Ariance (ANOVA), followed by Newman-Keuls multiple comparison tests using software (Prism 4.0, GraphPad Software). In the case of single mean comparison, data were analyzed by t test. p values#0.05 are regarded as statistically significant.Results TNF-a induces STAT3 activation in human NPCs at delayed time pointsPrevious work in our laboratory has demonstrated that TNF-a increases astrocytic differentiation and inhibits neuronal differentiation of human 18325633 NPCs. Furthermore, TNF-a induces astrogliogenesis through STAT3 signaling, since siRNA specifically 
targeting STAT3 (siSTAT3) inhibited TNF-a-induced astrogliogenesis [17,18]. To elucidate the additional mechanism involved in TNF-a-induced STAT3 activation and subsequent astrogliogenesis, we treated human NPCs with TNF-a and studied STAT3 phosphorylation at different time points (30 min, 6 h, and 24 h) (Figure 1A). TNF-a did not induce immediate STAT3 phosphorylation at 30 min. However, TNF-a induced STAT3 phosphorylation at 6 h and continued to induce even stronger STAT3 phosphorylation at 24 h (Figure 1A). The delayed STAT3 activation by TNF-a indicates that TNF-a may play an indirect role on STAT3 activation: secreted factors produced by TNF-a-treated NPCs activated the STAT3 pathway at later time points (6 h and 24 h). To test this hypothesis, human NPCs were treated with TNF-a for 30 min, 6 h and 24 h, and supernatants were collected as conditioned medium (CM). Parallelcultured NPCs were then treated with these different time point conditioned 1662274 media (TNF-a-treated (TNF-a-CM) or control NPCCM (Con-CM)) for 30 min and cell lysates were collected for Western blot. TNF-a-CM collected at 30 min did not induce a significant increase of STAT3 phosphorylation. In contrast, TNFa-CM collected at 6 h moderately increased STAT3 phosphorylation; and TNF-a-CM collected at 24 h showed a significant increase of STAT3 phosphorylation as compared with Con-CM treatment (Figure 1B). This result suggests that TNF-a-induced soluble factors, which are highly produced at 24 h, subsequently induce STAT3 phosphorylation in human NPCs in an HIF-2��-IN-1 autocrine manner. We next studied the kinetics of CM-mediated STAT3 phosphorylation in NPCs. To exclude the effect of residual TNF-a in CM, human NPCs were treated with TNF-a for 6 h, rinsed twice with X-Vivo 15 and then maintained in fresh X-Vivo 15 medium. Twenty-four hours later, the TNF-a-free cell supernatants were collected as TNF-a-free-CM. TNF-a-free-CM treatment induced an immediate STAT3 phosphorylation at 30 min, but not at 6 h or 24 h (Figure 1C). This result suggests that secreted factors produced by TNF-a-treated NPCs have differential kinetics in activating the STAT3 pathway compared to TNF-a. To further characterize TNF-a-induced STAT3 activation in NPCs, we performed immunocytochemical studies with NPC culture using antibodies against phospho-STAT3 and nestin, a neural progenitor cell marker. Consistent with the Western blot result, TNF-a did not increase STAT3 phosphorylation or nucleus translocation at the early time point (30 min). However, at 24 h following TNF-a treatment, we observed apparent STATFigure 1. TNF-a induces delayed STAT3 activation in human NPCs. A. Human NPCs were treated with 20 ng/ml TNF-a for 30 min, 6 h, and 24 h. Expression of phospho-STAT3 (P-STAT3) and get 117793 total-STAT3 (T-STAT3) were detected by Western blotting. b-actin was used as a loading control. B. Human NPCs were treated with 20 ng/ml TNF-a for 30 min, 6 h, and 24 h. Supernata.Ariance (ANOVA), followed by Newman-Keuls multiple comparison tests using software (Prism 4.0, GraphPad Software). In the case of single mean comparison, data were analyzed by t test. p values#0.05 are regarded as statistically significant.Results TNF-a induces STAT3 activation in human NPCs at delayed time pointsPrevious work in our laboratory has demonstrated that TNF-a increases astrocytic differentiation and inhibits neuronal differentiation of human 18325633 NPCs. Furthermore, TNF-a induces astrogliogenesis through STAT3 signaling, since siRNA specifically targeting STAT3 (siSTAT3) inhibited TNF-a-induced astrogliogenesis [17,18]. To elucidate the additional mechanism involved in TNF-a-induced STAT3 activation and subsequent astrogliogenesis, we treated human NPCs with TNF-a and studied STAT3 phosphorylation at different time points (30 min, 6 h, and 24 h) (Figure 1A). TNF-a did not induce immediate STAT3 phosphorylation at 30 min. However, TNF-a induced STAT3 phosphorylation at 6 h and continued to induce even stronger STAT3 phosphorylation at 24 h (Figure 1A). The delayed STAT3 activation by TNF-a indicates that TNF-a may play an indirect role on STAT3 activation: secreted factors produced by TNF-a-treated NPCs activated the STAT3 pathway at later time points (6 h and 24 h). To test this hypothesis, human NPCs were treated with TNF-a for 30 min, 6 h and 24 h, and supernatants were collected as conditioned medium (CM). Parallelcultured NPCs were then treated with these different time point conditioned 1662274 media (TNF-a-treated (TNF-a-CM) or control NPCCM (Con-CM)) for 30 min and cell lysates were collected for Western blot. TNF-a-CM collected at 30 min did not induce a significant increase of STAT3 phosphorylation. In contrast, TNFa-CM collected at 6 h moderately increased STAT3 phosphorylation; and TNF-a-CM collected at 24 h showed a significant increase of STAT3 phosphorylation as compared with Con-CM treatment (Figure 1B). This result suggests that TNF-a-induced soluble factors, which are highly produced at 24 h, subsequently induce STAT3 phosphorylation in human NPCs in an autocrine manner. We next studied the kinetics of CM-mediated STAT3 phosphorylation in NPCs. To exclude the effect of residual TNF-a in CM, human NPCs were treated with TNF-a for 6 h, rinsed twice with X-Vivo 15 and then maintained in fresh X-Vivo 15 medium. Twenty-four hours later, the TNF-a-free cell supernatants were collected as TNF-a-free-CM. TNF-a-free-CM treatment induced an immediate STAT3 phosphorylation at 30 min, but not at 6 h or 24 h (Figure 1C). This result suggests that secreted factors produced by TNF-a-treated NPCs have differential kinetics in activating the STAT3 
pathway compared to TNF-a. To further characterize TNF-a-induced STAT3 activation in NPCs, we performed immunocytochemical studies with NPC culture using antibodies against phospho-STAT3 and nestin, a neural progenitor cell marker. Consistent with the Western blot result, TNF-a did not increase STAT3 phosphorylation or nucleus translocation at the early time point (30 min). However, at 24 h following TNF-a treatment, we observed apparent STATFigure 1. TNF-a induces delayed STAT3 activation in human NPCs. A. Human NPCs were treated with 20 ng/ml TNF-a for 30 min, 6 h, and 24 h. Expression of phospho-STAT3 (P-STAT3) and total-STAT3 (T-STAT3) were detected by Western blotting. b-actin was used as a loading control. B. Human NPCs were treated with 20 ng/ml TNF-a for 30 min, 6 h, and 24 h. Supernata.