Rapid Replacement of Somatic Linker Histones with the Oocyte-Specific Linker Histone H1foo in Nuclear Transfer

Developmental Biology 266 (2004) 76–86 www.elsevier.com/locate/ydbio

Rapid replacement of somatic linker histones with the oocyte-specific linker histone H1foo in nuclear transfer

Takahide Teranishi,a,* Mamoru Tanaka,a Shingo Kimoto,b Yukiko Ono,b Kei Miyakoshi,a Tomohiro Kono,b and Yasunori Yoshimuraa

a Department of Obstetrics and Gynecology, Keio University School of Medicine, Shinjuku, Tokyo 160-8582, Japan b Department of BioScience, Tokyo University of Agriculture, Setagaya, Tokyo 156-8502, Japan Received for publication 26 May 2003, revised 22 September 2003, accepted 7 October 2003

Abstract

The most distinctive feature of oocyte-specific linker histones is the specific timing of their expression during embryonic development. In Xenopus nuclear transfer, somatic linker histones in the donor nucleus are replaced with oocyte-specific linker histone B4, leading to the involvement of oocyte-specific linker histones in nuclear reprogramming. We recently have discovered a mouse oocyte-specific linker histone, named H1foo, and demonstrated its expression pattern in normal preimplantation embryos. The present study was undertaken to determine whether the replacement of somatic linker histones with H1foo occurs during the process of mouse nuclear transfer. H1foo was detected in the donor nucleus soon after transplantation. Thereafter, H1foo was restricted to the chromatin in up to two-cell stage embryos. After fusion of an oocyte with a cell expressing GFP (green fluorescent protein)-tagged somatic linker histone H1c, immediate release of H1c in the donor nucleus was observed. In addition, we used fluorescence recovery after photobleaching (FRAP), and found that H1foo is more mobile than H1c in living cells. The greater mobility of H1foo may contribute to its rapid replacement and decreased stability of the embryonic chromatin structure. These results suggest that rapid replacement of H1c with H1foo may play an important role in nuclear remodeling. D 2003 Elsevier Inc. All rights reserved.

Keywords: Linker histone; H1foo; Nuclear remodeling; Fluorescence recovery after photobleaching (FRAP)

Introduction Shi et al., 2003; Wade and Kikyo, 2002). Accumulating evidence shows that the initial transcriptional activity of the Successful generation of cloned animals by nuclear donor cell nucleus is controlled predominantly by the egg transfer of somatic cells has been reported in several cytoplasm so that appropriate chromatin remodeling occurs species of mammals (Wakayama et al., 1998; Wilmut et (Reik et al., 2001; Rideout et al., 2001). Epigenetic al., 1997). These reports proved that somatic nuclei could modification of the genome ensures proper gene activation reverse their differentiated state to recover totipotency during development and involves (i) genomic methylation when introduced into oocyte cytoplasm. Transferred nuclei changes, (ii) the assembly of histones and histone variants change their gene expression pattern to that of early into nucleosomes, and (iii) remodeling of other chromatin- embryonic nuclei to permit successful development. This associated proteins, such as polycomb group proteins, change is generally termed reprogramming, and many nuclear scaffold proteins, and transcription factors (Latham, researchers have tried to elucidate the mechanism by which 1999). Whereas DNA methylation has been extensively nuclear reprogramming occurs (Kikyo and Wolffe, 2000; studied in mammalian development (Reik et al., 2001; Rideout et al., 2001), the molecular mechanisms involved in remodeling of chromatin-associated proteins remains to * Corresponding author. Department of Obstetrics and Gynecology, be elucidated. Keio University School of Medicine, 35 Shinanomachi, Shinjuku, Tokyo 160-8582, Japan. Fax: +81-3-3226-1667. DNA is tightly assembled with histone and nonhistone E-mail address: [email protected] (T. Teranishi). proteins to form chromatin in eukaryotic cells. Compaction

0012-1606/$ - see front matter D 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.ydbio.2003.10.004 T. Teranishi et al. / Developmental Biology 266 (2004) 76–86 77 of DNA into chromatin affects a large number of essential nuclear processes, including DNA replication, repair, re- combination, and transcription. The fundamental subunit of chromatin is the nucleosome, which contains 146 bp of DNA wrapped around an octamer core composed of two each of the histone proteins H2A, H2B, H3, and H4 (Wolffe, 1995). Because it links adjacent nucleosomes, H1 is often referred to as a linker histone. H1 has many subtypes, with mammalian somatic cells having six subtypes (H1a, H1b, H1c, H1d, H1e, and H1j).Thereasonwhysomany subtypes of linker histones exist remains uncertain, but the expression level of each subtype seems to be associated with differentiation (Zlatanova and Doenecke, 1994). According to studies knocking out a subtype(s) of linker histone, no one subtype is essential as members can compensate for each other’s absence (Fan et al., 2001). The association of histone H1 with DNA may stabilize the interaction between the core histone octamer and DNA, and facilitate assembly of the nucleosome array into a higher order structure (Felsenfeld and McGhee, 1986; Thoma et al., 1979).Itis currently accepted that H1 plays a regulatory role in transcription through modulation of higher structure of the chromatin. Indeed, phosphorylation of H1 has been shown to regulate chromatin remodeling enzymes (Horn et al., 2002).InXenopus oocytes and embryos, oocyte-specific linker histone B4 exists rather than somatic type H1 during the first divisions after fertilization. B4 is replaced by somatic H1 at the midblastula transition (Dworkin-Rastl et al., 1994; Smith et al., 1988). Accumulation of H1 is a rate- limiting factor for the loss of mesodermal competence (Steinbach et al., 1997). The major difference between H1 and B4 lies in the stability with which these proteins are incorporated into chromatin (Ura et al., 1996). In the mouse, histone synthesis is also developmentally regulated (Clarke et al., 1992; Weikowski et al., 1997). These findings suggest that linker histones play an important role in early development. Recently, during the course of a differential screening project, we discovered a mammalian oocyte-specific linker histone, H1foo, which is homologous to B4. H1foo is localized to the nucleus of germinal vesicle stage oocytes, metaphase II (MII) arrested oocytes, and the first polar body. Early one-cell stage embryos displayed H1foo immunoreactivity in condensed maternal metaphase chromatin, but not in the sperm head. However, following the extrusion of a second polar body, H1foo was detected in the swollen sperm head. Nuclear staining was somewhat Fig. 1. Indirect immunofluorescence of H1foo in reconstructed oocytes. reduced in two-cell embryos and was no longer detectable Anti-H1foo staining (A–G) and Hoechst 33342 staining (H–N). (A, H) MII oocyte and donor embryonic fibroblast cell without fusion. (B, I) in four-cell embryos (Tanaka et al., 2001). The expression Reconstructed oocyte 10 min after fusion. (C, J) Thirty minutes after pattern of H1foo in preimplantation embryos is develop- fusion. (D, K) Sixty minutes after fusion. (E, L) Six hours after activation; mentally regulated, as is Xenopus B4. An experiment of polar body (PB). (F, M) Two embryos in two-cell stage. H1foo was detected nuclear transfer in Xenopus showed that somatic type H1 in both embryos, but the immunoreactivity was reduced in the left embryo. in a donor cell is replaced by oocyte-type B4 soon after (G, N) Four-cell stage embryo. H1foo was localized to the donor nucleus shortly after transplantation and was restricted to the chromatin up to the transplantation into an oocyte, and that replacement is two-cell stage embryo. The transition of H1foo in a reconstructed oocyte is mediated by nucleoplasmin, a molecular chaperone that similar to that of the normal embryo. Scale bar = 10 Am. RN: recipient MII contributes to the acquisition of transcriptional competence oocyte nucleus, DN: donor fibroblast nucleus, PB: polar body. 78 T. Teranishi et al. / Developmental Biology 266 (2004) 76–86

Fig. 2. Subcellular distribution of H1foo-GFP throughout the cell cycle. GFP (A–D) and Hoechst 33342 staining (E–H). (A, E) Interphase. (B, F) Prometaphase. (C, G) Metaphase. (D, H) Telophase. H1foo-GFP fluorescence co-localized with chromosomes throughout the cell cycle. Scale bar = 5 Am.

(Dimitrov and Wolffe, 1996). H1 removal after nuclear virus. Following brief culture, the oocytes were artificially transfer also has been observed in mammals (Adenot et activated with 10 mM strontium for 6 h and then placed in al., 2000; Bordignon et al., 1999); however, replacement of CZB medium. H1 with oocyte-type linker histone has not yet been reported. Indirect immunofluorescence Another molecular characteristic of linker histones has been clarified. Recent studies of fluorescence recovery after At various time points after successful fusion with a photobleaching (FRAP) show that histone H1 is continu- donor cell, oocytes were fixed with freshly prepared 2% ously and rapidly exchanged between chromatin segments paraformaldehyde in phosphate-buffered saline (PBS) for 20 (Lever et al., 2000; Misteli et al., 2000). In contrast to this min. The fixed oocytes then were stained with anti-H1foo rapid exchange of H1, the association of core histones, antibody and counterstained with Hoechst 33342 as previ- especially H3 and H4, with DNA is stable (Kimura and ously described (Tanaka et al., 2001). Cook, 2001). Thus, the rapid removal of H1 in nuclear transfer may result from the greater mobility of H1. As the Plasmids immobile fraction of linker histone stabilizes chromatin structure, the different mobilities among linker histones Mouse full-length H1foo cDNAs were synthesized by may alter the chromatin structure. reverse transcriptase–polymerase chain reaction (RT-PCR) This study sought to determine whether replacement of and inserted into pcDNAII (Invitrogen, Carlsbad, CA). All the somatic linker histone with H1foo occurs during nuclear PCR primers are listed in Table 1 and annealed at 68jC. transfer in the mouse. To clarify the difference in molecular Full-length H1c cDNA was synthesized by nested PCR and dynamics between somatic and oocyte-type linker histones also inserted into pcDNAII. The correct sequences of the in vivo, we used FRAP to examine whether the mobility of desired inserts were confirmed by sequencing. The linker histones contributes to this replacement machinery pcDNAII/H1foo was subcloned into pEGFP-N3 (Clontech, and the stability of chromatin structure. Palo Alto, CA) via the NheI and EcoRI sites. The constructed pEGFP/H1foo plasmid was excised with NheI and NotI and ligated into the NheI and NotI sites of plasmid Materials and methods pTRE2hyg (Clontech). The pcDNAII/H1c was excised with BamHI and ligated into the BamHI site of pTRE2hyg/EGFP. Oocyte collection and nuclear transfer pTRE2hyg/EGFP was constructed as follows: pEGFP-N3 was excised with NheI and NotI, and ligated to the same site The procedure for oocyte collection and nuclear transfer of pTRE2hyg. To express C-His-tagged recombinant pro- has been described previously (Kono et al., 1996; Ono et al., tein, pET24b/H1foo was generated by inserting full-length 2001). Briefly, micromanipulations were performed in M2 H1foo, obtained by cutting pEGFP/H1foo with NheI and medium containing 5 Ag/ml cytochalasin B and 0.4 Ag/ml EcoRI, into pET24b (Novagen, Madison, WI). The pET24b/ nocodazole. After removal of the MII plate, a fibroblast cell H1c was produced by insertion of newly synthesized full- arrested in metaphase was introduced into the perivitelline length H1c that contained the terminal EcoRI and SalI site space of an enucleated oocyte using inactivated Sendai into pET24b. T. Teranishi et al. / Developmental Biology 266 (2004) 76–86 79

Table 1 PCR primer list Primer Sequence Additional restriction enzyme (underlined) H1foo forward 5V-GCTAGCCATATGGCTCCTGGGAGTGTCTCCAG-3V NheI H1foo reverse 5V-GCCTGAGTGTTCTCAGGGGTCTTTG-3V None H1c forward 5V-TGTAACAAACACAAGTTGGGCG-3V None H1c reverse 5V-AGCAGCCCAGCACAAACAAGTC-3V None H1c forward for nesting 5V-GGCTAGCATCATGTCTGAGGCTGCTCCTGCTG-3V NheI H1c reverse for nesting 5V-CGGATCCCTTTTTCTTGGCTGCAACCTTCTTG-3V BamHI H1c forward for protein expression 5V-CGAATTCCATGTCTGAGGCTGCTCCTGCTG-3V EcoRI H1c reverse for protein expression 5V-CGTCGACCTTTTTCTTGGCTGCAACCTTCTTG3V SalI To subclone the cDNA into the vector in-frame, the underlined restriction enzyme site was added to the 5Vend of the primer.

Cell culture, transfection, and selection of stable clones grated plasmid and induced protein expression were confirmed by PCR of genomic DNA and Western blotting, Mouse MEF/3T3 Tet-off cells (Clontech) were grown respectively. in Dulbecco’s modified Eagle’s medium, supplemented with 10% Tet system-approved fetal bovine serum, 100 Isolation of total histone proteins and immunoblottting IU/ml penicillin, 100 Ag/ml streptomycin, 2 mM L-gluta- mine, and 100 Ag/ml G418, and maintained at 37jC under Total histone proteins were isolated by the method of 5% CO2. Cells were grown in 10-cm dishes to 50% conflu- Brown et al. (1996) with slight modifications. Crude ence and transfected with lipofectamine (Gibco-BRL, Rock- histones were subjected to SDS-PAGE, using 15% re- ville, MD) using 24 Ag of plasmid and 36 Alof solving gels and 4% stacking gels. The gels were stained lipofectamine per dish. Twenty-four hours after transfection, with GelCode Blue Stain Reagent (Pierce, Rockford, IL) 4.5 Â 104 cells were re-plated in the presence of doxycy- or transferred to a polyvinylidene fluoride membrane. cline (1 Ag/ml) and hygromycin B (250 Ag/ml) in a 10-cm The membrane was blocked with 5% non-fat milk in dish. After 2 weeks, hygromycin-resistant colonies were TTBS (10 mM Tris–HCl, 150 mM NaCl, 0.05% Tween expanded in a 24-well plate. In the absence of doxycycline, 20) for 1 h. Primary antibody, mouse monoclonal anti- H1foo or H1c gene expression was detected by the presence body against green fluorescent protein (GFP; Santa Cruz of enhanced green fluorescent protein (EGFP). The inte- Biotechnology, Santa Cruz, CA), was diluted 1/200 in

Fig. 3. Effects of H1-GFP fusion expression on total histone proteins. Total histones were extracted from either parental 3T3 cells or a cell line stably expressing H1foo-GFP or H1c-GFP, and separated by SDS-PAGE. (A) Coomassie blue staining. (B) Immunodetection of GFP-tagged proteins with anti-GFP monoclonal antibody. The H1c-GFP protein level was less than 10% of total H1 protein calculated by coomassie blue staining. H1foo-GFP protein was detected by immunoblot analysis, although it was not visible by coomassie blue staining. 80 T. Teranishi et al. / Developmental Biology 266 (2004) 76–86

PBS containing 1% bovine serum albumin (BSA). Horse- FRAP radish peroxidase-conjugated goat anti-mouse secondary antibody (Santa Cruz Biotechnology) was diluted 1/5000 FRAP experiments were performed using a Laser Con- in 1% non-fat milk/TTBS. The signal was detected focal Microscope LSM 510 (Carl Zeiss, Jena, Germany). using the ECL system (Amersham Pharmacia Biotech, Piscataway, NJ).

Protein expression and purification

C-His-tagged protein was expressed in Escherichia coli BL21 (DE3) (Novagen) in LB medium. Protein expression was induced by adding isopropylthio-h-D-galactoside (IPTG) to a final concentration of 1 mM and growing the bacteria for 3 h. The bacterial pellets were resuspended in BugBuster reagent (Novagen). Cell suspension was incubated on rotating mixer at room temperature for 20 min and centrifuged at 12000 Â g for 20 min. The insoluble pellets were resuspended in buffer A (50 mM Tris–HCl pH 7.4, 4 M urea, 1 M NaCl, and 5% glycerol) including 20 mM imidazole and incubated on ice for 20 min with frequent vigorous vortexing. Cell debris was removed by centrifu- gation at 12000 Â g for 10 min. For protein purification, the supernatant was added to Ni-NTA resin (Novagen) that was equilibrated with buffer A. After gentle mixing for 30 min at 4jC, the supernatant was discarded, and the resin was washed three times with buffer A. His-tagged proteins were then eluted in buffer A containing 250 mM imidazole. The eluted sample was transferred to a dialysis bag and dialyzed against TEP buffer (5 mM Tris–HCl pH 7.5, 1 mM EDTA, and 0.1 mM phenylmethylsulfonyl fluoride) at 4jC overnight.

H1-depleted chromatin, reconstitution of chromatin, and micrococcal nuclease digestion

H1-depleted chromatin was prepared as described previously (Ura and Kaneda, 2001). Purified C-His-tagged H1c or H1foo, NaCl, and BSA were added to H1-depleted chromatin to a final concentration of 4 AM, 0.6 M, 0.5 mg/ml, respectively. To create stable reconstituted chromatin, the salt concentration of the dialysis buffer was decreased stepwise to 0.45 M, 0.3 M, and finally 0.15 M (Higurashi and Cole, 1991). A 1/100 volume of 0.3 M CaCl2 was added to the reconstituted chromatin, and the samples were pre-incubated at 37jC for 5 min. Micro- coccal nuclease (MNase) (Worthington Biochemical, Lakewood, NJ) was added (0.1 unit/Al), and digestion was carried out for 5, 10, or 30 min. The reaction was stopped by adding 100 mM EDTA, 10% SDS, and proteinase K to a final concentration of 10 mM, 0.5%, Fig. 4. H1c-GFP fluorescence in reconstructed oocytes without enucleation. and 100 Ag/ml, respectively. Total DNA was purified by H1c-GFP fluorescence (A–E) and Hoechst 33342 staining (F–J). (A, F) three extractions with phenol/chloroform/isoamyl alcohol MII oocyte and donor cell without fusion. (B, G) Reconstructed oocyte 5 and one extraction with chloroform, and was collected by min after fusion. (C, H) Thirty minutes after fusion. (D, I) Sixty minutes after fusion. (E, J) Three hours after fusion. Most of the H1c was released ethanol precipitation. MNase digested DNA was electro- within 30 min after fusion. However, faint fluorescence was detectable up phoresed on 2.2% agarose gel and stained with ethidium to 3 h after fusion. Scale bar = 10 Am. RN: recipient MII oocyte nucleus, bromide. DN: donor fibroblast nucleus. T. Teranishi et al. / Developmental Biology 266 (2004) 76–86 81

H1foo-EGFP and H1c-EGFP stable transfectants (1.5 Â 105 power, 20 images obtained after photobleaching were stored cells) were plated on a poly-L-lysine-coated 35-mm cover and analyzed with software controlling the LSM510. Fifteen glass bottom dish. Cells were imaged and photobleached cells of both H1foo-EGFP and H1c-EGFP were photo- using 488 nm argon laser with a Plan-Apochromat 63 Â oil bleached, the relative fluorescence intensity (RFI) was objective. The heterochromatin region of the nucleus was calculated as described (Misteli et al., 2000), and the mean photobleached at a high laser power, which resulted in 80% time for 50% recovery of fluorescence (t50) was determined. reduction of fluorescence intensity. Fluorescence recovery The mobile fraction was calculated by comparing the was observed during 7-s interval scanning at a low laser fluorescence in the bleached region after full recovery ( Ff)

Fig. 5. Micrococcal nuclease (MNase) digestion of reconstituted chromatin. (A) H1-depleted chromatin, purified recombinant H1s, and the dialyzed mixture of H1-depleted chromatin and H1c or H1foo were subjected to SDS-PAGE. (B) H1-depleted chromatin, reconstituted chromatin with H1c or H1foo were digested with MNase for the indicated time. Extracted DNA was electrophoresed on 2.2% agarose gel which was subsequently stained with ethidium bromide. H1foo was incorporated into nucleosome as well as H1c. 82 T. Teranishi et al. / Developmental Biology 266 (2004) 76–86 with the fluorescence before bleaching ( Fi) and just after H1foo-GFP binds to chromatin throughout the cell cycle bleaching ( F0). The mobile fraction was defined as R =(Ff (Fig. 2). H1c-GFP bound to chromatin in the same manner À F0)/( Fi À F0) (Reits and Neefjes, 2001). as H1foo-GFP (data not shown). On the basis of SDS- PAGE analysis of total histone protein extracted from cells stably expressing H1foo-GFP or H1c-GFP, we estimated Results that the cell lines overexpress less than 10% H1-GFP in addition to endogenous H1 (Fig. 3A). H1foo-GFP and H1foo is rapidly incorporated into the donor nucleus after H1c-GFP were detected in an acid-soluble nuclear fraction nuclear transfer (Fig. 3B). However, neither protein was identified in the other fraction by Western blotting (data not shown). These To investigate the possible involvement of the oocyte- results clearly demonstrate that the two GFP-H1 fusions specific linker histone H1foo during nuclear transfer, we were properly incorporated into chromatin. In addition, the analyzed H1foo immunoreactivity of reconstructed embryos expression of neither H1-GFP altered histone protein at different stages of development. Enucleated oocytes and composition, based on coomassie brilliant blue staining embryonic fibroblasts arrested in metaphase were fused (Fig. 3). using the standard micromanipulation procedure as described elsewhere (Kono et al., 1996; Ono et al., 2001). Release of somatic linker histone H1c from the donor After fusion of the fibroblast with an enucleated oocyte, a chromosome single metaphase plate was reassembled in the oocyte cortex. The detection of H1foo in the nuclear-transferred The removal of somatic linker histones after nuclear embryos is illustrated in Fig. 1. H1foo was localized to the transfer has been previously shown by indirect immuno- metaphase-arrested donor nucleus as soon as 10 min after fluorescence using anti-H1 antibody (Adenot et al., 2000). fusion. H1foo was detected in a polar body and pronucleus To visualize this phenomenon directly in vivo, MII-arrested (PN), as soon as the artificially activated, reconstructed oocyte and H1c-GFP transfectant were fused as described oocytes showed extrusion of the polar body and PN forma- except for enucleation of the recipient nucleus, and subse- tion. The polar body remained brightly fluorescent through- quent release of H1c-GFP was observed. As shown in Fig. out early embryogenesis. Nuclear staining, however, was 4, most of the H1c was released within 30 min after fusion. somewhat less in two-cell than one-cell embryos. Nuclear In addition to this rapid release, faint fluorescence was staining was no longer detectable at the four-cell stage of still detectable for 3 h. Thus, the reduction in H1c was embryonic development, but a strong signal persisted in the biphasic. polar body (Fig. 1). H1foo is incorporated into nucleosome Construction of stable transfectants To determine whether the exchange of linker histones To visualize linker histones in living cells, we fused the influences chromatin structure, we investigated the effect of coding region of the EGFP to that of mouse H1foo or MNase digestion of reconstituted chromatin obtained from somatic H1c, and established stable cell lines using mouse the addition of recombinant H1s to H1-depleted chromatin. 3T3 fibroblasts. Fluorescence microscopy showed that The purified recombinant H1s were of good quality, and an

Fig. 6. Representative images of H1c-GFP and H1foo-GFP before and during recovery after photobleaching. (A) H1c-GFP and (B) H1foo-GFP. Arrowhead: bleaching spot. Movement of H1-GFPs occurs from the unbleached to the bleached region. Scale bar = 5 Am. T. Teranishi et al. / Developmental Biology 266 (2004) 76–86 83 equal molar concentration of purified H1 was added to H1- Table 2 depleted chromatin (Fig. 5A). The nucleosomal repeat Fluorescence recovery after photobleaching (FRAP) analysis of H1foo-GFP and H1c-GFP length of the reconstituted chromatin which contained H1c F or H1foo was clearly greater than that of H1-depleted Experiment t50, time (s) SE Mobile fractions, n % F SE chromatin (Fig. 5B). H1foo was incorporated into the F F nucleosome and exerted a similar constraint on MNase H1foo 21.6 2.0* 91.7 1.8* 15 H1c 29.4 F 1.6 85.0 F 1.3 15 digestion as on H1c. The mean time for 50% recovery of fluorescence and the percent mobile fraction during the course of the experiment were determined. Statistical H1foo is more mobile than somatic type H1c in living cell significance was determined by Student’s t test (* P < 0.01). t50: the time nuclei for 50% recovery of fluorescence, SE: standard error, n: cell numbers.

To understand the molecular basis of the rapid exchange of linker histone and the differences between that H1foo is more mobile than somatic H1c in the nucleus H1c and H1foo, we investigated the dynamics of GFP- in vivo. tagged histone in unperturbed chromatin by FRAP, which can be used to define the mobility of molecules in living cells (Lever et al., 2000; Misteli et al., 2000). H1foo-GFP Discussion and H1c-GFP presented the same distribution pattern in the nucleus, and the movement of H1foo-GFP and H1c- This report shows that H1foo is rapidly accumulated GFP occurred from the unbleached to the bleached region into the donor nucleus and persists there until the two- (Fig. 6). cell stage embryo, then disappeared during the four-cell Upon bleaching the heterochromatin area, H1foo-GFP stage. We also have shown that H1foo is readily detected fluorescence recovered relatively rapidly and reached a in the swollen sperm head shortly after fertilization in plateau after 100 s. This finding clearly shows that normal preimplantation embryos, and that nuclear staining H1foo-GFP is continuously exchanged in the chromatin of H1foo is somewhat reduced in two-cell embryos and regions of the cell nucleus in a similar manner to somatic is no longer detectable in four-cell embryos (Tanaka et linker histone H1s. The recovery of H1foo-GFP was faster al., 2001). The developmentally regulated presence of than that of H1c-GFP (Fig. 7), and the mean time for 50% H1foo in a cloned embryo is therefore similar to that recovery of H1foo-GFP fluorescence was less than that of of a normal preimplantation embryo. Significantly, H1foo somatic H1c (Table 2). In addition, the immobile fraction of was detected in the donor nucleus 10 min after fusion of H1c was greater than that of H1foo. These results indicate the donor cell. In Xenopus, the midblastula transition and the activation of zygotic gene expression are associated with a dramatic decrease in B4 content and a simultaneous increase in somatic H1 (Dimitrov et al., 1993; Dworkin-Rastl et al., 1994). In nuclear transfer, the uptake of oocyte-type B4 into donor chromatin and the release of H1 is rapid, taking as little as 15 min from the time when it is mixed with the egg extract (Dimitrov and Wolffe, 1996). In the mouse, zygotic gene activation occurs immediately after formation of a two-cell embryo, a point when H1foo expression begins to decrease. Simultaneous zygotic gene activation and the transition from oocyte-type linker histone to a somatic one strongly suggest that linker histones play an important role in early development. H1 is thought to be a general repressor of transcription (Paranjape et al., 1994), but recent reports suggest that H1 has selective function in transcriptional regulation (Dou and Gorovsky, 2000; Dou et al., 1999; Shen and Gor- ovski, 1996). The replacement of B4 with somatic H1 leads to dominant and specific repression of oocyte 5S rRNA gene transcription (Bouvet et al., 1994). Somatic H1 subtypes have been shown to regulate specific gene Fig. 7. Quantitative FRAP analysis of H1foo-GFP and H1c-GFP. The values expression in the early period of Xenopus development are mean F SE. The recovery kinetics of H1foo-GFP is significantly faster (Bouvet et al., 1994; Steinbach et al., 1997). The zygote than H1c-GFP. remodels the paternal genome shortly after fertilization, 84 T. Teranishi et al. / Developmental Biology 266 (2004) 76–86 before embryonic genome activation occurs (Rideout et We demonstrated replacement of oocyte-type linker al., 2001). During the cloning process, the somatic nuclei histone with a somatic histone occurs rapidly. The greater transferred into an oocyte must be quickly reprogrammed mobility of H1foo suggests its involvement in linker to express genes required for early development. Thus, histone replacement. However, other mechanisms may H1foo might be one of the oocyte-derived factors required participate in this replacement. The large amount of for genetic reprogramming in mammals, as with B4 in H1foo and the difference in volume between donor and Xenopus. recipient cells should be taken into consideration. H1foo In FRAP, fluorescent molecules are irreversibly photo- rapidly accumulates in the donor nucleus transplanted bleached in a small area of the cell by a high-powered into the enucleated oocyte, albeit the removed nucleus focused laser beam. Subsequent diffusion of surrounding has a considerable amount of H1foo. As the nuclear non-bleached fluorescent molecules into the bleached area envelope of the recipient oocyte disintegrates in meta- leads to a recovery of fluorescence, which is recorded at phase, nuclear proteins may diffusely relocate in the low laser power. FRAP experiments provide information oocyte cytoplasm. H1foo has been identified clearly by concerning the mobility of fluorescent molecules in a Western blot analysis using only 50 oocytes, which defined compartment. In control experiments in which suggests that the oocyte originally has an excess amount whole nuclei were bleached, no recovery of fluorescence of H1foo (Tanaka et al., 2001). Thus, the large amount was detected (data not shown). This eliminates the possi- of H1foo dispersed in the reconstructed oocyte cyto- bility that de novo synthesis of H1foo-GFP or a cytosolic plasm, and it may easily attach to the externally derived counterpart of H1foo accounts for any of the fluorescence donor nucleus. On the other hand, somatic histone recovery in these experiments. Two parameters can be preferentially diffuses into oocyte cytoplasm, because an deduced from FRAP: the mobile fraction of fluorescent oocyte has about a 1000-fold greater volume than a molecules and their rate of mobility (Reits and Neefjes, fibroblast cell. 2001). Despite the similar constraint of MNase digestion in Histone modification in an oocyte alters histone mo- vitro, we have established the in vivo difference in mobil- bility. Lever et al. (2000) have shown that kinase inhib- ity between somatic and oocyte-type linker histones by itors change the mobility of H1. Although it has not been FRAP. In Xenopus, somatic H1 binds to nucleosomal DNA corroborated that H1foo can be phosphorylated, there are with an excess affinity relative to B4 (Ura et al., 1996). numerous serine and threonine moieties in H1foo that are Our FRAP data, which show that the immobile fraction of putative phosphorylation sites. The phosphorylation state H1c is greater than that of H1foo, are consistent with this of H1foo in an oocyte may affect its mobility. Another previous report. A higher immobile fraction implies tight possibility is that proteins regulate H1foo replacement assembly into chromatin with enhanced stability. We sug- directly. In Xenopus, nucleoplasmin has been shown to gest that the replacement of linker histones in the donor regulate the selective removal of somatic linker histones nucleus may destabilize the chromatin assembly, facilitat- from erythrocyte nuclei (Dimitrov and Wolffe, 1996), and ing assembly of embryonic chromatin structures that are recently, mammalian nucleoplasmin has been demonstrat- more easily replicated and transcribed when zygotic gene ed in mouse oocytes (Burns et al., 2003). Oocyte-specific activation occurs. mouse nucleoplasmin 2 (NPM2) has been found in the On the other hand, an intriguing feature of this study nucleus in oocytes and eight-cell embryos, and it is was the finding that the reduction in H1 in the recon- crucial to heterochromatin formation in early embryos. structed oocyte which was fused with H1c-GFP trans- On the basis of the common features of oocyte-specific fectant was biphasic. FRAP, which demonstrated that linker histone in frogs and mice, NPM2 is thought to 85% of H1c-GFP was mobile and 15% was immobile, regulate the removal of somatic H1 and cooperate with provides a compelling explanation to account for the H1foo to remodel nuclear function. Alekseev et al. (2002) biphasic reduction pattern. Most of the H1 that is lost have suggested that NASP (nuclear autoantigenic sperm was from the mobile fraction that can be replaced rapidly protein) might be one of the H1 regulating proteins. with H1foo or diffused into the oocyte cytoplasm, whereas Certainly, other yet to be determined proteins may partic- residual H1 represents the immobile fraction. Interestingly, ipate in the replacement process. Further investigation is the time interval between the injection of the somatic cell needed to clarify the mechanism responsible for this nucleus into the enucleated oocyte and oocyte activation phenomenon. affects the rate of development (Wakayama et al., 1998, In conclusion, we demonstrated that somatic type linker 2000). Activation immediately after nucleus injection histone H1 in a donor nucleus transplanted into an oocyte is resulted in significantly less progression to the morulae/ rapidly replaced with oocyte-type H1foo. The greater mo- blastocyst stage in vitro than when activation followed a bility of H1foo, compared with H1, may contribute to this delay of 1 to 6 h (Wakayama et al., 1998). This interval is rapid replacement and the instability of chromatin struc- consistent with the H1 removal time. Therefore, remaining tures. These findings suggest that the rapid replacement of somatic H1 may disturb appropriate gene expression and H1 with H1foo may play an important role in nuclear lead to failure of development. remodeling. T. Teranishi et al. / Developmental Biology 266 (2004) 76–86 85

Acknowledgments Horn, P.J., Carruthers, L.M., Logie, C., Hill, D.A., Solomon, M.J., Wade, P.A., Imbalzano, A.N., Hansen, J.C., Peterson, C.L., 2002. Phosphory- lation of linker histones regulates ATP-dependent chromatin remodeling We thank Dr. Kiyoe Ura for her valuable advice and enzymes. Nat. Struct. Biol. 9, 263–267. critical reading of the manuscript. We also thank Dr. Eli Y. Kikyo, N., Wolffe, A.P., 2000. Reprogramming nuclei: insights from clon- Adashi for his gift of H1foo antibody. This work was ing, nuclear transfer and heterokaryons. J. Cell Sci. 113, 11–20. supported, in part, by the Ministry of Education, Science, Kimura, H., Cook, P.R., 2001. Kinetics of core histones in living human Sports and Culture, a Grant-in-Aid for Scientific Research cells: little exchange of H3 and H4 and some rapid exchange of H2B. J. Cell Biol. 153, 1341–1353. (C), 14571584, 2002, and the Keio Gijuku Academic Kono, T., Obata, Y., Yoshimzu, T., Nakahara, T., Carroll, J., 1996. Development Fund 2001 (to M. Tanaka). Epigenetic modifications during oocyte growth correlates with ex- tended parthenogenetic development in the mouse. Nat. Genet. 13, 91–94. Latham, K.E., 1999. Mechanisms and control of embryonic genome References activation in mammalian embryos. In: Jeon, K.W. (Ed.), Interna- tional Review of Cytology, vol. 193. Academic Press, San Diego, Adenot, P.G., Campion, E., Legouy, E., Allis, C.D., Dimitrov, S., Renard, pp. 71–124. J., Thompson, E.M., 2000. Somatic linker histone H1 is present Lever, M.A., Th’ng, J.P.H., Sun, X., Hendzel, M.J., 2000. Rapid ex- throughout mouse embryogenesis and is not replaced by variant H10. change of histone H1.1 on chromatin in living human cells. Nature J. Cell Sci. 113, 2897–2907. 408, 873–876. Alekseev, O.M., Bencic, D.C., Richardson, R.T., Widgren, E.E., O’Rand, Misteli, T., Gunjan, A., Hock, R., Bustin, M., Brown, D.T., 2000. Dy- M.G., 2003. Overexpression of the linker histone-binding protein namic binding of histone H1 to chromatin in living cells. Nature 408, tNASP affects progression through the cell cycle. J. Biol. Chem. 278, 877–881. 8846–8852. Ono, Y., Shimozawa, N., Ito, M., Kono, T., 2001. Cloned mice from fetal Bordignon, V., Clarke, H.J., Smith, L.C., 1999. Developmentally regu- fibroblast cells arrested at metaphase by a serial nuclear transfer. Biol. lated loss and reappearance of immunoreactive somatic histone H1 on Reprod. 64, 44–50. chromatin of bovine morula-stage nuclei following transplantation into Paranjape, S.M., Kamakaka, R.T., Kadonaga, J.T., 1994. Role of chro- oocytes. Biol. Reprod. 61, 22–30. matin structure in the regulation of transcription by RNA polymerase Bouvet, P., Dimitrov, S., Wolffe, A.P., 1994. Specific regulation of Xenopus II. Annu. Rev. Biochem. 63, 265–297. chromosomal 5S rRNA gene transcription in vivo by histone H1. Genes Reik, W., Dean, W., Walter, J., 2001. Epigenetic reprogramming in mam- Dev. 8, 1147–1159. malian development. Science 293, 1089–1093. Brown, D.T., Alexander, B.T., Sittman, D.B., 1996. Differential effect of Reits, E.A., Neefjes, J.J., 2001. From fixed to FRAP: measuring H1 variant overexpression on cell cycle progression and gene expres- protein mobility and activity in living cells. Nat. Cell Biol. 3, sion. Nucleic Acids Res. 24, 486–493. E145–E147. Burns, K.H., Viveiros, M.M., Ren, Y., Wang, P., DeMayo, F.J., Frail, Rideout III, W.M., Eggan, K., Jaenisch, R., 2001. Nuclear cloning and D.E., Eppig, J.J., Matzuk, M.M., 2003. Roles of NPM2 in chromatin epigenetic reprogramming of the genome. Science 293, 1093–1098. and nucleolar organization in oocytes and embryos. Science 300, Shen, X., Gorovski, M.A., 1996. Linker histone H1 regulates specific gene 633–636. expression but not global transcription in vivo. Cell 86, 475–483. Clarke, H.J., Oblin, C., Bustin, M., 1992. Developmental regulation of Shi, W., Zakhartchenko, V., Wolf, E., 2003. Epigenetic reprogramming in chromatin composition during mouse embryogenesis: somatic his- mammalian nuclear transfer. Differentiation 71, 91–113. tone H1 is first detectable at the 4-cell stage. Development 115, Smith, R.C., Dworkin-Rastl, E., Dworkin, M.B., 1988. Expression of a 791–799. histone H1-like protein is restricted to early Xenopus development. Dimitrov, S., Wolffe, A.P., 1996. Remodeling somatic nuclei in Xenopus Genes Dev. 2, 1284–1295. laevis egg extracts: molecular mechanisms for the selective release of Steinbach, O.C., Wolffe, A.P., Rupp, R.A., 1997. Somatic linker histo- histones H1 and H1j from chromatin and the acquisition of transcrip- nes cause loss of mesodermal competence in Xenopus. Nature 389, tional competence. EMBO J. 15, 5897–5906. 395–399. Dimitrov, S., Almouzni, G., Dasso, M., Wolffe, A.P., 1993. Chromatin Tanaka, M., Hennebold, J.D., Macfarlane, J., Adashi, E.Y., 2001. A transitions during early Xenopus embryogenesis: changes in histone mammalian oocyte-specific linker histone gene H1oo: homology with H4 acetylation and in linker histone type. Dev. Biol. 160, 214–227. the genes for the oocyte-specific cleavage stage histone (cs-H1) of Dou, Y., Gorovsky, M.A., 2000. Phosphorylation of linker histone H1 sea urchin and the B4/H1M histone of the frog. Development 128, regulates gene expression in vivo by creating a charge patch. Mol. Cell 655–664. 6, 225–231. Thoma, F., Koller, T., Klug, A., 1979. Involvement of histone H1 in the Dou, Y., Mizzen, C.A., Abrams, M., Allis, C.D., Gorovsky, M.A., 1999. organization of the nucleosome and of the salt-dependent superstruc- Phosphorylation of linker histone H1 regulates gene expression in vivo tures of chromatin. J. Cell Biol. 83, 403–427. by mimicking H1 removal. Mol. Cell 4, 641–647. Ura, K., Kaneda, Y., 2001. Reconstitution of chromatin in vitro. Methods Dworkin-Rastl, E., Kandolf, H., Smith, R.C., 1994. The maternal histone Mol. Biol. 181, 309–325. H1 variant, H1M (B4 protein), is the predominant H1 histone in Xeno- Ura, K., Nightingale, K., Wolffe, A.P., 1996. Different association of pus pregastrula embryos. Dev. Biol. 161, 425–439. HMG1 and linker histones B4 and H1 with dinucleosomal DNA: Fan, Y., Sirotkin, A., Russell, R.G., Ayala, J., Skoultchi, A.I., 2001. Indi- structural transitions and transcriptional repression. EMBO J. 15, vidual somatic H1 subtypes are dispensable for mouse development 4959–4969. even in mice lacking the H1j replacement subtype. Mol. Cell. Biol. Wade, P.A., Kikyo, N., 2002. Chromatin remodeling in nuclear cloning. 21, 7933–7943. Eur. J. Biochem. 269, 2284–2287. Felsenfeld, G., McGhee, J.D., 1986. Structure of 30 nm chromatin fiber. Wakayama, T., Perry, A.C.F., Zuccotti, M., Johnson, K.R., Yanagimachi, Cell 44, 375–377. R., 1998. Full-term development of mice from enucleated oocytes in- Higurashi, M., Cole, R.D., 1991. The combination of DNA methylation jected with cumulus cell nuclei. Nature 394, 369–374. and H1 histone binding inhibits the action of a restriction nuclease on Wakayama, T., Tateno, H., Mombaerts, P., Yanagimachi, R., 2000. Nuclear plasmid DNA. J. Biol. Chem. 266, 8619–8625. transfer into mouse zygotes. Nat. Genet. 24, 108–109. 86 T. Teranishi et al. / Developmental Biology 266 (2004) 76–86

Weikowski, M., Miranda, M., Nothias, J.Y., Depamphilis, M.L., 1997. Viable offspring derived from fetal and adult mammalian cells. 1997. Changes in histone synthesis and modification at the begin- Nature 385, 810–813. ning of mouse development correlate with the establishment of Wolffe, A.P., 1995. Chromatin Structure and Function. Academic Press, chromatin mediated repression of transcription. J. Cell Sci. 110, London. 1147–1158. Zlatanova, J., Doenecke, D., 1994. Histone H10: a major player in cell Wilmut, I., Schnieke, A.E., McWhir, J., Kind, A.J., Campbell, K.H.S., differentiation? FASEB J. 8, 1260–1268.