Biomaterials xxx (2014) 1e10

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Reprogramming of mouse somatic cells into pluripotent stem-like cells using a combination of small molecules Phil Jun Kang a,1, Jai-Hee Moon a,1, Byung Sun Yoon a, b,1, Solji Hyeon a, Eun Kyoung Jun a, b, Gyuman Park a, Wonjin Yun a, Jiyong Park a, Minji Park a, Aeree Kim c, Kwang Youn Whang d, Gou Young Koh e, Sejong Oh f, Seungkwon You a, * a Laboratory of Cell Function Regulation, Division of Biotechnology, College of Life Sciences and Biotechnology, Korea University, Seoul 136-701, Republic of Korea b StemLab, Venture Incubation Center Korea University, Seoul 136-701, Republic of Korea c Department of Pathology, College of Medicine, Korea University Guro Hospital, Seoul 152-703, Republic of Korea d Division of Biotechnology, College of Life Sciences and Biotechnology, Korea University, Seoul 136-701, Republic of Korea e National Research Laboratory of Vascular Biology and Stem Cells, Korea Advanced Institute of Science and Technology, Daejeon 305-701, Republic of Korea f Division of Animal Science, Chonnam National University, Gwangju 500-757, Republic of Korea

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 April 2014 Accepted 5 May 2014 Available online xxx

Somatic cells can be reprogrammed to generate induced pluripotent stem cells (iPSCs) by overexpression of four transcription factors, Oct4, Klf4, Sox2, and c-Myc. However, exogenous expression of pluripotency factors raised concerns for clinical applications. Here, we show that iPS-like cells (iPSLCs) were generated from mouse somatic cells in two steps with small molecule compounds. In the first step, stable intermediate cells were generated from mouse astrocytes by Bmi1. These cells called induced epiblast stem cell (EpiSC)-like cells (iEpiSCLCs) are similar to EpiSCs in terms of expression of specific markers, epigenetic state, and ability to differentiate into three germ layers. In the second step, treatment with MEK/ERK and GSK3 pathway inhibitors in the presence of leukemia inhibitory factor resulted in conversion of iEpiSCLCs into iPSLCs that were similar to mESCs, suggesting that Bmi1 is sufficient to reprogram astrocytes to partially reprogrammed pluripotency. Next, Bmi1 function was replaced with Shh activators (oxysterol and purmorphamine), which demonstrating that combinations of small molecules can compensate for reprogramming factors and are sufficient to directly reprogram mouse somatic cells into iPSLCs. The chemically induced pluripotent stem cell-like cells (ciPSLCs) showed similar gene expression profiles, epigenetic status, and differentiation potentials to mESCs. Ó 2014 Elsevier Ltd. All rights reserved.

Keywords: Bmi1 Induced pluripotent stem cell (iPSC) Small molecule compounds Reprogramming

1. Introduction Somatic cells have been reprogrammed into induced pluripotent stem cells (iPSCs) that closely resemble embryonic stem cells (ESCs) by the introduction of a defined set of transcription factors (Oct4, Sox2, Klf4 and c-Myc (OSKM) [1]. Over the past few years, various approaches which enhance reprogramming efficiency and/ or replace reprogramming factors have successfully been developed to create iPSCs [2e6]. For example, CHIR99021 and PD0325901 enhance the completion and efficiency of the * Corresponding author. Tel.: þ82 232 903 057; fax: þ82 232 903 507. E-mail address: [email protected] (S. You). 1 These authors contributed equally to this work.

reprogramming process [7]. BIX01294 together with BayK8644 compensate for viral transduction of reprogramming factors and improve reprogramming efficiency [8]. Even though safety issue still needs to be addressed for small molecule-based reprogramming, it can overcome some limitations of transcription factorbased reprogramming methods. Therefore, identification of small molecules to substitute for reprogramming factors would provide alternative approaches to move toward a more efficient and safe reprogramming process [6]. Recently, Oct4 was replaced by small molecule which make possible to generate iPSCs only with small molecules combination in mouse somatic cells [9,10]. Mouse embryonic stem cells (mESCs) and post-implantation epiblast stem cells (EpiSCs) represent two phases in the ontogeny of pluripotent stem cells. Both mESCs and EpiSCs exhibit

http://dx.doi.org/10.1016/j.biomaterials.2014.05.015 0142-9612/Ó 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Kang PJ, et al., Reprogramming of mouse somatic cells into pluripotent stem-like cells using a combination of small molecules, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.05.015

2

P.J. Kang et al. / Biomaterials xxx (2014) 1e10

pluripotency; however, EpiSCs, unlike mESCs, readily form teratomas, but rarely form chimeras [11,12]. Thus, the pluripotent state of mESCs is termed ‘naïve’ and that of EpiSCs is termed ‘primed’ [13]. The development of ground state culture conditions utilizing small molecule inhibitors of MEK/ERK and GSK3 signaling (2i) has enhanced the generation of both mESCs and iPSCs. Furthermore, mESCs can be induced to differentiate into EpiSCs by exposure to activin A and bFGF, but the reverse transition requires transfection with Klf4 or other reprogramming factors [14,15]. Recent studies indicate that the transcription factors OSKM can not only generate iPSCs under mESC culture conditions, but also induce the formation of EpiSCs (termed induced EpiSCs or iEpiSCs) under EpiSC culture conditions [1,16]. These results demonstrate that a simple modification of the extrinsic signaling conditions can determine whether overexpression of OSKM will give rise to iPSCs or iEpiSCs. Furthermore, OSKM also termed reprogramming factors, can result in the efficient trans-differentiation of fibroblasts into functional neural stem/progenitor cells or cardiomyocytes provided that appropriate culture conditions are used [17,18]. Thus, the culture environment in transcription factor-mediated reprogramming determines the fate of the reprogrammed cell. This study aimed to find small molecules that can replace and/or enhance expression of exogenous reprogramming factors and enable the generation of iPS-like cells (iPSLC) from somatic cells in chemically defined conditions. We first established stable induced epiblast stem cell-like cells (iEpiSCLCs) during dedifferentiation of mouse astrocytes by Bmi1, which is not only essential for selfrenewal of stem cells but is also required for the cellular reprogramming process itself [19e25]. Using these stable iEpiSCLCs, we identified combinations of two small molecules that can enable reprogramming of iEpiSCLCs into iPSLCs. Next, we reprogrammed mouse somatic cells into iPSLCs using the same two small molecules in combination with activators of the Shh signaling pathway [22,24]. This study highlights the usefulness of our phenotypic screening approach in identifying small molecules that can effectively compensate for reprogramming factors. 2. Materials and methods

For PM2i-iPSLC, OM2i-iPSLC generation, MEFs (2  105) are seeded in six-well plate before induction. At day 1, medium is replaced in NSC/iEpiSCLC medium containing purmorphamine (0.5, 1 mM; Calbiochem), or 25-hydroxycholesterol (oxysterol; 0.1, 0.5 mM; Sigma) and incubated for 3 days. Next, transfer the cells to uncoated culture plate and form sphere for 7 days and dissociate carefully. Single cells were seeded in gelatin coated plate with 2i/LIF condition and incubated for 7 days. Finally, cells were seeded on MEF feeder with conventional mESC condition. mESC like colonies were observed at 7 days after seeding on feeder. 2.2. Retroviral transduction Astrocytes were infected with a retrovirus produced from the PT67 amphotropic packaging cell line (Clontech, Palo Alto, CA, USA) transfected with retroviral vectors [22,23], using Lipofectamine 2000 (Life Technologies), according to the manufacturer’s protocol. Astrocytes plated 24 h earlier at a density of 1  106 cells/10 cm dish were transduced by reseeding them with pre-filtered (0.45 mm) retroviral supernatant containing 6 mg/ml polybrene (SigmaeAldrich, St Louis, MO, USA). This step was repeated twice. 2.3. Flow cytometry Astrocytes, mESCs, and iEpiSCLCs were trypsinized and 1  106 cells were incubated with anti-SSEA1 antibody (EMD Millipore, Billerica, MA, USA) at 4  C for 1 h, and resuspended in 100 ml of Cy3-labeled secondary antibody (1:200, Jackson Immunoresearch). The cells were then washed with PBS containing 1% FBS and resuspended in fixative solution for FACS analysis. Immunoglobulin G and Cy3labeled secondary antibodies (1:200, Jackson Immunoresearch) were used to determine nonspecific signals. Cells to be probed for internal markers such as Oct4 were fixed with 0.1% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA, USA) for 10 min and permeabilized with 90% methanol for 30 min on ice. Fixed cells (1e5  105) were probed for 1 h with a 1:10 dilution of anti-Oct4 antibody or isotype control antibody diluted in PBS containing 1% FBS and analyzed using a FACS Calibur flow cytometer. The data were analyzed with the Cell Quest 3.0 software (BD Biosciences). All treatments were performed in triplicate. 2.4. RT-PCR and qRT-PCR Total RNA was extracted using the Trizol reagent (Life Technologies) and used to synthesize cDNA using an RT premix (Bioneer, Daejeon, South Korea) and oligo-dT (Life Technologies), according to the manufacturer’s instructions. For qRT-PCR, first-strand cDNA was generated using 500 ng RNA, RT premix (Bioneer), and oligo-dT (Life Technologies) using the SYBR supermix (Bio-Rad, Hercules, CA, USA). Subsequent PCRs were performed in a final volume of 20 ml containing 1 ml of cDNA and 1 ml of 10 pM primers in PCR-Premix (Bioneer). Primer sequences and the reaction conditions used in this study are listed in Supplementary Table S4. For each sample, the ratio of the mRNA level to that of the GAPDH mRNA was calculated. At least three independent samples were analyzed.

2.1. Cell culture

2.5. Western blot analysis

Astrocytes were isolated from the forebrain of individual embryonic (E13.5e 18.5), neonatal (1e5 days old), or adult (CF1 strain; 10e12 weeks old) mice, as described previously [26], and cultured in DMEM (Life Technologies, Gaithersburg, MD, USA) containing 10% FBS (Life Technologies), 2 mM L-glutamine, and 3 g/l Dglucose. The dedifferentiation process was observed by following the morphological changes as flat polygonal cells changed into bipolar cells and spherical aggregates in the final culture [22,27]. Mouse fetal NSCs were isolated from the subventricular zone of the brain of E13.5 embryos as described previously [22,27]. Primary astrocytes were plated at a clonal density of 2500e5000 cells/cm2 in NSC culture medium. NSC/iEpiSCLC medium consisted of DMEM/F-12 (1:1) containing N2 (Life Technologies), penicillin/streptomycin (Cambrex Bioscience, Walkersville, MD, USA), 10 ng/ml bFGF (R&D Systems, Minneapolis, MN, USA), with/without 10 ng/ml EGF (R&D Systems), respectively. Primary NSCs and Bmi1-expressing neurospheres were passaged by dissociation of the spheres into single cells using trituration through a fire polished pipette [22,27]. Floating cells were cultured as suspensions in untreated 60 mm diameter plates (BD Biosciences, Franklin Lakes, NJ, USA) at densities of 1e3  105 cells/plate in NSC medium. Cells were maintained in this medium and the growth factor was replaced daily. The expansion protocol was repeated every 3e4 days, as previously described [23,26,28]. For secondary sphere formation assays, single cells from spheres were plated at a density of 100 cells/well in 12-well plates and the number of single cell-derived spheres was counted after 14 days [23]. Six of these independently derived cultures were studied in the dedifferentiation experiments. MEFs were isolated from the uteri of pregnant mice (E13.5). Embryos were washed with phosphate-buffered saline (PBS) and the head, heart, and spinal cord were dissected from the isolated embryos. The remainder of the embryo was minced using scissors and forceps, treated with trypsin/EDTA solution, and incubated at 37  C for 2e3 min. After incubation, the sample was briefly triturated to produce a single cell suspension. Cells from each embryo were cultured in a T75 flask with fresh medium at 37  C and in 5% CO2. MEFs collected within the first three passages were used to avoid replicative senescence.

Total protein was extracted using RIPA buffer containing a protease inhibitor cocktail (Roche Molecular Diagnostics, Pleasanton, CA, USA). The proteins were separated by SDS-PAGE on a 4e12% gradient-precast gel and transferred onto a PVDF membrane (EMD Millipore). The membrane was incubated with the indicated primary antibody (Supplementary Table S5) followed by HRP-conjugated secondary antibodies against mouse, rabbit, or goat immunoglobulin G. The secondary antibodies were detected with the SuperSignal West Pico Kit (Thermo Fisher Scientific, Rockford, IL, USA). 2.6. Alkaline phosphatase (AP) staining and immunofluorescence analysis AP staining was performed using the Alkaline Phosphatase Detection Kit (EMD Millipore) according to the manufacturer’s instructions. Immunofluorescence analysis was performed as previously described [23]. The primary antibodies used in this study are listed in Supplementary Table S5. 2.7. Microarray analysis A microarray analysis was performed using the Mouse II Genome 430 2.0 GeneChip arrays (Affymetrix, Santa Clara, CA, USA) essentially as described previously [29]. The experiment was performed in triplicate for mESCs, astrocytes, uninduced cells (Bmi1-expressing astrocytes), and iEpiSCLCs. Normalization was calculated with the RMA algorithm26 and implemented in the Bioconductor software (Bioconductor, Seattle, WA, USA). 2.8. Bisulfite genomic sequencing analysis A Genomic DNA Purification Kit (Promega, Fitchburg, WI, USA) and EpiTect Bisulfite Kit (Qiagen, Hilden, Germany) were used for the isolation and sodium bisulfite conversion of genomic DNA from MEFs, mESCs, un-induced cells, and iEpiSCLCs. Treated DNA was amplified and cloned using the pGEM-T Easy vector (Promega) and sequenced using the T7 forward and SP6 reverse primers. The

Please cite this article in press as: Kang PJ, et al., Reprogramming of mouse somatic cells into pluripotent stem-like cells using a combination of small molecules, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.05.015

P.J. Kang et al. / Biomaterials xxx (2014) 1e10

A

B

3

C

OCT4 Akt

SOX2 OCT4

KLF4 C-MYC NANOG

p-Akt

SOX2

MEK

KLF4

p-MEK

REX1

ERK

C-MYC

SAL4

p-ERK

tubulin

E-RAS

β-actin

SOX15 FBX15 GAPDH

bFGF

+

U0126

+

Ly294002

-

p-Akt Akt p-MEK MEK

+

+

10μM 20μM -

-

+

+

-

-

10μM 20μM

E

80 60 50

**

40 30

**

20

**

**

0

ERK

bFGF

+

U0126

-

Ly294002

-

OCT4

β-actin

70

10

p-ERK

SOX2

90

Number of colonies (diameter >200μm)

D

+

+

10μM 20μM -

-

+

+

-

-

10μM 20μM

Fig. 1. NSCLCs undergo rapid but incomplete conversion toward a pluripotent state. (A) Expression levels of mESC marker genes in mESCs, bFGF-dependent NSCLCs, and EGFdependent NSCLCs. (B) Protein levels of pluripotency marker genes (Oct4, Sox2, c-Myc, and Klf4) in mESCs, bFGF-dependent NSCLCs and EGF-dependent NSCLCs. (C) Activation of bFGF intracellular signaling pathway in bFGF-dependent NSCLCs. Western blot analysis of the phosphorylation levels of Akt, MEK, and ERK in bFGF- and EGF-dependent NSCLCs. (D) Western blot analysis of the phosphorylation levels of Akt, MEK, and ERK in bFGF-dependent NSCLCs grown under NSC culture conditions in the presence of the PI3K inhibitor LY294002 or the MEK inhibitor U0126. (E) Number of neurospheres formed by NSCLCs grown under NSC culture conditions and treated with U0126 or LY294002. **P < 0.01 compared to control. primers used for promoter fragment amplification are listed in Supplementary Table S4. 2.9. ChIP assay ChIP was performed on astrocytes, mESCs, iEpiSCLCs, and iPSLCs using antiacetyl H3 and anti-dimethyl K9 H3 antibodies with the EZ ChIP Kit (EMD Millipore) according to the manufacturer’s instructions. 2.10. In vitro differentiation of iEpiSCLCs To assess the multipotency of iEpiSCLCs in vitro, cells were plated at a density of 2.5  104 cells/cm2 in 4-well plates (Nalge Nunc International Corporation, Rochester, NY, USA) pre-coated with poly-L-ornithine and laminin (SigmaeAldrich), as described previously [30e34]. Briefly, for differentiation into astrocytes, cells were plated in astrocytic induction medium composed of DMEM containing 10% FBS, bFGF (10 ng/ml), and CNTF (10 ng/ml) (Upstate Biotechnology, Lake Placid, NY) or EGF (10 ng/ml) for 3e7 days. For neuronal differentiation, the cells were plated in neuronal induction medium composed of N2 medium supplemented with B27 and containing either 1 mM retinoic acid (SigmaeAldrich), 10 mM retinoic acid, 1 mM valproic acid (Sigma), or 10 mM valproic acid for 7e14 days [31]. Oligodendrocytes were differentiated in N2 medium supplemented with PDGF-AA (10 ng/ml), T3 (30 nM), bFGF (10 ng/ml), and EGF (10 ng/ml) (R&D) for 21 days, or with NT-3 (10 ng/ ml) (Upstate Biotechnology) and bFGF (10 ng/ml) for 21 days [30,32,33]. The cells were fixed and then stained with corresponding primary antibodies, which were detected with secondary antibodies (Supplementary Table S5). The cells were counterstained with DAPI (SigmaeAldrich) to identify nuclei. For adipogenic differentiation, the cells were plated in adipogenic medium consisting of DMEM

supplemented with 10% FBS, 10 ng/ml insulin, 1 M dexamethasone, 0.5 mM 3isobutyl-1-methylxanthine, 1 M hydrocortisone, and 0.1 mM indomethacin (all chemicals from SigmaeAldrich) for 21 days [35], and then stained with Oil Red O (SigmaeAldrich). For osteogenic differentiation, cells were seeded in osteogenic medium consisting of IMDM supplemented with 0.1 M dexamethasone, 10 mM bglycerolphosphate, and 0.2 mM ascorbic acid (SigmaeAldrich) for 21 days and then stained with alizarin red S. For pancreatic b-cell differentiation, cells were cultured in DMEM supplemented with 1  ITS, 1  N2 supplement, 20 ng/ml bFGF, and 10 mM nicotinamide for 7 days, and then stained with dithizone. For hepatocytic differentiation, cells were seeded in hepatocyte medium and analyzed by immunofluorescence staining [36]. 2.11. Teratoma formation and chimera formation iEpiSCLCs, GFP-iEpiSCLCs, and Oct4p-GFP-iEpiSCLCs (1  106 per mouse) were injected under the kidney capsule or testis of Balb/c nude mice [22]. The mice were sacrificed 8e10 weeks later and the teratomas were harvested. For histological analysis, tissues were embedded in paraffin, sectioned, and stained with hematoxylin and eosin [22,27]. For chimera formation, 4e5-week-old female mice (C57BL/6) were induced to superovulate by intraperitoneal injection of 7.5 IU PMSG and 48 h later by intraperitoneal injection of 7.5 IU hCG, and were mated with a male mouse (C57BL/6). Blastocysts were collected 3.5 days after a vaginal plug was detected by flushing the oviducts into H-CZB medium. Either GFP-iEpiSCLCs or Oct4p-GFPiEpiSCLCs were expelled from the injection pipette against the ICM of a blastocyst. Injected blastocysts were transferred into the uterine horn of 2.5 day post-coitum pseudopregnant CD1 female mice that had been mated with vasectomized male mice [22].

Please cite this article in press as: Kang PJ, et al., Reprogramming of mouse somatic cells into pluripotent stem-like cells using a combination of small molecules, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.05.015

E-RAS FBx15 Utf-1 Dppa4 Dppa5 Ftl-17 Gdf3 GAPDH

SSEA1/DAPI

103

Counts

RT(-)

Induced

SSEA1

Counts

0.56% M1

101

102 FL2-H

103

0 100

104

101

200 160 120 80 40 0 100

102 FL2-H

104

103

104

16.88% M1

101

102 FL2-H

103

104

H

Oct4

Nanog

Fbx15

6.5% (5/77)

22.7% (10/44)

11.1% (4/36)

14 SOX2

12 10 8 6

Oct4 MYC

NANOG Klf4

4

Oct4 Uninduced iEpiSCLCs

Nanog Uninduced iEpiSCLCs

Fbx15 Uninduced iEpiSCLCs

2 0 2

6 8 10 12 4 ES (log2 intensity) 89.6% 90.5% (69/77) (95/105)

iEpiSCLCs

87.5% 64.6% (42/48) (31/48)

76.6% 66.7% (49/64) (32/48)

diMeK9H3

IgG

diMeK9H3

AcH3

Uninduced

F

AP

AcH3

AP

ES

IgG

Sox2/DAPI

IgG

SSEA1/DAPI

G

Oct4/DAPI

Sox2/DAPI

AcH3

Oct4/DAPI

80 40 0 100

102 FL2-H

M1

80 40

ES

Ectoderm

Bmi1-neurosphere

ES

200 160 120

Mesoderm

diMeK9H3

E

GAPDH

101

15.27%

120

M1

80 40 0 100

Rex1 TGDF1

0.09%

120

Epiblast

200 160

Counts

SAL4

Endoderm Lefty1 Otx2

Oct4

REX1

Klf4

Bmi1-neurosphere

Uninduced 200 160

Counts

NANOG

Eomes

Bmi1-neurosphere (log 2 intensity)

SOX2

D

Acr2b

GATA4 SOX7 MIXL1 CXCR4 Foxa2 CER1 SOX17 ES Brachyury FLT-1 CAD11 KDR MEOX1 Keratin5 SOX1 Nefl GAPDH

OCT4

EpiSC

C

RT(-)

ES

Induced

Uninduced

B RT(-)

ES

Ast

A

Induced

P.J. Kang et al. / Biomaterials xxx (2014) 1e10 Uninduced

4

Oct4 Nanog Sox2 Fig. 2. Molecular characterization of Bmi1-neurospheres. (A, B) Expression of mESC and germ layer marker genes in mESCs, astrocytes, un-induced cells (Bmi1-transduced astrocytes cultured in 10% FBS), and Bmi1-neurospheres, as detected by RT-PCR. (C) Expression of epiblast marker gene in mEpiSCs and Bmi1-neurospheres, as detected by RT-PCR. (D) FACS analysis demonstrated the expression of SSEA1 and Oct4 in un-induced cells and Bmi1-neurosphere. (E) AP staining and immunocytochemistry revealed the expression of pluripotent markers in mESCs and Bmi1-neurospheres. Scale bar, 200 mm. (F) Scatter-plot comparison of the global gene expression profiles of iEpiSCLCs and mESCs. (G) The Oct4, Nanog, and Sox2 promoters were analyzed by ChIP for the dimethylation status of lysine 9 of histone H3 and the acetylation status of histone H3 in mESCs, un-induced cells, and iEpiSCLCs. Data was normalized using b-actin. (H) Bisulfite genomic sequencing of the Oct4, Nanog, and Fbx15 promoter regions in iEpiSCLCs, un-induced cells, and mESCs.

3. Results 3.1. Induction of astrocytes to iEpiSCLCs with Bmi1 To identify small molecules that can replace and/or enhance expression of pluripotent genes and enable the generation of iPSCs from somatic cells in chemically defined conditions, we first established stable induced epiblast stem cell-like cells (iEpiSCLCs) during dedifferentiation of mouse astrocytes by Bmi1, which is not only essential for self-renewal of stem cells but is also required for the cellular reprogramming process itself [21e23]. We previously induced astrocytes into neural stem cell-like cells (NSCLCs) with Bmi1 [23]. Under NSC sphere culture conditions, the addition of bFGF, but not EGF, markedly increased the number of neurosphere formation, the expression of mESC and EpiSC markers, including Oct4, as well as the phosphorylation of AKT and MEK/ERK (Fig. 1AeC). However, treatment of cells with the MEK/ERK inhibitor U0126 or the AKT inhibitor LY294002 decreased the expression of Oct4 and Sox2 and the number of neurospheres (Fig. 1D and E). To better define the developmental state of Bmi1-neurospheres, gene expression was

assessed by performing RT-PCR and Western blot analyses. Bmi1neurospheres expressed the undifferentiated mESC markers at day 7. By contrast, levels of the astrocyte markers GFAP and S100 were reduced by day 7 (Fig. 2A; Supplementary Fig. S1). Transcripts from genes associated with the epiblast and the early germ layers, such as Otx2, Eomes, Foxa2, Brachyury (T), Sox17, and Cer were also expressed at higher levels in Bmi1-neurospheres than in uninduced cells or mESCs (Fig. 2B and C). Expression of markers shared with mESCs, including Oct4, Sox2, Nanog, and SSEA1, were also detected in Bmi1-neurospheres by immunocytochemistry and FACS analysis (Fig. 2D and E). The global gene expression profiles of Bmi1-neurospheres were similar to that of mESCs, but distinct from that of astrocytes (Fig. 2F and Supplementary Fig. S2), and the patterns of acetylation of histone H3 and of dimethylation of lysine 9 of histone H3 in the Oct4, Nanog, and Sox2 promoters of induced cells were similar to those of mESCs, but differed from those of uninduced cells (Fig. 2G). Furthermore, induced cells were capable to differentiate into cell and tissues of all three germ layers (Fig. 3 and Supplementary Fig. S3), suggesting that these bFGF dependentpluripotent stage cells had similar molecular characteristics to those of EpiSCs and designated them herein as iEpiSCLCs. However,

Please cite this article in press as: Kang PJ, et al., Reprogramming of mouse somatic cells into pluripotent stem-like cells using a combination of small molecules, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.05.015

P.J. Kang et al. / Biomaterials xxx (2014) 1e10

3.2. Generation of homogenous single cell-derived, GFP-tagged iEpiSCLC clones (GFP-iEpiSCLCs) To generate homogenous, single cell-derived GFP-expressing colonies (GFP-iEpiSCLCs) were selected, and subsequently induced by culture in NSC culture conditions as previously described

A

B

adipocytes

(Supplementary Fig. S6A and B). Ten GFP-iEpiSCLC clones were established, and each clone was subjected to RT-PCR and Western blot analyses to examine the expression of markers for the three embryonic germ layers (Supplementary Fig. S6CeE). When GFPiEpiSCLCs were transplanted into nude mice, teratomas were produced, and all three embryonic germ layers were histologically detected in the teratomas (Supplementary Fig. S6F). To further assess the developmental potential of GFP-iEpiSCLCs, cells were injected into E3.5 C57BL/6 blastocysts. As judged by the coat-color contribution, however, no chimeras were present in the five pups (Supplementary Fig. S6G). When we examined their potential to contribute to embryo development via morula aggregation, integration of GFP-iEpiSCLCs into the preimplantation embryo was not detected (Supplementary Fig. S6H). 3.3. Generation of Oct4 promoter-derived GFP-iEpiSCLCs (Oct4pGFP-iEpiSCLCs) To understand why iEpiSCLCs do not generate chimeric mice, and to investigate whether the lack of chimera formation might be due to the presence of a small proportion of pluripotent iEpiSCLCs, we first introduced Oct4 promoter-derived-GFP (Oct4p-GFP) reporter genes into iEpiSCLCs and then examined the expression of the GFP transgene [16]. GFP expression was evident in iEpiSCLCs, and we purified GFP-positive cells by single cell cloning as

C

osteoblasts

SS/Glucagon/DAPI Insulin/PDX1/DAPI

the methylation status of Oct4, Nanog, and Fbx15 promoters was predominantly demethylated in the mESC, whereas the methylation status of these promoters in iEpiSCLC was partially demethylated. (Fig. 2H), indicating this intermediate stage would not be fully reprogrammed yet. Similar results were observed with neonatal astrocytes (N-iEpiSCLCs) or adult astrocytes (A-iEpiSCLCs) (Supplementary Fig. S4). We next asked whether Bmi1 could convert astrocytes into iEpiSCLCs and generate homogeneous cell populations, a doxycycline (DOX)-inducible retroviral vector expressing a Bmi1-EGFP fusion (pBI-Bmi1-EGFP) was transduced into astrocytes. When cells were treated with DOX under NSC culture conditions, neurospheres expanded in the presence of DOX, expressed epiblast marker genes, and exhibited reorganization of the methylation pattern at the Oct4, Nanog, and Fbx15 gene promoters (Supplementary Fig. S5). These results indicate that DOX-inducible iEpiSCLCs (I-iEpiSCLCs) exhibit similar characteristics to those of iEpiSCLCs (Supplementary Fig. S5).

5

DTZ staining

Undiff

Diff

Diff

TH/DAPI

O1/DAPI

Tuj1/DAPI

GFAP/DAPI

Undiff

D E

Ectoderm

Endoderm

Mesoderm

Mesoderm

SMA/Map2/DAPI

SMA/AFP/DAPI

Perilipin/DAPI

SMA/NG2/CD31/DAPI

Albumin/CK18/DAPI Fig. 3. Differentiation potential of iEpiSCLCs. (A) Immunofluorescence staining of ectodermal lineage markers in differentiated iEpiSCLCs. Scale bar, 200 mm. (B) Oil Red O and alizarin red S staining of differentiated cells was used to identify adipocytes and osteoblasts, respectively. Scale bar, 200 mm. (C) Dithizone (DTZ) staining and immunofluorescence staining for insulin, Pdx1, somatostatin, and glucagon. Scale bar, 200 mm. (D) The hepatic differentiation of iEpiSCLCs. Morphological changes of iEpiSCLCs at different stages of differentiation. At day 14, the differentiated cells expressed the hepatic cell markers albumin and CK18 as shown by immunofluorescence staining. Scale bar, 200 mm. (E) The in vivo developmental potential of iEpiSCLCs. Teratomas generated from iEpiSCLCs differentiated into neural epithelium (ectoderm), gut epithelium (endoderm), and endothelial cells, muscle and fat (mesoderm). Hematoxylin and eosin-stained sections (top panels) and immunohistochemistry (bottom panels) of teratomas derived from iEpiSCLCs at 8e10 weeks after injection into a nude mouse. Scale bar, 200 mm.

Please cite this article in press as: Kang PJ, et al., Reprogramming of mouse somatic cells into pluripotent stem-like cells using a combination of small molecules, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.05.015

ES

200

20

80

160

15 10

60

M1

40

0 10

0

10

10 FL2-H

SSEA1/DAPI

OCT4 SOX2 NANOG REX1 SAL4 E-RAS Fbx15 Utf-1 Dppa4 Dppa5 Ftl-17 Gdf3 GAPDH

AP

E Uninduced

Oct4p-GFPiEpiSCLCs

Uninduced

Oct4p-GFPiEpiSCLCs

G

0 10

10

10 FL2-H

10

10

Oct4

ES

Stella GDF3 NANOS C-KIT DAZL VASA GATA4 SOX7 MIXL1 CXCR4 Foxa2 SOX17 Brachyury FLT-1 CAD11 KDR MEOX1 Keratin5 SOX1 Nefl GAPDH

Germ cell

Endoderm

Mesoderm

Ectoderm

Mesoderm

Oct4p-GFPiEpiSCLCs Brain WT

E

Lung

E

Liver

E

Kidney

WT

WT

WT

Endoderm

iEpiSCLCs-GFP

Ectoderm

WT

F

10

Intestin

Spleen E

Oct4p-GFPiEpiSCLCs

Fbx15

E

Uninduced

Nanog

WT

Oct4

E

AP

10

GFP GAPDH

TH/DAPI

CK18/AFP/DAPI

SMA/NG2/CD31/DAPI

Fig. 4. Generation and characterization of Oct4p-GFP-iEpiSCLCs. (A) Generation of Oct4p-GFP-iEpiSCLCs from iEpiSCLCs. Scale bar, 200 mm. (B) mESC marker gene expression in mESCs and GFP subpopulations was determined by AP staining and immunofluorescence analysis. Scale bar, 200 mm. (C) FACS analysis verified the expression of SSEA1 and Oct4 in Oct4p-GFP-iEpiSCLCs. (D) Expression of mESC marker and germ layer marker genes in Oct4p-GFP-iEpiSCLCs. (E) The epigenetic status of Oct4, Nanog, and Fbx15 promoters in Oct4p-GFP-iEpiSCLCs. (F) Formation of teratomas derived from Oct4p-GFP-iEpiSCLCs. Scale bar, 200 mm. (G) Coat-color chimera from Oct4p-GFP-iEpiSCLCs (top) and analysis of GFP DNA integration into the chromosomes of tissues by PCR using genomic DNA (bottom).

P.J. Kang et al. / Biomaterials xxx (2014) 1e10

SSEA1/DAPI

Uninduced iEpiSCLCs Oct4p-GFP-iEpiSCLCs RT(-)

Sox2/DAPI

ES Ast

Oct4/DAPI

WT

Sox2/DAPI

M1

80

SSEA1

Oct4p-GFPiEpiSCLCs

Oct4p-GFP- iEpiSCLCs

120

40

20

5

D

Oct4/DAPI

Counts

100

ES Ast Uninduced iEpiSCLCs Oct4p-GFP-iEpiSCLCs RT(-)

B

25

Counts

GFP+ Sphere forming efficiency (%)

Oct4p-GFPiEpiSCLCs

C

30

6

Please cite this article in press as: Kang PJ, et al., Reprogramming of mouse somatic cells into pluripotent stem-like cells using a combination of small molecules, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.05.015

A

DAY 10

DAY 21-28

B Phase

AP

SSEA1/DAPI

Oct4/DAPI

Nanog/DAPI

Sox2/DAPI

2iR-iPSLC

ES

DAY 3

E

**

**

** Oct4

**

FL2-H

Oct4

Sox2

0.5

M1

Tubulin

2iR-iPSLC 0.0% M1

FL2-H

96.24% M1

73.86% M1

FL2-H

0.2% M1

95.58%

Counts

**

85.69%

Counts

**

MEF

2iR-iPSLC BC1

1.5

MEF

Counts

ES

MEF MEF

Counts

**

1

D

ES ES

Counts

**

2

Counts

iPSLC

ES

Relative gene expression

C

iEpiSCLC

2iR-iPSLC

Fibroblast / Astrocyte

SSEA1

Reprogramming

M1

0 Oct3/4 Sox2 Nanog c-Myc

ES

Oct4

Klf4

MEF

E-Ras Rex1

2iR-iPSLC

Sal4

G

FL2-H

FL2-H

FL2-H

2iR-iPSLC

F

Nestin/DAPI endoderm

mesoderm

SMA/DAPI

P.J. Kang et al. / Biomaterials xxx (2014) 1e10

Foxa2/DAPI

ectoderm

Nanog 2iR-iPSLC

Please cite this article in press as: Kang PJ, et al., Reprogramming of mouse somatic cells into pluripotent stem-like cells using a combination of small molecules, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.05.015

A DAY 0

7

Fig. 5. Generation and characterization of 2iR-iPSLCs. (A) Schematic representation of the reprogramming process. (B) AP staining and immunofluorescence analysis of mESC markers in mESCs and 2iR-iPSLCs. Scale bar, 200 mm. (C, D) Gene expression of mESCs and 2iR-iPSLCs determined by qRT-PCR and Western blot analyses. **P < 0.01 compared to MEFs. (E) FACS analysis revealed the expression of SSEA1 and Oct4 in mESCs, MEFs, and 2iR-iPSLCs. (F) Bisulfite genomic sequencing of the Oct4 and Nanog promoter regions in mESCs, MEFs and 2iR-iPSLCs. (G) In vitro differentiation (top panel) and in vivo teratoma formation (bottom panel) by 2iR-iPSLCs. Scale bar, 200 mm.

8

P.J. Kang et al. / Biomaterials xxx (2014) 1e10

A

DAY 3

DAY 10

B

DAY 21-28

PM2i-iPSLC

ES

DAY 0

Fibroblast / Astrocyte

1.5 1

ES

** **

MEF

iEpiSCLC PM2i-iPSLC

**

**

** **

0.5 0

**

ES MEF OM2i-iPSLC

**

1

** **

**

0.5

AP

F

ES

SSEA1/DAPI

MEF

PM2i iPSLC

Oct4/DAPI

Sox2/DAPI

Nanog/DAPI

OM2i iPSLC

0

2 1 c s g l4 /4 f4 t3 Sox ano -My Kl -Ra Rex Sa c E Oc N

E ES

PM2i-IiPSLC

MEF

D

OM2i-iPSLC

1 2 c s g l4 /4 f4 t3 Sox ano -My Kl -Ra Rex Sa c E Oc N

iPSLC

**

1.5

PM 2i-iPSLC

**

2

Relative gene expression

Relative gene expression

C

OM2i-iPSLC

Reprogramming

Oct4

Oct4 Nanog

16 Sox2

14

Klf4

Nanog

C-Myc

12 10 8 8

10

12

14

16

ES Nanog

Sox2

Klf4

14 Oct4 12

C-Myc

10 8 6 4

4

6

8

10

12

14

Mesoderm

Ectoderm

Endoderm

PM 2i- iPSLC

PM 2i-iPSLC

16

G

16

OM 2i- iPSLC

6 6

MEF

Fig. 6. Generation and characterization of chemically induced iPSLCs. (A) Schematic representation of the small molecule-based reprogramming process. (B) AP staining and immunofluorescence analysis of mESC markers in mESCs, PM2i-iPSLCs and OM2i-iPSLCs. Scale bar, 200 mm. (C) Gene expression of mESCs, MEFs, PM2i-iPSLCs and OM2i-iPSLCs determined by qRT-PCR analyses. **P < 0.01 compared to MEFs. (D) Microarray analysis demonstrates the similarity between PM2i-iPSLCs, OM2i-iPSLCs and mESCs in terms of gene expression. (E) Scatter-plot comparison of the global gene expression profiles of mESCs, MEFs and PM2i-iPSLCs. (F) Bisulfite genomic sequencing of the Oct4 and Nanog promoter regions in mESCs, MEFs, PM2i-iPSLCs and OM2i-iPSLCs. (G) In vivo teratoma formation assay by PM2i-iPSLCs and OM2i-iPSLCs. Scale bar, 200 mm.

described above (Fig. 4A). The Oct4p-GFP-iEpiSCLCs population was comparable to that of mESCs with respect to the expression of mESC markers (Fig. 4BeD), methylation patterns of Oct4 promoters (Fig. 4E), and the potential to differentiate into three germ layers (Fig. 4F). Furthermore, injection of GFP-positive cells into blastocysts generated a few chimeras, but none of these showed germ line transmission (Fig. 4G), confirming that full pluripotency had not been attained. Taken together, these results suggest that Bmi1 reprogrammed mouse astrocytes into partially pluripotent Oct4pGFP-iEpiSCLCs. Nevertheless, the lack of a naïve state in Oct4p-GFPiEpiSCLCs is not surprising because EpiSCs also display Oct4 expression [11,12]. 3.4. Generation of iPSLCs from iEpiSCLCs To attain more naïve pluripotency, recent studies suggest that the additional introduction of Klf4 and a change in culture conditions can revert primed EpiSCs to a naïve mESC-like state [7,16]. However, unlike wild type EpiSCs, iEpiSCLCs expressed Klf4

(Fig. 2C). Therefore, we determined whether the Oct4p-GFPiEpiSCLCs could be induced to attain ground state pluripotency by changing culture conditions. After transfer under several mESC culture conditions, Oct4p-GFP-iEpiSCLCs underwent massive differentiation and death (Supplementary Fig. S7). It has been suggested that DZNep may account for its role in Oct4 activation and in the presence of 2i, Oct4 and Sox2 may activate other pluripotentrelated genes, along with the activation of Nanog [9]. Moreover, 2i/LIF condition creates an optimal environment for promoting the reversion of EpiSCs into iPSCs [7,16]. Consequently, culture of Oct4p-GFP-iEpiSCLCs in 2i/LIF condition produced a few colonies (Fig. 5A). Two colonies were picked, expanded and exhibited an undifferentiated morphology (Fig. 5B). Cells derived from these two colonies were designated 2i-reverted-iPS-like cells (2iR-iPSLCs), which were then analyzed by the same tests described above. The 2iR-iPSLCs retained mESC-specific gene expression and the epigenetic modifications of the Oct4 and Nanog promoters were similar to those in mESCs (Fig. 5CeF). The 2iR-iPSLCs differentiated into derivatives of the three germ layers, as revealed by the formation of

Please cite this article in press as: Kang PJ, et al., Reprogramming of mouse somatic cells into pluripotent stem-like cells using a combination of small molecules, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.05.015

P.J. Kang et al. / Biomaterials xxx (2014) 1e10

teratomas (Fig. 5G), however, no chimeras were formed following injection of 2iR-iPSLCs into blastocysts (data not shown). 3.5. iPSLCs obtained from Bmi1-transduced MEFs cultured under NSC conditions and treated with 2i/LIF have pluripotent properties characteristic of mESCs To confirm that induction of iEpiSCLCs by Bmi1 transduction and 2i/LIF treatment can induce MEFs to become iPSLCs, we used essentially the same experiment, with the primary MEFs were used instead of astrocytes. Similar to astrocytes, MEFs transduced with Bmi1 using a combination of 2i/LIF treatment generated iPSLCs, and these colonies (designated Bmi1-transduced MEFs with 2i-treatediPSLCs; BM2i-iPSLCs) expressed typical pluripotency markers and formed derivatives of all three germ layers (Supplementary Fig. S8). 3.6. Generation of iPSLCs from MEFs by treatment with Shh agonists and 2i/LIF and culture under NSC conditions We recently demonstrated that oxysterol and purmorphamine not only stimulate the Shh pathway but also replace the factor of Bmi1 [22]. Similar to generation of BM2i-iPSLCs, treatment of MEFs with oxysterol or purmorphamine under NSC culture conditions activated the Shh pathway and reprogrammed MEFs into NSCLCs that exhibited gene expression profiles characteristic of NSCs [22]. Furthermore, subsequent treatment of NSCLCs with 2i/LIF enabled the reprogramming of MEFs to pluripotent stem cells [combinations of oxysterol or purmorphamine-treated MEFs treated with 2i/ LIF, hereafter designated OM2i-iPSLCs (0.0003%) or PM2i-iPSLCs (0.0023%), respectively] (Fig. 6A), although the efficiency of reprogramming was very low compared to that of reprogramming achieved by small molecule combinations [9]. The PM2i-iPSLCs and OM2i-iPSLCs retained mESC-specific gene expression (Fig. 6B and C), and similar with mESCs in global gene expression (Fig. 6D and E). Furthermore, these cells were similar to mESCs in the epigenetic modifications of the Oct4 and Nanog promoters and able to differentiate into cells of the three primary germ layers (Fig. 6F and G), but no chimeras were formed following injections. Taken together, these results demonstrated that MEFs can be reprogrammed to pluripotency by combinations of activators of the Shh pathway (oxysterol or purmorphamine) and 2i/LIF. 4. Discussion Our data support two main conclusions. First, Bmi1 not only replaces Oct4, Sox2, Klf4, and c-Myc, but also represses expression of p16Ink4a and p19Arf during reprogramming of mouse astrocytes into iEpiSCLCs. Moreover, iEpiSCLCs can be further reprogrammed into more ground state by culture in 2i/LIF-containing medium (2iR-iPSLCs and BM2i-iPSLCs). Second, a combination of small molecules reprograms MEFs into iPSLCs in the absence of any other genetic reprogramming factors. We previously demonstrated that mouse fibroblasts and astrocytes can be converted into NSCLCs by Bmi1 under NSC culture conditions [23]. Moreover, these cells can be further induced to form iPSCs upon Oct4 expression [22,24]. In this study, we demonstrated that a stable intermediate state, iEpiSCLCs, was generated from astrocytes by Bmi1 in the presence of bFGF. The important role played by Bmi1 in cellular reprogramming is consistent with previous studies showing that Bmi1 can directly reprogram fibroblasts into NSC-like cells and can replace the reprogramming factors Sox2, c-Myc, and Klf4 [22,23]. The present data indicate that activation of bFGF signaling is critical not only for induction but also for self-renewal of iEpiSCLCs [37]. The precise mechanism underlying bFGF action in the self-renewal of iEpiSCLCs

9

is not completely understood. Consistent with recent reports [11,12,16], iEpiSCLCs had similar epigenetic and developmental potentials to those of EpiSCs and iEpiSCs [16]. Furthermore, this study demonstrates that single cell-derived iEpiSCLCs clones can generate derivatives of all three germ layers in vitro and in vivo, thereby excluding the possibility of a mixed cell population. However, Oct4p-GFP-iEpiSCLCs engaged in a low level of chimera formation without germ line transmission, suggesting that full potency was not achieved. The limited contribution of iEpiSCLCs to chimeras, as seen previously with EpiSCs [11,12], may be caused by developmental heterogenicity, which limits the ability of these cells to colonize host embryos. EpiSCs and iEpiSCs can revert back to an mESC-like state if they are forced to overexpress Klf4 or are treated with small molecule inhibitors under strict mESC culture conditions [14,16,37,38]. Inhibition of signaling pathways by small molecules, including inhibitors of bFGF/ERK and GSK3 signaling, in cooperation with LIF/ STAT3 activation, results in the reversion of EpiSCs to a mESC-like state. Cells in the initial phase of this reversion, iEpiSCLCs and Oct4p-GFP-iEpiSCLCs, could be further converted into 2iR-iPSLCs in cultures supplemented with 2i/LIF, and thus acquire an iPSC-like status [13,16]. These cells were comparable to mESCs or iPSCs in terms of their morphology and gene expression, and could be maintained extensively and clonally without spontaneous differentiation. However, these cells were characterized by incomplete reprogramming (2iR-iPSLCs) and were unable to contribute to chimera formation, as observed with the original Fbx15-iPSCs [1]. Moreover, we also performed the chimeric assay of chemically induced iPSLC clones (OM2i-iPSLCs and PM2i-iPSLCs) that have differentiation potential for three germ layers in vitro and in vivo using at least 100 blastocysts injections per each clones. We, however, could not obtain chimeras from OM2i-iPSLCs and PM2iiPSLCs because of abortions or infanticides. These results indicated iPSLCs have partial pluripotency that makes impossible to contribute for individual organism but makes possible to differentiate into several lineage types of somatic cells. Therefore our next studies will focus on elevating pluripotent stage of these iPSLCs using other chemical compounds such as epigenetic modulators. Signaling through the bFGF/ERK pathway may be essential for the induction of the neuroectodermal lineage from mESCs [39]. However, bFGF/ERK signaling must be suppressed by ERK inhibitors before iEpiSCLCs are induced into a 2iR-iPSC-like state. Thus, we propose the existence of two distinct steps in the process of reprogramming from astrocytes to 2iR-iPSLCs via iEpiSCLCs. The first step (somatic cells to iEpiSCLCs) is promoted by bFGF/ERK and the second step (iEpiSCLCs to 2iR-iPSLCs) is promoted by inhibition of bFGF/ERK signaling. Taken together, our results demonstrate that stable intermediate iEpiSCLCs can be induced by Bmi1 in the presence of bFGF/ERK signaling, and that these cells can be further induced to generate 2iR-iPSLCs under 2i/LIF culture conditions. 5. Conclusions Recently, it has been reported that pluripotent stem cells were generated from mouse somatic cells using a combination of seven small molecule compounds [9]. These findings, with our study, increase our understanding about reprogramming toward pluripotency using only a combination of small molecules. Furthermore, our phenotypic screening based reprogramming strategy can be used for screening chemical compounds that can effectively compensate for reprogramming factors. During reprogramming of somatic cells into iPSCs, pluripotent iPSCs could be isolated from cultured somatic cells by the selection of the rare reprogrammed iPSCs based upon the reactivation of Fbx15, Oct4, or Nanog. These

Please cite this article in press as: Kang PJ, et al., Reprogramming of mouse somatic cells into pluripotent stem-like cells using a combination of small molecules, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.05.015

10

P.J. Kang et al. / Biomaterials xxx (2014) 1e10

approaches are of potential therapeutic interest, but translation of human systems would be hindered by the requirement for genetic modification. The usefulness of phenotypic screening would minimize or avoid the genetic modifications for reprogramming of somatic cells into iPSCs. Acknowledgments This work was supported by the Bio & Medical Technology Development Program of the National Research Foundation of Korea funded by the Korean Ministry of Science, ICT & Future Planning (MSIP) NRF-2010-0020347, a grant of the Korea Health Technology R&D Project, Ministry of Health & Welfare Grant A120392 and School of Life Sciences and Biotechnology for BK21 PLUS, Korea University. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.biomaterials.2014.05.015. References [1] Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 2006;126: 663e76. [2] Kim D, Kim CH, Moon JI, Chung YG, Chang MY, Han BS, et al. Generation of human induced pluripotent stem cells by direct delivery of reprogramming proteins. Cell Stem Cell 2009;4:472e6. [3] Okita K, Nakagawa M, Hyenjong H, Ichisaka T, Yamanaka S. Generation of mouse induced pluripotent stem cells without viral vectors. Science 2008;322:949e53. [4] Stadtfeld M, Nagaya M, Utikal J, Weir G, Hochedlinger K. Induced pluripotent stem cells generated without viral integration. Science 2008;322:945e9. [5] Zhou H, Wu S, Joo JY, Zhu S, Han DW, Lin T, et al. Generation of induced pluripotent stem cells using recombinant proteins. Cell Stem Cell 2009;4: 381e4. [6] Zhu S, Li W, Zhou H, Wei W, Ambasudhan R, Lin T, et al. Reprogramming of human primary somatic cells by OCT4 and chemical compounds. Cell Stem Cell 2010;7:651e5. [7] Silva J, Barrandon O, Nichols J, Kawaguchi J, Theunissen TW, Smith A. Promotion of reprogramming to ground state pluripotency by signal inhibition. PLoS Biol 2008;6:e253. [8] Shi Y, Desponts C, Do JT, Hahm HS, Scholer HR, Ding S. Induction of pluripotent stem cells from mouse embryonic fibroblasts by Oct4 and Klf4 with small-molecule compounds. Cell Stem Cell 2008;3:568e74. [9] Hou P, Li Y, Zhang X, Liu C, Guan J, Li H, et al. Pluripotent stem cells induced from mouse somatic cells by small-molecule compounds. Science 2013;341: 651e4. [10] Li W, Tian E, Chen ZX, Sun G, Ye P, Yang S, et al. Identification of Oct4activating compounds that enhance reprogramming efficiency. Proc Natl Acad Sci U S A 2012;109:20853e8. [11] Brons IG, Smithers LE, Trotter MW, Rugg-Gunn P, Sun B, Chuva de Sousa Lopes SM, et al. Derivation of pluripotent epiblast stem cells from mammalian embryos. Nature 2007;448:191e5. [12] Tesar PJ, Chenoweth JG, Brook FA, Davies TJ, Evans EP, Mack DL, et al. New cell lines from mouse epiblast share defining features with human embryonic stem cells. Nature 2007;448:196e9. [13] Nichols J, Smith A. Naive and primed pluripotent states. Cell Stem Cell 2009;4: 487e92. [14] Guo G, Yang J, Nichols J, Hall JS, Eyres I, Mansfield W, et al. Klf4 reverts developmentally programmed restriction of ground state pluripotency. Development 2009;136:1063e9. [15] Yang J, van Oosten AL, Theunissen TW, Guo G, Silva JC, Smith A. Stat3 activation is limiting for reprogramming to ground state pluripotency. Cell Stem Cell 2010;7:319e28.

[16] Han DW, Greber B, Wu G, Tapia N, Arauzo-Bravo MJ, Ko K, et al. Direct reprogramming of fibroblasts into epiblast stem cells. Nat Cell Biol 2011;13: 66e71. [17] Efe JA, Hilcove S, Kim J, Zhou H, Ouyang K, Wang G, et al. Conversion of mouse fibroblasts into cardiomyocytes using a direct reprogramming strategy. Nat Cell Biol 2011;13:215e22. [18] Kim J, Efe JA, Zhu S, Talantova M, Yuan X, Wang S, et al. Direct reprogramming of mouse fibroblasts to neural progenitors. Proc Natl Acad Sci U S A 2011;108: 7838e43. [19] Esteban MA, Wang T, Qin B, Yang J, Qin D, Cai J, et al. Vitamin C enhances the generation of mouse and human induced pluripotent stem cells. Cell Stem Cell 2010;6:71e9. [20] Liu J, Cao L, Chen J, Song S, Lee IH, Quijano C, et al. Bmi1 regulates mitochondrial function and the DNA damage response pathway. Nature 2009;459: 387e92. [21] Molofsky AV, Pardal R, Iwashita T, Park IK, Clarke MF, Morrison SJ. Bmi-1 dependence distinguishes neural stem cell self-renewal from progenitor proliferation. Nature 2003;425:962e7. [22] Moon JH, Heo JS, Kim JS, Jun EK, Lee JH, Kim A, et al. Reprogramming fibroblasts into induced pluripotent stem cells with Bmi1. Cell Res 2011;21:1305e 15. [23] Moon JH, Yoon BS, Kim B, Park G, Jung HY, Maeng I, et al. Induction of neural stem cell-like cells (NSCLCs) from mouse astrocytes by Bmi1. Biochem Biophys Res Commun 2008;371:267e72. [24] Moon JH, Yun W, Kim J, Hyeon S, Kang PJ, Park G, et al. Reprogramming of mouse fibroblasts into induced pluripotent stem cells with Nanog. Biochem Biophys Res Commun 2013;431:444e9. [25] Park IK, Qian D, Kiel M, Becker MW, Pihalja M, Weissman IL, et al. Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells. Nature 2003;423:302e5. [26] McCarthy KD, de Vellis J. Preparation of separate astroglial and oligodendroglial cell cultures from rat cerebral tissue. J Cell Biol 1980;85:890e902. [27] Moon JH, Heo JS, Kwon S, Kim J, Hwang J, Kang PJ, et al. Two-step generation of induced pluripotent stem cells from mouse fibroblasts using Id3 and Oct4. J Mol Cell Biol 2012;4:59e62. [28] Reynolds BA, Tetzlaff W, Weiss S. A multipotent EGF-responsive striatal embryonic progenitor cell produces neurons and astrocytes. J Neurosci 1992;12: 4565e74. [29] Ruau D, Ensenat-Waser R, Dinger TC, Vallabhapurapu DS, Rolletschek A, Hacker C, et al. Pluripotency associated genes are reactivated by chromatinmodifying agents in neurosphere cells. Stem Cells 2008;26:920e6. [30] Chojnacki A, Weiss S. Isolation of a novel platelet-derived growth factorresponsive precursor from the embryonic ventral forebrain. J Neurosci 2004;24:10888e99. [31] Hsieh J, Nakashima K, Kuwabara T, Mejia E, Gage FH. Histone deacetylase inhibition-mediated neuronal differentiation of multipotent adult neural progenitor cells. Proc Natl Acad Sci U S A 2004;101:16659e64. [32] Hu X, Jin L, Feng L. Erk1/2 but not PI3K pathway is required for neurotrophin 3-induced oligodendrocyte differentiation of post-natal neural stem cells. J Neurochem 2004;90:1339e47. [33] Rogister B, Ben-Hur T, Dubois-Dalcq M. From neural stem cells to myelinating oligodendrocytes. Mol Cell Neurosci 1999;14:287e300. [34] Song MR, Ghosh A. FGF2-induced chromatin remodeling regulates CNTFmediated gene expression and astrocyte differentiation. Nat Neurosci 2004;7:229e35. [35] Moon JH, Kwak SS, Park G, Jung HY, Yoon BS, Park J, et al. Isolation and characterization of multipotent human keloid-derived mesenchymal-like stem cells. Stem Cells Dev 2008;17:713e24. [36] Zulewski H, Abraham EJ, Gerlach MJ, Daniel PB, Moritz W, Muller B, et al. Multipotential nestin-positive stem cells isolated from adult pancreatic islets differentiate ex vivo into pancreatic endocrine, exocrine, and hepatic phenotypes. Diabetes 2001;50:521e33. [37] Greber B, Wu G, Bernemann C, Joo JY, Han DW, Ko K, et al. Conserved and divergent roles of FGF signaling in mouse epiblast stem cells and human embryonic stem cells. Cell Stem Cell 2010;6:215e26. [38] Hanna J, Markoulaki S, Mitalipova M, Cheng AW, Cassady JP, Staerk J, et al. Metastable pluripotent states in NOD-mouse-derived ESCs. Cell Stem Cell 2009;4:513e24. [39] Ying QL, Stavridis M, Griffiths D, Li M, Smith A. Conversion of embryonic stem cells into neuroectodermal precursors in adherent monoculture. Nat Biotechnol 2003;21:183e6.

Please cite this article in press as: Kang PJ, et al., Reprogramming of mouse somatic cells into pluripotent stem-like cells using a combination of small molecules, Biomaterials (2014), http://dx.doi.org/10.1016/j.biomaterials.2014.05.015