Translation of incenp During Oocyte Maturation Is Required for Embryonic Development in Xenopus laevis

Abstract

The chromosome passenger complex (CPC) consists of Aurora-B kinase and several other subunits. One of these, incenp, binds Aurora-B and regulates its kinase activity. During Xenopus oocyte maturation, incenp accumulates through translation, contributing to aurora-b activation. A previous study has demonstrated that inhibition of incenp translation during oocyte maturation diminishes aurora-b activation but does not interfere with oocyte maturation, characterized by normal maturation-specific cyclin-b phosphorylation, degradation, and resynthesis. Here we have extended these findings, showing that inhibition of incenp translation during oocyte maturation did not interfere with meiosis I or II, as indicated by the normal emission of the first polar body and metaphase II arrest, followed by the successful emission of the second polar body upon parthenogenetic egg activation. Most importantly, however, when transferred to host frogs and subsequently ovulated, the incenp-deficient eggs were fertilized but failed to undergo mitotic cleavage. Thus, translation of incenp during oocyte maturation appears to be part of oocyte cytoplasmic maturation, preparing the egg for the rapid mitosis following fertilization.

Translation of incenp, a subunit of the chromosome passenger complex, during Xenopus oocyte maturation is not required for completion of meiosis, but it is necessary for mitotic cleavage of the embryo.

INTRODUCTION

In most vertebrates, the development of a diploid somatic cell to a haploid mature egg begins at the embryonic stage. During early embryogenesis, somatic gonadal precursors actively migrate to the developing gonad [1]. In the fetal ovaries, these cells, known as primordial germ cells, proliferate through mitotic cell division to increase their number before committing to a meiotic fate [2], proceeding to meiotic S phase, followed by homolog paring, synapsis, and homologous recombination (crossover), and finally arresting at the end of meiotic prophase (diplotene stage) as primary oocytes [3]. Primary oocytes remain arrested in prophase during the lengthy period of oogenesis, until the female becomes sexually mature. Under the influence of luteinizing hormone, the primary oocytes reinitiate meiosis (oocyte maturation). Oocyte maturation encompasses germinal vesicle (nuclear envelope) breakdown (GVBD), chromosome congression (metaphase I), homologue separation (anaphase I), and asymmetric cytokinesis to extrude the first polar body. Mature eggs in most vertebrate species are arrested and ovulated in metaphase II. Fertilization triggers anaphase II and the emission of the second polar body, producing a haploid pronucleus ready to fuse with the haploid sperm nucleus.

The proper completion of meiosis I and metaphase II arrest, as indicated by the progression of orderly chromosomal events (also referred to as nuclear maturation), is easily defined and has been the focus of research in oocyte maturation. In contrast, cytoplasmic maturation, a term broadly defined as extra-chromosomal readiness of the metaphase II eggs to support fertilization and the subsequent embryonic development, is much less well understood [4]. One example might be the maturation of calcium signaling capacities: only the mature eggs but not the oocytes are capable of sperm-induced intracellular calcium signaling that leads to cortical granule exocytosis and other egg activation events; and this change in calcium signaling capacity is independent of chromosome changes during oocyte maturation [5]. We have recently demonstrated that translation and accumulation of ornithine decarboxylase, a rate-limiting enzyme in cellular synthesis of polyamines (putrescine, spermidine, and spermine), during Xenopus oocyte maturation, although not required for first polar body emission or metaphase II arrest, is essential for producing healthy eggs competent for embryonic development [6].

In this study, we examined the possibility that translation and accumulation of incenp, a subunit of the chromosome passenger complex (CPC) is another component of oocyte cytoplasmic maturation. The CPC is composed of aurora-b, incenp, survivin and borealin [7]. While survivin and borealin act as scaffolding proteins, incenp regulates aurora-b kinase, which is expected to be essential for oocyte maturation [8–10]. Indeed, during Xenopus oocyte maturation, aurora-b is activated, peaking at the time of germinal vesicle breakdown (GVBD) [11]. Of all four subunits of the CPC, only incenp and survivin protein levels increase during oocyte maturation, coinciding with aurora-b activation. Antisense-mediated inhibition of incenp translation reduces but does not eliminate aurora-b activation, due to the presence of maternal incenp protein which is not affected by the antisense oligonucleotides. The incenp-deficient oocytes progress normally through oocyte maturation, as judged by normal GVBD and normal biochemical maturation: full activation of maturation promoting factor at GVBD, partial and transient degradation of cyclin-b shortly after GVBD, and resynthesis of cyclin-b thereafter [11]. The objective of this study was twofold: to extend the findings of Yamamoto et al. [11] by analyzing polar body emission and to determine if the accumulation of incenp during oocyte maturation was necessary for the subsequent embryonic cleavage. To this end, we employed morpholino oligonucleotides antisense to incenp translation start site, incenp_morpho, to inhibit incenp protein translation during progesterone-induced oocyte maturation. We found that these incenp-deficient oocytes indeed completed meiosis I and meiosis II successfully, thus supporting the earlier conclusion based on biochemical analyses [11]. However, following surgical transfer to ovulating host frogs to acquire the jelly coat necessary for fertilization [12], these incenp-deficient eggs were activated upon sperm binding but failed to undergo embryo cleavage. Therefore, incenp translation during oocyte maturation may be a component of cytoplasmic maturation, preparing the eggs for postfertilization mitosis.

MATERIALS AND METHODS

Chemicals were purchased from Sigma unless indicated otherwise. All morpholino oligonucleotides were purchased from Gene Tools (Philomath, OR). The primary antibody used was a polyclonal antibody raised in rabbit and targeting amino acids 884–901 of human INCENP (catalog no. ab12187; Abcam). This antibody also recognizes Xenopus laevis incenp [11]. Antibodies against β-tubulin were from Developmental Studies, Hybridoma Bank at the University of Iowa.

Xenopus incenp Expression Construct

The sequence of incenp (National Center for Biotechnology Information [NCBI] reference no. NM_001088421) was obtained from the NCBI nucleotide database. This sequence was used to probe for clones in the National Institutes for Basic Biology's (NIBB) Xenopus database (XDB) version 3.2. The incenp-a cDNA (XDB clone XL157o01) was returned and served as a template for PCR amplification using the following two primers: (forward) 5-CCGGAATTCAACGATGCAGAGTGCCT-3′; and (reverse) 5′-CCGGAATTCGTATTTGAGGCCATAACCC-3′). The PCR product was digested with EcoRI and ligated into pCS2+HA vector [13] that was similarly digested with EcoRI, creating HA-incenp. The plasmid was linearized with NotI, and mRNA was transcribed using the mMessage mMachine in vitro transcription kit (catalog no. AM1340; Ambion). Messenger RNA was then quantified using a Qubit fluorometer (catalog no. 32852; Invitrogen) and stored at 1 mg/ml at −80°C.

Host Transfer Experiments

All animal protocols were approved by the Animal Care Committee of Ottawa Hospital Research Institute. All our oocyte and embryo experiments were carried out in a room maintained at 20°C. We essentially followed the protocol described by Heasman et al. [12], plus the detailed protocols dealing with frog ovulation, fertilization, and embryogenesis [14]. Oocyte donors were primed with 50 U of pregnant mare's serum gonadotropin 3–10 days prior to oocyte retrieval. Oocytes were manually isolated from excised ovarian tissues [15] in OCM (prepared daily by mixing 480 ml of Leibovitz L-15 medium, 320 ml sterile water, 0.32 g of bovine serum albumin, and gentamicin to 0.5 mg/ml, pH 7.6–7.8). Oocytes were injected with ctrl_morpho (10 pmol per oocyte), or incenp_morpho (10 pmol per oocyte) or co-injected with incenp_morpho (10 pmol per oocyte) and 5 ng of HA-incenp mRNA. The incenp_morpho has the following sequence, 5′GGGACAGGCACTCTGCATCGTTCAT3′, antisense to the region surrounding the start codon (underlined) of Xenopus incenp mRNA [11]. The ctrl_morpho contains the same nucleotides but in a scrambled sequence by Gene Tools (5′GGAGCAGGAGACGCCTCACTTTTCT3′). The injected oocytes were cultured in OCM with 1 μM progesterone for 10–12 h at 20°C. Eggs were stained for 15 min with neutral red (0.25%, catalog no. N-6634), Bismarck brown (2%, catalog no. B-2759), or Nile blue (0.1%, catalog no. N-0766), and rinsed in OCM before host transfer. Recipient females were primed with human chorionic gonadotropin (50 U per frog; catalog no. CG10) 5–15 days prior to the day of egg transfer. Twelve hours prior to egg transfer, two or three frogs were each injected with 800 U of human chorionic gonadotropin and left in the 20°C room. In the morning, the female laying the best quality eggs [12] was chosen for egg transfer. Eggs were then surgically transferred into the abdominal cavity of an ovulating female anesthetized with MS222 (ethyl 3-aminobenzoate methane sulfate salt; catalog no. A5040). At 2.5–4 h after surgery, colored eggs were collected alongside host eggs and fertilized with a testis removed from a euthanized male. An intact and unused testis can be stored in fetal bovine serum with 20% high salt modified Barth solution (MBS) supplemented with gentamicin at 4°C for seven days without losing fertilization potency. We removed the jelly coat by incubation with 2% l-cysteine (pH 8, catalog no. W326305), 90 min after fertilization, when host eggs started to cleave. The de-jellied embryos were cultured in 0.1X MBS, monitored, and photographed.

For embryo injection experiments, embryos were de-jellied 20–30 min after fertilization and immediately, while the embryos were still at the 1-cell stage, injected with 10 pmol of morpholino oligomers in 3% Ficoll (catalog no. F4312). Embryos were cultured in 0.1X MBS at room temperature and cleavage of oocytes was scored as normal or abnormal at the indicated time points.

Confocal Fluorescence Microscopy

GV oocytes were injected with mRNA coding for proteins tagged with fluorescent peptides. For visualization of chromosomes, mRNA coding for histone H2B tagged with a red fluorescent protein (mRFP) [16] was used. Each oocyte received 5–10 nl of a 500–1000 μg/ml solution of mRNA. Imaging was performed using a 60× oil objective on a Zeiss Axiovert microscope with a Bio-Rad 1024 laser scanning confocal imaging system. Images were rendered to three dimensions (3D) and analyzed using Volocity imaging software (PerkinElmer) [17].

Parthenogenetic Activation of Metaphase II Eggs

Mature, metaphase II-arrested oocytes were stimulated to emit the second polar body by pricking them in the animal hemisphere with a fine needle (needle tip diameter, approximately 10 μm) in OR2 medium (82.50 mM NaCl, 2.5 mM KCl, 1 mM CaCl₂, 1 mM MgCl₂, 1 mM Na₂HPO₄, 5 mM N-2-hydroxyethylpiperazine-N′9-2-ethanesulfonic acid [HEPES], pH 7.8) [17]. Sister chromatid separation (anaphase II) occurred approximately 12 min after pricking and completion of second polar body emission takes an additional 15 min [17]. Oocytes are either scanned for the duration of second polar body emission, or examined at least 30 min after prick-activation.

Immunoblotting

Oocytes or de-jellied embryos were lysed by being forced through pipette tips in lysis buffer (20 mM HEPES, pH 7.3, 80 mM glycerolphosphate, 20 mM EGTA, 15 mM MgCl₂, 1 mM dithiothreitol, 0.5 mM phenylmethanesulphonyl fluoride, 10 μg/ml leupeptin, 1 μM okadaic acid; 5 μl per oocyte). The lysates were centrifuged for 30 min at maximum speed in a table-top microfuge, and the supernatants were mixed at a 1:1 ratio with 2× SDS sample buffer. Typically, one oocyte worth of extracts was analyzed, with anti-INCENP, and with anti-β-tubulin.

Statistics

Graphs and images are representative of at least three independent experiments. The only exception is the data shown in Figure 3, b and c, which is representative of two independent experiments. All statistical analyses were performed using GraphPad Prism version 5.02. Data were analyzed by using the Fisher exact test (two-tailed). Differences between groups was considered statistically significant if a P value of <0.05 was returned.

FIG. 3. .

The incenp mRNA rescues the cleavage defects caused by incenp_morpho. A) GV oocytes were left uninjected (lanes 1 and 2), or injected with ctrl_morpho (lane 3) or incenp_morpho (lane 4), or co-injected with incenp_morpho and incenp mRNA (lane 5). Lysates from 10 oocytes were pooled and one oocyte's worth was immunoblotted using anti-INCENP or anti-β-tubulin. B) Numbers of embryos with normal cleavage, abnormal cleavage, or no cleavage of the three treatment groups: Ctrl_morpho, incenp_morpho, and incenp_morpho plus incenp mRNA. The embryos were scored at 2.5 h pf. C) The most common phenotype of each of the three groups is shown at 2.5 h pf.

RESULTS

incenp_morpho Did Not Interfere with Meiosis I or Meiosis II

We used morpholino oligonucleotides antisense to Xenopus incenp mRNA [11], the incenp_morpho, to interfere with incenp translation. ctrl_morpho, which contains the same nucleotides as incenp_morpho but in a scrambled sequence, was used as control. As shown previously [11], GV oocytes contain incenp protein (Fig. 1A, lane 1). Progesterone triggers oocyte maturation, accompanied by further increase of incenp as well as an upward shift on SDS-PAGE, indicative of M-phase-specific phosphorylation [18] (Fig. 1A, lane 2). Injection of ctrl_morpho (Fig. 1A, lane 3) did not affect the increase in incenp protein nor its gel shift. Injection of incenp_morpho (Fig. 1A, lane 4) inhibited the increase in incenp levels but did not affect the shift. These results demonstrated that incenp_morpho, like antisense incenp oligonucleotides [11], efficiently inhibited incenp translation during oocyte maturation.

FIG. 1. .

Inhibition of incenp translation during oocyte maturation does not affect meiosis. Immature (GV) oocytes were uninjected or injected with ctrl_morpho or incenp_morpho, stimulated with progesterone and analyzed for incenp protein levels and polar body emission. A) Oocytes were left uninjected (lanes 1 and 2) or injected with ctrl_morpho (lane 3) or incenp_morpho (lane 4). Oocytes were incubated overnight in OCM (lane 1) or OCM containing 1 μM progesterone (lanes 2–4). Lysates from 10 oocytes were pooled and one oocyte's worth was analyzed by immunoblotting using anti-INCENP or anti-β-tubulin. Data shown are representative of five experiments. B) Shown are the numbers of oocytes, over the total numbers examined, having emitted the first polar body after overnight incubation with progesterone. Some metaphase II oocytes were prick-activated, followed by either time lapse imaging (shown in D), or scanned for the presence of both first and second polar bodies at least 30 min after prick-activation. Shown are numbers of the pricked oocytes having emitted the second polar body, over the total number of pricked-activated metaphase II oocytes (Fisher exact test: P  =  1.000 for first polar body; P  =  0.629 for second polar body). C) Typical confocal chromosome (RFP-H2B) images of oocytes having emitted the first polar body and arrested at metaphase II. The same oocyte is depicted in top view (top panels) and side view (bottom panels). Schematic depicts the side view of such an oocyte with horizontal lines representing the plasma membrane, and polar body above and metaphase II spindle below. D) Typical time series, side view, of a ctrl_morpho oocyte (top row) and an incenp_morpho oocyte (bottom row) during prick activation. Left images represent metaphase II oocytes prior to prick-activation. Middle images represent anaphase II oocytes (12–15 min after prick activation). Note the separation of sister chromatids at anaphase with the haploid egg chromatids appearing fainter due to their greater distance from the cell surface [17]. Right images represent oocytes having successfully emitted the second polar body. Schematics depict the process of second polar body emission with horizontal lines depicting plasma membrane.

Yamamoto et al. [11] demonstrated that inhibiting incenp translation does not interfere with the maturation-specific cyclin-b behavior: cyclin-b phosphorylation at the time of GVBD, partial degradation shortly after GVBD and resynthesis before reaching metaphase II arrest. To confirm that meiosis I was indeed not affected by inhibiting incenp translation, we injected GV oocytes with mRNA coding fluorescent histone (RFP-H2B) to allow imaging of chromosome behavior (Fig. 1B, summary). Figure 1C shows fluorescent images of typical oocytes after injection of ctrl_morpho or incenp_morpho and treatment overnight with progesterone. In both groups, a metaphase II chromosome array can be seen alongside the first polar body. Inhibition of incenp translation thus did not interfere with the completion of meiosis I, confirming the earlier conclusion drawn from biochemical analyses of protein kinases [11].

Oocytes arrested at metaphase II were pricked with a fine needle to mimic fertilization and release the oocyte from metaphase II arrest, resulting in the completion of meiosis II via emission of the second polar body. We carried out live cell imaging [17] to determine the effect of incenp translation during oocyte maturation on a second polar body emission. Figure 1D shows the progression of oocytes injected with either ctrl_morpho or incenp_morpho from a metaphase II arrest (left column) through the separation of the chromosomes at anaphase II (middle column), and finally after the emission of the second polar body (right column). In both groups, the second polar body can be seen and is properly emitted, suggesting that accumulated incenp is not necessary for second polar body emission. Figure 1B summarizes these imaging experiments, indicating that incenp_morpho-mediated incenp translation inhibition did not affect meiosis I or meiosis II.

Translation of incenp During Oocyte Maturation Is Required for Mitotic Cleavage Following Fertilization

To determine the developmental capacity of in vitro-manipulated oocytes/eggs, we used the host transfer strategy developed by Heasman et al. [12]. GV oocytes were injected with incenp_morpho or ctrl_morpho and matured in vitro before being surgically inserted into the abdominal cavity of an ovulating host frog to allow the acquisition of the jelly coat necessary for fertilization. After fertilization, the embryos were monitored for any developmental defects. At 2.5 h postfertilization (pf), 80% (47 of 59) of the ctrl_morpho embryos were at the expected 8-cell stage and were developing normally, whereas only 3% (2 of 69) of incenp_morpho embryos were normal (Fig. 2, normal cleavage). Twenty percent (12 of 59) of ctrl_morpho embryos and 25% (17 of 69) of incenp_morpho embryos exhibited asymmetrical or partial cleavage patterns. In these cases, embryos contained either an odd number of cells, displayed furrows that never completely cleaved the embryo, or both (Fig. 2, abnormal cleavage). The majority (72% [50 of 69]) of incenp_morpho embryos exhibited no cleavage furrow (Fig. 2, no cleavage). These embryos (as did all other groups) showed physiological signs of activation, such as lifting of the vitelline membrane from the egg or contraction of the animal hemisphere pigmentation (not shown) [14].

FIG. 2. .

Inhibition of incenp translation during oocyte maturation inhibits first cleavage in the embryos. GV oocytes were injected with either ctrl_morpho or incenp_morpho and stimulated with progesterone overnight. Mature (metaphase II) eggs were dyed before being transferred to an ovulating frog to acquire the jelly coat. The uncolored host eggs and colored experimental eggs were fertilized and the resulting embryos were examined at 2.5 h pf and were scored as normal (normal 8-cell embryo), abnormal (embryos with an abnormal number of cells or abnormal cleavage patterns), or no cleavage (embryos showing no signs of cleavage). The graph summarizes five independent experiments. Fisher exact test: P < 0.0001.

To further confirm the specificity of incenp_morpho, we performed rescue experiments in which HA-incenp mRNA was co-injected with incenp_morpho. The presence of the HA-coding sequence 5′ to the incenp coding sequence renders this rescue construct insensitive to incenp_morpho [6, 19]. Indeed, co-injection of HA-incenp mRNA along with incenp_morpho (Fig. 3A, lanes 5) rescues the level of incenp accumulated after oocyte maturation to levels similar to those in non-injected and ctrl_morpho injected oocytes (Fig. 3A, lanes 2 and 3, respectively). These rescue oocytes were also subjected to host transfer in order to test their developmental competence. Three groups of oocytes were injected with either ctrl_morpho (number of recovered embryos from this group: n = 4), incenp_morpho (n = 14; two experiments), or incenp_morpho + HA-incenp (n = 11; two experiments). All four (4 of 4) ctrl_morpho embryos were developing normally at 2.5 h pf, as those shown in Figure 2. At that stage, all of the 14 incenp_morpho embryos either cleaved abnormally (3/14) or exhibited no cleavage at all (11/14), similarly to that shown in Figure 2. In contrast, 9 out of 11 of the incenp_morpho + HA-incenp embryos appeared to be developing normally at 2.5 h pf while the other two (2/11) were cleaving abnormally. These experiments clearly indicated that injection of HA-incenp restored the levels of incenp proteins in the egg and rescued the cleavage defects caused by incenp_morpho.

The incenp_morpho injected into oocytes was expected to remain effective in the resulting embryos to inhibit incenp translation [20], which could contribute to the phenotype observed in incenp_morpho embryos. To rule out this possibility, we injected ctrl_morpho or incenp_morpho directly into 1-cell embryos. Analyzing incenp protein by immunoblotting indicated a gradual decrease of incenp protein following fertilization (Fig. 4A, lanes 3, 4, 5, 6, and 9), consistent with the observed incenp degradation in egg extracts after Ca²⁺-mediated activation [11]. However, the levels of incenp protein were indistinguishable between control embryos (Fig. 4A, lanes 6 and 9) and time-matched embryos injected with ctrl_morpho (Fig. 4A, lanes 7 and 10) or with incenp_morpho (Fig. 4A, lanes 8 and 11). These results suggest the absence of incenp translation during early embryogenesis, in contrast to the maternal translation observed during oocyte maturation. Not surprisingly, embryos injected with ctrl_morpho or with incenp_morpho all developed normally, at least until 7 h pf (Fig. 4B). These results support the conclusion that the no cleavage phenotype in incenp_morpho embryos is the result of inhibition of incenp translation during oocyte maturation, not during early embryogenesis.

FIG. 4. .

Injection of incenp_morpho after fertilization does not affect early embryogenesis. Embryos were left uninjected (-) or injected with ctrl_morpho (lanes 7 and 10) or incenp_morpho (lanes 8 and 11) at 20 min pf. A) Oocytes (lane 2 and lane 3 representing, respectively, in vitro matured and ovulated metaphase II eggs) or embryos were lysed at indicated pf time points and samples were immunoblotted for INCENP and β-tubulin. B) Embryos were assessed for cleavage patterns at 2.5 and 7 h pf. Shown are typical images. C) Percentage of embryos showing normal cleavage patterns is shown. Fisher exact test: P  =  1.000 for 2.5 h pf; P  =  0.528 for 7 h pf.

Finally, we have followed some incenp_morpho embryos that showed no cleavage phenotype at 2.5 h pf farther and observed that, between 5 h pf (Fig. 5A) and 7h pf (not shown), these embryos exhibited a surface pigmentation pattern similar to the cleavage pattern of control embryos at the same time of development. The significance of this observation will be discussed.

FIG. 5. .

The incenp_morpho embryos form furrow-like array. GV oocytes were injected with either ctrl_morpho or incenp_morpho, matured, subjected to the host transfer technique, and then fertilized. A) Top: A typical ctrl_morpho embryo photographed at 2.5 h pf and 5 h pf showing normal cleavage patterns at the respective embryonic stage. Bottom: A typical incenp_morpho embryo photographed at 2.5 h pf and 5 h pf. B) Hypothetical model where incenp_morpho embryos continue to divide nuclei in the absence of cytokinesis, resulting in a single cell containing multiple spindles. Once these spindles reach the surface of the embryo, they begin to direct the formation of furrows. Chromosomes are shown in red; microtubule spindles are shown in green.

DISCUSSION

We report here that incenp_morpho specifically and efficiently inhibited incenp translation during oocyte maturation. However, incenp_morpho did not affect the stability or the M phase-specific phosphorylation of the maternal incenp protein. These results confirmed those of Yamamoto et al. [11], who used antisense oligonucleotides to inhibit incenp translation. Yamamoto et al. [11] reported the apparently normal activation and dynamics of various maturation specific protein kinases in these oocytes in the absence of incenp protein translation. We extended those findings by demonstrating that oocytes injected with incenp_morpho underwent normal meiosis, emitting the first polar body and arresting at metaphase II. Furthermore, we demonstrated that the metaphase II eggs could be parthenogenetically activated to emit the second polar body. Therefore, translation of incenp during oocyte maturation is apparently dispensable for completion of meiosis I and meiosis II. In contrast, a recent study in mouse oocytes reported that Incenp-specific siRNA interferes with chromosome alignment and prevents polar body emission [21]. The simplest explanation might be that prophase mouse oocytes lack maternal INCENP protein, and therefore require de novo translation to complete meiosis, in contrast to prophase Xenopusoocytes which contain sufficient maternal incenp protein to support meiosis [11] (Fig. 1A).

In contrast to their ability to complete meiosis I and meiosis II, incenp_morpho eggs, upon fertilization, are incapable of embryonic cleavage. Therefore, higher levels of incenp protein, through translation of maternal mRNA during oocyte maturation, are specifically required for embryonic mitosis. What purpose would an excess of incenp protein serve? We speculate that it might have to do with the function of the CPC in cytokinesis and the differential localization of the spindle during meiosis and during first cleavage, respectively. The CPC is localized to the centromeres at metaphase and translocated to the central spindle at anaphase. However, to promote furrowing, the CPC must be translocated to the cortex of the plasma membrane [7, 22], or bridging the gap between the spindle midzone and the cortex [23]. During meiosis, the spindle is closely associated with the membrane [24, 25]. However, in 1-cell Xenopus embryos the spindle lies deeper inside the animal hemisphere cytoplasm and is farther away from the furrowing cortex [26]. The level of incenp protein accumulated during oogenesis might support sufficient CPC activity for furrowing when the spindle is closely associated with the membrane but significantly higher levels of incenp, accumulated through translation during oocyte maturation, may be required to promote furrowing when the spindle is much further away from the plasma membrane. In support of this notion was our observation that many of the no cleavage incenp_morpho embryos eventually exhibited surface pigmentation pattern similar to the cleavage pattern of ctrl_morpho embryos (Fig. 5A). In these embryos, it appeared as if nuclear divisions continued undisrupted but furrowing failed. As the number of nuclei increased and migrated closer to the cortex, the mitotic spindles were eventually positioned close enough to support furrowing (Fig. 5B). This model does not dispute other possible defects, including spindle assembly and bipolar chromosome attachment [22], due to suboptimal CPC activities.

In addition to incenp, Xenopus oocytes also accumulate several other mitotic proteins such as aurora-a [27] and the ran activator rcc1 [28] during oocyte maturation. Interestingly, inhibition of aurora-a translation (Chunqi Ma and X.J. Liu, unpublished results) or rcc1 translation [28] similarly does not appear to interfere with first polar body emission or metaphase II arrest. It remains to be determined if translation of these proteins during oocyte maturation is also required for embryonic mitosis. Postfertilization Xenopuszygotes undergo rapid (every 30 min) and synchronous mitosis until mid-blastula stage (12 cell cycles or 4096 cells) [29]. This rapid cell division may impose constraint on the protein translation machinery such that stockpiling of some essential mitotic proteins during the preceding oocyte maturation becomes necessary.