Interaction between ATM and PARP-1 in response to DNA damage and sensitization of ATM deficient cells through PARP inhibition

Abstract

ATM and PARP-1 are two of the most important players in the cell's response to DNA damage. PARP-1 and ATM recognize and bound to both single and double strand DNA breaks in response to different triggers. Here we report that ATM and PARP-1 form a molecular complex in vivo in undamaged cells and this association increases after gamma-irradiation. ATM is also modified by PARP-1 during DNA damage. We have also evaluated the impact of PARP-1 absence or inhibition on ATM-kinase activity and have found that while PARP-1 deficient cells display a defective ATM-kinase activity and reduced gamma-H2AX foci formation in response to gamma-irradiation, PARP inhibition on itself is able to activate ATM-kinase. PARP inhibition induced gamma H2AX foci accumulation, in an ATM-dependent manner. Inhibition of PARP also induces DNA double strand breaks which were dependent on the presence of ATM. As consequence ATM deficient cells display an increased sensitivity to PARP inhibition. In summary our results show that while PARP-1 is needed in the response of ATM to gamma irradiation, the inhibition of PARP induces DNA double strand breaks (which are resolved in and ATM-dependent pathway) and activates ATM kinase.

Figure 1

ATM form a complex with PARP-1 in DNA that is much more evident after treatment with DNA damaging agents. A: ATM was immunoprecipitated in human melanoma cell lines G361 and HT144 (ATM deficient) as explained under methods and the presence of PARP-1 was tested by immunoblot analysis. Cells were treated with 10 Gy IR or 2 mM MNU for 30 min. In the left panel, PARP-1 was immunoprecipitated from G361 cells treated or not with 2 mM MNU and western blot was performed to reveal ATM. B: Double indirect immunofluorescence in 3T3 fibroblasts (parp-1+/+)of PARP-1 (red signal) and ATM (green). Yellow signal correspond with co-localization of both proteins.

Figure 2

ATM is modified by poly(ADP)ribose. A: ATM was immunoprecipitated in G361 cells and the presence of poly(ADP ribose) was tested by immunoblot analysis in a time course, reaching a maximum after 15 min of IR (10 Gy). Equal loading was normalized by the input of ATM. B: Double indirect immunofluorescence in 3T3 fibroblasts (parp-1+/+) of poly(ADP-ribose) (red signal) and ATM (green). Yellow signal correspond with poly(ADP-ribosyl)ation of ATM. C: PARP-1 is needed for optimal activation of ATM. In vitro ATM-kinase assay. ATM activity increases after γ-irradiation in wild-type mouse splenocytes, but not in parp-1 knockout mouse splenocytes, where ATM is activated in control (upper panels). In G361 cells and C3ABR (middle and lower panel) irradiated with or without ANI co-treatment, ATM kinase activity increases after γ-irradiation and/or PARP inhibitors. Equal loading was checked with coomasie blue staining. Normalized signal respect to coomassie blue is shown below. Results are representative of three independent experiments.

Figure 3

ATM is activated by PARP inhibitors. Immunostaining for γ-H2AX in G361 (ATM wild type) and HT144 (ATM deficient) cells exposed to γ-irradiation (positive control) or ANI (time course after drug exposure). H2AX (green signal) is phophorilated by ATM after ANI treatment in absence of γ-irradiation in G361 cells (ATM-proficient). DAPI is shown in blue and both views are in coincidence.

Figure 4

PARP inhibitors activate DSB repair by homologous recombination. A: neutral comet assay for detection of double strand breaks. For scoring the comet pattern, 800 nuclei from each slide were counted. HT144 ATM-deficient cells display more DSBs than G361 ATM-proficient cells after ANI treatment and only ATM wild type cells were able to completely resolve DSBs. Experiment is one representative of three similars. B: Analysis of cell death. Sub-G1 analysis was performed by flow cytometry using the propidium iodide (PI) DNA-staining method. Increased cell death with ANI is observed after long times of exposure (72 hours) in ATM-deficient cells (HT144, grey bars and ATM-proficient cells, G361, white bars) treated either with ANI alone or in combination with γ-irradiation respect ATM-proficient cells. Results represent the average ± SEM of three independent observations. * p < 0,05; **p < 0,01 respect to untreated HT144.