Oct4^+/− Cells Express an Elevated Level of NANOG and Lack a Nanog-Negative Population
(A) Immunoblot analysis of Oct4^+/+ (E14Tg2a, CGR8, and ZIN40) and of Oct4^+/− (OKO160 [] and ZHTc6 [cultured in 1 μg/ml doxycycline []) cell lines. See for a detailed summary of all cell lines used in this study. ZHBTc4 (Oct4^−/−) cells (), which express an Oct4 transgene at heterozygous levels (when cultured without doxycycline) also showed increased Nanog. The parietal endoderm cell line D7A3/PE () provides a negative control.
(B) Quantitative PCR (qPCR) analysis of Oct4 and Nanog mRNAs in Oct4^+/+ (E14Tg2a and CGR8) and Oct4^+/− (OKO160 and ZHTc6 [cultured in 1 μg/ml doxycycline]) cells. Error bars represent SD; n = 3.
(C) Immunofluorescence analysis of wild-type (WT; E14Tg2a) and Oct4 mutant (OKO160, ZHTc6 [cultured in 1 μg/ml doxycycline], and ZHBTc4 [cultured without doxycycline]) cells for Nanog (green) and Oct4 (red).
(D) Quantitative analysis of Oct4 protein levels in individual cells in colonies of Oct4^+/+ (E14Tg2a, green) and Oct4^+/− (OKO160, brown) cells.
(E) Intracellular FACS quantitation of Oct4 and Nanog protein levels in individual cells of Oct4^+/+ (E14Tg2a, red) and Oct4^+/− (OKO160, blue) cultures.
(F) Top, immunofluorescence analysis of blastocyst expression of Oct4 protein following the aggregation of GFP-marked Oct4^+/− ESCs with WT morula. Bottom, quantitation of immunofluorescence for Oct4 in the ICM for WT and GFP-marked Oct4^+/− cells.
(G) FACS analysis for Nanog:GFP in Oct4 WT (Tg2a-Nanog:GFP) and Oct4 mutant cells (OKO-Nanog:GFP and ZHTc-Nanog:GFP [cultured in 1 μg/ml doxycycline] derived from the parental lines OKO160 and ZHTc6, respectively; Oct4 genotypes are indicated) at the indicated times following release from puromycin selection (day 0). The percentage of cells in Nanog-low, Nanog-middle, and Nanog-high populations are shown (red) with the coefficient of variation (CV) for Nanog:GFP indicated.
See also and and .

Titrated Elevation of Oct4 Levels Restores the Heterogeneous Expression of Nanog, Esrrb, and Klf4
(A) Immunofluorescence (Oct4, red) and fluorescence (Nanog:GFP, green) in ZHTc-Nanog:GFP cells treated with the indicated doxycycline concentrations for 6 days.
(B) Intracellular FACS analysis for Oct4 and Nanog:GFP at the indicated times of treatment with the indicated doxycycline dose.
(C) Immunofluorescence analysis at day 4 of doxycycline treatment of ZHTc-Nanog:GFP cells for Esrrb, Klf4, and Nanog:GFP.
(D) GFP-low ZHTc-Nanog:GFP cells remain undifferentiated and revert to a GFP-expressing state. Left, an experimental scheme; Oct4 expression was increased in Oct4^+/− (ZHTc-Nanog:GFP) cells in order to induce Nanog heterogeneity, and cells were sorted into Nanog:GFP-high and Nanog:GFP-low populations. Right, reanalysis of the sorted populations showed that they were >99% pure (day 0). Cells were replated in GMEMβ-FCS-LIF-doxycycline (1,000 ng/ml) in order to restore Oct4 protein to a Oct4^+/− state and the emergence of the GFP-high population (indicated as a percentage) monitored daily.
See also and .

ESCs Expressing Reduced Oct4 Levels Show Delayed Differentiation
(A) Morphology of Oct4^+/+ (E14Tg2a and Oct4GiP [which carries an Oct4 promoter-driven GFP-ires-pac-pA cassette as an additive transgene]) () and Oct4^+/− (OKO160 and ZHTc6) cell lines at the indicated times after LIF withdrawal.
(B) qPCR analysis of mRNA expression in the indicated cell lines on the days shown after LIF withdrawal. Error bars represent SD; n = 3.
(C) qPCR analysis of mRNA expression in the indicated cell lines on the days shown after the initiation of monolayer neural differentiation. Error bars represent SD; n = 3.
(D) Immunofluorescence for Nestin, Oct4, or βIII Tubulin at day 8 of neural differentiation. Cells were counterstained with DAPI.
(E) FACS analysis of Nanog:GFP and Pecam expression after the replating of Oct4^+/+ (Tg2a-Nanog:GFP) and Oct4^+/− (OKO-Nanog:GFP) ESCs in activin-FGF for the allowance of differentiation to an EpiSC state.
(F) qPCR analysis of mRNA expression in the indicated cell lines on the days shown after the replating of ESCs in activin-FGF2. Error bars represent SD; n = 3.
(G) Functional assessment of differentiation status. Following the outlined scheme (top), the differentiation status of ES-like colonies was assessed after 6 days (Oct4^+/+, E14Tg2a; Oct4^+/−, OKO160). Error bars represent SD; n = 4.
(H) JAK inhibitor blocks ESC self-renewal in the presence of LIF. ESCs were treated with JAK inhibitor for 6 days, replated at clonal density without JAK inhibitor in GMEMβ-FCS-LIF, cultured for an additional 6 days, and alkaline phosphatase-positive counted. Error bars represent SD; n = 3.
(I) EpiSC lines of the indicated genotypes (Oct4^+/+, E14Tg2a; Oct4^+/−, OKO160) were passaged twice in the presence of the same JAK inhibitor concentration used in (H), replated in GMEMβ-FCS-LIF either in the presence or the absence of exogenous Esrrb, and assessed for colony-forming capacity after 6 days. Error bars represent SD; n = 4.
See also and and and .

Pluripotency of Oct4^+/− Cells Is Supported in the Absence of BMP
(A) Alkaline phosphatase staining of colonies formed by plating the indicated cell lines at clonal density and culturing for 7 days in GMEMβ with 10% FCS and 100 U/ml LIF.
(B) Quantitation of colony types formed by cell lines shown in (A). Error bars represent SD; n = 3.
(C) Quantitation of colony types formed by plating Oct4^+/+ (E14Tg2a) and Oct4^+/− (OKO160) ESCs at clonal density in GMEMβ with 10% FCS and the indicated LIF concentrations (U/ml) for 7 days. Error bars represent SD; n = 3.
(D) FACS analysis of Tg2a-Nanog:GFP and OKO-Nanog:GFP cells treated with the indicated LIF concentrations (U/ml) for 3 days. Before starting (day 0), cell lines were selected for Nanog:GFP expression with puromycin.
(E) Oct4^+/+ (E14Tg2a) or Oct4^+/− ESCs (OKO160-GFP) were plated at clonal density, cultured in N2B27 (with the indicated cytokines) for 7 days, and stained for alkaline phosphatase.
(F) Oct4^+/+ (AGFP7) or Oct4^+/− (OKO160-GFP) ESCs were cultured in N2B27-LIF (100 U/ml) for 7 days at clonal density, replated for a further 7 days in the same conditions, and stained for alkaline phosphatase.
(G) Oct4^+/− ESCs (OKO160-GFP, which express a constitutive GFP transgene) cultured as described in (F) were injected into a host blastocyst and midgestation embryos examined for the presence of ubiquitous GFP-expressing Oct4^+/− cells by fluorescence microscopy.
(H) Oct4^+/− ESCs (OKO-AGFP, which express a constitutive GFP transgene) were treated as described in (F) and (G). E11.5 genital ridges were analyzed by immunofluorescence for Oct4 protein and GFP expression.
(I) Adult chimera from a blastocyst obtained as described in (G).

Signaling Influences in Oct4^+/− ESCs
(A) The Nanog promoter sequence around the octamer site (red). Binding sites for Essrb (blue) and Sox2 (green) are also shown as is the sequence of the Oct4 site mutation (red).
(B and C) Analysis of Tg2a-Nanog:GFP and a representative cell line containing a mutated Oct4 binding site (mTg2a-Nanog:GFP) after release from puromycin treatment by FACS at days 6 and 18 (B) and fluorescent microscopy at day 18 (C).
(D) qPCR analysis of Oct4 and GFP transcripts in Tg2a-Nanog:GFP and mTg2a-Nanog:GFP cells at day 18 after puromycin removal. Error bars represent SD; n = 3.
(E) FACS analysis of ZHTc-Nanog:GFP treated with or without doxycycline and in parallel with FGFR inhibitor (PD173074), MEK inhibitor (PD0325901), or GSK3 inhibitor (CHIR99021) for 6 days.
(F) FACS analysis of the indicated Nanog:GFP cell lines treated with or without FGF2 (10 ng/ml) for 6 days.
(G) Immunoblot analysis of phospho-ERK and total ERK in the indicated cell lysates prepared after overnight removal of FCS-LIF and stimulation with FGF2 (10 ng/ml) or LIF (100 U/ml) for the indicated time (min); PD indicates ESCs treated with PD0325901 for 4 hr.
(H) Immunofluorescence analysis for phospho-STAT3 and Nanog in the indicated cell lines.
(I) Immunofluorescence analysis for phospho-STAT3 and Nanog in ZHTc-Nanog:GFP cells after treatment with the indicated doxycycline concentrations.
(J) Immunoblot analysis of phospho-STAT3 in the indicated cell lysates prepared after overnight removal of LIF and stimulation with LIF (100 U/ml) for the indicated time.
See also and and .

Wnt and Transcription Factor Distinctions between Oct4^+/+ and Oct4^+/− ESCs
(A) qPCR analysis of E14Tg2a and OKO160 ESCs for the indicated transcripts (left) and ZHTc-Nanog:GFP ESCs treated with the indicated doxycycline dose for 4 days (right). Error bars represent SD; n = 3.
(B) FACS analysis of Nanog:GFP expression in Tg2a-Nanog:GFP cells treated with the indicated Wnt3a concentrations for 3 days.
(C) FACS analysis of Nanog:GFP expression in ZHTc-Nanog:GFP cells treated without doxycycline in the presence or absence of Wnt3a (80 ng/ml).
(D) FACS analysis of Nanog:GFP expression in FGF-Nanog:GFP cells treated with FGF2 in the presence or absence of Wnt3a (80 ng/ml).
(E) A comparison of the mean-quantile-normalized gene-expression levels in Oct4^+/+ (E14Tg2a and OctGIP) and Oct4^+/− (OKO160 and ZHTc6) ESCs. Genes differentially and consistently overexpressing (“induced” by Oct4) or underexpressing (“repressed” by Oct4) in Oct4^+/+ versus Oct4^+/− (|log2FC| ≥ log2(1.25), FDR ≤ 0.1; see ) are highlighted in blue and orange, respectively. Boundaries for a 1.25× fold change are shown in red.
(F) Average quantile-normalized expression intensities for Wnt signaling and pluripotency-related (G) genes. Error bars represent SEM.
See also and , , and .

Genome-Wide Analysis for Oct4 and Nanog Occupancy in Oct4^+/− ESCs
(A–D) Chromatin immunoprecipitation sequencing profiles of Oct4 and Nanog for Esrrb (A), Tbx3 (B), Oct4 (C), and Slc7a10 (D) gene loci in Oct4^+/+ ESCs and Oct4^+/− ESCs . Read counts are normalized by the number of uniquely mapped reads.
(E) Genes with increased Oct4 binding within 50 kb of the transcription start site (TSS) in Oct4^+/− ESCs are enriched in the gene set showing increased expression in Oct4^+/− ESCs; significance was estimated by Fisher’s exact test.
(F) Genes with reduced levels of H3K27me3 in Oct4^+/− ESCs are enriched in the gene set showing increased expression in Oct4^+/− ESCs; significance was estimated by Fisher’s exact test.
(G–I) Average read count for p300 (G), H3K4me1 (H), and H3K4me3 (I) at Oct4 binding sites shared between Oct4^+/+ and Oct4^+/− ESCs (black) or showing increased Oct4 occupancy in Oct4^+/− ESCs (red).
(J) A model of the pluripotency GRN. ESCs express a range of Oct4 concentrations (red circles), only the lower of which are present in Oct4^+/− ESCs (green box). In this lower state, Oct4 binds strongly to key regulatory genes, encoding heterogeneously expressed transcription factors and Wnt signaling components. The increased Wnt signaling results in enhanced phospho-STAT3 levels in response to LIF, further reinforcing the expression of heterogeneously expressed pluripotency transcription factors. This produces upward pressure on the Oct4 gene, driving cells into the high Oct4 concentration range (yellow arrow 1). At these high Oct4 concentrations (white box), the Wnt system is not activated and Oct4 binding to transcription factor loci is weakened. Exit of cells from this state can result either from the residual GRN activity, causing downward fluctuation in the Oct4 expression (yellow arrow 2) or from the cell experiencing an effective commitment signal (yellow arrow 3).
See also and , , and .