Positive feedback regulation of p53 transactivity by DNA damage-induced ISG15 modification

Abstract

p53 plays a pivotal role in tumour suppression under stresses, such as DNA damage. ISG15 has been implicated in the control of tumorigenesis. Intriguingly, the expression of ISG15, UBE1L and UBCH8 is induced by DNA-damaging agents, such as ultraviolet and doxorubicin, which are known to induce p53. Here, we show that the genes encoding ISG15, UBE1L, UBCH8 and EFP, have the p53-responsive elements and their expression is induced in a p53-dependent fashion under DNA damage conditions. Furthermore, DNA damage induces ISG15 conjugation to p53 and this modification markedly enhances the binding of p53 to the promoters of its target genes (for example, CDKN1 and BAX) as well as of its own gene by promoting phosphorylation and acetylation, leading to suppression of cell growth and tumorigenesis. These findings establish a novel feedback circuit between p53 and ISG15-conjugating system for positive regulation of the tumour suppressive function of p53 under DNA damage conditions.

Figure 1. ATM/ATR kinases induce the expression of ISG15-conjugating system.

p53^+/+ HCT116 cells were treated with 1 μM doxorubicin (DOX) (a), 0.25 μM camptothecin (CPT) (b) and 30 J m⁻² ultraviolet (UV) (c). Immediately after the treatment, the cells were incubated with 5 mM caffeine for increasing periods. The cells were washed with iced PBS, and lysed in 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1 mM NEM, 1 mM sodium orthovanadate, 1 mM NaF, 1 mM PMSF, 0.2% (v/v) Triton X-100 and 1 X protease inhibitor cocktail. The samples were centrifuged for 30 min at 16,000g and the supernatant fractions (cell lysates) were subjected to immunoblot with anti-ISG15, anti-UBE1L anti-UBCH8, anti-EFP, anti-phospho-p53 (p-p53), anti-p53, anti-phospho-Chk1 (p-Chk1) and anti-β-actin antibodies (from the top to the bottom, respectively).

Figure 2. p53REs in the promoter regions of the genes for ISG15-conjugating system.

Deletions of the promoter regions in the ISG15 (a), UBE1L (b), UBCH8 (c) and EFP genes (d) were fused to a Luc reporter vector (pGL3) (left panels). The resulting vectors (P1–P4) were transfected to p53^−/− HCT116 cells with pcDNA-HA-p53. p53^+/+ HCT116 cells transfected with the same P1–P4 vectors were treated with and without ultraviolet (UV). p21-Luc was expressed in the cells as a control. These cells were then subjected to assay for the luciferase activity (right panels). The activity seen without any expression or treatment was expressed as 1.0 and the others were expressed as its relative values. Error bar, ±s.d. (n=3).

Figure 3. p53-induced expression of ISG15-conjugating system is independent of type-I IFNs.

In P1 reporter vectors (see Fig. 2), RE3 of the ISG15 gene (a), RE1/RE2 of UBE1L (b), RE1 of UBCH8 (c) and RE1/RE2 of EFP (d) were mutated as shown in Supplementary Fig. 5 and the mutated REs were marked by the asterisks. ISREs of the genes were also mutated and marked by the asterisks. After transfection of the vectors, p53 were expressed in p53^−/− HCT116 cells or 1,000 U ml⁻¹ of interferon-α (IFNα) was treated to the cells (left panels). p53^+/+ HCT116 cells transfected with the same reporter vectors were treated with and without IFNα or ultraviolet (UV; right panels). The cells were then subjected to assay for the luciferase activity. The activity seen without any expression or treatment was expressed as 1.0 and the others were expressed as its relative values. Error bar, ±s.d. (n=3).

Figure 4. DNA damage induces p53 ISGylation.

(a) HA-p53 and Flag-ISG15, Myc-UBE1L (E1) and Myc-UBCH8 (E2) were expressed in HEK293T cells with or without Flag-UBP43. The cell lysates were subjected to immunoprecipitation (IP) with anti-HA antibody followed by immunoblot with anti-Flag or anti-HA antibody. They (lysates) were also directly probed with anti-HA, anti-Flag or anti-Myc antibody. (b) p53^+/+ HCT116 cells that had been transfected with shNS or shISG15 were treated with doxorubicin (DOX) or camptothecin (CPT) for 24 h. They were also irradiated with ultraviolet (UV), and then incubated for 24 h. The cell lysates were subjected to immunoprecipitation with anti-p53 antibody followed by immunoblot with anti-ISG15 antibody. They were also directly probed with respective antibodies. (c) Deletions of p53 (pΔ1–pΔ4) were tagged with HisMax to their N-termini, and expressed in HEK293T cells with Flag-ISG15, Myc-UBE1L and Myc-UBCH8. The cell lysates were subjected to pull-down with NTA resins followed by immunoblot with anti-Flag antibody. (d) Wild-type p53 (Wt) or its K-to-R mutants were expressed in HEK293T cells with Flag-ISG15, Myc-UBE1L and Myc-UBCH8. The cell lysates were subjected to immunoprecipitation with anti-Xpress antibody followed by immunoblot with anti-Flag or anti-Xpress antibody. (e) HCT116 (p53^+/+) cells were transfected with shNS or shISG15. After exposure to ultraviolet, the cells were subjected to incubation with 0.2 mg ml⁻¹ cycloheximide (CHX) for increasing periods followed by immunoblot analysis. (f) Experiments in e were repeated and the band intensities were scanned by using a densitometer and normalized by those of GAPDH. The normalized densities seen at ‘0' time points were expressed as 1.0 and the others were expressed as its relative values. Error bar, ±s.d. (n=3).

Figure 5. EFP serves as a p53-specific ISG15 E3 ligase.

(a) HA-EFP or its inactive C13/16S mutant was expressed in HEK293T cells with HisMax-p53. The cell lysates were subjected to immunoprecipitation with anti-HA antibody or pull-down with NTA resins followed by immunoblot with anti-Xpress or anti-HA antibody. The asterisk indicates IgG heavy chain. (b) HisMax-p53, Flag-ISG15, Myc-UBE1L (E1) and Myc-UBCH8 (E2) were expressed in HEK293T cells with Myc-tagged EFP or its C13/16S mutant. The cell lysates were subjected to immunoprecipitation with anti-Xpress antibody followed by immunoblot anti-Flag or anti-Xpress antibody. (c) p53^+/+ HCT116 cells that had been transfected with shNS or shEFP were treated with doxorubicin or camptothecin for 24 h. They were also treated with ultraviolet (UV), and then incubated for 24 h. The cell lysates were subjected to immunoprecipitation with anti-p53 antibody followed by immunoblot with anti-ISG15 antibody. (d) Myc-HERC5 was expressed in HEK293T cells with HisMax-p53. The cell lysates were subjected to pull-down with NTA resins followed by immunoblot with anti-Myc or anti-Xpress antibody. (e) After treatment with DNA-damaging agents, the HCT116 cells were subjected to incubation for 24 h followed by immunoblot with anti-HERC5, anti-EFP or anti-p53 antibody. (f) p53^+/+ HCT116 cells expressing shNS or shHERC5 were exposed to ultraviolet. After incubation for 24 h, the cells were subjected to immunoprecipitation with anti-p53 followed by immunoblot with anti-ISG15 antibody.

Figure 6. ISGylation promotes p53 transactivity.

(a) Wild-type p53 (Wt), its 2KR mutant, and an empty vector (Mock) were expressed in H1299 cells with PG13-Luc, p21-Luc and BAX-Luc. After exposure to ultraviolet (UV), they were incubated for various periods, and subjected to assay for the luciferase activity. Transfection efficiency was normalized by using β-galactosidase constructs. The luciferase activity seen without any treatment (that is, ‘0' time in Mock) was expressed as 1.0 and the others were as its relative values. Error bar, ±s.d. (n=3). (b) shNS, shISG15 or shEFP was expressed in p53^+/+ HCT116 cells with and without shRNA-insensitive Flag-ISG15 or Myc-EFP. After exposure to ultraviolet, the cells were incubated for 24 h and subjected to the assay for the luciferase activity as in a. Error bar, ±s.d. (n=3). (c) HisMax-p53 and ISG15-conjugating system (Es/Flag-ISG15) were expressed in p53^−/− HCT116 cells with and without Flag-UBP43. The cell lysates were subjected to ChIP assay using anti-p53 antibody or anti-mouse IgG. Bound DNAs were subjected to PCR using the probes for p53REs of CDKN1, MDM2, BAX and ISG15. (d) HisMax-p53 and HisMax-2KR were expressed in p53^−/− HCT116 cells with or without ISG15-conjugating system. The cell lysates were subjected to ChIP as in c. (e) shNS or shISG15 were expressed in p53^+/+ HCT116 cells with and without shRNA-insensitive Flag-ISG15. After exposure to ultraviolet, the cells were incubated for 24 h, and then subjected to ChIP as in c. Note that shISG15 was directed to a 5′-UTR region. a, b and e, similar results were obtained when the cells were treated with doxorubicin or camptothecin in place of ultraviolet.

Figure 7. p53 ISGylation suppresses cell growth and tumour development.

(a) p53^−/− HCT116 cells expressing wild-type p53 (Wt), its 2KR mutant, or an empty vector (Mock) were transfected with shNS or shISG15. After treatment with doxorubicin for increasing periods, viable cells were counted by staining with trypan blue exclusion. Error bar, ±s.d. (n=3). (b) p53^−/− HCT116 cells prepared as in a were incubated with or without doxorubicin for 10 days. The colonies were then stained with crystal violet. (c) Experiments were repeated as in b and the colonies were counted. Error bar, ±s.d. (n=3). (d) shNS or shISG15 were expressed in both p53^+/+ and p53^−/− HCT116 cells with and without shRNA-insensitive Flag-ISG15. After treatment with doxorubicin, experiments were then performed as in b. Error bar, ±s.d. (n=3). (e) p53^+/+ HCT116 cells expressing shNS or shISG15 (top) and p53^−/− HCT116 cells expressing wild-type p53 or its 2KR mutant (bottom) were injected to BALB/c nude mice. ‘Mock' in the bottom panel indicates the cells transfected with an empty vector. After treatment with PBS or doxorubicin, tumour volumes were determined. Error bar, ±s.d. (n=5). (f) A model for positive feedback regulation of p53 transactivity by ISGylation under DNA damage conditions. ‘Ac' and ‘P' indicate acetylated and phosphorylated, respectively.