DOC1R: a MAP kinase substrate that control microtubule organization of metaphase II mouse oocytes

After ovulation, mouse oocytes arrest for several hours in metaphase II(MII) of the second meiotic division with a stable spindle. Typically, mouse meiotic spindles are barrel shaped and are devoid of astral microtubules. These meiotic spindles organize from MTOCs (microtubule organizing centers)that lack centrioles. During the arrest in metaphase II, chromosomes remain aligned on the metaphase plate, waiting for fertilization to trigger chromosome segregation. Not much is known about the precise mechanisms that allow the second meiotic spindle to remain correctly organized during the so-called CSF (cytostatic factor) (Masui and Markert, 1971) arrest in metaphase II of vertebrate oocytes. This CSF arrest is mediated by the MOS/.../MAPK (mitogen activated protein kinase) pathway and has been mainly studied on the aspect of MPF (M-phase promoting factor) (Masui and Markert,1971) stabilization. Indeed, during the metaphase II arrest, MPF(a complex between CDC2 and cyclin B) levels remain high, owing to the CSF activity generated by the MOS/.../MAPK pathway(Colledge et al., 1994; Dupre et al., 2002; Haccard et al., 1993; Hashimoto et al., 1994; Nebreda and Hunt, 1993; Posada et al., 1993; Verlhac et al., 1996). In Xenopus oocytes, MAPK phosphorylates p90rsk which in turns phosphorylates Bub1, which would inhibit the APC/C (anaphase promoting complex/cyclosome) to target cyclin B for degradation therefore maintaining a high MPF activity (Gross et al.,1999; Gross et al.,2000; Tunquist et al.,2002). However, the MOS/.../MAPK pathway also controls microtubule organization, as mouse Mos^-/- oocytes are not arrested in metaphase II, instead they activate spontaneously and remain arrested in a third metaphase with monopolar spindles(Verlhac et al., 1996). Furthermore, using a two-hybrid screen with MAPK as a bait on a mouse oocyte cDNA library, we have recently identified MISS (MAPK interacting and spindle stabilizing) a substrate of the MOS/.../MAPK pathway which mediates metaphase II spindle stability during the CSF arrest(Lefebvre et al., 2002). For the success of fertilization and future embryo development, it is therefore a key issue to understand at a cellular level how the MOS/.../MAPK pathway maintains a spindle correctly organized in metaphase II.

We describe the isolation of another MAPK partner using the same two-hybrid screen: DOC1R (D19ERTD144E - Mouse Genome Informatics), a murine homologue of a potential human tumor suppressor gene DOC1R (deleted in oral cancer one related) (Zhang et al., 1999). Until now, little was known about this protein that has been found on the basis of its homology with human DOC1, a human tumor suppressor gene(Todd et al., 1995). DOC1R is related to a coiled-coil region of a kinesin (KIF14) of unknown function. Like MISS, DOC1R is rich in proline residues in its N terminus. DOC1R has a perfect consensus site for MAPK phosphorylation, a potential CDK2 binding site and a potential cyclin/CDK binding site (Zhang et al., 1999). We show here that DOC1R has been conserved from Xenopus laevis to human, which suggests that it performs important functions in vertebrate species. In mouse oocytes, DOC1R is present at all stages of meiotic maturation and is regulated by multiple phosphorylations. Both cyclin B/CDC2 and MAPK are able to phosphorylate DOC1R in vitro, and the MOS/.../MAPK pathway phosphorylates DOC1R in vivo. This protein is localized in dots on microtubules especially on metaphase I and II spindles. Consistently, a DOC1RGFP fusion localizes to the metaphase II spindle. The depletion of DOC1R by microinjection of antisense (asRNA) or double-stranded(dsRNA) RNA directed against its endogenous mRNA has a strong effect on the metaphase II spindle morphology. Injected oocytes harbor spindles with astral microtubules, as well as numerous asters in the cytoplasm, suggesting that DOC1R regulates microtubule organization during the CSF arrest of metaphase II oocytes. We show that this phenotype is specific to DOC1R as it can be rescued by the overexpression of the Xenopus protein. Thus, we have discovered a new class of proline-rich proteins, MISS and DOC1R, substrates of MAPK that regulate microtubule organization during the CSF arrest of mouse oocytes.

Immature oocytes arrested in prophase I of meiosis were obtained by removing ovaries from 11-week-old OF1 and Mos^-/- female mice. Oocytes were removed and cultured as previously described(Verlhac et al., 1996). Nocodazole was diluted in the culture medium at 10 μM.

The two-hybrid screen was performed as previously described(Lefebvre et al., 2002).

For preparing RNA from mouse ovaries, total RNA were extracted using the Rneasy mini Kit (Qiagen). Immature mouse oocytes were lysed in PBS without total RNA extraction. Then 500 ng of RNA from ovaries, or 30 immature oocytes in sterile PBS were treated with 2 U of RQ1 DNAse (Promega) for 20 minutes at 37°C and heated for 5 minutes at 85°C. The first strand cDNA synthesis was performed with 50 U of MmuLV Superscript (Life Technologies) using 2.5μM random hexamer (pdN6, Pharmacia), 1 mM dNTPs (Promega), in 10 mM Tris-HCl pH 8.3, 5 mM MgCl₂, 50 mM KCl for 1 hour at 37°C,followed by 5 minutes at 95°C. The PCR amplification was performed using 5′-ATGCCTCGAGATGACGTACAAGCCAATCGC and 5′-ATGCGAATTCCCGTGCGGGCATTGCGTTCT primers, at 55°C for 30 cycles.

The pRN3MYC2-DOC1R was constructed by RT-PCR amplification of the DOC1R open reading frame from RNA isolated from mouse ovaries. Total RNA were extracted using the Rneasy Mini Kit (Qiagen). The reverse transcription was performed on 500 ng of RNA. The PCR amplification was done on 50 ng of RNA/DNA using 5′-GATCGAATTCATGWSNTAYAARCCNATHGCN and 5′-GATCGCGGCCGCTTACGTGCGGGCATTGCGTTC primers. The PCR product was then cloned into the pRN3MYC2 vector. The pRN3DOC1R-GFP was obtained by PCR subcloning at XhoI/EcoRI sites using 5′-ATGCCTCGAGATGACGTACAAGCCAATCGC and 5′-ATGCGAATTCCCGTGCGGGCATTGCGTTCT primers. The pET-DOC1R plasmid was constructed by subcloning at EcoRI/Not1 sites into the pET30a vector (Novagen). The DOC1R protein expressed from the pET30a vector contains a 6His repeat inside 51 additional amino acids, which makes it bigger of about 5 kDa.

The Xenopus cDNA was isolated from the Xenopus oocyte cDNA library (Iouzalem et al.,1998) by PCR using the 5′-ATGCGAATTCATGTCGTATAAACCAT and 3′-ATGCGCGGCCGCTCATGTGCGGGCACTGCGTTCTGT primers on 200 ng of library. The PCR product was further cloned into pRN3 at the EcoRI/NotI sites.

The in vitro synthesis of capped RNA, asRNA and dsRNA was performed using linearized plasmids with the mMessage mMachine kit for capped RNAs (Ambion) or with the Megascript Kit for as and dsRNA (Ambion). The capped RNAs and asRNA were then purified on RNeasy columns (Qiagen) and eluted in the injection buffer (10 mM Tris, 0.1 mM EDTA, pH 7.4) at a final concentration of 0.5μg/μl. Aliquots were then stored at -80°C. For the production of dsRNA, each strand of complementary RNA was first precipitated with ethanol then washed in phenol/chloroform and further incubated at 85°C for 5 minutes. After annealing for 3 hours at 37°C, 4 μl aliquots of dsRNA were stored at - 80°C.

Microinjection into mouse oocytes was performed as described(Verlhac et al., 2000).

Xenopus oocyte microinjection as well as oocyte extraction was performed as previously described (Gavin et al., 1999). Samples of 30 oocytes were extracted in 300 μl of lysis buffer (80 mM β-glycerophosphate, pH 7.4, 20 mM EGTA, 15 mM MgCl₂, 100 μg/ml leupeptin, 100 μg/ml aprotinin, 1 mM sodium orthovanadate, 2 mM PMSF). The oocyte extract was then clarified by centrifugation at 13,000 rpm at 4°C for 20 minutes. The supernatant was removed from the overlying lipid layer and the yolk protein and was further clarified by a second centrifugation at 13,000 rpm at 4°C for 10 minutes. The oocyte extracts were stored at -70°C. For the immunoprecipitation of endogenous xp42^mapk, we used an anti-ERK2 antibody conjugated to agarose (Santa Cruz Biotechnology). Xenopus oocyte extracts (300μl) were pre-cleared with 20 μl of protein A coupled to agarose beads for 30 minutes at 4°C. The cleared extracts were then incubated for 2 hours at 4°C with the antibody coupled to 20 μl of agarose beads. The beads were washed four times in 1 ml lysis buffer supplemented with 150 mM NaCl and then processed for immunoblotting.

Just after collection and lysis, oocytes were incubated with or without 400 U of λ-phosphatase (New England Biolabs) in λ-phosphatase buffer at 37°C for 1 hour and then processed for immunoblotting.

The DOC1R recombinant protein was prepared from the HMS174 E. colitransformed with the pET-DOC1R plasmid. The protein was purified from bacteria under native conditions as described in the QIAexpressionist handbook(Qiagen).

The in vitro kinase assays were performed on 0.1 μg of purified DOC1R or 2.5 μg of Histone H1 in 10 μl of Histone H1 kinase buffer [80 mMβ-glycerophosphate, 20 mM EGTA, 15 mM MgCl₂, 2 μg/ml leupeptin, 2 μg/ml aprotinin, 2.5 mM benzamidine, 1 mM 4-(2-aminoethyl)-benzenesulfonyl fluoride and 1 mM DTT, pH 7.4] with 0.025μg of commercial recombinant active rat ERK2 (BIOMOL, Ref SE-137) or 0.025μg of cyclin B/CDC2 (Biolabs) supplemented with 6.25 μCi of[γ-³²P]-ATP (3000 Ci/mmole) and 50 μM cold ATP, for 30 minutes at 37°C. Reactions were stopped by the addition of sample buffer(Laemmli, 1970) and were analyzed by SDS-PAGE followed by autoradiography.

Oocytes microinjected with MYC-DOC1R encoding RNA were collected 14 hours after GVBD and stored at -80°C in water containing 1 mM PMSF and 10μg/ml leupeptin/pepstatin/aprotinin. They were then processed as previously described (Louvet-Vallee et al.,2001).

Immunocytochemistry was performed as previously described(Brunet et al., 1999). To visualize endogenous DOC1R, oocytes were first fixed for 30 minutes in 3.7%formaldehyde at 30°C then treated for 10 minutes in 0.25% TritonX-100 in PBS, washed in 0.1% Tween 20/PBS, incubated for 1 hour with the primary affinity-purified anti-DOC1R antibody (at 1:50) in 3% BSA/0.1% Tween 20/PBS. They were then washed in 0.1% Tween 20/PBS and further incubated for 1 hour with the secondary anti-rabbit FITC (1:80) in 3% BSA/0.1% Tween 20/PBS. After incubation with the secondary antibody, all samples were washed in 0.1% Tween 20/PBS, incubated for 5 minutes in Propidium Iodide (5 μg/ml in 0.1%Tween20/PBS), then washed three times in PBS before mounting in Citifluor(Chem. Lab., UCK).

Oocytes at the appropriate stage of maturation were collected in sample buffer (Laemmli, 1970) and heated for 3 minutes at 100°C. We used the following antibodies: an affinity-purified anti-DOC1R antibody directed against the LVRECLAETERNART peptide (Sigma Immunochemicals), the anti-MYC 9E10 monoclonal antibody (sc-40;Santa Cruz Biotechnology), an anti-Ezrin antibody(Louvet-Vallee et al., 2001)and the anti-Erk antibody (sc-94; Santa Cruz Biotechnology).

MISS was isolated from a two-hybrid screen using rat ERK2 as a bait on a mouse oocyte cDNA library that we had prepared in the laboratory(Lefebvre et al., 2002). During the same screen, we had identified another partial cDNA of about 500 bp encoding a protein domain that interacts with rat ERK2. The homology of this partial cDNA with the human DOC1R (Zhang et al., 1999) coding sequence allowed us to isolate the full-length ORF of the murine DOC1R. The full-length DOC1R protein specifically interacts with ERK2 in a two-hybrid assay, as yeast strains co-transformed with plasmids expressing LexAERK2 and B42-DOC1R can induce expression of the two reporter genes (Leu2 and β-galactosidase; Fig. 1A). A MYC-tagged DOC1R protein also co-immunoprecipitates with endogenous MAPK from Xenopusoocyte extracts, whereas a negative control encoding for MYC-WNT11 does not(Fig. 1B compare lanes 3 and 4). All these data show that DOC1R is a MAPK partner. DOC1R, together with MISS (Lefebvre et al., 2002)and p90^rsk (Kalab et al.,1996) are the only known MAPK substrates in mouse oocytes.

DOC1R potentially contains one MAPK phosphorylation site(Fig. 1C, blue), one CDK2 binding site (red) and a cyclin/CDK binding site (green)(Shintani et al., 2000). Like MISS, this protein is very rich in proline residues in its N-terminal end(Fig. 1C)(Lefebvre et al., 2002).

By homology searches in databases, we also isolated ESTs encoding homologues of the DOC1R protein (Fig. 1C). The DOC1R and human DOC1R protein are almost identical (they present 95% of identities, Fig. 1D). Interestingly, the protein is highly conserved from Xenopus laevis to human (about 70% of identities, Fig. 1D). This conservation suggests that DOC1R mediates important function(s) in vertebrate species.

To investigate if the DOC1R mRNA was expressed in mouse ovaries and oocytes, we performed RT-PCR using specific primers on total RNA from ovaries and immature oocytes. The presence of the DOC1R mRNA in ovaries and oocytes is shown in Fig. 2A (lanes 2 and 3).

To examine if the DOC1R protein was present in mouse oocytes, an affinity-purified anti-DOC1R antibody was produced and tested on a DOC1R protein produced in bacteria. A specific band of apparent molecular weight of 28 kDa is recognized with the antibody(Fig. 2B, lane 4) that disappears when the antibody is pre-incubated with the immunogenic peptide(Fig. 2B, lane 3). As the expected molecular weight of the tagged-DOC1R is ∼19 kDa, the conformation of the protein must therefore modify its migration in SDS PAGE (see below). The antibody recognizes one band of about 28 kDa in mature mouse oocytes(Fig. 2B, lane 2) that also disappears when the antibody is pre-incubated with the immunogenic peptide(Fig. 2B, lane 1). The antibody seems highly specific for DOC1R, as it does not recognize any other protein on the immunoblot. As one mouse oocyte contains only 23 ng of total protein, this certainly explains why the antibody recognizes only a very faint band in mouse oocyte extracts (200 oocytes).

Therefore these experiments show that both DOC1R mRNA and protein are expressed in mouse oocytes.

Because it required 200 oocytes to see a faint band corresponding to DOC1R on immunoblot, we followed the behavior of an overexpressed DOC1R protein during meiotic maturation after microinjection of RNA encoding MYCDOC1R into immature oocytes. Samples of microinjected oocytes were collected at different times during meiosis resumption. In contrast to MYC-MISS protein, the MYCDOC1R protein accumulates in GV oocytes and at all stages of meiotic maturation(Fig. 3A)(Lefebvre et al., 2002). The apparent increase in MYC-DOC1R protein amount during meiosis is due to the progressive translation of injected RNA. In immature oocytes, the protein migrates to an apparent molecular weight of 26 kDa(Fig. 3A, lane 1). By contrast,during meiosis resumption the MYC-DOC1R protein up-shifts, migrating more slowly at about 30 kDa, suggesting that it is regulated by post-translational modifications.

We tested if these modifications could be due to phosphorylation. Immature oocytes in GV were injected with RNA encoding the MYC-DOC1R protein and collected in metaphase II. Half of the sample was treated withλ-phosphatase and the other half served as a control. As shown in Fig. 3B, the phosphatase treatment induces a down-shift of MYC-DOC1R electrophoretic mobility (compare lanes 1 and 2). This result indicates that DOC1R effectively undergoes phosphorylation during meiosis resumption.

The down-shift is not complete, suggesting that the treatment withλ-phosphatase was not totally efficient, or that the protein undergoes post-translational modifications other than phosphorylation.

As MYC-DOC1R undergoes post-translational modifications in oocytes collected one hour after GVBD, when MPF is active and MAPK inactive(Verlhac et al., 1994), we tested the ability of both kinases to phosphorylate DOC1R in vitro. For that,we incubated either purified cyclin B/CDC2 or active ERK2 in the presence of[γ-³²P]-ATP and purified DOC1R protein. As shown on the autoradiograph, both purified cyclin B/CDC2 and active MAPK are able to in vitro phosphorylate the DOC1R protein (Fig. 4A, lanes 1 and 4) to levels close to the histone H1 phosphorylation (Fig. 4A, lanes 2 and 5). This is consistent with the prediction of one MAPK phosphorylation site and one cyclin/CDK-binding site(Shintani et al., 2000) in the DOC1R protein sequence.

We took advantage of the Mos^-/- mice, which do not activate MAPK (Colledge et al.,1994; Hashimoto et al.,1994; Verlhac et al.,1996), to show that MYC-DOC1R is an in vivo substrate of the MOS/.../MAPK pathway. As shown in Fig. 4B, the SDS PAGE migration profile of MYC-DOC1R from Mos^-/- oocytes taken 12 hours after GVBD is completely different from wild-type oocytes and is characterized by three different bands from 26 to 30 kDa (compare lanes 1 and 2). The molecular weight of the lower band is the same as the DOC1R protein present in immature oocytes and the upper band migrates at an apparent molecular weight identical to the DOC1R protein present in mature wild-type oocytes.

To check whether the upper migrating forms of MYCDOC1R present both in wild-type and Mos^-/- oocytes share the same post-translational modifications, we performed 2D gel analysis. We observed that the upper band from wild-type oocytes resolved into two major isoforms(Fig. 4C, top panel). To position the different isoforms of MYC-DOC1R from one sample to the other, we used Ezrin as an internal control (not shown)(Louvet-Vallee et al., 2001). In metaphase II Mos^-/- oocytes, the isoforms corresponding to the upper migrating band were shifted towards the basic pole (OH-, Fig. 4C, bottom panel). These results show first that the band migrating in 1D at an apparent molecular weight of 30 kDa corresponds to different isoforms of DOC1R. Second these results show that MYC-DOC1R is less phosphorylated in Mos^-/- oocytes, as phosphorylations confer negative charges to proteins. These experiments suggest that the protein is effectively phosphorylated by the MOS/.../MAPK pathway and that other kinases are also responsible for DOC1R phosphorylation.

We followed the endogenous DOC1R localization during meiotic maturation. We show that DOC1R is present in immature oocytes(Fig. 5A) and during all stages of meiotic maturation (Fig. 5B-E), confirming the immunoblotting analysis of the exogenous MYC-DOC1R protein. DOC1R accumulates in immature oocytes in the nucleus(Fig. 5A), as previously described in interphasic human cells(Zhang et al., 1999). During the first meiotic division (Fig. 5B-D), DOC1R accumulates in the cytoplasm and localizes in dots in the vicinity of the chromosomes in a region enriched in microtubules. Microtubule re-organization has been well described during mouse oocyte maturation (Brunet et al.,1999); microtubules form early during the first division around the chromosomes and organize into a bipolar spindle about 2 hours after meiosis resumption. DOC1R localization follows microtubule organization during metaphase I. In metaphase II (Fig. 5E), DOC1R also accumulates in the cytoplasm and on the spindle. To prove that the DOC1R protein is associated with microtubules, we treated metaphase II oocytes with nocodazole, which induces microtubule depolymerization (Fig. 5F). In these oocytes, the DOC1R protein was diffusely located in the cytoplasm, which proves that DOC1R associates specifically with spindle microtubules. As a control, we performed immunofluorescence after blocking the purified antibody with the immunogenic peptide and no staining was observed(Fig. 5G). Furthermore, like the endogenous DOC1R protein, a DOC1R-GFP fusion localizes in the germinal vesicle of immature oocytes (Fig. 5H) and on the metaphase II spindle(Fig. 5I).

To investigate the role of DOC1R, we microinjected antisense RNA (asRNA),or double-stranded RNA (dsRNA) targeting the endogenous DOC1R mRNA. Oocytes were injected at the GV stage, collected at different stages of meiotic maturation and immunocytochemistry was performed to examine the chromosome and microtubule morphology (Fig. 6A-F). The injection of control dsRNA targeting the Xenopus Frizzled 7 mRNA did not affect meiotic maturation or spindle organization (Fig. 6B,K), with injected oocytes looking like non-injected ones in metaphase II(Fig. 6A,K). Typically,metaphase II-arrested oocytes present barrel-shaped spindles with no astral microtubules and no cytoplasmic asters of microtubules(Maro et al., 1985). When we injected asRNA or dsRNA against DOC1R mRNA, oocytes underwent their first meiotic division normally, extruded first polar bodies of normal size and with the same percentage as control oocytes (84.2±8.2% of polar body extrusion in the case of dsDOC1R versus 84±12% in the control oocytes),but were arrested in metaphase II with severe phenotypes: spindles showing astral microtubules and numerous cytoplasmic asters(Fig. 6D,E). The injection of the asRNA induced 50% of such phenotypes while the injection of dsRNA was more efficient and induced 70% of elongated spindles(Fig. 6K).

To determine the efficiency of our dsRNA, we checked the presence of the endogenous DOC1R in the injected oocytes. The injection of dsRNA against DOC1R induces a loss of DOC1R localization on the MII spindle but not on the metaphase I spindles (Fig. 6 compare G with I and H with J).

To prove that the phenotype we observed was specific, we injected the dsRNA together with RNA encoding the Xenopus DOC1R protein. Despite the amino acid conservation between the murine and the Xenopus DOC1R proteins, the Xenopus cDNA diverges from the mouse, and thus is corresponding cDNA cannot be targeted by the dsRNA. As shown on Fig. 6F,K, the Xenopusprotein complements the microtubule defect induced by the dsRNA. The injection of the RNA coding the Xenopus protein has no effect by itself on microtubule organization (Fig. 6C). So the cytoplasmic asters as well as nucleation of microtubules from the spindle poles is solely due to DOC1R depletion.

Altogether, our experiments demonstrate that DOC1R regulates microtubule organization at least in metaphase II (see Discussion). The depletion of DOC1R induces severe damage to the microtubule cytoskeleton, which may compromise chromosome segregation after fertilization and hence compromise further embryo development.

DOC1R was isolated in the same screen that identified MISS, a protein required for metaphase II spindle stability. We show here that DOC1R is a MAPK partner as it interacts in a two-hybrid approach and co-immunoprecipitates with endogenous MAPK from Xenopus oocyte extracts.

DOC1R has a consensus site for MAPK phosphorylation, a potential CDK2 binding site and a potential cyclin/CDK binding site(Shintani et al., 2000). We show that DOC1R is expressed at all stages of meiotic maturation and that it is regulated by multiple phosphorylations. First, cyclin B/CDC2 is able to phosphorylate DOC1R in vitro, which is consistent with the presence of one potential cyclin/CDK binding site in its coding sequence. Furthermore, DOC1R becomes phosphorylated early in metaphase I, when MPF is active but not MAPK. Second, MAPK is able to phosphorylate DOC1R in vitro and in vivo, in agreement with the presence of one consensus site for MAPK phosphorylation in its sequence.

By 2D gel analysis, we show that the post-translational modifications that affect DOC1R during meiotic maturation are quite complex. Some modifications can be attributed to MPF activation, some to the MOS/.../MAPK pathway and it cannot be excluded that DOC1R is also modified by post-translational modifications other than phosphorylations. The characterization of DOC1R modifications by phosphorylation will be the object of further studies.

Endogenous DOC1R is localized in dots on microtubules during all stages of meiotic maturation in particular on metaphase I and II spindles. These dots could reflect either an accumulation of DOC1R into vesicular structures associated with microtubules or macromolecular complexes containing DOC1R multimers.

Consistent with DOC1R association with microtubules, its depletion leads to drastic phenotypes in metaphase II arrested oocytes: elongated spindles enriched in astral microtubules and numerous asters of microtubules in the cytoplasm. The phenotype is specific to DOC1R depletion (1) because it is observed only when the endogenous protein can no longer be detected on the metaphase II spindle and (2) because overexpression of the Xenopusprotein complements the defects observed after dsRNA injection.

The phenotype can be explained by extensive microtubule polymerization from MTOCs (microtubule organizing centers) that in mouse oocytes are present at spindle poles as well as foci in the cytoplasm(Maro et al., 1985). It can be interpreted as a reduced ability of the chromatin to stabilize microtubules in its vicinity. Normally, meiotic spindles are devoid of astral microtubules and cytoplasmic MTOCs do not form microtubule asters. The depletion of DOC1R promotes microtubule nucleation and/or elongation. This suggests that DOC1R normally increases microtubule dynamics in metaphase II arrested oocytes. We cannot exclude that DOC1R has a similar function in metaphase I, because we could not completely deplete the endogenous pool of DOC1R during the first meiotic division. The absence of a detectable phenotype in metaphase I could be explained by a low turnover of the protein and therefore a lack of full efficiency of the dsRNA during metaphase I.

DOC1R localization and the phenotype observed after its depletion are consistent with the potential tumor suppressor role of the DOC1R gene(Zhang et al., 1999). Indeed,human tumors are characterized by chromosomal instability primary resulting from spindle organization defects(Saunders et al., 2000).

As for MISS, we could not find obvious invertebrate homologues of DOC1R. However, the protein sequence of DOC1R has been highly conserved from Xenopus laevis to human. Moreover, we show here that the Xenopus protein can functionally complement the mouse DOC1R. So we believe that both DOC1R and MISS mediate important functions specific to vertebrate species.

The meiotic metaphase II spindle harbors an unusual location in the cell:it is closely associated with the cortex, which is enriched in actin microfilaments. This association allows spindle rotation at fertilization, a prerequisite to second polar body extrusion that is essential for further embryo development. It is interesting that DOC1R is related to a coiled-coil region in a kinesin (KIF14) (for a review, see Miki et al., 2001) of unknown function that contains both a kinesin as well as a myosin domain. We can imagine that DOC1R somehow regulates interactions between spindle microtubules and microfilaments of the cortex that are necessary for maintaining proper MII spindle organization.

The discovery of new MAPK substrates, such as DOC1R and MISS, involved in the regulation of microtubule organization, is of crucial importance for our understanding of the processes controlling the metaphase II arrest of vertebrate oocytes. It suggests that the MOS/.../MAPK pathway not only controls CSF arrest by maintaining a high MPF activity through p90^rsk activity (Bhatt and Ferrell, 1999; Gross et al.,1999; Gross et al.,2000) but also ensures that the spindle is properly organized during the arrest for the success of fertilization. Our findings are consistent with the unpublished data showing that the maintenance of bipolar spindles assembled in Xenopus egg extracts requires MAPK activity, but not p90^rsk activity (Horne et al., 2003).

In conclusion, we have discovered two proteins that seem essential for mouse embryo development as their absence results in severe damage to the microtubule network of metaphase II mouse oocytes. Indeed C. elegans mei-1 and mei-2 mutants (proteins that regulate microtubule dynamics in meiosis and induce severing of microtubules to produce small meiotic spindles) extrude large polar bodies and show aneuploidy(Srayko et al., 2000).

We thank Paul Hossenlopp (Université Pierre et Marie Curie) for his help in the choice of the DOC1R immunogenic peptide. We thank Richard Schwartzmann (CNRS) for his help with confocal microscopy. We thank Olivier Gavet, Katia Wassmann and Julien Dumont for critical reading of the manuscript. This work was supported by grants from the Association pour la Recherche sur le Cancer (ARC 9913 and ARC 5792 to M.-H.V.). M.E.T. and C.L. are recipients of fellowships from the Ministère de l'Education et de la Recherche Technologique.