In the present study we tested five different pharmacologica

In the present study, we tested five different pharmacological agents that are commonly used as selective inhibitors of Nodal/Activin signaling, namely SB431542, LY364947, A-83-01, SB505124, and SB525334 (Inman et al., 2002; Callahan et al., 2002; Laping et al., 2002; DaCosta Byfield et al., 2004; Laping et al., 2007; Sawyer et al., 2003; Peng et al., 2005; Tojo et al., 2005; Vogt et al., 2011), and found that only LY364947 was fully reversible such that Nodal/Activin signaling was activated efficiently after its removal from the culture medium. To our knowledge, this is the first study to demonstrate this unique characteristic of LY364947. It is unclear which chemical and/or physical characteristics of LY364947 enable effective and swift reversal of its negative impact on Nodal/Activin signaling. It may simply be due to its smaller molecular weight (MW=272.30) compared to the other inhibitors, namely SB431542 (MW=384.39), A-83-01 (MW=421.52), SB505124 (MW=335.40), and SB525334 (MW=343.42), that is, LY364947 may diffuse out of the Dihydromyricetin much more easily than the others. Regardless, this unique property of LY364947 allowed us in the present study to investigate critically the duration of Nodal/Activin signaling inhibition that impairs the developmental competence toward DE, without sustained impairment on Nodal/Activin signaling. Compared to SOX17 and FOXA2, the response of CXCR4 to Nodal/Activin signaling was not substantially impaired by the inhibition of Nodal/Activin signaling (Figs. 2B,D). This is interesting in light of a recent study that identified conserved triple SOX17-binding sites within the CXCR4 proximal enhancer, which implicates that the expression of CXCR4 is under the control of SOX17 (Katoh and Katoh, 2010). Nonetheless, CXCR4 was still expressed when the expression of SOX17 was significantly downregulated after the inhibition of Nodal/Activin signaling. It is possible that even with reduced SOX17 expression, CXCR4 expression was maintained by POU5F1 as well as Activin A, because the CXCR4 enhancer also contains a POU-binding site and SMAD-binding sites (Katoh and Katoh, 2010). We only used inhibition of Nodal/Activin signaling to induce NE in this study, because the main purpose of the study was to investigate specifically how Nodal/Activin signaling controls developmental commitment and competence toward DE and NE. Several studies have shown that the inhibition of Nodal/Activin signaling alone is indeed sufficient to induce NE lineages in hESCs (Smith et al., 2008; Patani et al., 2009). However, regulation of other signaling pathways is also involved in the promotion of NE development in hESCs, namely the activation of FGF signaling and the inhibition of BMP signaling (Chambers et al., 2009; Chng et al., 2010). Additionally, enhancement of other pathways, such as Wnt signaling can promote differentiation toward the DE lineage (Payne et al., 2011; Mayhew and Wells, 2010; Bone et al., 2011). Future studies are essential to determine how these other signaling pathways can impact the commitment and competence toward DE and NE. For example, does the activation of FGF signaling or inhibition of BMP signaling impair the competence toward DE? Also, does the activation of Wnt signaling impair competence toward NE? These studies should shed light on the molecular nature of the machinery that is responsible for the maintenance of DE and NE competence. A recent study shows that SIP1 controls cell-fate decisions between the NE and mesendoderm lineages in hESCs (Chng et al., 2010). SIP1 antagonizes TGFβ family signaling via direct interaction with MH2 domains of SMAD1/2/3. The expression of SIP1 is inhibited by the pluripotency regulators, POU5F1 and NANOG, which suppresses NE differentiation. On the other hand, SIP1 expression is activated by SOX2, which limits the mesendoderm-inducing effects of Nodal/Activin signaling. Inhibition of Nodal/Activin signaling results in the upregulation of SIP1, which further interferes with this signaling through positive-feedback to promote NE lineage development (Chng et al., 2010). In addition, the repression of SIP1 expression by POU5F1 and NANOG is relieved when these pluripotency regulators are downregulated by the inhibition of Nodal/Activin signaling. Is the loss of DE competence by LY364947 treatment, as shown in the present study, due to the inhibition of Nodal/Activin signaling by SIP1? This is unlikely the case, because even after LY364947 treatment, hESCs were still able to activate LEFTY1 and NODAL, the downstream target genes of Nodal/Activin signaling, in response to Activin A (Figure 3D). Therefore, the cause of DE competence loss cannot be attributed to impaired transduction of Nodal/Activin signaling.