gammaH2AX foci form preferentially in euchromatin after ionising-radiation

Abstract

Background: The histone variant histone H2A.X comprises up to 25% of the H2A complement in mammalian cells. It is rapidly phosphorylated following exposure of cells to double-strand break (DSB) inducing agents such as ionising radiation. Within minutes of DSB generation, H2AX molecules are phosphorylated in large chromatin domains flanking DNA double-strand breaks (DSBs); these domains can be observed by immunofluorescence microscopy and are termed gammaH2AX foci. H2AX phosphorylation is believed to have a role mounting an efficient cellular response to DNA damage. Theoretical considerations suggest an essentially random chromosomal distribution of X-ray induced DSBs, and experimental evidence does not consistently indicate otherwise. However, we observed an apparently uneven distribution of gammaH2AX foci following X-irradiation with regions of the nucleus devoid of foci.

Methodology/principle findings: Using immunofluorescence microscopy, we show that focal phosphorylation of histone H2AX occurs preferentially in euchromatic regions of the genome following X-irradiation. H2AX phosphorylation has also been demonstrated previously to occur at stalled replication forks induced by UV radiation or exposure to agents such as hydroxyurea. In this study, treatment of S-phase cells with hydroxyurea lead to efficient H2AX phosphorylation in both euchromatin and heterochromatin at times when these chromatin compartments were undergoing replication. This suggests a block to H2AX phosphorylation in heterochromatin that is at least partially relieved by ongoing DNA replication.

Conclusions/significance: We discuss a number of possible mechanisms that could account for the observed pattern of H2AX phosphorylation. Since gammaH2AX is regarded as forming a platform for the recruitment or retention of other DNA repair and signaling molecules, these findings imply that the processing of DSBs in heterochromatin differs from that in euchromatic regions. The differential responses of heterochromatic and euchromatic compartments of the genome to DSBs will have implications for understanding the processes of DNA repair in relation to nuclear and chromatin organization.

Figure 1. γH2AX foci induced by ionizing radiation are absent from HP1α-staining nuclear domains.

MCF7 cells were fixed 30 minutes after X-irradiation (2Gy, panels a–l & p–r) and were processed for immunofluorescence for γH2AX (green) and either HP1α or H3K9Me3 (red). Panels m–o, non-irradiated cells. DNA was stained with DAPI. Panels a, d, g, j, m & p, γH2AX; panels b, e, h, k & n, HP1α; Panel q, H3K9Me3; panels c, f, I, l, o & r merged images. Overlapping red and green signals appear yellow. The top row shows a group of cells with typical appearance. Individual nuclei are shown magnified in rows 2–6. For the nuclei shown in rows 2–4, line traces were generated (shown on the right) with the line drawn through the brightest HP1α regions. For γH2AX, HP1α and H3K9Me3, images were first adjusted using levels such that fainter interfocal nuclear fluorescence was not included.

Figure 2. γH2AX foci generated by etoposide treatment do not appear in HP1α-staining regions in MCF7 cells.

MCF7 cells were incubated in medium containing etoposide and processed as in Figure 1. Panels a–c, two hour exposure to 5 µM etoposide. Panels d–i, 24 hour exposure to 1 µM etoposide. Panels j–l, untreated cells. Cells were fixed immediately after etoposide treatment.

Figure 3. Spatial relationship between γH2AX foci, HP1α-staining heterochromatic regions and nucleoli.

Panel b–d and f–h were obtained from a single nucleus fixed 30 minutes after X-irradiation (2Gy) and processed for γH2AX and HP1α immunofluorescence. Panels, a & e were obtained from a separate irradiated nucleus processed for nucleolin and γH2AX immunofluorescence. Panel a, DAPI; b, γH2AX; c, HP1α; d, DAPI; e, merged DAPI (blue) and nucleolin (red), γH2AX (green) images; f, γH2AX (green)/DAPI (blue); g, HP1α (red)/DAPI (blue); h, γH2AX/HP1α/DAPI. In panels e–h, the red and green chanels were reduced to binary images, retaining as much detail as possible, before overlaying on the DAPI image. DAPI staining was carried out under optimum conditions to reveal nuclear structure.

Figure 4. Replication stress can induce phosphorylation of histone H2AX in heterochromatin.

Panels a–l, subconfluent asynchronous MCF7 cells were exposed to hydroxyurea (2 mM) for one hour immediately prior to fixation and processing for γH2AX (green) and HP1α (red) immunofluorescence. Panels a–c, representative nuclei displaying non-S phase, early to mid S-phase (S-E) and late S-phase (S-L) γH2AX staining respectively. Panels d–f & g–I, single S-L nuclei; j–l, single S-E nucleus. Panels d, g, j, γH2AX; e, h, k, HP1α; f, i, l, merged γH2AX/HP1α images. Line traces are presented on the right. Lines were drawn across the nucleus through heterochromatic (HP1α staining) regions in each case, including the DAPI channel. Panels m–r, subconfluent MCF7 cells were exposed to cisplatin (50 µM) for one hour, 38 hours after release from serum starvation. Cells were fixed immediately after cisplatin treatment and processed for γH2AX immunofluorescence. Panels m & p, γH2AX; n & q, DAPI; o & r merged γH2AX (green)/DAPI (red) images.

Figure 5. G₁ and S-phase H2AX phosphorylation.

MCF7 cells were serum starved (0.05% FCS) for 24 hours before release into medium containing 20% FCS. Cells were fixed at different times such that fixed cell populations were predominantly in G₁, early S (S-E) or late S-phase (S-L). (6.5 hr, 24 hr and 38 hr respectively). HU or cisplatin were added to the cells, to 2 mM or 50 µM respectively, at the indicated times prior to fixation. Irradiated cells were fixed 30 minutes after irradiation.