exportin Having confirmed that the Nanog

Having confirmed that the Nanog−/− iPSCs expressed key markers of pluripotency, we sought to determine the extent to which the global expression profile of Nanog−/− iPSCs recapitulated that of ESCs. To this end, we performed RNA sequencing (RNA-seq) of two replicates each of control Nanog+/+ ESC, iPSC, MEFs, and partially reprogrammed iPSC as well as Nanog−/− ESC and two iPSC clones, G2 and G5. We observed RNA-seq reads aligning to Nanog exon1, but not exons 2–4, in both the Nanog−/− ESC and iPSC clones (Figure S3). This confirms the absence of Nanog expression and indicates that the endogenous Nanog promoter is activated in these cells. As expected, we observed many RNA-seq reads mapping to all exons in control Nanog+/+ ESCs and iPSCs, but not in control MEFs or partially reprogrammed iPSCs (Figure S3). Unsupervised hierarchical clustering of the samples based on the expression of all genes revealed that all pluripotent exportin clustered together and apart from both MEFs and partially reprogrammed iPSCs. As expected, both Nanog−/− iPSC lines showed a high degree of similarity to Nanog−/− ESCs (Figure 3A). Pairwise comparisons further revealed that relative to MEFs, Nanog−/− iPSCs were as similar to Nanog−/− ESCs as control, Nanog+/+ iPSCs were to control Nanog+/+ ESCs (Figure 3C). Analysis of a wide range of reported pluripotency markers revealed that Nanog−/− iPSCs expressed all markers with a high degree of similarity to both Nanog−/− ESCs and control ESCs and iPSCs (Figure 3B). Moreover, Nanog−/− iPSCs expressed low levels of ectoderm, mesoderm, and fibroblast markers similar to Nanog−/− ESCs. Interestingly, as previously reported in Nanog−/− ESCs (Chambers et al., 2007, Niakan et al., 2010), each of the Nanog−/− iPSC lines expressed increased levels of early endoderm markers including Sox17, Gata4, and Gata6 when compared to Nanog+/+ ESCs or Nanog+/+ iPSCs. Finally, to definitively test whether these putative Nanog−/− iPSCs were indeed pluripotent, we asked whether they could colonize chimeric embryos and contribute differentiated progeny to the three embryonic germ layers. We injected cells from putative Nanog−/− iPSC lines G5, 3.1, and 3.2 into blastocysts and found that they contributed to E12.5 embryos by green fluorescence and to resulting chimeric adults by green fluorescence and coat color (Figures 4A–4C). In the case of the Nanog−/− iPSC lines reprogrammed with KSO, 12 out of 16 and 3 out of 8 embryos recovered were chimeric, and for the Nanog−/− iPSC line made with KSOM (G5), 3 out of 14 embryos were chimeric. Coat-color analysis of adult mice revealed that for the KSO Nanog−/− iPSC lines, 8 out of 15 and 8 out of 14 animals were chimeric, and for the KSOM Nanog−/−, 3 out of 14 animals were chimeric. Importantly, Nanog−/− cells contributed substantially to tissues from the three germ layers in adult chimeras, including the brain, heart, lung, and liver (Figures 4A and S4A). To evaluate if the Nanog−/− iPSCs could contribute to the germline and generate mature germ cells, we crossed chimeric Nanog−/− GFP+ iPSC males with C57BL/6 females. Genotyping for the GFP transgene in the resulting adult progeny revealed 7 out of 22 positive animals (Figure 4E). The genotyping strategy was further confirmed by detection of GFP expression in the tissues of transgene-positive animals, for example, F1 #4, but not their transgene-negative littermates (F1 #3, Figure 4F). Partially reprogrammed cells (piPS B1), on the other hand, did not contribute to embryonic or adult chimeras (Figure 4D). These experiments confirmed that unlike the partially reprogrammed Nanog−/− cell lines previously derived (Silva et al., 2009), the Nanog−/− iPSC lines we report here were pluripotent and fully reprogrammed.
Discussion Although our results seem to contradict previous reports (Silva et al., 2009), we believe that these incongruences are likely explained by a higher efficiency of reprogramming in our hands, which allowed us to observe relatively rare Nanog-independent reprogramming events that were previously undetected. Regardless, our findings underscore the redundant and pliable nature of reprogramming in vitro, further confirming that there are distinct routes to a pluripotent state. One the one hand, this is not surprising in light of recent studies showing that redundant factors within the pluripotency transcriptional network can compensate for loss of Nanog, and lineage-specific transcription factors can replace all canonical reprogramming factors when expressed in the right combinations (Festuccia et al., 2012; Martello et al., 2012; Shu et al., 2013). On the other hand, recent reports that Nanog expression within pluripotent stem cell cultures is not as heterogeneous as previously believed make the finding that it is not required for transition to or maintenance in the pluripotent state surprising (Faddah et al., 2013; Filipczyk et al., 2013).