Xtr in the fertilized eggs of Xenopus has been demonstrated to be a member of a messenger ribonucleoprotein (mRNP) complex that plays a crucial role in karyokinesis during cleavage. Since the Xtr is also present both in oocytes and spermatocytes and its amount increases immediately after spematogenic cells enter into the meiotic phase, this protein was also predicted to act during meiotic progression. Taking advantage of Xenopus oocytes’ large size to microinject anti-Xtr antibody into them for inhibition of Xtr function, we examined the role of Xtr in meiotic progression of oocytes. Microinjection of anti-Xtr antibody into immature oocytes followed by reinitiation of oocyte maturation did not affect germinal vesicle break down and the oscillation of Cdc2/cyclin B activity during meiotic progression but caused abnormal spindle formation and chromosomal alignment at meiotic metaphase I and II. Immunoprecipitation of Xtr showed the association of Xtr with FRGY2 and mRNAs such as RCC1 and XL-INCENP mRNAs, which are involved in the progression of karyokinesis. When anti-Xtr antibody was injected into oocytes, translation of XL-INCENP mRNA, which is known to be repressed in immature oocytes and induced after reinitiation of oocyte maturation, was inhibited even if the oocytes were treated with progesterone. A similar translational regulation was observed in oocytes injected with a reporter mRNA, which was composed of an enhanced green fluorescent protein open reading frame followed by the 3′ untranslational region (3′UTR) of XL-INCENP mRNA. These results indicate that Xtr regulates the translation of XL-INCENP mRNA through its 3′UTR during meiotic progression of oocyte.
The germ cell lineage arises from an originally small number of diploid cells called the primordial germ cells (PGCs) at the early stage of embryogenesis. In some animals, such as in insects and amphibians, the PGCs inherit specific cytoplasm, termed germ plasm, which is localized asymmetrically in a clearly defined region of the egg (Mahowald 1962; Okada et al. 1974; Whitington & Dixon 1975; Ikenishi et al. 1986). Germ plasm has common features among some organisms studied such as the presence of crowded mitochondria and electron-dense germinal granules including abundant RNAs and proteins. In mice, on the other hand, although the egg does not have a clearly defined germ plasm, germinal granules appear in the PGCs that are differentiated from the pluripotent epiblast cells by the induction from neighboring extraembryonic ectoderm (Lawson et al. 1999; Yoshimizu et al. 2001). Therefore, germ plasm is thought to contain germ cell determinants (Smith 1966; Illmensee & Mahowald 1974).
Germ plasm components have been studied and several proteins have been found. Among these, Tudor in Drosophila is well-known to be localized at germ plasm and is indispensable for germ cell (pole cell) formation (Boswell & Mahowald 1985). Since Tudor appears to possess no other domain apart from 11 repeating tudor domains (Thomson & Lasko 2004; Arkov et al. 2006), the crucial function of the Tudor is thought to be ascribable to these repeating domains. The tudor domain, a conserved motif of 50 amino acids, is also found in other proteins of various organisms (Ponting 1997). Among them, human Survival of Mortor Neuron (SMN) protein was first demonstrated to interact with Sm protein in snRNPs through direct binding of the tudor domain of SMN protein with symmetric dimethyl-arginine (sDMA) in the arginine-glycine-rich tail of Sm protein (Liu et al. 1997; Bühler et al. 1999; Selenko et al. 2001). Recent studies have shown that Drosophila Tudor associates with Aubergine, a member of the Drosophila Piwi family proteins, in an sDMA-dependent manner and that this association is necessary for the localization of these proteins in the germ plasm (Nishida et al. 2009; Kirino et al. 2010). Piwi family proteins are expressed in the germline and associate with piRNAs which are small RNAs that are thought to play an important role in transposon silencing (Malone & Hannon 2009). Mutation of the tudor gene in Drosophila causes alteration of a piRNA population associated with Aubergine and disruption of the localization of Aubergine in germ plasm, resulting in a lack of germ cells, but not derepression of transposons (Nishida et al. 2009; Kirino et al. 2010), suggesting that the Tudor–Aubergine–piRNAs complex functions in a different manner from transposon-silencing and its exact role in germ cell formation remains unclear.
In mice, tudor domain-containing proteins including Tdrd1, Tdrd5, Tdrd6, Tdrd7, and Tdrd9 occur in germinal granules of spermatogenic cells and are involved in maintenance of germinal granule architecture and spermatogenesis (Chuma et al. 2003, 2006; Hosokawa et al. 2007; Shoji et al. 2009; Vasileva et al. 2009; Tanaka et al. 2011; Yabuta et al. 2011). Most of these tudor proteins are shown to associate with mouse Piwi proteins, Mili, Miwi, and/or Miwi2 in an sDMA-dependent manner (Shoji et al. 2009; Vagin et al. 2009; Vasileva et al. 2009; Kirino et al. 2010). Since loss-of-function of Tdrd1, −5, −7, and −9 by targeting their genes causes the upregulation of transposable elements, these proteins are thought to play an important role in retrotransposon silencing (Shoji et al. 2009; Tanaka et al. 2011; Yabuta et al. 2011). On the other hand, disruption of the Tdrd6 gene does not cause significant upregulation of retrotransposon activity (Vasileva et al. 2009) and the exact function of Tdrd6 remains elusive.
Previously, we identified Xtr, a Xenopus tudor repeat protein, in X. laevis (Ikema et al. 2002). This protein is the Xenopus homologue of Tdrd6 in mice and associates with Xiwi, the Xenopus homologue of Piwi (Lau et al. 2009) and several kinds of maternal mRNAs (Mostafa et al. 2009). Since Xiwi contains sDMA (Kirino et al. 2009), the Xtr–Xiwi interaction may be mediated in a similar manner to the sDMA-dependent Tudor–Aubergine interaction in Drosophila. Xtr is found in germline cells as well as early embryonic cells as a maternal factor until the tailbud stage (Ikema et al. 2002; Hiyoshi et al. 2005) with an especially large accumulation of Xtr in the germ plasm of oocytes and cleavage stage embryos (Lau et al. 2009; Mostafa et al. 2009). Since no Xtr in the adult somatic cells has been observed, this protein is expected to be involved in germ cell specification and/or totipotency. Loss-of-function of the Xtr in fertilized eggs caused the inhibition of both microtubule assembly around the nucleus and of karyokinesis progression after prophase (Hiyoshi et al. 2005). Therefore, mitotic cycles in germline cells and in early embryonic cells may be regulated by a unique mechanism that does not exist in predestined somatic cells. Although Xtr is thought to play an important role in this mechanism, how it controls the cell cycle remains largely unknown.
In spermatogenesis, Xtr mRNA occurs in the spermatogenic cells at all stages except for the round spermatid stage and thereafter, and its amount clearly increases in the late-secondary spermatogonia stage (Ikema et al. 2002). Since the amount of Xtr increases immediately after entering into the meiotic phase, it is concluded that after entering the meiotic phase, synthesis of the Xtr starts using Xtr mRNA that has been transcribed at the late-secondary spermatogonial stage (Hiyoshi et al. 2005). Thus, Xtr is proposed to be involved in both the progression of the germ cell meiotic cycle as well as mitotic cycles during early embryonic stages. In this study, we show that ablation of Xtr function by microinjection of anti-Xtr antibody in oocytes causes abnormal spindle formation and chromosome alignment at meiotic metaphase I and II. Furthermore, we suggest that these abnormalities are ascribable to the translational arrest of maternal mRNAs which are associated with Xtr.
Materials and methods
Preparation of oocytes and mature eggs
Sexually mature South African clawed frogs, X. laevis, were purchased from a dealer in Hyogo Prefecture, Japan. The ovaries were obtained from females anesthetized with 0.2% ethyl m-aminobenzoate methanesulfonate (Nacalai tesque). Fully-grown oocytes at stage VI (Dumont 1972) were defolliculated manually or by treatment of dissected ovary with 0.2% collagenase (S-1; Nitta Gelatine) for 2 h, and stored in 70% Leibovitz L-15 media (Sigma) until use. Oocyte maturation was induced by treatment of fully-grown oocytes with 10 μg/mL progesterone (Nacalai tesque) in Marc's Modified Ringers (MMR) at 22°C. Mature eggs were obtained by injection of human chorionic gonadotropin (Asuka Pharmaceutical) into females. Jelly envelopes were removed by treatment of the eggs with 2% cysteine solution (pH 8.0).
Construction of recombinant plasmid and in vitro transcription
The enhanced green fluorescent protein (EGFP) open reading frame (ORF) was obtained by polymerase chain reaction (PCR) using pEGFP-N1 (Clontech) as a template and the primer set: 5′-GGTCTAGACCACCATGGTGAGCAAGG-3′ and 5′-GGCCCGGGTTACTTGTACAGCTCGTCCAT-3′. The amplified DNA fragment was digested with XbaI and SmaI and inserted into pT7G(UK II+) vector (Okamoto et al. 2002). Construction of the EGFP/XL-INCENP 3′UTR reporter construct was carried out as follows: total RNA was extracted from ovary by acid guanidinium thiocyanate-phenol-chloroform method (AGPC method: Chomczynski & Sacchi 1987). Following the synthesis of first strand cDNA using oligo dT primer (Takara) and SuperScript III reverse transcriptase (Invitrogen), the 3′UTR of XL-INCENP was amplified by PCR using KOD-Plus-Ver.2 (Toyobo) and the primer set: 5′-GGCCCGGGGGAAAGGGCAGACTTTAC-3′ and 5′-CCCTCGAGGGTAAAAGGTGCTCTCCC-3′. After digestion of the amplified DNA fragment with SmaI and XhoI, it was inserted between the EGFP ORF and 3′UTR of Xenopus β-globin cDNA in the pT7G(UK II+) vector. For microinjection into Xenopus oocytes, these constructs were digested with NotI and in vitro transcribed using T7 mMESSAGE mMACHINE Kit (Ambion).
Five oocytes with or without progesterone treatment were homogenized in 50 μL of homogenizing buffer (HB: 50 mmol/L sucrose, 100 mmol/L KCl, 5 mmol/L ethylene glycol bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid [EGTA], 2 mmol/L MgCl2, 10 mmol/L N-(2-hydroxyethyl)piperazine- N′-3-propanesulfonic acid [HEPES]-NaOH; pH 7.7) containing 100 μmol/L each of N-tosyl-phenylalanyl chloromethyl ketone (TPCK; Nacalai tesque) and N-tosyl-l-lysyl chloromethyl ketone (TLCK; Nacalai tesque) and centrifuged at 15 000 g for 5 min. Forty microliters of the supernatant were recovered and an equal volume of 2× Laemmli's sample buffer (Laemmli 1970) was added to the supernatant. To examine for the presence or absence of Xtr in germinal vesicles (GVs), 27 GVs (a volume equivalent to that of one oocyte) were collected according to the method of Dettlaff et al. (1964), washed thoroughly with MMR, and dissolved in 20 μL of Laemmli's sample buffer. For the detection of Xtr, XL-INCENP, and EGFP, 20 μL of the samples were electrophoresed in a sodium dodecyl sulfate–polyacrylamide gel (SDS–PAGE) (a 6% gel for Xtr; a 10% gel for XL-INCENP and EGFP) and blotted onto sheets (for Xtr, Hybond ECL [GE Healthcare]; for XL-INCENP and EGFP, Immobilon P [Millipore]). The transblotted sheets were incubated with 1:1000 diluted anti-Xtr rabbit antiserum (Hiyoshi et al. 2005), 1:2000 diluted anti-INCENP rabbit antibody (Abcam), or 1:2000 diluted anti-GFP mouse monoclonal antibody (Clontech). To evaluate the amount of the sample applied, anti-actin mouse monoclonal antibody (1:2000 dilution; Millipore) was also used. Following treatment of the sheets with alkaline phosphatase-conjugated anti-mouse-IgG antibody (1:2000 dilution; Invitrogen) or alkaline phosphatase-conjugated anti-rabbit-IgG antibody (1:2000 dilution; Chemicon International), signals were detected with NBT/BCIP (Roche) or CSPD (Roche), according to the methods of the supplier. The intensity of the signals was quantified by using Image J (http://rsbweb.nih.gov/ij/).
Immunoprecipitation and detection of Xtr-associated FRGY2 and mRNAs
Immunoprecipitation of Xtr from Xenopus oocytes was performed with the same method as previously reported (Mostafa et al. 2009). Briefly, lysate obtained by centrifuging the homogenate of 20 oocytes was incubated with anti-Xtr monoclonal antibody (Mostafa et al. 2009) or normal mouse IgG (NMI: UPC-10; Sigma) conjugated to Protein A Sepharose (GE Healthcare) for 30 min at 18°C, followed by elution of Xtr and Xtr-associated molecules with 50 μL of 0.1 mol/L glycine/HCl (pH 2.5). For detection of Xtr, FRGY2, and actin, 5 μL of the eluate was mixed with an equal volume of 2× Laemmli's sample buffer and subjected to Western blotting using anti-Xtr rabbit antiserum (1:1000 dilution), anti-FRGY2 rabbit antiserum (kindly provided by Dr K. Matsumoto; 1:3000 dilution), or anti-actin mouse monoclonal antibody (1:2000 dilution) as described above. For detection of the Xtr-associated mRNAs, following the addition of spermatogenic cell-specific SP4 mRNA (10 pg) to 10 μL of the eluates for evaluating the recovery of RNA and efficiency of RT–PCR, RNA in the eluates were purified by the AGPC method and subjected to RT–PCR. To detect the presence of XL-INCENP, RCC1, XMAD2, XRHAMM, and SP4 mRNAs in the eluates, the same gene-specific primer sets reported in Mostafa et al. (2009) were used.
Oocytes and mature eggs were placed in 70% Leibovitz L-15 and MMR, both of which contained 6% Ficoll (Sigma), respectively, and injected with affinity-purified antibody (Hiyoshi et al. 2005; 230 ng/46 nL into cytoplasm or 23 ng/4.6 nL into GV) in injection buffer (88 mmol/L NaCl, 1 mmol/L KCl, 15 mmol/L Tris–HCl; pH 7.4) or synthesized mRNA (2.3 ng/4.6 nL into cytoplasm) in H2O. When the demembranated sperm nuclei (approximately 50 nuclei prepared by the method of Ohsumi et al. (2006)) were injected into mature eggs, 10 mmol/L EGTA was added in the injection buffer for inhibiting the egg activation (Zhang & Masui 1992).
Immunohistochemical observations were performed according to the method of Wakabayashi and Shinagawa (2001) with modification. Briefly, 7 μm sections were successively treated with 1:500 diluted anti-tubulin antibody (clone DM1A; NeoMarker) in digoxygenin (DIG) II buffer (1.5% blocking reagent [Roche] in DIG I buffer [150 mmol/L NaCl, 100 mmol/L maleic acid; pH 7.5]), 1:500 diluted Alexa 568-conjugated anti-mouse IgG antibody (Invitrogen) in DIG II buffer, and 10 μg/mL of Hoechst (Nacalai tesque). After washing thoroughly with DIG I buffer, the sections were observed with a BZ-9000 fluorescence microscope (Keyence).
Histone H1 kinase assay
A frozen oocyte was homogenized on ice in 15 μL of the extraction buffer (80 mmol/L β-glycerophosphate, 20 mmol/L EGTA, 15 mmol/L MgCl2, 1 mmol/L dithiothreitol; pH 7.3) and centrifuged briefly. Five microliters of the supernatant was subjected to a histone H1 kinase assay (in the presence of [γ-32P] ATP), followed by SDS–PAGE, essentially as described by Furuno et al. (1994).
Data were obtained as the mean ± standard deviation. For statistical comparison, Student's t-test was used. P-values < 0.05 were considered to be statistically significant.
Effect of anti-Xtr antibody on meiotic progression
Previously, we prepared affinity-purified anti-Xtr antibody. Since this antibody recognized only Xtr in the ovary extract and the cleavage arrest of the blastomere of the two-cell stage embryos induced by injection of this antibody into it could be rescued by co-injection of Xtr-fusion protein that was used for production of this antibody as an antigen (Hiyoshi et al. 2005), this affinity-purified anti-Xtr antibody is judged to affect the Xtr function specifically. In the present study, we used this affinity-purified anti-Xtr antibody for inhibition of the Xtr function in the oocytes. To examine whether and how Xtr protein functions during meiotic progression, anti-Xtr antibody was injected into cytoplasm of oocytes and their maturation was induced by progesterone treatment. When the reinitiation of meiosis was evaluated by the appearance of a white spot at the animal pole, which indicates the occurrence of germinal vesicle break down (GVBD), all oocytes injected with anti-Xtr antibody began the GVBD in a similar way to those injected with normal rabbit IgG (NRI) (Fig. 1A). At 30 min after GVBD, the anti-Xtr antibody-injected oocytes as well as NRI-injected oocytes showed the chromosomes with fully developed spindles at meiotic metaphase I (MI) (Fig. 1B, upper panel, Table 1). When oocytes with chromosomes at meiotic metaphase II (MII) were observed at 150 min after GVBD, however, anti-Xtr antibody-injected oocytes did not show the microtubule around the chromosomes in contrast with the typical spindle formation in the NRI-injected oocytes (Fig. 1B, lower panel, Table 1). This lack of a spindle around the chromosomes in the anti-Xtr antibody-injected oocytes was also observed at 12 h after GVBD. These results suggest that the functional Xtr is indispensable for the spindle formation at MII. Oocyte maturation is well known to be controlled by Cdc2/cyclin B activity (Kishimoto 2003). Therefore, the effect of anti-Xtr antibody on the Cdc2/cyclin B activity was examined by Histone H1 kinase assay (Moreno & Nurse 1990). In both anti-Xtr antibody- and NRI-injected oocytes, the first peak of H1 kinase activity was observed at 3 h after progesterone treatment, at which time most oocytes showed the white spot (Fig. 1A). Following a brief decrease of the H1 kinase activity, it increased in both anti-Xtr antibody- and NRI-injected oocytes (Fig. 1C). This result clearly indicates that lack of spindle formation at MII in the anti-Xtr antibody-injected oocytes was not ascribable to the effect of the antibody on the oscillation of Cdc2/cyclin B activity during meiotic progression.
Table 1. Effect of anti-Xtr antibody injected into cytoplasm of oocytes on the spindle formation during oocyte maturation
Time after GVBD
Number of oocytes showing normal spindle around chromosomes/Number of oocytes examined (percentage of normal spindle formation).
When the demembranated sperm nuclei are injected into mature eggs under conditions inhibitory to egg activation by chelating the free calcium in the cytoplasm, the injected nuclei transform into chromosomes with spindles at MII (Zhang & Masui 1992). To examine whether Xtr protein acts directly on the spindle formation at MII, the demembranated sperm nuclei suspended in EGTA solution were injected into mature eggs with either anti-Xtr antibody or NRI. Ninety minutes after injection of sperm nuclei, they transformed into chromosomes and fully developed spindles were formed around them (Fig. 2A). On the other hand, injected sperm nuclei swelled a little but did not transform into chromosomes when they were injected into fully matured oocytes, which had been injected with anti-Xtr antibody before progesterone treatment (Fig. 2B). These results indicate that Xtr does not act directly on the spindle formation and that the lack of spindles around chromosomes at MII in anti-Xtr antibody-injected oocytes is ascribable to the inhibition of Xtr function in the process of oocyte maturation.
Injection of anti-Xtr antibody into the cytoplasm of oocytes followed by progesterone treatment prevented spindle formation at MII but not at MI, which initially indicated that Xtr function was unnecessary for progression through MI. However, since the germinal vesicle included a small amount of Xtr (Fig. 3A), the effect of anti-Xtr antibody on progression of MI was re-examined by its injection into both the oocyte cytoplasm and germinal vesicle. Thirty minutes after GVBD, at the timing of MI, fully developed spindles were observed around chromosomes in NRI-injected oocytes. In contrast, anti-Xtr atibody-injected oocytes did not show spindles around chromosomes and the spindles were not formed even 150 min after GVBD (Fig. 3B, Table 2). These results clearly show the necessity of Xtr for the spindle formation at both MI and MII.
Table 2. Effect of anti-Xtr antibody injected into both germinal vesicle and cytoplasm of oocytes on the spindle formation during oocyte maturation
Time after GVBD
Number of oocytes showing normal spindle around chromosomes/Number of oocytes examined (percentage of normal spindle formation).
Xtr in oocytes associated with mRNAs and regulated their translation
Our previous report showed that Xtr is a member of a messenger ribonucleoprotein (mRNP) in fertilized eggs associated with several kinds of proteins such as FRGY2 and maternal mRNAs including RCC1 and XL-INCENP mRNAs (Mostafa et al. 2009), of which translational products have been demonstrated to be involved in progression of karyokinesis (Ohtsubo et al. 1989; Yamamoto et al. 2008). To examine whether Xtr in oocytes also associates with FRGY2 and mRNAs, immunoprecipitation with anti-Xtr monoclonal antibody was performed. Western blot analyses clearly showed the association of Xtr with FRGY2 but not with actin in the oocytes (Fig. 4A,c), suggesting that the immunoprecipitate included only the molecules associating with Xtr. When gene specific primer sets were used for RT–PCR, XL-INCENP and other mRNAs, which had been shown to associate with Xtr in the fertilized eggs (Mostafa et al. 2009), were detected (Fig. 4B). These results indicate that Xtr in the oocytes is also a member of mRNP. Taken together with the lack of microtubule assembly around chromosomes in anti-Xtr antibody-injected oocytes, this result led us to expect that inhibition of spindle formation at MI and MII by the action of anti-Xtr antibody was due to the prevention of translation of these mRNAs by the inactivation of Xtr.
Translation of XL-INCENP mRNA has been shown to be repressed in oocytes and begin after reinitiation of oocyte maturation (Yamamoto et al. 2008). Therefore, the effect of anti-Xtr antibody on the translation of XL-INCENP mRNA was examined by its injection into both the cytoplasm and the germinal vesicle. In contrast to no effect of NRI on XL-INCENP translation, it was inhibited by injection of anti-Xtr antibody (Fig. 5). These results suggest that Xtr is indispensable for the translation of XL-INCENP mRNA.
The 3′UTR of XL-INCENP mRNA was necessary and sufficient for regulation of its translation by Xtr
In some mRNAs, their translation is well known to be regulated through their 3′UTR (de Moor et al. 2005). To determine whether translational regulation of XL-INCENP mRNA before and after reinitiation of oocyte maturation is mediated by its 3′UTR, recombinant reporter mRNAs with and without the 3′UTR of XL-INCENP mRNA were synthesized and their translational regulation were examined by their injection into oocytes. For the synthesis of the recombinant reporter mRNAs, EGFP ORF with or without XL-INCENP 3′UTR was inserted into pT7G(UK II+) vector, so that the mRNAs included β-globin 5′UTR and 3′UTR at upstream of EGFP ORF and downstream of XL-INCENP 3′UTR, respectively (Fig. 6A, cf. Okamoto et al. 2002). When EGFP mRNA (without XL-INCENP 3′UTR) was injected into oocytes followed by their incubation for 12 h, translational product from recombinant mRNA, EGFP, was detected in the oocytes by Western blot using anti-GFP antibody (Fig. 6B, without −). On the other hand, the oocytes injected with the EGFP/XL-INCENP 3′UTR mRNA and incubated for the same period showed a faint signal of EGFP (Fig. 6B, with −), and this amount was apparently smaller than that of EGFP in the oocytes injected with EGFP mRNA (Fig. 6C, 3′UTR+/PG− and 3′UTR−/PG−). Its amount increased after reinitiation of oocyte maturation by progesterone treatment (Fig. 6B, with +; 6C, 3′UTR+/PG+). These results suggest that the translation of exogenous EGFP/XL-INCENP 3′UTR mRNA was regulated like endogenous XL-INCENP mRNA. To examine the involvement of Xtr in the translational regulation of recombinant mRNA, oocytes injected with antibody into both the cytoplasm and germinal vesicle were injected with reporter mRNA. When the oocytes were injected with EGFP mRNA and incubated for 2 h, the NRI- and anti-Xtr antibody-injected oocytes expressed EGFP and its amount increased after progesterone treatment (Fig. 6D, without; 6E, closed circles and closed triangles). On the other hand, the oocytes injected with NRI and EGFP/XL-INCENP 3′UTR mRNA did not show EGFP before progesterone treatment and it was detected soon after progesterone treatment (Fig. 6D, with/NRI; 6E, open circles). In contrast to the accumulation of EGFP after progesterone treatment in NRI- and EGFP/XL-INCENP 3′UTR mRNA-injected oocytes, translation of EGFP/XL-INCENP 3′UTR mRNA after progesterone treatment was inhibited when the oocytes had been injected with anti-Xtr antibody instead of NRI (Fig. 6D, with/α Xtr; 6E, open triangles; 6F, 3′UTR+/α Xtr). This translational regulation of exogenous mRNA was consistent with that of endogenous XL-INCENP mRNA (Fig. 5). These results clearly indicate that 3′UTR of XL-INCENP mRNA is necessary and sufficient for its translational regulation and that Xtr plays an important role in this process.
In previous studies, we have shown the exclusive expression of the Xtr gene in germ line cells of Xenopus and the occurrence of Xtr in germ line cells as well as early embryonic cells as a maternal factor (Ikema et al. 2002; Hiyoshi et al. 2005). Loss-of-function of Xtr in fertilized eggs using anti-Xtr antibody caused the lack of chromosome condensation and microtubule assembly, resulting in cleavage arrest (Hiyoshi et al. 2005). Since Xtr is a member of mRNP complex associated with mRNAs encoding the proteins such as XL-INCENP and RCC1 (Mostafa et al. 2009), which play an important role in karyokinesis (Ohtsubo et al. 1989; Mackay et al. 1998; Adams et al. 2001), the inhibition of karyokinesis progression induced by ablation of Xtr function was expected to be ascribable to translational suppression of these mRNAs. In Xenopus spermatogenesis, the amount of Xtr increases immediately after spermatogenic cells enter into meiotic phase (Hiyoshi et al. 2005). Therefore, Xtr was also thought to be involved in meiotic control. In this study, we took advantage of microinjection of an anti-Xtr antibody into oocytes to determine the effect of Xtr inhibition during meiotic progression. Here we showed that inhibition of Xtr function in the oocytes caused prevention of microtubule assembly around scattered chromosomes at meiotic metaphase I and II (MI and MII) and translational suppression of XL-INCENP mRNA, of which translation is known to begin after reinitiation of oocyte maturation (Yamamoto et al. 2008). XL-INCENP is a chromosome passenger protein involved in metaphase chromosome alignment and chromosome segregation by cooperation with Aurora B (Mackay et al. 1993; Adams et al. 2001; Honda et al. 2003). Therefore, disalignment of chromosomes at MI and MII by inactivation of Xtr in oocytes may be due to a lack of translation of XL-INCENP mRNA. Since treatment of sperm nuclei with Xenopus egg extract from which XL-INCENP has been depleted induces the transformation of the sperm nuclei into chromosomes, XL-INCENP is thought to be unnecessary for chromosome condensation (MacCallum et al. 2002). However, sperm nuclei injected into fully matured oocytes, in which Xtr function had been inhibited by anti-Xtr antibody before progesterone treatment, did not transform into chromosomes (Fig. 2). This result shows that the Xtr function-depleted oocytes lacked some proteins as well as XL-INCENP that are indispensable for karyokinesis progression. Since Xtr was associated with several kinds of mRNAs, it is expected to regulate their translation as a member of mRNP complex.
When the XL-INCENP 3′UTR was inserted downstream of the EGFP ORF and the resulting mRNA was injected into fully-grown oocytes, EGFP translation was arrested until oocyte maturation was induced by progesterone treatment. In addition, translation of EGFP after progesterone treatment was inhibited by depletion of Xtr function. On the other hand, the control EGFP mRNA lacking the XL-INCENP 3′UTR was translated regardless of depletion of Xtr function. These results clearly indicate that translational arrest of XL-INCENP mRNA in immature oocytes and the beginning of its translation after reinitiation of oocyte maturation, which are controlled by Xtr, are mediated by its 3′UTR. Translational arrest and activation of some dormant maternal mRNAs such as cyclin B mRNA before and after reinitiation of oocyte maturation are well known to be regulated by the length of poly A tails at their 3′ termini (Méndez & Richter 2001; Richter 2007). This regulation is mediated by a cytoplasmic polyadenylation element (CPE; consensus U4-6 A1-3 U) in their 3′UTRs. The CPE recruits CPE-binding protein (CPEB), which makes a complex with PARN, xGld2, Symplekin, Maskin, and Pumilio (Stebbins-Boaz et al. 1999; Nakahata et al. 2003; Barnard et al. 2004; Kim & Richter 2006; Padmanabhan & Richter 2006; Ota et al. 2011) and the length of the poly A tail is kept short. After reinitiation of oocyte maturation, dissociation of Pumilio from CPEB and phosphorylation of CPEB by Aurora A trigger the polyadenylation of these mRNAs, resulting in the beginning of translation (Sarkissian et al. 2004; Padmanabhan & Richter 2006; Ota et al. 2011). In this translational regulation, the CPE at the 3′UTR of the mRNA is indispensable. XL-INCENP mRNA does not contain a consensus CRE sequence in its 3′UTR. Even if Xtr function was inhibited by injection of anti-Xtr antibody into oocytes, followed by progesterone treatment, GVBD and oscillation of H1 kinase activity were found to occur normally. These results indicate that inhibition of Xtr function has no effect on the synthesis of key proteins for oocyte maturation such as Mos and cyclin B of which translation is regulated by CPEB (Stebbins-Boaz et al. 1996). Thus, the translational regulation of transcripts by Xtr is expected to be by a novel mechanism distinguishable from CPEB-mediated regulation of translation.
When anti-Xtr antibody was injected into cytoplasm, spindle formation and chromosome alignment became abnormal at MII but not MI. On the other hand, abnormal progression of MI was observed by injection of the antibody into both the GV and cytoplasm. At present, the reason why injection of anti-Xtr antibody into GV as well as cytoplasm is necessary for inhibition of MI-progression remains unclear. CPEB, which like Xtr is not abundant in the GV, has been demonstrated to shuttle between the cytoplasm and GV and interact with lampbrush chromosomes and CPE-containing mRNA such as cyclin B1 and mos mRNAs in the GV (Lin et al. 2010). On the other hand, CPEB and CPE-containing mRNAs encoding proteins that play a crucial role in spindle assembly and chromosome segregation localize on the spindles and chromosomes at MI and disruption of their localization by injection of excess amounts of competitor RNA containing multiple CPE sequences into oocytes has been shown to cause abnormal spindle formation and chromosome alignment (Eliscovich et al. 2008). INCENP mRNA is known to localize to spindles at mitotic metaphase in XL177 cells (Sharp et al. 2011). Therefore, the transport of the mRNAs encoding the proteins indispensable for spindle formation and chromosome segregation into the GV and their localization on the spindle and/or chromosomes seems to be prerequisite for their translation. Since Xtr possesses a putative nuclear localization signal at amino acids 1432 to 1465 (cNLS Mapper; http://nls-mapper.iab.keio.ac.jp/cgi-bin/NLS_Mapper_form.cgi), Xtr may transport some mRNAs such as XL-INCENP mRNA into the GV and regulate their translation.
Xtr is present in germ line cells and early embryonic cells but not in destined somatic cells. On the other hand, XL-INCENP is an indispensable protein for cell division, so dividing somatic cells express this protein. In observation of Xenopus A6 cells by RT–PCR, which are derived from an adult male kidney, XL-INCENP mRNA but not Xtr mRNA was found (data not shown). This observation indicates that translation of XL-INCENP mRNA is regulated in a different manner between germ line cells/early embryonic cells and destined somatic cells. Elucidating which mRNAs are translationally regulated by Xtr, how Xtr regulates it in germ line and early embryonic cells, and why Xtr must disappear from destined somatic cells should lead us to understand the characteristic of germ line and early embryonic cells having pluripotency.
We thank Dr Ken Matsumoto (RIKEN, Japan) for providing us with anti-FRGY2 rabbit antiserum, Dr William J. Ratzan (University of Connecticut, USA) for his critical proofreading of this manuscript, and Mr Tomokazu Yamaguchi and Mr Yoshihiro Inoue (Kumamoto University, Japan) for their technical support. This work was supported in part by Grants-in-aid-for-scientific-research incentive from Kumamoto University of Japan (KT).