Palmdelphin, a novel target of p53 with Ser46 phosphorylation, controls cell death in response to DNA damage

Abstract

The tumor suppressor gene p53 regulates apoptosis in response to DNA damage. Promoter selectivity of p53 depends on mainly its phosphorylation. Particularly, the phosphorylation at serine-46 of p53 is indispensable in promoting pro-apoptotic genes that are, however, poorly determined. In the current study, we identified palmdelphin as a pro-apoptotic gene induced by p53 in a phosphorylated serine-46-specific manner. Upregulation of palmdelphin was observed in wild-type p53-transfected cells, but not in serine-46-mutated cells. Expression of palmdelphin was induced by p53 in response to DNA damage. In turn, palmdelphin induced apoptosis. Intriguingly, downregulation of palmdelphin resulted in necroptosis-like cell death via ATP depletion. Upon DNA damage, palmdelphin dominantly accumulated in the nucleus to induce apoptosis. These findings define palmdelphin as a target of serine-46-phosphorylated p53 that controls cell death in response to DNA damage.

Figure 1

PALMD is induced by p53 when ser46 is phosphorylated. (a) A genome-wide search of the pro-apoptotic genes promoted by phosphorylated p53 at ser46. Results of microarray and ChIP sequencing were overlapped to identify novel targets of phospho-ser46 of p53. (b) Control vector (control), wild-type p53 (wt-p53), and mutant p53 at serine-46 (p53S46A) were transfected into SaOS2 cells. Real-time PCR was performed and data represents fold change in PALMD relative to the GAPDH. *P<0.01. (c) Samples described in b were analyzed by western blotting. Protein-specific bands were detected with anti-PALMD (top panel), anti-phospho-p53(Ser46) (second panel), anti-p53 (third panel), or anti-tubulin (bottom panel). (d) The highest peak region of PALMD by ChIP sequencing that includes p53 consensus sites (red color). This region was amplified in the ChIP assay. (e) The binding of p53 onto PALMD was identified by the ChIP assay. Cell lysates from samples described in b were immunoprecipitated with anti-p53 or anti-immunoglobulin (Ig)G. ChIP assay and subsequent real-time PCR were performed. The data were normalized with the level of input control. *P<0.01. (f) DYRK2-depleted cells failed to induce PALMD expression. U2OS cells were transfected with non-silencing siRNA or siDYRK2. Cells were treated with ADR or left untreated. Cell lysates were immune-stained with anti-PALMD, anti-phospho-p53(Ser46), anti-p53, anti-DYRK2, or anti-tubulin

Figure 2

PALMD is promoted by p53 in response to DNA damage. SaOS2 (p53-deficient) and U2OS (p53-proficient) cells were treated with ADR (ADR or A) or left untreated (control or C). (a) mRNA expression of PALMD in response to DNA damage. Data represent fold change of PALMD in ADR-treated cells relative to the control. *P<0.01. (b) Protein expression level of PALMD in response to ADR exposure. Cell lysates were immunoblotted against anti-PALMD (top panel), anti-phospho-p53(Ser46) (second panel), anti-p53 (third panel), or anti-tubulin (bottom panel). (c) The binding of endogenously expressing p53 onto PALMD under stressed condition. Cell lysates were immunoprecipitated with anti-p53 or anti-immunoglobulin (Ig)G. ChIP assay and subsequent real-time PCR were performed. The data were normalized with the level of input control. *P<0.01. (d) PALMD level in p53-depleted cells. U2OS cells were transfected with non-silencing siRNA or p53 siRNA (sip53). Cells were treated or left untreated with ADR and harvested at indicated times. Cell lysates were immunoblotted with anti-PALMD (top panel), anti-phospho-p53(Ser46) (second panel), anti-p53 (third panel), or anti-tubulin (bottom panel)

Figure 3

PALMD induces apoptosis in response to DNA damage. (a) The specificity of the siRNAs of PALMD. Three different siRNAs, siPALMD-1, siPALMD-2, and siPALMD-3 or non-silencing siRNA were transfected into U2OS cells. After 24 h incubation, cells were harvested and total RNA was isolated. mRNA (upper panel, data represent fold expression of PALMD relative to the non-silencing siRNA) and protein expression (lower panel) were examined. (b) Apoptotic assay in PALMD-depleted cells. U2OS cells were seeded into four-well chamber slides and transfected with non-silencing siRNA, siPALMD-1, siPALMD-3, or sip53. Cells were treated with ADR or left untreated and incubated for 24 h. The results represent the percentage of the TUNEL (terminal deoxynucleotidyl transferase dUTP nick end labeling)-positive cells. *P<0.01. (c) Apoptotic assay in PALMD-overexpressing cells. Empty vector as a control (green fluorescent protein (GFP)) and GFP-tagged PALMD (GFP-PALMD) were transfected into U2OS cells. Cells were left untreated or treated with ADR and incubated for 6 h. Apoptotic cells were analyzed with TUNEL-positive cells. *P<0.01. (d) The transfection efficiency of GFP and GFP-PALMD into U2OS cells. (e) Cell viability in PALMD-depleted cells. U2OS cells were seeded in 96-well plates and transfected with non-silencing siRNA or siPALMD-1 and treated with ADR. The MTS assay was performed after 24 h incubation. The data represent the mean±S.D. from three independent experiments, each performed in triplicates

Figure 4

Loss of PALMD causes necroptosis-like cell death in response to DNA damage through ATP depletion. (a) ATP concentration in PALMD-depleted cells. U2OS cells were transfected with non-silencing siRNA or siPALMD-1 (siPALMD) and treated or left untreated with ADR. The data represent the mean±S.D. from three independent experiments, each performed in triplicates. *P<0.01. (b, left) Cell death morphology in PALMD-overexpressing cells. U2OS cells were co-transfected with green fluorescent protein (GFP)-tagged PALMD and non-silencing siRNA and treated with ADR. The morphologies of the cell death were illustrated by time-lapse imaging and cell image was captured at 5-min intervals from treatment. Upper panels present illustration of the cells just after treatment (alive) and lower panels show cell death (dead). The nuclei were stained with Hoechst. The scale bars indicate 20 μm. (b, right) The transfection efficiency of GFP-PALMD and non-silencing siRNA was monitored. Cell lysates were immunoblotted with anti-PALMD, anti-GFP, or anti-tubulin. (c, left) Cell death morphology in PALMD-depleted cells. U2OS cells were co-transfected with GFP vector and siPALMD-1. Further procedure was performed as described in b. (d) PALMD-involved cell death was confirmed by fluorescence-activated cell sorting analysis. U2OS cells were transfected with non-silencing siRNA or siPALMD-1 (siPALMD), and treated or left untreated with ADR. Cells were double stained with PI and AnV. PI⁻AnV⁻ populations were defined as viable cells, PI⁻AnV⁺ populations were identified as apoptotic and PI-positive populations indicated as necrotic/necroptotic cells. (e) Percentage of viable (left panel), apoptotic (middle panel), or necrotic/necroptotic (right panel) cells from the data described in d was calculated. The data represent the mean±S.D. from three independent experiments. *P<0.01

Figure 5

Accumulation of PALMD is required to induce apoptosis. (a) PALMD localization by immunoflourescence in U2OS cells in normal condition (upper panels) and in stressed condition by ADR (lower panels). U2OS cells were seeded in four-well chamber slides and treated or left untreated with ADR. The nuclei were stained with DAPI. The scale bars indicate 20 μm. (b) PALMD localization by the subcellular fractionation assay. U2OS cells were treated or left untreated with ADR and fractionated as cytoplasmic and nucleic fractions. Both fractions were immunoblotted with anti-PALMD (top panel), anti-ORC2 (second panel), or anti-tubulin (bottom panel). (c) The map of the NLS construct. To design mutant NLS (NLS-mut), arginine and lysine indicated in red within NLS region of wild-type PALMD (wt-PALMD) were substituted with threonine and tagged to the green fluorescent protein (GFP) vector. (d) Subcellular fractionation assay of wt-PALMD and NLS-mut-transfected-cells. U2OS cells were transfected with wt-PALMD (wt) or NLS-mut. Cells were treated (ADR, right panels) or left untreated with ADR (control, left panels) and incubated for 24 h. Cell lysates were fractionated into cytoplasmic (Cyto) and nucleic (Nuc) fractions. Immunoblotting was performed using anti-GFP (top panel), anti-ORC2 (middle panel), or anti-tubulin (bottom panel). (e) Apoptotic assay of wt-PALMD or NLS-mut-transfected cells. U2OS cells were transfected with empty GFP vector (GFP), wt-PALMD (wt) or NLS-mut. Cells were treated with ADR and incubated for 6 h. The percentage of apoptotic cells was quantified with TUNEL (terminal deoxynucleotidyl transferase dUTP nick end labeling)-positive cells. (f) Transfection efficiency of GFP, wt-PALMD, or NLS-mut. (g) Cells undergo death with wt-PALMD or NLS-mut transfection when endogenous PALMD is depleted. The siRNA resistant-PALMD (si-wt-PALMD) or NLS-mut (si-NLS-mut) were co-transfected with siPALMD-1 and incubated for 24 h. The nuclei were illustrated with Hoechst. Cell death was defined with cell blebbings and apoptotic body (arrow), and chromatin condensation (arrow head). The scale bars indicate 40 μm (yellow) or 10 μm (white). The percentage of the cell death in GFP-positive cells is shown in the right panel. The data represent the mean±S.D. from three independent experiments, each performed in triplicates. *P<0.01

Figure 6

Proposed model of the molecular function of PALMD. In response to DNA damage, p53 is stabilized and activated by phosphorylation at ser46. Activated p53 promotes PALMD expression. Accumulated PALMD in the cytoplasm localizes into the nucleus in response to DNA damage to induce apoptosis. In the other side, if PALMD is not induced, cells die with necroptosis-like cell death through ATP depletion