BioMed Research International Volume 2016 (2016), Article ID 8279560,
Analysis of Residual DSBs in Ataxia-Telangiectasia Lymphoblast Cells Initiating Apoptosis
Authors: Teresa Anglada, Mariona Terradas, Laia Hernández, Anna Genescà and Marta Martín
Affiliations: Departament de Biologia Cel·lular, Fisiologia i Immunologia, Universitat Autònoma de Barcelona, Edifici C, Bellaterra, 08193 Cerdanyola del Vallès, Spain
Exposure of cells to ionizing radiation generates DNA double-strand breaks (DSBs). The presence of DNA damage activates the DNA damage response (DDR) to halt cell cycle progression by activating the cell cycle checkpoints, and to repair broken DNA. Histone H2AX is phosphorylated in order to signal the damage and it accumulates at the DSBs sites forming foci. These foci can be detected by immunofluorescence, thus allowing DNA damage visualization. gH2AX accumulation favors DNA repair by allowing recruitment of other DNA repair proteins (1). Once the DSBs have been repaired, gH2AX foci disappear. When repair is not possible, cells experience a prolonged cell cycle arrest and they might eventually undergo programmed cell death by apoptosis (2).
The goal of this study is to analyze the relationship between persistence of unresolved DNA damage and onset of cell death. We have used a radiosensitive cell line deficient in the Ataxia-Telangiectasia Mutated protein (ATM); a key regulator of the DDR after DSBs induction (3). These cells provide an ideal model because they show DDR defects involving DNA repair and failure in checkpoint activation that allow the progression of unrepaired DSBs through the cell cycle.
ATM-deficient cells undergo high levels of apoptosis and deficient G1-checkpoint activation after DNA damage infliction
An ATM-deficient (hereafter named AT cells) and a proficient lymphoblastoid cell line were exposed to 5Gy of g-rays. Scoring of gH2AX foci and of apoptosis rates were quantified at 24, 48 and 72 hours after DNA damage induction.
We first determined the ability of AT cells to undergo apoptosis after g-rays irradiation. Despite the ATM deficiency, these cells were able to efficiently activate the p21- and caspase-dependent pathway that leads to apoptosis, that was analyzed using two different methods. First we used Annexin-V fluorescent detection of phosphatidylserine in the outer leaflet of the cell membrane, an early event in the apoptotic process. After that, apoptosis was measured using the TUNEL methodology, which detects a final stage of the apoptotic process characterized, among other features, by extensive degradation of the genome into ~180bp DNA fragments. Both methods demonstrated that the levels of IR-induced apoptosis were higher in AT cells than in its normal counterparts at all times analyzed (Figure 1A).
Although both Annexin-V/PI and TUNEL methodologies measured apoptosis, they seemed to detect correlative stages of this process. At 24 hours post-irradiation (pIR), there was a significant increase in cells undergoing early apoptosis, while yet very few cells were positive for TUNEL staining. The rate of early apoptotic cells positive for Annexin-V staining reaches a plateau at 48 hours pIR and decreases thereafter, while the frequency of TUNEL-positive events increases up to 72 hours after irradiation (Figure 1A). Because TUNEL methodology detects extensive DNA fragmentation, TUNEL-positive cells might undergo a later apoptotic stage than those signalled with Annexin-V. In this way, the combination of the results obtained with the Annexin-V and the TUNEL procedures renders a dynamic picture of the apoptotic process in the cells analysed.
Thus, AT cells effectively underwent high levels of apoptosis after irradiation, presumably due to inefficient checkpoint activation and subsequent progression of DNA damage through the cell cycle. So, our next step was to check cell cycle progression after irradiation. As expected, and contrary to normal cells, AT cells did show an impairment of the G1 checkpoint that allowed damaged cells to enter S phase and replicate their DNA, even if unrepaired. AT cells continued progression up to G2/M phase, where efficient arrest was detected (Figure 1B). Also, at 72h pIR a small but significant fraction of the irradiated AT cells showed a tetraploid DNA content, suggesting that they had re-entered the cycle and re-duplicated their DNA without successfully undergoing cytokinesis. This tetraploid population correlated with a significant fraction of AT cells with more than 2 centrosomes, the spindle poles organizers that duplicate –along with DNA– in each cell division. The presence of tetraploid cells and of abnormal centrosome number is frequently associated with mitotic catastrophe, an oncosupressive mechanism of cell death that is activated as a response to defects in cell cycle checkpoint activation, persistence of DNA damage and inability to resolve cell division (4). The observations suggest that this mechanism also contributes to the apoptosis levels scored in irradiated AT cells.
Figure 1 A. Radiation-induced apoptosis rates measured with Annexin-V and TUNEL methodologies. B. Comparison of the cell cycle distribution of normal and AT cells before irradiation and 72h pIR. The x-axis corresponds to the DNA content measured by PI staining. Percentages highlight the fraction of cells arrested in G2/M and the population of cells with 4N (tetraploid) DNA content.
AT cells accumulate more gH2AX foci and undergo higher apoptosis rates after irradiation
We next focused on analyzing the frequency of radiation-induced DSBs accumulating in viable cells and compare it to the frequency of cells undergoing apoptosis. The TUNEL methodology allowed for clear detection of viable cells (negative for TUNEL staining) and apoptotic cells (positive for TUNEL staining) (Figure 2A). As expected, the frequency of unrepaired DSBs -detected as gH2AX foci-, was higher in AT cells than in normal cells at all post-irradiation times analyzed (Figure 2B). Irradiated normal cells efficiently reduced the frequency of cells carrying DSBs, and reached basal levels (~20%) at 72h pIR (Figure 1). At 24h after irradiation some normal cells with DSBs accumulate >10 gH2AX foci/cell, but this fraction rapidly declines and most of these cells with DSBs had less than 10 gH2AX foci/cell at all times before and after irradiation. In normal cells, apoptotic levels decline when both the frequency of cells carrying DSBs and the frequency of cells with >10 gH2AX foci/cell also decrease. Conversely, more than 80% of the AT cells had gH2AX foci immediately after irradiation, and this frequency was still of 46,5% at 72h pIR (Figure 2B). Not only more AT cells had gH2AX foci, but also most of them had >10 gH2AX foci/cell, even at long times after DNA damage induction (Figure 2B). In AT cells, apoptotic levels were always higher than in normal cells, and they only stabilized at 72h pIR, when the frequency of cells with >10 gH2AX foci/cell also started to decrease. These results suggest that apoptotic levels are consistent with an increasing frequency of cells carrying several unrepaired DSBs for long times after DNA damage infliction.
An important fraction of cells carrying many unrepaired DSBs is found in early apoptosis
We then aimed to quantify the number of gH2AX foci in cells undergoing apoptosis, but no gH2AX foci were scored in AT or in normal TUNEL positive cells. Most probably, gH2AX foci formation is abolished in the later apoptotic stages due to extensive DNA condensation. We then reasoned that if Annexin-V staining detects an earlier apoptotic stage, this method might serve to select apoptotic cells in which unrepaired DSBs might still be visible. After flipping phosphatidylserine in the outer leaflet of the cell membrane –detected with Annexin-V–, and before undergoing extensive DNA fragmentation –detected with TUNEL–, cells also lose membrane integrity, allowing the incorporation of propidium iodide (PI), a DNA affinity staining. Thus, 48 hours after irradiation flow cytometry was used to select those cells positive for Annexin-V staining (Ann+/PI-), and those positive for both Annexin-V and propidium iodide (Ann+/PI+), which are cells presumably ahead in the apoptotic process. This time was chosen because it coincides with the highest rate of early apoptosis scored in irradiated AT and normal cells (Figure 1A). It was possible to detect unrepaired DSBs as gH2AX foci in both of these cell populations, allowing for quantification of unrepaired DSBs in apoptotic cells (Figure 2C), in which the number of gH2AX foci was scored. The frequency of apoptotic cells with >10gH2AX foci/cell was significantly higher in apoptotic AT cells (55-75%) than in apoptotic normal cells (26-41%), which agrees with the results observed in viable cells: because of their checkpoint deficiency and their DNA repair defect, AT cells always accumulate more unrepaired DSBs than the normal cells. Nonetheless it is important to note the evolution of the fraction of cells with more DSBs in normal and AT cells. As apoptosis begins (Ann+/PI-), 26% of the normal cells and 55% of the AT cells carry >10gH2AX foci/cell (Figure 2C), a fraction almost identical to that detected in viable cells at 48h pIR (Figure 2B). As apoptosis evolves (Ann+/PI+), the fraction of cells with more unrepaired DSBs increases in both cell types, 41% of normal cells and 75% with more than 10 gH2AX foci/cell (Figure 2C). These results are in line with others that have described that cells carrying DNA repair protein foci of Rad51 at 24h pIR are the ones more likely to die (5). Briefly, the results presented here suggest that the accumulation of unrepaired DSBs over time might serve as an efficient indicator of cell death, as cells that accumulate more DSBs per cell show higher apoptotic rates and those cells that progress through the apoptotic pathway are those carrying more DSBs.
Why is our study important?
This study proposes that the scoring of unrepaired DNA double strand breaks in the form of gH2AX foci in cells undergoing early apoptotic stages might be a powerful tool to estimate rates of radiation-induced apoptosis.
Figure 2 A. Scoring of TUNEL-positive cells and immunodetection of gH2AX in lymphoblasts. The upper image is a general view of irradiated cells under a microscope (40x) after immunostaining. Cells marked with white arrowheads are detailed in the panel below. Cell nuclei are stained with DAPI (blue), cells positive for TUNEL staining are shown in green and gH2AX foci are displayed in red. B. Frequency of viable (TUNEL-negative) cells with gH2AX foci. Cells with gH2AX foci in the viable population are represented with bars. Within them, the number of cells with <10 foci and with >10 foci is shown. The evolution of apoptosis rates with time is depicted as a solid line for each cell type in the graph. C. Number of gH2AX foci in apoptotic (Annexin-V positive) cells. The An+/PI− and An+/PI+ populations were isolated 48h after irradiation using flow cytometry. The percentages of cells with <10 foci and cells with >10 foci are calculated based on the total number of An+/PI− and An+/PI+ cells with gH2AX foci.
- E. P. Rogakou, C. Boon, C. Redon, and W. M. Bonner, “Megabase chromatin domains involved in DNA double-strand breaks in vivo,” The Journal of Cell Biology, vol. 146, no. 5, pp. 905–916, 1999.
- C. J. Bakkenist and M. B. Kastan, “Initiating cellular stress responses,” Cell, vol. 118, no. 1, pp. 9–17, 2004.
- Y. Shiloh, “ATM and related protein kinases: safeguarding genome integrity,” Nature Reviews Cancer, vol. 3, no. 3, pp. 155–168, 2003.
- I. Vitale, L. Galluzzi, M. Castedo, and G. Kroemer, “Mitotic catastrophe: a mechanism for avoiding genomic instability,” Nature Reviews. Molecular cell biology, vol. 12, no. 6, pp. 385– 392, 2011.
- M. Gatei, B.-B. Zhou, K. Hobson, S. Scott, D. Young, and K. K. Khanna, “Ataxia telangiectasia mutated (ATM) kinase and ATM and Rad3 related kinase mediate phosphorylation of Brca1 at distinct and overlapping sites. In vivo assessment using phospho-specific antibodies,” Journal of Biological Chemistry, vol. 276, no. 20, pp. 17276–17280, 2001.