Deregulated BCL-2 family proteins impact on repair of DNA double-strand breaks and are targets to overcome radioresistance in lung cancer

Abstract

Purpose: DNA damage-induced cell death is a major effector mechanism of radiotherapy. Aberrant expression of anti-apoptotic BCL-2 family proteins is frequently observed in lung cancers. Against this background, we studied radioresistance mediated by BCL-2 family proteins at the mechanistic level and its potential as target for radiochemotherapy.

Methods: Lung cancer models stably expressing BCL-xL or MCL-1 were irradiated to study cell death, clonogenic survival, and DNA repair kinetics in vitro, and growth suppression of established tumors in vivo. Additionally, endogenous BCL-xL and MCL-1 were targeted by shRNA or pharmacologic agents prior to irradiation.

Results: Radiation exposure induced apoptosis at negligible levels. Yet, anti-apoptotic BCL-xL and MCL-1 expression conferred short-term and long-term radioresistance in vitro and in vivo. Radioresistance correlated with pertubations in homologous recombination repair and repair of DNA double-strand breaks by error-prone, alternative end-joining. Notably, genetic or pharmacologic targeting of BCL-xL or MCL-1 effectively sensitized lung cancer cells to radiotherapy.

Conclusions: In addition to directly suppressing apoptosis, BCL-2 family proteins confer long-term survival benefits to irradiated cancer cells associated with utilization of error-prone repair pathways. Targeting BCL-xL and MCL-1 is an attractive strategy for improving lung cancer radiotherapy.

Keywords: Alternative end-joining (alt-EJ); BCL-2 family; DNA double-strand break (DSB) repair; Homologous recombination repair (HRR); Lung cancer; Radioresistance.

Fig. 1

Enforced expression of BCL-xL and MCL-1 mediates radioresistance in vitro. Expression levels of A431 cells stably overexpressing BCL-xL (a) or MCL-1 (b) compared to parental cells (P) evaluated by Western Blotting, actin served as loading control. Irradiation-induced cell death of parental A431 cells, A431 cells containing an empty vector control (vector ctrl.) and A431 cells stably overexpressing BCL-xL (c) or MCL-1 (d). Cells were stained with propidium iodide (PI) 72 h after irradiation (0–10 Gy) and the fraction of PI-positive cells was determined by flow cytometry (mean ± SD, n = 3). Long-term survival of BCL-xL (e) or MCL-1 (f) overexpressing and control cells measured by colony formation after irradiation with different doses (mean ± SD, n = 3). *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 (compared to parental/ctrl. cells treated with the same dose, Two-way ANOVA)

Fig. 2

Enforced expression of BCL-xL mediates radioresistance in vivo. Relative impact of irradiation on tumor growth of BCL-xL expressing A431 xenografts compared to control. NOD/SCID mice received subcutaneous injections of A431 cells stably overexpressing BCL-xL (BCL-xL) or an empty vector control (Ctrl.). Established tumors were irradiated with 4 Gy. Tumor growth was measured using a caliper and the impact of irradiation on relative tumor growth was calculated. BCL-xL expression in tumors was assessed by Western Blotting, actin served as a loading control. *p ≤ 0.05 (compared to ctrl. cells. Two-way ANOVA)

Fig. 3

Impact of increased BCL-xL expression on cellular repair kinetics. a Dose response estimation in A431 control cells and A431 cells with enforced BCL-xL expression was measured by PFGE. Left diagram Image of representative ethidium bromide stained gel, which was analyzed to estimate the fraction of DNA released (FDR) from the well into the lane after irradiation (regions defined as indicated). Right diagram Dose response curves for the induction of DSBs in cells exposed to different doses of IR. b Repair kinetics of IR-induced DSBs after irradiation with 20 Gy. FDR at each time point after irradiation was converted to DEQ (dose equivalent) with the help of dose response curves, as exemplarily shown in A (mean ± SD). c Repair kinetics of IR-induced DSBs in A431 control cells or A431 cells expressing BCL-xL was measured in the presence or absence of the DNA-PK inhibitor NU7441. Cells were pretreated with 10 µM NU7441 for 1 h prior to irradiation with 20 Gy. FDR of each time point was converted in DEQ using dose response curves (mean ± standard error, n = 2 with 4 samples each). d Long-term survival of A431 cells after combined treatment with irradiation and NU7441. A431 cells stably expressing either vector control (Ctrl.) or BCL-xL vector (BCL-xL) were pretreated with 2 µM NU7441 or DMSO for 1 h before irradiation with different doses (0–10 Gy). After 24 h, the inhibitor was removed and colonies were grown for 11 days (compare Online Resource 3 for indicators of significance). e Long-term survival of A431 cells after combined treatment with irradiation and the RAD51 inhibitor B02. A431 control and BCL-xL cells were pretreated with 50 µM B02 or DMSO for 1 h before irradiation with different doses (0–10 Gy). After 5–7 h, B02 was removed and colonies were grown for 10–11 days (compare Online Resource 3 for indicators of significance). f Chromosomal aberrations in A431 cells stably expressing either control vector (Ctrl.) or BCL-xL (BCL-xL) in the presence (right diagram) or absence (left diagram) of NU7441. Cells were pretreated with 5 µM NU7441 for 1 h, irradiated with 1 Gy and fixed after 8 and 20 h. At least 50 metaphases of each sample were analyzed for chromosomal aberrations (chromatid breaks, sister unions, dicentrics, translocations, acentric fragments). ***p ≤ 0.001 (compared to ctrl. cells of the same datapoint, Two-way ANOVA)

Fig. 4

Targeting BCL-xL and MCL-1 to sensitize lung cancer cells to radiation therapy. a Irradiation-induced cell death of tet-on BAK cells in the presence or absence of doxycycline. A549 cells stably expressing a conditional BAK transgene (tet-on BAK) were treated with doxycycline (100 ng/ml). Cells were irradiated with 0 Gy (sham irradiation) or 10 Gy 24 h after BAK induction. Cell death was measured 72 h after irradiation by determining the cell fraction with subgenomic DNA content using flow cytometry (mean ± SD, n = 4). b Long-term survival of tet-on BAK cells after irradiation with indicated doses (mean ± SD, n = 3). Irradiation-induced cell death of A549 cells in the presence or absence of the BH3-mimetics ABT-737 (c) or ABT-263 (e). A549 cells were pretreated with indicated concentrations of ABT-737 or ABT-263 for 1 h prior to irradiation with 0 Gy (sham irradiation) or 10 Gy. Cell death was determined 72 h after irradiation by quantifying the cell fraction with subgenomic DNA content using flow cytometry (mean ± SD, n = 3). Long-term survival of A549 cells treated with 12.5 µM ABT-737 (d) or 1 µM ABT-263 (f) and different doses of irradiation (0–10 Gy). ABT-737 was washed out 24 h after irradiation. ABT-263-treated cells were seeded for clonogenic survival in medium without ABT-263 24 h after irradiation. *p ≤ 0.05, ***p ≤ 0.001 (compared to non-irradiated controls/ctrl. cells treated with the same dose, Two-way ANOVA)

Fig. 5

Co-targeting of BCL-xL and MCL-1 as radiosensitization strategy. BCL-xL (a) and MCL-1 (b) protein levels in parental A549 cells (P), A549 cells stably expressing a lentiviral shRNA expression vector targeting BCL-xL (shRNA BCL-xL) or MCL-1 (shRNA MCL-1) or a control shRNA vector (shRNA scr.). Actin served as a loading control. Irradiation-induced cell death of control cells (parental, shRNA scr.) and BCL-xL (c) or MCL-1 (d) knockdown cells (shRNA BCL-xL, shRNA MCL-1) 72 h after irradiation with 0 Gy (sham irradiation) or 10 Gy. Cell death was determined by staining with PI followed by flow cytometry analysis (mean ± SD, n = 2). e, f Long-term survival of BCL-xL or MCL-1 knockdown cells (shRNA BCL-xL, shRNA MCL-1) compared to control cells (shRNA scr.) measured by colony formation after irradiation with different doses (mean ± SD, n = 3). g A549 cells stably expressing a lentiviral shRNA expression vector targeting BCL-xL (shRNA BCL-xL) or a control shRNA vector (shRNA scr.) were pretreated with 100 nM flavopiridol for 3 h followed by irradiation with 0 Gy (sham irradiation) or 10 Gy. Cell death was determined 72 h after irradiation by measuring the cell fraction with subgenomic DNA content using flow cytometry (mean ± SD, n = 4). h A549 cells stably expressing a lentiviral shRNA expression vector targeting MCL-1 (shRNA MCL-1) or a control shRNA vector (shRNA scr.) were pretreated with 5 µM ABT-737 for 1 h followed by irradiation with 0 Gy (sham irradiation) or 10 Gy. Cell death was determined as in A (mean ± SD, n = 3). p ≥ 0.05, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 (compared to non-irradiated controls/ctrl. cells treated with the same dose or as indicated, two-way ANOVA)