High glucose dephosphorylates serine 46 and inhibits p53 apoptotic activity

Abstract

Background: In response to diverse genotoxic stimuli p53 is activated as transcription factor to exert its tumor-suppressor function. P53 activation requires protein stabilization, nuclear localization and posttranslational modifications in key residues that may influence p53 selection of target genes. Among them, serine 46 (Ser46) phosphorylation is considered specific for apoptotic activation. Hyperglicaemia, the high blood glucose condition, may negatively affect tumor response to therapies through several mechanisms, conferring resistance to drug-induced cell death. However, whether high glucose might modify p53Ser46 phosphorylation has never been addressed.

Methods and results: Here, we performed biochemical and molecular analyses in different cancer cell lines treated with chemotherapy in the presence or absence of high glucose condition. Analyses of p53 posttranslational modifications showed that drug-induced p53Ser46 phosphorylation was reduced by high glucose. Such reduction depended by high glucose-induced calyculin A-sensitive phosphatase(s), able to specifically target p53Ser46 phosphorylation. The specific effect on Ser46 phosphorylation was addressed by analysing Ser15 phosphorylation that instead was not modified by high glucose. In agreement, a constitutively phosphorylated Ser46D p53 mutant was resistant to high glucose. As a consequence of phosphoSer46 impairment, high glucose reduced the tumor cell response to drugs, correlating with reduced p53 apoptotic transactivation. The drug-induced apoptotic cell death, reduced by high glucose, was finally restored by the phosphatase inhibitor calyculin A.

Conclusions: These data indicate that high glucose specifically inhibited Ser46 phosphorylation thus reducing p53 apoptotic activity. These results uncover a new mechanism of p53 inactivation providing an interesting novel molecular link between metabolic diseases such as diabetes or obesity and tumor progression and resistance to therapies.

Figure 1

High glucose (HG) culture condition reduces the cytotoxic effects of drugs in wtp53-carrying tumor cells. (A) Cells were plated at subconfluence in culture media containing 10% FBS and 1 g/L D-glucose. The day after, medium was changed with a medium containing 2% FBS with either 1 g/L D-glucose or 4.5 g/L D-glucose, respectively low and high glucose (HG), for 24 h before adding chemotherapeutic drugs Adriamycin (ADR) (to RKO, HCT116 and HCT116-p53^-/- cells) at 2 μg/ml or cisplatin (CDDP) (to A549 and 2008 cells) at 5 μg/ml. Twenty-four hours later, the percentage of dead cells was scored by trypan blue staining. Error bars show standard deviation. *P = 0.002. (B) HCT116 cells were treated with ADR (2 μg/ml) with or without HG culture condition. Equal amount of total cell extracts was analysed by western immunoblotting with anti-PARP antibody specific for the cleaved form (cleav). Anti-β-actin was used as protein loading control.

Figure 2

High glucose (HG) culture condition reduces the p53/DNA binding and transactivation functions mainly on apoptotic genes. (A) ChIP analyses performed with anti-p53 antibody on HCT116 cells treated with ADR (2 μg/ml) in 2% FBS culture medium with or without HG. PCR analyses were performed on the immunoprecipitated DNA samples using specific primers for the p53 target promoters. A sample representing linear amplification of the total chromatin (Input) was included as control. Additional controls included immunoprecipitation performed with nonspecific immunoglobulins (No Ab). (B) H1299 cells were transiently co-transfected with PG13-luc or p53AIP1-luc reporters and wtp53 plasmid. Twenty-four hours after transfection culture medium was changed with a medium containing 2% FBS with or without HG. Results, normalized to β-gal activity are the mean ± S.D. of three independent experiments performed in duplicate. *P = 0.001. (C) RNA samples were extracted by HCT116 cells treated as in Figure 1 and used for RT-PCR. The mRNA level of specific p53 target genes was analysed by densitometry and plotted as the expression ratio to control 28S expression. Data are the mean ± S.D. of two independent experiments. *P = 0.001.

Figure 3

High glucose (HG) culture condition does not affect drug-induced p53 stability and nuclear localization, rather it inhibits p53Ser46 phosphorylation. (A) HCT116 and A549 cells were treated with ADR (2 μg/ml) or CDDP (5 μg/ml), respectively, with or without HG culture condition. Equal amount of nuclear and cytoplasmic extracts was analysed by western immunoblotting with specific anti-p53 antibody. Anti-lamin A and anti-tubulin antibodies were used as control of efficient nuclear and cytoplasmic extraction, respectively. (B) HCT116, 2008 and A549 cells were treated with ADR or CDDP, as shown, with or without HG. Equal amount of total cell extracts was analysed by western immunboltting with phospho-specific Ser46 and Ser15 antibodies and total p53 antibody. Anti-β-actin was used as protein loading control. (C) H1299 cells were transiently co-transfected with p53AIP1-luc reporter and wtp53 or Ser46D mutant plasmids. Twenty-four hours after transfection culture medium was changed with a medium containing 2% FBS with or without HG. Results, normalized to β-gal activity are the mean ± S.D. of three independent experiments performed in duplicate. *P = 0.001; NSS: not statistically significant.

Figure 4

High glucose (HG) induces calyculin A-sensitive phospahatase(s) able to target p-Ser46. (A) HCT116, cells were treated with ADR with or without HG; phosphatase inhibitor calyculin A was added at 1 nM along with ADR. Equal amount of total cell extracts was analysed by western immunboltting with phospho-specific Ser46 antibody and total p53 antibody. Anti-β-actin was used as protein loading control. (B) RNA samples were extracted by HCT116 cells treated as in (A) and used for RT-PCR of p53 target gene expression. 28S levels were analyzed as internal control (left panel). Densitometric analysis is shown in the right panel and plotted as the expression ratio to control 28S expression. (C) Tunel assay of HCT116 and 2008 cells treated with ADR and CDDP, respectively, with or without HG condition. Phosphatase inhibitor calyculin A was added at 1 nM along with drugs. The results shown are the percentage of Tunel positive cells and are representative of two independent experiments performed in duplicates, ± S.D. *P = 0.001.