Analysis of Residual DSBs in Ataxia-Telangiectasia Lymphoblast Cells Initiating Apoptosis

Abstract

In order to examine the relationship between accumulation of residual DNA double-strand breaks (DSBs) and cell death, we have used a control and an ATM (Ataxia-Telangiectasia Mutated) defective cell line, as Ataxia-Telangiectasia (AT) cells tend to accumulate residual DSBs at long times after damage infliction. After irradiation, AT cells showed checkpoint impairment and a fraction of cells displayed an abnormal centrosome number and tetraploid DNA content, and this fraction increased along with apoptosis rates. At all times analyzed, AT cells displayed a significantly higher rate of radiation-induced apoptosis than normal cells. Besides apoptosis, 70-85% of the AT viable cells (TUNEL-negative) carried ≥ 10 γH2AX foci/cell, while only 12-27% of normal cells did. The fraction of AT and normal cells undergoing early and late apoptosis were isolated by flow cytometry and residual DSBs were concretely scored in these populations. Half of the γH2AX-positive AT cells undergoing early apoptosis carried ≥ 10 γH2AX foci/cell and this fraction increased to 75% in late apoptosis. The results suggest that retention of DNA damage-induced γH2AX foci is an indicative of lethal DNA damage, as cells undergoing apoptosis are those accumulating more DSBs. Scoring of residual γH2AX foci might function as a predictive tool to assess radiation-induced apoptosis.

Figure 1

(a) Radiation-induced apoptosis measured by means of Annexin-V/PI and TUNEL methodologies. Cytometry plots were used for gating cells stained using Annexin-V (An) and propidium iodide (PI) before and after irradiation. In all plots, the lower left quadrant corresponds to the viable, nonapoptotic cell population (An−/PI−). The lower right quadrant corresponds to the cell population An+PI–, which is undergoing early apoptosis (EA) and is shown in green. The upper right quadrant corresponds to the cell population An+PI+, which is undergoing late apoptosis (LA) and is shown in red. Frequencies of EA and LA are shown in each graph at 0, 24, 48, and 72 hours after irradiation in normal and AT cells and they correspond to the mean of 3 different experiments with two replicas each. A minimal number of 10000 cells were analyzed in each experiment. The asterisks indicate statistical differences in the apoptotic levels between AT and normal cells when comparing the EA fraction, the LA fraction, and the sum of Annexin-V-positive cells (EA + LA). In all cases, χ ² test was applied and the p values were <0.005. Frequencies of TUNEL-positive cells for each cell type at 0, 24, 48, and 72 hours pIR are shown over each cytometry plot. The asterisks indicate statistical differences between AT and normal cells (χ ² test; p values < 0.007). The values for TUNEL were obtained after scoring 1000 cells for each time point and each cell line using an epifluorescence microscope. (b) Scoring of TUNEL-positive cells. On the left, a general view under the microscope (40x) showing irradiated cells in which a combination of TUNEL staining (green) and γH2AX immunofluorescence (red) has been applied. DNA is stained with DAPI (blue). TUNEL-positive cells (white arrowheads) depict intense TUNEL staining and they show the morphological features of apoptotic cells (right panel): smaller nuclei with highly condensed chromatin—intensely stained with blue—undergoing variable levels of nuclear fragmentation. Also, TUNEL-positive cells could depict a pan-nuclear γH2AX staining but never had γH2AX foci. (c) Western blot detection of apoptotic markers. Normal and AT cells were irradiated with 5 Gy of γ-rays and expression of p53, its activated form phospho-p53 (Ser15), and other p53 targets such as p21, Bax, and the cleaved fraction of caspase 3 were analyzed at 0, 24, 48, and 72 hours after irradiation. Proteins were detected in two different experiments and GADPH was used as the housekeeping gene.

Figure 2

(a) Cell cycle analysis. The histograms show the cell cycle distribution of normal and AT cells before irradiation and at 24, 48, and 72 hours after irradiation. Cell cycle distribution was obtained by means of PI staining, which measures DNA content. The frequency of cells entering in S-phase for each cell type and each time point is shown, evidencing lack of IR-induced G1 checkpoint arrest in AT cells. The fraction of cells arrested in G2/M after irradiation and the tetraploid population (4N) arising after irradiation have also been highlighted. The frequencies displayed are the mean of two independent experiments in which a minimum number of 10000 cells were analyzed. (b) Tetraploidization and centrosome number. The image shows an AT lymphoblast (probably a metaphase) with 3 pericentrin signals (green; white arrowheads). The DNA is stained with DAPI and the red staining corresponds to α-tubulin. The bars in the graph show the fraction of tetraploid cells scored in AT and normal lymphoblasts before and after irradiation. The values are the mean of two experiments, and the error bars show the standard deviation. The lines in the graph depict the fraction of cells with an abnormal centrosome number (>2) within the same time points. The values for centrosome number were obtained after analyzing a minimal number of 400 cells for each cell type and each time point. The asterisk indicates statistical differences between normal and AT lymphoblasts in the frequency of cells with more than 2 centrosomes (χ ² test; p values < 0.002).

Figure 3

(a) Immunodetection of γH2AX in lymphoblasts. DSBs were scored by γH2AX foci detection in TUNEL-negative, An+/PI−, and An+/PI+ cells. Pan-nuclear γH2AX staining was scored in TUNEL-negative, An+/PI−, and An+/PI+ and in TUNEL-positive cells. (b) γH2AX-labeling in viable (TUNEL-negative) cells. The number and frequency of viable cells with γH2AX foci are reflected in the bars. Within this fraction, the frequency of cells with <10 foci or with ≥10 foci is shown inside the bars. The asterisks indicate statistical differences between normal and AT lymphoblasts in the fraction of cells with γH2AX foci or in the fraction of cells with ≥10 γH2AX foci (χ ² test; p values from p = 0.0270 to p < 0.0001). The frequencies for each category are calculated over the total number of TUNEL-negative scored cells. A minimal number of 350 TUNEL-negative cells were analyzed for each cell type and each time point. The apoptotic rate measured with TUNEL is depicted in the graph as a continuous line. Values for TUNEL-positive cells are given under the x-axis and are those corresponding to Figure 1(a). (c) γH2AX-labeling in Annexin-positive cells. AT and normal cells were irradiated and fractions corresponding to EA and LA were isolated by cell sorting. An+/PI− and An+/PI+ cells were classified into those with or without γH2AX foci and those with pan-nuclear γH2AX staining. The frequency of cells with γH2AX foci is depicted next to the bar. Within this fraction, the frequency of cells with less than 10 foci (light pink) or with 10 or more foci (pink) is shown inside the bars. The asterisks indicate statistical differences between normal and AT lymphoblasts in the frequency of cells with ≥10 γH2AX foci (χ ² test; p values from p = 0.0020 to p < 0.0001). The frequencies are calculated over the total number of An+/PI− and An+/PI+ sorted cells. A minimal number of 400 cells were analyzed for each cell type and each time point.