Elevated ornithine decarboxylase levels activate ataxia telangiectasia mutated-DNA damage signaling in normal keratinocytes

Abstract

We examined the effect of increased expression of ornithine decarboxylase (ODC), a key rate-limiting enzyme in polyamine biosynthesis, on cell survival in primary cultures of keratinocytes isolated from the skin of K6/ODC transgenic mice (Ker/ODC) and their normal littermates (Ker/Norm). Although elevated levels of ODC and polyamines stimulate proliferation of keratinocytes, Ker/ODC undergo apoptotic cell death within days of primary culture unlike Ker/Norm that continue to proliferate. Phosphorylation of ataxia telangiectasia mutated (ATM) and its substrate p53 are significantly induced both in Ker/ODC and in K6/ODC transgenic skin. Chromatin immunoprecipitation analyses show that the increased level of p53 in Ker/ODC is accompanied by increased recruitment of p53 to the Bax proximal promoter. ATM activation is polyamine dependent because alpha-difluoromethylornithine, a specific inhibitor of ODC activity, blocks its phosphorylation. Ker/ODC also displays increased generation of H(2)O(2), acrolein-lysine conjugates, and protein oxidation products as well as polyamine-dependent DNA damage, as measured by the comet assay and the expression of the phosphorylated form of the histone variant gamma H2AX. Both reactive oxygen species generation and apoptotic cell death of Ker/ODC may, at least in part, be due to induction of a polyamine catabolic pathway that generates both H(2)O(2) and cytotoxic aldehydes, because spermine oxidase (SMO) levels are induced in Ker/ODC. In addition, treatment with MDL 72,527, an inhibitor of SMO, blocks the production of H(2)O(2) and increases the survival of Ker/ODC. These results show a novel activation of the ATM-DNA damage signaling pathway in response to increased ODC activity in nontumorigenic keratinocytes.

Figure 1. Polyamine-dependent p53 protein accumulation and apoptotic cell death in ODC-overexpressing primary keratinocytes

Primary keratinocytes were isolated from K6/ODC transgenic mice (T) and their normal littermates (N) and plated. Some of the K6/ODC keratinocytes were treated with 3 mM DFMO. A) After 5 days in culture, cells were harvested in RIPA buffer and equal amounts of protein were resolved by SDS-PAGE and transferred to PVDF membranes. PCNA, p53, p21^Waf1, Bax, and Bcl-2 were detected by immunoblotting and the membranes were reprobed for tubulin protein as a loading control. B) Representative photomicrographs of Ker/Norm and Ker/ODC following 6 days in culture. C) Keratinocytes were stained immunohistochemically using an anti-caspase 3 (active form) antibody. The values for the percent caspase 3-stained keratinocytes are the percent of caspase 3-stained cells out of the total cells per coverslip. At least 1000 total cells were counted per group. D) The Bax promoter was assessed by ChIP analysis for p53 binding in crosslinked protein-DNA complexes precipitated from Ker/Norm, Ker/ODC and Ker/ODC + DFMO treated keratinocytes using an anti-p53 antibody or rabbit IgG as a control. PCR was performed with primers specific for the Bax promoter which included the p53 binding site. To verify equal amounts of starting material, PCR was performed on aliquots of the chromatin removed prior to immunoprecipitation (input). Amplified DNA was visualized by agarose gel electrophoresis.

Figure 2. ODC overexpression activates ATM in primary keratinocytes and in K6/ODC transgenic skin

A) Primary keratinocytes were isolated from K6/ODC transgenic mice (Ker/ODC) and their normal littermates (Ker/Norm) and plated. Some of the Ker/ODC were treated with 3 mM DFMO. Nuclear extracts and RIPA lysates were prepared and equal amounts of protein were resolved by SDS-PAGE and transferred to PVDF membranes. Nuclear extracts were probed for phospho-p53 (Ser15) and lamin B. Total cellular lysates were probed for ATM, γH2AX, and p19^Arf by immunoblotting and then reprobed for tubulin protein as a loading control. B) Immunofluorescence of normal littermate skin and K6/ODC transgenic mouse skin using an antibody specific for ATM phosphorylated at serine-1981 (pATM) in the presence or absence of a peptide for ATM phosphorylated at serine-1981. DAPI was used for the same views to identify all nuclei. C) Normal and K6/ODC skins were resolved by SDS-PAGE, transferred to PVDF and probed for phospho-p53 (Ser15) and β-actin by immunoblotting.

Figure 3. DNA damage induced by elevated ODC activity in primary keratinocytes

Primary keratinocytes were isolated from K6/ODC transgenic mice (Ker/ODC) and their normal littermates (Ker/Norm) and plated. Some of the Ker/ODC were treated with 3 mM DFMO. Following 5 days in culture, keratinocytes were harvested, embedded in low-melting point agarose, and processed for the alkaline comet assay as described in Material and Methods. Each slide was stained with SYBR Green, photographed using a fluorescent microscope, and the average tail lengths ± SD scored and analyzed using Cometscore software to determine the average olive moments ± SD. *p < 0.0001, significantly different compared with Ker/Norm or with DFMO-treated Ker/ODC.

Figure 4. Elevated ODC activity induces spermine oxidase

A) After 3 days in culture, Ker/Norm, Ker/ODC, and Ker/ODC treated with 3 mM DFMO were harvested. Immunoblot analysis of total cellular protein lysates in which blots were probed with an antibody specific for SMO and reprobed for tubulin as a control. B) RNA was prepared from Ker/Norm and Ker/ODC three days after isolation from the skin of K6/ODC transgenic mice and their normal littermates. Reverse transcription-PCR was performed to analyze levels of SMO, SSAT, APAO, and AdoMetDC at the mRNA level. GAPDH mRNA was simultaneously monitored as an internal PCR control for each reaction. C) HPLC analyses of total intracellular polyamine levels in Ker/Norm and Ker/ODC harvested three days in culture. Polyamines (Put, putrescine; Spd, spermidine; Spm, spermine) are expressed in nmol/mg DNA ± SD.

Figure 5. Elevated ODC activity leads to increased ROS generation

A) The levels of H₂O₂ production in Ker/Norm and Ker/ODC were compared using flow cytometry. Four days following culture, cells were incubated with 10 μM CM-H₂DCFdA at 37°C for 25 min, and fluorescence was measured using a BD FACS Canto II flow cytometer. X axis, fluorescent intensity; Y axis, cell number. Representative results of four experiments that gave similar results. B) Four days following culture, Ker/Norm, Ker/ODC, and Ker/ODC treated with 3 mM DFMO were harvested by freeze-thaw in PBS and nmol acrolein-lysine conjugates/mg protein ± SD was determined by direct ELISA assay. C) Relative level of protein carbonyl groups in Ker/Norm, Ker/ODC and Ker/ODC treated with 3 mM DFMO was measured using an Oxyblot protein oxidation kit after proteins were resolved by SDS-PAGE and transferred to PVDF.

Figure 6. DFMO and MDL72,527 prevent H₂O₂ generation in primary keratinocytes and increase survival of Ker/ODC

Primary keratinocytes were isolated from K6/ODC transgenic mice and their normal littermates as well as from K6/ODC transgenic mice and normal littermates bred to a p53^−/− background. Some Ker/ODC were cultured with 3 mM DFMO or 25 μM MDL 72,527 in the medium. A) The levels of H₂O₂ production in Ker/Norm (shaded curve) and Ker/ODC (open curve) treated with and without 3 mM DFMO and 25 μM MDL 72,527 were compared using flow cytometry. Four days following culture, cells were incubated with 10 μM CM-H₂DCFdA at 37°C for 25 min, and fluorescence was measured using a BD FACS Canto II flow cytometer. X axis, fluorescent intensity; Y axis, cell number. B) Representative phase contrast images were taken of the keratinocytes after 14 days in culture to capture the difference in survival between Ker/Norm and Ker/ODC treated with and without 3 mM DFMO and 25 μM MDL 72,527. Whereas no viable Ker/ODC, p53^+/+ remained by 7 days in culture, Ker/ODC, p53^−/− were still proliferating after 14 days of culture.