Establishment of SSCs from Oct4-GFP/LacZ transgenic mice (C57BL/6 background) and Oct4-GFP transgenic mice (C57BL/6 background) was described previously (9, 11, 14) SSCs, gPSCs, feeder-free system for derivation of gPS cells (FF-gPSCs), and ESCs were generated and characterized in previous studies and were cultured according to protocols presented previously (9, 12, 15).

For screening of small molecules inducing pluripotency, SSCs were seeded in Matrigel (Corning, Manassas, VA, USA)-coated 96-well plates. After 1 day, the Screen-Well Epigenetics library (Enzo Life Sciences, Doral, FL, USA) and 7 molecules (16) were each added to the culture medium (Supplementary Table S1). The library and the 7 molecules were dissolved in DMSO (Sigma-Aldrich, Saint Louis, MO, USA).

For pluripotency induction SSCs were seed at a density of 1×10⁵ cells per well in 24-well plates coated with Matrigel and cultured in expansion medium (9). The next day, medium was replaced with a medium containing 5 μM CI-994, 0.3 μM mocetinostat, or 1 μM entinostat and was changed every day for 4 days. Then the medium containing the three small molecules was replaced with expansion medium, which was changed every 2 days until Oct4-GFP colonies were formed.

SSCs were plated at 2.5×10⁵ cells per well in 24-well plates or 5×10⁵ cells per well in 12-well plates and cultured to observe gPSC conversion. After 4 weeks, GFP-positive colonies were counted to calculate self-reprogramming efficiency.

Total RNA was isolated by using the RNeasy Mini Kit (Qiagen, Hilden, Germany). Total RNA (500 ng) was reverse-transcribed by using the OmniscriptRT Kit (Qiagen) in a total volume of 20 μl. PCR analysis was performed with gene-specific primers and Takara Ex Taq DNA polymerase (TaKaRa Bio Inc., Kusatsu, Shiga, Japan) according to the manufacturer’s instructions. The PCR conditions were as follows: 32 cycles at 94℃ for 30 s, 50∼65℃ for 30 s, and 72℃ for 30 s. The RT-PCR products were analyzed by electrophoresis in 1% agarose gels. Primer sequences are listed in Supplementary Table S2.

The microarray study was carried out using Illumina Mouse Ref-8v2.0 Expression BeadChips (Illumina Inc., San Diego, CA, USA) as previously described (9, 11, 14). Quality-checked cDNA samples were hybridized for 18 h with the BeadChips, washed, stained, and scanned following the guidelines and using materials and instrumentation supplied or suggested by the manufacturer.

SSCs were blocked with 0.5% BSA (Sigma-Aldrich) in PBS and stained with Alexa Fluor^Ⓡ 647 Mouse Anti-SSEA-1 (BD Biosciences, Franklin Lakes, NJ, USA) for 30 min at room temperature (RT). Samples were analyzed using FACSAria IIu (BD Biosciences) and FlowJo v10 software (BD Biosciences).

Cells were fixed in 4% paraformaldehyde (Sigma-Aldrich) for 15 min at RT, washed 3 times with DPBS, and then incubated in DPBS containing 0.5% Triton X-100 (Sigma-Aldrich), 1% BSA fraction V (Sigma-Aldrich), and 10% fetal bovine serum (FBS; Gibco, Carlsbad, CA, USA) for 10 min at RT. The cells were briefly rinsed with DPBS and incubated with primary antibodies overnight at 4℃ or for 1 h at RT. The cells were then washed with 0.5% BSA (Sigma-Aldrich) in PBS and incubated with secondary antibodies (anti-mouse IgG or anti-mouse IgM; 1：200; R&D Systems Inc., Minnesota, USA) for 1 h at RT. Nuclei were stained with 4’-6-diamidino-2-phenylindole (Dapi; Sigma-Aldrich). Information on antibodies is provided in Supplementary Table S3.

Cells were washed with PBS, and genomic DNA was isolated using the Total DNA Extraction kit (Intron, Liberty Lake, WA, USA) according to the manufacturer’s protocol. Genomic DNA was treated with Epi-Tech Bisulfite (Qiagen) according to the manufacturer’s recommendations and used for PCR amplification. The PCR products were cloned using the PCR Cloning kit (Qiagen). Primer sequences are listed in Supplementary Table S4.

To differentiate CI-gPSCs into three germ layers, previously described protocols (17, 18) were applied to embryoid bodies derived from gPSCs. Embryoid bodies were attached to a gelatin-coated plate and cultured in MEF medium until beating cells were formed. MEF medium was Dulbecco’s modified Eagle’s medium (DMEM) low-glucose (Welgene, Gyeongsan, Korea) with the following supplements: 10% FBS (Gibco), 50 μM β-mercaptoethanol (Gibco), 1×penicillin–streptomycin (Welgene), and 1×minimal essential medium non-essential amino acids (Gibco).

CI-gPSCs were aggregated with denuded post-compacted 8-cell-stage embryos retrieved from mice (C57BL/6 albino) at 2.5 days post coitum (dpc). Clumps of loosely connected gPSCs (10∼20 cells) after short trypsin treatment were selected and transferred into microdrops of KSOM mouse embryo medium containing 10% FBS (Gibco) under mineral oil. Each clump was placed in a depression in the microdrop. Batches of 30 to 40 embryos were briefly incubated with acidified Tyrode’s solution until the zona pellucida was degraded. A single embryo was placed into each clump. All aggregates were cultured at 37℃ in a humidified atmosphere containing of 5% CO₂. After 24 h, the majority of the aggregates had formed blastocysts. Approximately 11 to 14 aggregated embryos were transferred into the uterine horn of 2.5-dpc pseudopregnant mouse.

To induce pluripotency in SSCs by epigenetic regulation, SSCs were treated with of Screen-Well Epigenetics library or 7 molecules at 1∼10 μM (16) and small molecules involved in reprogramming were selected according to the intensity of Oct4-GFP. The Oct4-GFP-positive gPSC population increased in CI-994, a HDAC class I inhibitor, -treatedSSCs (Supplementary Fig. S1). To determine the optimal concentration, SSCs were treated with CI-994 at 0.1∼100 μM to find the concentration at which cell death did not occur even with continuous exposure (Supplementary Fig. S2). Treatment for 4 days at a highest concentration of 5 μM did not induce cell death, and GFP-positive colonies appeared 8 days after removal the small molecules (Fig. 1A and 1B). Oct4-GFP-positive CI-gPSCs had grown similar morphology as mESCs in feeder-free culture (Fig. 1C). Positivity for alkaline phosphatase, SSEA1, and Nanog indicated that CI-gPSCs expressed pluripotency markers (Fig. 1C). To verify the reprogramming efficiency of CI-994 under the optimized conditions, we compared Oct4-GFP-positive colony formation with or without CI-994 treatment (Fig. 1D). We observed that the number of Oct4-GFP-positive colonies in CI-994 was higher than control group. These results suggest that CI-994, induces and promotes self-reprogramming of SSCs.

RT-PCR results indicated that CI-gPSCs expressed the pluripotency-associated genes Oct4, Nanog, Esg1, Fgf4, Cripto, Rex1, Eras, and Utf1 (Fig. 2A). Scatter plots of microarray analysis demonstrated the difference in gene expression patterns between CI-gPSCs and SSCs and similarity between CI-gPSCs and ESCs (Fig. 2B). Hierar-chical clustering and heatmap analysis revealed that global gene expression of CI-gPSCs was more similar to that of gPSCs and ESCs than to that of FF-gPSCs, SSCs, and MEFs (Fig. 2C). Bisulfite sequencing analysis showed that Oct4 and Nanog were hypomethylated in ESCs, gPSCs, FF-PSCs, and CI-PSCs, but were hypermethylated in SSCs (Fig. 2D). Taken together, these results demonstrate that the cellular and molecular characteristics of CI-gPSCs were very similar to those of ESCs and other gPSCs.

We carried out in vitro and in vivo differentiation to confirm the pluripotency of CI-gPSCs. We investigated differentiation of CI-gPSCs in embryoid bodies into three germ layers and observed differentiated cells positive for AFP, an endoderm lineage marker α-SMA, a mesenchymal cell marker; and Tuj1, a neuronal cell marker (Fig. 3A∼C). The presence of the three germ layers proved that CI-gPSCs could differentiate in vitro. To confirm the in vivo differentiation potential of CI-gPSCs, we transplanted them into immunodeficient mice to generate teratomas. Within 4 weeks of transplantation, teratomas formed. In the H&E analysis, staining of a teratoma section revealed that tissues contained all three germ layers, confirming that CI-gPSCs differentiated into the endoderm (gland), mesoderm (muscle), and ectoderm (neural rosette) (Fig. 3D∼F). The presence of teratomas containing three germ layers showed that CI-gPSCs could differentiate in vivo. To further assess the pluripotency of CI-gPSCs, we investigated chimera formation. The chimera analysis showed the contribution of CI-gPSCs to all three germ layers—endoderm (gut), mesoderm (skeletal muscle), ectoderm (skin)—and to germ cells, as confirmed by the presence of Oct4-GFP-positive cells in the fetal gonads of chimeric embryos (Fig. 3G∼I). These results clearly show that CI-gPSCs not only differentiate into the three germ layers, but also have pluripotency to contribute to the germline.

The DNA methylation patterns of differentially methylated regions (DMRs) are known not to change even after SSCs are converted to gPSCs (9). Therefore, we sought to confirm that methylation of DMRs of the paternally imprinted gene H19 and maternally imprinted gene Snrpn were maintained in the same CI-gPSCs as gPSCs and feeder free derived gPSCs generated previously (19). Bisulfite sequencing analysis revealed that H19 was hypermethylated, whereas Snrpn was hypomethylated (Supplementary Fig. S3) in CI-gPSCs.

CI-994 belong to class I HDAC inhibitors (Supplementary Table S5). To test whether various HDAC inhibitors would affect reprogramming, were carried out SSC reprogramming experiments using HDAC inhibitors of various classes. Oct4-GFP colonies were appear after 4 days of treatment with 0.3 μM mocetinostat or 1 μM entinostat, which are also class I HDAC inhibitors, followed by a week of incubation without HDAC inhibitors (Fig. 4A and 4B), whereas Oct4-GFP was not observed after treatment with inhibitors of other HDAC classes such as valproic acid (data not shown). After treatment of SSCs with CI-994, mocetinostat, and entinostat the number of colonies was approximately 4.5, 11.6, and 2.2 times that of untreated SSCs, respectively (Fig. 4C). These results indicate that class I HDAC could act as a barrier in SSC reprogramming and their inhibition by small molecules such as mocetinostat improves the efficiency of reprogramming.