Generation of MEF and Lung Cancer Cells

Supplementary Methods

Mice

All animal work was done in accordance with the Stanford University Administrative Panel for Laboratory Animal Care guidelines.

Generation of MEF and lung cancer cells

MEFs were derived from E13.5 embryos and cultured as described previously (Johnson et al. 2005). For Atg5fl/fl cells, recombination was verified by genotyping PCR using the following primers: Check2 (ACAACGTCGAGCACAGCTGCGCAAGG), 5L2 (CAGGGAATGGTGTCTCCCAC) and Short2 (GTACTGCATAATGGTTTAACTCTTGC) (Hara et al. 2006).

Mouse non-small cell lung cancer (NSCLC) cell lines were derived from compound mutant mice carrying a spontaneously recombining oncogenic mutant Ras (KrasG12D) knock-in allele and homozygous for p53 alleles silenced by a floxed transcriptional stop element (Johnson et al. 2001; Brady et al. 2011). KrasG12D;p53LSL-wt/LSL-wt mice develop spontaneous, early-onset lung tumors, which were isolated from eleven week old mice. Cell lines were established after culturing and sorting tumor cells for the epithelial cell adhesion molecule (EpCAM).

Supplementary References

Brady CA, Jiang D, Mello SS, Johnson TM, Jarvis LA, Kozak MM, Kenzelmann Broz D, Basak S, Park EJ, McLaughlin ME et al. 2011. Distinct p53 transcriptional programs dictate acute DNA-damage responses and tumor suppression. Cell 145: 571-583.

Hara T, Nakamura K, Matsui M, Yamamoto A, Nakahara Y, Suzuki-Migishima R, Yokoyama M, Mishima K, Saito I, Okano H et al. 2006. Suppression of basal autophagy in neural cells causes neurodegenerative disease in mice. Nature 441: 885-889.

Johnson L, Mercer K, Greenbaum D, Bronson RT, Crowley D, Tuveson DA, Jacks T. 2001. Somatic activation of the K-ras oncogene causes early onset lung cancer in mice. Nature 410: 1111-1116.

Johnson TM, Hammond EM, Giaccia A, Attardi LD. 2005. The p53QS transactivation-deficient mutant shows stress-specific apoptotic activity and induces embryonic lethality. Nat Genet 37: 145-152.

Supplementary Figure Legends

Supp Figure 1: Set-up, analysis, and validation of p53 ChIP-seq experiment

A) Induction of p53 in wild-type MEFs after 6h of doxorubicin treatment. p53-/- MEFs provide a control for antibody specificity. b-actin serves as a loading control. B) Time course qRT-PCR analysis of p53 target gene induction (p21, Mdm2, Noxa, and Prl3) in wild-type MEFs by doxorubicin treatment. The graph shows the average expression +/- SD of technical replicates, with normalization to b-actin. C) p53 ChIP-seq profiles showing p53 binding to known p53 target genes. The schematics show the p53 ChIP-seq profile and the gene organization, with the direction of transcription indicated by the orientation of the exon connectors, upward for 5’end of gene to the left, and downward for the 5’end of the gene to the right. Inverted triangles mark “called” peaks, and the numbers indicate ChIP enrichment. D) List of known p53 target genes in a wide range of functional categories detected as p53-bound in our ChIP-seq experiment. Genes with asterisks were included in the list but have binding sites located >10kb from the gene. E) p53 ChIP-qPCR validation of 35 p53 ChIP-seq peaks, including IgG ChIP in wild-type MEFs and p53 ChIP in p53-/- MEFs as negative controls. Fold enrichment is relative to a negative control genomic region with no p53 binding. Error bars denote standard deviation of technical replicates.

Supp Figure 2: Global characterization of p53-bound regions in the genome

A) Conservation of sequences flanking all centered p53 ChIP-seq peaks in vertebrates, indicating that p53 binding sites are more conserved than surrounding sequences. This is the case both for peaks near genes, and with a lower level of overall conservation, for intergenic peaks. Conservation in vertebrates was calculated with PhastCons. B) Percentage of p53 ChIP-seq peaks with p53 matrix recognition, as determined by matrix scan analysis of all peaks with a stringent p-value of 0.005, showing little difference between the frequency of p53 consensus sites in peaks near genes and intergenic peaks. MEME analysis of peaks with no p53 matrix recognition still detected p53-like consensus sequences as most enriched motifs, indicating that p53 response elements are still present in some of these peaks. C) KEGG pathway GO-terms enriched in our p53-bound gene dataset, with p-values as calculated by DAVID using a modified Fisher Exact Test.

Supp Figure 3: RNA-seq analysis and validation

A) Table of RNA-seq data for selected genes activated by doxorubicin in a p53-dependent manner, showing average tag counts in untreated (ut) and doxorubicin (dox)-treated cells, fold change (FC) with doxorubicin treatment compared to untreated cells, and both p-value and adjusted p-value for multiple testing using the Benjamini-Hochberg method in wt and p53-/- MEFs. Inf denotes infinitely-induced genes, reflecting lack of expression in untreated cells. B) qRT-PCR validation of the genes listed in (A), showing the average fold change +/- SD in MEFs treated with doxorubicin for 6 h relative to untreated MEFs, after normalization to b-actin. Three independent wild-type and p53-/- MEF lines were used. * denotes significantly induced genes in wild-type MEFs, while n.s. indicates no significant induction according to the Student’s t-test. C) Table of RNA-seq data for selected genes highlighted in the text. Shown are average tag counts in untreated (ut) and doxorubicin (dox)-treated wild-type MEFs and p53-/- MEFs, fold change (FC) with doxorubicin treatment compared to untreated cells, and both p-value and adjusted p-value.

Supp Figure 4: Basal p53 regulated genes

A) (Left) Overlap between all p53-bound genes and genes differentially expressed between untreated wild-type and p53-/- cells, with the overlap defining a set of 253 p53-bound and basal p53-regulated genes. (Right) Summary of percentages of basal p53-repressed (higher expression levels in p53-/- MEFs) and induced (lower expression level in p53-/- MEFs) genes in basal p53 regulated and in p53-bound and basal p53 regulated gene lists. B) KEGG pathways and GO-terms for biological processes enriched in our p53-bound and basal p53 regulated gene data set, with p-values as calculated by DAVID using a modified Fisher Exact Test. The genes within each category are shown.

Supp Figure 5: p53 regulates a program of autophagy genes

A) p53 ChIP-seq profiles showing p53 binding to genes encoding proteins involved in upstream regulation of autophagy and lysosomal function. The schematics show the p53 ChIP-seq profile and the gene organization, with the direction of transcription indicated by the orientation of the exon connectors, upward for 5’end of gene to the left, and downward for the 5’end of the gene to the right. Inverted triangles mark “called” peaks, and the numbers indicate ChIP enrichment. Inverted triangles mark “called” peaks and the numbers indicate ChIP enrichment. B) Table of ChIP-seq peak coordinates for Atg10 and Ulk2 with peak enrichments which are not represented in the ChIP-seq peak list showin in Supp Table 2. C) qRT-PCR examining p21 induction in wild-type and p53-/- MEFs in response to 24h DNA damage (doxorubicin), as a control for p53-dependent activation. Expression levels represent the average +/- SD of technical triplicates after normalization to expression in untreated cells and to β-actin. D) qRT-PCR examining p21 induction in human primary fibroblasts transduced with control (ctr) or p53 shRNAs in response to 24h DNA damage (doxorubicin), as a control for p53-dependent activation. Expression levels represent the average +/- SD of technical triplicates after normalization to expression in untreated cells and to β-actin. E) (Top) qRT-PCR analysis of human fibroblasts transduced with a control (ctr) or a second p53 shRNA confirms p53-dependent induction of p53-bound core autophagy genes after 24h doxorubicin treatment. Expression levels represent the average +/- SD of technical triplicates after normalization to expression in untreated cells and to β-actin. (Bottom) Analysis of average p53 and p21 expression levels +/- SD of technical triplicates after normalization to β-actin confirms efficient p53 knockdown and decreased p53 target gene induction. F) qRT-PCR analysis of p21 expression in UVC-treated wild-type and p53-/- MEFs 24h post treatment. Expression levels represent the average +/- SD of technical triplicates after normalization to expression in untreated cells and to β-actin. G) qRT-PCR analysis of p21 expression in doxorubicin-treated wild-type and p53-deficient HCT116 human colon carcinoma cells after 24h. Expression levels represent the average +/- SD of technical triplicates after normalization to expression in untreated cells and to β-actin.

Supp Figure 6: p53 family members contribute to regulation of the autophagy program

qRT-PCR showing p21 expression in untreated (ut) or doxorubicin (dox)-treated wild-type and p53-/- MEFs treated with either control (ctr) or with both p63 and p73 siRNAs. Expression levels represent the average +/- SD of technical triplicates, after normalization to b-actin.

Supp Figure 7: Activation of p53 results in autophagy target gene induction and autophagy in a transactivation-dependent manner.

A) qRT-PCR analysis of p53 target gene expression to confirm p53 activation in wild-type MEFs by Nutlin-3a and DNA damage (doxorubicin) treatment. Expression levels represent the average +/- SD of technical triplicates, after normalization to b-actin. B) Widespread p53 activation in p53LSL-wt/LSL-wt, p53LSL-25,26,53,54/LSL-25,26,53,54, and p53LSL-R270H/LSL-R270H MEFs or KrasG12D;p53LSL-wt/LSL-wt lung cancer cells (2 different lines) infected with Ad-Cre but not Ad-empty was verified by p53 immunofluorescence. p53-positive cells were counted to determine the percentage of cells with recombination of the STOP cassette. Shown are the statistics for the western blot experiments shown in Fig 5 C) and D) and representative immunofluorescence images with p53 staining in green and DAPI in blue. C) qRT-PCR analysis of p21, Ulk1 and Ulk2 expression in p53LSL-wt/LSL-wt lung cancer cells with (Cre) or without (empty) activation of p53 expression. RNA was collected 72h after transduction of cells with Ad-Cre or Ad-empty. Expression levels represent the average +/- SD of technical triplicates, after normalization to b-actin.

Supp Figure 8: Autophagy-deficiency does not compromise DNA damage-induced cell cycle arrest and survival, but impairs p53-dependent apoptosis and suppression of transformation

A) (Top) Schematic drawing of the floxed Atg5 locus, in which exon 3 of Atg5 is flanked by two loxP sites, and which gets removed after Cre activity (Hara et al. 2006). The positions of the genotyping primers used to verify recombination of the floxed allele and the length of the PCR products are indicated. (Bottom) Genotyping PCRs analyzed by ethidium bromide staining confirm the recombination of the floxed allele after infection with Ad-Cre, but not Ad-empty B) (Top) Western blot analysis of Atg5fl/fl MEFs confirming activation of p53 by doxorubicin (dox) and decreased autophagy after Ad-Cre infection relative to Ad-empty infection, as determined by LC3-II levels +/- Bafilomycin A1. β-actin serves as loading control. (Bottom) Quantification of LC3-II protein levels after normalization to β-actin. C) Two dimensional EdU/PI FACS plots from a representative experiment showing intact p53-induced cell cycle arrest in Atg5-deficient MEFs in response to 5Gy γ-irradiation (γ-IR) in primary MEF 18h post treatment. D) (Top) Western blot analysis of E1A;HrasV12;Atg5fl/fl MEFs confirming decreased autophagy after Ad-Cre infection relative to Ad-empty infection, as determined by LC3-II levels +/- Bafilomycin A1. Activation of p53 by doxorubicin (dox) is verified and β-actin serves as loading control. (Bottom) Quantification of LC3-II protein levels after normalization to β-actin. E) Representative two dimensional PI/AnnexinV FACS plots from the analysis of p53-dependent apoptosis in E1A;HrasV12;Atg5fl/fl MEFs infected with Ad-empty (wt) and with Ad-Cre (Atg5-/-), showing the extent of apoptosis in response to doxorubicin. F) qRT-PCR analysis of Ulk1 and Ulk2 expression levels in untreated E1A;HrasV12;wild-type and E1A;HrasV12;Ulk1-/- MEFs transduced with GFP or Ulk2 shRNAs. Expression levels represent the average +/- SD of technical triplicates after normalization to β-actin.

Supplementary Tables

Supp Table 1: ChIP-seq peak list

List of called p53 ChIP-seq peaks in DNA damage-treated wild-type primary MEFs, including information on genomic coordinates, enrichment, q-values, and genes located within 10kb.

Supp Table 2: p53 bound genes

List of 3193 p53-bound genes (RefSeq) derived from p53 ChIP-seq analysis.

Supp Table 3: p53-dependent, DNA damage regulated genes

List of 1323 p53-dependent, DNA damage-regulated genes, including information on tag counts, fold change, p-values, and variances.

Supp Table 4: p53-bound and p53-dependent, DNA damage regulated genes

List of 432 bound and p53-dependent, DNA damage-regulated genes, including information on tag counts, fold change, p-values, and variances.

Supp Table 5: p53-bound and p53-dependent, DNA damage induced genes

List of 365 p53-bound and p53-dependent, DNA damage-induced genes with fold changes, and p-values.

Supp Table 6: p53-bound and p53-dependent, DNA damage induced genes

List of 67 bound and p53-dependent, DNA damage-induced genes with fold changes and p-values.

Supp Table 7: basal p53 regulated genes

List of 1269 basal p53 regulated genes, including information on tag counts, fold change, p-values, and variances. Note that genes upregulated by p53 will have a fold change <1 in p53-/- cells compared to wild-type cells.

Supp Table 8: p53-bound and basal p53 regulated genes

List of 253 p53-bound and basal p53 regulated genes, including information on tag counts, fold change, p-values, and variances. Note that genes upregulated by p53 will have a fold change <1 in p53-/- cells compared to wild-type cells.

Supp Table 9: Full KEGG pathway annotation of p53-bound genes

Full DAVID analysis of KEGG pathways in p53-bound genes, including enrichment, p-values, and specific p53-bound genes within each pathway. KEGG pathways shown in Supp Fig 2 are underlined.

Supp Table 10: Full GO-term biological function annotation of p53-bound genes

Full DAVID analysis of GO-terms for biological function in p53-bound genes, including enrichment, p-values, and specific genes within each biological function category. GO-terms for biological functions shown in Fig 1 are underlined.

Supp Table 11: Full KEGG pathway annotation of p53-bound and p53-dependent DNA damage regulated genes

Full DAVID analysis of KEGG pathways in p53-bound and p53-dependent DNA damage-regulated genes, including enrichment, p-values, and specific genes within each pathway.

Supp Table 12: Full GO-term biological function annotation of p53-bound and p53-dependent DNA damage regulated genes

Full DAVID analysis of GO-terms for biological function in p53-bound and p53-dependent DNA-damage regulated genes, including enrichment, p-values, and specific genes within each biological function category. GO-terms for biological functions shown in Fig 2 are underlined.

Supp Table 13: Full KEGG pathway annotation of p53-bound and basal p53 regulated genes

Full DAVID analysis of KEGG pathways in p53-bound and basal p53 regulated genes, including enrichment, p-values, and specific genes within each pathway. KEGG pathways shown in Supp Fig 4 are underlined.

Supp Table 14: Full GO-term biological function annotation of p53-bound and basal p53 regulated genes

Full DAVID analysis of GO-terms for biological function in p53-bound and basal p53 regulated genes, including enrichment, p-values, and specific genes within each biological function category. GO-terms for biological functions shown in Supp Fig 4 are underlined.

Supp Table 15: Primer lists

List of primers used in this study, including primers for ChIP-seq validation, RNA-seq validation, autophagy gene qRT-PCR, and genotyping.