Inadequate histone deacetylation during oocytemeiosis causes aneuploidy and embryo death in miceTomohiko Akiyama, Masao Nagata, and Fugaku Aoki*

Department of Integrated Biosciences, Graduate School of Frontier Sciences, University of Tokyo, Chiba 277-8562, Japan

Edited by Barbara J. Meyer, University of California, Berkeley, CA, and approved March 10, 2006 (received for review December 20, 2005)

Errors in meiotic chromosome segregation are the leading cause ofspontaneous abortions and birth defects. Almost all such aneu-ploidy derives from meiotic errors in females, with increasingmaternal age representing a major risk factor. It was recentlyreported that histones are globally deacetylated in mammalianoocytes during meiosis but not mitosis. In the present study,inhibition of meiotic histone deacetylation was found to induceaneuploidy in fertilized mouse oocytes, which resulted in embry-onic death in utero at an early stage of development. In addition,a histone remained acetylated in the oocytes of older (10-month-old) female mice, suggesting that the function for histone deacety-lation is decreased in the oocytes of such mice. Thus, histonedeacetylation may be involved in the fair distribution of chromo-somes during meiotic division. The high incidence of aneuploidy inthe embryos of older females may be due to inadequate meiotichistone deacetylation.

acetylation � meiotic maturation � chromosome segregation

Mammalian female meiosis consists of specialized programsof cell division. Oocytes arrested at meiotic prophase I for

long periods of time resume meiosis in response to hormonalstimulation and then proceed through meiosis I. After comple-tion of this stage, during which the first polar body is extruded,oocytes successively enter meiosis II and are arrested again atmetaphase II (MII). After fertilization, the second stage ofmeiosis is completed, with extrusion of the second polar body.Because, in this process, chromosomes are fairly distributedbetween the oocytes themselves and the polar bodies, thereappears to be a meiosis-specific mechanism that functions insupporting a distinct pattern of chromosome distribution. In-deed, the oocytes of older females, in which many cellularfunctions have deteriorated (1), frequently fail in the fair dis-tribution of chromosomes, which results in aneuploidy (2).

Although a mechanism supporting a fair distribution of chro-mosomes has not yet been identified, meiosis-specific changes inhigher-order chromosome architecture may be involved in sucha process. During extended meiotic prophase, chromosomemorphology changes dynamically, allowing prophase to be di-vided into five stages (leptotene, zygotene, pachytene, diplotene,and diakinesis). Accurate chromosome segregation at meiosis Irequires physical connections between homologous chromo-somes to form hotspots for meiotic recombination (3). Chro-matin at these regions displays a meiosis-specific increase innuclease sensitivity, suggesting that recombinant events requirealterations in chromatin structure (4). However, the mechanismresponsible for a meiosis-specific chromatin configuration hasnot been clearly elucidated.

Posttranslational histone modifications are increasingly beingrecognized as playing important roles in chromosome structureand segregation during meiosis. Recently, a correlation betweenchanges in histone modifications and chromosome structureduring meiosis was reported in fungi and worms. Histones H3and H4 in the vicinity of ade6-M26, a well characterized site ofmeiotic recombination, were hyperacetylated during meiosis infission yeast (5). In Caenorhabditis elegans, him-17-null mutantsdefective for meiotic recombination and chromosome segrega-

tion display a reduced and delayed accumulation of methylationon histone H3, Lys-9 (6).

In mammalian oocytes, the N-terminals of the nucleosomecore histones, H3 and H4, are acetylated at prophase I, whereasthey are deacetylated globally at MI and MII by histone deacety-lase (HDAC) activity (7–9). However, the function of histonedeacetylation during meiotic metaphase in oocytes is unknown.Although histone acetylation is involved in the regulation of geneexpression (10–12), a transcriptionally inactive state has beenalready established in the oocytes at prophase I and is main-tained during meiosis (13). Therefore, meiotic global histonedeacetylation may instead be associated with changes in chro-matin structure that regulate cellular functions other than geneexpression.

In the present study, the role of histone deacetylation in themeiosis of mouse oocytes was analyzed. The results showed thatinhibition of histone deacetylation during meiosis induces a highfrequency of aneuploidy and embryo death. In addition, histonein oocytes obtained from older mice frequently remained acety-lated during meiosis, suggesting that a deficiency in the mech-anism regulating histone deacetylation is associated with a highfrequency of aneuploidy in oocytes from older females.

ResultsProphase-I-arrested mouse oocytes with germinal vesicles (GVs)spontaneously resume and complete meiosis I in vitro andthereafter arrest at MII stage. Resumption and completion ofmeiosis I is marked by GV breakdown and extrusion of the firstpolar body, respectively. We therefore first investigated whetherinhibition of histone deacetylation affects this progression ofmeiotic maturation. Deacetylation was inhibited by culturing theoocytes in vitro with trichostatin A (TSA), a histone deacetylase-specific inhibitor (14). Inhibition of histone deacetylation wasconfirmed by immunostaining with an antibody (Ab) that reactsspecifically with acetylated Lys-12 of histone H4 (H4K12) (Fig.1). All of the oocytes treated with TSA underwent GV break-down, and 66% emitted the first polar body (Table 1). Theseindices for the progression of meiosis differed little from thosemeasured in oocytes cultured without TSA, demonstrating thatinhibition of histone deacetylation did not influence meioticmaturation.

Next, we examined the effect of inhibiting meiotic histonedeacetylation on preimplantation development. GV-stage oo-cytes that had been allowed to proceed with meiotic maturationand reached the MII stage in the presence of TSA were fertilizedin vitro, and their development until blastocyst stage was fol-lowed (Table 2). TSA treatment continued for 3 h after insem-ination, by which time the second polar body was emitted, andmeiosis was completed. As a control, the oocytes were allowed

Conflict of interest statement: No conflicts declared.

This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: GV, germinal vesicle; MII, metaphase II; TSA, trichostatin A.

*To whom correspondence should be addressed at: Department of Integrated Biosciences,Graduate School of Frontier Sciences, University of Tokyo, Kashiwa, Chiba 277-8562,Japan. E-mail: aokif@k.u-tokyo.ac.jp.

© 2006 by The National Academy of Sciences of the USA

www.pnas.org�cgi�doi�10.1073�pnas.0510946103 PNAS � May 9, 2006 � vol. 103 � no. 19 � 7339–7344

DEV

ELO

PMEN

TAL

BIO

LOG

Y

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0

to mature to MII stage in vitro without TSA and then weretreated with TSA for 3 h beginning immediately after insemi-nation. In these control embryos, histones were deacetylatedbecause they had undergone meiotic maturation without TSAtreatment. The results showed that both the oocytes treated withTSA during meiotic maturation and control oocytes were fer-tilized in vitro at frequencies of �90%. After fertilization, 72%of the TSA-treated embryos and 61% of the control embryosreached blastocyst stage. Therefore, inhibiting histone deacety-lation during meiosis did not influence fertilization and preim-plantational development.

We then examined the functional role(s) of meiotic histonedeacetylation in embryonic growth after implantation. TSA-treated embryos developing into the two-cell stage were trans-ferred into the oviducts of pseudopregnant females. Recipientfemales transplanted with TSA-treated embryos exhibited asharp reduction in litter size, generating on average 3.7 offspringper female compared with 8.3 offspring per female in micetransplanted with control embryos (Fig. 2A). Furthermore, ofthe seven females transplanted with TSA-treated embryos, onlytwo gave birth normally on day 19. Natural birth on day 20 wasobserved in three females, and in two females, cesarean sectionwas performed after noon on day 20. Among the 26 offspringobtained from the seven females transplanted with the TSA-treated embryos, 1 died soon after birth, and 2 died by 2 or 3weeks after birth. However, abnormalities were not observed inthe other live pups, and their rate of growth was not significantlydifferent from that of control pups (see Fig. 5, which is publishedas supporting information on the PNAS web site). Embryo deathin utero was investigated on day 11 by examining the embryostransferred into recipients (Fig. 2B). The result revealed thatfemales transplanted with TSA-treated embryos contained anaverage of 5.3 resorbed embryos, whereas in females trans-planted with control embryos the average was 1.8 resorbedembryos.

We postulated that the death of TSA-treated embryos afterimplantation was due to the generation of aneuploid oocytesresulting from chromosomal segregation errors. To address thispossibility, karyotypic status was analyzed in one-cell zygotesthat had completed oocyte meiosis (Fig. 3A). Hypo- and hyper-ploidy was observed in 32.7% and 24.5% of the TSA-treatedone-cell zygotes (n � 49) compared with 8.3% and 10% ofcontrol zygotes, respectively (n � 60) (Fig. 3B). Furthermore,34% of the TSA-treated oocytes that matured to the MII stage(n � 32) showed aberrant chromosome alignment: some chro-mosomes were scattered apart from the metaphase plate in thecytoplasm (Fig. 3C). There was no abnormality in the controloocytes treated without reagent (n � 34). Moreover, in most ofthe TSA-treated oocytes, chromosomal condensation appearedto be looser than that in the control oocytes. These results

Fig. 2. Inhibition of histone deacetylation during meiosis affects embryosurvival. (A) Two-cell stage embryos treated with TSA during meiosis weretransferred into pseudopregnant females (TSA�), and the number of pupsborn was counted. As a control, two-cell-stage embryos treated with TSA onlyat fertilization were transferred into pseudopregnant females (TSA�). Foreach treatment, we used seven pseudopregnant females and transferred 7–11embryos per oviduct. Females transplanted with TSA-treated embryos deliv-ered significantly fewer offspring than females transplanted with controlembryos (P � 0.005, Student’s t test). (B) Females transplanted with TSA-treated (TSA�) or control (TSA�) embryos were killed at 12 days postcoitus.Three or more pseudopregnant females were used for each treatment, and 10embryos per oviduct were transferred. Embryo resorption in females trans-planted with TSA-treated embryos was significantly higher than in femalestransplanted with control embryos (P � 0.05, Student’s t test). Data representthe mean � SE.

Table 2. Effect of TSA treatment on preimplantationdevelopment

Condition*

No. ofembryos

examined

No. (%) of embryos developed into

2-Cellstage

4-Cellstage Morula Blastocyst

TSA� 33 29 (87.9) 24 (72.7) 22 (66.7) 20 (60.6)TSA� 50 46 (92.0) 42 (84.0) 40 (80.0) 36 (72.0)

*Oocytes were treated with TSA from the GV stage until 3 h after insemination(TSA�) or only immediately after insemination until 3 h later (TSA�).

Fig. 1. Inhibition of histone deacetylation during meiotic maturation byTSA. GV-stage oocytes (GV) were cultured in vitro with (TSA�) or without(TSA�) TSA and allowed to undergo meiotic maturation. Eighteen hours afterisolation from the follicles, at which time the oocytes had reached the MIIstage, the oocytes were collected and immunostained with Ab against acety-lated H4K12 (AcH4K12). The Ab was localized with FITC-conjugated secondaryAb (green), and DNA was stained with DAPI (blue). The experiment wasperformed three times with similar results; representative examples areshown.

Table 1. Effect of TSA treatment on meiotic maturation

Condition*No. of oocytes

cultured

No. (%) of oocytes

GVBD MII

TSA� 159 159 (100) 107 (67.3)TSA� 122 122 (100) 80 (65.6)

*The GV-stage oocytes were cultured for 18 h with TSA (TSA�) or without TSA(TSA�).

7340 � www.pnas.org�cgi�doi�10.1073�pnas.0510946103 Akiyama et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0

demonstrate that the inhibition of histone deacetylation duringoocyte meiosis induced aneuploidy, which led to embryo death.

In humans and mice, aneuploidy is predominantly caused byaberrant female meiotic chromosome segregation, and its fre-quency increases dramatically with maternal age (15–18). There-fore, the acetylation state of histone H4K12 during oocytemeiosis was studied in older (10-month-old) mice (Fig. 4). In40% of MII-stage oocytes obtained from these mice (n � 81),histone H4K12 remained acetylated, whereas it was completelydeacetylated in oocytes from young (3-week-old) mice (n � 74).We also examined the acetylation state of histones H3K14,H4K8, and H4K16. Histone H4K8 was distinctly acetylated inmost of the MII-stage oocytes obtained from older mice, al-though only a faint level of acetylation was detected in almost allof the oocytes from young mice. When their signal intensitieswere quantified, a significant difference was detected (P � 0.005,Student’s t test; see Fig. 6, which is published as supportinginformation on the PNAS web site). Histones H3K14 and H4K16were deacetylated in the oocytes from older mice as well asyoung mice. These experiments were performed by using super-ovulated oocytes. To exclude the possibility that the treatmentfor superovulation affected the acetylation state of H4K12, werepeated the same analysis for histone acetylation in the oldermice but with naturally ovulated oocytes. The results wereessentially the same as those using superovulated oocytes (seeFig. 7, which is published as supporting information on the PNASweb site). These results suggest that the mechanism-generatinghistone deacetylation was disrupted during meiosis in olderfemales, thus inducing a high frequency of aneuploidy.

DiscussionThis study demonstrates that inhibition of histone deacetylationduring female meiosis induces aneuploidy and embryo death inmice and that a histone remained acetylated during meiosisin older female mice. These results suggest that the increase inaneuploidy of oocytes from older females is caused by a defi-ciency in the mechanism that regulates histone deacetylationduring meiosis.

The analysis of fertilized oocytes showed that TSA treatmentduring meiotic maturation induced a significant increase in thefrequency of chromosomal abnormalities (Fig. 3). Thus, histonedeacetylation seems to play a key role in preventing aneuploidyduring female meiosis. In somatic cells and mammalian malegerm cells, a spindle assembly checkpoint monitors chromosomealignment and spindle integrity during cell division (19, 20). Inthe absence of proper chromosome alignment, the cell cycle isarrested or anaphase is delayed, allowing the cell to correcterrors that might otherwise produce aneuploid progeny. How-ever, this checkpoint is missing or less stringent in mammalianoocytes. Disturbances in the alignment of chromosomes duringfemale meiosis caused by either chemical exposure (21) ormutations that disrupt folliculogenesis (22) do not induce mei-otic arrest or a delay in anaphase onset. When the spindlecheckpoint is absent, a fair distribution of chromosomes duringfemale meiosis requires that there be a female meiosis-specificmechanism for aligning chromosomes correctly and preventinganeuploidy. We propose that histone deacetylation is one suchmechanism. Although TSA treatment had no effect on theprogression of cell division during meiosis, it induced a highfrequency of unequal chromosome distribution during meiosis,

Fig. 3. One-cell zygotes treated with TSA during meiosis showed chromosomal aneuploidy. Oocytes were treated with TSA from the GV stage until 3 h afterinsemination (TSA�) or only immediately after insemination until 3 h later (TSA�). (A) Zygotes at the one-cell stage were mitotically arrested by colcemidtreatment, and the chromosomal karyotypes were examined. Individual pronuclei are indicated by dotted lines; the arrow indicates a pronucleus with one extrachromosome (n � 21 chromosomes). (B) We analyzed the number of chromosomes at the one-cell stage. One-cell zygotes treated with TSA during meioticmaturation showed a significantly higher incidence of aneuploidy (the sum of hypo- and hyperploidy) compared with control zygotes (P � 0.01, �2 test). (C) TSAtreatment caused aberrant chromosomal arrangement in MII-stage oocytes. GV-stage oocytes were cultured in vitro with (TSA�) or without (TSA�) TSA. Theoocytes that had reached the MII stage were immunostained with an Ab against acetylated H4K12 (AcH4K12, green) and �-tubulin (red), and the DNA was stainedwith DAPI (blue).

Akiyama et al. PNAS � May 9, 2006 � vol. 103 � no. 19 � 7341

DEV

ELO

PMEN

TAL

BIO

LOG

Y

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0

resulting in aneuploidy (Fig. 3). These observations support thehypothesis that female meiosis is regulated by mechanismsdifferent from those of mitosis or male meiosis, in whichchromosomal errors are corrected or aneuploidic cells areeliminated. Instead, in female meiosis, a fair distribution ofchromosomes is ensured by the deacetylation of nucleosomalhistones and alterations in chromatin structure.

We previously reported that TSA treatment had no influenceon the progression of meiotic maturation (7). This result wasconfirmed in the present study by using a different mouse strainand different culture medium. Recently, De La Fuente et al. (23)reported that exposure to TSA during meiotic maturation in-duces abnormal chromosome alignment at the MII stage inmouse oocytes. This result is consistent with our finding thatTSA treatment during female meiosis induces aneuploidy infirst-cleavage embryos. However, De La Fuente et al. (23) alsofound that TSA treatment inhibited the progression of meioticmaturation in half of the oocytes examined. This discrepancy inthe effect of TSA treatment on meiotic maturation remains to beelucidated.

TSA seems to inhibit deacetylation of histones specificallyduring female meiosis. To our knowledge, TSA inhibits deacety-lation of two molecules as well as histones. First, it appears thatTSA can increase the levels of acetylated p53 protein in somaticcells (24). Thus, changes in some proteins whose expressions arecontrolled by p53 might be involved in the chromosomal abnor-mality during meiosis. However, because there is no transcrip-tional activity during meiosis (13), it is unlikely that changes inacetylation of p53 would be associated with aneuploidy by TSAtreatment in oocyte. Second, it was reported that TSA promotesthe cellular acetylation of �-tubulin (25). However, immunocy-tochemistry with a specific Ab against acetylated �-tubulinshowed that TSA did not affect acetylation level of �-tubulinduring female meiosis (F.A., unpublished data). Therefore, it is

unlikely that aneuploidy in one-cell zygotes was caused by theincreased acetylation of p53 or �-tubulin with TSA treatment.

In the present study, we prepared the control oocytes for TSAtreatment during meiosis by treating the oocytes with TSA at thetime of insemination (3-h periods after insemination). Thiscontrol was designed to exclude the effect of TSA on fertiliza-tion. In this control, aneuploidy was detected at the frequency of18.3% (Fig. 3). Although this value is markedly lower than thatin the oocytes that had been treated with TSA during meiosis, itwas much higher than those that have been reported in someprevious studies (�1.5%) (16, 26). This high incidence ofaneuploidy in our controls may have been due to the effects ofin vitro maturation and fertilization. In previous studies thatreported a low frequency of aneuploidy, one-cell zygotes wereobtained from mated females: thus, they underwent meioticmaturation and fertilization in vivo (16, 26). However, severalstudies have shown that when GV-stage oocytes are matured tothe MII stage in vitro, aneuploidy is observed in 11–33% of them(27–29). Moreover, Fraser and Maudlin (16) reported an in-creased incidence of aneuploidy (�4.3%) in one-cell zygotesfertilized in vitro compared with those fertilized in vivo (�1.5%).Therefore, it is likely that in vitro maturation and fertilizationincrease the frequency of aneuploidy. Indeed, we examined thefrequency of aneuploidy in one-cell zygotes that had beenmatured to the MII stage in vitro and fertilized in vitro withoutany TSA treatment and found that aneuploidy occurred in15.9%. Culturing the oocytes in vitro may induce the reductionof histone-deacetylating activity, resulting in the increased fre-quency of aneuploidy.

Aneuploidy has been estimated to occur in 10–25% of fertil-ized human oocytes and 0.3% of all human newborns; thus,numerical chromosomal abnormalities are the leading cause ofpregnancy loss, congenital defects, and mental retardation (2).Almost all instances of aneuploidy derive from female meioticerrors, and increasing maternal age represents a well docu-mented risk factor. An important finding in the present study wasthat histone acetylation persisted in the oocytes from olderfemale mice, suggesting that decreased histone deacetylationcontributes to maternal-age-related aneuploidy in mammals. Ithas been reported that the frequency of meiotic chromosomeerrors markedly increases with maternal age. In mice, a signif-icantly higher incidence of aneuploidy was observed in fertilizedoocytes from older animals than in those from young ones (7.5%vs. 3.3%) (16). In humans, the chance of generating chromo-somally abnormal fetuses increases from 6.8% in women 35–39years of age to �50% in those 45 years of age (30). Moreover,the risk of having a baby with Down’s syndrome, trisomy 18, orother chromosomal abnormalities increases with increasing ma-ternal age. Based on the age-associated increase of spontaneousabortions and numerical meiotic division errors, it has beensuggested that an age-related decline in oocyte quality is asso-ciated with a higher incidence of aneuploidy (31). However,neither a molecular mechanism explaining the high incidence ofaneuploidy in older mothers nor the presence of another factordetermining oocyte quality and the even distribution of chro-mosomes has been reported. We have shown that histone H4K12remained acetylated in 40% of the MII-stage oocytes of oldermice, whereas histone deacetylation was complete in the oocytesof young mice (Fig. 4). Moreover, a higher signal of acetylatedH4K8 was detected in MII-stage oocytes from older mice thanyoung ones, although some other lysines (H3K14 and H4K16)were deacetylated in the oocytes from older mice (Fig. 6). Theseresults suggested that the ability of deacetylating histones haddecreased, but not disappeared, in the oocytes from olderfemales. Furthermore, in young females, the maintenance ofhistone acetylation after TSA treatment led to �50% aneuploidyin fertilized oocytes. Thus, remaining acetylation during meiosisseems to increase the incidence of aneuploidy in the oocytes

Fig. 4. Histone acetylation persists in MII-stage oocytes from older mice. (A)MII-stage oocytes obtained from young (3 weeks) and older (10 months)female mice were immunostained with Ab against acetylated H4K12(AcH4K12). The Ab was localized with FITC-conjugated secondary Ab (green),and DNA was stained with DAPI (blue). (B) Histone acetylation was completelyundetectable in all oocytes from young mice but was frequently observed inthe oocytes from old mice.

7342 � www.pnas.org�cgi�doi�10.1073�pnas.0510946103 Akiyama et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0

from older mice. In these oocytes, however, the level of histoneacetylation was lower at the MII stage than the GV-stage,indicating that histones were deacetylated to some extent duringmeiotic maturation. This finding may be one reason why thefrequency of aneuploid embryos (�10% of the frequency re-ported by most studies) is much lower than that of the embryoswith acetylated histone H4K12 (40%) in older female mice.Taking these results together, we propose a mechanism in whichhistone deacetylation during meiosis is frequently disrupted inthe oocytes of older females, resulting in a significant increase inthe incidence of aneuploidy in their offspring.

Materials and MethodsCollection and Culture of Oocytes. Fully grown oocytes arrested atprophase I of meiosis were obtained by superovulating 3-week-old BDF1 female mice (CLEA Japan, Tokyo) by using 5 units ofpregnant mare’s serum gonadotrophin (PMSG). The ovarieswere removed from the mice at 47 h after PMSG treatment andtransferred into Hepes-buffered Whitten’s medium (32) supple-mented with 3 mg�ml BSA (Sigma). The ovarian follicles werepunctured with 27-gauge needles to release oocytes enclosedwith cumulus cells. The cumulus cells surrounding the oocyteswere removed by gentle pipetting through a narrow-bore glasspipette, and the oocytes with a GV were collected. The oocytesthen were transferred into fresh Whitten’s medium and incu-bated in a humidified atmosphere of 5% CO2�95% air at 38°C.For oocytes fertilized in vitro after the progression of meiosis,GV-stage oocytes were cultured in Waymouth’s medium insteadof Whitten’s medium as described in ref. 33.

In Vitro Fertilization and Embryo Cultures. Sperm were collectedfrom the caudal epididymis of adult ICR male mice (SLC Japan,Shizuoka, Japan) and capacitated by preincubation for 1 h in humantubal fluid (HTF) medium (34) supplemented with 30 mg�ml BSA.MII stage oocytes, cultured in HTF medium for 18 h, wereinseminated with capacitated sperm in a humidified atmosphere of5% CO2�95% air at 38°C. Three hours after insemination, thefertilized oocytes were washed and cultured in K-modified simplexoptimized medium (KSOM) (35) containing 3 mg�ml BSA.

TSA Treatment. Histone deacetylase activity during meiosis wasinhibited by treating GV-stage oocytes with 100 nM TSA(Sigma) added to the culture medium for 18 h. A stock solutionof TSA (1 mM) dissolved in water was diluted with culturemedium immediately before use. Oocytes fertilized after meiosisprogression in vitro were inseminated with sperm in the presenceof TSA and cultured for 3 h, during which time the second polarbody was emitted, and meiosis was completed. The fertilizedoocytes then were washed with K�-modified simplex optimizedmedium and cultured. The presence of TSA did not affect therate of fertilization. As a control, GV-stage oocytes were cul-tured in vitro in the absence of TSA, and then oocytes that hadreached MII stage were treated with TSA from the timeimmediately after insemination until 3 h later.

Embryo Transfer. Embryos that had reached the two-cell stagewere transferred into the oviducts (7–11 embryos per oviduct) ofICR females (SLC Japan) mated during the previous night withvasectomized ICR males. On day 19 (day 0 was the day when thevaginal plug was observed), the number of pups born was

recorded, and live pups were nursed by lactating ICR females.Recipient females that did not give birth naturally until noon onday 20 were killed; the pups were quickly removed from theuteruses, and live pups were raised by lactating females. Thebody weights of the pups were measured weekly. The number ofembryos lost was assessed by killing the recipients on day 11 andthen recording the number of implantation sites and fetuses.

Chromosomal Analysis of One-Cell Zygotes. At 11 h after insemi-nation, one-cell zygotes were incubated in K�-modified simplexoptimized medium containing 0.1 �g�ml colcemid to arrest thecells at the first-cleavage metaphase. Cytogenetic preparation ofthe zygotes was performed according to procedures described inref. 36 with some modifications. In brief, 6 h after colcemidtreatment, zygotes were transferred into a hypotonic solution of0.9% sodium citrate for 10 min at room temperature and thenexposed to a freshly prepared fixative mixture of 3:1 methanol:acetic acid. Chromosomes were air-dried and stained withGiemsa (Sigma), and the number of chromosomes was scored bytwo independent observers.

Collection of Oocytes from Older Females. MII-arrested oocytesfrom older mice were obtained by superovulating 10-month-oldICR female mice (Charles River Japan, Kanagawa, Japan) with5 units of pregnant mare’s serum gonadotrophin (PMSG),followed 48 h later with 5 units of human chorionic gonadotropin(hCG). Cumulus–oocyte complexes were collected from theampullae of oviducts 15 h after hCG injection, placed in Whit-ten’s medium, and freed of cumulus cells with hyaluronidase. Asa control, oocytes from 3-week-old ICR female mice (CharlesRiver Japan) were used. These younger mice had been subjectedto the same hormonal treatment as the older ones.

In some experiments, naturally ovulated oocytes instead ofsuperovulated ones were used. The vaginal smears of the femaleswere examined daily to check their estrous cycles, and naturallyovulated oocytes were obtained during the next morning of theproestrus stage. In these experiments, 5- to 7-week-old mice wereused as controls instead of 3-week-old mice.

Immunocytochemistry. The oocytes were fixed with 3.7% para-formaldehyde overnight, permeabilized with 0.5% Triton X-100for 15 min, and then immunostained with an Ab against acety-lated H4K12, H3K14, H4K8, or H4K16 (all from UpstateBiotechnology, Lake Placid, NY) for 1 h at room temperature.Ab that bound to the oocytes was probed with FITC-conjugatedanti-rabbit IgG (Jackson ImmunoResearch). To visualize micro-tubules, some oocytes were double-stained with antiacetylatedH4K12 Ab and anti-�-tubulin Ab (Sigma), which were probedwith rhodamine-conjugated anti-mouse IgG (Chemicon). DNAwas visualized by counterstaining the oocytes with diamidino-2-phenylindole (DAPI). Fluorescence was detected by using theLeica spectral confocal scanning system.

We thank H. Miki and A. Ogura (RIKEN Bioresource Center) forhelpful comments on the chromosomal analysis and T. Hata, O. Suzuki,and J. Matsuda (National Institute of Biomedical Innovation) fortechnical advice on embryo transfer. This work was supported in part byMinistry of Education, Science, and Culture Grants HD 14360164 and16045203 (to F.A.).

1. Hamatani, T., Falco, G., Carter, M. G., Akutsu, H., Stagg, C. A., Sharov, A. A.,Dudekula, D. B., VanBuren, V. & Ko, M. S. (2004) Hum. Mol. Genet. 13,2263–2278.

2. Hassold, T. & Hunt, P. (2001) Nat. Rev. Genet. 2, 280–291.3. Klein, F., Mahr, P., Galova, M., Buonomo, S. B., Michaelis, C., Nairz, K. &

Nasmyth, K. (1999) Cell 98, 91–103.4. Petes, T. D. (2001) Nat. Rev. Genet. 2, 360–369.

5. Yamada, T., Mizuno, K., Hirota, K., Kon, N., Wahls, W. P., Hartsuiker, E.,Murofushi, H., Shibata, T. & Ohta, K. (2004) EMBO J. 23, 1792–1803.

6. Reddy, K. C. & Villeneuve, A. M. (2004) Cell 118, 439–452.7. Kim, J. M., Liu, H., Tazaki, M., Nagata, M. & Aoki, F. (2003) J. Cell Biol. 162,

37–46.8. Akiyama, T., Kim, J. M., Nagata, M. & Aoki, F. (2004) Mol. Reprod. Dev. 69,

222–227.

Akiyama et al. PNAS � May 9, 2006 � vol. 103 � no. 19 � 7343

DEV

ELO

PMEN

TAL

BIO

LOG

Y

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0

9. Endo, T., Naito, K., Aoki, F., Kume, S. & Tojo, H. (2005) Mol. Reprod. Dev.71, 123–128.

10. Grunstein, M. (1997) Nature 389, 349–352.11. Mizzen, C. A. & Allis, C. D. (1998) Cell. Mol. Life Sci. 54, 6–20.12. Turner, B. M. (1998) Cell. Mol. Life Sci. 54, 21–31.13. Bachvarova, R. (1985) Dev. Biol. 1, 453–524.14. Yoshida, M., Kijima, M., Akita, M. & Beppu, T. (1990) J. Biol. Chem. 265,

17174–17179.15. Hassold, T., Jacobs, P., Kline, J., Stein, Z. & Warburton, D. (1980) Ann. Hum.

Genet. 44, 29–36.16. Fraser, L. R. & Maudlin, I. (1979) Environ. Health Perspect. 31, 141–149.17. Santalo, J., Badenas, J., Catala, V. & Egozcue, J. (1987) Hum. Reprod. 2,

717–719.18. Sakurada, K., Ishikawa, H. & Endo, A. (1996) Cytogenet. Cell Genet. 72, 46–49.19. Burke, D. J. (2000) Curr. Opin. Genet. Dev. 10, 26–31.20. Chandley, A. C., Maclean, N., Edmond, P., Fletcher, J. & Watson, G. S. (1976)

Ann. Hum. Genet. 40, 165–176.21. Yin, H., Cukurcam, S., Betzendahl, I., Adler, I. D. & Eichenlaub-Ritter, U.

(1998) Chromosoma 107, 514–522.22. Hodges, C. A., Ilagan, A., Jennings, D., Keri, R., Nilson, J. & Hunt, P. A. (2002)

Hum. Reprod. 17, 1171–1180.

23. De La Fuente, R., Viveiros, M. M., Burns, K. H., Adashi, E. Y., Matzuk, M. M.& Eppig, J. J. (2004) Dev. Biol. 275, 447–458.

24. Sakaguchi, K., Herrera, J. E., Saito, S., Miki, T., Bustin, M., Vassilev, A.,Anderson, C. W. & Appella, E. (1998) Genes Dev. 12, 2831–2841.

25. Koeller, K. M., Haggarty, S. J., Perkins, B. D., Leykin, I., Wong, J. C., Kao,M. C. & Schreiber, S. L. (2003) Chem. Biol. 10, 397–410.

26. Yuan, L., Liu, J. G., Hoja, M. R., Wilbertz, J., Nordqvist, K. & Hoog, C. (2002)Science 296, 1115–1118.

27. Cukurcam, S., Hegele-Hartung, C. & Eichenlaub-Ritter, U. (2003) Hum.Reprod. 18, 1908–1917.

28. Mailhes, J. B., Mastromatteo, C. & Fuseler, J. W. (2004) Mutat. Res. 559, 153–167.29. Roberts, R., Iatropoulou, A., Ciantar, D., Stark, J., Becker, D. L., Franks, S.

& Hardy, K. (2005) Biol. Reprod. 72, 107–118.30. Hassold, T. & Chiu, D. (1985) Hum. Genet. 70, 11–17.31. Eichenlaub-Ritter, U. (1996) Environ. Mol. Mutagen. 28, 211–236.32. Whitten, W. K. (1971) Adv. Biosci. 6, 129–139.33. Aono, N., Naganuma, T., Abe, Y., Hara, K., Sasada, H., Sato, E. & Yoshida,

H. (2003) J. Reprod. Dev. 49, 501–506.34. Quinn, P. & Begley, A. J. (1984) Aust. J. Biol. Sci. 37, 147–152.35. Lawitts, J. A. & Biggers, J. D. (1993) Methods Enzymol. 225, 153–164.36. Evans, E. P. (1979) Genetics 92, S97–S103.

7344 � www.pnas.org�cgi�doi�10.1073�pnas.0510946103 Akiyama et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 11

, 202

0