Article

• The EMBO Journal (1997) 16, 6407 - 6413
• doi:10.1093/emboj/16.21.6407

MAPK inactivation is required for the G₂ to M-phase transition of the first mitotic cell cycle

Ariane Abrieu¹, Daniel Fisher¹, Marie-Noëlle Simon², Marcel Dorée¹ and André Picard²

1. Centre de Recherches de Biochimie Macromoléculaire, 1919 route de Mende, 340293 Montpellier cedex 5, France
2. Laboratoire Arago, 66650 Banyuls sur Mer, France

Correspondence to:

Marcel Dorée, E-mail: doree@crbm.cnrs-mop.fr

Received 24 June 1997; Revised 26 August 1997

• Keywords:

  • cyclin B–cdc2 kinase,
  • fertilization,
  • G₂  to M-phase transition,
  • MAP kinase inactivation,
  • mitosis

Introduction

In the animal kingdom, the cell cycle proceeds very rapidly during early embryogenesis, and consists only of alternating S- and M-phases, without intervening gap phases. In many species, both vertebrate and invertebrate, the first mitotic cell cycle after fertilization is exceptionally long compared with the following ones. The increased length of the first mitotic cell cycle is only partly accounted for by a reduced rate of DNA replication. In the fertilized Xenopus egg at 20°C, for example, DNA replication starts 29 min after fertilization, before pronuclei fusion, and it is completed within the next 23 min (Gerhart, 1980). Nonetheless, the zygote nucleus envelope does not break down earlier than 69 min after fertilization. The interval of time from completion of the first round of DNA replication to initiation of the second one is about 34 min, whilst the total length of the 11 forthcoming cell cycles is as short as 20 min. This is due in part to the presence of a 17 min G₂ phase in the first mitotic cell cycle only. As observed in oocytes and somatic cells, G₂ phase is associated with tyrosine phosphorylation of cdc2 and inactivation of cyclin B–cdc2 kinase. However, the mechanism responsible for tyrosine phosphorylation of cdc2 in the first mitotic cell cycle only remains elusive.

Mitogen activated protein kinases (MAPKs) or extracellular signal-regulated kinases (ERKs) are activated through a cascade of conserved kinases in response to a variety of extracellular signals including growth factors and hormones, and couple extracellular stimuli to transcriptional activation in somatic cells (reviewed by Karin and Hunter, 1995; Waskiewiez and Cooper, 1995). MAPKs are also activated during meiotic maturation and have been shown to play a role in mechanisms controlling MPF activity (reviewed by Kosako et al., 1994; Sagata, 1997). At variance with the first mitotic cell cycle that follows fertilization, the 11 forthcoming ones proceed in the absence of MAPK activity in Xenopus embryos, and MAPK is reactivated only when intervening gap phases reappear in the cell cycle (Ferrell et al., 1991). MAPK activity also drops during the first mitotic cell cycle in mammals, molluscs and echinoderms, and does not reappear during the forthcoming early ones which proceed more rapidly than the first one (Shibuya et al., 1992; Verlhac et al., 1994; Picard et al., 1996).

We recently reported that fertilized eggs of the starfish Marthasterias glacialis microinjected with a constitutively active Ste 11–N mutant, a MAPK kinase kinase that clamps MAPK at a high level of activity, replicate DNA, but fail to enter mitosis and thus arrest at G₂ (Picard et al., 1996). Mature oocytes of the same starfish species replicate DNA after second polar body emission, then arrest with high MAP kinase activity if not fertilized. Taken together, these observations suggested that MAPK, which is active during part of the first, slow mitotic cell cycle, inactive during the forthcoming rapid ones and again reactivated later on when intervening gap phases reappear, may act as a brake on cell cycle progression.

The aim of the present study was to examine this possibility. We find that MAPK inactivation is essential for embryos to enter, then to complete first mitosis. However, advancing the time of MAPK inactivation after fertilization does not speed up the schedule for entry into mitosis, thus MAPK is not responsible for the increased length of the first cell cycle as compared with the following ones in the early embryo. The possibility that unrestrained activity of MAPK might inhibit progression of somatic cells into mitosis, as it does in embryonic cells, is discussed.

MAPK inactivation is required for Xenopus eggs to enter first mitosis

MAPK has been shown to be inactivated before entry into first mitosis in fertilized or parthenogenetically-activated eggs, but whether this inactivation is actually required for progression into mitosis has not been investigated. We have previously reported that activation of MAPK through addition of recombinant c-mos in interphase egg extracts suppresses activation of the cyclin degradation pathway, but not the ability of recombinant cyclin B to form active complexes with endogenous cdc2 (Abrieu et al., 1996). However, in these experiments the use of recombinant cyclin B in excess could have displaced an equilibrium (Solomon et al., 1990) and prevented detection of a MAPK-dependent block to cyclin B–cdc2 kinase activation. Indeed we have found that microinjection of the same recombinant c-mos protein in progesterone-matured Xenopus oocytes suppressed the reappearance of cdc2 kinase activity after its drop following parthenogenetic activation (data not shown; Karsenti et al., 1987).

In contrast, in another in vitro experiment no exogenous cyclin was added, and translation from mRNA of endogenous cyclins was used to drive the cell cycle from interphase into mitosis. Part of the cycling extract was clamped with high MAPK activity by adding to the extract, soon after its preparation, a non-degradable c-mos or Ste 11–N fusion protein, both acting as potent MAPKKK (Figure 1A). Under these conditions, pronuclei assembled and replicated DNA, as observed in control cycling extracts devoid of MAPK activity (Figure 1B and C).

Figure 1.

Suppression of MAPK inactivation in Xenopus cell cycle extracts causes arrest of in vitro assembled sperm nuclei at G₂. (A) GST–Ste11–N (lanes 2 and 3) or mal E-mos (lanes 4 and 5) fusion proteins were added or not (lane 1) soon after recovering, after centrifugation at 4°C, a cell cycle extract prepared from parthenogenetically-activated eggs, which was further incubated at 20°C. Samples containing identical amounts of egg extract were taken after 20 (lanes 1,2 and 4) and 60 min (lanes 3 and 5) and analysed for MAPK phosphorylation of MBP by p42^MAPK, using the in gel-MBP kinase assay. (B) and (C) Permeabilized sperm heads were added (500/l) to a cell cycle extract prepared from Xenopus eggs containing (+Mos) or not (-, control) the recombinant proto-oncogene (similar results were obtained when GST–Ste11–N was used instead of mal E-mos to suppress MAPK inactivation). (B) Samples were taken at the indicated times (in min) after sperm head addition and examined by Hoechst staining for assembly of sperm pronuclei and chromosome condensation. (C) [-³²P]dCTP was added, either simultaneously with sperm heads (0), or 20, 50 or 80 min later. Samples were collected at 20, 50, 80 or 110 min after sperm head addition and incorporation of ³²P–dCTP in DNA was evaluated by gel electrophoresis and autoradiography in the indicated time windows.

View full figure (47 KB)

However, pronuclei neither condensed chromosomes nor underwent nuclear envelope breakdown (NEBD). Instead, the G₂-arrested pronuclei continued to enlarge after completion of DNA replication, to reach, after 2 h, a diameter at least 2-fold that of control pronuclei at the time of NEBD, i.e. 15 min after completion of DNA replication (Figure 1B). Similar results were obtained by others, using a Ras-leu 61 mutant to arrest Xenopus egg extracts at interphase (Chen and Pan, 1994; Pan et al., 1994).

We also investigated the time course of H1 histone kinase activity in anti-cyclin A and anti-cyclin B immunoprecipitates prepared from the same extracts. Cyclin A is exclusively associated with cdc2 in the early Xenopus embryo (Roy et al., 1991; Devault et al., 1992; Rempel et al., 1995), thus only cdc2 kinases activities were monitored in these experiments. As shown in Figure 2A, cyclin A–cdc2 kinase was activated with roughly the same kinetics in extracts containing c-mos and in control extracts. However, cyclin A–cdc2 kinase did not undergo degradation in extracts clamped with high MAPK activity, whilst it dropped abruptly after 50 min in control cycling extracts, due to cyclin A degradation at prometaphase. In contrast to cyclin A–cdc2 kinase, cyclin B1–cdc2 kinase was not normally activated in extracts containing high MAPK activity, although it was activated very late and to a limited extent in some extracts as compared with control cycling extracts (Figure 2B). Similar results were obtained when H1 kinase was measured in anti-cyclin B2–cdc2 kinase (Figure 2C), which activates later on and more abruptly in cycling extracts, as it does in the fertilized egg (Minshull et al., 1990).

Figure 2.

Time-course of cdc2 kinase activities in cell cycle extracts prevented (+Mos) or not (-Mos) from inactivating MAPK by addition of the non-degradable Mos fusion protein. (A) H1 kinase activities were monitored by autoradiography after SDS–PAGE of anti-cyclin A immunoprecipitates prepared from aliquots taken after 20 (lane 1), 40 (lane 2), 50 (lane 3) 60 (lane 4) or 70 min (lane 5) of incubation at 20°C. (B) The same experiment, but H1 kinase activities were measured in anti-cyclin B1 immunoprecipitates. After diluting aliquots for immunoprecipitation in RIPA buffer, a tracer amount of ³⁵S-labelled cyclin B1 was added in each aliquot to evaluate immunoprecipitation recovery. ³⁵S-labelled cyclin B1 was also added before immunoprecipitation with anti-cyclin A in panel A experiments or anti-cyclin B2 in panel C experiments to control specificity of immunoprecipitation. (C) Same experiment, but H1 kinase activities were measured in anti-cyclin B2 immunoprecipitates. (D) Same experiment as in (B) (control, Mos), but H1 kinase activities of anti-cyclin B1 immunoprecipitates were quantified by liquid scintillation counting after SDS–PAGE, and are presented as percentages of control activity after 1 h (4). In parallel to these experiments, a sample of the same cell cycle extract received the CL100 recombinant protein, added simultaneously with Mos, at the time of transfer at 20°C. (E) The same experiment as in (D), but samples of egg extracts, taken 30, 75 and 120 min after transferring the cell cycle extract at 20°C, were analysed for immunoreactivity to anti-phosphotyrosine antibodies after SDS–PAGE and Western blotting. We previously showed that the prominent 55 kDa component recognized by our polyclonal anti-PTyr antibody in extracts containing the recombinant Mos protein is in fact tubulin, phosphorylated on serine residues exclusively (Abrieu et al., 1996).

View full figure (69 KB)

To establish that suppression of cyclin B–cdc2 kinase was actually due to MAPK activity in the above experiments, not merely to the toxic effect of adding a protein of bacterial origin or to some unexpected property of c-mos, the same experiments were repeated, but the CL100 protein phosphatase (Alessi et al., 1993) was added simultaneously with c-mos to suppress MAPK activity. As shown in Figure 2D, CL100 completely suppressed the inhibitory effect of c-mos and restored cyclin B–cdc2 kinase activation.

Finally, we found that prevention of cyclin B–cdc2 kinase activation by clamping MAPK early in its active tyrosine-phosphorylated form was associated with accumulation in egg extracts of a tyrosine-phosphorylated 34 kDa protein (Figure 2E) identified as cdc2, as it could also be immunoprecipitated with specific anti-cdc2 antibodies (not shown). Accumulation of tyrosine-phosphorylated p34^cdc2 was not observed in extracts to which both c-mos and CL100 were added simultaneously, even though they escaped from CL100 inhibition when incubated for more than 1 h at room temperature. This shows that MAPK no longer prevents newly synthesized cyclin B from forming active complexes with cdc2 once cdc2 kinase activation has started. Figure 2E further shows that CL100 suppressed MAPK, but not c-mos activity in egg extracts, as the tyrosine phosphatase did not prevent c-mos from phosphorylating a 55 kDa protein previously identified as tubulin (see Abrieu et al., 1996).

In the above experiments, recombinant mal E-mos (or GST–Ste11–N) was added soon after centrifugation of homogenates at 4°C, so that MAPK targets were not expected to undergo dephosphorylation when the extracts were further incubated at 20°C. However, sperm nuclei did not systematically arrest at G₂, and were observed in some cases to rather arrest at mitosis with condensed chromosomes and high cdc2 kinase activity; this occurred when the recombinant MAPKKK was added 10 min or more after incubating extracts at 20°C, in agreement with previous reports (Abrieu et al., 1996; Jones and Smythe, 1996). In this case, MAPK activity first dropped, then reappeared in egg extracts (not shown). We conclude that suppression of MAPK inactivation arrests cell cycle extracts at G₂, but reactivation of MAPK after its drop may arrest them either at G₂ or at metaphase.

Early inactivation of MAP kinase does not induce premature entry into first mitosis in Xenopus egg extracts

The above experiments suggested that MAP kinase may in some conditions act as a brake at the G₂ to M-phase transition. To investigate if MAP kinase actually plays this role in normal conditions, it was necessary to find a way to inactivate MAP kinase in advance of its normal schedule. In preliminary experiments, we failed to completely inactivate MAP kinase in CSF extracts using either the Pyst1 (Groom et al., 1996) or the CL100 tyrosine phosphatases, unless MAP kinase was first partially depleted (about 50%, data not shown) from extracts using specific antibodies. Using this specific double procedure, we first confirmed that inactivation of p42^{MAP kinase} releases the cyclin degradation pathway from a CSF block in the absence of a Ca²⁺ transient (not shown). Minshull et al. previously reached the same conclusion using the CL100 tyrosine phosphatase, which could have inactivated other MAP kinases besides p42^{MAP kinase} (Minshull et al., 1994).

As even mock-immunodepletion abrogates spontaneous cell cycling in egg extracts (probably by depressing cyclin synthesis), the following procedure was designed to investigate the effect on the first mitotic cell cycle of prematurely inactivating MAP kinase (Figure 3). A CSF extract was divided into two parts. One was partially immunodepleted of MAPK content, the other mock-depleted. Thirty minutes later, both received 0.5 mM CaCl₂. In addition, the immunodepleted extract received recombinant Pyst1 protein. Fifteen minutes later, recombinant sea urchin cyclin B was added to both extracts, then samples were collected as a function of time and used for determination of H1 and MAP kinase activities.

Figure 3.

Early inactivation of MAPK does not speed up cyclin B–cdc2 kinase activation in Xenopus egg extracts. Recombinant sea urchin cyclin B was added, 15 min after addition of 0.5 mM CaCl₂ to a CSF extract, in which early MAPK inactivation was induced (X) or not (l) through combined immunodepletion and Pyst-1 treatment. Samples were collected at the indicated times and analyzed for H1 kinase activities in anti-sea urchin cyclin B immunoprecipitates (A) or tyrosine phosphorylation of MAPK by immunoblotting (B).

View full figure (48 KB)

No difference was detected in the timing of H1 kinase activation, as determined in immunoprecipitates using specific antibodies against sea urchin cyclin B (Figure 3A), even though tyrosine dephosphorylation and thus inactivation of MAPK occurred at least 20 min earlier in the Pyst1-treated extract (Figure 3B).

We conclude that early inactivation of MAP kinase does not speed up activation of cyclin B–cdc2 kinase or entry into mitosis when recombinant cyclin B is added to extracts first driven from metaphase to interphase by Ca²⁺ addition. This suggests that the time of MAP kinase inactivation may not be rate-limiting for the fertilized or parthenogenetically-activated Xenopus egg to progress into first mitosis. Unfortunately, we were unable to validate this conclusion in more physiological conditions, as the combined procedure we had to use to completely and reproducibly inactivate MAPK in advance of its normal schedule could not be applied to intact eggs or to cycling extracts.

Thus we turned to another, more favourable biological system. We have previously reported that the length of the first mitotic cell cycle is longer than the following ones in the starfish M.glacialis (Picard et al., 1988), and this holds true also for the starfish Astropecten aranciacus (not shown). We also found that MAPK activity drops at the time of second polar body emission in fertilized but not unfertilized eggs of A.aranciacus, and does not reappear later on (Picard et al., 1996), at least during several cell cycles, as observed in Xenopus. However, an advantage of starfish, compared with most vertebrates, is that oocytes can be fertilized at various times during meiotic maturation (Cayer et al., 1975; Schuetz, 1975; Peaucellier and Dorée, 1981). As we observed that the timing of MAPK inactivation somewhat depends on the time of fertilization, this provided us with the opportunity to examine in physiological conditions whether early inactivation of MAPK may reduce the length of the first mitotic cell cycle.

Early inactivation of MAPK does not reduce the length of the first mitotic cell cycle in starfish eggs

In the next experiment, oocytes of the starfish A.aranciacus were fertilized either 20 or 75 min after addition of 10^-5 M 1-methyladenine, the natural inducer of meiotic maturation in starfish (Kanatani et al., 1969). In both cases a single sperm penetrated each maturing oocyte, as the block to polyspermy is established shortly after hormone addition (Cayer et al., 1975; Schuetz, 1975). The timing of maturation did not depend on the time of fertilization, and in both cases germinal vesicle breakdown (GVBD), first polar body emission and second polar body emission respectively occurred at the same time. In contrast, the time of MAPK inactivation was different (Figure 4). Whilst MAPK underwent inactivation later than 110 min post-hormone addition (p.h.a.) when oocytes were fertilized at 20 min p.h.a., MAPK inactivation occurred at least 10 min earlier when they were fertilized at 75 min p.h.a. In spite of this, the time of entry into the first mitosis (time of NEBD) and that of the first mitotic cleavage were identical in both groups of eggs. We conclude that the time of MAPK inactivation is not rate-limiting for progression of the first mitotic cell cycle in the intact starfish egg, as observed in the Xenopus in vitro assay.

Figure 4.

The time of entry into first mitosis (NEBD, nuclear envelope breakdown) does not depend on the time of MAPK inactivation following fertilization. Prophase-blocked oocytes of the starfish Astropecten aranciacus were induced to mature with 0.1 M 1 MeAde (time 0). Half of them were fertilized at 20 min post-hormone addition (p.h.a.) and the other half at 75 min p.h.a. At 80 min, 150 eggs of each batch were selected for a good fertilization membrane and homogenous 250 m diameter. Upper autoradiogram: samples of 10 oocytes were taken at 85 min (lane 5), 95 min (lanes 1 and 6), 105 min (lanes 2 and 7), 115 min (lanes 3 and 8) and 125 min p.h.a. (lane 4), and processed for detection of in-gel MBP kinase activity. Lower table: each indicated cell cycle event was scored by examining samples every 5 min. The values are the interpolated times p.h.a. for 50% eggs having undergone each event in either batch.

View full figure (39 KB)

In the present study, we demonstrate that MAPK inactivation is essential for eggs, both vertebrate and invertebrate, to enter first mitosis. Down-regulation of MAPK is not required for Ca²⁺ to trigger degradation of mitotic cyclins at fertilization, and in fact it occurs much after eggs have exited meiotic metaphase (Lorca et al., 1991, 1993; Watanabe et al., 1991; Weber et al., 1991). In contrast, suppressing down-regulation of MAPK, for example by microinjecting a constitutively active MAPKKK, suppresses entry into first mitosis. Conversely, inactivation of MAPK by the specific MAPK phosphatase Pyst1 is sufficient to drive the unfertilized mature egg into first mitosis.

Eggs of the starfish A.aranciacus and M.glacialis, as well as those of the amphibian Xenopus, do not require MAPK inactivation to replicate DNA in the first cell cycle. If prevented from inactivating MAPK, they replicate DNA, then arrest at G₂ with cyclin B–cdc2 kinase maintained in an inactive form, at least in part through phosphorylation of cdc2 on inhibitory residues. Besides this major effect, MAPK may slow down the first round of DNA replication. As a matter of fact, we have previously shown that the first round of DNA replication is completed much more rapidly (15 min instead of 45 min) in fertilized eggs of M.glacialis, which inactivate MAPK, than in fully-mature, unfertilized eggs which keep a high level of MAPK activity (see Figure 1 in Picard et al., 1996). In the starfish Asterina pectinifera, MAPK has even been shown to completely block the first round of DNA replication in 70% of mature oocytes (Tachibana et al., 1997). In this species, 30% of the eggs injected with constitutively active MAPKK escape G₁ arrest; in this case they arrest at G₂ after the first round of DNA replication (Tachibana et al., 1997; K.Tachibana and T.Kishimoto, personal communication). It thus appears that the most general effect of maintaining MAPK activity, observed in all investigated species, vertebrate and invertebrate, is prevention of the G₂ to M-phase transition, in the first mitotic cell cycle at least. If eggs are first allowed to inactivate MAPK following fertilization, and MAPK is then inappropriately reactivated, the cell cycle may arrest at metaphase, at least in vertebrates (Sagata et al., 1989; Haccard et al., 1993; MacNicol et al., 1995). This is, however, not observed in starfish and other invertebrates (A.Picard, unpublished results), that arrest at G₂ of the following cell cycle.

The mechanism responsible for MAPK-dependent inhibition of the G₂ to mitosis transition remains unknown. We show, using cycling Xenopus egg extracts, that MAPK specifically prevents cyclin B–cdc2 kinase activation, and has no effect on activation of cyclin A–cdc2. Moreover, prevention of cyclin B–cdc2 kinase activation by MAPK is associated with tyrosine phosphorylation of cdc2. However, a mutant of cdc2 that cannot be inhibited by phosphorylation has also been reported to be susceptible to inactivation in Xenopus egg extracts, demonstrating that inhibitory mechanisms independent of threonine–14 and tyrosine–15 phosphorylation may exist (Kumagai and Dunphy, 1995; Lee and Kirschner, 1996). Work is in progress to determine the respective contribution of this inhibitory pathway and of kinases that mediate the inhibitory phosphorylations on cdc2 in the MAPK-dependent G₂ to mitosis block.

A main objective of this work was to determine whether MAPK, active in the first but not the following cell cycles in early development of vertebrate and invertebrate embryos, acts as a brake in the fertilized or parthenogenetically-activated egg and is responsible for slowing down its progression into first mitosis. To address this question, we had to find a way to specifically inactivate MAPK in advance of its normal schedule, and investigate whether this would result in shortening of the first mitotic cell cycle. Microinjection of the MAPK-specific phosphatase Pyst1 was not sufficient in our hands to inactivate MAPK prematurely in parthenogenetically-activated Xenopus eggs or in extracts derived from them. However, the Pyst1 phosphatase readily inactivated MAPK prematurely in extracts undergoing parthenogenetic activation in vitro, provided they were partially immunodepleted of MAPK before parthenogenetic treatment. Using this dual procedure, we compared the time course of cyclin B–cdc2 kinase activation and onset of mitotic events following addition of recombinant cyclin B in Xenopus egg extracts undergoing early (10 min) or late (40 min) inactivation of MAPK following parthenogenetic activation. No difference was detected between 'early' and 'late' extracts.

These results argued against, but did not disprove, the hypothesis that MAPK might function as a brake on cell cycle progression during the first mitotic cell cycle. Indeed the dual procedure used for early MAPK inactivation could not be applied to intact Xenopus eggs. Fortunately, we were able to make complete the demonstration in physiological conditions, using starfish eggs which, like Xenopus eggs, have MAPK activity associated with the first mitotic cell cycle only, which proceeds more slowly than the following ones. At variance with Xenopus eggs, starfish eggs can be readily fertilized at various times during meiotic maturation. We found that the time of MAPK inactivation depends on the time when eggs are fertilized. However, the length of the first mitotic cell cycle did not depend on the time of MAPK inactivation. More specifically, advancing the time of MAPK inactivation did not speed up the schedule for entry of fertilized eggs into first mitosis. By inference, sustained activity of MAPK during part of the first mitotic cell cycle is not responsible for late entry of fertilized eggs into first mitosis, the reason for which remains unknown. The possibility that translation of mitotic cyclins or other regulators of mitotic progression might be rate-limiting during the first mitotic cell is unlikely, as suppressing protein synthesis after a 'point of no return' in the first S-phase hardly delays entry of both vertebrate and invertebrate eggs into first mitosis (Wagenaar, 1983; Solomon et al., 1990; Genevière-Garrigues et al., 1995).

The fact that (early) suppression of MAPK activity has no effect on cell cycle progression, whereas its prolongation or inappropriate reactivation arrests cell cycle progression at, possibly, different stages of the cell cycle, including entry into S-phase (A.pectinifera), entry into mitosis (other starfish species and Xenopus) or exit from mitosis (Xenopus) is reminiscent of checkpoint mechanisms that operate to arrest or delay cell cycle progression if a defect that would compromise genetic stability is detected. It is already established that MAPK activity is required for the spindle assembly checkpoint that prevents cells whose spindles are defective or whose chromosomes are misaligned from initiating anaphase (reviewed by Minshull et al., 1994; Murray, 1995; Takenaka et al., 1997; Wang et al., 1997). MAPK may act in this checkpoint by preventing MPF from turning on the cyclin degradation pathway in vitro (Abrieu et al., 1996; Jones and Smythe, 1996; Takenaka et al., 1997). Permanent MAPK activation mimics activation of the spindle assembly checkpoint and indeed arrests cell cycling in vivo (Takenaka et al., 1997). The c-mos proto-oncogene, whose expression is limited to meiotic maturation of vertebrate oocytes, also arrests cell cycling at metaphase in unfertilized eggs through activation of the MAPK cascade (reviewed by Sagata, 1997; see also Furuno et al., 1997). It would not be surprising that other checkpoints mechanisms use MAPK to arrest cell cycle at G₁/S or G₂/M in response to as yet uncharacterized deleterious stimuli. Inability of embryos to properly reactivate MAPK in many species during early cleavage could explain why they lack checkpoint mechanisms (Hartwell and Weinert, 1989; Clute and Masui, 1992; Murray, 1994; Clute and Masui, 1997) before a developmental transition in early development, which corresponds to MAPK reactivation. In agreement with this view, it was shown in Caenorhabditis that transgenic animals, expressing under control of a heat shock promoter a constitutively active MAPKK mutant with Ser223 mutated to Glu and Ser227 mutated to Asp, arrest eggs early in development when the heat shock was applied during early embryogenesis (Wu et al., 1995).

Xenopus egg extracts

Cycling and CSF-arrested extracts were prepared exactly as described in Morin et al. (1994), according to minor modifications of procedures described by Murray and Kirschner (1989). Procedures for assembly of sperm nuclei, cytological observations of nuclei and assay of DNA replication were as previously reported (Morin et al., 1994).

Starfish oocytes

The starfish A.aranciacus was collected during its breeding season near the marine biological station of Banyuls. Fully grown oocytes (260 m in diameter) were used throughout this work. Procedure for fertilization has been described previously (Picard et al., 1988).

Recombinant proteins and mRNAs

The plasmids encoding mal E-mos, residues 370–717 of Ste11 in fusion with GST and the CL100 and Pyst1 MAPK phosphatases were generous gifts of Drs Hunt (London), Nishida (Kyoto) and Keyse (Dundee), respectively. Construction of the sea urchin GST–cyclin B has been described previously (Abrieu et al., 1996). Recombinant proteins and mRNAs were prepared according to standard procedures.

Immunological procedures

The polyclonal antibody directed against phosphotyrosine has been described previously (Abrieu et al., 1997). Polyclonal antibodies against Xenopus cyclin B1, cyclin B2 and cyclin A were raised by immunizing rabbits with the corresponding recombinant proteins. Immunoprecipitations were performed after dilution in RIPA buffer (Lorca et al., 1992) and immunoprecipitates, collected on Protein A-Sepharose, were washed with 50 mM Tris (pH 7.5) before H1 kinase activities were assayed. Immunoblots were analysed by ECL.

Kinase assays

H1 kinase activities were assayed according to Labbé et al. (1991). Incubations were terminated by addition of Laemmli buffer, then proteins were separated by SDS–PAGE and phosphorylated H1 histone estimated either directly by liquid scintillation counting, or by autoradiography after Western blotting. In gel MBP-kinase activities were assayed exactly as described by Shibuya et al. (1992).

Abrieu A, Lorca T, Labbé JC, Morin N, Keyse S and Dorée M (1996) MAP kinase does not inactivate, but rather prevents the cyclin degradation pathway from being turned on in Xenopus egg extracts. J Cell Sci, 109, 239–246. | PubMed | ISI | ChemPort |

Abrieu A, Dorée M and Picard A (1997) Mitogen-activated protein-kinase activation down-regulates a mechanism that inactivates cyclin B–cdc2 kinase in G₂-arrested oocytes. Mol Biol Cell, 8, 249–261. | PubMed | ISI | ChemPort |

Alessi DR, Smythe C and Keyse SM (1993) The human CL100 gene encodes a Tyr/Thr protein phosphatase which potently and specifically inactivates MAP kinase and suppresses its activation by oncogenic ras in Xenopus oocyte extracts. Oncogene, 8, 2015–2020. | PubMed | ISI | ChemPort |

Cayer ML, Kishimoto T and Kanatani H (1975) Formation of the fertilization membrane by insemination of immature starfish oocytes pretreated with calcium-free sea water. Dev Growth Differ, 17, 119–125.

Chen CT and Pan BT (1994) Oncogenic ras stimulates a 96 kDa histone H2b kinase activity in activated Xenopus egg extracts. Correlation with the suppression of p34^cdc2 kinase. J Biol Chem, 269, 28034–28043. | PubMed | ChemPort |

Clute P and Masui Y (1992) Development of microtubule-dependence of the chromosome cycle at the midblastula transition in Xenopus laevis embryos. Dev Growth Differ, 34, 27–36.

Clute P and Masui Y (1997) Microtubule dependence of chromosome cycles in Xenopus laevis blastomeres under the influence of a DNA synthesis inhibitor, aphidicolin. Dev Biol, 185, 1–13. | Article | PubMed | ISI | ChemPort |

Devault A et al. (1992) Cyclin A potentiates maturation-promoting factor activation in the early Xenopus embryo via inhibition of the tyrosine kinase that phosphorylates cdc2. J Cell Biol, 118, 1105–1120.

Ferrell JE,Jr, Wu M, Gerhart JC and Martin GS (1991) Cell cycle tyrosine phosphorylation of p34^cdc2 and a microtubule-associated protein kinase homolog in Xenopus oocytes and eggs. Mol Cell Biol, 11, 1965–1971. | PubMed | ISI | ChemPort |

Furuno N, Ogawa Y, Iwashita J, Nakajo N and Sagata N (1997) Meiotic cell cycle in Xenopus oocytes is independent of cdk2 kinase. EMBO J, 16, 3860–3865. | Article | PubMed | ISI | ChemPort |

Genevière-Garrigues AM, Barakat A, Dorée M, Moreau JL and Picard A (1995) Active cyclin B–cdc2 kinase does not inhibit DNA replication and cannot drive prematurely fertilized sea urchin eggs into mitosis. J Cell Sci, 108, 2693–2703. | PubMed |

Gerhart JC (1980) Mechanisms regulating pattern formation in the amphibian egg and early embryo. In Goldbeter,R.F. (ed.), Biological Regulation and Development. Plenum Press, New York, vol. 2, 133–316.

Groom LA, Sneddon AA, Alessi DR, Dowd S and Keyse SM (1996) Differential regulation of the MAP, SAP and RK/p38 kinases by Pyst-1, a novel cytosolic dual-specificity phosphatase. EMBO J, 15, 3621–3632. | PubMed | ISI | ChemPort |

Haccard O, Sarcevic B, Lewellyn A, Hartley L, Roy T, Izyumi E, Erikson E and Maller JL (1993) Induction of metaphase arrest in cleaving Xenopus embryos by MAP kinase. Science, 262, 1262–1265. | Article | PubMed | ISI | ChemPort |

Hartwell L and Weinert T (1989) Checkpoints: controls that ensure the order of cell cycle events. Science, 6, 872–876.

Jones C and Smythe C (1996) Activation of the Xenopus cyclin degradation machinery by full-length cyclin A. J Cell Sci, 109, 1071–1079. | PubMed | ISI | ChemPort |

Kanatani H, Shirai H, Nakanishi K and Kurokava T (1969) Isolation and identification of meiosis-inducing substance in starfish Asterias anurensis. Nature, 211, 273–277.

Karin M and Hunter T (1995) Transcriptional control by protein phosphorylation: signal transmission from the cell surface to the nucleus. Curr Biol, 5, 747–757. | Article | PubMed | ISI | ChemPort |

Karsenti E, Bravo R and Kirschner M (1987) Phosphorylation changes associated with the early cell cycle in Xenopus eggs. Dev Biol, 119, 442–453. | PubMed | ChemPort |

Kosako H, Gotoh Y and Nishida E (1994) Regulation and function of the MAP kinase cascade is Xenopus oocytes. J Cell Sci Suppl, 18, 115–119. | PubMed | ChemPort |

Kumagai A and Dunphy W (1995) Control of the cdc2/cyclin B complex in Xenopus egg extracts arrested at G₂/M checkpoint with DNA synthesis inhibitors. Mol Biol Cell, 6, 199–213. | PubMed | ISI | ChemPort |

Labbé JC, Cavadore JC and Dorée M (1991) M-phase specific cdc2 kinase: preparation from starfish and properties. Methods Enzymol, 201, 291–301.

Lee T and Kirschner MW (1996) An inhibitor of p34cdc2/cyclin B that regulates the G₂/M transition in Xenopus extracts. Proc Natl Acad Sci USA, 93, 352–356. | Article | PubMed | ChemPort |

Lorca T, Galas S, Fesquet D, Devault A, Cavadore JC and Dorée M (1991) Degradation of the proto-oncogene product p39 mos is not necessary for cyclin proteolysis and exit from meiotic metaphase: requirement for a Ca²⁺-calmodulin-dependent event. EMBO J, 10, 2087–2093. | PubMed | ISI | ChemPort |

Lorca T et al. (1992) Cyclin A–cdc2 kinase does not trigger but delays cyclin degradation in interphase extracts of amphibian eggs. J Cell Sci, 102, 56–62.

Lorca T, Cruzalegui FH, Fesquet D, Cavadore JC, Méry J, Means A and Dorée M (1993) Calmodulin-dependent protein kinase II mediates inactivation of MPF and CSF upon fertilization of Xenopus eggs. Nature, 366, 270–273. | Article | PubMed | ISI | ChemPort |

MacNicol AM, Muslin AJ, Howard EL, Kikuchi A, MacNicol MC and Williams LT (1995) Regulation of Raf–1-dependent signaling during early Xenopus development. Mol Cell Biol, 15, 6686–6693. | PubMed | ChemPort |

Minshull J, Golsteyn R, Hill CS and Hunt T (1990) The A-and B-type cyclin associated cdc2 kinases in Xenopus turn on and off at different times in the cell cycle. EMBO J, 9, 2865–2875. | PubMed | ISI | ChemPort |

Minshull JH, Sun H, Tonks NK and Murray AW (1994) A MAP kinase-dependent spindle assembly checkpoint in Xenopus egg extracts. Cell, 79, 475–486. | Article | PubMed | ISI | ChemPort |

Morin N, Abrieu A, Lorca T and Dorée M (1994) The proteolysis-dependent metaphase to anaphase transition: calcium/calmodulin-dependent protein kinase II mediates onset of anaphase in extracts prepared from unfertilized Xenopus eggs. EMBO J, 13, 4343–4352. | PubMed | ISI | ChemPort |

Murray AW (1994) Cell cycle checkpoints. Curr Opin Cell Biol, 6, 872–876. | Article | PubMed | ISI | ChemPort |

Murray AW (1995) The genetics of cell cycle checkpoints. Curr Opin Genet Develop, 5, 5–11. | ChemPort |

Murray AW and Kirschner MW (1989) Cyclin synthesis drives the early embryonic cell cycle. Nature, 339, 275–280. | Article | PubMed | ISI | ChemPort |

Pan BP, Chen CT and Lin SM (1994) Oncogenic Ras blocks cell cycle progression and inhibits p34^cdc2 kinase in activated Xenopus egg extracts. J Biol Chem, 269, 5968–5975. | PubMed | ISI | ChemPort |

Peaucellier G and Dorée M (1981) Acid release at activation and fertilization of starfish oocytes. Dev Growth Differ, 23, 287–296. | ChemPort |

Picard A, Harricane MC, Labbé JC and Dorée M (1988) Germinal vesicle components are not required for the cell cycle oscillator of the early starfish embryo. Dev Biol, 128, 121–128. | PubMed | ChemPort |

Picard A, Galas S and Dorée M (1996) Newly assembled cyclin B–cdc2 kinase is required to suppress DNA replication between meiosis I and meiosis II in starfish oocytes. EMBO J, 15, 3590–3598. | PubMed | ISI | ChemPort |

Rempel RE, Sleight SB and Maller JL (1995) Maternal Xenopus cdk2–cyclin E complexes function during meiotic and early embryonic cell cycles that lack a G₁ phase. J Biol Chem, 270, 6843–6855. | Article | PubMed | ISI | ChemPort |

Roy LM, Swenson KI, Walker DH and Maller JL (1991) Activation of p34cdc2 by cyclin A. J Cell Biol, 113, 507–514. | Article | PubMed | ISI | ChemPort |

Sagata N (1997) What does Mos do in oocytes and somatic cells? Bio Essays, 19, 13–21. | ChemPort |

Sagata N, Watanabe N, Van de Woude WG and Ikawa Y (1989) The c-mos proto-oncogene product is a cytostatic factor responsible for meiotic arrest in vertebrate eggs. Nature, 342, 512–518. | Article | PubMed | ISI | ChemPort |

Schuetz AW (1975) Cytoplasmic activation of starfish oocytes by sperm and divalent ionophore A–23187. J Cell Biol, 66, 86–94. | Article | PubMed | ChemPort |

Shibuya EK, Boulton TG, Cobb MH and Ruderman JV (1992) Activation of p42 MAP kinase and the release of oocytes from cell cycle arrest. EMBO J, 11, 3963–3975. | PubMed | ISI | ChemPort |

Solomon MJ, Glotzer M, Lee TH, Philippe M and Kirschner MW (1990) Cyclin activation of p34^cdc2. Cell, 63, 1013–1024. | Article | PubMed | ISI | ChemPort |

Tachibana K, Machida T, Nomura Y and Kishimoto T (1997) MAP kinase links the fertilization signal transduction pathway to the G₁/S phase transition in starfish eggs. EMBO J, 16, 4333–4339. | Article | PubMed | ISI | ChemPort |

Takenaka K, Gotoh Y and Nishida E (1997) MAP kinase is required for the spindle assembly checkpoint but is dispensable for the normal M-phase entry and exit in Xenopus egg cell cycle extracts. J Cell Biol, 136, 1091–1098. | Article | PubMed | ISI | ChemPort |

Verlhac MH, Kubiak JZ, Clarke HJ and Maro B (1994) Microtubule and chromatin behavior follow MAP kinase activity but not MPF activity during meiosis in mouse oocytes. Development, 120, 1017–1025. | PubMed | ISI | ChemPort |

Wagenaar EB (1983) The timing of synthesis of proteins required for mitosis in the cell cycle of the sea urchin embryo. Exp Cell Res, 144, 393–403. | PubMed | ISI | ChemPort |

Wang XM, Zhai Y and Ferrell JEJr (1997) A role for mitogen-activated protein kinase in the spindle assembly checkpoint in XTC cells. J Cell Biol, 137, 433–443. | Article | PubMed | ISI | ChemPort |

Waskiewicz AJ and Cooper JA (1995) Mitogen and stress response pathways: MAP kinase cascades and phosphatase regulation in mammals and yeast. Curr Opin Cell Biol, 7, 798–805. | Article | PubMed | ISI | ChemPort |

Watanabe N, Hunt T, Ikawa Y and Sagata N (1991) Independent inactivation of MPF and cytostatic factor (Mos) upon fertilization of Xenopus eggs. Nature, 352, 247–248. | Article | PubMed | ISI | ChemPort |

Weber M, Kubiak JZ, Arlinghaus RB, Pines J and Maro B (1991) C-mos proto-oncogene product is partly degraded after release from meiotic arrest and persists during interphase in mouse zygotes. Dev Biol, 148, 393–397. | PubMed | ChemPort |

Wu Y, Han M and Guan KL (1995) MEK–2, a Caenorhabditis elegans MAP kinase kinase, functions in Ras-mediated vulval induction and other developmental events. Genes Dev, 9, 742–755. | PubMed | ChemPort |

Main navigation

• Search EMBO Journal

  Search

• Advanced search

• The EMBO Journal

  • Home
  • Aims and scope
  • Current issue
  • Advance online publication
  • Web focuses
  • Archive:
  • Browse by issue
  • Browse by subject
  • Browse by category
  • Free online sample issue
  • Press releases

• Authors and Referees

  • Editorial process
  • Guide for authors
  • Submit an article
  • Guide for referees
  • Editors and Editorial Board
  • Contact editorial office

• Customer Service

  • Subscribe
  • Order sample copy
  • Purchase articles
  • Reprints and permissions
  • Contact NPG
  • Advertising