Journal of Reproduction and Development, Vol. 53, No. 4, 2007

—Full Paper—

H1foo is Indispensable for Meiotic Maturation of the Mouse Oocyte

Masataka FURUYA1), Mamoru TANAKA1), Takahide TERANISHI2), Kazuya MATSUMOTO3), Yoshihiko HOSOI3), Kazuhiro SAEKI3), Hitoshi ISHIMOTO1), Kazuhiro MINEGISHI1), Akira IRITANI3) and Yasunori YOSHIMURA1)

1)Department of Obstetrics and Gynecology, Keio University School of Medicine, Tokyo 160- 8582, 2)Department of Obstetrics and Gynecology, Ota General Hospital, Ota 373-8585 and 3)Division of Biological Science, Graduate School of Biology-Oriented Science and Technology, Kinki University, Wakayama 649-6493, Japan

Abstract. Oocyte-specific linker H1foo is localized in the oocyte nucleus, either diffusely or bound to chromatin, during the processes of meiotic maturation and fertilization. This expression pattern suggests that H1foo plays a key role in the control of expression and chromatin modification during oogenesis and early embryogenesis. To reveal the function of H1foo, we microinjected antisense morpholino oligonucleotides (MO) against H1foo into mouse germinal- vesicle stage oocytes. The rate of in vitro maturation of the antisense MO group was significantly lower than that of the control group. Eggs that failed to extrude a first polar body following injection of antisense MO arrested at metaphase I. Additionally, co-injection of in vitro synthesized H1foo mRNA along with antisense MO successfully rescued expression of H1foo and improved the in vitro maturation rate. There was no difference in the rate of parthenogenesis between the antisense MO and control groups. These results indicate that H1foo is essential for maturation of germinal vesicle- stage oocytes. Key words: H1foo, Linker histone, Meiotic maturation, Morpholino, Mouse, Oocyte (J. Reprod. Dev. 53: 895–902, 2007)

n , the basic chromatin unit is more condensed, higher order forms of chromatin. commonly defined as the nucleosome, which is Mammals express eight H1 subtypes, including composed of DNA wrapped in two turns around H1fa through H1fe and H1f0, which are found in an octamer of core histone [1]. Association somatic cells in addition to germ cell-specific H1ft of DNA with the histone octamer to form and H1foo [2–4]. Expression of each mammalian nucleosomes is the first level of chromatin H1 subtype depends on the tissue, phase of the cell condensation and results in a configuration cycle and developmental stage. Knockout of commonly referred to as “beads on a string”. The individual H1 subtype in mice is DNA between the beads is referred to as linker compensated for by increased expression of the DNA. Linker , termed H1, bind to linker remaining subtypes [5]. However, simultaneous DNA and facilitate a shift in equilibrium towards inactivation of multiple H1 subtype genes causes embryonic death and therefore linker histones are Accepted for publication: April 15, 2007 essential for mammalian development [6]. Published online: May 23, 2007 H1foo is a linker histone specifically expressed in Correspondence: M. Tanaka (e-mail: [email protected]) murine oocytes [4]. Oocyte-specific linker histones 896 FURUYA et al. have been found across a wide range of species, Materials and Methods including the sea urchin, frog, zebrafish, mouse and human, and they are very well conserved in all Morpholino antisense oligonucleotides species. We have previously shown that mouse Morpholino oligonucleotides (MO) were made to H1foo is localized to the nucleus of oocytes at the order (Gene Tools; LLC, Corvallis, OR, USA). The germinal vesicle (GV) stage, the chromatin during complementary MO for H1foo was designed to metaphase-II (MII), and the first polar body. After target a sequence containing the initiation codon. fertilization, H1foo is also detected in the swollen The sequence of this MO was 5’-ACTGGAGAC- sperm head and second polar body. Expression of ACTCCCAGGAGCCATG-3’. Another MO with H1foo is somewhat reduced in two-cell embryos the inverse of the antisense sequence, 5’- and is not detectable in four-cell embryos. In GTACCGAGGACCCTCACAGAGGTCA-3’, was Xenopus, oocyte-specific linker histone B4 is used as a control for nonspecific MO toxicity. Stock replaced by H1 during the midblastula transition solutions of the antisense MO and control MO were [7, 8]. The expression pattern of H1foo in prepared in sterile water. The MO stock solutions preimplantation embryos is similar to that of were diluted to a concentration of 500 µM and Xenopus B4. Furthermore, replacement of the heated to 65 C for 5 min prior to use. somatic linker histone with H1foo occurs rapidly during nuclear transfer in the mouse [9, 10]. The Oocyte collection and microinjection of MOs greater mobility of H1foo compared with H1 is B6D2F1 female mice, age 8 to 12 weeks from thought to contribute to this rapid replacement, Sankyo Labo Service (Tokyo, Japan), were injected which provides flexibility for the chromatin with 5 IU of eCG (Teikokuzoki, Tokyo, Japan) and structure. These findings suggest that oocyte- sacrificed 48 h after eCG administration. Fully specific linker histones play a key role in control of grown, GV-intact oocytes were obtained by gene expression and chromatin modification puncture of antral follicles in M2 medium (Sigma- during oogenesis, early embryogenesis and nuclear Aldrich, St. Louis, MO, USA) containing 240 µM remodeling. However, the function of H1foo dibutyryl cyclic AMP (dbcAMP; Sigma). After the remains to be elucidated. oocytes were collected, the surrounding cumulus Reverse genetic approaches have become cells were mechanically removed using a glass increasingly attractive in developmental biology pipette. The denuded oocytes were microinjected because they represent an efficient means of with 10 pl of the MO solutions in M2 medium identifying the roles of interesting candidate genes containing 240 µM dbcAMP at room temperature from genomic databases. Recently, one strategy for using Narishige manipulators and an IM-300 inhibiting gene function was shown to be extremely Microinjector (Narishige, Tokyo, Japan). Following effective. Antisense morpholino oligonucleotides microinjection, the injected oocytes were incubated (MO) were first developed for clinical therapeutic for 10 min in M2 medium containing 240 µM applications, an area in which previous antisense dbcAMP at room temperature. Oocytes that were approaches have proven to be seriously flawed judged to be morphologically normal after [11]. They were first introduced into incubation were used for further experiments. developmental biology early in 2000 [12]. Antisense MOs block translation by interfering Culture and in vitro maturation of GV oocytes with the binding of ribosomes to target mRNA. For analysis, the injected and non- The use of antisense MOs has proven to be a injected GV oocytes were cultured in 100 µl valuable tool for investigating gene function in droplets of TYH medium (Mitsubishi, Tokyo, zebrafish and xenopus embryos [13]. However, Japan) supplemented with 5% heat inactivated fetal few studies have utilized this technology for the bovine serum (FBS; Invitrogen, Carlsbad, CA, USA) study of mammalian oocytes [14, 15]. In this work, and 240 µM dbcAMP under paraffin oil at 37 C in we investigated the feasibility of using antisense an atmosphere of 5% CO2 in humidified air. MOs to suppress mouse gene function and Twenty or 44 h after culture, oocytes were used for revealed the function of H1foo by microinjection of immunoblotting and immunofluorescence staining. an antisense MO against H1foo into mouse GV- For analysis of meiotic maturation, injected and stage oocytes. non-injected GV oocytes were cultured in 100 µl H1FOO IS INDISPENSABLE FOR IVM 897 droplets of TYH medium containing 5% FBS diluted 1/1000 in dilution buffer (0.5% TritonX-100 without dbcAMP. Twenty hours after culture, the in PBS). Following 3 washes in PBS with 0.1% rate of meiotic maturation was calculated based on TritonX-100, the oocytes were incubated with Alexa the presence or absence of a first polar body. Fluor 488 goat anti-rabbit IgG conjugate and Alexa Fluor 546 goat anti-mouse IgG conjugate for 1 h at Parthenogenesis of oocytes room temperature. The oocytes were washed again The injected and non-injected oocytes that and transferred into a drop of Vectashield matured in vitro were selected for evaluation of mounting medium with DAPI (Vector developmental capacity. After 3 washes in Ca2+- Laboratories, Burlingame, CA, USA) and covered free KSOM medium, the eggs were transferred into with a glass slip. The samples were observed with 2+ Ca -free KSOM medium containing 10 mM SrCl2 an AX-70 fluorescence microscope (Olympus, and activated for one hour. Following further Tokyo, Japan), and images were captured using a culture in KSOM medium for 5 h, the rate of SPOT RT Monochrome Digital Camera (Diagnostic parthenogenesis was calculated by formation of Instruments, Sterling Heights, MI, USA). pronuclei. Histone H1 kinase activity Immunoblotting of H1foo and cyclin B1 Histone H1 kinase activity was determined in Twenty-five oocytes from each group were HK buffer comprised of 80 mM β- washed in PBS, extracted in 4 µl of Laemmli sample glycerophosphate, 20 mM EGTA (pH 7.3), 15 mM buffer (Bio-RAD Laboratories, Hercules, CA, USA) MgCl2 , 1 mM dithiothreitol (DTT), 1 mM with 5% β-mercaptoethanol and boiled for 5 phenylmethylsulfonyl fluoride (PMSF), 10 µg/ml minutes. The extracts were resolved on 12% SDS- leupeptin, 10 µg/ml pepstatin, 10 µg/ml aprotinin PAGE gels and blotted onto a polyvinylidene using exogenous histone H1 (Sigma) as described fluoride membrane. The membrane was blocked previously [16]. Samples containing oocytes in M2 with 5% non-fat milk in TTBS (10 mM Tris-HCl, 150 medium were lysed by freezing and thawing three mM NaCl, 0.05% Tween20) for 1 h and incubated times, diluted twice in 2X concentrated HK buffer with 1 µg/ml of rabbit anti-H1foo polyclonal and incubated for 30 min at 37 C in the presence of antibody (the kind gift of Dr. E. Y. Adashi) or 3.3 mg/ml histone H1, 1 mM ATP and 30 µCi/ml monoclonal mouse anti-cyclin B1 IgG [γ- 32P]ATP. The reaction was stopped by (Dakocytomation, Carpinteria, CA, USA) at room incubation for 2 min at 90 C immediately following temperature for 1 h. After washing in TTBS, the the addition of concentrated sample buffer. The membrane was incubated with horseradish samples were then electrophoresed on 12% SDS- peroxidase-conjugated goat anti-rabbit or anti- polyacrylamide gel. After drying, the gels were mouse secondary antibody (Santa Cruz exposed to x-ray film at –80 C. Biotechnology, Santa Cruz, CA, USA) diluted 1/ 5,000 in 5% non-fat milk/TTBS for 1 h at room In vitro synthesis and microinjection of H1foo-GFP temperature. The signal was detected using the mRNA ECL system (Amersham Pharmacia, Piscataway, The constructed plasmid containing cDNA NJ, USA). encoding H1foo and EGFP without the MO recognition sequence was used as a template for in Immunofluorescence staining of H1foo, microtubules vitro transcription. RNA was synthesized using an and nuclei SP6 Message Machine Kit (Ambion, Austin, TX, Oocytes were fixed for 20 min at room USA). temperature in 2% paraformaldehyde in PBS (pH Synthesized RNA was polyadenylated with a 7.4). After blocking for 1 h at room temperature in Poly(A) Tailing Kit (Ambion). To avoid folding of blocking solution (10% goat serum and 0.5% the RNA, the polyadenylated RNA was heated at TritonX-100 in PBS), the oocytes were incubated for 90 C for 1 min and cooled on ice. A mixture of MO 1 h at room temperature with rabbit anti-H1foo and the RNA (final concentration 1 µg/µl) was polyclonal antibody at a concentration of 1 µg/ml used for microinjection. and mouse anti-α-tubulin monoclonal IgG (Sigma) 898 FURUYA et al.

Fig. 2. Depletion of H1foo suppressed the rate of IVM. The rate of IVM was calculated by scoring eggs for presence/absence of a first polar body. Only 33.1% of H1foo-depleted oocytes successfully extruded a first polar body and progressed into meiosis II, compared with 81.1% of the control MO treatment group oocytes and 89.0% of the non-injected oocytes. Bars represent means ± the standard deviation of five independent experiments. Student’s t-test revealed significant differences (*: P<0.05). Fig. 1. Microinjection of antisense MO successfully depleted H1foo protein in mouse oocytes. (A) GV oocytes were stained with anti-H1foo antibody (green) and DAPI (blue) 20 h (upper) and 44 h (lower) after MO, and after 44 h the signal was no longer microinjection of antisense MO (left) or control MO detectable (Fig. 1A). To quantitatively compare the (right). Expression of H1foo protein in oocytes injected with antisene MO was remarkably reduced 20 expression of H1foo protein, immunoblotting h after injection and was absent by 44 h after injection analysis was performed using lysates from 50 compared with oocytes injected with a control MO. oocytes for each lane. A significant reduction in (B) Oocytes collected 20 h (upper) and 44 h (lower) H1foo expression was observed at 20 h and 44 h after microinjection were analyzed by immunoblotting using either an anti-H1foo antibody after injection in the antisense MO treatment group (left) or anti-α-tubulin antibody (right). Anti-α- compared with the control MO group (Fig. 1B). tubulin antibodies were used as controls. These results demonstrated that the antisense MO successfully produced a specific reduction of H1foo in the mouse oocytes. Results Knockdown of H1foo impairs meiotic maturation Antisense MO injection completely abolishes H1foo H1foo transcript and protein is expressed in GV- expression in GV oocytes and metaphase II-stage oocytes and one-cell-stage Immunocytostaining and immunoblotting were embryos. However, H1foo expression is mostly performed to test whether microinjection of extinguished by the two-cell stage. We antisense MO into GV-stage oocytes suppressed investigated whether knockdown of H1foo by the expression of H1foo protein. After injection of antisense MO impaired oocyte maturation and MO, oocytes were cultured in the presence of development. Following culture for 20 h after dbcAMP for inhibition of GV breakdown (GVBD). microinjection of antisense or control MO, the rate Expression of H1foo was compared among oocytes of in vitro maturation (IVM) was calculated by that maintained the germinal vesicle. Twenty counting the number of eggs that extruded a first hours after microinjection, immunocytostaining polar body. In five separate experiments, only 33.1 showed that the expression of H1foo protein in the ± 12.1% of H1foo-depleted oocytes successfully oocytes injected with antisene MO was markedly extruded a first polar body and progressed into reduced compared with those injected with control meiosis II compared with 81.1 ± 7.0% of the control H1FOO IS INDISPENSABLE FOR IVM 899

Fig. 4. Co-injection of H1foo mRNA with H1foo antisense MO largely restored the IVM rate. (A) The H1foo mRNA construct lacks the MO target sequence and includes the SP6 promoter for in vitro transcription and an EGFP sequence at the 3’ end for recognition of expression and localization in oocytes. In vitro transcription and poly(A) addition were performed using this construct. (B) Fluorescent microscopic images at 4 h (upper) and 18 h (lower) after co- injection of antisense MO and transcribed RNA. Fig. 3. Analysis of oocytes that failed IVM after injection of Note that H1foo-EGFP protein could be clearly antisense MO. (A) Immunofluorescence staining visualized. (C) The rate of IVM was markedly higher with anti-α-tubulin antibody (red), DAPI (blue) and in oocytes co-injected with antisense MO and RNA anti-H1foo antibody (green). The oocyte is arrested compared with oocytes injected with antisense MO. at metaphase I. No morphological defects were The bars represent means ± the standard deviation of observed in spindle formation or four independent experiments. (3) Student’s t-tests condensation. (B) Results of immunoblotting for revealed significant differences (*: P<0.05). cyclin B1 (upper) and H1 kinase assays (lower) performed on lysates from arrested oocytes and normal oocytes at the GV stage, at 8 h after GVBD with or without polar bodies and at metaphase II. H1foo-depleted oocytes arrest at metaphase I Expression of cyclin B1 was maintained, but the H1 Eggs with failed meiotic maturation following kinase assay showed low MPF activity in H1foo- depleted oocytes. Based on this discrepancy between injection of antisense MO were fixed and cyclin B1 expression and MPF activity, suppression immunocytostained for further evaluation. of H1foo may prevent the activation of MPF. Immunocytostaining of α-tubulin and nuclear staining revealed that treatment with antisense MO MO treatment group oocytes and 89.0 ± 5.3% of the caused the eggs to arrest at metaphase I (Fig. 3A). non-injected oocytes (Fig. 2). Eggs that matured However, depletion of H1foo had no effect on after microinjection of MO were selected and chromatin condensation or spindle organization activated by strontium chloride to trigger (Fig. 3A). Considering that no morphological parthenogenesis in order to evaluate the defects were observed in the metaphase I-arrested developmental capacity of H1foo-depleted oocytes. H1foo-depleted oocytes, the observed decreased The rate of parthenogenesis, as calculated by the IVM rate could be the result of functional defects in formation of pronuclei, was 88.9% (24/27) for the other factors that would be necessary for antisense MO group, 93.8% (15/16) for the control progression of meiosis in the absence of H1foo. MO group and 89.9% (71/79) for the non-injection Maturation promoting factor (MPF), which is a group. These results suggest that H1foo plays an complex of cdc2 and a B-type cyclin, is responsible important role during first meiosis but not in th e for driving oocytes through meiotic maturation. We formation of pronuclei. evaluated the activity of MPF and the expression of 900 FURUYA et al. cyclin B1 in eggs that arrested at metaphase I due to on GVBD, but did impair extrusion of the first polar injection of antisense MO. Cyclin B1 was body leading to a decrease in the IVM rate. To undetectable and MPF activity was low in the confirm the specificity of the antisense MO, we control prophase oocytes (Fig. 3B). Expression of rescued this phenotype by co-injection of H1foo cyclin B1 began juct before Gon and MPF was mRNA lacking the MO recognition sequence, activated until metaphase I (Fig. 3B). Subsequently, leading to improvement of IVM. These results MPF activity fell during the metaphase I and show that oocyte-specific linker histone H1foo is anaphase I transition due to temporal degradation involved in the progression of meiosis. of cyclin B1 and then rose again leading to entry into Linker histones are known to be involved mainly meiosis II with concomitant accumulation of cyclin in chromatin organization, with ensuing B1 (Fig. 3B). In eggs injected with antisense MO, consequences on transcription. However, the cyclin B1 expression was maintained, but MPF biological functions of linker histones are more activity remained low as revealed by the H1 kinase varied than previously proposed. Recently, some assay (Fig. 3B). This discrepancy between cyclin B1 reports have indicated that linker histones may expression and MPF activity suggests that play a role in aging, DNA repair and apoptosis [17– suppression of H1foo might prevent the activation 19]. In tobacco, deficiency of the linker histone of MPF during first meiosis after GVBD or induce subtypes H1A and H1B is linked with male sterility inactivation of MPF without degradation of cyclin resulting from disturbances in correct pairing or B1. segregation of homologous during meiosis [20]. Furthermore, in mitosis, Exogenous H1foo rescues the phenotype produced by overexpression of H1f0 leads to transient inhibition the H1foo antisense MO of both G1 and S phase progression [21], and To determine the specificity of the effect of the immunodepletion of histone H1 resulted in a H1foo antisense MO, we attempted to rescue lengthening of chromosomes that prevents their knockdown of H1foo by the simultaneous injection proper alignment and segregation [22]. Our results of exogenous H1foo mRNA. The in vitro also indicate a relationship between linker histones synthesized mRNA lacked the first 24 nucleotides and the cell cycle. Eggs lacking H1foo arrested at of the H1foo ORF (so that it would not be targeted metaphase I despite exhibiting normal compaction, by the antisense MO) and included a sequence of alignment of chromosomes and spindle formation. enhanced green fluorescent protein (EGFP) There is a possibility that H1foo may play a conjugated at the 3’ end to aid in recognition of functional rather than structural role in the expression and localization of exogenous H1foo segregation of chromosomes. (Fig. 4A). The H1foo-EGFP mRNA was efficiently MPF is a heterodimer consisting of a kinase, cdc2, translated into protein that was detectable roughly and its regulatory partner, cyclin B1 [23, 24]. 4 h after injection and was localized on chromatin Increased MPF activity leads to GVBD and entry during meiotic maturation in a similar manner to into the first meiotic metaphase [25, 26]. After endogenous H1foo (Fig. 4B). When H1foo-EGFP metaphase I, segregation of chromosomes requires mRNA was injected along with antisense MO, a transient decline of MPF activity resulting from significant rescue of the IVM rate was observed anaphase-promoting complex (APC)-mediated (Fig. 4C). Thus, the phenotype induced by the destruction of cyclin B1 [27]. The MPF activity of H1foo antisense MO was purely attributable to its the injected oocytes could not be maintained for designed activity and was not attributable to a non- enough time before the MI-anaphase I transition. It specific secondary effect. is possible that this leads to arrest at the MI stage. The H1foo-depleted oocytes exhibited low MPF activity despite the persistent presence of Discussion reasonable amounts of cyclin B1. This suggests that H1foo may be essential for maintenance of MPF In the present study, we demonstrated that activity by cyclin B1 and for destruction of cyclin B1 antisense morpholinos could specifically block the by APC. It is notable that the phenotype resulting translation of H1foo in mouse oocytes. from injection of H1foo antisense MO bears a close Suppression of H1foo did not have any influence resemblance to that produced by injection of c-Mos H1FOO IS INDISPENSABLE FOR IVM 901 antisense oligonucleotides. Following about 10 µM, which is a typical dose used for the microinjection of c-Mos antisense oligonucleotides, microinjection of zebrafish and Xenopus oocytes. oocytes undergo GVBD but are blocked at or about Although there are few published reports on use of the first meiosis [28]. The c-Mos protein kinase is this or similar technology for mouse oocytes, this an efficient activator of mitogen-activated protein concentration matched well with those reports. In kinase (MAPK) kinase that in turn activates MAPK addition, we prepared a control MO with the [29, 30]. During each phase of meiotic maturation, inverse sequence of the antisense MO and injected a dynamic cross-talk exists between MPF activity it at the same concentration in order to rule out and the c-Mos/MAPK pathway [31]. Thus, nonspecific toxic side effects. Since there was no reduced MPF activity in H1foo-depleted oocytes difference between the control MO group and the may interfere with the c-Mos/MAPK pathway. noninjected group with regard to the expression of Moreover, antibody array assays have suggested H1foo, the dose was appeared to be appropriate for that H1foo may interact with many kinds of cell analyzing specific knockdown of H1foo. This cycle-related proteins including CDC25, an study clearly establishes the usefulness of antisense activator of MPF (data not shown). Recent studies MOs for investigating gene function in the mouse. of fluorescence recovery after photobleaching Herein, we have successfully knocked down the (FRAP) revealed that H1foo has high mobility [10, expression of oocyte-specific linker histone H1foo 32], potentially facilitating effective recruitment of in mouse oocytes using a specific antisense MO and cell cycle-related factors on chromatin. It is revealed the critical role played by this protein in reasonable to speculate that H1foo has greater oocyte maturation. Further studies are necessary to mobility than somatic linker histones within fully analyze the relationship between H1foo and oocytes, which have significantly more cytosol the cell cycle machinery. space than somatic cells. Our results are consistent with H1foo having a role in the regulation of MPF, although it is still unknown whether H1foo directly Acknowledgements interacts with the c-Mos/MAPK pathway. In this report, we demonstrated that translation We would like to thank E. Y. Adashi for H1foo of H1foo could effectively be blocked in mouse antiserum and T. Kono, Y. Obata, S. Miyazaki and oocytes via microinjection of a specific MO. It is F. Aoki for technical assistance. This work was important to note that use of antisense technology supported in part by Grant-in-Aid for Scientific such as MOs requires care to avoid serious Research (14571584, 16591684 and 18390452) from nonspecific toxic side effects. Particularly, the the Japan Society for the Promotion of Science and concentration of antisense MO to be used requires by Keio Gijuku Academic Development fund 2001 serious consideration. In our experiments, we to M. T. injected the antisense MO to a final concentration of

References

1. Luger K, Mader AW, Richmond RK, Sargent DF, gene H1oo: homology with the genes for the oocyte- Richmond TJ. Crystal structure of the nucleosome specific cleavage stage histone (cs-H1) of sea urchin core particle at 2.8 A resolution. Nature 1997; 389: and the B4/H1M histone of the frog. Development 251–260. 2001; 128: 655–664. 2. Doenecke D, Albig W, Bode C, Drabent B, Franke 5. Fan Y, Sirotkin A, Russell RG, Ayala J, Skoultchi K, Gavenis K, Witt O. Histones: genetic diversity AI. Individual somatic H1 subtypes are dispensable and tissue-specific gene expression. Histochem Cell for mouse development even in mice lacking the Biol 1997; 107: 1–10. H1(0) replacement subtype. Mol Cell Biol 2001; 21: 3. Seyedin SM, Cole RD, Kistler WS. H1 histones 7933–7943. from mammalian testes. The widespread occurrence 6. Fan Y, Nikitina T, Morin-Kensicki EM, Zhao J, of H1t. Exp Cell Res 1981; 136: 399–405. Magnuson TR, Woodcock CL, Skoultchi AI. H1 4. Tanaka M, Hennebold JD, Macfarlane J, Adashi linker histones are essential for mouse development EY. A mammalian oocyte-specific linker histone and affect nucleosome spacing in vivo. Mol Cell Biol 902 FURUYA et al.

2003; 23: 4559–4572. Jerzmanowski A. Linker histones play a role in 7. Smith RC, Dworkin-Rastl E, Dworkin MB. male meiosis and the development of pollen grains Expression of a histone H1-like protein is restricted in tobacco. Plant Cell 1999; 11: 2317–2329. to early Xenopus development. Genes Dev 1988; 2: 21. Brown DT, Alexander BT, Sittman DB. Differential 1284–1295. effect of H1 variant overexpression on cell cycle 8. Dworkin-Rastl E, Kandolf H, Smith RC. The progression and gene expression. Nucleic Acids Res maternal histone H1 variant, H1M (B4 protein), is 1996; 24: 486–493. the predominant H1 histone in Xenopus pregastrula 22. Maresca TJ, Freedman BS, Heald R. Histone H1 is embryos. Dev Biol 1994; 161: 425–439. essential for mitotic chromosome architecture and 9. Gao S, Chung YG, Parseghian MH, King GJ, Adashi segregation in Xenopus laevis egg extracts. J Cell Biol EY, Latham KE. Rapid H1 linker histone transitions 2005; 169: 859–869. following fertilization or somatic cell nuclear transfer: 23. Gautier J, Minshull J, Lohka M, Glotzer M, Hunt evidence for a uniform developmental program in T, Maller JL. Cyclin is a component of maturation- mice. Dev Biol 2004; 266: 62–75. promoting factor from Xenopus. Cell 1990; 60: 487– 10. Teranishi T, Tanaka M, Kimoto S, Ono Y, 494. Miyakoshi K, Kono T, Yoshimura Y. Rapid 24. Labbe JC, Picard A, Peaucellier G, Cavadore JC, replacement of somatic linker histones with the Nurse P, Doree M. Purification of MPF from oocyte-specific linker histone H1foo in nuclear starfish: identification as the H1 histone kinase transfer. Dev Biol 2004; 266: 76–86. p34cdc2 and a possible mechanism for its periodic 11. Summerton J, Weller D. Morpholino antisense activation. Cell 1989; 57: 253–263. oligomers: design, preparation, and properties. 25. Choi T, Aoki F, Mori M, Yamashita M, Nagahama Antisense Nucleic Acid Drug Dev 1997; 7: 187–195. Y, Kohmoto K. Activation of p34cdc2 protein kinase 12. Heasman J, Kofron M, Wylie C. Beta-catenin activity in meiotic and mitotic cell cycles in mouse signaling activity dissected in the early Xenopus oocytes and embryos. Development 1991; 113: 789– embryo: a novel antisense approach. Dev Biol 2000; 795. 222: 124–134. 26. Hashimoto N, Kishimoto T. Regulation of meiotic 13. Heasman J. Morpholino oligos: making sense of metaphase by a cytoplasmic maturation-promoting antisense? Dev Biol 2002; 243: 209–214. factor during mouse oocyte maturation. Dev Biol 14. Coonrod SA, Bolling LC, Wright PW, Visconti PE, 1988; 126: 242–252. Herr JC. A morpholino phenocopy of the mouse 27. Herbert M, Levasseur M, Homer H, Yallop K, mos mutation. Genesis 2001; 30: 198–200. Murdoch A, McDougall A. Homologue disjunction 15. Lefebvre C, Terret ME, Djiane A, Rassinier P, in mouse oocytes requires proteolysis of securin and Maro B, Verlhac MH. Meiotic spindle stability cyclin B1. Nat Cell Biol 2003; 5: 1023–1025. depends on MAPK-interacting and spindle- 28. Paules RS, Buccione R, Moschel RC, Vande stabilizing protein (MISS), a new MAPK substrate. J Woude GF, Eppig JJ. Mouse Mos protooncogene Cell Biol 2002; 157: 603–613. product is present and functions during oogenesis. 16. Goudet G, Bezard J, Belin F, Duchamp G, Palmer Proc Natl Acad Sci USA 1989; 86: 5395–5399. E, Gerard N. Oocyte competence for in vitro 29. Nebreda AR, Hunt T. The c-mos proto-oncogene maturation is associated with histone H1 kinase protein kinase turns on and maintains the activity of activity and is influenced by estrous cycle stage in MAP kinase, but not MPF, in cell-free extracts of the mare. Biol Reprod 1998; 59: 456–462. Xenopus oocytes and eggs. Embo J 1993; 12: 1979– 17. Barra JL, Rhounim L, Rossignol JL, Faugeron G. 1986. Histone H1 is dispensable for methylation- 30. Posada J, Yew N, Ahn NG, Vande Woude GF, associated gene silencing in Ascobolus immersus Cooper JA. Mos stimulates MAP kinase in Xenopus and essential for long life span. Mol Cell Biol 2000; 20: oocytes and activates a MAP kinase kinase in vitro. 61–69. Mol Cell Biol 1993; 13: 2546–2553. 18. Downs JA, Kosmidou E, Morgan A, Jackson SP. 31. Frank-Vaillant M, Haccard O, Ozon R, Jessus C. Suppression of homologous recombination by the Interplay between Cdc2 kinase and the c-Mos/ Saccharomyces cerevisiae linker histone. Mol Cell MAPK pathway between metaphase I and 2003; 11: 1685–1692. metaphase II in Xenopus oocytes. Dev Biol 2001; 231: 19. Konishi A, Shimizu S, Hirota J, Takao T, Fan Y, 279–288. Matsuoka Y, Zhang L, Yoneda Y, Fujii Y, Skoultchi 32. Becker M, Becker A, Miyara F, Han Z, Kihara M, AI. Involvement of histone H1.2 in apoptosis Brown DT, Hager GL, Latham K, Adashi EY, induced by DNA double-strand breaks. Cell 2003; Misteli T. Differential in vivo binding dynamics of 114: 673–688. somatic and oocyte-specific linker histones in 20. Prymakowska-Bosak M, Przewloka MR, oocytes and during ES cell nuclear transfer. Mol Biol Slusarczyk J, Kuras M, Lichota J, Kilianczyk B, Cell 2005; 16: 3887–3895.