Supplemental Data: Figure Legends

SUPPLEMENTAL DATA: FIGURE LEGENDS

Figure S1. Alignment of the amino acid sequences of the p53 C-terminal region from different vertebrate species. The arrow shows the position of S392 in human and the corresponding serine in the other species. The percentage of residue conservation is indicated by the histogram. Sequences for Human (P04637), Japanese macaque MACFU (P61260), Rat (P10361), Mouse (P02340), Pig (Q9TUB2), Bovin (P67939), Dog (Q29537), Salmon (P25035) and Chick (P10360) p53 are from UniProt.

Figure S2. Strategy used to mutate S392 into alanine in the HCT-116 cell line by CRISPR/Cas9 genome editing. (a) Step 1 of the strategy. Genome editing by CRISPR/Cas9 technology involves creation of targeted DNA breaks to promote the integration of a DNA repair template encoding for the mutation of interest by homologous recombination. The upper part of the figure displays a schematic representation of the TP53 gene. It is composed of 11 exons, the full-length p53 protein being encoded by exons 2 to 11. S392 is located in the last exon. The sgRNAs guides needed to induce the DNA double-strand breaks are indicated in orange and dark-pink and the red arrows indicate the cutting sites. The repair template (RT) used to introduce the S392A mutation is a synthetic single-stranded DNA oligonucleotide (ssODN) of 200 bases. It contains the desired mutation (S392A) and four silent mutations near and in the PAM sequence areas recognized by the two sgRNAs guides. These mutations prevent that the ssODN becomes a substrate for the Cas9n enzyme. The green lines indicate the homology arms of the RT with p53 genomic sequence. We first transfected the HCT-116 cell line with the repair template and two pX462 vectors coding for the Cas9n nickase and the sgRNA guides 1 and 2. Because the homologous recombination is a rare event, we created pools of 50 cells/well (in 96-well plates) with the cells recovered after a puromycin selection of 2 days. The genomic DNA of these pools was screened by PCR with primers specific for the RT sequence. Positives pools were then subcloned and allowed to develop. Clones developing from individual cells were again screened by PCR on genomic DNA and the positive clones further characterized by sequencing of exon 11, cDNA expression analysis and sequencing as well as protein expression analysis by western blot. At the end of this first step, the best clone contained one p53 allele with a correct integration of the RT and a second allele with a deletion of 23 bases and a partial insertion of the repair template. This indel mutation generates a stop codon resulting in a truncated p53 form of 372 amino acids (p53 1-372). Protein expression analysis showed that the two p53 forms, p53 S392A and p53 1-372, are expressed in the selected HCT-116 clone. (b) Step 2 of the strategy. The goal of this second step is to perform a second round of CRISPR/Cas9n editing, on the clone selected in step 1, to knock out byindel mutations the incorrect allele encoding for p53 1-372. The second ATG of p53 at position 40 (exon 4) is targeted in this approach. The new sgRNAs guides are indicated in orange and dark-pink and the red arrows indicate the cutting site. The selected HCT-116 clone was transfected with the pX462 plasmids without repair template. Since indel mutations resulting from DNA breaks are a common event, the transfected cells were directly cloned individually. The genomic DNA of clones was next screened by PCR with primers surrounding the cutting area. Further characterization of positive clones involved sequencing of exon 4, cDNA expression analysis and sequencing, and ultimately analysis of protein expression by western blot. The selected clone after this step 2 displays an allele encoding for the desired mutation (S392A) and a second non-functional allele with an insertion of 25 bases (in red and highlighted in the p53wt sequence) that generates a frameshift and formation of a stop codon in exon 4. As expected, western blot analysis for p53 expression confirmed that the incorrect allele (expressing p53 1-372) is no longer functional.

Figure S3. Genomic sequence of the HCT-116 p53 KO cell line generated by CRISPR/Cas9n.The ATG40 of p53 is highlighted by the green square. The targets of the sgRNA guides 3 and 4 are indicated in orange and dark-pink. Sequence analysis revealed in both alleles a homozygous deletion of 8 bases resulting in the formation of a premature stop codon and absence of p53 protein expression.

Figure S4. Analysis of the p53 mRNA level in HCT-116 p53 KO, p53 wt+/+ and p53 S392A+/- cells by RT-qPCR. Cells were treated for 24 h with 1 µM CPT before RNA extraction. The primers used for p53 are the following: 5’-AGATAGCGATGGTCTGGCCC-3’ (forward) and 5’-AACCTCAGGCGGCTCATAGG-3’ (reverse). PCR product length is 107 bp.56 Values are means ± SD of 6 independent experiments. The Kruskal–Wallis test was used to compare independent groups of numerical data. Since the Kruskal–Wallis test was significant, a nonparametric post hoc test (Steel-Dwass all pairs test) was used to compare the group pairs of interest (, p<0.05; N.S., non-significant). Although lower, the p53 mRNA level of S392A cells was not significantly different from that of p53 wt cells. The mRNA level of p53 KO cells is markedly and significantly lower than that of the two other cell lines, probably due to the surveillance mechanism of nonsense-mediated mRNA decay.

Figure S5. Measure of the expression level of the genes encoding for p21waf1, BAX, PIG3, BCL-2 and PUMA by RT-qPCR in p53 KO HCT-116 cells and cells expressing p53 wt or S392A.Cells were treated for 24 h with 1 µM CPT before RNA extraction. Values are means ± SD of 6 independent experiments. The Kruskal–Wallis test was used to compare independent groups of numerical data. When the Kruskal–Wallis test was significant, a nonparametric post hoc test (Steel-Dwass all pairs test) was used to compare the group pairs of interest (*, p<0.05; N.S., non-significant). Exact P-values and test details are available in the tables on the right.

Figure S6. No difference in the MDM2 protein level of HCT-116 p53 KO, p53 wt+/+ and p53 S392A+/- cells. Western blot analysis of p53 and MDM2 levels after a 24 h treatment with 1 µM CPT. Antibodies: mouse anti-p53 (DO-1) (dilution of 1:2,000; Millipore OP43) and mouse anti-MDM2 (Ab-1) (dilution of 1:500; Millipore OP46).

Figure S7. Presence or absence of S392 phosphorylation does not impact on the phosphorylation of S15, S46 or S315 and the acetylation of K382 in H1299 cells. H1299 inducible (Tet-On) clones expressing p53 wt, S392A or S392E as well as cells transfected with the empty vector were incubated for 8 h with 2 µg/mL doxycycline to induce p53 expression and an additional 24h with 2 µg/mL doxycycline and 1 µM CPT. After treatment, total protein extracts were isolated and analyzed for different phosphorylations of p53 (S15, S46, S315 and S392). β-actin was used as loading control. Primary antibodies employed: mouse anti-p53 (DO-1) (1:2,000 dilution; Millipore OP43); rabbit anti-pSer15 p53(1:1,000 dilution; Cell Signaling 9284); rabbit anti-pSer46 p53(1:1,000 dilution; Cell Signaling 2521); rabbit anti-pSer315 p53(1:1,000 dilution; Cell Signaling 2528), rabbit anti-pSer392 p53(1:1,000 dilution; Cell Signaling 9281) and rabbit anti-acetyl-K382 p53 (1:1,000 dilution; Cell Signaling 2525).

Figure S8. p53wt and the S392 mutant display a similar profile of ubiquitinationin HCT-116 cells. HCT-116 cells expressing p53 wt, the S392A mutant as well as p53 KO cells were treated with 1 µM CPT for 24 h. Detection of p53 was assessed by western blot analysis. A long exposure of the p53 blot allowed comparingp53 ubiquitination in cells expressing p53wt and the S392A mutant.

Figure S9. The inhibition of p38 MAPK (p38 mitogen-activated protein kinases), CKII (casein kinase II) or CDK9 (Cyclin-dependent kinase 9) has no impact on S392 phosphorylation. In the left panel, HCT-116 p53 wt cells received 20 µM of the indicated kinase inhibitors (or DMSO) 1 h before addition of CPT for 16h. The expression of p53 and its form phosphorylated on S392 were then assessed by western blot analysis. In the right panel, p53 wt H1299 Tet-On cells were treated overnight with 2 µg/mL doxycycline in the presence or not (NT) of 20 µM of the indicated inhibitors. The next day, we added 1 µM CPT for an additional 4 h. Then, total protein extracts were prepared. Expression of p53 and its form phosphorylated on S392 were assessed by western blot analysis. The following inhibitors were used: the p38 MAPK inhibitor SB203580 (Cell Signaling, 5633), the CKII inhibitor TBB (Santa Cruz, sc-202830) and CDK9 inhibitor II (Santa Cruz, sc-203326).

Figure S10. Decreased mitochondrial localization of p53 S392A in presence of a genotoxic stress with 10 µM CPT in H1299 cells. H1299 Tet-On clones expressing p53wt, S392A and S392E were incubated overnight with 2 µg/mL doxycycline to induce p53 expression. The next day, cells were treated with 10 µM camptothecin for 4 h. Total protein extracts, mitochondrial and cytosolic fractions were then isolated. Detection of p53, β-actin, GRP75 and PCNA was performed by western blot analysis.

Figure S11. No other phosphorylation and acetylation examined was found to be enriched in the mitochondrial fraction of both HCT-116 and PA-1 cell lines treated with 1 µM DOX. Western blot analysis of four different phosphorylated and two acetylated forms of p53 in the mitochondrial (M) and nuclear (Nuc) fractions as well as total protein extracts (Tot) of HCT-116 and PA-1 cells treated for 4 h with 1 µM doxorubicin. In the PA-1 cell line (but not in HCT-116 cells), P-S315 p53 was clearly present in the mitochondrial fraction while absent from the nucleus. Phosphorylation of S315 was previously described to negatively impact on p53 pro-apoptotic mitochondrial activity.4 Primary antibodies employed: mouse anti-p53 (1:2,000 dilution; Millipore OP43);rabbit anti-pSer15 p53(1:1,000; Cell Signaling 9284); rabbit anti-pSer33 p53(1:1,000; Cell Signaling 2526); rabbit anti-pSer37 p53(1:1,000; Cell Signaling 9289); rabbit anti-pSer315 p53(1:1,000; Cell Signaling 2528), rabbit anti-acetyl-K379 p53 (1:1,000; Cell Signaling 2570) and rabbit anti-acetyl-K382 p53 (1:1,000; Cell Signaling 2525).

Figure S12. The S392A mutant weakly interacts with BAK in mitochondrial fractions isolated from H1299 Cells. H1299 Tet-On clones expressing the empty vector as well as cells expressing p53, S392A and S392E were incubated overnight with 2 µg/mL doxycycline to induce p53 expression. The next day, cells were treated with 1 µM CPT for 4 h. Mitochondrial and cytosolic fractions were isolated as described in the materials and methods. For co-immunoprecipitation, the mitochondrial pellet was lysed in NP40 lysis buffer supplemented with protease and phosphatase inhibitors. Next, 500 µg of proteins were incubated overnight with 1 µg anti-BAK antibody in a final volume of 600 µL. Protein complexes were immunoprecipitated with 20 µL of agarose protein G slurry (Pierce, 20398) for 30 min at 4°C, followed by an additional incubation of 30 min at room temperature. The beads were washed 4 times with 700 µL of NP40 lysis buffer. Protein complexes were eluted with 20 µL of 1x Laemmli sample buffer and analyzed by western blot. Detection of p53 and BAK was performed by western blot analysis. Detection of GRP75 and PCNA were performed to control for the loading and the purity of the fractions. Total protein extracts was performed to control for the level of the expression of p53wt and its S392A mutant.

Figure S13. The S392A mutant weakly interacts with BAK in mitochondrial fractions isolated from HCT-116 Cells. HCT-116 cells expressing p53 wt, the S392A mutant as well as p53 KO cells were treated with 1 µM CPT for 4 h. After incubation, the mitochondrial and cytosolic fractions were isolated as described in the materials and methods. For co-immunoprecipitation, the mitochondrial pellet was lysed in NP40 lysis buffer supplemented with protease and phosphatase inhibitors. Next, 100 µg of proteins were incubated overnight with 1 µg of anti-BAK antibody in a final volume of 600 µL. Protein complexes were immunoprecipitated with 20 µL of agarose protein G slurry (Pierce, 20398) for 30 min at 4°C, followed by an additional incubation of 30 min at room temperature. The beads were washes 4 times with 700 µL of NP40 lysis buffer. Protein complexes were eluted with 20 µL of 1x Laemmli sample buffer and analyzed by western blot. Detection of p53 and BAK were assessed by western blot analysis. Detection of GRP75 and PCNA was performed to control for the loading and the purity of the fractions. Total protein extracts were also performed to control for the induction of p53 wt and S392A.

Figure S14. Phosphorylation of S392 does not modulate the p53-BAX interaction in H1299 cells. H1299 Tet-On clones expressing p53 wt, S392A and S392E as well as cells transfected with the empty vector were incubated overnight with 2 µg/mL doxycycline and 1 µM of CPT. The next day, total protein extracts were prepared and immunoprecipitated with an anti-BAX antibody. The interaction with p53 was detected by western blot.

1