Supportingfigure Legends

Supportingfigure legends

Fig. S1 Selection of the experimental system. (A) Scheme showing the strategy for the selection of the experimental system. (B) MCF-7 cells, SJSA cells, HCT116 TPp53+/+ cells and HCT116 TP53-/- cells were treated with five p53 activating compounds, RITA, nutlin, actinomycin D (ActD), cisplatin (CDDP) and 5-fluorouracil (5-FU), with or without pretreatment with p53 inhibitor PFT-α. Apoptosis was assayed by FACS of Annexin V-propidium iodide (PI) stained cells, Rhodamine 123 (Rho123)-PI stained cells or Sub-G1 population detection using PI staining. Cell cycle was assessed by FACS of PI-stained cells.

Fig. S2 Hierarchical clustering analysis of the shRNA screen data from present study (MCF-7) and the previously published shRNA screen data (in HCT116 cells, Sullivan et.al., 2012). Shown is the heatmap in which the abundance of individual shRNAs is indicated by blue color: light blue, low abundance; dark blue, high abundance.

Fig. S3 Network analysis of SLNs and RNs revealed Sp1 as the only RN at the center of the network with p53.

Fig. S4 Sp1 is a key determinant of p53-mediated apoptosis. The depletion of Sp1 in MCF-7 (A) and HCT116 cells (B) with two different Sp1 shRNAs rescued cells from inhibition by RITA as assessed by a colony formation assay. (C) Ectopic expression of Sp1 promotes apoptosis induced by nutlin in MCF-7 and HCT116 cells, but not in HCT116 TP53-/- cells as assessed by FACS of Annexin V-PI stained cells.

Fig. S5 Effect of Sp1 depletion on the gene expression profile induced by RITA in MCF-7 cells.The plot presents the fold change of gene expression significantly affected by RITA in MCF-7 cells. Data are presented as Log2(RITA/DMSO). p < 0.001.

Fig. S6 shRNA screen data for p53-induced pro-apoptotic genes. The experimental procedure of genome-wide short hairpin RNA (shRNA) screen in MCF7 cellsand data processing were performed as described inSI Materials and Methods. The blue bars indicate the ratio of read counts between RITA-treated cells and DMSO-treated cells (RITA versus DMSO). The red curve indicates -Lg(p value).

Fig. S7Comparison of gene expression profiles induced by RITA and nutlin in Sp1-depeleted and non-depleted cells.Venn diagrams show the intersection of genes significantly changed by RITA or nutlin treatment in MCF-7 cells without(A) or with (B)Sp1 depletion. p < 0.001.(C) Heatmap representation of expression profiles of cell cycle arrest genesin MCF-7 cells with or without Sp1 depletion upon RITA or nutlin treatment. Columns indicate arrays and rows indicate genes.

Fig. S8Pathways enriched with SLNs.Pathway analysisof SLNs identified via analysis of the genome-wideshRNA screendata was performed using DAVIDtool.p < 0.05.

Fig. S9Pathways enriched with227 p53 target genes co-repressed by p53 and Sp1upon RITA treatment.Pathway analysis of these 227 genes was performed using DAVIDtool.p < 0.05.

Fig. S10Network analysis of 227 p53-targets co-repressed by p53 and Sp1. Analysis of functional connections between proteins encoded by these 227 genes was performed using STRING software. Only nodes having at least two connections are shown.

Fig. S11 Binding profiles of p53 on selected genes.Binding profiles of p53 on selected genes in MCF-7 cells treated with DMSO, RITA or nutlin were plottedusing ChIP-seq data. Regions amplified by ChIP-PCR are indicated in each gene.

Fig. S12. Ectopic expression of Sp1 promotes the binding of Sp1 to p53 target genes and p53-mediated pro-apoptotic transcriptional repression in MCF-7 cells upon nutlin treatment. (A) Binding of Sp1 to selected genes (described in Fig. 4) was assessed by ChIP-PCR in nutlin-treated MCF-7 cells with or without ectopic expression of Sp1. (B) The mRNA level of selected genes was examined by qPCR. In (A) and (B) data are presented as mean ± SD, n = 3.

Fig. S13. Sp1 protein level is important for the formation of p53/Sp1 complex. (A) The amount of Sp1 bound to p53 upon RITA or nutlin treatment of MCF-7 cells was detected by p53 co-immunoprecipitation followed by immunoblotting. Right panel: equal amount of immunoprecipitated p53 was loaded to assess bound Sp1. (B) The amount of Sp1 bound to p53 upon nutlin treatment of MCF-7 cells with or without ectopic Sp1 expression was detected by p53 co-immunoprecipitation followed by immunoblotting. Equal amount of immunoprecipitated p53 was loaded to assess bound Sp1.

Fig. S14. Sp1 is not required for the induction of MDM2 transcription by p53.mRNA level of MDM2 in MCF-7 cells with or without Sp1 depletion upon RITA or nutlin treatment was examined by qPCR. Data are presented as mean ± SD, n = 3.

Table S1 Raw data of the genome-wide shRNA screen

Table S2 Validation of genome-wide shRNA screen. 15 genes were selected from screen based on threecriteria (p value < 0.1, FDR < 0.3, p(wZ) < 0.1) and re-tested for validation purpose. Blue indicates reduced viability (synthetic lethal node, SLN), purple indicates enhanced viability (resistance node, RN). Results were compared with statistic analysis of screen data.

Table S3 List of RNs and SLNs. These resistance nodes and synthetic lethal nodes were selected based on a stringent criteria: p value < 0.05, FDR < 0.1, p(wZ) < 0.1.

Table S4List of RNs that have been identified as direct transcriptional targets of Sp1.

Table S5Prediction of Sp1 response elements in 173 'RITA/nutlin common' targets using oPOSSUM software.

Table S6Prediction of Sp1 response elements in 54 ‘RITA-specific’ p53 targets using oPOSSUM software.

Table S7 List of primers