Coactivation of ATM/ERK/NF-kappaB in the low-dose radiation-induced radioadaptive response in human skin keratinocytes

Abstract

Elucidating the molecular mechanism of the low-dose radiation (LDR)-mediated radioadaptive response is crucial for inventing potential therapeutic approaches to improving normal tissue protection in radiation therapy. ATM, a DNA-damage sensor, is known to activate the stress-sensitive transcription factor NF-kappaB upon exposure to ionizing radiation. This study provides evidence of the cooperative functions of ATM, ERK, and NF-kappaB in inducing a survival advantage through a radioadaptive response as a result of LDR treatment (10 cGy X-rays). By using p53-inhibited human skin keratinocytes, we show that phosphorylation of ATM, MEK, and ERK (but not JNK or p38) is enhanced along with a twofold increase in NF-kappaB luciferase activity at 24 h post-LDR. However, NF-kappaB reporter gene transactivation without a significant enhancement of p65 or p50 protein level suggests that NF-kappaB is activated as a rapid protein response via ATM without involving the transcriptional activation of NF-kappaB subunit genes. A direct interaction between ATM and NF-kappaB p65 is detected in the resting cells and this interaction is significantly increased with LDR treatment. Inhibition of ATM with caffeine, KU-55933, or siRNA or inhibition of the MEK/ERK pathway can block the LDR-induced NF-kappaB activation and eliminate the LDR-induced survival advantage. Altogether, these results suggest a p53-independent prosurvival network involving the coactivation of the ATM, MEK/ERK, and NF-kappaB pathways in LDR-treated human skin keratinocytes, which is absent from mutant IkappaB cells (HK18/mIkappaB), which fail to express NF-kappaB activity.

Fig. 1.

NF-κB is involved in LDR-induced radioadaptive response in human skin keratinocytes. (A) Human skin keratinocytes HK18 and (B) HK18/mIκB, the stable transfectants of HK18 cells containing mutant IκB, were irradiated with LDR (10 cGy X-rays) and then incubated for 6 h before exposure to a challenging dose of 2 Gy γ-rays. Cell radiosensitivity was determined by cell proliferation (MTT, left) and clonogenic survival (right) assays after radiation. Cells treated with sham LDR were included as control (n=3/group).

Fig. 2.

NF-κB activation without induction of p65 and p50 in LDR-treated HK18 cells. (A) HK18 and (B) HK18/mIκB cells were cotransfected with the NF-κB luciferase reporter and β-galactosidase reporter for 6 h, and NF-κB luciferase activity was measured at the indicated time points after exposure to sham or a single low dose of 10 cGy X-rays. Luciferase reporter activity was normalized to β-galactosidase (n=3/group). (C) Exponentially growing HK18 cells were irradiated with a single dose of 10 cGy X-rays and NF-κB p50 and p65 levels were measured by Western analysis (C, sham-LDR control; β-actin served as loading control).

Fig. 3.

ATM is required for LDR-induced NF-κB activation and radioadaptive response. (A) LDR-induced ATM phosphorylation (pATM). HK18 cells were irradiated with 10 cGy X-rays; total and phosphorylated ATM (Ser-1981) levels were detected by Western blotting (right shows densitometry data of the relative expression levels of pATM normalized to the expression levels of β-actin; C, sham-LDR control). (B) HK18 cells were cotransfected with NF-κB luciferase and β-galactosidase reporters for 6 h and then incubated with 5 mM caffeine (left) for 2 h or 10 μM KU-55933 (right) for 1 h. Luciferase activity was measured 24 h after exposure to sham-IR or irradiation with a single dose of 10 cGy X-rays and normalized to β-galactosidase (n=3). (C) HK18 cells were treated with caffeine, KU-55933, or ATM siRNA and then exposed to 2 Gy γ-irradiation with or without preexposure to 10 cGy X-rays. Radiosensitivity was determined by clonogenic survival (n=3; inset on the right shows a Western blot of HK18 cells treated with ATM siRNA; Ab, antibody).

Fig. 4.

MEK/ERK activation is required for LDR-induced NF-κB activity. (A) HK18 cells were irradiated with a single dose of 10 cGy X-rays; total and phosphorylated MEK and ERK levels were detected by Western blot at the indicated times. Sham-LDR cells (0 cGy) were collected at the same times as irradiated cells. Right shows densitometry results of p-MEK1/ 2 and p-ERK1/2 normalized to β-actin (C, sham-LDR control, β-actin as a loading control). (B) HK18 cells were treated as in (A), and total and phosphorylated p38 and JNK levels were detected by Western blot. (C) HK18 cells were cotransfected with NF-κB luciferase and β-galactosidase reporters for 6 h and then incubated with 50 μM PD98059 (top) or 10 μM U0126 (bottom) for 2 h, followed by exposure to sham-LDR or a single dose of 10 cGy X-rays. Luciferase reporter activity was measured at 24 h postirradiation and normalized to β-galactosidase (*P=0.02 compared to cells treated with control DMSO; n=3).

Fig. 5.

LDR-induced MEK/ERK phosphorylation was either eliminated by ATM inhibitors or absent from ATM-deficient GM05849 cells. (A) HK18 cells were treated with 5 mM ATM inhibitor caffeine for 2 h or (B) 10 μM KU-55933 for 1 h before exposure to 10 cGy X-rays (C, sham-LDR control). The expression levels of the indicated proteins were detected by Western analysis. No detectable ATM phosphorylation was found in either ATM inhibitor-treated cells (A and B) or (C) the ATM-deficient GM05849 cell line.

Fig. 6.

LDR increased ATM/NF-κB p65 complex formation. HK18 and HK18/mIκB cells were irradiated with sham (0 cGy) or 10 cGy X-rays. Whole-cell lysates were immunoprecipitated (IP) with anti-ATM antibody followed by immunoblotting (IB) with p65 or ATM antibody. Total lysates extracted from cells treated with 10 cGy and preincubated with ATM antibody served as a negative (Neg.) control.