|
|
|
|
|
|
|
||
Article |
Correspondence to Michael J. Kruhlak: kruhlakm{at}mail.nih.gov; or André Nussenzweig: andre_nussenzweig{at}nih.gov
 Top Abstract IntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Abbreviations used in this paper: ATM, ataxia telangiectasia mutated; DSB, double-strand break; EFTEM, energy-filtering transmission EM; ESI, electron spectroscopic imaging; GR, gluccocorticoid receptor; IRIF, irradiation-induced foci; MEF, mouse embryo fibroblast; MMTV, mouse mammary tumor virus; PAGFP, photoactivatable GFP; WT, wild type.
 |
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
The best characterized DNA damage–induced histone modification is the phosphorylation of the core histone variant H2AX (
-H2AX; Rogakou et al., 1998; Fernandez-Capetillo et al., 2004b), which extends to large chromatin domains flanking each DSB (Rogakou et al., 1999; Downs et al., 2004; Shroff et al., 2004). Although the presence of H2AX is not required for the initial recognition of DSBs (Celeste et al., 2003), -H2AX organizes the dynamic assembly of multiprotein complexes into cytologically visible nuclear foci (Paull et al., 2000; Bassing et al., 2002; Celeste et al., 2002). The physical structure and composition of -H2AX–containing chromatin domains is unknown, but the phosphorylation of H2AX has been linked to an increased chromatin relaxation in Saccharomyces cerevisiae (Downs et al., 2000), whereas in contrast, mouse H2AX is required for chromatin condensation and transcriptional silencing of the male sex chromosomes during spermatogenesis (Fernandez-Capetillo et al., 2003). In addition, several chromatin remodeling complexes, including INO80 (Morrison et al., 2004; van Attikum et al., 2004) and NuA4/Tip60 (Downs et al., 2004; Kusch et al., 2004; Tsukuda et al., 2005), as well as key structural components (such as cohesin; Strom et al., 2004; Unal et al., 2004) assemble on chromatin in a -H2AX–dependent manner. Together, these observations indicate that the phosphorylation of H2AX may directly or indirectly modulate chromatin architecture in the vicinity of a DSB, which is a hypothesis tested in this study.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
-H2AX foci within tracks of DSBs in mammalian cells (Aten et al., 2004). To directly monitor the mobility of chromatin containing DSBs in vivo, we expressed histone H2B tagged with a photoactivatable version of GFP (PAGFP; Patterson and Lippincott-Schwartz, 2002) in wild-type (WT) and H2AX–/– mouse embryo fibroblasts (MEFs; Celeste et al., 2002). All of the core histones, including H2B, are tightly bound to DNA and are immobile over relatively long time periods and, thereby, provide excellent markers for chromatin in living cells (Kanda et al., 1998; Kimura and Cook, 2001; Siino et al., 2002). By using the 364-nm emission from a UV laser on a confocal microscope, we simultaneously introduced localized DNA DSBs within the nucleus of cells and photoactivated H2B-PAGFP. As expected, the introduction of DSBs, monitored by the formation of -H2AX and the recruitment of Nbs1 in fixed cells, was dependent on sensitizing the cells with the Hoechst 33342 DNA-binding dye (WT + Hoechst and H2AX–/– + Hoechst; Celeste et al., 2003), whereas H2B-PAGFP was photoactivated regardless of the presence of the dye (Fig. 1 A).
|
To examine the mobility of -irradiation–induced foci (IRIF) immediately after they form, we introduced DNA damage by -irradiation and monitored the spatial position of GFP-53BP1 (Huyen et al., 2004). Like endogenous 53BP1, GFP-53BP1 forms foci within minutes at sites of DSBs as determined by staining with -H2AX (Fig. S1; Huyen et al., 2004). After their initial appearance, GFP-53BP1 IRIF exhibited limited diffusional motion with a mean squared displacement of 0.9 µm2 over the 50-min time period monitored (3 x 10–4 µm2/s). This is similar to the range of movement of individual loci observed in undamaged human cells (Chubb et al., 2002). Although IRIF did not congregate within common sites in the nucleus, they did exhibit local small-scale dynamics, with IRIF within close proximity to each other frequently interacting but subsequently separating to their own nuclear space (Fig. 1 D, Fig. S1, and Video 3, available at http://www.jcb.org/cgi/content/full/jcb.200510015/DC1). This behavior occurred in a cell cycle–independent manner. In cells (n = 10) that we followed through mitosis and into early G1 and subsequently irradiated, we found that 53BP1 foci did not exhibit significant mobility compared with other irradiated interphase cells, nor did the foci appear to exhibit an increased propensity to associate in distinct subregions of the nucleus (Fig. S1). Similarly, early interphase cells exposed to localized UV laser microirradiation did not exhibit increased chromatin mobility compared with other interphase cells (not depicted). Collectively, our results indicate that DSBs generated by UV laser or irradiation do not exhibit large-scale mobility or strong cohesiveness.
Changes in chromatin upon the introduction of DNA DSBs
Despite the lack of large-scale genomic repositioning, simple visual inspection revealed that the photoactivated and DSB-containing chromatin regions underwent local expansion of photoactivated chromatin outside of the initial damaged zone (Fig. 2 A, Fig. S2, and Video 1). This expansion occurred immediately upon the introduction of DSBs and continued until 180 s after exposure to UV microirradiation, corresponding to a time interval in which DNA repair factors such as Nbs1 (Fig. 1 A) and ataxia telangiectasia mutated (ATM; see Fig. 4) are recruited to sites of DSBs. The persistence of photoactivated H2B in defined DNA damaged regions 4 h after UV laser microirradiation indicates that minimal amounts of H2B are released from chromatin containing DSBs (Fig. S1). Furthermore, we observed a similar expansion of H3-GFP in chromatin containing DSBs, which was confined to areas immediately surrounding regions exposed to UV laser microirradiation (Fig. 2 B). Importantly, these changes in chromatin structure were dependent on the presence of DSBs because they did not occur in photoactivated regions in cells lacking Hoescht sensitizing dye (Fig. S2 and Video 2). The change in area occupied by photoactivated chromatin corresponded to a 30% increase relative to the initial damaged area and occurred similarly in WT, H2AX–/–, and ATM–/– cells (Fig. 2 F). Thus, DSBs induce a local chromatin expansion that does not depend on the presence of either H2AX or ATM.
|
The observed change in chromatin gene array density may be caused by DNA damage–induced chromatin opening such that the bound GFP-GR signal is no longer concentrated sufficiently to detect above background nucleoplasmic levels of GFP-GR fluorescence. A nonmutually exclusive possibility is that GFP-GR may dissociate from the array. To discriminate between these possibilities, we performed DNA FISH with gene array–specific probes and measured the relative area occupied by the nondamaged (pre-UV) and damaged regions within the same nuclei (Fig. 2 D). As summarized in Fig. 2 G, the FISH signal (180 s after UV) occupied a greater area than that occupied by GFP-GR before exposure to UV microirradiation and a greater area than control FISH signals in cells not exposed to UV microirradiation (Fig. 2 E). The finding that damaged gene arrays occupied a greater area than before UV laser microirradiation indicates that specific chromatin regions that are transcriptionally competent undergo further remodeling upon the introduction of DSBs. We conclude that the changes observed in the gene arrays in living cells are the result, at least in part, of chromatin decondensation.
The laser microirradiation experiments were performed at the lowest doses (
0.86 nJ/pixel) that yielded a consistent -H2AX signal. By measuring the whole cell -H2AX response to irradiation, we established a standard curve that was used to calibrate the extent of DNA damage induced by the UV laser (Fig. S4, available at http://www.jcb.org/cgi/content/full/jcb.200510015/DC1). By this method, we estimate that the level of irradiation required to induce the same density of H2AX phosphorylation throughout the nucleus as laser microirradiation within the circular regions (Fig. 2, A, C, and D) is 2.5 Gy. Assuming that 1 Gy produces 35 DSBs, this equates to 85 DSBs localized to a specific subregion of the nucleus, corresponding to a density of 1 DSB for every 0.95 Mb DNA. This degree of DNA damage would not be expected to extensively fragment the chromatin; otherwise, H2B-PAGFP would have been mobile, and the signal would have eventually dispersed. Instead, we found that the H2B-PAGFP signal remained within the locally irradiated region and that the integrated intensity of fluorescence did not change over time. The most likely explanation for the expansion is that DSBs induce a local decondensation of chromatin. Examining the damaged regions in greater detail by electron microscopy further supported this interpretation.
Changes in chromatin containing DSBs monitored at high resolution by EFTEM
To monitor chromatin changes associated with DSBs at high resolution, we used correlative fluorescence and energy-filtering transmission EM (EFTEM; Spencer et al., 2000; Ren et al., 2003; Dellaire et al., 2004). In brief, electron spectroscopic imaging (ESI) with an energy-filtering transmission electron microscope generates element-specific maps of the specimen. Nucleic acid–containing structures are both phosphorus and nitrogen rich, whereas protein-based structures are nitrogen rich and phosphorus poor. By obtaining phosphorus and nitrogen image maps and examining the phosphorus and nitrogen content, nucleic acid–containing structures such as chromatin can be distinguished from primarily protein-based structures in intact nuclei (Bazett-Jones and Hendzel, 1999; Hendzel et al., 1999). Combining fluorescence microscopy with ESI, we were able to monitor specific subregions of the nucleus in living cells and examine the same regions at high resolution by EM. WT and H2AX –/– cells exposed to irradiation or UV laser microirradiation were analyzed using this correlative ESI method.
After introducing DSBs with the UV laser, WT and H2AX–/– cells were fluorescently labeled with anti–-H2AX or -Nbs1 antibodies and embedded, and ultrathin sections of the cells were imaged using the light microscope to collect a fluorescence map of the chromatin regions containing DSBs. The same sections were then imaged using the ESI mode of EFTEM, and nitrogen (Fig. 3, green) and phosphorus (Fig. 3, red) elemental maps were generated. The fluorescence image of the -H2AX or Nbs1 signal was resampled and overlaid onto a low magnification composite ESI-generated net phosphorus image map (Fig. 3, grayscale; white signal on black background). Subsequently, specific regions were examined at high magnification that were sufficient to resolve individual chromatin fibers.
|
-H2AX to mark DSBs, which are shown by green fluorescence track overlaid onto the low magnification ESI net phosphorus image in the top left panel. The ESI image is shown again in the top right panel of Fig. 3 A but with three yellow asterisks denoting the right-most boundary of the track exposed to UV laser microirradiation (the two gray lines traversing the laser track are EM sectioning artifacts). In this low magnification image, a general decrease in phosphorus density (lower signal intensity) was found specifically in the -H2AX–containing domains compared with the neighboring undamaged regions (higher signal intensity). This was confirmed by line scans through the UV microirradiated regions (Fig. S5, available at http://www.jcb.org/cgi/content/full/jcb.200510015/DC1). To determine whether this change affected chromatin structure specifically, we examined at higher magnification an area in which DSBs were introduced by the UV laser in only a subregion of heterochromatin (Fig. 3 A, bottom; arrows; the yellow box in the top panel denotes the region imaged at higher magnification by ESI). We found the regions containing DSBs displayed 10–30 nm chromatin fibers and appeared less condensed (Fig. 3 A, yellow arrow) relative to neighboring heterochromatin that was not exposed to laser-directed damage (Fig. 3 A, white arrow). In addition, the chromatin fibers in the DNA-damaged regions had significant accompanying nitrogen-only content (Fig. 3 A, green) surrounding the chromatin fibers compared with the adjacent region not containing DSBs, which likely represents the accumulation of DNA damage response proteins.
To determine whether H2AX is required for chromatin opening at DSBs, we examined the ultrastructure of chromatin containing DSBs in H2AX–/– MEFs. Because Nbs1 is recruited to DSBs independently of H2AX (Fig. 1 A; Celeste et al., 2003), UV laser–irradiated H2AX–/– cells were immunolabeled against Nbs1 to mark sites of DSBs (Fig. 3 B). Introducing DSBs specifically in regions of heterochromatin (Hoechst-rich regions in mouse cells) allowed us to directly compare the difference in chromatin opening within the damaged regions (Fig. 3 B, yellow arrows) to an equivalent heterochromatin region not containing DSBs (Fig. 3 B, red arrow). We found that DSBs induced a decondensation of heterochromatin characterized by the same 10–30-nm chromatin fibers observed in the damaged WT cells. Moreover, there was significantly more protein surrounding these decondensed chromatin fibers relative to the undamaged 2-µm diameter condensed heterochromatin blocks. Thus, consistent with the local chromatin expansion observed in live H2AX–/– cells (Fig. 2 F), the initial remodeling of damaged chromatin determined at high resolution occurs independently of H2AX.
We then compared the chromatin topology observed by UV laser–induced DNA damage to that introduced by irradiation, which results in a lower density of DSBs. WT MEFs were incubated 16 h after 8 Gy irradiation and were subsequently fixed and labeled for -H2AX (Fig. 3 C). Individual IRIF were easily discernable by immunofluorescence microscopy (Fig. 3 C, left). Immunogold labeling with silver enhancement was used to locate the IRIF regions within the nucleus in the high magnification ESI images. The gold particles are shown in the net nitrogen and net phosphorus overlay image as white dots (Fig. 3 C, middle). This cluster of white dots represents one IRIF observed by immunofluorescence microscopy (Fig. 3 C, arrow), which likely corresponds to a single DSB. Decondensed chromatin fibers on the order of 10–30 nm and nitrogen-rich/phosphorus-poor regions were characteristic of the focus (Fig. 3 C, right), which is consistent with the nature of the chromatin observed in the UV laser–irradiated regions in the aforementioned WT and H2AX–/– cells. Thus, a similar relaxation in higher order chromatin structure is observed both within IRIF and in UV laser–irradiated regions.
To quantify the density of chromatin fibers in DNA-damaged areas, we calculated the range of phosphorus/nitrogen intensity ratios that were representative of chromatin from nonirradiated control cells (not depicted). The phosphorus/nitrogen ratios measured for nondamaged chromatin was used as a minimum threshold for display purposes, and any pixel within subsequent combined net phosphorus/nitrogen overlay images that contained a phosphorus/nitrogen ratio within this range (representative of chromatin) was displayed as white (Fig. 3). In both WT and H2AX–/– cells, the damaged regions were more open, with chromatin fibers shown as white and phosphorus-poor and nitrogen-rich areas interdispersed among the chromatin fibers. By measuring the number of pixels containing the phosphorus/nitrogen ratio representative of chromatin compared with the total number of pixels within both damaged and nondamaged regions of the image (for example, see yellow outlines in Fig. 3, A–C), chromatin density coefficients (see Materials and methods) were calculated. In both UV laser–irradiated WT and H2AX–/– cells, the chromatin density coefficient decreased by 0.4 (on a scale of 0–1.0) for the regions containing DSBs compared with the nondamaged regions (Fig. 3 D), indicating that the density of chromatin fibers in the vicinity of DSBs is reduced by 40%. In addition, in the -irradiated WT cells, the chromatin density coefficient in the IRIF focus was 40% lower than in the region immediately outside of the focus (Fig. 3 D). Thus, the ultrastructure of chromatin within IRIF and UV laser–irradiated regions represents a consistent change that is related to the presence of DSBs.
Phosphorylation of ATM is initially restricted near sites of DNA damage
Given that the decrease in chromatin density was confined to a region surrounding DSBs, regardless of the method used to introduce DSBs, we were interested in determining how these changes related to the activation of DNA damage signaling. Specifically, it has been documented that alterations in chromatin structure rapidly induce the global nuclear activation of the ATM kinase (Bakkenist and Kastan, 2003). A marker of ATM activation is the autophosphorylation at residue Ser 1981 (Bakkenist and Kastan, 2003). To determine the spatio-temporal dynamics of ATM phosphorylation relative to the DNA damage–induced chromatin decondensation, HeLa cells expressing H2B-PAGFP were exposed to UV laser microirradiation in specific subregions of the nucleus, fixed with PFA at different time points after UV, and costained for ATM-1981P and -H2AX. Cells were relocated after staining based on the presence of the photoactivated H2B-PAGFP, and confocal images were collected using identical imaging parameters. As shown in Fig. 4 A, the phosphorylated ATM was initially confined to the laser-irradiated region, and both ATM-1981P and -H2AX accumulated at the DSB sites during the first minute. Similarly, the chromatin expansion was detectable within 20 s after UV microirradiation and continued until 180 s after irradiation (Fig. 2 A and not depicted). At 5 min after UV exposure, the phosphorylated form of ATM began to spread out from the damaged zone to the surrounding nucleoplasm, whereas the -H2AX signal was restricted to the original UV-exposed region (Fig. 4 A).
|
Remodeling of chromatin upon the introduction of DSBs is energy dependent
DNA damage response proteins are thought to move in a passive, energy-independent manner, scanning the nuclear volume for high-affinity binding sites (Nelms et al., 1998; Houtsmuller et al., 1999). To determine whether DSB-induced chromatin expansion occurs via an energy-dependent mechanism, we treated cells with the metabolic inhibitors 2-deoxyglucose and sodium azide (Platani et al., 2002). In contrast to normal cells, when energy-depleted cells were subjected to DSBs, a minimal amount of H2B-PAGFP expanded to outside the initial damaged region after 180 s (Figs. 5 A and 2 F). Thus, DSB-induced chromatin expansion is dependent on ATP. To determine whether DNA damage response proteins can still detect DSBs under energy-depleting conditions, we examined the recruitment of Nbs1 and 53BP1 to UV laser–irradiated regions and IRIF, respectively. Although -H2AX formed weakly in ATP-depleted cells (not depicted; but also see Fig. 5 C), Nbs1 (Fig. 5 B) and 53BP1 (Fig. 5 C) failed to accumulate at sites of DSBs. Moreover, combined irradiation followed by GFP photobleaching experiments confirmed that ATP depletion did not significantly alter the mobility of Nbs1 and 53BP1 (Fig. 5 B and not depicted). Thus, both 53BP1 and Nbs1 diffuse through the nuclear space but do not accumulate on chromatin containing DSBs in energy-depleted cells possibly because of a lack of available high-affinity binding sites or the impediment of ATP-dependent chromatin remodeling complexes. Alternatively, PI-3 kinase–related phosphorylation events may be required, although we have demonstrated chromatin decondensation in ATM–/– MEFs. Regardless, initial changes in chromatin that occur as a result of the presence of DSBs are dependent on processes that metabolize ATP.
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Although DNA damage generated by UV laser or irradiation does not alter the nuclear volume explored by chromosomes, chromatin containing DSBs undergoes a local expansion in the vicinity of the break. As determined by correlative fluorescence and EFTEM, these nuclear regions are characterized by lower order fibers that are morphologically reminiscent of euchromatin regions found in nondamaged nuclei. Although the mechanism that determines the boundary of this DSB-induced chromatin opening or that limits the expanse of H2AX phosphorylation is unclear, neighboring heterochromatin may provide a physical block to the spreading of the damaged chromatin domain. Because a DSB interrupts the continuous chromosome fiber, it is possible that the unfolding of the damaged chromatin region is established by the break itself, which could relieve some of the torsional stress imposed during the packaging of DNA. However, we have found that the DSB-induced chromatin alteration is dependent on ATP. Therefore, it is more likely that primary DSB sensors with ATP-dependent chromatin unwinding activities, which associate with DSBs in an H2AX-independent manner, play an active role in the initial alteration of chromatin folding. Such damage sensors would thereby link the recognition of a DSB to chromatin decondensation and the local activation of the DNA damage response, which is indicated by the initial accumulation of phosphorylated ATM within close proximity to regions containing DSBs.
H2AX is dispensable for the initial recruitment of DSB response proteins to DSBs and the signaling of DNA damage (Celeste et al., 2003). As shown in this study, the decondensation of chromatin at sites of DSBs also occurs independently of H2AX. Nevertheless, we have found that after the initial recruitment of DNA damage response proteins, they fail to stably associate with DSBs in H2AX–/– cells and "fall off" of chromatin (Celeste et al., 2003). Furthermore, H2AX dephosphorylation turns off DNA damage signaling, thereby triggering exit from cell cycle arrest (Keogh et al., 2005). Therefore, we speculate that -H2AX, although dispensable for the initial remodeling of chromatin, may be essential for maintaining the decondensed accessible state of chromatin. At late time points after DNA damage, the attenuation of -H2AX may allow chromatin to revert to a configuration that no longer permits access to DNA damage response proteins, thereby interrupting damage signaling even in cases in which DNA damage has not been adequately repaired (Keogh et al., 2005; Nussenzweig and Paull, 2006). This hypothesis can now be tested with the aforementioned methodology, which permits direct assessment of the physical state of damaged chromatin.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Photoactivating GFP and introducing DNA DSBs
The 364-nm emission of the UV laser described above was used to photoactivate GFP and to induce DNA damage. Incubation of cells with 7.5 µg/ml of the DNA-binding dye Hoechst 33342 (Sigma-Aldrich) was necessary for the UV laser to induce DSBs. However, the Hoechst dye was not necessary to photoconvert the PAGFP to the activated form using the UV laser. Cells were incubated in phenol red–free DME containing 10% FBS (Invitrogen) with or without Hoechst dye while imaging experiments were performed. The bleaching algorithm provided with the LSM510 software (Carl Zeiss MicroImaging, Inc.) was used to perform GFP photobleaching, PAGFP photoactivation, and to introduce DNA damage at specific sites within nuclei. The UV laser intensity was set to 0.86 nJ/pixel (measured at the specimen; see next section) for introducing DSBs and photoactivating PAGFP. We generally observed an 25-fold increase in PAGFP signal intensity upon photoconversion.
The calculated equivalence of UV laser microirradiation to Gy of irradiation
Multiple populations of cells were irradiated either with irradiation over a range of doses (0–20 Gy) or exposed to different levels of UV laser microirradiation (50% laser output with 5, 10, 20, or 50% Acuosto optical tunable filter–modulated transmission; equivalent to 0.29, 0.48, 0.86, and 2.05 nJ/pixel, respectively, measured at the specimen plane) in specific subregions of the nucleus, fixed and immunolabeled against -H2AX, and imaged using consistent imaging parameters on the LSM 510 confocal microscope (described in Microscopy). The resultant confocal fluorescence images were corrected for background, and the integrated fluorescence intensity was measured for the entire nucleus in the case of -irradiated cells or for each individual subnuclear region in the case of UV laser–irradiated cells. A standard curve of relative integrated intensity to the amount of irradiation (Gy) was plotted (Fig. S4). A linear best-fit curve was calculated, and the resultant integrated intensity values were measured from the UV laser–irradiated samples fitted to the curve. The equivalent number of Gy was obtained from the equation of the best-fit curve. The 0.86 nJ/pixel level of UV laser microirradiation was used consistently for the experiments performed, which corresponds to the introduction of one DSB for every 941 kb of DNA. The density of DSBs within the UV laser–irradiated regions was measured as follows: the density of DNA within the nucleus was calculated by taking the volume of the cell nucleus to be 523.60 µm3 based on a 10-µm diameter sphere and the amount of DNA present in a nonreplicated nucleus to be 6 x 109 bp DNA, therefore resulting in 1.14 x 107 bp DNA/um3. The volume of the subregions used and measured in the test sample was 15.2 µm3, thereby providing 1.73 x 108 bp DNA exposed to UV laser microirradiation. From the irradiation (Gy) standard curve, the DNA within this subnuclear region was exposed to 5.25 Gy of irradiation, which corresponds to 184 DSBs given that 1 Gy introduces 35 DSBs in whole-cell -irradiated cells. This provides a density of DSBs in the UV laser–irradiated subregions of one DSB for every 941 kb DNA. The UV laser–irradiated regions in the H2B-PAGFP–expressing cells were approximately half (7.08 µm3) that used in the aforementioned test cells, which, based on the same UV laser microirradiation conditions, corresponded to 2.45 Gy of irradiation in the subregions of H2B-PAGFP–expressing cells or 85 DSBs in 81 Mb DNA.
Cell lines, GFP-tagged protein constructs, FISH, immunolabeling, and ATP depletion
WT, H2AX–/–, and ATM–/– MEFs as well as HeLa and mouse 3617 cells were used for imaging experiments. The H2AX–/– and ATM –/– cell lines were generated from embryonic day 13.5 embryos derived from intercrosses between H2AX+/– or ATM+/– mice as previously described (Celeste et al., 2002). The mouse 3617 mammary tumor virus (MMTV) gene tandem array–containing cell line has been described previously (Walker et al., 1999). Cells were grown overnight in the presence of tetracycline to allow expression of the GFP-tagged GR protein. The following day and 30 min before exposing cells to UV laser microirradiation, 100 nM dexamethasone was added to the culture medium to induce translocation of GFP-GR to the nucleus and subsequent binding to promoter sequences in the MMTV gene array. For DNA FISH, MMTV gene array–containing cells were grown on gridded coverslips (Bellco Glass, Inc.). After UV laser microirradiation, cells were fixed in 4% PFA, and DNA FISH was performed as described previously (Muller et al., 2001). The H2B-PAGFP gene construct was modified from the H2B-GFP and PAGFP constructs described previously (Kanda et al., 1998; Patterson and Lippincott-Schwartz, 2002), which were supplied by T. Misteli (National Cancer Institute [NCI]) and G. Patterson (National Institute of Child Health and Human Development [NICHD]), respectively. GFP-53BP1 was generously provided by T. Halazonetis (Wistar Institute, Philadelphia, PA; Huyen et al., 2004). H3-GFP was provided by H. Kimura (Kyoto University, Kyoto, Japan; Kimura and Cook, 2001). NBS1-GFP was generated by cloning GFP in frame with mouse Nbs1. All cell transfections were conducted using FuGENE6 (Roche) with cells growing on either gridded on nongridded chambered coverglass (MatTek Corp.). Cells were transfected with either GFP-53BP1, Nbs1-GFP, H3-GFP, or H2B-PAGFP and exposed to UV laser microirradiation or 1–10 Gy of irradiation. Antibodies used in the immunofluorescence labeling protocol included anti-Nbs1 (1:1,000; Chen et al., 2000), anti-H2AX (1:1,000; Upstate Biotechnology), and anti-ATM1981p (1:300; Rockland Immunochemicals). ATP depletion was achieved by incubating cells for 30 min before imaging in media composed of PBS supplemented with 10% FBS, 10 mM 2-deoxyglucose, 10 mM sodium azide (Platani et al., 2002), and 7.5 µg/ml Hoechst dye, and ATP depletion was confirmed by the morphology and staining of mitochondria with 50 nM Mitotracker red CMXRos (Invitrogen).
Image processing and analysis
The Medical Imaging, Processing, and Visualization (MIPAV) software package (Matthew McAuliffe and coworkers, Center for Information Technology [CIT]/National Institutes of Health [NIH]) was used for area, center of intensity mass, and mean squared displacement measurements. Image series registration was performed using an optimized automatic 2.5D registration algorithm based on bilinear interpolation and least squares cost function. Region of interest statistics were generated using the statistics generator function in MIPAV. Area and volume measurements were also made using Imaris image processing and analysis software (Bitplane AG). Background-subtracted mean fluorescence intensities, line scan profiles, and particle tracking were also measured using MetaMorph image processing and analysis software (Universal Imaging Corp.). For H2B-PAGFP time series images, the intensity histogram was stretched for each time point to optimize viewing of the morphology of photoactivated regions for display purposes. Measurements were made on background-corrected raw data before or after image registration, with threshold levels set at an intensity value 10% greater than background and regions demarcated using the levelset volume of interest tool in the MIPAV software. Maximum intensity projections and phosphorus/nitrogen ratio (chromatin density) coefficients were generated and measured using the colocalization module of the LSM510 software. Chromatin density coefficients were calculated as follows: the phosphorus and nitrogen signal intensities representative of chromatin (taken from images of control nondamaged cells) were used as minimum signal thresholds for subsequent phosphorous/nitrogen overlay image sets. Nuclear pore complexes were used as a protein-only internal control representative of legitimate high nitrogen signal and low phosphorus signal above background. Within specific regions of the image sets, either in the irradiated areas (damaged) or adjacent to the irradiated areas (nondamaged), the number of pixels containing the phosphorus and nitrogen signal intensities representative of chromatin were added and divided by the total number of pixels within the region, which yielded the chromatin density coefficient. This method is similar to the procedure used to calculate colocalization coefficients (Manders et al., 1993). Regions included in measuring the phosphorous/nitrogen chromatin density coefficients were chosen based on whether the region was inside the -H2AX or Nbs1 fluorescence map or contained immunogold particles, in the case of the -irradiated sample. Equivalent areas adjacent and outside the irradiated regions were chosen for calculating the chromatin density coefficients of nondamaged chromatin except for the H2AX–/– sample, where a distant region of heterochromatin was included.
Online supplemental material
Fig. S1 shows the mobility of chromatin regions containing DNA DSBs. Fig. S2 shows the mobility and expansion of chromatin in WT living cells. Fig. S3 shows the changes in specific chromatin structure upon UV laser microirradiation. Fig. S4 shows a comparison of irradiation (Gy) with UV laser microirradiation. Fig. S5 shows a line scan through a net phosphorus ESI micrograph. Video 1 shows the mobility of photoactivated chromatin containing DNA DSBs. Video 2 shows the mobility of photoactivated chromatin without DNA DSBs. Video 3 shows the mobility of GFP-53BP1–containing IRIF in living cells. Online supplemental material is available at http://www.jcb.org/cgi/content/full/jcb.200510015/DC1.
|
Acknowledgments |
|---|
D.P. Bazett-Jones holds a Canada Research Chair in Molecular and Cellular Imaging, and G. Dellaire holds a Senior Research Fellowship from the Canadian Institutes of Health Research. This research was supported in part by the Intramural Research Program of the NCI's Center for Cancer Research and a grant from the AT Children's Project to A. Nussenzweig.
Submitted: 6 October 2005
Accepted: 9 February 2006
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Abney, J.R., B. Cutler, M.L. Fillbach, D. Axelrod, and B.A. Scalettar. 1997. Chromatin dynamics in interphase nuclei and its implications for nuclear structure. J. Cell Biol. 137:1459–1468.
Aten, J.A., J. Stap, P.M. Krawczyk, C.H. van Oven, R.A. Hoebe, J. Essers, and R. Kanaar. 2004. Dynamics of DNA double-strand breaks revealed by clustering of damaged chromosome domains. Science. 303:92–95.
Bakkenist, C.J., and M.B. Kastan. 2003. DNA damage activates ATM through intermolecular autophosphorylation and dimer dissociation. Nature. 421:499–506.[CrossRef][Medline]
Bakkenist, C.J., and M.B. Kastan. 2004. Initiating cellular stress responses. Cell. 118:9–17.[CrossRef][Medline]
Bassing, C.H., K.F. Chua, J. Sekiguchi, H. Suh, S.R. Whitlow, J.C. Fleming, B.C. Monroe, D.N. Ciccone, C. Yan, K. Vlasakova, et al. 2002. Increased ionizing radiation sensitivity and genomic instability in the absence of histone H2AX. Proc. Natl. Acad. Sci. USA. 99:8173–8178.
Bazett-Jones, D.P., and M.J. Hendzel. 1999. Electron spectroscopic imaging of chromatin. Methods. 17:188–200.[CrossRef][Medline]
Bird, A.W., D.Y. Yu, M.G. Pray-Grant, Q. Qiu, K.E. Harmon, P.C. Megee, P.A. Grant, M.M. Smith, and M.F. Christman. 2002. Acetylation of histone H4 by Esa1 is required for DNA double-strand break repair. Nature. 419:411–415.[CrossRef][Medline]
Celeste, A., S. Petersen, P.J. Romanienko, O. Fernandez-Capetillo, H.T. Chen, O.A. Sedelnikova, B. Reina-San-Martin, V. Coppola, E. Meffre, M.J. Difilippantonio, et al. 2002. Genomic instability in mice lacking histone H2AX. Science. 296:922–927.
Celeste, A., O. Fernandez-Capetillo, M.J. Kruhlak, D.R. Pilch, D.W. Staudt, A. Lee, R.F. Bonner, W.M. Bonner, and A. Nussenzweig. 2003. Histone H2AX phosphorylation is dispensable for the initial recognition of DNA breaks. Nat. Cell Biol. 5:675–679.[CrossRef][Medline]
Chen, H.T., A. Bhandoola, M.J. Difilippantonio, J. Zhu, M.J. Brown, X. Tai, E.P. Rogakou, T.M. Brotz, W.M. Bonner, T. Ried, and A. Nussenzweig. 2000. Response to RAG-mediated VDJ cleavage by NBS1 and gamma-H2AX. Science. 290:1962–1965.
Chubb, J.R., S. Boyle, P. Perry, and W.A. Bickmore. 2002. Chromatin motion is constrained by association with nuclear compartments in human cells. Curr. Biol. 12:439–445.[CrossRef][Medline]
Dellaire, G., R. Nisman, and D.P. Bazett-Jones. 2004. Correlative light and electron spectroscopic imaging of chromatin in situ. Methods Enzymol. 375:456–478.[CrossRef][Medline]
Downs, J.A., N.F. Lowndes, and S.P. Jackson. 2000. A role for Saccharomyces cerevisiae histone H2A in DNA repair. Nature. 408:1001–1004.[CrossRef][Medline]
Downs, J.A., S. Allard, O. Jobin-Robitaille, A. Javaheri, A. Auger, N. Bouchard, S.J. Kron, S.P. Jackson, and J. Cote. 2004. Binding of chromatin-modifying activities to phosphorylated histone H2A at DNA damage sites. Mol. Cell. 16:979–990.[CrossRef][Medline]
Fernandez-Capetillo, O., S.K. Mahadevaiah, A. Celeste, P.J. Romanienko, R.D. Camerini-Otero, W.M. Bonner, K. Manova, P. Burgoyne, and A. Nussenzweig. 2003. H2AX is required for chromatin remodeling and inactivation of sex chromosomes in male mouse meiosis. Dev. Cell. 4:497–508.[CrossRef][Medline]
Fernandez-Capetillo, O., C.D. Allis, and A. Nussenzweig. 2004a. Phosphorylation of histone H2B at DNA double-strand breaks. J. Exp. Med. 199:1671–1677.
Fernandez-Capetillo, O., A. Lee, M. Nussenzweig, and A. Nussenzweig. 2004b. H2AX: the histone guardian of the genome. DNA Repair (Amst.). 3:959–967.[CrossRef][Medline]
Hendzel, M.J., F. Boisvert, and D.P. Bazett-Jones. 1999. Direct visualization of a protein nuclear architecture. Mol. Biol. Cell. 10:2051–2062.
Hittelman, W.N., and M. Pollard. 1984. Visualization of chromatin events associated with repair of ultraviolet light-induced damage by premature chromosome condensation. Carcinogenesis. 5:1277–1285.
Houtsmuller, A.B., S. Rademakers, A.L. Nigg, D. Hoogstraten, J.H. Hoeijmakers, and W. Vermeulen. 1999. Action of DNA repair endonuclease ERCC1/XPF in living cells. Science. 284:958–961.
Huyen, Y., O. Zgheib, R.A. Ditullio Jr., V.G. Gorgoulis, P. Zacharatos, T.J. Petty, E.A. Sheston, H.S. Mellert, E.S. Stavridi, and T.D. Halazonetis. 2004. Methylated lysine 79 of histone H3 targets 53BP1 to DNA double-strand breaks. Nature. 432:406–411.[CrossRef][Medline]
Jaberaboansari, A., G.B. Nelson, J.L. Roti Roti, and K.T. Wheeler. 1988. Postirradiation alterations of neuronal chromatin structure. Radiat. Res. 114:94–104.[CrossRef][Medline]
Jazayeri, A., A.D. McAinsh, and S.P. Jackson. 2004. Saccharomyces cerevisiae Sin3p facilitates DNA double-strand break repair. Proc. Natl. Acad. Sci. USA. 101:1644–1649.
Kanda, T., K.F. Sullivan, and G.M. Wahl. 1998. Histone-GFP fusion protein enables sensitive analysis of chromosome dynamics in living mammalian cells. Curr. Biol. 8:377–385.[CrossRef][Medline]
Keogh, M.C., J.A. Kim, M. Downey, J. Fillingham, D. Chowdhury, J.C. Harrison, M. Onishi, N. Datta, S. Galicia, A. Emili, et al. 2005. A phosphatase complex that dephosphorylates gammaH2AX regulates DNA damage checkpoint recovery. Nature. 439:497–501.
Kimura, H., and P.R. Cook. 2001. Kinetics of core histones in living human cells: little exchange of H3 and H4 and some rapid exchange of H2B. J. Cell Biol. 153:1341–1353.
Kusch, T., L. Florens, W.H. Macdonald, S.K. Swanson, R.L. Glaser, J.R. Yates III, S.M. Abmayr, M.P. Washburn, and J.L. Workman. 2004. Acetylation by Tip60 is required for selective histone variant exchange at DNA lesions. Science. 306:2084–2087.
Lisby, M., U.H. Mortensen, and R. Rothstein. 2003. Colocalization of multiple DNA double-strand breaks at a single Rad52 repair centre. Nat. Cell Biol. 5:572–577.[CrossRef][Medline]
Ljungman, M. 1989. Pretreatment with UV light renders the chromatin in human fibroblasts more susceptible to the DNA-damaging agents bleomycin, gamma radiation and 8-methoxypsoralen. Carcinogenesis. 10:447–451.
Manders, E.M.M., F.J. Verbeek, and J.A. Aten. 1993. Measurement of co-localization of objects in dual-colour confocal images. J. Microsc. 169:375–382.
McNally, J.G., W.G. Muller, D. Walker, R. Wolford, and G.L. Hager. 2000. The glucocorticoid receptor: rapid exchange with regulatory sites in living cells. Science. 287:1262–1265.
Morrison, A.J., and X. Shen. 2005. DNA repair in the context of chromatin. Cell Cycle. 4:568–571.[Medline]
Morrison, A.J., J. Highland, N.J. Krogan, A. Arbel-Eden, J.F. Greenblatt, J.E. Haber, and X. Shen. 2004. INO80 and gamma-H2AX interaction links ATP-dependent chromatin remodeling to DNA damage repair. Cell. 119:767–775.[CrossRef][Medline]
Muller, W.G., D. Walker, G.L. Hager, and J.G. McNally. 2001. Large-scale chromatin decondensation and recondensation regulated by transcription from a natural promoter. J. Cell Biol. 154:33–48.
Murr, R., J.I. Loizou, Y.G. Yang, C. Cuenin, H. Li, Z.Q. Wang, and Z. Herceg. 2006. Histone acetylation by Trrap-Tip60 modulates loading of repair proteins and repair of DNA double-strand breaks. Nat. Cell Biol. 8:91–99.[CrossRef][Medline]
Nelms, B.E., R.S. Maser, J.F. MacKay, M.G. Lagally, and J.H. Petrini. 1998. In situ visualization of DNA double-strand break repair in human fibroblasts. Science. 280:590–592.
Nussenzweig, A., and T. Paull. 2006. DNA repair: tails of histones lost. Nature. 439:406–407.[CrossRef][Medline]
Patterson, G.H., and J. Lippincott-Schwartz. 2002. A photoactivatable GFP for selective photolabeling of proteins and cells. Science. 297:1873–1877.
Paull, T.T., E.P. Rogakou, V. Yamazaki, C.U. Kirchgessner, M. Gellert, and W.M. Bonner. 2000. A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr. Biol. 10:886–895.[CrossRef][Medline]
Peterson, C.L., and J. Cote. 2004. Cellular machineries for chromosomal DNA repair. Genes Dev. 18:602–616.
Pilch, D.R., O.A. Sedelnikova, C. Redon, A. Celeste, A. Nussenzweig, and W.M. Bonner. 2003. Characteristics of gamma-H2AX foci at DNA double-strand breaks sites. Biochem. Cell Biol. 81:123–129.[CrossRef][Medline]
Platani, M., I. Goldberg, A.I. Lamond, and J.R. Swedlow. 2002. Cajal body dynamics and association with chromatin are ATP-dependent. Nat. Cell Biol. 4:502–508.[CrossRef][Medline]
Qin, S., and M.R. Parthun. 2002. Histone H3 and the histone acetyltransferase Hat1p contribute to DNA double-strand break repair. Mol. Cell. Biol. 22:8353–8365.
