Mos is required for MAP kinase activation and is involved in microtubule organization during meiotic maturation in the mouse

Transcription

1 Development 122, (1996) Printed in Great ritain The Company of iologists Limited 1996 DEV Mos is required for MP kinase activation and is involved in microtubule organization during meiotic maturation in the mouse Marie-Hélène Verlhac 1, *, Jacek Z. Kubiak 1, Michèle Weber 1, Gérard Géraud 1, William H. Colledge 2, Martin J. Evans 2 and ernard Maro 1, 1 Département de iologie du Développement, Institut Jacques Monod, CNRS Université Paris 7, 2 place Jussieu, F Paris, France 2 Wellcome/CRC Institute of Cancer and Developmental iology, University of Cambridge, Tennis Court Road, Cambridge C2 1QR, UK *Present address: Department of Growth and Development, University of California, San Francisco, C , US uthor for correspondence SUMMRY Mos is normally expressed during oocyte meiotic maturation in vertebrates. However, apart from its cytostatic factor (CSF) activity, its precise role during mouse meiosis is still unknown. First, we analyzed its role as a MP kinase kinase kinase. Mos is synthesized concomitantly with the activation of MP kinase in mouse oocytes. Moreover, MP kinase is not activated during meiosis in oocytes from mos / mice. This result implies that Mos is necessary for MP kinase activation in mouse oocytes. Raf-1, another MP kinase kinase kinase, is already present in immature oocytes, but does not seem to be active when MP kinase is activated. Moreover, the absence of MP kinase activation in mos / oocytes demonstrates that Raf-1 cannot compensate for the lack of Mos. These results suggest that Raf- 1 is not involved in MP kinase activation. Second, we analyzed the organization of the microtubules and chromosomes in oocytes from mos / mice. We observed that during the transition between two meiotic M-phases, the microtubules and chromosomes evolve towards an interphase-like state in mos / oocytes, while in the control mos +/ oocytes they remain in an M-phase configuration, as in the wild type. Moreover, after spontaneous activation, the majority of mos / oocytes are arrested for at least 10 hours in a third meiotic M-phase where they exhibit monopolar half-spindles. These observations present the first evidence, in intact oocytes, of a role for the Mos/.../MP kinase cascade in the control of microtubule and chromatin organization during meiosis. Key words: Mos, MP kinase, microtubules, MTOCs, chromatin, mouse, meiosis INTRODUCTION The resumption of oocyte meiotic maturation is linked to the activation of a universal factor, the maturation promoting factor (MPF; Masui and Markert, 1971), a complex of cdc2 and Cyclin (Lohka et al., 1988; Draetta et al., 1989; Labbé et al., 1989; Gautier et al., 1990). In Xenopus oocytes, the translation of mrn encoding the Mos kinase is required for the activation of MPF. In addition, Mos indirectly triggers p42mapk (Xenopus MP kinase) activation through MP kinase kinase phosphorylation (Nebreda and Hunt, 1993; Posada et al., 1993; Shibuya and Ruderman, 1993). oth Mos and p42mapk are involved in germinal vesicle breakdown (GVD; Sagata et al., 1989a; Carnero et al., 1994; Kosako et al., 1994b; Haccard et al., 1995) and in the metaphase II arrest (Sagata et al., 1989b; Haccard et al., 1993; Kosako et al., 1994a). However, the sequence of events differs during meiotic resumption in mouse oocytes where MPF activation occurs about two hours before the activation of MP kinases, p42erk2 and p44erk1 (Verlhac et al., 1993, 1994). Moreover, the analysis of mutant mos / mice has shown that, in this species, Mos is not required for GVD although it is necessary for the arrest in metaphase II (Colledge et al., 1994; Hashimoto et al., 1994). In Xenopus oocytes at least three pathways can lead to p42mapk activation. One pathway involves Mos, and is activated mainly after progesterone stimulation (Nebreda and Hunt, 1993; Posada et al., 1993; Shibuya and Ruderman, 1993). The two other pathways depend upon receptor tyrosine kinases and act through the Ras GDP-exchange factor. However, whilst one pathway is Raf-1-dependent (Fabian et al., 1993; Muslin et al., 1993), the other one is Raf-1-independent (Itoh et al., 1993; Kuroda et al., 1995). ll three pathways induce MP kinase activation via MP kinase kinase (MPKK or MEK) in vitro and in vivo (Dent et al., 1992; Howe et al., 1992; Kyriakis et al., 1992; Muslin et al., 1993; Posada et al., 1993). The proteins involved in the Raf-1- dependent pathway are conserved from yeast to vertebrates (for review see vruch et al., 1994). In contrast, Mos has been shown to be involved in MP kinase activation only in oocytes

2 816 M.-H. Verlhac and others from Xenopus. In addition, the relationships between the Raf- 1 and Mos pathways in the triggering of MP kinase activation in Xenopus oocytes are not clear and the data available are contradictory. Some results suggest that Raf-1 activation lies downstream of Mos (Muslin et al., 1993), while other results suggest that Raf-1 lies on the tyrosine receptor pathway at the same level as Mos (Fabian et al., 1993). In mouse oocytes little is known concerning MP kinase activation. This activation is protein synthesis dependent (Verlhac et al., 1993), but no data are available concerning the roles of the Mos and Raf-1 kinases. In addition to the roles of Mos and MP kinase in the initiation of meiotic maturation and the metaphase II arrest, other roles have been suggested for these kinases during meiosis. It has been proposed that Mos is required for MPF reactivation in metaphase II (O Keefe et al., 1991; Furuno et al., 1994; Gebauer et al., 1994) and that MP kinase could be involved in microtubule and chromatin organization during the transition between the two meiotic M-phases (Verlhac et al., 1994). role for Mos and MP kinase in microtubule organization was also suggested from studies performed in tissue culture cells where Mos was found to be partly located in the mitotic spindle (Zhou et al., 1991) and in Xenopus extracts where the addition of MP kinase to an interphase extract induced the shortening of the microtubules growing around exogenous centrosomes (Gotoh et al., 1991). We show here that the Mos protein accumulates concomitantly with MP kinase activation, suggesting that it might be involved in this event. This is demonstrated directly using oocytes from c-mos knock-out strains, where we observe no activation of MP kinase. We also show that Raf-1 is already present in immature mouse oocytes, and thus is unable to compensate for the lack of Mos in MP kinase activation. Moreover, in wild-type oocytes the electrophoretic mobility of Raf-1 does not change at the time of MP kinase activation, reinforcing the idea that Raf-1 is not involved in ERK1 and ERK2 activation during the resumption of meiosis in mouse oocytes. We show that the changes in MPF activity are similar in mos / compared to mos +/ oocytes demonstrating that Mos is not required for MPF reactivation during the M I to M II transition. The majority of oocytes from mos / mice are not arrested in metaphase II and extrude a second polar body as described by Colledge et al. (1994) and Hashimoto et al. (1994). However, we found that 76% of them do not stay in interphase, instead they undergo another transition from metaphase II to another M-phase (which we refer to as metaphase III, M III). During the transitions between M I/M II and M II/M III, we observe the formation of long microtubules resembling interphasic ones and partially decondensed chromosomes in the cytoplasm of mos / oocytes. These observations show that the Mos/.../MP kinase pathway is normally required for the maintenance of microtubules and chromatin in a metaphasic state during the meiotic divisions. Moreover, the mos / metaphase III oocytes present abnormal monopolar spindles and remain arrested for at least 10 hours at this stage, suggesting that microtubule organizing centers (MTOCs) are altered in this mutant strain. ll these data present the first evidence of a physiological role of the Mos/.../MP kinase cascade in the control of microtubules as well as chromatin organization during meiosis in the mouse. MTERILS ND METHODS Collection and culture of oocytes Immature oocytes arrested at prophase I of meiosis were obtained by removing ovaries from 5- to 6-week-old Swiss female mice (nimalerie Spécialisée de Villejuif, Centre National de la Recherche Scientifique, France), or from mutant mos / and mos +/ mice (Colledge et al., 1994). The ovaries were placed directly into warmed (37 C) M2 medium (Whittingham, 1971) and ovarian follicles were punctured to release the enclosed oocytes. Only those immature oocytes displaying a germinal vesicle (GV) were collected and cultured further in M2 medium under liquid paraffin oil at 37 C in an atmosphere of 5% CO 2 in air. The resumption of meiotic maturation was typically observed 1 hour after release from the follicules. The oocytes were scored for GVD, for the extrusion of the first polar body and for the extrusion of the second polar body. Since the polar bodies often fragment and disappear in mouse oocytes, we systematically isolated oocytes when they formed the first, then the second polar body. Immunoblotting Oocytes at the appropriate stage of maturation were collected in sample buffer (Laemmli, 1970) and heated to 100 C for 3 minutes. C Fig. 1. () Mos synthesis during maturation. Mos immunoblot of GV-arrested oocytes (lane 1), oocytes collected 1, 3 and 6 hours after GVD (lanes 2 to 4 respectively), and M II-arrested oocytes (lane 5). Each sample contains 300 oocytes. () ERK1 and ERK2 activation during maturation. ERK immunoblot of GV-arrested oocytes (lane 1), oocytes collected 1, 3 and 6 hours after GVD (lanes 2 to 4 respectively), and M II-arrested oocytes (lane 5). Each sample contains 10 oocytes. (C) Raf-1 electrophoretic mobility does not change during maturation. Raf-1 immunoblot of GV-arrested oocytes (lane 1), oocytes cultured in M2 medium until GVD (lane 2), 1, 3 and 5 hours after GVD (lanes 3 to 5 respectively), and M IIarrested oocytes, collected after superovulation (lane 6). Each sample contains 20 oocytes. Molecular mass markers, 10 3, are on the right. D, germinal vesicle breakdown.

3 Mos during meiotic maturation in the mouse 817 Fig. 2. () Kinetics of meiotic maturation in mos / and mos +/ mice. Percentage of oocytes undergoing germinal vesicle breakdown (GVD), first polar body extrusion (P1) and second polar body extrusion (P2) during meiosis resumption of oocytes from mos / and mos +/ mice. Total number of oocytes scored: 150 for both strains. () Kinetics of histone H1 kinase activity in mos / and mos +/ mice. Each sample contains 3 oocytes. The proteins were separated by electrophoresis in 10% polyacrylamide containing 0.1% SDS and electrically transferred onto nitrocellulose membranes (Schleicher and Schuell, pore size 0.45 µm). Following transfer and blocking for 2 hours in 3% skimmed milk in TS (10 mm Tris, ph 7.5, 140 mm NaCl), containing 0.1% Tween- 20 (TS/Tween), the membrane was incubated overnight at 4 C with one of the primary antibodies diluted 1:300 in blocking solution. The three kinases Raf-1, Mos, and MP kinases were detected using an anti-raf-1 monoclonal antibody (Transduction Laboratories), an anti- Mos polyclonal antiserum (Gallick et al., 1985), and an anti-erk polyclonal affinity purified antiserum (no. 691; Santa Cruz iotechnology, Inc.). We used either an anti-rabbit immunoglobulin (when the polyclonal antibodies were used as first layers) or an anti-mouse immunoglobulin (for the monoclonal antibody) second layer antibodies conjugated to horseradish peroxidase (mersham) diluted 1:1000 in 3% skimmed milk in TS/Tween. The membranes were washed three times in TS/Tween and then processed using the ECL detection system (mersham). ll experiments were repeated at least three times. Immunofluorescence The fixation and labeling of oocytes were performed as described by Kubiak et al. (1992). We used the rat monoclonal antibody YL1/2 specific for tyrosinated α-tubulin (Kilmartin et al., 1982) and the mouse monoclonal MPM-2 antibody, which stains MTOCs in M- phase (Rao et al., 1989). s second layers, we used either a fluorescein-conjugated anti-rat antibody (Miles) or a rhodamine-conjugated Fig. 3. MP kinase is not activated in mos / mice. () ERK immunoblot of maturing oocytes from a mos / mouse collected 3, 5 and 8 hours after GVD (lanes 1 to 3 respectively) and from a mos +/ mouse 3 and 5 hours after GVD (lanes 4 and 5), and in GV (lane 6). () MP kinase activities measured in maturing oocytes from mos / mice and from mos +/ mice. anti-mouse antibody (Miles). The chromatin was visualized using propidium iodide (Molecular Probes; 5 mg/ml in PS). Samples were observed with a io-rad MRC-600 confocal microscope. Kinase assays The MP kinase and histone H1 kinase assays were performed as described by Félix et al. (1989) and oulton and Cobb (1991). Three to five oocytes were transferred into 1 µl of M2 medium, frozen immediately in a mixture of dry ice and ethanol, and stored at 80 C. For the combined histone H1/MP kinase assays, 2 µl of twice concentrated reaction buffer (180 mm EGT, ph 7.3, 300 mm β-glycerophosphate, 40 mm MgCl 2, 4 mm DTT, 4 mm benzamidine, 20 mm NaF, 0.4 mm sodium orthovanadate, 4 mm TP) supplemented with protease inhibitors was added to the sample, which was then thawed and refrozen once to lyse the cells. The reaction was started by the addition of 2 µl of a solution containing either 1.5 mg/ml MP or 66 µg/ml histone H1 and 0.5 mci/ml [ 32 P]TP, and lasted 30 minutes at 30 C. The reaction was stopped by adding twice concentrated sample buffer (Laemmli, 1970) and boiling for 3 minutes. The samples were then analysed by electrophoresis in 15% SDS-PGE, followed by autoradiography. RESULTS Mos accumulates concomitantly with MP kinase activation during maturation of mouse oocytes We examined mouse oocytes at different stages of meiotic maturation using anti-mos polyclonal antibodies previously characterized for immunoblotting in mouse oocytes (Weber et al.,

4 818 M.-H. Verlhac and others C D Fig. 4. Organization of the microtubules and chromosomes during the M I to M II transition in mos +/ mice. Microtubules are shown in green and chromosomes in red. () n oocyte collected in metaphase I; () an oocyte collected during the M I to M II transition (8 hours after GVD); (C,D) an oocyte arrested in metaphase II (12 and 16 hours after GVD respectively). Total number of oocytes scored: ). We observed that Mos was synthesized after GVD. The Mos kinase could be detected by immunoblotting 3 hours after GVD and reached a maximal level in metaphase IIarrested oocytes (Fig. 1). t the same time the electrophoretic mobility of ERK1 and ERK2 was reduced, demonstrating that their activation had occurred through phosphorylation (Fig. 1). Thus, in mouse oocytes, Mos is synthesized concomitantly with ERK1 and ERK2 activation, well after GVD. The electrophoretic mobility of Raf-1 does not change during maturation previous study has shown that Raf-1 mrns are present in mouse oocytes (Pal et al., 1993). To determine if the Raf-1 kinase is present in mouse oocytes during maturation and whether it is phosphorylated, oocytes were collected at different stages of maturation and examined by immunoblotting using a monoclonal antibody directed against a sequence of human Raf-1. In GV oocytes, a unique protein of relative molecular mass (the molecular mass of vertebrate Raf-1) was detected by the antibody as shown in Fig. 1C. The electrophoretic mobility of this protein did not change during in vitro maturation (lanes 1-5). However, in M II-arrested oocytes, bands of higher molecular mass were recognized by the anti-raf-1 antibody (lane 6). Thus, the electrophoretic mobility of Raf-1 remains constant and high during the process of meiosis in mouse oocytes, suggesting that it is not phosphorylated and thus not active (Morrison et al., 1989), at least until the metaphase II stage. Fig. 5. Organization of the microtubules and chromatin during the M I to M II and M II to M III transition in mos / mice. Microtubules are shown in green and chromosomes in red. () n oocyte collected in M I; () an oocyte collected during the M I to M II transition (9 hours after GVD); (C, C and C ) higher magnification of the chromatin from oocytes collected during the M I to M II transition (C and C, mos / ; C, control mos +/ ); (D) an oocyte in M II; (E) an oocyte collected during the M II to M III transition; (F) an oocyte in M III. Total number of oocytes scored: 150. Mos is not involved in the timing of meiotic maturation and the control of MPF activity but is required for MP kinase activation To check whether Mos was involved in the activation of MP kinase in mouse oocytes, we used the mutant strain of mos / mice (Colledge et al., 1994). Maturing oocytes, either from mos / mice or from mos +/ mice, were collected and synchronized by scoring for GVD, which allowed us to follow precisely their progression into the meiotic cell cycles. We analysed the kinetics of meiotic maturation and samples were collected at various times after GVD to measure histone H1 kinase activity. We assessed the phosphorylation state of ERK1 and ERK2 and the activity of myelin basic protein (MP) kinase. Fig. 2 shows that the timing of GVD (occurring between 0.5 and 1 hour after the beginning of

5 Mos during meiotic maturation in the mouse 819 oocyte culture) and extrusion of the first polar body (occurring between 7 and 8 hours after GVD) was very similar in oocytes from mos / and mos +/ mice. The mos / oocytes do not arrest in M II (Colledge et al., 1994; Hashimoto et al., 1994). They remained in metaphase II for 2-4 hours and then extruded a second polar body hours after GVD (Fig. 2). We observed both in mos / and in mos +/ oocytes that the histone H1 kinase activity began to rise at the time of GVD and reached a plateau in metaphase I (Fig. 2). The MPF activity in both types of oocyte decreased during the M I to M II transition and increased again in M II (Fig. 2), as previously observed during maturation (Kubiak et al., 1992). However, in mos / oocytes, MPF activity decreased again at the time of second polar body extrusion (Fig. 2). Oocytes from mos / mice, collected in metaphase I at 3 and 5 hours after GVD (Fig. 4 lanes 1 and 2), or after extrusion of the first polar body (Fig. 3 lane 3) contained the fast migrating forms of ERK1 and ERK2, which correspond to the inactive forms of MP kinases normally present in immature oocytes (Fig. 3 lane 6). In contrast, control oocytes from mos +/ mice showed a normal pattern of ERK1 and ERK2 activation (Fig. 3 lanes 4 and 5). We never detected a significant rise in the activity of MP kinase in oocytes from mos / mice, while MP kinase activity increased normally in mos +/ oocytes (Fig. 3). These results show that ERK1 and ERK2 are not activated during meiosis in oocytes from mos / mice. Thus, Raf-1 cannot compensate for the lack of Mos in these oocytes. However, the lack of MP kinase activity in mos / oocytes does not interfere with the progression of meiotic maturation and the changes in MPF activity until the metaphase II arrest. Mos is necessary for normal organization of the microtubules and chromatin during the transitions between two meiotic M-phases To check for an eventual role of the Mos/.../MP kinase cascade in microtubules and chromatin organization during the M I to M II transition (Verlhac et al., 1994), we processed oocytes for immunofluorescent staining of the microtubules and the chromosomes. We observed the classical M I to M II progression in mos +/ mice: the M I spindle appeared 6 hours after GVD (Fig. 4), the first polar body was extruded 8 hours after GVD and at that time the microtubules were short and concentrated mainly in the midbody near the condensed chromosomes (Fig. 4), the M II spindle was visible 10 hours after GVD (Fig. 4C) and was still present 4 hours later, due to the CSF-induced arrest (Fig. 4D). In mos / mice, after the formation of the M I spindle (Fig. 5), the oocytes extruded the first polar body and then progressed to metaphase II (Fig. 5D). However, during the M I to M II transition, instead of short microtubules localized in the vicinity of the condensed chromosomes, we could observe long microtubules radiating from a mass of slightly decondensed chromatin (Fig. 5 and compare C and C with C ). Thus, in mos / oocytes, the organization of microtubules and chromatin evolves towards an interphase-like state during the M I to M II transition. We found that 76% of the mos / oocytes that had extruded a first polar body, expelled subsequently a second polar body and entered a transition from M II (Fig. 5D) to a third M-phase stage (Fig. 5F), that we will call metaphase III (M III). Here again, during the transition from M II to M III, we observed the presence of long microtubules and slightly decondensed chromosomes (Fig. 5E). The remaining 24% either did not extrude the second polar body, or its extrusion was greatly delayed (4 to 6 hours later). Of the remaining 24% mos / oocytes which did not extrude their second polar body in time, we observed that 2/3 were arrested at a very peculiar state 24 hours after GVD. sters of microtubules had gathered in the oocyte center, near to the condensed chromatin and no spindlelike structures were evident (Fig. 6C). Only 1/3 of the 24% were in interphase at this time and possessed a well formed network of cortical microtubules and nuclei with decondensed chromatin (Fig. 6D). These oocytes always possessed two pronuclei, indicating abnormal chromosomal segregation. Together these results suggest that in the absence of Mos, the microtubules and chromatin evolve towards an interphaselike state during the meiotic transitions. They also show that 24% of the mos / oocytes present severe defects in M II spindle formation and/or function. The metaphase III mos / oocytes are arrested with monopolar half-spindles second meiotic transition leading to a third meiotic M-phase, the so-called metaphase III (M III), can be induced in wildtype oocytes by ethanol activation of freshly ovulated oocytes (13-15 hours post-hcg; Kubiak, 1989). The organization of the microtubule network in M III-arrested mos / oocytes was very unusual compared to those from control mos +/ M III oocytes C Fig. 6. Organization of the microtubules and chromatin in mos / mice 24 hours after GVD. Microtubules are shown in green and chromosomes in red. () Higher magnification of a mos / oocyte which expelled its second polar body between 11 and 14 hours after GVD and was arrested for 10 hours in metaphase III, arboring a monopolar spindle. This oocyte was also stained with the MPM-2 monoclonal antibody (in blue). () Control mos +/ oocyte, which has been activated with ethanol 14 hours after GVD, and is arrested in M III. (C,D) mos / oocytes, which either did not extrude the second polar body or extruded it after a 4- to 6-hour delay. D

6 820 M.-H. Verlhac and others obtained after ethanol activation 14 hours post GVD (Kubiak, 1989). Monopolar half-spindles were observed in 90% of mos / oocytes and we could only detect a single pole after staining with the MPM-2 antibody (Fig. 6). This third M-phase is not a real metaphase since the spindles are not bipolar, however, to simplify the terminology we will refer to it as M III. ll the M III spindles present in control ethanolactivated mos +/ oocytes were found to be bipolar (Fig. 6). Thus the monopolar spindles present in the M III mos / oocytes are not an artefact due to in vitro culture, but their presence is related either directly or indirectly to the loss of Mos function. Moreover, the M III oocytes from mos / mice possessed unphosphorylated and inactive forms of ERK1 and ERK2, which was not the case for M III-arrested mos +/ oocytes obtained after the parthenogenetic activation of in vitro matured oocytes (Fig. 7). Finally, we observed that the M III mos / oocytes were arrested in M-phase for at least 10 hours, probably as a consequence of the spindle checkpoint, since no CSF is present in these oocytes. In conclusion, we must point out that all mos / oocytes showed defects in spindle microtubule organization during the late stages of meiotic maturation that may be due to alterations in MTOCs: 76% of them had monopolar M III spindles, while 24% had severe defects in M II spindle formation and/or function. DISCUSSION Fig. 7. Oocytes from mos / mice arrested in metaphase III do not show MP kinase activity. () ERK immunoblot of immature oocytes (lane 1) and M III-arrested oocytes (lane 2) from mos / mice, and of M II-arrested oocytes from mos +/ mice (lane 3). () MP kinase activity in immature oocytes (lane 1) and M IIIarrested oocytes (lane 2) from mos / mice, and in M II-arrested (lane 3) and M III-arrested (lane 4) oocytes from mos +/ mice. This work was performed to analyse the role of Mos and Raf- 1 in the control of MP kinase activation, and also to determine some physiological roles of Mos during mouse meiotic maturation. Mos, but not Raf-1, is required for MP kinase activation Two pathways can trigger the resumption of meiosis and activation of p42mapk in Xenopus oocytes: the physiological progesterone pathway and the receptor tyrosine kinase pathway. Mos would appear to play a role mainly in the progesterone pathway, and Raf-1 in the receptor tyrosine kinase pathway, with a possible interaction of these two pathways (Fabian et al., 1993). We observed a good correlation between the accumulation of Mos and MP kinase activation in mouse oocytes, suggesting that there is a link between Mos synthesis and MP kinase activation. This correlation has also been described in Xenopus oocytes (Nebreda and Hunt, 1993). We then showed that MP kinase activation does not occur in mos / mice, implying that Mos is normally required for MP kinase activation. We also demonstrate that, similar to Xenopus oocytes, Raf-1 is present in immature mouse oocytes (Fabian et al., 1993; Muslin et al., 1993), but does not compensate for Mos in the MP kinase activation pathway in oocytes from mos / mice, thus suggesting strongly that Raf-1 is not involved in this process in the mouse oocyte. In Xenopus oocytes, the electrophoretic mobility of Raf-1 shifts to the slower migrating form at the time of germinal vesicle breakdown, when MP kinase is activated (Fabian et al., 1993; Muslin et al., 1993). However, in mouse oocytes the electrophoretic mobility of Raf-1 does not change during the early stages of maturation and corresponds to the fast migrating form of the kinase. It is not clear whether the slow migrating form of Raf-1 corresponds to the active kinase, however the fast migrating species corresponds to the inactive form of the enzyme (at least in vertebrate cells; Morrison et al., 1989; Samuels et al., 1993). This likely lack of Raf-1 activation in mouse oocytes when MP kinase activation occurs normally strengthens the idea that Raf-1 is not required for MP kinase activation during the resumption of meiosis. This result is consistent with the fact that the synthesis of Ras, which lies upstream of Raf-1, is not required for meiosis in mouse oocytes, but is required for progression to the 2-cell stage (Yamauchi et al., 1994). However, we do not exclude the possibility that Raf-1 plays a role later during fertilization, since M II-arrested mouse oocytes were found to contain the slower migrating form that corresponds to the phosphorylated form of Raf-1. The fact that in mouse oocytes, Mos but not Raf-1, is required for MP kinase activation is an important and new result, which suggests that Mos might be involved specifically in the physiological control of MP kinase activation during vertebrate meiotic maturation. Mos and MP kinase are not involved in MPF reactivation during the M I to M II transition The timing of germinal vesicle breakdown and the extrusion of the first polar body, as well as changes in MPF activity, were similar in mos / and mos +/ oocytes. These results show that MPF reactivation occurs normally during the M I to M II transition. Moreover, the fact that most (76%) mos / oocytes undergo a second meiotic transition leading to M III reinforces the idea that Mos is not required for MPF reactivation. Together, our results prove that in mouse oocytes, in contradiction to previous studies performed both in Xenopus and mouse oocytes (O Keefe et al., 1991; Furuno et al., 1994; Gebauer et al., 1994), Mos is not required for the reactivation of MPF during meiosis.

7 Mos during meiotic maturation in the mouse 821 Mos is required for normal microtubule and chromatin organization during meiosis in mouse oocytes We did not observe any major differences in the formation of the first meiotic spindles in oocytes from mos / compared to mos +/ (data not shown). However, the organization of the microtubules and chromatin during the M I to M II (and M II to M III) transition was very different in mos / oocytes compared to the classical organization observed in mos +/ oocytes. The lack of Mos and/or active MP kinase in mos / oocytes induces the reorganization of the microtubules and chromatin towards an interphase-like state during these transitions. These results reinforce the hypothesis of a role for MP kinase in the maintenance of a metaphasic organization of the microtubules and chromatin in meiosis, when MPF is inactive (Verlhac et al., 1994). We have also presented here evidence that Mos is required for proper meiotic spindle formation. Surprisingly, mos / oocytes activate spontaneously in M II but do not directly enter interphase. Most of them (76%) undergo another meiotic transition (M II to M III) and remain arrested with monopolar spindles in M III. The occurrence of an M II to M III transition can be explained by the work of (Kubiak, 1989), which showed that ethanol activation of very young ovulated oocytes induces an M II to M III transition, whereas the same treatment on older ovulated oocytes induces an M II to interphase transition. The mos / oocytes also activate early in M II (since they do not arrest in M II), and consistently most of them then progress to metaphase III. The mos / oocytes which show a delay in extrusion of the second polar body have a tendency to enter interphase. However, M III spindles obtained after activation of in vitro or in vivo matured oocytes are normally bipolar (Kubiak et al., 1992; this work), and not monopolar as observed in mos / oocytes. This result suggests strongly that MTOCs function is altered in mos / oocytes and thus impairs the proper formation of the M III spindle. We also presented evidence that mos / oocytes show defects in M II spindle function, since 24% of them either cannot extrude the second polar body or extrude it after a 4- to 6-hour delay. The abnormal formation and/or function of the M II spindle in this population of oocytes is reinforced by the observation of abnormal chromosomal segregation (oocytes with two pronuclei of different sizes). The possibility that MTOC function is altered in mos / mice, which lack MP kinase activity, is consistent with previous observations showing that a fraction of MP kinase localizes to the MTOCs in mouse oocytes (Verlhac et al., 1993). ased on these microtubules abnormalities observed in vivo, our results strengthen the idea of an important role for Mos and activated MP kinase in controlling microtubule organization during meiotic maturation. Oocytes from mos / mice do not arrest in metaphase II but arrest in a third M-phase We observe that 76% of mos / oocytes are arrested for at least 10 hours at a stage that here we call metaphase III. In these M III-arrested mos / oocytes, the chromosomes remained condensed, the microtubules remained short as in M-phase, and we could observe a positive staining with the MPM-2 monoclonal antibody, indicating that they are indeed arrested in metaphase. This arrest explains why it takes so long for mos / oocytes to reform pronuclei (nuclei appear 24 hours after GVD) and why they do so rarely (in 8% of the cases after in vitro culture, and in 40% of the cases after in vivo culture; Colledge et al., 1994). It has been shown previously that mouse oocytes possess a spindle checkpoint which prevents them exiting from metaphase II after activation if the spindle is not properly formed (Kubiak et al., 1993; Winston et al., 1995). The majority of mos / oocytes are arrested in metaphase III not because of a CSF activity (Mos is absent and we show here that MP kinase is not active), but probably because their M III spindles are monopolar and so are not functional. The lack of CSF activity in these oocytes is somehow compensated for by the absence of a functional M III spindle. Whether this arrest in M III can lead to fertilized embryos possessing the normal ploidy requires further investigation. We thank N. Winston, E. Karsenti and H. J. Clarke for critical reading of the manuscript and R. Schwartzmann for his expert photographic work. We are grateful to R. rlinghaus for the gift of the anti-mos antibody. This work was supported by grants to M from La Ligue contre le Cancer, l ssociation pour la Recherche contre le Cancer and le Centre National de la Recherche Scientifique. REFERENCES vruch, J., Zhang, X.-F. and Kyriakis, J. M. (1994). Raf meets Ras: completing the framework of a signal transduction pathway. Trends iochem. Sci. 19, oulton, T. G. and Cobb, M. H. (1991). Identification of multiple extracellular signal-regulated kinases (ERKs) with antipeptide antibodies. Cell Regul. 2, Carnero,., Jimenez,. and Lacal, J. C. (1994). Progesterone but not ras requires MPF for in vivo activation of MPK and S6KII: MPK is an essential connexion point of both signaling pathways. J. Cell. iochem. 55, Colledge, W. H., Carlton, M.. L., Udy, G.. and Evans, M. J. (1994). Disruption of c-mos causes parthenogenetic development of unfertilized mouse eggs. Nature 370, Dent, P., Haser, W., Haystead, T.. J., Vincent, L.., Roberts, T. M. and Sturgill, T. W. (1992). ctivation of mitogen-activated protein kinase kinase by v-raf in NIH 3T3 cells and in vitro. Science 257, Draetta, G., Luca, F., Westendorf, J., rizuela, L., Ruderman, J. and each, D. (1989). Cdc2 protein kinase is complexed with both cyclin and cyclin : evidence for proteolytic inactivation of MPF. Cell 56, Fabian, J. R., Morrison, D. K. and Daar, I. O. (1993). Requirement for Raf and MP kinase function during the meiotic maturation of Xenopus oocytes. J. Cell iol. 122, Félix, M.., Pines, J., Hunt, T. and Karsenti, E. (1989). post-ribosomal supernatant from activated Xenopus eggs that displays post-translationally regulated oscillation of its cdc2+ mitotic kinase activity. EMO J. 8, Furuno, N., Nishizawa, M., Okazaki, K., Tanaka, H., Iwashita, J., Nakajo, N., Ogawa, Y. and Sagata, N. (1994). Suppression of DN replication via Mos function during meiotic division in Xenopus oocytes. EMO J. 13, Gallick, G. E., Sparrow, J. T., Singh,., Maxwell, S.., Stanker, L. H. and rlinghaus, R.. (1985). Recognition of mos-related proteins with an antiserum to a peptide of the v-mos gene product. J. Gen. Virol. 66, Gautier, J., Minshull, J., Lohka, M., Glotzer, M., Hunt, T. and Maller, J. L. (1990). Cyclin is a component of maturation-promoting factor from Xenopus. Cell 60, Gebauer, F., Xu, W., Cooper, G. M. and Richter, J. D. (1994). Translational control by cytoplasmic polyadenylation of c-mos mrn is necessary for oocyte maturation in the mouse. EMO J. 13, Gotoh, Y., Nishida, E., Matsuda, S., Shiina, N., Kosako, H., Shiokawa, K., kiyama, T., Ohta, K. and Sakai, H. (1991). In vitro effects on microtubule dynamics of purified Xenopus M phase-activated MP kinase. Nature 349,

8 822 M.-H. Verlhac and others Haccard, O., Lewellyn,., Hartley, R. S., Erikson, E. and Maller, J. L. (1995). Induction of Xenopus Oocyte meiotic maturation by MP kinase. Dev. iol. 168, Haccard, O., Sarcevic,., Lewellyn,., Hartley, R., Roy, L., Izumi, T., Erikson, E. and Maller, J. L. (1993). Induction of metaphase arrest in cleaving xenopus embryos by MP kinase. Science 262, Hashimoto, N., Watanabe, N., Furuta, Y., Tamemoto, H., Sagata, N., Yokoyama, M., Okazaki, K., Nagayoshi, M., Takeda, N., Ikawa, Y. and izawa, S. (1994). Parthenogenetic activation of oocytes in c-mos-deficient mice. Nature 370, Howe, L. R., Leevers, S. J., Gomez, N., Nakielny, S., Cohen, P. and Marshall, C. J. (1992). ctivation of the MP kinase pathway by the protein kinase raf. Cell 71, Itoh, T., Kaibuchi, K., Masuda, T., Yamamoto, T., Matsuura, Y., Maeda,., Shimizu, K. and Takai, Y. (1993). protein factor for ras p21- dependent activation of mitogen-activated protein (MP) kinase through MP kinase kinase. Proc. Nat. cad. Sci. US 90, Kilmartin, J. V., Wright,. and Milstein, C. (1982). Rat monoclonal antitubulin antibodies derived by using a new nonsecreting rat cell line. J. Cell iol. 93, Kosako, H., Gotoh, Y. and Nishida, E. (1994a). Mitogen-activated protein kinase kinase is required for the mos-induced metaphase arrest. J. iol. Chem. 269, Kosako, H., Gotoh, Y. and Nishida, E. (1994b). Requirement for the MP kinase kinase/mp kinase cascade in xenopus oocyte maturation. EMO J. 13, Kubiak, J. Z. (1989). Mouse oocytes gradually develop the capacity for activation during the metaphase II arrest. Dev. iol. 136, Kubiak, J. Z., Weber, M., de Pennart, H., Winston, N. and Maro,. (1993). The metaphase II arrest in mouse oocytes is controlled through microtubuledependent destruction of cyclin in the presence of CSF. EMO J. 12, Kubiak, J. Z., Weber, M., Géraud, G. and Maro,. (1992). Cell cycle modification during the transition between meiotic M-phases in mouse oocytes. J. Cell Sci. 102, Kuroda, S., Shimizu, K., Yamamori,., Matsuda, S., Imazumi, K., Kaibuchi, K. and Takai, Y. (1995). Purification and characterization of REKs fropm Xenopus eggs. J. iol. Chem. 270, Kyriakis, J. M., pp, H., Zhang, X. F., anerjee, P., rautigan, D. L., Rapp, U. R. and vruch, J. (1992). Raf-1 ctivates MP kinase-kinase. Nature 358, Labbé, J. C., Picard,., Peaucellier, G., Cavadore, J. C., Nurse, P. and Dorée, M. (1989). Purification of MPF from starfish: identification as the H1 histone kinase p34cdc2 and a possible mechanism for its periodic activation. Cell 57, Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, Lohka, M. J., Hayes, M. K. and Maller, J. L. (1988). Purification of maturation promoting factor, an intracellular regulator of early mitotic events. Proc. Nat. cad. Sci. US 85, Masui, Y. and Markert, C. L. (1971). Cytoplasmic control of nuclear behavior during meiotic maturation of frog oocytes. J. Exp. Zool. 117, Morrison, D. K., Kaplan, D. R., Escobedo, J.., Rapp, U. R., Roberts, T. M. and Williams, L. T. (1989). Direct activation of serine-threonine kinase activity of the protooncogene Raf-1 through tyrosine phosphorylation by the PDGF-beta receptor. Cell 58, Muslin,. J., Macnicol,. M. and Williams, L. T. (1993). Raf-1 protein kinase is important for progesterone-induced xenopus oocyte maturation and acts downstream of mos. Mol. Cell. iol. 13, Nebreda,. R. and Hunt, T. (1993). The c-mos proto-oncogene protein kinase turns on and maintains the activity of MP kinase, but not MPF, in cell-free extract of Xenopus oocytes and eggs. EMO J. 12, O Keefe, S. O., Kiessling,.. and Cooper, M. (1991). The c-mos gene product is required for cyclin accumulation during meiosis of mouse eggs. Proc. Nat. cad. Sci. US 88, Pal, S. K., Crowell, R., Kiessling,.. and Cooper, G. M. (1993). Expression of proto-oncogenes in mouse eggs and preimplantation embryos. Mol. Reprod. Dev. 35, Posada, J., Yew, N., hn, N. G., Vande Woude, G. F. and Cooper, J. (1993). Mos stimulates MP kinase in Xenopus oocytes and activates a MP kinase kinase in vitro. Mol. Cell. iol. 13, Rao, P. N., Zhao, J.-Y., Ganju, R. K. and shorn, C. L. (1989). Monoclonal antibody against centrosome. J. Cell Sci. 93, Sagata, N., Daar, I., Oskarsson, M., Showalter, S. D. and Vande Woude, G. F. (1989a). The product of the mos proto-oncogene as a candidate initiator for oocyte maturation. Science 245, Sagata, N., Watanabe, N., Van de Woude, G. F. and Ikawa, Y. (1989b). The c-mos proto-oncogene product is a cytostatic factor responsible for meiotic arrest in vertebrate eggs. Nature 342, Samuels, M. L., Weber, M. J., ishop, J. M. and McMahon, M. (1993). Conditional transformation of cells and rapid activation of the mitogenactivated protein kinase cascade by an estradiol-dependent human Raf-1 protein kinase. Mol. Cell. iol. 13, Shibuya, E. K. and Ruderman, J. V. (1993). Mos induces the in vitro activation of mitogen-activated protein kinases in lysates of frog oocytes and mammalian somatic cells. Mol. iol. Cell 4, Verlhac, M.-H., Kubiak, J. Z., Clarke, H. J. and Maro,. (1994). Microtubule and chromatin behavior follow MP kinase activity but not MPF activity during meiosis in mouse oocytes. Development 120, Verlhac, M. H., de Pennart, H., Maro,., Cobb, M. H. and Clarke, H. J. (1993). MP kinase becomes stably activated at metaphase and is associated with microtubule-organizing centers during meiotic maturation of mouse oocytes. Dev. iol. 158, Weber, M., Kubiak, J. Z., rlinghaus, R.., Pines, J. and Maro,. (1991). c-mos proto-oncogene product is partly degraded after release from meiotic arrest and persists during interphase in mouse zygotes. Dev. iol. 148, Whittingham, D. G. (1971). Culture of mouse ova. J. Reprod. Fert. 14 Suppl., Winston, N. J., McGuinness, O., Johnson, M. H. and Maro,. (1995). The exit of mouse oocytes from meiotic M-phase requires an intact spindle during intracellular calcium release. J. Cell Sci. 108, Yamauchi, N., Kiessling,.. and Cooper, G. M. (1994). The Ras/Raf signaling pathway is required for progression of mouse embryos through the two-cell stage. Mol. Cell. iol. 14, Zhou, R. P., Oskarsson, M., Paules, R. S., Schulz, N., Cleveland, D. and Vandewoude, G. F. (1991). bility of the c-mos product to associate with and phosphorylate tubulin. Science 251, (ccepted 7 December 1995)