Research Paper Volume 9, Issue 10 pp 2137—2162

Figure 1. Microarray analysis identifies SOCS1-dependent p53 target genes. (A) Growth curves. Normal human fibroblasts (IMR90) expressing viral oncoprotein E7 were retrovirally infected with either an empty vector (V) or with constitutively activated STAT5A (cS5A) and with either a control shRNA (shNTC) or a shRNA against SOCS1 (shS1 a). Cells were counted and plated for the growth assay. (B) SOCS1 mRNA levels were measured by qPCR using cells collected 7 days post infection, as in (A). (C) Western blots of IMR90 cells at day 7 post infection, as described in (A) for MCM6, phosphorylated Histone H3 (S10) and Tubulin. (D) Western blots of IMR90 cells described in (A) for p53, phosphorylated p53 at serine 15 (p-p53 S15) and SOCS1 levels. (E) DAVID analysis (Kegg pathway) of Affymetrix microarray experiment performed on triplicates of IMR90 cells expressing E7 and either constitutively active STAT5A (cS5A) combined with a control shRNA (NTC) versus cells expressing cS5A combined with an shRNA against SCOS1 (shS1), collected 7 days after infection. (F) Gene Set Enrichment Analysis (GSEA) of differentially regulated genes between the conditions in (D). (G) GSEA of differentially regulated genes. (H) DiRE analysis of genes differentially regulated between cS5A NTC and cS5A shS1 conditions of the Affymetrix microarray analysis. (I) QPCR validation in IMR90 cells expressing the same constructs as mentioned in (A), for the p53 target genes identified by the microarray analysis. All experiments were performed three times, error bars indicate SD of triplicates (growth curves) or standard errors of triplicates (QPCR), *= p<0.05, using the Student’s t test. **=p<0.01, ***=p<0.005.