Germline competency of parthenogeneticembryonic stem cells from immature oocytesof adult mouse ovary

Zhong Liu1, Zhe Hu1,2, Xinghua Pan3, Minshu Li2, Taiwo A. Togun4, David Tuck4,

Mattia Pelizzola5, Junjiu Huang1,6, Xiaoying Ye2, Yu Yin2, Mengyuan Liu2, Chao Li1,

Zhisheng Chen1, Fang Wang1,2, Lingjun Zhou1, Lingyi Chen2, David L. Keefe6,7 and Lin Liu2,6,∗

1School of Life Science, Sun Yat-Sen University, Guangzhou 510275, China, 2Key Laboratory of Bioactive Materials of

Ministry of Education, College of Life Sciences, Nankai University, Tianjin 300071, China, 3Department of Genetics,

Yale University School of Medicine, New Haven, CT 06519, USA, 4Computational Biology and Bioinformatics,

Department of Pathology and 5Department of Epidemiology and Public Health, Yale University School of Medicine,

New Haven, CT 06520, USA, 6Department of Obstetrics and Gynecology, University of South Florida College of

Medicine, Tampa, FL 33612, USA and 7Department of Obstetrics and Gynecology, New York University Langone

Medical Center, 550 First Avenue, New York, NY 10016, USA

Received October 6, 2010; Revised December 20, 2010; Accepted January 10, 2011

Parthenogenetic embryonic stem cells (pESCs) have been generated in several mammalian species fromparthenogenetic embryos that would otherwise die around mid-gestation. However, previous reports suggestthat pESCs derived from in vivo ovulated (IVO) mature oocytes show limited pluripotency, as evidenced bylow chimera production, high tissue preference and especially deficiency in germline competence, a criticaltest for genetic integrity and pluripotency of ESCs. Here, we report efficient generation of germline-competent pESC lines (named as IVM pESCs) from parthenogenetic embryos developed from immatureoocytes of adult mouse ovaries following in vitro maturation (IVM) and artificial activation. In contrast,pESCs derived from IVO oocytes show defective germline competence, consistent with previous reports.Further, IVM pESCs resemble more ESCs from fertilized embryos (fESCs) than do IVO pESCs on genome-wide DNA methylation and global protein profiles. In addition, IVM pESCs express higher levels of Blimp1,Lin28 and Stella, relative to fESCs, and in their embryoid bodies following differentiation. This may indicatedifferences in differentiation potentially to the germline. The mechanisms for acquisition of pluripotency andgermline competency of IVM pESCs from immature oocytes remain to be determined.

INTRODUCTION

Parthenogenetic embryonic stem cells (pESCs) can be inducedfrom parthenogenetic ‘embryos’ created by artificial activationof oocytes without fertilization and may provide an alternativeto (f)ESCs as a valuable source of autologous stem cells forregenerative medicine (1–3). Parthenogenetic ‘embryos’developed from artificial activation of oocytes, withoutpaternal genomes, do not survive beyond mid-gestation,because aberrant genomic imprinting and defective

placentation fail to support subsequent development (4–7).pESCs have been generated from parthenogenetic embryosof mice (2,8–12), monkeys (1,13) and humans (3,14–17), aswell as several other mammalian species. pESCs show exten-sive differentiation capacity in vitro in both mice and primates(18,19) and contribute to a variety of adult tissues in chimeras(9). However, the proliferation and differential potential ofpESCs or their derivatives remain controversial (9,19–22).One strict test of stem cell pluripotency routinely applied istransmission of cells through the germline of chimeric

∗To whom correspondence should be addressed at: College of Life Sciences, New Biological Station C301, Nankai University, 94 Weijin Road,Tianjin 300071, China. Tel: +86 2223500752; Fax: +86 2223500752; Email: liutelom@yahoo.com or liulin@nankai.edu.cn

# The Author 2011. Published by Oxford University Press. All rights reserved.For Permissions, please email: journals.permissions@oup.com

Human Molecular Genetics, 2011, Vol. 20, No. 7 1339–1352doi:10.1093/hmg/ddr016Advance Access published on January 14, 2011

animals (23–25). Notably, most pESC lines reported thus fargenerated from in vivo ovulated (IVO) mature oocytes exhibitlow chimera production and poor germline transmission, evenafter repeated cross-breeding (2,9–11,26), suggestive ofrestricted pluripotency. Genomic imprints are establishedduring gametogenesis and play important roles in fetalgrowth and development (27). Efficacy and safety issues cur-rently limit the use of pESCs in regenerative medicine, pre-sumably due to consequences of aberrant genomicimprinting (28,29).

Limited availability of human mature oocytes further con-strains potential application of pESCs to regenerative medi-cine (14). In vitro fertilization (IVF) clinics, however,routinely produce and discard immature oocytes as by-products of ovarian stimulation after fertility treatments,which represent up to 10–20% of those retrieved (30). Parthe-nogenetic embryos, despite limited development potential,could be produced from the in vitro maturation (IVM) ofimmature eggs left from IVF clinics or collected fromovarian tissues of patients undergoing routine oophorectomyfor benign disease or endometrial cancer (31,32). This strategywould enable patients undergoing chemotherapy or radiationtherapy (33) and women with other diseases, e.g. polycysticovarian syndrome, to have their immature oocytes at the germ-inal vesicle (GV) stage retrieved from ovaries not only to

preserve their germline and fertility (34), but also to generatehistocompatible pESCs without destruction of live embryosper se. We thought to investigate whether immature oocytesobtained from adult ovaries can produce pluripotent pESCsusing mice as a model. We found that IVM pESCs can be effi-ciently derived from immature oocytes following IVM andartificial activation by strontium, and surprisingly, the IVMpESCs exhibit high germline competency, in contrast to con-ventional IVO pESCs.

RESULTS

Parthenogenetic ‘embryos’ developed from immatureoocytes and derivation of IVM pESC lines

To generate IVM pESCs from immature oocytes, we collectedimmature oocytes at the GV stage as cumulus–oocyte com-plexes (COCs; Fig. 1A) from adult mouse ovaries [B6C3F1and C57BL/6(B6)X129F1 strains and also Nanog-EGFP trans-genic mice generated in our laboratory] (35). Oocytes werematured in vitro for 15–16 h, when they showed cumulusexpansion. Oocytes were freed from cumulus cells, and onlymature oocytes at the meiosis II (MII) stage with clear extru-sion of the first polar body (arrowheads) were chosen forparthenogenetic activation by strontium chloride, in

Figure 1. Derivation and characterization of IVM pESC lines from IVM oocytes. (A) Immature oocytes at the GV stage enclosed in compact cumulus complexes(COC) collected from an adult B6C3F1 mouse ovary exhibit cumulus expansion at the meiosis II (MII) stage following IVM for 15 h. IVM oocytes develop toparthenogenetic (PA) blastocysts following activation by strontium chloride (Sr2+). Arrowhead, polar bodies. Scale bar, 50 mm. (B) IVM pESC 1116 (C57BL/6XC3H F1) at passage 8 (P8) showing typical morphology characteristics of undifferentiated fESC BF10P9, with clear, smooth boundary around the colonieswith densely packed cells. (C and D) Like fESC BF10 cells, IVM pESCs 1116 express specific ESC markers, Oct4 (pink), SSEA-1 (pink) and alkaline phos-phatase activity (AP) (blue, D). Scale bars (B–D), 100 mm. (E) Genotyping by microsatellite analysis of IVM pESC 1116 cell line. Arrow, IVM pESC bands.fESC, fertilized embryonic stem cells; IVM pESC, parthenogenetic ESCs from IVM oocytes.

1340 Human Molecular Genetics, 2011, Vol. 20, No. 7

combination with cytochalasin D that inhibits extrusion of thesecond polar body to diploidize the genome. Some oocyteswere further confirmed at the MII stage as evidenced bychromosomes aligned on MII spindles using immunostainingand fluorescence microscopy, and showing 20 chromosomesby karyotypes. Activated oocytes were cultured in potassiumsimplex optimized medium added with amino acids(KSOMAA) and developed to blastocysts. We compared thedevelopmental capacity of embryos derived from oocytesmatured in vitro versus in vivo and found that the rate of clea-vage to two cells following activation did not differ in twohybrid F1 mouse strains (B6C3F1 and B6X129F1), althoughrates of development to blastocyst for IVM oocytes wereslightly lower than those of IVO oocytes (SupplementaryMaterial, Table S1). IVM oocytes developed to blastocystsat high rates following artificial activation (86 and 65%, forB6C3F1 and B6X129F1, respectively).

Blastocysts seeded onto mouse embryonic fibroblast (MEF)feeder layers allowed inner cell mass (ICM) outgrowth in theES medium supplemented with either knockout serum replace-ment (KSR) or fetal bovine serum (FBS). KSR improved ICMoutgrowth and generated primary ES-like colonies with higherefficiency than FBS (Supplementary Material, Table S2).Similar high efficiency was obtained to derive new cell linesfrom B6X129F1 hybrid mice using the KSR ES medium.Extension of ICM outgrowth to 10 days in primary cultureprovided optimal outcomes for isolation of IVM pESC lines(Supplementary Material, Fig. S1), whereas ICM outgrowthfor about 5 days commonly used for the isolation of IVOpESCs (12) and ESCs (36) failed to produce appropriateprimary colonies of ICM outgrowth from IVM blastocysts.In contrast, parthenogenetic embryos from IVO oocytesformed an appropriate size of primary clones at day 5, likeprimary clones of ESC from fertilized embryos (fESC). Yet,a further culture of ICM outgrowth for a total of 10 daysreduced the size of IVO primary clones which also showeda sign of differentiation. This is confirmed by our recentisolation of IVM pESCs and IVO pESCs from oocytes ofNanog-EGFP mice (35) (Supplementary Material, Fig. S2A).Notably, the efficiency for the isolation of primary clones byculture of 10 days for IVO pESCs was greatly reduced(Supplementary Material, Table S2).

We further characterized IVM pESC lines (IVM pESC 1116of B6C3F1, and IVM pESC 11191 of B6X129F1 background)and compared them with IVO pESC C3 or C2 (B6C3F1), andfESC BF10 (B6C3F1) and fESC F5 (B6X129F1) derived fromfertilized embryos. IVM pESC 1116 cell line proliferated formore than 50 passages and maintained optimal morphologywith normal karyotypes. IVM pESC 1116 and fESC BF10also showed similar typical ES cell morphology, with well-defined boundaries in round colonies, small cells, with large,clear nuclei and prominent nucleoli (Fig. 1B), and normalXX karyotypes. These cells also expressed octamer-bindingtranscription factor 4 (Oct4), stage-specific embryonicantigen 1 (SSEA-1) and alkaline phosphatase activity(Fig. 1C and D), like fESCs. Microsatellite genotyping analy-sis showed that IVM pESCs displayed homozygosity onD18Mit17 and D16Mit4 alleles, identical to C57/Bl6 mice,but heterozygosity on the D8Mit4 allele for crossover loci(Fig. 1E), consistent with heterozygosity of pESCs shown by

the genome-wide SNP analysis (2). IVM pESCs and IVOpESCs derived from Nanog-EGFP transgenic mice alsoexpressed ES pluripotent markers, Nanog and Oct4 (Sup-plementary Material, Fig. S2B).

Differentiation in vitro and in vivo of IVM pESCs

The established IVM pESCs were examined for their ability tospontaneously differentiate in vitro. Spontaneous rhythmicbeating appeared about 8–18 days after embryoid bodies(EBs) attached to 0.1% gelatin-coated dishes, consistent withdifferentiation to cardiomyocyte-like tissues (video availableupon request). Average rhythmic contractility of EBs fromIVM pESCs resembled that of fESCs (50+ 19 versus 48+17/min, n ¼ 8, respectively). EBs differentiated from IVMpESCs expressed GFAP (a marker for ectoderm and astro-cytes), MF20 (mesoderm and myocardium) and AFP (endo-derm) (Fig. 2A–C).

We evaluated differentiation capacity in vivo of IVM pESCsby injecting them subcutaneously into immunodeficient nudemice (n ¼ 8). Four weeks after injection, all mice injectedwith IVM pESCs formed teratomas (Fig. 2D and E),whereas mice injected with MEF, as controls, did not form ter-atomas (Fig. 2F). Teratomas contained all three embryonicgerm layers, including immature neural tissue, neural tube(ectoderm), muscle, cartilage (mesoderm), gland and ciliatedcolumnar epithelium (endoderm) (Fig. 2G–L). The differen-tiation capacities of IVM pESCs are similar to those of IVOpESCs (12).

Generation of germline-competent chimerasfrom IVM pESCs

To further determine developmental pluripotency in vivo, weinjected IVM pESCs of non-albino genetic background intohost blastocysts of albino background, examined for chimeraproduction and germline transmission, and compared withthose of IVO pESCs and fESCs. Regardless of early (passages5–6), middle (passage 21) and late passage (passage 54) cells,IVM pESCs 1116 of agouti background injected into albinoICR mouse blastocysts produced chimeras with high effi-ciency, based on coat color (Fig. 3A), compared with fESCs(Fig. 3B; Supplementary Material, Table S3). IVO pESCsalso produced chimeras with high efficiency, regardless ofpassage number. Inbred Balb/c blastocysts serving as hostsfor IVM pESCs produced significantly more chimeras thandid outbreed ICR blastocysts (69 versus 25%, P , 0.01).IVM pESC line 11191 injected into Balb/c blastocysts alsoresulted in high production of chimeras, comparable to fESCF5 (60 and 58%, respectively).

IVM pESCs contributed to all tissues and organs, as evi-denced by microsatellite genotyping analysis (Fig. 3C), a stan-dard test for the identification and contribution of ESCs inchimeras (12,37). Notably, contributions of IVM pESCs toboth gonads and somatic tissues of female chimeras exceededthose of male chimeras (Supplementary Material, Table S4).Chimeras produced from IVM pESCs, IVO pESCs or fESCswere mated with albino mice to test their germline compe-tence. Initially, IVM pESC 1116 hosted in ICR blastocystsgenerated chimeras with germline transmission at efficiency

Human Molecular Genetics, 2011, Vol. 20, No. 7 1341

of about 12% after several rounds of mating (Table 1).Notably, chimeras produced from IVM pESCs hosted inBalb/c blastocysts exhibited higher rates of germline trans-mission (Fig. 3D), with efficiency of 20–100% (average of55%) after only the first or second round of breeding withalbino mice. Chimeras generated from IVM pESCs ofB6X129F1 background also showed high germline trans-mission (40%) after the first-round breeding and delivery(Table 1). IVM pESC-4 from Nanog-EGFP mice injectedinto ICR blastocysts also generated germline-competentchimeras (10%).

In contrast, chimeras generated from IVO pESCs C2 or C3from B6C3F1 background hosted in either ICR or Balb/c blas-tocysts failed in germline transmission, even after repeatedmating. Notably, 30–100% of female chimeras (n ¼ 11) pro-duced from IVM pESCs, but no male IVM pESC chimeras(n ¼ 2) exhibited transmission through the germline. Com-paratively, female chimeras generated from fESCs resultedin low germline transmission (0–25%, n ¼ 5), comparedwith male fESC chimeras (67–78%, n ¼ 12). It is of interestthat it is the female chimeras with greater gonad contribution,possibly suggesting a memory for their oocyte heritage. One

Figure 2. Differentiation in vitro and in vivo of IVM pESCs. (A) Glial fibrillary acidic protein (GFAP), positive for astrocytes and ectoderm, is expressed in cellsdifferentiated from IVM pESC 1116 P10. (B) MF20, a marker for mesoderm and myocardium, is expressed in the cytoplasm of differentiated cells from IVMpESC 1116 P10. (C) a-Fetoprotein (AFP), a marker for endoderm and visceral endoderm, is expressed in the cytoplasm of differentiated cells from IVM pESC1116 P10. Nuclei in blue were stained with DAPI. (D) Teratoma (arrow) 1–2 cm in size formed in the nude mouse 4 weeks post-injection of IVM pESC 1116 P6cells. (E) Teratoma (arrow) formed from another IVM pESC line 1114 P13 under shoulder skin of nude mice. (F) No teratoma formed from MEF cells served asa negative control. (G–L) Histological sections stained with hematoxylin–eosin showing tissues of three embryonic germ layers in the teratomas differentiatedfrom IVM pESC 1116 P6 (G–I): (G) immature neural tissue, (H) muscle and (I) gland; or from 1114P13 teratoma (J–L): (J) neural tube, (K) cartilage and (L)ciliated columnar epithelium.

1342 Human Molecular Genetics, 2011, Vol. 20, No. 7

reason that female versus male chimeras from IVM pESCscould be fertile might reflect X chromosome constitution. Itis generally assumed that the inactivation of X chromosomesis associated with increased expression of Xist. We showthat Xist expression was increased following the differen-tiation of IVM pESC 1116 (P18), in contrast to IVO

pESC C3. The fESCs F1 did not show much change in theexpression of Xist following differentiation, likely becausethe male fESCs with XY chromosomes did not exhibit X inac-tivation (Supplementary Material, Fig. S3). Also, the contri-bution of IVM pESCs to gonads of the female chimerasmight be important for germline competency of IVM pESCs.

Molecular analysis of pluripotency and germlinecompetence of IVM pESCs

Mechanisms underlying the efficient production of pluripotentIVM pESCs from immature oocytes remain elusive, but likelyinvolve the alteration of genome-wide methylation ofimprinted and non-imprinted genes, histone modification ormicroRNA regulation that together influence global geneexpression (38). To determine the global DNA methylationlevels of IVM pESCs or IVO pESCs, we examined relativeDNA methylation profiles throughout the genome usingMeDIP enrichment on NimbleGen CpG island plus promoterarray and compared them with fESCs. Using the Mann–Whitney non-parametric paired test and taking significanceas unadjusted P , 0.01, only 60 of 22 621 (0.27%) regionsof interest (ROIs) were differentially methylated betweenfESCs and IVM pESCs; and 699 of 22 621 (3.1%) ROIs dif-ferentially methylated between fESCs and IVO pESCs(Fig. 4A; Supplementary Material, Tables S8–S11). TheROIs do vary from each other in length. The mean length ofthe entire 22621 ROIs examined was �1.6 kb and �97% ofall the ROIs had their lengths within 1 kb of the mean. Thegenome-wide heat map (Fig. 4A) displayed 100 randomlyselected 22 621 ROIs. Some ROIs exhibited preferentialdifferences only in specific chromosomes, e.g. chromosome16 (Fig. 4B). Observation of the methylation profiles asshown in the plots, however, showed overall similar methyl-ation patterns, suggesting, perhaps, that these differenceswere generally small. On the basis of the result of the pairedtest, IVO pESCs appear more different from fESCs thanIVM pESCs from fESCs. Of these differentially methylatedROIs, 53 (P , 0.05) or 8 (P , 0.01) were common to bothIVM pESCs versus fESCs and IVO pESCs versus fESCs(Fig. 4C). It remains to be determined whether these 53 differ-entially methylated ROIs have any meaningful functions inpESCs, compared with fESCs. In the individual comparisonbetween fESC and IVM pESC or between fESC and IVOpESC, the non-parametric Mann–Whitney paired test revealedthat none of the key pluripotency factors, Oct4, Esrrb, Sox2,Klf4, Nanog and c-Myc, were differentially methylated at the

Figure 3. Chimeras and germline transmission (GT) generated from IVMpESCs. (A) Typical chimeras produced from IVM pESC 1116 p11 injectedinto Balb/c blastocysts. (B) Chimeras produced from fESC BF10 p5 celllines injected into Balb/c blastocysts. (C) Microsatellite analysis of IVMpESC 1116 chimeras. Numbers 1–6 on the right indicate six different chi-meras with various contributions of IVM pESC (arrows on the right) injectedinto albino ICR recipient embryos detected by microsatellite primer D8Mit4(1, P5 female; 2, P6 female; 3, 4, P21 female; 5, P21 male; 6, P54 male).Numbers 7–11 on the right indicate five different chimeras with various con-tributions of IVM pESC (arrows on the right) injected into Balb/c recipientembryos detected by microsatellite primer D18Mit17 (7, P7 male; 8 and 9,P10 males; 10 and 11, P11 females). MW, molecular weight. (D) GT pups(black or agouti coat) of the first litter from mating of IVM pESC 1116 p11chimera with a albino Kunming (KM) male. (E) GT pups (black or agouticoat) from mating of fESC BF10 p5 chimera with a Balb/c albino male.

Table 1. Summary of GT from offspring of chimeras

Cell lines (P5-11) Hostblastocysts

No. of pupsfrom chimeras

No. (%) ofGT pupsTypes Mouse strain

IVM pESC B6C3F1 ICR 57 7 (12)Balb/c 20 11 (55)

B6X129F1 Balb/c 10 4 (40)IVO pESC B6C3F1 ICR 69 0

Balb/c 26 0fESC B6C3F1 Balb/c 60 52 (87)

B6X129F1 Balb/c 70 50 (71)

Human Molecular Genetics, 2011, Vol. 20, No. 7 1343

unadjusted P-value of 0.01, although an ROI onchr3:34547676–34550288, which contains Sox2, was revealedto be differentially methylated (unadjusted P-value ¼ 0.02;BH-FDR adjusted P-value ¼ 0.15) in the individual compari-son between fESC and IVO pESC.

We analyzed the expression of several key imprinted genesin oocytes and in IVM pESCs and IVO pESCs. Expression ofthe imprinted genes, H19, Igf2r and Slc38a4, was low orundetectable in both IVM and IVO oocytes. Yet, Impact washighly expressed, and Snrpn and the DNA methyltransferase

Figure 4. DNA methylation of genome-wide CpG-rich regions (ROIs). (A) Genome-wide methylation heat map of ROIs. ROIs (heat-map rows) were divided in10 bins (columns) and the average enrichment within each bin was determined. Blue and red represent enrichment and depletion in respect to input DNA. (B)MeDIP enrichment of one of the differentially methylated ROIs identified on chromosome 16 is shown (P , 0.05, Mann–Whitney paired test). (C) Differentiallymethylated ROIs. The 53 differentially methylated ROIs were identified genome-wide (P , 0.05, Mann–Whitney paired test). ROIs (heat-map rows) weredivided in 10 bins (columns) and the average MeDIP enrichment within each bin was determined. Blue and red represent enrichment and depletion inrespect to input DNA.

1344 Human Molecular Genetics, 2011, Vol. 20, No. 7

Dnmt3L were slightly increased in IVM oocytes comparedwith IVO oocytes (Fig. 5A). Expression of Igf2, Igf2r,Impact, Slc38a4 and Snrpn in IVM pESCs resembled that offESCs, although the expression of H19 was slightly higherin IVM pESCs than in fESCs (Fig. 5B), similar to ourrecent report (37). Notably, the induction of differentiationby EB formation increased the expression of H19, Igf2 andSnrpn in ESCs/pESCs in general (Supplementary Material,Fig. S4).

We further checked the methylation status of specificimprinted genes in IVM pESCs and IVO pESCs and comparedthose with fESCs, and compared with progenitors oocytes.The ICR of the maternally expressed imprinted gene H19was hypomethylated in IVM oocytes, and the methylationlevel was increased in IVM pESCs, relative to IVO pESCs,but still lower than in fESCs. Unfortunately, this gene wasnot represented on the global methylation array. The

differentially methylated regions (DMRs) of paternallyexpressed imprinted genes Snrpn were hypermethylated inIVM oocytes and showed significantly reduced methylationin IVM pESCs to levels resembling those of fESCs(Fig. 5C). Comparatively, the methylation of DMRs ofSnrpn was lower, coincident with its increased expression ofIVO pESCs.

Pluripotency of ESCs is characterized by appropriateexpression and the regulatory circuitry of key transcriptionfactors Nanog, Oct4, Sox2 and Esrrb (38–41). Consistently,these factors were highly expressed in fESCs than in MEFserved as controls. Moreover, IVO pESCs and IVM pESCsexpressed high levels of these factors similar to those offESCs, except for that Nanog was higher in IVO pESCs andSox2 relatively higher in IVM pESCs than in fESCs(Fig. 6A). Abnormally, high expression of Nanog could sup-press normal differentiation of ESCs (42).

Figure 5. Expression and DMRs of imprinted genes. (A) Relative expression levels of imprinted genes in oocytes. (B) Relative expression levels of imprintedgenes in ESCs. Shown are fESC BF10 P6 from fertilized embryos, IVO pESC C3 P9 from activated IVO oocytes and IVM pESC 1116 P7 from activated IVMoocytes. Expression levels of IVO oocytes and fESC BF10P6 cells served as controls are arbitrarily designated as 1.0, and comparison is made with controls,respectively. ∗P , 0.5; ∗∗P , 0.01. n ¼ three replicates. (C) Methylation status in the ICR of H19 and DMR1 of Snrpn from IVM oocytes, and IVM pESC1116P7, IVO pESC C3P9, compared with fESC BF10P6. Genomic DNA was treated with bisulfite, followed by PCR amplification and sequencing. Shownare cytosines for a number of independently sequenced templates (horizontal lines). Percentage indicates methylated cytosines. fESC, fertilized embryonicstem cells; IVO pESC, parthenogenetic ESCs from IVO oocytes; IVM pESC, parthenogenetic ESCs from IVM oocytes.

Human Molecular Genetics, 2011, Vol. 20, No. 7 1345

Specification of germ cell fate is fundamental in develop-ment (43). The germline competence of IVM pESCs mightbe related to intrinsic specification of germ cells. We testedwhether IVM pESCs express genes specific for germ cell spe-cification and progression, compared with those of fESCs andIVO pESCs, and particularly following differentiation intoEBs, which is commonly used for the specification and

derivation of primordial germ cells (PGCs) in vitro (44,45).Germ cell marker genes, Stella, Blimp1, Lin28, Dazl andMvh, all were expressed in fESCs, and Blimp1 and Lin28showed increased expression in EBs 3 days following differ-entiation (Fig. 6B). IVM pESCs showed increased expression(P , 0.05) of Stella, Blimp1 and Lin28, compared with thoseof fESCs, and even higher (P , 0.01) expression of Stella andBlimp1 in their EBs than those of EBs differentiated fromfESCs (Fig. 6B). Expression of Dazl, Mvh and Lin28 did notdiffer in their EBs from IVM pESCs and fESCs. In contrast,expression of these germ cell marker genes in IVO pESCsdid not increase in their EBs following differentiation induc-tion, but rather Dazl and Mvh were notably decreased intheir EBs. Expression of Lin28 was much reduced in theEBs from IVO pESCs compared with that of EBs fromfESCs (Fig. 6B).

Epigenetic alternations ultimately change proteinexpression and action (46). We performed global proteomicsanalysis of protein expression in IVM and IVO pESCs andcompared them with fESCs. High-sensitive fluorescence two-dimensional differential in-gel electrophoresis (2D-DIGE)techniques not only detect levels of protein expression, butalso post-translational modifications (47,48). By 2D-DIGE,1291–1466 protein spots were found in ESC samples fromthree repeated experiments. IVO pESCs showed 39 differen-tially expressed proteins/peptides compared with fESCs(Fig. 7A and B), whereas IVM pESCs and fESCs exhibitedonly 13 differentially expressed protein spots (Fig. 7C andD), when the average ratio value was set to 1.5 and signifi-cance defined as P , 0.05 by t-test (Supplementary Material,Table S5). Fewer proteins were down-regulated and changeswere smaller in IVM pESCs than in IVO pESCs, comparedwith fESCs, respectively. Thus, the protein profile of IVMpESCs closely resembles that of fESCs. Differentiallyexpressed protein spots were further characterized by matrix-assisted laser desorption/ionization-time-of-flight and liquidchromatography/mass spectrometry analysis and found toinvolve in cell signaling, metabolism, cytoskeleton, apoptosisand oxidative stress. Consistent with genome-wide methyl-ation data, protein profiling also detected no clear differencesin the expression of imprinted genes between IVM pESCsand fESCs.

DISCUSSION

We show that parthenogenetic pluripotent stem cells can beeffectively induced from immature oocytes of adult ovariesfollowing IVM and activation. The IVM pESCs exhibit highgermline competence, in contrast to conventional (IVO)pESCs derived from ovulated mature oocytes. Chimeras pro-duced from IVM pESCs continued germline transmission,even after several rounds of breeding, and produced fertile off-spring, demonstrating that IVM pESC-derived tissues survivelong term without visible defects.

Epigenetic modification during IVM of oocytes may play acritical role in the generation of germline-competent IVMpESCs. Production of epigenetically competent oocytes iscomplex and involves many epigenetic regulators of chroma-tin remodeling and embryo development (reviewed in 49). GV

Figure 6. Expression of genes for pluripotent stem cells and germ cell speci-fication. (A) Relative expression levels of pluripotent stem cell genes in IVMpESCs, compared with MEF, fESCs and pESCs. MEF, mouse embryonicfibroblasts. (B) Relative expression levels of genes for germ cell specificationand progression in IVM pESCs and in their EBs 3 days following differen-tiation, relative to those of MEF, fESCs and pESCs. fESC F1 P16-18 from fer-tilized embryos, pESC C3 from activated IVO oocytes and IVM pESC 1116from activated IVM oocytes. Expression levels of fESCs served as controlsare arbitrarily designated as 1.0, and comparison is made with fESC controlsor EBs differentiated from fESCs, respectively. ∗P , 0.05; ∗∗P , 0.01. n ¼3–4 replicates.

1346 Human Molecular Genetics, 2011, Vol. 20, No. 7

oocytes and cytoplasmic factors related to demethylation ofhistone H3-K9 enhance somatic cell nuclear reprogrammingafter nuclear transfer (NT) (50,51). Global gene expressionprofiling by microarray analysis demonstrates that IVMoocytes exhibit immature patterns of gene expression reminis-cent of immature GV oocytes, compared with IVO oocytes(52,53). Our data show that IVM oocytes exhibit elevatedexpression of Dnmt3L and particularly Impact. Impact isimprinted after the oocyte has grown to the final, antral folliclestage (54). DNA methyltransferase 3-like (Dnmt3L) is essen-tial for the establishment of maternal methylation imprints

during oogenesis (55,56). Overexpression of DNMT3Lincreases nuclear reprogramming (57). Together, oocytesmatured in vitro may maintain some immature state, whichfacilitates epigenetic reprogramming and derivation of IVMpESCs.

Full activation of oocytes by Sr2+ also may contribute toefficient derivation of pluripotent IVM pESCs. By mimickingsperm-induced Ca2+ oscillations during fertilization (58),Sr2+ induces Ca2+ oscillations, fully activates oocytes andenhances parthenogenetic embryo development, like fertilizedembryos (59,60). Sr2+-activated cytoplasts were used for

Figure 7. Proteomics analysis of IVM pESCs or IVO pESCs versus fESCs by 2D-DIGE. Shown are representative scanned gel images of whole-cell lysatesfrom: (A) fESC, (B) IVO pESC, (C) fESC and (D) IVM pESC. Separation of proteins was performed in the first dimension (horizontal) by IEF and, sub-sequently, in the second dimension (vertical) by SDS–PAGE. pH range used for IEF is indicated on top of the figure (3.0–10.0). Mw, molecular weight(kDa). Spots representing protein expressed differentially between IVO pESC and fESC and between IVM pESC and fESC are indicated by arrows withnumbers.

Human Molecular Genetics, 2011, Vol. 20, No. 7 1347

somatic cell NT to successfully produce cloned mice (61) andto improve NT-pESCs (10). In addition, an embryo culturecondition could affect parthenogenesis prior to isolation ofstem cells from ICM of parthenogenetic ‘blastocysts’.Embryos cultured in KSOMAA show global gene expression,patterns of genomic imprinting and embryo development,like in vivo developed embryos (60,62–64).

The ICMs of parthenogenetic ‘blastocysts’ derived fromIVM oocytes seemed to develop slower than did parthenoge-netic ‘blastocysts’ from IVO oocytes and fertilized blastocysts.ICM outgrowths were extended to about 10 days for IVM‘blastocysts’ to achieve primary colonies. ICM outgrowthfor shorter duration (5 days), which produce primary coloniesfor IVO pESCs, could not produce appropriate primary colo-nies for the isolation of IVM pESCs. Extension of culturetime for the isolation process may have allowed sufficient epi-genetic reprogramming of somatic cells to iPSCs (65,66).Likewise, prolonged in vitro culture alters expression ofimprinted genes in embryonic cells (67,68). Genome-widemethylation and protein expression profiles suggest that epige-netic modification and gene expression in both imprinted andnon-imprinted genes of IVM pESCs are similar to those offESCs. Specifically, expression of Igf2r, Snrpn, Impact, Igf2and Slc38a4 in IVM pESCs resembles that of fESCs. Appro-priate methylation and expression of Igf2r, Snrpn andU2af1-rs1 are associated with pluripotency of IVM pESCs(37). Specification and determination of PGCs requiresBlimp1 and its regulation by Lin28 (43,69,70). Blimp1 marksthe specification of PGCs at the earliest stages, whereasDazl and Mvh are expressed during progression of germcells, and Dazl regulates the pre-meiotic translation of Mvh(70–72). Stella is constantly expressed in blastocysts, andPGC specification and progression (71). These genes aremore commonly used to identify PGCs and germ cells (73).Notably, genes for germ cell development at later stagesMvh and Dazl show expression in the EBs of IVM pESCssimilar to those of fESCs. IVM, full activation of oocytes,embryo culture and/or extended ICM outgrowth collectivelymay have enhanced the efficiency of derivation of germline-competent IVM pESCs. However, the mechanisms underlyingpluripotency and germline competency of IVM pESCs fromimmature oocytes remain to be fully elucidated.

Remarkably, pESCs are able to repair damaged tissues fol-lowing transplantation (28,74). Moreover, pESCs from non-human primates restore function of neurons in experimentalParkinson’s disease and more excitingly no teratoma for-mation was observed (75,76). Parthenogenetic embryoscould be produced from IVM of immature oocytes collectedfrom ovarian tissues (31,32). If IVM pESCs can be efficientlyobtained from immature human oocytes (oocytes enclosed inCOCs or denuded oocytes in the absence of COCs), theycould open the option of banking autologous pluripotentstem cells for stem cell therapy in women.

MATERIALS AND METHODS

Oocyte IVM

All mice were maintained and cared for based on approvedanimal protocols by the Institutional Animal Care Committee.

Collection of immature oocytes from ovaries and IVM weresimilar to the procedure described previously with minormodification (77). Briefly, mice were primed with 5 IU ofpregnant mare serum gonadotrophin (PMSG, Calbiochem).COCs at the GV stage were collected by puncturing folliclesin 20 mM Hepes-buffered IVM medium 42–46 h after injec-tion of PMSG. COCs were cultured in 100 ml droplets ofIVM medium covered with mineral oil at 378C in an atmos-phere of 6.5% CO2 in humidified air for 15–16 h, whenIVM oocytes reached the metaphase II (MII) stage. TheIVM medium consisted of 95% MEM (Invitrogen), 5% FBS(Hyclone), 0.24 mM sodium pyruvate, 1.5 IU/ml human chor-ionic gonadotrophin (hCG) and 1 IU/ml PMSG.

Fertilized embryos and parthenogenetic ‘embryos’

Mice were superovulated by consecutive injections of 5 IUPMSG and 5 IU hCG 44–48 h apart. Successfully matedfemales 20–21 h after injection of 5 IU hCG were used forcollecting zygotes (day 0.5). Freshly ovulated metaphase II(IVO) oocytes were collected from oviduct ampullae 14 hafter injection of 5 IU hCG, then removed off cumulus cellsby brief exposure to 0.03% hyaluronidase and pipetting.Parthenogenetic ‘embryos’ were produced by the activationof oocytes with SrCl2 and cytochalasin D, as described pre-viously (60). Embryos were cultured in 50 ml droplets of pot-assium simplex optimized medium (KSOMAA) (78), coveredwith mineral oil at 378C in an atmosphere of 6.5% CO2 inhumidified air. IVO or IVM parthenogenetic blastocystswere obtained by culture for 96 h of activated IVO or IVMMII oocytes, respectively.

Isolation and culture of ES cells and IVM pESCs

Intact blastocysts 96 h cultured from zygotes or activatedoocytes were seeded on feeder layers of mitomycinC-treated MEF, prepared on 0.1% gelatin-treated four-wellculture dishes and cultured in the ESC medium consisting ofKnockout DMEM (Invitrogen) and 20% FBS (Hyclone) orKSR (Invitrogen), supplemented with 1000 U/ml mouseESGRO leukemia inhibitory factor (LIF, Chemicon), 0.1 mM

NEAA, 1 mM L-glutamine, 0.1 mM b-mercaptoethanol,50 IU/ml penicillin, 50 IU/ml streptomycin and 50 mM MAPkinase inhibitor PD98059 (Cell Signaling) (12). Half of themedium was changed daily, beginning from the second daywhen blastocysts were seeded. Approximately 5 days afterseeding of fertilized blastocysts and IVO blastocysts, or 10days after seeding of IVM blastocysts, ICM outgrowths weremechanically removed and disaggregated into small clumpsand reseeded on fresh feeder cells. Stable ES-like cells wereroutinely obtained after two or three passages. ESC or IVMpESC lines were passaged and cultured in the ESC mediumsupplemented with FBS but without PD98059 followingbrief digestion with 0.25% trypsin–EDTA (GIBCO). Forstorage, cell lines were kept in freezing medium consistingof 40% ES medium, 50% FBS and 10% DMSO and storedfrozen in liquid nitrogen.

1348 Human Molecular Genetics, 2011, Vol. 20, No. 7

Karyotype analysis and alkaline phosphatase assay

Metaphase chromosomes were prepared by exposing cells to0.4 mg/ml nocodazole for 2 h, followed by hypotonic treat-ment with 75 mM KCl solution, fixed with methanol:glacialacetic acid (3:1) and spread onto clean slides. Spreads werestained with Giemsa and �40 metaphase spreads examinedfrom each cell culture. Alkaline phosphatase assay was per-formed using the Vector blue kit from Vector Laboratories.

Immunocytochemistry and fluorescence microscopy

ESCs or pESCs were cultured in tissue culture grade four-wellplates for 2 days. Cells were fixed in freshly prepared parafor-maldehyde [4% w/v in phosphate-buffered saline (PBS)],rinsed with blocking buffer (3% goat serum in PBS), permea-bilized with 0.1% Triton X-100 in blocking buffer, washedand left in blocking solution for 1 h. Cells were incubatedovernight at 48C with mouse monoclonal antibody againstOct4 (Santa Cruz), SSEA-1 (DSHB, MC-480), GFAP (Invitro-gen), MF20 (kindly provided by D.A. Fischman, CornellUniversity) or AFP antibody (DAKO), diluted at 1:100 inblocking solution. Primary antibodies were omitted for con-trols. After washing, cells were incubated with appropriatesecondary antibodies conjugated to Alexa Fluor 568 (or 488)(Molecular Probes), washed and counterstained with 0.2 mg/ml Hoechst 33342 or DAPI in Vectashield mountingmedium. Fluorescence was imaged under a Leica fluorescencemicroscope.

Teratoma formation test

Approximately 2 × 106 cells were injected subcutaneouslyinto 4-week-old immunodeficient nude mice to evaluate tera-toma formation. Four weeks after injection, the resultant tera-tomas were excised, fixed in 4% paraformaldehyde, embeddedin paraffin and sectioned for histological examination.

Chimera generation, germline transmission and DNAmicrosatellite polymorphism analysis

Approximately 10–15 ESCs/pESCs were injected into blasto-coels of expanded host blastocysts (3.5 dpc) of differentstrains using a Piezo injector (12). Injected blastocysts weretransferred into uterine horns of 2.5 dpc surrogate mice1–4 h following ESC injection. Chimeras initially were ident-ified by coat color. The contribution of ESCs to various tissuesfrom chimeras was confirmed and estimated by standard DNAmicrosatellite genotyping analysis (12,24), using D8Mit4,D16Mit4 and D18Mit17 primers. Genomic DNA wasextracted from tissues collected from chimeras, recipienttissues and donor ESCs/pESCs. By screening over 36 microsa-tellite markers from MGI (http://www.informatics.jax.org),some located on different chromosomes were identified to bepolymorphic in our experiments, and therefore used for geno-typing the chimeras. Microsatellite analysis was performed bypolymerase chain reaction (PCR) amplification with primersdesigned from conserved sequences flanking each markerand subsequent electrophoresis using 15% polyacrylamidegels. The gels were silver-stained and scanned, and the

genotype was determined. Ratios of contribution of stemcells in chimeras were estimated by quantification of thedensity of the bands after subtraction of the backgroundnoise using Bio-Rad Quantity One software. Some chimeraswere randomly selected for breeding with albino strains, andfurther examined for transmission through the germline.

In vitro differentiation

ESCs/pESCs were trypsinized from six-well plates and incu-bated for 50 min once or 30 min twice to remove feedercells. Suspensions were centrifuged and ES cells transferredto a 100 cm Petri dish with the ESC medium without LIFfor 3 days, then large and round-shaped EBs were placedinto four-well plates, with three EBs per well, and culturedfor 18 days, when cardiac-like muscles beat rhythmically.Some cells from day 3 EBs were collected for RNA extractionand real-time PCR analysis of germ cell marker genes. Fix-ation, immunostaining and microscopy for characterizationof three embryonic germ layers were performed as describedabove.

Genome-wide methylation analysis by MeDIPand microarray

Genome-wide methylation was profiled by MeDIP coupledwith NimbleGen microarray. MeDIP was performed accordingto a NimbleGen protocol, with slight modification from a pre-vious study (79). Briefly, the DNA from pESCs and ESCswere digested with MseI, and the fragment size was verifiedon 2% agarose gel as 200 bp–1 kb. A monoclonal mouseanti-5-methylcytidine (Eurogentec) was used to attach thefragments containing methylated CpG, followed by proteinA agarose beads to pull down the beads:antibody complex.The pooled beads:antibody:DNA were then digested with pro-tease K and purified with phenol/chloroform/isoamyl alcohol.Each of the enriched immunoprecipitated (IP) DNAs (30 ng)obtained above and their corresponding input control (Input)were amplified with WGA2 (Sigma) and purified with QIA-quick spin PCR kit (Qiagen).

The IPs were labeled as cy5, and inputs as cy3. Probe-levelMeDIP enrichment was determined by the log2(cy5/cy3).Ratios were normalized within and between arrays using theR/Bioconductor limma library (http://www.bioconductor.org). Probe-level normalized enrichment was assigned toROIs based on the array design provided by NimbleGen.ROIs are designed to spam across CpG-rich regions, includinggene promoter and distal CpG islands. For the selection ofdifferential methylated ROIs, the Mann–Whitney non-parametric paired test was used. As a result, 53 ROIs with P, 0.05 on the comparison of fESCs with both IVM pESCsreplicated experiments and the comparison of two replicatesof fESCs and IVO pESCs were identified. Heat maps werecreated using the GPLOT (http://www.astro.rug.nl/~gipsy/gplot/profile2dset.html). ROI-specific plots were created bycustom R scripts. On average, about 15 probes made upeach ROI. Using the midpoints of the genomic locations ofall probes within an ROI, the ROI was subdivided into 10bins and the MeDIP enrichment of probes within each binwas averaged. Each row on the heat map therefore represented

Human Molecular Genetics, 2011, Vol. 20, No. 7 1349

an ROI divided in 10 bins (columns), each representing theaverage MeDIP enrichment within it.

Re-defining ROIs on the different platforms: the originalexperiment was designed to compare IVM pESCs andfESCs on the 380 K platform, and the ROIs were definedusing MM8 annotation. The new experiment compared IVOpESCs and fESCs on a 700 K platform using MM9-annotatedROIs. The MM8-annotated ROIs and probes were convertedto MM9 using the UCSC genome lift-over tool. Originally,the experiment on the 380 K platform included 23 425 ROIs(357 296 probes); after the genome lift-over, 22 621 ROIs(340 491 probes) were retained for analysis of IVM pESCsversus fESCs. Using the MM9-annotated 22 621 ROIs, theprobes on the 700 K platform were mapped to appropriateROIs based on genomic location. The same 22 621 ROIs(340 519 probes) were used in the comparison analysis ofIVO pESCs versus fESCs.

Real-time PCR

Total RNA was isolated from oocytes (50 per sample), fESCs/pESCs using RNeasy micro kit (Qiagen). Extracted RNA wasquantified using a spectrophotometer (Eppendorf), andsamples from which more than 1 mg of RNA was extractedwere subjected to cDNA synthesis using Reverse Transcrip-tion System. Total RNA and PCR products were separatedon a 2% agarose gel, stained with ethidium bromide, visual-ized and photographed on a UV transluminator. Primerswere designed using GeneTool software (SupplementaryMaterial, Table S6). Each sample was analyzed in triplicatewith b-actin as the internal control. The final PCR reactionvolume of 20 ml contained 10 ml SYBR Green PCR MasterMix (Realtime Master Mix, Toyobo), 1 ml cDNA template,1 ml primer mixture and 8 ml water. Thermal cycling wascarried out with a 5 min denaturation step at 948C, followedby 40 two-step cycles: 30 s at 948C, 30 s at 608C and 30 s at728C. Amplification data were collected by the ABI PRISM7900 and analyzed using the Sequence Detection System 2.0software (ABI) or collected and analyzed by the iCycler iQ52.0 Standard Edition Optical System and the Software V2(Bio-Rad, Hercules, CA, USA).

DNA methylation by bisulfite sequencing

Genomic DNA was extracted from samples of ESCs/pESCs oroocytes using QIAamp DNA Micro Kit or DNeasy Tissue Kit(Qiagen) according to the manufacturer’s instructions. Bisul-fite treatment of DNA was performed with the EpiTect Bisul-fite Kit (Qiagen). Bisulfite converted DNA was amplified byseminested PCR, using HotStarTaq DNA Polymerase(Qiagen). Primer sequences are detailed in SupplementaryMaterial, Table S7. The outside primer pairs were used forthe first-round PCR, whereas the inside primer pairs wereused for the second-round PCR. Thermal cycling wascarried out with a 10 min denaturation step at 948C, followedby 30–40 three-step cycles (30 s at 948C, 30 s at 55–588C and30 s at 728C) and final incubation at 728C for 10 min. For thesecond round of PCR, 1 ml of the first-round sample was usedand the conditions for the PCR were the same as above. PCRproducts were recovered from stained gels (QIAquick Gel

Extraction Kit, Qiagen), cloned into a pGEM-T Easy vector(Promega) and then sequenced using T7 or SP6 primersusing an ABI 3730 capillary genetic analyzer (Applied Biosys-tems) by BigDye terminator sequencing chemistry. Bisulfiteefficiency as the fraction of modified cytosines in non-CpGsequences exceeded 98%.

Statistical analysis

Percentages were transformed using arcsin transformation.Transformed percentage data and real-time PCR data wereanalyzed by ANOVA and means compared by Fisher’s pro-tected least-significant difference using the StatView softwarefrom SAS Institute Inc. (Cary, NC, USA).

Proteomics analysis by 2D-DIGE

Methods are detailed in Supplementary Material.

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG online.

ACKNOWLEDGEMENTS

We thank Dr Rudolf Jaenisch for encouraging advice onchimera and germline transmission experiments, and HaojiaWu, Keguo Li, Qifeng Lyu and Xiaoling Meng for technicalassistance.

Conflict of Interest statement. None declared.

FUNDING

This study was supported by grants Science and TechnologyDivision of Guangzhou, China (2005Z-E0141), ChinaNational Nature Science Foundation; MOST Major ScientificResearch Project (2009CB941000) (L.L.), USF Ob/GynDepartmental Stem Cell Research Fund (D.L.K.) and NIHR01 AG23111 and Yale University CT Innovations StemCell Grant #06SCE01 (X.P.).

REFERENCES

1. Cibelli, J.B., Grant, K.A., Chapman, K.B., Cunniff, K., Worst, T., Green,H.L., Walker, S.J., Gutin, P.H., Vilner, L., Tabar, V. et al. (2002)Parthenogenetic stem cells in nonhuman primates. Science, 295, 819.

2. Kim, K., Lerou, P., Yabuuchi, A., Lengerke, C., Ng, K., West, J., Kirby,A., Daly, M.J. and Daley, G.Q. (2007) Histocompatible embryonic stemcells by parthenogenesis. Science, 315, 482–486.

3. De Sousa, P.A. and Wilmut, I. (2007) Human parthenogenetic embryostem cells: appreciating what you have when you have it. Cell Stem Cell,1, 243–244.

4. Barton, S.C., Surani, M.A. and Norris, M.L. (1984) Role of paternal andmaternal genomes in mouse development. Nature, 311, 374–376.

5. McGrath, J. and Solter, D. (1984) Completion of mouse embryogenesisrequires both the maternal and paternal genomes. Cell, 37, 179–183.

6. Surani, M.A., Barton, S.C. and Norris, M.L. (1984) Development ofreconstituted mouse eggs suggests imprinting of the genome duringgametogenesis. Nature, 308, 548–550.

1350 Human Molecular Genetics, 2011, Vol. 20, No. 7

7. Sturm, K.S., Flannery, M.L. and Pedersen, R.A. (1994) Abnormaldevelopment of embryonic and extraembryonic cell lineages inparthenogenetic mouse embryos. Dev. Dyn., 201, 11–28.

8. Kaufman, M.H., Robertson, E.J., Handyside, A.H. and Evans, M.J. (1983)Establishment of pluripotential cell lines from haploid mouse embryos.J. Embryol. Exp. Morphol., 73, 249–261.

9. Allen, N.D., Barton, S.C., Hilton, K., Norris, M.L. and Surani, M.A.(1994) A functional analysis of imprinting in parthenogenetic embryonicstem cells. Development, 120, 1473–1482.

10. Hikichi, T., Wakayama, S., Mizutani, E., Takashima, Y., Kishigami, S.,Van Thuan, N., Ohta, H., Thuy Bui, H., Nishikawa, S. and Wakayama, T.(2007) Differentiation potential of parthenogenetic embryonic stem cellsis improved by nuclear transfer. Stem Cells, 25, 46–53.

11. Jiang, H., Sun, B., Wang, W., Zhang, Z., Gao, F., Shi, G., Cui, B., Kong,X., He, Z., Ding, X. et al. (2007) Activation of paternally expressedimprinted genes in newly derived germline-competent mouseparthenogenetic embryonic stem cell lines. Cell Res., 17, 792–803.

12. Chen, Z., Liu, Z., Huang, J., Amano, T., Li, C., Cao, S., Wu, C., Liu, B.,Zhou, L., Carter, M.G. et al. (2009) Birth of parthenote mice directly fromparthenogenetic embryonic stem cells. Stem Cells, 27, 2136–2145.

13. Dighe, V., Clepper, L., Pedersen, D., Byrne, J., Ferguson, B., Gokhale, S.,Penedo, M.C., Wolf, D. and Mitalipov, S. (2008) Heterozygousembryonic stem cell lines derived from nonhuman primate parthenotes.Stem Cells, 26, 756–766.

14. Kim, K., Ng, K., Rugg-Gunn, P.J., Shieh, J.H., Kirak, O., Jaenisch, R.,Wakayama, T., Moore, M.A., Pedersen, R.A. and Daley, G.Q. (2007)Recombination signatures distinguish embryonic stem cells derived byparthenogenesis and somatic cell nuclear transfer. Cell Stem Cell, 1,346–352.

15. Mai, Q., Yu, Y., Li, T., Wang, L., Chen, M.J., Huang, S.Z., Zhou, C. andZhou, Q. (2007) Derivation of human embryonic stem cell lines fromparthenogenetic blastocysts. Cell Res., 17, 1008–1019.

16. Revazova, E.S., Turovets, N.A., Kochetkova, O.D., Kindarova, L.B.,Kuzmichev, L.N., Janus, J.D. and Pryzhkova, M.V. (2007) Patient-specificstem cell lines derived from human parthenogenetic blastocysts. Cloning

Stem Cells, 9, 432–449.17. Lin, G., OuYang, Q., Zhou, X., Gu, Y., Yuan, D., Li, W., Liu, G., Liu, T.

and Lu, G. (2007) A highly homozygous and parthenogenetic humanembryonic stem cell line derived from a one-pronuclear oocyte followingin vitro fertilization procedure. Cell Res., 17, 999–1007.

18. Vrana, K.E., Hipp, J.D., Goss, A.M., McCool, B.A., Riddle, D.R., Walker,S.J., Wettstein, P.J., Studer, L.P., Tabar, V., Cunniff, K. et al. (2003)Nonhuman primate parthenogenetic stem cells. Proc. Natl Acad. Sci. USA,100(Suppl. 1), 11911–11916.

19. Lin, H., Lei, J., Wininger, D., Nguyen, M.T., Khanna, R., Hartmann, C.,Yan, W.L. and Huang, S.C. (2003) Multilineage potential of homozygousstem cells derived from metaphase II oocytes. Stem Cells, 21, 152–161.

20. Newman-Smith, E.D. and Werb, Z. (1995) Stem cell defects inparthenogenetic peri-implantation embryos. Development, 121,2069–2077.

21. Jagerbauer, E.M., Fraser, A., Herbst, E.W., Kothary, R. and Fundele, R.(1992) Parthenogenetic stem cells in postnatal mouse chimeras.Development, 116, 95–102.

22. Hernandez, L., Kozlov, S., Piras, G. and Stewart, C.L. (2003) Paternal andmaternal genomes confer opposite effects on proliferation, cell-cyclelength, senescence, and tumor formation. Proc. Natl Acad. Sci. USA, 100,13344–13349.

23. Daley, G.Q., Lensch, M.W., Jaenisch, R., Meissner, A., Plath, K. andYamanaka, S. (2009) Broader implications of defining standards for thepluripotency of iPSCs. Cell Stem Cell, 4, 200–201; author reply 202.

24. Nichols, J., Jones, K., Phillips, J.M., Newland, S.A., Roode, M.,Mansfield, W., Smith, A. and Cooke, A. (2009) Validatedgermline-competent embryonic stem cell lines from nonobese diabeticmice. Nat. Med., 15, 814–818.

25. Han, J., Yuan, P., Yang, H., Zhang, J., Soh, B.S., Li, P., Lim, S.L., Cao, S.,Tay, J., Orlov, Y.L. et al. (2010) Tbx3 improves the germ-linecompetency of induced pluripotent stem cells. Nature, 463, 1096–1100.

26. Narasimha, M., Barton, S.C. and Surani, M.A. (1997) The role of thepaternal genome in the development of the mouse germ line. Curr. Biol.,7, 881–884.

27. Bartolomei, M.S. (2003) Epigenetics: role of germ cell imprinting. Adv.

Exp. Med. Biol., 518, 239–245.

28. Eckardt, S., Leu, N.A., Bradley, H.L., Kato, H., Bunting, K.D. andMcLaughlin, K.J. (2007) Hematopoietic reconstitution with androgeneticand gynogenetic stem cells. Genes Dev., 21, 409–419.

29. Lengerke, C., Kim, K., Lerou, P. and Daley, G.Q. (2007) Differentiationpotential of histocompatible parthenogenetic embryonic stem cells. Ann.

NY Acad. Sci., 1106, 209–218.30. Nogueira, D., Staessen, C., Van de Velde, H. and Van Steirteghem, A.

(2000) Nuclear status and cytogenetics of embryos derived from invitro-matured oocytes. Fertil. Steril., 74, 295–298.

31. Lee, H.J. and Teixeira, J. (2009) Parthenogenesis in human oocytes thatwere collected from resected ovarian tissue and matured in vitro. StemCells Dev., 18, 941–946.

32. McElroy, S.L., Byrne, J.A., Chavez, S.L., Behr, B., Hsueh, A.J.,Westphal, L.M. and Pera, R.A. (2010) Parthenogenic blastocysts derivedfrom cumulus-free in vitro matured human oocytes. PLoS One, 5, e10979.

33. Demirtas, E., Elizur, S.E., Holzer, H., Gidoni, Y., Son, W.Y., Chian, R.C.and Tan, S.L. (2008) Immature oocyte retrieval in the luteal phase topreserve fertility in cancer patients. Reprod. Biomed. Online, 17,520–523.

34. Chian, R.C., Gilbert, L., Huang, J.Y., Demirtas, E., Holzer, H., Benjamin,A., Buckett, W.M., Tulandi, T. and Tan, S.L. (2009) Live birth aftervitrification of in vitro matured human oocytes. Fertil. Steril., 91,372–376.

35. Huang, J., Deng, K., Wu, H., Liu, Z., Chen, Z., Cao, S., Zhou, L., Ye, X.,Keefe, D.L. and Liu, L. (2008) Efficient production of mice fromembryonic stem cells injected into four- or eight-cell embryos by piezomicromanipulation. Stem Cells, 26, 1883–1890.

36. Nagy, A., Gertsenstein, M., Vintersten, K. and Behringer, R. (2003)Manipulating the Mouse Embryo—A Laboratory Manual, 3rd edn. ColdSpring Harbor Laboratory Press, Cold Spring Harbor, New York.

37. Li, C., Chen, Z., Liu, Z., Huang, J., Zhang, W., Zhou, L., Keefe, D.L. andLiu, L. (2009) Correlation of expression and methylation of imprintedgenes with pluripotency of parthenogenetic embryonic stem cells. Hum.

Mol. Genet., 18, 2177–2187.38. Jaenisch, R. and Young, R. (2008) Stem cells, the molecular circuitry of

pluripotency and nuclear reprogramming. Cell, 132, 567–582.39. Navarro, P., Chambers, I., Karwacki-Neisius, V., Chureau, C., Morey, C.,

Rougeulle, C. and Avner, P. (2008) Molecular coupling of Xist regulationand pluripotency. Science, 321, 1693–1695.

40. Masui, S., Nakatake, Y., Toyooka, Y., Shimosato, D., Yagi, R.,Takahashi, K., Okochi, H., Okuda, A., Matoba, R., Sharov, A.A. et al.(2007) Pluripotency governed by Sox2 via regulation of Oct3/4 expressionin mouse embryonic stem cells. Nat. Cell Biol., 9, 625–635.

41. Zhang, X., Zhang, J., Wang, T., Esteban, M.A. and Pei, D. (2008) Esrrbactivates Oct4 transcription and sustains self-renewal and pluripotency inembryonic stem cells. J. Biol. Chem., 283, 35825–35833.

42. Chen, L., Yang, M., Dawes, J. and Khillan, J.S. (2007) Suppression of EScell differentiation by retinol (vitamin A) via the overexpression ofNanog. Differentiation, 75, 682–693.

43. Kurimoto, K., Yabuta, Y., Ohinata, Y., Shigeta, M., Yamanaka, K. andSaitou, M. (2008) Complex genome-wide transcription dynamicsorchestrated by Blimp1 for the specification of the germ cell lineage inmice. Genes Dev., 22, 1617–1635.

44. Clark, A.T., Bodnar, M.S., Fox, M., Rodriquez, R.T., Abeyta, M.J., Firpo,M.T. and Pera, R.A. (2004) Spontaneous differentiation of germ cellsfrom human embryonic stem cells in vitro. Hum. Mol. Genet., 13, 727–739.

45. Wei, W., Qing, T., Ye, X., Liu, H., Zhang, D., Yang, W. and Deng, H.(2008) Primordial germ cell specification from embryonic stem cells.PLoS One, 3, e4013.

46. Selbach, M., Schwanhausser, B., Thierfelder, N., Fang, Z., Khanin, R. andRajewsky, N. (2008) Widespread changes in protein synthesis induced bymicroRNAs. Nature, 455, 58–63.

47. Tannu, N.S. and Hemby, S.E. (2006) Methods for proteomics inneuroscience. Prog. Brain Res., 158, 41–82.

48. Wu, T.L. (2006) Two-dimensional difference gel electrophoresis.Methods Mol. Biol., 328, 71–95.

49. Bromfield, J., Messamore, W. and Albertini, D.F. (2008) Epigeneticregulation during mammalian oogenesis. Reprod. Fertil. Dev., 20, 74–80.

50. Gao, S., Gasparrini, B., McGarry, M., Ferrier, T., Fletcher, J., Harkness,L., De Sousa, P. and Wilmut, I. (2002) Germinal vesicle material isessential for nucleus remodeling after nuclear transfer. Biol. Reprod., 67,928–934.

Human Molecular Genetics, 2011, Vol. 20, No. 7 1351

51. Bui, H.T., Wakayama, S., Kishigami, S., Kim, J.H., Van Thuan, N. andWakayama, T. (2008) The cytoplasm of mouse germinal vesicle stageoocytes can enhance somatic cell nuclear reprogramming. Development,135, 3935–3945.

52. Jones, G.M., Cram, D.S., Song, B., Magli, M.C., Gianaroli, L.,Lacham-Kaplan, O., Findlay, J.K., Jenkin, G. and Trounson, A.O. (2008)Gene expression profiling of human oocytes following in vivo or in vitromaturation. Hum. Reprod., 23, 1138–1144.

53. Wells, D. and Patrizio, P. (2008) Gene expression profiling of humanoocytes at different maturational stages and after in vitro maturation.Am. J. Obstet. Gynecol., 198, 455.e1–455.e11.

54. Obata, Y. and Kono, T. (2002) Maternal primary imprinting is establishedat a specific time for each gene throughout oocyte growth. J. Biol. Chem.,277, 5285–5289.

55. Schaefer, C.B., Ooi, S.K., Bestor, T.H. and Bourc’his, D. (2007)Epigenetic decisions in mammalian germ cells. Science, 316, 398–399.

56. Hata, K., Okano, M., Lei, H. and Li, E. (2002) Dnmt3L cooperates withthe Dnmt3 family of de novo DNA methyltransferases to establishmaternal imprints in mice. Development, 129, 1983–1993.

57. Gokul, G., Ramakrishna, G. and Khosla, S. (2009) Reprogramming ofHeLa cells upon DNMT3L overexpression mimics carcinogenesis.Epigenetics, 4, 322–329.

58. Swann, K. and Ozil, J.P. (1994) Dynamics of the calcium signal thattriggers mammalian egg activation. Int. Rev. Cytol., 152, 183–222.

59. Bos-Mikich, A., Swann, K. and Whittingham, D.G. (1995) Calciumoscillations and protein synthesis inhibition synergistically activate mouseoocytes. Mol. Reprod. Dev., 41, 84–90.

60. Liu, L., Trimarchi, J.R. and Keefe, D.L. (2002) Haploidy but notparthenogenetic activation leads to increased incidence of apoptosis inmouse embryos. Biol. Reprod., 66, 204–210.

61. Wakayama, T., Perry, A.C., Zuccotti, M., Johnson, K.R. andYanagimachi, R. (1998) Full-term development of mice from enucleatedoocytes injected with cumulus cell nuclei. Nature, 394, 369–374.

62. Doherty, A.S., Mann, M.R., Tremblay, K.D., Bartolomei, M.S. andSchultz, R.M. (2000) Differential effects of culture on imprinted H19expression in the preimplantation mouse embryo. Biol. Reprod., 62,1526–1535.

63. Rinaudo, P. and Schultz, R.M. (2004) Effects of embryo culture on globalpattern of gene expression in preimplantation mouse embryos.Reproduction, 128, 301–311.

64. Liu, L., Czerwiec, E. and Keefe, D.L. (2004) Effect of ploidy and parentalgenome composition on expression of Oct-4 protein in mouse embryos.Gene Expr. Patterns, 4, 433–441.

65. Takahashi, K. and Yamanaka, S. (2006) Induction of pluripotent stemcells from mouse embryonic and adult fibroblast cultures by definedfactors. Cell, 126, 663–676.

66. Hanna, J., Markoulaki, S., Schorderet, P., Carey, B.W., Beard, C., Wernig,M., Creyghton, M.P., Steine, E.J., Cassady, J.P., Foreman, R. et al. (2008)

Direct reprogramming of terminally differentiated mature B lymphocytesto pluripotency. Cell, 133, 250–264.

67. Dean, W., Bowden, L., Aitchison, A., Klose, J., Moore, T., Meneses, J.J.,Reik, W. and Feil, R. (1998) Altered imprinted gene methylation andexpression in completely ES cell-derived mouse fetuses: association withaberrant phenotypes. Development, 125, 2273–2282.

68. Humpherys, D., Eggan, K., Akutsu, H., Hochedlinger, K., Rideout,W.M. 3rd, Biniszkiewicz, D., Yanagimachi, R. and Jaenisch, R. (2001)Epigenetic instability in ES cells and cloned mice. Science, 293, 95–97.

69. Ohinata, Y., Payer, B., O’Carroll, D., Ancelin, K., Ono, Y., Sano, M.,Barton, S.C., Obukhanych, T., Nussenzweig, M., Tarakhovsky, A. et al.

(2005) Blimp1 is a critical determinant of the germ cell lineage in mice.Nature, 436, 207–213.

70. West, J.A., Viswanathan, S.R., Yabuuchi, A., Cunniff, K., Takeuchi, A.,Park, I.H., Sero, J.E., Zhu, H., Perez-Atayde, A., Frazier, A.L. et al.

(2009) A role for Lin28 in primordial germ-cell development andgerm-cell malignancy. Nature, 460, 909–913.

71. Hayashi, K., de Sousa Lopes, S.M. and Surani, M.A. (2007) Germ cellspecification in mice. Science, 316, 394–396.

72. Reynolds, N., Collier, B., Maratou, K., Bingham, V., Speed, R.M.,Taggart, M., Semple, C.A., Gray, N.K. and Cooke, H.J. (2005) Dazl bindsin vivo to specific transcripts and can regulate the pre-meiotic translationof Mvh in germ cells. Hum. Mol. Genet., 14, 3899–3909.

73. Saiti, D. and Lacham-Kaplan, O. (2007) Mouse germ cell developmentin-vivo and in-vitro. Biomark. Insights, 2, 241–252.

74. Koh, C.J., Delo, D.M., Lee, J.W., Siddiqui, M.M., Lanza, R.P., Soker, S.,Yoo, J.J. and Atala, A. (2009) Parthenogenesis-derived multipotent stemcells adapted for tissue engineering applications. Methods, 47, 90–97.

75. Sanchez-Pernaute, R., Studer, L., Ferrari, D., Perrier, A., Lee, H., Vinuela,A. and Isacson, O. (2005) Long-term survival of dopamine neuronsderived from parthenogenetic primate embryonic stem cells (cyno-1) aftertransplantation. Stem Cells, 23, 914–922.

76. Sanchez-Pernaute, R., Lee, H., Patterson, M., Reske-Nielsen, C.,Yoshizaki, T., Sonntag, K.C., Studer, L. and Isacson, O. (2008)Parthenogenetic dopamine neurons from primate embryonic stem cellsrestore function in experimental Parkinson’s disease. Brain, 131,2127–2139.

77. Eppig, J.J. and Telfer, E.E. (1993) Isolation and culture of oocytes.Methods Enzymol., 225, 77–84.

78. Ho, Y., Wigglesworth, K., Eppig, J.J. and Schultz, R.M. (1995)Preimplantation development of mouse embryos in KSOM: augmentationby amino acids and analysis of gene expression. Mol. Reprod. Dev., 41,232–238.

79. Weber, M., Davies, J.J., Wittig, D., Oakeley, E.J., Haase, M., Lam, W.L.and Schubeler, D. (2005) Chromosome-wide and promoter-specificanalyses identify sites of differential DNA methylation in normal andtransformed human cells. Nat. Genet., 37, 853–862.

1352 Human Molecular Genetics, 2011, Vol. 20, No. 7