EMBRYONIC STEM CELLS/INDUCED PLURIPOTENT STEM CELLS Neural Stem Cells Differentiated From iPS Cells Spontaneously Regain Pluripotency HYUN WOO CHOI, JONG SOO KIM, SOL CHOI, YEAN JU HONG, MIN JUNG KIM, HAN GEUK SEO, JEONG TAE DO Key Words. Pluripotent stem cells • Induced pluripotent stem cells • Reprogramming • Neural stem cells • Exogenous factors
Department of Animal Biotechnology, College of Animal Bioscience and Technology, Konkuk University, Seoul, Republic of Korea Correspondence: Jeong Tae Do, Ph.D., Department of Animal Biotechnology, College of Animal Bioscience and Technology, Konkuk University, Seoul 143-701, Republic of Korea. Telephone: 82-2-4503673; Fax: 82-2-455-1044; e-mail: [email protected] Received November 28, 2013; accepted for publication May 6, 2014; first published online in STEM CELLS EXPRESS June 4, 2014. C AlphaMed Press V
ABSTRACT Differentiated somatic cells can be reprogrammed into pluripotent stem cells by transduction of exogenous reprogramming factors. After induced pluripotent stem (iPS) cells are established, exogenous genes are silenced. In the pluripotent state, retroviral genes integrated in the host genome are kept inactive through epigenetic transcriptional regulation. In this study, we tried to determine whether exogenous genes remain silenced or are reactivated upon loss of pluripotency or on differentiation using an in vitro system. We induced differentiation of iPS cells into neural stem cells (NSCs) in vitro; the NSCs appeared morphologically indistinguishable from brain-derived NSCs and stained positive for the NSC markers Nestin and Sox2. These iPS cellderived NSCs (iPS-NSCs) were also capable of differentiating into all three neural subtypes. Interestingly, iPS-NSCs spontaneously formed aggregates on long-term culture and showed reactivation of the Oct4-GFP marker, which was followed by the formation of embryonic stem cell-like colonies. The spontaneously reverted green fluorescent protein (GFP)-positive (iPS-NSCGFP1) cells expressed high levels of pluripotency markers (Oct4 and Nanog) and formed germline chimeras, indicating that iPS-NSC-GFP1 cells had the same pluripotency as the original iPS cells. The reactivation of silenced exogenous genes was tightly correlated with the downregulation of DNA methyltransferases (Dnmts) during differentiation of iPS cells. This phenomenon was not observed in doxycycline-inducible iPS cells, where the reactivation of exogenous genes could be induced only by doxycycline treatment. These results indicate that pluripotency can be regained through reactivation of exogenous genes, which is associated with dynamic change of Dnmt levels during differentiation of iPS cells. STEM CELLS 2014;32:2596–2604
1066-5099/2014/$30.00/0 http://dx.doi.org/ 10.1002/stem.1757
INTRODUCTION Differentiated somatic cells acquire pluripotency by transduction of reprogramming factors and are referred to as induced pluripotent stem (iPS) cells [1]. iPS cells express pluripotency-related genes and can differentiate into cells of all three germ layers in vitro and in vivo [2, 3]. Owing to their differentiation potential and self-renewal ability, personalized iPS cells have been considered useful for studying disease mechanisms and in clinical applications. However, it has been argued that iPS cells generated using a viral system form tumorigenic chimera [4, 5], although the exogenous genes are silenced in fully reprogrammed cells [6]. To overcome this obstacle, nonviral systems have been developed, such as episomal introduction of DNA or RNA [7–9], protein delivery systems [10, 11], and small-molecule compounds [12]. However, viral systems are widely used to study the mechanism underlying reprogramming because of their efficient
STEM CELLS 2014;32:2596–2604 www.StemCells.com
induction of pluripotency. Virus-derived factors directly bind to their endogenous counterparts, whose expression, however, normally takes a long period (more than a week after transduction) [13–16]. Therefore, initial reprogramming by exogenous factors may enhance cell proliferation and chromatin structure remodeling [17]. Upon complete reprogramming, exogenous genes become silenced probably by DNA methylation via the DNA methyltransferases, Dnmt3a and Dnmt3b, which are highly expressed in the pluripotent state [18]. Subsequently, endogenous genes stabilize the pluripotent state of iPS cells by maintaining the endogenous autoregulatory system [19]. Here, we aimed to determine whether the silenced exogenous genes in iPS cells remain silenced even after loss of pluripotency by in vitro differentiation. We therefore developed an in vitro system to determine the possibility of activation of exogenous genes that had integrated into the host genome during transduction of the transcription factor cocktail [1]. We C AlphaMed Press 2014 V
Choi, Kim, Choi et al. generated iPS cells using retroviral transduction of Yamanaka factors; the iPS cells redifferentiated into neural stem cells (NSCs) and neural subtypes. We found that the silenced exogenous genes could be reactivated spontaneously in NSCs, which was followed by reversion to pluripotent cells. The reactivation of the silenced genes was associated with the downregulation of DNA methyltransferases during the differentiation of iPS cells.
MATERIALS
AND
METHODS
Generation of iPS Cells by Retrovirus pMX-based retroviral vectors encoding the mouse cDNAs of Oct4, Sox2, KLf4, and c-Myc were separately cotransfected by packaging defective helper plasmids into 293T cells using the Fugene 6 transfection reagent (Roche). Forty-eight hours after transfection, virus supernatants were collected, filtered, and concentrated as previously described [20]. Murine embryonic fibroblasts (MEFs) (OG21/2/ROSA261/2) were seeded at a density of 1 3 105 cells per six-well plate and incubated for 24 hours with virus-containing supernatants of the four factors (1:1:1:1) supplemented with 6 mg/ml protamine sulfate (Sigma). After retroviral infection, green fluorescent protein (GFP)-positive colonies were transferred onto inactivated MEF feeders, and then trypsinized, replated, and cultured in embryonic stem cell (ESC) medium supplemented with 15% fetal bovine serum (FBS), 13 penicillin/streptomycin/glutamine, 0.1 mM nonessential amino acids, 1 mM b-mercaptoethanol (Gibco BRL, Grand Island, NY, http://www.lifetechnologies.com), and 103 units/ml leukemia inhibitory factor (LIF; ESGRO, Millipore, Billerica, MA, http://www.millipore.com).
Differentiation of ESCs and iPS Cells into NSCs ESCs and iPS cells were cultured for 2–3 days with MEF medium supplemented with 15% FBS, 13 penicillin/streptomycin/glutamine, 0.1 mM nonessential amino acids, 1 mM bmercaptoethanol (Gibco BRL) in Dulbecco’s modified Eagle’s medium (DMEM). After 2–3 days, ESCs and iPS cells were cultured without feeder cells in N2B27 medium (20 ng/ml epidermal growth factor [EGF; Gibco BRL], 20 ng/ml basic fibroblast growth factor [bFGF], B27 supplement [Gibco BRL], 8 mM HEPES, 2 mM glutamine, 100 U/ml penicillin, and 100 mg/ml streptomycin in DMEM/F-12 medium [Gibco BRL]) for 2 days in the suspension culture dish and then plated onto a 0.1% gelatin-coated dish for 3 days with NSC expansion medium (NS-A medium [Euroclone] with N2 supplement, 10 ng/ml EGF, 10 ng/ml bFGF [Invitrogen], 50 mg/ml bovine serum albumin (BSA) [BSA fraction; Gibco BRL], and 13 penicillin/streptomycin/glutamine). For differentiation of iPS cell-derived NSCs (iPS-NSCs) into all three neural subtypes, the NSCs were cultured for 1 week in neuronal differentiation medium (DMEM/F-12 medium [GIBCO], neurobasal medium [GIBCO], 1% FBS, 13 N2 supplement [GIBCO], 13 penicillin/streptomycin/glutamine [Invitrogen], and 503 B27 supplement).
Generation of Doxycycline-Inducible iPS Cells MEF cultures were established from E13.5 embryos from reprogrammable mice carrying one copy of Oct4-GFP transgenes and two copies of the OKSM cassette, the ROSA26M2rtTA allele [21, 22]. Reprogramming was performed in ESC medium in the presence of doxycycline (5 mg/ml). Ten days
www.StemCells.com
2597
after doxycycline treatment, Oct4-GFP-positive colonies were observed. Only Oct4-GFP-positive colonies were selected and transferred onto inactivated MEFs. For reprogramming experiments, doxycycline-inducible iPS (Di-iPS) cells were differentiated into NSCs and cultured in NSC medium or ESC medium in the presence of doxycycline (5 mg/ml).
RNA Isolation and qRT-PCR Analysis Total RNA was isolated using the RNeasy Mini Kit (Qiagen, Venlo, Netherlands, http://www.qiagen.com) and was treated with DNase to remove genomic DNA contamination. One microgram of total RNA was reverse-transcribed with SuperScript III Reverse Transcriptase Kit (Invitrogen) and oligo(dT) primer (Invitrogen) according to the manufacturer’s instructions. Quantitative polymerase chain reaction (PCR) reactions were set up in duplicate with the Power SYBR Green Master Mix (Takara) and analyzed with the Roche LightCycler 5480 (Roche). The primers for qRT-PCR used were as follows: Oct4 (endo) sense 50 -GATGCTGTGAGCCAAGGCAAG-30 , Oct4 (endo) antisense 50 -GGCTCCTGATCAACAGCATCAC-30 ; Nanog (endo) sense 50 -CTTTCACCTATTAAGGTGCTTGC-30 , Nanog (endo) antisense 50 -TGGCATCGGTTCATCATGGTAC-30 ; Nestin (endo) sense 50 -GGAACCCAGAGAATGTGGAA-30 , Nestin (endo) antisense 50 CACATCCTCCCACCTCTGTT-30 ; Dnmt1 sense 50 -CGG 0 0 CTCAAAGACTTGGAAAG-3 , Dnmt1 antisense 5 -TAGCCAGGTAG CCTTCCTCA-30 ; Dnmt3a sense 50 -GATGCTGTGAGCCAAGGCAAG30 , Dnmt3a antisense 50 -TGCTTGTTCTGCACTTCCAC-30 ; Dnmt3b sense 50 -ACTTGGTGATTGGTGGAAGC-30 , Dnmt3b antisense 50 CCAGAAGAATGGACGGTTGT-30 ; Oct4 TG sense 50 -GACGGCATCG CAGCTTGGATA-30 , Oct4 TG antisense 50 -CCAATACCTCTGAGC CTGGT-30 ; Sox2 TG sense 50 -GACGGCATCGCAGCTTGGATA-30 , Sox2 TG antisense 50 -CGCTTGGCCTCGTCGATGAA-30 ; Klf4 TG sense 50 -GACGGCATCGCAGCTTGGATA-30 , Klf4 TG antisense 50 GGGAAGTCGCTTCATGTGAG-30 ; c-Myc TG sense 50 -GACGGCATC GCAGCTTGGATA-30 , c-Myc TG antisense 50 -ACCGCAACATAG GATGGAGA-30 ; ACTB sense 50 -CGCCATGGATGACGATATCG-30 , and ACTB TG antisense 50 -CGAAGCCGGCTTTGCACATG-30 .
Bisulfite Genomic Sequencing To differentiate between methylated and unmethylated CG dinucleotides, genomic DNA was treated with sodium bisulfite to convert all unmethylated cytosine residues into uracil residues using the EpiTect Bisulfite Kit (QIAGEN) according to the manufacturer’s protocol. Briefly, purified genomic DNA (0.5–1 mg) was denatured at 99 C and then incubated at 60 C. After desulfonation, neutralization, and desalting, the modified DNA was diluted in 20 ml of distilled water. Subsequently, bisulfite PCR (BS-PCR) amplification was carried out using 1–2-ml aliquots of modified DNA for each PCR reaction. The primers used for BS-PCR were as follows [23]: Oct4 first sense 50 TTTGTTTTTTTATTTATTTAGGGGG-30 , Oct4 first antisense 50 -ATC CCCAATACCTCTAAACCTAATC-30 ; Oct4 second sense 50 -GGGTTG GAGGTTAAGGTTAGAGGG-30 , Oct4 second antisense 50 -CCCCC ACCTAATAAAAATAAAAAAA-30 ; Oct4 TG sense 50 -TTGCAAAGGAT TTTATATAGTTTT-30 , and Oct4 TG antisense 50 -AATCCCCAA TACCTCTAAACCTAA-30 . Briefly, the amplified products were verified by electrophoresis on a 1% agarose gel. The desired PCR products were used for subcloning using the TA cloning vector (pGEM-T Easy Vector, Promega). The reconstructed plasmids were purified, and individual clones were sequenced (Solgent Corporation). C AlphaMed Press 2014 V
2598
Immunocytochemistry Experiments For immunocytochemistry, cells were fixed with 4% paraformaldehyde for 20 minutes at room temperature. After the cells were washed with PBS, they were treated with PBS containing 10% normal goat serum and 0.03% Triton X-100 for 45 minutes at room temperature. The primary antibodies used were anti-nanog (nanog; monoclonal, 1:200, Abcam, Cambridge, UK, http://www.abcam.com), anti-Nestin (Nestin; monoclonal, 1:500, Millipore), anti-Sox2 (Sox2; polyclonal, 1:500, Millipore), anti-tubulin, bIII (Tuj1; monoclonal, 1:1,000, Millipore), anti-glial fibrillary acidic protein (GFAP) (monoclonal, 1:500, Millipore), and anti-O4 (O4; monoclonal, 1:200, Millipore). For detection of primary antibodies, fluorescently labeled (Alexa Fluor 488 or 568; Molecular Probes, Eugene, OR) secondary antibodies were used according to the specifications of the manufacturer.
Chimera Formation iPS cells were aggregated with denuded postcompacted eightcell-stage embryos to obtain an aggregate chimera. Eight-cell embryos flushed from 2.5-dpc B6D2F1 female mice were cultured in microdrops of embryo culture medium under mineral oil. The cells were trypsinized for 10 seconds, and clumps of iPS cells (4–10 cells) were selected and transferred into microdrops containing zona-free eight-cell embryos. Morula-stage embryos aggregated with iPS cells were cultured overnight at 37 C under 5% CO2. The aggregated blastocysts were transferred into one uterine horn of 2.5-dpc pseudopregnant recipients.
X-gal Staining For whole fetal embryo staining, the fetuses collected were rinsed with PBS and fixed with 4% formaldehyde for 1 hour at 4 C. Fetuses were rinsed three times at room temperature in PBS supplemented with 5 mM EGTA, 0.01% deoxycholate, 0.02% Nonidet P40, and 2 mM MgCl2. The specimens were washed with PBS and then stained in X-gal staining solution (PBS supplemented with 1 mg/ml 5-bromo-4-chloro-3-indolylgalactosidase [X-gal; Promega, Seoul, Rep of Korea http:// kr.promega.com], 5 mM K2Fe(CN)6, 5 mM K4Fe(CN)6, and 1 mM MgCl2). The blue staining was visualized by light microscopy.
RESULTS Derivation of NSCs from ESCs and iPS Cells In Vitro We generated iPS cells using a retroviral system and compared their pluripotency with the ESC line E14. The E14 ESCs were well-characterized pluripotent stem cells derived from an inbred 129/Ola strain [24]. iPS cells were derived from OG2/ ROSA26 heterozygous female MEFs, which contained Oct4-GFP and neo/LacZ transgenes. iPS cells were efficiently generated by retroviral transduction of four transcription factors (Oct4, Sox2, Klf4, and c-Myc) (Fig. 1A). Oct4-GFP-positive iPS cells (Fig. 1B) expressed high levels of endogenous Oct4, Nanog, Esrrb, and Prdm14 (Fig. 1C and Supporting Information Fig. S1) and formed germline chimeras (Fig. 1D), suggesting that these iPS cells were fully reprogrammed to pluripotent ES-like cells. To investigate the possibility that the silenced exogenous factors are reactivated during differentiation of iPS cells, we C AlphaMed Press 2014 V
Reacquisition of Pluripotency from iPSC-NSCs
differentiated retrovirus-mediated iPS cells into a specialized cell type, NSCs, which can be easily reprogrammed with fewer factors [25–27]. We differentiated ESCs (as a control) and iPS cells to NSCs using a previously described protocol with some modifications (Fig. 1A) [28]. ESCs and iPS cells were cultured in the absence of LIF and feeder cells for 2–3 days, serially cultured in N2B27 medium for 2–3 days, and then cultured in NS expansion medium containing EGF and bFGF for 3–4 days. When spherical embryoid bodies (EBs) were formed, they were cultured onto the gelatin-coated dish. EBs from ESCs and iPS cells formed neural tube-like rosettes at days 10 and 30, respectively, after induction of neural differentiation (Supporting Information Fig. S2A, S2B). These rosettes were picked and cultured in NSC culture medium, where the rosettes formed a monolayer of bipolar-shaped NSCs. The bipolar NSCs were picked in a new dish (passage 0) and further passaged. These cells were morphologically indistinguishable from NSCs derived from brain tissue (Fig. 1E); the cells rapidly propagated in the presence of EGF and bFGF and stained positive for the NSC markers Nestin and Sox2 (Fig. 1F), indicating that NSCs derived from ESCs and iPS cells were indistinguishable from brain-derived NSCs. To investigate the multipotent differentiation potential of iPS-NSCs, we induced differentiation of iPS-NSCs into neurons, astrocytes, and oligodendrocytes in a medium without EGF and bFGF. At 8–10 days in culture, the bipolar morphology of NSCs changed into a variety of polypolar cell types. The iPSNSCs stained positive for the Tuj1, GFAP, and O4 markers for neurons, astrocytes, and oligodendrocytes, respectively (Fig. 1G), indicating that iPS-NSCs could differentiate into all three neural subtypes.
Reactivation of Exogenous Factors Causes Regainment of Pluripotency in iPS-NSCs NSCs show sustained self-renewal for an expanded period in vitro in the presence of EGF and bFGF [28]. NSCs derived from ESCs and iPS cells were passaged every 3 days. The ESCderived NSCs (ES-NSCs) were maintained and propagated for more than 30 passages, for more than 3 months. Established iPS-NSCs maintained a typical NSC morphology for approximately 4 weeks. However, the morphology of iPS-NSCs #1 was changed at passages 8–9, and aggregated cells were observed (Fig. 2A), but not iPS-NSCs #2 and #3 (data not shown). We first tested whether the aggregates were neurospheres. Interestingly, most of the aggregated cells stained positive for Sox2 but not for Nestin (Fig. 2B), indicating that the aggregated cells were neither neurospheres nor NSCs. These cell aggregates gradually changed into ESC-like colonies in NSC culture medium. We therefore hypothesized that iPSNSCs spontaneously revert to a pluripotent state. However, the colony-forming cells did not reactivate Oct4-GFP, which was reactivated only after 4 days under ESC culture conditions (Fig. 2C and Supporting Information Fig. S4). Aggregates of iPS-NSCs (iPS-NSC-R) under NSC culture conditions also did not express endogenous Oct4 and Nanog until Oct4-GFP was reactivated (Fig. 2D). These results indicate that suitable culture conditions are also required for regaining pluripotency, in addition to the activation of transgenes. The Oct4-GFP-positive (iPS-NSC-GFP1) cells expressed the pluripotency markers Oct4 and Nanog, but not Nestin, as observed in pluripotent ESCs (Fig. 2D). Bisulfite DNA sequencing analysis showed the STEM CELLS
Choi, Kim, Choi et al.
2599
Figure 1. Generation of iPS cells and differentiation of NSCs derived from ESCs and iPS cells. (A): Timeline of generation of iPS cells and differentiation of NSCs. (B): Phase and fluorescence (Oct4-GFP) images of iPS cells from MEF. Scale bar 5 100 mm. (C): Expression levels of the pluripotency markers Oct4 and Nanog in iPS cells by qRT-PCR. Data are presented as mean 6 SD of triplicates (n 5 3). (D): iPS cells formed germline chimeras. A chimeric embryo from iPS cells was stained by X-gal, and Oct4-GFP-positive cells were detected in the gonad of the chimeric embryo. Scale bar 5 100 mm. (E): Morphology of brain-, ESC-, and iPS cell-derived NSCs. iPS cell-derived NSCs were morphologically indistinguishable from control brain-derived NSCs. Scale bar 5 100 mm. (F): Immunocytochemical results for the expression of the NSC markers Nestin and Sox2 in ESC- and iPS cell-derived NSCs. Scale bar 5 50 mm. (G): Differentiation potential of iPS cell-derived NSCs. Differentiated cells from iPS cell-derived NSCs were stained for markers of neurons (Tuj1; scale bar 5 50 mm), astrocytes (GFAP; scale bar 5 20 mm), and oligodendrocytes (O4; scale bar 5 20 mm). Abbreviations: GFAP, glial fibrillary acidic protein; GFP, green fluorescent protein; iPS, induced pluripotent stem; LIF, leukemia inhibitory factor; MEF, murine embryonic fibroblast; NSC, neural stem cell.
dynamics of DNA methylation on the Oct4 promoter region (Fig. 2E). The hypermethylated status of Oct4 promoter region of MEFs was demethylated after become an iPS cells and was then hypermethylated in iPS-NSCs on differentiation. The Oct4 promoter regions were demethylated in iPS-NSC-GFP1 cells, as shown in ESCs (Fig. 2E). Next, we analyzed a stringent marker, that is, X chromosome reactivation, for ground-state pluripotency. Xist RNA FISH analysis showed that all iPS-NSCs contained only inactive X chromosomes (Xi; detected as Xist cloud), indicating that the iPS-NSC population was not contaminated by undifferenti-
www.StemCells.com
ated cells (Fig. 2F); Xi was not observed in undifferentiated pluripotent cells, such as ESCs or iPS cells [29]. When the iPSNSCs were changed into iPS-NSC-GFP1 cells, two active X chromosomes (Xa; detected as pinpoint signals) were detected, as shown in female iPS cells. This result indicated that iPS-NSCs were completely differentiated from iPS cells and that undifferentiated cells were not contaminated in the iPS-NSC population. Moreover, iPS-NSC-GFP1 cells could form germline chimeras (Fig. 2G), indicating that NSCs differentiated from iPS cells could regain pluripotency and be regenerated into iPS cells. C AlphaMed Press 2014 V
2600
Reacquisition of Pluripotency from iPSC-NSCs
Figure 2. Regainment of pluripotency of NSCs derived from iPS cells. (A): Morphology of aggregates (iPS-NSC-R) from iPS cell-derived NSCs at passage 8. Scale bar 5 50 mm. (B): Expression of the NSC markers Nestin and Sox2 in iPS-NSC-R. Scale bar 5 50 mm. (C): Phase and fluorescence (Oct4-GFP) images of iPS-NSC-GFP1 cells at day 4 and passage 10. Scale bar 5 100 mm. (D): Gene expression levels of the pluripotency markers Oct4 (endo) and Nanog and the NSC marker Nestin in reprogramming to differentiation and reprogramming of cells. Data are presented as mean 6 SD of triplicates (n 5 3). (E): Bisulfite genomic sequencing of the promoter regions of Oct4. (F): Xist RNA FISH in fMEFs, iPS cells, iPS-NSCs, and iPS-NSC-GFP1 cells, using a Xist probe (red signal). (G): Formation of germline chimeras by iPS-NSC-GFP1 cells. A chimeric embryo from iPS-NSC-GFP1 cells was stained by X-gal. Scale bar 5 100 mm. Oct4-GFP-positive cells were detected in the gonad of the chimeric embryo. Scale bar 5 50 mm. Abbreviations: ESC, embryonic stem cell; GFP, green fluorescent protein; iPS, induced pluripotent stem; MEF, murine embryonic fibroblast; NSC, neural stem cell.
However, when iPS-NSCs were terminally differentiated into neural subtypes, cell aggregates or morphologic changes were not observed (Supporting Information Fig. S2), suggesting that regaining of pluripotency might occur only in easily reprogrammable cell types, such as NSC, but not in cell types that are difficult to reprogram, such as fully differentiated neurons [30].
Reactivation of Retroviral Genes Precedes the Reactivation of Endogenous Pluripotency Genes Retrovirus-mediated genes were silenced in ES and embryonic carcinoma (EC) cells [31] and fully reprogrammed iPS cells [2, 3, 6]. Here, we hypothesized that regainment of pluripotency in iPS-NSCs might be caused by reactivation of retroviral genes, which were silenced in the pluripotent state. Therefore, we examined the expression levels of transgenes (Oct4, Sox2, Klf4, and c-Myc) during the transformation of iPS-NSCs to iPS-NSCGFP1 cells. As expected, retrovirus-mediated transgenes were expressed in NSCs or MEFs shortly after retroviral transduction of transcription factors, but were not expressed in iPS cells after reprogramming (Fig. 3A). The silenced transgenes in iPS cells were reactivated after differentiation into iPS-NSCs and C AlphaMed Press 2014 V
were then downregulated in iPS-NSC-R (Fig. 3A); endogenous Oct4 was not expressed in both iPS-NSCs and iPS-NSC-R (Fig. 2D). The Sox2 transgene, which is a marker of NSCs and iPS cells, was more highly expressed in iPS-NSC-R (Fig. 3A). iPSNSC-GFP1 cells did not express retroviral transgenes; the transgenes were resilenced once the cells became pluripotent (Fig. 3A). However, only iPS-NSCs #1 were re-expressed transgenes, but not iPS-NSC #2 (Supporting Information Fig. S5). These data showed that reactivation of endogenous pluripotency-related genes (regainment of pluripotency) was associated with the reactivation of retroviral transgenes on differentiation.
Reactivation of Silenced Transgenes Is Associated With the Level of DNA Methyltransferases Epigenetic modifications are involved in the silencing of retroviral genes [32], which includes DNA methylation by DNA methyltransferases (Dnmts), such as Dnmt3a, Dnmt3b, and Dnmt1 [33–35]. During iPS cell induction, Dnmt3a and Dnmt3b levels gradually increased [36]. Dnmt1, which plays a role in maintaining DNA methylation, is involved in the silencing of retroviral genes in fully reprogrammed iPS cells [6]. We STEM CELLS
Choi, Kim, Choi et al.
2601
Figure 3. Reactivation of retroviral transgenes in iPS cell-derived NSCs, which was associated with the level of DNA methyltransferases. (A): qRT-PCR results for retroviral transgene (OSKM) expression levels in iPS cells, iPS-NSCs, iPS-NSC-R, and iPS-NSC-GFP1 cells. NSCs (4F) and MEF (4F) were control, infected retroviral four factors in, respectively, NSCs and MEF. Data are presented as mean 6 SD of triplicates (n 5 3). (B): qRT-PCR results for expression levels of Dnmt1, Dnmt3a, and Dnmt3b. Data are presented as mean 6 SD of triplicates (n 5 3). (C): Bisulfite genomic sequencing of regions of the retroviral transgene Oct4. Abbreviations: ESC, embryonic stem cell; GFP, green fluorescent protein; iPS, induced pluripotent stem; MEF, murine embryonic fibroblast; NSC, neural stem cell.
therefore examined the expression levels of Dnmt1, Dnmt3a, and Dnmt3b in iPS cells, iPS-NSCs, iPS-NSC-R, and iPS-NSCGFP1 cells (Fig. 3B). Dnmt1, Dnmt3a, and Dnmt3b, which were expressed at extremely low levels in MEFs, were upregulated after reprogramming into iPS cells (similar to the level in ESCs). The expression levels of Dnmt1, Dnmt3a, and Dnmt3b were downregulated after differentiation of iPS cells into iPS-NSCs. Dnmt3b expression was almost silenced in iPSNSCs like ES-derived NSCs (Supporting Information Fig. S6). These three Dnmts were upregulated during the transformation of iPS-NSCs into iPS-NSC-GFP1 cells (Fig. 3B). Considering that these Dnmts are responsible for the silencing of viral genes by inducing DNA methylation [6, 34, 35], we speculated that the DNA methylation status would change according to the levels of Dnmts. Therefore, we investigated the methylation status of the exogenous Oct4 promoter region in iPS
www.StemCells.com
cells, iPS-NSCs, iPS-NSC-R, and iPS-NSC-GFP1 cells by bisulfite genomic sequencing (Fig. 3C). The Oct4 transgene was found to be hypermethylated in iPS cells (high level of Dnmts), completely or hypomethylated in iPS-NSCs and iPS-NSC-R (low level of Dnmts), and hypermethylated in iPS-NSC-GFP1 cells (high level of Dnmts). The methylation status of Oct4 transgene was closely correlated with the expression level of the Oct4 transgene during the differentiation of iPS cells to iPSNSCs and the regainment of pluripotency in iPS-NSC-GFP1 cells (Fig. 3A, 3C).
Regainment of iPS Cells From iPS-NSCs was Closely Associated With Transgene Reactivation To confirm that iPS-NSCs spontaneously regain pluripotency because of transgene reactivation, we generated Di-iPS cells carrying Oct4-GFP, the OSKM (Oct4, Sox2, Klf4, and c-Myc) C AlphaMed Press 2014 V
Reacquisition of Pluripotency from iPSC-NSCs
2602
Figure 4. Regainment of pluripotency was closely associated with transgene reactivation. (A): Phase-contrast and fluorescence (Oct4GFP) images of Di-iPS cells. Scale bar 5 100 mm. (B): Immunocytochemistry results for the expression of the pluripotency marker Nanog in Di-iPS cells. Scale bar 5 50 mm. (C): Morphology of NSCs derived from Di-iPS cells and expression of the NSC markers Nestin and Sox2. Scale bar 5 50 mm. (D): Oct4-GFP1 cells from Di-iPS-NSCs were first observed at 15 days after doxycycline treatment in ES medium. Scale bar 5 50 mm. (E): Gene expression levels of endogenous Oct4, and Nanog in MEFs, Di-iPS cells, Di-iPS-NSCs (Dox2), DiiPS-NSCs (Dox1, day 3), Di-iPS-NSC-GFP1 cells, ESCs, control NSCs, and DW by RT-PCR. Abbreviations: DW, distilled water; ESC, embryonic stem cell; GFP, green fluorescent protein; iPS, induced pluripotent stem; MEF, murine embryonic fibroblast; NSC, neural stem cell.
cassette, and the transcriptional activator within the ROSA26 locus (ROSA26-M2rtTA) [21, 22]. Oct4-GFP-positive Di-iPS cells were generated from reprogrammable MEFs at day 10 after doxycycline induction (Fig. 4A). Once the Di-iPS cells were established, they maintained pluripotency without doxycycline. The Di-iPS cells expressed the pluripotency marker Nanog (Fig. 4B) and formed germline chimeras (Supporting Information Fig. S9A–S9C). Next, we induced differentiation of Di-iPS cells to NSCs (Di-NSCs). Di-NSCs without doxycycline maintained a NSC morphology and expressed the NSC markers Nestin and Sox2 (Fig. 4C); we observed neither aggregated cells nor Oct4-GFP-positive cells in long-term cultures of Di-NSCs (for more than 20 passages). However, when we cultured the Di-NSCs in doxycycline-containing medium, the cells aggregated to form ESC-like colonies and expressed Oct4-GFP (Fig. 4D and Supporting Information Fig. S10). Although we were unable to observe iPS-NSC-R-like aggregates, we cannot exclude the possibility of an intermediate state during the doxycyclineinducible reprogramming. These Oct4-GFP-positive cells expressed the pluripotency markers Oct4 (endo) and Nanog (Fig. 4E), indicating that Di-NSCs were reprogrammed into pluC AlphaMed Press 2014 V
ripotent state by transgene reactivation. Similar results were observed in another Di-iPS cell line, that is, in PB-iPS cells (Supporting Information Fig. S11A–S11C) [37]. These results indicate that NSCs differentiated from iPS cells regain pluripotency upon reactivation of exogenous factors.
DISCUSSION Reprogramming into iPS cells by transduction of reprogramming factors (more than a week) is a much slower step than that in cell fusion between ESCs and somatic cells (1–2 days). Therefore, initial reprogramming by exogenous factors might enhance proliferation and chromatin remodeling [17]. Transcription factors such as Oct4, Sox2, and c-Myc interact with chromatin remodeling factors [19], which leads to upregulation of endogenous pluripotency-related genes and repression of tissue-specific genes [38]. Once pluripotency is achieved, reprogrammed cells are controlled by endogenous regulatory circuits [17–19], and exogenous factors become inactive. In this study, we showed that the silenced exogenous factors STEM CELLS
Choi, Kim, Choi et al.
2603
were derepressed as the iPS cells lose pluripotency, which results in regainment of pluripotency. A previous study showed that transgenes were silenced in iPS-NSCs, which were propagated up to 67 passages [39]. Conversely, retroviral c-Myc often became reactivated in differentiated cells derived from iPS cells in chimeric mice, which subsequently led to tumor formation [4, 5]. Our in vitro system suggested that retroviral genes became reactivated following downregulation of dnmts during differentiation of iPS cells, which subsequently regained pluripotency. Therefore, silenced genes in a pluripotent environment can be activated through changes in the cell state. iPS cell-derived neurospheres containing undifferentiated Nanog-GFP-positive cells could form significantly larger teratomas when transplanted into NOD/SCID mice [40]. However, the regainment of the pluripotency of iPS-NSCs should not be attributed to contamination of undifferentiated cells in the iPS-NSC population. First, iPS-NSCs were cultured under NSC monolayer culture conditions for approximately 4 weeks, and a bipolar morphology was maintained. Moreover, to obtain a pure NSC population, only NSCs were picked and cultured at the initial stage of NSC derivation. Second, cell aggregate formation was observed in most of the NSC colonies simultaneously (Supporting Information Fig. S2). If a small number of undifferentiated cells were present, aggregates should be formed only in a few sites and in a separate colony from NSCs. Third, Xist RNA FISH analysis showed that the iPS-NSC population contained only differentiated cells (Fig. 1F); all cells contained Xi, which was not observed in undifferentiated pluripotent cells [29]. Taken together, these findings showed that iPS-NSCs were a completely differentiated cell population and that no contaminating undifferentiated cells were present in the iPS-NSC population. We were able to induce differentiation of iPS cells into NSCs and establish three NSC cell lines (iPS-NSCs #1, #2, and #3). iPS-NSCs #2 and #3 were maintained and proliferated over 30 passages. However, iPS-NSC #1 spontaneously aggregated and showed reactivation of Oct4-GFP after 8–9 passages; that is, only iPS-NSCs #1 had transgene reactivation. The mechanism underlying this phenomenon is unclear; however, it is possible that transgene integration sites affect the reactivation of silenced transgenes. Four retroviral factors were silenced again in reprogrammed iPS-NSC-GFP1 cells. However, terminally differentiated cells maintain their morphology without proliferation (Supporting Information Fig. S2). Therefore, regainment of pluripotency in iPS cell-derived differentiated cells might occur only in easily reprogrammable cell types, such as somatic stem cells, including NSCs. It has been suggested that retrovirus-mediated genes are silenced in ESCs, EC, and iPS cells during reprogramming [2, 3, 6, 31]. The silencing of retroviral transgenes is associated with de novo DNA methylation on CpG dinucleotides [41–43]. Silenced M-MLV proviruses introduced in EC cells could be reactivated after differentiation by combination treatment with retinoic acid and 5-azacytidine, a DNA demethylating agent [44]. DNA methylation is crucial for maintenance of
REFERENCES 1 Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 2006;126:663–676.
www.StemCells.com
gene silencing rather than establishing silencing [34, 35]. In this study, we also showed that DNA methylation on retroviral transgene was important for maintenance of silencing of the transgenes in the iPS cell-derived differentiated cells. Oct4 transgene regions were hypomethylated in iPS-NSCs and in aggregates (iPS-NSC-R; the initial stage of regainment of pluripotency), but were hypermethylated in reprogrammed iPSNSC-GFP1 cells (Fig. 3C). Previous research has demonstrated that retroviral silencing is independent of Dnmt3a and Dnmt3b in pluripotent stem cells [33]. However, we clearly showed that low levels of Dnmt3a and Dnmt3b were closely correlated with the DNA methylation status of retroviral transgenes in differentiated cells from iPS cells.
CONCLUSION In this study, iPS-NSCs were compared with NSCs derived from ESCs and brain tissue. NSCs derived from ESCs and iPS cells were morphologically indistinguishable from brain-derived NSCs, expressed the NSC markers Nestin and Sox2, and could be differentiated into all three neural subtypes: neurons, astrocytes, and oligodendrocytes. These iPS-NSCs spontaneously aggregated upon long-term culture in vitro, formed ESC-like colonies, and finally reactivated Oct4-GFP. The spontaneously reverted GFPpositive cells showed characteristics similar to those of ESCs and iPS cells. This phenomenon could be attributed to transgene reactivation during the differentiation of iPS cells. On the basis of these results, we suggest that exogenous gene-free iPS cells should be considered for therapeutic application.
ACKNOWLEDGMENTS This work was supported by the Biomedical Technology Development Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (Grant No. 20110019489) and the Next-Generation BioGreen 21 Program (Grant No. PJ00956202) funded by the Rural Development Administration, Republic of Korea.
AUTHOR CONTRIBUTIONS H.W.C.: conception and design, collection and/or assembly of data, data analysis and interpretation, and manuscript writing; J.S.K., S.C., Y.J.H., and M.J.K.: collection and/or assembly of data; H.G.S.: provision of study material or patients; J.T.D.: conception and design, financial support, data analysis and interpretation, manuscript writing, and final approval of manuscript.
DISCLOSURE
OF
POTENTIAL CONFLICTS
OF INTEREST
The authors indicate no potential conflicts of interest.
2 Maherali N, Sridharan R, Xie W et al. Directly reprogrammed fibroblasts show global epigenetic remodeling and widespread tissue contribution. Cell Stem Cell 2007;1:55– 70.
3 tion tent 4 tion
Okita K, Ichisaka T, Yamanaka S. Generaof germline-competent induced pluripostem cells. Nature 2007;448:313–317. Aoi T, Yae K, Nakagawa M et al. Generaof pluripotent stem cells from adult
C AlphaMed Press 2014 V
Reacquisition of Pluripotency from iPSC-NSCs
2604
mouse liver and stomach cells. Science 2008; 321:699–702. 5 Nakagawa M, Koyanagi M, Tanabe K et al. Generation of induced pluripotent stem cells without Myc from mouse and human fibroblasts. Nat Biotechnol 2008;26: 101–106. 6 Wernig M, Meissner A, Foreman R et al. In vitro reprogramming of fibroblasts into a pluripotent ES-cell-like state. Nature 2007; 448:318–324. 7 Kaji K, Norrby K, Paca A et al. Virus-free induction of pluripotency and subsequent excision of reprogramming factors. Nature 2009;458:771–775. 8 Warren L, Manos PD, Ahfeldt T et al. Highly efficient reprogramming to pluripotency and directed differentiation of human cells with synthetic modified mRNA. Cell Stem Cell 2010;7:618–630. 9 Yu J, Hu K, Smuga-Otto K et al. Human induced pluripotent stem cells free of vector and transgene sequences. Science 2009;324: 797–801. 10 Kim D, Kim CH, Moon JI et al. Generation of human induced pluripotent stem cells by direct delivery of reprogramming proteins. Cell Stem Cell 2009;4:472–476. 11 Zhou H, Wu S, Joo JY et al. Generation of induced pluripotent stem cells using recombinant proteins. Cell Stem Cell 2009;4: 381–384. 12 Hou P, Li Y, Zhang X et al. Pluripotent stem cells induced from mouse somatic cells by small-molecule compounds. Science 2013; 341:651–654. 13 Brambrink T, Foreman R, Welstead GG et al. Sequential expression of pluripotency markers during direct reprogramming of mouse somatic cells. Cell Stem Cell 2008;2: 151–159. 14 Okita K, Nakagawa M, Hyenjong H et al. Generation of mouse induced pluripotent stem cells without viral vectors. Science 2008;322:949–953. 15 Zhang H, Jiao W, Sun L et al. Intrachromosomal looping is required for activation of endogenous pluripotency genes during reprogramming. Cell Stem Cell 2013;13:30– 35. 16 Apostolou E, Ferrari F, Walsh RM et al. Genome-wide chromatin interactions of the Nanog locus in pluripotency, differentiation, and reprogramming. Cell Stem Cell 2013;12: 699–712.
17 Do JT, Scholer HR. Regulatory circuits underlying pluripotency and reprogramming. Trends Pharmacol Sci 2009;30:296–302. 18 Jaenisch R, Young R. Stem cells, the molecular circuitry of pluripotency and nuclear reprogramming. Cell 2008;132:567–582. 19 Loh YH, Wu Q, Chew JL et al. The Oct4 and Nanog transcription network regulates pluripotency in mouse embryonic stem cells. Nat Genet 2006;38:431–440. 20 Zaehres H, Daley GQ. Transgene expression and RNA interference in embryonic stem cells. Methods Enzymol 2006;420:49–64. 21 Stadtfeld M, Maherali N, Borkent M et al. A reprogrammable mouse strain from gene-targeted embryonic stem cells. Nat Methods 2010;7:53–55. 22 Polo JM, Anderssen E, Walsh RM et al. A molecular roadmap of reprogramming somatic cells into iPS cells. Cell 2012;151:1617–1632. 23 Kim MJ, Choi HW, Jang HJ et al. Conversion of genomic imprinting by reprogramming and redifferentiation. J Cell Sci 2013; 126:2516–2524. 24 Hooper M, Hardy K, Handyside A et al. HPRT-deficient (Lesch-Nyhan) mouse embryos derived from germline colonization by cultured cells. Nature 1987;326:292–295. 25 Kim JB, Sebastiano V, Wu G et al. Oct4induced pluripotency in adult neural stem cells. Cell 2009;136:411–419. 26 Shi Y, Desponts C, Do JT et al. Induction of pluripotent stem cells from mouse embryonic fibroblasts by Oct4 and Klf4 with smallmolecule compounds. Cell Stem Cell 2008;3: 568–574. 27 Do JT, Joo JY, Han DW et al. Generation of parthenogenetic induced pluripotent stem cells from parthenogenetic neural stem cells. Stem Cells 2009;27:2962–2968. 28 Conti L, Pollard SM, Gorba T et al. Niche-independent symmetrical self-renewal of a mammalian tissue stem cell. PLoS Biol 2005;3:e283. 29 Yang X, Smith SL, Tian XC et al. Nuclear reprogramming of cloned embryos and its implications for therapeutic cloning. Nat Genet 2007;39:295–302. 30 Kim J, Lengner CJ, Kirak O et al. Reprogramming of postnatal neurons into induced pluripotent stem cells by defined factors. Stem Cells 2011;29:992–1000. 31 Pannell D, Ellis J. Silencing of gene expression: Implications for design of retrovirus vectors. Rev Med Virol 2001;11:205–217.
32 Ellis J, Yao S. Retrovirus silencing and vector design: Relevance to normal and cancer stem cells? Curr Gene Ther 2005;5:367– 373. 33 Pannell D, Osborne CS, Yao S et al. Retrovirus vector silencing is de novo methylase independent and marked by a repressive histone code. EMBO J 2000;19:5884–5894. 34 Yao S, Sukonnik T, Kean T et al. Retrovirus silencing, variegation, extinction, and memory are controlled by a dynamic interplay of multiple epigenetic modifications. Mol Ther 2004;10:27–36. 35 Ramunas J, Montgomery HJ, Kelly L et al. Real-time fluorescence tracking of dynamic transgene variegation in stem cells. Mol Ther 2007;15:810–817. 36 Stadtfeld M, Maherali N, Breault DT et al. Defining molecular cornerstones during fibroblast to iPS cell reprogramming in mouse. Cell Stem Cell 2008;2:230–240. 37 Woltjen K, Michael IP, Mohseni P et al. piggyBac transposition reprograms fibroblasts to induced pluripotent stem cells. Nature 2009;458:766–770. 38 Liang J, Wan M, Zhang Y et al. Nanog and Oct4 associate with unique transcriptional repression complexes in embryonic stem cells. Nat Cell Biol 2008;10:731–739. 39 Onorati M, Camnasio S, Binetti M et al. Neuropotent self-renewing neural stem (NS) cells derived from mouse induced pluripotent stem (iPS) cells. Mol Cell Neurosci 2010;43:287–295. 40 Miura K, Okada Y, Aoi T et al. Variation in the safety of induced pluripotent stem cell lines. Nat Biotechnol 2009;27:743–745. 41 Jahner D, Stuhlmann H, Stewart CL et al. De novo methylation and expression of retroviral genomes during mouse embryogenesis. Nature 1982;298:623–628. 42 Challita PM, Kohn DB. Lack of expression from a retroviral vector after transduction of murine hematopoietic stem cells is associated with methylation in vivo. Proc Natl Acad Sci USA 1994;91:2567–2571. 43 Stewart CL, Stuhlmann H, Jahner D et al. De novo methylation, expression, and infectivity of retroviral genomes introduced into embryonal carcinoma cells. Proc Natl Acad Sci USA 1982;79:4098–4102. 44 Niwa O, Yokota Y, Ishida H et al. Independent mechanisms involved in suppression of the Moloney leukemia virus genome during differentiation of murine teratocarcinoma cells. Cell 1983;32:1105–1113.
See www.StemCells.com for supporting information available online.
C AlphaMed Press 2014 V
STEM CELLS
Department of Animal Biotechnology, College of Animal Bioscience and Technology, Konkuk University, Seoul, Republic of Korea Correspondence: Jeong Tae Do, Ph.D., Department of Animal Biotechnology, College of Animal Bioscience and Technology, Konkuk University, Seoul 143-701, Republic of Korea. Telephone: 82-2-4503673; Fax: 82-2-455-1044; e-mail: [email protected] Received November 28, 2013; accepted for publication May 6, 2014; first published online in STEM CELLS EXPRESS June 4, 2014. C AlphaMed Press V
ABSTRACT Differentiated somatic cells can be reprogrammed into pluripotent stem cells by transduction of exogenous reprogramming factors. After induced pluripotent stem (iPS) cells are established, exogenous genes are silenced. In the pluripotent state, retroviral genes integrated in the host genome are kept inactive through epigenetic transcriptional regulation. In this study, we tried to determine whether exogenous genes remain silenced or are reactivated upon loss of pluripotency or on differentiation using an in vitro system. We induced differentiation of iPS cells into neural stem cells (NSCs) in vitro; the NSCs appeared morphologically indistinguishable from brain-derived NSCs and stained positive for the NSC markers Nestin and Sox2. These iPS cellderived NSCs (iPS-NSCs) were also capable of differentiating into all three neural subtypes. Interestingly, iPS-NSCs spontaneously formed aggregates on long-term culture and showed reactivation of the Oct4-GFP marker, which was followed by the formation of embryonic stem cell-like colonies. The spontaneously reverted green fluorescent protein (GFP)-positive (iPS-NSCGFP1) cells expressed high levels of pluripotency markers (Oct4 and Nanog) and formed germline chimeras, indicating that iPS-NSC-GFP1 cells had the same pluripotency as the original iPS cells. The reactivation of silenced exogenous genes was tightly correlated with the downregulation of DNA methyltransferases (Dnmts) during differentiation of iPS cells. This phenomenon was not observed in doxycycline-inducible iPS cells, where the reactivation of exogenous genes could be induced only by doxycycline treatment. These results indicate that pluripotency can be regained through reactivation of exogenous genes, which is associated with dynamic change of Dnmt levels during differentiation of iPS cells. STEM CELLS 2014;32:2596–2604
1066-5099/2014/$30.00/0 http://dx.doi.org/ 10.1002/stem.1757
INTRODUCTION Differentiated somatic cells acquire pluripotency by transduction of reprogramming factors and are referred to as induced pluripotent stem (iPS) cells [1]. iPS cells express pluripotency-related genes and can differentiate into cells of all three germ layers in vitro and in vivo [2, 3]. Owing to their differentiation potential and self-renewal ability, personalized iPS cells have been considered useful for studying disease mechanisms and in clinical applications. However, it has been argued that iPS cells generated using a viral system form tumorigenic chimera [4, 5], although the exogenous genes are silenced in fully reprogrammed cells [6]. To overcome this obstacle, nonviral systems have been developed, such as episomal introduction of DNA or RNA [7–9], protein delivery systems [10, 11], and small-molecule compounds [12]. However, viral systems are widely used to study the mechanism underlying reprogramming because of their efficient
STEM CELLS 2014;32:2596–2604 www.StemCells.com
induction of pluripotency. Virus-derived factors directly bind to their endogenous counterparts, whose expression, however, normally takes a long period (more than a week after transduction) [13–16]. Therefore, initial reprogramming by exogenous factors may enhance cell proliferation and chromatin structure remodeling [17]. Upon complete reprogramming, exogenous genes become silenced probably by DNA methylation via the DNA methyltransferases, Dnmt3a and Dnmt3b, which are highly expressed in the pluripotent state [18]. Subsequently, endogenous genes stabilize the pluripotent state of iPS cells by maintaining the endogenous autoregulatory system [19]. Here, we aimed to determine whether the silenced exogenous genes in iPS cells remain silenced even after loss of pluripotency by in vitro differentiation. We therefore developed an in vitro system to determine the possibility of activation of exogenous genes that had integrated into the host genome during transduction of the transcription factor cocktail [1]. We C AlphaMed Press 2014 V
Choi, Kim, Choi et al. generated iPS cells using retroviral transduction of Yamanaka factors; the iPS cells redifferentiated into neural stem cells (NSCs) and neural subtypes. We found that the silenced exogenous genes could be reactivated spontaneously in NSCs, which was followed by reversion to pluripotent cells. The reactivation of the silenced genes was associated with the downregulation of DNA methyltransferases during the differentiation of iPS cells.
MATERIALS
AND
METHODS
Generation of iPS Cells by Retrovirus pMX-based retroviral vectors encoding the mouse cDNAs of Oct4, Sox2, KLf4, and c-Myc were separately cotransfected by packaging defective helper plasmids into 293T cells using the Fugene 6 transfection reagent (Roche). Forty-eight hours after transfection, virus supernatants were collected, filtered, and concentrated as previously described [20]. Murine embryonic fibroblasts (MEFs) (OG21/2/ROSA261/2) were seeded at a density of 1 3 105 cells per six-well plate and incubated for 24 hours with virus-containing supernatants of the four factors (1:1:1:1) supplemented with 6 mg/ml protamine sulfate (Sigma). After retroviral infection, green fluorescent protein (GFP)-positive colonies were transferred onto inactivated MEF feeders, and then trypsinized, replated, and cultured in embryonic stem cell (ESC) medium supplemented with 15% fetal bovine serum (FBS), 13 penicillin/streptomycin/glutamine, 0.1 mM nonessential amino acids, 1 mM b-mercaptoethanol (Gibco BRL, Grand Island, NY, http://www.lifetechnologies.com), and 103 units/ml leukemia inhibitory factor (LIF; ESGRO, Millipore, Billerica, MA, http://www.millipore.com).
Differentiation of ESCs and iPS Cells into NSCs ESCs and iPS cells were cultured for 2–3 days with MEF medium supplemented with 15% FBS, 13 penicillin/streptomycin/glutamine, 0.1 mM nonessential amino acids, 1 mM bmercaptoethanol (Gibco BRL) in Dulbecco’s modified Eagle’s medium (DMEM). After 2–3 days, ESCs and iPS cells were cultured without feeder cells in N2B27 medium (20 ng/ml epidermal growth factor [EGF; Gibco BRL], 20 ng/ml basic fibroblast growth factor [bFGF], B27 supplement [Gibco BRL], 8 mM HEPES, 2 mM glutamine, 100 U/ml penicillin, and 100 mg/ml streptomycin in DMEM/F-12 medium [Gibco BRL]) for 2 days in the suspension culture dish and then plated onto a 0.1% gelatin-coated dish for 3 days with NSC expansion medium (NS-A medium [Euroclone] with N2 supplement, 10 ng/ml EGF, 10 ng/ml bFGF [Invitrogen], 50 mg/ml bovine serum albumin (BSA) [BSA fraction; Gibco BRL], and 13 penicillin/streptomycin/glutamine). For differentiation of iPS cell-derived NSCs (iPS-NSCs) into all three neural subtypes, the NSCs were cultured for 1 week in neuronal differentiation medium (DMEM/F-12 medium [GIBCO], neurobasal medium [GIBCO], 1% FBS, 13 N2 supplement [GIBCO], 13 penicillin/streptomycin/glutamine [Invitrogen], and 503 B27 supplement).
Generation of Doxycycline-Inducible iPS Cells MEF cultures were established from E13.5 embryos from reprogrammable mice carrying one copy of Oct4-GFP transgenes and two copies of the OKSM cassette, the ROSA26M2rtTA allele [21, 22]. Reprogramming was performed in ESC medium in the presence of doxycycline (5 mg/ml). Ten days
www.StemCells.com
2597
after doxycycline treatment, Oct4-GFP-positive colonies were observed. Only Oct4-GFP-positive colonies were selected and transferred onto inactivated MEFs. For reprogramming experiments, doxycycline-inducible iPS (Di-iPS) cells were differentiated into NSCs and cultured in NSC medium or ESC medium in the presence of doxycycline (5 mg/ml).
RNA Isolation and qRT-PCR Analysis Total RNA was isolated using the RNeasy Mini Kit (Qiagen, Venlo, Netherlands, http://www.qiagen.com) and was treated with DNase to remove genomic DNA contamination. One microgram of total RNA was reverse-transcribed with SuperScript III Reverse Transcriptase Kit (Invitrogen) and oligo(dT) primer (Invitrogen) according to the manufacturer’s instructions. Quantitative polymerase chain reaction (PCR) reactions were set up in duplicate with the Power SYBR Green Master Mix (Takara) and analyzed with the Roche LightCycler 5480 (Roche). The primers for qRT-PCR used were as follows: Oct4 (endo) sense 50 -GATGCTGTGAGCCAAGGCAAG-30 , Oct4 (endo) antisense 50 -GGCTCCTGATCAACAGCATCAC-30 ; Nanog (endo) sense 50 -CTTTCACCTATTAAGGTGCTTGC-30 , Nanog (endo) antisense 50 -TGGCATCGGTTCATCATGGTAC-30 ; Nestin (endo) sense 50 -GGAACCCAGAGAATGTGGAA-30 , Nestin (endo) antisense 50 CACATCCTCCCACCTCTGTT-30 ; Dnmt1 sense 50 -CGG 0 0 CTCAAAGACTTGGAAAG-3 , Dnmt1 antisense 5 -TAGCCAGGTAG CCTTCCTCA-30 ; Dnmt3a sense 50 -GATGCTGTGAGCCAAGGCAAG30 , Dnmt3a antisense 50 -TGCTTGTTCTGCACTTCCAC-30 ; Dnmt3b sense 50 -ACTTGGTGATTGGTGGAAGC-30 , Dnmt3b antisense 50 CCAGAAGAATGGACGGTTGT-30 ; Oct4 TG sense 50 -GACGGCATCG CAGCTTGGATA-30 , Oct4 TG antisense 50 -CCAATACCTCTGAGC CTGGT-30 ; Sox2 TG sense 50 -GACGGCATCGCAGCTTGGATA-30 , Sox2 TG antisense 50 -CGCTTGGCCTCGTCGATGAA-30 ; Klf4 TG sense 50 -GACGGCATCGCAGCTTGGATA-30 , Klf4 TG antisense 50 GGGAAGTCGCTTCATGTGAG-30 ; c-Myc TG sense 50 -GACGGCATC GCAGCTTGGATA-30 , c-Myc TG antisense 50 -ACCGCAACATAG GATGGAGA-30 ; ACTB sense 50 -CGCCATGGATGACGATATCG-30 , and ACTB TG antisense 50 -CGAAGCCGGCTTTGCACATG-30 .
Bisulfite Genomic Sequencing To differentiate between methylated and unmethylated CG dinucleotides, genomic DNA was treated with sodium bisulfite to convert all unmethylated cytosine residues into uracil residues using the EpiTect Bisulfite Kit (QIAGEN) according to the manufacturer’s protocol. Briefly, purified genomic DNA (0.5–1 mg) was denatured at 99 C and then incubated at 60 C. After desulfonation, neutralization, and desalting, the modified DNA was diluted in 20 ml of distilled water. Subsequently, bisulfite PCR (BS-PCR) amplification was carried out using 1–2-ml aliquots of modified DNA for each PCR reaction. The primers used for BS-PCR were as follows [23]: Oct4 first sense 50 TTTGTTTTTTTATTTATTTAGGGGG-30 , Oct4 first antisense 50 -ATC CCCAATACCTCTAAACCTAATC-30 ; Oct4 second sense 50 -GGGTTG GAGGTTAAGGTTAGAGGG-30 , Oct4 second antisense 50 -CCCCC ACCTAATAAAAATAAAAAAA-30 ; Oct4 TG sense 50 -TTGCAAAGGAT TTTATATAGTTTT-30 , and Oct4 TG antisense 50 -AATCCCCAA TACCTCTAAACCTAA-30 . Briefly, the amplified products were verified by electrophoresis on a 1% agarose gel. The desired PCR products were used for subcloning using the TA cloning vector (pGEM-T Easy Vector, Promega). The reconstructed plasmids were purified, and individual clones were sequenced (Solgent Corporation). C AlphaMed Press 2014 V
2598
Immunocytochemistry Experiments For immunocytochemistry, cells were fixed with 4% paraformaldehyde for 20 minutes at room temperature. After the cells were washed with PBS, they were treated with PBS containing 10% normal goat serum and 0.03% Triton X-100 for 45 minutes at room temperature. The primary antibodies used were anti-nanog (nanog; monoclonal, 1:200, Abcam, Cambridge, UK, http://www.abcam.com), anti-Nestin (Nestin; monoclonal, 1:500, Millipore), anti-Sox2 (Sox2; polyclonal, 1:500, Millipore), anti-tubulin, bIII (Tuj1; monoclonal, 1:1,000, Millipore), anti-glial fibrillary acidic protein (GFAP) (monoclonal, 1:500, Millipore), and anti-O4 (O4; monoclonal, 1:200, Millipore). For detection of primary antibodies, fluorescently labeled (Alexa Fluor 488 or 568; Molecular Probes, Eugene, OR) secondary antibodies were used according to the specifications of the manufacturer.
Chimera Formation iPS cells were aggregated with denuded postcompacted eightcell-stage embryos to obtain an aggregate chimera. Eight-cell embryos flushed from 2.5-dpc B6D2F1 female mice were cultured in microdrops of embryo culture medium under mineral oil. The cells were trypsinized for 10 seconds, and clumps of iPS cells (4–10 cells) were selected and transferred into microdrops containing zona-free eight-cell embryos. Morula-stage embryos aggregated with iPS cells were cultured overnight at 37 C under 5% CO2. The aggregated blastocysts were transferred into one uterine horn of 2.5-dpc pseudopregnant recipients.
X-gal Staining For whole fetal embryo staining, the fetuses collected were rinsed with PBS and fixed with 4% formaldehyde for 1 hour at 4 C. Fetuses were rinsed three times at room temperature in PBS supplemented with 5 mM EGTA, 0.01% deoxycholate, 0.02% Nonidet P40, and 2 mM MgCl2. The specimens were washed with PBS and then stained in X-gal staining solution (PBS supplemented with 1 mg/ml 5-bromo-4-chloro-3-indolylgalactosidase [X-gal; Promega, Seoul, Rep of Korea http:// kr.promega.com], 5 mM K2Fe(CN)6, 5 mM K4Fe(CN)6, and 1 mM MgCl2). The blue staining was visualized by light microscopy.
RESULTS Derivation of NSCs from ESCs and iPS Cells In Vitro We generated iPS cells using a retroviral system and compared their pluripotency with the ESC line E14. The E14 ESCs were well-characterized pluripotent stem cells derived from an inbred 129/Ola strain [24]. iPS cells were derived from OG2/ ROSA26 heterozygous female MEFs, which contained Oct4-GFP and neo/LacZ transgenes. iPS cells were efficiently generated by retroviral transduction of four transcription factors (Oct4, Sox2, Klf4, and c-Myc) (Fig. 1A). Oct4-GFP-positive iPS cells (Fig. 1B) expressed high levels of endogenous Oct4, Nanog, Esrrb, and Prdm14 (Fig. 1C and Supporting Information Fig. S1) and formed germline chimeras (Fig. 1D), suggesting that these iPS cells were fully reprogrammed to pluripotent ES-like cells. To investigate the possibility that the silenced exogenous factors are reactivated during differentiation of iPS cells, we C AlphaMed Press 2014 V
Reacquisition of Pluripotency from iPSC-NSCs
differentiated retrovirus-mediated iPS cells into a specialized cell type, NSCs, which can be easily reprogrammed with fewer factors [25–27]. We differentiated ESCs (as a control) and iPS cells to NSCs using a previously described protocol with some modifications (Fig. 1A) [28]. ESCs and iPS cells were cultured in the absence of LIF and feeder cells for 2–3 days, serially cultured in N2B27 medium for 2–3 days, and then cultured in NS expansion medium containing EGF and bFGF for 3–4 days. When spherical embryoid bodies (EBs) were formed, they were cultured onto the gelatin-coated dish. EBs from ESCs and iPS cells formed neural tube-like rosettes at days 10 and 30, respectively, after induction of neural differentiation (Supporting Information Fig. S2A, S2B). These rosettes were picked and cultured in NSC culture medium, where the rosettes formed a monolayer of bipolar-shaped NSCs. The bipolar NSCs were picked in a new dish (passage 0) and further passaged. These cells were morphologically indistinguishable from NSCs derived from brain tissue (Fig. 1E); the cells rapidly propagated in the presence of EGF and bFGF and stained positive for the NSC markers Nestin and Sox2 (Fig. 1F), indicating that NSCs derived from ESCs and iPS cells were indistinguishable from brain-derived NSCs. To investigate the multipotent differentiation potential of iPS-NSCs, we induced differentiation of iPS-NSCs into neurons, astrocytes, and oligodendrocytes in a medium without EGF and bFGF. At 8–10 days in culture, the bipolar morphology of NSCs changed into a variety of polypolar cell types. The iPSNSCs stained positive for the Tuj1, GFAP, and O4 markers for neurons, astrocytes, and oligodendrocytes, respectively (Fig. 1G), indicating that iPS-NSCs could differentiate into all three neural subtypes.
Reactivation of Exogenous Factors Causes Regainment of Pluripotency in iPS-NSCs NSCs show sustained self-renewal for an expanded period in vitro in the presence of EGF and bFGF [28]. NSCs derived from ESCs and iPS cells were passaged every 3 days. The ESCderived NSCs (ES-NSCs) were maintained and propagated for more than 30 passages, for more than 3 months. Established iPS-NSCs maintained a typical NSC morphology for approximately 4 weeks. However, the morphology of iPS-NSCs #1 was changed at passages 8–9, and aggregated cells were observed (Fig. 2A), but not iPS-NSCs #2 and #3 (data not shown). We first tested whether the aggregates were neurospheres. Interestingly, most of the aggregated cells stained positive for Sox2 but not for Nestin (Fig. 2B), indicating that the aggregated cells were neither neurospheres nor NSCs. These cell aggregates gradually changed into ESC-like colonies in NSC culture medium. We therefore hypothesized that iPSNSCs spontaneously revert to a pluripotent state. However, the colony-forming cells did not reactivate Oct4-GFP, which was reactivated only after 4 days under ESC culture conditions (Fig. 2C and Supporting Information Fig. S4). Aggregates of iPS-NSCs (iPS-NSC-R) under NSC culture conditions also did not express endogenous Oct4 and Nanog until Oct4-GFP was reactivated (Fig. 2D). These results indicate that suitable culture conditions are also required for regaining pluripotency, in addition to the activation of transgenes. The Oct4-GFP-positive (iPS-NSC-GFP1) cells expressed the pluripotency markers Oct4 and Nanog, but not Nestin, as observed in pluripotent ESCs (Fig. 2D). Bisulfite DNA sequencing analysis showed the STEM CELLS
Choi, Kim, Choi et al.
2599
Figure 1. Generation of iPS cells and differentiation of NSCs derived from ESCs and iPS cells. (A): Timeline of generation of iPS cells and differentiation of NSCs. (B): Phase and fluorescence (Oct4-GFP) images of iPS cells from MEF. Scale bar 5 100 mm. (C): Expression levels of the pluripotency markers Oct4 and Nanog in iPS cells by qRT-PCR. Data are presented as mean 6 SD of triplicates (n 5 3). (D): iPS cells formed germline chimeras. A chimeric embryo from iPS cells was stained by X-gal, and Oct4-GFP-positive cells were detected in the gonad of the chimeric embryo. Scale bar 5 100 mm. (E): Morphology of brain-, ESC-, and iPS cell-derived NSCs. iPS cell-derived NSCs were morphologically indistinguishable from control brain-derived NSCs. Scale bar 5 100 mm. (F): Immunocytochemical results for the expression of the NSC markers Nestin and Sox2 in ESC- and iPS cell-derived NSCs. Scale bar 5 50 mm. (G): Differentiation potential of iPS cell-derived NSCs. Differentiated cells from iPS cell-derived NSCs were stained for markers of neurons (Tuj1; scale bar 5 50 mm), astrocytes (GFAP; scale bar 5 20 mm), and oligodendrocytes (O4; scale bar 5 20 mm). Abbreviations: GFAP, glial fibrillary acidic protein; GFP, green fluorescent protein; iPS, induced pluripotent stem; LIF, leukemia inhibitory factor; MEF, murine embryonic fibroblast; NSC, neural stem cell.
dynamics of DNA methylation on the Oct4 promoter region (Fig. 2E). The hypermethylated status of Oct4 promoter region of MEFs was demethylated after become an iPS cells and was then hypermethylated in iPS-NSCs on differentiation. The Oct4 promoter regions were demethylated in iPS-NSC-GFP1 cells, as shown in ESCs (Fig. 2E). Next, we analyzed a stringent marker, that is, X chromosome reactivation, for ground-state pluripotency. Xist RNA FISH analysis showed that all iPS-NSCs contained only inactive X chromosomes (Xi; detected as Xist cloud), indicating that the iPS-NSC population was not contaminated by undifferenti-
www.StemCells.com
ated cells (Fig. 2F); Xi was not observed in undifferentiated pluripotent cells, such as ESCs or iPS cells [29]. When the iPSNSCs were changed into iPS-NSC-GFP1 cells, two active X chromosomes (Xa; detected as pinpoint signals) were detected, as shown in female iPS cells. This result indicated that iPS-NSCs were completely differentiated from iPS cells and that undifferentiated cells were not contaminated in the iPS-NSC population. Moreover, iPS-NSC-GFP1 cells could form germline chimeras (Fig. 2G), indicating that NSCs differentiated from iPS cells could regain pluripotency and be regenerated into iPS cells. C AlphaMed Press 2014 V
2600
Reacquisition of Pluripotency from iPSC-NSCs
Figure 2. Regainment of pluripotency of NSCs derived from iPS cells. (A): Morphology of aggregates (iPS-NSC-R) from iPS cell-derived NSCs at passage 8. Scale bar 5 50 mm. (B): Expression of the NSC markers Nestin and Sox2 in iPS-NSC-R. Scale bar 5 50 mm. (C): Phase and fluorescence (Oct4-GFP) images of iPS-NSC-GFP1 cells at day 4 and passage 10. Scale bar 5 100 mm. (D): Gene expression levels of the pluripotency markers Oct4 (endo) and Nanog and the NSC marker Nestin in reprogramming to differentiation and reprogramming of cells. Data are presented as mean 6 SD of triplicates (n 5 3). (E): Bisulfite genomic sequencing of the promoter regions of Oct4. (F): Xist RNA FISH in fMEFs, iPS cells, iPS-NSCs, and iPS-NSC-GFP1 cells, using a Xist probe (red signal). (G): Formation of germline chimeras by iPS-NSC-GFP1 cells. A chimeric embryo from iPS-NSC-GFP1 cells was stained by X-gal. Scale bar 5 100 mm. Oct4-GFP-positive cells were detected in the gonad of the chimeric embryo. Scale bar 5 50 mm. Abbreviations: ESC, embryonic stem cell; GFP, green fluorescent protein; iPS, induced pluripotent stem; MEF, murine embryonic fibroblast; NSC, neural stem cell.
However, when iPS-NSCs were terminally differentiated into neural subtypes, cell aggregates or morphologic changes were not observed (Supporting Information Fig. S2), suggesting that regaining of pluripotency might occur only in easily reprogrammable cell types, such as NSC, but not in cell types that are difficult to reprogram, such as fully differentiated neurons [30].
Reactivation of Retroviral Genes Precedes the Reactivation of Endogenous Pluripotency Genes Retrovirus-mediated genes were silenced in ES and embryonic carcinoma (EC) cells [31] and fully reprogrammed iPS cells [2, 3, 6]. Here, we hypothesized that regainment of pluripotency in iPS-NSCs might be caused by reactivation of retroviral genes, which were silenced in the pluripotent state. Therefore, we examined the expression levels of transgenes (Oct4, Sox2, Klf4, and c-Myc) during the transformation of iPS-NSCs to iPS-NSCGFP1 cells. As expected, retrovirus-mediated transgenes were expressed in NSCs or MEFs shortly after retroviral transduction of transcription factors, but were not expressed in iPS cells after reprogramming (Fig. 3A). The silenced transgenes in iPS cells were reactivated after differentiation into iPS-NSCs and C AlphaMed Press 2014 V
were then downregulated in iPS-NSC-R (Fig. 3A); endogenous Oct4 was not expressed in both iPS-NSCs and iPS-NSC-R (Fig. 2D). The Sox2 transgene, which is a marker of NSCs and iPS cells, was more highly expressed in iPS-NSC-R (Fig. 3A). iPSNSC-GFP1 cells did not express retroviral transgenes; the transgenes were resilenced once the cells became pluripotent (Fig. 3A). However, only iPS-NSCs #1 were re-expressed transgenes, but not iPS-NSC #2 (Supporting Information Fig. S5). These data showed that reactivation of endogenous pluripotency-related genes (regainment of pluripotency) was associated with the reactivation of retroviral transgenes on differentiation.
Reactivation of Silenced Transgenes Is Associated With the Level of DNA Methyltransferases Epigenetic modifications are involved in the silencing of retroviral genes [32], which includes DNA methylation by DNA methyltransferases (Dnmts), such as Dnmt3a, Dnmt3b, and Dnmt1 [33–35]. During iPS cell induction, Dnmt3a and Dnmt3b levels gradually increased [36]. Dnmt1, which plays a role in maintaining DNA methylation, is involved in the silencing of retroviral genes in fully reprogrammed iPS cells [6]. We STEM CELLS
Choi, Kim, Choi et al.
2601
Figure 3. Reactivation of retroviral transgenes in iPS cell-derived NSCs, which was associated with the level of DNA methyltransferases. (A): qRT-PCR results for retroviral transgene (OSKM) expression levels in iPS cells, iPS-NSCs, iPS-NSC-R, and iPS-NSC-GFP1 cells. NSCs (4F) and MEF (4F) were control, infected retroviral four factors in, respectively, NSCs and MEF. Data are presented as mean 6 SD of triplicates (n 5 3). (B): qRT-PCR results for expression levels of Dnmt1, Dnmt3a, and Dnmt3b. Data are presented as mean 6 SD of triplicates (n 5 3). (C): Bisulfite genomic sequencing of regions of the retroviral transgene Oct4. Abbreviations: ESC, embryonic stem cell; GFP, green fluorescent protein; iPS, induced pluripotent stem; MEF, murine embryonic fibroblast; NSC, neural stem cell.
therefore examined the expression levels of Dnmt1, Dnmt3a, and Dnmt3b in iPS cells, iPS-NSCs, iPS-NSC-R, and iPS-NSCGFP1 cells (Fig. 3B). Dnmt1, Dnmt3a, and Dnmt3b, which were expressed at extremely low levels in MEFs, were upregulated after reprogramming into iPS cells (similar to the level in ESCs). The expression levels of Dnmt1, Dnmt3a, and Dnmt3b were downregulated after differentiation of iPS cells into iPS-NSCs. Dnmt3b expression was almost silenced in iPSNSCs like ES-derived NSCs (Supporting Information Fig. S6). These three Dnmts were upregulated during the transformation of iPS-NSCs into iPS-NSC-GFP1 cells (Fig. 3B). Considering that these Dnmts are responsible for the silencing of viral genes by inducing DNA methylation [6, 34, 35], we speculated that the DNA methylation status would change according to the levels of Dnmts. Therefore, we investigated the methylation status of the exogenous Oct4 promoter region in iPS
www.StemCells.com
cells, iPS-NSCs, iPS-NSC-R, and iPS-NSC-GFP1 cells by bisulfite genomic sequencing (Fig. 3C). The Oct4 transgene was found to be hypermethylated in iPS cells (high level of Dnmts), completely or hypomethylated in iPS-NSCs and iPS-NSC-R (low level of Dnmts), and hypermethylated in iPS-NSC-GFP1 cells (high level of Dnmts). The methylation status of Oct4 transgene was closely correlated with the expression level of the Oct4 transgene during the differentiation of iPS cells to iPSNSCs and the regainment of pluripotency in iPS-NSC-GFP1 cells (Fig. 3A, 3C).
Regainment of iPS Cells From iPS-NSCs was Closely Associated With Transgene Reactivation To confirm that iPS-NSCs spontaneously regain pluripotency because of transgene reactivation, we generated Di-iPS cells carrying Oct4-GFP, the OSKM (Oct4, Sox2, Klf4, and c-Myc) C AlphaMed Press 2014 V
Reacquisition of Pluripotency from iPSC-NSCs
2602
Figure 4. Regainment of pluripotency was closely associated with transgene reactivation. (A): Phase-contrast and fluorescence (Oct4GFP) images of Di-iPS cells. Scale bar 5 100 mm. (B): Immunocytochemistry results for the expression of the pluripotency marker Nanog in Di-iPS cells. Scale bar 5 50 mm. (C): Morphology of NSCs derived from Di-iPS cells and expression of the NSC markers Nestin and Sox2. Scale bar 5 50 mm. (D): Oct4-GFP1 cells from Di-iPS-NSCs were first observed at 15 days after doxycycline treatment in ES medium. Scale bar 5 50 mm. (E): Gene expression levels of endogenous Oct4, and Nanog in MEFs, Di-iPS cells, Di-iPS-NSCs (Dox2), DiiPS-NSCs (Dox1, day 3), Di-iPS-NSC-GFP1 cells, ESCs, control NSCs, and DW by RT-PCR. Abbreviations: DW, distilled water; ESC, embryonic stem cell; GFP, green fluorescent protein; iPS, induced pluripotent stem; MEF, murine embryonic fibroblast; NSC, neural stem cell.
cassette, and the transcriptional activator within the ROSA26 locus (ROSA26-M2rtTA) [21, 22]. Oct4-GFP-positive Di-iPS cells were generated from reprogrammable MEFs at day 10 after doxycycline induction (Fig. 4A). Once the Di-iPS cells were established, they maintained pluripotency without doxycycline. The Di-iPS cells expressed the pluripotency marker Nanog (Fig. 4B) and formed germline chimeras (Supporting Information Fig. S9A–S9C). Next, we induced differentiation of Di-iPS cells to NSCs (Di-NSCs). Di-NSCs without doxycycline maintained a NSC morphology and expressed the NSC markers Nestin and Sox2 (Fig. 4C); we observed neither aggregated cells nor Oct4-GFP-positive cells in long-term cultures of Di-NSCs (for more than 20 passages). However, when we cultured the Di-NSCs in doxycycline-containing medium, the cells aggregated to form ESC-like colonies and expressed Oct4-GFP (Fig. 4D and Supporting Information Fig. S10). Although we were unable to observe iPS-NSC-R-like aggregates, we cannot exclude the possibility of an intermediate state during the doxycyclineinducible reprogramming. These Oct4-GFP-positive cells expressed the pluripotency markers Oct4 (endo) and Nanog (Fig. 4E), indicating that Di-NSCs were reprogrammed into pluC AlphaMed Press 2014 V
ripotent state by transgene reactivation. Similar results were observed in another Di-iPS cell line, that is, in PB-iPS cells (Supporting Information Fig. S11A–S11C) [37]. These results indicate that NSCs differentiated from iPS cells regain pluripotency upon reactivation of exogenous factors.
DISCUSSION Reprogramming into iPS cells by transduction of reprogramming factors (more than a week) is a much slower step than that in cell fusion between ESCs and somatic cells (1–2 days). Therefore, initial reprogramming by exogenous factors might enhance proliferation and chromatin remodeling [17]. Transcription factors such as Oct4, Sox2, and c-Myc interact with chromatin remodeling factors [19], which leads to upregulation of endogenous pluripotency-related genes and repression of tissue-specific genes [38]. Once pluripotency is achieved, reprogrammed cells are controlled by endogenous regulatory circuits [17–19], and exogenous factors become inactive. In this study, we showed that the silenced exogenous factors STEM CELLS
Choi, Kim, Choi et al.
2603
were derepressed as the iPS cells lose pluripotency, which results in regainment of pluripotency. A previous study showed that transgenes were silenced in iPS-NSCs, which were propagated up to 67 passages [39]. Conversely, retroviral c-Myc often became reactivated in differentiated cells derived from iPS cells in chimeric mice, which subsequently led to tumor formation [4, 5]. Our in vitro system suggested that retroviral genes became reactivated following downregulation of dnmts during differentiation of iPS cells, which subsequently regained pluripotency. Therefore, silenced genes in a pluripotent environment can be activated through changes in the cell state. iPS cell-derived neurospheres containing undifferentiated Nanog-GFP-positive cells could form significantly larger teratomas when transplanted into NOD/SCID mice [40]. However, the regainment of the pluripotency of iPS-NSCs should not be attributed to contamination of undifferentiated cells in the iPS-NSC population. First, iPS-NSCs were cultured under NSC monolayer culture conditions for approximately 4 weeks, and a bipolar morphology was maintained. Moreover, to obtain a pure NSC population, only NSCs were picked and cultured at the initial stage of NSC derivation. Second, cell aggregate formation was observed in most of the NSC colonies simultaneously (Supporting Information Fig. S2). If a small number of undifferentiated cells were present, aggregates should be formed only in a few sites and in a separate colony from NSCs. Third, Xist RNA FISH analysis showed that the iPS-NSC population contained only differentiated cells (Fig. 1F); all cells contained Xi, which was not observed in undifferentiated pluripotent cells [29]. Taken together, these findings showed that iPS-NSCs were a completely differentiated cell population and that no contaminating undifferentiated cells were present in the iPS-NSC population. We were able to induce differentiation of iPS cells into NSCs and establish three NSC cell lines (iPS-NSCs #1, #2, and #3). iPS-NSCs #2 and #3 were maintained and proliferated over 30 passages. However, iPS-NSC #1 spontaneously aggregated and showed reactivation of Oct4-GFP after 8–9 passages; that is, only iPS-NSCs #1 had transgene reactivation. The mechanism underlying this phenomenon is unclear; however, it is possible that transgene integration sites affect the reactivation of silenced transgenes. Four retroviral factors were silenced again in reprogrammed iPS-NSC-GFP1 cells. However, terminally differentiated cells maintain their morphology without proliferation (Supporting Information Fig. S2). Therefore, regainment of pluripotency in iPS cell-derived differentiated cells might occur only in easily reprogrammable cell types, such as somatic stem cells, including NSCs. It has been suggested that retrovirus-mediated genes are silenced in ESCs, EC, and iPS cells during reprogramming [2, 3, 6, 31]. The silencing of retroviral transgenes is associated with de novo DNA methylation on CpG dinucleotides [41–43]. Silenced M-MLV proviruses introduced in EC cells could be reactivated after differentiation by combination treatment with retinoic acid and 5-azacytidine, a DNA demethylating agent [44]. DNA methylation is crucial for maintenance of
REFERENCES 1 Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 2006;126:663–676.
www.StemCells.com
gene silencing rather than establishing silencing [34, 35]. In this study, we also showed that DNA methylation on retroviral transgene was important for maintenance of silencing of the transgenes in the iPS cell-derived differentiated cells. Oct4 transgene regions were hypomethylated in iPS-NSCs and in aggregates (iPS-NSC-R; the initial stage of regainment of pluripotency), but were hypermethylated in reprogrammed iPSNSC-GFP1 cells (Fig. 3C). Previous research has demonstrated that retroviral silencing is independent of Dnmt3a and Dnmt3b in pluripotent stem cells [33]. However, we clearly showed that low levels of Dnmt3a and Dnmt3b were closely correlated with the DNA methylation status of retroviral transgenes in differentiated cells from iPS cells.
CONCLUSION In this study, iPS-NSCs were compared with NSCs derived from ESCs and brain tissue. NSCs derived from ESCs and iPS cells were morphologically indistinguishable from brain-derived NSCs, expressed the NSC markers Nestin and Sox2, and could be differentiated into all three neural subtypes: neurons, astrocytes, and oligodendrocytes. These iPS-NSCs spontaneously aggregated upon long-term culture in vitro, formed ESC-like colonies, and finally reactivated Oct4-GFP. The spontaneously reverted GFPpositive cells showed characteristics similar to those of ESCs and iPS cells. This phenomenon could be attributed to transgene reactivation during the differentiation of iPS cells. On the basis of these results, we suggest that exogenous gene-free iPS cells should be considered for therapeutic application.
ACKNOWLEDGMENTS This work was supported by the Biomedical Technology Development Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (Grant No. 20110019489) and the Next-Generation BioGreen 21 Program (Grant No. PJ00956202) funded by the Rural Development Administration, Republic of Korea.
AUTHOR CONTRIBUTIONS H.W.C.: conception and design, collection and/or assembly of data, data analysis and interpretation, and manuscript writing; J.S.K., S.C., Y.J.H., and M.J.K.: collection and/or assembly of data; H.G.S.: provision of study material or patients; J.T.D.: conception and design, financial support, data analysis and interpretation, manuscript writing, and final approval of manuscript.
DISCLOSURE
OF
POTENTIAL CONFLICTS
OF INTEREST
The authors indicate no potential conflicts of interest.
2 Maherali N, Sridharan R, Xie W et al. Directly reprogrammed fibroblasts show global epigenetic remodeling and widespread tissue contribution. Cell Stem Cell 2007;1:55– 70.
3 tion tent 4 tion
Okita K, Ichisaka T, Yamanaka S. Generaof germline-competent induced pluripostem cells. Nature 2007;448:313–317. Aoi T, Yae K, Nakagawa M et al. Generaof pluripotent stem cells from adult
C AlphaMed Press 2014 V
Reacquisition of Pluripotency from iPSC-NSCs
2604
mouse liver and stomach cells. Science 2008; 321:699–702. 5 Nakagawa M, Koyanagi M, Tanabe K et al. Generation of induced pluripotent stem cells without Myc from mouse and human fibroblasts. Nat Biotechnol 2008;26: 101–106. 6 Wernig M, Meissner A, Foreman R et al. In vitro reprogramming of fibroblasts into a pluripotent ES-cell-like state. Nature 2007; 448:318–324. 7 Kaji K, Norrby K, Paca A et al. Virus-free induction of pluripotency and subsequent excision of reprogramming factors. Nature 2009;458:771–775. 8 Warren L, Manos PD, Ahfeldt T et al. Highly efficient reprogramming to pluripotency and directed differentiation of human cells with synthetic modified mRNA. Cell Stem Cell 2010;7:618–630. 9 Yu J, Hu K, Smuga-Otto K et al. Human induced pluripotent stem cells free of vector and transgene sequences. Science 2009;324: 797–801. 10 Kim D, Kim CH, Moon JI et al. Generation of human induced pluripotent stem cells by direct delivery of reprogramming proteins. Cell Stem Cell 2009;4:472–476. 11 Zhou H, Wu S, Joo JY et al. Generation of induced pluripotent stem cells using recombinant proteins. Cell Stem Cell 2009;4: 381–384. 12 Hou P, Li Y, Zhang X et al. Pluripotent stem cells induced from mouse somatic cells by small-molecule compounds. Science 2013; 341:651–654. 13 Brambrink T, Foreman R, Welstead GG et al. Sequential expression of pluripotency markers during direct reprogramming of mouse somatic cells. Cell Stem Cell 2008;2: 151–159. 14 Okita K, Nakagawa M, Hyenjong H et al. Generation of mouse induced pluripotent stem cells without viral vectors. Science 2008;322:949–953. 15 Zhang H, Jiao W, Sun L et al. Intrachromosomal looping is required for activation of endogenous pluripotency genes during reprogramming. Cell Stem Cell 2013;13:30– 35. 16 Apostolou E, Ferrari F, Walsh RM et al. Genome-wide chromatin interactions of the Nanog locus in pluripotency, differentiation, and reprogramming. Cell Stem Cell 2013;12: 699–712.
17 Do JT, Scholer HR. Regulatory circuits underlying pluripotency and reprogramming. Trends Pharmacol Sci 2009;30:296–302. 18 Jaenisch R, Young R. Stem cells, the molecular circuitry of pluripotency and nuclear reprogramming. Cell 2008;132:567–582. 19 Loh YH, Wu Q, Chew JL et al. The Oct4 and Nanog transcription network regulates pluripotency in mouse embryonic stem cells. Nat Genet 2006;38:431–440. 20 Zaehres H, Daley GQ. Transgene expression and RNA interference in embryonic stem cells. Methods Enzymol 2006;420:49–64. 21 Stadtfeld M, Maherali N, Borkent M et al. A reprogrammable mouse strain from gene-targeted embryonic stem cells. Nat Methods 2010;7:53–55. 22 Polo JM, Anderssen E, Walsh RM et al. A molecular roadmap of reprogramming somatic cells into iPS cells. Cell 2012;151:1617–1632. 23 Kim MJ, Choi HW, Jang HJ et al. Conversion of genomic imprinting by reprogramming and redifferentiation. J Cell Sci 2013; 126:2516–2524. 24 Hooper M, Hardy K, Handyside A et al. HPRT-deficient (Lesch-Nyhan) mouse embryos derived from germline colonization by cultured cells. Nature 1987;326:292–295. 25 Kim JB, Sebastiano V, Wu G et al. Oct4induced pluripotency in adult neural stem cells. Cell 2009;136:411–419. 26 Shi Y, Desponts C, Do JT et al. Induction of pluripotent stem cells from mouse embryonic fibroblasts by Oct4 and Klf4 with smallmolecule compounds. Cell Stem Cell 2008;3: 568–574. 27 Do JT, Joo JY, Han DW et al. Generation of parthenogenetic induced pluripotent stem cells from parthenogenetic neural stem cells. Stem Cells 2009;27:2962–2968. 28 Conti L, Pollard SM, Gorba T et al. Niche-independent symmetrical self-renewal of a mammalian tissue stem cell. PLoS Biol 2005;3:e283. 29 Yang X, Smith SL, Tian XC et al. Nuclear reprogramming of cloned embryos and its implications for therapeutic cloning. Nat Genet 2007;39:295–302. 30 Kim J, Lengner CJ, Kirak O et al. Reprogramming of postnatal neurons into induced pluripotent stem cells by defined factors. Stem Cells 2011;29:992–1000. 31 Pannell D, Ellis J. Silencing of gene expression: Implications for design of retrovirus vectors. Rev Med Virol 2001;11:205–217.
32 Ellis J, Yao S. Retrovirus silencing and vector design: Relevance to normal and cancer stem cells? Curr Gene Ther 2005;5:367– 373. 33 Pannell D, Osborne CS, Yao S et al. Retrovirus vector silencing is de novo methylase independent and marked by a repressive histone code. EMBO J 2000;19:5884–5894. 34 Yao S, Sukonnik T, Kean T et al. Retrovirus silencing, variegation, extinction, and memory are controlled by a dynamic interplay of multiple epigenetic modifications. Mol Ther 2004;10:27–36. 35 Ramunas J, Montgomery HJ, Kelly L et al. Real-time fluorescence tracking of dynamic transgene variegation in stem cells. Mol Ther 2007;15:810–817. 36 Stadtfeld M, Maherali N, Breault DT et al. Defining molecular cornerstones during fibroblast to iPS cell reprogramming in mouse. Cell Stem Cell 2008;2:230–240. 37 Woltjen K, Michael IP, Mohseni P et al. piggyBac transposition reprograms fibroblasts to induced pluripotent stem cells. Nature 2009;458:766–770. 38 Liang J, Wan M, Zhang Y et al. Nanog and Oct4 associate with unique transcriptional repression complexes in embryonic stem cells. Nat Cell Biol 2008;10:731–739. 39 Onorati M, Camnasio S, Binetti M et al. Neuropotent self-renewing neural stem (NS) cells derived from mouse induced pluripotent stem (iPS) cells. Mol Cell Neurosci 2010;43:287–295. 40 Miura K, Okada Y, Aoi T et al. Variation in the safety of induced pluripotent stem cell lines. Nat Biotechnol 2009;27:743–745. 41 Jahner D, Stuhlmann H, Stewart CL et al. De novo methylation and expression of retroviral genomes during mouse embryogenesis. Nature 1982;298:623–628. 42 Challita PM, Kohn DB. Lack of expression from a retroviral vector after transduction of murine hematopoietic stem cells is associated with methylation in vivo. Proc Natl Acad Sci USA 1994;91:2567–2571. 43 Stewart CL, Stuhlmann H, Jahner D et al. De novo methylation, expression, and infectivity of retroviral genomes introduced into embryonal carcinoma cells. Proc Natl Acad Sci USA 1982;79:4098–4102. 44 Niwa O, Yokota Y, Ishida H et al. Independent mechanisms involved in suppression of the Moloney leukemia virus genome during differentiation of murine teratocarcinoma cells. Cell 1983;32:1105–1113.
See www.StemCells.com for supporting information available online.
C AlphaMed Press 2014 V
STEM CELLS
