(A) An Oct4∷GFP+ iPS line that was derived from a culture of RepSox-treated Oct4, Klf4, and cMyc-infected MEFs (OKM + RepSox line 1) displays the characteristic mES-like morphology and self-renewal properties. Passage 11. Scale bars = 500 μm.
(B) Antibody staining of OKM + RepSox line 1 cells shows that they express markers of pluripotent stem cells Sox2 and Nanog. Scale bars = 100 μm.
(C) Microarray scatter plots showing that the global gene expression profile of OKM + RepSox line 1 is highly similar to that of mES line V6.5 and very different from that of somatic MEFs.
(D) Motor neurons differentiated in vitro from OKM + RepSox line 1. Scale bar = 200 μm.
(E) Teratomas containing cells of all three germ layers formed by injection of OKM + RepSox line 1 cells into nude mice.
(F) E12.5 chimeric mouse embryo (left, vs. non-chimeric littermate on the right) showing a high amount of contribution from OKM + RepSox line 1 cells constitutively expressing the dTomato red fluorescent protein.
(G) 8 week-old chimeric mouse formed by injection of OK + RepSox line 1 cells (C57BL6 genetic background) into an ICR blastocyst.
(H) Oct4∷GFP+ cells derived from an OKM + RepSox cell line are present in the genital ridge of a male embryo at 13.5 d.p.c.

(A) Graph showing the number of Oct4∷GFP+ colonies induced by various timings of RepSox treatment of Oct4, cMyc, and Sox2-infected MEFs in mES medium. Colonies were counted at 24 days post-infection.
(B) Timecourse of RepSox treatment showing the number of Oct4∷GFP+ colonies induced by a 24-hr pulse of RepSox on Oct4, cMyc, and Sox2-infected MEFs in serum-free mES medium with knockout serum replacement (KSR mES). Colonies were counted at 24 days post-infection. Shown are average colony numbers +/− the standard deviation.

(A)RepSox treatment of RepSox-responsive line OKMS 6 strongly increases Nanog mRNA levels. Data were generated by microarray analysis and are relative to untreated controls. Nanog is induced faster and more significantly than Sox2, indicating it is upregulated before fully reprogrammed cells form.
(B) RT-PCR analysis showing that inhibition of Tgf-β signaling increases Nanog expression in the RepSox-responsive intermediate line OKMS 7.
(C) A pulse of RepSox induces a persistent increase in Nanog expression in the RepSox-responsive intermediate line OKM 10. OKM 10 cells were treated with 25 μM RepSox for 48 hours and RNA samples were taken at 0, 48, and 96 hours (48 hours after removal of RepSox) and analyzed by RT-PCR.
(D) shRNA-mediated knockdown of Nanog in OKM 10 cells inhibits replacement of Sox2 by RepSox.
(E) Pictures of reprogrammed Oct4∷GFP+ colonies induced by Sox2 (A) or Nanog (B) transduction of line OKM 10. Scale bars = 200 μm.
(F) Nanog transduction can reprogram line OKM 10 at a similar efficiency as Sox2 transduction.
(G) Nanog can substitute for Sox2 in defined-factor reprogramming of somatic fibroblasts.
(H) Picture of a reprogrammed Oct4∷GFP+ colony induced by Oct4, Klf4, cMyc and Nanog-transduction of MEFs. Scale bars = 100 μm.

(A) Oct4∷GFP+ colonies form readily in Oct4, Klf4, cMyc, and Sox2-infected MEF cultures and do not form in Oct4, Klf4, and cMyc-infected MEF cultures. Scale bars = 500 μm.
(B) Overview of chemical screen for replacement of Sox2.
(C) A P0 colony from Oct4, Klf4, and cMyc-infected MEFs + RepSox that displays a mES-like morphology and is Oct4∷GFP+. Scale bars = 200 μm.
(D) Number of Oct4∷GFP+ colonies detected for each hit in the primary screen after transduction of Oct4, Klf4, and cMyc and VPA treatment.
(E) Chemical structures of E-616452, E-616451, and EI-275, with the optimal concentrations for reprogramming listed.
(F) Quantification of small molecule replacement of Sox2 in Oct4, Klf4, and cMyc-infected MEFs with and without VPA treatment.
(G) Sox2 replacement by RepSox is not dependent on cMyc (no VPA treatment).

(A) Chemical structure of SB431542, an inhibitor of Tgfbr1 activity.
(B) Inhibition of Tgf-β signaling by treatment of Oct4, cMyc, and Sox2-infected MEFs with SB431542 or TGF-β neutralizing antibodies replaces Sox2.
(C) RepSox does not increase the efficiency of Oct4∷GFP+ colony induction in Oct4, Klf4, cMyc, and Sox2-infected MEFs.
(D) Inhibition of Tgf-β signaling by TGF-β neutralizing antibodies does not increase the efficiency of Oct4∷GFP+ colony induction in Oct4, Klf4, cMyc, and Sox2-infected MEFs.
(E) RepSox does not replace transgenic Oct4 or transgenic Klf4 in reprogramming. We observed no Oct4∷GFP+ colonies in RepSox-treated Klf4, cMyc, Sox2-infected MEFs or Oct4, cMyc, Sox2-infected MEFs out of 30,000 cells plated both with and without VPA treatment. We routinely observe 30-40 Oct4∷GFP+ colonies when we plate the same number of Oct4, Klf4, cMyc-infected MEFs and treat with RepSox.
(F) RepSox can replace cMyc in reprogramming. Cells were transduced with Oct4, Klf4, and cMyc and treated with RepSox continuously starting at day 5 post-infection.
(G) Inhibition of Tgf-β signaling can replace cMyc in reprogramming. Cells were transduced with Oct4, Klf4, and cMyc and treated with inhibitors of Tgf-β signaling continuously starting at day 5 post-infection.

(A) Stable Oct4∷GFP-negative cell lines derived from Oct4∷GFP-negative colonies in Oct4, Klf4, and cMyc-infected MEF cultures can be reprogrammed by RepSox. Scale bars in “OKM line 10 + RepSox” panels = 500 μm, all other scale bars = 200 μm.
(B) 2 of 10 stable, non-pluripotent intermediate cell lines derived from MEFs transduced with Oct4, Klf4, and cMyc can be reprogrammed with RepSox treatment but none can be reprogrammed with AZA treatment.
(C) Western blot for phospho-Smad3 showing that RepSox inhibits Tgf-β signaling in line OKM 10 (OKM 10) cells.
(D) RepSox does not increase the proliferation of OKM 10 cells.
(E) Line OKM 10 can be reprogrammed with RepSox treatment but not with AZA or 2i, indicating it is distinct from cell lines that can be reprogrammed by AZA or 2i.
(F) Stable Oct4∷GFP-negative cell lines derived from Oct4∷GFP-negative colonies in Oct4, Klf4, cMyc and Sox2-infected MEF cultures can be reprogrammed by RepSox or by AZA, but lines responsive to RepSox are not responsive to AZA alone and lines responsive to AZA are not responsive to RepSox alone, indicating the presence of two different types of stable intermediates in the reprogramming cultures.