Nanog Is the Gateway to the Pluripotent Ground State

Summary Pluripotency is generated naturally during mammalian development through formation of the epiblast, founder tissue of the embryo proper. Pluripotency can be recreated by somatic cell reprogramming. Here we present evidence that the homeodomain protein Nanog mediates acquisition of both embryonic and induced pluripotency. Production of pluripotent hybrids by cell fusion is promoted by and dependent on Nanog. In transcription factor-induced molecular reprogramming, Nanog is initially dispensable but becomes essential for dedifferentiated intermediates to transit to ground state pluripotency. In the embryo, Nanog specifically demarcates the nascent epiblast, coincident with the domain of X chromosome reprogramming. Without Nanog, pluripotency does not develop, and the inner cell mass is trapped in a pre-pluripotent, indeterminate state that is ultimately nonviable. These findings suggest that Nanog choreographs synthesis of the naive epiblast ground state in the embryo and that this function is recapitulated in the culmination of somatic cell reprogramming.

PD0325901 (1μM) obtained from the Division of Signal Transduction Therapy, University of Dundee.

Cell fusion
Mek inhibitor treatment of dsRed2 ES cells was performed using either PD184352 or PD0325901 at 3μM. For fusions, cells of each type (generally 1x10 7 ) were combined in serum-free GMEM in a conical tube, pelleted, and supernatant aspirated. The pellet was broken by gentle tapping and 300 μl of 50% PEG 1500 (ROCHE) pre-warmed to 37°C were gently added. Cells were left in 50% PEG over a 3 min period with occasional stirring. Then 1 ml of medium was added over a period of 3 min. Subsequently, a further 3 ml of medium were added. Cells were spun down and supernatant discarded. The pellet was resuspended in complete ES cell medium and plated. Selection was applied 48 hours later for reactivation of the Oct4 reporter transgene using puromycin (1μg/ml).
For fusions between NS cells constitutively expressing tauGFP and puromycin resistance and ES cells expressing dsRed2 and hygromycin resistance, primary fusion products were purified by flow cytometry 24 hours after PEG treatment using a Dako Cytomation MoFlo MLS sorter. An aliquot of the sorted population was analysed by flow cytometry to determine purity and the number of plated hybrids was calculated from these. Please note that sorting at this point precedes reprogramming (Do and Scholer, 2004;Silva et al., 2006;Tada et al., 2001). This strategy normalises for differences in cell fusion rate between different replicates representing an accurate quantification of somatic cell reprogramming by cell fusion. To eliminate non hybrids puromycin (1μg/ml) and hygromycin (250μg/ml) selection was applied on sorted cells 72hrs after plating.
Selection for hybrids in fusions between Nanog null ES and Nanog null NS cells depended on constitutively expressing drug resistance genes, hygromycin (150μg/ml) and puromycin (1μg/ml) respectively. This selection strategy was inefficient due to cell confluence and to eliminate persisting unfused cells 2i+LIF medium supplemented with puromycin was used.

Retroviral infection and iPS cell induction
Retroviral infection was performed as described (Takahashi et al., 2007;Takahashi and Yamanaka, 2006) with minor modifications. Plat-E cells were seeded at 4 10 6 cells per 100-mm dish. The following day, 9 μg of pMX-based retroviral vectors for Oct4, Sox2, Klf4, or c-MycT58 were introduced separately into Plat-E cultures using 27 μl of FuGENE 6 transfection reagent. After 24 h, the medium was replaced with 10 ml of DMEM containing 10% FCS. Target cells (MEFs and NS cells) were seeded at 1.2 x 10 5 cells per 35-mm dish coated with gelatin. The following day, virus-containing supernatants from Plat-E cultures were filtered through a 0.45-μm cellulose acetate filter. Equal volumes of the supernatants were mixed and supplemented with polybrene at the final concentration of 4μg ml -1 . Cells were incubated in the virus/polybrenecontaining supernatants for 24 h, then restored to NS cell culture medium. Three days after transduction, cultures were changed into ES cell medium. For further expansion pre-iPS cells were replated onto feeders at day 5 in medium containing serum and LIF. pMXs-gw plasmids; pMXs-Oct4, pMXs-Klf4 and pMXs-cMycT58 were obtained from Addgene repository.

Embryo collection
The Nanog null mutation generated by homologous recombination has been described previously (Mitsui et al., 2003). Genotypes of intercross blastocysts were inferred from presence or absence of Nanog immunostaining. For cultured ICMs genotypes were determined by PCR analysis of trophectoderm lysates (Nichols et al., 1998) using primers aatgggctgaccgcttcctcgtgctt, agtacctcagcctccagcagatgc and cagaatgcagacaggtctacagcccc. Diapause was induced by injecting pregnant female mice intraperitoneally with Tamoxifen (Sigma, T5648-1G; 10μg per mouse) and subcutaneously with Depo Provera (Pharmacia, MEDEP01); 3mg per mouse) at 2.5 days post coitum (dpc). Blastocysts were flushed from uteri four days later. Embryos were harvested at embryonic day (E) 3.5 or 3.75 and ICMs isolated by immunosurgery with collection of trophectoderm lysate for genotyping. ICMs were plated individually into gelatinised 4 well plates containing GMEM with 20% foetal calf serum and incubated for 8 days.

Immunofluorescent staining
Blastocysts were fixed for 15 minutes in 4% PFA, rinsed in PBS containing 3mg/ml PVP, permeabilised in PBS/PVP with 0.25% triton X 100 for 30 minutes, and blocked in PBS containing 0.1% BSA, 0.01% Tween 20 and 2% donkey serum for 15 minutes, all at room temperature. Primary antibody solutions were made up in blocking buffer and embryos were incubated overnight at 4 o C. They were rinsed three times for 15 minutes each in blocking buffer then incubated for an hour at room temperature in the appropriate secondary antibodies at 1 in 500 in blocking buffer. After three rinses for 15 minutes each they were incubated briefly in increasing concentrations of Vectashield Antibodies for NS cell staining were Nestin (1:10) from DSHB and Tuj1 (1:400) from Covance, and for pre-iPS cells were SSEA-1 (1:10) and Ecad (1:100) from DSHB.
Alexa fluorescent secondary antibodies from Molecular Probes were used in all cases.
Slides were analyzed by confocal microscopy (Leica TCS SP5) and processed with Leica software and Adobe Photoshop. Phase images of cells and blastocysts were collected using a Leica CTR microscope.

RT-PCR and qRT-PCR
For RT-PCR total RNA was extracted using the RNeasy kit (Qiagen), and cDNA generated using Superscript II (Invitrogen). For qRT-PCR total RNA was extracted using the RNeasy kit (Qiagen), and cDNA generated using Superscript III (Invitrogen).  Table for details of primers and TaqMan gene expression assays used in this study.
(C) Flow cytometry dot plots of Oct4-GFP expression during pre-iPS to iPS cell conversion in 2i/LIF. These data were used to derive histogram in Figure 2J.
(D) Chimera generated from MEF pre-iPS cell clone derived iPS cells with C57BL/6 mate and pups. The recipient blastocyst strain was C57BL/6 so agouti offspring demonstrates transmission of the iPS cell haploid genome. Without reprogramming, EpiSCs are completely unable to colonise the developing embryo after either morula aggregation (Guo et al., 2009) or blastocyst injection (Tesar et al., 2007). Cells from two Epi-iPS cell clones were combined with E2.5 morulae.
Aggregates were maintained in culture for 48 hours then transferred to pseudopregnant recipients. Embryos were harvested at E6.5 and examined by fluorescence microscopy for Oct4-GFP reporter expression. Images show two representative embryos with GFP expression throughout the egg cylinder demonstrating incorporation of Epi-iPS cells. Figure S9. Nanog negative blastocysts express Oct4 throughout the ICM at E3. 5 (A) Nanog and Oct4 immunostaining of E3.5 blastocysts from intercrosses of Nanog +/mice.

Figure S10
Loss of Nanog expression does not acutely trigger X inactivation.
Immunostaining for Nanog, Oct4 and Eed of XX ES cells cultured in 2i/LIF+N2B27 (0hrs) or in N2B27 alone for 48 and 96hrs.