Oxidative stress, radiation and other external insults have been shown to damage DNA molecules in cells derived from organisms as diverse as bacteria, yeast, drosophila, rodents and man (Friedberg et al. 2005). The presence of DNA damage may lead to cell cycle checkpoint arrest to allow time for DNA repair processes to occur. If however, the system becomes overwhelmed or the DNA repair mechanisms are impaired, the cell either enters the apoptosis pathway or become cancerous due to the accumulation of mutations resulting from replication of damaged DNA. Therefore, a complete understanding of DNA repair mechanisms is of great interest in the study of cancer prevention and treatment. In addition, this process has been implicated in cellular senescence and aging (Sedelnikova et al. 2004). Following induction of DNA double strand breaks, the specialized histone protein H2AX becomes phosphorylated and rapidly (within minutes) accumulates at the sites of DNA damage forming distinct foci (Paull et al. 2000).
H2AX foci formation is followed by recruitment of many other proteins involved in the DNA repair process including the p53 binding protein 53BP1 (Schulz et al ., 2000). Over long time periods (hours), the number of foci declines with the progression of DNA repair. Detection and quantification of foci development as an indicator of DNA damage and repair is of great interest to groups investigating these pathways (Kim et al . 2005). The assay is generally performed using coverslips and chamber slides and imaged on fluorescence microscopes (Sedelnikova et al . 2004). The demands for high spatial resolution and sophisticated image analysis, however, have hindered the transfer of this assay into a multi-well plate format and thus into the drug screening arena. Our goal was to establish image acquisition and analysis methods allowing quantification of foci develop ment both over time and with increasing concentrations of H2O2, a known inducer of DNA damage.
HT1080 cells (ATCC, CCL-121, human fibrosarcoma derived cell line) were grown in 37C, 5%CO2 humidified incubators according to the suppliers protocol. Cells were seeded at 10,000 cell/well into 96 well plates (Cat. No. 353219) and incubated overnight. H2O2 (Sigma, H-1009) was used to initiate DNA double strand breaks. Immediately prior to the experiment, H2O2 was diluted in pre-warmed media and dose-response curves were obtained by serial dilution. Cells were exposed to chemical insult for 10 minutes at room temperature after which they were washed and replenished with fresh media. To allow for DNA repair processes to occur, the cells were returned to the incubator and left for 30, 90, or 180 minutes. In some experiments, cells were left to recover for 16 hours. At those time points, cells were fixed for 10 minutes with pre-warmed formaldehyde, 3.7% in phosphatebuffered saline (PBS), washed with PBS and permeabilized with ice-cold 90% methanol (10 minutes). Cells were washed twice with PBS and blocked using 1% BSA (in PBS) for 30 minutes. For the detection of phosphorylated H2AX, the mouse monoclonal anti-phospho serine 139 H2AX antibody (Chemicon, 05-636) at 2 μg/mL in 1% BSA block was used. 53BP1 was detected using the anti-53BP1 rabbit polyclonal antibody (Novus Biologicals, NB 100-304) diluted as described for H2AX.
After 1h at room temperature, the primary antibody was removed and cells were incubated with Tween 20 (0.05% in PBS, 10 minutes). Two PBS washes prepared the cells for the addition of the secondary antibodies, 5 μg/mL Alexa Fluor 555-conjugated anti-mouse IgG (Cat. No. A21424) for H2AX and 5 μg/mL Alexa Fluor 488-conjugated anti-rabbit IgG (Cat. No. A1 1034) for 53BP1, respectively. The DNA intercalating dye Hoechst 33342 (Invitrogen, H3570) was added at this point at 2 μg/mL and removed with the secondary antibodies after 1h at room temperature (light protected). Cells were washed as described above and wells replenished with 200 μL of PBS. Plates were sealed with adhesive foil and imaged on the BD Pathway Bioimager. Analysis of the image data was performed using BD IPLab for Pathway (Cat. No. 340989) and plotted using BD Image Data Explorer (Cat. No. 341039).
Treatment of HT1080 cells with H2O2 rapidly induced the formation of phosphorylated H2AX and 53BP1 foci (Figure 1 ). There is a marked increase in the number and intensity of foci in H202 treated cells compared to control cells. The two foci shared a high degree of co-localization within nuclei as observed in the merged images.
Preliminary results had shown that the nuclear foci were not evenly distributed in the same focal plane ( Figure 2 ). We used the ability of the BD Pathway Bioimager to acquire 9 sections separated by 1.5 μm steps in confocal mode. To achieve the required spatial resolution, we used a 40 (NA 0.90) objective. Using the novel collapsed-stack acquisition feature, images were then automatically projected into a single focal plane with all objects appearing in focus.
This allowed segmentation and identification of objects within the nucleus using a customized image analysis routine written in BD IPLab for Pathway. The collapsed stack images were compared to images acquired in non-confocal mode ( Figure 2 shows part of an image containing a single nucleus). The analysis tool generated specific parameters such as foci count, foci area as percent nuclear area, average foci intensity and average foci size. On the cellular level, the response appeared highly heterogeneous and cells could be found that sh owed very few foci next to cells containing a large number of foci ( Figure 1 ). We therefore sampled on average 120 cells/well in a 33 montage. Data was imported into BD Image Data Explorer where dose-response curves and Z data were generated ( Figures 3 and 4 ).
The ability to acquire confocal, collapsed stack images has allowed us to develop an automated assay relying on identification of small sub-nuclear features, namely foci of DNA damage and repair. The improved statistical performance of the assay in confocal, collapsed stack mode suggests that this assay indeed can be applied to a high-content screening environment.
We noted that even untreated cells contained some foci, suggesting that under basal conditions, some DNA doublestrand breaks are present and repair occurs. Phosphorylated H2AX accumulation was seen immediately after chemical treatment confirming the rapid time course of the activation and recruitment of this protein to the sites of DNA damage.
The subsequent accumulation of 53BP1 foci peaked between 30 and 90 minutes and is in agreement with published reports (Schultz et al . 2000). To our knowledge, the EC50 for H2O2-induced H2AX and 53BP1 foci formation has not been published and we believe that our assay can aid the characterization of other chemical agents and DNA repair proteins in a similar manner. The appearance of the 53BP1 foci in the same spatial location after 30 minutes confirms reports of rapid activation of the repair signaling cascade (Schultz et al . 2000)
Although we did not investigate the signaling properties between these two proteins, the assay is very suitable to investigate whether intermediate molecules are involved or whether 53BP1 is recruited by activated H2AX directly.
The comparatively straight-forward a ssay protocol using only 2 primary antibodies, 1 dye and a commercially available cell line allow this assay to be used in automated high-content screening applications where hands-off operation is required. The BD Pathway Bioimager with its confocal capability, white-light illumination and choice of 16 excitation filters allows development of multi-color assays and would even accommodate a 4th imaging channel. The complexity of change in the cellular phenotype in this assay mandate sophisticated and specialized image and data analysis algorithms. The combination of BD IPLab for Pathway and BD Image Data Explorer allows generation and analysis of high-content data in a drug screening environment.
1. Friedberg, E.C, Walker, G.C., Siede, et al . 2005 DNA Repair and Mutagenesis, ASM Press; 2nd edition ISBN : 1555813194
2. Kim, J-S, Krasieva, J_S, Kurumizaka, H, et al . 2005. Independent and sequential recruitment of NHEJ and HR factors to DNA damage sites in mammalian cells. JCB , 170, 341-347
3. Paull T.T., Rogakou E.P., Yamazaki V., Kirchgessner C.U., Gellert M., Bonner W.M..2000. A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr Biol. 10, 886-95.
4. Schultz, L.B, Chehab, N.H., Malikzay, A., Halazonetis, T.H., 2000. p53 Binding Protein 1 (53BP1) Is an Early Participant in the Cellular Response to DNA Double-Strand Breaks. JCB , 7, 1381-1390
5. Sedelnikova O.A., Horikawa I., Zimonjic D.B., Popescu N.C., Bonner W.M., Barrett J.C. 2004. Senescing human cells and ageing mice accumulate DNA lesions with unrepairable double-strand breaks. Nature Cell Biology. 6, 168-70.