Targeting codon 158 p53-mutant cancers via the induction of p53 acetylation

Gain of function (GOF) DNA binding domain (DBD) mutations of TP53 upregulate chromatin regulatory genes that promote genome-wide histone methylation and acetylation. Here, we therapeutically exploit the oncogenic GOF mechanisms of p53 codon 158 (Arg158) mutation, a DBD mutant found to be prevalent in lung carcinomas. Using high throughput compound screening and combination analyses, we uncover that acetylating mutp53R158G could render cancers susceptible to cisplatin-induced DNA stress. Acetylation of mutp53R158G alters DNA binding motifs and upregulates TRAIP, a RING domain-containing E3 ubiquitin ligase which dephosphorylates IĸB and impedes nuclear translocation of RelA (p65), thus repressing oncogenic nuclear factor kappa-B (NF-ĸB) signaling and inducing apoptosis. Given that this mechanism of cytotoxic vulnerability appears inapt in p53 wild-type (WT) or other hotspot GOF mutp53 cells, our work provides a therapeutic opportunity specific to Arg158-mutp53 tumors utilizing a regimen consisting of DNA-damaging agents and mutp53 acetylators, which is currently being pursued clinically.

High-content screening of anti-cancer compounds and epigenetic modulators on isogenic p53 clones. The efficacy of the tested compounds was quantified relative to the mean viability of vehicle-treated cells (384-well format, in triplicates) (n = 1 independent experiment). Heatmap shows compounds filtered for specificity to p53 -/cells (A) or p53 wt cells (B) (growth inhibition > 50% in specific cell type).
(B-C) Electrophoretic mobility shift assay (EMSA) was performed using proteins expressed with in vitro translation. WT and Arg 158 p53 were expressed from their respective plasmids, and protein acetylation was performed using the HAT domain of p300/HAT and acetyl-CoA. Western blot detects the expressed and modified proteins (B). The expressed proteins were incubated with 50ng of DNA oligos, and separated on a 6% non-denaturing TBE gel. SYBR Green (left) and SYPRO Ruby (right) were used for detection of nucleic acid and protein respectively (n = 2 independent experiments).
(E) Extent of apoptosis was quantified with Annexin V staining in vector control (shNT and shLuc) or p53 knockdown (shp53.1, shp53.2) stable cells. Data are represented as mean ± SD (n = 3 independent experiments). Two tailed Student's ttest; *P < 0.05. Aetylated p53 (lys379)   Figure 4J). 8 fields were taken in each treatment group. Integrated density for each nucleus was determined by ImageJ. Data are represented as mean of each field± SD in a representative experiment (n = 3 independent experiments).
(C) Scatter plots of individual nuclei for p53 R158G are shown (X: log signal for acetyl-p53; Y: log signal for p-p53). Each scatter plot consists of > 400 nuclei. Gating was determined from vehicle control, upward shift indicates increase in phosphorylated p53, rightward shift indicates acetylated-p53.
(D) Tabulation of percentage of cell distribution for each p53 R158G clones is displayed (Q1: p-p53 negative, acetyl-p53 negative; Q2: p-p53 negative, acetyl-p53 positive; Q3: p-p53 positive, acetyl-p53 negative; and Q4: p-p53 positive, acetyl-p53 positive), expressed as mean ± SD (n = 3 independent experiments).  with key residues shown. The gray colored cartoon depicts the DBD while the orange cartoon is the DNA backbone with sticks representing the nucelobases in A; the black sphere represents a Zinc ion which is required for stabilizing the DBD; the R158 sidechain is shown as grey sphere (carbon atoms) and blue spheres (nitrogen atoms); the Lysine residues are shown as gray sticks with the terminal nitrogen colored in blue while upon acetylation, the sidechain oxygen is shown as red; the other sidechains shown are colored as: carbongrey, oxygenred, nitrogenblue; the dimerization helix is shown as green.  Heatmaps showing the enrichment of p53 ChIP-Seq peaks (± 3,500 bp from peak center) identified from p53 wt or p53 R158G cell line in vehicle-or belinostat/cisplatin-treated conditions (row), over the peaks of p53 occupancy in all four conditions (column).      (C-D) Distribution of p65 (in cytoplasm or nucleus) was determined by Western blot (C) after nuclear-cytosolic fractionation. TATA-box binding protein (TBP) (nuclear) and α-tubulin (cytoplasmic) were used as loading controls. Nuclear p65 signal was quantified with densitometry after normalizing to TBP (D). Data are presented as mean ± SD (n = 3 independent experiments). Two tailed Student's t-test; *P < 0.05.  Total p65    Growth curve analysis of Calu-1 p53 wt (A), p53 -/-(B), p53 R158G (C) and p53 R158G(k20A) (D) in xenografts treated with vehicle or cisplatin (CDDP; 4 mg/kg). Tumor sizes are presented as mean ± SEM. Two-way ANOVA with Bonferroni correction; *P < 0.05; **P < 0.01; ***P < 0.001. (E) Western blots demonstrating changes of the indicated proteins in tumors of respective treatment (n = 3 independent tumours). β-actin shown as loading control. Densitometric quantification of TRAIP expression was tabulated on the right. Relative fold change is normalized to β-actin, relative to vehicle control tumors and presented at mean ± SEM. Two tailed Student's t-test; *P < 0.05. (F-G) Immunohistochemistry staining analyses of intracellular expressions of p65 and Ki67 in respective tumors. Representative images at 20× showing p65 staining (F) and Ki67 staining (G). Scale bar, 50 µm. Quantification of positively-stained cells (%) in p53 wt and p53 R158G respectively was tabulated on the right. Data are represented as percentage of positive cells ± SD (n = 5 independent tumours). Two tailed Student's t-test; for xenograft models, *P < 0.05, **P < 0.01.