Conversion of genomic imprinting by reprogramming and ...

1 downloads 0 Views 5MB Size Report
2CHA Bio and Diostech Co Ltd., Seoul, Republic of Korea. 3Department of Cell and Developmental Biology, Max Planck Institute for Molecular Biomedicine, ...
2516

Research Article

Conversion of genomic imprinting by reprogramming and redifferentiation Min Jung Kim1, Hyun Woo Choi1, Hyo Jin Jang1, Hyung Min Chung2, Marcos J. Arauzo-Bravo3, Hans R. Scho¨ler3 and Jeong Tae Do1,* 1 Department of Animal Biotechnology, College of Animal Bioscience and Technology, Konkuk University, 120 Neungdong-ro, Gwangjin-gu, Seoul 133-702, Republic of Korea 2 CHA Bio and Diostech Co Ltd., Seoul, Republic of Korea 3 Department of Cell and Developmental Biology, Max Planck Institute for Molecular Biomedicine, Ro¨ntgenstrasse 20, 48149 Mu¨nster, Germany

*Author for correspondence ([email protected])

Journal of Cell Science

Accepted 5 March 2013 Journal of Cell Science 126, 2516–2524 ß 2013. Published by The Company of Biologists Ltd doi: 10.1242/jcs.122754

Summary Induced pluripotent stem cells (iPSCs), generated from somatic cells by overexpression of transcription factors Oct4, Sox2, Klf4 and cMyc have the same characteristics as pluripotent embryonic stem cells (ESCs). iPSCs reprogrammed from differentiated cells undergo epigenetic modification during reprogramming, and ultimately acquire a similar epigenetic state to that of ESCs. In this study, these epigenetic changes were observed in reprogramming of uniparental parthenogenetic somatic cells. The parthenogenetic pattern of imprinted genes changes during the generation of parthenogenetic maternal iPSCs (miPSCs), a process referred to as pluripotent reprogramming. We determined whether altered imprinted genes are maintained or revert to the parthenogenetic state when the reprogrammed cells are redifferentiated into specialized cell types. To address this question, we redifferentiated miPSCs into neural stem cells (miPS-NSCs) and compared them with biparental female NSCs (fNSCs) and parthenogenetic NSCs (pNSCs). We found that pluripotent reprogramming of parthenogenetic somatic cells could reset parthenogenetic DNA methylation patterns in imprinted genes, and that alterations in DNA methylation were maintained even after miPSCs were redifferentiated into miPS-NSCs. Notably, maternally methylated imprinted genes (Peg1, Peg3, Igf2r, Snrpn and Ndn), whose differentially methylated regions were fully methylated in pNSCs, were demethylated and their expression levels were found to be close to the levels in normal biparental fNSCs after reprogramming and redifferentiation. Our findings suggest that pluripotent reprogramming of parthenogenetic somatic cells followed by redifferentiation leads to changes in DNA methylation of imprinted genes and the reestablishment of gene expression levels to those of normal biparental cells. Key words: Reprogramming, Parthenogenetic cells, Imprinted genes, Induced pluripotent stem cells

Introduction Pluripotent stem cells can differentiate into all cell types of an organism and self-renew without losing their differentiation potential (Do and Scho¨ler, 2009). Induced pluripotent stem cells (iPSCs), generated from somatic cells by overexpression of transcription factors, such as Oct4, Sox2, Klf4 and c-Myc (Hochedlinger and Plath, 2009; Takahashi and Yamanaka, 2006), have the same characteristics as pluripotent embryonic stem cells (ESCs). iPSCs are indistinguishable from ESCs in morphology, differentiation potential, proliferation and teratoma formation ability. These cells have the capacity for unlimited selfrenewal and contribute to the formation of all three germ layers (Takahashi and Yamanaka, 2006). iPSCs reprogrammed from differentiated cells experience epigenetic modification during reprogramming and ultimately acquire a similar epigenetic state to that of ESCs. Epigenetic changes, including DNA methylation, histone modification and reactivation of the inactive X chromosome (Xi), control various cellular states, including proliferation, survival, pluripotency and differentiation (Meissner, 2010). Such epigenetic changes have also been observed in reprogramming of uniparental parthenogenetic somatic cells (Do et al., 2009).

Genomic imprinting is an epigenetic alteration that results in monoallelic gene expression. Therefore, imprinted genes are expressed in a monoallelic manner according to their parent-oforigin signature and are known for their roles in the regulation of fetal and/or placental growth and development. Aberrant allelespecific expression of imprinted genes disrupts fetal development, and is associated with genetic diseases, some cancers and several neurological disorders (Horsthemke and Wagstaff, 2008; Jelinic and Shaw, 2007; Khosla et al., 2001). To date, approximately more than 80 imprinted genes have been identified in mice and humans (Mitalipov, 2006). Several known paternally expressed imprinted genes are among the most downregulated genes in parthenogenetic somatic cells compared with bi-parental somatic cells (Morison et al., 2005). However, the partial loss of imprinting has been observed in undifferentiated ESCs derived from parthenogenetic embryos (Horii et al., 2008). Likewise, changes in parthenogenetic imprinting patterns have been observed during the establishment of iPSCs from parthenogenetic neural stem cells (pNSCs) (Do et al., 2009). pNSCs containing only maternal alleles regain biparental imprinting patterns after reprogramming. Thus, we have previously suggested that changes in genomic

Journal of Cell Science

Pluripotential reprogramming and genomic imprinting change

2517

imprinting could be a marker for pluripotency (Do et al., 2009). However, we have yet to define whether altered imprinted genes are maintained or reverted to the parthenogenetic state when the reprogrammed cells are differentiated into specialized cell types. Here, we sought to expand on our previous observations by analyzing genomic imprinting during reprogramming and redifferentiation processes. We compared genome-wide expression profiles of biparental female NSCs (fNSCs), pNSCs and NSCs differentiated from parthenogenetic maternal iPSCs (miPS-NSCs), and further analyzed the dynamics of methylation imprints during reprogramming and redifferentiation of pNSCs.

analysis. No difference was found in expression levels of maternally expressed imprinted genes between biparental fNSCs and pNSCs (Fig. 2A). However, the expression of several important paternally expressed imprinted genes such as Peg1 (also known as Mest), Ndn and Snurf was low in pNSCs compared to that in control fNSCs. (Fig. 2B). The reprogrammed miPSCs showed a similar imprinted gene expression pattern to that of pESCs; paternally imprinted genes, including Peg1, Ndn and Snurf, were upregulated after reprogramming (Fig. 2B). Collectively, these data suggest that miPSCs are epigenetically and genetically pluripotent, similar in state to that of pESCs.

Results

miPSCs can differentiate into stable somatic stem cell lines

Generation of parthenogenetic iPSCs

As we showed in our previous study, somatic cells derived from parthenogenetic embryos (10.5 dpc) exhibited growth retardation, parthenogenetic fibroblast cells underwent senescence, and pNSCs exhibited a slower proliferation rate than normal fNSCs (Do et al., 2009). Therefore, somatic cells differentiated from miPSCs may have the same growth defect as observed in parthenogenetic embryo-derived somatic cells, or miPSCs may not differentiate into a specialized somatic cell line. To test this, we first asked whether miPSCs could differentiate into stable somatic stem cells, NSCs, and if so, whether the differentiated cells could grow normally. Furthermore, we examined whether imprinted genes in miPS-NSCs are reverted to the pNSC state (parthenogenetic patterns) or maintained in the miPSC state (loss of parthenogenetic patterns). We were able to derive NSCs from miPSCs in a serum-free adherent monolayer culture system (Ying et al., 2003). Interestingly, these NSCs, which were differentiated from miPSCs (miPS-NSCs), were morphologically very similar to normal fNSCs (Fig. 3). Immunocytochemical staining confirmed the uniform expression of the NSC marker Nestin Sox2 and Musashi, indicating that the miPSNSCs appeared to be a highly homogeneous population (Fig. 3A). Bisulfite DNA sequencing analysis showed that the Oct4 promoter region, which was hypomethylated in miPSCs, was hypermethylated after differentiation into miPSC-NSCs (Fig. 3B). We found that miPS-NSCs could self-renew, even after repeated passaging (more than 20 passages), and grow much faster than pNSCs (Fig. 3C), indicating that characteristics of miPS-NSCs more closely reflected an fNSC-like state than a pNSC-like state. These data confirm that the miPS-NSCs, which were differentiated from miPSC cells, possess phenotypic and epigenetic characteristics of NSCs. To define the transcriptional signature of miPS-NSCs, we performed a large-scale gene expression analysis of miPS-NSCs, fNSCs and pNSCs, using the Illumina gene expression array. Hierarchical cluster analysis revealed a high degree of similarity between all NSC lines independent of their origin. Although there were slight differences in the global gene expression profiles of the three NSC lines tested, all of them were clearly distinct from mouse embryo fibroblasts (MEFs) (Fig. 4A). In Fig. 4B, we identified 26 imprinted genes expressed differently between samples from the Illumina MouseRef-8 v2 Expression BeadChip that were assigned to known imprinted genes (http://www.geneimprint.org). The expression pattern of 30 probe sets (imprinted genes) analyzed using a microarray indicated that the gene expression patterns of miPS-NSCs were more similar to those observed in fNSCs than those observed in pNSCs, indicating that imprinted genes in pNSCs underwent gene expression changes during reprogramming and redifferentiation (Fig. 4B). Moreover, fNSCs and miPS-NSCs showed similar expression levels of paternally expressed imprinted

Parthenogenetic maternal iPSCs (miPSCs) were generated from parthenogenetic neural stem cells (pNSCs; Fig. 1A), which were derived from Oct4-GFP transgenic parthenogenetic embryos (10.5 days post-coitum; dpc) (McGrath and Solter, 1984; Surani and Barton, 1983). We generated miPSCs using retroviral vectors encoding Oct4, Sox2, Klf4 and c-Myc. Reprogrammed colonies were initially identified as GFP-positive cells, and expanded by mouse ESC culture conditions. The miPSCs expressing Oct4GFP were morphologically indistinguishable from mouse ESCs (Fig. 1A). Moreover, miPSCs and parthenogenetic ESCs (pESCs) expressed similar levels of pluripotency markers, such as Nanog, Oct4 and Dnmt3b (Fig. 1B). To investigate the developmental potential of miPSCs in vitro, we determined whether miPSCs could differentiate into three germ layers by embryoid body (EB) differentiation. First of all, we determined the gene expression of EBs derived from miPSCs by RT-PCR analysis of the ectoderm marker GluR6, the endoderm marker afetoprotein and the mesoderm marker GATA4. The results showed that pESCs and miPSCs did not express the three germ layer-specific markers. However, the differentiated cells did (Fig. 1C). Immunocytochemical analysis confirmed that differentiated miPSCs expressed markers for all three germ layers; ectoderm (Tuj1+), mesoderm (Brachyury+) and endoderm (HNF3b+) (Fig. 1D). They could differentiate into ectodermal (secretory epithelium), mesodermal (cartilage) and endodermal (gut epithelium) derivatives in teratoma (Fig. 1E). Next, we evaluated the naive pluripotency marker by evaluating reactivation of the inactive X chromosome (Xi) (McBurney and Strutt, 1980). To determine X chromosome status, we conducted Xist RNA fluorescence in situ hybridization (FISH) using pNSCs, pESCs and miPSCs. A single Xi was detected as a large red signal in pNSCs, but the Xi disappeared during reprogramming (Fig. 1F). miPSCs contained two active X chromosomes (Xas), which were detected as a pinpoint signal, indicating that the Xi of pNSCs was transformed into an Xa. The Xist RNA expression pattern in miPSCs closely resembled that in pESCs (Fig. 1F). These results are consistent with those of previous studies, which demonstrated that 1 Xa and 1 Xi are present in somatic cells of female mammals, whereas both X chromosomes are active (XaXa) in female pluripotent cells (Lyon, 1961; Panning et al., 1997). Global gene expression profiles also showed that miPSCs were similar to pESCs but very different from pNSCs (Fig. 1G). Interestingly, changes in expression of genomic imprintingrelated genes were also observed during the induction of pluripotency (Fig. 2). We compared the expression levels of maternally and paternally imprinted genes by using microarray

Journal of Cell Science 126 (11)

Journal of Cell Science

2518

Fig. 1. Generation of parthenogenetic miPSCs and their pluripotent characteristics. (A) Phase-contrast images of pNSCs (left), phase-contrast images of miPSC colonies (middle) and fluorescence image of miPSC colonies (right). Scale bars: 100 mm. (B) qPCR for pluripotent genes in pNSCs, pES, miPSCs and miPSC-NSCs. Values refer to control pNSCs. (C) RT-PCR analysis of differentiation markers for the three germ layers: ectoderm (GluR6), mesoderm (GATA4) and endoderm (AFP). (D) Immunocytochemistry of miPSCs using markers for the three germ layers (ectoderm, Tuj1; mesoderm, Brachyury; endoderm, HNF3b) (E) In vivo differentiation potential of miPSCs (ectoderm, secretory epithelium; mesoderm, cartilage; endoderm, gut). Each tissue is indicated by arrow(s). (F) Xist RNA FISH analysis using pNSCs, pESCs and miPSCs. Xist RNA was detected as a large red signal in the inactive X chromosome (Xi) and as a red pinpoint signal in the active X chromosome. The Xi in miPSCs was not detected by Xist RNA FISH, indicating that the Xi of pNSCs was successfully reactivated after reprogramming. (G) Heat map of global gene expression patterns in the tested samples.

genes such as Peg1, Snrpn and Mkrn3, which were not expressed in pNSCs (Fig. 4C). Genomic imprinting patterns in miPSCs and differentiated cells (miPS-NSCs)

To investigate changes in imprinted genes during in vitro differentiation of miPSCs, we conducted DNA bisulfite

sequencing analysis to compare the DNA methylation status of paternally imprinted genes [H19 and insulin-like growth factor (Igf2)], maternally imprinted genes (Peg1, Peg3 and Igf2r), and Prader-Willi syndrome (PWS)-related maternally imprinted genes (Snrpn and Ndn.) (Fig. 5A). These seven genes were chosen because expression of these genes is developmentally regulated and they play an important role in embryonic or disease

Journal of Cell Science

Pluripotential reprogramming and genomic imprinting change

2519

Fig. 2. Expression levels of maternally and paternally imprinted genes. (A,B) Heat map of the maternally (A) and paternally (B) expressed imprinted genes in female neural stem cells (fNSCs), pNSCs, two miPSC lines (miPSC and miPSC-B), pESCs and female mouse embryo fibroblasts (fMEFs). The expression level of each gene is indicated within each box. Gene expression levels are shown on a log2 scale.

development (Andrieu et al., 2003; Cattanach et al., 1992; Dowdy et al., 2005; Poirier et al., 1991). Normal biparental fNSCs displayed differential imprinting patterns, which consisted of completely methylated and completely unmethylated alleles (see fNSCs in Fig. 5B). However, uniparental pNSCs showed methylation patterns consistent with parthenogenetic imprinting: for example, H19 and Igf2 were found to be completely unmethylated in pNSCs, whereas Peg1, Peg3, Igf2r, Snrpn and Ndn were completely methylated. However, after the reprogramming of pNSCs into miPSCs, de novo DNA methylation was observed in paternally imprinted genes (H19 and Igf2 regions) and the loss of DNA methylation was observed in maternally imprinted genes (Peg1, Igf2r, Snrpn and Ndn), excluding Peg3, indicating that the loss of parthenogenetic imprinting occurs during pluripotent reprogramming (Fig. 5). Next, we investigated whether imprinting patterns of miPSCs are reverted to the pNSC state (parthenogenetic patterns) or maintained in the miPSC state (loss of parthenogenetic patterns) when miPSCs differentiate into NSCs. miPS-NSCs showed a

similar imprinting pattern to miPSCs. The most striking change in DNA methylation pattern was observed in the PWS-related imprinted genes Snrpn and Ndn. Following reprogramming (miPSCs), genes that had been completely methylated in pNSCs (Snrpn and Ndn) showed differentially methylated patterns similar to those observed in biparental fNSCs and ESCs (Fig. 5A,B). This phenomenon was also observed in another miPSC line induced by the two factors Oct4 and Sox2 (supplementary material Fig. S1). These data clearly indicate that imprinting patterns that arise during cellular reprogramming are not reverted to the pNSC state after redifferentiation into NSCs. We next compared these results with those observed during the reprogramming of normal biparental fNSCs, and found that the methylation pattern of imprinted genes did not change dramatically (Fig. 5B). The Snrpn gene, which is expressed in fNSCs, maintained the same DNA methylation pattern during reprogramming and redifferentiation. Together with results obtained during pNSC reprogramming, the methylation pattern of Snrpn may represent the default pluripotent state.

Journal of Cell Science

2520

Journal of Cell Science 126 (11)

Fig. 3. Differentiation of miPSCs into NSCs (miPS-NSCs). (A) Fluorescence image of miPSC colonies and phase-contrast images of miPS-NSCs (scale bars: 100 mm). Immunocytochemistry of miPSC-derived NSCs (miPS-NSCs) using the NSC markers Nestin, Sox2 and Musashi (scale bar: 50 mm). (B) Bisulfite DNA sequencing of the Oct4 promoter region in pNSCs, pESCs and miPSCs. Open and filled circles indicate unmethylated and methylated CpGs, respectively. (C) Cell proliferation rate of pNSCs compared with that of fNSCs and miPS-NSCs. Shown are phase contrast images of pNSCs, fNSCs and miPS-NSCs at day 6 after culture (scale bar: 100 mm).

DNA methylation and gene expression regulation of imprinted genes during reprogramming and redifferentiation

Usually, DNA methylation leads to silencing of gene expression and DNA demethylation leads to activation of gene expression. Thus, we analyzed expression levels of imprinted genes by using real-time RT-PCR in the context of their DNA methylation status. Consistent with the results of a previous study (Li et al., 2009b), methylation of the DMRs for the paternally imprinted genes H19 and Igf2 did not correlate with the expression levels of these genes. For example, H19 and Igf2 were completely unmethylated in pNSCs, but were expressed at very low levels. Therefore, the methylation status of DMRs for H19 and Igf2 did not correlate with their expression levels in pNSCs. However, after reprogramming and redifferentiation, the expression of tested imprinted genes in miPS-NSCs was in accordance with DNA methylation status (Figs 5, 6). For example, the H19 gene in miPS-NSCs was almost completely methylated and was not expressed. In contrast to paternally imprinted genes, DNA methylation status was correlated with gene expression in the case of maternally imprinted genes (Peg1, Peg3, Snrpn and Ndn), which were completely methylated in pNSCs and were not expressed (Figs 5, 6). Maternally imprinted genes were silenced or their expression levels reduced in pNSCs, whereas these genes in miPS-NSCs were not only activated, but their expression levels were close to those observed in fNSCs. Particularly, the PWSrelated imprinted genes Snrpn and Ndn, which were not expressed in pNSCs (completely methylated), were upregulated

in miPS-NSCs (unmethylated). Therefore, the expression levels of PWS-related imprinted genes (Snrpn and Ndn) were highly correlated with their methylation status. Collectively, these data indicate that changes in DNA methylation patterns at imprinted loci during reprogramming and redifferentiation result in changes in gene expression. Discussion Pluripotent reprogramming resets the parthenogenetic imprinting patterns to the pESC state. In the present study, we further analyzed miPSCs and found that reprogrammed miPSCs could differentiate into a stable NSC line, which was very similar to normal fNSCs but not to the donor pNSCs. miPS-NSCs display a self-renewing ability and an imprinting pattern that is different from that observed for pNSCs (original somatic cell source for reprogramming), but similar to that of biparental fNSCs. Normal embryogenesis requires contributions from both maternal and paternal genomes. Thus, mammalian parthenogenetic embryos are not viable for full-term development and die from defects in genomic imprinting (McGrath and Solter, 1984; Surani et al., 1984). However, the pluripotent state of pESCs and miPSCs could generate germline chimeras. The production of normal chimeras from pESCs depends on the appropriate expression of imprinted genes dictated by epigenetic changes (Li et al., 2009a). Likewise, these epigenetic changes in miPSCs may partially explain why miPSCs can lead to the formation of germline chimeras (Do et al., 2009). The mouse Igf2 and H19 genes were among the first imprinted genes characterized in detail (Bartolomei et al., 1991; DeChiara

2521

Journal of Cell Science

Pluripotential reprogramming and genomic imprinting change

Fig. 4. Gene expression profiles of miPS-NSCs, fNSCs and pNSCs. (A) Heat map and hierarchical clustering of the global gene expression profile of miPS-NSCs, fNSCs and pNSCs. (B) Heat map of paternally and maternally imprinted genes in fNSCs, pNSCs, miPS-NSCs and MEFs. (C) Pairwise scatter plot of global gene expression patterns comparing miPS-NSCs with pNSCs and fNSCs.

et al., 1991). The reciprocal imprinting of the Igf2 and H19 genes is mechanistically coupled. H19 is maternally expressed and Igf2 is paternally expressed. Two imprinting control regions (ICRs) exist for Igf2 (DMR1 and DMR2) and both are paternally methylated (Constaˆncia et al., 2000). In mice, the receptor of Igf2, Igf2r, is maternally expressed, displaying a reciprocal pattern of imprinting to that of Igf2. Two ICRs are also known for Igf2r; the first, DMR1, is located in the Igf2r promoter region and is paternally methylated, and the second, DMR2, lies within the second intron of Igf2r and is maternally methylated (MacDonald, 2012; Wilkins and Haig, 2003). Snrpn and Ndn genes, which play important roles in embryonic development, are implicated in genetic disorders. Mice with deletions of Snrpn and Ndn show abnormal neuronal development, a phenotype also characteristic of PWS (Shemer et al., 2000). PWS is a human genetic disease that arises from the lack of expression of imprinted genes on the paternally derived chromosome 15q11–q13 (Nicholls and Knepper, 2001). PWS candidate genes, Snrpn and Ndn, are paternally expressed and maternally silenced, and are involved directly or indirectly in brain development and function. Our data

demonstrated that reprogramming of parthenogenetic somatic cells could change the DNA methylation and gene expression status of Snrpn and Ndn. Recent studies have shown that mouse germ cell markers are expressed during induced pluripotency (Xu et al., 2011) and that the Snrpn imprint is erased in fetal germ cells (Smith et al., 2011), which could be an explanation for why erasure or partial erasure of the imprint occurs in mouse miPSCs. Furthermore, alterations in DNA methylation of imprinted genes are maintained when the reprogrammed cells are differentiated into specialized cell types. Specifically, Snrpn and Ndn in miPSNSCs were not only activated, but their expression levels were also similar to those observed in normal fNSCs. Thus, it is possible that imprinting patterns could be rescued by in vitro reprogramming and redifferentiation. In vitro reprogramming and redifferentiation may be a valuable strategy for genomic imprinting studies. However, recent studies have suggested that iPSCs from PWS patients maintained their imprinting pattern even after reprogramming (Chamberlain et al., 2010; Yang et al., 2010). This discrepancy between mouse and human iPSCs may be attributable to

Journal of Cell Science 126 (11)

Journal of Cell Science

2522

Fig. 5. Comparison of DNA methylation status of paternally and maternally imprinted genes. (A,B) DNA methylation pattern of differentially methylated regions (H19, Igf2, Peg1, Peg3, Igf2r and Prader-Willi syndrome (PWS)-related imprinted genes Snrpn and Ndn) in (A) parthenogenetic cell lines (pNSCs, pESCs, miPSCs and miPS-NSCs) and (B) normal cell lines (fNSCs, ESCs, iPSCs and iPS-NSCs). Each line represents a separate clone. Black and white circles represent methylated and unmethylated CpGs, respectively.

differences in the pluripotent state of iPSCs: mouse ESCs and iPSCs appear to be in a developmentally less-advanced pluripotent state than human ESCs and iPSCs. Although mouse

and human ESCs are both derived from the ICM of the blastocyst, mouse ESCs appear to represent a developmentally ground state of pluripotency or ‘naive’ pluripotency, whereas

Fig. 6. Quantitative RT-PCR analysis of the expression of imprinted genes. The RNA levels of imprinted genes in parthenogenetic cell lines (pNSCs, pESCs, miPSCs and miPSNSCs) and normal cell lines (fNSCs and ESCs) were determined by quantitative real-time RT-PCR using primers specific for imprinted genes. The expression levels of parthenogenetic cell lines and normal cell lines were compared with those in pNSCs; expression in pNSCs was set to 1 for all samples. The transcript levels were normalized to b-actin levels. Means 6 s.e.m. are shown for three independent experiments.

Journal of Cell Science

Pluripotential reprogramming and genomic imprinting change human ESCs display a mouse epiblast stem cell (EpiSC)-like ‘primed’ post-implantation state (Nichols and Smith, 2009; Tesar et al., 2007). However, if iPSCs derived from PWS patients develop naive pluripotency under special culture conditions, imprinting patterns could be changed to be more similar to those of normal cells. This possibility was observed for Xa and Xi patterns in human ESCs. In female cells, naive pluripotent cells do not contain an Xi, but primed pluripotent cells have one Xa and one Xi (Nichols and Smith, 2009). Thus, the reactivation of the Xi in somatic cells is regarded to be a marker of complete reprogramming (to the naive pluripotent state). Xi chromosomes have also been found in undifferentiated human ESCs and iPSCs and are maintained upon the induction of differentiation (Tchieu et al., 2010). However, a few recent studies have shown two Xas in human ESCs cultured under hypoxic culture conditions (Lengner et al., 2010), in ESCs treated with a histone deacetylase inhibitor (Ware et al., 2009), and in ESCs overexpressing KLF4 (Hanna et al., 2010). Together with our findings, these data suggest that impaired imprinting genes could be altered by reprogramming and redifferentiation if naive pluripotency were induced in human somatic cells. If so, imprinting-related diseases could be cured by cell therapy. To date, many researchers have reported discoveries regarding the mechanisms involved in the establishment, maintenance and erasure of epigenetic imprints (Hajkova et al., 2010; Ito et al., 2010; Kaneda et al., 2004; Li et al., 1993; Popp et al., 2010). Nevertheless, many unanswered questions remain. Reprogramming and differentiation of parthenogenetic cells could be an excellent model system to study the mechanism of epigenetic imprinting, particularly that concerning DNA demethylation and imprint erasure.

2523

70% ethanol and stored in 70% ethanol at 4 ˚C until use in FISH. Xist RNA was detected with a Cy3-labeled Xist RNA FISH probe spanning the entire Xist cDNA. Xist RNA FISH was performed as previously described (Do et al., 2008). Images were obtained using an applied spectral imaging (ASI) camera and analyzed with FishView software (Applied Spectral Imaging; ASI GmBH). Immunofluorescence staining

Cells were fixed with 4% paraformaldehyde in PBS for 15 minutes at room temperature, washed with PBS and then incubated with blocking solution (10% FCS and 0.05% Triton X-100) for 45 minutes. Cells were incubated overnight at 4 ˚C with primary antibodies (rabbit anti-human HNF3b, mouse anti-mouse Tuj1 and goat anti-mouse Brachury; Millipore Corporation, Billerica, MA) diluted in 10% BSA in PBS. They were then washed with PBS and incubated with secondary antibodies (Alexa Fluor 594 goat anti-rabbit IgG and anti-mouse IgG2a) in 1% FCS in PBS for 1 hour at room temperature. Nuclei were counterstained with Hoechst. Whole genome expression analysis

RNA samples for microarrays were prepared using Qiagen RNeasy columns with on-column DNA digestion. Three hundred nanograms of total RNA per sample was used as starting material for the linear amplification protocol (Ambion, Austin, TX, http://www.ambion.com), which involved synthesis of T7-linked doublestranded cDNA and 12 hours of in vitro transcription incorporating biotin-labeled nucleotides. Purified and labeled cRNA was then hybridized for 18 hours onto MouseRef-8 v2 gene expression BeadChips (Illumina, Inc., San Diego, http:// www.illumina.com) following the manufacturer’s instructions. After the chips were washed, they were stained with streptavidine-Cy3 (GE Healthcare, Chalfont St. Giles, UK, http://www.gehealthcare.com) and scanned using the iScan reader and accompanying software. Samples were exclusively hybridized as biological replicates. Microarray analysis

Brain tissue was collected from 10.5-dpc parthenogenetic embryos (OG2+/2). Neurospheres were cultured as described in detail in our previous article (Do and Scho¨ler, 2004). Primary neurospheres were replated onto gelatinized dishes in NSC expansion medium. Outgrowing cells were trypsinized, replated, and cultured in NSC expansion medium. NSCs were established by dissociation and were replated onto gelatin-coated dishes in NSC expansion medium, comprising NS-A media (Euroclone, Siziano, Italy, http://www.euroclonegroup.it) supplemented with N2 supplement, 10 ng/ml of each epidermal growth factor and basic fibroblast growth factor (Invitrogen, Carlsbad, CA, http://www.invitrogen.com), 50 mg/ml BSA (Fraction V; Gibco BRL, Gaithersburg, MD, http://www.gibcob.com), 16 penicillin/streptomycin/glutamine and 16 nonessential amino acids (Gibco BRL).

The quality of hybridization and overall chip performance were monitored by visual inspection of both internal quality control checks and the raw scanned data. Raw data were extracted using the software provided by the manufacturer [Illumina GenomeStudio v2009.2 (Gene Expression Module v1.5.4)]. Array data were filtered using a detection P-value of ,0.05 (similar to signal to noise) in at least 50% of the samples. (We applied a filtering criterion for data analysis; higher signal value was required to obtain a detection P-value ,0.05.) Selected probe signal values were log transformed and normalized using the quantile method. The comparative analysis between test and control groups was carried out using the Local Pooled Error (LPE) test and fold-change. A false discovery rate (FDR) was applied to the data using the Benjamini-Hochberg algorithm. Hierarchical cluster analysis was performed using complete linkage and Euclidean distance as a measure of similarity. K-means cluster analysis was performed using differential distance as a measure of similarity. A list of maternally and paternally imprinted genes was compiled from the literature (Bartolomei and Tilghman, 1997; Coan et al., 2005; Kaneko-Ishino et al., 2003; Raefski and O’Neill, 2005). Microarray data were deposited in the GENE Expression Omnibus (GEO) and can be accessed at the following link: http://www.ncbi.nlm.nih.gov/geo/query/acc. cgi?acc5GSE37541.

Generation and culture of iPSCs

Methylation analysis

pMX-based retroviral vectors encoding the mouse cDNAs of Oct4, Sox2, Klf4 and cMyc (Takahashi and Yamanaka, 2006) were separately co-transfected by packaging defective helper plasmids into 293T cells by using Fugene 6 transfection reagent (Roche Diagnostics, Basel, Switzerland, http://www.roche-applied-science.com). Forty-eight hours post-infection, virus supernatants were collected, filtered and concentrated as previously described (Zaehres and Daley, 2006). fNSCs or pNSCs (OG2+/2) were seeded at a density of 16105 cells per six-well plate and incubated with virus-containing supernatants of the 4 factors (4F; 1:1:1:1) supplemented with 6 mg/ml of protamine sulfate (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com) for 24 hours. Cells were replated onto MMC-treated MEF feeders in ESC medium. Oct4GFP+ iPSCs were sorted by FACS and subcultured onto MMC-treated MEF feeders. miPSCs at passages 12–15 were used for further analysis.

Bisulfite treatment of DNA was performed with the EpiTect Bisulfite Kit (Qiagen) according to the manufacturer’s instructions. The bisulfite-converted DNA was amplified by nested PCR. Primer sequences are detailed in supplementary material Table S1. The outside primer pairs were used for the first-round PCR, whereas the inside primer pairs were used for the second-round PCR. The amplification consisted of a total of 45 cycles at 95 ˚C for 10 seconds, 60 ˚C for 30 seconds and 72 ˚C for 60 seconds with the first denaturation at 94 ˚C for 5 minutes and a final extension at 72 ˚C for 10 minutes. The amplified products were verified by electrophoresis on a 1% agarose gel. The desired PCR products were subcloned into the pGEM-T Easy vector (Promega, Madison, WI) and were sequenced using T7 primers. We used the BiQ Analyzer software (Max Planck Society, Germany) for the visualization and quantification of the bisulfite sequence data.

Teratoma formation in vivo

Quantitative real-time PCR analysis

miPS cells were harvested by dissociation solution treatment and washed twice with PBS. Prepared cells were injected into each testis of a severe combined immunodeficiency (SCID) mouse. After five weeks, mice were sacrificed, tumors were embedded in paraffin, and sections stained with hematoxylin and eosin and histologically analyzed.

RNA was isolated with Trizol (GIBCO) according to the manufacturer’s protocol. cDNA was synthesized from ,1 mg of total RNA by using SuperScript III reverse transcriptase (Invitrogen, Grand Island, NY). For real-time PCR, b-actin was used as a reference control. We corrected for differences in PCR efficiency between target and reference loci by using the efficiency correction within the Relative Quantification Software (Roche LC 480). Standard curves were created for each gene by using known quantities of total cDNA from other cells. Thermal cycling was carried out at 40 three-step cycles: 30 seconds at 94 ˚C, 30 seconds at 55 ˚C and 30 seconds at 72 ˚C. Real-time PCR primer sequences are listed in supplementary material Table S2.

Materials and Methods Generation of parthenogenetic NSCs

Xist RNA FISH

pNSCs and miPS cells (GFP-positive) on day 30 of culture were sorted by FACS and placed onto Roboz slides. The slides were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) for 10 minutes at room temperature, rinsed in

2524

Journal of Cell Science 126 (11)

Author contributions

M.J.K. contributed to conception and design of experiments, collection and/or assembly of data, data analysis and interpretation, manuscript writing and final approval of the manuscript; H.W.C. and H.J.J. contributed to the collection and/or assembly of data; H.M.C. and H.R.S. contributed to the provision of study material; M.J.A.-B. contributed to data analysis and interpretation; J.T.D. contributed to the conception and design of experiments, financial support, manuscript writing and final approval of the manuscript. Funding

This work was supported by the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology [grant numbers 20110019489 and 20100028247]. Supplementary material available online at http://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.122754/-/DC1

Journal of Cell Science

References Andrieu, D., Watrin, F., Niinobe, M., Yoshikawa, K., Muscatelli, F. and Fernandez, P. A. (2003). Expression of the Prader-Willi gene Necdin during mouse nervous system development correlates with neuronal differentiation and p75NTR expression. Gene Expr. Patterns 3, 761-765. Bartolomei, M. S. and Tilghman, S. M. (1997). Genomic imprinting in mammals. Annu. Rev. Genet. 31, 493-525. Bartolomei, M. S., Zemel, S. and Tilghman, S. M. (1991). Parental imprinting of the mouse H19 gene. Nature 351, 153-155. Cattanach, B. M., Barr, J. A., Evans, E. P., Burtenshaw, M., Beechey, C. V., Leff, S. E., Brannan, C. I., Copeland, N. G., Jenkins, N. A. and Jones, J. (1992). A candidate mouse model for Prader-Willi syndrome which shows an absence of Snrpn expression. Nat. Genet. 2, 270-274. Chamberlain, S. J., Chen, P. F., Ng, K. Y., Bourgois-Rocha, F., Lemtiri-Chlieh, F., Levine, E. S. and Lalande, M. (2010). Induced pluripotent stem cell models of the genomic imprinting disorders Angelman and Prader-Willi syndromes. Proc. Natl. Acad. Sci. USA 107, 17668-17673. Coan, P. M., Burton, G. J. and Ferguson-Smith, A. C. (2005). Imprinted genes in the placenta–a review. Placenta 26 Suppl. A , 10-20. Constaˆncia, M., Dean, W., Lopes, S., Moore, T., Kelsey, G. and Reik, W. (2000). Deletion of a silencer element in Igf2 results in loss of imprinting independent of H19. Nat. Genet. 26, 203-206. DeChiara, T. M., Robertson, E. J. and Efstratiadis, A. (1991). Parental imprinting of the mouse insulin-like growth factor II gene. Cell 64, 849-859. Do, J. T. and Scho¨ler, H. R. (2004). Nuclei of embryonic stem cells reprogram somatic cells. Stem Cells 22, 941-949. Do, J. T. and Scho¨ler, H. R. (2009). Regulatory circuits underlying pluripotency and reprogramming. Trends Pharmacol. Sci. 30, 296-302. Do, J. T., Han, D. W., Gentile, L., Sobek-Klocke, I., Stehling, M. and Scho¨ler, H. R. (2008). Enhanced reprogramming of Xist by induced upregulation of Tsix and Dnmt3a. Stem Cells 26, 2821-2831. Do, J. T., Joo, J. Y., Han, D. W., Arau´zo-Bravo, M. J., Kim, M. J., Greber, B., Zaehres, H., Sobek-Klocke, I., Chung, H. M. and Scho¨ler, H. R. (2009). Generation of parthenogenetic induced pluripotent stem cells from parthenogenetic neural stem cells. Stem Cells 27, 2962-2968. Dowdy, S. C., Gostout, B. S., Shridhar, V., Wu, X., Smith, D. I., Podratz, K. C. and Jiang, S. W. (2005). Biallelic methylation and silencing of paternally expressed gene 3 (PEG3) in gynecologic cancer cell lines. Gynecol. Oncol. 99, 126-134. Hajkova, P., Jeffries, S. J., Lee, C., Miller, N., Jackson, S. P. and Surani, M. A. (2010). Genome-wide reprogramming in the mouse germ line entails the base excision repair pathway. Science 329, 78-82. Hanna, J., Cheng, A. W., Saha, K., Kim, J., Lengner, C. J., Soldner, F., Cassady, J. P., Muffat, J., Carey, B. W. and Jaenisch, R. (2010). Human embryonic stem cells with biological and epigenetic characteristics similar to those of mouse ESCs. Proc. Natl. Acad. Sci. USA 107, 9222-9227. Hochedlinger, K. and Plath, K. (2009). Epigenetic reprogramming and induced pluripotency. Development 136, 509-523. Horii, T., Kimura, M., Morita, S., Nagao, Y. and Hatada, I. (2008). Loss of genomic imprinting in mouse parthenogenetic embryonic stem cells. Stem Cells 26, 79-88. Horsthemke, B. and Wagstaff, J. (2008). Mechanisms of imprinting of the PraderWilli/Angelman region. Am. J. Med. Genet. A. 146A, 2041-2052. Ito, S., D’Alessio, A. C., Taranova, O. V., Hong, K., Sowers, L. C. and Zhang, Y. (2010). Role of Tet proteins in 5mC to 5hmC conversion, ES-cell self-renewal and inner cell mass specification. Nature 466, 1129-1133. Jelinic, P. and Shaw, P. (2007). Loss of imprinting and cancer. J. Pathol. 211, 261-268. Kaneda, M., Okano, M., Hata, K., Sado, T., Tsujimoto, N., Li, E. and Sasaki, H. (2004). Essential role for de novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting. Nature 429, 900-903. Kaneko-Ishino, T., Kohda, T. and Ishino, F. (2003). The regulation and biological significance of genomic imprinting in mammals. J. Biochem. 133, 699-711.

Khosla, S., Dean, W., Reik, W. and Feil, R. (2001). Culture of preimplantation embryos and its long-term effects on gene expression and phenotype. Hum. Reprod. Update 7, 419-427. Lengner, C. J., Gimelbrant, A. A., Erwin, J. A., Cheng, A. W., Guenther, M. G., Welstead, G. G., Alagappan, R., Frampton, G. M., Xu, P., Muffat, J. et al. (2010). Derivation of pre-X inactivation human embryonic stem cells under physiological oxygen concentrations. Cell 141, 872-883. Li, E., Beard, C. and Jaenisch, R. (1993). Role for DNA methylation in genomic imprinting. Nature 366, 362-365. Li, C., Chen, Z., Liu, Z., Huang, J., Zhang, W., Zhou, L., Keefe, D. L. and Liu, L. (2009a). Correlation of expression and methylation of imprinted genes with pluripotency of parthenogenetic embryonic stem cells. Hum. Mol. Genet. 18, 2177-2187. Li, C., Chen, Z., Liu, Z., Huang, J., Zhang, W., Zhou, L., Keefe, D. L. and Liu, L. (2009b). Correlation of expression and methylation of imprinted genes with pluripotency of parthenogenetic embryonic stem cells. Hum. Mol. Genet. 18, 2177-2187. Lyon, M. F. (1961). Gene action in the X-chromosome of the mouse (Mus musculus L.). Nature 190, 372-373. MacDonald, W. A. (2012). Epigenetic mechanisms of genomic imprinting: common themes in the regulation of imprinted regions in mammals, plants, and insects. Genet. Res. Int. 2012, 585024. McBurney, M. W. and Strutt, B. J. (1980). Genetic activity of X chromosomes in pluripotent female teratocarcinoma cells and their differentiated progeny. Cell 21, 357-364. McGrath, J. and Solter, D. (1984). Completion of mouse embryogenesis requires both the maternal and paternal genomes. Cell 37, 179-183. Meissner, A. (2010). Epigenetic modifications in pluripotent and differentiated cells. Nat. Biotechnol. 28, 1079-1088. Mitalipov, S. M. (2006). Genomic imprinting in primate embryos and embryonic stem cells. Reprod. Fertil. Dev. 18, 817-821. Morison, I. M., Ramsay, J. P. and Spencer, H. G. (2005). A census of mammalian imprinting. Trends Genet. 21, 457-465. Nicholls, R. D. and Knepper, J. L. (2001). Genome organization, function, and imprinting in Prader-Willi and Angelman syndromes. Annu. Rev. Genomics Hum. Genet. 2, 153-175. Nichols, J. and Smith, A. (2009). Naive and primed pluripotent states. Cell Stem Cell 4, 487-492. Panning, B., Dausman, J. and Jaenisch, R. (1997). X chromosome inactivation is mediated by Xist RNA stabilization. Cell 90, 907-916. Poirier, F., Chan, C. T., Timmons, P. M., Robertson, E. J., Evans, M. J. and Rigby, P. W. (1991). The murine H19 gene is activated during embryonic stem cell differentiation in vitro and at the time of implantation in the developing embryo. Development 113, 1105-1114. Popp, C., Dean, W., Feng, S., Cokus, S. J., Andrews, S., Pellegrini, M., Jacobsen, S. E. and Reik, W. (2010). Genome-wide erasure of DNA methylation in mouse primordial germ cells is affected by AID deficiency. Nature 463, 1101-1105. Raefski, A. S. and O’Neill, M. J. (2005). Identification of a cluster of X-linked imprinted genes in mice. Nat. Genet. 37, 620-624. Shemer, R., Hershko, A. Y., Perk, J., Mostoslavsky, R., Tsuberi, B., Cedar, H., Buiting, K. and Razin, A. (2000). The imprinting box of the Prader-Willi/Angelman syndrome domain. Nat. Genet. 26, 440-443. Smith, E. Y., Futtner, C. R., Chamberlain, S. J., Johnstone, K. A. and Resnick, J. L. (2011). Transcription is required to establish maternal imprinting at the Prader-Willi syndrome and Angelman syndrome locus. PLoS Genet. 7, e1002422. Surani, M. A. and Barton, S. C. (1983). Development of gynogenetic eggs in the mouse: implications for parthenogenetic embryos. Science 222, 1034-1036. Surani, M. A., Barton, S. C. and Norris, M. L. (1984). Development of reconstituted mouse eggs suggests imprinting of the genome during gametogenesis. Nature 308, 548-550. Takahashi, K. and Yamanaka, S. (2006). Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, 663-676. Tchieu, J., Kuoy, E., Chin, M. H., Trinh, H., Patterson, M., Sherman, S. P., Aimiuwu, O., Lindgren, A., Hakimian, S., Zack, J. A. et al. (2010). Female human iPSCs retain an inactive X chromosome. Cell Stem Cell 7, 329-342. Tesar, P. J., Chenoweth, J. G., Brook, F. A., Davies, T. J., Evans, E. P., Mack, D. L., Gardner, R. L. and McKay, R. D. (2007). New cell lines from mouse epiblast share defining features with human embryonic stem cells. Nature 448, 196-199. Ware, C. B., Wang, L., Mecham, B. H., Shen, L., Nelson, A. M., Bar, M., Lamba, D. A., Dauphin, D. S., Buckingham, B., Askari, B. et al. (2009). Histone deacetylase inhibition elicits an evolutionarily conserved self-renewal program in embryonic stem cells. Cell Stem Cell 4, 359-369. Wilkins, J. F. and Haig, D. (2003). What good is genomic imprinting: the function of parent-specific gene expression. Nat. Rev. Genet. 4, 359-368. Xu, X., Pantakani, D. V., Lu¨hrig, S., Tan, X., Khromov, T., Nolte, J., Dressel, R., Zechner, U. and Engel, W. (2011). Stage-specific germ-cell marker genes are expressed in all mouse pluripotent cell types and emerge early during induced pluripotency. PLoS ONE 6, e22413. Yang, J., Cai, J., Zhang, Y., Wang, X., Li, W., Xu, J., Li, F., Guo, X., Deng, K., Zhong, M. et al. (2010). Induced pluripotent stem cells can be used to model the genomic imprinting disorder Prader-Willi syndrome. J. Biol. Chem. 285, 40303-40311. Ying, Q. L., Stavridis, M., Griffiths, D., Li, M. and Smith, A. (2003). Conversion of embryonic stem cells into neuroectodermal precursors in adherent monoculture. Nat. Biotechnol. 21, 183-186. Zaehres, H. and Daley, G. Q. (2006). Transgene expression and RNA interference in embryonic stem cells. Methods Enzymol. 420, 49-64.