in spermatogonia and juvenile spermatocytes and that U0126 inhibited endogenous MAPK activation in these cells. But we also found that MEK1 is absent in pachytene spermatocytes and that U0126 was not an effective inhibitor of MAPK activation in enriched preparations of these cells. Instead, data from OA treatment of pachytene spermatocytes in vitro suggest a role for CDKs in activation of the ERK1/2 MAPKs. However it is not known if CDKs play a physiologically significant role or if their apparent involvement is an artifact of OA treatment, which may abrogate some of the normal regulatory steps of the G2/MI transition.
Although it is not clear exactly how the MAPKs are activated in spermatocytes in vivo and in vitro, their activation is consistent with a role, similar to the one they play in oocytes, of either initiating or maintaining chromatin condensation. In spermatocytes, MAPKs have been implicated as activating NEK2, a serine-threonine kinase ortholog of the Aspergillus NIMA kinase required for the G2/MI transition [17]. Additionally, NIMA in pachytene spermatocytes interacts with a high-mobility group chromosomal protein, HMGA2, which could play a role in initiating chromosome condensation [48].
These studies thus are a step in elucidating the cascade of events that leads to the first meiotic metaphase in spermatocytes. The challenges ahead include dissecting out the regulators of both MPF and MAPK activation, determining the physiological significance of activation of MAPKs and MPF and their roles in initiating and maintaining chromosome condensation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence and current address: Mary Ann Handel, The Jackson Laboratory, 600 Main St., Bar Harbor, ME 04609. FAX: 207 288 6073; mahandel{at}jax.org
3 Current address: Laboratory for Reproductive and Developmental Toxicology, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, NC 27709
Received: 28 January 2004.
First decision: 16 February 2004.
Accepted: 1 April 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S C Mizrak, F Renault-Mihara, M Parraga, J Bogerd, H J G van de Kant, P P Lopez-Casas, M Paz, J del Mazo, and D G de Rooij Phosphoprotein enriched in astrocytes-15 is expressed in mouse testis and protects spermatocytes from apoptosis Reproduction, April 1, 2007; 133(4): 743 - 751. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
