Xp42Mpk1 Activation Is Not Required for Germinal Vescicle Breakdown but for Raf Complete Phosphorylation in Insulin-stimulated Xenopus Oocytes*

Fully grown G2-arrested Xenopus oocytes resume meiosis in vitro upon exposure to hormonal stimulation. Progesterone triggers oocyte meiosis resumption through a Ras-independent pathway that involves a p39Mos-dependent activation of the mitogen-activated protein (MAP) kinases. Insulin also triggers meiosis resumption through a tyrosine kinase receptor that activates a Ras-dependent pathway leading to the MAP kinases activation. Antisense phosphorothioate oligonucleotides were used to prevent p39Mos accumulation and Erk-like Xp42Mpk1 activation during insulin-induced Xenopus oocytes maturation. In contrast to previous works, prevention of p39Mos-induced activation of Xp42Mpk1 in insulin-treated oocytes did not inhibit but delayed meiotic resumption, like in progesterone-stimulated oocytes. Activations of Xp42Mpk1, the unique Erk of the oocyte, and of its downstream target p90Rsk, were impaired and phosphorylation of the MAPKK kinase Raf was partially inhibited. Similarly, oocytes treated with the MEK inhibitor U0126, stimulated by insulin exhibited delayed germinal vesicle breakdown, absence of Xp42Mpk1 activation, and partial phosphorylation of Raf. To summarize, whereas p39Mos-induced activation of MEK/MAPK pathway is dispensable for insulin-induced germinal vesicle breakdown, Xp42Mpk1 activation induced by insulin is dependent upon p39Mos synthesis. Raf complete phosphorylation appears to require the MEK/MAPK pathway activation both in progesterone and insulin-stimulated oocytes.

Immature Xenopus oocytes are physiologically arrested at the G 2 stage of the first meiotic division. Meiosis resumes after stimulation by the steroid hormone progesterone and is marked by dissolution or breakdown of the germinal vesicle (GVBD), 1 resulting in the formation of a white spot at the animal pole. At the biochemical level, GVBD is initiated through the activation of maturation-M phase promoting factor (MPF), a protein complex made up of p34 Cdc2 kinase and cyclin B (1). Progesterone induces mRNA polyadenylation (2,3) and synthesis of the Ser/Thr kinase p39 Mos . Consequently, the mitogen-activated protein kinase (MAPK) pathway, including the MAPK/Erk kinases MEK 1 and 2 and the Erk Xp42 Mpk1 , is activated. Xp42 Mpk1 phosphorylates and activates ribosomal S6 kinase (p90 Rsk ) that in turn inactivates Myt1, a negative regulator of MPF (4).
p39 Mos and its downstream targets induce meiotic resumption in the absence of progesterone when microinjected into prophase-arrested Xenopus oocytes (5)(6)(7)(8). There are conflicting reports regarding the necessity of p39 Mos in progesterone-induced GVBD. Whereas injection of phosphodiester antisense oligonucleotides against p39 Mos mRNA has been shown to not only inhibit p39 Mos accumulation but also progesterone-induced GVBD (9,10), inhibition of p39 Mos synthesis by morpholino antisense oligonucleotides has been shown recently to impair neither resumption of meiosis nor activation of MPF into progesterone-stimulated Xenopus oocytes (11). Moreover, other reports show that MEK 1/2 activity is not required for MPF activation induced by progesterone (12,13).
Insulin and insulin growth factor can also trigger meiotic resumption (14,15), through a tyrosine kinase receptor (16), inducing the activation of the MEK/MAPK pathway via the GTP-binding protein Ras (17,18). Xp42 Mpk1 activation induced by Ras can be impaired by dominant-negative forms of Raf in Xenopus oocytes and in oocytes extracts (19,20), leading to the conclusion that activation of MEK in these cases occurs through Raf activity, like in somatic cells (21). However, p39 Mos synthesis and accumulation has also been reported after fibroblast growth factor 1 stimulation in Xenopus oocytes expressing fibroblast growth factor receptors (22) and in oocytes treated with insulin (23).
Ras activation is also known to stimulate Raf-independent pathways (for a review in somatic cells, see Ref. 24). In Xenopus oocytes, the phosphoinositide 3-kinase pathway (25)(26)(27), which leads to activation of protein kinase B/Akt, has been well described because recent data demonstrate a crucial role for protein kinase B/Akt in the insulin-but not progesterone-stimulated resumption of meiosis (28,29).
Injections in Xenopus oocytes of V12 H-Ras (30) and Raf (31) can trigger meiotic resumption without progesterone stimulation, independently of p39 Mos . Moreover, dominant-negative forms of Raf can prevent Mos injection-induced GVBD (31). Nevertheless, other results showed that dominant-negative forms of Raf do not block Mos-induced MAPK activation in oocyte extracts (32). In fact, Raf activation subsequently to Mos injection would be under the control of the MAPK pathway itself (32). Despite all of these results, the hypothesis that p39 Mos might be dispensable to GVBD and MAP kinase activation induced by insulin has not yet been tested in Xenopus oocytes.
Therefore, we assessed the role of p39 Mos and activation of the Erk-Rsk pathway in maturation induced by insulin stimulation of immature Xenopus oocytes using phosphorothioate antisense oligodeoxynucleotides (PS-AS) to prevent p39 Mos synthesis and accumulation. Here, we show that activation of Xp42 Mpk1 induced by p39 Mos is not essential for MPF activation and GVBD induced by insulin. These observations also highlight that the active MEK1/2-Xp42 Mpk1 pathway is necessary for full activation of Raf, independently of p39 Mos accumulation.

MATERIALS AND METHODS
Handling of Oocytes-Adult Xenopus females were purchased from the University of Rennes I, France. After anesthesia with 1 g/liter MS222 (tricaine methanesulfonate, Sandoz), ovarian lobes were surgically removed and placed in ND96 medium (96 mM NaCl, 2 mM KCl, 1.8 mM CaCl 2 , 1 mM MgCl 2 , 5 mM HEPES, pH 7.5 (NaOH)). Full-grown stage VI oocytes (33) were isolated and follicles were removed by collagenase treatment for 30 min (1 mg/ml collagenase A, Roche Applied Science), followed by manual microdissection. Oocytes were stored at 14°C in ND96 medium until experiments.
Experimental Conditions-Phosphorothioate deoxyoligonucleotides were purchased by Eurogentec. The sequence of the antisense against p39 Mos mRNA (PS-AS) was AAGGCATTGCTGTGTGACTCGCTGA-AAC. As a control we used the inverted sequence GTTTCAGCGAGTC- After overnight incubation, oocytes were stimulated by progesterone. Appearance of the white spot was monitored every 30 min. B, Western blot analysis. At the end of the maturation, 3 oocytes were taken off, homogenized, and immunoblotted with antibodies against p39 Mos , Xp42 Mpk1 , p90 Rsk , Raf, and cyclin B2. Arrows show phosphorylation states. In immature oocytes (prophase), p39 Mos was not detected, p42 Mpk1 , p90 Rsk , and Raf were found under their non-phosphorylated isoforms (down arrows) and cyclin B2 was found as a doublet of two isoforms. In control mature oocytes injected with water (H 2 O/Pg), p39 Mos was synthesized and Xp42 Mpk1 , p90 Rsk , Raf, and cyclin B2 were found under their completely phosphorylated isoforms (up arrows). PS-S injection (PS-S/Pg) had no effects compared with water-injected oocytes (H 2 O/Pg). PS-AS injection abolished p39 Mos synthesis, resulting in total inhibition of Xp42 Mpk1 and p90 Rsk phosphorylations and partial inhibition of Raf phosphorylation. It did not prevent cyclin B2 phosphorylation but we observed lower amounts of cyclin B2. Ability of a purified mu-Mos to overcome the effects of inhibition of p39 Mos synthesis was tested by injecting mu-Mos in PS-AS-treated oocytes just before hormonal stimulation (PS-AS/mu-Mos/Pg). Injection of mu-Mos restored normal phosphorylation of Raf, Xp42 Mpk1 , and p90 Rsk and normal amounts of cyclin B2 compared with control mature oocytes.
Purified murine Mos protein (mu-Mos) was kindly provided by Dr. Vande Woude and Dr. Ahn. 37.5 ng (50 nl) were injected into PS-ASinjected oocytes just before hormonal treatment. Control oocytes were injected with 50 nl of water. U0126 (Promega) was made soluble in Me 2 SO to obtain a stock solution at 50 mM and was used at a final concentration of 50 M.
Treatment began 1 h before insulin or progesterone addition. Control oocytes were treated with Me 2 SO, 1/1000.
Batches were constituted from 30 to 60 oocytes that were observed every half-hour and scored for the appearance of white spot. GVBD 50 (time at which 50% of the oocytes showed a white spot) were estimated. The value in hours of GVBD 50 can be variable from one female to another. Each experiment was performed on at least 3 females. 3 oocytes were taken off at each time point for kinetic biochemical analysis or at the end of the maturation, respecting the ratio of the white spot, and conserved at Ϫ20°C until homogenization (see below).
Electrophoresis and Western Blotting-Oocytes were taken off and homogenized in homogenization buffer (34) and then centrifuged for 5 min at 10,000 ϫ g (4°C) to eliminate yolk platelets. Proteins were then After overnight incubation, oocytes were stimulated with insulin. Appearance of the white spot was monitored every 30 min. B, Western blot analysis. At the end of the maturation, 3 oocytes were taken off, homogenized, and immunoblotted with antibodies against p39 Mos , Xp42 Mpk1 , p90 Rsk , Raf, and cyclin B2. Arrows show phosphorylation states. In immature oocytes (prophase), p39 Mos was not detected, p42 Mpk1 , p90 Rsk , and Raf were found under their non-phosphorylated isoforms (down arrows) and cyclin B2 was found as a doublet of two isoforms. In control mature oocytes injected with water (H 2 O/Ins), p39 Mos was synthesized and Xp42 Mpk1 , p90 Rsk , Raf, and cyclin B2 were found under their completely phosphorylated isoforms (up arrows). PS-S injection (PS-S/Ins) had no effect compared with water-injected oocytes (H 2 O/Ins). PS-AS injection abolished p39 Mos synthesis, resulting in total inhibition of Xp42 Mpk1 and p90 Rsk phosphorylations and partial inhibition of Raf phosphorylation. It did not prevent cyclin B2 phosphorylation but we observed lower amounts of cyclin B2. Ability of a purified mu-Mos to overcome the effects of inhibition of p39 Mos synthesis was tested by injecting mu-Mos in PS-AS-treated oocytes just before hormonal stimulation (PS-AS/mu-Mos/Ins). Injection of mu-Mos restored normal phosphorylation of Raf, Xp42 Mpk1 , and p90 Rsk and normal amounts of cyclin B2 compared with control mature oocytes. separated by 10% mini-SDS-PAGE for p39 Mos immunodetection, by 12.5% mini-SDS-PAGE for Raf and cyclin B2 immunodetection, or by 17.5% modified mini-SDS-PAGE (23) for Xp42 Mpk1 and p90 Rsk immunodetection. Such gels allowed good discrimination between active and inactive proteins (35). Primary antibodies were diluted at 1/1000 in Tris-buffered saline. Xp42 Mpk1 was detected using the monoclonal antibody D-2 (Santa Cruz Biotechnology). Cyclin B2 detection was performed using the rabbit polyclonal antibody JG103 (a gift of Dr. J. Gannon, ICRF, South Mimms, United Kingdom). p39 Mos was detected using the polyclonal antibody C238 (Santa Cruz Biotechnology) and p90 Rsk detection was performed using the polyclonal antibody C21 (Santa Cruz Biotechnology). Raf was detected using the polyclonal antibody C20 (Santa Cruz Biotechnology). Signals were revealed using the ECL chemiluminescence system (Amersham Biosciences).

RESULTS
p39 Mos Inhibition by Antisense Phosphorothioate Oligonucleotides during Progesterone-induced Maturation-We have first investigated the effects of PS-AS oligodeoxynucleotides against p39 Mos mRNA on progesterone-induced GVBD. Compared with classical phosphodiester oligodeoxynucleotides (PO), phosphorothioate oligodeoxynucleotides (PS) have a sulfur atom instead of an oxygen in the phosphate group, resulting in better nuclease resistance. This extends in vivo longevity of PS and allows better specificity. 10 ng of PS-AS were injected in each oocyte. Control oocytes were injected either with water or 10 ng of PS-S. Oocytes were then incubated overnight at 20°C in OR2 medium before stimulation by progesterone. In our hands, inhibition of p39 Mos synthesis by PS-AS injection did not block GVBD but strongly delayed it (Fig. 1A). To verify if MPF was activated despite of inhibition of p39 Mos synthesis, cyclin B2 phosphorylation was assessed, because the latter is relevant for MPF activity (36,37). In prophase-arrested oocytes, cyclin B2 is detected as a doublet of two isoforms, corresponding to the non-phosphorylated form of the protein whereas, upon hormonal stimulation, cyclin B2 is only detected under its sole phosphorylated form. Inhibition of translation of p39 Mos did not affect phosphorylation of cyclin B2. However, detected amounts of cyclin B2 were drastically lower in antisense-injected oocytes than in control injected either with water or with sense oligonucleotides (Fig. 1B).
We verified on Western blot the activation of the MAP kinase pathway by detecting p39 Mos synthesis, the phosphorylation of Xp42 Mpk1 , and the hyperphosphorylation of p90 Rsk . Phosphorylation of Xp42 Mpk1 is a marker for its activation, which is confirmed by the hyperphosphorylation of p90 Rsk . In PS-ASinjected oocytes, p39 Mos was minimally detectable and Xp42 Mpk1 and p90 Rsk remained unactivated as demonstrated by the absence of an electrophoretic shift, whereas water or sense injection had no or little effect (Fig. 1B). In parallel, shifting up of Raf was also partially inhibited following PS-AS treatment (Fig. 1B). Indeed, on Western blot, Raf was detected in immature oocytes under its non-phosphorylated form. In mature oocytes injected with water or with PS-S, Raf was detected under a shifted up phosphorylated form. But in PS-AS-treated oocytes, progesterone stimulation resulted in detection of Raf as an intermediary shift between non-phosphorylated and completely phosphorylated forms (Fig. 1B). To control for eventual nonspecific effects of PS-AS, purified mu-Mos protein was injected before progesterone stimulation into PS-AS-injected oocytes. We observed that mu-Mos was able to revert all the effects of PS-AS on progesterone treatment (Fig.  1B). Although PS-AS injection appeared to strongly prevent p39 Mos accumulation, a small amount of Mos may still accumulate in PS-AS-injected oocytes either stimulated by insulin or progesterone (Figs. 1 and 2). However, it failed to activate Xp42 Mpk1 and p90 Rsk because we were unable to detect activated isoforms of Xp42 Mpk1 and p90 Rsk .
Our results show that phosphorylation of Xp42 Mpk1 and of its downstream target p90 Rsk induced by progesterone are dependent upon p39 Mos synthesis. Moreover, progesterone-induced GVBD occurs even when activation of these pathways are almost completely suppressed. p39 Mos Inhibition by Antisense Phosphorothioate Oligonucleotides during Insulin-induced Maturation-To test the effects of the inhibition of p39 Mos accumulation on GVBD and Raf phosphorylation induced by insulin, oocytes were injected with 10 ng of PS-AS and incubated overnight before stimulation by insulin. Minimal accumulation of p39 Mos but no activation of Xp42 Mpk1 nor p90 Rsk , even transitory, was observed (Figs. 2B and 3). Injection of PS-S had little or no effect on the rate or extent of GVBD in comparison to control oocytes injected with deionized water. In contrast, injection of PS-AS significantly delayed GVBD (Fig. 2A). GVBD 50 of PS-S-injected oocytes was 1.12 Ϯ 0.02-fold of GVBD 50 of control oocytes, whereas GVBD 50 of PS-AS-injected oocytes happened almost two times later than in control water-injected oocytes (1.92 Ϯ 0.09 GVBD 50 ).
Inhibition of p39 Mos synthesis by PS-AS did not prevent phosphorylation of cyclin B2 (Fig. 2B) but delayed it (not shown). Nevertheless, amounts of cyclin B2 detected in matured oocytes injected with PS-AS were dramatically lower than in control and immature oocytes (Fig. 2B).
To control for eventual nonspecific effects of PS-AS, we injected purified mu-Mos protein just before insulin stimulation into oocytes that had been treated with PS-AS. We observed At the end of maturation, 3 oocytes were taken off, homogenized, and immunoblotted with antibodies against p39 Mos , Xp42 Mpk1 , p90 Rsk , Raf, and cyclin B2. Arrows show phosphorylation states. In immature oocytes (prophase), p39 Mos was not detected, Xp42 Mpk1 , p90 Rsk , and Raf were found under their non-phosphorylated isoforms (down arrows) and cyclin B2 was found as a doublet of two isoforms. In control mature oocytes incubated in Me 2 SO (DMSO/Ins), p39 Mos was synthesized and Xp42 Mpk1 , p90 Rsk , Raf, and cyclin B2 were found under their completely phosphorylated isoforms (up arrows). U0126 treatment resulted in total inhibition of Xp42 Mpk1 and p90 Rsk phosphorylations and partial inhibition of Raf phosphorylation. It did not prevent cyclin B2 phosphorylation but we observed lower amounts of cyclin B2. At last, U0126 did not prevent synthesis of p39 Mos . that mu-Mos was able to revert the effects of PS-AS on insulin treatment. GVBD delay was abolished (data not shown), electrophoretic shifts of Xp42 Mpk1 and p90 Rsk were observed demonstrating their phosphorylation and normal amounts of cyclin B2 were detected as in control oocytes (Fig. 2B).
Surprisingly, phosphorylation of Raf was partially inhibited by p39 Mos synthesis inhibition in insulin-stimulated oocytes injected with PS-AS (Fig. 2B). Complete phosphorylation was restored by injection of purified mu-Mos protein that also restored Xp42 Mpk1 and p90 Rsk activation (Figs. 1B and 2B).
To determine whether complete phosphorylation of Raf is dependent upon MAP kinase activity, oocytes were treated with MEK inhibitor U0126 and stimulated by insulin. Whereas U0126 prevented insulin-induced activation of both Xp42 Mpk1 and p90 Rsk (Fig. 4B), it did not prevent insulin-induced GVBD. Rather, GVBD was delayed (1.82 GVBD 50 of control oocytes; Fig. 4A). In accordance with the previous results, U0126 also reduced cyclin B2 amounts but did not prevent cyclin B2 phosphorylation (Fig. 4B).
Partial inhibition of Raf phosphorylation in PS-AS-injected oocytes appeared to result from the absence of Xp42 Mpk1 activity rather than the consequence of p39 Mos absence itself. Indeed, when U0126-treated oocytes were stimulated with insulin, it triggered p39 Mos synthesis even if Xp42 Mpk1 activation was prevented. In such conditions, Raf phosphorylation was partial, even in the presence of p39 Mos , suggesting that complete phosphorylation of Raf is dependent upon Xp42 Mpk1 activity or its downstream effectors (Fig. 4B).
These results suggest first that p39 Mos is required for insulin-induced activation of MAPK and that insulin can induce GVBD independently of MEK/MAPK activation. Second, activity of the MAP kinases pathway, following p39 Mos activation, is required for complete phosphorylation of the MAPKK kinase Raf.

DISCUSSION
Two contradictory approaches have been used before ours to prevent p39 Mos synthesis and accumulation during progesterone-induced maturation of Xenopus oocytes: first, PO-AS against p39 Mos mRNA were used, which completely inhibited GVBD (9,10). Second, morpholino strategy revealed that the p39 Mos -Xp42 Mpk1 pathway was not required for progesteroneinduced GVBD and that PO-AS inhibition of maturation resulted from a nonspecific effect (11). In our hands, injection of PS-AS against p39 Mos mRNA, followed by progesterone stimulation, greatly diminished p39 Mos accumulation and Xp42 Mpk1 activation as shown by the absence of phosphorylation of both Xp42 Mpk1 and p90 Rsk . However, PS-AS were unable to block GVBD, according to results obtained with morpholinos (Ref. 11 and data not shown) and suggesting that, like morpholino oligonucleotides, PS-AS do not have nonspecific effects. We cannot exclude the possibility that the minimal amount of p39 Mos may influence GVBD as has been shown to occur previously (38).
Results obtained with PO-AS strategy that were used to prevent p39 Mos accumulation in insulin-stimulated oocytes have to be reconsidered. Injection of such PO-AS resulted in complete inhibition of GVBD (9,10). We assessed the effects of the more specific PS-AS on insulin-induced maturation. In contrast to PO-AS (9, 10), PS-AS injection did not block insulininduced GVBD. However, insulin failed to activate Xp42 Mpk1 independently of p39 Mos . This observation was surprising because p39 Mos and Raf have been shown in cell-free extracts to independently act to stimulate the Xp42 Mpk1 pathway (32). Moreover, dominant-negative forms of Raf inhibit Xp42 Mpk1 activation induced by oncogenic Ras protein both in Xenopus oocytes and in oocyte extracts (19,32,39,40). All these observations led to the conclusion that, during Ras-induced meiotic resumption, Xp42 Mpk1 is under control of Raf. In opposite, following progesterone stimulation, which is independent of Ras, Xp42 Mpk1 is under the control of p39 Mos (11). In contrast to these views, our results indicate that p39 Mos is required for Xp42 Mpk1 activation following stimulation of the tyrosine kinase receptor by insulin (Fig. 5). Results obtained by inhibition of Raf with dominant-negative forms might be explained by the fact that dominant-negative forms were truncated from their catalytic domains but still contained the Ras-binding domain. Also, these dominant-negative forms may have an inhibitory effect directly on Ras (32).
Most strikingly, Xp42 Mpk1 activation by p39 Mos but not p39 Mos itself appeared to be necessary for complete phosphorylation of Raf because PS-AS partially prevented Raf phosphorylation. Total phosphorylation of Raf was rescued by injection of exogenous mu-Mos protein. Raf has been shown to be activated in response to Ras activity after growth factor stimulation in somatic cells (for a review, see Ref. 21) and it has been proposed that the human oncogenic pathway involved Raf/MEK1/Xp42 Mpk1 cascade (10,19,41,42). Insulin-induced activation of Ras might account for partial phosphorylation of Raf observed in PS-AS-injected oocytes. However, interaction between Ras and Raf have been demonstrated to be insufficient to stimulate Raf activity in vitro (43,44). Our results in Xenopus oocytes suggest that Raf activation is a multistep process involving activation of Xp42 Mpk1 activity (Fig. 5).
In PS-AS-treated oocytes, MPF activation as assessed by cyclin B2 phosphorylation, and GVBD were both delayed but not prevented in contrast to results obtained with PO-AS (9, 10). As well, U0126 treatment resulted in such a delay in GVBD and did not prevent MPF activation. p90 Rsk and its Both insulin and progesterone pathways require p39 Mos synthesis to promote phosphorylation of Xp42 Mpk1 and p90 Rsk . In both cases, partial Raf phosphorylation can occur independently of this MAP kinase pathway but achievement of Raf phosphorylation requires activity of the p39 Mos / Xp42 Mpk1 pathway. upstream activator Xp42 Mpk1 have been suggested to indirectly activate pre-MPF by inhibiting Myt1 activity (4), and their activities, if not necessary for GVBD, are required for timely maturation to occur. In addition, Xp42 Mpk1 and p90 Rsk have been demonstrated to be responsible for suppression of replication between meiosis I and II and in cyclin B2 stabilization at metaphase II (11,45). Thus, absence of activity of these proteins could explain the decrease of the cyclin B2 level in the cells.
As observed in U0126-treated oocytes, p39 Mos synthesis induced by insulin is mainly controlled, if not totally, by a Xp42 Mpk1 -independent mechanism. In contrast to reported involvement of MAPK in a feedback loop that controls p39 Mos synthesis (46), our results strengthen the observations that MPF-dependent mechanisms might positively regulate p39 Mos synthesis (47,48). Indeed, p39 Mos accumulation can be observed in U0126-treated oocytes stimulated to resume meiosis either with progesterone (13) or by egg cytoplasm injection. 2 MPF could exert its positive control on p39 Mos accumulation by phosphorylating and stabilizing the p39 Mos oncoprotein (49) or by phosphorylating directly or indirectly the cytoplasmic polyadenylation element-binding protein whose activation is necessary for p39 Mos mRNA polyadenylation. Indeed, p39 Mos mRNA polyadenylation has been shown to be a crucial step in p39 Mos synthesis (3). Then, insulin is able to induce p39 Mos accumulation in U0126-treated oocytes through activation of MPF independently of Xp42 Mpk1 .
Alternate pathways that can be induced by Ras independently of Raf (50) might explain MPF activation observed in the absence of Xp42 Mpk1 activity. The best candidate is the phosphoinositide 3-kinase/Akt pathway (27)(28)(29)51). Akt has been shown to be acting through phosphorylation of Myt1 (52).
Nevertheless, involvement of MAPK family members in insulin-induced G 2 /M transition other than Xp42 Mpk1 should not be discarded because p38 MAPK have been involved in proliferation in many cell types (53) and because c-Jun kinase is activated during oocytes maturation (54). Then, both might be able to indirectly activate pre-MPF stored in prophase-arrested oocytes by the above mentioned mechanisms.
To conclude, our results showed that Xenopus oocyte provides us the first known model of the tyrosine kinase receptor signaling pathway implying the proto-oncogene p39 Mos in the MAP kinases pathway activation. It gives us unique opportunities for analyzing linear pathways and interdependences of complex signal transduction of Ras-dependent and independent pathways.